From solipsis at pitrou.net  Thu Jul  1 01:23:55 2010
From: solipsis at pitrou.net (solipsis at pitrou.net)
Date: Thu,  1 Jul 2010 01:23:55 +0200 (CEST)
Subject: [Python-checkins] Daily py3k reference leaks (r82407): sum=0
Message-ID: <20100630232355.4AB2A1770A@ns6635.ovh.net>

py3k results for svn r82407 (hg cset 40435013f940)
--------------------------------------------------



Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/antoine/py3k/refleaks/reflogluNgCL', '-x']

From python-checkins at python.org  Thu Jul  1 01:58:07 2010
From: python-checkins at python.org (ezio.melotti)
Date: Thu,  1 Jul 2010 01:58:07 +0200 (CEST)
Subject: [Python-checkins] r82412 - in tracker/instances/python-dev/html:
	issue.item.html issue.item.js
Message-ID: <20100630235807.73769EEA48@mail.python.org>

Author: ezio.melotti
Date: Thu Jul  1 01:58:07 2010
New Revision: 82412

Log:
Avoid using display:block -- the resize works with display:inline too. Also close the span properly in issue.item.html.

Modified:
   tracker/instances/python-dev/html/issue.item.html
   tracker/instances/python-dev/html/issue.item.js

Modified: tracker/instances/python-dev/html/issue.item.html
==============================================================================
--- tracker/instances/python-dev/html/issue.item.html	(original)
+++ tracker/instances/python-dev/html/issue.item.html	Thu Jul  1 01:58:07 2010
@@ -145,7 +145,7 @@
         tal:define="current_user request/user/username"
         tal:condition="python:request.user.username != 'anonymous' and current_user not in str(context.nosy).replace(' ','').split(',')"
         tal:attributes="onclick string:add_to_nosy('$current_user')"
-        onclick="add_to_nosy(the_current_username)" />
+        onclick="add_to_nosy(the_current_username)"></span>
  </td>
 </tr>
 <tr>

Modified: tracker/instances/python-dev/html/issue.item.js
==============================================================================
--- tracker/instances/python-dev/html/issue.item.js	(original)
+++ tracker/instances/python-dev/html/issue.item.js	Thu Jul  1 01:58:07 2010
@@ -38,6 +38,6 @@
     // hide the button and resize the list to fill the void
     var new_width = nosy.offsetWidth + add_me_button.offsetWidth;
     add_me_button.style.display = 'none';
-    nosy.style.display = 'block';
+    nosy.style.display = 'inline';
     nosy.style.width = new_width + "px";
 }

From python-checkins at python.org  Thu Jul  1 09:32:02 2010
From: python-checkins at python.org (ezio.melotti)
Date: Thu,  1 Jul 2010 09:32:02 +0200 (CEST)
Subject: [Python-checkins] r82413 - in python/branches/py3k:
	Lib/test/test_codeccallbacks.py Lib/test/test_unicode.py
	Misc/NEWS Objects/unicodeobject.c
Message-ID: <20100701073202.76D7CEEAEF@mail.python.org>

Author: ezio.melotti
Date: Thu Jul  1 09:32:02 2010
New Revision: 82413

Log:
Update PyUnicode_DecodeUTF8 from RFC 2279 to RFC 3629.

1) #8271: when a byte sequence is invalid, only the start byte and all the
   valid continuation bytes are now replaced by U+FFFD, instead of replacing
   the number of bytes specified by the start byte.
   See http://www.unicode.org/versions/Unicode5.2.0/ch03.pdf (pages 94-95);
2) 5- and 6-bytes-long UTF-8 sequences are now considered invalid (no changes
   in behavior);
3) Change the error messages "unexpected code byte" to "invalid start byte"
   and "invalid data" to "invalid continuation byte";
4) Add an extensive set of tests in test_unicode;
5) Fix test_codeccallbacks because it was failing after this change.


Modified:
   python/branches/py3k/   (props changed)
   python/branches/py3k/Lib/test/test_codeccallbacks.py
   python/branches/py3k/Lib/test/test_unicode.py
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/Objects/unicodeobject.c

Modified: python/branches/py3k/Lib/test/test_codeccallbacks.py
==============================================================================
--- python/branches/py3k/Lib/test/test_codeccallbacks.py	(original)
+++ python/branches/py3k/Lib/test/test_codeccallbacks.py	Thu Jul  1 09:32:02 2010
@@ -153,28 +153,30 @@
             sout += bytes("\\U%08x" % sys.maxunicode, "ascii")
         self.assertEqual(sin.encode("iso-8859-15", "backslashreplace"), sout)
 
-    def test_decoderelaxedutf8(self):
-        # This is the test for a decoding callback handler,
-        # that relaxes the UTF-8 minimal encoding restriction.
-        # A null byte that is encoded as "\xc0\x80" will be
-        # decoded as a null byte. All other illegal sequences
-        # will be handled strictly.
+    def test_decoding_callbacks(self):
+        # This is a test for a decoding callback handler
+        # that allows the decoding of the invalid sequence
+        # "\xc0\x80" and returns "\x00" instead of raising an error.
+        # All other illegal sequences will be handled strictly.
         def relaxedutf8(exc):
             if not isinstance(exc, UnicodeDecodeError):
                 raise TypeError("don't know how to handle %r" % exc)
-            if exc.object[exc.start:exc.end].startswith(b"\xc0\x80"):
+            if exc.object[exc.start:exc.start+2] == b"\xc0\x80":
                 return ("\x00", exc.start+2) # retry after two bytes
             else:
                 raise exc
 
-        codecs.register_error(
-            "test.relaxedutf8", relaxedutf8)
+        codecs.register_error("test.relaxedutf8", relaxedutf8)
 
+        # all the "\xc0\x80" will be decoded to "\x00"
         sin = b"a\x00b\xc0\x80c\xc3\xbc\xc0\x80\xc0\x80"
         sout = "a\x00b\x00c\xfc\x00\x00"
         self.assertEqual(sin.decode("utf-8", "test.relaxedutf8"), sout)
+
+        # "\xc0\x81" is not valid and a UnicodeDecodeError will be raised
         sin = b"\xc0\x80\xc0\x81"
-        self.assertRaises(UnicodeError, sin.decode, "utf-8", "test.relaxedutf8")
+        self.assertRaises(UnicodeDecodeError, sin.decode,
+                          "utf-8", "test.relaxedutf8")
 
     def test_charmapencode(self):
         # For charmap encodings the replacement string will be

Modified: python/branches/py3k/Lib/test/test_unicode.py
==============================================================================
--- python/branches/py3k/Lib/test/test_unicode.py	(original)
+++ python/branches/py3k/Lib/test/test_unicode.py	Thu Jul  1 09:32:02 2010
@@ -942,6 +942,159 @@
         # * strict decoding testing for all of the
         #   UTF8_ERROR cases in PyUnicode_DecodeUTF8
 
+    def test_utf8_decode_valid_sequences(self):
+        sequences = [
+            # single byte
+            (b'\x00', '\x00'), (b'a', 'a'), (b'\x7f', '\x7f'),
+            # 2 bytes
+            (b'\xc2\x80', '\x80'), (b'\xdf\xbf', '\u07ff'),
+            # 3 bytes
+            (b'\xe0\xa0\x80', '\u0800'), (b'\xed\x9f\xbf', '\ud7ff'),
+            (b'\xee\x80\x80', '\uE000'), (b'\xef\xbf\xbf', '\uffff'),
+            # 4 bytes
+            (b'\xF0\x90\x80\x80', '\U00010000'),
+            (b'\xf4\x8f\xbf\xbf', '\U0010FFFF')
+        ]
+        for seq, res in sequences:
+            self.assertEqual(seq.decode('utf-8'), res)
+
+
+    def test_utf8_decode_invalid_sequences(self):
+        # continuation bytes in a sequence of 2, 3, or 4 bytes
+        continuation_bytes = [bytes([x]) for x in range(0x80, 0xC0)]
+        # start bytes of a 2-byte sequence equivalent to codepoints < 0x7F
+        invalid_2B_seq_start_bytes = [bytes([x]) for x in range(0xC0, 0xC2)]
+        # start bytes of a 4-byte sequence equivalent to codepoints > 0x10FFFF
+        invalid_4B_seq_start_bytes = [bytes([x]) for x in range(0xF5, 0xF8)]
+        invalid_start_bytes = (
+            continuation_bytes + invalid_2B_seq_start_bytes +
+            invalid_4B_seq_start_bytes + [bytes([x]) for x in range(0xF7, 0x100)]
+        )
+
+        for byte in invalid_start_bytes:
+            self.assertRaises(UnicodeDecodeError, byte.decode, 'utf-8')
+
+        for sb in invalid_2B_seq_start_bytes:
+            for cb in continuation_bytes:
+                self.assertRaises(UnicodeDecodeError, (sb+cb).decode, 'utf-8')
+
+        for sb in invalid_4B_seq_start_bytes:
+            for cb1 in continuation_bytes[:3]:
+                for cb3 in continuation_bytes[:3]:
+                    self.assertRaises(UnicodeDecodeError,
+                                      (sb+cb1+b'\x80'+cb3).decode, 'utf-8')
+
+        for cb in [bytes([x]) for x in range(0x80, 0xA0)]:
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xE0'+cb+b'\x80').decode, 'utf-8')
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xE0'+cb+b'\xBF').decode, 'utf-8')
+        # surrogates
+        for cb in [bytes([x]) for x in range(0xA0, 0xC0)]:
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xED'+cb+b'\x80').decode, 'utf-8')
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xED'+cb+b'\xBF').decode, 'utf-8')
+        for cb in [bytes([x]) for x in range(0x80, 0x90)]:
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xF0'+cb+b'\x80\x80').decode, 'utf-8')
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xF0'+cb+b'\xBF\xBF').decode, 'utf-8')
+        for cb in [bytes([x]) for x in range(0x90, 0xC0)]:
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xF4'+cb+b'\x80\x80').decode, 'utf-8')
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xF4'+cb+b'\xBF\xBF').decode, 'utf-8')
+
+    def test_issue8271(self):
+        # Issue #8271: during the decoding of an invalid UTF-8 byte sequence,
+        # only the start byte and the continuation byte(s) are now considered
+        # invalid, instead of the number of bytes specified by the start byte.
+        # See http://www.unicode.org/versions/Unicode5.2.0/ch03.pdf (page 95,
+        # table 3-8, Row 2) for more information about the algorithm used.
+        FFFD = '\ufffd'
+        sequences = [
+            # invalid start bytes
+            (b'\x80', FFFD), # continuation byte
+            (b'\x80\x80', FFFD*2), # 2 continuation bytes
+            (b'\xc0', FFFD),
+            (b'\xc0\xc0', FFFD*2),
+            (b'\xc1', FFFD),
+            (b'\xc1\xc0', FFFD*2),
+            (b'\xc0\xc1', FFFD*2),
+            # with start byte of a 2-byte sequence
+            (b'\xc2', FFFD), # only the start byte
+            (b'\xc2\xc2', FFFD*2), # 2 start bytes
+            (b'\xc2\xc2\xc2', FFFD*3), # 2 start bytes
+            (b'\xc2\x41', FFFD+'A'), # invalid continuation byte
+            # with start byte of a 3-byte sequence
+            (b'\xe1', FFFD), # only the start byte
+            (b'\xe1\xe1', FFFD*2), # 2 start bytes
+            (b'\xe1\xe1\xe1', FFFD*3), # 3 start bytes
+            (b'\xe1\xe1\xe1\xe1', FFFD*4), # 4 start bytes
+            (b'\xe1\x80', FFFD), # only 1 continuation byte
+            (b'\xe1\x41', FFFD+'A'), # invalid continuation byte
+            (b'\xe1\x41\x80', FFFD+'A'+FFFD), # invalid cb followed by valid cb
+            (b'\xe1\x41\x41', FFFD+'AA'), # 2 invalid continuation bytes
+            (b'\xe1\x80\x41', FFFD+'A'), # only 1 valid continuation byte
+            (b'\xe1\x80\xe1\x41', FFFD*2+'A'), # 1 valid and the other invalid
+            (b'\xe1\x41\xe1\x80', FFFD+'A'+FFFD), # 1 invalid and the other valid
+            # with start byte of a 4-byte sequence
+            (b'\xf1', FFFD), # only the start byte
+            (b'\xf1\xf1', FFFD*2), # 2 start bytes
+            (b'\xf1\xf1\xf1', FFFD*3), # 3 start bytes
+            (b'\xf1\xf1\xf1\xf1', FFFD*4), # 4 start bytes
+            (b'\xf1\xf1\xf1\xf1\xf1', FFFD*5), # 5 start bytes
+            (b'\xf1\x80', FFFD), # only 1 continuation bytes
+            (b'\xf1\x80\x80', FFFD), # only 2 continuation bytes
+            (b'\xf1\x80\x41', FFFD+'A'), # 1 valid cb and 1 invalid
+            (b'\xf1\x80\x41\x41', FFFD+'AA'), # 1 valid cb and 1 invalid
+            (b'\xf1\x80\x80\x41', FFFD+'A'), # 2 valid cb and 1 invalid
+            (b'\xf1\x41\x80', FFFD+'A'+FFFD), # 1 invalid cv and 1 valid
+            (b'\xf1\x41\x80\x80', FFFD+'A'+FFFD*2), # 1 invalid cb and 2 invalid
+            (b'\xf1\x41\x80\x41', FFFD+'A'+FFFD+'A'), # 2 invalid cb and 1 invalid
+            (b'\xf1\x41\x41\x80', FFFD+'AA'+FFFD), # 1 valid cb and 1 invalid
+            (b'\xf1\x41\xf1\x80', FFFD+'A'+FFFD),
+            (b'\xf1\x41\x80\xf1', FFFD+'A'+FFFD*2),
+            (b'\xf1\xf1\x80\x41', FFFD*2+'A'),
+            (b'\xf1\x41\xf1\xf1', FFFD+'A'+FFFD*2),
+            # with invalid start byte of a 4-byte sequence (rfc2279)
+            (b'\xf5', FFFD), # only the start byte
+            (b'\xf5\xf5', FFFD*2), # 2 start bytes
+            (b'\xf5\x80', FFFD*2), # only 1 continuation byte
+            (b'\xf5\x80\x80', FFFD*3), # only 2 continuation byte
+            (b'\xf5\x80\x80\x80', FFFD*4), # 3 continuation bytes
+            (b'\xf5\x80\x41', FFFD*2+'A'), #  1 valid cb and 1 invalid
+            (b'\xf5\x80\x41\xf5', FFFD*2+'A'+FFFD),
+            (b'\xf5\x41\x80\x80\x41', FFFD+'A'+FFFD*2+'A'),
+            # with invalid start byte of a 5-byte sequence (rfc2279)
+            (b'\xf8', FFFD), # only the start byte
+            (b'\xf8\xf8', FFFD*2), # 2 start bytes
+            (b'\xf8\x80', FFFD*2), # only one continuation byte
+            (b'\xf8\x80\x41', FFFD*2 + 'A'), # 1 valid cb and 1 invalid
+            (b'\xf8\x80\x80\x80\x80', FFFD*5), # invalid 5 bytes seq with 5 bytes
+            # with invalid start byte of a 6-byte sequence (rfc2279)
+            (b'\xfc', FFFD), # only the start byte
+            (b'\xfc\xfc', FFFD*2), # 2 start bytes
+            (b'\xfc\x80\x80', FFFD*3), # only 2 continuation bytes
+            (b'\xfc\x80\x80\x80\x80\x80', FFFD*6), # 6 continuation bytes
+            # invalid start byte
+            (b'\xfe', FFFD),
+            (b'\xfe\x80\x80', FFFD*3),
+            # other sequences
+            (b'\xf1\x80\x41\x42\x43', '\ufffd\x41\x42\x43'),
+            (b'\xf1\x80\xff\x42\x43', '\ufffd\ufffd\x42\x43'),
+            (b'\xf1\x80\xc2\x81\x43', '\ufffd\x81\x43'),
+            (b'\x61\xF1\x80\x80\xE1\x80\xC2\x62\x80\x63\x80\xBF\x64',
+             '\x61\uFFFD\uFFFD\uFFFD\x62\uFFFD\x63\uFFFD\uFFFD\x64'),
+        ]
+        for n, (seq, res) in enumerate(sequences):
+            self.assertRaises(UnicodeDecodeError, seq.decode, 'utf-8', 'strict')
+            self.assertEqual(seq.decode('utf-8', 'replace'), res)
+            self.assertEqual((seq+b'b').decode('utf-8', 'replace'), res+'b')
+            self.assertEqual(seq.decode('utf-8', 'ignore'),
+                             res.replace('\uFFFD', ''))
+
     def test_codecs_idna(self):
         # Test whether trailing dot is preserved
         self.assertEqual("www.python.org.".encode("idna"), b"www.python.org.")

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Thu Jul  1 09:32:02 2010
@@ -12,6 +12,14 @@
 Core and Builtins
 -----------------
 
+- Issue #8271: during the decoding of an invalid UTF-8 byte sequence, only the
+  start byte and the continuation byte(s) are now considered invalid, instead
+  of the number of bytes specified by the start byte.
+  E.g.: '\xf1\x80AB'.decode('utf-8', 'replace') now returns u'\ufffdAB' and
+  replaces with U+FFFD only the start byte ('\xf1') and the continuation byte
+  ('\x80') even if '\xf1' is the start byte of a 4-bytes sequence.
+  Previous versions returned a single u'\ufffd'.
+
 - Issue #9011: A negated imaginary literal (e.g., "-7j") now has real
   part -0.0 rather than 0.0.  So "-7j" is now exactly equivalent to
   "-(7j)".

Modified: python/branches/py3k/Objects/unicodeobject.c
==============================================================================
--- python/branches/py3k/Objects/unicodeobject.c	(original)
+++ python/branches/py3k/Objects/unicodeobject.c	Thu Jul  1 09:32:02 2010
@@ -2285,24 +2285,24 @@
 
 static
 char utf8_code_length[256] = {
-    /* Map UTF-8 encoded prefix byte to sequence length.  zero means
-       illegal prefix.  see RFC 2279 for details */
+    /* Map UTF-8 encoded prefix byte to sequence length.  Zero means
+       illegal prefix.  See RFC 3629 for details */
+    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 00-0F */
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
-    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
-    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
-    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 70-7F */
+    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 80-8F */
     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
-    2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
-    3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
-    4, 4, 4, 4, 4, 4, 4, 4, 5, 5, 5, 5, 6, 6, 0, 0
+    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* B0-BF */
+    0, 0, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, /* C0-C1 + C2-CF */
+    2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, /* D0-DF */
+    3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, /* E0-EF */
+    4, 4, 4, 4, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0  /* F0-F4 + F5-FF */
 };
 
 PyObject *PyUnicode_DecodeUTF8(const char *s,
@@ -2332,6 +2332,7 @@
 {
     const char *starts = s;
     int n;
+    int k;
     Py_ssize_t startinpos;
     Py_ssize_t endinpos;
     Py_ssize_t outpos;
@@ -2415,7 +2416,9 @@
             else {
                 errmsg = "unexpected end of data";
                 startinpos = s-starts;
-                endinpos = size;
+                endinpos = startinpos+1;
+                for (k=1; (k < size-startinpos) && ((s[k]&0xC0) == 0x80); k++)
+                    endinpos++;
                 goto utf8Error;
             }
         }
@@ -2423,7 +2426,7 @@
         switch (n) {
 
         case 0:
-            errmsg = "unexpected code byte";
+            errmsg = "invalid start byte";
             startinpos = s-starts;
             endinpos = startinpos+1;
             goto utf8Error;
@@ -2436,63 +2439,67 @@
 
         case 2:
             if ((s[1] & 0xc0) != 0x80) {
-                errmsg = "invalid data";
+                errmsg = "invalid continuation byte";
                 startinpos = s-starts;
-                endinpos = startinpos+2;
+                endinpos = startinpos + 1;
                 goto utf8Error;
             }
             ch = ((s[0] & 0x1f) << 6) + (s[1] & 0x3f);
-            if (ch < 0x80) {
-                startinpos = s-starts;
-                endinpos = startinpos+2;
-                errmsg = "illegal encoding";
-                goto utf8Error;
-            }
-            else
-                *p++ = (Py_UNICODE)ch;
+            assert ((ch > 0x007F) && (ch <= 0x07FF));
+            *p++ = (Py_UNICODE)ch;
             break;
 
         case 3:
+            /* XXX: surrogates shouldn't be valid UTF-8!
+               see http://www.unicode.org/versions/Unicode5.2.0/ch03.pdf
+               (table 3-7) and http://www.rfc-editor.org/rfc/rfc3629.txt
+               Uncomment the 2 lines below to make them invalid,
+               codepoints: d800-dfff; UTF-8: \xed\xa0\x80-\xed\xbf\xbf. */
             if ((s[1] & 0xc0) != 0x80 ||
-                (s[2] & 0xc0) != 0x80) {
-                errmsg = "invalid data";
+                (s[2] & 0xc0) != 0x80 ||
+                ((unsigned char)s[0] == 0xE0 &&
+                 (unsigned char)s[1] < 0xA0) ||
+                ((unsigned char)s[0] == 0xED &&
+                 (unsigned char)s[1] > 0x9F)) {
+                errmsg = "invalid continuation byte";
                 startinpos = s-starts;
-                endinpos = startinpos+3;
+                endinpos = startinpos + 1;
+
+                /* if s[1] first two bits are 1 and 0, then the invalid
+                   continuation byte is s[2], so increment endinpos by 1,
+                   if not, s[1] is invalid and endinpos doesn't need to
+                   be incremented. */
+                if ((s[1] & 0xC0) == 0x80)
+                    endinpos++;
                 goto utf8Error;
             }
             ch = ((s[0] & 0x0f) << 12) + ((s[1] & 0x3f) << 6) + (s[2] & 0x3f);
-            if (ch < 0x0800 || (ch >= 0xd800 && ch <= 0xDFFF)) {
-                errmsg = "illegal encoding";
-                startinpos = s-starts;
-                endinpos = startinpos+3;
-                goto utf8Error;
-            }
-            else
-                *p++ = (Py_UNICODE)ch;
+            assert ((ch > 0x07FF) && (ch <= 0xFFFF));
+            *p++ = (Py_UNICODE)ch;
             break;
 
         case 4:
             if ((s[1] & 0xc0) != 0x80 ||
                 (s[2] & 0xc0) != 0x80 ||
-                (s[3] & 0xc0) != 0x80) {
-                errmsg = "invalid data";
+                (s[3] & 0xc0) != 0x80 ||
+                ((unsigned char)s[0] == 0xF0 &&
+                 (unsigned char)s[1] < 0x90) ||
+                ((unsigned char)s[0] == 0xF4 &&
+                 (unsigned char)s[1] > 0x8F)) {
+                errmsg = "invalid continuation byte";
                 startinpos = s-starts;
-                endinpos = startinpos+4;
+                endinpos = startinpos + 1;
+                if ((s[1] & 0xC0) == 0x80) {
+                    endinpos++;
+                    if ((s[2] & 0xC0) == 0x80)
+                        endinpos++;
+                }
                 goto utf8Error;
             }
             ch = ((s[0] & 0x7) << 18) + ((s[1] & 0x3f) << 12) +
-                ((s[2] & 0x3f) << 6) + (s[3] & 0x3f);
-            /* validate and convert to UTF-16 */
-            if ((ch < 0x10000)        /* minimum value allowed for 4
-                                         byte encoding */
-                || (ch > 0x10ffff))   /* maximum value allowed for
-                                         UTF-16 */
-            {
-                errmsg = "illegal encoding";
-                startinpos = s-starts;
-                endinpos = startinpos+4;
-                goto utf8Error;
-            }
+                 ((s[2] & 0x3f) << 6) + (s[3] & 0x3f);
+            assert ((ch > 0xFFFF) && (ch <= 0x10ffff));
+
 #ifdef Py_UNICODE_WIDE
             *p++ = (Py_UNICODE)ch;
 #else
@@ -2508,13 +2515,6 @@
             *p++ = (Py_UNICODE)(0xDC00 + (ch & 0x03FF));
 #endif
             break;
-
-        default:
-            /* Other sizes are only needed for UCS-4 */
-            errmsg = "unsupported Unicode code range";
-            startinpos = s-starts;
-            endinpos = startinpos+n;
-            goto utf8Error;
         }
         s += n;
         continue;

From python-checkins at python.org  Thu Jul  1 17:07:15 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Thu,  1 Jul 2010 17:07:15 +0200 (CEST)
Subject: [Python-checkins] r82415 - python/trunk/Doc/library/math.rst
Message-ID: <20100701150715.51429F6915@mail.python.org>

Author: benjamin.peterson
Date: Thu Jul  1 17:07:15 2010
New Revision: 82415

Log:
remove docs about delegating to special methods; it does no such thing

Modified:
   python/trunk/Doc/library/math.rst

Modified: python/trunk/Doc/library/math.rst
==============================================================================
--- python/trunk/Doc/library/math.rst	(original)
+++ python/trunk/Doc/library/math.rst	Thu Jul  1 17:07:15 2010
@@ -57,9 +57,6 @@
    Return the floor of *x* as a float, the largest integer value less than or equal
    to *x*.
 
-   .. versionchanged:: 2.6
-      Added :meth:`__floor__` delegation.
-
 
 .. function:: fmod(x, y)
 
@@ -137,7 +134,7 @@
 .. function:: trunc(x)
 
    Return the :class:`Real` value *x* truncated to an :class:`Integral` (usually
-   a long integer). Delegates to ``x.__trunc__()``.
+   a long integer).
 
    .. versionadded:: 2.6
 

From python-checkins at python.org  Thu Jul  1 17:09:15 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Thu,  1 Jul 2010 17:09:15 +0200 (CEST)
Subject: [Python-checkins] r82416 - python/branches/py3k
Message-ID: <20100701150915.09235EEB64@mail.python.org>

Author: benjamin.peterson
Date: Thu Jul  1 17:09:14 2010
New Revision: 82416

Log:
Blocked revisions 82415 via svnmerge

........
  r82415 | benjamin.peterson | 2010-07-01 10:07:15 -0500 (Thu, 01 Jul 2010) | 1 line
  
  remove docs about delegating to special methods; it does no such thing
........


Modified:
   python/branches/py3k/   (props changed)

From python-checkins at python.org  Thu Jul  1 17:16:55 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Thu,  1 Jul 2010 17:16:55 +0200 (CEST)
Subject: [Python-checkins] r82417 - in python/branches/py3k:
	Lib/test/test_descr.py Misc/NEWS Modules/mathmodule.c
Message-ID: <20100701151655.6E378EEBB1@mail.python.org>

Author: benjamin.peterson
Date: Thu Jul  1 17:16:55 2010
New Revision: 82417

Log:
correctly lookup __trunc__ and __floor__

Modified:
   python/branches/py3k/Lib/test/test_descr.py
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/Modules/mathmodule.c

Modified: python/branches/py3k/Lib/test/test_descr.py
==============================================================================
--- python/branches/py3k/Lib/test/test_descr.py	(original)
+++ python/branches/py3k/Lib/test/test_descr.py	Thu Jul  1 17:16:55 2010
@@ -1,6 +1,7 @@
 import builtins
 import sys
 import types
+import math
 import unittest
 
 from copy import deepcopy
@@ -1578,6 +1579,8 @@
             ("__exit__", run_context, swallow, set(), {"__enter__" : iden}),
             ("__complex__", complex, complex_num, set(), {}),
             ("__format__", format, format_impl, set(), {}),
+            ("__floor__", math.floor, zero, set(), {}),
+            ("__trunc__", math.trunc, zero, set(), {}),
             ]
 
         class Checker(object):

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Thu Jul  1 17:16:55 2010
@@ -1374,6 +1374,9 @@
 Extension Modules
 -----------------
 
+- In the math module, correctly lookup __trunc__ and __floor__ as special
+  methods.
+
 - Issue #9005: Prevent utctimetuple() from producing year 0 or year
   10,000.  Prior to this change, timezone adjustment in utctimetuple()
   could produce tm_year value of 0 or 10,000.  Now an OverflowError is

Modified: python/branches/py3k/Modules/mathmodule.c
==============================================================================
--- python/branches/py3k/Modules/mathmodule.c	(original)
+++ python/branches/py3k/Modules/mathmodule.c	Thu Jul  1 17:16:55 2010
@@ -883,17 +883,13 @@
     static PyObject *floor_str = NULL;
     PyObject *method;
 
-    if (floor_str == NULL) {
-        floor_str = PyUnicode_InternFromString("__floor__");
-        if (floor_str == NULL)
+    method = _PyObject_LookupSpecial(number, "__floor__", &floor_str);
+    if (method == NULL) {
+        if (PyErr_Occurred())
             return NULL;
-    }
-
-    method = _PyType_Lookup(Py_TYPE(number), floor_str);
-    if (method == NULL)
         return math_1_to_int(number, floor, 0);
-    else
-        return PyObject_CallFunction(method, "O", number);
+    }
+    return PyObject_CallFunctionObjArgs(method, NULL);
 }
 
 PyDoc_STRVAR(math_floor_doc,
@@ -1427,20 +1423,15 @@
             return NULL;
     }
 
-    if (trunc_str == NULL) {
-        trunc_str = PyUnicode_InternFromString("__trunc__");
-        if (trunc_str == NULL)
-            return NULL;
-    }
-
-    trunc = _PyType_Lookup(Py_TYPE(number), trunc_str);
+    trunc = _PyObject_LookupSpecial(number, "__trunc__", &trunc_str);
     if (trunc == NULL) {
-        PyErr_Format(PyExc_TypeError,
-                     "type %.100s doesn't define __trunc__ method",
-                     Py_TYPE(number)->tp_name);
+        if (!PyErr_Occurred())
+            PyErr_Format(PyExc_TypeError,
+                         "type %.100s doesn't define __trunc__ method",
+                         Py_TYPE(number)->tp_name);
         return NULL;
     }
-    return PyObject_CallFunctionObjArgs(trunc, number, NULL);
+    return PyObject_CallFunctionObjArgs(trunc, NULL);
 }
 
 PyDoc_STRVAR(math_trunc_doc,

From python-checkins at python.org  Thu Jul  1 18:16:24 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Thu,  1 Jul 2010 18:16:24 +0200 (CEST)
Subject: [Python-checkins] r82418 - in python/branches/release26-maint:
	Doc/library/math.rst
Message-ID: <20100701161624.EE662F692F@mail.python.org>

Author: benjamin.peterson
Date: Thu Jul  1 18:16:24 2010
New Revision: 82418

Log:
Merged revisions 82415 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/trunk

........
  r82415 | benjamin.peterson | 2010-07-01 10:07:15 -0500 (Thu, 01 Jul 2010) | 1 line
  
  remove docs about delegating to special methods; it does no such thing
........


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Doc/library/math.rst

Modified: python/branches/release26-maint/Doc/library/math.rst
==============================================================================
--- python/branches/release26-maint/Doc/library/math.rst	(original)
+++ python/branches/release26-maint/Doc/library/math.rst	Thu Jul  1 18:16:24 2010
@@ -57,9 +57,6 @@
    Return the floor of *x* as a float, the largest integer value less than or equal
    to *x*.
 
-   .. versionchanged:: 2.6
-      Added :meth:`__floor__` delegation.
-
 
 .. function:: fmod(x, y)
 
@@ -137,7 +134,7 @@
 .. function:: trunc(x)
 
    Return the :class:`Real` value *x* truncated to an :class:`Integral` (usually
-   a long integer). Delegates to ``x.__trunc__()``.
+   a long integer).
 
    .. versionadded:: 2.6
 

From python-checkins at python.org  Thu Jul  1 18:16:35 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Thu,  1 Jul 2010 18:16:35 +0200 (CEST)
Subject: [Python-checkins] r82419 - in sandbox/branches/py3k-datetime:
	datetime.py test_datetime.py
Message-ID: <20100701161635.88A6CEED3A@mail.python.org>

Author: alexander.belopolsky
Date: Thu Jul  1 18:16:35 2010
New Revision: 82419

Log:
Fixed pickling to pass a more rigid test suit.

Modified:
   sandbox/branches/py3k-datetime/datetime.py
   sandbox/branches/py3k-datetime/test_datetime.py

Modified: sandbox/branches/py3k-datetime/datetime.py
==============================================================================
--- sandbox/branches/py3k-datetime/datetime.py	(original)
+++ sandbox/branches/py3k-datetime/datetime.py	Thu Jul  1 18:16:35 2010
@@ -269,6 +269,8 @@
     raise ValueError("%s()=%d, must be in -1439..1439" % (name, offset))
 
 def _check_date_fields(year, month, day):
+    if not isinstance(year, int):
+        raise TypeError('int expected')
     if not MINYEAR <= year <= MAXYEAR:
         raise ValueError('year must be in %d..%d' % (MINYEAR, MAXYEAR), year)
     if not 1 <= month <= 12:
@@ -278,6 +280,8 @@
         raise ValueError('day must be in 1..%d' % dim, day)
 
 def _check_time_fields(hour, minute, second, microsecond):
+    if not isinstance(hour, int):
+        raise TypeError('int expected')
     if not 0 <= hour <= 23:
         raise ValueError('hour must be in 0..23', hour)
     if not 0 <= minute <= 59:
@@ -660,7 +664,7 @@
 
         year, month, day (required, base 1)
         """
-        if isinstance(year, bytes):
+        if isinstance(year, bytes) and len(year) == 4:
             # Pickle support
             self = object.__new__(cls)
             self.__setstate(year)
@@ -1008,7 +1012,7 @@
         tzinfo (default to None)
         """
         self = object.__new__(cls)
-        if isinstance(hour, str):
+        if isinstance(hour, bytes) and len(hour) == 6:
             # Pickle support
             self.__setstate(hour, minute or None)
             return self
@@ -1256,18 +1260,18 @@
     def __getstate(self):
         us2, us3 = divmod(self.__microsecond, 256)
         us1, us2 = divmod(us2, 256)
-        basestate = ("%c" * 6) % (self.__hour, self.__minute, self.__second,
-                                  us1, us2, us3)
+        basestate = bytes([self.__hour, self.__minute, self.__second,
+                           us1, us2, us3])
         if self._tzinfo is None:
             return (basestate,)
         else:
             return (basestate, self._tzinfo)
 
     def __setstate(self, string, tzinfo):
-        if len(string) != 6 or ord(string[0]) >= 24:
+        if len(string) != 6 or string[0] >= 24:
             raise TypeError("an integer is required")
-        self.__hour, self.__minute, self.__second, us1, us2, us3 = \
-                                                            map(ord, string)
+        (self.__hour, self.__minute, self.__second,
+         us1, us2, us3) = string
         self.__microsecond = (((us1 << 8) | us2) << 8) | us3
         self._tzinfo = tzinfo
 
@@ -1287,7 +1291,7 @@
 
     def __new__(cls, year, month=None, day=None, hour=0, minute=0, second=0,
                 microsecond=0, tzinfo=None):
-        if isinstance(year, bytes):
+        if isinstance(year, bytes) and len(year) == 10:
             # Pickle support
             self = date.__new__(cls, year[:4])
             self.__setstate(year, month)

Modified: sandbox/branches/py3k-datetime/test_datetime.py
==============================================================================
--- sandbox/branches/py3k-datetime/test_datetime.py	(original)
+++ sandbox/branches/py3k-datetime/test_datetime.py	Thu Jul  1 18:16:35 2010
@@ -261,6 +261,33 @@
                              t.replace(tzinfo=tz).dst())
 
 #############################################################################
+# Base clase for testing pickling state of timedelta, time, date and
+# datetime instances.
+
+class ReduceStructure:
+    initargs = 1, 1, 1
+    statetype = (bytes, tzinfo)
+    def test_reduce(self):
+        obj = self.theclass(*self.initargs)
+        factory, state = obj.__reduce__()[:2]
+        self.assertIs(factory, self.theclass)
+        for v, t in zip(state, self.statetype):
+            self.assertIsInstance(v, t)
+
+    def test_construct_from_state(self):
+        obj = self.theclass(*self.initargs)
+        factory, state = obj.__reduce__()[:2]
+        initbytes = state[0]
+        # This test should be overriden in tests for classes that
+        # don't produce their state as bytes.  Curently, timedelta
+        # and timezone
+        assert isinstance(initbytes, bytes)
+        obj = factory(initbytes)
+        self.assertIs(type(obj), self.theclass)
+        self.assertRaises(TypeError, factory, initbytes[:-1]) 
+        self.assertRaises(TypeError, factory, initbytes + b'x') 
+
+#############################################################################
 # Base clase for testing a particular aspect of timedelta, time, date and
 # datetime comparisons.
 
@@ -297,10 +324,10 @@
 #############################################################################
 # timedelta tests
 
-class TestTimeDelta(HarmlessMixedComparison, unittest.TestCase):
+class TestTimeDelta(HarmlessMixedComparison, ReduceStructure, unittest.TestCase):
 
     theclass = timedelta
-
+    statetype = (int, int, int)
     def test_constructor(self):
         eq = self.assertEqual
         td = timedelta
@@ -326,6 +353,9 @@
         eq(td(seconds=0.001), td(milliseconds=1))
         eq(td(milliseconds=0.001), td(microseconds=1))
 
+    def test_construct_from_state(self):
+        pass
+
     def test_computations(self):
         eq = self.assertEqual
         td = timedelta
@@ -758,11 +788,12 @@
 class SubclassDate(date):
     sub_var = 1
 
-class TestDate(HarmlessMixedComparison, unittest.TestCase):
+class TestDate(HarmlessMixedComparison, ReduceStructure, unittest.TestCase):
     # Tests here should pass for both dates and datetimes, except for a
     # few tests that TestDateTime overrides.
 
     theclass = date
+    statetype = (bytes,)
 
     def test_basic_attributes(self):
         dt = self.theclass(2002, 3, 1)
@@ -1387,6 +1418,7 @@
 class TestDateTime(TestDate):
 
     theclass = datetime
+    statetype = (bytes, tzinfo)
 
     def test_basic_attributes(self):
         dt = self.theclass(2002, 3, 1, 12, 0)
@@ -1735,6 +1767,8 @@
         # than one microsecond smaller than an integer.
         self.assertEqual(self.theclass.fromtimestamp(0.9999999),
                          self.theclass.fromtimestamp(1))
+        self.assertEqual(self.theclass.utcfromtimestamp(0.9999999),
+                         self.theclass.utcfromtimestamp(1))
 
     def test_insane_fromtimestamp(self):
         # It's possible that some platform maps time_t to double,
@@ -1962,7 +1996,7 @@
 class SubclassTime(time):
     sub_var = 1
 
-class TestTime(HarmlessMixedComparison, unittest.TestCase):
+class TestTime(HarmlessMixedComparison, ReduceStructure, unittest.TestCase):
 
     theclass = time
 

From python-checkins at python.org  Thu Jul  1 19:45:52 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Thu,  1 Jul 2010 19:45:52 +0200 (CEST)
Subject: [Python-checkins] r82420 - python/trunk/Doc/library/2to3.rst
Message-ID: <20100701174552.DC93FEEB23@mail.python.org>

Author: benjamin.peterson
Date: Thu Jul  1 19:45:52 2010
New Revision: 82420

Log:
fix fixer name

Modified:
   python/trunk/Doc/library/2to3.rst

Modified: python/trunk/Doc/library/2to3.rst
==============================================================================
--- python/trunk/Doc/library/2to3.rst	(original)
+++ python/trunk/Doc/library/2to3.rst	Thu Jul  1 19:45:52 2010
@@ -276,7 +276,7 @@
 
    Converts the :keyword:`print` statement to the :func:`print` function.
 
-.. 2to3fixer:: raises
+.. 2to3fixer:: raise
 
    Converts ``raise E, V`` to ``raise E(V)``, and ``raise E, V, T`` to ``raise
    E(V).with_traceback(T)``.  If ``E`` is a tuple, the translation will be

From python-checkins at python.org  Thu Jul  1 19:48:17 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Thu,  1 Jul 2010 19:48:17 +0200 (CEST)
Subject: [Python-checkins] r82421 - in python/branches/release26-maint:
	Doc/library/2to3.rst
Message-ID: <20100701174817.2423AEE9F4@mail.python.org>

Author: benjamin.peterson
Date: Thu Jul  1 19:48:16 2010
New Revision: 82421

Log:
Merged revisions 82420 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/trunk

........
  r82420 | benjamin.peterson | 2010-07-01 12:45:52 -0500 (Thu, 01 Jul 2010) | 1 line
  
  fix fixer name
........


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Doc/library/2to3.rst

Modified: python/branches/release26-maint/Doc/library/2to3.rst
==============================================================================
--- python/branches/release26-maint/Doc/library/2to3.rst	(original)
+++ python/branches/release26-maint/Doc/library/2to3.rst	Thu Jul  1 19:48:16 2010
@@ -276,7 +276,7 @@
 
    Converts the :keyword:`print` statement to the :func:`print` function.
 
-.. 2to3fixer:: raises
+.. 2to3fixer:: raise
 
    Converts ``raise E, V`` to ``raise E(V)``, and ``raise E, V, T`` to ``raise
    E(V).with_traceback(T)``.  If ``E`` is a tuple, the translation will be

From python-checkins at python.org  Thu Jul  1 19:49:01 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Thu,  1 Jul 2010 19:49:01 +0200 (CEST)
Subject: [Python-checkins] r82422 - in python/branches/py3k:
	Doc/library/2to3.rst
Message-ID: <20100701174901.61836EEA84@mail.python.org>

Author: benjamin.peterson
Date: Thu Jul  1 19:49:01 2010
New Revision: 82422

Log:
Merged revisions 82420 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/trunk

........
  r82420 | benjamin.peterson | 2010-07-01 12:45:52 -0500 (Thu, 01 Jul 2010) | 1 line
  
  fix fixer name
........


Modified:
   python/branches/py3k/   (props changed)
   python/branches/py3k/Doc/library/2to3.rst

Modified: python/branches/py3k/Doc/library/2to3.rst
==============================================================================
--- python/branches/py3k/Doc/library/2to3.rst	(original)
+++ python/branches/py3k/Doc/library/2to3.rst	Thu Jul  1 19:49:01 2010
@@ -276,7 +276,7 @@
 
    Converts the :keyword:`print` statement to the :func:`print` function.
 
-.. 2to3fixer:: raises
+.. 2to3fixer:: raise
 
    Converts ``raise E, V`` to ``raise E(V)``, and ``raise E, V, T`` to ``raise
    E(V).with_traceback(T)``.  If ``E`` is a tuple, the translation will be

From python-checkins at python.org  Thu Jul  1 19:51:42 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Thu,  1 Jul 2010 19:51:42 +0200 (CEST)
Subject: [Python-checkins] r82423 - in python/branches/release31-maint:
	Doc/library/2to3.rst
Message-ID: <20100701175142.879D4EEA65@mail.python.org>

Author: benjamin.peterson
Date: Thu Jul  1 19:51:42 2010
New Revision: 82423

Log:
Merged revisions 82422 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

................
  r82422 | benjamin.peterson | 2010-07-01 12:49:01 -0500 (Thu, 01 Jul 2010) | 9 lines
  
  Merged revisions 82420 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82420 | benjamin.peterson | 2010-07-01 12:45:52 -0500 (Thu, 01 Jul 2010) | 1 line
    
    fix fixer name
  ........
................


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Doc/library/2to3.rst

Modified: python/branches/release31-maint/Doc/library/2to3.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/2to3.rst	(original)
+++ python/branches/release31-maint/Doc/library/2to3.rst	Thu Jul  1 19:51:42 2010
@@ -276,7 +276,7 @@
 
    Converts the :keyword:`print` statement to the :func:`print` function.
 
-.. 2to3fixer:: raises
+.. 2to3fixer:: raise
 
    Converts ``raise E, V`` to ``raise E(V)``, and ``raise E, V, T`` to ``raise
    E(V).with_traceback(T)``.  If ``E`` is a tuple, the translation will be

From python-checkins at python.org  Thu Jul  1 21:16:58 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Thu,  1 Jul 2010 21:16:58 +0200 (CEST)
Subject: [Python-checkins] r82424 -
	sandbox/branches/py3k-datetime/datetime.py
Message-ID: <20100701191658.3B67CEE9A4@mail.python.org>

Author: alexander.belopolsky
Date: Thu Jul  1 21:16:58 2010
New Revision: 82424

Log:
added __slots__ specifications

Modified:
   sandbox/branches/py3k-datetime/datetime.py

Modified: sandbox/branches/py3k-datetime/datetime.py
==============================================================================
--- sandbox/branches/py3k-datetime/datetime.py	(original)
+++ sandbox/branches/py3k-datetime/datetime.py	Thu Jul  1 21:16:58 2010
@@ -316,6 +316,7 @@
     Representation: (days, seconds, microseconds).  Why?  Because I
     felt like it.
     """
+    __slots__ = '__days', '__seconds', '__microseconds'
 
     def __new__(cls, days=0, seconds=0, microseconds=0,
                 # XXX The following should only be used as keyword args:
@@ -656,6 +657,7 @@
     Properties (readonly):
     year, month, day
     """
+    __slots__ = '__year', '__month', '__day'
 
     def __new__(cls, year, month=None, day=None):
         """Constructor.
@@ -916,7 +918,7 @@
 
     Subclasses must override the name(), utcoffset() and dst() methods.
     """
-
+    __slots__ = ()
     def tzname(self, dt):
         "datetime -> string name of time zone."
         raise NotImplementedError("tzinfo subclass must override tzname()")
@@ -1289,6 +1291,9 @@
     # XXX needs docstrings
     # See http://www.zope.org/Members/fdrake/DateTimeWiki/TimeZoneInfo
 
+    __slots__ = date.__slots__ + (
+        '__hour', '__minute', '__second',
+        '__microsecond', '_tzinfo')
     def __new__(cls, year, month=None, day=None, hour=0, minute=0, second=0,
                 microsecond=0, tzinfo=None):
         if isinstance(year, bytes) and len(year) == 10:
@@ -1765,6 +1770,8 @@
     return week1monday
 
 class timezone(tzinfo):
+    __slots__ = '__offset', '__name'
+
     def __init__(self, offset, *args):
         # Reproduce C behavior
         n = len(args)

From python-checkins at python.org  Thu Jul  1 22:10:47 2010
From: python-checkins at python.org (ezio.melotti)
Date: Thu,  1 Jul 2010 22:10:47 +0200 (CEST)
Subject: [Python-checkins] r82425 - in tracker/instances/python-dev/html:
	issue.item.html issue.search.html page.html
Message-ID: <20100701201047.DA893EE98A@mail.python.org>

Author: ezio.melotti
Date: Thu Jul  1 22:10:47 2010
New Revision: 82425

Log:
#343: Remove the "(list)" for superseder and dependencies.

Modified:
   tracker/instances/python-dev/html/issue.item.html
   tracker/instances/python-dev/html/issue.search.html
   tracker/instances/python-dev/html/page.html

Modified: tracker/instances/python-dev/html/issue.item.html
==============================================================================
--- tracker/instances/python-dev/html/issue.item.html	(original)
+++ tracker/instances/python-dev/html/issue.item.html	Thu Jul  1 22:10:47 2010
@@ -93,8 +93,8 @@
 <tr tal:condition="context/id">
  <th>
   <tal:block i18n:translate="">Dependencies</tal:block>:
-  <span tal:condition="context/dependencies/is_edit_ok"
-        tal:replace="structure python:db.issue.classhelp('id,title,status', filter='status=0,1', property='dependencies')" />
+  <!-- <span tal:condition="context/dependencies/is_edit_ok"
+        tal:replace="structure python:db.issue.classhelp('id,title,status', filter='status=0,1', property='dependencies')" /> -->
  </th>
  <td>
   <span tal:replace="structure python:context.dependencies.field(showid=1,size=20)" />
@@ -104,8 +104,8 @@
  </td>
  <th i18n:translate="">
   <tal:block i18n:translate="">Superseder</tal:block>:
-  <span tal:condition="context/superseder/is_edit_ok"
-        tal:replace="structure python:db.issue.classhelp('id,title', filter='status=0,1', property='superseder')" />
+  <!-- <span tal:condition="context/superseder/is_edit_ok"
+        tal:replace="structure python:db.issue.classhelp('id,title', filter='status=0,1', property='superseder')" /> -->
  </th>
  <td>
   <span tal:replace="structure python:context.superseder.field(showid=1, size=20)" />

Modified: tracker/instances/python-dev/html/issue.search.html
==============================================================================
--- tracker/instances/python-dev/html/issue.search.html	(original)
+++ tracker/instances/python-dev/html/issue.search.html	Thu Jul  1 22:10:47 2010
@@ -42,6 +42,15 @@
   <td>&nbsp;</td>
 </tr>
 
+<tr tal:define="name string:ignore">
+  <th i18n:translate="">Ignore attachments:</th>
+  <td><input type="checkbox" checked="checked"
+             name="ignore" id="ignore" value="file:content"/></td>
+  <td>&nbsp;</td>
+  <td>&nbsp;</td>
+  <td>&nbsp;</td>
+</tr>
+
 <tr tal:define="name string:title">
   <th i18n:translate="">Title:</th>
   <td metal:use-macro="search_input"></td>

Modified: tracker/instances/python-dev/html/page.html
==============================================================================
--- tracker/instances/python-dev/html/page.html	(original)
+++ tracker/instances/python-dev/html/page.html	Thu Jul  1 22:10:47 2010
@@ -42,6 +42,7 @@
        <input type="hidden" name="@sort" value="-activity" />
        <input type="hidden" name="@filter" value="status"/>
        <input type="hidden" name="@action" value="searchid"/>
+       <input type="hidden" name="ignore" value="file:content"/>
        <input class="input-text" id="search-text" name="@search_text" size="10" />
        <input type="submit" id="submit" value="search" name="submit"
               class="input-button"/>
@@ -145,7 +146,8 @@
      })"
        i18n:translate="">Show Having Patch</a></li>
         <li>
-         <a href="issue?@template=search&status=1" i18n:translate="">Search</a>
+         <a href="issue?@template=search&status=1&ignore=file:content"
+            i18n:translate="">Search</a>
         </li>
         <li>
          <form method="post" action="#">

From python-checkins at python.org  Thu Jul  1 22:16:23 2010
From: python-checkins at python.org (ezio.melotti)
Date: Thu,  1 Jul 2010 22:16:23 +0200 (CEST)
Subject: [Python-checkins] r82426 - in tracker/instances/python-dev/html:
	issue.search.html page.html
Message-ID: <20100701201623.64B72EE98A@mail.python.org>

Author: ezio.melotti
Date: Thu Jul  1 22:16:23 2010
New Revision: 82426

Log:
Revert accidental changes in two files that were unrelated.

Modified:
   tracker/instances/python-dev/html/issue.search.html
   tracker/instances/python-dev/html/page.html

Modified: tracker/instances/python-dev/html/issue.search.html
==============================================================================
--- tracker/instances/python-dev/html/issue.search.html	(original)
+++ tracker/instances/python-dev/html/issue.search.html	Thu Jul  1 22:16:23 2010
@@ -42,15 +42,6 @@
   <td>&nbsp;</td>
 </tr>
 
-<tr tal:define="name string:ignore">
-  <th i18n:translate="">Ignore attachments:</th>
-  <td><input type="checkbox" checked="checked"
-             name="ignore" id="ignore" value="file:content"/></td>
-  <td>&nbsp;</td>
-  <td>&nbsp;</td>
-  <td>&nbsp;</td>
-</tr>
-
 <tr tal:define="name string:title">
   <th i18n:translate="">Title:</th>
   <td metal:use-macro="search_input"></td>

Modified: tracker/instances/python-dev/html/page.html
==============================================================================
--- tracker/instances/python-dev/html/page.html	(original)
+++ tracker/instances/python-dev/html/page.html	Thu Jul  1 22:16:23 2010
@@ -42,7 +42,6 @@
        <input type="hidden" name="@sort" value="-activity" />
        <input type="hidden" name="@filter" value="status"/>
        <input type="hidden" name="@action" value="searchid"/>
-       <input type="hidden" name="ignore" value="file:content"/>
        <input class="input-text" id="search-text" name="@search_text" size="10" />
        <input type="submit" id="submit" value="search" name="submit"
               class="input-button"/>
@@ -146,8 +145,7 @@
      })"
        i18n:translate="">Show Having Patch</a></li>
         <li>
-         <a href="issue?@template=search&status=1&ignore=file:content"
-            i18n:translate="">Search</a>
+         <a href="issue?@template=search&status=1" i18n:translate="">Search</a>
         </li>
         <li>
          <form method="post" action="#">

From python-checkins at python.org  Thu Jul  1 22:23:17 2010
From: python-checkins at python.org (ezio.melotti)
Date: Thu,  1 Jul 2010 22:23:17 +0200 (CEST)
Subject: [Python-checkins] r82427 -
	tracker/instances/python-dev/html/issue.item.html
Message-ID: <20100701202317.8EDC5D2CC@mail.python.org>

Author: ezio.melotti
Date: Thu Jul  1 22:23:17 2010
New Revision: 82427

Log:
#251: Remove the nosy count from the issue page.

Modified:
   tracker/instances/python-dev/html/issue.item.html

Modified: tracker/instances/python-dev/html/issue.item.html
==============================================================================
--- tracker/instances/python-dev/html/issue.item.html	(original)
+++ tracker/instances/python-dev/html/issue.item.html	Thu Jul  1 22:23:17 2010
@@ -135,7 +135,7 @@
  <td tal:condition="not:context/assignee/is_edit_ok">
   <span tal:replace="structure context/assignee/plain" />
  </td>
- <th><tal:block i18n:translate="">Nosy List</tal:block><span tal:condition="context/nosy_count" tal:replace="python: ' (%d)' % context.nosy_count" />:
+ <th><tal:block i18n:translate="">Nosy List</tal:block><!-- <span tal:condition="context/nosy_count" tal:replace="python: ' (%d)' % context.nosy_count" /> -->:
   <span tal:condition="context/nosy/is_edit_ok"
         tal:replace="structure python:db.user.classhelp('username,realname,address', property='nosy')" />
  </th>

From python-checkins at python.org  Thu Jul  1 22:26:54 2010
From: python-checkins at python.org (ezio.melotti)
Date: Thu,  1 Jul 2010 22:26:54 +0200 (CEST)
Subject: [Python-checkins] r82428 -
	tracker/instances/python-dev/html/issue.item.html
Message-ID: <20100701202654.A3D90EDB4@mail.python.org>

Author: ezio.melotti
Date: Thu Jul  1 22:26:54 2010
New Revision: 82428

Log:
Remove trailing whitespace and add a missing colon after "Keywords".

Modified:
   tracker/instances/python-dev/html/issue.item.html

Modified: tracker/instances/python-dev/html/issue.item.html
==============================================================================
--- tracker/instances/python-dev/html/issue.item.html	(original)
+++ tracker/instances/python-dev/html/issue.item.html	Thu Jul  1 22:26:54 2010
@@ -116,8 +116,8 @@
         tal:attributes="href string:issue${sup/id}; title sup/title;"></a> -->
    <br><span i18n:translate="">View</span>:
      <a tal:content="context/superseder/id"
-        tal:attributes="href string:issue${context/superseder/id}; title context/superseder/title;"></a> 
-  </span> 
+        tal:attributes="href string:issue${context/superseder/id}; title context/superseder/title;"></a>
+  </span>
  </td>
  </tr>
  <tr>
@@ -126,7 +126,7 @@
   <select name="assignee">
    <option value="-1">nobody</option>
    <tal:block tal:repeat="userdata python:db._db.user.filter_sql('select id,_username from _user where _roles like \'%Developer%\' order by _username')">
-    <option tal:attributes="value python:userdata[0]; 
+    <option tal:attributes="value python:userdata[0];
                             selected python:str(userdata[0]) == context.assignee._value"
             tal:content="python:userdata[1]"></option>
    </tal:block>
@@ -153,7 +153,7 @@
    <span tal:replace="structure python:db.priority.classhelp('id,name,description',label='Priority')" />:
  </th>
  <td tal:content="structure context/priority/menu">priority</td>
- <th tal:content="structure python:db.keyword.classhelp('id,name,description',label='Keywords')" />
+ <th tal:content="structure python:db.keyword.classhelp('id,name,description',label='Keywords:')" />
  <td tal:content="structure python:context['keywords'].menu(height=5)">keywords</td>
 
 
@@ -242,7 +242,7 @@
  <tal:block tal:repeat="msg python:context.messages.sorted('creation')">
   <tr>
     <th tal:define="mid string:msg${msg/id}">
-     <a tal:content="mid" tal:attributes="id mid; 
+     <a tal:content="mid" tal:attributes="id mid;
         href string:#$mid;"></a> - <a tal:attributes="href mid">(view)</a></th>
    <th i18n:translate="">Author: <tal:x replace="python:msg.author.realname.plain(unchecked=1)"
        i18n:name="author" /> (<tal:x replace="msg/author"/>)

From solipsis at pitrou.net  Fri Jul  2 01:24:57 2010
From: solipsis at pitrou.net (solipsis at pitrou.net)
Date: Fri,  2 Jul 2010 01:24:57 +0200 (CEST)
Subject: [Python-checkins] Daily py3k reference leaks (r82422): sum=13614
Message-ID: <20100701232457.799371770A@ns6635.ovh.net>

py3k results for svn r82422 (hg cset 68d89477054a)
--------------------------------------------------

test_decimal leaked [4104, 4104, 4104] references, sum=12312
test_descr leaked [280, 280, 280] references, sum=840
test_fractions leaked [70, 70, 70] references, sum=210
test_math leaked [84, 84, 84] references, sum=252


Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/antoine/py3k/refleaks/refloggTsoEh', '-x']

From python-checkins at python.org  Fri Jul  2 01:35:37 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Fri,  2 Jul 2010 01:35:37 +0200 (CEST)
Subject: [Python-checkins] r82429 - python/trunk/Doc/library/math.rst
Message-ID: <20100701233537.89B7EC35B@mail.python.org>

Author: benjamin.peterson
Date: Fri Jul  2 01:35:37 2010
New Revision: 82429

Log:
trunc does use the special method, though

Modified:
   python/trunk/Doc/library/math.rst

Modified: python/trunk/Doc/library/math.rst
==============================================================================
--- python/trunk/Doc/library/math.rst	(original)
+++ python/trunk/Doc/library/math.rst	Fri Jul  2 01:35:37 2010
@@ -134,7 +134,7 @@
 .. function:: trunc(x)
 
    Return the :class:`Real` value *x* truncated to an :class:`Integral` (usually
-   a long integer).
+   a long integer).  Uses the ``__trunc__`` method.
 
    .. versionadded:: 2.6
 

From python-checkins at python.org  Fri Jul  2 01:39:27 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Fri,  2 Jul 2010 01:39:27 +0200 (CEST)
Subject: [Python-checkins] r82430 - python/branches/py3k
Message-ID: <20100701233927.06334EE982@mail.python.org>

Author: benjamin.peterson
Date: Fri Jul  2 01:39:26 2010
New Revision: 82430

Log:
Blocked revisions 82429 via svnmerge

........
  r82429 | benjamin.peterson | 2010-07-01 18:35:37 -0500 (Thu, 01 Jul 2010) | 1 line
  
  trunc does use the special method, though
........


Modified:
   python/branches/py3k/   (props changed)

From python-checkins at python.org  Fri Jul  2 01:41:09 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Fri,  2 Jul 2010 01:41:09 +0200 (CEST)
Subject: [Python-checkins] r82431 - in python/branches/release26-maint:
	Doc/library/math.rst
Message-ID: <20100701234109.7F291EE988@mail.python.org>

Author: benjamin.peterson
Date: Fri Jul  2 01:41:09 2010
New Revision: 82431

Log:
Merged revisions 82429 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/trunk

........
  r82429 | benjamin.peterson | 2010-07-01 18:35:37 -0500 (Thu, 01 Jul 2010) | 1 line
  
  trunc does use the special method, though
........


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Doc/library/math.rst

Modified: python/branches/release26-maint/Doc/library/math.rst
==============================================================================
--- python/branches/release26-maint/Doc/library/math.rst	(original)
+++ python/branches/release26-maint/Doc/library/math.rst	Fri Jul  2 01:41:09 2010
@@ -134,7 +134,7 @@
 .. function:: trunc(x)
 
    Return the :class:`Real` value *x* truncated to an :class:`Integral` (usually
-   a long integer).
+   a long integer).  Uses the ``__trunc__`` method.
 
    .. versionadded:: 2.6
 

From python-checkins at python.org  Fri Jul  2 09:33:50 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri,  2 Jul 2010 09:33:50 +0200 (CEST)
Subject: [Python-checkins] r82433 -
	python/trunk/Doc/library/simplehttpserver.rst
Message-ID: <20100702073350.3DEA7EE999@mail.python.org>

Author: georg.brandl
Date: Fri Jul  2 09:33:50 2010
New Revision: 82433

Log:
Grammar and markup fixes.

Modified:
   python/trunk/Doc/library/simplehttpserver.rst

Modified: python/trunk/Doc/library/simplehttpserver.rst
==============================================================================
--- python/trunk/Doc/library/simplehttpserver.rst	(original)
+++ python/trunk/Doc/library/simplehttpserver.rst	Fri Jul  2 09:33:50 2010
@@ -82,34 +82,35 @@
       ``text/`` the file is opened in text mode; otherwise binary mode is used.
 
       The :func:`test` function in the :mod:`SimpleHTTPServer` module is an
-      example which creates a server using the
-      :class:`SimpleHTTPRequestHandler` as the Handler.
+      example which creates a server using the :class:`SimpleHTTPRequestHandler`
+      as the Handler.
 
       .. versionadded:: 2.5
          The ``'Last-Modified'`` header.
 
 
-The :mod:`SimpleHTTPServer` module can be used the following manner in order to
-set up a very basic web server serving files relative to the current
-directory.::
+The :mod:`SimpleHTTPServer` module can be used in the following manner in order
+to set up a very basic web server serving files relative to the current
+directory. ::
 
-        import SimpleHTTPServer
-        import SocketServer
+   import SimpleHTTPServer
+   import SocketServer
 
-        PORT = 8000
+   PORT = 8000
 
-        Handler = SimpleHTTPServer.SimpleHTTPRequestHandler
+   Handler = SimpleHTTPServer.SimpleHTTPRequestHandler
 
-        httpd = SocketServer.TCPServer(("", PORT), Handler)
+   httpd = SocketServer.TCPServer(("", PORT), Handler)
 
-        print "serving at port", PORT
-        httpd.serve_forever()
+   print "serving at port", PORT
+   httpd.serve_forever()
 
-:mod:`SimpleHTTPServer` module can also be invoked directly using the ``-m``
-switch of interpreter with a ``port number`` argument. Similar to previous
-example, this serves the files relative to the current directory.::
+The :mod:`SimpleHTTPServer` module can also be invoked directly using the
+:cmdoption:`-m` switch of the interpreter with a ``port number`` argument.
+Similar to the previous example, this serves the files relative to the
+current directory. ::
 
-        python -m SimpleHTTPServer 8000
+   python -m SimpleHTTPServer 8000
 
 .. seealso::
 

From python-checkins at python.org  Fri Jul  2 09:41:51 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri,  2 Jul 2010 09:41:51 +0200 (CEST)
Subject: [Python-checkins] r82434 - in python/branches/py3k:
	Doc/library/http.server.rst
Message-ID: <20100702074151.A97FFEE9BD@mail.python.org>

Author: georg.brandl
Date: Fri Jul  2 09:41:51 2010
New Revision: 82434

Log:
Merged revisions 82433 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/trunk

........
  r82433 | georg.brandl | 2010-07-02 09:33:50 +0200 (Fr, 02 Jul 2010) | 1 line
  
  Grammar and markup fixes.
........


Modified:
   python/branches/py3k/   (props changed)
   python/branches/py3k/Doc/library/http.server.rst

Modified: python/branches/py3k/Doc/library/http.server.rst
==============================================================================
--- python/branches/py3k/Doc/library/http.server.rst	(original)
+++ python/branches/py3k/Doc/library/http.server.rst	Fri Jul  2 09:41:51 2010
@@ -284,30 +284,30 @@
       For example usage, see the implementation of the :func:`test` function
       invocation in the :mod:`http.server` module.
 
-The :class:`SimpleHTTPRequestHandler` class can be invoked the following manner
-with the :mod:`http.server` to create a very basic webserver serving files
-relative to the current directory.::
 
-        import http.server
-        import socketserver
+The :class:`SimpleHTTPRequestHandler` class can be used in the following
+manner in order to create a very basic webserver serving files relative to
+the current directory. ::
 
-        PORT = 8000
+   import http.server
+   import socketserver
 
-        Handler = http.server.SimpleHTTPRequestHandler
+   PORT = 8000
 
-        httpd = socketserver.TCPServer(("", PORT), Handler)
+   Handler = http.server.SimpleHTTPRequestHandler
 
-        print("serving at port", PORT)
-        httpd.serve_forever()
+   httpd = socketserver.TCPServer(("", PORT), Handler)
 
-:mod:`http.server` can also be invoked directly using the ``-m`` switch of
-interpreter a with ``port number`` argument which uses
-:class:`SimpleHTTPRequestHandler` as the default request Handler. Similar to
-the previous example, even this serves files relative to the current
-directory.::
+   print("serving at port", PORT)
+   httpd.serve_forever()
+
+:mod:`http.server` can also be invoked directly using the :cmdoption:`-m`
+switch of the interpreter a with ``port number`` argument.  Similar to
+the previous example, this serves files relative to the current directory. ::
 
         python -m http.server 8000
 
+
 .. class:: CGIHTTPRequestHandler(request, client_address, server)
 
    This class is used to serve either files or output of CGI scripts from the

From python-checkins at python.org  Fri Jul  2 10:51:24 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri,  2 Jul 2010 10:51:24 +0200 (CEST)
Subject: [Python-checkins] r82435 - peps/trunk/pep-0392.txt
Message-ID: <20100702085124.E122EEE9CD@mail.python.org>

Author: georg.brandl
Date: Fri Jul  2 10:51:24 2010
New Revision: 82435

Log:
New tentative schedule.

Modified:
   peps/trunk/pep-0392.txt

Modified: peps/trunk/pep-0392.txt
==============================================================================
--- peps/trunk/pep-0392.txt	(original)
+++ peps/trunk/pep-0392.txt	Fri Jul  2 10:51:24 2010
@@ -17,7 +17,7 @@
 Python 3.2.  The schedule primarily concerns itself with PEP-sized
 items.  Small features may be added up to and including the first beta
 release.  Bugs may be fixed until the final release, which is planned
-for December 2010.
+for January 2011.
 
 
 Release Manager and Crew
@@ -34,17 +34,17 @@
 
 The current schedule is:
 
-- 3.2 alpha 1: July 3, 2010
-- 3.2 alpha 2: July 24, 2010
-- 3.2 alpha 3: August 21, 2010
-- 3.2 beta 1: September 18, 2010
+- 3.2 alpha 1: August 7, 2010
+- 3.2 alpha 2: September 4, 2010
+- 3.2 alpha 3: October 2, 2010
+- 3.2 beta 1: October 30, 2010
 
 (No new features beyond this point.)
 
-- 3.2 beta 2: October 16, 2010
-- 3.2 candidate 1: November 13, 2010
-- 3.2 candidate 2: November 27, 2010
-- 3.2 final: December 11, 2010
+- 3.2 beta 2: November 27, 2010
+- 3.2 candidate 1: December 11, 2010
+- 3.2 candidate 2: January 1, 2010
+- 3.2 final: January 15, 2010
 
 .. don't forget to update final date above as well
 

From python-checkins at python.org  Fri Jul  2 15:35:17 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Fri,  2 Jul 2010 15:35:17 +0200 (CEST)
Subject: [Python-checkins] r82437 - python/branches/py3k/Modules/mathmodule.c
Message-ID: <20100702133517.B3EFECA6A@mail.python.org>

Author: benjamin.peterson
Date: Fri Jul  2 15:35:17 2010
New Revision: 82437

Log:
account for different ref counting semantics of _PyObject_LookupSpecial

Modified:
   python/branches/py3k/Modules/mathmodule.c

Modified: python/branches/py3k/Modules/mathmodule.c
==============================================================================
--- python/branches/py3k/Modules/mathmodule.c	(original)
+++ python/branches/py3k/Modules/mathmodule.c	Fri Jul  2 15:35:17 2010
@@ -881,7 +881,7 @@
 
 static PyObject * math_floor(PyObject *self, PyObject *number) {
     static PyObject *floor_str = NULL;
-    PyObject *method;
+    PyObject *method, *result;
 
     method = _PyObject_LookupSpecial(number, "__floor__", &floor_str);
     if (method == NULL) {
@@ -889,7 +889,9 @@
             return NULL;
         return math_1_to_int(number, floor, 0);
     }
-    return PyObject_CallFunctionObjArgs(method, NULL);
+    result = PyObject_CallFunctionObjArgs(method, NULL);
+    Py_DECREF(method);
+    return result;
 }
 
 PyDoc_STRVAR(math_floor_doc,
@@ -1416,7 +1418,7 @@
 math_trunc(PyObject *self, PyObject *number)
 {
     static PyObject *trunc_str = NULL;
-    PyObject *trunc;
+    PyObject *trunc, *result;
 
     if (Py_TYPE(number)->tp_dict == NULL) {
         if (PyType_Ready(Py_TYPE(number)) < 0)
@@ -1431,7 +1433,9 @@
                          Py_TYPE(number)->tp_name);
         return NULL;
     }
-    return PyObject_CallFunctionObjArgs(trunc, NULL);
+    result = PyObject_CallFunctionObjArgs(trunc, NULL);
+    Py_DECREF(trunc);
+    return result;
 }
 
 PyDoc_STRVAR(math_trunc_doc,

From python-checkins at python.org  Fri Jul  2 15:46:42 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Fri,  2 Jul 2010 15:46:42 +0200 (CEST)
Subject: [Python-checkins] r82438 - in python/branches/py3k:
	Lib/test/test_descr.py Misc/NEWS Modules/mathmodule.c
Message-ID: <20100702134642.CA809EE982@mail.python.org>

Author: benjamin.peterson
Date: Fri Jul  2 15:46:42 2010
New Revision: 82438

Log:
fix lookup of __ceil__

Modified:
   python/branches/py3k/Lib/test/test_descr.py
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/Modules/mathmodule.c

Modified: python/branches/py3k/Lib/test/test_descr.py
==============================================================================
--- python/branches/py3k/Lib/test/test_descr.py	(original)
+++ python/branches/py3k/Lib/test/test_descr.py	Fri Jul  2 15:46:42 2010
@@ -1581,6 +1581,7 @@
             ("__format__", format, format_impl, set(), {}),
             ("__floor__", math.floor, zero, set(), {}),
             ("__trunc__", math.trunc, zero, set(), {}),
+            ("__ceil__", math.ceil, zero, set(), {}),
             ]
 
         class Checker(object):

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Fri Jul  2 15:46:42 2010
@@ -1374,8 +1374,8 @@
 Extension Modules
 -----------------
 
-- In the math module, correctly lookup __trunc__ and __floor__ as special
-  methods.
+- In the math module, correctly lookup __trunc__, __ceil__, and __floor__ as
+  special methods.
 
 - Issue #9005: Prevent utctimetuple() from producing year 0 or year
   10,000.  Prior to this change, timezone adjustment in utctimetuple()

Modified: python/branches/py3k/Modules/mathmodule.c
==============================================================================
--- python/branches/py3k/Modules/mathmodule.c	(original)
+++ python/branches/py3k/Modules/mathmodule.c	Fri Jul  2 15:46:42 2010
@@ -843,17 +843,17 @@
     static PyObject *ceil_str = NULL;
     PyObject *method;
 
-    if (ceil_str == NULL) {
-        ceil_str = PyUnicode_InternFromString("__ceil__");
-        if (ceil_str == NULL)
+    method = _PyObject_LookupSpecial(number, "__ceil__", &ceil_str);
+    if (method == NULL) {
+        if (PyErr_Occurred())
             return NULL;
-    }
-
-    method = _PyType_Lookup(Py_TYPE(number), ceil_str);
-    if (method == NULL)
         return math_1_to_int(number, ceil, 0);
-    else
-        return PyObject_CallFunction(method, "O", number);
+    }
+    else {
+        PyObject *result = PyObject_CallFunctionObjArgs(method, NULL);
+        Py_DECREF(method);
+        return result;
+    }
 }
 
 PyDoc_STRVAR(math_ceil_doc,

From python-checkins at python.org  Fri Jul  2 17:50:14 2010
From: python-checkins at python.org (nick.coghlan)
Date: Fri,  2 Jul 2010 17:50:14 +0200 (CEST)
Subject: [Python-checkins] r82439 - python/trunk/Doc/whatsnew/2.7.rst
Message-ID: <20100702155014.8E2D9EE99E@mail.python.org>

Author: nick.coghlan
Date: Fri Jul  2 17:50:14 2010
New Revision: 82439

Log:
Fix a couple of minor nits in What's New before the release goes out

Modified:
   python/trunk/Doc/whatsnew/2.7.rst

Modified: python/trunk/Doc/whatsnew/2.7.rst
==============================================================================
--- python/trunk/Doc/whatsnew/2.7.rst	(original)
+++ python/trunk/Doc/whatsnew/2.7.rst	Fri Jul  2 17:50:14 2010
@@ -861,10 +861,10 @@
 
 When running a module using the interpreter's :option:`-m` switch,
 ``sys.argv[0]`` will now be set to the string ``'-m'`` while the
-module is being located, while executing the :file:`__init__.py` files
-for any parent packages of the module to be executed.
-(Suggested by Michael Foord; implemented by Nick Coghlan;
-:issue:`8202`.)
+module is being located.  This change is visible to user code while
+executing the :file:`__init__.py` files for any parent packages of
+the module to be executed.  (Suggested by Michael Foord; implemented
+by Nick Coghlan; :issue:`8202`.)
 
 .. ======================================================================
 
@@ -1420,9 +1420,9 @@
   ``sys.path`` and the module :mod:`__main__` will be imported.  It's
   expected that the directory or zip contains a :file:`__main__.py`;
   if it doesn't, some other :file:`__main__.py` might be imported from
-  a location later in ``sys.path``.  This makes some of the machinery
+  a location later in ``sys.path``.  This makes more of the machinery
   of :mod:`runpy` available to scripts that want to mimic the way
-  Python's :option:`-m` processes an explicit path name.
+  Python's command line processes an explicit path name.
   (Added by Nick Coghlan; :issue:`6816`.)
 
 * New function: in the :mod:`shutil` module, :func:`~shutil.make_archive`

From python-checkins at python.org  Fri Jul  2 17:57:50 2010
From: python-checkins at python.org (nick.coghlan)
Date: Fri,  2 Jul 2010 17:57:50 +0200 (CEST)
Subject: [Python-checkins] r82440 - in python/branches/py3k:
	Doc/whatsnew/2.7.rst
Message-ID: <20100702155750.F1CBCEE98B@mail.python.org>

Author: nick.coghlan
Date: Fri Jul  2 17:57:50 2010
New Revision: 82440

Log:
Merged revisions 82439 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/trunk

........
  r82439 | nick.coghlan | 2010-07-03 01:50:14 +1000 (Sat, 03 Jul 2010) | 1 line
  
  Fix a couple of minor nits in What's New before the release goes out
........


Modified:
   python/branches/py3k/   (props changed)
   python/branches/py3k/Doc/whatsnew/2.7.rst

Modified: python/branches/py3k/Doc/whatsnew/2.7.rst
==============================================================================
--- python/branches/py3k/Doc/whatsnew/2.7.rst	(original)
+++ python/branches/py3k/Doc/whatsnew/2.7.rst	Fri Jul  2 17:57:50 2010
@@ -861,10 +861,10 @@
 
 When running a module using the interpreter's :option:`-m` switch,
 ``sys.argv[0]`` will now be set to the string ``'-m'`` while the
-module is being located, while executing the :file:`__init__.py` files
-for any parent packages of the module to be executed.
-(Suggested by Michael Foord; implemented by Nick Coghlan;
-:issue:`8202`.)
+module is being located.  This change is visible to user code while
+executing the :file:`__init__.py` files for any parent packages of
+the module to be executed.  (Suggested by Michael Foord; implemented
+by Nick Coghlan; :issue:`8202`.)
 
 .. ======================================================================
 
@@ -1420,9 +1420,9 @@
   ``sys.path`` and the module :mod:`__main__` will be imported.  It's
   expected that the directory or zip contains a :file:`__main__.py`;
   if it doesn't, some other :file:`__main__.py` might be imported from
-  a location later in ``sys.path``.  This makes some of the machinery
+  a location later in ``sys.path``.  This makes more of the machinery
   of :mod:`runpy` available to scripts that want to mimic the way
-  Python's :option:`-m` processes an explicit path name.
+  Python's command line processes an explicit path name.
   (Added by Nick Coghlan; :issue:`6816`.)
 
 * New function: in the :mod:`shutil` module, :func:`~shutil.make_archive`

From python-checkins at python.org  Fri Jul  2 18:01:53 2010
From: python-checkins at python.org (nick.coghlan)
Date: Fri,  2 Jul 2010 18:01:53 +0200 (CEST)
Subject: [Python-checkins] r82441 - python/trunk/Doc/whatsnew/2.7.rst
Message-ID: <20100702160153.75707CA5E@mail.python.org>

Author: nick.coghlan
Date: Fri Jul  2 18:01:53 2010
New Revision: 82441

Log:
Issue 8202 actually caused an unintended behavioural change and was reverted in 2.7 - remove reference from What's New

Modified:
   python/trunk/Doc/whatsnew/2.7.rst

Modified: python/trunk/Doc/whatsnew/2.7.rst
==============================================================================
--- python/trunk/Doc/whatsnew/2.7.rst	(original)
+++ python/trunk/Doc/whatsnew/2.7.rst	Fri Jul  2 18:01:53 2010
@@ -859,13 +859,6 @@
 
   export PYTHONWARNINGS=all,error:::Cookie:0
 
-When running a module using the interpreter's :option:`-m` switch,
-``sys.argv[0]`` will now be set to the string ``'-m'`` while the
-module is being located.  This change is visible to user code while
-executing the :file:`__init__.py` files for any parent packages of
-the module to be executed.  (Suggested by Michael Foord; implemented
-by Nick Coghlan; :issue:`8202`.)
-
 .. ======================================================================
 
 

From python-checkins at python.org  Fri Jul  2 18:05:16 2010
From: python-checkins at python.org (mark.dickinson)
Date: Fri,  2 Jul 2010 18:05:16 +0200 (CEST)
Subject: [Python-checkins] r82442 - python/branches/py3k/Modules/mathmodule.c
Message-ID: <20100702160516.11BB8CA3F@mail.python.org>

Author: mark.dickinson
Date: Fri Jul  2 18:05:15 2010
New Revision: 82442

Log:
Style/consistency nit: make math_floor and math_ceil code look the same.

Modified:
   python/branches/py3k/Modules/mathmodule.c

Modified: python/branches/py3k/Modules/mathmodule.c
==============================================================================
--- python/branches/py3k/Modules/mathmodule.c	(original)
+++ python/branches/py3k/Modules/mathmodule.c	Fri Jul  2 18:05:15 2010
@@ -841,7 +841,7 @@
 
 static PyObject * math_ceil(PyObject *self, PyObject *number) {
     static PyObject *ceil_str = NULL;
-    PyObject *method;
+    PyObject *method, *result;
 
     method = _PyObject_LookupSpecial(number, "__ceil__", &ceil_str);
     if (method == NULL) {
@@ -849,11 +849,9 @@
             return NULL;
         return math_1_to_int(number, ceil, 0);
     }
-    else {
-        PyObject *result = PyObject_CallFunctionObjArgs(method, NULL);
-        Py_DECREF(method);
-        return result;
-    }
+    result = PyObject_CallFunctionObjArgs(method, NULL);
+    Py_DECREF(method);
+    return result;
 }
 
 PyDoc_STRVAR(math_ceil_doc,

From python-checkins at python.org  Fri Jul  2 18:14:06 2010
From: python-checkins at python.org (nick.coghlan)
Date: Fri,  2 Jul 2010 18:14:06 +0200 (CEST)
Subject: [Python-checkins] r82443 - in python/branches/py3k:
	Doc/whatsnew/2.7.rst
Message-ID: <20100702161406.0D111EE982@mail.python.org>

Author: nick.coghlan
Date: Fri Jul  2 18:14:05 2010
New Revision: 82443

Log:
Merged revisions 82441 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/trunk

........
  r82441 | nick.coghlan | 2010-07-03 02:01:53 +1000 (Sat, 03 Jul 2010) | 1 line
  
  Issue 8202 actually caused an unintended behavioural change and was reverted in 2.7 - remove reference from What's New
........


Modified:
   python/branches/py3k/   (props changed)
   python/branches/py3k/Doc/whatsnew/2.7.rst

Modified: python/branches/py3k/Doc/whatsnew/2.7.rst
==============================================================================
--- python/branches/py3k/Doc/whatsnew/2.7.rst	(original)
+++ python/branches/py3k/Doc/whatsnew/2.7.rst	Fri Jul  2 18:14:05 2010
@@ -859,13 +859,6 @@
 
   export PYTHONWARNINGS=all,error:::Cookie:0
 
-When running a module using the interpreter's :option:`-m` switch,
-``sys.argv[0]`` will now be set to the string ``'-m'`` while the
-module is being located.  This change is visible to user code while
-executing the :file:`__init__.py` files for any parent packages of
-the module to be executed.  (Suggested by Michael Foord; implemented
-by Nick Coghlan; :issue:`8202`.)
-
 .. ======================================================================
 
 

From python-checkins at python.org  Fri Jul  2 18:17:17 2010
From: python-checkins at python.org (nick.coghlan)
Date: Fri,  2 Jul 2010 18:17:17 +0200 (CEST)
Subject: [Python-checkins] r82444 - python/branches/release26-maint
Message-ID: <20100702161717.65CF7CA5E@mail.python.org>

Author: nick.coghlan
Date: Fri Jul  2 18:17:17 2010
New Revision: 82444

Log:
Blocked revisions 82439,82441 via svnmerge

........
  r82439 | nick.coghlan | 2010-07-03 01:50:14 +1000 (Sat, 03 Jul 2010) | 1 line
  
  Fix a couple of minor nits in What's New before the release goes out
........
  r82441 | nick.coghlan | 2010-07-03 02:01:53 +1000 (Sat, 03 Jul 2010) | 1 line
  
  Issue 8202 actually caused an unintended behavioural change and was reverted in 2.7 - remove reference from What's New
........


Modified:
   python/branches/release26-maint/   (props changed)

From python-checkins at python.org  Fri Jul  2 18:21:37 2010
From: python-checkins at python.org (nick.coghlan)
Date: Fri,  2 Jul 2010 18:21:37 +0200 (CEST)
Subject: [Python-checkins] r82445 - python/branches/release31-maint
Message-ID: <20100702162137.736B9CA3F@mail.python.org>

Author: nick.coghlan
Date: Fri Jul  2 18:21:37 2010
New Revision: 82445

Log:
Blocked revisions 82440,82443 via svnmerge

................
  r82440 | nick.coghlan | 2010-07-03 01:57:50 +1000 (Sat, 03 Jul 2010) | 9 lines
  
  Merged revisions 82439 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82439 | nick.coghlan | 2010-07-03 01:50:14 +1000 (Sat, 03 Jul 2010) | 1 line
    
    Fix a couple of minor nits in What's New before the release goes out
  ........
................
  r82443 | nick.coghlan | 2010-07-03 02:14:05 +1000 (Sat, 03 Jul 2010) | 9 lines
  
  Merged revisions 82441 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82441 | nick.coghlan | 2010-07-03 02:01:53 +1000 (Sat, 03 Jul 2010) | 1 line
    
    Issue 8202 actually caused an unintended behavioural change and was reverted in 2.7 - remove reference from What's New
  ........
................


Modified:
   python/branches/release31-maint/   (props changed)

From ncoghlan at gmail.com  Fri Jul  2 18:39:26 2010
From: ncoghlan at gmail.com (Nick Coghlan)
Date: Sat, 3 Jul 2010 02:39:26 +1000
Subject: [Python-checkins] r82348 - in
	tracker/instances/python-dev/html: issue.item.html issue.item.js
In-Reply-To: <20100628212916.C0CE0EEC4C@mail.python.org>
References: <20100628212916.C0CE0EEC4C@mail.python.org>
Message-ID: <AANLkTikqyS04AZXgunChwbYn0D6D5Cv2tnBrr3gzb0ds@mail.gmail.com>

On Tue, Jun 29, 2010 at 7:29 AM, ezio.melotti
<python-checkins at python.org> wrote:
> Author: ezio.melotti
> Date: Mon Jun 28 23:29:16 2010
> New Revision: 82348
>
> Log:
> #266: Add an "Add me to nosy" button next to the nosy list.

Thank you, thank you, thank you, thank you, etc... :)

Cheers,
Nick.

-- 
Nick Coghlan   |   ncoghlan at gmail.com   |   Brisbane, Australia

From python-checkins at python.org  Fri Jul  2 20:06:52 2010
From: python-checkins at python.org (mark.dickinson)
Date: Fri,  2 Jul 2010 20:06:52 +0200 (CEST)
Subject: [Python-checkins] r82446 - python/trunk/Doc/library/sys.rst
Message-ID: <20100702180652.72B68F094@mail.python.org>

Author: mark.dickinson
Date: Fri Jul  2 20:06:52 2010
New Revision: 82446

Log:
Clarify sys.float_info documentation.

Modified:
   python/trunk/Doc/library/sys.rst

Modified: python/trunk/Doc/library/sys.rst
==============================================================================
--- python/trunk/Doc/library/sys.rst	(original)
+++ python/trunk/Doc/library/sys.rst	Fri Jul  2 20:06:52 2010
@@ -296,43 +296,65 @@
 .. data:: float_info
 
    A structseq holding information about the float type. It contains low level
-   information about the precision and internal representation. Please study
-   your system's :file:`float.h` for more information.
+   information about the precision and internal representation.  The values
+   correspond to the various floating-point constants defined in the standard
+   header file :file:`float.h` for the 'C' programming language; see section
+   5.2.4.2.2 of the 1999 ISO/IEC C standard [C99]_, 'Characteristics of
+   floating types', for details.
+
+   +---------------------+----------------+--------------------------------------------------+
+   | attribute           | float.h macro  | explanation                                      |
+   +=====================+================+==================================================+
+   | :const:`epsilon`    | DBL_MAX        | difference between 1 and the least value greater |
+   |                     |                | than 1 that is representable as a float          |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`dig`        | DBL_DIG        | maximum number of decimal digits that can be     |
+   |                     |                | faithfully represented in a float;  see below    |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`mant_dig`   | DBL_MANT_DIG   | float precision: the number of base-``radix``    |
+   |                     |                | digits in the significand of a float             |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`max`        | DBL_MAX        | maximum representable finite float               |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`max_exp`    | DBL_MAX_EXP    | maximum integer e such that ``radix**(e-1)`` is  |
+   |                     |                | a representable finite float                     |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`max_10_exp` | DBL_MAX_10_EXP | maximum integer e such that ``10**e`` is in the  |
+   |                     |                | range of representable finite floats             |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`min`        | DBL_MIN        | minimum positive normalized float                |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`min_exp`    | DBL_MIN_EXP    | minimum integer e such that ``radix**(e-1)`` is  |
+   |                     |                | a normalized float                               |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`min_10_exp` | DBL_MIN_10_EXP | minimum integer e such that ``10**e`` is a       |
+   |                     |                | normalized float                                 |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`radix`      | FLT_RADIX      | radix of exponent representation                 |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`rounds`     | FLT_ROUNDS     | constant representing rounding mode              |
+   |                     |                | used for arithmetic operations                   |
+   +---------------------+----------------+--------------------------------------------------+
+
+   The attribute :attr:`sys.float_info.dig` needs further explanation.  If
+   ``s`` is any string representing a decimal number with at most
+   :attr:`sys.float_info.dig` significant digits, then converting ``s`` to a
+   float and back again will recover a string representing the same decimal
+   value::
 
-   +---------------------+--------------------------------------------------+
-   | attribute           |  explanation                                     |
-   +=====================+==================================================+
-   | :const:`epsilon`    | Difference between 1 and the next representable  |
-   |                     | floating point number                            |
-   +---------------------+--------------------------------------------------+
-   | :const:`dig`        | digits (see :file:`float.h`)                     |
-   +---------------------+--------------------------------------------------+
-   | :const:`mant_dig`   | mantissa digits (see :file:`float.h`)            |
-   +---------------------+--------------------------------------------------+
-   | :const:`max`        | maximum representable finite float               |
-   +---------------------+--------------------------------------------------+
-   | :const:`max_exp`    | maximum int e such that radix**(e-1) is in the   |
-   |                     | range of finite representable floats             |
-   +---------------------+--------------------------------------------------+
-   | :const:`max_10_exp` | maximum int e such that 10**e is in the          |
-   |                     | range of finite representable floats             |
-   +---------------------+--------------------------------------------------+
-   | :const:`min`        | Minimum positive normalizer float                |
-   +---------------------+--------------------------------------------------+
-   | :const:`min_exp`    | minimum int e such that radix**(e-1) is a        |
-   |                     | normalized float                                 |
-   +---------------------+--------------------------------------------------+
-   | :const:`min_10_exp` | minimum int e such that 10**e is a normalized    |
-   |                     | float                                            |
-   +---------------------+--------------------------------------------------+
-   | :const:`radix`      | radix of exponent                                |
-   +---------------------+--------------------------------------------------+
-   | :const:`rounds`     | addition rounds (see :file:`float.h`)            |
-   +---------------------+--------------------------------------------------+
-
-   .. note::
-
-      The information in the table is simplified.
+      >>> import sys
+      >>> sys.float_info.dig
+      15
+      >>> s = '3.14159265358979'    # decimal string with 15 significant digits
+      >>> format(float(s), '.15g')  # convert to float and back -> same value
+      '3.14159265358979'
+
+   But for strings with more than :attr:`sys.float_info.dig` significant digits,
+   this isn't always true::
+
+      >>> s = '9876543211234567'    # 16 significant digits is too many!
+      >>> format(float(s), '.16g')  # conversion changes value
+      '9876543211234568'
 
    .. versionadded:: 2.6
 
@@ -992,3 +1014,8 @@
    first three characters of :const:`version`.  It is provided in the :mod:`sys`
    module for informational purposes; modifying this value has no effect on the
    registry keys used by Python. Availability: Windows.
+
+.. rubric:: Citations
+
+.. [C99] ISO/IEC 9899:1999.  "Programming languages -- C."  A public draft of this standard is available at http://www.open-std.org/jtc1/sc22/wg14/www/docs/n1256.pdf .
+

From python-checkins at python.org  Fri Jul  2 21:41:40 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Fri,  2 Jul 2010 21:41:40 +0200 (CEST)
Subject: [Python-checkins] r82447 - python/trunk/Objects/floatobject.c
Message-ID: <20100702194140.13745C9A6@mail.python.org>

Author: benjamin.peterson
Date: Fri Jul  2 21:41:39 2010
New Revision: 82447

Log:
add space

Modified:
   python/trunk/Objects/floatobject.c

Modified: python/trunk/Objects/floatobject.c
==============================================================================
--- python/trunk/Objects/floatobject.c	(original)
+++ python/trunk/Objects/floatobject.c	Fri Jul  2 21:41:39 2010
@@ -1347,7 +1347,7 @@
         return float_str((PyFloatObject *)v);
 
     if (x == 0.0) {
-        if(copysign(1.0, x) == -1.0)
+        if (copysign(1.0, x) == -1.0)
             return PyString_FromString("-0x0.0p+0");
         else
             return PyString_FromString("0x0.0p+0");

From python-checkins at python.org  Fri Jul  2 21:45:07 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Fri,  2 Jul 2010 21:45:07 +0200 (CEST)
Subject: [Python-checkins] r82448 - in python/branches/py3k:
	Objects/floatobject.c
Message-ID: <20100702194507.59EDCCA21@mail.python.org>

Author: benjamin.peterson
Date: Fri Jul  2 21:45:07 2010
New Revision: 82448

Log:
Merged revisions 82447 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/trunk

........
  r82447 | benjamin.peterson | 2010-07-02 14:41:39 -0500 (Fri, 02 Jul 2010) | 1 line
  
  add space
........


Modified:
   python/branches/py3k/   (props changed)
   python/branches/py3k/Objects/floatobject.c

Modified: python/branches/py3k/Objects/floatobject.c
==============================================================================
--- python/branches/py3k/Objects/floatobject.c	(original)
+++ python/branches/py3k/Objects/floatobject.c	Fri Jul  2 21:45:07 2010
@@ -1172,7 +1172,7 @@
         return float_str((PyFloatObject *)v);
 
     if (x == 0.0) {
-        if(copysign(1.0, x) == -1.0)
+        if (copysign(1.0, x) == -1.0)
             return PyUnicode_FromString("-0x0.0p+0");
         else
             return PyUnicode_FromString("0x0.0p+0");

From python-checkins at python.org  Fri Jul  2 21:45:12 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Fri,  2 Jul 2010 21:45:12 +0200 (CEST)
Subject: [Python-checkins] r82449 - in python/branches/release26-maint:
	Objects/floatobject.c
Message-ID: <20100702194512.0FDB9EE98B@mail.python.org>

Author: benjamin.peterson
Date: Fri Jul  2 21:45:11 2010
New Revision: 82449

Log:
Merged revisions 82447 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/trunk

........
  r82447 | benjamin.peterson | 2010-07-02 14:41:39 -0500 (Fri, 02 Jul 2010) | 1 line
  
  add space
........


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Objects/floatobject.c

Modified: python/branches/release26-maint/Objects/floatobject.c
==============================================================================
--- python/branches/release26-maint/Objects/floatobject.c	(original)
+++ python/branches/release26-maint/Objects/floatobject.c	Fri Jul  2 21:45:11 2010
@@ -1216,7 +1216,7 @@
         return float_str((PyFloatObject *)v);
 
     if (x == 0.0) {
-        if(copysign(1.0, x) == -1.0)
+        if (copysign(1.0, x) == -1.0)
             return PyString_FromString("-0x0.0p+0");
         else
             return PyString_FromString("0x0.0p+0");

From python-checkins at python.org  Fri Jul  2 21:50:16 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Fri,  2 Jul 2010 21:50:16 +0200 (CEST)
Subject: [Python-checkins] r82450 - in python/branches/release31-maint:
	Objects/floatobject.c
Message-ID: <20100702195016.67B62D37A@mail.python.org>

Author: benjamin.peterson
Date: Fri Jul  2 21:50:16 2010
New Revision: 82450

Log:
Merged revisions 82448 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

................
  r82448 | benjamin.peterson | 2010-07-02 14:45:07 -0500 (Fri, 02 Jul 2010) | 9 lines
  
  Merged revisions 82447 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82447 | benjamin.peterson | 2010-07-02 14:41:39 -0500 (Fri, 02 Jul 2010) | 1 line
    
    add space
  ........
................


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Objects/floatobject.c

Modified: python/branches/release31-maint/Objects/floatobject.c
==============================================================================
--- python/branches/release31-maint/Objects/floatobject.c	(original)
+++ python/branches/release31-maint/Objects/floatobject.c	Fri Jul  2 21:50:16 2010
@@ -1110,7 +1110,7 @@
         return float_str((PyFloatObject *)v);
 
     if (x == 0.0) {
-        if(copysign(1.0, x) == -1.0)
+        if (copysign(1.0, x) == -1.0)
             return PyUnicode_FromString("-0x0.0p+0");
         else
             return PyUnicode_FromString("0x0.0p+0");

From python-checkins at python.org  Fri Jul  2 22:24:32 2010
From: python-checkins at python.org (mark.dickinson)
Date: Fri,  2 Jul 2010 22:24:32 +0200 (CEST)
Subject: [Python-checkins] r82451 - in python/branches/release26-maint:
	Doc/library/sys.rst
Message-ID: <20100702202432.7D959EFCB@mail.python.org>

Author: mark.dickinson
Date: Fri Jul  2 22:24:32 2010
New Revision: 82451

Log:
Merged revisions 82446 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/trunk

........
  r82446 | mark.dickinson | 2010-07-02 19:06:52 +0100 (Fri, 02 Jul 2010) | 1 line
  
  Clarify sys.float_info documentation.
........


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Doc/library/sys.rst

Modified: python/branches/release26-maint/Doc/library/sys.rst
==============================================================================
--- python/branches/release26-maint/Doc/library/sys.rst	(original)
+++ python/branches/release26-maint/Doc/library/sys.rst	Fri Jul  2 22:24:32 2010
@@ -296,43 +296,65 @@
 .. data:: float_info
 
    A structseq holding information about the float type. It contains low level
-   information about the precision and internal representation. Please study
-   your system's :file:`float.h` for more information.
+   information about the precision and internal representation.  The values
+   correspond to the various floating-point constants defined in the standard
+   header file :file:`float.h` for the 'C' programming language; see section
+   5.2.4.2.2 of the 1999 ISO/IEC C standard [C99]_, 'Characteristics of
+   floating types', for details.
+
+   +---------------------+----------------+--------------------------------------------------+
+   | attribute           | float.h macro  | explanation                                      |
+   +=====================+================+==================================================+
+   | :const:`epsilon`    | DBL_MAX        | difference between 1 and the least value greater |
+   |                     |                | than 1 that is representable as a float          |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`dig`        | DBL_DIG        | maximum number of decimal digits that can be     |
+   |                     |                | faithfully represented in a float;  see below    |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`mant_dig`   | DBL_MANT_DIG   | float precision: the number of base-``radix``    |
+   |                     |                | digits in the significand of a float             |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`max`        | DBL_MAX        | maximum representable finite float               |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`max_exp`    | DBL_MAX_EXP    | maximum integer e such that ``radix**(e-1)`` is  |
+   |                     |                | a representable finite float                     |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`max_10_exp` | DBL_MAX_10_EXP | maximum integer e such that ``10**e`` is in the  |
+   |                     |                | range of representable finite floats             |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`min`        | DBL_MIN        | minimum positive normalized float                |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`min_exp`    | DBL_MIN_EXP    | minimum integer e such that ``radix**(e-1)`` is  |
+   |                     |                | a normalized float                               |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`min_10_exp` | DBL_MIN_10_EXP | minimum integer e such that ``10**e`` is a       |
+   |                     |                | normalized float                                 |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`radix`      | FLT_RADIX      | radix of exponent representation                 |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`rounds`     | FLT_ROUNDS     | constant representing rounding mode              |
+   |                     |                | used for arithmetic operations                   |
+   +---------------------+----------------+--------------------------------------------------+
+
+   The attribute :attr:`sys.float_info.dig` needs further explanation.  If
+   ``s`` is any string representing a decimal number with at most
+   :attr:`sys.float_info.dig` significant digits, then converting ``s`` to a
+   float and back again will recover a string representing the same decimal
+   value::
 
-   +---------------------+--------------------------------------------------+
-   | attribute           |  explanation                                     |
-   +=====================+==================================================+
-   | :const:`epsilon`    | Difference between 1 and the next representable  |
-   |                     | floating point number                            |
-   +---------------------+--------------------------------------------------+
-   | :const:`dig`        | digits (see :file:`float.h`)                     |
-   +---------------------+--------------------------------------------------+
-   | :const:`mant_dig`   | mantissa digits (see :file:`float.h`)            |
-   +---------------------+--------------------------------------------------+
-   | :const:`max`        | maximum representable finite float               |
-   +---------------------+--------------------------------------------------+
-   | :const:`max_exp`    | maximum int e such that radix**(e-1) is in the   |
-   |                     | range of finite representable floats             |
-   +---------------------+--------------------------------------------------+
-   | :const:`max_10_exp` | maximum int e such that 10**e is in the          |
-   |                     | range of finite representable floats             |
-   +---------------------+--------------------------------------------------+
-   | :const:`min`        | Minimum positive normalizer float                |
-   +---------------------+--------------------------------------------------+
-   | :const:`min_exp`    | minimum int e such that radix**(e-1) is a        |
-   |                     | normalized float                                 |
-   +---------------------+--------------------------------------------------+
-   | :const:`min_10_exp` | minimum int e such that 10**e is a normalized    |
-   |                     | float                                            |
-   +---------------------+--------------------------------------------------+
-   | :const:`radix`      | radix of exponent                                |
-   +---------------------+--------------------------------------------------+
-   | :const:`rounds`     | addition rounds (see :file:`float.h`)            |
-   +---------------------+--------------------------------------------------+
-
-   .. note::
-
-      The information in the table is simplified.
+      >>> import sys
+      >>> sys.float_info.dig
+      15
+      >>> s = '3.14159265358979'    # decimal string with 15 significant digits
+      >>> format(float(s), '.15g')  # convert to float and back -> same value
+      '3.14159265358979'
+
+   But for strings with more than :attr:`sys.float_info.dig` significant digits,
+   this isn't always true::
+
+      >>> s = '9876543211234567'    # 16 significant digits is too many!
+      >>> format(float(s), '.16g')  # conversion changes value
+      '9876543211234568'
 
    .. versionadded:: 2.6
 
@@ -930,3 +952,8 @@
    first three characters of :const:`version`.  It is provided in the :mod:`sys`
    module for informational purposes; modifying this value has no effect on the
    registry keys used by Python. Availability: Windows.
+
+.. rubric:: Citations
+
+.. [C99] ISO/IEC 9899:1999.  "Programming languages -- C."  A public draft of this standard is available at http://www.open-std.org/jtc1/sc22/wg14/www/docs/n1256.pdf .
+

From python-checkins at python.org  Fri Jul  2 22:26:08 2010
From: python-checkins at python.org (mark.dickinson)
Date: Fri,  2 Jul 2010 22:26:08 +0200 (CEST)
Subject: [Python-checkins] r82452 - in python/branches/py3k:
	Doc/library/sys.rst
Message-ID: <20100702202608.25ABDEFCB@mail.python.org>

Author: mark.dickinson
Date: Fri Jul  2 22:26:07 2010
New Revision: 82452

Log:
Merged revisions 82446 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/trunk

........
  r82446 | mark.dickinson | 2010-07-02 19:06:52 +0100 (Fri, 02 Jul 2010) | 1 line
  
  Clarify sys.float_info documentation.
........


Modified:
   python/branches/py3k/   (props changed)
   python/branches/py3k/Doc/library/sys.rst

Modified: python/branches/py3k/Doc/library/sys.rst
==============================================================================
--- python/branches/py3k/Doc/library/sys.rst	(original)
+++ python/branches/py3k/Doc/library/sys.rst	Fri Jul  2 22:26:07 2010
@@ -223,44 +223,65 @@
 .. data:: float_info
 
    A structseq holding information about the float type. It contains low level
-   information about the precision and internal representation. Please study
-   your system's :file:`float.h` for more information.
-
-   +---------------------+--------------------------------------------------+
-   | attribute           |  explanation                                     |
-   +=====================+==================================================+
-   | :const:`epsilon`    | Difference between 1 and the next representable  |
-   |                     | floating point number                            |
-   +---------------------+--------------------------------------------------+
-   | :const:`dig`        | digits (see :file:`float.h`)                     |
-   +---------------------+--------------------------------------------------+
-   | :const:`mant_dig`   | mantissa digits (see :file:`float.h`)            |
-   +---------------------+--------------------------------------------------+
-   | :const:`max`        | maximum representable finite float               |
-   +---------------------+--------------------------------------------------+
-   | :const:`max_exp`    | maximum int e such that radix**(e-1) is in the   |
-   |                     | range of finite representable floats             |
-   +---------------------+--------------------------------------------------+
-   | :const:`max_10_exp` | maximum int e such that 10**e is in the          |
-   |                     | range of finite representable floats             |
-   +---------------------+--------------------------------------------------+
-   | :const:`min`        | Minimum positive normalizer float                |
-   +---------------------+--------------------------------------------------+
-   | :const:`min_exp`    | minimum int e such that radix**(e-1) is a        |
-   |                     | normalized float                                 |
-   +---------------------+--------------------------------------------------+
-   | :const:`min_10_exp` | minimum int e such that 10**e is a normalized    |
-   |                     | float                                            |
-   +---------------------+--------------------------------------------------+
-   | :const:`radix`      | radix of exponent                                |
-   +---------------------+--------------------------------------------------+
-   | :const:`rounds`     | addition rounds (see :file:`float.h`)            |
-   +---------------------+--------------------------------------------------+
-
-   .. note::
-
-      The information in the table is simplified.
+   information about the precision and internal representation.  The values
+   correspond to the various floating-point constants defined in the standard
+   header file :file:`float.h` for the 'C' programming language; see section
+   5.2.4.2.2 of the 1999 ISO/IEC C standard [C99]_, 'Characteristics of
+   floating types', for details.
+
+   +---------------------+----------------+--------------------------------------------------+
+   | attribute           | float.h macro  | explanation                                      |
+   +=====================+================+==================================================+
+   | :const:`epsilon`    | DBL_MAX        | difference between 1 and the least value greater |
+   |                     |                | than 1 that is representable as a float          |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`dig`        | DBL_DIG        | maximum number of decimal digits that can be     |
+   |                     |                | faithfully represented in a float;  see below    |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`mant_dig`   | DBL_MANT_DIG   | float precision: the number of base-``radix``    |
+   |                     |                | digits in the significand of a float             |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`max`        | DBL_MAX        | maximum representable finite float               |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`max_exp`    | DBL_MAX_EXP    | maximum integer e such that ``radix**(e-1)`` is  |
+   |                     |                | a representable finite float                     |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`max_10_exp` | DBL_MAX_10_EXP | maximum integer e such that ``10**e`` is in the  |
+   |                     |                | range of representable finite floats             |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`min`        | DBL_MIN        | minimum positive normalized float                |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`min_exp`    | DBL_MIN_EXP    | minimum integer e such that ``radix**(e-1)`` is  |
+   |                     |                | a normalized float                               |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`min_10_exp` | DBL_MIN_10_EXP | minimum integer e such that ``10**e`` is a       |
+   |                     |                | normalized float                                 |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`radix`      | FLT_RADIX      | radix of exponent representation                 |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`rounds`     | FLT_ROUNDS     | constant representing rounding mode              |
+   |                     |                | used for arithmetic operations                   |
+   +---------------------+----------------+--------------------------------------------------+
+
+   The attribute :attr:`sys.float_info.dig` needs further explanation.  If
+   ``s`` is any string representing a decimal number with at most
+   :attr:`sys.float_info.dig` significant digits, then converting ``s`` to a
+   float and back again will recover a string representing the same decimal
+   value::
 
+      >>> import sys
+      >>> sys.float_info.dig
+      15
+      >>> s = '3.14159265358979'    # decimal string with 15 significant digits
+      >>> format(float(s), '.15g')  # convert to float and back -> same value
+      '3.14159265358979'
+
+   But for strings with more than :attr:`sys.float_info.dig` significant digits,
+   this isn't always true::
+
+      >>> s = '9876543211234567'    # 16 significant digits is too many!
+      >>> format(float(s), '.16g')  # conversion changes value
+      '9876543211234568'
 
 .. data:: float_repr_style
 
@@ -957,3 +978,8 @@
    first three characters of :const:`version`.  It is provided in the :mod:`sys`
    module for informational purposes; modifying this value has no effect on the
    registry keys used by Python. Availability: Windows.
+
+.. rubric:: Citations
+
+.. [C99] ISO/IEC 9899:1999.  "Programming languages -- C."  A public draft of this standard is available at http://www.open-std.org/jtc1/sc22/wg14/www/docs/n1256.pdf .
+

From python-checkins at python.org  Fri Jul  2 22:26:51 2010
From: python-checkins at python.org (mark.dickinson)
Date: Fri,  2 Jul 2010 22:26:51 +0200 (CEST)
Subject: [Python-checkins] r82453 - in python/branches/release31-maint:
	Doc/library/sys.rst
Message-ID: <20100702202651.D0408EE98C@mail.python.org>

Author: mark.dickinson
Date: Fri Jul  2 22:26:51 2010
New Revision: 82453

Log:
Merged revisions 82452 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

................
  r82452 | mark.dickinson | 2010-07-02 21:26:07 +0100 (Fri, 02 Jul 2010) | 9 lines
  
  Merged revisions 82446 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82446 | mark.dickinson | 2010-07-02 19:06:52 +0100 (Fri, 02 Jul 2010) | 1 line
    
    Clarify sys.float_info documentation.
  ........
................


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Doc/library/sys.rst

Modified: python/branches/release31-maint/Doc/library/sys.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/sys.rst	(original)
+++ python/branches/release31-maint/Doc/library/sys.rst	Fri Jul  2 22:26:51 2010
@@ -224,44 +224,65 @@
 .. data:: float_info
 
    A structseq holding information about the float type. It contains low level
-   information about the precision and internal representation. Please study
-   your system's :file:`float.h` for more information.
-
-   +---------------------+--------------------------------------------------+
-   | attribute           |  explanation                                     |
-   +=====================+==================================================+
-   | :const:`epsilon`    | Difference between 1 and the next representable  |
-   |                     | floating point number                            |
-   +---------------------+--------------------------------------------------+
-   | :const:`dig`        | digits (see :file:`float.h`)                     |
-   +---------------------+--------------------------------------------------+
-   | :const:`mant_dig`   | mantissa digits (see :file:`float.h`)            |
-   +---------------------+--------------------------------------------------+
-   | :const:`max`        | maximum representable finite float               |
-   +---------------------+--------------------------------------------------+
-   | :const:`max_exp`    | maximum int e such that radix**(e-1) is in the   |
-   |                     | range of finite representable floats             |
-   +---------------------+--------------------------------------------------+
-   | :const:`max_10_exp` | maximum int e such that 10**e is in the          |
-   |                     | range of finite representable floats             |
-   +---------------------+--------------------------------------------------+
-   | :const:`min`        | Minimum positive normalizer float                |
-   +---------------------+--------------------------------------------------+
-   | :const:`min_exp`    | minimum int e such that radix**(e-1) is a        |
-   |                     | normalized float                                 |
-   +---------------------+--------------------------------------------------+
-   | :const:`min_10_exp` | minimum int e such that 10**e is a normalized    |
-   |                     | float                                            |
-   +---------------------+--------------------------------------------------+
-   | :const:`radix`      | radix of exponent                                |
-   +---------------------+--------------------------------------------------+
-   | :const:`rounds`     | addition rounds (see :file:`float.h`)            |
-   +---------------------+--------------------------------------------------+
-
-   .. note::
-
-      The information in the table is simplified.
+   information about the precision and internal representation.  The values
+   correspond to the various floating-point constants defined in the standard
+   header file :file:`float.h` for the 'C' programming language; see section
+   5.2.4.2.2 of the 1999 ISO/IEC C standard [C99]_, 'Characteristics of
+   floating types', for details.
+
+   +---------------------+----------------+--------------------------------------------------+
+   | attribute           | float.h macro  | explanation                                      |
+   +=====================+================+==================================================+
+   | :const:`epsilon`    | DBL_MAX        | difference between 1 and the least value greater |
+   |                     |                | than 1 that is representable as a float          |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`dig`        | DBL_DIG        | maximum number of decimal digits that can be     |
+   |                     |                | faithfully represented in a float;  see below    |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`mant_dig`   | DBL_MANT_DIG   | float precision: the number of base-``radix``    |
+   |                     |                | digits in the significand of a float             |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`max`        | DBL_MAX        | maximum representable finite float               |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`max_exp`    | DBL_MAX_EXP    | maximum integer e such that ``radix**(e-1)`` is  |
+   |                     |                | a representable finite float                     |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`max_10_exp` | DBL_MAX_10_EXP | maximum integer e such that ``10**e`` is in the  |
+   |                     |                | range of representable finite floats             |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`min`        | DBL_MIN        | minimum positive normalized float                |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`min_exp`    | DBL_MIN_EXP    | minimum integer e such that ``radix**(e-1)`` is  |
+   |                     |                | a normalized float                               |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`min_10_exp` | DBL_MIN_10_EXP | minimum integer e such that ``10**e`` is a       |
+   |                     |                | normalized float                                 |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`radix`      | FLT_RADIX      | radix of exponent representation                 |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`rounds`     | FLT_ROUNDS     | constant representing rounding mode              |
+   |                     |                | used for arithmetic operations                   |
+   +---------------------+----------------+--------------------------------------------------+
+
+   The attribute :attr:`sys.float_info.dig` needs further explanation.  If
+   ``s`` is any string representing a decimal number with at most
+   :attr:`sys.float_info.dig` significant digits, then converting ``s`` to a
+   float and back again will recover a string representing the same decimal
+   value::
 
+      >>> import sys
+      >>> sys.float_info.dig
+      15
+      >>> s = '3.14159265358979'    # decimal string with 15 significant digits
+      >>> format(float(s), '.15g')  # convert to float and back -> same value
+      '3.14159265358979'
+
+   But for strings with more than :attr:`sys.float_info.dig` significant digits,
+   this isn't always true::
+
+      >>> s = '9876543211234567'    # 16 significant digits is too many!
+      >>> format(float(s), '.16g')  # conversion changes value
+      '9876543211234568'
 
 .. data:: float_repr_style
 
@@ -875,3 +896,8 @@
    first three characters of :const:`version`.  It is provided in the :mod:`sys`
    module for informational purposes; modifying this value has no effect on the
    registry keys used by Python. Availability: Windows.
+
+.. rubric:: Citations
+
+.. [C99] ISO/IEC 9899:1999.  "Programming languages -- C."  A public draft of this standard is available at http://www.open-std.org/jtc1/sc22/wg14/www/docs/n1256.pdf .
+

From python-checkins at python.org  Fri Jul  2 23:25:47 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Fri,  2 Jul 2010 23:25:47 +0200 (CEST)
Subject: [Python-checkins] r82454 -
	sandbox/branches/py3k-datetime/datetime.py
Message-ID: <20100702212547.5A091EE4D@mail.python.org>

Author: alexander.belopolsky
Date: Fri Jul  2 23:25:47 2010
New Revision: 82454

Log:
Cleaned up some XXXs

Modified:
   sandbox/branches/py3k-datetime/datetime.py

Modified: sandbox/branches/py3k-datetime/datetime.py
==============================================================================
--- sandbox/branches/py3k-datetime/datetime.py	(original)
+++ sandbox/branches/py3k-datetime/datetime.py	Fri Jul  2 23:25:47 2010
@@ -319,7 +319,6 @@
     __slots__ = '__days', '__seconds', '__microseconds'
 
     def __new__(cls, days=0, seconds=0, microseconds=0,
-                # XXX The following should only be used as keyword args:
                 milliseconds=0, minutes=0, hours=0, weeks=0):
         # Doing this efficiently and accurately in C is going to be difficult
         # and error-prone, due to ubiquitous overflow possibilities, and that
@@ -329,7 +328,7 @@
         # guide the C implementation; it's way more convoluted than speed-
         # ignoring auto-overflow-to-long idiomatic Python could be.
 
-        # XXX Check that all inputs are ints, longs or floats.
+        # XXX Check that all inputs are ints or floats.
 
         # Final values, all integer.
         # s and us fit in 32-bit signed ints; d isn't bounded.
@@ -839,8 +838,7 @@
             if 0 < o <= _MAXORDINAL:
                 return date.fromordinal(o)
             raise OverflowError("result out of range")
-        raise TypeError
-        # XXX Should be 'return NotImplemented', but there's a bug in 2.2...
+        return NotImplemented
 
     __radd__ = __add__
 
@@ -1112,7 +1110,6 @@
                        (other.__hour, other.__minute, other.__second,
                         other.__microsecond))
         if myoff is None or otoff is None:
-            # XXX Buggy in 2.2.2.
             raise TypeError("cannot compare naive and aware times")
         myhhmm = self.__hour * 60 + self.__minute - myoff
         othhmm = other.__hour * 60 + other.__minute - otoff
@@ -1287,9 +1284,11 @@
 time.resolution = timedelta(microseconds=1)
 
 class datetime(date):
+    """datetime(year, month, day[, hour[, minute[, second[, microsecond[,tzinfo]]]]])
 
-    # XXX needs docstrings
-    # See http://www.zope.org/Members/fdrake/DateTimeWiki/TimeZoneInfo
+    The year, month and day arguments are required. tzinfo may be None, or an
+    instance of a tzinfo subclass. The remaining arguments may be ints or longs.
+    """
 
     __slots__ = date.__slots__ + (
         '__hour', '__minute', '__second',
@@ -1667,7 +1666,6 @@
                         other.__hour, other.__minute, other.__second,
                         other.__microsecond))
         if myoff is None or otoff is None:
-            # XXX Buggy in 2.2.2.
             raise TypeError("cannot compare naive and aware datetimes")
         # XXX What follows could be done more efficiently...
         diff = self - other     # this will take offsets into account

From python-checkins at python.org  Fri Jul  2 23:44:16 2010
From: python-checkins at python.org (eric.smith)
Date: Fri,  2 Jul 2010 23:44:16 +0200 (CEST)
Subject: [Python-checkins] r82455 - python/trunk/Doc/library/string.rst
Message-ID: <20100702214416.CB4C2D280@mail.python.org>

Author: eric.smith
Date: Fri Jul  2 23:44:16 2010
New Revision: 82455

Log:
Moved period outside paren, where it belongs.

Modified:
   python/trunk/Doc/library/string.rst

Modified: python/trunk/Doc/library/string.rst
==============================================================================
--- python/trunk/Doc/library/string.rst	(original)
+++ python/trunk/Doc/library/string.rst	Fri Jul  2 23:44:16 2010
@@ -212,7 +212,7 @@
 
 The :meth:`str.format` method and the :class:`Formatter` class share the same
 syntax for format strings (although in the case of :class:`Formatter`,
-subclasses can define their own format string syntax.)
+subclasses can define their own format string syntax).
 
 Format strings contain "replacement fields" surrounded by curly braces ``{}``.
 Anything that is not contained in braces is considered literal text, which is

From python-checkins at python.org  Sat Jul  3 00:03:00 2010
From: python-checkins at python.org (brett.cannon)
Date: Sat,  3 Jul 2010 00:03:00 +0200 (CEST)
Subject: [Python-checkins] r82456 - in python/branches/py3k:
	Doc/glossary.rst Doc/library/dis.rst Misc/NEWS
Message-ID: <20100702220300.A3D3EF2EA@mail.python.org>

Author: brett.cannon
Date: Sat Jul  3 00:03:00 2010
New Revision: 82456

Log:
Some people have mistaken Python bytecode as being stable and unchanging. In
reality it's simply an implementation detail for CPython. This point is now
clearly documented in both the docs for dis and the glossary.

Closes issue #7829. Thanks to Terry Reedy for some initial suggestions on
wording.


Modified:
   python/branches/py3k/Doc/glossary.rst
   python/branches/py3k/Doc/library/dis.rst
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Doc/glossary.rst
==============================================================================
--- python/branches/py3k/Doc/glossary.rst	(original)
+++ python/branches/py3k/Doc/glossary.rst	Sat Jul  3 00:03:00 2010
@@ -57,11 +57,14 @@
 
    bytecode
       Python source code is compiled into bytecode, the internal representation
-      of a Python program in the interpreter.  The bytecode is also cached in
-      ``.pyc`` and ``.pyo`` files so that executing the same file is faster the
-      second time (recompilation from source to bytecode can be avoided).  This
-      "intermediate language" is said to run on a :term:`virtual machine`
-      that executes the machine code corresponding to each bytecode.
+      of a Python program in the CPython interpreter.  The bytecode is also
+      cached in ``.pyc`` and ``.pyo`` files so that executing the same file is
+      faster the second time (recompilation from source to bytecode can be
+      avoided).  This "intermediate language" is said to run on a
+      :term:`virtual machine` that executes the machine code corresponding to
+      each bytecode. Do note that bytecodes are not expected to work between
+      different Python virtual machines, nor to be stable between Python
+      releases.
 
    class
       A template for creating user-defined objects. Class definitions

Modified: python/branches/py3k/Doc/library/dis.rst
==============================================================================
--- python/branches/py3k/Doc/library/dis.rst	(original)
+++ python/branches/py3k/Doc/library/dis.rst	Sat Jul  3 00:03:00 2010
@@ -5,12 +5,13 @@
    :synopsis: Disassembler for Python bytecode.
 
 
-The :mod:`dis` module supports the analysis of Python :term:`bytecode` by
-disassembling it.  Since there is no Python assembler, this module defines the
-Python assembly language.  The Python bytecode which this module takes as an
+The :mod:`dis` module supports the analysis of CPython :term:`bytecode` by
+disassembling it. The CPython bytecode which this module takes as an
 input is defined in the file :file:`Include/opcode.h` and used by the compiler
 and the interpreter.
 
+.. warning:: Bytecode is an implementation detail of the CPython interpreter!  No guarantees are made that bytecode will not be added, removed, or changed between versions of Python. Use of this module should not be considered to work across Python VMs or Python releases.
+
 Example: Given the function :func:`myfunc`::
 
    def myfunc(alist):

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sat Jul  3 00:03:00 2010
@@ -468,6 +468,9 @@
 Library
 -------
 
+- Issue #7829: Clearly document that the dis module is exposing an
+  implementation detail that is not stable between Python VMs or releases.
+
 - Issue #6589: cleanup asyncore.socket_map in case smtpd.SMTPServer constructor 
   raises an exception.
 

From python-checkins at python.org  Sat Jul  3 00:17:29 2010
From: python-checkins at python.org (ezio.melotti)
Date: Sat,  3 Jul 2010 00:17:29 +0200 (CEST)
Subject: [Python-checkins] r82457 - python/trunk/Doc/library/string.rst
Message-ID: <20100702221729.626C9EDA6@mail.python.org>

Author: ezio.melotti
Date: Sat Jul  3 00:17:29 2010
New Revision: 82457

Log:
#9139: Add examples for str.format().

Modified:
   python/trunk/Doc/library/string.rst

Modified: python/trunk/Doc/library/string.rst
==============================================================================
--- python/trunk/Doc/library/string.rst	(original)
+++ python/trunk/Doc/library/string.rst	Sat Jul  3 00:17:29 2010
@@ -201,7 +201,7 @@
    .. method:: convert_field(value, conversion)
 
       Converts the value (returned by :meth:`get_field`) given a conversion type
-      (as in the tuple returned by the :meth:`parse` method.)  The default
+      (as in the tuple returned by the :meth:`parse` method).  The default
       version understands 'r' (repr) and 's' (str) conversion types.
 
 
@@ -238,6 +238,8 @@
 preceded by an exclamation point ``'!'``, and a *format_spec*, which is preceded
 by a colon ``':'``.  These specify a non-default format for the replacement value.
 
+See also the :ref:`formatspec` section.
+
 The *field_name* itself begins with an *arg_name* that is either either a number or a
 keyword.  If it's a number, it refers to a positional argument, and if it's a keyword,
 it refers to a named keyword argument.  If the numerical arg_names in a format string
@@ -248,6 +250,10 @@
 attribute using :func:`getattr`, while an expression of the form ``'[index]'``
 does an index lookup using :func:`__getitem__`.
 
+.. versionchanged:: 2.7
+   The positional argument specifiers can be omitted, so ``'{} {}'`` is
+   equivalent to ``'{0} {1}'``.
+
 Some simple format string examples::
 
    "First, thou shalt count to {0}" # References first positional argument
@@ -286,26 +292,7 @@
 format_spec are substituted before the *format_spec* string is interpreted.
 This allows the formatting of a value to be dynamically specified.
 
-For example, suppose you wanted to have a replacement field whose field width is
-determined by another variable::
-
-   "A man with two {0:{1}}".format("noses", 10)
-
-This would first evaluate the inner replacement field, making the format string
-effectively::
-
-   "A man with two {0:10}"
-
-Then the outer replacement field would be evaluated, producing::
-
-   "noses     "
-
-Which is substituted into the string, yielding::
-
-   "A man with two noses     "
-
-(The extra space is because we specified a field width of 10, and because left
-alignment is the default for strings.)
+See the :ref:`formatexamples` section for some examples.
 
 
 .. _formatspec:
@@ -315,7 +302,7 @@
 
 "Format specifications" are used within replacement fields contained within a
 format string to define how individual values are presented (see
-:ref:`formatstrings`.)  They can also be passed directly to the built-in
+:ref:`formatstrings`).  They can also be passed directly to the built-in
 :func:`format` function.  Each formattable type may define how the format
 specification is to be interpreted.
 
@@ -349,7 +336,7 @@
    | Option  | Meaning                                                  |
    +=========+==========================================================+
    | ``'<'`` | Forces the field to be left-aligned within the available |
-   |         | space (This is the default.)                             |
+   |         | space (this is the default).                             |
    +---------+----------------------------------------------------------+
    | ``'>'`` | Forces the field to be right-aligned within the          |
    |         | available space.                                         |
@@ -505,6 +492,148 @@
    +---------+----------------------------------------------------------+
 
 
+
+.. _formatexamples:
+
+Format examples
+^^^^^^^^^^^^^^^
+
+This section contains examples of the new format syntax and comparison with
+the old ``%``-formatting.
+
+In most of the cases the syntax is similar to the old ``%``-formatting, with the
+addition of the ``{}`` and with ``:`` used instead of ``%``.
+For example, ``'%03.2f'`` can be translated to ``'{:03.2f}'``.
+
+The new format syntax also supports new and different options, shown in the
+follow examples.
+
+Accessing arguments by position::
+
+   >>> '{0}, {1}, {2}'.format('a', 'b', 'c')
+   'a, b, c'
+   >>> '{}, {}, {}'.format('a', 'b', 'c')  # 2.7+ only
+   'a, b, c'
+   >>> '{2}, {1}, {0}'.format('a', 'b', 'c')
+   'c, b, a'
+   >>> '{2}, {1}, {0}'.format(*'abc')      # unpacking argument sequence
+   'c, b, a'
+   >>> '{0}{1}{0}'.format('abra', 'cad')   # arguments' indices can be repeated
+   'abracadabra'
+
+Accessing arguments by name::
+
+   >>> 'Coordinates: {latitude}, {longitude}'.format(latitude='37.24N', longitude='-115.81W')
+   'Coordinates: 37.24N, -115.81W'
+   >>> coord = {'latitude': '37.24N', 'longitude': '-115.81W'}
+   >>> 'Coordinates: {latitude}, {longitude}'.format(**coord)
+   'Coordinates: 37.24N, -115.81W'
+
+Accessing arguments' attributes::
+
+   >>> ('The complex number {0} is formed from the real part {0.real} '
+   ...  'and the imaginary part {0.imag}.').format(c)
+   'The complex number (3-5j) is formed from the real part 3.0 and the imaginary part -5.0.'
+   >>> class Point(object):
+   ...     def __init__(self, x, y):
+   ...         self.x, self.y = x, y
+   ...     def __str__(self):
+   ...         return 'Point({self.x}, {self.y})'.format(self=self)
+   ...
+   >>> str(Point(4, 2))
+   'Point(4, 2)'
+
+
+Accessing arguments' items::
+
+   >>> coord = (3, 5)
+   >>> 'X: {0[0]};  Y: {0[1]}'.format(coord)
+   'X: 3;  Y: 5'
+
+Replacing ``%s`` and ``%r``::
+
+   >>> "repr() shows quotes: {!r}; str() doesn't: {!s}".format('test1', 'test2')
+   "repr() shows quotes: 'test1'; str() doesn't: test2"
+
+Aligning the text and specifying a width::
+
+   >>> '{:<30}'.format('left aligned')
+   'left aligned                  '
+   >>> '{:>30}'.format('right aligned')
+   '                 right aligned'
+   >>> '{:^30}'.format('centered')
+   '           centered           '
+   >>> '{:*^30}'.format('centered')  # use '*' as a fill char
+   '***********centered***********'
+
+Replacing ``%+f``, ``%-f``, and ``% f`` and specifying a sign::
+
+   >>> '{:+f}; {:+f}'.format(3.14, -3.14)  # show it always
+   '+3.140000; -3.140000'
+   >>> '{: f}; {: f}'.format(3.14, -3.14)  # show a space for positive numbers
+   ' 3.140000; -3.140000'
+   >>> '{:-f}; {:-f}'.format(3.14, -3.14)  # show only the minus -- same as '{:f}; {:f}'
+   '3.140000; -3.140000'
+
+Replacing ``%x`` and ``%o`` and converting the value to different bases::
+
+   >>> # format also supports binary numbers
+   >>> "int: {0:d};  hex: {0:x};  oct: {0:o};  bin: {0:b}".format(42)
+   'int: 42;  hex: 2a;  oct: 52;  bin: 101010'
+   >>> # with 0x, 0o, or 0b as prefix:
+   >>> "int: {0:d};  hex: {0:#x};  oct: {0:#o};  bin: {0:#b}".format(42)
+   'int: 42;  hex: 0x2a;  oct: 0o52;  bin: 0b101010'
+
+Using the comma as a thousands separator::
+
+   >>> '{:,}'.format(1234567890)
+   '1,234,567,890'
+
+Expressing a percentage::
+
+   >>> points = 19.5
+   >>> total = 22
+   >>> 'Correct answers: {:.2%}.'.format(points/total)
+   'Correct answers: 88.64%'
+
+Using type-specific formatting::
+
+   >>> import datetime
+   >>> d = datetime.datetime(2010, 7, 4, 12, 15, 58)
+   >>> '{:%Y-%m-%d %H:%M:%S}'.format(d)
+   '2010-07-04 12:15:58'
+
+Nesting arguments and more complex examples::
+
+   >>> for align, text in zip('<^>', ['left', 'center', 'right']):
+   ...     '{0:{align}{fill}16}'.format(text, fill=align, align=align)
+   ...
+   'left<<<<<<<<<<<<'
+   '^^^^^center^^^^^'
+   '>>>>>>>>>>>right'
+   >>>
+   >>> octets = [192, 168, 0, 1]
+   >>> '{:02X}{:02X}{:02X}{:02X}'.format(*octets)
+   'C0A80001'
+   >>> int(_, 16)
+   3232235521
+   >>>
+   >>> width = 5
+   >>> for num in range(5,12):
+   ...     for base in 'dXob':
+   ...         print '{0:{width}{base}}'.format(num, base=base, width=width),
+   ...     print
+   ...
+       5     5     5   101
+       6     6     6   110
+       7     7     7   111
+       8     8    10  1000
+       9     9    11  1001
+      10     A    12  1010
+      11     B    13  1011
+
+
+
 Template strings
 ----------------
 

From python-checkins at python.org  Sat Jul  3 00:24:55 2010
From: python-checkins at python.org (eric.smith)
Date: Sat,  3 Jul 2010 00:24:55 +0200 (CEST)
Subject: [Python-checkins] r82458 - in python/branches/release26-maint:
	Doc/library/string.rst
Message-ID: <20100702222455.0D699EE2C@mail.python.org>

Author: eric.smith
Date: Sat Jul  3 00:24:54 2010
New Revision: 82458

Log:
Merged revisions 82455 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/trunk

........
  r82455 | eric.smith | 2010-07-02 17:44:16 -0400 (Fri, 02 Jul 2010) | 1 line
  
  Moved period outside paren, where it belongs.
........


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Doc/library/string.rst

Modified: python/branches/release26-maint/Doc/library/string.rst
==============================================================================
--- python/branches/release26-maint/Doc/library/string.rst	(original)
+++ python/branches/release26-maint/Doc/library/string.rst	Sat Jul  3 00:24:54 2010
@@ -212,7 +212,7 @@
 
 The :meth:`str.format` method and the :class:`Formatter` class share the same
 syntax for format strings (although in the case of :class:`Formatter`,
-subclasses can define their own format string syntax.)
+subclasses can define their own format string syntax).
 
 Format strings contain "replacement fields" surrounded by curly braces ``{}``.
 Anything that is not contained in braces is considered literal text, which is

From python-checkins at python.org  Sat Jul  3 00:50:39 2010
From: python-checkins at python.org (ezio.melotti)
Date: Sat,  3 Jul 2010 00:50:39 +0200 (CEST)
Subject: [Python-checkins] r82459 - python/trunk/Doc/library/string.rst
Message-ID: <20100702225039.8B07DECE2@mail.python.org>

Author: ezio.melotti
Date: Sat Jul  3 00:50:39 2010
New Revision: 82459

Log:
#9139: the thousands separator is new in 2.7.  Also add a missing variable in the example.

Modified:
   python/trunk/Doc/library/string.rst

Modified: python/trunk/Doc/library/string.rst
==============================================================================
--- python/trunk/Doc/library/string.rst	(original)
+++ python/trunk/Doc/library/string.rst	Sat Jul  3 00:50:39 2010
@@ -378,6 +378,9 @@
 For a locale aware separator, use the ``'n'`` integer presentation type
 instead.
 
+.. versionchanged:: 2.7
+   Added the ``','`` option (see also :pep:`378`).
+
 *width* is a decimal integer defining the minimum field width.  If not
 specified, then the field width will be determined by the content.
 
@@ -531,6 +534,7 @@
 
 Accessing arguments' attributes::
 
+   >>> c = 3-5j
    >>> ('The complex number {0} is formed from the real part {0.real} '
    ...  'and the imaginary part {0.imag}.').format(c)
    'The complex number (3-5j) is formed from the real part 3.0 and the imaginary part -5.0.'

From python-checkins at python.org  Sat Jul  3 00:58:12 2010
From: python-checkins at python.org (ezio.melotti)
Date: Sat,  3 Jul 2010 00:58:12 +0200 (CEST)
Subject: [Python-checkins] r82460 - in python/branches/release26-maint:
	Doc/library/string.rst
Message-ID: <20100702225812.E6C31DB2A@mail.python.org>

Author: ezio.melotti
Date: Sat Jul  3 00:58:12 2010
New Revision: 82460

Log:
Merged revisions 82457,82459 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/trunk

........
  r82457 | ezio.melotti | 2010-07-03 01:17:29 +0300 (Sat, 03 Jul 2010) | 1 line
  
  #9139: Add examples for str.format().
........
  r82459 | ezio.melotti | 2010-07-03 01:50:39 +0300 (Sat, 03 Jul 2010) | 1 line
  
  #9139: the thousands separator is new in 2.7.  Also add a missing variable in the example.
........


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Doc/library/string.rst

Modified: python/branches/release26-maint/Doc/library/string.rst
==============================================================================
--- python/branches/release26-maint/Doc/library/string.rst	(original)
+++ python/branches/release26-maint/Doc/library/string.rst	Sat Jul  3 00:58:12 2010
@@ -201,7 +201,7 @@
    .. method:: convert_field(value, conversion)
 
       Converts the value (returned by :meth:`get_field`) given a conversion type
-      (as in the tuple returned by the :meth:`parse` method.)  The default
+      (as in the tuple returned by the :meth:`parse` method).  The default
       version understands 'r' (repr) and 's' (str) conversion types.
 
 
@@ -236,6 +236,8 @@
 preceded by an exclamation point ``'!'``, and a *format_spec*, which is preceded
 by a colon ``':'``.
 
+See also the :ref:`formatspec` section.
+
 The *field_name* itself begins with either a number or a keyword.  If it's a
 number, it refers to a positional argument, and if it's a keyword it refers to a
 named keyword argument.  This can be followed by any number of index or
@@ -279,26 +281,7 @@
 format_spec are substituted before the *format_spec* string is interpreted.
 This allows the formatting of a value to be dynamically specified.
 
-For example, suppose you wanted to have a replacement field whose field width is
-determined by another variable::
-
-   "A man with two {0:{1}}".format("noses", 10)
-
-This would first evaluate the inner replacement field, making the format string
-effectively::
-
-   "A man with two {0:10}"
-
-Then the outer replacement field would be evaluated, producing::
-
-   "noses     "
-
-Which is substituted into the string, yielding::
-
-   "A man with two noses     "
-
-(The extra space is because we specified a field width of 10, and because left
-alignment is the default for strings.)
+See the :ref:`formatexamples` section for some examples.
 
 
 .. _formatspec:
@@ -308,7 +291,7 @@
 
 "Format specifications" are used within replacement fields contained within a
 format string to define how individual values are presented (see
-:ref:`formatstrings`.)  They can also be passed directly to the built-in
+:ref:`formatstrings`).  They can also be passed directly to the built-in
 :func:`format` function.  Each formattable type may define how the format
 specification is to be interpreted.
 
@@ -342,7 +325,7 @@
    | Option  | Meaning                                                  |
    +=========+==========================================================+
    | ``'<'`` | Forces the field to be left-aligned within the available |
-   |         | space (This is the default.)                             |
+   |         | space (this is the default).                             |
    +---------+----------------------------------------------------------+
    | ``'>'`` | Forces the field to be right-aligned within the          |
    |         | available space.                                         |
@@ -494,6 +477,142 @@
    +---------+----------------------------------------------------------+
 
 
+
+.. _formatexamples:
+
+Format examples
+^^^^^^^^^^^^^^^
+
+This section contains examples of the new format syntax and comparison with
+the old ``%``-formatting.
+
+In most of the cases the syntax is similar to the old ``%``-formatting, with the
+addition of the ``{}`` and with ``:`` used instead of ``%``.
+For example, ``'%03.2f'`` can be translated to ``'{0:03.2f}'``.
+
+The new format syntax also supports new and different options, shown in the
+follow examples.
+
+Accessing arguments by position::
+
+   >>> '{0}, {1}, {2}'.format('a', 'b', 'c')
+   'a, b, c'
+   >>> '{2}, {1}, {0}'.format('a', 'b', 'c')
+   'c, b, a'
+   >>> '{2}, {1}, {0}'.format(*'abc')      # unpacking argument sequence
+   'c, b, a'
+   >>> '{0}{1}{0}'.format('abra', 'cad')   # arguments' indices can be repeated
+   'abracadabra'
+
+Accessing arguments by name::
+
+   >>> 'Coordinates: {latitude}, {longitude}'.format(latitude='37.24N', longitude='-115.81W')
+   'Coordinates: 37.24N, -115.81W'
+   >>> coord = {'latitude': '37.24N', 'longitude': '-115.81W'}
+   >>> 'Coordinates: {latitude}, {longitude}'.format(**coord)
+   'Coordinates: 37.24N, -115.81W'
+
+Accessing arguments' attributes::
+
+   >>> c = 3-5j
+   >>> ('The complex number {0} is formed from the real part {0.real} '
+   ...  'and the imaginary part {0.imag}.').format(c)
+   'The complex number (3-5j) is formed from the real part 3.0 and the imaginary part -5.0.'
+   >>> class Point(object):
+   ...     def __init__(self, x, y):
+   ...         self.x, self.y = x, y
+   ...     def __str__(self):
+   ...         return 'Point({self.x}, {self.y})'.format(self=self)
+   ...
+   >>> str(Point(4, 2))
+   'Point(4, 2)'
+
+
+Accessing arguments' items::
+
+   >>> coord = (3, 5)
+   >>> 'X: {0[0]};  Y: {0[1]}'.format(coord)
+   'X: 3;  Y: 5'
+
+Replacing ``%s`` and ``%r``::
+
+   >>> "repr() shows quotes: {0!r}; str() doesn't: {1!s}".format('test1', 'test2')
+   "repr() shows quotes: 'test1'; str() doesn't: test2"
+
+Aligning the text and specifying a width::
+
+   >>> '{0:<30}'.format('left aligned')
+   'left aligned                  '
+   >>> '{0:>30}'.format('right aligned')
+   '                 right aligned'
+   >>> '{0:^30}'.format('centered')
+   '           centered           '
+   >>> '{0:*^30}'.format('centered')  # use '*' as a fill char
+   '***********centered***********'
+
+Replacing ``%+f``, ``%-f``, and ``% f`` and specifying a sign::
+
+   >>> '{0:+f}; {0:+f}'.format(3.14, -3.14)  # show it always
+   '+3.140000; -3.140000'
+   >>> '{0: f}; {0: f}'.format(3.14, -3.14)  # show a space for positive numbers
+   ' 3.140000; -3.140000'
+   >>> '{0:-f}; {0:-f}'.format(3.14, -3.14)  # show only the minus -- same as '{0:f}; {0:f}'
+   '3.140000; -3.140000'
+
+Replacing ``%x`` and ``%o`` and converting the value to different bases::
+
+   >>> # format also supports binary numbers
+   >>> "int: {0:d};  hex: {0:x};  oct: {0:o};  bin: {0:b}".format(42)
+   'int: 42;  hex: 2a;  oct: 52;  bin: 101010'
+   >>> # with 0x, 0o, or 0b as prefix:
+   >>> "int: {0:d};  hex: {0:#x};  oct: {0:#o};  bin: {0:#b}".format(42)
+   'int: 42;  hex: 0x2a;  oct: 0o52;  bin: 0b101010'
+
+Expressing a percentage::
+
+   >>> points = 19.5
+   >>> total = 22
+   >>> 'Correct answers: {0:.2%}.'.format(points/total)
+   'Correct answers: 88.64%'
+
+Using type-specific formatting::
+
+   >>> import datetime
+   >>> d = datetime.datetime(2010, 7, 4, 12, 15, 58)
+   >>> '{0:%Y-%m-%d %H:%M:%S}'.format(d)
+   '2010-07-04 12:15:58'
+
+Nesting arguments and more complex examples::
+
+   >>> for align, text in zip('<^>', ['left', 'center', 'right']):
+   ...     '{0:{align}{fill}16}'.format(text, fill=align, align=align)
+   ...
+   'left<<<<<<<<<<<<'
+   '^^^^^center^^^^^'
+   '>>>>>>>>>>>right'
+   >>>
+   >>> octets = [192, 168, 0, 1]
+   >>> '{0:02X}{1:02X}{2:02X}{3:02X}'.format(*octets)
+   'C0A80001'
+   >>> int(_, 16)
+   3232235521
+   >>>
+   >>> width = 5
+   >>> for num in range(5,12):
+   ...     for base in 'dXob':
+   ...         print '{0:{width}{base}}'.format(num, base=base, width=width),
+   ...     print
+   ...
+       5     5     5   101
+       6     6     6   110
+       7     7     7   111
+       8     8    10  1000
+       9     9    11  1001
+      10     A    12  1010
+      11     B    13  1011
+
+
+
 Template strings
 ----------------
 

From python-checkins at python.org  Sat Jul  3 01:05:28 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sat,  3 Jul 2010 01:05:28 +0200 (CEST)
Subject: [Python-checkins] r82461 - python/trunk/Lib/test/test_float.py
Message-ID: <20100702230528.2BC9FEDE7@mail.python.org>

Author: benjamin.peterson
Date: Sat Jul  3 01:05:27 2010
New Revision: 82461

Log:
don't require the presence of __getformat__ or __setformat__; use requires_IEEE_754 globally

Modified:
   python/trunk/Lib/test/test_float.py

Modified: python/trunk/Lib/test/test_float.py
==============================================================================
--- python/trunk/Lib/test/test_float.py	(original)
+++ python/trunk/Lib/test/test_float.py	Sat Jul  3 01:05:27 2010
@@ -12,8 +12,13 @@
 INF = float("inf")
 NAN = float("nan")
 
+have_getformat = hasattr(float, "__getformat__")
+requires_getformat = unittest.skipUnless(have_getformat,
+                                         "requires __getformat__")
+requires_setformat = unittest.skipUnless(hasattr(float, "__setformat__"),
+                                         "requires __setformat__")
 # decorator for skipping tests on non-IEEE 754 platforms
-requires_IEEE_754 = unittest.skipUnless(
+requires_IEEE_754 = unittest.skipUnless(have_getformat and
     float.__getformat__("double").startswith("IEEE"),
     "test requires IEEE 754 doubles")
 
@@ -357,6 +362,7 @@
             #self.assertTrue(0.0 > pow_op(-2.0, -1047) > -1e-315)
 
 
+ at requires_setformat
 class FormatFunctionsTestCase(unittest.TestCase):
 
     def setUp(self):
@@ -407,6 +413,7 @@
 # on non-IEEE platforms, attempting to unpack a bit pattern
 # representing an infinity or a NaN should raise an exception.
 
+ at requires_setformat
 class UnknownFormatTestCase(unittest.TestCase):
     def setUp(self):
         self.save_formats = {'double':float.__getformat__('double'),
@@ -439,41 +446,42 @@
 # let's also try to guarantee that -0.0 and 0.0 don't get confused.
 
 class IEEEFormatTestCase(unittest.TestCase):
-    if float.__getformat__("double").startswith("IEEE"):
-        def test_double_specials_do_unpack(self):
-            for fmt, data in [('>d', BE_DOUBLE_INF),
-                              ('>d', BE_DOUBLE_NAN),
-                              ('<d', LE_DOUBLE_INF),
-                              ('<d', LE_DOUBLE_NAN)]:
-                struct.unpack(fmt, data)
-
-    if float.__getformat__("float").startswith("IEEE"):
-        def test_float_specials_do_unpack(self):
-            for fmt, data in [('>f', BE_FLOAT_INF),
-                              ('>f', BE_FLOAT_NAN),
-                              ('<f', LE_FLOAT_INF),
-                              ('<f', LE_FLOAT_NAN)]:
-                struct.unpack(fmt, data)
-
-    if float.__getformat__("double").startswith("IEEE"):
-        def test_negative_zero(self):
-            def pos_pos():
-                return 0.0, math.atan2(0.0, -1)
-            def pos_neg():
-                return 0.0, math.atan2(-0.0, -1)
-            def neg_pos():
-                return -0.0, math.atan2(0.0, -1)
-            def neg_neg():
-                return -0.0, math.atan2(-0.0, -1)
-            self.assertEquals(pos_pos(), neg_pos())
-            self.assertEquals(pos_neg(), neg_neg())
-
-    if float.__getformat__("double").startswith("IEEE"):
-        def test_underflow_sign(self):
-            # check that -1e-1000 gives -0.0, not 0.0
-            self.assertEquals(math.atan2(-1e-1000, -1), math.atan2(-0.0, -1))
-            self.assertEquals(math.atan2(float('-1e-1000'), -1),
-                              math.atan2(-0.0, -1))
+
+    @requires_IEEE_754
+    def test_double_specials_do_unpack(self):
+        for fmt, data in [('>d', BE_DOUBLE_INF),
+                          ('>d', BE_DOUBLE_NAN),
+                          ('<d', LE_DOUBLE_INF),
+                          ('<d', LE_DOUBLE_NAN)]:
+            struct.unpack(fmt, data)
+
+    @requires_IEEE_754
+    def test_float_specials_do_unpack(self):
+        for fmt, data in [('>f', BE_FLOAT_INF),
+                          ('>f', BE_FLOAT_NAN),
+                          ('<f', LE_FLOAT_INF),
+                          ('<f', LE_FLOAT_NAN)]:
+            struct.unpack(fmt, data)
+
+    @requires_IEEE_754
+    def test_negative_zero(self):
+        def pos_pos():
+            return 0.0, math.atan2(0.0, -1)
+        def pos_neg():
+            return 0.0, math.atan2(-0.0, -1)
+        def neg_pos():
+            return -0.0, math.atan2(0.0, -1)
+        def neg_neg():
+            return -0.0, math.atan2(-0.0, -1)
+        self.assertEquals(pos_pos(), neg_pos())
+        self.assertEquals(pos_neg(), neg_neg())
+
+    @requires_IEEE_754
+    def test_underflow_sign(self):
+        # check that -1e-1000 gives -0.0, not 0.0
+        self.assertEquals(math.atan2(-1e-1000, -1), math.atan2(-0.0, -1))
+        self.assertEquals(math.atan2(float('-1e-1000'), -1),
+                          math.atan2(-0.0, -1))
 
     def test_format(self):
         # these should be rewritten to use both format(x, spec) and
@@ -530,8 +538,7 @@
         self.assertEqual('{0:f}'.format(NAN), 'nan')
         self.assertEqual('{0:F}'.format(NAN), 'NAN')
 
-    @unittest.skipUnless(float.__getformat__("double").startswith("IEEE"),
-                         "test requires IEEE 754 doubles")
+    @requires_IEEE_754
     def test_format_testfile(self):
         for line in open(format_testfile):
             if line.startswith('--'):
@@ -613,8 +620,8 @@
             self.assertEqual(s, repr(float(s)))
             self.assertEqual(negs, repr(float(negs)))
 
- at unittest.skipUnless(float.__getformat__("double").startswith("IEEE"),
-                     "test requires IEEE 754 doubles")
+
+ at requires_IEEE_754
 class RoundTestCase(unittest.TestCase):
     def test_second_argument_type(self):
         # any type with an __index__ method should be permitted as
@@ -752,8 +759,8 @@
         self.assertAlmostEqual(round(0.5e22, -22), 1e22)
         self.assertAlmostEqual(round(1.5e22, -22), 2e22)
 
-    @unittest.skipUnless(float.__getformat__("double").startswith("IEEE"),
-                         "test requires IEEE 754 doubles")
+
+    @requires_IEEE_754
     def test_format_specials(self):
         # Test formatting of nans and infs.
 

From python-checkins at python.org  Sat Jul  3 01:18:52 2010
From: python-checkins at python.org (ezio.melotti)
Date: Sat,  3 Jul 2010 01:18:52 +0200 (CEST)
Subject: [Python-checkins] r82462 - in python/branches/py3k:
	Doc/library/string.rst
Message-ID: <20100702231852.00B8EEDB4@mail.python.org>

Author: ezio.melotti
Date: Sat Jul  3 01:18:51 2010
New Revision: 82462

Log:
Merged revisions 82455,82457,82459 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/trunk

........
  r82455 | eric.smith | 2010-07-03 00:44:16 +0300 (Sat, 03 Jul 2010) | 1 line
  
  Moved period outside paren, where it belongs.
........
  r82457 | ezio.melotti | 2010-07-03 01:17:29 +0300 (Sat, 03 Jul 2010) | 1 line
  
  #9139: Add examples for str.format().
........
  r82459 | ezio.melotti | 2010-07-03 01:50:39 +0300 (Sat, 03 Jul 2010) | 1 line
  
  #9139: the thousands separator is new in 2.7.  Also add a missing variable in the example.
........


Modified:
   python/branches/py3k/   (props changed)
   python/branches/py3k/Doc/library/string.rst

Modified: python/branches/py3k/Doc/library/string.rst
==============================================================================
--- python/branches/py3k/Doc/library/string.rst	(original)
+++ python/branches/py3k/Doc/library/string.rst	Sat Jul  3 01:18:51 2010
@@ -174,7 +174,7 @@
    .. method:: convert_field(value, conversion)
 
       Converts the value (returned by :meth:`get_field`) given a conversion type
-      (as in the tuple returned by the :meth:`parse` method.)  The default
+      (as in the tuple returned by the :meth:`parse` method).  The default
       version understands 'r' (repr) and 's' (str) conversion types.
 
 
@@ -185,7 +185,7 @@
 
 The :meth:`str.format` method and the :class:`Formatter` class share the same
 syntax for format strings (although in the case of :class:`Formatter`,
-subclasses can define their own format string syntax.)
+subclasses can define their own format string syntax).
 
 Format strings contain "replacement fields" surrounded by curly braces ``{}``.
 Anything that is not contained in braces is considered literal text, which is
@@ -211,6 +211,8 @@
 preceded by an exclamation point ``'!'``, and a *format_spec*, which is preceded
 by a colon ``':'``.  These specify a non-default format for the replacement value.
 
+See also the :ref:`formatspec` section.
+
 The *field_name* itself begins with an *arg_name* that is either either a number or a
 keyword.  If it's a number, it refers to a positional argument, and if it's a keyword,
 it refers to a named keyword argument.  If the numerical arg_names in a format string
@@ -221,6 +223,10 @@
 attribute using :func:`getattr`, while an expression of the form ``'[index]'``
 does an index lookup using :func:`__getitem__`.
 
+.. versionchanged:: 3.1
+   The positional argument specifiers can be omitted, so ``'{} {}'`` is
+   equivalent to ``'{0} {1}'``.
+
 Some simple format string examples::
 
    "First, thou shalt count to {0}" # References first positional argument
@@ -261,26 +267,7 @@
 format_spec are substituted before the *format_spec* string is interpreted.
 This allows the formatting of a value to be dynamically specified.
 
-For example, suppose you wanted to have a replacement field whose field width is
-determined by another variable::
-
-   "A man with two {0:{1}}".format("noses", 10)
-
-This would first evaluate the inner replacement field, making the format string
-effectively::
-
-   "A man with two {0:10}"
-
-Then the outer replacement field would be evaluated, producing::
-
-   "noses     "
-
-Which is substituted into the string, yielding::
-
-   "A man with two noses     "
-
-(The extra space is because we specified a field width of 10, and because left
-alignment is the default for strings.)
+See the :ref:`formatexamples` section for some examples.
 
 
 .. _formatspec:
@@ -290,7 +277,7 @@
 
 "Format specifications" are used within replacement fields contained within a
 format string to define how individual values are presented (see
-:ref:`formatstrings`.)  They can also be passed directly to the built-in
+:ref:`formatstrings`).  They can also be passed directly to the built-in
 :func:`format` function.  Each formattable type may define how the format
 specification is to be interpreted.
 
@@ -324,7 +311,7 @@
    | Option  | Meaning                                                  |
    +=========+==========================================================+
    | ``'<'`` | Forces the field to be left-aligned within the available |
-   |         | space (This is the default.)                             |
+   |         | space (this is the default).                             |
    +---------+----------------------------------------------------------+
    | ``'>'`` | Forces the field to be right-aligned within the          |
    |         | available space.                                         |
@@ -366,6 +353,9 @@
 For a locale aware separator, use the ``'n'`` integer presentation type
 instead.
 
+.. versionchanged:: 3.1
+   Added the ``','`` option (see also :pep:`378`).
+
 *width* is a decimal integer defining the minimum field width.  If not
 specified, then the field width will be determined by the content.
 
@@ -484,6 +474,147 @@
    +---------+----------------------------------------------------------+
 
 
+.. _formatexamples:
+
+Format examples
+^^^^^^^^^^^^^^^
+
+This section contains examples of the new format syntax and comparison with
+the old ``%``-formatting.
+
+In most of the cases the syntax is similar to the old ``%``-formatting, with the
+addition of the ``{}`` and with ``:`` used instead of ``%``.
+For example, ``'%03.2f'`` can be translated to ``'{:03.2f}'``.
+
+The new format syntax also supports new and different options, shown in the
+follow examples.
+
+Accessing arguments by position::
+
+   >>> '{0}, {1}, {2}'.format('a', 'b', 'c')
+   'a, b, c'
+   >>> '{}, {}, {}'.format('a', 'b', 'c')  # 3.1+ only
+   'a, b, c'
+   >>> '{2}, {1}, {0}'.format('a', 'b', 'c')
+   'c, b, a'
+   >>> '{2}, {1}, {0}'.format(*'abc')      # unpacking argument sequence
+   'c, b, a'
+   >>> '{0}{1}{0}'.format('abra', 'cad')   # arguments' indices can be repeated
+   'abracadabra'
+
+Accessing arguments by name::
+
+   >>> 'Coordinates: {latitude}, {longitude}'.format(latitude='37.24N', longitude='-115.81W')
+   'Coordinates: 37.24N, -115.81W'
+   >>> coord = {'latitude': '37.24N', 'longitude': '-115.81W'}
+   >>> 'Coordinates: {latitude}, {longitude}'.format(**coord)
+   'Coordinates: 37.24N, -115.81W'
+
+Accessing arguments' attributes::
+
+   >>> c = 3-5j
+   >>> ('The complex number {0} is formed from the real part {0.real} '
+   ...  'and the imaginary part {0.imag}.').format(c)
+   'The complex number (3-5j) is formed from the real part 3.0 and the imaginary part -5.0.'
+   >>> class Point:
+   ...     def __init__(self, x, y):
+   ...         self.x, self.y = x, y
+   ...     def __str__(self):
+   ...         return 'Point({self.x}, {self.y})'.format(self=self)
+   ...
+   >>> str(Point(4, 2))
+   'Point(4, 2)'
+
+Accessing arguments' items::
+
+   >>> coord = (3, 5)
+   >>> 'X: {0[0]};  Y: {0[1]}'.format(coord)
+   'X: 3;  Y: 5'
+
+Replacing ``%s`` and ``%r``::
+
+   >>> "repr() shows quotes: {!r}; str() doesn't: {!s}".format('test1', 'test2')
+   "repr() shows quotes: 'test1'; str() doesn't: test2"
+
+Aligning the text and specifying a width::
+
+   >>> '{:<30}'.format('left aligned')
+   'left aligned                  '
+   >>> '{:>30}'.format('right aligned')
+   '                 right aligned'
+   >>> '{:^30}'.format('centered')
+   '           centered           '
+   >>> '{:*^30}'.format('centered')  # use '*' as a fill char
+   '***********centered***********'
+
+Replacing ``%+f``, ``%-f``, and ``% f`` and specifying a sign::
+
+   >>> '{:+f}; {:+f}'.format(3.14, -3.14)  # show it always
+   '+3.140000; -3.140000'
+   >>> '{: f}; {: f}'.format(3.14, -3.14)  # show a space for positive numbers
+   ' 3.140000; -3.140000'
+   >>> '{:-f}; {:-f}'.format(3.14, -3.14)  # show only the minus -- same as '{:f}; {:f}'
+   '3.140000; -3.140000'
+
+Replacing ``%x`` and ``%o`` and converting the value to different bases::
+
+   >>> # format also supports binary numbers
+   >>> "int: {0:d};  hex: {0:x};  oct: {0:o};  bin: {0:b}".format(42)
+   'int: 42;  hex: 2a;  oct: 52;  bin: 101010'
+   >>> # with 0x, 0o, or 0b as prefix:
+   >>> "int: {0:d};  hex: {0:#x};  oct: {0:#o};  bin: {0:#b}".format(42)
+   'int: 42;  hex: 0x2a;  oct: 0o52;  bin: 0b101010'
+
+Using the comma as a thousands separator::
+
+   >>> '{:,}'.format(1234567890)
+   '1,234,567,890'
+
+Expressing a percentage::
+
+   >>> points = 19
+   >>> total = 22
+   >>> 'Correct answers: {:.2%}.'.format(points/total)
+   'Correct answers: 86.36%'
+
+Using type-specific formatting::
+
+   >>> import datetime
+   >>> d = datetime.datetime(2010, 7, 4, 12, 15, 58)
+   >>> '{:%Y-%m-%d %H:%M:%S}'.format(d)
+   '2010-07-04 12:15:58'
+
+Nesting arguments and more complex examples::
+
+   >>> for align, text in zip('<^>', ['left', 'center', 'right']):
+   ...     '{0:{align}{fill}16}'.format(text, fill=align, align=align)
+   ...
+   'left<<<<<<<<<<<<'
+   '^^^^^center^^^^^'
+   '>>>>>>>>>>>right'
+   >>>
+   >>> octets = [192, 168, 0, 1]
+   >>> '{:02X}{:02X}{:02X}{:02X}'.format(*octets)
+   'C0A80001'
+   >>> int(_, 16)
+   3232235521
+   >>>
+   >>> width = 5
+   >>> for num in range(5,12):
+   ...     for base in 'dXob':
+   ...         print('{0:{width}{base}}'.format(num, base=base, width=width), end=' ')
+   ...     print()
+   ...
+       5     5     5   101
+       6     6     6   110
+       7     7     7   111
+       8     8    10  1000
+       9     9    11  1001
+      10     A    12  1010
+      11     B    13  1011
+
+
+
 .. _template-strings:
 
 Template strings

From python-checkins at python.org  Sat Jul  3 01:22:04 2010
From: python-checkins at python.org (ezio.melotti)
Date: Sat,  3 Jul 2010 01:22:04 +0200 (CEST)
Subject: [Python-checkins] r82463 - in python/branches/release31-maint:
	Doc/library/string.rst
Message-ID: <20100702232204.0A0D4EF9F@mail.python.org>

Author: ezio.melotti
Date: Sat Jul  3 01:22:03 2010
New Revision: 82463

Log:
Merged revisions 82462 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

................
  r82462 | ezio.melotti | 2010-07-03 02:18:51 +0300 (Sat, 03 Jul 2010) | 17 lines
  
  Merged revisions 82455,82457,82459 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82455 | eric.smith | 2010-07-03 00:44:16 +0300 (Sat, 03 Jul 2010) | 1 line
    
    Moved period outside paren, where it belongs.
  ........
    r82457 | ezio.melotti | 2010-07-03 01:17:29 +0300 (Sat, 03 Jul 2010) | 1 line
    
    #9139: Add examples for str.format().
  ........
    r82459 | ezio.melotti | 2010-07-03 01:50:39 +0300 (Sat, 03 Jul 2010) | 1 line
    
    #9139: the thousands separator is new in 2.7.  Also add a missing variable in the example.
  ........
................


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Doc/library/string.rst

Modified: python/branches/release31-maint/Doc/library/string.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/string.rst	(original)
+++ python/branches/release31-maint/Doc/library/string.rst	Sat Jul  3 01:22:03 2010
@@ -174,7 +174,7 @@
    .. method:: convert_field(value, conversion)
 
       Converts the value (returned by :meth:`get_field`) given a conversion type
-      (as in the tuple returned by the :meth:`parse` method.)  The default
+      (as in the tuple returned by the :meth:`parse` method).  The default
       version understands 'r' (repr) and 's' (str) conversion types.
 
 
@@ -185,7 +185,7 @@
 
 The :meth:`str.format` method and the :class:`Formatter` class share the same
 syntax for format strings (although in the case of :class:`Formatter`,
-subclasses can define their own format string syntax.)
+subclasses can define their own format string syntax).
 
 Format strings contain "replacement fields" surrounded by curly braces ``{}``.
 Anything that is not contained in braces is considered literal text, which is
@@ -211,6 +211,8 @@
 preceded by an exclamation point ``'!'``, and a *format_spec*, which is preceded
 by a colon ``':'``.  These specify a non-default format for the replacement value.
 
+See also the :ref:`formatspec` section.
+
 The *field_name* itself begins with an *arg_name* that is either either a number or a
 keyword.  If it's a number, it refers to a positional argument, and if it's a keyword,
 it refers to a named keyword argument.  If the numerical arg_names in a format string
@@ -221,6 +223,10 @@
 attribute using :func:`getattr`, while an expression of the form ``'[index]'``
 does an index lookup using :func:`__getitem__`.
 
+.. versionchanged:: 3.1
+   The positional argument specifiers can be omitted, so ``'{} {}'`` is
+   equivalent to ``'{0} {1}'``.
+
 Some simple format string examples::
 
    "First, thou shalt count to {0}" # References first positional argument
@@ -261,26 +267,7 @@
 format_spec are substituted before the *format_spec* string is interpreted.
 This allows the formatting of a value to be dynamically specified.
 
-For example, suppose you wanted to have a replacement field whose field width is
-determined by another variable::
-
-   "A man with two {0:{1}}".format("noses", 10)
-
-This would first evaluate the inner replacement field, making the format string
-effectively::
-
-   "A man with two {0:10}"
-
-Then the outer replacement field would be evaluated, producing::
-
-   "noses     "
-
-Which is substituted into the string, yielding::
-
-   "A man with two noses     "
-
-(The extra space is because we specified a field width of 10, and because left
-alignment is the default for strings.)
+See the :ref:`formatexamples` section for some examples.
 
 
 .. _formatspec:
@@ -290,7 +277,7 @@
 
 "Format specifications" are used within replacement fields contained within a
 format string to define how individual values are presented (see
-:ref:`formatstrings`.)  They can also be passed directly to the built-in
+:ref:`formatstrings`).  They can also be passed directly to the built-in
 :func:`format` function.  Each formattable type may define how the format
 specification is to be interpreted.
 
@@ -324,7 +311,7 @@
    | Option  | Meaning                                                  |
    +=========+==========================================================+
    | ``'<'`` | Forces the field to be left-aligned within the available |
-   |         | space (This is the default.)                             |
+   |         | space (this is the default).                             |
    +---------+----------------------------------------------------------+
    | ``'>'`` | Forces the field to be right-aligned within the          |
    |         | available space.                                         |
@@ -366,6 +353,9 @@
 For a locale aware separator, use the ``'n'`` integer presentation type
 instead.
 
+.. versionchanged:: 3.1
+   Added the ``','`` option (see also :pep:`378`).
+
 *width* is a decimal integer defining the minimum field width.  If not
 specified, then the field width will be determined by the content.
 
@@ -484,6 +474,147 @@
    +---------+----------------------------------------------------------+
 
 
+.. _formatexamples:
+
+Format examples
+^^^^^^^^^^^^^^^
+
+This section contains examples of the new format syntax and comparison with
+the old ``%``-formatting.
+
+In most of the cases the syntax is similar to the old ``%``-formatting, with the
+addition of the ``{}`` and with ``:`` used instead of ``%``.
+For example, ``'%03.2f'`` can be translated to ``'{:03.2f}'``.
+
+The new format syntax also supports new and different options, shown in the
+follow examples.
+
+Accessing arguments by position::
+
+   >>> '{0}, {1}, {2}'.format('a', 'b', 'c')
+   'a, b, c'
+   >>> '{}, {}, {}'.format('a', 'b', 'c')  # 3.1+ only
+   'a, b, c'
+   >>> '{2}, {1}, {0}'.format('a', 'b', 'c')
+   'c, b, a'
+   >>> '{2}, {1}, {0}'.format(*'abc')      # unpacking argument sequence
+   'c, b, a'
+   >>> '{0}{1}{0}'.format('abra', 'cad')   # arguments' indices can be repeated
+   'abracadabra'
+
+Accessing arguments by name::
+
+   >>> 'Coordinates: {latitude}, {longitude}'.format(latitude='37.24N', longitude='-115.81W')
+   'Coordinates: 37.24N, -115.81W'
+   >>> coord = {'latitude': '37.24N', 'longitude': '-115.81W'}
+   >>> 'Coordinates: {latitude}, {longitude}'.format(**coord)
+   'Coordinates: 37.24N, -115.81W'
+
+Accessing arguments' attributes::
+
+   >>> c = 3-5j
+   >>> ('The complex number {0} is formed from the real part {0.real} '
+   ...  'and the imaginary part {0.imag}.').format(c)
+   'The complex number (3-5j) is formed from the real part 3.0 and the imaginary part -5.0.'
+   >>> class Point:
+   ...     def __init__(self, x, y):
+   ...         self.x, self.y = x, y
+   ...     def __str__(self):
+   ...         return 'Point({self.x}, {self.y})'.format(self=self)
+   ...
+   >>> str(Point(4, 2))
+   'Point(4, 2)'
+
+Accessing arguments' items::
+
+   >>> coord = (3, 5)
+   >>> 'X: {0[0]};  Y: {0[1]}'.format(coord)
+   'X: 3;  Y: 5'
+
+Replacing ``%s`` and ``%r``::
+
+   >>> "repr() shows quotes: {!r}; str() doesn't: {!s}".format('test1', 'test2')
+   "repr() shows quotes: 'test1'; str() doesn't: test2"
+
+Aligning the text and specifying a width::
+
+   >>> '{:<30}'.format('left aligned')
+   'left aligned                  '
+   >>> '{:>30}'.format('right aligned')
+   '                 right aligned'
+   >>> '{:^30}'.format('centered')
+   '           centered           '
+   >>> '{:*^30}'.format('centered')  # use '*' as a fill char
+   '***********centered***********'
+
+Replacing ``%+f``, ``%-f``, and ``% f`` and specifying a sign::
+
+   >>> '{:+f}; {:+f}'.format(3.14, -3.14)  # show it always
+   '+3.140000; -3.140000'
+   >>> '{: f}; {: f}'.format(3.14, -3.14)  # show a space for positive numbers
+   ' 3.140000; -3.140000'
+   >>> '{:-f}; {:-f}'.format(3.14, -3.14)  # show only the minus -- same as '{:f}; {:f}'
+   '3.140000; -3.140000'
+
+Replacing ``%x`` and ``%o`` and converting the value to different bases::
+
+   >>> # format also supports binary numbers
+   >>> "int: {0:d};  hex: {0:x};  oct: {0:o};  bin: {0:b}".format(42)
+   'int: 42;  hex: 2a;  oct: 52;  bin: 101010'
+   >>> # with 0x, 0o, or 0b as prefix:
+   >>> "int: {0:d};  hex: {0:#x};  oct: {0:#o};  bin: {0:#b}".format(42)
+   'int: 42;  hex: 0x2a;  oct: 0o52;  bin: 0b101010'
+
+Using the comma as a thousands separator::
+
+   >>> '{:,}'.format(1234567890)
+   '1,234,567,890'
+
+Expressing a percentage::
+
+   >>> points = 19
+   >>> total = 22
+   >>> 'Correct answers: {:.2%}.'.format(points/total)
+   'Correct answers: 86.36%'
+
+Using type-specific formatting::
+
+   >>> import datetime
+   >>> d = datetime.datetime(2010, 7, 4, 12, 15, 58)
+   >>> '{:%Y-%m-%d %H:%M:%S}'.format(d)
+   '2010-07-04 12:15:58'
+
+Nesting arguments and more complex examples::
+
+   >>> for align, text in zip('<^>', ['left', 'center', 'right']):
+   ...     '{0:{align}{fill}16}'.format(text, fill=align, align=align)
+   ...
+   'left<<<<<<<<<<<<'
+   '^^^^^center^^^^^'
+   '>>>>>>>>>>>right'
+   >>>
+   >>> octets = [192, 168, 0, 1]
+   >>> '{:02X}{:02X}{:02X}{:02X}'.format(*octets)
+   'C0A80001'
+   >>> int(_, 16)
+   3232235521
+   >>>
+   >>> width = 5
+   >>> for num in range(5,12):
+   ...     for base in 'dXob':
+   ...         print('{0:{width}{base}}'.format(num, base=base, width=width), end=' ')
+   ...     print()
+   ...
+       5     5     5   101
+       6     6     6   110
+       7     7     7   111
+       8     8    10  1000
+       9     9    11  1001
+      10     A    12  1010
+      11     B    13  1011
+
+
+
 .. _template-strings:
 
 Template strings

From solipsis at pitrou.net  Sat Jul  3 01:24:54 2010
From: solipsis at pitrou.net (solipsis at pitrou.net)
Date: Sat,  3 Jul 2010 01:24:54 +0200 (CEST)
Subject: [Python-checkins] Daily py3k reference leaks (r82452): sum=0
Message-ID: <20100702232454.3921F1770A@ns6635.ovh.net>

py3k results for svn r82452 (hg cset 799649587872)
--------------------------------------------------



Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/antoine/py3k/refleaks/reflog8d1xYz', '-x']

From python-checkins at python.org  Sat Jul  3 01:25:44 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sat,  3 Jul 2010 01:25:44 +0200 (CEST)
Subject: [Python-checkins] r82464 - in python/branches/py3k:
	Lib/test/test_float.py
Message-ID: <20100702232544.950A4EE98A@mail.python.org>

Author: benjamin.peterson
Date: Sat Jul  3 01:25:44 2010
New Revision: 82464

Log:
Merged revisions 82461 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/trunk

........
  r82461 | benjamin.peterson | 2010-07-02 18:05:27 -0500 (Fri, 02 Jul 2010) | 1 line
  
  don't require the presence of __getformat__ or __setformat__; use requires_IEEE_754 globally
........


Modified:
   python/branches/py3k/   (props changed)
   python/branches/py3k/Lib/test/test_float.py

Modified: python/branches/py3k/Lib/test/test_float.py
==============================================================================
--- python/branches/py3k/Lib/test/test_float.py	(original)
+++ python/branches/py3k/Lib/test/test_float.py	Sat Jul  3 01:25:44 2010
@@ -11,8 +11,13 @@
 INF = float("inf")
 NAN = float("nan")
 
+have_getformat = hasattr(float, "__getformat__")
+requires_getformat = unittest.skipUnless(have_getformat,
+                                         "requires __getformat__")
+requires_setformat = unittest.skipUnless(hasattr(float, "__setformat__"),
+                                         "requires __setformat__")
 # decorator for skipping tests on non-IEEE 754 platforms
-requires_IEEE_754 = unittest.skipUnless(
+requires_IEEE_754 = unittest.skipUnless(have_getformat and
     float.__getformat__("double").startswith("IEEE"),
     "test requires IEEE 754 doubles")
 
@@ -377,6 +382,7 @@
             #self.assertTrue(0.0 > pow_op(-2.0, -1047) > -1e-315)
 
 
+ at requires_setformat
 class FormatFunctionsTestCase(unittest.TestCase):
 
     def setUp(self):
@@ -427,6 +433,7 @@
 # on non-IEEE platforms, attempting to unpack a bit pattern
 # representing an infinity or a NaN should raise an exception.
 
+ at requires_setformat
 class UnknownFormatTestCase(unittest.TestCase):
     def setUp(self):
         self.save_formats = {'double':float.__getformat__('double'),
@@ -459,36 +466,25 @@
 # let's also try to guarantee that -0.0 and 0.0 don't get confused.
 
 class IEEEFormatTestCase(unittest.TestCase):
-    if float.__getformat__("double").startswith("IEEE"):
-        def test_double_specials_do_unpack(self):
-            for fmt, data in [('>d', BE_DOUBLE_INF),
-                              ('>d', BE_DOUBLE_NAN),
-                              ('<d', LE_DOUBLE_INF),
-                              ('<d', LE_DOUBLE_NAN)]:
-                struct.unpack(fmt, data)
-
-    if float.__getformat__("float").startswith("IEEE"):
-        def test_float_specials_do_unpack(self):
-            for fmt, data in [('>f', BE_FLOAT_INF),
-                              ('>f', BE_FLOAT_NAN),
-                              ('<f', LE_FLOAT_INF),
-                              ('<f', LE_FLOAT_NAN)]:
-                struct.unpack(fmt, data)
-
-    if float.__getformat__("double").startswith("IEEE"):
-        def test_negative_zero(self):
-            def pos_pos():
-                return 0.0, math.atan2(0.0, -1)
-            def pos_neg():
-                return 0.0, math.atan2(-0.0, -1)
-            def neg_pos():
-                return -0.0, math.atan2(0.0, -1)
-            def neg_neg():
-                return -0.0, math.atan2(-0.0, -1)
-            self.assertEquals(pos_pos(), neg_pos())
-            self.assertEquals(pos_neg(), neg_neg())
+
+    @requires_IEEE_754
+    def test_double_specials_do_unpack(self):
+        for fmt, data in [('>d', BE_DOUBLE_INF),
+                          ('>d', BE_DOUBLE_NAN),
+                          ('<d', LE_DOUBLE_INF),
+                          ('<d', LE_DOUBLE_NAN)]:
+            struct.unpack(fmt, data)
+
+    @requires_IEEE_754
+    def test_float_specials_do_unpack(self):
+        for fmt, data in [('>f', BE_FLOAT_INF),
+                          ('>f', BE_FLOAT_NAN),
+                          ('<f', LE_FLOAT_INF),
+                          ('<f', LE_FLOAT_NAN)]:
+            struct.unpack(fmt, data)
 
 class FormatTestCase(unittest.TestCase):
+
     def test_format(self):
         # these should be rewritten to use both format(x, spec) and
         # x.__format__(spec)
@@ -542,8 +538,7 @@
         self.assertEqual(format(INF, 'f'), 'inf')
         self.assertEqual(format(INF, 'F'), 'INF')
 
-    @unittest.skipUnless(float.__getformat__("double").startswith("IEEE"),
-                         "test requires IEEE 754 doubles")
+    @requires_IEEE_754
     def test_format_testfile(self):
         for line in open(format_testfile):
             if line.startswith('--'):
@@ -623,9 +618,10 @@
             self.assertEqual(s, repr(float(s)))
             self.assertEqual(negs, repr(float(negs)))
 
+
+ at requires_IEEE_754
 class RoundTestCase(unittest.TestCase):
-    @unittest.skipUnless(float.__getformat__("double").startswith("IEEE"),
-                         "test requires IEEE 754 doubles")
+
     def test_inf_nan(self):
         self.assertRaises(OverflowError, round, INF)
         self.assertRaises(OverflowError, round, -INF)
@@ -635,8 +631,6 @@
         self.assertRaises(TypeError, round, NAN, "ceci n'est pas un integer")
         self.assertRaises(TypeError, round, -0.0, 1j)
 
-    @unittest.skipUnless(float.__getformat__("double").startswith("IEEE"),
-                         "test requires IEEE 754 doubles")
     def test_large_n(self):
         for n in [324, 325, 400, 2**31-1, 2**31, 2**32, 2**100]:
             self.assertEqual(round(123.456, n), 123.456)
@@ -649,8 +643,6 @@
         self.assertEqual(round(1e150, 309), 1e150)
         self.assertEqual(round(1.4e-315, 315), 1e-315)
 
-    @unittest.skipUnless(float.__getformat__("double").startswith("IEEE"),
-                         "test requires IEEE 754 doubles")
     def test_small_n(self):
         for n in [-308, -309, -400, 1-2**31, -2**31, -2**31-1, -2**100]:
             self.assertEqual(round(123.456, n), 0.0)
@@ -658,8 +650,6 @@
             self.assertEqual(round(1e300, n), 0.0)
             self.assertEqual(round(1e-320, n), 0.0)
 
-    @unittest.skipUnless(float.__getformat__("double").startswith("IEEE"),
-                         "test requires IEEE 754 doubles")
     def test_overflow(self):
         self.assertRaises(OverflowError, round, 1.6e308, -308)
         self.assertRaises(OverflowError, round, -1.7e308, -308)
@@ -707,9 +697,6 @@
             self.assertEqual(float(format(x, '.2f')), round(x, 2))
             self.assertEqual(float(format(x, '.3f')), round(x, 3))
 
-
-    @unittest.skipUnless(float.__getformat__("double").startswith("IEEE"),
-                         "test requires IEEE 754 doubles")
     def test_format_specials(self):
         # Test formatting of nans and infs.
 

From python-checkins at python.org  Sat Jul  3 03:32:48 2010
From: python-checkins at python.org (brett.cannon)
Date: Sat,  3 Jul 2010 03:32:48 +0200 (CEST)
Subject: [Python-checkins] r82465 -
	python/branches/py3k/Lib/test/test_import.py
Message-ID: <20100703013248.A037FEE98E@mail.python.org>

Author: brett.cannon
Date: Sat Jul  3 03:32:48 2010
New Revision: 82465

Log:
Make test_import a little bit more robust for cleaning up after itself in the
face of a failure.


Modified:
   python/branches/py3k/Lib/test/test_import.py

Modified: python/branches/py3k/Lib/test/test_import.py
==============================================================================
--- python/branches/py3k/Lib/test/test_import.py	(original)
+++ python/branches/py3k/Lib/test/test_import.py	Sat Jul  3 03:32:48 2010
@@ -147,22 +147,24 @@
         filename = module + '.py'
 
         # Create a file with a list of 65000 elements.
-        with open(filename, 'w+') as f:
+        with open(filename, 'w') as f:
             f.write('d = [\n')
             for i in range(65000):
                 f.write('"",\n')
             f.write(']')
 
-        # Compile & remove .py file; we only need .pyc (or .pyo).
-        # Bytecode must be relocated from the PEP 3147 bytecode-only location.
-        py_compile.compile(filename)
-        unlink(filename)
-        make_legacy_pyc(filename)
+        try:
+            # Compile & remove .py file; we only need .pyc (or .pyo).
+            # Bytecode must be relocated from the PEP 3147 bytecode-only location.
+            py_compile.compile(filename)
+        finally:
+            unlink(filename)
 
         # Need to be able to load from current dir.
         sys.path.append('')
 
         try:
+            make_legacy_pyc(filename)
             # This used to crash.
             exec('import ' + module)
         finally:

From alexander.belopolsky at gmail.com  Sat Jul  3 04:12:35 2010
From: alexander.belopolsky at gmail.com (Alexander Belopolsky)
Date: Fri, 2 Jul 2010 22:12:35 -0400
Subject: [Python-checkins] r81681 - python/trunk/Modules/datetimemodule.c
In-Reply-To: <20100604015127.1C135EEBFB@mail.python.org>
References: <20100604015127.1C135EEBFB@mail.python.org>
Message-ID: <AANLkTin7jluKBNv2xGKYaTA2bxnMkSjiNqnQMIWUh06d@mail.gmail.com>

On Thu, Jun 3, 2010 at 9:51 PM, sean.reifschneider
<python-checkins at python.org> wrote:
> Author: sean.reifschneider
> Date: Fri Jun ?4 03:51:26 2010
> New Revision: 81681
>
> Log:
> Issue8810: Clearing up docstring for tzinfo.utcoffset.
..
> ? ? {"fromutc", ? ? ? ? (PyCFunction)tzinfo_fromutc, ? ? ? ? ? ?METH_O,
> - ? ? PyDoc_STR("datetime in UTC -> datetime in local time.")},
> + ? ? PyDoc_STR("datetime -> timedelta showing offset from UTC, negative "
> + ? ? ? ? ? "values indicating West of UTC")},

This is wrong.  The fromutc method returns datetime, not timedelta.
See http://docs.python.org/dev/library/datetime.html#datetime.tzinfo.fromutc

From python-checkins at python.org  Sat Jul  3 05:27:13 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Sat,  3 Jul 2010 05:27:13 +0200 (CEST)
Subject: [Python-checkins] r82466 - python/trunk/Modules/datetimemodule.c
Message-ID: <20100703032713.1D5F1F1AE@mail.python.org>

Author: alexander.belopolsky
Date: Sat Jul  3 05:27:12 2010
New Revision: 82466

Log:
Revert r81681 (issue 8810).

Modified:
   python/trunk/Modules/datetimemodule.c

Modified: python/trunk/Modules/datetimemodule.c
==============================================================================
--- python/trunk/Modules/datetimemodule.c	(original)
+++ python/trunk/Modules/datetimemodule.c	Sat Jul  3 05:27:12 2010
@@ -3026,8 +3026,7 @@
      PyDoc_STR("datetime -> DST offset in minutes east of UTC.")},
 
     {"fromutc",         (PyCFunction)tzinfo_fromutc,            METH_O,
-     PyDoc_STR("datetime -> timedelta showing offset from UTC, negative "
-           "values indicating West of UTC")},
+     PyDoc_STR("datetime in UTC -> datetime in local time.")},
 
     {"__reduce__",  (PyCFunction)tzinfo_reduce,             METH_NOARGS,
      PyDoc_STR("-> (cls, state)")},

From python-checkins at python.org  Sat Jul  3 05:35:27 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Sat,  3 Jul 2010 05:35:27 +0200 (CEST)
Subject: [Python-checkins] r82467 - in python/branches/py3k:
	Modules/datetimemodule.c
Message-ID: <20100703033527.5FDA8EE990@mail.python.org>

Author: alexander.belopolsky
Date: Sat Jul  3 05:35:27 2010
New Revision: 82467

Log:
Merged revisions 82466 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/trunk

........
  r82466 | alexander.belopolsky | 2010-07-02 23:27:12 -0400 (Fri, 02 Jul 2010) | 1 line
  
  Revert r81681 (issue 8810).
........


Modified:
   python/branches/py3k/   (props changed)
   python/branches/py3k/Modules/datetimemodule.c

Modified: python/branches/py3k/Modules/datetimemodule.c
==============================================================================
--- python/branches/py3k/Modules/datetimemodule.c	(original)
+++ python/branches/py3k/Modules/datetimemodule.c	Sat Jul  3 05:35:27 2010
@@ -3277,8 +3277,7 @@
      PyDoc_STR("datetime -> DST offset in minutes east of UTC.")},
 
     {"fromutc",         (PyCFunction)tzinfo_fromutc,            METH_O,
-     PyDoc_STR("datetime -> timedelta showing offset from UTC, negative "
-           "values indicating West of UTC")},
+     PyDoc_STR("datetime in UTC -> datetime in local time.")},
 
     {"__reduce__",  (PyCFunction)tzinfo_reduce,             METH_NOARGS,
      PyDoc_STR("-> (cls, state)")},

From python-checkins at python.org  Sat Jul  3 06:52:19 2010
From: python-checkins at python.org (ezio.melotti)
Date: Sat,  3 Jul 2010 06:52:19 +0200 (CEST)
Subject: [Python-checkins] r82468 -
	python/branches/py3k/Objects/unicodeobject.c
Message-ID: <20100703045219.96D74EFCB@mail.python.org>

Author: ezio.melotti
Date: Sat Jul  3 06:52:19 2010
New Revision: 82468

Log:
Update comment about surrogates.

Modified:
   python/branches/py3k/Objects/unicodeobject.c

Modified: python/branches/py3k/Objects/unicodeobject.c
==============================================================================
--- python/branches/py3k/Objects/unicodeobject.c	(original)
+++ python/branches/py3k/Objects/unicodeobject.c	Sat Jul  3 06:52:19 2010
@@ -2450,11 +2450,11 @@
             break;
 
         case 3:
-            /* XXX: surrogates shouldn't be valid UTF-8!
-               see http://www.unicode.org/versions/Unicode5.2.0/ch03.pdf
-               (table 3-7) and http://www.rfc-editor.org/rfc/rfc3629.txt
-               Uncomment the 2 lines below to make them invalid,
-               codepoints: d800-dfff; UTF-8: \xed\xa0\x80-\xed\xbf\xbf. */
+            /* Decoding UTF-8 sequences in range \xed\xa0\x80-\xed\xbf\xbf
+               will result in surrogates in range d800-dfff. Surrogates are
+               not valid UTF-8 so they are rejected.
+               See http://www.unicode.org/versions/Unicode5.2.0/ch03.pdf
+               (table 3-7) and http://www.rfc-editor.org/rfc/rfc3629.txt */
             if ((s[1] & 0xc0) != 0x80 ||
                 (s[2] & 0xc0) != 0x80 ||
                 ((unsigned char)s[0] == 0xE0 &&

From python-checkins at python.org  Sat Jul  3 07:18:50 2010
From: python-checkins at python.org (ezio.melotti)
Date: Sat,  3 Jul 2010 07:18:50 +0200 (CEST)
Subject: [Python-checkins] r82469 - in python/branches/release31-maint:
	Lib/test/test_codeccallbacks.py Lib/test/test_unicode.py
	Misc/NEWS Objects/unicodeobject.c
Message-ID: <20100703051850.DA94BF2CA@mail.python.org>

Author: ezio.melotti
Date: Sat Jul  3 07:18:50 2010
New Revision: 82469

Log:
Merged revisions 82413,82468 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82413 | ezio.melotti | 2010-07-01 10:32:02 +0300 (Thu, 01 Jul 2010) | 13 lines
  
  Update PyUnicode_DecodeUTF8 from RFC 2279 to RFC 3629.
  
  1) #8271: when a byte sequence is invalid, only the start byte and all the
     valid continuation bytes are now replaced by U+FFFD, instead of replacing
     the number of bytes specified by the start byte.
     See http://www.unicode.org/versions/Unicode5.2.0/ch03.pdf (pages 94-95);
  2) 5- and 6-bytes-long UTF-8 sequences are now considered invalid (no changes
     in behavior);
  3) Change the error messages "unexpected code byte" to "invalid start byte"
     and "invalid data" to "invalid continuation byte";
  4) Add an extensive set of tests in test_unicode;
  5) Fix test_codeccallbacks because it was failing after this change.
........
  r82468 | ezio.melotti | 2010-07-03 07:52:19 +0300 (Sat, 03 Jul 2010) | 1 line
  
  Update comment about surrogates.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/test/test_codeccallbacks.py
   python/branches/release31-maint/Lib/test/test_unicode.py
   python/branches/release31-maint/Misc/NEWS
   python/branches/release31-maint/Objects/unicodeobject.c

Modified: python/branches/release31-maint/Lib/test/test_codeccallbacks.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_codeccallbacks.py	(original)
+++ python/branches/release31-maint/Lib/test/test_codeccallbacks.py	Sat Jul  3 07:18:50 2010
@@ -153,28 +153,30 @@
             sout += bytes("\\U%08x" % sys.maxunicode, "ascii")
         self.assertEqual(sin.encode("iso-8859-15", "backslashreplace"), sout)
 
-    def test_decoderelaxedutf8(self):
-        # This is the test for a decoding callback handler,
-        # that relaxes the UTF-8 minimal encoding restriction.
-        # A null byte that is encoded as "\xc0\x80" will be
-        # decoded as a null byte. All other illegal sequences
-        # will be handled strictly.
+    def test_decoding_callbacks(self):
+        # This is a test for a decoding callback handler
+        # that allows the decoding of the invalid sequence
+        # "\xc0\x80" and returns "\x00" instead of raising an error.
+        # All other illegal sequences will be handled strictly.
         def relaxedutf8(exc):
             if not isinstance(exc, UnicodeDecodeError):
                 raise TypeError("don't know how to handle %r" % exc)
-            if exc.object[exc.start:exc.end].startswith(b"\xc0\x80"):
+            if exc.object[exc.start:exc.start+2] == b"\xc0\x80":
                 return ("\x00", exc.start+2) # retry after two bytes
             else:
                 raise exc
 
-        codecs.register_error(
-            "test.relaxedutf8", relaxedutf8)
+        codecs.register_error("test.relaxedutf8", relaxedutf8)
 
+        # all the "\xc0\x80" will be decoded to "\x00"
         sin = b"a\x00b\xc0\x80c\xc3\xbc\xc0\x80\xc0\x80"
         sout = "a\x00b\x00c\xfc\x00\x00"
         self.assertEqual(sin.decode("utf-8", "test.relaxedutf8"), sout)
+
+        # "\xc0\x81" is not valid and a UnicodeDecodeError will be raised
         sin = b"\xc0\x80\xc0\x81"
-        self.assertRaises(UnicodeError, sin.decode, "utf-8", "test.relaxedutf8")
+        self.assertRaises(UnicodeDecodeError, sin.decode,
+                          "utf-8", "test.relaxedutf8")
 
     def test_charmapencode(self):
         # For charmap encodings the replacement string will be

Modified: python/branches/release31-maint/Lib/test/test_unicode.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_unicode.py	(original)
+++ python/branches/release31-maint/Lib/test/test_unicode.py	Sat Jul  3 07:18:50 2010
@@ -948,6 +948,159 @@
         # * strict decoding testing for all of the
         #   UTF8_ERROR cases in PyUnicode_DecodeUTF8
 
+    def test_utf8_decode_valid_sequences(self):
+        sequences = [
+            # single byte
+            (b'\x00', '\x00'), (b'a', 'a'), (b'\x7f', '\x7f'),
+            # 2 bytes
+            (b'\xc2\x80', '\x80'), (b'\xdf\xbf', '\u07ff'),
+            # 3 bytes
+            (b'\xe0\xa0\x80', '\u0800'), (b'\xed\x9f\xbf', '\ud7ff'),
+            (b'\xee\x80\x80', '\uE000'), (b'\xef\xbf\xbf', '\uffff'),
+            # 4 bytes
+            (b'\xF0\x90\x80\x80', '\U00010000'),
+            (b'\xf4\x8f\xbf\xbf', '\U0010FFFF')
+        ]
+        for seq, res in sequences:
+            self.assertEqual(seq.decode('utf-8'), res)
+
+
+    def test_utf8_decode_invalid_sequences(self):
+        # continuation bytes in a sequence of 2, 3, or 4 bytes
+        continuation_bytes = [bytes([x]) for x in range(0x80, 0xC0)]
+        # start bytes of a 2-byte sequence equivalent to codepoints < 0x7F
+        invalid_2B_seq_start_bytes = [bytes([x]) for x in range(0xC0, 0xC2)]
+        # start bytes of a 4-byte sequence equivalent to codepoints > 0x10FFFF
+        invalid_4B_seq_start_bytes = [bytes([x]) for x in range(0xF5, 0xF8)]
+        invalid_start_bytes = (
+            continuation_bytes + invalid_2B_seq_start_bytes +
+            invalid_4B_seq_start_bytes + [bytes([x]) for x in range(0xF7, 0x100)]
+        )
+
+        for byte in invalid_start_bytes:
+            self.assertRaises(UnicodeDecodeError, byte.decode, 'utf-8')
+
+        for sb in invalid_2B_seq_start_bytes:
+            for cb in continuation_bytes:
+                self.assertRaises(UnicodeDecodeError, (sb+cb).decode, 'utf-8')
+
+        for sb in invalid_4B_seq_start_bytes:
+            for cb1 in continuation_bytes[:3]:
+                for cb3 in continuation_bytes[:3]:
+                    self.assertRaises(UnicodeDecodeError,
+                                      (sb+cb1+b'\x80'+cb3).decode, 'utf-8')
+
+        for cb in [bytes([x]) for x in range(0x80, 0xA0)]:
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xE0'+cb+b'\x80').decode, 'utf-8')
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xE0'+cb+b'\xBF').decode, 'utf-8')
+        # surrogates
+        for cb in [bytes([x]) for x in range(0xA0, 0xC0)]:
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xED'+cb+b'\x80').decode, 'utf-8')
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xED'+cb+b'\xBF').decode, 'utf-8')
+        for cb in [bytes([x]) for x in range(0x80, 0x90)]:
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xF0'+cb+b'\x80\x80').decode, 'utf-8')
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xF0'+cb+b'\xBF\xBF').decode, 'utf-8')
+        for cb in [bytes([x]) for x in range(0x90, 0xC0)]:
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xF4'+cb+b'\x80\x80').decode, 'utf-8')
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xF4'+cb+b'\xBF\xBF').decode, 'utf-8')
+
+    def test_issue8271(self):
+        # Issue #8271: during the decoding of an invalid UTF-8 byte sequence,
+        # only the start byte and the continuation byte(s) are now considered
+        # invalid, instead of the number of bytes specified by the start byte.
+        # See http://www.unicode.org/versions/Unicode5.2.0/ch03.pdf (page 95,
+        # table 3-8, Row 2) for more information about the algorithm used.
+        FFFD = '\ufffd'
+        sequences = [
+            # invalid start bytes
+            (b'\x80', FFFD), # continuation byte
+            (b'\x80\x80', FFFD*2), # 2 continuation bytes
+            (b'\xc0', FFFD),
+            (b'\xc0\xc0', FFFD*2),
+            (b'\xc1', FFFD),
+            (b'\xc1\xc0', FFFD*2),
+            (b'\xc0\xc1', FFFD*2),
+            # with start byte of a 2-byte sequence
+            (b'\xc2', FFFD), # only the start byte
+            (b'\xc2\xc2', FFFD*2), # 2 start bytes
+            (b'\xc2\xc2\xc2', FFFD*3), # 2 start bytes
+            (b'\xc2\x41', FFFD+'A'), # invalid continuation byte
+            # with start byte of a 3-byte sequence
+            (b'\xe1', FFFD), # only the start byte
+            (b'\xe1\xe1', FFFD*2), # 2 start bytes
+            (b'\xe1\xe1\xe1', FFFD*3), # 3 start bytes
+            (b'\xe1\xe1\xe1\xe1', FFFD*4), # 4 start bytes
+            (b'\xe1\x80', FFFD), # only 1 continuation byte
+            (b'\xe1\x41', FFFD+'A'), # invalid continuation byte
+            (b'\xe1\x41\x80', FFFD+'A'+FFFD), # invalid cb followed by valid cb
+            (b'\xe1\x41\x41', FFFD+'AA'), # 2 invalid continuation bytes
+            (b'\xe1\x80\x41', FFFD+'A'), # only 1 valid continuation byte
+            (b'\xe1\x80\xe1\x41', FFFD*2+'A'), # 1 valid and the other invalid
+            (b'\xe1\x41\xe1\x80', FFFD+'A'+FFFD), # 1 invalid and the other valid
+            # with start byte of a 4-byte sequence
+            (b'\xf1', FFFD), # only the start byte
+            (b'\xf1\xf1', FFFD*2), # 2 start bytes
+            (b'\xf1\xf1\xf1', FFFD*3), # 3 start bytes
+            (b'\xf1\xf1\xf1\xf1', FFFD*4), # 4 start bytes
+            (b'\xf1\xf1\xf1\xf1\xf1', FFFD*5), # 5 start bytes
+            (b'\xf1\x80', FFFD), # only 1 continuation bytes
+            (b'\xf1\x80\x80', FFFD), # only 2 continuation bytes
+            (b'\xf1\x80\x41', FFFD+'A'), # 1 valid cb and 1 invalid
+            (b'\xf1\x80\x41\x41', FFFD+'AA'), # 1 valid cb and 1 invalid
+            (b'\xf1\x80\x80\x41', FFFD+'A'), # 2 valid cb and 1 invalid
+            (b'\xf1\x41\x80', FFFD+'A'+FFFD), # 1 invalid cv and 1 valid
+            (b'\xf1\x41\x80\x80', FFFD+'A'+FFFD*2), # 1 invalid cb and 2 invalid
+            (b'\xf1\x41\x80\x41', FFFD+'A'+FFFD+'A'), # 2 invalid cb and 1 invalid
+            (b'\xf1\x41\x41\x80', FFFD+'AA'+FFFD), # 1 valid cb and 1 invalid
+            (b'\xf1\x41\xf1\x80', FFFD+'A'+FFFD),
+            (b'\xf1\x41\x80\xf1', FFFD+'A'+FFFD*2),
+            (b'\xf1\xf1\x80\x41', FFFD*2+'A'),
+            (b'\xf1\x41\xf1\xf1', FFFD+'A'+FFFD*2),
+            # with invalid start byte of a 4-byte sequence (rfc2279)
+            (b'\xf5', FFFD), # only the start byte
+            (b'\xf5\xf5', FFFD*2), # 2 start bytes
+            (b'\xf5\x80', FFFD*2), # only 1 continuation byte
+            (b'\xf5\x80\x80', FFFD*3), # only 2 continuation byte
+            (b'\xf5\x80\x80\x80', FFFD*4), # 3 continuation bytes
+            (b'\xf5\x80\x41', FFFD*2+'A'), #  1 valid cb and 1 invalid
+            (b'\xf5\x80\x41\xf5', FFFD*2+'A'+FFFD),
+            (b'\xf5\x41\x80\x80\x41', FFFD+'A'+FFFD*2+'A'),
+            # with invalid start byte of a 5-byte sequence (rfc2279)
+            (b'\xf8', FFFD), # only the start byte
+            (b'\xf8\xf8', FFFD*2), # 2 start bytes
+            (b'\xf8\x80', FFFD*2), # only one continuation byte
+            (b'\xf8\x80\x41', FFFD*2 + 'A'), # 1 valid cb and 1 invalid
+            (b'\xf8\x80\x80\x80\x80', FFFD*5), # invalid 5 bytes seq with 5 bytes
+            # with invalid start byte of a 6-byte sequence (rfc2279)
+            (b'\xfc', FFFD), # only the start byte
+            (b'\xfc\xfc', FFFD*2), # 2 start bytes
+            (b'\xfc\x80\x80', FFFD*3), # only 2 continuation bytes
+            (b'\xfc\x80\x80\x80\x80\x80', FFFD*6), # 6 continuation bytes
+            # invalid start byte
+            (b'\xfe', FFFD),
+            (b'\xfe\x80\x80', FFFD*3),
+            # other sequences
+            (b'\xf1\x80\x41\x42\x43', '\ufffd\x41\x42\x43'),
+            (b'\xf1\x80\xff\x42\x43', '\ufffd\ufffd\x42\x43'),
+            (b'\xf1\x80\xc2\x81\x43', '\ufffd\x81\x43'),
+            (b'\x61\xF1\x80\x80\xE1\x80\xC2\x62\x80\x63\x80\xBF\x64',
+             '\x61\uFFFD\uFFFD\uFFFD\x62\uFFFD\x63\uFFFD\uFFFD\x64'),
+        ]
+        for n, (seq, res) in enumerate(sequences):
+            self.assertRaises(UnicodeDecodeError, seq.decode, 'utf-8', 'strict')
+            self.assertEqual(seq.decode('utf-8', 'replace'), res)
+            self.assertEqual((seq+b'b').decode('utf-8', 'replace'), res+'b')
+            self.assertEqual(seq.decode('utf-8', 'ignore'),
+                             res.replace('\uFFFD', ''))
+
     def test_codecs_idna(self):
         # Test whether trailing dot is preserved
         self.assertEqual("www.python.org.".encode("idna"), b"www.python.org.")

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Sat Jul  3 07:18:50 2010
@@ -12,6 +12,14 @@
 Core and Builtins
 -----------------
 
+- Issue #8271: during the decoding of an invalid UTF-8 byte sequence, only the
+  start byte and the continuation byte(s) are now considered invalid, instead
+  of the number of bytes specified by the start byte.
+  E.g.: '\xf1\x80AB'.decode('utf-8', 'replace') now returns u'\ufffdAB' and
+  replaces with U+FFFD only the start byte ('\xf1') and the continuation byte
+  ('\x80') even if '\xf1' is the start byte of a 4-bytes sequence.
+  Previous versions returned a single u'\ufffd'.
+
 - Issue #6543: Write the traceback in the terminal encoding instead of utf-8.
   Fix the encoding of the modules filename. Patch written by Amaury Forgeot
   d'Arc.

Modified: python/branches/release31-maint/Objects/unicodeobject.c
==============================================================================
--- python/branches/release31-maint/Objects/unicodeobject.c	(original)
+++ python/branches/release31-maint/Objects/unicodeobject.c	Sat Jul  3 07:18:50 2010
@@ -2147,24 +2147,24 @@
 
 static
 char utf8_code_length[256] = {
-    /* Map UTF-8 encoded prefix byte to sequence length.  zero means
-       illegal prefix.  see RFC 2279 for details */
+    /* Map UTF-8 encoded prefix byte to sequence length.  Zero means
+       illegal prefix.  See RFC 3629 for details */
+    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 00-0F */
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
-    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
-    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
-    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 70-7F */
+    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 80-8F */
     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
-    2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
-    3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
-    4, 4, 4, 4, 4, 4, 4, 4, 5, 5, 5, 5, 6, 6, 0, 0
+    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* B0-BF */
+    0, 0, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, /* C0-C1 + C2-CF */
+    2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, /* D0-DF */
+    3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, /* E0-EF */
+    4, 4, 4, 4, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0  /* F0-F4 + F5-FF */
 };
 
 PyObject *PyUnicode_DecodeUTF8(const char *s,
@@ -2194,6 +2194,7 @@
 {
     const char *starts = s;
     int n;
+    int k;
     Py_ssize_t startinpos;
     Py_ssize_t endinpos;
     Py_ssize_t outpos;
@@ -2277,7 +2278,9 @@
             else {
                 errmsg = "unexpected end of data";
                 startinpos = s-starts;
-                endinpos = size;
+                endinpos = startinpos+1;
+                for (k=1; (k < size-startinpos) && ((s[k]&0xC0) == 0x80); k++)
+                    endinpos++;
                 goto utf8Error;
             }
         }
@@ -2285,7 +2288,7 @@
         switch (n) {
 
         case 0:
-            errmsg = "unexpected code byte";
+            errmsg = "invalid start byte";
             startinpos = s-starts;
             endinpos = startinpos+1;
             goto utf8Error;
@@ -2298,63 +2301,67 @@
 
         case 2:
             if ((s[1] & 0xc0) != 0x80) {
-                errmsg = "invalid data";
+                errmsg = "invalid continuation byte";
                 startinpos = s-starts;
-                endinpos = startinpos+2;
+                endinpos = startinpos + 1;
                 goto utf8Error;
             }
             ch = ((s[0] & 0x1f) << 6) + (s[1] & 0x3f);
-            if (ch < 0x80) {
-                startinpos = s-starts;
-                endinpos = startinpos+2;
-                errmsg = "illegal encoding";
-                goto utf8Error;
-            }
-            else
-                *p++ = (Py_UNICODE)ch;
+            assert ((ch > 0x007F) && (ch <= 0x07FF));
+            *p++ = (Py_UNICODE)ch;
             break;
 
         case 3:
+            /* Decoding UTF-8 sequences in range \xed\xa0\x80-\xed\xbf\xbf
+               will result in surrogates in range d800-dfff. Surrogates are
+               not valid UTF-8 so they are rejected.
+               See http://www.unicode.org/versions/Unicode5.2.0/ch03.pdf
+               (table 3-7) and http://www.rfc-editor.org/rfc/rfc3629.txt */
             if ((s[1] & 0xc0) != 0x80 ||
-                (s[2] & 0xc0) != 0x80) {
-                errmsg = "invalid data";
+                (s[2] & 0xc0) != 0x80 ||
+                ((unsigned char)s[0] == 0xE0 &&
+                 (unsigned char)s[1] < 0xA0) ||
+                ((unsigned char)s[0] == 0xED &&
+                 (unsigned char)s[1] > 0x9F)) {
+                errmsg = "invalid continuation byte";
                 startinpos = s-starts;
-                endinpos = startinpos+3;
+                endinpos = startinpos + 1;
+
+                /* if s[1] first two bits are 1 and 0, then the invalid
+                   continuation byte is s[2], so increment endinpos by 1,
+                   if not, s[1] is invalid and endinpos doesn't need to
+                   be incremented. */
+                if ((s[1] & 0xC0) == 0x80)
+                    endinpos++;
                 goto utf8Error;
             }
             ch = ((s[0] & 0x0f) << 12) + ((s[1] & 0x3f) << 6) + (s[2] & 0x3f);
-            if (ch < 0x0800 || (ch >= 0xd800 && ch <= 0xDFFF)) {
-                errmsg = "illegal encoding";
-                startinpos = s-starts;
-                endinpos = startinpos+3;
-                goto utf8Error;
-            }
-            else
-                *p++ = (Py_UNICODE)ch;
+            assert ((ch > 0x07FF) && (ch <= 0xFFFF));
+            *p++ = (Py_UNICODE)ch;
             break;
 
         case 4:
             if ((s[1] & 0xc0) != 0x80 ||
                 (s[2] & 0xc0) != 0x80 ||
-                (s[3] & 0xc0) != 0x80) {
-                errmsg = "invalid data";
+                (s[3] & 0xc0) != 0x80 ||
+                ((unsigned char)s[0] == 0xF0 &&
+                 (unsigned char)s[1] < 0x90) ||
+                ((unsigned char)s[0] == 0xF4 &&
+                 (unsigned char)s[1] > 0x8F)) {
+                errmsg = "invalid continuation byte";
                 startinpos = s-starts;
-                endinpos = startinpos+4;
+                endinpos = startinpos + 1;
+                if ((s[1] & 0xC0) == 0x80) {
+                    endinpos++;
+                    if ((s[2] & 0xC0) == 0x80)
+                        endinpos++;
+                }
                 goto utf8Error;
             }
             ch = ((s[0] & 0x7) << 18) + ((s[1] & 0x3f) << 12) +
-                ((s[2] & 0x3f) << 6) + (s[3] & 0x3f);
-            /* validate and convert to UTF-16 */
-            if ((ch < 0x10000)        /* minimum value allowed for 4
-                                         byte encoding */
-                || (ch > 0x10ffff))   /* maximum value allowed for
-                                         UTF-16 */
-            {
-                errmsg = "illegal encoding";
-                startinpos = s-starts;
-                endinpos = startinpos+4;
-                goto utf8Error;
-            }
+                 ((s[2] & 0x3f) << 6) + (s[3] & 0x3f);
+            assert ((ch > 0xFFFF) && (ch <= 0x10ffff));
+
 #ifdef Py_UNICODE_WIDE
             *p++ = (Py_UNICODE)ch;
 #else
@@ -2370,13 +2377,6 @@
             *p++ = (Py_UNICODE)(0xDC00 + (ch & 0x03FF));
 #endif
             break;
-
-        default:
-            /* Other sizes are only needed for UCS-4 */
-            errmsg = "unsupported Unicode code range";
-            startinpos = s-starts;
-            endinpos = startinpos+n;
-            goto utf8Error;
         }
         s += n;
         continue;

From python-checkins at python.org  Sat Jul  3 07:34:39 2010
From: python-checkins at python.org (ezio.melotti)
Date: Sat,  3 Jul 2010 07:34:39 +0200 (CEST)
Subject: [Python-checkins] r82470 - in python/branches/release26-maint:
	Lib/test/test_codeccallbacks.py Lib/test/test_unicode.py
	Misc/NEWS Objects/unicodeobject.c
Message-ID: <20100703053439.74482EE996@mail.python.org>

Author: ezio.melotti
Date: Sat Jul  3 07:34:39 2010
New Revision: 82470

Log:
Merged revisions 81758-81759 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/trunk

........
  r81758 | ezio.melotti | 2010-06-05 20:51:07 +0300 (Sat, 05 Jun 2010) | 15 lines
  
  Update PyUnicode_DecodeUTF8 from RFC 2279 to RFC 3629.
  
  1) #8271: when a byte sequence is invalid, only the start byte and all the
     valid continuation bytes are now replaced by U+FFFD, instead of replacing
     the number of bytes specified by the start byte.
     See http://www.unicode.org/versions/Unicode5.2.0/ch03.pdf (pages 94-95);
  2) 5- and 6-bytes-long UTF-8 sequences are now considered invalid (no changes
     in behavior);
  3) Add code and tests to reject surrogates (U+D800-U+DFFF) as defined in
     RFC 3629, but leave it commented out since it's not backward compatible;
  4) Change the error messages "unexpected code byte" to "invalid start byte"
     and "invalid data" to "invalid continuation byte";
  5) Add an extensive set of tests in test_unicode;
  6) Fix test_codeccallbacks because it was failing after this change.
........
  r81759 | ezio.melotti | 2010-06-05 22:21:32 +0300 (Sat, 05 Jun 2010) | 1 line
  
  Add a NEWS entry for r81758 and clarify a comment.
........


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Lib/test/test_codeccallbacks.py
   python/branches/release26-maint/Lib/test/test_unicode.py
   python/branches/release26-maint/Misc/NEWS
   python/branches/release26-maint/Objects/unicodeobject.c

Modified: python/branches/release26-maint/Lib/test/test_codeccallbacks.py
==============================================================================
--- python/branches/release26-maint/Lib/test/test_codeccallbacks.py	(original)
+++ python/branches/release26-maint/Lib/test/test_codeccallbacks.py	Sat Jul  3 07:34:39 2010
@@ -153,28 +153,30 @@
             sout += "\\U%08x" % sys.maxunicode
         self.assertEqual(sin.encode("iso-8859-15", "backslashreplace"), sout)
 
-    def test_decoderelaxedutf8(self):
-        # This is the test for a decoding callback handler,
-        # that relaxes the UTF-8 minimal encoding restriction.
-        # A null byte that is encoded as "\xc0\x80" will be
-        # decoded as a null byte. All other illegal sequences
-        # will be handled strictly.
+    def test_decoding_callbacks(self):
+        # This is a test for a decoding callback handler
+        # that allows the decoding of the invalid sequence
+        # "\xc0\x80" and returns "\x00" instead of raising an error.
+        # All other illegal sequences will be handled strictly.
         def relaxedutf8(exc):
             if not isinstance(exc, UnicodeDecodeError):
                 raise TypeError("don't know how to handle %r" % exc)
-            if exc.object[exc.start:exc.end].startswith("\xc0\x80"):
+            if exc.object[exc.start:exc.start+2] == "\xc0\x80":
                 return (u"\x00", exc.start+2) # retry after two bytes
             else:
                 raise exc
 
-        codecs.register_error(
-            "test.relaxedutf8", relaxedutf8)
+        codecs.register_error("test.relaxedutf8", relaxedutf8)
 
+        # all the "\xc0\x80" will be decoded to "\x00"
         sin = "a\x00b\xc0\x80c\xc3\xbc\xc0\x80\xc0\x80"
         sout = u"a\x00b\x00c\xfc\x00\x00"
         self.assertEqual(sin.decode("utf-8", "test.relaxedutf8"), sout)
+
+        # "\xc0\x81" is not valid and a UnicodeDecodeError will be raised
         sin = "\xc0\x80\xc0\x81"
-        self.assertRaises(UnicodeError, sin.decode, "utf-8", "test.relaxedutf8")
+        self.assertRaises(UnicodeDecodeError, sin.decode,
+                          "utf-8", "test.relaxedutf8")
 
     def test_charmapencode(self):
         # For charmap encodings the replacement string will be

Modified: python/branches/release26-maint/Lib/test/test_unicode.py
==============================================================================
--- python/branches/release26-maint/Lib/test/test_unicode.py	(original)
+++ python/branches/release26-maint/Lib/test/test_unicode.py	Sat Jul  3 07:34:39 2010
@@ -587,6 +587,164 @@
         # * strict decoding testing for all of the
         #   UTF8_ERROR cases in PyUnicode_DecodeUTF8
 
+    def test_utf8_decode_valid_sequences(self):
+        sequences = [
+            # single byte
+            ('\x00', u'\x00'), ('a', u'a'), ('\x7f', u'\x7f'),
+            # 2 bytes
+            ('\xc2\x80', u'\x80'), ('\xdf\xbf', u'\u07ff'),
+            # 3 bytes
+            ('\xe0\xa0\x80', u'\u0800'), ('\xed\x9f\xbf', u'\ud7ff'),
+            ('\xee\x80\x80', u'\uE000'), ('\xef\xbf\xbf', u'\uffff'),
+            # 4 bytes
+            ('\xF0\x90\x80\x80', u'\U00010000'),
+            ('\xf4\x8f\xbf\xbf', u'\U0010FFFF')
+        ]
+        for seq, res in sequences:
+            self.assertEqual(seq.decode('utf-8'), res)
+
+        for ch in map(unichr, range(0, sys.maxunicode)):
+            self.assertEqual(ch, ch.encode('utf-8').decode('utf-8'))
+
+    def test_utf8_decode_invalid_sequences(self):
+        # continuation bytes in a sequence of 2, 3, or 4 bytes
+        continuation_bytes = map(chr, range(0x80, 0xC0))
+        # start bytes of a 2-byte sequence equivalent to codepoints < 0x7F
+        invalid_2B_seq_start_bytes = map(chr, range(0xC0, 0xC2))
+        # start bytes of a 4-byte sequence equivalent to codepoints > 0x10FFFF
+        invalid_4B_seq_start_bytes = map(chr, range(0xF5, 0xF8))
+        invalid_start_bytes = (
+            continuation_bytes + invalid_2B_seq_start_bytes +
+            invalid_4B_seq_start_bytes + map(chr, range(0xF7, 0x100))
+        )
+
+        for byte in invalid_start_bytes:
+            self.assertRaises(UnicodeDecodeError, byte.decode, 'utf-8')
+
+        for sb in invalid_2B_seq_start_bytes:
+            for cb in continuation_bytes:
+                self.assertRaises(UnicodeDecodeError, (sb+cb).decode, 'utf-8')
+
+        for sb in invalid_4B_seq_start_bytes:
+            for cb1 in continuation_bytes[:3]:
+                for cb3 in continuation_bytes[:3]:
+                    self.assertRaises(UnicodeDecodeError,
+                                      (sb+cb1+'\x80'+cb3).decode, 'utf-8')
+
+        for cb in map(chr, range(0x80, 0xA0)):
+            self.assertRaises(UnicodeDecodeError,
+                              ('\xE0'+cb+'\x80').decode, 'utf-8')
+            self.assertRaises(UnicodeDecodeError,
+                              ('\xE0'+cb+'\xBF').decode, 'utf-8')
+        # XXX: surrogates shouldn't be valid UTF-8!
+        # see http://www.unicode.org/versions/Unicode5.2.0/ch03.pdf
+        # (table 3-7) and http://www.rfc-editor.org/rfc/rfc3629.txt
+        #for cb in map(chr, range(0xA0, 0xC0)):
+            #sys.__stdout__.write('\\xED\\x%02x\\x80\n' % ord(cb))
+            #self.assertRaises(UnicodeDecodeError,
+                              #('\xED'+cb+'\x80').decode, 'utf-8')
+            #self.assertRaises(UnicodeDecodeError,
+                              #('\xED'+cb+'\xBF').decode, 'utf-8')
+        for cb in map(chr, range(0x80, 0x90)):
+            self.assertRaises(UnicodeDecodeError,
+                              ('\xF0'+cb+'\x80\x80').decode, 'utf-8')
+            self.assertRaises(UnicodeDecodeError,
+                              ('\xF0'+cb+'\xBF\xBF').decode, 'utf-8')
+        for cb in map(chr, range(0x90, 0xC0)):
+            self.assertRaises(UnicodeDecodeError,
+                              ('\xF4'+cb+'\x80\x80').decode, 'utf-8')
+            self.assertRaises(UnicodeDecodeError,
+                              ('\xF4'+cb+'\xBF\xBF').decode, 'utf-8')
+
+    def test_issue8271(self):
+        # Issue #8271: during the decoding of an invalid UTF-8 byte sequence,
+        # only the start byte and the continuation byte(s) are now considered
+        # invalid, instead of the number of bytes specified by the start byte.
+        # See http://www.unicode.org/versions/Unicode5.2.0/ch03.pdf (page 95,
+        # table 3-8, Row 2) for more information about the algorithm used.
+        FFFD = u'\ufffd'
+        sequences = [
+            # invalid start bytes
+            ('\x80', FFFD), # continuation byte
+            ('\x80\x80', FFFD*2), # 2 continuation bytes
+            ('\xc0', FFFD),
+            ('\xc0\xc0', FFFD*2),
+            ('\xc1', FFFD),
+            ('\xc1\xc0', FFFD*2),
+            ('\xc0\xc1', FFFD*2),
+            # with start byte of a 2-byte sequence
+            ('\xc2', FFFD), # only the start byte
+            ('\xc2\xc2', FFFD*2), # 2 start bytes
+            ('\xc2\xc2\xc2', FFFD*3), # 2 start bytes
+            ('\xc2\x41', FFFD+'A'), # invalid continuation byte
+            # with start byte of a 3-byte sequence
+            ('\xe1', FFFD), # only the start byte
+            ('\xe1\xe1', FFFD*2), # 2 start bytes
+            ('\xe1\xe1\xe1', FFFD*3), # 3 start bytes
+            ('\xe1\xe1\xe1\xe1', FFFD*4), # 4 start bytes
+            ('\xe1\x80', FFFD), # only 1 continuation byte
+            ('\xe1\x41', FFFD+'A'), # invalid continuation byte
+            ('\xe1\x41\x80', FFFD+'A'+FFFD), # invalid cb followed by valid cb
+            ('\xe1\x41\x41', FFFD+'AA'), # 2 invalid continuation bytes
+            ('\xe1\x80\x41', FFFD+'A'), # only 1 valid continuation byte
+            ('\xe1\x80\xe1\x41', FFFD*2+'A'), # 1 valid and the other invalid
+            ('\xe1\x41\xe1\x80', FFFD+'A'+FFFD), # 1 invalid and the other valid
+            # with start byte of a 4-byte sequence
+            ('\xf1', FFFD), # only the start byte
+            ('\xf1\xf1', FFFD*2), # 2 start bytes
+            ('\xf1\xf1\xf1', FFFD*3), # 3 start bytes
+            ('\xf1\xf1\xf1\xf1', FFFD*4), # 4 start bytes
+            ('\xf1\xf1\xf1\xf1\xf1', FFFD*5), # 5 start bytes
+            ('\xf1\x80', FFFD), # only 1 continuation bytes
+            ('\xf1\x80\x80', FFFD), # only 2 continuation bytes
+            ('\xf1\x80\x41', FFFD+'A'), # 1 valid cb and 1 invalid
+            ('\xf1\x80\x41\x41', FFFD+'AA'), # 1 valid cb and 1 invalid
+            ('\xf1\x80\x80\x41', FFFD+'A'), # 2 valid cb and 1 invalid
+            ('\xf1\x41\x80', FFFD+'A'+FFFD), # 1 invalid cv and 1 valid
+            ('\xf1\x41\x80\x80', FFFD+'A'+FFFD*2), # 1 invalid cb and 2 invalid
+            ('\xf1\x41\x80\x41', FFFD+'A'+FFFD+'A'), # 2 invalid cb and 1 invalid
+            ('\xf1\x41\x41\x80', FFFD+'AA'+FFFD), # 1 valid cb and 1 invalid
+            ('\xf1\x41\xf1\x80', FFFD+'A'+FFFD),
+            ('\xf1\x41\x80\xf1', FFFD+'A'+FFFD*2),
+            ('\xf1\xf1\x80\x41', FFFD*2+'A'),
+            ('\xf1\x41\xf1\xf1', FFFD+'A'+FFFD*2),
+            # with invalid start byte of a 4-byte sequence (rfc2279)
+            ('\xf5', FFFD), # only the start byte
+            ('\xf5\xf5', FFFD*2), # 2 start bytes
+            ('\xf5\x80', FFFD*2), # only 1 continuation byte
+            ('\xf5\x80\x80', FFFD*3), # only 2 continuation byte
+            ('\xf5\x80\x80\x80', FFFD*4), # 3 continuation bytes
+            ('\xf5\x80\x41', FFFD*2+'A'), #  1 valid cb and 1 invalid
+            ('\xf5\x80\x41\xf5', FFFD*2+'A'+FFFD),
+            ('\xf5\x41\x80\x80\x41', FFFD+'A'+FFFD*2+'A'),
+            # with invalid start byte of a 5-byte sequence (rfc2279)
+            ('\xf8', FFFD), # only the start byte
+            ('\xf8\xf8', FFFD*2), # 2 start bytes
+            ('\xf8\x80', FFFD*2), # only one continuation byte
+            ('\xf8\x80\x41', FFFD*2 + 'A'), # 1 valid cb and 1 invalid
+            ('\xf8\x80\x80\x80\x80', FFFD*5), # invalid 5 bytes seq with 5 bytes
+            # with invalid start byte of a 6-byte sequence (rfc2279)
+            ('\xfc', FFFD), # only the start byte
+            ('\xfc\xfc', FFFD*2), # 2 start bytes
+            ('\xfc\x80\x80', FFFD*3), # only 2 continuation bytes
+            ('\xfc\x80\x80\x80\x80\x80', FFFD*6), # 6 continuation bytes
+            # invalid start byte
+            ('\xfe', FFFD),
+            ('\xfe\x80\x80', FFFD*3),
+            # other sequences
+            ('\xf1\x80\x41\x42\x43', u'\ufffd\x41\x42\x43'),
+            ('\xf1\x80\xff\x42\x43', u'\ufffd\ufffd\x42\x43'),
+            ('\xf1\x80\xc2\x81\x43', u'\ufffd\x81\x43'),
+            ('\x61\xF1\x80\x80\xE1\x80\xC2\x62\x80\x63\x80\xBF\x64',
+             u'\x61\uFFFD\uFFFD\uFFFD\x62\uFFFD\x63\uFFFD\uFFFD\x64'),
+        ]
+        for n, (seq, res) in enumerate(sequences):
+            self.assertRaises(UnicodeDecodeError, seq.decode, 'utf-8', 'strict')
+            self.assertEqual(seq.decode('utf-8', 'replace'), res)
+            self.assertEqual((seq+'b').decode('utf-8', 'replace'), res+'b')
+            self.assertEqual(seq.decode('utf-8', 'ignore'),
+                             res.replace(u'\uFFFD', ''))
+
     def test_codecs_idna(self):
         # Test whether trailing dot is preserved
         self.assertEqual(u"www.python.org.".encode("idna"), "www.python.org.")

Modified: python/branches/release26-maint/Misc/NEWS
==============================================================================
--- python/branches/release26-maint/Misc/NEWS	(original)
+++ python/branches/release26-maint/Misc/NEWS	Sat Jul  3 07:34:39 2010
@@ -12,6 +12,14 @@
 Core and Builtins
 -----------------
 
+- Issue #8271: during the decoding of an invalid UTF-8 byte sequence, only the
+  start byte and the continuation byte(s) are now considered invalid, instead
+  of the number of bytes specified by the start byte.
+  E.g.: '\xf1\x80AB'.decode('utf-8', 'replace') now returns u'\ufffdAB' and
+  replaces with U+FFFD only the start byte ('\xf1') and the continuation byte
+  ('\x80') even if '\xf1' is the start byte of a 4-bytes sequence.
+  Previous versions returned a single u'\ufffd'.
+
 - Issue #9058: Remove assertions about INT_MAX in UnicodeDecodeError.
 
 - Issue #8941: decoding big endian UTF-32 data in UCS-2 builds could crash

Modified: python/branches/release26-maint/Objects/unicodeobject.c
==============================================================================
--- python/branches/release26-maint/Objects/unicodeobject.c	(original)
+++ python/branches/release26-maint/Objects/unicodeobject.c	Sat Jul  3 07:34:39 2010
@@ -1735,24 +1735,24 @@
 
 static
 char utf8_code_length[256] = {
-    /* Map UTF-8 encoded prefix byte to sequence length.  zero means
-       illegal prefix.  see RFC 2279 for details */
+    /* Map UTF-8 encoded prefix byte to sequence length.  Zero means
+       illegal prefix.  See RFC 3629 for details */
+    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 00-0F */
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
-    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
-    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
-    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 70-7F */
+    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 80-8F */
     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
-    2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
-    3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
-    4, 4, 4, 4, 4, 4, 4, 4, 5, 5, 5, 5, 6, 6, 0, 0
+    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* B0-BF */
+    0, 0, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, /* C0-C1 + C2-CF */
+    2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, /* D0-DF */
+    3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, /* E0-EF */
+    4, 4, 4, 4, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0  /* F0-F4 + F5-FF */
 };
 
 PyObject *PyUnicode_DecodeUTF8(const char *s,
@@ -1769,6 +1769,7 @@
 {
     const char *starts = s;
     int n;
+    int k;
     Py_ssize_t startinpos;
     Py_ssize_t endinpos;
     Py_ssize_t outpos;
@@ -1811,7 +1812,9 @@
             else {
                 errmsg = "unexpected end of data";
                 startinpos = s-starts;
-                endinpos = size;
+                endinpos = startinpos+1;
+                for (k=1; (k < size-startinpos) && ((s[k]&0xC0) == 0x80); k++)
+                    endinpos++;
                 goto utf8Error;
             }
         }
@@ -1819,7 +1822,7 @@
         switch (n) {
 
         case 0:
-            errmsg = "unexpected code byte";
+            errmsg = "invalid start byte";
             startinpos = s-starts;
             endinpos = startinpos+1;
             goto utf8Error;
@@ -1832,70 +1835,67 @@
 
         case 2:
             if ((s[1] & 0xc0) != 0x80) {
-                errmsg = "invalid data";
+                errmsg = "invalid continuation byte";
                 startinpos = s-starts;
-                endinpos = startinpos+2;
+                endinpos = startinpos + 1;
                 goto utf8Error;
             }
             ch = ((s[0] & 0x1f) << 6) + (s[1] & 0x3f);
-            if (ch < 0x80) {
-                startinpos = s-starts;
-                endinpos = startinpos+2;
-                errmsg = "illegal encoding";
-                goto utf8Error;
-            }
-            else
-                *p++ = (Py_UNICODE)ch;
+            assert ((ch > 0x007F) && (ch <= 0x07FF));
+            *p++ = (Py_UNICODE)ch;
             break;
 
         case 3:
+            /* XXX: surrogates shouldn't be valid UTF-8!
+               see http://www.unicode.org/versions/Unicode5.2.0/ch03.pdf
+               (table 3-7) and http://www.rfc-editor.org/rfc/rfc3629.txt
+               Uncomment the 2 lines below to make them invalid,
+               codepoints: d800-dfff; UTF-8: \xed\xa0\x80-\xed\xbf\xbf. */
             if ((s[1] & 0xc0) != 0x80 ||
-                (s[2] & 0xc0) != 0x80) {
-                errmsg = "invalid data";
+                (s[2] & 0xc0) != 0x80 ||
+                ((unsigned char)s[0] == 0xE0 &&
+                 (unsigned char)s[1] < 0xA0)/* ||
+                ((unsigned char)s[0] == 0xED &&
+                 (unsigned char)s[1] > 0x9F)*/) {
+                errmsg = "invalid continuation byte";
                 startinpos = s-starts;
-                endinpos = startinpos+3;
+                endinpos = startinpos + 1;
+
+                /* if s[1] first two bits are 1 and 0, then the invalid
+                   continuation byte is s[2], so increment endinpos by 1,
+                   if not, s[1] is invalid and endinpos doesn't need to
+                   be incremented. */
+                if ((s[1] & 0xC0) == 0x80)
+                    endinpos++;
                 goto utf8Error;
             }
             ch = ((s[0] & 0x0f) << 12) + ((s[1] & 0x3f) << 6) + (s[2] & 0x3f);
-            if (ch < 0x0800) {
-                /* Note: UTF-8 encodings of surrogates are considered
-                   legal UTF-8 sequences;
-
-                   XXX For wide builds (UCS-4) we should probably try
-                   to recombine the surrogates into a single code
-                   unit.
-                */
-                errmsg = "illegal encoding";
-                startinpos = s-starts;
-                endinpos = startinpos+3;
-                goto utf8Error;
-            }
-            else
-                *p++ = (Py_UNICODE)ch;
+            assert ((ch > 0x07FF) && (ch <= 0xFFFF));
+            *p++ = (Py_UNICODE)ch;
             break;
 
         case 4:
             if ((s[1] & 0xc0) != 0x80 ||
                 (s[2] & 0xc0) != 0x80 ||
-                (s[3] & 0xc0) != 0x80) {
-                errmsg = "invalid data";
+                (s[3] & 0xc0) != 0x80 ||
+                ((unsigned char)s[0] == 0xF0 &&
+                 (unsigned char)s[1] < 0x90) ||
+                ((unsigned char)s[0] == 0xF4 &&
+                 (unsigned char)s[1] > 0x8F)) {
+                errmsg = "invalid continuation byte";
                 startinpos = s-starts;
-                endinpos = startinpos+4;
+                endinpos = startinpos + 1;
+                if ((s[1] & 0xC0) == 0x80) {
+                    endinpos++;
+                    if ((s[2] & 0xC0) == 0x80)
+                        endinpos++;
+                }
                 goto utf8Error;
             }
             ch = ((s[0] & 0x7) << 18) + ((s[1] & 0x3f) << 12) +
-                ((s[2] & 0x3f) << 6) + (s[3] & 0x3f);
-            /* validate and convert to UTF-16 */
-            if ((ch < 0x10000)        /* minimum value allowed for 4
-                                         byte encoding */
-                || (ch > 0x10ffff))   /* maximum value allowed for
-                                         UTF-16 */
-            {
-                errmsg = "illegal encoding";
-                startinpos = s-starts;
-                endinpos = startinpos+4;
-                goto utf8Error;
-            }
+                 ((s[2] & 0x3f) << 6) + (s[3] & 0x3f);
+            assert ((ch > 0xFFFF) && (ch <= 0x10ffff));
+
 #ifdef Py_UNICODE_WIDE
             *p++ = (Py_UNICODE)ch;
 #else
@@ -1911,13 +1911,6 @@
             *p++ = (Py_UNICODE)(0xDC00 + (ch & 0x03FF));
 #endif
             break;
-
-        default:
-            /* Other sizes are only needed for UCS-4 */
-            errmsg = "unsupported Unicode code range";
-            startinpos = s-starts;
-            endinpos = startinpos+n;
-            goto utf8Error;
         }
         s += n;
         continue;

From python-checkins at python.org  Sat Jul  3 09:36:51 2010
From: python-checkins at python.org (nick.coghlan)
Date: Sat,  3 Jul 2010 09:36:51 +0200 (CEST)
Subject: [Python-checkins] r82471 - in python/branches/py3k:
	Doc/library/dis.rst Lib/dis.py Lib/test/test_dis.py Misc/ACKS
	Misc/NEWS
Message-ID: <20100703073651.93017EE982@mail.python.org>

Author: nick.coghlan
Date: Sat Jul  3 09:36:51 2010
New Revision: 82471

Log:
Issue 6507: accept source strings directly in dis.dis(). Original patch by Daniel Urban

Modified:
   python/branches/py3k/Doc/library/dis.rst
   python/branches/py3k/Lib/dis.py
   python/branches/py3k/Lib/test/test_dis.py
   python/branches/py3k/Misc/ACKS
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Doc/library/dis.rst
==============================================================================
--- python/branches/py3k/Doc/library/dis.rst	(original)
+++ python/branches/py3k/Doc/library/dis.rst	Sat Jul  3 09:36:51 2010
@@ -33,10 +33,13 @@
 .. function:: dis(x=None)
 
    Disassemble the *x* object. *x* can denote either a module, a
-   class, a method, a function, or a code object.   For a module, it disassembles
-   all functions.  For a class, it disassembles all methods.  For a single code
-   sequence, it prints one line per bytecode instruction.  If no object is
-   provided, it disassembles the last traceback.
+   class, a method, a function, a code object, a string of source code or a
+   byte sequence of raw bytecode.   For a module, it disassembles all
+   functions.  For a class, it disassembles all methods.  For a code object
+   or sequence of raw bytecode, it prints one line per bytecode instruction.
+   Strings are first compiled to code objects with the :func:`compile`
+   built-in function before being disassembled.  If no object is provided,
+   this function disassembles the last traceback.
 
 
 .. function:: distb(tb=None)

Modified: python/branches/py3k/Lib/dis.py
==============================================================================
--- python/branches/py3k/Lib/dis.py	(original)
+++ python/branches/py3k/Lib/dis.py	Sat Jul  3 09:36:51 2010
@@ -12,6 +12,22 @@
 
 _have_code = (types.MethodType, types.FunctionType, types.CodeType, type)
 
+def _try_compile(source, name):
+    """Attempts to compile the given source, first as an expression and
+       then as a statement if the first approach fails.
+
+       Utility function to accept strings in functions that otherwise
+       expect code objects
+    """
+    # ncoghlan: currently only used by dis(), but plan to add an
+    # equivalent for show_code() as well (but one that returns a
+    # string rather than printing directly to the console)
+    try:
+        c = compile(source, name, 'eval')
+    except SyntaxError:
+        c = compile(source, name, 'exec')
+    return c
+
 def dis(x=None):
     """Disassemble classes, methods, functions, or code.
 
@@ -38,7 +54,9 @@
     elif hasattr(x, 'co_code'):
         disassemble(x)
     elif isinstance(x, (bytes, bytearray)):
-        disassemble_string(x)
+        _disassemble_bytes(x)
+    elif isinstance(x, str):
+        _disassemble_str(x)
     else:
         raise TypeError("don't know how to disassemble %s objects" %
                         type(x).__name__)
@@ -157,7 +175,7 @@
                 print('(' + free[oparg] + ')', end=' ')
         print()
 
-def disassemble_string(code, lasti=-1, varnames=None, names=None,
+def _disassemble_bytes(code, lasti=-1, varnames=None, names=None,
                        constants=None):
     labels = findlabels(code)
     n = len(code)
@@ -196,6 +214,10 @@
                 print('(' + cmp_op[oparg] + ')', end=' ')
         print()
 
+def _disassemble_str(source):
+    """Compile the source string, then disassemble the code object."""
+    disassemble(_try_compile(source, '<dis>'))
+
 disco = disassemble                     # XXX For backwards compatibility
 
 def findlabels(code):

Modified: python/branches/py3k/Lib/test/test_dis.py
==============================================================================
--- python/branches/py3k/Lib/test/test_dis.py	(original)
+++ python/branches/py3k/Lib/test/test_dis.py	Sat Jul  3 09:36:51 2010
@@ -96,6 +96,46 @@
 
 """
 
+expr_str = "x + 1"
+
+dis_expr_str = """\
+  1           0 LOAD_NAME                0 (x)
+              3 LOAD_CONST               0 (1)
+              6 BINARY_ADD
+              7 RETURN_VALUE
+"""
+
+simple_stmt_str = "x = x + 1"
+
+dis_simple_stmt_str = """\
+  1           0 LOAD_NAME                0 (x)
+              3 LOAD_CONST               0 (1)
+              6 BINARY_ADD
+              7 STORE_NAME               0 (x)
+             10 LOAD_CONST               1 (None)
+             13 RETURN_VALUE
+"""
+
+compound_stmt_str = """\
+x = 0
+while 1:
+    x += 1"""
+# Trailing newline has been deliberately omitted
+
+dis_compound_stmt_str = """\
+  1           0 LOAD_CONST               0 (0)
+              3 STORE_NAME               0 (x)
+
+  2           6 SETUP_LOOP              13 (to 22)
+
+  3     >>    9 LOAD_NAME                0 (x)
+             12 LOAD_CONST               1 (1)
+             15 INPLACE_ADD
+             16 STORE_NAME               0 (x)
+             19 JUMP_ABSOLUTE            9
+        >>   22 LOAD_CONST               2 (None)
+             25 RETURN_VALUE
+"""
 
 class DisTests(unittest.TestCase):
     def do_disassembly_test(self, func, expected):
@@ -166,6 +206,11 @@
         from test import dis_module
         self.do_disassembly_test(dis_module, dis_module_expected_results)
 
+    def test_disassemble_str(self):
+        self.do_disassembly_test(expr_str, dis_expr_str)
+        self.do_disassembly_test(simple_stmt_str, dis_simple_stmt_str)
+        self.do_disassembly_test(compound_stmt_str, dis_compound_stmt_str)
+
 def test_main():
     run_unittest(DisTests)
 

Modified: python/branches/py3k/Misc/ACKS
==============================================================================
--- python/branches/py3k/Misc/ACKS	(original)
+++ python/branches/py3k/Misc/ACKS	Sat Jul  3 09:36:51 2010
@@ -797,6 +797,7 @@
 Eren T?rkay
 Lionel Ulmer
 Roger Upole
+Daniel Urban
 Michael Urman
 Hector Urtubia
 Andi Vajda

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sat Jul  3 09:36:51 2010
@@ -468,6 +468,8 @@
 Library
 -------
 
+- Issue #6507: Accept source strings in dis.dis()
+
 - Issue #7829: Clearly document that the dis module is exposing an
   implementation detail that is not stable between Python VMs or releases.
 

From python-checkins at python.org  Sat Jul  3 09:38:29 2010
From: python-checkins at python.org (nick.coghlan)
Date: Sat,  3 Jul 2010 09:38:29 +0200 (CEST)
Subject: [Python-checkins] r82472 - python/branches/py3k/Misc/NEWS
Message-ID: <20100703073829.0CD27EF59@mail.python.org>

Author: nick.coghlan
Date: Sat Jul  3 09:38:28 2010
New Revision: 82472

Log:
Issue 6507: missing patch submitter from NEWS entry in previous checkin

Modified:
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sat Jul  3 09:38:28 2010
@@ -468,7 +468,8 @@
 Library
 -------
 
-- Issue #6507: Accept source strings in dis.dis()
+- Issue #6507: Accept source strings in dis.dis(). Original patch by
+  Daniel Urban.
 
 - Issue #7829: Clearly document that the dis module is exposing an
   implementation detail that is not stable between Python VMs or releases.

From python-checkins at python.org  Sat Jul  3 09:43:04 2010
From: python-checkins at python.org (ezio.melotti)
Date: Sat,  3 Jul 2010 09:43:04 +0200 (CEST)
Subject: [Python-checkins] r82473 - in python/trunk/Doc: using/cmdline.rst
	whatsnew/2.7.rst
Message-ID: <20100703074304.DFEA8F416@mail.python.org>

Author: ezio.melotti
Date: Sat Jul  3 09:43:04 2010
New Revision: 82473

Log:
Update whatsnew/2.7.rst, add a few links and fix a couple of XXX.  Also add a paragraph about DeprecationWarnings in cmdline.rst.

Modified:
   python/trunk/Doc/using/cmdline.rst
   python/trunk/Doc/whatsnew/2.7.rst

Modified: python/trunk/Doc/using/cmdline.rst
==============================================================================
--- python/trunk/Doc/using/cmdline.rst	(original)
+++ python/trunk/Doc/using/cmdline.rst	Sat Jul  3 09:43:04 2010
@@ -315,6 +315,10 @@
    :option:`-W` options are ignored (though, a warning message is printed about
    invalid options when the first warning is issued).
 
+   Starting from Python 2.7, :exc:`DeprecationWarning` and its descendants
+   are ignored by default.  The :option:`-Wd` option can be used to re-enable
+   them.
+
    Warnings can also be controlled from within a Python program using the
    :mod:`warnings` module.
 

Modified: python/trunk/Doc/whatsnew/2.7.rst
==============================================================================
--- python/trunk/Doc/whatsnew/2.7.rst	(original)
+++ python/trunk/Doc/whatsnew/2.7.rst	Sat Jul  3 09:43:04 2010
@@ -9,7 +9,6 @@
 ..  hyperlink all the methods & functions.
 
 .. T_STRING_INPLACE not described in main docs
-.. "Format String Syntax" in string.rst could use many more examples.
 
 .. $Id$
    Rules for maintenance:
@@ -57,12 +56,12 @@
 schedule is described in :pep:`373`.
 
 Numeric handling has been improved in many ways, for both
-floating-point numbers and for the :class:`Decimal` class.  There are
-some useful additions to the standard library, such as a greatly
-enhanced :mod:`unittest` module, the :mod:`argparse` module for
-parsing command-line options, convenient ordered-dictionary and
-:class:`Counter` classes in the :mod:`collections` module, and many
-other improvements.
+floating-point numbers and for the :class:`~decimal.Decimal` class.
+There are some useful additions to the standard library, such as a
+greatly enhanced :mod:`unittest` module, the :mod:`argparse` module
+for parsing command-line options, convenient :class:`~collections.OrderedDict`
+and :class:`~collections.Counter` classes in the :mod:`collections` module,
+and many other improvements.
 
 Python 2.7 is planned to be the last of the 2.x releases, so we worked
 on making it a good release for the long term.  To help with porting
@@ -117,8 +116,8 @@
   with responding to these concerns.
 
   You can re-enable display of :exc:`DeprecationWarning` messages by
-  running Python with the :option:`-Wdefault` (short form:
-  :option:`-Wd`) switch, or by setting the :envvar:`PYTHONWARNINGS`
+  running Python with the :option:`-Wdefault <-W>` (short form:
+  :option:`-Wd <-W>`) switch, or by setting the :envvar:`PYTHONWARNINGS`
   environment variable to ``"default"`` (or ``"d"``) before running
   Python.  Python code can also re-enable them
   by calling ``warnings.simplefilter('default')``.
@@ -135,7 +134,7 @@
 A partial list of 3.1 features that were backported to 2.7:
 
 * The syntax for set literals (``{1,2,3}`` is a mutable set).
-* Dictionary and set comprehensions (``{ i: i*2 for i in range(3)}``).
+* Dictionary and set comprehensions (``{i: i*2 for i in range(3)}``).
 * Multiple context managers in a single :keyword:`with` statement.
 * A new version of the :mod:`io` library, rewritten in C for performance.
 * The ordered-dictionary type described in :ref:`pep-0372`.
@@ -155,7 +154,7 @@
 * :func:`operator.isCallable` and :func:`operator.sequenceIncludes`,
   which are not supported in 3.x, now trigger warnings.
 * The :option:`-3` switch now automatically
-  enables the :option:`-Qwarn` switch that causes warnings
+  enables the :option:`-Qwarn <-Q>` switch that causes warnings
   about using classic division with integers and long integers.
 
 
@@ -390,7 +389,8 @@
 
 .. seealso::
 
-   `argparse module documentation <http://docs.python.org/dev/library/argparse.html>`__
+   `argparse documentation <http://docs.python.org/dev/library/argparse.html>`__
+     The documentation page of the argparse module.
 
    `Upgrading optparse code to use argparse <http://docs.python.org/dev/library/argparse.html#upgrading-optparse-code>`__
      Part of the Python documentation, describing how to convert
@@ -402,8 +402,6 @@
 PEP 391: Dictionary-Based Configuration For Logging
 ====================================================
 
-.. XXX not documented in library reference yet; add link here once it's added.
-
 The :mod:`logging` module is very flexible; applications can define
 a tree of logging subsystems, and each logger in this tree can filter
 out certain messages, format them differently, and direct messages to
@@ -412,21 +410,21 @@
 All this flexibility can require a lot of configuration.  You can
 write Python statements to create objects and set their properties,
 but a complex set-up requires verbose but boring code.
-:mod:`logging` also supports a :func:`~logging.config.fileConfig`
+:mod:`logging` also supports a :func:`~logging.fileConfig`
 function that parses a file, but the file format doesn't support
 configuring filters, and it's messier to generate programmatically.
 
-Python 2.7 adds a :func:`~logging.config.dictConfig` function that
+Python 2.7 adds a :func:`~logging.dictConfig` function that
 uses a dictionary to configure logging.  There are many ways to
 produce a dictionary from different sources: construct one with code;
 parse a file containing JSON; or use a YAML parsing library if one is
-installed.
+installed.  For more information see :ref:`logging-config-api`.
 
 The following example configures two loggers, the root logger and a
-logger named "network".   Messages sent to the root logger will be
+logger named "network".  Messages sent to the root logger will be
 sent to the system log using the syslog protocol, and messages
 to the "network" logger will be written to a :file:`network.log` file
-that will be rotated once the log reaches 1Mb.
+that will be rotated once the log reaches 1MB.
 
 ::
 
@@ -445,7 +443,7 @@
                          'filename': '/logs/network.log',
                          'formatter': 'standard',
                          'level': 'INFO',
-                         'maxBytes': 1024*1024},
+                         'maxBytes': 1000000},
                   'syslog': {'class': 'logging.handlers.SysLogHandler',
                              'formatter': 'standard',
                              'level': 'ERROR'}},
@@ -483,16 +481,19 @@
   for UDP or :const:`socket.SOCK_STREAM` for TCP.  The default
   protocol remains UDP.
 
-* :class:`Logger` instances gained a :meth:`getChild` method that retrieves a
-  descendant logger using a relative path.  For example,
-  once you retrieve a logger by doing ``log = getLogger('app')``,
+* :class:`~logging.Logger` instances gained a :meth:`~logging.Logger.getChild`
+  method that retrieves a descendant logger using a relative path.
+  For example, once you retrieve a logger by doing ``log = getLogger('app')``,
   calling ``log.getChild('network.listen')`` is equivalent to
   ``getLogger('app.network.listen')``.
 
-* The :class:`LoggerAdapter` class gained a :meth:`isEnabledFor` method
-  that takes a *level* and returns whether the underlying logger would
+* The :class:`~logging.LoggerAdapter` class gained a
+  :meth:`~logging.LoggerAdapter.isEnabledFor` method that takes a
+  *level* and returns whether the underlying logger would
   process a message of that level of importance.
 
+.. XXX: Logger objects don't have a class declaration so the link don't work
+
 .. seealso::
 
    :pep:`391` - Dictionary-Based Configuration For Logging
@@ -501,14 +502,15 @@
 PEP 3106: Dictionary Views
 ====================================================
 
-The dictionary methods :meth:`keys`, :meth:`values`, and :meth:`items`
-are different in Python 3.x.  They return an object called a :dfn:`view`
-instead of a fully materialized list.
-
-It's not possible to change the return values of :meth:`keys`,
-:meth:`values`, and :meth:`items` in Python 2.7 because too much code
-would break.  Instead the 3.x versions were added under the new names
-:meth:`viewkeys`, :meth:`viewvalues`, and :meth:`viewitems`.
+The dictionary methods :meth:`~dict.keys`, :meth:`~dict.values`, and
+:meth:`~dict.items` are different in Python 3.x.  They return an object
+called a :dfn:`view` instead of a fully materialized list.
+
+It's not possible to change the return values of :meth:`~dict.keys`,
+:meth:`~dict.values`, and :meth:`~dict.items` in Python 2.7 because
+too much code would break.  Instead the 3.x versions were added
+under the new names :meth:`~dict.viewkeys`, :meth:`~dict.viewvalues`,
+and :meth:`~dict.viewitems`.
 
 ::
 
@@ -550,8 +552,8 @@
     RuntimeError: dictionary changed size during iteration
 
 You can use the view methods in Python 2.x code, and the 2to3
-converter will change them to the standard :meth:`keys`,
-:meth:`values`, and :meth:`items` methods.
+converter will change them to the standard :meth:`~dict.keys`,
+:meth:`~dict.values`, and :meth:`~dict.items` methods.
 
 .. seealso::
 
@@ -624,7 +626,7 @@
   ``{}`` continues to represent an empty dictionary; use
   ``set()`` for an empty set.
 
-    >>> {1,2,3,4,5}
+    >>> {1, 2, 3, 4, 5}
     set([1, 2, 3, 4, 5])
     >>> set() # empty set
     set([])
@@ -794,7 +796,7 @@
   ``None`` as its first argument.  (Fixed by Georg Brandl;
   :issue:`4759`.)
 
-  .. bytearray doesn't seem to be documented
+  .. XXX bytearray doesn't seem to be documented
 
 * When using ``@classmethod`` and ``@staticmethod`` to wrap
   methods as class or static methods, the wrapper object now
@@ -1054,7 +1056,7 @@
   :meth:`~collections.deque.count` method that returns the number of
   contained elements equal to the supplied argument *x*, and a
   :meth:`~collections.deque.reverse` method that reverses the elements
-  of the deque in-place.  :class:`deque` also exposes its maximum
+  of the deque in-place.  :class:`~collections.deque` also exposes its maximum
   length as the read-only :attr:`~collections.deque.maxlen` attribute.
   (Both features added by Raymond Hettinger.)
 
@@ -1135,15 +1137,14 @@
   ``Decimal('0.1000000000000000055511151231257827021181583404541015625')``.
   (Implemented by Raymond Hettinger; :issue:`4796`.)
 
-  Comparing instances of :class:`Decimal` with floating-point
+  Comparing instances of :class:`~decimal.Decimal` with floating-point
   numbers now produces sensible results based on the numeric values
   of the operands.  Previously such comparisons would fall back to
   Python's default rules for comparing objects, which produced arbitrary
   results based on their type.  Note that you still cannot combine
   :class:`Decimal` and floating-point in other operations such as addition,
   since you should be explicitly choosing how to convert between float and
-  :class:`Decimal`.
-  (Fixed by Mark Dickinson; :issue:`2531`.)
+  :class:`~decimal.Decimal`.  (Fixed by Mark Dickinson; :issue:`2531`.)
 
   The constructor for :class:`~decimal.Decimal` now accepts
   floating-point numbers (added by Raymond Hettinger; :issue:`8257`)
@@ -1195,8 +1196,8 @@
 
   Ordering comparisons (``<``, ``<=``, ``>``, ``>=``) between
   fractions and complex numbers now raise a :exc:`TypeError`.
-  This fixes an oversight, making the :class:`Fraction` match the other
-  numeric types.
+  This fixes an oversight, making the :class:`~fractions.Fraction`
+  match the other numeric types.
 
   .. revision 79455
 
@@ -1210,7 +1211,7 @@
   uploads thanks to an added *rest* parameter (patch by Pablo Mouzo;
   :issue:`6845`.)
 
-* New class decorator: :func:`total_ordering` in the :mod:`functools`
+* New class decorator: :func:`~functools.total_ordering` in the :mod:`functools`
   module takes a class that defines an :meth:`__eq__` method and one of
   :meth:`__lt__`, :meth:`__le__`, :meth:`__gt__`, or :meth:`__ge__`,
   and generates the missing comparison methods.  Since the
@@ -1218,7 +1219,7 @@
   this decorator makes it easier to define ordered classes.
   (Added by Raymond Hettinger; :issue:`5479`.)
 
-  New function: :func:`cmp_to_key` will take an old-style comparison
+  New function: :func:`~functools.cmp_to_key` will take an old-style comparison
   function that expects two arguments and return a new callable that
   can be used as the *key* parameter to functions such as
   :func:`sorted`, :func:`min` and :func:`max`, etc.  The primary
@@ -1345,7 +1346,7 @@
   with any object literal that decodes to a list of pairs.
   (Contributed by Raymond Hettinger; :issue:`5381`.)
 
-* The :mod:`mailbox` module's :class:`Maildir` class now records the
+* The :mod:`mailbox` module's :class:`~mailbox.Maildir` class now records the
   timestamp on the directories it reads, and only re-reads them if the
   modification time has subsequently changed.  This improves
   performance by avoiding unneeded directory scans.  (Fixed by
@@ -1466,10 +1467,10 @@
   defaults to False; if overridden to be True,
   new request connections will have the TCP_NODELAY option set to
   prevent buffering many small sends into a single TCP packet.
-  The :attr:`~SocketServer.TCPServer.timeout` class attribute can hold
+  The :attr:`~SocketServer.BaseServer.timeout` class attribute can hold
   a timeout in seconds that will be applied to the request socket; if
-  no request is received within that time, :meth:`handle_timeout`
-  will be called and :meth:`handle_request` will return.
+  no request is received within that time, :meth:`~SocketServer.BaseServer.handle_timeout`
+  will be called and :meth:`~SocketServer.BaseServer.handle_request` will return.
   (Contributed by Kristj?n Valur J?nsson; :issue:`6192` and :issue:`6267`.)
 
 * Updated module: the :mod:`sqlite3` module has been updated to
@@ -1479,7 +1480,7 @@
   and then call :meth:`~sqlite3.Connection.load_extension` to load a particular shared library.
   (Updated by Gerhard H?ring.)
 
-* The :mod:`ssl` module's :class:`ssl.SSLSocket` objects now support the
+* The :mod:`ssl` module's :class:`~ssl.SSLSocket` objects now support the
   buffer API, which fixed a test suite failure (fix by Antoine Pitrou;
   :issue:`7133`) and automatically set
   OpenSSL's :cmacro:`SSL_MODE_AUTO_RETRY`, which will prevent an error
@@ -1535,7 +1536,7 @@
   on receiving an :const:`EINTR` signal.  (Reported by several people; final
   patch by Gregory P. Smith in :issue:`1068268`.)
 
-* New function: :func:`~symtable.is_declared_global` in the :mod:`symtable` module
+* New function: :func:`~symtable.Symbol.is_declared_global` in the :mod:`symtable` module
   returns true for variables that are explicitly declared to be global,
   false for ones that are implicitly global.
   (Contributed by Jeremy Hylton.)
@@ -1716,7 +1717,7 @@
   Makefile and the :file:`pyconfig.h` file.
 * :func:`~sysconfig.get_config_vars` returns a dictionary containing
   all of the configuration variables.
-* :func:`~sysconfig.getpath` returns the configured path for
+* :func:`~sysconfig.get_path` returns the configured path for
   a particular type of module: the standard library,
   site-specific modules, platform-specific modules, etc.
 * :func:`~sysconfig.is_python_build` returns true if you're running a
@@ -1778,7 +1779,7 @@
 Consult the :mod:`unittest` module documentation for more details.
 (Developed in :issue:`6001`.)
 
-The :func:`main` function supports some other new options:
+The :func:`~unittest.main` function supports some other new options:
 
 * :option:`-b` or :option:`--buffer` will buffer the standard output
   and standard error streams during each test.  If the test passes,
@@ -1796,7 +1797,7 @@
   being tested or the tests being run have defined a signal handler of
   their own, by noticing that a signal handler was already set and
   calling it.  If this doesn't work for you, there's a
-  :func:`removeHandler` decorator that can be used to mark tests that
+  :func:`~unittest.removeHandler` decorator that can be used to mark tests that
   should have the control-C handling disabled.
 
 * :option:`-f` or :option:`--failfast` makes
@@ -1923,7 +1924,7 @@
 
 :func:`unittest.main` now takes an optional ``exit`` argument.  If
 False, :func:`~unittest.main` doesn't call :func:`sys.exit`, allowing
-:func:`main` to be used from the interactive interpreter.
+:func:`~unittest.main` to be used from the interactive interpreter.
 (Contributed by J. Pablo Fern?ndez; :issue:`3379`.)
 
 :class:`~unittest.TestResult` has new :meth:`~unittest.TestResult.startTestRun` and
@@ -2117,7 +2118,7 @@
   :cmacro:`Py_ISSPACE`,
   :cmacro:`Py_ISUPPER`,
   :cmacro:`Py_ISXDIGIT`,
-  and :cmacro:`Py_TOLOWER`, :cmacro:`Py_TOUPPER`.
+  :cmacro:`Py_TOLOWER`, and :cmacro:`Py_TOUPPER`.
   All of these functions are analogous to the C
   standard macros for classifying characters, but ignore the current
   locale setting, because in
@@ -2263,11 +2264,11 @@
   (Contributed by David Cournapeau; :issue:`4365`.)
 
 * The :mod:`_winreg` module for accessing the registry now implements
-  the :func:`CreateKeyEx` and :func:`DeleteKeyEx` functions, extended
-  versions of previously-supported functions that take several extra
-  arguments.  The :func:`DisableReflectionKey`,
-  :func:`EnableReflectionKey`, and :func:`QueryReflectionKey` were also
-  tested and documented.
+  the :func:`~_winreg.CreateKeyEx` and :func:`~_winreg.DeleteKeyEx`
+  functions, extended versions of previously-supported functions that
+  take several extra arguments.  The :func:`~_winreg.DisableReflectionKey`,
+  :func:`~_winreg.EnableReflectionKey`, and :func:`~_winreg.QueryReflectionKey`
+  were also tested and documented.
   (Implemented by Brian Curtin: :issue:`7347`.)
 
 * The new :cfunc:`_beginthreadex` API is used to start threads, and
@@ -2384,20 +2385,20 @@
 
 In the standard library:
 
-* Operations with :class:`datetime` instances that resulted in a year
+* Operations with :class:`~datetime.datetime` instances that resulted in a year
   falling outside the supported range didn't always raise
   :exc:`OverflowError`.  Such errors are now checked more carefully
   and will now raise the exception. (Reported by Mark Leander, patch
   by Anand B. Pillai and Alexander Belopolsky; :issue:`7150`.)
 
-* When using :class:`Decimal` instances with a string's
+* When using :class:`~decimal.Decimal` instances with a string's
   :meth:`format` method, the default alignment was previously
   left-alignment.  This has been changed to right-alignment, which might
   change the output of your programs.
   (Changed by Mark Dickinson; :issue:`6857`.)
 
   Comparisons involving a signaling NaN value (or ``sNAN``) now signal
-  :const:`InvalidOperation` instead of silently returning a true or
+  :const:`~decimal.InvalidOperation` instead of silently returning a true or
   false value depending on the comparison operator.  Quiet NaN values
   (or ``NaN``) are now hashable.  (Fixed by Mark Dickinson;
   :issue:`7279`.)
@@ -2408,7 +2409,7 @@
   or comment (which looks like `<!-- comment -->`).
   (Patch by Neil Muller; :issue:`2746`.)
 
-* The :meth:`readline` method of :class:`StringIO` objects now does
+* The :meth:`~StringIO.StringIO.readline` method of :class:`~StringIO.StringIO` objects now does
   nothing when a negative length is requested, as other file-like
   objects do.  (:issue:`7348`).
 

From python-checkins at python.org  Sat Jul  3 10:40:13 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat,  3 Jul 2010 10:40:13 +0200 (CEST)
Subject: [Python-checkins] r82474 -
	python/trunk/Doc/library/simplehttpserver.rst
Message-ID: <20100703084013.71674F800@mail.python.org>

Author: georg.brandl
Date: Sat Jul  3 10:40:13 2010
New Revision: 82474

Log:
Fix role name.

Modified:
   python/trunk/Doc/library/simplehttpserver.rst

Modified: python/trunk/Doc/library/simplehttpserver.rst
==============================================================================
--- python/trunk/Doc/library/simplehttpserver.rst	(original)
+++ python/trunk/Doc/library/simplehttpserver.rst	Sat Jul  3 10:40:13 2010
@@ -106,7 +106,7 @@
    httpd.serve_forever()
 
 The :mod:`SimpleHTTPServer` module can also be invoked directly using the
-:cmdoption:`-m` switch of the interpreter with a ``port number`` argument.
+:option:`-m` switch of the interpreter with a ``port number`` argument.
 Similar to the previous example, this serves the files relative to the
 current directory. ::
 

From python-checkins at python.org  Sat Jul  3 10:42:33 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat,  3 Jul 2010 10:42:33 +0200 (CEST)
Subject: [Python-checkins] r82475 - python/trunk/Doc/whatsnew/2.7.rst
Message-ID: <20100703084233.B673EF800@mail.python.org>

Author: georg.brandl
Date: Sat Jul  3 10:42:33 2010
New Revision: 82475

Log:
Use internal links.

Modified:
   python/trunk/Doc/whatsnew/2.7.rst

Modified: python/trunk/Doc/whatsnew/2.7.rst
==============================================================================
--- python/trunk/Doc/whatsnew/2.7.rst	(original)
+++ python/trunk/Doc/whatsnew/2.7.rst	Sat Jul  3 10:42:33 2010
@@ -389,10 +389,10 @@
 
 .. seealso::
 
-   `argparse documentation <http://docs.python.org/dev/library/argparse.html>`__
+   :mod:`argparse` documentation
      The documentation page of the argparse module.
 
-   `Upgrading optparse code to use argparse <http://docs.python.org/dev/library/argparse.html#upgrading-optparse-code>`__
+   :ref:`argparse-from-optparse`
      Part of the Python documentation, describing how to convert
      code that uses :mod:`optparse`.
 

From python-checkins at python.org  Sat Jul  3 11:15:10 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sat,  3 Jul 2010 11:15:10 +0200 (CEST)
Subject: [Python-checkins] r82476 - python/trunk/Doc/library/sys.rst
Message-ID: <20100703091510.0BF2BF416@mail.python.org>

Author: mark.dickinson
Date: Sat Jul  3 11:15:09 2010
New Revision: 82476

Log:
Fix typo in sys.float_info docs.

Modified:
   python/trunk/Doc/library/sys.rst

Modified: python/trunk/Doc/library/sys.rst
==============================================================================
--- python/trunk/Doc/library/sys.rst	(original)
+++ python/trunk/Doc/library/sys.rst	Sat Jul  3 11:15:09 2010
@@ -305,7 +305,7 @@
    +---------------------+----------------+--------------------------------------------------+
    | attribute           | float.h macro  | explanation                                      |
    +=====================+================+==================================================+
-   | :const:`epsilon`    | DBL_MAX        | difference between 1 and the least value greater |
+   | :const:`epsilon`    | DBL_EPSILON    | difference between 1 and the least value greater |
    |                     |                | than 1 that is representable as a float          |
    +---------------------+----------------+--------------------------------------------------+
    | :const:`dig`        | DBL_DIG        | maximum number of decimal digits that can be     |

From python-checkins at python.org  Sat Jul  3 11:16:01 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sat,  3 Jul 2010 11:16:01 +0200 (CEST)
Subject: [Python-checkins] r82477 - in python/branches/release26-maint:
	Doc/library/sys.rst
Message-ID: <20100703091601.823AAEE9A7@mail.python.org>

Author: mark.dickinson
Date: Sat Jul  3 11:16:01 2010
New Revision: 82477

Log:
Merged revisions 82476 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/trunk

........
  r82476 | mark.dickinson | 2010-07-03 10:15:09 +0100 (Sat, 03 Jul 2010) | 1 line
  
  Fix typo in sys.float_info docs.
........


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Doc/library/sys.rst

Modified: python/branches/release26-maint/Doc/library/sys.rst
==============================================================================
--- python/branches/release26-maint/Doc/library/sys.rst	(original)
+++ python/branches/release26-maint/Doc/library/sys.rst	Sat Jul  3 11:16:01 2010
@@ -305,7 +305,7 @@
    +---------------------+----------------+--------------------------------------------------+
    | attribute           | float.h macro  | explanation                                      |
    +=====================+================+==================================================+
-   | :const:`epsilon`    | DBL_MAX        | difference between 1 and the least value greater |
+   | :const:`epsilon`    | DBL_EPSILON    | difference between 1 and the least value greater |
    |                     |                | than 1 that is representable as a float          |
    +---------------------+----------------+--------------------------------------------------+
    | :const:`dig`        | DBL_DIG        | maximum number of decimal digits that can be     |

From python-checkins at python.org  Sat Jul  3 11:17:16 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sat,  3 Jul 2010 11:17:16 +0200 (CEST)
Subject: [Python-checkins] r82478 - in python/branches/py3k:
	Doc/library/sys.rst
Message-ID: <20100703091716.4B7A5F416@mail.python.org>

Author: mark.dickinson
Date: Sat Jul  3 11:17:16 2010
New Revision: 82478

Log:
Merged revisions 82476 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/trunk

........
  r82476 | mark.dickinson | 2010-07-03 10:15:09 +0100 (Sat, 03 Jul 2010) | 1 line
  
  Fix typo in sys.float_info docs.
........


Modified:
   python/branches/py3k/   (props changed)
   python/branches/py3k/Doc/library/sys.rst

Modified: python/branches/py3k/Doc/library/sys.rst
==============================================================================
--- python/branches/py3k/Doc/library/sys.rst	(original)
+++ python/branches/py3k/Doc/library/sys.rst	Sat Jul  3 11:17:16 2010
@@ -232,7 +232,7 @@
    +---------------------+----------------+--------------------------------------------------+
    | attribute           | float.h macro  | explanation                                      |
    +=====================+================+==================================================+
-   | :const:`epsilon`    | DBL_MAX        | difference between 1 and the least value greater |
+   | :const:`epsilon`    | DBL_EPSILON    | difference between 1 and the least value greater |
    |                     |                | than 1 that is representable as a float          |
    +---------------------+----------------+--------------------------------------------------+
    | :const:`dig`        | DBL_DIG        | maximum number of decimal digits that can be     |

From python-checkins at python.org  Sat Jul  3 11:18:26 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sat,  3 Jul 2010 11:18:26 +0200 (CEST)
Subject: [Python-checkins] r82479 - in python/branches/release31-maint:
	Doc/library/sys.rst
Message-ID: <20100703091826.687DCEE9A7@mail.python.org>

Author: mark.dickinson
Date: Sat Jul  3 11:18:26 2010
New Revision: 82479

Log:
Merged revisions 82478 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

................
  r82478 | mark.dickinson | 2010-07-03 10:17:16 +0100 (Sat, 03 Jul 2010) | 9 lines
  
  Merged revisions 82476 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82476 | mark.dickinson | 2010-07-03 10:15:09 +0100 (Sat, 03 Jul 2010) | 1 line
    
    Fix typo in sys.float_info docs.
  ........
................


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Doc/library/sys.rst

Modified: python/branches/release31-maint/Doc/library/sys.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/sys.rst	(original)
+++ python/branches/release31-maint/Doc/library/sys.rst	Sat Jul  3 11:18:26 2010
@@ -233,7 +233,7 @@
    +---------------------+----------------+--------------------------------------------------+
    | attribute           | float.h macro  | explanation                                      |
    +=====================+================+==================================================+
-   | :const:`epsilon`    | DBL_MAX        | difference between 1 and the least value greater |
+   | :const:`epsilon`    | DBL_EPSILON    | difference between 1 and the least value greater |
    |                     |                | than 1 that is representable as a float          |
    +---------------------+----------------+--------------------------------------------------+
    | :const:`dig`        | DBL_DIG        | maximum number of decimal digits that can be     |

From python-checkins at python.org  Sat Jul  3 12:21:51 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat,  3 Jul 2010 12:21:51 +0200 (CEST)
Subject: [Python-checkins] r82480 - python/branches/py3k/Doc/library/dis.rst
Message-ID: <20100703102151.21FB7EFCB@mail.python.org>

Author: georg.brandl
Date: Sat Jul  3 12:21:50 2010
New Revision: 82480

Log:
Wrap and use the correct directive.

Modified:
   python/branches/py3k/Doc/library/dis.rst

Modified: python/branches/py3k/Doc/library/dis.rst
==============================================================================
--- python/branches/py3k/Doc/library/dis.rst	(original)
+++ python/branches/py3k/Doc/library/dis.rst	Sat Jul  3 12:21:50 2010
@@ -10,7 +10,13 @@
 input is defined in the file :file:`Include/opcode.h` and used by the compiler
 and the interpreter.
 
-.. warning:: Bytecode is an implementation detail of the CPython interpreter!  No guarantees are made that bytecode will not be added, removed, or changed between versions of Python. Use of this module should not be considered to work across Python VMs or Python releases.
+.. impl-detail::
+
+   Bytecode is an implementation detail of the CPython interpreter!  No
+   guarantees are made that bytecode will not be added, removed, or changed
+   between versions of Python.  Use of this module should not be considered to
+   work across Python VMs or Python releases.
+
 
 Example: Given the function :func:`myfunc`::
 

From python-checkins at python.org  Sat Jul  3 12:22:11 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat,  3 Jul 2010 12:22:11 +0200 (CEST)
Subject: [Python-checkins] r82481 -
	python/branches/py3k/Doc/library/http.server.rst
Message-ID: <20100703102211.1E6B3EE986@mail.python.org>

Author: georg.brandl
Date: Sat Jul  3 12:22:10 2010
New Revision: 82481

Log:
Use the right role.

Modified:
   python/branches/py3k/Doc/library/http.server.rst

Modified: python/branches/py3k/Doc/library/http.server.rst
==============================================================================
--- python/branches/py3k/Doc/library/http.server.rst	(original)
+++ python/branches/py3k/Doc/library/http.server.rst	Sat Jul  3 12:22:10 2010
@@ -301,7 +301,7 @@
    print("serving at port", PORT)
    httpd.serve_forever()
 
-:mod:`http.server` can also be invoked directly using the :cmdoption:`-m`
+:mod:`http.server` can also be invoked directly using the :option:`-m`
 switch of the interpreter a with ``port number`` argument.  Similar to
 the previous example, this serves files relative to the current directory. ::
 

From python-checkins at python.org  Sat Jul  3 12:25:47 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat,  3 Jul 2010 12:25:47 +0200 (CEST)
Subject: [Python-checkins] r82482 - python/trunk/Doc/library/argparse.rst
Message-ID: <20100703102547.432ABF0BA@mail.python.org>

Author: georg.brandl
Date: Sat Jul  3 12:25:47 2010
New Revision: 82482

Log:
Add missing target.

Modified:
   python/trunk/Doc/library/argparse.rst

Modified: python/trunk/Doc/library/argparse.rst
==============================================================================
--- python/trunk/Doc/library/argparse.rst	(original)
+++ python/trunk/Doc/library/argparse.rst	Sat Jul  3 12:25:47 2010
@@ -1694,6 +1694,8 @@
             yield arg
 
 
+.. _argparse-from-optparse:
+
 Upgrading optparse code
 -----------------------
 

From python-checkins at python.org  Sat Jul  3 12:25:54 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat,  3 Jul 2010 12:25:54 +0200 (CEST)
Subject: [Python-checkins] r82483 - python/trunk/Doc/glossary.rst
Message-ID: <20100703102554.2B838EE996@mail.python.org>

Author: georg.brandl
Date: Sat Jul  3 12:25:54 2010
New Revision: 82483

Log:
Add link to bytecode docs.

Modified:
   python/trunk/Doc/glossary.rst

Modified: python/trunk/Doc/glossary.rst
==============================================================================
--- python/trunk/Doc/glossary.rst	(original)
+++ python/trunk/Doc/glossary.rst	Sat Jul  3 12:25:54 2010
@@ -63,6 +63,9 @@
       "intermediate language" is said to run on a :term:`virtual machine`
       that executes the machine code corresponding to each bytecode.
 
+      A list of bytecode instructions can be found in the documentation for
+      :ref:`the dis module <bytecodes>`.
+
    class
       A template for creating user-defined objects. Class definitions
       normally contain method definitions which operate on instances of the

From python-checkins at python.org  Sat Jul  3 12:26:18 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat,  3 Jul 2010 12:26:18 +0200 (CEST)
Subject: [Python-checkins] r82484 - python/branches/py3k
Message-ID: <20100703102618.28E1FF0BA@mail.python.org>

Author: georg.brandl
Date: Sat Jul  3 12:26:17 2010
New Revision: 82484

Log:
Recorded merge of revisions 82474 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/trunk

........
  r82474 | georg.brandl | 2010-07-03 10:40:13 +0200 (Sa, 03 Jul 2010) | 1 line
  
  Fix role name.
........


Modified:
   python/branches/py3k/   (props changed)

From python-checkins at python.org  Sat Jul  3 12:26:54 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat,  3 Jul 2010 12:26:54 +0200 (CEST)
Subject: [Python-checkins] r82485 - in python/branches/py3k: Doc/glossary.rst
Message-ID: <20100703102654.62ED2F0BA@mail.python.org>

Author: georg.brandl
Date: Sat Jul  3 12:26:54 2010
New Revision: 82485

Log:
Merged revisions 82483 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/trunk

........
  r82483 | georg.brandl | 2010-07-03 12:25:54 +0200 (Sa, 03 Jul 2010) | 1 line
  
  Add link to bytecode docs.
........


Modified:
   python/branches/py3k/   (props changed)
   python/branches/py3k/Doc/glossary.rst

Modified: python/branches/py3k/Doc/glossary.rst
==============================================================================
--- python/branches/py3k/Doc/glossary.rst	(original)
+++ python/branches/py3k/Doc/glossary.rst	Sat Jul  3 12:26:54 2010
@@ -66,6 +66,9 @@
       different Python virtual machines, nor to be stable between Python
       releases.
 
+      A list of bytecode instructions can be found in the documentation for
+      :ref:`the dis module <bytecodes>`.
+
    class
       A template for creating user-defined objects. Class definitions
       normally contain method definitions which operate on instances of the

From python-checkins at python.org  Sat Jul  3 12:29:21 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat,  3 Jul 2010 12:29:21 +0200 (CEST)
Subject: [Python-checkins] r82486 -
	python/branches/release26-maint/Doc/library/urllib.rst
Message-ID: <20100703102921.0D863EE982@mail.python.org>

Author: georg.brandl
Date: Sat Jul  3 12:29:20 2010
New Revision: 82486

Log:
Fix markup.

Modified:
   python/branches/release26-maint/Doc/library/urllib.rst

Modified: python/branches/release26-maint/Doc/library/urllib.rst
==============================================================================
--- python/branches/release26-maint/Doc/library/urllib.rst	(original)
+++ python/branches/release26-maint/Doc/library/urllib.rst	Sat Jul  3 12:29:20 2010
@@ -245,7 +245,7 @@
    two-element tuples is used as the *query* argument, the first element of
    each tuple is a key and the second is a value. The value element in itself
    can be a sequence and in that case, if the optional parameter *doseq* is
-   evaluates to *True*, individual ``key=value`` pairs separated by ``'&'``are
+   evaluates to *True*, individual ``key=value`` pairs separated by ``'&'`` are
    generated for each element of the value sequence for the key.  The order of
    parameters in the encoded string will match the order of parameter tuples in
    the sequence. The :mod:`urlparse` module provides the functions

From python-checkins at python.org  Sat Jul  3 12:33:26 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat,  3 Jul 2010 12:33:26 +0200 (CEST)
Subject: [Python-checkins] r82487 - python/branches/py3k/Doc/glossary.rst
Message-ID: <20100703103326.AF6BAEE982@mail.python.org>

Author: georg.brandl
Date: Sat Jul  3 12:33:26 2010
New Revision: 82487

Log:
Fix markup.

Modified:
   python/branches/py3k/Doc/glossary.rst

Modified: python/branches/py3k/Doc/glossary.rst
==============================================================================
--- python/branches/py3k/Doc/glossary.rst	(original)
+++ python/branches/py3k/Doc/glossary.rst	Sat Jul  3 12:33:26 2010
@@ -428,7 +428,6 @@
       modules, respectively.
 
    nested scope
-
       The ability to refer to a variable in an enclosing definition.  For
       instance, a function defined inside another function can refer to
       variables in the outer function.  Note that nested scopes by default work

From python-checkins at python.org  Sat Jul  3 12:41:33 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat,  3 Jul 2010 12:41:33 +0200 (CEST)
Subject: [Python-checkins] r82488 - in python/branches/py3k/Doc:
	library/dis.rst tools/sphinxext/pyspecific.py
Message-ID: <20100703104133.CC9D9EE98B@mail.python.org>

Author: georg.brandl
Date: Sat Jul  3 12:41:33 2010
New Revision: 82488

Log:
Remove the need for a "()" empty argument list after opcodes.

Modified:
   python/branches/py3k/Doc/library/dis.rst
   python/branches/py3k/Doc/tools/sphinxext/pyspecific.py

Modified: python/branches/py3k/Doc/library/dis.rst
==============================================================================
--- python/branches/py3k/Doc/library/dis.rst	(original)
+++ python/branches/py3k/Doc/library/dis.rst	Sat Jul  3 12:41:33 2010
@@ -38,9 +38,9 @@
 
 .. function:: dis(x=None)
 
-   Disassemble the *x* object. *x* can denote either a module, a
+   Disassemble the *x* object.  *x* can denote either a module, a
    class, a method, a function, a code object, a string of source code or a
-   byte sequence of raw bytecode.   For a module, it disassembles all
+   byte sequence of raw bytecode.  For a module, it disassembles all
    functions.  For a class, it disassembles all methods.  For a code object
    or sequence of raw bytecode, it prints one line per bytecode instruction.
    Strings are first compiled to code objects with the :func:`compile`
@@ -50,14 +50,15 @@
 
 .. function:: distb(tb=None)
 
-   Disassembles the top-of-stack function of a traceback, using the last traceback
-   if none was passed.  The instruction causing the exception is indicated.
+   Disassemble the top-of-stack function of a traceback, using the last
+   traceback if none was passed.  The instruction causing the exception is
+   indicated.
 
 
 .. function:: disassemble(code, lasti=-1)
               disco(code, lasti=-1)
 
-   Disassembles a code object, indicating the last instruction if *lasti* was
+   Disassemble a code object, indicating the last instruction if *lasti* was
    provided.  The output is divided in the following columns:
 
    #. the line number, for the first instruction of each line
@@ -143,225 +144,233 @@
 The Python compiler currently generates the following bytecode instructions.
 
 
-.. opcode:: STOP_CODE ()
+**General instructions**
+
+.. opcode:: STOP_CODE
 
    Indicates end-of-code to the compiler, not used by the interpreter.
 
 
-.. opcode:: NOP ()
+.. opcode:: NOP
 
    Do nothing code.  Used as a placeholder by the bytecode optimizer.
 
 
-.. opcode:: POP_TOP ()
+.. opcode:: POP_TOP
 
    Removes the top-of-stack (TOS) item.
 
 
-.. opcode:: ROT_TWO ()
+.. opcode:: ROT_TWO
 
    Swaps the two top-most stack items.
 
 
-.. opcode:: ROT_THREE ()
+.. opcode:: ROT_THREE
 
    Lifts second and third stack item one position up, moves top down to position
    three.
 
 
-.. opcode:: ROT_FOUR ()
+.. opcode:: ROT_FOUR
 
    Lifts second, third and forth stack item one position up, moves top down to
    position four.
 
 
-.. opcode:: DUP_TOP ()
+.. opcode:: DUP_TOP
 
    Duplicates the reference on top of the stack.
 
-Unary Operations take the top of the stack, apply the operation, and push the
-result back on the stack.
 
+**Unary operations**
+
+Unary operations take the top of the stack, apply the operation, and push the
+result back on the stack.
 
-.. opcode:: UNARY_POSITIVE ()
+.. opcode:: UNARY_POSITIVE
 
    Implements ``TOS = +TOS``.
 
 
-.. opcode:: UNARY_NEGATIVE ()
+.. opcode:: UNARY_NEGATIVE
 
    Implements ``TOS = -TOS``.
 
 
-.. opcode:: UNARY_NOT ()
+.. opcode:: UNARY_NOT
 
    Implements ``TOS = not TOS``.
 
 
-.. opcode:: UNARY_INVERT ()
+.. opcode:: UNARY_INVERT
 
    Implements ``TOS = ~TOS``.
 
 
-.. opcode:: GET_ITER ()
+.. opcode:: GET_ITER
 
    Implements ``TOS = iter(TOS)``.
 
+
+**Binary operations**
+
 Binary operations remove the top of the stack (TOS) and the second top-most
 stack item (TOS1) from the stack.  They perform the operation, and put the
 result back on the stack.
 
-
-.. opcode:: BINARY_POWER ()
+.. opcode:: BINARY_POWER
 
    Implements ``TOS = TOS1 ** TOS``.
 
 
-.. opcode:: BINARY_MULTIPLY ()
+.. opcode:: BINARY_MULTIPLY
 
    Implements ``TOS = TOS1 * TOS``.
 
 
-.. opcode:: BINARY_FLOOR_DIVIDE ()
+.. opcode:: BINARY_FLOOR_DIVIDE
 
    Implements ``TOS = TOS1 // TOS``.
 
 
-.. opcode:: BINARY_TRUE_DIVIDE ()
+.. opcode:: BINARY_TRUE_DIVIDE
 
    Implements ``TOS = TOS1 / TOS``.
 
 
-.. opcode:: BINARY_MODULO ()
+.. opcode:: BINARY_MODULO
 
    Implements ``TOS = TOS1 % TOS``.
 
 
-.. opcode:: BINARY_ADD ()
+.. opcode:: BINARY_ADD
 
    Implements ``TOS = TOS1 + TOS``.
 
 
-.. opcode:: BINARY_SUBTRACT ()
+.. opcode:: BINARY_SUBTRACT
 
    Implements ``TOS = TOS1 - TOS``.
 
 
-.. opcode:: BINARY_SUBSCR ()
+.. opcode:: BINARY_SUBSCR
 
    Implements ``TOS = TOS1[TOS]``.
 
 
-.. opcode:: BINARY_LSHIFT ()
+.. opcode:: BINARY_LSHIFT
 
    Implements ``TOS = TOS1 << TOS``.
 
 
-.. opcode:: BINARY_RSHIFT ()
+.. opcode:: BINARY_RSHIFT
 
    Implements ``TOS = TOS1 >> TOS``.
 
 
-.. opcode:: BINARY_AND ()
+.. opcode:: BINARY_AND
 
    Implements ``TOS = TOS1 & TOS``.
 
 
-.. opcode:: BINARY_XOR ()
+.. opcode:: BINARY_XOR
 
    Implements ``TOS = TOS1 ^ TOS``.
 
 
-.. opcode:: BINARY_OR ()
+.. opcode:: BINARY_OR
 
    Implements ``TOS = TOS1 | TOS``.
 
+
+**In-place operations**
+
 In-place operations are like binary operations, in that they remove TOS and
 TOS1, and push the result back on the stack, but the operation is done in-place
 when TOS1 supports it, and the resulting TOS may be (but does not have to be)
 the original TOS1.
 
-
-.. opcode:: INPLACE_POWER ()
+.. opcode:: INPLACE_POWER
 
    Implements in-place ``TOS = TOS1 ** TOS``.
 
 
-.. opcode:: INPLACE_MULTIPLY ()
+.. opcode:: INPLACE_MULTIPLY
 
    Implements in-place ``TOS = TOS1 * TOS``.
 
 
-.. opcode:: INPLACE_FLOOR_DIVIDE ()
+.. opcode:: INPLACE_FLOOR_DIVIDE
 
    Implements in-place ``TOS = TOS1 // TOS``.
 
 
-.. opcode:: INPLACE_TRUE_DIVIDE ()
+.. opcode:: INPLACE_TRUE_DIVIDE
 
    Implements in-place ``TOS = TOS1 / TOS``.
 
 
-.. opcode:: INPLACE_MODULO ()
+.. opcode:: INPLACE_MODULO
 
    Implements in-place ``TOS = TOS1 % TOS``.
 
 
-.. opcode:: INPLACE_ADD ()
+.. opcode:: INPLACE_ADD
 
    Implements in-place ``TOS = TOS1 + TOS``.
 
 
-.. opcode:: INPLACE_SUBTRACT ()
+.. opcode:: INPLACE_SUBTRACT
 
    Implements in-place ``TOS = TOS1 - TOS``.
 
 
-.. opcode:: INPLACE_LSHIFT ()
+.. opcode:: INPLACE_LSHIFT
 
    Implements in-place ``TOS = TOS1 << TOS``.
 
 
-.. opcode:: INPLACE_RSHIFT ()
+.. opcode:: INPLACE_RSHIFT
 
    Implements in-place ``TOS = TOS1 >> TOS``.
 
 
-.. opcode:: INPLACE_AND ()
+.. opcode:: INPLACE_AND
 
    Implements in-place ``TOS = TOS1 & TOS``.
 
 
-.. opcode:: INPLACE_XOR ()
+.. opcode:: INPLACE_XOR
 
    Implements in-place ``TOS = TOS1 ^ TOS``.
 
 
-.. opcode:: INPLACE_OR ()
+.. opcode:: INPLACE_OR
 
    Implements in-place ``TOS = TOS1 | TOS``.
 
 
-.. opcode:: STORE_SUBSCR ()
+.. opcode:: STORE_SUBSCR
 
    Implements ``TOS1[TOS] = TOS2``.
 
 
-.. opcode:: DELETE_SUBSCR ()
+.. opcode:: DELETE_SUBSCR
 
    Implements ``del TOS1[TOS]``.
 
-Miscellaneous opcodes.
 
+**Miscellaneous opcodes**
 
-.. opcode:: PRINT_EXPR ()
+.. opcode:: PRINT_EXPR
 
    Implements the expression statement for the interactive mode.  TOS is removed
    from the stack and printed.  In non-interactive mode, an expression statement is
    terminated with ``POP_STACK``.
 
 
-.. opcode:: BREAK_LOOP ()
+.. opcode:: BREAK_LOOP
 
    Terminates a loop due to a :keyword:`break` statement.
 
@@ -387,36 +396,35 @@
    Calls ``dict.setitem(TOS1[-i], TOS, TOS1)``.  Used to implement dict
    comprehensions.
 
-
 For all of the SET_ADD, LIST_APPEND and MAP_ADD instructions, while the
 added value or key/value pair is popped off, the container object remains on
 the stack so that it is available for further iterations of the loop.
 
 
-.. opcode:: RETURN_VALUE ()
+.. opcode:: RETURN_VALUE
 
    Returns with TOS to the caller of the function.
 
 
-.. opcode:: YIELD_VALUE ()
+.. opcode:: YIELD_VALUE
 
    Pops ``TOS`` and yields it from a :term:`generator`.
 
 
-.. opcode:: IMPORT_STAR ()
+.. opcode:: IMPORT_STAR
 
    Loads all symbols not starting with ``'_'`` directly from the module TOS to the
    local namespace. The module is popped after loading all names. This opcode
    implements ``from module import *``.
 
 
-.. opcode:: POP_BLOCK ()
+.. opcode:: POP_BLOCK
 
    Removes one block from the block stack.  Per frame, there is a  stack of blocks,
    denoting nested loops, try statements, and such.
 
 
-.. opcode:: POP_EXCEPT ()
+.. opcode:: POP_EXCEPT
 
    Removes one block from the block stack. The popped block must be an exception
    handler block, as implicitly created when entering an except handler.
@@ -424,14 +432,14 @@
    last three popped values are used to restore the exception state.
 
 
-.. opcode:: END_FINALLY ()
+.. opcode:: END_FINALLY
 
    Terminates a :keyword:`finally` clause.  The interpreter recalls whether the
    exception has to be re-raised, or whether the function returns, and continues
    with the outer-next block.
 
 
-.. opcode:: LOAD_BUILD_CLASS ()
+.. opcode:: LOAD_BUILD_CLASS
 
    Pushes :func:`builtins.__build_class__` onto the stack.  It is later called
    by ``CALL_FUNCTION`` to construct a class.
@@ -449,7 +457,7 @@
    :opcode:`UNPACK_SEQUENCE`).
 
 
-.. opcode:: WITH_CLEANUP ()
+.. opcode:: WITH_CLEANUP
 
    Cleans up the stack when a :keyword:`with` statement block exits.  TOS is
    the context manager's :meth:`__exit__` bound method. Below TOS are 1--3
@@ -659,7 +667,7 @@
    Pushes a try block from a try-except clause onto the block stack. *delta* points
    to the finally block.
 
-.. opcode:: STORE_MAP ()
+.. opcode:: STORE_MAP
 
    Store a key and value pair in a dictionary.  Pops the key and value while leaving
    the dictionary on the stack.
@@ -775,7 +783,7 @@
    variable-arguments tuple, followed by explicit keyword and positional arguments.
 
 
-.. opcode:: HAVE_ARGUMENT ()
+.. opcode:: HAVE_ARGUMENT
 
    This is not really an opcode.  It identifies the dividing line between opcodes
    which don't take arguments ``< HAVE_ARGUMENT`` and those which do ``>=

Modified: python/branches/py3k/Doc/tools/sphinxext/pyspecific.py
==============================================================================
--- python/branches/py3k/Doc/tools/sphinxext/pyspecific.py	(original)
+++ python/branches/py3k/Doc/tools/sphinxext/pyspecific.py	Sat Jul  3 12:41:33 2010
@@ -5,7 +5,7 @@
 
     Sphinx extension with Python doc-specific markup.
 
-    :copyright: 2008, 2009 by Georg Brandl.
+    :copyright: 2008, 2009, 2010 by Georg Brandl.
     :license: Python license.
 """
 
@@ -149,7 +149,7 @@
 import re
 from sphinx import addnodes
 
-opcode_sig_re = re.compile(r'(\w+(?:\+\d)?)\s*\((.*)\)')
+opcode_sig_re = re.compile(r'(\w+(?:\+\d)?)(?:\s*\((.*)\))?')
 
 def parse_opcode_signature(env, sig, signode):
     """Transform an opcode signature into RST nodes."""
@@ -158,9 +158,10 @@
         raise ValueError
     opname, arglist = m.groups()
     signode += addnodes.desc_name(opname, opname)
-    paramlist = addnodes.desc_parameterlist()
-    signode += paramlist
-    paramlist += addnodes.desc_parameter(arglist, arglist)
+    if arglist is not None:
+        paramlist = addnodes.desc_parameterlist()
+        signode += paramlist
+        paramlist += addnodes.desc_parameter(arglist, arglist)
     return opname.strip()
 
 

From python-checkins at python.org  Sat Jul  3 14:26:03 2010
From: python-checkins at python.org (jesse.noller)
Date: Sat,  3 Jul 2010 14:26:03 +0200 (CEST)
Subject: [Python-checkins] r82489 - in python/trunk/Lib/multiprocessing:
	connection.py forking.py heap.py reduction.py
Message-ID: <20100703122603.18CE0F1AE@mail.python.org>

Author: jesse.noller
Date: Sat Jul  3 14:26:02 2010
New Revision: 82489

Log:
Submit patch for bug 9144; broken imports due to r81380

Modified:
   python/trunk/Lib/multiprocessing/connection.py
   python/trunk/Lib/multiprocessing/forking.py
   python/trunk/Lib/multiprocessing/heap.py
   python/trunk/Lib/multiprocessing/reduction.py

Modified: python/trunk/Lib/multiprocessing/connection.py
==============================================================================
--- python/trunk/Lib/multiprocessing/connection.py	(original)
+++ python/trunk/Lib/multiprocessing/connection.py	Sat Jul  3 14:26:02 2010
@@ -173,7 +173,7 @@
 
 else:
 
-    from ._multiprocessing import win32
+    from _multiprocessing import win32
 
     def Pipe(duplex=True):
         '''

Modified: python/trunk/Lib/multiprocessing/forking.py
==============================================================================
--- python/trunk/Lib/multiprocessing/forking.py	(original)
+++ python/trunk/Lib/multiprocessing/forking.py	Sat Jul  3 14:26:02 2010
@@ -155,7 +155,7 @@
     import _subprocess
     import time
 
-    from ._multiprocessing import win32, Connection, PipeConnection
+    from _multiprocessing import win32, Connection, PipeConnection
     from .util import Finalize
 
     #try:

Modified: python/trunk/Lib/multiprocessing/heap.py
==============================================================================
--- python/trunk/Lib/multiprocessing/heap.py	(original)
+++ python/trunk/Lib/multiprocessing/heap.py	Sat Jul  3 14:26:02 2010
@@ -26,7 +26,7 @@
 
 if sys.platform == 'win32':
 
-    from ._multiprocessing import win32
+    from _multiprocessing import win32
 
     class Arena(object):
 

Modified: python/trunk/Lib/multiprocessing/reduction.py
==============================================================================
--- python/trunk/Lib/multiprocessing/reduction.py	(original)
+++ python/trunk/Lib/multiprocessing/reduction.py	Sat Jul  3 14:26:02 2010
@@ -34,7 +34,7 @@
 
 if sys.platform == 'win32':
     import _subprocess
-    from ._multiprocessing import win32
+    from _multiprocessing import win32
 
     def send_handle(conn, handle, destination_pid):
         process_handle = win32.OpenProcess(

From python-checkins at python.org  Sat Jul  3 15:34:32 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sat,  3 Jul 2010 15:34:32 +0200 (CEST)
Subject: [Python-checkins] r82490 - in python/trunk: Doc/license.rst LICENSE
Message-ID: <20100703133432.E3A1DC37F@mail.python.org>

Author: benjamin.peterson
Date: Sat Jul  3 15:34:32 2010
New Revision: 82490

Log:
add 2.7 to the license list

Modified:
   python/trunk/Doc/license.rst
   python/trunk/LICENSE

Modified: python/trunk/Doc/license.rst
==============================================================================
--- python/trunk/Doc/license.rst	(original)
+++ python/trunk/Doc/license.rst	Sat Jul  3 15:34:32 2010
@@ -102,6 +102,8 @@
 +----------------+--------------+-----------+------------+-----------------+
 | 2.6.4          | 2.6.3        | 2009      | PSF        | yes             |
 +----------------+--------------+-----------+------------+-----------------+
+| 2.7            | 2.6          | 2010      | PSF        | yes             |
++----------------+--------------+-----------+------------+-----------------+
 
 .. note::
 

Modified: python/trunk/LICENSE
==============================================================================
--- python/trunk/LICENSE	(original)
+++ python/trunk/LICENSE	Sat Jul  3 15:34:32 2010
@@ -63,6 +63,7 @@
     2.6.3           2.6.2       2009        PSF         yes
     2.6.4           2.6.3       2009        PSF         yes
     2.6.5           2.6.4       2010        PSF         yes
+    2.7             2.6         2010        PSF         yes
 
 Footnotes:
 

From python-checkins at python.org  Sat Jul  3 15:35:13 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sat,  3 Jul 2010 15:35:13 +0200 (CEST)
Subject: [Python-checkins] r82491 - python/trunk/Doc/license.rst
Message-ID: <20100703133513.D1E70C37F@mail.python.org>

Author: benjamin.peterson
Date: Sat Jul  3 15:35:13 2010
New Revision: 82491

Log:
fix date

Modified:
   python/trunk/Doc/license.rst

Modified: python/trunk/Doc/license.rst
==============================================================================
--- python/trunk/Doc/license.rst	(original)
+++ python/trunk/Doc/license.rst	Sat Jul  3 15:35:13 2010
@@ -100,7 +100,7 @@
 +----------------+--------------+-----------+------------+-----------------+
 | 2.6.3          | 2.6.2        | 2009      | PSF        | yes             |
 +----------------+--------------+-----------+------------+-----------------+
-| 2.6.4          | 2.6.3        | 2009      | PSF        | yes             |
+| 2.6.4          | 2.6.3        | 2010      | PSF        | yes             |
 +----------------+--------------+-----------+------------+-----------------+
 | 2.7            | 2.6          | 2010      | PSF        | yes             |
 +----------------+--------------+-----------+------------+-----------------+

From python-checkins at python.org  Sat Jul  3 15:36:20 2010
From: python-checkins at python.org (victor.stinner)
Date: Sat,  3 Jul 2010 15:36:20 +0200 (CEST)
Subject: [Python-checkins] r82492 - in python/trunk:
	Lib/test/test_audioop.py Misc/NEWS Modules/audioop.c
Message-ID: <20100703133620.302BCC37F@mail.python.org>

Author: victor.stinner
Date: Sat Jul  3 15:36:19 2010
New Revision: 82492

Log:
Issue #7673: Fix security vulnerability (CVE-2010-2089) in the audioop module,
ensure that the input string length is a multiple of the frame size


Modified:
   python/trunk/Lib/test/test_audioop.py
   python/trunk/Misc/NEWS
   python/trunk/Modules/audioop.c

Modified: python/trunk/Lib/test/test_audioop.py
==============================================================================
--- python/trunk/Lib/test/test_audioop.py	(original)
+++ python/trunk/Lib/test/test_audioop.py	Sat Jul  3 15:36:19 2010
@@ -20,6 +20,12 @@
 
 data = [gendata1(), gendata2(), gendata4()]
 
+INVALID_DATA = [
+    ('abc', 0),
+    ('abc', 2),
+    ('abc', 4),
+]
+
 
 class TestAudioop(unittest.TestCase):
 
@@ -166,6 +172,33 @@
         self.assertRaises(audioop.error,
             audioop.findmax, ''.join( chr(x) for x in xrange(256)), -2392392)
 
+    def test_issue7673(self):
+        state = None
+        for data, size in INVALID_DATA:
+            size2 = size
+            self.assertRaises(audioop.error, audioop.getsample, data, size, 0)
+            self.assertRaises(audioop.error, audioop.max, data, size)
+            self.assertRaises(audioop.error, audioop.minmax, data, size)
+            self.assertRaises(audioop.error, audioop.avg, data, size)
+            self.assertRaises(audioop.error, audioop.rms, data, size)
+            self.assertRaises(audioop.error, audioop.avgpp, data, size)
+            self.assertRaises(audioop.error, audioop.maxpp, data, size)
+            self.assertRaises(audioop.error, audioop.cross, data, size)
+            self.assertRaises(audioop.error, audioop.mul, data, size, 1.0)
+            self.assertRaises(audioop.error, audioop.tomono, data, size, 0.5, 0.5)
+            self.assertRaises(audioop.error, audioop.tostereo, data, size, 0.5, 0.5)
+            self.assertRaises(audioop.error, audioop.add, data, data, size)
+            self.assertRaises(audioop.error, audioop.bias, data, size, 0)
+            self.assertRaises(audioop.error, audioop.reverse, data, size)
+            self.assertRaises(audioop.error, audioop.lin2lin, data, size, size2)
+            self.assertRaises(audioop.error, audioop.ratecv, data, size, 1, 1, 1, state)
+            self.assertRaises(audioop.error, audioop.lin2ulaw, data, size)
+            self.assertRaises(audioop.error, audioop.ulaw2lin, data, size)
+            self.assertRaises(audioop.error, audioop.lin2alaw, data, size)
+            self.assertRaises(audioop.error, audioop.alaw2lin, data, size)
+            self.assertRaises(audioop.error, audioop.lin2adpcm, data, size, state)
+            self.assertRaises(audioop.error, audioop.adpcm2lin, data, size, state)
+
 def test_main():
     run_unittest(TestAudioop)
 

Modified: python/trunk/Misc/NEWS
==============================================================================
--- python/trunk/Misc/NEWS	(original)
+++ python/trunk/Misc/NEWS	Sat Jul  3 15:36:19 2010
@@ -17,6 +17,9 @@
 Library
 -------
 
+- Issue #7673: Fix security vulnerability (CVE-2010-2089) in the audioop
+  module, ensure that the input string length is a multiple of the frame size
+
 - Issue #9075: In the ssl module, remove the setting of a ``debug`` flag
   on an OpenSSL structure.
 
@@ -51,7 +54,7 @@
 Library
 -------
 
-- Issue #6589: cleanup asyncore.socket_map in case smtpd.SMTPServer constructor 
+- Issue #6589: cleanup asyncore.socket_map in case smtpd.SMTPServer constructor
   raises an exception.
 
 - Issue #8959: fix regression caused by using unmodified libffi

Modified: python/trunk/Modules/audioop.c
==============================================================================
--- python/trunk/Modules/audioop.c	(original)
+++ python/trunk/Modules/audioop.c	Sat Jul  3 15:36:19 2010
@@ -295,6 +295,29 @@
 
 static PyObject *AudioopError;
 
+static int
+audioop_check_size(int size)
+{
+    if (size != 1 && size != 2 && size != 4) {
+        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
+        return 0;
+    }
+    else
+        return 1;
+}
+
+static int
+audioop_check_parameters(int len, int size)
+{
+    if (!audioop_check_size(size))
+        return 0;
+    if (len % size != 0) {
+        PyErr_SetString(AudioopError, "not a whole number of frames");
+        return 0;
+    }
+    return 1;
+}
+
 static PyObject *
 audioop_getsample(PyObject *self, PyObject *args)
 {
@@ -304,10 +327,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#ii:getsample", &cp, &len, &size, &i) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     if ( i < 0 || i >= len/size ) {
         PyErr_SetString(AudioopError, "Index out of range");
         return 0;
@@ -328,10 +349,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:max", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     for ( i=0; i<len; i+= size) {
         if ( size == 1 )      val = (int)*CHARP(cp, i);
         else if ( size == 2 ) val = (int)*SHORTP(cp, i);
@@ -352,10 +371,8 @@
 
     if (!PyArg_ParseTuple(args, "s#i:minmax", &cp, &len, &size))
         return NULL;
-    if (size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
+    if (!audioop_check_parameters(len, size))
         return NULL;
-    }
     for (i = 0; i < len; i += size) {
         if (size == 1) val = (int) *CHARP(cp, i);
         else if (size == 2) val = (int) *SHORTP(cp, i);
@@ -376,10 +393,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:avg", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     for ( i=0; i<len; i+= size) {
         if ( size == 1 )      val = (int)*CHARP(cp, i);
         else if ( size == 2 ) val = (int)*SHORTP(cp, i);
@@ -403,10 +418,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:rms", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     for ( i=0; i<len; i+= size) {
         if ( size == 1 )      val = (int)*CHARP(cp, i);
         else if ( size == 2 ) val = (int)*SHORTP(cp, i);
@@ -612,10 +625,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:avgpp", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     /* Compute first delta value ahead. Also automatically makes us
     ** skip the first extreme value
     */
@@ -669,10 +680,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:maxpp", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     /* Compute first delta value ahead. Also automatically makes us
     ** skip the first extreme value
     */
@@ -720,10 +729,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:cross", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     ncross = -1;
     prevval = 17; /* Anything <> 0,1 */
     for ( i=0; i<len; i+= size) {
@@ -748,6 +755,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#id:mul", &cp, &len, &size, &factor ) )
         return 0;
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     if ( size == 1 ) maxval = (double) 0x7f;
     else if ( size == 2 ) maxval = (double) 0x7fff;
@@ -790,6 +799,12 @@
     if ( !PyArg_ParseTuple(args, "s#idd:tomono",
                            &cp, &len, &size, &fac1, &fac2 ) )
         return 0;
+    if (!audioop_check_parameters(len, size))
+        return NULL;
+    if (((len / size) & 1) != 0) {
+        PyErr_SetString(AudioopError, "not a whole number of frames");
+        return NULL;
+    }
 
     if ( size == 1 ) maxval = (double) 0x7f;
     else if ( size == 2 ) maxval = (double) 0x7fff;
@@ -835,6 +850,8 @@
     if ( !PyArg_ParseTuple(args, "s#idd:tostereo",
                            &cp, &len, &size, &fac1, &fac2 ) )
         return 0;
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     if ( size == 1 ) maxval = (double) 0x7f;
     else if ( size == 2 ) maxval = (double) 0x7fff;
@@ -893,7 +910,8 @@
     if ( !PyArg_ParseTuple(args, "s#s#i:add",
                       &cp1, &len1, &cp2, &len2, &size ) )
         return 0;
-
+    if (!audioop_check_parameters(len1, size))
+        return NULL;
     if ( len1 != len2 ) {
         PyErr_SetString(AudioopError, "Lengths should be the same");
         return 0;
@@ -948,10 +966,8 @@
                       &cp, &len, &size , &bias) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     rv = PyString_FromStringAndSize(NULL, len);
     if ( rv == 0 )
@@ -984,10 +1000,8 @@
                       &cp, &len, &size) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     rv = PyString_FromStringAndSize(NULL, len);
     if ( rv == 0 )
@@ -1021,11 +1035,10 @@
                       &cp, &len, &size, &size2) )
         return 0;
 
-    if ( (size != 1 && size != 2 && size != 4) ||
-         (size2 != 1 && size2 != 2 && size2 != 4)) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
+    if (!audioop_check_size(size2))
+        return NULL;
 
     if (len/size > INT_MAX/size2) {
         PyErr_SetString(PyExc_MemoryError,
@@ -1075,10 +1088,8 @@
                           &nchannels, &inrate, &outrate, &state,
                           &weightA, &weightB))
         return NULL;
-    if (size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
+    if (!audioop_check_size(size))
         return NULL;
-    }
     if (nchannels < 1) {
         PyErr_SetString(AudioopError, "# of channels should be >= 1");
         return NULL;
@@ -1255,10 +1266,8 @@
                            &cp, &len, &size) )
         return 0 ;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     rv = PyString_FromStringAndSize(NULL, len/size);
     if ( rv == 0 )
@@ -1289,10 +1298,8 @@
                            &cp, &len, &size) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     if (len > INT_MAX/size) {
         PyErr_SetString(PyExc_MemoryError,
@@ -1328,10 +1335,8 @@
                            &cp, &len, &size) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     rv = PyString_FromStringAndSize(NULL, len/size);
     if ( rv == 0 )
@@ -1362,10 +1367,8 @@
                            &cp, &len, &size) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     if (len > INT_MAX/size) {
         PyErr_SetString(PyExc_MemoryError,
@@ -1402,11 +1405,8 @@
                            &cp, &len, &size, &state) )
         return 0;
 
-
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     str = PyString_FromStringAndSize(NULL, len/(size*2));
     if ( str == 0 )
@@ -1509,10 +1509,8 @@
                            &cp, &len, &size, &state) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     /* Decode state, should have (value, step) */
     if ( state == Py_None ) {

From python-checkins at python.org  Sat Jul  3 15:37:39 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sat,  3 Jul 2010 15:37:39 +0200 (CEST)
Subject: [Python-checkins] r82493 - python/trunk/Doc/whatsnew/2.7.rst
Message-ID: <20100703133739.42A72C37F@mail.python.org>

Author: benjamin.peterson
Date: Sat Jul  3 15:37:39 2010
New Revision: 82493

Log:
add release date

Modified:
   python/trunk/Doc/whatsnew/2.7.rst

Modified: python/trunk/Doc/whatsnew/2.7.rst
==============================================================================
--- python/trunk/Doc/whatsnew/2.7.rst	(original)
+++ python/trunk/Doc/whatsnew/2.7.rst	Sat Jul  3 15:37:39 2010
@@ -51,9 +51,8 @@
    This saves the maintainer some effort going through the SVN logs
    when researching a change.
 
-This article explains the new features in Python 2.7.  The final
-release of 2.7 is currently scheduled for July 2010; the detailed
-schedule is described in :pep:`373`.
+This article explains the new features in Python 2.7.  Python 2.7 was released
+on July 7, 2010.
 
 Numeric handling has been improved in many ways, for both
 floating-point numbers and for the :class:`~decimal.Decimal` class.

From python-checkins at python.org  Sat Jul  3 15:39:22 2010
From: python-checkins at python.org (victor.stinner)
Date: Sat,  3 Jul 2010 15:39:22 +0200 (CEST)
Subject: [Python-checkins] r82494 - in python/branches/release26-maint:
	Lib/test/test_audioop.py Misc/NEWS Modules/audioop.c
Message-ID: <20100703133922.5F4D4F1AE@mail.python.org>

Author: victor.stinner
Date: Sat Jul  3 15:39:22 2010
New Revision: 82494

Log:
Merged revisions 82492 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/trunk

........
  r82492 | victor.stinner | 2010-07-03 15:36:19 +0200 (sam., 03 juil. 2010) | 3 lines
  
  Issue #7673: Fix security vulnerability (CVE-2010-2089) in the audioop module,
  ensure that the input string length is a multiple of the frame size
........


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Lib/test/test_audioop.py
   python/branches/release26-maint/Misc/NEWS
   python/branches/release26-maint/Modules/audioop.c

Modified: python/branches/release26-maint/Lib/test/test_audioop.py
==============================================================================
--- python/branches/release26-maint/Lib/test/test_audioop.py	(original)
+++ python/branches/release26-maint/Lib/test/test_audioop.py	Sat Jul  3 15:39:22 2010
@@ -20,6 +20,12 @@
 
 data = [gendata1(), gendata2(), gendata4()]
 
+INVALID_DATA = [
+    ('abc', 0),
+    ('abc', 2),
+    ('abc', 4),
+]
+
 
 class TestAudioop(unittest.TestCase):
 
@@ -166,6 +172,33 @@
         self.assertRaises(audioop.error,
             audioop.findmax, ''.join( chr(x) for x in xrange(256)), -2392392)
 
+    def test_issue7673(self):
+        state = None
+        for data, size in INVALID_DATA:
+            size2 = size
+            self.assertRaises(audioop.error, audioop.getsample, data, size, 0)
+            self.assertRaises(audioop.error, audioop.max, data, size)
+            self.assertRaises(audioop.error, audioop.minmax, data, size)
+            self.assertRaises(audioop.error, audioop.avg, data, size)
+            self.assertRaises(audioop.error, audioop.rms, data, size)
+            self.assertRaises(audioop.error, audioop.avgpp, data, size)
+            self.assertRaises(audioop.error, audioop.maxpp, data, size)
+            self.assertRaises(audioop.error, audioop.cross, data, size)
+            self.assertRaises(audioop.error, audioop.mul, data, size, 1.0)
+            self.assertRaises(audioop.error, audioop.tomono, data, size, 0.5, 0.5)
+            self.assertRaises(audioop.error, audioop.tostereo, data, size, 0.5, 0.5)
+            self.assertRaises(audioop.error, audioop.add, data, data, size)
+            self.assertRaises(audioop.error, audioop.bias, data, size, 0)
+            self.assertRaises(audioop.error, audioop.reverse, data, size)
+            self.assertRaises(audioop.error, audioop.lin2lin, data, size, size2)
+            self.assertRaises(audioop.error, audioop.ratecv, data, size, 1, 1, 1, state)
+            self.assertRaises(audioop.error, audioop.lin2ulaw, data, size)
+            self.assertRaises(audioop.error, audioop.ulaw2lin, data, size)
+            self.assertRaises(audioop.error, audioop.lin2alaw, data, size)
+            self.assertRaises(audioop.error, audioop.alaw2lin, data, size)
+            self.assertRaises(audioop.error, audioop.lin2adpcm, data, size, state)
+            self.assertRaises(audioop.error, audioop.adpcm2lin, data, size, state)
+
 def test_main():
     run_unittest(TestAudioop)
 

Modified: python/branches/release26-maint/Misc/NEWS
==============================================================================
--- python/branches/release26-maint/Misc/NEWS	(original)
+++ python/branches/release26-maint/Misc/NEWS	Sat Jul  3 15:39:22 2010
@@ -81,7 +81,10 @@
 Library
 -------
 
-- Issue #6589: cleanup asyncore.socket_map in case smtpd.SMTPServer constructor 
+- Issue #7673: Fix security vulnerability (CVE-2010-2089) in the audioop
+  module, ensure that the input string length is a multiple of the frame size
+
+- Issue #6589: cleanup asyncore.socket_map in case smtpd.SMTPServer constructor
   raises an exception.
 
 - Issue #9125: Add recognition of 'except ... as ...' syntax to parser module.

Modified: python/branches/release26-maint/Modules/audioop.c
==============================================================================
--- python/branches/release26-maint/Modules/audioop.c	(original)
+++ python/branches/release26-maint/Modules/audioop.c	Sat Jul  3 15:39:22 2010
@@ -295,6 +295,29 @@
 
 static PyObject *AudioopError;
 
+static int
+audioop_check_size(int size)
+{
+    if (size != 1 && size != 2 && size != 4) {
+        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
+        return 0;
+    }
+    else
+        return 1;
+}
+
+static int
+audioop_check_parameters(int len, int size)
+{
+    if (!audioop_check_size(size))
+        return 0;
+    if (len % size != 0) {
+        PyErr_SetString(AudioopError, "not a whole number of frames");
+        return 0;
+    }
+    return 1;
+}
+
 static PyObject *
 audioop_getsample(PyObject *self, PyObject *args)
 {
@@ -304,10 +327,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#ii:getsample", &cp, &len, &size, &i) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     if ( i < 0 || i >= len/size ) {
         PyErr_SetString(AudioopError, "Index out of range");
         return 0;
@@ -328,10 +349,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:max", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     for ( i=0; i<len; i+= size) {
         if ( size == 1 )      val = (int)*CHARP(cp, i);
         else if ( size == 2 ) val = (int)*SHORTP(cp, i);
@@ -352,10 +371,8 @@
 
     if (!PyArg_ParseTuple(args, "s#i:minmax", &cp, &len, &size))
         return NULL;
-    if (size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
+    if (!audioop_check_parameters(len, size))
         return NULL;
-    }
     for (i = 0; i < len; i += size) {
         if (size == 1) val = (int) *CHARP(cp, i);
         else if (size == 2) val = (int) *SHORTP(cp, i);
@@ -376,10 +393,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:avg", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     for ( i=0; i<len; i+= size) {
         if ( size == 1 )      val = (int)*CHARP(cp, i);
         else if ( size == 2 ) val = (int)*SHORTP(cp, i);
@@ -403,10 +418,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:rms", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     for ( i=0; i<len; i+= size) {
         if ( size == 1 )      val = (int)*CHARP(cp, i);
         else if ( size == 2 ) val = (int)*SHORTP(cp, i);
@@ -614,10 +627,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:avgpp", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     /* Compute first delta value ahead. Also automatically makes us
     ** skip the first extreme value
     */
@@ -671,10 +682,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:maxpp", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     /* Compute first delta value ahead. Also automatically makes us
     ** skip the first extreme value
     */
@@ -722,10 +731,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:cross", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     ncross = -1;
     prevval = 17; /* Anything <> 0,1 */
     for ( i=0; i<len; i+= size) {
@@ -750,6 +757,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#id:mul", &cp, &len, &size, &factor ) )
         return 0;
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     if ( size == 1 ) maxval = (double) 0x7f;
     else if ( size == 2 ) maxval = (double) 0x7fff;
@@ -792,6 +801,12 @@
     if ( !PyArg_ParseTuple(args, "s#idd:tomono",
                            &cp, &len, &size, &fac1, &fac2 ) )
         return 0;
+    if (!audioop_check_parameters(len, size))
+        return NULL;
+    if (((len / size) & 1) != 0) {
+        PyErr_SetString(AudioopError, "not a whole number of frames");
+        return NULL;
+    }
 
     if ( size == 1 ) maxval = (double) 0x7f;
     else if ( size == 2 ) maxval = (double) 0x7fff;
@@ -837,6 +852,8 @@
     if ( !PyArg_ParseTuple(args, "s#idd:tostereo",
                            &cp, &len, &size, &fac1, &fac2 ) )
         return 0;
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     if ( size == 1 ) maxval = (double) 0x7f;
     else if ( size == 2 ) maxval = (double) 0x7fff;
@@ -895,7 +912,8 @@
     if ( !PyArg_ParseTuple(args, "s#s#i:add",
                       &cp1, &len1, &cp2, &len2, &size ) )
         return 0;
-
+    if (!audioop_check_parameters(len1, size))
+        return NULL;
     if ( len1 != len2 ) {
         PyErr_SetString(AudioopError, "Lengths should be the same");
         return 0;
@@ -950,10 +968,8 @@
                       &cp, &len, &size , &bias) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     rv = PyString_FromStringAndSize(NULL, len);
     if ( rv == 0 )
@@ -986,10 +1002,8 @@
                       &cp, &len, &size) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     rv = PyString_FromStringAndSize(NULL, len);
     if ( rv == 0 )
@@ -1023,11 +1037,10 @@
                       &cp, &len, &size, &size2) )
         return 0;
 
-    if ( (size != 1 && size != 2 && size != 4) ||
-         (size2 != 1 && size2 != 2 && size2 != 4)) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
+    if (!audioop_check_size(size2))
+        return NULL;
 
     if (len/size > INT_MAX/size2) {
         PyErr_SetString(PyExc_MemoryError,
@@ -1077,10 +1090,8 @@
                           &nchannels, &inrate, &outrate, &state,
                           &weightA, &weightB))
         return NULL;
-    if (size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
+    if (!audioop_check_size(size))
         return NULL;
-    }
     if (nchannels < 1) {
         PyErr_SetString(AudioopError, "# of channels should be >= 1");
         return NULL;
@@ -1257,10 +1268,8 @@
                            &cp, &len, &size) )
         return 0 ;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     rv = PyString_FromStringAndSize(NULL, len/size);
     if ( rv == 0 )
@@ -1291,10 +1300,8 @@
                            &cp, &len, &size) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     if (len > INT_MAX/size) {
         PyErr_SetString(PyExc_MemoryError,
@@ -1330,10 +1337,8 @@
                            &cp, &len, &size) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     rv = PyString_FromStringAndSize(NULL, len/size);
     if ( rv == 0 )
@@ -1364,10 +1369,8 @@
                            &cp, &len, &size) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     if (len > INT_MAX/size) {
         PyErr_SetString(PyExc_MemoryError,
@@ -1404,11 +1407,8 @@
                            &cp, &len, &size, &state) )
         return 0;
 
-
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     str = PyString_FromStringAndSize(NULL, len/(size*2));
     if ( str == 0 )
@@ -1512,10 +1512,8 @@
                            &cp, &len, &size, &state) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     /* Decode state, should have (value, step) */
     if ( state == Py_None ) {

From python-checkins at python.org  Sat Jul  3 15:44:22 2010
From: python-checkins at python.org (victor.stinner)
Date: Sat,  3 Jul 2010 15:44:22 +0200 (CEST)
Subject: [Python-checkins] r82495 - in python/branches/py3k:
	Lib/test/test_audioop.py Misc/NEWS Modules/audioop.c
Message-ID: <20100703134422.437CFED67@mail.python.org>

Author: victor.stinner
Date: Sat Jul  3 15:44:22 2010
New Revision: 82495

Log:
Merged revisions 82492 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/trunk

........
  r82492 | victor.stinner | 2010-07-03 15:36:19 +0200 (sam., 03 juil. 2010) | 3 lines
  
  Issue #7673: Fix security vulnerability (CVE-2010-2089) in the audioop module,
  ensure that the input string length is a multiple of the frame size
........


Modified:
   python/branches/py3k/   (props changed)
   python/branches/py3k/Lib/test/test_audioop.py
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/Modules/audioop.c

Modified: python/branches/py3k/Lib/test/test_audioop.py
==============================================================================
--- python/branches/py3k/Lib/test/test_audioop.py	(original)
+++ python/branches/py3k/Lib/test/test_audioop.py	Sat Jul  3 15:44:22 2010
@@ -20,6 +20,12 @@
 
 data = [gendata1(), gendata2(), gendata4()]
 
+INVALID_DATA = [
+    ('abc', 0),
+    ('abc', 2),
+    ('abc', 4),
+]
+
 
 class TestAudioop(unittest.TestCase):
 
@@ -168,6 +174,33 @@
         self.assertRaises(audioop.error,
             audioop.findmax, ''.join(chr(x) for x in range(256)), -2392392)
 
+    def test_issue7673(self):
+        state = None
+        for data, size in INVALID_DATA:
+            size2 = size
+            self.assertRaises(audioop.error, audioop.getsample, data, size, 0)
+            self.assertRaises(audioop.error, audioop.max, data, size)
+            self.assertRaises(audioop.error, audioop.minmax, data, size)
+            self.assertRaises(audioop.error, audioop.avg, data, size)
+            self.assertRaises(audioop.error, audioop.rms, data, size)
+            self.assertRaises(audioop.error, audioop.avgpp, data, size)
+            self.assertRaises(audioop.error, audioop.maxpp, data, size)
+            self.assertRaises(audioop.error, audioop.cross, data, size)
+            self.assertRaises(audioop.error, audioop.mul, data, size, 1.0)
+            self.assertRaises(audioop.error, audioop.tomono, data, size, 0.5, 0.5)
+            self.assertRaises(audioop.error, audioop.tostereo, data, size, 0.5, 0.5)
+            self.assertRaises(audioop.error, audioop.add, data, data, size)
+            self.assertRaises(audioop.error, audioop.bias, data, size, 0)
+            self.assertRaises(audioop.error, audioop.reverse, data, size)
+            self.assertRaises(audioop.error, audioop.lin2lin, data, size, size2)
+            self.assertRaises(audioop.error, audioop.ratecv, data, size, 1, 1, 1, state)
+            self.assertRaises(audioop.error, audioop.lin2ulaw, data, size)
+            self.assertRaises(audioop.error, audioop.ulaw2lin, data, size)
+            self.assertRaises(audioop.error, audioop.lin2alaw, data, size)
+            self.assertRaises(audioop.error, audioop.alaw2lin, data, size)
+            self.assertRaises(audioop.error, audioop.lin2adpcm, data, size, state)
+            self.assertRaises(audioop.error, audioop.adpcm2lin, data, size, state)
+
 def test_main():
     run_unittest(TestAudioop)
 

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sat Jul  3 15:44:22 2010
@@ -468,13 +468,16 @@
 Library
 -------
 
+- Issue #7673: Fix security vulnerability (CVE-2010-2089) in the audioop
+  module, ensure that the input string length is a multiple of the frame size
+
 - Issue #6507: Accept source strings in dis.dis(). Original patch by
   Daniel Urban.
 
 - Issue #7829: Clearly document that the dis module is exposing an
   implementation detail that is not stable between Python VMs or releases.
 
-- Issue #6589: cleanup asyncore.socket_map in case smtpd.SMTPServer constructor 
+- Issue #6589: cleanup asyncore.socket_map in case smtpd.SMTPServer constructor
   raises an exception.
 
 - Issue #9110: Addition of ContextDecorator to contextlib, for creating APIs

Modified: python/branches/py3k/Modules/audioop.c
==============================================================================
--- python/branches/py3k/Modules/audioop.c	(original)
+++ python/branches/py3k/Modules/audioop.c	Sat Jul  3 15:44:22 2010
@@ -297,6 +297,29 @@
 
 static PyObject *AudioopError;
 
+static int
+audioop_check_size(int size)
+{
+    if (size != 1 && size != 2 && size != 4) {
+        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
+        return 0;
+    }
+    else
+        return 1;
+}
+
+static int
+audioop_check_parameters(int len, int size)
+{
+    if (!audioop_check_size(size))
+        return 0;
+    if (len % size != 0) {
+        PyErr_SetString(AudioopError, "not a whole number of frames");
+        return 0;
+    }
+    return 1;
+}
+
 static PyObject *
 audioop_getsample(PyObject *self, PyObject *args)
 {
@@ -306,10 +329,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#in:getsample", &cp, &len, &size, &i) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     if ( i < 0 || i >= len/size ) {
         PyErr_SetString(AudioopError, "Index out of range");
         return 0;
@@ -330,10 +351,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:max", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     for ( i=0; i<len; i+= size) {
         if ( size == 1 )      val = (int)*CHARP(cp, i);
         else if ( size == 2 ) val = (int)*SHORTP(cp, i);
@@ -354,10 +373,8 @@
 
     if (!PyArg_ParseTuple(args, "s#i:minmax", &cp, &len, &size))
         return NULL;
-    if (size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
+    if (!audioop_check_parameters(len, size))
         return NULL;
-    }
     for (i = 0; i < len; i += size) {
         if (size == 1) val = (int) *CHARP(cp, i);
         else if (size == 2) val = (int) *SHORTP(cp, i);
@@ -378,10 +395,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:avg", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     for ( i=0; i<len; i+= size) {
         if ( size == 1 )      val = (int)*CHARP(cp, i);
         else if ( size == 2 ) val = (int)*SHORTP(cp, i);
@@ -405,10 +420,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:rms", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     for ( i=0; i<len; i+= size) {
         if ( size == 1 )      val = (int)*CHARP(cp, i);
         else if ( size == 2 ) val = (int)*SHORTP(cp, i);
@@ -616,10 +629,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:avgpp", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     /* Compute first delta value ahead. Also automatically makes us
     ** skip the first extreme value
     */
@@ -673,10 +684,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:maxpp", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     /* Compute first delta value ahead. Also automatically makes us
     ** skip the first extreme value
     */
@@ -725,10 +734,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:cross", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     ncross = -1;
     prevval = 17; /* Anything <> 0,1 */
     for ( i=0; i<len; i+= size) {
@@ -753,6 +760,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#id:mul", &cp, &len, &size, &factor ) )
         return 0;
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     if ( size == 1 ) maxval = (double) 0x7f;
     else if ( size == 2 ) maxval = (double) 0x7fff;
@@ -798,6 +807,12 @@
         return 0;
     cp = pcp.buf;
     len = pcp.len;
+    if (!audioop_check_parameters(len, size))
+        return NULL;
+    if (((len / size) & 1) != 0) {
+        PyErr_SetString(AudioopError, "not a whole number of frames");
+        return NULL;
+    }
 
     if ( size == 1 ) maxval = (double) 0x7f;
     else if ( size == 2 ) maxval = (double) 0x7fff;
@@ -845,6 +860,8 @@
     if ( !PyArg_ParseTuple(args, "s#idd:tostereo",
                            &cp, &len, &size, &fac1, &fac2 ) )
         return 0;
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     if ( size == 1 ) maxval = (double) 0x7f;
     else if ( size == 2 ) maxval = (double) 0x7fff;
@@ -903,7 +920,8 @@
     if ( !PyArg_ParseTuple(args, "s#s#i:add",
                       &cp1, &len1, &cp2, &len2, &size ) )
         return 0;
-
+    if (!audioop_check_parameters(len1, size))
+        return NULL;
     if ( len1 != len2 ) {
         PyErr_SetString(AudioopError, "Lengths should be the same");
         return 0;
@@ -958,10 +976,8 @@
                       &cp, &len, &size , &bias) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     rv = PyBytes_FromStringAndSize(NULL, len);
     if ( rv == 0 )
@@ -994,10 +1010,8 @@
                       &cp, &len, &size) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     rv = PyBytes_FromStringAndSize(NULL, len);
     if ( rv == 0 )
@@ -1031,11 +1045,10 @@
                       &cp, &len, &size, &size2) )
         return 0;
 
-    if ( (size != 1 && size != 2 && size != 4) ||
-         (size2 != 1 && size2 != 2 && size2 != 4)) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
+    if (!audioop_check_size(size2))
+        return NULL;
 
     if (len/size > PY_SSIZE_T_MAX/size2) {
         PyErr_SetString(PyExc_MemoryError,
@@ -1086,10 +1099,8 @@
                           &nchannels, &inrate, &outrate, &state,
                           &weightA, &weightB))
         return NULL;
-    if (size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
+    if (!audioop_check_size(size))
         return NULL;
-    }
     if (nchannels < 1) {
         PyErr_SetString(AudioopError, "# of channels should be >= 1");
         return NULL;
@@ -1265,10 +1276,8 @@
                            &cp, &len, &size) )
         return 0 ;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     rv = PyBytes_FromStringAndSize(NULL, len/size);
     if ( rv == 0 )
@@ -1299,10 +1308,8 @@
                            &cp, &len, &size) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     if (len > PY_SSIZE_T_MAX/size) {
         PyErr_SetString(PyExc_MemoryError,
@@ -1338,10 +1345,8 @@
                            &cp, &len, &size) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     rv = PyBytes_FromStringAndSize(NULL, len/size);
     if ( rv == 0 )
@@ -1372,10 +1377,8 @@
                            &cp, &len, &size) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     if (len > PY_SSIZE_T_MAX/size) {
         PyErr_SetString(PyExc_MemoryError,
@@ -1413,11 +1416,8 @@
                            &cp, &len, &size, &state) )
         return 0;
 
-
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     str = PyBytes_FromStringAndSize(NULL, len/(size*2));
     if ( str == 0 )
@@ -1522,10 +1522,8 @@
                            &cp, &len, &size, &state) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     /* Decode state, should have (value, step) */
     if ( state == Py_None ) {

From python-checkins at python.org  Sat Jul  3 15:46:02 2010
From: python-checkins at python.org (victor.stinner)
Date: Sat,  3 Jul 2010 15:46:02 +0200 (CEST)
Subject: [Python-checkins] r82496 - in python/branches/release31-maint:
	Lib/test/test_audioop.py Misc/NEWS Modules/audioop.c
Message-ID: <20100703134602.3247DED67@mail.python.org>

Author: victor.stinner
Date: Sat Jul  3 15:46:01 2010
New Revision: 82496

Log:
Merged revisions 82495 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

................
  r82495 | victor.stinner | 2010-07-03 15:44:22 +0200 (sam., 03 juil. 2010) | 10 lines
  
  Merged revisions 82492 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82492 | victor.stinner | 2010-07-03 15:36:19 +0200 (sam., 03 juil. 2010) | 3 lines
    
    Issue #7673: Fix security vulnerability (CVE-2010-2089) in the audioop module,
    ensure that the input string length is a multiple of the frame size
  ........
................


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/test/test_audioop.py
   python/branches/release31-maint/Misc/NEWS
   python/branches/release31-maint/Modules/audioop.c

Modified: python/branches/release31-maint/Lib/test/test_audioop.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_audioop.py	(original)
+++ python/branches/release31-maint/Lib/test/test_audioop.py	Sat Jul  3 15:46:01 2010
@@ -20,6 +20,12 @@
 
 data = [gendata1(), gendata2(), gendata4()]
 
+INVALID_DATA = [
+    ('abc', 0),
+    ('abc', 2),
+    ('abc', 4),
+]
+
 
 class TestAudioop(unittest.TestCase):
 
@@ -168,6 +174,33 @@
         self.assertRaises(audioop.error,
             audioop.findmax, ''.join(chr(x) for x in range(256)), -2392392)
 
+    def test_issue7673(self):
+        state = None
+        for data, size in INVALID_DATA:
+            size2 = size
+            self.assertRaises(audioop.error, audioop.getsample, data, size, 0)
+            self.assertRaises(audioop.error, audioop.max, data, size)
+            self.assertRaises(audioop.error, audioop.minmax, data, size)
+            self.assertRaises(audioop.error, audioop.avg, data, size)
+            self.assertRaises(audioop.error, audioop.rms, data, size)
+            self.assertRaises(audioop.error, audioop.avgpp, data, size)
+            self.assertRaises(audioop.error, audioop.maxpp, data, size)
+            self.assertRaises(audioop.error, audioop.cross, data, size)
+            self.assertRaises(audioop.error, audioop.mul, data, size, 1.0)
+            self.assertRaises(audioop.error, audioop.tomono, data, size, 0.5, 0.5)
+            self.assertRaises(audioop.error, audioop.tostereo, data, size, 0.5, 0.5)
+            self.assertRaises(audioop.error, audioop.add, data, data, size)
+            self.assertRaises(audioop.error, audioop.bias, data, size, 0)
+            self.assertRaises(audioop.error, audioop.reverse, data, size)
+            self.assertRaises(audioop.error, audioop.lin2lin, data, size, size2)
+            self.assertRaises(audioop.error, audioop.ratecv, data, size, 1, 1, 1, state)
+            self.assertRaises(audioop.error, audioop.lin2ulaw, data, size)
+            self.assertRaises(audioop.error, audioop.ulaw2lin, data, size)
+            self.assertRaises(audioop.error, audioop.lin2alaw, data, size)
+            self.assertRaises(audioop.error, audioop.alaw2lin, data, size)
+            self.assertRaises(audioop.error, audioop.lin2adpcm, data, size, state)
+            self.assertRaises(audioop.error, audioop.adpcm2lin, data, size, state)
+
 def test_main():
     run_unittest(TestAudioop)
 

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Sat Jul  3 15:46:01 2010
@@ -75,7 +75,10 @@
 Library
 -------
 
-- Issue #6589: cleanup asyncore.socket_map in case smtpd.SMTPServer constructor 
+- Issue #7673: Fix security vulnerability (CVE-2010-2089) in the audioop
+  module, ensure that the input string length is a multiple of the frame size
+
+- Issue #6589: cleanup asyncore.socket_map in case smtpd.SMTPServer constructor
   raises an exception.
 
 - Issue #9075: In the ssl module, remove the setting of a ``debug`` flag

Modified: python/branches/release31-maint/Modules/audioop.c
==============================================================================
--- python/branches/release31-maint/Modules/audioop.c	(original)
+++ python/branches/release31-maint/Modules/audioop.c	Sat Jul  3 15:46:01 2010
@@ -295,6 +295,29 @@
 
 static PyObject *AudioopError;
 
+static int
+audioop_check_size(int size)
+{
+    if (size != 1 && size != 2 && size != 4) {
+        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
+        return 0;
+    }
+    else
+        return 1;
+}
+
+static int
+audioop_check_parameters(int len, int size)
+{
+    if (!audioop_check_size(size))
+        return 0;
+    if (len % size != 0) {
+        PyErr_SetString(AudioopError, "not a whole number of frames");
+        return 0;
+    }
+    return 1;
+}
+
 static PyObject *
 audioop_getsample(PyObject *self, PyObject *args)
 {
@@ -304,10 +327,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#ii:getsample", &cp, &len, &size, &i) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     if ( i < 0 || i >= len/size ) {
         PyErr_SetString(AudioopError, "Index out of range");
         return 0;
@@ -328,10 +349,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:max", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     for ( i=0; i<len; i+= size) {
         if ( size == 1 )      val = (int)*CHARP(cp, i);
         else if ( size == 2 ) val = (int)*SHORTP(cp, i);
@@ -352,10 +371,8 @@
 
     if (!PyArg_ParseTuple(args, "s#i:minmax", &cp, &len, &size))
         return NULL;
-    if (size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
+    if (!audioop_check_parameters(len, size))
         return NULL;
-    }
     for (i = 0; i < len; i += size) {
         if (size == 1) val = (int) *CHARP(cp, i);
         else if (size == 2) val = (int) *SHORTP(cp, i);
@@ -376,10 +393,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:avg", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     for ( i=0; i<len; i+= size) {
         if ( size == 1 )      val = (int)*CHARP(cp, i);
         else if ( size == 2 ) val = (int)*SHORTP(cp, i);
@@ -403,10 +418,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:rms", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     for ( i=0; i<len; i+= size) {
         if ( size == 1 )      val = (int)*CHARP(cp, i);
         else if ( size == 2 ) val = (int)*SHORTP(cp, i);
@@ -614,10 +627,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:avgpp", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     /* Compute first delta value ahead. Also automatically makes us
     ** skip the first extreme value
     */
@@ -671,10 +682,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:maxpp", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     /* Compute first delta value ahead. Also automatically makes us
     ** skip the first extreme value
     */
@@ -722,10 +731,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:cross", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     ncross = -1;
     prevval = 17; /* Anything <> 0,1 */
     for ( i=0; i<len; i+= size) {
@@ -750,6 +757,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#id:mul", &cp, &len, &size, &factor ) )
         return 0;
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     if ( size == 1 ) maxval = (double) 0x7f;
     else if ( size == 2 ) maxval = (double) 0x7fff;
@@ -795,6 +804,12 @@
         return 0;
     cp = pcp.buf;
     len = pcp.len;
+    if (!audioop_check_parameters(len, size))
+        return NULL;
+    if (((len / size) & 1) != 0) {
+        PyErr_SetString(AudioopError, "not a whole number of frames");
+        return NULL;
+    }
 
     if ( size == 1 ) maxval = (double) 0x7f;
     else if ( size == 2 ) maxval = (double) 0x7fff;
@@ -842,6 +857,8 @@
     if ( !PyArg_ParseTuple(args, "s#idd:tostereo",
                            &cp, &len, &size, &fac1, &fac2 ) )
         return 0;
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     if ( size == 1 ) maxval = (double) 0x7f;
     else if ( size == 2 ) maxval = (double) 0x7fff;
@@ -900,7 +917,8 @@
     if ( !PyArg_ParseTuple(args, "s#s#i:add",
                       &cp1, &len1, &cp2, &len2, &size ) )
         return 0;
-
+    if (!audioop_check_parameters(len1, size))
+        return NULL;
     if ( len1 != len2 ) {
         PyErr_SetString(AudioopError, "Lengths should be the same");
         return 0;
@@ -955,10 +973,8 @@
                       &cp, &len, &size , &bias) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     rv = PyBytes_FromStringAndSize(NULL, len);
     if ( rv == 0 )
@@ -991,10 +1007,8 @@
                       &cp, &len, &size) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     rv = PyBytes_FromStringAndSize(NULL, len);
     if ( rv == 0 )
@@ -1028,11 +1042,10 @@
                       &cp, &len, &size, &size2) )
         return 0;
 
-    if ( (size != 1 && size != 2 && size != 4) ||
-         (size2 != 1 && size2 != 2 && size2 != 4)) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
+    if (!audioop_check_size(size2))
+        return NULL;
 
     if (len/size > INT_MAX/size2) {
         PyErr_SetString(PyExc_MemoryError,
@@ -1082,10 +1095,8 @@
                           &nchannels, &inrate, &outrate, &state,
                           &weightA, &weightB))
         return NULL;
-    if (size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
+    if (!audioop_check_size(size))
         return NULL;
-    }
     if (nchannels < 1) {
         PyErr_SetString(AudioopError, "# of channels should be >= 1");
         return NULL;
@@ -1261,10 +1272,8 @@
                            &cp, &len, &size) )
         return 0 ;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     rv = PyBytes_FromStringAndSize(NULL, len/size);
     if ( rv == 0 )
@@ -1295,10 +1304,8 @@
                            &cp, &len, &size) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     if (len > INT_MAX/size) {
         PyErr_SetString(PyExc_MemoryError,
@@ -1334,10 +1341,8 @@
                            &cp, &len, &size) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     rv = PyBytes_FromStringAndSize(NULL, len/size);
     if ( rv == 0 )
@@ -1368,10 +1373,8 @@
                            &cp, &len, &size) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     if (len > INT_MAX/size) {
         PyErr_SetString(PyExc_MemoryError,
@@ -1408,11 +1411,8 @@
                            &cp, &len, &size, &state) )
         return 0;
 
-
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     str = PyBytes_FromStringAndSize(NULL, len/(size*2));
     if ( str == 0 )
@@ -1516,10 +1516,8 @@
                            &cp, &len, &size, &state) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     /* Decode state, should have (value, step) */
     if ( state == Py_None ) {

From python-checkins at python.org  Sat Jul  3 15:49:03 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sat,  3 Jul 2010 15:49:03 +0200 (CEST)
Subject: [Python-checkins] r82497 - python/trunk/Lib/idlelib/NEWS.txt
Message-ID: <20100703134903.2B054ED67@mail.python.org>

Author: benjamin.peterson
Date: Sat Jul  3 15:49:03 2010
New Revision: 82497

Log:
add release date

Modified:
   python/trunk/Lib/idlelib/NEWS.txt

Modified: python/trunk/Lib/idlelib/NEWS.txt
==============================================================================
--- python/trunk/Lib/idlelib/NEWS.txt	(original)
+++ python/trunk/Lib/idlelib/NEWS.txt	Sat Jul  3 15:49:03 2010
@@ -1,7 +1,7 @@
-What's New in IDLE 2.7a0?
-=========================
+What's New in IDLE 2.7?
+=======================
 
-*Release date: XX-XXX-2010*
+*Release date: 07-03-2010*
 
 - idle.py modified and simplified to better support developing experimental
   versions of IDLE which are not installed in the standard location.

From python-checkins at python.org  Sat Jul  3 15:49:24 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sat,  3 Jul 2010 15:49:24 +0200 (CEST)
Subject: [Python-checkins] r82498 - python/trunk/Misc/NEWS
Message-ID: <20100703134924.3EB9DED67@mail.python.org>

Author: benjamin.peterson
Date: Sat Jul  3 15:49:23 2010
New Revision: 82498

Log:
add period; rewrap

Modified:
   python/trunk/Misc/NEWS

Modified: python/trunk/Misc/NEWS
==============================================================================
--- python/trunk/Misc/NEWS	(original)
+++ python/trunk/Misc/NEWS	Sat Jul  3 15:49:23 2010
@@ -17,8 +17,8 @@
 Library
 -------
 
-- Issue #7673: Fix security vulnerability (CVE-2010-2089) in the audioop
-  module, ensure that the input string length is a multiple of the frame size
+- Issue #7673: Fix security vulnerability (CVE-2010-2089) in the audioop module,
+  ensure that the input string length is a multiple of the frame size.
 
 - Issue #9075: In the ssl module, remove the setting of a ``debug`` flag
   on an OpenSSL structure.

From python-checkins at python.org  Sat Jul  3 15:53:45 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sat,  3 Jul 2010 15:53:45 +0200 (CEST)
Subject: [Python-checkins] r82499 - python/trunk/Lib/pydoc_data/topics.py
Message-ID: <20100703135345.CCBF7EFBC@mail.python.org>

Author: benjamin.peterson
Date: Sat Jul  3 15:53:45 2010
New Revision: 82499

Log:
update pydoc-topics

Modified:
   python/trunk/Lib/pydoc_data/topics.py

Modified: python/trunk/Lib/pydoc_data/topics.py
==============================================================================
--- python/trunk/Lib/pydoc_data/topics.py	(original)
+++ python/trunk/Lib/pydoc_data/topics.py	Sat Jul  3 15:53:45 2010
@@ -1,4 +1,4 @@
-# Autogenerated by Sphinx on Sun Jun 20 10:55:52 2010
+# Autogenerated by Sphinx on Sat Jul  3 08:52:04 2010
 topics = {'assert': u'\nThe ``assert`` statement\n************************\n\nAssert statements are a convenient way to insert debugging assertions\ninto a program:\n\n   assert_stmt ::= "assert" expression ["," expression]\n\nThe simple form, ``assert expression``, is equivalent to\n\n   if __debug__:\n      if not expression: raise AssertionError\n\nThe extended form, ``assert expression1, expression2``, is equivalent\nto\n\n   if __debug__:\n      if not expression1: raise AssertionError(expression2)\n\nThese equivalences assume that ``__debug__`` and ``AssertionError``\nrefer to the built-in variables with those names.  In the current\nimplementation, the built-in variable ``__debug__`` is ``True`` under\nnormal circumstances, ``False`` when optimization is requested\n(command line option -O).  The current code generator emits no code\nfor an assert statement when optimization is requested at compile\ntime.  Note that it is unnecessary to include the source code for the\nexpression that failed in the error message; it will be displayed as\npart of the stack trace.\n\nAssignments to ``__debug__`` are illegal.  The value for the built-in\nvariable is determined when the interpreter starts.\n',
  'assignment': u'\nAssignment statements\n*********************\n\nAssignment statements are used to (re)bind names to values and to\nmodify attributes or items of mutable objects:\n\n   assignment_stmt ::= (target_list "=")+ (expression_list | yield_expression)\n   target_list     ::= target ("," target)* [","]\n   target          ::= identifier\n              | "(" target_list ")"\n              | "[" target_list "]"\n              | attributeref\n              | subscription\n              | slicing\n\n(See section *Primaries* for the syntax definitions for the last three\nsymbols.)\n\nAn assignment statement evaluates the expression list (remember that\nthis can be a single expression or a comma-separated list, the latter\nyielding a tuple) and assigns the single resulting object to each of\nthe target lists, from left to right.\n\nAssignment is defined recursively depending on the form of the target\n(list). When a target is part of a mutable object (an attribute\nreference, subscription or slicing), the mutable object must\nultimately perform the assignment and decide about its validity, and\nmay raise an exception if the assignment is unacceptable.  The rules\nobserved by various types and the exceptions raised are given with the\ndefinition of the object types (see section *The standard type\nhierarchy*).\n\nAssignment of an object to a target list is recursively defined as\nfollows.\n\n* If the target list is a single target: The object is assigned to\n  that target.\n\n* If the target list is a comma-separated list of targets: The object\n  must be an iterable with the same number of items as there are\n  targets in the target list, and the items are assigned, from left to\n  right, to the corresponding targets. (This rule is relaxed as of\n  Python 1.5; in earlier versions, the object had to be a tuple.\n  Since strings are sequences, an assignment like ``a, b = "xy"`` is\n  now legal as long as the string has the right length.)\n\nAssignment of an object to a single target is recursively defined as\nfollows.\n\n* If the target is an identifier (name):\n\n  * If the name does not occur in a ``global`` statement in the\n    current code block: the name is bound to the object in the current\n    local namespace.\n\n  * Otherwise: the name is bound to the object in the current global\n    namespace.\n\n  The name is rebound if it was already bound.  This may cause the\n  reference count for the object previously bound to the name to reach\n  zero, causing the object to be deallocated and its destructor (if it\n  has one) to be called.\n\n* If the target is a target list enclosed in parentheses or in square\n  brackets: The object must be an iterable with the same number of\n  items as there are targets in the target list, and its items are\n  assigned, from left to right, to the corresponding targets.\n\n* If the target is an attribute reference: The primary expression in\n  the reference is evaluated.  It should yield an object with\n  assignable attributes; if this is not the case, ``TypeError`` is\n  raised.  That object is then asked to assign the assigned object to\n  the given attribute; if it cannot perform the assignment, it raises\n  an exception (usually but not necessarily ``AttributeError``).\n\n  Note: If the object is a class instance and the attribute reference\n  occurs on both sides of the assignment operator, the RHS expression,\n  ``a.x`` can access either an instance attribute or (if no instance\n  attribute exists) a class attribute.  The LHS target ``a.x`` is\n  always set as an instance attribute, creating it if necessary.\n  Thus, the two occurrences of ``a.x`` do not necessarily refer to the\n  same attribute: if the RHS expression refers to a class attribute,\n  the LHS creates a new instance attribute as the target of the\n  assignment:\n\n     class Cls:\n         x = 3             # class variable\n     inst = Cls()\n     inst.x = inst.x + 1   # writes inst.x as 4 leaving Cls.x as 3\n\n  This description does not necessarily apply to descriptor\n  attributes, such as properties created with ``property()``.\n\n* If the target is a subscription: The primary expression in the\n  reference is evaluated.  It should yield either a mutable sequence\n  object (such as a list) or a mapping object (such as a dictionary).\n  Next, the subscript expression is evaluated.\n\n  If the primary is a mutable sequence object (such as a list), the\n  subscript must yield a plain integer.  If it is negative, the\n  sequence\'s length is added to it. The resulting value must be a\n  nonnegative integer less than the sequence\'s length, and the\n  sequence is asked to assign the assigned object to its item with\n  that index.  If the index is out of range, ``IndexError`` is raised\n  (assignment to a subscripted sequence cannot add new items to a\n  list).\n\n  If the primary is a mapping object (such as a dictionary), the\n  subscript must have a type compatible with the mapping\'s key type,\n  and the mapping is then asked to create a key/datum pair which maps\n  the subscript to the assigned object.  This can either replace an\n  existing key/value pair with the same key value, or insert a new\n  key/value pair (if no key with the same value existed).\n\n* If the target is a slicing: The primary expression in the reference\n  is evaluated.  It should yield a mutable sequence object (such as a\n  list).  The assigned object should be a sequence object of the same\n  type.  Next, the lower and upper bound expressions are evaluated,\n  insofar they are present; defaults are zero and the sequence\'s\n  length.  The bounds should evaluate to (small) integers.  If either\n  bound is negative, the sequence\'s length is added to it. The\n  resulting bounds are clipped to lie between zero and the sequence\'s\n  length, inclusive.  Finally, the sequence object is asked to replace\n  the slice with the items of the assigned sequence.  The length of\n  the slice may be different from the length of the assigned sequence,\n  thus changing the length of the target sequence, if the object\n  allows it.\n\n**CPython implementation detail:** In the current implementation, the\nsyntax for targets is taken to be the same as for expressions, and\ninvalid syntax is rejected during the code generation phase, causing\nless detailed error messages.\n\nWARNING: Although the definition of assignment implies that overlaps\nbetween the left-hand side and the right-hand side are \'safe\' (for\nexample ``a, b = b, a`` swaps two variables), overlaps *within* the\ncollection of assigned-to variables are not safe!  For instance, the\nfollowing program prints ``[0, 2]``:\n\n   x = [0, 1]\n   i = 0\n   i, x[i] = 1, 2\n   print x\n\n\nAugmented assignment statements\n===============================\n\nAugmented assignment is the combination, in a single statement, of a\nbinary operation and an assignment statement:\n\n   augmented_assignment_stmt ::= augtarget augop (expression_list | yield_expression)\n   augtarget                 ::= identifier | attributeref | subscription | slicing\n   augop                     ::= "+=" | "-=" | "*=" | "/=" | "//=" | "%=" | "**="\n             | ">>=" | "<<=" | "&=" | "^=" | "|="\n\n(See section *Primaries* for the syntax definitions for the last three\nsymbols.)\n\nAn augmented assignment evaluates the target (which, unlike normal\nassignment statements, cannot be an unpacking) and the expression\nlist, performs the binary operation specific to the type of assignment\non the two operands, and assigns the result to the original target.\nThe target is only evaluated once.\n\nAn augmented assignment expression like ``x += 1`` can be rewritten as\n``x = x + 1`` to achieve a similar, but not exactly equal effect. In\nthe augmented version, ``x`` is only evaluated once. Also, when\npossible, the actual operation is performed *in-place*, meaning that\nrather than creating a new object and assigning that to the target,\nthe old object is modified instead.\n\nWith the exception of assigning to tuples and multiple targets in a\nsingle statement, the assignment done by augmented assignment\nstatements is handled the same way as normal assignments. Similarly,\nwith the exception of the possible *in-place* behavior, the binary\noperation performed by augmented assignment is the same as the normal\nbinary operations.\n\nFor targets which are attribute references, the same *caveat about\nclass and instance attributes* applies as for regular assignments.\n',
  'atom-identifiers': u'\nIdentifiers (Names)\n*******************\n\nAn identifier occurring as an atom is a name.  See section\n*Identifiers and keywords* for lexical definition and section *Naming\nand binding* for documentation of naming and binding.\n\nWhen the name is bound to an object, evaluation of the atom yields\nthat object. When a name is not bound, an attempt to evaluate it\nraises a ``NameError`` exception.\n\n**Private name mangling:** When an identifier that textually occurs in\na class definition begins with two or more underscore characters and\ndoes not end in two or more underscores, it is considered a *private\nname* of that class. Private names are transformed to a longer form\nbefore code is generated for them.  The transformation inserts the\nclass name in front of the name, with leading underscores removed, and\na single underscore inserted in front of the class name.  For example,\nthe identifier ``__spam`` occurring in a class named ``Ham`` will be\ntransformed to ``_Ham__spam``.  This transformation is independent of\nthe syntactical context in which the identifier is used.  If the\ntransformed name is extremely long (longer than 255 characters),\nimplementation defined truncation may happen.  If the class name\nconsists only of underscores, no transformation is done.\n',
@@ -36,7 +36,7 @@
  'exprlists': u'\nExpression lists\n****************\n\n   expression_list ::= expression ( "," expression )* [","]\n\nAn expression list containing at least one comma yields a tuple.  The\nlength of the tuple is the number of expressions in the list.  The\nexpressions are evaluated from left to right.\n\nThe trailing comma is required only to create a single tuple (a.k.a. a\n*singleton*); it is optional in all other cases.  A single expression\nwithout a trailing comma doesn\'t create a tuple, but rather yields the\nvalue of that expression. (To create an empty tuple, use an empty pair\nof parentheses: ``()``.)\n',
  'floating': u'\nFloating point literals\n***********************\n\nFloating point literals are described by the following lexical\ndefinitions:\n\n   floatnumber   ::= pointfloat | exponentfloat\n   pointfloat    ::= [intpart] fraction | intpart "."\n   exponentfloat ::= (intpart | pointfloat) exponent\n   intpart       ::= digit+\n   fraction      ::= "." digit+\n   exponent      ::= ("e" | "E") ["+" | "-"] digit+\n\nNote that the integer and exponent parts of floating point numbers can\nlook like octal integers, but are interpreted using radix 10.  For\nexample, ``077e010`` is legal, and denotes the same number as\n``77e10``. The allowed range of floating point literals is\nimplementation-dependent. Some examples of floating point literals:\n\n   3.14    10.    .001    1e100    3.14e-10    0e0\n\nNote that numeric literals do not include a sign; a phrase like ``-1``\nis actually an expression composed of the unary operator ``-`` and the\nliteral ``1``.\n',
  'for': u'\nThe ``for`` statement\n*********************\n\nThe ``for`` statement is used to iterate over the elements of a\nsequence (such as a string, tuple or list) or other iterable object:\n\n   for_stmt ::= "for" target_list "in" expression_list ":" suite\n                ["else" ":" suite]\n\nThe expression list is evaluated once; it should yield an iterable\nobject.  An iterator is created for the result of the\n``expression_list``.  The suite is then executed once for each item\nprovided by the iterator, in the order of ascending indices.  Each\nitem in turn is assigned to the target list using the standard rules\nfor assignments, and then the suite is executed.  When the items are\nexhausted (which is immediately when the sequence is empty), the suite\nin the ``else`` clause, if present, is executed, and the loop\nterminates.\n\nA ``break`` statement executed in the first suite terminates the loop\nwithout executing the ``else`` clause\'s suite.  A ``continue``\nstatement executed in the first suite skips the rest of the suite and\ncontinues with the next item, or with the ``else`` clause if there was\nno next item.\n\nThe suite may assign to the variable(s) in the target list; this does\nnot affect the next item assigned to it.\n\nThe target list is not deleted when the loop is finished, but if the\nsequence is empty, it will not have been assigned to at all by the\nloop.  Hint: the built-in function ``range()`` returns a sequence of\nintegers suitable to emulate the effect of Pascal\'s ``for i := a to b\ndo``; e.g., ``range(3)`` returns the list ``[0, 1, 2]``.\n\nNote: There is a subtlety when the sequence is being modified by the loop\n  (this can only occur for mutable sequences, i.e. lists). An internal\n  counter is used to keep track of which item is used next, and this\n  is incremented on each iteration.  When this counter has reached the\n  length of the sequence the loop terminates.  This means that if the\n  suite deletes the current (or a previous) item from the sequence,\n  the next item will be skipped (since it gets the index of the\n  current item which has already been treated).  Likewise, if the\n  suite inserts an item in the sequence before the current item, the\n  current item will be treated again the next time through the loop.\n  This can lead to nasty bugs that can be avoided by making a\n  temporary copy using a slice of the whole sequence, e.g.,\n\n     for x in a[:]:\n         if x < 0: a.remove(x)\n',
- 'formatstrings': u'\nFormat String Syntax\n********************\n\nThe ``str.format()`` method and the ``Formatter`` class share the same\nsyntax for format strings (although in the case of ``Formatter``,\nsubclasses can define their own format string syntax.)\n\nFormat strings contain "replacement fields" surrounded by curly braces\n``{}``. Anything that is not contained in braces is considered literal\ntext, which is copied unchanged to the output.  If you need to include\na brace character in the literal text, it can be escaped by doubling:\n``{{`` and ``}}``.\n\nThe grammar for a replacement field is as follows:\n\n      replacement_field ::= "{" [field_name] ["!" conversion] [":" format_spec] "}"\n      field_name        ::= arg_name ("." attribute_name | "[" element_index "]")*\n      arg_name          ::= (identifier | integer)?\n      attribute_name    ::= identifier\n      element_index     ::= integer | index_string\n      index_string      ::= <any source character except "]"> +\n      conversion        ::= "r" | "s"\n      format_spec       ::= <described in the next section>\n\nIn less formal terms, the replacement field can start with a\n*field_name* that specifies the object whose value is to be formatted\nand inserted into the output instead of the replacement field. The\n*field_name* is optionally followed by a  *conversion* field, which is\npreceded by an exclamation point ``\'!\'``, and a *format_spec*, which\nis preceded by a colon ``\':\'``.  These specify a non-default format\nfor the replacement value.\n\nThe *field_name* itself begins with an *arg_name* that is either\neither a number or a keyword.  If it\'s a number, it refers to a\npositional argument, and if it\'s a keyword, it refers to a named\nkeyword argument.  If the numerical arg_names in a format string are\n0, 1, 2, ... in sequence, they can all be omitted (not just some) and\nthe numbers 0, 1, 2, ... will be automatically inserted in that order.\nThe *arg_name* can be followed by any number of index or attribute\nexpressions. An expression of the form ``\'.name\'`` selects the named\nattribute using ``getattr()``, while an expression of the form\n``\'[index]\'`` does an index lookup using ``__getitem__()``.\n\nSome simple format string examples:\n\n   "First, thou shalt count to {0}" # References first positional argument\n   "Bring me a {}"                  # Implicitly references the first positional argument\n   "From {} to {}"                  # Same as "From {0} to {1}"\n   "My quest is {name}"             # References keyword argument \'name\'\n   "Weight in tons {0.weight}"      # \'weight\' attribute of first positional arg\n   "Units destroyed: {players[0]}"  # First element of keyword argument \'players\'.\n\nThe *conversion* field causes a type coercion before formatting.\nNormally, the job of formatting a value is done by the\n``__format__()`` method of the value itself.  However, in some cases\nit is desirable to force a type to be formatted as a string,\noverriding its own definition of formatting.  By converting the value\nto a string before calling ``__format__()``, the normal formatting\nlogic is bypassed.\n\nTwo conversion flags are currently supported: ``\'!s\'`` which calls\n``str()`` on the value, and ``\'!r\'`` which calls ``repr()``.\n\nSome examples:\n\n   "Harold\'s a clever {0!s}"        # Calls str() on the argument first\n   "Bring out the holy {name!r}"    # Calls repr() on the argument first\n\nThe *format_spec* field contains a specification of how the value\nshould be presented, including such details as field width, alignment,\npadding, decimal precision and so on.  Each value type can define its\nown "formatting mini-language" or interpretation of the *format_spec*.\n\nMost built-in types support a common formatting mini-language, which\nis described in the next section.\n\nA *format_spec* field can also include nested replacement fields\nwithin it. These nested replacement fields can contain only a field\nname; conversion flags and format specifications are not allowed.  The\nreplacement fields within the format_spec are substituted before the\n*format_spec* string is interpreted. This allows the formatting of a\nvalue to be dynamically specified.\n\nFor example, suppose you wanted to have a replacement field whose\nfield width is determined by another variable:\n\n   "A man with two {0:{1}}".format("noses", 10)\n\nThis would first evaluate the inner replacement field, making the\nformat string effectively:\n\n   "A man with two {0:10}"\n\nThen the outer replacement field would be evaluated, producing:\n\n   "noses     "\n\nWhich is substituted into the string, yielding:\n\n   "A man with two noses     "\n\n(The extra space is because we specified a field width of 10, and\nbecause left alignment is the default for strings.)\n\n\nFormat Specification Mini-Language\n==================================\n\n"Format specifications" are used within replacement fields contained\nwithin a format string to define how individual values are presented\n(see *Format String Syntax*.)  They can also be passed directly to the\nbuilt-in ``format()`` function.  Each formattable type may define how\nthe format specification is to be interpreted.\n\nMost built-in types implement the following options for format\nspecifications, although some of the formatting options are only\nsupported by the numeric types.\n\nA general convention is that an empty format string (``""``) produces\nthe same result as if you had called ``str()`` on the value. A non-\nempty format string typically modifies the result.\n\nThe general form of a *standard format specifier* is:\n\n   format_spec ::= [[fill]align][sign][#][0][width][,][.precision][type]\n   fill        ::= <a character other than \'}\'>\n   align       ::= "<" | ">" | "=" | "^"\n   sign        ::= "+" | "-" | " "\n   width       ::= integer\n   precision   ::= integer\n   type        ::= "b" | "c" | "d" | "e" | "E" | "f" | "F" | "g" | "G" | "n" | "o" | "s" | "x" | "X" | "%"\n\nThe *fill* character can be any character other than \'}\' (which\nsignifies the end of the field).  The presence of a fill character is\nsignaled by the *next* character, which must be one of the alignment\noptions. If the second character of *format_spec* is not a valid\nalignment option, then it is assumed that both the fill character and\nthe alignment option are absent.\n\nThe meaning of the various alignment options is as follows:\n\n   +-----------+------------------------------------------------------------+\n   | Option    | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'<\'``   | Forces the field to be left-aligned within the available   |\n   |           | space (This is the default.)                               |\n   +-----------+------------------------------------------------------------+\n   | ``\'>\'``   | Forces the field to be right-aligned within the available  |\n   |           | space.                                                     |\n   +-----------+------------------------------------------------------------+\n   | ``\'=\'``   | Forces the padding to be placed after the sign (if any)    |\n   |           | but before the digits.  This is used for printing fields   |\n   |           | in the form \'+000000120\'. This alignment option is only    |\n   |           | valid for numeric types.                                   |\n   +-----------+------------------------------------------------------------+\n   | ``\'^\'``   | Forces the field to be centered within the available       |\n   |           | space.                                                     |\n   +-----------+------------------------------------------------------------+\n\nNote that unless a minimum field width is defined, the field width\nwill always be the same size as the data to fill it, so that the\nalignment option has no meaning in this case.\n\nThe *sign* option is only valid for number types, and can be one of\nthe following:\n\n   +-----------+------------------------------------------------------------+\n   | Option    | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'+\'``   | indicates that a sign should be used for both positive as  |\n   |           | well as negative numbers.                                  |\n   +-----------+------------------------------------------------------------+\n   | ``\'-\'``   | indicates that a sign should be used only for negative     |\n   |           | numbers (this is the default behavior).                    |\n   +-----------+------------------------------------------------------------+\n   | space     | indicates that a leading space should be used on positive  |\n   |           | numbers, and a minus sign on negative numbers.             |\n   +-----------+------------------------------------------------------------+\n\nThe ``\'#\'`` option is only valid for integers, and only for binary,\noctal, or hexadecimal output.  If present, it specifies that the\noutput will be prefixed by ``\'0b\'``, ``\'0o\'``, or ``\'0x\'``,\nrespectively.\n\nThe ``\',\'`` option signals the use of a comma for a thousands\nseparator. For a locale aware separator, use the ``\'n\'`` integer\npresentation type instead.\n\n*width* is a decimal integer defining the minimum field width.  If not\nspecified, then the field width will be determined by the content.\n\nIf the *width* field is preceded by a zero (``\'0\'``) character, this\nenables zero-padding.  This is equivalent to an *alignment* type of\n``\'=\'`` and a *fill* character of ``\'0\'``.\n\nThe *precision* is a decimal number indicating how many digits should\nbe displayed after the decimal point for a floating point value\nformatted with ``\'f\'`` and ``\'F\'``, or before and after the decimal\npoint for a floating point value formatted with ``\'g\'`` or ``\'G\'``.\nFor non-number types the field indicates the maximum field size - in\nother words, how many characters will be used from the field content.\nThe *precision* is not allowed for integer values.\n\nFinally, the *type* determines how the data should be presented.\n\nThe available string presentation types are:\n\n   +-----------+------------------------------------------------------------+\n   | Type      | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'s\'``   | String format. This is the default type for strings and    |\n   |           | may be omitted.                                            |\n   +-----------+------------------------------------------------------------+\n   | None      | The same as ``\'s\'``.                                       |\n   +-----------+------------------------------------------------------------+\n\nThe available integer presentation types are:\n\n   +-----------+------------------------------------------------------------+\n   | Type      | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'b\'``   | Binary format. Outputs the number in base 2.               |\n   +-----------+------------------------------------------------------------+\n   | ``\'c\'``   | Character. Converts the integer to the corresponding       |\n   |           | unicode character before printing.                         |\n   +-----------+------------------------------------------------------------+\n   | ``\'d\'``   | Decimal Integer. Outputs the number in base 10.            |\n   +-----------+------------------------------------------------------------+\n   | ``\'o\'``   | Octal format. Outputs the number in base 8.                |\n   +-----------+------------------------------------------------------------+\n   | ``\'x\'``   | Hex format. Outputs the number in base 16, using lower-    |\n   |           | case letters for the digits above 9.                       |\n   +-----------+------------------------------------------------------------+\n   | ``\'X\'``   | Hex format. Outputs the number in base 16, using upper-    |\n   |           | case letters for the digits above 9.                       |\n   +-----------+------------------------------------------------------------+\n   | ``\'n\'``   | Number. This is the same as ``\'d\'``, except that it uses   |\n   |           | the current locale setting to insert the appropriate       |\n   |           | number separator characters.                               |\n   +-----------+------------------------------------------------------------+\n   | None      | The same as ``\'d\'``.                                       |\n   +-----------+------------------------------------------------------------+\n\nIn addition to the above presentation types, integers can be formatted\nwith the floating point presentation types listed below (except\n``\'n\'`` and None). When doing so, ``float()`` is used to convert the\ninteger to a floating point number before formatting.\n\nThe available presentation types for floating point and decimal values\nare:\n\n   +-----------+------------------------------------------------------------+\n   | Type      | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'e\'``   | Exponent notation. Prints the number in scientific         |\n   |           | notation using the letter \'e\' to indicate the exponent.    |\n   +-----------+------------------------------------------------------------+\n   | ``\'E\'``   | Exponent notation. Same as ``\'e\'`` except it uses an upper |\n   |           | case \'E\' as the separator character.                       |\n   +-----------+------------------------------------------------------------+\n   | ``\'f\'``   | Fixed point. Displays the number as a fixed-point number.  |\n   +-----------+------------------------------------------------------------+\n   | ``\'F\'``   | Fixed point. Same as ``\'f\'``.                              |\n   +-----------+------------------------------------------------------------+\n   | ``\'g\'``   | General format.  For a given precision ``p >= 1``, this    |\n   |           | rounds the number to ``p`` significant digits and then     |\n   |           | formats the result in either fixed-point format or in      |\n   |           | scientific notation, depending on its magnitude.  The      |\n   |           | precise rules are as follows: suppose that the result      |\n   |           | formatted with presentation type ``\'e\'`` and precision     |\n   |           | ``p-1`` would have exponent ``exp``.  Then if ``-4 <= exp  |\n   |           | < p``, the number is formatted with presentation type      |\n   |           | ``\'f\'`` and precision ``p-1-exp``. Otherwise, the number   |\n   |           | is formatted with presentation type ``\'e\'`` and precision  |\n   |           | ``p-1``. In both cases insignificant trailing zeros are    |\n   |           | removed from the significand, and the decimal point is     |\n   |           | also removed if there are no remaining digits following    |\n   |           | it.  Postive and negative infinity, positive and negative  |\n   |           | zero, and nans, are formatted as ``inf``, ``-inf``, ``0``, |\n   |           | ``-0`` and ``nan`` respectively, regardless of the         |\n   |           | precision.  A precision of ``0`` is treated as equivalent  |\n   |           | to a precision of ``1``.                                   |\n   +-----------+------------------------------------------------------------+\n   | ``\'G\'``   | General format. Same as ``\'g\'`` except switches to ``\'E\'`` |\n   |           | if the number gets too large. The representations of       |\n   |           | infinity and NaN are uppercased, too.                      |\n   +-----------+------------------------------------------------------------+\n   | ``\'n\'``   | Number. This is the same as ``\'g\'``, except that it uses   |\n   |           | the current locale setting to insert the appropriate       |\n   |           | number separator characters.                               |\n   +-----------+------------------------------------------------------------+\n   | ``\'%\'``   | Percentage. Multiplies the number by 100 and displays in   |\n   |           | fixed (``\'f\'``) format, followed by a percent sign.        |\n   +-----------+------------------------------------------------------------+\n   | None      | The same as ``\'g\'``.                                       |\n   +-----------+------------------------------------------------------------+\n',
+ 'formatstrings': u'\nFormat String Syntax\n********************\n\nThe ``str.format()`` method and the ``Formatter`` class share the same\nsyntax for format strings (although in the case of ``Formatter``,\nsubclasses can define their own format string syntax).\n\nFormat strings contain "replacement fields" surrounded by curly braces\n``{}``. Anything that is not contained in braces is considered literal\ntext, which is copied unchanged to the output.  If you need to include\na brace character in the literal text, it can be escaped by doubling:\n``{{`` and ``}}``.\n\nThe grammar for a replacement field is as follows:\n\n      replacement_field ::= "{" [field_name] ["!" conversion] [":" format_spec] "}"\n      field_name        ::= arg_name ("." attribute_name | "[" element_index "]")*\n      arg_name          ::= [identifier | integer]\n      attribute_name    ::= identifier\n      element_index     ::= integer | index_string\n      index_string      ::= <any source character except "]"> +\n      conversion        ::= "r" | "s"\n      format_spec       ::= <described in the next section>\n\nIn less formal terms, the replacement field can start with a\n*field_name* that specifies the object whose value is to be formatted\nand inserted into the output instead of the replacement field. The\n*field_name* is optionally followed by a  *conversion* field, which is\npreceded by an exclamation point ``\'!\'``, and a *format_spec*, which\nis preceded by a colon ``\':\'``.  These specify a non-default format\nfor the replacement value.\n\nSee also the *Format Specification Mini-Language* section.\n\nThe *field_name* itself begins with an *arg_name* that is either\neither a number or a keyword.  If it\'s a number, it refers to a\npositional argument, and if it\'s a keyword, it refers to a named\nkeyword argument.  If the numerical arg_names in a format string are\n0, 1, 2, ... in sequence, they can all be omitted (not just some) and\nthe numbers 0, 1, 2, ... will be automatically inserted in that order.\nThe *arg_name* can be followed by any number of index or attribute\nexpressions. An expression of the form ``\'.name\'`` selects the named\nattribute using ``getattr()``, while an expression of the form\n``\'[index]\'`` does an index lookup using ``__getitem__()``.\n\nChanged in version 2.7: The positional argument specifiers can be\nomitted, so ``\'{} {}\'`` is equivalent to ``\'{0} {1}\'``.\n\nSome simple format string examples:\n\n   "First, thou shalt count to {0}" # References first positional argument\n   "Bring me a {}"                  # Implicitly references the first positional argument\n   "From {} to {}"                  # Same as "From {0} to {1}"\n   "My quest is {name}"             # References keyword argument \'name\'\n   "Weight in tons {0.weight}"      # \'weight\' attribute of first positional arg\n   "Units destroyed: {players[0]}"  # First element of keyword argument \'players\'.\n\nThe *conversion* field causes a type coercion before formatting.\nNormally, the job of formatting a value is done by the\n``__format__()`` method of the value itself.  However, in some cases\nit is desirable to force a type to be formatted as a string,\noverriding its own definition of formatting.  By converting the value\nto a string before calling ``__format__()``, the normal formatting\nlogic is bypassed.\n\nTwo conversion flags are currently supported: ``\'!s\'`` which calls\n``str()`` on the value, and ``\'!r\'`` which calls ``repr()``.\n\nSome examples:\n\n   "Harold\'s a clever {0!s}"        # Calls str() on the argument first\n   "Bring out the holy {name!r}"    # Calls repr() on the argument first\n\nThe *format_spec* field contains a specification of how the value\nshould be presented, including such details as field width, alignment,\npadding, decimal precision and so on.  Each value type can define its\nown "formatting mini-language" or interpretation of the *format_spec*.\n\nMost built-in types support a common formatting mini-language, which\nis described in the next section.\n\nA *format_spec* field can also include nested replacement fields\nwithin it. These nested replacement fields can contain only a field\nname; conversion flags and format specifications are not allowed.  The\nreplacement fields within the format_spec are substituted before the\n*format_spec* string is interpreted. This allows the formatting of a\nvalue to be dynamically specified.\n\nSee the *Format examples* section for some examples.\n\n\nFormat Specification Mini-Language\n==================================\n\n"Format specifications" are used within replacement fields contained\nwithin a format string to define how individual values are presented\n(see *Format String Syntax*).  They can also be passed directly to the\nbuilt-in ``format()`` function.  Each formattable type may define how\nthe format specification is to be interpreted.\n\nMost built-in types implement the following options for format\nspecifications, although some of the formatting options are only\nsupported by the numeric types.\n\nA general convention is that an empty format string (``""``) produces\nthe same result as if you had called ``str()`` on the value. A non-\nempty format string typically modifies the result.\n\nThe general form of a *standard format specifier* is:\n\n   format_spec ::= [[fill]align][sign][#][0][width][,][.precision][type]\n   fill        ::= <a character other than \'}\'>\n   align       ::= "<" | ">" | "=" | "^"\n   sign        ::= "+" | "-" | " "\n   width       ::= integer\n   precision   ::= integer\n   type        ::= "b" | "c" | "d" | "e" | "E" | "f" | "F" | "g" | "G" | "n" | "o" | "s" | "x" | "X" | "%"\n\nThe *fill* character can be any character other than \'}\' (which\nsignifies the end of the field).  The presence of a fill character is\nsignaled by the *next* character, which must be one of the alignment\noptions. If the second character of *format_spec* is not a valid\nalignment option, then it is assumed that both the fill character and\nthe alignment option are absent.\n\nThe meaning of the various alignment options is as follows:\n\n   +-----------+------------------------------------------------------------+\n   | Option    | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'<\'``   | Forces the field to be left-aligned within the available   |\n   |           | space (this is the default).                               |\n   +-----------+------------------------------------------------------------+\n   | ``\'>\'``   | Forces the field to be right-aligned within the available  |\n   |           | space.                                                     |\n   +-----------+------------------------------------------------------------+\n   | ``\'=\'``   | Forces the padding to be placed after the sign (if any)    |\n   |           | but before the digits.  This is used for printing fields   |\n   |           | in the form \'+000000120\'. This alignment option is only    |\n   |           | valid for numeric types.                                   |\n   +-----------+------------------------------------------------------------+\n   | ``\'^\'``   | Forces the field to be centered within the available       |\n   |           | space.                                                     |\n   +-----------+------------------------------------------------------------+\n\nNote that unless a minimum field width is defined, the field width\nwill always be the same size as the data to fill it, so that the\nalignment option has no meaning in this case.\n\nThe *sign* option is only valid for number types, and can be one of\nthe following:\n\n   +-----------+------------------------------------------------------------+\n   | Option    | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'+\'``   | indicates that a sign should be used for both positive as  |\n   |           | well as negative numbers.                                  |\n   +-----------+------------------------------------------------------------+\n   | ``\'-\'``   | indicates that a sign should be used only for negative     |\n   |           | numbers (this is the default behavior).                    |\n   +-----------+------------------------------------------------------------+\n   | space     | indicates that a leading space should be used on positive  |\n   |           | numbers, and a minus sign on negative numbers.             |\n   +-----------+------------------------------------------------------------+\n\nThe ``\'#\'`` option is only valid for integers, and only for binary,\noctal, or hexadecimal output.  If present, it specifies that the\noutput will be prefixed by ``\'0b\'``, ``\'0o\'``, or ``\'0x\'``,\nrespectively.\n\nThe ``\',\'`` option signals the use of a comma for a thousands\nseparator. For a locale aware separator, use the ``\'n\'`` integer\npresentation type instead.\n\nChanged in version 2.7: Added the ``\',\'`` option (see also **PEP\n378**).\n\n*width* is a decimal integer defining the minimum field width.  If not\nspecified, then the field width will be determined by the content.\n\nIf the *width* field is preceded by a zero (``\'0\'``) character, this\nenables zero-padding.  This is equivalent to an *alignment* type of\n``\'=\'`` and a *fill* character of ``\'0\'``.\n\nThe *precision* is a decimal number indicating how many digits should\nbe displayed after the decimal point for a floating point value\nformatted with ``\'f\'`` and ``\'F\'``, or before and after the decimal\npoint for a floating point value formatted with ``\'g\'`` or ``\'G\'``.\nFor non-number types the field indicates the maximum field size - in\nother words, how many characters will be used from the field content.\nThe *precision* is not allowed for integer values.\n\nFinally, the *type* determines how the data should be presented.\n\nThe available string presentation types are:\n\n   +-----------+------------------------------------------------------------+\n   | Type      | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'s\'``   | String format. This is the default type for strings and    |\n   |           | may be omitted.                                            |\n   +-----------+------------------------------------------------------------+\n   | None      | The same as ``\'s\'``.                                       |\n   +-----------+------------------------------------------------------------+\n\nThe available integer presentation types are:\n\n   +-----------+------------------------------------------------------------+\n   | Type      | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'b\'``   | Binary format. Outputs the number in base 2.               |\n   +-----------+------------------------------------------------------------+\n   | ``\'c\'``   | Character. Converts the integer to the corresponding       |\n   |           | unicode character before printing.                         |\n   +-----------+------------------------------------------------------------+\n   | ``\'d\'``   | Decimal Integer. Outputs the number in base 10.            |\n   +-----------+------------------------------------------------------------+\n   | ``\'o\'``   | Octal format. Outputs the number in base 8.                |\n   +-----------+------------------------------------------------------------+\n   | ``\'x\'``   | Hex format. Outputs the number in base 16, using lower-    |\n   |           | case letters for the digits above 9.                       |\n   +-----------+------------------------------------------------------------+\n   | ``\'X\'``   | Hex format. Outputs the number in base 16, using upper-    |\n   |           | case letters for the digits above 9.                       |\n   +-----------+------------------------------------------------------------+\n   | ``\'n\'``   | Number. This is the same as ``\'d\'``, except that it uses   |\n   |           | the current locale setting to insert the appropriate       |\n   |           | number separator characters.                               |\n   +-----------+------------------------------------------------------------+\n   | None      | The same as ``\'d\'``.                                       |\n   +-----------+------------------------------------------------------------+\n\nIn addition to the above presentation types, integers can be formatted\nwith the floating point presentation types listed below (except\n``\'n\'`` and None). When doing so, ``float()`` is used to convert the\ninteger to a floating point number before formatting.\n\nThe available presentation types for floating point and decimal values\nare:\n\n   +-----------+------------------------------------------------------------+\n   | Type      | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'e\'``   | Exponent notation. Prints the number in scientific         |\n   |           | notation using the letter \'e\' to indicate the exponent.    |\n   +-----------+------------------------------------------------------------+\n   | ``\'E\'``   | Exponent notation. Same as ``\'e\'`` except it uses an upper |\n   |           | case \'E\' as the separator character.                       |\n   +-----------+------------------------------------------------------------+\n   | ``\'f\'``   | Fixed point. Displays the number as a fixed-point number.  |\n   +-----------+------------------------------------------------------------+\n   | ``\'F\'``   | Fixed point. Same as ``\'f\'``.                              |\n   +-----------+------------------------------------------------------------+\n   | ``\'g\'``   | General format.  For a given precision ``p >= 1``, this    |\n   |           | rounds the number to ``p`` significant digits and then     |\n   |           | formats the result in either fixed-point format or in      |\n   |           | scientific notation, depending on its magnitude.  The      |\n   |           | precise rules are as follows: suppose that the result      |\n   |           | formatted with presentation type ``\'e\'`` and precision     |\n   |           | ``p-1`` would have exponent ``exp``.  Then if ``-4 <= exp  |\n   |           | < p``, the number is formatted with presentation type      |\n   |           | ``\'f\'`` and precision ``p-1-exp``. Otherwise, the number   |\n   |           | is formatted with presentation type ``\'e\'`` and precision  |\n   |           | ``p-1``. In both cases insignificant trailing zeros are    |\n   |           | removed from the significand, and the decimal point is     |\n   |           | also removed if there are no remaining digits following    |\n   |           | it.  Postive and negative infinity, positive and negative  |\n   |           | zero, and nans, are formatted as ``inf``, ``-inf``, ``0``, |\n   |           | ``-0`` and ``nan`` respectively, regardless of the         |\n   |           | precision.  A precision of ``0`` is treated as equivalent  |\n   |           | to a precision of ``1``.                                   |\n   +-----------+------------------------------------------------------------+\n   | ``\'G\'``   | General format. Same as ``\'g\'`` except switches to ``\'E\'`` |\n   |           | if the number gets too large. The representations of       |\n   |           | infinity and NaN are uppercased, too.                      |\n   +-----------+------------------------------------------------------------+\n   | ``\'n\'``   | Number. This is the same as ``\'g\'``, except that it uses   |\n   |           | the current locale setting to insert the appropriate       |\n   |           | number separator characters.                               |\n   +-----------+------------------------------------------------------------+\n   | ``\'%\'``   | Percentage. Multiplies the number by 100 and displays in   |\n   |           | fixed (``\'f\'``) format, followed by a percent sign.        |\n   +-----------+------------------------------------------------------------+\n   | None      | The same as ``\'g\'``.                                       |\n   +-----------+------------------------------------------------------------+\n\n\nFormat examples\n===============\n\nThis section contains examples of the new format syntax and comparison\nwith the old ``%``-formatting.\n\nIn most of the cases the syntax is similar to the old\n``%``-formatting, with the addition of the ``{}`` and with ``:`` used\ninstead of ``%``. For example, ``\'%03.2f\'`` can be translated to\n``\'{:03.2f}\'``.\n\nThe new format syntax also supports new and different options, shown\nin the follow examples.\n\nAccessing arguments by position:\n\n   >>> \'{0}, {1}, {2}\'.format(\'a\', \'b\', \'c\')\n   \'a, b, c\'\n   >>> \'{}, {}, {}\'.format(\'a\', \'b\', \'c\')  # 2.7+ only\n   \'a, b, c\'\n   >>> \'{2}, {1}, {0}\'.format(\'a\', \'b\', \'c\')\n   \'c, b, a\'\n   >>> \'{2}, {1}, {0}\'.format(*\'abc\')      # unpacking argument sequence\n   \'c, b, a\'\n   >>> \'{0}{1}{0}\'.format(\'abra\', \'cad\')   # arguments\' indices can be repeated\n   \'abracadabra\'\n\nAccessing arguments by name:\n\n   >>> \'Coordinates: {latitude}, {longitude}\'.format(latitude=\'37.24N\', longitude=\'-115.81W\')\n   \'Coordinates: 37.24N, -115.81W\'\n   >>> coord = {\'latitude\': \'37.24N\', \'longitude\': \'-115.81W\'}\n   >>> \'Coordinates: {latitude}, {longitude}\'.format(**coord)\n   \'Coordinates: 37.24N, -115.81W\'\n\nAccessing arguments\' attributes:\n\n   >>> c = 3-5j\n   >>> (\'The complex number {0} is formed from the real part {0.real} \'\n   ...  \'and the imaginary part {0.imag}.\').format(c)\n   \'The complex number (3-5j) is formed from the real part 3.0 and the imaginary part -5.0.\'\n   >>> class Point(object):\n   ...     def __init__(self, x, y):\n   ...         self.x, self.y = x, y\n   ...     def __str__(self):\n   ...         return \'Point({self.x}, {self.y})\'.format(self=self)\n   ...\n   >>> str(Point(4, 2))\n   \'Point(4, 2)\'\n\nAccessing arguments\' items:\n\n   >>> coord = (3, 5)\n   >>> \'X: {0[0]};  Y: {0[1]}\'.format(coord)\n   \'X: 3;  Y: 5\'\n\nReplacing ``%s`` and ``%r``:\n\n   >>> "repr() shows quotes: {!r}; str() doesn\'t: {!s}".format(\'test1\', \'test2\')\n   "repr() shows quotes: \'test1\'; str() doesn\'t: test2"\n\nAligning the text and specifying a width:\n\n   >>> \'{:<30}\'.format(\'left aligned\')\n   \'left aligned                  \'\n   >>> \'{:>30}\'.format(\'right aligned\')\n   \'                 right aligned\'\n   >>> \'{:^30}\'.format(\'centered\')\n   \'           centered           \'\n   >>> \'{:*^30}\'.format(\'centered\')  # use \'*\' as a fill char\n   \'***********centered***********\'\n\nReplacing ``%+f``, ``%-f``, and ``% f`` and specifying a sign:\n\n   >>> \'{:+f}; {:+f}\'.format(3.14, -3.14)  # show it always\n   \'+3.140000; -3.140000\'\n   >>> \'{: f}; {: f}\'.format(3.14, -3.14)  # show a space for positive numbers\n   \' 3.140000; -3.140000\'\n   >>> \'{:-f}; {:-f}\'.format(3.14, -3.14)  # show only the minus -- same as \'{:f}; {:f}\'\n   \'3.140000; -3.140000\'\n\nReplacing ``%x`` and ``%o`` and converting the value to different\nbases:\n\n   >>> # format also supports binary numbers\n   >>> "int: {0:d};  hex: {0:x};  oct: {0:o};  bin: {0:b}".format(42)\n   \'int: 42;  hex: 2a;  oct: 52;  bin: 101010\'\n   >>> # with 0x, 0o, or 0b as prefix:\n   >>> "int: {0:d};  hex: {0:#x};  oct: {0:#o};  bin: {0:#b}".format(42)\n   \'int: 42;  hex: 0x2a;  oct: 0o52;  bin: 0b101010\'\n\nUsing the comma as a thousands separator:\n\n   >>> \'{:,}\'.format(1234567890)\n   \'1,234,567,890\'\n\nExpressing a percentage:\n\n   >>> points = 19.5\n   >>> total = 22\n   >>> \'Correct answers: {:.2%}.\'.format(points/total)\n   \'Correct answers: 88.64%\'\n\nUsing type-specific formatting:\n\n   >>> import datetime\n   >>> d = datetime.datetime(2010, 7, 4, 12, 15, 58)\n   >>> \'{:%Y-%m-%d %H:%M:%S}\'.format(d)\n   \'2010-07-04 12:15:58\'\n\nNesting arguments and more complex examples:\n\n   >>> for align, text in zip(\'<^>\', [\'left\', \'center\', \'right\']):\n   ...     \'{0:{align}{fill}16}\'.format(text, fill=align, align=align)\n   ...\n   \'left<<<<<<<<<<<<\'\n   \'^^^^^center^^^^^\'\n   \'>>>>>>>>>>>right\'\n   >>>\n   >>> octets = [192, 168, 0, 1]\n   >>> \'{:02X}{:02X}{:02X}{:02X}\'.format(*octets)\n   \'C0A80001\'\n   >>> int(_, 16)\n   3232235521\n   >>>\n   >>> width = 5\n   >>> for num in range(5,12):\n   ...     for base in \'dXob\':\n   ...         print \'{0:{width}{base}}\'.format(num, base=base, width=width),\n   ...     print\n   ...\n       5     5     5   101\n       6     6     6   110\n       7     7     7   111\n       8     8    10  1000\n       9     9    11  1001\n      10     A    12  1010\n      11     B    13  1011\n',
  'function': u'\nFunction definitions\n********************\n\nA function definition defines a user-defined function object (see\nsection *The standard type hierarchy*):\n\n   decorated      ::= decorators (classdef | funcdef)\n   decorators     ::= decorator+\n   decorator      ::= "@" dotted_name ["(" [argument_list [","]] ")"] NEWLINE\n   funcdef        ::= "def" funcname "(" [parameter_list] ")" ":" suite\n   dotted_name    ::= identifier ("." identifier)*\n   parameter_list ::= (defparameter ",")*\n                      (  "*" identifier [, "**" identifier]\n                      | "**" identifier\n                      | defparameter [","] )\n   defparameter   ::= parameter ["=" expression]\n   sublist        ::= parameter ("," parameter)* [","]\n   parameter      ::= identifier | "(" sublist ")"\n   funcname       ::= identifier\n\nA function definition is an executable statement.  Its execution binds\nthe function name in the current local namespace to a function object\n(a wrapper around the executable code for the function).  This\nfunction object contains a reference to the current global namespace\nas the global namespace to be used when the function is called.\n\nThe function definition does not execute the function body; this gets\nexecuted only when the function is called. [3]\n\nA function definition may be wrapped by one or more *decorator*\nexpressions. Decorator expressions are evaluated when the function is\ndefined, in the scope that contains the function definition.  The\nresult must be a callable, which is invoked with the function object\nas the only argument. The returned value is bound to the function name\ninstead of the function object.  Multiple decorators are applied in\nnested fashion. For example, the following code:\n\n   @f1(arg)\n   @f2\n   def func(): pass\n\nis equivalent to:\n\n   def func(): pass\n   func = f1(arg)(f2(func))\n\nWhen one or more top-level parameters have the form *parameter* ``=``\n*expression*, the function is said to have "default parameter values."\nFor a parameter with a default value, the corresponding argument may\nbe omitted from a call, in which case the parameter\'s default value is\nsubstituted.  If a parameter has a default value, all following\nparameters must also have a default value --- this is a syntactic\nrestriction that is not expressed by the grammar.\n\n**Default parameter values are evaluated when the function definition\nis executed.**  This means that the expression is evaluated once, when\nthe function is defined, and that that same "pre-computed" value is\nused for each call.  This is especially important to understand when a\ndefault parameter is a mutable object, such as a list or a dictionary:\nif the function modifies the object (e.g. by appending an item to a\nlist), the default value is in effect modified. This is generally not\nwhat was intended.  A way around this  is to use ``None`` as the\ndefault, and explicitly test for it in the body of the function, e.g.:\n\n   def whats_on_the_telly(penguin=None):\n       if penguin is None:\n           penguin = []\n       penguin.append("property of the zoo")\n       return penguin\n\nFunction call semantics are described in more detail in section\n*Calls*. A function call always assigns values to all parameters\nmentioned in the parameter list, either from position arguments, from\nkeyword arguments, or from default values.  If the form\n"``*identifier``" is present, it is initialized to a tuple receiving\nany excess positional parameters, defaulting to the empty tuple.  If\nthe form "``**identifier``" is present, it is initialized to a new\ndictionary receiving any excess keyword arguments, defaulting to a new\nempty dictionary.\n\nIt is also possible to create anonymous functions (functions not bound\nto a name), for immediate use in expressions.  This uses lambda forms,\ndescribed in section *Lambdas*.  Note that the lambda form is merely a\nshorthand for a simplified function definition; a function defined in\na "``def``" statement can be passed around or assigned to another name\njust like a function defined by a lambda form.  The "``def``" form is\nactually more powerful since it allows the execution of multiple\nstatements.\n\n**Programmer\'s note:** Functions are first-class objects.  A "``def``"\nform executed inside a function definition defines a local function\nthat can be returned or passed around.  Free variables used in the\nnested function can access the local variables of the function\ncontaining the def.  See section *Naming and binding* for details.\n',
  'global': u'\nThe ``global`` statement\n************************\n\n   global_stmt ::= "global" identifier ("," identifier)*\n\nThe ``global`` statement is a declaration which holds for the entire\ncurrent code block.  It means that the listed identifiers are to be\ninterpreted as globals.  It would be impossible to assign to a global\nvariable without ``global``, although free variables may refer to\nglobals without being declared global.\n\nNames listed in a ``global`` statement must not be used in the same\ncode block textually preceding that ``global`` statement.\n\nNames listed in a ``global`` statement must not be defined as formal\nparameters or in a ``for`` loop control target, ``class`` definition,\nfunction definition, or ``import`` statement.\n\n**CPython implementation detail:** The current implementation does not\nenforce the latter two restrictions, but programs should not abuse\nthis freedom, as future implementations may enforce them or silently\nchange the meaning of the program.\n\n**Programmer\'s note:** the ``global`` is a directive to the parser.\nIt applies only to code parsed at the same time as the ``global``\nstatement. In particular, a ``global`` statement contained in an\n``exec`` statement does not affect the code block *containing* the\n``exec`` statement, and code contained in an ``exec`` statement is\nunaffected by ``global`` statements in the code containing the\n``exec`` statement.  The same applies to the ``eval()``,\n``execfile()`` and ``compile()`` functions.\n',
  'id-classes': u'\nReserved classes of identifiers\n*******************************\n\nCertain classes of identifiers (besides keywords) have special\nmeanings.  These classes are identified by the patterns of leading and\ntrailing underscore characters:\n\n``_*``\n   Not imported by ``from module import *``.  The special identifier\n   ``_`` is used in the interactive interpreter to store the result of\n   the last evaluation; it is stored in the ``__builtin__`` module.\n   When not in interactive mode, ``_`` has no special meaning and is\n   not defined. See section *The import statement*.\n\n   Note: The name ``_`` is often used in conjunction with\n     internationalization; refer to the documentation for the\n     ``gettext`` module for more information on this convention.\n\n``__*__``\n   System-defined names.  These names are defined by the interpreter\n   and its implementation (including the standard library);\n   applications should not expect to define additional names using\n   this convention.  The set of names of this class defined by Python\n   may be extended in future versions. See section *Special method\n   names*.\n\n``__*``\n   Class-private names.  Names in this category, when used within the\n   context of a class definition, are re-written to use a mangled form\n   to help avoid name clashes between "private" attributes of base and\n   derived classes. See section *Identifiers (Names)*.\n',

From python-checkins at python.org  Sat Jul  3 15:56:13 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sat,  3 Jul 2010 15:56:13 +0200 (CEST)
Subject: [Python-checkins] r82500 - in python/trunk: Include/patchlevel.h
	Lib/distutils/__init__.py Lib/idlelib/idlever.py Misc/NEWS
	Misc/RPM/python-2.7.spec README
Message-ID: <20100703135613.CAE51F2EA@mail.python.org>

Author: benjamin.peterson
Date: Sat Jul  3 15:56:13 2010
New Revision: 82500

Log:
update to 2.7 final

Modified:
   python/trunk/Include/patchlevel.h
   python/trunk/Lib/distutils/__init__.py
   python/trunk/Lib/idlelib/idlever.py
   python/trunk/Misc/NEWS
   python/trunk/Misc/RPM/python-2.7.spec
   python/trunk/README

Modified: python/trunk/Include/patchlevel.h
==============================================================================
--- python/trunk/Include/patchlevel.h	(original)
+++ python/trunk/Include/patchlevel.h	Sat Jul  3 15:56:13 2010
@@ -23,11 +23,11 @@
 #define PY_MAJOR_VERSION	2
 #define PY_MINOR_VERSION	7
 #define PY_MICRO_VERSION	0
-#define PY_RELEASE_LEVEL	PY_RELEASE_LEVEL_GAMMA
-#define PY_RELEASE_SERIAL	2
+#define PY_RELEASE_LEVEL	PY_RELEASE_LEVEL_FINAL
+#define PY_RELEASE_SERIAL	0
 
 /* Version as a string */
-#define PY_VERSION      	"2.7rc2+"
+#define PY_VERSION      	"2.7"
 /*--end constants--*/
 
 /* Subversion Revision number of this file (not of the repository) */

Modified: python/trunk/Lib/distutils/__init__.py
==============================================================================
--- python/trunk/Lib/distutils/__init__.py	(original)
+++ python/trunk/Lib/distutils/__init__.py	Sat Jul  3 15:56:13 2010
@@ -15,5 +15,5 @@
 # Updated automatically by the Python release process.
 #
 #--start constants--
-__version__ = "2.7rc2"
+__version__ = "2.7"
 #--end constants--

Modified: python/trunk/Lib/idlelib/idlever.py
==============================================================================
--- python/trunk/Lib/idlelib/idlever.py	(original)
+++ python/trunk/Lib/idlelib/idlever.py	Sat Jul  3 15:56:13 2010
@@ -1 +1 @@
-IDLE_VERSION = "2.7rc2"
+IDLE_VERSION = "2.7"

Modified: python/trunk/Misc/NEWS
==============================================================================
--- python/trunk/Misc/NEWS	(original)
+++ python/trunk/Misc/NEWS	Sat Jul  3 15:56:13 2010
@@ -7,7 +7,7 @@
 What's New in Python 2.7?
 =========================
 
-*Release date: XXXX-XX-XX*
+*Release date: 2010-07-03*
 
 Core and Builtins
 -----------------

Modified: python/trunk/Misc/RPM/python-2.7.spec
==============================================================================
--- python/trunk/Misc/RPM/python-2.7.spec	(original)
+++ python/trunk/Misc/RPM/python-2.7.spec	Sat Jul  3 15:56:13 2010
@@ -39,7 +39,7 @@
 
 %define name python
 #--start constants--
-%define version 2.7rc2
+%define version 2.7
 %define libver 2.7
 #--end constants--
 %define release 1pydotorg

Modified: python/trunk/README
==============================================================================
--- python/trunk/README	(original)
+++ python/trunk/README	Sat Jul  3 15:56:13 2010
@@ -1,5 +1,5 @@
-This is Python version 2.7 release candidate 2
-==============================================
+This is Python version 2.7
+===========================
 
 Copyright (c) 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010
 Python Software Foundation.

From python-checkins at python.org  Sat Jul  3 15:56:57 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sat,  3 Jul 2010 15:56:57 +0200 (CEST)
Subject: [Python-checkins] r82501 - python/trunk/Misc/NEWS
Message-ID: <20100703135657.0229DF2EA@mail.python.org>

Author: benjamin.peterson
Date: Sat Jul  3 15:56:56 2010
New Revision: 82501

Log:
add another line between sections

Modified:
   python/trunk/Misc/NEWS

Modified: python/trunk/Misc/NEWS
==============================================================================
--- python/trunk/Misc/NEWS	(original)
+++ python/trunk/Misc/NEWS	Sat Jul  3 15:56:56 2010
@@ -25,6 +25,7 @@
 
 - Issue #9125: Add recognition of 'except ... as ...' syntax to parser module.
 
+
 What's New in Python 2.7 release candidate 2?
 =============================================
 

From python-checkins at python.org  Sat Jul  3 15:57:30 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sat,  3 Jul 2010 15:57:30 +0200 (CEST)
Subject: [Python-checkins] r82502 - python/trunk/Misc/NEWS
Message-ID: <20100703135730.727D9F2EA@mail.python.org>

Author: benjamin.peterson
Date: Sat Jul  3 15:57:30 2010
New Revision: 82502

Log:
move sections

Modified:
   python/trunk/Misc/NEWS

Modified: python/trunk/Misc/NEWS
==============================================================================
--- python/trunk/Misc/NEWS	(original)
+++ python/trunk/Misc/NEWS	Sat Jul  3 15:57:30 2010
@@ -17,14 +17,17 @@
 Library
 -------
 
+- Issue #9125: Add recognition of 'except ... as ...' syntax to parser module.
+
+Extension Modules
+-----------------
+
 - Issue #7673: Fix security vulnerability (CVE-2010-2089) in the audioop module,
   ensure that the input string length is a multiple of the frame size.
 
 - Issue #9075: In the ssl module, remove the setting of a ``debug`` flag
   on an OpenSSL structure.
 
-- Issue #9125: Add recognition of 'except ... as ...' syntax to parser module.
-
 
 What's New in Python 2.7 release candidate 2?
 =============================================

From python-checkins at python.org  Sat Jul  3 16:00:41 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sat,  3 Jul 2010 16:00:41 +0200 (CEST)
Subject: [Python-checkins] r82503 - python/branches/release27-maint
Message-ID: <20100703140041.C4855EE982@mail.python.org>

Author: benjamin.peterson
Date: Sat Jul  3 16:00:41 2010
New Revision: 82503

Log:
make 2.7 maintenance branch


Added:
   python/branches/release27-maint/
      - copied from r82502, /python/trunk/

From python-checkins at python.org  Sat Jul  3 16:02:14 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sat,  3 Jul 2010 16:02:14 +0200 (CEST)
Subject: [Python-checkins] r82504 - python/tags/r27
Message-ID: <20100703140214.85959EE982@mail.python.org>

Author: benjamin.peterson
Date: Sat Jul  3 16:02:14 2010
New Revision: 82504

Log:
Tag Python 2.7 final!


Added:
   python/tags/r27/
      - copied from r82503, /python/trunk/

From python-checkins at python.org  Sat Jul  3 16:19:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 03 Jul 2010 16:19:06 +0200
Subject: [Python-checkins] distutils2: Added default
	implementation/documentation for the	external API
Message-ID: <E1OV3Yg-0002JW-Aa@dinsdale.python.org>

tarek.ziade pushed 8fb85c777536 to distutils2:

http://hg.python.org/distutils2/rev/8fb85c777536
changeset:   231:8fb85c777536
parent:      188:b7fd258b33e7
user:        Jeremy Kloth <jeremy.kloth at gmail.com>
date:        Thu Jun 03 15:42:12 2010 -0600
summary:     Added default implementation/documentation for the external API
files:       src/distutils2/cmd.py

diff --git a/src/distutils2/cmd.py b/src/distutils2/cmd.py
--- a/src/distutils2/cmd.py
+++ b/src/distutils2/cmd.py
@@ -189,6 +189,31 @@
         """
         log.log(level, msg)
 
+    # -- External interface --------------------------------------------
+    # (called by outsiders)
+
+    def get_source_files(self):
+        """Return the list of files that are used as inputs to this command,
+        i.e. the files used to generate the output files.  The result is used
+        by the `sdist` command in determining the set of default files.
+
+        Command classes should implement this method if they operate on files
+        from the source tree.
+        """
+        return []
+
+    def get_outputs(self):
+        """Return the list of files that would be produced if this command
+        were actually run.  Not affected by the "dry-run" flag or whether
+        any other commands have been run.
+
+        Command classes should implement this method if they produce any
+        output files that get consumed by another command.  e.g., `build_ext`
+        returns the list of built extension modules, but not any temporary
+        files used in the compilation process.
+        """
+        return []
+
     # -- Option validation methods -------------------------------------
     # (these are very handy in writing the 'finalize_options()' method)
     #

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul  3 16:19:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 03 Jul 2010 16:19:06 +0200
Subject: [Python-checkins] distutils2: Don?t use deprecated name (see
	comment in unittest2.case)
Message-ID: <E1OV3Yg-0002JK-9m@dinsdale.python.org>

tarek.ziade pushed 0b1b57c04665 to distutils2:

http://hg.python.org/distutils2/rev/0b1b57c04665
changeset:   230:0b1b57c04665
user:        ?ric Araujo <merwok at netwok.org>
date:        Thu Jun 24 16:36:34 2010 +0200
summary:     Don?t use deprecated name (see comment in unittest2.case)
files:       src/distutils2/_backport/tests/test_sysconfig.py, src/distutils2/tests/test_bdist.py, src/distutils2/tests/test_bdist_dumb.py, src/distutils2/tests/test_build.py, src/distutils2/tests/test_build_clib.py, src/distutils2/tests/test_build_ext.py, src/distutils2/tests/test_ccompiler.py, src/distutils2/tests/test_check.py, src/distutils2/tests/test_cmd.py, src/distutils2/tests/test_config.py, src/distutils2/tests/test_config_cmd.py, src/distutils2/tests/test_converter.py, src/distutils2/tests/test_cygwinccompiler.py, src/distutils2/tests/test_dist.py, src/distutils2/tests/test_install.py, src/distutils2/tests/test_install_data.py, src/distutils2/tests/test_install_headers.py, src/distutils2/tests/test_install_lib.py, src/distutils2/tests/test_manifest.py, src/distutils2/tests/test_metadata.py, src/distutils2/tests/test_msvc9compiler.py, src/distutils2/tests/test_register.py, src/distutils2/tests/test_sdist.py, src/distutils2/tests/test_spawn.py, src/distutils2/tests/test_upload.py, src/distutils2/tests/test_util.py, src/distutils2/tests/test_version.py

diff --git a/src/distutils2/_backport/tests/test_sysconfig.py b/src/distutils2/_backport/tests/test_sysconfig.py
--- a/src/distutils2/_backport/tests/test_sysconfig.py
+++ b/src/distutils2/_backport/tests/test_sysconfig.py
@@ -107,7 +107,7 @@
         wanted.sort()
         scheme = scheme.items()
         scheme.sort()
-        self.assertEquals(scheme, wanted)
+        self.assertEqual(scheme, wanted)
 
     def test_get_config_vars(self):
         cvars = get_config_vars()
@@ -120,21 +120,21 @@
         sys.version = ('2.4.4 (#71, Oct 18 2006, 08:34:43) '
                        '[MSC v.1310 32 bit (Intel)]')
         sys.platform = 'win32'
-        self.assertEquals(get_platform(), 'win32')
+        self.assertEqual(get_platform(), 'win32')
 
         # windows XP, amd64
         os.name = 'nt'
         sys.version = ('2.4.4 (#71, Oct 18 2006, 08:34:43) '
                        '[MSC v.1310 32 bit (Amd64)]')
         sys.platform = 'win32'
-        self.assertEquals(get_platform(), 'win-amd64')
+        self.assertEqual(get_platform(), 'win-amd64')
 
         # windows XP, itanium
         os.name = 'nt'
         sys.version = ('2.4.4 (#71, Oct 18 2006, 08:34:43) '
                        '[MSC v.1310 32 bit (Itanium)]')
         sys.platform = 'win32'
-        self.assertEquals(get_platform(), 'win-ia64')
+        self.assertEqual(get_platform(), 'win-ia64')
 
         # macbook
         os.name = 'posix'
@@ -153,9 +153,9 @@
         maxint = sys.maxint
         try:
             sys.maxint = 2147483647
-            self.assertEquals(get_platform(), 'macosx-10.3-ppc')
+            self.assertEqual(get_platform(), 'macosx-10.3-ppc')
             sys.maxint = 9223372036854775807
-            self.assertEquals(get_platform(), 'macosx-10.3-ppc64')
+            self.assertEqual(get_platform(), 'macosx-10.3-ppc64')
         finally:
             sys.maxint = maxint
 
@@ -173,9 +173,9 @@
         maxint = sys.maxint
         try:
             sys.maxint = 2147483647
-            self.assertEquals(get_platform(), 'macosx-10.3-i386')
+            self.assertEqual(get_platform(), 'macosx-10.3-i386')
             sys.maxint = 9223372036854775807
-            self.assertEquals(get_platform(), 'macosx-10.3-x86_64')
+            self.assertEqual(get_platform(), 'macosx-10.3-x86_64')
         finally:
             sys.maxint = maxint
 
@@ -186,33 +186,33 @@
                                        '-fno-strict-aliasing -fno-common '
                                        '-dynamic -DNDEBUG -g -O3')
 
-        self.assertEquals(get_platform(), 'macosx-10.4-fat')
+        self.assertEqual(get_platform(), 'macosx-10.4-fat')
 
         get_config_vars()['CFLAGS'] = ('-arch x86_64 -arch i386 -isysroot '
                                        '/Developer/SDKs/MacOSX10.4u.sdk  '
                                        '-fno-strict-aliasing -fno-common '
                                        '-dynamic -DNDEBUG -g -O3')
 
-        self.assertEquals(get_platform(), 'macosx-10.4-intel')
+        self.assertEqual(get_platform(), 'macosx-10.4-intel')
 
         get_config_vars()['CFLAGS'] = ('-arch x86_64 -arch ppc -arch i386 -isysroot '
                                        '/Developer/SDKs/MacOSX10.4u.sdk  '
                                        '-fno-strict-aliasing -fno-common '
                                        '-dynamic -DNDEBUG -g -O3')
-        self.assertEquals(get_platform(), 'macosx-10.4-fat3')
+        self.assertEqual(get_platform(), 'macosx-10.4-fat3')
 
         get_config_vars()['CFLAGS'] = ('-arch ppc64 -arch x86_64 -arch ppc -arch i386 -isysroot '
                                        '/Developer/SDKs/MacOSX10.4u.sdk  '
                                        '-fno-strict-aliasing -fno-common '
                                        '-dynamic -DNDEBUG -g -O3')
-        self.assertEquals(get_platform(), 'macosx-10.4-universal')
+        self.assertEqual(get_platform(), 'macosx-10.4-universal')
 
         get_config_vars()['CFLAGS'] = ('-arch x86_64 -arch ppc64 -isysroot '
                                        '/Developer/SDKs/MacOSX10.4u.sdk  '
                                        '-fno-strict-aliasing -fno-common '
                                        '-dynamic -DNDEBUG -g -O3')
 
-        self.assertEquals(get_platform(), 'macosx-10.4-fat64')
+        self.assertEqual(get_platform(), 'macosx-10.4-fat64')
 
         for arch in ('ppc', 'i386', 'x86_64', 'ppc64'):
             get_config_vars()['CFLAGS'] = ('-arch %s -isysroot '
@@ -220,7 +220,7 @@
                                            '-fno-strict-aliasing -fno-common '
                                            '-dynamic -DNDEBUG -g -O3'%(arch,))
 
-            self.assertEquals(get_platform(), 'macosx-10.4-%s'%(arch,))
+            self.assertEqual(get_platform(), 'macosx-10.4-%s'%(arch,))
 
         # linux debian sarge
         os.name = 'posix'
@@ -230,7 +230,7 @@
         self._set_uname(('Linux', 'aglae', '2.6.21.1dedibox-r7',
                     '#1 Mon Apr 30 17:25:38 CEST 2007', 'i686'))
 
-        self.assertEquals(get_platform(), 'linux-i686')
+        self.assertEqual(get_platform(), 'linux-i686')
 
         # XXX more platforms to tests here
 
@@ -241,7 +241,7 @@
     def test_get_scheme_names(self):
         wanted = ('nt', 'nt_user', 'os2', 'os2_home', 'posix_home',
                   'posix_prefix', 'posix_user')
-        self.assertEquals(get_scheme_names(), wanted)
+        self.assertEqual(get_scheme_names(), wanted)
 
     def test_expand_globals(self):
 
@@ -254,8 +254,8 @@
 
         _expand_globals(config)
 
-        self.assertEquals(config.get('posix', 'foo'), 'ok')
-        self.assertEquals(config.get('posix', 'more'), '/etc/ok')
+        self.assertEqual(config.get('posix', 'foo'), 'ok')
+        self.assertEqual(config.get('posix', 'more'), '/etc/ok')
 
         # we might not have globals after all
         # extending again (==no more globals section)
diff --git a/src/distutils2/tests/test_bdist.py b/src/distutils2/tests/test_bdist.py
--- a/src/distutils2/tests/test_bdist.py
+++ b/src/distutils2/tests/test_bdist.py
@@ -23,7 +23,7 @@
         cmd = bdist(dist)
         cmd.formats = ['msi']
         cmd.ensure_finalized()
-        self.assertEquals(cmd.formats, ['msi'])
+        self.assertEqual(cmd.formats, ['msi'])
 
         # what format bdist offers ?
         # XXX an explicit list in bdist is
@@ -33,7 +33,7 @@
         formats.sort()
         found = cmd.format_command.keys()
         found.sort()
-        self.assertEquals(found, formats)
+        self.assertEqual(found, formats)
 
 def test_suite():
     return unittest.makeSuite(BuildTestCase)
diff --git a/src/distutils2/tests/test_bdist_dumb.py b/src/distutils2/tests/test_bdist_dumb.py
--- a/src/distutils2/tests/test_bdist_dumb.py
+++ b/src/distutils2/tests/test_bdist_dumb.py
@@ -78,7 +78,7 @@
             base = base.replace(':', '-')
 
         wanted = ['%s.zip' % base]
-        self.assertEquals(dist_created, wanted)
+        self.assertEqual(dist_created, wanted)
 
         # now let's check what we have in the zip file
         # XXX to be done
@@ -87,16 +87,16 @@
         pkg_dir, dist = self.create_dist()
         os.chdir(pkg_dir)
         cmd = bdist_dumb(dist)
-        self.assertEquals(cmd.bdist_dir, None)
+        self.assertEqual(cmd.bdist_dir, None)
         cmd.finalize_options()
 
         # bdist_dir is initialized to bdist_base/dumb if not set
         base = cmd.get_finalized_command('bdist').bdist_base
-        self.assertEquals(cmd.bdist_dir, os.path.join(base, 'dumb'))
+        self.assertEqual(cmd.bdist_dir, os.path.join(base, 'dumb'))
 
         # the format is set to a default value depending on the os.name
         default = cmd.default_format[os.name]
-        self.assertEquals(cmd.format, default)
+        self.assertEqual(cmd.format, default)
 
 def test_suite():
     return unittest.makeSuite(BuildDumbTestCase)
diff --git a/src/distutils2/tests/test_build.py b/src/distutils2/tests/test_build.py
--- a/src/distutils2/tests/test_build.py
+++ b/src/distutils2/tests/test_build.py
@@ -20,11 +20,11 @@
         cmd.finalize_options()
 
         # if not specified, plat_name gets the current platform
-        self.assertEquals(cmd.plat_name, get_platform())
+        self.assertEqual(cmd.plat_name, get_platform())
 
         # build_purelib is build + lib
         wanted = os.path.join(cmd.build_base, 'lib')
-        self.assertEquals(cmd.build_purelib, wanted)
+        self.assertEqual(cmd.build_purelib, wanted)
 
         # build_platlib is 'build/lib.platform-x.x[-pydebug]'
         # examples:
@@ -34,21 +34,21 @@
             self.assertTrue(cmd.build_platlib.endswith('-pydebug'))
             plat_spec += '-pydebug'
         wanted = os.path.join(cmd.build_base, 'lib' + plat_spec)
-        self.assertEquals(cmd.build_platlib, wanted)
+        self.assertEqual(cmd.build_platlib, wanted)
 
         # by default, build_lib = build_purelib
-        self.assertEquals(cmd.build_lib, cmd.build_purelib)
+        self.assertEqual(cmd.build_lib, cmd.build_purelib)
 
         # build_temp is build/temp.<plat>
         wanted = os.path.join(cmd.build_base, 'temp' + plat_spec)
-        self.assertEquals(cmd.build_temp, wanted)
+        self.assertEqual(cmd.build_temp, wanted)
 
         # build_scripts is build/scripts-x.x
         wanted = os.path.join(cmd.build_base, 'scripts-' +  sys.version[0:3])
-        self.assertEquals(cmd.build_scripts, wanted)
+        self.assertEqual(cmd.build_scripts, wanted)
 
         # executable is os.path.normpath(sys.executable)
-        self.assertEquals(cmd.executable, os.path.normpath(sys.executable))
+        self.assertEqual(cmd.executable, os.path.normpath(sys.executable))
 
 def test_suite():
     return unittest.makeSuite(BuildTestCase)
diff --git a/src/distutils2/tests/test_build_clib.py b/src/distutils2/tests/test_build_clib.py
--- a/src/distutils2/tests/test_build_clib.py
+++ b/src/distutils2/tests/test_build_clib.py
@@ -55,14 +55,14 @@
         self.assertRaises(DistutilsSetupError, cmd.get_source_files)
 
         cmd.libraries = [('name', {'sources': ['a', 'b']})]
-        self.assertEquals(cmd.get_source_files(), ['a', 'b'])
+        self.assertEqual(cmd.get_source_files(), ['a', 'b'])
 
         cmd.libraries = [('name', {'sources': ('a', 'b')})]
-        self.assertEquals(cmd.get_source_files(), ['a', 'b'])
+        self.assertEqual(cmd.get_source_files(), ['a', 'b'])
 
         cmd.libraries = [('name', {'sources': ('a', 'b')}),
                          ('name2', {'sources': ['c', 'd']})]
-        self.assertEquals(cmd.get_source_files(), ['a', 'b', 'c', 'd'])
+        self.assertEqual(cmd.get_source_files(), ['a', 'b', 'c', 'd'])
 
     def test_build_libraries(self):
 
@@ -91,11 +91,11 @@
 
         cmd.include_dirs = 'one-dir'
         cmd.finalize_options()
-        self.assertEquals(cmd.include_dirs, ['one-dir'])
+        self.assertEqual(cmd.include_dirs, ['one-dir'])
 
         cmd.include_dirs = None
         cmd.finalize_options()
-        self.assertEquals(cmd.include_dirs, [])
+        self.assertEqual(cmd.include_dirs, [])
 
         cmd.distribution.libraries = 'WONTWORK'
         self.assertRaises(DistutilsSetupError, cmd.finalize_options)
diff --git a/src/distutils2/tests/test_build_ext.py b/src/distutils2/tests/test_build_ext.py
--- a/src/distutils2/tests/test_build_ext.py
+++ b/src/distutils2/tests/test_build_ext.py
@@ -80,11 +80,11 @@
         for attr in ('error', 'foo', 'new', 'roj'):
             self.assertTrue(hasattr(xx, attr))
 
-        self.assertEquals(xx.foo(2, 5), 7)
-        self.assertEquals(xx.foo(13,15), 28)
-        self.assertEquals(xx.new().demo(), None)
+        self.assertEqual(xx.foo(2, 5), 7)
+        self.assertEqual(xx.foo(13,15), 28)
+        self.assertEqual(xx.new().demo(), None)
         doc = 'This is a template module just for instruction.'
-        self.assertEquals(xx.__doc__, doc)
+        self.assertEqual(xx.__doc__, doc)
         self.assertTrue(isinstance(xx.Null(), xx.Null))
         self.assertTrue(isinstance(xx.Str(), xx.Str))
 
@@ -194,7 +194,7 @@
         cmd = build_ext(dist)
         cmd.libraries = 'my_lib'
         cmd.finalize_options()
-        self.assertEquals(cmd.libraries, ['my_lib'])
+        self.assertEqual(cmd.libraries, ['my_lib'])
 
         # make sure cmd.library_dirs is turned into a list
         # if it's a string
@@ -208,7 +208,7 @@
         cmd = build_ext(dist)
         cmd.rpath = os.pathsep.join(['one', 'two'])
         cmd.finalize_options()
-        self.assertEquals(cmd.rpath, ['one', 'two'])
+        self.assertEqual(cmd.rpath, ['one', 'two'])
 
         # XXX more tests to perform for win32
 
@@ -217,25 +217,25 @@
         cmd = build_ext(dist)
         cmd.define = 'one,two'
         cmd.finalize_options()
-        self.assertEquals(cmd.define, [('one', '1'), ('two', '1')])
+        self.assertEqual(cmd.define, [('one', '1'), ('two', '1')])
 
         # make sure undef is turned into a list of
         # strings if they are ','-separated strings
         cmd = build_ext(dist)
         cmd.undef = 'one,two'
         cmd.finalize_options()
-        self.assertEquals(cmd.undef, ['one', 'two'])
+        self.assertEqual(cmd.undef, ['one', 'two'])
 
         # make sure swig_opts is turned into a list
         cmd = build_ext(dist)
         cmd.swig_opts = None
         cmd.finalize_options()
-        self.assertEquals(cmd.swig_opts, [])
+        self.assertEqual(cmd.swig_opts, [])
 
         cmd = build_ext(dist)
         cmd.swig_opts = '1 2'
         cmd.finalize_options()
-        self.assertEquals(cmd.swig_opts, ['1', '2'])
+        self.assertEqual(cmd.swig_opts, ['1', '2'])
 
     def test_check_extensions_list(self):
         dist = Distribution()
@@ -271,7 +271,7 @@
         # check_extensions_list adds in ext the values passed
         # when they are in ('include_dirs', 'library_dirs', 'libraries'
         # 'extra_objects', 'extra_compile_args', 'extra_link_args')
-        self.assertEquals(ext.libraries, 'foo')
+        self.assertEqual(ext.libraries, 'foo')
         self.assertTrue(not hasattr(ext, 'some'))
 
         # 'macros' element of build info dict must be 1- or 2-tuple
@@ -281,15 +281,15 @@
 
         exts[0][1]['macros'] = [('1', '2'), ('3',)]
         cmd.check_extensions_list(exts)
-        self.assertEquals(exts[0].undef_macros, ['3'])
-        self.assertEquals(exts[0].define_macros, [('1', '2')])
+        self.assertEqual(exts[0].undef_macros, ['3'])
+        self.assertEqual(exts[0].define_macros, [('1', '2')])
 
     def test_get_source_files(self):
         modules = [Extension('foo', ['xxx'], optional=False)]
         dist = Distribution({'name': 'xx', 'ext_modules': modules})
         cmd = build_ext(dist)
         cmd.ensure_finalized()
-        self.assertEquals(cmd.get_source_files(), ['xxx'])
+        self.assertEqual(cmd.get_source_files(), ['xxx'])
 
     def test_compiler_option(self):
         # cmd.compiler is an option and
@@ -300,7 +300,7 @@
         cmd.compiler = 'unix'
         cmd.ensure_finalized()
         cmd.run()
-        self.assertEquals(cmd.compiler, 'unix')
+        self.assertEqual(cmd.compiler, 'unix')
 
     def test_get_outputs(self):
         tmp_dir = self.mkdtemp()
@@ -311,7 +311,7 @@
                              'ext_modules': [ext]})
         cmd = build_ext(dist)
         cmd.ensure_finalized()
-        self.assertEquals(len(cmd.get_outputs()), 1)
+        self.assertEqual(len(cmd.get_outputs()), 1)
 
         if os.name == "nt":
             cmd.debug = sys.executable.endswith("_d.exe")
@@ -331,19 +331,19 @@
         finally:
             os.chdir(old_wd)
         self.assertTrue(os.path.exists(so_file))
-        self.assertEquals(os.path.splitext(so_file)[-1],
+        self.assertEqual(os.path.splitext(so_file)[-1],
                           sysconfig.get_config_var('SO'))
         so_dir = os.path.dirname(so_file)
-        self.assertEquals(so_dir, other_tmp_dir)
+        self.assertEqual(so_dir, other_tmp_dir)
 
         cmd.inplace = 0
         cmd.run()
         so_file = cmd.get_outputs()[0]
         self.assertTrue(os.path.exists(so_file))
-        self.assertEquals(os.path.splitext(so_file)[-1],
+        self.assertEqual(os.path.splitext(so_file)[-1],
                           sysconfig.get_config_var('SO'))
         so_dir = os.path.dirname(so_file)
-        self.assertEquals(so_dir, cmd.build_lib)
+        self.assertEqual(so_dir, cmd.build_lib)
 
         # inplace = 0, cmd.package = 'bar'
         build_py = cmd.get_finalized_command('build_py')
@@ -351,7 +351,7 @@
         path = cmd.get_ext_fullpath('foo')
         # checking that the last directory is the build_dir
         path = os.path.split(path)[0]
-        self.assertEquals(path, cmd.build_lib)
+        self.assertEqual(path, cmd.build_lib)
 
         # inplace = 1, cmd.package = 'bar'
         cmd.inplace = 1
@@ -365,7 +365,7 @@
         # checking that the last directory is bar
         path = os.path.split(path)[0]
         lastdir = os.path.split(path)[-1]
-        self.assertEquals(lastdir, 'bar')
+        self.assertEqual(lastdir, 'bar')
 
     def test_ext_fullpath(self):
         ext = sysconfig.get_config_vars()['SO']
@@ -381,14 +381,14 @@
         curdir = os.getcwd()
         wanted = os.path.join(curdir, 'src', 'lxml', 'etree' + ext)
         path = cmd.get_ext_fullpath('lxml.etree')
-        self.assertEquals(wanted, path)
+        self.assertEqual(wanted, path)
 
         # building lxml.etree not inplace
         cmd.inplace = 0
         cmd.build_lib = os.path.join(curdir, 'tmpdir')
         wanted = os.path.join(curdir, 'tmpdir', 'lxml', 'etree' + ext)
         path = cmd.get_ext_fullpath('lxml.etree')
-        self.assertEquals(wanted, path)
+        self.assertEqual(wanted, path)
 
         # building twisted.runner.portmap not inplace
         build_py = cmd.get_finalized_command('build_py')
@@ -397,13 +397,13 @@
         path = cmd.get_ext_fullpath('twisted.runner.portmap')
         wanted = os.path.join(curdir, 'tmpdir', 'twisted', 'runner',
                               'portmap' + ext)
-        self.assertEquals(wanted, path)
+        self.assertEqual(wanted, path)
 
         # building twisted.runner.portmap inplace
         cmd.inplace = 1
         path = cmd.get_ext_fullpath('twisted.runner.portmap')
         wanted = os.path.join(curdir, 'twisted', 'runner', 'portmap' + ext)
-        self.assertEquals(wanted, path)
+        self.assertEqual(wanted, path)
 
 def test_suite():
     src = _get_source_filename()
diff --git a/src/distutils2/tests/test_ccompiler.py b/src/distutils2/tests/test_ccompiler.py
--- a/src/distutils2/tests/test_ccompiler.py
+++ b/src/distutils2/tests/test_ccompiler.py
@@ -31,7 +31,7 @@
         opts = gen_lib_options(compiler, libdirs, runlibdirs, libs)
         wanted = ['-Llib1', '-Llib2', '-cool', '-Rrunlib1', 'found',
                   '-lname2']
-        self.assertEquals(opts, wanted)
+        self.assertEqual(opts, wanted)
 
     def test_customize_compiler(self):
 
@@ -51,7 +51,7 @@
 
         comp = compiler()
         customize_compiler(comp)
-        self.assertEquals(comp.exes['archiver'], 'my_ar -arflags')
+        self.assertEqual(comp.exes['archiver'], 'my_ar -arflags')
 
 def test_suite():
     return unittest.makeSuite(CCompilerTestCase)
diff --git a/src/distutils2/tests/test_check.py b/src/distutils2/tests/test_check.py
--- a/src/distutils2/tests/test_check.py
+++ b/src/distutils2/tests/test_check.py
@@ -37,7 +37,7 @@
                     'name': 'xxx', 'version': 'xxx'
                     }
         cmd = self._run(metadata)
-        self.assertEquals(len(cmd._warnings), 0)
+        self.assertEqual(len(cmd._warnings), 0)
 
         # now with the strict mode, we should
         # get an error if there are missing metadata
@@ -45,7 +45,7 @@
 
         # and of course, no error when all metadata are present
         cmd = self._run(metadata, strict=1)
-        self.assertEquals(len(cmd._warnings), 0)
+        self.assertEqual(len(cmd._warnings), 0)
 
     def test_check_restructuredtext(self):
         if not _HAS_DOCUTILS: # won't test without docutils
@@ -55,7 +55,7 @@
         pkg_info, dist = self.create_dist(description=broken_rest)
         cmd = check(dist)
         cmd.check_restructuredtext()
-        self.assertEquals(len(cmd._warnings), 1)
+        self.assertEqual(len(cmd._warnings), 1)
 
         # let's see if we have an error with strict=1
         metadata = {'home_page': 'xxx', 'author': 'xxx',
@@ -69,7 +69,7 @@
         # and non-broken rest
         metadata['description'] = 'title\n=====\n\ntest'
         cmd = self._run(metadata, strict=1, restructuredtext=1)
-        self.assertEquals(len(cmd._warnings), 0)
+        self.assertEqual(len(cmd._warnings), 0)
 
     def test_check_all(self):
 
diff --git a/src/distutils2/tests/test_cmd.py b/src/distutils2/tests/test_cmd.py
--- a/src/distutils2/tests/test_cmd.py
+++ b/src/distutils2/tests/test_cmd.py
@@ -43,7 +43,7 @@
 
         # making sure execute gets called properly
         def _execute(func, args, exec_msg, level):
-            self.assertEquals(exec_msg, 'generating out from in')
+            self.assertEqual(exec_msg, 'generating out from in')
         cmd.force = True
         cmd.execute = _execute
         cmd.make_file(infiles='in', outfile='out', func='func', args=())
@@ -62,7 +62,7 @@
 
         wanted = ["command options for 'MyCmd':", '  option1 = 1',
                   '  option2 = 1']
-        self.assertEquals(msgs, wanted)
+        self.assertEqual(msgs, wanted)
 
     def test_ensure_string(self):
         cmd = self.cmd
@@ -80,7 +80,7 @@
         cmd = self.cmd
         cmd.option1 = 'ok,dok'
         cmd.ensure_string_list('option1')
-        self.assertEquals(cmd.option1, ['ok', 'dok'])
+        self.assertEqual(cmd.option1, ['ok', 'dok'])
 
         cmd.option2 = ['xxx', 'www']
         cmd.ensure_string_list('option2')
diff --git a/src/distutils2/tests/test_config.py b/src/distutils2/tests/test_config.py
--- a/src/distutils2/tests/test_config.py
+++ b/src/distutils2/tests/test_config.py
@@ -89,7 +89,7 @@
         expected = [('password', 'secret'), ('realm', 'pypi'),
                     ('repository', 'http://pypi.python.org/pypi'),
                     ('server', 'server1'), ('username', 'me')]
-        self.assertEquals(config, expected)
+        self.assertEqual(config, expected)
 
         # old format
         self.write_file(self.rc, PYPIRC_OLD)
@@ -99,7 +99,7 @@
         expected = [('password', 'secret'), ('realm', 'pypi'),
                     ('repository', 'http://pypi.python.org/pypi'),
                     ('server', 'server-login'), ('username', 'tarek')]
-        self.assertEquals(config, expected)
+        self.assertEqual(config, expected)
 
     def test_server_empty_registration(self):
         cmd = self._cmd(self.dist)
@@ -108,7 +108,7 @@
         cmd._store_pypirc('tarek', 'xxx')
         self.assertTrue(os.path.exists(rc))
         content = open(rc).read()
-        self.assertEquals(content, WANTED)
+        self.assertEqual(content, WANTED)
 
 def test_suite():
     return unittest.makeSuite(PyPIRCCommandTestCase)
diff --git a/src/distutils2/tests/test_config_cmd.py b/src/distutils2/tests/test_config_cmd.py
--- a/src/distutils2/tests/test_config_cmd.py
+++ b/src/distutils2/tests/test_config_cmd.py
@@ -34,7 +34,7 @@
             f.close()
 
         dump_file(this_file, 'I am the header')
-        self.assertEquals(len(self._logs), numlines+1)
+        self.assertEqual(len(self._logs), numlines+1)
 
     def test_search_cpp(self):
         if sys.platform == 'win32':
@@ -44,10 +44,10 @@
 
         # simple pattern searches
         match = cmd.search_cpp(pattern='xxx', body='// xxx')
-        self.assertEquals(match, 0)
+        self.assertEqual(match, 0)
 
         match = cmd.search_cpp(pattern='_configtest', body='// xxx')
-        self.assertEquals(match, 1)
+        self.assertEqual(match, 1)
 
     def test_finalize_options(self):
         # finalize_options does a bit of transformation
@@ -59,9 +59,9 @@
         cmd.library_dirs = 'three%sfour' % os.pathsep
         cmd.ensure_finalized()
 
-        self.assertEquals(cmd.include_dirs, ['one', 'two'])
-        self.assertEquals(cmd.libraries, ['one'])
-        self.assertEquals(cmd.library_dirs, ['three', 'four'])
+        self.assertEqual(cmd.include_dirs, ['one', 'two'])
+        self.assertEqual(cmd.libraries, ['one'])
+        self.assertEqual(cmd.library_dirs, ['three', 'four'])
 
     def test_clean(self):
         # _clean removes files
diff --git a/src/distutils2/tests/test_converter.py b/src/distutils2/tests/test_converter.py
--- a/src/distutils2/tests/test_converter.py
+++ b/src/distutils2/tests/test_converter.py
@@ -30,7 +30,7 @@
             wanted = file_.replace('before', 'after')
             wanted = _read_file(os.path.join(convdir, wanted))
             res = ref.refactor_string(original, 'setup.py')
-            self.assertEquals(str(res), wanted)
+            self.assertEqual(str(res), wanted)
 
 def test_suite():
     return unittest.makeSuite(ConverterTestCase)
diff --git a/src/distutils2/tests/test_cygwinccompiler.py b/src/distutils2/tests/test_cygwinccompiler.py
--- a/src/distutils2/tests/test_cygwinccompiler.py
+++ b/src/distutils2/tests/test_cygwinccompiler.py
@@ -44,48 +44,48 @@
         sys.version = ('2.6.1 (r261:67515, Dec  6 2008, 16:42:21) \n[GCC '
                        '4.0.1 (Apple Computer, Inc. build 5370)]')
 
-        self.assertEquals(check_config_h()[0], CONFIG_H_OK)
+        self.assertEqual(check_config_h()[0], CONFIG_H_OK)
 
         # then it tries to see if it can find "__GNUC__" in pyconfig.h
         sys.version = 'something without the *CC word'
 
         # if the file doesn't exist it returns  CONFIG_H_UNCERTAIN
-        self.assertEquals(check_config_h()[0], CONFIG_H_UNCERTAIN)
+        self.assertEqual(check_config_h()[0], CONFIG_H_UNCERTAIN)
 
         # if it exists but does not contain __GNUC__, it returns CONFIG_H_NOTOK
         self.write_file(self.python_h, 'xxx')
-        self.assertEquals(check_config_h()[0], CONFIG_H_NOTOK)
+        self.assertEqual(check_config_h()[0], CONFIG_H_NOTOK)
 
         # and CONFIG_H_OK if __GNUC__ is found
         self.write_file(self.python_h, 'xxx __GNUC__ xxx')
-        self.assertEquals(check_config_h()[0], CONFIG_H_OK)
+        self.assertEqual(check_config_h()[0], CONFIG_H_OK)
 
     def test_get_msvcr(self):
 
         # none
         sys.version  = ('2.6.1 (r261:67515, Dec  6 2008, 16:42:21) '
                         '\n[GCC 4.0.1 (Apple Computer, Inc. build 5370)]')
-        self.assertEquals(get_msvcr(), None)
+        self.assertEqual(get_msvcr(), None)
 
         # MSVC 7.0
         sys.version = ('2.5.1 (r251:54863, Apr 18 2007, 08:51:08) '
                        '[MSC v.1300 32 bits (Intel)]')
-        self.assertEquals(get_msvcr(), ['msvcr70'])
+        self.assertEqual(get_msvcr(), ['msvcr70'])
 
         # MSVC 7.1
         sys.version = ('2.5.1 (r251:54863, Apr 18 2007, 08:51:08) '
                        '[MSC v.1310 32 bits (Intel)]')
-        self.assertEquals(get_msvcr(), ['msvcr71'])
+        self.assertEqual(get_msvcr(), ['msvcr71'])
 
         # VS2005 / MSVC 8.0
         sys.version = ('2.5.1 (r251:54863, Apr 18 2007, 08:51:08) '
                        '[MSC v.1400 32 bits (Intel)]')
-        self.assertEquals(get_msvcr(), ['msvcr80'])
+        self.assertEqual(get_msvcr(), ['msvcr80'])
 
         # VS2008 / MSVC 9.0
         sys.version = ('2.5.1 (r251:54863, Apr 18 2007, 08:51:08) '
                        '[MSC v.1500 32 bits (Intel)]')
-        self.assertEquals(get_msvcr(), ['msvcr90'])
+        self.assertEqual(get_msvcr(), ['msvcr90'])
 
         # unknown
         sys.version = ('2.5.1 (r251:54863, Apr 18 2007, 08:51:08) '
diff --git a/src/distutils2/tests/test_dist.py b/src/distutils2/tests/test_dist.py
--- a/src/distutils2/tests/test_dist.py
+++ b/src/distutils2/tests/test_dist.py
@@ -71,11 +71,11 @@
         sys.argv.append("build")
 
         __, stdout = captured_stdout(self.create_distribution, files)
-        self.assertEquals(stdout, '')
+        self.assertEqual(stdout, '')
         distutils2.dist.DEBUG = True
         try:
             __, stdout = captured_stdout(self.create_distribution, files)
-            self.assertEquals(stdout, '')
+            self.assertEqual(stdout, '')
         finally:
             distutils2.dist.DEBUG = False
 
@@ -174,7 +174,7 @@
         finally:
             warnings.warn = old_warn
 
-        self.assertEquals(len(warns), 0)
+        self.assertEqual(len(warns), 0)
 
     def test_finalize_options(self):
 
@@ -185,20 +185,20 @@
         dist.finalize_options()
 
         # finalize_option splits platforms and keywords
-        self.assertEquals(dist.metadata['platform'], ['one', 'two'])
-        self.assertEquals(dist.metadata['keywords'], ['one', 'two'])
+        self.assertEqual(dist.metadata['platform'], ['one', 'two'])
+        self.assertEqual(dist.metadata['keywords'], ['one', 'two'])
 
     def test_get_command_packages(self):
         dist = Distribution()
-        self.assertEquals(dist.command_packages, None)
+        self.assertEqual(dist.command_packages, None)
         cmds = dist.get_command_packages()
-        self.assertEquals(cmds, ['distutils2.command'])
-        self.assertEquals(dist.command_packages,
+        self.assertEqual(cmds, ['distutils2.command'])
+        self.assertEqual(dist.command_packages,
                           ['distutils2.command'])
 
         dist.command_packages = 'one,two'
         cmds = dist.get_command_packages()
-        self.assertEquals(cmds, ['distutils2.command', 'one', 'two'])
+        self.assertEqual(cmds, ['distutils2.command', 'one', 'two'])
 
 
     def test_announce(self):
@@ -238,7 +238,7 @@
             os.path.expanduser = old_expander
 
         # make sure --no-user-cfg disables the user cfg file
-        self.assertEquals(len(all_files)-1, len(files))
+        self.assertEqual(len(all_files)-1, len(files))
 
 
 class MetadataTestCase(support.TempdirManager, support.EnvironGuard,
@@ -365,8 +365,8 @@
     def test_fix_help_options(self):
         help_tuples = [('a', 'b', 'c', 'd'), (1, 2, 3, 4)]
         fancy_options = fix_help_options(help_tuples)
-        self.assertEquals(fancy_options[0], ('a', 'b', 'c'))
-        self.assertEquals(fancy_options[1], (1, 2, 3))
+        self.assertEqual(fancy_options[0], ('a', 'b', 'c'))
+        self.assertEqual(fancy_options[1], (1, 2, 3))
 
     def test_show_help(self):
         # smoke test, just makes sure some help is displayed
@@ -412,14 +412,14 @@
         PKG_INFO.seek(0)
 
         metadata.read_file(PKG_INFO)
-        self.assertEquals(metadata['name'], "package")
-        self.assertEquals(metadata['version'], "1.0")
-        self.assertEquals(metadata['summary'], "xxx")
-        self.assertEquals(metadata['download_url'], 'http://example.com')
-        self.assertEquals(metadata['keywords'], ['one', 'two'])
-        self.assertEquals(metadata['platform'], [])
-        self.assertEquals(metadata['obsoletes'], [])
-        self.assertEquals(metadata['requires-dist'], ['foo'])
+        self.assertEqual(metadata['name'], "package")
+        self.assertEqual(metadata['version'], "1.0")
+        self.assertEqual(metadata['summary'], "xxx")
+        self.assertEqual(metadata['download_url'], 'http://example.com')
+        self.assertEqual(metadata['keywords'], ['one', 'two'])
+        self.assertEqual(metadata['platform'], [])
+        self.assertEqual(metadata['obsoletes'], [])
+        self.assertEqual(metadata['requires-dist'], ['foo'])
 
 def test_suite():
     suite = unittest.TestSuite()
diff --git a/src/distutils2/tests/test_install.py b/src/distutils2/tests/test_install.py
--- a/src/distutils2/tests/test_install.py
+++ b/src/distutils2/tests/test_install.py
@@ -139,23 +139,23 @@
 
         # two elements
         cmd.handle_extra_path()
-        self.assertEquals(cmd.extra_path, ['path', 'dirs'])
-        self.assertEquals(cmd.extra_dirs, 'dirs')
-        self.assertEquals(cmd.path_file, 'path')
+        self.assertEqual(cmd.extra_path, ['path', 'dirs'])
+        self.assertEqual(cmd.extra_dirs, 'dirs')
+        self.assertEqual(cmd.path_file, 'path')
 
         # one element
         cmd.extra_path = ['path']
         cmd.handle_extra_path()
-        self.assertEquals(cmd.extra_path, ['path'])
-        self.assertEquals(cmd.extra_dirs, 'path')
-        self.assertEquals(cmd.path_file, 'path')
+        self.assertEqual(cmd.extra_path, ['path'])
+        self.assertEqual(cmd.extra_dirs, 'path')
+        self.assertEqual(cmd.path_file, 'path')
 
         # none
         dist.extra_path = cmd.extra_path = None
         cmd.handle_extra_path()
-        self.assertEquals(cmd.extra_path, None)
-        self.assertEquals(cmd.extra_dirs, '')
-        self.assertEquals(cmd.path_file, None)
+        self.assertEqual(cmd.extra_path, None)
+        self.assertEqual(cmd.extra_dirs, '')
+        self.assertEqual(cmd.path_file, None)
 
         # three elements (no way !)
         cmd.extra_path = 'path,dirs,again'
@@ -199,7 +199,7 @@
         # line (the egg info file)
         f = open(cmd.record)
         try:
-            self.assertEquals(len(f.readlines()), 1)
+            self.assertEqual(len(f.readlines()), 1)
         finally:
             f.close()
 
diff --git a/src/distutils2/tests/test_install_data.py b/src/distutils2/tests/test_install_data.py
--- a/src/distutils2/tests/test_install_data.py
+++ b/src/distutils2/tests/test_install_data.py
@@ -27,14 +27,14 @@
         self.write_file(two, 'xxx')
 
         cmd.data_files = [one, (inst2, [two])]
-        self.assertEquals(cmd.get_inputs(), [one, (inst2, [two])])
+        self.assertEqual(cmd.get_inputs(), [one, (inst2, [two])])
 
         # let's run the command
         cmd.ensure_finalized()
         cmd.run()
 
         # let's check the result
-        self.assertEquals(len(cmd.get_outputs()), 2)
+        self.assertEqual(len(cmd.get_outputs()), 2)
         rtwo = os.path.split(two)[-1]
         self.assertTrue(os.path.exists(os.path.join(inst2, rtwo)))
         rone = os.path.split(one)[-1]
@@ -47,7 +47,7 @@
         cmd.run()
 
         # let's check the result
-        self.assertEquals(len(cmd.get_outputs()), 2)
+        self.assertEqual(len(cmd.get_outputs()), 2)
         self.assertTrue(os.path.exists(os.path.join(inst2, rtwo)))
         self.assertTrue(os.path.exists(os.path.join(inst, rone)))
         cmd.outfiles = []
@@ -65,7 +65,7 @@
         cmd.run()
 
         # let's check the result
-        self.assertEquals(len(cmd.get_outputs()), 4)
+        self.assertEqual(len(cmd.get_outputs()), 4)
         self.assertTrue(os.path.exists(os.path.join(inst2, rtwo)))
         self.assertTrue(os.path.exists(os.path.join(inst, rone)))
 
diff --git a/src/distutils2/tests/test_install_headers.py b/src/distutils2/tests/test_install_headers.py
--- a/src/distutils2/tests/test_install_headers.py
+++ b/src/distutils2/tests/test_install_headers.py
@@ -23,7 +23,7 @@
 
         pkg_dir, dist = self.create_dist(headers=headers)
         cmd = install_headers(dist)
-        self.assertEquals(cmd.get_inputs(), headers)
+        self.assertEqual(cmd.get_inputs(), headers)
 
         # let's run the command
         cmd.install_dir = os.path.join(pkg_dir, 'inst')
@@ -31,7 +31,7 @@
         cmd.run()
 
         # let's check the results
-        self.assertEquals(len(cmd.get_outputs()), 2)
+        self.assertEqual(len(cmd.get_outputs()), 2)
 
 def test_suite():
     return unittest.makeSuite(InstallHeadersTestCase)
diff --git a/src/distutils2/tests/test_install_lib.py b/src/distutils2/tests/test_install_lib.py
--- a/src/distutils2/tests/test_install_lib.py
+++ b/src/distutils2/tests/test_install_lib.py
@@ -25,8 +25,8 @@
         cmd = install_lib(dist)
 
         cmd.finalize_options()
-        self.assertEquals(cmd.compile, 1)
-        self.assertEquals(cmd.optimize, 0)
+        self.assertEqual(cmd.compile, 1)
+        self.assertEqual(cmd.optimize, 0)
 
         # optimize must be 0, 1, or 2
         cmd.optimize = 'foo'
@@ -36,7 +36,7 @@
 
         cmd.optimize = '2'
         cmd.finalize_options()
-        self.assertEquals(cmd.optimize, 2)
+        self.assertEqual(cmd.optimize, 2)
 
     @unittest.skipIf(no_bytecode, 'byte-compile not supported')
     def test_byte_compile(self):
@@ -82,7 +82,7 @@
         cmd.distribution.script_name = 'setup.py'
 
         # get_input should return 2 elements
-        self.assertEquals(len(cmd.get_inputs()), 2)
+        self.assertEqual(len(cmd.get_inputs()), 2)
 
     @unittest.skipUnless(bytecode_support, 'sys.dont_write_bytecode not supported')
     def test_dont_write_bytecode(self):
diff --git a/src/distutils2/tests/test_manifest.py b/src/distutils2/tests/test_manifest.py
--- a/src/distutils2/tests/test_manifest.py
+++ b/src/distutils2/tests/test_manifest.py
@@ -44,7 +44,7 @@
 
         # the manifest should have been read
         # and 3 warnings issued (we ddidn't provided the files)
-        self.assertEquals(len(warns), 3)
+        self.assertEqual(len(warns), 3)
         for warn in warns:
             self.assertIn('warning: no files found matching', warn)
 
diff --git a/src/distutils2/tests/test_metadata.py b/src/distutils2/tests/test_metadata.py
--- a/src/distutils2/tests/test_metadata.py
+++ b/src/distutils2/tests/test_metadata.py
@@ -75,14 +75,14 @@
         content = content % sys.platform
         metadata = DistributionMetadata(platform_dependent=True)
         metadata.read_file(StringIO(content))
-        self.assertEquals(metadata['Requires-Dist'], ['bar'])
+        self.assertEqual(metadata['Requires-Dist'], ['bar'])
 
         # test with context
         context = {'sys.platform': 'okook'}
         metadata = DistributionMetadata(platform_dependent=True,
                                         execution_context=context)
         metadata.read_file(StringIO(content))
-        self.assertEquals(metadata['Requires-Dist'], ['foo'])
+        self.assertEqual(metadata['Requires-Dist'], ['foo'])
 
     def test_description(self):
         PKG_INFO = os.path.join(os.path.dirname(__file__), 'PKG-INFO')
@@ -94,14 +94,14 @@
         # see if we can read the description now
         DESC = os.path.join(os.path.dirname(__file__), 'LONG_DESC.txt')
         wanted = open(DESC).read()
-        self.assertEquals(wanted, metadata['Description'])
+        self.assertEqual(wanted, metadata['Description'])
 
         # save the file somewhere and make sure we can read it back
         out = StringIO()
         metadata.write_file(out)
         out.seek(0)
         metadata.read_file(out)
-        self.assertEquals(wanted, metadata['Description'])
+        self.assertEqual(wanted, metadata['Description'])
 
     def test_mapper_apis(self):
         PKG_INFO = os.path.join(os.path.dirname(__file__), 'PKG-INFO')
@@ -116,25 +116,25 @@
     def test_versions(self):
         metadata = DistributionMetadata()
         metadata['Obsoletes'] = 'ok'
-        self.assertEquals(metadata['Metadata-Version'], '1.1')
+        self.assertEqual(metadata['Metadata-Version'], '1.1')
 
         del metadata['Obsoletes']
         metadata['Obsoletes-Dist'] = 'ok'
-        self.assertEquals(metadata['Metadata-Version'], '1.2')
+        self.assertEqual(metadata['Metadata-Version'], '1.2')
 
         del metadata['Obsoletes-Dist']
         metadata['Version'] = '1'
-        self.assertEquals(metadata['Metadata-Version'], '1.0')
+        self.assertEqual(metadata['Metadata-Version'], '1.0')
 
         PKG_INFO = os.path.join(os.path.dirname(__file__),
                                 'SETUPTOOLS-PKG-INFO')
         metadata.read_file(StringIO(open(PKG_INFO).read()))
-        self.assertEquals(metadata['Metadata-Version'], '1.0')
+        self.assertEqual(metadata['Metadata-Version'], '1.0')
 
         PKG_INFO = os.path.join(os.path.dirname(__file__),
                                 'SETUPTOOLS-PKG-INFO2')
         metadata.read_file(StringIO(open(PKG_INFO).read()))
-        self.assertEquals(metadata['Metadata-Version'], '1.1')
+        self.assertEqual(metadata['Metadata-Version'], '1.1')
 
     def test_warnings(self):
         metadata = DistributionMetadata()
@@ -162,7 +162,7 @@
 
         # we should have a certain amount of warnings
         num_wanted = len(values)
-        self.assertEquals(num_wanted, res)
+        self.assertEqual(num_wanted, res)
 
     def test_multiple_predicates(self):
         metadata = DistributionMetadata()
@@ -185,29 +185,29 @@
             res = m.warns
             del m.warns
 
-        self.assertEquals(res, 0)
+        self.assertEqual(res, 0)
 
     def test_project_url(self):
         metadata = DistributionMetadata()
         metadata['Project-URL'] = [('one', 'http://ok')]
-        self.assertEquals(metadata['Project-URL'],
+        self.assertEqual(metadata['Project-URL'],
                           [('one', 'http://ok')])
-        self.assertEquals(metadata.version, '1.2')
+        self.assertEqual(metadata.version, '1.2')
 
     def test_check(self):
         metadata = DistributionMetadata()
         metadata['Version'] = 'rr'
         metadata['Requires-dist'] = ['Foo (a)']
         missing, warnings = metadata.check()
-        self.assertEquals(missing, ['Name', 'Home-page'])
-        self.assertEquals(len(warnings), 2)
+        self.assertEqual(missing, ['Name', 'Home-page'])
+        self.assertEqual(len(warnings), 2)
 
     def test_best_choice(self):
         metadata = DistributionMetadata()
         metadata['Version'] = '1.0'
-        self.assertEquals(metadata.version, PKG_INFO_PREFERRED_VERSION)
+        self.assertEqual(metadata.version, PKG_INFO_PREFERRED_VERSION)
         metadata['Classifier'] = ['ok']
-        self.assertEquals(metadata.version, '1.2')
+        self.assertEqual(metadata.version, '1.2')
 
     def test_project_urls(self):
         # project-url is a bit specific, make sure we write it
@@ -215,7 +215,7 @@
         metadata = DistributionMetadata()
         metadata['Version'] = '1.0'
         metadata['Project-Url'] = [('one', 'http://ok')]
-        self.assertEquals(metadata['Project-Url'], [('one', 'http://ok')])
+        self.assertEqual(metadata['Project-Url'], [('one', 'http://ok')])
         file_ = StringIO()
         metadata.write_file(file_)
         file_.seek(0)
@@ -225,7 +225,7 @@
         file_.seek(0)
         metadata = DistributionMetadata()
         metadata.read_file(file_)
-        self.assertEquals(metadata['Project-Url'], [('one', 'http://ok')])
+        self.assertEqual(metadata['Project-Url'], [('one', 'http://ok')])
 
 
 def test_suite():
diff --git a/src/distutils2/tests/test_msvc9compiler.py b/src/distutils2/tests/test_msvc9compiler.py
--- a/src/distutils2/tests/test_msvc9compiler.py
+++ b/src/distutils2/tests/test_msvc9compiler.py
@@ -105,7 +105,7 @@
         import _winreg
         HKCU = _winreg.HKEY_CURRENT_USER
         keys = Reg.read_keys(HKCU, 'xxxx')
-        self.assertEquals(keys, None)
+        self.assertEqual(keys, None)
 
         keys = Reg.read_keys(HKCU, r'Control Panel')
         self.assertTrue('Desktop' in keys)
@@ -129,7 +129,7 @@
         f.close()
 
         # makes sure the manifest was properly cleaned
-        self.assertEquals(content, _CLEANED_MANIFEST)
+        self.assertEqual(content, _CLEANED_MANIFEST)
 
 
 def test_suite():
diff --git a/src/distutils2/tests/test_register.py b/src/distutils2/tests/test_register.py
--- a/src/distutils2/tests/test_register.py
+++ b/src/distutils2/tests/test_register.py
@@ -124,7 +124,7 @@
 
         # with the content similar to WANTED_PYPIRC
         content = open(self.rc).read()
-        self.assertEquals(content, WANTED_PYPIRC)
+        self.assertEqual(content, WANTED_PYPIRC)
 
         # now let's make sure the .pypirc file generated
         # really works : we shouldn't be asked anything
@@ -141,7 +141,7 @@
         self.assertTrue(self.conn.reqs, 2)
         req1 = dict(self.conn.reqs[0].headers)
         req2 = dict(self.conn.reqs[1].headers)
-        self.assertEquals(req2['Content-length'], req1['Content-length'])
+        self.assertEqual(req2['Content-length'], req1['Content-length'])
         self.assertTrue('xxx' in self.conn.reqs[1].data)
 
     def test_password_not_in_file(self):
@@ -154,7 +154,7 @@
 
         # dist.password should be set
         # therefore used afterwards by other commands
-        self.assertEquals(cmd.distribution.password, 'password')
+        self.assertEqual(cmd.distribution.password, 'password')
 
     def test_registering(self):
         # this test runs choice 2
@@ -171,7 +171,7 @@
         self.assertTrue(self.conn.reqs, 1)
         req = self.conn.reqs[0]
         headers = dict(req.headers)
-        self.assertEquals(headers['Content-length'], '608')
+        self.assertEqual(headers['Content-length'], '608')
         self.assertTrue('tarek' in req.data)
 
     def test_password_reset(self):
@@ -189,7 +189,7 @@
         self.assertTrue(self.conn.reqs, 1)
         req = self.conn.reqs[0]
         headers = dict(req.headers)
-        self.assertEquals(headers['Content-length'], '290')
+        self.assertEqual(headers['Content-length'], '290')
         self.assertTrue('tarek' in req.data)
 
     @unittest.skipUnless(DOCUTILS_SUPPORT, 'needs docutils')
@@ -246,8 +246,8 @@
         cmd.ensure_finalized()
         cmd.distribution.metadata['Requires-Dist'] = ['lxml']
         data = cmd.build_post_data('submit')
-        self.assertEquals(data['metadata_version'], '1.2')
-        self.assertEquals(data['requires_dist'], ['lxml'])
+        self.assertEqual(data['metadata_version'], '1.2')
+        self.assertEqual(data['requires_dist'], ['lxml'])
 
 def test_suite():
     return unittest.makeSuite(RegisterTestCase)
diff --git a/src/distutils2/tests/test_sdist.py b/src/distutils2/tests/test_sdist.py
--- a/src/distutils2/tests/test_sdist.py
+++ b/src/distutils2/tests/test_sdist.py
@@ -127,7 +127,7 @@
         # now let's check what we have
         dist_folder = join(self.tmp_dir, 'dist')
         files = os.listdir(dist_folder)
-        self.assertEquals(files, ['fake-1.0.zip'])
+        self.assertEqual(files, ['fake-1.0.zip'])
 
         zip_file = zipfile.ZipFile(join(dist_folder, 'fake-1.0.zip'))
         try:
@@ -136,7 +136,7 @@
             zip_file.close()
 
         # making sure everything has been pruned correctly
-        self.assertEquals(len(content), 4)
+        self.assertEqual(len(content), 4)
 
     @unittest.skipUnless(zlib, "requires zlib")
     def test_make_distribution(self):
@@ -158,7 +158,7 @@
         dist_folder = join(self.tmp_dir, 'dist')
         result = os.listdir(dist_folder)
         result.sort()
-        self.assertEquals(result,
+        self.assertEqual(result,
                           ['fake-1.0.tar', 'fake-1.0.tar.gz'] )
 
         os.remove(join(dist_folder, 'fake-1.0.tar'))
@@ -172,7 +172,7 @@
 
         result = os.listdir(dist_folder)
         result.sort()
-        self.assertEquals(result,
+        self.assertEqual(result,
                 ['fake-1.0.tar', 'fake-1.0.tar.gz'])
 
     @unittest.skipUnless(zlib, "requires zlib")
@@ -222,7 +222,7 @@
         # now let's check what we have
         dist_folder = join(self.tmp_dir, 'dist')
         files = os.listdir(dist_folder)
-        self.assertEquals(files, ['fake-1.0.zip'])
+        self.assertEqual(files, ['fake-1.0.zip'])
 
         zip_file = zipfile.ZipFile(join(dist_folder, 'fake-1.0.zip'))
         try:
@@ -231,11 +231,11 @@
             zip_file.close()
 
         # making sure everything was added
-        self.assertEquals(len(content), 11)
+        self.assertEqual(len(content), 11)
 
         # checking the MANIFEST
         manifest = open(join(self.tmp_dir, 'MANIFEST')).read()
-        self.assertEquals(manifest, MANIFEST % {'sep': os.sep})
+        self.assertEqual(manifest, MANIFEST % {'sep': os.sep})
 
     @unittest.skipUnless(zlib, "requires zlib")
     def test_metadata_check_option(self):
@@ -247,7 +247,7 @@
         cmd.ensure_finalized()
         cmd.run()
         warnings = self.get_logs(WARN)
-        self.assertEquals(len(warnings), 1)
+        self.assertEqual(len(warnings), 1)
 
         # trying with a complete set of metadata
         self.clear_logs()
@@ -259,7 +259,7 @@
         # removing manifest generated warnings
         warnings = [warn for warn in warnings if
                     not warn.endswith('-- skipping')]
-        self.assertEquals(len(warnings), 0)
+        self.assertEqual(len(warnings), 0)
 
 
     def test_show_formats(self):
@@ -269,7 +269,7 @@
         num_formats = len(get_archive_formats())
         output = [line for line in stdout.split('\n')
                   if line.strip().startswith('--formats=')]
-        self.assertEquals(len(output), num_formats)
+        self.assertEqual(len(output), num_formats)
 
     def test_finalize_options(self):
 
@@ -277,9 +277,9 @@
         cmd.finalize_options()
 
         # default options set by finalize
-        self.assertEquals(cmd.manifest, 'MANIFEST')
-        self.assertEquals(cmd.template, 'MANIFEST.in')
-        self.assertEquals(cmd.dist_dir, 'dist')
+        self.assertEqual(cmd.manifest, 'MANIFEST')
+        self.assertEqual(cmd.template, 'MANIFEST.in')
+        self.assertEqual(cmd.dist_dir, 'dist')
 
         # formats has to be a string splitable on (' ', ',') or
         # a stringlist
@@ -316,8 +316,8 @@
         archive = tarfile.open(archive_name)
         try:
             for member in archive.getmembers():
-                self.assertEquals(member.uid, 0)
-                self.assertEquals(member.gid, 0)
+                self.assertEqual(member.uid, 0)
+                self.assertEqual(member.gid, 0)
         finally:
             archive.close()
 
@@ -338,7 +338,7 @@
         # rights (see #7408)
         try:
             for member in archive.getmembers():
-                self.assertEquals(member.uid, os.getuid())
+                self.assertEqual(member.uid, os.getuid())
         finally:
             archive.close()
 
diff --git a/src/distutils2/tests/test_spawn.py b/src/distutils2/tests/test_spawn.py
--- a/src/distutils2/tests/test_spawn.py
+++ b/src/distutils2/tests/test_spawn.py
@@ -20,7 +20,7 @@
                                (['nochange', 'nospace'],
                                 ['nochange', 'nospace'])):
             res = _nt_quote_args(args)
-            self.assertEquals(res, wanted)
+            self.assertEqual(res, wanted)
 
 
     @unittest.skipUnless(os.name in ('nt', 'posix'),
diff --git a/src/distutils2/tests/test_upload.py b/src/distutils2/tests/test_upload.py
--- a/src/distutils2/tests/test_upload.py
+++ b/src/distutils2/tests/test_upload.py
@@ -80,7 +80,7 @@
         for attr, expected in (('username', 'me'), ('password', 'secret'),
                                ('realm', 'pypi'),
                                ('repository', 'http://pypi.python.org/pypi')):
-            self.assertEquals(getattr(cmd, attr), expected)
+            self.assertEqual(getattr(cmd, attr), expected)
 
     def test_saved_password(self):
         # file with no password
@@ -90,14 +90,14 @@
         dist = Distribution()
         cmd = upload(dist)
         cmd.finalize_options()
-        self.assertEquals(cmd.password, None)
+        self.assertEqual(cmd.password, None)
 
         # make sure we get it as well, if another command
         # initialized it at the dist level
         dist.password = 'xxx'
         cmd = upload(dist)
         cmd.finalize_options()
-        self.assertEquals(cmd.password, 'xxx')
+        self.assertEqual(cmd.password, 'xxx')
 
     def test_upload(self):
         tmp = self.mkdtemp()
@@ -118,8 +118,8 @@
         headers = dict(self.last_open.req.headers)
         self.assertTrue(int(headers['Content-length']) < 2000)
         self.assertTrue(headers['Content-type'].startswith('multipart/form-data'))
-        self.assertEquals(self.last_open.req.get_method(), 'POST')
-        self.assertEquals(self.last_open.req.get_full_url(),
+        self.assertEqual(self.last_open.req.get_method(), 'POST')
+        self.assertEqual(self.last_open.req.get_full_url(),
                           'http://pypi.python.org/pypi')
         self.assertTrue('xxx' in self.last_open.req.data)
         auth = self.last_open.req.headers['Authorization']
diff --git a/src/distutils2/tests/test_util.py b/src/distutils2/tests/test_util.py
--- a/src/distutils2/tests/test_util.py
+++ b/src/distutils2/tests/test_util.py
@@ -100,7 +100,7 @@
             return '/'.join(path)
         os.path.join = _join
 
-        self.assertEquals(convert_path('/home/to/my/stuff'),
+        self.assertEqual(convert_path('/home/to/my/stuff'),
                           '/home/to/my/stuff')
 
         # win
@@ -112,9 +112,9 @@
         self.assertRaises(ValueError, convert_path, '/home/to/my/stuff')
         self.assertRaises(ValueError, convert_path, 'home/to/my/stuff/')
 
-        self.assertEquals(convert_path('home/to/my/stuff'),
+        self.assertEqual(convert_path('home/to/my/stuff'),
                           'home\\to\\my\\stuff')
-        self.assertEquals(convert_path('.'),
+        self.assertEqual(convert_path('.'),
                           os.curdir)
 
     def test_change_root(self):
@@ -127,9 +127,9 @@
             return '/'.join(path)
         os.path.join = _join
 
-        self.assertEquals(change_root('/root', '/old/its/here'),
+        self.assertEqual(change_root('/root', '/old/its/here'),
                           '/root/old/its/here')
-        self.assertEquals(change_root('/root', 'its/here'),
+        self.assertEqual(change_root('/root', 'its/here'),
                           '/root/its/here')
 
         # windows
@@ -146,9 +146,9 @@
             return '\\'.join(path)
         os.path.join = _join
 
-        self.assertEquals(change_root('c:\\root', 'c:\\old\\its\\here'),
+        self.assertEqual(change_root('c:\\root', 'c:\\old\\its\\here'),
                           'c:\\root\\old\\its\\here')
-        self.assertEquals(change_root('c:\\root', 'its\\here'),
+        self.assertEqual(change_root('c:\\root', 'its\\here'),
                           'c:\\root\\its\\here')
 
         # BugsBunny os (it's a great os)
@@ -159,7 +159,7 @@
         # XXX platforms to be covered: os2, mac
 
     def test_split_quoted(self):
-        self.assertEquals(split_quoted('""one"" "two" \'three\' \\four'),
+        self.assertEqual(split_quoted('""one"" "two" \'three\' \\four'),
                           ['one', 'two', 'three', 'four'])
 
     def test_strtobool(self):
@@ -177,7 +177,7 @@
         res = rfc822_escape(header)
         wanted = ('I am a%(8s)spoor%(8s)slonesome%(8s)s'
                   'header%(8s)s') % {'8s': '\n'+8*' '}
-        self.assertEquals(res, wanted)
+        self.assertEqual(res, wanted)
 
     def test_find_exe_version(self):
         # the ld version scheme under MAC OS is:
@@ -195,7 +195,7 @@
                                 ('@(#)PROGRAM:ld  PROJECT:ld64-95.2.12',
                                  '95.2.12')):
             result = _MAC_OS_X_LD_VERSION.search(output)
-            self.assertEquals(result.group(1), version)
+            self.assertEqual(result.group(1), version)
 
     def _find_executable(self, name):
         if name in self._exes:
@@ -205,43 +205,43 @@
     def test_get_compiler_versions(self):
         # get_versions calls distutils.spawn.find_executable on
         # 'gcc', 'ld' and 'dllwrap'
-        self.assertEquals(get_compiler_versions(), (None, None, None))
+        self.assertEqual(get_compiler_versions(), (None, None, None))
 
         # Let's fake we have 'gcc' and it returns '3.4.5'
         self._exes['gcc'] = 'gcc (GCC) 3.4.5 (mingw special)\nFSF'
         res = get_compiler_versions()
-        self.assertEquals(str(res[0]), '3.4.5')
+        self.assertEqual(str(res[0]), '3.4.5')
 
         # and let's see what happens when the version
         # doesn't match the regular expression
         # (\d+\.\d+(\.\d+)*)
         self._exes['gcc'] = 'very strange output'
         res = get_compiler_versions()
-        self.assertEquals(res[0], None)
+        self.assertEqual(res[0], None)
 
         # same thing for ld
         if sys.platform != 'darwin':
             self._exes['ld'] = 'GNU ld version 2.17.50 20060824'
             res = get_compiler_versions()
-            self.assertEquals(str(res[1]), '2.17.50')
+            self.assertEqual(str(res[1]), '2.17.50')
             self._exes['ld'] = '@(#)PROGRAM:ld  PROJECT:ld64-77'
             res = get_compiler_versions()
-            self.assertEquals(res[1], None)
+            self.assertEqual(res[1], None)
         else:
             self._exes['ld'] = 'GNU ld version 2.17.50 20060824'
             res = get_compiler_versions()
-            self.assertEquals(res[1], None)
+            self.assertEqual(res[1], None)
             self._exes['ld'] = '@(#)PROGRAM:ld  PROJECT:ld64-77'
             res = get_compiler_versions()
-            self.assertEquals(str(res[1]), '77')
+            self.assertEqual(str(res[1]), '77')
 
         # and dllwrap
         self._exes['dllwrap'] = 'GNU dllwrap 2.17.50 20060824\nFSF'
         res = get_compiler_versions()
-        self.assertEquals(str(res[2]), '2.17.50')
+        self.assertEqual(str(res[2]), '2.17.50')
         self._exes['dllwrap'] = 'Cheese Wrap'
         res = get_compiler_versions()
-        self.assertEquals(res[2], None)
+        self.assertEqual(res[2], None)
 
     @unittest.skipUnless(hasattr(sys, 'dont_write_bytecode'),
                           'no dont_write_bytecode support')
@@ -294,7 +294,7 @@
         self.write_file(os.path.join(pkg5, '__init__.py'))
 
         res = find_packages([root], ['pkg1.pkg2'])
-        self.assertEquals(set(res), set(['pkg1', 'pkg5', 'pkg1.pkg3', 'pkg1.pkg3.pkg6']))
+        self.assertEqual(set(res), set(['pkg1', 'pkg5', 'pkg1.pkg3', 'pkg1.pkg3.pkg6']))
 
 
 def test_suite():
diff --git a/src/distutils2/tests/test_version.py b/src/distutils2/tests/test_version.py
--- a/src/distutils2/tests/test_version.py
+++ b/src/distutils2/tests/test_version.py
@@ -25,15 +25,15 @@
     def test_basic_versions(self):
 
         for v, s in self.versions:
-            self.assertEquals(str(v), s)
+            self.assertEqual(str(v), s)
 
     def test_from_parts(self):
 
         for v, s in self.versions:
             parts = v.parts
             v2 = V.from_parts(*v.parts)
-            self.assertEquals(v, v2)
-            self.assertEquals(str(v), str(v2))
+            self.assertEqual(v, v2)
+            self.assertEqual(str(v), str(v2))
 
     def test_irrational_versions(self):
 
@@ -96,34 +96,34 @@
 
     def test_suggest_normalized_version(self):
 
-        self.assertEquals(suggest('1.0'), '1.0')
-        self.assertEquals(suggest('1.0-alpha1'), '1.0a1')
-        self.assertEquals(suggest('1.0c2'), '1.0c2')
-        self.assertEquals(suggest('walla walla washington'), None)
-        self.assertEquals(suggest('2.4c1'), '2.4c1')
+        self.assertEqual(suggest('1.0'), '1.0')
+        self.assertEqual(suggest('1.0-alpha1'), '1.0a1')
+        self.assertEqual(suggest('1.0c2'), '1.0c2')
+        self.assertEqual(suggest('walla walla washington'), None)
+        self.assertEqual(suggest('2.4c1'), '2.4c1')
 
         # from setuptools
-        self.assertEquals(suggest('0.4a1.r10'), '0.4a1.post10')
-        self.assertEquals(suggest('0.7a1dev-r66608'), '0.7a1.dev66608')
-        self.assertEquals(suggest('0.6a9.dev-r41475'), '0.6a9.dev41475')
-        self.assertEquals(suggest('2.4preview1'), '2.4c1')
-        self.assertEquals(suggest('2.4pre1') , '2.4c1')
-        self.assertEquals(suggest('2.1-rc2'), '2.1c2')
+        self.assertEqual(suggest('0.4a1.r10'), '0.4a1.post10')
+        self.assertEqual(suggest('0.7a1dev-r66608'), '0.7a1.dev66608')
+        self.assertEqual(suggest('0.6a9.dev-r41475'), '0.6a9.dev41475')
+        self.assertEqual(suggest('2.4preview1'), '2.4c1')
+        self.assertEqual(suggest('2.4pre1') , '2.4c1')
+        self.assertEqual(suggest('2.1-rc2'), '2.1c2')
 
         # from pypi
-        self.assertEquals(suggest('0.1dev'), '0.1.dev0')
-        self.assertEquals(suggest('0.1.dev'), '0.1.dev0')
+        self.assertEqual(suggest('0.1dev'), '0.1.dev0')
+        self.assertEqual(suggest('0.1.dev'), '0.1.dev0')
 
         # we want to be able to parse Twisted
         # development versions are like post releases in Twisted
-        self.assertEquals(suggest('9.0.0+r2363'), '9.0.0.post2363')
+        self.assertEqual(suggest('9.0.0+r2363'), '9.0.0.post2363')
 
         # pre-releases are using markers like "pre1"
-        self.assertEquals(suggest('9.0.0pre1'), '9.0.0c1')
+        self.assertEqual(suggest('9.0.0pre1'), '9.0.0c1')
 
         # we want to be able to parse Tcl-TK
         # they us "p1" "p2" for post releases
-        self.assertEquals(suggest('1.4p1'), '1.4.post1')
+        self.assertEqual(suggest('1.4p1'), '1.4.post1')
 
     def test_predicate(self):
         # VersionPredicate knows how to parse stuff like:

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul  3 16:19:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 03 Jul 2010 16:19:06 +0200
Subject: [Python-checkins] distutils2: Add jkloth to contributors and sort
	list again
Message-ID: <E1OV3Yg-0002Jd-C9@dinsdale.python.org>

tarek.ziade pushed 658ed3f08df8 to distutils2:

http://hg.python.org/distutils2/rev/658ed3f08df8
changeset:   233:658ed3f08df8
user:        ?ric Araujo <merwok at netwok.org>
date:        Thu Jun 24 17:39:52 2010 +0200
summary:     Add jkloth to contributors and sort list again
files:       src/CONTRIBUTORS.txt

diff --git a/src/CONTRIBUTORS.txt b/src/CONTRIBUTORS.txt
--- a/src/CONTRIBUTORS.txt
+++ b/src/CONTRIBUTORS.txt
@@ -5,7 +5,7 @@
 Distutils2 is a project that was started and that is maintained by
 Tarek Ziad??, and many people are contributing to the project.
 
-If you did, please add your name below in alphabetical order !
+If you did, please add your name below in alphabetical order!
 
 Thanks to:
 
@@ -15,6 +15,8 @@
 - Nicolas Cadou
 - Josip Djolonga
 - Yannick Gringas
+- Jeremy Kloth
+- Martin von L??wis
 - Carl Meyer
 - Michael Mulich
 - George Peris
@@ -22,5 +24,3 @@
 - Erik Rose
 - Brian Rosner
 - Alexandre Vassalotti
-- Martin von L??wis
-

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul  3 16:19:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 03 Jul 2010 16:19:06 +0200
Subject: [Python-checkins] distutils2: Pull changeset with new docstrings
	from jkloth
Message-ID: <E1OV3Yg-0002Ja-BY@dinsdale.python.org>

tarek.ziade pushed ba4aa1e016db to distutils2:

http://hg.python.org/distutils2/rev/ba4aa1e016db
changeset:   232:ba4aa1e016db
parent:      230:0b1b57c04665
parent:      231:8fb85c777536
user:        ?ric Araujo <merwok at netwok.org>
date:        Thu Jun 24 17:39:49 2010 +0200
summary:     Pull changeset with new docstrings from jkloth
files:       src/distutils2/command/cmd.py

diff --git a/src/distutils2/command/cmd.py b/src/distutils2/command/cmd.py
--- a/src/distutils2/command/cmd.py
+++ b/src/distutils2/command/cmd.py
@@ -189,6 +189,31 @@
         """
         log.log(level, msg)
 
+    # -- External interface --------------------------------------------
+    # (called by outsiders)
+
+    def get_source_files(self):
+        """Return the list of files that are used as inputs to this command,
+        i.e. the files used to generate the output files.  The result is used
+        by the `sdist` command in determining the set of default files.
+
+        Command classes should implement this method if they operate on files
+        from the source tree.
+        """
+        return []
+
+    def get_outputs(self):
+        """Return the list of files that would be produced if this command
+        were actually run.  Not affected by the "dry-run" flag or whether
+        any other commands have been run.
+
+        Command classes should implement this method if they produce any
+        output files that get consumed by another command.  e.g., `build_ext`
+        returns the list of built extension modules, but not any temporary
+        files used in the compilation process.
+        """
+        return []
+
     # -- Option validation methods -------------------------------------
     # (these are very handy in writing the 'finalize_options()' method)
     #

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul  3 16:19:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 03 Jul 2010 16:19:06 +0200
Subject: [Python-checkins] distutils2: Make all classes new-style classes.
Message-ID: <E1OV3Yg-0002Jm-Du@dinsdale.python.org>

tarek.ziade pushed 1c2ab153640e to distutils2:

http://hg.python.org/distutils2/rev/1c2ab153640e
changeset:   235:1c2ab153640e
user:        ?ric Araujo <merwok at netwok.org>
date:        Sat Jun 26 01:43:30 2010 +0200
summary:     Make all classes new-style classes.
files:       src/distutils2/_backport/pkgutil.py, src/distutils2/_backport/tarfile.py, src/distutils2/command/cmd.py, src/distutils2/compiler/ccompiler.py, src/distutils2/compiler/msvc9compiler.py, src/distutils2/compiler/msvccompiler.py, src/distutils2/extension.py, src/distutils2/fancy_getopt.py, src/distutils2/log.py, src/distutils2/mkpkg.py, src/distutils2/tests/__init__.py

diff --git a/src/distutils2/_backport/pkgutil.py b/src/distutils2/_backport/pkgutil.py
--- a/src/distutils2/_backport/pkgutil.py
+++ b/src/distutils2/_backport/pkgutil.py
@@ -179,7 +179,7 @@
 iter_importer_modules = simplegeneric(iter_importer_modules)
 
 
-class ImpImporter:
+class ImpImporter(object):
     """:pep:`302` Importer that wraps Python's "classic" import algorithm
 
     ``ImpImporter(dirname)`` produces a :pep:`302` importer that searches that
@@ -242,7 +242,7 @@
                 yield prefix + modname, ispkg
 
 
-class ImpLoader:
+class ImpLoader(object):
     """:pep:`302` Loader that wraps Python's "classic" import algorithm """
 
     code = source = None
@@ -732,6 +732,9 @@
     def __eq__(self, other):
         return isinstance(other, Distribution) and self.path == other.path
 
+    # See http://docs.python.org/reference/datamodel#object.__hash__
+    __hash__ = object.__hash__
+
 
 class EggInfoDistribution(object):
     """Created with the *path* of the ``.egg-info`` directory or file provided
@@ -847,6 +850,9 @@
         return isinstance(other, EggInfoDistribution) and \
                self.path == other.path
 
+    # See http://docs.python.org/reference/datamodel#object.__hash__
+    __hash__ = object.__hash__
+
 
 def _normalize_dist_name(name):
     """Returns a normalized name from the given *name*.
diff --git a/src/distutils2/_backport/tarfile.py b/src/distutils2/_backport/tarfile.py
--- a/src/distutils2/_backport/tarfile.py
+++ b/src/distutils2/_backport/tarfile.py
@@ -370,7 +370,7 @@
 #---------------------------
 # internal stream interface
 #---------------------------
-class _LowLevelFile:
+class _LowLevelFile(object):
     """Low-level file object. Supports reading and writing.
        It is used instead of a regular file object for streaming
        access.
@@ -394,7 +394,7 @@
     def write(self, s):
         os.write(self.fd, s)
 
-class _Stream:
+class _Stream(object):
     """Class that serves as an adapter between TarFile and
        a stream-like object.  The stream-like object only
        needs to have a read() or write() method and is accessed
@@ -2423,7 +2423,7 @@
             print >> sys.stderr, msg
 # class TarFile
 
-class TarIter:
+class TarIter(object):
     """Iterator Class.
 
        for tarinfo in TarFile(...):
@@ -2460,7 +2460,7 @@
         return tarinfo
 
 # Helper classes for sparse file support
-class _section:
+class _section(object):
     """Base class for _data and _hole.
     """
     def __init__(self, offset, size):
@@ -2507,7 +2507,7 @@
 #---------------------------------------------
 TAR_PLAIN = 0           # zipfile.ZIP_STORED
 TAR_GZIPPED = 8         # zipfile.ZIP_DEFLATED
-class TarFileCompat:
+class TarFileCompat(object):
     """TarFile class compatible with standard module zipfile's
        ZipFile class.
     """
diff --git a/src/distutils2/command/cmd.py b/src/distutils2/command/cmd.py
--- a/src/distutils2/command/cmd.py
+++ b/src/distutils2/command/cmd.py
@@ -19,7 +19,7 @@
 except ImportError:
     from distutils2._backport.shutil import make_archive
 
-class Command:
+class Command(object):
     """Abstract base class for defining command classes, the "worker bees"
     of the Distutils.  A useful analogy for command classes is to think of
     them as subroutines with local variables called "options".  The options
diff --git a/src/distutils2/compiler/ccompiler.py b/src/distutils2/compiler/ccompiler.py
--- a/src/distutils2/compiler/ccompiler.py
+++ b/src/distutils2/compiler/ccompiler.py
@@ -76,7 +76,7 @@
 
         compiler.shared_lib_extension = so_ext
 
-class CCompiler:
+class CCompiler(object):
     """Abstract base class to define the interface that must be implemented
     by real compiler classes.  Also has some utility methods used by
     several compiler classes.
diff --git a/src/distutils2/compiler/msvc9compiler.py b/src/distutils2/compiler/msvc9compiler.py
--- a/src/distutils2/compiler/msvc9compiler.py
+++ b/src/distutils2/compiler/msvc9compiler.py
@@ -50,7 +50,7 @@
     'win-ia64' : 'ia64',
 }
 
-class Reg:
+class Reg(object):
     """Helper class to read values from the registry
     """
 
@@ -112,7 +112,7 @@
         return s
     convert_mbcs = staticmethod(convert_mbcs)
 
-class MacroExpander:
+class MacroExpander(object):
 
     def __init__(self, version):
         self.macros = {}
diff --git a/src/distutils2/compiler/msvccompiler.py b/src/distutils2/compiler/msvccompiler.py
--- a/src/distutils2/compiler/msvccompiler.py
+++ b/src/distutils2/compiler/msvccompiler.py
@@ -104,7 +104,7 @@
             pass
     return s
 
-class MacroExpander:
+class MacroExpander(object):
 
     def __init__(self, version):
         self.macros = {}
diff --git a/src/distutils2/extension.py b/src/distutils2/extension.py
--- a/src/distutils2/extension.py
+++ b/src/distutils2/extension.py
@@ -23,7 +23,7 @@
 # import that large-ish module (indirectly, through distutils.core) in
 # order to do anything.
 
-class Extension:
+class Extension(object):
     """Just a collection of attributes that describes an extension
     module and everything needed to build it (hopefully in a portable
     way, but there are hooks that let you be as unportable as you need).
diff --git a/src/distutils2/fancy_getopt.py b/src/distutils2/fancy_getopt.py
--- a/src/distutils2/fancy_getopt.py
+++ b/src/distutils2/fancy_getopt.py
@@ -30,7 +30,7 @@
 # (for use as attributes of some object).
 longopt_xlate = string.maketrans('-', '_')
 
-class FancyGetopt:
+class FancyGetopt(object):
     """Wrapper around the standard 'getopt()' module that provides some
     handy extra functionality:
       * short and long options are tied together
@@ -473,7 +473,7 @@
     return string.translate(opt, longopt_xlate)
 
 
-class OptionDummy:
+class OptionDummy(object):
     """Dummy class just used as a place to hold command-line option
     values as instance attributes."""
 
diff --git a/src/distutils2/log.py b/src/distutils2/log.py
--- a/src/distutils2/log.py
+++ b/src/distutils2/log.py
@@ -11,7 +11,7 @@
 
 import sys
 
-class Log:
+class Log(object):
 
     def __init__(self, threshold=WARN):
         self.threshold = threshold
diff --git a/src/distutils2/mkpkg.py b/src/distutils2/mkpkg.py
--- a/src/distutils2/mkpkg.py
+++ b/src/distutils2/mkpkg.py
@@ -675,7 +675,7 @@
 troveDict = buildTroveDict(troveList)
 
 
-class SetupClass:
+class SetupClass(object):
     def __init__(self):
         self.config = None
         self.classifierDict = {}
diff --git a/src/distutils2/tests/__init__.py b/src/distutils2/tests/__init__.py
--- a/src/distutils2/tests/__init__.py
+++ b/src/distutils2/tests/__init__.py
@@ -45,7 +45,7 @@
     """Test failed."""
 
 
-class BasicTestRunner:
+class BasicTestRunner(object):
     def run(self, test):
         result = unittest.TestResult()
         test(result)

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul  3 16:19:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 03 Jul 2010 16:19:06 +0200
Subject: [Python-checkins] distutils2: Tweak a docstring
Message-ID: <E1OV3Yg-0002Jt-F8@dinsdale.python.org>

tarek.ziade pushed 0f564dcc007a to distutils2:

http://hg.python.org/distutils2/rev/0f564dcc007a
changeset:   237:0f564dcc007a
user:        ?ric Araujo <merwok at netwok.org>
date:        Sat Jun 26 12:43:07 2010 +0200
summary:     Tweak a docstring
files:       src/distutils2/core.py

diff --git a/src/distutils2/core.py b/src/distutils2/core.py
--- a/src/distutils2/core.py
+++ b/src/distutils2/core.py
@@ -2,8 +2,9 @@
 
 The only module that needs to be imported to use the Distutils; provides
 the 'setup' function (which is to be called from the setup script).  Also
-indirectly provides the Distribution and Command classes, although they are
-really defined in distutils2.dist and distutils2.cmd.
+exports useful classes so that setup scripts can import them from here
+although they are really defined in other modules: Distribution, Command,
+PyPIRCommand, Extension.
 """
 
 __revision__ = "$Id: core.py 77704 2010-01-23 09:23:15Z tarek.ziade $"

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul  3 16:19:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 03 Jul 2010 16:19:06 +0200
Subject: [Python-checkins] distutils2: Fix some test classes
Message-ID: <E1OV3Yg-0002Ji-Cg@dinsdale.python.org>

tarek.ziade pushed 8c7fee19ab4d to distutils2:

http://hg.python.org/distutils2/rev/8c7fee19ab4d
changeset:   234:8c7fee19ab4d
user:        ?ric Araujo <merwok at netwok.org>
date:        Sat Jun 26 01:14:57 2010 +0200
summary:     Fix some test classes
files:       src/distutils2/_backport/tests/test_pkgutil.py, src/distutils2/tests/test_depgraph.py

diff --git a/src/distutils2/_backport/tests/test_pkgutil.py b/src/distutils2/_backport/tests/test_pkgutil.py
--- a/src/distutils2/_backport/tests/test_pkgutil.py
+++ b/src/distutils2/_backport/tests/test_pkgutil.py
@@ -1,6 +1,5 @@
 # -*- coding: utf-8 -*-
 """Tests for PEP 376 pkgutil functionality"""
-import unittest2
 import sys
 import os
 import csv
@@ -14,6 +13,7 @@
     from md5 import md5
 
 from test.test_support import run_unittest, TESTFN
+from distutils2.tests.support import unittest
 
 from distutils2._backport import pkgutil
 
@@ -23,13 +23,15 @@
 # TODO Add a test for absolute pathed RECORD items (e.g. /etc/myapp/config.ini)
 
 # Adapted from Python 2.7's trunk
-class TestPkgUtilData(unittest2.TestCase):
+class TestPkgUtilData(unittest.TestCase):
 
     def setUp(self):
+        super(TestPkgUtilData, self).setUp()
         self.dirname = tempfile.mkdtemp()
         sys.path.insert(0, self.dirname)
 
     def tearDown(self):
+        super(TestPkgUtilData, self).tearDown()
         del sys.path[0]
         shutil.rmtree(self.dirname)
 
@@ -91,7 +93,7 @@
         del sys.modules[pkg]
 
 # Adapted from Python 2.7's trunk
-class TestPkgUtilPEP302(unittest2.TestCase):
+class TestPkgUtilPEP302(unittest.TestCase):
 
     class MyTestLoader(object):
         def load_module(self, fullname):
@@ -113,9 +115,11 @@
             return TestPkgUtilPEP302.MyTestLoader()
 
     def setUp(self):
+        super(TestPkgUtilPEP302, self).setUp()
         sys.meta_path.insert(0, self.MyTestImporter())
 
     def tearDown(self):
+        super(TestPkgUtilPEP302, self).setUp()
         del sys.meta_path[0]
 
     def test_getdata_pep302(self):
@@ -134,10 +138,11 @@
         del sys.modules['foo']
 
 
-class TestPkgUtilDistribution(unittest2.TestCase):
+class TestPkgUtilDistribution(unittest.TestCase):
     """Tests the pkgutil.Distribution class"""
 
     def setUp(self):
+        super(TestPkgUtilDistribution, self).setUp()
         self.fake_dists_path = os.path.abspath(
             os.path.join(os.path.dirname(__file__), 'fake_dists'))
 
@@ -182,6 +187,7 @@
             self.records[distinfo_dir] = dict(record_data)
 
     def tearDown(self):
+        super(TestPkgUtilDistribution, self).setUp()
         self.records = None
         for distinfo_dir in self.distinfo_dirs:
             record_file = os.path.join(distinfo_dir, 'RECORD')
@@ -288,10 +294,11 @@
         self.assertEqual(sorted(found), sorted(distinfo_record_paths))
 
 
-class TestPkgUtilPEP376(unittest2.TestCase):
+class TestPkgUtilPEP376(unittest.TestCase):
     """Tests for the new functionality added in PEP 376."""
 
     def setUp(self):
+        super(TestPkgUtilPEP376, self).setUp()
         # Setup the path environment with our fake distributions
         current_path = os.path.abspath(os.path.dirname(__file__))
         self.sys_path = sys.path[:]
@@ -299,6 +306,7 @@
         sys.path[0:0] = [self.fake_dists_path]
 
     def tearDown(self):
+        super(TestPkgUtilPEP376, self).setUp()
         sys.path[:] = self.sys_path
 
     def test_distinfo_dirname(self):
@@ -527,8 +535,8 @@
 
 
 def test_suite():
-    suite = unittest2.TestSuite()
-    testcase_loader = unittest2.loader.defaultTestLoader.loadTestsFromTestCase
+    suite = unittest.TestSuite()
+    testcase_loader = unittest.loader.defaultTestLoader.loadTestsFromTestCase
     suite.addTest(testcase_loader(TestPkgUtilData))
     suite.addTest(testcase_loader(TestPkgUtilDistribution))
     suite.addTest(testcase_loader(TestPkgUtilPEP302))
diff --git a/src/distutils2/tests/test_depgraph.py b/src/distutils2/tests/test_depgraph.py
--- a/src/distutils2/tests/test_depgraph.py
+++ b/src/distutils2/tests/test_depgraph.py
@@ -27,7 +27,7 @@
         self.assertListEqual(sorted(l1), sorted(l2))
 
     def setUp(self):
-        super(unittest.TestCase, self).setUp()
+        super(DepGraphTestCase, self).setUp()
         path = os.path.join(os.path.dirname(__file__), '..', '_backport',
                             'tests', 'fake_dists')
         path = os.path.abspath(path)
@@ -177,7 +177,7 @@
         self.checkLists(matches, expected)
 
     def tearDown(self):
-        super(unittest.TestCase, self).tearDown()
+        super(DepGraphTestCase, self).tearDown()
         sys.path = self.sys_path
 
 def test_suite():

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul  3 16:19:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 03 Jul 2010 16:19:06 +0200
Subject: [Python-checkins] distutils2: Fix my hasty copy-paste
Message-ID: <E1OV3Yg-0002K2-G2@dinsdale.python.org>

tarek.ziade pushed e9f471d1a4f6 to distutils2:

http://hg.python.org/distutils2/rev/e9f471d1a4f6
changeset:   239:e9f471d1a4f6
user:        ?ric Araujo <merwok at netwok.org>
date:        Sun Jun 27 02:19:23 2010 +0200
summary:     Fix my hasty copy-paste
files:       src/distutils2/_backport/tests/test_pkgutil.py

diff --git a/src/distutils2/_backport/tests/test_pkgutil.py b/src/distutils2/_backport/tests/test_pkgutil.py
--- a/src/distutils2/_backport/tests/test_pkgutil.py
+++ b/src/distutils2/_backport/tests/test_pkgutil.py
@@ -119,8 +119,8 @@
         sys.meta_path.insert(0, self.MyTestImporter())
 
     def tearDown(self):
-        super(TestPkgUtilPEP302, self).setUp()
         del sys.meta_path[0]
+        super(TestPkgUtilPEP302, self).tearDown()
 
     def test_getdata_pep302(self):
         # Use a dummy importer/loader
@@ -187,11 +187,11 @@
             self.records[distinfo_dir] = dict(record_data)
 
     def tearDown(self):
-        super(TestPkgUtilDistribution, self).setUp()
         self.records = None
         for distinfo_dir in self.distinfo_dirs:
             record_file = os.path.join(distinfo_dir, 'RECORD')
             open(record_file, 'w').close()
+        super(TestPkgUtilDistribution, self).tearDown()
 
     def test_instantiation(self):
         """Test the Distribution class's instantiation provides us with usable
@@ -306,8 +306,8 @@
         sys.path[0:0] = [self.fake_dists_path]
 
     def tearDown(self):
-        super(TestPkgUtilPEP376, self).setUp()
         sys.path[:] = self.sys_path
+        super(TestPkgUtilPEP376, self).tearDown()
 
     def test_distinfo_dirname(self):
         """Given a name and a version, we expect the distinfo_dirname function

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul  3 16:19:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 03 Jul 2010 16:19:06 +0200
Subject: [Python-checkins] distutils2: Rename yet another ?distutils?
Message-ID: <E1OV3Yg-0002K9-Gq@dinsdale.python.org>

tarek.ziade pushed f2be51417dad to distutils2:

http://hg.python.org/distutils2/rev/f2be51417dad
changeset:   241:f2be51417dad
user:        ?ric Araujo <merwok at netwok.org>
date:        Sun Jun 27 02:22:14 2010 +0200
summary:     Rename yet another ?distutils?
files:       src/distutils2/util.py

diff --git a/src/distutils2/util.py b/src/distutils2/util.py
--- a/src/distutils2/util.py
+++ b/src/distutils2/util.py
@@ -363,7 +363,7 @@
                 script = open(script_name, "w")
 
             script.write("""\
-from distutils.util import byte_compile
+from distutils2.util import byte_compile
 files = [
 """)
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul  3 16:19:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 03 Jul 2010 16:19:06 +0200
Subject: [Python-checkins] distutils2: Remove costly shortcut in __init__
	and improve the	docstring
Message-ID: <E1OV3Yg-0002Jx-FW@dinsdale.python.org>

tarek.ziade pushed 80d665209636 to distutils2:

http://hg.python.org/distutils2/rev/80d665209636
changeset:   238:80d665209636
user:        ?ric Araujo <merwok at netwok.org>
date:        Sun Jun 27 02:15:52 2010 +0200
summary:     Remove costly shortcut in __init__ and improve the docstring
files:       src/distutils2/__init__.py

diff --git a/src/distutils2/__init__.py b/src/distutils2/__init__.py
--- a/src/distutils2/__init__.py
+++ b/src/distutils2/__init__.py
@@ -1,16 +1,20 @@
 """distutils
 
-The main package for the Python Module Distribution Utilities.  Normally
-used from a setup script as
+The main package for the Python Distribution Utilities 2.  Setup
+scripts should import the setup function from distutils2.core:
 
-   from distutils.core import setup
+    from distutils2.core import setup
 
-   setup (...)
+    setup(name=..., version=..., ...)
+
+Third-party tools can use parts of Distutils2 as building blocks
+without causing the other modules to be imported:
+
+    import distutils2.version
+    import distutils2.pypi.simple
+    import distutils2.tests.pypi_server
 """
-__all__ = ['__version__', 'setup']
+__all__ = ['__version__']
 
 __revision__ = "$Id: __init__.py 78020 2010-02-06 16:37:32Z benjamin.peterson $"
 __version__ = "1.0a2"
-
-from distutils2.core import setup
-

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul  3 16:19:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 03 Jul 2010 16:19:06 +0200
Subject: [Python-checkins] distutils2: Make unittest2 import error go to
	stderr again (2.4 to	3.x-compatible)
Message-ID: <E1OV3Yg-0002Kf-Le@dinsdale.python.org>

tarek.ziade pushed 216e2cb8d125 to distutils2:

http://hg.python.org/distutils2/rev/216e2cb8d125
changeset:   248:216e2cb8d125
user:        ?ric Araujo <merwok at netwok.org>
date:        Wed Jun 30 18:47:33 2010 +0200
summary:     Make unittest2 import error go to stderr again (2.4 to 3.x-compatible)
files:       src/runtests.py

diff --git a/src/runtests.py b/src/runtests.py
--- a/src/runtests.py
+++ b/src/runtests.py
@@ -28,7 +28,7 @@
     try:
         from distutils2.tests.support import unittest
     except ImportError:
-        print('Error: You have to install unittest2')
+        sys.stderr.write('Error: You have to install unittest2')
         sys.exit(1)
 
     sys.exit(test_main())

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul  3 16:19:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 03 Jul 2010 16:19:06 +0200
Subject: [Python-checkins] distutils2: Catch logging output in test_metadata
Message-ID: <E1OV3Yg-0002KY-Kk@dinsdale.python.org>

tarek.ziade pushed d0e993b9e892 to distutils2:

http://hg.python.org/distutils2/rev/d0e993b9e892
changeset:   246:d0e993b9e892
user:        ?ric Araujo <merwok at netwok.org>
date:        Wed Jun 30 18:14:57 2010 +0200
summary:     Catch logging output in test_metadata
files:       src/distutils2/tests/test_metadata.py

diff --git a/src/distutils2/tests/test_metadata.py b/src/distutils2/tests/test_metadata.py
--- a/src/distutils2/tests/test_metadata.py
+++ b/src/distutils2/tests/test_metadata.py
@@ -5,9 +5,9 @@
 
 from distutils2.metadata import (DistributionMetadata, _interpret,
                                  PKG_INFO_PREFERRED_VERSION)
-from distutils2.tests.support import unittest
+from distutils2.tests.support import unittest, LoggingSilencer
 
-class DistributionMetadataTestCase(unittest.TestCase):
+class DistributionMetadataTestCase(LoggingSilencer, unittest.TestCase):
 
 
     def test_interpret(self):

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul  3 16:19:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 03 Jul 2010 16:19:06 +0200
Subject: [Python-checkins] distutils2: s/klass/cls/g for PEP 8 and PEP 20
	compliance
Message-ID: <E1OV3Yg-0002Jq-Ee@dinsdale.python.org>

tarek.ziade pushed b1766b6037b9 to distutils2:

http://hg.python.org/distutils2/rev/b1766b6037b9
changeset:   236:b1766b6037b9
user:        ?ric Araujo <merwok at netwok.org>
date:        Sat Jun 26 12:37:06 2010 +0200
summary:     s/klass/cls/g for PEP 8 and PEP 20 compliance
files:       src/distutils2/compiler/ccompiler.py, src/distutils2/core.py, src/distutils2/dist.py, src/distutils2/tests/test_dist.py

diff --git a/src/distutils2/compiler/ccompiler.py b/src/distutils2/compiler/ccompiler.py
--- a/src/distutils2/compiler/ccompiler.py
+++ b/src/distutils2/compiler/ccompiler.py
@@ -1048,7 +1048,7 @@
         module_name = "distutils2.compiler." + module_name
         __import__ (module_name)
         module = sys.modules[module_name]
-        klass = vars(module)[class_name]
+        cls = vars(module)[class_name]
     except ImportError:
         raise DistutilsModuleError, \
               "can't compile C/C++ code: unable to load module '%s'" % \
@@ -1061,7 +1061,7 @@
     # XXX The None is necessary to preserve backwards compatibility
     # with classes that expect verbose to be the first positional
     # argument.
-    return klass(None, dry_run, force)
+    return cls(None, dry_run, force)
 
 
 def gen_preprocess_options(macros, include_dirs):
diff --git a/src/distutils2/core.py b/src/distutils2/core.py
--- a/src/distutils2/core.py
+++ b/src/distutils2/core.py
@@ -94,11 +94,7 @@
 
     # Determine the distribution class -- either caller-supplied or
     # our Distribution (see below).
-    klass = attrs.get('distclass')
-    if klass:
-        del attrs['distclass']
-    else:
-        klass = Distribution
+    distclass = attrs.pop('distclass', Distribution)
 
     if 'script_name' not in attrs:
         attrs['script_name'] = os.path.basename(sys.argv[0])
@@ -108,7 +104,7 @@
     # Create the Distribution instance, using the remaining arguments
     # (ie. everything except distclass) to initialize it
     try:
-        _setup_distribution = dist = klass(attrs)
+        _setup_distribution = dist = distclass(attrs)
     except DistutilsSetupError, msg:
         if 'name' in attrs:
             raise SystemExit, "error in %s setup command: %s" % \
diff --git a/src/distutils2/dist.py b/src/distutils2/dist.py
--- a/src/distutils2/dist.py
+++ b/src/distutils2/dist.py
@@ -627,16 +627,16 @@
 
         for command in self.commands:
             if isinstance(command, type) and issubclass(command, Command):
-                klass = command
+                cls = command
             else:
-                klass = self.get_command_class(command)
-            if (hasattr(klass, 'help_options') and
-                isinstance(klass.help_options, list)):
-                parser.set_option_table(klass.user_options +
-                                        fix_help_options(klass.help_options))
+                cls = self.get_command_class(command)
+            if (hasattr(cls, 'help_options') and
+                isinstance(cls.help_options, list)):
+                parser.set_option_table(cls.user_options +
+                                        fix_help_options(cls.help_options))
             else:
-                parser.set_option_table(klass.user_options)
-            parser.print_help("Options for '%s' command:" % klass.__name__)
+                parser.set_option_table(cls.user_options)
+            parser.print_help("Options for '%s' command:" % cls.__name__)
             print('')
 
         print(gen_usage(self.script_name))
@@ -688,11 +688,11 @@
         print(header + ":")
 
         for cmd in commands:
-            klass = self.cmdclass.get(cmd)
-            if not klass:
-                klass = self.get_command_class(cmd)
+            cls = self.cmdclass.get(cmd)
+            if not cls:
+                cls = self.get_command_class(cmd)
             try:
-                description = klass.description
+                description = cls.description
             except AttributeError:
                 description = "(no description available)"
 
@@ -754,11 +754,11 @@
 
         rv = []
         for cmd in (std_commands + extra_commands):
-            klass = self.cmdclass.get(cmd)
-            if not klass:
-                klass = self.get_command_class(cmd)
+            cls = self.cmdclass.get(cmd)
+            if not cls:
+                cls = self.get_command_class(cmd)
             try:
-                description = klass.description
+                description = cls.description
             except AttributeError:
                 description = "(no description available)"
             rv.append((cmd, description))
@@ -790,13 +790,13 @@
         Raises DistutilsModuleError if the expected module could not be
         found, or if that module does not define the expected class.
         """
-        klass = self.cmdclass.get(command)
-        if klass:
-            return klass
+        cls = self.cmdclass.get(command)
+        if cls:
+            return cls
 
         for pkgname in self.get_command_packages():
             module_name = "%s.%s" % (pkgname, command)
-            klass_name = command
+            class_name = command
 
             try:
                 __import__ (module_name)
@@ -805,14 +805,14 @@
                 continue
 
             try:
-                klass = getattr(module, klass_name)
+                cls = getattr(module, class_name)
             except AttributeError:
                 raise DistutilsModuleError, \
                       "invalid command '%s' (no class '%s' in module '%s')" \
-                      % (command, klass_name, module_name)
+                      % (command, class_name, module_name)
 
-            self.cmdclass[command] = klass
-            return klass
+            self.cmdclass[command] = cls
+            return cls
 
         raise DistutilsModuleError("invalid command '%s'" % command)
 
@@ -828,8 +828,8 @@
             log.debug("Distribution.get_command_obj(): " \
                       "creating '%s' command object" % command)
 
-            klass = self.get_command_class(command)
-            cmd_obj = self.command_obj[command] = klass(self)
+            cls = self.get_command_class(command)
+            cmd_obj = self.command_obj[command] = cls(self)
             self.have_run[command] = 0
 
             # Set any options that were supplied in config files
diff --git a/src/distutils2/tests/test_dist.py b/src/distutils2/tests/test_dist.py
--- a/src/distutils2/tests/test_dist.py
+++ b/src/distutils2/tests/test_dist.py
@@ -129,13 +129,13 @@
         # Check DistributionMetadata handling of Unicode fields
         tmp_dir = self.mkdtemp()
         my_file = os.path.join(tmp_dir, 'f')
-        klass = Distribution
+        cls = Distribution
 
-        dist = klass(attrs={'author': u'Mister Caf??',
-                            'name': 'my.package',
-                            'maintainer': u'Caf?? Junior',
-                            'summary': u'Caf?? torr??fi??',
-                            'description': u'H??h??h??'})
+        dist = cls(attrs={'author': u'Mister Caf??',
+                          'name': 'my.package',
+                          'maintainer': u'Caf?? Junior',
+                          'summary': u'Caf?? torr??fi??',
+                          'description': u'H??h??h??'})
 
 
         # let's make sure the file can be written
@@ -144,11 +144,11 @@
         dist.metadata.write_file(open(my_file, 'w'))
 
         # regular ascii is of course always usable
-        dist = klass(attrs={'author': 'Mister Cafe',
-                            'name': 'my.package',
-                            'maintainer': 'Cafe Junior',
-                            'summary': 'Cafe torrefie',
-                            'description': 'Hehehe'})
+        dist = cls(attrs={'author': 'Mister Cafe',
+                          'name': 'my.package',
+                          'maintainer': 'Cafe Junior',
+                          'summary': 'Cafe torrefie',
+                          'description': 'Hehehe'})
 
         my_file2 = os.path.join(tmp_dir, 'f2')
         dist.metadata.write_file(open(my_file, 'w'))
@@ -156,7 +156,7 @@
     def test_empty_options(self):
         # an empty options dictionary should not stay in the
         # list of attributes
-        klass = Distribution
+        cls = Distribution
 
         # catching warnings
         warns = []
@@ -166,11 +166,11 @@
         old_warn = warnings.warn
         warnings.warn = _warn
         try:
-            dist = klass(attrs={'author': 'xxx',
-                                'name': 'xxx',
-                                'version': 'xxx',
-                                'url': 'xxxx',
-                                'options': {}})
+            dist = cls(attrs={'author': 'xxx',
+                              'name': 'xxx',
+                              'version': 'xxx',
+                              'url': 'xxxx',
+                              'options': {}})
         finally:
             warnings.warn = old_warn
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul  3 16:19:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 03 Jul 2010 16:19:06 +0200
Subject: [Python-checkins] distutils2: Remove some pre-2.4 compatibility code
Message-ID: <E1OV3Yg-0002KH-IF@dinsdale.python.org>

tarek.ziade pushed ca50d669c9f4 to distutils2:

http://hg.python.org/distutils2/rev/ca50d669c9f4
changeset:   243:ca50d669c9f4
user:        ?ric Araujo <merwok at netwok.org>
date:        Sun Jun 27 02:28:37 2010 +0200
summary:     Remove some pre-2.4 compatibility code
files:       src/distutils2/util.py

diff --git a/src/distutils2/util.py b/src/distutils2/util.py
--- a/src/distutils2/util.py
+++ b/src/distutils2/util.py
@@ -349,12 +349,8 @@
     # "Indirect" byte-compilation: write a temporary script and then
     # run it with the appropriate flags.
     if not direct:
-        try:
-            from tempfile import mkstemp
-            (script_fd, script_name) = mkstemp(".py")
-        except ImportError:
-            from tempfile import mktemp
-            (script_fd, script_name) = None, mktemp(".py")
+        from tempfile import mkstemp
+        script_fd, script_name = mkstemp(".py")
         log.info("writing byte-compilation script '%s'", script_name)
         if not dry_run:
             if script_fd is not None:

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul  3 16:19:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 03 Jul 2010 16:19:06 +0200
Subject: [Python-checkins] distutils2: Add shortcut to util.find_packages in
	core
Message-ID: <E1OV3Yg-0002K6-GQ@dinsdale.python.org>

tarek.ziade pushed d235a3dcb404 to distutils2:

http://hg.python.org/distutils2/rev/d235a3dcb404
changeset:   240:d235a3dcb404
user:        ?ric Araujo <merwok at netwok.org>
date:        Sun Jun 27 02:20:58 2010 +0200
summary:     Add shortcut to util.find_packages in core
files:       src/distutils2/core.py

diff --git a/src/distutils2/core.py b/src/distutils2/core.py
--- a/src/distutils2/core.py
+++ b/src/distutils2/core.py
@@ -4,7 +4,7 @@
 the 'setup' function (which is to be called from the setup script).  Also
 exports useful classes so that setup scripts can import them from here
 although they are really defined in other modules: Distribution, Command,
-PyPIRCommand, Extension.
+PyPIRCommand, Extension, find_packages.
 """
 
 __revision__ = "$Id: core.py 77704 2010-01-23 09:23:15Z tarek.ziade $"
@@ -21,6 +21,7 @@
 from distutils2.command.cmd import Command
 from distutils2.config import PyPIRCCommand
 from distutils2.extension import Extension
+from distutils2.util import find_packages
 
 # This is a barebones help message generated displayed when the user
 # runs the setup script with no arguments at all.  More useful help

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul  3 16:19:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 03 Jul 2010 16:19:06 +0200
Subject: [Python-checkins] distutils2: Fix one indented line and add a
	marker for later grepping
Message-ID: <E1OV3Yg-0002Kj-MC@dinsdale.python.org>

tarek.ziade pushed c64e0613ed52 to distutils2:

http://hg.python.org/distutils2/rev/c64e0613ed52
changeset:   249:c64e0613ed52
user:        ?ric Araujo <merwok at netwok.org>
date:        Wed Jun 30 18:50:53 2010 +0200
summary:     Fix one indented line and add a marker for later grepping
files:       src/runtests.py

diff --git a/src/runtests.py b/src/runtests.py
--- a/src/runtests.py
+++ b/src/runtests.py
@@ -8,8 +8,7 @@
     import distutils2.tests
     from distutils2.tests import run_unittest, reap_children, TestFailed
     from distutils2._backport.tests import test_suite as btest_suite
-    # just supporting -q right now
-    # to enable detailed/quiet output
+    # XXX just supporting -q right now to enable detailed/quiet output
     if len(sys.argv) > 1:
         verbose = sys.argv[-1] != '-q'
     else:
@@ -17,7 +16,7 @@
     try:
         try:
             run_unittest([distutils2.tests.test_suite(), btest_suite()],
-                    verbose_=verbose)
+                         verbose_=verbose)
             return 0
         except TestFailed:
             return 1

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul  3 16:19:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 03 Jul 2010 16:19:06 +0200
Subject: [Python-checkins] distutils2: Wrap I/O operations in try/except
	blocks to ensure	handles are closed
Message-ID: <E1OV3Yg-0002KO-Jf@dinsdale.python.org>

tarek.ziade pushed c61f4eff8a25 to distutils2:

http://hg.python.org/distutils2/rev/c61f4eff8a25
changeset:   244:c61f4eff8a25
user:        ?ric Araujo <merwok at netwok.org>
date:        Sun Jun 27 02:42:50 2010 +0200
summary:     Wrap I/O operations in try/except blocks to ensure handles are closed
files:       src/distutils2/_backport/sysconfig.py, src/distutils2/_backport/tarfile.py, src/distutils2/_backport/tests/test_pkgutil.py, src/distutils2/compiler/ccompiler.py, src/distutils2/compiler/emxccompiler.py, src/distutils2/mkpkg.py, src/distutils2/tests/test_build_py.py, src/distutils2/tests/test_build_scripts.py, src/distutils2/tests/test_dist.py, src/distutils2/tests/test_install_scripts.py, src/distutils2/tests/test_metadata.py, src/distutils2/tests/test_msvc9compiler.py, src/distutils2/util.py

diff --git a/src/distutils2/_backport/sysconfig.py b/src/distutils2/_backport/sysconfig.py
--- a/src/distutils2/_backport/sysconfig.py
+++ b/src/distutils2/_backport/sysconfig.py
@@ -580,10 +580,11 @@
                 # behaviour.
                 pass
             else:
-                m = re.search(
-                        r'<key>ProductUserVisibleVersion</key>\s*' +
-                        r'<string>(.*?)</string>', f.read())
-                f.close()
+                try:
+                    m = re.search(r'<key>ProductUserVisibleVersion</key>\s*'
+                                  r'<string>(.*?)</string>', f.read())
+                finally:
+                    f.close()
                 if m is not None:
                     macrelease = '.'.join(m.group(1).split('.')[:2])
                 # else: fall back to the default behaviour
diff --git a/src/distutils2/_backport/tarfile.py b/src/distutils2/_backport/tarfile.py
--- a/src/distutils2/_backport/tarfile.py
+++ b/src/distutils2/_backport/tarfile.py
@@ -2003,8 +2003,10 @@
         # Append the tar header and data to the archive.
         if tarinfo.isreg():
             f = bltn_open(name, "rb")
-            self.addfile(tarinfo, f)
-            f.close()
+            try:
+                self.addfile(tarinfo, f)
+            finally:
+                f.close()
 
         elif tarinfo.isdir():
             self.addfile(tarinfo)
@@ -2214,9 +2216,11 @@
         """
         source = self.extractfile(tarinfo)
         target = bltn_open(targetpath, "wb")
-        copyfileobj(source, target)
-        source.close()
-        target.close()
+        try:
+            copyfileobj(source, target)
+        finally:
+            source.close()
+            target.close()
 
     def makeunknown(self, tarinfo, targetpath):
         """Make a file from a TarInfo object with an unknown type
@@ -2564,8 +2568,10 @@
        are able to handle, else return False.
     """
     try:
-        t = open(name)
-        t.close()
+        try:
+            t = open(name)
+        finally:
+            t.close()
         return True
     except TarError:
         return False
diff --git a/src/distutils2/_backport/tests/test_pkgutil.py b/src/distutils2/_backport/tests/test_pkgutil.py
--- a/src/distutils2/_backport/tests/test_pkgutil.py
+++ b/src/distutils2/_backport/tests/test_pkgutil.py
@@ -46,15 +46,22 @@
         os.mkdir(package_dir)
         # Empty init.py
         f = open(os.path.join(package_dir, '__init__.py'), "wb")
-        f.close()
+        try:
+            pass
+        finally:
+            f.close()
         # Resource files, res.txt, sub/res.txt
         f = open(os.path.join(package_dir, 'res.txt'), "wb")
-        f.write(RESOURCE_DATA)
-        f.close()
+        try:
+            f.write(RESOURCE_DATA)
+        finally:
+            f.close()
         os.mkdir(os.path.join(package_dir, 'sub'))
         f = open(os.path.join(package_dir, 'sub', 'res.txt'), "wb")
-        f.write(RESOURCE_DATA)
-        f.close()
+        try:
+            f.write(RESOURCE_DATA)
+        finally:
+            f.close()
 
         # Check we can read the resources
         res1 = pkgutil.get_data(pkg, 'res.txt')
@@ -74,13 +81,14 @@
         # Make a package with some resources
         zip_file = os.path.join(self.dirname, zip)
         z = zipfile.ZipFile(zip_file, 'w')
-
-        # Empty init.py
-        z.writestr(pkg + '/__init__.py', "")
-        # Resource files, res.txt, sub/res.txt
-        z.writestr(pkg + '/res.txt', RESOURCE_DATA)
-        z.writestr(pkg + '/sub/res.txt', RESOURCE_DATA)
-        z.close()
+        try:
+            # Empty init.py
+            z.writestr(pkg + '/__init__.py', "")
+            # Resource files, res.txt, sub/res.txt
+            z.writestr(pkg + '/res.txt', RESOURCE_DATA)
+            z.writestr(pkg + '/sub/res.txt', RESOURCE_DATA)
+        finally:
+            z.close()
 
         # Check we can read the resources
         sys.path.insert(0, zip_file)
diff --git a/src/distutils2/compiler/ccompiler.py b/src/distutils2/compiler/ccompiler.py
--- a/src/distutils2/compiler/ccompiler.py
+++ b/src/distutils2/compiler/ccompiler.py
@@ -800,14 +800,16 @@
             library_dirs = []
         fd, fname = tempfile.mkstemp(".c", funcname, text=True)
         f = os.fdopen(fd, "w")
-        for incl in includes:
-            f.write("""#include "%s"\n""" % incl)
-        f.write("""\
+        try:
+            for incl in includes:
+                f.write("""#include "%s"\n""" % incl)
+            f.write("""\
 main (int argc, char **argv) {
     %s();
 }
 """ % funcname)
-        f.close()
+        finally:
+            f.close()
         try:
             objects = self.compile([fname], include_dirs=include_dirs)
         except CompileError:
diff --git a/src/distutils2/compiler/emxccompiler.py b/src/distutils2/compiler/emxccompiler.py
--- a/src/distutils2/compiler/emxccompiler.py
+++ b/src/distutils2/compiler/emxccompiler.py
@@ -272,8 +272,10 @@
         # It would probably better to read single lines to search.
         # But we do this only once, and it is fast enough
         f = open(fn)
-        s = f.read()
-        f.close()
+        try:
+            s = f.read()
+        finally:
+            f.close()
 
     except IOError, exc:
         # if we can't read this file, we cannot say it is wrong
diff --git a/src/distutils2/mkpkg.py b/src/distutils2/mkpkg.py
--- a/src/distutils2/mkpkg.py
+++ b/src/distutils2/mkpkg.py
@@ -717,14 +717,16 @@
 
     def inspectFile(self, path):
         fp = open(path, 'r')
-        for line in [ fp.readline() for x in range(10) ]:
-            m = re.match(r'^#!.*python((?P<major>\d)(\.\d+)?)?$', line)
-            if m:
-                if m.group('major') == '3':
-                    self.classifierDict['Programming Language :: Python :: 3'] = 1
-                else:
-                    self.classifierDict['Programming Language :: Python :: 2'] = 1
-        fp.close()
+        try:
+            for line in [ fp.readline() for x in range(10) ]:
+                m = re.match(r'^#!.*python((?P<major>\d)(\.\d+)?)?$', line)
+                if m:
+                    if m.group('major') == '3':
+                        self.classifierDict['Programming Language :: Python :: 3'] = 1
+                    else:
+                        self.classifierDict['Programming Language :: Python :: 2'] = 1
+        finally:
+            fp.close()
 
 
     def inspectDirectory(self):
@@ -885,38 +887,33 @@
         if os.path.exists('setup.py'): shutil.move('setup.py', 'setup.py.old')
 
         fp = open('setup.py', 'w')
-        fp.write('#!/usr/bin/env python\n\n')
-        fp.write('from distutils2.core import setup\n\n')
+        try:
+            fp.write('#!/usr/bin/env python\n\n')
+            fp.write('from distutils2.core import setup\n\n')
+            fp.write('setup(name=%s,\n' % repr(self.setupData['name']))
+            fp.write('      version=%s,\n' % repr(self.setupData['version']))
+            fp.write('      description=%s,\n'
+                    % repr(self.setupData['description']))
+            fp.write('      author=%s,\n' % repr(self.setupData['author']))
+            fp.write('      author_email=%s,\n'
+                    % repr(self.setupData['author_email']))
+            if self.setupData['url']:
+                fp.write('      url=%s,\n' % repr(self.setupData['url']))
+            if self.setupData['classifier']:
+                fp.write('      classifier=[\n')
+                for classifier in sorted(self.setupData['classifier'].keys()):
+                    fp.write('            %s,\n' % repr(classifier))
+                fp.write('         ],\n')
+            if self.setupData['packages']:
+                fp.write('      packages=%s,\n'
+                        % repr(self._dotted_packages(self.setupData['packages'])))
+                fp.write('      package_dir=%s,\n'
+                        % repr(self.setupData['packages']))
+            fp.write('      #scripts=[\'path/to/script\']\n')
 
-        fp.write('from sys import version\n')
-        fp.write('if version < \'2.2.3\':\n')
-        fp.write('    from distutils2.dist import DistributionMetadata\n')
-        fp.write('    DistributionMetadata.classifier = None\n')
-        fp.write('    DistributionMetadata.download_url = None\n')
-
-        fp.write('setup(name = %s,\n' % repr(self.setupData['name']))
-        fp.write('        version = %s,\n' % repr(self.setupData['version']))
-        fp.write('        description = %s,\n'
-                % repr(self.setupData['description']))
-        fp.write('        author = %s,\n' % repr(self.setupData['author']))
-        fp.write('        author_email = %s,\n'
-                % repr(self.setupData['author_email']))
-        if self.setupData['url']:
-            fp.write('        url = %s,\n' % repr(self.setupData['url']))
-        if self.setupData['classifier']:
-            fp.write('        classifier = [\n')
-            for classifier in sorted(self.setupData['classifier'].keys()):
-                fp.write('              %s,\n' % repr(classifier))
-            fp.write('           ],\n')
-        if self.setupData['packages']:
-            fp.write('        packages = %s,\n'
-                    % repr(self._dotted_packages(self.setupData['packages'])))
-            fp.write('        package_dir = %s,\n'
-                    % repr(self.setupData['packages']))
-        fp.write('        #scripts = [\'path/to/script\']\n')
-
-        fp.write('        )\n')
-        fp.close()
+            fp.write('      )\n')
+        finally:
+            fp.close()
         os.chmod('setup.py', 0755)
 
         print 'Wrote "setup.py".'
diff --git a/src/distutils2/tests/test_build_py.py b/src/distutils2/tests/test_build_py.py
--- a/src/distutils2/tests/test_build_py.py
+++ b/src/distutils2/tests/test_build_py.py
@@ -19,11 +19,15 @@
     def test_package_data(self):
         sources = self.mkdtemp()
         f = open(os.path.join(sources, "__init__.py"), "w")
-        f.write("# Pretend this is a package.")
-        f.close()
+        try:
+            f.write("# Pretend this is a package.")
+        finally:
+            f.close()
         f = open(os.path.join(sources, "README.txt"), "w")
-        f.write("Info about this package")
-        f.close()
+        try:
+            f.write("Info about this package")
+        finally:
+            f.close()
 
         destination = self.mkdtemp()
 
diff --git a/src/distutils2/tests/test_build_scripts.py b/src/distutils2/tests/test_build_scripts.py
--- a/src/distutils2/tests/test_build_scripts.py
+++ b/src/distutils2/tests/test_build_scripts.py
@@ -74,8 +74,10 @@
 
     def write_script(self, dir, name, text):
         f = open(os.path.join(dir, name), "w")
-        f.write(text)
-        f.close()
+        try:
+            f.write(text)
+        finally:
+            f.close()
 
     def test_version_int(self):
         source = self.mkdtemp()
diff --git a/src/distutils2/tests/test_dist.py b/src/distutils2/tests/test_dist.py
--- a/src/distutils2/tests/test_dist.py
+++ b/src/distutils2/tests/test_dist.py
@@ -340,8 +340,10 @@
         temp_dir = self.mkdtemp()
         user_filename = os.path.join(temp_dir, user_filename)
         f = open(user_filename, 'w')
-        f.write('.')
-        f.close()
+        try:
+            f.write('.')
+        finally:
+            f.close()
 
         try:
             dist = Distribution()
diff --git a/src/distutils2/tests/test_install_scripts.py b/src/distutils2/tests/test_install_scripts.py
--- a/src/distutils2/tests/test_install_scripts.py
+++ b/src/distutils2/tests/test_install_scripts.py
@@ -42,8 +42,10 @@
         def write_script(name, text):
             expected.append(name)
             f = open(os.path.join(source, name), "w")
-            f.write(text)
-            f.close()
+            try:
+                f.write(text)
+            finally:
+                f.close()
 
         write_script("script1.py", ("#! /usr/bin/env python2.3\n"
                                     "# bogus script w/ Python sh-bang\n"
diff --git a/src/distutils2/tests/test_metadata.py b/src/distutils2/tests/test_metadata.py
--- a/src/distutils2/tests/test_metadata.py
+++ b/src/distutils2/tests/test_metadata.py
@@ -64,9 +64,12 @@
         res.seek(0)
         res = res.read()
         f = open(PKG_INFO)
-        wanted = f.read()
+        try:
+            # XXX this is not used
+            wanted = f.read()
+        finally:
+            f.close()
         self.assertTrue('Keywords: keyring,password,crypt' in res)
-        f.close()
 
     def test_metadata_markers(self):
         # see if we can be platform-aware
diff --git a/src/distutils2/tests/test_msvc9compiler.py b/src/distutils2/tests/test_msvc9compiler.py
--- a/src/distutils2/tests/test_msvc9compiler.py
+++ b/src/distutils2/tests/test_msvc9compiler.py
@@ -116,17 +116,21 @@
         tempdir = self.mkdtemp()
         manifest = os.path.join(tempdir, 'manifest')
         f = open(manifest, 'w')
-        f.write(_MANIFEST)
-        f.close()
+        try:
+            f.write(_MANIFEST)
+        finally:
+            f.close()
 
         compiler = MSVCCompiler()
         compiler._remove_visual_c_ref(manifest)
 
         # see what we got
         f = open(manifest)
-        # removing trailing spaces
-        content = '\n'.join([line.rstrip() for line in f.readlines()])
-        f.close()
+        try:
+            # removing trailing spaces
+            content = '\n'.join([line.rstrip() for line in f.readlines()])
+        finally:
+            f.close()
 
         # makes sure the manifest was properly cleaned
         self.assertEqual(content, _CLEANED_MANIFEST)
diff --git a/src/distutils2/util.py b/src/distutils2/util.py
--- a/src/distutils2/util.py
+++ b/src/distutils2/util.py
@@ -358,34 +358,36 @@
             else:
                 script = open(script_name, "w")
 
-            script.write("""\
+            try:
+                script.write("""\
 from distutils2.util import byte_compile
 files = [
 """)
 
-            # XXX would be nice to write absolute filenames, just for
-            # safety's sake (script should be more robust in the face of
-            # chdir'ing before running it).  But this requires abspath'ing
-            # 'prefix' as well, and that breaks the hack in build_lib's
-            # 'byte_compile()' method that carefully tacks on a trailing
-            # slash (os.sep really) to make sure the prefix here is "just
-            # right".  This whole prefix business is rather delicate -- the
-            # problem is that it's really a directory, but I'm treating it
-            # as a dumb string, so trailing slashes and so forth matter.
+                # XXX would be nice to write absolute filenames, just for
+                # safety's sake (script should be more robust in the face of
+                # chdir'ing before running it).  But this requires abspath'ing
+                # 'prefix' as well, and that breaks the hack in build_lib's
+                # 'byte_compile()' method that carefully tacks on a trailing
+                # slash (os.sep really) to make sure the prefix here is "just
+                # right".  This whole prefix business is rather delicate -- the
+                # problem is that it's really a directory, but I'm treating it
+                # as a dumb string, so trailing slashes and so forth matter.
 
-            #py_files = map(os.path.abspath, py_files)
-            #if prefix:
-            #    prefix = os.path.abspath(prefix)
+                #py_files = map(os.path.abspath, py_files)
+                #if prefix:
+                #    prefix = os.path.abspath(prefix)
 
-            script.write(",\n".join(map(repr, py_files)) + "]\n")
-            script.write("""
+                script.write(",\n".join(map(repr, py_files)) + "]\n")
+                script.write("""
 byte_compile(files, optimize=%r, force=%r,
              prefix=%r, base_dir=%r,
              verbose=%r, dry_run=0,
              direct=1)
 """ % (optimize, force, prefix, base_dir, verbose))
 
-            script.close()
+            finally:
+                script.close()
 
         cmd = [sys.executable, script_name]
         if optimize == 1:
@@ -534,10 +536,12 @@
     """Create a file with the specified name and write 'contents' (a
     sequence of strings without line terminators) to it.
     """
-    f = open(filename, "w")
-    for line in contents:
-        f.write(line + "\n")
-    f.close()
+    try:
+        f = open(filename, "w")
+        for line in contents:
+            f.write(line + "\n")
+    finally:
+        f.close()
 
 def _is_package(path):
     """Returns True if path is a package (a dir with an __init__ file."""

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul  3 16:19:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 03 Jul 2010 16:19:06 +0200
Subject: [Python-checkins] distutils2: Allow importing
	distutils2.test.support without loading a	ton of modules
Message-ID: <E1OV3Yg-0002Kb-LD@dinsdale.python.org>

tarek.ziade pushed 5886d4219749 to distutils2:

http://hg.python.org/distutils2/rev/5886d4219749
changeset:   247:5886d4219749
user:        ?ric Araujo <merwok at netwok.org>
date:        Wed Jun 30 18:28:17 2010 +0200
summary:     Allow importing distutils2.test.support without loading a ton of modules
files:       src/distutils2/tests/support.py

diff --git a/src/distutils2/tests/support.py b/src/distutils2/tests/support.py
--- a/src/distutils2/tests/support.py
+++ b/src/distutils2/tests/support.py
@@ -14,7 +14,6 @@
 
 from distutils2 import log
 from distutils2.log import DEBUG, INFO, WARN, ERROR, FATAL
-from distutils2.core import Distribution
 
 if sys.version_info >= (2, 7):
     # improved unittest package from 2.7's standard library
@@ -105,6 +104,7 @@
         It returns the package directory and the distribution
         instance.
         """
+        from distutils2.dist import Distribution
         tmp_dir = self.mkdtemp()
         pkg_dir = os.path.join(tmp_dir, pkg_name)
         os.mkdir(pkg_dir)

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul  3 16:19:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 03 Jul 2010 16:19:06 +0200
Subject: [Python-checkins] distutils2: Improve imports indentation
Message-ID: <E1OV3Yg-0002KE-Hm@dinsdale.python.org>

tarek.ziade pushed aae4336cb757 to distutils2:

http://hg.python.org/distutils2/rev/aae4336cb757
changeset:   242:aae4336cb757
user:        ?ric Araujo <merwok at netwok.org>
date:        Sun Jun 27 02:26:57 2010 +0200
summary:     Improve imports indentation
files:       src/distutils2/command/sdist.py, src/distutils2/compiler/bcppcompiler.py, src/distutils2/compiler/ccompiler.py, src/distutils2/compiler/emxccompiler.py, src/distutils2/compiler/msvc9compiler.py, src/distutils2/compiler/msvccompiler.py, src/distutils2/compiler/unixccompiler.py, src/distutils2/core.py, src/distutils2/dist.py

diff --git a/src/distutils2/command/sdist.py b/src/distutils2/command/sdist.py
--- a/src/distutils2/command/sdist.py
+++ b/src/distutils2/command/sdist.py
@@ -20,7 +20,7 @@
 from distutils2.core import Command
 from distutils2 import util
 from distutils2.errors import (DistutilsPlatformError, DistutilsOptionError,
-                              DistutilsTemplateError)
+                               DistutilsTemplateError)
 from distutils2.manifest import Manifest
 from distutils2 import log
 from distutils2.util import convert_path, newer
diff --git a/src/distutils2/compiler/bcppcompiler.py b/src/distutils2/compiler/bcppcompiler.py
--- a/src/distutils2/compiler/bcppcompiler.py
+++ b/src/distutils2/compiler/bcppcompiler.py
@@ -16,7 +16,7 @@
 import os
 
 from distutils2.errors import (DistutilsExecError, CompileError, LibError,
-                              LinkError, UnknownFileError)
+                               LinkError, UnknownFileError)
 from distutils2.compiler.ccompiler import CCompiler, gen_preprocess_options
 from distutils2.file_util import write_file
 from distutils2.dep_util import newer
diff --git a/src/distutils2/compiler/ccompiler.py b/src/distutils2/compiler/ccompiler.py
--- a/src/distutils2/compiler/ccompiler.py
+++ b/src/distutils2/compiler/ccompiler.py
@@ -10,7 +10,7 @@
 import re
 
 from distutils2.errors import (CompileError, LinkError, UnknownFileError,
-                              DistutilsPlatformError, DistutilsModuleError)
+                               DistutilsPlatformError, DistutilsModuleError)
 from distutils2.spawn import spawn
 from distutils2.util import split_quoted, execute, newer_group
 from distutils2 import log
diff --git a/src/distutils2/compiler/emxccompiler.py b/src/distutils2/compiler/emxccompiler.py
--- a/src/distutils2/compiler/emxccompiler.py
+++ b/src/distutils2/compiler/emxccompiler.py
@@ -25,9 +25,8 @@
 from warnings import warn
 
 from distutils2.compiler.unixccompiler import UnixCCompiler
-from distutils2.util import write_file
 from distutils2.errors import DistutilsExecError, CompileError, UnknownFileError
-from distutils2.util import get_compiler_versions
+from distutils2.util import get_compiler_versions, write_file
 
 class EMXCCompiler (UnixCCompiler):
 
diff --git a/src/distutils2/compiler/msvc9compiler.py b/src/distutils2/compiler/msvc9compiler.py
--- a/src/distutils2/compiler/msvc9compiler.py
+++ b/src/distutils2/compiler/msvc9compiler.py
@@ -20,7 +20,7 @@
 import re
 
 from distutils2.errors import (DistutilsExecError, DistutilsPlatformError,
-                              CompileError, LibError, LinkError)
+                               CompileError, LibError, LinkError)
 from distutils2.compiler.ccompiler import CCompiler, gen_lib_options
 from distutils2 import log
 from distutils2.util import get_platform
diff --git a/src/distutils2/compiler/msvccompiler.py b/src/distutils2/compiler/msvccompiler.py
--- a/src/distutils2/compiler/msvccompiler.py
+++ b/src/distutils2/compiler/msvccompiler.py
@@ -15,7 +15,7 @@
 import string
 
 from distutils2.errors import (DistutilsExecError, DistutilsPlatformError,
-                              CompileError, LibError, LinkError)
+                               CompileError, LibError, LinkError)
 from distutils2.compiler.ccompiler import CCompiler, gen_lib_options
 from distutils2 import log
 
diff --git a/src/distutils2/compiler/unixccompiler.py b/src/distutils2/compiler/unixccompiler.py
--- a/src/distutils2/compiler/unixccompiler.py
+++ b/src/distutils2/compiler/unixccompiler.py
@@ -19,10 +19,10 @@
 from types import StringType, NoneType
 
 from distutils2.util import newer
-from distutils2.compiler.ccompiler import \
-     CCompiler, gen_preprocess_options, gen_lib_options
-from distutils2.errors import \
-     DistutilsExecError, CompileError, LibError, LinkError
+from distutils2.compiler.ccompiler import (CCompiler, gen_preprocess_options,
+                                           gen_lib_options)
+from distutils2.errors import (DistutilsExecError, CompileError,
+                               LibError, LinkError)
 from distutils2 import log
 
 try:
diff --git a/src/distutils2/core.py b/src/distutils2/core.py
--- a/src/distutils2/core.py
+++ b/src/distutils2/core.py
@@ -13,7 +13,7 @@
 import os
 
 from distutils2.errors import (DistutilsSetupError, DistutilsArgError,
-                              DistutilsError, CCompilerError)
+                               DistutilsError, CCompilerError)
 from distutils2.util import grok_environment_error
 
 # Mainly import these so setup scripts can "from distutils2.core import" them.
diff --git a/src/distutils2/dist.py b/src/distutils2/dist.py
--- a/src/distutils2/dist.py
+++ b/src/distutils2/dist.py
@@ -16,7 +16,7 @@
 from ConfigParser import RawConfigParser
 
 from distutils2.errors import (DistutilsOptionError, DistutilsArgError,
-                              DistutilsModuleError, DistutilsClassError)
+                               DistutilsModuleError, DistutilsClassError)
 from distutils2.fancy_getopt import FancyGetopt, translate_longopt
 from distutils2.util import check_environ, strtobool
 from distutils2 import log

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul  3 16:19:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 03 Jul 2010 16:19:06 +0200
Subject: [Python-checkins] distutils2: Remove redundant conversions
Message-ID: <E1OV3Yg-0002Kn-NK@dinsdale.python.org>

tarek.ziade pushed b2ec3efa9496 to distutils2:

http://hg.python.org/distutils2/rev/b2ec3efa9496
changeset:   250:b2ec3efa9496
user:        ?ric Araujo <merwok at netwok.org>
date:        Wed Jun 30 19:00:34 2010 +0200
summary:     Remove redundant conversions
files:       src/distutils2/dist.py, src/distutils2/log.py, src/distutils2/metadata.py, src/distutils2/tests/support.py

diff --git a/src/distutils2/dist.py b/src/distutils2/dist.py
--- a/src/distutils2/dist.py
+++ b/src/distutils2/dist.py
@@ -250,7 +250,7 @@
                 elif hasattr(self, key):
                     setattr(self, key, val)
                 else:
-                    msg = "Unknown distribution option: %s" % repr(key)
+                    msg = "Unknown distribution option: %r" % key
                     if warnings is not None:
                         warnings.warn(msg)
                     else:
diff --git a/src/distutils2/log.py b/src/distutils2/log.py
--- a/src/distutils2/log.py
+++ b/src/distutils2/log.py
@@ -18,7 +18,7 @@
 
     def _log(self, level, msg, args):
         if level not in (DEBUG, INFO, WARN, ERROR, FATAL):
-            raise ValueError('%s wrong log level' % str(level))
+            raise ValueError('%s wrong log level' % level)
 
         if level >= self.threshold:
             if args:
diff --git a/src/distutils2/metadata.py b/src/distutils2/metadata.py
--- a/src/distutils2/metadata.py
+++ b/src/distutils2/metadata.py
@@ -517,7 +517,7 @@
             raise NameError(value)
 
     def _nonsense_op(self):
-        msg = 'This operation is not supported : "%s"' % str(self)
+        msg = 'This operation is not supported : "%s"' % self
         raise SyntaxError(msg)
 
     def __call__(self):
diff --git a/src/distutils2/tests/support.py b/src/distutils2/tests/support.py
--- a/src/distutils2/tests/support.py
+++ b/src/distutils2/tests/support.py
@@ -41,7 +41,7 @@
 
     def _log(self, level, msg, args):
         if level not in (DEBUG, INFO, WARN, ERROR, FATAL):
-            raise ValueError('%s wrong log level' % str(level))
+            raise ValueError('%s wrong log level' % level)
         self.logs.append((level, msg, args))
 
     def get_logs(self, *levels):

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul  3 16:19:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 03 Jul 2010 16:19:06 +0200
Subject: [Python-checkins] distutils2: Get more info from tests and also run
	with Python 2.7.
Message-ID: <E1OV3Yg-0002KT-KF@dinsdale.python.org>

tarek.ziade pushed 408a80b82cd0 to distutils2:

http://hg.python.org/distutils2/rev/408a80b82cd0
changeset:   245:408a80b82cd0
user:        ?ric Araujo <merwok at netwok.org>
date:        Wed Jun 30 18:10:13 2010 +0200
summary:     Get more info from tests and also run with Python 2.7.
files:       src/tests.sh

diff --git a/src/tests.sh b/src/tests.sh
--- a/src/tests.sh
+++ b/src/tests.sh
@@ -1,6 +1,6 @@
 #!/bin/sh
 echo -n "Running tests for Python 2.4... "
-python2.4 runtests.py -q > /dev/null 2> /dev/null
+python2.4 -Wd runtests.py -q 2> /dev/null
 if [ $? -ne 0 ];then
     echo "Failed"
     exit 1
@@ -9,7 +9,7 @@
 fi
 
 echo -n "Running tests for Python 2.5... "
-python2.5 runtests.py -q > /dev/null 2> /dev/null
+python2.5 -Wd runtests.py -q 2> /dev/null
 if [ $? -ne 0 ];then
     echo "Failed"
     exit 1
@@ -18,7 +18,7 @@
 fi
 
 echo -n "Running tests for Python 2.6... "
-python2.6 runtests.py -q > /dev/null 2> /dev/null
+python2.6 -Wd -bb -3 runtests.py -q 2> /dev/null
 if [ $? -ne 0 ];then
     echo "Failed"
     exit 1
@@ -26,3 +26,11 @@
     echo "Success"
 fi
 
+echo -n "Running tests for Python 2.7... "
+python2.7 -Wd -bb -3 runtests.py -q 2> /dev/null
+if [ $? -ne 0 ];then
+    echo "Failed"
+    exit 1
+else
+    echo "Success"
+fi

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul  3 16:19:08 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 03 Jul 2010 16:19:08 +0200
Subject: [Python-checkins] distutils2: now spawn takes the environment when
	required (like byte	compilation here, was
Message-ID: <E1OV3Yi-0002LP-5V@dinsdale.python.org>

tarek.ziade pushed 32559ebbec5c to distutils2:

http://hg.python.org/distutils2/rev/32559ebbec5c
changeset:   251:32559ebbec5c
tag:         tip
user:        Tarek Ziade <tarek at ziade.org>
date:        Sat Jul 03 16:18:51 2010 +0200
summary:     now spawn takes the environment when required (like byte compilation here, was failing)
files:       src/distutils2/spawn.py, src/distutils2/util.py

diff --git a/src/distutils2/spawn.py b/src/distutils2/spawn.py
--- a/src/distutils2/spawn.py
+++ b/src/distutils2/spawn.py
@@ -14,7 +14,7 @@
 from distutils2.errors import DistutilsPlatformError, DistutilsExecError
 from distutils2 import log
 
-def spawn(cmd, search_path=1, verbose=0, dry_run=0):
+def spawn(cmd, search_path=1, verbose=0, dry_run=0, env=None):
     """Run another program, specified as a command list 'cmd', in a new process.
 
     'cmd' is just the argument list for the new process, ie.
@@ -27,15 +27,18 @@
     must be the exact path to the executable.  If 'dry_run' is true,
     the command will not actually be run.
 
+    If 'env' is given, it's a environment dictionary used for the execution
+    environment.
+
     Raise DistutilsExecError if running the program fails in any way; just
     return on success.
     """
     if os.name == 'posix':
-        _spawn_posix(cmd, search_path, dry_run=dry_run)
+        _spawn_posix(cmd, search_path, dry_run=dry_run, env=env)
     elif os.name == 'nt':
-        _spawn_nt(cmd, search_path, dry_run=dry_run)
+        _spawn_nt(cmd, search_path, dry_run=dry_run, env=env)
     elif os.name == 'os2':
-        _spawn_os2(cmd, search_path, dry_run=dry_run)
+        _spawn_os2(cmd, search_path, dry_run=dry_run, env=env)
     else:
         raise DistutilsPlatformError, \
               "don't know how to spawn programs on platform '%s'" % os.name
@@ -56,7 +59,7 @@
             args[i] = '"%s"' % arg
     return args
 
-def _spawn_nt(cmd, search_path=1, verbose=0, dry_run=0):
+def _spawn_nt(cmd, search_path=1, verbose=0, dry_run=0, env=None):
     executable = cmd[0]
     cmd = _nt_quote_args(cmd)
     if search_path:
@@ -66,7 +69,11 @@
     if not dry_run:
         # spawn for NT requires a full path to the .exe
         try:
-            rc = os.spawnv(os.P_WAIT, executable, cmd)
+            if env is None:
+                rc = os.spawnv(os.P_WAIT, executable, cmd)
+            else:
+                rc = os.spawnve(os.P_WAIT, executable, cmd, env)
+
         except OSError, exc:
             # this seems to happen when the command isn't found
             raise DistutilsExecError, \
@@ -76,7 +83,7 @@
             raise DistutilsExecError, \
                   "command '%s' failed with exit status %d" % (cmd[0], rc)
 
-def _spawn_os2(cmd, search_path=1, verbose=0, dry_run=0):
+def _spawn_os2(cmd, search_path=1, verbose=0, dry_run=0, env=None):
     executable = cmd[0]
     if search_path:
         # either we find one or it stays the same
@@ -85,7 +92,11 @@
     if not dry_run:
         # spawnv for OS/2 EMX requires a full path to the .exe
         try:
-            rc = os.spawnv(os.P_WAIT, executable, cmd)
+            if env is None:
+                rc = os.spawnv(os.P_WAIT, executable, cmd)
+            else:
+                rc = os.spawnve(os.P_WAIT, executable, cmd, env)
+
         except OSError, exc:
             # this seems to happen when the command isn't found
             raise DistutilsExecError, \
@@ -97,16 +108,24 @@
                   "command '%s' failed with exit status %d" % (cmd[0], rc)
 
 
-def _spawn_posix(cmd, search_path=1, verbose=0, dry_run=0):
+def _spawn_posix(cmd, search_path=1, verbose=0, dry_run=0, env=None):
     log.info(' '.join(cmd))
     if dry_run:
         return
-    exec_fn = search_path and os.execvp or os.execv
+
+    if env is None:
+        exec_fn = search_path and os.execvp or os.execv
+    else:
+        exec_fn = search_path and os.execvpe or os.execve
+
     pid = os.fork()
 
     if pid == 0:  # in the child
         try:
-            exec_fn(cmd[0], cmd)
+            if env is None:
+                exec_fn(cmd[0], cmd)
+            else:
+                exec_fn(cmd[0], cmd, env)
         except OSError, e:
             sys.stderr.write("unable to execute %s: %s\n" %
                              (cmd[0], e.strerror))
diff --git a/src/distutils2/util.py b/src/distutils2/util.py
--- a/src/distutils2/util.py
+++ b/src/distutils2/util.py
@@ -6,6 +6,7 @@
 __revision__ = "$Id: util.py 77761 2010-01-26 22:46:15Z tarek.ziade $"
 
 import sys, os, string, re
+from copy import copy
 from fnmatch import fnmatchcase
 
 from distutils2.errors import (DistutilsPlatformError, DistutilsFileError,
@@ -394,9 +395,14 @@
             cmd.insert(1, "-O")
         elif optimize == 2:
             cmd.insert(1, "-OO")
-        spawn(cmd, dry_run=dry_run)
-        execute(os.remove, (script_name,), "removing %s" % script_name,
-                dry_run=dry_run)
+
+        env = copy(os.environ)
+        env['PYTHONPATH'] = ':'.join(sys.path)
+        try:
+            spawn(cmd, dry_run=dry_run, env=env)
+        finally:
+            execute(os.remove, (script_name,), "removing %s" % script_name,
+                    dry_run=dry_run)
 
     # "Direct" byte-compilation: use the py_compile module to compile
     # right here, right now.  Note that the script generated in indirect

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul  3 16:46:24 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sat,  3 Jul 2010 16:46:24 +0200 (CEST)
Subject: [Python-checkins] r82505 - sandbox/trunk/release/release.py
Message-ID: <20100703144624.2ABD7EE9D0@mail.python.org>

Author: benjamin.peterson
Date: Sat Jul  3 16:46:24 2010
New Revision: 82505

Log:
warn if dist already exists

Modified:
   sandbox/trunk/release/release.py

Modified: sandbox/trunk/release/release.py
==============================================================================
--- sandbox/trunk/release/release.py	(original)
+++ sandbox/trunk/release/release.py	Sat Jul  3 16:46:24 2010
@@ -206,7 +206,9 @@
     try:
         os.mkdir('dist')
     except OSError:
-        if not os.path.isdir('dist'):
+        if os.path.isdir('dist'):
+            print('WARNING: dist already exists', file=sys.stderr)
+        else:
             error('dist/ is not a directory')
     else:
         print('created dist directory')

From python-checkins at python.org  Sat Jul  3 16:51:25 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sat,  3 Jul 2010 16:51:25 +0200 (CEST)
Subject: [Python-checkins] r82506 - in python/branches/release27-maint:
	Include/patchlevel.h Lib/distutils/__init__.py
	Lib/idlelib/idlever.py Misc/NEWS Misc/RPM/python-2.7.spec README
Message-ID: <20100703145125.A8DB8EE9B3@mail.python.org>

Author: benjamin.peterson
Date: Sat Jul  3 16:51:25 2010
New Revision: 82506

Log:
prepare for 2.7.1

Modified:
   python/branches/release27-maint/Include/patchlevel.h
   python/branches/release27-maint/Lib/distutils/__init__.py
   python/branches/release27-maint/Lib/idlelib/idlever.py
   python/branches/release27-maint/Misc/NEWS
   python/branches/release27-maint/Misc/RPM/python-2.7.spec
   python/branches/release27-maint/README

Modified: python/branches/release27-maint/Include/patchlevel.h
==============================================================================
--- python/branches/release27-maint/Include/patchlevel.h	(original)
+++ python/branches/release27-maint/Include/patchlevel.h	Sat Jul  3 16:51:25 2010
@@ -22,12 +22,12 @@
 /*--start constants--*/
 #define PY_MAJOR_VERSION	2
 #define PY_MINOR_VERSION	7
-#define PY_MICRO_VERSION	0
-#define PY_RELEASE_LEVEL	PY_RELEASE_LEVEL_FINAL
+#define PY_MICRO_VERSION	1
+#define PY_RELEASE_LEVEL	PY_RELEASE_LEVEL_ALPHA
 #define PY_RELEASE_SERIAL	0
 
 /* Version as a string */
-#define PY_VERSION      	"2.7"
+#define PY_VERSION      	"2.7.1a0"
 /*--end constants--*/
 
 /* Subversion Revision number of this file (not of the repository) */

Modified: python/branches/release27-maint/Lib/distutils/__init__.py
==============================================================================
--- python/branches/release27-maint/Lib/distutils/__init__.py	(original)
+++ python/branches/release27-maint/Lib/distutils/__init__.py	Sat Jul  3 16:51:25 2010
@@ -15,5 +15,5 @@
 # Updated automatically by the Python release process.
 #
 #--start constants--
-__version__ = "2.7"
+__version__ = "2.7.1a0"
 #--end constants--

Modified: python/branches/release27-maint/Lib/idlelib/idlever.py
==============================================================================
--- python/branches/release27-maint/Lib/idlelib/idlever.py	(original)
+++ python/branches/release27-maint/Lib/idlelib/idlever.py	Sat Jul  3 16:51:25 2010
@@ -1 +1 @@
-IDLE_VERSION = "2.7"
+IDLE_VERSION = "2.7.1a0"

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Sat Jul  3 16:51:25 2010
@@ -4,6 +4,23 @@
 
 (editors: check NEWS.help for information about editing NEWS using ReST.)
 
+What's New in Python 2.7.1?
+===========================
+
+*Release date: XXXX-XX-XX*
+
+Core and Builtins
+-----------------
+
+
+Library
+-------
+
+
+Extension Modules
+-----------------
+
+
 What's New in Python 2.7?
 =========================
 

Modified: python/branches/release27-maint/Misc/RPM/python-2.7.spec
==============================================================================
--- python/branches/release27-maint/Misc/RPM/python-2.7.spec	(original)
+++ python/branches/release27-maint/Misc/RPM/python-2.7.spec	Sat Jul  3 16:51:25 2010
@@ -39,7 +39,7 @@
 
 %define name python
 #--start constants--
-%define version 2.7
+%define version 2.7.1a0
 %define libver 2.7
 #--end constants--
 %define release 1pydotorg

Modified: python/branches/release27-maint/README
==============================================================================
--- python/branches/release27-maint/README	(original)
+++ python/branches/release27-maint/README	Sat Jul  3 16:51:25 2010
@@ -1,5 +1,5 @@
-This is Python version 2.7
-===========================
+This is Python version 2.7.1 release candidate 1
+================================================
 
 Copyright (c) 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010
 Python Software Foundation.

From python-checkins at python.org  Sat Jul  3 17:09:20 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sat,  3 Jul 2010 17:09:20 +0200 (CEST)
Subject: [Python-checkins] r82507 - python/branches/py3k
Message-ID: <20100703150920.95347EE98F@mail.python.org>

Author: benjamin.peterson
Date: Sat Jul  3 17:09:20 2010
New Revision: 82507

Log:
Removed merge tracking for "svnmerge" for 
svn+ssh://pythondev at svn.python.org/python/branches/py3k/python/trunk


Modified:
   python/branches/py3k/   (props changed)

From python-checkins at python.org  Sat Jul  3 17:38:12 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sat,  3 Jul 2010 17:38:12 +0200 (CEST)
Subject: [Python-checkins] r82508 - python/branches/release27-maint
Message-ID: <20100703153812.990F3EE986@mail.python.org>

Author: benjamin.peterson
Date: Sat Jul  3 17:38:12 2010
New Revision: 82508

Log:
Initialized merge tracking via "svnmerge" with revisions "0-82507" from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k


Modified:
   python/branches/release27-maint/   (props changed)

From python-checkins at python.org  Sat Jul  3 19:45:27 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Sat,  3 Jul 2010 19:45:27 +0200 (CEST)
Subject: [Python-checkins] r82509 -
	sandbox/branches/py3k-datetime/datetime.py
Message-ID: <20100703174527.213EEEE983@mail.python.org>

Author: alexander.belopolsky
Date: Sat Jul  3 19:45:26 2010
New Revision: 82509

Log:
Use single underscore prefix for private attributes and methods.
Note that tests will fail if __setstate is renamed to _setstate.


Modified:
   sandbox/branches/py3k-datetime/datetime.py

Modified: sandbox/branches/py3k-datetime/datetime.py
==============================================================================
--- sandbox/branches/py3k-datetime/datetime.py	(original)
+++ sandbox/branches/py3k-datetime/datetime.py	Sat Jul  3 19:45:26 2010
@@ -316,7 +316,7 @@
     Representation: (days, seconds, microseconds).  Why?  Because I
     felt like it.
     """
-    __slots__ = '__days', '__seconds', '__microseconds'
+    __slots__ = '_days', '_seconds', '_microseconds'
 
     def __new__(cls, days=0, seconds=0, microseconds=0,
                 milliseconds=0, minutes=0, hours=0, weeks=0):
@@ -423,36 +423,36 @@
 
         self = object.__new__(cls)
 
-        self.__days = d
-        self.__seconds = s
-        self.__microseconds = us
+        self._days = d
+        self._seconds = s
+        self._microseconds = us
         if abs(d) > 999999999:
             raise OverflowError("timedelta # of days is too large: %d" % d)
 
         return self
 
     def __repr__(self):
-        if self.__microseconds:
+        if self._microseconds:
             return "%s(%d, %d, %d)" % ('datetime.' + self.__class__.__name__,
-                                       self.__days,
-                                       self.__seconds,
-                                       self.__microseconds)
-        if self.__seconds:
+                                       self._days,
+                                       self._seconds,
+                                       self._microseconds)
+        if self._seconds:
             return "%s(%d, %d)" % ('datetime.' + self.__class__.__name__,
-                                   self.__days,
-                                   self.__seconds)
-        return "%s(%d)" % ('datetime.' + self.__class__.__name__, self.__days)
+                                   self._days,
+                                   self._seconds)
+        return "%s(%d)" % ('datetime.' + self.__class__.__name__, self._days)
 
     def __str__(self):
-        mm, ss = divmod(self.__seconds, 60)
+        mm, ss = divmod(self._seconds, 60)
         hh, mm = divmod(mm, 60)
         s = "%d:%02d:%02d" % (hh, mm, ss)
-        if self.__days:
+        if self._days:
             def plural(n):
                 return n, abs(n) != 1 and "s" or ""
-            s = ("%d day%s, " % plural(self.__days)) + s
-        if self.__microseconds:
-            s = s + ".%06d" % self.__microseconds
+            s = ("%d day%s, " % plural(self._days)) + s
+        if self._microseconds:
+            s = s + ".%06d" % self._microseconds
         return s
 
     def total_seconds(self):
@@ -464,25 +464,25 @@
     @property
     def days(self):
         """days"""
-        return self.__days
+        return self._days
 
     @property
     def seconds(self):
         """seconds"""
-        return self.__seconds
+        return self._seconds
 
     @property
     def microseconds(self):
         """microseconds"""
-        return self.__microseconds
+        return self._microseconds
 
     def __add__(self, other):
         if isinstance(other, timedelta):
             # for CPython compatibility, we cannot use
             # our __class__ here, but need a real timedelta
-            return timedelta(self.__days + other.__days,
-                             self.__seconds + other.__seconds,
-                             self.__microseconds + other.__microseconds)
+            return timedelta(self._days + other._days,
+                             self._seconds + other._seconds,
+                             self._microseconds + other._microseconds)
         return NotImplemented
 
     __radd__ = __add__
@@ -500,15 +500,15 @@
     def __neg__(self):
             # for CPython compatibility, we cannot use
             # our __class__ here, but need a real timedelta
-            return timedelta(-self.__days,
-                             -self.__seconds,
-                             -self.__microseconds)
+            return timedelta(-self._days,
+                             -self._seconds,
+                             -self._microseconds)
 
     def __pos__(self):
         return self
 
     def __abs__(self):
-        if self.__days < 0:
+        if self._days < 0:
             return -self
         else:
             return self
@@ -517,9 +517,9 @@
         if isinstance(other, int):
             # for CPython compatibility, we cannot use
             # our __class__ here, but need a real timedelta
-            return timedelta(self.__days * other,
-                             self.__seconds * other,
-                             self.__microseconds * other)
+            return timedelta(self._days * other,
+                             self._seconds * other,
+                             self._microseconds * other)
         if isinstance(other, float):
             a, b = other.as_integer_ratio()
             return self * a / b
@@ -528,8 +528,8 @@
     __rmul__ = __mul__
 
     def _to_microseconds(self):
-        return ((self.__days * (24*3600) + self.__seconds) * 1000000 +
-                self.__microseconds)
+        return ((self._days * (24*3600) + self._seconds) * 1000000 +
+                self._microseconds)
 
     def __floordiv__(self, other):
         if not isinstance(other, (int, timedelta)):
@@ -569,59 +569,59 @@
 
     def __eq__(self, other):
         if isinstance(other, timedelta):
-            return self.__cmp(other) == 0
+            return self._cmp(other) == 0
         else:
             return False
 
     def __ne__(self, other):
         if isinstance(other, timedelta):
-            return self.__cmp(other) != 0
+            return self._cmp(other) != 0
         else:
             return True
 
     def __le__(self, other):
         if isinstance(other, timedelta):
-            return self.__cmp(other) <= 0
+            return self._cmp(other) <= 0
         else:
             _cmperror(self, other)
 
     def __lt__(self, other):
         if isinstance(other, timedelta):
-            return self.__cmp(other) < 0
+            return self._cmp(other) < 0
         else:
             _cmperror(self, other)
 
     def __ge__(self, other):
         if isinstance(other, timedelta):
-            return self.__cmp(other) >= 0
+            return self._cmp(other) >= 0
         else:
             _cmperror(self, other)
 
     def __gt__(self, other):
         if isinstance(other, timedelta):
-            return self.__cmp(other) > 0
+            return self._cmp(other) > 0
         else:
             _cmperror(self, other)
 
-    def __cmp(self, other):
+    def _cmp(self, other):
         assert isinstance(other, timedelta)
-        return _cmp(self.__getstate(), other.__getstate())
+        return _cmp(self._getstate(), other._getstate())
 
     def __hash__(self):
-        return hash(self.__getstate())
+        return hash(self._getstate())
 
     def __bool__(self):
-        return (self.__days != 0 or
-                self.__seconds != 0 or
-                self.__microseconds != 0)
+        return (self._days != 0 or
+                self._seconds != 0 or
+                self._microseconds != 0)
 
     # Pickle support.
 
-    def __getstate(self):
-        return (self.__days, self.__seconds, self.__microseconds)
+    def _getstate(self):
+        return (self._days, self._seconds, self._microseconds)
 
     def __reduce__(self):
-        return (self.__class__, self.__getstate())
+        return (self.__class__, self._getstate())
 
 timedelta.min = timedelta(-999999999)
 timedelta.max = timedelta(days=999999999, hours=23, minutes=59, seconds=59,
@@ -656,7 +656,7 @@
     Properties (readonly):
     year, month, day
     """
-    __slots__ = '__year', '__month', '__day'
+    __slots__ = '_year', '_month', '_day'
 
     def __new__(cls, year, month=None, day=None):
         """Constructor.
@@ -672,9 +672,9 @@
             return self
         _check_date_fields(year, month, day)
         self = object.__new__(cls)
-        self.__year = year
-        self.__month = month
-        self.__day = day
+        self._year = year
+        self._month = month
+        self._day = day
         return self
 
     # Additional constructors
@@ -706,9 +706,9 @@
     def __repr__(self):
         "Convert to formal string, for repr()."
         return "%s(%d, %d, %d)" % ('datetime.' + self.__class__.__name__,
-                                   self.__year,
-                                   self.__month,
-                                   self.__day)
+                                   self._year,
+                                   self._month,
+                                   self._day)
     # XXX These shouldn't depend on time.localtime(), because that
     # clips the usable dates to [1970 .. 2038).  At least ctime() is
     # easily done without using strftime() -- that's better too because
@@ -720,8 +720,8 @@
         weekday = self.toordinal() % 7 or 7
         return "%s %s %2d 00:00:00 %04d" % (
             _DAYNAMES[weekday],
-            _MONTHNAMES[self.__month],
-            self.__day, self.__year)
+            _MONTHNAMES[self._month],
+            self._day, self._year)
 
     def strftime(self, fmt):
         "Format using strftime()."
@@ -741,7 +741,7 @@
         - http://www.w3.org/TR/NOTE-datetime
         - http://www.cl.cam.ac.uk/~mgk25/iso-time.html
         """
-        return "%04d-%02d-%02d" % (self.__year, self.__month, self.__day)
+        return "%04d-%02d-%02d" % (self._year, self._month, self._day)
 
     __str__ = isoformat
 
@@ -749,23 +749,23 @@
     @property
     def year(self):
         """year (1-9999)"""
-        return self.__year
+        return self._year
 
     @property
     def month(self):
         """month (1-12)"""
-        return self.__month
+        return self._month
 
     @property
     def day(self):
         """day (1-31)"""
-        return self.__day
+        return self._day
 
     # Standard conversions, __cmp__, __hash__ (and helpers)
 
     def timetuple(self):
         "Return local time tuple compatible with time.localtime()."
-        return _build_struct_time(self.__year, self.__month, self.__day,
+        return _build_struct_time(self._year, self._month, self._day,
                                   0, 0, 0, -1)
 
     def toordinal(self):
@@ -774,16 +774,16 @@
         January 1 of year 1 is day 1.  Only the year, month and day values
         contribute to the result.
         """
-        return _ymd2ord(self.__year, self.__month, self.__day)
+        return _ymd2ord(self._year, self._month, self._day)
 
     def replace(self, year=None, month=None, day=None):
         """Return a new date with new values for the specified fields."""
         if year is None:
-            year = self.__year
+            year = self._year
         if month is None:
-            month = self.__month
+            month = self._month
         if day is None:
-            day = self.__day
+            day = self._day
         _check_date_fields(year, month, day)
         return date(year, month, day)
 
@@ -791,43 +791,43 @@
 
     def __eq__(self, other):
         if isinstance(other, date):
-            return self.__cmp(other) == 0
+            return self._cmp(other) == 0
         return NotImplemented
 
     def __ne__(self, other):
         if isinstance(other, date):
-            return self.__cmp(other) != 0
+            return self._cmp(other) != 0
         return NotImplemented
 
     def __le__(self, other):
         if isinstance(other, date):
-            return self.__cmp(other) <= 0
+            return self._cmp(other) <= 0
         return NotImplemented
 
     def __lt__(self, other):
         if isinstance(other, date):
-            return self.__cmp(other) < 0
+            return self._cmp(other) < 0
         return NotImplemented
 
     def __ge__(self, other):
         if isinstance(other, date):
-            return self.__cmp(other) >= 0
+            return self._cmp(other) >= 0
         return NotImplemented
 
     def __gt__(self, other):
         if isinstance(other, date):
-            return self.__cmp(other) > 0
+            return self._cmp(other) > 0
         return NotImplemented
 
-    def __cmp(self, other):
+    def _cmp(self, other):
         assert isinstance(other, date)
-        y, m, d = self.__year, self.__month, self.__day
-        y2, m2, d2 = other.__year, other.__month, other.__day
+        y, m, d = self._year, self._month, self._day
+        y2, m2, d2 = other._year, other._month, other._day
         return _cmp((y, m, d), (y2, m2, d2))
 
     def __hash__(self):
         "Hash."
-        return hash(self.__getstate())
+        return hash(self._getstate())
 
     # Computations
 
@@ -875,9 +875,9 @@
         ISO calendar algorithm taken from
         http://www.phys.uu.nl/~vgent/calendar/isocalendar.htm
         """
-        year = self.__year
+        year = self._year
         week1monday = _isoweek1monday(year)
-        today = _ymd2ord(self.__year, self.__month, self.__day)
+        today = _ymd2ord(self._year, self._month, self._day)
         # Internally, week and day have origin 0
         week, day = divmod(today - week1monday, 7)
         if week < 0:
@@ -892,18 +892,18 @@
 
     # Pickle support.
 
-    def __getstate(self):
-        yhi, ylo = divmod(self.__year, 256)
-        return bytes([yhi, ylo, self.__month, self.__day]),
+    def _getstate(self):
+        yhi, ylo = divmod(self._year, 256)
+        return bytes([yhi, ylo, self._month, self._day]),
 
     def __setstate(self, string):
         if len(string) != 4 or not (1 <= string[2] <= 12):
             raise TypeError("not enough arguments")
-        yhi, ylo, self.__month, self.__day = string
-        self.__year = yhi * 256 + ylo
+        yhi, ylo, self._month, self._day = string
+        self._year = yhi * 256 + ylo
 
     def __reduce__(self):
-        return (self.__class__, self.__getstate())
+        return (self.__class__, self._getstate())
 
 _date_class = date  # so functions w/ args named "date" can get at the class
 
@@ -1018,10 +1018,10 @@
             return self
         _check_tzinfo_arg(tzinfo)
         _check_time_fields(hour, minute, second, microsecond)
-        self.__hour = hour
-        self.__minute = minute
-        self.__second = second
-        self.__microsecond = microsecond
+        self._hour = hour
+        self._minute = minute
+        self._second = second
+        self._microsecond = microsecond
         self._tzinfo = tzinfo
         return self
 
@@ -1029,22 +1029,22 @@
     @property
     def hour(self):
         """hour (0-23)"""
-        return self.__hour
+        return self._hour
 
     @property
     def minute(self): 
         """minute (0-59)"""
-        return self.__minute
+        return self._minute
 
     @property
     def second(self):
         """second (0-59)"""
-        return self.__second
+        return self._second
     
     @property
     def microsecond(self):
         """microsecond (0-999999)"""
-        return self.__microsecond
+        return self._microsecond
 
     @property
     def tzinfo(self):
@@ -1057,41 +1057,41 @@
 
     def __eq__(self, other):
         if isinstance(other, time):
-            return self.__cmp(other) == 0
+            return self._cmp(other) == 0
         else:
             return False
 
     def __ne__(self, other):
         if isinstance(other, time):
-            return self.__cmp(other) != 0
+            return self._cmp(other) != 0
         else:
             return True
 
     def __le__(self, other):
         if isinstance(other, time):
-            return self.__cmp(other) <= 0
+            return self._cmp(other) <= 0
         else:
             _cmperror(self, other)
 
     def __lt__(self, other):
         if isinstance(other, time):
-            return self.__cmp(other) < 0
+            return self._cmp(other) < 0
         else:
             _cmperror(self, other)
 
     def __ge__(self, other):
         if isinstance(other, time):
-            return self.__cmp(other) >= 0
+            return self._cmp(other) >= 0
         else:
             _cmperror(self, other)
 
     def __gt__(self, other):
         if isinstance(other, time):
-            return self.__cmp(other) > 0
+            return self._cmp(other) > 0
         else:
             _cmperror(self, other)
 
-    def __cmp(self, other):
+    def _cmp(self, other):
         assert isinstance(other, time)
         mytz = self._tzinfo
         ottz = other._tzinfo
@@ -1105,22 +1105,22 @@
             base_compare = myoff == otoff
 
         if base_compare:
-            return _cmp((self.__hour, self.__minute, self.__second,
-                         self.__microsecond),
-                       (other.__hour, other.__minute, other.__second,
-                        other.__microsecond))
+            return _cmp((self._hour, self._minute, self._second,
+                         self._microsecond),
+                       (other._hour, other._minute, other._second,
+                        other._microsecond))
         if myoff is None or otoff is None:
             raise TypeError("cannot compare naive and aware times")
-        myhhmm = self.__hour * 60 + self.__minute - myoff
-        othhmm = other.__hour * 60 + other.__minute - otoff
-        return _cmp((myhhmm, self.__second, self.__microsecond),
-                    (othhmm, other.__second, other.__microsecond))
+        myhhmm = self._hour * 60 + self._minute - myoff
+        othhmm = other._hour * 60 + other._minute - otoff
+        return _cmp((myhhmm, self._second, self._microsecond),
+                    (othhmm, other._second, other._microsecond))
 
     def __hash__(self):
         """Hash."""
         tzoff = self._utcoffset()
         if not tzoff: # zero or None
-            return hash(self.__getstate()[0])
+            return hash(self._getstate()[0])
         h, m = divmod(self.hour * 60 + self.minute - tzoff, 60)
         if 0 <= h < 24:
             return hash(time(h, m, self.second, self.microsecond))
@@ -1144,14 +1144,14 @@
 
     def __repr__(self):
         """Convert to formal string, for repr()."""
-        if self.__microsecond != 0:
-            s = ", %d, %d" % (self.__second, self.__microsecond)
-        elif self.__second != 0:
-            s = ", %d" % self.__second
+        if self._microsecond != 0:
+            s = ", %d, %d" % (self._second, self._microsecond)
+        elif self._second != 0:
+            s = ", %d" % self._second
         else:
             s = ""
         s= "%s(%d, %d%s)" % ('datetime.' + self.__class__.__name__,
-                             self.__hour, self.__minute, s)
+                             self._hour, self._minute, s)
         if self._tzinfo is not None:
             assert s[-1:] == ")"
             s = s[:-1] + ", tzinfo=%r" % self._tzinfo + ")"
@@ -1163,8 +1163,8 @@
         This is 'HH:MM:SS.mmmmmm+zz:zz', or 'HH:MM:SS+zz:zz' if
         self.microsecond == 0.
         """
-        s = _format_time(self.__hour, self.__minute, self.__second,
-                         self.__microsecond)
+        s = _format_time(self._hour, self._minute, self._second,
+                         self._microsecond)
         tz = self._tzstr()
         if tz:
             s += tz
@@ -1179,7 +1179,7 @@
         # The year must be >= 1900 else Python's strftime implementation
         # can raise a bogus exception.
         timetuple = (1900, 1, 1,
-                     self.__hour, self.__minute, self.__second,
+                     self._hour, self._minute, self._second,
                      0, 1, -1)
         return _wrap_strftime(self, fmt, timetuple)
 
@@ -1256,10 +1256,10 @@
 
     # Pickle support.
 
-    def __getstate(self):
-        us2, us3 = divmod(self.__microsecond, 256)
+    def _getstate(self):
+        us2, us3 = divmod(self._microsecond, 256)
         us1, us2 = divmod(us2, 256)
-        basestate = bytes([self.__hour, self.__minute, self.__second,
+        basestate = bytes([self._hour, self._minute, self._second,
                            us1, us2, us3])
         if self._tzinfo is None:
             return (basestate,)
@@ -1269,13 +1269,13 @@
     def __setstate(self, string, tzinfo):
         if len(string) != 6 or string[0] >= 24:
             raise TypeError("an integer is required")
-        (self.__hour, self.__minute, self.__second,
+        (self._hour, self._minute, self._second,
          us1, us2, us3) = string
-        self.__microsecond = (((us1 << 8) | us2) << 8) | us3
+        self._microsecond = (((us1 << 8) | us2) << 8) | us3
         self._tzinfo = tzinfo
 
     def __reduce__(self):
-        return (time, self.__getstate())
+        return (time, self._getstate())
 
 _time_class = time  # so functions w/ args named "time" can get at the class
 
@@ -1291,8 +1291,8 @@
     """
 
     __slots__ = date.__slots__ + (
-        '__hour', '__minute', '__second',
-        '__microsecond', '_tzinfo')
+        '_hour', '_minute', '_second',
+        '_microsecond', '_tzinfo')
     def __new__(cls, year, month=None, day=None, hour=0, minute=0, second=0,
                 microsecond=0, tzinfo=None):
         if isinstance(year, bytes) and len(year) == 10:
@@ -1303,14 +1303,10 @@
         _check_tzinfo_arg(tzinfo)
         _check_time_fields(hour, minute, second, microsecond)
         self = date.__new__(cls, year, month, day)
-        # XXX This duplicates __year, __month, __day for convenience :-(
-        self.__year = year
-        self.__month = month
-        self.__day = day
-        self.__hour = hour
-        self.__minute = minute
-        self.__second = second
-        self.__microsecond = microsecond
+        self._hour = hour
+        self._minute = minute
+        self._second = second
+        self._microsecond = microsecond
         self._tzinfo = tzinfo
         return self
 
@@ -1318,22 +1314,22 @@
     @property
     def hour(self):
         """hour (0-23)"""
-        return self.__hour
+        return self._hour
 
     @property
     def minute(self): 
         """minute (0-59)"""
-        return self.__minute
+        return self._minute
 
     @property
     def second(self):
         """second (0-59)"""
-        return self.__second
+        return self._second
     
     @property
     def microsecond(self):
         """microsecond (0-999999)"""
-        return self.__microsecond
+        return self._microsecond
 
     @property
     def tzinfo(self):
@@ -1426,7 +1422,7 @@
 
     def date(self):
         "Return the date part."
-        return date(self.__year, self.__month, self.__day)
+        return date(self._year, self._month, self._day)
 
     def time(self):
         "Return the time part, with tzinfo None."
@@ -1489,10 +1485,10 @@
         weekday = self.toordinal() % 7 or 7
         return "%s %s %2d %02d:%02d:%02d %04d" % (
             _DAYNAMES[weekday],
-            _MONTHNAMES[self.__month],
-            self.__day,
-            self.__hour, self.__minute, self.__second,
-            self.__year)
+            _MONTHNAMES[self._month],
+            self._day,
+            self._hour, self._minute, self._second,
+            self._year)
 
     def isoformat(self, sep='T'):
         """Return the time formatted according to ISO.
@@ -1506,10 +1502,10 @@
         Optional argument sep specifies the separator between date and
         time, default 'T'.
         """
-        s = ("%04d-%02d-%02d%c" % (self.__year, self.__month, self.__day,
+        s = ("%04d-%02d-%02d%c" % (self._year, self._month, self._day,
                                   sep) +
-                _format_time(self.__hour, self.__minute, self.__second,
-                             self.__microsecond))
+                _format_time(self._hour, self._minute, self._second,
+                             self._microsecond))
         off = self._utcoffset()
         if off is not None:
             if off < 0:
@@ -1523,8 +1519,8 @@
 
     def __repr__(self):
         "Convert to formal string, for repr()."
-        L = [self.__year, self.__month, self.__day, # These are never zero
-             self.__hour, self.__minute, self.__second, self.__microsecond]
+        L = [self._year, self._month, self._day, # These are never zero
+             self._hour, self._minute, self._second, self._microsecond]
         if L[-1] == 0:
             del L[-1]
         if L[-1] == 0:
@@ -1597,7 +1593,7 @@
 
     def __eq__(self, other):
         if isinstance(other, datetime):
-            return self.__cmp(other) == 0
+            return self._cmp(other) == 0
         elif not isinstance(other, date):
             return NotImplemented
         else:
@@ -1605,7 +1601,7 @@
 
     def __ne__(self, other):
         if isinstance(other, datetime):
-            return self.__cmp(other) != 0
+            return self._cmp(other) != 0
         elif not isinstance(other, date):
             return NotImplemented
         else:
@@ -1613,7 +1609,7 @@
 
     def __le__(self, other):
         if isinstance(other, datetime):
-            return self.__cmp(other) <= 0
+            return self._cmp(other) <= 0
         elif not isinstance(other, date):
             return NotImplemented
         else:
@@ -1621,7 +1617,7 @@
 
     def __lt__(self, other):
         if isinstance(other, datetime):
-            return self.__cmp(other) < 0
+            return self._cmp(other) < 0
         elif not isinstance(other, date):
             return NotImplemented
         else:
@@ -1629,7 +1625,7 @@
 
     def __ge__(self, other):
         if isinstance(other, datetime):
-            return self.__cmp(other) >= 0
+            return self._cmp(other) >= 0
         elif not isinstance(other, date):
             return NotImplemented
         else:
@@ -1637,13 +1633,13 @@
 
     def __gt__(self, other):
         if isinstance(other, datetime):
-            return self.__cmp(other) > 0
+            return self._cmp(other) > 0
         elif not isinstance(other, date):
             return NotImplemented
         else:
             _cmperror(self, other)
 
-    def __cmp(self, other):
+    def _cmp(self, other):
         assert isinstance(other, datetime)
         mytz = self._tzinfo
         ottz = other._tzinfo
@@ -1659,12 +1655,12 @@
             base_compare = myoff == otoff
 
         if base_compare:
-            return _cmp((self.__year, self.__month, self.__day,
-                         self.__hour, self.__minute, self.__second,
-                         self.__microsecond),
-                       (other.__year, other.__month, other.__day,
-                        other.__hour, other.__minute, other.__second,
-                        other.__microsecond))
+            return _cmp((self._year, self._month, self._day,
+                         self._hour, self._minute, self._second,
+                         self._microsecond),
+                       (other._year, other._month, other._day,
+                        other._hour, other._minute, other._second,
+                        other._microsecond))
         if myoff is None or otoff is None:
             raise TypeError("cannot compare naive and aware datetimes")
         # XXX What follows could be done more efficiently...
@@ -1678,10 +1674,10 @@
         if not isinstance(other, timedelta):
             return NotImplemented
         delta = timedelta(self.toordinal(),
-                          hours=self.__hour,
-                          minutes=self.__minute,
-                          seconds=self.__second,
-                          microseconds=self.__microsecond)
+                          hours=self._hour,
+                          minutes=self._minute,
+                          seconds=self._second,
+                          microseconds=self._microsecond)
         delta += other
         hour, rem = divmod(delta.seconds, 3600)
         minute, second = divmod(rem, 60)
@@ -1703,11 +1699,11 @@
 
         days1 = self.toordinal()
         days2 = other.toordinal()
-        secs1 = self.__second + self.__minute * 60 + self.__hour * 3600
-        secs2 = other.__second + other.__minute * 60 + other.__hour * 3600
+        secs1 = self._second + self._minute * 60 + self._hour * 3600
+        secs2 = other._second + other._minute * 60 + other._hour * 3600
         base = timedelta(days1 - days2,
                          secs1 - secs2,
-                         self.__microsecond - other.__microsecond)
+                         self._microsecond - other._microsecond)
         if self._tzinfo is other._tzinfo:
             return base
         myoff = self._utcoffset()
@@ -1721,19 +1717,19 @@
     def __hash__(self):
         tzoff = self._utcoffset()
         if tzoff is None:
-            return hash(self.__getstate()[0])
+            return hash(self._getstate()[0])
         days = _ymd2ord(self.year, self.month, self.day)
         seconds = self.hour * 3600 + (self.minute - tzoff) * 60 + self.second
         return hash(timedelta(days, seconds, self.microsecond))
 
     # Pickle support.
 
-    def __getstate(self):
-        yhi, ylo = divmod(self.__year, 256)
-        us2, us3 = divmod(self.__microsecond, 256)
+    def _getstate(self):
+        yhi, ylo = divmod(self._year, 256)
+        us2, us3 = divmod(self._microsecond, 256)
         us1, us2 = divmod(us2, 256)
-        basestate = bytes([yhi, ylo, self.__month, self.__day,
-                           self.__hour, self.__minute, self.__second,
+        basestate = bytes([yhi, ylo, self._month, self._day,
+                           self._hour, self._minute, self._second,
                            us1, us2, us3])
         if self._tzinfo is None:
             return (basestate,)
@@ -1741,14 +1737,14 @@
             return (basestate, self._tzinfo)
 
     def __setstate(self, string, tzinfo):
-        (yhi, ylo, self.__month, self.__day, self.__hour,
-         self.__minute, self.__second, us1, us2, us3) = string
-        self.__year = yhi * 256 + ylo
-        self.__microsecond = (((us1 << 8) | us2) << 8) | us3
+        (yhi, ylo, self._month, self._day, self._hour,
+         self._minute, self._second, us1, us2, us3) = string
+        self._year = yhi * 256 + ylo
+        self._microsecond = (((us1 << 8) | us2) << 8) | us3
         self._tzinfo = tzinfo
 
     def __reduce__(self):
-        return (self.__class__, self.__getstate())
+        return (self.__class__, self._getstate())
 
 
 datetime.min = datetime(1, 1, 1)
@@ -1768,7 +1764,7 @@
     return week1monday
 
 class timezone(tzinfo):
-    __slots__ = '__offset', '__name'
+    __slots__ = '_offset', '_name'
 
     def __init__(self, offset, *args):
         # Reproduce C behavior
@@ -1788,40 +1784,40 @@
                     offset.seconds % 60 != 0):
                     raise ValueError("offset must be whole"
                                     " number of minutes")
-                self.__offset = offset
+                self._offset = offset
             else:
                 raise ValueError("offset out of range")
         else:
             raise TypeError("offset must be timedelta")
 
-        self.__name = name
+        self._name = name
 
     def __getinitargs__(self):
         """pickle support"""
-        if self.__name is None:
-            return (self.__offset,)
-        return (self.__offset, self.__name)
+        if self._name is None:
+            return (self._offset,)
+        return (self._offset, self._name)
 
     def __eq__(self, other):
-        return self.__offset == other.__offset
+        return self._offset == other._offset
     
     def __hash__(self):
-        return hash(self.__offset)
+        return hash(self._offset)
     
     def __str__(self):
         return self.tzname(None)
 
     def utcoffset(self, dt):
         if isinstance(dt, datetime) or dt is None:
-            return self.__offset
+            return self._offset
         raise TypeError("utcoffset() argument must be a datetime instance"
                         " or None")
 
     def tzname(self, dt):
         if isinstance(dt, datetime) or dt is None:
-            if self.__name is None:
-                return self._name_from_offset(self.__offset)
-            return self.__name
+            if self._name is None:
+                return self._name_from_offset(self._offset)
+            return self._name
         raise TypeError("tzname() argument must be a datetime instance"
                         " or None")
 
@@ -1836,7 +1832,7 @@
             if dt.tzinfo is not self:
                 raise ValueError("fromutc: dt.tzinfo "
                                  "is not self")
-            return dt + self.__offset
+            return dt + self._offset
         raise TypeError("fromutc() argument must be a datetime instance"
                         " or None")
 

From python-checkins at python.org  Sat Jul  3 19:48:22 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Sat,  3 Jul 2010 19:48:22 +0200 (CEST)
Subject: [Python-checkins] r82510 - in python/branches/py3k:
	Doc/library/urllib.parse.rst Lib/test/test_urllib.py
	Lib/urllib/parse.py Misc/NEWS
Message-ID: <20100703174822.C912BCA5E@mail.python.org>

Author: senthil.kumaran
Date: Sat Jul  3 19:48:22 2010
New Revision: 82510

Log:
Fix Issue5468 - urlencode to handle bytes and other alternate encodings.
(Extensive tests provided). Patch by Dan Mahn.



Modified:
   python/branches/py3k/Doc/library/urllib.parse.rst
   python/branches/py3k/Lib/test/test_urllib.py
   python/branches/py3k/Lib/urllib/parse.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Doc/library/urllib.parse.rst
==============================================================================
--- python/branches/py3k/Doc/library/urllib.parse.rst	(original)
+++ python/branches/py3k/Doc/library/urllib.parse.rst	Sat Jul  3 19:48:22 2010
@@ -310,23 +310,29 @@
    ``b'a&\xef'``.
 
 
-.. function:: urlencode(query, doseq=False)
+.. function:: urlencode(query, doseq=False, safe='', encoding=None, errors=None)
 
-   Convert a mapping object or a sequence of two-element tuples to a
-   "url-encoded" string, suitable to pass to :func:`urlopen` above as the
-   optional *data* argument.  This is useful to pass a dictionary of form
-   fields to a ``POST`` request.  The resulting string is a series of
-   ``key=value`` pairs separated by ``'&'`` characters, where both *key* and
-   *value* are quoted using :func:`quote_plus` above. When a sequence of
-   two-element tuples is used as the *query* argument, the first element of
-   each tuple is a key and the second is a value. The value element in itself
-   can be a sequence and in that case, if the optional parameter *doseq* is
-   evaluates to *True*, individual ``key=value`` pairs separated by ``'&'`` are
-   generated for each element of the value sequence for the key.  The order of
-   parameters in the encoded string will match the order of parameter tuples in
-   the sequence. This module provides the functions :func:`parse_qs` and
-   :func:`parse_qsl` which are used to parse query strings into Python data
-   structures.
+   Convert a mapping object or a sequence of two-element tuples, which may
+   either be a :class:`str` or a :class:`bytes`,  to a "url-encoded" string,
+   suitable to pass to :func:`urlopen` above as the optional *data* argument.
+   This is useful to pass a dictionary of form fields to a ``POST`` request.
+   The resulting string is a series of ``key=value`` pairs separated by ``'&'``
+   characters, where both *key* and *value* are quoted using :func:`quote_plus`
+   above. When a sequence of two-element tuples is used as the *query*
+   argument, the first element of each tuple is a key and the second is a
+   value. The value element in itself can be a sequence and in that case, if
+   the optional parameter *doseq* is evaluates to *True*, individual
+   ``key=value`` pairs separated by ``'&'`` are generated for each element of
+   the value sequence for the key.  The order of parameters in the encoded
+   string will match the order of parameter tuples in the sequence. This module
+   provides the functions :func:`parse_qs` and :func:`parse_qsl` which are used
+   to parse query strings into Python data structures.
+
+   When *query* parameter is a :class:`str`, the *safe*, *encoding* and *error*
+   parameters are sent the :func:`quote_plus` for encoding.
+
+   .. versionchanged:: 3.2
+      query paramater supports bytes and string.
 
 
 .. seealso::

Modified: python/branches/py3k/Lib/test/test_urllib.py
==============================================================================
--- python/branches/py3k/Lib/test/test_urllib.py	(original)
+++ python/branches/py3k/Lib/test/test_urllib.py	Sat Jul  3 19:48:22 2010
@@ -795,6 +795,116 @@
         self.assertEqual("a=a&a=b",
                          urllib.parse.urlencode({"a": {"a": 1, "b": 1}}, True))
 
+    def test_urlencode_encoding(self):
+        # ASCII encoding. Expect %3F with errors="replace'
+        given = (('\u00a0', '\u00c1'),)
+        expect = '%3F=%3F'
+        result = urllib.parse.urlencode(given, encoding="ASCII", errors="replace")
+        self.assertEqual(expect, result)
+
+        # Default is UTF-8 encoding.
+        given = (('\u00a0', '\u00c1'),)
+        expect = '%C2%A0=%C3%81'
+        result = urllib.parse.urlencode(given)
+        self.assertEqual(expect, result)
+
+        # Latin-1 encoding.
+        given = (('\u00a0', '\u00c1'),)
+        expect = '%A0=%C1'
+        result = urllib.parse.urlencode(given, encoding="latin-1")
+        self.assertEqual(expect, result)
+
+    def test_urlencode_encoding_doseq(self):
+        # ASCII Encoding. Expect %3F with errors="replace'
+        given = (('\u00a0', '\u00c1'),)
+        expect = '%3F=%3F'
+        result = urllib.parse.urlencode(given, doseq=True,
+                                        encoding="ASCII", errors="replace")
+        self.assertEqual(expect, result)
+
+        # ASCII Encoding. On a sequence of values.
+        given = (("\u00a0", (1, "\u00c1")),)
+        expect = '%3F=1&%3F=%3F'
+        result = urllib.parse.urlencode(given, True,
+                                        encoding="ASCII", errors="replace")
+        self.assertEqual(expect, result)
+
+        # Utf-8
+        given = (("\u00a0", "\u00c1"),)
+        expect = '%C2%A0=%C3%81'
+        result = urllib.parse.urlencode(given, True)
+        self.assertEqual(expect, result)
+
+        given = (("\u00a0", (42, "\u00c1")),)
+        expect = '%C2%A0=42&%C2%A0=%C3%81'
+        result = urllib.parse.urlencode(given, True)
+        self.assertEqual(expect, result)
+
+        # latin-1
+        given = (("\u00a0", "\u00c1"),)
+        expect = '%A0=%C1'
+        result = urllib.parse.urlencode(given, True, encoding="latin-1")
+        self.assertEqual(expect, result)
+
+        given = (("\u00a0", (42, "\u00c1")),)
+        expect = '%A0=42&%A0=%C1'
+        result = urllib.parse.urlencode(given, True, encoding="latin-1")
+        self.assertEqual(expect, result)
+
+    def test_urlencode_bytes(self):
+        given = ((b'\xa0\x24', b'\xc1\x24'),)
+        expect = '%A0%24=%C1%24'
+        result = urllib.parse.urlencode(given)
+        self.assertEqual(expect, result)
+        result = urllib.parse.urlencode(given, True)
+        self.assertEqual(expect, result)
+
+        # Sequence of values
+        given = ((b'\xa0\x24', (42, b'\xc1\x24')),)
+        expect = '%A0%24=42&%A0%24=%C1%24'
+        result = urllib.parse.urlencode(given, True)
+        self.assertEqual(expect, result)
+
+    def test_urlencode_encoding_safe_parameter(self):
+
+        # Send '$' (\x24) as safe character
+        # Default utf-8 encoding
+
+        given = ((b'\xa0\x24', b'\xc1\x24'),)
+        result = urllib.parse.urlencode(given, safe=":$")
+        expect = '%A0$=%C1$'
+        self.assertEqual(expect, result)
+
+        given = ((b'\xa0\x24', b'\xc1\x24'),)
+        result = urllib.parse.urlencode(given, doseq=True, safe=":$")
+        expect = '%A0$=%C1$'
+        self.assertEqual(expect, result)
+
+        # Safe parameter in sequence
+        given = ((b'\xa0\x24', (b'\xc1\x24', 0xd, 42)),)
+        expect = '%A0$=%C1$&%A0$=13&%A0$=42'
+        result = urllib.parse.urlencode(given, True, safe=":$")
+        self.assertEqual(expect, result)
+
+        # Test all above in latin-1 encoding
+
+        given = ((b'\xa0\x24', b'\xc1\x24'),)
+        result = urllib.parse.urlencode(given, safe=":$",
+                                        encoding="latin-1")
+        expect = '%A0$=%C1$'
+        self.assertEqual(expect, result)
+
+        given = ((b'\xa0\x24', b'\xc1\x24'),)
+        expect = '%A0$=%C1$'
+        result = urllib.parse.urlencode(given, doseq=True, safe=":$",
+                                        encoding="latin-1")
+
+        given = ((b'\xa0\x24', (b'\xc1\x24', 0xd, 42)),)
+        expect = '%A0$=%C1$&%A0$=13&%A0$=42'
+        result = urllib.parse.urlencode(given, True, safe=":$",
+                                        encoding="latin-1")
+        self.assertEqual(expect, result)
+
 class Pathname_Tests(unittest.TestCase):
     """Test pathname2url() and url2pathname()"""
 

Modified: python/branches/py3k/Lib/urllib/parse.py
==============================================================================
--- python/branches/py3k/Lib/urllib/parse.py	(original)
+++ python/branches/py3k/Lib/urllib/parse.py	Sat Jul  3 19:48:22 2010
@@ -559,7 +559,7 @@
         _safe_quoters[safe] = quoter = Quoter(safe).__getitem__
     return ''.join([quoter(char) for char in bs])
 
-def urlencode(query, doseq=False):
+def urlencode(query, doseq=False, safe='', encoding=None, errors=None):
     """Encode a sequence of two-element tuples or dictionary into a URL query string.
 
     If any values in the query arg are sequences and doseq is true, each
@@ -568,6 +568,10 @@
     If the query arg is a sequence of two-element tuples, the order of the
     parameters in the output will match the order of parameters in the
     input.
+
+    The query arg may be either a string or a bytes type. When query arg is a
+    string, the safe, encoding and error parameters are sent the quote_plus for
+    encoding.
     """
 
     if hasattr(query, "items"):
@@ -592,14 +596,28 @@
     l = []
     if not doseq:
         for k, v in query:
-            k = quote_plus(str(k))
-            v = quote_plus(str(v))
+            if isinstance(k, bytes):
+                k = quote_plus(k, safe)
+            else:
+                k = quote_plus(str(k), safe, encoding, errors)
+
+            if isinstance(v, bytes):
+                v = quote_plus(v, safe)
+            else:
+                v = quote_plus(str(v), safe, encoding, errors)
             l.append(k + '=' + v)
     else:
         for k, v in query:
-            k = quote_plus(str(k))
-            if isinstance(v, str):
-                v = quote_plus(v)
+            if isinstance(k, bytes):
+                k = quote_plus(k, safe)
+            else:
+                k = quote_plus(str(k), safe, encoding, errors)
+
+            if isinstance(v, bytes):
+                v = quote_plus(v, safe)
+                l.append(k + '=' + v)
+            elif isinstance(v, str):
+                v = quote_plus(v, safe, encoding, errors)
                 l.append(k + '=' + v)
             else:
                 try:
@@ -607,12 +625,16 @@
                     x = len(v)
                 except TypeError:
                     # not a sequence
-                    v = quote_plus(str(v))
+                    v = quote_plus(str(v), safe, encoding, errors)
                     l.append(k + '=' + v)
                 else:
                     # loop over the sequence
                     for elt in v:
-                        l.append(k + '=' + quote_plus(str(elt)))
+                        if isinstance(elt, bytes):
+                            elt = quote_plus(elt, safe)
+                        else:
+                            elt = quote_plus(str(elt), safe, encoding, errors)
+                        l.append(k + '=' + elt)
     return '&'.join(l)
 
 # Utilities to parse URLs (most of these return None for missing parts):

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sat Jul  3 19:48:22 2010
@@ -468,6 +468,9 @@
 Library
 -------
 
+- Issue #5468: urlencode to handle bytes type and other encodings in its query
+  parameter. Patch by Dan Mahn.
+
 - Issue #7673: Fix security vulnerability (CVE-2010-2089) in the audioop
   module, ensure that the input string length is a multiple of the frame size
 

From python-checkins at python.org  Sat Jul  3 19:55:41 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Sat,  3 Jul 2010 19:55:41 +0200 (CEST)
Subject: [Python-checkins] r82511 - in python/branches/release31-maint:
	Doc/library/urllib.parse.rst Lib/test/test_urllib.py
	Lib/urllib/parse.py Misc/NEWS
Message-ID: <20100703175541.A9F43ED67@mail.python.org>

Author: senthil.kumaran
Date: Sat Jul  3 19:55:41 2010
New Revision: 82511

Log:
Merged revisions 82510 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82510 | senthil.kumaran | 2010-07-03 23:18:22 +0530 (Sat, 03 Jul 2010) | 4 lines
  
  Fix Issue5468 - urlencode to handle bytes and other alternate encodings.
  (Extensive tests provided). Patch by Dan Mahn.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Doc/library/urllib.parse.rst
   python/branches/release31-maint/Lib/test/test_urllib.py
   python/branches/release31-maint/Lib/urllib/parse.py
   python/branches/release31-maint/Misc/NEWS

Modified: python/branches/release31-maint/Doc/library/urllib.parse.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/urllib.parse.rst	(original)
+++ python/branches/release31-maint/Doc/library/urllib.parse.rst	Sat Jul  3 19:55:41 2010
@@ -307,23 +307,29 @@
    ``b'a&\xef'``.
 
 
-.. function:: urlencode(query, doseq=False)
+.. function:: urlencode(query, doseq=False, safe='', encoding=None, errors=None)
 
-   Convert a mapping object or a sequence of two-element tuples  to a
-   "url-encoded" string, suitable to pass to :func:`urlopen` above as the
-   optional *data* argument.  This is useful to pass a dictionary of form
-   fields to a ``POST`` request.  The resulting string is a series of
-   ``key=value`` pairs separated by ``'&'`` characters, where both *key* and
-   *value* are quoted using :func:`quote_plus` above. When a sequence of
-   two-element tuples is used as the *query* argument, the first element of
-   each tuple is a key and the second is a value. The value element in itself
-   can be a sequence and in that case, if the optional parameter *doseq* is
-   evaluates to *True*, individual ``key=value`` pairs separated by ``'&'``are
-   generated for each element of the value sequence for the key.  The order of
-   parameters in the encoded string will match the order of parameter tuples in
-   the sequence. This module provides the functions :func:`parse_qs` and
-   :func:`parse_qsl` which are used to parse query strings into Python data
-   structures.
+   Convert a mapping object or a sequence of two-element, which may either be a
+   :class:`str` or a :class:`bytes` tuples,  to a "url-encoded" string,
+   suitable to pass to :func:`urlopen` above as the optional *data* argument.
+   This is useful to pass a dictionary of form fields to a ``POST`` request.
+   The resulting string is a series of ``key=value`` pairs separated by ``'&'``
+   characters, where both *key* and *value* are quoted using :func:`quote_plus`
+   above. When a sequence of two-element tuples is used as the *query*
+   argument, the first element of each tuple is a key and the second is a
+   value. The value element in itself can be a sequence and in that case, if
+   the optional parameter *doseq* is evaluates to *True*, individual
+   ``key=value`` pairs separated by ``'&'`` are generated for each element of
+   the value sequence for the key.  The order of parameters in the encoded
+   string will match the order of parameter tuples in the sequence. This module
+   provides the functions :func:`parse_qs` and :func:`parse_qsl` which are used
+   to parse query strings into Python data structures.
+
+   When *query* parameter is a :class:`str`, the *safe*, *encoding* and *error*
+   parameters are sent the :func:`quote_plus` for encoding.
+
+   .. versionchanged:: 3.2
+      query paramater supports bytes and string.
 
 
 .. seealso::

Modified: python/branches/release31-maint/Lib/test/test_urllib.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_urllib.py	(original)
+++ python/branches/release31-maint/Lib/test/test_urllib.py	Sat Jul  3 19:55:41 2010
@@ -797,6 +797,116 @@
         self.assertEqual("a=a&a=b",
                          urllib.parse.urlencode({"a": {"a": 1, "b": 1}}, True))
 
+    def test_urlencode_encoding(self):
+        # ASCII encoding. Expect %3F with errors="replace'
+        given = (('\u00a0', '\u00c1'),)
+        expect = '%3F=%3F'
+        result = urllib.parse.urlencode(given, encoding="ASCII", errors="replace")
+        self.assertEqual(expect, result)
+
+        # Default is UTF-8 encoding.
+        given = (('\u00a0', '\u00c1'),)
+        expect = '%C2%A0=%C3%81'
+        result = urllib.parse.urlencode(given)
+        self.assertEqual(expect, result)
+
+        # Latin-1 encoding.
+        given = (('\u00a0', '\u00c1'),)
+        expect = '%A0=%C1'
+        result = urllib.parse.urlencode(given, encoding="latin-1")
+        self.assertEqual(expect, result)
+
+    def test_urlencode_encoding_doseq(self):
+        # ASCII Encoding. Expect %3F with errors="replace'
+        given = (('\u00a0', '\u00c1'),)
+        expect = '%3F=%3F'
+        result = urllib.parse.urlencode(given, doseq=True,
+                                        encoding="ASCII", errors="replace")
+        self.assertEqual(expect, result)
+
+        # ASCII Encoding. On a sequence of values.
+        given = (("\u00a0", (1, "\u00c1")),)
+        expect = '%3F=1&%3F=%3F'
+        result = urllib.parse.urlencode(given, True,
+                                        encoding="ASCII", errors="replace")
+        self.assertEqual(expect, result)
+
+        # Utf-8
+        given = (("\u00a0", "\u00c1"),)
+        expect = '%C2%A0=%C3%81'
+        result = urllib.parse.urlencode(given, True)
+        self.assertEqual(expect, result)
+
+        given = (("\u00a0", (42, "\u00c1")),)
+        expect = '%C2%A0=42&%C2%A0=%C3%81'
+        result = urllib.parse.urlencode(given, True)
+        self.assertEqual(expect, result)
+
+        # latin-1
+        given = (("\u00a0", "\u00c1"),)
+        expect = '%A0=%C1'
+        result = urllib.parse.urlencode(given, True, encoding="latin-1")
+        self.assertEqual(expect, result)
+
+        given = (("\u00a0", (42, "\u00c1")),)
+        expect = '%A0=42&%A0=%C1'
+        result = urllib.parse.urlencode(given, True, encoding="latin-1")
+        self.assertEqual(expect, result)
+
+    def test_urlencode_bytes(self):
+        given = ((b'\xa0\x24', b'\xc1\x24'),)
+        expect = '%A0%24=%C1%24'
+        result = urllib.parse.urlencode(given)
+        self.assertEqual(expect, result)
+        result = urllib.parse.urlencode(given, True)
+        self.assertEqual(expect, result)
+
+        # Sequence of values
+        given = ((b'\xa0\x24', (42, b'\xc1\x24')),)
+        expect = '%A0%24=42&%A0%24=%C1%24'
+        result = urllib.parse.urlencode(given, True)
+        self.assertEqual(expect, result)
+
+    def test_urlencode_encoding_safe_parameter(self):
+
+        # Send '$' (\x24) as safe character
+        # Default utf-8 encoding
+
+        given = ((b'\xa0\x24', b'\xc1\x24'),)
+        result = urllib.parse.urlencode(given, safe=":$")
+        expect = '%A0$=%C1$'
+        self.assertEqual(expect, result)
+
+        given = ((b'\xa0\x24', b'\xc1\x24'),)
+        result = urllib.parse.urlencode(given, doseq=True, safe=":$")
+        expect = '%A0$=%C1$'
+        self.assertEqual(expect, result)
+
+        # Safe parameter in sequence
+        given = ((b'\xa0\x24', (b'\xc1\x24', 0xd, 42)),)
+        expect = '%A0$=%C1$&%A0$=13&%A0$=42'
+        result = urllib.parse.urlencode(given, True, safe=":$")
+        self.assertEqual(expect, result)
+
+        # Test all above in latin-1 encoding
+
+        given = ((b'\xa0\x24', b'\xc1\x24'),)
+        result = urllib.parse.urlencode(given, safe=":$",
+                                        encoding="latin-1")
+        expect = '%A0$=%C1$'
+        self.assertEqual(expect, result)
+
+        given = ((b'\xa0\x24', b'\xc1\x24'),)
+        expect = '%A0$=%C1$'
+        result = urllib.parse.urlencode(given, doseq=True, safe=":$",
+                                        encoding="latin-1")
+
+        given = ((b'\xa0\x24', (b'\xc1\x24', 0xd, 42)),)
+        expect = '%A0$=%C1$&%A0$=13&%A0$=42'
+        result = urllib.parse.urlencode(given, True, safe=":$",
+                                        encoding="latin-1")
+        self.assertEqual(expect, result)
+
 class Pathname_Tests(unittest.TestCase):
     """Test pathname2url() and url2pathname()"""
 

Modified: python/branches/release31-maint/Lib/urllib/parse.py
==============================================================================
--- python/branches/release31-maint/Lib/urllib/parse.py	(original)
+++ python/branches/release31-maint/Lib/urllib/parse.py	Sat Jul  3 19:55:41 2010
@@ -533,7 +533,7 @@
         _safe_quoters[cachekey] = quoter
     return ''.join([quoter[char] for char in bs])
 
-def urlencode(query, doseq=False):
+def urlencode(query, doseq=False, safe='', encoding=None, errors=None):
     """Encode a sequence of two-element tuples or dictionary into a URL query string.
 
     If any values in the query arg are sequences and doseq is true, each
@@ -542,6 +542,10 @@
     If the query arg is a sequence of two-element tuples, the order of the
     parameters in the output will match the order of parameters in the
     input.
+
+    The query arg may be either a string or a bytes type. When query arg is a
+    string, the safe, encoding and error parameters are sent the quote_plus for
+    encoding.
     """
 
     if hasattr(query, "items"):
@@ -566,14 +570,28 @@
     l = []
     if not doseq:
         for k, v in query:
-            k = quote_plus(str(k))
-            v = quote_plus(str(v))
+            if isinstance(k, bytes):
+                k = quote_plus(k, safe)
+            else:
+                k = quote_plus(str(k), safe, encoding, errors)
+
+            if isinstance(v, bytes):
+                v = quote_plus(v, safe)
+            else:
+                v = quote_plus(str(v), safe, encoding, errors)
             l.append(k + '=' + v)
     else:
         for k, v in query:
-            k = quote_plus(str(k))
-            if isinstance(v, str):
-                v = quote_plus(v)
+            if isinstance(k, bytes):
+                k = quote_plus(k, safe)
+            else:
+                k = quote_plus(str(k), safe, encoding, errors)
+
+            if isinstance(v, bytes):
+                v = quote_plus(v, safe)
+                l.append(k + '=' + v)
+            elif isinstance(v, str):
+                v = quote_plus(v, safe, encoding, errors)
                 l.append(k + '=' + v)
             else:
                 try:
@@ -581,12 +599,16 @@
                     x = len(v)
                 except TypeError:
                     # not a sequence
-                    v = quote_plus(str(v))
+                    v = quote_plus(str(v), safe, encoding, errors)
                     l.append(k + '=' + v)
                 else:
                     # loop over the sequence
                     for elt in v:
-                        l.append(k + '=' + quote_plus(str(elt)))
+                        if isinstance(elt, bytes):
+                            elt = quote_plus(elt, safe)
+                        else:
+                            elt = quote_plus(str(elt), safe, encoding, errors)
+                        l.append(k + '=' + elt)
     return '&'.join(l)
 
 # Utilities to parse URLs (most of these return None for missing parts):

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Sat Jul  3 19:55:41 2010
@@ -75,6 +75,9 @@
 Library
 -------
 
+- Issue #5468: urlencode to handle bytes type and other encodings in its query
+  parameter. Patch by Dan Mahn.
+
 - Issue #7673: Fix security vulnerability (CVE-2010-2089) in the audioop
   module, ensure that the input string length is a multiple of the frame size
 

From python-checkins at python.org  Sat Jul  3 20:04:31 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Sat,  3 Jul 2010 20:04:31 +0200 (CEST)
Subject: [Python-checkins] r82512 -
	python/branches/release31-maint/Doc/library/urllib.parse.rst
Message-ID: <20100703180431.9B106EE99D@mail.python.org>

Author: senthil.kumaran
Date: Sat Jul  3 20:04:31 2010
New Revision: 82512

Log:
Minor change in the wording.



Modified:
   python/branches/release31-maint/Doc/library/urllib.parse.rst

Modified: python/branches/release31-maint/Doc/library/urllib.parse.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/urllib.parse.rst	(original)
+++ python/branches/release31-maint/Doc/library/urllib.parse.rst	Sat Jul  3 20:04:31 2010
@@ -309,8 +309,8 @@
 
 .. function:: urlencode(query, doseq=False, safe='', encoding=None, errors=None)
 
-   Convert a mapping object or a sequence of two-element, which may either be a
-   :class:`str` or a :class:`bytes` tuples,  to a "url-encoded" string,
+   Convert a mapping object or a sequence of two-element tuples, which may
+   either be a :class:`str` or a :class:`bytes`,  to a "url-encoded" string,
    suitable to pass to :func:`urlopen` above as the optional *data* argument.
    This is useful to pass a dictionary of form fields to a ``POST`` request.
    The resulting string is a series of ``key=value`` pairs separated by ``'&'``

From python-checkins at python.org  Sat Jul  3 20:24:52 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sat,  3 Jul 2010 20:24:52 +0200 (CEST)
Subject: [Python-checkins] r82513 -
	python/branches/release27-maint/Doc/whatsnew/2.7.rst
Message-ID: <20100703182452.C96F9EE99E@mail.python.org>

Author: benjamin.peterson
Date: Sat Jul  3 20:24:52 2010
New Revision: 82513

Log:
fix date

Modified:
   python/branches/release27-maint/Doc/whatsnew/2.7.rst

Modified: python/branches/release27-maint/Doc/whatsnew/2.7.rst
==============================================================================
--- python/branches/release27-maint/Doc/whatsnew/2.7.rst	(original)
+++ python/branches/release27-maint/Doc/whatsnew/2.7.rst	Sat Jul  3 20:24:52 2010
@@ -52,7 +52,7 @@
    when researching a change.
 
 This article explains the new features in Python 2.7.  Python 2.7 was released
-on July 7, 2010.
+on July 3, 2010.
 
 Numeric handling has been improved in many ways, for both
 floating-point numbers and for the :class:`~decimal.Decimal` class.

From python-checkins at python.org  Sat Jul  3 22:35:53 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Sat,  3 Jul 2010 22:35:53 +0200 (CEST)
Subject: [Python-checkins] r82514 - in python/branches/py3k:
	Lib/pickletools.py Misc/NEWS
Message-ID: <20100703203553.2FB99EE982@mail.python.org>

Author: alexander.belopolsky
Date: Sat Jul  3 22:35:53 2010
New Revision: 82514

Log:
Issue #9094: Make python -m pickletools disassemble pickles given in
the command line.


Modified:
   python/branches/py3k/Lib/pickletools.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/pickletools.py
==============================================================================
--- python/branches/py3k/Lib/pickletools.py	(original)
+++ python/branches/py3k/Lib/pickletools.py	Sat Jul  3 22:35:53 2010
@@ -2330,4 +2330,43 @@
     return doctest.testmod()
 
 if __name__ == "__main__":
-    _test()
+    import sys, argparse
+    parser = argparse.ArgumentParser(
+        description='disassemble one or more pickle files')
+    parser.add_argument(
+        'pickle_file', type=argparse.FileType('br'),
+        nargs='*', help='the pickle file')
+    parser.add_argument(
+        '-o', '--output', default=sys.stdout, type=argparse.FileType('w'),
+        help='the file where the output should be written')
+    parser.add_argument(
+        '-m', '--memo', action='store_true',
+        help='preserve memo between disassemblies')
+    parser.add_argument(
+        '-l', '--indentlevel', default=4, type=int,
+        help='the number of blanks by which to indent a new MARK level')
+    parser.add_argument(
+        '-p', '--preamble', default="==> {name} <==",
+        help='if more than one pickle file is specified, print this before'
+        ' each disassembly')
+    parser.add_argument(
+        '-t', '--test', action='store_true',
+        help='run self-test suite')
+    parser.add_argument(
+        '-v', action='store_true',
+        help='run verbosely; only affects self-test run')
+    args = parser.parse_args()
+    if args.test:
+        _test()
+    else:
+        if not args.pickle_file:
+            parser.print_help()
+        elif len(args.pickle_file) == 1:
+            dis(args.pickle_file[0], args.output,
+                indentlevel=args.indentlevel)
+        else:
+            memo = {} if args.memo else None
+            for f in args.pickle_file:
+                preamble = args.preamble.format(name=f.name)
+                args.output.write(preamble + '\n')
+                dis(f, args.output, memo, args.indentlevel)

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sat Jul  3 22:35:53 2010
@@ -468,6 +468,10 @@
 Library
 -------
 
+- Issue #9094: python -m pickletools will now disassemble pickle files
+  listed in the command line arguments.  See output of python -m
+  pickletools -h for more details.
+
 - Issue #5468: urlencode to handle bytes type and other encodings in its query
   parameter. Patch by Dan Mahn.
 

From python-checkins at python.org  Sat Jul  3 23:00:51 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sat,  3 Jul 2010 23:00:51 +0200 (CEST)
Subject: [Python-checkins] r82515 - python/branches/py3k/Tools/README
Message-ID: <20100703210051.ADAD3EE989@mail.python.org>

Author: mark.dickinson
Date: Sat Jul  3 23:00:51 2010
New Revision: 82515

Log:
Remove nonexistent tools from Tools README.

Modified:
   python/branches/py3k/Tools/README

Modified: python/branches/py3k/Tools/README
==============================================================================
--- python/branches/py3k/Tools/README	(original)
+++ python/branches/py3k/Tools/README	Sat Jul  3 23:00:51 2010
@@ -1,15 +1,6 @@
 This directory contains a number of Python programs that are useful
 while building or extending Python.
 
-audiopy		Audiopy is a program to control the Solaris audio
-		device, allowing you to choose both the input and
-		output devices, and to set the output volume, that can
-		be run either as a command-line script, or as a
-		Tkinter application.
-
-compiler	Tools used to maintain the compiler package in the
-		standard library.
-
 faqwiz		FAQ Wizard.
 		See http://www.python.org/cgi-bin/faqw.py
 		for a live example.
@@ -36,8 +27,6 @@
 versioncheck	A tool to automate checking whether you have the latest
 		version of a package (by Jack Jansen).
 
-webchecker	A link checker for web sites.
-
 world		Script to take a list of Internet addresses and print
 		out where in the world those addresses originate from,
 		based on the top-level domain country code found in

From python-checkins at python.org  Sat Jul  3 23:37:28 2010
From: python-checkins at python.org (brett.cannon)
Date: Sat,  3 Jul 2010 23:37:28 +0200 (CEST)
Subject: [Python-checkins] r82516 - python/branches/py3k/Tools/scripts
Message-ID: <20100703213728.40331EE982@mail.python.org>

Author: brett.cannon
Date: Sat Jul  3 23:37:28 2010
New Revision: 82516

Log:
Ignore __pycache__.


Modified:
   python/branches/py3k/Tools/scripts/   (props changed)

From python-checkins at python.org  Sat Jul  3 23:42:47 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Sat,  3 Jul 2010 23:42:47 +0200 (CEST)
Subject: [Python-checkins] r82517 -
	python/branches/py3k/Demo/classes/Dates.py
Message-ID: <20100703214247.8E726EE982@mail.python.org>

Author: alexander.belopolsky
Date: Sat Jul  3 23:42:47 2010
New Revision: 82517

Log:
Issue #9151: Demo/classes/Dates.py does not work in 3.x
Made minimal changes to make included test pass.


Modified:
   python/branches/py3k/Demo/classes/Dates.py

Modified: python/branches/py3k/Demo/classes/Dates.py
==============================================================================
--- python/branches/py3k/Demo/classes/Dates.py	(original)
+++ python/branches/py3k/Demo/classes/Dates.py	Sat Jul  3 23:42:47 2010
@@ -39,7 +39,7 @@
 # Note that as of Python 2.3, a datetime module is included in the stardard
 # library.
 
-# vi:set tabsize=8:
+import functools
 
 _MONTH_NAMES = [ 'January', 'February', 'March', 'April', 'May',
                  'June', 'July', 'August', 'September', 'October',
@@ -57,8 +57,6 @@
     dbm = dbm + dim
 del dbm, dim
 
-_INT_TYPES = type(1), type(1)
-
 def _is_leap(year):           # 1 if leap year, else 0
     if year % 4 != 0: return 0
     if year % 400 == 0: return 1
@@ -85,7 +83,7 @@
 _DI400Y = _days_before_year(400)      # number of days in 400 years
 
 def _num2date(n):             # return date with ordinal n
-    if type(n) not in _INT_TYPES:
+    if not isinstance(n, int):
         raise TypeError('argument must be integer: %r' % type(n))
 
     ans = Date(1,1,1)   # arguments irrelevant; just getting a Date obj
@@ -116,7 +114,7 @@
 def _num2day(n):      # return weekday name of day with ordinal n
     return _DAY_NAMES[ int(n % 7) ]
 
-
+ at functools.total_ordering
 class Date:
     def __init__(self, month, day, year):
         if not 1 <= month <= 12:
@@ -133,8 +131,11 @@
             raise AttributeError('read-only attribute ' + name)
         self.__dict__[name] = value
 
-    def __cmp__(self, other):
-        return cmp(self.ord, other.ord)
+    def __eq__(self, other):
+        return self.ord == other.ord
+
+    def __lt__(self, other):
+        return self.ord < other.ord
 
     # define a hash function so dates can be used as dictionary keys
     def __hash__(self):
@@ -150,14 +151,14 @@
 
     # Python 1.1 coerces neither int+date nor date+int
     def __add__(self, n):
-        if type(n) not in _INT_TYPES:
+        if not isinstance(n, int):
             raise TypeError('can\'t add %r to date' % type(n))
         return _num2date(self.ord + n)
     __radd__ = __add__ # handle int+date
 
     # Python 1.1 coerces neither date-int nor date-date
     def __sub__(self, other):
-        if type(other) in _INT_TYPES:           # date-int
+        if isinstance(other, int):           # date-int
             return _num2date(self.ord - other)
         else:
             return self.ord - other.ord         # date-date

From python-checkins at python.org  Sat Jul  3 23:48:26 2010
From: python-checkins at python.org (brett.cannon)
Date: Sat,  3 Jul 2010 23:48:26 +0200 (CEST)
Subject: [Python-checkins] r82518 - in python/branches/py3k:
	Doc/library/importlib.rst Lib/importlib/__init__.py
	Lib/importlib/_bootstrap.py Lib/importlib/abc.py
	Lib/importlib/test/extension/test_case_sensitivity.py
	Lib/importlib/test/extension/test_finder.py
	Lib/importlib/test/extension/test_loader.py
	Lib/importlib/test/extension/test_path_hook.py
	Lib/importlib/test/import_/test_path.py
	Lib/importlib/test/regrtest.py
	Lib/importlib/test/source/test_abc_loader.py
	Lib/importlib/test/source/test_case_sensitivity.py
	Lib/importlib/test/source/test_file_loader.py
	Lib/importlib/test/source/test_finder.py
	Lib/importlib/test/source/test_path_hook.py
	Lib/importlib/test/source/test_source_encoding.py
	Lib/test/test_runpy.py
Message-ID: <20100703214826.3EAF9EA60@mail.python.org>

Author: brett.cannon
Date: Sat Jul  3 23:48:25 2010
New Revision: 82518

Log:
Make importlib.abc.SourceLoader the primary mechanism for importlib.

This required moving the class from importlib/abc.py into
importlib/_bootstrap.py and jiggering some code to work better with the class.
This included changing how the file finder worked to better meet import
semantics. This also led to fixing importlib to handle the empty string from
sys.path as import currently does (and making me wish we didn't support that
instead just required people to insert '.' instead to represent cwd).

It also required making the new set_data abstractmethod create
any needed subdirectories implicitly thanks to __pycache__ (it was either this
or grow the SourceLoader ABC to gain an 'exists' method and either a mkdir
method or have set_data with no data arg mean to create a directory).

Lastly, as an optimization the file loaders cache the file path where the
finder found something to use for loading (this is thanks to having a
sourceless loader separate from the source loader to simplify the code and
cut out stat calls).
Unfortunately test_runpy assumed a loader would always work for a module, even
if you changed from underneath it what it was expected to work with. By simply
dropping the previous loader in test_runpy so the proper loader can be returned
by the finder fixed the failure.

At this point importlib deviates from import on two points:

1. The exception raised when trying to import a file is different (import does
an explicit file check to print a special message, importlib just says the path
cannot be imported as if it was just some module name).

2. the co_filename on a code object is not being set to where bytecode was
actually loaded from instead of where the marshalled code object originally
came from (a solution for this has already been agreed upon on python-dev but has
not been implemented yet; issue8611).


Modified:
   python/branches/py3k/Doc/library/importlib.rst
   python/branches/py3k/Lib/importlib/__init__.py
   python/branches/py3k/Lib/importlib/_bootstrap.py
   python/branches/py3k/Lib/importlib/abc.py
   python/branches/py3k/Lib/importlib/test/extension/test_case_sensitivity.py
   python/branches/py3k/Lib/importlib/test/extension/test_finder.py
   python/branches/py3k/Lib/importlib/test/extension/test_loader.py
   python/branches/py3k/Lib/importlib/test/extension/test_path_hook.py
   python/branches/py3k/Lib/importlib/test/import_/test_path.py
   python/branches/py3k/Lib/importlib/test/regrtest.py
   python/branches/py3k/Lib/importlib/test/source/test_abc_loader.py
   python/branches/py3k/Lib/importlib/test/source/test_case_sensitivity.py
   python/branches/py3k/Lib/importlib/test/source/test_file_loader.py
   python/branches/py3k/Lib/importlib/test/source/test_finder.py
   python/branches/py3k/Lib/importlib/test/source/test_path_hook.py
   python/branches/py3k/Lib/importlib/test/source/test_source_encoding.py
   python/branches/py3k/Lib/test/test_runpy.py

Modified: python/branches/py3k/Doc/library/importlib.rst
==============================================================================
--- python/branches/py3k/Doc/library/importlib.rst	(original)
+++ python/branches/py3k/Doc/library/importlib.rst	Sat Jul  3 23:48:25 2010
@@ -247,8 +247,11 @@
     .. method:: set_data(self, path, data)
 
         Optional abstract method which writes the specified bytes to a file
-        path. When writing to the path fails because the path is read-only, do
-        not propagate the exception.
+        path. Any intermediate directories which do not exist are to be created
+        automatically.
+
+        When writing to the path fails because the path is read-only
+        (:attr:`errno.EACCES`), do not propagate the exception.
 
     .. method:: get_code(self, fullname)
 

Modified: python/branches/py3k/Lib/importlib/__init__.py
==============================================================================
--- python/branches/py3k/Lib/importlib/__init__.py	(original)
+++ python/branches/py3k/Lib/importlib/__init__.py	Sat Jul  3 23:48:25 2010
@@ -36,7 +36,7 @@
     """
     if 'PYTHONCASEOK' in os.environ:
         return True
-    elif check in os.listdir(directory):
+    elif check in os.listdir(directory if directory else os.getcwd()):
         return True
     return False
 

Modified: python/branches/py3k/Lib/importlib/_bootstrap.py
==============================================================================
--- python/branches/py3k/Lib/importlib/_bootstrap.py	(original)
+++ python/branches/py3k/Lib/importlib/_bootstrap.py	Sat Jul  3 23:48:25 2010
@@ -22,7 +22,7 @@
 def _path_join(*args):
     """Replacement for os.path.join."""
     return path_sep.join(x[:-len(path_sep)] if x.endswith(path_sep) else x
-                            for x in args)
+                            for x in args if x)
 
 
 def _path_exists(path):
@@ -53,6 +53,8 @@
 # XXX Could also expose Modules/getpath.c:isdir()
 def _path_isdir(path):
     """Replacement for os.path.isdir."""
+    if not path:
+        path = _os.getcwd()
     return _path_is_mode_type(path, 0o040000)
 
 
@@ -99,6 +101,8 @@
     new.__dict__.update(old.__dict__)
 
 
+code_type = type(_wrap.__code__)
+
 # Finder/loader utility code ##################################################
 
 def set_package(fxn):
@@ -138,7 +142,7 @@
     the second argument.
 
     """
-    def decorated(self, fullname):
+    def decorated(self, fullname, *args, **kwargs):
         module = sys.modules.get(fullname)
         is_reload = bool(module)
         if not is_reload:
@@ -148,7 +152,7 @@
             module = imp.new_module(fullname)
             sys.modules[fullname] = module
         try:
-            return fxn(self, module)
+            return fxn(self, module, *args, **kwargs)
         except:
             if not is_reload:
                 del sys.modules[fullname]
@@ -301,7 +305,59 @@
         return imp.is_frozen_package(fullname)
 
 
-class SourceLoader:
+class _LoaderBasics:
+
+    """Base class of common code needed by both SourceLoader and
+    _SourcelessFileLoader."""
+
+    def is_package(self, fullname):
+        """Concrete implementation of InspectLoader.is_package by checking if
+        the path returned by get_filename has a filename of '__init__.py'."""
+        filename = self.get_filename(fullname).rpartition(path_sep)[2]
+        return filename.rsplit('.', 1)[0] == '__init__'
+
+    def _bytes_from_bytecode(self, fullname, data, source_mtime):
+        """Return the marshalled bytes from bytecode, verifying the magic
+        number and timestamp alon the way.
+
+        If source_mtime is None then skip the timestamp check.
+
+        """
+        magic = data[:4]
+        raw_timestamp = data[4:8]
+        if len(magic) != 4 or magic != imp.get_magic():
+            raise ImportError("bad magic number in {}".format(fullname))
+        elif len(raw_timestamp) != 4:
+            raise EOFError("bad timestamp in {}".format(fullname))
+        elif source_mtime is not None:
+            if marshal._r_long(raw_timestamp) != source_mtime:
+                raise ImportError("bytecode is stale for {}".format(fullname))
+        # Can't return the code object as errors from marshal loading need to
+        # propagate even when source is available.
+        return data[8:]
+
+    @module_for_loader
+    def _load_module(self, module, *, sourceless=False):
+        """Helper for load_module able to handle either source or sourceless
+        loading."""
+        name = module.__name__
+        code_object = self.get_code(name)
+        module.__file__ = self.get_filename(name)
+        if not sourceless:
+            module.__cached__ = imp.cache_from_source(module.__file__)
+        else:
+            module.__cached__ = module.__file__
+        module.__package__ = name
+        if self.is_package(name):
+            module.__path__ = [module.__file__.rsplit(path_sep, 1)[0]]
+        else:
+            module.__package__ = module.__package__.rpartition('.')[0]
+        module.__loader__ = self
+        exec(code_object, module.__dict__)
+        return module
+
+
+class SourceLoader(_LoaderBasics):
 
     def path_mtime(self, path:str) -> int:
         """Optional method that returns the modification time for the specified
@@ -320,11 +376,6 @@
         """
         raise NotImplementedError
 
-    def is_package(self, fullname):
-        """Concrete implementation of InspectLoader.is_package by checking if
-        the path returned by get_filename has a filename of '__init__.py'."""
-        filename = self.get_filename(fullname).rsplit(path_sep, 1)[1]
-        return filename.rsplit('.', 1)[0] == '__init__'
 
     def get_source(self, fullname):
         """Concrete implementation of InspectLoader.get_source."""
@@ -359,12 +410,18 @@
                 except IOError:
                     pass
                 else:
-                    magic = data[:4]
-                    raw_timestamp = data[4:8]
-                    if (len(magic) == 4 and len(raw_timestamp) == 4 and
-                            magic == imp.get_magic() and
-                            marshal._r_long(raw_timestamp) == source_mtime):
-                        return marshal.loads(data[8:])
+                    try:
+                        bytes_data = self._bytes_from_bytecode(fullname, data,
+                                                               source_mtime)
+                    except (ImportError, EOFError):
+                        pass
+                    else:
+                        found = marshal.loads(bytes_data)
+                        if isinstance(found, code_type):
+                            return found
+                        else:
+                            msg = "Non-code object in {}"
+                            raise ImportError(msg.format(bytecode_path))
         source_bytes = self.get_data(source_path)
         code_object = compile(source_bytes, source_path, 'exec',
                                 dont_inherit=True)
@@ -382,8 +439,7 @@
                 pass
         return code_object
 
-    @module_for_loader
-    def load_module(self, module):
+    def load_module(self, fullname):
         """Concrete implementation of Loader.load_module.
 
         Requires ExecutionLoader.get_filename and ResourceLoader.get_data to be
@@ -391,31 +447,17 @@
         get_code uses/writes bytecode.
 
         """
-        name = module.__name__
-        code_object = self.get_code(name)
-        module.__file__ = self.get_filename(name)
-        module.__cached__ = imp.cache_from_source(module.__file__)
-        module.__package__ = name
-        is_package = self.is_package(name)
-        if is_package:
-            module.__path__ = [module.__file__.rsplit(path_sep, 1)[0]]
-        else:
-            module.__package__ = module.__package__.rpartition('.')[0]
-        module.__loader__ = self
-        exec(code_object, module.__dict__)
-        return module
-
+        return self._load_module(fullname)
 
-class _SourceFileLoader(SourceLoader):
 
-    """Concrete implementation of SourceLoader.
+class _FileLoader:
 
-    NOT A PUBLIC CLASS! Do not expect any API stability from this class, so DO
-    NOT SUBCLASS IN YOUR OWN CODE!
-
-    """
+    """Base file loader class which implements the loader protocol methods that
+    require file system usage."""
 
     def __init__(self, fullname, path):
+        """Cache the module name and the path to the file found by the
+        finder."""
         self._name = fullname
         self._path = path
 
@@ -424,272 +466,66 @@
         """Return the path to the source file as found by the finder."""
         return self._path
 
+    def get_data(self, path):
+        """Return the data from path as raw bytes."""
+        with _closing(_io.FileIO(path, 'r')) as file:
+            return file.read()
+
+
+class _SourceFileLoader(_FileLoader, SourceLoader):
+
+    """Concrete implementation of SourceLoader using the file system."""
+
     def path_mtime(self, path):
         """Return the modification time for the path."""
         return int(_os.stat(path).st_mtime)
 
-    def set_data(self, data, path):
+    def set_data(self, path, data):
         """Write bytes data to a file."""
         try:
-            with _closing(_io.FileIO(bytecode_path, 'w')) as file:
+            with _closing(_io.FileIO(path, 'wb')) as file:
                 file.write(data)
         except IOError as exc:
-            if exc.errno != errno.EACCES:
+            if exc.errno == errno.ENOENT:
+                directory, _, filename = path.rpartition(path_sep)
+                sub_directories = []
+                while not _path_isdir(directory):
+                    directory, _, sub_dir = directory.rpartition(path_sep)
+                    sub_directories.append(sub_dir)
+                    for part in reversed(sub_directories):
+                        directory = _path_join(directory, part)
+                        try:
+                            _os.mkdir(directory)
+                        except IOError as exc:
+                            if exc.errno != errno.EACCES:
+                                raise
+                            else:
+                                return
+                return self.set_data(path, data)
+            elif exc.errno != errno.EACCES:
                 raise
 
 
-class PyLoader:
-
-    """Loader base class for Python source code.
-
-    Subclasses need to implement the methods:
-
-    - source_path
-    - get_data
-    - is_package
-
-    """
-
-    @module_for_loader
-    def load_module(self, module):
-        """Initialize the module."""
-        name = module.__name__
-        code_object = self.get_code(module.__name__)
-        module.__file__ = self.get_filename(name)
-        if self.is_package(name):
-            module.__path__  = [module.__file__.rsplit(path_sep, 1)[0]]
-        module.__package__ = module.__name__
-        if not hasattr(module, '__path__'):
-            module.__package__ = module.__package__.rpartition('.')[0]
-        module.__loader__ = self
-        exec(code_object, module.__dict__)
-        return module
-
-    def get_filename(self, fullname):
-        """Return the path to the source file, else raise ImportError."""
-        path = self.source_path(fullname)
-        if path is not None:
-            return path
-        else:
-            raise ImportError("no source path available for "
-                                "{0!r}".format(fullname))
-
-    def get_code(self, fullname):
-        """Get a code object from source."""
-        source_path = self.source_path(fullname)
-        if source_path is None:
-            message = "a source path must exist to load {0}".format(fullname)
-            raise ImportError(message)
-        source = self.get_data(source_path)
-        return compile(source, source_path, 'exec', dont_inherit=True)
-
-    # Never use in implementing import! Imports code within the method.
-    def get_source(self, fullname):
-        """Return the source code for a module.
-
-        self.source_path() and self.get_data() are used to implement this
-        method.
-
-        """
-        path = self.source_path(fullname)
-        if path is None:
-            return None
-        try:
-            source_bytes = self.get_data(path)
-        except IOError:
-            return ImportError("source not available through get_data()")
-        import io
-        import tokenize
-        encoding = tokenize.detect_encoding(io.BytesIO(source_bytes).readline)
-        return source_bytes.decode(encoding[0])
-
-
-class PyPycLoader(PyLoader):
-
-    """Loader base class for Python source and bytecode.
+class _SourcelessFileLoader(_FileLoader, _LoaderBasics):
 
-    Requires implementing the methods needed for PyLoader as well as
-    source_mtime, bytecode_path, and write_bytecode.
+    """Loader which handles sourceless file imports."""
 
-    """
-
-    def get_filename(self, fullname):
-        """Return the source or bytecode file path."""
-        path = self.source_path(fullname)
-        if path is not None:
-            return path
-        path = self.bytecode_path(fullname)
-        if path is not None:
-            return path
-        raise ImportError("no source or bytecode path available for "
-                            "{0!r}".format(fullname))
+    def load_module(self, fullname):
+        return self._load_module(fullname, sourceless=True)
 
     def get_code(self, fullname):
-        """Get a code object from source or bytecode."""
-        # XXX Care enough to make sure this call does not happen if the magic
-        #     number is bad?
-        source_timestamp = self.source_mtime(fullname)
-        # Try to use bytecode if it is available.
-        bytecode_path = self.bytecode_path(fullname)
-        if bytecode_path:
-            data = self.get_data(bytecode_path)
-            try:
-                magic = data[:4]
-                if len(magic) < 4:
-                    raise ImportError("bad magic number in {}".format(fullname))
-                raw_timestamp = data[4:8]
-                if len(raw_timestamp) < 4:
-                    raise EOFError("bad timestamp in {}".format(fullname))
-                pyc_timestamp = marshal._r_long(raw_timestamp)
-                bytecode = data[8:]
-                # Verify that the magic number is valid.
-                if imp.get_magic() != magic:
-                    raise ImportError("bad magic number in {}".format(fullname))
-                # Verify that the bytecode is not stale (only matters when
-                # there is source to fall back on.
-                if source_timestamp:
-                    if pyc_timestamp < source_timestamp:
-                        raise ImportError("bytecode is stale")
-            except (ImportError, EOFError):
-                # If source is available give it a shot.
-                if source_timestamp is not None:
-                    pass
-                else:
-                    raise
-            else:
-                # Bytecode seems fine, so try to use it.
-                # XXX If the bytecode is ill-formed, would it be beneficial to
-                #     try for using source if available and issue a warning?
-                return marshal.loads(bytecode)
-        elif source_timestamp is None:
-            raise ImportError("no source or bytecode available to create code "
-                                "object for {0!r}".format(fullname))
-        # Use the source.
-        code_object = super().get_code(fullname)
-        # Generate bytecode and write it out.
-        if not sys.dont_write_bytecode:
-            data = bytearray(imp.get_magic())
-            data.extend(marshal._w_long(source_timestamp))
-            data.extend(marshal.dumps(code_object))
-            self.write_bytecode(fullname, data)
-        return code_object
-
-
-class _PyFileLoader(PyLoader):
-
-    """Load a Python source file."""
-
-    def __init__(self, name, path, is_pkg):
-        self._name = name
-        self._is_pkg = is_pkg
-        # Figure out the base path based on whether it was source or bytecode
-        # that was found.
-        try:
-            self._base_path = _path_without_ext(path, imp.PY_SOURCE)
-        except ValueError:
-            self._base_path = _path_without_ext(path, imp.PY_COMPILED)
-
-    def _find_path(self, ext_type):
-        """Find a path from the base path and the specified extension type that
-        exists, returning None if one is not found."""
-        for suffix in _suffix_list(ext_type):
-            path = self._base_path + suffix
-            if _path_exists(path):
-                return path
+        path = self.get_filename(fullname)
+        data = self.get_data(path)
+        bytes_data = self._bytes_from_bytecode(fullname, data, None)
+        found = marshal.loads(bytes_data)
+        if isinstance(found, code_type):
+            return found
         else:
-            return None
-
-    @_check_name
-    def source_path(self, fullname):
-        """Return the path to an existing source file for the module, or None
-        if one cannot be found."""
-        # Not a property so that it is easy to override.
-        return self._find_path(imp.PY_SOURCE)
-
-    def get_data(self, path):
-        """Return the data from path as raw bytes."""
-        return _io.FileIO(path, 'r').read()  # Assuming bytes.
-
-    @_check_name
-    def is_package(self, fullname):
-        """Return a boolean based on whether the module is a package.
-
-        Raises ImportError (like get_source) if the loader cannot handle the
-        package.
-
-        """
-        return self._is_pkg
-
-
-class _PyPycFileLoader(PyPycLoader, _PyFileLoader):
-
-    """Load a module from a source or bytecode file."""
-
-    def _find_path(self, ext_type):
-        """Return PEP 3147 path if ext_type is PY_COMPILED, otherwise
-        super()._find_path() is called."""
-        if ext_type == imp.PY_COMPILED:
-            # We don't really care what the extension on self._base_path is,
-            # as long as it has exactly one dot.
-            source_path = self._base_path + '.py'
-            pycache_path = imp.cache_from_source(source_path)
-            legacy_path = self._base_path + '.pyc'
-            # The rule is: if the source file exists, then Python always uses
-            # the __pycache__/foo.<tag>.pyc file.  If the source file does not
-            # exist, then Python uses the legacy path.
-            pyc_path = (pycache_path
-                        if _path_exists(source_path)
-                        else legacy_path)
-            return (pyc_path if _path_exists(pyc_path) else None)
-        return super()._find_path(ext_type)
-
-    @_check_name
-    def source_mtime(self, name):
-        """Return the modification time of the source for the specified
-        module."""
-        source_path = self.source_path(name)
-        if not source_path:
-            return None
-        return int(_os.stat(source_path).st_mtime)
-
-    @_check_name
-    def bytecode_path(self, fullname):
-        """Return the path to a bytecode file, or None if one does not
-        exist."""
-        # Not a property for easy overriding.
-        return self._find_path(imp.PY_COMPILED)
+            raise ImportError("Non-code object in {}".format(path))
 
-    @_check_name
-    def write_bytecode(self, name, data):
-        """Write out 'data' for the specified module, returning a boolean
-        signifying if the write-out actually occurred.
-
-        Raises ImportError (just like get_source) if the specified module
-        cannot be handled by the loader.
-
-        """
-        bytecode_path = self.bytecode_path(name)
-        if not bytecode_path:
-            source_path = self.source_path(name)
-            bytecode_path = imp.cache_from_source(source_path)
-            # Ensure that the __pycache__ directory exists.  We can't use
-            # os.path.dirname() here.
-            dirname, sep, basename = bytecode_path.rpartition(path_sep)
-            try:
-                _os.mkdir(dirname)
-            except OSError as error:
-                if error.errno != errno.EEXIST:
-                    raise
-        try:
-            # Assuming bytes.
-            with _closing(_io.FileIO(bytecode_path, 'w')) as bytecode_file:
-                bytecode_file.write(data)
-                return True
-        except IOError as exc:
-            if exc.errno == errno.EACCES:
-                return False
-            else:
-                raise
+    def get_source(self, fullname):
+        """Return None as there is no source code."""
+        return None
 
 
 class _ExtensionFileLoader:
@@ -700,7 +536,7 @@
 
     """
 
-    def __init__(self, name, path, is_pkg):
+    def __init__(self, name, path):
         """Initialize the loader.
 
         If is_pkg is True then an exception is raised as extension modules
@@ -709,8 +545,6 @@
         """
         self._name = name
         self._path = path
-        if is_pkg:
-            raise ValueError("extension modules cannot be packages")
 
     @_check_name
     @set_package
@@ -808,147 +642,88 @@
             return None
 
 
-class _ChainedFinder:
-
-    """Finder that sequentially calls other finders."""
-
-    def __init__(self, *finders):
-        self._finders = finders
-
-    def find_module(self, fullname, path=None):
-        for finder in self._finders:
-            result = finder.find_module(fullname, path)
-            if result:
-                return result
-        else:
-            return None
-
-
 class _FileFinder:
 
-    """Base class for file finders.
-
-    Subclasses are expected to define the following attributes:
-
-        * _suffixes
-            Sequence of file suffixes whose order will be followed.
-
-        * _possible_package
-            True if importer should check for packages.
+    """File-based finder.
 
-        * _loader
-            A callable that takes the module name, a file path, and whether
-            the path points to a package and returns a loader for the module
-            found at that path.
+    Constructor takes a list of objects detailing what file extensions their
+    loader supports along with whether it can be used for a package.
 
     """
 
-    def __init__(self, path_entry):
-        """Initialize an importer for the passed-in sys.path entry (which is
-        assumed to have already been verified as an existing directory).
+    def __init__(self, path, *details):
+        """Initialize with finder details."""
+        packages = []
+        modules = []
+        for detail in details:
+            modules.extend((suffix, detail.loader) for suffix in detail.suffixes)
+            if detail.supports_packages:
+                packages.extend((suffix, detail.loader)
+                                for suffix in detail.suffixes)
+        self.packages = packages
+        self.modules = modules
+        self.path = path
 
-        Can be used as an entry on sys.path_hook.
-
-        """
-        absolute_path = _path_absolute(path_entry)
-        if not _path_isdir(absolute_path):
-            raise ImportError("only directories are supported")
-        self._path_entry = absolute_path
-
-    def find_module(self, fullname, path=None):
+    def find_module(self, fullname):
+        """Try to find a loader for the specified module."""
         tail_module = fullname.rpartition('.')[2]
-        package_directory = None
-        if self._possible_package:
-            for ext in self._suffixes:
-                package_directory = _path_join(self._path_entry, tail_module)
-                init_filename = '__init__' + ext
-                package_init = _path_join(package_directory, init_filename)
-                if (_path_isfile(package_init) and
-                        _case_ok(self._path_entry, tail_module) and
-                        _case_ok(package_directory, init_filename)):
-                    return self._loader(fullname, package_init, True)
-        for ext in self._suffixes:
-            file_name = tail_module + ext
-            file_path = _path_join(self._path_entry, file_name)
-            if (_path_isfile(file_path) and
-                    _case_ok(self._path_entry, file_name)):
-                return self._loader(fullname, file_path, False)
-        else:
-            # Raise a warning if it matches a directory w/o an __init__ file.
-            if (package_directory is not None and
-                    _path_isdir(package_directory) and
-                    _case_ok(self._path_entry, tail_module)):
-                _warnings.warn("Not importing directory %s: missing __init__"
-                                % package_directory, ImportWarning)
-            return None
-
-
-class _PyFileFinder(_FileFinder):
-
-    """Importer for source/bytecode files."""
-
-    _possible_package = True
-    _loader = _PyFileLoader
-
-    def __init__(self, path_entry):
-        # Lack of imp during class creation means _suffixes is set here.
-        # Make sure that Python source files are listed first!  Needed for an
-        # optimization by the loader.
-        self._suffixes = _suffix_list(imp.PY_SOURCE)
-        super().__init__(path_entry)
-
-
-class _PyPycFileFinder(_PyFileFinder):
+        base_path = _path_join(self.path, tail_module)
+        if _path_isdir(base_path) and _case_ok(self.path, tail_module):
+            for suffix, loader in self.packages:
+                init_filename = '__init__' + suffix
+                full_path = _path_join(base_path, init_filename)
+                if (_path_isfile(full_path) and
+                        _case_ok(base_path, init_filename)):
+                    return loader(fullname, full_path)
+            else:
+                msg = "Not importing directory {}: missing __init__"
+                _warnings.warn(msg.format(base_path), ImportWarning)
+        for suffix, loader in self.modules:
+            mod_filename = tail_module + suffix
+            full_path = _path_join(self.path, mod_filename)
+            if _path_isfile(full_path) and _case_ok(self.path, mod_filename):
+                return loader(fullname, full_path)
+        return None
 
-    """Finder for source and bytecode files."""
+class _SourceFinderDetails:
 
-    _loader = _PyPycFileLoader
+    loader = _SourceFileLoader
+    supports_packages = True
 
-    def __init__(self, path_entry):
-        super().__init__(path_entry)
-        self._suffixes += _suffix_list(imp.PY_COMPILED)
+    def __init__(self):
+        self.suffixes = _suffix_list(imp.PY_SOURCE)
 
+class _SourcelessFinderDetails:
 
+    loader = _SourcelessFileLoader
+    supports_packages = True
 
+    def __init__(self):
+        self.suffixes = _suffix_list(imp.PY_COMPILED)
 
-class _ExtensionFileFinder(_FileFinder):
 
-    """Importer for extension files."""
+class _ExtensionFinderDetails:
 
-    _possible_package = False
-    _loader = _ExtensionFileLoader
+    loader = _ExtensionFileLoader
+    supports_packages = False
 
-    def __init__(self, path_entry):
-        # Assigning to _suffixes here instead of at the class level because
-        # imp is not imported at the time of class creation.
-        self._suffixes = _suffix_list(imp.C_EXTENSION)
-        super().__init__(path_entry)
+    def __init__(self):
+        self.suffixes = _suffix_list(imp.C_EXTENSION)
 
 
 # Import itself ###############################################################
 
-def _chained_path_hook(*path_hooks):
-    """Create a closure which sequentially checks path hooks to see which ones
-    (if any) can work with a path."""
-    def path_hook(entry):
-        """Check to see if 'entry' matches any of the enclosed path hooks."""
-        finders = []
-        for hook in path_hooks:
-            try:
-                finder = hook(entry)
-            except ImportError:
-                continue
-            else:
-                finders.append(finder)
-        if not finders:
-            raise ImportError("no finder found")
-        else:
-            return _ChainedFinder(*finders)
-
-    return path_hook
+def _file_path_hook(path):
+    """If the path is a directory, return a file-based finder."""
+    if _path_isdir(path):
+        return _FileFinder(path, _ExtensionFinderDetails(),
+                           _SourceFinderDetails(),
+                           _SourcelessFinderDetails())
+    else:
+        raise ImportError("only directories are supported")
 
 
-_DEFAULT_PATH_HOOK = _chained_path_hook(_ExtensionFileFinder, _PyPycFileFinder)
+_DEFAULT_PATH_HOOK = _file_path_hook
 
 class _DefaultPathFinder(PathFinder):
 

Modified: python/branches/py3k/Lib/importlib/abc.py
==============================================================================
--- python/branches/py3k/Lib/importlib/abc.py	(original)
+++ python/branches/py3k/Lib/importlib/abc.py	Sat Jul  3 23:48:25 2010
@@ -182,8 +182,6 @@
         else:
             return path
 
-PyLoader.register(_bootstrap.PyLoader)
-
 
 class PyPycLoader(PyLoader):
 
@@ -266,7 +264,6 @@
             self.write_bytecode(fullname, data)
         return code_object
 
-
     @abc.abstractmethod
     def source_mtime(self, fullname:str) -> int:
         """Abstract method which when implemented should return the
@@ -285,5 +282,3 @@
         bytecode for the module, returning a boolean representing whether the
         bytecode was written or not."""
         raise NotImplementedError
-
-PyPycLoader.register(_bootstrap.PyPycLoader)

Modified: python/branches/py3k/Lib/importlib/test/extension/test_case_sensitivity.py
==============================================================================
--- python/branches/py3k/Lib/importlib/test/extension/test_case_sensitivity.py	(original)
+++ python/branches/py3k/Lib/importlib/test/extension/test_case_sensitivity.py	Sat Jul  3 23:48:25 2010
@@ -13,7 +13,8 @@
         good_name = ext_util.NAME
         bad_name = good_name.upper()
         assert good_name != bad_name
-        finder = _bootstrap._ExtensionFileFinder(ext_util.PATH)
+        finder = _bootstrap._FileFinder(ext_util.PATH,
+                                        _bootstrap._ExtensionFinderDetails())
         return finder.find_module(bad_name)
 
     def test_case_sensitive(self):

Modified: python/branches/py3k/Lib/importlib/test/extension/test_finder.py
==============================================================================
--- python/branches/py3k/Lib/importlib/test/extension/test_finder.py	(original)
+++ python/branches/py3k/Lib/importlib/test/extension/test_finder.py	Sat Jul  3 23:48:25 2010
@@ -9,7 +9,8 @@
     """Test the finder for extension modules."""
 
     def find_module(self, fullname):
-        importer = _bootstrap._ExtensionFileFinder(util.PATH)
+        importer = _bootstrap._FileFinder(util.PATH,
+                                          _bootstrap._ExtensionFinderDetails())
         return importer.find_module(fullname)
 
     def test_module(self):

Modified: python/branches/py3k/Lib/importlib/test/extension/test_loader.py
==============================================================================
--- python/branches/py3k/Lib/importlib/test/extension/test_loader.py	(original)
+++ python/branches/py3k/Lib/importlib/test/extension/test_loader.py	Sat Jul  3 23:48:25 2010
@@ -13,7 +13,7 @@
 
     def load_module(self, fullname):
         loader = _bootstrap._ExtensionFileLoader(ext_util.NAME,
-                                                ext_util.FILEPATH, False)
+                                                ext_util.FILEPATH)
         return loader.load_module(fullname)
 
     def test_module(self):

Modified: python/branches/py3k/Lib/importlib/test/extension/test_path_hook.py
==============================================================================
--- python/branches/py3k/Lib/importlib/test/extension/test_path_hook.py	(original)
+++ python/branches/py3k/Lib/importlib/test/extension/test_path_hook.py	Sat Jul  3 23:48:25 2010
@@ -14,7 +14,7 @@
     # XXX Should it only work for directories containing an extension module?
 
     def hook(self, entry):
-        return _bootstrap._ExtensionFileFinder(entry)
+        return _bootstrap._file_path_hook(entry)
 
     def test_success(self):
         # Path hook should handle a directory where a known extension module

Modified: python/branches/py3k/Lib/importlib/test/import_/test_path.py
==============================================================================
--- python/branches/py3k/Lib/importlib/test/import_/test_path.py	(original)
+++ python/branches/py3k/Lib/importlib/test/import_/test_path.py	Sat Jul  3 23:48:25 2010
@@ -5,6 +5,7 @@
 import imp
 import os
 import sys
+import tempfile
 from test import support
 from types import MethodType
 import unittest
@@ -80,23 +81,28 @@
 
     def test_implicit_hooks(self):
         # Test that the implicit path hooks are used.
-        existing_path = os.path.dirname(support.TESTFN)
         bad_path = '<path>'
         module = '<module>'
         assert not os.path.exists(bad_path)
-        with util.import_state():
-            nothing = _bootstrap._DefaultPathFinder.find_module(module,
-                                                           path=[existing_path])
-            self.assertTrue(nothing is None)
-            self.assertTrue(existing_path in sys.path_importer_cache)
-            self.assertTrue(not isinstance(sys.path_importer_cache[existing_path],
-                                        imp.NullImporter))
-            nothing = _bootstrap._DefaultPathFinder.find_module(module,
-                                                                path=[bad_path])
-            self.assertTrue(nothing is None)
-            self.assertTrue(bad_path in sys.path_importer_cache)
-            self.assertTrue(isinstance(sys.path_importer_cache[bad_path],
-                                    imp.NullImporter))
+        existing_path = tempfile.mkdtemp()
+        try:
+            with util.import_state():
+                nothing = _bootstrap._DefaultPathFinder.find_module(module,
+                                                        path=[existing_path])
+                self.assertTrue(nothing is None)
+                self.assertTrue(existing_path in sys.path_importer_cache)
+                result = isinstance(sys.path_importer_cache[existing_path],
+                                    imp.NullImporter)
+                self.assertFalse(result)
+                nothing = _bootstrap._DefaultPathFinder.find_module(module,
+                                                            path=[bad_path])
+                self.assertTrue(nothing is None)
+                self.assertTrue(bad_path in sys.path_importer_cache)
+                self.assertTrue(isinstance(sys.path_importer_cache[bad_path],
+                                           imp.NullImporter))
+        finally:
+            os.rmdir(existing_path)
+
 
     def test_path_importer_cache_has_None(self):
         # Test that the default hook is used when sys.path_importer_cache

Modified: python/branches/py3k/Lib/importlib/test/regrtest.py
==============================================================================
--- python/branches/py3k/Lib/importlib/test/regrtest.py	(original)
+++ python/branches/py3k/Lib/importlib/test/regrtest.py	Sat Jul  3 23:48:25 2010
@@ -6,9 +6,11 @@
 this script.
 
 XXX FAILING
-    test_import
-        execution bit
+  * test_import
+    - test_incorrect_code_name
         file name differing between __file__ and co_filename (r68360 on trunk)
+    - test_import_by_filename
+        exception for trying to import by file name does not match
 
 """
 import importlib

Modified: python/branches/py3k/Lib/importlib/test/source/test_abc_loader.py
==============================================================================
--- python/branches/py3k/Lib/importlib/test/source/test_abc_loader.py	(original)
+++ python/branches/py3k/Lib/importlib/test/source/test_abc_loader.py	Sat Jul  3 23:48:25 2010
@@ -815,6 +815,7 @@
     def test_Loader(self):
         self.raises_NotImplementedError(self.Loader(), 'load_module')
 
+    # XXX misplaced; should be somewhere else
     def test_Finder(self):
         self.raises_NotImplementedError(self.Finder(), 'find_module')
 

Modified: python/branches/py3k/Lib/importlib/test/source/test_case_sensitivity.py
==============================================================================
--- python/branches/py3k/Lib/importlib/test/source/test_case_sensitivity.py	(original)
+++ python/branches/py3k/Lib/importlib/test/source/test_case_sensitivity.py	Sat Jul  3 23:48:25 2010
@@ -19,7 +19,9 @@
     assert name != name.lower()
 
     def find(self, path):
-        finder = _bootstrap._PyPycFileFinder(path)
+        finder = _bootstrap._FileFinder(path,
+                                        _bootstrap._SourceFinderDetails(),
+                                        _bootstrap._SourcelessFinderDetails())
         return finder.find_module(self.name)
 
     def sensitivity_test(self):
@@ -27,7 +29,7 @@
         sensitive_pkg = 'sensitive.{0}'.format(self.name)
         insensitive_pkg = 'insensitive.{0}'.format(self.name.lower())
         context = source_util.create_modules(insensitive_pkg, sensitive_pkg)
-        with  context as mapping:
+        with context as mapping:
             sensitive_path = os.path.join(mapping['.root'], 'sensitive')
             insensitive_path = os.path.join(mapping['.root'], 'insensitive')
             return self.find(sensitive_path), self.find(insensitive_path)
@@ -37,7 +39,7 @@
             env.unset('PYTHONCASEOK')
             sensitive, insensitive = self.sensitivity_test()
             self.assertTrue(hasattr(sensitive, 'load_module'))
-            self.assertIn(self.name, sensitive._base_path)
+            self.assertIn(self.name, sensitive.get_filename(self.name))
             self.assertIsNone(insensitive)
 
     def test_insensitive(self):
@@ -45,9 +47,9 @@
             env.set('PYTHONCASEOK', '1')
             sensitive, insensitive = self.sensitivity_test()
             self.assertTrue(hasattr(sensitive, 'load_module'))
-            self.assertIn(self.name, sensitive._base_path)
+            self.assertIn(self.name, sensitive.get_filename(self.name))
             self.assertTrue(hasattr(insensitive, 'load_module'))
-            self.assertIn(self.name, insensitive._base_path)
+            self.assertIn(self.name, insensitive.get_filename(self.name))
 
 
 def test_main():

Modified: python/branches/py3k/Lib/importlib/test/source/test_file_loader.py
==============================================================================
--- python/branches/py3k/Lib/importlib/test/source/test_file_loader.py	(original)
+++ python/branches/py3k/Lib/importlib/test/source/test_file_loader.py	Sat Jul  3 23:48:25 2010
@@ -4,6 +4,7 @@
 from . import util as source_util
 
 import imp
+import marshal
 import os
 import py_compile
 import stat
@@ -23,8 +24,7 @@
     # [basic]
     def test_module(self):
         with source_util.create_modules('_temp') as mapping:
-            loader = _bootstrap._PyPycFileLoader('_temp', mapping['_temp'],
-                                                    False)
+            loader = _bootstrap._SourceFileLoader('_temp', mapping['_temp'])
             module = loader.load_module('_temp')
             self.assertTrue('_temp' in sys.modules)
             check = {'__name__': '_temp', '__file__': mapping['_temp'],
@@ -34,9 +34,8 @@
 
     def test_package(self):
         with source_util.create_modules('_pkg.__init__') as mapping:
-            loader = _bootstrap._PyPycFileLoader('_pkg',
-                                                 mapping['_pkg.__init__'],
-                                                 True)
+            loader = _bootstrap._SourceFileLoader('_pkg',
+                                                 mapping['_pkg.__init__'])
             module = loader.load_module('_pkg')
             self.assertTrue('_pkg' in sys.modules)
             check = {'__name__': '_pkg', '__file__': mapping['_pkg.__init__'],
@@ -48,8 +47,8 @@
 
     def test_lacking_parent(self):
         with source_util.create_modules('_pkg.__init__', '_pkg.mod')as mapping:
-            loader = _bootstrap._PyPycFileLoader('_pkg.mod',
-                                                    mapping['_pkg.mod'], False)
+            loader = _bootstrap._SourceFileLoader('_pkg.mod',
+                                                    mapping['_pkg.mod'])
             module = loader.load_module('_pkg.mod')
             self.assertTrue('_pkg.mod' in sys.modules)
             check = {'__name__': '_pkg.mod', '__file__': mapping['_pkg.mod'],
@@ -63,8 +62,7 @@
 
     def test_module_reuse(self):
         with source_util.create_modules('_temp') as mapping:
-            loader = _bootstrap._PyPycFileLoader('_temp', mapping['_temp'],
-                                                    False)
+            loader = _bootstrap._SourceFileLoader('_temp', mapping['_temp'])
             module = loader.load_module('_temp')
             module_id = id(module)
             module_dict_id = id(module.__dict__)
@@ -74,7 +72,7 @@
             # everything that has happened above can be too fast;
             # force an mtime on the source that is guaranteed to be different
             # than the original mtime.
-            loader.source_mtime = self.fake_mtime(loader.source_mtime)
+            loader.path_mtime = self.fake_mtime(loader.path_mtime)
             module = loader.load_module('_temp')
             self.assertTrue('testing_var' in module.__dict__,
                          "'testing_var' not in "
@@ -94,8 +92,7 @@
                 setattr(orig_module, attr, value)
             with open(mapping[name], 'w') as file:
                 file.write('+++ bad syntax +++')
-            loader = _bootstrap._PyPycFileLoader('_temp', mapping['_temp'],
-                                                    False)
+            loader = _bootstrap._SourceFileLoader('_temp', mapping['_temp'])
             with self.assertRaises(SyntaxError):
                 loader.load_module(name)
             for attr in attributes:
@@ -106,8 +103,7 @@
         with source_util.create_modules('_temp') as mapping:
             with open(mapping['_temp'], 'w') as file:
                 file.write('=')
-            loader = _bootstrap._PyPycFileLoader('_temp', mapping['_temp'],
-                                                    False)
+            loader = _bootstrap._SourceFileLoader('_temp', mapping['_temp'])
             with self.assertRaises(SyntaxError):
                 loader.load_module('_temp')
             self.assertTrue('_temp' not in sys.modules)
@@ -116,7 +112,7 @@
 class BadBytecodeTest(unittest.TestCase):
 
     def import_(self, file, module_name):
-        loader = _bootstrap._PyPycFileLoader(module_name, file, False)
+        loader = self.loader(module_name, file)
         module = loader.load_module(module_name)
         self.assertTrue(module_name in sys.modules)
 
@@ -129,101 +125,156 @@
         except KeyError:
             pass
         py_compile.compile(mapping[name])
-        bytecode_path = imp.cache_from_source(mapping[name])
-        with open(bytecode_path, 'rb') as file:
-            bc = file.read()
-        new_bc = manipulator(bc)
-        with open(bytecode_path, 'wb') as file:
-            if new_bc:
-                file.write(new_bc)
-        if del_source:
+        if not del_source:
+            bytecode_path = imp.cache_from_source(mapping[name])
+        else:
             os.unlink(mapping[name])
-            make_legacy_pyc(mapping[name])
+            bytecode_path = make_legacy_pyc(mapping[name])
+        if manipulator:
+            with open(bytecode_path, 'rb') as file:
+                bc = file.read()
+                new_bc = manipulator(bc)
+            with open(bytecode_path, 'wb') as file:
+                if new_bc is not None:
+                    file.write(new_bc)
         return bytecode_path
 
-    @source_util.writes_bytecode_files
-    def test_empty_file(self):
-        # When a .pyc is empty, regenerate it if possible, else raise
-        # ImportError.
+    def _test_empty_file(self, test, *, del_source=False):
         with source_util.create_modules('_temp') as mapping:
             bc_path = self.manipulate_bytecode('_temp', mapping,
-                                                lambda bc: None)
-            self.import_(mapping['_temp'], '_temp')
-            with open(bc_path, 'rb') as file:
-                self.assertGreater(len(file.read()), 8)
-            self.manipulate_bytecode('_temp', mapping, lambda bc: None,
-                                        del_source=True)
-            with self.assertRaises(ImportError):
-                self.import_(mapping['_temp'], '_temp')
+                                                lambda bc: b'',
+                                                del_source=del_source)
+            test('_temp', mapping, bc_path)
 
     @source_util.writes_bytecode_files
-    def test_partial_magic(self):
+    def _test_partial_magic(self, test, *, del_source=False):
         # When their are less than 4 bytes to a .pyc, regenerate it if
         # possible, else raise ImportError.
         with source_util.create_modules('_temp') as mapping:
             bc_path = self.manipulate_bytecode('_temp', mapping,
-                                                lambda bc: bc[:3])
-            self.import_(mapping['_temp'], '_temp')
-            with open(bc_path, 'rb') as file:
-                self.assertGreater(len(file.read()), 8)
-            self.manipulate_bytecode('_temp', mapping, lambda bc: bc[:3],
-                                        del_source=True)
+                                                lambda bc: bc[:3],
+                                                del_source=del_source)
+            test('_temp', mapping, bc_path)
+
+    def _test_magic_only(self, test, *, del_source=False):
+        with source_util.create_modules('_temp') as mapping:
+            bc_path = self.manipulate_bytecode('_temp', mapping,
+                                                lambda bc: bc[:4],
+                                                del_source=del_source)
+            test('_temp', mapping, bc_path)
+
+    def _test_partial_timestamp(self, test, *, del_source=False):
+        with source_util.create_modules('_temp') as mapping:
+            bc_path = self.manipulate_bytecode('_temp', mapping,
+                                                lambda bc: bc[:7],
+                                                del_source=del_source)
+            test('_temp', mapping, bc_path)
+
+    def _test_no_marshal(self, *, del_source=False):
+        with source_util.create_modules('_temp') as mapping:
+            bc_path = self.manipulate_bytecode('_temp', mapping,
+                                                lambda bc: bc[:8],
+                                                del_source=del_source)
+            file_path = mapping['_temp'] if not del_source else bc_path
+            with self.assertRaises(EOFError):
+                self.import_(file_path, '_temp')
+
+    def _test_non_code_marshal(self, *, del_source=False):
+        with source_util.create_modules('_temp') as mapping:
+            bytecode_path = self.manipulate_bytecode('_temp', mapping,
+                                    lambda bc: bc[:8] + marshal.dumps(b'abcd'),
+                                    del_source=del_source)
+            file_path = mapping['_temp'] if not del_source else bytecode_path
             with self.assertRaises(ImportError):
-                self.import_(mapping['_temp'], '_temp')
+                self.import_(file_path, '_temp')
+
+    def _test_bad_marshal(self, *, del_source=False):
+        with source_util.create_modules('_temp') as mapping:
+            bytecode_path = self.manipulate_bytecode('_temp', mapping,
+                                                lambda bc: bc[:8] + b'<test>',
+                                                del_source=del_source)
+            file_path = mapping['_temp'] if not del_source else bytecode_path
+            with self.assertRaises(ValueError):
+                self.import_(file_path, '_temp')
+
+    def _test_bad_magic(self, test, *, del_source=False):
+        with source_util.create_modules('_temp') as mapping:
+            bc_path = self.manipulate_bytecode('_temp', mapping,
+                                    lambda bc: b'\x00\x00\x00\x00' + bc[4:])
+            test('_temp', mapping, bc_path)
+
+
+class SourceLoaderBadBytecodeTest(BadBytecodeTest):
+
+    loader = _bootstrap._SourceFileLoader
+
+    @source_util.writes_bytecode_files
+    def test_empty_file(self):
+        # When a .pyc is empty, regenerate it if possible, else raise
+        # ImportError.
+        def test(name, mapping, bytecode_path):
+            self.import_(mapping[name], name)
+            with open(bytecode_path, 'rb') as file:
+                self.assertGreater(len(file.read()), 8)
+
+        self._test_empty_file(test)
+
+    def test_partial_magic(self):
+        def test(name, mapping, bytecode_path):
+            self.import_(mapping[name], name)
+            with open(bytecode_path, 'rb') as file:
+                self.assertGreater(len(file.read()), 8)
+
+        self._test_partial_magic(test)
 
     @source_util.writes_bytecode_files
     def test_magic_only(self):
         # When there is only the magic number, regenerate the .pyc if possible,
         # else raise EOFError.
-        with source_util.create_modules('_temp') as mapping:
-            bc_path = self.manipulate_bytecode('_temp', mapping,
-                                                lambda bc: bc[:4])
-            self.import_(mapping['_temp'], '_temp')
-            with open(bc_path, 'rb') as file:
+        def test(name, mapping, bytecode_path):
+            self.import_(mapping[name], name)
+            with open(bytecode_path, 'rb') as file:
                 self.assertGreater(len(file.read()), 8)
-            self.manipulate_bytecode('_temp', mapping, lambda bc: bc[:4],
-                                        del_source=True)
-            with self.assertRaises(EOFError):
-                self.import_(mapping['_temp'], '_temp')
+
+    @source_util.writes_bytecode_files
+    def test_bad_magic(self):
+        # When the magic number is different, the bytecode should be
+        # regenerated.
+        def test(name, mapping, bytecode_path):
+            self.import_(mapping[name], name)
+            with open(bytecode_path, 'rb') as bytecode_file:
+                self.assertEqual(bytecode_file.read(4), imp.get_magic())
+
+        self._test_bad_magic(test)
 
     @source_util.writes_bytecode_files
     def test_partial_timestamp(self):
         # When the timestamp is partial, regenerate the .pyc, else
         # raise EOFError.
-        with source_util.create_modules('_temp') as mapping:
-            bc_path = self.manipulate_bytecode('_temp', mapping,
-                                                lambda bc: bc[:7])
-            self.import_(mapping['_temp'], '_temp')
+        def test(name, mapping, bc_path):
+            self.import_(mapping[name], name)
             with open(bc_path, 'rb') as file:
                 self.assertGreater(len(file.read()), 8)
-            self.manipulate_bytecode('_temp', mapping, lambda bc: bc[:7],
-                                        del_source=True)
-            with self.assertRaises(EOFError):
-                self.import_(mapping['_temp'], '_temp')
 
     @source_util.writes_bytecode_files
     def test_no_marshal(self):
         # When there is only the magic number and timestamp, raise EOFError.
-        with source_util.create_modules('_temp') as mapping:
-            bc_path = self.manipulate_bytecode('_temp', mapping,
-                                                lambda bc: bc[:8])
-            with self.assertRaises(EOFError):
-                self.import_(mapping['_temp'], '_temp')
+        self._test_no_marshal()
 
     @source_util.writes_bytecode_files
-    def test_bad_magic(self):
-        # When the magic number is different, the bytecode should be
-        # regenerated.
-        with source_util.create_modules('_temp') as mapping:
-            bc_path = self.manipulate_bytecode('_temp', mapping,
-                                    lambda bc: b'\x00\x00\x00\x00' + bc[4:])
-            self.import_(mapping['_temp'], '_temp')
-            with open(bc_path, 'rb') as bytecode_file:
-                self.assertEqual(bytecode_file.read(4), imp.get_magic())
+    def test_non_code_marshal(self):
+        self._test_non_code_marshal()
+        # XXX ImportError when sourceless
+
+    # [bad marshal]
+    @source_util.writes_bytecode_files
+    def test_bad_marshal(self):
+        # Bad marshal data should raise a ValueError.
+        self._test_bad_marshal()
 
     # [bad timestamp]
     @source_util.writes_bytecode_files
-    def test_bad_bytecode(self):
+    def test_old_timestamp(self):
         # When the timestamp is older than the source, bytecode should be
         # regenerated.
         zeros = b'\x00\x00\x00\x00'
@@ -240,23 +291,6 @@
                 bytecode_file.seek(4)
                 self.assertEqual(bytecode_file.read(4), source_timestamp)
 
-    # [bad marshal]
-    @source_util.writes_bytecode_files
-    def test_bad_marshal(self):
-        # Bad marshal data should raise a ValueError.
-        with source_util.create_modules('_temp') as mapping:
-            bytecode_path = imp.cache_from_source(mapping['_temp'])
-            source_mtime = os.path.getmtime(mapping['_temp'])
-            source_timestamp = importlib._w_long(source_mtime)
-            source_util.ensure_bytecode_path(bytecode_path)
-            with open(bytecode_path, 'wb') as bytecode_file:
-                bytecode_file.write(imp.get_magic())
-                bytecode_file.write(source_timestamp)
-                bytecode_file.write(b'AAAA')
-            with self.assertRaises(ValueError):
-                self.import_(mapping['_temp'], '_temp')
-            self.assertTrue('_temp' not in sys.modules)
-
     # [bytecode read-only]
     @source_util.writes_bytecode_files
     def test_read_only_bytecode(self):
@@ -279,9 +313,57 @@
                 os.chmod(bytecode_path, stat.S_IWUSR)
 
 
+class SourcelessLoaderBadBytecodeTest(BadBytecodeTest):
+
+    loader = _bootstrap._SourcelessFileLoader
+
+    def test_empty_file(self):
+        def test(name, mapping, bytecode_path):
+            with self.assertRaises(ImportError):
+                self.import_(bytecode_path, name)
+
+        self._test_empty_file(test, del_source=True)
+
+    def test_partial_magic(self):
+        def test(name, mapping, bytecode_path):
+            with self.assertRaises(ImportError):
+                self.import_(bytecode_path, name)
+        self._test_partial_magic(test, del_source=True)
+
+    def test_magic_only(self):
+        def test(name, mapping, bytecode_path):
+            with self.assertRaises(EOFError):
+                self.import_(bytecode_path, name)
+
+        self._test_magic_only(test, del_source=True)
+
+    def test_bad_magic(self):
+        def test(name, mapping, bytecode_path):
+            with self.assertRaises(ImportError):
+                self.import_(bytecode_path, name)
+
+        self._test_bad_magic(test, del_source=True)
+
+    def test_partial_timestamp(self):
+        def test(name, mapping, bytecode_path):
+            with self.assertRaises(EOFError):
+                self.import_(bytecode_path, name)
+
+        self._test_partial_timestamp(test, del_source=True)
+
+    def test_no_marshal(self):
+        self._test_no_marshal(del_source=True)
+
+    def test_non_code_marshal(self):
+        self._test_non_code_marshal(del_source=True)
+
+
 def test_main():
     from test.support import run_unittest
-    run_unittest(SimpleTest, BadBytecodeTest)
+    run_unittest(SimpleTest,
+                 SourceLoaderBadBytecodeTest,
+                 SourcelessLoaderBadBytecodeTest
+                 )
 
 
 if __name__ == '__main__':

Modified: python/branches/py3k/Lib/importlib/test/source/test_finder.py
==============================================================================
--- python/branches/py3k/Lib/importlib/test/source/test_finder.py	(original)
+++ python/branches/py3k/Lib/importlib/test/source/test_finder.py	Sat Jul  3 23:48:25 2010
@@ -34,7 +34,9 @@
     """
 
     def import_(self, root, module):
-        finder = _bootstrap._PyPycFileFinder(root)
+        finder = _bootstrap._FileFinder(root,
+                                        _bootstrap._SourceFinderDetails(),
+                                        _bootstrap._SourcelessFinderDetails())
         return finder.find_module(module)
 
     def run_test(self, test, create=None, *, compile_=None, unlink=None):
@@ -116,7 +118,7 @@
         # XXX This is not a blackbox test!
         name = '_temp'
         loader = self.run_test(name, {'{0}.__init__'.format(name), name})
-        self.assertTrue('__init__' in loader._base_path)
+        self.assertTrue('__init__' in loader.get_filename(name))
 
 
     def test_failure(self):

Modified: python/branches/py3k/Lib/importlib/test/source/test_path_hook.py
==============================================================================
--- python/branches/py3k/Lib/importlib/test/source/test_path_hook.py	(original)
+++ python/branches/py3k/Lib/importlib/test/source/test_path_hook.py	Sat Jul  3 23:48:25 2010
@@ -8,9 +8,8 @@
     """Test the path hook for source."""
 
     def test_success(self):
-        # XXX Only work on existing directories?
         with source_util.create_modules('dummy') as mapping:
-            self.assertTrue(hasattr(_bootstrap._FileFinder(mapping['.root']),
+            self.assertTrue(hasattr(_bootstrap._file_path_hook(mapping['.root']),
                                  'find_module'))
 
 

Modified: python/branches/py3k/Lib/importlib/test/source/test_source_encoding.py
==============================================================================
--- python/branches/py3k/Lib/importlib/test/source/test_source_encoding.py	(original)
+++ python/branches/py3k/Lib/importlib/test/source/test_source_encoding.py	Sat Jul  3 23:48:25 2010
@@ -35,8 +35,8 @@
         with source_util.create_modules(self.module_name) as mapping:
             with open(mapping[self.module_name], 'wb') as file:
                 file.write(source)
-            loader = _bootstrap._PyPycFileLoader(self.module_name,
-                                       mapping[self.module_name], False)
+            loader = _bootstrap._SourceFileLoader(self.module_name,
+                                       mapping[self.module_name])
             return loader.load_module(self.module_name)
 
     def create_source(self, encoding):
@@ -97,8 +97,8 @@
         with source_util.create_modules(module_name) as mapping:
             with open(mapping[module_name], 'wb') as file:
                 file.write(source)
-            loader = _bootstrap._PyPycFileLoader(module_name,
-                                                 mapping[module_name], False)
+            loader = _bootstrap._SourceFileLoader(module_name,
+                                                 mapping[module_name])
             return loader.load_module(module_name)
 
     # [cr]

Modified: python/branches/py3k/Lib/test/test_runpy.py
==============================================================================
--- python/branches/py3k/Lib/test/test_runpy.py	(original)
+++ python/branches/py3k/Lib/test/test_runpy.py	Sat Jul  3 23:48:25 2010
@@ -6,7 +6,7 @@
 import re
 import tempfile
 import py_compile
-from test.support import forget, make_legacy_pyc, run_unittest, verbose
+from test.support import forget, make_legacy_pyc, run_unittest, unload, verbose
 from test.script_helper import (
     make_pkg, make_script, make_zip_pkg, make_zip_script, temp_dir)
 
@@ -174,6 +174,7 @@
             __import__(mod_name)
             os.remove(mod_fname)
             make_legacy_pyc(mod_fname)
+            unload(mod_name)  # In case loader caches paths
             if verbose: print("Running from compiled:", mod_name)
             d2 = run_module(mod_name) # Read from bytecode
             self.assertIn("x", d2)
@@ -197,6 +198,7 @@
             __import__(mod_name)
             os.remove(mod_fname)
             make_legacy_pyc(mod_fname)
+            unload(mod_name)  # In case loader caches paths
             if verbose: print("Running from compiled:", pkg_name)
             d2 = run_module(pkg_name) # Read from bytecode
             self.assertIn("x", d2)
@@ -252,6 +254,7 @@
             __import__(mod_name)
             os.remove(mod_fname)
             make_legacy_pyc(mod_fname)
+            unload(mod_name)  # In case the loader caches paths
             if verbose: print("Running from compiled:", mod_name)
             d2 = run_module(mod_name, run_name=run_name) # Read from bytecode
             self.assertIn("__package__", d2)
@@ -405,7 +408,11 @@
 
 
 def test_main():
-    run_unittest(RunModuleCodeTest, RunModuleTest, RunPathTest)
+    run_unittest(
+                 RunModuleCodeTest,
+                 RunModuleTest,
+                 RunPathTest
+                 )
 
 if __name__ == "__main__":
     test_main()

From python-checkins at python.org  Sat Jul  3 23:50:52 2010
From: python-checkins at python.org (brett.cannon)
Date: Sat,  3 Jul 2010 23:50:52 +0200 (CEST)
Subject: [Python-checkins] r82519 - python/branches/py3k/Misc/Vim/vimrc
Message-ID: <20100703215052.48B17EC40@mail.python.org>

Author: brett.cannon
Date: Sat Jul  3 23:50:52 2010
New Revision: 82519

Log:
Trailing whitespace is bad for .rst files.

Modified:
   python/branches/py3k/Misc/Vim/vimrc

Modified: python/branches/py3k/Misc/Vim/vimrc
==============================================================================
--- python/branches/py3k/Misc/Vim/vimrc	(original)
+++ python/branches/py3k/Misc/Vim/vimrc	Sat Jul  3 23:50:52 2010
@@ -43,7 +43,7 @@
 " Display tabs at the beginning of a line in Python mode as bad.
 au BufRead,BufNewFile *.py,*.pyw match BadWhitespace /^\t\+/
 " Make trailing whitespace be flagged as bad.
-au BufRead,BufNewFile *.py,*.pyw,*.c,*.h match BadWhitespace /\s\+$/
+au BufRead,BufNewFile *.py,*.pyw,*.c,*.h,*.rst match BadWhitespace /\s\+$/
 
 " Wrap text after a certain number of characters
 " Python: 79 

From python-checkins at python.org  Sun Jul  4 00:03:16 2010
From: python-checkins at python.org (brett.cannon)
Date: Sun,  4 Jul 2010 00:03:16 +0200 (CEST)
Subject: [Python-checkins] r82520 -
	python/branches/py3k/Lib/importlib/test/test_api.py
Message-ID: <20100703220316.E611EF2DA@mail.python.org>

Author: brett.cannon
Date: Sun Jul  4 00:03:16 2010
New Revision: 82520

Log:
Fix a spelling mistake in a comment.

Modified:
   python/branches/py3k/Lib/importlib/test/test_api.py

Modified: python/branches/py3k/Lib/importlib/test/test_api.py
==============================================================================
--- python/branches/py3k/Lib/importlib/test/test_api.py	(original)
+++ python/branches/py3k/Lib/importlib/test/test_api.py	Sun Jul  4 00:03:16 2010
@@ -27,7 +27,7 @@
                 self.assertEqual(module.__name__, name)
 
     def test_shallow_relative_package_import(self):
-        # Test importing a module from a package through a relatve import.
+        # Test importing a module from a package through a relative import.
         pkg_name = 'pkg'
         pkg_long_name = '{0}.__init__'.format(pkg_name)
         module_name = 'mod'

From python-checkins at python.org  Sun Jul  4 00:05:41 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Sun,  4 Jul 2010 00:05:41 +0200 (CEST)
Subject: [Python-checkins] r82521 -
	python/branches/py3k/Demo/classes/Dates.py
Message-ID: <20100703220541.3BBA1EE9A4@mail.python.org>

Author: alexander.belopolsky
Date: Sun Jul  4 00:05:41 2010
New Revision: 82521

Log:
Issue #9151: Demo/classes/Dates.py does not work in 3.x Converted
descriptive comment into a docstring.  Cast attributes to int in
__init__.  Use __new__ instead of deleting attributes to
"uninitialize".


Modified:
   python/branches/py3k/Demo/classes/Dates.py

Modified: python/branches/py3k/Demo/classes/Dates.py
==============================================================================
--- python/branches/py3k/Demo/classes/Dates.py	(original)
+++ python/branches/py3k/Demo/classes/Dates.py	Sun Jul  4 00:05:41 2010
@@ -1,43 +1,45 @@
-# Class Date supplies date objects that support date arithmetic.
-#
-# Date(month,day,year) returns a Date object.  An instance prints as,
-# e.g., 'Mon 16 Aug 1993'.
-#
-# Addition, subtraction, comparison operators, min, max, and sorting
-# all work as expected for date objects:  int+date or date+int returns
-# the date `int' days from `date'; date+date raises an exception;
-# date-int returns the date `int' days before `date'; date2-date1 returns
-# an integer, the number of days from date1 to date2; int-date raises an
-# exception; date1 < date2 is true iff date1 occurs before date2 (&
-# similarly for other comparisons); min(date1,date2) is the earlier of
-# the two dates and max(date1,date2) the later; and date objects can be
-# used as dictionary keys.
-#
-# Date objects support one visible method, date.weekday().  This returns
-# the day of the week the date falls on, as a string.
-#
-# Date objects also have 4 read-only data attributes:
-#   .month  in 1..12
-#   .day    in 1..31
-#   .year   int or long int
-#   .ord    the ordinal of the date relative to an arbitrary staring point
-#
-# The Dates module also supplies function today(), which returns the
-# current date as a date object.
-#
-# Those entranced by calendar trivia will be disappointed, as no attempt
-# has been made to accommodate the Julian (etc) system.  On the other
-# hand, at least this package knows that 2000 is a leap year but 2100
-# isn't, and works fine for years with a hundred decimal digits <wink>.
+"""
+Class Date supplies date objects that support date arithmetic.
 
-# Tim Peters   tim at ksr.com
-# not speaking for Kendall Square Research Corp
+Date(month,day,year) returns a Date object.  An instance prints as,
+e.g., 'Mon 16 Aug 1993'.
 
-# Adapted to Python 1.1 (where some hacks to overcome coercion are unnecessary)
-# by Guido van Rossum
-
-# Note that as of Python 2.3, a datetime module is included in the stardard
-# library.
+Addition, subtraction, comparison operators, min, max, and sorting
+all work as expected for date objects:  int+date or date+int returns
+the date `int' days from `date'; date+date raises an exception;
+date-int returns the date `int' days before `date'; date2-date1 returns
+an integer, the number of days from date1 to date2; int-date raises an
+exception; date1 < date2 is true iff date1 occurs before date2 (&
+similarly for other comparisons); min(date1,date2) is the earlier of
+the two dates and max(date1,date2) the later; and date objects can be
+used as dictionary keys.
+
+Date objects support one visible method, date.weekday().  This returns
+the day of the week the date falls on, as a string.
+
+Date objects also have 4 read-only data attributes:
+  .month  in 1..12
+  .day    in 1..31
+  .year   int or long int
+  .ord    the ordinal of the date relative to an arbitrary staring point
+
+The Dates module also supplies function today(), which returns the
+current date as a date object.
+
+Those entranced by calendar trivia will be disappointed, as no attempt
+has been made to accommodate the Julian (etc) system.  On the other
+hand, at least this package knows that 2000 is a leap year but 2100
+isn't, and works fine for years with a hundred decimal digits <wink>.
+
+Tim Peters   tim at ksr.com
+not speaking for Kendall Square Research Corp
+
+Adapted to Python 1.1 (where some hacks to overcome coercion are unnecessary)
+by Guido van Rossum
+
+Note that as of Python 2.3, a datetime module is included in the stardard
+library.
+"""
 
 import functools
 
@@ -86,8 +88,9 @@
     if not isinstance(n, int):
         raise TypeError('argument must be integer: %r' % type(n))
 
-    ans = Date(1,1,1)   # arguments irrelevant; just getting a Date obj
-    del ans.ord, ans.month, ans.day, ans.year # un-initialize it
+    # Get uninitialized Date object.  This is necesary because once
+    # attributes are set, they cannot be changed.
+    ans = Date.__new__(Date)
     ans.ord = n
 
     n400 = (n-1)//_DI400Y                # # of 400-year blocks preceding
@@ -97,11 +100,7 @@
     if dby >= n:
         more = more - 1
         dby = dby - _days_in_year(more)
-    year, n = year + more, int(n - dby)
-
-    try: year = int(year)               # chop to int, if it fits
-    except (ValueError, OverflowError): pass
-
+    year, n = year + more, n - dby
     month = min(n//29 + 1, 12)
     dbm = _days_before_month(month, year)
     if dbm >= n:
@@ -112,7 +111,7 @@
     return ans
 
 def _num2day(n):      # return weekday name of day with ordinal n
-    return _DAY_NAMES[ int(n % 7) ]
+    return _DAY_NAMES[n % 7]
 
 @functools.total_ordering
 class Date:
@@ -122,7 +121,7 @@
         dim = _days_in_month(month, year)
         if not 1 <= day <= dim:
             raise ValueError('day must be in 1..%r: %r' % (dim, day))
-        self.month, self.day, self.year = month, day, year
+        self.month, self.day, self.year = map(int, (month, day, year))
         self.ord = _date2num(self)
 
     # don't allow setting existing attributes

From python-checkins at python.org  Sun Jul  4 00:18:47 2010
From: python-checkins at python.org (brett.cannon)
Date: Sun,  4 Jul 2010 00:18:47 +0200 (CEST)
Subject: [Python-checkins] r82522 - in
	python/branches/py3k/Lib/importlib/test/source:
	test_file_loader.py test_finder.py test_path_hook.py
Message-ID: <20100703221847.C9A28CA36@mail.python.org>

Author: brett.cannon
Date: Sun Jul  4 00:18:47 2010
New Revision: 82522

Log:
Make importlib.abc.SourceLoader the primary mechanism for importlib.

This required moving the class from importlib/abc.py into
importlib/_bootstrap.py and jiggering some code to work better with the class.
This included changing how the file finder worked to better meet import
semantics. This also led to fixing importlib to handle the empty string from
sys.path as import currently does (and making me wish we didn't support that
instead just required people to insert '.' instead to represent cwd).

It also required making the new set_data abstractmethod create
any needed subdirectories implicitly thanks to __pycache__ (it was either this
or grow the SourceLoader ABC to gain an 'exists' method and either a mkdir
method or have set_data with no data arg mean to create a directory).

Lastly, as an optimization the file loaders cache the file path where the
finder found something to use for loading (this is thanks to having a
sourceless loader separate from the source loader to simplify the code and
cut out stat calls).
Unfortunately test_runpy assumed a loader would always work for a module, even
if you changed from underneath it what it was expected to work with. By simply
dropping the previous loader in test_runpy so the proper loader can be returned
by the finder fixed the failure.

At this point importlib deviates from import on two points:

1. The exception raised when trying to import a file is different (import does
an explicit file check to print a special message, importlib just says the path
cannot be imported as if it was just some module name).

2. the co_filename on a code object is not being set to where bytecode was
actually loaded from instead of where the marshalled code object originally
came from (a solution for this has already been agreed upon on python-dev but has
not been implemented yet; issue8611).


Modified:
   python/branches/py3k/Lib/importlib/test/source/test_file_loader.py
   python/branches/py3k/Lib/importlib/test/source/test_finder.py
   python/branches/py3k/Lib/importlib/test/source/test_path_hook.py

Modified: python/branches/py3k/Lib/importlib/test/source/test_file_loader.py
==============================================================================
--- python/branches/py3k/Lib/importlib/test/source/test_file_loader.py	(original)
+++ python/branches/py3k/Lib/importlib/test/source/test_file_loader.py	Sun Jul  4 00:18:47 2010
@@ -1,6 +1,7 @@
 import importlib
 from importlib import _bootstrap
 from .. import abc
+from .. import util
 from . import util as source_util
 
 import imp
@@ -108,6 +109,22 @@
                 loader.load_module('_temp')
             self.assertTrue('_temp' not in sys.modules)
 
+    def test_file_from_empty_string_dir(self):
+        # Loading a module found from an empty string entry on sys.path should
+        # not only work, but keep all attributes relative.
+        with open('_temp.py', 'w') as file:
+            file.write("# test file for importlib")
+        try:
+            with util.uncache('_temp'):
+                loader = _bootstrap._SourceFileLoader('_temp', '_temp.py')
+                mod = loader.load_module('_temp')
+                self.assertEqual('_temp.py', mod.__file__)
+                self.assertEqual(imp.cache_from_source('_temp.py'),
+                                 mod.__cached__)
+
+        finally:
+            os.unlink('_temp.py')
+
 
 class BadBytecodeTest(unittest.TestCase):
 

Modified: python/branches/py3k/Lib/importlib/test/source/test_finder.py
==============================================================================
--- python/branches/py3k/Lib/importlib/test/source/test_finder.py	(original)
+++ python/branches/py3k/Lib/importlib/test/source/test_finder.py	Sun Jul  4 00:18:47 2010
@@ -115,7 +115,6 @@
 
     # [package over modules]
     def test_package_over_module(self):
-        # XXX This is not a blackbox test!
         name = '_temp'
         loader = self.run_test(name, {'{0}.__init__'.format(name), name})
         self.assertTrue('__init__' in loader.get_filename(name))
@@ -133,6 +132,17 @@
             with self.assertRaises(ImportWarning):
                 self.run_test('pkg', {'pkg.__init__'}, unlink={'pkg.__init__'})
 
+    def test_empty_string_for_dir(self):
+        # The empty string from sys.path means to search in the cwd.
+        finder = _bootstrap._FileFinder('', _bootstrap._SourceFinderDetails())
+        with open('mod.py', 'w') as file:
+            file.write("# test file for importlib")
+        try:
+            loader = finder.find_module('mod')
+            self.assertTrue(hasattr(loader, 'load_module'))
+        finally:
+            os.unlink('mod.py')
+
 
 def test_main():
     from test.support import run_unittest

Modified: python/branches/py3k/Lib/importlib/test/source/test_path_hook.py
==============================================================================
--- python/branches/py3k/Lib/importlib/test/source/test_path_hook.py	(original)
+++ python/branches/py3k/Lib/importlib/test/source/test_path_hook.py	Sun Jul  4 00:18:47 2010
@@ -12,6 +12,10 @@
             self.assertTrue(hasattr(_bootstrap._file_path_hook(mapping['.root']),
                                  'find_module'))
 
+    def test_empty_string(self):
+        # The empty string represents the cwd.
+        self.assertTrue(hasattr(_bootstrap._file_path_hook(''), 'find_module'))
+
 
 def test_main():
     from test.support import run_unittest

From python-checkins at python.org  Sun Jul  4 00:32:41 2010
From: python-checkins at python.org (brett.cannon)
Date: Sun,  4 Jul 2010 00:32:41 +0200 (CEST)
Subject: [Python-checkins] r82523 - in python/branches/py3k/Lib/importlib:
	_bootstrap.py test/source/test_abc_loader.py
Message-ID: <20100703223241.E5CE7EFBA@mail.python.org>

Author: brett.cannon
Date: Sun Jul  4 00:32:41 2010
New Revision: 82523

Log:
Make importlib.abc.SourceLoader the primary mechanism for importlib.

This required moving the class from importlib/abc.py into
importlib/_bootstrap.py and jiggering some code to work better with the class.
This included changing how the file finder worked to better meet import
semantics. This also led to fixing importlib to handle the empty string from
sys.path as import currently does (and making me wish we didn't support that
instead just required people to insert '.' instead to represent cwd).

It also required making the new set_data abstractmethod create
any needed subdirectories implicitly thanks to __pycache__ (it was either this
or grow the SourceLoader ABC to gain an 'exists' method and either a mkdir
method or have set_data with no data arg mean to create a directory).

Lastly, as an optimization the file loaders cache the file path where the
finder found something to use for loading (this is thanks to having a
sourceless loader separate from the source loader to simplify the code and
cut out stat calls).
Unfortunately test_runpy assumed a loader would always work for a module, even
if you changed from underneath it what it was expected to work with. By simply
dropping the previous loader in test_runpy so the proper loader can be returned
by the finder fixed the failure.

At this point importlib deviates from import on two points:

1. The exception raised when trying to import a file is different (import does
an explicit file check to print a special message, importlib just says the path
cannot be imported as if it was just some module name).

2. the co_filename on a code object is not being set to where bytecode was
actually loaded from instead of where the marshalled code object originally
came from (a solution for this has already been agreed upon on python-dev but has
not been implemented yet; issue8611).


Modified:
   python/branches/py3k/Lib/importlib/_bootstrap.py
   python/branches/py3k/Lib/importlib/test/source/test_abc_loader.py

Modified: python/branches/py3k/Lib/importlib/_bootstrap.py
==============================================================================
--- python/branches/py3k/Lib/importlib/_bootstrap.py	(original)
+++ python/branches/py3k/Lib/importlib/_bootstrap.py	Sun Jul  4 00:32:41 2010
@@ -386,8 +386,8 @@
         except IOError:
             raise ImportError("source not available through get_data()")
         encoding = tokenize.detect_encoding(_io.BytesIO(source_bytes).readline)
-        # XXX Universal newlines?
-        return source_bytes.decode(encoding[0])
+        newline_decoder = _io.IncrementalNewlineDecoder(None, True)
+        return newline_decoder.decode(source_bytes.decode(encoding[0]))
 
     def get_code(self, fullname):
         """Concrete implementation of InspectLoader.get_code.

Modified: python/branches/py3k/Lib/importlib/test/source/test_abc_loader.py
==============================================================================
--- python/branches/py3k/Lib/importlib/test/source/test_abc_loader.py	(original)
+++ python/branches/py3k/Lib/importlib/test/source/test_abc_loader.py	Sun Jul  4 00:32:41 2010
@@ -7,6 +7,7 @@
 
 import imp
 import inspect
+import io
 import marshal
 import os
 import sys
@@ -327,30 +328,6 @@
             mock.get_filename(name)
 
 
-class PyLoaderGetSourceTests(unittest.TestCase):
-
-    """Tests for importlib.abc.PyLoader.get_source()."""
-
-    def test_default_encoding(self):
-        # Should have no problems with UTF-8 text.
-        name = 'mod'
-        mock = PyLoaderMock({name: os.path.join('path', 'to', 'mod')})
-        source = 'x = "?"'
-        mock.source = source.encode('utf-8')
-        returned_source = mock.get_source(name)
-        self.assertEqual(returned_source, source)
-
-    def test_decoded_source(self):
-        # Decoding should work.
-        name = 'mod'
-        mock = PyLoaderMock({name: os.path.join('path', 'to', 'mod')})
-        source = "# coding: Latin-1\nx='?'"
-        assert source.encode('latin-1') != source.encode('utf-8')
-        mock.source = source.encode('latin-1')
-        returned_source = mock.get_source(name)
-        self.assertEqual(returned_source, source)
-
-
 class PyPycLoaderTests(PyLoaderTests):
 
     """Tests for importlib.abc.PyPycLoader."""
@@ -752,6 +729,39 @@
         code_object = self.loader.get_code(self.name)
         self.verify_code(code_object)
 
+
+class SourceLoaderGetSourceTests(unittest.TestCase):
+
+    """Tests for importlib.abc.SourceLoader.get_source()."""
+
+    def test_default_encoding(self):
+        # Should have no problems with UTF-8 text.
+        name = 'mod'
+        mock = SourceOnlyLoaderMock('mod.file')
+        source = 'x = "?"'
+        mock.source = source.encode('utf-8')
+        returned_source = mock.get_source(name)
+        self.assertEqual(returned_source, source)
+
+    def test_decoded_source(self):
+        # Decoding should work.
+        name = 'mod'
+        mock = SourceOnlyLoaderMock("mod.file")
+        source = "# coding: Latin-1\nx='?'"
+        assert source.encode('latin-1') != source.encode('utf-8')
+        mock.source = source.encode('latin-1')
+        returned_source = mock.get_source(name)
+        self.assertEqual(returned_source, source)
+
+    def test_universal_newlines(self):
+        # PEP 302 says universal newlines should be used.
+        name = 'mod'
+        mock = SourceOnlyLoaderMock('mod.file')
+        source = "x = 42\r\ny = -13\r\n"
+        mock.source = source.encode('utf-8')
+        expect = io.IncrementalNewlineDecoder(None, True).decode(source)
+        self.assertEqual(mock.get_source(name), expect)
+
 class AbstractMethodImplTests(unittest.TestCase):
 
     """Test the concrete abstractmethod implementations."""
@@ -852,12 +862,13 @@
 def test_main():
     from test.support import run_unittest
     run_unittest(PyLoaderTests, PyLoaderCompatTests,
-                    PyLoaderInterfaceTests, PyLoaderGetSourceTests,
+                    PyLoaderInterfaceTests,
                     PyPycLoaderTests, PyPycLoaderInterfaceTests,
                     SkipWritingBytecodeTests, RegeneratedBytecodeTests,
                     BadBytecodeFailureTests, MissingPathsTests,
                     SourceOnlyLoaderTests,
                     SourceLoaderBytecodeTests,
+                    SourceLoaderGetSourceTests,
                     AbstractMethodImplTests)
 
 

From python-checkins at python.org  Sun Jul  4 00:36:07 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Sun,  4 Jul 2010 00:36:07 +0200 (CEST)
Subject: [Python-checkins] r82524 -
	python/branches/py3k/Demo/classes/Complex.py
Message-ID: <20100703223607.250BDEFBA@mail.python.org>

Author: alexander.belopolsky
Date: Sun Jul  4 00:36:06 2010
New Revision: 82524

Log:
Made minimal modifications to pass included tests

Modified:
   python/branches/py3k/Demo/classes/Complex.py

Modified: python/branches/py3k/Demo/classes/Complex.py
==============================================================================
--- python/branches/py3k/Demo/classes/Complex.py	(original)
+++ python/branches/py3k/Demo/classes/Complex.py	Sun Jul  4 00:36:06 2010
@@ -151,13 +151,9 @@
             raise ValueError("can't convert Complex with nonzero im to float")
         return float(self.re)
 
-    def __cmp__(self, other):
+    def __eq__(self, other):
         other = ToComplex(other)
-        return cmp((self.re, self.im), (other.re, other.im))
-
-    def __rcmp__(self, other):
-        other = ToComplex(other)
-        return cmp(other, self)
+        return (self.re, self.im) == (other.re, other.im)
 
     def __bool__(self):
         return not (self.re == self.im == 0)
@@ -190,14 +186,14 @@
 
     __rmul__ = __mul__
 
-    def __div__(self, other):
+    def __truediv__(self, other):
         other = ToComplex(other)
         d = float(other.re*other.re + other.im*other.im)
         if not d: raise ZeroDivisionError('Complex division')
         return Complex((self.re*other.re + self.im*other.im) / d,
                        (self.im*other.re - self.re*other.im) / d)
 
-    def __rdiv__(self, other):
+    def __rtruediv__(self, other):
         other = ToComplex(other)
         return other / self
 
@@ -226,8 +222,9 @@
     print('       ', a, 'and', b, end=' ')
     try:
         result = eval(expr)
-    except:
-        result = sys.exc_info()[0]
+    except Exception as e:
+        print('!!\t!!\t!! error: {}'.format(e))
+        return
     print('->', result)
     if isinstance(result, str) or isinstance(value, str):
         ok = (result == value)
@@ -295,13 +292,6 @@
                     (Complex(1), Complex(0,10), 1),
                     (2, Complex(4,0), 16),
             ],
-            'cmp(a,b)': [
-                    (1, 10, -1),
-                    (1, Complex(0,10), 1),
-                    (Complex(0,10), 1, -1),
-                    (Complex(0,10), Complex(1), -1),
-                    (Complex(1), Complex(0,10), 1),
-            ],
     }
     for expr in sorted(testsuite):
         print(expr + ':')

From solipsis at pitrou.net  Sun Jul  4 01:25:33 2010
From: solipsis at pitrou.net (solipsis at pitrou.net)
Date: Sun,  4 Jul 2010 01:25:33 +0200 (CEST)
Subject: [Python-checkins] Daily py3k reference leaks (r82515): sum=9
Message-ID: <20100703232533.C23591771C@ns6635.ovh.net>

py3k results for svn r82515 (hg cset 25d975bb6fd5)
--------------------------------------------------

test_audioop leaked [3, 3, 3] references, sum=9


Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/antoine/py3k/refleaks/reflog8VuWBG', '-x']

From python-checkins at python.org  Sun Jul  4 03:35:45 2010
From: python-checkins at python.org (ezio.melotti)
Date: Sun,  4 Jul 2010 03:35:45 +0200 (CEST)
Subject: [Python-checkins] r82525 - in tracker/roundup-src:
	roundup/cgi/templating.py test/test_templating.py
Message-ID: <20100704013545.D70B8EE9A0@mail.python.org>

Author: ezio.melotti
Date: Sun Jul  4 03:35:45 2010
New Revision: 82525

Log:
#254: now URLs inside parentheses (e.g. www.python.org) are linkified properly.

Modified:
   tracker/roundup-src/roundup/cgi/templating.py
   tracker/roundup-src/test/test_templating.py

Modified: tracker/roundup-src/roundup/cgi/templating.py
==============================================================================
--- tracker/roundup-src/roundup/cgi/templating.py	(original)
+++ tracker/roundup-src/roundup/cgi/templating.py	Sun Jul  4 03:35:45 2010
@@ -1343,16 +1343,21 @@
             u = s = match.group('url')
             if not self.protocol_re.search(s):
                 u = 'http://' + s
-            # catch an escaped ">" at the end of the URL
             if s.endswith('&gt;'):
+                # catch an escaped ">" at the end of the URL
                 u = s = s[:-4]
                 e = '&gt;'
+            elif s.count('(') != s.count(')'):
+                # don't include extraneous ')' in the link
+                pos = s.rfind(')')
+                e = s[pos:]
+                u = s = s[:pos]
             else:
                 e = ''
-            return '<a href="%s">%s</a>%s'%(u, s, e)
+            return '<a href="%s">%s</a>%s' % (u, s, e)
         elif match.group('email'):
             s = match.group('email')
-            return '<a href="mailto:%s">%s</a>'%(s, s)
+            return '<a href="mailto:%s">%s</a>' % (s, s)
         elif len(match.group('id')) < 10:
             return self._hyper_repl_item(match,
                 '<a href="%(cls)s%(id)s">%(item)s</a>')

Modified: tracker/roundup-src/test/test_templating.py
==============================================================================
--- tracker/roundup-src/test/test_templating.py	(original)
+++ tracker/roundup-src/test/test_templating.py	Sun Jul  4 03:35:45 2010
@@ -147,9 +147,23 @@
         p = StringHTMLProperty(self.client, 'test', '1', None, 'test', '')
         def t(s): return p.hyper_re.sub(p._hyper_repl, s)
         ae = self.assertEquals
-        ae(t('http://roundup.net/'), '<a href="http://roundup.net/">http://roundup.net/</a>')
-        ae(t('&lt;HTTP://roundup.net/&gt;'), '&lt;<a href="HTTP://roundup.net/">HTTP://roundup.net/</a>&gt;')
-        ae(t('&lt;www.roundup.net&gt;'), '&lt;<a href="http://www.roundup.net">www.roundup.net</a>&gt;')
+        ae(t('http://roundup.net/'),
+           '<a href="http://roundup.net/">http://roundup.net/</a>')
+        ae(t('&lt;HTTP://roundup.net/&gt;'),
+           '&lt;<a href="HTTP://roundup.net/">HTTP://roundup.net/</a>&gt;')
+        ae(t('&lt;www.roundup.net&gt;'),
+           '&lt;<a href="http://www.roundup.net">www.roundup.net</a>&gt;')
+        ae(t('(www.roundup.net)'),
+           '(<a href="http://www.roundup.net">www.roundup.net</a>)')
+        ae(t('foo http://msdn.microsoft.com/en-us/library/ms741540(VS.85).aspx bar'),
+           'foo <a href="http://msdn.microsoft.com/en-us/library/ms741540(VS.85).aspx">'
+           'http://msdn.microsoft.com/en-us/library/ms741540(VS.85).aspx</a> bar')
+        ae(t('(e.g. http://en.wikipedia.org/wiki/Python_(programming_language))'),
+           '(e.g. <a href="http://en.wikipedia.org/wiki/Python_(programming_language)">'
+           'http://en.wikipedia.org/wiki/Python_(programming_language)</a>)')
+        ae(t('(e.g. http://en.wikipedia.org/wiki/Python_(programming_language)).'),
+           '(e.g. <a href="http://en.wikipedia.org/wiki/Python_(programming_language)">'
+           'http://en.wikipedia.org/wiki/Python_(programming_language)</a>).')
 
 '''
 class HTMLPermissions:

From python-checkins at python.org  Sun Jul  4 09:25:18 2010
From: python-checkins at python.org (georg.brandl)
Date: Sun,  4 Jul 2010 09:25:18 +0200 (CEST)
Subject: [Python-checkins] r82526 - python/branches/py3k/Demo/classes/Dbm.py
Message-ID: <20100704072518.A0C34EECF@mail.python.org>

Author: georg.brandl
Date: Sun Jul  4 09:25:18 2010
New Revision: 82526

Log:
Remove old and unsafe Dbm demo class.

Removed:
   python/branches/py3k/Demo/classes/Dbm.py

Deleted: python/branches/py3k/Demo/classes/Dbm.py
==============================================================================
--- python/branches/py3k/Demo/classes/Dbm.py	Sun Jul  4 09:25:18 2010
+++ (empty file)
@@ -1,66 +0,0 @@
-# A wrapper around the (optional) built-in class dbm, supporting keys
-# and values of almost any type instead of just string.
-# (Actually, this works only for keys and values that can be read back
-# correctly after being converted to a string.)
-
-
-class Dbm:
-
-    def __init__(self, filename, mode, perm):
-        import dbm.ndbm
-        self.db = dbm.ndbm.open(filename, mode, perm)
-
-    def __repr__(self):
-        s = ''
-        for key in self.keys():
-            t = repr(key) + ': ' + repr(self[key])
-            if s: t = ', ' + t
-            s = s + t
-        return '{' + s + '}'
-
-    def __len__(self):
-        return len(self.db)
-
-    def __getitem__(self, key):
-        return eval(self.db[repr(key)])
-
-    def __setitem__(self, key, value):
-        self.db[repr(key)] = repr(value)
-
-    def __delitem__(self, key):
-        del self.db[repr(key)]
-
-    def keys(self):
-        res = []
-        for key in self.db.keys():
-            res.append(eval(key))
-        return res
-
-    def has_key(self, key):
-        return repr(key) in self.db
-
-
-def test():
-    d = Dbm('@dbm', 'rw', 0o600)
-    print(d)
-    while 1:
-        try:
-            key = eval(input('key: '))
-            if key in d:
-                value = d[key]
-                print('currently:', value)
-            value = eval(input('value: '))
-            if value is None:
-                del d[key]
-            else:
-                d[key] = value
-        except KeyboardInterrupt:
-            print('')
-            print(d)
-        except EOFError:
-            print('[eof]')
-            break
-    print(d)
-
-
-test()

From python-checkins at python.org  Sun Jul  4 11:16:45 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:16:45 +0200
Subject: [Python-checkins] distutils2: [mq]: util.Mixin2to3,
	util.run_2to3 added(wrappers). tests added.
Message-ID: <E1OVLJd-0008FM-UC@dinsdale.python.org>

tarek.ziade pushed bf6057787636 to distutils2:

http://hg.python.org/distutils2/rev/bf6057787636
changeset:   253:bf6057787636
user:        Zubin Mithra <zubin.mithra at gmail.com>
date:        Sun Jul 04 02:46:38 2010 +0530
summary:     [mq]: util.Mixin2to3, util.run_2to3 added(wrappers). tests added.
files:       src/distutils2/tests/test_util.py, src/distutils2/util.py

diff --git a/src/distutils2/tests/test_util.py b/src/distutils2/tests/test_util.py
--- a/src/distutils2/tests/test_util.py
+++ b/src/distutils2/tests/test_util.py
@@ -296,6 +296,35 @@
         res = find_packages([root], ['pkg1.pkg2'])
         self.assertEqual(set(res), set(['pkg1', 'pkg5', 'pkg1.pkg3', 'pkg1.pkg3.pkg6']))
 
+    def test_run_2to3_on_code(self):
+        content = "print 'test'"
+        converted_content = "print('test')"
+        file_handle = tempfile.NamedTemporaryFile(delete=True)
+        file_name = file_handle.name
+        file_handle.write(content)
+        file_handle.flush()
+        file_handle.seek(0)
+        from distutils2.util import run_2to3
+        run_2to3([file_name])
+        new_content = "".join(file_handle.read())
+        file_handle.close()
+        self.assertEquals(new_content, converted_content)
+
+    def test_run_2to3_on_doctests(self):
+        # to check if text files containing doctests only get converted.
+        content = ">>> print 'test'\ntest\n"
+        converted_content = ">>> print('test')\ntest\n\n"
+        file_handle = tempfile.NamedTemporaryFile(delete=True)
+        file_name = file_handle.name
+        file_handle.write(content)
+        file_handle.flush()
+        file_handle.seek(0)
+        from distutils2.util import run_2to3
+        run_2to3([file_name], doctests_only=True)
+        new_content = "".join(file_handle.readlines())
+        file_handle.close()
+        self.assertEquals(new_content, converted_content)
+
 
 def test_suite():
     return unittest.makeSuite(UtilTestCase)
diff --git a/src/distutils2/util.py b/src/distutils2/util.py
--- a/src/distutils2/util.py
+++ b/src/distutils2/util.py
@@ -616,3 +616,48 @@
                 packages.append(package_name)
     return packages
 
+
+# utility functions for 2to3 support
+
+def run_2to3(files, doctests_only=False, fixer_names=None, options=None, 
+                                                            explicit=None):
+    """ Wrapper function around the refactor() class which
+	performs the conversions on a list of python files.
+	Invoke 2to3 on a list of Python files. The files should all come 
+	from the build area, as the modification is done in-place."""
+
+    if not files:
+        return
+
+    # Make this class local, to delay import of 2to3
+    from lib2to3.refactor import get_fixers_from_package
+    from distutils2.converter.refactor import DistutilsRefactoringTool
+
+    if fixer_names is None:
+        fixer_names = get_fixers_from_package('lib2to3.fixes')
+
+    r = DistutilsRefactoringTool(fixer_names, options=options)
+    if doctests_only:
+        r.refactor(files, doctests_only=True, write=True)
+    else:
+        r.refactor(files, write=True)
+
+class Mixin2to3:
+    """ Wrapper class for commands that run 2to3.
+    To configure 2to3, setup scripts may either change
+    the class variables, or inherit from this class
+    to override how 2to3 is invoked.
+    """
+    # provide list of fixers to run.
+    # defaults to all from lib2to3.fixers
+    fixer_names = None
+
+    # options dictionary
+    options = None
+
+    # list of fixers to invoke even though they are marked as explicit
+    explicit = None
+
+    def run_2to3(self, files, doctests_only=False):
+        """ Issues a call to util.run_2to3. """
+        return run_2to3(files, doctests_only, self.fixer_names, self.options, self.explicit)

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:16:45 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:16:45 +0200
Subject: [Python-checkins] distutils2: [mq]: test for build_py.Mixin2to3
	added
Message-ID: <E1OVLJd-0008FT-VK@dinsdale.python.org>

tarek.ziade pushed 47a7885ad97a to distutils2:

http://hg.python.org/distutils2/rev/47a7885ad97a
changeset:   255:47a7885ad97a
user:        Zubin Mithra <zubin.mithra at gmail.com>
date:        Sun Jul 04 02:51:59 2010 +0530
summary:     [mq]: test for build_py.Mixin2to3 added
files:       src/distutils2/tests/test_Mixin2to3.py

diff --git a/src/distutils2/tests/test_Mixin2to3.py b/src/distutils2/tests/test_Mixin2to3.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/test_Mixin2to3.py
@@ -0,0 +1,47 @@
+"""Tests for distutils.command.build_py."""
+
+import tempfile
+import distutils2
+from distutils2.tests import support
+from distutils2.tests.support import unittest
+from distutils2.command.build_py import Mixin2to3
+
+class Mixin2to3TestCase(unittest.TestCase):
+    def test_convert_code_only(self):
+        # used to check if code gets converted properly.
+        code_content = "print 'test'\n"
+        code_handle = tempfile.NamedTemporaryFile(delete=True)
+        code_name = code_handle.name
+
+        code_handle.write(code_content)
+        code_handle.flush()
+
+        mixin2to3 = Mixin2to3() 
+        mixin2to3._run_2to3([code_name])
+        converted_code_content = "print('test')\n"
+        new_code_content = "".join(open(code_name).readlines())
+
+        self.assertEquals(new_code_content, converted_code_content)
+
+    def test_doctests_only(self):
+        # used to check if doctests gets converted properly.
+        doctest_content = '"""\n>>> print test\ntest\n"""\nprint test\n\n'
+        doctest_handle = tempfile.NamedTemporaryFile(delete=True)
+        doctest_name = doctest_handle.name
+
+        doctest_handle.write(doctest_content)
+        doctest_handle.flush()
+
+        mixin2to3 = Mixin2to3()
+        mixin2to3._run_2to3([doctest_name])
+
+        converted_doctest_content = '"""\n>>> print(test)\ntest\n"""\nprint(test)\n\n\n'
+        new_doctest_content = "".join(open(doctest_name).readlines())
+
+        self.assertEquals(new_doctest_content, converted_doctest_content)
+
+def test_suite():
+    return unittest.makeSuite(Mixin2to3TestCase)
+
+if __name__ == "__main__":
+    unittest.main(defaultTest="test_suite")

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:16:45 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:16:45 +0200
Subject: [Python-checkins] distutils2: [mq]: core.py,
	dist.py modified to include use_2to3, convert_2to3_doctests
Message-ID: <E1OVLJd-0008FC-TU@dinsdale.python.org>

tarek.ziade pushed 961a8257a8b9 to distutils2:

http://hg.python.org/distutils2/rev/961a8257a8b9
changeset:   252:961a8257a8b9
user:        Zubin Mithra <zubin.mithra at gmail.com>
date:        Sun Jul 04 02:44:25 2010 +0530
summary:     [mq]: core.py, dist.py modified to include use_2to3, convert_2to3_doctests
files:       src/distutils2/core.py, src/distutils2/dist.py

diff --git a/src/distutils2/core.py b/src/distutils2/core.py
--- a/src/distutils2/core.py
+++ b/src/distutils2/core.py
@@ -49,7 +49,8 @@
                   'maintainer', 'maintainer_email', 'url', 'license',
                   'description', 'long_description', 'keywords',
                   'platforms', 'classifiers', 'download_url',
-                  'requires', 'provides', 'obsoletes',
+                  'requires', 'provides', 'obsoletes', 'use_2to3',
+                  'convert_2to3_doctests',
                   )
 
 # Legal keyword arguments for the Extension constructor
diff --git a/src/distutils2/dist.py b/src/distutils2/dist.py
--- a/src/distutils2/dist.py
+++ b/src/distutils2/dist.py
@@ -112,7 +112,11 @@
         ('requires', None,
          "print the list of packages/modules required"),
         ('obsoletes', None,
-         "print the list of packages/modules made obsolete")
+         "print the list of packages/modules made obsolete"),
+        ('use-2to3', None,
+         "use 2to3 to make source python 3.x compatible"),
+        ('convert-2to3-doctests', None,
+         "use 2to3 to convert doctests in seperate text files"), 
         ]
     display_option_names = map(lambda x: translate_longopt(x[0]),
                                display_options)
@@ -206,6 +210,8 @@
         self.scripts = None
         self.data_files = None
         self.password = ''
+        self.use_2to3 = False
+        self.convert_2to3_doctests = []
 
         # And now initialize bookkeeping stuff that can't be supplied by
         # the caller at all.  'command_obj' maps command names to
@@ -581,15 +587,11 @@
         instance, analogous to the .finalize_options() method of Command
         objects.
         """
-
-        # XXX conversion -- removed
-        #for attr in ('keywords', 'platforms'):
-        #    value = self.metadata.get_field(attr)
-        #    if value is None:
-        #        continue
-        #    if isinstance(value, str):
-        #        value = [elm.strip() for elm in value.split(',')]
-        #        setattr(self.metadata, attr, value)
+        if getattr(self, 'convert_2to3_doctests', None):
+            self.convert_2to3_doctests = [os.path.join(p) 
+                                for p in self.convert_2to3_doctests]
+        else:
+            self.convert_2to3_doctests = []
 
     def _show_help(self, parser, global_options=1, display_options=1,
                    commands=[]):

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:16:45 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:16:45 +0200
Subject: [Python-checkins] distutils2: [mq]: use_2to3,
	convert_2to3_doctests added to build_py.py
Message-ID: <E1OVLJd-0008FQ-Ul@dinsdale.python.org>

tarek.ziade pushed 1dae0574c333 to distutils2:

http://hg.python.org/distutils2/rev/1dae0574c333
changeset:   254:1dae0574c333
user:        Zubin Mithra <zubin.mithra at gmail.com>
date:        Sun Jul 04 02:49:48 2010 +0530
summary:     [mq]: use_2to3, convert_2to3_doctests added to build_py.py
files:       src/distutils2/__init__.py, src/distutils2/command/build_py.py

diff --git a/src/distutils2/__init__.py b/src/distutils2/__init__.py
--- a/src/distutils2/__init__.py
+++ b/src/distutils2/__init__.py
@@ -18,3 +18,9 @@
 
 __revision__ = "$Id: __init__.py 78020 2010-02-06 16:37:32Z benjamin.peterson $"
 __version__ = "1.0a2"
+
+
+# when set to True, converts doctests by default too 
+run_2to3_on_doctests = True
+# Standard package names for fixer packages
+lib2to3_fixer_packages = ['lib2to3.fixes']
diff --git a/src/distutils2/command/build_py.py b/src/distutils2/command/build_py.py
--- a/src/distutils2/command/build_py.py
+++ b/src/distutils2/command/build_py.py
@@ -6,14 +6,65 @@
 
 import os
 import sys
+import logging
 from glob import glob
 
 from distutils2.core import Command
 from distutils2.errors import DistutilsOptionError, DistutilsFileError
 from distutils2.util import convert_path
-from distutils2 import log
+from distutils2.converter.refactor import DistutilsRefactoringTool
 
-class build_py(Command):
+# marking public APIs
+__all__ = ['Mixin2to3', 'build_py']
+
+try:
+    from distutils2.util import Mixin2to3 as _Mixin2to3
+    from lib2to3.refactor import get_fixers_from_package
+    _CONVERT = True
+    _KLASS = _Mixin2to3
+except ImportError:
+    _CONVERT = False
+    _KLASS = object
+
+class Mixin2to3(_KLASS):
+    """ The base class which can be used for refactoring. When run under
+    Python 3.0, the run_2to3 method provided by Mixin2to3 is overridden. 
+    When run on Python 2.x, it merely creates a class which overrides run_2to3, 
+    yet does nothing in particular with it. 
+    """
+    if _CONVERT:
+        def _run_2to3(self, files, doctests=[]):
+            """ Takes a list of files and doctests, and performs conversion
+            on those. 
+              - First, the files which contain the code(`files`) are converted.
+              - Second, the doctests in `files` are converted.
+              - Thirdly, the doctests in `doctests` are converted.
+            """
+
+            # Convert the ".py" files.
+            logging.info("Converting Python code")
+            _KLASS.run_2to3(self, files)
+
+            # Convert the doctests in the ".py" files.
+            logging.info("Converting doctests with '.py' files")
+            _KLASS.run_2to3(self, files, doctests_only=True)
+
+            # If the following conditions are met, then convert:-
+            # 1. User has specified the 'convert_2to3_doctests' option. So, we
+            #    can expect that the list 'doctests' is not empty.
+            # 2. The default is allow distutils2 to allow conversion of text files
+            #    containing doctests. It is set as 
+            #    distutils2.run_2to3_on_doctests
+
+            if doctests != [] and distutils2.run_2to3_on_doctests:
+                logging.info("Converting text files which contain doctests")
+                _KLASS.run_2to3(self, doctests, doctests_only=True)
+    else:
+        # If run on Python 2.x, there is nothing to do.
+        def _run_2to3(self, files, doctests=[]):
+            pass
+
+class build_py(Command, Mixin2to3):
 
     description = "\"build\" pure Python modules (copy to build directory)"
 
@@ -39,6 +90,8 @@
         self.compile = 0
         self.optimize = 0
         self.force = None
+        self._updated_files = []
+        self._doctests_2to3 = []
 
     def finalize_options(self):
         self.set_undefined_options('build',
@@ -93,6 +146,9 @@
             self.build_packages()
             self.build_package_data()
 
+        if self.distribution.use_2to3 and self_updated_files:
+            self.run_2to3(self._updated_files, self._doctests_2to3)
+
         self.byte_compile(self.get_outputs(include_bytecode=0))
 
     def get_data_files(self):
@@ -137,8 +193,10 @@
             for filename in filenames:
                 target = os.path.join(build_dir, filename)
                 self.mkpath(os.path.dirname(target))
-                self.copy_file(os.path.join(src_dir, filename), target,
-                               preserve_mode=False)
+                outf, copied = self.copy_file(os.path.join(src_dir, filename),
+                               target, preserve_mode=False)
+                if copied and srcfile in self.distribution.convert_2to3.doctests:
+                    self._doctests_2to3.append(outf)
 
     def get_package_dir(self, package):
         """Return the directory, relative to the top of the source

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:16:45 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:16:45 +0200
Subject: [Python-checkins] distutils2: Making sure the code works on all
	Python versions.
Message-ID: <E1OVLJd-0008FY-W3@dinsdale.python.org>

tarek.ziade pushed f696d39e9e86 to distutils2:

http://hg.python.org/distutils2/rev/f696d39e9e86
changeset:   256:f696d39e9e86
tag:         tip
user:        Tarek Ziade <tarek at ziade.org>
date:        Sun Jul 04 11:16:37 2010 +0200
summary:     Making sure the code works on all Python versions.
files:       src/distutils2/command/build_py.py, src/distutils2/tests/support.py, src/distutils2/tests/test_Mixin2to3.py, src/distutils2/tests/test_util.py, src/distutils2/util.py

diff --git a/src/distutils2/command/build_py.py b/src/distutils2/command/build_py.py
--- a/src/distutils2/command/build_py.py
+++ b/src/distutils2/command/build_py.py
@@ -28,14 +28,14 @@
 
 class Mixin2to3(_KLASS):
     """ The base class which can be used for refactoring. When run under
-    Python 3.0, the run_2to3 method provided by Mixin2to3 is overridden. 
-    When run on Python 2.x, it merely creates a class which overrides run_2to3, 
-    yet does nothing in particular with it. 
+    Python 3.0, the run_2to3 method provided by Mixin2to3 is overridden.
+    When run on Python 2.x, it merely creates a class which overrides run_2to3,
+    yet does nothing in particular with it.
     """
     if _CONVERT:
         def _run_2to3(self, files, doctests=[]):
             """ Takes a list of files and doctests, and performs conversion
-            on those. 
+            on those.
               - First, the files which contain the code(`files`) are converted.
               - Second, the doctests in `files` are converted.
               - Thirdly, the doctests in `doctests` are converted.
@@ -53,7 +53,7 @@
             # 1. User has specified the 'convert_2to3_doctests' option. So, we
             #    can expect that the list 'doctests' is not empty.
             # 2. The default is allow distutils2 to allow conversion of text files
-            #    containing doctests. It is set as 
+            #    containing doctests. It is set as
             #    distutils2.run_2to3_on_doctests
 
             if doctests != [] and distutils2.run_2to3_on_doctests:
@@ -258,8 +258,8 @@
             if os.path.isfile(init_py):
                 return init_py
             else:
-                log.warn(("package init file '%s' not found " +
-                          "(or not a regular file)"), init_py)
+                logging.warning(("package init file '%s' not found " +
+                                 "(or not a regular file)"), init_py)
 
         # Either not in a package at all (__init__.py not expected), or
         # __init__.py doesn't exist -- so don't return the filename.
@@ -267,7 +267,8 @@
 
     def check_module(self, module, module_file):
         if not os.path.isfile(module_file):
-            log.warn("file %s (for module %s) not found", module_file, module)
+            logging.warning("file %s (for module %s) not found",
+                            module_file, module)
             return False
         else:
             return True
diff --git a/src/distutils2/tests/support.py b/src/distutils2/tests/support.py
--- a/src/distutils2/tests/support.py
+++ b/src/distutils2/tests/support.py
@@ -64,12 +64,22 @@
     def setUp(self):
         super(TempdirManager, self).setUp()
         self.tempdirs = []
+        self.tempfiles = []
 
     def tearDown(self):
         super(TempdirManager, self).tearDown()
         while self.tempdirs:
             d = self.tempdirs.pop()
             shutil.rmtree(d, os.name in ('nt', 'cygwin'))
+        for file_ in self.tempfiles:
+            if os.path.exists(file_):
+                os.remove(file_)
+
+    def mktempfile(self):
+        """Create a temporary file that will be cleaned up."""
+        tempfile_ = tempfile.NamedTemporaryFile()
+        self.tempfiles.append(tempfile_.name)
+        return tempfile_
 
     def mkdtemp(self):
         """Create a temporary directory that will be cleaned up.
diff --git a/src/distutils2/tests/test_Mixin2to3.py b/src/distutils2/tests/test_Mixin2to3.py
--- a/src/distutils2/tests/test_Mixin2to3.py
+++ b/src/distutils2/tests/test_Mixin2to3.py
@@ -1,32 +1,37 @@
 """Tests for distutils.command.build_py."""
+import sys
+import tempfile
 
-import tempfile
 import distutils2
 from distutils2.tests import support
 from distutils2.tests.support import unittest
 from distutils2.command.build_py import Mixin2to3
 
-class Mixin2to3TestCase(unittest.TestCase):
+
+class Mixin2to3TestCase(support.TempdirManager, unittest.TestCase):
+
+    @unittest.skipUnless(sys.version > '2.6', 'Need >= 2.6')
     def test_convert_code_only(self):
         # used to check if code gets converted properly.
         code_content = "print 'test'\n"
-        code_handle = tempfile.NamedTemporaryFile(delete=True)
+        code_handle = self.mktempfile()
         code_name = code_handle.name
 
         code_handle.write(code_content)
         code_handle.flush()
 
-        mixin2to3 = Mixin2to3() 
+        mixin2to3 = Mixin2to3()
         mixin2to3._run_2to3([code_name])
         converted_code_content = "print('test')\n"
         new_code_content = "".join(open(code_name).readlines())
 
         self.assertEquals(new_code_content, converted_code_content)
 
+    @unittest.skipUnless(sys.version > '2.6', 'Need >= 2.6')
     def test_doctests_only(self):
         # used to check if doctests gets converted properly.
         doctest_content = '"""\n>>> print test\ntest\n"""\nprint test\n\n'
-        doctest_handle = tempfile.NamedTemporaryFile(delete=True)
+        doctest_handle = self.mktempfile()
         doctest_name = doctest_handle.name
 
         doctest_handle.write(doctest_content)
@@ -35,7 +40,9 @@
         mixin2to3 = Mixin2to3()
         mixin2to3._run_2to3([doctest_name])
 
-        converted_doctest_content = '"""\n>>> print(test)\ntest\n"""\nprint(test)\n\n\n'
+        converted_doctest_content = ['"""', '>>> print(test)', 'test', '"""',
+                                     'print(test)', '', '', '']
+        converted_doctest_content = '\n'.join(converted_doctest_content)
         new_doctest_content = "".join(open(doctest_name).readlines())
 
         self.assertEquals(new_doctest_content, converted_doctest_content)
diff --git a/src/distutils2/tests/test_util.py b/src/distutils2/tests/test_util.py
--- a/src/distutils2/tests/test_util.py
+++ b/src/distutils2/tests/test_util.py
@@ -296,10 +296,11 @@
         res = find_packages([root], ['pkg1.pkg2'])
         self.assertEqual(set(res), set(['pkg1', 'pkg5', 'pkg1.pkg3', 'pkg1.pkg3.pkg6']))
 
+    @unittest.skipUnless(sys.version > '2.6', 'Need Python 2.6 or more')
     def test_run_2to3_on_code(self):
         content = "print 'test'"
         converted_content = "print('test')"
-        file_handle = tempfile.NamedTemporaryFile(delete=True)
+        file_handle = self.mktempfile()
         file_name = file_handle.name
         file_handle.write(content)
         file_handle.flush()
@@ -310,11 +311,12 @@
         file_handle.close()
         self.assertEquals(new_content, converted_content)
 
+    @unittest.skipUnless(sys.version > '2.6', 'Need Python 2.6 or more')
     def test_run_2to3_on_doctests(self):
         # to check if text files containing doctests only get converted.
         content = ">>> print 'test'\ntest\n"
         converted_content = ">>> print('test')\ntest\n\n"
-        file_handle = tempfile.NamedTemporaryFile(delete=True)
+        file_handle = self.mktempfile()
         file_name = file_handle.name
         file_handle.write(content)
         file_handle.flush()
diff --git a/src/distutils2/util.py b/src/distutils2/util.py
--- a/src/distutils2/util.py
+++ b/src/distutils2/util.py
@@ -5,7 +5,10 @@
 
 __revision__ = "$Id: util.py 77761 2010-01-26 22:46:15Z tarek.ziade $"
 
-import sys, os, string, re
+import sys
+import os
+import string
+import re
 from copy import copy
 from fnmatch import fnmatchcase
 
@@ -17,6 +20,7 @@
 
 _PLATFORM = None
 
+
 def newer(source, target):
     """Tells if the target is newer than the source.
 
@@ -37,6 +41,7 @@
 
     return os.stat(source).st_mtime > os.stat(target).st_mtime
 
+
 def get_platform():
     """Return a string that identifies the current platform.
 
@@ -48,6 +53,7 @@
         _PLATFORM = _sysconfig.get_platform()
     return _PLATFORM
 
+
 def set_platform(identifier):
     """Sets the platform string identifier returned by get_platform().
 
@@ -57,6 +63,7 @@
     global _PLATFORM
     _PLATFORM = identifier
 
+
 def convert_path(pathname):
     """Return 'pathname' as a name that will work on the native filesystem.
 
@@ -125,6 +132,7 @@
 
 _environ_checked = 0
 
+
 def check_environ():
     """Ensure that 'os.environ' has all the environment variables needed.
 
@@ -147,6 +155,7 @@
 
     _environ_checked = 1
 
+
 def subst_vars(s, local_vars):
     """Perform shell/Perl-style variable substitution on 'string'.
 
@@ -158,7 +167,8 @@
     variables not found in either 'local_vars' or 'os.environ'.
     """
     check_environ()
-    def _subst (match, local_vars=local_vars):
+
+    def _subst(match, local_vars=local_vars):
         var_name = match.group(1)
         if var_name in local_vars:
             return str(local_vars[var_name])
@@ -170,6 +180,7 @@
     except KeyError, var:
         raise ValueError("invalid variable '$%s'" % var)
 
+
 def grok_environment_error(exc, prefix="error: "):
     """Generate a useful error message from an EnvironmentError.
 
@@ -196,12 +207,14 @@
 # Needed by 'split_quoted()'
 _wordchars_re = _squote_re = _dquote_re = None
 
+
 def _init_regex():
     global _wordchars_re, _squote_re, _dquote_re
     _wordchars_re = re.compile(r'[^\\\'\"%s ]*' % string.whitespace)
     _squote_re = re.compile(r"'(?:[^'\\]|\\.)*'")
     _dquote_re = re.compile(r'"(?:[^"\\]|\\.)*"')
 
+
 def split_quoted(s):
     """Split a string up according to Unix shell-like rules for quotes and
     backslashes.
@@ -217,7 +230,8 @@
     # This is a nice algorithm for splitting up a single string, since it
     # doesn't require character-by-character examination.  It was a little
     # bit of a brain-bender to get it working right, though...
-    if _wordchars_re is None: _init_regex()
+    if _wordchars_re is None:
+        _init_regex()
 
     s = s.strip()
     words = []
@@ -237,8 +251,8 @@
 
         elif s[end] == '\\':            # preserve whatever is being escaped;
                                         # will become part of the current word
-            s = s[:end] + s[end+1:]
-            pos = end+1
+            s = s[:end] + s[end + 1:]
+            pos = end + 1
 
         else:
             if s[end] == "'":           # slurp singly-quoted string
@@ -253,7 +267,7 @@
                 raise ValueError("bad string (mismatched %s quotes?)" % s[end])
 
             (beg, end) = m.span()
-            s = s[:beg] + s[beg+1:end-1] + s[end:]
+            s = s[:beg] + s[beg + 1:end - 1] + s[end:]
             pos = m.end() - 2
 
         if pos >= len(s):
@@ -296,7 +310,7 @@
     elif val in ('n', 'no', 'f', 'false', 'off', '0'):
         return 0
     else:
-        raise ValueError, "invalid truth value %r" % (val,)
+        raise ValueError("invalid truth value %r" % (val,))
 
 
 def byte_compile(py_files, optimize=0, force=0, prefix=None, base_dir=None,
@@ -450,7 +464,9 @@
     return sep.join(lines)
 
 _RE_VERSION = re.compile('(\d+\.\d+(\.\d+)*)')
-_MAC_OS_X_LD_VERSION = re.compile('^@\(#\)PROGRAM:ld  PROJECT:ld64-((\d+)(\.\d+)*)')
+_MAC_OS_X_LD_VERSION = re.compile('^@\(#\)PROGRAM:ld  '
+                                  'PROJECT:ld64-((\d+)(\.\d+)*)')
+
 
 def _find_ld_version():
     """Finds the ld version. The version scheme differs under Mac OSX."""
@@ -459,6 +475,7 @@
     else:
         return _find_exe_version('ld -v')
 
+
 def _find_exe_version(cmd, pattern=_RE_VERSION):
     """Find the version of an executable by running `cmd` in the shell.
 
@@ -488,6 +505,7 @@
         return None
     return result.group(1)
 
+
 def get_compiler_versions():
     """Returns a tuple providing the versions of gcc, ld and dllwrap
 
@@ -499,6 +517,7 @@
     dllwrap = _find_exe_version('dllwrap --version')
     return gcc, ld, dllwrap
 
+
 def newer_group(sources, target, missing='error'):
     """Return true if 'target' is out-of-date with respect to any file
     listed in 'sources'.
@@ -538,6 +557,7 @@
 
     return False
 
+
 def write_file(filename, contents):
     """Create a file with the specified name and write 'contents' (a
     sequence of strings without line terminators) to it.
@@ -549,12 +569,14 @@
     finally:
         f.close()
 
+
 def _is_package(path):
     """Returns True if path is a package (a dir with an __init__ file."""
     if not os.path.isdir(path):
         return False
     return os.path.isfile(os.path.join(path, '__init__.py'))
 
+
 def _under(path, root):
     path = path.split(os.sep)
     root = root.split(os.sep)
@@ -565,12 +587,14 @@
             return False
     return True
 
+
 def _package_name(root_path, path):
     """Returns a dotted package name, given a subpath."""
     if not _under(path, root_path):
         raise ValueError('"%s" is not a subpath of "%s"' % (path, root_path))
     return path[len(root_path) + 1:].replace(os.sep, '.')
 
+
 def find_packages(paths=('.',), exclude=()):
     """Return a list all Python packages found recursively within
     directories 'paths'
@@ -585,6 +609,7 @@
     """
     packages = []
     discarded = []
+
     def _discarded(path):
         for discard in discarded:
             if _under(path, discard):
@@ -619,12 +644,12 @@
 
 # utility functions for 2to3 support
 
-def run_2to3(files, doctests_only=False, fixer_names=None, options=None, 
+def run_2to3(files, doctests_only=False, fixer_names=None, options=None,
                                                             explicit=None):
     """ Wrapper function around the refactor() class which
-	performs the conversions on a list of python files.
-	Invoke 2to3 on a list of Python files. The files should all come 
-	from the build area, as the modification is done in-place."""
+    performs the conversions on a list of python files.
+    Invoke 2to3 on a list of Python files. The files should all come
+    from the build area, as the modification is done in-place."""
 
     if not files:
         return
@@ -642,6 +667,7 @@
     else:
         r.refactor(files, write=True)
 
+
 class Mixin2to3:
     """ Wrapper class for commands that run 2to3.
     To configure 2to3, setup scripts may either change
@@ -660,4 +686,5 @@
 
     def run_2to3(self, files, doctests_only=False):
         """ Issues a call to util.run_2to3. """
-        return run_2to3(files, doctests_only, self.fixer_names, self.options, self.explicit)
+        return run_2to3(files, doctests_only, self.fixer_names,
+                        self.options, self.explicit)

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:38 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:38 +0200
Subject: [Python-checkins] distutils2: two typos fixed
Message-ID: <E1OVLoU-0000fv-IG@dinsdale.python.org>

tarek.ziade pushed 2f6b9df54955 to distutils2:

http://hg.python.org/distutils2/rev/2f6b9df54955
changeset:   257:2f6b9df54955
parent:      82:342424580a5f
user:        Konrad Delong <konryd at gmail.com>
date:        Fri Mar 19 09:38:42 2010 -0400
summary:     two typos fixed
files:       docs/design/pep-0376.txt, docs/design/wiki.rst

diff --git a/docs/design/pep-0376.txt b/docs/design/pep-0376.txt
--- a/docs/design/pep-0376.txt
+++ b/docs/design/pep-0376.txt
@@ -633,7 +633,7 @@
 
 Distributions installed using existing, pre-standardization formats do not have
 the necessary metadata available for the new API, and thus will be
-ignored. Third-party tools may of course to continue to support previous
+ignored. Third-party tools may of course continue to support previous
 formats in addition to the new format, in order to ease the transition.
 
 
diff --git a/docs/design/wiki.rst b/docs/design/wiki.rst
--- a/docs/design/wiki.rst
+++ b/docs/design/wiki.rst
@@ -282,7 +282,7 @@
   mailman/etc/*               = {config}                # 8
   mailman/foo/**/bar/*.cfg    = {config}/baz            # 9
   mailman/foo/**/*.cfg        = {config}/hmm            # 9, 10
-  some-new-semantic.txt       = {funky-crazy-category}  # 11
+  some-new-semantic.sns       = {funky-crazy-category}  # 11
 
 The glob definitions are relative paths that match files from the top
 of the source tree (the location of ``setup.cfg``). Forward slashes

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:38 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:38 +0200
Subject: [Python-checkins] distutils2: docs updated
Message-ID: <E1OVLoU-0000gI-OR@dinsdale.python.org>

tarek.ziade pushed dc5fbc82c929 to distutils2:

http://hg.python.org/distutils2/rev/dc5fbc82c929
changeset:   261:dc5fbc82c929
user:        Konrad Delong <konryd at gmail.com>
date:        Sun May 09 14:10:36 2010 -0400
summary:     docs updated
files:       docs/source/index.rst, docs/source/new_commands.rst

diff --git a/docs/source/index.rst b/docs/source/index.rst
--- a/docs/source/index.rst
+++ b/docs/source/index.rst
@@ -12,6 +12,7 @@
    :maxdepth: 2
 
    metadata
+   new_commands
 
 Indices and tables
 ==================
diff --git a/docs/source/new_commands.rst b/docs/source/new_commands.rst
new file mode 100644
--- /dev/null
+++ b/docs/source/new_commands.rst
@@ -0,0 +1,64 @@
+========
+Commands
+========
+
+Distutils2 provides a set of commands that are not present in distutils itself.
+You might recognize some of them from other projects, like Distribute or
+Setuptools.
+
+``upload_docs`` - Upload package documentation to PyPI
+======================================================
+
+PyPI now supports uploading project documentation to the dedicated URL
+http://packages.python.org/<project>/.
+
+The ``upload_docs`` command will create the necessary zip file out of a
+documentation directory and will post to the repository.
+
+Note that to upload the documentation of a project, the corresponding version
+must already be registered with PyPI, using the distutils ``register``
+command -- just like the ``upload`` command.
+
+Assuming there is an ``Example`` project with documentation in the
+subdirectory ``docs``, e.g.::
+
+  Example/
+  |-- example.py
+  |-- setup.cfg
+  |-- setup.py
+  |-- docs
+  |   |-- build
+  |   |   `-- html
+  |   |   |   |-- index.html
+  |   |   |   `-- tips_tricks.html
+  |   |-- conf.py
+  |   |-- index.txt
+  |   `-- tips_tricks.txt
+
+You can simply pass the documentation directory path to the ``upload_docs``
+command::
+
+    python setup.py upload_docs --upload-dir=docs/build/html
+
+As with any other ``setuptools`` based command, you can define useful
+defaults in the ``setup.cfg`` of your Python project, e.g.:
+
+.. code-block:: ini
+
+    [upload_docs]
+    upload-dir = docs/build/html
+
+The ``upload_docs`` command has the following options:
+
+``--upload-dir``
+    The directory to be uploaded to the repository.
+
+``--show-response``
+    Display the full response text from server; this is useful for debugging
+    PyPI problems.
+
+``--repository=URL, -r URL``
+    The URL of the repository to upload to.  Defaults to
+    http://pypi.python.org/pypi (i.e., the main PyPI installation).
+
+

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:38 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:38 +0200
Subject: [Python-checkins] distutils2: merged upstream
Message-ID: <E1OVLoU-0000g6-M5@dinsdale.python.org>

tarek.ziade pushed 0b00b7e6e0b8 to distutils2:

http://hg.python.org/distutils2/rev/0b00b7e6e0b8
changeset:   258:0b00b7e6e0b8
parent:      257:2f6b9df54955
parent:      122:69952f1f6af5
user:        Konrad Delong <konryd at gmail.com>
date:        Fri Apr 30 13:47:23 2010 -0400
summary:     merged upstream
files:       docs/design/wiki.rst

diff --git a/docs/design/wiki.rst b/docs/design/wiki.rst
--- a/docs/design/wiki.rst
+++ b/docs/design/wiki.rst
@@ -332,6 +332,7 @@
 
 What happens?
 ~~~~~~~~~~~~~
+
 As an example, ``mailman/database/schemas/blah.schema``:
 
 1. The file ``mailman/database/schemas/blah.schema`` in the source
@@ -397,8 +398,17 @@
   be nice to not need an equivalent of ``setup.py develop``. Can we just walk up
   the folder hierarchy from the module until we find a setup.cfg? A setup.cfg is
   necessary if you use distutils2, is it not?
+
+  -> information founded in setup.cfg will be put in the *FILES* file upon
+  installation in the egg-info directory. 
+  IOW in the unbuit-egg case, we would need to create that dir, then use 
+  pkgutil APIs.
+
 * If sysconfig.cfg lands in Python 2.7, what happens when we run distutils2 in
   2.4?
+
+  -> A backport of sysconfig.cfg is provided within the distutils2 distribution.
+  
 * Our new glob-based [resources] section is much more compact (and consistent
   with other systems, like bash) than the explicit MANIFEST.in directives, but
   they don't offer some of the old features. Is it okay to lose exclude,
@@ -406,6 +416,8 @@
   cover their behavior? I think we could probably use a [resource:exclude]
   section with additional exclude globs in it.
 
+  -> let's list the use cases we don't cover, and see
+
 API
 ===
 
diff --git a/src/distutils2/_backport/pkgutil.py b/src/distutils2/_backport/pkgutil.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/pkgutil.py
@@ -0,0 +1,791 @@
+"""Utilities to support packages."""
+
+# NOTE: This module must remain compatible with Python 2.3, as it is shared
+# by setuptools for distribution with Python 2.3 and up.
+
+import os
+import sys
+import imp
+import os.path
+from csv import reader as csv_reader
+from types import ModuleType
+from distutils2.errors import DistutilsError
+from distutils2.metadata import DistributionMetadata
+from distutils2.version import suggest_normalized_version
+
+__all__ = [
+    'get_importer', 'iter_importers', 'get_loader', 'find_loader',
+    'walk_packages', 'iter_modules',
+    'ImpImporter', 'ImpLoader', 'read_code', 'extend_path',
+    'Distribution', 'distinfo_dirname', 'get_distributions',
+    'get_distribution', 'get_file_users', 
+]
+
+def read_code(stream):
+    # This helper is needed in order for the PEP 302 emulation to
+    # correctly handle compiled files
+    import marshal
+
+    magic = stream.read(4)
+    if magic != imp.get_magic():
+        return None
+
+    stream.read(4) # Skip timestamp
+    return marshal.load(stream)
+
+
+def simplegeneric(func):
+    """Make a trivial single-dispatch generic function"""
+    registry = {}
+    def wrapper(*args, **kw):
+        ob = args[0]
+        try:
+            cls = ob.__class__
+        except AttributeError:
+            cls = type(ob)
+        try:
+            mro = cls.__mro__
+        except AttributeError:
+            try:
+                class cls(cls, object):
+                    pass
+                mro = cls.__mro__[1:]
+            except TypeError:
+                mro = object,   # must be an ExtensionClass or some such  :(
+        for t in mro:
+            if t in registry:
+                return registry[t](*args, **kw)
+        else:
+            return func(*args, **kw)
+    try:
+        wrapper.__name__ = func.__name__
+    except (TypeError, AttributeError):
+        pass    # Python 2.3 doesn't allow functions to be renamed
+
+    def register(typ, func=None):
+        if func is None:
+            return lambda f: register(typ, f)
+        registry[typ] = func
+        return func
+
+    wrapper.__dict__ = func.__dict__
+    wrapper.__doc__ = func.__doc__
+    wrapper.register = register
+    return wrapper
+
+
+def walk_packages(path=None, prefix='', onerror=None):
+    """Yields (module_loader, name, ispkg) for all modules recursively
+    on path, or, if path is None, all accessible modules.
+
+    'path' should be either None or a list of paths to look for
+    modules in.
+
+    'prefix' is a string to output on the front of every module name
+    on output.
+
+    Note that this function must import all *packages* (NOT all
+    modules!) on the given path, in order to access the __path__
+    attribute to find submodules.
+
+    'onerror' is a function which gets called with one argument (the
+    name of the package which was being imported) if any exception
+    occurs while trying to import a package.  If no onerror function is
+    supplied, ImportErrors are caught and ignored, while all other
+    exceptions are propagated, terminating the search.
+
+    Examples:
+
+    # list all modules python can access
+    walk_packages()
+
+    # list all submodules of ctypes
+    walk_packages(ctypes.__path__, ctypes.__name__+'.')
+    """
+
+    def seen(p, m={}):
+        if p in m:
+            return True
+        m[p] = True
+
+    for importer, name, ispkg in iter_modules(path, prefix):
+        yield importer, name, ispkg
+
+        if ispkg:
+            try:
+                __import__(name)
+            except ImportError:
+                if onerror is not None:
+                    onerror(name)
+            except Exception:
+                if onerror is not None:
+                    onerror(name)
+                else:
+                    raise
+            else:
+                path = getattr(sys.modules[name], '__path__', None) or []
+
+                # don't traverse path items we've seen before
+                path = [p for p in path if not seen(p)]
+
+                for item in walk_packages(path, name+'.', onerror):
+                    yield item
+
+
+def iter_modules(path=None, prefix=''):
+    """Yields (module_loader, name, ispkg) for all submodules on path,
+    or, if path is None, all top-level modules on sys.path.
+
+    'path' should be either None or a list of paths to look for
+    modules in.
+
+    'prefix' is a string to output on the front of every module name
+    on output.
+    """
+
+    if path is None:
+        importers = iter_importers()
+    else:
+        importers = map(get_importer, path)
+
+    yielded = {}
+    for i in importers:
+        for name, ispkg in iter_importer_modules(i, prefix):
+            if name not in yielded:
+                yielded[name] = 1
+                yield i, name, ispkg
+
+
+#@simplegeneric
+def iter_importer_modules(importer, prefix=''):
+    if not hasattr(importer, 'iter_modules'):
+        return []
+    return importer.iter_modules(prefix)
+
+iter_importer_modules = simplegeneric(iter_importer_modules)
+
+
+class ImpImporter:
+    """PEP 302 Importer that wraps Python's "classic" import algorithm
+
+    ImpImporter(dirname) produces a PEP 302 importer that searches that
+    directory.  ImpImporter(None) produces a PEP 302 importer that searches
+    the current sys.path, plus any modules that are frozen or built-in.
+
+    Note that ImpImporter does not currently support being used by placement
+    on sys.meta_path.
+    """
+
+    def __init__(self, path=None):
+        self.path = path
+
+    def find_module(self, fullname, path=None):
+        # Note: we ignore 'path' argument since it is only used via meta_path
+        subname = fullname.split(".")[-1]
+        if subname != fullname and self.path is None:
+            return None
+        if self.path is None:
+            path = None
+        else:
+            path = [os.path.realpath(self.path)]
+        try:
+            file, filename, etc = imp.find_module(subname, path)
+        except ImportError:
+            return None
+        return ImpLoader(fullname, file, filename, etc)
+
+    def iter_modules(self, prefix=''):
+        if self.path is None or not os.path.isdir(self.path):
+            return
+
+        yielded = {}
+        import inspect
+
+        filenames = os.listdir(self.path)
+        filenames.sort()  # handle packages before same-named modules
+
+        for fn in filenames:
+            modname = inspect.getmodulename(fn)
+            if modname=='__init__' or modname in yielded:
+                continue
+
+            path = os.path.join(self.path, fn)
+            ispkg = False
+
+            if not modname and os.path.isdir(path) and '.' not in fn:
+                modname = fn
+                for fn in os.listdir(path):
+                    subname = inspect.getmodulename(fn)
+                    if subname=='__init__':
+                        ispkg = True
+                        break
+                else:
+                    continue    # not a package
+
+            if modname and '.' not in modname:
+                yielded[modname] = 1
+                yield prefix + modname, ispkg
+
+
+class ImpLoader:
+    """PEP 302 Loader that wraps Python's "classic" import algorithm
+    """
+    code = source = None
+
+    def __init__(self, fullname, file, filename, etc):
+        self.file = file
+        self.filename = filename
+        self.fullname = fullname
+        self.etc = etc
+
+    def load_module(self, fullname):
+        self._reopen()
+        try:
+            mod = imp.load_module(fullname, self.file, self.filename, self.etc)
+        finally:
+            if self.file:
+                self.file.close()
+        # Note: we don't set __loader__ because we want the module to look
+        # normal; i.e. this is just a wrapper for standard import machinery
+        return mod
+
+    def get_data(self, pathname):
+        return open(pathname, "rb").read()
+
+    def _reopen(self):
+        if self.file and self.file.closed:
+            mod_type = self.etc[2]
+            if mod_type==imp.PY_SOURCE:
+                self.file = open(self.filename, 'rU')
+            elif mod_type in (imp.PY_COMPILED, imp.C_EXTENSION):
+                self.file = open(self.filename, 'rb')
+
+    def _fix_name(self, fullname):
+        if fullname is None:
+            fullname = self.fullname
+        elif fullname != self.fullname:
+            raise ImportError("Loader for module %s cannot handle "
+                              "module %s" % (self.fullname, fullname))
+        return fullname
+
+    def is_package(self, fullname):
+        fullname = self._fix_name(fullname)
+        return self.etc[2]==imp.PKG_DIRECTORY
+
+    def get_code(self, fullname=None):
+        fullname = self._fix_name(fullname)
+        if self.code is None:
+            mod_type = self.etc[2]
+            if mod_type==imp.PY_SOURCE:
+                source = self.get_source(fullname)
+                self.code = compile(source, self.filename, 'exec')
+            elif mod_type==imp.PY_COMPILED:
+                self._reopen()
+                try:
+                    self.code = read_code(self.file)
+                finally:
+                    self.file.close()
+            elif mod_type==imp.PKG_DIRECTORY:
+                self.code = self._get_delegate().get_code()
+        return self.code
+
+    def get_source(self, fullname=None):
+        fullname = self._fix_name(fullname)
+        if self.source is None:
+            mod_type = self.etc[2]
+            if mod_type==imp.PY_SOURCE:
+                self._reopen()
+                try:
+                    self.source = self.file.read()
+                finally:
+                    self.file.close()
+            elif mod_type==imp.PY_COMPILED:
+                if os.path.exists(self.filename[:-1]):
+                    f = open(self.filename[:-1], 'rU')
+                    self.source = f.read()
+                    f.close()
+            elif mod_type==imp.PKG_DIRECTORY:
+                self.source = self._get_delegate().get_source()
+        return self.source
+
+
+    def _get_delegate(self):
+        return ImpImporter(self.filename).find_module('__init__')
+
+    def get_filename(self, fullname=None):
+        fullname = self._fix_name(fullname)
+        mod_type = self.etc[2]
+        if self.etc[2]==imp.PKG_DIRECTORY:
+            return self._get_delegate().get_filename()
+        elif self.etc[2] in (imp.PY_SOURCE, imp.PY_COMPILED, imp.C_EXTENSION):
+            return self.filename
+        return None
+
+
+try:
+    import zipimport
+    from zipimport import zipimporter
+
+    def iter_zipimport_modules(importer, prefix=''):
+        dirlist = zipimport._zip_directory_cache[importer.archive].keys()
+        dirlist.sort()
+        _prefix = importer.prefix
+        plen = len(_prefix)
+        yielded = {}
+        import inspect
+        for fn in dirlist:
+            if not fn.startswith(_prefix):
+                continue
+
+            fn = fn[plen:].split(os.sep)
+
+            if len(fn)==2 and fn[1].startswith('__init__.py'):
+                if fn[0] not in yielded:
+                    yielded[fn[0]] = 1
+                    yield fn[0], True
+
+            if len(fn)!=1:
+                continue
+
+            modname = inspect.getmodulename(fn[0])
+            if modname=='__init__':
+                continue
+
+            if modname and '.' not in modname and modname not in yielded:
+                yielded[modname] = 1
+                yield prefix + modname, False
+
+    iter_importer_modules.register(zipimporter, iter_zipimport_modules)
+
+except ImportError:
+    pass
+
+
+def get_importer(path_item):
+    """Retrieve a PEP 302 importer for the given path item
+
+    The returned importer is cached in sys.path_importer_cache
+    if it was newly created by a path hook.
+
+    If there is no importer, a wrapper around the basic import
+    machinery is returned. This wrapper is never inserted into
+    the importer cache (None is inserted instead).
+
+    The cache (or part of it) can be cleared manually if a
+    rescan of sys.path_hooks is necessary.
+    """
+    try:
+        importer = sys.path_importer_cache[path_item]
+    except KeyError:
+        for path_hook in sys.path_hooks:
+            try:
+                importer = path_hook(path_item)
+                break
+            except ImportError:
+                pass
+        else:
+            importer = None
+        sys.path_importer_cache.setdefault(path_item, importer)
+
+    if importer is None:
+        try:
+            importer = ImpImporter(path_item)
+        except ImportError:
+            importer = None
+    return importer
+
+
+def iter_importers(fullname=""):
+    """Yield PEP 302 importers for the given module name
+
+    If fullname contains a '.', the importers will be for the package
+    containing fullname, otherwise they will be importers for sys.meta_path,
+    sys.path, and Python's "classic" import machinery, in that order.  If
+    the named module is in a package, that package is imported as a side
+    effect of invoking this function.
+
+    Non PEP 302 mechanisms (e.g. the Windows registry) used by the
+    standard import machinery to find files in alternative locations
+    are partially supported, but are searched AFTER sys.path. Normally,
+    these locations are searched BEFORE sys.path, preventing sys.path
+    entries from shadowing them.
+
+    For this to cause a visible difference in behaviour, there must
+    be a module or package name that is accessible via both sys.path
+    and one of the non PEP 302 file system mechanisms. In this case,
+    the emulation will find the former version, while the builtin
+    import mechanism will find the latter.
+
+    Items of the following types can be affected by this discrepancy:
+        imp.C_EXTENSION, imp.PY_SOURCE, imp.PY_COMPILED, imp.PKG_DIRECTORY
+    """
+    if fullname.startswith('.'):
+        raise ImportError("Relative module names not supported")
+    if '.' in fullname:
+        # Get the containing package's __path__
+        pkg = '.'.join(fullname.split('.')[:-1])
+        if pkg not in sys.modules:
+            __import__(pkg)
+        path = getattr(sys.modules[pkg], '__path__', None) or []
+    else:
+        for importer in sys.meta_path:
+            yield importer
+        path = sys.path
+    for item in path:
+        yield get_importer(item)
+    if '.' not in fullname:
+        yield ImpImporter()
+
+def get_loader(module_or_name):
+    """Get a PEP 302 "loader" object for module_or_name
+
+    If the module or package is accessible via the normal import
+    mechanism, a wrapper around the relevant part of that machinery
+    is returned.  Returns None if the module cannot be found or imported.
+    If the named module is not already imported, its containing package
+    (if any) is imported, in order to establish the package __path__.
+
+    This function uses iter_importers(), and is thus subject to the same
+    limitations regarding platform-specific special import locations such
+    as the Windows registry.
+    """
+    if module_or_name in sys.modules:
+        module_or_name = sys.modules[module_or_name]
+    if isinstance(module_or_name, ModuleType):
+        module = module_or_name
+        loader = getattr(module, '__loader__', None)
+        if loader is not None:
+            return loader
+        fullname = module.__name__
+    else:
+        fullname = module_or_name
+    return find_loader(fullname)
+
+def find_loader(fullname):
+    """Find a PEP 302 "loader" object for fullname
+
+    If fullname contains dots, path must be the containing package's __path__.
+    Returns None if the module cannot be found or imported. This function uses
+    iter_importers(), and is thus subject to the same limitations regarding
+    platform-specific special import locations such as the Windows registry.
+    """
+    for importer in iter_importers(fullname):
+        loader = importer.find_module(fullname)
+        if loader is not None:
+            return loader
+
+    return None
+
+
+def extend_path(path, name):
+    """Extend a package's path.
+
+    Intended use is to place the following code in a package's __init__.py:
+
+        from pkgutil import extend_path
+        __path__ = extend_path(__path__, __name__)
+
+    This will add to the package's __path__ all subdirectories of
+    directories on sys.path named after the package.  This is useful
+    if one wants to distribute different parts of a single logical
+    package as multiple directories.
+
+    It also looks for *.pkg files beginning where * matches the name
+    argument.  This feature is similar to *.pth files (see site.py),
+    except that it doesn't special-case lines starting with 'import'.
+    A *.pkg file is trusted at face value: apart from checking for
+    duplicates, all entries found in a *.pkg file are added to the
+    path, regardless of whether they are exist the filesystem.  (This
+    is a feature.)
+
+    If the input path is not a list (as is the case for frozen
+    packages) it is returned unchanged.  The input path is not
+    modified; an extended copy is returned.  Items are only appended
+    to the copy at the end.
+
+    It is assumed that sys.path is a sequence.  Items of sys.path that
+    are not (unicode or 8-bit) strings referring to existing
+    directories are ignored.  Unicode items of sys.path that cause
+    errors when used as filenames may cause this function to raise an
+    exception (in line with os.path.isdir() behavior).
+    """
+
+    if not isinstance(path, list):
+        # This could happen e.g. when this is called from inside a
+        # frozen package.  Return the path unchanged in that case.
+        return path
+
+    pname = os.path.join(*name.split('.')) # Reconstitute as relative path
+    # Just in case os.extsep != '.'
+    sname = os.extsep.join(name.split('.'))
+    sname_pkg = sname + os.extsep + "pkg"
+    init_py = "__init__" + os.extsep + "py"
+
+    path = path[:] # Start with a copy of the existing path
+
+    for dir in sys.path:
+        if not isinstance(dir, basestring) or not os.path.isdir(dir):
+            continue
+        subdir = os.path.join(dir, pname)
+        # XXX This may still add duplicate entries to path on
+        # case-insensitive filesystems
+        initfile = os.path.join(subdir, init_py)
+        if subdir not in path and os.path.isfile(initfile):
+            path.append(subdir)
+        # XXX Is this the right thing for subpackages like zope.app?
+        # It looks for a file named "zope.app.pkg"
+        pkgfile = os.path.join(dir, sname_pkg)
+        if os.path.isfile(pkgfile):
+            try:
+                f = open(pkgfile)
+            except IOError, msg:
+                sys.stderr.write("Can't open %s: %s\n" %
+                                 (pkgfile, msg))
+            else:
+                for line in f:
+                    line = line.rstrip('\n')
+                    if not line or line.startswith('#'):
+                        continue
+                    path.append(line) # Don't check for existence!
+                f.close()
+
+    return path
+
+def get_data(package, resource):
+    """Get a resource from a package.
+
+    This is a wrapper round the PEP 302 loader get_data API. The package
+    argument should be the name of a package, in standard module format
+    (foo.bar). The resource argument should be in the form of a relative
+    filename, using '/' as the path separator. The parent directory name '..'
+    is not allowed, and nor is a rooted name (starting with a '/').
+
+    The function returns a binary string, which is the contents of the
+    specified resource.
+
+    For packages located in the filesystem, which have already been imported,
+    this is the rough equivalent of
+
+        d = os.path.dirname(sys.modules[package].__file__)
+        data = open(os.path.join(d, resource), 'rb').read()
+
+    If the package cannot be located or loaded, or it uses a PEP 302 loader
+    which does not support get_data(), then None is returned.
+    """
+
+    loader = get_loader(package)
+    if loader is None or not hasattr(loader, 'get_data'):
+        return None
+    mod = sys.modules.get(package) or loader.load_module(package)
+    if mod is None or not hasattr(mod, '__file__'):
+        return None
+
+    # Modify the resource name to be compatible with the loader.get_data
+    # signature - an os.path format "filename" starting with the dirname of
+    # the package's __file__
+    parts = resource.split('/')
+    parts.insert(0, os.path.dirname(mod.__file__))
+    resource_name = os.path.join(*parts)
+    return loader.get_data(resource_name)
+
+##########################
+# PEP 376 Implementation #
+##########################
+
+DIST_FILES = ('INSTALLER', 'METADATA', 'RECORD', 'REQUESTED',)
+
+class Distribution(object):
+    """Created with the *path* of the ``.dist-info`` directory provided to the
+    constructor. It reads the metadata contained in METADATA when it is
+    instantiated."""
+
+    # Attribute documenting for Sphinx style documentation, see for more info:
+    #   http://sphinx.pocoo.org/ext/autodoc.html#dir-autoattribute
+    name = ''
+    """The name of the distribution."""
+    metadata = None
+    """A :class:`distutils2.metadata.DistributionMetadata` instance loaded with 
+    the distribution's METADATA file."""
+    requested = False
+    """A boolean that indicates whether the REQUESTED metadata file is present
+    (in other words, whether the package was installed by user request)."""
+
+    def __init__(self, path):
+        self.path = path
+        metadata_path = os.path.join(path, 'METADATA')
+        self.metadata = DistributionMetadata(path=metadata_path)
+        self.name = self.metadata['name']
+
+    def _get_records(self, local=False):
+        RECORD = os.path.join(self.path, 'RECORD')
+        record_reader = csv_reader(open(RECORD, 'rb'), delimiter=',')
+        for row in record_reader:
+            path, md5, size = row[:] + [ None for i in xrange(len(row), 3) ]
+            if local:
+                path = path.replace('/', os.sep)
+                path = os.path.join(sys.prefix, path)
+            yield path, md5, size
+
+    def get_installed_files(self, local=False):
+        """
+        Iterates over the RECORD entries and returns a tuple (path, md5, size)
+        for each line. If *local* is ``True``, the returned path is transformed
+        into a local absolute path. Otherwise the raw value from RECORD is
+        returned.
+
+        A local absolute path is an absolute path in which occurrences of
+        ``'/'`` have been replaced by the system separator given by ``os.sep``.
+
+        :parameter local: flag to say if the path should be returned a local
+                          absolute path
+        :type local: boolean
+        :returns: iterator of (path, md5, size)
+        """
+        return self._get_records(local)
+
+
+    def uses(self, path):
+        """
+        Returns ``True`` if path is listed in RECORD. *path* can be a local 
+        absolute path or a relative ``'/'``-separated path.
+
+        :rtype: boolean
+        """
+        for p, md5, size in self._get_records():
+            local_absolute = os.path.join(sys.prefix, p)
+            if path == p or path == local_absolute:
+                return True
+        return False
+
+    def get_distinfo_file(self, path, binary=False):
+        """
+        Returns a file located under the ``.dist-info`` directory. Returns a
+        ``file`` instance for the file pointed by *path*.
+
+        :parameter path: a ``'/'``-separated path relative to the ``.dist-info``
+                         directory or an absolute path; If *path* is an absolute 
+                         path and doesn't start with the ``.dist-info``
+                         directory path, a :class:`DistutilsError` is raised
+        :type path: string
+        :parameter binary: If *binary* is ``True``, opens the file in read-only
+                           binary mode (rb), otherwise opens it in read-only
+                           mode (r).
+        :rtype: file object
+        """
+        open_flags = 'r'
+        if binary:
+            open_flags += 'b'
+
+        # Check if it is an absolute path
+        if path.find(os.sep) >= 0:
+            # it's an absolute path?
+            distinfo_dirname, path = path.split(os.sep)[-2:]
+            if distinfo_dirname != self.path.split(os.sep)[-1]:
+                raise DistutilsError("Requested dist-info file does not "
+                    "belong to the %s distribution. '%s' was requested." \
+                    % (self.name, os.sep.join([distinfo_dirname, path])))
+
+        # The file must be relative
+        if path not in DIST_FILES:
+            raise DistutilsError("Requested an invalid dist-info file: "
+                "%s" % path)
+
+        # Convert the relative path back to absolute
+        path = os.path.join(self.path, path)
+        return open(path, open_flags)
+
+    def get_distinfo_files(self, local=False):
+        """
+        Iterates over the RECORD entries and returns paths for each line if the 
+        path is pointing to a file located in the ``.dist-info`` directory or 
+        one of its subdirectories.
+
+        :parameter local: If *local* is ``True``, each returned path is
+                          transformed into a local absolute path. Otherwise the
+                          raw value from RECORD is returned.
+        :type local: boolean
+        :returns: iterator of paths
+        """
+        for path, md5, size in self._get_records(local):
+            yield path
+
+
+def _normalize_dist_name(name):
+    """Returns a normalized name from the given *name*.
+    :rtype: string"""
+    return name.replace('-', '_')
+
+def distinfo_dirname(name, version):
+    """
+    The *name* and *version* parameters are converted into their
+    filename-escaped form, i.e. any ``'-'`` characters are replaced with ``'_'``
+    other than the one in ``'dist-info'`` and the one separating the name from 
+    the version number.
+
+    :parameter name: is converted to a standard distribution name by replacing
+                     any runs of non- alphanumeric characters with a single
+                     ``'-'``.
+    :type name: string
+    :parameter version: is converted to a standard version string. Spaces become
+                        dots, and all other non-alphanumeric characters (except
+                        dots) become dashes, with runs of multiple dashes
+                        condensed to a single dash.
+    :type version: string
+    :returns: directory name
+    :rtype: string"""
+    file_extension = '.dist-info'
+    name = _normalize_dist_name(name)
+    normalized_version = suggest_normalized_version(version)
+    # Because this is a lookup procedure, something will be returned even if
+    #   it is a version that cannot be normalized
+    if normalized_version is None:
+        # Unable to achieve normality?
+        normalized_version = version
+    return '-'.join([name, normalized_version]) + file_extension
+
+def get_distributions():
+    """
+    Provides an iterator that looks for ``.dist-info`` directories in
+    ``sys.path`` and returns :class:`Distribution` instances for each one of
+    them.
+
+    :rtype: iterator of :class:`Distribution` instances"""
+    for path in sys.path:
+        realpath = os.path.realpath(path)
+        if not os.path.isdir(realpath):
+            continue
+        for dir in os.listdir(realpath):
+            if dir.endswith('.dist-info'):
+                dist = Distribution(os.path.join(realpath, dir))
+                yield dist
+
+def get_distribution(name):
+    """
+    Scans all elements in ``sys.path`` and looks for all directories ending with
+    ``.dist-info``. Returns a :class:`Distribution` corresponding to the 
+    ``.dist-info`` directory that contains the METADATA that matches *name* for
+    the *name* metadata.
+
+    This function only returns the first result founded, as no more than one
+    value is expected. If the directory is not found, ``None`` is returned.
+
+    :rtype: :class:`Distribution` or None"""
+    found = None
+    for dist in get_distributions():
+        if dist.name == name:
+            found = dist
+            break
+    return found
+
+def get_file_users(path):
+    """
+    Iterates over all distributions to find out which distributions uses
+    *path*.
+
+    :parameter path: can be a local absolute path or a relative
+                     ``'/'``-separated path.
+    :type path: string
+    :rtype: iterator of :class:`Distribution` instances"""
+    for dist in get_distributions():
+        if dist.uses(path):
+            yield dist
diff --git a/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info/INSTALLER b/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info/INSTALLER
new file mode 100644
diff --git a/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info/METADATA b/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info/METADATA
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info/METADATA
@@ -0,0 +1,6 @@
+Metadata-Version: 1.2
+Name: choxie
+Version: 2.0.0.9
+Summary: Chocolate with a kick!
+Requires-Dist: towel-stuff (0.1)
+Provides-Dist: truffles (1.0)
diff --git a/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info/RECORD b/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info/RECORD
new file mode 100644
diff --git a/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info/REQUESTED b/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info/REQUESTED
new file mode 100644
diff --git a/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9/choxie/__init__.py b/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9/choxie/__init__.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9/choxie/__init__.py
@@ -0,0 +1,2 @@
+# -*- coding: utf-8 -*-
+
diff --git a/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9/choxie/chocolate.py b/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9/choxie/chocolate.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9/choxie/chocolate.py
@@ -0,0 +1,10 @@
+# -*- coding: utf-8 -*-
+from towel_stuff import Towel
+
+class Chocolate(object):
+    """A piece of chocolate."""
+
+    def wrap_with_towel(self):
+        towel = Towel()
+        towel.wrap(self)
+        return towel
diff --git a/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9/truffles.py b/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9/truffles.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9/truffles.py
@@ -0,0 +1,5 @@
+# -*- coding: utf-8 -*-
+from choxie.chocolate import Chocolate
+
+class Truffle(Chocolate):
+    """A truffle."""
diff --git a/src/distutils2/_backport/tests/fake_dists/grammar-1.0a4.dist-info/INSTALLER b/src/distutils2/_backport/tests/fake_dists/grammar-1.0a4.dist-info/INSTALLER
new file mode 100644
diff --git a/src/distutils2/_backport/tests/fake_dists/grammar-1.0a4.dist-info/METADATA b/src/distutils2/_backport/tests/fake_dists/grammar-1.0a4.dist-info/METADATA
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/tests/fake_dists/grammar-1.0a4.dist-info/METADATA
@@ -0,0 +1,3 @@
+Metadata-Version: 1.2
+Name: grammar
+Version: 1.0a4
diff --git a/src/distutils2/_backport/tests/fake_dists/grammar-1.0a4.dist-info/RECORD b/src/distutils2/_backport/tests/fake_dists/grammar-1.0a4.dist-info/RECORD
new file mode 100644
diff --git a/src/distutils2/_backport/tests/fake_dists/grammar-1.0a4.dist-info/REQUESTED b/src/distutils2/_backport/tests/fake_dists/grammar-1.0a4.dist-info/REQUESTED
new file mode 100644
diff --git a/src/distutils2/_backport/tests/fake_dists/grammar-1.0a4/grammar/__init__.py b/src/distutils2/_backport/tests/fake_dists/grammar-1.0a4/grammar/__init__.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/tests/fake_dists/grammar-1.0a4/grammar/__init__.py
@@ -0,0 +1,2 @@
+# -*- coding: utf-8 -*-
+
diff --git a/src/distutils2/_backport/tests/fake_dists/grammar-1.0a4/grammar/utils.py b/src/distutils2/_backport/tests/fake_dists/grammar-1.0a4/grammar/utils.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/tests/fake_dists/grammar-1.0a4/grammar/utils.py
@@ -0,0 +1,8 @@
+# -*- coding: utf-8 -*-
+from random import randint
+
+def is_valid_grammar(sentence):
+    if randint(0, 10) < 2:
+        return False
+    else:
+        return True
diff --git a/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1.dist-info/INSTALLER b/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1.dist-info/INSTALLER
new file mode 100644
diff --git a/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1.dist-info/METADATA b/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1.dist-info/METADATA
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1.dist-info/METADATA
@@ -0,0 +1,3 @@
+Metadata-Version: 1.2
+Name: towel-stuff
+Version: 0.1
diff --git a/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1.dist-info/RECORD b/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1.dist-info/RECORD
new file mode 100644
diff --git a/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1.dist-info/REQUESTED b/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1.dist-info/REQUESTED
new file mode 100644
diff --git a/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1/towel_stuff/__init__.py b/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1/towel_stuff/__init__.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1/towel_stuff/__init__.py
@@ -0,0 +1,18 @@
+# -*- coding: utf-8 -*-
+
+class Towel(object):
+    """A towel, that one should never be without."""
+
+    def __init__(self, color='tie-dye'):
+        self.color = color
+        self.wrapped_obj = None
+
+    def wrap(self, obj):
+        """Wrap an object up in our towel."""
+        self.wrapped_obj = obj
+
+    def unwrap(self):
+        """Unwrap whatever is in our towel and return whatever it is."""
+        obj = self.wrapped_obj
+        self.wrapped_obj = None
+        return obj
diff --git a/src/distutils2/_backport/tests/test_pkgutil.py b/src/distutils2/_backport/tests/test_pkgutil.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/tests/test_pkgutil.py
@@ -0,0 +1,275 @@
+# -*- coding: utf-8 -*-
+"""Tests for PEP 376 pkgutil functionality"""
+import unittest2
+import sys
+import os
+import csv
+try:
+    from hashlib import md5
+except ImportError:
+    from md5 import md5
+
+from test.test_support import run_unittest, TESTFN
+
+import distutils2._backport.pkgutil
+
+# TODO Add a test for getting a distribution that is provided by another
+#   distribution.
+
+# TODO Add a test for absolute pathed RECORD items (e.g. /etc/myapp/config.ini)
+
+class TestPkgUtilDistribution(unittest2.TestCase):
+    """Tests the pkgutil.Distribution class"""
+
+    def setUp(self):
+        super(TestPkgUtilDistribution, self).setUp()
+
+        self.fake_dists_path = os.path.abspath(
+            os.path.join(os.path.dirname(__file__), 'fake_dists'))
+        self.distinfo_dirs = [ os.path.join(self.fake_dists_path, dir)
+            for dir in os.listdir(self.fake_dists_path)
+            if dir.endswith('.dist-info')
+            ]
+
+        def get_hexdigest(file):
+            md5_hash = md5()
+            md5_hash.update(open(file).read())
+            return md5_hash.hexdigest()
+        def record_pieces(file):
+            path = os.path.relpath(file, sys.prefix)
+            digest = get_hexdigest(file)
+            size = os.path.getsize(file)
+            return [path, digest, size]
+
+        self.records = {}
+        for distinfo_dir in self.distinfo_dirs:
+            # Setup the RECORD file for this dist
+            record_file = os.path.join(distinfo_dir, 'RECORD')
+            record_writer = csv.writer(open(record_file, 'w'), delimiter=',',
+                quoting=csv.QUOTE_NONE)
+            dist_location = distinfo_dir.replace('.dist-info', '')
+
+            for path, dirs, files in os.walk(dist_location):
+                for f in files:
+                    record_writer.writerow(record_pieces(os.path.join(path, f)))
+            for file in ['INSTALLER', 'METADATA', 'REQUESTED']:
+                record_writer.writerow(record_pieces(
+                    os.path.join(distinfo_dir, file)))
+            record_writer.writerow([os.path.relpath(record_file, sys.prefix)])
+            del record_writer # causes the RECORD file to close
+            record_reader = csv.reader(open(record_file, 'rb'))
+            record_data = []
+            for row in record_reader:
+                path, md5_, size = row[:] + [ None for i in xrange(len(row), 3) ]
+                record_data.append([path, (md5_, size,)])
+            self.records[distinfo_dir] = dict(record_data)
+
+    def tearDown(self):
+        self.records = None
+        for distinfo_dir in self.distinfo_dirs:
+            record_file = os.path.join(distinfo_dir, 'RECORD')
+            open(record_file, 'w').close()
+        super(TestPkgUtilDistribution, self).tearDown()
+
+    def test_instantiation(self):
+        """Test the Distribution class's instantiation provides us with usable
+        attributes."""
+        # Import the Distribution class
+        from distutils2._backport.pkgutil import distinfo_dirname, Distribution
+
+        here = os.path.abspath(os.path.dirname(__file__))
+        name = 'choxie'
+        version = '2.0.0.9'
+        dist_path = os.path.join(here, 'fake_dists',
+            distinfo_dirname(name, version))
+        dist = Distribution(dist_path)
+
+        self.assertEqual(dist.name, name)
+        from distutils2.metadata import DistributionMetadata
+        self.assertTrue(isinstance(dist.metadata, DistributionMetadata))
+        self.assertEqual(dist.metadata['version'], version)
+        self.assertTrue(isinstance(dist.requested, type(bool())))
+
+    def test_installed_files(self):
+        """Test the iteration of installed files."""
+        # Test the distribution's installed files
+        from distutils2._backport.pkgutil import Distribution
+        for distinfo_dir in self.distinfo_dirs:
+            dist = Distribution(distinfo_dir)
+            for path, md5_, size in dist.get_installed_files():
+                record_data = self.records[dist.path]
+                self.assertTrue(path in record_data.keys())
+                self.assertEqual(md5_, record_data[path][0])
+                self.assertEqual(size, record_data[path][1])
+
+    def test_uses(self):
+        """Test to determine if a distribution uses a specified file."""
+        # Criteria to test against
+        distinfo_name = 'grammar-1.0a4'
+        distinfo_dir = os.path.join(self.fake_dists_path,
+            distinfo_name + '.dist-info')
+        true_path = [self.fake_dists_path, distinfo_name, 'grammar', 'utils.py']
+        true_path = os.path.relpath(os.path.join(*true_path), sys.prefix)
+        false_path = [self.fake_dists_path, 'towel_stuff-0.1', 'towel_stuff',
+            '__init__.py']
+        false_path = os.path.relpath(os.path.join(*false_path), sys.prefix)
+
+        # Test if the distribution uses the file in question
+        from distutils2._backport.pkgutil import Distribution
+        dist = Distribution(distinfo_dir)
+        self.assertTrue(dist.uses(true_path))
+        self.assertFalse(dist.uses(false_path))
+
+    def test_get_distinfo_file(self):
+        """Test the retrieval of dist-info file objects."""
+        from distutils2._backport.pkgutil import Distribution
+        distinfo_name = 'choxie-2.0.0.9'
+        other_distinfo_name = 'grammar-1.0a4'
+        distinfo_dir = os.path.join(self.fake_dists_path,
+            distinfo_name + '.dist-info')
+        dist = Distribution(distinfo_dir)
+        # Test for known good file matches
+        distinfo_files = [
+            # Relative paths
+            'INSTALLER', 'METADATA',
+            # Absolute paths
+            os.path.join(distinfo_dir, 'RECORD'),
+            os.path.join(distinfo_dir, 'REQUESTED'),
+            ]
+
+        for distfile in distinfo_files:
+            value = dist.get_distinfo_file(distfile)
+            self.assertTrue(isinstance(value, file))
+            # Is it the correct file?
+            self.assertEqual(value.name, os.path.join(distinfo_dir, distfile))
+
+        from distutils2.errors import DistutilsError
+        # Test an absolute path that is part of another distributions dist-info
+        other_distinfo_file = os.path.join(self.fake_dists_path,
+            other_distinfo_name + '.dist-info', 'REQUESTED')
+        self.assertRaises(DistutilsError, dist.get_distinfo_file,
+            other_distinfo_file)
+        # Test for a file that does not exist and should not exist
+        self.assertRaises(DistutilsError, dist.get_distinfo_file, 'ENTRYPOINTS')
+
+    def test_get_distinfo_files(self):
+        """Test for the iteration of RECORD path entries."""
+        from distutils2._backport.pkgutil import Distribution
+        distinfo_name = 'towel_stuff-0.1'
+        distinfo_dir = os.path.join(self.fake_dists_path,
+            distinfo_name + '.dist-info')
+        dist = Distribution(distinfo_dir)
+        # Test for the iteration of the raw path
+        distinfo_record_paths = self.records[distinfo_dir].keys()
+        found = [ path for path in dist.get_distinfo_files() ]
+        self.assertEqual(sorted(found), sorted(distinfo_record_paths))
+        # Test for the iteration of local absolute paths
+        distinfo_record_paths = [ os.path.join(sys.prefix, path)
+            for path in self.records[distinfo_dir].keys()
+            ]
+        found = [ path for path in dist.get_distinfo_files(local=True) ]
+        self.assertEqual(sorted(found), sorted(distinfo_record_paths))
+
+
+class TestPkgUtilFunctions(unittest2.TestCase):
+    """Tests for the new functionality added in PEP 376."""
+
+    def setUp(self):
+        super(TestPkgUtilFunctions, self).setUp()
+        # Setup the path environment with our fake distributions
+        current_path = os.path.abspath(os.path.dirname(__file__))
+        self.sys_path = sys.path[:]
+        self.fake_dists_path = os.path.join(current_path, 'fake_dists')
+        sys.path[0:0] = [self.fake_dists_path]
+
+    def tearDown(self):
+        super(TestPkgUtilFunctions, self).tearDown()
+        sys.path[:] = self.sys_path
+
+    def test_distinfo_dirname(self):
+        """Given a name and a version, we expect the distinfo_dirname function
+        to return a standard distribution information directory name."""
+
+        items = [ # (name, version, standard_dirname)
+            # Test for a very simple single word name and decimal version number
+            ('docutils', '0.5', 'docutils-0.5.dist-info'),
+            # Test for another except this time with a '-' in the name, which
+            #   needs to be transformed during the name lookup
+            ('python-ldap', '2.5', 'python_ldap-2.5.dist-info'),
+            # Test for both '-' in the name and a funky version number
+            ('python-ldap', '2.5 a---5', 'python_ldap-2.5 a---5.dist-info'),
+            ]
+
+        # Import the function in question
+        from distutils2._backport.pkgutil import distinfo_dirname
+
+        # Loop through the items to validate the results
+        for name, version, standard_dirname in items:
+            dirname = distinfo_dirname(name, version)
+            self.assertEqual(dirname, standard_dirname)
+
+    def test_get_distributions(self):
+        """Lookup all distributions found in the ``sys.path``."""
+        # This test could potentially pick up other installed distributions
+        fake_dists = [('grammar', '1.0a4'), ('choxie', '2.0.0.9'),
+            ('towel-stuff', '0.1')]
+        found_dists = []
+
+        # Import the function in question
+        from distutils2._backport.pkgutil import get_distributions, Distribution
+
+        # Verify the fake dists have been found.
+        dists = [ dist for dist in get_distributions() ]
+        for dist in dists:
+            if not isinstance(dist, Distribution):
+                self.fail("item received was not a Distribution instance: "
+                    "%s" % type(dist))
+            if dist.name in dict(fake_dists).keys():
+                found_dists.append((dist.name, dist.metadata['version'],))
+            # otherwise we don't care what other distributions are found
+
+        # Finally, test that we found all that we were looking for
+        self.assertListEqual(sorted(found_dists), sorted(fake_dists))
+
+    def test_get_distribution(self):
+        """Test for looking up a distribution by name."""
+        # Test the lookup of the towel-stuff distribution
+        name = 'towel-stuff' # Note: This is different from the directory name
+
+        # Import the function in question
+        from distutils2._backport.pkgutil import get_distribution, Distribution
+
+        # Lookup the distribution
+        dist = get_distribution(name)
+        self.assertTrue(isinstance(dist, Distribution))
+        self.assertEqual(dist.name, name)
+
+        # Verify that an unknown distribution returns None
+        self.assertEqual(None, get_distribution('bogus'))
+
+        # Verify partial name matching doesn't work
+        self.assertEqual(None, get_distribution('towel'))
+
+    def test_get_file_users(self):
+        """Test the iteration of distributions that use a file."""
+        from distutils2._backport.pkgutil import get_file_users, Distribution
+        name = 'towel_stuff-0.1'
+        path = os.path.join(self.fake_dists_path, name,
+            'towel_stuff', '__init__.py')
+        for dist in get_file_users(path):
+            self.assertTrue(isinstance(dist, Distribution))
+            self.assertEqual(dist.name, name)
+
+
+def test_suite():
+    suite = unittest2.TestSuite()
+    testcase_loader = unittest2.loader.defaultTestLoader.loadTestsFromTestCase
+    suite.addTest(testcase_loader(TestPkgUtilFunctions))
+    suite.addTest(testcase_loader(TestPkgUtilDistribution))
+    return suite
+
+def test_main():
+    run_unittest(test_suite())
+
+if __name__ == "__main__":
+    test_main()
diff --git a/src/distutils2/command/check.py b/src/distutils2/command/check.py
--- a/src/distutils2/command/check.py
+++ b/src/distutils2/command/check.py
@@ -25,14 +25,14 @@
         self.restructuredtext = 0
         self.metadata = 1
         self.strict = 0
-        self._warnings = 0
+        self._warnings = []
 
     def finalize_options(self):
         pass
 
     def warn(self, msg):
         """Counts the number of warnings that occurs."""
-        self._warnings += 1
+        self._warnings.append(msg)
         return Command.warn(self, msg)
 
     def run(self):
@@ -48,8 +48,9 @@
 
         # let's raise an error in strict mode, if we have at least
         # one warning
-        if self.strict and self._warnings > 0:
-            raise DistutilsSetupError('Please correct your package.')
+        if self.strict and len(self._warnings) > 0:
+            msg = '\n'.join(self._warnings)
+            raise DistutilsSetupError(msg)
 
     def check_metadata(self):
         """Ensures that all required elements of meta-data are supplied.
diff --git a/src/distutils2/command/register.py b/src/distutils2/command/register.py
--- a/src/distutils2/command/register.py
+++ b/src/distutils2/command/register.py
@@ -224,39 +224,11 @@
     def build_post_data(self, action):
         # figure the data to send - the metadata plus some additional
         # information used by the package server
-        meta = self.distribution.metadata
-        data = {
-            ':action': action,
-            'metadata_version' : meta.version,
-            'name': meta['Name'],
-            'version': meta['Version'],
-            'summary': meta['Summary'],
-            'home_page': meta['Home-page'],
-            'author': meta['Author'],
-            'author_email': meta['Author-email'],
-            'license': meta['License'],
-            'description': meta['Description'],
-            'keywords': meta['Keywords'],
-            'platform': meta['Platform'],
-            'classifier': meta['Classifier'],
-            'download_url': meta['Download-URL'],
-        }
-
-        if meta.version == '1.2':
-            data['requires_dist'] = meta['Requires-Dist']
-            data['requires_python'] = meta['Requires-Python']
-            data['requires_external'] = meta['Requires-External']
-            data['provides_dist'] = meta['Provides-Dist']
-            data['obsoletes_dist'] = meta['Obsoletes-Dist']
-            data['project_url'] = meta['Project-Url']
-
-        elif meta.version == '1.1':
-            data['provides'] = meta['Provides']
-            data['requires'] = meta['Requires']
-            data['obsoletes'] = meta['Obsoletes']
-
+        data = self._metadata_to_pypy_dict()
+        data[':action'] = action
         return data
 
+    # XXX to be refactored with upload.upload_file
     def post_to_server(self, data, auth=None):
         ''' Post a query to the server, and return a string response.
         '''
diff --git a/src/distutils2/command/upload.py b/src/distutils2/command/upload.py
--- a/src/distutils2/command/upload.py
+++ b/src/distutils2/command/upload.py
@@ -7,7 +7,7 @@
 from urllib2 import urlopen, Request, HTTPError
 from base64 import standard_b64encode
 import urlparse
-import cStringIO as StringIO
+import StringIO as StringIO
 try:
     from hashlib import md5
 except ImportError:
@@ -62,6 +62,7 @@
         for command, pyversion, filename in self.distribution.dist_files:
             self.upload_file(command, pyversion, filename)
 
+    # XXX to be refactored with register.post_to_server
     def upload_file(self, command, pyversion, filename):
         # Makes sure the repository URL is compliant
         schema, netloc, url, params, query, fragments = \
@@ -83,42 +84,17 @@
         # Fill in the data - send all the meta-data in case we need to
         # register a new release
         content = open(filename,'rb').read()
-        meta = self.distribution.metadata
 
-        data = {
-            # action
-            ':action': 'file_upload',
-            'protcol_version': '1',
+        data = self._metadata_to_pypy_dict()
 
-            # identify release
-            'name': meta['Name'],
-            'version': meta['Version'],
+        # extra upload infos
+        data[':action'] = 'file_upload'
+        data['protcol_version'] = '1'
+        data['content'] = [os.path.basename(filename), content]
+        data['filetype'] = command
+        data['pyversion'] = pyversion
+        data['md5_digest'] = md5(content).hexdigest()
 
-            # file content
-            'content': (os.path.basename(filename),content),
-            'filetype': command,
-            'pyversion': pyversion,
-            'md5_digest': md5(content).hexdigest(),
-
-            # additional meta-data
-            # XXX Implement 1.1
-            'metadata_version' : '1.0',
-            'name': meta['Name'],
-            'version': meta['Version'],
-            'summary': meta['Summary'],
-            'home_page': meta['Home-page'],
-            'author': meta['Author'],
-            'author_email': meta['Author-email'],
-            'license': meta['License'],
-            'description': meta['Description'],
-            'keywords': meta['Keywords'],
-            'platform': meta['Platform'],
-            'classifiers': meta['Classifier'],
-            'download_url': meta['Download-URL'],
-            #'provides': meta['Provides'],
-            #'requires': meta['Requires'],
-            #'obsoletes': meta['Obsoletes'],
-            }
         comment = ''
         if command == 'bdist_rpm':
             dist, version, id = platform.dist()
@@ -129,8 +105,8 @@
         data['comment'] = comment
 
         if self.sign:
-            data['gpg_signature'] = (os.path.basename(filename) + ".asc",
-                                     open(filename+".asc").read())
+            data['gpg_signature'] = [(os.path.basename(filename) + ".asc",
+                                      open(filename+".asc").read())]
 
         # set up the authentication
         auth = "Basic " + standard_b64encode(self.username + ":" +
@@ -141,29 +117,41 @@
         sep_boundary = '\n--' + boundary
         end_boundary = sep_boundary + '--'
         body = StringIO.StringIO()
-        for key, value in data.items():
+        file_fields = ('content', 'gpg_signature')
+
+        for key, values in data.items():
             # handle multiple entries for the same name
-            if not isinstance(value, list):
-                value = [value]
-            for value in value:
-                if isinstance(value, tuple):
-                    fn = ';filename="%s"' % value[0]
-                    value = value[1]
-                else:
-                    fn = ""
+            if not isinstance(values, (tuple, list)):
+                values = [values]
 
+            content_dispo = 'Content-Disposition: form-data; name="%s"' % key
+
+            if key in file_fields:
+                filename_, content = values
+                filename_ = ';filename="%s"' % filename_
                 body.write(sep_boundary)
-                body.write('\nContent-Disposition: form-data; name="%s"'%key)
-                body.write(fn)
+                body.write("\n")
+                body.write(content_dispo)
+                body.write(filename_)
                 body.write("\n\n")
-                body.write(value)
-                if value and value[-1] == '\r':
-                    body.write('\n')  # write an extra newline (lurve Macs)
+                body.write(content)
+            else:
+                for value in values:
+                    body.write(sep_boundary)
+                    body.write("\n")
+                    body.write(content_dispo)
+                    body.write("\n\n")
+                    body.write(value)
+                    if value and value[-1] == '\r':
+                        # write an extra newline (lurve Macs)
+                        body.write('\n')
+
         body.write(end_boundary)
         body.write("\n")
         body = body.getvalue()
 
-        self.announce("Submitting %s to %s" % (filename, self.repository), log.INFO)
+        self.announce("Submitting %s to %s" % (filename, self.repository),
+                      log.INFO)
 
         # build the Request
         headers = {'Content-type':
diff --git a/src/distutils2/config.py b/src/distutils2/config.py
--- a/src/distutils2/config.py
+++ b/src/distutils2/config.py
@@ -107,6 +107,40 @@
 
         return {}
 
+    def _metadata_to_pypy_dict(self):
+        meta = self.distribution.metadata
+        data = {
+            'metadata_version' : meta.version,
+            'name': meta['Name'],
+            'version': meta['Version'],
+            'summary': meta['Summary'],
+            'home_page': meta['Home-page'],
+            'author': meta['Author'],
+            'author_email': meta['Author-email'],
+            'license': meta['License'],
+            'description': meta['Description'],
+            'keywords': meta['Keywords'],
+            'platform': meta['Platform'],
+            'classifier': meta['Classifier'],
+            'download_url': meta['Download-URL'],
+        }
+
+        if meta.version == '1.2':
+            data['requires_dist'] = meta['Requires-Dist']
+            data['requires_python'] = meta['Requires-Python']
+            data['requires_external'] = meta['Requires-External']
+            data['provides_dist'] = meta['Provides-Dist']
+            data['obsoletes_dist'] = meta['Obsoletes-Dist']
+            data['project_url'] = [','.join(url) for url in
+                                   meta['Project-URL']]
+
+        elif meta.version == '1.1':
+            data['provides'] = meta['Provides']
+            data['requires'] = meta['Requires']
+            data['obsoletes'] = meta['Obsoletes']
+
+        return data
+
     def initialize_options(self):
         """Initialize options."""
         self.repository = None
diff --git a/src/distutils2/metadata.py b/src/distutils2/metadata.py
--- a/src/distutils2/metadata.py
+++ b/src/distutils2/metadata.py
@@ -77,8 +77,8 @@
                'Requires-Python', 'Requires-External')
 
 _345_MARKERS = ('Provides-Dist', 'Requires-Dist', 'Requires-Python',
-        'Obsoletes-Dist', 'Requires-External', 'Maintainer',
-        'Maintainer-email')
+                'Obsoletes-Dist', 'Requires-External', 'Maintainer',
+                'Maintainer-email', 'Project-URL')
 
 _ALL_FIELDS = []
 
@@ -153,6 +153,7 @@
         'Requires', 'Provides', 'Obsoletes-Dist',
         'Provides-Dist', 'Requires-Dist', 'Requires-External',
         'Project-URL')
+_LISTTUPLEFIELDS = ('Project-URL',)
 
 _ELEMENTSFIELD = ('Keywords',)
 
@@ -355,7 +356,11 @@
                 valid, val = self._platform(val)
                 if not valid:
                     continue
-                res.append(self._encode_field(val))
+                if name not in _LISTTUPLEFIELDS:
+                    res.append(self._encode_field(val))
+                else:
+                    # That's for Project-URL
+                    res.append((self._encode_field(val[0]), val[1]))
             return res
 
         elif name in _ELEMENTSFIELD:
@@ -382,6 +387,30 @@
             warnings = self._check_rst_data(self['Description'])
         else:
             warnings = []
+
+        # checking metadata 1.2 (XXX needs to check 1.1, 1.0)
+        if self['Metadata-Version'] != '1.2':
+            return missing, warnings
+
+
+        def is_valid_predicates(value):
+            for v in value:
+                if not is_valid_predicate(v.split(';')[0]):
+                    return False
+            return True
+
+        for fields, controller in ((_PREDICATE_FIELDS, is_valid_predicates),
+                                  (_VERSIONS_FIELDS, is_valid_versions),
+                                  (_VERSION_FIELDS, is_valid_version)):
+            for field in fields:
+                value = self[field]
+                if value == 'UNKNOWN':
+                    continue
+
+                if not controller(value):
+                    warnings.append('Wrong value for "%s": %s' \
+                            % (field, value))
+
         return missing, warnings
 
     def keys(self):
diff --git a/src/distutils2/mkpkg.py b/src/distutils2/mkpkg.py
--- a/src/distutils2/mkpkg.py
+++ b/src/distutils2/mkpkg.py
@@ -73,774 +73,776 @@
 }
 
 troveList = [
-      'Development Status :: 1 - Planning',
-		'Development Status :: 2 - Pre-Alpha',
-		'Development Status :: 3 - Alpha',
-		'Development Status :: 4 - Beta',
-		'Development Status :: 5 - Production/Stable',
-		'Development Status :: 6 - Mature',
-		'Development Status :: 7 - Inactive',
-		'Environment :: Console',
-		'Environment :: Console :: Curses',
-		'Environment :: Console :: Framebuffer',
-		'Environment :: Console :: Newt',
-		'Environment :: Console :: svgalib',
-		'Environment :: Handhelds/PDA\'s',
-		'Environment :: MacOS X',
-		'Environment :: MacOS X :: Aqua',
-		'Environment :: MacOS X :: Carbon',
-		'Environment :: MacOS X :: Cocoa',
-		'Environment :: No Input/Output (Daemon)',
-		'Environment :: Other Environment',
-		'Environment :: Plugins',
-		'Environment :: Web Environment',
-		'Environment :: Web Environment :: Buffet',
-		'Environment :: Web Environment :: Mozilla',
-		'Environment :: Web Environment :: ToscaWidgets',
-		'Environment :: Win32 (MS Windows)',
-		'Environment :: X11 Applications',
-		'Environment :: X11 Applications :: Gnome',
-		'Environment :: X11 Applications :: GTK',
-		'Environment :: X11 Applications :: KDE',
-		'Environment :: X11 Applications :: Qt',
-		'Framework :: BFG',
-		'Framework :: Buildout',
-		'Framework :: Chandler',
-		'Framework :: CubicWeb',
-		'Framework :: Django',
-		'Framework :: IDLE',
-		'Framework :: Paste',
-		'Framework :: Plone',
-		'Framework :: Pylons',
-		'Framework :: Setuptools Plugin',
-		'Framework :: Trac',
-		'Framework :: TurboGears',
-		'Framework :: TurboGears :: Applications',
-		'Framework :: TurboGears :: Widgets',
-		'Framework :: Twisted',
-		'Framework :: ZODB',
-		'Framework :: Zope2',
-		'Framework :: Zope3',
-		'Intended Audience :: Customer Service',
-		'Intended Audience :: Developers',
-		'Intended Audience :: Education',
-		'Intended Audience :: End Users/Desktop',
-		'Intended Audience :: Financial and Insurance Industry',
-		'Intended Audience :: Healthcare Industry',
-		'Intended Audience :: Information Technology',
-		'Intended Audience :: Legal Industry',
-		'Intended Audience :: Manufacturing',
-		'Intended Audience :: Other Audience',
-		'Intended Audience :: Religion',
-		'Intended Audience :: Science/Research',
-		'Intended Audience :: System Administrators',
-		'Intended Audience :: Telecommunications Industry',
-		'License :: Aladdin Free Public License (AFPL)',
-		'License :: DFSG approved',
-		'License :: Eiffel Forum License (EFL)',
-		'License :: Free For Educational Use',
-		'License :: Free For Home Use',
-		'License :: Free for non-commercial use',
-		'License :: Freely Distributable',
-		'License :: Free To Use But Restricted',
-		'License :: Freeware',
-		'License :: Netscape Public License (NPL)',
-		'License :: Nokia Open Source License (NOKOS)',
-		'License :: OSI Approved',
-		'License :: OSI Approved :: Academic Free License (AFL)',
-		'License :: OSI Approved :: Apache Software License',
-		'License :: OSI Approved :: Apple Public Source License',
-		'License :: OSI Approved :: Artistic License',
-		'License :: OSI Approved :: Attribution Assurance License',
-		'License :: OSI Approved :: BSD License',
-		'License :: OSI Approved :: Common Public License',
-		'License :: OSI Approved :: Eiffel Forum License',
-		'License :: OSI Approved :: European Union Public Licence 1.0 (EUPL 1.0)',
-		'License :: OSI Approved :: European Union Public Licence 1.1 (EUPL 1.1)',
-		'License :: OSI Approved :: GNU Affero General Public License v3',
-		'License :: OSI Approved :: GNU Free Documentation License (FDL)',
-		'License :: OSI Approved :: GNU General Public License (GPL)',
-		'License :: OSI Approved :: GNU Library or Lesser General Public License (LGPL)',
-		'License :: OSI Approved :: IBM Public License',
-		'License :: OSI Approved :: Intel Open Source License',
-		'License :: OSI Approved :: ISC License (ISCL)',
-		'License :: OSI Approved :: Jabber Open Source License',
-		'License :: OSI Approved :: MIT License',
-		'License :: OSI Approved :: MITRE Collaborative Virtual Workspace License (CVW)',
-		'License :: OSI Approved :: Motosoto License',
-		'License :: OSI Approved :: Mozilla Public License 1.0 (MPL)',
-		'License :: OSI Approved :: Mozilla Public License 1.1 (MPL 1.1)',
-		'License :: OSI Approved :: Nethack General Public License',
-		'License :: OSI Approved :: Nokia Open Source License',
-		'License :: OSI Approved :: Open Group Test Suite License',
-		'License :: OSI Approved :: Python License (CNRI Python License)',
-		'License :: OSI Approved :: Python Software Foundation License',
-		'License :: OSI Approved :: Qt Public License (QPL)',
-		'License :: OSI Approved :: Ricoh Source Code Public License',
-		'License :: OSI Approved :: Sleepycat License',
-		'License :: OSI Approved :: Sun Industry Standards Source License (SISSL)',
-		'License :: OSI Approved :: Sun Public License',
-		'License :: OSI Approved :: University of Illinois/NCSA Open Source License',
-		'License :: OSI Approved :: Vovida Software License 1.0',
-		'License :: OSI Approved :: W3C License',
-		'License :: OSI Approved :: X.Net License',
-		'License :: OSI Approved :: zlib/libpng License',
-		'License :: OSI Approved :: Zope Public License',
-		'License :: Other/Proprietary License',
-		'License :: Public Domain',
-		'License :: Repoze Public License',
-		'Natural Language :: Afrikaans',
-		'Natural Language :: Arabic',
-		'Natural Language :: Bengali',
-		'Natural Language :: Bosnian',
-		'Natural Language :: Bulgarian',
-		'Natural Language :: Catalan',
-		'Natural Language :: Chinese (Simplified)',
-		'Natural Language :: Chinese (Traditional)',
-		'Natural Language :: Croatian',
-		'Natural Language :: Czech',
-		'Natural Language :: Danish',
-		'Natural Language :: Dutch',
-		'Natural Language :: English',
-		'Natural Language :: Esperanto',
-		'Natural Language :: Finnish',
-		'Natural Language :: French',
-		'Natural Language :: German',
-		'Natural Language :: Greek',
-		'Natural Language :: Hebrew',
-		'Natural Language :: Hindi',
-		'Natural Language :: Hungarian',
-		'Natural Language :: Icelandic',
-		'Natural Language :: Indonesian',
-		'Natural Language :: Italian',
-		'Natural Language :: Japanese',
-		'Natural Language :: Javanese',
-		'Natural Language :: Korean',
-		'Natural Language :: Latin',
-		'Natural Language :: Latvian',
-		'Natural Language :: Macedonian',
-		'Natural Language :: Malay',
-		'Natural Language :: Marathi',
-		'Natural Language :: Norwegian',
-		'Natural Language :: Panjabi',
-		'Natural Language :: Persian',
-		'Natural Language :: Polish',
-		'Natural Language :: Portuguese',
-		'Natural Language :: Portuguese (Brazilian)',
-		'Natural Language :: Romanian',
-		'Natural Language :: Russian',
-		'Natural Language :: Serbian',
-		'Natural Language :: Slovak',
-		'Natural Language :: Slovenian',
-		'Natural Language :: Spanish',
-		'Natural Language :: Swedish',
-		'Natural Language :: Tamil',
-		'Natural Language :: Telugu',
-		'Natural Language :: Thai',
-		'Natural Language :: Turkish',
-		'Natural Language :: Ukranian',
-		'Natural Language :: Urdu',
-		'Natural Language :: Vietnamese',
-		'Operating System :: BeOS',
-		'Operating System :: MacOS',
-		'Operating System :: MacOS :: MacOS 9',
-		'Operating System :: MacOS :: MacOS X',
-		'Operating System :: Microsoft',
-		'Operating System :: Microsoft :: MS-DOS',
-		'Operating System :: Microsoft :: Windows',
-		'Operating System :: Microsoft :: Windows :: Windows 3.1 or Earlier',
-		'Operating System :: Microsoft :: Windows :: Windows 95/98/2000',
-		'Operating System :: Microsoft :: Windows :: Windows CE',
-		'Operating System :: Microsoft :: Windows :: Windows NT/2000',
-		'Operating System :: OS/2',
-		'Operating System :: OS Independent',
-		'Operating System :: Other OS',
-		'Operating System :: PalmOS',
-		'Operating System :: PDA Systems',
-		'Operating System :: POSIX',
-		'Operating System :: POSIX :: AIX',
-		'Operating System :: POSIX :: BSD',
-		'Operating System :: POSIX :: BSD :: BSD/OS',
-		'Operating System :: POSIX :: BSD :: FreeBSD',
-		'Operating System :: POSIX :: BSD :: NetBSD',
-		'Operating System :: POSIX :: BSD :: OpenBSD',
-		'Operating System :: POSIX :: GNU Hurd',
-		'Operating System :: POSIX :: HP-UX',
-		'Operating System :: POSIX :: IRIX',
-		'Operating System :: POSIX :: Linux',
-		'Operating System :: POSIX :: Other',
-		'Operating System :: POSIX :: SCO',
-		'Operating System :: POSIX :: SunOS/Solaris',
-		'Operating System :: Unix',
-		'Programming Language :: Ada',
-		'Programming Language :: APL',
-		'Programming Language :: ASP',
-		'Programming Language :: Assembly',
-		'Programming Language :: Awk',
-		'Programming Language :: Basic',
-		'Programming Language :: C',
-		'Programming Language :: C#',
-		'Programming Language :: C++',
-		'Programming Language :: Cold Fusion',
-		'Programming Language :: Cython',
-		'Programming Language :: Delphi/Kylix',
-		'Programming Language :: Dylan',
-		'Programming Language :: Eiffel',
-		'Programming Language :: Emacs-Lisp',
-		'Programming Language :: Erlang',
-		'Programming Language :: Euler',
-		'Programming Language :: Euphoria',
-		'Programming Language :: Forth',
-		'Programming Language :: Fortran',
-		'Programming Language :: Haskell',
-		'Programming Language :: Java',
-		'Programming Language :: JavaScript',
-		'Programming Language :: Lisp',
-		'Programming Language :: Logo',
-		'Programming Language :: ML',
-		'Programming Language :: Modula',
-		'Programming Language :: Objective C',
-		'Programming Language :: Object Pascal',
-		'Programming Language :: OCaml',
-		'Programming Language :: Other',
-		'Programming Language :: Other Scripting Engines',
-		'Programming Language :: Pascal',
-		'Programming Language :: Perl',
-		'Programming Language :: PHP',
-		'Programming Language :: Pike',
-		'Programming Language :: Pliant',
-		'Programming Language :: PL/SQL',
-		'Programming Language :: PROGRESS',
-		'Programming Language :: Prolog',
-		'Programming Language :: Python',
-		'Programming Language :: Python :: 2',
-		'Programming Language :: Python :: 2.3',
-		'Programming Language :: Python :: 2.4',
-		'Programming Language :: Python :: 2.5',
-		'Programming Language :: Python :: 2.6',
-		'Programming Language :: Python :: 2.7',
-		'Programming Language :: Python :: 3',
-		'Programming Language :: Python :: 3.0',
-		'Programming Language :: Python :: 3.1',
-		'Programming Language :: Python :: 3.2',
-		'Programming Language :: REBOL',
-		'Programming Language :: Rexx',
-		'Programming Language :: Ruby',
-		'Programming Language :: Scheme',
-		'Programming Language :: Simula',
-		'Programming Language :: Smalltalk',
-		'Programming Language :: SQL',
-		'Programming Language :: Tcl',
-		'Programming Language :: Unix Shell',
-		'Programming Language :: Visual Basic',
-		'Programming Language :: XBasic',
-		'Programming Language :: YACC',
-		'Programming Language :: Zope',
-		'Topic :: Adaptive Technologies',
-		'Topic :: Artistic Software',
-		'Topic :: Communications',
-		'Topic :: Communications :: BBS',
-		'Topic :: Communications :: Chat',
-		'Topic :: Communications :: Chat :: AOL Instant Messenger',
-		'Topic :: Communications :: Chat :: ICQ',
-		'Topic :: Communications :: Chat :: Internet Relay Chat',
-		'Topic :: Communications :: Chat :: Unix Talk',
-		'Topic :: Communications :: Conferencing',
-		'Topic :: Communications :: Email',
-		'Topic :: Communications :: Email :: Address Book',
-		'Topic :: Communications :: Email :: Email Clients (MUA)',
-		'Topic :: Communications :: Email :: Filters',
-		'Topic :: Communications :: Email :: Mailing List Servers',
-		'Topic :: Communications :: Email :: Mail Transport Agents',
-		'Topic :: Communications :: Email :: Post-Office',
-		'Topic :: Communications :: Email :: Post-Office :: IMAP',
-		'Topic :: Communications :: Email :: Post-Office :: POP3',
-		'Topic :: Communications :: Fax',
-		'Topic :: Communications :: FIDO',
-		'Topic :: Communications :: File Sharing',
-		'Topic :: Communications :: File Sharing :: Gnutella',
-		'Topic :: Communications :: File Sharing :: Napster',
-		'Topic :: Communications :: Ham Radio',
-		'Topic :: Communications :: Internet Phone',
-		'Topic :: Communications :: Telephony',
-		'Topic :: Communications :: Usenet News',
-		'Topic :: Database',
-		'Topic :: Database :: Database Engines/Servers',
-		'Topic :: Database :: Front-Ends',
-		'Topic :: Desktop Environment',
-		'Topic :: Desktop Environment :: File Managers',
-		'Topic :: Desktop Environment :: Gnome',
-		'Topic :: Desktop Environment :: GNUstep',
-		'Topic :: Desktop Environment :: K Desktop Environment (KDE)',
-		'Topic :: Desktop Environment :: K Desktop Environment (KDE) :: Themes',
-		'Topic :: Desktop Environment :: PicoGUI',
-		'Topic :: Desktop Environment :: PicoGUI :: Applications',
-		'Topic :: Desktop Environment :: PicoGUI :: Themes',
-		'Topic :: Desktop Environment :: Screen Savers',
-		'Topic :: Desktop Environment :: Window Managers',
-		'Topic :: Desktop Environment :: Window Managers :: Afterstep',
-		'Topic :: Desktop Environment :: Window Managers :: Afterstep :: Themes',
-		'Topic :: Desktop Environment :: Window Managers :: Applets',
-		'Topic :: Desktop Environment :: Window Managers :: Blackbox',
-		'Topic :: Desktop Environment :: Window Managers :: Blackbox :: Themes',
-		'Topic :: Desktop Environment :: Window Managers :: CTWM',
-		'Topic :: Desktop Environment :: Window Managers :: CTWM :: Themes',
-		'Topic :: Desktop Environment :: Window Managers :: Enlightenment',
-		'Topic :: Desktop Environment :: Window Managers :: Enlightenment :: Epplets',
-		'Topic :: Desktop Environment :: Window Managers :: Enlightenment :: Themes DR15',
-		'Topic :: Desktop Environment :: Window Managers :: Enlightenment :: Themes DR16',
-		'Topic :: Desktop Environment :: Window Managers :: Enlightenment :: Themes DR17',
-		'Topic :: Desktop Environment :: Window Managers :: Fluxbox',
-		'Topic :: Desktop Environment :: Window Managers :: Fluxbox :: Themes',
-		'Topic :: Desktop Environment :: Window Managers :: FVWM',
-		'Topic :: Desktop Environment :: Window Managers :: FVWM :: Themes',
-		'Topic :: Desktop Environment :: Window Managers :: IceWM',
-		'Topic :: Desktop Environment :: Window Managers :: IceWM :: Themes',
-		'Topic :: Desktop Environment :: Window Managers :: MetaCity',
-		'Topic :: Desktop Environment :: Window Managers :: MetaCity :: Themes',
-		'Topic :: Desktop Environment :: Window Managers :: Oroborus',
-		'Topic :: Desktop Environment :: Window Managers :: Oroborus :: Themes',
-		'Topic :: Desktop Environment :: Window Managers :: Sawfish',
-		'Topic :: Desktop Environment :: Window Managers :: Sawfish :: Themes 0.30',
-		'Topic :: Desktop Environment :: Window Managers :: Sawfish :: Themes pre-0.30',
-		'Topic :: Desktop Environment :: Window Managers :: Waimea',
-		'Topic :: Desktop Environment :: Window Managers :: Waimea :: Themes',
-		'Topic :: Desktop Environment :: Window Managers :: Window Maker',
-		'Topic :: Desktop Environment :: Window Managers :: Window Maker :: Applets',
-		'Topic :: Desktop Environment :: Window Managers :: Window Maker :: Themes',
-		'Topic :: Desktop Environment :: Window Managers :: XFCE',
-		'Topic :: Desktop Environment :: Window Managers :: XFCE :: Themes',
-		'Topic :: Documentation',
-		'Topic :: Education',
-		'Topic :: Education :: Computer Aided Instruction (CAI)',
-		'Topic :: Education :: Testing',
-		'Topic :: Games/Entertainment',
-		'Topic :: Games/Entertainment :: Arcade',
-		'Topic :: Games/Entertainment :: Board Games',
-		'Topic :: Games/Entertainment :: First Person Shooters',
-		'Topic :: Games/Entertainment :: Fortune Cookies',
-		'Topic :: Games/Entertainment :: Multi-User Dungeons (MUD)',
-		'Topic :: Games/Entertainment :: Puzzle Games',
-		'Topic :: Games/Entertainment :: Real Time Strategy',
-		'Topic :: Games/Entertainment :: Role-Playing',
-		'Topic :: Games/Entertainment :: Side-Scrolling/Arcade Games',
-		'Topic :: Games/Entertainment :: Simulation',
-		'Topic :: Games/Entertainment :: Turn Based Strategy',
-		'Topic :: Home Automation',
-		'Topic :: Internet',
-		'Topic :: Internet :: File Transfer Protocol (FTP)',
-		'Topic :: Internet :: Finger',
-		'Topic :: Internet :: Log Analysis',
-		'Topic :: Internet :: Name Service (DNS)',
-		'Topic :: Internet :: Proxy Servers',
-		'Topic :: Internet :: WAP',
-		'Topic :: Internet :: WWW/HTTP',
-		'Topic :: Internet :: WWW/HTTP :: Browsers',
-		'Topic :: Internet :: WWW/HTTP :: Dynamic Content',
-		'Topic :: Internet :: WWW/HTTP :: Dynamic Content :: CGI Tools/Libraries',
-		'Topic :: Internet :: WWW/HTTP :: Dynamic Content :: Message Boards',
-		'Topic :: Internet :: WWW/HTTP :: Dynamic Content :: News/Diary',
-		'Topic :: Internet :: WWW/HTTP :: Dynamic Content :: Page Counters',
-		'Topic :: Internet :: WWW/HTTP :: HTTP Servers',
-		'Topic :: Internet :: WWW/HTTP :: Indexing/Search',
-		'Topic :: Internet :: WWW/HTTP :: Site Management',
-		'Topic :: Internet :: WWW/HTTP :: Site Management :: Link Checking',
-		'Topic :: Internet :: WWW/HTTP :: WSGI',
-		'Topic :: Internet :: WWW/HTTP :: WSGI :: Application',
-		'Topic :: Internet :: WWW/HTTP :: WSGI :: Middleware',
-		'Topic :: Internet :: WWW/HTTP :: WSGI :: Server',
-		'Topic :: Internet :: Z39.50',
-		'Topic :: Multimedia',
-		'Topic :: Multimedia :: Graphics',
-		'Topic :: Multimedia :: Graphics :: 3D Modeling',
-		'Topic :: Multimedia :: Graphics :: 3D Rendering',
-		'Topic :: Multimedia :: Graphics :: Capture',
-		'Topic :: Multimedia :: Graphics :: Capture :: Digital Camera',
-		'Topic :: Multimedia :: Graphics :: Capture :: Scanners',
-		'Topic :: Multimedia :: Graphics :: Capture :: Screen Capture',
-		'Topic :: Multimedia :: Graphics :: Editors',
-		'Topic :: Multimedia :: Graphics :: Editors :: Raster-Based',
-		'Topic :: Multimedia :: Graphics :: Editors :: Vector-Based',
-		'Topic :: Multimedia :: Graphics :: Graphics Conversion',
-		'Topic :: Multimedia :: Graphics :: Presentation',
-		'Topic :: Multimedia :: Graphics :: Viewers',
-		'Topic :: Multimedia :: Sound/Audio',
-		'Topic :: Multimedia :: Sound/Audio :: Analysis',
-		'Topic :: Multimedia :: Sound/Audio :: Capture/Recording',
-		'Topic :: Multimedia :: Sound/Audio :: CD Audio',
-		'Topic :: Multimedia :: Sound/Audio :: CD Audio :: CD Playing',
-		'Topic :: Multimedia :: Sound/Audio :: CD Audio :: CD Ripping',
-		'Topic :: Multimedia :: Sound/Audio :: CD Audio :: CD Writing',
-		'Topic :: Multimedia :: Sound/Audio :: Conversion',
-		'Topic :: Multimedia :: Sound/Audio :: Editors',
-		'Topic :: Multimedia :: Sound/Audio :: MIDI',
-		'Topic :: Multimedia :: Sound/Audio :: Mixers',
-		'Topic :: Multimedia :: Sound/Audio :: Players',
-		'Topic :: Multimedia :: Sound/Audio :: Players :: MP3',
-		'Topic :: Multimedia :: Sound/Audio :: Sound Synthesis',
-		'Topic :: Multimedia :: Sound/Audio :: Speech',
-		'Topic :: Multimedia :: Video',
-		'Topic :: Multimedia :: Video :: Capture',
-		'Topic :: Multimedia :: Video :: Conversion',
-		'Topic :: Multimedia :: Video :: Display',
-		'Topic :: Multimedia :: Video :: Non-Linear Editor',
-		'Topic :: Office/Business',
-		'Topic :: Office/Business :: Financial',
-		'Topic :: Office/Business :: Financial :: Accounting',
-		'Topic :: Office/Business :: Financial :: Investment',
-		'Topic :: Office/Business :: Financial :: Point-Of-Sale',
-		'Topic :: Office/Business :: Financial :: Spreadsheet',
-		'Topic :: Office/Business :: Groupware',
-		'Topic :: Office/Business :: News/Diary',
-		'Topic :: Office/Business :: Office Suites',
-		'Topic :: Office/Business :: Scheduling',
-		'Topic :: Other/Nonlisted Topic',
-		'Topic :: Printing',
-		'Topic :: Religion',
-		'Topic :: Scientific/Engineering',
-		'Topic :: Scientific/Engineering :: Artificial Intelligence',
-		'Topic :: Scientific/Engineering :: Astronomy',
-		'Topic :: Scientific/Engineering :: Atmospheric Science',
-		'Topic :: Scientific/Engineering :: Bio-Informatics',
-		'Topic :: Scientific/Engineering :: Chemistry',
-		'Topic :: Scientific/Engineering :: Electronic Design Automation (EDA)',
-		'Topic :: Scientific/Engineering :: GIS',
-		'Topic :: Scientific/Engineering :: Human Machine Interfaces',
-		'Topic :: Scientific/Engineering :: Image Recognition',
-		'Topic :: Scientific/Engineering :: Information Analysis',
-		'Topic :: Scientific/Engineering :: Interface Engine/Protocol Translator',
-		'Topic :: Scientific/Engineering :: Mathematics',
-		'Topic :: Scientific/Engineering :: Medical Science Apps.',
-		'Topic :: Scientific/Engineering :: Physics',
-		'Topic :: Scientific/Engineering :: Visualization',
-		'Topic :: Security',
-		'Topic :: Security :: Cryptography',
-		'Topic :: Sociology',
-		'Topic :: Sociology :: Genealogy',
-		'Topic :: Sociology :: History',
-		'Topic :: Software Development',
-		'Topic :: Software Development :: Assemblers',
-		'Topic :: Software Development :: Bug Tracking',
-		'Topic :: Software Development :: Build Tools',
-		'Topic :: Software Development :: Code Generators',
-		'Topic :: Software Development :: Compilers',
-		'Topic :: Software Development :: Debuggers',
-		'Topic :: Software Development :: Disassemblers',
-		'Topic :: Software Development :: Documentation',
-		'Topic :: Software Development :: Embedded Systems',
-		'Topic :: Software Development :: Internationalization',
-		'Topic :: Software Development :: Interpreters',
-		'Topic :: Software Development :: Libraries',
-		'Topic :: Software Development :: Libraries :: Application Frameworks',
-		'Topic :: Software Development :: Libraries :: Java Libraries',
-		'Topic :: Software Development :: Libraries :: Perl Modules',
-		'Topic :: Software Development :: Libraries :: PHP Classes',
-		'Topic :: Software Development :: Libraries :: Pike Modules',
-		'Topic :: Software Development :: Libraries :: pygame',
-		'Topic :: Software Development :: Libraries :: Python Modules',
-		'Topic :: Software Development :: Libraries :: Ruby Modules',
-		'Topic :: Software Development :: Libraries :: Tcl Extensions',
-		'Topic :: Software Development :: Localization',
-		'Topic :: Software Development :: Object Brokering',
-		'Topic :: Software Development :: Object Brokering :: CORBA',
-		'Topic :: Software Development :: Pre-processors',
-		'Topic :: Software Development :: Quality Assurance',
-		'Topic :: Software Development :: Testing',
-		'Topic :: Software Development :: Testing :: Traffic Generation',
-		'Topic :: Software Development :: User Interfaces',
-		'Topic :: Software Development :: Version Control',
-		'Topic :: Software Development :: Version Control :: CVS',
-		'Topic :: Software Development :: Version Control :: RCS',
-		'Topic :: Software Development :: Version Control :: SCCS',
-		'Topic :: Software Development :: Widget Sets',
-		'Topic :: System',
-		'Topic :: System :: Archiving',
-		'Topic :: System :: Archiving :: Backup',
-		'Topic :: System :: Archiving :: Compression',
-		'Topic :: System :: Archiving :: Mirroring',
-		'Topic :: System :: Archiving :: Packaging',
-		'Topic :: System :: Benchmark',
-		'Topic :: System :: Boot',
-		'Topic :: System :: Boot :: Init',
-		'Topic :: System :: Clustering',
-		'Topic :: System :: Console Fonts',
-		'Topic :: System :: Distributed Computing',
-		'Topic :: System :: Emulators',
-		'Topic :: System :: Filesystems',
-		'Topic :: System :: Hardware',
-		'Topic :: System :: Hardware :: Hardware Drivers',
-		'Topic :: System :: Hardware :: Mainframes',
-		'Topic :: System :: Hardware :: Symmetric Multi-processing',
-		'Topic :: System :: Installation/Setup',
-		'Topic :: System :: Logging',
-		'Topic :: System :: Monitoring',
-		'Topic :: System :: Networking',
-		'Topic :: System :: Networking :: Firewalls',
-		'Topic :: System :: Networking :: Monitoring',
-		'Topic :: System :: Networking :: Monitoring :: Hardware Watchdog',
-		'Topic :: System :: Networking :: Time Synchronization',
-		'Topic :: System :: Operating System',
-		'Topic :: System :: Operating System Kernels',
-		'Topic :: System :: Operating System Kernels :: BSD',
-		'Topic :: System :: Operating System Kernels :: GNU Hurd',
-		'Topic :: System :: Operating System Kernels :: Linux',
-		'Topic :: System :: Power (UPS)',
-		'Topic :: System :: Recovery Tools',
-		'Topic :: System :: Shells',
-		'Topic :: System :: Software Distribution',
-		'Topic :: System :: Systems Administration',
-		'Topic :: System :: Systems Administration :: Authentication/Directory',
-		'Topic :: System :: Systems Administration :: Authentication/Directory :: LDAP',
-		'Topic :: System :: Systems Administration :: Authentication/Directory :: NIS',
-		'Topic :: System :: System Shells',
-		'Topic :: Terminals',
-		'Topic :: Terminals :: Serial',
-		'Topic :: Terminals :: Telnet',
-		'Topic :: Terminals :: Terminal Emulators/X Terminals',
-		'Topic :: Text Editors',
-		'Topic :: Text Editors :: Documentation',
-		'Topic :: Text Editors :: Emacs',
-		'Topic :: Text Editors :: Integrated Development Environments (IDE)',
-		'Topic :: Text Editors :: Text Processing',
-		'Topic :: Text Editors :: Word Processors',
-		'Topic :: Text Processing',
-		'Topic :: Text Processing :: Filters',
-		'Topic :: Text Processing :: Fonts',
-		'Topic :: Text Processing :: General',
-		'Topic :: Text Processing :: Indexing',
-		'Topic :: Text Processing :: Linguistic',
-		'Topic :: Text Processing :: Markup',
-		'Topic :: Text Processing :: Markup :: HTML',
-		'Topic :: Text Processing :: Markup :: LaTeX',
-		'Topic :: Text Processing :: Markup :: SGML',
-		'Topic :: Text Processing :: Markup :: VRML',
-		'Topic :: Text Processing :: Markup :: XML',
-		'Topic :: Utilities',
+        'Development Status :: 1 - Planning',
+        'Development Status :: 2 - Pre-Alpha',
+        'Development Status :: 3 - Alpha',
+        'Development Status :: 4 - Beta',
+        'Development Status :: 5 - Production/Stable',
+        'Development Status :: 6 - Mature',
+        'Development Status :: 7 - Inactive',
+        'Environment :: Console',
+        'Environment :: Console :: Curses',
+        'Environment :: Console :: Framebuffer',
+        'Environment :: Console :: Newt',
+        'Environment :: Console :: svgalib',
+        'Environment :: Handhelds/PDA\'s',
+        'Environment :: MacOS X',
+        'Environment :: MacOS X :: Aqua',
+        'Environment :: MacOS X :: Carbon',
+        'Environment :: MacOS X :: Cocoa',
+        'Environment :: No Input/Output (Daemon)',
+        'Environment :: Other Environment',
+        'Environment :: Plugins',
+        'Environment :: Web Environment',
+        'Environment :: Web Environment :: Buffet',
+        'Environment :: Web Environment :: Mozilla',
+        'Environment :: Web Environment :: ToscaWidgets',
+        'Environment :: Win32 (MS Windows)',
+        'Environment :: X11 Applications',
+        'Environment :: X11 Applications :: Gnome',
+        'Environment :: X11 Applications :: GTK',
+        'Environment :: X11 Applications :: KDE',
+        'Environment :: X11 Applications :: Qt',
+        'Framework :: BFG',
+        'Framework :: Buildout',
+        'Framework :: Chandler',
+        'Framework :: CubicWeb',
+        'Framework :: Django',
+        'Framework :: IDLE',
+        'Framework :: Paste',
+        'Framework :: Plone',
+        'Framework :: Pylons',
+        'Framework :: Setuptools Plugin',
+        'Framework :: Trac',
+        'Framework :: TurboGears',
+        'Framework :: TurboGears :: Applications',
+        'Framework :: TurboGears :: Widgets',
+        'Framework :: Twisted',
+        'Framework :: ZODB',
+        'Framework :: Zope2',
+        'Framework :: Zope3',
+        'Intended Audience :: Customer Service',
+        'Intended Audience :: Developers',
+        'Intended Audience :: Education',
+        'Intended Audience :: End Users/Desktop',
+        'Intended Audience :: Financial and Insurance Industry',
+        'Intended Audience :: Healthcare Industry',
+        'Intended Audience :: Information Technology',
+        'Intended Audience :: Legal Industry',
+        'Intended Audience :: Manufacturing',
+        'Intended Audience :: Other Audience',
+        'Intended Audience :: Religion',
+        'Intended Audience :: Science/Research',
+        'Intended Audience :: System Administrators',
+        'Intended Audience :: Telecommunications Industry',
+        'License :: Aladdin Free Public License (AFPL)',
+        'License :: DFSG approved',
+        'License :: Eiffel Forum License (EFL)',
+        'License :: Free For Educational Use',
+        'License :: Free For Home Use',
+        'License :: Free for non-commercial use',
+        'License :: Freely Distributable',
+        'License :: Free To Use But Restricted',
+        'License :: Freeware',
+        'License :: Netscape Public License (NPL)',
+        'License :: Nokia Open Source License (NOKOS)',
+        'License :: OSI Approved',
+        'License :: OSI Approved :: Academic Free License (AFL)',
+        'License :: OSI Approved :: Apache Software License',
+        'License :: OSI Approved :: Apple Public Source License',
+        'License :: OSI Approved :: Artistic License',
+        'License :: OSI Approved :: Attribution Assurance License',
+        'License :: OSI Approved :: BSD License',
+        'License :: OSI Approved :: Common Public License',
+        'License :: OSI Approved :: Eiffel Forum License',
+        'License :: OSI Approved :: European Union Public Licence 1.0 (EUPL 1.0)',
+        'License :: OSI Approved :: European Union Public Licence 1.1 (EUPL 1.1)',
+        'License :: OSI Approved :: GNU Affero General Public License v3',
+        'License :: OSI Approved :: GNU Free Documentation License (FDL)',
+        'License :: OSI Approved :: GNU General Public License (GPL)',
+        'License :: OSI Approved :: GNU Library or Lesser General Public License (LGPL)',
+        'License :: OSI Approved :: IBM Public License',
+        'License :: OSI Approved :: Intel Open Source License',
+        'License :: OSI Approved :: ISC License (ISCL)',
+        'License :: OSI Approved :: Jabber Open Source License',
+        'License :: OSI Approved :: MIT License',
+        'License :: OSI Approved :: MITRE Collaborative Virtual Workspace License (CVW)',
+        'License :: OSI Approved :: Motosoto License',
+        'License :: OSI Approved :: Mozilla Public License 1.0 (MPL)',
+        'License :: OSI Approved :: Mozilla Public License 1.1 (MPL 1.1)',
+        'License :: OSI Approved :: Nethack General Public License',
+        'License :: OSI Approved :: Nokia Open Source License',
+        'License :: OSI Approved :: Open Group Test Suite License',
+        'License :: OSI Approved :: Python License (CNRI Python License)',
+        'License :: OSI Approved :: Python Software Foundation License',
+        'License :: OSI Approved :: Qt Public License (QPL)',
+        'License :: OSI Approved :: Ricoh Source Code Public License',
+        'License :: OSI Approved :: Sleepycat License',
+        'License :: OSI Approved :: Sun Industry Standards Source License (SISSL)',
+        'License :: OSI Approved :: Sun Public License',
+        'License :: OSI Approved :: University of Illinois/NCSA Open Source License',
+        'License :: OSI Approved :: Vovida Software License 1.0',
+        'License :: OSI Approved :: W3C License',
+        'License :: OSI Approved :: X.Net License',
+        'License :: OSI Approved :: zlib/libpng License',
+        'License :: OSI Approved :: Zope Public License',
+        'License :: Other/Proprietary License',
+        'License :: Public Domain',
+        'License :: Repoze Public License',
+        'Natural Language :: Afrikaans',
+        'Natural Language :: Arabic',
+        'Natural Language :: Bengali',
+        'Natural Language :: Bosnian',
+        'Natural Language :: Bulgarian',
+        'Natural Language :: Catalan',
+        'Natural Language :: Chinese (Simplified)',
+        'Natural Language :: Chinese (Traditional)',
+        'Natural Language :: Croatian',
+        'Natural Language :: Czech',
+        'Natural Language :: Danish',
+        'Natural Language :: Dutch',
+        'Natural Language :: English',
+        'Natural Language :: Esperanto',
+        'Natural Language :: Finnish',
+        'Natural Language :: French',
+        'Natural Language :: German',
+        'Natural Language :: Greek',
+        'Natural Language :: Hebrew',
+        'Natural Language :: Hindi',
+        'Natural Language :: Hungarian',
+        'Natural Language :: Icelandic',
+        'Natural Language :: Indonesian',
+        'Natural Language :: Italian',
+        'Natural Language :: Japanese',
+        'Natural Language :: Javanese',
+        'Natural Language :: Korean',
+        'Natural Language :: Latin',
+        'Natural Language :: Latvian',
+        'Natural Language :: Macedonian',
+        'Natural Language :: Malay',
+        'Natural Language :: Marathi',
+        'Natural Language :: Norwegian',
+        'Natural Language :: Panjabi',
+        'Natural Language :: Persian',
+        'Natural Language :: Polish',
+        'Natural Language :: Portuguese',
+        'Natural Language :: Portuguese (Brazilian)',
+        'Natural Language :: Romanian',
+        'Natural Language :: Russian',
+        'Natural Language :: Serbian',
+        'Natural Language :: Slovak',
+        'Natural Language :: Slovenian',
+        'Natural Language :: Spanish',
+        'Natural Language :: Swedish',
+        'Natural Language :: Tamil',
+        'Natural Language :: Telugu',
+        'Natural Language :: Thai',
+        'Natural Language :: Turkish',
+        'Natural Language :: Ukranian',
+        'Natural Language :: Urdu',
+        'Natural Language :: Vietnamese',
+        'Operating System :: BeOS',
+        'Operating System :: MacOS',
+        'Operating System :: MacOS :: MacOS 9',
+        'Operating System :: MacOS :: MacOS X',
+        'Operating System :: Microsoft',
+        'Operating System :: Microsoft :: MS-DOS',
+        'Operating System :: Microsoft :: Windows',
+        'Operating System :: Microsoft :: Windows :: Windows 3.1 or Earlier',
+        'Operating System :: Microsoft :: Windows :: Windows 95/98/2000',
+        'Operating System :: Microsoft :: Windows :: Windows CE',
+        'Operating System :: Microsoft :: Windows :: Windows NT/2000',
+        'Operating System :: OS/2',
+        'Operating System :: OS Independent',
+        'Operating System :: Other OS',
+        'Operating System :: PalmOS',
+        'Operating System :: PDA Systems',
+        'Operating System :: POSIX',
+        'Operating System :: POSIX :: AIX',
+        'Operating System :: POSIX :: BSD',
+        'Operating System :: POSIX :: BSD :: BSD/OS',
+        'Operating System :: POSIX :: BSD :: FreeBSD',
+        'Operating System :: POSIX :: BSD :: NetBSD',
+        'Operating System :: POSIX :: BSD :: OpenBSD',
+        'Operating System :: POSIX :: GNU Hurd',
+        'Operating System :: POSIX :: HP-UX',
+        'Operating System :: POSIX :: IRIX',
+        'Operating System :: POSIX :: Linux',
+        'Operating System :: POSIX :: Other',
+        'Operating System :: POSIX :: SCO',
+        'Operating System :: POSIX :: SunOS/Solaris',
+        'Operating System :: Unix',
+        'Programming Language :: Ada',
+        'Programming Language :: APL',
+        'Programming Language :: ASP',
+        'Programming Language :: Assembly',
+        'Programming Language :: Awk',
+        'Programming Language :: Basic',
+        'Programming Language :: C',
+        'Programming Language :: C#',
+        'Programming Language :: C++',
+        'Programming Language :: Cold Fusion',
+        'Programming Language :: Cython',
+        'Programming Language :: Delphi/Kylix',
+        'Programming Language :: Dylan',
+        'Programming Language :: Eiffel',
+        'Programming Language :: Emacs-Lisp',
+        'Programming Language :: Erlang',
+        'Programming Language :: Euler',
+        'Programming Language :: Euphoria',
+        'Programming Language :: Forth',
+        'Programming Language :: Fortran',
+        'Programming Language :: Haskell',
+        'Programming Language :: Java',
+        'Programming Language :: JavaScript',
+        'Programming Language :: Lisp',
+        'Programming Language :: Logo',
+        'Programming Language :: ML',
+        'Programming Language :: Modula',
+        'Programming Language :: Objective C',
+        'Programming Language :: Object Pascal',
+        'Programming Language :: OCaml',
+        'Programming Language :: Other',
+        'Programming Language :: Other Scripting Engines',
+        'Programming Language :: Pascal',
+        'Programming Language :: Perl',
+        'Programming Language :: PHP',
+        'Programming Language :: Pike',
+        'Programming Language :: Pliant',
+        'Programming Language :: PL/SQL',
+        'Programming Language :: PROGRESS',
+        'Programming Language :: Prolog',
+        'Programming Language :: Python',
+        'Programming Language :: Python :: 2',
+        'Programming Language :: Python :: 2.3',
+        'Programming Language :: Python :: 2.4',
+        'Programming Language :: Python :: 2.5',
+        'Programming Language :: Python :: 2.6',
+        'Programming Language :: Python :: 2.7',
+        'Programming Language :: Python :: 3',
+        'Programming Language :: Python :: 3.0',
+        'Programming Language :: Python :: 3.1',
+        'Programming Language :: Python :: 3.2',
+        'Programming Language :: REBOL',
+        'Programming Language :: Rexx',
+        'Programming Language :: Ruby',
+        'Programming Language :: Scheme',
+        'Programming Language :: Simula',
+        'Programming Language :: Smalltalk',
+        'Programming Language :: SQL',
+        'Programming Language :: Tcl',
+        'Programming Language :: Unix Shell',
+        'Programming Language :: Visual Basic',
+        'Programming Language :: XBasic',
+        'Programming Language :: YACC',
+        'Programming Language :: Zope',
+        'Topic :: Adaptive Technologies',
+        'Topic :: Artistic Software',
+        'Topic :: Communications',
+        'Topic :: Communications :: BBS',
+        'Topic :: Communications :: Chat',
+        'Topic :: Communications :: Chat :: AOL Instant Messenger',
+        'Topic :: Communications :: Chat :: ICQ',
+        'Topic :: Communications :: Chat :: Internet Relay Chat',
+        'Topic :: Communications :: Chat :: Unix Talk',
+        'Topic :: Communications :: Conferencing',
+        'Topic :: Communications :: Email',
+        'Topic :: Communications :: Email :: Address Book',
+        'Topic :: Communications :: Email :: Email Clients (MUA)',
+        'Topic :: Communications :: Email :: Filters',
+        'Topic :: Communications :: Email :: Mailing List Servers',
+        'Topic :: Communications :: Email :: Mail Transport Agents',
+        'Topic :: Communications :: Email :: Post-Office',
+        'Topic :: Communications :: Email :: Post-Office :: IMAP',
+        'Topic :: Communications :: Email :: Post-Office :: POP3',
+        'Topic :: Communications :: Fax',
+        'Topic :: Communications :: FIDO',
+        'Topic :: Communications :: File Sharing',
+        'Topic :: Communications :: File Sharing :: Gnutella',
+        'Topic :: Communications :: File Sharing :: Napster',
+        'Topic :: Communications :: Ham Radio',
+        'Topic :: Communications :: Internet Phone',
+        'Topic :: Communications :: Telephony',
+        'Topic :: Communications :: Usenet News',
+        'Topic :: Database',
+        'Topic :: Database :: Database Engines/Servers',
+        'Topic :: Database :: Front-Ends',
+        'Topic :: Desktop Environment',
+        'Topic :: Desktop Environment :: File Managers',
+        'Topic :: Desktop Environment :: Gnome',
+        'Topic :: Desktop Environment :: GNUstep',
+        'Topic :: Desktop Environment :: K Desktop Environment (KDE)',
+        'Topic :: Desktop Environment :: K Desktop Environment (KDE) :: Themes',
+        'Topic :: Desktop Environment :: PicoGUI',
+        'Topic :: Desktop Environment :: PicoGUI :: Applications',
+        'Topic :: Desktop Environment :: PicoGUI :: Themes',
+        'Topic :: Desktop Environment :: Screen Savers',
+        'Topic :: Desktop Environment :: Window Managers',
+        'Topic :: Desktop Environment :: Window Managers :: Afterstep',
+        'Topic :: Desktop Environment :: Window Managers :: Afterstep :: Themes',
+        'Topic :: Desktop Environment :: Window Managers :: Applets',
+        'Topic :: Desktop Environment :: Window Managers :: Blackbox',
+        'Topic :: Desktop Environment :: Window Managers :: Blackbox :: Themes',
+        'Topic :: Desktop Environment :: Window Managers :: CTWM',
+        'Topic :: Desktop Environment :: Window Managers :: CTWM :: Themes',
+        'Topic :: Desktop Environment :: Window Managers :: Enlightenment',
+        'Topic :: Desktop Environment :: Window Managers :: Enlightenment :: Epplets',
+        'Topic :: Desktop Environment :: Window Managers :: Enlightenment :: Themes DR15',
+        'Topic :: Desktop Environment :: Window Managers :: Enlightenment :: Themes DR16',
+        'Topic :: Desktop Environment :: Window Managers :: Enlightenment :: Themes DR17',
+        'Topic :: Desktop Environment :: Window Managers :: Fluxbox',
+        'Topic :: Desktop Environment :: Window Managers :: Fluxbox :: Themes',
+        'Topic :: Desktop Environment :: Window Managers :: FVWM',
+        'Topic :: Desktop Environment :: Window Managers :: FVWM :: Themes',
+        'Topic :: Desktop Environment :: Window Managers :: IceWM',
+        'Topic :: Desktop Environment :: Window Managers :: IceWM :: Themes',
+        'Topic :: Desktop Environment :: Window Managers :: MetaCity',
+        'Topic :: Desktop Environment :: Window Managers :: MetaCity :: Themes',
+        'Topic :: Desktop Environment :: Window Managers :: Oroborus',
+        'Topic :: Desktop Environment :: Window Managers :: Oroborus :: Themes',
+        'Topic :: Desktop Environment :: Window Managers :: Sawfish',
+        'Topic :: Desktop Environment :: Window Managers :: Sawfish :: Themes 0.30',
+        'Topic :: Desktop Environment :: Window Managers :: Sawfish :: Themes pre-0.30',
+        'Topic :: Desktop Environment :: Window Managers :: Waimea',
+        'Topic :: Desktop Environment :: Window Managers :: Waimea :: Themes',
+        'Topic :: Desktop Environment :: Window Managers :: Window Maker',
+        'Topic :: Desktop Environment :: Window Managers :: Window Maker :: Applets',
+        'Topic :: Desktop Environment :: Window Managers :: Window Maker :: Themes',
+        'Topic :: Desktop Environment :: Window Managers :: XFCE',
+        'Topic :: Desktop Environment :: Window Managers :: XFCE :: Themes',
+        'Topic :: Documentation',
+        'Topic :: Education',
+        'Topic :: Education :: Computer Aided Instruction (CAI)',
+        'Topic :: Education :: Testing',
+        'Topic :: Games/Entertainment',
+        'Topic :: Games/Entertainment :: Arcade',
+        'Topic :: Games/Entertainment :: Board Games',
+        'Topic :: Games/Entertainment :: First Person Shooters',
+        'Topic :: Games/Entertainment :: Fortune Cookies',
+        'Topic :: Games/Entertainment :: Multi-User Dungeons (MUD)',
+        'Topic :: Games/Entertainment :: Puzzle Games',
+        'Topic :: Games/Entertainment :: Real Time Strategy',
+        'Topic :: Games/Entertainment :: Role-Playing',
+        'Topic :: Games/Entertainment :: Side-Scrolling/Arcade Games',
+        'Topic :: Games/Entertainment :: Simulation',
+        'Topic :: Games/Entertainment :: Turn Based Strategy',
+        'Topic :: Home Automation',
+        'Topic :: Internet',
+        'Topic :: Internet :: File Transfer Protocol (FTP)',
+        'Topic :: Internet :: Finger',
+        'Topic :: Internet :: Log Analysis',
+        'Topic :: Internet :: Name Service (DNS)',
+        'Topic :: Internet :: Proxy Servers',
+        'Topic :: Internet :: WAP',
+        'Topic :: Internet :: WWW/HTTP',
+        'Topic :: Internet :: WWW/HTTP :: Browsers',
+        'Topic :: Internet :: WWW/HTTP :: Dynamic Content',
+        'Topic :: Internet :: WWW/HTTP :: Dynamic Content :: CGI Tools/Libraries',
+        'Topic :: Internet :: WWW/HTTP :: Dynamic Content :: Message Boards',
+        'Topic :: Internet :: WWW/HTTP :: Dynamic Content :: News/Diary',
+        'Topic :: Internet :: WWW/HTTP :: Dynamic Content :: Page Counters',
+        'Topic :: Internet :: WWW/HTTP :: HTTP Servers',
+        'Topic :: Internet :: WWW/HTTP :: Indexing/Search',
+        'Topic :: Internet :: WWW/HTTP :: Site Management',
+        'Topic :: Internet :: WWW/HTTP :: Site Management :: Link Checking',
+        'Topic :: Internet :: WWW/HTTP :: WSGI',
+        'Topic :: Internet :: WWW/HTTP :: WSGI :: Application',
+        'Topic :: Internet :: WWW/HTTP :: WSGI :: Middleware',
+        'Topic :: Internet :: WWW/HTTP :: WSGI :: Server',
+        'Topic :: Internet :: Z39.50',
+        'Topic :: Multimedia',
+        'Topic :: Multimedia :: Graphics',
+        'Topic :: Multimedia :: Graphics :: 3D Modeling',
+        'Topic :: Multimedia :: Graphics :: 3D Rendering',
+        'Topic :: Multimedia :: Graphics :: Capture',
+        'Topic :: Multimedia :: Graphics :: Capture :: Digital Camera',
+        'Topic :: Multimedia :: Graphics :: Capture :: Scanners',
+        'Topic :: Multimedia :: Graphics :: Capture :: Screen Capture',
+        'Topic :: Multimedia :: Graphics :: Editors',
+        'Topic :: Multimedia :: Graphics :: Editors :: Raster-Based',
+        'Topic :: Multimedia :: Graphics :: Editors :: Vector-Based',
+        'Topic :: Multimedia :: Graphics :: Graphics Conversion',
+        'Topic :: Multimedia :: Graphics :: Presentation',
+        'Topic :: Multimedia :: Graphics :: Viewers',
+        'Topic :: Multimedia :: Sound/Audio',
+        'Topic :: Multimedia :: Sound/Audio :: Analysis',
+        'Topic :: Multimedia :: Sound/Audio :: Capture/Recording',
+        'Topic :: Multimedia :: Sound/Audio :: CD Audio',
+        'Topic :: Multimedia :: Sound/Audio :: CD Audio :: CD Playing',
+        'Topic :: Multimedia :: Sound/Audio :: CD Audio :: CD Ripping',
+        'Topic :: Multimedia :: Sound/Audio :: CD Audio :: CD Writing',
+        'Topic :: Multimedia :: Sound/Audio :: Conversion',
+        'Topic :: Multimedia :: Sound/Audio :: Editors',
+        'Topic :: Multimedia :: Sound/Audio :: MIDI',
+        'Topic :: Multimedia :: Sound/Audio :: Mixers',
+        'Topic :: Multimedia :: Sound/Audio :: Players',
+        'Topic :: Multimedia :: Sound/Audio :: Players :: MP3',
+        'Topic :: Multimedia :: Sound/Audio :: Sound Synthesis',
+        'Topic :: Multimedia :: Sound/Audio :: Speech',
+        'Topic :: Multimedia :: Video',
+        'Topic :: Multimedia :: Video :: Capture',
+        'Topic :: Multimedia :: Video :: Conversion',
+        'Topic :: Multimedia :: Video :: Display',
+        'Topic :: Multimedia :: Video :: Non-Linear Editor',
+        'Topic :: Office/Business',
+        'Topic :: Office/Business :: Financial',
+        'Topic :: Office/Business :: Financial :: Accounting',
+        'Topic :: Office/Business :: Financial :: Investment',
+        'Topic :: Office/Business :: Financial :: Point-Of-Sale',
+        'Topic :: Office/Business :: Financial :: Spreadsheet',
+        'Topic :: Office/Business :: Groupware',
+        'Topic :: Office/Business :: News/Diary',
+        'Topic :: Office/Business :: Office Suites',
+        'Topic :: Office/Business :: Scheduling',
+        'Topic :: Other/Nonlisted Topic',
+        'Topic :: Printing',
+        'Topic :: Religion',
+        'Topic :: Scientific/Engineering',
+        'Topic :: Scientific/Engineering :: Artificial Intelligence',
+        'Topic :: Scientific/Engineering :: Astronomy',
+        'Topic :: Scientific/Engineering :: Atmospheric Science',
+        'Topic :: Scientific/Engineering :: Bio-Informatics',
+        'Topic :: Scientific/Engineering :: Chemistry',
+        'Topic :: Scientific/Engineering :: Electronic Design Automation (EDA)',
+        'Topic :: Scientific/Engineering :: GIS',
+        'Topic :: Scientific/Engineering :: Human Machine Interfaces',
+        'Topic :: Scientific/Engineering :: Image Recognition',
+        'Topic :: Scientific/Engineering :: Information Analysis',
+        'Topic :: Scientific/Engineering :: Interface Engine/Protocol Translator',
+        'Topic :: Scientific/Engineering :: Mathematics',
+        'Topic :: Scientific/Engineering :: Medical Science Apps.',
+        'Topic :: Scientific/Engineering :: Physics',
+        'Topic :: Scientific/Engineering :: Visualization',
+        'Topic :: Security',
+        'Topic :: Security :: Cryptography',
+        'Topic :: Sociology',
+        'Topic :: Sociology :: Genealogy',
+        'Topic :: Sociology :: History',
+        'Topic :: Software Development',
+        'Topic :: Software Development :: Assemblers',
+        'Topic :: Software Development :: Bug Tracking',
+        'Topic :: Software Development :: Build Tools',
+        'Topic :: Software Development :: Code Generators',
+        'Topic :: Software Development :: Compilers',
+        'Topic :: Software Development :: Debuggers',
+        'Topic :: Software Development :: Disassemblers',
+        'Topic :: Software Development :: Documentation',
+        'Topic :: Software Development :: Embedded Systems',
+        'Topic :: Software Development :: Internationalization',
+        'Topic :: Software Development :: Interpreters',
+        'Topic :: Software Development :: Libraries',
+        'Topic :: Software Development :: Libraries :: Application Frameworks',
+        'Topic :: Software Development :: Libraries :: Java Libraries',
+        'Topic :: Software Development :: Libraries :: Perl Modules',
+        'Topic :: Software Development :: Libraries :: PHP Classes',
+        'Topic :: Software Development :: Libraries :: Pike Modules',
+        'Topic :: Software Development :: Libraries :: pygame',
+        'Topic :: Software Development :: Libraries :: Python Modules',
+        'Topic :: Software Development :: Libraries :: Ruby Modules',
+        'Topic :: Software Development :: Libraries :: Tcl Extensions',
+        'Topic :: Software Development :: Localization',
+        'Topic :: Software Development :: Object Brokering',
+        'Topic :: Software Development :: Object Brokering :: CORBA',
+        'Topic :: Software Development :: Pre-processors',
+        'Topic :: Software Development :: Quality Assurance',
+        'Topic :: Software Development :: Testing',
+        'Topic :: Software Development :: Testing :: Traffic Generation',
+        'Topic :: Software Development :: User Interfaces',
+        'Topic :: Software Development :: Version Control',
+        'Topic :: Software Development :: Version Control :: CVS',
+        'Topic :: Software Development :: Version Control :: RCS',
+        'Topic :: Software Development :: Version Control :: SCCS',
+        'Topic :: Software Development :: Widget Sets',
+        'Topic :: System',
+        'Topic :: System :: Archiving',
+        'Topic :: System :: Archiving :: Backup',
+        'Topic :: System :: Archiving :: Compression',
+        'Topic :: System :: Archiving :: Mirroring',
+        'Topic :: System :: Archiving :: Packaging',
+        'Topic :: System :: Benchmark',
+        'Topic :: System :: Boot',
+        'Topic :: System :: Boot :: Init',
+        'Topic :: System :: Clustering',
+        'Topic :: System :: Console Fonts',
+        'Topic :: System :: Distributed Computing',
+        'Topic :: System :: Emulators',
+        'Topic :: System :: Filesystems',
+        'Topic :: System :: Hardware',
+        'Topic :: System :: Hardware :: Hardware Drivers',
+        'Topic :: System :: Hardware :: Mainframes',
+        'Topic :: System :: Hardware :: Symmetric Multi-processing',
+        'Topic :: System :: Installation/Setup',
+        'Topic :: System :: Logging',
+        'Topic :: System :: Monitoring',
+        'Topic :: System :: Networking',
+        'Topic :: System :: Networking :: Firewalls',
+        'Topic :: System :: Networking :: Monitoring',
+        'Topic :: System :: Networking :: Monitoring :: Hardware Watchdog',
+        'Topic :: System :: Networking :: Time Synchronization',
+        'Topic :: System :: Operating System',
+        'Topic :: System :: Operating System Kernels',
+        'Topic :: System :: Operating System Kernels :: BSD',
+        'Topic :: System :: Operating System Kernels :: GNU Hurd',
+        'Topic :: System :: Operating System Kernels :: Linux',
+        'Topic :: System :: Power (UPS)',
+        'Topic :: System :: Recovery Tools',
+        'Topic :: System :: Shells',
+        'Topic :: System :: Software Distribution',
+        'Topic :: System :: Systems Administration',
+        'Topic :: System :: Systems Administration :: Authentication/Directory',
+        'Topic :: System :: Systems Administration :: Authentication/Directory :: LDAP',
+        'Topic :: System :: Systems Administration :: Authentication/Directory :: NIS',
+        'Topic :: System :: System Shells',
+        'Topic :: Terminals',
+        'Topic :: Terminals :: Serial',
+        'Topic :: Terminals :: Telnet',
+        'Topic :: Terminals :: Terminal Emulators/X Terminals',
+        'Topic :: Text Editors',
+        'Topic :: Text Editors :: Documentation',
+        'Topic :: Text Editors :: Emacs',
+        'Topic :: Text Editors :: Integrated Development Environments (IDE)',
+        'Topic :: Text Editors :: Text Processing',
+        'Topic :: Text Editors :: Word Processors',
+        'Topic :: Text Processing',
+        'Topic :: Text Processing :: Filters',
+        'Topic :: Text Processing :: Fonts',
+        'Topic :: Text Processing :: General',
+        'Topic :: Text Processing :: Indexing',
+        'Topic :: Text Processing :: Linguistic',
+        'Topic :: Text Processing :: Markup',
+        'Topic :: Text Processing :: Markup :: HTML',
+        'Topic :: Text Processing :: Markup :: LaTeX',
+        'Topic :: Text Processing :: Markup :: SGML',
+        'Topic :: Text Processing :: Markup :: VRML',
+        'Topic :: Text Processing :: Markup :: XML',
+        'Topic :: Utilities',
    ]
 
 def askYn(question, default = None, helptext = None):
-	while True:
-		answer = ask(question, default, helptext, required = True)
-		if answer and answer[0].lower() in 'yn':
-			return(answer[0].lower())
+    while True:
+        answer = ask(question, default, helptext, required = True)
+        if answer and answer[0].lower() in 'yn':
+            return(answer[0].lower())
 
-		print '\nERROR: You must select "Y" or "N".\n'
+        print '\nERROR: You must select "Y" or "N".\n'
 
 
 def ask(question, default = None, helptext = None, required = True,
         lengthy = False, multiline = False):
-	prompt = '%s: ' % ( question, )
-	if default:
-		prompt = '%s [%s]: ' % ( question, default )
-		if default and len(question) + len(default) > 70:
-			prompt = '%s\n    [%s]: ' % ( question, default )
-	if lengthy or multiline:
-		prompt += '\n   >'
+    prompt = '%s: ' % ( question, )
+    if default:
+        prompt = '%s [%s]: ' % ( question, default )
+        if default and len(question) + len(default) > 70:
+            prompt = '%s\n    [%s]: ' % ( question, default )
+    if lengthy or multiline:
+        prompt += '\n   >'
 
-	if not helptext: helptext = 'No additional help available.'
-	if helptext[0] == '\n': helptext = helptext[1:]
-	if helptext[-1] == '\n': helptext = helptext[:-1]
+    if not helptext: helptext = 'No additional help available.'
+    if helptext[0] == '\n': helptext = helptext[1:]
+    if helptext[-1] == '\n': helptext = helptext[:-1]
 
-	while True:
-		sys.stdout.write(prompt)
-		sys.stdout.flush()
+    while True:
+        sys.stdout.write(prompt)
+        sys.stdout.flush()
 
-		line = sys.stdin.readline().strip()
-		if line == '?':
-			print '=' * 70
-			print helptext
-			print '=' * 70
-			continue
-		if default and not line: return(default)
-		if not line and required:
-			print '*' * 70
-			print 'This value cannot be empty.'
-			print '==========================='
-			if helptext: print helptext
-			print '*' * 70
-			continue
-		return(line)
+        line = sys.stdin.readline().strip()
+        if line == '?':
+            print '=' * 70
+            print helptext
+            print '=' * 70
+            continue
+        if default and not line: return(default)
+        if not line and required:
+            print '*' * 70
+            print 'This value cannot be empty.'
+            print '==========================='
+            if helptext: print helptext
+            print '*' * 70
+            continue
+        return(line)
 
 
 def buildTroveDict(troveList):
-	dict = {}
-	for key in troveList:
-		subDict = dict
-		for subkey in key.split(' :: '):
-			if not subkey in subDict: subDict[subkey] = {}
-			subDict = subDict[subkey]
-	return(dict)
+    dict = {}
+    for key in troveList:
+        subDict = dict
+        for subkey in key.split(' :: '):
+            if not subkey in subDict: subDict[subkey] = {}
+            subDict = subDict[subkey]
+    return(dict)
 troveDict = buildTroveDict(troveList)
 
 
 class SetupClass:
-	def __init__(self):
-		self.config = None
-		self.classifierDict = {}
-		self.setupData = {}
-		self.setupData['classifier'] = self.classifierDict
-		self.setupData['packages'] = {}
+    def __init__(self):
+        self.config = None
+        self.classifierDict = {}
+        self.setupData = {}
+        self.setupData['classifier'] = self.classifierDict
+        self.setupData['packages'] = {}
 
-		self.loadConfigFile()
+        self.loadConfigFile()
 
 
-	def lookupOption(self, key):
-		if not self.config.has_option('DEFAULT', key): return(None)
-		return(self.config.get('DEFAULT', key))
+    def lookupOption(self, key):
+        if not self.config.has_option('DEFAULT', key): return(None)
+        return(self.config.get('DEFAULT', key))
 
 
-	def loadConfigFile(self):
-		self.config = ConfigParser.RawConfigParser()
-		self.config.read(os.path.expanduser('~/.mkpkgpy'))
-		self.setupData['author'] = self.lookupOption('author')
-		self.setupData['author_email'] = self.lookupOption('author_email')
+    def loadConfigFile(self):
+        self.config = ConfigParser.RawConfigParser()
+        self.config.read(os.path.expanduser('~/.mkpkgpy'))
+        self.setupData['author'] = self.lookupOption('author')
+        self.setupData['author_email'] = self.lookupOption('author_email')
 
 
-	def updateConfigFile(self):
-		valuesDifferent = False
-		for compareKey in ('author', 'author_email'):
-			if self.lookupOption(compareKey) != self.setupData[compareKey]:
-				valuesDifferent = True
-				self.config.set('DEFAULT', compareKey, self.setupData[compareKey])
+    def updateConfigFile(self):
+        valuesDifferent = False
+        for compareKey in ('author', 'author_email'):
+            if self.lookupOption(compareKey) != self.setupData[compareKey]:
+                valuesDifferent = True
+                self.config.set('DEFAULT', compareKey,
+                    self.setupData[compareKey])
 
-		if not valuesDifferent: return
+        if not valuesDifferent: return
 
-		self.config.write(open(os.path.expanduser('~/.pygiver'), 'w'))
+        self.config.write(open(os.path.expanduser('~/.pygiver'), 'w'))
 
 
-	def loadExistingSetup(self):
-		raise NotImplementedError
+    def loadExistingSetup(self):
+        raise NotImplementedError
 
 
-	def inspectFile(self, path):
-		fp = open(path, 'r')
-		for line in [ fp.readline() for x in range(10) ]:
-			m = re.match(r'^#!.*python((?P<major>\d)(\.\d+)?)?$', line)
-			if m:
-				if m.group('major') == '3':
-					self.classifierDict['Programming Language :: Python :: 3'] = 1
-				else:
-					self.classifierDict['Programming Language :: Python :: 2'] = 1
-		fp.close()
+    def inspectFile(self, path):
+        fp = open(path, 'r')
+        for line in [ fp.readline() for x in range(10) ]:
+            m = re.match(r'^#!.*python((?P<major>\d)(\.\d+)?)?$', line)
+            if m:
+                if m.group('major') == '3':
+                    self.classifierDict['Programming Language :: Python :: 3'] = 1
+                else:
+                    self.classifierDict['Programming Language :: Python :: 2'] = 1
+        fp.close()
 
 
-	def inspectDirectory(self):
-		dirName = os.path.basename(os.getcwd())
-		self.setupData['name'] = dirName
-		m = re.match(r'(.*)-(\d.+)', dirName)
-		if m:
-			self.setupData['name'] = m.group(1)
-			self.setupData['version'] = m.group(2)
+    def inspectDirectory(self):
+        dirName = os.path.basename(os.getcwd())
+        self.setupData['name'] = dirName
+        m = re.match(r'(.*)-(\d.+)', dirName)
+        if m:
+            self.setupData['name'] = m.group(1)
+            self.setupData['version'] = m.group(2)
 
-		for root, dirs, files in os.walk('.'):
-			for file in files:
-				if root == '.' and file == 'setup.py': continue
-				fileName = os.path.join(root, file)
-				self.inspectFile(fileName)
+        for root, dirs, files in os.walk('.'):
+            for file in files:
+                if root == '.' and file == 'setup.py': continue
+                fileName = os.path.join(root, file)
+                self.inspectFile(fileName)
 
-				if file == '__init__.py':
-					trySrc = os.path.join('.', 'src')
-					tmpRoot = root
-					if tmpRoot.startswith(trySrc):
-						tmpRoot = tmpRoot[len(trySrc):]
-					if tmpRoot.startswith(os.path.sep):
-						tmpRoot = tmpRoot[len(os.path.sep):]
+                if file == '__init__.py':
+                    trySrc = os.path.join('.', 'src')
+                    tmpRoot = root
+                    if tmpRoot.startswith(trySrc):
+                        tmpRoot = tmpRoot[len(trySrc):]
+                    if tmpRoot.startswith(os.path.sep):
+                        tmpRoot = tmpRoot[len(os.path.sep):]
 
-					self.setupData['packages'][tmpRoot] = root[1 + len(os.path.sep):]
+                    self.setupData['packages'][tmpRoot] = root[1 + len(os.path.sep):]
 
 
-	def queryUser(self):
-		self.setupData['name'] = ask('Package name', self.setupData['name'],
-		      helpText['name'])
-		self.setupData['version'] = ask('Current version number',
-		      self.setupData.get('version'), helpText['version'])
-		self.setupData['description'] = ask('Package description',
-		      self.setupData.get('description'), helpText['description'],
-		      lengthy = True)
-		self.setupData['author'] = ask('Author name',
-		      self.setupData.get('author'), helpText['author'])
-		self.setupData['author_email'] = ask('Author e-mail address',
-		      self.setupData.get('author_email'), helpText['author_email'])
-		self.setupData['url'] = ask('Project URL',
-		      self.setupData.get('url'), helpText['url'], required = False)
+    def queryUser(self):
+        self.setupData['name'] = ask('Package name', self.setupData['name'],
+              helpText['name'])
+        self.setupData['version'] = ask('Current version number',
+              self.setupData.get('version'), helpText['version'])
+        self.setupData['description'] = ask('Package description',
+              self.setupData.get('description'), helpText['description'],
+              lengthy = True)
+        self.setupData['author'] = ask('Author name',
+              self.setupData.get('author'), helpText['author'])
+        self.setupData['author_email'] = ask('Author e-mail address',
+              self.setupData.get('author_email'), helpText['author_email'])
+        self.setupData['url'] = ask('Project URL',
+              self.setupData.get('url'), helpText['url'], required = False)
 
-		if (askYn('Do you want to set Trove classifiers?',
-				helptext = helpText['do_classifier']) == 'y'):
-			self.setTroveClassifier()
+        if (askYn('Do you want to set Trove classifiers?',
+                helptext = helpText['do_classifier']) == 'y'):
+            self.setTroveClassifier()
 
 
-	def setTroveClassifier(self):
-		self.setTroveDevStatus(self.classifierDict)
-		self.setTroveLicense(self.classifierDict)
-		self.setTroveOther(self.classifierDict)
+    def setTroveClassifier(self):
+        self.setTroveDevStatus(self.classifierDict)
+        self.setTroveLicense(self.classifierDict)
+        self.setTroveOther(self.classifierDict)
 
 
-	def setTroveOther(self, classifierDict):
-		if askYn('Do you want to set other trove identifiers', 'n',
-				helpText['trove_generic']) != 'y': return
+    def setTroveOther(self, classifierDict):
+        if askYn('Do you want to set other trove identifiers', 'n',
+                helpText['trove_generic']) != 'y': return
 
-		self.walkTrove(classifierDict, [troveDict], '')
-	
-	def walkTrove(self, classifierDict, trovePath, desc):
-		trove = trovePath[-1]
+        self.walkTrove(classifierDict, [troveDict], '')
 
-		if not trove:
-			return
+    def walkTrove(self, classifierDict, trovePath, desc):
+        trove = trovePath[-1]
 
-		for key in sorted(trove.keys()):
-			if len(trove[key]) == 0:
-				if askYn('Add "%s"' % desc[4:] + ' :: ' + key, 'n') == 'y':
-					classifierDict[desc[4:] + ' :: ' + key] = 1
-				continue
+        if not trove:
+            return
 
-			if askYn('Do you want to set items under\n   "%s" (%d sub-items)'
-					% ( key, len(trove[key]) ), 'n',
-					helpText['trove_generic']) == 'y':
-				self.walkTrove(classifierDict, trovePath + [trove[key]],
-						desc + ' :: ' + key)
+        for key in sorted(trove.keys()):
+            if len(trove[key]) == 0:
+                if askYn('Add "%s"' % desc[4:] + ' :: ' + key, 'n') == 'y':
+                    classifierDict[desc[4:] + ' :: ' + key] = 1
+                continue
 
+            if askYn('Do you want to set items under\n   "%s" (%d sub-items)'
+                    % ( key, len(trove[key]) ), 'n',
+                    helpText['trove_generic']) == 'y':
+                self.walkTrove(classifierDict, trovePath + [trove[key]],
+                        desc + ' :: ' + key)
 
-	def setTroveLicense(self, classifierDict):
-		while True:
-			license = ask('What license do you use',
-			      helptext = helpText['trove_license'], required = False)
-			if not license: return
 
-			licenseWords = license.lower().split(' ')
+    def setTroveLicense(self, classifierDict):
+        while True:
+            license = ask('What license do you use',
+                  helptext = helpText['trove_license'], required = False)
+            if not license: return
 
-			foundList = []
-			for index in range(len(troveList)):
-				troveItem = troveList[index]
-				if not troveItem.startswith('License :: '): continue
-				troveItem = troveItem[11:].lower()
+            licenseWords = license.lower().split(' ')
 
-				allMatch = True
-				for word in licenseWords:
-					if not word in troveItem:
-						allMatch = False
-						break
-				if allMatch: foundList.append(index)
+            foundList = []
+            for index in range(len(troveList)):
+                troveItem = troveList[index]
+                if not troveItem.startswith('License :: '): continue
+                troveItem = troveItem[11:].lower()
 
-			question = 'Matching licenses:\n\n'
-			for i in xrange(1, len(foundList) + 1):
-				question += '   %s) %s\n' % ( i, troveList[foundList[i - 1]] )
-			question += ('\nType the number of the license you wish to use or '
-			         '? to try again:')
-			troveLicense = ask(question, required = False)
+                allMatch = True
+                for word in licenseWords:
+                    if not word in troveItem:
+                        allMatch = False
+                        break
+                if allMatch: foundList.append(index)
 
-			if troveLicense == '?': continue
-			if troveLicense == '': return
-			foundIndex = foundList[int(troveLicense) - 1]
-			classifierDict[troveList[foundIndex]] = 1
-			try:
-				return
-			except IndexError:
-				print 'ERROR: Invalid selection, type a number from the list above.'
+            question = 'Matching licenses:\n\n'
+            for i in xrange(1, len(foundList) + 1):
+                question += '   %s) %s\n' % ( i, troveList[foundList[i - 1]] )
+            question += ('\nType the number of the license you wish to use or '
+                     '? to try again:')
+            troveLicense = ask(question, required = False)
 
+            if troveLicense == '?': continue
+            if troveLicense == '': return
+            foundIndex = foundList[int(troveLicense) - 1]
+            classifierDict[troveList[foundIndex]] = 1
+            try:
+                return
+            except IndexError:
+                print("ERROR: Invalid selection, type a number from the list "
+                    "above.")
 
-	def setTroveDevStatus(self, classifierDict):
-		while True:
-			devStatus = ask('''Please select the project status:
+
+    def setTroveDevStatus(self, classifierDict):
+        while True:
+            devStatus = ask('''Please select the project status:
 
 1 - Planning
 2 - Pre-Alpha
@@ -851,71 +853,80 @@
 7 - Inactive
 
 Status''', required = False)
-			if devStatus:
-				try:
-					key = {
-						   '1' : 'Development Status :: 1 - Planning',
-						   '2' : 'Development Status :: 2 - Pre-Alpha',
-						   '3' : 'Development Status :: 3 - Alpha',
-						   '4' : 'Development Status :: 4 - Beta',
-						   '5' : 'Development Status :: 5 - Production/Stable',
-						   '6' : 'Development Status :: 6 - Mature',
-						   '7' : 'Development Status :: 7 - Inactive',
-						   }[devStatus]
-					classifierDict[key] = 1
-					return
-				except KeyError:
-					print 'ERROR: Invalid selection, type a single digit number.'
+            if devStatus:
+                try:
+                    key = {
+                           '1' : 'Development Status :: 1 - Planning',
+                           '2' : 'Development Status :: 2 - Pre-Alpha',
+                           '3' : 'Development Status :: 3 - Alpha',
+                           '4' : 'Development Status :: 4 - Beta',
+                           '5' : 'Development Status :: 5 - Production/Stable',
+                           '6' : 'Development Status :: 6 - Mature',
+                           '7' : 'Development Status :: 7 - Inactive',
+                           }[devStatus]
+                    classifierDict[key] = 1
+                    return
+                except KeyError:
+                    print("ERROR: Invalid selection, type a single digit "
+                        "number.")
 
+    def _dotted_packages(self, data):
+        packages = sorted(data.keys())
+        modified_pkgs = []
+        for pkg in packages:
+            pkg = pkg.lstrip('./')
+            pkg = pkg.replace('/', '.')
+            modified_pkgs.append(pkg)
+        return modified_pkgs
 
-	def writeSetup(self):
-		if os.path.exists('setup.py'): shutil.move('setup.py', 'setup.py.old')
+    def writeSetup(self):
+        if os.path.exists('setup.py'): shutil.move('setup.py', 'setup.py.old')
 
-		fp = open('setup.py', 'w')
-		fp.write('#!/usr/bin/env python\n\n')
-		fp.write('from distutils2.core import setup\n\n')
+        fp = open('setup.py', 'w')
+        fp.write('#!/usr/bin/env python\n\n')
+        fp.write('from distutils2.core import setup\n\n')
 
-		fp.write('from sys import version\n')
-		fp.write('if version < \'2.2.3\':\n')
-		fp.write('    from distutils2.dist import DistributionMetadata\n')
-		fp.write('    DistributionMetadata.classifier = None\n')
-		fp.write('    DistributionMetadata.download_url = None\n')
+        fp.write('from sys import version\n')
+        fp.write('if version < \'2.2.3\':\n')
+        fp.write('    from distutils2.dist import DistributionMetadata\n')
+        fp.write('    DistributionMetadata.classifier = None\n')
+        fp.write('    DistributionMetadata.download_url = None\n')
 
-		fp.write('setup(name = %s,\n' % repr(self.setupData['name']))
-		fp.write('        version = %s,\n' % repr(self.setupData['version']))
-		fp.write('        description = %s,\n'
-				% repr(self.setupData['description']))
-		fp.write('        author = %s,\n' % repr(self.setupData['author']))
-		fp.write('        author_email = %s,\n'
-				% repr(self.setupData['author_email']))
-		if self.setupData['url']:
-			fp.write('        url = %s,\n' % repr(self.setupData['url']))
-		if self.setupData['classifier']:
-			fp.write('        classifier = [\n')
-			for classifier in sorted(self.setupData['classifier'].keys()):
-				fp.write('              %s,\n' % repr(classifier))
-			fp.write('           ],\n')
-		if self.setupData['packages']:
-			fp.write('        packages = %s,\n'
-					% repr(sorted(self.setupData['packages'].keys())))
-			fp.write('        package_dir = %s,\n'
-					% repr(self.setupData['packages']))
-		fp.write('        #scripts = [\'path/to/script\']\n')
+        fp.write('setup(name = %s,\n' % repr(self.setupData['name']))
+        fp.write('        version = %s,\n' % repr(self.setupData['version']))
+        fp.write('        description = %s,\n'
+                % repr(self.setupData['description']))
+        fp.write('        author = %s,\n' % repr(self.setupData['author']))
+        fp.write('        author_email = %s,\n'
+                % repr(self.setupData['author_email']))
+        if self.setupData['url']:
+            fp.write('        url = %s,\n' % repr(self.setupData['url']))
+        if self.setupData['classifier']:
+            fp.write('        classifier = [\n')
+            for classifier in sorted(self.setupData['classifier'].keys()):
+                fp.write('              %s,\n' % repr(classifier))
+            fp.write('           ],\n')
+        if self.setupData['packages']:
+            fp.write('        packages = %s,\n'
+                    % repr(self._dotted_packages(self.setupData['packages'])))
+            fp.write('        package_dir = %s,\n'
+                    % repr(self.setupData['packages']))
+        fp.write('        #scripts = [\'path/to/script\']\n')
 
-		fp.write('        )\n')
-		fp.close()
-		os.chmod('setup.py', 0755)
+        fp.write('        )\n')
+        fp.close()
+        os.chmod('setup.py', 0755)
 
-		print 'Wrote "setup.py".'
+        print 'Wrote "setup.py".'
 
 
 def main():
-	setup = SetupClass()
-	setup.inspectDirectory()
-	setup.queryUser()
-	setup.updateConfigFile()
-	setup.writeSetup()
+    setup = SetupClass()
+    setup.inspectDirectory()
+    setup.queryUser()
+    setup.updateConfigFile()
+    setup.writeSetup()
 
 
 if __name__ == '__main__':
-	main()
+    main()
diff --git a/src/distutils2/tests/test_check.py b/src/distutils2/tests/test_check.py
--- a/src/distutils2/tests/test_check.py
+++ b/src/distutils2/tests/test_check.py
@@ -27,7 +27,7 @@
         # by default, check is checking the metadata
         # should have some warnings
         cmd = self._run()
-        self.assert_(cmd._warnings > 0)
+        self.assert_(len(cmd._warnings) > 0)
 
         # now let's add the required fields
         # and run it again, to make sure we don't get
@@ -37,7 +37,7 @@
                     'name': 'xxx', 'version': 'xxx'
                     }
         cmd = self._run(metadata)
-        self.assertEquals(cmd._warnings, 0)
+        self.assertEquals(len(cmd._warnings), 0)
 
         # now with the strict mode, we should
         # get an error if there are missing metadata
@@ -45,34 +45,35 @@
 
         # and of course, no error when all metadata are present
         cmd = self._run(metadata, strict=1)
-        self.assertEquals(cmd._warnings, 0)
+        self.assertEquals(len(cmd._warnings), 0)
 
     def test_check_restructuredtext(self):
         if not _HAS_DOCUTILS: # won't test without docutils
             return
         # let's see if it detects broken rest in long_description
         broken_rest = 'title\n===\n\ntest'
-        pkg_info, dist = self.create_dist(long_description=broken_rest)
+        pkg_info, dist = self.create_dist(description=broken_rest)
         cmd = check(dist)
         cmd.check_restructuredtext()
-        self.assertEquals(cmd._warnings, 1)
+        self.assertEquals(len(cmd._warnings), 1)
 
         # let's see if we have an error with strict=1
-        metadata = {'url': 'xxx', 'author': 'xxx',
+        metadata = {'home_page': 'xxx', 'author': 'xxx',
                     'author_email': 'xxx',
                     'name': 'xxx', 'version': 'xxx',
-                    'long_description': broken_rest}
+                    'description': broken_rest}
+
         self.assertRaises(DistutilsSetupError, self._run, metadata,
                           **{'strict': 1, 'restructuredtext': 1})
 
         # and non-broken rest
-        metadata['long_description'] = 'title\n=====\n\ntest'
+        metadata['description'] = 'title\n=====\n\ntest'
         cmd = self._run(metadata, strict=1, restructuredtext=1)
-        self.assertEquals(cmd._warnings, 0)
+        self.assertEquals(len(cmd._warnings), 0)
 
     def test_check_all(self):
 
-        metadata = {'url': 'xxx', 'author': 'xxx'}
+        metadata = {'home_page': 'xxx', 'author': 'xxx'}
         self.assertRaises(DistutilsSetupError, self._run,
                           {}, **{'strict': 1,
                                  'restructuredtext': 1})
diff --git a/src/distutils2/tests/test_metadata.py b/src/distutils2/tests/test_metadata.py
--- a/src/distutils2/tests/test_metadata.py
+++ b/src/distutils2/tests/test_metadata.py
@@ -185,6 +185,22 @@
 
         self.assertEquals(res, 0)
 
+    def test_project_url(self):
+        metadata = DistributionMetadata()
+        metadata['Project-URL'] = [('one', 'http://ok')]
+        self.assertEquals(metadata['Project-URL'],
+                          [('one', 'http://ok')])
+        self.assertEquals(metadata.version, '1.2')
+
+    def test_check(self):
+        metadata = DistributionMetadata()
+        metadata['Version'] = 'rr'
+        metadata['Requires-dist'] = ['Foo (a)']
+        missing, warnings = metadata.check()
+        self.assertEquals(missing, ['Name', 'Home-page'])
+        self.assertEquals(len(warnings), 2)
+
+
 def test_suite():
     return unittest2.makeSuite(DistributionMetadataTestCase)
 
diff --git a/src/distutils2/tests/test_register.py b/src/distutils2/tests/test_register.py
--- a/src/distutils2/tests/test_register.py
+++ b/src/distutils2/tests/test_register.py
@@ -7,6 +7,12 @@
 import urllib2
 import warnings
 
+try:
+    import docutils
+    DOCUTILS_SUPPORT = True
+except ImportError:
+    DOCUTILS_SUPPORT = False
+
 from distutils2.command import register as register_module
 from distutils2.command.register import register
 from distutils2.core import Distribution
@@ -186,6 +192,7 @@
         self.assertEquals(headers['Content-length'], '290')
         self.assertTrue('tarek' in req.data)
 
+    @unittest2.skipUnless(DOCUTILS_SUPPORT, 'needs docutils')
     def test_strict(self):
         # testing the script option
         # when on, the register command stops if
@@ -198,26 +205,20 @@
         cmd.strict = 1
         self.assertRaises(DistutilsSetupError, cmd.run)
 
-        # we don't test the reSt feature if docutils
-        # is not installed
-        try:
-            import docutils
-        except ImportError:
-            return
-
         # metadata are OK but long_description is broken
-        metadata = {'url': 'xxx', 'author': 'xxx',
+        metadata = {'home_page': 'xxx', 'author': 'xxx',
                     'author_email': u'??x??x??',
                     'name': 'xxx', 'version': 'xxx',
-                    'long_description': 'title\n==\n\ntext'}
+                    'description': 'title\n==\n\ntext'}
 
         cmd = self._get_cmd(metadata)
         cmd.ensure_finalized()
         cmd.strict = 1
+
         self.assertRaises(DistutilsSetupError, cmd.run)
 
         # now something that works
-        metadata['long_description'] = 'title\n=====\n\ntext'
+        metadata['description'] = 'title\n=====\n\ntext'
         cmd = self._get_cmd(metadata)
         cmd.ensure_finalized()
         cmd.strict = 1
diff --git a/src/distutils2/tests/test_util.py b/src/distutils2/tests/test_util.py
--- a/src/distutils2/tests/test_util.py
+++ b/src/distutils2/tests/test_util.py
@@ -6,7 +6,10 @@
 from StringIO import StringIO
 import subprocess
 
-from distutils2.errors import DistutilsPlatformError, DistutilsByteCompileError
+from distutils2.errors import (DistutilsPlatformError,
+                               DistutilsByteCompileError,
+                               DistutilsFileError)
+
 from distutils2.util import (convert_path, change_root,
                             check_environ, split_quoted, strtobool,
                             rfc822_escape, get_compiler_versions,
@@ -250,6 +253,10 @@
         finally:
             sys.dont_write_bytecode = old_dont_write_bytecode
 
+    def test_newer(self):
+        self.assertRaises(DistutilsFileError, util.newer, 'xxx', 'xxx')
+
+
 def test_suite():
     return unittest2.makeSuite(UtilTestCase)
 
diff --git a/src/distutils2/util.py b/src/distutils2/util.py
--- a/src/distutils2/util.py
+++ b/src/distutils2/util.py
@@ -8,11 +8,10 @@
 
 import sys, os, string, re
 
-from distutils2.errors import DistutilsPlatformError
+from distutils2.errors import (DistutilsPlatformError, DistutilsFileError,
+                               DistutilsByteCompileError)
 from distutils2.spawn import spawn, find_executable
 from distutils2 import log
-from distutils2.errors import DistutilsByteCompileError
-
 from distutils2._backport import sysconfig as _sysconfig
 
 _PLATFORM = None
diff --git a/src/setup.py b/src/setup.py
--- a/src/setup.py
+++ b/src/setup.py
@@ -65,7 +65,7 @@
 
 setup_kwargs = {}
 if sys.version < '2.6':
-    kwargs['scripts'] = 'distutils2/mkpkg.py'
+    setup_kwargs['scripts'] = ['distutils2/mkpkg.py']
 
 setup (name="Distutils2",
        version=VERSION,

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:38 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:38 +0200
Subject: [Python-checkins] distutils2: updated docs
Message-ID: <E1OVLoU-0000gh-RV@dinsdale.python.org>

tarek.ziade pushed d7dc7573e8ef to distutils2:

http://hg.python.org/distutils2/rev/d7dc7573e8ef
changeset:   267:d7dc7573e8ef
user:        Konrad Delong <konryd at gmail.com>
date:        Thu May 13 20:06:08 2010 +0200
summary:     updated docs
files:       docs/source/new_commands.rst

diff --git a/docs/source/new_commands.rst b/docs/source/new_commands.rst
--- a/docs/source/new_commands.rst
+++ b/docs/source/new_commands.rst
@@ -51,7 +51,8 @@
 The ``upload_docs`` command has the following options:
 
 ``--upload-dir``
-    The directory to be uploaded to the repository.
+    The directory to be uploaded to the repository. The default value is
+    ``docs`` in project root.
 
 ``--show-response``
     Display the full response text from server; this is useful for debugging

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:38 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:38 +0200
Subject: [Python-checkins] distutils2: fixed imports
Message-ID: <E1OVLoU-0000gV-Px@dinsdale.python.org>

tarek.ziade pushed 642169797dfc to distutils2:

http://hg.python.org/distutils2/rev/642169797dfc
changeset:   264:642169797dfc
user:        Konrad Delong <konryd at gmail.com>
date:        Thu May 13 14:10:25 2010 +0200
summary:     fixed imports
files:       src/distutils2/tests/pypi_server.py, src/distutils2/tests/test_upload.py

diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
--- a/src/distutils2/tests/pypi_server.py
+++ b/src/distutils2/tests/pypi_server.py
@@ -1,13 +1,5 @@
-import threading
-import copy
-import Queue
-try:
-    from urllib.request import urlopen
-except ImportError:
-    from urllib2 import urlopen
-
-import time
-from wsgiref.simple_server import make_server, demo_app
+import Queue, threading, time
+from wsgiref.simple_server import make_server
 
 class PyPIServer(threading.Thread):
     """Thread that wraps a wsgi app"""
diff --git a/src/distutils2/tests/test_upload.py b/src/distutils2/tests/test_upload.py
--- a/src/distutils2/tests/test_upload.py
+++ b/src/distutils2/tests/test_upload.py
@@ -1,16 +1,12 @@
 """Tests for distutils.command.upload."""
 # -*- encoding: utf8 -*-
-import sys
-import os
-import unittest2
+import os, unittest2
 
-from distutils2.command import upload as upload_mod
 from distutils2.command.upload import upload
 from distutils2.core import Distribution
 
-from distutils2.tests import support
+from distutils2.tests.pypi_server import PyPIServer
 from distutils2.tests.test_config import PYPIRC, PyPIRCCommandTestCase
-from distutils2.tests.pypi_server import PyPIServer
 
 
 PYPIRC_NOPASSWORD = """\

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:38 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:38 +0200
Subject: [Python-checkins] distutils2: moved the upload_docs command
Message-ID: <E1OVLoU-0000gD-Nv@dinsdale.python.org>

tarek.ziade pushed fa9aa6f78f8f to distutils2:

http://hg.python.org/distutils2/rev/fa9aa6f78f8f
changeset:   260:fa9aa6f78f8f
user:        Konrad Delong <konryd at gmail.com>
date:        Sun May 09 13:37:11 2010 -0400
summary:     moved the upload_docs command
files:       src/distutils2/command/upload_docs.py, src/distutils2/tests/test_upload_docs.py

diff --git a/src/distutils2/command/upload_docs.py b/src/distutils2/command/upload_docs.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/command/upload_docs.py
@@ -0,0 +1,174 @@
+# -*- coding: utf-8 -*-
+"""upload_docs
+
+Implements a Distutils2 'upload_docs' subcommand (upload documentation to
+PyPI's packages.python.org).
+"""
+
+import os
+import socket
+import zipfile
+import httplib
+import base64
+import urlparse
+import tempfile
+import sys
+
+from distutils2 import log
+from distutils2.errors import DistutilsOptionError
+from distutils2.command.upload import upload
+
+_IS_PYTHON3 = sys.version > '3'
+
+try:
+    bytes
+except NameError:
+    bytes = str
+
+def b(str_or_bytes):
+    """Return bytes by either encoding the argument as ASCII or simply return
+    the argument as-is."""
+    if not isinstance(str_or_bytes, bytes):
+        return str_or_bytes.encode('ascii')
+    else:
+        return str_or_bytes
+
+
+class upload_docs(upload):
+
+    description = 'Upload documentation to PyPI'
+
+    user_options = [
+        ('repository=', 'r',
+         "url of repository [default: %s]" % upload.DEFAULT_REPOSITORY),
+        ('show-response', None,
+         'display full response text from server'),
+        ('upload-dir=', None, 'directory to upload'),
+        ]
+    boolean_options = upload.boolean_options
+
+    def initialize_options(self):
+        upload.initialize_options(self)
+        self.upload_dir = None
+
+    def finalize_options(self):
+        upload.finalize_options(self)
+        if self.upload_dir is None:
+            build = self.get_finalized_command('build')
+            self.upload_dir = os.path.join(build.build_base, 'docs')
+            self.mkpath(self.upload_dir)
+        self.ensure_dirname('upload_dir')
+        self.announce('Using upload directory %s' % self.upload_dir)
+
+    def create_zipfile(self):
+        name = self.distribution.metadata['Name']
+        tmp_dir = tempfile.mkdtemp()
+        tmp_file = os.path.join(tmp_dir, "%s.zip" % name)
+        zip_file = zipfile.ZipFile(tmp_file, "w")
+        for root, dirs, files in os.walk(self.upload_dir):
+            if root == self.upload_dir and not files:
+                raise DistutilsOptionError(
+                    "no files found in upload directory '%s'"
+                    % self.upload_dir)
+            for name in files:
+                full = os.path.join(root, name)
+                relative = root[len(self.upload_dir):].lstrip(os.path.sep)
+                dest = os.path.join(relative, name)
+                zip_file.write(full, dest)
+        zip_file.close()
+        return tmp_file
+
+    def run(self):
+        zip_file = self.create_zipfile()
+        self.upload_file(zip_file)
+
+    def upload_file(self, filename):
+        content = open(filename, 'rb').read()
+        meta = self.distribution.metadata
+        data = {
+            ':action': 'doc_upload',
+            'name': meta['Name'],
+            'content': (os.path.basename(filename), content),
+        }
+        # set up the authentication
+        credentials = self.username + ':' + self.password
+        if _IS_PYTHON3:  # base64 only works with bytes in Python 3.
+            encoded_creds = base64.encodebytes(credentials.encode('utf8'))
+            auth = bytes("Basic ")
+        else:
+            encoded_creds = base64.encodestring(credentials)
+            auth = "Basic "
+        auth += encoded_creds.strip()
+
+        # Build up the MIME payload for the POST data
+        boundary = b('--------------GHSKFJDLGDS7543FJKLFHRE75642756743254')
+        sep_boundary = b('\n--') + boundary
+        end_boundary = sep_boundary + b('--')
+        body = []
+        for key, values in data.items():
+            # handle multiple entries for the same name
+            if type(values) != type([]):
+                values = [values]
+            for value in values:
+                if type(value) is tuple:
+                    fn = b(';filename="%s"' % value[0])
+                    value = value[1]
+                else:
+                    fn = b("")
+                body.append(sep_boundary)
+                body.append(b('\nContent-Disposition: form-data; name="%s"'%key))
+                body.append(fn)
+                body.append(b("\n\n"))
+                body.append(b(value))
+                if value and value[-1] == b('\r'):
+                    body.append(b('\n'))  # write an extra newline (lurve Macs)
+        body.append(end_boundary)
+        body.append(b("\n"))
+        body = b('').join(body)
+
+        self.announce("Submitting documentation to %s" % (self.repository),
+                      log.INFO)
+
+        # build the Request
+        # We can't use urllib2 since we need to send the Basic
+        # auth right with the first request
+        schema, netloc, url, params, query, fragments = \
+            urlparse.urlparse(self.repository)
+        assert not params and not query and not fragments
+        if schema == 'http':
+            conn = httplib.HTTPConnection(netloc)
+        elif schema == 'https':
+            conn = httplib.HTTPSConnection(netloc)
+        else:
+            raise AssertionError("unsupported schema "+schema)
+
+        data = ''
+        loglevel = log.INFO
+        try:
+            conn.connect()
+            conn.putrequest("POST", url)
+            conn.putheader('Content-type',
+                           'multipart/form-data; boundary=%s'%boundary)
+            conn.putheader('Content-length', str(len(body)))
+            conn.putheader('Authorization', auth)
+            conn.endheaders()
+            conn.send(body)
+        except socket.error, e:
+            self.announce(str(e), log.ERROR)
+            return
+
+        r = conn.getresponse()
+        if r.status == 200:
+            self.announce('Server response (%s): %s' % (r.status, r.reason),
+                          log.INFO)
+        elif r.status == 301:
+            location = r.getheader('Location')
+            if location is None:
+                location = 'http://packages.python.org/%s/' % meta['Name']
+            self.announce('Upload successful. Visit %s' % location,
+                          log.INFO)
+        else:
+            self.announce('Upload failed (%s): %s' % (r.status, r.reason),
+                          log.ERROR)
+        if self.show_response:
+            print '-'*75, r.read(), '-'*75
diff --git a/src/distutils2/tests/test_upload_docs.py b/src/distutils2/tests/test_upload_docs.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/test_upload_docs.py
@@ -0,0 +1,69 @@
+"""build_ext tests
+"""
+import sys, os, shutil, tempfile, unittest2, site, zipfile
+from distutils2.command.upload_docs import upload_docs
+from distutils2.dist import Distribution
+
+SETUP_PY = """\
+from distutils2.core import setup
+
+setup(name='foo')
+"""
+
+class TestUploadDocsTest(unittest2.TestCase):
+    def setUp(self):
+        self.dir = tempfile.mkdtemp()
+        setup = os.path.join(self.dir, 'setup.py')
+        f = open(setup, 'w')
+        f.write(SETUP_PY)
+        f.close()
+        self.old_cwd = os.getcwd()
+        os.chdir(self.dir)
+        
+        self.upload_dir = os.path.join(self.dir, 'build')
+        os.mkdir(self.upload_dir)
+        
+        # A test document.
+        f = open(os.path.join(self.upload_dir, 'index.html'), 'w')
+        f.write("Hello world.")
+        f.close()
+        
+        # An empty folder.
+        os.mkdir(os.path.join(self.upload_dir, 'empty'))
+        
+        if sys.version >= "2.6":
+            self.old_base = site.USER_BASE
+            site.USER_BASE = upload_docs.USER_BASE = tempfile.mkdtemp()
+            self.old_site = site.USER_SITE
+            site.USER_SITE = upload_docs.USER_SITE = tempfile.mkdtemp()
+    
+    def tearDown(self):
+        os.chdir(self.old_cwd)
+        shutil.rmtree(self.dir)
+        if sys.version >= "2.6":
+            shutil.rmtree(site.USER_BASE)
+            shutil.rmtree(site.USER_SITE)
+            site.USER_BASE = self.old_base
+            site.USER_SITE = self.old_site
+
+    def test_create_zipfile(self):
+        # Test to make sure zipfile creation handles common cases.
+        # This explicitly includes a folder containing an empty folder.
+        
+        dist = Distribution()
+        
+        cmd = upload_docs(dist)
+        cmd.upload_dir = self.upload_dir
+        zip_file = cmd.create_zipfile()
+        
+        assert zipfile.is_zipfile(zip_file)
+        
+        zip_f = zipfile.ZipFile(zip_file) # woh...
+        
+        assert zip_f.namelist() == ['index.html']
+
+def test_suite():
+    return unittest2.makeSuite(TestUploadDocsTest)
+
+if __name__ == "__main__":
+    unittest2.main(defaultTest="test_suite")

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:38 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:38 +0200
Subject: [Python-checkins] distutils2: refactored out PyPIServerTestCase and
	renamed	assertEquals to assertEqual
Message-ID: <E1OVLoU-0000gd-R1@dinsdale.python.org>

tarek.ziade pushed 4a0c2ff468c6 to distutils2:

http://hg.python.org/distutils2/rev/4a0c2ff468c6
changeset:   266:4a0c2ff468c6
user:        Konrad Delong <konryd at gmail.com>
date:        Thu May 13 20:02:35 2010 +0200
summary:     refactored out PyPIServerTestCase and renamed assertEquals to assertEqual
files:       src/distutils2/tests/test_upload.py

diff --git a/src/distutils2/tests/test_upload.py b/src/distutils2/tests/test_upload.py
--- a/src/distutils2/tests/test_upload.py
+++ b/src/distutils2/tests/test_upload.py
@@ -19,17 +19,18 @@
 username:me
 """
 
-class UploadTestCase(PyPIRCCommandTestCase):
+class PyPIServerTestCase(unittest2.TestCase):
 
     def setUp(self):
-        super(UploadTestCase, self).setUp()
-        assert not hasattr(self, "pypi")
+        super(PyPIServerTestCase, self).setUp()
         self.pypi = PyPIServer()
         self.pypi.start()
 
     def tearDown(self):
         self.pypi.stop()
 
+class UploadTestCase(PyPIServerTestCase, PyPIRCCommandTestCase):
+
     def test_finalize_options(self):
         # new format
         self.write_file(self.rc, PYPIRC)
@@ -39,7 +40,7 @@
         for attr, waited in (('username', 'me'), ('password', 'secret'),
                              ('realm', 'pypi'),
                              ('repository', 'http://pypi.python.org/pypi')):
-            self.assertEquals(getattr(cmd, attr), waited)
+            self.assertEqual(getattr(cmd, attr), waited)
 
     def test_saved_password(self):
         # file with no password
@@ -49,14 +50,14 @@
         dist = Distribution()
         cmd = upload(dist)
         cmd.ensure_finalized()
-        self.assertEquals(cmd.password, None)
+        self.assertEqual(cmd.password, None)
 
         # make sure we get it as well, if another command
         # initialized it at the dist level
         dist.password = 'xxx'
         cmd = upload(dist)
         cmd.finalize_options()
-        self.assertEquals(cmd.password, 'xxx')
+        self.assertEqual(cmd.password, 'xxx')
 
     def test_upload(self):
         path = os.path.join(self.tmp_dir, 'xxx')
@@ -77,7 +78,7 @@
         self.assertIn('xxx', request_data)
         self.assert_(environ['CONTENT_LENGTH'] > 2000)
         self.assertTrue(environ['CONTENT_TYPE'].startswith('multipart/form-data'))
-        self.assertEquals(environ['REQUEST_METHOD'], 'POST')
+        self.assertEqual(environ['REQUEST_METHOD'], 'POST')
         self.assertNotIn('\n', environ['HTTP_AUTHORIZATION'])
 
 def test_suite():

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:38 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:38 +0200
Subject: [Python-checkins] distutils2: removed upload_docs shot
Message-ID: <E1OVLoU-0000gM-P2@dinsdale.python.org>

tarek.ziade pushed ed6149b35a96 to distutils2:

http://hg.python.org/distutils2/rev/ed6149b35a96
changeset:   262:ed6149b35a96
user:        Konrad Delong <konryd at gmail.com>
date:        Thu May 13 08:04:08 2010 -0400
summary:     removed upload_docs shot
files:       src/distutils2/command/upload_docs.py, src/distutils2/tests/test_upload_docs.py

diff --git a/src/distutils2/command/upload_docs.py b/src/distutils2/command/upload_docs.py
deleted file mode 100644
--- a/src/distutils2/command/upload_docs.py
+++ /dev/null
@@ -1,174 +0,0 @@
-# -*- coding: utf-8 -*-
-"""upload_docs
-
-Implements a Distutils2 'upload_docs' subcommand (upload documentation to
-PyPI's packages.python.org).
-"""
-
-import os
-import socket
-import zipfile
-import httplib
-import base64
-import urlparse
-import tempfile
-import sys
-
-from distutils2 import log
-from distutils2.errors import DistutilsOptionError
-from distutils2.command.upload import upload
-
-_IS_PYTHON3 = sys.version > '3'
-
-try:
-    bytes
-except NameError:
-    bytes = str
-
-def b(str_or_bytes):
-    """Return bytes by either encoding the argument as ASCII or simply return
-    the argument as-is."""
-    if not isinstance(str_or_bytes, bytes):
-        return str_or_bytes.encode('ascii')
-    else:
-        return str_or_bytes
-
-
-class upload_docs(upload):
-
-    description = 'Upload documentation to PyPI'
-
-    user_options = [
-        ('repository=', 'r',
-         "url of repository [default: %s]" % upload.DEFAULT_REPOSITORY),
-        ('show-response', None,
-         'display full response text from server'),
-        ('upload-dir=', None, 'directory to upload'),
-        ]
-    boolean_options = upload.boolean_options
-
-    def initialize_options(self):
-        upload.initialize_options(self)
-        self.upload_dir = None
-
-    def finalize_options(self):
-        upload.finalize_options(self)
-        if self.upload_dir is None:
-            build = self.get_finalized_command('build')
-            self.upload_dir = os.path.join(build.build_base, 'docs')
-            self.mkpath(self.upload_dir)
-        self.ensure_dirname('upload_dir')
-        self.announce('Using upload directory %s' % self.upload_dir)
-
-    def create_zipfile(self):
-        name = self.distribution.metadata['Name']
-        tmp_dir = tempfile.mkdtemp()
-        tmp_file = os.path.join(tmp_dir, "%s.zip" % name)
-        zip_file = zipfile.ZipFile(tmp_file, "w")
-        for root, dirs, files in os.walk(self.upload_dir):
-            if root == self.upload_dir and not files:
-                raise DistutilsOptionError(
-                    "no files found in upload directory '%s'"
-                    % self.upload_dir)
-            for name in files:
-                full = os.path.join(root, name)
-                relative = root[len(self.upload_dir):].lstrip(os.path.sep)
-                dest = os.path.join(relative, name)
-                zip_file.write(full, dest)
-        zip_file.close()
-        return tmp_file
-
-    def run(self):
-        zip_file = self.create_zipfile()
-        self.upload_file(zip_file)
-
-    def upload_file(self, filename):
-        content = open(filename, 'rb').read()
-        meta = self.distribution.metadata
-        data = {
-            ':action': 'doc_upload',
-            'name': meta['Name'],
-            'content': (os.path.basename(filename), content),
-        }
-        # set up the authentication
-        credentials = self.username + ':' + self.password
-        if _IS_PYTHON3:  # base64 only works with bytes in Python 3.
-            encoded_creds = base64.encodebytes(credentials.encode('utf8'))
-            auth = bytes("Basic ")
-        else:
-            encoded_creds = base64.encodestring(credentials)
-            auth = "Basic "
-        auth += encoded_creds.strip()
-
-        # Build up the MIME payload for the POST data
-        boundary = b('--------------GHSKFJDLGDS7543FJKLFHRE75642756743254')
-        sep_boundary = b('\n--') + boundary
-        end_boundary = sep_boundary + b('--')
-        body = []
-        for key, values in data.items():
-            # handle multiple entries for the same name
-            if type(values) != type([]):
-                values = [values]
-            for value in values:
-                if type(value) is tuple:
-                    fn = b(';filename="%s"' % value[0])
-                    value = value[1]
-                else:
-                    fn = b("")
-                body.append(sep_boundary)
-                body.append(b('\nContent-Disposition: form-data; name="%s"'%key))
-                body.append(fn)
-                body.append(b("\n\n"))
-                body.append(b(value))
-                if value and value[-1] == b('\r'):
-                    body.append(b('\n'))  # write an extra newline (lurve Macs)
-        body.append(end_boundary)
-        body.append(b("\n"))
-        body = b('').join(body)
-
-        self.announce("Submitting documentation to %s" % (self.repository),
-                      log.INFO)
-
-        # build the Request
-        # We can't use urllib2 since we need to send the Basic
-        # auth right with the first request
-        schema, netloc, url, params, query, fragments = \
-            urlparse.urlparse(self.repository)
-        assert not params and not query and not fragments
-        if schema == 'http':
-            conn = httplib.HTTPConnection(netloc)
-        elif schema == 'https':
-            conn = httplib.HTTPSConnection(netloc)
-        else:
-            raise AssertionError("unsupported schema "+schema)
-
-        data = ''
-        loglevel = log.INFO
-        try:
-            conn.connect()
-            conn.putrequest("POST", url)
-            conn.putheader('Content-type',
-                           'multipart/form-data; boundary=%s'%boundary)
-            conn.putheader('Content-length', str(len(body)))
-            conn.putheader('Authorization', auth)
-            conn.endheaders()
-            conn.send(body)
-        except socket.error, e:
-            self.announce(str(e), log.ERROR)
-            return
-
-        r = conn.getresponse()
-        if r.status == 200:
-            self.announce('Server response (%s): %s' % (r.status, r.reason),
-                          log.INFO)
-        elif r.status == 301:
-            location = r.getheader('Location')
-            if location is None:
-                location = 'http://packages.python.org/%s/' % meta['Name']
-            self.announce('Upload successful. Visit %s' % location,
-                          log.INFO)
-        else:
-            self.announce('Upload failed (%s): %s' % (r.status, r.reason),
-                          log.ERROR)
-        if self.show_response:
-            print '-'*75, r.read(), '-'*75
diff --git a/src/distutils2/tests/test_upload_docs.py b/src/distutils2/tests/test_upload_docs.py
deleted file mode 100644
--- a/src/distutils2/tests/test_upload_docs.py
+++ /dev/null
@@ -1,69 +0,0 @@
-"""build_ext tests
-"""
-import sys, os, shutil, tempfile, unittest2, site, zipfile
-from distutils2.command.upload_docs import upload_docs
-from distutils2.dist import Distribution
-
-SETUP_PY = """\
-from distutils2.core import setup
-
-setup(name='foo')
-"""
-
-class TestUploadDocsTest(unittest2.TestCase):
-    def setUp(self):
-        self.dir = tempfile.mkdtemp()
-        setup = os.path.join(self.dir, 'setup.py')
-        f = open(setup, 'w')
-        f.write(SETUP_PY)
-        f.close()
-        self.old_cwd = os.getcwd()
-        os.chdir(self.dir)
-        
-        self.upload_dir = os.path.join(self.dir, 'build')
-        os.mkdir(self.upload_dir)
-        
-        # A test document.
-        f = open(os.path.join(self.upload_dir, 'index.html'), 'w')
-        f.write("Hello world.")
-        f.close()
-        
-        # An empty folder.
-        os.mkdir(os.path.join(self.upload_dir, 'empty'))
-        
-        if sys.version >= "2.6":
-            self.old_base = site.USER_BASE
-            site.USER_BASE = upload_docs.USER_BASE = tempfile.mkdtemp()
-            self.old_site = site.USER_SITE
-            site.USER_SITE = upload_docs.USER_SITE = tempfile.mkdtemp()
-    
-    def tearDown(self):
-        os.chdir(self.old_cwd)
-        shutil.rmtree(self.dir)
-        if sys.version >= "2.6":
-            shutil.rmtree(site.USER_BASE)
-            shutil.rmtree(site.USER_SITE)
-            site.USER_BASE = self.old_base
-            site.USER_SITE = self.old_site
-
-    def test_create_zipfile(self):
-        # Test to make sure zipfile creation handles common cases.
-        # This explicitly includes a folder containing an empty folder.
-        
-        dist = Distribution()
-        
-        cmd = upload_docs(dist)
-        cmd.upload_dir = self.upload_dir
-        zip_file = cmd.create_zipfile()
-        
-        assert zipfile.is_zipfile(zip_file)
-        
-        zip_f = zipfile.ZipFile(zip_file) # woh...
-        
-        assert zip_f.namelist() == ['index.html']
-
-def test_suite():
-    return unittest2.makeSuite(TestUploadDocsTest)
-
-if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:38 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:38 +0200
Subject: [Python-checkins] distutils2: merged upstream
Message-ID: <E1OVLoU-0000gA-NN@dinsdale.python.org>

tarek.ziade pushed be8f96b7c538 to distutils2:

http://hg.python.org/distutils2/rev/be8f96b7c538
changeset:   259:be8f96b7c538
parent:      258:0b00b7e6e0b8
parent:      133:d519643faee6
user:        Konrad Delong <konryd at gmail.com>
date:        Sun May 09 13:02:57 2010 -0400
summary:     merged upstream
files:       

diff --git a/.hgtags b/.hgtags
new file mode 100644
--- /dev/null
+++ b/.hgtags
@@ -0,0 +1,1 @@
+5847fb2b7549fb301882c1054eb8b3d0893e3570 1.0a1
diff --git a/src/CHANGES.txt b/src/CHANGES.txt
--- a/src/CHANGES.txt
+++ b/src/CHANGES.txt
@@ -2,9 +2,16 @@
 CHANGES
 =======
 
-1.0a1
------
+1.0a2 - ?
+--------
 
-- Initial import from the stdlib
+-
 
+1.0a1 - 2010-05-06
+------------------
 
+- Initial import from the stdlib [tarek]
+- Added support for PEP 376 in distutils2.version [tarek]
+- Added support for PEP 345 in distutils2.metadata [tarek]
+- Added distutils2.util.find_package [tarek]
+
diff --git a/src/README.txt b/src/README.txt
--- a/src/README.txt
+++ b/src/README.txt
@@ -1,5 +1,15 @@
-======
-README
-======
+==========
+Distutils2
+==========
 
+Welcome to Distutils2 !
 
+Distutils2 is the new version of Distutils. It's not backward compatible with
+Distutils but provides more features, and implement most new packaging
+standards.
+
+See the documentation at http://packages.python.org/Distutils2 for more info.
+
+**Beware that Distutils2 is its in early stage and should not be used in
+production. Its API is subject to changes**
+
diff --git a/src/distutils2/__init__.py b/src/distutils2/__init__.py
--- a/src/distutils2/__init__.py
+++ b/src/distutils2/__init__.py
@@ -10,7 +10,7 @@
 __all__ = ['__version__', 'setup']
 
 __revision__ = "$Id: __init__.py 78020 2010-02-06 16:37:32Z benjamin.peterson $"
-__version__ = "1.0a1"
+__version__ = "1.0a2"
 
 from distutils2.core import setup
 
diff --git a/src/distutils2/metadata.py b/src/distutils2/metadata.py
--- a/src/distutils2/metadata.py
+++ b/src/distutils2/metadata.py
@@ -103,20 +103,44 @@
             if marker in keys:
                 return True
         return False
+
     keys = fields.keys()
-    is_1_1 = _has_marker(keys, _314_MARKERS)
-    is_1_2 = _has_marker(keys, _345_MARKERS)
+    possible_versions = ['1.0', '1.1', '1.2']
+
+
+    # first let's try to see if a field is not part of one of the version
+    for key in keys:
+        if key not in _241_FIELDS and '1.0' in possible_versions:
+            possible_versions.remove('1.0')
+        if key not in _314_FIELDS and '1.1' in possible_versions:
+            possible_versions.remove('1.1')
+        if key not in _345_FIELDS and '1.2' in possible_versions:
+            possible_versions.remove('1.2')
+
+    # possible_version contains qualified versions
+    if len(possible_versions) == 1:
+        return possible_versions[0]   # found !
+    elif len(possible_versions) == 0:
+        raise MetadataConflictError('Unknown metadata set')
+
+    # let's see if one unique marker is found
+    is_1_1 = '1.1' in possible_versions and _has_marker(keys, _314_MARKERS)
+    is_1_2 = '1.2' in possible_versions and _has_marker(keys, _345_MARKERS)
     if is_1_1 and is_1_2:
-        raise MetadataConflictError('You used both 1.1 and 1.2 fields')
+        raise MetadataConflictError('You used incompatible 1.1 and 1.2 fields')
 
     # we have the choice, either 1.0, or 1.2
     #   - 1.0 has a broken Summary field but work with all tools
     #   - 1.1 is to avoid
     #   - 1.2 fixes Summary but is not spreaded yet
     if not is_1_1 and not is_1_2:
-        return PKG_INFO_PREFERRED_VERSION
+        # we couldn't find any specific marker
+        if PKG_INFO_PREFERRED_VERSION in possible_versions:
+            return PKG_INFO_PREFERRED_VERSION
     if is_1_1:
         return '1.1'
+
+    # default marker when 1.0 is disqualified
     return '1.2'
 
 _ATTR2FIELD = {'metadata_version': 'Metadata-Version',
@@ -269,6 +293,8 @@
             if field in _LISTFIELDS:
                 # we can have multiple lines
                 values = msg.get_all(field)
+                if field in _LISTTUPLEFIELDS and values is not None:
+                    values = [tuple(value.split(',')) for value in values]
                 self.set(field, values)
             else:
                 # single line
@@ -299,6 +325,9 @@
                     values = values.replace('\n', '\n       |')
                 values = [values]
 
+            if field in _LISTTUPLEFIELDS:
+                values = [','.join(value) for value in values]
+
             for value in values:
                 self._write_field(fileobject, field, value)
 
diff --git a/src/distutils2/tests/test_metadata.py b/src/distutils2/tests/test_metadata.py
--- a/src/distutils2/tests/test_metadata.py
+++ b/src/distutils2/tests/test_metadata.py
@@ -4,7 +4,8 @@
 import sys
 from StringIO import StringIO
 
-from distutils2.metadata import DistributionMetadata, _interpret
+from distutils2.metadata import (DistributionMetadata, _interpret,
+                                 PKG_INFO_PREFERRED_VERSION)
 
 class DistributionMetadataTestCase(unittest2.TestCase):
 
@@ -200,6 +201,31 @@
         self.assertEquals(missing, ['Name', 'Home-page'])
         self.assertEquals(len(warnings), 2)
 
+    def test_best_choice(self):
+        metadata = DistributionMetadata()
+        metadata['Version'] = '1.0'
+        self.assertEquals(metadata.version, PKG_INFO_PREFERRED_VERSION)
+        metadata['Classifier'] = ['ok']
+        self.assertEquals(metadata.version, '1.2')
+
+    def test_project_urls(self):
+        # project-url is a bit specific, make sure we write it
+        # properly in PKG-INFO
+        metadata = DistributionMetadata()
+        metadata['Version'] = '1.0'
+        metadata['Project-Url'] = [('one', 'http://ok')]
+        self.assertEquals(metadata['Project-Url'], [('one', 'http://ok')])
+        file_ = StringIO()
+        metadata.write_file(file_)
+        file_.seek(0)
+        res = file_.read().split('\n')
+        self.assertIn('Project-URL: one,http://ok', res)
+
+        file_.seek(0)
+        metadata = DistributionMetadata()
+        metadata.read_file(file_)
+        self.assertEquals(metadata['Project-Url'], [('one', 'http://ok')])
+
 
 def test_suite():
     return unittest2.makeSuite(DistributionMetadataTestCase)
diff --git a/src/distutils2/tests/test_util.py b/src/distutils2/tests/test_util.py
--- a/src/distutils2/tests/test_util.py
+++ b/src/distutils2/tests/test_util.py
@@ -14,7 +14,7 @@
                             check_environ, split_quoted, strtobool,
                             rfc822_escape, get_compiler_versions,
                             _find_exe_version, _MAC_OS_X_LD_VERSION,
-                            byte_compile)
+                            byte_compile, find_packages)
 from distutils2 import util
 from distutils2.tests import support
 
@@ -32,7 +32,9 @@
             self.stdout = StringIO(exes[self.cmd])
             self.stderr = StringIO()
 
-class UtilTestCase(support.EnvironGuard, unittest2.TestCase):
+class UtilTestCase(support.EnvironGuard,
+                   support.TempdirManager,
+                   unittest2.TestCase):
 
     def setUp(self):
         super(UtilTestCase, self).setUp()
@@ -257,6 +259,44 @@
         self.assertRaises(DistutilsFileError, util.newer, 'xxx', 'xxx')
 
 
+    def test_find_packages(self):
+        # let's create a structure we want to scan:
+        #
+        #   pkg1
+        #     __init__
+        #     pkg2
+        #       __init__
+        #     pkg3
+        #       __init__
+        #       pkg6
+        #           __init__
+        #     pkg4    <--- not a pkg
+        #       pkg8
+        #          __init__
+        #   pkg5
+        #     __init__
+        #
+        root = self.mkdtemp()
+        pkg1 = os.path.join(root, 'pkg1')
+        os.mkdir(pkg1)
+        self.write_file(os.path.join(pkg1, '__init__.py'))
+        os.mkdir(os.path.join(pkg1, 'pkg2'))
+        self.write_file(os.path.join(pkg1, 'pkg2', '__init__.py'))
+        os.mkdir(os.path.join(pkg1, 'pkg3'))
+        self.write_file(os.path.join(pkg1, 'pkg3', '__init__.py'))
+        os.mkdir(os.path.join(pkg1, 'pkg3', 'pkg6'))
+        self.write_file(os.path.join(pkg1, 'pkg3', 'pkg6', '__init__.py'))
+        os.mkdir(os.path.join(pkg1, 'pkg4'))
+        os.mkdir(os.path.join(pkg1, 'pkg4', 'pkg8'))
+        self.write_file(os.path.join(pkg1, 'pkg4', 'pkg8', '__init__.py'))
+        pkg5 = os.path.join(root, 'pkg5')
+        os.mkdir(pkg5)
+        self.write_file(os.path.join(pkg5, '__init__.py'))
+
+        res = find_packages([root], ['pkg1.pkg2'])
+        self.assertEquals(res, ['pkg1', 'pkg5', 'pkg1.pkg3', 'pkg1.pkg3.pkg6'])
+
+
 def test_suite():
     return unittest2.makeSuite(UtilTestCase)
 
diff --git a/src/distutils2/util.py b/src/distutils2/util.py
--- a/src/distutils2/util.py
+++ b/src/distutils2/util.py
@@ -7,6 +7,7 @@
 __revision__ = "$Id: util.py 77761 2010-01-26 22:46:15Z tarek.ziade $"
 
 import sys, os, string, re
+from fnmatch import fnmatchcase
 
 from distutils2.errors import (DistutilsPlatformError, DistutilsFileError,
                                DistutilsByteCompileError)
@@ -543,3 +544,70 @@
         f.write(line + "\n")
     f.close()
 
+def _is_package(path):
+    """Returns True if path is a package (a dir with an __init__ file."""
+    if not os.path.isdir(path):
+        return False
+    return os.path.isfile(os.path.join(path, '__init__.py'))
+
+def _under(path, root):
+    path = path.split(os.sep)
+    root = root.split(os.sep)
+    if len(root) > len(path):
+        return False
+    for pos, part in enumerate(root):
+        if path[pos] != part:
+            return False
+    return True
+
+def _package_name(root_path, path):
+    """Returns a dotted package name, given a subpath."""
+    if not _under(path, root_path):
+        raise ValueError('"%s" is not a subpath of "%s"' % (path, root_path))
+    return path[len(root_path) + 1:].replace(os.sep, '.')
+
+def find_packages(paths=('.',), exclude=()):
+    """Return a list all Python packages found recursively within
+    directories 'paths'
+
+    'paths' should be supplied as a sequence of "cross-platform"
+    (i.e. URL-style) path; it will be converted to the appropriate local
+    path syntax.
+
+    'exclude' is a sequence of package names to exclude; '*' can be used as
+    a wildcard in the names, such that 'foo.*' will exclude all subpackages
+    of 'foo' (but not 'foo' itself).
+    """
+    packages = []
+    discarded = []
+    def _discarded(path):
+        for discard in discarded:
+            if _under(path, discard):
+                return True
+        return False
+
+    for path in paths:
+        path = convert_path(path)
+        for root, dirs, files in os.walk(path):
+            for dir_ in dirs:
+                fullpath = os.path.join(root, dir_)
+                if _discarded(fullpath):
+                    continue
+                # we work only with Python packages
+                if not _is_package(fullpath):
+                    discarded.append(fullpath)
+                    continue
+                # see if it's excluded
+                excluded = False
+                package_name = _package_name(path, fullpath)
+                for pattern in exclude:
+                    if fnmatchcase(package_name, pattern):
+                        excluded = True
+                        break
+                if excluded:
+                    continue
+
+                # adding it to the list
+                packages.append(package_name)
+    return packages
+
diff --git a/src/setup.py b/src/setup.py
--- a/src/setup.py
+++ b/src/setup.py
@@ -8,6 +8,7 @@
 from distutils2.command.sdist import sdist
 from distutils2.command.install import install
 from distutils2 import __version__ as VERSION
+from distutils2.util import find_packages
 
 f = open('README.txt')
 try:
@@ -67,20 +68,36 @@
 if sys.version < '2.6':
     setup_kwargs['scripts'] = ['distutils2/mkpkg.py']
 
+_CLASSIFIERS = """\
+Development Status :: 3 - Alpha
+Intended Audience :: Developers
+License :: OSI Approved :: Python Software Foundation License
+Operating System :: OS Independent
+Programming Language :: Python
+Topic :: Software Development :: Libraries :: Python Modules
+Topic :: System :: Archiving :: Packaging
+Topic :: System :: Systems Administration
+Topic :: Utilities"""
+
 setup (name="Distutils2",
        version=VERSION,
-       description="Python Distribution Utilities",
+       summary="Python Distribution Utilities",
+       keywords=['packaging', 'distutils'],
        author="Tarek Ziade",
        author_email="tarek at ziade.org",
-       url="http://www.python.org/sigs/distutils-sig",
+       home_page="http://bitbucket.org/tarek/distutils2/wiki/Home",
        license="PSF",
-       long_description=README,
-       packages=['distutils2',
-                 'distutils2.command',
-                 'distutils2.tests',
-                 'distutils2._backport'],
+       description=README,
+       classifier=_CLASSIFIERS.split('\n'),
+       packages=find_packages(),
        cmdclass={'sdist': sdist_hg, 'install': install_hg},
        package_data={'distutils2._backport': ['sysconfig.cfg']},
+       project_url=[('Mailing-list',
+                    'http://mail.python.org/mailman/listinfo/distutils-sig/'),
+                    ('Documentation',
+                     'http://packages.python.org/Distutils2'),
+                    ('Repository', 'http://hg.python.org/distutils2'),
+                    ('Bug tracker', 'http://bugs.python.org')],
        **setup_kwargs
        )
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:38 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:38 +0200
Subject: [Python-checkins] distutils2: added PYPiServer and refactored
	upload test to make use	of it
Message-ID: <E1OVLoU-0000gQ-PX@dinsdale.python.org>

tarek.ziade pushed 00f3215f67f9 to distutils2:

http://hg.python.org/distutils2/rev/00f3215f67f9
changeset:   263:00f3215f67f9
user:        Konrad Delong <konryd at gmail.com>
date:        Thu May 13 14:06:00 2010 +0200
summary:     added PYPiServer and refactored upload test to make use of it
files:       src/distutils2/tests/pypi_server.py, src/distutils2/tests/test_upload.py

diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypi_server.py
@@ -0,0 +1,49 @@
+import threading
+import copy
+import Queue
+try:
+    from urllib.request import urlopen
+except ImportError:
+    from urllib2 import urlopen
+
+import time
+from wsgiref.simple_server import make_server, demo_app
+
+class PyPIServer(threading.Thread):
+    """Thread that wraps a wsgi app"""
+    def __init__(self):
+        threading.Thread.__init__(self)
+        self.httpd = make_server('', 0, self.pypi_app)
+        self.httpd.RequestHandlerClass.log_request = lambda *_: None
+        self.address = self.httpd.server_address
+        self.request_queue = Queue.Queue()
+        self._requests = []
+
+    def run(self):
+        self.httpd.serve_forever()
+
+    def stop(self):
+        self.httpd.shutdown()
+        self.join()
+        time.sleep(0.2)
+
+    def pypi_app(self, environ, start_response):
+        status = '200 OK' # HTTP Status
+        headers = [('Content-type', 'text/plain')] # HTTP Headers
+        start_response(status, headers)
+        request_data = environ.pop('wsgi.input').read(int(environ['CONTENT_LENGTH']))
+        self.request_queue.put((environ, request_data))
+        return ["hello"]
+
+    @property
+    def requests(self):
+        while True:
+            try:
+                self._requests.append(self.request_queue.get_nowait())
+            except Queue.Empty:
+                break
+        return self._requests
+
+    @property
+    def full_address(self):
+        return "http://%s:%s" % self.address
diff --git a/src/distutils2/tests/test_upload.py b/src/distutils2/tests/test_upload.py
--- a/src/distutils2/tests/test_upload.py
+++ b/src/distutils2/tests/test_upload.py
@@ -10,24 +10,7 @@
 
 from distutils2.tests import support
 from distutils2.tests.test_config import PYPIRC, PyPIRCCommandTestCase
-
-PYPIRC_LONG_PASSWORD = """\
-[distutils]
-
-index-servers =
-    server1
-    server2
-
-[server1]
-username:me
-password:aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
-
-[server2]
-username:meagain
-password: secret
-realm:acme
-repository:http://another.pypi/
-"""
+from distutils2.tests.pypi_server import PyPIServer
 
 
 PYPIRC_NOPASSWORD = """\
@@ -40,38 +23,15 @@
 username:me
 """
 
-class FakeOpen(object):
-
-    def __init__(self, url):
-        self.url = url
-        if not isinstance(url, str):
-            self.req = url
-        else:
-            self.req = None
-        self.msg = 'OK'
-
-    def getcode(self):
-        return 200
-
-
-class uploadTestCase(PyPIRCCommandTestCase):
+class UploadTestCase(PyPIRCCommandTestCase):
 
     def setUp(self):
-        super(uploadTestCase, self).setUp()
-        self.old_open = upload_mod.urlopen
-        upload_mod.urlopen = self._urlopen
-        self.last_open = None
-
-    def tearDown(self):
-        upload_mod.urlopen = self.old_open
-        super(uploadTestCase, self).tearDown()
-
-    def _urlopen(self, url):
-        self.last_open = FakeOpen(url)
-        return self.last_open
+        super(UploadTestCase, self).setUp()
+        assert not hasattr(self, "pypi")
+        self.pypi = PyPIServer()
+        self.pypi.start()
 
     def test_finalize_options(self):
-
         # new format
         self.write_file(self.rc, PYPIRC)
         dist = Distribution()
@@ -89,7 +49,7 @@
         # make sure it passes
         dist = Distribution()
         cmd = upload(dist)
-        cmd.finalize_options()
+        cmd.ensure_finalized()
         self.assertEquals(cmd.password, None)
 
         # make sure we get it as well, if another command
@@ -100,33 +60,29 @@
         self.assertEquals(cmd.password, 'xxx')
 
     def test_upload(self):
-        tmp = self.mkdtemp()
-        path = os.path.join(tmp, 'xxx')
+        path = os.path.join(self.tmp_dir, 'xxx')
         self.write_file(path)
         command, pyversion, filename = 'xxx', '2.6', path
         dist_files = [(command, pyversion, filename)]
-        self.write_file(self.rc, PYPIRC_LONG_PASSWORD)
 
         # lets run it
         pkg_dir, dist = self.create_dist(dist_files=dist_files, author=u'd??d??')
         cmd = upload(dist)
         cmd.ensure_finalized()
+        cmd.repository = self.pypi.full_address
         cmd.run()
 
         # what did we send ?
-        self.assertIn('d??d??', self.last_open.req.data)
-        headers = dict(self.last_open.req.headers)
-        self.assert_(headers['Content-length'] > 2000)
-        self.assertTrue(headers['Content-type'].startswith('multipart/form-data'))
-        self.assertEquals(self.last_open.req.get_method(), 'POST')
-        self.assertEquals(self.last_open.req.get_full_url(),
-                          'http://pypi.python.org/pypi')
-        self.assertTrue('xxx' in self.last_open.req.data)
-        auth = self.last_open.req.headers['Authorization']
-        self.assertFalse('\n' in auth)
+        environ, request_data = self.pypi.requests[-1]
+        self.assertIn('d??d??', request_data)
+        self.assertIn('xxx', request_data)
+        self.assert_(environ['CONTENT_LENGTH'] > 2000)
+        self.assertTrue(environ['CONTENT_TYPE'].startswith('multipart/form-data'))
+        self.assertEquals(environ['REQUEST_METHOD'], 'POST')
+        self.assertNotIn('\n', environ['HTTP_AUTHORIZATION'])
 
 def test_suite():
-    return unittest2.makeSuite(uploadTestCase)
+    return unittest2.makeSuite(UploadTestCase)
 
 if __name__ == "__main__":
     unittest2.main(defaultTest="test_suite")

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:38 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:38 +0200
Subject: [Python-checkins] distutils2: started work on upload_docs
Message-ID: <E1OVLoU-0000gl-Rv@dinsdale.python.org>

tarek.ziade pushed adec99dc940d to distutils2:

http://hg.python.org/distutils2/rev/adec99dc940d
changeset:   268:adec99dc940d
user:        Konrad Delong <konryd at gmail.com>
date:        Thu May 13 20:26:12 2010 +0200
summary:     started work on upload_docs
files:       src/distutils2/command/upload_docs.py, src/distutils2/tests/test_upload_docs.py

diff --git a/src/distutils2/command/upload_docs.py b/src/distutils2/command/upload_docs.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/command/upload_docs.py
@@ -0,0 +1,19 @@
+import os.path
+from distutils2.core import Command
+
+class upload_docs(Command):
+
+    user_options = [
+        ('upload-dir=', None, 'directory to upload'),
+        ]
+
+    def initialize_options(self):
+        self.upload_dir = None
+
+    def finalize_options(self):
+        if self.upload_dir == None:
+            build = self.get_finalized_command('build')
+            self.upload_dir = os.path.join(build.build_base, "docs")
+
+    def run(self):
+        pass
diff --git a/src/distutils2/tests/test_upload_docs.py b/src/distutils2/tests/test_upload_docs.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/test_upload_docs.py
@@ -0,0 +1,34 @@
+"""Tests for distutils.command.upload_docs."""
+# -*- encoding: utf8 -*-
+import os, os.path, unittest2
+
+from distutils2.command.upload_docs import upload_docs
+from distutils2.core import Distribution
+
+from distutils2.tests.pypi_server import PyPIServer
+from distutils2.tests.test_upload import PyPIServerTestCase
+from distutils2.tests.test_config import PYPIRC, PyPIRCCommandTestCase
+
+
+class UploadDocsTestCase(PyPIServerTestCase, PyPIRCCommandTestCase):
+
+    def setUp(self):
+        super(UploadDocsTestCase, self).setUp()
+        self.dist = Distribution()
+        self.cmd = upload_docs(self.dist)
+
+    def test_generates_uploaddir_if_none(self):
+        self.cmd.ensure_finalized()
+        self.assertEqual(self.cmd.upload_dir, os.path.join("build", "docs"))
+
+    def test_zip_dir(self):
+        pass
+
+    def test_upload(self):
+        pass
+
+def test_suite():
+    return unittest2.makeSuite(UploadDocsTestCase)
+
+if __name__ == "__main__":
+    unittest2.main(defaultTest="test_suite")

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:38 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:38 +0200
Subject: [Python-checkins] distutils2: stopping pypi_server in teardown &
	tested pypi_server	separately
Message-ID: <E1OVLoU-0000gZ-QY@dinsdale.python.org>

tarek.ziade pushed 169e3d4e7247 to distutils2:

http://hg.python.org/distutils2/rev/169e3d4e7247
changeset:   265:169e3d4e7247
user:        Konrad Delong <konryd at gmail.com>
date:        Thu May 13 14:43:38 2010 +0200
summary:     stopping pypi_server in teardown & tested pypi_server separately
files:       src/distutils2/tests/pypi_server.py, src/distutils2/tests/test_pypi_server.py, src/distutils2/tests/test_upload.py

diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
--- a/src/distutils2/tests/pypi_server.py
+++ b/src/distutils2/tests/pypi_server.py
@@ -23,7 +23,10 @@
         status = '200 OK' # HTTP Status
         headers = [('Content-type', 'text/plain')] # HTTP Headers
         start_response(status, headers)
-        request_data = environ.pop('wsgi.input').read(int(environ['CONTENT_LENGTH']))
+        if environ.get("CONTENT_LENGTH"):
+            request_data = environ.pop('wsgi.input').read(int(environ['CONTENT_LENGTH']))
+        else:
+            request_data = environ.pop('wsgi.input').read()
         self.request_queue.put((environ, request_data))
         return ["hello"]
 
diff --git a/src/distutils2/tests/test_pypi_server.py b/src/distutils2/tests/test_pypi_server.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/test_pypi_server.py
@@ -0,0 +1,27 @@
+"""Tests for distutils.command.bdist."""
+import unittest2, urllib, urllib2
+from distutils2.tests.pypi_server import PyPIServer
+
+class PyPIServerTest(unittest2.TestCase):
+
+    def test_records_requests(self):
+        server = PyPIServer()
+        server.start()
+        self.assertEqual(len(server.requests), 0)
+
+        data = "Rock Around The Bunker"
+        headers = {"X-test-header": "Mister Iceberg"}
+
+        request = urllib2.Request(server.full_address, data, headers)
+        urllib2.urlopen(request)
+        self.assertEqual(len(server.requests), 1)
+        environ, request_data = server.requests[-1]
+        self.assertIn("Rock Around The Bunker", request_data)
+        self.assertEqual(environ["HTTP_X_TEST_HEADER"], "Mister Iceberg")
+        server.stop()
+
+def test_suite():
+    return unittest2.makeSuite(PyPIServerTest)
+
+if __name__ == '__main__':
+    unittest2.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_upload.py b/src/distutils2/tests/test_upload.py
--- a/src/distutils2/tests/test_upload.py
+++ b/src/distutils2/tests/test_upload.py
@@ -27,6 +27,9 @@
         self.pypi = PyPIServer()
         self.pypi.start()
 
+    def tearDown(self):
+        self.pypi.stop()
+
     def test_finalize_options(self):
         # new format
         self.write_file(self.rc, PYPIRC)

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:38 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:38 +0200
Subject: [Python-checkins] distutils2: refactored the pypi server a little
	bit (also, moved the PyPIServerTestCase to
Message-ID: <E1OVLoU-0000h6-UM@dinsdale.python.org>

tarek.ziade pushed ed1fc0d836b6 to distutils2:

http://hg.python.org/distutils2/rev/ed1fc0d836b6
changeset:   273:ed1fc0d836b6
user:        Konrad Delong <konryd at gmail.com>
date:        Mon May 17 01:08:19 2010 +0200
summary:     refactored the pypi server a little bit (also, moved the PyPIServerTestCase to that file)
files:       src/distutils2/tests/pypi_server.py, src/distutils2/tests/test_upload.py, src/distutils2/tests/test_upload_docs.py

diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
--- a/src/distutils2/tests/pypi_server.py
+++ b/src/distutils2/tests/pypi_server.py
@@ -1,6 +1,17 @@
-import Queue, threading, time
+import Queue, threading, time, unittest2
 from wsgiref.simple_server import make_server
 
+class PyPIServerTestCase(unittest2.TestCase):
+
+    def setUp(self):
+        super(PyPIServerTestCase, self).setUp()
+        self.pypi = PyPIServer()
+        self.pypi.start()
+
+    def tearDown(self):
+        super(PyPIServerTestCase, self).tearDown()
+        self.pypi.stop()
+
 class PyPIServer(threading.Thread):
     """Thread that wraps a wsgi app"""
     def __init__(self):
@@ -10,6 +21,9 @@
         self.address = self.httpd.server_address
         self.request_queue = Queue.Queue()
         self._requests = []
+        self._default_response_status = "200 OK"
+        self._default_response_headers = [('Content-type', 'text/plain')]
+        self._default_response_data = ["hello"]
 
     def run(self):
         self.httpd.serve_forever()
@@ -19,15 +33,21 @@
         self.join()
 
     def pypi_app(self, environ, start_response):
-        status = '200 OK' # HTTP Status
-        headers = [('Content-type', 'text/plain')] # HTTP Headers
-        start_response(status, headers)
+        # record the request
         if environ.get("CONTENT_LENGTH"):
             request_data = environ.pop('wsgi.input').read(int(environ['CONTENT_LENGTH']))
         else:
             request_data = environ.pop('wsgi.input').read()
         self.request_queue.put((environ, request_data))
-        return ["hello"]
+        # send back a response
+        status, headers, data = self.get_next_response()
+        start_response(status, headers)
+        return data
+
+    def get_next_response(self):
+        return (self._default_response_status,
+                self._default_response_headers,
+                self._default_response_data)
 
     @property
     def requests(self):
diff --git a/src/distutils2/tests/test_upload.py b/src/distutils2/tests/test_upload.py
--- a/src/distutils2/tests/test_upload.py
+++ b/src/distutils2/tests/test_upload.py
@@ -5,7 +5,7 @@
 from distutils2.command.upload import upload
 from distutils2.core import Distribution
 
-from distutils2.tests.pypi_server import PyPIServer
+from distutils2.tests.pypi_server import PyPIServer, PyPIServerTestCase
 from distutils2.tests.test_config import PYPIRC, PyPIRCCommandTestCase
 
 
@@ -19,16 +19,6 @@
 username:me
 """
 
-class PyPIServerTestCase(unittest2.TestCase):
-
-    def setUp(self):
-        super(PyPIServerTestCase, self).setUp()
-        self.pypi = PyPIServer()
-        self.pypi.start()
-
-    def tearDown(self):
-        self.pypi.stop()
-
 class UploadTestCase(PyPIServerTestCase, PyPIRCCommandTestCase):
 
     def test_finalize_options(self):
diff --git a/src/distutils2/tests/test_upload_docs.py b/src/distutils2/tests/test_upload_docs.py
--- a/src/distutils2/tests/test_upload_docs.py
+++ b/src/distutils2/tests/test_upload_docs.py
@@ -8,8 +8,7 @@
 from distutils2.core import Distribution
 
 from distutils2.tests import support
-from distutils2.tests.pypi_server import PyPIServer
-from distutils2.tests.test_upload import PyPIServerTestCase
+from distutils2.tests.pypi_server import PyPIServer, PyPIServerTestCase
 from distutils2.tests.test_config import PYPIRC, PyPIRCCommandTestCase
 
 
@@ -108,6 +107,9 @@
     def test_honours_dry_run(self):
         pass
 
+    def test_reads_pypirc_data(self):
+        pass
+
 def test_suite():
     return unittest2.makeSuite(UploadDocsTestCase)
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:38 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:38 +0200
Subject: [Python-checkins] distutils2: implemented run for the simplest case
Message-ID: <E1OVLoU-0000h2-Tv@dinsdale.python.org>

tarek.ziade pushed 4c79b30a4bc4 to distutils2:

http://hg.python.org/distutils2/rev/4c79b30a4bc4
changeset:   272:4c79b30a4bc4
user:        Konrad Delong <konryd at gmail.com>
date:        Fri May 14 00:42:49 2010 +0200
summary:     implemented run for the simplest case
files:       src/distutils2/command/upload_docs.py, src/distutils2/tests/test_upload_docs.py

diff --git a/src/distutils2/command/upload_docs.py b/src/distutils2/command/upload_docs.py
--- a/src/distutils2/command/upload_docs.py
+++ b/src/distutils2/command/upload_docs.py
@@ -1,10 +1,11 @@
-import os.path, tempfile, zipfile
+import base64, httplib, os.path, tempfile, urlparse, zipfile
+from cStringIO import StringIO
+from distutils2 import log
 from distutils2.core import Command
 
-def zip_dir_into(directory, destination):
-    """Compresses recursively contents of directory into a zipfile located
-    under given destination.
-    """
+def zip_dir(directory):
+    """Compresses recursively contents of directory into a StringIO object"""
+    destination = StringIO()
     zip_file = zipfile.ZipFile(destination, "w")
     for root, dirs, files in os.walk(directory):
         for name in files:
@@ -13,14 +14,14 @@
             dest = os.path.join(relative, name)
             zip_file.write(full, dest)
     zip_file.close()
+    return destination
 
 # grabbed from
 #    http://code.activestate.com/recipes/146306-http-client-to-post-using-multipartform-data/
 def encode_multipart(fields, files, boundary=None):
     """
     fields is a sequence of (name, value) elements for regular form fields.
-    files is a sequence of (name, filename, content_type, value)
-                            elements for data to be uploaded as files
+    files is a sequence of (name, filename, value) elements for data to be uploaded as files
     Return (content_type, body) ready for httplib.HTTP instance
     """
     if boundary is None:
@@ -33,11 +34,10 @@
             'Content-Disposition: form-data; name="%s"' % key,
             '',
             value])
-    for (key, filename, content_type, value) in files:
+    for (key, filename, value) in files:
         l.extend([
             '--' + boundary,
             'Content-Disposition: form-data; name="%s"; filename="%s"' % (key, filename),
-            'Content-Type: %s' % content_type,
             '',
             value])
     l.append('--' + boundary + '--')
@@ -61,4 +61,34 @@
             self.upload_dir = os.path.join(build.build_base, "docs")
 
     def run(self):
-        pass
+        tmp_dir = tempfile.mkdtemp()
+        name = self.distribution.metadata['Name']
+        zip_file = zip_dir(self.upload_dir)
+
+        fields = {':action': 'doc_upload', 'name': name}.items()
+        files = [('content', name, zip_file.getvalue())]
+        content_type, body = encode_multipart(fields, files)
+
+        credentials = self.username + ':' + self.password
+        auth = "Basic " + base64.encodestring(credentials).strip()
+
+        self.announce("Submitting documentation to %s" % (self.repository),
+                      log.INFO)
+
+        schema, netloc, url, params, query, fragments = \
+            urlparse.urlparse(self.repository)
+        conn = httplib.HTTPConnection(netloc)
+
+        try:
+            conn.connect()
+            conn.putrequest("POST", url)
+            conn.putheader('Content-type', content_type)
+            conn.putheader('Content-length', str(len(body)))
+            conn.putheader('Authorization', auth)
+            conn.endheaders()
+            conn.send(body)
+        except socket.error, e:
+            self.announce(str(e), log.ERROR)
+            return
+
+        r = conn.getresponse()
diff --git a/src/distutils2/tests/test_upload_docs.py b/src/distutils2/tests/test_upload_docs.py
--- a/src/distutils2/tests/test_upload_docs.py
+++ b/src/distutils2/tests/test_upload_docs.py
@@ -1,8 +1,9 @@
 """Tests for distutils.command.upload_docs."""
 # -*- encoding: utf8 -*-
 import os, os.path, tempfile, unittest2, zipfile
+from cStringIO import StringIO
 
-from distutils2.command.upload_docs import (upload_docs, zip_dir_into,
+from distutils2.command.upload_docs import (upload_docs, zip_dir,
                                     encode_multipart)
 from distutils2.core import Distribution
 
@@ -23,14 +24,12 @@
 'd',
 '---x',
 'Content-Disposition: form-data; name="e"; filename="f"',
-'Content-Type: g',
 '',
-'h',
+'g',
 '---x',
-'Content-Disposition: form-data; name="i"; filename="j"',
-'Content-Type: k',
+'Content-Disposition: form-data; name="h"; filename="i"',
 '',
-'l',
+'j',
 '---x--',
 '',
 ])
@@ -42,33 +41,68 @@
     def setUp(self):
         super(UploadDocsTestCase, self).setUp()
         self.dist = Distribution()
+        self.dist.metadata['Name'] = "distr-name"
         self.cmd = upload_docs(self.dist)
 
     def test_generates_uploaddir_name_not_given(self):
         self.cmd.ensure_finalized()
         self.assertEqual(self.cmd.upload_dir, os.path.join("build", "docs"))
 
-    def test_zip_dir_into(self):
-        source_dir = tempfile.mkdtemp()
-        os.mkdir(os.path.join(source_dir, "some_dir"))
-        some_file = open(os.path.join(source_dir, "some_dir", "some_file"), "w")
+    def prepare_sample_dir(self):
+        sample_dir = tempfile.mkdtemp()
+        os.mkdir(os.path.join(sample_dir, "some_dir"))
+        some_file = open(os.path.join(sample_dir, "some_dir", "some_file"), "w")
         some_file.write("Ce mortel ennui")
         some_file.close()
-        dest_dir = tempfile.mkdtemp()
-        destination = os.path.join(dest_dir, "archive.zip")
-        zip_dir_into(source_dir, destination)
+        return sample_dir
 
-        self.assertTrue(zipfile.is_zipfile(destination))
-        zip_f = zipfile.ZipFile(destination)
+    def test_zip_dir(self):
+        source_dir = self.prepare_sample_dir()
+        compressed = zip_dir(source_dir)
+
+        zip_f = zipfile.ZipFile(compressed)
         self.assertEqual(zip_f.namelist(), ['some_dir/some_file'])
 
     def test_encode_multipart(self):
         fields = [("a", "b"), ("c", "d")]
-        files = [("e", "f", "g", "h"), ("i", "j", "k", "l")]
-        self.assertEqual(encode_multipart(fields, files, "-x")[0], "multipart/form-data; boundary=-x")
-        self.assertEqual(encode_multipart(fields, files, "-x")[1], EXPECTED_MULTIPART_OUTPUT)
+        files = [("e", "f", "g"), ("h", "i", "j")]
+        content_type, body = encode_multipart(fields, files, "-x")
+        self.assertEqual(content_type, "multipart/form-data; boundary=-x")
+        self.assertEqual(body, EXPECTED_MULTIPART_OUTPUT)
 
     def test_upload(self):
+        self.cmd.ensure_finalized()
+        self.cmd.repository = self.pypi.full_address
+        self.cmd.upload_dir = self.prepare_sample_dir()
+        self.cmd.username = "username"
+        self.cmd.password = "password"
+        self.cmd.run()
+
+        self.assertEqual(len(self.pypi.requests), 1)
+        environ, request_data = self.pypi.requests[-1]
+
+        self.assertIn("content", request_data)
+        self.assertIn("Basic", environ['HTTP_AUTHORIZATION'])
+        self.assertTrue(environ['CONTENT_TYPE'].startswith('multipart/form-data;'))
+
+        action, name, content =\
+            request_data.split("----------------GHSKFJDLGDS7543FJKLFHRE75642756743254")[1:4]
+
+        # check that we picked the right chunks
+        self.assertIn('name=":action"', action)
+        self.assertIn('name="name"', name)
+        self.assertIn('name="content"', content)
+
+        # check their contents
+        self.assertIn("doc_upload", action)
+        self.assertIn("distr-name", name)
+        self.assertIn("some_dir/some_file", content)
+        self.assertIn("Ce mortel ennui", content)
+
+    def test_https_connection(self):
+        pass
+
+    def test_handling_response(self):
         pass
 
     def test_honours_dry_run(self):

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:38 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:38 +0200
Subject: [Python-checkins] distutils2: added zip_dir_into
Message-ID: <E1OVLoU-0000gp-SS@dinsdale.python.org>

tarek.ziade pushed 3804252d385c to distutils2:

http://hg.python.org/distutils2/rev/3804252d385c
changeset:   269:3804252d385c
user:        Konrad Delong <konryd at gmail.com>
date:        Thu May 13 22:02:57 2010 +0200
summary:     added zip_dir_into
files:       src/distutils2/command/upload_docs.py, src/distutils2/tests/test_upload_docs.py

diff --git a/src/distutils2/command/upload_docs.py b/src/distutils2/command/upload_docs.py
--- a/src/distutils2/command/upload_docs.py
+++ b/src/distutils2/command/upload_docs.py
@@ -1,6 +1,19 @@
-import os.path
+import os.path, tempfile, zipfile
 from distutils2.core import Command
 
+def zip_dir_into(directory, destination):
+    """Compresses recursively contents of directory into a zipfile located
+    under given destination.
+    """
+    zip_file = zipfile.ZipFile(destination, "w")
+    for root, dirs, files in os.walk(directory):
+        for name in files:
+            full = os.path.join(root, name)
+            relative = root[len(directory):].lstrip(os.path.sep)
+            dest = os.path.join(relative, name)
+            zip_file.write(full, dest)
+    zip_file.close()
+
 class upload_docs(Command):
 
     user_options = [
diff --git a/src/distutils2/tests/test_upload_docs.py b/src/distutils2/tests/test_upload_docs.py
--- a/src/distutils2/tests/test_upload_docs.py
+++ b/src/distutils2/tests/test_upload_docs.py
@@ -1,32 +1,52 @@
 """Tests for distutils.command.upload_docs."""
 # -*- encoding: utf8 -*-
-import os, os.path, unittest2
+import os, os.path, tempfile, unittest2, zipfile
 
-from distutils2.command.upload_docs import upload_docs
+from distutils2.command.upload_docs import upload_docs, zip_dir_into
 from distutils2.core import Distribution
 
+from distutils2.tests import support
 from distutils2.tests.pypi_server import PyPIServer
 from distutils2.tests.test_upload import PyPIServerTestCase
 from distutils2.tests.test_config import PYPIRC, PyPIRCCommandTestCase
 
 
-class UploadDocsTestCase(PyPIServerTestCase, PyPIRCCommandTestCase):
+class UploadDocsTestCase(PyPIServerTestCase,
+                         support.TempdirManager,
+                         unittest2.TestCase):
 
     def setUp(self):
         super(UploadDocsTestCase, self).setUp()
         self.dist = Distribution()
         self.cmd = upload_docs(self.dist)
 
-    def test_generates_uploaddir_if_none(self):
+    def test_generates_uploaddir_name_not_given(self):
         self.cmd.ensure_finalized()
         self.assertEqual(self.cmd.upload_dir, os.path.join("build", "docs"))
 
+    def test_zip_dir_into(self):
+        source_dir = tempfile.mkdtemp()
+        os.mkdir(os.path.join(source_dir, "some_dir"))
+        some_file = open(os.path.join(source_dir, "some_dir", "some_file"), "w")
+        some_file.write("Ce mortel ennui")
+        some_file.close()
+        dest_dir = tempfile.mkdtemp()
+        destination = os.path.join(dest_dir, "archive.zip")
+        zip_dir_into(source_dir, destination)
+
+        self.assertTrue(zipfile.is_zipfile(destination))
+        zip_f = zipfile.ZipFile(destination)
+        self.assertEqual(zip_f.namelist(), ['some_dir/some_file'])
+
     def test_zip_dir(self):
         pass
 
     def test_upload(self):
         pass
 
+    def test_honours_dry_run(self):
+        pass
+
 def test_suite():
     return unittest2.makeSuite(UploadDocsTestCase)
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: show_response option enabled
Message-ID: <E1OVLoV-0000hP-0Z@dinsdale.python.org>

tarek.ziade pushed 03bc7496c3b2 to distutils2:

http://hg.python.org/distutils2/rev/03bc7496c3b2
changeset:   278:03bc7496c3b2
user:        Konrad Delong <konryd at gmail.com>
date:        Tue Jun 01 17:09:32 2010 +0200
summary:     show_response option enabled
files:       src/distutils2/command/upload_docs.py, src/distutils2/tests/test_upload_docs.py

diff --git a/src/distutils2/command/upload_docs.py b/src/distutils2/command/upload_docs.py
--- a/src/distutils2/command/upload_docs.py
+++ b/src/distutils2/command/upload_docs.py
@@ -50,10 +50,8 @@
 class upload_docs(PyPIRCCommand):
 
     user_options = [
-        #('repository=', 'r',
-         #"url of repository [default: %s]" % upload.DEFAULT_REPOSITORY),
-        #('show-response', None,
-         #'display full response text from server'),
+        ('repository=', 'r', "url of repository [default: %s]" % upload.DEFAULT_REPOSITORY),
+        ('show-response', None, 'display full response text from server'),
         ('upload-dir=', None, 'directory to upload'),
         ]
 
@@ -132,3 +130,6 @@
         else:
             self.announce('Upload failed (%s): %s' % (r.status, r.reason),
                           log.ERROR)
+
+        if self.show_response:
+            print "\n".join(['-'*75, r.read(), '-'*75])
diff --git a/src/distutils2/tests/test_upload_docs.py b/src/distutils2/tests/test_upload_docs.py
--- a/src/distutils2/tests/test_upload_docs.py
+++ b/src/distutils2/tests/test_upload_docs.py
@@ -1,6 +1,6 @@
 """Tests for distutils.command.upload_docs."""
 # -*- encoding: utf8 -*-
-import httplib, os, os.path, shutil, tempfile, unittest2, zipfile
+import httplib, os, os.path, shutil, sys, tempfile, unittest2, zipfile
 from cStringIO import StringIO
 
 from distutils2.command import upload_docs as upload_docs_mod
@@ -174,6 +174,22 @@
         shutil.rmtree(os.path.join(self.cmd.upload_dir))
         self.assertRaises(DistutilsOptionError, self.cmd.ensure_finalized)
 
+    def test_show_response(self):
+        orig_stdout = sys.stdout
+        write_args = []
+        class MockStdIn(object):
+            def write(self, arg):
+                write_args.append(arg)
+        sys.stdout = MockStdIn()
+        try:
+            self.prepare_command()
+            self.cmd.show_response = True
+            self.cmd.run()
+        finally:
+            sys.stdout = orig_stdout
+        self.assertTrue(write_args[0], "should report the response")
+        self.assertIn("\n".join(self.pypi.default_response_data), write_args[0])
+
 def test_suite():
     return unittest2.makeSuite(UploadDocsTestCase)
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:38 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:38 +0200
Subject: [Python-checkins] distutils2: removed voodoo sleep
Message-ID: <E1OVLoU-0000gt-Su@dinsdale.python.org>

tarek.ziade pushed 5948b354eed6 to distutils2:

http://hg.python.org/distutils2/rev/5948b354eed6
changeset:   270:5948b354eed6
user:        Konrad Delong <konryd at gmail.com>
date:        Thu May 13 22:22:13 2010 +0200
summary:     removed voodoo sleep
files:       src/distutils2/tests/pypi_server.py

diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
--- a/src/distutils2/tests/pypi_server.py
+++ b/src/distutils2/tests/pypi_server.py
@@ -17,7 +17,6 @@
     def stop(self):
         self.httpd.shutdown()
         self.join()
-        time.sleep(0.2)
 
     def pypi_app(self, environ, start_response):
         status = '200 OK' # HTTP Status

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:38 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:38 +0200
Subject: [Python-checkins] distutils2: added pypi server documentation
Message-ID: <E1OVLoU-0000h9-Ul@dinsdale.python.org>

tarek.ziade pushed 3c1b00037fb0 to distutils2:

http://hg.python.org/distutils2/rev/3c1b00037fb0
changeset:   274:3c1b00037fb0
user:        Konrad Delong <konryd at gmail.com>
date:        Mon May 17 01:55:02 2010 +0200
summary:     added pypi server documentation
files:       docs/source/index.rst, docs/source/test_framework.rst, src/distutils2/tests/pypi_server.py

diff --git a/docs/source/index.rst b/docs/source/index.rst
--- a/docs/source/index.rst
+++ b/docs/source/index.rst
@@ -13,6 +13,7 @@
 
    metadata
    new_commands
+   test_framework
 
 Indices and tables
 ==================
diff --git a/docs/source/test_framework.rst b/docs/source/test_framework.rst
new file mode 100644
--- /dev/null
+++ b/docs/source/test_framework.rst
@@ -0,0 +1,34 @@
+==============
+Test Framework
+==============
+
+When you are testing code that works with distutils, you might find these tools
+useful.
+
+
+``PyPIServer``
+==============
+
+PyPIServer is a class that implements an HTTP server running in a separate
+thread. All it does is record the requests for further inspection. The recorded
+data is available under ``requests`` attribute. The default
+HTTP response can be overriden with the ``default_response_status``,
+``default_response_headers`` and ``default_response_data`` attributes.
+
+
+``PyPIServerTestCase``
+======================
+
+``PyPIServerTestCase`` is a test case class with setUp and tearDown methods that
+take care of a single PyPIServer instance attached as a ``pypi`` attribute on
+the test class. Use it as one of the base classes in you test case::
+
+  class UploadTestCase(PyPIServerTestCase):
+      def test_something(self):
+          cmd = self.prepare_command()
+          cmd.ensure_finalized()
+          cmd.repository = self.pypi.full_address
+          cmd.run()
+
+          environ, request_data = self.pypi.requests[-1]
+          self.assertEqual(request_data, EXPECTED_REQUEST_DATA)
diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
--- a/src/distutils2/tests/pypi_server.py
+++ b/src/distutils2/tests/pypi_server.py
@@ -21,9 +21,9 @@
         self.address = self.httpd.server_address
         self.request_queue = Queue.Queue()
         self._requests = []
-        self._default_response_status = "200 OK"
-        self._default_response_headers = [('Content-type', 'text/plain')]
-        self._default_response_data = ["hello"]
+        self.default_response_status = "200 OK"
+        self.default_response_headers = [('Content-type', 'text/plain')]
+        self.default_response_data = ["hello"]
 
     def run(self):
         self.httpd.serve_forever()
@@ -45,9 +45,9 @@
         return data
 
     def get_next_response(self):
-        return (self._default_response_status,
-                self._default_response_headers,
-                self._default_response_data)
+        return (self.default_response_status,
+                self.default_response_headers,
+                self.default_response_data)
 
     @property
     def requests(self):

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:38 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:38 +0200
Subject: [Python-checkins] distutils2: added encode_multipart
Message-ID: <E1OVLoU-0000gx-TN@dinsdale.python.org>

tarek.ziade pushed bffc38f3b3ad to distutils2:

http://hg.python.org/distutils2/rev/bffc38f3b3ad
changeset:   271:bffc38f3b3ad
user:        Konrad Delong <konryd at gmail.com>
date:        Thu May 13 23:20:53 2010 +0200
summary:     added encode_multipart
files:       src/distutils2/command/upload_docs.py, src/distutils2/tests/test_upload_docs.py

diff --git a/src/distutils2/command/upload_docs.py b/src/distutils2/command/upload_docs.py
--- a/src/distutils2/command/upload_docs.py
+++ b/src/distutils2/command/upload_docs.py
@@ -14,6 +14,38 @@
             zip_file.write(full, dest)
     zip_file.close()
 
+# grabbed from
+#    http://code.activestate.com/recipes/146306-http-client-to-post-using-multipartform-data/
+def encode_multipart(fields, files, boundary=None):
+    """
+    fields is a sequence of (name, value) elements for regular form fields.
+    files is a sequence of (name, filename, content_type, value)
+                            elements for data to be uploaded as files
+    Return (content_type, body) ready for httplib.HTTP instance
+    """
+    if boundary is None:
+        boundary = '--------------GHSKFJDLGDS7543FJKLFHRE75642756743254'
+    CRLF = '\r\n'
+    l = []
+    for (key, value) in fields:
+        l.extend([
+            '--' + boundary,
+            'Content-Disposition: form-data; name="%s"' % key,
+            '',
+            value])
+    for (key, filename, content_type, value) in files:
+        l.extend([
+            '--' + boundary,
+            'Content-Disposition: form-data; name="%s"; filename="%s"' % (key, filename),
+            'Content-Type: %s' % content_type,
+            '',
+            value])
+    l.append('--' + boundary + '--')
+    l.append('')
+    body = CRLF.join(l)
+    content_type = 'multipart/form-data; boundary=%s' % boundary
+    return content_type, body
+
 class upload_docs(Command):
 
     user_options = [
diff --git a/src/distutils2/tests/test_upload_docs.py b/src/distutils2/tests/test_upload_docs.py
--- a/src/distutils2/tests/test_upload_docs.py
+++ b/src/distutils2/tests/test_upload_docs.py
@@ -2,7 +2,8 @@
 # -*- encoding: utf8 -*-
 import os, os.path, tempfile, unittest2, zipfile
 
-from distutils2.command.upload_docs import upload_docs, zip_dir_into
+from distutils2.command.upload_docs import (upload_docs, zip_dir_into,
+                                    encode_multipart)
 from distutils2.core import Distribution
 
 from distutils2.tests import support
@@ -11,6 +12,29 @@
 from distutils2.tests.test_config import PYPIRC, PyPIRCCommandTestCase
 
 
+EXPECTED_MULTIPART_OUTPUT = "\r\n".join([
+'---x',
+'Content-Disposition: form-data; name="a"',
+'',
+'b',
+'---x',
+'Content-Disposition: form-data; name="c"',
+'',
+'d',
+'---x',
+'Content-Disposition: form-data; name="e"; filename="f"',
+'Content-Type: g',
+'',
+'h',
+'---x',
+'Content-Disposition: form-data; name="i"; filename="j"',
+'Content-Type: k',
+'',
+'l',
+'---x--',
+'',
+])
+
 class UploadDocsTestCase(PyPIServerTestCase,
                          support.TempdirManager,
                          unittest2.TestCase):
@@ -38,8 +62,11 @@
         zip_f = zipfile.ZipFile(destination)
         self.assertEqual(zip_f.namelist(), ['some_dir/some_file'])
 
-    def test_zip_dir(self):
-        pass
+    def test_encode_multipart(self):
+        fields = [("a", "b"), ("c", "d")]
+        files = [("e", "f", "g", "h"), ("i", "j", "k", "l")]
+        self.assertEqual(encode_multipart(fields, files, "-x")[0], "multipart/form-data; boundary=-x")
+        self.assertEqual(encode_multipart(fields, files, "-x")[1], EXPECTED_MULTIPART_OUTPUT)
 
     def test_upload(self):
         pass

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:38 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:38 +0200
Subject: [Python-checkins] distutils2: removed overengineered constant :)
Message-ID: <E1OVLoV-0000hM-03@dinsdale.python.org>

tarek.ziade pushed b2983b08480f to distutils2:

http://hg.python.org/distutils2/rev/b2983b08480f
changeset:   277:b2983b08480f
user:        Konrad Delong <konryd at gmail.com>
date:        Tue Jun 01 15:31:37 2010 +0200
summary:     removed overengineered constant :)
files:       src/distutils2/command/upload_docs.py

diff --git a/src/distutils2/command/upload_docs.py b/src/distutils2/command/upload_docs.py
--- a/src/distutils2/command/upload_docs.py
+++ b/src/distutils2/command/upload_docs.py
@@ -28,7 +28,6 @@
     """
     if boundary is None:
         boundary = '--------------GHSKFJDLGDS7543FJKLFHRE75642756743254'
-    CRLF = '\r\n'
     l = []
     for (key, value) in fields:
         l.extend([
@@ -44,7 +43,7 @@
             value])
     l.append('--' + boundary + '--')
     l.append('')
-    body = CRLF.join(l)
+    body =  '\r\n'.join(l)
     content_type = 'multipart/form-data; boundary=%s' % boundary
     return content_type, body
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:38 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:38 +0200
Subject: [Python-checkins] distutils2: added verification to finalize_options
Message-ID: <E1OVLoU-0000hI-Vq@dinsdale.python.org>

tarek.ziade pushed 4d2e03cda556 to distutils2:

http://hg.python.org/distutils2/rev/4d2e03cda556
changeset:   276:4d2e03cda556
user:        Konrad Delong <konryd at gmail.com>
date:        Tue Jun 01 15:28:49 2010 +0200
summary:     added verification to finalize_options
files:       src/distutils2/command/upload_docs.py, src/distutils2/tests/test_upload_docs.py

diff --git a/src/distutils2/command/upload_docs.py b/src/distutils2/command/upload_docs.py
--- a/src/distutils2/command/upload_docs.py
+++ b/src/distutils2/command/upload_docs.py
@@ -1,6 +1,7 @@
 import base64, httplib, os.path, socket, tempfile, urlparse, zipfile
 from cStringIO import StringIO
 from distutils2 import log
+from distutils2.command.upload import upload
 from distutils2.core import PyPIRCCommand
 from distutils2.errors import DistutilsFileError
 
@@ -50,17 +51,24 @@
 class upload_docs(PyPIRCCommand):
 
     user_options = [
+        #('repository=', 'r',
+         #"url of repository [default: %s]" % upload.DEFAULT_REPOSITORY),
+        #('show-response', None,
+         #'display full response text from server'),
         ('upload-dir=', None, 'directory to upload'),
         ]
 
     def initialize_options(self):
-        self.upload_dir = None
+        PyPIRCCommand.initialize_options(self)
+        self.upload_dir = "build/docs"
 
     def finalize_options(self):
         PyPIRCCommand.finalize_options(self)
         if self.upload_dir == None:
             build = self.get_finalized_command('build')
             self.upload_dir = os.path.join(build.build_base, "docs")
+        self.announce('Using upload directory %s' % self.upload_dir)
+        self.verify_upload_dir(self.upload_dir)
         config = self._read_pypirc()
         if config != {}:
             self.username = config['username']
@@ -69,16 +77,15 @@
             self.realm = config['realm']
 
     def verify_upload_dir(self, upload_dir):
+        self.ensure_dirname('upload_dir')
         index_location = os.path.join(upload_dir, "index.html")
         if not os.path.exists(index_location):
             mesg = "No 'index.html found in docs directory (%s)"
             raise DistutilsFileError(mesg % upload_dir)
 
-
     def run(self):
         tmp_dir = tempfile.mkdtemp()
         name = self.distribution.metadata['Name']
-        self.verify_upload_dir(self.upload_dir)
         zip_file = zip_dir(self.upload_dir)
 
         fields = {':action': 'doc_upload', 'name': name}.items()
diff --git a/src/distutils2/tests/test_upload_docs.py b/src/distutils2/tests/test_upload_docs.py
--- a/src/distutils2/tests/test_upload_docs.py
+++ b/src/distutils2/tests/test_upload_docs.py
@@ -1,6 +1,6 @@
 """Tests for distutils.command.upload_docs."""
 # -*- encoding: utf8 -*-
-import httplib, os, os.path, tempfile, unittest2, zipfile
+import httplib, os, os.path, shutil, tempfile, unittest2, zipfile
 from cStringIO import StringIO
 
 from distutils2.command import upload_docs as upload_docs_mod
@@ -8,7 +8,7 @@
                                     encode_multipart)
 from distutils2.core import Distribution
 
-from distutils2.errors import DistutilsFileError
+from distutils2.errors import DistutilsFileError, DistutilsOptionError
 
 from distutils2.tests import support
 from distutils2.tests.pypi_server import PyPIServer, PyPIServerTestCase
@@ -54,15 +54,23 @@
         self.dist.metadata['Name'] = "distr-name"
         self.cmd = upload_docs(self.dist)
 
-    def test_generates_uploaddir_if_not_given(self):
-        self.assertEqual(self.cmd.upload_dir, None)
-        self.cmd.ensure_finalized()
-        self.assertEqual(self.cmd.upload_dir, os.path.join("build", "docs"))
+    def test_default_uploaddir(self):
+        sandbox = tempfile.mkdtemp()
+        previous = os.getcwd()
+        os.chdir(sandbox)
+        try:
+            os.mkdir("build")
+            self.prepare_sample_dir("build")
+            self.cmd.ensure_finalized()
+            self.assertEqual(self.cmd.upload_dir, os.path.join("build", "docs"))
+        finally:
+            os.chdir(previous)
 
-    def prepare_sample_dir(self):
-        sample_dir = tempfile.mkdtemp()
-        os.mkdir(os.path.join(sample_dir, "some_dir"))
-        self.write_file(os.path.join(sample_dir, "some_dir", "index.html"), "Ce mortel ennui")
+    def prepare_sample_dir(self, sample_dir=None):
+        if sample_dir is None:
+            sample_dir = tempfile.mkdtemp()
+        os.mkdir(os.path.join(sample_dir, "docs"))
+        self.write_file(os.path.join(sample_dir, "docs", "index.html"), "Ce mortel ennui")
         self.write_file(os.path.join(sample_dir, "index.html"), "Oh la la")
         return sample_dir
 
@@ -71,7 +79,7 @@
         compressed = zip_dir(source_dir)
 
         zip_f = zipfile.ZipFile(compressed)
-        self.assertEqual(zip_f.namelist(), ['index.html', 'some_dir/index.html'])
+        self.assertEqual(zip_f.namelist(), ['index.html', 'docs/index.html'])
 
     def test_encode_multipart(self):
         fields = [("a", "b"), ("c", "d")]
@@ -81,9 +89,9 @@
         self.assertEqual(body, EXPECTED_MULTIPART_OUTPUT)
 
     def prepare_command(self):
+        self.cmd.upload_dir = self.prepare_sample_dir()
         self.cmd.ensure_finalized()
         self.cmd.repository = self.pypi.full_address
-        self.cmd.upload_dir = self.prepare_sample_dir()
         self.cmd.username = "username"
         self.cmd.password = "password"
 
@@ -109,7 +117,7 @@
         # check their contents
         self.assertIn("doc_upload", action)
         self.assertIn("distr-name", name)
-        self.assertIn("some_dir/index.html", content)
+        self.assertIn("docs/index.html", content)
         self.assertIn("Ce mortel ennui", content)
 
     def test_https_connection(self):
@@ -151,14 +159,20 @@
     def test_reads_pypirc_data(self):
         self.write_file(self.rc, PYPIRC % self.pypi.full_address)
         self.cmd.repository = self.pypi.full_address
+        self.cmd.upload_dir = self.prepare_sample_dir()
         self.cmd.ensure_finalized()
         self.assertEqual(self.cmd.username, "real_slim_shady")
         self.assertEqual(self.cmd.password, "long_island")
 
     def test_checks_index_html_presence(self):
-        self.prepare_command()
+        self.cmd.upload_dir = self.prepare_sample_dir()
         os.remove(os.path.join(self.cmd.upload_dir, "index.html"))
-        self.assertRaises(DistutilsFileError, self.cmd.run)
+        self.assertRaises(DistutilsFileError, self.cmd.ensure_finalized)
+
+    def test_checks_upload_dir(self):
+        self.cmd.upload_dir = self.prepare_sample_dir()
+        shutil.rmtree(os.path.join(self.cmd.upload_dir))
+        self.assertRaises(DistutilsOptionError, self.cmd.ensure_finalized)
 
 def test_suite():
     return unittest2.makeSuite(UploadDocsTestCase)

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:38 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:38 +0200
Subject: [Python-checkins] distutils2: other schemas support and informing
	about errors
Message-ID: <E1OVLoU-0000hE-VH@dinsdale.python.org>

tarek.ziade pushed 944a00b90bde to distutils2:

http://hg.python.org/distutils2/rev/944a00b90bde
changeset:   275:944a00b90bde
user:        Konrad Delong <konryd at gmail.com>
date:        Mon May 17 16:58:40 2010 +0200
summary:     other schemas support and informing about errors
files:       src/distutils2/command/upload_docs.py, src/distutils2/tests/test_upload_docs.py

diff --git a/src/distutils2/command/upload_docs.py b/src/distutils2/command/upload_docs.py
--- a/src/distutils2/command/upload_docs.py
+++ b/src/distutils2/command/upload_docs.py
@@ -1,7 +1,8 @@
-import base64, httplib, os.path, tempfile, urlparse, zipfile
+import base64, httplib, os.path, socket, tempfile, urlparse, zipfile
 from cStringIO import StringIO
 from distutils2 import log
-from distutils2.core import Command
+from distutils2.core import PyPIRCCommand
+from distutils2.errors import DistutilsFileError
 
 def zip_dir(directory):
     """Compresses recursively contents of directory into a StringIO object"""
@@ -46,7 +47,7 @@
     content_type = 'multipart/form-data; boundary=%s' % boundary
     return content_type, body
 
-class upload_docs(Command):
+class upload_docs(PyPIRCCommand):
 
     user_options = [
         ('upload-dir=', None, 'directory to upload'),
@@ -56,13 +57,28 @@
         self.upload_dir = None
 
     def finalize_options(self):
+        PyPIRCCommand.finalize_options(self)
         if self.upload_dir == None:
             build = self.get_finalized_command('build')
             self.upload_dir = os.path.join(build.build_base, "docs")
+        config = self._read_pypirc()
+        if config != {}:
+            self.username = config['username']
+            self.password = config['password']
+            self.repository = config['repository']
+            self.realm = config['realm']
+
+    def verify_upload_dir(self, upload_dir):
+        index_location = os.path.join(upload_dir, "index.html")
+        if not os.path.exists(index_location):
+            mesg = "No 'index.html found in docs directory (%s)"
+            raise DistutilsFileError(mesg % upload_dir)
+
 
     def run(self):
         tmp_dir = tempfile.mkdtemp()
         name = self.distribution.metadata['Name']
+        self.verify_upload_dir(self.upload_dir)
         zip_file = zip_dir(self.upload_dir)
 
         fields = {':action': 'doc_upload', 'name': name}.items()
@@ -77,7 +93,12 @@
 
         schema, netloc, url, params, query, fragments = \
             urlparse.urlparse(self.repository)
-        conn = httplib.HTTPConnection(netloc)
+        if schema == "http":
+            conn = httplib.HTTPConnection(netloc)
+        elif schema == "https":
+            conn = httplib.HTTPSConnection(netloc)
+        else:
+            raise AssertionError("unsupported schema "+schema)
 
         try:
             conn.connect()
@@ -92,3 +113,16 @@
             return
 
         r = conn.getresponse()
+
+        if r.status == 200:
+            self.announce('Server response (%s): %s' % (r.status, r.reason),
+                          log.INFO)
+        elif r.status == 301:
+            location = r.getheader('Location')
+            if location is None:
+                location = 'http://packages.python.org/%s/' % meta.get_name()
+            self.announce('Upload successful. Visit %s' % location,
+                          log.INFO)
+        else:
+            self.announce('Upload failed (%s): %s' % (r.status, r.reason),
+                          log.ERROR)
diff --git a/src/distutils2/tests/test_upload_docs.py b/src/distutils2/tests/test_upload_docs.py
--- a/src/distutils2/tests/test_upload_docs.py
+++ b/src/distutils2/tests/test_upload_docs.py
@@ -1,15 +1,18 @@
 """Tests for distutils.command.upload_docs."""
 # -*- encoding: utf8 -*-
-import os, os.path, tempfile, unittest2, zipfile
+import httplib, os, os.path, tempfile, unittest2, zipfile
 from cStringIO import StringIO
 
+from distutils2.command import upload_docs as upload_docs_mod
 from distutils2.command.upload_docs import (upload_docs, zip_dir,
                                     encode_multipart)
 from distutils2.core import Distribution
 
+from distutils2.errors import DistutilsFileError
+
 from distutils2.tests import support
 from distutils2.tests.pypi_server import PyPIServer, PyPIServerTestCase
-from distutils2.tests.test_config import PYPIRC, PyPIRCCommandTestCase
+from distutils2.tests.test_config import PyPIRCCommandTestCase
 
 
 EXPECTED_MULTIPART_OUTPUT = "\r\n".join([
@@ -33,9 +36,17 @@
 '',
 ])
 
-class UploadDocsTestCase(PyPIServerTestCase,
-                         support.TempdirManager,
-                         unittest2.TestCase):
+PYPIRC = """\
+[distutils]
+index-servers = server1
+
+[server1]
+repository = %s
+username = real_slim_shady
+password = long_island
+"""
+
+class UploadDocsTestCase(PyPIServerTestCase, PyPIRCCommandTestCase):
 
     def setUp(self):
         super(UploadDocsTestCase, self).setUp()
@@ -43,16 +54,16 @@
         self.dist.metadata['Name'] = "distr-name"
         self.cmd = upload_docs(self.dist)
 
-    def test_generates_uploaddir_name_not_given(self):
+    def test_generates_uploaddir_if_not_given(self):
+        self.assertEqual(self.cmd.upload_dir, None)
         self.cmd.ensure_finalized()
         self.assertEqual(self.cmd.upload_dir, os.path.join("build", "docs"))
 
     def prepare_sample_dir(self):
         sample_dir = tempfile.mkdtemp()
         os.mkdir(os.path.join(sample_dir, "some_dir"))
-        some_file = open(os.path.join(sample_dir, "some_dir", "some_file"), "w")
-        some_file.write("Ce mortel ennui")
-        some_file.close()
+        self.write_file(os.path.join(sample_dir, "some_dir", "index.html"), "Ce mortel ennui")
+        self.write_file(os.path.join(sample_dir, "index.html"), "Oh la la")
         return sample_dir
 
     def test_zip_dir(self):
@@ -60,7 +71,7 @@
         compressed = zip_dir(source_dir)
 
         zip_f = zipfile.ZipFile(compressed)
-        self.assertEqual(zip_f.namelist(), ['some_dir/some_file'])
+        self.assertEqual(zip_f.namelist(), ['index.html', 'some_dir/index.html'])
 
     def test_encode_multipart(self):
         fields = [("a", "b"), ("c", "d")]
@@ -69,12 +80,15 @@
         self.assertEqual(content_type, "multipart/form-data; boundary=-x")
         self.assertEqual(body, EXPECTED_MULTIPART_OUTPUT)
 
-    def test_upload(self):
+    def prepare_command(self):
         self.cmd.ensure_finalized()
         self.cmd.repository = self.pypi.full_address
         self.cmd.upload_dir = self.prepare_sample_dir()
         self.cmd.username = "username"
         self.cmd.password = "password"
+
+    def test_upload(self):
+        self.prepare_command()
         self.cmd.run()
 
         self.assertEqual(len(self.pypi.requests), 1)
@@ -95,20 +109,56 @@
         # check their contents
         self.assertIn("doc_upload", action)
         self.assertIn("distr-name", name)
-        self.assertIn("some_dir/some_file", content)
+        self.assertIn("some_dir/index.html", content)
         self.assertIn("Ce mortel ennui", content)
 
     def test_https_connection(self):
-        pass
+        https_called = False
+        orig_https = upload_docs_mod.httplib.HTTPSConnection
+        def https_conn_wrapper(*args):
+            https_called = True
+            return upload_docs_mod.httplib.HTTPConnection(*args) # the testing server is http
+        upload_docs_mod.httplib.HTTPSConnection = https_conn_wrapper
+        try:
+            self.prepare_command()
+            self.cmd.run()
+            self.assertFalse(https_called)
+
+            self.cmd.repository = self.cmd.repository.replace("http", "https")
+            self.cmd.run()
+            self.assertFalse(https_called)
+        finally:
+            upload_docs_mod.httplib.HTTPSConnection = orig_https
 
     def test_handling_response(self):
-        pass
+        calls = []
+        def aggr(*args):
+            calls.append(args)
+        self.pypi.default_response_status = '403 Forbidden'
+        self.prepare_command()
+        self.cmd.announce = aggr
+        self.cmd.run()
+        message, _ = calls[-1]
+        self.assertIn('Upload failed (403): Forbidden', message)
 
-    def test_honours_dry_run(self):
-        pass
+        calls = []
+        self.pypi.default_response_status = '301 Moved Permanently'
+        self.pypi.default_response_headers.append(("Location", "brand_new_location"))
+        self.cmd.run()
+        message, _ = calls[-1]
+        self.assertIn('brand_new_location', message)
 
     def test_reads_pypirc_data(self):
-        pass
+        self.write_file(self.rc, PYPIRC % self.pypi.full_address)
+        self.cmd.repository = self.pypi.full_address
+        self.cmd.ensure_finalized()
+        self.assertEqual(self.cmd.username, "real_slim_shady")
+        self.assertEqual(self.cmd.password, "long_island")
+
+    def test_checks_index_html_presence(self):
+        self.prepare_command()
+        os.remove(os.path.join(self.cmd.upload_dir, "index.html"))
+        self.assertRaises(DistutilsFileError, self.cmd.run)
 
 def test_suite():
     return unittest2.makeSuite(UploadDocsTestCase)

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: typo fix in Metadata documentation
Message-ID: <E1OVLoV-0000hX-5Z@dinsdale.python.org>

tarek.ziade pushed 53f360a56b5a to distutils2:

http://hg.python.org/distutils2/rev/53f360a56b5a
changeset:   280:53f360a56b5a
parent:      160:30ad3af43b5f
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Mon May 24 10:00:10 2010 +0200
summary:     typo fix in Metadata documentation
files:       docs/source/metadata.rst

diff --git a/docs/source/metadata.rst b/docs/source/metadata.rst
--- a/docs/source/metadata.rst
+++ b/docs/source/metadata.rst
@@ -71,8 +71,8 @@
     >>> metadata.write('/to/my/PKG-INFO')
 
 The class will pick the best version for the metadata, depending on the values
-provided. If all the values provided exists in all versions, teh class will
-used :attr:`metadata.PKG_INFO_PREFERRED_VERSION`. It is set by default to 1.0.
+provided. If all the values provided exists in all versions, the class will
+use :attr:`metadata.PKG_INFO_PREFERRED_VERSION`. It is set by default to 1.0.
 
 
 Conflict checking and best version

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: Add build path to hgignore
Message-ID: <E1OVLoV-0000hb-64@dinsdale.python.org>

tarek.ziade pushed fdac92d97dcc to distutils2:

http://hg.python.org/distutils2/rev/fdac92d97dcc
changeset:   281:fdac92d97dcc
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Thu May 27 12:28:09 2010 +0200
summary:     Add build path to hgignore
files:       .hgignore

diff --git a/.hgignore b/.hgignore
--- a/.hgignore
+++ b/.hgignore
@@ -2,4 +2,5 @@
 .*\.pyo$
 ^docs/build
 .*\.swp$
+src/build/*
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: fixing typos while reading
Message-ID: <E1OVLoV-0000hg-6j@dinsdale.python.org>

tarek.ziade pushed e34cd12c1cef to distutils2:

http://hg.python.org/distutils2/rev/e34cd12c1cef
changeset:   282:e34cd12c1cef
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Thu May 27 12:29:02 2010 +0200
summary:     fixing typos while reading
files:       src/distutils2/cmd.py, src/distutils2/dist.py

diff --git a/src/distutils2/cmd.py b/src/distutils2/cmd.py
--- a/src/distutils2/cmd.py
+++ b/src/distutils2/cmd.py
@@ -57,8 +57,7 @@
     def __init__(self, dist):
         """Create and initialize a new Command object.  Most importantly,
         invokes the 'initialize_options()' method, which is the real
-        initializer and depends on the actual command being
-        instantiated.
+        initializer and depends on the actual command being instantiated.
         """
         # late import because of mutual dependence between these classes
         from distutils2.dist import Distribution
diff --git a/src/distutils2/dist.py b/src/distutils2/dist.py
--- a/src/distutils2/dist.py
+++ b/src/distutils2/dist.py
@@ -144,8 +144,8 @@
         # information here (and enough command-line options) that it's
         # worth it.  Also delegate 'get_XXX()' methods to the 'metadata'
         # object in a sneaky and underhanded (but efficient!) way.
+        self.metadata = DistributionMetadata()
 
-        self.metadata = DistributionMetadata()
         #for basename in self.metadata._METHOD_BASENAMES:
         #    method_name = "get_" + basename
         #    setattr(self, method_name, getattr(self.metadata, method_name))

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: Don't forget to commit the files
	related to PyPIServer	tests.
Message-ID: <E1OVLoV-0000hv-9X@dinsdale.python.org>

tarek.ziade pushed 119680b67694 to distutils2:

http://hg.python.org/distutils2/rev/119680b67694
changeset:   286:119680b67694
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Tue Jun 01 09:21:53 2010 +0200
summary:     Don't forget to commit the files related to PyPIServer tests.
files:       src/distutils2/tests/pypiserver/index.html

diff --git a/src/distutils2/tests/pypiserver/index.html b/src/distutils2/tests/pypiserver/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/index.html
@@ -0,0 +1,1 @@
+Yeah

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: Update .hgignore
Message-ID: <E1OVLoV-0000hz-A3@dinsdale.python.org>

tarek.ziade pushed 6577b806d932 to distutils2:

http://hg.python.org/distutils2/rev/6577b806d932
changeset:   287:6577b806d932
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Tue Jun 01 09:26:22 2010 +0200
summary:     Update .hgignore
files:       .hgignore

diff --git a/.hgignore b/.hgignore
--- a/.hgignore
+++ b/.hgignore
@@ -3,4 +3,5 @@
 ^docs/build
 .*\.swp$
 src/build/*
+.*\.pida-metadata
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: create pypi submodule and import
	package_index.py from	distribute to it
Message-ID: <E1OVLoV-0000hj-7i@dinsdale.python.org>

tarek.ziade pushed d6099bdc57cd to distutils2:

http://hg.python.org/distutils2/rev/d6099bdc57cd
changeset:   283:d6099bdc57cd
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Thu May 27 12:30:12 2010 +0200
summary:     create pypi submodule and import package_index.py from distribute to it
files:       src/distutils2/pypi/__init__.py, src/distutils2/pypi/package_index.py

diff --git a/src/distutils2/pypi/__init__.py b/src/distutils2/pypi/__init__.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/pypi/__init__.py
@@ -0,0 +1,7 @@
+"""distutils2.pypi
+
+Package containing ways to interact with the PyPI APIs.
+""" 
+
+__all__ = ['package_index',
+}
diff --git a/src/distutils2/pypi/package_index.py b/src/distutils2/pypi/package_index.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/pypi/package_index.py
@@ -0,0 +1,823 @@
+"""PyPI and direct package downloading"""
+import sys, os.path, re, urlparse, urllib2, shutil, random, socket, cStringIO
+import httplib
+from pkg_resources import *
+from distutils import log
+from distutils.errors import DistutilsError
+try:
+    from hashlib import md5
+except ImportError:
+    from md5 import md5
+from fnmatch import translate
+
+EGG_FRAGMENT = re.compile(r'^egg=([-A-Za-z0-9_.]+)$')
+HREF = re.compile("""href\\s*=\\s*['"]?([^'"> ]+)""", re.I)
+# this is here to fix emacs' cruddy broken syntax highlighting
+PYPI_MD5 = re.compile(
+    '<a href="([^"#]+)">([^<]+)</a>\n\s+\\(<a (?:title="MD5 hash"\n\s+)'
+    'href="[^?]+\?:action=show_md5&amp;digest=([0-9a-f]{32})">md5</a>\\)'
+)
+URL_SCHEME = re.compile('([-+.a-z0-9]{2,}):',re.I).match
+EXTENSIONS = ".tar.gz .tar.bz2 .tar .zip .tgz".split()
+
+__all__ = [
+    'PackageIndex', 'distros_for_url', 'parse_bdist_wininst',
+    'interpret_distro_name',
+]
+
+_SOCKET_TIMEOUT = 15
+
+def parse_bdist_wininst(name):
+    """Return (base,pyversion) or (None,None) for possible .exe name"""
+
+    lower = name.lower()
+    base, py_ver = None, None
+
+    if lower.endswith('.exe'):
+        if lower.endswith('.win32.exe'):
+            base = name[:-10]
+        elif lower.startswith('.win32-py',-16):
+            py_ver = name[-7:-4]
+            base = name[:-16]
+
+    return base,py_ver
+
+def egg_info_for_url(url):
+    scheme, server, path, parameters, query, fragment = urlparse.urlparse(url)
+    base = urllib2.unquote(path.split('/')[-1])
+    if '#' in base: base, fragment = base.split('#',1)
+    return base,fragment
+
+def distros_for_url(url, metadata=None):
+    """Yield egg or source distribution objects that might be found at a URL"""
+    base, fragment = egg_info_for_url(url)
+    for dist in distros_for_location(url, base, metadata): yield dist
+    if fragment:
+        match = EGG_FRAGMENT.match(fragment)
+        if match:
+            for dist in interpret_distro_name(
+                url, match.group(1), metadata, precedence = CHECKOUT_DIST
+            ):
+                yield dist
+
+def distros_for_location(location, basename, metadata=None):
+    """Yield egg or source distribution objects based on basename"""
+    if basename.endswith('.egg.zip'):
+        basename = basename[:-4]    # strip the .zip
+    if basename.endswith('.egg') and '-' in basename:
+        # only one, unambiguous interpretation
+        return [Distribution.from_location(location, basename, metadata)]
+
+    if basename.endswith('.exe'):
+        win_base, py_ver = parse_bdist_wininst(basename)
+        if win_base is not None:
+            return interpret_distro_name(
+                location, win_base, metadata, py_ver, BINARY_DIST, "win32"
+            )
+
+    # Try source distro extensions (.zip, .tgz, etc.)
+    #
+    for ext in EXTENSIONS:
+        if basename.endswith(ext):
+            basename = basename[:-len(ext)]
+            return interpret_distro_name(location, basename, metadata)
+    return []  # no extension matched
+
+def distros_for_filename(filename, metadata=None):
+    """Yield possible egg or source distribution objects based on a filename"""
+    return distros_for_location(
+        normalize_path(filename), os.path.basename(filename), metadata
+    )
+
+
+def interpret_distro_name(location, basename, metadata,
+    py_version=None, precedence=SOURCE_DIST, platform=None
+):
+    """Generate alternative interpretations of a source distro name
+
+    Note: if `location` is a filesystem filename, you should call
+    ``pkg_resources.normalize_path()`` on it before passing it to this
+    routine!
+    """
+    # Generate alternative interpretations of a source distro name
+    # Because some packages are ambiguous as to name/versions split
+    # e.g. "adns-python-1.1.0", "egenix-mx-commercial", etc.
+    # So, we generate each possible interepretation (e.g. "adns, python-1.1.0"
+    # "adns-python, 1.1.0", and "adns-python-1.1.0, no version").  In practice,
+    # the spurious interpretations should be ignored, because in the event
+    # there's also an "adns" package, the spurious "python-1.1.0" version will
+    # compare lower than any numeric version number, and is therefore unlikely
+    # to match a request for it.  It's still a potential problem, though, and
+    # in the long run PyPI and the distutils should go for "safe" names and
+    # versions in distribution archive names (sdist and bdist).
+
+    parts = basename.split('-')
+    if not py_version:
+        for i,p in enumerate(parts[2:]):
+            if len(p)==5 and p.startswith('py2.'):
+                return # It's a bdist_dumb, not an sdist -- bail out
+
+    for p in range(1,len(parts)+1):
+        yield Distribution(
+            location, metadata, '-'.join(parts[:p]), '-'.join(parts[p:]),
+            py_version=py_version, precedence = precedence,
+            platform = platform
+        )
+
+REL = re.compile("""<([^>]*\srel\s*=\s*['"]?([^'">]+)[^>]*)>""", re.I)
+# this line is here to fix emacs' cruddy broken syntax highlighting
+
+def find_external_links(url, page):
+    """Find rel="homepage" and rel="download" links in `page`, yielding URLs"""
+
+    for match in REL.finditer(page):
+        tag, rel = match.groups()
+        rels = map(str.strip, rel.lower().split(','))
+        if 'homepage' in rels or 'download' in rels:
+            for match in HREF.finditer(tag):
+                yield urlparse.urljoin(url, htmldecode(match.group(1)))
+
+    for tag in ("<th>Home Page", "<th>Download URL"):
+        pos = page.find(tag)
+        if pos!=-1:
+            match = HREF.search(page,pos)
+            if match:
+                yield urlparse.urljoin(url, htmldecode(match.group(1)))
+
+user_agent = "Python-urllib/%s distribute/%s" % (
+    sys.version[:3], require('distribute')[0].version
+)
+
+
+class PackageIndex(Environment):
+    """A distribution index that scans web pages for download URLs"""
+
+    def __init__(self, index_url="http://pypi.python.org/simple", hosts=('*',),
+        *args, **kw
+    ):
+        Environment.__init__(self,*args,**kw)
+        self.index_url = index_url + "/"[:not index_url.endswith('/')]
+        self.scanned_urls = {}
+        self.fetched_urls = {}
+        self.package_pages = {}
+        self.allows = re.compile('|'.join(map(translate,hosts))).match
+        self.to_scan = []
+
+
+
+    def process_url(self, url, retrieve=False):
+        """Evaluate a URL as a possible download, and maybe retrieve it"""
+        if url in self.scanned_urls and not retrieve:
+            return
+        self.scanned_urls[url] = True
+        if not URL_SCHEME(url):
+            self.process_filename(url)
+            return
+        else:
+            dists = list(distros_for_url(url))
+            if dists:
+                if not self.url_ok(url):
+                    return
+                self.debug("Found link: %s", url)
+
+        if dists or not retrieve or url in self.fetched_urls:
+            map(self.add, dists)
+            return  # don't need the actual page
+
+        if not self.url_ok(url):
+            self.fetched_urls[url] = True
+            return
+
+        self.info("Reading %s", url)
+        f = self.open_url(url, "Download error: %s -- Some packages may not be found!")
+        if f is None: return
+        self.fetched_urls[url] = self.fetched_urls[f.url] = True
+
+        if 'html' not in f.headers.get('content-type', '').lower():
+            f.close()   # not html, we can't process it
+            return
+
+        base = f.url     # handle redirects
+        page = f.read()
+        if sys.version_info >= (3,):
+            charset = f.headers.get_param('charset') or 'latin-1'
+            page = page.decode(charset, "ignore")
+        f.close()
+        for match in HREF.finditer(page):
+            link = urlparse.urljoin(base, htmldecode(match.group(1)))
+            self.process_url(link)
+        if url.startswith(self.index_url) and getattr(f,'code',None)!=404:
+            page = self.process_index(url, page)
+
+    def process_filename(self, fn, nested=False):
+        # process filenames or directories
+        if not os.path.exists(fn):
+            self.warn("Not found: %s", fn)
+            return
+
+        if os.path.isdir(fn) and not nested:
+            path = os.path.realpath(fn)
+            for item in os.listdir(path):
+                self.process_filename(os.path.join(path,item), True)
+
+        dists = distros_for_filename(fn)
+        if dists:
+            self.debug("Found: %s", fn)
+            map(self.add, dists)
+
+    def url_ok(self, url, fatal=False):
+        s = URL_SCHEME(url)
+        if (s and s.group(1).lower()=='file') or self.allows(urlparse.urlparse(url)[1]):
+            return True
+        msg = "\nLink to % s ***BLOCKED*** by --allow-hosts\n"
+        if fatal:
+            raise DistutilsError(msg % url)
+        else:
+            self.warn(msg, url)
+
+    def scan_egg_links(self, search_path):
+        for item in search_path:
+            if os.path.isdir(item):
+                for entry in os.listdir(item):
+                    if entry.endswith('.egg-link'):
+                        self.scan_egg_link(item, entry)
+
+    def scan_egg_link(self, path, entry):
+        lines = filter(None, map(str.strip, open(os.path.join(path, entry))))
+        if len(lines)==2:
+            for dist in find_distributions(os.path.join(path, lines[0])):
+                dist.location = os.path.join(path, *lines)
+                dist.precedence = SOURCE_DIST
+                self.add(dist)
+
+    def process_index(self,url,page):
+        """Process the contents of a PyPI page"""
+        def scan(link):
+            # Process a URL to see if it's for a package page
+            if link.startswith(self.index_url):
+                parts = map(
+                    urllib2.unquote, link[len(self.index_url):].split('/')
+                )
+                if len(parts)==2 and '#' not in parts[1]:
+                    # it's a package page, sanitize and index it
+                    pkg = safe_name(parts[0])
+                    ver = safe_version(parts[1])
+                    self.package_pages.setdefault(pkg.lower(),{})[link] = True
+                    return to_filename(pkg), to_filename(ver)
+            return None, None
+
+        # process an index page into the package-page index
+        for match in HREF.finditer(page):
+            try:
+                scan( urlparse.urljoin(url, htmldecode(match.group(1))) )
+            except ValueError:
+                pass
+
+        pkg, ver = scan(url)   # ensure this page is in the page index
+        if pkg:
+            # process individual package page
+            for new_url in find_external_links(url, page):
+                # Process the found URL
+                base, frag = egg_info_for_url(new_url)
+                if base.endswith('.py') and not frag:
+                    if ver:
+                        new_url+='#egg=%s-%s' % (pkg,ver)
+                    else:
+                        self.need_version_info(url)
+                self.scan_url(new_url)
+
+            return PYPI_MD5.sub(
+                lambda m: '<a href="%s#md5=%s">%s</a>' % m.group(1,3,2), page
+            )
+        else:
+            return ""   # no sense double-scanning non-package pages
+
+
+
+    def need_version_info(self, url):
+        self.scan_all(
+            "Page at %s links to .py file(s) without version info; an index "
+            "scan is required.", url
+        )
+
+    def scan_all(self, msg=None, *args):
+        if self.index_url not in self.fetched_urls:
+            if msg: self.warn(msg,*args)
+            self.info(
+                "Scanning index of all packages (this may take a while)"
+            )
+        self.scan_url(self.index_url)
+
+    def find_packages(self, requirement):
+        self.scan_url(self.index_url + requirement.unsafe_name+'/')
+
+        if not self.package_pages.get(requirement.key):
+            # Fall back to safe version of the name
+            self.scan_url(self.index_url + requirement.project_name+'/')
+
+        if not self.package_pages.get(requirement.key):
+            # We couldn't find the target package, so search the index page too
+            self.not_found_in_index(requirement)
+
+        for url in list(self.package_pages.get(requirement.key,())):
+            # scan each page that might be related to the desired package
+            self.scan_url(url)
+
+    def obtain(self, requirement, installer=None):
+        self.prescan(); self.find_packages(requirement)
+        for dist in self[requirement.key]:
+            if dist in requirement:
+                return dist
+            self.debug("%s does not match %s", requirement, dist)
+        return super(PackageIndex, self).obtain(requirement,installer)
+
+
+
+
+
+    def check_md5(self, cs, info, filename, tfp):
+        if re.match('md5=[0-9a-f]{32}$', info):
+            self.debug("Validating md5 checksum for %s", filename)
+            if cs.hexdigest()<>info[4:]:
+                tfp.close()
+                os.unlink(filename)
+                raise DistutilsError(
+                    "MD5 validation failed for "+os.path.basename(filename)+
+                    "; possible download problem?"
+                )
+
+    def add_find_links(self, urls):
+        """Add `urls` to the list that will be prescanned for searches"""
+        for url in urls:
+            if (
+                self.to_scan is None        # if we have already "gone online"
+                or not URL_SCHEME(url)      # or it's a local file/directory
+                or url.startswith('file:')
+                or list(distros_for_url(url))   # or a direct package link
+            ):
+                # then go ahead and process it now
+                self.scan_url(url)
+            else:
+                # otherwise, defer retrieval till later
+                self.to_scan.append(url)
+
+    def prescan(self):
+        """Scan urls scheduled for prescanning (e.g. --find-links)"""
+        if self.to_scan:
+            map(self.scan_url, self.to_scan)
+        self.to_scan = None     # from now on, go ahead and process immediately
+
+    def not_found_in_index(self, requirement):
+        if self[requirement.key]:   # we've seen at least one distro
+            meth, msg = self.info, "Couldn't retrieve index page for %r"
+        else:   # no distros seen for this name, might be misspelled
+            meth, msg = (self.warn,
+                "Couldn't find index page for %r (maybe misspelled?)")
+        meth(msg, requirement.unsafe_name)
+        self.scan_all()
+
+    def download(self, spec, tmpdir):
+        """Locate and/or download `spec` to `tmpdir`, returning a local path
+
+        `spec` may be a ``Requirement`` object, or a string containing a URL,
+        an existing local filename, or a project/version requirement spec
+        (i.e. the string form of a ``Requirement`` object).  If it is the URL
+        of a .py file with an unambiguous ``#egg=name-version`` tag (i.e., one
+        that escapes ``-`` as ``_`` throughout), a trivial ``setup.py`` is
+        automatically created alongside the downloaded file.
+
+        If `spec` is a ``Requirement`` object or a string containing a
+        project/version requirement spec, this method returns the location of
+        a matching distribution (possibly after downloading it to `tmpdir`).
+        If `spec` is a locally existing file or directory name, it is simply
+        returned unchanged.  If `spec` is a URL, it is downloaded to a subpath
+        of `tmpdir`, and the local filename is returned.  Various errors may be
+        raised if a problem occurs during downloading.
+        """
+        if not isinstance(spec,Requirement):
+            scheme = URL_SCHEME(spec)
+            if scheme:
+                # It's a url, download it to tmpdir
+                found = self._download_url(scheme.group(1), spec, tmpdir)
+                base, fragment = egg_info_for_url(spec)
+                if base.endswith('.py'):
+                    found = self.gen_setup(found,fragment,tmpdir)
+                return found
+            elif os.path.exists(spec):
+                # Existing file or directory, just return it
+                return spec
+            else:
+                try:
+                    spec = Requirement.parse(spec)
+                except ValueError:
+                    raise DistutilsError(
+                        "Not a URL, existing file, or requirement spec: %r" %
+                        (spec,)
+                    )
+        return getattr(self.fetch_distribution(spec, tmpdir),'location',None)
+
+
+    def fetch_distribution(self,
+        requirement, tmpdir, force_scan=False, source=False, develop_ok=False
+    ):
+        """Obtain a distribution suitable for fulfilling `requirement`
+
+        `requirement` must be a ``pkg_resources.Requirement`` instance.
+        If necessary, or if the `force_scan` flag is set, the requirement is
+        searched for in the (online) package index as well as the locally
+        installed packages.  If a distribution matching `requirement` is found,
+        the returned distribution's ``location`` is the value you would have
+        gotten from calling the ``download()`` method with the matching
+        distribution's URL or filename.  If no matching distribution is found,
+        ``None`` is returned.
+
+        If the `source` flag is set, only source distributions and source
+        checkout links will be considered.  Unless the `develop_ok` flag is
+        set, development and system eggs (i.e., those using the ``.egg-info``
+        format) will be ignored.
+        """
+
+        # process a Requirement
+        self.info("Searching for %s", requirement)
+        skipped = {}
+
+        def find(req):
+            # Find a matching distribution; may be called more than once
+
+            for dist in self[req.key]:
+
+                if dist.precedence==DEVELOP_DIST and not develop_ok:
+                    if dist not in skipped:
+                        self.warn("Skipping development or system egg: %s",dist)
+                        skipped[dist] = 1
+                    continue
+
+                if dist in req and (dist.precedence<=SOURCE_DIST or not source):
+                    self.info("Best match: %s", dist)
+                    return dist.clone(
+                        location=self.download(dist.location, tmpdir)
+                    )
+
+        if force_scan:
+            self.prescan()
+            self.find_packages(requirement)
+
+        dist = find(requirement)
+        if dist is None and self.to_scan is not None:
+            self.prescan()
+            dist = find(requirement)
+
+        if dist is None and not force_scan:
+            self.find_packages(requirement)
+            dist = find(requirement)
+
+        if dist is None:
+            self.warn(
+                "No local packages or download links found for %s%s",
+                (source and "a source distribution of " or ""),
+                requirement,
+            )
+        return dist
+
+    def fetch(self, requirement, tmpdir, force_scan=False, source=False):
+        """Obtain a file suitable for fulfilling `requirement`
+
+        DEPRECATED; use the ``fetch_distribution()`` method now instead.  For
+        backward compatibility, this routine is identical but returns the
+        ``location`` of the downloaded distribution instead of a distribution
+        object.
+        """
+        dist = self.fetch_distribution(requirement,tmpdir,force_scan,source)
+        if dist is not None:
+            return dist.location
+        return None
+
+
+
+
+
+
+
+
+    def gen_setup(self, filename, fragment, tmpdir):
+        match = EGG_FRAGMENT.match(fragment)
+        dists = match and [d for d in
+            interpret_distro_name(filename, match.group(1), None) if d.version
+        ] or []
+
+        if len(dists)==1:   # unambiguous ``#egg`` fragment
+            basename = os.path.basename(filename)
+
+            # Make sure the file has been downloaded to the temp dir.
+            if os.path.dirname(filename) != tmpdir:
+                dst = os.path.join(tmpdir, basename)
+                from setuptools.command.easy_install import samefile
+                if not samefile(filename, dst):
+                    shutil.copy2(filename, dst)
+                    filename=dst
+
+            file = open(os.path.join(tmpdir, 'setup.py'), 'w')
+            file.write(
+                "from setuptools import setup\n"
+                "setup(name=%r, version=%r, py_modules=[%r])\n"
+                % (
+                    dists[0].project_name, dists[0].version,
+                    os.path.splitext(basename)[0]
+                )
+            )
+            file.close()
+            return filename
+
+        elif match:
+            raise DistutilsError(
+                "Can't unambiguously interpret project/version identifier %r; "
+                "any dashes in the name or version should be escaped using "
+                "underscores. %r" % (fragment,dists)
+            )
+        else:
+            raise DistutilsError(
+                "Can't process plain .py files without an '#egg=name-version'"
+                " suffix to enable automatic setup script generation."
+            )
+
+    dl_blocksize = 8192
+    def _download_to(self, url, filename):
+        self.info("Downloading %s", url)
+        # Download the file
+        fp, tfp, info = None, None, None
+        try:
+            if '#' in url:
+                url, info = url.split('#', 1)
+            fp = self.open_url(url)
+            if isinstance(fp, urllib2.HTTPError):
+                raise DistutilsError(
+                    "Can't download %s: %s %s" % (url, fp.code,fp.msg)
+                )
+            cs = md5()
+            headers = fp.info()
+            blocknum = 0
+            bs = self.dl_blocksize
+            size = -1
+            if "content-length" in headers:
+                size = int(headers["Content-Length"])
+                self.reporthook(url, filename, blocknum, bs, size)
+            tfp = open(filename,'wb')
+            while True:
+                block = fp.read(bs)
+                if block:
+                    cs.update(block)
+                    tfp.write(block)
+                    blocknum += 1
+                    self.reporthook(url, filename, blocknum, bs, size)
+                else:
+                    break
+            if info: self.check_md5(cs, info, filename, tfp)
+            return headers
+        finally:
+            if fp: fp.close()
+            if tfp: tfp.close()
+
+    def reporthook(self, url, filename, blocknum, blksize, size):
+        pass    # no-op
+
+
+    def open_url(self, url, warning=None):
+        if url.startswith('file:'):
+            return local_open(url)
+        try:
+            return open_with_auth(url)
+        except (ValueError, httplib.InvalidURL), v:
+            msg = ' '.join([str(arg) for arg in v.args])
+            if warning:
+                self.warn(warning, msg)
+            else:
+                raise DistutilsError('%s %s' % (url, msg))
+        except urllib2.HTTPError, v:
+            return v
+        except urllib2.URLError, v:
+            if warning:
+                self.warn(warning, v.reason)
+            else:
+                raise DistutilsError("Download error for %s: %s"
+                                     % (url, v.reason))
+        except httplib.BadStatusLine, v:
+            if warning:
+                self.warn(warning, v.line)
+            else:
+                raise DistutilsError('%s returned a bad status line. '
+                                     'The server might be down, %s' % \
+                                             (url, v.line))
+        except httplib.HTTPException, v:
+            if warning:
+                self.warn(warning, v)
+            else:
+                raise DistutilsError("Download error for %s: %s"
+                                     % (url, v))
+
+    def _download_url(self, scheme, url, tmpdir):
+        # Determine download filename
+        #
+        name = filter(None,urlparse.urlparse(url)[2].split('/'))
+        if name:
+            name = name[-1]
+            while '..' in name:
+                name = name.replace('..','.').replace('\\','_')
+        else:
+            name = "__downloaded__"    # default if URL has no path contents
+
+        if name.endswith('.egg.zip'):
+            name = name[:-4]    # strip the extra .zip before download
+
+        filename = os.path.join(tmpdir,name)
+
+        # Download the file
+        #
+        if scheme=='svn' or scheme.startswith('svn+'):
+            return self._download_svn(url, filename)
+        elif scheme=='file':
+            return urllib2.url2pathname(urlparse.urlparse(url)[2])
+        else:
+            self.url_ok(url, True)   # raises error if not allowed
+            return self._attempt_download(url, filename)
+
+
+
+    def scan_url(self, url):
+        self.process_url(url, True)
+
+
+    def _attempt_download(self, url, filename):
+        headers = self._download_to(url, filename)
+        if 'html' in headers.get('content-type','').lower():
+            return self._download_html(url, headers, filename)
+        else:
+            return filename
+
+    def _download_html(self, url, headers, filename):
+        file = open(filename)
+        for line in file:
+            if line.strip():
+                # Check for a subversion index page
+                if re.search(r'<title>([^- ]+ - )?Revision \d+:', line):
+                    # it's a subversion index page:
+                    file.close()
+                    os.unlink(filename)
+                    return self._download_svn(url, filename)
+                break   # not an index page
+        file.close()
+        os.unlink(filename)
+        raise DistutilsError("Unexpected HTML page found at "+url)
+
+    def _download_svn(self, url, filename):
+        url = url.split('#',1)[0]   # remove any fragment for svn's sake
+        self.info("Doing subversion checkout from %s to %s", url, filename)
+        os.system("svn checkout -q %s %s" % (url, filename))
+        return filename
+
+    def debug(self, msg, *args):
+        log.debug(msg, *args)
+
+    def info(self, msg, *args):
+        log.info(msg, *args)
+
+    def warn(self, msg, *args):
+        log.warn(msg, *args)
+
+# This pattern matches a character entity reference (a decimal numeric
+# references, a hexadecimal numeric reference, or a named reference).
+entity_sub = re.compile(r'&(#(\d+|x[\da-fA-F]+)|[\w.:-]+);?').sub
+
+def uchr(c):
+    if not isinstance(c, int):
+        return c
+    if c>255: return unichr(c)
+    return chr(c)
+
+def decode_entity(match):
+    what = match.group(1)
+    if what.startswith('#x'):
+        what = int(what[2:], 16)
+    elif what.startswith('#'):
+        what = int(what[1:])
+    else:
+        from htmlentitydefs import name2codepoint
+        what = name2codepoint.get(what, match.group(0))
+    return uchr(what)
+
+def htmldecode(text):
+    """Decode HTML entities in the given text."""
+    return entity_sub(decode_entity, text)
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+def socket_timeout(timeout=15):
+    def _socket_timeout(func):
+        def _socket_timeout(*args, **kwargs):
+            old_timeout = socket.getdefaulttimeout()
+            socket.setdefaulttimeout(timeout)
+            try:
+                return func(*args, **kwargs)
+            finally:
+                socket.setdefaulttimeout(old_timeout)
+        return _socket_timeout
+    return _socket_timeout
+
+
+def open_with_auth(url):
+    """Open a urllib2 request, handling HTTP authentication"""
+
+    scheme, netloc, path, params, query, frag = urlparse.urlparse(url)
+
+    if scheme in ('http', 'https'):
+        auth, host = urllib2.splituser(netloc)
+    else:
+        auth = None
+
+    if auth:
+        auth = "Basic " + urllib2.unquote(auth).encode('base64').strip()
+        new_url = urlparse.urlunparse((scheme,host,path,params,query,frag))
+        request = urllib2.Request(new_url)
+        request.add_header("Authorization", auth)
+    else:
+        request = urllib2.Request(url)
+
+    request.add_header('User-Agent', user_agent)
+    fp = urllib2.urlopen(request)
+
+    if auth:
+        # Put authentication info back into request URL if same host,
+        # so that links found on the page will work
+        s2, h2, path2, param2, query2, frag2 = urlparse.urlparse(fp.url)
+        if s2==scheme and h2==host:
+            fp.url = urlparse.urlunparse((s2,netloc,path2,param2,query2,frag2))
+
+    return fp
+
+# adding a timeout to avoid freezing package_index
+open_with_auth = socket_timeout(_SOCKET_TIMEOUT)(open_with_auth)
+
+
+
+
+
+
+
+
+
+
+
+def fix_sf_url(url):
+    return url      # backward compatibility
+
+def local_open(url):
+    """Read a local path, with special support for directories"""
+    scheme, server, path, param, query, frag = urlparse.urlparse(url)
+    filename = urllib2.url2pathname(path)
+    if os.path.isfile(filename):
+        return urllib2.urlopen(url)
+    elif path.endswith('/') and os.path.isdir(filename):
+        files = []
+        for f in os.listdir(filename):
+            if f=='index.html':
+                fp = open(os.path.join(filename,f),'rb')
+                body = fp.read()
+                fp.close()
+                break
+            elif os.path.isdir(os.path.join(filename,f)):
+                f+='/'
+            files.append("<a href=%r>%s</a>" % (f,f))
+        else:
+            body = ("<html><head><title>%s</title>" % url) + \
+                "</head><body>%s</body></html>" % '\n'.join(files)
+        status, message = 200, "OK"
+    else:
+        status, message, body = 404, "Path not found", "Not found"
+
+    return urllib2.HTTPError(url, status, message,
+            {'content-type':'text/html'}, cStringIO.StringIO(body))
+
+
+
+
+
+
+
+
+
+
+
+
+
+# this line is a kludge to keep the trailing blank lines for pje's editor

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: merge konryd works about pypi test
	server
Message-ID: <E1OVLoV-0000ho-8W@dinsdale.python.org>

tarek.ziade pushed c217e36b984a to distutils2:

http://hg.python.org/distutils2/rev/c217e36b984a
changeset:   284:c217e36b984a
parent:      283:d6099bdc57cd
parent:      275:944a00b90bde
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Fri May 28 11:24:53 2010 +0200
summary:     merge konryd works about pypi test server
files:       

diff --git a/docs/design/pep-0376.txt b/docs/design/pep-0376.txt
--- a/docs/design/pep-0376.txt
+++ b/docs/design/pep-0376.txt
@@ -633,7 +633,7 @@
 
 Distributions installed using existing, pre-standardization formats do not have
 the necessary metadata available for the new API, and thus will be
-ignored. Third-party tools may of course to continue to support previous
+ignored. Third-party tools may of course continue to support previous
 formats in addition to the new format, in order to ease the transition.
 
 
diff --git a/docs/design/wiki.rst b/docs/design/wiki.rst
--- a/docs/design/wiki.rst
+++ b/docs/design/wiki.rst
@@ -282,7 +282,7 @@
   mailman/etc/*               = {config}                # 8
   mailman/foo/**/bar/*.cfg    = {config}/baz            # 9
   mailman/foo/**/*.cfg        = {config}/hmm            # 9, 10
-  some-new-semantic.txt       = {funky-crazy-category}  # 11
+  some-new-semantic.sns       = {funky-crazy-category}  # 11
 
 The glob definitions are relative paths that match files from the top
 of the source tree (the location of ``setup.cfg``). Forward slashes
diff --git a/docs/source/index.rst b/docs/source/index.rst
--- a/docs/source/index.rst
+++ b/docs/source/index.rst
@@ -12,6 +12,8 @@
    :maxdepth: 2
 
    metadata
+   new_commands
+   test_framework
 
 Indices and tables
 ==================
diff --git a/docs/source/new_commands.rst b/docs/source/new_commands.rst
new file mode 100644
--- /dev/null
+++ b/docs/source/new_commands.rst
@@ -0,0 +1,65 @@
+========
+Commands
+========
+
+Distutils2 provides a set of commands that are not present in distutils itself.
+You might recognize some of them from other projects, like Distribute or
+Setuptools.
+
+``upload_docs`` - Upload package documentation to PyPI
+======================================================
+
+PyPI now supports uploading project documentation to the dedicated URL
+http://packages.python.org/<project>/.
+
+The ``upload_docs`` command will create the necessary zip file out of a
+documentation directory and will post to the repository.
+
+Note that to upload the documentation of a project, the corresponding version
+must already be registered with PyPI, using the distutils ``register``
+command -- just like the ``upload`` command.
+
+Assuming there is an ``Example`` project with documentation in the
+subdirectory ``docs``, e.g.::
+
+  Example/
+  |-- example.py
+  |-- setup.cfg
+  |-- setup.py
+  |-- docs
+  |   |-- build
+  |   |   `-- html
+  |   |   |   |-- index.html
+  |   |   |   `-- tips_tricks.html
+  |   |-- conf.py
+  |   |-- index.txt
+  |   `-- tips_tricks.txt
+
+You can simply pass the documentation directory path to the ``upload_docs``
+command::
+
+    python setup.py upload_docs --upload-dir=docs/build/html
+
+As with any other ``setuptools`` based command, you can define useful
+defaults in the ``setup.cfg`` of your Python project, e.g.:
+
+.. code-block:: ini
+
+    [upload_docs]
+    upload-dir = docs/build/html
+
+The ``upload_docs`` command has the following options:
+
+``--upload-dir``
+    The directory to be uploaded to the repository. The default value is
+    ``docs`` in project root.
+
+``--show-response``
+    Display the full response text from server; this is useful for debugging
+    PyPI problems.
+
+``--repository=URL, -r URL``
+    The URL of the repository to upload to.  Defaults to
+    http://pypi.python.org/pypi (i.e., the main PyPI installation).
+
+
diff --git a/docs/source/test_framework.rst b/docs/source/test_framework.rst
new file mode 100644
--- /dev/null
+++ b/docs/source/test_framework.rst
@@ -0,0 +1,34 @@
+==============
+Test Framework
+==============
+
+When you are testing code that works with distutils, you might find these tools
+useful.
+
+
+``PyPIServer``
+==============
+
+PyPIServer is a class that implements an HTTP server running in a separate
+thread. All it does is record the requests for further inspection. The recorded
+data is available under ``requests`` attribute. The default
+HTTP response can be overriden with the ``default_response_status``,
+``default_response_headers`` and ``default_response_data`` attributes.
+
+
+``PyPIServerTestCase``
+======================
+
+``PyPIServerTestCase`` is a test case class with setUp and tearDown methods that
+take care of a single PyPIServer instance attached as a ``pypi`` attribute on
+the test class. Use it as one of the base classes in you test case::
+
+  class UploadTestCase(PyPIServerTestCase):
+      def test_something(self):
+          cmd = self.prepare_command()
+          cmd.ensure_finalized()
+          cmd.repository = self.pypi.full_address
+          cmd.run()
+
+          environ, request_data = self.pypi.requests[-1]
+          self.assertEqual(request_data, EXPECTED_REQUEST_DATA)
diff --git a/src/distutils2/command/upload_docs.py b/src/distutils2/command/upload_docs.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/command/upload_docs.py
@@ -0,0 +1,128 @@
+import base64, httplib, os.path, socket, tempfile, urlparse, zipfile
+from cStringIO import StringIO
+from distutils2 import log
+from distutils2.core import PyPIRCCommand
+from distutils2.errors import DistutilsFileError
+
+def zip_dir(directory):
+    """Compresses recursively contents of directory into a StringIO object"""
+    destination = StringIO()
+    zip_file = zipfile.ZipFile(destination, "w")
+    for root, dirs, files in os.walk(directory):
+        for name in files:
+            full = os.path.join(root, name)
+            relative = root[len(directory):].lstrip(os.path.sep)
+            dest = os.path.join(relative, name)
+            zip_file.write(full, dest)
+    zip_file.close()
+    return destination
+
+# grabbed from
+#    http://code.activestate.com/recipes/146306-http-client-to-post-using-multipartform-data/
+def encode_multipart(fields, files, boundary=None):
+    """
+    fields is a sequence of (name, value) elements for regular form fields.
+    files is a sequence of (name, filename, value) elements for data to be uploaded as files
+    Return (content_type, body) ready for httplib.HTTP instance
+    """
+    if boundary is None:
+        boundary = '--------------GHSKFJDLGDS7543FJKLFHRE75642756743254'
+    CRLF = '\r\n'
+    l = []
+    for (key, value) in fields:
+        l.extend([
+            '--' + boundary,
+            'Content-Disposition: form-data; name="%s"' % key,
+            '',
+            value])
+    for (key, filename, value) in files:
+        l.extend([
+            '--' + boundary,
+            'Content-Disposition: form-data; name="%s"; filename="%s"' % (key, filename),
+            '',
+            value])
+    l.append('--' + boundary + '--')
+    l.append('')
+    body = CRLF.join(l)
+    content_type = 'multipart/form-data; boundary=%s' % boundary
+    return content_type, body
+
+class upload_docs(PyPIRCCommand):
+
+    user_options = [
+        ('upload-dir=', None, 'directory to upload'),
+        ]
+
+    def initialize_options(self):
+        self.upload_dir = None
+
+    def finalize_options(self):
+        PyPIRCCommand.finalize_options(self)
+        if self.upload_dir == None:
+            build = self.get_finalized_command('build')
+            self.upload_dir = os.path.join(build.build_base, "docs")
+        config = self._read_pypirc()
+        if config != {}:
+            self.username = config['username']
+            self.password = config['password']
+            self.repository = config['repository']
+            self.realm = config['realm']
+
+    def verify_upload_dir(self, upload_dir):
+        index_location = os.path.join(upload_dir, "index.html")
+        if not os.path.exists(index_location):
+            mesg = "No 'index.html found in docs directory (%s)"
+            raise DistutilsFileError(mesg % upload_dir)
+
+
+    def run(self):
+        tmp_dir = tempfile.mkdtemp()
+        name = self.distribution.metadata['Name']
+        self.verify_upload_dir(self.upload_dir)
+        zip_file = zip_dir(self.upload_dir)
+
+        fields = {':action': 'doc_upload', 'name': name}.items()
+        files = [('content', name, zip_file.getvalue())]
+        content_type, body = encode_multipart(fields, files)
+
+        credentials = self.username + ':' + self.password
+        auth = "Basic " + base64.encodestring(credentials).strip()
+
+        self.announce("Submitting documentation to %s" % (self.repository),
+                      log.INFO)
+
+        schema, netloc, url, params, query, fragments = \
+            urlparse.urlparse(self.repository)
+        if schema == "http":
+            conn = httplib.HTTPConnection(netloc)
+        elif schema == "https":
+            conn = httplib.HTTPSConnection(netloc)
+        else:
+            raise AssertionError("unsupported schema "+schema)
+
+        try:
+            conn.connect()
+            conn.putrequest("POST", url)
+            conn.putheader('Content-type', content_type)
+            conn.putheader('Content-length', str(len(body)))
+            conn.putheader('Authorization', auth)
+            conn.endheaders()
+            conn.send(body)
+        except socket.error, e:
+            self.announce(str(e), log.ERROR)
+            return
+
+        r = conn.getresponse()
+
+        if r.status == 200:
+            self.announce('Server response (%s): %s' % (r.status, r.reason),
+                          log.INFO)
+        elif r.status == 301:
+            location = r.getheader('Location')
+            if location is None:
+                location = 'http://packages.python.org/%s/' % meta.get_name()
+            self.announce('Upload successful. Visit %s' % location,
+                          log.INFO)
+        else:
+            self.announce('Upload failed (%s): %s' % (r.status, r.reason),
+                          log.ERROR)
diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypi_server.py
@@ -0,0 +1,63 @@
+import Queue, threading, time, unittest2
+from wsgiref.simple_server import make_server
+
+class PyPIServerTestCase(unittest2.TestCase):
+
+    def setUp(self):
+        super(PyPIServerTestCase, self).setUp()
+        self.pypi = PyPIServer()
+        self.pypi.start()
+
+    def tearDown(self):
+        super(PyPIServerTestCase, self).tearDown()
+        self.pypi.stop()
+
+class PyPIServer(threading.Thread):
+    """Thread that wraps a wsgi app"""
+    def __init__(self):
+        threading.Thread.__init__(self)
+        self.httpd = make_server('', 0, self.pypi_app)
+        self.httpd.RequestHandlerClass.log_request = lambda *_: None
+        self.address = self.httpd.server_address
+        self.request_queue = Queue.Queue()
+        self._requests = []
+        self.default_response_status = "200 OK"
+        self.default_response_headers = [('Content-type', 'text/plain')]
+        self.default_response_data = ["hello"]
+
+    def run(self):
+        self.httpd.serve_forever()
+
+    def stop(self):
+        self.httpd.shutdown()
+        self.join()
+
+    def pypi_app(self, environ, start_response):
+        # record the request
+        if environ.get("CONTENT_LENGTH"):
+            request_data = environ.pop('wsgi.input').read(int(environ['CONTENT_LENGTH']))
+        else:
+            request_data = environ.pop('wsgi.input').read()
+        self.request_queue.put((environ, request_data))
+        # send back a response
+        status, headers, data = self.get_next_response()
+        start_response(status, headers)
+        return data
+
+    def get_next_response(self):
+        return (self.default_response_status,
+                self.default_response_headers,
+                self.default_response_data)
+
+    @property
+    def requests(self):
+        while True:
+            try:
+                self._requests.append(self.request_queue.get_nowait())
+            except Queue.Empty:
+                break
+        return self._requests
+
+    @property
+    def full_address(self):
+        return "http://%s:%s" % self.address
diff --git a/src/distutils2/tests/test_pypi_server.py b/src/distutils2/tests/test_pypi_server.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/test_pypi_server.py
@@ -0,0 +1,27 @@
+"""Tests for distutils.command.bdist."""
+import unittest2, urllib, urllib2
+from distutils2.tests.pypi_server import PyPIServer
+
+class PyPIServerTest(unittest2.TestCase):
+
+    def test_records_requests(self):
+        server = PyPIServer()
+        server.start()
+        self.assertEqual(len(server.requests), 0)
+
+        data = "Rock Around The Bunker"
+        headers = {"X-test-header": "Mister Iceberg"}
+
+        request = urllib2.Request(server.full_address, data, headers)
+        urllib2.urlopen(request)
+        self.assertEqual(len(server.requests), 1)
+        environ, request_data = server.requests[-1]
+        self.assertIn("Rock Around The Bunker", request_data)
+        self.assertEqual(environ["HTTP_X_TEST_HEADER"], "Mister Iceberg")
+        server.stop()
+
+def test_suite():
+    return unittest2.makeSuite(PyPIServerTest)
+
+if __name__ == '__main__':
+    unittest2.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_upload.py b/src/distutils2/tests/test_upload.py
--- a/src/distutils2/tests/test_upload.py
+++ b/src/distutils2/tests/test_upload.py
@@ -1,34 +1,13 @@
 """Tests for distutils.command.upload."""
 # -*- encoding: utf8 -*-
-import sys
-import os
-import unittest2
+import os, unittest2
 
-from distutils2.command import upload as upload_mod
 from distutils2.command.upload import upload
 from distutils2.core import Distribution
 
-from distutils2.tests import support
+from distutils2.tests.pypi_server import PyPIServer, PyPIServerTestCase
 from distutils2.tests.test_config import PYPIRC, PyPIRCCommandTestCase
 
-PYPIRC_LONG_PASSWORD = """\
-[distutils]
-
-index-servers =
-    server1
-    server2
-
-[server1]
-username:me
-password:aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
-
-[server2]
-username:meagain
-password: secret
-realm:acme
-repository:http://another.pypi/
-"""
-
 
 PYPIRC_NOPASSWORD = """\
 [distutils]
@@ -40,38 +19,9 @@
 username:me
 """
 
-class FakeOpen(object):
-
-    def __init__(self, url):
-        self.url = url
-        if not isinstance(url, str):
-            self.req = url
-        else:
-            self.req = None
-        self.msg = 'OK'
-
-    def getcode(self):
-        return 200
-
-
-class uploadTestCase(PyPIRCCommandTestCase):
-
-    def setUp(self):
-        super(uploadTestCase, self).setUp()
-        self.old_open = upload_mod.urlopen
-        upload_mod.urlopen = self._urlopen
-        self.last_open = None
-
-    def tearDown(self):
-        upload_mod.urlopen = self.old_open
-        super(uploadTestCase, self).tearDown()
-
-    def _urlopen(self, url):
-        self.last_open = FakeOpen(url)
-        return self.last_open
+class UploadTestCase(PyPIServerTestCase, PyPIRCCommandTestCase):
 
     def test_finalize_options(self):
-
         # new format
         self.write_file(self.rc, PYPIRC)
         dist = Distribution()
@@ -80,7 +30,7 @@
         for attr, waited in (('username', 'me'), ('password', 'secret'),
                              ('realm', 'pypi'),
                              ('repository', 'http://pypi.python.org/pypi')):
-            self.assertEquals(getattr(cmd, attr), waited)
+            self.assertEqual(getattr(cmd, attr), waited)
 
     def test_saved_password(self):
         # file with no password
@@ -89,44 +39,40 @@
         # make sure it passes
         dist = Distribution()
         cmd = upload(dist)
-        cmd.finalize_options()
-        self.assertEquals(cmd.password, None)
+        cmd.ensure_finalized()
+        self.assertEqual(cmd.password, None)
 
         # make sure we get it as well, if another command
         # initialized it at the dist level
         dist.password = 'xxx'
         cmd = upload(dist)
         cmd.finalize_options()
-        self.assertEquals(cmd.password, 'xxx')
+        self.assertEqual(cmd.password, 'xxx')
 
     def test_upload(self):
-        tmp = self.mkdtemp()
-        path = os.path.join(tmp, 'xxx')
+        path = os.path.join(self.tmp_dir, 'xxx')
         self.write_file(path)
         command, pyversion, filename = 'xxx', '2.6', path
         dist_files = [(command, pyversion, filename)]
-        self.write_file(self.rc, PYPIRC_LONG_PASSWORD)
 
         # lets run it
         pkg_dir, dist = self.create_dist(dist_files=dist_files, author=u'd??d??')
         cmd = upload(dist)
         cmd.ensure_finalized()
+        cmd.repository = self.pypi.full_address
         cmd.run()
 
         # what did we send ?
-        self.assertIn('d??d??', self.last_open.req.data)
-        headers = dict(self.last_open.req.headers)
-        self.assert_(headers['Content-length'] > 2000)
-        self.assertTrue(headers['Content-type'].startswith('multipart/form-data'))
-        self.assertEquals(self.last_open.req.get_method(), 'POST')
-        self.assertEquals(self.last_open.req.get_full_url(),
-                          'http://pypi.python.org/pypi')
-        self.assertTrue('xxx' in self.last_open.req.data)
-        auth = self.last_open.req.headers['Authorization']
-        self.assertFalse('\n' in auth)
+        environ, request_data = self.pypi.requests[-1]
+        self.assertIn('d??d??', request_data)
+        self.assertIn('xxx', request_data)
+        self.assert_(environ['CONTENT_LENGTH'] > 2000)
+        self.assertTrue(environ['CONTENT_TYPE'].startswith('multipart/form-data'))
+        self.assertEqual(environ['REQUEST_METHOD'], 'POST')
+        self.assertNotIn('\n', environ['HTTP_AUTHORIZATION'])
 
 def test_suite():
-    return unittest2.makeSuite(uploadTestCase)
+    return unittest2.makeSuite(UploadTestCase)
 
 if __name__ == "__main__":
     unittest2.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_upload_docs.py b/src/distutils2/tests/test_upload_docs.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/test_upload_docs.py
@@ -0,0 +1,167 @@
+"""Tests for distutils.command.upload_docs."""
+# -*- encoding: utf8 -*-
+import httplib, os, os.path, tempfile, unittest2, zipfile
+from cStringIO import StringIO
+
+from distutils2.command import upload_docs as upload_docs_mod
+from distutils2.command.upload_docs import (upload_docs, zip_dir,
+                                    encode_multipart)
+from distutils2.core import Distribution
+
+from distutils2.errors import DistutilsFileError
+
+from distutils2.tests import support
+from distutils2.tests.pypi_server import PyPIServer, PyPIServerTestCase
+from distutils2.tests.test_config import PyPIRCCommandTestCase
+
+
+EXPECTED_MULTIPART_OUTPUT = "\r\n".join([
+'---x',
+'Content-Disposition: form-data; name="a"',
+'',
+'b',
+'---x',
+'Content-Disposition: form-data; name="c"',
+'',
+'d',
+'---x',
+'Content-Disposition: form-data; name="e"; filename="f"',
+'',
+'g',
+'---x',
+'Content-Disposition: form-data; name="h"; filename="i"',
+'',
+'j',
+'---x--',
+'',
+])
+
+PYPIRC = """\
+[distutils]
+index-servers = server1
+
+[server1]
+repository = %s
+username = real_slim_shady
+password = long_island
+"""
+
+class UploadDocsTestCase(PyPIServerTestCase, PyPIRCCommandTestCase):
+
+    def setUp(self):
+        super(UploadDocsTestCase, self).setUp()
+        self.dist = Distribution()
+        self.dist.metadata['Name'] = "distr-name"
+        self.cmd = upload_docs(self.dist)
+
+    def test_generates_uploaddir_if_not_given(self):
+        self.assertEqual(self.cmd.upload_dir, None)
+        self.cmd.ensure_finalized()
+        self.assertEqual(self.cmd.upload_dir, os.path.join("build", "docs"))
+
+    def prepare_sample_dir(self):
+        sample_dir = tempfile.mkdtemp()
+        os.mkdir(os.path.join(sample_dir, "some_dir"))
+        self.write_file(os.path.join(sample_dir, "some_dir", "index.html"), "Ce mortel ennui")
+        self.write_file(os.path.join(sample_dir, "index.html"), "Oh la la")
+        return sample_dir
+
+    def test_zip_dir(self):
+        source_dir = self.prepare_sample_dir()
+        compressed = zip_dir(source_dir)
+
+        zip_f = zipfile.ZipFile(compressed)
+        self.assertEqual(zip_f.namelist(), ['index.html', 'some_dir/index.html'])
+
+    def test_encode_multipart(self):
+        fields = [("a", "b"), ("c", "d")]
+        files = [("e", "f", "g"), ("h", "i", "j")]
+        content_type, body = encode_multipart(fields, files, "-x")
+        self.assertEqual(content_type, "multipart/form-data; boundary=-x")
+        self.assertEqual(body, EXPECTED_MULTIPART_OUTPUT)
+
+    def prepare_command(self):
+        self.cmd.ensure_finalized()
+        self.cmd.repository = self.pypi.full_address
+        self.cmd.upload_dir = self.prepare_sample_dir()
+        self.cmd.username = "username"
+        self.cmd.password = "password"
+
+    def test_upload(self):
+        self.prepare_command()
+        self.cmd.run()
+
+        self.assertEqual(len(self.pypi.requests), 1)
+        environ, request_data = self.pypi.requests[-1]
+
+        self.assertIn("content", request_data)
+        self.assertIn("Basic", environ['HTTP_AUTHORIZATION'])
+        self.assertTrue(environ['CONTENT_TYPE'].startswith('multipart/form-data;'))
+
+        action, name, content =\
+            request_data.split("----------------GHSKFJDLGDS7543FJKLFHRE75642756743254")[1:4]
+
+        # check that we picked the right chunks
+        self.assertIn('name=":action"', action)
+        self.assertIn('name="name"', name)
+        self.assertIn('name="content"', content)
+
+        # check their contents
+        self.assertIn("doc_upload", action)
+        self.assertIn("distr-name", name)
+        self.assertIn("some_dir/index.html", content)
+        self.assertIn("Ce mortel ennui", content)
+
+    def test_https_connection(self):
+        https_called = False
+        orig_https = upload_docs_mod.httplib.HTTPSConnection
+        def https_conn_wrapper(*args):
+            https_called = True
+            return upload_docs_mod.httplib.HTTPConnection(*args) # the testing server is http
+        upload_docs_mod.httplib.HTTPSConnection = https_conn_wrapper
+        try:
+            self.prepare_command()
+            self.cmd.run()
+            self.assertFalse(https_called)
+
+            self.cmd.repository = self.cmd.repository.replace("http", "https")
+            self.cmd.run()
+            self.assertFalse(https_called)
+        finally:
+            upload_docs_mod.httplib.HTTPSConnection = orig_https
+
+    def test_handling_response(self):
+        calls = []
+        def aggr(*args):
+            calls.append(args)
+        self.pypi.default_response_status = '403 Forbidden'
+        self.prepare_command()
+        self.cmd.announce = aggr
+        self.cmd.run()
+        message, _ = calls[-1]
+        self.assertIn('Upload failed (403): Forbidden', message)
+
+        calls = []
+        self.pypi.default_response_status = '301 Moved Permanently'
+        self.pypi.default_response_headers.append(("Location", "brand_new_location"))
+        self.cmd.run()
+        message, _ = calls[-1]
+        self.assertIn('brand_new_location', message)
+
+    def test_reads_pypirc_data(self):
+        self.write_file(self.rc, PYPIRC % self.pypi.full_address)
+        self.cmd.repository = self.pypi.full_address
+        self.cmd.ensure_finalized()
+        self.assertEqual(self.cmd.username, "real_slim_shady")
+        self.assertEqual(self.cmd.password, "long_island")
+
+    def test_checks_index_html_presence(self):
+        self.prepare_command()
+        os.remove(os.path.join(self.cmd.upload_dir, "index.html"))
+        self.assertRaises(DistutilsFileError, self.cmd.run)
+
+def test_suite():
+    return unittest2.makeSuite(UploadDocsTestCase)
+
+if __name__ == "__main__":
+    unittest2.main(defaultTest="test_suite")

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: make mocked PyPIServer returning
	static files when	accessing the server with
Message-ID: <E1OVLoV-0000hs-93@dinsdale.python.org>

tarek.ziade pushed f90efe197ac3 to distutils2:

http://hg.python.org/distutils2/rev/f90efe197ac3
changeset:   285:f90efe197ac3
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Mon May 31 22:43:03 2010 +0200
summary:     make mocked PyPIServer returning static files when accessing the server with the /simple/ prefix
files:       docs/source/test_framework.rst, src/distutils2/tests/pypi_server.py, src/distutils2/tests/test_pypi_server.py

diff --git a/docs/source/test_framework.rst b/docs/source/test_framework.rst
--- a/docs/source/test_framework.rst
+++ b/docs/source/test_framework.rst
@@ -5,7 +5,6 @@
 When you are testing code that works with distutils, you might find these tools
 useful.
 
-
 ``PyPIServer``
 ==============
 
@@ -15,13 +14,16 @@
 HTTP response can be overriden with the ``default_response_status``,
 ``default_response_headers`` and ``default_response_data`` attributes.
 
+When accessing the server with urls beginning with `/simple/`, the server also
+record your requests, but will look for files under the /tests/pypi_server/
+path.
 
 ``PyPIServerTestCase``
 ======================
 
 ``PyPIServerTestCase`` is a test case class with setUp and tearDown methods that
 take care of a single PyPIServer instance attached as a ``pypi`` attribute on
-the test class. Use it as one of the base classes in you test case::
+the test class. Use it as one of the base classes in your test case::
 
   class UploadTestCase(PyPIServerTestCase):
       def test_something(self):
diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
--- a/src/distutils2/tests/pypi_server.py
+++ b/src/distutils2/tests/pypi_server.py
@@ -1,5 +1,8 @@
 import Queue, threading, time, unittest2
 from wsgiref.simple_server import make_server
+import os.path
+
+PYPI_HTML_BASE_PATH = os.path.dirname(os.path.abspath(__file__)) + "/pypiserver"
 
 class PyPIServerTestCase(unittest2.TestCase):
 
@@ -33,16 +36,37 @@
         self.join()
 
     def pypi_app(self, environ, start_response):
-        # record the request
-        if environ.get("CONTENT_LENGTH"):
-            request_data = environ.pop('wsgi.input').read(int(environ['CONTENT_LENGTH']))
+        # record the request. Read the input only on PUT or POST requests
+        if environ["REQUEST_METHOD"] in ("PUT", "POST"):
+            if environ.get("CONTENT_LENGTH"):
+                request_data = environ.pop('wsgi.input').read(int(environ['CONTENT_LENGTH']))
+            else:
+                request_data = environ.pop('wsgi.input').read()
+        elif environ["REQUEST_METHOD"] in ("GET", "DELETE"):
+            request_data = ''
+            
+        self.request_queue.put((environ, request_data))
+        
+        relative_path = environ["PATH_INFO"].replace(self.full_address, '')
+        
+        # serve the content from local disc if we request /simple
+        if relative_path.startswith("/simple/"):
+            file_to_serve = relative_path.replace("/simple", PYPI_HTML_BASE_PATH)
+            try:
+                file = open(file_to_serve)
+                data = file.read()
+                start_response("200 OK", [('Content-type', 'text/plain')])
+            except IOError:
+                start_response("404 NOT FOUND", [('Content-type', 'text/plain')])
+                data = "Not Found"
+            return data
+
+        # otherwise serve the content from default_response* parameters
         else:
-            request_data = environ.pop('wsgi.input').read()
-        self.request_queue.put((environ, request_data))
-        # send back a response
-        status, headers, data = self.get_next_response()
-        start_response(status, headers)
-        return data
+            # send back a response
+            status, headers, data = self.get_next_response()
+            start_response(status, headers)
+            return data
 
     def get_next_response(self):
         return (self.default_response_status,
diff --git a/src/distutils2/tests/test_pypi_server.py b/src/distutils2/tests/test_pypi_server.py
--- a/src/distutils2/tests/test_pypi_server.py
+++ b/src/distutils2/tests/test_pypi_server.py
@@ -5,6 +5,7 @@
 class PyPIServerTest(unittest2.TestCase):
 
     def test_records_requests(self):
+        """We expect that PyPIServer can log our requests"""
         server = PyPIServer()
         server.start()
         self.assertEqual(len(server.requests), 0)
@@ -20,6 +21,26 @@
         self.assertEqual(environ["HTTP_X_TEST_HEADER"], "Mister Iceberg")
         server.stop()
 
+    def test_serve_static_content(self):
+        """We expect that when accessing the test PyPI server, files can 
+        be served statically."""
+        server = PyPIServer()
+        server.start()
+        
+        # the file does not exists on the disc, so it might not be served
+        url = server.full_address + "/simple/unexisting_page"
+        request = urllib2.Request(url)
+        try:
+            urllib2.urlopen(request)
+        except urllib2.HTTPError as e:
+            self.assertEqual(e.getcode(), 404)
+
+        # now try serving a content that do exists
+        url = server.full_address + "/simple/index.html"
+        request = urllib2.Request(url)
+        f = urllib2.urlopen(request)
+        self.assertEqual(f.read(), "Yeah\n")
+
 def test_suite():
     return unittest2.makeSuite(PyPIServerTest)
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: Rename package_index module from
	Distribute to	pypi.simple in Distutils2
Message-ID: <E1OVLoV-0000i4-BE@dinsdale.python.org>

tarek.ziade pushed a8087e4ed26c to distutils2:

http://hg.python.org/distutils2/rev/a8087e4ed26c
changeset:   288:a8087e4ed26c
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Tue Jun 01 09:27:47 2010 +0200
summary:     Rename package_index module from Distribute to pypi.simple in Distutils2
files:       src/distutils2/pypi/__init__.py, src/distutils2/pypi/package_index.py, src/distutils2/pypi/pkg_resources.py, src/distutils2/pypi/simple.py, src/distutils2/tests/test_pypi_simple.py

diff --git a/src/distutils2/pypi/__init__.py b/src/distutils2/pypi/__init__.py
--- a/src/distutils2/pypi/__init__.py
+++ b/src/distutils2/pypi/__init__.py
@@ -4,4 +4,4 @@
 """ 
 
 __all__ = ['package_index',
-}
+]
diff --git a/src/distutils2/pypi/pkg_resources.py b/src/distutils2/pypi/pkg_resources.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/pypi/pkg_resources.py
@@ -0,0 +1,2693 @@
+"""Package resource API
+--------------------
+
+A resource is a logical file contained within a package, or a logical
+subdirectory thereof.  The package resource API expects resource names
+to have their path parts separated with ``/``, *not* whatever the local
+path separator is.  Do not use os.path operations to manipulate resource
+names being passed into the API.
+
+The package resource API is designed to work with normal filesystem packages,
+.egg files, and unpacked .egg files.  It can also work in a limited way with
+.zip files and with custom PEP 302 loaders that support the ``get_data()``
+method.
+"""
+
+import sys, os, zipimport, time, re, imp, types
+from urlparse import urlparse, urlunparse
+
+try:
+    frozenset
+except NameError:
+    from sets import ImmutableSet as frozenset
+
+# capture these to bypass sandboxing
+from os import utime
+try:
+    from os import mkdir, rename, unlink
+    WRITE_SUPPORT = True
+except ImportError:
+    # no write support, probably under GAE
+    WRITE_SUPPORT = False
+
+from os import open as os_open
+from os.path import isdir, split
+
+# This marker is used to simplify the process that checks is the
+# setuptools package was installed by the Setuptools project
+# or by the Distribute project, in case Setuptools creates
+# a distribution with the same version.
+#
+# The bootstrapping script for instance, will check if this
+# attribute is present to decide wether to reinstall the package
+_distribute = True
+
+def _bypass_ensure_directory(name, mode=0777):
+    # Sandbox-bypassing version of ensure_directory()
+    if not WRITE_SUPPORT:
+        raise IOError('"os.mkdir" not supported on this platform.')
+    dirname, filename = split(name)
+    if dirname and filename and not isdir(dirname):
+        _bypass_ensure_directory(dirname)
+        mkdir(dirname, mode)
+
+
+
+
+
+
+
+
+def get_supported_platform():
+    """Return this platform's maximum compatible version.
+
+    distutils.util.get_platform() normally reports the minimum version
+    of Mac OS X that would be required to *use* extensions produced by
+    distutils.  But what we want when checking compatibility is to know the
+    version of Mac OS X that we are *running*.  To allow usage of packages that
+    explicitly require a newer version of Mac OS X, we must also know the
+    current version of the OS.
+
+    If this condition occurs for any other platform with a version in its
+    platform strings, this function should be extended accordingly.
+    """
+    plat = get_build_platform(); m = macosVersionString.match(plat)
+    if m is not None and sys.platform == "darwin":
+        try:
+            plat = 'macosx-%s-%s' % ('.'.join(_macosx_vers()[:2]), m.group(3))
+        except ValueError:
+            pass    # not Mac OS X
+    return plat
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+__all__ = [
+    # Basic resource access and distribution/entry point discovery
+    'require', 'run_script', 'get_provider',  'get_distribution',
+    'load_entry_point', 'get_entry_map', 'get_entry_info', 'iter_entry_points',
+    'resource_string', 'resource_stream', 'resource_filename',
+    'resource_listdir', 'resource_exists', 'resource_isdir',
+
+    # Environmental control
+    'declare_namespace', 'working_set', 'add_activation_listener',
+    'find_distributions', 'set_extraction_path', 'cleanup_resources',
+    'get_default_cache',
+
+    # Primary implementation classes
+    'Environment', 'WorkingSet', 'ResourceManager',
+    'Distribution', 'Requirement', 'EntryPoint',
+
+    # Exceptions
+    'ResolutionError','VersionConflict','DistributionNotFound','UnknownExtra',
+    'ExtractionError',
+
+    # Parsing functions and string utilities
+    'parse_requirements', 'parse_version', 'safe_name', 'safe_version',
+    'get_platform', 'compatible_platforms', 'yield_lines', 'split_sections',
+    'safe_extra', 'to_filename',
+
+    # filesystem utilities
+    'ensure_directory', 'normalize_path',
+
+    # Distribution "precedence" constants
+    'EGG_DIST', 'BINARY_DIST', 'SOURCE_DIST', 'CHECKOUT_DIST', 'DEVELOP_DIST',
+
+    # "Provider" interfaces, implementations, and registration/lookup APIs
+    'IMetadataProvider', 'IResourceProvider', 'FileMetadata',
+    'PathMetadata', 'EggMetadata', 'EmptyProvider', 'empty_provider',
+    'NullProvider', 'EggProvider', 'DefaultProvider', 'ZipProvider',
+    'register_finder', 'register_namespace_handler', 'register_loader_type',
+    'fixup_namespace_packages', 'get_importer',
+
+    # Deprecated/backward compatibility only
+    'run_main', 'AvailableDistributions',
+]
+class ResolutionError(Exception):
+    """Abstract base for dependency resolution errors"""
+    def __repr__(self):
+        return self.__class__.__name__+repr(self.args)
+
+class VersionConflict(ResolutionError):
+    """An already-installed version conflicts with the requested version"""
+
+class DistributionNotFound(ResolutionError):
+    """A requested distribution was not found"""
+
+class UnknownExtra(ResolutionError):
+    """Distribution doesn't have an "extra feature" of the given name"""
+_provider_factories = {}
+
+PY_MAJOR = sys.version[:3]
+EGG_DIST    = 3
+BINARY_DIST = 2
+SOURCE_DIST = 1
+CHECKOUT_DIST = 0
+DEVELOP_DIST = -1
+
+def register_loader_type(loader_type, provider_factory):
+    """Register `provider_factory` to make providers for `loader_type`
+
+    `loader_type` is the type or class of a PEP 302 ``module.__loader__``,
+    and `provider_factory` is a function that, passed a *module* object,
+    returns an ``IResourceProvider`` for that module.
+    """
+    _provider_factories[loader_type] = provider_factory
+
+def get_provider(moduleOrReq):
+    """Return an IResourceProvider for the named module or requirement"""
+    if isinstance(moduleOrReq,Requirement):
+        return working_set.find(moduleOrReq) or require(str(moduleOrReq))[0]
+    try:
+        module = sys.modules[moduleOrReq]
+    except KeyError:
+        __import__(moduleOrReq)
+        module = sys.modules[moduleOrReq]
+    loader = getattr(module, '__loader__', None)
+    return _find_adapter(_provider_factories, loader)(module)
+
+def _macosx_vers(_cache=[]):
+    if not _cache:
+        import platform
+        version = platform.mac_ver()[0]
+        # fallback for MacPorts
+        if version == '':
+            import plistlib
+            plist = '/System/Library/CoreServices/SystemVersion.plist'
+            if os.path.exists(plist):
+                if hasattr(plistlib, 'readPlist'):
+                    plist_content = plistlib.readPlist(plist)
+                    if 'ProductVersion' in plist_content:
+                        version = plist_content['ProductVersion']
+
+        _cache.append(version.split('.'))
+    return _cache[0]
+
+def _macosx_arch(machine):
+    return {'PowerPC':'ppc', 'Power_Macintosh':'ppc'}.get(machine,machine)
+
+def get_build_platform():
+    """Return this platform's string for platform-specific distributions
+
+    XXX Currently this is the same as ``distutils.util.get_platform()``, but it
+    needs some hacks for Linux and Mac OS X.
+    """
+    try:
+        from distutils.util import get_platform
+    except ImportError:
+        from sysconfig import get_platform
+
+    plat = get_platform()
+    if sys.platform == "darwin" and not plat.startswith('macosx-'):
+        try:
+            version = _macosx_vers()
+            machine = os.uname()[4].replace(" ", "_")
+            return "macosx-%d.%d-%s" % (int(version[0]), int(version[1]),
+                _macosx_arch(machine))
+        except ValueError:
+            # if someone is running a non-Mac darwin system, this will fall
+            # through to the default implementation
+            pass
+    return plat
+
+macosVersionString = re.compile(r"macosx-(\d+)\.(\d+)-(.*)")
+darwinVersionString = re.compile(r"darwin-(\d+)\.(\d+)\.(\d+)-(.*)")
+get_platform = get_build_platform   # XXX backward compat
+
+def compatible_platforms(provided,required):
+    """Can code for the `provided` platform run on the `required` platform?
+
+    Returns true if either platform is ``None``, or the platforms are equal.
+
+    XXX Needs compatibility checks for Linux and other unixy OSes.
+    """
+    if provided is None or required is None or provided==required:
+        return True     # easy case
+
+    # Mac OS X special cases
+    reqMac = macosVersionString.match(required)
+    if reqMac:
+        provMac = macosVersionString.match(provided)
+
+        # is this a Mac package?
+        if not provMac:
+            # this is backwards compatibility for packages built before
+            # setuptools 0.6. All packages built after this point will
+            # use the new macosx designation.
+            provDarwin = darwinVersionString.match(provided)
+            if provDarwin:
+                dversion = int(provDarwin.group(1))
+                macosversion = "%s.%s" % (reqMac.group(1), reqMac.group(2))
+                if dversion == 7 and macosversion >= "10.3" or \
+                    dversion == 8 and macosversion >= "10.4":
+
+                    #import warnings
+                    #warnings.warn("Mac eggs should be rebuilt to "
+                    #    "use the macosx designation instead of darwin.",
+                    #    category=DeprecationWarning)
+                    return True
+            return False    # egg isn't macosx or legacy darwin
+
+        # are they the same major version and machine type?
+        if provMac.group(1) != reqMac.group(1) or \
+            provMac.group(3) != reqMac.group(3):
+            return False
+
+
+
+        # is the required OS major update >= the provided one?
+        if int(provMac.group(2)) > int(reqMac.group(2)):
+            return False
+
+        return True
+
+    # XXX Linux and other platforms' special cases should go here
+    return False
+
+
+def run_script(dist_spec, script_name):
+    """Locate distribution `dist_spec` and run its `script_name` script"""
+    ns = sys._getframe(1).f_globals
+    name = ns['__name__']
+    ns.clear()
+    ns['__name__'] = name
+    require(dist_spec)[0].run_script(script_name, ns)
+
+run_main = run_script   # backward compatibility
+
+def get_distribution(dist):
+    """Return a current distribution object for a Requirement or string"""
+    if isinstance(dist,basestring): dist = Requirement.parse(dist)
+    if isinstance(dist,Requirement): dist = get_provider(dist)
+    if not isinstance(dist,Distribution):
+        raise TypeError("Expected string, Requirement, or Distribution", dist)
+    return dist
+
+def load_entry_point(dist, group, name):
+    """Return `name` entry point of `group` for `dist` or raise ImportError"""
+    return get_distribution(dist).load_entry_point(group, name)
+
+def get_entry_map(dist, group=None):
+    """Return the entry point map for `group`, or the full entry map"""
+    return get_distribution(dist).get_entry_map(group)
+
+def get_entry_info(dist, group, name):
+    """Return the EntryPoint object for `group`+`name`, or ``None``"""
+    return get_distribution(dist).get_entry_info(group, name)
+
+
+class IMetadataProvider:
+
+    def has_metadata(name):
+        """Does the package's distribution contain the named metadata?"""
+
+    def get_metadata(name):
+        """The named metadata resource as a string"""
+
+    def get_metadata_lines(name):
+        """Yield named metadata resource as list of non-blank non-comment lines
+
+       Leading and trailing whitespace is stripped from each line, and lines
+       with ``#`` as the first non-blank character are omitted."""
+
+    def metadata_isdir(name):
+        """Is the named metadata a directory?  (like ``os.path.isdir()``)"""
+
+    def metadata_listdir(name):
+        """List of metadata names in the directory (like ``os.listdir()``)"""
+
+    def run_script(script_name, namespace):
+        """Execute the named script in the supplied namespace dictionary"""
+
+
+
+
+
+
+
+
+
+
+class IResourceProvider(IMetadataProvider):
+    """An object that provides access to package resources"""
+
+    def get_resource_filename(manager, resource_name):
+        """Return a true filesystem path for `resource_name`
+
+        `manager` must be an ``IResourceManager``"""
+
+    def get_resource_stream(manager, resource_name):
+        """Return a readable file-like object for `resource_name`
+
+        `manager` must be an ``IResourceManager``"""
+
+    def get_resource_string(manager, resource_name):
+        """Return a string containing the contents of `resource_name`
+
+        `manager` must be an ``IResourceManager``"""
+
+    def has_resource(resource_name):
+        """Does the package contain the named resource?"""
+
+    def resource_isdir(resource_name):
+        """Is the named resource a directory?  (like ``os.path.isdir()``)"""
+
+    def resource_listdir(resource_name):
+        """List of resource names in the directory (like ``os.listdir()``)"""
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+class WorkingSet(object):
+    """A collection of active distributions on sys.path (or a similar list)"""
+
+    def __init__(self, entries=None):
+        """Create working set from list of path entries (default=sys.path)"""
+        self.entries = []
+        self.entry_keys = {}
+        self.by_key = {}
+        self.callbacks = []
+
+        if entries is None:
+            entries = sys.path
+
+        for entry in entries:
+            self.add_entry(entry)
+
+
+    def add_entry(self, entry):
+        """Add a path item to ``.entries``, finding any distributions on it
+
+        ``find_distributions(entry,True)`` is used to find distributions
+        corresponding to the path entry, and they are added.  `entry` is
+        always appended to ``.entries``, even if it is already present.
+        (This is because ``sys.path`` can contain the same value more than
+        once, and the ``.entries`` of the ``sys.path`` WorkingSet should always
+        equal ``sys.path``.)
+        """
+        self.entry_keys.setdefault(entry, [])
+        self.entries.append(entry)
+        for dist in find_distributions(entry, True):
+            self.add(dist, entry, False)
+
+
+    def __contains__(self,dist):
+        """True if `dist` is the active distribution for its project"""
+        return self.by_key.get(dist.key) == dist
+
+
+
+
+
+    def find(self, req):
+        """Find a distribution matching requirement `req`
+
+        If there is an active distribution for the requested project, this
+        returns it as long as it meets the version requirement specified by
+        `req`.  But, if there is an active distribution for the project and it
+        does *not* meet the `req` requirement, ``VersionConflict`` is raised.
+        If there is no active distribution for the requested project, ``None``
+        is returned.
+        """
+        dist = self.by_key.get(req.key)
+        if dist is not None and dist not in req:
+            raise VersionConflict(dist,req)     # XXX add more info
+        else:
+            return dist
+
+    def iter_entry_points(self, group, name=None):
+        """Yield entry point objects from `group` matching `name`
+
+        If `name` is None, yields all entry points in `group` from all
+        distributions in the working set, otherwise only ones matching
+        both `group` and `name` are yielded (in distribution order).
+        """
+        for dist in self:
+            entries = dist.get_entry_map(group)
+            if name is None:
+                for ep in entries.values():
+                    yield ep
+            elif name in entries:
+                yield entries[name]
+
+    def run_script(self, requires, script_name):
+        """Locate distribution for `requires` and run `script_name` script"""
+        ns = sys._getframe(1).f_globals
+        name = ns['__name__']
+        ns.clear()
+        ns['__name__'] = name
+        self.require(requires)[0].run_script(script_name, ns)
+
+
+
+    def __iter__(self):
+        """Yield distributions for non-duplicate projects in the working set
+
+        The yield order is the order in which the items' path entries were
+        added to the working set.
+        """
+        seen = {}
+        for item in self.entries:
+            for key in self.entry_keys[item]:
+                if key not in seen:
+                    seen[key]=1
+                    yield self.by_key[key]
+
+    def add(self, dist, entry=None, insert=True):
+        """Add `dist` to working set, associated with `entry`
+
+        If `entry` is unspecified, it defaults to the ``.location`` of `dist`.
+        On exit from this routine, `entry` is added to the end of the working
+        set's ``.entries`` (if it wasn't already present).
+
+        `dist` is only added to the working set if it's for a project that
+        doesn't already have a distribution in the set.  If it's added, any
+        callbacks registered with the ``subscribe()`` method will be called.
+        """
+        if insert:
+            dist.insert_on(self.entries, entry)
+
+        if entry is None:
+            entry = dist.location
+        keys = self.entry_keys.setdefault(entry,[])
+        keys2 = self.entry_keys.setdefault(dist.location,[])
+        if dist.key in self.by_key:
+            return      # ignore hidden distros
+
+        self.by_key[dist.key] = dist
+        if dist.key not in keys:
+            keys.append(dist.key)
+        if dist.key not in keys2:
+            keys2.append(dist.key)
+        self._added_new(dist)
+
+    def resolve(self, requirements, env=None, installer=None, replacement=True):
+        """List all distributions needed to (recursively) meet `requirements`
+
+        `requirements` must be a sequence of ``Requirement`` objects.  `env`,
+        if supplied, should be an ``Environment`` instance.  If
+        not supplied, it defaults to all distributions available within any
+        entry or distribution in the working set.  `installer`, if supplied,
+        will be invoked with each requirement that cannot be met by an
+        already-installed distribution; it should return a ``Distribution`` or
+        ``None``.
+        """
+
+        requirements = list(requirements)[::-1]  # set up the stack
+        processed = {}  # set of processed requirements
+        best = {}  # key -> dist
+        to_activate = []
+
+        while requirements:
+            req = requirements.pop(0)   # process dependencies breadth-first
+            if _override_setuptools(req) and replacement:
+                req = Requirement.parse('distribute')
+
+            if req in processed:
+                # Ignore cyclic or redundant dependencies
+                continue
+            dist = best.get(req.key)
+            if dist is None:
+                # Find the best distribution and add it to the map
+                dist = self.by_key.get(req.key)
+                if dist is None:
+                    if env is None:
+                        env = Environment(self.entries)
+                    dist = best[req.key] = env.best_match(req, self, installer)
+                    if dist is None:
+                        #msg = ("The '%s' distribution was not found on this "
+                        #       "system, and is required by this application.")
+                        #raise DistributionNotFound(msg % req)
+
+                        # unfortunately, zc.buildout uses a str(err)
+                        # to get the name of the distribution here..
+                        raise DistributionNotFound(req)
+                to_activate.append(dist)
+            if dist not in req:
+                # Oops, the "best" so far conflicts with a dependency
+                raise VersionConflict(dist,req) # XXX put more info here
+            requirements.extend(dist.requires(req.extras)[::-1])
+            processed[req] = True
+
+        return to_activate    # return list of distros to activate
+
+    def find_plugins(self,
+        plugin_env, full_env=None, installer=None, fallback=True
+    ):
+        """Find all activatable distributions in `plugin_env`
+
+        Example usage::
+
+            distributions, errors = working_set.find_plugins(
+                Environment(plugin_dirlist)
+            )
+            map(working_set.add, distributions)  # add plugins+libs to sys.path
+            print 'Could not load', errors        # display errors
+
+        The `plugin_env` should be an ``Environment`` instance that contains
+        only distributions that are in the project's "plugin directory" or
+        directories. The `full_env`, if supplied, should be an ``Environment``
+        contains all currently-available distributions.  If `full_env` is not
+        supplied, one is created automatically from the ``WorkingSet`` this
+        method is called on, which will typically mean that every directory on
+        ``sys.path`` will be scanned for distributions.
+
+        `installer` is a standard installer callback as used by the
+        ``resolve()`` method. The `fallback` flag indicates whether we should
+        attempt to resolve older versions of a plugin if the newest version
+        cannot be resolved.
+
+        This method returns a 2-tuple: (`distributions`, `error_info`), where
+        `distributions` is a list of the distributions found in `plugin_env`
+        that were loadable, along with any other distributions that are needed
+        to resolve their dependencies.  `error_info` is a dictionary mapping
+        unloadable plugin distributions to an exception instance describing the
+        error that occurred. Usually this will be a ``DistributionNotFound`` or
+        ``VersionConflict`` instance.
+        """
+
+        plugin_projects = list(plugin_env)
+        plugin_projects.sort()  # scan project names in alphabetic order
+
+        error_info = {}
+        distributions = {}
+
+        if full_env is None:
+            env = Environment(self.entries)
+            env += plugin_env
+        else:
+            env = full_env + plugin_env
+
+        shadow_set = self.__class__([])
+        map(shadow_set.add, self)   # put all our entries in shadow_set
+
+        for project_name in plugin_projects:
+
+            for dist in plugin_env[project_name]:
+
+                req = [dist.as_requirement()]
+
+                try:
+                    resolvees = shadow_set.resolve(req, env, installer)
+
+                except ResolutionError,v:
+                    error_info[dist] = v    # save error info
+                    if fallback:
+                        continue    # try the next older version of project
+                    else:
+                        break       # give up on this project, keep going
+
+                else:
+                    map(shadow_set.add, resolvees)
+                    distributions.update(dict.fromkeys(resolvees))
+
+                    # success, no need to try any more versions of this project
+                    break
+
+        distributions = list(distributions)
+        distributions.sort()
+
+        return distributions, error_info
+
+
+
+
+
+    def require(self, *requirements):
+        """Ensure that distributions matching `requirements` are activated
+
+        `requirements` must be a string or a (possibly-nested) sequence
+        thereof, specifying the distributions and versions required.  The
+        return value is a sequence of the distributions that needed to be
+        activated to fulfill the requirements; all relevant distributions are
+        included, even if they were already activated in this working set.
+        """
+
+        needed = self.resolve(parse_requirements(requirements))
+
+        for dist in needed:
+            self.add(dist)
+
+        return needed
+
+
+    def subscribe(self, callback):
+        """Invoke `callback` for all distributions (including existing ones)"""
+        if callback in self.callbacks:
+            return
+        self.callbacks.append(callback)
+        for dist in self:
+            callback(dist)
+
+
+    def _added_new(self, dist):
+        for callback in self.callbacks:
+            callback(dist)
+
+
+
+
+
+
+
+
+
+
+
+class Environment(object):
+    """Searchable snapshot of distributions on a search path"""
+
+    def __init__(self, search_path=None, platform=get_supported_platform(), python=PY_MAJOR):
+        """Snapshot distributions available on a search path
+
+        Any distributions found on `search_path` are added to the environment.
+        `search_path` should be a sequence of ``sys.path`` items.  If not
+        supplied, ``sys.path`` is used.
+
+        `platform` is an optional string specifying the name of the platform
+        that platform-specific distributions must be compatible with.  If
+        unspecified, it defaults to the current platform.  `python` is an
+        optional string naming the desired version of Python (e.g. ``'2.4'``);
+        it defaults to the current version.
+
+        You may explicitly set `platform` (and/or `python`) to ``None`` if you
+        wish to map *all* distributions, not just those compatible with the
+        running platform or Python version.
+        """
+        self._distmap = {}
+        self._cache = {}
+        self.platform = platform
+        self.python = python
+        self.scan(search_path)
+
+    def can_add(self, dist):
+        """Is distribution `dist` acceptable for this environment?
+
+        The distribution must match the platform and python version
+        requirements specified when this environment was created, or False
+        is returned.
+        """
+        return (self.python is None or dist.py_version is None
+            or dist.py_version==self.python) \
+           and compatible_platforms(dist.platform,self.platform)
+
+    def remove(self, dist):
+        """Remove `dist` from the environment"""
+        self._distmap[dist.key].remove(dist)
+
+    def scan(self, search_path=None):
+        """Scan `search_path` for distributions usable in this environment
+
+        Any distributions found are added to the environment.
+        `search_path` should be a sequence of ``sys.path`` items.  If not
+        supplied, ``sys.path`` is used.  Only distributions conforming to
+        the platform/python version defined at initialization are added.
+        """
+        if search_path is None:
+            search_path = sys.path
+
+        for item in search_path:
+            for dist in find_distributions(item):
+                self.add(dist)
+
+    def __getitem__(self,project_name):
+        """Return a newest-to-oldest list of distributions for `project_name`
+        """
+        try:
+            return self._cache[project_name]
+        except KeyError:
+            project_name = project_name.lower()
+            if project_name not in self._distmap:
+                return []
+
+        if project_name not in self._cache:
+            dists = self._cache[project_name] = self._distmap[project_name]
+            _sort_dists(dists)
+
+        return self._cache[project_name]
+
+    def add(self,dist):
+        """Add `dist` if we ``can_add()`` it and it isn't already added"""
+        if self.can_add(dist) and dist.has_version():
+            dists = self._distmap.setdefault(dist.key,[])
+            if dist not in dists:
+                dists.append(dist)
+                if dist.key in self._cache:
+                    _sort_dists(self._cache[dist.key])
+
+
+    def best_match(self, req, working_set, installer=None):
+        """Find distribution best matching `req` and usable on `working_set`
+
+        This calls the ``find(req)`` method of the `working_set` to see if a
+        suitable distribution is already active.  (This may raise
+        ``VersionConflict`` if an unsuitable version of the project is already
+        active in the specified `working_set`.)  If a suitable distribution
+        isn't active, this method returns the newest distribution in the
+        environment that meets the ``Requirement`` in `req`.  If no suitable
+        distribution is found, and `installer` is supplied, then the result of
+        calling the environment's ``obtain(req, installer)`` method will be
+        returned.
+        """
+        dist = working_set.find(req)
+        if dist is not None:
+            return dist
+        for dist in self[req.key]:
+            if dist in req:
+                return dist
+        return self.obtain(req, installer) # try and download/install
+
+    def obtain(self, requirement, installer=None):
+        """Obtain a distribution matching `requirement` (e.g. via download)
+
+        Obtain a distro that matches requirement (e.g. via download).  In the
+        base ``Environment`` class, this routine just returns
+        ``installer(requirement)``, unless `installer` is None, in which case
+        None is returned instead.  This method is a hook that allows subclasses
+        to attempt other ways of obtaining a distribution before falling back
+        to the `installer` argument."""
+        if installer is not None:
+            return installer(requirement)
+
+    def __iter__(self):
+        """Yield the unique project names of the available distributions"""
+        for key in self._distmap.keys():
+            if self[key]: yield key
+
+
+
+
+    def __iadd__(self, other):
+        """In-place addition of a distribution or environment"""
+        if isinstance(other,Distribution):
+            self.add(other)
+        elif isinstance(other,Environment):
+            for project in other:
+                for dist in other[project]:
+                    self.add(dist)
+        else:
+            raise TypeError("Can't add %r to environment" % (other,))
+        return self
+
+    def __add__(self, other):
+        """Add an environment or distribution to an environment"""
+        new = self.__class__([], platform=None, python=None)
+        for env in self, other:
+            new += env
+        return new
+
+
+AvailableDistributions = Environment    # XXX backward compatibility
+
+
+class ExtractionError(RuntimeError):
+    """An error occurred extracting a resource
+
+    The following attributes are available from instances of this exception:
+
+    manager
+        The resource manager that raised this exception
+
+    cache_path
+        The base directory for resource extraction
+
+    original_error
+        The exception instance that caused extraction to fail
+    """
+
+
+
+
+class ResourceManager:
+    """Manage resource extraction and packages"""
+    extraction_path = None
+
+    def __init__(self):
+        self.cached_files = {}
+
+    def resource_exists(self, package_or_requirement, resource_name):
+        """Does the named resource exist?"""
+        return get_provider(package_or_requirement).has_resource(resource_name)
+
+    def resource_isdir(self, package_or_requirement, resource_name):
+        """Is the named resource an existing directory?"""
+        return get_provider(package_or_requirement).resource_isdir(
+            resource_name
+        )
+
+    def resource_filename(self, package_or_requirement, resource_name):
+        """Return a true filesystem path for specified resource"""
+        return get_provider(package_or_requirement).get_resource_filename(
+            self, resource_name
+        )
+
+    def resource_stream(self, package_or_requirement, resource_name):
+        """Return a readable file-like object for specified resource"""
+        return get_provider(package_or_requirement).get_resource_stream(
+            self, resource_name
+        )
+
+    def resource_string(self, package_or_requirement, resource_name):
+        """Return specified resource as a string"""
+        return get_provider(package_or_requirement).get_resource_string(
+            self, resource_name
+        )
+
+    def resource_listdir(self, package_or_requirement, resource_name):
+        """List the contents of the named resource directory"""
+        return get_provider(package_or_requirement).resource_listdir(
+            resource_name
+        )
+
+    def extraction_error(self):
+        """Give an error message for problems extracting file(s)"""
+
+        old_exc = sys.exc_info()[1]
+        cache_path = self.extraction_path or get_default_cache()
+
+        err = ExtractionError("""Can't extract file(s) to egg cache
+
+The following error occurred while trying to extract file(s) to the Python egg
+cache:
+
+  %s
+
+The Python egg cache directory is currently set to:
+
+  %s
+
+Perhaps your account does not have write access to this directory?  You can
+change the cache directory by setting the PYTHON_EGG_CACHE environment
+variable to point to an accessible directory.
+"""         % (old_exc, cache_path)
+        )
+        err.manager        = self
+        err.cache_path     = cache_path
+        err.original_error = old_exc
+        raise err
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+    def get_cache_path(self, archive_name, names=()):
+        """Return absolute location in cache for `archive_name` and `names`
+
+        The parent directory of the resulting path will be created if it does
+        not already exist.  `archive_name` should be the base filename of the
+        enclosing egg (which may not be the name of the enclosing zipfile!),
+        including its ".egg" extension.  `names`, if provided, should be a
+        sequence of path name parts "under" the egg's extraction location.
+
+        This method should only be called by resource providers that need to
+        obtain an extraction location, and only for names they intend to
+        extract, as it tracks the generated names for possible cleanup later.
+        """
+        extract_path = self.extraction_path or get_default_cache()
+        target_path = os.path.join(extract_path, archive_name+'-tmp', *names)
+        try:
+            _bypass_ensure_directory(target_path)
+        except:
+            self.extraction_error()
+
+        self.cached_files[target_path] = 1
+        return target_path
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+    def postprocess(self, tempname, filename):
+        """Perform any platform-specific postprocessing of `tempname`
+
+        This is where Mac header rewrites should be done; other platforms don't
+        have anything special they should do.
+
+        Resource providers should call this method ONLY after successfully
+        extracting a compressed resource.  They must NOT call it on resources
+        that are already in the filesystem.
+
+        `tempname` is the current (temporary) name of the file, and `filename`
+        is the name it will be renamed to by the caller after this routine
+        returns.
+        """
+
+        if os.name == 'posix':
+            # Make the resource executable
+            mode = ((os.stat(tempname).st_mode) | 0555) & 07777
+            os.chmod(tempname, mode)
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+    def set_extraction_path(self, path):
+        """Set the base path where resources will be extracted to, if needed.
+
+        If you do not call this routine before any extractions take place, the
+        path defaults to the return value of ``get_default_cache()``.  (Which
+        is based on the ``PYTHON_EGG_CACHE`` environment variable, with various
+        platform-specific fallbacks.  See that routine's documentation for more
+        details.)
+
+        Resources are extracted to subdirectories of this path based upon
+        information given by the ``IResourceProvider``.  You may set this to a
+        temporary directory, but then you must call ``cleanup_resources()`` to
+        delete the extracted files when done.  There is no guarantee that
+        ``cleanup_resources()`` will be able to remove all extracted files.
+
+        (Note: you may not change the extraction path for a given resource
+        manager once resources have been extracted, unless you first call
+        ``cleanup_resources()``.)
+        """
+        if self.cached_files:
+            raise ValueError(
+                "Can't change extraction path, files already extracted"
+            )
+
+        self.extraction_path = path
+
+    def cleanup_resources(self, force=False):
+        """
+        Delete all extracted resource files and directories, returning a list
+        of the file and directory names that could not be successfully removed.
+        This function does not have any concurrency protection, so it should
+        generally only be called when the extraction path is a temporary
+        directory exclusive to a single process.  This method is not
+        automatically called; you must call it explicitly or register it as an
+        ``atexit`` function if you wish to ensure cleanup of a temporary
+        directory used for extractions.
+        """
+        # XXX
+
+
+
+def get_default_cache():
+    """Determine the default cache location
+
+    This returns the ``PYTHON_EGG_CACHE`` environment variable, if set.
+    Otherwise, on Windows, it returns a "Python-Eggs" subdirectory of the
+    "Application Data" directory.  On all other systems, it's "~/.python-eggs".
+    """
+    try:
+        return os.environ['PYTHON_EGG_CACHE']
+    except KeyError:
+        pass
+
+    if os.name!='nt':
+        return os.path.expanduser('~/.python-eggs')
+
+    app_data = 'Application Data'   # XXX this may be locale-specific!
+    app_homes = [
+        (('APPDATA',), None),       # best option, should be locale-safe
+        (('USERPROFILE',), app_data),
+        (('HOMEDRIVE','HOMEPATH'), app_data),
+        (('HOMEPATH',), app_data),
+        (('HOME',), None),
+        (('WINDIR',), app_data),    # 95/98/ME
+    ]
+
+    for keys, subdir in app_homes:
+        dirname = ''
+        for key in keys:
+            if key in os.environ:
+                dirname = os.path.join(dirname, os.environ[key])
+            else:
+                break
+        else:
+            if subdir:
+                dirname = os.path.join(dirname,subdir)
+            return os.path.join(dirname, 'Python-Eggs')
+    else:
+        raise RuntimeError(
+            "Please set the PYTHON_EGG_CACHE enviroment variable"
+        )
+
+def safe_name(name):
+    """Convert an arbitrary string to a standard distribution name
+
+    Any runs of non-alphanumeric/. characters are replaced with a single '-'.
+    """
+    return re.sub('[^A-Za-z0-9.]+', '-', name)
+
+
+def safe_version(version):
+    """Convert an arbitrary string to a standard version string
+
+    Spaces become dots, and all other non-alphanumeric characters become
+    dashes, with runs of multiple dashes condensed to a single dash.
+    """
+    version = version.replace(' ','.')
+    return re.sub('[^A-Za-z0-9.]+', '-', version)
+
+
+def safe_extra(extra):
+    """Convert an arbitrary string to a standard 'extra' name
+
+    Any runs of non-alphanumeric characters are replaced with a single '_',
+    and the result is always lowercased.
+    """
+    return re.sub('[^A-Za-z0-9.]+', '_', extra).lower()
+
+
+def to_filename(name):
+    """Convert a project or version name to its filename-escaped form
+
+    Any '-' characters are currently replaced with '_'.
+    """
+    return name.replace('-','_')
+
+
+
+
+
+
+
+
+class NullProvider:
+    """Try to implement resources and metadata for arbitrary PEP 302 loaders"""
+
+    egg_name = None
+    egg_info = None
+    loader = None
+
+    def __init__(self, module):
+        self.loader = getattr(module, '__loader__', None)
+        self.module_path = os.path.dirname(getattr(module, '__file__', ''))
+
+    def get_resource_filename(self, manager, resource_name):
+        return self._fn(self.module_path, resource_name)
+
+    def get_resource_stream(self, manager, resource_name):
+        return StringIO(self.get_resource_string(manager, resource_name))
+
+    def get_resource_string(self, manager, resource_name):
+        return self._get(self._fn(self.module_path, resource_name))
+
+    def has_resource(self, resource_name):
+        return self._has(self._fn(self.module_path, resource_name))
+
+    def has_metadata(self, name):
+        return self.egg_info and self._has(self._fn(self.egg_info,name))
+
+    if sys.version_info <= (3,):
+        def get_metadata(self, name):
+            if not self.egg_info:
+                return ""
+            return self._get(self._fn(self.egg_info,name))
+    else:
+        def get_metadata(self, name):
+            if not self.egg_info:
+                return ""
+            return self._get(self._fn(self.egg_info,name)).decode("utf-8")
+
+    def get_metadata_lines(self, name):
+        return yield_lines(self.get_metadata(name))
+
+    def resource_isdir(self,resource_name):
+        return self._isdir(self._fn(self.module_path, resource_name))
+
+    def metadata_isdir(self,name):
+        return self.egg_info and self._isdir(self._fn(self.egg_info,name))
+
+
+    def resource_listdir(self,resource_name):
+        return self._listdir(self._fn(self.module_path,resource_name))
+
+    def metadata_listdir(self,name):
+        if self.egg_info:
+            return self._listdir(self._fn(self.egg_info,name))
+        return []
+
+    def run_script(self,script_name,namespace):
+        script = 'scripts/'+script_name
+        if not self.has_metadata(script):
+            raise ResolutionError("No script named %r" % script_name)
+        script_text = self.get_metadata(script).replace('\r\n','\n')
+        script_text = script_text.replace('\r','\n')
+        script_filename = self._fn(self.egg_info,script)
+        namespace['__file__'] = script_filename
+        if os.path.exists(script_filename):
+            execfile(script_filename, namespace, namespace)
+        else:
+            from linecache import cache
+            cache[script_filename] = (
+                len(script_text), 0, script_text.split('\n'), script_filename
+            )
+            script_code = compile(script_text,script_filename,'exec')
+            exec script_code in namespace, namespace
+
+    def _has(self, path):
+        raise NotImplementedError(
+            "Can't perform this operation for unregistered loader type"
+        )
+
+    def _isdir(self, path):
+        raise NotImplementedError(
+            "Can't perform this operation for unregistered loader type"
+        )
+
+    def _listdir(self, path):
+        raise NotImplementedError(
+            "Can't perform this operation for unregistered loader type"
+        )
+
+    def _fn(self, base, resource_name):
+        if resource_name:
+            return os.path.join(base, *resource_name.split('/'))
+        return base
+
+    def _get(self, path):
+        if hasattr(self.loader, 'get_data'):
+            return self.loader.get_data(path)
+        raise NotImplementedError(
+            "Can't perform this operation for loaders without 'get_data()'"
+        )
+
+register_loader_type(object, NullProvider)
+
+
+class EggProvider(NullProvider):
+    """Provider based on a virtual filesystem"""
+
+    def __init__(self,module):
+        NullProvider.__init__(self,module)
+        self._setup_prefix()
+
+    def _setup_prefix(self):
+        # we assume here that our metadata may be nested inside a "basket"
+        # of multiple eggs; that's why we use module_path instead of .archive
+        path = self.module_path
+        old = None
+        while path!=old:
+            if path.lower().endswith('.egg'):
+                self.egg_name = os.path.basename(path)
+                self.egg_info = os.path.join(path, 'EGG-INFO')
+                self.egg_root = path
+                break
+            old = path
+            path, base = os.path.split(path)
+
+
+
+
+
+
+class DefaultProvider(EggProvider):
+    """Provides access to package resources in the filesystem"""
+
+    def _has(self, path):
+        return os.path.exists(path)
+
+    def _isdir(self,path):
+        return os.path.isdir(path)
+
+    def _listdir(self,path):
+        return os.listdir(path)
+
+    def get_resource_stream(self, manager, resource_name):
+        return open(self._fn(self.module_path, resource_name), 'rb')
+
+    def _get(self, path):
+        stream = open(path, 'rb')
+        try:
+            return stream.read()
+        finally:
+            stream.close()
+
+register_loader_type(type(None), DefaultProvider)
+
+
+class EmptyProvider(NullProvider):
+    """Provider that returns nothing for all requests"""
+
+    _isdir = _has = lambda self,path: False
+    _get          = lambda self,path: ''
+    _listdir      = lambda self,path: []
+    module_path   = None
+
+    def __init__(self):
+        pass
+
+empty_provider = EmptyProvider()
+
+
+
+
+class ZipProvider(EggProvider):
+    """Resource support for zips and eggs"""
+
+    eagers = None
+
+    def __init__(self, module):
+        EggProvider.__init__(self,module)
+        self.zipinfo = zipimport._zip_directory_cache[self.loader.archive]
+        self.zip_pre = self.loader.archive+os.sep
+
+    def _zipinfo_name(self, fspath):
+        # Convert a virtual filename (full path to file) into a zipfile subpath
+        # usable with the zipimport directory cache for our target archive
+        if fspath.startswith(self.zip_pre):
+            return fspath[len(self.zip_pre):]
+        raise AssertionError(
+            "%s is not a subpath of %s" % (fspath,self.zip_pre)
+        )
+
+    def _parts(self,zip_path):
+        # Convert a zipfile subpath into an egg-relative path part list
+        fspath = self.zip_pre+zip_path  # pseudo-fs path
+        if fspath.startswith(self.egg_root+os.sep):
+            return fspath[len(self.egg_root)+1:].split(os.sep)
+        raise AssertionError(
+            "%s is not a subpath of %s" % (fspath,self.egg_root)
+        )
+
+    def get_resource_filename(self, manager, resource_name):
+        if not self.egg_name:
+            raise NotImplementedError(
+                "resource_filename() only supported for .egg, not .zip"
+            )
+        # no need to lock for extraction, since we use temp names
+        zip_path = self._resource_to_zip(resource_name)
+        eagers = self._get_eager_resources()
+        if '/'.join(self._parts(zip_path)) in eagers:
+            for name in eagers:
+                self._extract_resource(manager, self._eager_to_zip(name))
+        return self._extract_resource(manager, zip_path)
+
+    def _extract_resource(self, manager, zip_path):
+
+        if zip_path in self._index():
+            for name in self._index()[zip_path]:
+                last = self._extract_resource(
+                    manager, os.path.join(zip_path, name)
+                )
+            return os.path.dirname(last)  # return the extracted directory name
+
+        zip_stat = self.zipinfo[zip_path]
+        t,d,size = zip_stat[5], zip_stat[6], zip_stat[3]
+        date_time = (
+            (d>>9)+1980, (d>>5)&0xF, d&0x1F,                      # ymd
+            (t&0xFFFF)>>11, (t>>5)&0x3F, (t&0x1F) * 2, 0, 0, -1   # hms, etc.
+        )
+        timestamp = time.mktime(date_time)
+
+        try:
+            if not WRITE_SUPPORT:
+                raise IOError('"os.rename" and "os.unlink" are not supported '
+                              'on this platform')
+
+            real_path = manager.get_cache_path(
+                self.egg_name, self._parts(zip_path)
+            )
+
+            if os.path.isfile(real_path):
+                stat = os.stat(real_path)
+                if stat.st_size==size and stat.st_mtime==timestamp:
+                    # size and stamp match, don't bother extracting
+                    return real_path
+
+            outf, tmpnam = _mkstemp(".$extract", dir=os.path.dirname(real_path))
+            os.write(outf, self.loader.get_data(zip_path))
+            os.close(outf)
+            utime(tmpnam, (timestamp,timestamp))
+            manager.postprocess(tmpnam, real_path)
+
+            try:
+                rename(tmpnam, real_path)
+
+            except os.error:
+                if os.path.isfile(real_path):
+                    stat = os.stat(real_path)
+
+                    if stat.st_size==size and stat.st_mtime==timestamp:
+                        # size and stamp match, somebody did it just ahead of
+                        # us, so we're done
+                        return real_path
+                    elif os.name=='nt':     # Windows, del old file and retry
+                        unlink(real_path)
+                        rename(tmpnam, real_path)
+                        return real_path
+                raise
+
+        except os.error:
+            manager.extraction_error()  # report a user-friendly error
+
+        return real_path
+
+    def _get_eager_resources(self):
+        if self.eagers is None:
+            eagers = []
+            for name in ('native_libs.txt', 'eager_resources.txt'):
+                if self.has_metadata(name):
+                    eagers.extend(self.get_metadata_lines(name))
+            self.eagers = eagers
+        return self.eagers
+
+    def _index(self):
+        try:
+            return self._dirindex
+        except AttributeError:
+            ind = {}
+            for path in self.zipinfo:
+                parts = path.split(os.sep)
+                while parts:
+                    parent = os.sep.join(parts[:-1])
+                    if parent in ind:
+                        ind[parent].append(parts[-1])
+                        break
+                    else:
+                        ind[parent] = [parts.pop()]
+            self._dirindex = ind
+            return ind
+
+    def _has(self, fspath):
+        zip_path = self._zipinfo_name(fspath)
+        return zip_path in self.zipinfo or zip_path in self._index()
+
+    def _isdir(self,fspath):
+        return self._zipinfo_name(fspath) in self._index()
+
+    def _listdir(self,fspath):
+        return list(self._index().get(self._zipinfo_name(fspath), ()))
+
+    def _eager_to_zip(self,resource_name):
+        return self._zipinfo_name(self._fn(self.egg_root,resource_name))
+
+    def _resource_to_zip(self,resource_name):
+        return self._zipinfo_name(self._fn(self.module_path,resource_name))
+
+register_loader_type(zipimport.zipimporter, ZipProvider)
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+class FileMetadata(EmptyProvider):
+    """Metadata handler for standalone PKG-INFO files
+
+    Usage::
+
+        metadata = FileMetadata("/path/to/PKG-INFO")
+
+    This provider rejects all data and metadata requests except for PKG-INFO,
+    which is treated as existing, and will be the contents of the file at
+    the provided location.
+    """
+
+    def __init__(self,path):
+        self.path = path
+
+    def has_metadata(self,name):
+        return name=='PKG-INFO'
+
+    def get_metadata(self,name):
+        if name=='PKG-INFO':
+            f = open(self.path,'rU')
+            metadata = f.read()
+            f.close()
+            return metadata
+        raise KeyError("No metadata except PKG-INFO is available")
+
+    def get_metadata_lines(self,name):
+        return yield_lines(self.get_metadata(name))
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+class PathMetadata(DefaultProvider):
+    """Metadata provider for egg directories
+
+    Usage::
+
+        # Development eggs:
+
+        egg_info = "/path/to/PackageName.egg-info"
+        base_dir = os.path.dirname(egg_info)
+        metadata = PathMetadata(base_dir, egg_info)
+        dist_name = os.path.splitext(os.path.basename(egg_info))[0]
+        dist = Distribution(basedir,project_name=dist_name,metadata=metadata)
+
+        # Unpacked egg directories:
+
+        egg_path = "/path/to/PackageName-ver-pyver-etc.egg"
+        metadata = PathMetadata(egg_path, os.path.join(egg_path,'EGG-INFO'))
+        dist = Distribution.from_filename(egg_path, metadata=metadata)
+    """
+
+    def __init__(self, path, egg_info):
+        self.module_path = path
+        self.egg_info = egg_info
+
+
+class EggMetadata(ZipProvider):
+    """Metadata provider for .egg files"""
+
+    def __init__(self, importer):
+        """Create a metadata provider from a zipimporter"""
+
+        self.zipinfo = zipimport._zip_directory_cache[importer.archive]
+        self.zip_pre = importer.archive+os.sep
+        self.loader = importer
+        if importer.prefix:
+            self.module_path = os.path.join(importer.archive, importer.prefix)
+        else:
+            self.module_path = importer.archive
+        self._setup_prefix()
+
+
+class ImpWrapper:
+    """PEP 302 Importer that wraps Python's "normal" import algorithm"""
+
+    def __init__(self, path=None):
+        self.path = path
+
+    def find_module(self, fullname, path=None):
+        subname = fullname.split(".")[-1]
+        if subname != fullname and self.path is None:
+            return None
+        if self.path is None:
+            path = None
+        else:
+            path = [self.path]
+        try:
+            file, filename, etc = imp.find_module(subname, path)
+        except ImportError:
+            return None
+        return ImpLoader(file, filename, etc)
+
+
+class ImpLoader:
+    """PEP 302 Loader that wraps Python's "normal" import algorithm"""
+
+    def __init__(self, file, filename, etc):
+        self.file = file
+        self.filename = filename
+        self.etc = etc
+
+    def load_module(self, fullname):
+        try:
+            mod = imp.load_module(fullname, self.file, self.filename, self.etc)
+        finally:
+            if self.file: self.file.close()
+        # Note: we don't set __loader__ because we want the module to look
+        # normal; i.e. this is just a wrapper for standard import machinery
+        return mod
+
+
+
+
+def get_importer(path_item):
+    """Retrieve a PEP 302 "importer" for the given path item
+
+    If there is no importer, this returns a wrapper around the builtin import
+    machinery.  The returned importer is only cached if it was created by a
+    path hook.
+    """
+    try:
+        importer = sys.path_importer_cache[path_item]
+    except KeyError:
+        for hook in sys.path_hooks:
+            try:
+                importer = hook(path_item)
+            except ImportError:
+                pass
+            else:
+                break
+        else:
+            importer = None
+
+    sys.path_importer_cache.setdefault(path_item,importer)
+    if importer is None:
+        try:
+            importer = ImpWrapper(path_item)
+        except ImportError:
+            pass
+    return importer
+
+try:
+    from pkgutil import get_importer, ImpImporter
+except ImportError:
+    pass    # Python 2.3 or 2.4, use our own implementation
+else:
+    ImpWrapper = ImpImporter    # Python 2.5, use pkgutil's implementation
+    del ImpLoader, ImpImporter
+
+
+
+
+
+
+_distribution_finders = {}
+
+def register_finder(importer_type, distribution_finder):
+    """Register `distribution_finder` to find distributions in sys.path items
+
+    `importer_type` is the type or class of a PEP 302 "Importer" (sys.path item
+    handler), and `distribution_finder` is a callable that, passed a path
+    item and the importer instance, yields ``Distribution`` instances found on
+    that path item.  See ``pkg_resources.find_on_path`` for an example."""
+    _distribution_finders[importer_type] = distribution_finder
+
+
+def find_distributions(path_item, only=False):
+    """Yield distributions accessible via `path_item`"""
+    importer = get_importer(path_item)
+    finder = _find_adapter(_distribution_finders, importer)
+    return finder(importer, path_item, only)
+
+def find_in_zip(importer, path_item, only=False):
+    metadata = EggMetadata(importer)
+    if metadata.has_metadata('PKG-INFO'):
+        yield Distribution.from_filename(path_item, metadata=metadata)
+    if only:
+        return  # don't yield nested distros
+    for subitem in metadata.resource_listdir('/'):
+        if subitem.endswith('.egg'):
+            subpath = os.path.join(path_item, subitem)
+            for dist in find_in_zip(zipimport.zipimporter(subpath), subpath):
+                yield dist
+
+register_finder(zipimport.zipimporter, find_in_zip)
+
+def StringIO(*args, **kw):
+    """Thunk to load the real StringIO on demand"""
+    global StringIO
+    try:
+        from cStringIO import StringIO
+    except ImportError:
+        from StringIO import StringIO
+    return StringIO(*args,**kw)
+
+def find_nothing(importer, path_item, only=False):
+    return ()
+register_finder(object,find_nothing)
+
+def find_on_path(importer, path_item, only=False):
+    """Yield distributions accessible on a sys.path directory"""
+    path_item = _normalize_cached(path_item)
+
+    if os.path.isdir(path_item) and os.access(path_item, os.R_OK):
+        if path_item.lower().endswith('.egg'):
+            # unpacked egg
+            yield Distribution.from_filename(
+                path_item, metadata=PathMetadata(
+                    path_item, os.path.join(path_item,'EGG-INFO')
+                )
+            )
+        else:
+            # scan for .egg and .egg-info in directory
+            for entry in os.listdir(path_item):
+                lower = entry.lower()
+                if lower.endswith('.egg-info'):
+                    fullpath = os.path.join(path_item, entry)
+                    if os.path.isdir(fullpath):
+                        # egg-info directory, allow getting metadata
+                        metadata = PathMetadata(path_item, fullpath)
+                    else:
+                        metadata = FileMetadata(fullpath)
+                    yield Distribution.from_location(
+                        path_item,entry,metadata,precedence=DEVELOP_DIST
+                    )
+                elif not only and lower.endswith('.egg'):
+                    for dist in find_distributions(os.path.join(path_item, entry)):
+                        yield dist
+                elif not only and lower.endswith('.egg-link'):
+                    for line in open(os.path.join(path_item, entry)):
+                        if not line.strip(): continue
+                        for item in find_distributions(os.path.join(path_item,line.rstrip())):
+                            yield item
+                        break
+register_finder(ImpWrapper,find_on_path)
+
+_namespace_handlers = {}
+_namespace_packages = {}
+
+def register_namespace_handler(importer_type, namespace_handler):
+    """Register `namespace_handler` to declare namespace packages
+
+    `importer_type` is the type or class of a PEP 302 "Importer" (sys.path item
+    handler), and `namespace_handler` is a callable like this::
+
+        def namespace_handler(importer,path_entry,moduleName,module):
+            # return a path_entry to use for child packages
+
+    Namespace handlers are only called if the importer object has already
+    agreed that it can handle the relevant path item, and they should only
+    return a subpath if the module __path__ does not already contain an
+    equivalent subpath.  For an example namespace handler, see
+    ``pkg_resources.file_ns_handler``.
+    """
+    _namespace_handlers[importer_type] = namespace_handler
+
+def _handle_ns(packageName, path_item):
+    """Ensure that named package includes a subpath of path_item (if needed)"""
+    importer = get_importer(path_item)
+    if importer is None:
+        return None
+    loader = importer.find_module(packageName)
+    if loader is None:
+        return None
+    module = sys.modules.get(packageName)
+    if module is None:
+        module = sys.modules[packageName] = types.ModuleType(packageName)
+        module.__path__ = []; _set_parent_ns(packageName)
+    elif not hasattr(module,'__path__'):
+        raise TypeError("Not a package:", packageName)
+    handler = _find_adapter(_namespace_handlers, importer)
+    subpath = handler(importer,path_item,packageName,module)
+    if subpath is not None:
+        path = module.__path__; path.append(subpath)
+        loader.load_module(packageName); module.__path__ = path
+    return subpath
+
+def declare_namespace(packageName):
+    """Declare that package 'packageName' is a namespace package"""
+
+    imp.acquire_lock()
+    try:
+        if packageName in _namespace_packages:
+            return
+
+        path, parent = sys.path, None
+        if '.' in packageName:
+            parent = '.'.join(packageName.split('.')[:-1])
+            declare_namespace(parent)
+            __import__(parent)
+            try:
+                path = sys.modules[parent].__path__
+            except AttributeError:
+                raise TypeError("Not a package:", parent)
+
+        # Track what packages are namespaces, so when new path items are added,
+        # they can be updated
+        _namespace_packages.setdefault(parent,[]).append(packageName)
+        _namespace_packages.setdefault(packageName,[])
+
+        for path_item in path:
+            # Ensure all the parent's path items are reflected in the child,
+            # if they apply
+            _handle_ns(packageName, path_item)
+
+    finally:
+        imp.release_lock()
+
+def fixup_namespace_packages(path_item, parent=None):
+    """Ensure that previously-declared namespace packages include path_item"""
+    imp.acquire_lock()
+    try:
+        for package in _namespace_packages.get(parent,()):
+            subpath = _handle_ns(package, path_item)
+            if subpath: fixup_namespace_packages(subpath,package)
+    finally:
+        imp.release_lock()
+
+def file_ns_handler(importer, path_item, packageName, module):
+    """Compute an ns-package subpath for a filesystem or zipfile importer"""
+
+    subpath = os.path.join(path_item, packageName.split('.')[-1])
+    normalized = _normalize_cached(subpath)
+    for item in module.__path__:
+        if _normalize_cached(item)==normalized:
+            break
+    else:
+        # Only return the path if it's not already there
+        return subpath
+
+register_namespace_handler(ImpWrapper,file_ns_handler)
+register_namespace_handler(zipimport.zipimporter,file_ns_handler)
+
+
+def null_ns_handler(importer, path_item, packageName, module):
+    return None
+
+register_namespace_handler(object,null_ns_handler)
+
+
+def normalize_path(filename):
+    """Normalize a file/dir name for comparison purposes"""
+    return os.path.normcase(os.path.realpath(filename))
+
+def _normalize_cached(filename,_cache={}):
+    try:
+        return _cache[filename]
+    except KeyError:
+        _cache[filename] = result = normalize_path(filename)
+        return result
+
+def _set_parent_ns(packageName):
+    parts = packageName.split('.')
+    name = parts.pop()
+    if parts:
+        parent = '.'.join(parts)
+        setattr(sys.modules[parent], name, sys.modules[packageName])
+
+
+def yield_lines(strs):
+    """Yield non-empty/non-comment lines of a ``basestring`` or sequence"""
+    if isinstance(strs,basestring):
+        for s in strs.splitlines():
+            s = s.strip()
+            if s and not s.startswith('#'):     # skip blank lines/comments
+                yield s
+    else:
+        for ss in strs:
+            for s in yield_lines(ss):
+                yield s
+
+LINE_END = re.compile(r"\s*(#.*)?$").match         # whitespace and comment
+CONTINUE = re.compile(r"\s*\\\s*(#.*)?$").match    # line continuation
+DISTRO   = re.compile(r"\s*((\w|[-.])+)").match    # Distribution or extra
+VERSION  = re.compile(r"\s*(<=?|>=?|==|!=)\s*((\w|[-.])+)").match  # ver. info
+COMMA    = re.compile(r"\s*,").match               # comma between items
+OBRACKET = re.compile(r"\s*\[").match
+CBRACKET = re.compile(r"\s*\]").match
+MODULE   = re.compile(r"\w+(\.\w+)*$").match
+EGG_NAME = re.compile(
+    r"(?P<name>[^-]+)"
+    r"( -(?P<ver>[^-]+) (-py(?P<pyver>[^-]+) (-(?P<plat>.+))? )? )?",
+    re.VERBOSE | re.IGNORECASE
+).match
+
+component_re = re.compile(r'(\d+ | [a-z]+ | \.| -)', re.VERBOSE)
+replace = {'pre':'c', 'preview':'c','-':'final-','rc':'c','dev':'@'}.get
+
+def _parse_version_parts(s):
+    for part in component_re.split(s):
+        part = replace(part,part)
+        if not part or part=='.':
+            continue
+        if part[:1] in '0123456789':
+            yield part.zfill(8)    # pad for numeric comparison
+        else:
+            yield '*'+part
+
+    yield '*final'  # ensure that alpha/beta/candidate are before final
+
+def parse_version(s):
+    """Convert a version string to a chronologically-sortable key
+
+    This is a rough cross between distutils' StrictVersion and LooseVersion;
+    if you give it versions that would work with StrictVersion, then it behaves
+    the same; otherwise it acts like a slightly-smarter LooseVersion. It is
+    *possible* to create pathological version coding schemes that will fool
+    this parser, but they should be very rare in practice.
+
+    The returned value will be a tuple of strings.  Numeric portions of the
+    version are padded to 8 digits so they will compare numerically, but
+    without relying on how numbers compare relative to strings.  Dots are
+    dropped, but dashes are retained.  Trailing zeros between alpha segments
+    or dashes are suppressed, so that e.g. "2.4.0" is considered the same as
+    "2.4". Alphanumeric parts are lower-cased.
+
+    The algorithm assumes that strings like "-" and any alpha string that
+    alphabetically follows "final"  represents a "patch level".  So, "2.4-1"
+    is assumed to be a branch or patch of "2.4", and therefore "2.4.1" is
+    considered newer than "2.4-1", which in turn is newer than "2.4".
+
+    Strings like "a", "b", "c", "alpha", "beta", "candidate" and so on (that
+    come before "final" alphabetically) are assumed to be pre-release versions,
+    so that the version "2.4" is considered newer than "2.4a1".
+
+    Finally, to handle miscellaneous cases, the strings "pre", "preview", and
+    "rc" are treated as if they were "c", i.e. as though they were release
+    candidates, and therefore are not as new as a version string that does not
+    contain them, and "dev" is replaced with an '@' so that it sorts lower than
+    than any other pre-release tag.
+    """
+    parts = []
+    for part in _parse_version_parts(s.lower()):
+        if part.startswith('*'):
+            if part<'*final':   # remove '-' before a prerelease tag
+                while parts and parts[-1]=='*final-': parts.pop()
+            # remove trailing zeros from each series of numeric parts
+            while parts and parts[-1]=='00000000':
+                parts.pop()
+        parts.append(part)
+    return tuple(parts)
+
+class EntryPoint(object):
+    """Object representing an advertised importable object"""
+
+    def __init__(self, name, module_name, attrs=(), extras=(), dist=None):
+        if not MODULE(module_name):
+            raise ValueError("Invalid module name", module_name)
+        self.name = name
+        self.module_name = module_name
+        self.attrs = tuple(attrs)
+        self.extras = Requirement.parse(("x[%s]" % ','.join(extras))).extras
+        self.dist = dist
+
+    def __str__(self):
+        s = "%s = %s" % (self.name, self.module_name)
+        if self.attrs:
+            s += ':' + '.'.join(self.attrs)
+        if self.extras:
+            s += ' [%s]' % ','.join(self.extras)
+        return s
+
+    def __repr__(self):
+        return "EntryPoint.parse(%r)" % str(self)
+
+    def load(self, require=True, env=None, installer=None):
+        if require: self.require(env, installer)
+        entry = __import__(self.module_name, globals(),globals(), ['__name__'])
+        for attr in self.attrs:
+            try:
+                entry = getattr(entry,attr)
+            except AttributeError:
+                raise ImportError("%r has no %r attribute" % (entry,attr))
+        return entry
+
+    def require(self, env=None, installer=None):
+        if self.extras and not self.dist:
+            raise UnknownExtra("Can't require() without a distribution", self)
+        map(working_set.add,
+            working_set.resolve(self.dist.requires(self.extras),env,installer))
+
+
+
+    #@classmethod
+    def parse(cls, src, dist=None):
+        """Parse a single entry point from string `src`
+
+        Entry point syntax follows the form::
+
+            name = some.module:some.attr [extra1,extra2]
+
+        The entry name and module name are required, but the ``:attrs`` and
+        ``[extras]`` parts are optional
+        """
+        try:
+            attrs = extras = ()
+            name,value = src.split('=',1)
+            if '[' in value:
+                value,extras = value.split('[',1)
+                req = Requirement.parse("x["+extras)
+                if req.specs: raise ValueError
+                extras = req.extras
+            if ':' in value:
+                value,attrs = value.split(':',1)
+                if not MODULE(attrs.rstrip()):
+                    raise ValueError
+                attrs = attrs.rstrip().split('.')
+        except ValueError:
+            raise ValueError(
+                "EntryPoint must be in 'name=module:attrs [extras]' format",
+                src
+            )
+        else:
+            return cls(name.strip(), value.strip(), attrs, extras, dist)
+
+    parse = classmethod(parse)
+
+
+
+
+
+
+
+
+    #@classmethod
+    def parse_group(cls, group, lines, dist=None):
+        """Parse an entry point group"""
+        if not MODULE(group):
+            raise ValueError("Invalid group name", group)
+        this = {}
+        for line in yield_lines(lines):
+            ep = cls.parse(line, dist)
+            if ep.name in this:
+                raise ValueError("Duplicate entry point", group, ep.name)
+            this[ep.name]=ep
+        return this
+
+    parse_group = classmethod(parse_group)
+
+    #@classmethod
+    def parse_map(cls, data, dist=None):
+        """Parse a map of entry point groups"""
+        if isinstance(data,dict):
+            data = data.items()
+        else:
+            data = split_sections(data)
+        maps = {}
+        for group, lines in data:
+            if group is None:
+                if not lines:
+                    continue
+                raise ValueError("Entry points must be listed in groups")
+            group = group.strip()
+            if group in maps:
+                raise ValueError("Duplicate group name", group)
+            maps[group] = cls.parse_group(group, lines, dist)
+        return maps
+
+    parse_map = classmethod(parse_map)
+
+
+def _remove_md5_fragment(location):
+    if not location:
+        return ''
+    parsed = urlparse(location)
+    if parsed[-1].startswith('md5='):
+        return urlunparse(parsed[:-1] + ('',))
+    return location
+
+
+class Distribution(object):
+    """Wrap an actual or potential sys.path entry w/metadata"""
+    def __init__(self,
+        location=None, metadata=None, project_name=None, version=None,
+        py_version=PY_MAJOR, platform=None, precedence = EGG_DIST
+    ):
+        self.project_name = safe_name(project_name or 'Unknown')
+        if version is not None:
+            self._version = safe_version(version)
+        self.py_version = py_version
+        self.platform = platform
+        self.location = location
+        self.precedence = precedence
+        self._provider = metadata or empty_provider
+
+    #@classmethod
+    def from_location(cls,location,basename,metadata=None,**kw):
+        project_name, version, py_version, platform = [None]*4
+        basename, ext = os.path.splitext(basename)
+        if ext.lower() in (".egg",".egg-info"):
+            match = EGG_NAME(basename)
+            if match:
+                project_name, version, py_version, platform = match.group(
+                    'name','ver','pyver','plat'
+                )
+        return cls(
+            location, metadata, project_name=project_name, version=version,
+            py_version=py_version, platform=platform, **kw
+        )
+    from_location = classmethod(from_location)
+
+
+    hashcmp = property(
+        lambda self: (
+            getattr(self,'parsed_version',()),
+            self.precedence,
+            self.key,
+            _remove_md5_fragment(self.location),
+            self.py_version,
+            self.platform
+        )
+    )
+    def __hash__(self): return hash(self.hashcmp)
+    def __lt__(self, other):
+        return self.hashcmp < other.hashcmp
+    def __le__(self, other):
+        return self.hashcmp <= other.hashcmp
+    def __gt__(self, other):
+        return self.hashcmp > other.hashcmp
+    def __ge__(self, other):
+        return self.hashcmp >= other.hashcmp
+    def __eq__(self, other):
+        if not isinstance(other, self.__class__):
+            # It's not a Distribution, so they are not equal
+            return False
+        return self.hashcmp == other.hashcmp
+    def __ne__(self, other):
+        return not self == other
+
+    # These properties have to be lazy so that we don't have to load any
+    # metadata until/unless it's actually needed.  (i.e., some distributions
+    # may not know their name or version without loading PKG-INFO)
+
+    #@property
+    def key(self):
+        try:
+            return self._key
+        except AttributeError:
+            self._key = key = self.project_name.lower()
+            return key
+    key = property(key)
+
+    #@property
+    def parsed_version(self):
+        try:
+            return self._parsed_version
+        except AttributeError:
+            self._parsed_version = pv = parse_version(self.version)
+            return pv
+
+    parsed_version = property(parsed_version)
+
+    #@property
+    def version(self):
+        try:
+            return self._version
+        except AttributeError:
+            for line in self._get_metadata('PKG-INFO'):
+                if line.lower().startswith('version:'):
+                    self._version = safe_version(line.split(':',1)[1].strip())
+                    return self._version
+            else:
+                raise ValueError(
+                    "Missing 'Version:' header and/or PKG-INFO file", self
+                )
+    version = property(version)
+
+
+
+
+    #@property
+    def _dep_map(self):
+        try:
+            return self.__dep_map
+        except AttributeError:
+            dm = self.__dep_map = {None: []}
+            for name in 'requires.txt', 'depends.txt':
+                for extra,reqs in split_sections(self._get_metadata(name)):
+                    if extra: extra = safe_extra(extra)
+                    dm.setdefault(extra,[]).extend(parse_requirements(reqs))
+            return dm
+    _dep_map = property(_dep_map)
+
+    def requires(self,extras=()):
+        """List of Requirements needed for this distro if `extras` are used"""
+        dm = self._dep_map
+        deps = []
+        deps.extend(dm.get(None,()))
+        for ext in extras:
+            try:
+                deps.extend(dm[safe_extra(ext)])
+            except KeyError:
+                raise UnknownExtra(
+                    "%s has no such extra feature %r" % (self, ext)
+                )
+        return deps
+
+    def _get_metadata(self,name):
+        if self.has_metadata(name):
+            for line in self.get_metadata_lines(name):
+                yield line
+
+    def activate(self,path=None):
+        """Ensure distribution is importable on `path` (default=sys.path)"""
+        if path is None: path = sys.path
+        self.insert_on(path)
+        if path is sys.path:
+            fixup_namespace_packages(self.location)
+            map(declare_namespace, self._get_metadata('namespace_packages.txt'))
+
+
+    def egg_name(self):
+        """Return what this distribution's standard .egg filename should be"""
+        filename = "%s-%s-py%s" % (
+            to_filename(self.project_name), to_filename(self.version),
+            self.py_version or PY_MAJOR
+        )
+
+        if self.platform:
+            filename += '-'+self.platform
+        return filename
+
+    def __repr__(self):
+        if self.location:
+            return "%s (%s)" % (self,self.location)
+        else:
+            return str(self)
+
+    def __str__(self):
+        try: version = getattr(self,'version',None)
+        except ValueError: version = None
+        version = version or "[unknown version]"
+        return "%s %s" % (self.project_name,version)
+
+    def __getattr__(self,attr):
+        """Delegate all unrecognized public attributes to .metadata provider"""
+        if attr.startswith('_'):
+            raise AttributeError,attr
+        return getattr(self._provider, attr)
+
+    #@classmethod
+    def from_filename(cls,filename,metadata=None, **kw):
+        return cls.from_location(
+            _normalize_cached(filename), os.path.basename(filename), metadata,
+            **kw
+        )
+    from_filename = classmethod(from_filename)
+
+    def as_requirement(self):
+        """Return a ``Requirement`` that matches this distribution exactly"""
+        return Requirement.parse('%s==%s' % (self.project_name, self.version))
+
+    def load_entry_point(self, group, name):
+        """Return the `name` entry point of `group` or raise ImportError"""
+        ep = self.get_entry_info(group,name)
+        if ep is None:
+            raise ImportError("Entry point %r not found" % ((group,name),))
+        return ep.load()
+
+    def get_entry_map(self, group=None):
+        """Return the entry point map for `group`, or the full entry map"""
+        try:
+            ep_map = self._ep_map
+        except AttributeError:
+            ep_map = self._ep_map = EntryPoint.parse_map(
+                self._get_metadata('entry_points.txt'), self
+            )
+        if group is not None:
+            return ep_map.get(group,{})
+        return ep_map
+
+    def get_entry_info(self, group, name):
+        """Return the EntryPoint object for `group`+`name`, or ``None``"""
+        return self.get_entry_map(group).get(name)
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+    def insert_on(self, path, loc = None):
+        """Insert self.location in path before its nearest parent directory"""
+
+        loc = loc or self.location
+
+        if self.project_name == 'setuptools':
+            try:
+                version = self.version
+            except ValueError:
+                version = ''
+            if '0.7' in version:
+                raise ValueError(
+                    "A 0.7-series setuptools cannot be installed "
+                    "with distribute. Found one at %s" % str(self.location))
+
+        if not loc:
+            return
+
+        if path is sys.path:
+            self.check_version_conflict()
+
+        nloc = _normalize_cached(loc)
+        bdir = os.path.dirname(nloc)
+        npath= map(_normalize_cached, path)
+
+        bp = None
+        for p, item in enumerate(npath):
+            if item==nloc:
+                break
+            elif item==bdir and self.precedence==EGG_DIST:
+                # if it's an .egg, give it precedence over its directory
+                path.insert(p, loc)
+                npath.insert(p, nloc)
+                break
+        else:
+            path.append(loc)
+            return
+
+        # p is the spot where we found or inserted loc; now remove duplicates
+        while 1:
+            try:
+                np = npath.index(nloc, p+1)
+            except ValueError:
+                break
+            else:
+                del npath[np], path[np]
+                p = np  # ha!
+
+        return
+
+
+
+    def check_version_conflict(self):
+        if self.key=='distribute':
+            return      # ignore the inevitable setuptools self-conflicts  :(
+
+        nsp = dict.fromkeys(self._get_metadata('namespace_packages.txt'))
+        loc = normalize_path(self.location)
+        for modname in self._get_metadata('top_level.txt'):
+            if (modname not in sys.modules or modname in nsp
+                or modname in _namespace_packages
+            ):
+                continue
+            if modname in ('pkg_resources', 'setuptools', 'site'):
+                continue
+            fn = getattr(sys.modules[modname], '__file__', None)
+            if fn and (normalize_path(fn).startswith(loc) or
+                       fn.startswith(self.location)):
+                continue
+            issue_warning(
+                "Module %s was already imported from %s, but %s is being added"
+                " to sys.path" % (modname, fn, self.location),
+            )
+
+    def has_version(self):
+        try:
+            self.version
+        except ValueError:
+            issue_warning("Unbuilt egg for "+repr(self))
+            return False
+        return True
+
+    def clone(self,**kw):
+        """Copy this distribution, substituting in any changed keyword args"""
+        for attr in (
+            'project_name', 'version', 'py_version', 'platform', 'location',
+            'precedence'
+        ):
+            kw.setdefault(attr, getattr(self,attr,None))
+        kw.setdefault('metadata', self._provider)
+        return self.__class__(**kw)
+
+
+
+
+    #@property
+    def extras(self):
+        return [dep for dep in self._dep_map if dep]
+    extras = property(extras)
+
+
+def issue_warning(*args,**kw):
+    level = 1
+    g = globals()
+    try:
+        # find the first stack frame that is *not* code in
+        # the pkg_resources module, to use for the warning
+        while sys._getframe(level).f_globals is g:
+            level += 1
+    except ValueError:
+        pass
+    from warnings import warn
+    warn(stacklevel = level+1, *args, **kw)
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+def parse_requirements(strs):
+    """Yield ``Requirement`` objects for each specification in `strs`
+
+    `strs` must be an instance of ``basestring``, or a (possibly-nested)
+    iterable thereof.
+    """
+    # create a steppable iterator, so we can handle \-continuations
+    lines = iter(yield_lines(strs))
+
+    def scan_list(ITEM,TERMINATOR,line,p,groups,item_name):
+
+        items = []
+
+        while not TERMINATOR(line,p):
+            if CONTINUE(line,p):
+                try:
+                    line = lines.next(); p = 0
+                except StopIteration:
+                    raise ValueError(
+                        "\\ must not appear on the last nonblank line"
+                    )
+
+            match = ITEM(line,p)
+            if not match:
+                raise ValueError("Expected "+item_name+" in",line,"at",line[p:])
+
+            items.append(match.group(*groups))
+            p = match.end()
+
+            match = COMMA(line,p)
+            if match:
+                p = match.end() # skip the comma
+            elif not TERMINATOR(line,p):
+                raise ValueError(
+                    "Expected ',' or end-of-list in",line,"at",line[p:]
+                )
+
+        match = TERMINATOR(line,p)
+        if match: p = match.end()   # skip the terminator, if any
+        return line, p, items
+
+    for line in lines:
+        match = DISTRO(line)
+        if not match:
+            raise ValueError("Missing distribution spec", line)
+        project_name = match.group(1)
+        p = match.end()
+        extras = []
+
+        match = OBRACKET(line,p)
+        if match:
+            p = match.end()
+            line, p, extras = scan_list(
+                DISTRO, CBRACKET, line, p, (1,), "'extra' name"
+            )
+
+        line, p, specs = scan_list(VERSION,LINE_END,line,p,(1,2),"version spec")
+        specs = [(op,safe_version(val)) for op,val in specs]
+        yield Requirement(project_name, specs, extras)
+
+
+def _sort_dists(dists):
+    tmp = [(dist.hashcmp,dist) for dist in dists]
+    tmp.sort()
+    dists[::-1] = [d for hc,d in tmp]
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+class Requirement:
+    def __init__(self, project_name, specs, extras):
+        """DO NOT CALL THIS UNDOCUMENTED METHOD; use Requirement.parse()!"""
+        self.unsafe_name, project_name = project_name, safe_name(project_name)
+        self.project_name, self.key = project_name, project_name.lower()
+        index = [(parse_version(v),state_machine[op],op,v) for op,v in specs]
+        index.sort()
+        self.specs = [(op,ver) for parsed,trans,op,ver in index]
+        self.index, self.extras = index, tuple(map(safe_extra,extras))
+        self.hashCmp = (
+            self.key, tuple([(op,parsed) for parsed,trans,op,ver in index]),
+            frozenset(self.extras)
+        )
+        self.__hash = hash(self.hashCmp)
+
+    def __str__(self):
+        specs = ','.join([''.join(s) for s in self.specs])
+        extras = ','.join(self.extras)
+        if extras: extras = '[%s]' % extras
+        return '%s%s%s' % (self.project_name, extras, specs)
+
+    def __eq__(self,other):
+        return isinstance(other,Requirement) and self.hashCmp==other.hashCmp
+
+    def __contains__(self,item):
+        if isinstance(item,Distribution):
+            if item.key <> self.key: return False
+            if self.index: item = item.parsed_version  # only get if we need it
+        elif isinstance(item,basestring):
+            item = parse_version(item)
+        last = None
+        compare = lambda a, b: (a > b) - (a < b) # -1, 0, 1
+        for parsed,trans,op,ver in self.index:
+            action = trans[compare(item,parsed)] # Indexing: 0, 1, -1
+            if action=='F':     return False
+            elif action=='T':   return True
+            elif action=='+':   last = True
+            elif action=='-' or last is None:   last = False
+        if last is None: last = True    # no rules encountered
+        return last
+
+
+    def __hash__(self):
+        return self.__hash
+
+    def __repr__(self): return "Requirement.parse(%r)" % str(self)
+
+    #@staticmethod
+    def parse(s, replacement=True):
+        reqs = list(parse_requirements(s))
+        if reqs:
+            if len(reqs) == 1:
+                founded_req = reqs[0]
+                # if asked for setuptools distribution
+                # and if distribute is installed, we want to give
+                # distribute instead
+                if _override_setuptools(founded_req) and replacement:
+                    distribute = list(parse_requirements('distribute'))
+                    if len(distribute) == 1:
+                        return distribute[0]
+                    return founded_req
+                else:
+                    return founded_req
+
+            raise ValueError("Expected only one requirement", s)
+        raise ValueError("No requirements found", s)
+
+    parse = staticmethod(parse)
+
+state_machine = {
+    #       =><
+    '<' :  '--T',
+    '<=':  'T-T',
+    '>' :  'F+F',
+    '>=':  'T+F',
+    '==':  'T..',
+    '!=':  'F++',
+}
+
+
+def _override_setuptools(req):
+    """Return True when distribute wants to override a setuptools dependency.
+
+    We want to override when the requirement is setuptools and the version is
+    a variant of 0.6.
+
+    """
+    if req.project_name == 'setuptools':
+        if not len(req.specs):
+            # Just setuptools: ok
+            return True
+        for comparator, version in req.specs:
+            if comparator in ['==', '>=', '>']:
+                if '0.7' in version:
+                    # We want some setuptools not from the 0.6 series.
+                    return False
+        return True
+    return False
+
+
+def _get_mro(cls):
+    """Get an mro for a type or classic class"""
+    if not isinstance(cls,type):
+        class cls(cls,object): pass
+        return cls.__mro__[1:]
+    return cls.__mro__
+
+def _find_adapter(registry, ob):
+    """Return an adapter factory for `ob` from `registry`"""
+    for t in _get_mro(getattr(ob, '__class__', type(ob))):
+        if t in registry:
+            return registry[t]
+
+
+def ensure_directory(path):
+    """Ensure that the parent directory of `path` exists"""
+    dirname = os.path.dirname(path)
+    if not os.path.isdir(dirname):
+        os.makedirs(dirname)
+
+def split_sections(s):
+    """Split a string or iterable thereof into (section,content) pairs
+
+    Each ``section`` is a stripped version of the section header ("[section]")
+    and each ``content`` is a list of stripped lines excluding blank lines and
+    comment-only lines.  If there are any such lines before the first section
+    header, they're returned in a first ``section`` of ``None``.
+    """
+    section = None
+    content = []
+    for line in yield_lines(s):
+        if line.startswith("["):
+            if line.endswith("]"):
+                if section or content:
+                    yield section, content
+                section = line[1:-1].strip()
+                content = []
+            else:
+                raise ValueError("Invalid section heading", line)
+        else:
+            content.append(line)
+
+    # wrap up last segment
+    yield section, content
+
+def _mkstemp(*args,**kw):
+    from tempfile import mkstemp
+    old_open = os.open
+    try:
+        os.open = os_open   # temporarily bypass sandboxing
+        return mkstemp(*args,**kw)
+    finally:
+        os.open = old_open  # and then put it back
+
+
+# Set up global resource manager
+_manager = ResourceManager()
+def _initialize(g):
+    for name in dir(_manager):
+        if not name.startswith('_'):
+            g[name] = getattr(_manager, name)
+_initialize(globals())
+
+# Prepare the master working set and make the ``require()`` API available
+working_set = WorkingSet()
+try:
+    # Does the main program list any requirements?
+    from __main__ import __requires__
+except ImportError:
+    pass # No: just use the default working set based on sys.path
+else:
+    # Yes: ensure the requirements are met, by prefixing sys.path if necessary
+    try:
+        working_set.require(__requires__)
+    except VersionConflict:     # try it without defaults already on sys.path
+        working_set = WorkingSet([])    # by starting with an empty path
+        for dist in working_set.resolve(
+            parse_requirements(__requires__), Environment()
+        ):
+            working_set.add(dist)
+        for entry in sys.path:  # add any missing entries from sys.path
+            if entry not in working_set.entries:
+                working_set.add_entry(entry)
+        sys.path[:] = working_set.entries   # then copy back to sys.path
+
+require = working_set.require
+iter_entry_points = working_set.iter_entry_points
+add_activation_listener = working_set.subscribe
+run_script = working_set.run_script
+run_main = run_script   # backward compatibility
+# Activate all distributions already on sys.path, and ensure that
+# all distributions added to the working set in the future (e.g. by
+# calling ``require()``) will get activated as well.
+add_activation_listener(lambda dist: dist.activate())
+working_set.entries=[]; map(working_set.add_entry,sys.path) # match order
+
diff --git a/src/distutils2/pypi/package_index.py b/src/distutils2/pypi/simple.py
rename from src/distutils2/pypi/package_index.py
rename to src/distutils2/pypi/simple.py
--- a/src/distutils2/pypi/package_index.py
+++ b/src/distutils2/pypi/simple.py
@@ -1,15 +1,43 @@
-"""PyPI and direct package downloading"""
-import sys, os.path, re, urlparse, urllib2, shutil, random, socket, cStringIO
+"""distutils2.pypi.package_index
+
+Implement a simple way to use the PyPI API's, by using known URIs from 
+PyPI. Works with the API defined on 
+http://peak.telecommunity.com/DevCenter/EasyInstall#package-index-api , 
+and available at http://pypi.python.org/simple/ .
+
+"""
+# XXX Make a explaination document on how this API is made.
+
+import sys
+import os.path
+import re
+import urlparse
+import urllib2
+import shutil
+import socket
+import cStringIO
 import httplib
-from pkg_resources import *
-from distutils import log
-from distutils.errors import DistutilsError
+import logging
+
+from distutils2.errors import DistutilsError
+from distutils2 import __version__ as __distutils2_version__
+
 try:
     from hashlib import md5
 except ImportError:
     from md5 import md5
 from fnmatch import translate
 
+# -- Constants ----------------------------------------------------------------
+
+__all__ = ['PackageIndex', 
+           'distros_for_url', 
+           'parse_bdist_wininst',
+           'interpret_distro_name',
+]
+
+_SOCKET_TIMEOUT = 15
+SOURCE_DIST = 1
 EGG_FRAGMENT = re.compile(r'^egg=([-A-Za-z0-9_.]+)$')
 HREF = re.compile("""href\\s*=\\s*['"]?([^'"> ]+)""", re.I)
 # this is here to fix emacs' cruddy broken syntax highlighting
@@ -20,15 +48,19 @@
 URL_SCHEME = re.compile('([-+.a-z0-9]{2,}):',re.I).match
 EXTENSIONS = ".tar.gz .tar.bz2 .tar .zip .tgz".split()
 
-__all__ = [
-    'PackageIndex', 'distros_for_url', 'parse_bdist_wininst',
-    'interpret_distro_name',
-]
+# -- Dependencies from distribute ---------------------------------------------
 
-_SOCKET_TIMEOUT = 15
+# XXX Remove these dependencies, refactor the package_index codebase to not 
+# depends anymore on distribute classes and internal workings.
+from pkg_resources import *
+
+# -- Base functions -----------------------------------------------------------
 
 def parse_bdist_wininst(name):
-    """Return (base,pyversion) or (None,None) for possible .exe name"""
+    """Parse a bdist_win package name, and return it's package base name 
+    and version number if prossible.
+
+    Return a tuple of (base,pyversion) for possible .exe names"""
 
     lower = name.lower()
     base, py_ver = None, None
@@ -43,21 +75,24 @@
     return base,py_ver
 
 def egg_info_for_url(url):
+    """
+    """
     scheme, server, path, parameters, query, fragment = urlparse.urlparse(url)
     base = urllib2.unquote(path.split('/')[-1])
-    if '#' in base: base, fragment = base.split('#',1)
+    if '#' in base:
+        base, fragment = base.split('#',1)
     return base,fragment
 
 def distros_for_url(url, metadata=None):
     """Yield egg or source distribution objects that might be found at a URL"""
     base, fragment = egg_info_for_url(url)
-    for dist in distros_for_location(url, base, metadata): yield dist
+    for dist in distros_for_location(url, base, metadata): 
+        yield dist
     if fragment:
         match = EGG_FRAGMENT.match(fragment)
         if match:
-            for dist in interpret_distro_name(
-                url, match.group(1), metadata, precedence = CHECKOUT_DIST
-            ):
+            for dist in interpret_distro_name(url, match.group(1), metadata, 
+                precedence = CHECKOUT_DIST):
                 yield dist
 
 def distros_for_location(location, basename, metadata=None):
@@ -71,12 +106,10 @@
     if basename.endswith('.exe'):
         win_base, py_ver = parse_bdist_wininst(basename)
         if win_base is not None:
-            return interpret_distro_name(
-                location, win_base, metadata, py_ver, BINARY_DIST, "win32"
-            )
+            return interpret_distro_name(location, win_base, metadata, py_ver, 
+                BINARY_DIST, "win32")
 
     # Try source distro extensions (.zip, .tgz, etc.)
-    #
     for ext in EXTENSIONS:
         if basename.endswith(ext):
             basename = basename[:-len(ext)]
@@ -84,15 +117,14 @@
     return []  # no extension matched
 
 def distros_for_filename(filename, metadata=None):
-    """Yield possible egg or source distribution objects based on a filename"""
-    return distros_for_location(
-        normalize_path(filename), os.path.basename(filename), metadata
-    )
-
+    """Yield possible egg or source distribution objects based on a 
+    filename.
+    """
+    return distros_for_location(normalize_path(filename), 
+        os.path.basename(filename), metadata)
 
 def interpret_distro_name(location, basename, metadata,
-    py_version=None, precedence=SOURCE_DIST, platform=None
-):
+    py_version=None, precedence=SOURCE_DIST, platform=None):
     """Generate alternative interpretations of a source distro name
 
     Note: if `location` is a filesystem filename, you should call
@@ -121,11 +153,9 @@
         yield Distribution(
             location, metadata, '-'.join(parts[:p]), '-'.join(parts[p:]),
             py_version=py_version, precedence = precedence,
-            platform = platform
-        )
+            platform = platform)
 
 REL = re.compile("""<([^>]*\srel\s*=\s*['"]?([^'">]+)[^>]*)>""", re.I)
-# this line is here to fix emacs' cruddy broken syntax highlighting
 
 def find_external_links(url, page):
     """Find rel="homepage" and rel="download" links in `page`, yielding URLs"""
@@ -144,17 +174,16 @@
             if match:
                 yield urlparse.urljoin(url, htmldecode(match.group(1)))
 
-user_agent = "Python-urllib/%s distribute/%s" % (
-    sys.version[:3], require('distribute')[0].version
+user_agent = "Python-urllib/%s distutils2/%s" % (
+    sys.version[:3], __distutils2_version__
 )
 
-
 class PackageIndex(Environment):
     """A distribution index that scans web pages for download URLs"""
 
     def __init__(self, index_url="http://pypi.python.org/simple", hosts=('*',),
-        *args, **kw
-    ):
+        *args, **kw):
+        self._init_logging()
         Environment.__init__(self,*args,**kw)
         self.index_url = index_url + "/"[:not index_url.endswith('/')]
         self.scanned_urls = {}
@@ -163,7 +192,9 @@
         self.allows = re.compile('|'.join(map(translate,hosts))).match
         self.to_scan = []
 
-
+    def _init_logging(self):
+        import sys
+#        logging.basicConfig(stream=sys.stdout, level=logging.DEBUG)
 
     def process_url(self, url, retrieve=False):
         """Evaluate a URL as a possible download, and maybe retrieve it"""
@@ -207,6 +238,8 @@
             link = urlparse.urljoin(base, htmldecode(match.group(1)))
             self.process_url(link)
         if url.startswith(self.index_url) and getattr(f,'code',None)!=404:
+            from ipdb import set_trace
+            set_trace()
             page = self.process_index(url, page)
 
     def process_filename(self, fn, nested=False):
@@ -234,6 +267,7 @@
             raise DistutilsError(msg % url)
         else:
             self.warn(msg, url)
+        return False
 
     def scan_egg_links(self, search_path):
         for item in search_path:
@@ -292,8 +326,6 @@
         else:
             return ""   # no sense double-scanning non-package pages
 
-
-
     def need_version_info(self, url):
         self.scan_all(
             "Page at %s links to .py file(s) without version info; an index "
@@ -331,10 +363,6 @@
             self.debug("%s does not match %s", requirement, dist)
         return super(PackageIndex, self).obtain(requirement,installer)
 
-
-
-
-
     def check_md5(self, cs, info, filename, tfp):
         if re.match('md5=[0-9a-f]{32}$', info):
             self.debug("Validating md5 checksum for %s", filename)
@@ -343,8 +371,7 @@
                 os.unlink(filename)
                 raise DistutilsError(
                     "MD5 validation failed for "+os.path.basename(filename)+
-                    "; possible download problem?"
-                )
+                    "; possible download problem?")
 
     def add_find_links(self, urls):
         """Add `urls` to the list that will be prescanned for searches"""
@@ -416,7 +443,6 @@
                     )
         return getattr(self.fetch_distribution(spec, tmpdir),'location',None)
 
-
     def fetch_distribution(self,
         requirement, tmpdir, force_scan=False, source=False, develop_ok=False
     ):
@@ -436,7 +462,7 @@
         set, development and system eggs (i.e., those using the ``.egg-info``
         format) will be ignored.
         """
-
+        
         # process a Requirement
         self.info("Searching for %s", requirement)
         skipped = {}
@@ -491,14 +517,7 @@
         if dist is not None:
             return dist.location
         return None
-
-
-
-
-
-
-
-
+    
     def gen_setup(self, filename, fragment, tmpdir):
         match = EGG_FRAGMENT.match(fragment)
         dists = match and [d for d in
@@ -536,7 +555,7 @@
             )
         else:
             raise DistutilsError(
-                "Can't process plain .py files without an '#egg=name-version'"
+                "Can't process plain .py without an '#egg=name-version'"
                 " suffix to enable automatic setup script generation."
             )
 
@@ -580,7 +599,6 @@
     def reporthook(self, url, filename, blocknum, blksize, size):
         pass    # no-op
 
-
     def open_url(self, url, warning=None):
         if url.startswith('file:'):
             return local_open(url)
@@ -616,7 +634,6 @@
 
     def _download_url(self, scheme, url, tmpdir):
         # Determine download filename
-        #
         name = filter(None,urlparse.urlparse(url)[2].split('/'))
         if name:
             name = name[-1]
@@ -631,7 +648,6 @@
         filename = os.path.join(tmpdir,name)
 
         # Download the file
-        #
         if scheme=='svn' or scheme.startswith('svn+'):
             return self._download_svn(url, filename)
         elif scheme=='file':
@@ -640,12 +656,9 @@
             self.url_ok(url, True)   # raises error if not allowed
             return self._attempt_download(url, filename)
 
-
-
     def scan_url(self, url):
         self.process_url(url, True)
 
-
     def _attempt_download(self, url, filename):
         headers = self._download_to(url, filename)
         if 'html' in headers.get('content-type','').lower():
@@ -675,13 +688,13 @@
         return filename
 
     def debug(self, msg, *args):
-        log.debug(msg, *args)
+        logging.debug(msg, *args)
 
     def info(self, msg, *args):
-        log.info(msg, *args)
+        logging.info(msg, *args)
 
     def warn(self, msg, *args):
-        log.warn(msg, *args)
+        logging.warn(msg, *args)
 
 # This pattern matches a character entity reference (a decimal numeric
 # references, a hexadecimal numeric reference, or a named reference).
@@ -690,7 +703,8 @@
 def uchr(c):
     if not isinstance(c, int):
         return c
-    if c>255: return unichr(c)
+    if c>255: 
+        return unichr(c)
     return chr(c)
 
 def decode_entity(match):
@@ -708,20 +722,6 @@
     """Decode HTML entities in the given text."""
     return entity_sub(decode_entity, text)
 
-
-
-
-
-
-
-
-
-
-
-
-
-
-
 def socket_timeout(timeout=15):
     def _socket_timeout(func):
         def _socket_timeout(*args, **kwargs):
@@ -734,7 +734,6 @@
         return _socket_timeout
     return _socket_timeout
 
-
 def open_with_auth(url):
     """Open a urllib2 request, handling HTTP authentication"""
 
@@ -765,19 +764,10 @@
 
     return fp
 
+
 # adding a timeout to avoid freezing package_index
 open_with_auth = socket_timeout(_SOCKET_TIMEOUT)(open_with_auth)
 
-
-
-
-
-
-
-
-
-
-
 def fix_sf_url(url):
     return url      # backward compatibility
 
@@ -807,17 +797,3 @@
 
     return urllib2.HTTPError(url, status, message,
             {'content-type':'text/html'}, cStringIO.StringIO(body))
-
-
-
-
-
-
-
-
-
-
-
-
-
-# this line is a kludge to keep the trailing blank lines for pje's editor
diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -0,0 +1,121 @@
+"""Tests for pypi.simple module.
+
+"""
+# More would be better!
+import sys
+import os
+import shutil 
+import tempfile 
+import unittest2
+import urllib2
+
+from distutils2.tests.pypi_server import PyPIServer
+from distutils2.pypi import simple
+
+class NoneTestCase(unittest2.TestCase):
+    pass
+
+class PackageIndexTestCase(unittest2.TestCase):
+
+    def test_bad_urls(self):
+        index = simple.PackageIndex()
+        url = 'http://127.0.0.1:0/nonesuch/test_simple'
+        try:
+            v = index.open_url(url)
+        except Exception, v:
+            self.assert_(url in str(v))
+        else:
+            self.assert_(isinstance(v,urllib2.HTTPError))
+
+        # issue 16
+        # easy_install inquant.contentmirror.plone breaks because of a typo
+        # in its home URL
+        index = simple.PackageIndex(
+            hosts=('www.example.com',)
+        )
+
+        url = 'url:%20https://svn.plone.org/svn/collective/inquant.contentmirror.plone/trunk'
+        try:
+            v = index.open_url(url)
+        except Exception, v:
+            self.assert_(url in str(v))
+        else:
+            self.assert_(isinstance(v, urllib2.HTTPError))
+
+        def _urlopen(*args):
+            import httplib
+            raise httplib.BadStatusLine('line')
+
+        old_urlopen = urllib2.urlopen
+        urllib2.urlopen = _urlopen
+        url = 'http://example.com'
+        try:
+            try:
+                v = index.open_url(url)
+            except Exception, v:
+                self.assert_('line' in str(v))
+            else:
+                raise AssertionError('Should have raise here!')
+        finally:
+            urllib2.urlopen = old_urlopen
+
+        # issue 20
+        url = 'http://http://svn.pythonpaste.org/Paste/wphp/trunk'
+        try:
+            index.open_url(url)
+        except Exception, v:
+            self.assert_('nonnumeric port' in str(v))
+
+        # issue #160
+        if sys.version_info[0] == 2 and sys.version_info[1] == 7:
+            # this should not fail
+            url = 'http://example.com'
+            page = ('<a href="http://www.famfamfam.com]('
+                    'http://www.famfamfam.com/">')
+            index.process_index(url, page)
+
+    def test_url_ok(self):
+        index = simple.PackageIndex(
+            hosts=('www.example.com',)
+        )
+        url = 'file:///tmp/test_simple'
+        self.assert_(index.url_ok(url, True))
+
+    def test_links_priority(self):
+        """
+        Download links from the pypi simple index should be used before
+        external download links.
+        http://bitbucket.org/tarek/distribute/issue/163/md5-validation-error
+
+        Usecase :
+        - someone uploads a package on pypi, a md5 is generated
+        - someone manually copies this link (with the md5 in the url) onto an
+          external page accessible from the package page.
+        - someone reuploads the package (with a different md5)
+        - while easy_installing, an MD5 error occurs because the external link
+          is used
+        -> Distribute should use the link from pypi, not the external one.
+        """
+        # start an index server
+        server = PyPIServer()
+        server.start()
+        index_url = server.full_address + 'test_links_priority/simple/'
+
+        # scan a test index
+        pi = simple.PackageIndex(index_url)
+        requirement = simple.Requirement.parse('foobar')
+        pi.find_packages(requirement)
+        server.stop()
+
+        # the distribution has been found
+        self.assert_('foobar' in pi)
+        # we have only one link, because links are compared without md5
+        self.assert_(len(pi['foobar'])==1)
+        # the link should be from the index
+        self.assert_('correct_md5' in pi['foobar'][0].location)
+
+def test_suite():
+    return unittest2.makeSuite(NoneTestCase)
+
+if __name__ == '__main__':
+    unittest2.main(defaultTest="test_suite")

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: merged tarek's changes
Message-ID: <E1OVLoV-0000hU-4n@dinsdale.python.org>

tarek.ziade pushed ec619c341cd2 to distutils2:

http://hg.python.org/distutils2/rev/ec619c341cd2
changeset:   279:ec619c341cd2
parent:      278:03bc7496c3b2
parent:      188:b7fd258b33e7
user:        Konrad Delong <konryd at gmail.com>
date:        Tue Jun 01 11:23:39 2010 -0400
summary:     merged tarek's changes
files:       src/distutils2/filelist.py, src/distutils2/tests/test_filelist.py, src/distutils2/tests/test_text_file.py, src/distutils2/text_file.py

diff --git a/.hgignore b/.hgignore
--- a/.hgignore
+++ b/.hgignore
@@ -1,3 +1,5 @@
 .*\.pyc$
 .*\.pyo$
 ^docs/build
+.*\.swp$
+
diff --git a/src/CONTRIBUTORS.txt b/src/CONTRIBUTORS.txt
new file mode 100644
--- /dev/null
+++ b/src/CONTRIBUTORS.txt
@@ -0,0 +1,25 @@
+============
+Contributors
+============
+
+Distutils2 is a project that was started and that is maintained by
+Tarek Ziad??, and many people are contributing to the project.
+
+If you did, please add your name below in alphabetical order !
+
+Thanks to:
+
+- Pior Bastida
+- Titus Brown
+- Nicolas Cadou
+- Josip Djolonga
+- Yannick Gringas
+- Carl Meyer
+- Michael Mulich
+- George Peris
+- Sean Reifschneider
+- Erik Rose
+- Brian Rosner
+- Alexandre Vassalotti
+- Martin von L??wis
+
diff --git a/src/DEVNOTES.txt b/src/DEVNOTES.txt
new file mode 100644
--- /dev/null
+++ b/src/DEVNOTES.txt
@@ -0,0 +1,10 @@
+Notes for developers
+====================
+
+- Distutils2 runs from 2.4 to 3.2 (3.x not implemented yet), so
+  make sure you don't use a syntax that doesn't work under
+  a specific Python version.
+
+- Always run tests.sh before you push a change. This implies
+  that you have all Python versions installed.
+
diff --git a/src/distutils2/_backport/pkgutil.py b/src/distutils2/_backport/pkgutil.py
--- a/src/distutils2/_backport/pkgutil.py
+++ b/src/distutils2/_backport/pkgutil.py
@@ -11,16 +11,18 @@
 from types import ModuleType
 from distutils2.errors import DistutilsError
 from distutils2.metadata import DistributionMetadata
-from distutils2.version import suggest_normalized_version
+from distutils2.version import suggest_normalized_version, VersionPredicate
 
 __all__ = [
     'get_importer', 'iter_importers', 'get_loader', 'find_loader',
     'walk_packages', 'iter_modules',
     'ImpImporter', 'ImpLoader', 'read_code', 'extend_path',
-    'Distribution', 'distinfo_dirname', 'get_distributions',
-    'get_distribution', 'get_file_users', 
+    'Distribution', 'EggInfoDistribution', 'distinfo_dirname',
+    'get_distributions', 'get_distribution', 'get_file_users',
+    'provides_distribution', 'obsoletes_distribution',
 ]
 
+
 def read_code(stream):
     # This helper is needed in order for the PEP 302 emulation to
     # correctly handle compiled files
@@ -37,6 +39,7 @@
 def simplegeneric(func):
     """Make a trivial single-dispatch generic function"""
     registry = {}
+
     def wrapper(*args, **kw):
         ob = args[0]
         try:
@@ -47,6 +50,7 @@
             mro = cls.__mro__
         except AttributeError:
             try:
+
                 class cls(cls, object):
                     pass
                 mro = cls.__mro__[1:]
@@ -128,7 +132,7 @@
                 # don't traverse path items we've seen before
                 path = [p for p in path if not seen(p)]
 
-                for item in walk_packages(path, name+'.', onerror):
+                for item in walk_packages(path, name + '.', onerror):
                     yield item
 
 
@@ -206,7 +210,7 @@
 
         for fn in filenames:
             modname = inspect.getmodulename(fn)
-            if modname=='__init__' or modname in yielded:
+            if modname == '__init__' or modname in yielded:
                 continue
 
             path = os.path.join(self.path, fn)
@@ -216,7 +220,7 @@
                 modname = fn
                 for fn in os.listdir(path):
                     subname = inspect.getmodulename(fn)
-                    if subname=='__init__':
+                    if subname == '__init__':
                         ispkg = True
                         break
                 else:
@@ -255,7 +259,7 @@
     def _reopen(self):
         if self.file and self.file.closed:
             mod_type = self.etc[2]
-            if mod_type==imp.PY_SOURCE:
+            if mod_type == imp.PY_SOURCE:
                 self.file = open(self.filename, 'rU')
             elif mod_type in (imp.PY_COMPILED, imp.C_EXTENSION):
                 self.file = open(self.filename, 'rb')
@@ -270,22 +274,22 @@
 
     def is_package(self, fullname):
         fullname = self._fix_name(fullname)
-        return self.etc[2]==imp.PKG_DIRECTORY
+        return self.etc[2] == imp.PKG_DIRECTORY
 
     def get_code(self, fullname=None):
         fullname = self._fix_name(fullname)
         if self.code is None:
             mod_type = self.etc[2]
-            if mod_type==imp.PY_SOURCE:
+            if mod_type == imp.PY_SOURCE:
                 source = self.get_source(fullname)
                 self.code = compile(source, self.filename, 'exec')
-            elif mod_type==imp.PY_COMPILED:
+            elif mod_type == imp.PY_COMPILED:
                 self._reopen()
                 try:
                     self.code = read_code(self.file)
                 finally:
                     self.file.close()
-            elif mod_type==imp.PKG_DIRECTORY:
+            elif mod_type == imp.PKG_DIRECTORY:
                 self.code = self._get_delegate().get_code()
         return self.code
 
@@ -293,29 +297,28 @@
         fullname = self._fix_name(fullname)
         if self.source is None:
             mod_type = self.etc[2]
-            if mod_type==imp.PY_SOURCE:
+            if mod_type == imp.PY_SOURCE:
                 self._reopen()
                 try:
                     self.source = self.file.read()
                 finally:
                     self.file.close()
-            elif mod_type==imp.PY_COMPILED:
+            elif mod_type == imp.PY_COMPILED:
                 if os.path.exists(self.filename[:-1]):
                     f = open(self.filename[:-1], 'rU')
                     self.source = f.read()
                     f.close()
-            elif mod_type==imp.PKG_DIRECTORY:
+            elif mod_type == imp.PKG_DIRECTORY:
                 self.source = self._get_delegate().get_source()
         return self.source
 
-
     def _get_delegate(self):
         return ImpImporter(self.filename).find_module('__init__')
 
     def get_filename(self, fullname=None):
         fullname = self._fix_name(fullname)
         mod_type = self.etc[2]
-        if self.etc[2]==imp.PKG_DIRECTORY:
+        if self.etc[2] == imp.PKG_DIRECTORY:
             return self._get_delegate().get_filename()
         elif self.etc[2] in (imp.PY_SOURCE, imp.PY_COMPILED, imp.C_EXTENSION):
             return self.filename
@@ -339,16 +342,16 @@
 
             fn = fn[plen:].split(os.sep)
 
-            if len(fn)==2 and fn[1].startswith('__init__.py'):
+            if len(fn) == 2 and fn[1].startswith('__init__.py'):
                 if fn[0] not in yielded:
                     yielded[fn[0]] = 1
                     yield fn[0], True
 
-            if len(fn)!=1:
+            if len(fn) != 1:
                 continue
 
             modname = inspect.getmodulename(fn[0])
-            if modname=='__init__':
+            if modname == '__init__':
                 continue
 
             if modname and '.' not in modname and modname not in yielded:
@@ -436,6 +439,7 @@
     if '.' not in fullname:
         yield ImpImporter()
 
+
 def get_loader(module_or_name):
     """Get a PEP 302 "loader" object for module_or_name
 
@@ -461,6 +465,7 @@
         fullname = module_or_name
     return find_loader(fullname)
 
+
 def find_loader(fullname):
     """Find a PEP 302 "loader" object for fullname
 
@@ -551,6 +556,7 @@
 
     return path
 
+
 def get_data(package, resource):
     """Get a resource from a package.
 
@@ -594,6 +600,7 @@
 
 DIST_FILES = ('INSTALLER', 'METADATA', 'RECORD', 'REQUESTED',)
 
+
 class Distribution(object):
     """Created with the *path* of the ``.dist-info`` directory provided to the
     constructor. It reads the metadata contained in METADATA when it is
@@ -604,7 +611,7 @@
     name = ''
     """The name of the distribution."""
     metadata = None
-    """A :class:`distutils2.metadata.DistributionMetadata` instance loaded with 
+    """A :class:`distutils2.metadata.DistributionMetadata` instance loaded with
     the distribution's METADATA file."""
     requested = False
     """A boolean that indicates whether the REQUESTED metadata file is present
@@ -620,7 +627,7 @@
         RECORD = os.path.join(self.path, 'RECORD')
         record_reader = csv_reader(open(RECORD, 'rb'), delimiter=',')
         for row in record_reader:
-            path, md5, size = row[:] + [ None for i in xrange(len(row), 3) ]
+            path, md5, size = row[:] + [None for i in xrange(len(row), 3)]
             if local:
                 path = path.replace('/', os.sep)
                 path = os.path.join(sys.prefix, path)
@@ -635,7 +642,6 @@
 
         A local absolute path is an absolute path in which occurrences of
         ``'/'`` have been replaced by the system separator given by ``os.sep``.
-
         :parameter local: flag to say if the path should be returned a local
                           absolute path
         :type local: boolean
@@ -643,10 +649,9 @@
         """
         return self._get_records(local)
 
-
     def uses(self, path):
         """
-        Returns ``True`` if path is listed in RECORD. *path* can be a local 
+        Returns ``True`` if path is listed in RECORD. *path* can be a local
         absolute path or a relative ``'/'``-separated path.
 
         :rtype: boolean
@@ -663,8 +668,8 @@
         ``file`` instance for the file pointed by *path*.
 
         :parameter path: a ``'/'``-separated path relative to the ``.dist-info``
-                         directory or an absolute path; If *path* is an absolute 
-                         path and doesn't start with the ``.dist-info``
+                         directory or an absolute path; If *path* is an
+                         absolute path and doesn't start with the ``.dist-info``
                          directory path, a :class:`DistutilsError` is raised
         :type path: string
         :parameter binary: If *binary* is ``True``, opens the file in read-only
@@ -696,8 +701,8 @@
 
     def get_distinfo_files(self, local=False):
         """
-        Iterates over the RECORD entries and returns paths for each line if the 
-        path is pointing to a file located in the ``.dist-info`` directory or 
+        Iterates over the RECORD entries and returns paths for each line if the
+        path is pointing to a file located in the ``.dist-info`` directory or
         one of its subdirectories.
 
         :parameter local: If *local* is ``True``, each returned path is
@@ -710,16 +715,42 @@
             yield path
 
 
+class EggInfoDistribution(object):
+    """Created with the *path* of the ``.egg-info`` directory or file provided
+    to the constructor. It reads the metadata contained in the file itself, or
+    if the given path happens to be a directory, the metadata is read from the
+    file PKG-INFO under that directory."""
+
+    name = ''
+    """The name of the distribution."""
+    metadata = None
+    """A :class:`distutils2.metadata.DistributionMetadata` instance loaded with
+    the distribution's METADATA file."""
+
+    def __init__(self, path):
+        if os.path.isdir(path):
+            path = os.path.join(path, 'PKG-INFO')
+        self.metadata = DistributionMetadata(path=path)
+        self.name = self.metadata['name']
+
+    def get_installed_files(self, local=False):
+        return []
+
+    def uses(self, path):
+        return False
+
+
 def _normalize_dist_name(name):
     """Returns a normalized name from the given *name*.
     :rtype: string"""
     return name.replace('-', '_')
 
+
 def distinfo_dirname(name, version):
     """
     The *name* and *version* parameters are converted into their
     filename-escaped form, i.e. any ``'-'`` characters are replaced with ``'_'``
-    other than the one in ``'dist-info'`` and the one separating the name from 
+    other than the one in ``'dist-info'`` and the one separating the name from
     the version number.
 
     :parameter name: is converted to a standard distribution name by replacing
@@ -743,13 +774,16 @@
         normalized_version = version
     return '-'.join([name, normalized_version]) + file_extension
 
-def get_distributions():
+
+def get_distributions(use_egg_info=False):
     """
     Provides an iterator that looks for ``.dist-info`` directories in
     ``sys.path`` and returns :class:`Distribution` instances for each one of
-    them.
+    them. If the parameters *use_egg_info* is ``True``, then the ``.egg-info``
+    files and directores are iterated as well.
 
-    :rtype: iterator of :class:`Distribution` instances"""
+    :rtype: iterator of :class:`Distribution` and :class:`EggInfoDistribution`
+            instances"""
     for path in sys.path:
         realpath = os.path.realpath(path)
         if not os.path.isdir(realpath):
@@ -758,25 +792,117 @@
             if dir.endswith('.dist-info'):
                 dist = Distribution(os.path.join(realpath, dir))
                 yield dist
+            elif use_egg_info and dir.endswith('.egg-info'):
+                dist = EggInfoDistribution(os.path.join(realpath, dir))
+                yield dist
 
-def get_distribution(name):
+
+def get_distribution(name, use_egg_info=False):
     """
     Scans all elements in ``sys.path`` and looks for all directories ending with
-    ``.dist-info``. Returns a :class:`Distribution` corresponding to the 
+    ``.dist-info``. Returns a :class:`Distribution` corresponding to the
     ``.dist-info`` directory that contains the METADATA that matches *name* for
-    the *name* metadata.
+    the *name* metadata field.
+    If no distribution exists with the given *name* and the parameter
+    *use_egg_info* is set to ``True``, then all files and directories ending
+    with ``.egg-info`` are scanned. A :class:`EggInfoDistribution` instance is
+    returned if one is found that has metadata that matches *name* for the
+    *name* metadata field.
 
     This function only returns the first result founded, as no more than one
     value is expected. If the directory is not found, ``None`` is returned.
 
-    :rtype: :class:`Distribution` or None"""
+    :rtype: :class:`Distribution` or :class:`EggInfoDistribution: or None"""
     found = None
     for dist in get_distributions():
         if dist.name == name:
             found = dist
             break
+    if use_egg_info:
+        for dist in get_distributions(True):
+            if dist.name == name:
+                found = dist
+                break
     return found
 
+
+def obsoletes_distribution(name, version=None, use_egg_info=False):
+    """
+    Iterates over all distributions to find which distributions obsolete *name*.
+    If a *version* is provided, it will be used to filter the results.
+    If the argument *use_egg_info* is set to ``True``, then ``.egg-info``
+    distributions will be considered as well.
+
+    :type name: string
+    :type version: string
+    :parameter name:
+    """
+    for dist in get_distributions(use_egg_info):
+        obsoleted = dist.metadata['Obsoletes-Dist'] + dist.metadata['Obsoletes']
+        for obs in obsoleted:
+            o_components = obs.split(' ', 1)
+            if len(o_components) == 1 or version is None:
+                if name == o_components[0]:
+                    yield dist
+                    break
+            else:
+                try:
+                    predicate = VersionPredicate(obs)
+                except ValueError:
+                    raise DistutilsError(('Distribution %s has ill formed' +
+                                          ' obsoletes field') % (dist.name,))
+                if name == o_components[0] and predicate.match(version):
+                    yield dist
+                    break
+
+
+def provides_distribution(name, version=None, use_egg_info=False):
+    """
+    Iterates over all distributions to find which distributions provide *name*.
+    If a *version* is provided, it will be used to filter the results. Scans
+    all elements in ``sys.path``  and looks for all directories ending with
+    ``.dist-info``. Returns a :class:`Distribution`  corresponding to the
+    ``.dist-info`` directory that contains a ``METADATA`` that matches *name*
+    for the name metadata. If the argument *use_egg_info* is set to ``True``,
+    then all files and directories ending with ``.egg-info`` are considered
+    as well and returns an :class:`EggInfoDistribution` instance.
+
+    This function only returns the first result founded, since no more than
+    one values are expected. If the directory is not found, returns ``None``.
+
+    :parameter version: a version specifier that indicates the version
+                        required, conforming to the format in ``PEP-345``
+
+    :type name: string
+    :type version: string
+    """
+    predicate = None
+    if not version is None:
+        try:
+            predicate = VersionPredicate(name + ' (' + version + ')')
+        except ValueError:
+            raise DistutilsError('Invalid name or version')
+
+    for dist in get_distributions(use_egg_info):
+        provided = dist.metadata['Provides-Dist'] + dist.metadata['Provides']
+
+        for p in provided:
+            p_components = p.split(' ', 1)
+            if len(p_components) == 1 or predicate is None:
+                if name == p_components[0]:
+                    yield dist
+                    break
+            else:
+                p_name, p_ver = p_components
+                if len(p_ver) < 2 or p_ver[0] != '(' or p_ver[-1] != ')':
+                    raise DistutilsError(('Distribution %s has invalid ' +
+                                          'provides field') % (dist.name,))
+                p_ver = p_ver[1:-1] # trim off the parenthesis
+                if p_name == name and predicate.match(p_ver):
+                    yield dist
+                    break
+
+
 def get_file_users(path):
     """
     Iterates over all distributions to find out which distributions uses
diff --git a/src/distutils2/_backport/tests/fake_dists/bacon-0.1.egg-info/PKG-INFO b/src/distutils2/_backport/tests/fake_dists/bacon-0.1.egg-info/PKG-INFO
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/tests/fake_dists/bacon-0.1.egg-info/PKG-INFO
@@ -0,0 +1,5 @@
+Metadata-Version: 1.2
+Name: bacon
+Version: 0.1
+Provides-Dist: truffles (2.0)
+Obsoletes-Dist: truffles (>=0.9,<=1.5)
diff --git a/src/distutils2/_backport/tests/fake_dists/cheese-2.0.2.egg-info b/src/distutils2/_backport/tests/fake_dists/cheese-2.0.2.egg-info
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/tests/fake_dists/cheese-2.0.2.egg-info
@@ -0,0 +1,5 @@
+Metadata-Version: 1.2
+Name: cheese
+Version: 2.0.2
+Provides-Dist: truffles (1.0.2)
+Obsoletes-Dist: truffles (!=1.2,<=2.0)
diff --git a/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info/METADATA b/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info/METADATA
--- a/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info/METADATA
+++ b/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info/METADATA
@@ -4,3 +4,5 @@
 Summary: Chocolate with a kick!
 Requires-Dist: towel-stuff (0.1)
 Provides-Dist: truffles (1.0)
+Obsoletes-Dist: truffles (<=0.8,>=0.5)
+Obsoletes-Dist: truffles (<=0.9,>=0.6)
diff --git a/src/distutils2/_backport/tests/fake_dists/grammar-1.0a4.dist-info/METADATA b/src/distutils2/_backport/tests/fake_dists/grammar-1.0a4.dist-info/METADATA
--- a/src/distutils2/_backport/tests/fake_dists/grammar-1.0a4.dist-info/METADATA
+++ b/src/distutils2/_backport/tests/fake_dists/grammar-1.0a4.dist-info/METADATA
@@ -1,3 +1,4 @@
 Metadata-Version: 1.2
 Name: grammar
 Version: 1.0a4
+Requires-Dist: truffles (>=1.2)
diff --git a/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1.dist-info/METADATA b/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1.dist-info/METADATA
--- a/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1.dist-info/METADATA
+++ b/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1.dist-info/METADATA
@@ -1,3 +1,5 @@
 Metadata-Version: 1.2
 Name: towel-stuff
 Version: 0.1
+Provides-Dist: truffles (1.1.2)
+Obsoletes-Dist: truffles (!=0.8,<1.0)
diff --git a/src/distutils2/_backport/tests/test_pkgutil.py b/src/distutils2/_backport/tests/test_pkgutil.py
--- a/src/distutils2/_backport/tests/test_pkgutil.py
+++ b/src/distutils2/_backport/tests/test_pkgutil.py
@@ -216,7 +216,9 @@
         found_dists = []
 
         # Import the function in question
-        from distutils2._backport.pkgutil import get_distributions, Distribution
+        from distutils2._backport.pkgutil import get_distributions, \
+                                                 Distribution, \
+                                                 EggInfoDistribution
 
         # Verify the fake dists have been found.
         dists = [ dist for dist in get_distributions() ]
@@ -231,13 +233,31 @@
         # Finally, test that we found all that we were looking for
         self.assertListEqual(sorted(found_dists), sorted(fake_dists))
 
+        # Now, test if the egg-info distributions are found correctly as well
+        fake_dists += [('bacon', '0.1'), ('cheese', '2.0.2')]
+        found_dists = []
+
+        dists = [ dist for dist in get_distributions(use_egg_info=True) ]
+        for dist in dists:
+            if not (isinstance(dist, Distribution) or \
+                    isinstance(dist, EggInfoDistribution)):
+                self.fail("item received was not a Distribution or "
+                          "EggInfoDistribution instance: %s" % type(dist))
+            if dist.name in dict(fake_dists).keys():
+                found_dists.append((dist.name, dist.metadata['version']))
+
+        self.assertListEqual(sorted(fake_dists), sorted(found_dists))
+
+
     def test_get_distribution(self):
         """Test for looking up a distribution by name."""
         # Test the lookup of the towel-stuff distribution
         name = 'towel-stuff' # Note: This is different from the directory name
 
         # Import the function in question
-        from distutils2._backport.pkgutil import get_distribution, Distribution
+        from distutils2._backport.pkgutil import get_distribution, \
+                                                 Distribution, \
+                                                 EggInfoDistribution
 
         # Lookup the distribution
         dist = get_distribution(name)
@@ -250,6 +270,21 @@
         # Verify partial name matching doesn't work
         self.assertEqual(None, get_distribution('towel'))
 
+        # Verify that it does not find egg-info distributions, when not
+        # instructed to
+        self.assertEqual(None, get_distribution('bacon'))
+        self.assertEqual(None, get_distribution('cheese'))
+
+        # Now check that it works well in both situations, when egg-info
+        # is a file and directory respectively.
+        dist = get_distribution('cheese', use_egg_info=True)
+        self.assertTrue(isinstance(dist, EggInfoDistribution))
+        self.assertEqual(dist.name, 'cheese')
+
+        dist = get_distribution('bacon', use_egg_info=True)
+        self.assertTrue(isinstance(dist, EggInfoDistribution))
+        self.assertEqual(dist.name, 'bacon')
+
     def test_get_file_users(self):
         """Test the iteration of distributions that use a file."""
         from distutils2._backport.pkgutil import get_file_users, Distribution
@@ -260,6 +295,80 @@
             self.assertTrue(isinstance(dist, Distribution))
             self.assertEqual(dist.name, name)
 
+    def test_provides(self):
+        """ Test for looking up distributions by what they provide """
+        from distutils2._backport.pkgutil import provides_distribution
+        from distutils2.errors import DistutilsError
+
+        checkLists = lambda x,y: self.assertListEqual(sorted(x), sorted(y))
+
+        l = [dist.name for dist in provides_distribution('truffles')]
+        checkLists(l, ['choxie', 'towel-stuff'])
+
+        l = [dist.name for dist in provides_distribution('truffles', '1.0')]
+        checkLists(l, ['choxie'])
+
+        l = [dist.name for dist in provides_distribution('truffles', '1.0',
+                                                         use_egg_info=True)]
+        checkLists(l, ['choxie', 'cheese'])
+
+        l = [dist.name for dist in provides_distribution('truffles', '1.1.2')]
+        checkLists(l, ['towel-stuff'])
+
+        l = [dist.name for dist in provides_distribution('truffles', '1.1')]
+        checkLists(l, ['towel-stuff'])
+
+        l = [dist.name for dist in provides_distribution('truffles', '!=1.1,<=2.0')]
+        checkLists(l, ['choxie'])
+
+        l = [dist.name for dist in provides_distribution('truffles', '!=1.1,<=2.0',
+                                                          use_egg_info=True)]
+        checkLists(l, ['choxie', 'bacon', 'cheese'])
+
+        l = [dist.name for dist in provides_distribution('truffles', '>1.0')]
+        checkLists(l, ['towel-stuff'])
+
+        l = [dist.name for dist in provides_distribution('truffles', '>1.5')]
+        checkLists(l, [])
+
+        l = [dist.name for dist in provides_distribution('truffles', '>1.5',
+                                                         use_egg_info=True)]
+        checkLists(l, ['bacon'])
+
+        l = [dist.name for dist in provides_distribution('truffles', '>=1.0')]
+        checkLists(l, ['choxie', 'towel-stuff'])
+
+    def test_obsoletes(self):
+        """ Test looking for distributions based on what they obsolete """
+        from distutils2._backport.pkgutil import obsoletes_distribution
+        from distutils2.errors import DistutilsError
+
+        checkLists = lambda x,y: self.assertListEqual(sorted(x), sorted(y))
+
+        l = [dist.name for dist in obsoletes_distribution('truffles', '1.0')]
+        checkLists(l, [])
+
+        l = [dist.name for dist in obsoletes_distribution('truffles', '1.0',
+                                                          use_egg_info=True)]
+        checkLists(l, ['cheese', 'bacon'])
+
+
+        l = [dist.name for dist in obsoletes_distribution('truffles', '0.8')]
+        checkLists(l, ['choxie'])
+
+        l = [dist.name for dist in obsoletes_distribution('truffles', '0.8',
+                                                          use_egg_info=True)]
+        checkLists(l, ['choxie', 'cheese'])
+
+        l = [dist.name for dist in obsoletes_distribution('truffles', '0.9.6')]
+        checkLists(l, ['choxie', 'towel-stuff'])
+
+        l = [dist.name for dist in obsoletes_distribution('truffles', '0.5.2.3')]
+        checkLists(l, ['choxie', 'towel-stuff'])
+
+        l = [dist.name for dist in obsoletes_distribution('truffles', '0.2')]
+        checkLists(l, ['towel-stuff'])
+
 
 def test_suite():
     suite = unittest2.TestSuite()
@@ -273,3 +382,16 @@
 
 if __name__ == "__main__":
     test_main()
+
+def test_suite():
+    suite = unittest2.TestSuite()
+    testcase_loader = unittest2.loader.defaultTestLoader.loadTestsFromTestCase
+    suite.addTest(testcase_loader(TestPkgUtilFunctions))
+    suite.addTest(testcase_loader(TestPkgUtilDistribution))
+    return suite
+
+def test_main():
+    run_unittest(test_suite())
+
+if __name__ == "__main__":
+    test_main()
diff --git a/src/distutils2/command/__init__.py b/src/distutils2/command/__init__.py
--- a/src/distutils2/command/__init__.py
+++ b/src/distutils2/command/__init__.py
@@ -20,7 +20,6 @@
            'register',
            'bdist',
            'bdist_dumb',
-           'bdist_rpm',
            'bdist_wininst',
            'upload',
            'check',
diff --git a/src/distutils2/command/bdist.py b/src/distutils2/command/bdist.py
--- a/src/distutils2/command/bdist.py
+++ b/src/distutils2/command/bdist.py
@@ -55,9 +55,6 @@
          "lists available distribution formats", show_formats),
         ]
 
-    # The following commands do not take a format option from bdist
-    no_format_option = ('bdist_rpm',)
-
     # This won't do in reality: will need to distinguish RPM-ish Linux,
     # Debian-ish Linux, Solaris, FreeBSD, ..., Windows, Mac OS.
     default_format = {'posix': 'gztar',
@@ -69,8 +66,7 @@
                        'wininst', 'zip', 'msi']
 
     # And the real information.
-    format_command = {'rpm':   ('bdist_rpm',  "RPM distribution"),
-                      'gztar': ('bdist_dumb', "gzip'ed tar file"),
+    format_command = {'gztar': ('bdist_dumb', "gzip'ed tar file"),
                       'bztar': ('bdist_dumb', "bzip2'ed tar file"),
                       'ztar':  ('bdist_dumb', "compressed tar file"),
                       'tar':   ('bdist_dumb', "tar file"),
@@ -131,8 +127,6 @@
         for i in range(len(self.formats)):
             cmd_name = commands[i]
             sub_cmd = self.reinitialize_command(cmd_name)
-            if cmd_name not in self.no_format_option:
-                sub_cmd.format = self.formats[i]
 
             # passing the owner and group names for tar archiving
             if cmd_name == 'bdist_dumb':
diff --git a/src/distutils2/command/sdist.py b/src/distutils2/command/sdist.py
--- a/src/distutils2/command/sdist.py
+++ b/src/distutils2/command/sdist.py
@@ -10,6 +10,8 @@
 from glob import glob
 from warnings import warn
 from shutil import rmtree
+import re
+
 try:
     from shutil import get_archive_formats
 except ImportError:
@@ -17,10 +19,9 @@
 
 from distutils2.core import Command
 from distutils2 import util
-from distutils2.text_file import TextFile
 from distutils2.errors import (DistutilsPlatformError, DistutilsOptionError,
                               DistutilsTemplateError)
-from distutils2.filelist import FileList
+from distutils2.manifest import Manifest
 from distutils2 import log
 from distutils2.util import convert_path, newer
 
@@ -36,6 +37,10 @@
     FancyGetopt(formats).print_help(
         "List of available source distribution formats:")
 
+# a \ followed by some spaces + EOL
+_COLLAPSE_PATTERN = re.compile('\\\w\n', re.M)
+_COMMENTED_LINE = re.compile('^#.*\n$|^\w*\n$', re.M)
+
 class sdist(Command):
 
     description = "create a source distribution (tarball, zip file, etc.)"
@@ -63,10 +68,7 @@
         ('no-prune', None,
          "don't automatically exclude anything"),
         ('manifest-only', 'o',
-         "just regenerate the manifest and then stop "
-         "(implies --force-manifest)"),
-        ('force-manifest', 'f',
-         "forcibly regenerate the manifest and carry on as usual"),
+         "just regenerate the manifest and then stop "),
         ('formats=', None,
          "formats for source distribution (comma-separated list)"),
         ('keep-temp', 'k',
@@ -85,8 +87,7 @@
         ]
 
     boolean_options = ['use-defaults', 'prune',
-                       'manifest-only', 'force-manifest',
-                       'keep-temp', 'metadata-check']
+                       'manifest-only', 'keep-temp', 'metadata-check']
 
     help_options = [
         ('help-formats', None,
@@ -111,10 +112,7 @@
         # in the manifest
         self.use_defaults = 1
         self.prune = 1
-
         self.manifest_only = 0
-        self.force_manifest = 0
-
         self.formats = None
         self.keep_temp = 0
         self.dist_dir = None
@@ -123,6 +121,7 @@
         self.metadata_check = 1
         self.owner = None
         self.group = None
+        self.filelist = None
 
     def _check_archive_formats(self, formats):
         supported_formats = [name for name, desc in get_archive_formats()]
@@ -154,10 +153,14 @@
         if self.dist_dir is None:
             self.dist_dir = "dist"
 
+        if self.filelist is None:
+            self.filelist = Manifest()
+
+
     def run(self):
         # 'filelist' contains the list of files that will make up the
         # manifest
-        self.filelist = FileList()
+        self.filelist.clear()
 
         # Run sub commands
         for cmd_name in self.get_sub_commands():
@@ -189,64 +192,24 @@
         distribution, and put it in 'self.filelist'.  This might involve
         reading the manifest template (and writing the manifest), or just
         reading the manifest, or just using the default file set -- it all
-        depends on the user's options and the state of the filesystem.
+        depends on the user's options.
         """
-        # If we have a manifest template, see if it's newer than the
-        # manifest; if so, we'll regenerate the manifest.
         template_exists = os.path.isfile(self.template)
+        if not template_exists:
+            self.warn(("manifest template '%s' does not exist " +
+                        "(using default file list)") %
+                        self.template)
+
+        self.filelist.findall()
+
+        if self.use_defaults:
+            self.add_defaults()
         if template_exists:
-            template_newer = newer(self.template, self.manifest)
+            self.filelist.read_template(self.template)
+        if self.prune:
+            self.prune_file_list()
 
-        # The contents of the manifest file almost certainly depend on the
-        # setup script as well as the manifest template -- so if the setup
-        # script is newer than the manifest, we'll regenerate the manifest
-        # from the template.  (Well, not quite: if we already have a
-        # manifest, but there's no template -- which will happen if the
-        # developer elects to generate a manifest some other way -- then we
-        # can't regenerate the manifest, so we don't.)
-        setup_newer = newer(self.distribution.script_name,
-                            self.manifest)
-
-        # cases:
-        #   1) no manifest, template exists: generate manifest
-        #      (covered by 2a: no manifest == template newer)
-        #   2) manifest & template exist:
-        #      2a) template or setup script newer than manifest:
-        #          regenerate manifest
-        #      2b) manifest newer than both:
-        #          do nothing (unless --force or --manifest-only)
-        #   3) manifest exists, no template:
-        #      do nothing (unless --force or --manifest-only)
-        #   4) no manifest, no template: generate w/ warning ("defaults only")
-
-        manifest_outofdate = (template_exists and
-                              (template_newer or setup_newer))
-        force_regen = self.force_manifest or self.manifest_only
-        manifest_exists = os.path.isfile(self.manifest)
-        neither_exists = (not template_exists and not manifest_exists)
-
-        # Regenerate the manifest if necessary (or if explicitly told to)
-        if manifest_outofdate or neither_exists or force_regen:
-            if not template_exists:
-                self.warn(("manifest template '%s' does not exist " +
-                           "(using default file list)") %
-                          self.template)
-            self.filelist.findall()
-
-            if self.use_defaults:
-                self.add_defaults()
-            if template_exists:
-                self.read_template()
-            if self.prune:
-                self.prune_file_list()
-
-            self.filelist.sort()
-            self.filelist.remove_duplicates()
-            self.write_manifest()
-
-        # Don't regenerate the manifest, just read it in.
-        else:
-            self.read_manifest()
+        self.filelist.write(self.manifest)
 
     def add_defaults(self):
         """Add all the default files to self.filelist:
@@ -330,32 +293,6 @@
             build_scripts = self.get_finalized_command('build_scripts')
             self.filelist.extend(build_scripts.get_source_files())
 
-    def read_template(self):
-        """Read and parse manifest template file named by self.template.
-
-        (usually "MANIFEST.in") The parsing and processing is done by
-        'self.filelist', which updates itself accordingly.
-        """
-        log.info("reading manifest template '%s'", self.template)
-        template = TextFile(self.template,
-                            strip_comments=1,
-                            skip_blanks=1,
-                            join_lines=1,
-                            lstrip_ws=1,
-                            rstrip_ws=1,
-                            collapse_join=1)
-
-        while 1:
-            line = template.readline()
-            if line is None:            # end of file
-                break
-
-            try:
-                self.filelist.process_template_line(line)
-            except DistutilsTemplateError, msg:
-                self.warn("%s, line %d: %s" % (template.filename,
-                                               template.current_line,
-                                               msg))
 
     def prune_file_list(self):
         """Prune off branches that might slip into the file list as created
@@ -383,30 +320,6 @@
         vcs_ptrn = r'(^|%s)(%s)(%s).*' % (seps, '|'.join(vcs_dirs), seps)
         self.filelist.exclude_pattern(vcs_ptrn, is_regex=1)
 
-    def write_manifest(self):
-        """Write the file list in 'self.filelist' (presumably as filled in
-        by 'add_defaults()' and 'read_template()') to the manifest file
-        named by 'self.manifest'.
-        """
-        self.execute(util.write_file,
-                     (self.manifest, self.filelist.files),
-                     "writing manifest file '%s'" % self.manifest)
-
-    def read_manifest(self):
-        """Read the manifest file (named by 'self.manifest') and use it to
-        fill in 'self.filelist', the list of files to include in the source
-        distribution.
-        """
-        log.info("reading manifest file '%s'", self.manifest)
-        manifest = open(self.manifest)
-        while 1:
-            line = manifest.readline()
-            if line == '':              # end of file
-                break
-            if line[-1] == '\n':
-                line = line[0:-1]
-            self.filelist.append(line)
-        manifest.close()
 
     def make_release_tree(self, base_dir, files):
         """Create the directory tree that will become the source
diff --git a/src/distutils2/command/upload.py b/src/distutils2/command/upload.py
--- a/src/distutils2/command/upload.py
+++ b/src/distutils2/command/upload.py
@@ -96,11 +96,7 @@
         data['md5_digest'] = md5(content).hexdigest()
 
         comment = ''
-        if command == 'bdist_rpm':
-            dist, version, id = platform.dist()
-            if dist:
-                comment = 'built for %s %s' % (dist, version)
-        elif command == 'bdist_dumb':
+        if command == 'bdist_dumb':
             comment = 'built for %s' % platform.platform(terse=1)
         data['comment'] = comment
 
diff --git a/src/distutils2/converter/__init__.py b/src/distutils2/converter/__init__.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/converter/__init__.py
@@ -0,0 +1,8 @@
+"""distutils2.converter
+
+This package provide a refactoring tool to transform a
+setuptools or distutils project into a distutils2 one.
+"""
+
+from distutils2.converter.refactor import DistutilsRefactoringTool
+
diff --git a/src/distutils2/converter/fixers/__init__.py b/src/distutils2/converter/fixers/__init__.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/converter/fixers/__init__.py
@@ -0,0 +1,4 @@
+"""distutils2.converter.fixers
+
+Contains all fixers for the converter.
+"""
diff --git a/src/distutils2/converter/fixers/fix_imports.py b/src/distutils2/converter/fixers/fix_imports.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/converter/fixers/fix_imports.py
@@ -0,0 +1,47 @@
+"""distutils2.converter.fixers.fix_imports
+
+Fixer for import statements in setup.py
+"""
+from lib2to3.fixer_base import BaseFix
+from lib2to3.fixer_util import syms
+
+
+class FixImports(BaseFix):
+    """Makes sure all import in setup.py are translated"""
+
+    PATTERN = """
+    import_from< 'from' imp=any 'import' ['('] any [')'] >
+    |
+    import_name< 'import' imp=any >
+    """
+
+    def transform(self, node, results):
+        imp = results['imp']
+        if node.type != syms.import_from:
+            return
+
+        if not hasattr(imp, "next_sibling"):
+            imp.next_sibling = imp.get_next_sibling()
+
+        while not hasattr(imp, 'value'):
+            imp = imp.children[0]
+
+        if imp.value == 'distutils':
+            imp.value = 'distutils2'
+            imp.changed()
+            return node
+
+        if imp.value == 'setuptools':
+            # catching "from setuptools import setup"
+            pattern = []
+            next = imp.next_sibling
+            while next is not None:
+                pattern.append(next.value)
+                if not hasattr(next, "next_sibling"):
+                    next.next_sibling = next.get_next_sibling()
+                next = next.next_sibling
+            if pattern == ['import', 'setup']:
+                imp.value = 'distutils2.core'
+                imp.changed()
+
+            return node
diff --git a/src/distutils2/converter/fixers/fix_setup_options.py b/src/distutils2/converter/fixers/fix_setup_options.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/converter/fixers/fix_setup_options.py
@@ -0,0 +1,85 @@
+"""Fixer for setup() options.
+
+All distutils or setuptools options are translated
+into PEP 345-style options.
+"""
+from lib2to3.pytree import Leaf, Node
+from lib2to3.pgen2 import token
+from lib2to3.fixer_base import BaseFix
+
+# XXX where is that defined ?
+_ARG = 260
+
+# name mapping : we want to convert
+# all old-style options to distutils2 style
+_OLD_NAMES = {'url': 'home_page',
+              'long_description': 'description',
+              'description': 'summary',
+              'install_requires': 'requires_dist'}
+
+_SEQUENCE_NAMES = ['requires_dist']
+
+
+class FixSetupOptions(BaseFix):
+
+    # XXX need to find something better here :
+    # identify a setup call, whatever alias is used
+    PATTERN = """
+            power< name='setup' trailer< '(' [any] ')' > any* >
+              """
+
+    def _get_list(self, *nodes):
+        """A List node, filled"""
+        lbrace = Leaf(token.LBRACE, u"[")
+        lbrace.prefix = u" "
+        if len(nodes) > 0:
+            nodes[0].prefix = u""
+        return Node(self.syms.trailer,
+                    [lbrace] +
+                    [node.clone() for node in nodes] +
+                    [Leaf(token.RBRACE, u"]")])
+
+    def _fix_name(self, argument, remove_list):
+        name = argument.children[0]
+
+        if not hasattr(name, "next_sibling"):
+            name.next_sibling = name.get_next_sibling()
+
+        sibling = name.next_sibling
+        if sibling is None or sibling.type != token.EQUAL:
+            return False
+
+        if name.value in _OLD_NAMES:
+            name.value = _OLD_NAMES[name.value]
+            if name.value in _SEQUENCE_NAMES:
+                if not hasattr(sibling, "next_sibling"):
+                    sibling.next_sibling = sibling.get_next_sibling()
+                right_operand = sibling.next_sibling
+                # replacing string -> list[string]
+                if right_operand.type == token.STRING:
+                    # we want this to be a list now
+                    new_node = self._get_list(right_operand)
+                    right_operand.replace(new_node)
+
+
+            return True
+
+        return False
+
+    def transform(self, node, results):
+        arglist = node.children[1].children[1]
+        remove_list = []
+        changed = False
+
+        for subnode in arglist.children:
+            if subnode.type != _ARG:
+                continue
+            if self._fix_name(subnode, remove_list) and not changed:
+                changed = True
+
+        for subnode in remove_list:
+            subnode.remove()
+
+        if changed:
+            node.changed()
+        return node
diff --git a/src/distutils2/converter/refactor.py b/src/distutils2/converter/refactor.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/converter/refactor.py
@@ -0,0 +1,28 @@
+"""distutils2.converter.refactor
+
+Provides DistutilsRefactoringTool, a class that register fixers used
+to refactor distutils or setuptools packages into distutils2 ones.
+"""
+try:
+    from lib2to3.refactor import RefactoringTool
+    _LIB2TO3 = True
+except ImportError:
+    # we need 2.6 at least to run this
+    _LIB2TO3 = False
+
+_DISTUTILS_FIXERS = ['distutils2.converter.fixers.fix_imports',
+                     'distutils2.converter.fixers.fix_setup_options']
+
+if _LIB2TO3:
+    class DistutilsRefactoringTool(RefactoringTool):
+
+        def __init__(self, fixer_names=_DISTUTILS_FIXERS, options=None,
+                    explicit=None):
+
+            super(DistutilsRefactoringTool, self).__init__(fixer_names, options,
+                                                            explicit)
+else:
+    class DistutilsRefactoringTool(object):
+        def __init__(self, *args, **kw):
+            raise NotImplementedError('Not available if run from Python < 2.6')
+
diff --git a/src/distutils2/depgraph.py b/src/distutils2/depgraph.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/depgraph.py
@@ -0,0 +1,181 @@
+"""
+A dependency graph generator. The graph is represented as an instance of
+:class:`DependencyGraph`, and DOT output is possible as well.
+"""
+
+from distutils2._backport import pkgutil
+from distutils2.errors import DistutilsError
+from distutils2.version import VersionPredicate
+
+__all__ = ['DependencyGraph', 'generate_graph']
+
+
+class DependencyGraph(object):
+    """
+    Represents a dependency graph between distributions.
+
+    The depedency relationships are stored in an *adjacency_list* that maps
+    distributions to a list of ``(other, label)`` tuples where  ``other``
+    is a distribution and the edge is labelled with ``label`` (i.e. the version
+    specifier, if such was provided). If any missing depencies are found,
+    they are stored in ``missing``. It maps distributions to a list of
+    requirements that were not provided by any other distributions.
+    """
+
+    def __init__(self):
+        self.adjacency_list = {}
+        self.missing = {}
+
+    def add_distribution(self, distribution):
+        """
+        Add distribution *x* to the graph.
+
+        :type distribution: :class:`pkgutil.Distribution` or
+                            :class:`pkgutil.EggInfoDistribution`
+        """
+        self.adjacency_list[distribution] = list()
+        self.missing[distribution] = list()
+
+    def add_edge(self, x, y, label=None):
+        """
+        Add an edge from distribution *x* to distribution *y* with the given
+        *label*.
+
+
+        :type x: :class:`pkgutil.Distribution` or
+                 :class:`pkgutil.EggInfoDistribution`
+        :type y: :class:`pkgutil.Distribution` or
+                 :class:`pkgutil.EggInfoDistribution`
+        :type label: ``str`` or ``None``
+        """
+        self.adjacency_list[x].append((y, label))
+
+    def add_missing(self, distribution, requirement):
+        """
+        Add a missing *requirement* for the given *distribution*.
+
+        :type distribution: :class:`pkgutil.Distribution` or
+                            :class:`pkgutil.EggInfoDistribution`
+        :type requirement: ``str``
+        """
+        self.missing[distribution].append(requirement)
+
+    def to_dot(self, f, skip_disconnected=True):
+        """
+        Writes a DOT output for the graph to the provided *file*.
+        If *skip_disconnected* is set to ``True``, then all distributions
+        that are not dependent on any other distributions are skipped.
+
+        :type f: ``file``
+        ;type skip_disconnected: ``bool``
+        """
+        if not isinstance(f, file):
+            raise TypeError('the argument has to be of type file')
+
+        disconnected = []
+
+        f.write("digraph dependencies {\n")
+        for dist, adjs in self.adjacency_list.iteritems():
+            if len(adjs) == 0 and not skip_disconnected:
+                disconnected.append(dist)
+            for (other, label) in adjs:
+                if not label is None:
+                    f.write('"%s" -> "%s" [label="%s"]\n' %
+                                                (dist.name, other.name, label))
+                else:
+                    f.write('"%s" -> "%s"\n' % (dist.name, other.name))
+        if not skip_disconnected and len(disconnected) > 0:
+            f.write('subgraph disconnected {\n')
+            f.write('label = "Disconnected"\n')
+            f.write('bgcolor = red\n')
+
+            for dist in disconnected:
+                f.write('"%s"' % dist.name)
+                f.write('\n')
+            f.write('}\n')
+        f.write('}\n')
+
+
+def generate_graph(dists):
+    """
+    Generates a dependency graph from the given distributions.
+
+    :parameter dists: a list of distributions
+    :type dists: list of :class:`pkgutil.Distribution` and
+                         :class:`pkgutil.EggInfoDistribution` instances
+    :rtype: an :class:`DependencyGraph` instance
+    """
+    graph = DependencyGraph()
+    provided = {} # maps names to lists of (version, dist) tuples
+    dists = list(dists) # maybe use generator_tools in future
+
+    # first, build the graph and find out the provides
+    for dist in dists:
+        graph.add_distribution(dist)
+        provides = dist.metadata['Provides-Dist'] + dist.metadata['Provides']
+
+        for p in provides:
+            comps = p.split(" ", 1)
+            name = comps[0]
+            version = None
+            if len(comps) == 2:
+                version = comps[1]
+                if len(version) < 3 or version[0] != '(' or version[-1] != ')':
+                    raise DistutilsError('Distribution %s has ill formed' \
+                                         'provides field: %s' % (dist.name, p))
+                version = version[1:-1] # trim off parenthesis
+            if not name in provided:
+                provided[name] = []
+            provided[name].append((version, dist))
+
+    # now make the edges
+    for dist in dists:
+        requires = dist.metadata['Requires-Dist'] + dist.metadata['Requires']
+        for req in requires:
+            predicate = VersionPredicate(req)
+            comps = req.split(" ", 1)
+            name = comps[0]
+
+            if not name in provided:
+                graph.add_missing(dist, req)
+            else:
+                for (version, provider) in provided[name]:
+                    if predicate.match(version):
+                        graph.add_edge(dist, provider, req)
+
+    return graph
+
+
+def dependent_dists(dists, dist):
+    """
+    Recursively generate a list of distributions from *dists* that are
+    dependent on *dist*.
+
+    :param dists: a list of distributions
+    :param dist: a distribution, member of *dists* for which we are interested
+    """
+    if not dist in dists:
+        raise ValueError('The given distribution is not a member of the list')
+    graph = generate_graph(dists)
+
+    dep = [dist]
+    fringe = [dist] # list of nodes we should expand
+    while not len(fringe) == 0:
+        next = graph.adjacency_list[fringe.pop()]
+        for (dist, label) in next:
+            if not dist in dep: # avoid infinite loops
+                dep.append(dist)
+                fringe.append(dist)
+
+    dep.pop()
+    return dep
+
+if __name__ == '__main__':
+    dists = list(pkgutil.get_distributions(use_egg_info=True))
+    graph = generate_graph(dists)
+    for dist, reqs in graph.missing.iteritems():
+        if len(reqs) > 0:
+            print("Missing dependencies for %s: %s" % (dist.name,
+                                                       ", ".join(reqs)))
+    f = open('output.dot', 'w')
+    graph.to_dot(f, True)
diff --git a/src/distutils2/filelist.py b/src/distutils2/filelist.py
deleted file mode 100644
--- a/src/distutils2/filelist.py
+++ /dev/null
@@ -1,306 +0,0 @@
-"""distutils.filelist
-
-Provides the FileList class, used for poking about the filesystem
-and building lists of files.
-"""
-
-__revision__ = "$Id: filelist.py 75196 2009-10-03 00:07:35Z tarek.ziade $"
-
-import os, re
-import fnmatch
-from distutils2.util import convert_path
-from distutils2.errors import DistutilsTemplateError, DistutilsInternalError
-from distutils2 import log
-
-class FileList(object):
-    """A list of files built by on exploring the filesystem and filtered by
-    applying various patterns to what we find there.
-
-    Instance attributes:
-      dir
-        directory from which files will be taken -- only used if
-        'allfiles' not supplied to constructor
-      files
-        list of filenames currently being built/filtered/manipulated
-      allfiles
-        complete list of files under consideration (ie. without any
-        filtering applied)
-    """
-
-    def __init__(self):
-        self.allfiles = None
-        self.files = []
-
-    def findall(self, dir=os.curdir):
-        self.allfiles = findall(dir)
-
-    # -- List-like methods ---------------------------------------------
-
-    def append(self, item):
-        self.files.append(item)
-
-    def extend(self, items):
-        self.files.extend(items)
-
-    def sort(self):
-        # Not a strict lexical sort!
-        sortable_files = map(os.path.split, self.files)
-        sortable_files.sort()
-        self.files = []
-        for sort_tuple in sortable_files:
-            self.files.append(os.path.join(*sort_tuple))
-
-
-    # -- Other miscellaneous utility methods ---------------------------
-
-    def remove_duplicates(self):
-        # Assumes list has been sorted!
-        for i in range(len(self.files) - 1, 0, -1):
-            if self.files[i] == self.files[i - 1]:
-                del self.files[i]
-
-
-    # -- "File template" methods ---------------------------------------
-
-    def _parse_template_line(self, line):
-        words = line.split()
-        action = words[0]
-
-        patterns = dir = dir_pattern = None
-
-        if action in ('include', 'exclude',
-                      'global-include', 'global-exclude'):
-            if len(words) < 2:
-                raise DistutilsTemplateError, \
-                      "'%s' expects <pattern1> <pattern2> ..." % action
-
-            patterns = map(convert_path, words[1:])
-
-        elif action in ('recursive-include', 'recursive-exclude'):
-            if len(words) < 3:
-                raise DistutilsTemplateError, \
-                      "'%s' expects <dir> <pattern1> <pattern2> ..." % action
-
-            dir = convert_path(words[1])
-            patterns = map(convert_path, words[2:])
-
-        elif action in ('graft', 'prune'):
-            if len(words) != 2:
-                raise DistutilsTemplateError, \
-                     "'%s' expects a single <dir_pattern>" % action
-
-            dir_pattern = convert_path(words[1])
-
-        else:
-            raise DistutilsTemplateError, "unknown action '%s'" % action
-
-        return (action, patterns, dir, dir_pattern)
-
-    def process_template_line(self, line):
-        # Parse the line: split it up, make sure the right number of words
-        # is there, and return the relevant words.  'action' is always
-        # defined: it's the first word of the line.  Which of the other
-        # three are defined depends on the action; it'll be either
-        # patterns, (dir and patterns), or (dir_pattern).
-        action, patterns, dir, dir_pattern = self._parse_template_line(line)
-
-        # OK, now we know that the action is valid and we have the
-        # right number of words on the line for that action -- so we
-        # can proceed with minimal error-checking.
-        if action == 'include':
-            for pattern in patterns:
-                if not self.include_pattern(pattern, anchor=1):
-                    log.warn("warning: no files found matching '%s'",
-                             pattern)
-
-        elif action == 'exclude':
-            for pattern in patterns:
-                if not self.exclude_pattern(pattern, anchor=1):
-                    log.warn(("warning: no previously-included files "
-                              "found matching '%s'"), pattern)
-
-        elif action == 'global-include':
-            for pattern in patterns:
-                if not self.include_pattern(pattern, anchor=0):
-                    log.warn(("warning: no files found matching '%s' " +
-                              "anywhere in distribution"), pattern)
-
-        elif action == 'global-exclude':
-            for pattern in patterns:
-                if not self.exclude_pattern(pattern, anchor=0):
-                    log.warn(("warning: no previously-included files matching "
-                              "'%s' found anywhere in distribution"),
-                             pattern)
-
-        elif action == 'recursive-include':
-            for pattern in patterns:
-                if not self.include_pattern(pattern, prefix=dir):
-                    log.warn(("warning: no files found matching '%s' " +
-                                "under directory '%s'"),
-                             pattern, dir)
-
-        elif action == 'recursive-exclude':
-            for pattern in patterns:
-                if not self.exclude_pattern(pattern, prefix=dir):
-                    log.warn(("warning: no previously-included files matching "
-                              "'%s' found under directory '%s'"),
-                             pattern, dir)
-
-        elif action == 'graft':
-            if not self.include_pattern(None, prefix=dir_pattern):
-                log.warn("warning: no directories found matching '%s'",
-                         dir_pattern)
-
-        elif action == 'prune':
-            if not self.exclude_pattern(None, prefix=dir_pattern):
-                log.warn(("no previously-included directories found " +
-                          "matching '%s'"), dir_pattern)
-        else:
-            raise DistutilsInternalError, \
-                  "this cannot happen: invalid action '%s'" % action
-
-    # -- Filtering/selection methods -----------------------------------
-
-    def include_pattern(self, pattern, anchor=1, prefix=None, is_regex=0):
-        """Select strings (presumably filenames) from 'self.files' that
-        match 'pattern', a Unix-style wildcard (glob) pattern.
-
-        Patterns are not quite the same as implemented by the 'fnmatch'
-        module: '*' and '?'  match non-special characters, where "special"
-        is platform-dependent: slash on Unix; colon, slash, and backslash on
-        DOS/Windows; and colon on Mac OS.
-
-        If 'anchor' is true (the default), then the pattern match is more
-        stringent: "*.py" will match "foo.py" but not "foo/bar.py".  If
-        'anchor' is false, both of these will match.
-
-        If 'prefix' is supplied, then only filenames starting with 'prefix'
-        (itself a pattern) and ending with 'pattern', with anything in between
-        them, will match.  'anchor' is ignored in this case.
-
-        If 'is_regex' is true, 'anchor' and 'prefix' are ignored, and
-        'pattern' is assumed to be either a string containing a regex or a
-        regex object -- no translation is done, the regex is just compiled
-        and used as-is.
-
-        Selected strings will be added to self.files.
-
-        Return 1 if files are found.
-        """
-        files_found = 0
-        pattern_re = translate_pattern(pattern, anchor, prefix, is_regex)
-        # delayed loading of allfiles list
-        if self.allfiles is None:
-            self.findall()
-
-        for name in self.allfiles:
-            if pattern_re.search(name):
-                self.files.append(name)
-                files_found = 1
-
-        return files_found
-
-
-    def exclude_pattern(self, pattern, anchor=1, prefix=None, is_regex=0):
-        """Remove strings (presumably filenames) from 'files' that match
-        'pattern'.
-
-        Other parameters are the same as for 'include_pattern()', above.
-        The list 'self.files' is modified in place. Return 1 if files are
-        found.
-        """
-        files_found = 0
-        pattern_re = translate_pattern(pattern, anchor, prefix, is_regex)
-        for i in range(len(self.files)-1, -1, -1):
-            if pattern_re.search(self.files[i]):
-                del self.files[i]
-                files_found = 1
-
-        return files_found
-
-
-# ----------------------------------------------------------------------
-# Utility functions
-
-def findall(dir=os.curdir):
-    """Find all files under 'dir' and return the list of full filenames
-    (relative to 'dir').
-    """
-    from stat import ST_MODE, S_ISREG, S_ISDIR, S_ISLNK
-
-    list = []
-    stack = [dir]
-    pop = stack.pop
-    push = stack.append
-
-    while stack:
-        dir = pop()
-        names = os.listdir(dir)
-
-        for name in names:
-            if dir != os.curdir:        # avoid the dreaded "./" syndrome
-                fullname = os.path.join(dir, name)
-            else:
-                fullname = name
-
-            # Avoid excess stat calls -- just one will do, thank you!
-            stat = os.stat(fullname)
-            mode = stat[ST_MODE]
-            if S_ISREG(mode):
-                list.append(fullname)
-            elif S_ISDIR(mode) and not S_ISLNK(mode):
-                push(fullname)
-
-    return list
-
-
-def glob_to_re(pattern):
-    """Translate a shell-like glob pattern to a regular expression.
-
-    Return a string containing the regex.  Differs from
-    'fnmatch.translate()' in that '*' does not match "special characters"
-    (which are platform-specific).
-    """
-    pattern_re = fnmatch.translate(pattern)
-
-    # '?' and '*' in the glob pattern become '.' and '.*' in the RE, which
-    # IMHO is wrong -- '?' and '*' aren't supposed to match slash in Unix,
-    # and by extension they shouldn't match such "special characters" under
-    # any OS.  So change all non-escaped dots in the RE to match any
-    # character except the special characters.
-    # XXX currently the "special characters" are just slash -- i.e. this is
-    # Unix-only.
-    pattern_re = re.sub(r'((?<!\\)(\\\\)*)\.', r'\1[^/]', pattern_re)
-
-    return pattern_re
-
-
-def translate_pattern(pattern, anchor=1, prefix=None, is_regex=0):
-    """Translate a shell-like wildcard pattern to a compiled regular
-    expression.
-
-    Return the compiled regex.  If 'is_regex' true,
-    then 'pattern' is directly compiled to a regex (if it's a string)
-    or just returned as-is (assumes it's a regex object).
-    """
-    if is_regex:
-        if isinstance(pattern, str):
-            return re.compile(pattern)
-        else:
-            return pattern
-
-    if pattern:
-        pattern_re = glob_to_re(pattern)
-    else:
-        pattern_re = ''
-
-    if prefix is not None:
-        # ditch end of pattern character
-        empty_pattern = glob_to_re('')
-        prefix_re = glob_to_re(prefix)[:-len(empty_pattern)]
-        pattern_re = "^" + os.path.join(prefix_re, ".*" + pattern_re)
-    else:                               # no prefix -- respect anchor flag
-        if anchor:
-            pattern_re = "^" + pattern_re
-
-    return re.compile(pattern_re)
diff --git a/src/distutils2/manifest.py b/src/distutils2/manifest.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/manifest.py
@@ -0,0 +1,361 @@
+"""distutils2.manifest
+
+Provides a Manifest class that can be used to:
+
+ - read or write a MANIFEST file
+ - read a template file and find out the file list
+
+Basically, Manifest *is* the file list.
+
+XXX todo: document + add tests
+"""
+import re
+import os
+import fnmatch
+import logging
+
+from distutils2.util import write_file, convert_path
+from distutils2.errors import (DistutilsTemplateError,
+                               DistutilsInternalError)
+
+__all__ = ['Manifest']
+
+# a \ followed by some spaces + EOL
+_COLLAPSE_PATTERN = re.compile('\\\w*\n', re.M)
+_COMMENTED_LINE = re.compile('#.*?(?=\n)|^\w*\n|\n(?=$)', re.M|re.S)
+
+class Manifest(object):
+    """A list of files built by on exploring the filesystem and filtered by
+    applying various patterns to what we find there.
+    """
+
+    def __init__(self):
+        self.allfiles = None
+        self.files = []
+
+    #
+    # Public API
+    #
+
+    def findall(self, dir=os.curdir):
+        self.allfiles = _findall(dir)
+
+    def append(self, item):
+        self.files.append(item)
+
+    def extend(self, items):
+        self.files.extend(items)
+
+    def sort(self):
+        # Not a strict lexical sort!
+        sortable_files = map(os.path.split, self.files)
+        sortable_files.sort()
+        self.files = []
+        for sort_tuple in sortable_files:
+            self.files.append(os.path.join(*sort_tuple))
+
+    def clear(self):
+        """Clear all collected files."""
+        self.files = []
+        if self.allfiles is not None:
+            self.allfiles = []
+
+    def remove_duplicates(self):
+        # Assumes list has been sorted!
+        for i in range(len(self.files) - 1, 0, -1):
+            if self.files[i] == self.files[i - 1]:
+                del self.files[i]
+
+    def read_template(self, path):
+        """Read and parse a manifest template file.
+
+        Updates the list accordingly.
+        """
+        f = open(path)
+        try:
+            content = f.read()
+            # first, let's unwrap collapsed lines
+            content = _COLLAPSE_PATTERN.sub('', content)
+
+            # next, let's remove commented lines and empty lines
+            content = _COMMENTED_LINE.sub('', content)
+
+            # now we have our cleaned up lines
+            lines = [line.strip() for line in content.split('\n')]
+        finally:
+            f.close()
+
+        for line in lines:
+            try:
+                self._process_template_line(line)
+            except DistutilsTemplateError, msg:
+                logging.warning("%s, %s" % (path, msg))
+
+    def write(self, path):
+        """Write the file list in 'self.filelist' (presumably as filled in
+        by 'add_defaults()' and 'read_template()') to the manifest file
+        named by 'self.manifest'.
+        """
+        self.sort()
+        self.remove_duplicates()
+        write_file(path, self.files)
+
+    def read(self, path):
+        """Read the manifest file (named by 'self.manifest') and use it to
+        fill in 'self.filelist', the list of files to include in the source
+        distribution.
+        """
+        logging.info("reading manifest file '%s'" % path)
+        manifest = open(path)
+        try:
+            for line in manifest.readlines():
+                self.append(line)
+        finally:
+            manifest.close()
+
+    def exclude_pattern(self, pattern, anchor=1, prefix=None, is_regex=0):
+        """Remove strings (presumably filenames) from 'files' that match
+        'pattern'.
+
+        Other parameters are the same as for 'include_pattern()', above.
+        The list 'self.files' is modified in place. Return 1 if files are
+        found.
+        """
+        files_found = 0
+        pattern_re = _translate_pattern(pattern, anchor, prefix, is_regex)
+        for i in range(len(self.files)-1, -1, -1):
+            if pattern_re.search(self.files[i]):
+                del self.files[i]
+                files_found = 1
+
+        return files_found
+
+    #
+    # Private API
+    #
+
+    def _parse_template_line(self, line):
+        words = line.split()
+        action = words[0]
+
+        patterns = dir = dir_pattern = None
+
+        if action in ('include', 'exclude',
+                      'global-include', 'global-exclude'):
+            if len(words) < 2:
+                raise DistutilsTemplateError(
+                      "'%s' expects <pattern1> <pattern2> ..." % action)
+
+            patterns = map(convert_path, words[1:])
+
+        elif action in ('recursive-include', 'recursive-exclude'):
+            if len(words) < 3:
+                raise DistutilsTemplateError(
+                      "'%s' expects <dir> <pattern1> <pattern2> ..." % action)
+
+            dir = convert_path(words[1])
+            patterns = map(convert_path, words[2:])
+
+        elif action in ('graft', 'prune'):
+            if len(words) != 2:
+                raise DistutilsTemplateError(
+                     "'%s' expects a single <dir_pattern>" % action)
+
+            dir_pattern = convert_path(words[1])
+
+        else:
+            raise DistutilsTemplateError("unknown action '%s'" % action)
+
+        return action, patterns, dir, dir_pattern
+
+    def _process_template_line(self, line):
+        # Parse the line: split it up, make sure the right number of words
+        # is there, and return the relevant words.  'action' is always
+        # defined: it's the first word of the line.  Which of the other
+        # three are defined depends on the action; it'll be either
+        # patterns, (dir and patterns), or (dir_pattern).
+        action, patterns, dir, dir_pattern = self._parse_template_line(line)
+
+        # OK, now we know that the action is valid and we have the
+        # right number of words on the line for that action -- so we
+        # can proceed with minimal error-checking.
+        if action == 'include':
+            for pattern in patterns:
+                if not self._include_pattern(pattern, anchor=1):
+                    logging.warning("warning: no files found matching '%s'" %
+                             pattern)
+
+        elif action == 'exclude':
+            for pattern in patterns:
+                if not self.exclude_pattern(pattern, anchor=1):
+                    logging.warning(("warning: no previously-included files "
+                              "found matching '%s'") % pattern)
+
+        elif action == 'global-include':
+            for pattern in patterns:
+                if not self._include_pattern(pattern, anchor=0):
+                    logging.warning(("warning: no files found matching '%s' " +
+                              "anywhere in distribution") % pattern)
+
+        elif action == 'global-exclude':
+            for pattern in patterns:
+                if not self.exclude_pattern(pattern, anchor=0):
+                    logging.warning(("warning: no previously-included files "
+                              "matching '%s' found anywhere in distribution") %
+                             pattern)
+
+        elif action == 'recursive-include':
+            for pattern in patterns:
+                if not self._include_pattern(pattern, prefix=dir):
+                    logging.warning(("warning: no files found matching '%s' "
+                                "under directory '%s'" % (pattern, dir)))
+
+        elif action == 'recursive-exclude':
+            for pattern in patterns:
+                if not self.exclude_pattern(pattern, prefix=dir):
+                    logging.warning(("warning: no previously-included files "
+                              "matching '%s' found under directory '%s'") %
+                             (pattern, dir))
+
+        elif action == 'graft':
+            if not self._include_pattern(None, prefix=dir_pattern):
+                logging.warning("warning: no directories found matching '%s'" %
+                         dir_pattern)
+
+        elif action == 'prune':
+            if not self.exclude_pattern(None, prefix=dir_pattern):
+                logging.warning(("no previously-included directories found " +
+                          "matching '%s'") % dir_pattern)
+        else:
+            raise DistutilsInternalError(
+                  "this cannot happen: invalid action '%s'" % action)
+
+    def _include_pattern(self, pattern, anchor=1, prefix=None, is_regex=0):
+        """Select strings (presumably filenames) from 'self.files' that
+        match 'pattern', a Unix-style wildcard (glob) pattern.
+
+        Patterns are not quite the same as implemented by the 'fnmatch'
+        module: '*' and '?'  match non-special characters, where "special"
+        is platform-dependent: slash on Unix; colon, slash, and backslash on
+        DOS/Windows; and colon on Mac OS.
+
+        If 'anchor' is true (the default), then the pattern match is more
+        stringent: "*.py" will match "foo.py" but not "foo/bar.py".  If
+        'anchor' is false, both of these will match.
+
+        If 'prefix' is supplied, then only filenames starting with 'prefix'
+        (itself a pattern) and ending with 'pattern', with anything in between
+        them, will match.  'anchor' is ignored in this case.
+
+        If 'is_regex' is true, 'anchor' and 'prefix' are ignored, and
+        'pattern' is assumed to be either a string containing a regex or a
+        regex object -- no translation is done, the regex is just compiled
+        and used as-is.
+
+        Selected strings will be added to self.files.
+
+        Return 1 if files are found.
+        """
+        files_found = 0
+        pattern_re = _translate_pattern(pattern, anchor, prefix, is_regex)
+
+        # delayed loading of allfiles list
+        if self.allfiles is None:
+            self.findall()
+
+        for name in self.allfiles:
+            if pattern_re.search(name):
+                self.files.append(name)
+                files_found = 1
+
+        return files_found
+
+
+
+#
+# Utility functions
+#
+
+def _findall(dir=os.curdir):
+    """Find all files under 'dir' and return the list of full filenames
+    (relative to 'dir').
+    """
+    from stat import ST_MODE, S_ISREG, S_ISDIR, S_ISLNK
+
+    list = []
+    stack = [dir]
+    pop = stack.pop
+    push = stack.append
+
+    while stack:
+        dir = pop()
+        names = os.listdir(dir)
+
+        for name in names:
+            if dir != os.curdir:        # avoid the dreaded "./" syndrome
+                fullname = os.path.join(dir, name)
+            else:
+                fullname = name
+
+            # Avoid excess stat calls -- just one will do, thank you!
+            stat = os.stat(fullname)
+            mode = stat[ST_MODE]
+            if S_ISREG(mode):
+                list.append(fullname)
+            elif S_ISDIR(mode) and not S_ISLNK(mode):
+                push(fullname)
+
+    return list
+
+
+
+def _glob_to_re(pattern):
+    """Translate a shell-like glob pattern to a regular expression.
+
+    Return a string containing the regex.  Differs from
+    'fnmatch.translate()' in that '*' does not match "special characters"
+    (which are platform-specific).
+    """
+    pattern_re = fnmatch.translate(pattern)
+
+    # '?' and '*' in the glob pattern become '.' and '.*' in the RE, which
+    # IMHO is wrong -- '?' and '*' aren't supposed to match slash in Unix,
+    # and by extension they shouldn't match such "special characters" under
+    # any OS.  So change all non-escaped dots in the RE to match any
+    # character except the special characters.
+    # XXX currently the "special characters" are just slash -- i.e. this is
+    # Unix-only.
+    pattern_re = re.sub(r'((?<!\\)(\\\\)*)\.', r'\1[^/]', pattern_re)
+
+    return pattern_re
+
+
+def _translate_pattern(pattern, anchor=1, prefix=None, is_regex=0):
+    """Translate a shell-like wildcard pattern to a compiled regular
+    expression.
+
+    Return the compiled regex.  If 'is_regex' true,
+    then 'pattern' is directly compiled to a regex (if it's a string)
+    or just returned as-is (assumes it's a regex object).
+    """
+    if is_regex:
+        if isinstance(pattern, str):
+            return re.compile(pattern)
+        else:
+            return pattern
+
+    if pattern:
+        pattern_re = _glob_to_re(pattern)
+    else:
+        pattern_re = ''
+
+    if prefix is not None:
+        # ditch end of pattern character
+        empty_pattern = _glob_to_re('')
+        prefix_re = _glob_to_re(prefix)[:-len(empty_pattern)]
+        pattern_re = "^" + os.path.join(prefix_re, ".*" + pattern_re)
+    else:                               # no prefix -- respect anchor flag
+        if anchor:
+            pattern_re = "^" + pattern_re
+
+    return re.compile(pattern_re)
diff --git a/src/distutils2/metadata.py b/src/distutils2/metadata.py
--- a/src/distutils2/metadata.py
+++ b/src/distutils2/metadata.py
@@ -13,7 +13,6 @@
 from tokenize import tokenize, NAME, OP, STRING, ENDMARKER
 
 from distutils2.log import warn
-from distutils2.util import rfc822_escape
 from distutils2.version import (is_valid_predicate, is_valid_version,
                                 is_valid_versions)
 from distutils2.errors import (MetadataConflictError,
@@ -25,7 +24,6 @@
     from docutils.parsers.rst import Parser
     from docutils import frontend
     from docutils import nodes
-    from StringIO import StringIO
 
     class SilentReporter(Reporter):
 
diff --git a/src/distutils2/mkpkg.py b/src/distutils2/mkpkg.py
--- a/src/distutils2/mkpkg.py
+++ b/src/distutils2/mkpkg.py
@@ -72,6 +72,8 @@
 ''',
 }
 
+# XXX this list should be asked at PyPI (it changes)
+# then cached, rather than hardcoded
 troveList = [
         'Development Status :: 1 - Planning',
         'Development Status :: 2 - Pre-Alpha',
diff --git a/src/distutils2/tests/__init__.py b/src/distutils2/tests/__init__.py
--- a/src/distutils2/tests/__init__.py
+++ b/src/distutils2/tests/__init__.py
@@ -22,7 +22,6 @@
 
 verbose = 1
 
-
 def test_suite():
     suite = unittest2.TestSuite()
     for fn in os.listdir(here):
@@ -48,9 +47,11 @@
         return result
 
 
-def _run_suite(suite):
+def _run_suite(suite, verbose_=1):
     """Run tests from a unittest2.TestSuite-derived class."""
-    if verbose:
+    global verbose
+    verbose = verbose_
+    if verbose_:
         runner = unittest2.TextTestRunner(sys.stdout, verbosity=2)
     else:
         runner = BasicTestRunner()
@@ -66,7 +67,7 @@
         raise TestFailed(err)
 
 
-def run_unittest(*classes):
+def run_unittest(classes, verbose_=1):
     """Run tests from unittest2.TestCase-derived classes.
 
     Extracted from stdlib test.test_support and modified to support unittest2.
@@ -83,7 +84,7 @@
             suite.addTest(cls)
         else:
             suite.addTest(unittest2.makeSuite(cls))
-    _run_suite(suite)
+    _run_suite(suite, verbose_)
 
 
 def reap_children():
diff --git a/src/distutils2/tests/conversions/01_after.py b/src/distutils2/tests/conversions/01_after.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/conversions/01_after.py
@@ -0,0 +1,4 @@
+from distutils2.core import setup
+
+setup(name='Foo')
+
diff --git a/src/distutils2/tests/conversions/01_before.py b/src/distutils2/tests/conversions/01_before.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/conversions/01_before.py
@@ -0,0 +1,4 @@
+from distutils.core import setup
+
+setup(name='Foo')
+
diff --git a/src/distutils2/tests/conversions/02_after.py b/src/distutils2/tests/conversions/02_after.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/conversions/02_after.py
@@ -0,0 +1,46 @@
+# -*- encoding: utf8 -*-
+import sys
+import os
+from distutils2.core import setup, Extension
+from distutils2.errors import CCompilerError, DistutilsError, CompileError
+from distutils2.command.build_ext import build_ext as distutils_build_ext
+
+VERSION = "0.1"
+
+class build_ext(distutils_build_ext):
+
+    def build_extension(self, ext):
+        try:
+            return distutils_build_ext.build_extension(self, ext)
+        except (CCompilerError, DistutilsError, CompileError), e:
+            pass
+
+def _get_ext_modules():
+    levenshtein = Extension('_levenshtein',
+                            sources=[os.path.join('texttools',
+                                                  '_levenshtein.c')])
+    return [levenshtein]
+
+with open('README.txt') as f:
+    LONG_DESCRIPTION = f.read()
+
+setup(name="TextTools", version=VERSION, author="Tarek Ziade",
+      author_email="tarek at ziade.org",
+      home_page="http://bitbucket.org/tarek/texttools",
+      summary="Text manipulation utilities",
+      description=LONG_DESCRIPTION,
+      keywords="text,guess,levenshtein",
+      classifiers=[
+         'Development Status :: 4 - Beta',
+         'Intended Audience :: Developers',
+         'License :: OSI Approved :: Python Software Foundation License'
+      ],
+      cmdclass={'build_ext': build_ext},
+      packages=['texttools'],
+      package_dir={'texttools': 'texttools'},
+      package_data={'texttools': [os.path.join('samples', '*.txt')]},
+      scripts=[os.path.join('scripts', 'levenshtein.py'),
+               os.path.join('scripts', 'guesslang.py')],
+      ext_modules=_get_ext_modules()
+      )
+
diff --git a/src/distutils2/tests/conversions/02_before.py b/src/distutils2/tests/conversions/02_before.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/conversions/02_before.py
@@ -0,0 +1,46 @@
+# -*- encoding: utf8 -*-
+import sys
+import os
+from distutils.core import setup, Extension
+from distutils.errors import CCompilerError, DistutilsError, CompileError
+from distutils.command.build_ext import build_ext as distutils_build_ext
+
+VERSION = "0.1"
+
+class build_ext(distutils_build_ext):
+
+    def build_extension(self, ext):
+        try:
+            return distutils_build_ext.build_extension(self, ext)
+        except (CCompilerError, DistutilsError, CompileError), e:
+            pass
+
+def _get_ext_modules():
+    levenshtein = Extension('_levenshtein',
+                            sources=[os.path.join('texttools',
+                                                  '_levenshtein.c')])
+    return [levenshtein]
+
+with open('README.txt') as f:
+    LONG_DESCRIPTION = f.read()
+
+setup(name="TextTools", version=VERSION, author="Tarek Ziade",
+      author_email="tarek at ziade.org",
+      url="http://bitbucket.org/tarek/texttools",
+      description="Text manipulation utilities",
+      long_description=LONG_DESCRIPTION,
+      keywords="text,guess,levenshtein",
+      classifiers=[
+         'Development Status :: 4 - Beta',
+         'Intended Audience :: Developers',
+         'License :: OSI Approved :: Python Software Foundation License'
+      ],
+      cmdclass={'build_ext': build_ext},
+      packages=['texttools'],
+      package_dir={'texttools': 'texttools'},
+      package_data={'texttools': [os.path.join('samples', '*.txt')]},
+      scripts=[os.path.join('scripts', 'levenshtein.py'),
+               os.path.join('scripts', 'guesslang.py')],
+      ext_modules=_get_ext_modules()
+      )
+
diff --git a/src/distutils2/tests/conversions/03_after.py b/src/distutils2/tests/conversions/03_after.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/conversions/03_after.py
@@ -0,0 +1,93 @@
+##############################################################################
+#
+# Copyright (c) 2006-2009 Zope Corporation and Contributors.
+# All Rights Reserved.
+#
+# This software is subject to the provisions of the Zope Public License,
+# Version 2.1 (ZPL).  A copy of the ZPL should accompany this distribution.
+# THIS SOFTWARE IS PROVIDED "AS IS" AND ANY AND ALL EXPRESS OR IMPLIED
+# WARRANTIES ARE DISCLAIMED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+# WARRANTIES OF TITLE, MERCHANTABILITY, AGAINST INFRINGEMENT, AND FITNESS
+# FOR A PARTICULAR PURPOSE.
+#
+##############################################################################
+name = "zc.buildout"
+version = "1.5.0dev"
+
+import os
+from distutils2.core import setup
+
+def read(*rnames):
+    return open(os.path.join(os.path.dirname(__file__), *rnames)).read()
+
+long_description=(
+        read('README.txt')
+        + '\n' +
+        'Detailed Documentation\n'
+        '**********************\n'
+        + '\n' +
+        read('src', 'zc', 'buildout', 'buildout.txt')
+        + '\n' +
+        read('src', 'zc', 'buildout', 'unzip.txt')
+        + '\n' +
+        read('src', 'zc', 'buildout', 'repeatable.txt')
+        + '\n' +
+        read('src', 'zc', 'buildout', 'download.txt')
+        + '\n' +
+        read('src', 'zc', 'buildout', 'downloadcache.txt')
+        + '\n' +
+        read('src', 'zc', 'buildout', 'extends-cache.txt')
+        + '\n' +
+        read('src', 'zc', 'buildout', 'setup.txt')
+        + '\n' +
+        read('src', 'zc', 'buildout', 'update.txt')
+        + '\n' +
+        read('src', 'zc', 'buildout', 'debugging.txt')
+        + '\n' +
+        read('src', 'zc', 'buildout', 'testing.txt')
+        + '\n' +
+        read('src', 'zc', 'buildout', 'easy_install.txt')
+        + '\n' +
+        read('src', 'zc', 'buildout', 'distribute.txt')
+        + '\n' +
+        read('CHANGES.txt')
+        + '\n' +
+        'Download\n'
+        '**********************\n'
+        )
+
+entry_points = """
+[console_scripts]
+buildout = %(name)s.buildout:main
+
+[zc.buildout]
+debug = %(name)s.testrecipes:Debug
+
+""" % dict(name=name)
+
+setup(
+    name = name,
+    version = version,
+    author = "Jim Fulton",
+    author_email = "jim at zope.com",
+    summary = "System for managing development buildouts",
+    description=long_description,
+    license = "ZPL 2.1",
+    keywords = "development build",
+    home_page='http://buildout.org',
+
+    data_files = [('.', ['README.txt'])],
+    packages = ['zc', 'zc.buildout'],
+    package_dir = {'': 'src'},
+    namespace_packages = ['zc'],
+    requires_dist = ['setuptools'],
+    include_package_data = True,
+    entry_points = entry_points,
+    zip_safe=False,
+    classifiers = [
+       'Intended Audience :: Developers',
+       'License :: OSI Approved :: Zope Public License',
+       'Topic :: Software Development :: Build Tools',
+       'Topic :: Software Development :: Libraries :: Python Modules',
+       ],
+    )
diff --git a/src/distutils2/tests/conversions/03_before.py b/src/distutils2/tests/conversions/03_before.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/conversions/03_before.py
@@ -0,0 +1,93 @@
+##############################################################################
+#
+# Copyright (c) 2006-2009 Zope Corporation and Contributors.
+# All Rights Reserved.
+#
+# This software is subject to the provisions of the Zope Public License,
+# Version 2.1 (ZPL).  A copy of the ZPL should accompany this distribution.
+# THIS SOFTWARE IS PROVIDED "AS IS" AND ANY AND ALL EXPRESS OR IMPLIED
+# WARRANTIES ARE DISCLAIMED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+# WARRANTIES OF TITLE, MERCHANTABILITY, AGAINST INFRINGEMENT, AND FITNESS
+# FOR A PARTICULAR PURPOSE.
+#
+##############################################################################
+name = "zc.buildout"
+version = "1.5.0dev"
+
+import os
+from setuptools import setup
+
+def read(*rnames):
+    return open(os.path.join(os.path.dirname(__file__), *rnames)).read()
+
+long_description=(
+        read('README.txt')
+        + '\n' +
+        'Detailed Documentation\n'
+        '**********************\n'
+        + '\n' +
+        read('src', 'zc', 'buildout', 'buildout.txt')
+        + '\n' +
+        read('src', 'zc', 'buildout', 'unzip.txt')
+        + '\n' +
+        read('src', 'zc', 'buildout', 'repeatable.txt')
+        + '\n' +
+        read('src', 'zc', 'buildout', 'download.txt')
+        + '\n' +
+        read('src', 'zc', 'buildout', 'downloadcache.txt')
+        + '\n' +
+        read('src', 'zc', 'buildout', 'extends-cache.txt')
+        + '\n' +
+        read('src', 'zc', 'buildout', 'setup.txt')
+        + '\n' +
+        read('src', 'zc', 'buildout', 'update.txt')
+        + '\n' +
+        read('src', 'zc', 'buildout', 'debugging.txt')
+        + '\n' +
+        read('src', 'zc', 'buildout', 'testing.txt')
+        + '\n' +
+        read('src', 'zc', 'buildout', 'easy_install.txt')
+        + '\n' +
+        read('src', 'zc', 'buildout', 'distribute.txt')
+        + '\n' +
+        read('CHANGES.txt')
+        + '\n' +
+        'Download\n'
+        '**********************\n'
+        )
+
+entry_points = """
+[console_scripts]
+buildout = %(name)s.buildout:main
+
+[zc.buildout]
+debug = %(name)s.testrecipes:Debug
+
+""" % dict(name=name)
+
+setup(
+    name = name,
+    version = version,
+    author = "Jim Fulton",
+    author_email = "jim at zope.com",
+    description = "System for managing development buildouts",
+    long_description=long_description,
+    license = "ZPL 2.1",
+    keywords = "development build",
+    url='http://buildout.org',
+
+    data_files = [('.', ['README.txt'])],
+    packages = ['zc', 'zc.buildout'],
+    package_dir = {'': 'src'},
+    namespace_packages = ['zc'],
+    install_requires = 'setuptools',
+    include_package_data = True,
+    entry_points = entry_points,
+    zip_safe=False,
+    classifiers = [
+       'Intended Audience :: Developers',
+       'License :: OSI Approved :: Zope Public License',
+       'Topic :: Software Development :: Build Tools',
+       'Topic :: Software Development :: Libraries :: Python Modules',
+       ],
+    )
diff --git a/src/distutils2/tests/conversions/04_after.py b/src/distutils2/tests/conversions/04_after.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/conversions/04_after.py
@@ -0,0 +1,69 @@
+import sys, os
+try:
+    from distutils2.core import setup
+    kw = {'entry_points':
+          """[console_scripts]\nvirtualenv = virtualenv:main\n""",
+          'zip_safe': False}
+except ImportError:
+    from distutils2.core import setup
+    if sys.platform == 'win32':
+        print('Note: without Setuptools installed you will have to use "python -m virtualenv ENV"')
+    else:
+        kw = {'scripts': ['scripts/virtualenv']}
+import re
+
+here = os.path.dirname(os.path.abspath(__file__))
+
+## Figure out the version from virtualenv.py:
+version_re = re.compile(
+    r'virtualenv_version = "(.*?)"')
+fp = open(os.path.join(here, 'virtualenv.py'))
+version = None
+for line in fp:
+    match = version_re.search(line)
+    if match:
+        version = match.group(1)
+        break
+else:
+    raise Exception("Cannot find version in virtualenv.py")
+fp.close()
+
+## Get long_description from index.txt:
+f = open(os.path.join(here, 'docs', 'index.txt'))
+long_description = f.read().strip()
+long_description = long_description.split('split here', 1)[1]
+f.close()
+
+## A warning just for Ian (related to distribution):
+try:
+    import getpass
+except ImportError:
+    is_ianb = False
+else:
+    is_ianb = getpass.getuser() == 'ianb'
+
+if is_ianb and 'register' in sys.argv:
+    if 'hg tip\n~~~~~~' in long_description:
+        print >> sys.stderr, (
+            "WARNING: hg tip is in index.txt")
+
+setup(name='virtualenv',
+      version=version,
+      summary="Virtual Python Environment builder",
+      description=long_description,
+      classifiers=[
+        'Development Status :: 4 - Beta',
+        'Intended Audience :: Developers',
+        'License :: OSI Approved :: MIT License',
+      ],
+      keywords='setuptools deployment installation distutils',
+      author='Ian Bicking',
+      author_email='ianb at colorstudy.com',
+      home_page='http://virtualenv.openplans.org',
+      license='MIT',
+      use_2to3=True,
+      py_modules=['virtualenv'],
+      packages=['virtualenv_support'],
+      package_data={'virtualenv_support': ['*-py%s.egg' % sys.version[:3], '*.tar.gz']},
+      **kw
+      )
diff --git a/src/distutils2/tests/conversions/04_before.py b/src/distutils2/tests/conversions/04_before.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/conversions/04_before.py
@@ -0,0 +1,69 @@
+import sys, os
+try:
+    from setuptools import setup
+    kw = {'entry_points':
+          """[console_scripts]\nvirtualenv = virtualenv:main\n""",
+          'zip_safe': False}
+except ImportError:
+    from distutils.core import setup
+    if sys.platform == 'win32':
+        print('Note: without Setuptools installed you will have to use "python -m virtualenv ENV"')
+    else:
+        kw = {'scripts': ['scripts/virtualenv']}
+import re
+
+here = os.path.dirname(os.path.abspath(__file__))
+
+## Figure out the version from virtualenv.py:
+version_re = re.compile(
+    r'virtualenv_version = "(.*?)"')
+fp = open(os.path.join(here, 'virtualenv.py'))
+version = None
+for line in fp:
+    match = version_re.search(line)
+    if match:
+        version = match.group(1)
+        break
+else:
+    raise Exception("Cannot find version in virtualenv.py")
+fp.close()
+
+## Get long_description from index.txt:
+f = open(os.path.join(here, 'docs', 'index.txt'))
+long_description = f.read().strip()
+long_description = long_description.split('split here', 1)[1]
+f.close()
+
+## A warning just for Ian (related to distribution):
+try:
+    import getpass
+except ImportError:
+    is_ianb = False
+else:
+    is_ianb = getpass.getuser() == 'ianb'
+
+if is_ianb and 'register' in sys.argv:
+    if 'hg tip\n~~~~~~' in long_description:
+        print >> sys.stderr, (
+            "WARNING: hg tip is in index.txt")
+
+setup(name='virtualenv',
+      version=version,
+      description="Virtual Python Environment builder",
+      long_description=long_description,
+      classifiers=[
+        'Development Status :: 4 - Beta',
+        'Intended Audience :: Developers',
+        'License :: OSI Approved :: MIT License',
+      ],
+      keywords='setuptools deployment installation distutils',
+      author='Ian Bicking',
+      author_email='ianb at colorstudy.com',
+      url='http://virtualenv.openplans.org',
+      license='MIT',
+      use_2to3=True,
+      py_modules=['virtualenv'],
+      packages=['virtualenv_support'],
+      package_data={'virtualenv_support': ['*-py%s.egg' % sys.version[:3], '*.tar.gz']},
+      **kw
+      )
diff --git a/src/distutils2/tests/test_bdist.py b/src/distutils2/tests/test_bdist.py
--- a/src/distutils2/tests/test_bdist.py
+++ b/src/distutils2/tests/test_bdist.py
@@ -31,8 +31,7 @@
         # XXX an explicit list in bdist is
         # not the best way to  bdist_* commands
         # we should add a registry
-        formats = ['rpm', 'zip', 'gztar', 'bztar', 'ztar',
-                   'tar', 'wininst', 'msi']
+        formats = ['zip', 'gztar', 'bztar', 'ztar', 'tar', 'wininst', 'msi']
         formats.sort()
         founded = cmd.format_command.keys()
         founded.sort()
diff --git a/src/distutils2/tests/test_converter.py b/src/distutils2/tests/test_converter.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/test_converter.py
@@ -0,0 +1,39 @@
+"""Tests for distutils.converter."""
+import os
+import sys
+import unittest2
+from distutils2.converter import DistutilsRefactoringTool
+
+_CURDIR = os.path.dirname(__file__)
+
+def _read_file(path):
+    # yes, distutils2 is 2.4 compatible, so, no with...
+    f = open(path)
+    try:
+        return f.read()
+    finally:
+        f.close()
+
+
+class ConverterTestCase(unittest2.TestCase):
+
+    @unittest2.skipUnless(not sys.version < '2.6', 'Needs Python >=2.6')
+    def test_conversions(self):
+        # for all XX_before in the conversions/ dir
+        # we run the refactoring tool
+        ref = DistutilsRefactoringTool()
+        convdir = os.path.join(_CURDIR, 'conversions')
+        for file_ in os.listdir(convdir):
+            if 'after' in file_ or not file_.endswith('py'):
+                continue
+            original = _read_file(os.path.join(convdir, file_))
+            wanted = file_.replace('before', 'after')
+            wanted = _read_file(os.path.join(convdir, wanted))
+            res = ref.refactor_string(original, 'setup.py')
+            self.assertEquals(str(res), wanted)
+
+def test_suite():
+    return unittest2.makeSuite(ConverterTestCase)
+
+if __name__ == '__main__':
+    unittest2.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_core.py b/src/distutils2/tests/test_core.py
--- a/src/distutils2/tests/test_core.py
+++ b/src/distutils2/tests/test_core.py
@@ -60,6 +60,19 @@
         distutils2.core.run_setup(
             self.write_setup(setup_using___file__))
 
+    def test_run_setup_stop_after(self):
+        f = self.write_setup(setup_using___file__)
+        for s in ['init', 'config', 'commandline', 'run']:
+            distutils2.core.run_setup(f, stop_after=s)
+        self.assertRaises(ValueError, distutils2.core.run_setup, 
+                          f, stop_after='bob')
+
+    def test_run_setup_args(self):
+        f = self.write_setup(setup_using___file__)
+        d = distutils2.core.run_setup(f, script_args=["--help"], 
+                                        stop_after="init")
+        self.assertEqual(['--help'], d.script_args)
+
     def test_run_setup_uses_current_dir(self):
         # This tests that the setup script is run with the current directory
         # as its own current directory; this was temporarily broken by a
diff --git a/src/distutils2/tests/test_filelist.py b/src/distutils2/tests/test_filelist.py
deleted file mode 100644
--- a/src/distutils2/tests/test_filelist.py
+++ /dev/null
@@ -1,69 +0,0 @@
-"""Tests for distutils.filelist."""
-from os.path import join
-import unittest2
-from distutils2.tests import captured_stdout
-
-from distutils2.filelist import glob_to_re, FileList
-
-MANIFEST_IN = """\
-include ok
-include xo
-exclude xo
-include foo.tmp
-global-include *.x
-global-include *.txt
-global-exclude *.tmp
-recursive-include f *.oo
-recursive-exclude global *.x
-graft dir
-prune dir3
-"""
-
-class FileListTestCase(unittest2.TestCase):
-
-    # this only works on 2.7
-    def _test_glob_to_re(self):
-        # simple cases
-        self.assertEquals(glob_to_re('foo*'), 'foo[^/]*\\Z(?ms)')
-        self.assertEquals(glob_to_re('foo?'), 'foo[^/]\\Z(?ms)')
-        self.assertEquals(glob_to_re('foo??'), 'foo[^/][^/]\\Z(?ms)')
-
-        # special cases
-        self.assertEquals(glob_to_re(r'foo\\*'), r'foo\\\\[^/]*\Z(?ms)')
-        self.assertEquals(glob_to_re(r'foo\\\*'), r'foo\\\\\\[^/]*\Z(?ms)')
-        self.assertEquals(glob_to_re('foo????'), r'foo[^/][^/][^/][^/]\Z(?ms)')
-        self.assertEquals(glob_to_re(r'foo\\??'), r'foo\\\\[^/][^/]\Z(?ms)')
-
-    def test_process_template_line(self):
-        # testing  all MANIFEST.in template patterns
-        file_list = FileList()
-
-        # simulated file list
-        file_list.allfiles = ['foo.tmp', 'ok', 'xo', 'four.txt',
-                              join('global', 'one.txt'),
-                              join('global', 'two.txt'),
-                              join('global', 'files.x'),
-                              join('global', 'here.tmp'),
-                              join('f', 'o', 'f.oo'),
-                              join('dir', 'graft-one'),
-                              join('dir', 'dir2', 'graft2'),
-                              join('dir3', 'ok'),
-                              join('dir3', 'sub', 'ok.txt')
-                              ]
-
-        for line in MANIFEST_IN.split('\n'):
-            if line.strip() == '':
-                continue
-            file_list.process_template_line(line)
-
-        wanted = ['ok', 'four.txt', join('global', 'one.txt'),
-                  join('global', 'two.txt'), join('f', 'o', 'f.oo'),
-                  join('dir', 'graft-one'), join('dir', 'dir2', 'graft2')]
-
-        self.assertEquals(file_list.files, wanted)
-
-def test_suite():
-    return unittest2.makeSuite(FileListTestCase)
-
-if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_manifest.py b/src/distutils2/tests/test_manifest.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/test_manifest.py
@@ -0,0 +1,57 @@
+"""Tests for distutils.manifest."""
+import unittest2
+import os
+import sys
+import logging
+
+from distutils2.tests import support
+from distutils2.manifest import Manifest
+
+_MANIFEST = """\
+recursive-include foo *.py   # ok
+# nothing here
+
+#
+
+recursive-include bar \\
+  *.dat   *.txt
+"""
+
+class ManifestTestCase(support.TempdirManager,
+                       unittest2.TestCase):
+
+    def test_manifest_reader(self):
+
+        tmpdir = self.mkdtemp()
+        MANIFEST = os.path.join(tmpdir, 'MANIFEST.in')
+        f = open(MANIFEST, 'w')
+        try:
+            f.write(_MANIFEST)
+        finally:
+            f.close()
+        manifest = Manifest()
+
+        warns = []
+        def _warn(msg):
+            warns.append(msg)
+
+        old_warn = logging.warning
+        logging.warning = _warn
+        try:
+            manifest.read_template(MANIFEST)
+        finally:
+            logging.warning = old_warn
+
+        # the manifest should have been read
+        # and 3 warnings issued (we ddidn't provided the files)
+        self.assertEquals(len(warns), 3)
+        for warn in warns:
+            self.assertIn('warning: no files found matching', warn)
+
+
+
+def test_suite():
+    return unittest2.makeSuite(ManifestTestCase)
+
+if __name__ == '__main__':
+    run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_sdist.py b/src/distutils2/tests/test_sdist.py
--- a/src/distutils2/tests/test_sdist.py
+++ b/src/distutils2/tests/test_sdist.py
@@ -257,6 +257,9 @@
         cmd.metadata_check = 0
         cmd.run()
         warnings = self.get_logs(WARN)
+        # removing manifest generated warnings
+        warnings = [warn for warn in warnings if
+                    not warn.endswith('-- skipping')]
         self.assertEquals(len(warnings), 0)
 
 
@@ -340,6 +343,19 @@
         finally:
             archive.close()
 
+    def test_get_file_list(self):
+        dist, cmd = self.get_cmd()
+        cmd.finalize_options()
+        cmd.template = os.path.join(self.tmp_dir, 'MANIFEST.in')
+        f = open(cmd.template, 'w')
+        try:
+            f.write('include MANIFEST.in\n')
+        finally:
+            f.close()
+
+        cmd.get_file_list()
+        self.assertIn('MANIFEST.in', cmd.filelist.files)
+
 def test_suite():
     return unittest2.makeSuite(SDistTestCase)
 
diff --git a/src/distutils2/tests/test_text_file.py b/src/distutils2/tests/test_text_file.py
deleted file mode 100644
--- a/src/distutils2/tests/test_text_file.py
+++ /dev/null
@@ -1,88 +0,0 @@
-"""Tests for distutils.text_file."""
-import os
-import unittest2
-from distutils2.text_file import TextFile
-from distutils2.tests import support
-
-TEST_DATA = """# test file
-
-line 3 \\
-# intervening comment
-  continues on next line
-"""
-
-class TextFileTestCase(support.TempdirManager, unittest2.TestCase):
-
-    def test_class(self):
-        # old tests moved from text_file.__main__
-        # so they are really called by the buildbots
-
-        # result 1: no fancy options
-        result1 = ['# test file\n', '\n', 'line 3 \\\n',
-                   '# intervening comment\n',
-                   '  continues on next line\n']
-
-        # result 2: just strip comments
-        result2 = ["\n",
-                   "line 3 \\\n",
-                   "  continues on next line\n"]
-
-        # result 3: just strip blank lines
-        result3 = ["# test file\n",
-                   "line 3 \\\n",
-                   "# intervening comment\n",
-                   "  continues on next line\n"]
-
-        # result 4: default, strip comments, blank lines,
-        # and trailing whitespace
-        result4 = ["line 3 \\",
-                   "  continues on next line"]
-
-        # result 5: strip comments and blanks, plus join lines (but don't
-        # "collapse" joined lines
-        result5 = ["line 3   continues on next line"]
-
-        # result 6: strip comments and blanks, plus join lines (and
-        # "collapse" joined lines
-        result6 = ["line 3 continues on next line"]
-
-        def test_input(count, description, file, expected_result):
-            result = file.readlines()
-            self.assertEquals(result, expected_result)
-
-        tmpdir = self.mkdtemp()
-        filename = os.path.join(tmpdir, "test.txt")
-        out_file = open(filename, "w")
-        try:
-            out_file.write(TEST_DATA)
-        finally:
-            out_file.close()
-
-        in_file = TextFile (filename, strip_comments=0, skip_blanks=0,
-                            lstrip_ws=0, rstrip_ws=0)
-        test_input (1, "no processing", in_file, result1)
-
-        in_file = TextFile (filename, strip_comments=1, skip_blanks=0,
-                            lstrip_ws=0, rstrip_ws=0)
-        test_input (2, "strip comments", in_file, result2)
-
-        in_file = TextFile (filename, strip_comments=0, skip_blanks=1,
-                            lstrip_ws=0, rstrip_ws=0)
-        test_input (3, "strip blanks", in_file, result3)
-
-        in_file = TextFile (filename)
-        test_input (4, "default processing", in_file, result4)
-
-        in_file = TextFile (filename, strip_comments=1, skip_blanks=1,
-                            join_lines=1, rstrip_ws=1)
-        test_input (5, "join lines without collapsing", in_file, result5)
-
-        in_file = TextFile (filename, strip_comments=1, skip_blanks=1,
-                            join_lines=1, rstrip_ws=1, collapse_join=1)
-        test_input (6, "join lines with collapsing", in_file, result6)
-
-def test_suite():
-    return unittest2.makeSuite(TextFileTestCase)
-
-if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_util.py b/src/distutils2/tests/test_util.py
--- a/src/distutils2/tests/test_util.py
+++ b/src/distutils2/tests/test_util.py
@@ -294,7 +294,7 @@
         self.write_file(os.path.join(pkg5, '__init__.py'))
 
         res = find_packages([root], ['pkg1.pkg2'])
-        self.assertEquals(res, ['pkg1', 'pkg5', 'pkg1.pkg3', 'pkg1.pkg3.pkg6'])
+        self.assertEquals(set(res), set(['pkg1', 'pkg5', 'pkg1.pkg3', 'pkg1.pkg3.pkg6']))
 
 
 def test_suite():
diff --git a/src/distutils2/tests/test_version.py b/src/distutils2/tests/test_version.py
--- a/src/distutils2/tests/test_version.py
+++ b/src/distutils2/tests/test_version.py
@@ -139,13 +139,17 @@
         for predicate in predicates:
             v = VersionPredicate(predicate)
 
-        assert VersionPredicate('Hey (>=2.5,<2.7)').match('2.6')
-        assert VersionPredicate('Ho').match('2.6')
-        assert not VersionPredicate('Hey (>=2.5,!=2.6,<2.7)').match('2.6')
-        assert VersionPredicate('Ho (<3.0)').match('2.6')
-        assert VersionPredicate('Ho (<3.0,!=2.5)').match('2.6.0')
-        assert not VersionPredicate('Ho (<3.0,!=2.6)').match('2.6.0')
-
+        self.assertTrue(VersionPredicate('Hey (>=2.5,<2.7)').match('2.6'))
+        self.assertTrue(VersionPredicate('Ho').match('2.6'))
+        self.assertFalse(VersionPredicate('Hey (>=2.5,!=2.6,<2.7)').match('2.6'))
+        self.assertTrue(VersionPredicate('Ho (<3.0)').match('2.6'))
+        self.assertTrue(VersionPredicate('Ho (<3.0,!=2.5)').match('2.6.0'))
+        self.assertFalse(VersionPredicate('Ho (<3.0,!=2.6)').match('2.6.0'))
+        self.assertTrue(VersionPredicate('Ho (2.5)').match('2.5.4'))
+        self.assertFalse(VersionPredicate('Ho (!=2.5)').match('2.5.2'))
+        self.assertTrue(VersionPredicate('Hey (<=2.5)').match('2.5.9'))
+        self.assertFalse(VersionPredicate('Hey (<=2.5)').match('2.6.0'))
+        self.assertTrue(VersionPredicate('Hey (>=2.5)').match('2.5.1'))
 
         # XXX need to silent the micro version in this case
         #assert not VersionPredicate('Ho (<3.0,!=2.6)').match('2.6.3')
diff --git a/src/distutils2/text_file.py b/src/distutils2/text_file.py
deleted file mode 100644
--- a/src/distutils2/text_file.py
+++ /dev/null
@@ -1,304 +0,0 @@
-"""text_file
-
-provides the TextFile class, which gives an interface to text files
-that (optionally) takes care of stripping comments, ignoring blank
-lines, and joining lines with backslashes."""
-
-__revision__ = "$Id: text_file.py 76956 2009-12-21 01:22:46Z tarek.ziade $"
-
-import sys
-
-
-class TextFile:
-
-    """Provides a file-like object that takes care of all the things you
-       commonly want to do when processing a text file that has some
-       line-by-line syntax: strip comments (as long as "#" is your
-       comment character), skip blank lines, join adjacent lines by
-       escaping the newline (ie. backslash at end of line), strip
-       leading and/or trailing whitespace.  All of these are optional
-       and independently controllable.
-
-       Provides a 'warn()' method so you can generate warning messages that
-       report physical line number, even if the logical line in question
-       spans multiple physical lines.  Also provides 'unreadline()' for
-       implementing line-at-a-time lookahead.
-
-       Constructor is called as:
-
-           TextFile (filename=None, file=None, **options)
-
-       It bombs (RuntimeError) if both 'filename' and 'file' are None;
-       'filename' should be a string, and 'file' a file object (or
-       something that provides 'readline()' and 'close()' methods).  It is
-       recommended that you supply at least 'filename', so that TextFile
-       can include it in warning messages.  If 'file' is not supplied,
-       TextFile creates its own using the 'open()' builtin.
-
-       The options are all boolean, and affect the value returned by
-       'readline()':
-         strip_comments [default: true]
-           strip from "#" to end-of-line, as well as any whitespace
-           leading up to the "#" -- unless it is escaped by a backslash
-         lstrip_ws [default: false]
-           strip leading whitespace from each line before returning it
-         rstrip_ws [default: true]
-           strip trailing whitespace (including line terminator!) from
-           each line before returning it
-         skip_blanks [default: true}
-           skip lines that are empty *after* stripping comments and
-           whitespace.  (If both lstrip_ws and rstrip_ws are false,
-           then some lines may consist of solely whitespace: these will
-           *not* be skipped, even if 'skip_blanks' is true.)
-         join_lines [default: false]
-           if a backslash is the last non-newline character on a line
-           after stripping comments and whitespace, join the following line
-           to it to form one "logical line"; if N consecutive lines end
-           with a backslash, then N+1 physical lines will be joined to
-           form one logical line.
-         collapse_join [default: false]
-           strip leading whitespace from lines that are joined to their
-           predecessor; only matters if (join_lines and not lstrip_ws)
-
-       Note that since 'rstrip_ws' can strip the trailing newline, the
-       semantics of 'readline()' must differ from those of the builtin file
-       object's 'readline()' method!  In particular, 'readline()' returns
-       None for end-of-file: an empty string might just be a blank line (or
-       an all-whitespace line), if 'rstrip_ws' is true but 'skip_blanks' is
-       not."""
-
-    default_options = { 'strip_comments': 1,
-                        'skip_blanks':    1,
-                        'lstrip_ws':      0,
-                        'rstrip_ws':      1,
-                        'join_lines':     0,
-                        'collapse_join':  0,
-                      }
-
-    def __init__ (self, filename=None, file=None, **options):
-        """Construct a new TextFile object.  At least one of 'filename'
-           (a string) and 'file' (a file-like object) must be supplied.
-           They keyword argument options are described above and affect
-           the values returned by 'readline()'."""
-
-        if filename is None and file is None:
-            raise RuntimeError, \
-                  "you must supply either or both of 'filename' and 'file'"
-
-        # set values for all options -- either from client option hash
-        # or fallback to default_options
-        for opt in self.default_options.keys():
-            if opt in options:
-                setattr (self, opt, options[opt])
-
-            else:
-                setattr (self, opt, self.default_options[opt])
-
-        # sanity check client option hash
-        for opt in options.keys():
-            if opt not in self.default_options:
-                raise KeyError, "invalid TextFile option '%s'" % opt
-
-        if file is None:
-            self.open (filename)
-        else:
-            self.filename = filename
-            self.file = file
-            self.current_line = 0       # assuming that file is at BOF!
-
-        # 'linebuf' is a stack of lines that will be emptied before we
-        # actually read from the file; it's only populated by an
-        # 'unreadline()' operation
-        self.linebuf = []
-
-
-    def open (self, filename):
-        """Open a new file named 'filename'.  This overrides both the
-           'filename' and 'file' arguments to the constructor."""
-
-        self.filename = filename
-        self.file = open (self.filename, 'r')
-        self.current_line = 0
-
-
-    def close (self):
-        """Close the current file and forget everything we know about it
-           (filename, current line number)."""
-
-        self.file.close ()
-        self.file = None
-        self.filename = None
-        self.current_line = None
-
-
-    def gen_error (self, msg, line=None):
-        outmsg = []
-        if line is None:
-            line = self.current_line
-        outmsg.append(self.filename + ", ")
-        if isinstance(line, (list, tuple)):
-            outmsg.append("lines %d-%d: " % tuple (line))
-        else:
-            outmsg.append("line %d: " % line)
-        outmsg.append(str(msg))
-        return ''.join(outmsg)
-
-
-    def error (self, msg, line=None):
-        raise ValueError, "error: " + self.gen_error(msg, line)
-
-    def warn (self, msg, line=None):
-        """Print (to stderr) a warning message tied to the current logical
-           line in the current file.  If the current logical line in the
-           file spans multiple physical lines, the warning refers to the
-           whole range, eg. "lines 3-5".  If 'line' supplied, it overrides
-           the current line number; it may be a list or tuple to indicate a
-           range of physical lines, or an integer for a single physical
-           line."""
-        sys.stderr.write("warning: " + self.gen_error(msg, line) + "\n")
-
-
-    def readline (self):
-        """Read and return a single logical line from the current file (or
-           from an internal buffer if lines have previously been "unread"
-           with 'unreadline()').  If the 'join_lines' option is true, this
-           may involve reading multiple physical lines concatenated into a
-           single string.  Updates the current line number, so calling
-           'warn()' after 'readline()' emits a warning about the physical
-           line(s) just read.  Returns None on end-of-file, since the empty
-           string can occur if 'rstrip_ws' is true but 'strip_blanks' is
-           not."""
-
-        # If any "unread" lines waiting in 'linebuf', return the top
-        # one.  (We don't actually buffer read-ahead data -- lines only
-        # get put in 'linebuf' if the client explicitly does an
-        # 'unreadline()'.
-        if self.linebuf:
-            line = self.linebuf[-1]
-            del self.linebuf[-1]
-            return line
-
-        buildup_line = ''
-
-        while 1:
-            # read the line, make it None if EOF
-            line = self.file.readline()
-            if line == '': line = None
-
-            if self.strip_comments and line:
-
-                # Look for the first "#" in the line.  If none, never
-                # mind.  If we find one and it's the first character, or
-                # is not preceded by "\", then it starts a comment --
-                # strip the comment, strip whitespace before it, and
-                # carry on.  Otherwise, it's just an escaped "#", so
-                # unescape it (and any other escaped "#"'s that might be
-                # lurking in there) and otherwise leave the line alone.
-
-                pos = line.find("#")
-                if pos == -1:           # no "#" -- no comments
-                    pass
-
-                # It's definitely a comment -- either "#" is the first
-                # character, or it's elsewhere and unescaped.
-                elif pos == 0 or line[pos-1] != "\\":
-                    # Have to preserve the trailing newline, because it's
-                    # the job of a later step (rstrip_ws) to remove it --
-                    # and if rstrip_ws is false, we'd better preserve it!
-                    # (NB. this means that if the final line is all comment
-                    # and has no trailing newline, we will think that it's
-                    # EOF; I think that's OK.)
-                    eol = (line[-1] == '\n') and '\n' or ''
-                    line = line[0:pos] + eol
-
-                    # If all that's left is whitespace, then skip line
-                    # *now*, before we try to join it to 'buildup_line' --
-                    # that way constructs like
-                    #   hello \\
-                    #   # comment that should be ignored
-                    #   there
-                    # result in "hello there".
-                    if line.strip() == "":
-                        continue
-
-                else:                   # it's an escaped "#"
-                    line = line.replace("\\#", "#")
-
-
-            # did previous line end with a backslash? then accumulate
-            if self.join_lines and buildup_line:
-                # oops: end of file
-                if line is None:
-                    self.warn ("continuation line immediately precedes "
-                               "end-of-file")
-                    return buildup_line
-
-                if self.collapse_join:
-                    line = line.lstrip()
-                line = buildup_line + line
-
-                # careful: pay attention to line number when incrementing it
-                if isinstance(self.current_line, list):
-                    self.current_line[1] = self.current_line[1] + 1
-                else:
-                    self.current_line = [self.current_line,
-                                         self.current_line+1]
-            # just an ordinary line, read it as usual
-            else:
-                if line is None:        # eof
-                    return None
-
-                # still have to be careful about incrementing the line number!
-                if isinstance(self.current_line, list):
-                    self.current_line = self.current_line[1] + 1
-                else:
-                    self.current_line = self.current_line + 1
-
-
-            # strip whitespace however the client wants (leading and
-            # trailing, or one or the other, or neither)
-            if self.lstrip_ws and self.rstrip_ws:
-                line = line.strip()
-            elif self.lstrip_ws:
-                line = line.lstrip()
-            elif self.rstrip_ws:
-                line = line.rstrip()
-
-            # blank line (whether we rstrip'ed or not)? skip to next line
-            # if appropriate
-            if (line == '' or line == '\n') and self.skip_blanks:
-                continue
-
-            if self.join_lines:
-                if line[-1] == '\\':
-                    buildup_line = line[:-1]
-                    continue
-
-                if line[-2:] == '\\\n':
-                    buildup_line = line[0:-2] + '\n'
-                    continue
-
-            # well, I guess there's some actual content there: return it
-            return line
-
-    # readline ()
-
-
-    def readlines (self):
-        """Read and return the list of all logical lines remaining in the
-           current file."""
-
-        lines = []
-        while 1:
-            line = self.readline()
-            if line is None:
-                return lines
-            lines.append (line)
-
-
-    def unreadline (self, line):
-        """Push 'line' (a string) onto an internal buffer that will be
-           checked by future 'readline()' calls.  Handy for implementing
-           a parser with line-at-a-time lookahead."""
-
-        self.linebuf.append (line)
diff --git a/src/distutils2/version.py b/src/distutils2/version.py
--- a/src/distutils2/version.py
+++ b/src/distutils2/version.py
@@ -330,10 +330,11 @@
 
     _operators = {"<": lambda x, y: x < y,
                   ">": lambda x, y: x > y,
-                  "<=": lambda x, y: x <= y,
-                  ">=": lambda x, y: x >= y,
-                  "==": lambda x, y: x == y,
-                  "!=": lambda x, y: x != y}
+                  "<=": lambda x, y: str(x).startswith(str(y)) or x < y,
+                  ">=": lambda x, y: str(x).startswith(str(y)) or x > y,
+                  "==": lambda x, y: str(x).startswith(str(y)),
+                  "!=": lambda x, y: not str(x).startswith(str(y)),
+                  }
 
     def __init__(self, predicate):
         predicate = predicate.strip()
diff --git a/src/runtests.py b/src/runtests.py
--- a/src/runtests.py
+++ b/src/runtests.py
@@ -2,17 +2,34 @@
 
 The tests for distutils2 are defined in the distutils2.tests package;
 """
-
-import distutils2.tests
-from distutils2.tests import run_unittest, reap_children
-
-from distutils2._backport.tests import test_suite as btest_suite
+import sys
 
 def test_main():
-    run_unittest(distutils2.tests.test_suite())
-    run_unittest(btest_suite())
-    reap_children()
-
+    import distutils2.tests
+    from distutils2.tests import run_unittest, reap_children, TestFailed
+    from distutils2._backport.tests import test_suite as btest_suite
+    # just supporting -q right now
+    # to enable detailed/quiet output
+    if len(sys.argv) > 1:
+        verbose = sys.argv[-1] != '-q'
+    else:
+        verbose = 1
+    try:
+        try:
+            run_unittest([distutils2.tests.test_suite(), btest_suite()],
+                    verbose_=verbose)
+            return 0
+        except TestFailed:
+            return 1
+    finally:
+        reap_children()
 
 if __name__ == "__main__":
-    test_main()
+    try:
+        import unittest2
+    except ImportError:
+        print('!!! You need to install unittest2')
+        sys.exit(1)
+
+    sys.exit(test_main())
+
diff --git a/src/tests.sh b/src/tests.sh
new file mode 100755
--- /dev/null
+++ b/src/tests.sh
@@ -0,0 +1,28 @@
+#!/bin/sh
+echo -n "Running tests for Python 2.4..."
+python2.4 runtests.py -q > /dev/null 2> /dev/null
+if [ $? -ne 0 ];then
+    echo "Failed"
+    exit $1
+else
+    echo "Success"
+fi
+
+echo -n "Running tests for Python 2.5..."
+python2.5 runtests.py -q > /dev/null 2> /dev/null
+if [ $? -ne 0 ];then
+    echo "Failed"
+    exit $1
+else
+    echo "Success"
+fi
+
+echo -n "Running tests for Python 2.6..."
+python2.6 runtests.py -q > /dev/null 2> /dev/null
+if [ $? -ne 0 ];then
+    echo "Failed"
+    exit $1
+else
+    echo "Success"
+fi
+

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: merge upstream
Message-ID: <E1OVLoV-0000i7-Cg@dinsdale.python.org>

tarek.ziade pushed 50b239007c3b to distutils2:

http://hg.python.org/distutils2/rev/50b239007c3b
changeset:   289:50b239007c3b
parent:      288:a8087e4ed26c
parent:      188:b7fd258b33e7
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Tue Jun 01 09:28:45 2010 +0200
summary:     merge upstream
files:       src/distutils2/dist.py

diff --git a/src/CONTRIBUTORS.txt b/src/CONTRIBUTORS.txt
new file mode 100644
--- /dev/null
+++ b/src/CONTRIBUTORS.txt
@@ -0,0 +1,25 @@
+============
+Contributors
+============
+
+Distutils2 is a project that was started and that is maintained by
+Tarek Ziad??, and many people are contributing to the project.
+
+If you did, please add your name below in alphabetical order !
+
+Thanks to:
+
+- Pior Bastida
+- Titus Brown
+- Nicolas Cadou
+- Josip Djolonga
+- Yannick Gringas
+- Carl Meyer
+- Michael Mulich
+- George Peris
+- Sean Reifschneider
+- Erik Rose
+- Brian Rosner
+- Alexandre Vassalotti
+- Martin von L??wis
+
diff --git a/src/DEVNOTES.txt b/src/DEVNOTES.txt
new file mode 100644
--- /dev/null
+++ b/src/DEVNOTES.txt
@@ -0,0 +1,10 @@
+Notes for developers
+====================
+
+- Distutils2 runs from 2.4 to 3.2 (3.x not implemented yet), so
+  make sure you don't use a syntax that doesn't work under
+  a specific Python version.
+
+- Always run tests.sh before you push a change. This implies
+  that you have all Python versions installed.
+
diff --git a/src/distutils2/_backport/pkgutil.py b/src/distutils2/_backport/pkgutil.py
--- a/src/distutils2/_backport/pkgutil.py
+++ b/src/distutils2/_backport/pkgutil.py
@@ -11,16 +11,18 @@
 from types import ModuleType
 from distutils2.errors import DistutilsError
 from distutils2.metadata import DistributionMetadata
-from distutils2.version import suggest_normalized_version
+from distutils2.version import suggest_normalized_version, VersionPredicate
 
 __all__ = [
     'get_importer', 'iter_importers', 'get_loader', 'find_loader',
     'walk_packages', 'iter_modules',
     'ImpImporter', 'ImpLoader', 'read_code', 'extend_path',
-    'Distribution', 'distinfo_dirname', 'get_distributions',
-    'get_distribution', 'get_file_users', 
+    'Distribution', 'EggInfoDistribution', 'distinfo_dirname',
+    'get_distributions', 'get_distribution', 'get_file_users',
+    'provides_distribution', 'obsoletes_distribution',
 ]
 
+
 def read_code(stream):
     # This helper is needed in order for the PEP 302 emulation to
     # correctly handle compiled files
@@ -37,6 +39,7 @@
 def simplegeneric(func):
     """Make a trivial single-dispatch generic function"""
     registry = {}
+
     def wrapper(*args, **kw):
         ob = args[0]
         try:
@@ -47,6 +50,7 @@
             mro = cls.__mro__
         except AttributeError:
             try:
+
                 class cls(cls, object):
                     pass
                 mro = cls.__mro__[1:]
@@ -128,7 +132,7 @@
                 # don't traverse path items we've seen before
                 path = [p for p in path if not seen(p)]
 
-                for item in walk_packages(path, name+'.', onerror):
+                for item in walk_packages(path, name + '.', onerror):
                     yield item
 
 
@@ -206,7 +210,7 @@
 
         for fn in filenames:
             modname = inspect.getmodulename(fn)
-            if modname=='__init__' or modname in yielded:
+            if modname == '__init__' or modname in yielded:
                 continue
 
             path = os.path.join(self.path, fn)
@@ -216,7 +220,7 @@
                 modname = fn
                 for fn in os.listdir(path):
                     subname = inspect.getmodulename(fn)
-                    if subname=='__init__':
+                    if subname == '__init__':
                         ispkg = True
                         break
                 else:
@@ -255,7 +259,7 @@
     def _reopen(self):
         if self.file and self.file.closed:
             mod_type = self.etc[2]
-            if mod_type==imp.PY_SOURCE:
+            if mod_type == imp.PY_SOURCE:
                 self.file = open(self.filename, 'rU')
             elif mod_type in (imp.PY_COMPILED, imp.C_EXTENSION):
                 self.file = open(self.filename, 'rb')
@@ -270,22 +274,22 @@
 
     def is_package(self, fullname):
         fullname = self._fix_name(fullname)
-        return self.etc[2]==imp.PKG_DIRECTORY
+        return self.etc[2] == imp.PKG_DIRECTORY
 
     def get_code(self, fullname=None):
         fullname = self._fix_name(fullname)
         if self.code is None:
             mod_type = self.etc[2]
-            if mod_type==imp.PY_SOURCE:
+            if mod_type == imp.PY_SOURCE:
                 source = self.get_source(fullname)
                 self.code = compile(source, self.filename, 'exec')
-            elif mod_type==imp.PY_COMPILED:
+            elif mod_type == imp.PY_COMPILED:
                 self._reopen()
                 try:
                     self.code = read_code(self.file)
                 finally:
                     self.file.close()
-            elif mod_type==imp.PKG_DIRECTORY:
+            elif mod_type == imp.PKG_DIRECTORY:
                 self.code = self._get_delegate().get_code()
         return self.code
 
@@ -293,29 +297,28 @@
         fullname = self._fix_name(fullname)
         if self.source is None:
             mod_type = self.etc[2]
-            if mod_type==imp.PY_SOURCE:
+            if mod_type == imp.PY_SOURCE:
                 self._reopen()
                 try:
                     self.source = self.file.read()
                 finally:
                     self.file.close()
-            elif mod_type==imp.PY_COMPILED:
+            elif mod_type == imp.PY_COMPILED:
                 if os.path.exists(self.filename[:-1]):
                     f = open(self.filename[:-1], 'rU')
                     self.source = f.read()
                     f.close()
-            elif mod_type==imp.PKG_DIRECTORY:
+            elif mod_type == imp.PKG_DIRECTORY:
                 self.source = self._get_delegate().get_source()
         return self.source
 
-
     def _get_delegate(self):
         return ImpImporter(self.filename).find_module('__init__')
 
     def get_filename(self, fullname=None):
         fullname = self._fix_name(fullname)
         mod_type = self.etc[2]
-        if self.etc[2]==imp.PKG_DIRECTORY:
+        if self.etc[2] == imp.PKG_DIRECTORY:
             return self._get_delegate().get_filename()
         elif self.etc[2] in (imp.PY_SOURCE, imp.PY_COMPILED, imp.C_EXTENSION):
             return self.filename
@@ -339,16 +342,16 @@
 
             fn = fn[plen:].split(os.sep)
 
-            if len(fn)==2 and fn[1].startswith('__init__.py'):
+            if len(fn) == 2 and fn[1].startswith('__init__.py'):
                 if fn[0] not in yielded:
                     yielded[fn[0]] = 1
                     yield fn[0], True
 
-            if len(fn)!=1:
+            if len(fn) != 1:
                 continue
 
             modname = inspect.getmodulename(fn[0])
-            if modname=='__init__':
+            if modname == '__init__':
                 continue
 
             if modname and '.' not in modname and modname not in yielded:
@@ -436,6 +439,7 @@
     if '.' not in fullname:
         yield ImpImporter()
 
+
 def get_loader(module_or_name):
     """Get a PEP 302 "loader" object for module_or_name
 
@@ -461,6 +465,7 @@
         fullname = module_or_name
     return find_loader(fullname)
 
+
 def find_loader(fullname):
     """Find a PEP 302 "loader" object for fullname
 
@@ -551,6 +556,7 @@
 
     return path
 
+
 def get_data(package, resource):
     """Get a resource from a package.
 
@@ -594,6 +600,7 @@
 
 DIST_FILES = ('INSTALLER', 'METADATA', 'RECORD', 'REQUESTED',)
 
+
 class Distribution(object):
     """Created with the *path* of the ``.dist-info`` directory provided to the
     constructor. It reads the metadata contained in METADATA when it is
@@ -604,7 +611,7 @@
     name = ''
     """The name of the distribution."""
     metadata = None
-    """A :class:`distutils2.metadata.DistributionMetadata` instance loaded with 
+    """A :class:`distutils2.metadata.DistributionMetadata` instance loaded with
     the distribution's METADATA file."""
     requested = False
     """A boolean that indicates whether the REQUESTED metadata file is present
@@ -620,7 +627,7 @@
         RECORD = os.path.join(self.path, 'RECORD')
         record_reader = csv_reader(open(RECORD, 'rb'), delimiter=',')
         for row in record_reader:
-            path, md5, size = row[:] + [ None for i in xrange(len(row), 3) ]
+            path, md5, size = row[:] + [None for i in xrange(len(row), 3)]
             if local:
                 path = path.replace('/', os.sep)
                 path = os.path.join(sys.prefix, path)
@@ -635,7 +642,6 @@
 
         A local absolute path is an absolute path in which occurrences of
         ``'/'`` have been replaced by the system separator given by ``os.sep``.
-
         :parameter local: flag to say if the path should be returned a local
                           absolute path
         :type local: boolean
@@ -643,10 +649,9 @@
         """
         return self._get_records(local)
 
-
     def uses(self, path):
         """
-        Returns ``True`` if path is listed in RECORD. *path* can be a local 
+        Returns ``True`` if path is listed in RECORD. *path* can be a local
         absolute path or a relative ``'/'``-separated path.
 
         :rtype: boolean
@@ -663,8 +668,8 @@
         ``file`` instance for the file pointed by *path*.
 
         :parameter path: a ``'/'``-separated path relative to the ``.dist-info``
-                         directory or an absolute path; If *path* is an absolute 
-                         path and doesn't start with the ``.dist-info``
+                         directory or an absolute path; If *path* is an
+                         absolute path and doesn't start with the ``.dist-info``
                          directory path, a :class:`DistutilsError` is raised
         :type path: string
         :parameter binary: If *binary* is ``True``, opens the file in read-only
@@ -696,8 +701,8 @@
 
     def get_distinfo_files(self, local=False):
         """
-        Iterates over the RECORD entries and returns paths for each line if the 
-        path is pointing to a file located in the ``.dist-info`` directory or 
+        Iterates over the RECORD entries and returns paths for each line if the
+        path is pointing to a file located in the ``.dist-info`` directory or
         one of its subdirectories.
 
         :parameter local: If *local* is ``True``, each returned path is
@@ -710,16 +715,42 @@
             yield path
 
 
+class EggInfoDistribution(object):
+    """Created with the *path* of the ``.egg-info`` directory or file provided
+    to the constructor. It reads the metadata contained in the file itself, or
+    if the given path happens to be a directory, the metadata is read from the
+    file PKG-INFO under that directory."""
+
+    name = ''
+    """The name of the distribution."""
+    metadata = None
+    """A :class:`distutils2.metadata.DistributionMetadata` instance loaded with
+    the distribution's METADATA file."""
+
+    def __init__(self, path):
+        if os.path.isdir(path):
+            path = os.path.join(path, 'PKG-INFO')
+        self.metadata = DistributionMetadata(path=path)
+        self.name = self.metadata['name']
+
+    def get_installed_files(self, local=False):
+        return []
+
+    def uses(self, path):
+        return False
+
+
 def _normalize_dist_name(name):
     """Returns a normalized name from the given *name*.
     :rtype: string"""
     return name.replace('-', '_')
 
+
 def distinfo_dirname(name, version):
     """
     The *name* and *version* parameters are converted into their
     filename-escaped form, i.e. any ``'-'`` characters are replaced with ``'_'``
-    other than the one in ``'dist-info'`` and the one separating the name from 
+    other than the one in ``'dist-info'`` and the one separating the name from
     the version number.
 
     :parameter name: is converted to a standard distribution name by replacing
@@ -743,13 +774,16 @@
         normalized_version = version
     return '-'.join([name, normalized_version]) + file_extension
 
-def get_distributions():
+
+def get_distributions(use_egg_info=False):
     """
     Provides an iterator that looks for ``.dist-info`` directories in
     ``sys.path`` and returns :class:`Distribution` instances for each one of
-    them.
+    them. If the parameters *use_egg_info* is ``True``, then the ``.egg-info``
+    files and directores are iterated as well.
 
-    :rtype: iterator of :class:`Distribution` instances"""
+    :rtype: iterator of :class:`Distribution` and :class:`EggInfoDistribution`
+            instances"""
     for path in sys.path:
         realpath = os.path.realpath(path)
         if not os.path.isdir(realpath):
@@ -758,25 +792,117 @@
             if dir.endswith('.dist-info'):
                 dist = Distribution(os.path.join(realpath, dir))
                 yield dist
+            elif use_egg_info and dir.endswith('.egg-info'):
+                dist = EggInfoDistribution(os.path.join(realpath, dir))
+                yield dist
 
-def get_distribution(name):
+
+def get_distribution(name, use_egg_info=False):
     """
     Scans all elements in ``sys.path`` and looks for all directories ending with
-    ``.dist-info``. Returns a :class:`Distribution` corresponding to the 
+    ``.dist-info``. Returns a :class:`Distribution` corresponding to the
     ``.dist-info`` directory that contains the METADATA that matches *name* for
-    the *name* metadata.
+    the *name* metadata field.
+    If no distribution exists with the given *name* and the parameter
+    *use_egg_info* is set to ``True``, then all files and directories ending
+    with ``.egg-info`` are scanned. A :class:`EggInfoDistribution` instance is
+    returned if one is found that has metadata that matches *name* for the
+    *name* metadata field.
 
     This function only returns the first result founded, as no more than one
     value is expected. If the directory is not found, ``None`` is returned.
 
-    :rtype: :class:`Distribution` or None"""
+    :rtype: :class:`Distribution` or :class:`EggInfoDistribution: or None"""
     found = None
     for dist in get_distributions():
         if dist.name == name:
             found = dist
             break
+    if use_egg_info:
+        for dist in get_distributions(True):
+            if dist.name == name:
+                found = dist
+                break
     return found
 
+
+def obsoletes_distribution(name, version=None, use_egg_info=False):
+    """
+    Iterates over all distributions to find which distributions obsolete *name*.
+    If a *version* is provided, it will be used to filter the results.
+    If the argument *use_egg_info* is set to ``True``, then ``.egg-info``
+    distributions will be considered as well.
+
+    :type name: string
+    :type version: string
+    :parameter name:
+    """
+    for dist in get_distributions(use_egg_info):
+        obsoleted = dist.metadata['Obsoletes-Dist'] + dist.metadata['Obsoletes']
+        for obs in obsoleted:
+            o_components = obs.split(' ', 1)
+            if len(o_components) == 1 or version is None:
+                if name == o_components[0]:
+                    yield dist
+                    break
+            else:
+                try:
+                    predicate = VersionPredicate(obs)
+                except ValueError:
+                    raise DistutilsError(('Distribution %s has ill formed' +
+                                          ' obsoletes field') % (dist.name,))
+                if name == o_components[0] and predicate.match(version):
+                    yield dist
+                    break
+
+
+def provides_distribution(name, version=None, use_egg_info=False):
+    """
+    Iterates over all distributions to find which distributions provide *name*.
+    If a *version* is provided, it will be used to filter the results. Scans
+    all elements in ``sys.path``  and looks for all directories ending with
+    ``.dist-info``. Returns a :class:`Distribution`  corresponding to the
+    ``.dist-info`` directory that contains a ``METADATA`` that matches *name*
+    for the name metadata. If the argument *use_egg_info* is set to ``True``,
+    then all files and directories ending with ``.egg-info`` are considered
+    as well and returns an :class:`EggInfoDistribution` instance.
+
+    This function only returns the first result founded, since no more than
+    one values are expected. If the directory is not found, returns ``None``.
+
+    :parameter version: a version specifier that indicates the version
+                        required, conforming to the format in ``PEP-345``
+
+    :type name: string
+    :type version: string
+    """
+    predicate = None
+    if not version is None:
+        try:
+            predicate = VersionPredicate(name + ' (' + version + ')')
+        except ValueError:
+            raise DistutilsError('Invalid name or version')
+
+    for dist in get_distributions(use_egg_info):
+        provided = dist.metadata['Provides-Dist'] + dist.metadata['Provides']
+
+        for p in provided:
+            p_components = p.split(' ', 1)
+            if len(p_components) == 1 or predicate is None:
+                if name == p_components[0]:
+                    yield dist
+                    break
+            else:
+                p_name, p_ver = p_components
+                if len(p_ver) < 2 or p_ver[0] != '(' or p_ver[-1] != ')':
+                    raise DistutilsError(('Distribution %s has invalid ' +
+                                          'provides field') % (dist.name,))
+                p_ver = p_ver[1:-1] # trim off the parenthesis
+                if p_name == name and predicate.match(p_ver):
+                    yield dist
+                    break
+
+
 def get_file_users(path):
     """
     Iterates over all distributions to find out which distributions uses
diff --git a/src/distutils2/_backport/tests/fake_dists/bacon-0.1.egg-info/PKG-INFO b/src/distutils2/_backport/tests/fake_dists/bacon-0.1.egg-info/PKG-INFO
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/tests/fake_dists/bacon-0.1.egg-info/PKG-INFO
@@ -0,0 +1,5 @@
+Metadata-Version: 1.2
+Name: bacon
+Version: 0.1
+Provides-Dist: truffles (2.0)
+Obsoletes-Dist: truffles (>=0.9,<=1.5)
diff --git a/src/distutils2/_backport/tests/fake_dists/cheese-2.0.2.egg-info b/src/distutils2/_backport/tests/fake_dists/cheese-2.0.2.egg-info
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/tests/fake_dists/cheese-2.0.2.egg-info
@@ -0,0 +1,5 @@
+Metadata-Version: 1.2
+Name: cheese
+Version: 2.0.2
+Provides-Dist: truffles (1.0.2)
+Obsoletes-Dist: truffles (!=1.2,<=2.0)
diff --git a/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info/METADATA b/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info/METADATA
--- a/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info/METADATA
+++ b/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info/METADATA
@@ -4,3 +4,5 @@
 Summary: Chocolate with a kick!
 Requires-Dist: towel-stuff (0.1)
 Provides-Dist: truffles (1.0)
+Obsoletes-Dist: truffles (<=0.8,>=0.5)
+Obsoletes-Dist: truffles (<=0.9,>=0.6)
diff --git a/src/distutils2/_backport/tests/fake_dists/grammar-1.0a4.dist-info/METADATA b/src/distutils2/_backport/tests/fake_dists/grammar-1.0a4.dist-info/METADATA
--- a/src/distutils2/_backport/tests/fake_dists/grammar-1.0a4.dist-info/METADATA
+++ b/src/distutils2/_backport/tests/fake_dists/grammar-1.0a4.dist-info/METADATA
@@ -1,3 +1,4 @@
 Metadata-Version: 1.2
 Name: grammar
 Version: 1.0a4
+Requires-Dist: truffles (>=1.2)
diff --git a/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1.dist-info/METADATA b/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1.dist-info/METADATA
--- a/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1.dist-info/METADATA
+++ b/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1.dist-info/METADATA
@@ -1,3 +1,5 @@
 Metadata-Version: 1.2
 Name: towel-stuff
 Version: 0.1
+Provides-Dist: truffles (1.1.2)
+Obsoletes-Dist: truffles (!=0.8,<1.0)
diff --git a/src/distutils2/_backport/tests/test_pkgutil.py b/src/distutils2/_backport/tests/test_pkgutil.py
--- a/src/distutils2/_backport/tests/test_pkgutil.py
+++ b/src/distutils2/_backport/tests/test_pkgutil.py
@@ -216,7 +216,9 @@
         found_dists = []
 
         # Import the function in question
-        from distutils2._backport.pkgutil import get_distributions, Distribution
+        from distutils2._backport.pkgutil import get_distributions, \
+                                                 Distribution, \
+                                                 EggInfoDistribution
 
         # Verify the fake dists have been found.
         dists = [ dist for dist in get_distributions() ]
@@ -231,13 +233,31 @@
         # Finally, test that we found all that we were looking for
         self.assertListEqual(sorted(found_dists), sorted(fake_dists))
 
+        # Now, test if the egg-info distributions are found correctly as well
+        fake_dists += [('bacon', '0.1'), ('cheese', '2.0.2')]
+        found_dists = []
+
+        dists = [ dist for dist in get_distributions(use_egg_info=True) ]
+        for dist in dists:
+            if not (isinstance(dist, Distribution) or \
+                    isinstance(dist, EggInfoDistribution)):
+                self.fail("item received was not a Distribution or "
+                          "EggInfoDistribution instance: %s" % type(dist))
+            if dist.name in dict(fake_dists).keys():
+                found_dists.append((dist.name, dist.metadata['version']))
+
+        self.assertListEqual(sorted(fake_dists), sorted(found_dists))
+
+
     def test_get_distribution(self):
         """Test for looking up a distribution by name."""
         # Test the lookup of the towel-stuff distribution
         name = 'towel-stuff' # Note: This is different from the directory name
 
         # Import the function in question
-        from distutils2._backport.pkgutil import get_distribution, Distribution
+        from distutils2._backport.pkgutil import get_distribution, \
+                                                 Distribution, \
+                                                 EggInfoDistribution
 
         # Lookup the distribution
         dist = get_distribution(name)
@@ -250,6 +270,21 @@
         # Verify partial name matching doesn't work
         self.assertEqual(None, get_distribution('towel'))
 
+        # Verify that it does not find egg-info distributions, when not
+        # instructed to
+        self.assertEqual(None, get_distribution('bacon'))
+        self.assertEqual(None, get_distribution('cheese'))
+
+        # Now check that it works well in both situations, when egg-info
+        # is a file and directory respectively.
+        dist = get_distribution('cheese', use_egg_info=True)
+        self.assertTrue(isinstance(dist, EggInfoDistribution))
+        self.assertEqual(dist.name, 'cheese')
+
+        dist = get_distribution('bacon', use_egg_info=True)
+        self.assertTrue(isinstance(dist, EggInfoDistribution))
+        self.assertEqual(dist.name, 'bacon')
+
     def test_get_file_users(self):
         """Test the iteration of distributions that use a file."""
         from distutils2._backport.pkgutil import get_file_users, Distribution
@@ -260,6 +295,80 @@
             self.assertTrue(isinstance(dist, Distribution))
             self.assertEqual(dist.name, name)
 
+    def test_provides(self):
+        """ Test for looking up distributions by what they provide """
+        from distutils2._backport.pkgutil import provides_distribution
+        from distutils2.errors import DistutilsError
+
+        checkLists = lambda x,y: self.assertListEqual(sorted(x), sorted(y))
+
+        l = [dist.name for dist in provides_distribution('truffles')]
+        checkLists(l, ['choxie', 'towel-stuff'])
+
+        l = [dist.name for dist in provides_distribution('truffles', '1.0')]
+        checkLists(l, ['choxie'])
+
+        l = [dist.name for dist in provides_distribution('truffles', '1.0',
+                                                         use_egg_info=True)]
+        checkLists(l, ['choxie', 'cheese'])
+
+        l = [dist.name for dist in provides_distribution('truffles', '1.1.2')]
+        checkLists(l, ['towel-stuff'])
+
+        l = [dist.name for dist in provides_distribution('truffles', '1.1')]
+        checkLists(l, ['towel-stuff'])
+
+        l = [dist.name for dist in provides_distribution('truffles', '!=1.1,<=2.0')]
+        checkLists(l, ['choxie'])
+
+        l = [dist.name for dist in provides_distribution('truffles', '!=1.1,<=2.0',
+                                                          use_egg_info=True)]
+        checkLists(l, ['choxie', 'bacon', 'cheese'])
+
+        l = [dist.name for dist in provides_distribution('truffles', '>1.0')]
+        checkLists(l, ['towel-stuff'])
+
+        l = [dist.name for dist in provides_distribution('truffles', '>1.5')]
+        checkLists(l, [])
+
+        l = [dist.name for dist in provides_distribution('truffles', '>1.5',
+                                                         use_egg_info=True)]
+        checkLists(l, ['bacon'])
+
+        l = [dist.name for dist in provides_distribution('truffles', '>=1.0')]
+        checkLists(l, ['choxie', 'towel-stuff'])
+
+    def test_obsoletes(self):
+        """ Test looking for distributions based on what they obsolete """
+        from distutils2._backport.pkgutil import obsoletes_distribution
+        from distutils2.errors import DistutilsError
+
+        checkLists = lambda x,y: self.assertListEqual(sorted(x), sorted(y))
+
+        l = [dist.name for dist in obsoletes_distribution('truffles', '1.0')]
+        checkLists(l, [])
+
+        l = [dist.name for dist in obsoletes_distribution('truffles', '1.0',
+                                                          use_egg_info=True)]
+        checkLists(l, ['cheese', 'bacon'])
+
+
+        l = [dist.name for dist in obsoletes_distribution('truffles', '0.8')]
+        checkLists(l, ['choxie'])
+
+        l = [dist.name for dist in obsoletes_distribution('truffles', '0.8',
+                                                          use_egg_info=True)]
+        checkLists(l, ['choxie', 'cheese'])
+
+        l = [dist.name for dist in obsoletes_distribution('truffles', '0.9.6')]
+        checkLists(l, ['choxie', 'towel-stuff'])
+
+        l = [dist.name for dist in obsoletes_distribution('truffles', '0.5.2.3')]
+        checkLists(l, ['choxie', 'towel-stuff'])
+
+        l = [dist.name for dist in obsoletes_distribution('truffles', '0.2')]
+        checkLists(l, ['towel-stuff'])
+
 
 def test_suite():
     suite = unittest2.TestSuite()
@@ -273,3 +382,16 @@
 
 if __name__ == "__main__":
     test_main()
+
+def test_suite():
+    suite = unittest2.TestSuite()
+    testcase_loader = unittest2.loader.defaultTestLoader.loadTestsFromTestCase
+    suite.addTest(testcase_loader(TestPkgUtilFunctions))
+    suite.addTest(testcase_loader(TestPkgUtilDistribution))
+    return suite
+
+def test_main():
+    run_unittest(test_suite())
+
+if __name__ == "__main__":
+    test_main()
diff --git a/src/distutils2/command/sdist.py b/src/distutils2/command/sdist.py
--- a/src/distutils2/command/sdist.py
+++ b/src/distutils2/command/sdist.py
@@ -121,6 +121,7 @@
         self.metadata_check = 1
         self.owner = None
         self.group = None
+        self.filelist = None
 
     def _check_archive_formats(self, formats):
         supported_formats = [name for name, desc in get_archive_formats()]
@@ -152,10 +153,14 @@
         if self.dist_dir is None:
             self.dist_dir = "dist"
 
+        if self.filelist is None:
+            self.filelist = Manifest()
+
+
     def run(self):
         # 'filelist' contains the list of files that will make up the
         # manifest
-        self.filelist = Manifest()
+        self.filelist.clear()
 
         # Run sub commands
         for cmd_name in self.get_sub_commands():
@@ -200,7 +205,7 @@
         if self.use_defaults:
             self.add_defaults()
         if template_exists:
-            self.read_template()
+            self.filelist.read_template(self.template)
         if self.prune:
             self.prune_file_list()
 
diff --git a/src/distutils2/converter/fixers/fix_imports.py b/src/distutils2/converter/fixers/fix_imports.py
--- a/src/distutils2/converter/fixers/fix_imports.py
+++ b/src/distutils2/converter/fixers/fix_imports.py
@@ -20,6 +20,9 @@
         if node.type != syms.import_from:
             return
 
+        if not hasattr(imp, "next_sibling"):
+            imp.next_sibling = imp.get_next_sibling()
+
         while not hasattr(imp, 'value'):
             imp = imp.children[0]
 
@@ -34,6 +37,8 @@
             next = imp.next_sibling
             while next is not None:
                 pattern.append(next.value)
+                if not hasattr(next, "next_sibling"):
+                    next.next_sibling = next.get_next_sibling()
                 next = next.next_sibling
             if pattern == ['import', 'setup']:
                 imp.value = 'distutils2.core'
diff --git a/src/distutils2/converter/fixers/fix_setup_options.py b/src/distutils2/converter/fixers/fix_setup_options.py
--- a/src/distutils2/converter/fixers/fix_setup_options.py
+++ b/src/distutils2/converter/fixers/fix_setup_options.py
@@ -41,6 +41,10 @@
 
     def _fix_name(self, argument, remove_list):
         name = argument.children[0]
+
+        if not hasattr(name, "next_sibling"):
+            name.next_sibling = name.get_next_sibling()
+
         sibling = name.next_sibling
         if sibling is None or sibling.type != token.EQUAL:
             return False
@@ -48,6 +52,8 @@
         if name.value in _OLD_NAMES:
             name.value = _OLD_NAMES[name.value]
             if name.value in _SEQUENCE_NAMES:
+                if not hasattr(sibling, "next_sibling"):
+                    sibling.next_sibling = sibling.get_next_sibling()
                 right_operand = sibling.next_sibling
                 # replacing string -> list[string]
                 if right_operand.type == token.STRING:
diff --git a/src/distutils2/depgraph.py b/src/distutils2/depgraph.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/depgraph.py
@@ -0,0 +1,181 @@
+"""
+A dependency graph generator. The graph is represented as an instance of
+:class:`DependencyGraph`, and DOT output is possible as well.
+"""
+
+from distutils2._backport import pkgutil
+from distutils2.errors import DistutilsError
+from distutils2.version import VersionPredicate
+
+__all__ = ['DependencyGraph', 'generate_graph']
+
+
+class DependencyGraph(object):
+    """
+    Represents a dependency graph between distributions.
+
+    The depedency relationships are stored in an *adjacency_list* that maps
+    distributions to a list of ``(other, label)`` tuples where  ``other``
+    is a distribution and the edge is labelled with ``label`` (i.e. the version
+    specifier, if such was provided). If any missing depencies are found,
+    they are stored in ``missing``. It maps distributions to a list of
+    requirements that were not provided by any other distributions.
+    """
+
+    def __init__(self):
+        self.adjacency_list = {}
+        self.missing = {}
+
+    def add_distribution(self, distribution):
+        """
+        Add distribution *x* to the graph.
+
+        :type distribution: :class:`pkgutil.Distribution` or
+                            :class:`pkgutil.EggInfoDistribution`
+        """
+        self.adjacency_list[distribution] = list()
+        self.missing[distribution] = list()
+
+    def add_edge(self, x, y, label=None):
+        """
+        Add an edge from distribution *x* to distribution *y* with the given
+        *label*.
+
+
+        :type x: :class:`pkgutil.Distribution` or
+                 :class:`pkgutil.EggInfoDistribution`
+        :type y: :class:`pkgutil.Distribution` or
+                 :class:`pkgutil.EggInfoDistribution`
+        :type label: ``str`` or ``None``
+        """
+        self.adjacency_list[x].append((y, label))
+
+    def add_missing(self, distribution, requirement):
+        """
+        Add a missing *requirement* for the given *distribution*.
+
+        :type distribution: :class:`pkgutil.Distribution` or
+                            :class:`pkgutil.EggInfoDistribution`
+        :type requirement: ``str``
+        """
+        self.missing[distribution].append(requirement)
+
+    def to_dot(self, f, skip_disconnected=True):
+        """
+        Writes a DOT output for the graph to the provided *file*.
+        If *skip_disconnected* is set to ``True``, then all distributions
+        that are not dependent on any other distributions are skipped.
+
+        :type f: ``file``
+        ;type skip_disconnected: ``bool``
+        """
+        if not isinstance(f, file):
+            raise TypeError('the argument has to be of type file')
+
+        disconnected = []
+
+        f.write("digraph dependencies {\n")
+        for dist, adjs in self.adjacency_list.iteritems():
+            if len(adjs) == 0 and not skip_disconnected:
+                disconnected.append(dist)
+            for (other, label) in adjs:
+                if not label is None:
+                    f.write('"%s" -> "%s" [label="%s"]\n' %
+                                                (dist.name, other.name, label))
+                else:
+                    f.write('"%s" -> "%s"\n' % (dist.name, other.name))
+        if not skip_disconnected and len(disconnected) > 0:
+            f.write('subgraph disconnected {\n')
+            f.write('label = "Disconnected"\n')
+            f.write('bgcolor = red\n')
+
+            for dist in disconnected:
+                f.write('"%s"' % dist.name)
+                f.write('\n')
+            f.write('}\n')
+        f.write('}\n')
+
+
+def generate_graph(dists):
+    """
+    Generates a dependency graph from the given distributions.
+
+    :parameter dists: a list of distributions
+    :type dists: list of :class:`pkgutil.Distribution` and
+                         :class:`pkgutil.EggInfoDistribution` instances
+    :rtype: an :class:`DependencyGraph` instance
+    """
+    graph = DependencyGraph()
+    provided = {} # maps names to lists of (version, dist) tuples
+    dists = list(dists) # maybe use generator_tools in future
+
+    # first, build the graph and find out the provides
+    for dist in dists:
+        graph.add_distribution(dist)
+        provides = dist.metadata['Provides-Dist'] + dist.metadata['Provides']
+
+        for p in provides:
+            comps = p.split(" ", 1)
+            name = comps[0]
+            version = None
+            if len(comps) == 2:
+                version = comps[1]
+                if len(version) < 3 or version[0] != '(' or version[-1] != ')':
+                    raise DistutilsError('Distribution %s has ill formed' \
+                                         'provides field: %s' % (dist.name, p))
+                version = version[1:-1] # trim off parenthesis
+            if not name in provided:
+                provided[name] = []
+            provided[name].append((version, dist))
+
+    # now make the edges
+    for dist in dists:
+        requires = dist.metadata['Requires-Dist'] + dist.metadata['Requires']
+        for req in requires:
+            predicate = VersionPredicate(req)
+            comps = req.split(" ", 1)
+            name = comps[0]
+
+            if not name in provided:
+                graph.add_missing(dist, req)
+            else:
+                for (version, provider) in provided[name]:
+                    if predicate.match(version):
+                        graph.add_edge(dist, provider, req)
+
+    return graph
+
+
+def dependent_dists(dists, dist):
+    """
+    Recursively generate a list of distributions from *dists* that are
+    dependent on *dist*.
+
+    :param dists: a list of distributions
+    :param dist: a distribution, member of *dists* for which we are interested
+    """
+    if not dist in dists:
+        raise ValueError('The given distribution is not a member of the list')
+    graph = generate_graph(dists)
+
+    dep = [dist]
+    fringe = [dist] # list of nodes we should expand
+    while not len(fringe) == 0:
+        next = graph.adjacency_list[fringe.pop()]
+        for (dist, label) in next:
+            if not dist in dep: # avoid infinite loops
+                dep.append(dist)
+                fringe.append(dist)
+
+    dep.pop()
+    return dep
+
+if __name__ == '__main__':
+    dists = list(pkgutil.get_distributions(use_egg_info=True))
+    graph = generate_graph(dists)
+    for dist, reqs in graph.missing.iteritems():
+        if len(reqs) > 0:
+            print("Missing dependencies for %s: %s" % (dist.name,
+                                                       ", ".join(reqs)))
+    f = open('output.dot', 'w')
+    graph.to_dot(f, True)
diff --git a/src/distutils2/manifest.py b/src/distutils2/manifest.py
--- a/src/distutils2/manifest.py
+++ b/src/distutils2/manifest.py
@@ -54,6 +54,12 @@
         for sort_tuple in sortable_files:
             self.files.append(os.path.join(*sort_tuple))
 
+    def clear(self):
+        """Clear all collected files."""
+        self.files = []
+        if self.allfiles is not None:
+            self.allfiles = []
+
     def remove_duplicates(self):
         # Assumes list has been sorted!
         for i in range(len(self.files) - 1, 0, -1):
diff --git a/src/distutils2/tests/test_converter.py b/src/distutils2/tests/test_converter.py
--- a/src/distutils2/tests/test_converter.py
+++ b/src/distutils2/tests/test_converter.py
@@ -36,4 +36,4 @@
     return unittest2.makeSuite(ConverterTestCase)
 
 if __name__ == '__main__':
-    run_unittest(test_suite())
+    unittest2.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_core.py b/src/distutils2/tests/test_core.py
--- a/src/distutils2/tests/test_core.py
+++ b/src/distutils2/tests/test_core.py
@@ -60,6 +60,19 @@
         distutils2.core.run_setup(
             self.write_setup(setup_using___file__))
 
+    def test_run_setup_stop_after(self):
+        f = self.write_setup(setup_using___file__)
+        for s in ['init', 'config', 'commandline', 'run']:
+            distutils2.core.run_setup(f, stop_after=s)
+        self.assertRaises(ValueError, distutils2.core.run_setup, 
+                          f, stop_after='bob')
+
+    def test_run_setup_args(self):
+        f = self.write_setup(setup_using___file__)
+        d = distutils2.core.run_setup(f, script_args=["--help"], 
+                                        stop_after="init")
+        self.assertEqual(['--help'], d.script_args)
+
     def test_run_setup_uses_current_dir(self):
         # This tests that the setup script is run with the current directory
         # as its own current directory; this was temporarily broken by a
diff --git a/src/distutils2/tests/test_sdist.py b/src/distutils2/tests/test_sdist.py
--- a/src/distutils2/tests/test_sdist.py
+++ b/src/distutils2/tests/test_sdist.py
@@ -343,6 +343,19 @@
         finally:
             archive.close()
 
+    def test_get_file_list(self):
+        dist, cmd = self.get_cmd()
+        cmd.finalize_options()
+        cmd.template = os.path.join(self.tmp_dir, 'MANIFEST.in')
+        f = open(cmd.template, 'w')
+        try:
+            f.write('include MANIFEST.in\n')
+        finally:
+            f.close()
+
+        cmd.get_file_list()
+        self.assertIn('MANIFEST.in', cmd.filelist.files)
+
 def test_suite():
     return unittest2.makeSuite(SDistTestCase)
 
diff --git a/src/distutils2/tests/test_version.py b/src/distutils2/tests/test_version.py
--- a/src/distutils2/tests/test_version.py
+++ b/src/distutils2/tests/test_version.py
@@ -139,13 +139,17 @@
         for predicate in predicates:
             v = VersionPredicate(predicate)
 
-        assert VersionPredicate('Hey (>=2.5,<2.7)').match('2.6')
-        assert VersionPredicate('Ho').match('2.6')
-        assert not VersionPredicate('Hey (>=2.5,!=2.6,<2.7)').match('2.6')
-        assert VersionPredicate('Ho (<3.0)').match('2.6')
-        assert VersionPredicate('Ho (<3.0,!=2.5)').match('2.6.0')
-        assert not VersionPredicate('Ho (<3.0,!=2.6)').match('2.6.0')
-
+        self.assertTrue(VersionPredicate('Hey (>=2.5,<2.7)').match('2.6'))
+        self.assertTrue(VersionPredicate('Ho').match('2.6'))
+        self.assertFalse(VersionPredicate('Hey (>=2.5,!=2.6,<2.7)').match('2.6'))
+        self.assertTrue(VersionPredicate('Ho (<3.0)').match('2.6'))
+        self.assertTrue(VersionPredicate('Ho (<3.0,!=2.5)').match('2.6.0'))
+        self.assertFalse(VersionPredicate('Ho (<3.0,!=2.6)').match('2.6.0'))
+        self.assertTrue(VersionPredicate('Ho (2.5)').match('2.5.4'))
+        self.assertFalse(VersionPredicate('Ho (!=2.5)').match('2.5.2'))
+        self.assertTrue(VersionPredicate('Hey (<=2.5)').match('2.5.9'))
+        self.assertFalse(VersionPredicate('Hey (<=2.5)').match('2.6.0'))
+        self.assertTrue(VersionPredicate('Hey (>=2.5)').match('2.5.1'))
 
         # XXX need to silent the micro version in this case
         #assert not VersionPredicate('Ho (<3.0,!=2.6)').match('2.6.3')
diff --git a/src/distutils2/version.py b/src/distutils2/version.py
--- a/src/distutils2/version.py
+++ b/src/distutils2/version.py
@@ -330,10 +330,11 @@
 
     _operators = {"<": lambda x, y: x < y,
                   ">": lambda x, y: x > y,
-                  "<=": lambda x, y: x <= y,
-                  ">=": lambda x, y: x >= y,
-                  "==": lambda x, y: x == y,
-                  "!=": lambda x, y: x != y}
+                  "<=": lambda x, y: str(x).startswith(str(y)) or x < y,
+                  ">=": lambda x, y: str(x).startswith(str(y)) or x > y,
+                  "==": lambda x, y: str(x).startswith(str(y)),
+                  "!=": lambda x, y: not str(x).startswith(str(y)),
+                  }
 
     def __init__(self, predicate):
         predicate = predicate.strip()
diff --git a/src/runtests.py b/src/runtests.py
--- a/src/runtests.py
+++ b/src/runtests.py
@@ -6,7 +6,7 @@
 
 def test_main():
     import distutils2.tests
-    from distutils2.tests import run_unittest, reap_children
+    from distutils2.tests import run_unittest, reap_children, TestFailed
     from distutils2._backport.tests import test_suite as btest_suite
     # just supporting -q right now
     # to enable detailed/quiet output
@@ -14,10 +14,15 @@
         verbose = sys.argv[-1] != '-q'
     else:
         verbose = 1
-
-    run_unittest([distutils2.tests.test_suite(), btest_suite()],
-                 verbose_=verbose)
-    reap_children()
+    try:
+        try:
+            run_unittest([distutils2.tests.test_suite(), btest_suite()],
+                    verbose_=verbose)
+            return 0
+        except TestFailed:
+            return 1
+    finally:
+        reap_children()
 
 if __name__ == "__main__":
     try:
@@ -26,4 +31,5 @@
         print('!!! You need to install unittest2')
         sys.exit(1)
 
-    test_main()
+    sys.exit(test_main())
+
diff --git a/src/tests.sh b/src/tests.sh
--- a/src/tests.sh
+++ b/src/tests.sh
@@ -1,12 +1,28 @@
 #!/bin/sh
-echo Testing with Python 2.4....
-python2.4 runtests.py -q
+echo -n "Running tests for Python 2.4..."
+python2.4 runtests.py -q > /dev/null 2> /dev/null
+if [ $? -ne 0 ];then
+    echo "Failed"
+    exit $1
+else
+    echo "Success"
+fi
 
-echo
-echo Testing with Python 2.5....
-python2.5 runtests.py -q
+echo -n "Running tests for Python 2.5..."
+python2.5 runtests.py -q > /dev/null 2> /dev/null
+if [ $? -ne 0 ];then
+    echo "Failed"
+    exit $1
+else
+    echo "Success"
+fi
 
-echo
-echo Testing with Python 2.6....
-python2.6 runtests.py -q
+echo -n "Running tests for Python 2.6..."
+python2.6 runtests.py -q > /dev/null 2> /dev/null
+if [ $? -ne 0 ];then
+    echo "Failed"
+    exit $1
+else
+    echo "Success"
+fi
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: PyPI mock server. Always search paths
	in reverse order	(last first).
Message-ID: <E1OVLoV-0000iJ-Em@dinsdale.python.org>

tarek.ziade pushed 3c3947befe9d to distutils2:

http://hg.python.org/distutils2/rev/3c3947befe9d
changeset:   292:3c3947befe9d
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Tue Jun 01 11:50:43 2010 +0200
summary:     PyPI mock server. Always search paths in reverse order (last first).
files:       docs/source/test_framework.rst, src/distutils2/tests/pypi_server.py

diff --git a/docs/source/test_framework.rst b/docs/source/test_framework.rst
--- a/docs/source/test_framework.rst
+++ b/docs/source/test_framework.rst
@@ -29,6 +29,12 @@
 
     server = PyPIServer(static_filesystem_paths=["path/to/your/dir"])
 
+Or add some paths to the existing ones, the server will alwas try to load paths
+in reverse order (e.g here, try "another/super/path" then the default one) ::
+
+    server = PyPIServer()
+    server.static_filesystem_paths.append("another/super/path")
+
 ``PyPIServerTestCase``
 ======================
 
diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
--- a/src/distutils2/tests/pypi_server.py
+++ b/src/distutils2/tests/pypi_server.py
@@ -35,8 +35,8 @@
         self.default_response_status = "200 OK"
         self.default_response_headers = [('Content-type', 'text/plain')]
         self.default_response_data = ["hello"]
-        self._static_uri_paths = static_uri_paths
-        self._static_filesystem_paths = static_filesystem_paths
+        self.static_uri_paths = static_uri_paths
+        self.static_filesystem_paths = static_filesystem_paths
 
     def run(self):
         self.httpd.serve_forever()
@@ -65,12 +65,16 @@
         self.request_queue.put((environ, request_data))
         
         # serve the content from local disc if we request an URL beginning 
-        # by a pattern defined in `_static_paths`
+        # by a pattern defined in `static_paths`
         relative_path = environ["PATH_INFO"].replace(self.full_address, '')
         url_parts = relative_path.split("/")
-        if len(url_parts) > 1 and url_parts[1] in self._static_uri_paths:
+        if len(url_parts) > 1 and url_parts[1] in self.static_uri_paths:
             data = None
-            for fs_path in self._static_filesystem_paths:
+            # always take the last first.
+            fs_paths = []
+            fs_paths.extend(self.static_filesystem_paths)
+            fs_paths.reverse()
+            for fs_path in fs_paths:
                 try:
                     file = open(fs_path + relative_path)
                     data = file.read()

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: PyPI mocked server can now serve
	static content for	specified paths
Message-ID: <E1OVLoV-0000iC-DQ@dinsdale.python.org>

tarek.ziade pushed 1c0062d03f12 to distutils2:

http://hg.python.org/distutils2/rev/1c0062d03f12
changeset:   290:1c0062d03f12
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Tue Jun 01 10:35:32 2010 +0200
summary:     PyPI mocked server can now serve static content for specified paths
files:       src/distutils2/tests/pypi_server.py, src/distutils2/tests/pypiserver/external/index.html, src/distutils2/tests/pypiserver/index.html, src/distutils2/tests/pypiserver/simple/index.html, src/distutils2/tests/test_pypi_server.py, src/distutils2/tests/test_pypi_simple.py

diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
--- a/src/distutils2/tests/pypi_server.py
+++ b/src/distutils2/tests/pypi_server.py
@@ -2,7 +2,7 @@
 from wsgiref.simple_server import make_server
 import os.path
 
-PYPI_HTML_BASE_PATH = os.path.dirname(os.path.abspath(__file__)) + "/pypiserver"
+PYPI_DEFAULT_STATIC_PATH =  os.path.dirname(os.path.abspath(__file__)) + "/pypiserver"
 
 class PyPIServerTestCase(unittest2.TestCase):
 
@@ -17,7 +17,15 @@
 
 class PyPIServer(threading.Thread):
     """Thread that wraps a wsgi app"""
-    def __init__(self):
+    
+    def __init__(self, static_uri_paths=["pypi"],
+            static_filesystem_path=PYPI_DEFAULT_STATIC_PATH):
+        """Initialize the server.
+
+        static_uri_paths and static_base_path are parameters used to provides
+        respectively the http_paths to serve statically, and where to find the
+        matching files on the filesystem.
+        """
         threading.Thread.__init__(self)
         self.httpd = make_server('', 0, self.pypi_app)
         self.httpd.RequestHandlerClass.log_request = lambda *_: None
@@ -27,6 +35,8 @@
         self.default_response_status = "200 OK"
         self.default_response_headers = [('Content-type', 'text/plain')]
         self.default_response_data = ["hello"]
+        self._static_uri_paths = static_uri_paths
+        self._static_filesystem_path = static_filesystem_path
 
     def run(self):
         self.httpd.serve_forever()
@@ -36,10 +46,17 @@
         self.join()
 
     def pypi_app(self, environ, start_response):
+        """Serve the content.
+
+        Also record the requests to be accessed later. If trying to access an
+        url matching `_static_paths`, serve static content, otherwise serve 
+        what is provided by the `get_next_response` method.
+        """
         # record the request. Read the input only on PUT or POST requests
         if environ["REQUEST_METHOD"] in ("PUT", "POST"):
             if environ.get("CONTENT_LENGTH"):
-                request_data = environ.pop('wsgi.input').read(int(environ['CONTENT_LENGTH']))
+                request_data = environ.pop('wsgi.input').read(
+                    int(environ['CONTENT_LENGTH']))
             else:
                 request_data = environ.pop('wsgi.input').read()
         elif environ["REQUEST_METHOD"] in ("GET", "DELETE"):
@@ -47,11 +64,12 @@
             
         self.request_queue.put((environ, request_data))
         
+        # serve the content from local disc if we request an URL beginning 
+        # by a pattern defined in `_static_paths`
         relative_path = environ["PATH_INFO"].replace(self.full_address, '')
-        
-        # serve the content from local disc if we request /simple
-        if relative_path.startswith("/simple/"):
-            file_to_serve = relative_path.replace("/simple", PYPI_HTML_BASE_PATH)
+        url_parts = relative_path.split("/")
+        if len(url_parts) > 1 and url_parts[1] in self._static_uri_paths:
+            file_to_serve = self._static_filesystem_path + relative_path
             try:
                 file = open(file_to_serve)
                 data = file.read()
@@ -61,7 +79,7 @@
                 data = "Not Found"
             return data
 
-        # otherwise serve the content from default_response* parameters
+        # otherwise serve the content from get_next_response
         else:
             # send back a response
             status, headers, data = self.get_next_response()
diff --git a/src/distutils2/tests/pypiserver/external/index.html b/src/distutils2/tests/pypiserver/external/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/external/index.html
@@ -0,0 +1,1 @@
+index.html from external server
diff --git a/src/distutils2/tests/pypiserver/index.html b/src/distutils2/tests/pypiserver/simple/index.html
rename from src/distutils2/tests/pypiserver/index.html
rename to src/distutils2/tests/pypiserver/simple/index.html
diff --git a/src/distutils2/tests/test_pypi_server.py b/src/distutils2/tests/test_pypi_server.py
--- a/src/distutils2/tests/test_pypi_server.py
+++ b/src/distutils2/tests/test_pypi_server.py
@@ -1,6 +1,7 @@
 """Tests for distutils.command.bdist."""
 import unittest2, urllib, urllib2
 from distutils2.tests.pypi_server import PyPIServer
+import os.path
 
 class PyPIServerTest(unittest2.TestCase):
 
@@ -22,9 +23,20 @@
         server.stop()
 
     def test_serve_static_content(self):
-        """We expect that when accessing the test PyPI server, files can 
-        be served statically."""
-        server = PyPIServer()
+        """PYPI Mocked server can serve static content from disk.
+        """
+
+        def uses_local_files_for(server, url_path):
+            """Test that files are served statically (eg. the output from the
+            server is the same than the one made by a simple file read.
+            """
+            url = server.full_address + url_path 
+            request = urllib2.Request(url)
+            response = urllib2.urlopen(request)
+            file = open(server._static_filesystem_path + url_path)
+            return response.read() == file.read()
+
+        server = PyPIServer(static_uri_paths=["simple", "external"])
         server.start()
         
         # the file does not exists on the disc, so it might not be served
@@ -36,10 +48,10 @@
             self.assertEqual(e.getcode(), 404)
 
         # now try serving a content that do exists
-        url = server.full_address + "/simple/index.html"
-        request = urllib2.Request(url)
-        f = urllib2.urlopen(request)
-        self.assertEqual(f.read(), "Yeah\n")
+        self.assertTrue(uses_local_files_for(server, "/simple/index.html"))
+
+        # and another one in another root path
+        self.assertTrue(uses_local_files_for(server, "/external/index.html"))
 
 def test_suite():
     return unittest2.makeSuite(PyPIServerTest)
diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -1,7 +1,6 @@
-"""Tests for pypi.simple module.
+"""Tests for the pypi.simple module.
 
 """
-# More would be better!
 import sys
 import os
 import shutil 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: merge konryd changes
Message-ID: <E1OVLoV-0000im-Iy@dinsdale.python.org>

tarek.ziade pushed 0840194e5a63 to distutils2:

http://hg.python.org/distutils2/rev/0840194e5a63
changeset:   299:0840194e5a63
parent:      298:031d3246aa0d
parent:      279:ec619c341cd2
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Thu Jun 03 17:45:59 2010 +0200
summary:     merge konryd changes
files:       

diff --git a/src/distutils2/command/upload_docs.py b/src/distutils2/command/upload_docs.py
--- a/src/distutils2/command/upload_docs.py
+++ b/src/distutils2/command/upload_docs.py
@@ -1,6 +1,7 @@
 import base64, httplib, os.path, socket, tempfile, urlparse, zipfile
 from cStringIO import StringIO
 from distutils2 import log
+from distutils2.command.upload import upload
 from distutils2.core import PyPIRCCommand
 from distutils2.errors import DistutilsFileError
 
@@ -27,7 +28,6 @@
     """
     if boundary is None:
         boundary = '--------------GHSKFJDLGDS7543FJKLFHRE75642756743254'
-    CRLF = '\r\n'
     l = []
     for (key, value) in fields:
         l.extend([
@@ -43,24 +43,29 @@
             value])
     l.append('--' + boundary + '--')
     l.append('')
-    body = CRLF.join(l)
+    body =  '\r\n'.join(l)
     content_type = 'multipart/form-data; boundary=%s' % boundary
     return content_type, body
 
 class upload_docs(PyPIRCCommand):
 
     user_options = [
+        ('repository=', 'r', "url of repository [default: %s]" % upload.DEFAULT_REPOSITORY),
+        ('show-response', None, 'display full response text from server'),
         ('upload-dir=', None, 'directory to upload'),
         ]
 
     def initialize_options(self):
-        self.upload_dir = None
+        PyPIRCCommand.initialize_options(self)
+        self.upload_dir = "build/docs"
 
     def finalize_options(self):
         PyPIRCCommand.finalize_options(self)
         if self.upload_dir == None:
             build = self.get_finalized_command('build')
             self.upload_dir = os.path.join(build.build_base, "docs")
+        self.announce('Using upload directory %s' % self.upload_dir)
+        self.verify_upload_dir(self.upload_dir)
         config = self._read_pypirc()
         if config != {}:
             self.username = config['username']
@@ -69,16 +74,15 @@
             self.realm = config['realm']
 
     def verify_upload_dir(self, upload_dir):
+        self.ensure_dirname('upload_dir')
         index_location = os.path.join(upload_dir, "index.html")
         if not os.path.exists(index_location):
             mesg = "No 'index.html found in docs directory (%s)"
             raise DistutilsFileError(mesg % upload_dir)
 
-
     def run(self):
         tmp_dir = tempfile.mkdtemp()
         name = self.distribution.metadata['Name']
-        self.verify_upload_dir(self.upload_dir)
         zip_file = zip_dir(self.upload_dir)
 
         fields = {':action': 'doc_upload', 'name': name}.items()
@@ -126,3 +130,6 @@
         else:
             self.announce('Upload failed (%s): %s' % (r.status, r.reason),
                           log.ERROR)
+
+        if self.show_response:
+            print "\n".join(['-'*75, r.read(), '-'*75])
diff --git a/src/distutils2/tests/test_upload_docs.py b/src/distutils2/tests/test_upload_docs.py
--- a/src/distutils2/tests/test_upload_docs.py
+++ b/src/distutils2/tests/test_upload_docs.py
@@ -1,6 +1,6 @@
 """Tests for distutils.command.upload_docs."""
 # -*- encoding: utf8 -*-
-import httplib, os, os.path, tempfile, unittest2, zipfile
+import httplib, os, os.path, shutil, sys, tempfile, unittest2, zipfile
 from cStringIO import StringIO
 
 from distutils2.command import upload_docs as upload_docs_mod
@@ -8,7 +8,7 @@
                                     encode_multipart)
 from distutils2.core import Distribution
 
-from distutils2.errors import DistutilsFileError
+from distutils2.errors import DistutilsFileError, DistutilsOptionError
 
 from distutils2.tests import support
 from distutils2.tests.pypi_server import PyPIServer, PyPIServerTestCase
@@ -54,15 +54,23 @@
         self.dist.metadata['Name'] = "distr-name"
         self.cmd = upload_docs(self.dist)
 
-    def test_generates_uploaddir_if_not_given(self):
-        self.assertEqual(self.cmd.upload_dir, None)
-        self.cmd.ensure_finalized()
-        self.assertEqual(self.cmd.upload_dir, os.path.join("build", "docs"))
+    def test_default_uploaddir(self):
+        sandbox = tempfile.mkdtemp()
+        previous = os.getcwd()
+        os.chdir(sandbox)
+        try:
+            os.mkdir("build")
+            self.prepare_sample_dir("build")
+            self.cmd.ensure_finalized()
+            self.assertEqual(self.cmd.upload_dir, os.path.join("build", "docs"))
+        finally:
+            os.chdir(previous)
 
-    def prepare_sample_dir(self):
-        sample_dir = tempfile.mkdtemp()
-        os.mkdir(os.path.join(sample_dir, "some_dir"))
-        self.write_file(os.path.join(sample_dir, "some_dir", "index.html"), "Ce mortel ennui")
+    def prepare_sample_dir(self, sample_dir=None):
+        if sample_dir is None:
+            sample_dir = tempfile.mkdtemp()
+        os.mkdir(os.path.join(sample_dir, "docs"))
+        self.write_file(os.path.join(sample_dir, "docs", "index.html"), "Ce mortel ennui")
         self.write_file(os.path.join(sample_dir, "index.html"), "Oh la la")
         return sample_dir
 
@@ -71,7 +79,7 @@
         compressed = zip_dir(source_dir)
 
         zip_f = zipfile.ZipFile(compressed)
-        self.assertEqual(zip_f.namelist(), ['index.html', 'some_dir/index.html'])
+        self.assertEqual(zip_f.namelist(), ['index.html', 'docs/index.html'])
 
     def test_encode_multipart(self):
         fields = [("a", "b"), ("c", "d")]
@@ -81,9 +89,9 @@
         self.assertEqual(body, EXPECTED_MULTIPART_OUTPUT)
 
     def prepare_command(self):
+        self.cmd.upload_dir = self.prepare_sample_dir()
         self.cmd.ensure_finalized()
         self.cmd.repository = self.pypi.full_address
-        self.cmd.upload_dir = self.prepare_sample_dir()
         self.cmd.username = "username"
         self.cmd.password = "password"
 
@@ -109,7 +117,7 @@
         # check their contents
         self.assertIn("doc_upload", action)
         self.assertIn("distr-name", name)
-        self.assertIn("some_dir/index.html", content)
+        self.assertIn("docs/index.html", content)
         self.assertIn("Ce mortel ennui", content)
 
     def test_https_connection(self):
@@ -151,14 +159,36 @@
     def test_reads_pypirc_data(self):
         self.write_file(self.rc, PYPIRC % self.pypi.full_address)
         self.cmd.repository = self.pypi.full_address
+        self.cmd.upload_dir = self.prepare_sample_dir()
         self.cmd.ensure_finalized()
         self.assertEqual(self.cmd.username, "real_slim_shady")
         self.assertEqual(self.cmd.password, "long_island")
 
     def test_checks_index_html_presence(self):
-        self.prepare_command()
+        self.cmd.upload_dir = self.prepare_sample_dir()
         os.remove(os.path.join(self.cmd.upload_dir, "index.html"))
-        self.assertRaises(DistutilsFileError, self.cmd.run)
+        self.assertRaises(DistutilsFileError, self.cmd.ensure_finalized)
+
+    def test_checks_upload_dir(self):
+        self.cmd.upload_dir = self.prepare_sample_dir()
+        shutil.rmtree(os.path.join(self.cmd.upload_dir))
+        self.assertRaises(DistutilsOptionError, self.cmd.ensure_finalized)
+
+    def test_show_response(self):
+        orig_stdout = sys.stdout
+        write_args = []
+        class MockStdIn(object):
+            def write(self, arg):
+                write_args.append(arg)
+        sys.stdout = MockStdIn()
+        try:
+            self.prepare_command()
+            self.cmd.show_response = True
+            self.cmd.run()
+        finally:
+            sys.stdout = orig_stdout
+        self.assertTrue(write_args[0], "should report the response")
+        self.assertIn("\n".join(self.pypi.default_response_data), write_args[0])
 
 def test_suite():
     return unittest2.makeSuite(UploadDocsTestCase)

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: Add a simpler way to use the
	PyPIServer on tests.
Message-ID: <E1OVLoV-0000iW-GZ@dinsdale.python.org>

tarek.ziade pushed 87cc41b3aaca to distutils2:

http://hg.python.org/distutils2/rev/87cc41b3aaca
changeset:   295:87cc41b3aaca
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Tue Jun 01 14:57:53 2010 +0200
summary:     Add a simpler way to use the PyPIServer on tests.
files:       docs/source/test_framework.rst, src/distutils2/tests/pypi_server.py, src/distutils2/tests/test_pypi_server.py, src/distutils2/tests/test_pypi_simple.py

diff --git a/docs/source/test_framework.rst b/docs/source/test_framework.rst
--- a/docs/source/test_framework.rst
+++ b/docs/source/test_framework.rst
@@ -35,18 +35,17 @@
 keeping in mind that the server will alwas try to load paths in reverse order 
 (e.g here, try "another/super/path" then the default one) ::
 
-    server = PyPIServer()
+    server = PyPIServer(test_static_path="another/super/path")
+    server = PyPIServer("another/super/path")
+    # or 
     server.static_filesystem_paths.append("another/super/path")
 
 As a result of what, in your tests, while you need to use the PyPIServer, in
 order to isolates the test cases, the best practice is to place the common files
 in the `default` folder, and to create a directory for each specific test case::
 
-    server = PyPIServer(static_uri_paths=["simple", "external"],
-        static_filesystem_paths = ["default", "test_pypi_server"])
-
-    # or just
-    server = PyPiServer(["simple", "external"], ["default", "test_pypi_server"])
+    server = PyPIServer(static_filesystem_paths = ["default", "test_pypi_server"],
+        static_uri_paths=["simple", "external"])
 
 ``PyPIServerTestCase``
 ======================
diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
--- a/src/distutils2/tests/pypi_server.py
+++ b/src/distutils2/tests/pypi_server.py
@@ -18,8 +18,8 @@
 class PyPIServer(threading.Thread):
     """Thread that wraps a wsgi app"""
     
-    def __init__(self, static_uri_paths=["pypi"],
-            static_filesystem_paths=["default"]):
+    def __init__(self, test_static_path=None, 
+            static_filesystem_paths=["default"], static_uri_paths=["simple"]):
         """Initialize the server.
 
         static_uri_paths and static_base_path are parameters used to provides
@@ -36,6 +36,8 @@
         self.default_response_headers = [('Content-type', 'text/plain')]
         self.default_response_data = ["hello"]
         self.static_uri_paths = static_uri_paths
+        if test_static_path is not None:
+            static_filesystem_paths.append(test_static_path)
         self.static_filesystem_paths = [PYPI_DEFAULT_STATIC_PATH + "/" + path 
             for path in static_filesystem_paths]
 
diff --git a/src/distutils2/tests/test_pypi_server.py b/src/distutils2/tests/test_pypi_server.py
--- a/src/distutils2/tests/test_pypi_server.py
+++ b/src/distutils2/tests/test_pypi_server.py
@@ -37,7 +37,8 @@
                url_path)
             return response.read() == file.read()
 
-        server = PyPIServer(["simple", "external"],["test_pypi_server"])
+        server = PyPIServer(static_uri_paths=["simple", "external"],
+            static_filesystem_paths = ["test_pypi_server"])
         server.start()
         
         # the file does not exists on the disc, so it might not be served
diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -77,6 +77,9 @@
         url = 'file:///tmp/test_simple'
         self.assert_(index.url_ok(url, True))
 
+    def test_found_links(self):
+        server = PyPIServer("test_found_links")
+
     def test_links_priority(self):
         """
         Download links from the pypi simple index should be used before
@@ -93,7 +96,7 @@
         -> Distribute should use the link from pypi, not the external one.
         """
         # start an index server
-        server = PyPIServer(["simple", "external"],["test_link_priority"])
+        server = PyPIServer(None, ["test_link_priority"], ["simple", "external"])
         server.start()
         index_url = server.full_address + '/simple/'
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: Mocked PyPIServer - Some work to
	isolate the tests cases	from each others
Message-ID: <E1OVLoV-0000iN-FN@dinsdale.python.org>

tarek.ziade pushed e29b378fc23a to distutils2:

http://hg.python.org/distutils2/rev/e29b378fc23a
changeset:   293:e29b378fc23a
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Tue Jun 01 12:15:53 2010 +0200
summary:     Mocked PyPIServer - Some work to isolate the tests cases from each others
files:       docs/source/test_framework.rst, src/distutils2/tests/pypi_server.py, src/distutils2/tests/pypiserver/external/index.html, src/distutils2/tests/pypiserver/simple/index.html, src/distutils2/tests/pypiserver/test_pypi_server/external/index.html, src/distutils2/tests/pypiserver/test_pypi_server/simple/index.html, src/distutils2/tests/test_pypi_server.py

diff --git a/docs/source/test_framework.rst b/docs/source/test_framework.rst
--- a/docs/source/test_framework.rst
+++ b/docs/source/test_framework.rst
@@ -23,14 +23,17 @@
 
     server = PyPIServer(static_uri_paths=["first_path", "second_path"])
 
-You need to create the content that will be served under the `/test/pypiserver`
-path. If you want to serve content from another place, you also can specify
-another filesystem path::
+You need to create the content that will be served under the 
+`/tests/pypiserver/default` path. If you want to serve content from another 
+place, you also can specify another filesystem path (wich need to be under
+`tests/pypiserver/`. This will replace the default behavior of the server, and
+it will not serve content from the `default` dir ::
 
     server = PyPIServer(static_filesystem_paths=["path/to/your/dir"])
 
-Or add some paths to the existing ones, the server will alwas try to load paths
-in reverse order (e.g here, try "another/super/path" then the default one) ::
+If you just need to add some paths to the existing ones, you can do as shown, 
+keeping in mind that the server will alwas try to load paths in reverse order 
+(e.g here, try "another/super/path" then the default one) ::
 
     server = PyPIServer()
     server.static_filesystem_paths.append("another/super/path")
diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
--- a/src/distutils2/tests/pypi_server.py
+++ b/src/distutils2/tests/pypi_server.py
@@ -19,7 +19,7 @@
     """Thread that wraps a wsgi app"""
     
     def __init__(self, static_uri_paths=["pypi"],
-            static_filesystem_paths=[PYPI_DEFAULT_STATIC_PATH]):
+            static_filesystem_paths=["default"]):
         """Initialize the server.
 
         static_uri_paths and static_base_path are parameters used to provides
@@ -36,7 +36,8 @@
         self.default_response_headers = [('Content-type', 'text/plain')]
         self.default_response_data = ["hello"]
         self.static_uri_paths = static_uri_paths
-        self.static_filesystem_paths = static_filesystem_paths
+        self.static_filesystem_paths = [PYPI_DEFAULT_STATIC_PATH + "/" + path 
+            for path in static_filesystem_paths]
 
     def run(self):
         self.httpd.serve_forever()
@@ -76,6 +77,7 @@
             fs_paths.reverse()
             for fs_path in fs_paths:
                 try:
+                    print fs_path + relative_path
                     file = open(fs_path + relative_path)
                     data = file.read()
                     start_response("200 OK", [('Content-type', 'text/plain')])
diff --git a/src/distutils2/tests/pypiserver/external/index.html b/src/distutils2/tests/pypiserver/external/index.html
deleted file mode 100644
--- a/src/distutils2/tests/pypiserver/external/index.html
+++ /dev/null
@@ -1,1 +0,0 @@
-index.html from external server
diff --git a/src/distutils2/tests/pypiserver/test_pypi_server/external/index.html b/src/distutils2/tests/pypiserver/test_pypi_server/external/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/test_pypi_server/external/index.html
@@ -0,0 +1,1 @@
+index.html from external server
diff --git a/src/distutils2/tests/pypiserver/simple/index.html b/src/distutils2/tests/pypiserver/test_pypi_server/simple/index.html
rename from src/distutils2/tests/pypiserver/simple/index.html
rename to src/distutils2/tests/pypiserver/test_pypi_server/simple/index.html
diff --git a/src/distutils2/tests/test_pypi_server.py b/src/distutils2/tests/test_pypi_server.py
--- a/src/distutils2/tests/test_pypi_server.py
+++ b/src/distutils2/tests/test_pypi_server.py
@@ -33,10 +33,12 @@
             url = server.full_address + url_path 
             request = urllib2.Request(url)
             response = urllib2.urlopen(request)
-            file = open(PYPI_DEFAULT_STATIC_PATH + url_path)
+            file = open(PYPI_DEFAULT_STATIC_PATH + "/test_pypi_server" + 
+               url_path)
             return response.read() == file.read()
 
-        server = PyPIServer(static_uri_paths=["simple", "external"])
+        server = PyPIServer(static_uri_paths=["simple", "external"],
+            static_filesystem_paths = ["test_pypi_server"])
         server.start()
         
         # the file does not exists on the disc, so it might not be served

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: Replace wsgiref by BaseHTTPServer to
	be python 2.4	compatible.
Message-ID: <E1OVLoV-0000ie-Hv@dinsdale.python.org>

tarek.ziade pushed d4fe9695bc65 to distutils2:

http://hg.python.org/distutils2/rev/d4fe9695bc65
changeset:   297:d4fe9695bc65
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Thu Jun 03 17:44:59 2010 +0200
summary:     Replace wsgiref by BaseHTTPServer to be python 2.4 compatible.
files:       src/distutils2/tests/pypi_server.py, src/distutils2/tests/test_pypi_server.py

diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
--- a/src/distutils2/tests/pypi_server.py
+++ b/src/distutils2/tests/pypi_server.py
@@ -10,6 +10,8 @@
 import time
 import unittest2
 import urllib2
+from BaseHTTPServer import HTTPServer
+from SimpleHTTPServer import SimpleHTTPRequestHandler
 from wsgiref.simple_server import make_server
 import os.path
 
@@ -39,7 +41,6 @@
         super(PyPIServerTestCase, self).tearDown()
         self.pypi.stop()
 
-
 class PyPIServer(threading.Thread):
     """PyPI Mocked server.
     Provides a mocked version of the PyPI API's, to ease tests.
@@ -57,14 +58,15 @@
         """
         threading.Thread.__init__(self)
         self._run = True
-        self.httpd = make_server('', 0, self.pypi_app)
+        self.httpd = HTTPServer(('', 0), PyPIRequestHandler) 
         self.httpd.RequestHandlerClass.log_request = lambda *_: None
+        self.httpd.RequestHandlerClass.pypi_server = self
         self.address = self.httpd.server_address
         self.request_queue = Queue.Queue()
         self._requests = []
-        self.default_response_status = "200 OK"
+        self.default_response_status = 200
         self.default_response_headers = [('Content-type', 'text/plain')]
-        self.default_response_data = ["hello"]
+        self.default_response_data = "hello"
         self.static_uri_paths = static_uri_paths
         if test_static_path is not None:
             static_filesystem_paths.append(test_static_path)
@@ -82,59 +84,6 @@
         """self shutdown is not supported for python < 2.6"""
         self._run = False
 
-    def pypi_app(self, environ, start_response):
-        """Serve the content.
-
-        Also record the requests to be accessed later. If trying to access an
-        url matching a static uri, serve static content, otherwise serve
-        what is provided by the `get_next_response` method.
-        """
-        # record the request. Read the input only on PUT or POST requests
-        if environ["REQUEST_METHOD"] in ("PUT", "POST"):
-            if environ.get("CONTENT_LENGTH"):
-                request_data = environ.pop('wsgi.input').read(
-                    int(environ['CONTENT_LENGTH']))
-            else:
-                request_data = environ.pop('wsgi.input').read()
-        elif environ["REQUEST_METHOD"] in ("GET", "DELETE"):
-            request_data = ''
-
-        self.request_queue.put((environ, request_data))
-
-        # serve the content from local disc if we request an URL beginning
-        # by a pattern defined in `static_paths`
-        relative_path = environ["PATH_INFO"].replace(self.full_address, '')
-        url_parts = relative_path.split("/")
-        if len(url_parts) > 1 and url_parts[1] in self.static_uri_paths:
-            data = None
-            # always take the last first.
-            fs_paths = []
-            fs_paths.extend(self.static_filesystem_paths)
-            fs_paths.reverse()
-            for fs_path in fs_paths:
-                try:
-                    if relative_path.endswith("/"):
-                        relative_path += "index.html"
-                    print fs_path + relative_path
-                    file = open(fs_path + relative_path)
-                    data = file.read()
-                    start_response("200 OK", [('Content-type', 'text/html')])
-                except IOError:
-                    pass
-
-            if data is None:
-                start_response("404 NOT FOUND",
-                    [('Content-type', 'text/html')])
-                data = "Not Found"
-            return data
-
-        # otherwise serve the content from get_next_response
-        else:
-            # send back a response
-            status, headers, data = self.get_next_response()
-            start_response(status, headers)
-            return data
-
     def get_next_response(self):
         return (self.default_response_status,
                 self.default_response_headers,
@@ -155,3 +104,83 @@
     @property
     def full_address(self):
         return "http://%s:%s" % self.address
+
+
+class PyPIRequestHandler(SimpleHTTPRequestHandler):
+    # we need to access the pypi server while serving the content
+    pypi_server = None
+
+    def do_POST(self):
+        return self.serve_request()
+    def do_GET(self):
+        return self.serve_request()
+    def do_DELETE(self):
+        return self.serve_request()
+    def do_PUT(self):
+        return self.serve_request()
+
+    def serve_request(self):
+        """Serve the content.
+
+        Also record the requests to be accessed later. If trying to access an
+        url matching a static uri, serve static content, otherwise serve
+        what is provided by the `get_next_response` method.
+        """
+        # record the request. Read the input only on PUT or POST requests
+        if self.command in ("PUT", "POST"):
+            if self.headers.dict.has_key("content-length"):
+                request_data = self.rfile.read(
+                    int(self.headers['content-length']))
+            else:
+                request_data = self.rfile.read()
+        elif self.command in ("GET", "DELETE"):
+            request_data = ''
+
+        self.pypi_server.request_queue.put((self, request_data))
+
+        # serve the content from local disc if we request an URL beginning
+        # by a pattern defined in `static_paths`
+        url_parts = self.path.split("/")
+        if (len(url_parts) > 1 and 
+                url_parts[1] in self.pypi_server.static_uri_paths):
+            data = None
+            # always take the last first.
+            fs_paths = []
+            fs_paths.extend(self.pypi_server.static_filesystem_paths)
+            fs_paths.reverse()
+            relative_path = self.path
+            for fs_path in fs_paths:
+                try:
+                    if self.path.endswith("/"):
+                        relative_path += "index.html"
+                    file = open(fs_path + relative_path)
+                    data = file.read()
+                    self.make_response(data)
+                except IOError:
+                    pass
+
+            if data is None:
+                self.make_response("Not found", 404)
+
+        # otherwise serve the content from get_next_response
+        else:
+            # send back a response
+            status, headers, data = self.pypi_server.get_next_response()
+            self.make_response(data, status, headers)
+
+    def make_response(self, data, status=200, 
+            headers=[('Content-type', 'text/html')]):
+        """Send the response to the HTTP client"""
+        if not isinstance(status, int):
+            try:
+                status = int(status)
+            except ValueError:
+                # we probably got something like YYY Codename. 
+                # Just get the first 3 digits
+                status = int(status[:3]) 
+
+        self.send_response(status)
+        for header, value in headers:
+            self.send_header(header, value)
+        self.end_headers()
+        self.wfile.write(data)
diff --git a/src/distutils2/tests/test_pypi_server.py b/src/distutils2/tests/test_pypi_server.py
--- a/src/distutils2/tests/test_pypi_server.py
+++ b/src/distutils2/tests/test_pypi_server.py
@@ -21,9 +21,11 @@
         request = urllib2.Request(server.full_address, data, headers)
         urllib2.urlopen(request)
         self.assertEqual(len(server.requests), 1)
-        environ, request_data = server.requests[-1]
+        handler, request_data = server.requests[-1]
         self.assertIn("Rock Around The Bunker", request_data)
-        self.assertEqual(environ["HTTP_X_TEST_HEADER"], "Mister Iceberg")
+        self.assertTrue(handler.headers.dict.has_key("x-test-header"))
+        self.assertEqual(handler.headers.dict["x-test-header"], 
+            "Mister Iceberg")
         server.stop()
 
     def test_serve_static_content(self):

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: removed pypi_server's blocking
	behaviour (mostly)
Message-ID: <E1OVLoV-0000iu-Jz@dinsdale.python.org>

tarek.ziade pushed e415685da4c5 to distutils2:

http://hg.python.org/distutils2/rev/e415685da4c5
changeset:   301:e415685da4c5
user:        Konrad Delong <konryd at gmail.com>
date:        Sat Jun 05 12:51:51 2010 +0200
summary:     removed pypi_server's blocking behaviour (mostly)
files:       src/distutils2/tests/pypi_server.py, src/distutils2/tests/test_pypi_server.py

diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
--- a/src/distutils2/tests/pypi_server.py
+++ b/src/distutils2/tests/pypi_server.py
@@ -12,8 +12,8 @@
 import urllib2
 from BaseHTTPServer import HTTPServer
 from SimpleHTTPServer import SimpleHTTPRequestHandler
-from wsgiref.simple_server import make_server
 import os.path
+import select
 
 PYPI_DEFAULT_STATIC_PATH = os.path.dirname(os.path.abspath(__file__)) + "/pypiserver"
 
@@ -61,7 +61,7 @@
         self.httpd = HTTPServer(('', 0), PyPIRequestHandler) 
         self.httpd.RequestHandlerClass.log_request = lambda *_: None
         self.httpd.RequestHandlerClass.pypi_server = self
-        self.address = self.httpd.server_address
+        self.address = (self.httpd.server_name, self.httpd.server_port)
         self.request_queue = Queue.Queue()
         self._requests = []
         self.default_response_status = 200
@@ -77,10 +77,10 @@
 
     def run(self):
         # loop because we can't stop it otherwise, for python < 2.6
-        while True:
-            self.httpd.handle_request()
-            if not self._run:
-                break
+        while self._run:
+            r, w, e = select.select([self.httpd], [], [], 0.5)
+            if r:
+                self.httpd.handle_request()
 
     def stop(self):
         """self shutdown is not supported for python < 2.6"""
diff --git a/src/distutils2/tests/test_pypi_server.py b/src/distutils2/tests/test_pypi_server.py
--- a/src/distutils2/tests/test_pypi_server.py
+++ b/src/distutils2/tests/test_pypi_server.py
@@ -60,6 +60,7 @@
 
         # and another one in another root path
         self.assertTrue(uses_local_files_for(server, "/external/index.html"))
+        server.stop()
 
 
 def test_suite():

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: Make a decorator for PyPIServer mock,
	and starting some tests
Message-ID: <E1OVLoV-0000ia-HJ@dinsdale.python.org>

tarek.ziade pushed e467c13bf12b to distutils2:

http://hg.python.org/distutils2/rev/e467c13bf12b
changeset:   296:e467c13bf12b
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Tue Jun 01 19:30:32 2010 +0200
summary:     Make a decorator for PyPIServer mock, and starting some tests
files:       docs/source/test_framework.rst, src/distutils2/command/upload.py, src/distutils2/tests/pypi_server.py, src/distutils2/tests/pypiserver/test_found_links/simple/foobar/index.html, src/distutils2/tests/pypiserver/test_found_links/simple/index.html, src/distutils2/tests/test_pypi_server.py, src/distutils2/tests/test_pypi_simple.py

diff --git a/docs/source/test_framework.rst b/docs/source/test_framework.rst
--- a/docs/source/test_framework.rst
+++ b/docs/source/test_framework.rst
@@ -63,3 +63,26 @@
 
           environ, request_data = self.pypi.requests[-1]
           self.assertEqual(request_data, EXPECTED_REQUEST_DATA)
+
+The ``use_pypi_server`` decorator
+=================================
+
+You also can use a decorator for your tests, if you do not need the same server
+instance along all you test case. So, you can specify, for each test method,
+some initialisation parameters for the server.
+
+For this, you need to add a `server` parameter to your method, like this::
+
+    class SampleTestCase(TestCase):
+        @use_pypi_server()
+        def test_somthing(self, server):
+            # your tests goes here
+
+The decorator will instanciate the server for you, and run and stop it just
+before and after your method call. You also can pass the server initializer,
+just like this::
+
+    class SampleTestCase(TestCase):
+        @use_pypi_server("test_case_name")
+        def test_something(self, server):
+            # something
diff --git a/src/distutils2/command/upload.py b/src/distutils2/command/upload.py
--- a/src/distutils2/command/upload.py
+++ b/src/distutils2/command/upload.py
@@ -160,7 +160,7 @@
         # send the data
         try:
             result = urlopen(request)
-            status = result.getcode()
+            status = result.code
             reason = result.msg
         except socket.error, e:
             self.announce(str(e), log.ERROR)
diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
--- a/src/distutils2/tests/pypi_server.py
+++ b/src/distutils2/tests/pypi_server.py
@@ -1,8 +1,32 @@
-import Queue, threading, time, unittest2
+"""Mocked PyPI Server implementation, to use in tests.
+
+This module also provides a simple test case to extend if you need to use
+the PyPIServer all along your test case. Be sure to read the documentation 
+before any use.
+"""
+
+import Queue
+import threading
+import time
+import unittest2
+import urllib2
 from wsgiref.simple_server import make_server
 import os.path
 
-PYPI_DEFAULT_STATIC_PATH =  os.path.dirname(os.path.abspath(__file__)) + "/pypiserver"
+PYPI_DEFAULT_STATIC_PATH = os.path.dirname(os.path.abspath(__file__)) + "/pypiserver"
+
+def use_pypi_server(*server_args, **server_kwargs):
+    """Decorator to make use of the PyPIServer for test methods, 
+    just when needed, and not for the entire duration of the testcase.
+    """
+    def wrapper(func):
+        def wrapped(*args, **kwargs):
+            server = PyPIServer(*server_args, **server_kwargs)
+            server.start()
+            func(server=server, *args, **kwargs)
+            server.stop()
+        return wrapped
+    return wrapper 
 
 class PyPIServerTestCase(unittest2.TestCase):
 
@@ -15,10 +39,15 @@
         super(PyPIServerTestCase, self).tearDown()
         self.pypi.stop()
 
+
 class PyPIServer(threading.Thread):
-    """Thread that wraps a wsgi app"""
-    
-    def __init__(self, test_static_path=None, 
+    """PyPI Mocked server.
+    Provides a mocked version of the PyPI API's, to ease tests.
+
+    Support serving static content and serving previously given text.
+    """
+
+    def __init__(self, test_static_path=None,
             static_filesystem_paths=["default"], static_uri_paths=["simple"]):
         """Initialize the server.
 
@@ -27,6 +56,7 @@
         matching files on the filesystem.
         """
         threading.Thread.__init__(self)
+        self._run = True
         self.httpd = make_server('', 0, self.pypi_app)
         self.httpd.RequestHandlerClass.log_request = lambda *_: None
         self.address = self.httpd.server_address
@@ -38,21 +68,25 @@
         self.static_uri_paths = static_uri_paths
         if test_static_path is not None:
             static_filesystem_paths.append(test_static_path)
-        self.static_filesystem_paths = [PYPI_DEFAULT_STATIC_PATH + "/" + path 
+        self.static_filesystem_paths = [PYPI_DEFAULT_STATIC_PATH + "/" + path
             for path in static_filesystem_paths]
 
     def run(self):
-        self.httpd.serve_forever()
+        # loop because we can't stop it otherwise, for python < 2.6
+        while True:
+            self.httpd.handle_request()
+            if not self._run:
+                break
 
     def stop(self):
-        self.httpd.shutdown()
-        self.join()
+        """self shutdown is not supported for python < 2.6"""
+        self._run = False
 
     def pypi_app(self, environ, start_response):
         """Serve the content.
 
         Also record the requests to be accessed later. If trying to access an
-        url matching a static uri, serve static content, otherwise serve 
+        url matching a static uri, serve static content, otherwise serve
         what is provided by the `get_next_response` method.
         """
         # record the request. Read the input only on PUT or POST requests
@@ -64,10 +98,10 @@
                 request_data = environ.pop('wsgi.input').read()
         elif environ["REQUEST_METHOD"] in ("GET", "DELETE"):
             request_data = ''
-            
+
         self.request_queue.put((environ, request_data))
-        
-        # serve the content from local disc if we request an URL beginning 
+
+        # serve the content from local disc if we request an URL beginning
         # by a pattern defined in `static_paths`
         relative_path = environ["PATH_INFO"].replace(self.full_address, '')
         url_parts = relative_path.split("/")
@@ -87,9 +121,10 @@
                     start_response("200 OK", [('Content-type', 'text/html')])
                 except IOError:
                     pass
-            
+
             if data is None:
-                start_response("404 NOT FOUND", [('Content-type', 'text/html')])
+                start_response("404 NOT FOUND",
+                    [('Content-type', 'text/html')])
                 data = "Not Found"
             return data
 
@@ -107,6 +142,9 @@
 
     @property
     def requests(self):
+        """Use this property to get all requests that have been made
+        to the server
+        """
         while True:
             try:
                 self._requests.append(self.request_queue.get_nowait())
diff --git a/src/distutils2/tests/pypiserver/test_found_links/simple/foobar/index.html b/src/distutils2/tests/pypiserver/test_found_links/simple/foobar/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/test_found_links/simple/foobar/index.html
@@ -0,0 +1,6 @@
+<html><head><title>Links for Foobar</title></head><body><h1>Links for Foobar</h1>
+<a href="../../packages/source/F/Foobar/Foobar-1.0.tar.gz#md5=98fa833fdabcdd78d00245aead66c174">Foobar-1.0.tar.gz</a><br/> 
+<a href="../../packages/source/F/Foobar/Foobar-1.0.1.tar.gz#md5=2351efb20f6b7b5d9ce80fa4cb1bd9ca">Foobar-1.0.1.tar.gz</a><br/> 
+<a href="../../packages/source/F/Foobar/Foobar-2.0.tar.gz#md5=98fa833fdabcdd78d00245aead66c274">Foobar-2.0.tar.gz</a><br/> 
+<a href="../../packages/source/F/Foobar/Foobar-2.0.1.tar.gz#md5=2352efb20f6b7b5d9ce80fa4cb2bd9ca">Foobar-2.0.1.tar.gz</a><br/> 
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/test_found_links/simple/index.html b/src/distutils2/tests/pypiserver/test_found_links/simple/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/test_found_links/simple/index.html
@@ -0,0 +1,1 @@
+<a href="foobar/">foobar/</a> 
diff --git a/src/distutils2/tests/test_pypi_server.py b/src/distutils2/tests/test_pypi_server.py
--- a/src/distutils2/tests/test_pypi_server.py
+++ b/src/distutils2/tests/test_pypi_server.py
@@ -1,7 +1,11 @@
 """Tests for distutils.command.bdist."""
-import unittest2, urllib, urllib2
+import unittest2
+import urllib
+import urllib2
+import os.path
+
 from distutils2.tests.pypi_server import PyPIServer, PYPI_DEFAULT_STATIC_PATH
-import os.path
+
 
 class PyPIServerTest(unittest2.TestCase):
 
@@ -30,24 +34,24 @@
             """Test that files are served statically (eg. the output from the
             server is the same than the one made by a simple file read.
             """
-            url = server.full_address + url_path 
+            url = server.full_address + url_path
             request = urllib2.Request(url)
             response = urllib2.urlopen(request)
-            file = open(PYPI_DEFAULT_STATIC_PATH + "/test_pypi_server" + 
+            file = open(PYPI_DEFAULT_STATIC_PATH + "/test_pypi_server" +
                url_path)
             return response.read() == file.read()
 
         server = PyPIServer(static_uri_paths=["simple", "external"],
-            static_filesystem_paths = ["test_pypi_server"])
+            static_filesystem_paths=["test_pypi_server"])
         server.start()
-        
+
         # the file does not exists on the disc, so it might not be served
         url = server.full_address + "/simple/unexisting_page"
         request = urllib2.Request(url)
         try:
             urllib2.urlopen(request)
-        except urllib2.HTTPError as e:
-            self.assertEqual(e.getcode(), 404)
+        except urllib2.HTTPError,e:
+            self.assertEqual(e.code, 404)
 
         # now try serving a content that do exists
         self.assertTrue(uses_local_files_for(server, "/simple/index.html"))
@@ -55,6 +59,7 @@
         # and another one in another root path
         self.assertTrue(uses_local_files_for(server, "/external/index.html"))
 
+
 def test_suite():
     return unittest2.makeSuite(PyPIServerTest)
 
diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -3,14 +3,15 @@
 """
 import sys
 import os
-import shutil 
-import tempfile 
+import shutil
+import tempfile
 import unittest2
 import urllib2
 
-from distutils2.tests.pypi_server import PyPIServer
+from distutils2.tests.pypi_server import use_pypi_server
 from distutils2.pypi import simple
 
+
 class PyPISimpleTestCase(unittest2.TestCase):
 
     def test_bad_urls(self):
@@ -19,24 +20,23 @@
         try:
             v = index.open_url(url)
         except Exception, v:
-            self.assert_(url in str(v))
+            self.assertTrue(url in str(v))
         else:
-            self.assert_(isinstance(v,urllib2.HTTPError))
+            self.assertTrue(isinstance(v, urllib2.HTTPError))
 
         # issue 16
         # easy_install inquant.contentmirror.plone breaks because of a typo
         # in its home URL
         index = simple.PackageIndex(
-            hosts=('www.example.com',)
-        )
+            hosts=('www.example.com',))
 
         url = 'url:%20https://svn.plone.org/svn/collective/inquant.contentmirror.plone/trunk'
         try:
             v = index.open_url(url)
         except Exception, v:
-            self.assert_(url in str(v))
+            self.assertTrue(url in str(v))
         else:
-            self.assert_(isinstance(v, urllib2.HTTPError))
+            self.assertTrue(isinstance(v, urllib2.HTTPError))
 
         def _urlopen(*args):
             import httplib
@@ -49,7 +49,7 @@
             try:
                 v = index.open_url(url)
             except Exception, v:
-                self.assert_('line' in str(v))
+                self.assertTrue('line' in str(v))
             else:
                 raise AssertionError('Should have raise here!')
         finally:
@@ -60,7 +60,7 @@
         try:
             index.open_url(url)
         except Exception, v:
-            self.assert_('nonnumeric port' in str(v))
+            self.assertTrue('nonnumeric port' in str(v))
 
         # issue #160
         if sys.version_info[0] == 2 and sys.version_info[1] == 7:
@@ -72,15 +72,41 @@
 
     def test_url_ok(self):
         index = simple.PackageIndex(
-            hosts=('www.example.com',)
-        )
+            hosts=('www.example.com',))
         url = 'file:///tmp/test_simple'
-        self.assert_(index.url_ok(url, True))
+        self.assertTrue(index.url_ok(url, True))
+    
+    @use_pypi_server("test_found_links")
+    def test_found_links(self, server):
+        """Browse the index, asking for a specified distribution version
+        """
+        # The PyPI index contains links for version 1.0, 1.1, 2.0 and 2.0.1
+        # We query only for the version 1.1, so all distributions must me
+        # filled in the package_index (as the url has been scanned), but
+        # obtain must only return the one we want.
+        pi = simple.PackageIndex(server.full_address + "/simple/")
+        last_distribution = pi.obtain(simple.Requirement.parse("foobar"))
 
-    def test_found_links(self):
-        server = PyPIServer("test_found_links")
+        # we have scanned the index page
+        self.assertTrue(pi.package_pages.has_key('foobar'))
+        self.assertIn(server.full_address + "/simple/foobar/",
+            pi.package_pages['foobar'])
+        
+        # we have found 4 distributions in this page
+        self.assertEqual(len(pi["foobar"]), 4)
 
-    def test_links_priority(self):
+        # and returned the most recent one
+        self.assertEqual(last_distribution.version, '2.0.1')
+
+    def test_process_external_pages(self):
+        """If the index provides links external to the pypi index, 
+        they need to be processed, in order to discover new distributions.
+        """
+        pass
+    
+    @use_pypi_server(static_filesystem_paths=["test_link_priority"],
+        static_uri_paths=["simple", "external"])
+    def test_links_priority(self, server):
         """
         Download links from the pypi simple index should be used before
         external download links.
@@ -96,8 +122,6 @@
         -> Distribute should use the link from pypi, not the external one.
         """
         # start an index server
-        server = PyPIServer(None, ["test_link_priority"], ["simple", "external"])
-        server.start()
         index_url = server.full_address + '/simple/'
 
         # scan a test index
@@ -107,11 +131,12 @@
         server.stop()
 
         # the distribution has been found
-        self.assert_('foobar' in pi)
+        self.assertTrue('foobar' in pi)
         # we have only one link, because links are compared without md5
-        self.assert_(len(pi['foobar'])==1)
+        self.assertEqual(len(pi['foobar']), 1)
         # the link should be from the index
-        self.assert_('correct_md5' in pi['foobar'][0].location)
+        self.assertTrue('correct_md5' in pi['foobar'][0].location)
+
 
 def test_suite():
     return unittest2.makeSuite(PyPISimpleTestCase)

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: Make existing tests working with the
	new mocked PyPI	server.
Message-ID: <E1OVLoV-0000iS-G2@dinsdale.python.org>

tarek.ziade pushed a611bae3505c to distutils2:

http://hg.python.org/distutils2/rev/a611bae3505c
changeset:   294:a611bae3505c
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Tue Jun 01 14:31:52 2010 +0200
summary:     Make existing tests working with the new mocked PyPI server.
files:       docs/source/test_framework.rst, src/distutils2/pypi/simple.py, src/distutils2/tests/pypi_server.py, src/distutils2/tests/pypiserver/test_link_priority/external/external.html, src/distutils2/tests/pypiserver/test_link_priority/simple/foobar/index.html, src/distutils2/tests/pypiserver/test_link_priority/simple/index.html, src/distutils2/tests/test_pypi_server.py, src/distutils2/tests/test_pypi_simple.py

diff --git a/docs/source/test_framework.rst b/docs/source/test_framework.rst
--- a/docs/source/test_framework.rst
+++ b/docs/source/test_framework.rst
@@ -38,6 +38,16 @@
     server = PyPIServer()
     server.static_filesystem_paths.append("another/super/path")
 
+As a result of what, in your tests, while you need to use the PyPIServer, in
+order to isolates the test cases, the best practice is to place the common files
+in the `default` folder, and to create a directory for each specific test case::
+
+    server = PyPIServer(static_uri_paths=["simple", "external"],
+        static_filesystem_paths = ["default", "test_pypi_server"])
+
+    # or just
+    server = PyPiServer(["simple", "external"], ["default", "test_pypi_server"])
+
 ``PyPIServerTestCase``
 ======================
 
diff --git a/src/distutils2/pypi/simple.py b/src/distutils2/pypi/simple.py
--- a/src/distutils2/pypi/simple.py
+++ b/src/distutils2/pypi/simple.py
@@ -238,8 +238,6 @@
             link = urlparse.urljoin(base, htmldecode(match.group(1)))
             self.process_url(link)
         if url.startswith(self.index_url) and getattr(f,'code',None)!=404:
-            from ipdb import set_trace
-            set_trace()
             page = self.process_index(url, page)
 
     def process_filename(self, fn, nested=False):
diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
--- a/src/distutils2/tests/pypi_server.py
+++ b/src/distutils2/tests/pypi_server.py
@@ -77,15 +77,17 @@
             fs_paths.reverse()
             for fs_path in fs_paths:
                 try:
+                    if relative_path.endswith("/"):
+                        relative_path += "index.html"
                     print fs_path + relative_path
                     file = open(fs_path + relative_path)
                     data = file.read()
-                    start_response("200 OK", [('Content-type', 'text/plain')])
+                    start_response("200 OK", [('Content-type', 'text/html')])
                 except IOError:
                     pass
             
             if data is None:
-                start_response("404 NOT FOUND", [('Content-type', 'text/plain')])
+                start_response("404 NOT FOUND", [('Content-type', 'text/html')])
                 data = "Not Found"
             return data
 
diff --git a/src/distutils2/tests/pypiserver/test_link_priority/external/external.html b/src/distutils2/tests/pypiserver/test_link_priority/external/external.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/test_link_priority/external/external.html
@@ -0,0 +1,3 @@
+<html><body>
+<a href="/foobar-0.1.tar.gz#md5=1__bad_md5___">bad old link</a>
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/test_link_priority/simple/foobar/index.html b/src/distutils2/tests/pypiserver/test_link_priority/simple/foobar/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/test_link_priority/simple/foobar/index.html
@@ -0,0 +1,4 @@
+<html><body>
+<a href="/foobar-0.1.tar.gz#md5=0_correct_md5">foobar-0.1.tar.gz</a><br/>
+<a href="../../external/external.html" rel="homepage">external homepage</a><br/>
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/test_link_priority/simple/index.html b/src/distutils2/tests/pypiserver/test_link_priority/simple/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/test_link_priority/simple/index.html
@@ -0,0 +1,1 @@
+<a href="foobar/">foobar/</a> 
diff --git a/src/distutils2/tests/test_pypi_server.py b/src/distutils2/tests/test_pypi_server.py
--- a/src/distutils2/tests/test_pypi_server.py
+++ b/src/distutils2/tests/test_pypi_server.py
@@ -37,8 +37,7 @@
                url_path)
             return response.read() == file.read()
 
-        server = PyPIServer(static_uri_paths=["simple", "external"],
-            static_filesystem_paths = ["test_pypi_server"])
+        server = PyPIServer(["simple", "external"],["test_pypi_server"])
         server.start()
         
         # the file does not exists on the disc, so it might not be served
diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -11,10 +11,7 @@
 from distutils2.tests.pypi_server import PyPIServer
 from distutils2.pypi import simple
 
-class NoneTestCase(unittest2.TestCase):
-    pass
-
-class PackageIndexTestCase(unittest2.TestCase):
+class PyPISimpleTestCase(unittest2.TestCase):
 
     def test_bad_urls(self):
         index = simple.PackageIndex()
@@ -96,9 +93,9 @@
         -> Distribute should use the link from pypi, not the external one.
         """
         # start an index server
-        server = PyPIServer()
+        server = PyPIServer(["simple", "external"],["test_link_priority"])
         server.start()
-        index_url = server.full_address + 'test_links_priority/simple/'
+        index_url = server.full_address + '/simple/'
 
         # scan a test index
         pi = simple.PackageIndex(index_url)
@@ -114,7 +111,7 @@
         self.assert_('correct_md5' in pi['foobar'][0].location)
 
 def test_suite():
-    return unittest2.makeSuite(NoneTestCase)
+    return unittest2.makeSuite(PyPISimpleTestCase)
 
 if __name__ == '__main__':
     unittest2.main(defaultTest="test_suite")

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: PyPIServer can now serve static files
	from multiple	directories.
Message-ID: <E1OVLoV-0000iG-Dt@dinsdale.python.org>

tarek.ziade pushed 7eaab6fefd5a to distutils2:

http://hg.python.org/distutils2/rev/7eaab6fefd5a
changeset:   291:7eaab6fefd5a
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Tue Jun 01 10:53:15 2010 +0200
summary:     PyPIServer can now serve static files from multiple directories.
files:       docs/source/test_framework.rst, src/distutils2/tests/pypi_server.py, src/distutils2/tests/test_pypi_server.py

diff --git a/docs/source/test_framework.rst b/docs/source/test_framework.rst
--- a/docs/source/test_framework.rst
+++ b/docs/source/test_framework.rst
@@ -14,9 +14,20 @@
 HTTP response can be overriden with the ``default_response_status``,
 ``default_response_headers`` and ``default_response_data`` attributes.
 
-When accessing the server with urls beginning with `/simple/`, the server also
-record your requests, but will look for files under the /tests/pypi_server/
-path.
+By default, when accessing the server with urls beginning with `/simple/`, 
+the server also record your requests, but will look for files under 
+the `/tests/pypiserver/simple/` path.
+
+You can tell the sever to serve static files for other paths. This could be 
+accomplished by using the `static_uri_paths` parameter, as below::
+
+    server = PyPIServer(static_uri_paths=["first_path", "second_path"])
+
+You need to create the content that will be served under the `/test/pypiserver`
+path. If you want to serve content from another place, you also can specify
+another filesystem path::
+
+    server = PyPIServer(static_filesystem_paths=["path/to/your/dir"])
 
 ``PyPIServerTestCase``
 ======================
diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
--- a/src/distutils2/tests/pypi_server.py
+++ b/src/distutils2/tests/pypi_server.py
@@ -19,7 +19,7 @@
     """Thread that wraps a wsgi app"""
     
     def __init__(self, static_uri_paths=["pypi"],
-            static_filesystem_path=PYPI_DEFAULT_STATIC_PATH):
+            static_filesystem_paths=[PYPI_DEFAULT_STATIC_PATH]):
         """Initialize the server.
 
         static_uri_paths and static_base_path are parameters used to provides
@@ -36,7 +36,7 @@
         self.default_response_headers = [('Content-type', 'text/plain')]
         self.default_response_data = ["hello"]
         self._static_uri_paths = static_uri_paths
-        self._static_filesystem_path = static_filesystem_path
+        self._static_filesystem_paths = static_filesystem_paths
 
     def run(self):
         self.httpd.serve_forever()
@@ -49,7 +49,7 @@
         """Serve the content.
 
         Also record the requests to be accessed later. If trying to access an
-        url matching `_static_paths`, serve static content, otherwise serve 
+        url matching a static uri, serve static content, otherwise serve 
         what is provided by the `get_next_response` method.
         """
         # record the request. Read the input only on PUT or POST requests
@@ -69,12 +69,16 @@
         relative_path = environ["PATH_INFO"].replace(self.full_address, '')
         url_parts = relative_path.split("/")
         if len(url_parts) > 1 and url_parts[1] in self._static_uri_paths:
-            file_to_serve = self._static_filesystem_path + relative_path
-            try:
-                file = open(file_to_serve)
-                data = file.read()
-                start_response("200 OK", [('Content-type', 'text/plain')])
-            except IOError:
+            data = None
+            for fs_path in self._static_filesystem_paths:
+                try:
+                    file = open(fs_path + relative_path)
+                    data = file.read()
+                    start_response("200 OK", [('Content-type', 'text/plain')])
+                except IOError:
+                    pass
+            
+            if data is None:
                 start_response("404 NOT FOUND", [('Content-type', 'text/plain')])
                 data = "Not Found"
             return data
diff --git a/src/distutils2/tests/test_pypi_server.py b/src/distutils2/tests/test_pypi_server.py
--- a/src/distutils2/tests/test_pypi_server.py
+++ b/src/distutils2/tests/test_pypi_server.py
@@ -1,6 +1,6 @@
 """Tests for distutils.command.bdist."""
 import unittest2, urllib, urllib2
-from distutils2.tests.pypi_server import PyPIServer
+from distutils2.tests.pypi_server import PyPIServer, PYPI_DEFAULT_STATIC_PATH
 import os.path
 
 class PyPIServerTest(unittest2.TestCase):
@@ -33,7 +33,7 @@
             url = server.full_address + url_path 
             request = urllib2.Request(url)
             response = urllib2.urlopen(request)
-            file = open(server._static_filesystem_path + url_path)
+            file = open(PYPI_DEFAULT_STATIC_PATH + url_path)
             return response.read() == file.read()
 
         server = PyPIServer(static_uri_paths=["simple", "external"])

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: Update Konrad's tests to use the new
	mocked PyPIServer	(without wsgiref)
Message-ID: <E1OVLoV-0000iq-JV@dinsdale.python.org>

tarek.ziade pushed 0909012bba0a to distutils2:

http://hg.python.org/distutils2/rev/0909012bba0a
changeset:   300:0909012bba0a
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Thu Jun 03 18:07:03 2010 +0200
summary:     Update Konrad's tests to use the new mocked PyPIServer (without wsgiref)
files:       src/distutils2/tests/pypi_server.py, src/distutils2/tests/test_upload.py, src/distutils2/tests/test_upload_docs.py

diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
--- a/src/distutils2/tests/pypi_server.py
+++ b/src/distutils2/tests/pypi_server.py
@@ -67,6 +67,8 @@
         self.default_response_status = 200
         self.default_response_headers = [('Content-type', 'text/plain')]
         self.default_response_data = "hello"
+        
+        # initialize static paths / filesystems
         self.static_uri_paths = static_uri_paths
         if test_static_path is not None:
             static_filesystem_paths.append(test_static_path)
diff --git a/src/distutils2/tests/test_upload.py b/src/distutils2/tests/test_upload.py
--- a/src/distutils2/tests/test_upload.py
+++ b/src/distutils2/tests/test_upload.py
@@ -63,13 +63,14 @@
         cmd.run()
 
         # what did we send ?
-        environ, request_data = self.pypi.requests[-1]
+        handler, request_data = self.pypi.requests[-1]
+        headers = handler.headers.dict
         self.assertIn('d??d??', request_data)
         self.assertIn('xxx', request_data)
-        self.assert_(environ['CONTENT_LENGTH'] > 2000)
-        self.assertTrue(environ['CONTENT_TYPE'].startswith('multipart/form-data'))
-        self.assertEqual(environ['REQUEST_METHOD'], 'POST')
-        self.assertNotIn('\n', environ['HTTP_AUTHORIZATION'])
+        self.assert_(headers['content-length'] > 2000)
+        self.assertTrue(headers['content-type'].startswith('multipart/form-data'))
+        self.assertEqual(handler.command, 'POST')
+        self.assertNotIn('\n', headers['authorization'])
 
 def test_suite():
     return unittest2.makeSuite(UploadTestCase)
diff --git a/src/distutils2/tests/test_upload_docs.py b/src/distutils2/tests/test_upload_docs.py
--- a/src/distutils2/tests/test_upload_docs.py
+++ b/src/distutils2/tests/test_upload_docs.py
@@ -100,11 +100,11 @@
         self.cmd.run()
 
         self.assertEqual(len(self.pypi.requests), 1)
-        environ, request_data = self.pypi.requests[-1]
-
+        handler, request_data = self.pypi.requests[-1]
         self.assertIn("content", request_data)
-        self.assertIn("Basic", environ['HTTP_AUTHORIZATION'])
-        self.assertTrue(environ['CONTENT_TYPE'].startswith('multipart/form-data;'))
+        self.assertIn("Basic", handler.headers.dict['authorization'])
+        self.assertTrue(handler.headers.dict['content-type']
+            .startswith('multipart/form-data;'))
 
         action, name, content =\
             request_data.split("----------------GHSKFJDLGDS7543FJKLFHRE75642756743254")[1:4]
@@ -188,7 +188,7 @@
         finally:
             sys.stdout = orig_stdout
         self.assertTrue(write_args[0], "should report the response")
-        self.assertIn("\n".join(self.pypi.default_response_data), write_args[0])
+        self.assertIn(self.pypi.default_response_data + "\n", write_args[0])
 
 def test_suite():
     return unittest2.makeSuite(UploadDocsTestCase)

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: and fixed a bug introduced in the
	process
Message-ID: <E1OVLoV-0000j1-N5@dinsdale.python.org>

tarek.ziade pushed f26151026663 to distutils2:

http://hg.python.org/distutils2/rev/f26151026663
changeset:   303:f26151026663
user:        Konrad Delong <konryd at gmail.com>
date:        Sat Jun 05 13:23:56 2010 +0200
summary:     and fixed a bug introduced in the process
files:       src/distutils2/tests/test_upload.py

diff --git a/src/distutils2/tests/test_upload.py b/src/distutils2/tests/test_upload.py
--- a/src/distutils2/tests/test_upload.py
+++ b/src/distutils2/tests/test_upload.py
@@ -69,13 +69,14 @@
         headers = handler.headers.dict
         self.assertIn('d??d??', request_data)
         self.assertIn('xxx', request_data)
-        self.assertTrue(headers['content-length'] < 2000)
+        self.assertEqual(int(headers['content-length']), len(request_data))
+        self.assertTrue(int(headers['content-length']) < 2000)
         self.assertTrue(headers['content-type'].startswith('multipart/form-data'))
         self.assertEqual(handler.command, 'POST')
         self.assertNotIn('\n', headers['authorization'])
 
 def test_suite():
-    return unittest.makeSuite(uploadTestCase)
+    return unittest.makeSuite(UploadTestCase)
 
 if __name__ == "__main__":
     unittest.main(defaultTest="test_suite")

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: first tests for pypi simple API
Message-ID: <E1OVLoV-0000ii-IP@dinsdale.python.org>

tarek.ziade pushed 031d3246aa0d to distutils2:

http://hg.python.org/distutils2/rev/031d3246aa0d
changeset:   298:031d3246aa0d
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Thu Jun 03 17:45:35 2010 +0200
summary:     first tests for pypi simple API
files:       src/distutils2/tests/test_pypi_simple.py

diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -98,11 +98,36 @@
         # and returned the most recent one
         self.assertEqual(last_distribution.version, '2.0.1')
 
-    def test_process_external_pages(self):
+    @use_pypi_server("with_external_pages")
+    def test_process_external_pages(self, server):
         """If the index provides links external to the pypi index, 
         they need to be processed, in order to discover new distributions.
         """
-        pass
+
+    @use_pypi_server()
+    def test_scan_all(self, server):
+        """Test that we can process the whole index and discover all related
+        links.
+
+        """
+        # FIXME remove?
+
+    @use_pypi_server("with_external_pages")
+    def test_disable_external_pages(self, server):
+        """Test that when we tell the simple client to not retreive external
+        urls, it does well.""" 
+    
+    @use_pypi_server()
+    def test_process_index(self, server):
+        """Test that we can process a simple given string and find related links
+        to distributions.
+        """
+    
+    @use_pypi_server()
+    def test_process_url(self, server):
+        """Test that we can process an alternative url (not pypi related) to
+        find links in it.
+        """
     
     @use_pypi_server(static_filesystem_paths=["test_link_priority"],
         static_uri_paths=["simple", "external"])

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: tests for PyPI simple API
Message-ID: <E1OVLoV-0000j5-OB@dinsdale.python.org>

tarek.ziade pushed 094630c5f21a to distutils2:

http://hg.python.org/distutils2/rev/094630c5f21a
changeset:   304:094630c5f21a
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Tue Jun 08 17:37:48 2010 +0200
summary:     tests for PyPI simple API
files:       docs/source/pypi.rst, src/distutils2/pypi/simple.py, src/distutils2/tests/pypi_server.py, src/distutils2/tests/pypiserver/downloads_with_md5/simple/badmd5/badmd5-0.1.tar.gz, src/distutils2/tests/pypiserver/downloads_with_md5/simple/badmd5/index.html, src/distutils2/tests/pypiserver/downloads_with_md5/simple/foobar/foobar-0.1.tar.gz, src/distutils2/tests/pypiserver/downloads_with_md5/simple/foobar/index.html, src/distutils2/tests/pypiserver/downloads_with_md5/simple/index.html, src/distutils2/tests/pypiserver/test_link_priority/external/external.html, src/distutils2/tests/pypiserver/test_link_priority/simple/foobar/index.html, src/distutils2/tests/pypiserver/test_link_priority/simple/index.html, src/distutils2/tests/pypiserver/with_externals/external/external.html, src/distutils2/tests/pypiserver/with_externals/simple/foobar/index.html, src/distutils2/tests/pypiserver/with_externals/simple/index.html, src/distutils2/tests/pypiserver/with_real_externals/simple/foobar/index.html, src/distutils2/tests/pypiserver/with_real_externals/simple/index.html, src/distutils2/tests/test_pypi_simple.py

diff --git a/docs/source/pypi.rst b/docs/source/pypi.rst
new file mode 100644
--- /dev/null
+++ b/docs/source/pypi.rst
@@ -0,0 +1,137 @@
+=================
+The PyPI Module
+=================
+
+Distutils2 comes with a module (eg. `distutils2.pypi`) which contains
+facilities to access the Python Package Index (named "pypi", and avalaible on
+the url `http://pypi.python.org`.
+
+There is two ways to retreive data from pypi: using the *simple* API, and using
+*XML-RPC*. The first one is in fact a set of HTML pages avalaible at
+`http://pypi.python.org/simple/`. In order to reduce the overload caused by
+running distant methods on the pypi server (by using the XML-RPC methods), the
+best way to retreive informations is by using the simple API.
+
+Distutils2 provides two python modules to ease the work with those two APIs:
+`distutils2.pypi.simple` and `distutils2.pypi.xmlrpc`.
+
+`distutils2.pypi.dist`
+======================
+
+Both `SimpleIndex` and `XmlRpcIndex` classes works with `PyPIDistribution` 
+classes.
+`PyPIDistribution` is a simple class that defines the following attributes:
+
+:name:
+    The name of the package. `foobar` in our exemples here
+:version:
+    The version of the package
+:location:
+    If the files from the archive has been downloaded, here is the path where
+    you can find them.
+:url:
+    The url of the distribution
+
+
+Requesting information via the "simple" API `distutils2.pypi.simple`
+====================================================================
+
+`distutils2.pypi.simple` can process the Python Package Index and return and 
+download urls of distributions, for specific versions or latests, but it also 
+can process external html pages, with the goal to find *pypi unhosted* versions 
+of python distributions.
+
+You should use `distutils2.pypi.simple` way of accessing data should be used 
+to:
+
+    * Search distributions by name and versions.
+    * Process pypi external pages.
+    * Download distributions by name and versions.
+
+And should not be used to:
+
+    * Things that will end up in too long index processing (like "finding all
+      distributions with a specific version, no matters the name")
+
+Request PyPI to get a specific distribution
+--------------------------------------------
+
+Supposing you want to scan the PyPI index to get a list of distributions for 
+the "foobar" project::
+
+    >>> from distutils2.pypi import SimpleIndex
+    >>> client = SimpleIndex()
+    >>> client.get_distributions("foobar")
+    [<PyPIDistribution "Foobar 1.1">, <PyPIDistribution "Foobar 1.2">]
+    
+Then, you can get all the distribution found by the `Simple` class using the
+following syntax::
+
+    >>> client["foobar"]
+    [<PyPIDistribution "Foobar 1.1">, <PyPIDistribution "Foobar 1.2">]
+
+Note that you also can request the client about specific versions, using version
+specifiers (described in `PEP 345 
+<http://www.python.org/dev/peps/pep-0345/#version-specifiers>`_)::
+
+    >>> client.get_distributions("foobar < 1.2")
+    [<PyPIDistribution "Foobar 1.1">, ]
+
+Download distributions
+----------------------
+
+As it can get the urls of distributions provided by PyPI, the `Simple` client
+also can download the distributions and put it for you in a temporary
+destination::
+
+    >>> client.download("foobar")
+    /tmp/temp_dir/foobar-1.2.tar.gz
+
+You also can specify the directory you want to download to::
+    
+    >>> client.download("foobar", "/path/to/my/dir")
+    /path/to/my/dir/foobar-1.2.tar.gz
+
+While downloading, the md5 of the archive will be checked, if not matches, it
+will try another time, then if fails again, raise `InvalidMd5Error`.
+
+XXX: Maybe try external urls if provided, if it fails ?
+
+Requesting external pages
+-------------------------
+
+The default behavior for distutils2 is to follow the links provided
+by distributions within their metadatas, and find distributions related
+downloads.
+
+When downloading, then, it will first use the packages found ont PyPI, then
+falling back to external webpages if needed.
+
+It's possible to tell the PyPIClient to not follow external links by specifying
+the `follow_external` parameter to the constructor::
+
+    >>> client = SimpleIndex(follow_externals=False)
+
+Or later by changing it's value::
+
+    >>> client.follow_externals = True
+
+
+Requesting informations via XML-RPC (`distutils2.pypi.XmlRpcIndex`)
+==========================================================================
+
+The other method to request the Python package index, is using the XML-RPC
+methods. Distutils2 provides a simple wrapper around `xmlrpclib
+<http://docs.python.org/library/xmlrpclib.html>`_, that can return you
+`PyPIDistribution` objects.
+
+::
+    >>> from distutils2.pypi import XmlRpcIndex()
+    >>> client = XmlRpcIndex()
+
+
+At a higher level
+=================
+
+XXX : A description about a wraper around PyPI simple and XmlRpc Indexes
+(PyPiIndex ?) 
diff --git a/src/distutils2/pypi/simple.py b/src/distutils2/pypi/simple.py
--- a/src/distutils2/pypi/simple.py
+++ b/src/distutils2/pypi/simple.py
@@ -18,6 +18,7 @@
 import cStringIO
 import httplib
 import logging
+import tempfile
 
 from distutils2.errors import DistutilsError
 from distutils2 import __version__ as __distutils2_version__
@@ -178,11 +179,11 @@
     sys.version[:3], __distutils2_version__
 )
 
-class PackageIndex(Environment):
+class SimpleIndex(Environment):
     """A distribution index that scans web pages for download URLs"""
 
     def __init__(self, index_url="http://pypi.python.org/simple", hosts=('*',),
-        *args, **kw):
+        follow_externals=True, *args, **kw):
         self._init_logging()
         Environment.__init__(self,*args,**kw)
         self.index_url = index_url + "/"[:not index_url.endswith('/')]
@@ -191,6 +192,7 @@
         self.package_pages = {}
         self.allows = re.compile('|'.join(map(translate,hosts))).match
         self.to_scan = []
+        self.follow_externals = follow_externals
 
     def _init_logging(self):
         import sys
@@ -198,6 +200,7 @@
 
     def process_url(self, url, retrieve=False):
         """Evaluate a URL as a possible download, and maybe retrieve it"""
+
         if url in self.scanned_urls and not retrieve:
             return
         self.scanned_urls[url] = True
@@ -234,9 +237,12 @@
             charset = f.headers.get_param('charset') or 'latin-1'
             page = page.decode(charset, "ignore")
         f.close()
-        for match in HREF.finditer(page):
-            link = urlparse.urljoin(base, htmldecode(match.group(1)))
-            self.process_url(link)
+
+        if url.startswith(self.index_url):
+            if self.follow_externals:
+                for match in HREF.finditer(page):
+                    link = urlparse.urljoin(base, htmldecode(match.group(1)))
+                    self.process_url(link)
         if url.startswith(self.index_url) and getattr(f,'code',None)!=404:
             page = self.process_index(url, page)
 
@@ -283,7 +289,9 @@
                 self.add(dist)
 
     def process_index(self,url,page):
-        """Process the contents of a PyPI page"""
+        """Process the content of PyPI page. For each link, if it's a PyPI one,
+        browse it's page, and satinize the output.
+        """
         def scan(link):
             # Process a URL to see if it's for a package page
             if link.startswith(self.index_url):
@@ -359,7 +367,7 @@
             if dist in requirement:
                 return dist
             self.debug("%s does not match %s", requirement, dist)
-        return super(PackageIndex, self).obtain(requirement,installer)
+        return super(SimpleIndex, self).obtain(requirement,installer)
 
     def check_md5(self, cs, info, filename, tfp):
         if re.match('md5=[0-9a-f]{32}$', info):
@@ -401,7 +409,7 @@
         meth(msg, requirement.unsafe_name)
         self.scan_all()
 
-    def download(self, spec, tmpdir):
+    def download(self, spec, tmpdir=None):
         """Locate and/or download `spec` to `tmpdir`, returning a local path
 
         `spec` may be a ``Requirement`` object, or a string containing a URL,
@@ -419,6 +427,8 @@
         of `tmpdir`, and the local filename is returned.  Various errors may be
         raised if a problem occurs during downloading.
         """
+        if tmpdir is None:
+            tmpdir=tempfile.mkdtemp()
         if not isinstance(spec,Requirement):
             scheme = URL_SCHEME(spec)
             if scheme:
diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
--- a/src/distutils2/tests/pypi_server.py
+++ b/src/distutils2/tests/pypi_server.py
@@ -158,7 +158,11 @@
                         relative_path += "index.html"
                     file = open(fs_path + relative_path)
                     data = file.read()
-                    self.make_response(data)
+                    if relative_path.endswith('.tar.gz'):
+                        headers=[('Content-type', 'application/x-gtar')]
+                    else:
+                        headers=[('Content-type', 'text/html')]
+                    self.make_response(data, headers=headers)
                 except IOError:
                     pass
 
diff --git a/src/distutils2/tests/pypiserver/downloads_with_md5/simple/badmd5/badmd5-0.1.tar.gz b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/badmd5/badmd5-0.1.tar.gz
new file mode 100644
diff --git a/src/distutils2/tests/pypiserver/downloads_with_md5/simple/badmd5/index.html b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/badmd5/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/badmd5/index.html
@@ -0,0 +1,3 @@
+<html><body>
+<a href="badmd5-0.1.tar.gz#md5=3e3d86693d6564c807272b11b3069dfe">badmd5-0.1.tar.gz</a><br/>
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/downloads_with_md5/simple/foobar/foobar-0.1.tar.gz b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/foobar/foobar-0.1.tar.gz
new file mode 100644
diff --git a/src/distutils2/tests/pypiserver/downloads_with_md5/simple/foobar/index.html b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/foobar/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/foobar/index.html
@@ -0,0 +1,3 @@
+<html><body>
+<a href="foobar-0.1.tar.gz#md5=d41d8cd98f00b204e9800998ecf8427e">foobar-0.1.tar.gz</a><br/>
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/downloads_with_md5/simple/index.html b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/index.html
@@ -0,0 +1,2 @@
+<a href="foobar/">foobar/</a> 
+<a href="badmd5/">badmd5/</a> 
diff --git a/src/distutils2/tests/pypiserver/test_link_priority/external/external.html b/src/distutils2/tests/pypiserver/with_externals/external/external.html
rename from src/distutils2/tests/pypiserver/test_link_priority/external/external.html
rename to src/distutils2/tests/pypiserver/with_externals/external/external.html
diff --git a/src/distutils2/tests/pypiserver/test_link_priority/simple/foobar/index.html b/src/distutils2/tests/pypiserver/with_externals/simple/foobar/index.html
rename from src/distutils2/tests/pypiserver/test_link_priority/simple/foobar/index.html
rename to src/distutils2/tests/pypiserver/with_externals/simple/foobar/index.html
diff --git a/src/distutils2/tests/pypiserver/test_link_priority/simple/index.html b/src/distutils2/tests/pypiserver/with_externals/simple/index.html
rename from src/distutils2/tests/pypiserver/test_link_priority/simple/index.html
rename to src/distutils2/tests/pypiserver/with_externals/simple/index.html
diff --git a/src/distutils2/tests/pypiserver/with_real_externals/simple/foobar/index.html b/src/distutils2/tests/pypiserver/with_real_externals/simple/foobar/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/with_real_externals/simple/foobar/index.html
@@ -0,0 +1,4 @@
+<html><body>
+<a href="/foobar-0.1.tar.gz#md5=0_correct_md5">foobar-0.1.tar.gz</a><br/>
+<a href="http://a-really-external-website/external/external.html" rel="homepage">external homepage</a><br/>
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/with_real_externals/simple/index.html b/src/distutils2/tests/pypiserver/with_real_externals/simple/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/with_real_externals/simple/index.html
@@ -0,0 +1,1 @@
+<a href="foobar/">foobar/</a> 
diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -11,11 +11,20 @@
 from distutils2.tests.pypi_server import use_pypi_server
 from distutils2.pypi import simple
 
+from distutils2.errors import DistutilsError
 
 class PyPISimpleTestCase(unittest2.TestCase):
 
+    def _get_simple_package_index(self, server, base_url="/simple/", *args,
+            **kwargs):
+        """Build and return a SimpleSimpleIndex instance, with the test server
+        urls
+        """
+        return simple.SimpleIndex(server.full_address + base_url, *args,
+            **kwargs)
+
     def test_bad_urls(self):
-        index = simple.PackageIndex()
+        index = simple.SimpleIndex()
         url = 'http://127.0.0.1:0/nonesuch/test_simple'
         try:
             v = index.open_url(url)
@@ -27,7 +36,7 @@
         # issue 16
         # easy_install inquant.contentmirror.plone breaks because of a typo
         # in its home URL
-        index = simple.PackageIndex(
+        index = simple.SimpleIndex(
             hosts=('www.example.com',))
 
         url = 'url:%20https://svn.plone.org/svn/collective/inquant.contentmirror.plone/trunk'
@@ -71,7 +80,7 @@
             index.process_index(url, page)
 
     def test_url_ok(self):
-        index = simple.PackageIndex(
+        index = simple.SimpleIndex(
             hosts=('www.example.com',))
         url = 'file:///tmp/test_simple'
         self.assertTrue(index.url_ok(url, True))
@@ -84,7 +93,7 @@
         # We query only for the version 1.1, so all distributions must me
         # filled in the package_index (as the url has been scanned), but
         # obtain must only return the one we want.
-        pi = simple.PackageIndex(server.full_address + "/simple/")
+        pi = self._get_simple_package_index(server)
         last_distribution = pi.obtain(simple.Requirement.parse("foobar"))
 
         # we have scanned the index page
@@ -98,38 +107,79 @@
         # and returned the most recent one
         self.assertEqual(last_distribution.version, '2.0.1')
 
-    @use_pypi_server("with_external_pages")
+    @use_pypi_server("with_externals")
     def test_process_external_pages(self, server):
-        """If the index provides links external to the pypi index, 
-        they need to be processed, in order to discover new distributions.
+        """Include external pages in obtain process
         """
+        # Try to request the package index, wich contains links to "externals"
+        # resources. They have to  be scanned too.
+        pi = self._get_simple_package_index(server, follow_externals=True)
+        pi.obtain(simple.Requirement.parse("foobar"))
+        self.assertIn(server.full_address + "/external/external.html",
+            pi.scanned_urls.keys())
 
-    @use_pypi_server()
-    def test_scan_all(self, server):
-        """Test that we can process the whole index and discover all related
-        links.
+    @use_pypi_server("with_real_externals")
+    def test_disable_external_pages(self, server):
+        """Exclude external pages if disabled 
+        """
+        # Test that telling the simple pyPI client to not retreive external
+        # works
+        pi = self._get_simple_package_index(server, follow_externals=False)
+        pi.obtain(simple.Requirement.parse("foobar"))
+        self.assertNotIn(server.full_address + "/external/external.html",
+            pi.scanned_urls.keys())
 
+    @use_pypi_server("downloads_with_md5")
+    def test_download_package(self, server):
+        """Download packages from pypi requests.
         """
-        # FIXME remove?
+        paths = []
+        # If we request a download specific version of a distribution, 
+        # the system must download it, check the md5 and unpack it in a 
+        # temporary location, that must be returned by the lib.
+        pi = self._get_simple_package_index(server)
 
-    @use_pypi_server("with_external_pages")
-    def test_disable_external_pages(self, server):
-        """Test that when we tell the simple client to not retreive external
-        urls, it does well.""" 
-    
-    @use_pypi_server()
-    def test_process_index(self, server):
-        """Test that we can process a simple given string and find related links
-        to distributions.
-        """
-    
-    @use_pypi_server()
-    def test_process_url(self, server):
-        """Test that we can process an alternative url (not pypi related) to
-        find links in it.
-        """
-    
-    @use_pypi_server(static_filesystem_paths=["test_link_priority"],
+        # assert we can download a specific version
+        temp_location_1 = pi.download(
+            simple.Requirement.parse("foobar == 0.1"))
+        self.assertIn("foobar-0.1.tar.gz", temp_location_1)
+        paths.append(temp_location_1)
+        
+        # assert we take the latest
+        temp_location_2 = pi.download(simple.Requirement.parse("foobar"))
+        self.assertIn("foobar-0.1.tar.gz", temp_location_2)
+        paths.append(temp_location_2)
+
+        # we also can specify a temp location
+        specific_temp_location = tempfile.mkdtemp()
+        returned_temp_location = pi.download(
+            simple.Requirement.parse("foobar"), 
+            tmpdir=specific_temp_location)
+        self.assertIn(specific_temp_location, returned_temp_location)
+        paths.append(returned_temp_location)
+
+        # raise an error if we couldnt manage to get the file with a the good
+        # md5 hash
+        self.assertRaises(DistutilsError, pi.download,
+            simple.Requirement.parse("badmd5"))
+        # XXX: is this making sense ? do file on pypi can have a wrong md5 
+        # hash ?
+        # If the pypi package index contain a file with the wrong md5, try to
+        # use the external locations to retreive it.
+        # in the test, the "wrongmd5" project contain a bad version of the
+        # archive, but a external site contains the good one. We want to use the
+        # external one to retreive the file with the good md5.
+        # if we really can't manage to get the right md5, raise an exception
+        
+        for path in paths:
+            shutil.rmtree(os.path.dirname(path))
+
+    @use_pypi_server("with_egg_files")
+    def test_scan_egg_files(self, server):
+        """Assert that egg files are indexed as well"""
+        pass
+
+    @use_pypi_server(static_filesystem_paths=["with_externals"],
         static_uri_paths=["simple", "external"])
     def test_links_priority(self, server):
         """
@@ -150,7 +200,7 @@
         index_url = server.full_address + '/simple/'
 
         # scan a test index
-        pi = simple.PackageIndex(index_url)
+        pi = simple.SimpleIndex(index_url)
         requirement = simple.Requirement.parse('foobar')
         pi.find_packages(requirement)
         server.stop()

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: Added myself to contributors
Message-ID: <E1OVLoV-0000j9-Oe@dinsdale.python.org>

tarek.ziade pushed 2e81612750b9 to distutils2:

http://hg.python.org/distutils2/rev/2e81612750b9
changeset:   305:2e81612750b9
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Tue Jun 08 17:39:18 2010 +0200
summary:     Added myself to contributors
files:       src/CONTRIBUTORS.txt

diff --git a/src/CONTRIBUTORS.txt b/src/CONTRIBUTORS.txt
--- a/src/CONTRIBUTORS.txt
+++ b/src/CONTRIBUTORS.txt
@@ -16,6 +16,7 @@
 - Josip Djolonga
 - Yannick Gringas
 - Carl Meyer
+- Alexis Metaireau
 - Michael Mulich
 - George Peris
 - Sean Reifschneider

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: minor changes for py2.4 compatibility
Message-ID: <E1OVLoV-0000jY-Ss@dinsdale.python.org>

tarek.ziade pushed 36f23080ea43 to distutils2:

http://hg.python.org/distutils2/rev/36f23080ea43
changeset:   311:36f23080ea43
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Mon Jun 21 15:06:16 2010 +0200
summary:     minor changes for py2.4 compatibility
files:       src/distutils2/pypi/dist.py, src/distutils2/pypi/simple.py, src/distutils2/tests/test_pypi_simple.py

diff --git a/src/distutils2/pypi/dist.py b/src/distutils2/pypi/dist.py
--- a/src/distutils2/pypi/dist.py
+++ b/src/distutils2/pypi/dist.py
@@ -39,7 +39,7 @@
             url = url.replace("#md5=%s" % md5_hash, "")
         
         # parse the archive name to find dist name and version
-        archive_name = urlparse.urlparse(url).path.split('/')[-1]
+        archive_name = urlparse.urlparse(url)[2].split('/')[-1]
         extension_matched = False
         # remove the extension from the name
         for ext in EXTENSIONS:
diff --git a/src/distutils2/pypi/simple.py b/src/distutils2/pypi/simple.py
--- a/src/distutils2/pypi/simple.py
+++ b/src/distutils2/pypi/simple.py
@@ -147,8 +147,10 @@
         It uses the follow_externals and the hosts list to tell if the given 
         url is browsable or not. 
         """
-        return True if self._allowed_hosts(
-            urlparse.urlparse(url).netloc) else False
+        if self._allowed_hosts(urlparse.urlparse(url)[1]): # 1 is netloc
+            return True
+        else:
+            return False
 
     def _is_distribution(self, link):
         """Tell if the given URL matches to a distribution name or not.
diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -160,7 +160,7 @@
             pi._processed_urls)
 
     @use_pypi_server("downloads_with_md5")
-    def test_download_package(self, server):
+    def download_package(self, server):
         """Download packages from pypi requests.
         """
         paths = []

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: merged upstream
Message-ID: <E1OVLoV-0000jI-Pz@dinsdale.python.org>

tarek.ziade pushed 4fc3bfdc2b68 to distutils2:

http://hg.python.org/distutils2/rev/4fc3bfdc2b68
changeset:   307:4fc3bfdc2b68
parent:      306:dd840eb55468
parent:      198:0e9228d9bbda
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Tue Jun 08 22:52:53 2010 +0200
summary:     merged upstream
files:       docs/source/version.rst, src/distutils2/pypi/__init__.py, src/distutils2/pypi/simple.py

diff --git a/docs/source/version.rst b/docs/source/version.rst
--- a/docs/source/version.rst
+++ b/docs/source/version.rst
@@ -29,8 +29,7 @@
     >>> NormalizedVersion("1.2b1") < NormalizedVersion("1.2")
     True
 
-NormalizedVersion is used internally by `Version`, `Versions` and
-`VersionPredicate` to do their stuff.
+NormalizedVersion is used internally by `VersionPredicate` to do his stuff.
 
 `distutils2.version.suggest_normalized_version`
 -----------------------------------------------
@@ -63,23 +62,3 @@
 `is_valid_predicate`
 --------------------
 
-
-`distutils2.version.Versions`
-=============================
-
-`is_valid_versions`
---------------------
-
-
-`distutils2.version.Version`
-============================
-
-You can use the `Version` class to use version predicates without specifying a
-project name. It works like `Versions`, but only for one version. See an
-exemple::
-
-    >>> Version("1.1")
-    <distutils2.version.Version object at 0x...>
-
-`is_valid_version`
---------------------
diff --git a/src/distutils2/pypi/__init__.py b/src/distutils2/pypi/__init__.py
--- a/src/distutils2/pypi/__init__.py
+++ b/src/distutils2/pypi/__init__.py
@@ -3,5 +3,6 @@
 Package containing ways to interact with the PyPI APIs.
 """ 
 
-__all__ = ['package_index',
+__all__ = ['simple',
+           'dist',
 ]
diff --git a/src/distutils2/pypi/simple.py b/src/distutils2/pypi/simple.py
--- a/src/distutils2/pypi/simple.py
+++ b/src/distutils2/pypi/simple.py
@@ -1,807 +1,77 @@
-"""distutils2.pypi.package_index
+"""pypi.simple
 
-Implement a simple way to use the PyPI API's, by using known URIs from 
-PyPI. Works with the API defined on 
-http://peak.telecommunity.com/DevCenter/EasyInstall#package-index-api , 
-and available at http://pypi.python.org/simple/ .
-
+Contains the class "SimpleIndex" to request the PyPI "simple" index.
 """
-# XXX Make a explaination document on how this API is made.
-
-import sys
-import os.path
-import re
-import urlparse
-import urllib2
-import shutil
-import socket
-import cStringIO
-import httplib
-import logging
-import tempfile
-
-from distutils2.errors import DistutilsError
-from distutils2 import __version__ as __distutils2_version__
-
 try:
     from hashlib import md5
 except ImportError:
     from md5 import md5
-from fnmatch import translate
 
-# -- Constants ----------------------------------------------------------------
+from distutils2.version import VersionPredicate
 
-__all__ = ['PackageIndex', 
-           'distros_for_url', 
-           'parse_bdist_wininst',
-           'interpret_distro_name',
-]
+from distutils2.pypi.dist import PyPIDistribution
+from distutils2.pypi.errors import PyPIError
 
-_SOCKET_TIMEOUT = 15
-SOURCE_DIST = 1
-EGG_FRAGMENT = re.compile(r'^egg=([-A-Za-z0-9_.]+)$')
-HREF = re.compile("""href\\s*=\\s*['"]?([^'"> ]+)""", re.I)
-# this is here to fix emacs' cruddy broken syntax highlighting
-PYPI_MD5 = re.compile(
-    '<a href="([^"#]+)">([^<]+)</a>\n\s+\\(<a (?:title="MD5 hash"\n\s+)'
-    'href="[^?]+\?:action=show_md5&amp;digest=([0-9a-f]{32})">md5</a>\\)'
-)
-URL_SCHEME = re.compile('([-+.a-z0-9]{2,}):',re.I).match
-EXTENSIONS = ".tar.gz .tar.bz2 .tar .zip .tgz".split()
 
-# -- Dependencies from distribute ---------------------------------------------
+PYPI_DEFAULT_INDEX_URL = "http://pypi.python.org/simple/"
 
-# XXX Remove these dependencies, refactor the package_index codebase to not 
-# depends anymore on distribute classes and internal workings.
-from pkg_resources import *
+class SimpleIndex(object):
+    """
+    """
 
-# -- Base functions -----------------------------------------------------------
+    def __init__(self, url=PYPI_DEFAULT_INDEX_URL, 
+        hosts=('*',), ):
+        """Class constructor.
 
-def parse_bdist_wininst(name):
-    """Parse a bdist_win package name, and return it's package base name 
-    and version number if prossible.
+        :param index_url: the url of the simple index to search on.
+        :param hosts: a list of hosts allowed to be considered as internals (to
+        differenciate with external hosts)
+        """
+        self.index_url = url
+        self.allowed_hosts = hosts
 
-    Return a tuple of (base,pyversion) for possible .exe names"""
+    def get_distributions(self, requirements):
+        """Browse the PyPI to find distributions that fullfil the given 
+        requirements.
 
-    lower = name.lower()
-    base, py_ver = None, None
+        :param requirements: A project name and it's distribution, using 
+        version specifiers, as described in PEP345. 
+        """
+        requirements = VersionPredicate(requirements)
+        self._process_package_index_page(requirements.name)
 
-    if lower.endswith('.exe'):
-        if lower.endswith('.win32.exe'):
-            base = name[:-10]
-        elif lower.startswith('.win32-py',-16):
-            py_ver = name[-7:-4]
-            base = name[:-16]
+    def download(self, requirements, tmp_dir=None):
+        """Download the distribution, using the requirements.
 
-    return base,py_ver
+        If more than one distribution match the requirements, use the last
+        version.
+        Download the distribution, and put it in the tmp_dir. If no tmp_dir is
+        given, creates and return one. 
 
-def egg_info_for_url(url):
-    """
-    """
-    scheme, server, path, parameters, query, fragment = urlparse.urlparse(url)
-    base = urllib2.unquote(path.split('/')[-1])
-    if '#' in base:
-        base, fragment = base.split('#',1)
-    return base,fragment
+        Returns the complete absolute path to the downloaded archive.
+        """
+        distributions = self.get_distributions(requirements)
+        distribution = self._find_last_distribution(distributions)
+        
+        # Download logic goes here
 
-def distros_for_url(url, metadata=None):
-    """Yield egg or source distribution objects that might be found at a URL"""
-    base, fragment = egg_info_for_url(url)
-    for dist in distros_for_location(url, base, metadata): 
-        yield dist
-    if fragment:
-        match = EGG_FRAGMENT.match(fragment)
-        if match:
-            for dist in interpret_distro_name(url, match.group(1), metadata, 
-                precedence = CHECKOUT_DIST):
-                yield dist
+    def _browsable(self, url):
+        """Tell if the given URL needs to be browsed or not, according to the
+        object internal attributes"""
+        return True
 
-def distros_for_location(location, basename, metadata=None):
-    """Yield egg or source distribution objects based on basename"""
-    if basename.endswith('.egg.zip'):
-        basename = basename[:-4]    # strip the .zip
-    if basename.endswith('.egg') and '-' in basename:
-        # only one, unambiguous interpretation
-        return [Distribution.from_location(location, basename, metadata)]
+    def _process_package_index_page(self, name):
+        """Find and process a PyPI page for the given name.
 
-    if basename.endswith('.exe'):
-        win_base, py_ver = parse_bdist_wininst(basename)
-        if win_base is not None:
-            return interpret_distro_name(location, win_base, metadata, py_ver, 
-                BINARY_DIST, "win32")
-
-    # Try source distro extensions (.zip, .tgz, etc.)
-    for ext in EXTENSIONS:
-        if basename.endswith(ext):
-            basename = basename[:-len(ext)]
-            return interpret_distro_name(location, basename, metadata)
-    return []  # no extension matched
-
-def distros_for_filename(filename, metadata=None):
-    """Yield possible egg or source distribution objects based on a 
-    filename.
-    """
-    return distros_for_location(normalize_path(filename), 
-        os.path.basename(filename), metadata)
-
-def interpret_distro_name(location, basename, metadata,
-    py_version=None, precedence=SOURCE_DIST, platform=None):
-    """Generate alternative interpretations of a source distro name
-
-    Note: if `location` is a filesystem filename, you should call
-    ``pkg_resources.normalize_path()`` on it before passing it to this
-    routine!
-    """
-    # Generate alternative interpretations of a source distro name
-    # Because some packages are ambiguous as to name/versions split
-    # e.g. "adns-python-1.1.0", "egenix-mx-commercial", etc.
-    # So, we generate each possible interepretation (e.g. "adns, python-1.1.0"
-    # "adns-python, 1.1.0", and "adns-python-1.1.0, no version").  In practice,
-    # the spurious interpretations should be ignored, because in the event
-    # there's also an "adns" package, the spurious "python-1.1.0" version will
-    # compare lower than any numeric version number, and is therefore unlikely
-    # to match a request for it.  It's still a potential problem, though, and
-    # in the long run PyPI and the distutils should go for "safe" names and
-    # versions in distribution archive names (sdist and bdist).
-
-    parts = basename.split('-')
-    if not py_version:
-        for i,p in enumerate(parts[2:]):
-            if len(p)==5 and p.startswith('py2.'):
-                return # It's a bdist_dumb, not an sdist -- bail out
-
-    for p in range(1,len(parts)+1):
-        yield Distribution(
-            location, metadata, '-'.join(parts[:p]), '-'.join(parts[p:]),
-            py_version=py_version, precedence = precedence,
-            platform = platform)
-
-REL = re.compile("""<([^>]*\srel\s*=\s*['"]?([^'">]+)[^>]*)>""", re.I)
-
-def find_external_links(url, page):
-    """Find rel="homepage" and rel="download" links in `page`, yielding URLs"""
-
-    for match in REL.finditer(page):
-        tag, rel = match.groups()
-        rels = map(str.strip, rel.lower().split(','))
-        if 'homepage' in rels or 'download' in rels:
-            for match in HREF.finditer(tag):
-                yield urlparse.urljoin(url, htmldecode(match.group(1)))
-
-    for tag in ("<th>Home Page", "<th>Download URL"):
-        pos = page.find(tag)
-        if pos!=-1:
-            match = HREF.search(page,pos)
-            if match:
-                yield urlparse.urljoin(url, htmldecode(match.group(1)))
-
-user_agent = "Python-urllib/%s distutils2/%s" % (
-    sys.version[:3], __distutils2_version__
-)
-
-class SimpleIndex(Environment):
-    """A distribution index that scans web pages for download URLs"""
-
-    def __init__(self, index_url="http://pypi.python.org/simple", hosts=('*',),
-        follow_externals=True, *args, **kw):
-        self._init_logging()
-        Environment.__init__(self,*args,**kw)
-        self.index_url = index_url + "/"[:not index_url.endswith('/')]
-        self.scanned_urls = {}
-        self.fetched_urls = {}
-        self.package_pages = {}
-        self.allows = re.compile('|'.join(map(translate,hosts))).match
-        self.to_scan = []
-        self.follow_externals = follow_externals
-
-    def _init_logging(self):
-        import sys
-#        logging.basicConfig(stream=sys.stdout, level=logging.DEBUG)
-
-    def process_url(self, url, retrieve=False):
-        """Evaluate a URL as a possible download, and maybe retrieve it"""
-
-        if url in self.scanned_urls and not retrieve:
-            return
-        self.scanned_urls[url] = True
-        if not URL_SCHEME(url):
-            self.process_filename(url)
-            return
-        else:
-            dists = list(distros_for_url(url))
-            if dists:
-                if not self.url_ok(url):
-                    return
-                self.debug("Found link: %s", url)
-
-        if dists or not retrieve or url in self.fetched_urls:
-            map(self.add, dists)
-            return  # don't need the actual page
-
-        if not self.url_ok(url):
-            self.fetched_urls[url] = True
-            return
-
-        self.info("Reading %s", url)
-        f = self.open_url(url, "Download error: %s -- Some packages may not be found!")
-        if f is None: return
-        self.fetched_urls[url] = self.fetched_urls[f.url] = True
-
-        if 'html' not in f.headers.get('content-type', '').lower():
-            f.close()   # not html, we can't process it
-            return
-
-        base = f.url     # handle redirects
-        page = f.read()
-        if sys.version_info >= (3,):
-            charset = f.headers.get_param('charset') or 'latin-1'
-            page = page.decode(charset, "ignore")
-        f.close()
-
-        if url.startswith(self.index_url):
-            if self.follow_externals:
-                for match in HREF.finditer(page):
-                    link = urlparse.urljoin(base, htmldecode(match.group(1)))
-                    self.process_url(link)
-        if url.startswith(self.index_url) and getattr(f,'code',None)!=404:
-            page = self.process_index(url, page)
-
-    def process_filename(self, fn, nested=False):
-        # process filenames or directories
-        if not os.path.exists(fn):
-            self.warn("Not found: %s", fn)
-            return
-
-        if os.path.isdir(fn) and not nested:
-            path = os.path.realpath(fn)
-            for item in os.listdir(path):
-                self.process_filename(os.path.join(path,item), True)
-
-        dists = distros_for_filename(fn)
-        if dists:
-            self.debug("Found: %s", fn)
-            map(self.add, dists)
-
-    def url_ok(self, url, fatal=False):
-        s = URL_SCHEME(url)
-        if (s and s.group(1).lower()=='file') or self.allows(urlparse.urlparse(url)[1]):
-            return True
-        msg = "\nLink to % s ***BLOCKED*** by --allow-hosts\n"
-        if fatal:
-            raise DistutilsError(msg % url)
-        else:
-            self.warn(msg, url)
-        return False
-
-    def scan_egg_links(self, search_path):
-        for item in search_path:
-            if os.path.isdir(item):
-                for entry in os.listdir(item):
-                    if entry.endswith('.egg-link'):
-                        self.scan_egg_link(item, entry)
-
-    def scan_egg_link(self, path, entry):
-        lines = filter(None, map(str.strip, open(os.path.join(path, entry))))
-        if len(lines)==2:
-            for dist in find_distributions(os.path.join(path, lines[0])):
-                dist.location = os.path.join(path, *lines)
-                dist.precedence = SOURCE_DIST
-                self.add(dist)
-
-    def process_index(self,url,page):
-        """Process the content of PyPI page. For each link, if it's a PyPI one,
-        browse it's page, and satinize the output.
+        :param name: the name of the project to find the page
         """
-        def scan(link):
-            # Process a URL to see if it's for a package page
-            if link.startswith(self.index_url):
-                parts = map(
-                    urllib2.unquote, link[len(self.index_url):].split('/')
-                )
-                if len(parts)==2 and '#' not in parts[1]:
-                    # it's a package page, sanitize and index it
-                    pkg = safe_name(parts[0])
-                    ver = safe_version(parts[1])
-                    self.package_pages.setdefault(pkg.lower(),{})[link] = True
-                    return to_filename(pkg), to_filename(ver)
-            return None, None
-
-        # process an index page into the package-page index
-        for match in HREF.finditer(page):
-            try:
-                scan( urlparse.urljoin(url, htmldecode(match.group(1))) )
-            except ValueError:
-                pass
-
-        pkg, ver = scan(url)   # ensure this page is in the page index
-        if pkg:
-            # process individual package page
-            for new_url in find_external_links(url, page):
-                # Process the found URL
-                base, frag = egg_info_for_url(new_url)
-                if base.endswith('.py') and not frag:
-                    if ver:
-                        new_url+='#egg=%s-%s' % (pkg,ver)
-                    else:
-                        self.need_version_info(url)
-                self.scan_url(new_url)
-
-            return PYPI_MD5.sub(
-                lambda m: '<a href="%s#md5=%s">%s</a>' % m.group(1,3,2), page
-            )
-        else:
-            return ""   # no sense double-scanning non-package pages
-
-    def need_version_info(self, url):
-        self.scan_all(
-            "Page at %s links to .py file(s) without version info; an index "
-            "scan is required.", url
-        )
-
-    def scan_all(self, msg=None, *args):
-        if self.index_url not in self.fetched_urls:
-            if msg: self.warn(msg,*args)
-            self.info(
-                "Scanning index of all packages (this may take a while)"
-            )
-        self.scan_url(self.index_url)
-
-    def find_packages(self, requirement):
-        self.scan_url(self.index_url + requirement.unsafe_name+'/')
-
-        if not self.package_pages.get(requirement.key):
-            # Fall back to safe version of the name
-            self.scan_url(self.index_url + requirement.project_name+'/')
-
-        if not self.package_pages.get(requirement.key):
-            # We couldn't find the target package, so search the index page too
-            self.not_found_in_index(requirement)
-
-        for url in list(self.package_pages.get(requirement.key,())):
-            # scan each page that might be related to the desired package
-            self.scan_url(url)
-
-    def obtain(self, requirement, installer=None):
-        self.prescan(); self.find_packages(requirement)
-        for dist in self[requirement.key]:
-            if dist in requirement:
-                return dist
-            self.debug("%s does not match %s", requirement, dist)
-        return super(SimpleIndex, self).obtain(requirement,installer)
-
-    def check_md5(self, cs, info, filename, tfp):
-        if re.match('md5=[0-9a-f]{32}$', info):
-            self.debug("Validating md5 checksum for %s", filename)
-            if cs.hexdigest()<>info[4:]:
-                tfp.close()
-                os.unlink(filename)
-                raise DistutilsError(
-                    "MD5 validation failed for "+os.path.basename(filename)+
-                    "; possible download problem?")
-
-    def add_find_links(self, urls):
-        """Add `urls` to the list that will be prescanned for searches"""
-        for url in urls:
-            if (
-                self.to_scan is None        # if we have already "gone online"
-                or not URL_SCHEME(url)      # or it's a local file/directory
-                or url.startswith('file:')
-                or list(distros_for_url(url))   # or a direct package link
-            ):
-                # then go ahead and process it now
-                self.scan_url(url)
-            else:
-                # otherwise, defer retrieval till later
-                self.to_scan.append(url)
-
-    def prescan(self):
-        """Scan urls scheduled for prescanning (e.g. --find-links)"""
-        if self.to_scan:
-            map(self.scan_url, self.to_scan)
-        self.to_scan = None     # from now on, go ahead and process immediately
-
-    def not_found_in_index(self, requirement):
-        if self[requirement.key]:   # we've seen at least one distro
-            meth, msg = self.info, "Couldn't retrieve index page for %r"
-        else:   # no distros seen for this name, might be misspelled
-            meth, msg = (self.warn,
-                "Couldn't find index page for %r (maybe misspelled?)")
-        meth(msg, requirement.unsafe_name)
-        self.scan_all()
-
-    def download(self, spec, tmpdir=None):
-        """Locate and/or download `spec` to `tmpdir`, returning a local path
-
-        `spec` may be a ``Requirement`` object, or a string containing a URL,
-        an existing local filename, or a project/version requirement spec
-        (i.e. the string form of a ``Requirement`` object).  If it is the URL
-        of a .py file with an unambiguous ``#egg=name-version`` tag (i.e., one
-        that escapes ``-`` as ``_`` throughout), a trivial ``setup.py`` is
-        automatically created alongside the downloaded file.
-
-        If `spec` is a ``Requirement`` object or a string containing a
-        project/version requirement spec, this method returns the location of
-        a matching distribution (possibly after downloading it to `tmpdir`).
-        If `spec` is a locally existing file or directory name, it is simply
-        returned unchanged.  If `spec` is a URL, it is downloaded to a subpath
-        of `tmpdir`, and the local filename is returned.  Various errors may be
-        raised if a problem occurs during downloading.
-        """
-        if tmpdir is None:
-            tmpdir=tempfile.mkdtemp()
-        if not isinstance(spec,Requirement):
-            scheme = URL_SCHEME(spec)
-            if scheme:
-                # It's a url, download it to tmpdir
-                found = self._download_url(scheme.group(1), spec, tmpdir)
-                base, fragment = egg_info_for_url(spec)
-                if base.endswith('.py'):
-                    found = self.gen_setup(found,fragment,tmpdir)
-                return found
-            elif os.path.exists(spec):
-                # Existing file or directory, just return it
-                return spec
-            else:
-                try:
-                    spec = Requirement.parse(spec)
-                except ValueError:
-                    raise DistutilsError(
-                        "Not a URL, existing file, or requirement spec: %r" %
-                        (spec,)
-                    )
-        return getattr(self.fetch_distribution(spec, tmpdir),'location',None)
-
-    def fetch_distribution(self,
-        requirement, tmpdir, force_scan=False, source=False, develop_ok=False
-    ):
-        """Obtain a distribution suitable for fulfilling `requirement`
-
-        `requirement` must be a ``pkg_resources.Requirement`` instance.
-        If necessary, or if the `force_scan` flag is set, the requirement is
-        searched for in the (online) package index as well as the locally
-        installed packages.  If a distribution matching `requirement` is found,
-        the returned distribution's ``location`` is the value you would have
-        gotten from calling the ``download()`` method with the matching
-        distribution's URL or filename.  If no matching distribution is found,
-        ``None`` is returned.
-
-        If the `source` flag is set, only source distributions and source
-        checkout links will be considered.  Unless the `develop_ok` flag is
-        set, development and system eggs (i.e., those using the ``.egg-info``
-        format) will be ignored.
-        """
+        # Browse and index the content of the given PyPI page.
+        # Put all informations about the processed pages in the 
+        # _processed_pages attribute.
+        url = self.index_url + name + "/"
         
-        # process a Requirement
-        self.info("Searching for %s", requirement)
-        skipped = {}
-
-        def find(req):
-            # Find a matching distribution; may be called more than once
-
-            for dist in self[req.key]:
-
-                if dist.precedence==DEVELOP_DIST and not develop_ok:
-                    if dist not in skipped:
-                        self.warn("Skipping development or system egg: %s",dist)
-                        skipped[dist] = 1
-                    continue
-
-                if dist in req and (dist.precedence<=SOURCE_DIST or not source):
-                    self.info("Best match: %s", dist)
-                    return dist.clone(
-                        location=self.download(dist.location, tmpdir)
-                    )
-
-        if force_scan:
-            self.prescan()
-            self.find_packages(requirement)
-
-        dist = find(requirement)
-        if dist is None and self.to_scan is not None:
-            self.prescan()
-            dist = find(requirement)
-
-        if dist is None and not force_scan:
-            self.find_packages(requirement)
-            dist = find(requirement)
-
-        if dist is None:
-            self.warn(
-                "No local packages or download links found for %s%s",
-                (source and "a source distribution of " or ""),
-                requirement,
-            )
-        return dist
-
-    def fetch(self, requirement, tmpdir, force_scan=False, source=False):
-        """Obtain a file suitable for fulfilling `requirement`
-
-        DEPRECATED; use the ``fetch_distribution()`` method now instead.  For
-        backward compatibility, this routine is identical but returns the
-        ``location`` of the downloaded distribution instead of a distribution
-        object.
-        """
-        dist = self.fetch_distribution(requirement,tmpdir,force_scan,source)
-        if dist is not None:
-            return dist.location
-        return None
-    
-    def gen_setup(self, filename, fragment, tmpdir):
-        match = EGG_FRAGMENT.match(fragment)
-        dists = match and [d for d in
-            interpret_distro_name(filename, match.group(1), None) if d.version
-        ] or []
-
-        if len(dists)==1:   # unambiguous ``#egg`` fragment
-            basename = os.path.basename(filename)
-
-            # Make sure the file has been downloaded to the temp dir.
-            if os.path.dirname(filename) != tmpdir:
-                dst = os.path.join(tmpdir, basename)
-                from setuptools.command.easy_install import samefile
-                if not samefile(filename, dst):
-                    shutil.copy2(filename, dst)
-                    filename=dst
-
-            file = open(os.path.join(tmpdir, 'setup.py'), 'w')
-            file.write(
-                "from setuptools import setup\n"
-                "setup(name=%r, version=%r, py_modules=[%r])\n"
-                % (
-                    dists[0].project_name, dists[0].version,
-                    os.path.splitext(basename)[0]
-                )
-            )
-            file.close()
-            return filename
-
-        elif match:
-            raise DistutilsError(
-                "Can't unambiguously interpret project/version identifier %r; "
-                "any dashes in the name or version should be escaped using "
-                "underscores. %r" % (fragment,dists)
-            )
-        else:
-            raise DistutilsError(
-                "Can't process plain .py without an '#egg=name-version'"
-                " suffix to enable automatic setup script generation."
-            )
-
-    dl_blocksize = 8192
-    def _download_to(self, url, filename):
-        self.info("Downloading %s", url)
-        # Download the file
-        fp, tfp, info = None, None, None
-        try:
-            if '#' in url:
-                url, info = url.split('#', 1)
-            fp = self.open_url(url)
-            if isinstance(fp, urllib2.HTTPError):
-                raise DistutilsError(
-                    "Can't download %s: %s %s" % (url, fp.code,fp.msg)
-                )
-            cs = md5()
-            headers = fp.info()
-            blocknum = 0
-            bs = self.dl_blocksize
-            size = -1
-            if "content-length" in headers:
-                size = int(headers["Content-Length"])
-                self.reporthook(url, filename, blocknum, bs, size)
-            tfp = open(filename,'wb')
-            while True:
-                block = fp.read(bs)
-                if block:
-                    cs.update(block)
-                    tfp.write(block)
-                    blocknum += 1
-                    self.reporthook(url, filename, blocknum, bs, size)
-                else:
-                    break
-            if info: self.check_md5(cs, info, filename, tfp)
-            return headers
-        finally:
-            if fp: fp.close()
-            if tfp: tfp.close()
-
-    def reporthook(self, url, filename, blocknum, blksize, size):
-        pass    # no-op
-
-    def open_url(self, url, warning=None):
-        if url.startswith('file:'):
-            return local_open(url)
-        try:
-            return open_with_auth(url)
-        except (ValueError, httplib.InvalidURL), v:
-            msg = ' '.join([str(arg) for arg in v.args])
-            if warning:
-                self.warn(warning, msg)
-            else:
-                raise DistutilsError('%s %s' % (url, msg))
-        except urllib2.HTTPError, v:
-            return v
-        except urllib2.URLError, v:
-            if warning:
-                self.warn(warning, v.reason)
-            else:
-                raise DistutilsError("Download error for %s: %s"
-                                     % (url, v.reason))
-        except httplib.BadStatusLine, v:
-            if warning:
-                self.warn(warning, v.line)
-            else:
-                raise DistutilsError('%s returned a bad status line. '
-                                     'The server might be down, %s' % \
-                                             (url, v.line))
-        except httplib.HTTPException, v:
-            if warning:
-                self.warn(warning, v)
-            else:
-                raise DistutilsError("Download error for %s: %s"
-                                     % (url, v))
-
-    def _download_url(self, scheme, url, tmpdir):
-        # Determine download filename
-        name = filter(None,urlparse.urlparse(url)[2].split('/'))
-        if name:
-            name = name[-1]
-            while '..' in name:
-                name = name.replace('..','.').replace('\\','_')
-        else:
-            name = "__downloaded__"    # default if URL has no path contents
-
-        if name.endswith('.egg.zip'):
-            name = name[:-4]    # strip the extra .zip before download
-
-        filename = os.path.join(tmpdir,name)
-
-        # Download the file
-        if scheme=='svn' or scheme.startswith('svn+'):
-            return self._download_svn(url, filename)
-        elif scheme=='file':
-            return urllib2.url2pathname(urlparse.urlparse(url)[2])
-        else:
-            self.url_ok(url, True)   # raises error if not allowed
-            return self._attempt_download(url, filename)
-
-    def scan_url(self, url):
-        self.process_url(url, True)
-
-    def _attempt_download(self, url, filename):
-        headers = self._download_to(url, filename)
-        if 'html' in headers.get('content-type','').lower():
-            return self._download_html(url, headers, filename)
-        else:
-            return filename
-
-    def _download_html(self, url, headers, filename):
-        file = open(filename)
-        for line in file:
-            if line.strip():
-                # Check for a subversion index page
-                if re.search(r'<title>([^- ]+ - )?Revision \d+:', line):
-                    # it's a subversion index page:
-                    file.close()
-                    os.unlink(filename)
-                    return self._download_svn(url, filename)
-                break   # not an index page
-        file.close()
-        os.unlink(filename)
-        raise DistutilsError("Unexpected HTML page found at "+url)
-
-    def _download_svn(self, url, filename):
-        url = url.split('#',1)[0]   # remove any fragment for svn's sake
-        self.info("Doing subversion checkout from %s to %s", url, filename)
-        os.system("svn checkout -q %s %s" % (url, filename))
-        return filename
-
-    def debug(self, msg, *args):
-        logging.debug(msg, *args)
-
-    def info(self, msg, *args):
-        logging.info(msg, *args)
-
-    def warn(self, msg, *args):
-        logging.warn(msg, *args)
-
-# This pattern matches a character entity reference (a decimal numeric
-# references, a hexadecimal numeric reference, or a named reference).
-entity_sub = re.compile(r'&(#(\d+|x[\da-fA-F]+)|[\w.:-]+);?').sub
-
-def uchr(c):
-    if not isinstance(c, int):
-        return c
-    if c>255: 
-        return unichr(c)
-    return chr(c)
-
-def decode_entity(match):
-    what = match.group(1)
-    if what.startswith('#x'):
-        what = int(what[2:], 16)
-    elif what.startswith('#'):
-        what = int(what[1:])
-    else:
-        from htmlentitydefs import name2codepoint
-        what = name2codepoint.get(what, match.group(0))
-    return uchr(what)
-
-def htmldecode(text):
-    """Decode HTML entities in the given text."""
-    return entity_sub(decode_entity, text)
-
-def socket_timeout(timeout=15):
-    def _socket_timeout(func):
-        def _socket_timeout(*args, **kwargs):
-            old_timeout = socket.getdefaulttimeout()
-            socket.setdefaulttimeout(timeout)
-            try:
-                return func(*args, **kwargs)
-            finally:
-                socket.setdefaulttimeout(old_timeout)
-        return _socket_timeout
-    return _socket_timeout
-
-def open_with_auth(url):
-    """Open a urllib2 request, handling HTTP authentication"""
-
-    scheme, netloc, path, params, query, frag = urlparse.urlparse(url)
-
-    if scheme in ('http', 'https'):
-        auth, host = urllib2.splituser(netloc)
-    else:
-        auth = None
-
-    if auth:
-        auth = "Basic " + urllib2.unquote(auth).encode('base64').strip()
-        new_url = urlparse.urlunparse((scheme,host,path,params,query,frag))
-        request = urllib2.Request(new_url)
-        request.add_header("Authorization", auth)
-    else:
-        request = urllib2.Request(url)
-
-    request.add_header('User-Agent', user_agent)
-    fp = urllib2.urlopen(request)
-
-    if auth:
-        # Put authentication info back into request URL if same host,
-        # so that links found on the page will work
-        s2, h2, path2, param2, query2, frag2 = urlparse.urlparse(fp.url)
-        if s2==scheme and h2==host:
-            fp.url = urlparse.urlunparse((s2,netloc,path2,param2,query2,frag2))
-
-    return fp
-
-
-# adding a timeout to avoid freezing package_index
-open_with_auth = socket_timeout(_SOCKET_TIMEOUT)(open_with_auth)
-
-def fix_sf_url(url):
-    return url      # backward compatibility
-
-def local_open(url):
-    """Read a local path, with special support for directories"""
-    scheme, server, path, param, query, frag = urlparse.urlparse(url)
-    filename = urllib2.url2pathname(path)
-    if os.path.isfile(filename):
-        return urllib2.urlopen(url)
-    elif path.endswith('/') and os.path.isdir(filename):
-        files = []
-        for f in os.listdir(filename):
-            if f=='index.html':
-                fp = open(os.path.join(filename,f),'rb')
-                body = fp.read()
-                fp.close()
-                break
-            elif os.path.isdir(os.path.join(filename,f)):
-                f+='/'
-            files.append("<a href=%r>%s</a>" % (f,f))
-        else:
-            body = ("<html><head><title>%s</title>" % url) + \
-                "</head><body>%s</body></html>" % '\n'.join(files)
-        status, message = 200, "OK"
-    else:
-        status, message, body = 404, "Path not found", "Not found"
-
-    return urllib2.HTTPError(url, status, message,
-            {'content-type':'text/html'}, cStringIO.StringIO(body))
+        # Search for external links here, and process them if needed.
+        # use the _browsable method to know if we can process the found links.
+        for link in found_links:
+            if self._browsable(link):
+                self._search_in_url(link)
diff --git a/src/distutils2/version.py b/src/distutils2/version.py
--- a/src/distutils2/version.py
+++ b/src/distutils2/version.py
@@ -3,6 +3,10 @@
 
 from distutils2.errors import IrrationalVersionError, HugeMajorVersionNumError
 
+__all__ = ['NormalizedVersion', 'suggest_normalized_version',
+           'VersionPredicate', 'is_valid_version', 'is_valid_versions',
+           'is_valid_predicate']
+
 # A marker used in the second and third parts of the `parts` tuple, for
 # versions that don't have those segments, to sort properly. An example
 # of versions in sort order ('highest' last):
@@ -18,9 +22,9 @@
 #                                                              |
 #   'dev' < 'f' ----------------------------------------------/
 # Other letters would do, but 'f' for 'final' is kind of nice.
-FINAL_MARKER = ('f',)
+_FINAL_MARKER = ('f',)
 
-VERSION_RE = re.compile(r'''
+_VERSION_RE = re.compile(r'''
     ^
     (?P<version>\d+\.\d+)          # minimum 'N.N'
     (?P<extraversion>(?:\.\d+)*)   # any number of extra '.N' segments
@@ -70,13 +74,13 @@
         self._parse(s, error_on_huge_major_num)
 
     @classmethod
-    def from_parts(cls, version, prerelease=FINAL_MARKER,
-                   devpost=FINAL_MARKER):
+    def from_parts(cls, version, prerelease=_FINAL_MARKER,
+                   devpost=_FINAL_MARKER):
         return cls(cls.parts_to_str((version, prerelease, devpost)))
 
     def _parse(self, s, error_on_huge_major_num=True):
         """Parses a string version into parts."""
-        match = VERSION_RE.search(s)
+        match = _VERSION_RE.search(s)
         if not match:
             raise IrrationalVersionError(s)
 
@@ -98,7 +102,7 @@
                                          pad_zeros_length=1)
             parts.append(tuple(block))
         else:
-            parts.append(FINAL_MARKER)
+            parts.append(_FINAL_MARKER)
 
         # postdev
         if groups.get('postdev'):
@@ -106,14 +110,14 @@
             dev = groups.get('dev')
             postdev = []
             if post is not None:
-                postdev.extend([FINAL_MARKER[0], 'post', int(post)])
+                postdev.extend([_FINAL_MARKER[0], 'post', int(post)])
                 if dev is None:
-                    postdev.append(FINAL_MARKER[0])
+                    postdev.append(_FINAL_MARKER[0])
             if dev is not None:
                 postdev.extend(['dev', int(dev)])
             parts.append(tuple(postdev))
         else:
-            parts.append(FINAL_MARKER)
+            parts.append(_FINAL_MARKER)
         self.parts = tuple(parts)
         if error_on_huge_major_num and self.parts[0][0] > 1980:
             raise HugeMajorVersionNumError("huge major version number, %r, "
@@ -154,10 +158,10 @@
         # XXX This doesn't check for invalid tuples
         main, prerel, postdev = parts
         s = '.'.join(str(v) for v in main)
-        if prerel is not FINAL_MARKER:
+        if prerel is not _FINAL_MARKER:
             s += prerel[0]
             s += '.'.join(str(v) for v in prerel[1:])
-        if postdev and postdev is not FINAL_MARKER:
+        if postdev and postdev is not _FINAL_MARKER:
             if postdev[0] == 'f':
                 postdev = postdev[1:]
             i = 0
@@ -364,7 +368,7 @@
                 return False
         return True
 
-class Versions(VersionPredicate):
+class _Versions(VersionPredicate):
     def __init__(self, predicate):
         predicate = predicate.strip()
         match = _PLAIN_VERSIONS.match(predicate)
@@ -375,7 +379,7 @@
         self.predicates = [_split_predicate(pred.strip())
                            for pred in predicates.split(',')]
 
-class Version(VersionPredicate):
+class _Version(VersionPredicate):
     def __init__(self, predicate):
         predicate = predicate.strip()
         match = _PLAIN_VERSIONS.match(predicate)
@@ -394,7 +398,7 @@
 
 def is_valid_versions(predicate):
     try:
-        Versions(predicate)
+        _Versions(predicate)
     except (ValueError, IrrationalVersionError):
         return False
     else:
@@ -402,7 +406,7 @@
 
 def is_valid_version(predicate):
     try:
-        Version(predicate)
+        _Version(predicate)
     except (ValueError, IrrationalVersionError):
         return False
     else:

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: finished refactoring. Updated tests
Message-ID: <E1OVLoV-0000jU-S7@dinsdale.python.org>

tarek.ziade pushed c598326257f0 to distutils2:

http://hg.python.org/distutils2/rev/c598326257f0
changeset:   310:c598326257f0
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Mon Jun 21 15:00:15 2010 +0200
summary:     finished refactoring. Updated tests
files:       src/distutils2/pypi/dist.py, src/distutils2/pypi/errors.py, src/distutils2/pypi/simple.py, src/distutils2/tests/pypiserver/with_externals/simple/foobar/index.html, src/distutils2/tests/test_pypi_dist.py, src/distutils2/tests/test_pypi_simple.py

diff --git a/src/distutils2/pypi/dist.py b/src/distutils2/pypi/dist.py
--- a/src/distutils2/pypi/dist.py
+++ b/src/distutils2/pypi/dist.py
@@ -47,28 +47,13 @@
                 archive_name = archive_name[:-len(ext)]
                 extension_matched = True
 
-        name = None
-        # get the name from probable_dist_name
-        if probable_dist_name is not None:
-            if probable_dist_name in archive_name:
-                name = probable_dist_name
-
-        if name is None:
-            # determine the name and the version
-            splits = archive_name.split("-")
-            version = splits[-1]
-            name = "-".join(splits[:-1])
-        else:
-            # just determine the version
-            version = archive_name[len(name):]
-            if version.startswith("-"):
-                version = version[1:]
-        
-        version = suggest_normalized_version(version)
+        name, version = split_archive_name(archive_name)
         if extension_matched is True:
             return PyPIDistribution(name, version, url=url, md5_hash=md5_hash)
 
     def __init__(self, name, version, type=None, url=None, md5_hash=None):
+        """Create a new instance of PyPIDistribution
+        """
         self.name = name
         self.version = version
         self.url = url
@@ -89,13 +74,42 @@
         # if we do not have downloaded it yet, do it.
         if self._downloaded_path is None:
             # download logic goes here
-            real_path = "" 
+            real_path = ""
         return real_path
 
-    def __str__(self):
-        """string representation of the PyPIDistribution"""
-        return "%s-%s" % (self.name, self.version)
+    def __repr__(self):
+        return "<%s %s (%s)>" % (self.__class__.__name__, self.name, self.version)
 
+    def _check_is_comparable(self, other):
+        if not isinstance(other, PyPIDistribution):
+            raise TypeError("cannot compare %s and %s" 
+                % (type(self).__name__, type(other).__name__))
+        elif self.name != other.name:
+            raise TypeError("cannot compare %s and %s"
+                % (self.name, other.name))
+
+    def __eq__(self, other):
+        self._check_is_comparable(other)
+        return self.version == other.version
+
+    def __lt__(self, other):
+        self._check_is_comparable(other)
+        return self.version < other.version
+
+    def __ne__(self, other):
+        return not self.__eq__(other)
+
+    def __gt__(self, other):
+        return not (self.__lt__(other) or self.__eq__(other))
+
+    def __le__(self, other):
+        return self.__eq__(other) or self.__lt__(other)
+
+    def __ge__(self, other):
+        return self.__eq__(other) or self.__gt__(other)
+
+    # See http://docs.python.org/reference/datamodel#object.__hash__
+    __hash__ = object.__hash__
 
 class PyPIDistributions(list):
     """A container of PyPIDistribution objects.
@@ -104,9 +118,53 @@
     """
 
     def filter(self, predicate):
-        """Filter the distributions and return just the one matching the given
-        predicate.
+        """Filter the distributions and return a subset of distributions that
+        match the given predicate
         """
-        return [dist for dist in self if dist.name == predicate.name and
-            predicate.match(dist.version)]
+        return PyPIDistributions(
+            [dist for dist in self if dist.name == predicate.name and
+            predicate.match(dist.version)])
 
+    def get_last(self, predicate):
+        """Return the most up to date version, that satisfy the given 
+        predicate
+        """
+        distributions = self.filter(predicate)
+        distributions.sort()
+        return distributions[-1]
+
+def split_archive_name(archive_name, probable_name=None):
+    """Split an archive name into two parts: name and version.
+
+    Return the tuple (name, version)
+    """
+    # Try to determine wich part is the name and wich is the version using the "-"
+    # separator. Take the larger part to be the version number then reduce if this
+    # not works.
+    def eager_split(str, maxsplit=2):
+        # split using the "-" separator
+        splits = str.rsplit("-", maxsplit)
+        name = splits[0]
+        version = "-".join(splits[1:])
+        if version.startswith("-"):
+            version = version[1:]
+        if suggest_normalized_version(version) is None and maxsplit >= 0:
+            # we dont get a good version number: recurse !
+            return eager_split(str, maxsplit-1)
+        else:
+            return (name, version)
+    if probable_name is not None:
+        probable_name = probable_name.lower()
+    name = None
+    if probable_name is not None and probable_name in archive_name:
+        # we get the name from probable_name, if given.
+        name = probable_name
+        version = archive_name.lstrip(name)
+    else:
+        name, version = eager_split(archive_name)
+    
+    version = suggest_normalized_version(version)
+    if version != "" and name != "":
+        return (name.lower(), version)
+    else:
+        raise CantParseArchiveName(archive_name) 
diff --git a/src/distutils2/pypi/errors.py b/src/distutils2/pypi/errors.py
--- a/src/distutils2/pypi/errors.py
+++ b/src/distutils2/pypi/errors.py
@@ -9,4 +9,7 @@
     """The base class for errors of the pypi python package."""
 
 class DistributionNotFound(PyPIError):
-    """Raised when no distribution match the given requirements."""
+    """No distribution match the given requirements."""
+
+class CantParseArchiveName(PyPIError):
+    """An archive name can't be parsed to find distribution name and version"""
diff --git a/src/distutils2/pypi/simple.py b/src/distutils2/pypi/simple.py
--- a/src/distutils2/pypi/simple.py
+++ b/src/distutils2/pypi/simple.py
@@ -9,6 +9,7 @@
 import sys
 import re
 import urllib2
+import httplib
 import socket
 try:
     from hashlib import md5
@@ -23,6 +24,7 @@
 
 # -- Constants -----------------------------------------------
 PYPI_DEFAULT_INDEX_URL = "http://pypi.python.org/simple/"
+DEFAULT_HOSTS = ("*.python.org",)
 SOCKET_TIMEOUT=15
 USER_AGENT = "Python-urllib/%s distutils2/%s" % (
     sys.version[:3], __distutils2_version__
@@ -59,8 +61,8 @@
     """Provides useful tools to request the Python Package Index simple API
     """
 
-    def __init__(self, url=PYPI_DEFAULT_INDEX_URL, 
-        hosts=('*',), follow_externals=False):
+    def __init__(self, url=PYPI_DEFAULT_INDEX_URL, hosts=DEFAULT_HOSTS, 
+        mirrors=[]):
         """Class constructor.
 
         :param index_url: the url of the simple index to search on.
@@ -68,10 +70,12 @@
         differenciate with external hosts)
         :param follow_externals: tell if follow external links is needed or not
         Default is False
+        :param mirrors: a list of mirrors to check out if problems occurs while
+        working with the one given in "url"
         """
         self.index_url = url
-        self.follow_externals=follow_externals
-         
+        self.mirrors = mirrors
+        self._hosts = hosts
         # create a regexp to match all given hosts
         self._allowed_hosts = re.compile('|'.join(map(translate,hosts))).match
 
@@ -81,14 +85,27 @@
         self._processed_urls = []
         self._distributions = {}
 
-    def get_distributions(self, requirements):
+    def get(self, requirements):
+        """Browse the PyPI index to find distributions that fullfil the 
+        given requirements, and return the most recent one.
+        """
+        predicate = self._get_version_predicate(requirements)
+        dists = self.find(predicate)
+
+        if len(dists) == 0:
+            raise DistributionNotFound(requirements)
+
+        return dists.get_last(predicate) 
+
+    def find(self, requirements):
         """Browse the PyPI to find distributions that fullfil the given 
         requirements.
 
         :param requirements: A project name and it's distribution, using 
-        version specifiers, as described in PEP345. 
+        version specifiers, as described in PEP345. You can pass either a 
+        version.VersionPredicate or a string.
         """
-        requirements = VersionPredicate(requirements)
+        requirements = self._get_version_predicate(requirements)
         
         # process the index for this project
         self._process_pypi_page(requirements.name)
@@ -99,10 +116,7 @@
         else:
             dists = []
 
-        if dists is not []:
-            return dists
-        else:
-            raise DistributionNotFound(requirements.name)
+        return dists
 
     def download(self, requirements, temp_path=None):
         """Download the distribution, using the requirements.
@@ -114,9 +128,17 @@
 
         Returns the complete absolute path to the downloaded archive.
         """
-        distributions = self.get_distributions(requirements)
-        return distributions.get_last(requirements) \
-            .download(temp_path=temp_path)
+        requirements = self._get_version_predicate(requirements)
+        distributions = self.find(requirements)
+        return distributions.get_last(requirements).download(path=temp_path)
+
+    def _get_version_predicate(self, requirements):
+        """Return a VersionPredicate object, from a string or an already 
+        existing object.
+        """
+        if isinstance(requirements, str):
+            requirements = VersionPredicate(requirements)
+        return requirements
 
     def _is_browsable(self, url):
         """Tell if the given URL needs to be browsed or not, according to the
@@ -125,14 +147,13 @@
         It uses the follow_externals and the hosts list to tell if the given 
         url is browsable or not. 
         """
-        if self.follow_externals is True:
-            return True
         return True if self._allowed_hosts(
             urlparse.urlparse(url).netloc) else False
 
     def _is_distribution(self, link):
         """Tell if the given URL matches to a distribution name or not.
         """
+        #XXX find a better way to check that links are distributions
         for ext in EXTENSIONS:
             if ext in link:
                 return True
@@ -189,31 +210,44 @@
     def _open_url(self, url):
         """Open a urllib2 request, handling HTTP authentication
         """
-        scheme, netloc, path, params, query, frag = urlparse.urlparse(url)
+        try:
+            scheme, netloc, path, params, query, frag = urlparse.urlparse(url)
 
-        if scheme in ('http', 'https'):
-            auth, host = urllib2.splituser(netloc)
-        else:
-            auth = None
+            if scheme in ('http', 'https'):
+                auth, host = urllib2.splituser(netloc)
+            else:
+                auth = None
 
-        if auth:
-            auth = "Basic " + urllib2.unquote(auth).encode('base64').strip()
-            new_url = urlparse.urlunparse((scheme,host,path,params,query,frag))
-            request = urllib2.Request(new_url)
-            request.add_header("Authorization", auth)
-        else:
-            request = urllib2.Request(url)
-        request.add_header('User-Agent', USER_AGENT)
-        fp = urllib2.urlopen(request)
+            if auth:
+                auth = "Basic " + urllib2.unquote(auth).encode('base64').strip()
+                new_url = urlparse.urlunparse((scheme,host,path,params,query,frag))
+                request = urllib2.Request(new_url)
+                request.add_header("Authorization", auth)
+            else:
+                request = urllib2.Request(url)
+            request.add_header('User-Agent', USER_AGENT)
+            fp = urllib2.urlopen(request)
 
-        if auth:
-            # Put authentication info back into request URL if same host,
-            # so that links found on the page will work
-            s2, h2, path2, param2, query2, frag2 = urlparse.urlparse(fp.url)
-            if s2==scheme and h2==host:
-                fp.url = urlparse.urlunparse((s2,netloc,path2,param2,query2,frag2))
+            if auth:
+                # Put authentication info back into request URL if same host,
+                # so that links found on the page will work
+                s2, h2, path2, param2, query2, frag2 = urlparse.urlparse(fp.url)
+                if s2==scheme and h2==host:
+                    fp.url = urlparse.urlunparse((s2,netloc,path2,param2,query2,frag2))
 
-        return fp
+            return fp
+        except (ValueError, httplib.InvalidURL), v:
+            msg = ' '.join([str(arg) for arg in v.args])
+            raise PyPIError('%s %s' % (url, msg))
+        except urllib2.HTTPError, v:
+            return v
+        except urllib2.URLError, v:
+            raise PyPIError("Download error for %s: %s" % (url, v.reason))
+        except httplib.BadStatusLine, v:
+            raise PyPIError('%s returned a bad status line. '
+                'The server might be down, %s' % (url, v.line))
+        except httplib.HTTPException, v:
+            raise PyPIError("Download error for %s: %s" % (url, v))
 
     def _decode_entity(self, match):
         what = match.group(1)
diff --git a/src/distutils2/tests/pypiserver/with_externals/simple/foobar/index.html b/src/distutils2/tests/pypiserver/with_externals/simple/foobar/index.html
--- a/src/distutils2/tests/pypiserver/with_externals/simple/foobar/index.html
+++ b/src/distutils2/tests/pypiserver/with_externals/simple/foobar/index.html
@@ -1,4 +1,4 @@
 <html><body>
-<a href="/foobar-0.1.tar.gz#md5=0_correct_md5">foobar-0.1.tar.gz</a><br/>
+<a href="/foobar-0.1.tar.gz#md5=12345678901234567">foobar-0.1.tar.gz</a><br/>
 <a href="../../external/external.html" rel="homepage">external homepage</a><br/>
 </body></html>
diff --git a/src/distutils2/tests/test_pypi_dist.py b/src/distutils2/tests/test_pypi_dist.py
--- a/src/distutils2/tests/test_pypi_dist.py
+++ b/src/distutils2/tests/test_pypi_dist.py
@@ -1,7 +1,8 @@
 """Tests for the distutils2.pypi.dist module.
 """
 
-from distutils2.pypi.dist import PyPIDistribution, PyPIDistributions
+from distutils2.pypi.dist import PyPIDistribution as Dist, \
+    PyPIDistributions as Dists, split_archive_name
 from distutils2.tests.support import unittest
 from distutils2.version import VersionPredicate
 
@@ -11,7 +12,7 @@
     def test_instanciation(self):
         """Test the Distribution class provides us the good attributes when
         given on construction"""
-        dist = PyPIDistribution("FooBar", "1.1")
+        dist = Dist("FooBar", "1.1")
         self.assertEqual("FooBar", dist.name)
         self.assertEqual("1.1", dist.version)
 
@@ -19,11 +20,11 @@
         """Test that the Distribution object can be built from a single URL"""
         url_list = {
             'FooBar-1.1.0.tar.gz': {
-                'name': 'FooBar',
+                'name': 'foobar', # lowercase the name
                 'version': '1.1.0',
             },
             'Foo-Bar-1.1.0.zip': {
-                'name': 'Foo-Bar',
+                'name': 'foo-bar', # keep the dash
                 'version': '1.1.0',
             },
             'foobar-1.1b2.tar.gz#md5=123123123123123': {
@@ -31,13 +32,41 @@
                 'version': '1.1b2',
                 'url':'http://test.tld/foobar-1.1b2.tar.gz', #without md5 hash
                 'md5_hash': '123123123123123',
+            }, 
+            'foobar-1.1-rc2.tar.gz': { # use suggested name
+                'name': 'foobar',
+                'version': '1.1c2',
+                'url':'http://test.tld/foobar-1.1-rc2.tar.gz',
             }
         }
 
         for url, attributes in url_list.items():
-            dist = PyPIDistribution.from_url("http://test.tld/"+url)
+            dist = Dist.from_url("http://test.tld/"+url)
             for attribute, value in attributes.items():
-                self.assertEqual(getattr(dist, attribute), value) 
+                self.assertEqual(getattr(dist, attribute), value)
+
+    def test_comparaison(self):
+        """Test that we can compare PyPIDistributions"""
+        foo1 = Dist("foo", "1.0")
+        foo2 = Dist("foo", "2.0")
+        bar = Dist("bar", "2.0")
+        # assert we use the version to compare
+        self.assertTrue(foo1 < foo2)
+        self.assertFalse(foo1 > foo2)
+        self.assertFalse(foo1 == foo2)
+
+        # assert we can't compare dists with different names
+        self.assertRaises(TypeError, foo1.__eq__,bar)
+
+    def test_split_archive_name(self):
+        """Test we can split the archive names"""
+        names = {
+            'foo-bar-baz-1.0-rc2': ('foo-bar-baz','1.0c2'),
+            'foo-bar-baz-1.0': ('foo-bar-baz','1.0'),
+            'foobarbaz-1.0': ('foobarbaz','1.0'),
+        }
+        for name, results in names.items():
+            self.assertEqual(results, split_archive_name(name))
 
 class TestPyPIDistributions(unittest.TestCase):
     """test the pypi.distr.PyPIDistributions class"""
@@ -45,11 +74,11 @@
     def test_filter(self):
         """Test we filter the distributions the right way, using version
         predicate match method"""
-        dists = PyPIDistributions((
-            PyPIDistribution("FooBar", "1.1"),
-            PyPIDistribution("FooBar", "1.1.1"),
-            PyPIDistribution("FooBar", "1.2"),
-            PyPIDistribution("FooBar", "1.2.1"),
+        dists = Dists((
+            Dist("FooBar", "1.1"),
+            Dist("FooBar", "1.1.1"),
+            Dist("FooBar", "1.2"),
+            Dist("FooBar", "1.2.1"),
         ))
         filtered = dists.filter(VersionPredicate("FooBar (<1.2)"))
         self.assertNotIn(dists[2], filtered)
diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -15,11 +15,14 @@
 
 class PyPISimpleTestCase(unittest2.TestCase):
 
-    def _get_simple_package_index(self, server, base_url="/simple/", *args,
-            **kwargs):
+    def _get_simple_index(self, server, base_url="/simple/", hosts=None, 
+        *args, **kwargs):
         """Build and return a SimpleSimpleIndex instance, with the test server
         urls
         """
+        if hosts is None:
+            hosts = (server.full_address,)
+        kwargs['hosts'] = hosts
         return simple.SimpleIndex(server.full_address + base_url, *args,
             **kwargs)
 
@@ -27,7 +30,7 @@
         index = simple.SimpleIndex()
         url = 'http://127.0.0.1:0/nonesuch/test_simple'
         try:
-            v = index.open_url(url)
+            v = index._open_url(url)
         except Exception, v:
             self.assertTrue(url in str(v))
         else:
@@ -41,7 +44,7 @@
 
         url = 'url:%20https://svn.plone.org/svn/collective/inquant.contentmirror.plone/trunk'
         try:
-            v = index.open_url(url)
+            v = index._open_url(url)
         except Exception, v:
             self.assertTrue(url in str(v))
         else:
@@ -56,7 +59,7 @@
         url = 'http://example.com'
         try:
             try:
-                v = index.open_url(url)
+                v = index._open_url(url)
             except Exception, v:
                 self.assertTrue('line' in str(v))
             else:
@@ -67,7 +70,7 @@
         # issue 20
         url = 'http://http://svn.pythonpaste.org/Paste/wphp/trunk'
         try:
-            index.open_url(url)
+            index._open_url(url)
         except Exception, v:
             self.assertTrue('nonnumeric port' in str(v))
 
@@ -80,7 +83,7 @@
             index.process_index(url, page)
 
     def test_browsable_urls(self):
-        """Test we can given PyPI index a list of allowed hosts
+        """Test we can give PyPI index a list of allowed hosts
 
         """
         index = simple.SimpleIndex(hosts=["pypi.python.org","test.org"])
@@ -107,33 +110,43 @@
         """Browse the index, asking for a specified distribution version
         """
         # The PyPI index contains links for version 1.0, 1.1, 2.0 and 2.0.1
-        # We query only for the version 1.1, so all distributions must me
+        # We query only for the version 1.1, so all distributions must be
         # filled in the package_index (as the url has been scanned), but
-        # obtain must only return the one we want.
-        pi = self._get_simple_package_index(server)
-        last_distribution = pi.obtain(simple.Requirement.parse("foobar"))
+        # "get" must only return the one we want.
+        pi = self._get_simple_index(server)
+        last_distribution = pi.get("foobar")
 
         # we have scanned the index page
-        self.assertTrue(pi.package_pages.has_key('foobar'))
         self.assertIn(server.full_address + "/simple/foobar/",
-            pi.package_pages['foobar'])
+            pi._processed_urls)
         
         # we have found 4 distributions in this page
-        self.assertEqual(len(pi["foobar"]), 4)
+        self.assertEqual(len(pi._distributions["foobar"]), 4)
 
         # and returned the most recent one
         self.assertEqual(last_distribution.version, '2.0.1')
 
+    def test_is_browsable(self):
+        # allow all hosts
+        index = simple.SimpleIndex(hosts=("*",))
+        self.assertTrue(index._is_browsable("http://an-external.link/path"))
+        self.assertTrue(index._is_browsable("pypi.test.tld/a/path"))
+        
+        # specify a list of hosts we want to allow
+        index = simple.SimpleIndex(hosts=("*.test.tld",))
+        self.assertFalse(index._is_browsable("http://an-external.link/path"))
+        self.assertTrue(index._is_browsable("http://pypi.test.tld/a/path"))
+
     @use_pypi_server("with_externals")
     def test_process_external_pages(self, server):
-        """Include external pages in obtain process
+        """Include external pages 
         """
         # Try to request the package index, wich contains links to "externals"
         # resources. They have to  be scanned too.
-        pi = self._get_simple_package_index(server, follow_externals=True)
-        pi.obtain(simple.Requirement.parse("foobar"))
+        pi = self._get_simple_index(server, hosts=("*",))
+        pi.get("foobar")
         self.assertIn(server.full_address + "/external/external.html",
-            pi.scanned_urls.keys())
+            pi._processed_urls)
 
     @use_pypi_server("with_real_externals")
     def test_disable_external_pages(self, server):
@@ -141,10 +154,10 @@
         """
         # Test that telling the simple pyPI client to not retreive external
         # works
-        pi = self._get_simple_package_index(server, follow_externals=False)
-        pi.obtain(simple.Requirement.parse("foobar"))
+        pi = self._get_simple_index(server, hosts=(server.full_address,))
+        pi.get("foobar")
         self.assertNotIn(server.full_address + "/external/external.html",
-            pi.scanned_urls.keys())
+            pi._processed_urls)
 
     @use_pypi_server("downloads_with_md5")
     def test_download_package(self, server):
@@ -154,32 +167,29 @@
         # If we request a download specific version of a distribution, 
         # the system must download it, check the md5 and unpack it in a 
         # temporary location, that must be returned by the lib.
-        pi = self._get_simple_package_index(server)
+        pi = self._get_simple_index(server)
 
         # assert we can download a specific version
-        temp_location_1 = pi.download(
-            simple.Requirement.parse("foobar == 0.1"))
+        temp_location_1 = pi.download("foobar (0.1)")
         self.assertIn("foobar-0.1.tar.gz", temp_location_1)
-        paths.append(temp_location_1)
+        paths.append(temp_location_1) # to delete later
         
         # assert we take the latest
-        temp_location_2 = pi.download(simple.Requirement.parse("foobar"))
+        temp_location_2 = pi.download("foobar")
         self.assertIn("foobar-0.1.tar.gz", temp_location_2)
         paths.append(temp_location_2)
 
         # we also can specify a temp location
         specific_temp_location = tempfile.mkdtemp()
-        returned_temp_location = pi.download(
-            simple.Requirement.parse("foobar"), 
-            tmpdir=specific_temp_location)
+        returned_temp_location = pi.download("foobar", specific_temp_location)
         self.assertIn(specific_temp_location, returned_temp_location)
         paths.append(returned_temp_location)
 
         # raise an error if we couldnt manage to get the file with a the good
         # md5 hash
-        self.assertRaises(DistutilsError, pi.download,
-            simple.Requirement.parse("badmd5"))
-        
+        self.assertRaises(DistutilsError, pi.download, "badmd5")
+       
+        # delete the temp paths
         for path in paths:
             shutil.rmtree(os.path.dirname(path))
 
@@ -203,24 +213,22 @@
         - someone reuploads the package (with a different md5)
         - while easy_installing, an MD5 error occurs because the external link
           is used
-        -> Distribute should use the link from pypi, not the external one.
+        -> The index should use the link from pypi, not the external one.
         """
         # start an index server
         index_url = server.full_address + '/simple/'
 
         # scan a test index
         pi = simple.SimpleIndex(index_url)
-        requirement = simple.Requirement.parse('foobar')
-        pi.find_packages(requirement)
+        dists = pi.find("foobar")
         server.stop()
 
         # the distribution has been found
-        self.assertTrue('foobar' in pi)
+        self.assertTrue(dists is not [])
         # we have only one link, because links are compared without md5
-        self.assertEqual(len(pi['foobar']), 1)
+        self.assertEqual(len(dists), 1)
         # the link should be from the index
-        self.assertTrue('correct_md5' in pi['foobar'][0].location)
-
+        self.assertEqual('12345678901234567', dists[0].md5_hash)
 
 def test_suite():
     return unittest2.makeSuite(PyPISimpleTestCase)

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: merged upstream
Message-ID: <E1OVLoV-0000iy-MA@dinsdale.python.org>

tarek.ziade pushed bee366433f37 to distutils2:

http://hg.python.org/distutils2/rev/bee366433f37
changeset:   302:bee366433f37
parent:      301:e415685da4c5
parent:      197:8a30e3667a75
user:        Konrad Delong <konryd at gmail.com>
date:        Sat Jun 05 13:19:28 2010 +0200
summary:     merged upstream
files:       .hgignore, src/distutils2/cmd.py, src/distutils2/command/cmd.py, src/distutils2/dist.py, src/distutils2/tests/pypi_server.py, src/distutils2/tests/test_pypi_server.py, src/distutils2/tests/test_upload.py, src/distutils2/tests/test_upload_docs.py

diff --git a/.hgignore b/.hgignore
--- a/.hgignore
+++ b/.hgignore
@@ -1,7 +1,5 @@
 .*\.pyc$
 .*\.pyo$
+^src/build
 ^docs/build
 .*\.swp$
-src/build/*
-.*\.pida-metadata
-
diff --git a/src/CONTRIBUTORS.txt b/src/CONTRIBUTORS.txt
--- a/src/CONTRIBUTORS.txt
+++ b/src/CONTRIBUTORS.txt
@@ -9,6 +9,7 @@
 
 Thanks to:
 
+- ??ric Araujo
 - Pior Bastida
 - Titus Brown
 - Nicolas Cadou
diff --git a/src/distutils2/_backport/tests/__init__.py b/src/distutils2/_backport/tests/__init__.py
--- a/src/distutils2/_backport/tests/__init__.py
+++ b/src/distutils2/_backport/tests/__init__.py
@@ -1,12 +1,13 @@
 import os
 import sys
-import unittest2
+
+from distutils2.tests.support import unittest
 
 
 here = os.path.dirname(__file__)
 
 def test_suite():
-    suite = unittest2.TestSuite()
+    suite = unittest.TestSuite()
     for fn in os.listdir(here):
         if fn.startswith("test") and fn.endswith(".py"):
             modname = "distutils2._backport.tests." + fn[:-3]
diff --git a/src/distutils2/_backport/tests/test_pkgutil.py b/src/distutils2/_backport/tests/test_pkgutil.py
--- a/src/distutils2/_backport/tests/test_pkgutil.py
+++ b/src/distutils2/_backport/tests/test_pkgutil.py
@@ -1,6 +1,5 @@
 # -*- coding: utf-8 -*-
 """Tests for PEP 376 pkgutil functionality"""
-import unittest2
 import sys
 import os
 import csv
@@ -12,13 +11,14 @@
 from test.test_support import run_unittest, TESTFN
 
 import distutils2._backport.pkgutil
+from distutils2.tests.support import unittest
 
 # TODO Add a test for getting a distribution that is provided by another
 #   distribution.
 
 # TODO Add a test for absolute pathed RECORD items (e.g. /etc/myapp/config.ini)
 
-class TestPkgUtilDistribution(unittest2.TestCase):
+class TestPkgUtilDistribution(unittest.TestCase):
     """Tests the pkgutil.Distribution class"""
 
     def setUp(self):
@@ -171,7 +171,7 @@
         self.assertEqual(sorted(found), sorted(distinfo_record_paths))
 
 
-class TestPkgUtilFunctions(unittest2.TestCase):
+class TestPkgUtilFunctions(unittest.TestCase):
     """Tests for the new functionality added in PEP 376."""
 
     def setUp(self):
@@ -371,8 +371,8 @@
 
 
 def test_suite():
-    suite = unittest2.TestSuite()
-    testcase_loader = unittest2.loader.defaultTestLoader.loadTestsFromTestCase
+    suite = unittest.TestSuite()
+    testcase_loader = unittest.loader.defaultTestLoader.loadTestsFromTestCase
     suite.addTest(testcase_loader(TestPkgUtilFunctions))
     suite.addTest(testcase_loader(TestPkgUtilDistribution))
     return suite
@@ -384,8 +384,8 @@
     test_main()
 
 def test_suite():
-    suite = unittest2.TestSuite()
-    testcase_loader = unittest2.loader.defaultTestLoader.loadTestsFromTestCase
+    suite = unittest.TestSuite()
+    testcase_loader = unittest.loader.defaultTestLoader.loadTestsFromTestCase
     suite.addTest(testcase_loader(TestPkgUtilFunctions))
     suite.addTest(testcase_loader(TestPkgUtilDistribution))
     return suite
diff --git a/src/distutils2/_backport/tests/test_sysconfig.py b/src/distutils2/_backport/tests/test_sysconfig.py
--- a/src/distutils2/_backport/tests/test_sysconfig.py
+++ b/src/distutils2/_backport/tests/test_sysconfig.py
@@ -4,7 +4,6 @@
 executing have not been removed.
 
 """
-import unittest2
 import sys
 import os
 import shutil
@@ -19,8 +18,9 @@
                        get_path, get_path_names,
                        _get_default_scheme, _subst_vars, _expand_vars,
                        get_scheme_names, _CONFIG_FILE)
+from distutils2.tests.support import unittest
 
-class TestSysConfig(unittest2.TestCase):
+class TestSysConfig(unittest.TestCase):
 
     def setUp(self):
         """Make a copy of sys.path"""
@@ -97,7 +97,7 @@
         substitution key (which would be weird).
 
         """
-        self.failUnlessEqual(_subst_vars('{py{version}}', {'version': '31'}), '{py31}')
+        self.assertEqual(_subst_vars('{py{version}}', {'version': '31'}), '{py31}')
 
     def test_get_paths(self):
         scheme = get_paths()
@@ -262,7 +262,7 @@
         _expand_globals(config)
 
 def test_suite():
-    return unittest2.makeSuite(TestSysConfig)
+    return unittest.makeSuite(TestSysConfig)
 
 def test_main():
     run_unittest(test_suite())
diff --git a/src/distutils2/cmd.py b/src/distutils2/command/cmd.py
rename from src/distutils2/cmd.py
rename to src/distutils2/command/cmd.py
diff --git a/src/distutils2/command/install_egg_info.py b/src/distutils2/command/install_egg_info.py
--- a/src/distutils2/command/install_egg_info.py
+++ b/src/distutils2/command/install_egg_info.py
@@ -4,7 +4,7 @@
 a package's PKG-INFO metadata."""
 
 
-from distutils2.cmd import Command
+from distutils2.command.cmd import Command
 from distutils2 import log
 from distutils2._backport.shutil import rmtree
 import os, sys, re
diff --git a/src/distutils2/config.py b/src/distutils2/config.py
--- a/src/distutils2/config.py
+++ b/src/distutils2/config.py
@@ -6,7 +6,7 @@
 import os
 from ConfigParser import ConfigParser
 
-from distutils2.cmd import Command
+from distutils2.command.cmd import Command
 
 DEFAULT_PYPIRC = """\
 [distutils]
diff --git a/src/distutils2/core.py b/src/distutils2/core.py
--- a/src/distutils2/core.py
+++ b/src/distutils2/core.py
@@ -17,7 +17,7 @@
 
 # Mainly import these so setup scripts can "from distutils2.core import" them.
 from distutils2.dist import Distribution
-from distutils2.cmd import Command
+from distutils2.command.cmd import Command
 from distutils2.config import PyPIRCCommand
 from distutils2.extension import Extension
 
diff --git a/src/distutils2/dist.py b/src/distutils2/dist.py
--- a/src/distutils2/dist.py
+++ b/src/distutils2/dist.py
@@ -493,7 +493,7 @@
         None if the user asked for help on this command.
         """
         # late import because of mutual dependence between these modules
-        from distutils2.cmd import Command
+        from distutils2.command.cmd import Command
 
         # Pull the current command from the head of the command line
         command = args[0]
@@ -607,7 +607,7 @@
         """
         # late import because of mutual dependence between these modules
         from distutils2.core import gen_usage
-        from distutils2.cmd import Command
+        from distutils2.command.cmd import Command
 
         if global_options:
             if display_options:
@@ -904,7 +904,7 @@
 
         Returns the reinitialized command object.
         """
-        from distutils2.cmd import Command
+        from distutils2.command.cmd import Command
         if not isinstance(command, Command):
             command_name = command
             command = self.get_command_obj(command_name)
diff --git a/src/distutils2/tests/__init__.py b/src/distutils2/tests/__init__.py
--- a/src/distutils2/tests/__init__.py
+++ b/src/distutils2/tests/__init__.py
@@ -1,20 +1,25 @@
-"""Test suite for distutils.
+"""Test suite for distutils2.
 
 This test suite consists of a collection of test modules in the
-distutils.tests package.  Each test module has a name starting with
+distutils2.tests package.  Each test module has a name starting with
 'test' and contains a function test_suite().  The function is expected
-to return an initialized unittest2.TestSuite instance.
+to return an initialized unittest.TestSuite instance.
 
-Tests for the command classes in the distutils.command package are
-included in distutils.tests as well, instead of using a separate
-distutils.command.tests package, since command identification is done
+Tests for the command classes in the distutils2.command package are
+included in distutils2.tests as well, instead of using a separate
+distutils2.command.tests package, since command identification is done
 by import rather than matching pre-defined names.
 
+Utility code is included in distutils2.tests.support.  Always import
+unittest from that module, it will be the right version (standard
+library unittest for 2.7 and higher, third-party unittest2 release for
+older versions).
 """
+
 import os
 import sys
 import warnings
-import unittest2
+from distutils2.tests.support import unittest
 
 from test.test_support import TESTFN    # use TESTFN from stdlib/test_support.
 
@@ -23,7 +28,7 @@
 verbose = 1
 
 def test_suite():
-    suite = unittest2.TestSuite()
+    suite = unittest.TestSuite()
     for fn in os.listdir(here):
         if fn.startswith("test") and fn.endswith(".py"):
             modname = "distutils2.tests." + fn[:-3]
@@ -42,17 +47,17 @@
 
 class BasicTestRunner:
     def run(self, test):
-        result = unittest2.TestResult()
+        result = unittest.TestResult()
         test(result)
         return result
 
 
 def _run_suite(suite, verbose_=1):
-    """Run tests from a unittest2.TestSuite-derived class."""
+    """Run tests from a unittest.TestSuite-derived class."""
     global verbose
     verbose = verbose_
     if verbose_:
-        runner = unittest2.TextTestRunner(sys.stdout, verbosity=2)
+        runner = unittest.TextTestRunner(sys.stdout, verbosity=2)
     else:
         runner = BasicTestRunner()
 
@@ -68,22 +73,22 @@
 
 
 def run_unittest(classes, verbose_=1):
-    """Run tests from unittest2.TestCase-derived classes.
+    """Run tests from unittest.TestCase-derived classes.
 
-    Extracted from stdlib test.test_support and modified to support unittest2.
+    Extracted from stdlib test.test_support and modified to support unittest.
     """
-    valid_types = (unittest2.TestSuite, unittest2.TestCase)
-    suite = unittest2.TestSuite()
+    valid_types = (unittest.TestSuite, unittest.TestCase)
+    suite = unittest.TestSuite()
     for cls in classes:
         if isinstance(cls, str):
             if cls in sys.modules:
-                suite.addTest(unittest2.findTestCases(sys.modules[cls]))
+                suite.addTest(unittest.findTestCases(sys.modules[cls]))
             else:
                 raise ValueError("str arguments must be keys in sys.modules")
         elif isinstance(cls, valid_types):
             suite.addTest(cls)
         else:
-            suite.addTest(unittest2.makeSuite(cls))
+            suite.addTest(unittest.makeSuite(cls))
     _run_suite(suite, verbose_)
 
 
@@ -128,4 +133,4 @@
 
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
--- a/src/distutils2/tests/pypi_server.py
+++ b/src/distutils2/tests/pypi_server.py
@@ -8,13 +8,14 @@
 import Queue
 import threading
 import time
-import unittest2
 import urllib2
 from BaseHTTPServer import HTTPServer
 from SimpleHTTPServer import SimpleHTTPRequestHandler
 import os.path
 import select
 
+from distutils2.tests.support import unittest
+
 PYPI_DEFAULT_STATIC_PATH = os.path.dirname(os.path.abspath(__file__)) + "/pypiserver"
 
 def use_pypi_server(*server_args, **server_kwargs):
@@ -30,7 +31,7 @@
         return wrapped
     return wrapper 
 
-class PyPIServerTestCase(unittest2.TestCase):
+class PyPIServerTestCase(unittest.TestCase):
 
     def setUp(self):
         super(PyPIServerTestCase, self).setUp()
diff --git a/src/distutils2/tests/support.py b/src/distutils2/tests/support.py
--- a/src/distutils2/tests/support.py
+++ b/src/distutils2/tests/support.py
@@ -1,5 +1,12 @@
-"""Support code for distutils test cases."""
+"""Support code for distutils2 test cases.
+
+Always import unittest from this module, it will be the right version
+(standard library unittest for 2.7 and higher, third-party unittest2
+release for older versions).
+"""
+
 import os
+import sys
 import shutil
 import tempfile
 from copy import deepcopy
@@ -9,6 +16,13 @@
 from distutils2.log import DEBUG, INFO, WARN, ERROR, FATAL
 from distutils2.core import Distribution
 
+if sys.version_info >= (2, 7):
+    # improved unittest package from 2.7's standard library
+    import unittest
+else:
+    # external release of same package for older versions
+    import unittest2 as unittest
+
 class LoggingSilencer(object):
 
     def setUp(self):
diff --git a/src/distutils2/tests/test_bdist.py b/src/distutils2/tests/test_bdist.py
--- a/src/distutils2/tests/test_bdist.py
+++ b/src/distutils2/tests/test_bdist.py
@@ -1,5 +1,4 @@
 """Tests for distutils.command.bdist."""
-import unittest2
 import sys
 import os
 import tempfile
@@ -10,12 +9,13 @@
 from distutils2.core import Distribution
 from distutils2.command.bdist import bdist
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 from distutils2.spawn import find_executable
 from distutils2 import spawn
 from distutils2.errors import DistutilsExecError
 
 class BuildTestCase(support.TempdirManager,
-                    unittest2.TestCase):
+                    unittest.TestCase):
 
     def test_formats(self):
 
@@ -38,7 +38,7 @@
         self.assertEquals(founded, formats)
 
 def test_suite():
-    return unittest2.makeSuite(BuildTestCase)
+    return unittest.makeSuite(BuildTestCase)
 
 if __name__ == '__main__':
     run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_bdist_dumb.py b/src/distutils2/tests/test_bdist_dumb.py
--- a/src/distutils2/tests/test_bdist_dumb.py
+++ b/src/distutils2/tests/test_bdist_dumb.py
@@ -1,6 +1,5 @@
 """Tests for distutils.command.bdist_dumb."""
 
-import unittest2
 import sys
 import os
 
@@ -12,6 +11,7 @@
     zlib = None
 
 from distutils2.tests import run_unittest
+from distutils2.tests.support import unittest
 
 from distutils2.core import Distribution
 from distutils2.command.bdist_dumb import bdist_dumb
@@ -29,7 +29,7 @@
 class BuildDumbTestCase(support.TempdirManager,
                         support.LoggingSilencer,
                         support.EnvironGuard,
-                        unittest2.TestCase):
+                        unittest.TestCase):
 
     def setUp(self):
         super(BuildDumbTestCase, self).setUp()
@@ -42,7 +42,7 @@
         sys.argv[:] = self.old_sys_argv[1]
         super(BuildDumbTestCase, self).tearDown()
 
-    @unittest2.skipUnless(zlib, "requires zlib")
+    @unittest.skipUnless(zlib, "requires zlib")
     def test_simple_built(self):
 
         # let's create a simple package
@@ -99,7 +99,7 @@
         self.assertEquals(cmd.format, default)
 
 def test_suite():
-    return unittest2.makeSuite(BuildDumbTestCase)
+    return unittest.makeSuite(BuildDumbTestCase)
 
 if __name__ == '__main__':
     run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_bdist_msi.py b/src/distutils2/tests/test_bdist_msi.py
--- a/src/distutils2/tests/test_bdist_msi.py
+++ b/src/distutils2/tests/test_bdist_msi.py
@@ -1,16 +1,16 @@
 """Tests for distutils.command.bdist_msi."""
-import unittest2
 import sys
 
 from distutils2.tests import run_unittest
 
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 class BDistMSITestCase(support.TempdirManager,
                        support.LoggingSilencer,
-                       unittest2.TestCase):
+                       unittest.TestCase):
 
-    @unittest2.skipUnless(sys.platform=="win32", "These tests are only for win32")
+    @unittest.skipUnless(sys.platform=="win32", "These tests are only for win32")
     def test_minial(self):
         # minimal test XXX need more tests
         from distutils2.command.bdist_msi import bdist_msi
@@ -19,7 +19,7 @@
         cmd.ensure_finalized()
 
 def test_suite():
-    return unittest2.makeSuite(BDistMSITestCase)
+    return unittest.makeSuite(BDistMSITestCase)
 
 if __name__ == '__main__':
     run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_bdist_wininst.py b/src/distutils2/tests/test_bdist_wininst.py
--- a/src/distutils2/tests/test_bdist_wininst.py
+++ b/src/distutils2/tests/test_bdist_wininst.py
@@ -1,14 +1,14 @@
 """Tests for distutils.command.bdist_wininst."""
-import unittest2
 
 from distutils2.tests import run_unittest
 
 from distutils2.command.bdist_wininst import bdist_wininst
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 class BuildWinInstTestCase(support.TempdirManager,
                            support.LoggingSilencer,
-                           unittest2.TestCase):
+                           unittest.TestCase):
 
     def test_get_exe_bytes(self):
 
@@ -26,7 +26,7 @@
         self.assertTrue(len(exe_file) > 10)
 
 def test_suite():
-    return unittest2.makeSuite(BuildWinInstTestCase)
+    return unittest.makeSuite(BuildWinInstTestCase)
 
 if __name__ == '__main__':
     run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_build.py b/src/distutils2/tests/test_build.py
--- a/src/distutils2/tests/test_build.py
+++ b/src/distutils2/tests/test_build.py
@@ -1,10 +1,10 @@
 """Tests for distutils.command.build."""
-import unittest2
 import os
 import sys
 
 from distutils2.command.build import build
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 try:
     from sysconfig import get_platform
 except ImportError:
@@ -12,7 +12,7 @@
 
 class BuildTestCase(support.TempdirManager,
                     support.LoggingSilencer,
-                    unittest2.TestCase):
+                    unittest.TestCase):
 
     def test_finalize_options(self):
         pkg_dir, dist = self.create_dist()
@@ -51,7 +51,7 @@
         self.assertEquals(cmd.executable, os.path.normpath(sys.executable))
 
 def test_suite():
-    return unittest2.makeSuite(BuildTestCase)
+    return unittest.makeSuite(BuildTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_build_clib.py b/src/distutils2/tests/test_build_clib.py
--- a/src/distutils2/tests/test_build_clib.py
+++ b/src/distutils2/tests/test_build_clib.py
@@ -1,5 +1,4 @@
 """Tests for distutils.command.build_clib."""
-import unittest2
 import os
 import sys
 
@@ -7,10 +6,11 @@
 from distutils2.errors import DistutilsSetupError
 from distutils2.tests import support
 from distutils2.spawn import find_executable
+from distutils2.tests.support import unittest
 
 class BuildCLibTestCase(support.TempdirManager,
                         support.LoggingSilencer,
-                        unittest2.TestCase):
+                        unittest.TestCase):
 
     def test_check_library_dist(self):
         pkg_dir, dist = self.create_dist()
@@ -137,7 +137,7 @@
         self.assertTrue('libfoo.a' in os.listdir(build_temp))
 
 def test_suite():
-    return unittest2.makeSuite(BuildCLibTestCase)
+    return unittest.makeSuite(BuildCLibTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_build_ext.py b/src/distutils2/tests/test_build_ext.py
--- a/src/distutils2/tests/test_build_ext.py
+++ b/src/distutils2/tests/test_build_ext.py
@@ -6,6 +6,7 @@
 import warnings
 
 import distutils2.tests
+from distutils2.tests.support import unittest
 from distutils2.core import Extension, Distribution
 from distutils2.command.build_ext import build_ext
 from distutils2.tests import support
@@ -17,7 +18,6 @@
 except ImportError:
     from distutils2._backport import sysconfig
 
-import unittest2
 
 # http://bugs.python.org/issue4373
 # Don't load the xx module more than once.
@@ -29,7 +29,7 @@
 
 class BuildExtTestCase(support.TempdirManager,
                        support.LoggingSilencer,
-                       unittest2.TestCase):
+                       unittest.TestCase):
     def setUp(self):
         # Create a simple test environment
         # Note that we're making changes to sys.path
@@ -46,7 +46,7 @@
             build_ext.USER_BASE = site.USER_BASE
 
     # XXX only works with 2.6 > -- dunno why yet
-    @unittest2.skipUnless(sys.version_info >= (2, 6,), 'works for >= 2.6')
+    @unittest.skipUnless(sys.version_info >= (2, 6,), 'works for >= 2.6')
     def test_build_ext(self):
         global ALREADY_TESTED
         xx_c = os.path.join(self.tmp_dir, 'xxmodule.c')
@@ -412,8 +412,8 @@
         if distutils2.tests.verbose:
             print ('test_build_ext: Cannot find source code (test'
                    ' must run in python build dir)')
-        return unittest2.TestSuite()
-    else: return unittest2.makeSuite(BuildExtTestCase)
+        return unittest.TestSuite()
+    else: return unittest.makeSuite(BuildExtTestCase)
 
 if __name__ == '__main__':
     distsutils2.tests.run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_build_py.py b/src/distutils2/tests/test_build_py.py
--- a/src/distutils2/tests/test_build_py.py
+++ b/src/distutils2/tests/test_build_py.py
@@ -3,18 +3,18 @@
 import os
 import sys
 import StringIO
-import unittest2
 
 from distutils2.command.build_py import build_py
 from distutils2.core import Distribution
 from distutils2.errors import DistutilsFileError
 
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 
 class BuildPyTestCase(support.TempdirManager,
                       support.LoggingSilencer,
-                      unittest2.TestCase):
+                      unittest.TestCase):
 
     def test_package_data(self):
         sources = self.mkdtemp()
@@ -90,7 +90,7 @@
             os.chdir(cwd)
             sys.stdout = old_stdout
 
-    @unittest2.skipUnless(hasattr(sys, 'dont_write_bytecode'),
+    @unittest.skipUnless(hasattr(sys, 'dont_write_bytecode'),
                          'dont_write_bytecode support')
     def test_dont_write_bytecode(self):
         # makes sure byte_compile is not used
@@ -109,7 +109,7 @@
         self.assertTrue('byte-compiling is disabled' in self.logs[0][1])
 
 def test_suite():
-    return unittest2.makeSuite(BuildPyTestCase)
+    return unittest.makeSuite(BuildPyTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_build_scripts.py b/src/distutils2/tests/test_build_scripts.py
--- a/src/distutils2/tests/test_build_scripts.py
+++ b/src/distutils2/tests/test_build_scripts.py
@@ -1,7 +1,6 @@
 """Tests for distutils.command.build_scripts."""
 
 import os
-import unittest2
 
 from distutils2.command.build_scripts import build_scripts
 from distutils2.core import Distribution
@@ -11,11 +10,12 @@
     from distutils2._backport import sysconfig
 
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 
 class BuildScriptsTestCase(support.TempdirManager,
                            support.LoggingSilencer,
-                           unittest2.TestCase):
+                           unittest.TestCase):
 
     def test_default_settings(self):
         cmd = self.get_build_scripts_cmd("/foo/bar", [])
@@ -106,7 +106,7 @@
             self.assertTrue(name in built)
 
 def test_suite():
-    return unittest2.makeSuite(BuildScriptsTestCase)
+    return unittest.makeSuite(BuildScriptsTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_ccompiler.py b/src/distutils2/tests/test_ccompiler.py
--- a/src/distutils2/tests/test_ccompiler.py
+++ b/src/distutils2/tests/test_ccompiler.py
@@ -1,11 +1,11 @@
 """Tests for distutils.ccompiler."""
 import os
-import unittest2
 from distutils2.tests import captured_stdout
 
 from distutils2.compiler.ccompiler import (gen_lib_options, CCompiler,
                                  get_default_compiler, customize_compiler)
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 class FakeCompiler(object):
     def library_dir_option(self, dir):
@@ -20,7 +20,7 @@
     def library_option(self, lib):
         return "-l" + lib
 
-class CCompilerTestCase(support.EnvironGuard, unittest2.TestCase):
+class CCompilerTestCase(support.EnvironGuard, unittest.TestCase):
 
     def test_gen_lib_options(self):
         compiler = FakeCompiler()
@@ -54,7 +54,7 @@
         self.assertEquals(comp.exes['archiver'], 'my_ar -arflags')
 
 def test_suite():
-    return unittest2.makeSuite(CCompilerTestCase)
+    return unittest.makeSuite(CCompilerTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_check.py b/src/distutils2/tests/test_check.py
--- a/src/distutils2/tests/test_check.py
+++ b/src/distutils2/tests/test_check.py
@@ -1,14 +1,14 @@
 """Tests for distutils.command.check."""
-import unittest2
 
 from distutils2.command.check import check
 from distutils2.metadata import _HAS_DOCUTILS
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 from distutils2.errors import DistutilsSetupError
 
 class CheckTestCase(support.LoggingSilencer,
                     support.TempdirManager,
-                    unittest2.TestCase):
+                    unittest.TestCase):
 
     def _run(self, metadata=None, **options):
         if metadata is None:
@@ -27,7 +27,7 @@
         # by default, check is checking the metadata
         # should have some warnings
         cmd = self._run()
-        self.assert_(len(cmd._warnings) > 0)
+        self.assertTrue(len(cmd._warnings) > 0)
 
         # now let's add the required fields
         # and run it again, to make sure we don't get
@@ -79,7 +79,7 @@
                                  'restructuredtext': 1})
 
 def test_suite():
-    return unittest2.makeSuite(CheckTestCase)
+    return unittest.makeSuite(CheckTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_clean.py b/src/distutils2/tests/test_clean.py
--- a/src/distutils2/tests/test_clean.py
+++ b/src/distutils2/tests/test_clean.py
@@ -1,15 +1,15 @@
 """Tests for distutils.command.clean."""
 import sys
 import os
-import unittest2
 import getpass
 
 from distutils2.command.clean import clean
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 class cleanTestCase(support.TempdirManager,
                     support.LoggingSilencer,
-                    unittest2.TestCase):
+                    unittest.TestCase):
 
     def test_simple_run(self):
         pkg_dir, dist = self.create_dist()
@@ -44,7 +44,7 @@
         cmd.run()
 
 def test_suite():
-    return unittest2.makeSuite(cleanTestCase)
+    return unittest.makeSuite(cleanTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_cmd.py b/src/distutils2/tests/test_cmd.py
--- a/src/distutils2/tests/test_cmd.py
+++ b/src/distutils2/tests/test_cmd.py
@@ -1,17 +1,17 @@
 """Tests for distutils.cmd."""
-import unittest2
 import os
 from distutils2.tests import captured_stdout, run_unittest
 
-from distutils2.cmd import Command
+from distutils2.command.cmd import Command
 from distutils2.dist import Distribution
 from distutils2.errors import DistutilsOptionError
+from distutils2.tests.support import unittest
 
 class MyCmd(Command):
     def initialize_options(self):
         pass
 
-class CommandTestCase(unittest2.TestCase):
+class CommandTestCase(unittest.TestCase):
 
     def setUp(self):
         dist = Distribution()
@@ -104,7 +104,7 @@
         self.assertRaises(DistutilsOptionError, cmd.ensure_dirname, 'option2')
 
 def test_suite():
-    return unittest2.makeSuite(CommandTestCase)
+    return unittest.makeSuite(CommandTestCase)
 
 if __name__ == '__main__':
     run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_config.py b/src/distutils2/tests/test_config.py
--- a/src/distutils2/tests/test_config.py
+++ b/src/distutils2/tests/test_config.py
@@ -1,7 +1,6 @@
 """Tests for distutils.pypirc.pypirc."""
 import sys
 import os
-import unittest2
 import tempfile
 import shutil
 
@@ -11,6 +10,7 @@
 from distutils2.log import WARN
 
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 PYPIRC = """\
 [distutils]
@@ -50,7 +50,7 @@
 class PyPIRCCommandTestCase(support.TempdirManager,
                             support.LoggingSilencer,
                             support.EnvironGuard,
-                            unittest2.TestCase):
+                            unittest.TestCase):
 
     def setUp(self):
         """Patches the environment."""
@@ -112,7 +112,7 @@
         self.assertEquals(content, WANTED)
 
 def test_suite():
-    return unittest2.makeSuite(PyPIRCCommandTestCase)
+    return unittest.makeSuite(PyPIRCCommandTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_config_cmd.py b/src/distutils2/tests/test_config_cmd.py
--- a/src/distutils2/tests/test_config_cmd.py
+++ b/src/distutils2/tests/test_config_cmd.py
@@ -1,15 +1,15 @@
 """Tests for distutils.command.config."""
-import unittest2
 import os
 import sys
 
 from distutils2.command.config import dump_file, config
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 from distutils2 import log
 
 class ConfigTestCase(support.LoggingSilencer,
                      support.TempdirManager,
-                     unittest2.TestCase):
+                     unittest.TestCase):
 
     def _info(self, msg, *args):
         for line in msg.splitlines():
@@ -83,7 +83,7 @@
             self.assertTrue(not os.path.exists(f))
 
 def test_suite():
-    return unittest2.makeSuite(ConfigTestCase)
+    return unittest.makeSuite(ConfigTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_converter.py b/src/distutils2/tests/test_converter.py
--- a/src/distutils2/tests/test_converter.py
+++ b/src/distutils2/tests/test_converter.py
@@ -1,8 +1,8 @@
 """Tests for distutils.converter."""
 import os
 import sys
-import unittest2
 from distutils2.converter import DistutilsRefactoringTool
+from distutils2.tests.support import unittest
 
 _CURDIR = os.path.dirname(__file__)
 
@@ -15,9 +15,9 @@
         f.close()
 
 
-class ConverterTestCase(unittest2.TestCase):
+class ConverterTestCase(unittest.TestCase):
 
-    @unittest2.skipUnless(not sys.version < '2.6', 'Needs Python >=2.6')
+    @unittest.skipUnless(not sys.version < '2.6', 'Needs Python >=2.6')
     def test_conversions(self):
         # for all XX_before in the conversions/ dir
         # we run the refactoring tool
@@ -33,7 +33,7 @@
             self.assertEquals(str(res), wanted)
 
 def test_suite():
-    return unittest2.makeSuite(ConverterTestCase)
+    return unittest.makeSuite(ConverterTestCase)
 
 if __name__ == '__main__':
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_core.py b/src/distutils2/tests/test_core.py
--- a/src/distutils2/tests/test_core.py
+++ b/src/distutils2/tests/test_core.py
@@ -6,8 +6,8 @@
 import shutil
 import sys
 from distutils2.tests import captured_stdout
-import unittest2
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 # setup script that uses __file__
 setup_using___file__ = """\
@@ -28,7 +28,7 @@
 """
 
 
-class CoreTestCase(support.EnvironGuard, unittest2.TestCase):
+class CoreTestCase(support.EnvironGuard, unittest.TestCase):
 
     def setUp(self):
         super(CoreTestCase, self).setUp()
@@ -92,7 +92,7 @@
         self.assertEqual(cwd, output)
 
 def test_suite():
-    return unittest2.makeSuite(CoreTestCase)
+    return unittest.makeSuite(CoreTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_cygwinccompiler.py b/src/distutils2/tests/test_cygwinccompiler.py
--- a/src/distutils2/tests/test_cygwinccompiler.py
+++ b/src/distutils2/tests/test_cygwinccompiler.py
@@ -1,5 +1,4 @@
 """Tests for distutils.cygwinccompiler."""
-import unittest2
 import sys
 import os
 import warnings
@@ -18,9 +17,10 @@
                                        get_msvcr, RE_VERSION)
 from distutils2.util import get_compiler_versions
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 class CygwinCCompilerTestCase(support.TempdirManager,
-                              unittest2.TestCase):
+                              unittest.TestCase):
 
     def setUp(self):
         super(CygwinCCompilerTestCase, self).setUp()
@@ -94,7 +94,7 @@
 
 
 def test_suite():
-    return unittest2.makeSuite(CygwinCCompilerTestCase)
+    return unittest.makeSuite(CygwinCCompilerTestCase)
 
 if __name__ == '__main__':
     run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_dist.py b/src/distutils2/tests/test_dist.py
--- a/src/distutils2/tests/test_dist.py
+++ b/src/distutils2/tests/test_dist.py
@@ -4,15 +4,15 @@
 import os
 import StringIO
 import sys
-import unittest2
 import warnings
 import textwrap
 
 from distutils2.dist import Distribution, fix_help_options, DistributionMetadata
-from distutils2.cmd import Command
+from distutils2.command.cmd import Command
 import distutils2.dist
 from distutils2.tests import TESTFN, captured_stdout
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 class test_dist(Command):
     """Sample distutils2 extension command."""
@@ -40,7 +40,7 @@
 class DistributionTestCase(support.TempdirManager,
                            support.LoggingSilencer,
                            support.EnvironGuard,
-                           unittest2.TestCase):
+                           unittest.TestCase):
 
     def setUp(self):
         super(DistributionTestCase, self).setUp()
@@ -242,7 +242,7 @@
 
 
 class MetadataTestCase(support.TempdirManager, support.EnvironGuard,
-                       unittest2.TestCase):
+                       unittest.TestCase):
 
     def setUp(self):
         super(MetadataTestCase, self).setUp()
@@ -422,10 +422,10 @@
         self.assertEquals(metadata['requires-dist'], ['foo'])
 
 def test_suite():
-    suite = unittest2.TestSuite()
-    suite.addTest(unittest2.makeSuite(DistributionTestCase))
-    suite.addTest(unittest2.makeSuite(MetadataTestCase))
+    suite = unittest.TestSuite()
+    suite.addTest(unittest.makeSuite(DistributionTestCase))
+    suite.addTest(unittest.makeSuite(MetadataTestCase))
     return suite
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_emxccompiler.py b/src/distutils2/tests/test_emxccompiler.py
--- a/src/distutils2/tests/test_emxccompiler.py
+++ b/src/distutils2/tests/test_emxccompiler.py
@@ -1,5 +1,4 @@
 """Tests for distutils.emxccompiler."""
-import unittest2
 import sys
 import os
 import warnings
@@ -10,14 +9,15 @@
 from distutils2.compiler.emxccompiler import get_versions
 from distutils2.util import get_compiler_versions
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 class EmxCCompilerTestCase(support.TempdirManager,
-                           unittest2.TestCase):
+                           unittest.TestCase):
 
     pass
 
 def test_suite():
-    return unittest2.makeSuite(EmxCCompilerTestCase)
+    return unittest.makeSuite(EmxCCompilerTestCase)
 
 if __name__ == '__main__':
     run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_extension.py b/src/distutils2/tests/test_extension.py
--- a/src/distutils2/tests/test_extension.py
+++ b/src/distutils2/tests/test_extension.py
@@ -1,16 +1,16 @@
 """Tests for distutils.extension."""
-import unittest2
 import os
 import warnings
 
 from distutils2.extension import Extension
+from distutils2.tests.support import unittest
 
-class ExtensionTestCase(unittest2.TestCase):
+class ExtensionTestCase(unittest.TestCase):
 
     pass
 
 def test_suite():
-    return unittest2.makeSuite(ExtensionTestCase)
+    return unittest.makeSuite(ExtensionTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_install.py b/src/distutils2/tests/test_install.py
--- a/src/distutils2/tests/test_install.py
+++ b/src/distutils2/tests/test_install.py
@@ -3,7 +3,6 @@
 import os
 import os.path
 import sys
-import unittest2
 import site
 
 from distutils2._backport import sysconfig
@@ -22,11 +21,12 @@
 from distutils2.errors import DistutilsOptionError
 
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 class InstallTestCase(support.TempdirManager,
                       support.EnvironGuard,
                       support.LoggingSilencer,
-                      unittest2.TestCase):
+                      unittest.TestCase):
 
     def test_home_installation_scheme(self):
         # This ensure two things:
@@ -214,7 +214,7 @@
         self.assertTrue(len(self.logs) > old_logs_len)
 
 def test_suite():
-    return unittest2.makeSuite(InstallTestCase)
+    return unittest.makeSuite(InstallTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_install_data.py b/src/distutils2/tests/test_install_data.py
--- a/src/distutils2/tests/test_install_data.py
+++ b/src/distutils2/tests/test_install_data.py
@@ -1,16 +1,16 @@
 """Tests for distutils.command.install_data."""
 import sys
 import os
-import unittest2
 import getpass
 
 from distutils2.command.install_data import install_data
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 class InstallDataTestCase(support.TempdirManager,
                           support.LoggingSilencer,
                           support.EnvironGuard,
-                          unittest2.TestCase):
+                          unittest.TestCase):
 
     def test_simple_run(self):
         pkg_dir, dist = self.create_dist()
@@ -70,7 +70,7 @@
         self.assertTrue(os.path.exists(os.path.join(inst, rone)))
 
 def test_suite():
-    return unittest2.makeSuite(InstallDataTestCase)
+    return unittest.makeSuite(InstallDataTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_install_headers.py b/src/distutils2/tests/test_install_headers.py
--- a/src/distutils2/tests/test_install_headers.py
+++ b/src/distutils2/tests/test_install_headers.py
@@ -1,16 +1,16 @@
 """Tests for distutils.command.install_headers."""
 import sys
 import os
-import unittest2
 import getpass
 
 from distutils2.command.install_headers import install_headers
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 class InstallHeadersTestCase(support.TempdirManager,
                              support.LoggingSilencer,
                              support.EnvironGuard,
-                             unittest2.TestCase):
+                             unittest.TestCase):
 
     def test_simple_run(self):
         # we have two headers
@@ -34,7 +34,7 @@
         self.assertEquals(len(cmd.get_outputs()), 2)
 
 def test_suite():
-    return unittest2.makeSuite(InstallHeadersTestCase)
+    return unittest.makeSuite(InstallHeadersTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_install_lib.py b/src/distutils2/tests/test_install_lib.py
--- a/src/distutils2/tests/test_install_lib.py
+++ b/src/distutils2/tests/test_install_lib.py
@@ -1,12 +1,12 @@
 """Tests for distutils.command.install_data."""
 import sys
 import os
-import unittest2
 
 from distutils2.command.install_lib import install_lib
 from distutils2.extension import Extension
 from distutils2.tests import support
 from distutils2.errors import DistutilsOptionError
+from distutils2.tests.support import unittest
 
 try:
     no_bytecode = sys.dont_write_bytecode
@@ -18,7 +18,7 @@
 class InstallLibTestCase(support.TempdirManager,
                          support.LoggingSilencer,
                          support.EnvironGuard,
-                         unittest2.TestCase):
+                         unittest.TestCase):
 
     def test_finalize_options(self):
         pkg_dir, dist = self.create_dist()
@@ -38,7 +38,7 @@
         cmd.finalize_options()
         self.assertEquals(cmd.optimize, 2)
 
-    @unittest2.skipIf(no_bytecode, 'byte-compile not supported')
+    @unittest.skipIf(no_bytecode, 'byte-compile not supported')
     def test_byte_compile(self):
         pkg_dir, dist = self.create_dist()
         cmd = install_lib(dist)
@@ -84,7 +84,7 @@
         # get_input should return 2 elements
         self.assertEquals(len(cmd.get_inputs()), 2)
 
-    @unittest2.skipUnless(bytecode_support, 'sys.dont_write_bytecode not supported')
+    @unittest.skipUnless(bytecode_support, 'sys.dont_write_bytecode not supported')
     def test_dont_write_bytecode(self):
         # makes sure byte_compile is not used
         pkg_dir, dist = self.create_dist()
@@ -102,7 +102,7 @@
         self.assertTrue('byte-compiling is disabled' in self.logs[0][1])
 
 def test_suite():
-    return unittest2.makeSuite(InstallLibTestCase)
+    return unittest.makeSuite(InstallLibTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_install_scripts.py b/src/distutils2/tests/test_install_scripts.py
--- a/src/distutils2/tests/test_install_scripts.py
+++ b/src/distutils2/tests/test_install_scripts.py
@@ -1,17 +1,17 @@
 """Tests for distutils.command.install_scripts."""
 
 import os
-import unittest2
 
 from distutils2.command.install_scripts import install_scripts
 from distutils2.core import Distribution
 
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 
 class InstallScriptsTestCase(support.TempdirManager,
                              support.LoggingSilencer,
-                             unittest2.TestCase):
+                             unittest.TestCase):
 
     def test_default_settings(self):
         dist = Distribution()
@@ -73,7 +73,7 @@
 
 
 def test_suite():
-    return unittest2.makeSuite(InstallScriptsTestCase)
+    return unittest.makeSuite(InstallScriptsTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_manifest.py b/src/distutils2/tests/test_manifest.py
--- a/src/distutils2/tests/test_manifest.py
+++ b/src/distutils2/tests/test_manifest.py
@@ -1,10 +1,10 @@
 """Tests for distutils.manifest."""
-import unittest2
 import os
 import sys
 import logging
 
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 from distutils2.manifest import Manifest
 
 _MANIFEST = """\
@@ -18,7 +18,7 @@
 """
 
 class ManifestTestCase(support.TempdirManager,
-                       unittest2.TestCase):
+                       unittest.TestCase):
 
     def test_manifest_reader(self):
 
@@ -51,7 +51,7 @@
 
 
 def test_suite():
-    return unittest2.makeSuite(ManifestTestCase)
+    return unittest.makeSuite(ManifestTestCase)
 
 if __name__ == '__main__':
     run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_metadata.py b/src/distutils2/tests/test_metadata.py
--- a/src/distutils2/tests/test_metadata.py
+++ b/src/distutils2/tests/test_metadata.py
@@ -1,13 +1,13 @@
 """Tests for distutils.command.bdist."""
-import unittest2
 import os
 import sys
 from StringIO import StringIO
 
 from distutils2.metadata import (DistributionMetadata, _interpret,
                                  PKG_INFO_PREFERRED_VERSION)
+from distutils2.tests.support import unittest
 
-class DistributionMetadataTestCase(unittest2.TestCase):
+class DistributionMetadataTestCase(unittest.TestCase):
 
 
     def test_interpret(self):
@@ -64,7 +64,7 @@
         res = res.read()
         f = open(PKG_INFO)
         wanted = f.read()
-        self.assert_('Keywords: keyring,password,crypt' in res)
+        self.assertTrue('Keywords: keyring,password,crypt' in res)
         f.close()
 
     def test_metadata_markers(self):
@@ -228,7 +228,7 @@
 
 
 def test_suite():
-    return unittest2.makeSuite(DistributionMetadataTestCase)
+    return unittest.makeSuite(DistributionMetadataTestCase)
 
 if __name__ == '__main__':
     run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_msvc9compiler.py b/src/distutils2/tests/test_msvc9compiler.py
--- a/src/distutils2/tests/test_msvc9compiler.py
+++ b/src/distutils2/tests/test_msvc9compiler.py
@@ -1,10 +1,10 @@
 """Tests for distutils.msvc9compiler."""
 import sys
-import unittest2
 import os
 
 from distutils2.errors import DistutilsPlatformError
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 _MANIFEST = """\
 <?xml version="1.0" encoding="UTF-8" standalone="yes"?>
@@ -62,9 +62,9 @@
 
 
 class msvc9compilerTestCase(support.TempdirManager,
-                            unittest2.TestCase):
+                            unittest.TestCase):
 
-    @unittest2.skipUnless(sys.platform=="win32", "These tests are only for win32")
+    @unittest.skipUnless(sys.platform=="win32", "These tests are only for win32")
     def test_no_compiler(self):
         # makes sure query_vcvarsall throws
         # a DistutilsPlatformError if the compiler
@@ -86,7 +86,7 @@
         finally:
             msvc9compiler.find_vcvarsall = old_find_vcvarsall
 
-    @unittest2.skipUnless(sys.platform=="win32", "These tests are only for win32")
+    @unittest.skipUnless(sys.platform=="win32", "These tests are only for win32")
     def test_reg_class(self):
         from distutils2.msvccompiler import get_build_version
         if get_build_version() < 8.0:
@@ -110,7 +110,7 @@
         keys = Reg.read_keys(HKCU, r'Control Panel')
         self.assertTrue('Desktop' in keys)
 
-    @unittest2.skipUnless(sys.platform=="win32", "These tests are only for win32")
+    @unittest.skipUnless(sys.platform=="win32", "These tests are only for win32")
     def test_remove_visual_c_ref(self):
         from distutils2.msvc9compiler import MSVCCompiler
         tempdir = self.mkdtemp()
@@ -133,7 +133,7 @@
 
 
 def test_suite():
-    return unittest2.makeSuite(msvc9compilerTestCase)
+    return unittest.makeSuite(msvc9compilerTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_pypi_server.py b/src/distutils2/tests/test_pypi_server.py
--- a/src/distutils2/tests/test_pypi_server.py
+++ b/src/distutils2/tests/test_pypi_server.py
@@ -1,13 +1,13 @@
 """Tests for distutils.command.bdist."""
-import unittest2
 import urllib
 import urllib2
 import os.path
 
 from distutils2.tests.pypi_server import PyPIServer, PYPI_DEFAULT_STATIC_PATH
+from distutils2.tests.support import unittest
 
 
-class PyPIServerTest(unittest2.TestCase):
+class PyPIServerTest(unittest.TestCase):
 
     def test_records_requests(self):
         """We expect that PyPIServer can log our requests"""
@@ -64,7 +64,7 @@
 
 
 def test_suite():
-    return unittest2.makeSuite(PyPIServerTest)
+    return unittest.makeSuite(PyPIServerTest)
 
 if __name__ == '__main__':
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_pypi_versions.py b/src/distutils2/tests/test_pypi_versions.py
--- a/src/distutils2/tests/test_pypi_versions.py
+++ b/src/distutils2/tests/test_pypi_versions.py
@@ -17,9 +17,9 @@
 
 import xmlrpclib
 import os.path
-import unittest2
 
 from distutils2.version import suggest_normalized_version
+from distutils2.tests.support import unittest
 
 def test_pypi():
     #
@@ -121,11 +121,11 @@
     print "Have Suggestion : ", have_sugg, pct % (have_sugg/total_versions*100,)
     print "No Suggestion   : ", no_sugg, pct % (no_sugg/total_versions*100,)
 
-class TestPyPI(unittest2.TestCase):
+class TestPyPI(unittest.TestCase):
     pass
 
 def test_suite():
-    return unittest2.makeSuite(TestPyPI)
+    return unittest.makeSuite(TestPyPI)
 
 if __name__ == '__main__':
     run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_register.py b/src/distutils2/tests/test_register.py
--- a/src/distutils2/tests/test_register.py
+++ b/src/distutils2/tests/test_register.py
@@ -2,7 +2,6 @@
 # -*- encoding: utf8 -*-
 import sys
 import os
-import unittest2
 import getpass
 import urllib2
 import warnings
@@ -19,6 +18,7 @@
 from distutils2.errors import DistutilsSetupError
 
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 from distutils2.tests.test_config import PYPIRC, PyPIRCCommandTestCase
 
 PYPIRC_NOPASSWORD = """\
@@ -192,7 +192,7 @@
         self.assertEquals(headers['Content-length'], '290')
         self.assertTrue('tarek' in req.data)
 
-    @unittest2.skipUnless(DOCUTILS_SUPPORT, 'needs docutils')
+    @unittest.skipUnless(DOCUTILS_SUPPORT, 'needs docutils')
     def test_strict(self):
         # testing the script option
         # when on, the register command stops if
@@ -250,7 +250,7 @@
         self.assertEquals(data['requires_dist'], ['lxml'])
 
 def test_suite():
-    return unittest2.makeSuite(RegisterTestCase)
+    return unittest.makeSuite(RegisterTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_sdist.py b/src/distutils2/tests/test_sdist.py
--- a/src/distutils2/tests/test_sdist.py
+++ b/src/distutils2/tests/test_sdist.py
@@ -1,6 +1,5 @@
 """Tests for distutils.command.sdist."""
 import os
-import unittest2
 import shutil
 import zipfile
 import tarfile
@@ -29,6 +28,7 @@
 from distutils2.command.sdist import sdist
 from distutils2.command.sdist import show_formats
 from distutils2.core import Distribution
+from distutils2.tests.support import unittest
 from distutils2.tests.test_config import PyPIRCCommandTestCase
 from distutils2.errors import DistutilsExecError, DistutilsOptionError
 from distutils2.spawn import find_executable
@@ -97,7 +97,7 @@
         cmd.warn = _warn
         return dist, cmd
 
-    @unittest2.skipUnless(zlib, "requires zlib")
+    @unittest.skipUnless(zlib, "requires zlib")
     def test_prune_file_list(self):
         # this test creates a package with some vcs dirs in it
         # and launch sdist to make sure they get pruned
@@ -139,7 +139,7 @@
         # making sure everything has been pruned correctly
         self.assertEquals(len(content), 4)
 
-    @unittest2.skipUnless(zlib, "requires zlib")
+    @unittest.skipUnless(zlib, "requires zlib")
     def test_make_distribution(self):
 
         # check if tar and gzip are installed
@@ -176,7 +176,7 @@
         self.assertEquals(result,
                 ['fake-1.0.tar', 'fake-1.0.tar.gz'])
 
-    @unittest2.skipUnless(zlib, "requires zlib")
+    @unittest.skipUnless(zlib, "requires zlib")
     def test_add_defaults(self):
 
         # http://bugs.python.org/issue2279
@@ -238,7 +238,7 @@
         manifest = open(join(self.tmp_dir, 'MANIFEST')).read()
         self.assertEquals(manifest, MANIFEST % {'sep': os.sep})
 
-    @unittest2.skipUnless(zlib, "requires zlib")
+    @unittest.skipUnless(zlib, "requires zlib")
     def test_metadata_check_option(self):
         # testing the `medata-check` option
         dist, cmd = self.get_cmd(metadata={})
@@ -293,8 +293,8 @@
         cmd.formats = 'supazipa'
         self.assertRaises(DistutilsOptionError, cmd.finalize_options)
 
-    @unittest2.skipUnless(zlib, "requires zlib")
-    @unittest2.skipUnless(UID_GID_SUPPORT, "Requires grp and pwd support")
+    @unittest.skipUnless(zlib, "requires zlib")
+    @unittest.skipUnless(UID_GID_SUPPORT, "Requires grp and pwd support")
     def test_make_distribution_owner_group(self):
 
         # check if tar and gzip are installed
@@ -357,7 +357,7 @@
         self.assertIn('MANIFEST.in', cmd.filelist.files)
 
 def test_suite():
-    return unittest2.makeSuite(SDistTestCase)
+    return unittest.makeSuite(SDistTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_spawn.py b/src/distutils2/tests/test_spawn.py
--- a/src/distutils2/tests/test_spawn.py
+++ b/src/distutils2/tests/test_spawn.py
@@ -1,5 +1,4 @@
 """Tests for distutils.spawn."""
-import unittest2
 import os
 import time
 from distutils2.tests import captured_stdout
@@ -8,10 +7,11 @@
 from distutils2.spawn import spawn, find_executable
 from distutils2.errors import DistutilsExecError
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 class SpawnTestCase(support.TempdirManager,
                     support.LoggingSilencer,
-                    unittest2.TestCase):
+                    unittest.TestCase):
 
     def test_nt_quote_args(self):
 
@@ -23,7 +23,7 @@
             self.assertEquals(res, wanted)
 
 
-    @unittest2.skipUnless(os.name in ('nt', 'posix'),
+    @unittest.skipUnless(os.name in ('nt', 'posix'),
                          'Runs only under posix or nt')
     def test_spawn(self):
         tmpdir = self.mkdtemp()
@@ -54,7 +54,7 @@
         spawn([exe])  # should work without any error
 
 def test_suite():
-    return unittest2.makeSuite(SpawnTestCase)
+    return unittest.makeSuite(SpawnTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_unixccompiler.py b/src/distutils2/tests/test_unixccompiler.py
--- a/src/distutils2/tests/test_unixccompiler.py
+++ b/src/distutils2/tests/test_unixccompiler.py
@@ -1,6 +1,5 @@
 """Tests for distutils.unixccompiler."""
 import sys
-import unittest2
 
 try:
     import sysconfig
@@ -8,8 +7,9 @@
     from distutils2._backport import sysconfig
 
 from distutils2.compiler.unixccompiler import UnixCCompiler
+from distutils2.tests.support import unittest
 
-class UnixCCompilerTestCase(unittest2.TestCase):
+class UnixCCompilerTestCase(unittest.TestCase):
 
     def setUp(self):
         self._backup_platform = sys.platform
@@ -127,7 +127,7 @@
 
 
 def test_suite():
-    return unittest2.makeSuite(UnixCCompilerTestCase)
+    return unittest.makeSuite(UnixCCompilerTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_upload.py b/src/distutils2/tests/test_upload.py
--- a/src/distutils2/tests/test_upload.py
+++ b/src/distutils2/tests/test_upload.py
@@ -1,11 +1,13 @@
 """Tests for distutils.command.upload."""
 # -*- encoding: utf8 -*-
-import os, unittest2
+import os, sys
 
 from distutils2.command.upload import upload
 from distutils2.core import Distribution
 
 from distutils2.tests.pypi_server import PyPIServer, PyPIServerTestCase
+from distutils2.tests import support
+from distutils2.tests.support import unittest
 from distutils2.tests.test_config import PYPIRC, PyPIRCCommandTestCase
 
 
@@ -67,13 +69,13 @@
         headers = handler.headers.dict
         self.assertIn('d??d??', request_data)
         self.assertIn('xxx', request_data)
-        self.assert_(headers['content-length'] > 2000)
+        self.assertTrue(headers['content-length'] < 2000)
         self.assertTrue(headers['content-type'].startswith('multipart/form-data'))
         self.assertEqual(handler.command, 'POST')
         self.assertNotIn('\n', headers['authorization'])
 
 def test_suite():
-    return unittest2.makeSuite(UploadTestCase)
+    return unittest.makeSuite(uploadTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_upload_docs.py b/src/distutils2/tests/test_upload_docs.py
--- a/src/distutils2/tests/test_upload_docs.py
+++ b/src/distutils2/tests/test_upload_docs.py
@@ -1,6 +1,6 @@
 """Tests for distutils.command.upload_docs."""
 # -*- encoding: utf8 -*-
-import httplib, os, os.path, shutil, sys, tempfile, unittest2, zipfile
+import httplib, os, os.path, shutil, sys, tempfile, zipfile
 from cStringIO import StringIO
 
 from distutils2.command import upload_docs as upload_docs_mod
@@ -13,6 +13,7 @@
 from distutils2.tests import support
 from distutils2.tests.pypi_server import PyPIServer, PyPIServerTestCase
 from distutils2.tests.test_config import PyPIRCCommandTestCase
+from distutils2.tests.support import unittest
 
 
 EXPECTED_MULTIPART_OUTPUT = "\r\n".join([
@@ -191,7 +192,7 @@
         self.assertIn(self.pypi.default_response_data + "\n", write_args[0])
 
 def test_suite():
-    return unittest2.makeSuite(UploadDocsTestCase)
+    return unittest.makeSuite(UploadDocsTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_util.py b/src/distutils2/tests/test_util.py
--- a/src/distutils2/tests/test_util.py
+++ b/src/distutils2/tests/test_util.py
@@ -1,7 +1,6 @@
 """Tests for distutils.util."""
 import os
 import sys
-import unittest2
 from copy import copy
 from StringIO import StringIO
 import subprocess
@@ -17,6 +16,7 @@
                             byte_compile, find_packages)
 from distutils2 import util
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 class FakePopen(object):
     test_class = None
@@ -34,7 +34,7 @@
 
 class UtilTestCase(support.EnvironGuard,
                    support.TempdirManager,
-                   unittest2.TestCase):
+                   unittest.TestCase):
 
     def setUp(self):
         super(UtilTestCase, self).setUp()
@@ -243,7 +243,7 @@
         res = get_compiler_versions()
         self.assertEquals(res[2], None)
 
-    @unittest2.skipUnless(hasattr(sys, 'dont_write_bytecode'),
+    @unittest.skipUnless(hasattr(sys, 'dont_write_bytecode'),
                           'no dont_write_bytecode support')
     def test_dont_write_bytecode(self):
         # makes sure byte_compile raise a DistutilsError
@@ -298,7 +298,7 @@
 
 
 def test_suite():
-    return unittest2.makeSuite(UtilTestCase)
+    return unittest.makeSuite(UtilTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_version.py b/src/distutils2/tests/test_version.py
--- a/src/distutils2/tests/test_version.py
+++ b/src/distutils2/tests/test_version.py
@@ -1,5 +1,4 @@
 """Tests for distutils.version."""
-import unittest
 import doctest
 import os
 
@@ -7,6 +6,7 @@
 from distutils2.version import IrrationalVersionError
 from distutils2.version import suggest_normalized_version as suggest
 from distutils2.version import VersionPredicate
+from distutils2.tests.support import unittest
 
 class VersionTestCase(unittest.TestCase):
 
diff --git a/src/distutils2/version.py b/src/distutils2/version.py
--- a/src/distutils2/version.py
+++ b/src/distutils2/version.py
@@ -197,6 +197,9 @@
     def __ge__(self, other):
         return self.__eq__(other) or self.__gt__(other)
 
+    # See http://docs.python.org/reference/datamodel#object.__hash__
+    __hash__ = object.__hash__
+
 def suggest_normalized_version(s):
     """Suggest a normalized version close to the given version string.
 
diff --git a/src/runtests.py b/src/runtests.py
--- a/src/runtests.py
+++ b/src/runtests.py
@@ -26,9 +26,9 @@
 
 if __name__ == "__main__":
     try:
-        import unittest2
+        from distutils2.tests.support import unittest
     except ImportError:
-        print('!!! You need to install unittest2')
+        print('Error: You have to install unittest2')
         sys.exit(1)
 
     sys.exit(test_main())
diff --git a/src/setup.py b/src/setup.py
--- a/src/setup.py
+++ b/src/setup.py
@@ -17,19 +17,17 @@
     f.close()
 
 def get_tip_revision(path=os.getcwd()):
+    from subprocess import Popen, PIPE
     try:
-        from mercurial.hg import repository
-        from mercurial.ui import ui
-        from mercurial import node
-        from mercurial.error import RepoError
-    except ImportError:
+        cmd = Popen(['hg', 'tip', '--template', '{rev}', '-R', path],
+                    stdout=PIPE, stderr=PIPE)
+    except OSError:
         return 0
-    try:
-        repo = repository(ui(), path)
-        tip = repo.changelog.tip()
-        return repo.changelog.rev(tip)
-    except RepoError:
+    rev = cmd.stdout.read()
+    if rev == '':
+        # there has been an error in the command
         return 0
+    return int(rev)
 
 DEV_SUFFIX = '.dev%d' % get_tip_revision('..')
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: some more tests + first fonctionnal
	version
Message-ID: <E1OVLoV-0000jQ-RO@dinsdale.python.org>

tarek.ziade pushed bd3fbdf672db to distutils2:

http://hg.python.org/distutils2/rev/bd3fbdf672db
changeset:   309:bd3fbdf672db
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Fri Jun 18 02:07:55 2010 +0200
summary:     some more tests + first fonctionnal version
files:       src/distutils2/pypi/dist.py, src/distutils2/pypi/errors.py, src/distutils2/pypi/simple.py, src/distutils2/tests/test_pypi_dist.py, src/distutils2/version.py

diff --git a/src/distutils2/pypi/dist.py b/src/distutils2/pypi/dist.py
--- a/src/distutils2/pypi/dist.py
+++ b/src/distutils2/pypi/dist.py
@@ -6,6 +6,8 @@
 import re
 import urlparse
 
+from distutils2.version import suggest_normalized_version 
+
 EXTENSIONS = ".tar.gz .tar.bz2 .tar .zip .tgz .egg".split()
 MD5_HASH = re.compile(r'^.*#md5=([a-f0-9]+)$')
 
@@ -49,15 +51,22 @@
         # get the name from probable_dist_name
         if probable_dist_name is not None:
             if probable_dist_name in archive_name:
-                name = probable_dist_name 
+                name = probable_dist_name
 
         if name is None:
-            version = archive_name.split("-")[-1]
-            name = archive_name[:-len(version)+1]
+            # determine the name and the version
+            splits = archive_name.split("-")
+            version = splits[-1]
+            name = "-".join(splits[:-1])
+        else:
+            # just determine the version
+            version = archive_name[len(name):]
+            if version.startswith("-"):
+                version = version[1:]
         
+        version = suggest_normalized_version(version)
         if extension_matched is True:
             return PyPIDistribution(name, version, url=url, md5_hash=md5_hash)
-        raise Exception("test")
 
     def __init__(self, name, version, type=None, url=None, md5_hash=None):
         self.name = name
@@ -83,6 +92,10 @@
             real_path = "" 
         return real_path
 
+    def __str__(self):
+        """string representation of the PyPIDistribution"""
+        return "%s-%s" % (self.name, self.version)
+
 
 class PyPIDistributions(list):
     """A container of PyPIDistribution objects.
@@ -94,7 +107,6 @@
         """Filter the distributions and return just the one matching the given
         predicate.
         """
-        dists = self._distributions
-        return filter(predicate.match, 
-            [d.version for d in dists if d.name == predicate.name])
+        return [dist for dist in self if dist.name == predicate.name and
+            predicate.match(dist.version)]
 
diff --git a/src/distutils2/pypi/errors.py b/src/distutils2/pypi/errors.py
--- a/src/distutils2/pypi/errors.py
+++ b/src/distutils2/pypi/errors.py
@@ -6,5 +6,7 @@
 
 
 class PyPIError(DistutilsError):
-    """The base class for errors of the pypi python package.
-    """
+    """The base class for errors of the pypi python package."""
+
+class DistributionNotFound(PyPIError):
+    """Raised when no distribution match the given requirements."""
diff --git a/src/distutils2/pypi/simple.py b/src/distutils2/pypi/simple.py
--- a/src/distutils2/pypi/simple.py
+++ b/src/distutils2/pypi/simple.py
@@ -16,8 +16,9 @@
     from md5 import md5
 
 from distutils2.version import VersionPredicate
-from distutils2.pypi.dist import PyPIDistribution, EXTENSIONS
-from distutils2.pypi.errors import PyPIError
+from distutils2.pypi.dist import PyPIDistribution, PyPIDistributions, \
+    EXTENSIONS
+from distutils2.pypi.errors import PyPIError, DistributionNotFound
 from distutils2 import __version__ as __distutils2_version__
 
 # -- Constants -----------------------------------------------
@@ -54,21 +55,6 @@
     return _socket_timeout
 
 
-class UrlProcessor(object):
-    """Tools to process urls and return links from them.
-    """
-    def __init__(self):
-        self._processed_urls = {}
-
-    def find_links(self, url, is_browsable=lambda x:True):
-        """Find links and return them, for a specific url.
-        
-        Use "is_browsable" to provides a method to tell if the found urls might
-        be browsed or not.
-        """
-        pass
-
-
 class SimpleIndex(object):
     """Provides useful tools to request the Python Package Index simple API
     """
@@ -89,11 +75,11 @@
         # create a regexp to match all given hosts
         self._allowed_hosts = re.compile('|'.join(map(translate,hosts))).match
 
-        # _current_requirements are used to know what the index is currently
-        # browsing for. This could be used to determine the distributon name
-        # while having multiple ambigous names/version couples
-        self._current_requirements = None
-        self._processed_pages = []
+        # we keep an index of pages we have processed, in order to avoid
+        # scanning them multple time (eg. if there is multiple pages pointing
+        # on one)
+        self._processed_urls = []
+        self._distributions = {}
 
     def get_distributions(self, requirements):
         """Browse the PyPI to find distributions that fullfil the given 
@@ -103,13 +89,20 @@
         version specifiers, as described in PEP345. 
         """
         requirements = VersionPredicate(requirements)
-        self.current_requirements = requirements
         
         # process the index for this project
-        distributions = self._process_pypi_page(requirements.name)
+        self._process_pypi_page(requirements.name)
         
         # filter with requirements and return the results
-        return distributions.filter(requirements)
+        if self._distributions.has_key(requirements.name):
+            dists = self._distributions[requirements.name].filter(requirements)
+        else:
+            dists = []
+
+        if dists is not []:
+            return dists
+        else:
+            raise DistributionNotFound(requirements.name)
 
     def download(self, requirements, temp_path=None):
         """Download the distribution, using the requirements.
@@ -134,23 +127,54 @@
         """
         if self.follow_externals is True:
             return True
-        return True if self._allowed_hosts(urlparse(url).netloc) else False
+        return True if self._allowed_hosts(
+            urlparse.urlparse(url).netloc) else False
 
     def _is_distribution(self, link):
         """Tell if the given URL matches to a distribution name or not.
         """
+        for ext in EXTENSIONS:
+            if ext in link:
+                return True
+        return False
          
+    def _register_dist(self, dist):
+        """Register a distribution as a part of fetched distributions for
+        SimpleIndex.
 
-    def _process_url(self, url):
-        """Process an url and return fetched links.
+        Return the PyPIDistributions object for the specified project name
+        """
+        # Internally, check if a entry exists with the project name, if not,
+        # create a new one, and if exists, add the dist to the pool.
+        if not self._distributions.has_key(dist.name):
+            self._distributions[dist.name] = PyPIDistributions()
+        self._distributions[dist.name].append(dist)
+        return self._distributions[dist.name]
+
+    def _process_url(self, url, project_name=None, follow_links=True):
+        """Process an url and search for distributions packages.
+
+        :param url: the url to analyse
+        :param project_name: the project name we are searching for.
+        :param follow_links: We do not want to follow links more than from one
+        level. This parameter tells if we want to follow the links we find (eg.
+        run recursively this method on it)
         """
         f = self._open_url(url)
         base = f.url
+        self._processed_urls.append(url)
         for match in HREF.finditer(f.read()):
             link = urlparse.urljoin(base, self._htmldecode(match.group(1)))
-            if self._is_distribution(link):
-                PyPIDistribution.from_url(link)
-            else:
+            if link not in self._processed_urls:
+                if self._is_distribution(link):
+                    # it's a distribution, so create a dist object
+                    self._processed_urls.append(link)
+                    self._register_dist(PyPIDistribution.from_url(link, 
+                        project_name))
+                else:
+                    if self._is_browsable(link) and follow_links:
+                        self._process_url(link, project_name, 
+                            follow_links=False)
 
     def _process_pypi_page(self, name):
         """Find and process a PyPI page for the given project name.
@@ -158,15 +182,8 @@
         :param name: the name of the project to find the page
         """
         # Browse and index the content of the given PyPI page.
-        # Put all informations about the processed pages in the 
-        # _processed_pages attribute.
         url = self.index_url + name + "/"
-        found_links = self._process_url(url)
-        
-        # Search for external links here, and process them if needed.
-        for link in found_links:
-            if self._is_browsable(link):
-                self._search_in_url(link)
+        self._process_url(url, name)
     
     @socket_timeout()
     def _open_url(self, url):
@@ -186,7 +203,6 @@
             request.add_header("Authorization", auth)
         else:
             request = urllib2.Request(url)
-
         request.add_header('User-Agent', USER_AGENT)
         fp = urllib2.urlopen(request)
 
diff --git a/src/distutils2/tests/test_pypi_dist.py b/src/distutils2/tests/test_pypi_dist.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/test_pypi_dist.py
@@ -0,0 +1,68 @@
+"""Tests for the distutils2.pypi.dist module.
+"""
+
+from distutils2.pypi.dist import PyPIDistribution, PyPIDistributions
+from distutils2.tests.support import unittest
+from distutils2.version import VersionPredicate
+
+class TestPyPIDistribution(unittest.TestCase):
+    """tests the pypi.dist.PyPIDistribution class"""
+
+    def test_instanciation(self):
+        """Test the Distribution class provides us the good attributes when
+        given on construction"""
+        dist = PyPIDistribution("FooBar", "1.1")
+        self.assertEqual("FooBar", dist.name)
+        self.assertEqual("1.1", dist.version)
+
+    def test_from_url(self):
+        """Test that the Distribution object can be built from a single URL"""
+        url_list = {
+            'FooBar-1.1.0.tar.gz': {
+                'name': 'FooBar',
+                'version': '1.1.0',
+            },
+            'Foo-Bar-1.1.0.zip': {
+                'name': 'Foo-Bar',
+                'version': '1.1.0',
+            },
+            'foobar-1.1b2.tar.gz#md5=123123123123123': {
+                'name': 'foobar',
+                'version': '1.1b2',
+                'url':'http://test.tld/foobar-1.1b2.tar.gz', #without md5 hash
+                'md5_hash': '123123123123123',
+            }
+        }
+
+        for url, attributes in url_list.items():
+            dist = PyPIDistribution.from_url("http://test.tld/"+url)
+            for attribute, value in attributes.items():
+                self.assertEqual(getattr(dist, attribute), value) 
+
+class TestPyPIDistributions(unittest.TestCase):
+    """test the pypi.distr.PyPIDistributions class"""
+
+    def test_filter(self):
+        """Test we filter the distributions the right way, using version
+        predicate match method"""
+        dists = PyPIDistributions((
+            PyPIDistribution("FooBar", "1.1"),
+            PyPIDistribution("FooBar", "1.1.1"),
+            PyPIDistribution("FooBar", "1.2"),
+            PyPIDistribution("FooBar", "1.2.1"),
+        ))
+        filtered = dists.filter(VersionPredicate("FooBar (<1.2)"))
+        self.assertNotIn(dists[2], filtered)
+        self.assertNotIn(dists[3], filtered)
+        self.assertIn(dists[0], filtered)
+        self.assertIn(dists[1], filtered)
+
+
+def test_suite():
+    suite = unittest.TestSuite()
+    suite.addTest(unittest.makeSuite(TestPyPIDistribution))
+    suite.addTest(unittest.makeSuite(TestPyPIDistributions))
+    return suite
+
+if __name__ == '__main__':
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/version.py b/src/distutils2/version.py
--- a/src/distutils2/version.py
+++ b/src/distutils2/version.py
@@ -215,7 +215,7 @@
     on observation of versions currently in use on PyPI. Given a dump of
     those version during PyCon 2009, 4287 of them:
     - 2312 (53.93%) match NormalizedVersion without change
-    - with the automatic suggestion
+      with the automatic suggestion
     - 3474 (81.04%) match when using this suggestion method
 
     @param s {str} An irrational version string.

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: start documentation about
	distutils2.version
Message-ID: <E1OVLoV-0000jE-P7@dinsdale.python.org>

tarek.ziade pushed dd840eb55468 to distutils2:

http://hg.python.org/distutils2/rev/dd840eb55468
changeset:   306:dd840eb55468
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Tue Jun 08 21:51:36 2010 +0200
summary:     start documentation about distutils2.version
files:       docs/source/version.rst

diff --git a/docs/source/version.rst b/docs/source/version.rst
new file mode 100644
--- /dev/null
+++ b/docs/source/version.rst
@@ -0,0 +1,85 @@
+======================
+Working with versions
+======================
+
+Distutils2 ships with a python package capable to work with version numbers.
+It's an implementation of version specifiers `as defined in PEP 345
+<http://www.python.org/dev/peps/pep-0345/#version-specifiers>`_ about
+Metadata.
+
+`distutils2.version.NormalizedVersion`
+======================================
+
+A Normalized version corresponds to a specific version of a distribution, as
+described in the PEP 345. So, you can work with the `NormalizedVersion` like
+this::
+
+    >>> NormalizedVersion("1.2b1")
+    NormalizedVersion('1.2b1')
+
+If you try to use irrational version specifiers, an `IrrationalVersionError`
+will be raised::
+
+    >>> NormalizedVersion("irrational_version_number")
+    ...
+    IrrationalVersionError: irrational_version_number
+
+You can compare NormalizedVersion objects, like this::
+
+    >>> NormalizedVersion("1.2b1") < NormalizedVersion("1.2")
+    True
+
+NormalizedVersion is used internally by `Version`, `Versions` and
+`VersionPredicate` to do their stuff.
+
+`distutils2.version.suggest_normalized_version`
+-----------------------------------------------
+
+You also can let the normalized version be suggested to you, using the
+`suggest_normalized_version` function::
+
+    >>> suggest_normalized_version('2.1-rc1') 
+    2.1c1
+
+If `suggest_normalized_version` can't actually suggest you a version, it will
+return `None`::
+
+    >>> print suggest_normalized_version('not a version')
+    None
+
+`distutils2.version.VersionPredicate`
+=====================================
+
+`VersionPredicate` knows how to parse stuff like "ProjectName (>=version)", the
+class also provides a `match` method to test if a version number is the version
+predicate::
+
+    >>> version = VersionPredicate("ProjectName (<1.2,>1.0")
+    >>> version.match("1.2.1")
+    False
+    >>> version.match("1.1.1")
+    True
+
+`is_valid_predicate`
+--------------------
+
+
+`distutils2.version.Versions`
+=============================
+
+`is_valid_versions`
+--------------------
+
+
+`distutils2.version.Version`
+============================
+
+You can use the `Version` class to use version predicates without specifying a
+project name. It works like `Versions`, but only for one version. See an
+exemple::
+
+    >>> Version("1.1")
+    <distutils2.version.Version object at 0x...>
+
+`is_valid_version`
+--------------------

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: Added PyPIDistribution,
	started refactoring SimpleIndex with this. WIP
Message-ID: <E1OVLoV-0000jM-Qk@dinsdale.python.org>

tarek.ziade pushed fd3496bd8989 to distutils2:

http://hg.python.org/distutils2/rev/fd3496bd8989
changeset:   308:fd3496bd8989
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Tue Jun 15 19:39:32 2010 +0200
summary:     Added PyPIDistribution, started refactoring SimpleIndex with this. WIP
files:       docs/source/index.rst, src/distutils2/pypi/dist.py, src/distutils2/pypi/errors.py, src/distutils2/pypi/oldsimple.py, src/distutils2/pypi/simple.py, src/distutils2/tests/test_pypi_simple.py

diff --git a/docs/source/index.rst b/docs/source/index.rst
--- a/docs/source/index.rst
+++ b/docs/source/index.rst
@@ -14,6 +14,8 @@
    metadata
    new_commands
    test_framework
+   pypi
+   version
 
 Indices and tables
 ==================
diff --git a/src/distutils2/pypi/dist.py b/src/distutils2/pypi/dist.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/pypi/dist.py
@@ -0,0 +1,100 @@
+"""distutils2.pypi.dist
+
+Provides the PyPIDistribution class thats represents a distribution retreived 
+on PyPI.
+"""
+import re
+import urlparse
+
+EXTENSIONS = ".tar.gz .tar.bz2 .tar .zip .tgz .egg".split()
+MD5_HASH = re.compile(r'^.*#md5=([a-f0-9]+)$')
+
+class PyPIDistribution(object):
+    """Represents a distribution retreived from PyPI.
+
+    This is a simple container for various attributes as name, version, 
+    location, url etc.
+
+    The PyPIDistribution class is used by the pypi.*Index class to return
+    information about distributions. 
+    """
+
+    @classmethod
+    def from_url(cls, url, probable_dist_name=None):
+        """Build a Distribution from a url archive (egg or zip or tgz).
+
+        :param url: complete url of the distribution
+        :param probable_dist_name: the probable name of the distribution. This 
+        could be useful when multiple name can be assumed from the archive 
+        name.
+        """
+        # if the url contains a md5 hash, get it.
+        md5_hash = None
+        match = MD5_HASH.match(url)
+        if match is not None:
+            md5_hash = match.group(1)
+            # remove the hash
+            url = url.replace("#md5=%s" % md5_hash, "")
+        
+        # parse the archive name to find dist name and version
+        archive_name = urlparse.urlparse(url).path.split('/')[-1]
+        extension_matched = False
+        # remove the extension from the name
+        for ext in EXTENSIONS:
+            if archive_name.endswith(ext):
+                archive_name = archive_name[:-len(ext)]
+                extension_matched = True
+
+        name = None
+        # get the name from probable_dist_name
+        if probable_dist_name is not None:
+            if probable_dist_name in archive_name:
+                name = probable_dist_name 
+
+        if name is None:
+            version = archive_name.split("-")[-1]
+            name = archive_name[:-len(version)+1]
+        
+        if extension_matched is True:
+            return PyPIDistribution(name, version, url=url, md5_hash=md5_hash)
+        raise Exception("test")
+
+    def __init__(self, name, version, type=None, url=None, md5_hash=None):
+        self.name = name
+        self.version = version
+        self.url = url
+        self.type = type
+        self.md5_hash = md5_hash
+        # set the downloaded path to Null by default. The goal here
+        # is to not download distributions multiple times 
+        self._downloaded_path = None
+
+    def download(self, url=None, path=None):
+        """Download the distribution to a path, and return it.
+
+        If the path is given in path, use this, otherwise, generates a new one
+        """
+        if url is not None:
+            self.url = url
+
+        # if we do not have downloaded it yet, do it.
+        if self._downloaded_path is None:
+            # download logic goes here
+            real_path = "" 
+        return real_path
+
+
+class PyPIDistributions(list):
+    """A container of PyPIDistribution objects.
+    
+    Contains methods and facilities to sort and filter distributions.
+    """
+
+    def filter(self, predicate):
+        """Filter the distributions and return just the one matching the given
+        predicate.
+        """
+        dists = self._distributions
+        return filter(predicate.match, 
+            [d.version for d in dists if d.name == predicate.name])
+
diff --git a/src/distutils2/pypi/errors.py b/src/distutils2/pypi/errors.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/pypi/errors.py
@@ -0,0 +1,10 @@
+"""distutils2.pypi.errors
+
+All errors and exceptions raised by PyPiIndex classes.
+"""
+from distutils2.errors import DistutilsError
+
+
+class PyPIError(DistutilsError):
+    """The base class for errors of the pypi python package.
+    """
diff --git a/src/distutils2/pypi/oldsimple.py b/src/distutils2/pypi/oldsimple.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/pypi/oldsimple.py
@@ -0,0 +1,806 @@
+"""distutils2.pypi.package_index
+
+Implement a simple way to use the PyPI API's, by using known URIs from 
+PyPI. Works with the API defined on 
+http://peak.telecommunity.com/DevCenter/EasyInstall#package-index-api , 
+and available at http://pypi.python.org/simple/ .
+
+"""
+# XXX Make a explaination document on how this API is made.
+
+import sys
+import os.path
+import re
+import urlparse
+import urllib2
+import shutil
+import socket
+import cStringIO
+import httplib
+import logging
+import tempfile
+
+from distutils2.errors import DistutilsError
+from distutils2 import __version__ as __distutils2_version__
+
+try:
+    from hashlib import md5
+except ImportError:
+    from md5 import md5
+from fnmatch import translate
+
+# -- Constants ----------------------------------------------------------------
+
+__all__ = ['PackageIndex', 
+           'distros_for_url', 
+           'parse_bdist_wininst',
+           'interpret_distro_name',
+]
+
+_SOCKET_TIMEOUT = 15
+SOURCE_DIST = 1
+EGG_FRAGMENT = re.compile(r'^egg=([-A-Za-z0-9_.]+)$')
+HREF = re.compile("""href\\s*=\\s*['"]?([^'"> ]+)""", re.I)
+PYPI_MD5 = re.compile(
+    '<a href="([^"#]+)">([^<]+)</a>\n\s+\\(<a (?:title="MD5 hash"\n\s+)'
+    'href="[^?]+\?:action=show_md5&amp;digest=([0-9a-f]{32})">md5</a>\\)'
+)
+URL_SCHEME = re.compile('([-+.a-z0-9]{2,}):',re.I).match
+EXTENSIONS = ".tar.gz .tar.bz2 .tar .zip .tgz".split()
+
+# -- Dependencies from distribute ---------------------------------------------
+
+# XXX Remove these dependencies, refactor the package_index codebase to not 
+# depends anymore on distribute classes and internal workings.
+from pkg_resources import *
+
+# -- Base functions -----------------------------------------------------------
+
+def parse_bdist_wininst(name):
+    """Parse a bdist_win package name, and return it's package base name 
+    and version number if possible.
+
+    Return a tuple of (base,pyversion) for possible .exe names"""
+
+    lower = name.lower()
+    base, py_ver = None, None
+
+    if lower.endswith('.exe'):
+        if lower.endswith('.win32.exe'):
+            base = name[:-10]
+        elif lower.startswith('.win32-py',-16):
+            py_ver = name[-7:-4]
+            base = name[:-16]
+
+    return base,py_ver
+
+def egg_info_for_url(url):
+    """
+    """
+    scheme, server, path, parameters, query, fragment = urlparse.urlparse(url)
+    base = urllib2.unquote(path.split('/')[-1])
+    if '#' in base:
+        base, fragment = base.split('#',1)
+    return base,fragment
+
+def distros_for_url(url, metadata=None):
+    """Yield egg or source distribution objects that might be found at a URL"""
+    base, fragment = egg_info_for_url(url)
+    for dist in distros_for_location(url, base, metadata): 
+        yield dist
+    if fragment:
+        match = EGG_FRAGMENT.match(fragment)
+        if match:
+            for dist in interpret_distro_name(url, match.group(1), metadata, 
+                precedence = CHECKOUT_DIST):
+                yield dist
+
+def distros_for_location(location, basename, metadata=None):
+    """Yield egg or source distribution objects based on basename"""
+    if basename.endswith('.egg.zip'):
+        basename = basename[:-4]    # strip the .zip
+    if basename.endswith('.egg') and '-' in basename:
+        # only one, unambiguous interpretation
+        return [Distribution.from_location(location, basename, metadata)]
+
+    if basename.endswith('.exe'):
+        win_base, py_ver = parse_bdist_wininst(basename)
+        if win_base is not None:
+            return interpret_distro_name(location, win_base, metadata, py_ver, 
+                BINARY_DIST, "win32")
+
+    # Try source distro extensions (.zip, .tgz, etc.)
+    for ext in EXTENSIONS:
+        if basename.endswith(ext):
+            basename = basename[:-len(ext)]
+            return interpret_distro_name(location, basename, metadata)
+    return []  # no extension matched
+
+def distros_for_filename(filename, metadata=None):
+    """Yield possible egg or source distribution objects based on a 
+    filename.
+    """
+    return distros_for_location(normalize_path(filename), 
+        os.path.basename(filename), metadata)
+
+def interpret_distro_name(location, basename, metadata,
+    py_version=None, precedence=SOURCE_DIST, platform=None):
+    """Generate alternative interpretations of a source distro name
+
+    Note: if `location` is a filesystem filename, you should call
+    ``pkg_resources.normalize_path()`` on it before passing it to this
+    routine!
+    """
+    # Generate alternative interpretations of a source distro name
+    # Because some packages are ambiguous as to name/versions split
+    # e.g. "adns-python-1.1.0", "egenix-mx-commercial", etc.
+    # So, we generate each possible interepretation (e.g. "adns, python-1.1.0"
+    # "adns-python, 1.1.0", and "adns-python-1.1.0, no version").  In practice,
+    # the spurious interpretations should be ignored, because in the event
+    # there's also an "adns" package, the spurious "python-1.1.0" version will
+    # compare lower than any numeric version number, and is therefore unlikely
+    # to match a request for it.  It's still a potential problem, though, and
+    # in the long run PyPI and the distutils should go for "safe" names and
+    # versions in distribution archive names (sdist and bdist).
+
+    parts = basename.split('-')
+    if not py_version:
+        for i,p in enumerate(parts[2:]):
+            if len(p)==5 and p.startswith('py2.'):
+                return # It's a bdist_dumb, not an sdist -- bail out
+
+    for p in range(1,len(parts)+1):
+        yield Distribution(
+            location, metadata, '-'.join(parts[:p]), '-'.join(parts[p:]),
+            py_version=py_version, precedence = precedence,
+            platform = platform)
+
+REL = re.compile("""<([^>]*\srel\s*=\s*['"]?([^'">]+)[^>]*)>""", re.I)
+
+def find_external_links(url, page):
+    """Find rel="homepage" and rel="download" links in `page`, yielding URLs"""
+
+    for match in REL.finditer(page):
+        tag, rel = match.groups()
+        rels = map(str.strip, rel.lower().split(','))
+        if 'homepage' in rels or 'download' in rels:
+            for match in HREF.finditer(tag):
+                yield urlparse.urljoin(url, htmldecode(match.group(1)))
+
+    for tag in ("<th>Home Page", "<th>Download URL"):
+        pos = page.find(tag)
+        if pos!=-1:
+            match = HREF.search(page,pos)
+            if match:
+                yield urlparse.urljoin(url, htmldecode(match.group(1)))
+
+user_agent = "Python-urllib/%s distutils2/%s" % (
+    sys.version[:3], __distutils2_version__
+)
+
+class SimpleIndex(Environment):
+    """A distribution index that scans web pages for download URLs"""
+
+    def __init__(self, index_url="http://pypi.python.org/simple", hosts=('*',),
+        follow_externals=True, *args, **kw):
+        self._init_logging()
+        Environment.__init__(self,*args,**kw)
+        self.index_url = index_url + "/"[:not index_url.endswith('/')]
+        self.scanned_urls = {}
+        self.fetched_urls = {}
+        self.package_pages = {}
+        self.allows = re.compile('|'.join(map(translate,hosts))).match
+        self.to_scan = []
+        self.follow_externals = follow_externals
+
+    def _init_logging(self):
+        import sys
+#        logging.basicConfig(stream=sys.stdout, level=logging.DEBUG)
+
+    def process_url(self, url, retrieve=False):
+        """Evaluate a URL as a possible download, and maybe retrieve it"""
+
+        if url in self.scanned_urls and not retrieve:
+            return
+        self.scanned_urls[url] = True
+        if not URL_SCHEME(url):
+            self.process_filename(url)
+            return
+        else:
+            dists = list(distros_for_url(url))
+            if dists:
+                if not self.url_ok(url):
+                    return
+                self.debug("Found link: %s", url)
+
+        if dists or not retrieve or url in self.fetched_urls:
+            map(self.add, dists)
+            return  # don't need the actual page
+
+        if not self.url_ok(url):
+            self.fetched_urls[url] = True
+            return
+
+        self.info("Reading %s", url)
+        f = self.open_url(url, "Download error: %s -- Some packages may not be found!")
+        if f is None: return
+        self.fetched_urls[url] = self.fetched_urls[f.url] = True
+
+        if 'html' not in f.headers.get('content-type', '').lower():
+            f.close()   # not html, we can't process it
+            return
+
+        base = f.url     # handle redirects
+        page = f.read()
+        if sys.version_info >= (3,):
+            charset = f.headers.get_param('charset') or 'latin-1'
+            page = page.decode(charset, "ignore")
+        f.close()
+
+        if url.startswith(self.index_url):
+            if self.follow_externals:
+                for match in HREF.finditer(page):
+                    link = urlparse.urljoin(base, htmldecode(match.group(1)))
+                    self.process_url(link)
+        if url.startswith(self.index_url) and getattr(f,'code',None)!=404:
+            page = self.process_index(url, page)
+
+    def process_filename(self, fn, nested=False):
+        # process filenames or directories
+        if not os.path.exists(fn):
+            self.warn("Not found: %s", fn)
+            return
+
+        if os.path.isdir(fn) and not nested:
+            path = os.path.realpath(fn)
+            for item in os.listdir(path):
+                self.process_filename(os.path.join(path,item), True)
+
+        dists = distros_for_filename(fn)
+        if dists:
+            self.debug("Found: %s", fn)
+            map(self.add, dists)
+
+    def url_ok(self, url, fatal=False):
+        s = URL_SCHEME(url)
+        if (s and s.group(1).lower()=='file') or self.allows(urlparse.urlparse(url)[1]):
+            return True
+        msg = "\nLink to % s ***BLOCKED*** by --allow-hosts\n"
+        if fatal:
+            raise DistutilsError(msg % url)
+        else:
+            self.warn(msg, url)
+        return False
+
+    def scan_egg_links(self, search_path):
+        for item in search_path:
+            if os.path.isdir(item):
+                for entry in os.listdir(item):
+                    if entry.endswith('.egg-link'):
+                        self.scan_egg_link(item, entry)
+
+    def scan_egg_link(self, path, entry):
+        lines = filter(None, map(str.strip, open(os.path.join(path, entry))))
+        if len(lines)==2:
+            for dist in find_distributions(os.path.join(path, lines[0])):
+                dist.location = os.path.join(path, *lines)
+                dist.precedence = SOURCE_DIST
+                self.add(dist)
+
+    def process_index(self,url,page):
+        """Process the content of PyPI page. For each link, if it's a PyPI one,
+        browse it's page, and satinize the output.
+        """
+        def scan(link):
+            # Process a URL to see if it's for a package page
+            if link.startswith(self.index_url):
+                parts = map(
+                    urllib2.unquote, link[len(self.index_url):].split('/')
+                )
+                if len(parts)==2 and '#' not in parts[1]:
+                    # it's a package page, sanitize and index it
+                    pkg = safe_name(parts[0])
+                    ver = safe_version(parts[1])
+                    self.package_pages.setdefault(pkg.lower(),{})[link] = True
+                    return to_filename(pkg), to_filename(ver)
+            return None, None
+
+        # process an index page into the package-page index
+        for match in HREF.finditer(page):
+            try:
+                scan( urlparse.urljoin(url, htmldecode(match.group(1))) )
+            except ValueError:
+                pass
+
+        pkg, ver = scan(url)   # ensure this page is in the page index
+        if pkg:
+            # process individual package page
+            for new_url in find_external_links(url, page):
+                # Process the found URL
+                base, frag = egg_info_for_url(new_url)
+                if base.endswith('.py') and not frag:
+                    if ver:
+                        new_url+='#egg=%s-%s' % (pkg,ver)
+                    else:
+                        self.need_version_info(url)
+                self.scan_url(new_url)
+
+            return PYPI_MD5.sub(
+                lambda m: '<a href="%s#md5=%s">%s</a>' % m.group(1,3,2), page
+            )
+        else:
+            return ""   # no sense double-scanning non-package pages
+
+    def need_version_info(self, url):
+        self.scan_all(
+            "Page at %s links to .py file(s) without version info; an index "
+            "scan is required.", url
+        )
+
+    def scan_all(self, msg=None, *args):
+        if self.index_url not in self.fetched_urls:
+            if msg: self.warn(msg,*args)
+            self.info(
+                "Scanning index of all packages (this may take a while)"
+            )
+        self.scan_url(self.index_url)
+
+    def find_packages(self, requirement):
+        self.scan_url(self.index_url + requirement.unsafe_name+'/')
+
+        if not self.package_pages.get(requirement.key):
+            # Fall back to safe version of the name
+            self.scan_url(self.index_url + requirement.project_name+'/')
+
+        if not self.package_pages.get(requirement.key):
+            # We couldn't find the target package, so search the index page too
+            self.not_found_in_index(requirement)
+
+        for url in list(self.package_pages.get(requirement.key,())):
+            # scan each page that might be related to the desired package
+            self.scan_url(url)
+
+    def obtain(self, requirement, installer=None):
+        self.prescan(); self.find_packages(requirement)
+        for dist in self[requirement.key]:
+            if dist in requirement:
+                return dist
+            self.debug("%s does not match %s", requirement, dist)
+        return super(SimpleIndex, self).obtain(requirement,installer)
+
+    def check_md5(self, cs, info, filename, tfp):
+        if re.match('md5=[0-9a-f]{32}$', info):
+            self.debug("Validating md5 checksum for %s", filename)
+            if cs.hexdigest()<>info[4:]:
+                tfp.close()
+                os.unlink(filename)
+                raise DistutilsError(
+                    "MD5 validation failed for "+os.path.basename(filename)+
+                    "; possible download problem?")
+
+    def add_find_links(self, urls):
+        """Add `urls` to the list that will be prescanned for searches"""
+        for url in urls:
+            if (
+                self.to_scan is None        # if we have already "gone online"
+                or not URL_SCHEME(url)      # or it's a local file/directory
+                or url.startswith('file:')
+                or list(distros_for_url(url))   # or a direct package link
+            ):
+                # then go ahead and process it now
+                self.scan_url(url)
+            else:
+                # otherwise, defer retrieval till later
+                self.to_scan.append(url)
+
+    def prescan(self):
+        """Scan urls scheduled for prescanning (e.g. --find-links)"""
+        if self.to_scan:
+            map(self.scan_url, self.to_scan)
+        self.to_scan = None     # from now on, go ahead and process immediately
+
+    def not_found_in_index(self, requirement):
+        if self[requirement.key]:   # we've seen at least one distro
+            meth, msg = self.info, "Couldn't retrieve index page for %r"
+        else:   # no distros seen for this name, might be misspelled
+            meth, msg = (self.warn,
+                "Couldn't find index page for %r (maybe misspelled?)")
+        meth(msg, requirement.unsafe_name)
+        self.scan_all()
+
+    def download(self, spec, tmpdir=None):
+        """Locate and/or download `spec` to `tmpdir`, returning a local path
+
+        `spec` may be a ``Requirement`` object, or a string containing a URL,
+        an existing local filename, or a project/version requirement spec
+        (i.e. the string form of a ``Requirement`` object).  If it is the URL
+        of a .py file with an unambiguous ``#egg=name-version`` tag (i.e., one
+        that escapes ``-`` as ``_`` throughout), a trivial ``setup.py`` is
+        automatically created alongside the downloaded file.
+
+        If `spec` is a ``Requirement`` object or a string containing a
+        project/version requirement spec, this method returns the location of
+        a matching distribution (possibly after downloading it to `tmpdir`).
+        If `spec` is a locally existing file or directory name, it is simply
+        returned unchanged.  If `spec` is a URL, it is downloaded to a subpath
+        of `tmpdir`, and the local filename is returned.  Various errors may be
+        raised if a problem occurs during downloading.
+        """
+        if tmpdir is None:
+            tmpdir=tempfile.mkdtemp()
+        if not isinstance(spec,Requirement):
+            scheme = URL_SCHEME(spec)
+            if scheme:
+                # It's a url, download it to tmpdir
+                found = self._download_url(scheme.group(1), spec, tmpdir)
+                base, fragment = egg_info_for_url(spec)
+                if base.endswith('.py'):
+                    found = self.gen_setup(found,fragment,tmpdir)
+                return found
+            elif os.path.exists(spec):
+                # Existing file or directory, just return it
+                return spec
+            else:
+                try:
+                    spec = Requirement.parse(spec)
+                except ValueError:
+                    raise DistutilsError(
+                        "Not a URL, existing file, or requirement spec: %r" %
+                        (spec,)
+                    )
+        return getattr(self.fetch_distribution(spec, tmpdir),'location',None)
+
+    def fetch_distribution(self,
+        requirement, tmpdir, force_scan=False, source=False, develop_ok=False
+    ):
+        """Obtain a distribution suitable for fulfilling `requirement`
+
+        `requirement` must be a ``pkg_resources.Requirement`` instance.
+        If necessary, or if the `force_scan` flag is set, the requirement is
+        searched for in the (online) package index as well as the locally
+        installed packages.  If a distribution matching `requirement` is found,
+        the returned distribution's ``location`` is the value you would have
+        gotten from calling the ``download()`` method with the matching
+        distribution's URL or filename.  If no matching distribution is found,
+        ``None`` is returned.
+
+        If the `source` flag is set, only source distributions and source
+        checkout links will be considered.  Unless the `develop_ok` flag is
+        set, development and system eggs (i.e., those using the ``.egg-info``
+        format) will be ignored.
+        """
+        
+        # process a Requirement
+        self.info("Searching for %s", requirement)
+        skipped = {}
+
+        def find(req):
+            # Find a matching distribution; may be called more than once
+
+            for dist in self[req.key]:
+
+                if dist.precedence==DEVELOP_DIST and not develop_ok:
+                    if dist not in skipped:
+                        self.warn("Skipping development or system egg: %s",dist)
+                        skipped[dist] = 1
+                    continue
+
+                if dist in req and (dist.precedence<=SOURCE_DIST or not source):
+                    self.info("Best match: %s", dist)
+                    return dist.clone(
+                        location=self.download(dist.location, tmpdir)
+                    )
+
+        if force_scan:
+            self.prescan()
+            self.find_packages(requirement)
+
+        dist = find(requirement)
+        if dist is None and self.to_scan is not None:
+            self.prescan()
+            dist = find(requirement)
+
+        if dist is None and not force_scan:
+            self.find_packages(requirement)
+            dist = find(requirement)
+
+        if dist is None:
+            self.warn(
+                "No local packages or download links found for %s%s",
+                (source and "a source distribution of " or ""),
+                requirement,
+            )
+        return dist
+
+    def fetch(self, requirement, tmpdir, force_scan=False, source=False):
+        """Obtain a file suitable for fulfilling `requirement`
+
+        DEPRECATED; use the ``fetch_distribution()`` method now instead.  For
+        backward compatibility, this routine is identical but returns the
+        ``location`` of the downloaded distribution instead of a distribution
+        object.
+        """
+        dist = self.fetch_distribution(requirement,tmpdir,force_scan,source)
+        if dist is not None:
+            return dist.location
+        return None
+    
+    def gen_setup(self, filename, fragment, tmpdir):
+        match = EGG_FRAGMENT.match(fragment)
+        dists = match and [d for d in
+            interpret_distro_name(filename, match.group(1), None) if d.version
+        ] or []
+
+        if len(dists)==1:   # unambiguous ``#egg`` fragment
+            basename = os.path.basename(filename)
+
+            # Make sure the file has been downloaded to the temp dir.
+            if os.path.dirname(filename) != tmpdir:
+                dst = os.path.join(tmpdir, basename)
+                from setuptools.command.easy_install import samefile
+                if not samefile(filename, dst):
+                    shutil.copy2(filename, dst)
+                    filename=dst
+
+            file = open(os.path.join(tmpdir, 'setup.py'), 'w')
+            file.write(
+                "from setuptools import setup\n"
+                "setup(name=%r, version=%r, py_modules=[%r])\n"
+                % (
+                    dists[0].project_name, dists[0].version,
+                    os.path.splitext(basename)[0]
+                )
+            )
+            file.close()
+            return filename
+
+        elif match:
+            raise DistutilsError(
+                "Can't unambiguously interpret project/version identifier %r; "
+                "any dashes in the name or version should be escaped using "
+                "underscores. %r" % (fragment,dists)
+            )
+        else:
+            raise DistutilsError(
+                "Can't process plain .py without an '#egg=name-version'"
+                " suffix to enable automatic setup script generation."
+            )
+
+    dl_blocksize = 8192
+    def _download_to(self, url, filename):
+        self.info("Downloading %s", url)
+        # Download the file
+        fp, tfp, info = None, None, None
+        try:
+            if '#' in url:
+                url, info = url.split('#', 1)
+            fp = self.open_url(url)
+            if isinstance(fp, urllib2.HTTPError):
+                raise DistutilsError(
+                    "Can't download %s: %s %s" % (url, fp.code,fp.msg)
+                )
+            cs = md5()
+            headers = fp.info()
+            blocknum = 0
+            bs = self.dl_blocksize
+            size = -1
+            if "content-length" in headers:
+                size = int(headers["Content-Length"])
+                self.reporthook(url, filename, blocknum, bs, size)
+            tfp = open(filename,'wb')
+            while True:
+                block = fp.read(bs)
+                if block:
+                    cs.update(block)
+                    tfp.write(block)
+                    blocknum += 1
+                    self.reporthook(url, filename, blocknum, bs, size)
+                else:
+                    break
+            if info: self.check_md5(cs, info, filename, tfp)
+            return headers
+        finally:
+            if fp: fp.close()
+            if tfp: tfp.close()
+
+    def reporthook(self, url, filename, blocknum, blksize, size):
+        pass    # no-op
+
+    def open_url(self, url, warning=None):
+        if url.startswith('file:'):
+            return local_open(url)
+        try:
+            return open_with_auth(url)
+        except (ValueError, httplib.InvalidURL), v:
+            msg = ' '.join([str(arg) for arg in v.args])
+            if warning:
+                self.warn(warning, msg)
+            else:
+                raise DistutilsError('%s %s' % (url, msg))
+        except urllib2.HTTPError, v:
+            return v
+        except urllib2.URLError, v:
+            if warning:
+                self.warn(warning, v.reason)
+            else:
+                raise DistutilsError("Download error for %s: %s"
+                                     % (url, v.reason))
+        except httplib.BadStatusLine, v:
+            if warning:
+                self.warn(warning, v.line)
+            else:
+                raise DistutilsError('%s returned a bad status line. '
+                                     'The server might be down, %s' % \
+                                             (url, v.line))
+        except httplib.HTTPException, v:
+            if warning:
+                self.warn(warning, v)
+            else:
+                raise DistutilsError("Download error for %s: %s"
+                                     % (url, v))
+
+    def _download_url(self, scheme, url, tmpdir):
+        # Determine download filename
+        name = filter(None,urlparse.urlparse(url)[2].split('/'))
+        if name:
+            name = name[-1]
+            while '..' in name:
+                name = name.replace('..','.').replace('\\','_')
+        else:
+            name = "__downloaded__"    # default if URL has no path contents
+
+        if name.endswith('.egg.zip'):
+            name = name[:-4]    # strip the extra .zip before download
+
+        filename = os.path.join(tmpdir,name)
+
+        # Download the file
+        if scheme=='svn' or scheme.startswith('svn+'):
+            return self._download_svn(url, filename)
+        elif scheme=='file':
+            return urllib2.url2pathname(urlparse.urlparse(url)[2])
+        else:
+            self.url_ok(url, True)   # raises error if not allowed
+            return self._attempt_download(url, filename)
+
+    def scan_url(self, url):
+        self.process_url(url, True)
+
+    def _attempt_download(self, url, filename):
+        headers = self._download_to(url, filename)
+        if 'html' in headers.get('content-type','').lower():
+            return self._download_html(url, headers, filename)
+        else:
+            return filename
+
+    def _download_html(self, url, headers, filename):
+        file = open(filename)
+        for line in file:
+            if line.strip():
+                # Check for a subversion index page
+                if re.search(r'<title>([^- ]+ - )?Revision \d+:', line):
+                    # it's a subversion index page:
+                    file.close()
+                    os.unlink(filename)
+                    return self._download_svn(url, filename)
+                break   # not an index page
+        file.close()
+        os.unlink(filename)
+        raise DistutilsError("Unexpected HTML page found at "+url)
+
+    def _download_svn(self, url, filename):
+        url = url.split('#',1)[0]   # remove any fragment for svn's sake
+        self.info("Doing subversion checkout from %s to %s", url, filename)
+        os.system("svn checkout -q %s %s" % (url, filename))
+        return filename
+
+    def debug(self, msg, *args):
+        logging.debug(msg, *args)
+
+    def info(self, msg, *args):
+        logging.info(msg, *args)
+
+    def warn(self, msg, *args):
+        logging.warn(msg, *args)
+
+# This pattern matches a character entity reference (a decimal numeric
+# references, a hexadecimal numeric reference, or a named reference).
+entity_sub = re.compile(r'&(#(\d+|x[\da-fA-F]+)|[\w.:-]+);?').sub
+
+def uchr(c):
+    if not isinstance(c, int):
+        return c
+    if c>255: 
+        return unichr(c)
+    return chr(c)
+
+def decode_entity(match):
+    what = match.group(1)
+    if what.startswith('#x'):
+        what = int(what[2:], 16)
+    elif what.startswith('#'):
+        what = int(what[1:])
+    else:
+        from htmlentitydefs import name2codepoint
+        what = name2codepoint.get(what, match.group(0))
+    return uchr(what)
+
+def htmldecode(text):
+    """Decode HTML entities in the given text."""
+    return entity_sub(decode_entity, text)
+
+def socket_timeout(timeout=15):
+    def _socket_timeout(func):
+        def _socket_timeout(*args, **kwargs):
+            old_timeout = socket.getdefaulttimeout()
+            socket.setdefaulttimeout(timeout)
+            try:
+                return func(*args, **kwargs)
+            finally:
+                socket.setdefaulttimeout(old_timeout)
+        return _socket_timeout
+    return _socket_timeout
+
+def _open_url(self, url):
+    """Open a urllib2 request, handling HTTP authentication"""
+
+    scheme, netloc, path, params, query, frag = urlparse.urlparse(url)
+
+    if scheme in ('http', 'https'):
+        auth, host = urllib2.splituser(netloc)
+    else:
+        auth = None
+
+    if auth:
+        auth = "Basic " + urllib2.unquote(auth).encode('base64').strip()
+        new_url = urlparse.urlunparse((scheme,host,path,params,query,frag))
+        request = urllib2.Request(new_url)
+        request.add_header("Authorization", auth)
+    else:
+        request = urllib2.Request(url)
+
+    request.add_header('User-Agent', user_agent)
+    fp = urllib2.urlopen(request)
+
+    if auth:
+        # Put authentication info back into request URL if same host,
+        # so that links found on the page will work
+        s2, h2, path2, param2, query2, frag2 = urlparse.urlparse(fp.url)
+        if s2==scheme and h2==host:
+            fp.url = urlparse.urlunparse((s2,netloc,path2,param2,query2,frag2))
+
+    return fp
+
+
+# adding a timeout to avoid freezing package_index
+open_with_auth = socket_timeout(_SOCKET_TIMEOUT)(open_with_auth)
+
+def fix_sf_url(url):
+    return url      # backward compatibility
+
+def local_open(url):
+    """Read a local path, with special support for directories"""
+    scheme, server, path, param, query, frag = urlparse.urlparse(url)
+    filename = urllib2.url2pathname(path)
+    if os.path.isfile(filename):
+        return urllib2.urlopen(url)
+    elif path.endswith('/') and os.path.isdir(filename):
+        files = []
+        for f in os.listdir(filename):
+            if f=='index.html':
+                fp = open(os.path.join(filename,f),'rb')
+                body = fp.read()
+                fp.close()
+                break
+            elif os.path.isdir(os.path.join(filename,f)):
+                f+='/'
+            files.append("<a href=%r>%s</a>" % (f,f))
+        else:
+            body = ("<html><head><title>%s</title>" % url) + \
+                "</head><body>%s</body></html>" % '\n'.join(files)
+        status, message = 200, "OK"
+    else:
+        status, message, body = 404, "Path not found", "Not found"
+
+    return urllib2.HTTPError(url, status, message,
+            {'content-type':'text/html'}, cStringIO.StringIO(body))
diff --git a/src/distutils2/pypi/simple.py b/src/distutils2/pypi/simple.py
--- a/src/distutils2/pypi/simple.py
+++ b/src/distutils2/pypi/simple.py
@@ -1,34 +1,99 @@
 """pypi.simple
 
-Contains the class "SimpleIndex" to request the PyPI "simple" index.
+Contains the class "SimpleIndex", a simple spider to find and retreive 
+distributions on the Python Package Index, using it's "simple" API, 
+avalaible at http://pypi.python.org/simple/
 """
+from fnmatch import translate
+import urlparse
+import sys
+import re
+import urllib2
+import socket
 try:
     from hashlib import md5
 except ImportError:
     from md5 import md5
 
 from distutils2.version import VersionPredicate
+from distutils2.pypi.dist import PyPIDistribution, EXTENSIONS
+from distutils2.pypi.errors import PyPIError
+from distutils2 import __version__ as __distutils2_version__
 
-from distutils2.pypi.dist import PyPIDistribution
-from distutils2.pypi.errors import PyPIError
+# -- Constants -----------------------------------------------
+PYPI_DEFAULT_INDEX_URL = "http://pypi.python.org/simple/"
+SOCKET_TIMEOUT=15
+USER_AGENT = "Python-urllib/%s distutils2/%s" % (
+    sys.version[:3], __distutils2_version__
+)
 
+# -- Regexps -------------------------------------------------
+EGG_FRAGMENT = re.compile(r'^egg=([-A-Za-z0-9_.]+)$')
+HREF = re.compile("""href\\s*=\\s*['"]?([^'"> ]+)""", re.I)
+PYPI_MD5 = re.compile(
+    '<a href="([^"#]+)">([^<]+)</a>\n\s+\\(<a (?:title="MD5 hash"\n\s+)'
+    'href="[^?]+\?:action=show_md5&amp;digest=([0-9a-f]{32})">md5</a>\\)'
+)
+URL_SCHEME = re.compile('([-+.a-z0-9]{2,}):',re.I).match
+# This pattern matches a character entity reference (a decimal numeric
+# references, a hexadecimal numeric reference, or a named reference).
+ENTITY_SUB = re.compile(r'&(#(\d+|x[\da-fA-F]+)|[\w.:-]+);?').sub
 
-PYPI_DEFAULT_INDEX_URL = "http://pypi.python.org/simple/"
+def socket_timeout(timeout=SOCKET_TIMEOUT):
+    """Decorator to add a socket timeout when requesting pages on PyPI.
+    """
+    def _socket_timeout(func):
+        def _socket_timeout(*args, **kwargs):
+            old_timeout = socket.getdefaulttimeout()
+            socket.setdefaulttimeout(timeout)
+            try:
+                return func(*args, **kwargs)
+            finally:
+                socket.setdefaulttimeout(old_timeout)
+        return _socket_timeout
+    return _socket_timeout
+
+
+class UrlProcessor(object):
+    """Tools to process urls and return links from them.
+    """
+    def __init__(self):
+        self._processed_urls = {}
+
+    def find_links(self, url, is_browsable=lambda x:True):
+        """Find links and return them, for a specific url.
+        
+        Use "is_browsable" to provides a method to tell if the found urls might
+        be browsed or not.
+        """
+        pass
+
 
 class SimpleIndex(object):
-    """
+    """Provides useful tools to request the Python Package Index simple API
     """
 
     def __init__(self, url=PYPI_DEFAULT_INDEX_URL, 
-        hosts=('*',), ):
+        hosts=('*',), follow_externals=False):
         """Class constructor.
 
         :param index_url: the url of the simple index to search on.
         :param hosts: a list of hosts allowed to be considered as internals (to
         differenciate with external hosts)
+        :param follow_externals: tell if follow external links is needed or not
+        Default is False
         """
         self.index_url = url
-        self.allowed_hosts = hosts
+        self.follow_externals=follow_externals
+         
+        # create a regexp to match all given hosts
+        self._allowed_hosts = re.compile('|'.join(map(translate,hosts))).match
+
+        # _current_requirements are used to know what the index is currently
+        # browsing for. This could be used to determine the distributon name
+        # while having multiple ambigous names/version couples
+        self._current_requirements = None
+        self._processed_pages = []
 
     def get_distributions(self, requirements):
         """Browse the PyPI to find distributions that fullfil the given 
@@ -38,30 +103,57 @@
         version specifiers, as described in PEP345. 
         """
         requirements = VersionPredicate(requirements)
-        self._process_package_index_page(requirements.name)
+        self.current_requirements = requirements
+        
+        # process the index for this project
+        distributions = self._process_pypi_page(requirements.name)
+        
+        # filter with requirements and return the results
+        return distributions.filter(requirements)
 
-    def download(self, requirements, tmp_dir=None):
+    def download(self, requirements, temp_path=None):
         """Download the distribution, using the requirements.
 
         If more than one distribution match the requirements, use the last
         version.
-        Download the distribution, and put it in the tmp_dir. If no tmp_dir is
-        given, creates and return one. 
+        Download the distribution, and put it in the temp_path. If no temp_path 
+        is given, creates and return one. 
 
         Returns the complete absolute path to the downloaded archive.
         """
         distributions = self.get_distributions(requirements)
-        distribution = self._find_last_distribution(distributions)
-        
-        # Download logic goes here
+        return distributions.get_last(requirements) \
+            .download(temp_path=temp_path)
 
-    def _browsable(self, url):
+    def _is_browsable(self, url):
         """Tell if the given URL needs to be browsed or not, according to the
-        object internal attributes"""
-        return True
+        object internal attributes.
 
-    def _process_package_index_page(self, name):
-        """Find and process a PyPI page for the given name.
+        It uses the follow_externals and the hosts list to tell if the given 
+        url is browsable or not. 
+        """
+        if self.follow_externals is True:
+            return True
+        return True if self._allowed_hosts(urlparse(url).netloc) else False
+
+    def _is_distribution(self, link):
+        """Tell if the given URL matches to a distribution name or not.
+        """
+         
+
+    def _process_url(self, url):
+        """Process an url and return fetched links.
+        """
+        f = self._open_url(url)
+        base = f.url
+        for match in HREF.finditer(f.read()):
+            link = urlparse.urljoin(base, self._htmldecode(match.group(1)))
+            if self._is_distribution(link):
+                PyPIDistribution.from_url(link)
+            else:
+
+    def _process_pypi_page(self, name):
+        """Find and process a PyPI page for the given project name.
 
         :param name: the name of the project to find the page
         """
@@ -69,9 +161,55 @@
         # Put all informations about the processed pages in the 
         # _processed_pages attribute.
         url = self.index_url + name + "/"
+        found_links = self._process_url(url)
         
         # Search for external links here, and process them if needed.
-        # use the _browsable method to know if we can process the found links.
         for link in found_links:
-            if self._browsable(link):
+            if self._is_browsable(link):
                 self._search_in_url(link)
+    
+    @socket_timeout()
+    def _open_url(self, url):
+        """Open a urllib2 request, handling HTTP authentication
+        """
+        scheme, netloc, path, params, query, frag = urlparse.urlparse(url)
+
+        if scheme in ('http', 'https'):
+            auth, host = urllib2.splituser(netloc)
+        else:
+            auth = None
+
+        if auth:
+            auth = "Basic " + urllib2.unquote(auth).encode('base64').strip()
+            new_url = urlparse.urlunparse((scheme,host,path,params,query,frag))
+            request = urllib2.Request(new_url)
+            request.add_header("Authorization", auth)
+        else:
+            request = urllib2.Request(url)
+
+        request.add_header('User-Agent', USER_AGENT)
+        fp = urllib2.urlopen(request)
+
+        if auth:
+            # Put authentication info back into request URL if same host,
+            # so that links found on the page will work
+            s2, h2, path2, param2, query2, frag2 = urlparse.urlparse(fp.url)
+            if s2==scheme and h2==host:
+                fp.url = urlparse.urlunparse((s2,netloc,path2,param2,query2,frag2))
+
+        return fp
+
+    def _decode_entity(self, match):
+        what = match.group(1)
+        if what.startswith('#x'):
+            what = int(what[2:], 16)
+        elif what.startswith('#'):
+            what = int(what[1:])
+        else:
+            from htmlentitydefs import name2codepoint
+            what = name2codepoint.get(what, match.group(0))
+        return unichr(what)
+
+    def _htmldecode(self, text):
+        """Decode HTML entities in the given text."""
+        return ENTITY_SUB(self._decode_entity, text)
diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -79,11 +79,28 @@
                     'http://www.famfamfam.com/">')
             index.process_index(url, page)
 
-    def test_url_ok(self):
-        index = simple.SimpleIndex(
-            hosts=('www.example.com',))
-        url = 'file:///tmp/test_simple'
-        self.assertTrue(index.url_ok(url, True))
+    def test_browsable_urls(self):
+        """Test we can given PyPI index a list of allowed hosts
+
+        """
+        index = simple.SimpleIndex(hosts=["pypi.python.org","test.org"])
+        good_urls = (
+            "http://pypi.python.org/foo/bar",
+            "http://pypi.python.org/simple/foobar",
+            "http://test.org",
+            "http://test.org/",
+            "http://test.org/simple/",
+        )
+        bad_urls = (
+            "http://python.org",
+            "http://test.tld",
+        )
+
+        for url in good_urls:
+            self.assertTrue(index._is_browsable(url))
+
+        for url in bad_urls:
+            self.assertFalse(index._is_browsable(url))
     
     @use_pypi_server("test_found_links")
     def test_found_links(self, server):
@@ -162,14 +179,6 @@
         # md5 hash
         self.assertRaises(DistutilsError, pi.download,
             simple.Requirement.parse("badmd5"))
-        # XXX: is this making sense ? do file on pypi can have a wrong md5 
-        # hash ?
-        # If the pypi package index contain a file with the wrong md5, try to
-        # use the external locations to retreive it.
-        # in the test, the "wrongmd5" project contain a bad version of the
-        # archive, but a external site contains the good one. We want to use the
-        # external one to retreive the file with the good md5.
-        # if we really can't manage to get the right md5, raise an exception
         
         for path in paths:
             shutil.rmtree(os.path.dirname(path))

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: pep8 compliance
Message-ID: <E1OVLoV-0000jg-UX@dinsdale.python.org>

tarek.ziade pushed c342370ec357 to distutils2:

http://hg.python.org/distutils2/rev/c342370ec357
changeset:   313:c342370ec357
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Mon Jun 21 18:46:32 2010 +0200
summary:     pep8 compliance
files:       src/distutils2/pypi/dist.py, src/distutils2/pypi/errors.py, src/distutils2/pypi/simple.py, src/distutils2/tests/test_pypi_dist.py, src/distutils2/tests/test_pypi_simple.py

diff --git a/src/distutils2/pypi/dist.py b/src/distutils2/pypi/dist.py
--- a/src/distutils2/pypi/dist.py
+++ b/src/distutils2/pypi/dist.py
@@ -1,6 +1,6 @@
 """distutils2.pypi.dist
 
-Provides the PyPIDistribution class thats represents a distribution retreived 
+Provides the PyPIDistribution class thats represents a distribution retreived
 on PyPI.
 """
 import re
@@ -12,20 +12,21 @@
 except ImportError:
     from md5 import md5
 
-from distutils2.version import suggest_normalized_version 
+from distutils2.version import suggest_normalized_version
 from distutils2.pypi.errors import MD5HashDoesNotMatch
 
 EXTENSIONS = ".tar.gz .tar.bz2 .tar .zip .tgz .egg".split()
 MD5_HASH = re.compile(r'^.*#md5=([a-f0-9]+)$')
 
+
 class PyPIDistribution(object):
     """Represents a distribution retreived from PyPI.
 
-    This is a simple container for various attributes as name, version, 
+    This is a simple container for various attributes as name, version,
     location, url etc.
 
     The PyPIDistribution class is used by the pypi.*Index class to return
-    information about distributions. 
+    information about distributions.
     """
 
     @classmethod
@@ -33,8 +34,8 @@
         """Build a Distribution from a url archive (egg or zip or tgz).
 
         :param url: complete url of the distribution
-        :param probable_dist_name: the probable name of the distribution. This 
-        could be useful when multiple name can be assumed from the archive 
+        :param probable_dist_name: the probable name of the distribution. This
+        could be useful when multiple name can be assumed from the archive
         name.
         """
         # if the url contains a md5 hash, get it.
@@ -44,7 +45,7 @@
             md5_hash = match.group(1)
             # remove the hash
             url = url.replace("#md5=%s" % md5_hash, "")
-        
+
         # parse the archive name to find dist name and version
         archive_name = urlparse.urlparse(url)[2].split('/')[-1]
         extension_matched = False
@@ -67,8 +68,8 @@
         self.type = type
         self.md5_hash = md5_hash
         # set the downloaded path to Null by default. The goal here
-        # is to not download distributions multiple times 
-        self._downloaded_path = None
+        # is to not download distributions multiple times
+        self.location = None
 
     def download(self, url=None, path=None):
         """Download the distribution to a path, and return it.
@@ -79,33 +80,35 @@
             self.url = url
 
         if path is None:
-            path=tempfile.mkdtemp()
-            
+            path = tempfile.mkdtemp()
+
         # if we do not have downloaded it yet, do it.
-        if self._downloaded_path is None:
+        if self.location is None:
             archive_name = urlparse.urlparse(self.url)[2].split('/')[-1]
-            filename, headers = urllib.urlretrieve(self.url, path + "/" + archive_name)
-            self._downloaded_path = filename
+            filename, headers = urllib.urlretrieve(
+                self.url, path + "/" + archive_name)
+            self.location = filename
             self._check_md5(filename)
-        return self._downloaded_path
+        return self.location
 
     def _check_md5(self, filename):
-        """Check that the md5 checksum of the given file matches the one in 
+        """Check that the md5 checksum of the given file matches the one in
         self._md5_hash."""
         if self.md5_hash is not None:
             f = open(filename)
             hash = md5()
             hash.update(f.read())
             if hash.hexdigest() != self.md5_hash:
-                raise MD5HashDoesNotMatch("%s instead of %s" 
-                    % (hash.hexdigest(), self.md5_hash)) 
+                raise MD5HashDoesNotMatch("%s instead of %s"
+                    % (hash.hexdigest(), self.md5_hash))
 
     def __repr__(self):
-        return "<%s %s (%s)>" % (self.__class__.__name__, self.name, self.version)
+        return "<%s %s (%s)>" \
+            % (self.__class__.__name__, self.name, self.version)
 
     def _check_is_comparable(self, other):
         if not isinstance(other, PyPIDistribution):
-            raise TypeError("cannot compare %s and %s" 
+            raise TypeError("cannot compare %s and %s"
                 % (type(self).__name__, type(other).__name__))
         elif self.name != other.name:
             raise TypeError("cannot compare %s and %s"
@@ -134,9 +137,10 @@
     # See http://docs.python.org/reference/datamodel#object.__hash__
     __hash__ = object.__hash__
 
+
 class PyPIDistributions(list):
     """A container of PyPIDistribution objects.
-    
+
     Contains methods and facilities to sort and filter distributions.
     """
 
@@ -149,21 +153,22 @@
             predicate.match(dist.version)])
 
     def get_last(self, predicate):
-        """Return the most up to date version, that satisfy the given 
+        """Return the most up to date version, that satisfy the given
         predicate
         """
         distributions = self.filter(predicate)
         distributions.sort()
         return distributions[-1]
 
+
 def split_archive_name(archive_name, probable_name=None):
     """Split an archive name into two parts: name and version.
 
     Return the tuple (name, version)
     """
-    # Try to determine wich part is the name and wich is the version using the "-"
-    # separator. Take the larger part to be the version number then reduce if this
-    # not works.
+    # Try to determine wich part is the name and wich is the version using the
+    # "-" separator. Take the larger part to be the version number then reduce
+    # if this not works.
     def eager_split(str, maxsplit=2):
         # split using the "-" separator
         splits = str.rsplit("-", maxsplit)
@@ -173,7 +178,7 @@
             version = version[1:]
         if suggest_normalized_version(version) is None and maxsplit >= 0:
             # we dont get a good version number: recurse !
-            return eager_split(str, maxsplit-1)
+            return eager_split(str, maxsplit - 1)
         else:
             return (name, version)
     if probable_name is not None:
@@ -185,9 +190,9 @@
         version = archive_name.lstrip(name)
     else:
         name, version = eager_split(archive_name)
-    
+
     version = suggest_normalized_version(version)
     if version != "" and name != "":
         return (name.lower(), version)
     else:
-        raise CantParseArchiveName(archive_name) 
+        raise CantParseArchiveName(archive_name)
diff --git a/src/distutils2/pypi/errors.py b/src/distutils2/pypi/errors.py
--- a/src/distutils2/pypi/errors.py
+++ b/src/distutils2/pypi/errors.py
@@ -8,15 +8,18 @@
 class PyPIError(DistutilsError):
     """The base class for errors of the pypi python package."""
 
+
 class DistributionNotFound(PyPIError):
     """No distribution match the given requirements."""
 
+
 class CantParseArchiveName(PyPIError):
     """An archive name can't be parsed to find distribution name and version"""
 
+
 class DownloadError(PyPIError):
     """An error has occurs while downloading"""
 
+
 class MD5HashDoesNotMatch(DownloadError):
     """Compared MD5 hashes does not match"""
-    
diff --git a/src/distutils2/pypi/simple.py b/src/distutils2/pypi/simple.py
--- a/src/distutils2/pypi/simple.py
+++ b/src/distutils2/pypi/simple.py
@@ -1,7 +1,7 @@
 """pypi.simple
 
-Contains the class "SimpleIndex", a simple spider to find and retreive 
-distributions on the Python Package Index, using it's "simple" API, 
+Contains the class "SimpleIndex", a simple spider to find and retreive
+distributions on the Python Package Index, using it's "simple" API,
 avalaible at http://pypi.python.org/simple/
 """
 from fnmatch import translate
@@ -21,23 +21,23 @@
 # -- Constants -----------------------------------------------
 PYPI_DEFAULT_INDEX_URL = "http://pypi.python.org/simple/"
 DEFAULT_HOSTS = ("*.python.org",)
-SOCKET_TIMEOUT=15
+SOCKET_TIMEOUT = 15
 USER_AGENT = "Python-urllib/%s distutils2/%s" % (
-    sys.version[:3], __distutils2_version__
-)
+    sys.version[:3], __distutils2_version__)
 
 # -- Regexps -------------------------------------------------
 EGG_FRAGMENT = re.compile(r'^egg=([-A-Za-z0-9_.]+)$')
 HREF = re.compile("""href\\s*=\\s*['"]?([^'"> ]+)""", re.I)
 PYPI_MD5 = re.compile(
     '<a href="([^"#]+)">([^<]+)</a>\n\s+\\(<a (?:title="MD5 hash"\n\s+)'
-    'href="[^?]+\?:action=show_md5&amp;digest=([0-9a-f]{32})">md5</a>\\)'
-)
-URL_SCHEME = re.compile('([-+.a-z0-9]{2,}):',re.I).match
+    'href="[^?]+\?:action=show_md5&amp;digest=([0-9a-f]{32})">md5</a>\\)')
+URL_SCHEME = re.compile('([-+.a-z0-9]{2,}):', re.I).match
+
 # This pattern matches a character entity reference (a decimal numeric
 # references, a hexadecimal numeric reference, or a named reference).
 ENTITY_SUB = re.compile(r'&(#(\d+|x[\da-fA-F]+)|[\w.:-]+);?').sub
 
+
 def socket_timeout(timeout=SOCKET_TIMEOUT):
     """Decorator to add a socket timeout when requesting pages on PyPI.
     """
@@ -57,7 +57,7 @@
     """Provides useful tools to request the Python Package Index simple API
     """
 
-    def __init__(self, url=PYPI_DEFAULT_INDEX_URL, hosts=DEFAULT_HOSTS, 
+    def __init__(self, url=PYPI_DEFAULT_INDEX_URL, hosts=DEFAULT_HOSTS,
         mirrors=[]):
         """Class constructor.
 
@@ -73,7 +73,7 @@
         self.mirrors = mirrors
         self._hosts = hosts
         # create a regexp to match all given hosts
-        self._allowed_hosts = re.compile('|'.join(map(translate,hosts))).match
+        self._allowed_hosts = re.compile('|'.join(map(translate, hosts))).match
 
         # we keep an index of pages we have processed, in order to avoid
         # scanning them multple time (eg. if there is multiple pages pointing
@@ -82,7 +82,7 @@
         self._distributions = {}
 
     def get(self, requirements):
-        """Browse the PyPI index to find distributions that fullfil the 
+        """Browse the PyPI index to find distributions that fullfil the
         given requirements, and return the most recent one.
         """
         predicate = self._get_version_predicate(requirements)
@@ -91,23 +91,23 @@
         if len(dists) == 0:
             raise DistributionNotFound(requirements)
 
-        return dists.get_last(predicate) 
+        return dists.get_last(predicate)
 
     def find(self, requirements):
-        """Browse the PyPI to find distributions that fullfil the given 
+        """Browse the PyPI to find distributions that fullfil the given
         requirements.
 
-        :param requirements: A project name and it's distribution, using 
-        version specifiers, as described in PEP345. You can pass either a 
+        :param requirements: A project name and it's distribution, using
+        version specifiers, as described in PEP345. You can pass either a
         version.VersionPredicate or a string.
         """
         requirements = self._get_version_predicate(requirements)
-        
+
         # process the index for this project
         self._process_pypi_page(requirements.name)
-        
+
         # filter with requirements and return the results
-        if self._distributions.has_key(requirements.name):
+        if requirements.name in self._distributions:
             dists = self._distributions[requirements.name].filter(requirements)
         else:
             dists = []
@@ -119,8 +119,8 @@
 
         If more than one distribution match the requirements, use the last
         version.
-        Download the distribution, and put it in the temp_path. If no temp_path 
-        is given, creates and return one. 
+        Download the distribution, and put it in the temp_path. If no temp_path
+        is given, creates and return one.
 
         Returns the complete absolute path to the downloaded archive.
         """
@@ -129,7 +129,7 @@
         return distributions.get_last(requirements).download(path=temp_path)
 
     def _get_version_predicate(self, requirements):
-        """Return a VersionPredicate object, from a string or an already 
+        """Return a VersionPredicate object, from a string or an already
         existing object.
         """
         if isinstance(requirements, str):
@@ -140,10 +140,10 @@
         """Tell if the given URL needs to be browsed or not, according to the
         object internal attributes.
 
-        It uses the follow_externals and the hosts list to tell if the given 
-        url is browsable or not. 
+        It uses the follow_externals and the hosts list to tell if the given
+        url is browsable or not.
         """
-        if self._allowed_hosts(urlparse.urlparse(url)[1]): # 1 is netloc
+        if self._allowed_hosts(urlparse.urlparse(url)[1]):  # 1 is netloc
             return True
         else:
             return False
@@ -156,7 +156,7 @@
             if ext in link:
                 return True
         return False
-         
+
     def _register_dist(self, dist):
         """Register a distribution as a part of fetched distributions for
         SimpleIndex.
@@ -165,7 +165,7 @@
         """
         # Internally, check if a entry exists with the project name, if not,
         # create a new one, and if exists, add the dist to the pool.
-        if not self._distributions.has_key(dist.name):
+        if not dist.name in self._distributions:
             self._distributions[dist.name] = PyPIDistributions()
         self._distributions[dist.name].append(dist)
         return self._distributions[dist.name]
@@ -188,11 +188,11 @@
                 if self._is_distribution(link):
                     # it's a distribution, so create a dist object
                     self._processed_urls.append(link)
-                    self._register_dist(PyPIDistribution.from_url(link, 
+                    self._register_dist(PyPIDistribution.from_url(link,
                         project_name))
                 else:
                     if self._is_browsable(link) and follow_links:
-                        self._process_url(link, project_name, 
+                        self._process_url(link, project_name,
                             follow_links=False)
 
     def _process_pypi_page(self, name):
@@ -203,7 +203,7 @@
         # Browse and index the content of the given PyPI page.
         url = self.index_url + name + "/"
         self._process_url(url, name)
-    
+
     @socket_timeout()
     def _open_url(self, url):
         """Open a urllib2 request, handling HTTP authentication
@@ -217,8 +217,10 @@
                 auth = None
 
             if auth:
-                auth = "Basic " + urllib2.unquote(auth).encode('base64').strip()
-                new_url = urlparse.urlunparse((scheme,host,path,params,query,frag))
+                auth = "Basic " + \
+                    urllib2.unquote(auth).encode('base64').strip()
+                new_url = urlparse.urlunparse((
+                    scheme, host, path, params, query, frag))
                 request = urllib2.Request(new_url)
                 request.add_header("Authorization", auth)
             else:
@@ -229,9 +231,11 @@
             if auth:
                 # Put authentication info back into request URL if same host,
                 # so that links found on the page will work
-                s2, h2, path2, param2, query2, frag2 = urlparse.urlparse(fp.url)
-                if s2==scheme and h2==host:
-                    fp.url = urlparse.urlunparse((s2,netloc,path2,param2,query2,frag2))
+                s2, h2, path2, param2, query2, frag2 = \
+                    urlparse.urlparse(fp.url)
+                if s2 == scheme and h2 == host:
+                    fp.url = urlparse.urlunparse(
+                        (s2, netloc, path2, param2, query2, frag2))
 
             return fp
         except (ValueError, httplib.InvalidURL), v:
diff --git a/src/distutils2/tests/test_pypi_dist.py b/src/distutils2/tests/test_pypi_dist.py
--- a/src/distutils2/tests/test_pypi_dist.py
+++ b/src/distutils2/tests/test_pypi_dist.py
@@ -27,28 +27,28 @@
         """Test that the Distribution object can be built from a single URL"""
         url_list = {
             'FooBar-1.1.0.tar.gz': {
-                'name': 'foobar', # lowercase the name
+                'name': 'foobar',  # lowercase the name
                 'version': '1.1.0',
             },
             'Foo-Bar-1.1.0.zip': {
-                'name': 'foo-bar', # keep the dash
+                'name': 'foo-bar',  # keep the dash
                 'version': '1.1.0',
             },
             'foobar-1.1b2.tar.gz#md5=123123123123123': {
                 'name': 'foobar',
                 'version': '1.1b2',
-                'url':'http://test.tld/foobar-1.1b2.tar.gz', #without md5 hash
+                'url': 'http://test.tld/foobar-1.1b2.tar.gz',  # without hash
                 'md5_hash': '123123123123123',
-            }, 
-            'foobar-1.1-rc2.tar.gz': { # use suggested name
+            },
+            'foobar-1.1-rc2.tar.gz': {  # use suggested name
                 'name': 'foobar',
                 'version': '1.1c2',
-                'url':'http://test.tld/foobar-1.1-rc2.tar.gz',
+                'url': 'http://test.tld/foobar-1.1-rc2.tar.gz',
             }
         }
 
         for url, attributes in url_list.items():
-            dist = Dist.from_url("http://test.tld/"+url)
+            dist = Dist.from_url("http://test.tld/" + url)
             for attribute, value in attributes.items():
                 self.assertEqual(getattr(dist, attribute), value)
 
@@ -63,14 +63,14 @@
         self.assertFalse(foo1 == foo2)
 
         # assert we can't compare dists with different names
-        self.assertRaises(TypeError, foo1.__eq__,bar)
+        self.assertRaises(TypeError, foo1.__eq__, bar)
 
     def test_split_archive_name(self):
         """Test we can split the archive names"""
         names = {
-            'foo-bar-baz-1.0-rc2': ('foo-bar-baz','1.0c2'),
-            'foo-bar-baz-1.0': ('foo-bar-baz','1.0'),
-            'foobarbaz-1.0': ('foobarbaz','1.0'),
+            'foo-bar-baz-1.0-rc2': ('foo-bar-baz', '1.0c2'),
+            'foo-bar-baz-1.0': ('foo-bar-baz', '1.0'),
+            'foobarbaz-1.0': ('foobarbaz', '1.0'),
         }
         for name, results in names.items():
             self.assertEqual(results, split_archive_name(name))
@@ -78,20 +78,20 @@
     @use_pypi_server("downloads_with_md5")
     def test_download(self, server):
         """Download is possible, and the md5 is checked if given"""
-        
+
         url = "%s/simple/foobar/foobar-0.1.tar.gz" % server.full_address
         # check md5 if given
-        dist = Dist("FooBar", "0.1", url=url, 
+        dist = Dist("FooBar", "0.1", url=url,
             md5_hash="d41d8cd98f00b204e9800998ecf8427e")
         dist.download()
-        
+
         # a wrong md5 fails
-        dist2 = Dist("FooBar", "0.1", url=url, 
+        dist2 = Dist("FooBar", "0.1", url=url,
             md5_hash="wrongmd5")
         self.assertRaises(MD5HashDoesNotMatch, dist2.download)
-        
+
         # we can omit the md5 hash
-        dist3 = Dist("FooBar", "0.1", url=url) 
+        dist3 = Dist("FooBar", "0.1", url=url)
         dist3.download()
 
         # and specify a temporary location
@@ -100,7 +100,7 @@
         dist3.download(path=path1)
         # and for a new one
         path2_base = tempfile.mkdtemp()
-        dist4 = Dist("FooBar", "0.1",url=url) 
+        dist4 = Dist("FooBar", "0.1", url=url)
         path2 = dist4.download(path=path2_base)
         self.assertTrue(path2_base in path2)
 
@@ -108,6 +108,7 @@
         shutil.rmtree(path1)
         shutil.rmtree(os.path.dirname(path2))
 
+
 class TestPyPIDistributions(unittest.TestCase):
     """test the pypi.distr.PyPIDistributions class"""
 
diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -13,9 +13,10 @@
 
 from distutils2.errors import DistutilsError
 
+
 class PyPISimpleTestCase(unittest2.TestCase):
 
-    def _get_simple_index(self, server, base_url="/simple/", hosts=None, 
+    def _get_simple_index(self, server, base_url="/simple/", hosts=None,
         *args, **kwargs):
         """Build and return a SimpleSimpleIndex instance, with the test server
         urls
@@ -86,7 +87,7 @@
         """Test we can give PyPI index a list of allowed hosts
 
         """
-        index = simple.SimpleIndex(hosts=["pypi.python.org","test.org"])
+        index = simple.SimpleIndex(hosts=["pypi.python.org", "test.org"])
         good_urls = (
             "http://pypi.python.org/foo/bar",
             "http://pypi.python.org/simple/foobar",
@@ -104,7 +105,7 @@
 
         for url in bad_urls:
             self.assertFalse(index._is_browsable(url))
-    
+
     @use_pypi_server("test_found_links")
     def test_found_links(self, server):
         """Browse the index, asking for a specified distribution version
@@ -119,7 +120,7 @@
         # we have scanned the index page
         self.assertIn(server.full_address + "/simple/foobar/",
             pi._processed_urls)
-        
+
         # we have found 4 distributions in this page
         self.assertEqual(len(pi._distributions["foobar"]), 4)
 
@@ -131,7 +132,7 @@
         index = simple.SimpleIndex(hosts=("*",))
         self.assertTrue(index._is_browsable("http://an-external.link/path"))
         self.assertTrue(index._is_browsable("pypi.test.tld/a/path"))
-        
+
         # specify a list of hosts we want to allow
         index = simple.SimpleIndex(hosts=("*.test.tld",))
         self.assertFalse(index._is_browsable("http://an-external.link/path"))
@@ -139,7 +140,7 @@
 
     @use_pypi_server("with_externals")
     def test_process_external_pages(self, server):
-        """Include external pages 
+        """Include external pages
         """
         # Try to request the package index, wich contains links to "externals"
         # resources. They have to  be scanned too.
@@ -150,7 +151,7 @@
 
     @use_pypi_server("with_real_externals")
     def test_disable_external_pages(self, server):
-        """Exclude external pages if disabled 
+        """Exclude external pages if disabled
         """
         # Test that telling the simple pyPI client to not retreive external
         # works
@@ -164,16 +165,16 @@
         """Download packages from pypi requests.
         """
         paths = []
-        # If we request a download specific version of a distribution, 
-        # the system must download it, check the md5 and unpack it in a 
+        # If we request a download specific version of a distribution,
+        # the system must download it, check the md5 and unpack it in a
         # temporary location, that must be returned by the lib.
         pi = self._get_simple_index(server)
 
         # assert we can download a specific version
         temp_location_1 = pi.download("foobar (0.1)")
         self.assertIn("foobar-0.1.tar.gz", temp_location_1)
-        paths.append(temp_location_1) # to delete later
-        
+        paths.append(temp_location_1)  # to delete later
+
         # assert we take the latest
         temp_location_2 = pi.download("foobar")
         self.assertIn("foobar-0.1.tar.gz", temp_location_2)
@@ -188,7 +189,7 @@
         # raise an error if we couldnt manage to get the file with a the good
         # md5 hash
         self.assertRaises(DistutilsError, pi.download, "badmd5")
-       
+
         # delete the temp paths
         for path in paths:
             shutil.rmtree(os.path.dirname(path))
@@ -230,6 +231,7 @@
         # the link should be from the index
         self.assertEqual('12345678901234567', dists[0].md5_hash)
 
+
 def test_suite():
     return unittest2.makeSuite(PyPISimpleTestCase)
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: remove download tests from
	test_pypi_simple, while the good place to test this
Message-ID: <E1OVLoW-0000jw-5h@dinsdale.python.org>

tarek.ziade pushed 790e93eb1d9b to distutils2:

http://hg.python.org/distutils2/rev/790e93eb1d9b
changeset:   317:790e93eb1d9b
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Tue Jun 22 11:42:32 2010 +0200
summary:     remove download tests from test_pypi_simple, while the good place to test this is in test_pypi_dist
files:       src/distutils2/pypi/simple.py, src/distutils2/tests/test_pypi_simple.py

diff --git a/src/distutils2/pypi/simple.py b/src/distutils2/pypi/simple.py
--- a/src/distutils2/pypi/simple.py
+++ b/src/distutils2/pypi/simple.py
@@ -125,9 +125,7 @@
 
         Returns the complete absolute path to the downloaded archive.
         """
-        requirements = self._get_version_predicate(requirements)
-        distributions = self.find(requirements)
-        return distributions.get_last(requirements).download(path=temp_path)
+        return self.get(requirements).download(path=temp_path)
 
     def _get_version_predicate(self, requirements):
         """Return a VersionPredicate object, from a string or an already
diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -160,41 +160,6 @@
         self.assertNotIn(server.full_address + "/external/external.html",
             index._processed_urls)
 
-    @use_pypi_server("downloads_with_md5")
-    def download_package(self, server):
-        """Download packages from pypi requests.
-        """
-        paths = []
-        # If we request a download specific version of a distribution,
-        # the system must download it, check the md5 and unpack it in a
-        # temporary location, that must be returned by the lib.
-        index = self._get_simple_index(server)
-
-        # assert we can download a specific version
-        temp_location_1 = index.download("foobar (0.1)")
-        self.assertIn("foobar-0.1.tar.gz", temp_location_1)
-        paths.append(temp_location_1)  # to delete later
-
-        # assert we take the latest
-        temp_location_2 = index.download("foobar")
-        self.assertIn("foobar-0.1.tar.gz", temp_location_2)
-        paths.append(temp_location_2)
-
-        # we also can specify a temp location
-        specific_temp_location = tempfile.mkdtemp()
-        returned_temp_location = index.download("foobar",
-            specific_temp_location)
-        self.assertIn(specific_temp_location, returned_temp_location)
-        paths.append(returned_temp_location)
-
-        # raise an error if we couldnt manage to get the file with a the good
-        # md5 hash
-        self.assertRaises(DistutilsError, index.download, "badmd5")
-
-        # delete the temp paths
-        for path in paths:
-            shutil.rmtree(os.path.dirname(path))
-
     @use_pypi_server("with_egg_files")
     def test_scan_egg_files(self, server):
         """Assert that egg files are indexed as well"""
@@ -257,7 +222,11 @@
         self.assertIn("%s/foobar-2.0.tar.gz" % server.full_address,
             index._processed_urls)  # linked from external homepage (rel)
 
-
+#    def test_uses_mirrors(self):
+#        """When the main repository seems down, try using the given mirrors"""
+#        server = 
+#        mirror = 
+#
 def test_suite():
     return unittest2.makeSuite(PyPISimpleTestCase)
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: PyPIDistribution Download and MD5
	checking
Message-ID: <E1OVLoV-0000jc-Ts@dinsdale.python.org>

tarek.ziade pushed 5419de89aa0c to distutils2:

http://hg.python.org/distutils2/rev/5419de89aa0c
changeset:   312:5419de89aa0c
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Mon Jun 21 18:07:37 2010 +0200
summary:     PyPIDistribution Download and MD5 checking
files:       src/distutils2/pypi/dist.py, src/distutils2/pypi/errors.py, src/distutils2/pypi/simple.py, src/distutils2/tests/test_pypi_dist.py

diff --git a/src/distutils2/pypi/dist.py b/src/distutils2/pypi/dist.py
--- a/src/distutils2/pypi/dist.py
+++ b/src/distutils2/pypi/dist.py
@@ -5,8 +5,15 @@
 """
 import re
 import urlparse
+import urllib
+import tempfile
+try:
+    from hashlib import md5
+except ImportError:
+    from md5 import md5
 
 from distutils2.version import suggest_normalized_version 
+from distutils2.pypi.errors import MD5HashDoesNotMatch
 
 EXTENSIONS = ".tar.gz .tar.bz2 .tar .zip .tgz .egg".split()
 MD5_HASH = re.compile(r'^.*#md5=([a-f0-9]+)$')
@@ -71,11 +78,27 @@
         if url is not None:
             self.url = url
 
+        if path is None:
+            path=tempfile.mkdtemp()
+            
         # if we do not have downloaded it yet, do it.
         if self._downloaded_path is None:
-            # download logic goes here
-            real_path = ""
-        return real_path
+            archive_name = urlparse.urlparse(self.url)[2].split('/')[-1]
+            filename, headers = urllib.urlretrieve(self.url, path + "/" + archive_name)
+            self._downloaded_path = filename
+            self._check_md5(filename)
+        return self._downloaded_path
+
+    def _check_md5(self, filename):
+        """Check that the md5 checksum of the given file matches the one in 
+        self._md5_hash."""
+        if self.md5_hash is not None:
+            f = open(filename)
+            hash = md5()
+            hash.update(f.read())
+            if hash.hexdigest() != self.md5_hash:
+                raise MD5HashDoesNotMatch("%s instead of %s" 
+                    % (hash.hexdigest(), self.md5_hash)) 
 
     def __repr__(self):
         return "<%s %s (%s)>" % (self.__class__.__name__, self.name, self.version)
diff --git a/src/distutils2/pypi/errors.py b/src/distutils2/pypi/errors.py
--- a/src/distutils2/pypi/errors.py
+++ b/src/distutils2/pypi/errors.py
@@ -13,3 +13,10 @@
 
 class CantParseArchiveName(PyPIError):
     """An archive name can't be parsed to find distribution name and version"""
+
+class DownloadError(PyPIError):
+    """An error has occurs while downloading"""
+
+class MD5HashDoesNotMatch(DownloadError):
+    """Compared MD5 hashes does not match"""
+    
diff --git a/src/distutils2/pypi/simple.py b/src/distutils2/pypi/simple.py
--- a/src/distutils2/pypi/simple.py
+++ b/src/distutils2/pypi/simple.py
@@ -11,10 +11,6 @@
 import urllib2
 import httplib
 import socket
-try:
-    from hashlib import md5
-except ImportError:
-    from md5 import md5
 
 from distutils2.version import VersionPredicate
 from distutils2.pypi.dist import PyPIDistribution, PyPIDistributions, \
diff --git a/src/distutils2/tests/test_pypi_dist.py b/src/distutils2/tests/test_pypi_dist.py
--- a/src/distutils2/tests/test_pypi_dist.py
+++ b/src/distutils2/tests/test_pypi_dist.py
@@ -1,10 +1,17 @@
 """Tests for the distutils2.pypi.dist module.
 """
 
+from distutils2.tests.pypi_server import use_pypi_server
+from distutils2.tests.support import unittest
+from distutils2.version import VersionPredicate
+from distutils2.pypi.errors import MD5HashDoesNotMatch
 from distutils2.pypi.dist import PyPIDistribution as Dist, \
     PyPIDistributions as Dists, split_archive_name
-from distutils2.tests.support import unittest
-from distutils2.version import VersionPredicate
+
+import tempfile
+import os
+import shutil
+
 
 class TestPyPIDistribution(unittest.TestCase):
     """tests the pypi.dist.PyPIDistribution class"""
@@ -68,6 +75,39 @@
         for name, results in names.items():
             self.assertEqual(results, split_archive_name(name))
 
+    @use_pypi_server("downloads_with_md5")
+    def test_download(self, server):
+        """Download is possible, and the md5 is checked if given"""
+        
+        url = "%s/simple/foobar/foobar-0.1.tar.gz" % server.full_address
+        # check md5 if given
+        dist = Dist("FooBar", "0.1", url=url, 
+            md5_hash="d41d8cd98f00b204e9800998ecf8427e")
+        dist.download()
+        
+        # a wrong md5 fails
+        dist2 = Dist("FooBar", "0.1", url=url, 
+            md5_hash="wrongmd5")
+        self.assertRaises(MD5HashDoesNotMatch, dist2.download)
+        
+        # we can omit the md5 hash
+        dist3 = Dist("FooBar", "0.1", url=url) 
+        dist3.download()
+
+        # and specify a temporary location
+        # for an already downloaded dist
+        path1 = tempfile.mkdtemp()
+        dist3.download(path=path1)
+        # and for a new one
+        path2_base = tempfile.mkdtemp()
+        dist4 = Dist("FooBar", "0.1",url=url) 
+        path2 = dist4.download(path=path2_base)
+        self.assertTrue(path2_base in path2)
+
+        # remove the temp folders
+        shutil.rmtree(path1)
+        shutil.rmtree(os.path.dirname(path2))
+
 class TestPyPIDistributions(unittest.TestCase):
     """test the pypi.distr.PyPIDistributions class"""
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:39 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:39 +0200
Subject: [Python-checkins] distutils2: update the documentation
Message-ID: <E1OVLoV-0000jk-V9@dinsdale.python.org>

tarek.ziade pushed e73d8ec4a658 to distutils2:

http://hg.python.org/distutils2/rev/e73d8ec4a658
changeset:   314:e73d8ec4a658
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Mon Jun 21 19:17:45 2010 +0200
summary:     update the documentation
files:       docs/source/pypi.rst

diff --git a/docs/source/pypi.rst b/docs/source/pypi.rst
--- a/docs/source/pypi.rst
+++ b/docs/source/pypi.rst
@@ -18,8 +18,12 @@
 `distutils2.pypi.dist`
 ======================
 
-Both `SimpleIndex` and `XmlRpcIndex` classes works with `PyPIDistribution` 
-classes.
+Both `SimpleIndex` and `XmlRpcIndex` classes works with the classes provided
+in the `pypi.dist` package.
+
+`PyPIDistribution`
+------------------
+
 `PyPIDistribution` is a simple class that defines the following attributes:
 
 :name:
@@ -33,6 +37,13 @@
     The url of the distribution
 
 
+`PyPIDistributions`
+-------------------
+
+The `dist` module also provides another class, to work with lists of 
+`PyPIDistribution` classes.
+
+
 Requesting information via the "simple" API `distutils2.pypi.simple`
 ====================================================================
 
@@ -57,25 +68,25 @@
 --------------------------------------------
 
 Supposing you want to scan the PyPI index to get a list of distributions for 
-the "foobar" project::
+the "foobar" project. You can use the "search" method for that.::
 
     >>> from distutils2.pypi import SimpleIndex
     >>> client = SimpleIndex()
-    >>> client.get_distributions("foobar")
+    >>> client.find("foobar")
     [<PyPIDistribution "Foobar 1.1">, <PyPIDistribution "Foobar 1.2">]
     
-Then, you can get all the distribution found by the `Simple` class using the
-following syntax::
-
-    >>> client["foobar"]
-    [<PyPIDistribution "Foobar 1.1">, <PyPIDistribution "Foobar 1.2">]
-
 Note that you also can request the client about specific versions, using version
 specifiers (described in `PEP 345 
 <http://www.python.org/dev/peps/pep-0345/#version-specifiers>`_)::
 
-    >>> client.get_distributions("foobar < 1.2")
-    [<PyPIDistribution "Foobar 1.1">, ]
+    >>> client.find("foobar < 1.2")
+    [<PyPIDistribution "foobar 1.1">, ]
+
+`find` returns a list of distributions, but you also can get the last
+distribution (the more up to date) that fullfil your requirements, like this::
+    
+    >>> client.get("foobar < 1.2")
+    <PyPIDistribution "foobar 1.1">
 
 Download distributions
 ----------------------
@@ -93,9 +104,7 @@
     /path/to/my/dir/foobar-1.2.tar.gz
 
 While downloading, the md5 of the archive will be checked, if not matches, it
-will try another time, then if fails again, raise `InvalidMd5Error`.
-
-XXX: Maybe try external urls if provided, if it fails ?
+will try another time, then if fails again, raise `MD5HashDoesNotMatchError`.
 
 Requesting external pages
 -------------------------
@@ -108,13 +117,9 @@
 falling back to external webpages if needed.
 
 It's possible to tell the PyPIClient to not follow external links by specifying
-the `follow_external` parameter to the constructor::
+a list of allowed hosts::
 
-    >>> client = SimpleIndex(follow_externals=False)
-
-Or later by changing it's value::
-
-    >>> client.follow_externals = True
+    >>> client = SimpleIndex(hosts=("*.python.org"))
 
 
 Requesting informations via XML-RPC (`distutils2.pypi.XmlRpcIndex`)
@@ -134,4 +139,4 @@
 =================
 
 XXX : A description about a wraper around PyPI simple and XmlRpc Indexes
-(PyPiIndex ?) 
+(PyPIIndex ?) 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: Udpate PyPI simple documentation to
	reflect new wanted	API.
Message-ID: <E1OVLoW-0000kC-8w@dinsdale.python.org>

tarek.ziade pushed d1d7547cd1bd to distutils2:

http://hg.python.org/distutils2/rev/d1d7547cd1bd
changeset:   321:d1d7547cd1bd
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Sun Jun 27 19:09:18 2010 +0200
summary:     Udpate PyPI simple documentation to reflect new wanted API.
files:       docs/source/pypi.rst

diff --git a/docs/source/pypi.rst b/docs/source/pypi.rst
--- a/docs/source/pypi.rst
+++ b/docs/source/pypi.rst
@@ -8,12 +8,15 @@
 
 There is two ways to retrieve data from pypi: using the *simple* API, and using
 *XML-RPC*. The first one is in fact a set of HTML pages avalaible at
-`http://pypi.python.org/simple/`. In order to reduce the overload caused by
-running distant methods on the pypi server (by using the XML-RPC methods), the
-best way to retrieve informations is by using the simple API.
+`http://pypi.python.org/simple/`, and the second one contains a set of XML-RPC
+methods. In order to reduce the overload caused by running distant methods on 
+the pypi server (by using the XML-RPC methods), the best way to retrieve 
+informations is by using the simple API, when it contains the information you 
+need.
 
 Distutils2 provides two python modules to ease the work with those two APIs:
-`distutils2.pypi.simple` and `distutils2.pypi.xmlrpc`.
+`distutils2.pypi.simple` and `distutils2.pypi.xmlrpc`. Both of them depends on 
+another python module: `distutils2.pypi.dist`.
 
 
 Requesting information via the "simple" API `distutils2.pypi.simple`
@@ -24,8 +27,7 @@
 can process external html pages, with the goal to find *pypi unhosted* versions 
 of python distributions.
 
-You should use `distutils2.pypi.simple` way of accessing data should be used 
-to:
+You should use `distutils2.pypi.simple` for: 
 
     * Search distributions by name and versions.
     * Process pypi external pages.
@@ -39,17 +41,22 @@
 API
 ----
 
+Here is a complete overview of the APIs of the SimpleIndex class.
+
 .. autoclass:: distutils2.pypi.simple.SimpleIndex
     :members:
 
 Usage Exemples
 ---------------
 
+To help you understand how using the `SimpleIndex` class, here are some basic
+usages.
+
 Request PyPI to get a specific distribution
 ++++++++++++++++++++++++++++++++++++++++++++
 
 Supposing you want to scan the PyPI index to get a list of distributions for 
-the "foobar" project. You can use the "search" method for that.::
+the "foobar" project. You can use the "find" method for that::
 
     >>> from distutils2.pypi import SimpleIndex
     >>> client = SimpleIndex()
@@ -72,8 +79,8 @@
 Download distributions
 +++++++++++++++++++++++
 
-As it can get the urls of distributions provided by PyPI, the `Simple` client
-also can download the distributions and put it for you in a temporary
+As it can get the urls of distributions provided by PyPI, the `SimpleIndex` 
+client also can download the distributions and put it for you in a temporary
 destination::
 
     >>> client.download("foobar")
@@ -87,26 +94,51 @@
 While downloading, the md5 of the archive will be checked, if not matches, it
 will try another time, then if fails again, raise `MD5HashDoesNotMatchError`.
 
-Requesting external pages
-+++++++++++++++++++++++++
+Internally, that's not the SimpleIndex which download the distributions, but the
+`PyPIDistribution` class. Please refer to this documentation for more details.
 
-The default behavior for distutils2 is to follow the links provided
+Following PyPI external links
+++++++++++++++++++++++++++++++
+
+The default behavior for distutils2 is to *not* follow the links provided
 by HTML pages in the "simple index", to find distributions related
 downloads.
 
-It's possible to tell the PyPIClient to not follow external links by specifying
-a list of allowed hosts::
+It's possible to tell the PyPIClient to follow external links by setting the 
+`follow_externals` attribute, on instanciation or after::
 
-    >>> client = SimpleIndex(hosts=("*.python.org"))
+    >>> client = SimpleIndex(follow_externals=True)
 
-Working with mirrors
-+++++++++++++++++++++
+or ::
 
-The SimpleClient implement a fallback mechanism to switch from one mirror to 
-another, the simple way. All you need to do is to provide a list of mirrors to
-it at the instanciation time::
+    >>> client = SimpleIndex()
+    >>> client.follow_externals = True
 
-    >>> client = SimpleClient(mirrors=['http://mirror1/', 'http://mirror2/'])
+Working with external indexes, and mirrors
++++++++++++++++++++++++++++++++++++++++++++
+
+The default `SimpleIndex` behavior is to rely on the Python Package index stored
+on PyPI (http://pypi.python.org/simple).
+
+As you can need to work with a local index, or private indexes, you can specify
+it using the index_url parameter::
+
+    >>> client = SimpleIndex(index_url="file://filesystem/path/")
+
+or ::
+
+    >>> client = SimpleIndex(index_url="http://some.specific.url/")
+
+You also can specify mirrors to fallback on in case the first index_url you
+provided doesnt respond, or not correctly. The default behavior for
+`SimpleIndex` is to use the list provided by Python.org DNS records, as
+described in the :pep:`381` about mirroring infrastructure.
+
+If you don't want to rely on these, you could specify the list of mirrors you
+want to try by specifying the `mirrors` attribute. It's a simple iterable::
+
+    >>> mirrors = ["http://first.mirror","http://second.mirror"]
+    >>> client = SimpleIndex(mirrors=mirrors)
 
 
 Requesting informations via XML-RPC (`distutils2.pypi.XmlRpcIndex`)

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: correct a typo
Message-ID: <E1OVLoW-0000k7-8N@dinsdale.python.org>

tarek.ziade pushed 1b75f9a8dd54 to distutils2:

http://hg.python.org/distutils2/rev/1b75f9a8dd54
changeset:   320:1b75f9a8dd54
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Tue Jun 22 16:17:44 2010 +0200
summary:     correct a typo
files:       docs/source/pypi.rst, src/distutils2/pypi/dist.py, src/distutils2/pypi/simple.py, src/distutils2/tests/test_pypi_simple.py

diff --git a/docs/source/pypi.rst b/docs/source/pypi.rst
--- a/docs/source/pypi.rst
+++ b/docs/source/pypi.rst
@@ -6,11 +6,11 @@
 facilities to access the Python Package Index (named "pypi", and avalaible on
 the url `http://pypi.python.org`.
 
-There is two ways to retreive data from pypi: using the *simple* API, and using
+There is two ways to retrieve data from pypi: using the *simple* API, and using
 *XML-RPC*. The first one is in fact a set of HTML pages avalaible at
 `http://pypi.python.org/simple/`. In order to reduce the overload caused by
 running distant methods on the pypi server (by using the XML-RPC methods), the
-best way to retreive informations is by using the simple API.
+best way to retrieve informations is by using the simple API.
 
 Distutils2 provides two python modules to ease the work with those two APIs:
 `distutils2.pypi.simple` and `distutils2.pypi.xmlrpc`.
diff --git a/src/distutils2/pypi/dist.py b/src/distutils2/pypi/dist.py
--- a/src/distutils2/pypi/dist.py
+++ b/src/distutils2/pypi/dist.py
@@ -1,6 +1,6 @@
 """distutils2.pypi.dist
 
-Provides the PyPIDistribution class thats represents a distribution retreived
+Provides the PyPIDistribution class thats represents a distribution retrieved
 on PyPI.
 """
 import re
@@ -20,7 +20,7 @@
 
 
 class PyPIDistribution(object):
-    """Represents a distribution retreived from PyPI.
+    """Represents a distribution retrieved from PyPI.
 
     This is a simple container for various attributes as name, version,
     location, url etc.
diff --git a/src/distutils2/pypi/simple.py b/src/distutils2/pypi/simple.py
--- a/src/distutils2/pypi/simple.py
+++ b/src/distutils2/pypi/simple.py
@@ -1,6 +1,6 @@
 """pypi.simple
 
-Contains the class "SimpleIndex", a simple spider to find and retreive
+Contains the class "SimpleIndex", a simple spider to find and retrieve
 distributions on the Python Package Index, using it's "simple" API,
 avalaible at http://pypi.python.org/simple/
 """
diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -153,7 +153,7 @@
     def test_disable_external_pages(self, server):
         """Exclude external pages if disabled
         """
-        # Test that telling the simple pyPI client to not retreive external
+        # Test that telling the simple pyPI client to not retrieve external
         # works
         index = self._get_simple_index(server, hosts=(server.full_address,))
         index.get("foobar")

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: Only retreive links that have
	rel="homepage" or	rel="download", from PyPI pages.
Message-ID: <E1OVLoW-0000js-4Y@dinsdale.python.org>

tarek.ziade pushed 4b8eaa414258 to distutils2:

http://hg.python.org/distutils2/rev/4b8eaa414258
changeset:   316:4b8eaa414258
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Tue Jun 22 11:14:06 2010 +0200
summary:     Only retreive links that have rel="homepage" or rel="download", from PyPI pages.
files:       src/distutils2/pypi/simple.py, src/distutils2/tests/pypiserver/downloads_with_md5/simple/badmd5/index.html, src/distutils2/tests/pypiserver/downloads_with_md5/simple/foobar/index.html, src/distutils2/tests/pypiserver/test_found_links/simple/foobar/index.html, src/distutils2/tests/pypiserver/with_externals/simple/foobar/index.html, src/distutils2/tests/pypiserver/with_norel_links/external/homepage.html, src/distutils2/tests/pypiserver/with_norel_links/external/nonrel.html, src/distutils2/tests/pypiserver/with_norel_links/simple/foobar/index.html, src/distutils2/tests/pypiserver/with_norel_links/simple/index.html, src/distutils2/tests/pypiserver/with_real_externals/simple/foobar/index.html, src/distutils2/tests/test_pypi_simple.py

diff --git a/src/distutils2/pypi/simple.py b/src/distutils2/pypi/simple.py
--- a/src/distutils2/pypi/simple.py
+++ b/src/distutils2/pypi/simple.py
@@ -36,6 +36,7 @@
 # This pattern matches a character entity reference (a decimal numeric
 # references, a hexadecimal numeric reference, or a named reference).
 ENTITY_SUB = re.compile(r'&(#(\d+|x[\da-fA-F]+)|[\w.:-]+);?').sub
+REL = re.compile("""<([^>]*\srel\s*=\s*['"]?([^'">]+)[^>]*)>""", re.I)
 
 
 def socket_timeout(timeout=SOCKET_TIMEOUT):
@@ -170,30 +171,60 @@
         self._distributions[dist.name].append(dist)
         return self._distributions[dist.name]
 
-    def _process_url(self, url, project_name=None, follow_links=True):
+    def _process_url(self, url, project_name=None,follow_links=True, 
+        filter_rel=False):
         """Process an url and search for distributions packages.
 
         :param url: the url to analyse
         :param project_name: the project name we are searching for.
-        :param follow_links: We do not want to follow links more than from one
+        :param follow_links: Do not want to follow links more than from one
         level. This parameter tells if we want to follow the links we find (eg.
         run recursively this method on it)
         """
         f = self._open_url(url)
-        base = f.url
-        self._processed_urls.append(url)
-        for match in HREF.finditer(f.read()):
-            link = urlparse.urljoin(base, self._htmldecode(match.group(1)))
-            if link not in self._processed_urls:
-                if self._is_distribution(link):
-                    # it's a distribution, so create a dist object
-                    self._processed_urls.append(link)
-                    self._register_dist(PyPIDistribution.from_url(link,
-                        project_name))
-                else:
-                    if self._is_browsable(link) and follow_links:
-                        self._process_url(link, project_name,
-                            follow_links=False)
+        base_url = f.url
+        if url not in self._processed_urls:
+            self._processed_urls.append(url)
+            link_matcher = self._get_link_matcher(url)
+            for link, is_download in link_matcher(f.read(), base_url):
+                if link not in self._processed_urls:
+                    if self._is_distribution(link) or is_download:
+                        # it's a distribution, so create a dist object
+                        self._processed_urls.append(link)
+                        self._register_dist(PyPIDistribution.from_url(link,
+                            project_name))
+                    else:
+                        if self._is_browsable(link) and follow_links:
+                            self._process_url(link, project_name,
+                                follow_links=False)
+    
+    def _get_link_matcher(self, url):
+        """Returns the right link matcher function of the given url
+        """
+        if self.index_url in url:
+            return self._simple_link_finder
+        else:
+            return self._default_link_finder
+
+    def _simple_link_finder(self, content, base_url):
+        """Yield all links with a rel="download" or rel="homepage".
+
+        This matches the simple index requirements for matching links
+        """
+        for match in REL.finditer(content):
+            tag, rel = match.groups()
+            rels = map(str.strip, rel.lower().split(','))
+            if 'homepage' in rels or 'download' in rels:
+                for match in HREF.finditer(tag):
+                    yield (urlparse.urljoin(base_url, 
+                        self._htmldecode(match.group(1))), "downloads" in rels)
+
+    def _default_link_finder(self, content, base_url):
+        """Yield all links found on the page.
+        """
+        for match in HREF.finditer(content):
+            yield (urlparse.urljoin(base_url, 
+                self._htmldecode(match.group(1))), False)
 
     def _process_pypi_page(self, name):
         """Find and process a PyPI page for the given project name.
diff --git a/src/distutils2/tests/pypiserver/downloads_with_md5/simple/badmd5/index.html b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/badmd5/index.html
--- a/src/distutils2/tests/pypiserver/downloads_with_md5/simple/badmd5/index.html
+++ b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/badmd5/index.html
@@ -1,3 +1,3 @@
 <html><body>
-<a href="badmd5-0.1.tar.gz#md5=3e3d86693d6564c807272b11b3069dfe">badmd5-0.1.tar.gz</a><br/>
+<a href="badmd5-0.1.tar.gz#md5=3e3d86693d6564c807272b11b3069dfe" rel="download">badmd5-0.1.tar.gz</a><br/>
 </body></html>
diff --git a/src/distutils2/tests/pypiserver/downloads_with_md5/simple/foobar/index.html b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/foobar/index.html
--- a/src/distutils2/tests/pypiserver/downloads_with_md5/simple/foobar/index.html
+++ b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/foobar/index.html
@@ -1,3 +1,3 @@
 <html><body>
-<a href="foobar-0.1.tar.gz#md5=d41d8cd98f00b204e9800998ecf8427e">foobar-0.1.tar.gz</a><br/>
+<a href="foobar-0.1.tar.gz#md5=d41d8cd98f00b204e9800998ecf8427e" rel="download">foobar-0.1.tar.gz</a><br/>
 </body></html>
diff --git a/src/distutils2/tests/pypiserver/test_found_links/simple/foobar/index.html b/src/distutils2/tests/pypiserver/test_found_links/simple/foobar/index.html
--- a/src/distutils2/tests/pypiserver/test_found_links/simple/foobar/index.html
+++ b/src/distutils2/tests/pypiserver/test_found_links/simple/foobar/index.html
@@ -1,6 +1,6 @@
 <html><head><title>Links for Foobar</title></head><body><h1>Links for Foobar</h1>
-<a href="../../packages/source/F/Foobar/Foobar-1.0.tar.gz#md5=98fa833fdabcdd78d00245aead66c174">Foobar-1.0.tar.gz</a><br/> 
-<a href="../../packages/source/F/Foobar/Foobar-1.0.1.tar.gz#md5=2351efb20f6b7b5d9ce80fa4cb1bd9ca">Foobar-1.0.1.tar.gz</a><br/> 
-<a href="../../packages/source/F/Foobar/Foobar-2.0.tar.gz#md5=98fa833fdabcdd78d00245aead66c274">Foobar-2.0.tar.gz</a><br/> 
-<a href="../../packages/source/F/Foobar/Foobar-2.0.1.tar.gz#md5=2352efb20f6b7b5d9ce80fa4cb2bd9ca">Foobar-2.0.1.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/Foobar/Foobar-1.0.tar.gz#md5=98fa833fdabcdd78d00245aead66c174">Foobar-1.0.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/Foobar/Foobar-1.0.1.tar.gz#md5=2351efb20f6b7b5d9ce80fa4cb1bd9ca">Foobar-1.0.1.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/Foobar/Foobar-2.0.tar.gz#md5=98fa833fdabcdd78d00245aead66c274">Foobar-2.0.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/Foobar/Foobar-2.0.1.tar.gz#md5=2352efb20f6b7b5d9ce80fa4cb2bd9ca">Foobar-2.0.1.tar.gz</a><br/> 
 </body></html>
diff --git a/src/distutils2/tests/pypiserver/with_externals/simple/foobar/index.html b/src/distutils2/tests/pypiserver/with_externals/simple/foobar/index.html
--- a/src/distutils2/tests/pypiserver/with_externals/simple/foobar/index.html
+++ b/src/distutils2/tests/pypiserver/with_externals/simple/foobar/index.html
@@ -1,4 +1,4 @@
 <html><body>
-<a href="/foobar-0.1.tar.gz#md5=12345678901234567">foobar-0.1.tar.gz</a><br/>
+<a rel ="download" href="/foobar-0.1.tar.gz#md5=12345678901234567">foobar-0.1.tar.gz</a><br/>
 <a href="../../external/external.html" rel="homepage">external homepage</a><br/>
 </body></html>
diff --git a/src/distutils2/tests/pypiserver/with_norel_links/external/homepage.html b/src/distutils2/tests/pypiserver/with_norel_links/external/homepage.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/with_norel_links/external/homepage.html
@@ -0,0 +1,7 @@
+<html>
+<body>
+<p>a rel=homepage HTML page</p>
+<a href="/foobar-2.0.tar.gz">foobar 2.0</a>
+</body>
+</html>
+
diff --git a/src/distutils2/tests/pypiserver/with_norel_links/external/nonrel.html b/src/distutils2/tests/pypiserver/with_norel_links/external/nonrel.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/with_norel_links/external/nonrel.html
@@ -0,0 +1,1 @@
+A page linked without rel="download" or rel="homepage" link.
diff --git a/src/distutils2/tests/pypiserver/with_norel_links/simple/foobar/index.html b/src/distutils2/tests/pypiserver/with_norel_links/simple/foobar/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/with_norel_links/simple/foobar/index.html
@@ -0,0 +1,6 @@
+<html><body>
+<a rel="download" href="/foobar-0.1.tar.gz" rel="download">foobar-0.1.tar.gz</a><br/>
+<a href="../../external/homepage.html" rel="homepage">external homepage</a><br/>
+<a href="../../external/nonrel.html">unrelated link</a><br/>
+<a href="/unrelated-0.2.tar.gz">unrelated download</a></br/>
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/with_norel_links/simple/index.html b/src/distutils2/tests/pypiserver/with_norel_links/simple/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/with_norel_links/simple/index.html
@@ -0,0 +1,1 @@
+<a href="foobar/">foobar/</a> 
diff --git a/src/distutils2/tests/pypiserver/with_real_externals/simple/foobar/index.html b/src/distutils2/tests/pypiserver/with_real_externals/simple/foobar/index.html
--- a/src/distutils2/tests/pypiserver/with_real_externals/simple/foobar/index.html
+++ b/src/distutils2/tests/pypiserver/with_real_externals/simple/foobar/index.html
@@ -1,4 +1,4 @@
 <html><body>
-<a href="/foobar-0.1.tar.gz#md5=0_correct_md5">foobar-0.1.tar.gz</a><br/>
+<a rel="download" href="/foobar-0.1.tar.gz#md5=0_correct_md5">foobar-0.1.tar.gz</a><br/>
 <a href="http://a-really-external-website/external/external.html" rel="homepage">external homepage</a><br/>
 </body></html>
diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -22,7 +22,7 @@
         urls
         """
         if hosts is None:
-            hosts = (server.full_address,)
+            hosts = (server.full_address.strip("http://"),)
         kwargs['hosts'] = hosts
         return simple.SimpleIndex(server.full_address + base_url, *args,
             **kwargs)
@@ -114,15 +114,15 @@
         # We query only for the version 1.1, so all distributions must be
         # filled in the package_index (as the url has been scanned), but
         # "get" must only return the one we want.
-        pi = self._get_simple_index(server)
-        last_distribution = pi.get("foobar")
+        index = self._get_simple_index(server)
+        last_distribution = index.get("foobar")
 
         # we have scanned the index page
         self.assertIn(server.full_address + "/simple/foobar/",
-            pi._processed_urls)
+            index._processed_urls)
 
         # we have found 4 distributions in this page
-        self.assertEqual(len(pi._distributions["foobar"]), 4)
+        self.assertEqual(len(index._distributions["foobar"]), 4)
 
         # and returned the most recent one
         self.assertEqual(last_distribution.version, '2.0.1')
@@ -144,10 +144,10 @@
         """
         # Try to request the package index, wich contains links to "externals"
         # resources. They have to  be scanned too.
-        pi = self._get_simple_index(server, hosts=("*",))
-        pi.get("foobar")
+        index = self._get_simple_index(server, hosts=("*",))
+        index.get("foobar")
         self.assertIn(server.full_address + "/external/external.html",
-            pi._processed_urls)
+            index._processed_urls)
 
     @use_pypi_server("with_real_externals")
     def test_disable_external_pages(self, server):
@@ -155,10 +155,10 @@
         """
         # Test that telling the simple pyPI client to not retreive external
         # works
-        pi = self._get_simple_index(server, hosts=(server.full_address,))
-        pi.get("foobar")
+        index = self._get_simple_index(server, hosts=(server.full_address,))
+        index.get("foobar")
         self.assertNotIn(server.full_address + "/external/external.html",
-            pi._processed_urls)
+            index._processed_urls)
 
     @use_pypi_server("downloads_with_md5")
     def download_package(self, server):
@@ -168,27 +168,28 @@
         # If we request a download specific version of a distribution,
         # the system must download it, check the md5 and unpack it in a
         # temporary location, that must be returned by the lib.
-        pi = self._get_simple_index(server)
+        index = self._get_simple_index(server)
 
         # assert we can download a specific version
-        temp_location_1 = pi.download("foobar (0.1)")
+        temp_location_1 = index.download("foobar (0.1)")
         self.assertIn("foobar-0.1.tar.gz", temp_location_1)
         paths.append(temp_location_1)  # to delete later
 
         # assert we take the latest
-        temp_location_2 = pi.download("foobar")
+        temp_location_2 = index.download("foobar")
         self.assertIn("foobar-0.1.tar.gz", temp_location_2)
         paths.append(temp_location_2)
 
         # we also can specify a temp location
         specific_temp_location = tempfile.mkdtemp()
-        returned_temp_location = pi.download("foobar", specific_temp_location)
+        returned_temp_location = index.download("foobar",
+            specific_temp_location)
         self.assertIn(specific_temp_location, returned_temp_location)
         paths.append(returned_temp_location)
 
         # raise an error if we couldnt manage to get the file with a the good
         # md5 hash
-        self.assertRaises(DistutilsError, pi.download, "badmd5")
+        self.assertRaises(DistutilsError, index.download, "badmd5")
 
         # delete the temp paths
         for path in paths:
@@ -209,8 +210,8 @@
 
         Usecase :
         - someone uploads a package on pypi, a md5 is generated
-        - someone manually copies this link (with the md5 in the url) onto an
-          external page accessible from the package page.
+        - someone manually coindexes this link (with the md5 in the url) onto
+          an external page accessible from the package page.
         - someone reuploads the package (with a different md5)
         - while easy_installing, an MD5 error occurs because the external link
           is used
@@ -220,8 +221,8 @@
         index_url = server.full_address + '/simple/'
 
         # scan a test index
-        pi = simple.SimpleIndex(index_url)
-        dists = pi.find("foobar")
+        index = simple.SimpleIndex(index_url)
+        dists = index.find("foobar")
         server.stop()
 
         # the distribution has been found
@@ -231,6 +232,31 @@
         # the link should be from the index
         self.assertEqual('12345678901234567', dists[0].md5_hash)
 
+    @use_pypi_server(static_filesystem_paths=["with_norel_links"],
+        static_uri_paths=["simple", "external"])
+    def test_not_scan_all_links(self, server):
+        """Do not follow all index page links.
+        The links not tagged with rel="download" and rel="homepage" have
+        to not be processed by the package index, while processing "pages".
+        """
+        # process the pages
+        index = self._get_simple_index(server)
+        index.find("foobar")
+        # now it should have processed only pages with links rel="download" 
+        # and rel="homepage"
+        self.assertIn("%s/simple/foobar/" % server.full_address, 
+            index._processed_urls)  # it's the simple index page
+        self.assertIn("%s/external/homepage.html" % server.full_address, 
+            index._processed_urls)  # the external homepage is rel="homepage"
+        self.assertNotIn("%s/external/nonrel.html" % server.full_address, 
+            index._processed_urls)  # this link contains no rel=*
+        self.assertNotIn("%s/unrelated-0.2.tar.gz" % server.full_address,
+            index._processed_urls)  # linked from simple index (no rel)
+        self.assertIn("%s/foobar-0.1.tar.gz" % server.full_address,
+            index._processed_urls)  # linked from simple index (rel)
+        self.assertIn("%s/foobar-2.0.tar.gz" % server.full_address,
+            index._processed_urls)  # linked from external homepage (rel)
+
 
 def test_suite():
     return unittest2.makeSuite(PyPISimpleTestCase)

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: merged upstream
Message-ID: <E1OVLoW-0000jo-39@dinsdale.python.org>

tarek.ziade pushed f39830d92939 to distutils2:

http://hg.python.org/distutils2/rev/f39830d92939
changeset:   315:f39830d92939
parent:      314:e73d8ec4a658
parent:      215:14006e536aff
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Mon Jun 21 19:21:55 2010 +0200
summary:     merged upstream
files:       docs/source/index.rst, docs/source/metadata.rst

diff --git a/docs/source/_static/depgraph_big.png b/docs/source/_static/depgraph_big.png
new file mode 100644
index 0000000000000000000000000000000000000000..8b8a27ccc4f695a149fc67f721f40e2798f8a8eb
GIT binary patch
literal 181858
zc$|EEcRbep_x~lLvPX7=hE+yF*(#M}W$#f|X7)~2l9fb}Eqi5V&y=#0dD*fuBQk&I
z<?g;e- at ks>t?oy4d0+4M>vhg^oO51+6y<LcpP)a1!C;7Q- at 2}h!H}9`FgRSt at ZoRd
z-)i!}|L~3O+`NuCK>sCHr$=Hir!lv$OR2cT{lvO at +<P!0zR!h?6|}18?pAhUPdi4G
z&8nzE8?Gv&C`cpumZu!EeeA at n=sf#HBKD<cT(O~SpGc0pQcb*3`pVOc5SysoEubA)
zTr07Kow;(mr<--reB7;8ZzgJLYAQ~|Ja%fhI)5i6D<>xhkNp4p>U|d2O?;a5|M#+x
zrR_fV|9wXmZa{I+)BoRVzV6N({_u&`*4D(ng`~a(L%~=jEv*xQjX0Qsp+#O=qVCd~
zni|FI-tNW3#M4__Tc?GEX|-~x;fpgWt3Cnu at Me{+6>4zv#<q1XY_Rk2M7O_QcN62}
z<cur5OiKipkzc5)s!HrKi~H{C;({)Kfj^8bECi!V?(8&nbmTK75dZJlvDQDGL>#W3
zl9iQJgqum}db+mu2=lnt!vFLA7o|~QM^79(_PDlILRwlnvDYHa%j&-ssmf8~Lf at sV
zd?fNS-dp*=)>c_eAO8cvSYpiB;$lc$oh0T!=$%ro8rzvOcq}X|D;pcCKd+FHEA1KH
zs-$<SJf*3rS?;pR`pDisA~Mo<ch}>-sp({E6j%A9=@W2eSfw<{-)$6BRDNS)27dnj
zEa%SQD=8_-N9SslNLM;7#x;ywKYS5?HYz+VBt-M!!v}V5?(4U2AAc4c{Pxw07Zg+{
zPfDft)SC%3Sd>Vo_i%LU<*E_t>+9p;;YnuqvT}11B_$;tJAPcs%q*jaf{7_~cGgV5
zW+V>B=D&?i9+Fd1B5!{y68QW%5!@l$`SS$P7j>nrtuL<q`lY0-JfTuB-m$W>B6;i9
zv4)0*&K%84#g+pUe0+RQgMy at VbQlMV&3#EuUA4vV{9SY<F%C}7hPj^fh}W-!e*UyX
z&)uDNr`+QY|A`YPQVo8$ojx)Y5fSk;B;*Eszmc&qIRiuR_grl at ad9R?LqnlyjqBIH
z*{Vbt&4&)X{CvORbL4rupZ$DsSM2d;Iucs^32DM3B3RmNv(>V at s$A{O!@V3F1X$VF
z$m!{y!CIy_kB&0?`1q{+{*6DT(?*Q>YI2F2I at qByH5JdPQa7vV!jn<AYNi`EZhUHM
zyP>BSPesAQ$@zkbfs)eq_u|l<FQ=S`ee!fGSvfhgeDuT)m#N~UCO7r{`}a??vEkD2
z=v|kW4|T9+eKK(N>{<D+<3vO`g6oVF6h}(5h++B71#gvTh5U_jcJ^C?fx^)gO_Mr9
zrh>6=zehz!H<}3?!N}gZV_2eP$V+R$|2kEZ`^E+Z`DtewP982U`#~GDVIK_S<cQt~
z*@OyO4JJA+d^f{ZT=pHl{@?3vIIK-q%1yhePvoiT>661B>~AS3+?0~S!PK?15Oj5Q
znRO)})hae)IjPZUQKG!TrbIk6I;ua<^-5h1fWzTUmCCw)Fy}BlMCY6XJ13{Ko*vUh
z{hH^nX&huEnE3ek?$cQLw^x(yL|>fk&8E5PFpD824a4{9{Wt61oC|J#V%!qSa`x=e
zxw$!E_uo19^uzP>^QmI2;lt71($R5j;0B#NeHtSrCAG4?jyERyQR%r7UfxMkNNQgC
z5M_yQjXb^TXDC!2JYa!H9~c-Y(8|R;e){y7!_qL_>gwt(Il1d$Ok$nY?#}WPH?Mzt
zidAE=%I<yE7)Ux<=YzX7875&n(R4gc!kdXC5Er6;wlfJAZuc>4Xo_y7li6sE*kGv@
zVXj)XL(f_2;G-xzZO4y5;Gg$z?MRRa`}$Q5BdMZ7RpGc0V2!OvvKg(J?aOCwXl#tK
z5<ZOU%bX at fbf)cb5%}?+uMN1mx^_E36f5_JW at NCZ-yzo{AtenA3k!SF@#6;v56?|Q
z!z2NvSG&r7-|QbfvfY?ho#ShZ6}q|bJvVL4DUjq;czF2Ztt}UTEv_xg|KaE at 1R73@
zgZ{n*q^z*2>GE%%V6AV$on3WadLDX8B-wtZ9X;K|<Rm#IC2o{#nudF>8rMa=s;8Gs
z+sD4Xz9j58Pt_hT7JM&V9<YIxo&A`Y+qz^&yjWMQw>J@^u;Lr4sK1{!>l7Uw9e7H_
zX$)3T*kg-VK3?>AY+JymPq)sTJI5+6u6N>EOG`_N>#rZrN`-`k)aMCg6%<-tUos7U
z^(r7+J-=aSh{1W-j=*hmDWr0#f^ZY-%uMJ(L4KLF%;DD1eSSmu*#1gqp5IH at VA!li
zi;`prW#y-u+;(<$vO=2*J#!AiUc18bal$?jN8|m4#=@>^X8^Y0`Oy8YFOM?&;uA~D
z$vy9GJ9`LO_y%P|5%KY%k>?dy#l%j*3N>WjOSfH}xS^t=QU}RzHB{<&g09TG2OnW6
zP%s%8S>pOVvzO0h*kGMzsjO+DX)fvj4 at 70fq6j;F{P?X~x9U*h`}yIyxVm!eQT=aA
ztj+!X{kJyy^b750<bMv95Yo}nAyi!+sqV^F=UM+bK#3OE(lV#}ZZ2hDqYo at R%jL at _
zIXF0O-M!lkDT;#;bN}5`y*U!>zFbY?JnAV_Rb743`oh16*$8Q{wOBS<VBA9Tv%e4*
zQt22G(e- at 2>eT(jk!tsp8^P4FLM8GSbu0W{zNBIl_n<`=zkORuQ<IK at lG1L}b8~#e
zeYsg|a~QWoE7##(&{J#eGM!sH6V1W>Rj!u+Lj0zt?!SzVPJvve<u$-8)$CZKwFQ`v
z6)LKy4Uu8I>Tv5s7PF`e*}2H$*b2wS0>egLqo!l)OIQYg0$HJdf at c2=JRyp-PZ4Zz
z2lWwbH*Eo(W7<q0lo5o(Vk(=JLzE)G>L9S at hxY@yq^6~fug~>#PXjX;7HT8{c$Yj>
zU!U*2ll6##nVA$`X|eMEq1(6A2d1XRi-RQ^W!9&hmPZh<t}pZ{&=(9=yKid0=PAv}
zA*ys<{s<|I_9af#l>)Q`o6Pe%h at sWK+WnLJ`}?v&k^e^eJI)Iinm>jxtp4hdF#huD
z0_cmI3F4l%Xz6NpmadJ~c-nk;!U+UVK+5#8VK$JIIUp>IG=P{PpmuN3hL+C=Pk=t8
zyX_2XMeZ;K%i^z at 4&&+8sIAo&X1ncmbGT&#NK0CA4`F;#rstiVifOU^X8ns!KSyhs
zAp{#<UC>53Q&7PBoJov6^4wi%W##S*e>&I?V_G|+*P?m6F)$)7?m27}+TjNe9-ts7
zv&P0r*TF&<(`p+6xIy4bT3BQiGm5!U$)u12(ajE(@!*p%K21$MgZ4^XTwGS@^?%Wu
z(Z$8Z)i~dqz1k);I<`6Lebda0T_%LypU?QC!KZMRm0!Q^RRj at V1l$E|Y8)(i2=_n)
ztNonr5<Wz7bE1 at QqH^X-Yu3Nf+C`k1kx^G_HMG)9qi^u*N1I}%3N8QrV;@6L1w*V~
z%kJgGHsPx(D?iy-=$o8L at cszzg)8uSY}tygHc`@wy6|2y0$r2%IiQ%3kg(8g!y-rL
z0|j8BbTIV=Uc&~w=ch!dtKBvx0&iV-U}9p_7(lF1?RMGgV9&X!scCFHkolu=m{>ql
z)Ah92n}^ZD4+}#;!dPchyBC11be at FA3>>hO@$eAEC!&1}z;q|8^{b>)&zP;SEo at fm
zyLS at +c>UGxLJ&R#-rnBBt}}0M$;vjs!`jZgbzU=-{Vx*FM80_Oq(kC>@!|IzpT_aJ
za=S_12dz|R&Ys0SI{9M%d2%xA?%Efgt at -Ty0`CK_VyhwQo^*wlZ<+Ui`!XGV{ksts
zXC)+<^}YU_t=V0>qf_B<d~?{I^xfUqkF8z&X=!O1wO(R1``Zt9e-GK14ZJ`5P9Y%_
zQYA&R$V65s?CI>Q-YMz9me5m;fy`b{qqx)?`ub=lKZZ0;eU3ypiC=V(da%D)t5ITc
zQY}v>sK~U#2teKC$7 at qwldrS^IEVK at mE1E|U^OIY*<XO8U1}+1XU7LP8}ag`pARkp
zdjHeVP_>HJ4wTc=(=1o6(4hod!dBY-_)6y&5HPX7v)uBOoQ>CERuPYYfD0 at 4KZkI`
za<t}@P4x!#i^#~;=-S<*J at 4)|rrwV1x6vX5(yXV^_wZS+Jz#>Uj=Ac{spm3&wAi9o
zcJAH3TKz*5$i>YOchWD>JU0~;$(X!%?c@@~PeJ70t1u54{PSyObvjyqBG#tHr{16r
zmUnd`n1}V^MR{}$?4vhwx7;BZU3M1B3Qaqh9Ok- at c8LBWI(PRC9zw((>)Rlz8{R+o
z0<eaU!f3!Wp&4R)Y<W4~q30i<cU8~VYbKT0eO&Ue&j{dsWo->b1n;A%ThD3vj$F_x
zR;&p6_87r%2!jxc%UkmD0a)vn$2hpWPCqrCGfObRqsR(H{4f1Rtgsa)0W4aWy?1?B
zt_6&xMe9N8zvynWc)?9gShGFPRRxcY<ncMVn8!jRF224FLI39Nr_Z13pByDDcr-1G
z?%8{P#+%o1US;dg9}aBsxn`q6gFkE2F2K4Gz~dbE-i-lGncAIZO^(gz3M}3P){WNp
zqTBhs*a&dV=l<K0h+g2<oC#KtpPzpP5^18SL;RYpEid&&ou}qK at 9Kw!wEu1iRvGRN
z0_E{?%?`;>sg<u*?#I#n-$KvnFP{Ky@&iv54+FH-U*jnvXg&P&V0T7>_qAO4!Tt`P
zdG{IgF7SK*cO_=uT?Jth at 4fG82`gg!@p(O0e$^9*E-q}2HfWp{wD8`$lg#LuLG1I|
zjELgl<8xvK{)Sk^UCmrIv{o9$X5`7q$pr at WN95iJ9M{v+lNHLnT{K{+QD}5j&wcSS
z6*aZd{_dZcOQs|q9v&~P>B*HM`^>Hl0&%-+&!rc-{xTT`Cf=HnI0zE=+-Aq-45 at i}
ziH~^vy6f?K!1QTwa3eStJ^-w=*c*pzH1$@*nFQij9CGV;<Y&L6Yk@|sFFA;YyB%M*
znQ-*idhR%Z?J}5X3W^bP7X%bd|6k?<X;Oh$RoC}C8BD_?1O93y-+M1Os%YW1tLqgY
zh5*=gT%fD1-CrGiK)HWE6<%epJ$HVO1IOHi`}@%r8+gV32S^-!*g|FSHPk$M&xJ-k
zD8Y_x0dCm!={j>@xmlCXwOyOTFMlM=sPX>(@d}5zCpkIXNLYwC{WOr&HUyu!HReld
zyEJ?SsTV|=kgAG`iiv+S;Y}(C7K8QKuGw#yl%+PKp?!Mp#30KNNCAhoJQmcA2H*TK
zj5!PvfmK9=A-`(v1bQPV)CnH~raDk9QU6f}B8 at As`P;u2gTo~DX_-CO$(8TjqwCDO
z32f!E|7XT-f7g|WNt8B4CZqw95<C&wwZASNALO8(>-0&0TyWYrNbv3c`32ui0BfdE
zV1Uyecf}X9G-zA0{OaHIj3j~GciC7i-qWj|T?AY4oI!}hdv}%KM at K>!DB1d!mYZ`X
z5e$fx(Ps(JukLM*8uu3%$_gDOUdC-m>yN=aHBUioA(3SN<7<6}Qp(tP*X)nZ<Z~Ix
z$wv at ZzBqfQz-Ck&R1E<T*TmS3|5a4@)$;X1Afb}9%dEEmHoHo#w9{f8hW>2U9)#2$
z>@7ci_N)O`va7=JVOs3JR8y8QF=2#w_hu+#JxI4zc(?A{IjkTj7uPBT!K6Qpo+OBt
z0;feSKR-YA!>vy5O2v>c3O|k)b8ooMKR(-);xZkjPTOhS6hO)xT3jrE^ak6_(54}q
z+7OwSo97R2(BYy^`Qt+47NgEYsbhqMjS$cU0GsgP0XS9_=Axpa2+)3FD`^lfgZalN
zruT#QPc--wB0l;L-((=9KSK2IKK#!8bD$^?$I2hAcS{&^piD}=t~1tOVt?z}+_`@P
z;Q~8E%ihlBO)#6v%F1w`2AokRKu}(nl|6ynCk`(D>GS7dhrn`O31L0H(nu;X7Vo|!
zb+Erz2n2iP%o%p<nXIEBGQQJaq9<YNTHu~m!2>pq+oB~wdP=6*q-Y%ILT0A at t=;dR
znwA6ak0QbYP>-^v$4-D>xv*FBR?M9m68j?{tU=Mh{~`_~`AI@*?mDp4tDny&jDd3U
z1EbV^w<5Rs<42>uJY8gmU%Yr>JKM?ff3`0if^FDy6T7mx*;3=V6La>SqLC5n at 8wZa
z9v&W0ucVMdD8 at 2wE2(no?7IMgxUAF at tTqtxjD-k|JeEyUet*Tm7vydTD5*&#bj~Mu
zG$gPc-qU07sB$s9IOgW&2m#~W<~c|>*=~TwM4%NTWWxkjU=>#DG$KWl|CPaTFyeov
zC`EsLy|mTgz3&T`=E5?*uoGr<nFu;xU^gklC~S*4<k{Q*s_&-Pa_5jHI(Fhj1HevF
zm)T2e3LJKj(49%wK)$26&*-&4&;IQUyVh-YSXh^>@jyf|yT1|+ayAX}Wz&pY%HvT|
zqGSO{7UoY#!--`)%#dcQ!Q$f})dHK1q+sJs+T#Wiajfk2?+vAihw3sz#v^7#kN^*`
z^82?8kW_{9a;Ry%Ye1A}*7)?aKe$ljkU>$*j#P^X+Kl8l6#Wl!3*0xYML}X|lv-+5
z;}`4hG*?$ubym6Bqv(YE`oOVMbJEq#Z36tjO at L%xw{^3ZQBfS&GymrB#5=jyM*!;+
zy!SO%A3GV(_L&n|VIC(Yo<@3)mzKz7z3XO_H3^VQlEzu9l&+_Fx|K#NWA$58kqY!v
zF%!6V#wW(hijh<R at a->oNP}EgTI at Bn#Hmj~5RiipXec)CDR7um22Q{1!Z1H*=zOUo
za(-bryP$v!rER at 0!NcLu8IQ|F8r(?Etzex;4V#~xq!qG`>%RMcDtRw^EpDI(0l8Q8
zt6gNnE$7X4TXl6cWGx91E$^7Z;*Q!<!uN%Z@`b!gUW;$+RU6;6WrYI&t>t8ZGarw*
z$y5(m(3yGfd|GS|`?e$fHHInqc+nuQy=}vd`Cc$M<=t&8tQPB3*{NdQ`yO!HhI8F%
zvX_14>yb8u0FZlo^(1xhjkzCWhjo*3*k)rOOO0|{7V*9HdpC{@)|Iqh?MoBfysUQe
z<VjiyFHsUE(Wek+lRXL&H%v^<wSM`cq^1 at C;N`Kq+5!eAzx%G$`tI&7@*HFEs|jRp
zA;jSSk`K}^^X#2h7>u6x-UhzNyIGpDp+!_n2q2<665AF4u_vg(@Fsa|%mAf}Muh`M
zd%Re-gZ;licKp2tH&P6Go*TU~<ek>s^h3h!Aoa$2)%BYJ*$hjx4mUd`{npvV^>qh_
zh3~{5tgZnb)Bh|Fu`-F!j)GuD><(Bo3E-nvapj*1u1tO(8y_Ee%FVEkpx!ea?6b@=
z?31^J+d(Jr0n7?K{7wWRDJ3WO_3qH-qUFcubihWq-QC?}jEp(|-6CixL1h&cKeb$~
z at yg|zMqE-+#pRalyyflIC at vtBCtA4yU1n985ZHGCC&0$n?yd!Q-(~q5A?JQT=rjYO
zJOOvmfTZ3JOnLRT<9vfj5$rjM%hHD#1mh?|zh!VJ-`+v*4`;bXFOYb_l$RDMPtg6>
zQo6=~;w4Q?-t`C~Y59`l9vK4zp^c4=pTEB$cuEk3?ooT0Dx4WAoB&T%Ypq;QgMxg(
zPJ;yW%dg$9X#E}XV05Foql46bro9=6RM_~$A=j%>b89>{X?t at RQ$lI-O|CbsmMIrn
zTU%Fv3{M#v8iFX#TI(N-bB~ISo?I*&l>vvGsFt0imfe-BEmXC$R9P@!+3m!oo`;92
zf=m at +?9>BY2xy(yYcamC5bU%#I1VA?r<d<MxFpj)FkDUTRCx+1vyON%Ksr?4N3mE>
zz}y6K#(MV3uqCs{3Jxn9+bWnmPAt`bNmtRf3wsSVy$R&?Nvptz%v*o~V0Ot)ogyw8
z(0Skvy05=q1J>c^@Q0$oz3-(5+QmtwlfEyP+~dIq0ct(ZR?q*wQ~2)=e_uoS4e3Hm
z-B~)c=yI^P*;Qn!07_UoSB;H><2Zw$rER5^)g$eaxnHip9!biX;W06={U_!MA)9W=
z$@$5qHC!VVJCmDlO=N1-BKn;GwgxF>ea{WmYUbdl8JQ1PWCzu<Pr{0C!2|HQu1y7q
z7O(qU22{w=uN|4IftzNO at M26#N|GO{7-~U*>b5 at 1x$~C)!HF6W`J7>YK#=fR50`Te
zaaAvGr}iy$<!D~oUGGt#<ukhO%XnCLxdj5()vV_ovDcqz8X`LW>vKQKRk6>*F1O*$
zE!E7o=QatJh;6>t%0&f)-#-T?ORIhHi5p6-hR`N}GM)e(F6_L-I9O(*zjJ=?pzPt=
z)MxOxQA0Z-&P&h1^q*#D|CG6xXK~fkep5>>RxtM?N$W at cUq9d)sTj}u_frvC{fCj)
zK|}!mh2h6;eJiOQ8n+#T%yt>~rzv>wg&IH-1CAFTgQyyn6Tl+`&Lnt`m`i+lb)gZg
zQm|gN+xW_OqxGh4^uEoX$%mK3Z_U-rz}gp^bsgcUUZ;rn+MaV0qyAUAUWFX)OuIva
znir=;oN_iN2P_4_OC}E{N9RqRdf3BrZ%8}=R3E6C#PV{D6Yp^Qr02uLx7i`uP?2+G
z#Wr2=sbH_;S7XoH3JNztPoPVFu%AJnqIB>3SX6Nslh_I0dhnq50_#a<Q61i4mMlk&
z3xO)?tAOCLsog6rZUgI!o^#gWDrDu<^z?X2n7H(V2WR*Hba+=_fx_Ka7xHTxj0#Wl
z at F;AP?Sn^%D}89WlSf)#Utc~MDw680_s?_9!*v6J35T6WJt0l&8AY6kU_<<REgVH=
z7u;s_?nGZ~21QCnLGh$FTfNbMx6Xjqpf^jk)1m}bI#cUilUZ--=;%aEKGi%k9Xz}?
z{iV>Nm$TTS_XMZ~U-|TFV2!{dbmi+ypza1pu$QH^5x+Q{^PA^;vvDw}*0A>L*SM{)
zB0wTBgP=gx>4UL)Uve_C$B41#-jV%hKCHkw90f;Xxib08>po~PRP+To=u=iE1a#qO
z9t-Sn1T*3~bBZ=Z#=N at MqB;3SFa`q~s$9~Nap#~+$Lb>i2X^!4__$B)?ljM>+qW|}
zjsLxTE*o4P2M33kq7y%FBOO&j=tJ=JJywsNAS49Zup6tx0b%{F^>@i!bf#5;LXk<^
z3YY?13~-HVzoj62f$u%QU_8K`DoYN%tq66^;Q?hDpp|PPK}O!?DByB- at g$d;!C<j@
zfmSZDs|MyksOS^Kc-%fmq2q!Y5rZJn+pCVK5RSpXMl|g#*M@<60VmvGU3O<^=g*&d
zOF{L4W!RI>T5oT2PkDI~BV%KLJ4G!m&RiSroje=L^YVTJmLDP48GgnbGMDUYK`Z+A
zMMZ^HLxO3s%DJ(+w^5OWR?t$TB<#b;2X&Cbb`wp<b2N+aNyXN~d&Gg8{BK8|L%v7Y
zZJm90Zx7U>fne<T%*@Phmq%;Ae=iOpOVNNtMy0b&`eM at uRUi>N at S+$Yr}iI9B?vG0
z`T0?msi~y}5htYi^-8D3tYUTZYXL-3kc_JU;Ozegh%L^5>FbH1N=i*_nk%h0c&wKn
z*f{?9^Zh4{;}{GT4b7dsOu)P8>1lQzp1{xdrS0ta#CLvPI&=QK at mQVD#=(8X5}tvc
z9!l6x)b|1WS3k=;lC92@`tF at kSMrS*5$B7K53-Z7AdDa$@9a1%-pPl+*j+85WBx|F
z9dFtm7X+%{%y|#?wt|0W0nQ8c^Zi at PG2a<I*waTVm9CIr9X<(aix^x|%d?YQyyo4h
z7G>FYYUOlyJY=5J`rM5bynVw6u*6(0PM931%wY(>t4yVjFKh#LSUyJBE?FRxg`8Zq
zcZ_8b)h2mrwpbmm>Rw}_t~>BF1M$AKvvX(R?m5aMM~<Y{|CZQY!96DzGbWgLx}1<5
z!p+6i^@G^E{u;i;${MlNCQ4L(Dz(NA8nhhBU*%Td55Lw{SwwG;-&#T<sa68?z_Hn$
zUpLyTudh_(<Pbr~%kQvrarw%m_|(=)$ep9qDzWg-$>9dHd-w6R0JcQv^l1#pkofp`
z`At<dHQ7Fpqe^;uUX3%BOM1amQ}^ZKMd_|wxgx&+A at K}AiaH`9qL(QpNo;G(cV&IO
ziBw`Y(c#{I`e)w1-oW-vIC=SZ(>bmCbrvPyc(3FZ9S#`CFMvzNhI)1_Y-D6)*lMY(
zso}@A{U~m?6l^pq6h5WSIAl_Edk4)G;QL%<B;mm5Lgee_=j8Z7qzqWf+_>Ri%!EO$
zV(p(j)@35q)KO7Us!V-~p-f`Wtg#iBKNVWc6FSdC>wDG5 at EMa)QwM+}gOyCO8Y=zN
z*48XE>fw8n*|T`ivoGyHdw*;2IjIGaX9seU at 9xH?;<Ws$RQoOg7BPytkkuaSyZbUK
z<f=8oO&ai?=H)FIT%7Obxz{J25EsV?Fr`U<GZP5@)vJKc&QAKUd(jXAi|r!I*GFqS
z3-5n^;U<AtVA^r-u615P0q at IKuOQ#m|EySWqe7U9n%V at abAEk&iJXRpKte(S_I_(?
zOJ%2gP&DCV`SLc>li}gv_!1WpltoY^r=&!@d6VUE$(pkdCpd5TXz{|Ax6UI$u&b8+
z1xX*nOGT3Jy?=)jVDxw at Bwqc4`vnkI;jdo at fu)}WW<(t?fTaS!B+)-pks4*zx^Lre
zdbYN;-GDoevc~^!E`_acrJ&2qfxc;mUfxOlhGu%tm7_YkjV<6Dvf_x!Nrfu28 at QK)
z9f)~`ECs=t?PYojoB`}kmx67HT3J~MU!w6bi;_L7dr+jpdD6qfBXx57&!5|yf8e(=
zPfySB at o^kdQc{!&#!6LHRbvYagv`v$E^cnJL+&mvrv(I525k%sPIGc{j*XAA5Fc^e
zJC&7@@wm;z7oq{^hmnNfxlGD>C2G<O#)2Wi2ag9PL}G7u;Y)^6ifzj&=dq?BN~F;M
zNW)`egUic>A!7^mLN#Mor9zUHube+mVA>Jy9~ejg_XuC2Vq$P#EPG6UwzlfNd?B<b
zp at -}M1?dOs_?yLLl&@dv>LfvDH-b8bbU^0U!9f6nMhV<<fUy8m0Zj??^+)mdn2-Kl
z&OKK?KE6+1zTB&z>&*&#u5D(96a;FtfiTNb&8Ekll-B(y`n_y;Sf}aZN5M}ZINxO9
zE8KRT>plwWdF#%dhXt=~br-RfG=Q-Y*2 at D$1eneW$A_4 at Kfh*xZ$UxyNaVeUj67H`
zjGeEsY5E*_o(~+sjnKP?C4)`~gq6AUga!j%$&@a3Y~0o3L`2s$HR&EbdPGJ^$(Ac&
zJ64x4xD4=tI`cp_sCk53Y>%;!_3$x_^68zfo6OB`E<Y?V>taPcAgjUROh*ya`2O}y
z1$Fp>f`TqvyBBq2!*=r8j!B~80wz2n!bi|*aI8<?yE<vFmjI*?y!~seki!2oKyUUt
z3TY?xBu%}`bQxMW6;)M?GAT+Gx=Kno)q2a at 6vGazyc{m>?jpqeDw!&1Tp||G<Spk{
zlrIoZJ`VyLaiV08f{+I31Zf49iG`O}(X9qp1x$&pFr#v_Is7Xq{W;S~5U=1fHMpta
z&Ha|840)U>YT4t9i^LF*u(P=3)K0%QHv=IminVg7w5U6D at nomNB5Kt@XZ4EXm#cBy
zl%wbHjtdJ5Ti*r=-CMykXuz8YnET+t1NI9S{Ppsqt)ru&{-Ha){F~<H9AJ at 951f_=
zAM_pMOYhcIixMsH4YI*%+5R#qQZ6pSM{se;|Cu<N+aD-0W#i(?mX`<%NL9{6#dOrZ
z02}nmi-f0Y<;i+cmb~tEt&Lx|X9pb}-RwZoc~Bp?!Of^9^8P*loac9=!aDc@^O%uw
z46Fwz6XnZWv8pAyehtk{xSUZZKn5`}F+p^{Y#3%GyZ8zMJH1vx$xj;IFq*ivL+R~~
zqW{E+M{YK3e>M7GP;l at yd3kx?lY2k=3)#Sd1#B$zjTRr at oFAOI<h-PVraVTyb}l@!
zw_okP!(&+g<fzNcTW1SVc@*I1!RteeqvlefAQ}1Pqr^<zOPxu$Xih0tyNr{%JN}CO
zIn*Ls!d6a(N$l4FIO1T~ZieDvAdJzZ6_S?m?n_qyIgDFEPr1xzrtcrd{KEY32YWO(
z5<txGeXQkhW=^>QP!CxjZv7fY5O&HHR~)-b9;&0hFRB|8FneHf)q*{Dej9GD&v{$a
z_K9r15w>GR6AuTQ2m8#JF_3};wz)-*!1i|wVvz1Z?URMReAJ*`S^L6c-1L<EX>hPF
z_`0+^(MZCVEdm4KzmtFaY`igWa-&bbB~i*3na;pQA5`lfsb<jk-n&e~C~SMLBO5hA
z!S++!zJ2>?Xs8SX3-V(aTl(4T;^tr)G{WZN<0JfN>KKHhfMRauU17<;ROKD|&`eX$
zeBkmr7wPfJfM>eCU)RQp9x7_-VFxaoag&~_#D9(T56 at TUs1PqFoDT|5051+Mn+yAj
zzzRqB=+S0z?GlS=6;Dm3!J~?4&vK5Dw3?S_Wgg)lV?5#{%m`HdI69i<tH~uWEI*w@
z03Sgyr*VF&D>>~3Hwxfv!2cQ?bz;w4>L|@>5xImmxin2JTl7WUEPGWNa=llzHLU$P
zVB04r*{bo^3f>>nvJjxpe3G9*c8ZK_Y+WmO at 3a1$)uQY>n2MUGed_5wNyWt~T15D`
zvUl#F3AbWBrn*;w?{#qx9!)q>@A(=XCn1rswB#0`4=HZP3Q9>ybuTVd>zNeMU42Ue
zUvAY*@H^_>qMGBU@>he~L!<iv@^7!A0S8Q8NL%LJQCCQv1=+b8)3@@0z67Kbxn(2L
z`Fho4OiZCBZEtRSdzYMT7{2;e-ffhukC2d1NlWV)xWJe;6RMLZ>n%#8^ZOdj1Y~>r
zZ_w)Oi>qm9xL}QgpRFaIC<m<)7}!Frnt0;Gi5DZWRn^sLo2U9V+5hZ>xp-4At4{1`
z>O9++P_7aRy0g0fw~zGg<*<E+m|~_%f1xo!0Ldw><(615 at +i0HE}4LiGFqSQLS+op
zq%0ftVtCk>m-|b6p?jvtxP`=HYdr9loE)mTfp`m55dVXRivD|>qhYD3sTTi;67!k%
zw?N8u5Qp;{!q>0geVe816ETi*2~~3A#N11zM4mTGFy9Pi3<A3*jHO{<2%c;SV-pad
z%CFhx1iezi8RiY2RcO*ik7+;nMz-U)sa!GA7xP;;VSmPZpWyT7&)~ew!zce4$ig4>
z+&l%V=?_==XGA9b*?c?C{@${;@$S~DYN_{B#7$H#RdtoSJHxTIlkdFrphSz$WyRnp
zwz6;03m6M^!_8MR>kaDsj#Czx_q?-6FqV|Wh;PrbiEhuP$S)*#{?_EyF3lNi37Wab
z47wG`nysHj9Tt`f*UBB{D6TrrPYLi2swygmQw@@kkOau2L|6y63QsuQx^)cb9Mz_}
z%*YrSxv?}<R4NZ1^Z#obC7Eu$A_$-MF?Ovhe^+p6OvnF&ctd+$RDv$ojZvaJmiw*=
zj~{NSGBG+4(|_^(oF`?H_=Aie1AJ6JI;vl>c+q8no|DQ)c-$>2*+y{H{=$U|H$b*P
zSf`Cfym&z|=jB_E$?kpD-QCTZWVU+gq=Yj>BPtZVzj(5=xkIw*A}cK|1qJJ#TvNy0
zDnZs5c0G~4#Nyo<Zm)m|)0u5IA=Nw_3-TOloCkldlIMJ%bC6${@@{;@T9DamktSZ2
zn0VSufZj~tjYYO*@X(sY_qETsc{f#hsI8*(qw^zq);5;Fb%1kv#jW|Is!G%^AYf!=
zL07wPVJNR1`JM- at RG2yjn<^iW82Zkmj*AC*U$#!C?Zij(=!br=pGo#U*ar#uws at qa
zg#dIDSO7Gn2e6xZmXlMqB44w}<mvP04%>0%15GpO^68Ny%o}di>rCb#*$OpMj}wEO
zU-x=gsDX+EKw8M&GxT*lbVed-dip5nNnriPFJJD^Gt5j+`@`aJ5G~Ek9S52Nx({xC
z0)GSFl>S*J?`GBc;7#S#*VM3LcOS7_Y<4eQa4#!=8Q5 at ROoxfddC&6v->M)vo6K_n
zP`sM`?FBc9S8XO}ma4#hM!syb@^{I%clz&2J;h~G$aO0mq`-%xE<1b+2J_cEvs}E0
z`e)JSG93B>218wQ6v9N;j>*c(W<AF}%gSo{%&&OEO$>=B)ZFh7+tl$rrS4%ix8e3R
zUJRu!ih`xeWpQrJ_jtg0__kJp6R5?Ercd&8K0Kkmpk=T&^*KhuyJnp<i-A~K=$&U@
z^skwY6!_*X&~T}mhW|`N5P&p9Gfk)l;JiG-=_?bZ*7K(sq*jRYh%3PXq#F6jlm0Et
z-i(Aa7hmF+jNH1l<2IvLS0eufsLH4}OBI#xK==-p*@VHf8Gg^vl<)OH1)H%Uo7(22
zYk_Du(#6Fk5hxOU6iR6L^}(vI{jJC<+|0!0sCP?mwmRzhfX*{OV=e1lH at hmGmt)=9
zomR%AJpN2aqxQ;0{Tej6<S_s3c;tD7k39+!xh@)-`F$x6MyRO|Y|l<CNEvktWPQ(L
z#gwjN)p|U{a+l~s*f}^N@`AHOWl{=1`dD~*dG+GgG&}OSsOWVK>5ggVs)?vR|Eo{l
z<qG1My4GpvZD=xyI_A!ElZt)0*1o1#B-&&NQCBvEBdl)pDzNA}`)?E5+^5|6p#x-w
zrmj(@eOaA(ET*pP74MuzTc=novU^YX){hBZ*Nxr5`WNKi^qan3CnzjjKB!o{^_Hww
zh&A;*=|a#b+iK-GuOE56wKg=Yk{3vof)fiL7HFvyIq9BHz>|`%`jLF3L{#S#BjZo1
zitlCu-5cTIWa*AVgUo8m%F0#+6%~#Xe5_;3t&l;7^Y&_PBzeIe)#Vp2?(OY$rQM-H
zB?j2Ms-ziJ8X@{_afey5?ca-pAu_&qqAy;lR`_1LT(R!m_rN3kg=dFT<tc2Xa|`fD
zAt**FYU<>~L<~4Cc6Rn~YP-sU0y6bHoqE`<w2ntoxoT{pqKsW#UCNI|`1vW?-dv7s
zz_NCpVqzL$ABk*oCUUJU*UJz5eE-q>HKEsX-JYPyPXvA0>2tjLmg+=)sBD_Fq9PeM
zuCbY!^VOs7Zf*^Ied-lCVDGm{42lNEAR^Y*9)adIZ=fj}AOel<TYuWK_Jwyx7I7E8
z9f!D3Q`&CD?xov0&z8?jyszM#$m?wk`desUS;((FoK*lNOioF8Z at Hytz!G^mG!B5q
zsaMw5(F`W9X*=U{#w!$(*zE_82RHtp9_+*KIa0M=d#DXMR|q*W+mlWT%Z-NDa<oda
zJ_>#_`D*M*e$)WfN$Y(H`iATvn~p-};*6yycr~E;HE=+QcVh)*1$^dx8pkD-l_`vj
zjnOpR-wvS)4``HU&z>RZg4~>J at F)D#+A8hjBnXZQS?;YZ7fWnKiBy(1S+A<0A<O=D
zUu|-6F;!+}rjVyJ4xhn at qwSk|16%sXYWK#5E_3Uh;&T~xw9Q^Pt=8A<IQGz(cI=<=
zIThw9LnCt8R|IdTS?-gQMG>S<tG-%b8~drnvE31_e*INo(DXQyYt(($3{578?R(Rz
zp3^=LTqQM`QeUZSz6#9Hz1EX=GBMLjgZmZ3d~`U$iq5s~eQaLL5az4ysX!KzgCWD-
zm3ZSeFSiChWKVEYljpn&%(_vaCH}!Tdk{aPnRPpRYQaY4s1pztC?SwRN#*7Ji>%vu
z!NI|A^$485SvZzR%@6wJPDct4XPgpoLh~tLTkYSB74f|m_q_TFIAbQkn`yavGrqe`
z>05W&ox=N{DdNOEE7~dXk+h<sx|Oup&>*c-?RFBvG@=1JQS;=<6V6B<a7qF;BaBK)
zN at gQfSCUgw(uTlKn|;rrGcq!Iqf=taH}`z2iag=YxyajZzFqSOcrABsPJET_?71DX
z1e>*G{iUbNXN)LROSD2D{eYpb-0eqxPeDNe9Zc!zxu+#6Dw^7n+}bJ&p3wXVMRqUe
zDSe$*-!Z}1yW4r}2kjaRYn)?i^W)y{o^5Sp1ycSqhp^`|=0gAyqVUnQY+9`JWz;OS
z6lD0Gr<;xHV)SCo(t{W at ssRFv8uSrFwqh2YLnK&KBk&Vje`XT?wh{ieN*#2K>fTgN
zuT6hJb5 at Y3N-8R6h{NsF`S8jGSH5?2km^=C-GFFn{c>25Utb(5^GP_Vw1K(_flO}4
z>Pn7(bK96tvkkUrDM>C{sI%zvDalg}ZDQqo|6y%oBbWbaX=##8hzy%p0B+i0vRO;4
zooHXHD?E1bOSagT#?P!P-1&MMLv_Vf%Ih1Uu544|My>`Q`0(PURTFiy#qaf~M~@XR
z;Mt_5rFBm>Tk*L*7Qw;6QI65d?v*XOp=wO7V^6FU%$eBhpF>wuYnPnR;%G-x`byn}
zI=HeRp{l#+;aSr at 1IN3nc{YMzh+xN2I29BW$aW55D<Hb&eo4xwbfF56RzpJrs^oQc
zDyGkzqM%SdYk6U$B}&k1*X6#63G0<Be*(Dt`=Y^9p>c>$U%p&(bGu4GLGjZ1g%QcA
ztA44eXR1~|GN8tM9H&noDn5YFNw>g)YnQ)cZG92=c4FNtzsoGO(+NxkI;X<Gz=fqY
zv|bgPRGYdQk at fk0*1e;)O4HgZnEbg*svbg=%df$XDJOi8NwL$S0kxq^3J&^7_x}Dq
zvP)rMVFKjgTY20Uo1G^?=B00dkJuK!2!AxR;~@194ti#BO>b4?^l8a$60Vx%XUiX1
zSXjC at RSR>stWVcm3brkGStf=n)-Zpmz4K3;VM+M;`O%St2VY(xpDHU9IE+LtY6r(%
zb<FCL&$)H|TMRpW_eBnl`Z^z6)JTOK{PsnnG-JGKz55OtC<Fa`?#$V<9RJLCRJ1{)
zmq{(a@|zdLdT_9_OM(H7sw>IBjM)mGme8e$!h_{NXLy>Ln^6%JO`BbH`uQ{_hSt23
z&-TLZ-fT*k(a!HhG}Xxmw?fi+=7uUn_Uc5l3vfPC@@SY4UdN6dlLW8cm2{1usih?a
ze0%reJ2kshLGYAF5aZM=HDH&rI6ef_WM{9i;J?D&x at _}jO#hJ6A1V0pT#wPji<y=(
zFKpXr__Eg8E<_QN_i#f(J-MPt^f5Q-<=~*iEJC?p-VD(xZt2r&EAK_gE<8E2mRMA2
zOwk)PU>Ts7A5CrNwOAqVHH1B5d6~R^?(t*H9HZ7SwU4%)c6Ki-FRx++L>TMQ>}IQq
zKi7{HRNPq4Ehp4iaJbrUS<1OX#qzLkzMa+9qk7sz#VwDxi)PV{;~XX`pAyGXeXYRJ
z#ijh-%Ag}T%g at fvP8rJL{Yzj0ZTobc(NLVk!R{sS&QaDB7~ze%QSXDj{$lgUC%H;$
zGBWtEp}tK`*U at M(swPTHAE|I$_)*Nx=PG;S#!;jQcXvIee*9pSkYK)Y<x2N-6qkDH
zoj*IzwQ^<T<esYpX{*dS<$h~BZkWdx+r~>v#LmTa?e^^;2kc>APcpsdN-kx>&YEKh
zu?GPGfr$Ty8xj>7=cUJ%w)>hVAF|f``oiVK04n;~7~tpC1hVLfiWuSBcn!^_8k(A~
zi*x&zH_nAbP|gsnbvirN+vchf0&wCH5Zp;xe)D8}B2L=N`k#VCYN~?79u-Jl#!_qS
zGbvwuh_x33ApOvsKB>emnS$u5-{E`;1r at P5uB4tG6Dod;jTz)<myr|}7Rm|+T0h2G
zxBMI^S_N#JL}dIaf(_Q~&{Eo=6JpO#<L7=L)cS(kXLPzWj at FL@pGSm|1PH`@#gB}N
z%5Qeotk at b$e?hzSI%M?3!~`mD>efcodw>#f0s7(f-uFN~<O3k&&T98f25-Zs)}37`
zH_><$qp+>y8-YYnN~Rt0&95(+qA^U=A4Y#{4yG|I8%k2n>`cBv=rG%f13C8c$ndGF
z4w5=L!&VY=k3lS*upKp<9Xdt+E?nK30yk_LH#}SWe8Z{31rmxp#`TTrS&in<H>$e4
z;06Ea>8Q6hXLv8<yPQ9g;CNh)@-j860g{oAOF%}72fY at i`M$7W-B^OjQaS%h-<c*Z
zFSFuuuSC&+j`1fgz2-q($A^4{{&8hx&QY at 8ONccdmT27=xs<DxO;u3yX!_YbiIUw4
zQIAFXIzl3%>n<+B2;l%wfpO;eLH)t1$}d<)6*%g?eD!K#wkrjdnL-(_Ji#R(Nd)ak
zD!PKH*`CjqS#<*)hq@`q^A#E$eIsO(=MYI{H{Ng*bvI3l=wSQIhDt90V=L<Eg at QO8
z-&&mn;fcY(c4(cj+??&gt#;dxzJEWpp=0bo=fk6;!17WcnEswG<7m1->KGj4@%Z8B
z at syMl@>8cmzzUrf6olW4CI5FAF(ZThXo6}F8>iEkM(i4|GYa+UoE*7%-)Jy<qwbD6
zxWjA9tP|p{%vHPMh5IIIWdMs2=blrr*xFg2Ip^Iy=U^+$*zL^aIy<DxtD}PNjMcY#
zYES1}Z)+#d_GFy(7!wuMnYV05j~=}(WosA8*6uAWEuA{{&q*_-yYZgCHFGoyk1X`%
zw_u$|a~<rD)h%Hw`P|mcMEAEQHQwJ3It+4tDpCk?H8pCPFs2a5;oTXD1Efn|T2u5Y
z1Basl-Ip(^(D8~CnGofF=KH=K8?Y2KAA3XFf3DUOyqCGHwv-eO2+cUAeC13a!ym<X
zczAU^J(Pf}-I7P&$_GAqa%5_1s?Mkog8`5MvFr<0HAyvFO<w9}f4 at et<-q$9g6*_-
z?<QWG#@CNG2CnuShmD5_jYv8;I8-x59xE5*8+>w<aHUOXG!cBHMyVy8+wY&6lg6NB
z(9yM?bOofimBV)+F}p|KY305;GK7w0$;n;s%hN?;vT8cnY)Y?(9PNlbzi`O~Eu1}j
zR{5yJ6;jeL&icg)zbcy_b(avBZ~tn#V)NiZn#uz;6=Qp1(a{{MsifHo6)(yuDHC!%
z>XV~QiTi_=Y|Q=M)mBqdW`)FEiw$N1$)GmPBM35deZQ~mth-eY8C<Ykw7Kn<1%Ckm
zYu(s4Q7YU~g+1?j at 5w*?V=&~YqXf=<y4l(Q;brv~m2!J at PXdO2U!vx-^M8R3E^jCN
z`t?gmQSmX7nNmN&%5YNpf;?0`uKor0yP=^Ws$OWFKSz at GkO<%R-sv5{B4n81d&%hN
zf|$hI5^shvJq-(M0W%viQtgfaYXXlrL{e7v1nLz3UK}$1@%43ot+xb9D&W0(qe97C
zwG{WwrP=;MwqmUdz<^*F&>^ke&Dw*b7<7IYKuHFy4eG$}PV;!1uS;mJdZ7juh?h+1
zO|b6aZ{LPM%1q77+!@;0-j+RZeij^T&+vnUm{@g%4qMT*$=G*X8GAQ})B4MD&%#D8
ze(VieP4W8Ym1AYQ;(t=lfYW at lp2-qqFYL;z6Swd|4LAQTZ&hmQ^$LYJk!QzC^KSap
zmuOw6s;*AzGP59ZdniC}{)C2{I^Uexi2uSrUHvzVBm_*ZQ{*|imDFIl44G~lKJ3eD
zZVIC0vmO at q*xTp>kEVjc`mRjMF>~C$eY;(VJj1>14K!`|qy6m=yGe7}JUWA+AhtmT
zA696My?W-{InICPJS|R7Pfsg<(jqdT^zFYst3~9(1v16XuCQmax~eJ;zxc+D8&`-Q
z!DrLK-WePoYJZlT%!*E`Ub;j{M at RSR>(`f0f|!_?z&L!Ms-rfvEG#zbN>+-~1_{25
zrH%b}7QhJXFB;N?W#seRcKG_{G9fUCN`<)tVQXt^1Gp7*zJ*mlfGU{!LPH~vSvr9X
z&B>FueXI&o)6!Vj*l_>q9K_#Fi_d`3IjX=*dqu|6eW38o$Rf1sbKu~#Dp$_#EjeLd
zsZGPkm{XUHi1V9mlfl2WKjr+sJE~us#+;Sf<cOZo^d>b+I{&HR<l;)>w9TuZtDrD6
zQEHY`33j4+me(f8x}p<nH5HW}l(B>Yw6g*fsO;7tc4Dbo8ZSo|_ZzFh;zf-Md6x`w
zMOQy4-uqNhv6H>yd5oC&#;se&q_15gr>ECSt8jE=S@@`Rw8v7A4KxO5W91bsIebo1
zqhK!!*YL0Z3|@3ooIE-GO?}Z#-<UK~Q1J56_iU`Jsgh3U<ZqFw0xWC8+^V^`Ijx{2
zEx=^~SY^~9PcoJM63x?M-ys&raxK7Mx;1KLb2ECWXlU`4f<m;{N3g at fb`yBq)PzTl
z96?8jVf&?1x&X~i!xqB7f~xlo3?!JCoJ<>Y0`v at ziRquh--iI^DDa%?KC4sXK{qfk
zQ10>PJ=Kp_2i|hIY9e=8e_s5G)flz6aa7ALX~oOR!?CeTSZK;#D<~83^76XuHKJ!T
z5&6->>ukGE^A#I&^GuaR?Ksjez9SPlnsE&OiavA<5&n&ex|T@^M<=f6JJtFt^*>JP
z96CR*;kDr6;$AsU`_HNR?3|n?kO>8r1N>F{+Y6|ydfwQ=fFuwX#Y=X6%`?p+lW|~+
z=>vkNPv<kb{EB at ZGUwfS#a-XvPlyg6pfcGkwz90djg<!CI_C02YPB2<pFn1>!xI^%
z?q?6jo^ycOE806cqVe-SU+Xu*qGHZ7?I&erWmE;!EDy;mLZ3gs4je1+39|GEU|7M>
z;wOk;e?pqn8+ZkmE?r8iC*q7c(bd&ujoXio;Ogk;L}qPU2*#pA48{ZRA0X at LODg_!
zpvd&saG>?!IqvGsk;V>*gXh2}C^sNoU-pcB*ftQLx10aQfrb%b>%s3gu1>ZnX=oh7
zAa}l0K1qU(dIbdqq4_j$FpA5I$qScgiTa3L)yvuqcvS`xGW0b1=CKnmRNtXFoQq8u
z8$R2i%S2X()w%_$oSKmui)(%{%?7r7pS1B}n#XMUIz`yokHyEwE1!P<Zp^`|l!bMA
zZe#nZe@?Se(LmPEJ9rig6#^xRL%FTiWhogMx3nx6im54`KS5+)fj2oiFX$}W5JhTy
z4`>_@=hUA0 at nd4XHyaI!AZfiY at xHkD+gu@<uH5T{b?V+gO`**of9p_4QPk<)*<5Z}
zu006-_U)Sm?f^Pp42ziiI~GtG?H4*ABC=fN3vBQP_Zowuk`j5m$J%Wqa)143YkZhr
zt(M(!GPdp9(XxB@?x7zM5GQPRL at oOyeF#3fSi1b%#4a<mwm-horM`P7o!@sYr7Kat
z)~oziyU5|e>m at 7$nrU0HT!;p~Q&Uq5F<g58p5NR(S1q`|zn{KwIOIyPdnF}~O;_MW
zYUhQig&Vb%m4<OOH8s|aLq-#C=a(q^Z2S)zX^HU78?>CIZ20KjRJE4H(B6%k{IWTI
zQPN9{_1MKH*ogx=ixPVD(*ZU%{AE%iwlPn9ebdA-|NCJUkgR=uWvw-bg}Y<OX|7BJ
z^CZLFqEk<x*s>iWtqG%f^gq7jxQEVUaO+jkUNZTrc*qRODJRScX?XOSC>2C~&@B4=
zw at hFM!>HHxa|g!D?TyKz9eY0B1DkI(W9aa&-&ay3dzpBe)0#H<wsE(P=3wz{`;xnl
z22;hvta{?@2jc{;SXsO&HSZCydhzbu<&aE13!S>7G*nWG;dk%H`fF<7(*_s at y#BK{
zzU<om^|1`)1C6B)$LhYXRbPoZGCu8yFaN4u&NPK&*j~s#Jkeuda2gHkfQV}DX=L>y
zlRm~0S)jpP6?>Ubw<x^Ce)f$g)>A6BwpM~vGx$1djcwZ%hqfbq(n%|CXu@*UxUPzb
zxE|YCoBp!vWomj}--RA@?bOu$uL%;~(#J$tSPazyet&FfSw-jl(3zOKcZpHncx+_r
zH}7qZD%=SFba!rfxkX}sl@!fS3)}x7Ld7)Gcvt+PQmbDs54=NUe7CpVqH_xi$w4@;
ztgcEudc at Cv>C&H4hdJf>ixR%i!@~51zV2zVEPrkwfW5p}zsPW$si~>yG(UeguJ;l2
zRCYqb!m|`l1Zjy11}tgm=_^cV-&{y`BE2^H$sNm19T58T>Ep68AutesE at +$W-FVX5
zzD!3>ZpU!ocbw&3n{`<w)usDKiQL`Y^BGD at 2P`ixPcAO*eQ5knB^xLHT!xKQYhlC9
zYgEh++t=4JqlcZ*#W{WE%+w{{?|giGlr_{$OiMlro)6bIH-E|ZahqX#dT>KzCMiBS
zExUj(KT at bc1YZ%i at Pe)%Nm4u(biKj1&vI0bbLvVYF)8UJ2vcFtZ2=%kB~8uGxTTQl
zKpd5GlD|v(Ed{H#CW5yHI=liPzbDX%ffr}*Sa+RgiA2Y2b{5N4Cz%gu(Gfq?Do##L
z=2?%2Y$PTo&dADY_+USysH*D!@uLh}sj0cy5FirGaW{@XmQTO-;KA97%1Tsl0X>GK
zJh%pp50B0^8LeK}PVEV<H+Za^S!EqFWGVP-hR|P{Y09ZK{#8uOt(h~IV(epfik at KG
zhwADu5`j)LjB8F3rCT{eAJ!{-ElMizstNG%U1HqL%#Ilw8`Cl|1$3EBN!h9mha9cF
zDI>!xD*9EVC|2&={zeDCCc$!Nr{c(*`}?{wcj34_U4`Fzhq3DSmCDi45glYo>a|$i
z{narUbUuD%YrAZ~!o!2nrfL=6Yb`QCk~Lxsg6*b>$-DHx8^>5KeJ>_5$5lVS at BFjh
zCt-g*eRgiHUa0oZ-QBGgi2z0R_4UplajUC#=#UCJA=ouE6bxub58i+HdCf67_*^o6
z%8-x{sr<f_^z=Z2+<}S9r5PcPbP&qUu2zzg7*gHfW+SEoClW)Ez&x9CKTQfX0BreK
z{IK7 at 2jt~l95Lqx>xY`k^77~m*w!AHoGZ+-BV{(sXfhXZ4h9_upUBqV;}o<S6yU<0
zp9*GRU})Um+gWWFSw=rL0~A?Zbj|wNLu2;gQCfPSv#ww)v6YpTXmMMGwX=gmopl)-
ze9(svAJ9*nk(VcVWN)9zKhOO at fL@}=c(ddP>5?ut$BHC8l1HhJ4-QkfAbrTiQ|ZH-
zh5nzoRm6mah3CX)LREwY9}ISMkOFH^Hgx9rzI3PE{)Lx~9StUvjwvkUV~T&^qexxj
zL}DOfvlQJkz&I2VZ?9;WIoD!cra+(LG~(v&e%`~=((>!Hr-z3J8MwHZ7}|a3#j56y
zALX|<2G%P--{+UgQ~_}$nWPA)E19X1oSseqTn+$jQ8IW|XLe?WAW89QdATrCO=xg%
zY=Hu6#MfsRD=z7}yNi(CPq^xM0%97{8y%*9c9JVRAt4Mh0#bT1;OYf$rrzFObOKjN
zS2yH*yvq at Zoq-NA1_pQSnm!#TU?OxzZMWkOfMU5-Z7rm<g at wgoc|^>eT2)nbc4;Xf
zI+{lL-o1#32p@>X7`b!BrKQO=HB7MLb8~YVl};BxUl+P<%wOsCR7>w!#Rw7FJ3Bfi
z6%|pwefzf4qGb1jl9CeT3D$d0-K`M at S(h!rk^?3vl#h;%N{kB=Y}GdKw~sG&oN&5%
z^JY)s@|phK<FWdD^dWsj9<gmdj*qWIM36rZ30cjrtnD*B4ocU+V2s!imz0zg9b5-x
z>LMvBEKCcZQx&6o|Nfb_4~J<IKJ at WE|G31#FNxL4(LEBR+Q7Y`hK~a2L$xbTKWWjx
zFFeo0=>Fp9%CCe2MgY=yCThqqvmU|JHVtlSRBPU9=JE2Wua^?tU1>y%32QMvm(=FA
zx3}l=t4(M%P+}KHOiavWceQ0TGGQ0b&dx5Li5}8L`cn>{px_UR0+YAbHCh(p<C$hR
z+{^^%iHL~Gs{{oEUh?P}pSp&31V{hd#|r)6>&ZMi1~XbxKS5omWM>CE6HuJ+j|?b?
zjEtOnvU^}!sG+E?-a^BKPEi8);$whtm^Vk<dFB at u?0|&PVW7#6A^q<k5Te|fY5Zxg
z>+LP^dwDbr>?isG4CwrLxg8sm+bj#3Pr2lk{0)qFULMKi%a=3q@?;zxFKcOQ8<q{R
zPz0UBE1Mg3ux6mApYX9NHQ1LC3~uHhuO6<9zDu29I|6ro;r<yvvPWMA7M+MlNE+_*
zulkoRZ4Mae1UFA~V?VnxFUTx1elpn+C)CzqIC--9T1VufG0wuLPq(5yL9kOk>4gnj
zSzCK_T#ZutQ?B6Et5*$~?xu9fX^q*sxb#n#fUv1 at Vc|0Qv{dvYU}?u%{aA#DS9_8h
zb-{Owl8a$)MRY1VijEx6NE?k?P&YY?Gdp-OGBAi_&xim=7i%p#`u0I4`TF|qUIro{
z3vLeoe4i}nI{c1iyHq&+ue17SaR0C$xbOB}D_7WiUlehTl9Eqrt1Rr&--C9aIy!C|
z8nS?J=bJtiq8T_d6YowUlR`ea`$K5;XHiG{V0-Jmxo?^GT;|fEQGb%^<jL_Ubv-FH
zweK&(u4VToXJxfM0D+90q>K#yeSW&1P9hZ*6}I4XFp^GAf{;K;Iy%o$+6oK1w6=)p
z#<ZC*^~YQ8<d8WlD^rXf!NnEZy8SdAvM;*xcgXPjou#U(syw=>8ntqFLF%H{vP-9`
z^81KA?%&5VDf;sUTQ}ep`Txjz4{)yAH*8!YdnPNJBxEIIW+X{bd1UV-J9|a4vnv%6
z(niQ0$)?9nLiQ-h9)*bib?N#2-{W}S<8ai0`hGv(`@XL8I?wyQ#s|si>jlT8diS%k
z-cC<XKchNaQ&U4qPCm}PFgj|EnRk4AL}2e*Kk?__RNme)scv-WIJ5*O&z_~I+8kOy
zqY*fAq$Vz4m6w5$(brf1nBl2Y!~siDW}~Au!T2lZz)d5*!-xHR^&vh8X>G<!SxF=<
zB3667-(*itPS%0XhRJZxo;_P!Sy@?Kv`H*4m)%;53#=d6+4frB+?>d7;;6f%tq=+d
zTbm{Zo!44jETye?&NZjYHobcqsk0cfu_+iMInV^AtbE#H4ZiWHKK;W9M(_IS$6NqE
z)1N&fZfk3^b#nR^G&?`<#8J-})$rGSuGReE!+Rn6<~KPx4SjvgPck#}{Wn}Oh43$}
zP}S0+3^dW!roqZxtit~K^(*tigR08Pdo<6U&9|uAXKig=Juom3oEAOk<XMqQ@(Y_i
zVCNw&E`piQIjYyL9VvF`Crr?ooVFO&tE;O+gIahE*cn`0EG5H6t-*JPtj5uB@~mQ1
zm#w>jh9|G`LRFRGgIKfb9<C<qp)MlAnh>3m6?KyeGH&Z{6{ZotqXyHmtr82;)47J1
znny;`M7^mfDPik~7UbK?xz82Hg}&~uy}Wh<Rz4NL%C^W*AW>5>?s<>h;iI14hfjz|
zNVwL at vsNsyH8YYD>C!s0N+fBC47H~Qf6ksL^DA+y7 at A+^7#ka_y!p&c+Mcn{X3D8O
zD7}n9<5-ebB}LZ7S885f(lizYZ~A0YW4#jAmY2svl?_G>c at j<vCh~}gP)j65?_?(=
zHCX1SzkW>)i-tuguV23wla(b8I(c(z%O8(bh4ivdIyuc39-+M^$+`r*mk!$(?Ay0*
z at q1?6IDoSVER%vlAxnc(36`Wbhz4$))z#I-^J=BvU2%CnD>h&lZZd3GH$mf2Fe3zH
zP9=jDv{!Ik9F4Bv0dOc<T3UQV1}3HnOxV}cV{B_UQI<qoqQ;{PZuG4qZGnFHyr$-p
zT2kwa(g){lO8D<ytMv;F4K?w4lgi#Cw~!+8KGP<vtLs8^Y;3K$m^t_puy2T3<HSTU
z>bzlLVeq;BRULL1wl_DwNm4~ZLSkYXsE{oH|B`QEm>p1V at 5ns&LSCL25i#*nyqAlM
z7-*}Fnf at EQekE~?UD2NzM7KO<qhceQrob%{DW^dV`@_!hLrzUK!6evaWm%~X8Jd8I
z$gkB-a0b03ac;-PF&9io#?(D*KLEmPFmUG&(!dKkIw$wfrd<LtGVHgYu-o3meVYDb
zclXk2xAugP*#<cTNqGhN-G_)fw(D7K<twYJtEKKVd)}C8J<e_^0w|19ofseIqK|w&
z$@8|~LH+F6sI$x|3LGpfdk9cjC at 7?c=U=*b(VI4falz~$MGprMwgU$a;3cVPZg^T6
z=i9e$;~zX|QthW;6fPZGVY>Sv1OOiW3G{V1n3KD^B>CSKL;#Z(AMPBIT#~8lcaSnJ
zb=lJt$2qkV23Em!_XF<uphvF>EWzWIsB at J2_O_=fbyPU^e)^RD?j1V;EhlG3 at WAZV
zqr$?ib-erqHO<YWzkmPk?Cm9aXWFrs&>?NJ7rtL#&~0pKnJM4nv=(|rSYvc%W+p at X
z9FSqo84gJa3Bur-YPoZ9(b49xog>mf0x8brGQHdu&KA-Ej&W{?hMe5h!23r)8m)>A
z#pl$^uXLA1-pQu6f4v`?pK2oH-A2wv9;#xxF2==Gt(q2FS}-uTU)FfWUG~9>;M!u5
zpMDc~Z5rnZFFCJy at 1qRlCgrIeU#X)Rqlh6gX+CpRp)~^H2duB8<bjeBa2PT&GI{?w
zj>_1$xNr@=C$$>|A#FCQj*drZuB6HP(DU;0V&k+gMfMkTbwj||pFVx+?fWwAnu7zS
z^ZP$i0U$9mV12-*y=U{Au-R7I#zyzd^!$9I+4eG1<)2>zIsy!Rh1d26tk-d3y4Yim
zFH|h-?5801V-vgX7uqXMfJ;^JEngvG-PFrE1sno`)_mp4`Fr>8Th4tcaw~876~oNJ
zqU`5a(c<~_-vfmuC7A^UcbOO|sHq=IR$`etA3y)c>&#_GK<3Ej&B;&A%XcXDf2$fi
zE+f--#dPSMM%<>nzk`FY$-?7(Vd3H85dFC)9mSZv#U6!j^jAJ|YaaOX6a4ez==)=9
zyN>i<;y`>nEe9nH#4B4{fxy{qk6#Z1N_Od<R?hyz=SP9cR%fV&V}))LhaaWc=9gZs
z1PytG!fR)8^d at ox{|Id_KX{*>l|_N=Na6ywtQZ&=ynSskc(s+r7auD!j7gRA8h^Jo
ztO`p)Pfzax3Ml>QQ=+WPxsSV<%$(J3+)w}>8IEbWiHV88E7su^qe(dZX<P0eh4_X}
zrs7lqS_%O9Wse=WnbvpLxxv(?cgb$v-s>I}#&V+cKR?|{2O=yfDVhH26&VIDJ3AE!
zy8vY7B?G8FAsXU2w1NWrkt0X8s at zUrb#-*OU<C0O^N8;r*0wY^C+O<x5)crm{?U_#
z9f_WueK7pdNgiHJo+x1T`vrJ+HFb3!QBfLrw29QvD-!)jHy=-az9g6keD53kla5Cx
zf`Qs<JiEwIWmaT&z$2(^5Aa&~yf&wb^77G%iQwpH%E_sziCDS$hvHFMA7;&^QcM_Q
zPmt{AA6WJM^K|Sz0MP at dK!=PZ9Rc_Lj#_dECmidAA+*a4;MbH?R2H_j5#cWYD#bzF
zau#qE`Q`ZnutJ3N&;Qok+<et at GBP@PGD at fpHkgjChRH5qIQ^&bkLoj|+;^vEW`Z at J
zja%r%sMl32zt(+CHsrHs at ky94tze?2iPIGc7UJ9&Nf1Uk3V{@Y+>p0zyKN|&wo at v4
zIs#rb!*0qcVtz_`wlci^`gYuqQ at DEOSZeN2aG!~ZNvEykR);*qZ;2#I_;M{{TUuH|
zdI-*EEE2B8pXsz*@ThLA8kW^%vYC$UgewvDS*bC<slitZPKb%qcvZ48)v>m!qa<B1
z6w7-$7TcH9&kt45F*8Rv4it}_HTz8u9B+D!1Pc>(`|<cM?uq44AkWv2KFxW9!EEPu
z at 72{2h3=l7hP?yrL&d79s#tLpTtg5T7>E^&MkP+CfMA%&m7VddU>8jM_fY9v<@8Tq
z0zAKYdhv+p at 7c<u`gu3W1>%ir$6So1t8==i`G=Gg73Y@~_w3n|W}uPiIQvEEczF_w
ze)E8}M2)$a7O78)7C$8oO+`qhL4rEty3NYn*`*?HKR at 2T$ISAd6#6F?7hM!-rneq?
zY*T~W0-tL;N{`+)nEv^*UggHO6f6bl92tqbbBEqxpqR?2*db}9A(v#|K5TD;-5UY<
zPXoK+xUqU+SF-1s2fYYCP0C&HC7Sx0oV}A1_GpMu at cWhi{P{lMqS$~CTLJ6qM-&zo
zo*21e^rCxg^21$hmx)>is2vt>gGZwF`J8RAmy?rIo8~Tpz$8V1VNU0?dnGT85z&u;
ztR6cPMly9ZHRF<&qLW;s>^{%CiQj_PVLP~(n3%2NjEp_3G5e}}&JSO^cFq6nxpSus
z4Otn5^-qSzINj;7LsX3AREkApsgK-`2R1qaY9I&xOB0Najy^JHhsPV(OmlU*`{k=w
z%2rmqjZIB-OiWFC2a=MKz{D}v(Bn#~<;27Uwj*X`W$o<jtd=`jK+Q?PBvOw}(kvUu
zvki;LKYaM`m}5B`2(ONTK_t-2jJI#;L7f2?4L-?rG9W+^ZxAXg<tM*>?HU)8aMaKF
zY`?G6pgG9VFo*HRjT<2GL>6kwTl1^;&PX&#1)UVjXkq(f$`#^I=P<@&G$<n}83GIN
zph@^7pzV4<aX#@K)C?nae@{B@?#t at 8Do#)E$hR<Y?(5Bb@?`zSkNaw*+|nK9VvA0d
zy+1z60S8LsWM`*}R3nU3yH#2$6EMS8#Tco^@Hl0k7XN-<KR=&p!`%b!#m(&WLw>Vt
zE=#EM^b1t(w at -1-kJ-!aluD(v?Mu`QZErvC?&%49yzZ-Bj$WNz5g~!UzdsjaR6=~b
zOB+|QzVPD}d%kZ^fJ3AEosQ$c2A>%{c9;eY3<|xYp`k&)&^F;kxR255+S&v*(RgE0
zfsF at 4LPF9D3mL$D8o*uY4j!DD=02a)*r<-hpqMS<^5fGH(9L5&80+`a$lq*Bk?#0W
zq?|qQE9uzP^{u?~FM`3tKy8^9^vB=37lQpn(Onbn-~Tqv-Qf+MDv~h=4_9 at BfxK_d
z)7!^Kng8yEn<?I{qobo9a_$vO26J<BU)d*<Ky}qOHUepGzw#zEhu6HRZkG9ByN69&
zMg})?jJhfF$3NmJBa*OxE2-^q{VU<&;kg at laAd49;LzTEgZiEVH5JuYx2Llk?Qzp1
zTo`3+<f0mGA);Uj`u3kqi-LZXTYmFpBnPFPLX1V$R%u@#hfgB`Z5s?!1Uzf;t7sq#
zHbvdq+LBjLQz1>-Z*xv#Ajvcrn<!yP;rQZM9WyiY&68Z!|0W7u08hkP_@(g%Qflg7
zt2)jHQ`*A7{dNEOXAHao``;#~V8wYjFcOF1N|qF9XZ784M@@`{u65aZGsjhX4Rh{?
zza6UU6&Dx3>p5;w;*jv&o!nI0=B?N-0NBp%?!BXS%zk6Fp}3hXbH+&y<?L+KBtL%q
z(9zQi!z(LrFr*#E#%wg!x!^6+)6?B{HJKl*Re&Ka2pXH1#K2==3E}4EO6>IP$Qeyd
z@;i6#yd>OWvX*4+8Xk_lf1gqE`0=H*F8GHanbv>*#-2tePMpBIB5=g`#KcokanB=q
z>>4&!rY?XJVyXnZvCb>q#kIB5PEI0NHE`<|A at Bw+9-i9nx<iv;No_U{MvL9zQ-}ro
z#wChJoJuMxPrJLXK4qWbpZUao at Zj0^qNGX0 at M(a_F6hp&J-69oKQ&t9SDk#L&0NfQ
z#V_yFEeM>X6cnl;CCsdu85tYje;Yo-FDx9bWTR|f?xID?UDCV0E^|-w5KWk;r|0rw
zkxIr%FRzlb*VCJuH6VfyM_%T?yOb_4pp=qZciu}XC5Fk at l}2A~<m$>xAdVurG-PB!
zAe)<ZvLg;rq~x8yHM`_M{d>rcI()B!f`YvJ99!f3_wP4VGDf*)L36gINYg|{Mrvqj
zO)6PIaD*@V$PmH>qUxTX;@DDPtS<aIkPLS(FRW+-&WXo?pWm27WoKufSSxtlJ&jGF
z{8zh9W6NP-Mha}7H$HBOK8sC^EiEnO6<RT1rRU^O9X|i48eAJIo*qBuh-k0`wz>Rf
z+^x*k8HCM$M>D}BG*5tOnw^f0jI2pMZol~Hl6J<+m(6v&sW%!M8#@7sTONp3gp at c-
z+PxpQ{a-FXb?T!>T1|=dR?+IbC=0|TQ-XV^-9?%{IZzKPd}8O|XjdI at 2pHOsNmj6F
z8g!fU`f2=)<BJqCrTiyHN!G(RS|2O*PuMU0{P}g|ZF{I_zm-(rK^B$|a{h81{SaRk
z|41sG#UsaQC!jWXmxur~TD^yoni}m34+ySb7YFR#ldnHJZ81y_3y!xN|21BMXU3q5
zloX4(LHx)*;18jtd6>xUcoiZ(#3~S3)!3PRyq*k1KDIiC*EKLO_%B8M;>C+x4_aRQ
zQ#!5=7?=s`iH&Q)EkHNL{r>IyP>Ua%-YY3_3SE9v|N5V-m#WN&AkPosaoxsm!xE4+
zhy2 at KU#JMr8FIdA7-|mC0!aC(w>J}_B)*!e>RwRV1mspDP7pLnXgKP at 0<cUwM~?!i
z4DPJyJJwUDPW_GXKHaz at gM-n}pYvn)TqFX(<G&0H_>k~r=wSa`G_f;h&TLib3Wn+m
zrmkdECM3|o#{=Gqnj$kZ*emQinB at 5vdtH9s@^W>RfZYRy87>;Q(bC-7s%B{TC3<tm
zws-~vPUQZ>F3Xc=@Z10x_3-d8`~&=?$Ruvvx at B<ej5eu@$JiUc%TC2FTNOszex2}A
zBYG1?JQ&Q{j_0sJXPamZD^;6wu1X|%H-Fi<Kg7w*?BX-)FDEbWKh)rtkdPoUW;Ye;
z+dlRKB);0IQ|G*$)YLGk_UfR#9Ct{oP4AEP8MbB{x4pv5tel)Qk!lhd$8wEDfH=Jj
zSLzzHDl*LIo=)$v>jJb(Jx`LTdHmW7`*)@b65&5)X5^-?mjj9RX;tAV^qz_8vMUl%
zGIOwZc0PUiG84 at L5fKp=H+N8YUTO&}U4$ABO(4C5N&lSZr%#{oVg%cu>0G+B*UZce
z4_)j(XV+kN;Rg>+PmP~fKYbe8CSwf!`t>W2|FAZjx~+|se20DsVWVP7KtO=quEK?4
z$C81r{{F})CeiW1GB-RH{^y?q09!mfJV<%4R)C4g3SipZ)RQVWI2dz16xT=ha at cvv
zupvbH`pV9~TQfzm2qA+ZxiSXqzy)Puaj~g%_0b0{33u+?`t(T`)XmZ+Q0`UdM_NMg
zN_boF at Dcm_f(_3pt6#&zrhw$zhN7aP9G^WOJ73*=6gNOjSd)E3QpG6-uWkA`SXh!)
z#VZ!JAA0P&8Y<U6c(Vd<OOd4YTi}rD5a8*4a*wy04jf73=kQ?DS`c3Fk(?Jb7Ilk?
zHvR344Q}(lvq0=R^yE|&7cHs0!r;uunV+_{wkQ7+#;;!s3NVmk&(Vh42Lwb!8p}Oq
z;!{%2x-o~)j$0HcK2=}|RSEj?#qd*4&r(ii=3W9?X6Ek~V~!m;LU!>L!}aUe<29er
zl80VWywKU(8x9Z$Ms8{_P*!UT0{doG)Rg}BM~*BkEUPW|k;-yd<?Hi@<ao%kGmwY2
z<mhF2_^=x>2;BX^MIZU5e~+J32)mq|+~JUXBSS+mMMY{*3c`kkq<=G{)NE|{L-U>Y
z-u(~)I&r)+Lls*Tfr+6ef!Kxph%wjk&Y`n54JtHQJ-3<6{OLq17YXl+nA{HBUP*%|
zmA92QSoHSxe)wKdQu5x8 at xRs*q at gO<B8H!TAL!1A+qYHR-6cn!Z>pSxSJ>X(CNnZK
z8wuITy{Gx??876%!jyzT1Yk#=WQ7Hrp!e2?*Q&zTMglp(k-V2b9A27p4)psXx|}kD
zl+k<>xWvn>tU3sC7cX7fZBq@}3S{^0?W7J_(nXA7spY9m56^v0=y+!Zk*W2mGOeCV
zxCad#d1ypz?61xjeAaeUR8;3k^|Tp_5x-sLw%G9ILgC%JcXTW)CoHU at zV#iYxO5u4
z_e)PQtx9vNq?lNBE~jakYr`S7AA_mVVqz{Ibp!?km}QIHW!d4&*VorOK{HXaA-I<{
zCQPZxKtqF+kDe?<W$dYiz23>lP}%03XBS_s7x}K4Cw;Ic#F9~NMa6*H4(6HIqlubV
z8~k~bSm6O=ycfB!B_ErtJE`wGY~7AW=+Dok0Bh4CmUOD^?4-rvAYg2|moH<6rj7su
z!JBrAr!QXc9z9C+>eVZ}r&i-pR#73Ww_LiH{T|P5fZ%<yJI+A9n_3QNi+6NoWh_=!
zR*OG>VrwPuX8(=luV5HYe=4Wl{a{_2;#?KUF6WtET>SZK>9w!_CgL~>&^7l(&CWMy
zJ&$m^S at G;y=!lfGbY^ZY4FN41TQr~y*`E1(_ZT3K_^j`zb-h6y9vohjD(6~o1EQIT
z+mgc#fr*6$n%%p6!oT<W`TDIC0~DbuKu-X$fR_1vzjP<5*;>NF)|M1ZLcVft>*3JV
z&7t=;V$5Qal9LxhstsIR^3L8I at QM9$VzBL3MPN*5_4LDV^a<%YYl$Tluc`Lby8eQz
zT|GU-qd$y0YeOj#9zSlMx~1P`n6K|YFy+EdO(Mxk1?vH1Nmf=CV-WH_DXC?*(U1@#
z&`wM7KIhL<w_Skkmppb@^~+sxW=dNpCv_vEl(Pio at 7}S)W^MUE4B4ucGV1$Q85$N=
z+uPfF<*hV?Y4|b0=takN2PrFecLu!3XJeZ at XcW>rFhCKbzPF&ibwudzSspgay>s_2
z;9c$Bukih))fcCbCX}^X=Hq?uvy10SJr?}uqW|%EEDZeX-^WyA_Fn3lbeKL9O2OpK
zTv1iU0V0Forq9Z^C9z-Q4N*Ae1$N;zpZ`G_HfrNaZ at T|+F`xzy+ia|DY%Kr${)rcT
z<mBWwLwtI`0#JPy78Q+cOr<q>z27tmc<ibG25485yko5W%Wb5$S309b(B*h at E8##;
zP!Nt=(AQ at sBqYQFdcahoef!GY4Wn=0cJ-lITM;6-tGC3#%uL|&bD<Yx;mT1)OI8k!
z#^z?yR-0oqR&perN${WQ>+4HHx`v`oY4FjnuC8*)$sN>I+~70y`I=~NN;*G1U9%9l
z<A=$0VkvFjSUcW&tb^<L at t?<bW+SJlXlOD){2n-P;J-Tslwb)7HFB6H%`GAEv0L?!
zwIK^EeTJjd`v%5`W#JV+ec!d%|JmB|F69#ui4E5r4Cc>{^3u_X4_7?I?Q-Ic`9O7b
z_0vrt5#FP|E-tj56*p=(H#Rr5N&f*nFgfNhR~1M2IfPbXlZA5M1D%&E$Ip2xbdQYe
zlylF5&ZP;w-S2E`8{t&}4?orfYW2&NG}GKEZZUJqP8B|SZWMQ%6`b_+<2Bkc(zcSU
zg at uJ$-)H;3eEB!H58~05p+ZGKN5E%m36{3GJED`IuCOWoui5@!oF3BDL_$nVjHk0W
zK%%u(OPl7p9?$?h%{MeO?ELhpzAjxb at pgE$|BV|PJ8iSZM+F6MJA*w&L_`p{6!8kC
z94j=;sU-Lg7#bEH4$(~vur10p-|Y<Jzg9qkuu_8K<dr?kupb6qR95AAhB+7-wpU*$
zjX{{m>vzE3I9;|OpcP%+-2Qh`4+x>lRkIhlbH__J?@f5{tv*eV8;iwmH|4mw$HLyk
z&`AvG=X at qxnaggLl<S3Hu9MXx#!M-a3+(ml`}gDH<K8b*Om?iNm}kA1xl0IVKDH4K
zX0N~W>r0vQqVX^A_|$7S%mHghKkoUM*NfVtz+<SL+qt;Q=Q!S-luWzJ4(k8{lASwu
z4tzDtFy~wE@}EC{HpB~7*npZa)YjIHE-ZA!9Ra&tH*N!K6fJL`FCdg at PtnUdrL9d9
z_~!3%874G5ejE!9#&i7m0SKA8(TO0Y3CORF5P<i}85B5u2DPe9E8`5l$QVTofoJQd
zurPsfDw}x^85kHXD=Rrf+z07if6L=(_wV1kdmK_L=hJPy%dDMBK-eRf)fL1 at pJVn)
zSLK!NvK9ALhoE%7mtB86-{@ZZY%2*NPu0YP-Ok?r!sW~1#l=#6bDnDI>d2j%o0~^1
zW}~Ai$#n;9YzkO_xgp8R^I9;4SA}4lZT6l6<9Ha=L7O?AE8%rPa|0z4WmAt;0|FGc
zx2BYMPn_7eAD9Nw`b@>e0`rgGzYFA6Do9C%jg47=B*3f+H~y5iMhYe|z%l<Zq5W)Z
z@)-|<-5O?Jz7$YgV&ULGW)Tt|&I^d9tgI|AGe5i`CoB8=d2IkT_^+c;ijFxbjpyYR
z^G`~FzV9Xw{p`7nYWw5Ii~b9KBXz*Soz^$$I*zf%$miVv&G_l_=h~JQ3T$y+U_*sB
z;QIO+6ntc5WD^q;B10<rH&Z+)ER>z}2%3G`sR&e5Rei3~PSH{rIW#gpBnz!IFd&T>
zWQ73ID<4~Cy6*dy=QuYv_x at PXM{y_KFcm)^;e;KynKc#H*4Bo1h9YKFvmc-Jo`_Iz
z9+o%lObS)Kd-pE)gO*pBpB%-cr1k-97a3B$GwlfeEKE99*|EV~Gca&aC4-hQ=q3rb
zez at G!Vv)YEy1RQJiPG5M(=J=%+w3wsmH8*H*xG&%8K1_argP^)fv7AV<|>G+8>=pJ
zn_9#c;uCR7jnB?L!0aEnh06VS6a~8_K>1 at 2eA4d#BSX>p;+Lj~i?Snb=jMu<)z8TH
z&an;KtQjPT6z54xNYKOP#cS~4p|lo%uol0ngF|+cqFkA8+bX8;rfsht2Fi&wX%JvK
zZ6$3T9Yf)Tu+9CGCp=h)nVx<#;JkvZzdsv0JNw;zr3MyDfX;Zg0F)w%@u-TSA*;l(
zV+r@}sWtr5oNXbMvQZXw{>4P-FN64s7cat at 5?w`ZvokOdfygT at Ej<Gf{|`?>MEi*k
z_rzIaMD{o_4tp+)&J8vU{IPGJDM{Vf*eDAr?*C>i^V8Rif|BylxDwUw4GnQt4>>F|
z at Z|Y<CsXDR3g!y(@~tV&1TJMAti3WLG!zsmw5vAuZ@#*o7 at wHnPS6;gnwrY*I8Vww
z8I=+Li(zSbxi*$vZgDi2#@AN>v*D8J`Wj>sHHkx2YMAbWQ&hzrRowe#ANLgr7Bhxg
z+2()0y8b1tssaq{NsC2t;3L0&d2-ZaCcwtZR7RIA0fB278z^a!fl8khdHEk at s+nEp
z3&ZdFN}M^V*=30e`lAO2jj^LSo@?ppF)j?t8(k|h#C|0xBRT at N-L^(al+vm9l=Ok7
z!2Xa=pFTC;+5VGepx@$DVzo9uwEi<UFsA#3HV%9n+1Xyh4v&Tel6$ErDMxV#ljkh+
zzb!3YeSHx(ZxY~1 at XY6&%s1*GgZudJlF6o0gWv{ABEnAAPk8KeYUjJf(|F}vJvST~
z?=f$)`MtKcwP#=;3PjWD&!0=1fL==TGczf)rJknI&02b`4L3%tJCd at d)DfA6tE&l;
zlao8)-M)RZ!NYKHlkY!%+^QRMA+s!?{e>ei%xehAEk{)JvJM#LFp`H7!b8BH at Ep5+
zroxzVW at l#HeAwlR-9^aj3905!1%F at ox4=Q_ATqAb&g#H#@D?r`%U1&0t8ewS9~8RD
z#TccqMb_+KQ_%0SwSFD at me)U4 at JrXr!igw3MkSUwCrDWM>5i%m_E7-@!dAxXjnY+P
z(_`@iPhY<#$MnYSrMN)s<%Yfl<ThPL0ACL19nNS0AzRt)sjR#QqJg}E`>m{|;hdZt
z<aSn8yf{e&__4dEXRT97-N*<hj&;)3Zh}|Ia~V at XmVM|@=r`Zy?5{2|>y_o2&&J*6
z;3MPb&A6D=BhqlmfLoAJz+y_xfjv at J@ara7lE|aI!_wvXmh$_8gP6vzGcq%Kw~4LA
z5LHdDQDEHKlwa)x{swH at +nWzoUp`3X%cT&NFKE4GsrrX@=I7_h*z^7oJOFGJXuL2Z
zMc_FtEz0*g{0SQC4H8o)J#}^Oktp at 7S{C$oe*Adr#}7N;;y~RQnM91MK{5FI`^P6I
z-v*Rf7W4G<l=67Lbd!QfWY>lZuF|~D%0k(P-NxTk%&T0xHvPy@*FaA%>@<fDQB6$^
zwqQU7tW9(6v)#38H6U!c7|1cB3<%s{=L}FUY>G9J8K-zli+?v!2XhI5&XoAnPkMIQ
zN+N?8S at l(@w-iU>@^hxw3K%FYDbZsNer~r$2s&o^jj{*d#tsViwJ|;_^ZL$6N+fSv
zSI&0+bE**+ZRf3x0b6(Xdn71#-*?nP at by#?7_ccQt2_-fbf at BWWF$80rXtR}>L<<#
zSaRufZtn^~v75%d0~-_+6cediI8g-yIzU=vWTg9}C1Mc!!$;rz&8Gx`vXy`FRfu|K
z(D at g^D_>oMP7>tnQ at +wA2Za))+X(o;N3SfHsKQ07)0BM;u at XxeRaN(j`d8oPf;frs
z0A6_OYFpXfce;W<R~J99!G`RXd7&Y{-~K83-CcHaauSCp)Byax937$5H86-gAf%TL
zY7CgmV-D)FwtXZd*dZP>3$eA?^2*B6pM^k908Kne!P_RQrdYsFgBRF3f!tC1SY~FX
zNR6W>B}UYxrR#WRedGvPQc{x0P)Ya4kF*R7lhb4SC9bHet9zeDFrII2yKVII*RSPX
z%;x96iyutn!Ag+;M2L9JXIqb=;*)NA{MeQsD3bLdAt4F0Y2Ng;a26!cMwC`!_{a|u
zCFcK_=?0H}FXOX|k$~C)4CDdtv$80_Bf{zMriMO}mXlNI8VT}a&~@c5{_ND at d%@*i
zzIrv5F}pI=QD-a3BPK?R4Zp!S-xd at kBqoLexg{bY@$>C-``FjVj2GHC0cE7>2Evf8
zAl`P)m(PeWFfnzMohpD=6n(!X5wM!cIfg@#W^&)3&low<+0_*^K5iKj5;D3tRwrHb
z5_3avdf4tfGE-Ai9JsK!+O4fB@?fNq?(RU_m`nPbH`FB5Y$pv2Sg_4Ltf%$Wt5y?@
zF#<=Ao-#G%@F+QKoy^R_k|bF_H=Jse)l=&rwbGmahP!>J><~Q}u-NsZ1K-yVf at F~w
ze%3v`{%!ClUUud7W|owAsdx`}bkOAI=lky`B?1T7x(=SavuPf;myl3?^%@y>Nb7Mx
zR6Nhk%BpLKV#0D;JTanf6&Dw;<Tdm4m8G%jE*ojPF2mOO_ePsk%zzf2qnyDx^7ifV
zCR>N7-Sow;C6 at UQW7&6+^7CiCVY-yid?O<x;QV-P4|0*4mp6DPJOAVX_PhJF^ZOl8
zMk#JBs^Wos=gd#63A(X16uvoMfoHIfIe0b*O?{^iZT%{FA8MHMYgMP2LC)g`15TM4
zs_?>A=sVjBJF&U`UxiR306_?H{CsTyM9bL+BBw4~p!<);^)kIU64ij`5FDd6q4zPP
zw8(H{sS$O0LP8KWz~klJ3+}*>%<J|AzyXB1uHXT)?pmp0GjQuE5-?hx1hCsBkwM{4
zZON1o4VHp0&)8IU99)!QrUbS9XjK73 at nNE}+{+I?`zY<wZ5r}x`&F6;)w$=6sBk+X
z-!N_gaHKIZ6N&)F+n6TG7`1Wg_4>NEsfRxQT_jAffaSA;PlYbOSrqfcbG|gCxSASe
z%=|?54GV-fLU-?C5(NCv$G6!n`Ga-si?o2-CKa9*fUjb#F*w<P#zxen+`ZlOB|8gC
z4fdLWU*kBGMz5+BcEHv17X%Y=mQUroch_h8@>Av?I*2hZzEJu at _I`6{A-%^A>45*v
z=E&*u=i41w#gjBacVM&PpuHq`jR4YJnTr<hyuF5f4&(f%hzL@=^LE`E78b at o*Pe6n
z)%wwye{D+!HfDHud4V}Wur;|gn0<+Nrj>bDk9w^D%9)hkDTtS&(9VyqFwAM}?2K=*
z2NKOqAL*T*9GY{AC!uAcuTLKs)9=7Lv?|So;*yf}s(lXjwzfALER|hc#E>U~`yM)b
zpG!f39mr$(v|D|dt&2;%=y0#C<W{L>hR(?s-OBJMxQ<=Xa&Jgj*mCHmL=q)W0;xn&
z>oeO}UBPEA^OTb=GM&T2?NJ=aKX3-Fk<r0yJt0uG;viD3t*uMlGqum2&CF9slEUFR
z*#8P!o*U+<J2{Ea*mR^wcXZeBmS~(gLx!UQ02lmx1?p9CrVaM#3fvqRLDtQ=y%@er
zA(l5ILW_k21zkXY;CtdQqkKzZ>=*>Gky~2&!JDP?Zs47cj*hkt4r*?0$9~Mrbai&#
zynmlDIxY_QLK;cQv17+J`0suIYN7I+s-T}TRAo<*VGJVUuGDk)@=|?v5tI5~bT3V9
zZf^35h)^#rEmbNj`}@nA at z%AIyb_Q|%3-4>S!kZ~m+SvLGNNSRR&}1(*rX*OXYtq7
zxfd!(6PkZs&`6<G`6fkvO`8r)o0Qw_ZjW8)?(N(MAXLhs&CMDd#Z`K=DU1(qi2J_d
zz`1Yz1y?Wp^G_&n1WHQEYT!f2?_s_En~MzlRy{H@(yhv0%BqNU7YGR0Iuq38sZ*y)
z*M|L&zG^(Xs7=!~0Zh(iZOAJVVkoAp=;fNuO>?)Mu`9w$9#SeQyo#q|Ve#o+01wBh
zkXyfd->ZX|y?y)k;%q_3lfuG#jLfYNlaKBWC0!VayM3FSpc*e_E?&F^>JUVTk9SB&
zyy)%F2L`1P9^UT5`lqJ>l(%IWB`x>xX6Q6?%7vGKjXKY)133VaA;}tZwlkUdZNK6>
zQ<|wG-{RwG0Uz*7ji<h$L0NzS6H_honL)kLM-r)I#J4+Q(F|#DaPZ6QY_)UesPIbJ
z$S4YQfD9WoDG?!77t6^V{O>kRQ<FT at p1*5kWZ~`N at C~{um5iHRwv!z+<mAbnUz3kk
zU5x|fcGWoOpu`D~24P|QK+IxEgolU6Zm-g?FdUIhOu at L=-X|QLdgqRcTZy$Hut#hL
zIecTy<Wp}i9_G?0u5)n&tXq)Vbf~~9M#RQ8BEXItjr_j<B3xBoE3*rS*VSUbzK0K)
zak$Pptgqtqf>4Zl at Y}Z%I9 at b6n^N-`J>V)%Y@~5~@%*{(=%x6MI?Md@{CqmBcgoFe
z0>}k3CU|y{j35F+$WSpDalZGGrLbY)_u`HPI*CgVkjg?z5_bt`d3pKz##8mov5vKs
zl?i31GGpp8<JkIoHLT+md0;hM>7!d3)oK9i`F&yGw3!(vjS~$er55XV&*6nZtq?LA
zn$zy?l0<}iEG;ehhv09~)`R=^t19_{S$|{rzhT<ila-zQ)YYqp;e9)YhGN>=&j$?j
z^bi*rwq at PfM@)R~yYC!FJv9~8&&9*ZCV&u#(yw0&?zZq~f28c^Cs$cni5<JYwmewT
z`E}>e+3(}klE)+^L$K_jjz*~-P}Cd@)TWIUQ<O0EAw(tPMYxP3^60`s1~8KfuLX5~
z|H>EPk}r-hkmKDx7*RMi`)`#f2M}gU{|e=zpu-8Q$2#PL|D(<&F7@@@K<KV7jw;U%
zl+dIF{5}FEdgtz491S;yMDX?jA%RKdW?`F>fs7tIpZFAym``>^TY9f_&wpoSVPk9E
zt2xhJ-Vcz%@19m{m;(xazsiMniXnH|)Rwj7wKXrY%Ta1P0ZQ;ZAi5Va(;2u!;*(l!
z3Nl{3B732o)`%_-G!v8@!AVzFiTlC^wRL at 9%es>DJKZxsbzbRiJ*@o`;N>N)e){yr
z3iy)G)UnURM->%0;Dhhp>JweKaDih>ZE|-jz<)t9V95}6C<NvT#z{&;gO^3#(!X(R
zw6!&tm{@0b9sgu`tIaW-o{T+AaiR|FJBRn^9xRn<ZfRLrU2JyL1vsj#lm~gnqo6R<
zUB}&x^HR==z6(*>+33Jl3z+SV^(ehJ&H?7qIC3iZGJtWxuTEdTE{^yNBrl?2{K#*j
zNj(;3=FuQW#^EaWX?oFdMdxAp2cw?q>iaHUy7cPO5pL;f%Y3beBe1+cB*$kw_wC!a
zzH*9;tTOU{xd1#}A&Z`tmOS6uQb at qA^YX-K0$5d^r-2Wh**kdS#*OcDbEnMA(pnmS
zm7S))6&&niE+-}R`B3#-K$^nyA~#IC5&VCQ*hYexQvT*pSJ&7_AYx1paiTLH-=b#a
z?2}kyk<%(Fgv6RJPQo4t3JGBiFsw?(^XL1ltgO)QaT4<1q^>bcw4~MJ?hdXouLg0P
zXWuJ=$x6V08yXt=iX9JPBOZVQZJO&}PPmK`VBbkRp`B_^Wr$AZ|MKMv*6HKK)oyL2
zKeef;tOpO~-`MoTafBHe8SB9AVh*SI-w7ybRq?D~$0j(#@BH%p)|_C6R0J5t;RPFq
zyhdVR&3`bcH?uwOvCHV5_O=~3(Tdd=x`G*9w&wZz>-G?-=@=M at L`6mQbMzjz*;HvX
z{yj6ptbAR;P%VB{2zfNd1&hbogXvUAlXCEyj~P3T$oTyd4hs*bqoWJj+VaP%SCBto
zFewEY7J4Vcj(T3s>hY(OH==cxO4(Qe{R4z1K0clxQmr;lv at d_(f!B|WKqn(DNKR(`
z at 2U)9cK!OgOo$5if58r!Ccrw7+r{PO<<&g0GK+KP&UsJ7NH3PW|3d{V6l<lh8VZw^
z0s;aab8sXPzW?2mi+Wczv#WE*%=#9D_N~QmrFyK^#f9GdHP_tUzTMi}|9>u%g at mwb
z6tGD`Lc-g&;^oVT*jF|J;dMdn?d at bHC4aAIW2>uAqU*@X4Lm!^aA4uk<Tom6>g6H@
zc6N4=b4;>wa$UW>O=$p?*gOiXhYKh-Jp4nYKt*i&A0L0R$une-AiPeTyAsnsf1lYq
zjK`%5`$66#Z6hLt7zm_0&Y)x-zK3rNcTRzEeo2p)f(Yoy at bNJZiaT1B=ZAx9ZpOw^
z7nYQ)%=gX~gQCMisWRYbn1UXW8XsRR`VpR?^XUS^v6MC$ulYx3ukyyIV|IbC!P}7V
z{FUY9_&ayTr^oKhFmrG;p60v}k}Bu*q_UC?gcpv}!XyBP{<l}!MQ&Bm?e_Tg9w+mk
zvp1{b+*v>Uqfo95b4&8_XrDcMh9gy-Kkz7|o)5>-mx1eboH!@4Q<Y{6=P7`6aIT5?
z;Qs03$6GjWI5SQWi+C{~91m5oH?*dvrY$!Iw_CJc)+q=Hm}Mex+oRq_*->g}2IBF{
zmjajy&BaA9JUpD&?;w_>1wukCFi<JbjFXcS)(X at s@&}Ozym)u?s!vgHpDHZgbHGqM
z=EK3;TkF>gi;HmxI|Y-7T0npz)^$%#P4ON-egIzfF~_{)SD>}s-4h=-41ogY=vA<p
z$uI>*#>Au+t`zNNxr7>vyjW~zdvWm<W+<Aj-#pM2efu_Wv^OpM4_Xuyj8zmL<mZcH
z6KFJ5oeLL&CMT`^Ck~EO!^hSnJ}{LI#0kXnL@<rwBB(UU|5b8gBO9DjxYe<<)p4$E
zCRvMLS66qj^A|V*CX~y~m+n1u=n&p7y?y)krA{^F)qH(C1cVQR$4OHiX-%ONOk$Ff
zM6e=HpFUlzmZET>AR^SJNr`81w6-QnO--$|*h}GDaKig at Bq~4GvPa=1L#l&?kAvbx
zZCqSvecuf~6Y_F)7A-V9O5-{6)Au`kxyxdjapf>D&qIZ6DuaYU1n>pSY+?U@*Y<}G
zh0sb2{6CEMF{oMK1Z|>E+}+(5b&6L`9dLGb*5IS(DEFMd*;bsM&P5ZLT$WQ(!i13!
z$I>rPo~f$V7ED|wF2u>kKz;w1v=IQS$9yF`J6>wytaB={>VAF>hv#B{nZ-pHL{q<?
zJ(sbfZ~Xnv=KH1YwCzt2ta>r=``*2KyW at +aM60}8hI_3g#ufs1<WLxc at WxyUoO~?h
z(07s|-$wk|#YoJ^+|4cpds+Ye8wVHvhuoSwUi1<a`ofG4PMKeRV?4GtFfd>t$-2kb
z#HCjD(!XC_;j`gkHE7>+!i;9p#y0u-#<vWBag&Ey)+$c5KO7qyi~nF9ufZo at -PXvR
zsOhsd*Uw?c3(M!m9HWlOPvC)HzkV(CUg9AnB*aO^*{1K3nbs*NC{|W=oXfzrKykZ!
zcw|&mq%b|RjypgedX at m-w8Fj0=KV-&-0|Id!^md$V}Mye0f18_9lHwKK&7OlBy3WC
zkEzSJ%Krq`Q%#T2FNYpU)ZE%$d*76{vvn070S9vK4!ZQ5Gfa+4$#jr92t05RXKsEt
zoVL2^(kwaUBD46}$a#!JX?@SFTelX!*ymcb+&`K&^UhtC9os4axy7Id%ocb&X2m@}
zmy#Fo$CsS5jUFQ5+}zyDoE#)7BbbnOQZTXHeE^_V{o+OXKc+mBj6f0|ik<HGHg|xW
z7~Y?NOHlAONio=W9MEzua8?RN;dD##kimU61xX|&MPREQa!NL1PO0Zff@?;XeFFmn
zK?MKR6#%Q4UHM85U?EUJLgJ%p8fs%LHH`d80|HMKP6<3chdJ?ztTACExkIXbf{Fir
zt!=LKS&6 at QPjXC(n|lw4O^$88dIJ!-f6KPp+S+j3H6R<dt4b8QjE#LGBO at Pk2$m_s
zXGAdv=89dcSAYJ at 0GOmvSfwp?98jlFcdBg>eIL!aHA><=SJb<h<9wq9Of}5{hbQ1Q
zI?e;f$`_|;zkxYVly~np+MKb88 at Ttl+{gxX)pBBP=lF)fo7`L)zqR?@eCh9<+}I8d
z6YEC;mGJcKhv(U5$<MvL($@$o-o0Zd4<!V~P~CGrv3*G9QH7hvvyVq8RKUlv!i9Cz
zW=LxD_wV(UKU-T|^pW1uPK=AUqoP!S`rutoqUPAtl=PvA!+4bogaJ#;aXgfe5ax1N
zlnh|R{x7-g>%WNw<NnuNT;jccTh`c5i5077?x~<;V_>L{y7;#C9&lDzsmkw~Cr{oC
zSE5UrZ)|G1aOqO8Ra0!nlP83jNxHZwR&Oc3yt0yS*+34w>H^(4iJ4D5J!8G^w~u?y
z50V9)y!qm>FL|88FJ_N6e$3kksKC}}yVvgFt$<`4Bb&o(i@=A*u|U;1U?nBx$7_Y8
z- at kovfGbX71i%?}k;(Jla2>uPcldBP`HwaU_<x`TMaJg-rXq)ghg*HPb4b206+Ry<
zKc<lYbGFvYH_QQwnk?2a@;7M}b4duW2^pAZDN#v3BOTq?n6`5BAdoHNayR=TL%`9O
zJ$CSCK$^!7o8D;xE{JCTzse0SB_`Z6zFWV?#Xt`0{eDrL9i>7X9^y)rT+r~lt%Cqd
zy0Hb^32$lOjf;z=I2y4MlB~K%^8v*EyJ2}#Z7JapE@=YkU5^6409W%MyJ9cNN<>V|
zEiKK2iN5)k#CmUx&-|l#HU0j5M&y<Y at 76dl{peH_<5)&G2M$CeDl?v+9&3}?ozv$x
ze*!Z-{&$&kaMf(*ry15vV}JtzH?XXLtB+;X*wPV0JFYn9iE%SVIk&9wc83Eco*$~f
zg9ebkSg?T1VKj(D=i)^PAwq$WR!I;yc-3?EECtZs>l at 2gs`Tw4DpmK09g_>ZPHkSW
ziRDS#4ASimQmU_X%R|ZxZW{ePugUCMbV`LFj)meaUn^I<z;G7^k349Bw|_N511ov(
zV{dOb&IiA%tyohNj0H{6g at uwF|C$!qv_+?AyUMVEdJ{>_H6W-cDA+$VG_>`Qu+(s{
z;l#~3w&wdSUm5uL(xU3dP6BXIv&&A>{|i4u-Tpb-Ah@<x<;R_KHiv*T?j at l%2kSys
zsi(*I*ykII{9^T9-Ngqk10~K)5QIom&!3*koqRsmUqAw}MuvxnC(o*h8hG^9{mO;W
zQ7)WKGg`d|%@i;cVl9UKN}ppM{f`^8`1k*qnK7vy`1$iEkRVcOYCM()%Di#^&K;Gp
z8!Arw!|V1y6a)J$EGo(<DmsvplY^J&z^N>X4EKP^bsj2BO-{ZId(D01NF=!s?Krno
zN=$f>pRccXdd1XEzSoGw1!KpZ>i4U!>si%kp9QD$a9I-NTP)-=MJq(+Z+$%HZ|Se~
zr0a4FtJSBM`BD6yIu~Zsxt=_UxFwKING>HyG5B7|Kk32S>iKf93YWk)Q509MUY$=m
z;r6XHsch=$(@_8ojdSNt+1Uw#N(uvL&vYjbOGuy-78W+Qvm<wNb9<PaOfD+!)c^KS
zYip~yjZNd)Jd0H}^#q?|BXib>%$bV&E_bA?qQ&)!K1TLkzkXds#S`=j{r>&C-LAxr
zAJ35_8yFmX<jnUv5_U&TO>GoZ29hj1VgwzGV`PErs$49tsE|`yeEJ-p<elG|C1|mT
zh=`j>Ni2=A>``R<`O{UysV&1_6ysb5STqYzNDt+G-u8`&ouH+|vEe{L04OrAwA?>l
zoDDXGi5&-}9ICS}=i+=}%#(ie=FRb9t>>~>*4v at 9gLfr(+!P%hEoR+Ji&M<ut-mGM
zvs`)$tHCQP%bK5`9}yGN0EGGa!mtA0;lq0gAe2}*JJZO?Rgj<QV`OCf`9oQVN9Avo
z#Nzsrm3+V!GeB$Y`@0}Gl9L6ab<(r4qV69xx_ at R=3Vi=gM{8T#MusJ<VAGwucX2*A
z<}!k?1k#K3B{;TR at y72@-;Y0#z*CeHCk!;IMb0#eYy`TxN&p`3Aut3rbnI9#_^ueC
z=k&<^`}f5pB<Ou&^i41+=h)SD(nyn6uU^5j=H`l)d(7+;5)#U<v(9d$rK9FljOGkj
z-_zCAwYaiU$7JRoWctoGlAZOEvT7J*5J1uPpEa|5RSP>o2=dru4g1ys5ovuWt~%W!
zz;XqXOfjA6_U+r=_vU<`LR=LW7l+-dF0$`+nfr2lu-t=!n3&kMFOPe0aB%e7w|!n-
zUf$l`E^=}ChL}oTUtcc~IdJI1Agq&shzK^enVp>-{r;Ww^y$+wCr;GV)a;?AriS&{
z?{|RNY6s^^ccdwuHZ){?Ebns*Hg_zvQTwrcP4^4!P_>08Pk6 at sVpxygN=izKuXc2D
ziiPNFrZFpWX7!Es8ZB7to6-NrVm)|%WuA@;p+0lz(xtvHUuqj0_en`f<pJnyJ32TB
zFB^)TVqj$Ck&w8k at h={f0Nn=%;Flkqnwz5{AtB*#JQ;I_j~=&fpxCkQ!`;Ic4i3>4
zbzv5 at A!In+TWNABc6)t-1_xyJm3s(n{rS^C9e!lAEkzpJftg!dhk$7Kw0>3qlr^4F
zV6S6f@?o%L-ynP_2k%>2($~ggi0Yudm`Q;?gL8(r)|VcNSrE#4&I)*XN547{sdlTv
zYhnCzj!|udB<!oknKOF`B(HrWI)DCrUP(zf$kN3H_^n&25Vl3Nn4DWC8~octNL^h$
zOzd4ohD9jR{GEi-vv(?}sHjdE^ENVD{d at 1uom+o4*LWd9f^ph at DLjH5ivhGVuXa2Z
zJWd{?#-nU$$wN<1zqmN2D(}CZr^Vx`sd-!MV{BDbKxjg)QHj%0KFbn^{yS#H^>)Ck
z8L%Py{=?_Fz*v9({*8^D(-d!X=9($V$jC&n^H>#)_vKsRr1&>x?_;qy@<`y0!j~fZ
z!=CzDrw%YNWnSxkj=u&3<Mou$ilL*!sZhekrL at OvTC0<f+uLQ&JxELp1rOj75Fl-e
z<786rpYMJ_vyX-&y6VQ}6Ij&aZylb?VjJp}Z|AtLZNjTZL`EuGSa2sxUc1PDZD44K
z{oA*1mf@}X?-Ei{geqx&wWrDf2%OT=qWqgtlm6@(PW7n%@PRBgHrB<(<z{9k at 0Vi7
zmX3?!&uhUgLe%E(5Y2!3q>EU93CHKpGpf&uWITID4AzE?m;pCx<FK8_0?_i?rFzlg
zjZeq^FB{GOeJdjV;X{Ez8G$pku|NYYEuVZ+b1*U*kT}!tzjE@=fddD|e*c_Nxrpi%
zb6OdX8j at zd0l8G$-hO^oCy9RWogWi&f8eG#at=_XdSrw%cpufRprD|>u^uvP%C|UH
zheHSaSK6d4!|R5zHf(TcNZe=H8ZENutrkx$h?hLO9zkF)HDA84;{5B$4|g{n8O6!_
zu9C`m&Q|}L8=&Llq+w#(9nCS(6c-M@*f%Er{30JcnbS~tD4;W<KRhmu)mGyPs;W|a
zl$zT0#zd~Pv^3qM!qc|*4Z9NQg(z27*Qfk0O0aTZqTZ|D>EH(;(+8!UjQN8qR&56n
z?2a9q=ptpd(hB`LH8s(kz*|c(rlqAN=Fle=&7r`m;JbPg{*`S$EO6qP+VdJ3wA(v?
z%o&`Y>g$OdzZBB7D}3Gm7HsQA*XN)fZM6`H5##V}gVN4ca7YMIUtgbjXS(wDff7-F
ze}CN{&q$(s3P(rH(a?bWFiYCfZf}n4xVlPZ>48;a|95ykoCX66_~RzviY0=;r&s$i
z`H7b|lhw>-`a-d&aVf#MbLZR_hS>pXYes8=yWU=DVT}uV753;6o7$;U+~60zZ%kCx
z)QILs-pAo|t>om|p&=8E!FhPGwNMGY7FbWPLLC0VA0+p;MOSBM=gq{#M3<4;h6duf
zfs*i|qT}z??u#G4sH4Nh=lo!BMN?B#aD`{$W8~nzefw;O%8x?`3T0w`uywDp`QfG)
zvsJd|bu_si=SNRX&dQug5pUEtxto?2S6V8AHncWBwBKppV1<{qHy4D55hnE-5d~JM
z!<M}Y4Gj&Y*C+P1KROW%a5f5J%zNb<)o|d>AFxvSADTmW*Lm|M0iF+H<IT5kYr%HV
zy7VCuu$!j3{+czwp@`tK_5UUcAx74KtKxtN6tJ*TuX?VGRCv*3>E&SN8b;IU?iW_S
zX8Ub>UjLKRSW|xhnC|<X?Soh~Hq)Eim33JZ&9E+<X6k``p6 at hZ7LqSaGNq#ng013G
zQsO+9EExo(10qys)eXNw`(Ene;^HU#!Q%au-b*TWgkIA-ObiSJ1VG?I!~YRJdbEYF
zD=}IF-d<>#71rmE+s^h4>|R;nIi~`V;rH4?T%6)>Qh?0dq6^P}J+<(eKG;c3O{F5}
z?CMg6$M1Z8?x4PbLCwGalA{%Fk32nd*D&%y{trx61;(>Jl@=KP=n>`Bt5+#G71?;z
zqenw@^+X$*nylsr%Sb6HaTfT}M4S at Z#BAQfWc at 1t69IpIUY7U!Majy_iq?md2-Y_?
zGCXEJW62=6g`H!+s9BYoxj7fceuQ5kAqpoaCuUi90l-B{CZ-35QK7HGq at 9P+KWl9V
z=;-J)G&BeZ`uh7bz}a!A%mWdV(b?HnQDti_BG_{DOQ(^lnD5_hG1`|n4XFTZ0Q8yL
z+D3rFegeL5<xUAKwXo;+XmMhYsbr!O5RcRW#9Ve}mL{4sw6ui5CK0p#Bz_HCKQ$jM
z{@nl3(>7+Sp9c?L>!m;c(FT6<#<a>m5;m5?+4AU_=79A at EF4eI$%#P(?&_ig1csok
z3Nd(WbW|rxI at VzBEWFL%wlR23ARoBR1~K<EJY&dHhr@>t0};|>M2p$}O#vDqSd9lE
zQ2ffj46EO}f=6nN_ksi*9UBV*6=B&F%f6L&nO(yEvoIEq2?~+}vJ_QqMMqOQ43^T(
ze9oyWxY{A+zwUyq#8TzF-u%>m=&h-KXt2nhiE!`UjNII~e5)qQ->cJGIucpkmR45R
zx3_$-<^XFj?jF|07CDzMU&aYg92^`jUS8obF)@7^Z&_8;a561`$>R4eUcN(z5;y)p
zoSppkjgUa{_;Cv`xoCgG0ESNx*;h1(V5!vbz~e^`rmB$w;nPZyqJr0dlhE}%Dk^FT
z0Pq!AA2ujIckaM*$KlGdcz2WSlF(4%_w}K#vLU8h0QMkLd6JnqHa>pqLZ-UKWOD*A
z2qroCMl}eAQDy-f#GtbgVBlRz%*@R4>n!b_2kvYm+50s=B;+ZO9qR*|<ojzjQay&1
zn~zWJ-sT);X8HXR#&ap`?f_&7aDZODZ~jY>(Ba9?msnU>N?pI+UUd8U;~#4qo7(1P
zQdo=cgJs8X2=4T>0dKEMw0J*W?m#du^_aOIU&+qSel)t;-o|DhtX^qZSxs9T)%^VY
zqwEj*>~|hN7U6Svu<CX{fbNQ^DF?wxGc!&-J-xi*;`_<<ZEIB!I|BY6Q~MzyA>lI7
zNQI4h`1$v(EsUJj(u!Ubdjgthc5V(>e;k;0?eH)MkW(<ai3<%saLRB<NC=mV4D$=^
zG at QXmNk!FFWPcf((PGPTKnIz~LTW*uhi)KQF!)79QP{LT47xtixMXzSUI&yTPLKpT
z7{w&&YGTv`J6sFQH|C(Ux`sxCt>}q(4L<DH`1+qLS{4>EQBhGewYo5>iSX<nNw{wS
z&nQg+Blf~tXr(E#Pc+6%FfE7`t&KNOU>jYWm3ia0<j0R6J6~Sl_4f121ah{xv=jtr
zcKC9>#ABsEQ~v96;^O&Q7uNtnL$5)g65?}u5dHE?xraS73rl1vKP=Jj2vYC}tnhRB
z^V<U>8TRf4emb&YINg(V`3imR*=;maBnF at X2(}k8>X$RkA8d{E_35)@{i~k?7FYvB
z(q(<of|;4w9LcQ26>@;#dg>VQ3)tNW`2|>6zHJwe|F8b5gCvcUdOA8c8)FVm1a40S
zin)E$%CEEV=mhrPm187bXxBrotgQTGn3IPGfONN1&vlX%thlT6x)mj}7(p-zjXG$(
zvK-?$<V^EJ6_|MekBd_u;JIk&va_>8LL6af3!O(eWMyS{=^uDHe{}H;_ at MZ*+kZB>
zcnWR1R5ZSboEdcLdJ@{leXQ|bbj0H7bT`Nu_*~wLqkDi3$HV{AvUhT#2AKfF7EuZe
z#>Lh37I47``)X#Z*D-esL6!te at XZ<4p7{doPd60UG%(C3e`h?TNIUa@{M1fUBql(O
z{FliysW%k^pMUo3(BYBK$1s6}h6YP8>Dbt!^YX+PfFJ|zejlp9z66UkB%F&W$7b?i
zImd8lY=&w$Hfie0JZ;q&&9WtS^(Yq)(j}0(m<bJ(pn*EF=O$n_4pIPg6#CYlD&#4H
z$KY7758<?~ww4g!uCef1_u{bM0;tBUE1(VVI)sprkbodlJ*F09^yt^Ghg_|rKZt-A
zVY>-HdpcHD%pgUJ2?z+lw#fwr1-7Ei<{ovh)d>$D>ht?lOzwanY;JC*eeU at 2>6y?A
zrDu1(xgE92R+PIwPK4R1&CME=?6M3y8y$hL%Oqvr5|Wc|14kJ<tf*+J at o&J!@*#qg
zmoDwMv9U?yw_FKs$`-*o)3LhneD~>#A>rY5y}eA`&(GBX;9EdI1o;#x2+<HN4gd{1
z^b4C+J(>x4o(IlY+tfs|w6p}E!6?V3cN)TQ=Gg}#I5*96wvXFW<`z&zHn3MD(ZIkn
zGBblY6nsfD9o#-r`vH<AB^1MgE&l##4~AY1y5jNU$KO9bqX2jJnKW(Gt%h9=37LZr
zWnAG&1 at FYg^Drri?99L+QR7i2xnp`ISFT>Ah#Y|5p#A`CnIC>9G`ME>=_TUP*yJQ1
zpOc)dZ0*O7cg!N%n6uV)ncENP^B><F+C?`lN>Itl?(T(){GTvX1C^tXBq}d2$Dt&~
z#>S%n^FZA*fqr8Oa|3lsq`;9QWQZ0XKNvvx<HG|_BjvraHrv0yE&2F<IW4_l5$S>0
z|K$SUm_ at g#^GJkBoJZ7wCx}@!Qbw~Lzs>nA`xJJ$zeP;p0&#eKeLaGmI!hVnuzq}Y
z_BzNf(A~G$sc)9h8027wo??eR1X_>fS=_$020;vLPRRfA3*H|);Gl>BKm-aBjofd%
zKGOEn%LEV at dNnIk9r?~9ikOy?kdTm<msfpr&t9CgjkXpR7G at dV{3D^YRSP2M7#PBp
zJ7s6pzX5v<0$Ypc)8B4_)jQ2%8QpURFN)>7L~uqH=tRt4{D=SQix)32{|tkHzklVT
zoT4I!^Kj)J0%lg$|IQ3lcsxI#yjIoL&Mnr9%QrlwhLarweor4Bd>_cE8cw}CeDZv%
zEclC+Ad-B5XEgi#D=mybeRXEzA0``ttcB<j2L2MsPM$S^^IyQ4%y9&RtjFuVF<n$}
z)zv4l{`~OyM<>sprv`B%<~pen%`AQ^P0pNPxJu*u at H;<J3W~Z!q04yu3tSLE0k|M&
z?Xb8wnsU!MazLg0x+~FRbAObtMGjzvfvc+u>|5uvvrOJTJ~gNvU0AG&CO1~5O5LWY
zk+LidD_~Rvy#b3vK!B|mWMoja;lK_&<oWS~+{(%-f%j^3{+BvLG`;uN?|2^F)wEW}
zU*+Y|dd&|iTU$SkkG=srKSmV>6wKxNb!CW;A)J1Gezx}ZQEVsn8~uNLy>(PpTh}*y
zVK-7LE!`jpD6MpNNUMO-D5W%tN=kQkhafFs&`6hrpmcZVH`h7$``q96ykmU)amG06
z#kKd^bN*tkxz|)u!a;S8#Kha^EaR at Ot~=b^;y`fk>&$BiwS>@7)9BUAQVZhs^>t(t
z&;U_VK6`w=V#gT}KmQE%a^ABBj?`a#@#t!ZB=ze0`rG(;GFIKnGgq%(Efm46a at aTr
zxz~W+?GFuo2lfDUWX<;H#r#TQ#{r*;#=N7q3en(;m*b-7E85%JpL#ajLf1yXr2c1j
zZQ>3mXWeX93bHa}M{!xjAXPZ1UPSoGDz0zL_c38m&m%K4Go#$NWp}D~liBxM2I~}*
z=!I-F3LO3hRn?G8z^npGQ>^X7v-1iF0FY<Ry1BKv`PjnZDjy#on9;BO{Wrjzd at 3w_
z19neZSA|`j*mAVW8x58MPZi1Uz)^F!F0<TWwCIL|g9e44y?o;<px1Dj)lAptC}qe!
z^gG;+wjzEdeRBL$P<#9x7U4PpL0UJPa$9Sw2<T-%aI{lv`26_~_$Y#MFC3~}nm at UI
z;0$i_^OK;{>fc|!iH4uTKT2`fn1-mjfm=3QV%A7y&5=$CDgreFz5DP1%>+SC*3yzS
zn#=4Q5G%#Y;oraG^p86p34}yNX at b}QEj>`6-w1qy4x4~G1rZkp%=+t7(yu3X8y<c9
zzlwH7(vp%7(7Yxva%5!wnVIzHVnHlg43~<5ly9Xr+K}|jEe1xL=*?0=-P&LRAdfF2
zNB`&Ro7sT^a%UHpBv3u}k-0Y at H>l|8FMxIALyhb}a(~ggX7Ev-K at n<L>KAr)?}2UQ
zWAA^%ju^m$b$ADs5mog-T+1!KV5Dbc{0`xRrVhegCmS^SqgE3TzV$P{j`Vl$;-T#Z
z2s4yG2#*q6wh6&={lu0j)VfJT<U{Uuh?5~7Ck$(ahQ)#FizUZAwfhbr$q4G}keZGT
zwRd0wzPH^{VFQ^QiRQ4!@xBKnOM$yaiUUpaA}0^n+_c4 at eiV>0j|~h?LrOidHHTFN
z_y$T;ndEvR>&(N$gT at aOx*j=}fBAymFo8QH;A3u36 at T~;D_t%I)!xU^fH82Q?K9Zv
z_=AIk#v|qIrRIZ_{I6Rd$sMfzzzzxyPPS%q%##IWjZWbNcTd7+kIuh9 at 5*N?r64=R
zX)&a_cs<G|NRqnTeu)UZHiXW at 0OWrSEKS61?u9M5`#^GJdv(<vc=6X?j)v9k@@m&3
z-tXVPqj#QAe=+DK^nyZ01|9Mf at d8dDj6+uF?y4)$(9uz$YDt0qP1lU{bYUQUfZ`W6
zHX>lF-rxMZ%Ef1p%ldyskeC2CWw5ux!uFz~XS#cOfV5A8-G1%p`2EL^)5w5XTaQoP
zQpdvi3wjgH7f0fSjO-c<3kzgXB3PYU_g~yLx3H+c7SfljQJ{ym-BDvY2*1u84L0ZX
zW;#?L8la$zflxq at 4$K<~<_XA849E)fGv5mchcH(uVTfg<tlW<G&I8W>Kwa at scb{k5
z{>*)L9S1$;60-qHdU`MY&2Oe^j*h&~I^sWq!b?IkkJlQ>LCRr+0n6=|q{an$Qv++0
z^5SkoU3gd+jh~mx$+1gaW8(x+6ROaFZN^|wZ(DIOzWDdjk`e+?lf}jF09cyLzD3Ua
z1Ozlf!lNyW!~KN<5am8v)aLB!+Sb<AcZ at r4BMl5ff3`0t;ul3BFEu32(ZP;{wDcue
zu&LKP`_O2iPbDSj-OM|tz4V1Yg3vI at w{PEmLR%aHj`-jLQDGbn{sJv(J$5D}yS)4s
z at WMA>n)h`}V4%K#{fdEyfSk#{hlX_~K0J-s>Zkz37NpoXYJIbrzJiN~_XM5TX=O!W
z5`)K1m@^j7 at 1P8*0>HzjSM}9trP?!J_e(O^93<F4>(pHfV{gF!{=d1Y2)Qee(TFi|
zaNeLwAS^2eqphutCOoTT$p16q(RTJLRb8PpsyCv8^boSVeY-l at j29Ul{S+a~a2XqT
z{L?^!g+B~S7C}{^*K+NbhR#}8SfFYY;M at lf4-a4%3{u<yWCEXn{&x=!P+%B=rg)Q*
zLWL&(fzI(El`lbiLzcgJQI!FVR3xVfE^4=gv at R4ephP{NXx2)ec9}3`&_5;$f{lcf
zwEo)1zK6h?C`d7n!6X9YB{bo%r#p at s;ai%SnUQ6Mv(w^vi at rUNP4)+zke9D7p6mF;
z#2VZS#yvfqmYId+iK(d%@MAKnwwIa{qWUJc#n1x at 35m&)$FfJL;sYYyxG$Rk9is~o
z>`jVmzz8XM{`C24bxXr#=$xElvjO?Vz(=GTeH!`{ckWyVb$EjhU%dmhDuTxl1#3N2
znqp&P^YU=_C2FvM#P9$U)DbVh4x$DP(ECWlh07<J{v9sw<y<!n#Y?2VAcPTU1y_Rl
zDJNl17bfF>?SA)p`=6gtchR8>kS5XG7DziKf;YZ(=P^dB)SSUS(vKY{tSWM9q{0U6
z`JoPIcnm*AMo1u`>H{f6{>;pr*VWZ66fyE+k9zUHs-cmb at 87@o^7E at hEpxcIZ_#N}
ze0;=hG50(HoOtc#si~iBv{9?aB_~tH3c3)Xh7fcT at pKz8D(sv-cLk|ZP)RLR*}W<t
zw!Rk`wW8jpU&F(wl8GK~t)@=B?cy&Sxft&HER~G9Ywx`cVNK9qS^lm}<F-22-qP|N
zvKe(@qw^c}YhBS(2WVulH<mr4si}$D;}M|1vqgn|-y1CBsaWZpCMG7RuJr!>KgXT|
zuKB|AGR??VDkq1v0}TCVSF_d6Tzh at Ez5N>8)lCPMTLvWzaEpb7h2>5=tp6D^2!ayz
zR#66_ldPtvr-kF57#L9X5~{1KH#Ik-Y69q*>$h$-x#0L!djxO40O5}Qm+h~iEPC4?
ziC0+mnKKsZR4A+<MbVJ4r=H*EqpGU-(bkC2(I#HHd>o_w{S{`2C{%6$>y3g0jqZN6
zGJ1>46kpu$*0&`R^}Ntfe4X+a4<$!xK&WC+1s*6!4CLb9rKOKMqfQEwl7t3dhedw^
z4uG)&ZV&;|g6H)P?bQAH^aO4A at yRwsHi6W`Atdxhx)ZnzC<4(QES2X)JX-L>DvM|z
zQTl(^jLuapNdmA6OG;iqht2 at QQ3C0+XcfB`6cnufAnlM<udlC1)dtjv3^<6Yk#W3s
z-{DE2D0f{j7Z4XkuP&pP%^|8#N8(px7v7vcZ(US#iTBkqN#&~*e=n~Kz}Ns4co%U{
zr7KUX<U4?9-8BhT_ySb#K+hd at A;KD<nQaaJq?bXa7yd9P$pw^k+FnF6wUA%#5IP~b
ze1q8?Z2={*bD`1-TpT*a0aS<wxT0Sa?CNxB85t9>?&WSLf*>&7jlF#N5}q&?1|5wp
z%#mJQEwH$_h=c}63{LZ3ij!QA{TLXjs7|x9vqQk-uiBsub6VQBUg)G?0G|Ks>w;)`
zk3PJ#^u*9mb<Wunn;hNTNcqdta?zYw^;}PLEkQE?&D at L^bn$_|lIMOR>jNKq(h|ym
zR%(A|h0Xqc?zNVtrU#(Y*8m5fHU(0k;&&3dx1^*but0cx_-QTe?F|6t at 84OFl9Hls
zP?Sq(J6ujqu296#^v_gl1ZqY=j}G8 at 96-CXw^!8j8fhmmE80gJ9*)r`X^8Y`YHj_2
z=AN$pd59eU+}zv?d;5mYBoX#!KW`OV46CE?V7$%4L-^&@N&}Gobz<VnV6>2(Jw86Z
z1B=VY4^qVaf7N}|-O^H1{~$&Izb&62_z`tsfUrhhBrtRd9KL;|Y#Z7h_|Mz)zzToC
z1hX47JYM`&h^nZ3{Irw+=cs97u*k^E+xr_J&Fa>GehR1ruq at ek9)HE4cWQw-A<snZ
zA(O5Uab5uF^4`<a^VeT`Y7kPudkr-%2c4->cu2wQ?oI at B6`QC4*OcTYCZgl{e*_2}
zd!g;;-<5X2b_KtluybwI-_(B$W){|a!qes!9;haeOGd$^E5E;Z9_*}4!0H3PV%*V@
zRomO!1mb^63k(0=EW49_%IgA}fc9!dFUoKcV8zhT5Q8D&G{*c-(M(CQskEr*E$9=S
zR&5Yi(?6p;(4_L+-QGGzc6BRsRu`%-0CWEtAE&gx at 0S$|a{EV54<4#f0l?OQDp5Lr
zjpS=2yFp-ld^=m~<76*?e^ir2r;Pxg75>P=DbTAHe((T;lp2^9R852 at wz{+9SyEE+
z>n}T3aC|)K;k+z`=BsyhcMF4HM^h-#E(^#5YyBCHhR}k{%nJzC!AtmhoX4hwNML1U
z{eD*wZS?SoHbh=RD){*Lz-n={nm8&d>K2FL1z64Zz*DP_Y5ds3zWpy{k|P8L_r>4u
zXHlOrL<njPsHqXGs;=&WFW;Q&IS*ej!c#YCzq721&WA^w1B&d4k<oj!<M1<*9b)tb
zgh$chqWK{N1e(Bv`X}IT$8!p5#*329J$?EV)l}e$*Jp{ZU{DDFlKDEFeC&0;&{yl5
zb1KNMBD$8BzeXz;{SNqVz13ts3AEZ%@XYsL{K25R089L}vr}>LQzTJ^-NKELk&#sZ
zu_xWBQjnInm?S8J{y-=wOZ<_Smp_N|5Y&2jcsPoNk`iRFii?Zu^sY9{<3k|t?!Nv{
zso}rW6NEO=(D1yWQVVTb!ePJ>06=BwD<Bp;^72 at wI}!rKd~Yxq#rySXtgWrRc=__%
z`cx~LU0nF%iR{$m)Kn69Gvn^mi|z3Ot<CDW4bF#ajpg7r^7X12?dfyLJ{J~#3uRD2
zXEqdyWW9v%Mg;)$z(84lYi+$8LZ{FGcK^xKr<C at uW!qBI(*FKOiz1<(no1oR8JP&r
zV0B|d1d<i;B8ZMYC$vH8yvMP<{M#3v&ICx^)y>U1g#RzcML8`;9wF$UZ0+c1L=&38
zcOj1iR{vYB=I1*u`?d&N7gvFGNa#$!6(6ND*kf`nJ3BiwdzwFtx$KaFO0K&mk`2!g
zb=6-b;(Xw7?&=@lKM~Js&0GP&M|4XP(ExGiz`>Ra35C2=G&E=7k=7+X^g#KJLJ7D4
zAuecIRAqvw{@cn{D++mHn%7QEdAQLbgyL1J*cb~W(k)`*ANBQT!Mn2O>r^N|+pknk
zh$v_+E-88P?AeDPDhbpzv$#5m!UtK2+0LZXv9YlzBG9Dda>p%3q}ga>!s%HqmSu0e
z$(`|frCVCm7sjC*HZr1v&h7v~ft)HDmkxr%0L*=CV%{l31`qH`bh&wX?8cpmj5A6R
z{&I=O at hQ?d=5eSo44u?BUhhjOn#TX^*|XiL2z_+oFq&@$vJbLByw>&TItj at YJBQ~w
zy8P#IqF2hsTrXd}dezv}^l7~hfpsH=h1T7>chOeg?c1n7)dT$g@@VyeB@=}Nb4g)A
zXjW;frks*`uIRhL&(D8AT;S2khdNZumoB}o)hJ-y=};?M;o at SlpU6=@i3K|jPLUk|
z7;V|0V}|B>GJ+x_y&xk;svPi811v0mSCNs7sF)b4I3gbbw};Ml19(EkZhhB18gwKC
zD&xbapB4}?W#n&`p$!&Lz3)m&xHdO8AH%=DdGqGaIKG=`CV0`$=xlNuYiraC8^`zR
z>9 at CMZwWfz2Y&#hdxnmV&X0)el6I*%63fmmF6f9VG%*x at UT-fFqO%%>j(1O^I&4Bh
zLN{(P)*pz=58mDvYxbsN7&9iU)9$kmU&waU^Z)p<9>TS&!sa<FA!D%%YTP;5IXNaD
zDzm~C5fPbPSoqQ3e<Q(p{bF!%@MOHcy*)RojG{r|4Gj%w+wk9M=YKBssN1`qK&buw
z^T!>y=F#$0M25VDj9r>fV<UK?Rqzo=`JfX~dwM)Re!Pl~0i>p*Yvp1yB6&_TX(FM~
zcdT3O$b~Qte131v84ERLGn&w~D<aRH(bgO+^8v9uP*A|8rKJUeB|dfPlwrIlrJbXi
z5eqf-8Td%(zUJJvECdOo^Jsy-P at 3+<fW5BXKx=f-d<pnZG~|<;n;RWI2}o>iZjQF1
zK#8kH>UAwRvauBqndxT}0fC`mj at aaaqAy<D0hP46H<OS8 at O10$vn!`gpH@;;MXWAY
zb#i=w{s=wXx;oLihK8rj!8B;ULPOu>Ecnkynwn35^mc(fgTT?-H%Zeh4Gs79;-l^n
z(0X;(6gaaT{Z$$&q=E#Wn}q(3&-Moc2Mn=xuu?k#!Gey%AR!@H{edm?_2<tkXrOlU
zek~Xbbh-?zDWh|LnU($%&@7|X-)EGRl;F=5U)Qa^st(ARtunPH%NAB!xbWxq at VA!v
z6Xn#}zq%_{pnXt_8`>2 at 9RRLJ`zUTv-919zmEo0;Cx=vy88MgXJ2d(kcq_RsnQ?x7
z?M#!Kz#~G3VNAAdZDb#5wz`|-#(4c@<c#5ZSZs1tw|C}M)g-*TcIq0r4NUzW<zXHG
z00o;l+6 at WJ-zY$BrBVtC0rtxy4EFQ{xqv*swx`@c_R?%PGsZ|59oRxTr4|$@1dt1<
z_h;Wp54RjCe<%~qoCE+J^x?xfLL#DWkZj;;<w9d)neAV1f<2oD$q?4~dT*jCzjsD0
zuK-duPRLDf at p?{ZNQemEtL0RH<&g038(fwnk<WfcJ_J*lmYkeSq-|+uBo=N{_k~q^
za9%npoK{FE0nC{M0FR-uv82As76C+*?Bb_v>gnU_W`lh!8k9kbAXrV{DfhLfu#(2i
zJNqj?BT;M10e3PvGm{ZIm5C-b^Mab+n(gwy1b+T}*J89P%G$=J{#u^2-P)~C?y>LR
zg`;`x7_n|!SXk`Cb0+XfsWb@*326qc3i{~*ct2J1vxcfM2hvfdUt;_4jp~U=NMyk_
z<CBuI3|RMB46nHsXz8vb<XDa!?C-1Z%InT1IIWGJh1?0VU;OnDTtq4eOhvF>Sr0wV
zeXk at YCRTy}G1;8$+Dte(^n}Er=jDxF-`J?XR;!nt9oui~=36u$k(o))!OotwwZE-9
zqq}|tx8M2p>}6vR=Y|#*a?#PzXWVV=?3xD$qs0A5q)>0)(r9%IK!!XI50A9YWi&L_
z3((BCHJqhSzI96aWecOmy7JUU5l5+u+Cf)c-8)JuDx-J}Yio|d;o(*t*3`JMm5~Y&
zA|fJuTwKbXO4SlqyCBBgm~1j~wXg;8k3?NyX2(qQd;-fzbu|xmASVpX&1GdGSk>sS
z*`;M0eU#)GkYv-Vx=lxyAw9Fvc5keSlU!_mv!z7~z62`~fQd08EiX at wmzOs*D2T#2
zJU*WFP?yk7&7%L6U3D%FLx$cL>#0kG#Kamz0fmL^u!to;U*T8h>jM%+MMt-g8t4W^
zMM=RUrqS2e at 9yoT27=ii_hkbJ(guCr8!1Ye1p2ZE>{)Y9PY7rsNezt%$DL(45asFR
z<@XCkoS&=3;c;<s&F!vDl#u!UQr#ob=ou(=-xPfAFhgbKG_Py;{CN-jzUIb8PfkwG
zrje1 at 8>~8#Acds08@}w$rUayxKBUI=?i7a%zwP4U($d<>bhqu~u;b)9CYr;@N4PaS
z!%y{IUS)}zSuo=O4ws9*bo?=!*gmmovj6L&eJ&RLnpp7mk`fZ0w6c*804t<pYIC!P
z$YAYl|9pL;6(B~?yEP#dwD7H^Y4}{`<sILJ=$8`>UF9z)ROfGnYRJ{)FFN>>@~O`k
zDi4n3D65gusE^^15Bp`YZRPIuUH>XIG(SHPAVf<=Mb$<2Ec+)28nh6YPV_o at z3<IU
zfD5f6!;4r}<So*Yl7S|A>&vydxy&4Q?j&{y9>{jB5~|4T?d{=1v;%b|!E)eXOuLfL
z$|VTK`I8A`TW{6<`R4KdS^=n4MI9ZIshJrSpriV_x=t|qSr8OPH7XkO8|w?`43 at 60
zF33p<OjBc{MBQvE6Dw;vBu1wn;oT;PLt9(hOwcm;*RLxkdDL+CsYhjIP4$9+(R%n0
z%VMNFNUOwDY=3{B^VPDBw459PA9Ivfg;1TAy<O|oOP44?ayag;CMnP4FmsiV*a=E7
z-1X0%wwv!|IQ0NrxJpZrs+_7S{@%esNPa#mm(^&j>+aw4(lRo5 at Qbnz*dMvpR8&-O
z+RiGTdhp at Hhn3Ac{><z*@$oxBl1izlgt(p at y?P&!=J!Z7sIoyfzNp^-L>MT^Pg?qx
z7AgbftvOj+nb=1*-C%NYF<DancGYgnkQ}zlBRoUpFR5JirXtX9FZ%qM3UKDsD6QbH
z!NIOiPkzh}Th*rNezEtC4%E{38XQZ`-A|i*|8RME*>-zTv!$ga`S8%W^3 at 92(ec5`
z<lnzZ?~DoPf+Xp=xx=rK+<ySNg6bm)Pm%1~U_e?(`>v)SDia_Oh#@TWF)3<Xr3u!R
z7^@lzAW}wV=3IL%R*i8b+atBDDp^r5vZ6wVtEYe0xM=^~+E`spv|H#`GB;;N4-;5E
zPQWRgpP&B%#=*&{l`{dra(#1?8e?c?CgX`so<2~ZucEGgBaYu8cz<g_<1b%#`^x8M
zZIl`s8WjU69Cm5H89x3CANl$D#ZT)m`S|!GgIH^6YfA%F90XniD{=YvmsWnHIedy!
zn9NT0;US^Zh<23f60?ouGzTOh`Y15H>2k!Eu3Sk)pKfnYpJ)uYykq?1$B&%20FWF;
zMoTNTC$s_r)zW1S&1;J5*4;pZ7ZXtEU=%@tgeq(FJ#*0*XeA1=o-YCUj)$2aDq(v6
z{=Jf=CF|7mw6d`=z1>{*drV+Nghtc~y`!$%>Ys<FzOu2Ulwm6aq(y>`yoiA?=^83A
zn}Y;GU^4>N*cf<{({a=I?|YqV*RE}XhgX!8^lH)dXcZnRd1Y>%`=X<iyFgnl>9>4p
z<;;D`sn{4s6}5!VoGs=G3^LBn{4NJeWrok5i9_m7PQ19k at _%{(CMFPJ;@jziFo?r+
zLbA0U9<_uq3D>#<lVK%7{)9$71F$&b4m^|8Sv18!)hCx at Y0(#*nr-*4Xy`vOBZ+na
z9kD%BXmH+cw(|`}5iI&_oj0DUj!y2Tk(5f!A!uC{W#!<&Pjlnr%1kdMeZ^$$j$SKI
zI)36<FU6YQKYD6cCpA=Zw*x2u$NiF>vyoeYw2 at tCq>F;H`Q7xI-p|t-A{aW8oRoW3
zqZg<qq$kG5g#iH4z<O%=`^|n?siDGfx*qOcO87`F6p!fy4aU|}g+J(a`usI&0|NsH
z^xy?2zpR*>Y<ivfx?fsGoJ|JuNnLh+zp~w$@6%fN8l<79cpb}X%=|U@^Qc-40G5`1
zDlV=t_p)Y*>~YkJU`h6+yu1?@NR3u0ue3!}jwK^M)Fip=o-{~9tD>SJN^o2IyAuHn
zJUp6<pC&2)BP3ppl)vmPG1Exv(9kq}$}{hZWJF(E>+j!j;(oW1LXEPrLqqZRm%prJ
zg4jvT$dG{8RrAGR>)pG>EHtK3pr=%3`TGI}Ng81hkudI+c`%zFfH;gno2(Se8r3CB
z_i`<-hZd4aNlB#w{XK7w&FY)-d8j+VT0q;MQnT3+wlmvB)eUR{D&KKq`eU!jyupc%
ztE=Ge@|Paqc4bhPj`3tu(B$GG^T_Q!AU-B0CZao!F$i3W?H1H8VPhBA5_9bRiz4-3
zzq(_*l%I#H>*&Z{ym--evI%D&9Gvmt?%L%1ytJdEqp)UaVPR5|h-Wv*ig#NYo}y at C
z0<2TAzT1&n`do#Mw?xpE#?*$jLyKiZs9$1sK>=IOCKeXf8TV*T6JZ7h1`f82%{>8-
zYDg- at +_O^WT)Z?d{Y8bSa6m53KL8iP0np;{<HXJ#g5T3Vr75jtZd8<%k|H7=*U0$k
z-BwLy!@v&Y2LTse!01+3r+t5(l$3PF9ZkJ_Kara+A7P(2wX&k{yV5SlHlC*O2hh-a
zLw8&dL)4gJi(&4xUbPanu|Cft_4wrB(#dG>e;EXVPV18wv2Ft#TuNZD)^!6YJ(b9@
z6Raj?Y at E5uVSD=YX}AGg-5O_zgGZ#d0Z^%_-_`7nU$%bn;`;JSpttB!t1<bGcmX^<
z(I{2*S^>9Xo&1H^XumY>taYAzej`gZ&HY%TG)X7v!AqAfBPFr<*OMHfa8zU at qBTRZ
zJk==a(nhD1Y6&qhvF~FMEZX82<NeJ!PN!}2Qx8s#R at _>1MSgs^dJ!Ywur8Wp)T$z$
z9itvMl+7g-w%`?A#U at 7WcA{aFy<nC3k(cuR{YdyM(q at y8T*IXnIcvvTEiG>1;^(le
z^uKFqt4=-2#cm8q9`NW3vLo+rJgCq13n?~x9#6Bul<-4oaEwU>M1uXlB7yE7k+8}w
zrA-;s=|jNO)YkUV at 8d@)ZS5GXV&jKY;(nmj9zPjsah2}C#uz<&*8MZ;ZWaiT)U-4j
zXV<lHvFJNbPL-Gs1{ar<pg-^Z^2&<WVcjsQoJ`P}-j*zUnFOTDhfjXQ_be~Cu`x2@
z2II>FY-?~VL70Gm5={PZl{gp_-c?dw)%i?!KKsQ)%ionj;ELzw=YzqAqqyFj at 5{L~
zoR__oySM at N{5pmMGJbzLt*VOObg+m9E3qd-L8Ed^dh|t^^^|m0LIkrugcevf6IfRe
zm`Zbpp;?nl(o;JOI9<&lbQj%!UKH}-bXL!O^blhVkJs8IfZgsoIk`fPdR{UxQAlYi
zH+;V`g2=Kl*AS;vl|p!!A26LALMO}vJJR!~y+wrqd6Dew?&<<o^^_h<&h`Ng+S1a}
zBEmY^d67}nesjI$=g-tU?Xuv20BjRu<A0pW@!_8B#<W}-i-;!UaJX6=U#bi3!f`|z
zOHe|B0)or!rQwh_Z?Fh&-BJWnnH$h|OMCnFt*Jp!nn3!j^(2qIp&Def)6O!Dlao_-
ze}8(GN=Co5F7<*vWpEx at V8%x#Ikl<mN^7dtqN%~L73FiSRs+GM<;6={1a^YlYHK7k
z3&lmDG at 1qVhME~}$9tZfTwGJq&ond!2M0H265N74JWivF1^>!GPcLbt at 0HTt_;_wF
zJvH@$RS@^Rdl^7TO(agj!l$kha-d;jNW7)X$4N&;MOo|rLsSqCa~mB(8vDoLTsOo@
zb?puo`*JiUA&0LLbEneEMRyh%wR+7iT_6Gf4E}@4v^(`8%cD<;-7)9Z^D}a{Qb54N
zCqiOkhzSV^gJ3;ClYI0i;Yse~4k}}AQn^>4UmG9IZNaE9_~$79_ at KIKzgtLSbiCg8
zA_iP|ez!@v+wzGdHLj1hcM=c`QoP2CzX}2N7^&q?C4$F=1t$k96M!b2MzHBmca=<1
zo4H*MT3mW+^!qMFa9%z`MJ<H_v359bakfo8sEzj>8M0BrWMpLDw;oIJ{RBq<-b+ef
zKf&3>g at Kkf{d-l8wwoFb5s@@xC^$Q+sPq}@v_}=$=EDr13q*<31lcrh_!fm$-T(UY
zr)>Hh$R~NstS7l^pczkE#|i}nMQT=7mLjgYT**M_ at BG~S*jC4pR^{|C!U)cJTj$WG
zmKI7(@24js?`sNXU#1)LRXQKM>djP|oR~;?F;b+b(5z)2`a469P>t2BrBWP)CWgRY
zF1hUBG^V1ml9A-Oon2+ISsvVc7b*wM at ynDRv_`N=fhPcuM9;$VvAerlQbxuX?hExE
zhmqa$*<^~apX;Up7fK-~C%3;;Hs&Q+<ERk5uh`$BQml9P-aWd+dCzqgW at bj`^d`Hb
zt%U&qv)8K+mhUfkz-<l{n}poClnQDEGC at 9lKa5?ouD-qtlDfI0Lm_jex~ggx!j5&r
zFD*MO*Of`Cn{|xLdEhw*<AH3{!1#J`b&JZBjgC<vy$7>Pdy#UgfLaJ#rSVit7y>~|
zrikaYVx!g at peOUfrtokEfBm|SskC2;UM+Jy+Q(Vl9$Y&**uhiWHWCDuron)%(|F>#
zE*L=9dNeuT0xGPPI3Sswv547ZVA-z3KePEjS#*8JO)b6h_`^^x6#$ah@$qqLPL9m<
z^mOr)A0LQh`tuiWIfA{%N=%oFkv at vgHL9V_C!W%D)m>my>0e8o-Fj?iXmNV==xLua
z??Y`um%f&kElS5S8waD%IAS`vxsL4U2<8F8DK#|- at 8ADaXXd5268>a*NV670j43Zk
zJ&4Ea0=1WOSalfi;lnvnUfWDr^YHNSGw#XB$?s}ev`hU!TmXrt{4_of*71xxGYbn-
zZbI>E_cH8TVVse^zP>r;&OndhdbO at BJC4!^Rg-n)PwRay9T>DlbDePqO`kntg-0uU
z8xy7%O3q{bPCS52clwDCDJdyX`SqpWUo=d=CloKGOHw6)=sDxAtfIp7^BxV&dkliI
z939ua`#L&0l87%#&7}Z_BOJdeQUZ%&0&8n)Wm_N1_WiD~@kBn$+TNZP1EESq#AnZp
zelv^Judv>wby~Zy-t(y5tA%~`%$W;4+17D-uA?ByZD-m^lQzC=itx4E*KbNLFOQT@
z5R`)?qyysa?&(PhqLS$DN|7*fye`Zfrh8r8E{T5WuGo&-Ua6>_9@*aRuBz5DxF5Y|
zPSU<&;oFhPtwB at 2f*J=uKhLt6X@^9Ug0D?I*jZuXBo}Kcikr+jZ)h?>pQ97TcCTe*
zY(?vw(O`a&;*5bxU%L{Qnt$tw<d+*cmT`d368`@FTiI>8fWwUe<QaO^j;I#K!^_*9
z;C8IJRH9~~GGp**(>X6U_X39J{(aTGXXEQH(=swH-;0($3H4VY^Snk%>EhxNefPP)
z&RC8=b0K}@Qugq3k8 at XpwG_Z#sO=sAHBS0#4{a9p2An;67JRdYkuGfqcTfg%lhe_I
z&Fwb_*g}Vj(-ss9+Y?9mL;pd*_<!sa+?gux)twHtg3L^>@$vE5)xSTAZReENcXsHo
zv9V1%6W at GNfNai^2q4>Rrc3Aq=QamE*YxfsWeX-Z64kyj<w17U8r at t!C))L1G^zef
zmCJo|$mL>}1N%z5g&PDnZb<7s*PQR{W_NUQirA}CN)b0Y9{c=U0dDFVc&@_kjmx*k
zxbm|1wuh~zq@%~J)m$pF^RH4`IQ6SmD1)Lj+HlkWyNZ9~hB!z&Dhx*eu0F>l^GA<v
z{l6+BT4k1UsPr(?ksx!VH at 9&BgqWv@`}ALkq*fjkSeb9}E}94yM_SYEXzM0c;%GG(
zRTm at nW|=N~+qnrUcJ?5|NWTgVE;X{WZiVt2NesHohbc`JrRSQZ-eA$Xe%pn5>C>k>
zr$$RZPtush=^bk^v9kJOI>F*G#l4zaQtB2$C%x<+?To$OPd&8luF=<Zp|c>%kdLDT
zvChH4fvXoyVO32h=R$85R!?1^mZJ{P<M;=hNN7ZagoA?vx=nOmC01gI*?{_PJmtU^
z4I9PGLjSy&xHwW)&A)y{Ub%YJq$QLg*)P9kJ|DnyACS>_zBkLz)KvQG at R?Z6&(FFF
z4I1hB__P<VpPgDL12+yq9^W}i%4Hh58pRRcoVj?{s5ESJs^$x0!bJA)hZW&)!I^pb
zq4;b$jeH#nH#awwJ%C$C at z1VJG;-C9fL_MKXw^9L;HZ(%eA&EwQN}+hO(p^t6U}87
zeuvyhAmDuH(q%>g3!_w*d}IFZY3-Pfh=D%d<n*C$KHL95<vI3wFGz--=Ls(2m6esq
zst=V~5KB?xrm?0R3Zh!i<fLJeY?xrbjG#v9&xjMH%}S-gjuUAWa(_Cpg7Z04X2k9*
zH*;=i1{Y6RSk;>4M{u-Hg(=xv#&Q<5%M8vq7aqj#{(2 at WkTs1*Vuy6=msFg+R<>G2
zQ&UC)0s<va!K4C?p@)ZuEwcJ*Q`+keAn0acWeG;-dkgf1E?v5Wx(k(_JfTPfF>}wY
zg)Ukyn$sK83Ag#Kc587kL`PRQ3;4unbVNNkSG}f)&AV}EC^}!ST6%A9&vvSXz_cg*
zDpq1a&AyKpYTy#ZWP&iUo$aJ>+?ppKz{l@|hx_isECeabhNJqHgQ8;Ki}lu)7Tn`_
z2!m(Oo}F<=Z+l9YvEu}Vgy3Oiu0OF$E{n?_F0;xIblDFCguku<Fkn1h=WRP)cQFfl
zqDqD;DZ5PKCO-akOm|;j9>?<hxG at M{WEkdO1qTOTczw{X+0d(IZqCf2SM^BpCpulC
zE$XhPu&{9GD3;3b=GN9l_YtczEIh~BkWSyeeM2z`){WD8Qu0(w%@%l#&Fpz)0DfM#
z<Erl);R7Lb3W$|y&CJYBJ<!wB+dNpQ^%WHUkjrQm(&VV5b8OYObTU7LOLLE&h|5&e
zX?rmV!WHS=-(Oz!Aiwiq(Eq7PVN*AmiY-qO315*H2r(c}9L)O0M#{oKLF<-_QJ at hY
zgBG@`c$c-BZbg>m==$QyD;AHmvIUUah}V at 1&S!)AOwW!>3DxzVs<;ti6o=#*1ejXQ
zgRFc6L~>CaV&!kpqRK4o;(PFtd!)@=cHi9Gq-S7|%HKMg&naL;ud~wDsr9idN1>8a
zem<+q$<dCrTUlj!+EJ*Y?XX%|2hbs^s99TE8-qqaQ?*1DxV<AE$J<QuN^MFqe7|r~
zF6Sx{XQEsTx00hHFDi~18`C4_vTQ-Apcz<JLqo}N_ihSg{+esS^Z(%S&{S*0MNElt
zS913mA<E?hCk<?J!PLaW+ggQB&%&?zSP8I9n^Y<lLOZ#%BzO1O&qUB&DlcAq5~hmk
zr%@v%o;uJrw7AJM5%0}!*OsGEfNK8)<5s^uQOL#fSA5r=V#*N#N<_8RV)H=`;8*h~
z2edD9p4|Dt=Vxun5TF<Dbus0%j%`J&_g%Sl$DLL_mVRSXCqFeIMy63imsoUY%rsjk
zYu%xu$bljIun1sY#O-j+^XJ$=PY)I5yU0-YG~H5_^8O*(!i(m2C>kjoP<4_UTx#ha
z)+F4i${Ctz at s}a8;}!H8-F{fRGwPI at l=N_>Jub)I(F9+h9p6GCLQrtJBSAr4ULKX4
zK&|R1G^^N~CM)O?{i{|%%oBl$y>km6UldpYZ56$F^F~-#{kO5MTAqrXo#UQap3uo*
z2sn=H+{@Bd0dah<a53v!TLi%*_De%*Pb$0d^H#2r at k=^7I`$@uUH0+w!%q(?8x2Vp
z3 at vGSs$3V&qTSTqo-aINZ*Pz423ccDla;RE08{hw@)VCW^s5~<KA;oR^7Hf0xbwbR
zmi=PClwF8}uXzL)E&KQH-_0M`LV;ht at PR=JOiH4-O2!|F4!>g4tGdorlRdA&##X?w
zGZwlgE3THH##PO40Y2*}O$q5&AiJV%9*vux$x96Esp;w6;8n83d~c+tq=<kEN(LQ1
z at N3<EWmFD>%HX1+_+(`An5O=I^~i><t{{veL>o at fDkTu`_Qf9h{lmBXE)Um3rHuF$
zD9rp?oNpB7_v5r!o^N$BZxs#)JYMjss}qeDbm28IGYhG%7FZhmobhwEcqO^W#n{*w
zRaCPAUocxmX>8l26eXO5rPq)s_~>A(OZ0QQGsfBso)F%%)QNg$Y%=*T8W$yeWR at GO
zcv<ACKDfDF8kduuQn&0hZV<!1l}e^y+l%#eIEIzxW*A&wn#0C)*K}J911qci{AN+C
z*QwK|(_CHmH%*YRjEas9E-kf>I+svW;~HfmGx|u4gG0&Qo_lU?t_zS;4sG<!Ei7EV
zd5+w5mjWvh#IEGWnaSo5qhr0`-Rap`Nm0 at G`1msq=3}JDg<NGH49a`Y_kXVJ`2G9$
zMGT8>rRuJ>)WnajuEL?18x0K&XA+0X)FJ>{#m)M2ar#RpCAySs7*l at Yoq)og`~Lni
zf<rd_npAgp3<dj32!G{Tu%1jh81{7e<*`}?)IeEcH5Tt)7T59lc}F^Wmmu%67(tcU
zsgd`WZ?+s(r~0$H{LBIITJu>1{FAJloJpM*4n7Xfn*2t%7%&bgpS|CQD}=&BLqomg
zFAbvkUa^WvNVIa1t0gwoo2%Wd{})e}hDxX~#jjRIFKwVVp0E<H0l*OYGB2@<MNqXC
zs)Q96bM8-uC?L at d+M6C6diqM=8O_p=<gEur_$(TIVKMX5EFH(kCyT!wbmWFlgjO}u
zy4`=Oe3*^CtD`zCJn&PjDUxENQFhvI2e+9YpCa_<4;sUGMv-!kG1;UG3`AU2Zf?7A
z8*=h??Q^l)grgmeDW0C6cLrx%wl&R`O8WK83J%l1<<69ddgJzHsnB^Zy{XEy8C5MY
z3r+4RWVT>8sZmi@#tmas^9g6x3~n9lGG7_1rMq|U9-~V76#`bBNb!$EVhZv6IGnV0
z?H-x|Oj??4- at d(3R#r|ut|>9?YvIZtS(mi*Clf#w%5l1%GaB2xPx^&$96EIMYBl at M
z=hoHLP3~8MSMGf|6YihNKrXhFDB!d``Fhllhps5~tY+&ld}WBg|20zn*LNBp(fl3L
zt#*4QRh;I=ngc==2LpNGn1Ar{B-XmhX~(}kUO)zowqZAD2#kv(4P({SiuDKSCUNcB
zHBN(uD?`;z_kh>fItXSt#kQ>(%8qZCEY$myKL_C7`19>Fl1^LmeKLoS1u&?Jzq2xS
zbA^skgibzI63j*#t8S%Yy$?bCfky75AV7S|<Jy3LfI8RIk(TwtQWLe0 at 88!exR^V1
zP1}YIm++W5JM=xIu1?RjO=}sLT3nB at D$0-pnTfDH-y$Oy-1)^_b_~y*JAt@(cnlzK
z<>UFKfOu&B4D{J}xz^NZYiV_lxt*}sEVg79FjVSErT%gntvH(Y at t481Yaz!b<G)^@
z%I~yoUJ*cLB^Xjh#?nHbo}P|m=IH3?o{x6)b||xn1fk1(aHh4EiYhS!Bcl(aTK1-l
z=9LHhZ}^-D1Bo?THy15f=;-L?JCi(A6CQ3F2pj+X{vH(*(3W&llUEHuL944~v&tx7
zkA$-j!-2>xBbFWNt`EiGaaB{OR8~=08SL%d%r at 6nSNHDO&q#_jinaCP1&ur{F$kbw
z9`WrTTMGkO`n9g84FNoV3gnxzt!;7V6ZWh4q@;2nXwdQ^GzQ4%A$L1`2y*2u{~5(@
zF1230YG4M11IDMa;+DhseK#UXm)N`+^(Ts2t%lV_hXiM;#skX)X;M+NBqSv4mYH>@
zUfkQ?m;PN75pFD2Pbuc32v?b>QxWV=j%|fUc=y>AEFoGBmy_c-Sy@>yC^H$>4 at cK7
z7j2%M+KLFHY*3LTPN8h1G&LnmT9g_p+IBOdvvBiZ48?&T&eyAtLBY{5F){Ii@~R<!
z7ph&8t(<nrVz_iE<H$&E!^iR-qvK%#^-y5Kqy_f0*buKf<0Rd-TWVA)i<t{wtxYw)
z05=DdSjWc at du{oHR;^d<d49N+E8ZKZf*>GnYmE8RNWBF3bj;~4u{i)<A%SS9Th9yP
z#BK1#_3ICKd3lxZB_{~E$zw9j2aAAaZrzf^%pUG-ppD8?4>A-IXepD6j9PIqj%(xZ
z>s}WLO+kdMWwg{57P4arog18;oAb7kHu|WcudkoI!qA2eeZpNe=nJtLt5LGHE_%IQ
z$f-b!do?l8Xnfieo80(lf6LI!ET^~OVU}{*1^2qJNl{Y4y}h4I8u`~T*@{VT>y`)!
z3IAzim^!}FaH(Zuro&{g$Y`97DuP8jyx}9U#7#E6-c0X7O|>lL_n6L!#sDM7gv_78
zy$pdzM9D2imAO~*4)xeX^=h1@;KLa{1Wc$3Ut%#F4sHN{v`{~A(&XoKKYjbcPDc(a
zMf6CNU7t-5Trds>yju&`4U>6;byt5*2Z`X+RBEkCTLvI0<-PT(=Eg>e)pe=p4QeZ%
zKs<*YHN8370)y6dr<r?KrH9Fye0gW&)V_!_aFDhR-3Zy{Lj}c_2)2T)9A0WVx*p)b
z=I-ubxU+if7~HqKGv9I>UG_H_!k9GtF^XlDzpY~*hYQ$u`Mlt%r;!n0{sJh2wx_`_
zHoIKGF?kxaGcvUmW8~&zqgT*MRF-;_24JU9NO1K^xM;WyrHoc!NtMgaGAY)sk^3XP
z-B%hbW4gE&3kfn(QaC~sd<*XcrOMs%AEOQV!Ozc|fBf)>jf=|y2sC{B__ll_613ku
z&i4S;p<?gTO2t>*;c7*_z638jL~)GTU@@}{8vV~4zNXxJcM&fWFlKybVQx+dcq$82
z%H;C0{KJP2@$v9BWSkvznW(G7$GG#dySln?3h)`O$|-1oNW0-y>cu%Qox7H$T&X;@
zU~TWya&N50B?Y3?=RC(pav{C!9RenGKO}Kx*@oYGcwF3FOY~eU(i{8u at uSJk@^6mD
z+}!=><b^O%pX!;oVXOREZg=YWU?KRYS-?VP+<{TT3b2~ir|&u}bR{1qfu;hlRh#-!
zb0R!lfJSCWNj#yFIy<%>e^A`<<X8ahKxV6E3BSKYkPN}ny*|}iZ-h#UM*a#z90F#V
zu~AV{My=sAuU_rw1jGp}Z6b8nS{bdj83~X)UY7 at uGablhd3if4cPgHGD7e2UxJ)O8
z-QXNn;tdup#TPH`(9qGTSX*<PJ%3(hW4i4krY(|vB4VexCx2zQY(>Gur+H>+xQx?k
zl%G+%l$r3>txk8W)3XpS!k+b<s?N^*U^tZ$1YJ;Z9n1`}nZ at 8S`8C2+Ll>vCRaNo9
z_`G)~XWFfHIZ&^X!!h8n9!}3JeeJsZC1;(Bl~1hL!f4K>-IsrhLCt!i;VRlDM4Jkb
zSj}y1ACr at l@$H?vqz99dlF-KIDOcCd!O?1``nk=a60<XhmAuJS6F2d at Zc>|k`Tw9=
zuH&bwBYn;0?y*E^qdLg_CQzrScLOX8hunj9dF6ctS=rRiPDRt<QWmU4fN$Tj=HThD
zE=$$TU45xy3Zdp<vY)=ib{Mr51=&gf$^Nnpz5%A%f%TCV;{QKr!;C*<FflQ)o$IDS
z_GxQrSi?D_Mk0VrXo+p*eI}T-Gl!1`QA_7I!Mv!#V&JA=mx>oGqI3wY93>tep0u1?
zT4*Rf2RC;*gGzcQNVyAv&Yxw5#EWOAUG&_1?qiFJMny%9XSAUch5Pe$nU1e@=UP%a
z@=ELF90HqtMEV at 0+~M4uD)1$zap!GbyZK(W)7GCG*p~FwlYN}FGg@(Q;&^O6U=&-!
zSvC&+d8Fjzj?%ir)Ibzr9opVVuS77#q at H`dAW$b;m4QgWkrVaDJ$3En1bYOzww}&x
zXfm5=9mQ%!b6zmv|MUWgFbNI#!sQRux(V^zzaI!Vp_Zxm=2=^`7?6vyTMc7K342U<
zIIT*`{e$hdk*O>`cgyx8i>&y43=Itf0|KzoKS$DZt~<>dn_1;URlCZ=u+t7{Om at +!
zEc=3}&ITzFl#-HCxBWn`m&Yo5h)X(X?ooefWbqi109USDX&M;NX!)Er2)H at hnItmK
z9aK`nwKP^+`(WYqeC+wpZp$9$u`>;ugU`6{Z!g7|4XcSpg%o>xA>R&iZgOQMp<qH1
z&zG`bufif?%1*N*iGM2PdWfA7i<Q`sqNYYt__>qi`J*o~$reX82ir?jw^s<-yvp6=
z37mm-zyJ8r+|dz$3N-*s<7$-d<o8-hNl4J9YO&b>Yl*&{_E3>@zvD~Sm8d3_ezez6
z?|ah(<k{rnqAUR3bs-#5%;2wIowHpj^#xO0FBIa<-WS^mnD`e+W9W^7OL<e7qy$~|
zMZX3R+$|#49)bup*RcNwy5t*`Lo+k!Pr1rI4*U=nMt~Kpq1m)VrqjRhC3jCsRC9~V
zNWME<H0X%g$MUpf^0mE*1h-=x>=t4H$KvnmUcq5iTgFPqojpA%Pkwx8Cb%19zQmx}
z<DY!cOQSSZoW96 at MO4(%((=r0{Seg+#-7jN^&r)5V;tm1f4T~aTJT|t0dz at rsDf*&
zdIkpQNJ&YDK0W!twnqOnxRg7c%b}G^=_ET<N-z054ruT72n*0sU at R5P2J*vr#A!99
zr7tawRH*Rhrvj4OF8*TwMxNfrsR5Bm%47ZR+*Kk`fKB8`<zl$grdlIps=w#(TQiP1
zZq7c$pvuGe06q_`NWbHkTss<*q2u68fSL at 0DNkmX_U_;`!89Mhm8P~fxx-KXNg|#t
zhk*`Kh`+xE6h2ShjNGEQps6ve_vtYfR^mlG+C$qb4vlQ|X2GTZB204QZIh><o46x#
zkMt>vw(n&=C44#4PP+DkU~zHLutq{I)F~Q*KmaRII*id<KZK@@Ddl(OEI*w$_avRe
zu!lp?wsX!?*Y0O6k*RW_wH11MZ3_-lG)C<oLrUZ*%y`1I32AJ{@MbI;dP5xjW;-zt
z+V6GgSG|6SY5Mi+(P8v#Gc}Lh{<Qji-*Z=uY9F0Bdv<EK=Ap+q91N;u_T}rw at V;7R
z9_<I$ldWCGy0^P~9Ye(b`i1|RVeN%DUOOL*A`qjuS6$rD- at kux4I-wTq(8 at b8J`><
zjf<xkyN^`bEu`){E2YqKa43&&gN_M-pYoRsrk%D+Lt)#?zhh>4lGl6gqeC;W65pIY
ze{4Mh|HZ{b-?NHW9a9NL%C7tN&v`)AGl%=sLJjZm2??dh$jA&03`9WNsTLVZS-*Ul
zl$dxsFETPNEc1}-?%fm+b<#&QBQ1T4!>Sds5g8(q{R?PA(`3FkOEq1V5LJXhM0a+j
zNc1^=Dxv0ylTn)BTxdH(iOLeKz4Yxh0tJ=tj5%ETu~;aR>Dr%F7gAejHR+M7OwLRe
zvv6{Tnwy(vDkceYT8`W=d1S<L;1FCw(v;)RWL4S1^m}q^u;_5qN17PuA{{smFKF_A
z;|uec34|B2##v8$X$K1Q|JhbS5SOF=u|H3nt`JYVtCt5`dVl28r#mm{_5vwHTq{hX
zT~4%0Os9w<a<}eCp;t>#>muA)tIHLWr%JZJDKnOes#q6zLerQ{etn|As!-SroQiD<
zzG28>CFH#O05jX2CZoK>_PzFt<H>TcvGz%<alM*?LI?lO)G<#V0UjQD`yCbaf<Aw~
ziyH81b1MY^_7MTWj}o#X6f$A}Iq^>Tn$3<K)CEPf3=G~NO-vvbIA8pEkgfSSefM>m
zOa$%u^XIMY><EO4*%oG`RX*TOS#emn4i{&OM{k6=xY@~SbO&!%%*W=d?p4P26yl57
z{%+&_D|nAe-0${-b?G4r3W_t5u6OMJi#B2dOy2lT9okAqZVvTfcxoNweD;giv8_q(
zS?bi#89)ChrM}*uBDQxL1K0XLcq4DWx1<sw8rBnC)Obf!Lc;gF5IL`{<nqL%TKaPq
z-+_Nyt{q0r+Fb>~I5dL<0kh-EI}_X<Q+3qhd5sizW}z=f<D^)@wTCsx{vY7!!k9^U
ztb?wTai5f0D=J+e7jy<=0g{5Fv1#3Ys)b;Fw3-j~RkTHOnSk344GyNXeI)CVmY%K*
zVzi>X+)KkGN558(MXQ*8%%W=%a3@~CNgOm9rGS9Inv1q(0Y%X$vC1=TouPx}Z0C=y
zbC32YWD-6X8)hkF6(MFUG&qm46!do1tCdmPg?^TbfLDLMVx79Jr_|luy&mCW+MD_C
zYe3jdtj-`R3D+UkmDl_x`)}8!6 at p8do;oKMvoks}Kedlm6sP~vMq_qzGsyYqYPOE}
zne*pE^72^V^1E06{-Eh!vkBIY#+U2n=H$e$rIm at +5dWey=- at BOw;ocp^sP$LsYZ=Q
z$b_nk+Cpqw4R6XSU*qBAR_<gJaYp8veNjXFe}M>3p_}#g*8t-FJgt&sR6LKlXGJd{
z5ckx;0FhR$>Q~XLt0JD)q+%DtwEyZP37<Xn04QQ}Z{xYNv~)7W`bqIy3H#wv3o4AE
zm6Za<7}j^(B6*jDgyaGSAZqefKw26NY9Xs}JCO&ur<$k5bqduuMMcqW-YM5BysKMV
zX-!R%(lRnBu#Ar2P#m0`R6ajzOWlv}HmBP!Y|(;Xu0IH8)zz9OU=$;WzI>B)a(@pH
zr?<#RW~j>HF4!8WY=fBrunhI5503X2*nB>IOaoX&ZVK!6>C>kjj}IN<8%;nJlHPye
zo}7}x0JkaVdMH^_Q!~za|K+%-_sW+Se@|iEe*XOVS+~0lhIlt`N(c%H_Ey at Nfp1P5
zulMbN)t~FlQX#shsjd#IMTb#M7QMK>xtTiCk<hcVGR6S8M<(PdJ32Z#e(^13MOe;Y
zSr#ZM1uCUz10c`1Q>RX40=u1Y&sNWqMsS+;BvZ<oHv*#W_Mb+dn3b{G3|Is7matl$
zmRh>_%k3qHKALbgefh5e8JCyGY7=mAap@^2-k{&~dTsoSyL#@UAe7MLDs_N{OFulZ
zIh}W3Ag9mA$G5*T=4QjsCLBD?->h|QQmrxUp6*nb<@&z4rt8XK`qeInJe7=VAm(Wr
z5he;)Wq1D<6X*7~7C;Q#d(SoKc5);EiigI)z<^OL8^5BWVya^3iEE}4S?Fd3Aj7!)
z3cEqWBvD3APEK7pGc)sKw&#@L^v@{Noeu;MzHroCm)vCLYB_A=pxbXlFQZvx+Q)e6
zLA?(_3RsaISU?6Yt}tO?;n at zM6KcxAc5h~rhyCl^JUmgD&O(F6RKc6w at j}PH)plM*
zh!XG5h>q5}DIf<0no(-&I^A-D3LD6CRJT4pJ|0&q6Wfyb)0oH)s29y`5#5qs$zq{N
zBUAkdXGgXuoC|Gv^<*f7L`2;Dc#B<_|EscU0|A5bJD^$_CXM`5Pi%5!aq)BYAmGru
z`yP~wOH1{Q;ptbxrbi~#w`a=aa!mdF{EQp at NtIU$uFWwPm_0%}XJByfadCwU3k!+j
z{y|a|=A!+H0HImXdz*iMoR?Nm$S5zrk9Gq=n&4snDPJ|>=w{@o<%nm<$0etxQlrfJ
z`t`v};+N}_%{(<7gM(2mp$t?n!(yuFb86CO9chA|t~fAdpRg5_TPL^d8NecrcXZ5+
z7VuBDJ^u%CQ(pZ2zUO%VGczD=w?D`JzY4F3TN(Id?r?7-G%`|Bzt%Mk64;ln7>uA!
zwx?f{HUi+2!wm%hOQ-k*8VlFJVtv>BFl>_hx5-}AT2D!@2^NJr1sls~wMvg#=K(RD
zaYx;H5CH(Fs2|hzXy2}=q=e>A<E6I2=Arxq9vj-eY&Dtb-AHqk8Hi0pL_`=h@=1=i
z2J{h-B>i|w#0tsSo*<+E*K<?<rfGk!6smNi<Bkq?E5fW$qZVcomEMc=MGUL&fz>Q8
zws4bD-6M)nVO|d|>EPiL>&@tWh<0HGogeXgH-G!~hE=bMe&Z$NIDP!PGwwP%I<AAy
zC!2!CmnthO at rj8Kdh8t<C!~6AQY+|(P&S;G{vVf_>G&ymzjkuTN^y&W^6uS`!=s}t
z>!}u0pakcX*xTF7KSswOM at vJaWNb{23O$&>l9H0wG0Y}<giGRL0ViRi#Q()&0U;CO
z|6*|j$3|uHbOVEFRb1|rs-;nAr;9>t8KH^-ztk6Zj1aAf?<F-UH at IWm6(QSMMehSc
z0X_NmpJQTT>UMZ}ct8`LfDp)VAs29 at 9;J1}&6#DRq!b?CIsM>wrQN}lO<P-=+k9^P
zohQQiGY^MW7~1TE0IGa<bSrI5;tzcPc|l-(bv3ZO8Z-{tv~xV#FX?Smr3T$W(vg%(
zM?S4C6}0)>$Y at K3F3`)<^E}pVH#fI5=X(v-8%GDPcpcWMK>!9o^k+hp^=2rL^SU0E
ztHx0bjV}xo^cEUOWVPy04xUTDo at EKp=c#Df=<ny3T46JzTwy&Wsh|+xxV<R#yggRQ
zx6gN-DZ_?A%Ypx33?!AIytZc9j9P at GloZY@bE*B$pFbaDkpylH(PalPjN24h$pA&5
z&b+87LbZIIC?r`M2XBUR&Uj0fTHh|<&99o1j^;JCOkinUi5n}XSaJy4-f5QCT~E)}
zO~DWIH?r_;-jB=8%{AVf?ZVx8F64Z;8+3#D(StPUu;lOWFZX0BrOY;m&|NwqvFq^Q
zaaji4)YR1Eh3O1sP_Z640mLx0v^;K|@gn;gIQ%MD?AbF~x!C(b;7pA{d~nq2M?b=M
zXzQ@}!dx(;MmCkfymx%WR|3v(V4}rZ-casxQD~W{=#3jUCYP7vd*;ID`D8%+nVu{U
znuwgldKr(a^wa6jd^nil{LvUdJ~vWfGdVXW1q7Tn=5`Wy<?7XoyDx3Zq(W#>zyJF7
zb~?z>hX*yO)$OgyZ0&jGfz$ELUtX=CmK)UXl+_YlwapG1obT1Lmykv at hibG<h_+ at t
zDwCOXCStY43uw<Dsj=&NOFCWcNw-a#A22d;Ws^FPaq<hMk?969N#nhSgWmbE=2x$F
z+?oYNURy=CuvHBfl4FVv$Ea at rU$@}r(g;rTXh(A9ac$Lji#Ov;K$WXLW^-*q!h0oX
z>k-dVLR{{aaq>wkIojEceNMR(No^5AvovY1H~Fs^BnlcIlH?&xQP0!DC68yGyI&DE
z5!+&$d?I7!>guXm^?EJgT+OuTkhMs+>)w(En$tQrH;04Hf+56d)?1cJUNVsjw`QTH
zp3DdvD>@-0CRRZOy~f})-GuOP at u`-uRvT*pmDt%?W7H!=z@`_AiV+QdgwihenGGlj
z2?-H)Mf2Kazg89o$TAr!HnA4^$y2t_3oDAEz}D6lz2n944+N=Sw^*xM{*trLC66;e
zi=(+88>W<?+*q>!=4Z`NT3cJ2*KspbmFcwaBTY^3jxG?Bu7fqtJN_T8{sW%N|9b<+
zg{&kiUfC<7%*;@Bc8DT-l)Va(><AG_GRqca7K)Hv(Xc{DqB2S-d(`io-k<OP@&EmL
zyg!dmgO~fhpXXfXy3X@?pNnH at 6_(O!QsT>jL;-ypLu<dmtlDY+pd4X|L$1Pk{Hb(Z
z9I~Rie@(5Q?_YEaB}v-6O{cA`UG3b<bL7Yo9>@TBDDUN4(|)3mfirXOA1XC!j!0F+
zU&<HG7Mo4$>$$egXK+Ov(rVr-DW(34fCqO*LD|bD*Szg_E2qv_r>+J*EWOZnw1c;-
zp-V7#^InXcmt;}7i4SQs8F9#1GbJTuE*<ZQn^94__a8Ee0q%{>wQ9#V8QJ$%*?<!P
z4=4l at a1)4=9`p0%Wnc5B-rVV#naNE}$5Ahhjg1LPOG^)}#d;2!J}?n_;qmR;x1%9S
zM2wuA1`tvAfk`FD#!})dk=C%V-G87c at oH*nJ}|4|nf?AfIVp+meVIinN_Sb$v72)s
zW3Bi=<q;Vs at o4bT1iaJ^3R1)c9xAsg3^7nq2wE4faqf+POT~sI6r>~vHN(D*`j8#@
zz3lS0|JQ5fF*Y{-chuLSR(NRti)WYSb@=)Dm*+=iqe3`4FZyh)df=J-6d~hl*irQR
zK8S}R&;=#B=+O~L$=T*`l7TV{S*TrE>m$Y$*2nNTVN_(~c6F%dSJ#uHCRM0z1-Gh{
zLjNG?DdIR5c=O&p4(zw%@dK~%hNO+BbHY!2>VWw^IQ at _b?$#Kf2hYBPKse{h==4cc
z--_qp!6;;WfU@@il>R-B3Y+Y)j~_$lvb1z`cqWDfa&#;>lUIFyZHDA|_{|M)Lru+4
z at Gm#FR4P1Xpx&2Ca{2v**e|-KeFd2f$1U!BQUE=`#}`@eJF80cOR=KPXu3ftPiIK*
z?CKj4kyj&3$twyT!<E>dd!I*5XKruL<1tblD=Z>{J8Jd*3nxv>Eschk-XGidZM6=t
zo6dkbQ_|^V!q)SEr3n0Yc$7$nQT|uhJnypF*Zl_23t#YUXlOVWm|a~ZI5+mS*6PmC
z at 42jZUT1$nDKW3RSh&8Gk(P%0p6p5?mgPP?Zss=&-G6>~EO7MbQSEKdqn|%n?S!uV
zsLl;Nsa`$es;nGllHI-Ilc8HZ2k`&Q1t`+MgVTqPNJ|6fXrG6mK at fl0-5uxosmaQg
z;Yp#0!}${C?u+lv0p?GV&zu+GTv=W&4%yipDAZ-ffPfX}fAQ!Wa2|ZS58p6M5j2We
zUS7s%!eRmJi}ApLjqbH1;D6KRxi4kP4qdo-F%Jr;cH{FNAb-CPy^+?->`QrIlX||y
z;QU9Whd<Wds6YR at JlijH=+L2|cPZx}Sn*s4UMR?s2S)j1cuJ}%WarR at zokQK@L5LB
zLw3$7?os-W<em4NJ~<e|q_O(<mGRpL7ysNWYBcIeu8uwTP?KTDY<3S3yJCQ0rzBUf
z-mO~g%uBa>?~BGBx-U(qQJoYK5m9V>|D;KQ79Dn*fyVa(nhQ!u7|l(+G%jM;u{j$A
zl9*T-B+I+s3c41aS|bK at J6*eW?V1ey1?d^9wR(K(NA<xnp$C at uCd*#V<9H<UWpi^F
zHep*^BTmE<R1C6 at rx22b=Qo45R+9h?EFb1P+TPe)Iea-F;9sYIcsQ*{Fa4I=f!4G<
zJd6A7O{V7aomJLdIiEtdZor>Xr at okJ|EfRvChKT25x#K%KiDlsc&!t6OACLd+<jm(
z#dvo5PT{+=2Udjqem<kwySEUtDj|UectGOnj}K}<`Y$(mO}58Q`UuT_R)r*^^U9X-
zFi&`q<fnK~Dp=*$vh*v~#DXcNFY6ceMR{Wwl{_uS|A<}QWQ`f9h{tzfR~9~HL(0E=
z`}XI3@{5iKXSbL|Ez=<QU4Skg6&2^w&oDjn%xDBScG~6MYuoC4SFRWu%^UNqbQ@%x
zl+w0H3VzSCk^WtyAgiJL<<^Dk1TO)%qX+d598~|rK!-Xzg@%qUt#5W_rup5weI#Vm
zwDOE)UnVDsIXO9TtUCt>$B{;S3k;`S>FH4elHF#^4b;}tQ=jcClmNhYAFfoH at lLxI
z7at$_{J9bu3<E*P2UNz%aA=1DctXX_!Jr({<?E98u}Gh_;mR8PRJzK?kK58D9B`K5
z%=hnymw%^SgtZ{Yapa)z;S0F6sFRYC(&YUP;iO#8kM%A6{W#7gqCGDbo(d7s++e|n
zN;?wZNn!7 at mG-bx8A}2x76V!<9y=xhL?pM(Xis{2I!@8UsZ|pbwq_O<4x7ufID-W8
zD3T`e70#eWQ<sLzzi}qZ%-kFmJN;pzpFe*tuC8{1ZjYjgqKw~~4(ZtO84-|_WPs?>
zFf$|I^vLPA>^SNe7}nMB at Lf>yqH*Tg-?^cReUM7IZEsT(Ih7_yYTPJjXy9LDy#dN`
z(x!SWJr0?1cXx09%OrG4ea8EIQMt!R0}dp?2}lo3E80IL?azwPG7YjibMMCx?gZkj
z*no{+g*~|vLKprR-fM4Hzb$A)4E!J_F>z?~cejiF<wtiqPCqbd=}I}g1e&g_Y{=W(
znB(lo3Q&^DM~_tdi}b79hRTVr-#nk1m4!nMOD*f at K@)mkxPUX&(5-{!hD#EVHyreX
zHW!()UWU&@f^n$Ut5=jj5LKh^pIK+Qub46DkMq}YyywDrW3fpIy`_~EMnuiR6%nnu
z;eB1_BYze99H3)ixh`Z}Gyz#fkZ~tKu`qgcPCK{l)a9(~+nRTOCl|5c2tkNHg9)v)
zYLlw?$%Z#+o0D{9dQ*)dJ1pVh;Y4Ad+oHR-|D1j7$j!@30uuO)ue0lS;F2$2zixk>
zE(uHIQVi&h-WE7<8Xy6E6(AV7`S?iDzzL+~<%rrlU0&(;GC|9=efq at NocOAdUExwY
zRQ?M~lN!q)9R!>o(($yy)5kHxM5vj=UqGQufTy2x@~%7%$ha^!AjGB=k|p3QNrWR}
zaiBZR)e>$=NlB>K%-Cp at XROJ6_%IHD6J2v~_76PQ&tJYon%B6ZofuR~dU<vJLeTGM
z`{3>GuetW~BONR2VPCCMqV0|8?0}`o-RO>$t7b<$^{39}!r!db-uLe{=0|EIoO`%R
zts7ZV#cWPkTC#z3jh`QVlCPgBgA-L?32>TLT1H0Rg{e-$(TjYw?jIs7o_KOgOP`*p
zNP5UDX5BnA#LTOipng~0&#*}EfSkPiBvgZ9t7kGeAVf2f<1?rU9O!|=j#lSKAATkE
z&|jU-4h5b^@4qm<U!Jju at 4vA02R7LdxI)$2+e at t3S<tbeb{VZd;^V1tD(iW&ZpzsH
zj9=$O;>4^QD6U+&Qg%0IFGM6iL?pm6-X)WWxd~BHRTkmE0UY9l=3WrI7nWy?%*@Rv
zL;r4Z4h at -wyNo@RP*hZ0oO^%RvhLz;AWH3!Y(|ZNemJupBKqsH;K8O&p~RvUtD!Or
zW}r~5pwoLwj0JEonzY*>+R4I7iuwlVX+MoO<nR;ab@%YVdG^PytWFG+7~}mzycC|<
z$3;s#E%4Q7db>UCfnmpy>$uYzXWA#q)o`m;{L#4Z_pv(3llK&bhAw=2dpB+mvnUrI
zpSn?Qi8MS6oa(QeDvFOt1x+4s9Qnq^=a)&m5UJ`<JU`-qQ{SIs=>jj0M5WB0Jt!c6
zp(rXUIyr>!aq}iEtlhV*bv?q_{_DB9+<<zga~SNeUcHI~;BaE}+pMGBetusCW)8*5
zANOEhH%fIKsTLZnuo3eW)Z1&-6m;OyrAq<Z>k_W-9~|_Ra~0y at A;w9kD=SeYB|<o8
zaB1pAzFztf)#%Ow at 9%>m5CA+PB6}f7G*U&aB57iU&Vpd?1^Fy&`P6X6Fs=D%=uf9H
zr56`(rKiV&#y4n~yaWLFbG$JQa!%F8#-?F%b>aD6QR at rOaE=&<(}@#gIC$x>X*n9b
z;2fsBygVEo<mTpvbI(M~tCi!JL<~#Ks)!RSxQg8}&#(aI<2b9!^sz)#hpd+T&K`Ya
z^QphD&#>OtrS$A$GSHg)W#ss!-*R@@xw*wbk#YF;Iz3I&DWT_cG3iOSF788d%p&HI
zz!I?UM38ih at 3Eu2-k)(%YqZ<|3Kk>}&arK(D6nq|{R{DM>EC$j>37)@E`5AY>+4%z
zzt)@SNxK8w at WsaOX*yckFf<Vved6WZU$hr(cYui at cB@7gKLQA3%NY0H`SylY*s7ip
z@;(fN36vrt0Re%TxnCJBM~#L;3`Kcws~LC2AU6FRt3&^SAM^87F8%zVg4Jd!O_(6m
zdTwtQXa^khSSfHl&xl2t8JpjV+7<+Pd3iyTVp_nBC<xOW5DTi&?S$@cuaX6Z>wE+*
zEsUcv4oEO!be_mQ9&FeUa3QzN;gEXL`M`7M_+-4s<I60b at NjeQ!U at Oy{c2EMHUe{Z
zLHRVSb~eIk)xCKdP>arn2YCXVIKIt_^%$+qwR(05ueBnp?A%-cZ*+_yFg93(Z}-X7
z#rZ{eBH&iZ_Oi6IU)W$|V7LmE%^pgNmX=oYO>Q-Dzd4hE(zC`hJ!TwDK^qaV3<CD=
z?j4)?^jryY`01ab7WhAWZ=@k;;}-c- at y@iYn3>erQU9M~bOvMP7EeS=%gQ=JK|F-t
zyLV630TgcYT?K#d*}j7~s(GlwMiXS#>({S)fVEb;3&+{={@+=E5A*;H>ToG`&d$X>
z at 6Mc_8!k$@6CF(sRDW at K*0?EzSzQ05>4 at v%M9VtxK?nkVU%`0FLy#lL|01ql-BbE5
z+vN!qK=ify{P}aSX~n&s&*CC7cCDoATU%3*Jli=&`%|*A;^DSiKYnD#VcF7dO#FIj
zM7+GbxoQsPpfdO4(90)YoC%yt37|xA3f0c$idrJaF?^J{YUlJEKw%O?5=3NYb3&>{
zMeSk6p`%J7!Yv^ae-0JB!ollC`6uw;M%u9}GQg-h;_AeUIs$NHtc&cu0xRWzzun13
zdkDn<S{wlm<kP6`;(0j*g|-hL4#1z7<rRxsaK=B##?Ldo`GQvUQV0U@$xC}Z?s$cS
zgrv!Mus}ed at ml%sDz-+TXCRE9Ui?Z0fd>$xr~f8Xb{`J*jAIrHXuV2|gX*A&!0I4`
z*I$~y4%g_zahe2@@&#Zia*x+(5+un21}7lAqG;lT%pkZxn at 3_8Ly1w1ZX-CbT#dm=
z$ndPknu1wny~f*NnQx}0QGgzn@*E=!4i4fZ`kESP_)fn5B}o9(h4Euc1wRm2z7JIh
zf?)pf<3|t3EJPJRkSX~xGJ31$5K6I-t9#gK99-w&rZM*5&o2&7NR%-_TDM#!pMs|)
zX!*1f*S6C0qdX3I<WvgzvU!N%_1ErH^x%Ocu<OZDfouIm%gMP9ejyHN#4$x>X0n3L
z2#0*6WM}C=MM+7C7Famf9sqx4b{6Mm|Ih58`WyG<c%e~v6v!@dnf9_sH3LYYcTaAc
zef<6$h!6Z=w5bQpJuyH3W=fyXe at YSO0avsBhXwH4FenZ`#_AYw{37V1-o`8b at dOEC
z8XB4)pG+ZrSJy+^J3BaW9VuyTZ7oi4XJcp2t*8)#s`q9G$O}Q%lTFm}X&5>#0l5(e
zHY+b&)GBJzM6IBpfM<S9gU?^QvbijT5ebDX7kGtgbiTw9NlC29z)y7=4)wzlqxD(&
z!v~dLzMKmP47C5+mD2P6eMf<IviRj+=NJ#4K_Dg8&yu6d&CSK3;1$Zv1P7hlz@>n_
zO#&ChzeK-2_#+Q+I{1Jsml_W at x9{2IPk^%Hr0hc7RBs?=Sn+L|g7BE6?LR+M1+)tT
zIz}9#cIGDNxCnSKk=L$Ku~Rv?iX1z}jtm$sQNm+HD!0u}PaJ(%;In9%m>}8pbdjcM
zYik3;iR{T07fHE!GYqgS2XwdG#V;`s4+nzQ7kA-1bATF#`6u?G!7J2M)##3DOLuoP
zmm&@g4aGTjMMXt8>7R^-Ee7<(h35(mRJ0r%G_ta?INMi7M#k*XIm(eIUShsx at m0jj
zKwtFI#CHi8+-tA2YeT#DkQp8a^J_S7G40Zn_Q>u?{B8AVP+HsD+m#-pGB|zj%$YOD
zoU`uUts{vLdjHIS0S9^qT)xbG{5VHKLIMt(1;|9^9#&l~1&n+O3Y+Q;7U_wvUva)0
zgqdse=qM{bh$mj3B59&Zj<T|{N;&rsjgOD3M&~5~@S*uODH+*y^0s~K@{A`=oB)1s
zk(QY`y2fn?N7e at h2BH}#&gMK-r~`3tj02sqFFM>5tbns}edXdMicZrrFwnu1zV{UE
zH`L){G&RZJKRB(He&qbkjEw8(cMF$&Q1^~H)FL}4C!91s0Fs$}(-n88b5Go0!P*8Q
zjTb(!qy{~Bp?m~@+W)86q9s&tdNW+rwbay9TtTBMdkuW9j~(C81Mw<^2ZgS?I0#hG
zv^)z|^e7T1pqJ=u2cLy53TQUN&g6hoT9uWFP_{x|wY0VpUnSeynxk?R7al69-uy?v
zCQnaKoSO@!dkM1WcK}ak_3LmFa*XwY!a_Toem?r7qK8|elZwIs$2HG)>Z(X|5{#b$
z6^4h}+S^M9kcy6?=C$szK(#*q+1WxnDuW(J9R^TH at JIAKf`924u5wt~TyX>5xJ2qv
zL<`R(3Lp?qap1H+J3G4|wJ2>7s8}F{Rij&^!jh92(C`dkBv-9h3$XeH+&LWwM?$T~
zsQuhPNwIm&Ato{Fco43X>>SBBpl=V8a5Kaa$eio&^*Edy2dV?^MbRXFHBLEnS{%w9
z2rF_SA)$X*ZQjw|P6||_6Qn|w!wVXm77xNn6PVW0_J$A6n1o^pLNpqLFiv*&^h}^X
zsDrawXBN}L4xK5D0rh4FtNKs<uyqcM6dUC)uFhM4&@`PfBrrgzTzhJBuPOpfO4Kzp
za$(_Bqg$(QCnoMgq63v<k<>%l(ZwYer*A at n2LMjZ4VJ}S{{4-cn_C5`+hLm~MMMO&
zsf9<16ZwOJf(#!!>PvZzv$4v!E4#aw_ShXMs?CsciD+(CMnfbJ$IjyoM{zvl%a>YE
zz~lGdJRZD>@q-puSy@><PrPh6oLUJE5a7H26<vv_%YU|yj*X2OLbdgsAK4GJBNDC@
zE$zO3je;zpVCU&Kba8RPF|Ckzy`^U*Aqvqf1;}7$o<`jIH)-RTfdLaVjD}JtXkIOd
znhy(S6xDuud8wikXh{gNU=KT4KM|DWPDmb<s<6z?plKpgQ<>o_D4Cg=ORb)rQ9Qfc
z5yz|={ella3Wpd&NsS2wyi_oGcxEq-D*pJ)KZ+*ll?!Ny_0 at USJHlsqfc{7nNuyQ)
zDS*R!k?ZQ*5sv-&@!1so59iwB413 at h=n#y|v(|$Z2j8QAS4YQgxDcSk2hk=AXOI5f
z{=LuM-ahh=c_|+5Mc6Q_v<-(mEgI}f6+`D%;j at n-IR`td`#@;l1oRDDnWJQ<>PJV?
zs~~xBeg)w4o}I0Q=UBs#lfJ#H0L9!m3<%pKTol5pjOSPk at RQGwE`ZDgqUNqHs-Hi9
zqA4iO75x;lg|~d7?&8-)oOKJ=hkk_edYuEOfOUC)?b?r)%4jHAW?qByZ`(UM0M#5y
ziJ*K at qiavPq!TTqCS2a<^`$BNA+YY9x3!{Uv}U-f!xIu{0Hkr~HOO=b2pxU>j?Zn;
zI1s)&RV+Vf;}?#ywY0Phpl30zhegM!xqx|UzP?iMnNixw{J9Xi0JBnh&d!2pEe_y}
zv$Y8dmxO)&HZE)GU4k0o+?%&s at z^o!afCtP%!jHuT<`a=C)fE#d%HT?QQ=H&pfgyL
zUuPXfLq8^YsXK9ti{8Lk9X>S$qZ3NV*1Rm1KA at pjQ&LiTK1&?<@!>H!V3!*F^l1e^
zt6Ko*Q}CA*R8*}ffLK}MuJ1qG+R~!3ZCdX;tC)z}tYgpo_d#GW{zZoq^++HJE&sE$
zuBp+0PrLlyV=ohd5QFgp2;w|ceisn9s22Daxvs7*7E_SB04iqT@)ssa>;1l8gJ+Yc
zmBa&oSKSP3ZE5kd8sw!uNY6W$w^l_mq}{HuNV}rZGR|g-ZfE?_f8PLEE4p52YHGq~
z-s6;eXL3?il^Bf&Ak3%6>U;#xloEgdNt}C0i`friN;Udr+6eA^LS?TESh6Q-hCG~@
zoFpb8AqnS&ypLR84*+Sr8|B>N$C76%tZx8S+XIC{j}odp1rcO3gB*P)!9jK6|H`$L
zzPdVTLt`Vllfk0zwyO7S0c^vBXgd4y_Rp2?L+H>^>-C9-2c3?vYM7Lq(g3`=rmdsX
zy1nrmjTW$VU~O#;VEv%SPD1)jQ_y8xOu#xOC~CX#QqX!9XQu-9mS-!&ap5?N86?Qs
z&yQCd8;=zkWG4VGTAE<o?+?Wny)MXU*pxvCLKV0Qkd^}_pxEL`IV`@A?`+?0uTM=H
zu=w%u at mN<o2dv=%B0t0#^$-ULu<<u;^cLx}E)}?Il$jp at Isj=O^H)_(4WeU at nwpxF
z>i`8>M%C6{D=Wjs8{@BZJ?ZU*(7$a-yqeyB{{-I_K6`4$n_z5##v}ld_l{j5la!PM
zYF65e-NUKQ8!_?mjBt|&5zWi3o}ue3EI!l5udjPXs$EEku!8`R$nUKV*d7rf?)aL3
zPa0%EJ6cDL6B}#8bD^N12m`W<j>Z6*4inAc1jPVxC`M4!Bx)Go+JObyLx^It@=U2&
zI9v%8ENy96mf~f4kouA3<)S#99b$-zo!&vO<^LnK^cZj$HvkTHhU<K$Z9P3{0MF1@
z`kyp^QLSudMgVn-<NI%m*>ErjoNR|sK&cI7g=OZ~*KV8`e%~M)#TGTIOuSdZ;~;H)
z{iviQI`k~JZVaSkr|Bocg#k*84z9zaqkGsB2Jz$-eliFQ9M=nv1x1b0hGhpG9jpdA
z)1kOn7;n4^;v6*@KFGIk6GGMYuRxi_Y3slhhaa7fgLLdn8+rBeC0;E&PVdzpXGzJ)
zc|i^AWtC~?4Bg&EPEKA8xYK;5He~h30aT(;$k7!CimH~bt{Q+;sJX{8m;R!Li$yHD
zCr at 5Ol?Fu~!u6Nh at dL@m5J{*Z;D&IH|G#uUAnz*(^%1>t0Wr;S@;d09R-h9-Wfrm!
z59X6VDR784whchv0WYzfxoNfwC_ at BDuEBDvxH}>i_M^2P#nz4WkR^BW0prlN at 6Vq<
zG1srF0s|C>Y+qjZl+7vJk`@ojLgGI|U7X&CCdTd_9w0|0&Jfu<I7ozS2mBa)vJY22
zI-2~`maX2^MVyG=?^&-(4thmzr7>`Y%4hm5uEpmsUq}Hakl2gaG)2PoAa_eiPmi5?
zk%;|Q9Quw<64+P*^h`S*EDv8nm70|m4^ZDSKF;AcHvr0WAG%iE%X#x_9;g==k4d at Z
zSqS!P;<B=u`6q68|N5d0_1$%T^a;-JzOeSw3VAv<Y(d`w#|j6uq+;jj#`=TE2=y0j
z?ps<~KykLF95zMZft;hq)dpsMJZPQE_o?t6A0MCpj3!uj?b`K^)46`$*;-p%T2jHr
zQIXyOG=S^w-UDR44Ps~U&(4*|s3_El&hW&;^?S&|0EN-7e7Mq%1V=IhSXwp)?w53Y
z9R_t9JIw&dxdBU)Xukb2iMJ|v>yl!?A`$4h-An at AoV>ieIPhFbit!(+0SUZQQX(|=
z at gp|GA|oRyK(Fuq_TjNp@!7{kJ$B+E7j`!Nh_F*gnh at y{2jr~UrJoG68}=p)dIotq
zJ3AK_7nK3RCMPG+a2D<mGy3QwxPY#d!~Zi{F{!X-hl^AOJi+#PmV$r6mMbCo6t3#R
zLgK`PEiGM!tXIM=5;87%dGnbbkg8|$)G2Wf1YI3~2=?Zvko>DjmH$XU8u6kIz`rSQ
z{gO=J_y`~tweBrI84SxUkNz{&)=c>F>U^nkK|z5nFdLl3k1gpsA1A#`nH#25=XLz1
zI`=L;V8dzc|Hj3}y4*xx6{}~L))zmM;K at 2)IX$(fb#<Sc!$|_R*Q<JREB}i9=b?c>
z>g(%^NT`<~70c;fZ~}t;0t64v(8rnC*y98mnuEqveT92iWZYlw^D^3n?&sI`vZA-p
z=E`^Mki?hVe*IbxTQKPMXlu*$bFA)aYt){$|NMv7rxKM6 at 1JDPZM!`Jv^vDxLqS-W
zIyN>|@KgZ>PLOwVJ8bQjIcol39ymFUJ0GmDK`Ux1?7lpN>_PT{&0_dW{CPU2|CP36
z&gB*=+`YLl*-oh#w2rN2CR|y#uIRo(U3AQY6=IO*<G=YonvA}tK<OdUi2qd(dtn1g
zN=odr0n5i*f;Gp!*W0&m8;&SC96`GpG at DE0RI>lv8ij0fd1VCwpy-^0gaq!7K<SqV
z{S$-(?f;RB4njA%_5Zu`QY1Rw6dZ at fFpvi*7`ImEad*-8```_Qzh>d<+Z&QH?!#dK
zo?P7AyHH8NlX&mCC~m9-0>G@^R|wS`06F>{g1DlW7cR29^m8m0o)XW|qnQ3=lzaDT
z0DRHX(Y0J<H_@}SWW#(yTQ-16$kP|`2P6%jcv?XnIRbdt3gyQeo*M=pls*IxxT at YN
z2hD1aQ6{K!O!9hVx08~>kud{lf9m(0X`lW1c$Y><+N`5KWZ~iAXgc>Ik<+M9XFqyJ
z0Zd=LdzT#weiV%{cVB=1(nJdhWaUKX4c4cZfAil{yiClaTI2fO*1>`I-qVkjc5TE~
z2eQ6;z&GM1Eip0i#LSE!iuOVvvUwppo9O)hkGe)n6zV{csH&2{{h at z#m!_W>o3-%G
zAeasU;lFy`(XsVEPZjjxL^{QYd!v<URC1wz6t7Vqh=3}K-r*ooZh+Y4uMse~XFL7&
zE*fZ{3-g21htbUofDe6e<P|$jtS??Xg7{+JK?Q=}7Pv29Ss7Fg^izf#Aj7u*$Oy}o
z)gPZF;^Ipr?xF9}$PED(W?KBM at p;@n_W4GQivggOckb0K?8Tf&V0)DuY!(jueJ|W#
zmEZR>1Qr(i?e(SgPk;ZsILswccx8Ep4!6JY8Uc9z{ry9SBZX!S0TSq%ygX|RRgi^`
zmzU&x*X@>w-Fw6 at OsYc_qz9da4Rf;^*Zt@)fQ}ka?$MT6UW%WWr}eIaKP?b!$SiI-
zIRacDI!Ljxu_5py(fy|$P0HvvIrmvuSfH9=7Pah2dlelo!C!{^dtY89<Jv=T5fPzd
zJv>4*Rsdc>$!6CQ8yAO)0E<3wJw)rOsw#0QDRp2 at 9fv(ehljmG?^AjE`E{V*<oco=
zFi8z0c)B4WA;`G?Y)o^$)zr&B7QD#_;eywT*bKME;<)<b!^}v{VJO47_wPsFzRj4H
zp56jAN$nanoxEaEZfjc`TEbsh`)T*;*1<Fh2eMD8=ae=e!%o~_iE?<6 at b9ms-I(sv
z6`lFt4jVQ_A&${10qMT`Fna4;@0WDAjO}TDgM0Dldx*LRs{Fq#Zp{C+g-_AR$qb-X
zD`E9Q4A~uAumArVaXKgnv5$IC_YULpOB2nzpg87S{MtnnhC=S;OA55Afd?H)l_F8}
z75ECCm~j8_Xt$VkLl|TcdJm$FWOK9f%wpP|00 at RN<(8D8p`o{hjLD4epX`A8xS?II
zH>++?lC2baoJGnx43Hn&U-+(TJDPn$bY+i_qMt3WDf1_uA|dm?r{AGEtl;SUbsh*V
z+VjOS2t?9EeMLVo0Prxl$8zgNDmOPb^k)YG69K3Rg#Rd7Gy<%mJv$0yG^HCDV4NSR
zX#>U=jy at d!8cHQgJUl!-4VQhe1^*&}eX3U2K`$TOqX54#YyeG`=H4H^Eo#NWF7Jy|
z#L-BD{-6#8dr?=w>W}n*zEwQyurPiM=yEPdo&UP{H35^g5Uf3)UGJjR89*T#@xd>Y
zhj*YRU4!dKr$9y9Oqa;wV!?P$C5~6GUST(bh>M?}47a=!IgYi at H@?eO$~VeC0kxm@
zh-J|(_2hQk4gwN6 at X&Oh&$rhtfFXuu=F;K$4jJE2e?2-!85A6B=i#x}%gbw^%)%5=
zWwhFb0iy%f$sWK{Ttb3Qp0D#969^kvCyXjPJ3C;S@>zH9e*W=8m_xIhko*IZ2YqnI
zpNA&G`k+@>(7t$yy{WfvM*_{m5`hZ at pkVaW+YTH2UzPjTdmvnhi155e#nQSVFiZ|b
zZ3{|t*~HfgYs9icI at 4(3)}x_@*aC;gT|fA#=o3Eh;IuACMJ6Z^QO}+!;D!>=KKsu2
z19-s3U+^(8b_?Yc6nFs)fImb&e0aDb at D<}BsM7iwQc13#MxS`)UzqCDNr-Vd0yhS7
zWdg!T9EgNG8-LWzo3sG|0TZCRu}=uqlFZ4;iQ`=1H#EOxWQ+v<l%*8P0d*EhcR<R#
zjK at e0Bo;(U5|Ti)^>!O7UtC%FO3o?XDb)ph1`kPO$z3337PD3d1V$59QLFkKKr9S}
z^78T?y9ZH_{2<cFao-0(2#<-t0%Nxr811%8-z7-ce|r}%S#HFL+qY978w2VJ7a9%V
zUv!C{fuXlRo8f=>7FJcYB(?=)Yti;((U}6Pvlj+S&zjY`r{BuA%P3cmrC$dzNu+FJ
zgU54mC-mW&GHeW?ac5;^C9RxJ(a(shSFw2)7DkkondyLDHd4;CPo~mLFrhRxH9^u@
z2PuV1`0eN5!2M%9GX>mP_1UF)w8#FR4>;(+G#wpnZ9Iij<^JIa5c~u1Os1pRj}+}j
zA4`bbma#E5xR-5jfld>~#C_-ECGM~r^qq+SIfFCVhJvZ7sj)eW at MCBaFX54T=gu{?
zXc{{Z%joZgdVXo at g~Q at R3(&9_NN*7pb*xi&#B36}Q^ndkJIR3QsYAM~pO7++`SRt9
zqJSV7DJhz{;Q=Y^r~=gNH}usT#qJ<Tn%91ud-m*U2w2Js<FU^;1hfO|5~{O8U1k99
zBmX;`QROxyN?>Jm`0_Fd3DqH!VlH22;cOrbc$@+aGO>lUXU|m-D7f;VsPlk4;czxR
zP`Wgy9W0-E?}DPx3AFa}{JhA8JQ)K+yZDl;K%_07hp_)gP1FMd<lyTOTc85wL+n_Z
z8s?;>5tejD^IoA>Jw$Npf5#3+2M3u>KQN(_x4faIj+K at i(i{^QxELek^uS4VbyBD~
zy-mSaAdJ at TQ<g{5u%pE#8i?-VJ*OvwhPl`GA6^8iWx4h&^zZS~(o!nHLq#R1YKx1D
zZLEjhzmF;}7lr%Wg@#RPpPGXAK*=hCDufp3P at 8an6_Fmhqww5*k#wh9jyGJs3IH<+
zC^Y;$RO!e3JP`e=AZ-xQK(#OZ7)Gmg;5nouWYl4CaeJT+aQOzrmlO<D*f^jOAy8PT
zYE_r!&42Pb>b2pKD0KP!_{<;wE}re^shOJu2l_>X6^FA+d>#i&jPl9wl}yXE%j=8W
z5GuyXbiij%`}#^Djllx}fTqU+mz&p at rtG}D#8&!e4s=(*(`S?O#)EH9{TDdoeP_Zz
z&HUP{UgQlwPZ^%q#E&1_-L;XCq`2AA)YSA3=M)tKI?fiE#4(F?LUBpDbBF)JGbqGp
zAuB8Ugg7DICGSj`c}rIpRi=ywy*wj-v_bZ9JOzd43yz2WiAP_bUteBrGoIf2YvZ>Q
zZmR)ov`6h>UYzOWMO&wl5pzWsSmHYZwA1}1#`p2W;PTIpTdDhNVkv+>w#Cr#_B?hn
z*p)ACa*Fi-&H|i^ud)f=UZ+!0QCWvP$0kR6XJ>0)A3c;=jr%9>+B!MCAlcZX<><%{
zvi;D#r@^4dBBP^uj~!!2-BD5N;pP^H63V$ZPc^!f4(=)XeD at tZg_z8*|8EoHEnFs^
zGD72az+Tcthltwzd|o2t%$2pZOya7Y1#1u$il#9*BX|)qgZO%k%OgPWZy9$$z$vR^
z-vS$5;z#a$1Qvt)Hh=$Y?c+Qiq^ze$4YZbyf#E8KedOj3Uq3&iQZoq*7m&VCuvB8{
z4~A?Uwr-Gv*u7 at oalPgBYidVF$GU{V_0S(5{mo3AMwAv8q-YCdXB)2$>?)OV(2S`f
zO at 5kjqw_r>IZaw3bNe-arv38!FVeMy{LbOLPU7JdQ5{WrEw8?AYBg5JZ)?$4-gZ;)
z^Akt^3MzLK71C^YpMG>&(f at M6DWQza%-+7fi28bYv`o#(ArcZ2(gL1uC>Yt;cuZAQ
zwG!~}(OX@`Cy+R3 at CQs2?JXB4<HU}z)2mQlhqc`X#fo;DzkdHl^ab>{?YO at GTpL{k
zp&2M@<%OveQnIoR at T~z;H)OST1XSYf8XFte*Pj2qQPH$X1wh^!PC~AAY?n&Bc){`8
zToe&AaZ)b(lEj{G(`N^-eh3q=WQxN+ at i`|aN!kM{v{M=&ub^&4!;+!DER?T~dcW_s
zuC86a&h2tMOmXM}8?>`BBzX968-(=|4zRx)7t4*iFAd9mx<s4%+O$ue>^xl25K&tz
zgI1j_EiHOc+VQ-*ckc$O{`aE;+>f1;ldK*cox=zCVHbcz-cC)GWT#qUtAF+kOJva2
z{F9emT at lZpD~(@SI7YgAclaAeNp=p27*{v9*=zGJUuxaHecQG*k`kNUpj7E-X^8;Y
zyz4BB)Y$#7`a<;<Hmf88kx`Jzg2x{A9bYGEo@${O85z;LrcFy4J~n0<85PB?sK^GI
zeZpG~t<ynlKxrpsWo0c=<NNb4KcA$eqy$wZpdY%%b#>`fesD*3%oi`BL*$MgP5QlN
zQ+;L46B#b6jC-^Hr(gkD_b9FyrPa25N~o)Fe>6amLsd<UJ4vhgv$wd at TQ<wOi&3+*
zl#}U4ft|?8GLxh{S`1JO+MK<Qwx0)L;t6i+LB~%S8)t6 at GsQ(!R*J(T*y~_Ix+wR~
zfBG#u<SXv>6ql9lPVJ6LN!c$fESz)yKKZd@#|Q+1vZ*Ob_MLLhD_5>?Gg9>q4w5Bl
zk-~rB%3<HWeG5q(96Uo2rCRr?qNwP^Up+rRKe#Ea!c*1N)j4HlH;yN?eQs`VC#|im
z^}clJT52jYke?HCM`Bb2l3$@Fz{aA|&T at a4S~sPA^{rRuV&dZ3KDR|<SJqpOtzr9D
zH#YW*^wL{>14t$EAf0W2{z3iO`L~M)kQq2<e$m>RqiN$CE8DTa+fqIZ?SlBycf;z4
z8)$YsW at dN({QCt$m07&O at eNX-Y62Dh5rh-}|8acl7QLB;#lAGTFBj|9=N?wz?mrn1
zFNMAE_nDbypySy5#Xj5k`1tVC3%tMqXf!l5irRQP!vL2jqnNGN+hS`ze*B2V1WQgv
z1`8hi1?m%3Q?nwjp%X8A<HijLTTLGyNjWwec+P0g2sGUmI67|lfnr4$=|`ixBuXu<
ztddev-Y+zTb=V_x2P{tPQcI$9&<7;Nf6L57N7|w`ZnBp7(I-_9HdxiqoH-M)(71+;
zZ at nzJq}iSrZbNJ91c-5U4UNxG;)_d4!mnSa0-iow<RL37+gJKGesXfM()+V2DJkg$
za7Bcc>}*axdP;~i+|jlP-AS1hr;H#t1zx!VdN&dH6Bdw;jt-=FhK96EqLwj7eZDDc
zXhhA5$2#Ho`M~@0TWbrv($f1?b7?_xMBcwI&^THP>#wJ$mqz11fVSP%SLUVhxEBES
z7iksR(7tC2P3X6p2P*tj$!+UI`zo`SFJEHCZlj}bfBd1&N8ry&Md;qAlA>`Pszye1
z!XhI7A{^6p|9<I-wWpS%*croZGar*FQ}id?_{wbA5I_fDADNUyx4p9?;qy&zeSN*T
zrNvImjRO!C(zvIuj~F8Q-aQTq(uj~3RUh1)U0qd;jTy|%%?*u>Ss$fQ*mJ>?kUsur
z-2ovbVEV3q8cj48nU$N{;geZ%;h^v6KzX5T!5+G)P4IsBUcG2dTHX^nI#fWEueP?T
z;!fh<?P;a0y@|^!D+rObo*%DQJpYrCs|Amky0EYib(D^-E~2Bbd9~J0AQ^v;|8H}C
z0_vNipiTrKTPs*lg_cx>TYp6DWq0=;fEJQ_-nzQFc8-pnhUZ9C;-z7IkPIbo9J>xA
zG4k3q+?82eU3HX~vZs!Xk8dwM`*;x`(QfVMM-q<y4<BONqW1y8Vt00SZU%bi at BXnq
zMUP70+E};P^LU7HRbeKQ082#%disbvcbHJ=!auY_!Pa(QkobXDxKC-j(nf_lwYx+1
zQ_xiq9FmL&BnANuX6EK5zgIl(%=Bt{<tx|D)z#hI2JjX=^&(O2?AZfeO&k2<AzK~<
z7M3IIEE17aR8%uX^Po<pWMouzb<v+5$trL;U}@<zKI+eaJbinYz3+H-VpbNLl$4Y*
zPg2h(rO-c$cZAQj+D5|tlSAO)F-9{p0-ntK_RUdFUf#YpPXk{QxGiM-GI)*vD0(w9
z^YD|=OP-#h5b+=gnx}sQUEV7uCKhnzikh);^6X8`(t-lM$jC@#3ky~d^r3S!q({=J
zf%WqYb&&SHKofZu7OHeT{?u&#tJh<Ciy9i(n3$MI2G0fr2J##|nx6YIriveg<LA$x
zcbeXQy@*F^tE;7sPz1-scGzR1kbED9x~eKM9$WRDJ%vY5I5;@|#m*oa&ktI7eSZAz
zc$+|WP^+&Tsl>Tkw{Cf_FWR9wJ$)>xSmQE1#M~Ym8=FVEf+`bBwmrB%5+~M#T at o73
z<J1NQ21LpxCXA3Lvkq$upAvy7OYT3ULVFecyuBJ%m at rHdk8e_BhyFPV;DoxAjg2~8
z?;79}mx98T=8v at ypDh1DX3kCidG2bbcO2+?c~QVpCKDgn*z?frmX{O}gxHbM(cZDK
z?bmNgPn9srdX6=(eF{y)qYT5t1nk7KdNu5_viiuKaEE}#NO+hvMx6)B7F1&RnzTPn
zT70b#Okr>v%3QRz^CLCu6Ib`yo;%lWJJ240wL#dT*1dUQ_WO5U9-fGks<iFi4*}Xh
zpN8#@I(_<dL0T5U1lR&vV}gcDr->pE1Nx%F at haSYQ4v3U0JYci9hrZX#krj}`PDf*
zJ`VgHTl2%iX4N${DxhRRLHhQ~e9$O;4qAGBa}%;)%_7^_)Kt^XjvJ~GA^s*;OjYrt
zNA$3~pvctVF+(V!xumS)=<u*OJCB53q$+n`U*FBFtau0|d^x4Gl$yx)zfMX0<<XBH
z*}J;BUXv+Cn4CFd4 at C9l+qa##@KA_W6~1k6{5E^y$qroSRiK;-(K>(tMqI3VpBFn-
zKm$`sb~ZL4$?|w(l)OOBAcUJ7DtzAg^&;v>_mS$@76`#F?@etwfEa$c=R}lKTpSZ0
zAAg at Fj{BaCfx$lPcp%Sv|NeczY+))M)?DA%NGei at 2snWUPUqe~!~)pos~LZJetsTR
zFgBUn+tr6hMka5_jyJ<s<levk0*Z9YQvf<7qLh at B0c#)qmli_*vKrq%Ss1NFoap0#
z{R(ey at wQ%i(xnD;kHuq7{r3%CKzTy$m1{^*VnW4{NDj%pz8-LdFOBOOWy!vS6cJp4
z1=?3A-VYA8LO~PFZTsxKxU=>txx*-e>kF_~)Z>pG-?V;s>=e2-47`Nx=tWZ8o4;^D
z0s<ei2z$9~G)`$e{GG`U9voWS2$GVP&V_QL$g%J6nbPKWN`I+y4PzRwtWA!3J=f6G
zB*$tJ`sa5xp8xB&W0+D0DvQTqM at HmrmvZy+cwk*$o=hgq#}hOl+culR-Z)|_8F-9?
z+!6ayELv#XAR9vKVC?TNi~iE^`S><K&~eU%hTk-IL at ZXL6}A8- at u)u9>)a5|PgCbn
zadpKb{b>6O&jCWMd0^nS=P(U5H6KYb(U0LO63*adqv!T+6HA^EaZDnu5YC;-i2Vlv
zpJr!gF&aVO!xiQ{cyJwXiU+$_Ou}Knru*&adDVl1kDWSo3ZVAOmvaE3VclO_U%iSQ
zcNv9>^7if9iHmi5(XWv<K`zpwJ2gHH927J34CtY_xVSP<8BnLR%*>h at W3d<JF;c7S
zJB(SY!(R+NuW0y1mB6kb)6hcKZiAM94Lhr&zG}M`1FUN5>sdB8H>ppYh(Cyk0F3V|
z*|RD$VLUJ7HdtyW{-_GF-VFupp+kojH@{and><?WI_D%W96?B__4<^6T#i}Pl1tye
z{V8yzD79!B?7BO9d1-2CQH%#n?R}jt87-AhH<h8s|IhFqc{?8W7ZOUc;I*wXGc+_r
z{TCDzM2Nr571MI_&Yf!qb?@Yqm(zgwp^fWUUK at 6o!OG at 8emS~@g3>;SdnK0?3a at fT
zb+jeaBYOvj*)QDDZ7LofA|DE)xhTxxH^jumHz6e_medcDkX#LUao*9<eqp>3**kpP
zpVcSN at ib$>(oMoC4-XGKp?Ezm?km$pGhrsIg4oo9ACD|^Jb2Bj?}r-oGq@@)MMcY>
z8y}c~rAI~WK62!UR+VT=8uqK;!!E9o{(F%O>Z|v;b~B&9e}eVy%J2XB^{cqF^m%zU
zdqR6hhaFr5?pgvjU;LKoWe<viWZZx#(SH!PL7?~~Wn}2Qz2Tq^fPg1VA at Cs40M<ES
zwejQ;+%WK`&(qVgKd%qNhwV{PQmTvi6P|VK3RCwT5!Rw&K>x!ayy)23 at q89wGjU2v
z3TkMei6B)c78d?Sw;5AJoH}#ngsg|a!Gqk82LLFBr%pxu_X@$xi-57%9U!7e``tY}
zF0B4IYi4FP0r}VgnZgCaXKLx6 at M#K4r<OR_Sy)g|;3Qi9&09`TRP<VU`UhG~4cd1V
zRaJBlU>DsUG&MB=)SZJ|6yBIt`Wp+$m-p-$Yx7h%6E`=voc~X at wxk>S($doSz(uVa
zV(<XVP`TAP_Dn9amTP-i7k|G~;yrTYmefbODIF&#0X)0|kDXgol!l+5A1~87bjOdt
z$_lDdMZBn>fYjRB+S$crR`%tMw$I9^Pw#*A>Q!ZJt)rl71Q%I12%t)@PffeX?;I8s
z#L{Z{jgcaP^1sSTq?NpSQvrlJhA$ujNFe0pgJl+a(HgYv6h_w8G at +rPCZ?up!c14Y
zzuH4mtiKEWLjoAUjV+z=#w&Ep%y`}xVfC<Sc_iqNwq6DYD_B(Qg#*5<xP^VaSSKze
zg)ew5y;WFH*42&66^vc_f#;DnzNY?Nl#|K4k(0v(&-gMSk0<6&MrJ0yz!NA-Bbpjl
zUoZb_T${Qc*fOx>$rqfV?V!<tnY4>PC@=5)v$M4{m8g_gQ?p<9j&KXyFBbUCw6||h
znwgvDL2_A&GUH*An3%AL2;!?r3~ITcj{O(LHO`&m1H{84%;V$M!yi828xTPL7JvU{
z?bp!J*VhNk2{Spg^O4_!DQ-82O-4pWWdj2*0xuthy_Bn~YwP=mrgTh9*MMv&3o_!7
zY5+!6C^ustKepIcrO_l4NT53Z`t at t(Zf)|dTQ}3wDXYw=;gbWOcpT5(=jF9<pJwKO
zF30hpn6KTbSf}xDLcQO2E?SD21s^#!ENt*sw&FSwws;oq_|^}}v3kEfmZEDVBqEm;
zIgsgt#NX!=NxC$IJu4$4qnlcMwvRuo&)(0j3|w(7YUAm=iCf)3wr76+B#UkfZ)!UJ
z=<(y++}svlXNR{Pam?6vbar#Y*FN#Zlzwiy_B5cYsO<0 at u#k|@x%FKVqWc*cC+=s5
zv_j#E2U3Ik_t<D at joQ=l^<ws{6JFWe+#Jw;j!9=qd*|Sl_og5<M~#0wK*_sy?HX_e
z{>&=cc0ndOy6}mKiI6seApFa{d-q}z69wAzDeVb2Hw`1WD2Ta0YMkVvRfeKyUqx`2
zVTG%2>q<GCP7}u!L&v40gj5OQ;nv}_-ku)iz(9E<6`UdKyI`I0RQ7n_cahf_(s7Je
z=-X8JlRqOJfub~%*%^+nA3$LubiSmRtlHcR#Fr1s%H~VrZZuEvSRGozcY5Jl-JYuQ
zQ5;oKQNb4*h-`tVK at 8c-@z|GAQBy|(*L68bxQ~6G&1ILMzdLJMy1FFU?_PA?V8xw}
zzZ=<qTi{_GlCALrjwN^b7hl~A!W3{bdU|?V1_mU3d<PGz2s5?y_0a=tET0_x_U)Tq
zw7fIcJrJf5Zyfi at 9c>WfNQ{Qdo#-pnZDV9tq0N5<$_b+mlyUvdgIruhz&!De|7qPM
zf&g4n(!F~LINxk&9>@_SHo!%{f`{>&QCJ<lXvLZO`g(j_3)gmKWo2hj_^dEf)8^b^
z6k)j(lo6Mqp0EFS{En81X?4KSWB2aew^LGX#O>PGKC=cg%ywn&{Y{<1iKPH%cXvLu
z(pZ(1H?p55`VEvcfOA0A0g*56F(Ose#$pUtgJ*%T(xUKc+F7*_JbYM^ok=DB=-s<_
z7k~Z2b{h{buQo4 at geNEzd?PG+A7|(M=#$)vip0|o&wK{P7l!SLKh{vefO`Qx(!=t%
zF-oXVm%{=AV&~-NPx&WKEd`(?hEIF(MV?0`-f#c@{rDoZq2TsMOS#N9oxu8V57sU|
zfvZ#O6sHb|T2L2uc+jM>Qe5!Rp}gGOT at M~S(2G{5C2hkK6Rms1Kon4$>2*3j&Mhg4
zm3AAP;QYHz{`&Q6!u8vP3hciFN%{MehXyjMu<+nTAD`^Yzl2q&@7%meYx(SwN3*J$
zXzCtdnIK$1Gzr@@u|X;585(lpIu#M{V|liJ`EBR at lXqLM?>|gPh#@>YzO!t+y*laz
zyxdWqF~SZH#5{J|2bF+m at 7}%DPoCsIer)`cz!VpSObPo0YDV;WGcz;IP;F>gS>tAA
zoFTMwL2N<ESX^7X?>|~wRdrxM$&^2N?{{T<r39N4PVeraV?%%<N8pt!+42+9a48gf
z_p;ZnLt#xXI3<MJmk?#xY!nmQXKrq8_SBo3NyI!79%0f-uzfpVh8pn3kJWc?-;M-|
zTWKbYUe7m1p(c6z__Re*GNjRL)TMyXdi(ZgiP}MkGi;GGY|cGwV>_S{FMRp(<@L!p
zG2HZr>Qg6e0?`iihg`w`=Q(lJhyWQ7))1`#TqgimMQxg1`)aile1P7$4_EF2TtEql
zZIU827UQ|^->U+a?7w{ZGBI4`5U|w2oK6$V)mdFyDh&F1DnIdfE1(fJQV6Ej)(P1m
zo0OUfY%2QtdmZ0oV4HfPgV`n$YQk6*$X0ys4N$!K>%}FY19l+hsWnd0hFUp03x at 2h
z3^~c)c`=2jHt{44REBGEN1))wnF- at TE{L4$*#O9!bB>O!Z{K>x at bXeb3`Eldn=8=X
z62sd8WFsZs1^3o7K2Ae~72MxlR~P5*A?d}VT6k;xC$okZiq2?g>;{~qqN2k6xq*QJ
zP*hq}O)(cV`IB>@@&Tuzja7&LW at cx@wfJr+{CqXUx-v4WYiykS$8WMd7QfBJ#N^tl
z*djIHkGTI&OW#YE`koE1bl6wr+`k{)*LMntsst_flBwalq1Dw(C`b>bsB+Q1+~G8y
zPHX}I_{#M6C>ss@^kzr}D6F7-A(hvzgzPJXWW~oDSmf}*T?RF;-}k}27fjl{XXodm
z3<V<%1=WGefKrF50=4Z+!uKCP)J;uO$A<Xz(s~wxbzu$IXvl$lBR7LcGC6%Z`(f4d
z1$(k6*LouplXg5)eeM;#u;o)wqb#8t at 3>=}R2rwBR97n>O35oL+uJ;)abfQGu7HLq
z=9HH^UsG*14~FbgT&Yp{c$g>*NW`(%QSzp=>NZcjK6Pdt^#QF;1ZYy!#(g4eZq7CG
z+BF^ti3`o2`H3n+|Av&>JP%3Z-EtUMsewzkuwV_ot9oFDC8~Zk$8e!+pr+>f^{z^5
zi$pRjD=RG>9aZ4<_V)I8KmwIhb2HI at JmP9ag%}$<d-nHtA3$0l^LNPzTwmlyQiq3q
zfQ~{4A4!kr(7|;8-4qcK!Oi)DmywE_DXw>9<dA=&mX-0cbdR*G_ at eYxP=JjLDc7!%
zLA>SU<@GJxOXiCCgEeS=L>f1`6cyP3B7pv#08xVHs(rrcJtEmJ;fima0Wn<W6%wKp
z5fK6CMqEryWnK?D4m@~bW=3v~Oia|GmO53;CS=Y}WHBJ%)5`ZD36w|i?zW*a+3S3_
z4EXMuYHL%X9?I8DjOeiMtzbWR at Sw?=Gc+Uh5O}!PAH1+yx{)N1J!KLSxd|b%pF!Xx
zpF(+p8U$%^MTI8BIG%j}@xwi6xmVNM-=BOj^c_18EVyuNg95=){}N8=xU#&QQ&K_+
zWC=9Z{dbMQ3l&@69^ngyxKBSi%G%r8%O|T815AdFjg1;8WK0aj at aU){JH1N$o~_mS
zHid--+-B3uK7M!Bo5EflcoFWjV=Do^6n+-I<Y at Nnk_a0c8_~-OS;NeS+rPTScpp_2
z0YfY+D|7bn&~S3PZ|Mn8s_SX%^C|0q-`)a<e7sa==lFU*?#Z?HLqlz2V_6|I-gTCB
zHDTcZV+ipA2KU&iRk}@f#7_Sng$%DU8xwn{dq7ZWwc4C;NU3vbX(p=z2&t8T9_^G0
z&}AMT9`B!HN0FQW=3!HEcJ{zx07oLZtgP(VGyjFUi&i{IF~|S<`uRc5r*I_vId}5e
z!bWtL-NTZS-f7MUWC>g+<ma+ZSUDKIyYk>kp;(d at X|7>R%3)LD0837tGOsm_ua}Db
z^kK#H(!|5Fx+&}<^9M7wRw<2YCT}-QpHOZ}*t7nejR2w#QV=vc#F#1>{|$+$dhKU$
z+ch;c#U&*wAj5H+a9}_^bZpHhNo!ys at 8Z+q;$meZBRU`$hQ`KM at 7!U+J(>J`wb0VS
z!rhQ)Se}N5380d`bm+uWz|+O=fbf89?in7AfehVrm@^GmR%X6HEe#L2K0q6y3Gfk{
zA$ZTQChY9&1Ofp+2MXI&^VdPYzv*MIRG4YsaL&gej8{Bcwe>jGNr0BZ at a$QkwFVD&
zclF at l#<6MQ at G$Fb4w?ZU)6=I at LdxMvA=V6j`Zzc^q-A9MDQs;>7BIk&c;{OVL}FRB
zT$zh2Dva0A!GRaZzTxT98q9 at KGBSC<0zyQMqYlAcEw`?F=P<DjynmmYmq!ej2Nkh@
zp<`rZWVz->QW9N!d^~PUjf|MXb at wRc-@i`|AB6`7akJy7&$pXAER*Rm>OA-fO-&qc
z-n at ZN$mfI<rKKRfXyfMQmaNNl?5uMa-`uxGFCHB$tHw`i at av}!d<jGdj at T%$yFg7B
ztB&1Ae8-2L9>o#G&D2zKC#UI<SMyE4$WUqm;pV<k2&@1C03Lo$MVXtcD|}ZN01Dy*
ztjY3~#f6S-#p3qkDHKxT;yD!$OA-b2Ge{*RCDSr96JxCe{!&+&VPCPmvlGv;;=7E1
z?;y?`-)2#QLSNt7YJ2sH(p_dNJ)}sTCpRt_H(uD}yu+#{H$WZYW)2b$OG~!KwkPU5
z;ZL5(qG`+8nm?>D1Q}>!V3q2+y4?%M<>lpj1_xzqjk~<_=*M^4ILG|oSpeW at 7-vOA
z-xiKjQ&a!9huG-!CzL;?SDhzm0*D^A4Usd2D~Aag7jg0Nk%Br4{Mo`tDshW(|9);j
zLfqU2^ak|+zj8o+nVOmg1O`UDo4)#Px~-+f&TxE9X?sx(UxNb;`R*_Cdd1e-0(RU&
zpL)J>eB3g%1YDkklQUao<FWSfB*HvgSf$&LD4x3j;c}{kaiM8T6j<+Q$csz9zW6iV
zzkd%nc>gyy1o_2m<Bg<nGHUAGfOGXZ$z9Q~rmw!PE-x?Jy1P at O@AmMqR&74|FVoY!
z0s`b38XC(e6X9D)XdVVK>FDV6^!A3=)g3J at E1SjBk3Z(;UrpSV0M5#F`0$>Njt)T)
z5e>lXuCA_`te?vIoT1ksx9_^uR95D24h|0DMrlUI2-l8&XD#p~&^>!rS66Yv3 at TX~
z(x<)Yde=am0|DTcluXXrqL`$FHNyRUd`ES1(jHGG;%O0b&LDBy<e8 at 6t;NL9Z3i=Z
zH!m;jTX%JL8=XEK1y#0we7ZMZ)7jIrh}DKX?)r5q#2BdEkb;B*2U>c1ZvGh_8DRk(
zRFHn<whL5)OP6GJv<wV1PoAVX)~uzXvJ0MQV`Jm;!sV^`Csj`GI5vL+S*i027x1xx
z<hhvbf9X=!`+o6Uug3UWx9IOK^!4?zi0?a5#!EndZJYMugG_OF{$dZa=mc;IfN?^l
zUFI7{2dEgskIw5p&+Z)@yaocK>uxB=vB1c#E}enF!M^G2mLLy<EIB%atG>R7*jP$f
zgjpY2zdu#g)qHQp+H+=SW)RO at -P}--4xQR}djYZ%DKg^NvuDxLCm-Y*TUoKA4RH6Y
zV){bw`}a|x#p7?>cx5i9vaS8!Mvv;*vj^^y6uC09valFhSj2U9o~W;|FW!nQ{%!_Q
zQvC3tnvG5N+h8Nk?{jml?d_zFj*jOXD(F04%eVj?$|)!yg#^cC_w$nu{;YNP+?I9Z
zuK>k5hT&uZsku&lyaax|oXju4y_97OkJVWif2wF)8^3}tmq2EayU*lYPb`Vhe-My7
zBJltK0}2vxI1z%AyE|LuVt$2oGJo%lB2e_=Qc{s6B|@d8rM50El)Sw8Dw!cC;%$x{
zI|gD(71oPt&mJQqBisgt00NLy^)!_{Bymz&eN7|_GC1G=S;)m{w+Bnb6#_P11}7);
zE}5`8JN<+L0;otqN?czF1@~lmPh{YYM`pr!c<A+O>Y}0|aR~`jTkakn_{KQ{L$cGK
zC-nhKJh)X}VzL?ihU~w-&r at Kcl7E9HKllN>R94bPMMVVw13Ps1u+iDG+5$^k)_OvR
z4t4EcH{>*aXL0`g`Fgn+d0E4E?*#nAbND|kA7rHZ%5<zm=+=ObnIK~y33c&wxuIZ?
z>r}uepG*MGU=>Gudtzf_V>1hjw!uNB+S=Nk41Ih}2{?9LK=IS9lY>BJJHEV1{yaBF
z#mvkscl2oZyXoZ4VLSfhS5Q?Bnl``}L_wgLn40E4dUVQDRdmI&C?k7IM$I^uc`i^B
zR^4;jjcA9z#z=oK_*mWW$wqRLo1^I#T(muVmItm#y;nXd5 at ERaOE9y1@w6E*ZP0X<
zR#u<Ce~*KT(DwE%O=oB4ml4M-hPV#jyho27(L<cI0)^ZBs4Ocb^|JAJ*ozx)gYTa(
zzVRCB)AEa8fu&OlrbY`J5bm~*KP>;8AE<6?@*HoVB*K;*dHi1&jk4f9&nzq~xKnh1
zsOn at NF1C8Mb&}`jsO3p%g;Narqa!1m9}_`T^v3xQ_+^xrivmvD+S(E^FfgR0r2$C}
z6OKqE9v>eEqTB{}joW6saPr&e6R*FXeeoUkDk>^=hAd|zRk?rt`gQZ}U45H;!$i$Q
zj@}nW3kn7 at T5G|73okVl6=~8ek8ggF?n+)L9Jm4rcNo5- at ok?!LG#mz3v*LVYpG;p
zWa6^2(RFoomUiWhjXO^XCIh3RxG#!zjGv!uX=w at AUKE7g$Ljg%X>wE3q5Dib9WP$s
zNFpMRPc{TrR(u0>;`c_|hUx5mAT at M!bodrSYilbJ5lS^6`m^CaD=RC=kvu#^?#BF#
z%WzffL7&7w$NhH3{2wUQ^w4a7Do+dE>-$E(W2d00Kf}V at eC9jx+;=GF%hFesxd at dp
z*8-QV9#34X>*?>m3h;pCt)jx9Lh at 8a+>+euFv8t)M-G69^WI!`$x`sAfuFdzx_Sa+
zDaJl8FK>|=UyNNZ++IV2VnIOxm$dZ$l9CcU(;=^*fIW- at l8!hstf{GKadB~7H|puY
z=4;$`w|aKz*xGkp{$$*WgpYmas|ZvOQukJJvYm&=&a)Q@)j{?)eA!ymwVRuPKgJtb
zKL!2X8}?$T3BFoTOic6q`GcV0zLfi4R=j5!Qs9*kIkG#_+_Z?Ln4DKhb-r9fI>Pw-
z%V3`?1BT9XQ-ExBTiDYHJk~un#s-`Y4-+>w9upQ3dF35pHwfIMxLA;nkB{r%!QBTA
z8~}w2%H`tT_fR1&zB at X1Z!S)(sX2jNVP;F<5VJ~IqKvF;{~Hp#n8SzFpz1Y>f{v27
zFsc6Z*|WC(eun=02AaysL at sRXpa1K`;p?n;&;~vnzU8VO|5d0O!otEcH7pRPGjnr&
zN5$a+H`%CPE|~W{c_{-53(&B<yxce3GdTFzU0G51NiHrz at Mv4?t8DG;b^(WYT33h1
z$kXKg=AN3dZc0M<PAn|!BPS=fYz$0cq+%c?;W!>dAs`@tZ+M3O-TIVH6U}vkuf*i~
z%a>Z)>r;ui?Mh_ZylY;VX%9{T!rJN_dg;m)g(F9f;7OjbG0Xp*&;yONoPE@#kFVb7
zhO2phUtivLLU9HeKv|%ZT~6r9Trbh^da^kSN#)=%Algv<v4K!u&w}5$$hv#?ZZmW9
z*70$U{~eo(jwU}4Aw>oM%B`TlN=ZqnrLV7X<_r^5L-kLo7H7|fzj4e51pYLV^x(nV
z!v90ndw})aw*TWwOFK~!8d}m+N_$GQG-zmO(a_#IG_ at pBT1p!YiP8`acapYL+B at x~
zlKju}zMt=L{QmdxJcr|;KA-pdy3X@;zFyaL4(h!tIyw)Y=RweW`qOTwh=%gD+DyVv
z(z?GdX`~~4HFbYl=%sy0kn)1aOr7*MYt=%p-glS at qi1hAnD_d|hPj>HrC~Q<$oDsd
zEtC^3<zV6(_KKPMa^&IT$1M&Y8!SqlC;p+svp+dG7Qpa-+S)Ea90S<;k-(+#fcbkS
zxJX9Oj%@yu&b+iBl78!|C7*Cf8el3N8v)(h;r)*uwoHPr5_)=imXwv<efA8*vh>Kr
zg!0 at sG6|q+larGGRthdI!a(6>ejc6cpK2a-EMPmkf6^C1*Y5 at A!M8q7fGB(nIq+w0
zvY^-ml-j-$=h2UWHT-pdO%k-LmVx25-wpb`_<gXWL-_hhiJo))k3qrw at OV~QDhy%X
z9Af0@)2BghJm{qoUPM*%^l7l{w04aHY*+q^7Y~1GJoV+x-#K5{8#9;l)OSEnPj4oW
z;_2UVhhevq!GZNB at _$|_^o*LL$nu<V1bGT-dXQOpqXJZDivu~HUIJ(IT|46D=7!0B
zfc(e~PEv=(%yBuytA>EtSlzt&tfb_=Ko-1iCjCEpqud^y!Vt!R(!gHKno&_whDp1y
zAB+t1_j<Ru=m;FGy(=pJlyA3Q4G?(PkS_7!d-v{@ynda{(tXJG=FOYxPeg51Ejd*(
zhK5Whrl#<^Swcbr_b4z>7MN%IzyJkxxRUrC9j6_zob~#K1}nXVb^|lBI6Ujo(|cN7
zo$4rC42lq331ZGZX at RUauuCsrqBzIijWjeg0KqMPs?7qoiLqQ#o$I%5)y;{CiTi1O
z=}PDrA|hDd27lA8(RcXpVLYLBt#obEol{tN4QyI^jw?&IDp+4nuf26TAmcs!$jrh5
zaY}0#5SYP|mzkNj1#7ny7jJ?p_y7x+o}RwG{h_Zfd8jdo>BQjR;6KgHSnPX0BSTBv
znvsE_Q+yd7)phBu9vi0zmFSbaK%Fkb=Y at r||E2eW;>DA-Kuz745m*D)Dofmv#zx|=
zU%zJ0I7Z{Yo9zdxw*LFr*`%jMMX`>&W3H9%W7$hZk`S<f*vD68e(UZ&me@;HzAST)
zp(a0N_jYcbcD)>k*O`khdKR&*H at bA03P{g3g!wMSXPus^+4y$tpVrp%dU}VtGHEHJ
z{>!{dpEaNk-+$#w&rLfkf)m`^<mBY{A3h8Lz7NK5Qc#cv)?(iYL1~?zQ~}JX!1KT(
zCCBn4-;|ekM3v>_<oqi-P<-^?OHMw5TQV_01YE}3&kxvzqdNC2=fMNmxZ1v~qT-zB
z*|Q4ZoLEJik-_305HLDBdr?p}XuabC4*Lxd$gbVn=rtSv^J|#pO^_g<-m6z<CMG94
z?2mT;b at B0$FnRc4)Olv at c+YF)muyEtIUlfda99wQcu@#qp}f3&J$4l6-1>6LyLXcB
zTwX#*+mq=ImYUme1P23 at Jtrxtc<oxshxb$pi0pV5js;SytFA!4*hEDSWoBl!xqm$w
zvGVloTgJY=zKEzOrCYc7#>U1j>+6%kYG6gL?0Y{S9|djg6|bPHK<s6HJp43%yJZrq
z)xTH!teKf+*xK0C1?{b|C||e$#O;A8Mb=_5=~2~&Nmp|#tB{EaD~Ks^PyEN6S+^Tj
zu!HE3aMZiK9CB at rg9i-w@Q7q&W=3^-9ihR_jvQPUR0@@Ww9JFE5XjnwYb74kX6RP3
zfj5ILLW~A&1~I!1*krThDr>wdN5s8*A2)vX;`lJD8#j<~quru>{q}7$IqbHy*30uw
zPJ(!z^nnBQu9}*XmKKe%v9XZ-H$KcX4h#&;ZVoX|OG~S8o2JHmB-lbLD=T at f1at#O
zM at Kv=Oi5wp(n^=M*`Mg^>zh97p>_5({LBbA1QHAQ__`Zqqf=91z%8zC&i9c~QXcD7
z`T{U9^7AJJV4m^&zyQ6TaM}igGV^+pknnK8AZPU=vx5!SZJnILYisv{Tb~95?5=$J
zV^^3Ss?3f9HY_YH;~zZ0K1g`=%KONoGy9^myL<M}H~CyPA|hpU+FGoK$g#V~yU5i3
zI&QG!<HwI*zIrtb9v9(#VId|sI5=?Q0g*_)#$yP>moHxq1~ZUkxoT>vDSnTKhX+rS
zjEs!d-2_-!_lb##A?pIP1_chvRE9l0;qFB~hG$@6Uwr^%pS^o$B*3CV<(wTR4ov6s
z<cE9 at OhgCvA6FT4(!w1WHXb}+l4e7La%Gq%(1>?aX?DTjOQDgG&1OfxOMX{or!!$5
z_;u`IMh4Aod7N2QQBl!8%_1i!_uwZpDeya#>+d)UJzs;G19Mz2ISwr3U3y1d-BYvi
zR}eNqork$oIiH(T0?#|j%slIMWjM at z&i>J(N3Pp~@TI;&#g85x22wgYKTip3C2TUR
z935~ll4yH-TlM0_%;A%`99OTB6zGv72QV;b?{(>>Ju$Zi;x6Xi&vWsr%F|W(sf->0
zS`MqE(u;pGS52v at sh!>2_Gh$l*T~h|=!x2@<>-LpvWbY$8yXq{Kq{bM|A=zGy!>oH
z=gc)8d9I*!{cG1w>j&!?8>1$d_TRdn9+3^6jABSq(qeU at 8F>eeo3CKv&el2gs;n#?
z;0f3{cIN;S10Hq<Xei*$jJVSrS8G$#6irlF(}St08(<i$f`W0ibYB4VAb5jk!Pbq8
zjuL2TX<1oYcSyNvR~QSNJ4Oww>g(tCd*Q{){QL-rYnTk4BJOnJ>-+wEP<i-`-Q9Y=
zV?5<0FM%DrEz#h2yk=k!U3DiVJ^jSPhY#C{4Uu`k_t at F-f){|rgPSWe9$Mb^_xACL
zi5#o%cv at L`q`A4-{q}_$H at H^U)_{gn!2W$77-%`mHqh7ip}*hv$I at o|_wTAS<}(8g
zrUnKPo131?-qLt+R`tFLcxp?_^I^@@hJP^OSt~ksPp%-ed~VKeUmSB@{?rI}N at xKa
z at y^bU75jGt06Y&kf1X%IG}Fn=ANcpb?Bxi^fSmv^BJ!(Oue!V1Ri}9+yKXI;W41rQ
zyA_*9>a|h};;*R^`$typ6Q76+2h_-;jvZU^#=2lv*E2YzxV9Gv_|}#;2RHXgVPQHz
z4_Y#!%b?~H66jM?Q%6Qd<d=kG5*b&m!C+fjW{=met*ynUrNyMC9s~6bPlh-&Fu)8V
zFG<+4DXG>J6Cb~RRl)lbOG`Rt=7HjZpV-5amv`=$U*9b;!u|VQ#u_3O9G2nLz&oG*
z4RUpJi_gf2yK{%ox1Fy1*^oR9Sof>+Ey at tZOyg%#kM^PP(AK6)Tkh at XxPABT{+e`U
zPC}qENA{R$Ee(OBYo<#_g4kt0eVPJ$w!w=lDk+VQjs2VC&pR<a-K8-gEG!KCD+KJ*
zWJ}a-N*y+IgSpmid)1oHwAx_tNy580IqzlGsn(QVKS}`zPD)9o)RL%aXyinN-2U`)
z>0o3?ED+-a|1mT9Rrof2ea1oWCy9wFOaE{&L_xe!m|owRQpGaqr%#1$EDex!4i0J*
zGwagW at pyn_1;viOy0}OUOvl^XJGgB+?)mQY8nBp%*x07NvYRfWABjc9#O}0Bcf~vh
zA;Tiv`rO3C1Pg|-nd<LNhwR*3MfhIK&_yW-X!7cfv^sDU+W*o8Fh2u$%<{_0hv*|B
zcsHb;Asf&@9c|<2h at Hgf7#YI>`(eA4U_ITYzF;ENox68`zbuZirlO+4nw8h(<t}Se
zt=i&bX{=38ApRN{8jk+`^Zsu*joj{iyk}|H+55dz>a%vk_xg5Mi2)z>hw}Ti_@~}4
zKRTLH`TTh{US4t+7Z=pq@$qz9{;~lXqu{SfO7$OhpHngy9=P-FuR3TBBcr1LlrLVs
z90lzZ6cmK%925J(3?O{TsHWIm{{nCW8V{U$mcV+y45&6$jaCs at Fc9GJ_O3y$`$RXK
z9Dosk&XS~2p*qKL_DqBGeN_%T^R=?#u&}VWfB*izcvT__3JUOH%oobfCj+`eMn#3`
zUrE&4T41nvUewvuHM^z-es6T`nz at ZlL~!uFon1?F^AGM74Q at 8JwwWpXZmzEPA3X{M
zhW+~O+q<nd>c4zZFX~JByaFo}v(?HTcFo at 2{{1fXhP|QT5db5-D_3TBsXZ1~*Vm7g
z{&XSI*3%1tkb%wyihXu=7SyhLTna!D;NEq6dkT0Bc*NKUHV!2Xo<p760(Q=Bnwr#u
z12=4J$Ye1;odz%unalcR%6PiIY!hv51;#%*f+sRBlL2`HtnEyY!qR6TZ*FdDE|26-
zTmm#$Z&U}cswu<#N`OlA?8!-D78Vv*lXFs1aZ%zgOEvizPI7XFYO*$c`t+>4{4fwu
zXD_dCEVPifvA0)axETmhkQBvfVqzEsY-}R~Ol{u at acRJTdI+-EP^`2x(`&UGi0!~+
zg8HjJmdNGSHWmk~u<}Bg{Wg$G4BJyvDvX_!-$XY`i(b7t2ulI~{FG4y0&d-;>*(Cv
zB}Sz$Ny2yevS|qI$TTc0*s&Q2*ccm#!!ya03=9naYkv=vh0 at E*3zqG{&osG<ypOUV
zo&qDpF7rs`VZS;%buevVV8C*&y8v7K<Z7mo`}z5;I!Q-EFaZWLaj&rK)vL>}Js at uY
zsX at MdrsDdq1b(C9*el^q4==B-s+P^;N$^7)kCR at Nmc~~-c7sRA$>HqFxp6-yI~)7(
zz#f#mdSxMb?2b9O;d)H~>|TJW|N0DF at Fyz1TVg5cgw*@O!on(o{>ow}1mJY&E&yA{
zs7VvJwQ^J9YW}^eRW+^u%lU6PK~#1&#|}U)RirW at gi|6C5_TaW+S=M$%i7ClAubMC
zO965>Ha2R>Bz_uK02+&#CO<0O{;p?mGDKk}D>!9krU~$-kri)j`UM6zKR-WlG_=`K
zQ>pmPn?%R^c at Osa`1q`k91jQx at b>lX{Iv-GVS22XjfEaC9_;3YIsKr^@bp^n{q7mO
z<2|LDd;ho09dAt#x9jqW- at h9`?5E{ppa4^FSM2TTqVFv-!P1NiOP`fSozH=5VunfW
z at 5L&GzQM6(e+vji00Cf>-QC at +v}C9w!7edp72?`N&Gq~D{n^Way at 4OgPpj(a=(G<G
zQo%#+y!+hRN(C`3n`h<7KUQZjdT-Os)n>hU!vH(Fyt?`k5zw-U5^skO6-nde+lAO;
zA2OA~krTm}V9LJQV*zZj-Zd(d0HhI%=BK}=De36o2 at z(kSMTlo94~D4tZ?`8!n=2F
z&D6S{=lYr1*&&r~(^&t3?*lYHiF!rQ_WzeKh@}wNv$OEZ^OEZy?g+~)XXfRR?(OZ_
z*x1M`DiZn((l(H-%;sf`{a$(x;jJS-z=<;=ljG4Pb93|C5ZUvn#TTPmZ(wKit?lWI
z$yEy(8JUR4$Pd^U_r!_V&z~=WV8&kbn2Yf17t5NIzfO5xM at I+76F~cpzP<y3f`TC-
zA*|=lF==XQs)tJyYM(98lQ9G2b#--RWr&*eYynuila_W|CO at XThw>#bA>Z)r+RBpM
zU4M5^&pV&SJ3NBAePh7ajyS4P-UBZ4qrzoBJR{l!@KSS2OQj4yLPA24V`QD*zKPv9
zYM7&Q>V&knd`1wqc<SsV1YFzM9iN;$?|VX}<m*?>?7X~_5)wx)Uc9KPso7K?PD8M-
zsi_Hg|3l`3KY?@1Ns~mXaws}F7X2V81n~xNpXnB4YHI54kW^LmJK{*)#b`lbY7sFp
z=iS{!0p>4Xxk3y!h^=A34Z$o%*0xK^%1&-X?Z*UeP#R!5_d#g at 0X)ROaHrAYLt|rT
zt<%QF250>dlI94TP60kXWr!K<Cr at IO?C&=MAlM02s;LB07f&RsRQUWkNEJo1eVp9f
z2#~Ntcq)7SIxBY386Q{jJj%q0l~$9V4`Zf=tn3MJIXsjuE*8Xe{$!DxAHRJ|Jj5qM
z8mW9|XU8``KOf~4zJ<TP^w{_~+aYocJ($1t;SO<<$W7|Lf-USB#VRRz6g?@1S$r9^
z5RfS(35~+Kd77b^eV-Y7N<#yixj)UQDkvzJSzHW at iJ`(-2Y5R$Je<LILiSl!7CZ6-
zamOE+r7lB at aJYVVepz}QLY~Xty#Nj_E^HQurR4qn{ZDgph=E{WN7=HQZCo)icU5^G
zvCGMsEgGExzye}2;#vuClIwL1Oz9&C)=#ZOWh?TarLwap3aGaKy at HZ0_$?<TCG|>m
z4=ZRh3k&Z7v9WA^z^NXt6n&ZSw!C9zn%%9#WJLQ;ERolZI{zo=_5sXc8(gcW-5C%}
zZX at yor@*WKOP9#^VR8uOL{@tX1Km0j>uR&mS4O(@{_oz>qVk6uWo!w}%Ro`Eplo+A
zE54~o1^dF3mX=od{laDd9bH}f&z?P7QB~D)SQ2P?yhkBjKFF2i<Yd=IWdOpNg#}#K
zT=itdVhtZ(-`H72PYY)d8`v%YGd;oX>6n-zz}v at j9-FdH^WR}ogir_OoJ;=>Xf!h8
zq9OtKu?(Y;FXJn97NuAM>JRESXwG9te>D<x8kRhM`^IJesXlmuK3oMvJf2r?%w53_
zw_x_yOAVYoJaX1f;yKIh+XT4HJv}|Ht;_JB_nWP-pHH>Vnvm0wG;jikA4TR2hYycp
z72RxCUS?L-9e6y;!FL`I$jl{QPwRjZ!Tecl;iAg%5$m9_<{>dLQ9WF2uBW^EdRw~8
z*2*UqydOn$+nDPiqozLL_A3B_85spdeW~Rq=kX>gcsa|3{_^jyZM*UZ&#u`5)wx<^
z92pTo1{Q?<3uo_>-PsZZ-Fy1<>FQrYep|io_ZqR%Yz&hF1Z%h0zuywe)|nOgbAmej
zZYTxgSg%<S`T%wU1sV(Vs=nj&Ki&VA3otN{;z;J^?tW}+=7LzE3lKp{Z1a+z9}EK6
z)MxVW3#0OT-DA>s3k!MQeG?TBpoTq!SkvCq6AA=ICaHCB at HlXW>^Bx6 at t1OV`1y0(
zu6KuN0plHUKIiQ0j6EoCd`Tj}k~adnl`r{zt=QNGv;dGz>;~-YEX2TXgjo`~N>&f`
zc+UeR3D|3HZN=-|L;;g~u+;TkT?w<apJd`yLx5u|U$}r-tazsp)d-U7Vap1ztxmpw
zfIcx)=59|+SYb+r^a&b>wC;btug`W}td|z%kd%~k#xC5-$|L>d_E=z>o67~yb{o6`
zn*sLpc>I{MCb=umiv1+${+DMAgr4^i{M%vdNDmJSoO|}<ZWNPkXHiiRjxnM=emq9X
zYw>ArE(v at CL<>YE83)T3oEA+7lhwmMepJlP&c;?g*xVp6P?k$GRrNi=*C%HYO(z;9
zFw3~%(<iLKprfPvr=<n$8^_H^xJ}_*6Bid3gcku-j)T6wz2)}3w9L#D`;sKwxUih}
z{(ag9Gk=R~>xn=V;)RX%=R<fhC35pCc0^A8BKk#AN@{jC4?qs1FK8tIRqVb`L`)o?
z*0uvo8M83|@SHkz8@|`ndz2U<;`75(5H&cWCMgpx<)ChdxN0&p(B8fen0WQ_M+zK2
zf1 at p3gW>RA2*5x?V<TpTL`FucYG{PYvj&5~{CA2(B`z)w^!Xz&SM{ainDrh1`0<6M
z`IOYuXowkLkxCjGSYn0OnINwK7B33_$(Wd$!qf@>l}1MOaK&h6TU%SaG88a*7gAj<
z&2;oA;B<Xm-9FfshOF~ZNz^XtbmXBJ#9 at h?L3`@y>(_U5(6O<xVb3{u7Wk3&R#<fi
z<868q4S<OA^U-e8Upqqo0KilP{C;E1Mjl>&8|)e_*2&2UZ@&8bpS)0ru>erUszZqG
znAr=$9%zv~C at 2V>I2fCh;(^CLLhs#F3BOG~OyUGJf}Kfw3a=QH+x6u4mBzC+Wx<+&
zD0uer<pNFIq-!M(2}qamD{-IgB#2|%Zs>@cp#S?WEQ|yb+4PHyKfJv57GR77SUA=T
z{BJnLkv*v|A&P7=LUW%VorY%$!352AF&-W=Fh4MqiF;8|uxMr at 4UzOEuU|KOdwH$Y
zqVb-vSK3ftUnIte-<u2IsEzi;HDL||Z{i1M|K3SWJ%$}(^0YF-Vq>Xs^bsOlXZA$^
zFg`mwJ1lp{M17lEw;Hf9%m-4I=;W!FHGaP=K+)n<C&K;v at ggpQ`_|yQ!|*vOE-u2S
zu5(3j<I7-p0ili=Zr83wL4d|;KFq{65KEyFcpL at PXkQ%Z at fNEySPT{PdtkW4du7bU
z&0$R+7!B5)<2f|%>C;#ej?59em-Zc+cn?$o)jL`rw&=08#vItVIGX<s&JX-|^C5Da
zcjXuep!@%&_P@(5EGwf2&{|&}V>I=?-EDgC4}Qx(@h9<HWY%v at etwiB7qpkuci4~T
z5CH$~_8N|~SY2JkcE$Mp-V5cuKto&Hy~hd9`mcXYlZuFrRtD(G);aa##}8Us+Ta4+
zypI5{slzH2z$XDmuz@%>Oa#gVth~CW2D`DovrU<|P~W`*bg9C9_Tb0*dKY+~_07$X
z;Q0y)3Wz5_5NFGn6c-j2aHtFz1)fO&tbh}+!?r`b_8tvoR2A6MVtT{1KuU9h6`<s?
zfQce#ZXo|7KYyxesb@(0obefrc7^DSID`c}&z@=P=n!HbQc}{B92`Uh1O(Xx;RK)=
zF%Sp-+0X}_{b6u0%`qDuh^x4Y?rBZUaV!%?6r!WM4I7F5p|N}r4^4nX=PDh$iWyE`
zi^nuGWMe>KVWQpM?sk&6(+PkJDt><cTqTYjJT-wAasIR8f%Xv-6N9Z+ at gDwf(B at Yu
zzI#6dez!3sni46{ejq$5>M45kyn at Cd?aR+f8AY4&wJi_&t;RIj+^Jz!i&jyN;lD at h
zr!RWojm^AGLs+mt&En7!%bus%XTobYJa2eSyP25kpD9b7PEVdrT}vH$TYNtKM|^Vf
z#oxb{@&M^|jg60;Iz$ft=?HE*^uI2+yt#TTN9Q7VV5a4#*p7~87iPyAqd4MKdHDFE
zch+Z>AoA05$tRArrii<5&KpXC)uc&!+{0FX4=&}*b;Vz)I`*yD%xor_E}u43QQ`02
z?_yu>rk2NoUGOUxGcR9$rV_)$!J%ApGQGs|Mk6JQZ0duAgxU5?C1W+Z5w>E}nyjQG
zrv3_-lYiFQ<POo&%GdaA;2_HbQfx=*_N_xu?_deqRRH#u1F<Kcl!TP(s1<|5hxK)L
zS1+$TFpjjpZ<NBRIUcvnLa?*Cc{5MeM1qGW0vqSudTL1^?YF5OU^j|oJrIknKxqI_
z&Yc(fsC>9(9H{DR&jlm=esC#mcpRL>eCXIQzHT!@Lg|sdvYV0`x!M|biK-m#KwIqE
zpWYTqw;>=KkDN?MPuIl9!g28{<DZ&=mrHsrzK|_9OnT8Im at XS|DV&BS7s9xt?>f)q
z^fVCdG at H6$g6^`LTEKS4jM$$6^y}#99pX1A3HJB*pBZp^pL2UZ>63sUCv9yyhrxG`
zfPSli#1ZwF(|dBpn!K;S-<3b+7Y8S2#|y(Uo331q?xza-aY`%|lTe*S%hsJcg#B+E
znE8z>?&DB_d(qL^@CqsG1wOdPzlMV{Fp|NOOd7=o2M3d+OZzB-iRYqm0CTHMJjr_f
z%mh3uI8;t{_GOUPCbj;_dt2i)IK9@;&`=u~9XPxuE?Ze#c(QVVUT1Rxw=Fyimfyd6
z`!-M5vZ?)hsbw>88G9cUPC>!Aujw++9d(B)-NJG9EkN;=xX|)jeGgu|cu|qaTgOgJ
zOgu+C0J8ZR1iAah0(n4G=K9M|gFe#+4CjACWj)LD=#r%K$Uc6PcMmnwBy-X)v=^Jn
z6`Iya_O7bOhJ+CHzqX~vC4;54nQ6~l at 6k8wkXov)FBdI4dHQq|HH)k|uy^<65e471
zDXOzJ?ZIHp%Eu+$pN@=NH~*Ru3GmJ=8L;!(H+6bA=r5<fzW#<nsL4Bbx~VlXy-SXc
z0t<sxkKf$tQ at jv&f|Hb#l$?qx2ly)darpBA>Pk>NjZICRuzxdvq7(D;>VJO!$^&iI
z1=_r=TzaT}j(q{2^OZFPm`ogIO5oDtN6^sZ;X{P0h`<_1y>=Rw?SJdU0xNH5;!J()
z-2IOqKZZ~fF<jKp$l3VWs|$K{ur?qC=#9-t-EGn9b^BI-{hBsPmnX(W3fK{CY-nf)
z3_c|#l_I;ltaMb^f(W*)<(8D^+UlwetmrzJS_fE0GbmW!UyB<1NhqIgE)1kBxTP?Y
zhu-;`Dxoy+#^ITa-{xgsU&(T(;dIcn1!nI9Z`4F-nwzuX3jnIlf|OBK?yhvZIoFj(
z6B-_#5EVu7iJX{HLEy3Oty_G+ at S|c+$XsHNk5E)nvi;{i+o2OD9OH#!6f>W~PPgM?
zZLN&spz8;!9P2}VOWLZcWT7D;mqCk!n8Xn?v2k;U3z$}`;`m~S=+C`+Zhd=k#AE(@
z{5!8Df%&HV+mkJib-{;Zg7#AFPZ3Iwf#ton>*1AjpP`kIl<cg#z5mpiGY=qWS0-}F
zbCO~Y`cr^zO>rz*0A*uFPlkXGB<yT1sOc2wnP~23-KVFgr~Jr6P0VR175(qHl;<HX
zR<g^W8js8P>eY~OadL*j+W(7@@M|xP%Eyed_NhVq)P{F{X;>B(%Oa;)TU#q?*Uhst
zk)RbqZ%0Jr-BcU!J72%p^wGDKgOMQ+J`Xc6JkTlNU}wK}?V78z^HYe}=F%M>OV(GE
z0hC=_Twa^klK{s#8Lz4g4w?sGrE6f2)R(4ts|YW`GnK-zMK_u>p*)BgQP&C8xz3!Z
zW+wq`)pGQ#-JzV^+)hM_k0A#b^u;h8;gxmE3>ajdT~A at h&c at FIfKKC<G0yGnZ5%d?
zqqA{lcHr*Hg}LGP?@J0g|7Pm(&H{4efs1Ry8fc_TCj)CVcOB0M33-T#Nfqdt`%F9A
z=x=he<s_haI-Z_luxQb+9fVw;K=93kvjk!n$PT}ynhObB8i?<P#(ZB$3JMBd`)=47
zmD?T8$;<1y&!#kkvmdv2b^zwiPPj^dCJ)Dgc3^l96B3ke- at bk5=+R5K7$BvR{0Fq_
zgdSev{Iflsflnamza_-U!GY78J?6S2eb;9cSOV7Lo$X!`Kh@*41qX*cI{#~N at L5TT
z(9X`!;!s(#uz#AHGiPROj9ytC7<?a;0j_4VIp5d)!ce3rCs`B_ at 3noeK!2qhcbc@%
z1NbsxQc{FU;Mu8w*%gUAb?i6~J?L*>f0dVDnuJ?egN1Ma at W$>cFDG;hiF4<i`L%;D
zU|THExEK94=QDwYl2cGTV^a#d43DEx6x4kXXke&s43;MpI1auOz4cvaXTGd0ANI+(
zWt=lwuu~x5s4hg)nejL|1Pz=NosdhU9swp*pmkUVf=p#)Wye#6yEe-sAG)i3+=?cX
zser>gMGsL|r|cUT=z_Q#>UHZQ-ev(Z&AuHDe0G<V=E2*y;sFrT*JtxGp0q9s#j(ic
zp0)qh9!AB?BO;RUFF=q|X9F~){AoDtmRC}8KlYe}qOh<qLcXh~XKwt-3!Q)kt%XW|
zfJD3w{Nsxu6(h)7$H9ESjqVyhuQDl4r4fEmv;QtuZGco`V`F at HIB5)%@Ew)dW44=l
zTayzL?eJOMQdVHOI<R2{YL5jD9XyCJ<FIEPy%0HUi2L5I52D<03D=M!lPVQ3*c?DR
zU1en=kJ--i-&unrAmFG0TU_7Wb^|UuW)!<GoaYcZ?qX|(Tt<;;%_SE9WrCQaXYaAI
zvxj`uNPm!#p#|XqqZ*DUAtoW|e0WMX-)m`z>n&Sj@%z0?;ExlFi<<D?#K?%er>7_W
zvAMbFKK=Fhy|}nMP)fXfeA%80{n`cwOdB&DO5d$CpDKdIfJx?S8P&LqlR)GQC!u7L
zN4WJG4tQ)INl3Z8vte&k<;f2)a?iBdTOl(uQ}o6cCNK|ruIC5HmLG!ObyWMTMMS9b
zK6+rlmkX3<Bg=u2k+D!Wj~e4xp?<MuTZSAyf$!FGrcgSMymI2%bRJ+tK%a6z-{=Bg
z_gMUqOhR))9Snu_!i5V(lLwgOZ`|O*paw{%&aIis8Wa?CROALF6^rbT2{CtmUf!%^
zk()?RHWmj{68W#)Soxz4rvD7sP8WcgO;hv{yahFn*A(8{UDG-$>wk_Ds0rv=+uf~|
zkGb}f*v{hg)fZ2f-v3Rv`}X1 at v$$h6c&;<Qntf9Y^96JT``#inf?ii5A|l=qkS5*b
z_WJKUelp;Ud3y<fkPkp5ajv at i*0MQz4FXY>=R!rIbt1wbusU1d-MH5~t1W`~lU%)?
zym3E%7$ac-%ZZ^rE~yNd3wxapj1A)zaK~u??Kp7VOr&Vu%d88-?=xVxJbr%Ds42~S
z`I5i)ATiU_@@T!N_2)zH0)C5~S5X0-m=KbXZm8&?vmGb5r;6j^z+>aGva-I|If at MI
z`xoZm)ogo<O!N#4R7Xcg|HXm+^5-Q5A)ya|Ga|72{c`MoA0&r-DnmBFW@~x0yU0Y!
zJVB1){FNzz|0vFRjTEsQbgW>4{r~jzJhJy8G+J$w``eB$Jl+)ZFeOC|oHw_)cIQ&W
zl}r1>&VeHL-Iyajefo68yLT2pDqOYz>60&+)<xrxU~Jq%OS_L<5q@{2MeseVoLq*z
zEmsRLV{9Q#ONU)s;qBRgHP{;xQa(sZiU7kuDj*<$GrJ*9<H)Zf>)BR#V;nl9%5kte
zUwc0e$^f3&`-;e544`IfG at JzwLn4*$;502*3*(kE5oI^qvS6oaqZ;w#1N(ff&*~Og
zA7DzxM(I;ul9x9(!wmRT04|)ty^E}cFPWO2z@~Lru-D)J@$+YLbXlC=0wN<07K3nK
zFiwN*du2V-T|f`<6es(FR>OAMPoD<c`=rLl#}kwqL_>&k;XfaI5;hlGn*r^UXODjx
z^mh-Z4!~o(Pk?!<4A%y*#jE1T1Jw1Gavs<llHeE)?0V|x=m<|vcHqDVFwFajiB09A
z1oTs|4_i|yPM51Z9Xx`@?;SpUKR*5g{EG~xNU6k~pu!Pc1v<pQmhr3)zKMwFH4bKA
zWygUQ_2D$1epI at xz#g4sV?!~zFJwiTU_rQdrD3`)!xHwd<n3E5*8#2)SLL<j%C8pu
z??eUgKP;aD at ZmaJ`}oO|djL9^g~cp(Ys83p2|F`kL!jCKzf75^w|_3<1Rg!!Mp71k
zTroHPW-C>7b>&a7$JriT62AHMSi;P&$hf$01HLAR-W7 at LtgKIoFXbHE+uPe(tXjgq
zsPj8>=8TZ<x=lo*1<qauDKlpDpz(f2MqDTb<Hy!kDgfceT#b~9L at U|@2M#a+<4jFW
zy>4wC3c-wxi;D!fg0ULCkoFl75p2Z*0ocIM at IznUk&`OXcpV1ptZrXqWF#3C6){~V
z#5A0`U72Vl43zeNqh3nhylE02_P1R9({?K%KK^!N6k~l`8;x=Z{3Ht)SE#YTw9%7&
zZK`T&^{^YXOiX0#IA#^+oi!*igeYd#eS3KgM4rsXvETIEnt(!e@~_IuN`Zn at uYHxG
zjZ`M1p%LHqP*AsMXLkryyp3Z^!3$x(?0c!DJm=-XCUtj+zNAJm3R-~E-fy<TM&V!x
zdFlBlFH3;a>uUMVeJ?!;4Df3?2SH-<=g&Bi1RJ=4WWe^2PD9nBMv32W at J7Jz%}}s;
zYzVygBNA{O9U>Am9G+0MJ~?|Ab_DpVhU01<NUrV_3p~CM6Jt1k{=9L^!I7k_EOs1Z
z11x(#T_#u==9Klmq8x)S#F3xICMICfc!UBHJE72xh%W)V3%4ZlW1D7W_S*ovwY#fo
zQ42)**WL<%6V{hg5>Q9P#a;OM?Ij+uS~QShB9{F{LXy>y5BqVxiA5>X%WH2{{QaeS
zAGj-KULFDLfISOU<-mTX3<AbHx7<1E&Vl)?tgJ9WH2E(NG(@m=ix8;JxY3hf1{|h?
zV<%KO6yc9j*U3v(*48fkYVz{(q at Po&Iik7AneItMbLVDf<M_ss^77EQI2zzir_ at zm
zv4SO*STvF&^x(Ktki|N>x|NAn1c0{SP{+A%FHfse76qq>JAE7;X2H3u3EWy0iS<VS
zZgGyti<d8P6l8*d)faKV1b`HRWc4rzXshpz>s`Ja43PB<G at h{<9W8D9ZG=DwK+{HH
z!TX9OfExfKBc`FDX=rF56Q;TVk(q>^OFbC&5%`B`t$%Yl$4d|x4ZxOq+`^BFJ4R;&
z?Irg*$PsjbZYO?z^`#=PxNbKluf-pyzL<+Dz$0L*Jc!9Stm$T3`XqQ3 at R2xdI{e`S
z0nT-!mT#8_jKwp2-YbQ59UXKyV+vGQWg_QwcDzi*XSjQJKkWI+!rNi;@ZZT2;OqLu
zjM!H(P0Fhxk)uwCJd_Yfj?GYYrEhGZ^HvH$)azgKeRn}^gI}-D_t7&77+=?U92~kb
z-zS12)Pe2bc%k{eviO9A`f|RZRbVSP-II^u+4(4jb3i(K9}q>WYid$s^FwiQag01*
zkkyOF5?dyjK!N};W6!Ahp_-IlU*hlQ3-l}id4Lnj!oM`d9&h=5z~k at kTAZ(oC;UD3
zNh#tyu>26VjGco+dEqof1b~`vhZq>xB_)k&+)QALA(Sn{6BgwhAe%WqOaB&|ZPL?Q
zYBM*i&-G9MyBdMrqQimSV4Z9SBWVvFWT!lq%%)#ttN;R2nVk-(ZPV{_MwyqdgFB%0
z!{14IEk*;+mCwBF%&$>Lh$98@!bVO`4pt{5JbWDV68Ig at d2MYv3k!?xgh=d`{r$T^
zW-2YM&`AVuF5$+N25?Kfn23#y#VlNGkDZdzQqD&Zi at D(sIMqfL7pb|qx%rG<M|dUV
zv&i^{%=J8f2D|OOF?YDv)w`|bu^=6QmzkLvSd0t*rx_fm;IlTBfA!_juNiV}<;N1H
zA#&ghB)|?Q at m7;k3*uwPkMnm96BDT?1KcZe96XS6gPhC;Ugr8-cL6}#*37fGkVbat
zQJ`nobs!~$8A$xc@>FuP*1{{#N5;qZL(q<s-BCRkYEox|GiCq)LjWj>iv<Jswr at r~
zp|||}&!j63zSJo=1bUoNOmq^jLddA8!yY|4jIA!qkC`q{wKA)xo}&{J6T=n2AO~gx
z;FiQ=Yis*2K^Pq+U}k2<Udz=B`{~=`K;IYAuz}&6b{wj1M&#7yef0gE$7#FXA}Y|w
z at X(*K#<kNJZHM1^P=Gy$sd9V-Via<miin9*M&`KN)E7+a^A8BX2{J{TE3vSZc*$#O
zYP$93*KlZP5yMx21R>i_4%HkTcVNSht6o3%e*@z4d$muQQ1ey!7>E$y{FrCP!oqS)
z(mlfdqBU;X`fL|lLUa2oYc(7q1#rlPE&V#O&Qttjb;f_KRlGW#?+JKpxqWXIFt_`c
za&QKufNAwVu<ZmYmEJ4tK)opo^vJ=OUR--y^n-rrOC<2vO!&WX3$4&Wa6EZX03)-r
zk+9~@{1<|)J(q^4w6wIkK{LF~vtyC<XTp%6%>Dr=X{gCV;&V!{x4>G+C at DiAP?VV0
zhYVixkq-=%#qKFvzke(7KC)<ks(?8b|8fr|WhSjd*XebSXaQ((Ve4swYu7A6?0#(S
zBea9SjK}owLa!YMpNOXfe4cd0DSyW$-D!8PVA2*gz)^`gsvz*#jsKGk*um>teFuRH
zPERtG7px8k{T)s78u|bl|C3bE=YRgektBGIP?SSPMmD1gP6ObCu@?Vg`QH5eEp%3Z
za4ci0NIZG at -isG}l*c6xUA=nM;C-M>uj^+}?O49s{qh>)nPf_YJOJhjfV at l&E-oCq
z0|6L5oOX(0Iklldx%&5F752H#QG1M^M-CAjKuUo<{`&e%2SZi6IxN_qoedm#QF!$w
z-nUYu#E}cB0D2_~niXvpQ!$H?{Z#uB48U67({p%Zu7_TX|CRje+8Pdv0vehHGzW7V
zdLNuq_<6pcn3xzxAYn`Nv-VtkItQ;m!a>Jp&KzWq|99cIG{+oH?1GM$S7ws-E-9gi
zmp_0Pdw{s-BqUaDoY4VB!G7|jf-=sW_&1Jm)ajbm_=euzPkJX}xtTj-`G$TSj&=a1
z^Y!ajgHnso_I-!ikHEWOd;l647#O(qv-o{Qq8?$6P9*FHCewcV_Kiu#SM(q~JwyoB
zI$jzUSsY1t-N7LSr at 7UKQQfGSMWHe}GExUo|73|x2Rrx$z#K)Mu_0}9b2E|zV2yxZ
zm5I-$fOF3Ls1%7H1MjFdyag1ozorMmD~4cUOA>N&a?oQg`6R{x{{E~9&C8gS^x-Zk
z;3X~Xby{<2?Yg98A8D7Bu|}YAQ$2*n1bf>4Y;wH|lJF)pvRidy3`a#m?QZ`LX9=|V
z_9uoHJ`wx9f+ii+yUKvG0D;OWDk^%PJ>viEh2iq*D%O4X>VRRdU;pPm8;&6BJwrs~
zeb-pv6Br>W#As~D2NLq|@_5|j#INoD%LR}Fcr*vPW2~m__VXJ**t=1UM5a)$N-%be
zWx<gQKuuD_9F{>Cxbk0oiXCIooN<aXqo66c=Tkz^ltd7-0SkNBsR>Vg1)Bkx?gncJ
z(5HQF)*$#S4uODm13zRJ61seD_k(hCTigFa((?9n21P(P!&kie1R(j|*T*I)`NUpA
z<-+sx{aP8a=PzG|_diy0qxGnH{Da^Qq-C&eql=5NCojZ0^M{LoOD+SJer#+c4%%Ho
z1?`Q;)U{h<6cDiZOskW0`A?lXMMh43=lS!~n at huO1q=jT<^UPa9v=Adot>TMdJY~x
zP6bfx{_`6ZX!~)!sNmT(&}Jc_p*Y}GSXkJ&g<$y&9qLnjw<IBjBxTO)F46%zhd at Yz
zg`EGrxzPIN?$Yk|8Y}xK2eZh{rsyM>rT6Am at _BVa2A)Wrf(8hCtnsSL;OVaXr0}86
zo}MZ=_7KDUBpI0vhnyS>Cei_fLog3bFo^+{K>~3e=lQ_en12j805&jYlnAa<Pc|M&
zX#62>r2#l|^a&a4IJ`F1iuuTSzC>GE at F)Bxe4mi#yunxh8?&7`IA4|a*%5giorCE5
zpi|xWLp?ByQ&3RQ)FIZ~Q95v!DLp+MlRJBL_U!{XhBG^{!`r`th`TT*Ak_0oaNU6H
zSnP2rV3uYE#`f&DhHH<*w?Ee}p3)14_k>7pe)DEH=nT+ww{>L*2`5*2&4MyPSA&SA
zx9~oJB?LIumXD7T_^5FUUHcvELp(Tm{4NRAtEBC7*q$Ahq5qA;>07eA*G2W1N3Z|_
z$PxbV?e%qc_nCH-q^WUj2bRM at E~ux7Q32pkFbYf<#cCJ0$8XMd<)NVf%XzZU;32 at u
zPHP)*zw&%ST3U>GeV76t!#JqUX`|Q<AD|q-Q7Iwb;RHdIx5WUt_Pqi)v%0H`9-tW8
z!-L)7<hzDXpB``3RBpYQlhFGM45PZbx;;;eWf#J$am(}jfQ^8)BY;Iy4SN98Bh%C2
zpqIynYW!464R`#mY{pRys9Os7T-0A2tjgQW2&Z9j&BEc}Ku-Wf!i3~)vzCSb@*E=f
zxN8YUlse19%li>RKL%g$5j;T^;Scsn91fmr(rOfbbaMw at 1E~`5cf98sY?yuhT9lSf
zGxePLWl8_7^f0 at TdXbV3jAs5w@qsoRZA_O)95|dK<<Jj52c(aI&){hpwdKy>=E4Bx
zeD&y?9ceVRb#-+WvT0|HP-aIWQ(0M=I<t at Wmu{`aymTW&ZuZ9BpWisy2HRuXc5xL=
zCh-Ha!BiEjs0X<NdJX$*+Z(3Vh3}f%3im`G5ov$t0VL?*D0~z4U2HGBavDCrX2{ms
z1K5m(h8Ac8QAg_f at D*j<#a_X$u!KT(-8?`5NFn+G(As=2VdGLy3YPo3n{YTKH}U=1
z58*2f5J*G9!f at 7nIQ8){<-1&8Mt>U-l2RW-Jpk)}W4+5~arixrl$6x74Z$Rlfj16!
zYieo?1ve!;=b5r~3O)qH9 at egwrJ<tYcaswpeOS7ACZpw_ at lSE}pFfjxYo$*Zk;5wO
z13KEIxheLzLZizKk2P at OJnY)JzJ~jPvJPJtN<%`3{I at 3`<5)WExZvPG1>g<utuozu
zI>|21wjKhmf^vo5ucE$*5BG#_vUE2*Rk+Lh<!{(lti+jH--Rlws_;hk#X#r7VIZVo
z5fNmm5^iB0zo*czK+8Ih)DdXTzX!=hM at I)N3y&w5qN3I_uf8;5WjC(zE&oyj;Whsc
zKL<zlmp`ZVi-IFAKflUfb4x_xs@>v>bT7yV?4yjOV{s>B?x&}NN{`|a5ut~$s0gap
zP|jSh%lmdz6ou^8$o^xp{z+E`CT)eS9e-31+`M^n=vGz}%I_p0b3!m4RL_SF9dhPx
zo)97q(o$2aANg>1bZY8qmW<AjhougP$w+$cI}teI&wQ`;0G{=VI`khpYya)gw<qK_
zRv>=?DUn0;>O`xA=1eyya3g-;NKz4zE`ulQ9}n at Ck0m`Zl`ZjHFamOhV?XUrsXWWi
zCp#|XiNpCMBu3AiJo)U)zMIwwA+3#XPy3plK7D$~>j5|!BT#Tm{Ee{_my%+}cZA)K
z at jduTETOOO$@i~`Z7FJ|oj8yWPo(zt_Eaxk6fRr<R*K+1 at YYGZ&_rys^Kr$i*6H0c
zkDpin$9GlU@;`3r$WdcNXa;z#Hi)*1oqT)hla~72NQ;QAlDRn>gP<uTFn0`)^CY?8
z7gR&qsWih~m#O@{NpkzDeMD~Ex)m+CGJLGXL1Pxv4imIw3Fs7fAC0{o4(w73Fa5Jv
zcY8ljyokt1;U^j+4$*Xn4&mhSBLSE7>6_NoxP(2^Ehuz5$*YggqyPzNun_*2a84#2
z7Bs7Ez-}5dGh`r2F`>u%&(7K2SLZ@$w(B2Q6ymHM;7H$JSyS}te6CYIVrH0 at X-mkZ
z>#OHtvm#7U8_n(7SKOwzhwM2OVDkihlJ0h0k9U at 0raTD!LjB at c``)4>J)YLfv%iI6
zDrE*9oz_SC9)I$}(H_V5fq!V}=*;y}rqn^e;<Wf8>ykV0y8!ao_44XM{^}b&-q1&n
zu0*;_cX#V;cyN8*SNr&+ZmzllZfaVRk%0k9zzwc2_3P9w*H&Kh?(~-1>jQTWqw{-b
zGj>qTF-}5C>Zf}lFa3hB!zmt~V0Er~U at VSeJ|)s$21(W2cDGzM6z{EeiZpnAi2<Ps
zyEO`YPNh^|I}mFSy_cVAyC3Sc_>3KWz&pc at 1;h&n^@~kWmmrhgUYm}wyGCChPD4tE
zBN#c07#SEoe)vG3?8Io@|NT3wpkQ%>6RPBYXPmFUdKtxXW%HU^#PoJurQ0+}%DZ)S
z^3}c at _v|GYS{|DPMv_n+Wy;wt%wMF7RK_~rpGPm*3wr=yBS1q0bLI~{{O}SdL at p9w
zS9ofm%bz<ha?9qXWGi at u&8 at hHUTGAb^WO>l{QM|cWPey_mWZE+- at u9T@Zn?&>bJ at J
z-a4^Zok?yymagqM>ofiJ_&F&l%P&d7T-p^bCP!xO70gajl$Mr)X^{w?5jxDJ at r20B
z%S+m4HH7{*APXxy;v44M;b0ulbpWxu5NYsWwg0uP?wHe;#<yy6SWkwTFrX!mD2D*5
zyX9+2VKUJ(G4e;3IDLODA^@7IzdM^K_{#c|@#=%xfBw@{z at b&5lmh?$@dAHpXljC}
zR66pnm;Kr;^I9?mqV>$v8wA*EyB_k|+S=5Nbdw{wrhzH0yA4eZ9J=gmYy`5hvN#ac
zkiXk;;0-^Lz9MU;=u<m~QgFN;H8nMM(C*byul*sA24q3NcXR##`<=A3G=&7qisgaB
zr?28MlGoDE*?TpZr$7Il4&<xY=Sceg>xGk{1#<#11+t%VHBx{_@#VI~YnWDh<7k+=
zZ{HYzCmEDnCs+~>F6U4`8#h-Eap1a>6HdM;%6XWnaF+x_Au9PT at dF1A5YUYR2eQ`C
zx_XrXA_ETK-nV;HYvPqzMuAgbspSP)I-}P%w{`P0v5aMVdmD4C0n$Sg`LCUMn&`82
zwbofUo$2slVu0Wu6)yEQV&$Pu!{Hadtn>lg6WQC_zc8-s<k`e97pZ&);$s)jCg1I?
zEpJ%Ke@`$^MbjTY9s at xI#Ma#_Qiato`)NHu3(olg=<n|s2rW)5$;T03MMXt;>~Qla
zniwf5sS4xK6r~@zr(R}dv16eru-vVkpT*{umiyTiA>c+coL>8o*IFa<?Ul6}PS?Q<
zK+xbIr*YO3N at V2_&xQWd+&p(bb$y%>2jnb8;{+d#+piB}6O)sfj)U)*&RD62&K{YQ
zno_)RgUjyQi%6ffsUxkqxx1cVy&KhpmT%U!FJEA~1SW?fq}1}$(bgNFe~ilQLcv$Y
zz)@AouF<84JCV>m>bvz5yquLCbu&m>BD&G}zB10ht{OitTDq)f&o=%aH&p~iJl9(s
z2g_X=c&|*}@ZiCNcx(@nley7!)<08^cV(iPp7Pi^TICQ>|3}V;bMfbX$-bD-USh6P
zCT+e`wN!gIA%R{{j9u6|!(2(K%l0aGk8%ir#^aeho+U|rY{m`-iSt`QuYETEnkoVO
zzv(V14RGu0gfEQB$vr(iF}Jc;$KZlxq=}wh2w=+73vnl^d92Dmbn{y2cC~PeIsPDn
zt^QFPkYa!8h3jl5C+u#jw9f<j-!3jLtiDw(&jptHjs2^tr at gmVb9`Ubo<af8{an9|
zC+1^P5)+`n_*82O at Cy>RYkU at wm}=hFcLYMB<=|C2JKi5ZexMBg=5T~rARzMeK<@n@
zBt8&8vU77^IjR-tW%>PD%-7AMe(>Od!P}cWxyFGhzPk-m;F6;t^zu!r-drpXJ&gwd
z=H}+cEkaDX8>ilo9XNoq)+-VT>Rz`jbk2fb6OoeMjlfBdQH+9T65j~V_B>~l^q8|c
zbge5-E4->o%89{df9^HYq{79{q*b>UrZq=_ic6+>axDM7c=6%~2yNKW%E`$IO%6L2
zwH)LLJvFUuf;Dn!X~)6a3swR_VbLBw916j%?Vrz|KTe4I>~0AIZ<(K;UpGCM^D-SM
zLOq~4h at I;H^G92IdkXfk(4aX&mKH|KSZ*#@Bpj9eYteczQFeLdw at _%M>;T}xbxX^;
z4GqfQ95j-I2ow|)UJR6q0y^FW#^cI=*990GNcpYX9^h;>F5 at b+bjYYSYBbX%Z%32>
zX^t}N#sjDBVzUg9-<g at 2OtSvxp4mPDXVWl(5EKeF6+wnAej9Q#pL}o98L+GyXw*{1
z at nu&x^7}18JxLBkL`T1HPN2|jx%;F<ltqs}-)Cj)e9r8VF{!DKFAU2tJry+Kvp&A1
zs_g7+CcPU%1&!URs;a>Te3+IQZQ<0Nuf4psHr{ifp-}P at NNMs=#ee`%?yT%Mu?+7~
zBJ!1YP4)D$V>Qxc{Bj>VnC#}i8aBEl-Np!JaQT at E6>!`z`%?@!idz^6`?J2&f at Xjf
zw<Exj8Y>xeIRMp$hK8`FxL1d1@?$7P$*Wg5ysjdVu<mt07}bJ$Ucsf}^(0fD-byRb
z=kU*pc*CMYbP0PlK#&-UjXbcN0Kl4Q-I7?%d-J=zWn_&-Be}PakAlEs=4)C<$SEjF
zab^|?3C3rgf<vaJrZa<8;viN at M*ixPKiMm|{3i&|1>3_{ynCm}9-qIIA>~C4ZW#9V
zt$4Q1DZD%w4QCnqC;lYoy0q`QYRz(%_xNUixxF1TUTvA_^Be-lC7~Pbcy_^h-O9 at 9
z`quKOv#aYJP#R!yhGSm5IJxlHF-ibxc<Qa~ZJZQlU8`-~W;pbRo?A0G0yt8E3uAM2
zwY0aN-`xo3-K~{NwMTzbo{(k&LOWqZ@|=j680CIy-_r1Vxd<}3y{%`OL3yjCu#vk<
zL4TDt=6cKlp at M^hUA2}*o&`N-?B?R;e#~@##BT`8DdOU2esa%GPEBP2Ua1W9m@$9;
z9d`Jvf#_h at x24yo+Sb<A=)fx=>ev0(i_NAp`@mFNhS?3a8ZQ3+?#P|~4$iit2^b=K
zdl?}fz34vzb^z<}$!8>QL7e3EX(^A$cu{Tb^!zW=<)ts9ijO`Ba`!22QY_HJMCI|O
z7|X$LPtMwp%+G&$-l%)n{PUzZ%Q&|~&~uISo;R)8OCl?wRbESoKJRx|fr2uP=4Avb
zF7%h59qM}H|AHs3dD0b!n&SnGo15;E^xwIj)9VWZjGLRA*fbkV?(@P(4;M?KKFLgf
zV*5tlHj_TAA0?35>b`=(hb;?Zv$ftUgc>Pg#B?}bdQx+~JzGXtm`=ubJ-DVu21xBn
zGj|5ajD$x7Ob*^V>!+;F9Qm12J6PeeAHXKnK|udRf2~)*ZuXhKR$1!Z8xrOGZfIrh
zvpQ&H|22amqo!85b?b%clCu8$UQW-^2W_&yDO;bMJ>9W%`m(UoQ1u5O^EepJ+uQp(
zFvx1 at oCuAOq<cZakLEa%8>1NQtBYT;9gJ)*GLZs at B>@?2l+!Az-b7J$vyFt#e5CI7
zi=T!S&TJ86@$vE1N3 at 7uFZGhXb{ao(R`Y<(<6b?>d++%8zU8L(iluj~%O=l|PLNy1
zaDxu`gx3GlllSgRrkGiSS7(B{#e&WZlA0t7Sa#tk(WT)G4=JELDtdZ{o{PD^94`Ok
zs2*1!)T|}<2QPiDc23><*<1Y6_`+<|=bL~C`;*m4fL%;XSYfRUd%QeX0{GtJYqhgo
zk4;k7rrYXaJ1t6?S`zv9KfUJjQ50}jFoI0n at yGR^ST4X0{B6K at ktanG4GxE8F1Sp6
zO`~A{c!4EAZz{dWbKT*W?f853)_~IPO{V3)jOPFa)2_%AZYIg547R08&~N<g6;F}Q
zKcKd%(9^KRsg`K-D4ng=lA7lb`PSA}{_x{co~5dM23si!iANQVGe2IfE($0A-4Y_r
z^Z5A<M+h3cbzA%)VQ}r*ILL$>gNa33zklD2xMZX2(12AyX;NNy at +ZPBJQg$~q02O_
z at wM(0xoVuX5TG<BY$7^msiym$fMqpQY<HvA47fA)|LoP_3j>S}&Y$Q%_~w+L6V9ns
z4gsEJw=3naM#${oXfp5C7cOquL at 8?9*}(HA`ZkE51TKwb_zX9G;(cX*YSdqD6DBR3
zy!>JPXyv`3o7T3r^e0^#8zSjDUKomeo!+E)m<=>_rni_S at 8}6>@8IQS=kEgpm;rH8
zC0brZMJ2Q}Gt*M1vCG0DFr3YH=2`O3(&FA?v--xyMjK{B*YnApWgwS<N2 at GF^Tg4e
zRM9iq`l&;3URhbs`~A9HysF-f at Nn?fI}unpw^0t<OM&Ac{15Z_=YGOAfD7GdeZtwv
zqeas(&|FveRW3&-5&%5$`O=`*@OfX~cVF8IW&=~i9ORV^DxA%emrDG%U1^c46>Pk<
zObDl80lZCG_{sxxn3erUl~<ey6|Y>c$Nejlx=LY`9W!N8p7WRgCbsRFNG|+WWvMn8
z-%}glw;@wIorX1cIGE|V*qd8 at gaEEqH*c2NU10od&tWsbdMZFe|NUu`^l2cI0lzmV
zcl>~uUJ|7}Yuk~l+1(o}=95SPd~|evJ{n@)s~aadIf=n_tG^%6OmTXl)@NvF)EM;a
zfXS+|ZN|+`eX3Tu{%U=J?^Rxcz}5|queW~rqFip*<NcoJ%{ibvYy5#0&#QU<Xz8Xz
zE^Vk;WA6qCHaAKWcEQL&E{HmMXE|&eXrG{<q3PWU*uQ^&i?LbRjhH4;-j?tuqB8 at D
z?UQ}mNA))r1|Ey&0nssWWYl^K#DJBZh=@qf`;LmONkKtDzV<VfBT|hXuD6A`)_%=)
zh3nVmE2jJ^D$dZG%20n8`A#-Z%vw`Hfk4u0 at ryxnUexeUBU86l@L7=k5rb7;Zv3IY
ztZv*G10I4C8?jska34D^Si9Vz)m;(HJu2-j6p{a<*$LDj7+A(mgGJU`?h7H|;aEHL
z$Wioihlu{f{=+$qMBqZibfe3o^}czTFVhpaweCcK<Y`M;QI>xj_-D0Ne^yL at e*EFP
ziE{J$u-h{W91HO$FZ^3G<T6fj at 7_I8*)p*Wf0vj?jUrEm at -qJf8(T!LG;nLB%d^Mh
zl*<==?Idq3L~|`K23NZL{_G!TYtl6JB^g^)AU4~JEe_RCuy1mcKOf2w+o%#4j+qrc
z%%vUkOU`>ac;$8FM<&a!y!V2XuDk%bZ3mqCQks_9U(K#z{j0Gym&Rv0o?%D3nwlD4
z5pt6jdNFLM314eio*4<^;TiD%Xss~*KSLVAj$P}=o?PaC^=oBC+sx9gh*EoULDppV
zab%@n;`b9R&CLqHO(5heD05Cx&E0w9tG7 at J?>u|Ua(i_$|LV&peG>{A1`lXFXW9us
ze&}V8n;81y8K3j~cVVFHfS$}fMIdc{Et+^F5509V+mPX!N_^xM*Sz~FT;d$aHyq*h
zc|p+p9vH=ujrD6LTkBm~3d$%tgJ0!YF%EkxO^R!4YqPRPpO7(1ixnP!&}s9oO<6kk
zr3%-y;2N8mO$&F1zmCO9?Y=5(Jq%&+Y+-9>V4fat{KJRYu?hDc6BPd5nF#ItyYAp7
zQj-|00bH0g>w$fQdNTOX2Ve%LPMtDVBdGHaD$ca8{XX;m0Mtt5qE71)wt4}Iwnx(w
z7JKt+>ZrQXrAy at 1)zzQ3Zmmc#HWs`XYR at yZv<M`nx1GsKrm-}$Y>LM81MW95=aa?#
z_I7<}>7L&Wc=vRU_V;XF#`5Z<8X!3kNAtgXe`bDEmhE^wlDIclTU;!tnR-t9#vNZE
zVK_>u(l{0%qS|{!QKnA!-CduV_DoiGKC at cmfR~`5ZN9xY0y5JmYb9$gTVSv<r*nLI
zIvi(T$;q)`u at tZj1!c^e34R;3KY|QA`{PiJAKT*=?aP;g-~ZiXb{Kd=OlLmVU0@`4
z^Zc*zu`zk&<>h6bQYM`6(eqrNl+L^@U1qY@(sJ#8CDK6KPBXa$mr38cbxU6PpVro%
z*(E*MtIx0UW3G>nkB|4t7%@)ZpryNW=T7(g$+95*L+(a~hL6bjw$i<dj4KJ*EjEaH
z0{`sT{@6}d`}~~owZr{~wO5meuUt)kE-CR`uiy?m7WR-zPEJOJZftBUujg8#nZV#M
zryQ_glW?;#>t<UzlhE~j_Oll*T&N!!O5bRv?|<4V{_X8NA5FlTvbC(4VgI|W!p!Xp
z1KyqAC#L`HdMhj}{Gt!<ps`94Xz90GLNTWv5Bo=DBnb$*+lnMB{?Va>R{)-?=<Y7s
zt3x@}6vNEQ9+OAN+&OgB#J=wpCpJV&O+ALk->mHCBqi;t_lEqIY}J)ABpWaP73-<E
zRG{~<r{|GAcTI|GrHQl&`<?Rgv+2 at 4SzT+I$!Bt+i?+7B%Zu)Pz9ME=c7qCtxZ!}t
zRe{I#pFbC>e-dEH&zC7yG?Od%BA3iVU*2K1`>X$<L?OKnj))KZ;|ubX9X1!|7{bPQ
z1aiOIswtX0%ffWd`RPs#*+Wh#DQ3JN>nXf~K>)sD#V%8ZkxcmTUyDK#s-wXXyg$5&
zy{r1_Kpx|<t}4gDatByM$|GlpAa;E_*P)wormA|is`@o0>K9ga5FSTFC9YhP(Mp%r
z{+lQ`9vSTTXQ%&-OEOblH)k=>*L8rL=s<36Ze<mfM2QayDZaH8wV(e#vfcu!s;%oA
z$4)>%1wo`+Is_>tBt$||Qd(N_AR#IuAR-_lB_Lf&NVjxLhjcd_y21ZkKF|Ap-#7ku
z+%fLG-gD00d#yQtvG!VXPTY9ZcX|2T6ls1vD(rDdUA>Pm+4lJfEl$#|Y}#K-h3=B&
z^|@?@@rsO!iL%$eYZZLtB}1<_eSEf#4RsbUI^|?!hRkxP-7xipamf?C{mpq#h+nc=
zwYAbR<L2vE2*kw1x-U_xPBKr2oY?<DgT9Rl1WiwSy?uLn%f^9Kp?@qsKxAU-Z3J5O
zu#5WwdMqR{QB3s~sw{3P6;mij{MXoE#g8tpUUo1|r-Tn}fAQ=}zpU!|9iZ;+Ad0~6
z- at mue<C|AKc3TCFdx~IubG}P6+0+Lcb_$h68ZLJ9w=SUm>-_g<XAN?D%O*TbORYmV
zX`O;%W-LWM3AEB(3qGU^gQRQrk}RjQ9T(NE5d5nZ>~Xxu;FIz;24oC{#9KdN_LHw%
zi1J50{^}f&8P?Dhe-4qWF&r>AvODGiMV60B{89&<M=x+S-%FwwO>SE~>JRds=<4R=
z<e(NiO_s9p7P?Q1TU$3ZPv);}Z`V!K`uUp&v!2Dj^#K%BhVkcban6!`8jbC|dX}36
zDNkMezvk~W9hAi`8}iNf{vdmYN8Xfo`So8za5v8#9E{v^65gY>3qI$MkB=8Lu<O&!
zz$xkN*zlXd|Et6R_0U4AG_8p{jzst;l{`agJw at L4V&dYc;4bQ<K}t+&u62U^u7mXC
z(PVu9s*n;L9Sy8U_vSGfStMb_aF+flA8B_ at MJFddm_|;=1r>A_laG&Y0ViJ}Muaks
zE15X0lg<5WX<Bs&APCL~p7Rv~Ho(IN8z8rvdD2*1e>bfZJnJ|q at irxeaVy$mocn+F
z0`RJ~D^<L7 at 1@^asGg1Su+A at GM9yI2Oy)Z_uDh<=mcEP39*zzUDtePS9--D>WoZeW
zU0mWokfKc&mt_YLjClGCdrhE?hQQZP9_mqSnn|l%4O5-LgoMoclidLu18Ty!)5q8M
z1VKnYr^5e*SraALazn>Ju%I$uqn-3_T9s#)8=JnEi)3Kl*lX(St^Fs*>Vmb=3sAXe
zq42y-sgF0!>`G!H?Dkp<c75m9#X|XlnY!6SDra at F9l?=!N^h9Cyvr?6?E_qmE8kw|
zNxzsW$j|@m|G?1#P(G>Eb6=*=!P+j>DE=L={c8LIWx3nnsxLaKs$~3ZobO7$FM)8I
zWj;p0{$0-|ZmffA`L5`-{e2nZ<3@?_6#wIiioT>4GqBG-=e6=JnRd$N*X147Ty>f)
zM6!$;2>~Q*Sv at gR)6 at EYej2ZCZoc*OJeBk)Qf)%w^1}x{sk_9=0{r~1d>jVuQ3?p$
z0FF|xAnFmt$H)KC)qXB%!V47OYag^_D}GLim50Dr;_M&8h)zdS$^V6#WD&HP32g}M
zkfI0)QyH6>K=0F2Jdz&?>uKv5uYE|ec(;AZ4tH?U7s89(wv$pl>smY~`NNYf9|zT^
zT;&!g-sD>iPWZ8xKV>~aJGf*+n7=1xiiPg$gktOJ!Y&3E78Yg|H;nZEcVXf|wyKbg
zsgcpIy1KJq_uy|3(+e`%2yH&oz&N~CVU~=4;NN69k!<mz>MUGVs>EqoBTp#;yrWO8
z0a~+M%Wb8z?xH{2^!xW&rAEnt{ecb!wyquz+^7QsndR)xmmt0UISuRP%JzISmO0()
zZhpKe at c2Oi(=2GKH~2}3iIu~{<fqQzbu$TzyVc)B0!biTFqWw;bzo<<Erxwf3pACg
zLqO_3O?BqX8A=0;NJ2JWUS}xqM9PK(dPR3z1BbdQIp$^BaVwzp{Y|G68lb0_mY2bi
z%YJ3F`B0ef)1y|O`c{`3P4GtOULboe9cTcFnY>hzn6=f_Uriw_{^mJxhiW+{tz6zG
z=OT#1tpCX=tARWVZ`;W_TZ2e5IkOefZ?4~s<|5en_Rp(uls%9ibe%pLY2W6IL at PQu
z{r*Y24O5N5Xh)qqFWQSuNqtk;CAVg4!u{6&5mNsPDfqX9a0tw_15DXIGs$@D&oh9J
zwU3pgNtTa#Q&4~jr{v_}$(0aS8obCTC|J@`wmeuMPD^MueBNz$mC8rDEbOQ~R-BR=
zm58bjpg7m at HwH)VhbHe80zx7p-UX&IvY<43-4M>DrKNkD`P?ehK`G~=-!JC<{Mor6
zkXKZJ!RlTi05=gB9_~9dq`hUG0>+Pd=g6+IJ5?E()(qn;F=uDz at Q*5n$~h*Ft`Jy{
zmjCD)O?CU8m6bK*%DlTYT6{Lrtxb?C{6kN$rp15O`?J54i_CQerZ94(w_urB+Eg#?
zyvOqYpa#t&L3hsN|8jFIK-1oudFI()j>9+&opa{B`9?KHf942nwi1Qxxfz=+PNct^
zp|F~iVMB)&`m+t6Jgfky+})iFc=9j?SFY%AM(eP~=tZnWWaHI^&?fXh9<lxXj7}S%
z^V~DVRLfR9d7UNHYEC*`yzUEE#g7J6CYe<|B-VcOXK+s2l{gopwl2xZ4F9%-F)}i?
z%a-KjIeHZrGGhq|3GYduig+yZ6hh8fLpv?J1Em=EqyMfXUDs*4*~8XyZCN6%S+JuP
z>&Tk9_Y18kgP+Q~Jc3RY#CROv5D^i{%FC-&*K_v*3vSEmr~Xf{FlTFArtX|q8H8j)
z?TNwCvaQ{Aiwe~F3hHg^!iid9?9WuGSjE+!R>?o;bp{IMJ}7f3QHZ$hUeVXrHy>oq
zO`Y{yOurr7;n3Rt@?M6%tB~#P>iFD19*f1i#ijp8i8u_ko%!$Dw8cETMivg-)c*?(
zXdn8e>$jiMS+SvYMAW-5zMpxS7 at a)a*Ier=<^3l at 0425-`}zt~6#CZR;-XLSn03BY
zB})X!s0T_X#O|Q+a;5eJpH=i^?+-0X7gtUHSf@?jW``|3@!4_>P1noL2e%Z^$>(2x
z_hw^q(xiHB5&6~L9Yj3T#EQ>bxr3I5Rx1f;9-udSZ84&K|AB at v^A(AC?E2xJ{g&A?
z)ijk#AjO`Yu>TPwdNX5yNoup=52U0*pN##wqrzaUSL|>B4D>mr#z(yXj!DOS04e7P
zLY_<+wG<}sM67TV1em1<HSRULH4+aP7#e!c%{|L7);Rt9_iy9tk~#iKZm7iP!%#L<
zSb&lm=1I(5A++?@gUHDFywjFMq@?c^ME2ei4a%Ve8QIcNRl&@WvfL=1X>PykyS%Mz
zyyNJ5Zp$i?#PiIh>siI`?%lgn_mSN1+c!#i-nqN#nQ<=Sqb~lwd*u5`v6&_!AKd8}
z7(9JYG5Czctao9usG4+RV`G8}^TW}Jw}F9ITn=|RqeR>zX4+yDz~IrDnwp~XqW+}3
zva_3z=}p>^!mfNLzFi=qZ-+-`x^pKKNAdps*L6H#gv4RmKY#ryht7`DNPP?}D-#Ap
z1;2FV&!0aXVD#sXj}FiW%FD}DN}Q~s@(2k|P*PKSiQT{G_2#oI9aD(xmLqjsRh8I}
zXyr)slDWIk)=>80L$vR^qu6nA4j6fAaq%HS^u-(f=p>a&nTsRZHBeF_n3$Mon>K#5
zGs-|6SLV92u(k6zHMes#EZAd10NXHa$IQq`39Q*`38;e(DqGBxc>*V1#H!C=>R!#?
z&E_ANdXnM1w@!yn_vUS;tENB5)Gae$b=PU1AzRdPT97q<QT6#LH<sk$;IR%&$v(E?
zug1d|{MFR|0~Io%rV}gyw}j&WdVtpn+}z#g8-p2)Gq{yRKo>=2aC^lbTcz2Cq^GkG
z1j*EoxG@*_DQD^m-+g(=j}E88AepU>RVFAh@^@67W~NBh%wuMmPv18L5nLy`@0Q5*
zcPdV9%2mhW`}U<^S~|L*f`U7JT>Q!3zP*B^klI at 4iT5R9cntS2SLt~!_gM$^zk__Y
zGOrjgAJ?j+tr!2?mQ_=`ZuRG<SFzKw?8wLn&&xySVuyJ`5E|`cm7Yv&Y#*aVJ#=TV
z64XKRfBQ{F%iU>1n^VoYk|2|5dY1_Zre|k~ItDS7p=%O4?TP-)Z)3QXD0)+sz0o;I
z35nU?QC-_g&A+@~S{qhxJI8r at pR?Hy&GTaCBqf^MuIsEt&6TI-=NmQmVX%44DNc1Q
z6N~t#KBMO44e42XVmP_7BG0Z at K2_sX@$!i0xPX3>Rf?AI!f=tjcU*k at jgA%`cHL5r
zV$mpm+c$4KJsX>vno^wn4By*onW-2)VAaYu>cLYcxw})_IbbLufrp2O4#*;tL-4O=
zgsN-Wc34?i*>8{Z#hW<FEuLS-Jkhz!mKM1*-Lm2bTGM~{YA`LR?2qmTGmvN<-~>{o
z0&Z>2w30<|8lIix+YHCXKRlkSzWEit^q4thYG$S`0+*egy{Wl*((RUwq{CeMYagGF
z6rvd`>=!RwP#RS~+VQ$XJO at lecl_YY=meqRg8qNK3EKU!zP|o8n}cS2VnWQW=&QvI
zBU4kUADVfO>(e%7+tf$9=-9J5W=t$tPiQ41CI<QWU1G=&Pl71>y0t}eO1#f4`w2s~
zj!jtCAr>rt_J9=IH?mu|Zk2uvV0qKX$wWVcnTvc^9<j%<ok*&{yC@))sBV?+ba{MZ
zZQR$_my){w&!5!HRHu`SHpCKXPcir`EG$<D?Ck9yYKyvTJ}dIL<GlLx3IW<S-wl2X
zNMgZNr^Kliq`4n18Q&`(4x@%ZyRDl7b_q3Gg at ycA2p0M>@o}DY#GUau|Mzf1LBK}*
zV77Zk1ZSx~`;>$YW?^G}{Um|@-ukrWpF7uO6A5`OdQo%AqAyh!(LuGy$VmAT7cT!M
zVjXSm=Q}GpKmKM=Q?R%Pv>pF=q`9ADWr;^2f_kXs=JNDfexMM&wLFk#$m;ISh$Xv=
z1hedt#6=v_-(Ne-`cgV~SXkO%f-h57eIyrbN{AiE`w7vkIjE{sEN|5LQDpBGE_yG^
zUcUP#Z!>^g at GSiB=g(JrElI at YZN6LldGh2#f(?3;#gg;GFY{-R2TyHq-Aqrq`W3%g
ze%p!D7JTwich6*NXqymQi5NCb-x-}?pl`mUGbop&SfNq_Y(jP24>d*4NaPShOH?;*
zr1seGh6EJ58R}##K2FXW{(>XNkdcNCnf;Dxme!Pe+BBU{UvlBXg at 3OT_2=4S6`4b>
zEnGf-K1GpOeR}w-fW2aYwN7uRI^$Jz>IH0iZ3M1+$4nVFxgOUOdUHQEWBIP6NB1SJ
zMKi|!uZXz!{5c0YQj~Z3H!9k`zqP28uEyG6#Pxw(@VbnQjCM1Zz}v9*gJEg+J<n27
z3%yvEm^|w5*8s<-S9VIIV=!u^1cUvE3X~N4<u2Id9wk50eE03H&Wp8)+OvuAe$juA
z4-}5HMEn`d)AeF(D<|ge$~UUGr!;2pZ3%tT6N-Ph_xK3~tvZRJkS%tkEU?jtQ~}FD
zZ&~|HuKp8F&g8b-)_wIPKeQT86(XB$Gb+++nKF{D&1Twq_M6sG#v+ErKh}_04xN{|
zLIBfgqEcc>c;kkz^>FcxS~atZhTp&60r`MJe^_H4;r4K}d<n=(oH^u;4;ZdbgcY3K
z?BTEcf{RPb%5L~_^6InMY8jJe=#-pgU|_gH;IuqooJH<F;J&|`NoB8h`-iaFO;y3h
zkrCa at ioQHywS8I@D}Kz%*Iw+oF^^l_Z|QcwmFMHGk=s8-{%w7ot9W?5ydD^cGDMat
zNV?npk^^gwBe|a;_um_?c&JPGeTnbeI<r$#sH at 0efsJYFyI6FZQ6fRaw`+8?qQwT>
z=sD^)0GdAM0kfoh59&TVY<lQ-L3sa}QM^Yfzgt*jk}rVVuaS}Rmige5OW&H^$dghU
z)b+~UE}&(zy!{Ba7q9*K?hL*NlDT>j9ehH2aOIb{TmNezGK at dJYCP8cmBp)9xBbYy
zw}F%Qj$gBz>J7AIHP@`G at C((d2Wq}Y=zDs8e*UYO^K2C6o=QGeC#I-HuGO!1x;WO1
zLm+gpAN3DLl1Bd09R{BAd5{qwZVGhvYw~nv=m?_nd`wIiBqb%$-oh`ppPZzA^W(>l
zeN5Wfsg>apPzQ?iQSWc2yKF5`QoHR8V^CWKd3ijLOFiVR5yvZ`dM{59#EN_4QBvR5
zDOPb;NW at m-i=HOXuYQL+y|A{kQx7TtEy^Mt_3!lvTKK?O1;O{mXKZoNZ}jB>u0piX
zeXw$Xun0)LYZc%DbC;{NrM9|{S>ZgXc<H`nGdw)}qL0MS&kyxV(JghZ%)6?F4x1Tq
zJqedTX-GssaDp105kUPWz40kf!yo|xfkO8KK44vMKC6MxyOHnRzHbVP92AIz5a?~l
z8yOk-J~%^GQ(rHB`}$- at ywvxB4zKlc{lR>z`x0}51?;-zZoVMLZ;HOWnP=Y9$->c*
zFBI_p{Yiq#)>d3}lonX}=g*&6?gvh2vz>&BL0lMp$w0TSTz|**AOFf~Z+t+Nwb>Yt
zpC<8HS*TjE at m-I?dy)Tt#3Ehj8>kDeGpT%YCgn*E{bH~>+=XA8$iW9yGC6~V={qt1
zgI8{udd!AX4MFGe at Mbj68h#+<y%=@Zk=b5na6M=lkHKo;YSJKK(oOY8Fw77urnIc|
z99wo;!Na?|&Rh5Dz#_AB3m7+D^)EK>%V;%w(U)tcjA{d`l5v~<t^vAeF8QczH6j?i
zR}y!y|JipS?qPTih?FY?=xsYp8)|y9x3`DRa<oMYpP1YJ?`lbgjI{JC3F^v!Cprxr
z&|7xf$p`=4`5phjpBclIeoy8<ug%S?YC&|d$bn{{Vq`pz-j+O8R%Uq=ihJ3g7>4~q
zs>;PwmG`0SdZdrq;FI?yRyQ}NTAiuCmA(9r8Cv+%7r|rp2DP?cTU$fH3u!cEConNF
z(M=S1*J19PyjG-(ZY0L=+5f^BCj-Mdv^TEMmd1iF_w~Pb>V!|7hN4qN=9(OAE8pRg
zDdw50PEAi|n at ob!36!Sw@$)nM53Ds~3=QSHm{fv$`gL(^<j@|;!oot5#IJpPqUQTi
zRPKKSF)Jzxq<UFTU+>>3{=y)qtf+_s3O!!_q{r&m2_JB=7rrOh;Mqnd%_{y8_Ba+o
zdp60)$Q%#V+_<IPm*=`xy6Qb}!w(EDENB0Y5;O$#m6n%(prk|q4~EK4|N3?B3IPbg
zs<d0Gu8`OIcL({OkG&fBC@}SJMP(uK- at 9~l0EjxsC$IR25c5784E;%hSa8lb4($JM
zj`{DJKHl%oOv+bCNE(Wx)+X!Yfh~;r9A_>uF)=OFYDpwy$E<WIh^mef`_;nGI{pJV
z+4y-8BK!AwJq=Tb#v)Uftbp#G4Hxs$%YSXiyXvPuT<rK9gSlfKlS;4Gy63Sp{6hP4
z&+)m{w#(I7sjuQ*QUCO9p#RDA^E~r=vS*qZsm3pIEPK=*W~$v`ZSE{~9K1}|!9}DZ
zA*psE#GT+i!4nCkEBBqk!m-!@*B37}*<Q5>aaX^n(B1f{&fT)d_4P$`j(eM#brM8v
zdxA{xd7Kw6E_Q8>vn7-PA{d)l!Uj}g%FNun6O!NsQ$@kGYuADl8C9GQ=JZ|cOGD?@
z?8>K}+;L@*$|&eHS=%df3H$KjGT}uphxtyra&i6R<h;C)9f9mTi#|deQUN<!urErk
zuE at o7Gz8I<sA7FxY;+tj$A2fwrED^$)c})|nA~U{OO{|6q9P(>4YD;dE-2|Z?))0l
zqexf#F`-*vQ2YL%d?piXFG|K|S%__8(=J?nEsB}%Omx-NQdImUj5$8sbB$(UV|zJ<
z<#pX&stB<#F>z}1Enu%WTo)Jh&Nz<Nc~RblDjf8^-0YAFa4Q2r&Km?uc>wFP-uaL{
zC?<v?^OO)~WhnOBw{L8HqH!ziamu~>!(EX<Kix5?43(y)ro;ZGSyAXU5)x at ZnD0Z)
z^J(Xfa0M~$$BU>Gg|K;;!<Ng)z<?|O7XJ6y<KyG*-rntYyAnHvC4al({Sua{_336N
zMn>tC>BH4(d?QpDf at roRE`2$AMO%=f=fT1}CXU4_%sXT_s(8O}t2jL-ow}#HI}ntt
z*!Av&<m6zD+o)h$bJ_QiZwq}k#ulPmy<2f5?)1hL6+$h*_0qSmD+9ca#%@NOi%J>n
zSv15P*A4Ie9WF{1TEWCJ?3^{WQV$6U3HtKodTnj(R?~qnn`(u%Xvz at aTxUlI70%}V
zVy5Wo7L1RPiOF_>-Q7FOW0NkJg>I(K-rg?<jRjj<TRaYPiUWnV@!HMv85wRXWqU_2
z`!|G(=$)n`wo}56_qAchA4Lo8#bO7e+7e2($_4u at M9_$Xyti>bUGCuEp!n<=<DdTi
zz@#L4;hiCS*}Mxe%yY6O6DxhgSW}A4Xq~;kI%QHKbF)3!kFJrD%Hp7US3B+_9UVa;
zds9TrEG({5u8rcDOkeziGQnVmlAkiO at go(2eRGhM1ULhwFLF9t7ziE9J~|`^4EA7m
z<LkzX_jO#mEiPZaj7l51j%EzamaP8~TG^Ubst~2PDiDLJ7Y9d0MQI0{QAp}q<&IcC
zB-~mlb+_ at qFUrg&X<=nWr=_JeQ0$mK)EvQ~|7M8TRCFolXwiOnuf%bSv{fo*DnwnE
zih?3W*F#TaxX{+~tr!)sjO>cB;}+JgqEC25V;hXHjJUWbPCG-4o0!CKMf7Gq5rcf-
zyLV?*)zzDSe5e2hpdlwGUznQ<Y(Btj*?ezV&X)@Lvv>aD#W>XZav8HIBOpA#vz2?i
z?Xdh%Z!SQ1Y4#j0sl;qs47T%NllE5JUu?kga7n>ng@=TMM8S$j(=mBzOu6DYimeg+
zb$&-F9cNe9u4eX%@9kU^+nChv-)Z53gy&Cx{_>@F*}@4kJ8_lMDb{Dr(N{tYG)f}5
zpi at KH_r<=<;Mmx4;R`xVb&ZX_2Zz;FRma<I8X6=!hn{>V)m2KSGH-1jPG&Hz$bTry
zdkGjMsqVg3OU(0p>|UM>hw2$eN5_Qtc=$!~#~R at QX<c{k-u?XfvqM{(!`46i!8BEj
zx-xfyDP^LieH;37+PL=pwW;OhA3Tv8J2TGg9h0+ejp^1-Tjf-lTch~zrNSYS0z*SX
zA6s2-J?u+*6xmQ&c|T(^7T<od4ws9Qb5E-*hS%cy3CbX8hqgruwF*~@{~LS~z+`oH
zcb8%d!|%QrV<<CeB?WmN-_#^4S+{Wql_;~Dmi?KVOB*CD;o{<gia_asRTTDrU0i%u
zv&@AT)vppPFH6583?udTa8|UjrR4(;FE1GP_k5Q9_pGd}@Q;dq`>u>vdyB3LaC7Gz
z99zyF^!D|2<+O=BeE04hB()S36_v{4$Lj at s^))r^;1$h)q7qV5Wng$vr72YLI!f3z
z|6zrxdNkHyrsc-Tdq)R56<DJadI^u2PzlNJKYqa7`eL6?Euc!5&z?!6j-#2j;D7dy
zc1F#%m;M0SRJo(}S7uuaJ)a{YT+n#b-`f1j)r*ZD`S5wZq^rAI0u=9Nx3U6LLMbRX
zLcdyvv0o>=KJPH%ACMg%ZnPou9uX12<F@;JAWQ$0ql3d{ihgI*;tLB47PRmJT1Q4p
zD-vd<;m at D&eCuHs1 at XM1qVQK0*tq^4R|ECt7v0!;O#D*8=u4Vx4LJ}*cI*$CLp~)Z
zGqi?rdiGVm%Q*J;Eq=M|rb6bMVmj)Z(Y+`24-WaaiUk1<rD+K-Uc5-n&d%;-RkuDJ
zs`m*7QC6-nZGEdGm at 6-)cw at lZ-Lz!T^zqC=bM?}n9B&>To at cYaqujhZ)5XNa;r0$}
z{#K;-Nxi`5cbs<ZA9TN<YNP%`tKj2pR at N6-`zxu(jtBbFw?tol&CSg{kAoT?<GZ=b
z$Xqyc?p$zm^ex<*{DIM~D+0y#vv}-%B?AMRKeMvFIWG3Xpv6lCd{Sgg5O&>WOeoEm
zTbLcu+M3dD+nI{k&5}|#EFQIAUS8h(V;W;ykqJMwcW|H#aymdHc%}DXC0A#+y`$rP
zWo0FgLCyKRyYnmhw1gLyN6I2Ct*m5d2|FNUTt{uqW;!$?W%mYD0EiRP(;w!Tw7N`L
z?6pTN9t~$!#QY8p*;yX!o3^94bBj9eNsdVnT!@#4hq2X-TB&bQ8na(m#;YLHorl$Q
za^m};R}l?<(QNG3>x9_Y`zXk#rX(^ej_-zqhW=gNnQwA0XIKtj(bg+oKVbP3H%p|Y
zso4P0jq0Wsg>G at Io5ycCX|3yT#g`Z=mI<6~#gvr|M(|po&j4jk>Gpz^Z5SHT&S1Ff
zxN|(xQxPdNqv9g2r9~DV9-cZ(LPMYg)}8gqPl=UjhphwWioUHnjqP&xgF?y4$U8Rl
zaVu>^n at N&>*E_xn*iUKJ>7l6-5)zu)K3-_IE;-nlb35-P#>K%wNlZ-aGVVoutEj^S
z(>!lhk;$-CkZhs1FXnMuKp<T0yT*N*2Oq{3B7%d7&~E at +0y)Dd#kn3B4sQV*K4&OR
zug0vt+!rHv>vk1wnCvEZ1WNE|e|6&29D0??vZ3S=3ZWZ`(+)b&C*`1TX{Duht9K^=
z%gY~?+{tWyTlDtr4-j#T)YM{B(~f;CnWa at ntXTB|7G~zc*}e>&gI_yJDkTZ&ZKA&;
zCRS>83ymEo4qQ8~AxsR<GhCRdCDx-=P*Cvd at jA%k%+1SdtgCw$?Q#5q`S$HnY at 6eB
z31y}TM!ngi-SuDpY%>cBWjD9ds^%}r3k&9?Tt?^1Haj1>d5cMu362cT+uD`;Oex=S
z&H39+W|8yR%(Ao#9;C6qKg`_Rd>7Aa&ku+PFmKXPehFX~)pYCZ=}A0s>TElAn~*p-
z33?nJ=hZP+-N8uv$HPZ>jg!JI_N)bQ-90@>(9Q47#?12KNji@)4EOdvn5^@USBihY
z<e>OQ36DiR!_6pJy90t)c%y~8yTZexskL=l at TS+?k`Ln|PC(ix(<8Mn*pG7b at n<ew
z5A^r<7bCiZ%B&U_7kfMuF)U`Fq4C5C%**2)ja{YZ<qdIgbi7uqR14NvMo%wBEkips
ztLBMn!exI)Rr(NFGaw6qzLeF|K93^#g6>g%^C!FWNrhSU*`LetOMsMRe~%Anubf9^
z6lVm#>b{<0UHG590K{Zu4{_$|{K?$B?eHasl}X9SiewlZJBEhj4C_9i3M|=9m#Tvr
zjd)Cd-y{`u`Yv{T9fRt*nmv2gju~}N&f>a}bcdIZ57pdk`1R|X!UqA=9aBN{sD{oo
zQzQEgn_hVfMx({>sa4v67Es6h!op)iLz>apl&HG82Oz!1uP<foNo*yiO}=t#WuRwZ
z2uAJ0H(KvpWC>PM0N+fRt?@mpC?7t}#l;oZ-md8NxsUh|p|X*YL~Cm+AlxGOYH4wC
z at nwY)Mfw*nUeFTyGBPko#ZIH0Mhdqhl&`tvHs3N`zO)$7*ro?zp9fy^XF&mWI^{Ed
zx~IoZp`%n3YMq1@>o6p8_^PU^L9wy1RR`>752|mi!9ayag;=w(%LCQ=+$mIKkPcuR
zTYkU>l2fL at W|Ur8S$QDa at Z4xDhcIn$V?$qG=oeP4It-&70R at GEL3Y^4i&=6o3M8Kz
zjdHDr?*hkw1gOFAx12(Y{z*wDp?rkupHbl?oWP0-QTQDU8g|Tzw%zmRiU*!9zPCOl
zJd$u&>}9ocbWDsD_kL_(5Eo});H;*pDVvm(bo!ndeV_G6sf==xT<puE{jH+n-QR*(
z)HuP*+k1ojy=LIQu#*P}LM4SaH#e`A7i!Q4cYLCInBwvG`0BN5iS0joUPD+y2)9o)
z24BEo)hqvs4Rp5MUDaRc%cKzgd$@5?;jXNl9IcLyPJ|%FI%Y(8SHJ(BQyW07qO|mR
zG`)6qT<n}^f)WxFC3SUm-KkG_=@(NoFnFU1hyw!yn*chi<R7)K5)%{QG&MAg34V5E
zoQUoO4*^P#0SDNhvhU~BA6Z#}`&)}rV22(@3>Vxu1`^aPvZLF=j>f!y|Ng4(alX}{
zH070lDGMG($$9T`m$HC>fRT-j0zEzb=a7(E48L7Zf4?-y7dl)Dk*_rdoWci7IsaT+
zwSmhC3*}+GCr at senVEI0443TTNiL)Fmq36k0v+M43~3CiH_uJEimZ84yrhg(ltynj
ztC~$zVqlC=t^wy|WpQ|YIqHz22!eclXtaRpof{b&OUAgZUl@&TXnk|$QbSu?fWH*e
zD+y}Q$P}8z4yjemx9&PHT8)<Lj8AjzB+ohhnapt4KjPdg()0K(_~rH!F<`^8n<4A^
zj;q*P1uH4RFe8D{(b1X(ech_c%0!UiAL$jMzpri?afN^A&D2vk`TCWv`<AMg^_Lch
z80L>=jSn9_yo4(Ao<4cPsz2*B_W6D;s3+v4l#}Jx$S-!Cvgpg8qoJWm%gC4!yjmpS
zyh;gy?y}y<kXl|IjeQ;_k_P{bcAmHZ!Fu(^jYbGo8q3l0C>TU`-vLl>==I-^A3s#o
z)ur_H^^KlBy@!J&$Hf5eqKfTpC(Tl4PFxa>SD at _o@T912{kfEynVGz|F&ivR8;2IF
zqS_N??g}R3pt727-r^DxOiWCu4D8SBY^8(RPdhZhjdP$wXkOpD2QFw*@MZ{o+y0N+
zqfz_&E)ha6xNeY=J{A)@$<B%TXP`3qU@(r4upS$Po at vD;w_sMc8c#Px31UDjokjdm
z%DN&gBSQtbib}Y`n646fav7fbBdq-Rv6#HPJTiE9?Po4$W;@U)08 at 0jV#|`hB){O0
zQ|GyJr>g{8{*XGz2Q(rhJ9_{&a4HF0KTbO&Z6k(pN3lkDdSyH0F54$j$Azn?M&@-^
zO$iV_Zr;_a2|s at LU%#z^1Z;9)A^*q(zuPVc?hUTDmX?-tAlFK3D>-zif*Vn(YLqtM
zjZ_wEeNiL0ac_#wu}V)F1%&`z*Ja+cl$0^S+eyUPd_A5>q at lNc<<*LUh=bswf0zH}
z0G!5~Prhp6irMW_5Dh|J6Nu=0Or66MNd6yy$}ok?19=a}rwej(QSsoi{e|?4>{`R6
zhwIHqd&8`{c~jp3;zkvFv(6k}%+jwu%f-i+1a62Wcc`~l4SW--K<xFUStyKp(6}sb
z+iR<SYV!V3b_S{FOQEv0I<hj7pmhVsRowtBO!p)3OYD&}oWcx2(v1B4VJ$5!&g7#b
zB^dVk#l_&99Pa7a*^TSxHlC;PCDE}1h9g0C_Uu_%$0lRFe3vcr&b~fu at ZuZQKVTp`
zO$nEX2nic%YThm_Ep at w)lgEEuUtfPV)ewZ at C!Z?!;**M(!WX>60F%ANn#)`d7goF@
zBgtSqnRu?x90Sjp?QhN_#~>pk<K`XDp|jwHT*HJPyR@=$@Zr%>$c%$=(hjmv5+_ca
zP<isC(c(4$eNvu9-zON+)^Kib>(TO73n90I?LiylZLhOwOM}~oU;8xmq#|UzB>2-O
zvHbjeBWvr7t9|n*;=vdr?6n>)24rGBfrHZH<f&6kbaY<W=U3B`lHOfs*Oe2C4MeA#
z%d2DGzrSc_Zy*2q)Va<8(c^~eAtPS6MwkGAA5>rVEZ$9N)CXs4LA6|*nwmOQMOF~A
zk^m|u(97$L`_WdPWwKsU+cRtHB#^cdEv>CcKa}sykbRO(bn~XefQQ%wpR_?<Cw$db
zf#CdUBTGv;fCbgT+BGm%NL6ghylXi^D<?5A6e!=!&6!8X+ab`H7#V#es0qq;C;U`Y
zRckTQmpAS-Ha7C;SDglLnE<z|VKsOHlvI+so_omt4psqOt at 0d2%GSz?`wa_a&Zf73
zu*9l2&Z*Sx)dVvrd at 3&&@uv`tcE=BsYHDjsww<gCdj0xDZB5PU_4+?KrZO-w at RZJe
z_j|m&yn>`@7l_D0B_$={-wZAtO*0A!eT5q$7k#Ncj{1-j;xzsKt>*lv<%sF7Oe8js
z{lnp9 at Gxj2N_q8DlEn7#=e70pGIBcr_sD*R&}XkTg|L9{egqD5vscT8sh0n=J-h0S
zu1}%BOI?vR<ubrtQ<Ib8iHR_NHHQkb?}&hC+zpoQk)yMd?$(B(kRV+(zC<!oQr;r_
zb2LskS7<PxQgY3@=t1qozIuh at Wuw+0R8B7F6bSqg!lM4z-Mvh#NyufB at yyw?Xb=E~
zf9B^W-5GUL)}Q9a4~>d?1P0Z}+&t~7JHO<D8apj3tN+mOaFN#+2<Y|qe)H8J^Nu#V
z6hN~+MJmar;^+%q{ZX4Q5d$i<3)A}DA$g4R&<sKW1u$Ir+c(N$r)APnD%-fYxMyI!
z(cD)YZIg|TbI3BoSiZY_L(Ig)#9^tQv%Gp7;h#lj*#p2oH+T0xcqwuP)6>&vw}P((
zwacXcozHf^(onN+mw7t0phiSQv=28l|2s<1&AXPvPDxKsK`Y;q%Fxgd4Fj)5?`G<m
z2Q*S26HtRxEC-1|V_O at 8tE(#)4^P6V$KNRIbCNTts=*Bsl0Z*S&&PSH##}f4JtKfL
z5+hbag+ZjGq`P>2`l$Z^Qm7~I9UmWm4hpKoNZSPgIc?0gG2r6jqHX{{wo!R&H{L1n
z^TjkUzWLe4pj-I-wUDr|%)^I%XrdP8=EekRA2j9`6m$XIOM_-Z74wPNwBe7D<OONd
z8P09Gizhk#^{(T4fKX)pv5Hq`Kv>g1dh`gDR7Sd1^mx}30?}rR>F(VK97P=+a`tKN
z%QO_m#>S`bG4UN3kr<;sumGT&D<xQFRgX`59(SyVD8byLxtp3Yg7J26aA*(Jdnxna
z!2_OW?Ubm;MP7dXI{Edkdej8nG{*e_pd*sNdhT1-z$Sy9IgU$B^+g5zA$~2bya_EV
zEaD-)J^|gLMH&F6=j7z%n~e-HiWM)9p4IpeST1=7MV>l`7g$szfEX*^dbnvuNkv0K
z;xr<s+pMfiR8*iHNTPvat;KmnMMareSSAH&y%x9Cl%j}6KzO2ih~%>}NZ#ivi;Iu<
ztg5bV{PQOqMgYl=WiAW_YD&M9(?*B`iji&bbp?KKqY>)I5ajQV3+Ug4+yX{=M-_BL
zoAA!f3#U(_O&P`}ChO#c at wA;?U1kHhX2`)%{N45LT+xLYz7M!Zd);q7F;q7WK(bui
z;IlRexgWY^5ojH%t1B2R!7ko>5bFE<qhMr0{*UW&Uf=%Eit2}bSp|h;2*5x97X-ja
znitzno*mU^8rof(oX_jiofN$M;A2E&Bw93r3qKpL@^ZHeOwVLwVfp9Tz+}M>D1yWq
zr5aYZx92`u)cZ?)>eMMjks!`~lH1(SO-M at WjFa?(k?RQKH2iFCpEW`qBt3^p(63F_
z6XB5FvA%f(++n$fTg_2wu;KfmG^NumPJ%iXQ3YB|>gWifrlHxx^Hn+tG7qq?VPqs4
z+~E91o2Yu4(;#?Sv-WTIKY#jEg<)i-VPf*#+}c9-?B*T!92f#2$wge;X~C;`Yw#5$
z{&30oR<HkSE(&m58q&t`!40hzfg4CED~rM^k~~O#>+Kz!pU>AC#wnhP_dx%+gy0i|
z$Dcoclx%G|N5_ZymIn*wkuHN^LKV>60r at O*#2$<Ttx0MXSkrNHbHk(|HNvudS8k>P
zdC<L?NHmbc#g4Lqqoe9NrOrX1qoje4P~l(<e<=bZ#Eqa=_xATYJL0AO)bX_#LZ~=Q
zH4v<CZ_|SeYz}2NU4G6B%z%6G!UZaj<X{dAJo-fF+1Ng!WY*G(M5n#~@mr0bHG>%j
z)iFl`_$)3dTFl&BZ)R5pWoT1Zw at Cm=n`=KS|KFv~OT=tOzV$k&AC{e+9ZZ!3%)@Q1
zSEKb``EB0}4Gl>%UhM?4q(W`RgDxi|gv>{<f{4tSGiTs(tF)q`F*B`^QlQ1b=_C^h
z(#s at 1-iI7-7#c#>14U<^Ygk2*-SowyNs7PVY~@g|DxlzWZHD;#{4bDyvQAD;r(eO<
zkd#JtuUy<P%(}>N@%r*mQLwhQ_IVt%1F!doR)|waK6MM|kS}aHO&F37CvOVaAw4M#
z)B~2FO`b!I2G9 at nCP0WxMn(p%X+ki8Ej1$}33bN#;O{SU`+65V5|W?*(<wzT?*loe
z_%EEDui&h&u71<=I8vP+m<_O79^f4)eZd_b9v%c at x;65SLaIEGKFK>Oh+8?V()1oG
z0_!s2%YeoNvWSVE{v9 at uBpU7fMiT>nia3_5;~sTTdzIHE8FlqgJPMH(48JTpHT4^a
z_GfDowXz|WnbKr;tZ8AcL5D2o#N4+Dl%_>5XM!Q;z;msUHVa})#SQB}PV=69r|&Wf
zajo$B^=mipN)Fk17!4yUtAD*4R#%^wSI=*N(wm<S)tl)hNfYa|H#Vk2!W3B*6bk_!
zhvLR{w at G?JLPL?wqy|^Z#li6nDDtNf<252uQrZ3eeIA$1XUG+dj&HtzG;~_(zh7Kj
z%;U7AkzaSj20Dqyu<kOTIZ~{y+e`Y9qMXCMBO~&$v9VKgbJ7qECcnQnX1JR(- at g4p
zkm?Ag+th-%X9%1S6DVPv_U=?=CI}rc+C7uB%ktnLa}De95NiRqLY$)w6WL_!8`4EZ
zMgN8(D2M>DQ3S8W-1>C0n|I=4NHNqo1OOYQ09rV6T1X at XqqTW}tq9J{WIyo at bvBTf
z#~amm^aK5fDyn_``W1yV%nOr<NHjK!<czPse<DM)^Bd5q0)m3+Q>j!|R#qrJgF#RN
zn1H3rmZc5nUI%N&2swZpPDDh6k(rqcupSMlku~xG!-q05v8w56K_F$2vxeB3ucz=>
z3TC*%orap3mj69=g?qma{}X)%^?``m-w*{qX5?qHABbWQ=nfITukQs!vddq7RDsxj
z^j`sirpRhAKh;Pp9t;@T#0Dan1_!;T0G*cH)FcZz)7=0MJlCVHM~5T_C{jtk>q<zy
z>TfZJ%RR`>&fc5~RzOl7_m)uD$jbvOW+tX_L7sQ|VcsaF6;iw03^X-KA-4%|*4Nfz
z!Mf#O?@_eF%pUAu8NkY)yvJ?Q@@;d|p}g9K5yUrY at C-8sV!>~~%Bh<Si890skc2LX
zd=8n!&^LvJh1Yj>Ox=TT6}s(l?d|QMrD}5i=dZBOpUrnBUL_=~!jN39-u$T)|DXe;
zNCW&=d at c%Pv`7k>0atx`Azvu4(dg=}Tk>jZYRr)wqwVeYM%^}02h^Z at l$1c!`)wY_
z)p!q{x&X at Ix*seTOwG<BXnGKE>(S`T16E!>zIAfaE^&}?sHY7GJ6*Rm$>J9q^rKW;
zGAOLUS)zgSINpe{Y}q~mgKP%(hhEWCd*dhLi|pQGx)W>k`618(YX4$nV)6ijA=mN`
z7uq!`c>BR`M4^klKTy8a)zu*l>=dBPJwHF6Yt%?c=qrJe=&sWe$tV?*SG|1aN_R)c
z{qgZ})PI8Tq8E=*!xa~BF;A)(?E?T6+KKhtDp?!ya&qEXwel4m9eGD%Q*$9)(4&fP
zZ*QM|MgJ&50-Yp_<WMInupW;4p;ds$3oTcE6|kS%obPh;-qJwXGFlN+T2^)e2enKB
zI;?kUQWA%W+QfSNjVuvC6a=nI<#@llC_Xealo%sASne)Jrh*kWH#eUH&`E~bA;5{?
z&_9V at Re<8!#rx5Rq}fQB>pV!TgyW-wU`0k+8ylN}Vn>#1*RHLR-|DK at 0o94tdUbVm
zX|G;1`}OJsdhw*>?930}-5`&C{9-l=gPLD}BtDZ4KwU86;`cutRKO7g)77K}g5uFE
zuqH?1<?`jrbIS#zYLRQfARN&u$TLs`@Y76`l&@ayj-Ts|HgQ43gcgY)RE9{ZAs4?p
zC7(e{*<j5i<D~-Vai*rHB~fn at F!qo;B=6?0gG_etDsE$-r=#m2QV_ZiQX6?|L;^Y$
zFQc$oVY*I7`;MpyS$ISQ(wHwcW>SX^N4A#-k)}?Pk7`<&)5*}QkOyx^&&taBZ+&!}
z(SK>l($2vl;pNeO2Ru;MEKy&H7=3UDI0dxE at ii(c#KdzO^?>1S6RQ4~32~6pC-41p
z>Qx-F>;P7EAIZ^DjM@*)M5oD?CT8?Upw_`Te__>1T8}9Kh?}er=mKr|=xBe7266x)
z4(k0v*Tps53)zDDN1=6ph;W#yuAj!ujKEo2CEah~yJ#U+?$IOE0Bo~EiXz*>>y37?
z!#{BiDRg*rggT93169P#%$WYgzP0slE7_kGeQaM&{dx{#^{g*LXDEq!W at ZL?Ud0zL
z1V&>yFRJP2JOYErKtn^*+1r~0AeoJutrDifoQUxF at bK_FPEm33V-1b4rr9%F^d|@i
zT%G`DRaI53HnUfJ%gVX}QGOl|uMUD4RTPg^>>rfmlKoy<dNU{}$j8U$o|~HxXgSmb
z7M&D<D0A~pR`WT}$;sJYWY2)M1p%8FfS%R(X><Y9#4cXI{5wb?2`Q<`K0S}xAR4J}
ziHTld8zw+oze!A_8y(+DgDFFm%g^KEdjtE=w1lHG3aHF_dw2Jgpit5|)PNUs6WZT_
z8g1e{1v1yjF~Nn|eDN`3XL at GlU&D3{4H8W)Ep(u<B}SB at va*tcgJTT7Pdr at iP6kha
z>Xs7_5uF0oL5=2tlHbI~(~OQYNw~TSPE1Uo&Xf%e(kg0dFeE3z?43V<{#Sxb at KY0$
zI!xFHQWkc0Dby7lK1oLAx4v!{79Os at z4Qkaq`G$x2Q=8IK64`K%m73oB`JAkG?qgM
zH7_YYSP?`ANK-O0>OS1yUTxuSLp at rq_)55W*AFGe$DdMCQbN`Hz$|kBER}8#SfFxq
zkeR<Q&Se=nSkzn|ZY?)2ukt_bJy8|n;9#OrDlXZSyu53^zP_kKsEW!}P+?Ea%zS~5
z^gH4t&3pJl8;yMY{cC|f-gutF7ZVph<wwk32jc#Z21PZ8`Ri5&E{tD5fRwE4MR>CP
z7B>U9b9BdOxKq;8t3g1ZN=T^dI0XA7nh*#<G+6Ej+v7F9L>xRkHTG at 7-aYB+=}U!X
zE-rV`#zz=~NszCVIc*|?rkQPA;g>}ZS5Kib=_EI9Pzed$%*e<9aiKCWIEV}@>JkIC
zYZqv942x~UXvl(WxMfqeg|;=Wg3O(O*;9|mUIuxN#^B$Y?N!idjp#rVn7pe1g6`WQ
zzd44ff*Z3;+is!Fro6nosN?I1^CWJ6fB!0ZgeDcWgrwv+K&+Iu_RW*0PN74>AIJs2
zVB1|(!Eu2jLfeSZF?)E()t#MsKy^-#U>Z-JtdUd4zkToLhZ?ScL`02aCEniPHhBsI
zjf$~>%-qEb3i6zpF+p{V(aA;B7};f>u5@?8kAsVALXam9by@*=G5+iIDRdeuIp*bi
zkP_?UUz+w7we#;$q80>iJUuyGw=FM&biiJ at aV8DIX1v-PHRA!WJBfo<%~AdZ(?2qB
z?-0o-D!Ku1(+>;+laFddCO!-`sC;v#y{~UpP-un#ZGVLEc<bdg4%3eA6fO0hK7AU-
zqciB?r|n at tM{mFgB1jTOM#fjEsZ1tq(f&_zOe#l5DL|Tpnyy{O2VO(nX0BcHf=C29
zu4-sFpJUQ0_WU`=n>TMX&MBRcm6b*9QOaC*;!;x11DGHfK#eT4w6w~r4OhgegD9Dp
zE*xwxR|EYi>gfeE1~Z_}^Qg%y0Lw0(QU)k7f3y!5J|u2nKz$h(m(ylMNNB5Dd39?`
zHA2R0<`^OdReeUEI)``jo{kPVd`d}6>k<P41M0GI;=~E`*}<A=&|f!2+(^);AT{6t
zV8T)9`O|04RMynI#a;=&m6DcLg_@|3{W?)oQ*)1&5EW2%UT<U|xOsCZd5}lqZA*(>
zO<f%|8yg;Q#4}6FxBmXPfYIYvEWUz*0)*8)$}6knZ1Hb=&ezt~qOMVdgoIa!iBXM+
zc#prwa4%jJ6{1OU;U|fSiPh?hDS3FVvX`!&)YR16#q%0RTlrA!N_2V~GSv%YY4po|
zv)h`v=MC#Vj0={Eure^5t2o|gf#g+ER`z64P8utkjTV=dmX4Xf1k64;)fmhVLI(kG
zx_V|+OUq at Lm?}uz at yW?*`n7#VYB7C%Ds)>Ax?y1?xw*N&T3d0^E?Y?J at tK(<wdL2S
zqy^j=7z;*rc5hH6YTq>~=~GDFDOp&GWMhj)-KcRkv7;1_x2T7=o0}VlNefBdsCL?q
zAGi=XfZtLu%AoEce%C;EppqB2^|cy73r+w`^cUDrpcWadUv*2JIZaw4&H!zfi at RB3
zR)#9U(yndw=~biM%$Sw8-rna%^_jZ=<e08Cv1C$ma}%Ln!F)C&-?!}x0%!yU$!sc)
z_~gIxzYYw%VmVU!g!PWy3OYPm`uF$%z5RfQ%!15LJg~I1vdVVbv-_ at B5sg&@?LmiH
zyTH?l{5`Zs?@U8FJ3HU#z1%ya=eBkRBACPU_s!DvCYIHW4dwK;XOb3XW|g42{&mI%
zYl(^rkGL$dqQb%0_9++)ypE0ze!Hpj=x!Dk%uxwn2qDx)A|L>t?9^gBs%8Q537OTw
zZ_cY at l>pwT3^&2mt2Hy;`+*S=q-an7-{YN%j{bfh<K{2=tr5JIo*VVFA+jjnmIn)H
zApESYt(_<R at 2tYmrXKC)P<QwBq0)$mn*sX)yg+gm4N<wey7s<&%q{Wu&6|@5#{f4_
zYjuDk7!=fW0`fZBm3ZTh#9LG%6sFu8k6cVj>Kshle~XcA&-3mz%?#vO at H?+k0>gRQ
z&;G6iP6U|cK at FQNEG){al{M`YI?D$EY=8?Yb>5yidE&nvtY5vtadvgh5gwAS{hz%6
zzPGQ at 1~&!<2Yc at BI-_*r<RpN3e+{{WpaylU9~>kD$aV7`FilKOMjMX%D?CKeQ42CM
zGE`6(m2#}8 at Q8V``r|`ddb;iAoDzDx`T&ahC_&8F%F0K&hhLro?z!`&=KTCgSn>B*
zu&TN`=*yS0V2b3zxxC1pKkn%|+x-ai66)Rtxc~zYo0 at t76=2PMMiA|`F7wRHjEa*p
zNN|luRFpz2_6i#toA$}%n3x#N==1&E-ETqAP5^eJbOSqrG62Slnu7!FsBiD>oe`9q
z5xJ`{?dR)T=}W{A7#Vp9F3!g33dH}tY#$ww0As2h$w5bVrnR*ds1wDykgzc7l7f$q
z9~Kd at 4p{OQl<h|tbShR>T-3lGvvPA`Vc}bP`Xz9ND>Y6S42F+?<P=06DzFdk?T_P!
z6RV^7+RvUbI$=lrCngL~7x~{`1;V&Z@#y8ke(kT?(TB(eef^61w^313pQNLs%gWBK
z1Z9e9YlTHdZje)B*Vot4u*}Up`zqp2y1cxMx?Z5ZoaY2}?g`7#5*is9jR7M{%gTy@
zV1Q}hd_HzhF6wT`!y)y8gd at n^v9U@|q-8#R`t;P?{5=pHDT~^7R77)hobz%u<JCw7
z6@`EZ1 at IRCsicb9Z2gWFM%~@#=I1M`t51!LjI6G&R~6b$cJFV7!Jpq0a=rrQwmwxk
zneytz@!45#M7xk>-*vvx0==PFK(r2b2EH{+=mpnxR?W8%R4L07IS3U1TczdX5ZIxE
z0C at _%A^_6>Cm&K$7<Ug23}=5wJ$G=J61*LP$Q>LTv4DU8N_BK#1Mqg0{5rdVfkBql
zARjvA4(wlBalFMY7s=<3&01*&CZurRB+zvpKZ778iP+(zaB^Z|WoxTH_BqJ|x<(_e
zn|B?~qy2-AA3sK2L#CTUbN0SkQ&Q8?(dmPl=YKwi12Q)Ygw}VRlKeJ$@+KoLG}x%Q
zJj5Co7Z>G~^J7a(m-Y1YcK7$I>g$u+GAL9rs;a7JG9EuBM19lYS^piS=K#fG?P=h&
z at NBeqXy`+GyCSNkQ|P=#jf%1w)Ly);TS{{393IH)kf6{|Z=Dh+eaJW at P+<(Sot8D7
z_LFbs<>ym_$N>OBkP0D*XJ}}MhO09{2BeYcIlQzK5R*WSA6<t^Ujz?22+-1=MzX|W
zGMvrA(sJt4tI3H8B>7<spGP|9IAu*38ylmtU+5R%p}u8i27tBuSI=T=%Q(`?;k8(B
zE*6#PrKP90BcA*E^`?&mbzw=#XmAau-ITOQmEo<pq@;HbBT!#DkWSARds#t<+g>s{
zdGh2_ef>&IN{gODwDXN?*J^e69UUEuBxE3I;1|S|m4mU0J!)1~te=8|i)wm++2<A(
z>@G9_wA5$mSMTB}r>dx`+FsgFh!*l47|`Uud)Lmfx4!;dwKx8(V2emBSRr)oja<az
zSm-R?O`l)C?tzvNUeBEa5T~W3eJ(S2=FFKZ7rmZ4I+ih*X${MtF|)Bj+8VI~FfUB$
zJwWQoI{(S**Iiz`$bQWo-QC+u&C7c&H8oXQUcL?<seNPw at 6Me&04xSTp^yp&WD%RM
zB&hG}>W&C6*KN$Up<YD0tK<I~!jY4svh3&1o$DVQ%xNnMcxGWCe*gZ7ix)2j!pt5X
zx=l?@je+n3yOC8;FuU&wXom{U$owgz6c#4Gb?X)*8=HyXA5M?p;NWVyOQy!gwk3Aw
zPM?0dKGkUXT85p8g{2xY4 at nvThVJg}Up+n7sHmvG_vj~wi%KXf6M=xNXH@~3ku)@<
zLB+;F85I^5B9Vb|r9D=B4=;57Y^uupswy!fV`C~Bnp01nJdsvVkb-Z&!>7Q5um;XR
zZFxX9nir5HCMKQ%S2UCw4$c9*9IE!3^eD29ujMJpyAyyR#sPvcjKPhHii)p3f4-WK
zkT3?>jUwFVe13jD!jHEXh&c4CPUl%N;QW2m|Iyd?zJmh~0RQvl0bUIaji(?=hHhTa
zYDpsT^72BbTOf(xFGWqwi`?AY*49Q%pP9 at f!`wvyX8Uu^hyYDJVH^yAoKbs?fn2 at B
z$4sDT-q at 7yon=snu7L}poeYVOQdFHJYNAve7rL3i at 11QE-qDWHP<aWFiu#__Q+P<o
z%j1C{eB-#-i&T<q*|$ev%zy)&7gCc!6zYfRuCCQn81AeLGcq$zfYDJxg}l#Sx>Ox4
z?5e1$>ILLtfZ$!&mG3qy>mHs$YJ(9MDqIDM0O?&BnG2w)yn&s!ee<7Udiwe%A+v|h
z=FiX1Lxgw&@d^nEp at PnZ#l=-Hq3_<kbMtO?5Qprzga0Kw+;?-+LA%`T2B_54)zz_?
zDxiV=?*5ZqjuLMnUubUMz6_8oWnsYr)A9~vSWC39)-qrI3NAn=%5J<{LO$QW-$r@}
z{JN#*wdku?uTlb@*xK5HbRC;P4NnRMZb78yX$fX50QsW|Ai#R#Lq+z|GBTBU7JaB|
zh;{2<<ic%i+At{}NDCbouYV*Ld<Q8xxw|&`?Afz1zSU at cTWU9V_l^n=kMe32bTUTU
zT0u_ka)xf%MGz}+Yh<WvXJ#gQGwTxl$fw-DY5DorFI>2QSfR*%HZUZ_>&utxB&4Kv
zcSe6p;DO}Gm)5(_%)(-OeE4E&dYTF{1xP|1h7|<v1Ao%mWo?QpSAJdpmGbSk@>jq`
zWJx+Z9|MKyAMUQ98pvlc{cN<tG$FFKYZH>Fa3TaKJovA at joG3E6$p65n^7hsxQ|n>
zLh_ at t!I%orAFqpw1P%`mIbI$*qxOF-VVvzfJ=22PA-$ju0pU^p0A2tPA)yYs4&(zW
z|AhM1KyC}`I at oW_Jj&1~rl1G_4G8e*?CktJIdOk?_nwqga;S&k%bxD;F|Y?<Hzdu?
zQEw>9;Krw>ru9*$xk1tJ^Ydrj_tr(11Nbm9HJyYBK{O3iO9lK3aARX*gF3k=C&}pt
z-?!@oVFX~BkeEmXRs{kTT;x573<+uiOH0eKsVUUj3ALTAt2 at ib$M<ht#v$n(@_NW5
z6&8|6N=i1gwMo5vDFOzYDoFZUe7t8zyi|2Z2LV!GgTYbn=qbQnkqFq)qSk5<x8>FP
zpEb<Pn8325YO8PGo<^1WdwY9P6)gyc@%i}+U_29brPD6^oQDgb^P*rW4GayH9zQ<m
z?(UB2DMR+59D4A91c-xK1{^Awi)4kEIusQUjt at iG?$ZZ at TRaU?pG45<HtNG_YU-^N
z{~&wVCFI_tN4RX-g|99Vvr)pM1Ew`JHr}_l=XP*(<b<Jw3ACGhym0saNJj^9?*^mg
z?)(mOS0W-K(Uc(?hJZJACwQkHCiQf~w at F56!`EZ at IVT9eM~@!OjaG=FG)9dAnwwG8
zYx4pPPoKc3DF4a&0Q<<)-oC!-{{Aqn20 at g`g at sw!x%DdQ>fTFBmZ%9RnWU<^y11Sm
zg@}j<@GI(4HTdk(F8~j8QWQxU6B85X<GooV{6mIY$`TSz>+9>Io`MJ^f!oHxm0@!j
zM4QE(ocPf87Hg$F0Ft%ME(ODZTr+cSwjr2&xEX|1KJVYZfq1-rTjNzvx_WhOtrvFn
zRw`H|02dVAV88e9y!FHC!3&^XJg8JXfG7RvIL8OgAlXFzmxuhvyOa3HwD<J%AbJCL
z(E1~J!Ef<|ApYu{Tvp+W4SJXY<YPeY|N8y=Ble2w1BkUdZT;axT54)%ak6rAr)H{_
zmY2uJ#&B3zSorYjaz1{<(=M_*Gusv;1>6V`6a*R1qMm_+bNlw~v4w?;ps06ucTtlj
z06Zb7)|3dqgTuYqnB9#wQQN~^TOMBCN|*wmBBU?1wY9rVh{{#~eozyvj})S4eeUy@
zFIAnLgov)Po}K`B+r<moqX?IK2U%NFTT2C^AJjY}3ybHwcW2QJg6M|}qkdF1J_P0^
zCingzF#yg3-KeG}DR|T8<5g#*6CZyd;XDbd0ridMH0z}07Z6CDPAdAa{tV;=1U0G?
z4(Sfle*W}HbUgp~_?SfWr4Y!y{)iImg}FHdJ75)G*VIUWvWm;iy?S)8;|+*-cyz>R
zKO>jFR%Q4&f_AxLLFh%MJ!T8xF!<d-Y1%7(wa5mb`-Myfdx;MEe);m{dv)h2Am@;f
zkV1!fbb_RiYzZ!6d$4VV^9v;2_{s_%=m2LImnu|S>+Iz*0lTCERTULVDykEJUP!m<
z>r?rhuYQMHV#*)_5=KK)bGz~5kS7dwY)XocmeE(ZV;DRhK0f47S5{Vr*x10pKlKeQ
zONfs at 1!Oife+Ct|&B%ZlI}2uid}0Evv!F3>`u~`E?|82J?|=MNNg6~$MF?e<k)((Q
zAxaq`TZ(L%*$N>cMK)R4D|>|O2$7Y&GqTFc{GG$~{`|iFxNg_grPu5Ed_2y%pZonh
z&N&ZD%ik4Kd*b8c`=vPQ8X82wgkQaS1z3k!R- at -dxw(9>K*KqV4PMlI<kZx4XAISg
zo>6pmcFxYtUDwb!h4~^(a!hvfUbxT?Vg<s|bwEs{0tE&9B(68`&PPCK6HCjNnVCGW
zM&LeQEYH}ufW`Uw`C-bSroR5Bni?(N*f{{%Cr{jQj91*$Yid5JDN4h{LqLk&zCAQN
zI%+!gy{PC4##|a28Ulflkr6c9(_3APV%{#Xv1h*3*K6wO)iMP?WAGI>Szod63kX2y
z1R)0mZj?4VJDWhrP&@$f4teW3pN at _Wd_cxx%ubv^&cwvz?&DKk;c*yTP?`c_4lEu7
zfRT%Vx_Wvqii<Y{<0$zB1j>P-O?B$3o&j#9q=ffM#>B-9!K>HQ)LgT-7o4A;M^<$^
z_`+Ub`Y%F4s0$ar-U8a!9yfl)H$Od{9AqOf;Pmu#6n&yTwDo6<l$Deoj(v}`&zY_j
z249tslzci;dldKpo;m<D<9LY_<y-<`Wo1Q0{qtWUa{&Q?><uIF(}(bC%FT`S<sqM|
z_mq|KvTiM~OXNu%#2yFAxqP12GqZW0Z(x=<eJJ#*Bkk9(Uy+9b<(80;_z)TSoLj4^
zw7oq*z)QSmIv4EK+S(fL>ozq#4=Uj4!o+>>7+mPRU3fKnZf><Z1*2l&66d4yCu(bJ
zfnS!Md90(YZR#0U|Lt4<&`|jqZ2^A%PmQyf at 0?v<>cgxV$S}N{b?x8Zxr)PgORK70
z0`euo?%*I9%<cj4FNGk|c;)W_JXiYkbu<xl(e6D5X)C6C3Mr_m_xp<DJ(hSUH4BUQ
zh|nsa(<y#_DopSA__)cjRStiBhcJBwzJV>9-oo1YGDNNCz{SeDx<Wu_zO=p#QcbbI
zoRy7 at twU7k0Hau#o41IF2#QFSvu9lZnyY;14mvtIV(f-^rkAfa3(5d<mV-k>zHi<*
zLA-yWrw8hC4}cg68JRZo>!m1ALL8cfe=%wX-6n>x)X~vlGNc|Y=2L~2V1mI*{hWFX
zfX%|nI#5*uE{33W<k+!ja8aOj-5syNOme4#4V&xcft+7NECik^=6&+|&6~T44;&|4
zy?Pa*F(&q56ZG<ftSu~fF)09ORZ(5d$j-iBgh)X_VcO*>SaosvqXO(JS^74I7s!T~
z9p^V^IeGFVkV(W4W#vp;<%2y%)`!70x(6h}+@@7Z9B4`%+s6wE3(fS)kS_rMghy+y
z6bM*NpW^1GKq at FMj#Cs)AKHtT%i`V3g at u@p^!D~XTRr6n5nEbLu3{uRHZG2lkMCGY
zN=o<3io?uO!z+mfqz*IEou{W(QS1SvZELUUp5y!iK{q1F0pFanvw4$KQZyew-VG56
zU=EYCRaI5?Ie)qfEcUmjsfz)X;ZQGrjf4oIZ~&)zl|Q?}_qH}-0^WX$GOxeCbbS*3
zL9DkLkIyY25JLU^Yi;eIWo1)ya~bS%zH2?~%FC3LbD*hGE%&SPXXoZdfBJMtAys97
zSRQufo}wZ|>3!rBD@!v60Vj>(ZosPoNniM>e~_-Wt}b6>sNp6rFld64!}`jX?JFJd
z=IP(Re`{!I?E-0(ost`tn3#yu-Bk+AabidJ&|sw}7|zbwKGll?#|$mnwBYx89~Rpz
zut|ZI#_Ok<n20f^UwOvGbT|Cl8N(#~L3^-aQ!_KX40~N!MFns4^%W=DzkmPhla~(x
z><vp%U*zLMhg|u_`lr6W9zy5PTVYRl4HcEwp2s+&l9LY{@*n~mLt+>j7KZ2#76Rch
zqh~?RMUHJ`qd3f`!P4OAQ*I??<sAh0RW5FBy!2O7Tbmdj5(BHS8<}fR(217^PV59G
z|K7dt4k;-q4jMP=E%HcGNE-wv!6>p(SX6WpT>Z`0M+AbQp&^GyZ&lT3c#Msinc2nJ
zNFtyPACi(hL7Hf_B|orpE01UK5P1O{+(_p1_NWR7r{914a0Xxmdj&~z5Qs4tEP}2f
z(>njVyPlq&`H%eIuc+ji?0`20)ox;HI{3C- at XD2b*!$GySKf-B21m=+c`kycwtv60
zy86NXk`Rx;kP!Hs7q*!B13JOQ%j at dl;URJE<NI9jV7dGE2RgI$S<amsIrJ$_CI1D8
z6{A071Pm+s`kp|x%{-q&{Aw-U>`4G*VwMUmjTCk>S)SEI3)QN6{Uz$AC!YQKOY(yT
zmp)@4l<}*8KZ1qQQ&UR|Z#};xAi&;g_wfGxqq!&M#(%`&P+{0oAmK=@K-%L7!knB@
zffh9dRaLLouMZd*od&uMydG8*v!|n?(;MyG-6tVfk&_VFxp5v7aY;;M^Yi!b;kjxG
zGU~|Dqmd~oq~KJbBbE$K4-XEOQLZdF$+6u88b)^VlGD|V<z28vKmsfLSY^4a|2%qT
zYx|dfhG at -dXb{uz{1*GTxw&B-ToV#laq=T at mU0LXXNK4nRaI$#NnuBlh?rP3NJQ95
z44DH%L(a&-8yci^FLF(fja`GKmywZq1ELRvhstN0qr3Kyi2%J0c#@iy20~R(1{@6*
z6`?{(N(yX&>zfu9baeOhWN1~B6$@(y3XQBROm^?U(_4h+dn|aTFp$w&Lp42P<Fxa)
zIKj3bKYm<OS0}8k&2;|!`8Pm^sQHXAg+xnBi|5n$a3dqFI&B=R&VZ~&(@E*+Z$kF7
zv#|}fwPHTCtZXN6wWV)vK_MYp5Q70s+?%s<|FwZ23E?vyM4RB`{54v}|EnD0a(?!E
zetv%RcfQTm{P&$eunoz-&PD+w0Y-KU#)Ahi;KUcQZtWdmAvy=CF1XK_6pApt!{pfR
z-MhP)j-EJi0=&fu_?U{SYW`^+kh=a;o;;f~{a>6FKilwDJmki9A`qq!u*179<U~g9
zAmF8bV7Wa!jar6=q~Mqb=!8z%*w|dzWLg7c8EkuI*)Z2F*l4!&YpLCiTTHB4DD%Cu
zv?1UyN-i54{&r5hv`>WLIWTtwD;=FZKvp at b>%pM5vKzfg5Yqv;?#alME$nWHxatFL
z2e$Rxed|WSgPlR&jkL6)bIe}UUW<L3MF)CzcL0a#pb>2t-&^2lK<`bNXZO0^I8IHy
zwUiKcUXoH*oy$3S;R!i0DV59fU3(t(mp(s!;mI2r&Qd?|)A1=OHw_KXVNb at oGT3}=
zt%ORUC24!As_^5-XW&`#8ZHt7gM&9xu0Mp=rlzJoe&WR3vNCt#hy45?>WcA`-|@P0
z at bdi1%fL%<awd3ez8k<5j+*}Byr;aTh6MaY!i=gAya4oQVp0+w6N2A>5{!mjIePRc
z?8h}--Elz=ac5%Ef?x9~4V6{$^700QYI|GT+D!K at gzY!xKg{3N)wN^I@$=O04;D<z
zC#b36v&X{SK3%x&PQuH}yL82<JMTHXLh6PkQ}epT8~`p&Kzi?k0DsLLJ<jA5pE5I1
ziX+;yvhIbQNzcqI-=6N`4)~BQRc2shR8{GDEMH`}!I4w7zy&POG>wOgtAAoHU9((6
zTl*zoj_HEc_Yc>*OKRV9wO9)lTTD{o`l*Jbsh4;|>@p%0K}55#u;^|iJ#n50!jsD2
z3IsnyA#9Yju=O6l%uo;~gl-A$6Q7(c0*j4g?*9GV339dTufbDZzkZz;V+bT0uS$YQ
zj7KrgpYJ$u;J}BpG{;8a)=8|uo|v4Rv2JN?jRX-31RK?7JMqNr+SG)E{J72UK_pin
zR*+yFNJqM+%4);6Z?6TKR31$Gd~*g;%>cv at g4C$M5mB$KtSpmToZufSgQgJa`Q1Wq
z-?_7Sallwl?-gz$a6@>AqPuTU`Xnaq<1^~M1{^3=wXkYow*Xz(*SMX}3E;@x9G|T=
zlO~qDiiZ(AItJ9~Mn*<d)G3x{zBYz=%7Hz;q!)f^V$t#a`>gT7`<F?b$2L0(XYyE5
z7aaTPY>p90OOuP-y!Pd##2&Anc<IA;qN)OqJ4#1$Gi7v7)}Q;RcsP1Ak&nCnePDOT
zQGb at 5F9ca$9DT#&E<SVC*1>UD!0dtI(VH~2zt(N0BtwIGtckU)bi&QT&4?h7>k9&x
zbxTe at i<y{Q<G%L-(eQ6#LsVZN*bIGE{cPrwnQ8W}_-bJUh`Qf>C3J;WGos?+B(OLA
zQ&V2Y)Sm6os`fUV%_KQQ!E0e~*Y5lj4lk={j+Mk3bG%%Q&S9DU<DHN;jXgSSpc=lu
zW2QK9!UONN6||*<-3F5rzIM$?B at fhf)9EfN;9|p3!$79{VZBTRJm!}!Ubx`*SCEF^
z9B=dm_}e2OP^dJ;NIGH_BcC*4=?l*^2=DP{K_8SJXawKj;8*kW=Knf-D=K2zt`#?z
zc#TjMb4w?+q9`G!qI%WmxJiSx6X(v|47q4HOoj~vf3Er5E%0LF<h)C}#>2rOHJuDX
zdAF>rEM}d+yzz#5o5kNhH5#TyrDA1JYh#JkvuBqW7#V>P?GhFi1_larj&`i~vp|9Q
z_ at 9;;!Ts<y3?LniFp-M4PuSUA0k)6C3 at CIy<FlFG&(tb~T|@Y2-wUmbyNj$dZ;?H@
z(GrS*Z^w=ud6rXDSO!Pt7XP5`+c&>7u~;zxPp7M^%i`;*c;E7WrwKoQf&>@{q7;yv
z(rPZ90`x~ll&bdm6_9c{Ik{`<>U2g%MpQKGwhI%jAIuo*{pC6oGiwxUoqLMym8bXb
zCm42SU7u`A8QFMi*p+j$&~i!xurSy0{$_s05&D3D01S?oFJI2?EQ8oKyr`K#lWaqG
zMO9U`qpM5baL>EWB5TfDqz at z{UR7B&y6!(VY}Q>>^|9 at mck`0Z=g;dr>nZ`gKCG;)
zAWgrI9kT^ZZ2hK^L$#nL^M_>L_4*9Kf`r<yU%!BOMeEU8nIBio*7I-Uf9L0S#4BoO
z;;Lk(@h&G<*WIYwb#&Yor at HE<1Z7se=Z%x(Q>e`QtO<XX<QFW%w^cBJ#V%c4T`cKp
zZf+iCFz9dGIH_7--qhLfLylQIdQ at _vMA80__ww>HSaw;XO~a<B-P`9g at lB2y_mkvJ
z>9~}YyNJ1;VYkC{cRu5iR*~&eYWLz9LHxu;PDx40yn=#S-s`ZkCRSEDh9NtHMSK^X
zfxUgb<^1j2HzgXfZ5|LQJ;n$;rfCfHm*x(Lmk^Zdc2fh0{Bhq0aWk8P__dc8FNa(b
z1iX9qK4jd{)lKLoQInH%`SZtg_}4E`86w5k&$6=%<Nxrz5j+*g59Zm?*;yNQGg1AX
zcwO+110nMUw{C^Oa#?H3CWy<e9E!J|rlz8*z1cN7I?Bbv;|xM!O%1;N9j6@#L*7FI
z&8v51;OnH_?<*>j_)EA#xB$|oedC_^|7rp7+7qCBEF2u7IBD<iU(i*jt$O376cmo7
zDCfR(c6L at V0FK3CB{klWp%s1N=4;yI#6*edc-hY^>iot$=g*tW^a^_Y8Uv~n<_qBv
zrxgARg$Q-C?)|xuzxTz&;@av%z~59Vm|Wr)*53sL?4c(DaRU<e1t1SfQy?_1uC6RB
zEN>cnV0*nDvCiuJ5m_k;*f*v}7X1APKM61cdIp9Y;g1DCy1<IgE-qF9gTWK;`Khkl
zrqeEy>{qX<Z*=>=l1thQ4G&LU=)5j*UQto8rv2i_j~~-NeQI7lvHX{(P<u|UUOG0B
z`Lfeh<3lmjPVfl8B`^nPZEX!){`#$3djY3Vn*sQ=*$B=4vK1|O at 5{eWe)DHjQ^uDA
zE7wUqC1T$i%glh$IR4Uk{P?TuSD9ZxZB$ZbDEb at CN^F{P=GC~)pMQ~%kYIP`A&46Q
z5PX(Q!xP7kKbh)(@V9{2X)Rw}R$AIO^z^+1IpCZkAirOEdXCr<jTc|HX}r%_ld7g6
z9~K*X2E;1wzBc7tBd at j;<+^qqaAjZWzTh&eJ?sDSH^<{52C*=h|A-zj{QakYN9&as
zJ^Mf>z_$Y(YwGB2X*mK;@4u&ki;vF at Tnx5pq)|gvwMS{`Me9|o*v4BP^Ospi9^(GP
zdq+hv{43^SX5M2p-JP&-!T`t+fKZCpbeMpZd{KO8I7hY;Dh5nOHG~P&eETLMDoXt6
z)2HFb5NN7C+b-P$Oq%=$M9-A at C>Z3{HXLc at b%F1OpnhTcM7CF_l39>BjY-L48CsD7
zA*9v73xHekntC-yGttn{fE26S)g^v&yj5UC+`9VX#oFIP5_j%Yh6~!JZ<o+r2O|Zl
zUjc^6Z8QJG`{Esu7V&h!xu&ke at RMOWB7wJK<As#AGoRgpcyul<dq0H;eaUt)`O1$)
z2v@!l5eLD|G2O9UT_=W^ZTBmA<z7<jpHpeg!Yd^$LNb}_w?P|5+l%P^;Nj)<@0yP(
z`dQv7)}m*T@;B3FRs8AG4ACw%*h~1QrR7CEOS3l(`U!VpYy=`&)XExIYVp{X&T%cW
z>GCqjSJUZCz#kKH^OhQg9KF_kz!*UXKTl0fy+SSH;2_kZm)$*j&z@@<6APCwpZNLn
zCt;)L>SSkjQn$4N8-s5AF3C4cum}3>saW|4(1Tguhgw<?9@!j;J32aQc2z0RC3gUm
zIj5B66dq1HzpyYe!<<0(d|p)<#FGAOe^%Cod at n^Yu{{uduPsgY42+L^z}l9Tm+yin
zxntVjTvH=XL_~xK(^H+<viv6*65Woe+ffh`6T^zTHAj;BQa*U_z}?3uVE^}+y<!la
zM?66E%|yr86<wGuy0Bc-(r{|wR=h*K3QyXbkP!AI*}D3A|2>wCb#?DN?ewN|&3-jd
z?Y86K;4rbUc)U4;mtE`xREHQ6&-)~;g#*t%ZI`BXL#YoXM8srg^8!=U06STV;|5EP
zjEt;0f6bPT%CIhR>0MLK?Fs64 at GZHXx9a}<`EyTRo+;N^chXJEzyK4m`Q)lp$r(1b
zfqu+97cL~X&jiIs-nx4?fVuE3 at gRg#(Uj}o at TL#PO+_UoJ-a=(W}c3JUXSm)+Dya<
z%Ok9$6kJsr3CnpcPAcpIG1qu=EZ)0DW!D=-+XrG}woTa>EfSkEfcY&droDUj&KbxG
zo&@vo$Zmuu_4d9C3sapG5LHl6;OKlmu at j=gh=Zi4sOXcXDA(?*N1Q^<5;v%LbxZ?H
z1K&>A2geA}0!?LsxF9V}ff*xsr}n+#H-GX at 5*;8V1(q$Fq*~qEOAGGpx_tr3%b1wc
z$*P5v at 0jk`B=<2(%ieqNzytm)Km7q!h4_l2wzhVPe$V at J%29VhE}9y;wm3?!*uCl9
zl-cCGRiD!vykICptq4(QZ6l!ba|{(<g1RmIm*{PH at i%)8CaQIf0w>YY(E&~(7~6MB
zFp)_;MpXe+K2@}5%UXU+OiWvVI$$W!$YHV*o&2}tIi=pV{4P~ozj0qL at BOQ2cxOz#
zX=!QY-%A8Ppkkbt-26oEOH}}|FkchkLRQPgcDye3!v_MqylLG21m~A6ifjo%?g<V9
z``6NzZ|Rw2XJz^AFNx(8Gcz;udJ!{0x^XAtx<9RtL~QdVh1eS<5JmFyg!y$OB_xKW
zgh6y=zY at th`dDGck7zBeMr6h44L@&CaB%JJ0f?P{4APtP*#%>E`j3tx**x2Y*-#*2
zw{PG6o~C{ktIAHDA`}-F*B1nS{Pam?T7$vjqObNhf8xwH{1<NRZas76%%5K-3sN7P
z7QLf*4<Bqbz7(TR4Iq?d at Kc6kXMmB0fq?<uUE~#1Y-nIWRpBnzVkchNxnz6f$dQ`L
zmS)=<6C~$UL*z;_QvVhEwiNv%&a(PWe4$UQ`JxP?zxg{(Lq^rjEU=Qc*Zr+ at XXEvX
zLqkKk^xMv!RZPD&e$V}$v^0HVOGr>q0=J2%np(IhcebCOTvBWH^zI}+y%vM{H$t4T
zBduw1=gyu*2V$0webJI**j3in_MUWFP6y(1wrF1ft&CI#EIX{1=`LlEMi>YmJYZEf
z)-Nn>%emdfE4+YsUCXYdq*PW~N&utyUc)ONG4)~P0Lh(v^6^`*JUqS>+pi)*W7Z5-
z_xMGF<9NS;FQb_0DOQFkrq@#**^eJR>i6U<F0JySj!aAQ`I)Js1r#hjJKH{L)mo`V
zj~Bpf<b*P?ysaawYt`N~O_ at 4(|1|-P213eyaxPQ${{1t#ZsL_`GV}%jz1<&~No!b1
z`B!5)#bT~MjeYJDlOw2Way{YBT^o}(U#i4nW#}(oeh at y#5C5EYIh|d1>tFlCk{GYR
z^;^GQv`n*;rnlvr{bJ&I#;prVAz%;DZ3&4RN=j51*A~+Ng}yoy at 24LwBPC!Ieg6FU
zl|UB{507r9GuiiV?OM9!yvQ4HJ#t6?_ekO6+z*KWf%Ggmn4+qNz6J4+kdRz6FgUBC
zqT)y9=1wcUa3Uo(Dk|8u6*!0K^!15S^>v}-(b5MXD!>2u at vf^X=7uplJ3H{dDAhvC
zSM~jONF&S})jzK%t!&oS*1m~|;9QFHWul8OKJdNQO|<J%@NN#?fLr!jF&>mLJBDK(
zH*0wC>TJHJy;9<l^^W+7IN+S8a<=`ZT!bQ#Q<Y$D&I}!fe`hbn2M at kfRf&cti>=T$
zUp%poR~39N at T#K_R^(wxI6!}6jJlSlW at L8`pjS#$8)IuK(6O%;=OC1t(lI5J$kgwc
z%g$bHNj6N_*m$$T^FUFtw4)=ht-l1ohH8qxxxBNpKx!8yP(3Z6hj?8po;VwH=lR|Y
z6<LBHoW`w{-Qpvyx9c5Kw5PZCOD``{7jg)}rd_Tv+RABuZtuknHJW2IWYk4<KmDLQ
zr&Kd_QbyUWg|bSisml3gr-+D1%P{c#!xR)0So<s at LUicRAuMPUvR^p`z6;xhRdSIG
z6coJd;o;$Uoc*3dHPM^}{;Jn2iP$nchqszh1AVgR>({SO+EZyny{KHdwW>5qopu-;
zp=XHF<^+i}*`8JnI#pUp2`{@JT3_mO1CHDrDfkJHzM`Vy6!23#Fur~Jz}vTP32T>?
z%^Jf6o01eb^PhE9R=VE&0FQclZL!N3kAxEw6F(#+C7E9mnGby7>^v|q at Eiy2(d?tZ
zAL(<L_-#p$#cv~}zNpci_f*2I<#gDkr#tbe9!zYu{NPoM;Tm5FH<1<=!=Ima8yC$H
zgYOJXOeC6LYH`XC4*?|`Tm%o7|3u)*m6LOGb2x-A*Qkg5$dRo>E+Qh67%m0|y=E+=
z9~p7w)o)|L2atAF0516E4LL5!o`bX^Mn)V37LzY=NxlVgk9FmyReke;<(rQi9DImf
zLzqtlZ^1n8S(Uu>rpA8ImCVPCBz-`IVJi>tP+D5rv8L$%RoULXd)MoM7Ca%9^Eu at l
zjjGot{=9x7(z5)!r|_8U$J<}LsQC~zW6Od3!_Qj?GXvPjf(4(iX$7f3d=J~#$;zv!
zyteAK-t)h%mb1z^x3w(Q8En!)@#N*_(}M-{*)N4P58T^T&0q+hq-TlQ5g<5NcL7W7
z?&0xVBK8KTc=F at NSsHVCGPJ~Eh0icB*gSlBkj~q3s#6nqwS?OZ*u7I&6TP{y-tpO%
z&ztTlnZLwt(EIo}*h#=|F%e_#j#ecKgg(cp=hfr7_w at AiJBF%#ss}4Q(G+5ZZ6AUH
zqM)Ys#1uJRZV2cAV%epr=nCPQ8yQXizySR*6WeFcJ^;u2-rlZo$rPS?>Y_o1laJ5}
ziPcQ+ef}d1HhADaJ3n7B{Po>jd&$QAaTE8gpGV{4<9H2bRCIKPUMm}xgN?rl_6J4x
zaP(Uc<}h-M`x-XpL at cs^_!O;97v?#v^MgG0Cv&5^M5zmYB%O)dA0N+RoioLw)9|6F
z=;Y$!BGt_<P{X0DUa(~;;4(K-<c9*(9i-u57@~+7fs6uBR>In1tyW-gFrGAH!Gfsw
zNe}rHB~}`6K<Vg@@<}){{Nc|}k1;pGWzfNe)i>az5^jqvKp`-rl97>NI=7gu-~K at +
z0gvj9dJ3>0d3t)<#LUdOuuy>C@;8Z~pkS;pGx(+vd=?f4g46ktxw_M0^W*FK5I#mB
zAzIAH;?yblAtGF|nP<TX+rS9d=6~EB`<WTnOwTK+kfv5y$64|k90eNyb`c-A$A3DC
zI0bZ}YJoZ4GnHY~!yj87DeANZHzK@%fddHb+i(NS6+{map0E7-hr`R^Wzf(C8t{d=
z!6opjIIx;68;S``Ok{%((ZJIBOr3*=4j*p%c!wHo2v2f9TzA&MNiKo#D|fufcY^sS
z<`|p+YL4~Rm7d46D%>gRA8LGi^=e;5W#!k(E->7()+7brw{Le6P<SRKu at _iQ?E{{P
zC#S8gM}gAApUYgb0o&(zfA+x(P-%nBv6A>WK4Z21>|?D-$FRL1QrL}+hHMLeC|t-D
z^G(le{0S(*0!>_AP>%j&Ne|f1ovREnF7{A&TtA8<*R&f#6U at a$NZUZ&%r%M8p}pf(
z6Z|Ljh!kV)AV`q2`R_Z%p>plfqtjS_3$8ajJY4#J&3)IdT_db4PSuf_frn|-2lMXd
zGwLpFiWc*S-E<!dVq;?~0|tS-0<Sj!LhcOj2!P=Xc=W+>bK`q!>wW&)V1?S)h!J7G
zc!G?Ep9la3posi1)9okkcRi$}q{QPxtiAAOS1=i=Jz7y+jdxOk0*AOO3 at l^hL9VKb
z%J&pyKCH!TZk8FJ^>B~(2;wuQ#m+2v>{wy?D6tRnNfgJAyGw=f54h1e+yia2^>gIt
z(E%^M-qKXn!gTOEbSyyTO)-+^*h%vB?U`qjDY at xjNXG$XXegzf-v$J@!RHR8<o1^7
zMA&}>%ob1I!t61Mz4Tb-dze}36axcd6kb~JKSsKHHwso at q+CFk`BmVn367My)m2sS
zZ#RGsy1R!L`1u{#D%M||D$mfW1|GMUj*brNg%GB;sEDt-d;HJBXLAfVN at HzhGHt9Q
zgApsn+SP2mc4-ba2IZ!ym!wMP+2ir at XPcuC@d-|gQ(br#4><7WTNlt1*RZg#vZlin
z(ptRsE9O{uc>1&mJi{0TUY{))z;V;g4sXvz>VrSQAQDcase^dLUey;bUiguDWZ|XU
zOn1W%7#ka}EtPD#!cQ|j1 at CbjZ<1s16R$or`Ti)8XA7Mu1b6Tm_kvxyT=`*n<Hk;q
zw-CE_J32Z7Suca7fp8+nF9*i{k&;8jeRkGFL`;kw!&=kfIGF_O;e^QM;^Klx^YP<H
zJl4l}FC8uVrL9eVoXf-AUEDK-&-mfLKO;Cd8_#9`{r#l{76<fCfj^5lE-`T*gweM;
zeR#4yI5>!B at Wa1;?ZRVRyo(n0wfyJL6Jh)oH$X9BzkryS*x2MGyWJ76 at q1o$S7|{?
zFOSDXAYx76ga at oK#Iar2^z=W-s_zcs(c!Y0e*(N$t36fK#KZ)k$;Qd^7$?@9<FO8?
z&+j>g%CW)>GTj(5KigU*Kgc8o*p>qi6#35sk6_vqD8PRl2{r|14}qnFHPLT#q at u_(
z3{g7TbAy%G1OWjlW{V)*#rzH)=;E5ae2GPI2#S9juli_~zoh5lBCo2dl5o2PziQzZ
z9`4<Eb at OD%MZ*`qz7+p;6q$VR=a;y9w;JI5&6_t)Son7Obj`s8CuMbYI&W|9>{q$5
z5f6^v6#?n9x>{aWw_m?48KX*~^V_$0*=8h!yp95kq~-zgEo#J*tKZIng-_1Jq<A at y
zhPL)cb8)#nup?!&1Le|cYPi99d3h`}hkT>3ZWEXrBNNlhU35PfHGMC_XEgC(?dCh?
zyVwzj`F`MRa%@Kn%*S6~FTu9xqY<#&wxcMZi?0CdToL|To&Xb}=jW$FUL-C~^6=q9
zjYp4McI+b70AxnR!*4acjo)&20jtVa7AEmFW2{0$d{~_+Impb+94oA!^(ixxcVBMO
zrXA=UaaH2ZEZu17ys6<~7l5x1!fs at +-A=x~zGY$%Htx#8{$ngzn=*Ek-C7;K-e<e8
z-W{Zi25 at v>lK9i|+Wy~20Md9|e{6%cv9SRp^1;Gb*qtj_OabdNKivbw!2JjXCFM(a
z_*17(6Tw#^Y(LsIni!|DvKt~dZ+^`*jJRw at 4E~><n=5wj9{GP=msVC*V-pi>G#tRI
zchtQ<SN at We`b}tP@|M|H;5Fz7K|pBA0Lt+s1>X9~=GtF8I|ah4si&8?wMz3xKpYHp
z8JAQ7KKp_R?LEwd{bLY9-I^nLSsy=soUPp}B`%H?p1=fLKYcoHtt=oQkpD#4<`O at D
z?bWW14wh%9VZGJykO~4CH-HghwHsb&{PAvt_t at Bz`H5D^|2R_oiB;ep7_6|vFHpcL
zwRzyX78@(;m%OJ#CM)y0*d^54+Y75Ue><U74{h*H$i++MV{d>)GFEKS$SC3u$;k(F
z8bYzbjQqq25BQEFl$2O_3Sn(_Zti(R1RYlQ96o$Fwj6v*bPQE*Ft7gB@{$HCrC at P$
zy5pt;U{U|tzrVos#56QcVKv!6xf?|8`&(Evc0wjW4xs`YhOjIV7P`F5bw9<CBQfSr
z>Tk%)Q_9H542zkXeyZBc1jTB2=guAd`r>m(pXM6((E at Yhx_D83_6dBF1D;WVv0~4z
zR<$>5HiL~*VBm2)3md8R59HN<|9?)l7*Mjdn_WH`FVDtie&9bgAdahD_8!J50KdC(
zeWtitlI{N1^95cuW8zRNA_0p)mgK(8l1Sfxv;x9{t&_kd;ZHD~2(UNSQ^;p$XGcy>
zPEc1@&ob<iA2)$OOVN?8xeMtD(8w(r<ohx-^yGhy!=*p~@FpMxLeM;DLpZ2BFeHSX
zh6$dR^)|_K=RF4zvSEXra&!4GmH|Jw20UndHaG>sxmVOwuxmF~sm8~@$u;gN$W}>L
z{rs#D%pNTpg5`snJ7D&}1L12J+1U082?-$-FZl)qc@*2PG61(z&NaIKU&z<18f}d5
zS(t3s0_zXF^fU^r=&;O#2Og}lA2s at l?UCIPlTyV1JG!fCX?amaRh9d3?Gd}V%46`U
zUPnB$JW&34K(Gdo9mMWo>p!Ql_zW20B>{oD-CYpn$J$b;0j&MV0^oN8!3;GA%3N#e
z>JS4M^hRCCLBD+2SYLfQ_MJWy{FI-cfA;U+@)}=8Zs69>o;_pPS|DDHrluxr!iNyP
z=h(4hcFO~9SV4V&l+>ThOOO0UKg297%~w*Ql983wUk?orS4(=M%+`^rRy6g0olWHc
zk6)tUi*11Q_5&k at hre{`5)05qU1 at pwm;L=3glme5$M at uH%eKn14?YpGE2LJyr(+2(
z*avemL{A*U?yx$=2ZZig at v$>{iM at 7z@8Di5q^dl}@DF}gx$S47i61@?q@|^?yA~N_
zgwR%BrLuB)b at d)tAYhQ#NDTC7YmJWN*sofC3+DAiD~xf^%DMnyYY at PMgp~9q_)2Gv
zp(h2S=oVOCctU>m+&O=;ceoEhpj4Ap@(*?88oQ;WoI^2*<DzL;sDi+3SXo)c at 7`@+
z&bOh at O-<#1Pr*KSc=ggRU+|u9r|4)#e)BN`E)&3!ogApZy4*eoPwT&<ksm%_qE382
zJrQ;r0j<Lt<|kb_DZp~Jy)7Fe#J3Q>g94VnOC>o!$CzK5_rp at LRt5D+2=U~H4($Yj
z15w_F4I3?he})NI4OSjg$0%R14cILr at p5b`Mb9W8D2TvH!4AC%4=3BTYZqWWHG<OL
zzs}eUZRp`1OpjcuDN{~P&Lusa>!hH~5iuXyN}$G|J`RAU#c~Ycmj|vlh6_eUM at O5-
z;};OceEf)gSIfW`Bje)S9%O0{^jP)XoNP-0BMK*{C&H5c(9qW{EpnLpzI^!vuG at C8
z_H^%Oc4K2eed(;J{{B<PEF|kL{pF`<+C-KP^t=3LrVcXxtmUz2pLClAz0fPxhXidG
z<^JbxA9D=5USijMQd%)KZi8SAKkqMrJy?TFeU8-tn%H at SixkDi&W`y)*dJ_!MRz;7
z?QX&8;K5k+3hsb`Uq$6)iykjf;{Kr_JiA{83ewu at Ur7P-YwNF!=_V#7mcg@)1bu>q
zh?Gg_A0PMl;=D&|w*SlQ(vlkhN^|5P*u=x{- at S8o++6Kb%r&Cfw{IVpVdn+A*)Mxc
zEG^~vFW3-ENJ@?{7$umjg6jD58i9y~oqZQ!Y<xV at es%G`(sfX_ at pHz&ppFPp0Uy(E
ziI>JxtoP at XktyoE(N0*M%I!l=_+MruN}CPH$d at l)mRI{Wy-U_cIN at y@#c$r+UFAdP
z1ER&n&8 at t=To__~c;HcX`F^_1<vlpDJx3&wZ<|o_GUDxB-rnzZy09Vlm8Yksp&==D
zxWZzzw;xkeQ{&?1mPV~%^CHK%&pWr*W`AX6<(8hkQvp(&f{N-Dwq%1=#^XGIQtoH7
z8V9H at kN`(}_;f!aG}f;5*)P#x$Ny<&=AqaCwXzqx_kZ6?iZHr9*4D1A?DFvNz^ZU;
z0X%#dB<+jZS_y(vP|$;vvZ;xyJdRQ#cf343_p3RsTUdT#($heb0L4OzCodoGgnaxT
z3Azoe0dKg%I#+C$0o>6^h6lj%KpPwWsOV at -W8*YH=7slQuD~j=Ic|AUeG^k>MB6ef
z{)B9TT%O6mPV7?uZ{EhEN2TAtlf!~wfG;Vzy0*5ap|780o(JsXARtF|%&m8&7YHCy
zQZaZ5SRO#PLIULHM`UJZ27LQ#=HtS`!tJo-dplNUAVA&W%d4a8<cAMC(+j`U0C+!s
z$>c+Gr1UO80N0k57NyFr<`N*m{<+zinLm^&dSWGyb^QFv`V7Q`Pij=uPEawZrf?u3
z<`W$kXZn2RhN_3uH&>2<28#c`Rly`0&aBYXq-Xe}J=E~OS^!4I%TAAZ{39ZKoSX=R
zYa${<fQv`H-N)wk(sP-Q-Rmy4r~cI#;RbB9nO^#E=@v3GG`|T9WI1LK0e(6+H+L_o
z)#T~Z=V at u2J9q9xZ-U4>Ha5nB1hJK3%ax?D>x7L>f}^$=mX%OiS0~9)3k2mQ&f)hx
zgxW5Ue5cz%x+)MvNK7H<;v6Gv_%gS!@F(-3 at +d1SQ*!NUZp6Mjbf2szO*`=v5xzPp
zDQPVetmE6aZ%2goF+_DXL|i?{$;k<Nqx9#`2f+SFN<j`|0{UB+z_AGX#i0M6?&OxN
zlTu$aC<RW|KRgTo9P<HU2Ug5hR_+D{;*^n*0dOt&9zyUym{sp}Ik_V%i&IHpFCyFa
z;!eu~>S%ioMAg at K%uj(yVtv8=BzYV at f;~QKYu|H?)xA&O6!|YdJ241vig~FU!op7n
zEB0bZglwD?n(cpuj8Tv4-f!!x`3~%N8L11v9<Q;Mgu_B>v%)x}=$oO;XA*i7P!TF6
z4yq2Q*4xw6t5cM7Kgc8y2xMerre7>UJ~U<P;|+Cqk2Yw6eRR7icoi^(^PW at P8*CRf
z@(zfSm4Vz^n8?K>{jN9!2nyea!0+5~H{fHAUp9 at wbSjQt!YuOc-2)KlwA4bV!iI+*
z0prks=fnv at p!$Q!w;`s9F(Em*!gc>KoNM6jP6E1uo`>hKsHo`JkGO-dF8!b==;-J=
zIy<)%#X24SJ>QF<^ee!UvGdEt)fM}d+uGWWb7?qTQNF0tgz)!qtTS65b5%J8KX+o2
zX%NpIoJx|Mf-xyL at YSn*r<s^)Z+795fCj~ey6(@m#|TJ-ftA$V>}u5+$jQi%UM6vm
zPp}=24ImIo*8X^w`I6%u*w~wki at dy4vRS`#%kCLuz>$t3x98*IGx+)G=D11xj?0$6
zkK;_nqK%bF{M{6AItwP3QV2iT<d&(^XiCtriP{d;06SZ*BL>UB!9jL|1T#)uUDVtC
zfv$yx0$b)D;2!_rAqv(V*p`DseSozOEYA%AQF9IoIsxlOe)8n&U*Wc7_AB!PPpZ+$
z4bWNj+fyq65AgsPW#Zq_`k`yQ^S6gbNA->vDnxh`f_5QjXlX at 5FrJ<!!|pc$0iHf7
zk%f=)NmzsBxHd!wy#6CN2(LbRBvx8LK*IqpySMngNKAZdl7d#{tE29op2A7;DX`sC
zfuC)c9(LzFT?T=b at uZ0Xf>h!=FKRw)lE$H7EA!*g;4u~dc}#d7pd{Wn3Ly-oEQDEU
z6_wYN=afFEmpI^30AB$CT^*zgzf6HP!FzXnd>m&n01%)MIeq5LmyV9(NCe5qs+ at N0
z!X(*oDk?S_HZ<HQ9J^%wR~wTey1Kg0fPBvlxY2!jN6dO7MV|4 at nVTFP#ik>#*35vS
zxR7Hp2CV*-D_8or+vdzC+eiVSB;4+bc-4&mh#Q!i@&YS2ANzhC at FX||-*>RW1LuFV
zrK;Bb&p_lH9E3tcLowfXJNybjPyBPG5(hz?ss?fL;r9>MAxh8VxI4_ at 7g_&JHy at XQ
zAjR~bc$#lXkYlT;s;YJE0-_==F7AY_ld7r%Lqm*>dG$YYJlw;*r at C^rYW-N_6BA`n
z4A{KF5jR8CKG=|rJ?bD$G at m?icUWK90(}z`uA<MMH8Hy!za`M#f<OZz{t5tU0Q4ug
z6`|L0)Bg2w?iZEkTTTHMVG8}f^2Uc+T2KG|c?=PdKuAtb#zd3(OfSwYK!pMec1eSc
zLU#d0HNaDz`S{!BNOE4~5I*Co%~i+EgLsh&UU;l8wY@^!%n^1kz^ZX<#k%$QbEeyY
zP9PI;0^j8)dg(yA(6nE?*cmEl`~L8Q at 3LREgN{r^PoWi#2Z3J>6|z?fH|zq*jnlC_
zbL{QxG<0<nFdBRH0?V3f5STf|$$1yEqGNk&8yc#BD-M_a2BtqYIT>x-w`DV>&vy+>
zPVSeLmEAh_VbC)NFvhc!lb212jEsyhmx9?zMn*;;Dv4*9nfE{}0Hc4mKX+(w5QobU
z(layRzhZ-OdwO~_wY5K*42#5^26VtQmE`-g()>rFw8!!IN{r6l1;?cAKgJ>V|D=w;
zeEqr?p_u7TFod3#(vjg|Y=U)ga9IBLD*~9Vc+wfk_gm-Y7@$T+F<mUOeZ=QZMFcF^
z^BCt~OM)yq84~}%z<BPi985ug_n^7ZVtyw5+1$e|Co;{?&nG!}Fu?PZmX=mzQj%xy
zXIsWgmyTn**?|Kt at PD^BEd}ohcvU>yv;5t$<NrZd9oG+{^ykjq;pl$9@!51_tI01q
z+S!{%K<|i=75I#5q2<W!Ln8I139 at l`sc%Jv$fe7dS!u|mGo$2_aF`elhEsQ3=S40T
zE4;lDI|<2ZpQ&)#IgUB8x3|Z9LO+N%tntT5m$!q?1KPS at KxPH<<O^Wj>}ZHF8x0w+
zX0ieOR$k`1x;h=eA=aB_Jg%dJ=#P`D3aw`Fs#_c at jlm!6Ha#b&YIdM}uF-MhFg-ne
zO=F`1{~D8|m>5nv^x}W{(&PZTi4QwObsT%Mht7 at wYyv_QA{;sxj at tu;g1zdP2m%tv
z^{|43^2t_~S`)rvd1G}3ClcViiji9XvG0+)MjOI#F=ESYAITKown00DwR#3&$7Ey8
zI<`EC$FLgcx0jIpibh-VgG)~bT^@4to(2rUH4s|v--VYh<0vAWxC3v48S~04-5(gw
z;9>f++tr-o;+U}>6MPhB?redv?UE6mlIvwFjhE}T1=rdxUEf$~FDa|878~c%%ePrz
z1C2p!Y-~I>F(K5px`sox!0>J6$98~c9H*v^mJHy)sdX|kGFd{^5Ud<myZR=mH|((~
zlw{w&htDrI at yIJFy~3Fo5O8q%(EcsW<sN_iXNJ}u at bG&{*9z%Jg(7lZ1}E)wg7?Pc
zGbJTudU`sZ0$I-Vs{Z;_T~V=>{{v!xT~XU%T5Iw5Pw_i<T313>M-s8+0^Hys;9X5^
zt<6}mH3#om(s31==Ks#1ZwGQ+Z`wHp!3FP at PEpEkuHn?t(HR0c?ds)qCuOaqdB*pp
zO at ok!%K8s}oy+eqDNMmj)ojouEiHY?e&qrv1mD0w5>-`Itp3Jev at M<igO8S6*WDM>
z7x^}qbipf<{^TxcVVpN|R<+<rt}zy40u(!a`*vGIRFq*m=Z^E_lu}+oU?Dzm+zCh>
zPO7?i at uGyA#8$*UN=b<;hQl<QfBcvm^yD0LJ}7i8Ldf2aEJcdekB^s&>)MkiEIW7Y
zikK*l5J^qt_}2$=R||v|sKdmijftp?UkvPQY<Li1T{VS6<hCc}k>4n7X*n`IJ-t=E
zZZWwv5~i at Qyth|v{CkM&$3MNFBmDf104>yjCkVWt+brAFIP`t7(6Obj(s{$-{rmSA
zY;BgNQ!?x@&&92^)sD9~)pfObgO-Vt^C0lcjv{Lw*xp$Y^$_=IPGD(x&V;#IFv!xb
zE=)9UA5Oy)fzYjjI6p8l5}T2l at k@yAzHq~Ryh{_zZNt7D)Z|QL+S~_~`|xd`0q}vn
zMh at 2AKYHv~c~1|`^XJbyK0THUTdX&`WlcJ#Vk6^1j$Ma1stZ8%!%a`h?1v8Wzc5Si
ziblY)<<-ncfO=Kq)lIv;U#ttXx;4Lk|7d7!{}noNdn?qoJQ2?{Yy at 0W&NaGu<HpYO
z=g-gfZEhT66!WeC<}wR0q!hjvQ_F}JX^!jr#k@}r7<BpEB54{sOn+zKxO=$E+qZ|d
z4)%ya>;U+>(>W~f_z`zk4EQqADbRQ)z_Kvh;0RD0&;`UF69rY?H2<*%*9VG<8ofo<
z*xiDsUcljS)B%>cg2_<b;gF~e*=7X~Yu?buCnwu6Wgfe7tJP-+z}x-x>k2$t0653&
z^-z_U{sqUwu_vgg${>#5*gm24d08y3*t>Tx_IU%B!2wfP<$3NLDbB=tNk+48yK$Eh
z%oC>)Lc}luQ8GL{%wm7(j=Ze(OdC#q2F%jhsz;IW1e4y_SSlHUm+ at JrvR_@Sgm{D9
z+}IVLo=#rvLnqc(Y>#z5+2uodPlqr{fP`shwoAd&F8DIOnm#^h*`uI-fv800<sFZc
z3Ii<JN{sPd!hwrGq^RnhV(!Wc*WW+G#PqN!>L$U-=jz7s%a<?LeEs^Rn$Gb6K{3yS
z{_Wehv at 86dZ5GNB?<dW|AAi96xMO2+E+C%mba!`)HQWyh3c{<YJ8}%EfkRGMy}5Is
zF;*vJ$q!=>j;nK68E5v|X?VuP#ieF5u~jd>egbx$;Pn0bJy7Jlfg97L3!%HiZ>MhP
zhGmA{+KR!!e%Mo1V7_-dQ&|Ow|HjYnbYspGC}BX=hMXSv3%SyN-NWDGc`yWHaVaUp
ziYRSK;Gw{?1_lQH82kis%XRUhD6AR9cDEKz87nKhfhlr48FqR3@?mS at eS9MvFy^>n
zhqE?8M at hKJ*RB6Cw3_a|-j!phsi*g9aBxtkU?y%(iJO}n at ENaE#5w9X1$CiU$6iZM
z4+~*BIuxx(r}FYH;n9M%wKaiYYikQ44(W%W&AgMF+nxf;soLj6bh5Djda{a&?m32C
zXn_Fm3MmWW;o&&x1uxA6Qaq4QaZlmyiWQFg16#tdk(HIzwjv!ZN`Ln30a$u29v)kU
z)w8Uu&q1j}ICJK&e<>`o0LUp+v@($doWBy_UK44Y_?FBN$;K8$OH0e;#V%w06n&}u
zMz at i~35GLgc4M{_9>!$6IhL8S>+KoI*{{qI8qG1c at VKz50BhJ_EpYtldk)nV#Wct9
z<}>qSO*n*sg_SiSdqXonu_;<?XtX{AtAHXSBirm}RctQQFQ_E?fJ(&#7G??Bg32Nm
zF0)MCO_A4cId{b%|Im!@XlPQ}%;mbEtq2!?i<^t<jZRl at vAqB$u6%u;V|@g+t%61x
z`I)1ft#_s|hlYyk6)@V^az|}BwxjZCYNWs?6crUqOG^n;o!R9;@7%%2tOaOj!jxQp
zH-B1NvqL?Zo|Pp60t*j-2&F)Oab`_mNJumgcg$HSO%jXSc!7}=%~esu?_#YzmR$m%
zwT>UhI;8%74XpO7sUZmyu&R2od%s?Nu-f0_6E?q(9zBYu0yc}k_d=Kt3=WRcX$(g>
z&Bo52JuMt7GZ5_~bOYiHC<ziEv`9WI_-Uhbt^+QQlY`+O1 at b8yC#4BiiKTTo`(t3>
zIlwSBnBZjEXU{I-KtozuZ}eGVH?qfD-Yw0Dm)_vn*I=dRSXVA5&LU-IX2wgFytG9%
z4R<MlSw94Ulb)OF_Wbz{0|Nsb7y_(ZM60l1xLePXD`6u(HPyPrZTx2 at GbM)#2?129
zSjvJ=u)1Bhini*X*VW!M&Y+xr|Nfmtsrx=40OB4k&T(H~>Z2Q9n!WI|qvI6NUQp*%
zRbJFM+qb)01*e#Oj}*q4jY>HN9(b;V1K?(7o$~W9+bm4DBAr7gw3%l)&%@&{<R_ZK
zT4ZH4)9VFcfQXn_PnC|E+7p%lX*~w4EM3{?e+Kr)b-!bZB4yIq*;!gn&IRyNFvA|4
zG9W%azI1;4M?{DIAz#ZIBzAwkGGo#Ux$(&>&))o}xLY*V=c?((dJ30z<t9jmXQk$|
z_BXDzKycX$hJ=mh5bJ@$Q{}(W=?VA?*1h at _@DWHd`SIf<u(mkO76`r1Jk>!n(9VQX
z{j|^W`e9kgLOc(wCR+9qrUkx_mYd$6Y)cU~H9Zgf^Y3^2IhBH|;gUX_5GJsY4H&>|
zukB2$l6rt$?)+v{6hqPaVpn=b#`EH0hm?g*Idugy0pLw3t4YOcniFr`()6!d8wQ)D
z7We*2KDCv}*Jfj5tAGGh*_o|h$(~lkz;8ZwihyJbDJ#x-p6jvd^8lF^DWBv|wv>Pa
zSBacfg--Kss at wmJmyXsGH2&2P7Iim5DEqx}cb at d5{%cyng0>u>$%7CLm%Og0$a^Y)
zkSNXoHF9V8*RQ^%e- at V)KYcn6xcaQm5Qx|*cnZBKRM1MfefefOf}0Nkhd_se{YIJ}
z7ZJIm^z>;4u%E!?>Z=>{c-{!oj)~w7$`S{lc=GDEx&QsuxZhtQc3V-;`^Cn^dAjaD
zwvU~i9e)?}J|obeSaZc2?w$Y6`%9bwP(~WDn*fiYsNWTg*H$Apw?>Pm%_%OnzG^r)
zG_;#A(BFTG)3$s4iWJ9R=|8WFi;H(UDQ4^4PZ?4yv`itKT3MW0EgdqfsjaKKdHeQj
zEL5p<qjO++dEf*?XO1EJCI@(8<%jFe6(H>^t@`Yl8zQc9*4EbgkriIoWt&ko1WkLE
z;2tgJqo-P8HKT^0qWSP4LrE%F>~UT_KRk050m2sMU}a%{rM7S2slT6}``FkMiX%r(
z6Wouo%RlLv28Y|}6v%I(_j!rcg&bVq=sBe<a-c5u3~6E_O92rPoK>$+Fun$Q0&L=c
zbwFd!K`;^#aK?AaBpvhJmGo}8n>SsLpMO-d%j|;0{D_ at mYU1e3R9u9^p=eRBL&U_y
ze;jV#xl>u;akvupcV at NFE>s2JFNKKU8x-_l(t_vwd7 at bzI{)fy>%ZDNt{tGgqByxy
zFr2vYuhDUzq@<*8SXhdMVB?voXB at 5Su&tmjd}in8k5pDxD&;2B&HQrs^`w5`pv}75
z)Z}&UC${TiU(4IGMX!a71||ixEQLPasvDv3UH<#>Pi~*X;hB=n;@PF80At$Zl$6RE
zU&hL-oB!@o96H2cYG&q7w&3x`dnU6{=rBaQsKi8x?n0}(zq at kp9yoDcq*tno^mSIU
zabGc3q(#NVh!idL+U|4$5%j&F2&9IRI_7`}0S}^zic~`REhNl;cSsY+_fwwDGd=);
zC7W;xI1x^)xv8cW4%k5s&Vb1RFp)fy0s9E`04?<#gCN2wz%N#i?l_;~rlw|ON3ne&
z2!=yW+HS16bW&1McrY53oP7Jyqet!mYBskB<Qp3s3u~Sft$k59Juz81?=mtx><&of
zv3oxy2^}3>zHi0XueVT$W8v>$tv~yAdRk;38w(4$y4{?siMjcam6er2;N6t)V6qoJ
z?y<1s0#3;+75XNZM#wG?ODSsV>JGE{OW-%B6|YD%9F&-NXYAL)Hxq{~l7v&YZ{MEy
z98vxCv5-Xg9Kf+ at TpV_%@q!796k1LZ5$3v#O9W at 1QAZ~ypQWIpV#9+aOi|*&pQeGq
zSptc at z5T%SGzXJZC<~UkWjJm+_+o`h&8sa3=<ey+?c}y$zkmP!N(fCXK!Dli=jVs>
z&YnGsQ!7No#ogPUH&Mw{rv<N0bYu+e^5m11ks%@gN-b6Q>R9k-e<O#0+BRk)A#ztu
zMJ05sGdt-?U-4)6SFdFL%kD(ENk~ai5-R at v{gs{S=yyd|)l3%b_N|!!Ze&eE1Lerb
z2m>3NMA2;7{u?BZS4 at S@D#Xep;FkxUfBcGzk1t>;AuWv)pUz=zPE1-+ at mXJc+TB!%
ziGZkcxw6vI^aK)N;c_?H=!<i6b7jCo$HvCo)6=<9x?0XC+?SI(4dD-Sj9IC_^5?H!
z2?$r|l_jA^#-^#S&yJ)yuczF!sCk8qhTpx<aZ?q at d;7nCoozkFt6~o}Sge-|M%S}-
zrSexM!YeM<ZRg<_&LN0HB&-PL)vrJhKS at Kgd-m6uerjqe#>+tvV26G+MjZ3^_m^;6
zt%-S51u_|*qk^x`j?oBETfcBm&snuRlj~~l{rwk8M at C0SmDGm-9;xbLz1 at wBjF!A2
z9XB_2I)NfLPkHvExf#|zRx)5!sDz_>dVHLf;2z<ylF%juBoJ^pT+p_yQ|#HZytZes
zgL<FInlcw5QavptFhtn^i4DedOy%nnJR(ceJ+APW-<z5!KnWZ+E?Lh($|!#4&Q>8Q
zgeWKAKb7#YB8}mK0M>wbmQ$TLDI(nZEU&<&vGH*a_`k0(3;=o1GoS_o6!oGygU~y_
zu)rIx0~Tnk5MdV`gOL8p({uNS2Qlk;tJa|^qtT-$J{J`M{(+dX4!Yo(6A}_~FRAt5
zmEMn0pFYV{R9Aa~AbW6Dtw<3orcAwQozQEqU%x&WX4Olv37Cc}e_>x&NFpfsR?!Kk
z<@6J|Nrh&=yfHcyIgU$5{8XKYDyAvX+o16R at ckCVGb3PDbfKf--4mTl4r_Vv?)fvX
z6-DMB;l%qgpgllbUcG*O;C~hOiSIxZ?6?M+=GNqj*w=S$ynL<EVh49zGw3N;D1OS?
zv_&u=7V4jr-P~B;isj%8v at 4wp3t9X0In|5Lg5AMOKARg{HBx9-X?5KAOQ-qh(dk2n
z4(X)~{b!1s6aMT9`@mD5{_e<dGg?hQnU=ND|Hav7ZEep&VcO*66XpNM0&Tv&<JYm+
z;AGi)L>u>S-)>{k3r>oJ1u<Ft{QzsQDk_NVSLUz7Q;mqZk<s{(Ej<Sfg^8h<xH$c@
zbQcYdV=^(zsE3LGf^-)l-l6N%IoYjB;H1B%hG{rs>+s>j*4DRf?fv%cn}&hG>*i*e
zltw-t9^qTJJc<tL<}aI(vk{ztmVz5(p*<QseE0(B1txRhsk90KtdQEnvz$3|hH4e`
zx(fvHs>M$42=Tqhi|y>Lz*FI|e!MgbCucPd?fWx)3)4pY{8SK1U44BiR_4cd0qBBE
z$H_e at O4%}7CBF=MA~%GE=Kx^`VXaI|P0PCTx4<v!Nz-7^?~>GiiC{q|K=;?~%C+?i
z?J8eX?JJ5vfWUM;@GU1$^)^$#o|*>v1nYEX>Kw$e*&Q7n{jj50i3TbhPZY6 at 7Q*|4
zy-)UKK<sjyGY{NWGJs<MXgi+#NJ~?gTUcP#E3KdnszzE`T8OMJnVFeuR=U4~Yi$*%
z1qB7;TlGLZUsq6AdS1=^wYIhlR9*%S@`muH)>~kq4=)fWl{R-NT&|%cKAr>uM>&Kb
zEXTvSv6Jm-Y1v^)DKEprY2oXn at 84hOSY<zV&K30O%F>LFVOLHovpGY7&!0c0SahLS
z;vm>-zod#4J$S{i+s!QoEyuif at 7}TQe4d7I!Tp?^X-bom!OZr4-F^uPtP+xvfUP8E
zW at af}Y}?j-(XjIrJG<mYNp3D5)16?Ky@#1<{$<5gwzX0I{P`2RpF=JhUTAx)tzFj8
zFzx$StEs68=Xm<Qd*_U^(TA$Mmg*DQcoEQtRi?aY1TM7AsjSe2OD;6p?kpWBcf(=Y
zcJMa>1+1opa~(VS?kcIPs2q@${UK(v;k@##y?aCY{Z$Ft;#6x#!(ipqja%WFXPcR}
ztpB!NHkCrE$`@eLBh at 2y0k9~$1q1}<C)*F=P)T^^qN1X$($E?z(3>~`j7O)TAE(IL
zEpz?(`xh&+e)oQ+E}e6lwfJ8xz>mTQ$#<(oS%dy?H7|B9Zl-HFc0anUYMkfc$7L}g
z2TbPR>e7q?rY?wCd3X-vy$k;~e-{4Di=Vh?(0^>TINqJ*#ls4r@=`*o;A78+KEeJ<
zALT!4eh at mR{eZ(z{$ozy<5#hP at Ab8F0wojDsA|t25MVGOF=Bb~`x{rul8;&gwa31F
zC*M!5{`;mg|1Icpd+CmWMXQL_S<U~hEVzIFzRA<4=-*pY0A~8=U?nIiDNU>K{3gm0
z6Y`kR!rEN$O3VX*9|7gNn at O7+xxnV==!iqz9XD6$KpPQTSy?G6Dcy+s at uj<at)M-x
z$3`pVWBlrR*B&}Mr^{xe{lKlO!7XOSE}d(NIp^Xe5sPE at DbFerV7W1$Vq#?_CM88i
z{dq!@>$u583+riSW=Ya+5MOv6bAWgkaPioeQ=OxRt~YjwiHYS$^C{W>>rC+p3!_1^
zla}5gDOpE7t)QgzYN{)DsBosZ9B6K2`{WM7e{<ikn#YaSCceS-(PmIdq|y8A1pN(c
zH}jzt{GK4o47jJZ^z7!%-8kP)N5>5)c+cWfJdws*aDavub)1yu!O^am8o<%f05V!t
zuYCm$ZP%bh`}z5mf;<L86nXY+*}{h}OR6)lQ_k#STklt;2Ibboc^kpS=^oFS;#D%d
zT1i)zm4M0RgF?%v at ad$ZMSEridv2xZ^_M<ho=huA*Kg+l!W+}zfQK(Y;kZHYgPcvj
zp0-}cS<>&p9DdE(`tnDquvfqUCqDk8l1=I?wilqa{muO(TZzSALRX+McGKrqb6C5e
ztgP(QM!}{2jB46)pv~0tp)W7$@-067tj4*1RZ2Fk(Aa>cnBHP@#XLsI{#``o!omJ6
zFNx}sR#IYHoay7l8-VA=qJ6M6X2-R_z(7o|VM#Tf_4&Si%P?>Ax%=tlCF?&igj2AJ
zv4(_-5=X;$V`Z(}cB8IreTA?oSezpSm$bAQ(g-@>`ULlumKNgAY1AVtagHTcu`<9&
z>O=U{b7w&85D_9}5~$2yzdRWce+IA48H8tlCqPUgo%Er7(p5eQ`)6VUxinv(7 at 6l$
z*3kF>#PJY_ at 59KH1@Hiy!h)cXrgqem^4y{6>1l{;B3D+%qVevFdds4hZAlQFh_E_R
zT4xpHfN|0&lMA`1sOV0o_=JRfqs@=0tGd}rn#RVQ1osr>+}^pTMse?dcJN1&- at yA?
z2;^b>7KeHYt&R~YenmL;en_6*M3L9gp@=gSFreU~lYq&No3#M1=6+Z~GB+Ax{OV4~
zMGviN?^av29Hyxe{am3f^-)t_pRzPZBVg&y9APJ;sH`kGgTe3sosfz_NBVUfAUD6D
zAMxY6vaniG0z__60-l5L*<U$zsz1jkL%W^|34+7wRBo38DMLZV!esk?TU%SyRRCYC
zR{wG8uF=6khWF=`<!7`=Q{G+tcoFaw1VN8gUk}&2llqs9?<cLqK`_qHEGOo+TQ(X0
z+}_rP_ljh|GgFV|g(<bdmlF}V9oCj7)vVT*=Rm~5-kvktKW9~6)6ybGAcr+A)8Oj<
zbJ;85{)q#`u$L8q+*%U`zw+g-hD&<EUwVq!il-1DJ%JYzU!BojFrqhF|CgPeeUu=g
zrluA)wTJ9v{8B!j&F3nEk{|qb_Gq4dO7>=<nZJ+xmkQBo3_q%zYjhMmAaS|Taf8vo
zz<`JV9?U#lI3t7?P|XebTz$2+wnkZJ4twElG1=w;lye^#KBzjpx{8fJ!p_dFTQa)x
zIKwg|E-vnp&3t@(YAU6e_eo!>7&Y6eY-%CLg^AXkP5 at NL02z+v=jQ|7d4N^`+Qm&l
zLvs#-sAKsrzTwZDDtY$^<f;Yc6d`=Zl(Dg~QCV4X2Ef+MCt5%sf+ZG=*p{?viJa42
z{{NV|4tT8F at BNe_2^mp at P%2wSLX at 41jF4=i%(9bEW{Oft*%{fZjF6C!GQuNeM97Sg
zRl@%q^#1;Re17A7#q+(tXI$60&$-UjxBm3hxcgw!;k+RsA=u?xgXIQ4YUx93;18=9
zZ>p<nB!9l*od5BKGsmcm%UuVbe5}$%BKGsEt1m`+D=G-MY^w%%byr%z0m_z-o at XJd
z!Rzr^@j``KO=DwzWU-=DM~l$Fo0^(7`=UCwysN8g>_E?@p4|DD0Q6|_5c3KO=m~ED
z6^--eyzCk1R_A+>@2be|4-P>C0Ib#4b)plmCwgaoDNs5Cyg*T++Y7xuCU;mQbr=8q
zQK7g3-mlR6`g#p9UArb^7tdRuY6nzQR0hL%FA12G%{>91CG1=I`>Q$nwL2##C&snu
zc0H3 at A0h>MXDnLwIusAdSiakte|^afRp%j&D`;wle=fgP3^0GQ?`{c~i{;Hr at O!rs
zrW=HtyLFATb$PIm09NTR{#)L~0x%<%KHt21r;Jy)9oK#KuXn-PA`j%!h%@|W`g4$-
zPTL;U%{AuBF)ZN#Q<S%}%fbTAPT9D^nBt7HJ?{V<)aaR9KR`;!v|sdw*j#4|L>NRm
zn at -@Wro_CwGmt+*kl_=63cNZ=Le9Wvu>bo~@9IiN- at wr#-9mtBK&lN>^B)ekW=|WI
z76Jq$*U->tDRwl`)X^cP4S&u0%J()~rh<fw3SPkHe2ba at 3P$-07g9c#x#x~Rka>|^
z=BUc@`YC#WKX^gxe_miF5vvXnKi|A&JIZQ*8y-G*Cp2^`;oHy<m*QPkzDK<0I98fh
zf6ElVN+eEXEPqdYeO_Mvt?N?zLN{}{m0VGW12cps*SeoyUZlQs`u<)mUIubo^y?s>
zNuQ*O-XG<9y1F6PCQF6`mmo-7cV(y(QxFPFE1_f-Q^iuo<ZgL+K5Sp*;NnsQoQSDm
z)S6E~KmgxQFi?~bi1fNL2STxYC6SY)bT{P6KrOkju<$nsMc6EUWwbSg>*@r=<e}zx
z6G!_W^#|i$Wk*Htl$Bcg@#1Lq_a610pH{_FQ<txRD~^Bdn%Fz+X7eO8TlkYfu|sS@
z^UbXk4!=thC6Uqh4-6={xD-5=zFzurIPO~V;UBfNwQoLrIE^-n!VgH=I at Ibe9$pK+
z=+cC at _Ya&N!$^7poBMIiL9y3r<J0_~*{>&#Uh|IRH^SxE<#oo-EpHL_y at vO$I4pf_
zJ`3kyJw~1e`5Wgt&i=YJQfqEuQDfxZb08)rW-Qin9XJyi_<)VREGz3+(w7 at g4oz1(
ztEYFHjMby>nJDA2PflMmblZ1YH^tDf0kdLJBYz#PA^m$WknJx1SLrhI`OJ*+UG(&Z
z3kIRtYvELfyM|6f>Pk=kResSO%_-x%cJ0C$i|ra3&0T*gUYu_)bDaC#^XjT|d9%--
z(NxDjLVt>D9UL#RtXSP$eKBbgA7-3c at xb{}?B=!4mwf^q?1lAR{XEkfyM at gCT_X4a
zG9Yp1;N-*#DJshd&wY2RsjUqGL`XqRO>Jgl!wumUOD&5(UNn#K5-;VLFxVFFVr0Ax
zr>>3>4=7Qv66;m at 7^$nzPv(|kq11Es;<vFeZb-fUIE1~hzzDelD``i)yybeYfyL{;
zy!`YGcLWZR>YqDD4!8ioN76Wh`i9jpi6{fjd(Yw*oL0ZQG at J%uaXSI8`F89UJkbzy
zOxv!jNX_t-hvwv3U0FWV4VW$Ff|e3zYcl{|&s3jqXlZFpj)*cF<hG8_p$rtm at oV$+
zR#sM66*ph``#UisW30&5K|w)*<H!*$oPT=!<VkyUO1^7FXLNpkK9=dYBi8CpxDVmz
zb)V4VO%tDvFl$?9xemUg*>vLbmoM#Img<eorh>sSF?*-0sbzwJrJ|lkh;g~+sd$Tn
z2M<nHl28Xf5w;G&!~~?$Dn6mp__f=(wxJ;}hP0_~)y%&&FYQswpBjh%-T%cpepk8Y
z+<oa!+XH0r%C8>?$x*3!_;=Wwk(64>xN8 at AfGj)xc7>#a^*dWaM_nYaVF=$AtM{TA
z&X<>Pu8%2;KDl#e$EFh(MWV)8Z7;?a2o)Y?UvbDWdgYO?BU6~4PqWbDZW2pjUN~#e
z^vO7rw*M!2JH{sf9dBxD6?g0vKLbad{q=hE^HbAqHfJ#tgQtPXBsGC$7E;-d8q&%J
zXJl}%<#~Wuw+HI(9vv$wbU6Fd;oY!1xH7iTz{!#yJ$e)q*3-7pDEpr4*@47w%mxd&
z8n!al-|bpow;!{eT{VtbD}Nh20sK9HJEc5l5B2x=C$wC|8C^0mGXH>pr;fd^TeHEV
zo$BoAA-NDvIIl`*>FSc6KYt!*N+SF at cUOVfIXBV82v5g`u(l*=Cryu7x$)hC0+CVc
z<c(H*!QDHMm{fQ#W7jd#!VQ|Fw1$jrXup5|jx7jS8^wkyYN;s#?#bN2EQX at kPIa~&
z1{sfjYoHv%(Oxe<c**AV_Yd&7LG({d<bZ$;U%+%Y*B2_F_<u>svCW$|CupT7=y22Q
z-n}hCWh>Ghi(<SN*DrJ9<2NeZ!pzKEGElJn!i5VXorOGjuYxmCl-|atPLZ>Ej<637
z52Ks_n6q8&LwOb at H#adZ^%xxGL8$=jL2o8szZ#xUfQQ>jOWT at eAb;(xLHnFTvF-)I
z-6&!Sp^3nwn{~0)KbsE1OV4D9j<z<c2b&Hz#o|IeFfZX}KVEt?8RxPO^G2_CDs^s)
zRVRa7EV{p at riLA8B;jMKM!bYt*U6J7vr7-BA7BvEJ$v>WPM`pT*#Sf{IVDB3D?hk%
zvy6<4)bGIuk7aybcWVX*Gw})tPy;!Oh<Yx{MEU;xdt6s?@7~Vz!)5S-fxd=I@~`o(
z7j&|w9XUpevuzr<AGG`WLeE;4D&(EZPvwO#jdHDtF!-#li2JNawYIfEcI at D&OX*5&
z8$}`FhY#uD9mr+2qu6UeKmdUNk?j!u_B0)C_`;d!>T18x>#O&pqbVSt;j#}%{tqS9
zGqF1r>x)PVQO;^M2HvXF->GUOtvDyf#ujg-eu}sP at fuh4qVNdd&^+&DPmgQYKCs+g
zpQb#wHUlVRL8=}r5#Z2MTF;^+mUY(XC%99gjZZG57`vddv;+7Ono5#8#_^)*$LD;K
zioY`yq|~h44TmN9<6d;wzB~4TYpt}LD3+C%m#5w~-;$)XQIXlQb?Yg*;_5eVE`!N`
zV9{%(rC?qyskE=Ux%IKIwfYN|+i84|#}1;P at vGU^(@sk492t7&&h4S4rA>yn{8Ai@
z3e{bKrnLJ1fJF1Aql1x)i)+Id?w?guB*igbARN`*-PpACSj?WOyQk-S*r%KV%Lnr^
zwo%kFe+9jl=Q^xpxJ2V$y^>15bRJF}jQ`lNW8iCf#l?|a_B;_Pc<fuYR6_8d{QNvh
z*qREEq6RkevZS0H>Bwbc^_n^fIXvOylF4hYJ#s-*H7TbfQGrG*oL2M(E^(O|tj7)y
zY!?Q{ql2LCRJYzCX(QI&-i~XvAbL#y9*+OOqNJO3w!QtcLAy-EqepKV8!4TK8mhoR
zJ9ms49F05eDmrh7t7~6g89+Y%7eK4sM~6*P*5fw+J8(BHNy+mzpQRxj6&UcjNU&hH
zOkSQK_T2B^kGp6tySY6&U~!t)PDD%$=ajHr7tZ4z1y@?N^X1Ex*v$)B^#e;l at HSGv
zO?&q2!3J5T=WVU6XTa}Rk9#UJMvcP56KzC+84QN0uAvGWBRGLf-Rw7vCSEtNsMw;!
zwkF20vb^+xCEMuYD3?8?@owkl<5<!2oE>TTQJ4Jatbswu`noaY#x5MB5)cSVPe1JC
z<&~U{u6ib22d`qf^V2DD;SG+|0 at e8oNO@{%>YgO6X0(;yZfv>6R?x3szhdXzW)hO`
zVH;qF<5P~&?Zev7_DM^#LW-OeC#vJ_OKfJoMR(8?5|R5}8&2H2h!wQn3N~z0LU8TA
zKhZOPy(1_sHZv>BUoq*Vq2XR!R?WtC7jh`!M6Pi;cJhDxcm_{bT>P=x8TpBa;o&qW
zyCh^$`wos!BH`C?GpDq)anyYl_<Q&7(*l>4%hu)O<yDQkKLM!N-EiGgPNlY)eR#x&
z(E}Gpz&k_pN8J?xQ)OaEJCTTU-?)uKMddzI at vmRMHWB#w`Qd<EnTn4eKaLk=&Q3Ob
zI5Gt8mgN5PE8hF|0ltA}5#tjUro*9ez at tL$lZFj($D?IVuh)x$IQ%Qi7rx>K?!R*#
zCU{RJNn1z)%o0~{1X42N<V4B98D<<pP-44I50C?=lmygOvza!+twaJ~p0{;%r`5Wq
zr>Ak&N>x?0Z(x8^)FCJ^kV8=L9tCx9=ga<nb6m~?fg5|4C8=l%o%M2!QCkAtdkI`|
z_ntkcWUPtm^}2`XUrDS7R$;7iuE|?04&24|N0i8ems(k0e`@lgJn_WVty at 3Z6yp9H
zc%fkT&!0a}=omM%wWUbWA}=)H6TM8fWy|F%JF)_sc5YnjbysR;ZQkF)f)fuXGcj>9
z1p{v-;O>Mohj3)^P=rc}#X2&n?_ZoA0JaBDiQh07CO3h)@|S at Qu7!dCgAzD7)Z`#1
zC&y7?lzt2kb_h}H{3llf*qm6v!pIo(_N_e1l|$;rJ}yA3#8(p&6O;fBN;v5>Ng$0^
zf#2h(edP@>VmbK<Wk{tWOq94z<niNRfB!82%m at Tr49mn60$FIZUx12=>d3*h;5Vj^
zgGCoUJo+}CTcLR2!T~J6sm4m|!iy&TUQQc7ZXO=&=n7RN3034&>){7?d-7xt{q`DH
z?P7a-`|n|DjVE$DEFfMjLVQ6>LH(1zuhm`1+$K74{2RN=ewHt=V!f4)j_$uxvYF^C
z^qBu*jbi+Fb_>aZt5Kmt`H at j8f@%Jl8$SrHmV)Dy=5M#bQ<yf#9fwo6FG<_Nx8Z5Q
zW03Q4cN{!t`c;U at iKV4X@OW_gC{;ytX at 6m@0VmZ$iSv8Cy=;dMA3k%4-l>i~Q7q&6
zb6SJ!q9*?zSy{ZhcaXY_f67*ub5mB{784u$eLYIauWCRh5wq{~2OfwUsHLT)@wEy!
zOZnjh5zF1+N!&mN_H*yy!!6mm6xq7i<BeBApFi)8i;Dwy+eSsDc=@s*s!X^|aJ$=;
zl3?i`1qT+V%y2rzqAn}~Ba<t at Tl8V<sYQ{SX6dCJ5cuv%GOYVu<}@6!1$<20N0XYG
z+I=OB1^Y^Y*M_P^i`9c=J%vDnAPxdCu5pfxj0DSK@>l!UMFc5CSzY~6RMEh|K*~@S
zpxV<9ifkQYC0y at I9--NUs1y!92w+ELPmj^0xIB?`WX=<i&MC?-;HNl!q at L=Rlw=WW
z^#kJA%aW3MmM3d(f=M~E9Xs9erd89(C=^19#TYRimH4|`g44+1S7fNS73pRwcza6|
zaD!!4RaI+&J>Wz at El=pQ<?!X>Zs=RL$nmy&3mMr5mRo6RM|SU61j>s%z-6>`ZA&zS
z7%q3-^@5;oD)=JW%*m6RsP>9inn}CV;oK)SM_M<>1>C-U8_sFhF-WS4VSCBpcVQyj
zyHnPF`$i}l2}2jVb?X)!tI~SN#_L1~ELCl7FVqLz&Yanqm6ZjFi-_xT$#6K@%#2+i
z?u2}Y(C!`Y-oMWfW|us6Y!49Gyuw1<@g^%Pn;dATp`igkq at HEnR42~>dqXX+k&%ly
zm)P7)t(~fleJ$r?_SvN3${H{s^s=u__4PNZOiw`68yB(9T$7=w0;`(*yd?pQ_UEf#
zR15;fcQlg}(QN~hl9IXsh9<*zAfv;@jSWwP at w`@653E^EiyBS_E_}Tvuu;f|r2LNM
zR_B^hT4|~Ir>t|2k6$Z-bBaKuj`x8(_8i&L>$7?sNb+EqJn at j-`pr`!mpg>9OPy|T
zP7r}nP3dt|l<d<p`$R-^Y`)GfEDUo9Bw7{OcZo@@EDn4epZn{{Io4-mV}o9roV<TO
z7ncgfi_<D9re<b0Dl01`=$>Eq@>;y^x0X0ba0if!iwGW5R1_2L5QC48Q-5efPa+;Z
zRHiGoO?9xgv^<g8R at K?bv^`)mWc}8`3143VaHRJvWpcmaPy{c$r$_BAAT987*YduO
zme$RyzlOfyB#rMTlBsq*ld9fclS%z`LSg>~FF09Q!vG1R7#Ls!C#6I&J8(qQd;<>`
zKa}X#Ha>@w5QrC1z7jyFpqSVrNe0cNe-X(5SBpaQO^F5oaP{gjIKZ9s^!r8agMi=x
zKX<A-V55g1141JzDniaTxTl^f=jkaS!W8OqT3uZc{#HX{<I^*Ta1N at k&(uFKP&QrX
zO+$m~<ma`|x<3g>;<u%r6?N|;c%HV~Jk`MqgB!RYrK;-Wq=!6F77z-=Z7xxJ+d=~X
zO9=%9^v|9>oBH)DQ`l8fTzvQZ{Jc}0gF9+Yz at eb34WiMIT|9l0ll<<yiDGr8f+%a~
zJlqtZo{IOeQ$JobJJE=*hayg}b+f^D;kX01Hk28+IhAl~GDpP3?1)eaSYBEf_<HSU
zgd|mcazX-Opf*$vB06?`BTsTt;wZJ>uqTdl7+z9SPq$%2!c0?cj3PPk$STDEX;2J<
z<w;tjL6O}aoPddu at iCq}Bu`Yuyc88>kjgFzC>)|%UQtn?R{BGx=YBVDLQ4L8Z*AC8
z{^%}9q7b{y+n%Wrd=JKXYzCiG(AR(5P#>&J1yQo at E?bad(hUepT3WRC at 87@td*%W-
zExswOfg*!!MeycQ*KscQA7ZC+JMiWVPb4Mf0D9o`Kpk+QG_}qO9~o?_LFEGuj?L(x
zF)=ZN|1hQ1Sv at _}C&ST9+uZU%_peuj#x+KY;31KV^m;F4cWWw>;wm!83Z at PYVK_qo
zW{oGs`Oouolh&6Xw-#W}_F7iQj?{H^CY)#lxE>Z32ASQ48IKBeDFNH#X>CzCeHusV
zAwpafVcLJ-0B%$DEJuZyZ98^6WRN2xBSTFMun?!Z_1`9jyVJ%ya+SavwY0Q=*W(p1
z-etwaoR)vr*chOj311Oo^Z@$^1U)^yw04`(H&-C2Uv_nkICf<mGV>H<rf+~0aZU78
z|7`)RO+kc=kYv(~cJ6;mj>uYOWswtj>NFtq>gHyRf9^P;2)`K56!@k^PpRwQ*xO7K
z8}biW1`-z+f5#GJ5L{(iiFdUq1j(s*r*f=4vxFh0bF0|e>@lQFx4E&ims(LR48qmD
z_;_l7PTj5 at Mu36+{QU5;#oaPd*LwcP7nCA{?->k}l at HZ1G`s^|1HtLVix;@d1qg2Y
zldx>vOk8S;D#jM;KVNbeGxKe<uU|urxYi2Z-%jvEcqt`GG at 3k45Fg>hM>jh&g_0gr
zk0B6VzI>@Xcc-)Tby-<i!>#XdNO>1NFkl}o&NJY~WQg0dqisrnQ|;9uW1y0mhV<_(
zI8z2_mWGv;|6l5JI}89H7(K4i!*AOoY*qR0o${nUQ4y!fmF}`)13U*OCjdjfeF6g1
z54hCcRQZwRTh`O6#!DMbD$A<{{(}ZXF*4kz)R+JD0{9r)e9kF1Ds>^Rt*xD!o(|`-
z5Aydvk)j1b*I at FUJl_x;wV<eIwsX}@h-8*lR#-_q!o`_*zP`CRl1t-Kp#d1$x54 at c
zJ7p`sd?{%nrb8(6 at bGxI0T?9GAkum*O%$WN3q(Jlhj<^1%NhH5Xx8?9H7Sf>PyK at j
zBz{%CIYwol(yykqk5OqZ?36X_Dzf*0zl;s{sd1j$;r#s&?X(1U?uBoD{CGDKJO~UP
z3q9`erf`%EUdh^@LR+wX3P%37Wj|k26dHu$ROa%Zxd)OAg*3Pyu(hpid`EnJQ&TMt
zlY9$en_2kdW?3CbiG=}FL!BG%07{1F<kY{g-x<)|ZJ2H*DI`P#I7#b8MtpocKzLeO
zlofj%9gR`!9JbIu7Bt&@=gu7nvtfX~2Hpfv;M9kX4m~Q7jE!w)WMrKE@#O(U>Mcv4
zB;gSeTLDrMi8^qYCjs9=I3 at Ib`?d*SQ;ODMK2<k#HqrH<yiCO+;ScD-?#uJ7R#sN1
zH4A2f*~eXk!{$32AZfPldF!!xR0jNK)I}_N_wLZpXm%3{O$Ts+IUku5#AIBniPD9A
z<KsN;B8g&`R#%oizDMZYi;4;f4&L_aSCiyrC#U=W&}mR&yUjEJwE<uv9y9d^(aM0d
z%Yiw*XOZ&QNi8+yTi|W;d*(aV;Cw3t{*3hX?Q at nb)eZ;<z>`IoeW~r4Sxcfq-t*`1
zb2T4WtdHl>6rKm?NfWj|pVZRVS!j!UFSOF-FJ4R&ah2S+&sVP}$&LN!QO0u5xr`T$
zpFitMRLprE-$6=3=mRWJ3pRLOic?cbYpr(hP@*j#YZK?r;T_K6 at 82g=wAxFKNl3ik
z0BUJ1+bJlfCyIL!&H}oV)z{w*=P2mVEspC60P}YSKYxA%%b!9*nl{}{4GsRrxk2US
z*JE#WLHwJVpAQ7a+nvw6Uq*)QL!<!4P!$yw`~j%(hXh24)8+w~8S*7;()jl68#*`M
ze*JT9s);%_pSM5{kT+r<ltB3`?HU?W$kuUl6T_ZYY|%Qz&3!XJUkDHu;cq!rYh$C@
zkiScr|4uBLST)6vJ3Big$Dlqwdk;`z6y*l!O{*3J=-Sluh|B)q{{1Jwva+&vJbLs9
z%L7~vON)z at Yk*Te^iz@&HDukDXG at qC;k{^ozZG at H+xs=~PIkKuW)+|uK%ao7tEhdx
zapT4r?uc#Mw)J)FTYqz>aTz)s0Jh!fpF8($G&R1KLf}>9;e2_Q^W5Cr_wL{40F)xa
z1SI1mV7gDA7`P);Y&vx_+ac^+I?H(BI6!v<T$mKq)zxv{jhL?e?Z*#Zw8~KkmTLl=
z-pihc=(nS!5S-?}qFLf9(t~)f8FCyFv9-0u7l6iQhul>$`^4F^yNVop`PCZ5L`4DU
zlsQxY8^jvty?c{*%n&`Xc^GB7mKS=5bzE!~KCcTo$g>y#^TS;*_wMZzbp+6ol##K$
zsHg}zYE)6XjVShDSz7LQw at ZwU62HPB=Fq*Z+}Pyj02db*&a4?<pKE7OS0qXU1llX(
z{i?~u4qW}_&CLYgva%)Dw*2E#QkLrLzl%5n{2<S&kqNunQq^PMTwmxxT~e$ftknoP
z6UDNC`vSk!YI8F;=g`vD#-=cM;zJQC>6Z}lt*or@{kc92ar^W03ly(c{@CJXPhQ>~
z0<RZ2a82y(d77SQ5EuL4NzrDzdwQk;#+m*2oWF*0nbyN+dlCd8)LO-{VK6VGtc$&?
zE4b{fBq>}~Jw_Zgo(Vp=W?8XOMr@&qehCQ)k3tmH65xPveH6Y66B84Ze*oG5DWC7=
zMjDb|O>7&z_vlgJ;-W|Tl`=Q(AU{76$PDl$D<?-oOZ%o)(#BOc+aTcGyLZeCLiYCd
zS0=lUeg5*L at bZzwqEz@~a0IRnxB)LK%YR)dV@$+?jFdDHklzNn;jlRi!NKk4*U;}D
zA_t0lz5TGRsS$Sv!rB_81s5leRS%k-bxFSjF=roOOjJS4%k!0edM~O-4nE4xogb4!
zyF^7raoXA3+&s^sjtcjXLCAY7;mQpxy2G-Lee7aLNJ!qr77nbUQ&CYpKI+JdiUR*H
z8)=fzGccHswx;0X*gK)2*m@(cpz!qC!X$)=J+J0IoyQ#JSkeorTj=7)1O`ak-|ot+
zoPeZ(^0e-~*BOPa8cZhwCFnPB4~^)F3l|OmZrjPmCZN{S^zI#2DN0>NIn+9H?7Jtw
zyh!a1Pbc}%)D&1;>|TBs2P1(5dO$omFTP82{Mq^i!@8K4Y-T%$sOYZY=6EuK%S7j{
zbsxvCL-gqBgPiNGrYPY&v;)PbJ?Cub>FG at ur@o=D%26MGd+M<~aS*sCu&Cr`_MwU-
zVCZ;5s;|FKQ1DE~1aMpk#G;<lSYU+@;up$%w7<XR_RH0kB_XpnTaZfvaI}Bsu1H*o
zh>WyA5$TbnyV#!FVdBHmmZ}jH5fRbdvOa7)B8I|}znW$0B2Bz;f2cXW6(I{!0f&xF
z3W|z2!gVw at c^Q&JKcshuDjgTenP0!AYndHQ0SMqwG1zwP?O$3!Z`bSO9En{zy1I})
zZ~OaixisbmkpV0Si2>i(b}FhsgY3Y}%uGB}+z<l!BNrvJ!vGVwjDJ!K)RE7w#(nmX
zIR^(VQR^BG6b}94s0;2Dh>vHZ- at X;TbhvlHe*zL14BPPgoDmlZyRITi)MH#(Uc!AX
zV5BH60ih7x-Yx6dGew*3oa_l;dOA8fl|YIifR+l{Dl2dE^76_l3w{gvJ`Wf<;sVqi
znV+v+oF2eyCM@#cnM&*E=?z3XzOKTZIrs0=V&e&FJMP`PcQ8zG-80v`_k59ERcp#=
zEI0$k#Rf>6{Nv^29Sl?AOL+E-6bq%`Sd_8qdOxM{yGL4*&WrC6)zQ?XLCS<!%c#t)
z=J|yWf=+|VB^EUG{~)Nm8VdTL2$;EfqR5q{x3}m3I at Z)EAQBzkHIP<<`oUerj=9c-
zryzApEp$8U=bH!IsSh?#2XlvfioJZIcr}e<$!n|xQUV0PEu^HfPtQ~W#eraX<lsS$
z_2MT4Z=9?|5DL)(8&_y{?W!T?mM5BnvBJ+wDfI^CaR0#rO(P at 9^XJcRAtS>&IqsPQ
zL&c@@amTOq<IM#C|Bn?uJ|(@IHKS8I#!)w%S3i$%_3Bmh0XSN6YisNOTwWN<QD9<$
z+2 at Pjw6xIRbY1eP2S8lmxEr&8=)tp5Lk!QgC5iz&R8`%9qb%AIrQu^^HoPXU_o6_u
z|4y8a6SSz^u?8?hm39BIY8xvB1qCc`LnaVmq683)I!fEv`ygqG+I5IIE{wIGo_t57
z!mxxR;<KLR1pu6O<J4qoYAPIQR2)z%+?E8vxS&7;;(*HS3xKM!va(RIDMjmC>fy4}
zk^D|d>!+FMbBKp$O=-mIw-ZuK$T;AZb;xLe#<)vyjG4Mfupr38Hc|jchxqve!^3I(
zcVf{DFxGmEf2y0=lB&U0!bw+n3gA1A{!Uksu_1i{+YuRRj&Dw%UAGsOcL5ZPd2SO1
zNOgI5C~OLunx&U`<2=U>yG(YUoAe;6@*BSNO-R at W;mGjl9V_rncpDFeSe75`+{}E`
z;cd(@XWWzxK$iq}DC}in0Z>5X&f5U>%A_4RMt+J(;Za4ROq9R{%>;M%4G!v9ce-7^
zEGQ)<H5jI9e at 8b{;mnzPv+lUd7}9a{QO8F at yINu;eO6u(gN?4mLm-`*al&$9bv4QM
zfX$mWZCa0?FPA3EaPH3CW#fj=-$(sS{)-pY09bLS;@9$Py)z|v!Gv?^Uh`k9^NNZb
zR`TSD4!Bk0=ht$aW}RO8S-}aXV{B~f at prmz8gOKFZ!cSYeLVm<YfDS?%2-Kv-0~<Z
zD~lTDkVUz)6Qk2|a%%B%A9o8ta7P0ayf*)Xboso-1_#%(v*XpxH4cSnESuVf^`6Io
zV{l6|1qDUPfF|8(H8llb1=z*xS4F at bDRme@I*hNW4Y_O~;O=#+#;5>@dDsbsFin49
z{~yMGS*nosz<~qn*Hd)B^I~K9_$L1h7uc(RPdG2Md&gxL7x`1CZWX8{wGc0LbJ;Y8
z+`c`oTANcYz{|UZfPvMfHYfJ}#PBc~_~w%b>pxUeyVPTv10e!x9s<8w{OPlL8~1jA
z8Tp5YYnC at Z5xGPQDI0H;v78RCf-fd2xbr^byY$0LeN}$IFDrdFZKI~HLsiG~-qeQ$
z1e741#@0?))P+@<3Zkl<O=nK*%O0TF44&5V+1e0+ at RuswX7~od2NhOvhIb84gs!J0
zgK)@sc0HxLgoK2U1O1qo5V9d2!kU;z-wm?0%SCw`4xg3Z2T_rs+?bkxTa+P7g#ku6
z|J9V`@y^YX?vrnW*t~<~f~aukseZvln$4Rx52D%r<DzN1a6CdcGvb;3K`t&1 at C<+{
zaF`F7ifyBdf0mBtF!QUYgbhRLs|8YsNfh21BLI#FN!k|IjEoFO{+g-mlE;s;tW|da
z*exzCaqio9qV#Ztiukc(^)WFUk91!>wHo}=ALXV|D=#vV5oMRH<F1m<&(F7d%Wgv*
zt*u+P;x=sXcGRK(bd&F(C{OSOZiH+RD=(-|F!=p at RZY!SfbV$nAD_y78yWFXsylS}
zaJbDMynTSwpsr5kzY}tbQtYz{$Ux$fl1E$0IBm*tvjz)GD4hP&w6tw_i7#r)U}$9Y
z{a^aZHuUzE#))`T at p~7}IRbf_)#vXOtTAI{Tvb(dFihEf1RL-n^RWBl$hWcAv;wa|
zf%2q23%vD86qbTRLv>T^2tegpY}=Z=DF5HRW^m_YaCU&MMOP(-h3P!YS;NA at fDvT0
zy#&ba?!3u{tk7fH9ChUgYpb#sdVMMyZl!T%=Hw^>sz5#w7soGZH9R<YB6qDL;q~ja
zc~dS+vMpN at Ap86mado)>MWA-nr(6 at PYpKSF(;yN8H~~SQXyY2Y913{S?z1571~V#k
z2|p}KQ~w>3EDF8PPjm&ztp24-Pfy1R#KnuaN;&Xbb&y<&mgqH>ayUk&P(_>0u)BBb
zz<V|kCMPFRNf*BH&6_t6N5&0(Z>ANtjL%WV>7)7&xSMJH{R5C3v3qv9(>C|%`C^C0
znEUId(wQV|-GovzpI=_#gwKb&th^R4<ED9B|6W*F*iADk#s~ZjyG{IeV%`Kt`W{9!
z19V~mxC}`Xj<TR|L+H`3CrnKfu->!fzi@*0p~|ODrMJC|jf<lK{{h2=l+7tFu5Z&l
zKQnW4jh3h1faDH*{EU{?&5)4oDAWic5+Z%(3ljx}t=v33DHUuo-g}*g8g4{H?85eA
z`{Lr_pj)@dfi>b9Gf7EFU>u?zCwH|p>YP1`dWHbzlr=Pzbafe8TU${&<<d9fc<Jkq
zDos+^yFPrlJ-lzDZ%qY|5b^~`lY at hUeXo}$QM3tf1jfe3B$lF{Jb4OC1rLdonw3|(
zzso{!_ZdS&l!iuWm4X6#iCFc{6~_w~E_9Up5FmL5Jb7|1)*csLzIZ`G`26{EC0GrZ
zouIJrgCe>@QTR at Amyw<L!mlsQvkM8)$Zi8fWIokbg>sXV<qKbRuY{SDIBg{-C&wEY
zFE70IM`E>`nYp^IPRXX5ijp!wCFLfBgYNKA23$G$d*%WzgHKN0;#qFo=^(h9kf472
zD}WgQEa+Sk?voI+lTrN*FoU4&r>y8n_tDk^2syAQ;_51zn3yQ&Jfx}~wqej*xI7Q1
zU{d3~Wg+)8G&WB6`BC?WE3sAu({L_IUSt+=9n-<p at EY;bcuksPT+V%5S4YRcQ|z73
zr!spxJ5vh_ENne>9&XCZ&ksLv?Ys`yCYTf{o6lb%6thIt;JFTyx*qplKCB=F{KKCU
zIuOGF>^*I{m|Qynpa-xwyl3`9hm_#}wr{_A_b&BXnCO|r=9S^5SUF>3)*Qo<o4DN=
zZ1ttnccAIplVu{ore40(H*o2qMIkE-3lda^?(BS1Ur!ExcoK{)MFtLID}eaRuC4)@
znLMc)uBN+o?E+eRVtv(oGWi!E`m(^=8X6k<Mn}m}qj-I1C5lB(Por9P_w;(yNn?S0
ziT6~m7yBtJt*lT}d~VX(+<dcNl`l3DIaVyY-4=Z_I_HT}HH`d*l_{qqQ8Nm+myES%
zx~}|fUV_Ml($f$i($do}g~!b07Zg0?pRop$S5i`P86&RU;x079BisPS3_RsyiL><=
zC+)iX@$q*7s8IMU&oP=?SilJ;w2h)lIDiv22yrqizv5R5o)}me8^;mzuFsBW?&soy
zBicPQG;}6ir?mis`QqZ@)XgkD<<vIxD1f4J4i5aN`i(u45K;~uI&{L!>?!f#^)etP
z8|zvU^>U3-S02#V<qjivlobP>tv2hxq_ORWp2_uHSp4y=5cqtqH-H4A2Nu)87;j`}
z^I0`KzLA!8Xr=7Lx&btz;5fuYVet7I$FI%fZ7V7p<(XCQWM!o)G-#9vX at KB?yKH{^
zu#1tt&IAr2er=xD$=Nv^jsU1;Vsi3e*j)=Vvm3z8 at tzKM>~#0^q;wg%_iUAssi5k8
z7aJQJn)~|CcKjH|xplZw<B6~}>aVO72LBb9h&M2}95|q0)2$Av&%l5wBQp~Pnge7v
z6&g at RsBA4X2=(<Ps1IG-!j8M=?t%Nn#O%e-1;hhC7qHqZ^FPi3FH~MSudh$M<h^1o
zT#%ieEpoE@{rl~ZqFuR6NN~C3*xrhw`akg~Qx1SonOl&89d{cMT)CW0%G at ZyG&eR<
zs+>NZLOOHv=1qyl!Lw)2wvL}acdn|ZhXo~rqNAC+ySq`?*~8Nl<)`2f5puoodX;_w
z0<B8uXAG35Ps at Rg!<(f9)adN&jQ86e9gF~sH{^0(m6zv6H?3cv$rEF+%7H{n_XBDJ
z;H_w5^id>Y?k)}3I at a(+_+4Y80$2-vC+Z;r?8BZ(+^z-RVC4RnS at c-PSNx={t+c at 2
zv4WhGl%%tiUsUvn-**UrhS5aLn>U+3efrdKVR`srxItlC$tGn+idUX<+`wDlsmDcb
z>*?qicAQyhmVqP5ZnBMzj>cs#@Z=|o9eVou`)^M6lqCU<f_ENC^wKxcO=O1jdJ^6o
z6`S;(&!6>Cp(|A*J}^D~@MaPcA=63%00;uMFTf8N4C~tOzY|5*Ao|iUG6pR!dRzoj
zfA(zE*RQO=Zp_Tgwoy>XK?a6X9t`W|NBdxB--KSbu_wKuK^3Cb<HwJ&!=n7!!lS4(
zTiYeT|A7x4&`G$CVGHi)s{YEc<7I9cZuaG`v$A%e>VANc!1^XKG#->Rgp{AMdLBKC
ze*4ykDhD9vBf)i<=q&sq!wP_MV|mIBQj$A&?x2J<1Z{&(rOb9z>B0&lmJi?a<f1Cm
zU+E1>!IuNv7Dz at FYN6rQcCb@y1qM!W)2uqs?A<+fzL_N#C0#8oj`rt3fKv*s3vH>(
zT*q-%6TmcDT24+54)Q{R*#SHRs~UJG>*wcJFyHd^t1*-OY%@j7ryM_CUfwKc9tau^
z4mmp2&CSh0>ZyK!K|L2HDSdo=0Mg23>!QNn#OoX@!QE3`jumU&0q_!0NdjOhs;3Ur
z%mq*G{0HwO#ADTifG1%8GZz<@#sCaLf<=R>sF1no=a*97rIJEhR8-U$m7bg|ZzJyz
ziikQcE(~ap<%J8U)V{!<w5+TxQwy8FbY=>amXthIy32|U+s$#uQw|dwm#=JDiF32N
zaG^ap_Z(!ezkj^|8?(#Em~fj8egA%n;=V0&6*zH9puL8M#^RqpvgtZAn+e%xu3Qnq
zRwM+F77(@3F?P!0zAeZYTEaz;zi@@i=;#;o2X*VHEZAWPuSh5aUVTicZ?m$p%FD~E
z>@e{OJs9WFAGm}YZzLs|QKWx*pe|*Ub=4tCSV#yJyZ~IcKnPAh_PbZ at eN$6YcIl++
z!5yT2M8m~qt~vO~K)RK?&>>O8>?-bzDl01^zU%>Zj}G6}#SDn at 6~UA&GBPp<8 at NvX
z24D|p|AwC*2~KveEH7DnOgK4$dl{m<liEh_#l#q<UO`+<>bHrXpC8gj=aP at f)Po#7
zBco6fGOE72GAor&<$`eC5a6ZRv380ZH*RQI6lkTth49<n&zL6VKDielT5{ktwvnc$
zCIR+?(`l_CkFp5A3ZMl0TkX4wPOa850r+WdrUnoGT<o~kZI2_<@B=+23OfX?8g_pQ
z-N?S}%4%vgKPy(l(4>Q?SZ*5U1{q}6l2A+e;=X{s${Q>hMZ~>Wjp!+JKh3DuuwId!
zEFI_H81Ua7u-V(&8}E)%QogI1<d&9})^{*6?I0Z-8sY#VcInb3T<Gb56VR@%*!?aD
z(L-wXquiN8^lF}7C0IqC8*4{RFU*-JDF0`6)MN2m&@rR9v|Imb8o at c*+S-1Qz1BMc
zPAw=XC;<Ks1cWPMUp5Z_F26pE4MHp=793Ak`%_wsdVa{i)YTQ8oQ?nwz84X(72q{4
zzFS-*UXn#6cL1>f=`9=iicy0Sjpi;}Zrj^9$aO$_A3u45Z+zo^X!~~Aix<;=#QzNI
z1MJai`D1q0#nRFei#SL)SK`8qyL^$?v$3%qqTlY};i0U at abTlL61uX~yNYt;ls2s^
z2f>k1+Yki-;Y^W12 at 9v(j%+6qaY7j)U{0HAH+VBn!IUbSoId)a at WO!u2OwY6q-&=p
zKU?VRLj8#8eg%S$3E?ArM%4IU*Pr(C@&XJ~ch_Cg<@@_x>Fj0mE8Qi|cv|o;W#!~j
zPJMxQW_V&z0Dj`fiy~`{&&AC#$4CfMv$GG^FADbV?H*)Yu)7Vwa0)U#UN=+&>po*_
zOckLLaMF*|3~U?~nAh0qRXIfKPJ`PpVN?fE4*(5L%gB)JeSICrS?}x?!upZB%N at XK
zQuCiJfIB;uufG1XxQK at biKG_G1@Q8x`k6M;XGX-MNB!s+xS-ClR~lkaRYSuLNG4cb
zztVes0i{;5V!IJ*!ms5Ot*|Yesl4^!UoU_GPzqG#c6C4wDLKXmHh40r1=V?jLqltT
zG9vbY1dXDnI1`RL3;X)^xf4l+ckjST2lAauZCsC|qb(pC0JB@`Oq>qxWckP4H2nZt
zszenTUGw?&W-GxrC at 5&n9BXP)TIxQI&$Vto4G<KkK$?#)7~bZ62|xK=L&IC(f+;)_
zRu?Xu+_nj)A0kvPnPblmn0|GQVyLCy?#nZSr!KBFK`Cp?05jMMOc|^e$0~M<U*3me
z?qFr!cO2Uyc=YathqH at Xbck8SOS)5`OghQ`mSLcKF5t5G)yVQ|cRx(8tP)mAGtZ|S
za=EN55$}30-QWCydA{1)+6J at WMk^~Zx3{WRR)J4m2spiay<hi!j3mv0Wd9J$kTau~
z+5Rl&D=j!)uaR}{Fe;C`Tetl9=XH~gxvNHB+!w05hOd`5zO0H^_t&Txc-8BIB at l@`
zO!y<^_#->YuHX_i$TA^TeNiUo4GhE<iBz7YWYnxEd90UXc+$%1S$_v+nv9H_j~+3A
zZ}trhk>Ya5($Y78S|Gs4Dy4dI^YY5u*ktsxKgFy+H{P*{09hc!svq7bBGN{;&E?OR
zfiH<W58dgHh>5ua_~xXX90^dDkW*ddo^z1%<ZeIqGBh?03l83f^{pUkwm|rmfB*hP
z**<I~R8&-yRq`?e6p5;8iPKUC50XJ7toxjAfl}QGkieT=%OMLQey^{uzo(L-;RSh~
z3OopGH~~ODr7QEgLCc4Z0hPs8)i4&xK%962!t#+w1e;>?x_%uu!9c(uW=Iz6w70jT
zq%U?NPfcCMW_9caUteQ}TY^QRCH{Q>5Q!U<fK(2=y+xOnlcRXz1Oc9jX7}#2+4^yK
z7DYgieSLjXhQpA=DDhI`5#RZHk&$EoN$@7dqV|s2a*Xn=f4E&<;5&T4!e{pJGCr&j
zm<Yb`K|e(mO^l6gcXs||Z7cxs95N<>5U#uhzO1dSaW^o847;hn at W!xv9a}k2&>hmp
z$F??XZRxKeZ+^efS49H20|$O^VT!VHCB%r9*VnJ%x(di5kOTcwQc^x at o>lkuer=vK
zgDekX7cK_^yu6iAaH&nP#n;~2n(W}ggP*&)lFk=9*!gboZ!5euLG5P|QBgBHyB+rS
z_GoxuDM-E4)u~Ww_s*RiQXW$T8yg$MJOBhd=f<|8S-pIjpq;8N+q?a^pAL5oE+&nN
z%D6bCrlzJ~X2wn=5>0`+I66AkkPs#&CIZ5``|H0uSz23n{8?7f&``p*-~<5cmz|xp
zMJ`m#yyrpTFmcyk at b(G~)Vfgd;@r7CE-o%;$KdNoF{-Oc;GOJ}aPUJNYwwlc*8Twj
z-vFv$VFP=pD=RCl%IYQHyt<0)G45Xe`StjYz2c~Ho7GV;4#vE;Y6+}>JzP1?Ge&to
z?p4dm%3_j`Ua#1M?*nQ at PuX?l$`xF3GyW+X(V;?{Tdr}r!Q1M|wad<)QdjJ_aQOO)
z9L at qSsyLOx{PBu!`e9B%0X at oy0~Vs8qa&fDq(og)AhnKDWF4pW)8hIkY=awWj1CH*
zU|bnL2I&NM<YLUkT}TibwWJDg9XVyx-Qcm>jXOyWo68G7dr>{eR;>r at DbSD~cbR<~
z8q&2LD+&z_HFx at nJKQd{CL6Sciin8pdr*%}C>TX+Yik8!vH|=H2?>RCZ0rlbeL5th
zJA5IP1Hl$QrEls~vNuw|L=g)@{r$L^6tEn2aTInKS0|kG6TIAaB1I;ACZ1K=YuoYT
z<%;Kj#v|B;oDXIKiH4n%Q%>mA7({7;FTBUYT;^R}UDfsV<jBE9MR)BN-Z<l(K(MaN
zj%db6xNgU?hl<K}ZXO=UQ8&JpU#kWti5rMZyBh|Fhk3^Ck&}~$kP(=enBIK&Kym%g
z><!4rKsTP|<nX(6Ta5#71gwO at IJi}?pg;t-(xY2^{OB7S%YvO&KYnCAzUcr=%H;=x
z26qH5W7&W=*aSM-df=Vq5o|Zd_y%!heJw=J!zjB;mk#4nwwW{aIIi{O`30<{0 at J`g
zG4qDULKh)Z`{kx&?58D9OiaXuu(;d|021C;ugPZ~>%sU15JN#-y-Aq4U{k``OkG^u
zted6(G;47q$lWaUUY5WB_wuDUF6k<E><xr~*nct{RhDLEXzfg%967LmziCa#F0_ at 8
z5CpIQJWw0z2e1BP%NA`z!#lvgaj!0l#$z)j>LcwIwIQC?plNPus(lyEc?Gg9JgT^)
zBqtwVc9R7vi9<9*pbO>}V91Yk__Zz&Kq*;%bfQQ8%h*YW`1sDMo=$&KY9xH)q at Erl
zm>`gtIxwayi&JL6O^;up%-7!m0RXQ;*}SGsmd7ie)6%;6?wvADWCe$W+*IX~i<iDm
z-}oAy&F;PK6)-r|e*#PauHpv-I4v_1_iX`Wd~=Uo?&C8x=9DZWK>N6TZ;Y6Z9tq+3
z;d82~1;tN1dO3%T!RY`BR)#Vk-4~(4PJ0UrMOb?S4h>cpVp7piO|e0KqB6$DY_Mxr
zLV^jmZK$gH8}bK1OhfT42vxW$Gu<cdU^{2z2b6O_>jc}wZGBG8&KP$wl&mjUlY`G4
z5Z1Xr;^D(XO!>&Bu at KRkavEYMfiN;YUft3{gZqkHT?3+{nIRu1q@*0U?39sv|2}Pv
z95vY8y at wC|CEX`aY-Cr;cetydu<-im&=}Y<kJw+JilXn1#{f3LF$h3aINpn0w1BJm
zR(@{A at +Phcf+&DneQ`^n``4>V)6OMV21s$RqhjUHxCwK?YWmBUyZ`+8^FCxFV<jH}
z-bhcqJ at oEfe1A`Y6p at jU)%eNcpqRCF>O-<t(LgN20Y_G_v*VRIejHbAA{>p1ijq~*
zJ`Cvy#e*Rt<D@!<G_*v0Uhhd;U{jRZ&s1;7b7&GR<vD0j7d6eIL&OR|9`;3pXJVb<
ztM`)q7BaHpv>;mAbf$l17H*2I-f?lpwqv*#{EUvyY{M^1CYLVVg8%=fqocUV0-O!P
zVV>)_?%TI-2?X4ZLr?EtS$V>N!KR;8%ENbgX~6=>HEKO8 at 4pE|9i@;UH!yO$eER<5
z%ZqK~<W*l at SZ^ZW;^aKDYU8meutX#k at 87?{IP&Dlt%e3woMl=Yw1v!wCiUqP1DF)5
z6o-aV{`m1DAv=4yb>Uy#>S07=BrcO&UH<8V9)Zh;hK7c4`5-`$>9NdQCl#^d>;Q#O
zNy}wqXBg}85WpLW;Cb7&ZL6I6R*j3}pK at lzK-dL-M9eU`04xdq!kTy;ydm*p$EvET
zHUX3hN=o8X1^cB`R|$`f&b-(O7NocyMW5#8Qfp{vARCU0i<_F7@|dK2dh+B+3>*7s
zZ{e&2Iw&bAsldZu5Snm<Q{J@&2S8DML2TY0W4)XOmbj0}z4v+$-}xfcisI4D^3Bf|
z>hA77bM|aNs{LUi#MaVMiHyulcJR8$51R;>(xjJX+iG8JT~AjW5toOXSa2(cv^49f
zQ>U;%`ToNPbR#SXeE4u$Q}WU~_Wig`9=ul&aFr)@+3%yHqoX+GNV6^!|E=ujYYU?k
zR at _CPaGSvl%nYv^82AkQHIo$E$GgFu{Qdn+;Z5$mC0}TBWNd6~5s+U?y7obM)2X@^
z^T9Nnc=aTm at JLzVKVGa8sLF+k(Qsg^bNY~^l~c6Pz5$X1q^BSDtw@#1I$ul`EoLtZ
za0UxjeSM~=(T(dIpF7^zGBwh|DX*xA!-^Q$!oz8>lyK?+7qG1#0XgS%bo`Gw51F<-
zQ#+%l7o@`- at a!2Ew%Rtt%TyS5V0Bej_Y>WNd|_c>yig|~&nz!5Pux>Bbiu}^v;Sb}
z&h=!ujCajQscbv=acBxqHS$NiZbqttGN8dkfB5*=-v^Nm{>n?Q-$xo68&SXlFWevu
zsWhX&Ua&mJ+B#s{<G1BGs^HM+sp%Ttzn4{1B*lA;{(f`930z4?y%Un2KbRPVEmJG2
zS|(izu5jhclikO#4R#%kzI2;lB2?Dal0wu9c?J-3*jie at t+kc5BiDqC5L{!9SW at Qi
z6v;EyCt5#!>cCX;uM8-n>Ng-g9F?4^Wp)(#a_em90ZvX59A>^jLN;dC0P#X}$BrGQ
zfKVVIzWMxlH_!<|5fT3zH#Xrvx3t_sM at NTBxxSM7)-6qoPgL|lR8UV1oPdA;HJHF<
zC#Rdm#l?wM=Jxi>ZwhatG#r37Tm=&)YAa}3>C4z;sH+=@`@SKo1GgpfUi=mmS!#Sb
zB0Bo^Rtg4ub9m0W*#=vgnWIx#edd3hLklT(=((xGeZAk=)zwv8LINAtrsn5$MXWkh
zKXGxiSN(%)T6 at 5fFT1&UKXpE+7Uds()Uh929E}%64!}^RPu<{aXpBCJ;#Igtee2e(
z7zY?lyf!&H{`8C$h)}_95A1AH0u$Izd#hg<_=Aj$46v6D3G9B{vu97CSRODq4tQ*!
zKRa*NV27EXpP#uzk)I!lot at oBYy>a&*=Ud*8tN`_UYscI&9iQ1MIks~x)4&ZR|7IX
z+0y|{VFXctve&jd0m_CrhbR)kMqpqdWs$n8t0)z#RPfJ=RaP*A#i3}&DfoF}pXcHg
zIl?hop<*Jy4WI<ODw^u+HHnp~YHGV0KLaX4_3-R$zEZb|jM+OxDS8G5Zl<CK0KtYD
z_DFdx?(FRB1WchP;&9E=lbb6CFV-NPXozgfOi#ziSFyUXR3Y7vke0R$!o^8QO<1SH
zE at 1eZDrL<>id+xA;Yt;K{oMwIHUYqg%=;>DfWe)})Xnw}Q`(5tgWrdTaTmiznQJ3(
zrlumt!#yy-gi&PxfZmvEBbGW29M}q0b+6tMPYMe=Jb^vL4zMaAw;%fW$!|08dFBKS
zY4*_4(cv0p!xE>hzlNLZAO_ir4gg8OTh2Gb at jg`H$~xA5ZlPdT0GE@#t{U$Q7yOS)
za&PPy=CC|+E;hbHwKMID$RPkV0EDDGXSWghMn=|Fj`)#L2bPw44)}9L?0-;yd8+Ru
zuBE}%YPb$1w_=59`On<7v14_2+3J7<A}M_LeMaJLOIurY%rWPCkl+xqTUqr3_6liU
zIY!w9*$v-NL4nuM!`&MpJR|%0bJ}m;zUdd*?1BeF at y~@`A2v?KJGdtC<8;~+05E%T
zRVZH30Kb5jge%g3B+k9*c&NM|7(W at tYRINFwe|Pz?aVSRFFolb1C+9)qy&dxFbb)C
zMw at ADJZm>AEGXz35wQyx8J1&gZEY2Vi^9Ux{VdrtAkBn_hihhDSpt{^uL4~O*{Fqu
zMb^0&&zgRTu5q3zAlCzRY(DZpEaJq6{{H at j?<^rFCoC^7FY?c-&1s;<$8lg3_f6vh
zN?KYDm;D<KVufQ7NIt<Pn&OV%e)y0cQt=7yh(z#-!l<J4y<fWHUQbj>Sy)(fbalDE
z(%r5(qpe-7XL3E at +zYFdr%#VR*@NNEvYyuK_xDJT<+%>juzqS$7xw;T55p<$2y^Uc
zo%tZ}ZFm?{EG~h6={!vJ_3ALoUP-qf>DTC2J0KcjJ>|ebuEfay7gTVH6nu{=Uq9>O
zTU|Xp6mT`QwWVNYX09QXCFmE~?ZHmPO(di^4H0<v?%jLSrxaL#Q=mEGT1?YQ!rwpR
zxtA|rKI7$$(yuf12k|l)U{>6pd%jFqyDu+4{$paoR at CB!@`h#3Iss=x;V$e)q-K>u
zyn at ZkvlCr<BNvIAu}`-yj0JayX=!PFd||DDav}Ka<JUu<KR<#I?&;H~_?UnUvqRl+
zz4`0wOJlCtkaq?L2lpL4$_O^`60ob2lM{Vz6dCzu1RcD5d}j9c*=ebtUd_(Ujjzh>
z)4B;sAKPl#**77uxc)~37-`y~<c(6uJMaZLmQC2I;fxeLc<QP=76<_i=j90&IQG8I
zcS!nzLK)ODe|_;<7TXpDD*kHOwyr}^8k)m(K=#(k8J)N%_lU4C`-6H at Q`3Zt-EQvg
zK|njP&HPmu03g6%Q*g{2f`WNX77#-)<?a=~On at iF5s%-)J_k;zs>)x!EC{EI!(O;p
z4fuX at R=_gKgA029T}+i8yceuP^6$<``NEpfO!aNDv$x+BP1 at MdV57#)FCb8Nb1TJE
zn}!T_O+l1ECbb2%@*xUM*B^|>=C{YkuZdYT$Aya8e(D3@<fp@(t at llPKdsr%ujRYN
z?DRSJb8XZT_5g9mX3Djp5+0sv_*~d2NR}A>9y>{aDq{l)uyKo)M1_%#++OXw0^Hnh
z-be7{d;NBL_~?=J+*4lw7RqXBTc=jlKCgPUtJbdca{W+`m8c>TQJ<>N7&0!duPi7m
zWS%EGkt`1}yQ(Qxa&_+8lgi%cJAAynD5jvipB5LIASl<6S>EF6I;W`I<%Fk*(F9^4
zbK&~Q1aHwYF!=NK?L<}<FTv-h#WjQD3TkSU0NuU4y&*(Jm>;5=au;5Vc6 at z{NYwe!
zRg7(tp-uvP=T4s71SBRrGLo$CCsSl#AgON!+v9_g9{|0^U!^QDZ2Ge!eEYxoBO1{?
z_4V~QZ-&b|0w@^GI}0v>hto+gh_b_P!#(GyV*<fdX5lw5A`}(F6+H)%IX8~?D%D<b
z-#x-s36P}O+1VdUy7L2e at efnbNgW)t#9=k1u)S4%Resn$kG+l4R~;Pqu}2tKjx2Y?
z%6P2=gcod?7ZCV)i|a>J?RaTaeqlkuL$|x~p^fV}%V+Wz5G=%<z~Xq=cWuHe+OHl1
zlnpV*=z4ut%RgS(puXEGla%fXUTWQs<<*Rgu_tvQHRjl8HUHxaemG2)ReofgfW#oM
z1I2MnCi0y(Hf9B1#<@<s7G%3VOM_e0!CiH9bP`^@+FMgogU!+B&h44(t+)%vfkUeR
zV+g(wHgOuPPm(<`)$G4xU8~9gq`8Zo8nzud3Tq@<UJars12(<db-$$1N~ecR8<HAg
zbMfK{BO{ixXV0c(T5{0d!v2zxk&)VsUU_+97T_0Lowu=F3- at unf!98wFF)rr_>RVF
zwngcs>o|jMmj0%rM~`A>7(AVM{%+b^$;KU;5U?Q;#$Bwnq7y5m2T}^a-afWazi%~l
z`>Fl3 at 4h<`e8JNp=(VJrW<-iH)bQjYShPj0gJxn<60U8+jy?$8yCkSKBF82~4RB;}
zORuD_|H0ukEFR$Q`FnTbQZh2O0}n$rX(y*6v$M1QVe;~n|I`}flV@;cV9jmD0lbw4
zU)OQ+EK=UT**DdJ%Xa%7oo}p64-EP1US0i#+8#JMw>b3;fEM-<;x%x1IPSE8H$zyw
z)@IkJSovs1cq5H#1!6NEF-`N(;lovUae1`YOZ%ggsAxD7oVDPQBY>5U;)oewABu-R
zeiL2U0CokGlpK?ilG?d<?+!LLwv$Ih>kL#B6t-gB5;eyS3=9ql2#`%J$PrfUZ_dl$
ztpWTfEnQWP2)F*X<co`oQq#R5ZWEnV at 855yqN2)s{`_V at z*bD{7cK+{S=9EeEH9yK
z4;Wv2-;<~h?y_<Q1s7>>`m3No#ATxM?1=bAeT*;aKe`rwH3mq%K5w8Qf9}~_Ofo3H
zdC}JPs7KL|A+^IHc$cV4URWqPKiR|MvX{eV- at i_Ij)?ghMibNDGvBeD?53YzK-e#x
zj5DT|59V*<>O$0a#OXT<Mt-mUU%0((tUVLY7#Pyb%nXiXjf{^|+<Ua{RcV)QhlIQV
ze%UuLfI}=mR`f at dHADuNAL#RL$eN at Q3z-vHe`PJD(d^ubyFU~&Z43N8wG9pT)`l_%
ziP?9WjuF$bWQR%=zlPL+q}9;(r5^KHkuo$gVipn#AH$;Ko9b#aM at MQN9v*!IgE!#+
z at P@K~EtX$f=*O9BFn+9jrC8eV_mWelmzFX?OvI{CkwXu+%f<rIL<YW%yNDH52 at Zza
z$lV{*69^E(1cimQiRW?QD-h>9Y%;8CE8G|RNK!6nw|53+V8Ok+rw5hXcd$t4z>6Ug
z{)bv^fh^!2BZww`m6a!+4x~vj95XSLm)~;p=FL3kVRcjn{_kZrYQe+fZb!&m0|1Eu
z$(DW$CrF*s4zIVwVjjS`og4Ijae5&1xtpBvX#n;BRq@|KQbHwXj6o=t<}|2W9#zUL
zq3!%{uQ?7f0GJQ@`_~KW=?B$gPQbf({^G@^zdeKJ&Yn#XUjSI36LVT#<KTn(J^;1A
z?QzM8%K6k1uY)25w;2UcU<)^)KX~xqzaVOm1t{0z at 8Wdcix+b6#0X0~reXoEmb#42
zygh~-c6|kWGO;v|yF*N;F;U?*)~S+HQ{~_foWSh;|DD<>JV{<*VGU3+OyJvCQcszN
z|N4$d9zKDhl@%B6OZxBSS$zjwfAjWjim#sIe1M}3M#KxHvY*X#nD`*HgxG{TQ<z-s
zFBtNOitdt at miCw(p#i(0p{Jj1wS8^(Y$!P|Zzos`Zm4vd?Ed|B7 at ndUFU0>YUcoK6
zk^DwCRk?7kL(uHamQA#DqI}qaPE6On0+xxlNteb&B7bhVu#Jie?;kuoUVha4?{qRU
zxC?CBGrrdKmN=6Q4GqCR^o@^?r+tv_Z7&yr at 9TWA1Gcc^bPGI|=P4UmvJDEK at u9fa
zn}Upt3{^2d?u9HMvph at ZIsfApc*qG}h9u|VCTt|Iv&-ri4_BUjtIcKm_1fYTJG$)1
z=x9}CCE at fV-c3^rwT<vLN5;n9z(4sl)OaoN#5rf&*QNScaFeH}=f$>Xl-SA*h?6+Z
zT4divYGPsnPAv!dT~}8ZQ~4A?TZs<Ouh()jG7t%2?@OtxEfzF#a&mA36MMMwn4uhZ
zL_Z2C!J&Q3ILq_7tIMM8nc7?_+l42hwhREDajJNGl;1|IIQ9nsp_zq+pNY@%*^xtY
zx*hW2xK9n|Ero<`K)?v1V!<T$`0-<dy7ttZndb2hqUu<NP<wIFFGu(gLe@=t36)_i
zl5vZFui7dL802PX at Ezvor{W0T7AxT at d-dvXx(*zKymgBla6#DKV>5;_WEX@)MC2_k
z_pi^kt{jy9`SB?UHh*9E at Q9s{j}nr2a9G%oyg{O8C$NCcPg1rq{_6$Uy!l~g!ZL5;
zgyc5`tbLGCv+5W;kl4s}%VootK9c$#zczou#Dp!kd_L+v?eTQ!!ht`Z&iYo^@}Te^
zE}SzmGQzWa{rYwAojW>v+lcdmHkU53A3C(HtJu-;;asqCiTHnYIdx4<8bGx)tgP7A
zur2CB|Jz%iffQ)V<o>pBHPX0|byWb+w(iNDi`V4N!P|j!Yg1kUky-W0k?*%`Av(?8
zYK-OL;=;zmrTK~RhBHP+p^$F&@$zOh{Rj&SL**Yx_Ty3&(hJ7t&S9%<jQEwrCJSss
zf_#3Gm!ZSJ0^ZB#QrB_1I>64%z<52zqTVym(A>a&^iSF6L*jWhj-fnecGl%(X(<OE
zpV|Rtt<_>KjtBL~uyOiLQ&SVA0D+PjkAE)B$#?&0jEYSsmdCg?%5)ot1<rS+pN8y$
zEv|*OdV}A;9}e>gcQsBno|7uv at 960Gwyw@%v^C|isI6}HfcNs89*!^J5ZC<&58^#5
zWEaOWbMt&wy_26klm4~vICv}<Fc>K*DYleDf{Q5Cb^fh>Ca|?tYc~@U2QM$GJ3edu
z#J&+!0tX<U9Xxn2%dkXi)V5rC;$@EQ-xoO?;mT9M5j(!Tq7<1pen;spYfa at 1lJNL=
zTs)eRt0e!v`qV$%;sj_2uMprEpC%o8wtac3ni^@I<&izOcoIUk=WrSPe=IQqM&#l0
z(})qLv9Ynyc(qy(YJhvkwX|rl?*I5PYCv|Ex>5jt(lA1bxQ&7WDFl)cFg_egM42am
zE7<I*aGRl=SS|pB6k~ju+XRl~0xC>MPDU-mj~yL#@V0=k%pA<QJ~Ma<@8+<z_0XX+
z^rCuM`kHs+(`L48*>Yk(ExUw-!N<M-5>!1Paip*jb{|(#qek~IF)`tIl8z4jTXP`>
zUi}1du6_GZjvr}Svesc7*T$5jr#ArbR&TFfeCd0c=yY4ADI4R;4;^k}<_->2eLsWt
zfAVPAc(=*OJqdZ1_4f_5&qG$v_n0;(VDWF-aG`fMMm4;~#Jx%q+cjT=$Hs=j*;QkM
z%qKhYUh?MCdU}kH9z6mh*B+DD_}s+HP4O~!S&w^)j780*vvd^aJO{%5_U+pQwP-Pf
z$sr*jsM5E%=z)xj5H-oF2+;VQ<&jSY!BLyk|3gW_*XHf^%6J<dc-99Lqz{0~)WSkg
zY%B}*l?)7!@bF|B)E&3IKfk`DoRn~gm)Ac$yhXdWqrF|>)Tw&94sn-}Guh9kMn*1T
z%iGtltdRmHIl?&nuyOTeT<+QaJ{l~Fj*bp@(ZbJc`ZezS6G$Zr1g_<$Ee}xW5Tz$v
zUC-L7WS_&PjghY6pLAD&lKFvIw$Gh!xrmpYxE%x%0F4OQ#?+<rH~658jEq%mP4gnF
zlKFB#CISNkaUK>h2aa-bOG`QpZo^q5+%2S_pm0wmC8W8apa8qI`o7z!h#D6aX-C4j
z%PA?5b#``oK<+=C{flERV>tUj`OJGBywkurS4jPY#nr=|^YimBT}JP#sAq~DIr0X_
z?MB;DDOdv1&f$d*UIR^ilS@$#A&qKwzxf>3jzXOJoM+~z)$z|EBNs+mI3c02b8~NB
zSy at pr`FYi*ZZ11>J-|>RgOdFDN_)6e?JBiJw3z+W2Z0LHH~!?I&sk{>ll at mj_>MY;
zd;YJj>yD>-|NloulroA2rz8=jVI`4Bk&$fKqhw2=?2!{HgrW!`vyfRSyOebuSw;3W
zN+^_BzONUD-|vs>(I2;aa?bg at ->>m}t><goHe3-vR=fQD)8RYYgz08yXA8Qfr>Eb8
zd~e>+)?ocHQZf0Qj!yRUyt85d&;Ix1)>iE|#(4?9<7ulGrHtXRA;#5yrZ)LkI at KW%
zLPjmE7FRFNF++WQ4%8$;K~hM_+01?nlTuV-^kTehTJI|GkjdmUXAe&?)+09o<5!lK
zQ$5wrWc`0vGdOq?!p*j5uLQ+!`Vlx(e>!x>pSu1yaC;6mrM7md2S8qKVwFopB`hXp
zUVY?^Lh(0On=|nHI0uW_YH;CLJOwRy={YZkw3?3nRZ_o<NEhtvf*CgPrv3ZGAu+`g
zFwQbu(V4rk4WbOc+~$Sj{s95|oQah9$j$YIeWz?&mSNc4y98BL)#PlQ?)IOP&02ad
z?1LvJu7F87p`-KSAM==r9j8HDF}%*RsJBMfb+p-kBW0jqs`O59u;cWYVS+h<;Nj_s
z+kpU>AuSmWnM1_KOe6*aIBgr*zG(Y4t6_daa>}{A5!GDqC#rT=RILpg^B&e}jRuE6
zckk#$LHmqwfL~=8Rcu``O{XDdMlhCO{w9BRawrLqzOY3dPLb5q)~bQ4y>!yxWV6<i
z<a*BkrkZh=Ef8w6+KW?Hd3nNUk&V4(X_1|3wBrTt55{T3yUL50ItMe0e5!!yqJyK|
zGBXy6smPKgn~daSWT+&Zzdf09rhvWS0gx>j9#;GfS^4w<*KYzfJX1SPzcL7UeQi7u
zz?$D6S7Oxr)x71=uC6W|m;npqAF`%r-0V?kC=;ft3}3tSF3e66d at 2K&j_BQt;8Sg#
z>s}&HdJq#M{fkpxUY>Bm&8^5R`m|y8rP&wmFLH9I2n(}A at v4v9d&;N?KAD+zp7v^c
zHnX$i at T2G46pO?<ilF6T=htrekP`=@))NPkkr$5u(epYE4h~BzD?A%(!*3cI84qjJ
z+8R=JSsz=x7a56^t%S1S#w266*e&TgQVxP=;6XvnXb1~aO{%(Yt{*1!G^t9Hboc*_
zI>ufFDh{5&mx`K>9yI+kfH^>+Q+!&D&83 at +njvIhE8R)Cy(jIMXaNK#=B6^xZT-GM
zLnl3b%XoL`V)}?^i_{%jT3XPa;nvI;xPsw5Ijg8BBs&{t7ICgWAql%PajUg`XA$i~
z<(qwdX7JX86G^t$Gq6mLEYaLyT>ShDO{uDD2{d$c$A3MtYD%FZU>S39VMrP`kAv4j
zmwW#{$$6wT;o5kYsuaQ@;heoZJT$$%y;z|<e|{T;@5#ExTB+Mx04 at q*^F9bRDMXgy
zaBOj)5=;b6mimWGs55TZKt;f+A0~uAbFf1C<)x!4r+#;|dcElMXvg$hn&zm?(dK9F
zl#O3;ad8- at Aoh9zkxg=&YnWTh?dI^COglHaySpQ|)#Jwmu-r>a>)V*yL#6))swJIv
zoMTX<&VBy;1n4c!y94$RDH!*^Ng0c2ZJ;C#qwiua{6pC8;G)6OmXP?=EFqWgx&-gG
z(>R?zKRu2_E<r&-*C*?bP>FO(-NDL6Q`3#tuU{`+LQn?BRr#Y7fs)EJF5O}5`HKzV
z2gv^V$Rvv^rwoxz1oLgJj*AQCE{5Y$<MX29v~+abLPEK%KbNtn#)8K;3Gb>q+ps(x
z?;0xlesO&E*Q at E88bwA<@gG5{#;v#%JK<yiL=#U_Ef%`n*)lX8$GfD&j~+Gs+4s(w
z-GlN-!|wzE^W0_^tN{P3X}{)@rwwbj$K-~=hWN#|^GT<Jap{_kVN!``GT3x;h=No(
zHLij%k4oMyEtPip{)x)d^N-$m=?95}>e+5xdg{S;9fka2{Wy;a|9Za7ovi}Q`H%9o
zJ8^4s!=t at t<r*B|yW{3WWh3>-!HL+%jEL_e_n-H}{nI1I0w|`zA6I<iXgtW&DR5VS
zKs{|x(!&&Z=;`UNTPL|XJ3B{oR&L$8H4wzBS9)<a^kW-0_dp-dinTa{%%JJ%{8b^u
z^$(SKbwm99X|MtbK_zx|g at kOx$Pjy9etl~Rj>PTX$h6g7V6&D;BqG2D3o_G2_nU2>
zHkX(fyNrwsR*L|<<rft64i5){I_%=-|C&D3I at Bd9DvJFY*eCbTKmU+Fe)KCW6nhjK
z%d&mD-cX937EeSqw*P}U>hHYPy_)P(%M&)QV#K at o(Nq5N+ISq<(dYxtQaF_UVTUVc
zzWr`xx<XLDP}0%X(?$^|Ar%zpA*`Uw(Or1!Y?M-n6A7N4o=D7zc_6o`y@`p5ah7DS
zWuTZ?z++=$vs{+f(9u*;R%S%rv}C2IT$vaJ2HeL5XJhc{%BkVKgy^no6}92D1Bm^3
zl^>MRQO+}wmj~b~got>l+wJPTpAw~24NvW7$9@&*zcWdRDbJFUu=yDq3{EtCsk}bg
zE{bS**REZ%^1Bl0%2v^SE_Q_i#N>kTZKO5RCXB;-`$+Yjix)4FTUtnwk?i1x?iWR7
zgw+mU?jC;bhKE<E<Wnr5;shX|snL!tT3T9xw{G>tm&c at 4J>pTvPH_N2)JPxz4=2`M
z^oxz1{U(TSe^Ti2its&3{vjuuZG%CsgdQ at 2eUg>8a;&JRz(%~jxnCoov71)Yuu=h7
zz at D4>iUn*C3O*;!HApK=Fp0gL_ONGF%+9m3v*X|~7-Z}a0o%Whn0`Lv*|WEBLj(c<
zi#4fM2Fc0EOO#XY)ApR=mwjR?DOz|>K~<GWK7`Y^q~y5U)UYqO)AhulXrC`%4A4Ch
zd;o&bKT0Xg%@snDWh{OG&II=(;KY+?ft2>2O1-7Qi10bxV;=y=jBPJS<SJ%KsYHp&
zU|VT$@CII$=;k!0?k7cdokeV6VPPIUd}_iKtIpL596k1FVTpEhG#IR3!fw%HdU|@u
z>OwG?8g7K=kUV1l<pjjCj)QgE-oAT>CSO*b0)VW4gW<*vQFl0wYy`-1>hk47V8IVM
z?N?Xh4x_iQuz;AUtf%}2Hu^j9klwzP1A)acgA>=Ui-Xz8ejs#KuErE_7t($7zYiVi
zW)oorGyrjTSzR6DPOq7S)jZlTZJ-=C=K8S)l6dO2e at IhethDzIEN#L?i=Vrhwr#a&
zkS=-f;6Ze0``IfyJBb`(wmqAO9N;C92MwjA;rP7ru(%@51feH`@Q~cxj2U)GNlE{Z
zmF_Gv at b{*(69a)kK{6g5lAJ{1&ynm>T-iKq`NU5|L<Gm(p6S0h3u@#ObHu at H;@@gq
zR{tzB6PqTN@|U`Df>m&XJp%)zva%C^H`WnT&p*w|0=p=T%`r$MYnG!stU!_VDZlUE
zzhhS~QqF)*CXpTWp6t`f;yQGQ8R;wG*FdP{HYjM;8;(-Gr`)}$DD$x|uduuB&6_tk
zrB>e9s5Y$I?2ldD$Y%<{GLEv6)^Cy3*ViA`T}hE^J33-)>LY5?3XpMYx7WNN+D}o@
z<5%M}m5`A~P|qJ<{P2MuYr&s#%#TXNC=ej<8SXBXZf<U7=HOU=^5jVr9wcLdI6eoy
zJ^Ju6AF5ShNu+yj`&kHGH#z{!5hwKLZ%t1}_Gf0an=CB8t(p3u3xV627qufOZW#N^
zk&GG3i>|ImQDc~n&kz84vr24FfBEK(B!(<FS1QL{jufH*v2tW607$`Pw7sA*_hK#P
zRj_e6f>*`w#tmxZ;D9KQff!1E{``3?xndJB at t6yjh{$I6Kggx^`LPNQGc`3e2)zt%
z1dg75{`^^1-dbBgK;V$q!gZ8H_vx{VH^f%>dM2v*;!Or>?jwoRVBJINbWH)LCl&{V
zgjjLkW<bDNq at Jd;OSA}FDc&^y3ra?P=FA!095Y((-MdY#tb)+T966!`@@01A$`vG3
z#V{8vF2q8zva)Lo3=K&ud=+wwnEk<yGfcq(*u!efg9i_AsOzQ2Y%cj{8Y!&$!g5A+
zCy3Q$fWAmDi5Bko!jF^r)-8&|fpUUz2?@LU`ufO)cKS$uhkFJk)blQnv<l*07%&|;
zjuRgrPY$CfW126?-`|8%xu;{?>a(!m0Ro_;p1v7ZYht3Xr^f^vctNc;OR%%>Q>_G<
zX=l_mG%D}$Yv5!#Qq=(rex8$44)N&?+&e}LrO^bf1bfmu`l9<L*fDHlbNgkDeJ6jW
zy91)P*Y2V`e at 0DJ6`f7%v+_*+7tb&E%3?M;BqRh?P&>;Y2*MzGVpw!e-lYb!-`_u@
zW7T&kCc*Xhf>pw{&ANxe2~VHybCTnp!$KeKxdE?_{Rg;D3AAcG at oDusCS27bg*g!1
zklLbS%qGFQQuRLrMh^+{0Z*eh2Dpp6^(3_O!lI+sIXF1rMiqcM at 1jL*xY%!#z_ciG
zXAFAFP!40;-Q{){Bs2q;{OR*&)0%tyD1g}Vn9#4b8$iO2Lfg-N1vYKm!or)1-KVRX
z(>1VV0Dj=KC^uqs!r56A8}PyZVR@%9Ngh*(AnrI52B-;7;$?om2)1p%b{VaJ_rs+#
z*q}H+_8QKHSz-rEOVeO|%o5Konam81`bKtkHaRT%PR9!eT#^c(Q-+mTK=m>ZVy#+{
zW_k$ak*E@%7GkM#cv=$&hl5Vfc%SCyZ${_~_kqOf*wEqP;v)N=n|sgyRnCtg*#Nj)
zEEFL>SwjO8T*Fogr at d3Zevv}tH>ACMxfMXSJopWy4oFTWA)_6HeK)bx@^eKhsc;sR
zmzNHrNl+1*nwlD%Gsjd?clXY*v9YCcJa78Tj~~@hR46UA_FrC7DJdz at tX*sV?}xC<
zj*cu6+$_b9L8Q^Pfb?%YY(`VE_}3k`brF>=;;h_Iymu8iGI|gic%-=c(G`2z{5kIu
z3oEN1oHV*c<ebF`kDO?n#zU6PgoK2<ipdxc@}7M}5AfDWgxi{CrPi2e@$}g<M)*uz
z`hc5ZK3ovO{&6 at C94rHkBZVj&>;a?>N?H*o<&M^O!-fs;j3#dbS at 6gJtir!pK~CBb
zWWIbUE_dn_9t;?_Ln0z?eW<7pS(3SCW at aw06QUyUurlBbE~p&-o7A(heg+5R!Rcch
z?BhcKxn4(%2;5FC at tntQlnWOxngZ~`4glO%;I#ifMzX&~+AxhN4^|o at bR;VP&nhcV
zX>Q=AsvfukQRHGd#O*i>2so!Fui^NOSFc`?8#T3Yhl1 at VABduHatPaNF2#G&fdNiP
zVo}>|wjwG8^f>!?{RbBo4VUTBv9pDG?Z&rdzSOvG5L`9`G2+`f_)1_-5XT=rd>C_s
zPEJmngFj;r9GD35VG15oIC1poQKT1kauNacLBbboAjO#xfJbl$ph!%C at 5~aC!oxdD
zy>~J&Fa(B%zORnpL(&AW5S0K_u)BYHIsh#2Qh{DkPLb!nB3sNYhwhT#-pT-6eSzct
z<VKpSL;ny_Nx*2yoH at gQOGh9YyV7)sVozGQ!xWJ189Qunqc+$njLwmQRz*c+___en
zUX=d%%^~K{#TX=qTXnG2GXTTS#y|zasvNPWLV)`7=TC$#0P>KZJ3SpW at VdKuB|M#x
zk&$J0NikVX|4DRkJ`OB+&WQjNL`KkS*RG+=;V|RGgzQoeq#{M01QZ1BV at 46)`PS)9
zP!UTT8$?+4_VyH#l1Gmo>6%+sU?`7kYD7iZz{%snf&2S(kY|jKk8kR0)7q}CE*xZr
z*T{}R3Ix_5Odreu0>)##?&fB;F!OUM|Ahu1)d}D`ZZHAdi)*m*ZQ3?MkU`JD;0vm<
zs{ab4>I*^$JRrUc3=SrdNVM4E0bzf6YpeF~_24WpBiL!r0!A`PJ`@Y$2p|0!Hj%Ro
zk0tc8g!!x%S_O>&{Dg2G^#kqU;o(8<mb<kB&-3zbP-cDl`qrQg{Pt}f at x^(5LBV(L
zZf^-&W^d%-<!x<O7#SViuyf~5<Z-f%=A*!%Es;@C0k~j!*DfkN7$v1j_$J^Q03;KZ
zN8~W?U07I%tJwMc((Z!Yz$vMH`;^p_6jfAC85!+Z+N_+Fm38^*Rcz?$>FxCuvume+
z6dkBp4-$dQf;jLDhwAI=i|c+C78atlMvV_dVIMXeG`gIzuHxIE(PkC?44COmNl78I
zP?RDqsJgE1Ed)q85WXPgR!a*Td`sNYjCANulF9u0_sfCCR8&^tU?k`VBT-^A^7!!G
z-W0ZG#n%V^l?NdY4-d!vM&RDyT>8pytVLq!<m5feGgge3E?p8A7e~C0F!+;z3U^o!
zd^8sA0j(k_0?1K+RWJt}zhzq%5zijLbXT$a)!l+CZdePvAH<kOHtVkH>21dx7Xf4P
z%uGRBTU+dOJa+6D04W at 0JOm~|+;fsAOaU7wu!nG&n+X$*Lc+q>?)A!GV~#H4zJ2?!
z9qBMjXMKGz)Aa7dA!LUJvtSCI5o2xuzu3cpWAx<5GAdVCHBC(sNBscrzzvWM1n{zx
z(_28}SV=6yHlx55kIAU1ql3-FSSg1`L5rAW_}UEK2km^R*B=ucJgeKHZDfR`5O1rh
z`r!0Ik9?P>W|83};bg#)jeXC<%~rV**Co24-N2KJj7*ui<l-WBL1-I<Dkc^dbT&3N
za0rzEnzXgG4~dGN0vxTUtTqM#Gm at i#WaZ}N;T+5N at 3ts&M~-Y+?ySPq3CQMSXo%gs
z5OrR?v6XW4>HY|g8o3_bzfY>GQ`XehUPAyEguD+A9zH~RuL2{iGlT2WSC$%fMKU3H
zgAWZ2^7i%z0OQ^Th2S`1kDZ)2z^2SgO>i0}Q{m7|L4npAfUPE`rsV;9Sk-R0&HUJf
zL+ at b3`@k|T*N<4<6c?l6zwYXagt+9!Miz+-92$d1M^?+ZDJz`G(#bOLO-bSP at bJJI
zrTK at uyAHu8N(fLML-gv`uU{#C0J#xrY80_$X4z8y`0>N1xVTucweF(OHh>GhfCIUf
zCmOH^2mEKzwegMK-ri_7K|n~6k&$TDNUOQnVOlA#g{|OzZ=L=(+S{nujSV7cEzZ>J
zY^-D<WiApX1qCr6V+nXY0BqRbd&@iH>C^Ss<yFJ0FYFZ%r~qV#!>LHB1%Qo~mKKw;
z_TXh85c+p$2uFc$P2Rb4$Hd$`$<p8j+}i#7Y_ at iG37`l)M3$9Ihcv|4_#1<R+;9dM
z;=|SBRm8;HhA<WbT*o(ASqJdT0K#$l{)q*y&<~;ncyF8{L<2lGHwWq%+t+4oZM~+e
zcQXCiGdV}coW5VSr5_du1X?ES1?VZKMxNcjf97AhPai;jgEnLDUO*dI>jev>KQfe*
znz~j)Lj&0#zI-wGIReTH|G`O=aAN*37MPYCUORx3nqb#hz>1(xmzLg%qs2##;20jB
zJjnGs?^2|$2~&7-Pm*cpaU=o62a1gizP;t(Ut{;_$%v}$B*z7oZFKv0mokc(4?nl@
zYggIQp8t0?IOtd+*nMxS;e4*XL*k6S-wZ`2BN?ta#u#PWjZa71ePhn!D!iVZZ|JD}
z`}gm^IZ52=*A3~nbZ}5zyLPSH&py)c>1prko0kaq^J9_uR2V>vsfht4B9Zv@$B%P=
zyo?_zpRTvi04ro&py#$QlOx%nk<@rDsqwB|$2FJKjt>3V5{GXy02s7Vn;O4<O$QN4
z77w}DQ5bjJGviW2+}GL3q4tiBh>@aeE_?*<9kT69ANIwUi{8n_#bQler~MJ<F_+it
z{7k;Ow^(a&aBy7VluuSaLz0z&;}M9eP0!;QxbiA_lz3lp&o4>NjT#MYZEY#LT*Nu8
zg&tZ8Jp at mtsJ5B^7I*OJ-6YdAuhP;f&#(H!Uz`mg2Dh at YQ35qrjFoC=f{RxzmR|aJ
zpYCd~jUL at m<MVS1vw3rWN{TwO+=}B$JZ8DOr|VrE$(@Pn=@(=>UKyMX=}$YicW-<{
ztt038$mA=p9%V3Of{eA8f-O at 1v-faJy~vKxsIfP?@!K~B3 at -9U1oerxT8ezM71Lm*
zxz+#4KbJ3}`r{wLTJ-`wtzx67k)rl5Uqps at p8fsNd-7G*Yt70vKE^Gxv#;7m3qwLf
zTkKyrU%8Ta#G&hf#A~~b3CRY=-n16$4EUq{G}li5otavAeqkYRN1;Y(ann~#cyz|f
z=Eu$DCXXj4uWG~_pFgjb)JXpQyD(1DHL<hPaCB at 8ewja}zN_Ij)zZ_?tOd)R8WFK6
zHZCrs>1$fk*I_`xMI|MXQ<-C7j1nr=0 at d#wIR_?P^EMBHYSs)~2``yG<@IMyE2&Yz
zk#l5X;v?fG6}npoHtN<}<V{Gbc64+MHdz<s=jH9L%^m9<CCq3rf270D>WQnf$kS|U
zF*$TDyRGTnJ5p7D4j=&KKnW|a;KUCim-p1?20iI{Y3b=HRS at f*WBHJ#bUn#W;QlQJ
z-zddBM?;eIGE?AY+NTVkw(M0{v-AK9Pf9yvi)SQb&x-B{Wjhs?(v<GUMfNzZ!%V~5
z%*XG1wr=Pu$@IaTK5v9tx~5){{3`=74wKq}-2JtRq9uYiD<@|cKK>s6*C1Up at +QNY
zviNU<gQo$Id<<aPIraCCdu(iMdfUiDiN%3|WV#!^@b*^XJ3`+_*IR^_FzFZUSAt7;
zXLeXCp!?G&(1mk(pEdJ7-y3qhR;0FW=?nIhx;Q$TM)r6t%&3)&#}p~D7H+I0H-Ff}
z&v>$a=<1p7n{gIG+l2F*3r97}UaP048D+;L3&#yrJ9lax+(TH<r?;Xe{MSzz(}yq9
zwEy`Mq34>c0>^*9<VDzJ_usF~P at O+BI$uKb->)tl_`60>I$_dln#utGe&*C!xhz=|
GzyAU5D?ecX

diff --git a/docs/source/conf.py b/docs/source/conf.py
--- a/docs/source/conf.py
+++ b/docs/source/conf.py
@@ -17,12 +17,13 @@
 # add these directories to sys.path here. If the directory is relative to the
 # documentation root, use os.path.abspath to make it absolute, like shown here.
 #sys.path.append(os.path.abspath('.'))
+sys.path.append(os.path.abspath(os.path.join(__file__, '..', '..', '..', 'src')))
 
 # -- General configuration -----------------------------------------------------
 
 # Add any Sphinx extension module names here, as strings. They can be extensions
 # coming with Sphinx (named 'sphinx.ext.*') or your custom ones.
-extensions = []
+extensions = ['sphinx.ext.autodoc']
 
 # Add any paths that contain templates here, relative to this directory.
 templates_path = ['_templates']
diff --git a/docs/source/depgraph.rst b/docs/source/depgraph.rst
new file mode 100644
--- /dev/null
+++ b/docs/source/depgraph.rst
@@ -0,0 +1,125 @@
+===============================
+Dependency Graph Builder Module
+===============================
+
+Introduction
+------------
+
+This module provides the means to analyse the dependencies between various
+distributions and furthermore to create a graph representing the relationships
+from a list of :class:`distutils2._backport.pkgutil.Distribution` and
+:class:`distutils2._backport.pkgutil.EggInfoDistribution` instances. The graph
+is represented by the :class:`distutils2.depgraph.DependencyGraph` class that
+keeps internally an adjacency list. Several functions are provided that
+generate a graph in different manners. First, all of them are documented and
+then several use case examples are provided along with
+`graphviz <http://www.graphviz.org/>`_ illustrations
+of the generated graphs.
+
+API
+---
+
+.. automodule:: distutils2.depgraph
+   :members:
+
+Example Usage
+-------------
+
+Depict all Dependenciess in the System
+++++++++++++++++++++++++++++++++++++++
+
+First, we shall generate a graph of all the distributions on the system
+and then create an image out of it using the tools provided by
+`graphviz <http://www.graphviz.org/>`_. For obtaining the list of
+installed distributions, we will use the functions provided by the module
+:mod:`distutils2._backport.pkgutil`::
+
+  from distutils2._backport.pkgutil import get_distributions
+  from distutils2.depgraph import generate_graph
+
+  dists = list(pkgutil.get_distributions())
+  graph = generate_graph(dists)
+
+Now, it would be of interest to print out the missing requirements. This
+can be done as follows::
+
+  for dist, reqs in graph.missing.iteritems():
+      if len(reqs) > 0:
+          reqs_s = " ,".join(reqs)
+          print("Missing dependencies for %s: %s" % (dist.name, reqs_s))
+
+Example output on my configuration is:
+
+.. code-block:: none
+
+  Missing dependencies for TurboCheetah: Cheetah
+  Missing dependencies for TurboGears: ConfigObj, DecoratorTools, RuleDispatch
+  Missing dependencies for jockey: PyKDE4.kdecore, PyKDE4.kdeui, PyQt4.QtCore, PyQt4.QtGui
+  Missing dependencies for TurboKid: kid
+  Missing dependencies for TurboJson: DecoratorTools, RuleDispatch
+
+Now, we proceed with generating a graphical representation of the graph. First
+we write it to a file, and then we generate a PNG image using the ``dot``
+command line tool::
+
+  from distutils2.depgraph import graph_to_dot
+  f = open('output.dot', 'w')
+  # we only show the interesting distributions, skipping the disconnected ones
+  graph_to_dot(graph, f, skip_disconnected=True)
+
+Now, we can create the actual picture using:
+
+.. code-block:: bash
+
+  $ dot -Tpng output.dot > output.png
+
+An example output image is:
+
+.. figure:: images/depgraph_output.png
+   :alt: An example dot output
+
+If you want to include ``.egg`` and ``.egg-info`` distributions as well, then
+the code requires only one change, namely the line::
+
+  dists = list(pkgutil.get_distributions())
+
+has to be replaced with::
+
+  dists = list(pkgutil.get_distributions(use_egg_info=True))
+
+Then, on most platforms, a richer graph is obtained because at the moment most
+distributions are provided in the ``.egg``/``.egg-info`` rather than the
+``.dist-info`` format. An example of a more involved graph for illustrative
+reasons can be seen `here <_static/depgraph_big.png>`_.
+
+
+List all Dependent Distributions
+++++++++++++++++++++++++++++++++
+
+We will list all distributions that are dependent on some given distibution.
+This time, ``.egg``/``.egg-info`` distributions will be considered as well::
+
+  from distutils2._backport.pkgutil import get_distributions
+  from distutils2.depgraph import dependent_dists
+  import sys
+
+  dists = list(get_distributions(use_egg_info=True))
+  dist = None
+  for d in dists:
+      if d.name == 'bacon':
+          dist = d
+          break
+  if dist is None:
+     print('No such distribution in the system')
+     sys.exit(1)
+  deps = dependent_dists(dists, dist)
+  deps_s = ", ".join([x.name for x in deps])
+  print("The following distributions depend on %s: %s" % (dist.name, deps_s))
+
+And this is example output with the dependency relationships as in the
+`previous section <_static/depgraph_big.png>`_:
+
+.. code-block:: none
+
+  The following distributions depend on bacon: towel-stuff, choxie, grammar
+
diff --git a/docs/source/images/depgraph_output.png b/docs/source/images/depgraph_output.png
new file mode 100644
index 0000000000000000000000000000000000000000..960bb1b5639e7ff615ddbd6cbf8a542be31d5fe7
GIT binary patch
literal 24719
zc%00<Wn5KT_xHU4ML<9rX#u4bK|lmS1f--vq`O<BLr`e}2|>D3x*H at FQBqn;>Fx%f
zvCg^v&-J;VcyqtH&%wiiz4ux(#u&fvZ_c?4Qk0j(!=c1Mp-_0zQsT-e6j~DeVaLLR
z-&~f}{RZDK4WCMiqy8fQrZ(n8p-?wa(&CR*T@%+P-LzHrFCJ};;GW)HV<`K|kcx%F
za>LfzBD>aDIw9D7-Soc4&y0!mXp=+d^#wI&_EX(*9+!0PYHg+VZ<k(Nd7?yyb~*Zo
z0fm2 at +uw_=p5vf<);KRkrh2~^(SAPmT-bH%jcapD>3xv$fRUa7-6(|1^pbKZhJFaw
zu$TRS0sP35=y=^Igd#L=6zy)OK(c6>$df0R+1S~Yo<G0hCyIu;!Jw+9hO=+XO~Jvz
zfkKgyk%e~|msD2=ynTy7O-+3xLP|U+Md%^_^4(!Si%QM8#l=4ZSx;FwIK-8d2yt+5
z?AOP6Nk~a&hYPiGb8~$(3vEAMtgWxlE-hg_fBt-LW0H!NmbO(vPfzdCrAt`S at 3_Qn
zKJ*AJEkZsZrJ(R7PbDWjI@)h-&Dz+^ETgCh|8tV>%*F=3f`WpptLx{sHVk=r`Bs)E
zhb1K?K1oSQeMOCpjd$NDv5<(q`IMjkRyj}n$&)7;k{TKmGBPrXs;a)}=?o-fWa*ii
zm&7j-6BE0-xutpky}-i76_=92VqjpH+1%{SxQXc#g>RHZL^%6B?jFD662ZXWU_?xe
z|Jj+RRF?A5 at i9FM3l<t08uGEf7w7%ec6y_uqk&|_GFhUpUNwc|_NF{~f0g{sOsn73
z+`>Y?#zrB3T#7e=fmq^_l16i1BVZKMJG$H3qytLj(@RPSP7gOF9UQpvX&(kA+_!6m
zenmt?wBieLaNt%|RT&11TUu69#>K_)zdGZGw!AnydXXsTg1 at k^a9!AwU*bJ&Xn%$z
z<HLuElA at EBm}xr2KO`i)OYy#-cyzu;97gvDUr<o6y`uwL at K2N1X)ujpJv%4omo%|}
z at WjNBxjAF)T8GOp5wYARIN7psZwMJBMA+toDzCB23xrT;JbOmUXEW~0Yc)L69z?t}
zluw at Qxg*{F<qIPd6DAWAlVO0lDmkYv3DG7 at S6}{fPT3S;8t?N}z5ZgoN2P5>>me|h
z5)RPjv+2L*bgzzPufQli{Cl=5B_%b(cv?TVu)x5 at ML6wsNVBuELqJUYmR!%1P~p)z
z&-d@&+wtF5Vn92bO5R?%_WRHKM-nnJIGERHLms?b&|mqPm#%xXxmKZhJ2EP&)P3L5
zdb)|m*4EbbbiI1J^(yD*fdS>Utx=4bHug?|ud{6dT-Kv^LuiF4q1{r_(k(&6%v%Fb
zIbXC|yjErFh~+Zcf;;@)neUFhYk)rW>eS}w=t#dKm~{Q1meJG`vc<~md9m&jPtS+H
zfB)9EvwL4r!9&e&k18!KH4GTO<>v1GJHNO|R9Balh=`~}zXjdb*H?eK$;&XHa{~h=
zW;{qOzXMW)kbuCa+^pBIH(6-3-0T`e2ZOWuwaet><QzJ6B#S*sV$HA4?>eukKivI(
zuf}OLW;CsoZ^lX>r1GuRNRdHj2*uI);q>-<?=uX$hHcF6KYn!JuclRb|9u4!dj$^<
zPrdn-kVGiuEoZ3$QrFD3Q49fxIjlEt-zq^4WJtdMo#Zk#>uYq$s57Kxc`%2Nl+=&Q
zs8hc`{fS{fyXI?{uxt8X!f37wI&;u^tZ^kibR&Y{Clk!H?U<UP4GIc6+VuYW?!E8{
zR&m1)4i9DMs6naemQFTBy at FPi^<z;{bX|{iUeA-|98W58!p5WD at 3ywKJ`XB9!t)b-
z?X)_)ecCU*%}dj{A*QKGX=r4$y-~kuINR!H7|?dXeET-ud!d8tx?cOw&Q|sQ5(!e9
zqL8s$I at w>v!p3fiQV<S$mai5>`EJ8p_q+F>9)h+>=Z#-MImzxg$lX0X8%AeKJfOrC
zgl5Yn`2YTGxHjFaw}*|9WDqb-VOTSglkzHfzWd{!-yQFCR*HU2q9K=`sh)n-1|eX7
zu%>Zzdist=KIhl03#Pa)gpE;qAfbAbXDwW}9=R?a10y4s=aGGZda3A|BFcWiq6adN
zkc#Tv(~oyrCMLAj>_`7 at eX!`%Ts}MAMM~M)Sb69`ie_P__r)%qZ$N<9osVKnQS^L#
zWP&amJnAmcd3zU^fZEyzCx at HwE3HOWhE(!Y)ys^rMa9JEqq=THbrDcf2Ag!pq@|^y
zp$O>ct_6v|LBqfl$sfY$9kQ&veperLlL$N6>p;VxR-A#0U&Z=QXJ}w(C{li)e9tT{
z2Dt4mq<#AINk~oMo&PnuFk!C~{lRQGiRJf|L|mPkCeQ{R{m-b3y)5NC^Y5`-mp;#k
zNgle0ioR)amzI_m)yYzp?(grHtD+z$mt>0$hMM}dzdr)i2NUDBSgchYhEjxxUS9Z-
z;Lf#VUa1)`;A9zfe(CaMrDxBcT`6u66h7M$16|;j(JOF5R6_=un4wW_N|99z^FfR%
zD=keM%6~pss8yZY)RY`XD<l)etU$K1>^wI$HRU+-8BLij+VT9v$sIxq8x>0ZKsmSG
zb=weK1Ro#2!lau_nS~@%k;Qy>;fJxEodU{yc_2$%yuzRj6J11DSU6KT4 at ytO`}fz-
zQsEYS!TVF0<xn-~vu!4-luS%+<*BlL`}#E%Dj)$W{*m`9q|Fdo(x}|*>`s>R(wG-8
z`O%oT*G1?&w_Yoq;9z6Zqbgqf=`>wkc0Q5}qqeAM$jz0K(iCWEX}RP>L`=LtK2~8t
zjF#Hhr)V&CwEYJgRj6GX*>tkPjxM5-BX7fvne_Fm6uO9og at wD5(~INN)BC4&<K1!h
z%&<^8HTEsYO}E9-J2kro2Z at nf*Y!T%$KQO)6qN>gh~tOuP}|VuUT4J$Nq?7 at msdUe
zl)SpE><vhtlM`=2VIkE}VNsEPXeeGmL4l%{)>RbCojVeWiUjbZ?*aF{dmzT6){lmV
zhe4 at Go1C?(6E~>m&{0r&Q<d|^%VajXVM1o-=lvf2JtGziz+d0GQ15nSKr<?wTUiMP
zJuPj!yENrHNJ~va(_e0;5E~blR#ipT)6?^VNrXnnu-}}g556ob;Vaa1beZY4V+GVO
zG}NzOzd{D+tSBJ&G`qUHuU@`<Sz<X~ulW^f<CkJZ<KFrNIjY3t(6+Hmr{0-&g&eie
zoAN#}k(xuZ0&BDB^iEH*(7VHF at 8*tIZ{EB?p`d6IOWgOT?Nn4$MBTf0PhKD?XuEfb
zYnayy#9(f2PMj_JQ`_mu-k<_DYHVz*Kc3erx1b<oWp#DA_t9S-{)o at v9^H<_*oldW
zEUc_Szz=zKb at 4ETT`etX{o-#huHc7 at Go(Svg-c|l)zt~??CssmQ<Y(5W&LqbZu)}~
zT_ia<xx#i*pw4}tjhBZfTsbfE9XaQ~{b^6JHh(;;FO@?T5bOP;rADCx<W at AGYL&_W
zTyY5rgGD;^0u7weGEXujLM`_{Scq-hR$+?<b{aBH0C7>((^HtApP$><kVlzAPg{4L
zqVOIH3FW1yr;j0(a){~F;?BZBrYc+4 at NiXsg&mVHa`_g$zbDV#_xdG62R_PVr9<^l
zvbVphp{W^JS;^bc**OqKBj_t&BP}5A;@l2xB^$3OD@&i5(bt5+Gc+<nOxXXjS4a|N
z4)rfIIe*BKm4V?6s&BXV(X=6MS63Ga9i76)#)h$}sf3!E8VL!B_{+r}UjFjJMkLCA
zO-{lP*0r4)7Jexoxo-6q;3nN}x=Fp>`8Jb6GKVvN0k(;W$<jzMUDMI;tFlP~&!BGV
zAca9f&vW}S*=<cktXg|}72x+mJd}8}-e)g<4h_-s@<um*FN~VaY*VLNcDC&3=r{u5
z);BX_c9vSZQ`~q!3G;TGni}<+bugqdA}9zuGBPrC$dc>TnX9U*Y9I|!TT<_p#=jS*
z>XnwH at EZTDqcX^Q{WWK2=cD}*Jta-e;H4#V5_0l)G|JW4{pMT0-*Fm_m6?n}HiKYw
z;0N&{GI`kb<;#`kzZZhCa&ib;WM*dv!v|EAmES;Bpl4;pu6Nm7D9+uOgI*n=w6(Rv
z<Ku&Y1Vbo!aW<R(3jegM95ryF@;F#~3<L`NgS at VkArVAFWQE8QZFphf?N=AaW}fH!
zBkHAwm;Ldm{EfbZ1<?$P6#~WrMRY=0;R06pN2M|Ghy*!?aS&8te!1|yr7Mb={$n?y
zRZ!3|MvC>w1q%X#WwKfk4ZJu%{WH_ at R!T;uwa#f(-^hs3*}Nbc8UHXS=>Z=;ker?E
z_Dadg`878`nsWVvmu$Y}qsm6wbi6<X#6I)(n($}#;)XZ7i at m%133mb?<+Q0^$;;1g
z`}8b-YdI(7Hz+|YuNB#p=dRp8lB)R;l)w+v&G2|<9=L|gv^P1_Wy<XeWDdee#L#1b
zSZEz2_WN5;V{UMAUW4#=-R>aY{;8tS(%LGzwk41S{WsWO8G?4B+b{l{r?P86b8&Hr
zU^KcGjId_<kmU=IiUQq6N<rr}9~i)7w*`uSiB;k#`^Ux#2>;RXVoF48Y#;<IJ#aPx
ztlQApT^%}F1{gi!#}*c>eAc5rnVC0 at jg3Edbzv(AAKy%N{f*t;-d?dnB~ULAD)AQ7
z34yqy?Jsl at adFt!u3eLj<GzfIjg58nDzTvWFd7UdQ(6A=V#C$p!Xp at x@R%4(5N at dX
z2!_6YxP#T&+6v;{j=%dFEEM=0MqXZ0NG*QPqr2a_yJtbRnUh`7x3{+gLqg`L9EFic
z8V*u at _6%Ea9)b57d$jEAY()MMlR+=$ufH|Zf`x_U6CE8rI?E at WKg96p5uI62;$wGr
zfq{X6 at Q4UBFn4BVW)FA1GG6DkBmo0~M at L6j+GZlrj#!d4n0%?HPusTteDR}k0shZa
zW^HI}M0oxZCMJ@>OOy5Yq5RB(7xVBq#eZMe)I<wTi-nD?71Sc)<HrC92}VIdDk!wM
z#l-=6d90ZD)Pc}zdJ9ZU%pLP=sy~ps<S<haU{&DFf50oXGBPvg^iLMeJ3K%x4leP^
zl>murPoF;hI`06wg6-tg{9H{<wE8I&ZeNLv$N!i+H at 63fxl>kV*fTRQcnwf<<L+Go
z#GYYciGV#LyncN&Pty4N5t4!7xkW{t&_pG5^}zpl!G;EHPK%!-BYzIoCqSuCeM>G{
zT3QBVNf0-nf_srM85xZ3`zx0 at IXVC1I)ywANY<Uoi~JzkaEXF0CGgq&E_ODh3A20K
zcHQW^Iozr9`aMdhz+9an{_v&{t_-ME|3|)58t_J(;k7sep<sw<^v92k^z>*#?t4g<
z|F*GodrHQk?qksAPr-3I|51;Cl=Lzc6&0A>8*FSi6IC`=Dh)8h>J6 at 3{`fTGpK9R0
z!PkV3QjaBqxQ5r!Qsol(2vIAa>ixn38yhD)FOC=e_%?0p^X7lBvasB^diCmf&5I1(
zF|!n}?2+V+Qp1j5wJ%mn$3HZ{zJN+8D=8^OIk_TDmMC-C)Sb-%`;>I<KG>a54%8PL
zAOAZqC)u*%;r;vMn+?18D8#|AaB?Cs0pf8975m6r7;35}shz(54Fo<xsj5C*s3-_s
zh9ran!=sZo#QL7=pb!1MW5_^ukx-%9#HV?=TCY>SE at STahi+xMV{a~B#X><(<ch7w
zD=aSgOnLu3KMiZVMa(37qgGpfXlMwt+YQ=?fs%zc9qG7u%}uS|WQ)dm8pH9}+?>Uj
zn*zQ7<4di_xOsSZu=aO$cj+Ufz&~CqRAc9O^l~AjZvlwUcV4rQ?BL)aT<Yf6-d;PB
zk<L&m at Dx`)JUrCZ)j#+4UIVTbd-iO|t68m5GZY*$$?WWGNoncpxHt;HDKG`^p&}U2
zR8$DZ*M~W}j5Y2%j!4F9O(g=~B2)(S?rs9Wg7TrbXSx)<977}QSu=J(CvI)c0l-uG
za#6mGbUbn6y3sWno);H%E;N&D(K5kdVYX%R+S=5B$7ynq2IwL=@<~K7w>8Rq!E#97
zB+{&N%quzjnWu8er=g+2*F>{0CR;vf+}Qi^<Ht~`vJNRVjSkkvZZ}QEeA8Mfue7QB
z89+anqri+pn7*-$R>++I#m36o&tCuPTmUUKS;(WxQiyloJ^tlVA3eIrJ+mGj^&Uo=
z&g&?|v9xZA|6D>trFdUF2%~<0e&xy)>#?%S0K at PFjNuQCM_E~U3o86o8>rSO)T`6=
z`vzkmcZA at f!ovv=Kd4 at mPfSI{&T$Ifv&3!J)YZeo2tpbuEizf65)zmaA=h6Q6mXoK
zPXBEx9;pL&f%?px;`t^rlE|Eg@<VjA0g%}Bd!`{J`0-kFHZL}J<1LFTy$~5j^;O&H
zHF8g4_4W5F6 at LfuKq+cyj4h1xX})^(3gtu0EPuVIZ^7})b&sV~(h)~+=GdrPOiU}$
z%gI!flx+Vr3JcZo<qI*PlAIi#hPrwm7)gt%1~+313+c+rN^x<7y-?;q;(1$l85tR|
zQCV48%lpq&RmD+Xr>C3yt<2Z&{P^-E0M!S)xeQ2?W7m8h<KW<cTl8^Z+`8zOqK3-;
z1d$+Ng!Ua~RH%fzfT^kJrPTgh70E1RLKOG?`*u!xHld|OMPaD$(9qCX=kZ_njMmZy
z<1>w*2`F6M8)_^hmwXT(Ve-64n;tdy`?sjLIPeY=%5iJvv5dSt6RHBTmVv?H#S41W
zhlB+7hk8s=aBVIwdSYVYNSUmg(K6Bu5m+zc<KlST at m#(Xot|z4$jiv!)YjLpM!%;K
zjFEZzG{JhX&z+cY8LS(+2xX{5vFzHQ6=kSDDqNL~m)gTBKsJPHn9S>B at A2kTBLZ&V
z!40OHyh^I7I=@CRayiUNeKqD at o2b6_;K2iy+qeC4a+p0mJ$)Ak|H^Y&@KC~Jf>&)^
zZg)}8HGelcs*bE<Zq2r-*EzBx;4nHiW&l@?g&=&}X#XoNw=V+n*+-CAa3f1_-uX}*
zF9|4&@6#Y5cnLHVSSifz2T)%x)oG}>^zsk(uje{%P8B|{vKqOA5@~*QPI<few5|t<
zMq*-&moHyJa^GNL!aP0PBrjTAT3G?ZjE3q-63lVI^VNK at o{?1t0v{L@)G}e$97^}<
z2o16i9YxJ=pZaVS12yOH0F_!(Q`5f#tjtI8++%BR?-J@)gPTe*$DKR)sFGLb?huM)
z^~Fh*_jP1-b#?#KJg{h}4cV?n7ZNU~6;;&dZqDWqI`1<=aFMxrc`er-o%z(hT;%(d
zkugzgGhTsBen*?&&H%>eSS}-ERTdIU!pX_eCXDV8?dnLeq=5nbPfK1Hf)~pJw-EMf
z3%J%?JVHT36B<f&KRq`Wi!v0Gl$3O|$xG0ulNbwf21P(bgpP9k6MA0}rtRd!o%*3G
z0St3PagC-GPz=ha`y=OS^p9jAs)dCG1idfLH@!2`)6wBNM}YjL7K2Q&2?>8Vy^o&;
zbU4p{BSnd*tH;jb*a(**xi#ZG;%NFKerT?eAWK;q)I?5x8)7-@<m5zRePM2n=hMrJ
zGq*GjcJ_RwvH9+g`@h}yhn{N*RUS<>Yq>Yd2`zk&WBh4l>qS6#>((s_jub8<i2&U}
zjcQvOv{ZOM6^GphAAgG1fkytR-FjW~#Lj#-Hp*tIVNCx4plocoBO5#WGAJh#R80f2
zpOb>*&<-$K+2FCtDk^TWv$Nyi<7a~sE#0D~0enOkX#~$(_PwnwqiZxWDvDKDI2n9Y
z*Z1!skZCL|EP)F>NfYL#T3XJ(yf2QK8 at F1o&i(l#1`%RxV<QK8um@!?3WQ{(vO+j8
zaJWE23|I(pXmH_dzpLcRVq#(eOi7<Uu|;+n^BA=HGOb!q)<r|LdkmqAT!c{~1X&)T
zESN}^Hf#08Mt+Cm*Q#GT=Z$x-vyKa-6Xb9?iPk7-<X7#>Qn0+gK*zh at 9dxRxs9*qD
zeuo*;^V*MY5%81?;OeYPMHnBZU#r at dRFI@)xKJw%%$WDZ8Ph7V6H$8SE3&%Lbj;Lr
zabi^}Mp&R-OJvu$Pv(BOF=<Sr!q^D}P(b5#w*AF0pzmv`(HG5vHbf1KB-*6m`mz20
z%YFtP^Z(XM-YNJ-li#)M8vRsq<XgJBq}30=R5EaL6M;KH`Ak&XwSD`B!~VR;Hz6TG
zJ*_N&04`P+=Q8a at 4l`=|OlL}{<e)QW<oc?q=I(kbM6B!I^TSXAyj5F381?%<K@;X6
z=CqjE*{{BS{aPY8(24xvfI$IO{DYTFfSR~5Af=G}FIo*~-T<ymB1ZdKrKU~E_xAQn
zhz2@{7u3%TT7@{N5E2sdzxeGtWsC;1(a`YF)y*x4gu)|sTPmQ`mJJ||iHRvRMmkFw
zmx9}u7=QIHbaFKtDQD)D;}YE<+bNnxG>BcAR4vwRBp at cvxIgZ8vYhh-T6G0^bEoO|
zSbFV2E-tReApP!pOY{z^YX+TQ5Ny|gx`-qM#C?Gx0;vf-Mv-MPs`ARpj+z&LP$&?R
zSRkjbgrxyVd825KWPnJm8J|AkKvX_^_AGQoRZWdAUf0sV04+|I>CDE^5FLmcXs)%h
z6BFT|?)kg- at 82I^)tAjuPRq{5thSr>|93$Pb-qrHVq;^QSzNq=+S=Z3nQHVfu(2tu
z{_*W!i!jlUltjb9!Exhbcj#{VsMw#hvqXHtS#UIfRFdH7z$@7AF38O)fklUKOoL`Y
z^Fa02`dncS;B&JxGblkp!M&BCt571B`ZEHy>dVXTK`i1y7K0i%nLjMB>a)r3=;#30
zB&RhoF>!Am`S~-osHjKZaM_%Xj}O_90{^hKwl)JchMAd}|G|S6V1c8F{(eHR5555b
zSG2XY^)0`BX_!>{C?_D3MMTc2EAH&fcSonr_w3A5yV0HW`1lyvO!M+;R2+eL94$Ru
z8za7R=Z?LzvwvY>VK{C<VIjDC_%Z;NMA_ThoBzxS4VBCA{xCj1E^)pTb{A+uQAZ~<
zI2h;W&!1R~x!=CYsH&>Y09lHnbKJh&0i{_{ON+h22rhsa95=U?i3zHysi~#u=0`AU
zSKyaYS)OVY`6?_VH4gJQ4liH&*&M(t$U5SS7hFX;^~2^9+`POt6~xURk{i%nD=Vu;
z!obi_Jh*{@kr9T7_vz)0MBRH_TrJaoFI+fu8?KX)k+n+wwXHLe$x at cG-QVAD4g3Ti
z68H at QgoT}*IC6FpduPK&n=QJvqvNuzt*!mbmv8Iq1+{f_X8-)b(9_e~TOPd92)^vI
zi%(G2ATtfPab%;=*!af1d&HnspP|-;r=$R4Uj>alIy%A<G)FcLJv^>@*VokWxV_Ko
z$%NqY^YeqUD)RX8>u`~)I5_Z;_uvdsV4q3R(a{MA2z)@P8F_dR*0y(YB0VN8 at fZTT
z2!{3Rw{M~_V=yF$i-pQCv+yGUwXx<oIXT&1?_yU|SJyf at sbeR(21Z#+YXU#C5^e}p
zmS$X1M~B9Csv!t&)p#jiOHEM`m8Fc03hy!&Mc>)qM;2VrF)+C7XG9t?M2DRm>oQ*+
z at 0es}Wz8OK|AEZEB%dVU7avav&UkNgnhwgzKNEJa{w_831`JJ(7VaP^&gHW5@`$9Q
z(5vKkZh#xt(9l4pkcScvaSODxVOv}FzJ7iNM_aS4fuT+fptgVDvq-m0O_{jw-FpG|
z%pI~+*VJsCp4J<Ey+Yg6)U<VQuweDdg`PpX-dXPY<&{@Z=@^)quRt20P>{E{*YBb*
zFmXi`6$wN|MG2^>f8D~rTQArj)syw~qy5}hLTX{p>v?&3ENpCVWSPYN-FPJ+BlCyW
z?5&O{Y>HX0^;`-BdjMREK|w)*bQiMX5>z^n&Gybt2B0OdGM<j3^0vQfUhnNWQa^e{
z{pCHKq`SMow{PF{EiKuoaq+1631D2X at bM*5guPrRY?wjAMU{KMeQO1-u?OL2A;AIm
z((g-s92giVA})RjSwU&?Iu&Jz2&Cbwf7c%+Z)0=&U~~GtPQ9~1Z?aI0^ExFpHT7LU
z%Jz;9dI14SdV2av!@U6|+v%pzM;AvJ&@X<6IV at z;wNULHSd9?CqbMdhQCC;~zxS$?
zA!G1q1boFF&X(r=&sl%~{8_+Fs80bwLD&#;ca6Uih>MHg0Of&koKbp*6^P~L=GOXJ
z#NEBl-oL%42Om%- at ZCExD9kX#NAbB|Ln0lKp#YzUxDqLOd4J792}egBCfQgYK-wt+
z0d**!VYEWQ7+3I-y}cU8Wg>`zy9TY7WaD|TfgSxJ6Gjc3DS at r{?PpNnpzOhXY1cVo
zJJeDmHv|IhOB3UA-O`^$7h3fM+UrSq^d3h4>|nwU*(VPU4xX8t^No$Y{?F3VJyYDX
z^Yae;%8BKYOqJ))1CZAJ=?ooh_I?D^+8IXkPCAOoU~RP2^OL0b$2&T$VKjmzK*?tp
zr<*0^<!vu}g#sXyjYdlh3<H>h- at HLb_ICxH*F<1~krh|ZBl}sb2i)9|lG8fc^73K!
zHzFb<-%#DRMV5eI)LZAeqChEIt~5*ZZAdS|Pe&{H#U)OwYCngEe{X-GlN1$2havhb
zIhfS9zdC{}2s*6}Bm4eahlgL_+xEayPNBWS8Dvd=xLEH!va6n>kRtJ(R>-%1+UUK|
z at Y37nwOLVdWN*I8ZaQpzqM8x(y{M=tmj49<vj3Qy%OZUKODUHIjK(;?FUSTV3I`7_
ztGMYT7@~DN!_D-jS#NSy^_2U&#Kc4?rl|hJ2QOI!1+~^n#-l~kQlUxxphMDZ(UA!W
zRE^u6R3GBwWh^T-*S2D%?&vkiLgMBawD}XFpmb!t7d{EZ!NF<AUr*Z=k(SKp?}Mt8
zXE)shZp-5AXdBE~%Rz8x5EMUugtt~kis=A6;-R at oLLO>sCA1Je_&?(A1;8XB&T~-V
zkzDh|2|F`o_99e09`6fJD0z%|sxojZmi4K|gbK~V(KI2F4Z-D`%*=zSkKanMMGq8c
zl+)HR8agkg8bZABK-hre<MB9HyW=wVF(yVrC0mvrt^tO44+wb&E3q<IvoKrF`#c)d
zNs=hY7hJJLn?GKra$Y-r;Zv)Z1FURp>Dzz45ThvX+ftjDnmUVX7ZCSbR#t$0-`U=#
z2b$KbFel{Y;R&QsE0<4xC=FrFgyMmGI)TZS&gc)MAzMxFNqiuNEIz2HsO;?Qq{F=L
zeq9u<WL>_=#1xs4!MMLZ at e!h5faT_}22}dNfvl%p- at XMiFfuN~{~88lF2aEFxNK-6
z{c&7U6mVP$AEhr1kb3n>_<Y^Yo4BetBm}oqY{fw{+7S$qMP~@b%?PQb=;yj=@$vCt
zVegy_p%jzQ(#mtkh45LA-f-rB144}4ySA=w;BeZzS2DezE%BjSWO at cN?GS{GfIm)#
z0Y6PDVjVwy`UL752C-reKDRS4$O(-tI(W)$ktR~=W`*amgOX8RetyP-mkY>_)rg*V
z5($o9?smz><NTok2+6NqT|uH^VuJHuo9|ijEr0m<QN`Sx1xidmXl4j0 at v11Y!Vfk*
z?Ra<b-GH}}5+NS#!`QX;bqkou^!C0Xv{dJlJ at EimcJ>TN(5R>=Vu&F6sq7dkmB)|K
z`~m`^8XJYqPgaX1Hk<P;c4!7HD|ud>x#HmBW`g1wb+*^VLZU>~);{n8MwVEwE-4B3
z^zvfm;rRfe6*_*{Sl1d>&q_n+;Zc2jw%dEvW@^$|r8`tEPjvt+N7t`kiLZ|5$(0P3
z*y23{$Zl%_IXOFmS{Ze=)#XOBKaZ4U$^c@%p4AK at N%_$A?h^?KWfPMur4ws>=A<y2
z!IqZSs6L=aDxG{Q%Ubm*k4+))Lyl)hFG|IhD{!YU7k_uWL$pI#nJC%ycTiX at DZ0lJ
za<&Bum8hnsX27)QTw(Q8E0~#xm>5hmvqk*%d|t=mk<ZO4IU|In)h-Ej2Njw&88-QO
z^M>)fR%B4!{Gm--h0VBJAA+)F<3xLfPYVcNrZhqR4z>poca4w7Tie)J9ITBwo*vk|
zl~k-V!SQ>e>%M%;!O`(1F)?x4S at ra*e})`bC at wx82^pc2r3_4FiKX#R1qGRL_sk*z
zJRdxG&>wm0X<*<I<kH^vd%fY6mAuF=JvJM2TzrRUZGI_ymuq}w#LCIZ?BU^oq=#Td
z%J=WG-1qP2K-M^$KPvR|kl+0LIWHNExn_eam!GdM_<mA|TqV%#+d6d>{Q-EpJ3Cm&
zb}H|Qc0V-Bk53^*_%_FgIghcKnIsDExL8S>iqAR~GyzzI5l9RvQP2r3J7e=D?@!3S
zH4?EnSFbYGYRiL-om*Lvfzpr$I#uEB#WS)vY4W3TNSV$MLYmRr+Z!@GVl>TB#(h%}
z#5688R?*V()AenF{-j<_0In4qSz=~+YrbVrwY)LBv#9_`0T3Pv>(z}W$M>0;H{n);
zS{Fc~)otVEPyZPO_v87GtbBY`VyCfOMwoK&NwHJ&cQMXBGFeClCKM2plaH`$1055h
z1e{hrlvsB~-hz~tVu<L2lHfO=BcH^8nwy`$#W4tx?zHlg=@eKp9ZYuPk=B2fJ~xsb
z!H*YeNea2|rH+qlLYz77O{G;@jYNiqhUOk-x~~>iW1#^0l{7Ueq421{x8su&z68gI
zLPE<kEpJuz^rmw?!E0BT_0qy at ->lS>8!6Vyfp{lE1qTM+9aRH}F)=e!0c<%x-}Gii
z85<cH*&m0~i;Xvio(9$Cjo8V8e%>G^j>1o<Dp1b*+S%z3x;9O}K3%-v><UJf2{kt}
zBcfGhP46fA7^08QZb}wR*`-t^hBz1;v{cTfBT;oe%bzTiJmv&`{{90XO at 8yphAmnu
zgda|~TCuJi>g&WrLY2*gM09jC6mS+s#<$1{1B&topLKW7IFwUVA9ISQC?Hv8zd5oI
zkDRkG>I|`PN+^^QxIMHf8?;~xX296eQby2agGBIyyyWJj`{{Z$>%)iIA?laTPxcv6
z72ba at B)SjRCq~vVYdsEa->%;R0fJP at dtR&yA01!H{W>r(ps&TOkW3OYS at R-eY~G=^
zVRE7-B{sqRyb~!I=zRv{%H?HcSg0gHmm;z8X1yQlXa8N1wHdDn@)2|~9XbBL*JS&X
zq=~`7AVL^_-;P#8QxiwBZ+m<DlFx8~2IIp=03;|yP0e^+?<Y^5T=D at 3(jq-=^z`(^
zLWR-@#^`#VJDmVBu`)B~$gO5%?F7&-6A)g$bgA#r#rCz*wj~IQLO|4kuMvzLflu at e
z3?8HQ{Ihv^d4Jf3`)_;*kBw#dK5h5IUb}j2Dnj>Z_e6Q<X=v at eU##gB6(3NJqswDP
zU&07ez0XChp)26Ruu;#RKVOYzCBZ=#QBzZEEd#GYfI^mK8PX@~U5HTvu3I-?<cRIf
zs>lc<-o3kuX3k at x{r_65rLs{0b1&C9_3I!eGE#J7W5azOZj}W?tXB9w<{Mp1CNTwt
zv|ZzV%&38*fV*DQNV%E9hv?|Z`kJSA_p-F;y!HqKv5*b=JGu>lb#(#+gsN=M%55Nx
zD=geRs$Mj6x%hL^)W7(H3H=D+e*XM93OO4SP%%yMkKyw;lmHWtcx12BgO>4eN=C`B
zb{Hr6sIC&bY2mrKxzvgZlH}y%t?g~XWQD&8T-p8Rsqi&qnKooe_2Z_Mxp__TsG*Tj
z1Q at -xfNMse>8=ZLX8LAk0dRFgfP-rXtuGd>vgycx)LIX|Z8+GGjaOQcLmPSO`pm8C
z2CeGG?h2e9ZrTkh5fDbi#ZjR3)!$dvny~ENxnh&iXh_!PGVLV^ibUfs*9Un(#;N<u
zV14 at 9Bui^wQh`&>NMvN>-nT02v2+*r&_MH(z2!?jP3H$%D`4S-E{^_)j at 7*Q1Ey4p
zAMay$c;8YjSWg>)%g3n`3bwX)&~C)ZGI9K~DA_gdfS!8TvJ2imz%Vm|s&EJI@&0n7
z7L+Y&8XAlB at k$2YmtPX>nphKEgaJ_VRM`aJ+wtjXwOh{W!h!_Ke74n3RZUHzva+%Q
zW>TX3kLF+uSCqU*2Z#R`T8|K2*I7UO^*?%lPspJ#Iep^PUmYuFMoR?_*4R8%`~NyG
z<9F@;lu(3E;L?O$^DR_s_4cca<3-W2U{ZE06rxqo*VB!9iC&#@Q#>>mrkDb)>aZf+
zMj23oOjWjJklno>_ow?x4BC88Y{p)tr>C0(1*)p6OIcRNf9vk{bKMYrGWzS}27`fU
z+r8e0Vn_{9Wn+EtAg-aYkuEs%p54?#s7>Jzf-;!Xh29stDI_E$H^(`02L_Z^hVnm7
zx&6T=W+Ff#s}`!wwRm<g2$;P}j}CvmOyXOvc>cZ988`s^Z$?2u0seZ<r%%CPmCVP=
zOlZaxw7$plcF%Z7I4+9ZVPpHI{{R&n8ai~U*WJ}+p}Kf>yek at K%k8&euvA13fNvRf
zK3U&bVRC9w)CfFQVfjBx$s<30t$vec78Fc?{5;0&lDm6k3;DUTwUs)e>+xfB8w-M5
z{Lh#7%yF_3RaI4cADhpfTrU$46ukdw;+KDpTK+QK-=iyTuCCz_fcwjX0c3ZyrS|vt
z|I-6i4Grmwi;KPA?>H&5np0k#1bzShUEs$^#&b2bfo{%bWFw$;xxapz`2TN3n#aw4
zad!4GD%WLmN-hR&#6zLt;J^(9 at E`byi;DmnxP19?KL8ok4{NLLgWbgx?}4kF4aiwL
zq;3xn4<nEZvEr5yVfJ+O?NU!5$Bi34&=TaL751~DmgeS51JXGS4T(@CEygP>3<Cx%
zc#mz*p*|Wzt+9{KQFqM26F6xLD6px}ye)J*7X_1P{FB1b`|s&ALQME{!qyeS6jt8~
zwW`g*%k$XJh$s~Yxj6rzWhhWDee?atk0;U(6&`M1ItR*uNXaWJi{!KY#o`&h0RzF~
z{_y+VXqhJ}Ir6u(w6wnU^zd0uQ-=DR_9n|gk*{!=m&rKzF*+%q$p0AnMM6LzGE$^N
zESM-?*zscd&E>1$b@)ykE4X-^ruaEHWPd9BD*syyserf`e0=;pNP{4vw`%$S7&BQ}
zS##(m=A)}^>XMp`?z|u#0{|UM00xEugUgjG#!KXSo{yFDK0xYS&cLJcKHvZ62-m)?
z+Sj(>o>?e^B-GThU@#pI)-+1n(&pxj9h{w+9zA*lRqEsDY>Dj+X667Ug=AT{2n5CG
z+ at AfgdVcX|o9U(`_k%TQ=tcSz?uVeKg=&L{1}`iukkHV`!BiRs+yrBhZI<E{<0F_7
z-t4-aju at M8@SvBkl7;*EUA~No=^Wq1q4Ojm+v`J2%mA^%L!_+>-{aaA at 6^YA_&`F*
zZH$F<s?4NYU_C-dMg|8MAreR*ag@#V$(8RWs_h1M=DWLwhN2Nm(Gg7gv#;-lv${@?
z*AHHx8%8&G_ekwKImqe0whfl=0s#dD9J5x$V57t!?7#%2`}<=eg9jxkLX|CxRV7=|
z-Mwzqp!7~|W~MK~@@3W4%v4lV?z=zkBj-^K1BOf{4=)`8b|FXG#wI4BK;+JMQ~qXB
z-M6hUvz9fcprGKf8RvnTkoEX2W`DAfM>~Gr4<1Uq^P at lSNeBraqgdG3f`^NAy06F-
zXh4AAQ}c at qX3H@?efkuXH0}MPi&!=`Hn6JQ9y1EhCL7$M$T-wR0QPRcJL2z|$p?5d
zBso!sN{BU|?Y!H{GJ1RES~l1|=W!Jan+T9o<YJJ1k=^q=R>K at Ja&lfQx>QzlA3621
zuL&Rd00c?{Zw3-Fn(wa+xgS0 at J}jOT{-&j(B9^aKnC|0)0{2HkO8NwT8aTDo9)6%x
zWB(c%jQ^ZejAD{Y7Yo1-qEQO~;L3cKuhtJm77JBdUq6Uc1dqdw`Q at X@zOiyMi~jT{
zBxGa~;KC91e0xp!uB!+A!OTr at m&wfr8UjM(Sp}!nVS|Y(8*_8>&zaIucfC#=A3}LA
zDJ!$=-5A9{*2=1ECxh{*_-+UZ(M(NENl8h4{vLO4c6F8A*;NABCOq5e=0x at bfgf&s
z?B0ffHX6v3M$RNboU%Fx79_Do%TNoslhiw}w;rrdL?k3!wX?H(xZO at 1`_PRW(40Pc
z#cTDd74v$s<A$1!&h53I;M{?jUiW868vSk$LJsn~!*D6-lrtp?IN<<Fw)Z3nzWCja
zjqK}w`@o72KWNO22&r~_C7!ITtkF`VE67R?4AtX2RffBF2{I%?MXjwl02*dzW^Opg
z7KBn#QijLG_(Qm7W at gUxKKd)}<t2oPN9ija#WWkloGfB)o;zybOhpq9%~DhI7xFmJ
zU+hU5t+t~h!MWVp+Dbr99sr(5{K=D+&d$#3X^!j=%TjW39kXo#5s8T*;F6C{P7IMA
z at V{Vy_UW6NGC9X4g(>H?ml}NukBoc+?W2c*wXCEBtJIFH%>8HG at b29bIlZ+n0=nSa
z!{{D`?d>^c=HxKmzKsV6iJXdpzR<eO5w^FtZ?UM|lz1xOqLQbIEL$FIOuhpqLKcC;
zrEXpazK1?@IW8&s`uVkVcXyAbIaz>AfBp8YHHet`pS9cv4ws<iC8ed0RaHrlcZ1Ep
zgoznUWBE}<Mdj`Ptcg0WUFYWJwg*@Hb>0E4M$f>24)HIcuYcp at aPu1Y!cwtUON4|w
zySpGOe&uGpI6(_g&+hW_c6@$~K}bOn6u~ITaOV!bt*xzL!1o)<1GOXT6V)%E4Vl^5
z47|Lg;KAN#7W$W$-$TZimNu-U<Sw!n4du?$7(2D9ij167w>?iK#~vo%*w}bxZLD08
zEgD&91b`$YCVmSOf=m1GI+#$yfHTE@%gP$pZH9rtLBFI)$O$Dqy=%~=qvK<Pna{7W
zaBxHw6$#MM(fe`~QslRzf9r&57D^Zx&^tIdfFuVYhuT0H`Mob50vpjsNX<aphX at DY
zQoMosvc3OD{*%hhBo7Gi#0M`i|Fg^uBJ&zPC|0P}*4m2x?;$pVFt*c!^{WaBy#X7Z
z^=kP~q@}TuEr+C}P-Mfy#Drc at kP3?T4TjP2aYN+r-_@(cL=`D1deu#QjAWIXp+Nh{
zrU*Fquiw9mLm)vDQ`Pd(P#_{(`}=;y#hk#P$fnBC(NQWWv;B(~%T<h-fJ+}fd_WFi
z*4Eb6cph`ZZC}iPBLx9YMV>;KnVo%|ntB8G--UMM2qHQLh5>lb9d`Y~A=aocdM76*
zWB~4H*W6h1K^}|h{>XWiDzmWgc0)tM%>4Xi$Yxh}_tv2yQsmTGem;9*Qqt_&+Ar23
z1AtnD<sl<$+;*9e&63tuQMff0vOkln%7*euN}@r=zuDV+)(i=xz2&xhHUbZ#W&EJW
zJnl=*xo&Q5xy8lgD|IzBso)5=wziOa{F<Dc{rwx=%*>2mK%lj&OZsRmPErm68#z0s
zp+SK_=Cs|fn%7n%MQJrP*TG{;LN3Y3$oQ^nGemrDYZK$-Iz2gQ?d`n=)|jE0`yjPW
zcECbLOblaZXUCUjA2|h?lEQhWqpUozltPK3rly{OJ0geY#>ODyF&%2D;S0E~q_Wa4
zC<vQ_gM;$iNJ~?bkFspK$%~3IRN|<<tc*z97fce2#0*rw+)tlGot$_rEiF?OS$r?4
zJbymI!UV|T>gtNsHPGGt^k^(cfv%>m&bg1$*l+;C_tm)okacimBoXo$-oKl6ot0K2
z$b)&H-suGevYWLrF)^a(Wo2b=U{(qW3k?hm(2x^n8X7GF14REGM+=w9f;1Vl`m<r9
znbD85C3NNI&=8bI^qPB&&+=4%Bh4!;^i$3gH!-=jva+K1{5kW at SdR%5%F4>Brl#ie
zj~{qo7ww&#`~m{5Kt=YOYV^oJretU5ji28Wj{?nMX{Y1NK}uwYSb_u?`JY*zZo2)@
zG3fC}7GOV1D=Q@(olwyJug2WScBZ5;3S@&SG~U|UTHn$V2Xt0lORKG?2OkDDynDW+
zs>&Zah3s92)F6sN-Jz+e8C+_qP7Y=eFO1sp(dx<y3?{NgdUE2f`s^8UN)8}8H#Zk$
zzeo}z&R|jTW2#;bqfb!I2E12dQyE+{E9=(U+L~|X0;mCX>ZBm$OGihNSBu82KDLpL
zwCw?OuL1$Mu_50K16oJD>g{LIAu*6M$sx2tnrrsz_dCfkG{d8!25p3sdTDj1z5fb<
zb=m&P_&$&ZXLW&pxnky9H2c=!A-N!K=SG5NVW&SH)m@`b++ at dA@h at 8XS6wsP>gqsL
zkOV0={Xry1(#gNf0~k$BN=o{0z(8jlYA14b5INT?W)Ko3|1$X9J2VtHT(Sp5og2Nt
z9b2o#&A&OlkKc`^-Np2Yk}rp_NAr0U-dR at wueN^shNF@#O9)p+o^MivX+ at p)-o88h
z%=7R0q3cHN;*qd%!Qf2hU`TK<4n)g#z_|Ly56Q`4FBf~Z^LU5z;8POtsrL@*ANS7>
zrx{sTup%NNU_ at f!X7>P<_?JuW<P{a+YgJj#S_~8n&SnncM$2?WVZn7ieDDQU`yR_B
zwx)R9$~Q{>DhZC?DE-}GQ$6o<-fY>pR=uw80yOdhw`We%%8Hm688McZmtE(hpSR=t
zaS0_lUibKloFbv8r at zi+=nGxU4nQ7jizqAOKHF-$7Axpd1iv&z&T!B&K%V;n+-A!s
zu{pDZP_#?uPyiH%HigBP_Q)2h{b~Mt&T0BR7OhtMjx+ziuglBJV>Txwm-d*i4i|py
z?rzI3ZbClH8w8z$SnHAwaH3#WEBK5YHU&0^StyXsHoVdox4+bn0itbUVq)*$KqN at r
z8Sl6@%0(^cjO%@IoCA#-=Ad2X5&LNFsQ2Qtx>=>;vNA4>013!FDNIIu2i&|);pHwX
z-*B8xlP3kx^gj+C?p+xO_jY7Vg1Q~SXi#8*9FT;f_%<imT?QFOLcKu?BzVEhhJXCX
zjC9ozDnWrpIcAF20Y#>=u)lE<on6*#P at T^m at 5npw-*UZ;xuX+IE99<!ur>yBr=Gth
zm2(+{36z8cf^!dMY6foK0zhv72?E1YiSyo3li>RGKv1XBHuFpfN^ETGrT&a7naZp;
ziLg;K;B{gJoY)c*6M?$+Xbm+>^wA)|nJjVrjm^xAwtshA7jQJsa2g{6Y3*!q+a(|)
z!>rZDzk2m5sPAm_bKS=V25Bve$gt3PZna=^G+B>vLoNU9N3M(Vk>3`k(lVPYLd|^C
zdxc;LQXY(q-sa}ASWi?H4)u~-)vwsM$2&nnqKnMW&&v&?iF9g51I9J<ElzqpI$bMw
z at 7^qA^nddPy{x1p71AM?OiX3mNQk$(hlZMZ2`JXR_rHgf`@K4#(2H`%g*2V*JZ<E!
zd)L#T<8?4*CTB|wAz=QD%Fz0GXP;M9-zl<;fm#A*V^<bd`gaU2hJ{+6tp8Y{Sy=LG
zYAP9}2o~FdhqC{7{RqP3V4+pDwAeLq<*;#aGp{{53+KLfZ_`5vkCG=ahEq=-peH>(
zo)QQn<NWk6s=4_QT53%Vf8(O+^XF0E&l^ux@()Q5wd at +7064)fnK_70z4)qn>U4B!
z9sIx>W20_sRs=%s=Ya`Fc2ZDZfBi}XrPs2S3|aqny{?z+1O=+R4CTVM?APeuwYrN9
zdHzC;JN^)QQL;=KT~W+b9CZBlH*V`T#Ny!Kgiyyd9q1~U^(4}x08dOxJX?LSZ+`4X
z9tM(I2TzS`&TuvzrT6b_d~nnpFVf6NP4yw;)RhI0wf^!o5b;eY#3s9&Af5=dcX7Ho
z?Q%*}sY7uULf9`&EZ{f96K3~@aKO}0UXsb_p`oFpiRlYWW at hHt`1qCR`$J7dYg6uZ
zRs=t8+jaf-9QADCHiVPnvuC${D&-Fy?F}jr($eCCEXT&hwSYuzZ*NP<%iC6+|M$Q&
zW_NFrU;uJ}4J=GxSXk$$XZf`8f+B$D|2!s#8}p^JGrg{k60C)_O8r&xJH81CR2K);
z%@iEZSx91qJorE?koW_yO$4kQx`OO)gBkJxC4>vi_!Nw*{Lfi{xjA3tN!-5|=kDuy
zLzZWYkN!S(b>*LPcwo#;0eDT-QK+0(V%$aK>+dhh6!n-Xik^puq&w!e82poQ;5d;D
zV$te%6?rrRpc*+F0>#Z{NWk2p-crWR?SWdMrr*+%xvPiApCp&5mfDw#<3A;yf at wmb
z?wNk?V;{UN=y|l|RXwdV5#1k{v|slIF}7!CXSgwGq$7%s-U`UR%Fe>LU0(5m;oYB2
zYWZ2f)O^PcK!Tq>ePUop2UeawtV~}X%ppbz*w6U1?9d20_m%ewKLsl!>S1bX%Jbs4
zxQwi<_4d!M^Mdy=<CdJr)8A;Rl5}1_^^akCu~0t}ACwQ-Bnx}RL)AEkiE=lQvoS+`
zZEJh`?c2A#m8nzUk_wv%zO-bStaKPNo_gVEnOjdJCGTr6-?<Z}$`-9V4Rr@|mHF1K
z${Jb-vkf&Zt*1djLCD=9i-=Dmx{UYM%6pNg;NrDE9+vAsX?}}Rl$H+Ctq0~~M*RdI
zLiWd2MiSB&iasfFjvQU2qoV_}TWefn`J2#x_hnWVso1POz0!&))@_VOg;!VKhY$*7
zV`rDnQdW|Xzzn_4^#+9;;tmS*_1)@XPJ!ECL!!ewGLi&ckTsxq&q;BYqBxZK_FkN5
zTI&Al$e_`CAqgfXCeiPFe0(3m!$rWN->TG{oU>ZBmPXbZk*8e2qa#aTB|QM0*eJZ~
zcT<Pf!)S%#{jbp#x!^=E{r$4Jy86_@!UB1U at JkpCBQOEF$nL at q;*~I}`(aSm#gI)J
zwftoOL}a-OU4)R3P_%Kr`{UKkww0m$WvJK4XC>d$lFHj4N$w*^UR`ZHCW_<t2JEcO
zRD(uMPYvi)BY$IK<1+X)WFZ%>wg>Y at jG}ztVEk at 1Eo)t-P;C&fkCmSO^_)X(Urt^g
ztI)xyQN^oscV!k5xvIZ6CZg|@P9~W<sj`uxxfHGbdLK4k6atFM&cnkK4!B>aQ$MWl
zW);=5F<C!xb`~Zh07!AkXJBApZ-tzcln}LZc*x8^<gAJ8N+28G$g=vkt}d=K!_BnQ
zfRWI6&5VkQ56I(&^b(VQ9*5y;Wi&?0gC(l1t0O|WA5OX?6g-D2xz?v;)SE1XY}+GG
zhpHF7^9C$;@5K;-GD{xw|JY?R7Xos}*%uWR#S$sM*Z#k59&P;{uRpDc-4Vi{Nr|&?
zkGKBRojx8|?}jxtD&r89GfJY?l|5SUrG(sNbFy)jl+%{A;MT{$Csd+=JBs4sawbA+
zHpFE0IP+<d$=PvTn-{Y^+8L7h%d&q?Up4luRtheToHdyx#%&&-_Z%i-+!Yk87kH7K
zm6ccVY|Z7z at x%D>Vgcv1+%h?Qe0&ZDhBx4|a=Q*+X|2cztc(|HQr${2MUH;#?(Y}6
z=)8=QH!+!~`rB9+8J3ntkMa2uIvO$7rI6mmd%`Kbt`C#T8WmZLL%3q%<G+D+tWr<b
z*4N)ZzO}G>jO!!AtD^#>(9UQ3>nk-4&4l(q*VnK5;HGg=3A|Qnjf_Noz;u{C+s9{2
z)6egGEN}8U<+iii#GGLH?d>H47DV}0+}cF7!u)h?DWgjVOLeKz>!v1Q48_f$)twy&
z%-rNXwLSW%F3Mm2**ukcjmeI1n=z>FTDO*SjojsQPD#mGmU#xNY$jq$OSz(?ZaNL~
z-d}E5c8`<IuFBk9dGKN3><G+*^Tw|SRqp$i3xEE!g at lGupcdxmhi06)4*oI at Jb&u3
zs-mmgV`yj?TVG%QEuuOif-nJaMbK?0vs0lM=9Mk#cx|kl>fAtoZNAojfi}W+wB$A5
zXDJBc{n6}NX-O{LO8Z$+=iLRl`R35yEDe>B@<tPVAF>bF<77MKlOJ{gYHFIf|CpRi
z>S%AzwQ;ur-;SFr_}e(7bZ*WVlgAE6>2cD3y7TAD`#6ys`cS!?hFQZXsRULmSKehF
zDL1@&JaW9dI27ZEiBJ6%RRTPP`N#76WW5W;){e2Y5y>NakemG?1t}^0rlzKEoTOV0
zzVoxQEr^a#)3BM#42;#j%;DDPNw{C!MTgm`8?%Nk*V{-=LQ>+qK3+OAGb1s<`rFRZ
zU5 at K~_vG3E)cF-pOlN0jT+~FlnZnz1w(OA=c2r<!sENq=pjD`im<mLj?lTRIF1STp
zK!9YDC_2<eGR~XkU6B?}8_%`VGIuedFzMPo2{dH!Ug}G8Uwn~p!oVv3>-6ZT1dKgX
zgp|6B49 at C!W&U_%4^-ysTV*x$0s`^#i;MSQ at Bkp|!5F>^4mJb?58>+6*49>+mBnRe
zVbOP}O^p8Vp=7i9Z&MGUwHL`V3A5hhLYGDcKHB$R{BbE-85kMI26GgsycC>{Go;P-
zSBAbd>(tmAl>aTNeI}u+tNXpPGkZ$uAfJS552OItg)2|>8)!vbWhF1hg`QMj)>Cp!
zpUYRSl>h#1sP^nx2N4m`hsCS9=eSL$8-kcV6jW5 at Aa*6lt12pO?S?O41n%9t=M1Q+
zrliyYu4fdY{sTGp$aAT_Q9homyz@`=3Ms-SP{*vlhSRgDgxp>pdM73(D!S6hzt at u_
zDA)A9?vTAmalj2jtJ>B?LPBB<CY`MqKGfKsk&^Aa2GsQP^bnZb@~J=wSCxLt8&jGV
z-l|u_1sb8ROaWnVQCjuRHjS$`Ha25GFZ}iVdvcNzq30evRSzMkdpua#+731*C-gl7
zuF=_6DCaftLsC7@`TP87_S?5_S3~<pMyh)bK*{)k4J01X1&O!9uv1$9Sy}m{TslT1
zodex?05$mw;6+vp;bWuf7L96Kouq at pM!gq7kv&u_?CeGi5rH;iWl!e!VOB~3Jh<7}
z*`sIcMPD{a2|BOogPk=NO$!I4E}NOruS_0Pl9HA-h+>j61TQAbLc)!Kfiaw;K*wh{
z_0Ys5L7l at CfJJ|@&dEqXKtT2;QR!=pD`g;>BOb}~&tU4l_w|J%yH?|2?AlDovlUE4
zL6i at xx6Y1Vni?3iWGkdxNK7cJsda)kj|&RICJqw+-rfClzKR{<=(b){qK3Nq-899o
z1%-tg_V#!4^YZdsP60-=%1!ZBM~dkNM at Gz`OCTJdWS>gkBm%I_wl>n&{|v(SrM0bX
z`h at 89iTTN^j?r!mOfxexxP7eDO~am^p0%ynw*4xbALqeV9~`F?_4RL*S643zgz%Yc
z{Hld!D;n8#w_NTJDtq(h4W;!x?+ef2G_fl^*;KqV>(8Bd&6<Bx-b%XvAt~txs^sTS
z)gm#Adn7Ix7u#RxYVn^KogMvoFRP$X+(JAHYH#$A_I*cZXO)Ao%@yRsm3kKFv_>JX
z8wl^;zrVDy!obQJw5m8BV%L052J_;l&aogs5<0WA6iCP at F@ya50ZC5R=!B8K=m>ep
zTwGG}C4tVw<J0I*od at i4kEIu{0Wf#|=b50#Z!urc;P*3BR#skr;Bb>x*z=!-5Qtgr
z0CA~z>6yD%9taAGDJWcnTg?lQq(1 at 4ECJ*Q>@dW=e%EK!9?Rmzp84BH=Ii5?{4N{R
zkpG=dEaFeK8(awy$ORO>MV(%0p1Dir>FH^|zw&HuZcao`FJ;wUipu%#g%>|R|L3VG
zT0j+>;qq`Si+3!`4SrXy(ITsVM?2k|rEs&NHhLRX4GjtiVHAq|j<yeI#TP=tbYY&x
z12RxGLL#C!z*3ByobQp#g8#be?d^@Mpn>Fnun!#0i0IM!R9NVjk-><Ii;J8jjwhqH
zGoEU2L;L~Y at 9fOXXD5E&T$)hFsYsC<BH%%}xw#1_D7u{ZGx5e6-0oFbkG%m5m*JmF
zdMurD8Cj=>{<Zz?c!!+1Mp}rh(v#iM4uo)!jpqqMhK84?aai=a(a~I#0{Hg;_UA>P
z+uMCt!vp<gz)b*hB2Vj9T8%i+;L{fdN6KGiyK at J57}DCtCL%UgYSBfC225EC@(?4y
zIU_fB<N at JMPug-NnX##<KiI}ZVK2>$O)v2ps77f29+o!(Q2b#3RWL|Gz_P at wm-f=7
zOZuk=>qlq1y|wrO<~~Rxz&L at q41D+Qt;mfTft#ZIX>CrZ at vptT*Dhba9Q*IXEi=o@
zT~1QIVj3D6$iZN0YHH-X2GV~CMjBq^*k*cq`VwV`%P3SgF56%HHFb5rfg~O4Z|?MD
zWM)W&a2=hV{(%|b_d5B%TDa~&s at pGqjqH_5GO}JGD-mzVxOmNLMzRuRMcJz)T&~TP
z8!da3n^hDcn~acKq(n#=$qdQ(o#*$ye^%e_z0dhR=QEz?oTKn#VWE~St||}p-=iBN
zrK^0v*MbL`;BPtl%!DF%mV;G)e5j<6k&!_&vIZ#>^nBXpW-uzq1~x=<mqhunyM(ml
zd$W)RGBRwa2D+|J#oWRIo_#a|a5k~9kaYf;%SBkQKN}(SoavzZJ}|K#>Vjm&j50BR
zDWSaqyg*q?tNrRqHEL2)EC9osf5!KclX#WP&npO(=N3>5KtutbqEn#YkdTYRW%iH_
zySK_;Z7~6tip$7Yh1knvpeixca)^cwz&lSC{!9CrS(m@$<m&GaF9Rf%1Lp9cMEQb#
zjyA_1lALj?NqXJe+t#?Za}9F~Y(J_vKbk^_k7wH4+!U0SW+kxkM`mOk?j&;N)jkSM
z5<ypJpjza>{QMZLy=~xr>VdX!L}{rstdqXhj!i8)rJ~}*=I`Ilt*tcC(b4F(E=Mz}
zhK<CL^p8G4>wDkBR*3#vK;X9a_FRi%V^{?bvZC7B+8Qz$k{3WIZc?&$X)P=)P(3~%
z^VA6}_9LBvG9N1FQ9X0U?ELvSV5_E)k-PJ$udhH(e3_h#JS^+*s9uJHA*Z;Qum_HJ
zW~@O)_wD|{=cYnbouTVL;!s29m!=4AMNe*|Ed%tv9s<L^w7N>(sJHd<k_K!1tv{Px
zIZ&h^M10l_<LvB=?xj#d<>@4#$wwt6$IDuUF}uI}!zXul{yZuz1zBJu)(VQI!sX39
ztK`Jg=W#{D3cZjQa+kXg`p*ndf!AukMMYD?rjNeh-x&ju at ImQ!?i?QmvNAgDd*b9t
zbB8ywoN<DjagI(-)UfhcZQ<&U;N?jj7JlWbk^D?F0cK5H2$z%Q&MNQwRnM+2kac%=
z*W1?@wFE$dTQ9B|>*(x*-Cq&B?8GMQm1!r_369Wven=V}6rjE(O--tm)z#%M%{1uL
zzP^6D^v_!gB?|nLvi}SQX!OQXGktd<{NkC<bsGZ4Mqub5JT%$;*SyndFxXs<lXqD#
zOB54zvq7N(#&mxLbe9jNAGKXH?krqG7+z*yH&E%UOA1uqUZS}pbv_b6R7grH@%DUJ
zcTW$xZi4q9bnF;Cx(;!0IO^fyfx(~|CTg2sXi$qtb2Ak%^@&rbEM>s6a>Qw$P*kML
z-t&FF(XcUQF{b<tGP_$~2bBETbLS*=_2Q?yZXZL-Gd5;J_mlwB8e+F2I;V24442JN
zirLYn)b;Dv(Iw5`;M1;!t#9|QEdh5bGZH$dGzB?e*X0K-j-jjIjue5En>$<8UpozA
zhw^4mu^!_}SRQfB&VBik$IQ&E+jX@*c$s==Xs9M=@#y9aHJ82zak%@6k9X8A=;-MD
z9WZwgAVP>M!DnYH;N8$r9)z- at x>{#(tO1SpL!%18fj_VdUz==a{`&p9h?3Ib#_bhm
zUq3&RAr8k1tT6$U*&IcW=KIZjdQsjzo~<5!v^9p+!k8?fMu9!v1o0Ug8{<~LsS1+9
zHtZ9F`@HyU6%|Mo#7z77`IFo!Md)G_lt$)>i;>;zQw93BZMM0#fKT;IOmxl7In&b8
z(B%#6GKflvl#7dt^ic6jbV1Zjl6>9W9ceA~PiUy}QPp4!Gc$8dT^$--=I!GXQCoZZ
zRz^M#KR-VUuObVuvIK<ef at _sNLx_C3|Ho83p0 at YF=>v3hbSD%P+DdI-oGl6m+(N?z
zwFqnt?-`=xV%FCDe~dQ@`1tq)ug&b=T3=ue3JOBs0m60Q{d=!C+~g?gqImEiv7q3u
zg2F;gvD at er4MMjGuMYK^)NhPkw<GAs4&@Y3I#h3Na|42RfEJvyFVD}9g7{5+sC0H4
zF8j>e5OndP5IDJ*g-<W?b8;{Mm1sgC(Eb2W+zWPx)<eL^b&IdGU at yW}&Vm1rMj~ix
zYNFQQTDBC9QVsoL)|K`3j<@-`Buk6(15DBIYGjSG)mSYoC}QI#Qd}YsZiJMON-n*b
z5%2;gTG`b-X9qqd4i1{5eq`I{p3Nl5b0^#e1beg91%Nti<hw=i%n-E9?d==Zwzf((
z9nAGcs^GhHAt$0X{0&V^I#;H9h-qomy1Kfw`}U!py3o+jIfjNe=A-t?z%}+S{hMEm
z-%Nlnd`r<_{7o~HOzYJPs+&~Uo<!`rRO9!Z4gB2FpT$P<lb)`w=9#zo=zd?X(7d3i
z2rI?qJooN#y7wH~b at 0V#c!_sVkglmIho`3}V%9P(k%Bk5L}OlaGqdH6 at 9uc4kP!~&
z1I8Cp)u}!HkoAkJ{Ayb>B2R3pnSi}L;;pX32eAX6qXLPEZ{K1dKN=_&M0`D?fwI@@
z^2XhGcp$%`f&;9F4!oQ!)etlcfn?h$&Jc|~(UJaQYH_qyzJyFhV-`RLDP3vP=DBnC
zZro|VZ>Z%E%`ZZu&q4(S01ylEMIGvcBZ^J3OQawNgIB&Xq6+z*4B33Y?|tZe6r{`W
zga>P_T{0>)nH?-dl_7cg`JdWO8REU8(~`MBlsBhKSa|>VAcrRqrpm*2B>CgW`YtYF
zVkTKEKxwcKKYaM$&Ce;s6PK2Jr=+B$1<!Z at r4V8WaFBL{f?x3o2 at _LOQKO@FAb?)>
zEd($KrTbT<oju+Tn~`+*^!nWUg!$C5BZzd?*JG at uI@%Ks&ptc$^%Oj23NfFYn(~=w
z*%z9465Vl#;qKw at D_s*-QQ__A=m3_C?SREz27$8Fe)xjmsZ)od=-I6d1>#q+Rr3l@
z^75#Ylan`Bv-c7~o`Cezb~b(UgiAtFvNcWAu;S`x<!VvBK~)gZu_-B?cjaAEQ&W+;
zbYtO}bt8svPYLMZa5`32cWlV-+zKc+a&vQI<Ki>{RQ>&D?#AIw2j6l+d}oGAc~SqD
zS^htdt7Q$8-y9?AIbOaT4O*i&Q!Wbh<pdwD+&vC{TVB|ob#6H)ZSt at 44i3ppO&Z+1
zy!rNZfdER_hkxqrL5u^1p&nOLas1u8NB&9TK|Ry&L)76_vGOW90$Bq!M&{-U+1cBJ
z^nBzgLNY)n?J6oN+&nxj0|Ok8&eqRda1ZOQt_G_v1*^WE_NAq#Zyg=w1A&F6j0XN3
zeSD|?YrxjlR;GeGC#WOJ at tys7%iM37!hiQ%u!gvk*Y at 3_a?Hxg>fW=G0{iu(rl!XD
z)9_(%`^kK#^E{7{3`QQA<@0#<b31d!;aP!^n-u|)oy{Lt8IQvfemz(B{-Qy?;Q)5z
zGIhi)K$L2OaJ<_|h- at 7BT=`^vikioJH9tQ;RFrb>9uMf`JaFMP;d^S}@JL=AiAT^3
zs7cuDsXf!;Qc`V2xcg1qT6%i1Pwlm}|KFtv(hrqM1#hiM06e4;i4n_FoirHKO;%N<
zkeQh&sHn&_Jv}WhE898NaQ*G8u<MZGPCHw_0w(LdC#Vrz#c4;jv|vGwOaAd>Ma`9<
zdQe%d)ocn!Qb)j$$hGn(PZFsdkC|k7l6IFGiCACXr`_}L$4eONa)*ZW^1_0G3ZJQt
zO87-`KmsIU3?SK4d$4s!FwcSQ=>9B#r=@8EAUWsyIqS at IZj6ho>$zin at Eb|bHx|cT
zG1ed<$AUlYiHnGctOH6`*+8g{AOCBB8vdD3SSZ@&B*(|k|9&hxFEyv6gbBE$dzf5O
z0^%sAygX%Y&I7TBMUjzv)fbJ*KG+11P_SwOk!aT7#md4WNYpbmGovEv6+AUw{y-0Y
z|E#6u<2H%gU2|uXs{`fmAX?B+I8nnpRp>Pxzvn=q=2r=%P3dfHC9s3}1zeMC;%;q;
zX7*YAh69-qB}e!8?8g!xZ~KyMpl-xC0KF3mY*+;<QGV$hGC2VvG69;f^y6YQ^Wn3*
z>Y$x51Esb~7}@MS8qCDR#O6{fdv19-I|dx7-NOy at QBrLsv&12G0~OHH4+FqtTZ9PU
z%&n~%?i$g77%S5f!utU-pnIUc2d60|VRZZUp&}g+VPJn4n3$+BiM&c|m;rG0n5Gg-
zGE$Zhp1;JzoIX~$2aNTMj|*T<TpLpbrG7wER5aC~5Q%+2E(`|U)M6Zej at F_E(94%a
zk4s8cqQS5Cl}rkg#Ps#`Bf!p{>5S#L#l%>JM%7X%6lc&F9*fL-Dw^t7M{EB1_f$vP
za{bpYONnTgAD6q<KvmtBx*O=@<6e<OCQG8jjsvdFp9XkTRaG+^8^g(KI<ZAy!$kk$
zWSsw4Eos|$v*F_(^KHir5~&gp!o|s{MO%3P>dJ}}C^F0Ew|k~Nf3HmUwzNbtMq^#P
zyxv4cMV-gta1H8P=8w}$&p(R+rI0c^KQB+zYXg)xkHtoI2gXE4OMK$y=B at xOSGlz0
zO~S8wJV^42dinWP*nM=unwZE^y`iILtMUGd%_Z2p5X=@}CZ+=i4ghlATDgaAKa)j-
zwB-TzgcIfdR#8E0@^|o(;F){(oaN-?kRM%HSxIj{gs-1}l&X=i^siGy6Vlw=#D^#9
zDX3-!%@_VP_dea5A!J_?pgaT;qz7sprKO|u<?mDZ4ul3ENsGn=0XF)8=jPF=xB-i{
zqt~>tf~0f2bP2zYnX9R{mqj3aTW(mWK^`=A%tR}pb!dpIE@&~vRQxO0Wd)`)mIj8>
z4)xrSF{rK#<USUQ{k!qxg- at blV#0!g7_hO(UT0+;9TwW3=QHylnd`nvkc^YFb7jb?
zOHFO<1V|^KGSXMy)2OEt`OKBCU8pG+OfHXxzHSdLOZ5n9U2}Jr`0?Wh2R}c~WZP2{
z%xKiJXYy#mb7m&D=g8COqIV4Lw4x7#H2Sm7fFMS02v&?Lv_e8c><I6REYF at rbGA@*
zAczXom<`4j+?Do}*<gq+_mLw=!oogB<x1$($nRUPa!Q6+61l9b1I_h2dL|T&yCD*Z
zHGpoYR3g9)Uj?RQOb)!Ewv`nZcp>j^y+?q!(LUc?crjkXhRe6KvJ!?*J`=ju^xp_&
zUZtyF5 at cWHBzQB1(D*;>Q!x^|{a;fy*osad at BIf4;(_oY`irObV^Benu!u<Im0_9J
zmrnNHJH?D at NyftW!jM?P<rQd|+S(?xwP|~f)gO_NkU- at _;H+<ig`HaYtuSY0V-t at W
zM!*vw!rl1!S<bbO4`MTTW`QBZFx@{k9$4S$&-wytp9a5n(bbh8VNujXp;(anZB;{7
zXdydM*82LMgvfy(hDB(IsAFw7MET8SJ5Yrrl5TCCom22SE2^s6K=n=h{F$7qMK~@n
zIK*Xn?i`w`2fqy&w&|V>fTa{zA9~kIYioEQ+~JFyc>&_HsV-ezU7G(IhSaRAWZ*9U
z#W8*mJm5Sb6=_0$%@q>W-+{A0@%5e^6h)PU^704HmDxoYiD`|$3Paya6wz&IZ`b*)
z=Y@?8H60YX_5^@^WMl-{k-$JjKzvYv)F8BAXScwvZn(bj->_LncXviZL&Iu^!hUWo
zP>Vl5*AXHksYh%5=utJ{%^hXPLO^_Q{q)G>oj%<>GIE5CjcsPIFyW(<F#es6KA|cW
z(quB@!p&ynEAHP<0F5^KGJqNUEyh7D5S>EIfaL7$?RE6<D6#ukeT6<OEDQhyg8?-1
z4+x;DSAEQ0&sVA%zRSlMM+;agE-j5JD8bJ!Z7w?>Ja|y!IEfnL at 9&SqkXqQ5cm|L#
zB<P)=TpA3W;<Fspp}^dzE;Tzl1&A^i1aNzMdp919nnGUI(h|)m!I(NYu+`eJVg7G=
z`{`3UwO}XY78e(%RTvo=A!m(Ddd1VHktEWgm;STV5gJX?E(cI at 5;(AW<XmM_EQ1hj
zV>=fHgAqiv0HrZ$Y(n@{E3JPYcuTb;Kdi}tWe8E)K3tKWdQ$O$+9*7W(bqB7F2g!T
F{tpk90)YSk

diff --git a/docs/source/index.rst b/docs/source/index.rst
--- a/docs/source/index.rst
+++ b/docs/source/index.rst
@@ -12,6 +12,8 @@
    :maxdepth: 2
 
    metadata
+   pkgutil
+   depgraph
    new_commands
    test_framework
    pypi
diff --git a/docs/source/metadata.rst b/docs/source/metadata.rst
--- a/docs/source/metadata.rst
+++ b/docs/source/metadata.rst
@@ -5,13 +5,13 @@
 Distutils2 provides a :class:`DistributionMetadata` class that can read and
 write Metadata files. This class is compatible with all versions of Metadata:
 
-- 1.0 : PEP 241
-- 1.1 : PEP 314
-- 1.2 : PEP 345
+* 1.0 : :pep:`241`
+* 1.1 : :pep:`314`
+* 1.2 : :pep:`345`
 
-The PEP 345 implementation supports the micro-language for the environment
+The :pep:`345` implementation supports the micro-language for the environment
 markers, and displays warnings when versions that are supposed to be
-PEP 386 are violating the scheme.
+:pep:`386` are violating the scheme.
 
 
 Reading metadata
@@ -78,7 +78,7 @@
 Conflict checking and best version
 ==================================
 
-Some fields in PEP 345 have to follow a version scheme in their versions
+Some fields in :pep:`345` have to follow a version scheme in their versions
 predicate. When the scheme is violated, a warning is emited::
 
     >>> from distutils2.metadata import DistributionMetadata
@@ -89,7 +89,7 @@
 
 
 
-XXX talk about check()
+.. TODO talk about check()
 
 
 
diff --git a/docs/source/pkgutil.rst b/docs/source/pkgutil.rst
new file mode 100644
--- /dev/null
+++ b/docs/source/pkgutil.rst
@@ -0,0 +1,143 @@
+=======
+pkgutil
+=======
+
+Introduction
+============
+
+This module provides the necessary functions to provide support for
+the "Importer Protocol" as described in :pep:`302` and for working with
+the database of installed Python distributions which is specified in
+:pep:`376`. In addition to the functions required in :pep:`376`, back support
+for older ``.egg`` and ``.egg-info`` distributions is provided as well. These
+distributions are represented by the class
+:class:`distutils2._backport.pkgutil.EggInfoDistribution` and
+most functions provide an extra argument ``use_egg_info`` which indicates if
+they should consider these old styled distributions. In this document,
+first a complete documentation of the functions and classes
+is provided and then several use cases are presented.
+
+API Reference
+=============
+
+.. automodule:: distutils2._backport.pkgutil
+   :members:
+
+Example Usage
+=============
+
+Print All Information About a Distribution
+++++++++++++++++++++++++++++++++++++++++++
+
+Given a path to a ``.dist-info`` distribution, we shall print out all
+information that can be obtained using functions provided in this module::
+
+  from distutils2._backport import pkgutil
+  import sys
+
+  path = raw_input() # read the path from the keyboard
+  # first create the Distribution instance
+  try:
+      dist = pkgutil.Distribution(path)
+  except IOError:
+      print('No such distribution')
+      sys.exit(1)
+
+  print('Information about %s' % dist.name)
+  print('Files')
+  print('=====')
+  for (path, md5, size) in dist.get_installed_files():
+      print('* Path: %s' % path)
+      print('  Hash %s, Size: %s bytes' % (md5, size)) 
+  print('Metadata')
+  print('========')
+  for key, value in dist.metadata.items():
+      print('%20s: %s' % (key, value))
+  print('Extra')
+  print('=====')
+  if dist.requested:
+      print('* It was installed by user request')
+  else:
+      print('* It was installed as a dependency')
+
+If we save the script above as ``print_info.py`` and we are intested in the
+distribution located at
+``/home/josip/dev/distutils2/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9``
+then by typing in the console:
+
+.. code-block:: bash
+
+  $ echo /home/josip/dev/distutils2/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info | python print_info.py
+
+we get the following output:
+
+.. code-block:: none
+
+  Information about choxie
+  Files
+  =====
+  * Path: ../home/josip/dev/distutils2/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9/truffles.py
+    Hash 5e052db6a478d06bad9ae033e6bc08af, Size: 111 bytes
+  * Path: ../home/josip/dev/distutils2/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9/choxie/chocolate.py
+    Hash ac56bf496d8d1d26f866235b95f31030, Size: 214 bytes
+  * Path: ../home/josip/dev/distutils2/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9/choxie/__init__.py
+    Hash 416aab08dfa846f473129e89a7625bbc, Size: 25 bytes
+  * Path: ../home/josip/dev/distutils2/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info/INSTALLER
+    Hash d41d8cd98f00b204e9800998ecf8427e, Size: 0 bytes
+  * Path: ../home/josip/dev/distutils2/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info/METADATA
+    Hash 696a209967fef3c8b8f5a7bb10386385, Size: 225 bytes
+  * Path: ../home/josip/dev/distutils2/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info/REQUESTED
+    Hash d41d8cd98f00b204e9800998ecf8427e, Size: 0 bytes
+  * Path: ../home/josip/dev/distutils2/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info/RECORD
+    Hash None, Size: None bytes
+  Metadata
+  ========
+      Metadata-Version: 1.2
+                  Name: choxie
+               Version: 2.0.0.9
+              Platform: []
+    Supported-Platform: UNKNOWN
+               Summary: Chocolate with a kick!
+           Description: UNKNOWN
+              Keywords: []
+             Home-page: UNKNOWN
+                Author: UNKNOWN
+          Author-email: UNKNOWN
+            Maintainer: UNKNOWN
+      Maintainer-email: UNKNOWN
+               License: UNKNOWN
+            Classifier: []
+          Download-URL: UNKNOWN
+        Obsoletes-Dist: ['truffles (<=0.8,>=0.5)', 'truffles (<=0.9,>=0.6)']
+           Project-URL: []
+         Provides-Dist: ['truffles (1.0)']
+         Requires-Dist: ['towel-stuff (0.1)']
+       Requires-Python: UNKNOWN
+     Requires-External: []
+  Extra
+  =====
+  * It was installed as a dependency
+
+Find Out Obsoleted Distributions
+++++++++++++++++++++++++++++++++
+
+Now, we take tackle a different problem, we are interested in finding out
+which distributions have been obsoleted. This can be easily done as follows::
+
+  from distutils2._backport import pkgutil
+
+  # iterate over all distributions in the system
+  for dist in pkgutil.get_distributions():
+      name = dist.name
+      version = dist.metadata['Version']
+      # find out which distributions obsolete this name/version combination
+      for obsoleted_by in pkgutil.obsoletes_distribution(name, version):
+          print('%s(%s) is obsoleted by %s' % (name, version, obsoleted_by.name))
+
+This is how the output might look like:
+
+.. code-block:: none
+
+  strawberry(0.6) is obsoleted by choxie
+  grammar(1.0a4) is obsoleted by towel-stuff
+
diff --git a/src/distutils2/_backport/pkgutil.py b/src/distutils2/_backport/pkgutil.py
--- a/src/distutils2/_backport/pkgutil.py
+++ b/src/distutils2/_backport/pkgutil.py
@@ -1,4 +1,4 @@
-"""Utilities to support packages."""
+""" Utilities to support packages. """
 
 # NOTE: This module must remain compatible with Python 2.3, as it is shared
 # by setuptools for distribution with Python 2.3 and up.
@@ -12,10 +12,17 @@
 from distutils2.errors import DistutilsError
 from distutils2.metadata import DistributionMetadata
 from distutils2.version import suggest_normalized_version, VersionPredicate
+import zipimport
+try:
+    import cStringIO as StringIO
+except ImportError:
+    import StringIO
+import re
+import warnings
 
 __all__ = [
     'get_importer', 'iter_importers', 'get_loader', 'find_loader',
-    'walk_packages', 'iter_modules',
+    'walk_packages', 'iter_modules', 'get_data',
     'ImpImporter', 'ImpLoader', 'read_code', 'extend_path',
     'Distribution', 'EggInfoDistribution', 'distinfo_dirname',
     'get_distributions', 'get_distribution', 'get_file_users',
@@ -24,7 +31,7 @@
 
 
 def read_code(stream):
-    # This helper is needed in order for the PEP 302 emulation to
+    # This helper is needed in order for the :pep:`302` emulation to
     # correctly handle compiled files
     import marshal
 
@@ -79,32 +86,34 @@
 
 
 def walk_packages(path=None, prefix='', onerror=None):
-    """Yields (module_loader, name, ispkg) for all modules recursively
-    on path, or, if path is None, all accessible modules.
+    """Yields ``(module_loader, name, ispkg)`` for all modules recursively
+    on *path*, or, if *path* is ``None``, all accessible modules.
 
-    'path' should be either None or a list of paths to look for
-    modules in.
+    :parameter path: should be either ``None`` or a list of paths to look for
+                     modules in.
+    :parameter prefix: is a string to output on the front of every module name
+                       on output.
 
-    'prefix' is a string to output on the front of every module name
-    on output.
-
-    Note that this function must import all *packages* (NOT all
-    modules!) on the given path, in order to access the __path__
+    Note that this function must import all packages (NOT all
+    modules!) on the given path, in order to access the ``__path__``
     attribute to find submodules.
 
-    'onerror' is a function which gets called with one argument (the
+    *onerror* is a function which gets called with one argument (the
     name of the package which was being imported) if any exception
     occurs while trying to import a package.  If no onerror function is
-    supplied, ImportErrors are caught and ignored, while all other
+    supplied, ``ImportErrors`` are caught and ignored, while all other
     exceptions are propagated, terminating the search.
 
     Examples:
 
-    # list all modules python can access
-    walk_packages()
+    * list all modules python can access::
 
-    # list all submodules of ctypes
-    walk_packages(ctypes.__path__, ctypes.__name__+'.')
+        walk_packages()
+
+    * list all submodules of ctypes::
+
+        walk_packages(ctypes.__path__, ctypes.__name__+'.')
+
     """
 
     def seen(p, m={}):
@@ -137,14 +146,14 @@
 
 
 def iter_modules(path=None, prefix=''):
-    """Yields (module_loader, name, ispkg) for all submodules on path,
-    or, if path is None, all top-level modules on sys.path.
+    """Yields ``(module_loader, name, ispkg)`` for all submodules on path,
+    or, if *path* is ``None``, all top-level modules on ``sys.path``.
 
-    'path' should be either None or a list of paths to look for
-    modules in.
+    :parameter path: should be either None or a list of paths to look for
+                     modules in.
+    :parameter prefix: is a string to output on the front of every module name
+                       on output.
 
-    'prefix' is a string to output on the front of every module name
-    on output.
     """
 
     if path is None:
@@ -162,6 +171,7 @@
 
 #@simplegeneric
 def iter_importer_modules(importer, prefix=''):
+    ""
     if not hasattr(importer, 'iter_modules'):
         return []
     return importer.iter_modules(prefix)
@@ -170,14 +180,15 @@
 
 
 class ImpImporter:
-    """PEP 302 Importer that wraps Python's "classic" import algorithm
+    """:pep:`302` Importer that wraps Python's "classic" import algorithm
 
-    ImpImporter(dirname) produces a PEP 302 importer that searches that
-    directory.  ImpImporter(None) produces a PEP 302 importer that searches
-    the current sys.path, plus any modules that are frozen or built-in.
+    ``ImpImporter(dirname)`` produces a :pep:`302` importer that searches that
+    directory. ``ImpImporter(None)`` produces a :pep:`302` importer that
+    searches the current ``sys.path``, plus any modules that are frozen
+    or built-in.
 
-    Note that ImpImporter does not currently support being used by placement
-    on sys.meta_path.
+    Note that :class:`ImpImporter` does not currently support being used by placement
+    on ``sys.meta_path``.
     """
 
     def __init__(self, path=None):
@@ -232,8 +243,8 @@
 
 
 class ImpLoader:
-    """PEP 302 Loader that wraps Python's "classic" import algorithm
-    """
+    """:pep:`302` Loader that wraps Python's "classic" import algorithm """
+
     code = source = None
 
     def __init__(self, fullname, file, filename, etc):
@@ -365,17 +376,17 @@
 
 
 def get_importer(path_item):
-    """Retrieve a PEP 302 importer for the given path item
+    """Retrieve a  :pep:`302` importer for the given path item
 
-    The returned importer is cached in sys.path_importer_cache
+    The returned importer is cached in ``sys.path_importer_cache``
     if it was newly created by a path hook.
 
     If there is no importer, a wrapper around the basic import
     machinery is returned. This wrapper is never inserted into
-    the importer cache (None is inserted instead).
+    the importer cache (``None`` is inserted instead).
 
     The cache (or part of it) can be cleared manually if a
-    rescan of sys.path_hooks is necessary.
+    rescan of ``sys.path_hooks`` is necessary.
     """
     try:
         importer = sys.path_importer_cache[path_item]
@@ -399,7 +410,7 @@
 
 
 def iter_importers(fullname=""):
-    """Yield PEP 302 importers for the given module name
+    """Yield :pep:`302` importers for the given module name
 
     If fullname contains a '.', the importers will be for the package
     containing fullname, otherwise they will be importers for sys.meta_path,
@@ -407,20 +418,20 @@
     the named module is in a package, that package is imported as a side
     effect of invoking this function.
 
-    Non PEP 302 mechanisms (e.g. the Windows registry) used by the
+    Non :pep:`302` mechanisms (e.g. the Windows registry) used by the
     standard import machinery to find files in alternative locations
-    are partially supported, but are searched AFTER sys.path. Normally,
-    these locations are searched BEFORE sys.path, preventing sys.path
+    are partially supported, but are searched AFTER ``sys.path``. Normally,
+    these locations are searched BEFORE sys.path, preventing ``sys.path``
     entries from shadowing them.
 
     For this to cause a visible difference in behaviour, there must
     be a module or package name that is accessible via both sys.path
-    and one of the non PEP 302 file system mechanisms. In this case,
+    and one of the non :pep:`302` file system mechanisms. In this case,
     the emulation will find the former version, while the builtin
     import mechanism will find the latter.
 
     Items of the following types can be affected by this discrepancy:
-        imp.C_EXTENSION, imp.PY_SOURCE, imp.PY_COMPILED, imp.PKG_DIRECTORY
+        ``imp.C_EXTENSION, imp.PY_SOURCE, imp.PY_COMPILED, imp.PKG_DIRECTORY``
     """
     if fullname.startswith('.'):
         raise ImportError("Relative module names not supported")
@@ -441,15 +452,15 @@
 
 
 def get_loader(module_or_name):
-    """Get a PEP 302 "loader" object for module_or_name
+    """Get a :pep:`302` "loader" object for module_or_name
 
     If the module or package is accessible via the normal import
     mechanism, a wrapper around the relevant part of that machinery
     is returned.  Returns None if the module cannot be found or imported.
     If the named module is not already imported, its containing package
-    (if any) is imported, in order to establish the package __path__.
+    (if any) is imported, in order to establish the package ``__path__``.
 
-    This function uses iter_importers(), and is thus subject to the same
+    This function uses :func:`iter_importers`, and is thus subject to the same
     limitations regarding platform-specific special import locations such
     as the Windows registry.
     """
@@ -467,12 +478,13 @@
 
 
 def find_loader(fullname):
-    """Find a PEP 302 "loader" object for fullname
+    """Find a :pep:`302` "loader" object for fullname
 
-    If fullname contains dots, path must be the containing package's __path__.
-    Returns None if the module cannot be found or imported. This function uses
-    iter_importers(), and is thus subject to the same limitations regarding
-    platform-specific special import locations such as the Windows registry.
+    If fullname contains dots, path must be the containing package's
+    ``__path__``. Returns ``None`` if the module cannot be found or imported.
+    This function uses :func:`iter_importers`, and is thus subject to the same
+    limitations regarding platform-specific special import locations such as
+    the Windows registry.
     """
     for importer in iter_importers(fullname):
         loader = importer.find_module(fullname)
@@ -485,21 +497,22 @@
 def extend_path(path, name):
     """Extend a package's path.
 
-    Intended use is to place the following code in a package's __init__.py:
+    Intended use is to place the following code in a package's
+    ``__init__.py``::
 
         from pkgutil import extend_path
         __path__ = extend_path(__path__, __name__)
 
-    This will add to the package's __path__ all subdirectories of
-    directories on sys.path named after the package.  This is useful
+    This will add to the package's ``__path__`` all subdirectories of
+    directories on ``sys.path`` named after the package.  This is useful
     if one wants to distribute different parts of a single logical
     package as multiple directories.
 
-    It also looks for *.pkg files beginning where * matches the name
-    argument.  This feature is similar to *.pth files (see site.py),
-    except that it doesn't special-case lines starting with 'import'.
-    A *.pkg file is trusted at face value: apart from checking for
-    duplicates, all entries found in a *.pkg file are added to the
+    It also looks for ``*.pkg`` files beginning where ``*`` matches the name
+    argument.  This feature is similar to ``*.pth`` files (see ``site.py``),
+    except that it doesn't special-case lines starting with ``import``.
+    A ``*.pkg`` file is trusted at face value: apart from checking for
+    duplicates, all entries found in a ``*.pkg`` file are added to the
     path, regardless of whether they are exist the filesystem.  (This
     is a feature.)
 
@@ -512,7 +525,7 @@
     are not (unicode or 8-bit) strings referring to existing
     directories are ignored.  Unicode items of sys.path that cause
     errors when used as filenames may cause this function to raise an
-    exception (in line with os.path.isdir() behavior).
+    exception (in line with ``os.path.isdir()`` behavior).
     """
 
     if not isinstance(path, list):
@@ -560,23 +573,23 @@
 def get_data(package, resource):
     """Get a resource from a package.
 
-    This is a wrapper round the PEP 302 loader get_data API. The package
+    This is a wrapper round the :pep:`302` loader get_data API. The package
     argument should be the name of a package, in standard module format
-    (foo.bar). The resource argument should be in the form of a relative
-    filename, using '/' as the path separator. The parent directory name '..'
-    is not allowed, and nor is a rooted name (starting with a '/').
+    (``foo.bar``). The resource argument should be in the form of a relative
+    filename, using ``'/'`` as the path separator. The parent directory name
+    ``'..'`` is not allowed, and nor is a rooted name (starting with a ``'/'``).
 
     The function returns a binary string, which is the contents of the
     specified resource.
 
     For packages located in the filesystem, which have already been imported,
-    this is the rough equivalent of
+    this is the rough equivalent of::
 
         d = os.path.dirname(sys.modules[package].__file__)
         data = open(os.path.join(d, resource), 'rb').read()
 
-    If the package cannot be located or loaded, or it uses a PEP 302 loader
-    which does not support get_data(), then None is returned.
+    If the package cannot be located or loaded, or it uses a :pep:`302` loader
+    which does not support :func:`get_data`, then ``None`` is returned.
     """
 
     loader = get_loader(package)
@@ -603,7 +616,7 @@
 
 class Distribution(object):
     """Created with the *path* of the ``.dist-info`` directory provided to the
-    constructor. It reads the metadata contained in METADATA when it is
+    constructor. It reads the metadata contained in ``METADATA`` when it is
     instantiated."""
 
     # Attribute documenting for Sphinx style documentation, see for more info:
@@ -612,9 +625,9 @@
     """The name of the distribution."""
     metadata = None
     """A :class:`distutils2.metadata.DistributionMetadata` instance loaded with
-    the distribution's METADATA file."""
+    the distribution's ``METADATA`` file."""
     requested = False
-    """A boolean that indicates whether the REQUESTED metadata file is present
+    """A boolean that indicates whether the ``REQUESTED`` metadata file is present
     (in other words, whether the package was installed by user request)."""
 
     def __init__(self, path):
@@ -635,15 +648,17 @@
 
     def get_installed_files(self, local=False):
         """
-        Iterates over the RECORD entries and returns a tuple (path, md5, size)
-        for each line. If *local* is ``True``, the returned path is transformed
-        into a local absolute path. Otherwise the raw value from RECORD is
-        returned.
+        Iterates over the ``RECORD`` entries and returns a tuple
+        ``(path, md5, size)`` for each line. If *local* is ``True``,
+        the returned path is transformed into a local absolute path.
+        Otherwise the raw value from RECORD is returned.
 
         A local absolute path is an absolute path in which occurrences of
         ``'/'`` have been replaced by the system separator given by ``os.sep``.
+
         :parameter local: flag to say if the path should be returned a local
                           absolute path
+
         :type local: boolean
         :returns: iterator of (path, md5, size)
         """
@@ -651,7 +666,7 @@
 
     def uses(self, path):
         """
-        Returns ``True`` if path is listed in RECORD. *path* can be a local
+        Returns ``True`` if path is listed in ``RECORD``. *path* can be a local
         absolute path or a relative ``'/'``-separated path.
 
         :rtype: boolean
@@ -673,8 +688,8 @@
                          directory path, a :class:`DistutilsError` is raised
         :type path: string
         :parameter binary: If *binary* is ``True``, opens the file in read-only
-                           binary mode (rb), otherwise opens it in read-only
-                           mode (r).
+                           binary mode (``rb``), otherwise opens it in read-only
+                           mode (``r``).
         :rtype: file object
         """
         open_flags = 'r'
@@ -701,37 +716,126 @@
 
     def get_distinfo_files(self, local=False):
         """
-        Iterates over the RECORD entries and returns paths for each line if the
-        path is pointing to a file located in the ``.dist-info`` directory or
-        one of its subdirectories.
+        Iterates over the ``RECORD`` entries and returns paths for each line if
+        the path is pointing to a file located in the ``.dist-info`` directory
+        or one of its subdirectories.
 
         :parameter local: If *local* is ``True``, each returned path is
                           transformed into a local absolute path. Otherwise the
-                          raw value from RECORD is returned.
+                          raw value from ``RECORD`` is returned.
         :type local: boolean
         :returns: iterator of paths
         """
         for path, md5, size in self._get_records(local):
             yield path
 
+    def __eq__(self, other):
+        return isinstance(other, Distribution) and self.path == other.path
+
 
 class EggInfoDistribution(object):
     """Created with the *path* of the ``.egg-info`` directory or file provided
     to the constructor. It reads the metadata contained in the file itself, or
     if the given path happens to be a directory, the metadata is read from the
-    file PKG-INFO under that directory."""
+    file ``PKG-INFO`` under that directory."""
 
     name = ''
     """The name of the distribution."""
     metadata = None
     """A :class:`distutils2.metadata.DistributionMetadata` instance loaded with
-    the distribution's METADATA file."""
+    the distribution's ``METADATA`` file."""
+    _REQUIREMENT = re.compile( \
+        r'(?P<name>[-A-Za-z0-9_.]+)\s*' \
+        r'(?P<first>(?:<|<=|!=|==|>=|>)[-A-Za-z0-9_.]+)?\s*' \
+        r'(?P<rest>(?:\s*,\s*(?:<|<=|!=|==|>=|>)[-A-Za-z0-9_.]+)*)\s*' \
+        r'(?P<extras>\[.*\])?')
 
     def __init__(self, path):
-        if os.path.isdir(path):
-            path = os.path.join(path, 'PKG-INFO')
-        self.metadata = DistributionMetadata(path=path)
-        self.name = self.metadata['name']
+        self.path = path
+
+        # reused from Distribute's pkg_resources
+        def yield_lines(strs):
+            """Yield non-empty/non-comment lines of a ``basestring`` or sequence"""
+            if isinstance(strs, basestring):
+                for s in strs.splitlines():
+                    s = s.strip()
+                    if s and not s.startswith('#'): # skip blank lines/comments
+                        yield s
+            else:
+                for ss in strs:
+                    for s in yield_lines(ss):
+                        yield s
+
+        requires = None
+        if path.endswith('.egg'):
+            if os.path.isdir(path):
+                meta_path = os.path.join(path, 'EGG-INFO', 'PKG-INFO')
+                self.metadata = DistributionMetadata(path=meta_path)
+                try:
+                    req_path = os.path.join(path, 'EGG-INFO', 'requires.txt')
+                    requires = open(req_path, 'r').read()
+                except IOError:
+                    requires = None
+            else:
+                zipf = zipimport.zipimporter(path)
+                fileobj = StringIO.StringIO(zipf.get_data('EGG-INFO/PKG-INFO'))
+                self.metadata = DistributionMetadata(fileobj=fileobj)
+                try:
+                    requires = zipf.get_data('EGG-INFO/requires.txt')
+                except IOError:
+                    requires = None
+            self.name = self.metadata['name']
+        elif path.endswith('.egg-info'):
+            if os.path.isdir(path):
+                path = os.path.join(path, 'PKG-INFO')
+                try:
+                    req_f = open(os.path.join(path, 'requires.txt'), 'r')
+                    requires = req_f.read()
+                except IOError:
+                    requires = None
+            self.metadata = DistributionMetadata(path=path)
+            self.name = self.metadata['name']
+        else:
+            raise ValueError('The path must end with .egg-info or .egg')
+
+        provides = "%s (%s)" % (self.metadata['name'],
+                                self.metadata['version'])
+        if self.metadata['Metadata-Version'] == '1.2':
+            self.metadata['Provides-Dist'] += (provides,)
+        else:
+            self.metadata['Provides'] += (provides,)
+        reqs = []
+        if requires is not None:
+            for line in yield_lines(requires):
+                if line[0] == '[':
+                    warnings.warn('distutils2 does not support extensions in requires.txt')
+                    break
+                else:
+                    match = self._REQUIREMENT.match(line.strip())
+                    if not match:
+                        raise ValueError('Distribution %s has ill formed '
+                                         'requires.txt file (%s)' %
+                                         (self.name, line))
+                    else:
+                        if match.group('extras'):
+                            s = 'Distribution %s uses extra requirements which'\
+                                ' are not supported in distutils' % (self.name)
+                            warnings.warn(s)
+                        name = match.group('name')
+                        version = None
+                        if match.group('first'):
+                            version = match.group('first')
+                            if match.group('rest'):
+                                version += match.group('rest')
+                            version = version.replace(' ', '') # trim spaces
+                        if version is None:
+                            reqs.append(name)
+                        else:
+                            reqs.append('%s (%s)' % (name, version))
+            if self.metadata['Metadata-Version'] == '1.2':
+                self.metadata['Requires-Dist'] += reqs
+            else:
+                self.metadata['Requires'] += reqs
 
     def get_installed_files(self, local=False):
         return []
@@ -739,6 +843,10 @@
     def uses(self, path):
         return False
 
+    def __eq__(self, other):
+        return isinstance(other, EggInfoDistribution) and \
+               self.path == other.path
+
 
 def _normalize_dist_name(name):
     """Returns a normalized name from the given *name*.
@@ -792,7 +900,8 @@
             if dir.endswith('.dist-info'):
                 dist = Distribution(os.path.join(realpath, dir))
                 yield dist
-            elif use_egg_info and dir.endswith('.egg-info'):
+            elif use_egg_info and (dir.endswith('.egg-info') or
+                                   dir.endswith('.egg')):
                 dist = EggInfoDistribution(os.path.join(realpath, dir))
                 yield dist
 
@@ -801,8 +910,8 @@
     """
     Scans all elements in ``sys.path`` and looks for all directories ending with
     ``.dist-info``. Returns a :class:`Distribution` corresponding to the
-    ``.dist-info`` directory that contains the METADATA that matches *name* for
-    the *name* metadata field.
+    ``.dist-info`` directory that contains the ``METADATA`` that matches *name*
+    for the *name* metadata field.
     If no distribution exists with the given *name* and the parameter
     *use_egg_info* is set to ``True``, then all files and directories ending
     with ``.egg-info`` are scanned. A :class:`EggInfoDistribution` instance is
@@ -812,7 +921,7 @@
     This function only returns the first result founded, as no more than one
     value is expected. If the directory is not found, ``None`` is returned.
 
-    :rtype: :class:`Distribution` or :class:`EggInfoDistribution: or None"""
+    :rtype: :class:`Distribution` or :class:`EggInfoDistribution` or None"""
     found = None
     for dist in get_distributions():
         if dist.name == name:
@@ -887,7 +996,7 @@
         provided = dist.metadata['Provides-Dist'] + dist.metadata['Provides']
 
         for p in provided:
-            p_components = p.split(' ', 1)
+            p_components = p.rsplit(' ', 1)
             if len(p_components) == 1 or predicate is None:
                 if name == p_components[0]:
                     yield dist
@@ -896,7 +1005,8 @@
                 p_name, p_ver = p_components
                 if len(p_ver) < 2 or p_ver[0] != '(' or p_ver[-1] != ')':
                     raise DistutilsError(('Distribution %s has invalid ' +
-                                          'provides field') % (dist.name,))
+                                          'provides field: %s') \
+                                           % (dist.name, p))
                 p_ver = p_ver[1:-1] # trim off the parenthesis
                 if p_name == name and predicate.match(p_ver):
                     yield dist
diff --git a/src/distutils2/_backport/tests/fake_dists/bacon-0.1.egg-info/PKG-INFO b/src/distutils2/_backport/tests/fake_dists/bacon-0.1.egg-info/PKG-INFO
--- a/src/distutils2/_backport/tests/fake_dists/bacon-0.1.egg-info/PKG-INFO
+++ b/src/distutils2/_backport/tests/fake_dists/bacon-0.1.egg-info/PKG-INFO
@@ -2,4 +2,5 @@
 Name: bacon
 Version: 0.1
 Provides-Dist: truffles (2.0)
+Provides-Dist: bacon (0.1)
 Obsoletes-Dist: truffles (>=0.9,<=1.5)
diff --git a/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/PKG-INFO b/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/PKG-INFO
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/PKG-INFO
@@ -0,0 +1,18 @@
+Metadata-Version: 1.0
+Name: banana
+Version: 0.4
+Summary: A yellow fruit
+Home-page: http://en.wikipedia.org/wiki/Banana
+Author: Josip Djolonga
+Author-email: foo at nbar.com
+License: BSD
+Description: A fruit
+Keywords: foo bar
+Platform: UNKNOWN
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: Intended Audience :: Science/Research
+Classifier: License :: OSI Approved :: BSD License
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python
+Classifier: Topic :: Scientific/Engineering :: GIS
diff --git a/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/SOURCES.txt b/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/SOURCES.txt
new file mode 100644
diff --git a/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/dependency_links.txt b/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/dependency_links.txt
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/dependency_links.txt
@@ -0,0 +1,1 @@
+
diff --git a/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/entry_points.txt b/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/entry_points.txt
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/entry_points.txt
@@ -0,0 +1,3 @@
+
+      # -*- Entry points: -*-
+      
\ No newline at end of file
diff --git a/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/not-zip-safe b/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/not-zip-safe
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/not-zip-safe
@@ -0,0 +1,1 @@
+
diff --git a/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/requires.txt b/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/requires.txt
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/requires.txt
@@ -0,0 +1,6 @@
+# this should be ignored
+
+strawberry >=0.5
+
+[section ignored]
+foo ==0.5
diff --git a/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/top_level.txt b/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/top_level.txt
new file mode 100644
diff --git a/src/distutils2/_backport/tests/fake_dists/strawberry-0.6.egg b/src/distutils2/_backport/tests/fake_dists/strawberry-0.6.egg
new file mode 100644
index 0000000000000000000000000000000000000000..6d160e8b161031ae52638514843592187925b757
GIT binary patch
literal 1402
zc$`yK)KALH(=X28%1l#;R!B%nEKbc!%uQ8LF-TCbRZuEUEh#N1$<NOz)-}*GOExsr
zEvPioGuBH at Pghci<toWY%~MFNNKIBKs4U6I&jU&+=q4*DW#$&-7nLX!R~G9i<QH3m
z7<vVXB^i2|dBv$kB^m}GVTed#QZb0uP**Wf*VMFDNGr<ERX|v)mz<xQo0ylPmzr2y
z84uK6l9-dD05v};Kfk27q$sffVnb?0W{Cz|OhZ#sNkK`)L5V9hr#O{MK_N9cBOmB5
zh0J1wy!;XcpxYIa^NWg7lS>r9UQWv|0ty0Ufu2)H%gjmDgJ}xL0otCbPy`8 at OrXVy
z$=M1e`3iV~M*-+)g_5F5g~as4%sjABpm0h{1UV)xlPkcRnT3l11j?rGw_!j6Vhl12
zuI}!-o_=or`X%`V at j0nwsX2Nj6(yk|oD9qiubF*7xU_<sfsy3}GXn#dK$usBW}XPL
zdBOgnLC&thaML(|CUIalO$4ZygTbxvn9cJCa*K8Yd7pq-5ZSZ<ZwLcyl*$Bd9}q?}
z%GU3+-(d%yJ- at a8FngYAJv`50)he!AtAbSdVkX?4u~u<++)e34e{3~_X5O|jF|J;B
zC3xwx-m|X72inV6*}Y^gtiD*1wpsGu-O%40%MDdGSw1j37$$7q`bcR1wJ4|Sg?Z{f
zrybfGuv%=l^lLM>i9RnLxBqqeU~_GC)B*8Q%^m+PITr6Z&8t31F+o4>xK^{d-p7?^
zx~J#&ZkWuS9<SEFoOjj3--Yb2mtJ3TW&iZ at 1KQ^&N4;mX5B=!V_JJD_rYtj1!&DME
z6jJkm at f=@}pP5%u3=ddwV4#ZQ4p;_;F99Leli98tF#_2jEDOYNBU4ffQu9($^O7s$
zb29U?!NvlU3?q{qGp<A<0cUS%tYT!qnS at v&NeIJT2(vL05VF~)kj+L(POOmRgw<qR
zi3wsdBiLjKCI)OtixrZz at HiMVb`TC;(kREwfG at 6CA#sJpco7ENF@*?^LS_Q-7U0dw
Q22#ojgxi7o*D`~60Pqom)c^nh

diff --git a/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1.dist-info/METADATA b/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1.dist-info/METADATA
--- a/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1.dist-info/METADATA
+++ b/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1.dist-info/METADATA
@@ -2,4 +2,6 @@
 Name: towel-stuff
 Version: 0.1
 Provides-Dist: truffles (1.1.2)
+Provides-Dist: towel-stuff (0.1)
 Obsoletes-Dist: truffles (!=0.8,<1.0)
+Requires-Dist: bacon (<=0.2)
diff --git a/src/distutils2/_backport/tests/test_pkgutil.py b/src/distutils2/_backport/tests/test_pkgutil.py
--- a/src/distutils2/_backport/tests/test_pkgutil.py
+++ b/src/distutils2/_backport/tests/test_pkgutil.py
@@ -1,8 +1,13 @@
 # -*- coding: utf-8 -*-
 """Tests for PEP 376 pkgutil functionality"""
+import unittest2
 import sys
 import os
 import csv
+import imp
+import tempfile
+import shutil
+import zipfile
 try:
     from hashlib import md5
 except ImportError:
@@ -10,31 +15,141 @@
 
 from test.test_support import run_unittest, TESTFN
 
-import distutils2._backport.pkgutil
-from distutils2.tests.support import unittest
+from distutils2._backport import pkgutil
 
 # TODO Add a test for getting a distribution that is provided by another
 #   distribution.
 
 # TODO Add a test for absolute pathed RECORD items (e.g. /etc/myapp/config.ini)
 
-class TestPkgUtilDistribution(unittest.TestCase):
+# Adapted from Python 2.7's trunk
+class TestPkgUtilData(unittest2.TestCase):
+
+    def setUp(self):
+        self.dirname = tempfile.mkdtemp()
+        sys.path.insert(0, self.dirname)
+
+    def tearDown(self):
+        del sys.path[0]
+        shutil.rmtree(self.dirname)
+
+    def test_getdata_filesys(self):
+        pkg = 'test_getdata_filesys'
+
+        # Include a LF and a CRLF, to test that binary data is read back
+        RESOURCE_DATA = 'Hello, world!\nSecond line\r\nThird line'
+
+        # Make a package with some resources
+        package_dir = os.path.join(self.dirname, pkg)
+        os.mkdir(package_dir)
+        # Empty init.py
+        f = open(os.path.join(package_dir, '__init__.py'), "wb")
+        f.close()
+        # Resource files, res.txt, sub/res.txt
+        f = open(os.path.join(package_dir, 'res.txt'), "wb")
+        f.write(RESOURCE_DATA)
+        f.close()
+        os.mkdir(os.path.join(package_dir, 'sub'))
+        f = open(os.path.join(package_dir, 'sub', 'res.txt'), "wb")
+        f.write(RESOURCE_DATA)
+        f.close()
+
+        # Check we can read the resources
+        res1 = pkgutil.get_data(pkg, 'res.txt')
+        self.assertEqual(res1, RESOURCE_DATA)
+        res2 = pkgutil.get_data(pkg, 'sub/res.txt')
+        self.assertEqual(res2, RESOURCE_DATA)
+
+        del sys.modules[pkg]
+
+    def test_getdata_zipfile(self):
+        zip = 'test_getdata_zipfile.zip'
+        pkg = 'test_getdata_zipfile'
+
+        # Include a LF and a CRLF, to test that binary data is read back
+        RESOURCE_DATA = 'Hello, world!\nSecond line\r\nThird line'
+
+        # Make a package with some resources
+        zip_file = os.path.join(self.dirname, zip)
+        z = zipfile.ZipFile(zip_file, 'w')
+
+        # Empty init.py
+        z.writestr(pkg + '/__init__.py', "")
+        # Resource files, res.txt, sub/res.txt
+        z.writestr(pkg + '/res.txt', RESOURCE_DATA)
+        z.writestr(pkg + '/sub/res.txt', RESOURCE_DATA)
+        z.close()
+
+        # Check we can read the resources
+        sys.path.insert(0, zip_file)
+        res1 = pkgutil.get_data(pkg, 'res.txt')
+        self.assertEqual(res1, RESOURCE_DATA)
+        res2 = pkgutil.get_data(pkg, 'sub/res.txt')
+        self.assertEqual(res2, RESOURCE_DATA)
+        del sys.path[0]
+
+        del sys.modules[pkg]
+
+# Adapted from Python 2.7's trunk
+class TestPkgUtilPEP302(unittest2.TestCase):
+
+    class MyTestLoader(object):
+        def load_module(self, fullname):
+            # Create an empty module
+            mod = sys.modules.setdefault(fullname, imp.new_module(fullname))
+            mod.__file__ = "<%s>" % self.__class__.__name__
+            mod.__loader__ = self
+            # Make it a package
+            mod.__path__ = []
+            # Count how many times the module is reloaded
+            mod.__dict__['loads'] = mod.__dict__.get('loads',0) + 1
+            return mod
+
+        def get_data(self, path):
+            return "Hello, world!"
+
+    class MyTestImporter(object):
+        def find_module(self, fullname, path=None):
+            return TestPkgUtilPEP302.MyTestLoader()
+
+    def setUp(self):
+        sys.meta_path.insert(0, self.MyTestImporter())
+
+    def tearDown(self):
+        del sys.meta_path[0]
+
+    def test_getdata_pep302(self):
+        # Use a dummy importer/loader
+        self.assertEqual(pkgutil.get_data('foo', 'dummy'), "Hello, world!")
+        del sys.modules['foo']
+
+    def test_alreadyloaded(self):
+        # Ensure that get_data works without reloading - the "loads" module
+        # variable in the example loader should count how many times a reload
+        # occurs.
+        import foo
+        self.assertEqual(foo.loads, 1)
+        self.assertEqual(pkgutil.get_data('foo', 'dummy'), "Hello, world!")
+        self.assertEqual(foo.loads, 1)
+        del sys.modules['foo']
+
+
+class TestPkgUtilDistribution(unittest2.TestCase):
     """Tests the pkgutil.Distribution class"""
 
     def setUp(self):
-        super(TestPkgUtilDistribution, self).setUp()
-
         self.fake_dists_path = os.path.abspath(
             os.path.join(os.path.dirname(__file__), 'fake_dists'))
-        self.distinfo_dirs = [ os.path.join(self.fake_dists_path, dir)
+
+        self.distinfo_dirs = [os.path.join(self.fake_dists_path, dir)
             for dir in os.listdir(self.fake_dists_path)
-            if dir.endswith('.dist-info')
-            ]
+            if dir.endswith('.dist-info')]
 
         def get_hexdigest(file):
             md5_hash = md5()
             md5_hash.update(open(file).read())
             return md5_hash.hexdigest()
+
         def record_pieces(file):
             path = os.path.relpath(file, sys.prefix)
             digest = get_hexdigest(file)
@@ -51,7 +166,8 @@
 
             for path, dirs, files in os.walk(dist_location):
                 for f in files:
-                    record_writer.writerow(record_pieces(os.path.join(path, f)))
+                    record_writer.writerow(record_pieces(
+                                           os.path.join(path, f)))
             for file in ['INSTALLER', 'METADATA', 'REQUESTED']:
                 record_writer.writerow(record_pieces(
                     os.path.join(distinfo_dir, file)))
@@ -60,7 +176,8 @@
             record_reader = csv.reader(open(record_file, 'rb'))
             record_data = []
             for row in record_reader:
-                path, md5_, size = row[:] + [ None for i in xrange(len(row), 3) ]
+                path, md5_, size = row[:] + \
+                                   [None for i in xrange(len(row), 3)]
                 record_data.append([path, (md5_, size,)])
             self.records[distinfo_dir] = dict(record_data)
 
@@ -69,7 +186,6 @@
         for distinfo_dir in self.distinfo_dirs:
             record_file = os.path.join(distinfo_dir, 'RECORD')
             open(record_file, 'w').close()
-        super(TestPkgUtilDistribution, self).tearDown()
 
     def test_instantiation(self):
         """Test the Distribution class's instantiation provides us with usable
@@ -108,7 +224,8 @@
         distinfo_name = 'grammar-1.0a4'
         distinfo_dir = os.path.join(self.fake_dists_path,
             distinfo_name + '.dist-info')
-        true_path = [self.fake_dists_path, distinfo_name, 'grammar', 'utils.py']
+        true_path = [self.fake_dists_path, distinfo_name, \
+                     'grammar', 'utils.py']
         true_path = os.path.relpath(os.path.join(*true_path), sys.prefix)
         false_path = [self.fake_dists_path, 'towel_stuff-0.1', 'towel_stuff',
             '__init__.py']
@@ -150,7 +267,8 @@
         self.assertRaises(DistutilsError, dist.get_distinfo_file,
             other_distinfo_file)
         # Test for a file that does not exist and should not exist
-        self.assertRaises(DistutilsError, dist.get_distinfo_file, 'ENTRYPOINTS')
+        self.assertRaises(DistutilsError, dist.get_distinfo_file, \
+                          'ENTRYPOINTS')
 
     def test_get_distinfo_files(self):
         """Test for the iteration of RECORD path entries."""
@@ -161,21 +279,19 @@
         dist = Distribution(distinfo_dir)
         # Test for the iteration of the raw path
         distinfo_record_paths = self.records[distinfo_dir].keys()
-        found = [ path for path in dist.get_distinfo_files() ]
+        found = [path for path in dist.get_distinfo_files()]
         self.assertEqual(sorted(found), sorted(distinfo_record_paths))
         # Test for the iteration of local absolute paths
-        distinfo_record_paths = [ os.path.join(sys.prefix, path)
-            for path in self.records[distinfo_dir].keys()
-            ]
-        found = [ path for path in dist.get_distinfo_files(local=True) ]
+        distinfo_record_paths = [os.path.join(sys.prefix, path)
+            for path in self.records[distinfo_dir].keys()]
+        found = [path for path in dist.get_distinfo_files(local=True)]
         self.assertEqual(sorted(found), sorted(distinfo_record_paths))
 
 
-class TestPkgUtilFunctions(unittest.TestCase):
+class TestPkgUtilPEP376(unittest2.TestCase):
     """Tests for the new functionality added in PEP 376."""
 
     def setUp(self):
-        super(TestPkgUtilFunctions, self).setUp()
         # Setup the path environment with our fake distributions
         current_path = os.path.abspath(os.path.dirname(__file__))
         self.sys_path = sys.path[:]
@@ -183,15 +299,15 @@
         sys.path[0:0] = [self.fake_dists_path]
 
     def tearDown(self):
-        super(TestPkgUtilFunctions, self).tearDown()
         sys.path[:] = self.sys_path
 
     def test_distinfo_dirname(self):
         """Given a name and a version, we expect the distinfo_dirname function
         to return a standard distribution information directory name."""
 
-        items = [ # (name, version, standard_dirname)
-            # Test for a very simple single word name and decimal version number
+        items = [# (name, version, standard_dirname)
+            # Test for a very simple single word name and decimal
+            # version number
             ('docutils', '0.5', 'docutils-0.5.dist-info'),
             # Test for another except this time with a '-' in the name, which
             #   needs to be transformed during the name lookup
@@ -221,7 +337,7 @@
                                                  EggInfoDistribution
 
         # Verify the fake dists have been found.
-        dists = [ dist for dist in get_distributions() ]
+        dists = [dist for dist in get_distributions()]
         for dist in dists:
             if not isinstance(dist, Distribution):
                 self.fail("item received was not a Distribution instance: "
@@ -234,10 +350,11 @@
         self.assertListEqual(sorted(found_dists), sorted(fake_dists))
 
         # Now, test if the egg-info distributions are found correctly as well
-        fake_dists += [('bacon', '0.1'), ('cheese', '2.0.2')]
+        fake_dists += [('bacon', '0.1'), ('cheese', '2.0.2'),
+                       ('banana', '0.4'), ('strawberry', '0.6')]
         found_dists = []
 
-        dists = [ dist for dist in get_distributions(use_egg_info=True) ]
+        dists = [dist for dist in get_distributions(use_egg_info=True)]
         for dist in dists:
             if not (isinstance(dist, Distribution) or \
                     isinstance(dist, EggInfoDistribution)):
@@ -248,7 +365,6 @@
 
         self.assertListEqual(sorted(fake_dists), sorted(found_dists))
 
-
     def test_get_distribution(self):
         """Test for looking up a distribution by name."""
         # Test the lookup of the towel-stuff distribution
@@ -274,6 +390,8 @@
         # instructed to
         self.assertEqual(None, get_distribution('bacon'))
         self.assertEqual(None, get_distribution('cheese'))
+        self.assertEqual(None, get_distribution('strawberry'))
+        self.assertEqual(None, get_distribution('banana'))
 
         # Now check that it works well in both situations, when egg-info
         # is a file and directory respectively.
@@ -285,6 +403,14 @@
         self.assertTrue(isinstance(dist, EggInfoDistribution))
         self.assertEqual(dist.name, 'bacon')
 
+        dist = get_distribution('banana', use_egg_info=True)
+        self.assertTrue(isinstance(dist, EggInfoDistribution))
+        self.assertEqual(dist.name, 'banana')
+
+        dist = get_distribution('strawberry', use_egg_info=True)
+        self.assertTrue(isinstance(dist, EggInfoDistribution))
+        self.assertEqual(dist.name, 'strawberry')
+
     def test_get_file_users(self):
         """Test the iteration of distributions that use a file."""
         from distutils2._backport.pkgutil import get_file_users, Distribution
@@ -300,7 +426,7 @@
         from distutils2._backport.pkgutil import provides_distribution
         from distutils2.errors import DistutilsError
 
-        checkLists = lambda x,y: self.assertListEqual(sorted(x), sorted(y))
+        checkLists = lambda x, y: self.assertListEqual(sorted(x), sorted(y))
 
         l = [dist.name for dist in provides_distribution('truffles')]
         checkLists(l, ['choxie', 'towel-stuff'])
@@ -318,10 +444,12 @@
         l = [dist.name for dist in provides_distribution('truffles', '1.1')]
         checkLists(l, ['towel-stuff'])
 
-        l = [dist.name for dist in provides_distribution('truffles', '!=1.1,<=2.0')]
+        l = [dist.name for dist in provides_distribution('truffles', \
+                                                         '!=1.1,<=2.0')]
         checkLists(l, ['choxie'])
 
-        l = [dist.name for dist in provides_distribution('truffles', '!=1.1,<=2.0',
+        l = [dist.name for dist in provides_distribution('truffles', \
+                                                         '!=1.1,<=2.0',
                                                           use_egg_info=True)]
         checkLists(l, ['choxie', 'bacon', 'cheese'])
 
@@ -338,12 +466,39 @@
         l = [dist.name for dist in provides_distribution('truffles', '>=1.0')]
         checkLists(l, ['choxie', 'towel-stuff'])
 
+        l = [dist.name for dist in provides_distribution('strawberry', '0.6',
+                                                         use_egg_info=True)]
+        checkLists(l, ['strawberry'])
+
+        l = [dist.name for dist in provides_distribution('strawberry', '>=0.5',
+                                                         use_egg_info=True)]
+        checkLists(l, ['strawberry'])
+
+
+        l = [dist.name for dist in provides_distribution('strawberry', '>0.6',
+                                                         use_egg_info=True)]
+        checkLists(l, [])
+
+
+        l = [dist.name for dist in provides_distribution('banana', '0.4',
+                                                         use_egg_info=True)]
+        checkLists(l, ['banana'])
+
+        l = [dist.name for dist in provides_distribution('banana', '>=0.3',
+                                                         use_egg_info=True)]
+        checkLists(l, ['banana'])
+
+
+        l = [dist.name for dist in provides_distribution('banana', '!=0.4',
+                                                         use_egg_info=True)]
+        checkLists(l, [])
+
     def test_obsoletes(self):
         """ Test looking for distributions based on what they obsolete """
         from distutils2._backport.pkgutil import obsoletes_distribution
         from distutils2.errors import DistutilsError
 
-        checkLists = lambda x,y: self.assertListEqual(sorted(x), sorted(y))
+        checkLists = lambda x, y: self.assertListEqual(sorted(x), sorted(y))
 
         l = [dist.name for dist in obsoletes_distribution('truffles', '1.0')]
         checkLists(l, [])
@@ -363,7 +518,8 @@
         l = [dist.name for dist in obsoletes_distribution('truffles', '0.9.6')]
         checkLists(l, ['choxie', 'towel-stuff'])
 
-        l = [dist.name for dist in obsoletes_distribution('truffles', '0.5.2.3')]
+        l = [dist.name for dist in obsoletes_distribution('truffles', \
+                                                          '0.5.2.3')]
         checkLists(l, ['choxie', 'towel-stuff'])
 
         l = [dist.name for dist in obsoletes_distribution('truffles', '0.2')]
@@ -371,27 +527,18 @@
 
 
 def test_suite():
-    suite = unittest.TestSuite()
-    testcase_loader = unittest.loader.defaultTestLoader.loadTestsFromTestCase
-    suite.addTest(testcase_loader(TestPkgUtilFunctions))
+    suite = unittest2.TestSuite()
+    testcase_loader = unittest2.loader.defaultTestLoader.loadTestsFromTestCase
+    suite.addTest(testcase_loader(TestPkgUtilData))
     suite.addTest(testcase_loader(TestPkgUtilDistribution))
+    suite.addTest(testcase_loader(TestPkgUtilPEP302))
+    suite.addTest(testcase_loader(TestPkgUtilPEP376))
     return suite
 
+
 def test_main():
     run_unittest(test_suite())
 
-if __name__ == "__main__":
-    test_main()
-
-def test_suite():
-    suite = unittest.TestSuite()
-    testcase_loader = unittest.loader.defaultTestLoader.loadTestsFromTestCase
-    suite.addTest(testcase_loader(TestPkgUtilFunctions))
-    suite.addTest(testcase_loader(TestPkgUtilDistribution))
-    return suite
-
-def test_main():
-    run_unittest(test_suite())
 
 if __name__ == "__main__":
     test_main()
diff --git a/src/distutils2/depgraph.py b/src/distutils2/depgraph.py
--- a/src/distutils2/depgraph.py
+++ b/src/distutils2/depgraph.py
@@ -1,47 +1,49 @@
-"""
-A dependency graph generator. The graph is represented as an instance of
-:class:`DependencyGraph`, and DOT output is possible as well.
+"""Analyse the relationships between the distributions in the system and generate
+a dependency graph.
 """
 
-from distutils2._backport import pkgutil
 from distutils2.errors import DistutilsError
 from distutils2.version import VersionPredicate
 
-__all__ = ['DependencyGraph', 'generate_graph']
+__all__ = ['DependencyGraph', 'generate_graph', 'dependent_dists',
+           'graph_to_dot']
 
 
 class DependencyGraph(object):
     """
     Represents a dependency graph between distributions.
 
-    The depedency relationships are stored in an *adjacency_list* that maps
+    The depedency relationships are stored in an ``adjacency_list`` that maps
     distributions to a list of ``(other, label)`` tuples where  ``other``
     is a distribution and the edge is labelled with ``label`` (i.e. the version
-    specifier, if such was provided). If any missing depencies are found,
-    they are stored in ``missing``. It maps distributions to a list of
-    requirements that were not provided by any other distributions.
+    specifier, if such was provided). Also, for more efficient traversal, for
+    every distribution ``x``, a list of predecessors is kept in
+    ``reverse_list[x]``. An edge from distribution ``a`` to
+    distribution ``b`` means that ``a`` depends on ``b``. If any missing
+    depencies are found, they are stored in ``missing``, which is a dictionary
+    that maps distributions to a list of requirements that were not provided by
+    any other distributions.
     """
 
     def __init__(self):
         self.adjacency_list = {}
+        self.reverse_list = {}
         self.missing = {}
 
     def add_distribution(self, distribution):
-        """
-        Add distribution *x* to the graph.
+        """Add the *distribution* to the graph.
 
         :type distribution: :class:`pkgutil.Distribution` or
                             :class:`pkgutil.EggInfoDistribution`
         """
         self.adjacency_list[distribution] = list()
+        self.reverse_list[distribution] = list()
         self.missing[distribution] = list()
 
     def add_edge(self, x, y, label=None):
-        """
-        Add an edge from distribution *x* to distribution *y* with the given
+        """Add an edge from distribution *x* to distribution *y* with the given
         *label*.
 
-
         :type x: :class:`pkgutil.Distribution` or
                  :class:`pkgutil.EggInfoDistribution`
         :type y: :class:`pkgutil.Distribution` or
@@ -49,6 +51,9 @@
         :type label: ``str`` or ``None``
         """
         self.adjacency_list[x].append((y, label))
+        # multiple edges are allowed, so be careful
+        if not x in self.reverse_list[y]:
+            self.reverse_list[y].append(x)
 
     def add_missing(self, distribution, requirement):
         """
@@ -60,45 +65,42 @@
         """
         self.missing[distribution].append(requirement)
 
-    def to_dot(self, f, skip_disconnected=True):
-        """
-        Writes a DOT output for the graph to the provided *file*.
-        If *skip_disconnected* is set to ``True``, then all distributions
-        that are not dependent on any other distributions are skipped.
 
-        :type f: ``file``
-        ;type skip_disconnected: ``bool``
-        """
-        if not isinstance(f, file):
-            raise TypeError('the argument has to be of type file')
+def graph_to_dot(graph, f, skip_disconnected=True):
+    """Writes a DOT output for the graph to the provided file *f*.
 
-        disconnected = []
+    If *skip_disconnected* is set to ``True``, then all distributions
+    that are not dependent on any other distribution are skipped.
 
-        f.write("digraph dependencies {\n")
-        for dist, adjs in self.adjacency_list.iteritems():
-            if len(adjs) == 0 and not skip_disconnected:
-                disconnected.append(dist)
-            for (other, label) in adjs:
-                if not label is None:
-                    f.write('"%s" -> "%s" [label="%s"]\n' %
-                                                (dist.name, other.name, label))
-                else:
-                    f.write('"%s" -> "%s"\n' % (dist.name, other.name))
-        if not skip_disconnected and len(disconnected) > 0:
-            f.write('subgraph disconnected {\n')
-            f.write('label = "Disconnected"\n')
-            f.write('bgcolor = red\n')
+    :type f: has to support ``file``-like operations
+    :type skip_disconnected: ``bool``
+    """
+    disconnected = []
 
-            for dist in disconnected:
-                f.write('"%s"' % dist.name)
-                f.write('\n')
-            f.write('}\n')
+    f.write("digraph dependencies {\n")
+    for dist, adjs in graph.adjacency_list.iteritems():
+        if len(adjs) == 0 and not skip_disconnected:
+            disconnected.append(dist)
+        for (other, label) in adjs:
+            if not label is None:
+                f.write('"%s" -> "%s" [label="%s"]\n' %
+                                            (dist.name, other.name, label))
+            else:
+                f.write('"%s" -> "%s"\n' % (dist.name, other.name))
+    if not skip_disconnected and len(disconnected) > 0:
+        f.write('subgraph disconnected {\n')
+        f.write('label = "Disconnected"\n')
+        f.write('bgcolor = red\n')
+
+        for dist in disconnected:
+            f.write('"%s"' % dist.name)
+            f.write('\n')
         f.write('}\n')
+    f.write('}\n')
 
 
 def generate_graph(dists):
-    """
-    Generates a dependency graph from the given distributions.
+    """Generates a dependency graph from the given distributions.
 
     :parameter dists: a list of distributions
     :type dists: list of :class:`pkgutil.Distribution` and
@@ -115,7 +117,7 @@
         provides = dist.metadata['Provides-Dist'] + dist.metadata['Provides']
 
         for p in provides:
-            comps = p.split(" ", 1)
+            comps = p.strip().rsplit(" ", 1)
             name = comps[0]
             version = None
             if len(comps) == 2:
@@ -133,22 +135,26 @@
         requires = dist.metadata['Requires-Dist'] + dist.metadata['Requires']
         for req in requires:
             predicate = VersionPredicate(req)
-            comps = req.split(" ", 1)
+            comps = req.strip().rsplit(" ", 1)
             name = comps[0]
 
             if not name in provided:
                 graph.add_missing(dist, req)
             else:
+                matched = False
                 for (version, provider) in provided[name]:
                     if predicate.match(version):
                         graph.add_edge(dist, provider, req)
+                        matched = True
+                        break
+                if not matched:
+                    graph.add_missing(dist, req)
 
     return graph
 
 
 def dependent_dists(dists, dist):
-    """
-    Recursively generate a list of distributions from *dists* that are
+    """Recursively generate a list of distributions from *dists* that are
     dependent on *dist*.
 
     :param dists: a list of distributions
@@ -158,24 +164,26 @@
         raise ValueError('The given distribution is not a member of the list')
     graph = generate_graph(dists)
 
-    dep = [dist]
-    fringe = [dist] # list of nodes we should expand
+    dep = [dist] # dependent distributions
+    fringe = graph.reverse_list[dist] # list of nodes we should inspect
+
     while not len(fringe) == 0:
-        next = graph.adjacency_list[fringe.pop()]
-        for (dist, label) in next:
-            if not dist in dep: # avoid infinite loops
-                dep.append(dist)
-                fringe.append(dist)
+        node = fringe.pop()
+        dep.append(node)
+        for prev in graph.reverse_list[node]:
+            if not prev in dep:
+                fringe.append(prev)
 
-    dep.pop()
+    dep.pop(0) # remove dist from dep, was there to prevent infinite loops
     return dep
 
 if __name__ == '__main__':
-    dists = list(pkgutil.get_distributions(use_egg_info=True))
+    from distutils2._backport.pkgutil import get_distributions
+    dists = list(get_distributions(use_egg_info=True))
     graph = generate_graph(dists)
     for dist, reqs in graph.missing.iteritems():
         if len(reqs) > 0:
             print("Missing dependencies for %s: %s" % (dist.name,
                                                        ", ".join(reqs)))
     f = open('output.dot', 'w')
-    graph.to_dot(f, True)
+    graph_to_dot(graph, f, True)
diff --git a/src/distutils2/metadata.py b/src/distutils2/metadata.py
--- a/src/distutils2/metadata.py
+++ b/src/distutils2/metadata.py
@@ -186,13 +186,15 @@
     """Distribution meta-data class (1.0 or 1.2).
     """
     def __init__(self, path=None, platform_dependant=False,
-                 execution_context=None):
+                 execution_context=None, fileobj=None):
         self._fields = {}
         self.version = None
         self.docutils_support = _HAS_DOCUTILS
         self.platform_dependant = platform_dependant
         if path is not None:
             self.read(path)
+        elif fileobj is not None:
+            self.read_file(fileobj)
         self.execution_context = execution_context
 
     def _set_best_version(self):
diff --git a/src/distutils2/tests/test_depgraph.py b/src/distutils2/tests/test_depgraph.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/test_depgraph.py
@@ -0,0 +1,187 @@
+"""Tests for distutils.depgraph """
+
+from distutils2.tests import support
+from distutils2.tests.support import unittest
+from distutils2 import depgraph
+from distutils2._backport import pkgutil
+
+import os
+import sys
+import re
+try:
+    import cStringIO as StringIO
+except ImportErorr:
+    import StringIO
+
+class DepGraphTestCase(support.LoggingSilencer,
+                       unittest.TestCase):
+
+    DISTROS_DIST = ('choxie', 'grammar', 'towel-stuff')
+    DISTROS_EGG  = ('bacon', 'banana', 'strawberry', 'cheese')
+    EDGE = re.compile(
+           r'"(?P<from>.*)" -> "(?P<to>.*)" \[label="(?P<label>.*)"\]'
+           )
+
+    def checkLists(self, l1, l2):
+        """ Compare two lists without taking the order into consideration """
+        self.assertListEqual(sorted(l1), sorted(l2))
+
+    def setUp(self):
+        super(unittest.TestCase, self).setUp()
+        path = os.path.join(os.path.dirname(__file__), '..', '_backport',
+                            'tests', 'fake_dists')
+        path = os.path.abspath(path)
+        self.sys_path = sys.path[:]
+        sys.path[0:0] = [path]
+
+    def test_generate_graph(self):
+        dists = []
+        for name in self.DISTROS_DIST:
+            dist = pkgutil.get_distribution(name)
+            self.assertNotEqual(dist, None)
+            dists.append(dist)
+
+        choxie, grammar, towel = dists
+
+        graph = depgraph.generate_graph(dists)
+
+        deps = [(x.name, y) for (x,y) in graph.adjacency_list[choxie]]
+        self.checkLists([('towel-stuff', 'towel-stuff (0.1)')], deps)
+        self.assertTrue(choxie in graph.reverse_list[towel])
+        self.checkLists(graph.missing[choxie], [])
+
+        deps = [(x.name, y) for (x,y) in graph.adjacency_list[grammar]]
+        self.checkLists([], deps)
+        self.checkLists(graph.missing[grammar], ['truffles (>=1.2)'])
+
+        deps = [(x.name, y) for (x,y) in graph.adjacency_list[towel]]
+        self.checkLists([], deps)
+        self.checkLists(graph.missing[towel], ['bacon (<=0.2)'])
+
+    def test_generate_graph_egg(self):
+        dists = []
+        for name in self.DISTROS_DIST + self.DISTROS_EGG:
+            dist = pkgutil.get_distribution(name, use_egg_info=True)
+            self.assertNotEqual(dist, None)
+            dists.append(dist)
+
+        choxie, grammar, towel, bacon, banana, strawberry, cheese = dists
+
+        graph = depgraph.generate_graph(dists)
+
+        deps = [(x.name, y) for (x,y) in graph.adjacency_list[choxie]]
+        self.checkLists([('towel-stuff', 'towel-stuff (0.1)')], deps)
+        self.assertTrue(choxie in graph.reverse_list[towel])
+        self.checkLists(graph.missing[choxie], [])
+
+        deps = [(x.name, y) for (x,y) in graph.adjacency_list[grammar]]
+        self.checkLists([('bacon', 'truffles (>=1.2)')], deps)
+        self.checkLists(graph.missing[grammar], [])
+        self.assertTrue(grammar in graph.reverse_list[bacon])
+
+        deps = [(x.name, y) for (x,y) in graph.adjacency_list[towel]]
+        self.checkLists([('bacon', 'bacon (<=0.2)')], deps)
+        self.checkLists(graph.missing[towel], [])
+        self.assertTrue(towel in graph.reverse_list[bacon])
+
+        deps = [(x.name, y) for (x,y) in graph.adjacency_list[bacon]]
+        self.checkLists([], deps)
+        self.checkLists(graph.missing[bacon], [])
+
+        deps = [(x.name, y) for (x,y) in graph.adjacency_list[banana]]
+        self.checkLists([('strawberry', 'strawberry (>=0.5)')], deps)
+        self.checkLists(graph.missing[banana], [])
+        self.assertTrue(banana in graph.reverse_list[strawberry])
+
+        deps = [(x.name, y) for (x,y) in graph.adjacency_list[strawberry]]
+        self.checkLists([], deps)
+        self.checkLists(graph.missing[strawberry], [])
+
+        deps = [(x.name, y) for (x,y) in graph.adjacency_list[cheese]]
+        self.checkLists([], deps)
+        self.checkLists(graph.missing[cheese], [])
+
+    def test_dependent_dists(self):
+        dists = []
+        for name in self.DISTROS_DIST:
+            dist = pkgutil.get_distribution(name)
+            self.assertNotEqual(dist, None)
+            dists.append(dist)
+
+        choxie, grammar, towel = dists
+
+        deps = [d.name for d in depgraph.dependent_dists(dists, choxie)]
+        self.checkLists([], deps)
+
+        deps = [d.name for d in depgraph.dependent_dists(dists, grammar)]
+        self.checkLists([], deps)
+
+        deps = [d.name for d in depgraph.dependent_dists(dists, towel)]
+        self.checkLists(['choxie'], deps)
+
+
+    def test_dependent_dists_egg(self):
+        dists = []
+        for name in self.DISTROS_DIST + self.DISTROS_EGG:
+            dist = pkgutil.get_distribution(name, use_egg_info=True)
+            self.assertNotEqual(dist, None)
+            dists.append(dist)
+
+        choxie, grammar, towel, bacon, banana, strawberry, cheese = dists
+
+        deps = [d.name for d in depgraph.dependent_dists(dists, choxie)]
+        self.checkLists([], deps)
+
+        deps = [d.name for d in depgraph.dependent_dists(dists, grammar)]
+        self.checkLists([], deps)
+
+        deps = [d.name for d in depgraph.dependent_dists(dists, towel)]
+        self.checkLists(['choxie'], deps)
+
+        deps = [d.name for d in depgraph.dependent_dists(dists, bacon)]
+        self.checkLists(['choxie', 'towel-stuff', 'grammar'], deps)
+
+        deps = [d.name for d in depgraph.dependent_dists(dists, strawberry)]
+        self.checkLists(['banana'], deps)
+
+        deps = [d.name for d in depgraph.dependent_dists(dists, cheese)]
+        self.checkLists([], deps)
+
+    def test_graph_to_dot(self):
+        expected = (
+            ('towel-stuff', 'bacon', 'bacon (<=0.2)'),
+            ('grammar', 'bacon', 'truffles (>=1.2)'),
+            ('choxie', 'towel-stuff', 'towel-stuff (0.1)'),
+            ('banana', 'strawberry', 'strawberry (>=0.5)')
+        )
+
+        dists = []
+        for name in self.DISTROS_DIST + self.DISTROS_EGG:
+            dist = pkgutil.get_distribution(name, use_egg_info=True)
+            self.assertNotEqual(dist, None)
+            dists.append(dist)
+
+        graph = depgraph.generate_graph(dists)
+        buf = StringIO.StringIO()
+        depgraph.graph_to_dot(graph, buf)
+        buf.seek(0)
+        matches = []
+        lines = buf.readlines()
+        for line in lines[1:-1]: # skip the first and the last lines
+            if line[-1] == '\n':
+                line = line[:-1]
+            match = self.EDGE.match(line.strip())
+            self.assertTrue(match is not None)
+            matches.append(match.groups())
+
+        self.checkLists(matches, expected)
+
+    def tearDown(self):
+        super(unittest.TestCase, self).tearDown()
+        sys.path = self.sys_path
+
+def test_suite():
+    return unittest.makeSuite(DepGraphTestCase)
+
+if __name__ == "__main__":
+    unittest.main(defaultTest="test_suite")

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: merge merwork changes
Message-ID: <E1OVLoW-0000kK-AW@dinsdale.python.org>

tarek.ziade pushed 5d0bb31fc8c3 to distutils2:

http://hg.python.org/distutils2/rev/5d0bb31fc8c3
changeset:   323:5d0bb31fc8c3
parent:      321:d1d7547cd1bd
parent:      322:2832841bf47c
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Mon Jun 28 10:20:13 2010 +0200
summary:     merge merwork changes
files:       

diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
--- a/src/distutils2/tests/pypi_server.py
+++ b/src/distutils2/tests/pypi_server.py
@@ -29,7 +29,7 @@
             func(server=server, *args, **kwargs)
             server.stop()
         return wrapped
-    return wrapper 
+    return wrapper
 
 class PyPIServerTestCase(unittest.TestCase):
 
@@ -50,7 +50,7 @@
     """
 
     def __init__(self, test_static_path=None,
-            static_filesystem_paths=["default"], static_uri_paths=["simple"]):
+                 static_filesystem_paths=["default"], static_uri_paths=["simple"]):
         """Initialize the server.
 
         static_uri_paths and static_base_path are parameters used to provides
@@ -59,7 +59,7 @@
         """
         threading.Thread.__init__(self)
         self._run = True
-        self.httpd = HTTPServer(('', 0), PyPIRequestHandler) 
+        self.httpd = HTTPServer(('', 0), PyPIRequestHandler)
         self.httpd.RequestHandlerClass.log_request = lambda *_: None
         self.httpd.RequestHandlerClass.pypi_server = self
         self.address = (self.httpd.server_name, self.httpd.server_port)
@@ -131,7 +131,7 @@
         """
         # record the request. Read the input only on PUT or POST requests
         if self.command in ("PUT", "POST"):
-            if self.headers.dict.has_key("content-length"):
+            if 'content-length' in self.headers.dict:
                 request_data = self.rfile.read(
                     int(self.headers['content-length']))
             else:
@@ -175,8 +175,8 @@
             status, headers, data = self.pypi_server.get_next_response()
             self.make_response(data, status, headers)
 
-    def make_response(self, data, status=200, 
-            headers=[('Content-type', 'text/html')]):
+    def make_response(self, data, status=200,
+                      headers=[('Content-type', 'text/html')]):
         """Send the response to the HTTP client"""
         if not isinstance(status, int):
             try:
@@ -184,7 +184,7 @@
             except ValueError:
                 # we probably got something like YYY Codename. 
                 # Just get the first 3 digits
-                status = int(status[:3]) 
+                status = int(status[:3])
 
         self.send_response(status)
         for header, value in headers:
diff --git a/src/distutils2/tests/test_pypi_dist.py b/src/distutils2/tests/test_pypi_dist.py
--- a/src/distutils2/tests/test_pypi_dist.py
+++ b/src/distutils2/tests/test_pypi_dist.py
@@ -1,16 +1,16 @@
-"""Tests for the distutils2.pypi.dist module.
-"""
+"""Tests for the distutils2.pypi.dist module."""
+
+import os
+import shutil
+import tempfile
 
 from distutils2.tests.pypi_server import use_pypi_server
 from distutils2.tests.support import unittest
 from distutils2.version import VersionPredicate
 from distutils2.pypi.errors import MD5HashDoesNotMatch
-from distutils2.pypi.dist import PyPIDistribution as Dist, \
-    PyPIDistributions as Dists, split_archive_name
-
-import tempfile
-import os
-import shutil
+from distutils2.pypi.dist import (PyPIDistribution as Dist,
+                                  PyPIDistributions as Dists,
+                                  split_archive_name)
 
 
 class TestPyPIDistribution(unittest.TestCase):
diff --git a/src/distutils2/tests/test_pypi_server.py b/src/distutils2/tests/test_pypi_server.py
--- a/src/distutils2/tests/test_pypi_server.py
+++ b/src/distutils2/tests/test_pypi_server.py
@@ -23,9 +23,9 @@
         self.assertEqual(len(server.requests), 1)
         handler, request_data = server.requests[-1]
         self.assertIn("Rock Around The Bunker", request_data)
-        self.assertTrue(handler.headers.dict.has_key("x-test-header"))
-        self.assertEqual(handler.headers.dict["x-test-header"], 
-            "Mister Iceberg")
+        self.assertIn("x-test-header", handler.headers.dict)
+        self.assertEqual(handler.headers.dict["x-test-header"],
+                         "Mister Iceberg")
         server.stop()
 
     def test_serve_static_content(self):

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: Minor style tweaks (whitespace,
	imports, methods)
Message-ID: <E1OVLoW-0000kG-9a@dinsdale.python.org>

tarek.ziade pushed 2832841bf47c to distutils2:

http://hg.python.org/distutils2/rev/2832841bf47c
changeset:   322:2832841bf47c
parent:      320:1b75f9a8dd54
user:        ?ric Araujo <merwok at netwok.org>
date:        Sun Jun 27 01:30:13 2010 +0200
summary:     Minor style tweaks (whitespace, imports, methods)
files:       src/distutils2/tests/pypi_server.py, src/distutils2/tests/test_pypi_dist.py, src/distutils2/tests/test_pypi_server.py

diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
--- a/src/distutils2/tests/pypi_server.py
+++ b/src/distutils2/tests/pypi_server.py
@@ -29,7 +29,7 @@
             func(server=server, *args, **kwargs)
             server.stop()
         return wrapped
-    return wrapper 
+    return wrapper
 
 class PyPIServerTestCase(unittest.TestCase):
 
@@ -50,7 +50,7 @@
     """
 
     def __init__(self, test_static_path=None,
-            static_filesystem_paths=["default"], static_uri_paths=["simple"]):
+                 static_filesystem_paths=["default"], static_uri_paths=["simple"]):
         """Initialize the server.
 
         static_uri_paths and static_base_path are parameters used to provides
@@ -59,7 +59,7 @@
         """
         threading.Thread.__init__(self)
         self._run = True
-        self.httpd = HTTPServer(('', 0), PyPIRequestHandler) 
+        self.httpd = HTTPServer(('', 0), PyPIRequestHandler)
         self.httpd.RequestHandlerClass.log_request = lambda *_: None
         self.httpd.RequestHandlerClass.pypi_server = self
         self.address = (self.httpd.server_name, self.httpd.server_port)
@@ -131,7 +131,7 @@
         """
         # record the request. Read the input only on PUT or POST requests
         if self.command in ("PUT", "POST"):
-            if self.headers.dict.has_key("content-length"):
+            if 'content-length' in self.headers.dict:
                 request_data = self.rfile.read(
                     int(self.headers['content-length']))
             else:
@@ -175,8 +175,8 @@
             status, headers, data = self.pypi_server.get_next_response()
             self.make_response(data, status, headers)
 
-    def make_response(self, data, status=200, 
-            headers=[('Content-type', 'text/html')]):
+    def make_response(self, data, status=200,
+                      headers=[('Content-type', 'text/html')]):
         """Send the response to the HTTP client"""
         if not isinstance(status, int):
             try:
@@ -184,7 +184,7 @@
             except ValueError:
                 # we probably got something like YYY Codename. 
                 # Just get the first 3 digits
-                status = int(status[:3]) 
+                status = int(status[:3])
 
         self.send_response(status)
         for header, value in headers:
diff --git a/src/distutils2/tests/test_pypi_dist.py b/src/distutils2/tests/test_pypi_dist.py
--- a/src/distutils2/tests/test_pypi_dist.py
+++ b/src/distutils2/tests/test_pypi_dist.py
@@ -1,16 +1,16 @@
-"""Tests for the distutils2.pypi.dist module.
-"""
+"""Tests for the distutils2.pypi.dist module."""
+
+import os
+import shutil
+import tempfile
 
 from distutils2.tests.pypi_server import use_pypi_server
 from distutils2.tests.support import unittest
 from distutils2.version import VersionPredicate
 from distutils2.pypi.errors import MD5HashDoesNotMatch
-from distutils2.pypi.dist import PyPIDistribution as Dist, \
-    PyPIDistributions as Dists, split_archive_name
-
-import tempfile
-import os
-import shutil
+from distutils2.pypi.dist import (PyPIDistribution as Dist,
+                                  PyPIDistributions as Dists,
+                                  split_archive_name)
 
 
 class TestPyPIDistribution(unittest.TestCase):
diff --git a/src/distutils2/tests/test_pypi_server.py b/src/distutils2/tests/test_pypi_server.py
--- a/src/distutils2/tests/test_pypi_server.py
+++ b/src/distutils2/tests/test_pypi_server.py
@@ -23,9 +23,9 @@
         self.assertEqual(len(server.requests), 1)
         handler, request_data = server.requests[-1]
         self.assertIn("Rock Around The Bunker", request_data)
-        self.assertTrue(handler.headers.dict.has_key("x-test-header"))
-        self.assertEqual(handler.headers.dict["x-test-header"], 
-            "Mister Iceberg")
+        self.assertIn("x-test-header", handler.headers.dict)
+        self.assertEqual(handler.headers.dict["x-test-header"],
+                         "Mister Iceberg")
         server.stop()
 
     def test_serve_static_content(self):

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: Add a fallback mechanism to use
	mirrors
Message-ID: <E1OVLoW-0000k0-6R@dinsdale.python.org>

tarek.ziade pushed 7d8b2a7faeca to distutils2:

http://hg.python.org/distutils2/rev/7d8b2a7faeca
changeset:   318:7d8b2a7faeca
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Tue Jun 22 15:46:13 2010 +0200
summary:     Add a fallback mechanism to use mirrors
files:       docs/source/pypi.rst, src/distutils2/pypi/dist.py, src/distutils2/pypi/errors.py, src/distutils2/pypi/simple.py, src/distutils2/tests/pypiserver/foo_bar_baz/simple/bar/index.html, src/distutils2/tests/pypiserver/foo_bar_baz/simple/baz/index.html, src/distutils2/tests/pypiserver/foo_bar_baz/simple/foo/index.html, src/distutils2/tests/pypiserver/foo_bar_baz/simple/index.html, src/distutils2/tests/test_pypi_simple.py

diff --git a/docs/source/pypi.rst b/docs/source/pypi.rst
--- a/docs/source/pypi.rst
+++ b/docs/source/pypi.rst
@@ -1,6 +1,6 @@
-=================
-The PyPI Module
-=================
+=========================================
+Tools to query PyPI: the PyPI package
+=========================================
 
 Distutils2 comes with a module (eg. `distutils2.pypi`) which contains
 facilities to access the Python Package Index (named "pypi", and avalaible on
@@ -15,34 +15,6 @@
 Distutils2 provides two python modules to ease the work with those two APIs:
 `distutils2.pypi.simple` and `distutils2.pypi.xmlrpc`.
 
-`distutils2.pypi.dist`
-======================
-
-Both `SimpleIndex` and `XmlRpcIndex` classes works with the classes provided
-in the `pypi.dist` package.
-
-`PyPIDistribution`
-------------------
-
-`PyPIDistribution` is a simple class that defines the following attributes:
-
-:name:
-    The name of the package. `foobar` in our exemples here
-:version:
-    The version of the package
-:location:
-    If the files from the archive has been downloaded, here is the path where
-    you can find them.
-:url:
-    The url of the distribution
-
-
-`PyPIDistributions`
--------------------
-
-The `dist` module also provides another class, to work with lists of 
-`PyPIDistribution` classes.
-
 
 Requesting information via the "simple" API `distutils2.pypi.simple`
 ====================================================================
@@ -64,8 +36,17 @@
     * Things that will end up in too long index processing (like "finding all
       distributions with a specific version, no matters the name")
 
+API
+----
+
+.. autoclass:: distutils2.pypi.simple.SimpleIndex
+    :members:
+
+Usage Exemples
+---------------
+
 Request PyPI to get a specific distribution
---------------------------------------------
+++++++++++++++++++++++++++++++++++++++++++++
 
 Supposing you want to scan the PyPI index to get a list of distributions for 
 the "foobar" project. You can use the "search" method for that.::
@@ -89,7 +70,7 @@
     <PyPIDistribution "foobar 1.1">
 
 Download distributions
-----------------------
++++++++++++++++++++++++
 
 As it can get the urls of distributions provided by PyPI, the `Simple` client
 also can download the distributions and put it for you in a temporary
@@ -107,20 +88,26 @@
 will try another time, then if fails again, raise `MD5HashDoesNotMatchError`.
 
 Requesting external pages
--------------------------
++++++++++++++++++++++++++
 
 The default behavior for distutils2 is to follow the links provided
-by distributions within their metadatas, and find distributions related
+by HTML pages in the "simple index", to find distributions related
 downloads.
 
-When downloading, then, it will first use the packages found ont PyPI, then
-falling back to external webpages if needed.
-
 It's possible to tell the PyPIClient to not follow external links by specifying
 a list of allowed hosts::
 
     >>> client = SimpleIndex(hosts=("*.python.org"))
 
+Working with mirrors
++++++++++++++++++++++
+
+The SimpleClient implement a fallback mechanism to switch from one mirror to 
+another, the simple way. All you need to do is to provide a list of mirrors to
+it at the instanciation time::
+
+    >>> client = SimpleClient(mirrors=['http://mirror1/', 'http://mirror2/'])
+
 
 Requesting informations via XML-RPC (`distutils2.pypi.XmlRpcIndex`)
 ==========================================================================
@@ -135,6 +122,40 @@
     >>> client = XmlRpcIndex()
 
 
+PyPI Distributions
+==================
+
+Both `SimpleIndex` and `XmlRpcIndex` classes works with the classes provided
+in the `pypi.dist` package.
+
+`PyPIDistribution`
+------------------
+
+`PyPIDistribution` is a simple class that defines the following attributes:
+
+:name:
+    The name of the package. `foobar` in our exemples here
+:version:
+    The version of the package
+:location:
+    If the files from the archive has been downloaded, here is the path where
+    you can find them.
+:url:
+    The url of the distribution
+
+.. autoclass:: distutils2.pypi.dist.PyPIDistribution
+    :members:
+
+`PyPIDistributions`
+-------------------
+
+The `dist` module also provides another class, to work with lists of 
+`PyPIDistribution` classes. It allow to filter results and is used as a 
+container of 
+
+.. autoclass:: distutils2.pypi.dist.PyPIDistributions
+    :members:
+
 At a higher level
 =================
 
diff --git a/src/distutils2/pypi/dist.py b/src/distutils2/pypi/dist.py
--- a/src/distutils2/pypi/dist.py
+++ b/src/distutils2/pypi/dist.py
@@ -34,9 +34,7 @@
         """Build a Distribution from a url archive (egg or zip or tgz).
 
         :param url: complete url of the distribution
-        :param probable_dist_name: the probable name of the distribution. This
-        could be useful when multiple name can be assumed from the archive
-        name.
+        :param probable_dist_name: A probable name of the distribution. 
         """
         # if the url contains a md5 hash, get it.
         md5_hash = None
diff --git a/src/distutils2/pypi/errors.py b/src/distutils2/pypi/errors.py
--- a/src/distutils2/pypi/errors.py
+++ b/src/distutils2/pypi/errors.py
@@ -23,3 +23,7 @@
 
 class MD5HashDoesNotMatch(DownloadError):
     """Compared MD5 hashes does not match"""
+
+
+class UnableToDownload(PyPIError):
+    """All mirrors have been tried, without success"""
diff --git a/src/distutils2/pypi/simple.py b/src/distutils2/pypi/simple.py
--- a/src/distutils2/pypi/simple.py
+++ b/src/distutils2/pypi/simple.py
@@ -15,7 +15,8 @@
 from distutils2.version import VersionPredicate
 from distutils2.pypi.dist import PyPIDistribution, PyPIDistributions, \
     EXTENSIONS
-from distutils2.pypi.errors import PyPIError, DistributionNotFound
+from distutils2.pypi.errors import PyPIError, DistributionNotFound, \
+    DownloadError, UnableToDownload
 from distutils2 import __version__ as __distutils2_version__
 
 # -- Constants -----------------------------------------------
@@ -43,11 +44,13 @@
     """Decorator to add a socket timeout when requesting pages on PyPI.
     """
     def _socket_timeout(func):
-        def _socket_timeout(*args, **kwargs):
+        def _socket_timeout(self, *args, **kwargs):
             old_timeout = socket.getdefaulttimeout()
+            if hasattr(self, "_timeout"):
+                timeout = self._timeout
             socket.setdefaulttimeout(timeout)
             try:
-                return func(*args, **kwargs)
+                return func(self, *args, **kwargs)
             finally:
                 socket.setdefaulttimeout(old_timeout)
         return _socket_timeout
@@ -59,7 +62,7 @@
     """
 
     def __init__(self, url=PYPI_DEFAULT_INDEX_URL, hosts=DEFAULT_HOSTS,
-        mirrors=[]):
+        mirrors=[], timeout=SOCKET_TIMEOUT):
         """Class constructor.
 
         :param index_url: the url of the simple index to search on.
@@ -70,9 +73,10 @@
         :param mirrors: a list of mirrors to check out if problems occurs while
         working with the one given in "url"
         """
-        self.index_url = url
-        self.mirrors = mirrors
-        self._hosts = hosts
+        self._index_urls = [url]
+        self._index_urls.extend(mirrors)
+        self._current_index_url = 0
+        self._timeout = timeout
         # create a regexp to match all given hosts
         self._allowed_hosts = re.compile('|'.join(map(translate, hosts))).match
 
@@ -99,8 +103,9 @@
         requirements.
 
         :param requirements: A project name and it's distribution, using
-        version specifiers, as described in PEP345. You can pass either a
-        version.VersionPredicate or a string.
+                             version specifiers, as described in PEP345. 
+        :type requirements:  You can pass either a version.VersionPredicate 
+                             or a string.
         """
         requirements = self._get_version_predicate(requirements)
 
@@ -124,6 +129,8 @@
         is given, creates and return one.
 
         Returns the complete absolute path to the downloaded archive.
+
+        :param requirements: The same as the find attribute of `find`. 
         """
         return self.get(requirements).download(path=temp_path)
 
@@ -134,7 +141,20 @@
         if isinstance(requirements, str):
             requirements = VersionPredicate(requirements)
         return requirements
+    
+    @property
+    def index_url(self):
+        return self._index_urls[self._current_index_url] 
 
+    def _switch_to_next_mirror(self):
+        """Switch to the next mirror (eg. point self.index_url to the next
+        url.
+        """
+        if self._current_index_url < len(self._index_urls):
+            self._current_index_url = self._current_index_url + 1
+        else:
+            raise UnableToDownload("All mirrors fails")
+    
     def _is_browsable(self, url):
         """Tell if the given URL needs to be browsed or not, according to the
         object internal attributes.
@@ -229,9 +249,16 @@
 
         :param name: the name of the project to find the page
         """
-        # Browse and index the content of the given PyPI page.
-        url = self.index_url + name + "/"
-        self._process_url(url, name)
+        try:
+            # Browse and index the content of the given PyPI page.
+            url = self.index_url + name + "/"
+            self._process_url(url, name)
+        except DownloadError:
+            # if an error occurs, try with the next index_url 
+            # (provided by the mirrors)
+            self._switch_to_next_mirror()
+            self._distributions.clear()
+            self._process_pypi_page(name)
 
     @socket_timeout()
     def _open_url(self, url):
@@ -273,12 +300,12 @@
         except urllib2.HTTPError, v:
             return v
         except urllib2.URLError, v:
-            raise PyPIError("Download error for %s: %s" % (url, v.reason))
+            raise DownloadError("Download error for %s: %s" % (url, v.reason))
         except httplib.BadStatusLine, v:
-            raise PyPIError('%s returned a bad status line. '
+            raise DownloadError('%s returned a bad status line. '
                 'The server might be down, %s' % (url, v.line))
         except httplib.HTTPException, v:
-            raise PyPIError("Download error for %s: %s" % (url, v))
+            raise DownloadError("Download error for %s: %s" % (url, v))
 
     def _decode_entity(self, match):
         what = match.group(1)
diff --git a/src/distutils2/tests/pypiserver/foo_bar_baz/simple/bar/index.html b/src/distutils2/tests/pypiserver/foo_bar_baz/simple/bar/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/foo_bar_baz/simple/bar/index.html
@@ -0,0 +1,6 @@
+<html><head><title>Links for bar</title></head><body><h1>Links for bar</h1>
+<a rel="download" href="../../packages/source/F/bar/bar-1.0.tar.gz">bar-1.0.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/bar/bar-1.0.1.tar.gz">bar-1.0.1.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/bar/bar-2.0.tar.gz">bar-2.0.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/bar/bar-2.0.1.tar.gz">bar-2.0.1.tar.gz</a><br/> 
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/foo_bar_baz/simple/baz/index.html b/src/distutils2/tests/pypiserver/foo_bar_baz/simple/baz/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/foo_bar_baz/simple/baz/index.html
@@ -0,0 +1,6 @@
+<html><head><title>Links for baz</title></head><body><h1>Links for baz</h1>
+<a rel="download" href="../../packages/source/F/baz/baz-1.0.tar.gz">baz-1.0.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/baz/baz-1.0.1.tar.gz">baz-1.0.1.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/baz/baz-2.0.tar.gz">baz-2.0.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/baz/baz-2.0.1.tar.gz">baz-2.0.1.tar.gz</a><br/> 
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/foo_bar_baz/simple/foo/index.html b/src/distutils2/tests/pypiserver/foo_bar_baz/simple/foo/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/foo_bar_baz/simple/foo/index.html
@@ -0,0 +1,6 @@
+<html><head><title>Links for foo</title></head><body><h1>Links for foo</h1>
+<a rel="download" href="../../packages/source/F/foo/foo-1.0.tar.gz">foo-1.0.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/foo/foo-1.0.1.tar.gz">foo-1.0.1.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/foo/foo-2.0.tar.gz">foo-2.0.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/foo/foo-2.0.1.tar.gz">foo-2.0.1.tar.gz</a><br/> 
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/foo_bar_baz/simple/index.html b/src/distutils2/tests/pypiserver/foo_bar_baz/simple/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/foo_bar_baz/simple/index.html
@@ -0,0 +1,3 @@
+<a href="foo/">foo/</a> 
+<a href="bar/">bar/</a> 
+<a href="baz/">baz/</a> 
diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -8,7 +8,7 @@
 import unittest2
 import urllib2
 
-from distutils2.tests.pypi_server import use_pypi_server
+from distutils2.tests.pypi_server import use_pypi_server, PyPIServer
 from distutils2.pypi import simple
 
 from distutils2.errors import DistutilsError
@@ -222,11 +222,24 @@
         self.assertIn("%s/foobar-2.0.tar.gz" % server.full_address,
             index._processed_urls)  # linked from external homepage (rel)
 
-#    def test_uses_mirrors(self):
-#        """When the main repository seems down, try using the given mirrors"""
-#        server = 
-#        mirror = 
-#
+    def test_uses_mirrors(self):
+        """When the main repository seems down, try using the given mirrors"""
+        server = PyPIServer("foo_bar_baz")
+        mirror = PyPIServer("foo_bar_baz")
+        mirror.start()  # we dont start the server here
+        
+        try:
+            # create the index using both servers
+            index = simple.SimpleIndex(server.full_address + "/simple/", 
+                hosts=('*',), timeout=1,  # set the timeout to 1s for the tests
+                mirrors=[mirror.full_address + "/simple/",])
+            
+            # this should not raise a timeout
+            self.assertEqual(4, len(index.find("foo")))
+        except Exception, e:
+            mirror.stop()
+            raise e
+
 def test_suite():
     return unittest2.makeSuite(PyPISimpleTestCase)
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: Fix some tests to use the
	compatibility layer of	unittest2 (useful for
Message-ID: <E1OVLoW-0000kR-Bt@dinsdale.python.org>

tarek.ziade pushed 5e1578fd8d46 to distutils2:

http://hg.python.org/distutils2/rev/5e1578fd8d46
changeset:   325:5e1578fd8d46
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Wed Jun 30 16:45:48 2010 +0200
summary:     Fix some tests to use the compatibility layer of unittest2 (useful for
files:       src/distutils2/_backport/tests/test_pkgutil.py

diff --git a/src/distutils2/_backport/tests/test_pkgutil.py b/src/distutils2/_backport/tests/test_pkgutil.py
--- a/src/distutils2/_backport/tests/test_pkgutil.py
+++ b/src/distutils2/_backport/tests/test_pkgutil.py
@@ -1,6 +1,7 @@
 # -*- coding: utf-8 -*-
 """Tests for PEP 376 pkgutil functionality"""
 import unittest2
+import unittest2.compatibility
 import sys
 import os
 import csv

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: Stop the pypi mocked server even if
	the tests fails.
Message-ID: <E1OVLoW-0000ka-Dt@dinsdale.python.org>

tarek.ziade pushed 7c7afc6f542f to distutils2:

http://hg.python.org/distutils2/rev/7c7afc6f542f
changeset:   327:7c7afc6f542f
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Thu Jul 01 12:44:02 2010 +0200
summary:     Stop the pypi mocked server even if the tests fails.
files:       src/distutils2/tests/pypi_server.py

diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
--- a/src/distutils2/tests/pypi_server.py
+++ b/src/distutils2/tests/pypi_server.py
@@ -26,8 +26,10 @@
         def wrapped(*args, **kwargs):
             server = PyPIServer(*server_args, **server_kwargs)
             server.start()
-            func(server=server, *args, **kwargs)
-            server.stop()
+            try:
+                func(server=server, *args, **kwargs)
+            finally:
+                server.stop()
         return wrapped
     return wrapper
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: local files support
Message-ID: <E1OVLoW-0000kq-Ga@dinsdale.python.org>

tarek.ziade pushed 187764b18476 to distutils2:

http://hg.python.org/distutils2/rev/187764b18476
changeset:   331:187764b18476
parent:      324:f28c41acac7c
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Tue Jun 29 13:45:26 2010 +0200
summary:     local files support
files:       src/distutils2/pypi/simple.py, src/distutils2/tests/test_pypi_simple.py

diff --git a/src/distutils2/pypi/simple.py b/src/distutils2/pypi/simple.py
--- a/src/distutils2/pypi/simple.py
+++ b/src/distutils2/pypi/simple.py
@@ -75,6 +75,8 @@
                         working with the one given in "url"
         :param timeout: time in seconds to consider a url has timeouted.
         """
+        if not index_url.endswith("/"):
+            index_url += "/"
         self._index_urls = [index_url]
         self._index_urls.extend(mirrors)
         self._current_index_url = 0
@@ -169,10 +171,8 @@
         """
         # if _index_url is contained in the given URL, we are browsing the
         # index, and it's always "browsable".
-        # We asume here that if the url starts with "." or "..", it's browsable
-        # too. This is useful as the simple index make heavy use of relative
-        # URLS.
-        if self.index_url in url:
+        # local files are always considered browable resources
+        if self.index_url in url or urlparse.urlparse(url)[0] == "file":
             return True
         elif self.follow_externals is True:
             if self._allowed_hosts(urlparse.urlparse(url)[1]):  # 1 is netloc
@@ -287,7 +287,8 @@
 
     @socket_timeout()
     def _open_url(self, url):
-        """Open a urllib2 request, handling HTTP authentication.
+        """Open a urllib2 request, handling HTTP authentication, and local
+        files support.
 
         """
         try:
@@ -297,6 +298,11 @@
                 auth, host = urllib2.splituser(netloc)
             else:
                 auth = None
+            
+            # add index.html automatically for filesystem paths
+            if scheme == 'file':
+                if url.endswith('/'):
+                    url += "index.html"
 
             if auth:
                 auth = "Basic " + \
diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -8,7 +8,8 @@
 import unittest2
 import urllib2
 
-from distutils2.tests.pypi_server import use_pypi_server, PyPIServer
+from distutils2.tests.pypi_server import use_pypi_server, PyPIServer, \
+                                         PYPI_DEFAULT_STATIC_PATH
 from distutils2.pypi import simple
 
 from distutils2.errors import DistutilsError
@@ -274,6 +275,14 @@
                          generator.next())
         self.assertRaises(StopIteration, generator.next)
 
+    def test_browse_local_files(self):
+        """Test that we can browse local files"""
+        index_path = os.sep.join(["file://" + PYPI_DEFAULT_STATIC_PATH,
+                                  "test_found_links", "simple"])
+        index = simple.SimpleIndex(index_path)
+        dists = index.find("foobar")
+        self.assertEqual(4, len(dists))
+
 def test_suite():
     return unittest2.makeSuite(PyPISimpleTestCase)
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: Remove the hardcoded md5 hash checks
	from pypi.dist.
Message-ID: <E1OVLoW-0000ke-Em@dinsdale.python.org>

tarek.ziade pushed 8e7ea96738eb to distutils2:

http://hg.python.org/distutils2/rev/8e7ea96738eb
changeset:   328:8e7ea96738eb
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Thu Jul 01 12:46:15 2010 +0200
summary:     Remove the hardcoded md5 hash checks from pypi.dist.
files:       src/distutils2/pypi/dist.py, src/distutils2/pypi/errors.py, src/distutils2/tests/test_pypi_dist.py, src/distutils2/tests/test_pypi_simple.py

diff --git a/src/distutils2/pypi/dist.py b/src/distutils2/pypi/dist.py
--- a/src/distutils2/pypi/dist.py
+++ b/src/distutils2/pypi/dist.py
@@ -8,12 +8,11 @@
 import urllib
 import tempfile
 try:
-    from hashlib import md5
+    import hashlib
 except ImportError:
-    from md5 import md5
-
+    from distutils2._backport import hashlib
 from distutils2.version import suggest_normalized_version
-from distutils2.pypi.errors import MD5HashDoesNotMatch
+from distutils2.pypi.errors import HashDoesNotMatch
 
 EXTENSIONS = ".tar.gz .tar.bz2 .tar .zip .tgz .egg".split()
 MD5_HASH = re.compile(r'^.*#md5=([a-f0-9]+)$')
@@ -57,19 +56,20 @@
 
         name, version = split_archive_name(archive_name)
         if extension_matched is True:
-            return PyPIDistribution(name, version, url=url, md5_hash=md5_hash, 
-                                    is_external=is_external)
+            return PyPIDistribution(name, version, url=url, hashname="md5", 
+                                    hashval=md5_hash, is_external=is_external)
 
-    def __init__(self, name, version, type=None, url=None, md5_hash=None, 
-                 is_external=True):
+    def __init__(self, name, version, type=None, url=None, hashname=None, 
+                 hashval=None, is_external=True):
         """Create a new instance of PyPIDistribution.
 
         :param name: the name of the distribution
         :param version: the version of the distribution
         :param type: the type of the dist (eg. source, bin-*, etc.)
         :param url: URL where we found this distribution
-        :param md5_hash: the MD5 hash of the distribution. That's used to 
-                         check that the download is good enough.
+        :param hashname: the name of the hash we want to use. Refer to the
+                         hashlib.new documentation for more information.
+        :param hashval: the hash value.
         :param is_external: we need to know if the provided url comes from an 
                             index browsing, or from an external resource.
 
@@ -88,13 +88,14 @@
         # order to dont make the selection process multiple times.
         self._urls = []
         self._url = None
-        self.add_url(url, md5_hash, is_external)
+        self.add_url(url, hashname, hashval, is_external)
     
-    def add_url(self, url, md5=None, is_external=True):
+    def add_url(self, url, hashname=None, hashval=None, is_external=True):
         """Add a new url to the list of urls"""
         self._urls.append({
             'url': url,
-            'md5': md5,
+            'hashname': hashname,
+            'hashval': hashval,
             'is_external': is_external,
         })
         # reset the url selection process
@@ -136,15 +137,16 @@
 
     def _check_md5(self, filename):
         """Check that the md5 checksum of the given file matches the one in
-        self._md5_hash."""
-        md5_hash = self.url['md5']
-        if md5_hash is not None:
+        url param"""
+        hashname = self.url['hashname']
+        expected_hashval = self.url['hashval']
+        if not None in (expected_hashval, hashname):
             f = open(filename)
-            hash = md5()
-            hash.update(f.read())
-            if hash.hexdigest() != md5_hash:
-                raise MD5HashDoesNotMatch("%s instead of %s"
-                    % (hash.hexdigest(), md5_hash))
+            hashval = hashlib.new(hashname)
+            hashval.update(f.read())
+            if hashval.hexdigest() != expected_hashval:
+                raise HashDoesNotMatch("got %s instead of %s"
+                    % (hashval.hexdigest(), expected_hashval))
 
     def __repr__(self):
         return "<%s %s (%s)>" \
diff --git a/src/distutils2/pypi/errors.py b/src/distutils2/pypi/errors.py
--- a/src/distutils2/pypi/errors.py
+++ b/src/distutils2/pypi/errors.py
@@ -21,7 +21,7 @@
     """An error has occurs while downloading"""
 
 
-class MD5HashDoesNotMatch(DownloadError):
+class HashDoesNotMatch(DownloadError):
     """Compared MD5 hashes does not match"""
 
 
diff --git a/src/distutils2/tests/test_pypi_dist.py b/src/distutils2/tests/test_pypi_dist.py
--- a/src/distutils2/tests/test_pypi_dist.py
+++ b/src/distutils2/tests/test_pypi_dist.py
@@ -7,7 +7,7 @@
 from distutils2.tests.pypi_server import use_pypi_server
 from distutils2.tests.support import unittest
 from distutils2.version import VersionPredicate
-from distutils2.pypi.errors import MD5HashDoesNotMatch
+from distutils2.pypi.errors import HashDoesNotMatch
 from distutils2.pypi.dist import (PyPIDistribution as Dist,
                                   PyPIDistributions as Dists,
                                   split_archive_name)
@@ -39,7 +39,8 @@
                 'version': '1.1b2',
                 'url': {
                     'url': 'http://test.tld/foobar-1.1b2.tar.gz',  # no hash
-                    'md5': '123123123123123',
+                    'hashval': '123123123123123',
+                    'hashname': 'md5',
                 }
             },
             'foobar-1.1-rc2.tar.gz': {  # use suggested name
@@ -68,7 +69,8 @@
         self.assertDictEqual(d.url, {
             "url": "test_url",
             "is_external": True,
-            "md5": None,
+            "hashname": None,
+            "hashval": None,
         })
 
         # add a new url
@@ -77,7 +79,8 @@
         self.assertDictEqual(d.url, {
             "url": "internal_url",
             "is_external": False,
-            "md5": None,
+            "hashname": None,
+            "hashval": None,
         })
         self.assertEqual(2, len(d._urls))
 
@@ -111,13 +114,13 @@
         url = "%s/simple/foobar/foobar-0.1.tar.gz" % server.full_address
         # check md5 if given
         dist = Dist("FooBar", "0.1", url=url,
-            md5_hash="d41d8cd98f00b204e9800998ecf8427e")
+            hashname="md5", hashval="d41d8cd98f00b204e9800998ecf8427e")
         dist.download()
 
         # a wrong md5 fails
         dist2 = Dist("FooBar", "0.1", url=url,
-            md5_hash="wrongmd5")
-        self.assertRaises(MD5HashDoesNotMatch, dist2.download)
+            hashname="md5", hashval="wrongmd5")
+        self.assertRaises(HashDoesNotMatch, dist2.download)
 
         # we can omit the md5 hash
         dist3 = Dist("FooBar", "0.1", url=url)
diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -196,7 +196,8 @@
         # we have only one link, because links are compared without md5
         self.assertEqual(len(dists), 1)
         # the link should be from the index
-        self.assertEqual('12345678901234567', dists[0].url['md5'])
+        self.assertEqual('12345678901234567', dists[0].url['hashval'])
+        self.assertEqual('md5', dists[0].url['hashname'])
 
     @use_pypi_server(static_filesystem_paths=["with_norel_links"],
         static_uri_paths=["simple", "external"])

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: Support of prefer_final and
	prefer_source in pypi.simple
Message-ID: <E1OVLoW-0000l6-Jn@dinsdale.python.org>

tarek.ziade pushed 2dfbdecc2dc3 to distutils2:

http://hg.python.org/distutils2/rev/2dfbdecc2dc3
changeset:   335:2dfbdecc2dc3
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Thu Jul 01 19:58:18 2010 +0200
summary:     Support of prefer_final and prefer_source in pypi.simple
files:       src/distutils2/pypi/dist.py

diff --git a/src/distutils2/pypi/dist.py b/src/distutils2/pypi/dist.py
--- a/src/distutils2/pypi/dist.py
+++ b/src/distutils2/pypi/dist.py
@@ -232,6 +232,9 @@
         else:
             super(PyPIDistributions, self).append(o)
 
+    def sort(self, prefer_source=None, prefer_final=None, *args, **kwargs):
+        return super(PyPIDistributions, self).sort(*args, **kwargs)
+
 
 def split_archive_name(archive_name, probable_name=None):
     """Split an archive name into two parts: name and version.

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: Backport hashlib to be used in python
	< 2.5.
Message-ID: <E1OVLoW-0000kW-D9@dinsdale.python.org>

tarek.ziade pushed 8302f35bca4f to distutils2:

http://hg.python.org/distutils2/rev/8302f35bca4f
changeset:   326:8302f35bca4f
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Thu Jul 01 11:42:11 2010 +0200
summary:     Backport hashlib to be used in python < 2.5.
files:       src/Modules/_hashopenssl.c, src/Modules/md5.c, src/Modules/md5.h, src/Modules/md5module.c, src/Modules/sha256module.c, src/Modules/sha512module.c, src/Modules/shamodule.c, src/distutils2/_backport/hashlib.py, src/setup.py

diff --git a/src/Modules/_hashopenssl.c b/src/Modules/_hashopenssl.c
new file mode 100644
--- /dev/null
+++ b/src/Modules/_hashopenssl.c
@@ -0,0 +1,524 @@
+/* Module that wraps all OpenSSL hash algorithms */
+
+/*
+ * Copyright (C) 2005   Gregory P. Smith (greg at krypto.org)
+ * Licensed to PSF under a Contributor Agreement.
+ *
+ * Derived from a skeleton of shamodule.c containing work performed by:
+ *
+ * Andrew Kuchling (amk at amk.ca)
+ * Greg Stein (gstein at lyra.org)
+ *
+ */
+
+#define PY_SSIZE_T_CLEAN
+
+#include "Python.h"
+#include "structmember.h"
+
+#if (PY_VERSION_HEX < 0x02050000)
+#define Py_ssize_t      int
+#endif
+
+/* EVP is the preferred interface to hashing in OpenSSL */
+#include <openssl/evp.h>
+
+#define MUNCH_SIZE INT_MAX
+
+
+#ifndef HASH_OBJ_CONSTRUCTOR
+#define HASH_OBJ_CONSTRUCTOR 0
+#endif
+
+typedef struct {
+    PyObject_HEAD
+    PyObject            *name;  /* name of this hash algorithm */
+    EVP_MD_CTX          ctx;    /* OpenSSL message digest context */
+} EVPobject;
+
+
+static PyTypeObject EVPtype;
+
+
+#define DEFINE_CONSTS_FOR_NEW(Name)  \
+    static PyObject *CONST_ ## Name ## _name_obj; \
+    static EVP_MD_CTX CONST_new_ ## Name ## _ctx; \
+    static EVP_MD_CTX *CONST_new_ ## Name ## _ctx_p = NULL;
+
+DEFINE_CONSTS_FOR_NEW(md5)
+DEFINE_CONSTS_FOR_NEW(sha1)
+DEFINE_CONSTS_FOR_NEW(sha224)
+DEFINE_CONSTS_FOR_NEW(sha256)
+DEFINE_CONSTS_FOR_NEW(sha384)
+DEFINE_CONSTS_FOR_NEW(sha512)
+
+
+static EVPobject *
+newEVPobject(PyObject *name)
+{
+    EVPobject *retval = (EVPobject *)PyObject_New(EVPobject, &EVPtype);
+
+    /* save the name for .name to return */
+    if (retval != NULL) {
+        Py_INCREF(name);
+        retval->name = name;
+    }
+
+    return retval;
+}
+
+/* Internal methods for a hash object */
+
+static void
+EVP_dealloc(PyObject *ptr)
+{
+    EVP_MD_CTX_cleanup(&((EVPobject *)ptr)->ctx);
+    Py_XDECREF(((EVPobject *)ptr)->name);
+    PyObject_Del(ptr);
+}
+
+
+/* External methods for a hash object */
+
+PyDoc_STRVAR(EVP_copy__doc__, "Return a copy of the hash object.");
+
+static PyObject *
+EVP_copy(EVPobject *self, PyObject *unused)
+{
+    EVPobject *newobj;
+
+    if ( (newobj = newEVPobject(self->name))==NULL)
+        return NULL;
+
+    EVP_MD_CTX_copy(&newobj->ctx, &self->ctx);
+    return (PyObject *)newobj;
+}
+
+PyDoc_STRVAR(EVP_digest__doc__,
+"Return the digest value as a string of binary data.");
+
+static PyObject *
+EVP_digest(EVPobject *self, PyObject *unused)
+{
+    unsigned char digest[EVP_MAX_MD_SIZE];
+    EVP_MD_CTX temp_ctx;
+    PyObject *retval;
+    unsigned int digest_size;
+
+    EVP_MD_CTX_copy(&temp_ctx, &self->ctx);
+    digest_size = EVP_MD_CTX_size(&temp_ctx);
+    EVP_DigestFinal(&temp_ctx, digest, NULL);
+
+    retval = PyString_FromStringAndSize((const char *)digest, digest_size);
+    EVP_MD_CTX_cleanup(&temp_ctx);
+    return retval;
+}
+
+PyDoc_STRVAR(EVP_hexdigest__doc__,
+"Return the digest value as a string of hexadecimal digits.");
+
+static PyObject *
+EVP_hexdigest(EVPobject *self, PyObject *unused)
+{
+    unsigned char digest[EVP_MAX_MD_SIZE];
+    EVP_MD_CTX temp_ctx;
+    PyObject *retval;
+    char *hex_digest;
+    unsigned int i, j, digest_size;
+
+    /* Get the raw (binary) digest value */
+    EVP_MD_CTX_copy(&temp_ctx, &self->ctx);
+    digest_size = EVP_MD_CTX_size(&temp_ctx);
+    EVP_DigestFinal(&temp_ctx, digest, NULL);
+
+    EVP_MD_CTX_cleanup(&temp_ctx);
+
+    /* Create a new string */
+    /* NOTE: not thread safe! modifying an already created string object */
+    /* (not a problem because we hold the GIL by default) */
+    retval = PyString_FromStringAndSize(NULL, digest_size * 2);
+    if (!retval)
+	    return NULL;
+    hex_digest = PyString_AsString(retval);
+    if (!hex_digest) {
+	    Py_DECREF(retval);
+	    return NULL;
+    }
+
+    /* Make hex version of the digest */
+    for(i=j=0; i<digest_size; i++) {
+        char c;
+        c = (digest[i] >> 4) & 0xf;
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+        c = (digest[i] & 0xf);
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+    }
+    return retval;
+}
+
+PyDoc_STRVAR(EVP_update__doc__,
+"Update this hash object's state with the provided string.");
+
+static PyObject *
+EVP_update(EVPobject *self, PyObject *args)
+{
+    unsigned char *cp;
+    Py_ssize_t len;
+
+    if (!PyArg_ParseTuple(args, "s#:update", &cp, &len))
+        return NULL;
+
+    if (len > 0 && len <= MUNCH_SIZE) {
+    EVP_DigestUpdate(&self->ctx, cp, Py_SAFE_DOWNCAST(len, Py_ssize_t,
+                                                      unsigned int));
+    } else {
+        Py_ssize_t offset = 0;
+        while (len) {
+            unsigned int process = len > MUNCH_SIZE ? MUNCH_SIZE : len;
+            EVP_DigestUpdate(&self->ctx, cp + offset, process);
+            len -= process;
+            offset += process;
+        }
+    }
+    Py_INCREF(Py_None);
+    return Py_None;
+}
+
+static PyMethodDef EVP_methods[] = {
+    {"update",	  (PyCFunction)EVP_update,    METH_VARARGS, EVP_update__doc__},
+    {"digest",	  (PyCFunction)EVP_digest,    METH_NOARGS,  EVP_digest__doc__},
+    {"hexdigest", (PyCFunction)EVP_hexdigest, METH_NOARGS,  EVP_hexdigest__doc__},
+    {"copy",	  (PyCFunction)EVP_copy,      METH_NOARGS,  EVP_copy__doc__},
+    {NULL,	  NULL}		/* sentinel */
+};
+
+static PyObject *
+EVP_get_block_size(EVPobject *self, void *closure)
+{
+    return PyInt_FromLong(EVP_MD_CTX_block_size(&((EVPobject *)self)->ctx));
+}
+
+static PyObject *
+EVP_get_digest_size(EVPobject *self, void *closure)
+{
+    return PyInt_FromLong(EVP_MD_CTX_size(&((EVPobject *)self)->ctx));
+}
+
+static PyMemberDef EVP_members[] = {
+    {"name", T_OBJECT, offsetof(EVPobject, name), READONLY, PyDoc_STR("algorithm name.")},
+    {NULL}  /* Sentinel */
+};
+
+static PyGetSetDef EVP_getseters[] = {
+    {"digest_size",
+     (getter)EVP_get_digest_size, NULL,
+     NULL,
+     NULL},
+    {"block_size",
+     (getter)EVP_get_block_size, NULL,
+     NULL,
+     NULL},
+    /* the old md5 and sha modules support 'digest_size' as in PEP 247.
+     * the old sha module also supported 'digestsize'.  ugh. */
+    {"digestsize",
+     (getter)EVP_get_digest_size, NULL,
+     NULL,
+     NULL},
+    {NULL}  /* Sentinel */
+};
+
+
+static PyObject *
+EVP_repr(PyObject *self)
+{
+    char buf[100];
+    PyOS_snprintf(buf, sizeof(buf), "<%s HASH object @ %p>",
+            PyString_AsString(((EVPobject *)self)->name), self);
+    return PyString_FromString(buf);
+}
+
+#if HASH_OBJ_CONSTRUCTOR
+static int
+EVP_tp_init(EVPobject *self, PyObject *args, PyObject *kwds)
+{
+    static char *kwlist[] = {"name", "string", NULL};
+    PyObject *name_obj = NULL;
+    char *nameStr;
+    unsigned char *cp = NULL;
+    Py_ssize_t len = 0;
+    const EVP_MD *digest;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwds, "O|s#:HASH", kwlist,
+                                     &name_obj, &cp, &len)) {
+        return -1;
+    }
+
+    if (!PyArg_Parse(name_obj, "s", &nameStr)) {
+        PyErr_SetString(PyExc_TypeError, "name must be a string");
+        return -1;
+    }
+
+    digest = EVP_get_digestbyname(nameStr);
+    if (!digest) {
+        PyErr_SetString(PyExc_ValueError, "unknown hash function");
+        return -1;
+    }
+    EVP_DigestInit(&self->ctx, digest);
+
+    self->name = name_obj;
+    Py_INCREF(self->name);
+
+    if (cp && len) {
+        if (len > 0 && len <= MUNCH_SIZE) {
+        EVP_DigestUpdate(&self->ctx, cp, Py_SAFE_DOWNCAST(len, Py_ssize_t,
+                                                          unsigned int));
+        } else {
+            Py_ssize_t offset = 0;
+            while (len) {
+                unsigned int process = len > MUNCH_SIZE ? MUNCH_SIZE : len;
+                EVP_DigestUpdate(&self->ctx, cp + offset, process);
+                len -= process;
+                offset += process;
+            }
+        }
+    }
+    
+    return 0;
+}
+#endif
+
+
+PyDoc_STRVAR(hashtype_doc,
+"A hash represents the object used to calculate a checksum of a\n\
+string of information.\n\
+\n\
+Methods:\n\
+\n\
+update() -- updates the current digest with an additional string\n\
+digest() -- return the current digest value\n\
+hexdigest() -- return the current digest as a string of hexadecimal digits\n\
+copy() -- return a copy of the current hash object\n\
+\n\
+Attributes:\n\
+\n\
+name -- the hash algorithm being used by this object\n\
+digest_size -- number of bytes in this hashes output\n");
+
+static PyTypeObject EVPtype = {
+    PyObject_HEAD_INIT(NULL)
+    0,			/*ob_size*/
+    "_hashlib.HASH",    /*tp_name*/
+    sizeof(EVPobject),	/*tp_basicsize*/
+    0,			/*tp_itemsize*/
+    /* methods */
+    EVP_dealloc,	/*tp_dealloc*/
+    0,			/*tp_print*/
+    0,                  /*tp_getattr*/
+    0,                  /*tp_setattr*/
+    0,                  /*tp_compare*/
+    EVP_repr,           /*tp_repr*/
+    0,                  /*tp_as_number*/
+    0,                  /*tp_as_sequence*/
+    0,                  /*tp_as_mapping*/
+    0,                  /*tp_hash*/
+    0,                  /*tp_call*/
+    0,                  /*tp_str*/
+    0,                  /*tp_getattro*/
+    0,                  /*tp_setattro*/
+    0,                  /*tp_as_buffer*/
+    Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE, /*tp_flags*/
+    hashtype_doc,       /*tp_doc*/
+    0,                  /*tp_traverse*/
+    0,			/*tp_clear*/
+    0,			/*tp_richcompare*/
+    0,			/*tp_weaklistoffset*/
+    0,			/*tp_iter*/
+    0,			/*tp_iternext*/
+    EVP_methods,	/* tp_methods */
+    EVP_members,	/* tp_members */
+    EVP_getseters,      /* tp_getset */
+#if 1
+    0,                  /* tp_base */
+    0,                  /* tp_dict */
+    0,                  /* tp_descr_get */
+    0,                  /* tp_descr_set */
+    0,                  /* tp_dictoffset */
+#endif
+#if HASH_OBJ_CONSTRUCTOR
+    (initproc)EVP_tp_init, /* tp_init */
+#endif
+};
+
+static PyObject *
+EVPnew(PyObject *name_obj,
+       const EVP_MD *digest, const EVP_MD_CTX *initial_ctx,
+       const unsigned char *cp, Py_ssize_t len)
+{
+    EVPobject *self;
+
+    if (!digest && !initial_ctx) {
+        PyErr_SetString(PyExc_ValueError, "unsupported hash type");
+        return NULL;
+    }
+
+    if ((self = newEVPobject(name_obj)) == NULL)
+        return NULL;
+
+    if (initial_ctx) {
+        EVP_MD_CTX_copy(&self->ctx, initial_ctx);
+    } else {
+        EVP_DigestInit(&self->ctx, digest);
+    }
+
+    if (cp && len) {
+        if (len > 0 && len <= MUNCH_SIZE) {
+            EVP_DigestUpdate(&self->ctx, cp, Py_SAFE_DOWNCAST(len, Py_ssize_t,
+                                                              unsigned int));
+        } else {
+            Py_ssize_t offset = 0;
+            while (len) {
+                unsigned int process = len > MUNCH_SIZE ? MUNCH_SIZE : len;
+                EVP_DigestUpdate(&self->ctx, cp + offset, process);
+                len -= process;
+                offset += process;
+            }
+        }
+    }
+
+    return (PyObject *)self;
+}
+
+
+/* The module-level function: new() */
+
+PyDoc_STRVAR(EVP_new__doc__,
+"Return a new hash object using the named algorithm.\n\
+An optional string argument may be provided and will be\n\
+automatically hashed.\n\
+\n\
+The MD5 and SHA1 algorithms are always supported.\n");
+
+static PyObject *
+EVP_new(PyObject *self, PyObject *args, PyObject *kwdict)
+{
+    static char *kwlist[] = {"name", "string", NULL};
+    PyObject *name_obj = NULL;
+    char *name;
+    const EVP_MD *digest;
+    unsigned char *cp = NULL;
+    Py_ssize_t len = 0;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwdict, "O|s#:new", kwlist,
+                                     &name_obj, &cp, &len)) {
+        return NULL;
+    }
+
+    if (!PyArg_Parse(name_obj, "s", &name)) {
+        PyErr_SetString(PyExc_TypeError, "name must be a string");
+        return NULL;
+    }
+
+    digest = EVP_get_digestbyname(name);
+
+    return EVPnew(name_obj, digest, NULL, cp, len);
+}
+
+/*
+ *  This macro generates constructor function definitions for specific
+ *  hash algorithms.  These constructors are much faster than calling
+ *  the generic one passing it a python string and are noticably
+ *  faster than calling a python new() wrapper.  Thats important for
+ *  code that wants to make hashes of a bunch of small strings.
+ */
+#define GEN_CONSTRUCTOR(NAME)  \
+    static PyObject * \
+    EVP_new_ ## NAME (PyObject *self, PyObject *args) \
+    { \
+        unsigned char *cp = NULL; \
+        Py_ssize_t len = 0; \
+     \
+        if (!PyArg_ParseTuple(args, "|s#:" #NAME , &cp, &len)) { \
+            return NULL; \
+        } \
+     \
+        return EVPnew( \
+                CONST_ ## NAME ## _name_obj, \
+                NULL, \
+                CONST_new_ ## NAME ## _ctx_p, \
+                cp, len); \
+    }
+
+/* a PyMethodDef structure for the constructor */
+#define CONSTRUCTOR_METH_DEF(NAME)  \
+    {"openssl_" #NAME, (PyCFunction)EVP_new_ ## NAME, METH_VARARGS, \
+        PyDoc_STR("Returns a " #NAME \
+                  " hash object; optionally initialized with a string") \
+    }
+
+/* used in the init function to setup a constructor */
+#define INIT_CONSTRUCTOR_CONSTANTS(NAME)  do { \
+    CONST_ ## NAME ## _name_obj = PyString_FromString(#NAME); \
+    if (EVP_get_digestbyname(#NAME)) { \
+        CONST_new_ ## NAME ## _ctx_p = &CONST_new_ ## NAME ## _ctx; \
+        EVP_DigestInit(CONST_new_ ## NAME ## _ctx_p, EVP_get_digestbyname(#NAME)); \
+    } \
+} while (0);
+
+GEN_CONSTRUCTOR(md5)
+GEN_CONSTRUCTOR(sha1)
+GEN_CONSTRUCTOR(sha224)
+GEN_CONSTRUCTOR(sha256)
+GEN_CONSTRUCTOR(sha384)
+GEN_CONSTRUCTOR(sha512)
+
+/* List of functions exported by this module */
+
+static struct PyMethodDef EVP_functions[] = {
+    {"new", (PyCFunction)EVP_new, METH_VARARGS|METH_KEYWORDS, EVP_new__doc__},
+    CONSTRUCTOR_METH_DEF(md5),
+    CONSTRUCTOR_METH_DEF(sha1),
+    CONSTRUCTOR_METH_DEF(sha224),
+    CONSTRUCTOR_METH_DEF(sha256),
+    CONSTRUCTOR_METH_DEF(sha384),
+    CONSTRUCTOR_METH_DEF(sha512),
+    {NULL,	NULL}		 /* Sentinel */
+};
+
+
+/* Initialize this module. */
+
+PyMODINIT_FUNC
+init_hashlib(void)
+{
+    PyObject *m;
+
+    OpenSSL_add_all_digests();
+
+    /* TODO build EVP_functions openssl_* entries dynamically based
+     * on what hashes are supported rather than listing many
+     * but having some be unsupported.  Only init appropriate
+     * constants. */
+
+    EVPtype.ob_type = &PyType_Type;
+    if (PyType_Ready(&EVPtype) < 0)
+        return;
+
+    m = Py_InitModule("_hashlib", EVP_functions);
+    if (m == NULL)
+        return;
+
+#if HASH_OBJ_CONSTRUCTOR
+    Py_INCREF(&EVPtype);
+    PyModule_AddObject(m, "HASH", (PyObject *)&EVPtype);
+#endif
+
+    /* these constants are used by the convenience constructors */
+    INIT_CONSTRUCTOR_CONSTANTS(md5);
+    INIT_CONSTRUCTOR_CONSTANTS(sha1);
+    INIT_CONSTRUCTOR_CONSTANTS(sha224);
+    INIT_CONSTRUCTOR_CONSTANTS(sha256);
+    INIT_CONSTRUCTOR_CONSTANTS(sha384);
+    INIT_CONSTRUCTOR_CONSTANTS(sha512);
+}
diff --git a/src/Modules/md5.c b/src/Modules/md5.c
new file mode 100644
--- /dev/null
+++ b/src/Modules/md5.c
@@ -0,0 +1,381 @@
+/*
+  Copyright (C) 1999, 2000, 2002 Aladdin Enterprises.  All rights reserved.
+
+  This software is provided 'as-is', without any express or implied
+  warranty.  In no event will the authors be held liable for any damages
+  arising from the use of this software.
+
+  Permission is granted to anyone to use this software for any purpose,
+  including commercial applications, and to alter it and redistribute it
+  freely, subject to the following restrictions:
+
+  1. The origin of this software must not be misrepresented; you must not
+     claim that you wrote the original software. If you use this software
+     in a product, an acknowledgment in the product documentation would be
+     appreciated but is not required.
+  2. Altered source versions must be plainly marked as such, and must not be
+     misrepresented as being the original software.
+  3. This notice may not be removed or altered from any source distribution.
+
+  L. Peter Deutsch
+  ghost at aladdin.com
+
+ */
+/* $Id: md5.c,v 1.6 2002/04/13 19:20:28 lpd Exp $ */
+/*
+  Independent implementation of MD5 (RFC 1321).
+
+  This code implements the MD5 Algorithm defined in RFC 1321, whose
+  text is available at
+	http://www.ietf.org/rfc/rfc1321.txt
+  The code is derived from the text of the RFC, including the test suite
+  (section A.5) but excluding the rest of Appendix A.  It does not include
+  any code or documentation that is identified in the RFC as being
+  copyrighted.
+
+  The original and principal author of md5.c is L. Peter Deutsch
+  <ghost at aladdin.com>.  Other authors are noted in the change history
+  that follows (in reverse chronological order):
+
+  2002-04-13 lpd Clarified derivation from RFC 1321; now handles byte order
+	either statically or dynamically; added missing #include <string.h>
+	in library.
+  2002-03-11 lpd Corrected argument list for main(), and added int return
+	type, in test program and T value program.
+  2002-02-21 lpd Added missing #include <stdio.h> in test program.
+  2000-07-03 lpd Patched to eliminate warnings about "constant is
+	unsigned in ANSI C, signed in traditional"; made test program
+	self-checking.
+  1999-11-04 lpd Edited comments slightly for automatic TOC extraction.
+  1999-10-18 lpd Fixed typo in header comment (ansi2knr rather than md5).
+  1999-05-03 lpd Original version.
+ */
+
+#include "md5.h"
+#include <string.h>
+
+#undef BYTE_ORDER	/* 1 = big-endian, -1 = little-endian, 0 = unknown */
+#ifdef ARCH_IS_BIG_ENDIAN
+#  define BYTE_ORDER (ARCH_IS_BIG_ENDIAN ? 1 : -1)
+#else
+#  define BYTE_ORDER 0
+#endif
+
+#define T_MASK ((md5_word_t)~0)
+#define T1 /* 0xd76aa478 */ (T_MASK ^ 0x28955b87)
+#define T2 /* 0xe8c7b756 */ (T_MASK ^ 0x173848a9)
+#define T3    0x242070db
+#define T4 /* 0xc1bdceee */ (T_MASK ^ 0x3e423111)
+#define T5 /* 0xf57c0faf */ (T_MASK ^ 0x0a83f050)
+#define T6    0x4787c62a
+#define T7 /* 0xa8304613 */ (T_MASK ^ 0x57cfb9ec)
+#define T8 /* 0xfd469501 */ (T_MASK ^ 0x02b96afe)
+#define T9    0x698098d8
+#define T10 /* 0x8b44f7af */ (T_MASK ^ 0x74bb0850)
+#define T11 /* 0xffff5bb1 */ (T_MASK ^ 0x0000a44e)
+#define T12 /* 0x895cd7be */ (T_MASK ^ 0x76a32841)
+#define T13    0x6b901122
+#define T14 /* 0xfd987193 */ (T_MASK ^ 0x02678e6c)
+#define T15 /* 0xa679438e */ (T_MASK ^ 0x5986bc71)
+#define T16    0x49b40821
+#define T17 /* 0xf61e2562 */ (T_MASK ^ 0x09e1da9d)
+#define T18 /* 0xc040b340 */ (T_MASK ^ 0x3fbf4cbf)
+#define T19    0x265e5a51
+#define T20 /* 0xe9b6c7aa */ (T_MASK ^ 0x16493855)
+#define T21 /* 0xd62f105d */ (T_MASK ^ 0x29d0efa2)
+#define T22    0x02441453
+#define T23 /* 0xd8a1e681 */ (T_MASK ^ 0x275e197e)
+#define T24 /* 0xe7d3fbc8 */ (T_MASK ^ 0x182c0437)
+#define T25    0x21e1cde6
+#define T26 /* 0xc33707d6 */ (T_MASK ^ 0x3cc8f829)
+#define T27 /* 0xf4d50d87 */ (T_MASK ^ 0x0b2af278)
+#define T28    0x455a14ed
+#define T29 /* 0xa9e3e905 */ (T_MASK ^ 0x561c16fa)
+#define T30 /* 0xfcefa3f8 */ (T_MASK ^ 0x03105c07)
+#define T31    0x676f02d9
+#define T32 /* 0x8d2a4c8a */ (T_MASK ^ 0x72d5b375)
+#define T33 /* 0xfffa3942 */ (T_MASK ^ 0x0005c6bd)
+#define T34 /* 0x8771f681 */ (T_MASK ^ 0x788e097e)
+#define T35    0x6d9d6122
+#define T36 /* 0xfde5380c */ (T_MASK ^ 0x021ac7f3)
+#define T37 /* 0xa4beea44 */ (T_MASK ^ 0x5b4115bb)
+#define T38    0x4bdecfa9
+#define T39 /* 0xf6bb4b60 */ (T_MASK ^ 0x0944b49f)
+#define T40 /* 0xbebfbc70 */ (T_MASK ^ 0x4140438f)
+#define T41    0x289b7ec6
+#define T42 /* 0xeaa127fa */ (T_MASK ^ 0x155ed805)
+#define T43 /* 0xd4ef3085 */ (T_MASK ^ 0x2b10cf7a)
+#define T44    0x04881d05
+#define T45 /* 0xd9d4d039 */ (T_MASK ^ 0x262b2fc6)
+#define T46 /* 0xe6db99e5 */ (T_MASK ^ 0x1924661a)
+#define T47    0x1fa27cf8
+#define T48 /* 0xc4ac5665 */ (T_MASK ^ 0x3b53a99a)
+#define T49 /* 0xf4292244 */ (T_MASK ^ 0x0bd6ddbb)
+#define T50    0x432aff97
+#define T51 /* 0xab9423a7 */ (T_MASK ^ 0x546bdc58)
+#define T52 /* 0xfc93a039 */ (T_MASK ^ 0x036c5fc6)
+#define T53    0x655b59c3
+#define T54 /* 0x8f0ccc92 */ (T_MASK ^ 0x70f3336d)
+#define T55 /* 0xffeff47d */ (T_MASK ^ 0x00100b82)
+#define T56 /* 0x85845dd1 */ (T_MASK ^ 0x7a7ba22e)
+#define T57    0x6fa87e4f
+#define T58 /* 0xfe2ce6e0 */ (T_MASK ^ 0x01d3191f)
+#define T59 /* 0xa3014314 */ (T_MASK ^ 0x5cfebceb)
+#define T60    0x4e0811a1
+#define T61 /* 0xf7537e82 */ (T_MASK ^ 0x08ac817d)
+#define T62 /* 0xbd3af235 */ (T_MASK ^ 0x42c50dca)
+#define T63    0x2ad7d2bb
+#define T64 /* 0xeb86d391 */ (T_MASK ^ 0x14792c6e)
+
+
+static void
+md5_process(md5_state_t *pms, const md5_byte_t *data /*[64]*/)
+{
+    md5_word_t
+	a = pms->abcd[0], b = pms->abcd[1],
+	c = pms->abcd[2], d = pms->abcd[3];
+    md5_word_t t;
+#if BYTE_ORDER > 0
+    /* Define storage only for big-endian CPUs. */
+    md5_word_t X[16];
+#else
+    /* Define storage for little-endian or both types of CPUs. */
+    md5_word_t xbuf[16];
+    const md5_word_t *X;
+#endif
+
+    {
+#if BYTE_ORDER == 0
+	/*
+	 * Determine dynamically whether this is a big-endian or
+	 * little-endian machine, since we can use a more efficient
+	 * algorithm on the latter.
+	 */
+	static const int w = 1;
+
+	if (*((const md5_byte_t *)&w)) /* dynamic little-endian */
+#endif
+#if BYTE_ORDER <= 0		/* little-endian */
+	{
+	    /*
+	     * On little-endian machines, we can process properly aligned
+	     * data without copying it.
+	     */
+	    if (!((data - (const md5_byte_t *)0) & 3)) {
+		/* data are properly aligned */
+		X = (const md5_word_t *)data;
+	    } else {
+		/* not aligned */
+		memcpy(xbuf, data, 64);
+		X = xbuf;
+	    }
+	}
+#endif
+#if BYTE_ORDER == 0
+	else			/* dynamic big-endian */
+#endif
+#if BYTE_ORDER >= 0		/* big-endian */
+	{
+	    /*
+	     * On big-endian machines, we must arrange the bytes in the
+	     * right order.
+	     */
+	    const md5_byte_t *xp = data;
+	    int i;
+
+#  if BYTE_ORDER == 0
+	    X = xbuf;		/* (dynamic only) */
+#  else
+#    define xbuf X		/* (static only) */
+#  endif
+	    for (i = 0; i < 16; ++i, xp += 4)
+		xbuf[i] = xp[0] + (xp[1] << 8) + (xp[2] << 16) + (xp[3] << 24);
+	}
+#endif
+    }
+
+#define ROTATE_LEFT(x, n) (((x) << (n)) | ((x) >> (32 - (n))))
+
+    /* Round 1. */
+    /* Let [abcd k s i] denote the operation
+       a = b + ((a + F(b,c,d) + X[k] + T[i]) <<< s). */
+#define F(x, y, z) (((x) & (y)) | (~(x) & (z)))
+#define SET(a, b, c, d, k, s, Ti)\
+  t = a + F(b,c,d) + X[k] + Ti;\
+  a = ROTATE_LEFT(t, s) + b
+    /* Do the following 16 operations. */
+    SET(a, b, c, d,  0,  7,  T1);
+    SET(d, a, b, c,  1, 12,  T2);
+    SET(c, d, a, b,  2, 17,  T3);
+    SET(b, c, d, a,  3, 22,  T4);
+    SET(a, b, c, d,  4,  7,  T5);
+    SET(d, a, b, c,  5, 12,  T6);
+    SET(c, d, a, b,  6, 17,  T7);
+    SET(b, c, d, a,  7, 22,  T8);
+    SET(a, b, c, d,  8,  7,  T9);
+    SET(d, a, b, c,  9, 12, T10);
+    SET(c, d, a, b, 10, 17, T11);
+    SET(b, c, d, a, 11, 22, T12);
+    SET(a, b, c, d, 12,  7, T13);
+    SET(d, a, b, c, 13, 12, T14);
+    SET(c, d, a, b, 14, 17, T15);
+    SET(b, c, d, a, 15, 22, T16);
+#undef SET
+
+     /* Round 2. */
+     /* Let [abcd k s i] denote the operation
+          a = b + ((a + G(b,c,d) + X[k] + T[i]) <<< s). */
+#define G(x, y, z) (((x) & (z)) | ((y) & ~(z)))
+#define SET(a, b, c, d, k, s, Ti)\
+  t = a + G(b,c,d) + X[k] + Ti;\
+  a = ROTATE_LEFT(t, s) + b
+     /* Do the following 16 operations. */
+    SET(a, b, c, d,  1,  5, T17);
+    SET(d, a, b, c,  6,  9, T18);
+    SET(c, d, a, b, 11, 14, T19);
+    SET(b, c, d, a,  0, 20, T20);
+    SET(a, b, c, d,  5,  5, T21);
+    SET(d, a, b, c, 10,  9, T22);
+    SET(c, d, a, b, 15, 14, T23);
+    SET(b, c, d, a,  4, 20, T24);
+    SET(a, b, c, d,  9,  5, T25);
+    SET(d, a, b, c, 14,  9, T26);
+    SET(c, d, a, b,  3, 14, T27);
+    SET(b, c, d, a,  8, 20, T28);
+    SET(a, b, c, d, 13,  5, T29);
+    SET(d, a, b, c,  2,  9, T30);
+    SET(c, d, a, b,  7, 14, T31);
+    SET(b, c, d, a, 12, 20, T32);
+#undef SET
+
+     /* Round 3. */
+     /* Let [abcd k s t] denote the operation
+          a = b + ((a + H(b,c,d) + X[k] + T[i]) <<< s). */
+#define H(x, y, z) ((x) ^ (y) ^ (z))
+#define SET(a, b, c, d, k, s, Ti)\
+  t = a + H(b,c,d) + X[k] + Ti;\
+  a = ROTATE_LEFT(t, s) + b
+     /* Do the following 16 operations. */
+    SET(a, b, c, d,  5,  4, T33);
+    SET(d, a, b, c,  8, 11, T34);
+    SET(c, d, a, b, 11, 16, T35);
+    SET(b, c, d, a, 14, 23, T36);
+    SET(a, b, c, d,  1,  4, T37);
+    SET(d, a, b, c,  4, 11, T38);
+    SET(c, d, a, b,  7, 16, T39);
+    SET(b, c, d, a, 10, 23, T40);
+    SET(a, b, c, d, 13,  4, T41);
+    SET(d, a, b, c,  0, 11, T42);
+    SET(c, d, a, b,  3, 16, T43);
+    SET(b, c, d, a,  6, 23, T44);
+    SET(a, b, c, d,  9,  4, T45);
+    SET(d, a, b, c, 12, 11, T46);
+    SET(c, d, a, b, 15, 16, T47);
+    SET(b, c, d, a,  2, 23, T48);
+#undef SET
+
+     /* Round 4. */
+     /* Let [abcd k s t] denote the operation
+          a = b + ((a + I(b,c,d) + X[k] + T[i]) <<< s). */
+#define I(x, y, z) ((y) ^ ((x) | ~(z)))
+#define SET(a, b, c, d, k, s, Ti)\
+  t = a + I(b,c,d) + X[k] + Ti;\
+  a = ROTATE_LEFT(t, s) + b
+     /* Do the following 16 operations. */
+    SET(a, b, c, d,  0,  6, T49);
+    SET(d, a, b, c,  7, 10, T50);
+    SET(c, d, a, b, 14, 15, T51);
+    SET(b, c, d, a,  5, 21, T52);
+    SET(a, b, c, d, 12,  6, T53);
+    SET(d, a, b, c,  3, 10, T54);
+    SET(c, d, a, b, 10, 15, T55);
+    SET(b, c, d, a,  1, 21, T56);
+    SET(a, b, c, d,  8,  6, T57);
+    SET(d, a, b, c, 15, 10, T58);
+    SET(c, d, a, b,  6, 15, T59);
+    SET(b, c, d, a, 13, 21, T60);
+    SET(a, b, c, d,  4,  6, T61);
+    SET(d, a, b, c, 11, 10, T62);
+    SET(c, d, a, b,  2, 15, T63);
+    SET(b, c, d, a,  9, 21, T64);
+#undef SET
+
+     /* Then perform the following additions. (That is increment each
+        of the four registers by the value it had before this block
+        was started.) */
+    pms->abcd[0] += a;
+    pms->abcd[1] += b;
+    pms->abcd[2] += c;
+    pms->abcd[3] += d;
+}
+
+void
+md5_init(md5_state_t *pms)
+{
+    pms->count[0] = pms->count[1] = 0;
+    pms->abcd[0] = 0x67452301;
+    pms->abcd[1] = /*0xefcdab89*/ T_MASK ^ 0x10325476;
+    pms->abcd[2] = /*0x98badcfe*/ T_MASK ^ 0x67452301;
+    pms->abcd[3] = 0x10325476;
+}
+
+void
+md5_append(md5_state_t *pms, const md5_byte_t *data, int nbytes)
+{
+    const md5_byte_t *p = data;
+    int left = nbytes;
+    int offset = (pms->count[0] >> 3) & 63;
+    md5_word_t nbits = (md5_word_t)(nbytes << 3);
+
+    if (nbytes <= 0)
+	return;
+
+    /* Update the message length. */
+    pms->count[1] += nbytes >> 29;
+    pms->count[0] += nbits;
+    if (pms->count[0] < nbits)
+	pms->count[1]++;
+
+    /* Process an initial partial block. */
+    if (offset) {
+	int copy = (offset + nbytes > 64 ? 64 - offset : nbytes);
+
+	memcpy(pms->buf + offset, p, copy);
+	if (offset + copy < 64)
+	    return;
+	p += copy;
+	left -= copy;
+	md5_process(pms, pms->buf);
+    }
+
+    /* Process full blocks. */
+    for (; left >= 64; p += 64, left -= 64)
+	md5_process(pms, p);
+
+    /* Process a final partial block. */
+    if (left)
+	memcpy(pms->buf, p, left);
+}
+
+void
+md5_finish(md5_state_t *pms, md5_byte_t digest[16])
+{
+    static const md5_byte_t pad[64] = {
+	0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
+    };
+    md5_byte_t data[8];
+    int i;
+
+    /* Save the length before padding. */
+    for (i = 0; i < 8; ++i)
+	data[i] = (md5_byte_t)(pms->count[i >> 2] >> ((i & 3) << 3));
+    /* Pad to 56 bytes mod 64. */
+    md5_append(pms, pad, ((55 - (pms->count[0] >> 3)) & 63) + 1);
+    /* Append the length. */
+    md5_append(pms, data, 8);
+    for (i = 0; i < 16; ++i)
+	digest[i] = (md5_byte_t)(pms->abcd[i >> 2] >> ((i & 3) << 3));
+}
diff --git a/src/Modules/md5.h b/src/Modules/md5.h
new file mode 100644
--- /dev/null
+++ b/src/Modules/md5.h
@@ -0,0 +1,91 @@
+/*
+  Copyright (C) 1999, 2002 Aladdin Enterprises.  All rights reserved.
+
+  This software is provided 'as-is', without any express or implied
+  warranty.  In no event will the authors be held liable for any damages
+  arising from the use of this software.
+
+  Permission is granted to anyone to use this software for any purpose,
+  including commercial applications, and to alter it and redistribute it
+  freely, subject to the following restrictions:
+
+  1. The origin of this software must not be misrepresented; you must not
+     claim that you wrote the original software. If you use this software
+     in a product, an acknowledgment in the product documentation would be
+     appreciated but is not required.
+  2. Altered source versions must be plainly marked as such, and must not be
+     misrepresented as being the original software.
+  3. This notice may not be removed or altered from any source distribution.
+
+  L. Peter Deutsch
+  ghost at aladdin.com
+
+ */
+/* $Id: md5.h 43594 2006-04-03 16:27:50Z matthias.klose $ */
+/*
+  Independent implementation of MD5 (RFC 1321).
+
+  This code implements the MD5 Algorithm defined in RFC 1321, whose
+  text is available at
+	http://www.ietf.org/rfc/rfc1321.txt
+  The code is derived from the text of the RFC, including the test suite
+  (section A.5) but excluding the rest of Appendix A.  It does not include
+  any code or documentation that is identified in the RFC as being
+  copyrighted.
+
+  The original and principal author of md5.h is L. Peter Deutsch
+  <ghost at aladdin.com>.  Other authors are noted in the change history
+  that follows (in reverse chronological order):
+
+  2002-04-13 lpd Removed support for non-ANSI compilers; removed
+	references to Ghostscript; clarified derivation from RFC 1321;
+	now handles byte order either statically or dynamically.
+  1999-11-04 lpd Edited comments slightly for automatic TOC extraction.
+  1999-10-18 lpd Fixed typo in header comment (ansi2knr rather than md5);
+	added conditionalization for C++ compilation from Martin
+	Purschke <purschke at bnl.gov>.
+  1999-05-03 lpd Original version.
+ */
+
+#ifndef md5_INCLUDED
+#  define md5_INCLUDED
+
+/*
+ * This package supports both compile-time and run-time determination of CPU
+ * byte order.  If ARCH_IS_BIG_ENDIAN is defined as 0, the code will be
+ * compiled to run only on little-endian CPUs; if ARCH_IS_BIG_ENDIAN is
+ * defined as non-zero, the code will be compiled to run only on big-endian
+ * CPUs; if ARCH_IS_BIG_ENDIAN is not defined, the code will be compiled to
+ * run on either big- or little-endian CPUs, but will run slightly less
+ * efficiently on either one than if ARCH_IS_BIG_ENDIAN is defined.
+ */
+
+typedef unsigned char md5_byte_t; /* 8-bit byte */
+typedef unsigned int md5_word_t; /* 32-bit word */
+
+/* Define the state of the MD5 Algorithm. */
+typedef struct md5_state_s {
+    md5_word_t count[2];	/* message length in bits, lsw first */
+    md5_word_t abcd[4];		/* digest buffer */
+    md5_byte_t buf[64];		/* accumulate block */
+} md5_state_t;
+
+#ifdef __cplusplus
+extern "C" 
+{
+#endif
+
+/* Initialize the algorithm. */
+void md5_init(md5_state_t *pms);
+
+/* Append a string to the message. */
+void md5_append(md5_state_t *pms, const md5_byte_t *data, int nbytes);
+
+/* Finish the message and return the digest. */
+void md5_finish(md5_state_t *pms, md5_byte_t digest[16]);
+
+#ifdef __cplusplus
+}  /* end extern "C" */
+#endif
+
+#endif /* md5_INCLUDED */
diff --git a/src/Modules/md5module.c b/src/Modules/md5module.c
new file mode 100644
--- /dev/null
+++ b/src/Modules/md5module.c
@@ -0,0 +1,312 @@
+
+/* MD5 module */
+
+/* This module provides an interface to the RSA Data Security,
+   Inc. MD5 Message-Digest Algorithm, described in RFC 1321.
+   It requires the files md5c.c and md5.h (which are slightly changed
+   from the versions in the RFC to avoid the "global.h" file.) */
+
+
+/* MD5 objects */
+
+#include "Python.h"
+#include "structmember.h"
+#include "md5.h"
+
+typedef struct {
+	PyObject_HEAD
+        md5_state_t	md5;		/* the context holder */
+} md5object;
+
+static PyTypeObject MD5type;
+
+#define is_md5object(v)		((v)->ob_type == &MD5type)
+
+static md5object *
+newmd5object(void)
+{
+	md5object *md5p;
+
+	md5p = PyObject_New(md5object, &MD5type);
+	if (md5p == NULL)
+		return NULL;
+
+	md5_init(&md5p->md5);	/* actual initialisation */
+	return md5p;
+}
+
+
+/* MD5 methods */
+
+static void
+md5_dealloc(md5object *md5p)
+{
+	PyObject_Del(md5p);
+}
+
+
+/* MD5 methods-as-attributes */
+
+static PyObject *
+md5_update(md5object *self, PyObject *args)
+{
+	unsigned char *cp;
+	int len;
+
+	if (!PyArg_ParseTuple(args, "s#:update", &cp, &len))
+		return NULL;
+
+	md5_append(&self->md5, cp, len);
+
+	Py_INCREF(Py_None);
+	return Py_None;
+}
+
+PyDoc_STRVAR(update_doc,
+"update (arg)\n\
+\n\
+Update the md5 object with the string arg. Repeated calls are\n\
+equivalent to a single call with the concatenation of all the\n\
+arguments.");
+
+
+static PyObject *
+md5_digest(md5object *self)
+{
+ 	md5_state_t mdContext;
+	unsigned char aDigest[16];
+
+	/* make a temporary copy, and perform the final */
+	mdContext = self->md5;
+	md5_finish(&mdContext, aDigest);
+
+	return PyString_FromStringAndSize((char *)aDigest, 16);
+}
+
+PyDoc_STRVAR(digest_doc,
+"digest() -> string\n\
+\n\
+Return the digest of the strings passed to the update() method so\n\
+far. This is a 16-byte string which may contain non-ASCII characters,\n\
+including null bytes.");
+
+
+static PyObject *
+md5_hexdigest(md5object *self)
+{
+ 	md5_state_t mdContext;
+	unsigned char digest[16];
+	unsigned char hexdigest[32];
+	int i, j;
+
+	/* make a temporary copy, and perform the final */
+	mdContext = self->md5;
+	md5_finish(&mdContext, digest);
+
+	/* Make hex version of the digest */
+	for(i=j=0; i<16; i++) {
+		char c;
+		c = (digest[i] >> 4) & 0xf;
+		c = (c>9) ? c+'a'-10 : c + '0';
+		hexdigest[j++] = c;
+		c = (digest[i] & 0xf);
+		c = (c>9) ? c+'a'-10 : c + '0';
+		hexdigest[j++] = c;
+	}
+	return PyString_FromStringAndSize((char*)hexdigest, 32);
+}
+
+
+PyDoc_STRVAR(hexdigest_doc,
+"hexdigest() -> string\n\
+\n\
+Like digest(), but returns the digest as a string of hexadecimal digits.");
+
+
+static PyObject *
+md5_copy(md5object *self)
+{
+	md5object *md5p;
+
+	if ((md5p = newmd5object()) == NULL)
+		return NULL;
+
+	md5p->md5 = self->md5;
+
+	return (PyObject *)md5p;
+}
+
+PyDoc_STRVAR(copy_doc,
+"copy() -> md5 object\n\
+\n\
+Return a copy (``clone'') of the md5 object.");
+
+
+static PyMethodDef md5_methods[] = {
+	{"update",    (PyCFunction)md5_update,    METH_VARARGS, update_doc},
+	{"digest",    (PyCFunction)md5_digest,    METH_NOARGS,  digest_doc},
+	{"hexdigest", (PyCFunction)md5_hexdigest, METH_NOARGS,  hexdigest_doc},
+	{"copy",      (PyCFunction)md5_copy,      METH_NOARGS,  copy_doc},
+	{NULL, NULL}			     /* sentinel */
+};
+
+static PyObject *
+md5_get_block_size(PyObject *self, void *closure)
+{
+    return PyInt_FromLong(64);
+}
+
+static PyObject *
+md5_get_digest_size(PyObject *self, void *closure)
+{
+    return PyInt_FromLong(16);
+}
+
+static PyObject *
+md5_get_name(PyObject *self, void *closure)
+{
+    return PyString_FromStringAndSize("MD5", 3);
+}
+
+static PyGetSetDef md5_getseters[] = {
+    {"digest_size",
+     (getter)md5_get_digest_size, NULL,
+     NULL,
+     NULL},
+    {"block_size",
+     (getter)md5_get_block_size, NULL,
+     NULL,
+     NULL},
+    {"name",
+     (getter)md5_get_name, NULL,
+     NULL,
+     NULL},
+    /* the old md5 and sha modules support 'digest_size' as in PEP 247.
+     * the old sha module also supported 'digestsize'.  ugh. */
+    {"digestsize",
+     (getter)md5_get_digest_size, NULL,
+     NULL,
+     NULL},
+    {NULL}  /* Sentinel */
+};
+
+
+PyDoc_STRVAR(module_doc,
+"This module implements the interface to RSA's MD5 message digest\n\
+algorithm (see also Internet RFC 1321). Its use is quite\n\
+straightforward: use the new() to create an md5 object. You can now\n\
+feed this object with arbitrary strings using the update() method, and\n\
+at any point you can ask it for the digest (a strong kind of 128-bit\n\
+checksum, a.k.a. ``fingerprint'') of the concatenation of the strings\n\
+fed to it so far using the digest() method.\n\
+\n\
+Functions:\n\
+\n\
+new([arg]) -- return a new md5 object, initialized with arg if provided\n\
+md5([arg]) -- DEPRECATED, same as new, but for compatibility\n\
+\n\
+Special Objects:\n\
+\n\
+MD5Type -- type object for md5 objects");
+
+PyDoc_STRVAR(md5type_doc,
+"An md5 represents the object used to calculate the MD5 checksum of a\n\
+string of information.\n\
+\n\
+Methods:\n\
+\n\
+update() -- updates the current digest with an additional string\n\
+digest() -- return the current digest value\n\
+hexdigest() -- return the current digest as a string of hexadecimal digits\n\
+copy() -- return a copy of the current md5 object");
+
+static PyTypeObject MD5type = {
+	PyObject_HEAD_INIT(NULL)
+	0,			  /*ob_size*/
+	"_md5.md5",		  /*tp_name*/
+	sizeof(md5object),	  /*tp_size*/
+	0,			  /*tp_itemsize*/
+	/* methods */
+	(destructor)md5_dealloc,  /*tp_dealloc*/
+	0,			  /*tp_print*/
+	0,                        /*tp_getattr*/
+	0,			  /*tp_setattr*/
+	0,			  /*tp_compare*/
+	0,			  /*tp_repr*/
+        0,			  /*tp_as_number*/
+	0,                        /*tp_as_sequence*/
+	0,			  /*tp_as_mapping*/
+	0, 			  /*tp_hash*/
+	0,			  /*tp_call*/
+	0,			  /*tp_str*/
+	0,			  /*tp_getattro*/
+	0,			  /*tp_setattro*/
+	0,	                  /*tp_as_buffer*/
+	Py_TPFLAGS_DEFAULT,	  /*tp_flags*/
+	md5type_doc,		  /*tp_doc*/
+        0,                        /*tp_traverse*/
+        0,			  /*tp_clear*/
+        0,			  /*tp_richcompare*/
+        0,			  /*tp_weaklistoffset*/
+        0,			  /*tp_iter*/
+        0,			  /*tp_iternext*/
+        md5_methods,	          /*tp_methods*/
+        0,      	          /*tp_members*/
+        md5_getseters,            /*tp_getset*/
+};
+
+
+/* MD5 functions */
+
+static PyObject *
+MD5_new(PyObject *self, PyObject *args)
+{
+	md5object *md5p;
+	unsigned char *cp = NULL;
+	int len = 0;
+
+	if (!PyArg_ParseTuple(args, "|s#:new", &cp, &len))
+		return NULL;
+
+	if ((md5p = newmd5object()) == NULL)
+		return NULL;
+
+	if (cp)
+		md5_append(&md5p->md5, cp, len);
+
+	return (PyObject *)md5p;
+}
+
+PyDoc_STRVAR(new_doc,
+"new([arg]) -> md5 object\n\
+\n\
+Return a new md5 object. If arg is present, the method call update(arg)\n\
+is made.");
+
+
+/* List of functions exported by this module */
+
+static PyMethodDef md5_functions[] = {
+	{"new",		(PyCFunction)MD5_new, METH_VARARGS, new_doc},
+	{NULL,		NULL}	/* Sentinel */
+};
+
+
+/* Initialize this module. */
+
+PyMODINIT_FUNC
+init_md5(void)
+{
+	PyObject *m, *d;
+
+        MD5type.ob_type = &PyType_Type;
+        if (PyType_Ready(&MD5type) < 0)
+            return;
+	m = Py_InitModule3("_md5", md5_functions, module_doc);
+	if (m == NULL)
+	    return;
+	d = PyModule_GetDict(m);
+	PyDict_SetItemString(d, "MD5Type", (PyObject *)&MD5type);
+	PyModule_AddIntConstant(m, "digest_size", 16);
+	/* No need to check the error here, the caller will do that */
+}
diff --git a/src/Modules/sha256module.c b/src/Modules/sha256module.c
new file mode 100644
--- /dev/null
+++ b/src/Modules/sha256module.c
@@ -0,0 +1,701 @@
+/* SHA256 module */
+
+/* This module provides an interface to NIST's SHA-256 and SHA-224 Algorithms */
+
+/* See below for information about the original code this module was
+   based upon. Additional work performed by:
+
+   Andrew Kuchling (amk at amk.ca)
+   Greg Stein (gstein at lyra.org)
+   Trevor Perrin (trevp at trevp.net)
+
+   Copyright (C) 2005   Gregory P. Smith (greg at krypto.org)
+   Licensed to PSF under a Contributor Agreement.
+
+*/
+
+/* SHA objects */
+
+#include "Python.h"
+#include "structmember.h"
+
+
+/* Endianness testing and definitions */
+#define TestEndianness(variable) {int i=1; variable=PCT_BIG_ENDIAN;\
+	if (*((char*)&i)==1) variable=PCT_LITTLE_ENDIAN;}
+
+#define PCT_LITTLE_ENDIAN 1
+#define PCT_BIG_ENDIAN 0
+
+/* Some useful types */
+
+typedef unsigned char SHA_BYTE;
+
+#if SIZEOF_INT == 4
+typedef unsigned int SHA_INT32;	/* 32-bit integer */
+#else
+/* not defined. compilation will die. */
+#endif
+
+/* The SHA block size and message digest sizes, in bytes */
+
+#define SHA_BLOCKSIZE    64
+#define SHA_DIGESTSIZE  32
+
+/* The structure for storing SHA info */
+
+typedef struct {
+    PyObject_HEAD
+    SHA_INT32 digest[8];		/* Message digest */
+    SHA_INT32 count_lo, count_hi;	/* 64-bit bit count */
+    SHA_BYTE data[SHA_BLOCKSIZE];	/* SHA data buffer */
+    int Endianness;
+    int local;				/* unprocessed amount in data */
+    int digestsize;
+} SHAobject;
+
+/* When run on a little-endian CPU we need to perform byte reversal on an
+   array of longwords. */
+
+static void longReverse(SHA_INT32 *buffer, int byteCount, int Endianness)
+{
+    SHA_INT32 value;
+
+    if ( Endianness == PCT_BIG_ENDIAN )
+	return;
+
+    byteCount /= sizeof(*buffer);
+    while (byteCount--) {
+        value = *buffer;
+        value = ( ( value & 0xFF00FF00L ) >> 8  ) | \
+                ( ( value & 0x00FF00FFL ) << 8 );
+        *buffer++ = ( value << 16 ) | ( value >> 16 );
+    }
+}
+
+static void SHAcopy(SHAobject *src, SHAobject *dest)
+{
+    dest->Endianness = src->Endianness;
+    dest->local = src->local;
+    dest->digestsize = src->digestsize;
+    dest->count_lo = src->count_lo;
+    dest->count_hi = src->count_hi;
+    memcpy(dest->digest, src->digest, sizeof(src->digest));
+    memcpy(dest->data, src->data, sizeof(src->data));
+}
+
+
+/* ------------------------------------------------------------------------
+ *
+ * This code for the SHA-256 algorithm was noted as public domain. The
+ * original headers are pasted below.
+ *
+ * Several changes have been made to make it more compatible with the
+ * Python environment and desired interface.
+ *
+ */
+
+/* LibTomCrypt, modular cryptographic library -- Tom St Denis
+ *
+ * LibTomCrypt is a library that provides various cryptographic
+ * algorithms in a highly modular and flexible manner.
+ *
+ * The library is free for all purposes without any express
+ * gurantee it works.
+ *
+ * Tom St Denis, tomstdenis at iahu.ca, http://libtomcrypt.org
+ */
+
+
+/* SHA256 by Tom St Denis */
+
+/* Various logical functions */
+#define ROR(x, y)\
+( ((((unsigned long)(x)&0xFFFFFFFFUL)>>(unsigned long)((y)&31)) | \
+((unsigned long)(x)<<(unsigned long)(32-((y)&31)))) & 0xFFFFFFFFUL)
+#define Ch(x,y,z)       (z ^ (x & (y ^ z)))
+#define Maj(x,y,z)      (((x | y) & z) | (x & y)) 
+#define S(x, n)         ROR((x),(n))
+#define R(x, n)         (((x)&0xFFFFFFFFUL)>>(n))
+#define Sigma0(x)       (S(x, 2) ^ S(x, 13) ^ S(x, 22))
+#define Sigma1(x)       (S(x, 6) ^ S(x, 11) ^ S(x, 25))
+#define Gamma0(x)       (S(x, 7) ^ S(x, 18) ^ R(x, 3))
+#define Gamma1(x)       (S(x, 17) ^ S(x, 19) ^ R(x, 10))
+
+
+static void
+sha_transform(SHAobject *sha_info)
+{
+    int i;
+	SHA_INT32 S[8], W[64], t0, t1;
+
+    memcpy(W, sha_info->data, sizeof(sha_info->data));
+    longReverse(W, (int)sizeof(sha_info->data), sha_info->Endianness);
+
+    for (i = 16; i < 64; ++i) {
+		W[i] = Gamma1(W[i - 2]) + W[i - 7] + Gamma0(W[i - 15]) + W[i - 16];
+    }
+    for (i = 0; i < 8; ++i) {
+        S[i] = sha_info->digest[i];
+    }
+
+    /* Compress */
+#define RND(a,b,c,d,e,f,g,h,i,ki)                    \
+     t0 = h + Sigma1(e) + Ch(e, f, g) + ki + W[i];   \
+     t1 = Sigma0(a) + Maj(a, b, c);                  \
+     d += t0;                                        \
+     h  = t0 + t1;
+
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],0,0x428a2f98);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],1,0x71374491);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],2,0xb5c0fbcf);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],3,0xe9b5dba5);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],4,0x3956c25b);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],5,0x59f111f1);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],6,0x923f82a4);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],7,0xab1c5ed5);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],8,0xd807aa98);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],9,0x12835b01);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],10,0x243185be);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],11,0x550c7dc3);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],12,0x72be5d74);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],13,0x80deb1fe);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],14,0x9bdc06a7);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],15,0xc19bf174);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],16,0xe49b69c1);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],17,0xefbe4786);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],18,0x0fc19dc6);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],19,0x240ca1cc);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],20,0x2de92c6f);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],21,0x4a7484aa);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],22,0x5cb0a9dc);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],23,0x76f988da);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],24,0x983e5152);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],25,0xa831c66d);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],26,0xb00327c8);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],27,0xbf597fc7);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],28,0xc6e00bf3);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],29,0xd5a79147);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],30,0x06ca6351);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],31,0x14292967);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],32,0x27b70a85);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],33,0x2e1b2138);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],34,0x4d2c6dfc);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],35,0x53380d13);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],36,0x650a7354);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],37,0x766a0abb);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],38,0x81c2c92e);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],39,0x92722c85);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],40,0xa2bfe8a1);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],41,0xa81a664b);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],42,0xc24b8b70);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],43,0xc76c51a3);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],44,0xd192e819);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],45,0xd6990624);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],46,0xf40e3585);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],47,0x106aa070);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],48,0x19a4c116);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],49,0x1e376c08);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],50,0x2748774c);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],51,0x34b0bcb5);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],52,0x391c0cb3);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],53,0x4ed8aa4a);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],54,0x5b9cca4f);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],55,0x682e6ff3);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],56,0x748f82ee);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],57,0x78a5636f);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],58,0x84c87814);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],59,0x8cc70208);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],60,0x90befffa);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],61,0xa4506ceb);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],62,0xbef9a3f7);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],63,0xc67178f2);
+
+#undef RND     
+    
+    /* feedback */
+    for (i = 0; i < 8; i++) {
+        sha_info->digest[i] = sha_info->digest[i] + S[i];
+    }
+
+}
+
+
+
+/* initialize the SHA digest */
+
+static void
+sha_init(SHAobject *sha_info)
+{
+    TestEndianness(sha_info->Endianness)
+    sha_info->digest[0] = 0x6A09E667L;
+    sha_info->digest[1] = 0xBB67AE85L;
+    sha_info->digest[2] = 0x3C6EF372L;
+    sha_info->digest[3] = 0xA54FF53AL;
+    sha_info->digest[4] = 0x510E527FL;
+    sha_info->digest[5] = 0x9B05688CL;
+    sha_info->digest[6] = 0x1F83D9ABL;
+    sha_info->digest[7] = 0x5BE0CD19L;
+    sha_info->count_lo = 0L;
+    sha_info->count_hi = 0L;
+    sha_info->local = 0;
+    sha_info->digestsize = 32;
+}
+
+static void
+sha224_init(SHAobject *sha_info)
+{
+    TestEndianness(sha_info->Endianness)
+    sha_info->digest[0] = 0xc1059ed8L;
+    sha_info->digest[1] = 0x367cd507L;
+    sha_info->digest[2] = 0x3070dd17L;
+    sha_info->digest[3] = 0xf70e5939L;
+    sha_info->digest[4] = 0xffc00b31L;
+    sha_info->digest[5] = 0x68581511L;
+    sha_info->digest[6] = 0x64f98fa7L;
+    sha_info->digest[7] = 0xbefa4fa4L;
+    sha_info->count_lo = 0L;
+    sha_info->count_hi = 0L;
+    sha_info->local = 0;
+    sha_info->digestsize = 28;
+}
+
+
+/* update the SHA digest */
+
+static void
+sha_update(SHAobject *sha_info, SHA_BYTE *buffer, int count)
+{
+    int i;
+    SHA_INT32 clo;
+
+    clo = sha_info->count_lo + ((SHA_INT32) count << 3);
+    if (clo < sha_info->count_lo) {
+        ++sha_info->count_hi;
+    }
+    sha_info->count_lo = clo;
+    sha_info->count_hi += (SHA_INT32) count >> 29;
+    if (sha_info->local) {
+        i = SHA_BLOCKSIZE - sha_info->local;
+        if (i > count) {
+            i = count;
+        }
+        memcpy(((SHA_BYTE *) sha_info->data) + sha_info->local, buffer, i);
+        count -= i;
+        buffer += i;
+        sha_info->local += i;
+        if (sha_info->local == SHA_BLOCKSIZE) {
+            sha_transform(sha_info);
+        }
+        else {
+            return;
+        }
+    }
+    while (count >= SHA_BLOCKSIZE) {
+        memcpy(sha_info->data, buffer, SHA_BLOCKSIZE);
+        buffer += SHA_BLOCKSIZE;
+        count -= SHA_BLOCKSIZE;
+        sha_transform(sha_info);
+    }
+    memcpy(sha_info->data, buffer, count);
+    sha_info->local = count;
+}
+
+/* finish computing the SHA digest */
+
+static void
+sha_final(unsigned char digest[SHA_DIGESTSIZE], SHAobject *sha_info)
+{
+    int count;
+    SHA_INT32 lo_bit_count, hi_bit_count;
+
+    lo_bit_count = sha_info->count_lo;
+    hi_bit_count = sha_info->count_hi;
+    count = (int) ((lo_bit_count >> 3) & 0x3f);
+    ((SHA_BYTE *) sha_info->data)[count++] = 0x80;
+    if (count > SHA_BLOCKSIZE - 8) {
+	memset(((SHA_BYTE *) sha_info->data) + count, 0,
+	       SHA_BLOCKSIZE - count);
+	sha_transform(sha_info);
+	memset((SHA_BYTE *) sha_info->data, 0, SHA_BLOCKSIZE - 8);
+    }
+    else {
+	memset(((SHA_BYTE *) sha_info->data) + count, 0,
+	       SHA_BLOCKSIZE - 8 - count);
+    }
+
+    /* GJS: note that we add the hi/lo in big-endian. sha_transform will
+       swap these values into host-order. */
+    sha_info->data[56] = (hi_bit_count >> 24) & 0xff;
+    sha_info->data[57] = (hi_bit_count >> 16) & 0xff;
+    sha_info->data[58] = (hi_bit_count >>  8) & 0xff;
+    sha_info->data[59] = (hi_bit_count >>  0) & 0xff;
+    sha_info->data[60] = (lo_bit_count >> 24) & 0xff;
+    sha_info->data[61] = (lo_bit_count >> 16) & 0xff;
+    sha_info->data[62] = (lo_bit_count >>  8) & 0xff;
+    sha_info->data[63] = (lo_bit_count >>  0) & 0xff;
+    sha_transform(sha_info);
+    digest[ 0] = (unsigned char) ((sha_info->digest[0] >> 24) & 0xff);
+    digest[ 1] = (unsigned char) ((sha_info->digest[0] >> 16) & 0xff);
+    digest[ 2] = (unsigned char) ((sha_info->digest[0] >>  8) & 0xff);
+    digest[ 3] = (unsigned char) ((sha_info->digest[0]      ) & 0xff);
+    digest[ 4] = (unsigned char) ((sha_info->digest[1] >> 24) & 0xff);
+    digest[ 5] = (unsigned char) ((sha_info->digest[1] >> 16) & 0xff);
+    digest[ 6] = (unsigned char) ((sha_info->digest[1] >>  8) & 0xff);
+    digest[ 7] = (unsigned char) ((sha_info->digest[1]      ) & 0xff);
+    digest[ 8] = (unsigned char) ((sha_info->digest[2] >> 24) & 0xff);
+    digest[ 9] = (unsigned char) ((sha_info->digest[2] >> 16) & 0xff);
+    digest[10] = (unsigned char) ((sha_info->digest[2] >>  8) & 0xff);
+    digest[11] = (unsigned char) ((sha_info->digest[2]      ) & 0xff);
+    digest[12] = (unsigned char) ((sha_info->digest[3] >> 24) & 0xff);
+    digest[13] = (unsigned char) ((sha_info->digest[3] >> 16) & 0xff);
+    digest[14] = (unsigned char) ((sha_info->digest[3] >>  8) & 0xff);
+    digest[15] = (unsigned char) ((sha_info->digest[3]      ) & 0xff);
+    digest[16] = (unsigned char) ((sha_info->digest[4] >> 24) & 0xff);
+    digest[17] = (unsigned char) ((sha_info->digest[4] >> 16) & 0xff);
+    digest[18] = (unsigned char) ((sha_info->digest[4] >>  8) & 0xff);
+    digest[19] = (unsigned char) ((sha_info->digest[4]      ) & 0xff);
+    digest[20] = (unsigned char) ((sha_info->digest[5] >> 24) & 0xff);
+    digest[21] = (unsigned char) ((sha_info->digest[5] >> 16) & 0xff);
+    digest[22] = (unsigned char) ((sha_info->digest[5] >>  8) & 0xff);
+    digest[23] = (unsigned char) ((sha_info->digest[5]      ) & 0xff);
+    digest[24] = (unsigned char) ((sha_info->digest[6] >> 24) & 0xff);
+    digest[25] = (unsigned char) ((sha_info->digest[6] >> 16) & 0xff);
+    digest[26] = (unsigned char) ((sha_info->digest[6] >>  8) & 0xff);
+    digest[27] = (unsigned char) ((sha_info->digest[6]      ) & 0xff);
+    digest[28] = (unsigned char) ((sha_info->digest[7] >> 24) & 0xff);
+    digest[29] = (unsigned char) ((sha_info->digest[7] >> 16) & 0xff);
+    digest[30] = (unsigned char) ((sha_info->digest[7] >>  8) & 0xff);
+    digest[31] = (unsigned char) ((sha_info->digest[7]      ) & 0xff);
+}
+
+/*
+ * End of copied SHA code.
+ *
+ * ------------------------------------------------------------------------
+ */
+
+static PyTypeObject SHA224type;
+static PyTypeObject SHA256type;
+
+
+static SHAobject *
+newSHA224object(void)
+{
+    return (SHAobject *)PyObject_New(SHAobject, &SHA224type);
+}
+
+static SHAobject *
+newSHA256object(void)
+{
+    return (SHAobject *)PyObject_New(SHAobject, &SHA256type);
+}
+
+/* Internal methods for a hash object */
+
+static void
+SHA_dealloc(PyObject *ptr)
+{
+    PyObject_Del(ptr);
+}
+
+
+/* External methods for a hash object */
+
+PyDoc_STRVAR(SHA256_copy__doc__, "Return a copy of the hash object.");
+
+static PyObject *
+SHA256_copy(SHAobject *self, PyObject *unused)
+{
+    SHAobject *newobj;
+
+    if (((PyObject*)self)->ob_type == &SHA256type) {
+        if ( (newobj = newSHA256object())==NULL)
+            return NULL;
+    } else {
+        if ( (newobj = newSHA224object())==NULL)
+            return NULL;
+    }
+
+    SHAcopy(self, newobj);
+    return (PyObject *)newobj;
+}
+
+PyDoc_STRVAR(SHA256_digest__doc__,
+"Return the digest value as a string of binary data.");
+
+static PyObject *
+SHA256_digest(SHAobject *self, PyObject *unused)
+{
+    unsigned char digest[SHA_DIGESTSIZE];
+    SHAobject temp;
+
+    SHAcopy(self, &temp);
+    sha_final(digest, &temp);
+    return PyString_FromStringAndSize((const char *)digest, self->digestsize);
+}
+
+PyDoc_STRVAR(SHA256_hexdigest__doc__,
+"Return the digest value as a string of hexadecimal digits.");
+
+static PyObject *
+SHA256_hexdigest(SHAobject *self, PyObject *unused)
+{
+    unsigned char digest[SHA_DIGESTSIZE];
+    SHAobject temp;
+    PyObject *retval;
+    char *hex_digest;
+    int i, j;
+
+    /* Get the raw (binary) digest value */
+    SHAcopy(self, &temp);
+    sha_final(digest, &temp);
+
+    /* Create a new string */
+    retval = PyString_FromStringAndSize(NULL, self->digestsize * 2);
+    if (!retval)
+	    return NULL;
+    hex_digest = PyString_AsString(retval);
+    if (!hex_digest) {
+	    Py_DECREF(retval);
+	    return NULL;
+    }
+
+    /* Make hex version of the digest */
+    for(i=j=0; i<self->digestsize; i++) {
+        char c;
+        c = (digest[i] >> 4) & 0xf;
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+        c = (digest[i] & 0xf);
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+    }
+    return retval;
+}
+
+PyDoc_STRVAR(SHA256_update__doc__,
+"Update this hash object's state with the provided string.");
+
+static PyObject *
+SHA256_update(SHAobject *self, PyObject *args)
+{
+    unsigned char *cp;
+    int len;
+
+    if (!PyArg_ParseTuple(args, "s#:update", &cp, &len))
+        return NULL;
+
+    sha_update(self, cp, len);
+
+    Py_INCREF(Py_None);
+    return Py_None;
+}
+
+static PyMethodDef SHA_methods[] = {
+    {"copy",	  (PyCFunction)SHA256_copy,      METH_NOARGS,  SHA256_copy__doc__},
+    {"digest",	  (PyCFunction)SHA256_digest,    METH_NOARGS,  SHA256_digest__doc__},
+    {"hexdigest", (PyCFunction)SHA256_hexdigest, METH_NOARGS,  SHA256_hexdigest__doc__},
+    {"update",	  (PyCFunction)SHA256_update,    METH_VARARGS, SHA256_update__doc__},
+    {NULL,	  NULL}		/* sentinel */
+};
+
+static PyObject *
+SHA256_get_block_size(PyObject *self, void *closure)
+{
+    return PyInt_FromLong(SHA_BLOCKSIZE);
+}
+
+static PyObject *
+SHA256_get_name(PyObject *self, void *closure)
+{
+    if (((SHAobject *)self)->digestsize == 32)
+        return PyString_FromStringAndSize("SHA256", 6);
+    else
+        return PyString_FromStringAndSize("SHA224", 6);
+}
+
+static PyGetSetDef SHA_getseters[] = {
+    {"block_size",
+     (getter)SHA256_get_block_size, NULL,
+     NULL,
+     NULL},
+    {"name",
+     (getter)SHA256_get_name, NULL,
+     NULL,
+     NULL},
+    {NULL}  /* Sentinel */
+};
+
+static PyMemberDef SHA_members[] = {
+    {"digest_size", T_INT, offsetof(SHAobject, digestsize), READONLY, NULL},
+    /* the old md5 and sha modules support 'digest_size' as in PEP 247.
+     * the old sha module also supported 'digestsize'.  ugh. */
+    {"digestsize", T_INT, offsetof(SHAobject, digestsize), READONLY, NULL},
+    {NULL}  /* Sentinel */
+};
+
+static PyTypeObject SHA224type = {
+    PyObject_HEAD_INIT(NULL)
+    0,			/*ob_size*/
+    "_sha256.sha224",	/*tp_name*/
+    sizeof(SHAobject),	/*tp_size*/
+    0,			/*tp_itemsize*/
+    /* methods */
+    SHA_dealloc,	/*tp_dealloc*/
+    0,			/*tp_print*/
+    0,          	/*tp_getattr*/
+    0,                  /*tp_setattr*/
+    0,                  /*tp_compare*/
+    0,                  /*tp_repr*/
+    0,                  /*tp_as_number*/
+    0,                  /*tp_as_sequence*/
+    0,                  /*tp_as_mapping*/
+    0,                  /*tp_hash*/
+    0,                  /*tp_call*/
+    0,                  /*tp_str*/
+    0,                  /*tp_getattro*/
+    0,                  /*tp_setattro*/
+    0,                  /*tp_as_buffer*/
+    Py_TPFLAGS_DEFAULT, /*tp_flags*/
+    0,                  /*tp_doc*/
+    0,                  /*tp_traverse*/
+    0,			/*tp_clear*/
+    0,			/*tp_richcompare*/
+    0,			/*tp_weaklistoffset*/
+    0,			/*tp_iter*/
+    0,			/*tp_iternext*/
+    SHA_methods,	/* tp_methods */
+    SHA_members,	/* tp_members */
+    SHA_getseters,      /* tp_getset */
+};
+
+static PyTypeObject SHA256type = {
+    PyObject_HEAD_INIT(NULL)
+    0,			/*ob_size*/
+    "_sha256.sha256",	/*tp_name*/
+    sizeof(SHAobject),	/*tp_size*/
+    0,			/*tp_itemsize*/
+    /* methods */
+    SHA_dealloc,	/*tp_dealloc*/
+    0,			/*tp_print*/
+    0,          	/*tp_getattr*/
+    0,                  /*tp_setattr*/
+    0,                  /*tp_compare*/
+    0,                  /*tp_repr*/
+    0,                  /*tp_as_number*/
+    0,                  /*tp_as_sequence*/
+    0,                  /*tp_as_mapping*/
+    0,                  /*tp_hash*/
+    0,                  /*tp_call*/
+    0,                  /*tp_str*/
+    0,                  /*tp_getattro*/
+    0,                  /*tp_setattro*/
+    0,                  /*tp_as_buffer*/
+    Py_TPFLAGS_DEFAULT, /*tp_flags*/
+    0,                  /*tp_doc*/
+    0,                  /*tp_traverse*/
+    0,			/*tp_clear*/
+    0,			/*tp_richcompare*/
+    0,			/*tp_weaklistoffset*/
+    0,			/*tp_iter*/
+    0,			/*tp_iternext*/
+    SHA_methods,	/* tp_methods */
+    SHA_members,	/* tp_members */
+    SHA_getseters,      /* tp_getset */
+};
+
+
+/* The single module-level function: new() */
+
+PyDoc_STRVAR(SHA256_new__doc__,
+"Return a new SHA-256 hash object; optionally initialized with a string.");
+
+static PyObject *
+SHA256_new(PyObject *self, PyObject *args, PyObject *kwdict)
+{
+    static char *kwlist[] = {"string", NULL};
+    SHAobject *new;
+    unsigned char *cp = NULL;
+    int len;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwdict, "|s#:new", kwlist,
+                                     &cp, &len)) {
+        return NULL;
+    }
+
+    if ((new = newSHA256object()) == NULL)
+        return NULL;
+
+    sha_init(new);
+
+    if (PyErr_Occurred()) {
+        Py_DECREF(new);
+        return NULL;
+    }
+    if (cp)
+        sha_update(new, cp, len);
+
+    return (PyObject *)new;
+}
+
+PyDoc_STRVAR(SHA224_new__doc__,
+"Return a new SHA-224 hash object; optionally initialized with a string.");
+
+static PyObject *
+SHA224_new(PyObject *self, PyObject *args, PyObject *kwdict)
+{
+    static char *kwlist[] = {"string", NULL};
+    SHAobject *new;
+    unsigned char *cp = NULL;
+    int len;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwdict, "|s#:new", kwlist,
+                                     &cp, &len)) {
+        return NULL;
+    }
+
+    if ((new = newSHA224object()) == NULL)
+        return NULL;
+
+    sha224_init(new);
+
+    if (PyErr_Occurred()) {
+        Py_DECREF(new);
+        return NULL;
+    }
+    if (cp)
+        sha_update(new, cp, len);
+
+    return (PyObject *)new;
+}
+
+
+/* List of functions exported by this module */
+
+static struct PyMethodDef SHA_functions[] = {
+    {"sha256", (PyCFunction)SHA256_new, METH_VARARGS|METH_KEYWORDS, SHA256_new__doc__},
+    {"sha224", (PyCFunction)SHA224_new, METH_VARARGS|METH_KEYWORDS, SHA224_new__doc__},
+    {NULL,	NULL}		 /* Sentinel */
+};
+
+
+/* Initialize this module. */
+
+#define insint(n,v) { PyModule_AddIntConstant(m,n,v); }
+
+PyMODINIT_FUNC
+init_sha256(void)
+{
+    PyObject *m;
+
+    SHA224type.ob_type = &PyType_Type;
+    if (PyType_Ready(&SHA224type) < 0)
+        return;
+    SHA256type.ob_type = &PyType_Type;
+    if (PyType_Ready(&SHA256type) < 0)
+        return;
+    m = Py_InitModule("_sha256", SHA_functions);
+    if (m == NULL)
+	return;
+}
diff --git a/src/Modules/sha512module.c b/src/Modules/sha512module.c
new file mode 100644
--- /dev/null
+++ b/src/Modules/sha512module.c
@@ -0,0 +1,769 @@
+/* SHA512 module */
+
+/* This module provides an interface to NIST's SHA-512 and SHA-384 Algorithms */
+
+/* See below for information about the original code this module was
+   based upon. Additional work performed by:
+
+   Andrew Kuchling (amk at amk.ca)
+   Greg Stein (gstein at lyra.org)
+   Trevor Perrin (trevp at trevp.net)
+
+   Copyright (C) 2005   Gregory P. Smith (greg at krypto.org)
+   Licensed to PSF under a Contributor Agreement.
+
+*/
+
+/* SHA objects */
+
+#include "Python.h"
+#include "structmember.h"
+
+#ifdef PY_LONG_LONG /* If no PY_LONG_LONG, don't compile anything! */
+
+/* Endianness testing and definitions */
+#define TestEndianness(variable) {int i=1; variable=PCT_BIG_ENDIAN;\
+	if (*((char*)&i)==1) variable=PCT_LITTLE_ENDIAN;}
+
+#define PCT_LITTLE_ENDIAN 1
+#define PCT_BIG_ENDIAN 0
+
+/* Some useful types */
+
+typedef unsigned char SHA_BYTE;
+
+#if SIZEOF_INT == 4
+typedef unsigned int SHA_INT32;	/* 32-bit integer */
+typedef unsigned PY_LONG_LONG SHA_INT64;	/* 64-bit integer */
+#else
+/* not defined. compilation will die. */
+#endif
+
+/* The SHA block size and message digest sizes, in bytes */
+
+#define SHA_BLOCKSIZE   128
+#define SHA_DIGESTSIZE  64
+
+/* The structure for storing SHA info */
+
+typedef struct {
+    PyObject_HEAD
+    SHA_INT64 digest[8];		/* Message digest */
+    SHA_INT32 count_lo, count_hi;	/* 64-bit bit count */
+    SHA_BYTE data[SHA_BLOCKSIZE];	/* SHA data buffer */
+    int Endianness;
+    int local;				/* unprocessed amount in data */
+    int digestsize;
+} SHAobject;
+
+/* When run on a little-endian CPU we need to perform byte reversal on an
+   array of longwords. */
+
+static void longReverse(SHA_INT64 *buffer, int byteCount, int Endianness)
+{
+    SHA_INT64 value;
+
+    if ( Endianness == PCT_BIG_ENDIAN )
+	return;
+
+    byteCount /= sizeof(*buffer);
+    while (byteCount--) {
+        value = *buffer;
+
+		((unsigned char*)buffer)[0] = (unsigned char)(value >> 56) & 0xff;
+		((unsigned char*)buffer)[1] = (unsigned char)(value >> 48) & 0xff;
+		((unsigned char*)buffer)[2] = (unsigned char)(value >> 40) & 0xff;
+		((unsigned char*)buffer)[3] = (unsigned char)(value >> 32) & 0xff;
+		((unsigned char*)buffer)[4] = (unsigned char)(value >> 24) & 0xff;
+		((unsigned char*)buffer)[5] = (unsigned char)(value >> 16) & 0xff;
+		((unsigned char*)buffer)[6] = (unsigned char)(value >>  8) & 0xff;
+		((unsigned char*)buffer)[7] = (unsigned char)(value      ) & 0xff;
+        
+		buffer++;
+    }
+}
+
+static void SHAcopy(SHAobject *src, SHAobject *dest)
+{
+    dest->Endianness = src->Endianness;
+    dest->local = src->local;
+    dest->digestsize = src->digestsize;
+    dest->count_lo = src->count_lo;
+    dest->count_hi = src->count_hi;
+    memcpy(dest->digest, src->digest, sizeof(src->digest));
+    memcpy(dest->data, src->data, sizeof(src->data));
+}
+
+
+/* ------------------------------------------------------------------------
+ *
+ * This code for the SHA-512 algorithm was noted as public domain. The
+ * original headers are pasted below.
+ *
+ * Several changes have been made to make it more compatible with the
+ * Python environment and desired interface.
+ *
+ */
+
+/* LibTomCrypt, modular cryptographic library -- Tom St Denis
+ *
+ * LibTomCrypt is a library that provides various cryptographic
+ * algorithms in a highly modular and flexible manner.
+ *
+ * The library is free for all purposes without any express
+ * gurantee it works.
+ *
+ * Tom St Denis, tomstdenis at iahu.ca, http://libtomcrypt.org
+ */
+
+
+/* SHA512 by Tom St Denis */
+
+/* Various logical functions */
+#define ROR64(x, y) \
+    ( ((((x) & 0xFFFFFFFFFFFFFFFFULL)>>((unsigned PY_LONG_LONG)(y) & 63)) | \
+      ((x)<<((unsigned PY_LONG_LONG)(64-((y) & 63))))) & 0xFFFFFFFFFFFFFFFFULL)
+#define Ch(x,y,z)       (z ^ (x & (y ^ z)))
+#define Maj(x,y,z)      (((x | y) & z) | (x & y)) 
+#define S(x, n)         ROR64((x),(n))
+#define R(x, n)         (((x) & 0xFFFFFFFFFFFFFFFFULL) >> ((unsigned PY_LONG_LONG)n))
+#define Sigma0(x)       (S(x, 28) ^ S(x, 34) ^ S(x, 39))
+#define Sigma1(x)       (S(x, 14) ^ S(x, 18) ^ S(x, 41))
+#define Gamma0(x)       (S(x, 1) ^ S(x, 8) ^ R(x, 7))
+#define Gamma1(x)       (S(x, 19) ^ S(x, 61) ^ R(x, 6))
+
+
+static void
+sha512_transform(SHAobject *sha_info)
+{
+    int i;
+    SHA_INT64 S[8], W[80], t0, t1;
+
+    memcpy(W, sha_info->data, sizeof(sha_info->data));
+    longReverse(W, (int)sizeof(sha_info->data), sha_info->Endianness);
+
+    for (i = 16; i < 80; ++i) {
+		W[i] = Gamma1(W[i - 2]) + W[i - 7] + Gamma0(W[i - 15]) + W[i - 16];
+    }
+    for (i = 0; i < 8; ++i) {
+        S[i] = sha_info->digest[i];
+    }
+
+    /* Compress */
+#define RND(a,b,c,d,e,f,g,h,i,ki)                    \
+     t0 = h + Sigma1(e) + Ch(e, f, g) + ki + W[i];   \
+     t1 = Sigma0(a) + Maj(a, b, c);                  \
+     d += t0;                                        \
+     h  = t0 + t1;
+
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],0,0x428a2f98d728ae22ULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],1,0x7137449123ef65cdULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],2,0xb5c0fbcfec4d3b2fULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],3,0xe9b5dba58189dbbcULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],4,0x3956c25bf348b538ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],5,0x59f111f1b605d019ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],6,0x923f82a4af194f9bULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],7,0xab1c5ed5da6d8118ULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],8,0xd807aa98a3030242ULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],9,0x12835b0145706fbeULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],10,0x243185be4ee4b28cULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],11,0x550c7dc3d5ffb4e2ULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],12,0x72be5d74f27b896fULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],13,0x80deb1fe3b1696b1ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],14,0x9bdc06a725c71235ULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],15,0xc19bf174cf692694ULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],16,0xe49b69c19ef14ad2ULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],17,0xefbe4786384f25e3ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],18,0x0fc19dc68b8cd5b5ULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],19,0x240ca1cc77ac9c65ULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],20,0x2de92c6f592b0275ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],21,0x4a7484aa6ea6e483ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],22,0x5cb0a9dcbd41fbd4ULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],23,0x76f988da831153b5ULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],24,0x983e5152ee66dfabULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],25,0xa831c66d2db43210ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],26,0xb00327c898fb213fULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],27,0xbf597fc7beef0ee4ULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],28,0xc6e00bf33da88fc2ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],29,0xd5a79147930aa725ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],30,0x06ca6351e003826fULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],31,0x142929670a0e6e70ULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],32,0x27b70a8546d22ffcULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],33,0x2e1b21385c26c926ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],34,0x4d2c6dfc5ac42aedULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],35,0x53380d139d95b3dfULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],36,0x650a73548baf63deULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],37,0x766a0abb3c77b2a8ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],38,0x81c2c92e47edaee6ULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],39,0x92722c851482353bULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],40,0xa2bfe8a14cf10364ULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],41,0xa81a664bbc423001ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],42,0xc24b8b70d0f89791ULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],43,0xc76c51a30654be30ULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],44,0xd192e819d6ef5218ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],45,0xd69906245565a910ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],46,0xf40e35855771202aULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],47,0x106aa07032bbd1b8ULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],48,0x19a4c116b8d2d0c8ULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],49,0x1e376c085141ab53ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],50,0x2748774cdf8eeb99ULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],51,0x34b0bcb5e19b48a8ULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],52,0x391c0cb3c5c95a63ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],53,0x4ed8aa4ae3418acbULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],54,0x5b9cca4f7763e373ULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],55,0x682e6ff3d6b2b8a3ULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],56,0x748f82ee5defb2fcULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],57,0x78a5636f43172f60ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],58,0x84c87814a1f0ab72ULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],59,0x8cc702081a6439ecULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],60,0x90befffa23631e28ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],61,0xa4506cebde82bde9ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],62,0xbef9a3f7b2c67915ULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],63,0xc67178f2e372532bULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],64,0xca273eceea26619cULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],65,0xd186b8c721c0c207ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],66,0xeada7dd6cde0eb1eULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],67,0xf57d4f7fee6ed178ULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],68,0x06f067aa72176fbaULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],69,0x0a637dc5a2c898a6ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],70,0x113f9804bef90daeULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],71,0x1b710b35131c471bULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],72,0x28db77f523047d84ULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],73,0x32caab7b40c72493ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],74,0x3c9ebe0a15c9bebcULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],75,0x431d67c49c100d4cULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],76,0x4cc5d4becb3e42b6ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],77,0x597f299cfc657e2aULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],78,0x5fcb6fab3ad6faecULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],79,0x6c44198c4a475817ULL);
+
+#undef RND     
+    
+    /* feedback */
+    for (i = 0; i < 8; i++) {
+        sha_info->digest[i] = sha_info->digest[i] + S[i];
+    }
+
+}
+
+
+
+/* initialize the SHA digest */
+
+static void
+sha512_init(SHAobject *sha_info)
+{
+    TestEndianness(sha_info->Endianness)
+    sha_info->digest[0] = 0x6a09e667f3bcc908ULL;
+    sha_info->digest[1] = 0xbb67ae8584caa73bULL;
+    sha_info->digest[2] = 0x3c6ef372fe94f82bULL;
+    sha_info->digest[3] = 0xa54ff53a5f1d36f1ULL;
+    sha_info->digest[4] = 0x510e527fade682d1ULL;
+    sha_info->digest[5] = 0x9b05688c2b3e6c1fULL;
+    sha_info->digest[6] = 0x1f83d9abfb41bd6bULL;
+    sha_info->digest[7] = 0x5be0cd19137e2179ULL;
+    sha_info->count_lo = 0L;
+    sha_info->count_hi = 0L;
+    sha_info->local = 0;
+    sha_info->digestsize = 64;
+}
+
+static void
+sha384_init(SHAobject *sha_info)
+{
+    TestEndianness(sha_info->Endianness)
+    sha_info->digest[0] = 0xcbbb9d5dc1059ed8ULL;
+    sha_info->digest[1] = 0x629a292a367cd507ULL;
+    sha_info->digest[2] = 0x9159015a3070dd17ULL;
+    sha_info->digest[3] = 0x152fecd8f70e5939ULL;
+    sha_info->digest[4] = 0x67332667ffc00b31ULL;
+    sha_info->digest[5] = 0x8eb44a8768581511ULL;
+    sha_info->digest[6] = 0xdb0c2e0d64f98fa7ULL;
+    sha_info->digest[7] = 0x47b5481dbefa4fa4ULL;
+    sha_info->count_lo = 0L;
+    sha_info->count_hi = 0L;
+    sha_info->local = 0;
+    sha_info->digestsize = 48;
+}
+
+
+/* update the SHA digest */
+
+static void
+sha512_update(SHAobject *sha_info, SHA_BYTE *buffer, int count)
+{
+    int i;
+    SHA_INT32 clo;
+
+    clo = sha_info->count_lo + ((SHA_INT32) count << 3);
+    if (clo < sha_info->count_lo) {
+        ++sha_info->count_hi;
+    }
+    sha_info->count_lo = clo;
+    sha_info->count_hi += (SHA_INT32) count >> 29;
+    if (sha_info->local) {
+        i = SHA_BLOCKSIZE - sha_info->local;
+        if (i > count) {
+            i = count;
+        }
+        memcpy(((SHA_BYTE *) sha_info->data) + sha_info->local, buffer, i);
+        count -= i;
+        buffer += i;
+        sha_info->local += i;
+        if (sha_info->local == SHA_BLOCKSIZE) {
+            sha512_transform(sha_info);
+        }
+        else {
+            return;
+        }
+    }
+    while (count >= SHA_BLOCKSIZE) {
+        memcpy(sha_info->data, buffer, SHA_BLOCKSIZE);
+        buffer += SHA_BLOCKSIZE;
+        count -= SHA_BLOCKSIZE;
+        sha512_transform(sha_info);
+    }
+    memcpy(sha_info->data, buffer, count);
+    sha_info->local = count;
+}
+
+/* finish computing the SHA digest */
+
+static void
+sha512_final(unsigned char digest[SHA_DIGESTSIZE], SHAobject *sha_info)
+{
+    int count;
+    SHA_INT32 lo_bit_count, hi_bit_count;
+
+    lo_bit_count = sha_info->count_lo;
+    hi_bit_count = sha_info->count_hi;
+    count = (int) ((lo_bit_count >> 3) & 0x7f);
+    ((SHA_BYTE *) sha_info->data)[count++] = 0x80;
+    if (count > SHA_BLOCKSIZE - 16) {
+	memset(((SHA_BYTE *) sha_info->data) + count, 0,
+	       SHA_BLOCKSIZE - count);
+	sha512_transform(sha_info);
+	memset((SHA_BYTE *) sha_info->data, 0, SHA_BLOCKSIZE - 16);
+    }
+    else {
+	memset(((SHA_BYTE *) sha_info->data) + count, 0,
+	       SHA_BLOCKSIZE - 16 - count);
+    }
+
+    /* GJS: note that we add the hi/lo in big-endian. sha512_transform will
+       swap these values into host-order. */
+    sha_info->data[112] = 0;
+    sha_info->data[113] = 0;
+    sha_info->data[114] = 0;
+    sha_info->data[115] = 0;
+    sha_info->data[116] = 0;
+    sha_info->data[117] = 0;
+    sha_info->data[118] = 0;
+    sha_info->data[119] = 0;
+    sha_info->data[120] = (hi_bit_count >> 24) & 0xff;
+    sha_info->data[121] = (hi_bit_count >> 16) & 0xff;
+    sha_info->data[122] = (hi_bit_count >>  8) & 0xff;
+    sha_info->data[123] = (hi_bit_count >>  0) & 0xff;
+    sha_info->data[124] = (lo_bit_count >> 24) & 0xff;
+    sha_info->data[125] = (lo_bit_count >> 16) & 0xff;
+    sha_info->data[126] = (lo_bit_count >>  8) & 0xff;
+    sha_info->data[127] = (lo_bit_count >>  0) & 0xff;
+    sha512_transform(sha_info);
+    digest[ 0] = (unsigned char) ((sha_info->digest[0] >> 56) & 0xff);
+    digest[ 1] = (unsigned char) ((sha_info->digest[0] >> 48) & 0xff);
+    digest[ 2] = (unsigned char) ((sha_info->digest[0] >> 40) & 0xff);
+    digest[ 3] = (unsigned char) ((sha_info->digest[0] >> 32) & 0xff);
+    digest[ 4] = (unsigned char) ((sha_info->digest[0] >> 24) & 0xff);
+    digest[ 5] = (unsigned char) ((sha_info->digest[0] >> 16) & 0xff);
+    digest[ 6] = (unsigned char) ((sha_info->digest[0] >>  8) & 0xff);
+    digest[ 7] = (unsigned char) ((sha_info->digest[0]      ) & 0xff);
+    digest[ 8] = (unsigned char) ((sha_info->digest[1] >> 56) & 0xff);
+    digest[ 9] = (unsigned char) ((sha_info->digest[1] >> 48) & 0xff);
+    digest[10] = (unsigned char) ((sha_info->digest[1] >> 40) & 0xff);
+    digest[11] = (unsigned char) ((sha_info->digest[1] >> 32) & 0xff);
+    digest[12] = (unsigned char) ((sha_info->digest[1] >> 24) & 0xff);
+    digest[13] = (unsigned char) ((sha_info->digest[1] >> 16) & 0xff);
+    digest[14] = (unsigned char) ((sha_info->digest[1] >>  8) & 0xff);
+    digest[15] = (unsigned char) ((sha_info->digest[1]      ) & 0xff);
+    digest[16] = (unsigned char) ((sha_info->digest[2] >> 56) & 0xff);
+    digest[17] = (unsigned char) ((sha_info->digest[2] >> 48) & 0xff);
+    digest[18] = (unsigned char) ((sha_info->digest[2] >> 40) & 0xff);
+    digest[19] = (unsigned char) ((sha_info->digest[2] >> 32) & 0xff);
+    digest[20] = (unsigned char) ((sha_info->digest[2] >> 24) & 0xff);
+    digest[21] = (unsigned char) ((sha_info->digest[2] >> 16) & 0xff);
+    digest[22] = (unsigned char) ((sha_info->digest[2] >>  8) & 0xff);
+    digest[23] = (unsigned char) ((sha_info->digest[2]      ) & 0xff);
+    digest[24] = (unsigned char) ((sha_info->digest[3] >> 56) & 0xff);
+    digest[25] = (unsigned char) ((sha_info->digest[3] >> 48) & 0xff);
+    digest[26] = (unsigned char) ((sha_info->digest[3] >> 40) & 0xff);
+    digest[27] = (unsigned char) ((sha_info->digest[3] >> 32) & 0xff);
+    digest[28] = (unsigned char) ((sha_info->digest[3] >> 24) & 0xff);
+    digest[29] = (unsigned char) ((sha_info->digest[3] >> 16) & 0xff);
+    digest[30] = (unsigned char) ((sha_info->digest[3] >>  8) & 0xff);
+    digest[31] = (unsigned char) ((sha_info->digest[3]      ) & 0xff);
+    digest[32] = (unsigned char) ((sha_info->digest[4] >> 56) & 0xff);
+    digest[33] = (unsigned char) ((sha_info->digest[4] >> 48) & 0xff);
+    digest[34] = (unsigned char) ((sha_info->digest[4] >> 40) & 0xff);
+    digest[35] = (unsigned char) ((sha_info->digest[4] >> 32) & 0xff);
+    digest[36] = (unsigned char) ((sha_info->digest[4] >> 24) & 0xff);
+    digest[37] = (unsigned char) ((sha_info->digest[4] >> 16) & 0xff);
+    digest[38] = (unsigned char) ((sha_info->digest[4] >>  8) & 0xff);
+    digest[39] = (unsigned char) ((sha_info->digest[4]      ) & 0xff);
+    digest[40] = (unsigned char) ((sha_info->digest[5] >> 56) & 0xff);
+    digest[41] = (unsigned char) ((sha_info->digest[5] >> 48) & 0xff);
+    digest[42] = (unsigned char) ((sha_info->digest[5] >> 40) & 0xff);
+    digest[43] = (unsigned char) ((sha_info->digest[5] >> 32) & 0xff);
+    digest[44] = (unsigned char) ((sha_info->digest[5] >> 24) & 0xff);
+    digest[45] = (unsigned char) ((sha_info->digest[5] >> 16) & 0xff);
+    digest[46] = (unsigned char) ((sha_info->digest[5] >>  8) & 0xff);
+    digest[47] = (unsigned char) ((sha_info->digest[5]      ) & 0xff);
+    digest[48] = (unsigned char) ((sha_info->digest[6] >> 56) & 0xff);
+    digest[49] = (unsigned char) ((sha_info->digest[6] >> 48) & 0xff);
+    digest[50] = (unsigned char) ((sha_info->digest[6] >> 40) & 0xff);
+    digest[51] = (unsigned char) ((sha_info->digest[6] >> 32) & 0xff);
+    digest[52] = (unsigned char) ((sha_info->digest[6] >> 24) & 0xff);
+    digest[53] = (unsigned char) ((sha_info->digest[6] >> 16) & 0xff);
+    digest[54] = (unsigned char) ((sha_info->digest[6] >>  8) & 0xff);
+    digest[55] = (unsigned char) ((sha_info->digest[6]      ) & 0xff);
+    digest[56] = (unsigned char) ((sha_info->digest[7] >> 56) & 0xff);
+    digest[57] = (unsigned char) ((sha_info->digest[7] >> 48) & 0xff);
+    digest[58] = (unsigned char) ((sha_info->digest[7] >> 40) & 0xff);
+    digest[59] = (unsigned char) ((sha_info->digest[7] >> 32) & 0xff);
+    digest[60] = (unsigned char) ((sha_info->digest[7] >> 24) & 0xff);
+    digest[61] = (unsigned char) ((sha_info->digest[7] >> 16) & 0xff);
+    digest[62] = (unsigned char) ((sha_info->digest[7] >>  8) & 0xff);
+    digest[63] = (unsigned char) ((sha_info->digest[7]      ) & 0xff);
+}
+
+/*
+ * End of copied SHA code.
+ *
+ * ------------------------------------------------------------------------
+ */
+
+static PyTypeObject SHA384type;
+static PyTypeObject SHA512type;
+
+
+static SHAobject *
+newSHA384object(void)
+{
+    return (SHAobject *)PyObject_New(SHAobject, &SHA384type);
+}
+
+static SHAobject *
+newSHA512object(void)
+{
+    return (SHAobject *)PyObject_New(SHAobject, &SHA512type);
+}
+
+/* Internal methods for a hash object */
+
+static void
+SHA512_dealloc(PyObject *ptr)
+{
+    PyObject_Del(ptr);
+}
+
+
+/* External methods for a hash object */
+
+PyDoc_STRVAR(SHA512_copy__doc__, "Return a copy of the hash object.");
+
+static PyObject *
+SHA512_copy(SHAobject *self, PyObject *unused)
+{
+    SHAobject *newobj;
+
+    if (((PyObject*)self)->ob_type == &SHA512type) {
+        if ( (newobj = newSHA512object())==NULL)
+            return NULL;
+    } else {
+        if ( (newobj = newSHA384object())==NULL)
+            return NULL;
+    }
+
+    SHAcopy(self, newobj);
+    return (PyObject *)newobj;
+}
+
+PyDoc_STRVAR(SHA512_digest__doc__,
+"Return the digest value as a string of binary data.");
+
+static PyObject *
+SHA512_digest(SHAobject *self, PyObject *unused)
+{
+    unsigned char digest[SHA_DIGESTSIZE];
+    SHAobject temp;
+
+    SHAcopy(self, &temp);
+    sha512_final(digest, &temp);
+    return PyString_FromStringAndSize((const char *)digest, self->digestsize);
+}
+
+PyDoc_STRVAR(SHA512_hexdigest__doc__,
+"Return the digest value as a string of hexadecimal digits.");
+
+static PyObject *
+SHA512_hexdigest(SHAobject *self, PyObject *unused)
+{
+    unsigned char digest[SHA_DIGESTSIZE];
+    SHAobject temp;
+    PyObject *retval;
+    char *hex_digest;
+    int i, j;
+
+    /* Get the raw (binary) digest value */
+    SHAcopy(self, &temp);
+    sha512_final(digest, &temp);
+
+    /* Create a new string */
+    retval = PyString_FromStringAndSize(NULL, self->digestsize * 2);
+    if (!retval)
+	    return NULL;
+    hex_digest = PyString_AsString(retval);
+    if (!hex_digest) {
+	    Py_DECREF(retval);
+	    return NULL;
+    }
+
+    /* Make hex version of the digest */
+    for (i=j=0; i<self->digestsize; i++) {
+        char c;
+        c = (digest[i] >> 4) & 0xf;
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+        c = (digest[i] & 0xf);
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+    }
+    return retval;
+}
+
+PyDoc_STRVAR(SHA512_update__doc__,
+"Update this hash object's state with the provided string.");
+
+static PyObject *
+SHA512_update(SHAobject *self, PyObject *args)
+{
+    unsigned char *cp;
+    int len;
+
+    if (!PyArg_ParseTuple(args, "s#:update", &cp, &len))
+        return NULL;
+
+    sha512_update(self, cp, len);
+
+    Py_INCREF(Py_None);
+    return Py_None;
+}
+
+static PyMethodDef SHA_methods[] = {
+    {"copy",	  (PyCFunction)SHA512_copy,      METH_NOARGS, SHA512_copy__doc__},
+    {"digest",	  (PyCFunction)SHA512_digest,    METH_NOARGS, SHA512_digest__doc__},
+    {"hexdigest", (PyCFunction)SHA512_hexdigest, METH_NOARGS, SHA512_hexdigest__doc__},
+    {"update",	  (PyCFunction)SHA512_update,    METH_VARARGS, SHA512_update__doc__},
+    {NULL,	  NULL}		/* sentinel */
+};
+
+static PyObject *
+SHA512_get_block_size(PyObject *self, void *closure)
+{
+    return PyInt_FromLong(SHA_BLOCKSIZE);
+}
+
+static PyObject *
+SHA512_get_name(PyObject *self, void *closure)
+{
+    if (((SHAobject *)self)->digestsize == 64)
+        return PyString_FromStringAndSize("SHA512", 6);
+    else
+        return PyString_FromStringAndSize("SHA384", 6);
+}
+
+static PyGetSetDef SHA_getseters[] = {
+    {"block_size",
+     (getter)SHA512_get_block_size, NULL,
+     NULL,
+     NULL},
+    {"name",
+     (getter)SHA512_get_name, NULL,
+     NULL,
+     NULL},
+    {NULL}  /* Sentinel */
+};
+
+static PyMemberDef SHA_members[] = {
+    {"digest_size", T_INT, offsetof(SHAobject, digestsize), READONLY, NULL},
+    /* the old md5 and sha modules support 'digest_size' as in PEP 247.
+     * the old sha module also supported 'digestsize'.  ugh. */
+    {"digestsize", T_INT, offsetof(SHAobject, digestsize), READONLY, NULL},
+    {NULL}  /* Sentinel */
+};
+
+static PyTypeObject SHA384type = {
+    PyObject_HEAD_INIT(NULL)
+    0,			/*ob_size*/
+    "_sha512.sha384",	/*tp_name*/
+    sizeof(SHAobject),	/*tp_size*/
+    0,			/*tp_itemsize*/
+    /* methods */
+    SHA512_dealloc,	/*tp_dealloc*/
+    0,			/*tp_print*/
+    0,          	/*tp_getattr*/
+    0,                  /*tp_setattr*/
+    0,                  /*tp_compare*/
+    0,                  /*tp_repr*/
+    0,                  /*tp_as_number*/
+    0,                  /*tp_as_sequence*/
+    0,                  /*tp_as_mapping*/
+    0,                  /*tp_hash*/
+    0,                  /*tp_call*/
+    0,                  /*tp_str*/
+    0,                  /*tp_getattro*/
+    0,                  /*tp_setattro*/
+    0,                  /*tp_as_buffer*/
+    Py_TPFLAGS_DEFAULT, /*tp_flags*/
+    0,                  /*tp_doc*/
+    0,                  /*tp_traverse*/
+    0,			/*tp_clear*/
+    0,			/*tp_richcompare*/
+    0,			/*tp_weaklistoffset*/
+    0,			/*tp_iter*/
+    0,			/*tp_iternext*/
+    SHA_methods,	/* tp_methods */
+    SHA_members,	/* tp_members */
+    SHA_getseters,      /* tp_getset */
+};
+
+static PyTypeObject SHA512type = {
+    PyObject_HEAD_INIT(NULL)
+    0,			/*ob_size*/
+    "_sha512.sha512",	/*tp_name*/
+    sizeof(SHAobject),	/*tp_size*/
+    0,			/*tp_itemsize*/
+    /* methods */
+    SHA512_dealloc,	/*tp_dealloc*/
+    0,			/*tp_print*/
+    0,          	/*tp_getattr*/
+    0,                  /*tp_setattr*/
+    0,                  /*tp_compare*/
+    0,                  /*tp_repr*/
+    0,                  /*tp_as_number*/
+    0,                  /*tp_as_sequence*/
+    0,                  /*tp_as_mapping*/
+    0,                  /*tp_hash*/
+    0,                  /*tp_call*/
+    0,                  /*tp_str*/
+    0,                  /*tp_getattro*/
+    0,                  /*tp_setattro*/
+    0,                  /*tp_as_buffer*/
+    Py_TPFLAGS_DEFAULT, /*tp_flags*/
+    0,                  /*tp_doc*/
+    0,                  /*tp_traverse*/
+    0,			/*tp_clear*/
+    0,			/*tp_richcompare*/
+    0,			/*tp_weaklistoffset*/
+    0,			/*tp_iter*/
+    0,			/*tp_iternext*/
+    SHA_methods,	/* tp_methods */
+    SHA_members,	/* tp_members */
+    SHA_getseters,      /* tp_getset */
+};
+
+
+/* The single module-level function: new() */
+
+PyDoc_STRVAR(SHA512_new__doc__,
+"Return a new SHA-512 hash object; optionally initialized with a string.");
+
+static PyObject *
+SHA512_new(PyObject *self, PyObject *args, PyObject *kwdict)
+{
+    static char *kwlist[] = {"string", NULL};
+    SHAobject *new;
+    unsigned char *cp = NULL;
+    int len;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwdict, "|s#:new", kwlist,
+                                     &cp, &len)) {
+        return NULL;
+    }
+
+    if ((new = newSHA512object()) == NULL)
+        return NULL;
+
+    sha512_init(new);
+
+    if (PyErr_Occurred()) {
+        Py_DECREF(new);
+        return NULL;
+    }
+    if (cp)
+        sha512_update(new, cp, len);
+
+    return (PyObject *)new;
+}
+
+PyDoc_STRVAR(SHA384_new__doc__,
+"Return a new SHA-384 hash object; optionally initialized with a string.");
+
+static PyObject *
+SHA384_new(PyObject *self, PyObject *args, PyObject *kwdict)
+{
+    static char *kwlist[] = {"string", NULL};
+    SHAobject *new;
+    unsigned char *cp = NULL;
+    int len;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwdict, "|s#:new", kwlist,
+                                     &cp, &len)) {
+        return NULL;
+    }
+
+    if ((new = newSHA384object()) == NULL)
+        return NULL;
+
+    sha384_init(new);
+
+    if (PyErr_Occurred()) {
+        Py_DECREF(new);
+        return NULL;
+    }
+    if (cp)
+        sha512_update(new, cp, len);
+
+    return (PyObject *)new;
+}
+
+
+/* List of functions exported by this module */
+
+static struct PyMethodDef SHA_functions[] = {
+    {"sha512", (PyCFunction)SHA512_new, METH_VARARGS|METH_KEYWORDS, SHA512_new__doc__},
+    {"sha384", (PyCFunction)SHA384_new, METH_VARARGS|METH_KEYWORDS, SHA384_new__doc__},
+    {NULL,	NULL}		 /* Sentinel */
+};
+
+
+/* Initialize this module. */
+
+#define insint(n,v) { PyModule_AddIntConstant(m,n,v); }
+
+PyMODINIT_FUNC
+init_sha512(void)
+{
+    PyObject *m;
+
+    SHA384type.ob_type = &PyType_Type;
+    if (PyType_Ready(&SHA384type) < 0)
+        return;
+    SHA512type.ob_type = &PyType_Type;
+    if (PyType_Ready(&SHA512type) < 0)
+        return;
+    m = Py_InitModule("_sha512", SHA_functions);
+    if (m == NULL)
+	return;
+}
+
+#endif
diff --git a/src/Modules/shamodule.c b/src/Modules/shamodule.c
new file mode 100644
--- /dev/null
+++ b/src/Modules/shamodule.c
@@ -0,0 +1,593 @@
+/* SHA module */
+
+/* This module provides an interface to NIST's Secure Hash Algorithm */
+
+/* See below for information about the original code this module was
+   based upon. Additional work performed by:
+
+   Andrew Kuchling (amk at amk.ca)
+   Greg Stein (gstein at lyra.org)
+
+   Copyright (C) 2005   Gregory P. Smith (greg at krypto.org)
+   Licensed to PSF under a Contributor Agreement.
+
+*/
+
+/* SHA objects */
+
+#include "Python.h"
+#include "structmember.h"
+
+
+/* Endianness testing and definitions */
+#define TestEndianness(variable) {int i=1; variable=PCT_BIG_ENDIAN;\
+	if (*((char*)&i)==1) variable=PCT_LITTLE_ENDIAN;}
+
+#define PCT_LITTLE_ENDIAN 1
+#define PCT_BIG_ENDIAN 0
+
+/* Some useful types */
+
+typedef unsigned char SHA_BYTE;
+
+#if SIZEOF_INT == 4
+typedef unsigned int SHA_INT32;	/* 32-bit integer */
+#else
+/* not defined. compilation will die. */
+#endif
+
+/* The SHA block size and message digest sizes, in bytes */
+
+#define SHA_BLOCKSIZE    64
+#define SHA_DIGESTSIZE  20
+
+/* The structure for storing SHS info */
+
+typedef struct {
+    PyObject_HEAD
+    SHA_INT32 digest[5];		/* Message digest */
+    SHA_INT32 count_lo, count_hi;	/* 64-bit bit count */
+    SHA_BYTE data[SHA_BLOCKSIZE];	/* SHA data buffer */
+    int Endianness;
+    int local;				/* unprocessed amount in data */
+} SHAobject;
+
+/* When run on a little-endian CPU we need to perform byte reversal on an
+   array of longwords. */
+
+static void longReverse(SHA_INT32 *buffer, int byteCount, int Endianness)
+{
+    SHA_INT32 value;
+
+    if ( Endianness == PCT_BIG_ENDIAN )
+	return;
+
+    byteCount /= sizeof(*buffer);
+    while (byteCount--) {
+        value = *buffer;
+        value = ( ( value & 0xFF00FF00L ) >> 8  ) | \
+                ( ( value & 0x00FF00FFL ) << 8 );
+        *buffer++ = ( value << 16 ) | ( value >> 16 );
+    }
+}
+
+static void SHAcopy(SHAobject *src, SHAobject *dest)
+{
+    dest->Endianness = src->Endianness;
+    dest->local = src->local;
+    dest->count_lo = src->count_lo;
+    dest->count_hi = src->count_hi;
+    memcpy(dest->digest, src->digest, sizeof(src->digest));
+    memcpy(dest->data, src->data, sizeof(src->data));
+}
+
+
+/* ------------------------------------------------------------------------
+ *
+ * This code for the SHA algorithm was noted as public domain. The original
+ * headers are pasted below.
+ *
+ * Several changes have been made to make it more compatible with the
+ * Python environment and desired interface.
+ *
+ */
+
+/* NIST Secure Hash Algorithm */
+/* heavily modified by Uwe Hollerbach <uh at alumni.caltech edu> */
+/* from Peter C. Gutmann's implementation as found in */
+/* Applied Cryptography by Bruce Schneier */
+/* Further modifications to include the "UNRAVEL" stuff, below */
+
+/* This code is in the public domain */
+
+/* UNRAVEL should be fastest & biggest */
+/* UNROLL_LOOPS should be just as big, but slightly slower */
+/* both undefined should be smallest and slowest */
+
+#define UNRAVEL
+/* #define UNROLL_LOOPS */
+
+/* The SHA f()-functions.  The f1 and f3 functions can be optimized to
+   save one boolean operation each - thanks to Rich Schroeppel,
+   rcs at cs.arizona.edu for discovering this */
+
+/*#define f1(x,y,z)	((x & y) | (~x & z))		// Rounds  0-19 */
+#define f1(x,y,z)	(z ^ (x & (y ^ z)))		/* Rounds  0-19 */
+#define f2(x,y,z)	(x ^ y ^ z)			/* Rounds 20-39 */
+/*#define f3(x,y,z)	((x & y) | (x & z) | (y & z))	// Rounds 40-59 */
+#define f3(x,y,z)	((x & y) | (z & (x | y)))	/* Rounds 40-59 */
+#define f4(x,y,z)	(x ^ y ^ z)			/* Rounds 60-79 */
+
+/* SHA constants */
+
+#define CONST1		0x5a827999L			/* Rounds  0-19 */
+#define CONST2		0x6ed9eba1L			/* Rounds 20-39 */
+#define CONST3		0x8f1bbcdcL			/* Rounds 40-59 */
+#define CONST4		0xca62c1d6L			/* Rounds 60-79 */
+
+/* 32-bit rotate */
+
+#define R32(x,n)	((x << n) | (x >> (32 - n)))
+
+/* the generic case, for when the overall rotation is not unraveled */
+
+#define FG(n)	\
+    T = R32(A,5) + f##n(B,C,D) + E + *WP++ + CONST##n;	\
+    E = D; D = C; C = R32(B,30); B = A; A = T
+
+/* specific cases, for when the overall rotation is unraveled */
+
+#define FA(n)	\
+    T = R32(A,5) + f##n(B,C,D) + E + *WP++ + CONST##n; B = R32(B,30)
+
+#define FB(n)	\
+    E = R32(T,5) + f##n(A,B,C) + D + *WP++ + CONST##n; A = R32(A,30)
+
+#define FC(n)	\
+    D = R32(E,5) + f##n(T,A,B) + C + *WP++ + CONST##n; T = R32(T,30)
+
+#define FD(n)	\
+    C = R32(D,5) + f##n(E,T,A) + B + *WP++ + CONST##n; E = R32(E,30)
+
+#define FE(n)	\
+    B = R32(C,5) + f##n(D,E,T) + A + *WP++ + CONST##n; D = R32(D,30)
+
+#define FT(n)	\
+    A = R32(B,5) + f##n(C,D,E) + T + *WP++ + CONST##n; C = R32(C,30)
+
+/* do SHA transformation */
+
+static void
+sha_transform(SHAobject *sha_info)
+{
+    int i;
+    SHA_INT32 T, A, B, C, D, E, W[80], *WP;
+
+    memcpy(W, sha_info->data, sizeof(sha_info->data));
+    longReverse(W, (int)sizeof(sha_info->data), sha_info->Endianness);
+
+    for (i = 16; i < 80; ++i) {
+	W[i] = W[i-3] ^ W[i-8] ^ W[i-14] ^ W[i-16];
+
+	/* extra rotation fix */
+	W[i] = R32(W[i], 1);
+    }
+    A = sha_info->digest[0];
+    B = sha_info->digest[1];
+    C = sha_info->digest[2];
+    D = sha_info->digest[3];
+    E = sha_info->digest[4];
+    WP = W;
+#ifdef UNRAVEL
+    FA(1); FB(1); FC(1); FD(1); FE(1); FT(1); FA(1); FB(1); FC(1); FD(1);
+    FE(1); FT(1); FA(1); FB(1); FC(1); FD(1); FE(1); FT(1); FA(1); FB(1);
+    FC(2); FD(2); FE(2); FT(2); FA(2); FB(2); FC(2); FD(2); FE(2); FT(2);
+    FA(2); FB(2); FC(2); FD(2); FE(2); FT(2); FA(2); FB(2); FC(2); FD(2);
+    FE(3); FT(3); FA(3); FB(3); FC(3); FD(3); FE(3); FT(3); FA(3); FB(3);
+    FC(3); FD(3); FE(3); FT(3); FA(3); FB(3); FC(3); FD(3); FE(3); FT(3);
+    FA(4); FB(4); FC(4); FD(4); FE(4); FT(4); FA(4); FB(4); FC(4); FD(4);
+    FE(4); FT(4); FA(4); FB(4); FC(4); FD(4); FE(4); FT(4); FA(4); FB(4);
+    sha_info->digest[0] += E;
+    sha_info->digest[1] += T;
+    sha_info->digest[2] += A;
+    sha_info->digest[3] += B;
+    sha_info->digest[4] += C;
+#else /* !UNRAVEL */
+#ifdef UNROLL_LOOPS
+    FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1);
+    FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1);
+    FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2);
+    FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2);
+    FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3);
+    FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3);
+    FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4);
+    FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4);
+#else /* !UNROLL_LOOPS */
+    for (i =  0; i < 20; ++i) { FG(1); }
+    for (i = 20; i < 40; ++i) { FG(2); }
+    for (i = 40; i < 60; ++i) { FG(3); }
+    for (i = 60; i < 80; ++i) { FG(4); }
+#endif /* !UNROLL_LOOPS */
+    sha_info->digest[0] += A;
+    sha_info->digest[1] += B;
+    sha_info->digest[2] += C;
+    sha_info->digest[3] += D;
+    sha_info->digest[4] += E;
+#endif /* !UNRAVEL */
+}
+
+/* initialize the SHA digest */
+
+static void
+sha_init(SHAobject *sha_info)
+{
+    TestEndianness(sha_info->Endianness)
+
+    sha_info->digest[0] = 0x67452301L;
+    sha_info->digest[1] = 0xefcdab89L;
+    sha_info->digest[2] = 0x98badcfeL;
+    sha_info->digest[3] = 0x10325476L;
+    sha_info->digest[4] = 0xc3d2e1f0L;
+    sha_info->count_lo = 0L;
+    sha_info->count_hi = 0L;
+    sha_info->local = 0;
+}
+
+/* update the SHA digest */
+
+static void
+sha_update(SHAobject *sha_info, SHA_BYTE *buffer, int count)
+{
+    int i;
+    SHA_INT32 clo;
+
+    clo = sha_info->count_lo + ((SHA_INT32) count << 3);
+    if (clo < sha_info->count_lo) {
+        ++sha_info->count_hi;
+    }
+    sha_info->count_lo = clo;
+    sha_info->count_hi += (SHA_INT32) count >> 29;
+    if (sha_info->local) {
+        i = SHA_BLOCKSIZE - sha_info->local;
+        if (i > count) {
+            i = count;
+        }
+        memcpy(((SHA_BYTE *) sha_info->data) + sha_info->local, buffer, i);
+        count -= i;
+        buffer += i;
+        sha_info->local += i;
+        if (sha_info->local == SHA_BLOCKSIZE) {
+            sha_transform(sha_info);
+        }
+        else {
+            return;
+        }
+    }
+    while (count >= SHA_BLOCKSIZE) {
+        memcpy(sha_info->data, buffer, SHA_BLOCKSIZE);
+        buffer += SHA_BLOCKSIZE;
+        count -= SHA_BLOCKSIZE;
+        sha_transform(sha_info);
+    }
+    memcpy(sha_info->data, buffer, count);
+    sha_info->local = count;
+}
+
+/* finish computing the SHA digest */
+
+static void
+sha_final(unsigned char digest[20], SHAobject *sha_info)
+{
+    int count;
+    SHA_INT32 lo_bit_count, hi_bit_count;
+
+    lo_bit_count = sha_info->count_lo;
+    hi_bit_count = sha_info->count_hi;
+    count = (int) ((lo_bit_count >> 3) & 0x3f);
+    ((SHA_BYTE *) sha_info->data)[count++] = 0x80;
+    if (count > SHA_BLOCKSIZE - 8) {
+	memset(((SHA_BYTE *) sha_info->data) + count, 0,
+	       SHA_BLOCKSIZE - count);
+	sha_transform(sha_info);
+	memset((SHA_BYTE *) sha_info->data, 0, SHA_BLOCKSIZE - 8);
+    }
+    else {
+	memset(((SHA_BYTE *) sha_info->data) + count, 0,
+	       SHA_BLOCKSIZE - 8 - count);
+    }
+
+    /* GJS: note that we add the hi/lo in big-endian. sha_transform will
+       swap these values into host-order. */
+    sha_info->data[56] = (hi_bit_count >> 24) & 0xff;
+    sha_info->data[57] = (hi_bit_count >> 16) & 0xff;
+    sha_info->data[58] = (hi_bit_count >>  8) & 0xff;
+    sha_info->data[59] = (hi_bit_count >>  0) & 0xff;
+    sha_info->data[60] = (lo_bit_count >> 24) & 0xff;
+    sha_info->data[61] = (lo_bit_count >> 16) & 0xff;
+    sha_info->data[62] = (lo_bit_count >>  8) & 0xff;
+    sha_info->data[63] = (lo_bit_count >>  0) & 0xff;
+    sha_transform(sha_info);
+    digest[ 0] = (unsigned char) ((sha_info->digest[0] >> 24) & 0xff);
+    digest[ 1] = (unsigned char) ((sha_info->digest[0] >> 16) & 0xff);
+    digest[ 2] = (unsigned char) ((sha_info->digest[0] >>  8) & 0xff);
+    digest[ 3] = (unsigned char) ((sha_info->digest[0]      ) & 0xff);
+    digest[ 4] = (unsigned char) ((sha_info->digest[1] >> 24) & 0xff);
+    digest[ 5] = (unsigned char) ((sha_info->digest[1] >> 16) & 0xff);
+    digest[ 6] = (unsigned char) ((sha_info->digest[1] >>  8) & 0xff);
+    digest[ 7] = (unsigned char) ((sha_info->digest[1]      ) & 0xff);
+    digest[ 8] = (unsigned char) ((sha_info->digest[2] >> 24) & 0xff);
+    digest[ 9] = (unsigned char) ((sha_info->digest[2] >> 16) & 0xff);
+    digest[10] = (unsigned char) ((sha_info->digest[2] >>  8) & 0xff);
+    digest[11] = (unsigned char) ((sha_info->digest[2]      ) & 0xff);
+    digest[12] = (unsigned char) ((sha_info->digest[3] >> 24) & 0xff);
+    digest[13] = (unsigned char) ((sha_info->digest[3] >> 16) & 0xff);
+    digest[14] = (unsigned char) ((sha_info->digest[3] >>  8) & 0xff);
+    digest[15] = (unsigned char) ((sha_info->digest[3]      ) & 0xff);
+    digest[16] = (unsigned char) ((sha_info->digest[4] >> 24) & 0xff);
+    digest[17] = (unsigned char) ((sha_info->digest[4] >> 16) & 0xff);
+    digest[18] = (unsigned char) ((sha_info->digest[4] >>  8) & 0xff);
+    digest[19] = (unsigned char) ((sha_info->digest[4]      ) & 0xff);
+}
+
+/*
+ * End of copied SHA code.
+ *
+ * ------------------------------------------------------------------------
+ */
+
+static PyTypeObject SHAtype;
+
+
+static SHAobject *
+newSHAobject(void)
+{
+    return (SHAobject *)PyObject_New(SHAobject, &SHAtype);
+}
+
+/* Internal methods for a hashing object */
+
+static void
+SHA_dealloc(PyObject *ptr)
+{
+    PyObject_Del(ptr);
+}
+
+
+/* External methods for a hashing object */
+
+PyDoc_STRVAR(SHA_copy__doc__, "Return a copy of the hashing object.");
+
+static PyObject *
+SHA_copy(SHAobject *self, PyObject *unused)
+{
+    SHAobject *newobj;
+
+    if ( (newobj = newSHAobject())==NULL)
+        return NULL;
+
+    SHAcopy(self, newobj);
+    return (PyObject *)newobj;
+}
+
+PyDoc_STRVAR(SHA_digest__doc__,
+"Return the digest value as a string of binary data.");
+
+static PyObject *
+SHA_digest(SHAobject *self, PyObject *unused)
+{
+    unsigned char digest[SHA_DIGESTSIZE];
+    SHAobject temp;
+
+    SHAcopy(self, &temp);
+    sha_final(digest, &temp);
+    return PyString_FromStringAndSize((const char *)digest, sizeof(digest));
+}
+
+PyDoc_STRVAR(SHA_hexdigest__doc__,
+"Return the digest value as a string of hexadecimal digits.");
+
+static PyObject *
+SHA_hexdigest(SHAobject *self, PyObject *unused)
+{
+    unsigned char digest[SHA_DIGESTSIZE];
+    SHAobject temp;
+    PyObject *retval;
+    char *hex_digest;
+    int i, j;
+
+    /* Get the raw (binary) digest value */
+    SHAcopy(self, &temp);
+    sha_final(digest, &temp);
+
+    /* Create a new string */
+    retval = PyString_FromStringAndSize(NULL, sizeof(digest) * 2);
+    if (!retval)
+	    return NULL;
+    hex_digest = PyString_AsString(retval);
+    if (!hex_digest) {
+	    Py_DECREF(retval);
+	    return NULL;
+    }
+
+    /* Make hex version of the digest */
+    for(i=j=0; i<sizeof(digest); i++) {
+        char c;
+        c = (digest[i] >> 4) & 0xf;
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+        c = (digest[i] & 0xf);
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+    }
+    return retval;
+}
+
+PyDoc_STRVAR(SHA_update__doc__,
+"Update this hashing object's state with the provided string.");
+
+static PyObject *
+SHA_update(SHAobject *self, PyObject *args)
+{
+    unsigned char *cp;
+    int len;
+
+    if (!PyArg_ParseTuple(args, "s#:update", &cp, &len))
+        return NULL;
+
+    sha_update(self, cp, len);
+
+    Py_INCREF(Py_None);
+    return Py_None;
+}
+
+static PyMethodDef SHA_methods[] = {
+    {"copy",	  (PyCFunction)SHA_copy,      METH_NOARGS,  SHA_copy__doc__},
+    {"digest",	  (PyCFunction)SHA_digest,    METH_NOARGS,  SHA_digest__doc__},
+    {"hexdigest", (PyCFunction)SHA_hexdigest, METH_NOARGS,  SHA_hexdigest__doc__},
+    {"update",	  (PyCFunction)SHA_update,    METH_VARARGS, SHA_update__doc__},
+    {NULL,	  NULL}		/* sentinel */
+};
+
+static PyObject *
+SHA_get_block_size(PyObject *self, void *closure)
+{
+    return PyInt_FromLong(SHA_BLOCKSIZE);
+}
+
+static PyObject *
+SHA_get_digest_size(PyObject *self, void *closure)
+{
+    return PyInt_FromLong(SHA_DIGESTSIZE);
+}
+
+static PyObject *
+SHA_get_name(PyObject *self, void *closure)
+{
+    return PyString_FromStringAndSize("SHA1", 4);
+}
+
+static PyGetSetDef SHA_getseters[] = {
+    {"digest_size",
+     (getter)SHA_get_digest_size, NULL,
+     NULL,
+     NULL},
+    {"block_size",
+     (getter)SHA_get_block_size, NULL,
+     NULL,
+     NULL},
+    {"name",
+     (getter)SHA_get_name, NULL,
+     NULL,
+     NULL},
+    /* the old md5 and sha modules support 'digest_size' as in PEP 247.
+     * the old sha module also supported 'digestsize'.  ugh. */
+    {"digestsize",
+     (getter)SHA_get_digest_size, NULL,
+     NULL,
+     NULL},
+    {NULL}  /* Sentinel */
+};
+
+static PyTypeObject SHAtype = {
+    PyObject_HEAD_INIT(NULL)
+    0,			/*ob_size*/
+    "_sha.sha",		/*tp_name*/
+    sizeof(SHAobject),	/*tp_size*/
+    0,			/*tp_itemsize*/
+    /* methods */
+    SHA_dealloc,	/*tp_dealloc*/
+    0,			/*tp_print*/
+    0,                  /*tp_getattr*/
+    0,                  /*tp_setattr*/
+    0,                  /*tp_compare*/
+    0,                  /*tp_repr*/
+    0,                  /*tp_as_number*/
+    0,                  /*tp_as_sequence*/
+    0,                  /*tp_as_mapping*/
+    0,                  /*tp_hash*/
+    0,                  /*tp_call*/
+    0,                  /*tp_str*/
+    0,                  /*tp_getattro*/
+    0,                  /*tp_setattro*/
+    0,                  /*tp_as_buffer*/
+    Py_TPFLAGS_DEFAULT, /*tp_flags*/
+    0,                  /*tp_doc*/
+    0,                  /*tp_traverse*/
+    0,			/*tp_clear*/
+    0,			/*tp_richcompare*/
+    0,			/*tp_weaklistoffset*/
+    0,			/*tp_iter*/
+    0,			/*tp_iternext*/
+    SHA_methods,	/* tp_methods */
+    0,                  /* tp_members */
+    SHA_getseters,      /* tp_getset */
+};
+
+
+/* The single module-level function: new() */
+
+PyDoc_STRVAR(SHA_new__doc__,
+"Return a new SHA hashing object.  An optional string argument\n\
+may be provided; if present, this string will be automatically\n\
+hashed.");
+
+static PyObject *
+SHA_new(PyObject *self, PyObject *args, PyObject *kwdict)
+{
+    static char *kwlist[] = {"string", NULL};
+    SHAobject *new;
+    unsigned char *cp = NULL;
+    int len;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwdict, "|s#:new", kwlist,
+                                     &cp, &len)) {
+        return NULL;
+    }
+
+    if ((new = newSHAobject()) == NULL)
+        return NULL;
+
+    sha_init(new);
+
+    if (PyErr_Occurred()) {
+        Py_DECREF(new);
+        return NULL;
+    }
+    if (cp)
+        sha_update(new, cp, len);
+
+    return (PyObject *)new;
+}
+
+
+/* List of functions exported by this module */
+
+static struct PyMethodDef SHA_functions[] = {
+    {"new", (PyCFunction)SHA_new, METH_VARARGS|METH_KEYWORDS, SHA_new__doc__},
+    {NULL,	NULL}		 /* Sentinel */
+};
+
+
+/* Initialize this module. */
+
+#define insint(n,v) { PyModule_AddIntConstant(m,n,v); }
+
+PyMODINIT_FUNC
+init_sha(void)
+{
+    PyObject *m;
+
+    SHAtype.ob_type = &PyType_Type;
+    if (PyType_Ready(&SHAtype) < 0)
+        return;
+    m = Py_InitModule("_sha", SHA_functions);
+    if (m == NULL)
+	return;
+
+    /* Add some symbolic constants to the module */
+    insint("blocksize", 1);  /* For future use, in case some hash
+                                functions require an integral number of
+                                blocks */ 
+    insint("digestsize", 20);
+    insint("digest_size", 20);
+}
diff --git a/src/distutils2/_backport/hashlib.py b/src/distutils2/_backport/hashlib.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/hashlib.py
@@ -0,0 +1,143 @@
+# $Id$
+#
+#  Copyright (C) 2005   Gregory P. Smith (greg at krypto.org)
+#  Licensed to PSF under a Contributor Agreement.
+#
+
+__doc__ = """hashlib module - A common interface to many hash functions.
+
+new(name, string='') - returns a new hash object implementing the
+                       given hash function; initializing the hash
+                       using the given string data.
+
+Named constructor functions are also available, these are much faster
+than using new():
+
+md5(), sha1(), sha224(), sha256(), sha384(), and sha512()
+
+More algorithms may be available on your platform but the above are
+guaranteed to exist.
+
+NOTE: If you want the adler32 or crc32 hash functions they are available in
+the zlib module.
+
+Choose your hash function wisely.  Some have known collision weaknesses.
+sha384 and sha512 will be slow on 32 bit platforms.
+
+Hash objects have these methods:
+ - update(arg): Update the hash object with the string arg. Repeated calls
+                are equivalent to a single call with the concatenation of all
+                the arguments.
+ - digest():    Return the digest of the strings passed to the update() method
+                so far. This may contain non-ASCII characters, including
+                NUL bytes.
+ - hexdigest(): Like digest() except the digest is returned as a string of
+                double length, containing only hexadecimal digits.
+ - copy():      Return a copy (clone) of the hash object. This can be used to
+                efficiently compute the digests of strings that share a common
+                initial substring.
+
+For example, to obtain the digest of the string 'Nobody inspects the
+spammish repetition':
+
+    >>> import hashlib
+    >>> m = hashlib.md5()
+    >>> m.update("Nobody inspects")
+    >>> m.update(" the spammish repetition")
+    >>> m.digest()
+    '\\xbbd\\x9c\\x83\\xdd\\x1e\\xa5\\xc9\\xd9\\xde\\xc9\\xa1\\x8d\\xf0\\xff\\xe9'
+
+More condensed:
+
+    >>> hashlib.sha224("Nobody inspects the spammish repetition").hexdigest()
+    'a4337bc45a8fc544c03f52dc550cd6e1e87021bc896588bd79e901e2'
+
+"""
+
+# This tuple and __get_builtin_constructor() must be modified if a new
+# always available algorithm is added.
+__always_supported = ('md5', 'sha1', 'sha224', 'sha256', 'sha384', 'sha512')
+
+algorithms = __always_supported
+
+__all__ = __always_supported + ('new', 'algorithms')
+
+
+def __get_builtin_constructor(name):
+    if name in ('SHA1', 'sha1'):
+        import _sha
+        return _sha.new
+    elif name in ('MD5', 'md5'):
+        import _md5
+        return _md5.new
+    elif name in ('SHA256', 'sha256', 'SHA224', 'sha224'):
+        import _sha256
+        bs = name[3:]
+        if bs == '256':
+            return _sha256.sha256
+        elif bs == '224':
+            return _sha256.sha224
+    elif name in ('SHA512', 'sha512', 'SHA384', 'sha384'):
+        import _sha512
+        bs = name[3:]
+        if bs == '512':
+            return _sha512.sha512
+        elif bs == '384':
+            return _sha512.sha384
+
+    raise ValueError('unsupported hash type %s' % name)
+
+
+def __get_openssl_constructor(name):
+    try:
+        f = getattr(_hashlib, 'openssl_' + name)
+        # Allow the C module to raise ValueError.  The function will be
+        # defined but the hash not actually available thanks to OpenSSL.
+        f()
+        # Use the C function directly (very fast)
+        return f
+    except (AttributeError, ValueError):
+        return __get_builtin_constructor(name)
+
+
+def __py_new(name, string=''):
+    """new(name, string='') - Return a new hashing object using the named algorithm;
+    optionally initialized with a string.
+    """
+    return __get_builtin_constructor(name)(string)
+
+
+def __hash_new(name, string=''):
+    """new(name, string='') - Return a new hashing object using the named algorithm;
+    optionally initialized with a string.
+    """
+    try:
+        return _hashlib.new(name, string)
+    except ValueError:
+        # If the _hashlib module (OpenSSL) doesn't support the named
+        # hash, try using our builtin implementations.
+        # This allows for SHA224/256 and SHA384/512 support even though
+        # the OpenSSL library prior to 0.9.8 doesn't provide them.
+        return __get_builtin_constructor(name)(string)
+
+
+try:
+    import _hashlib
+    new = __hash_new
+    __get_hash = __get_openssl_constructor
+except ImportError:
+    new = __py_new
+    __get_hash = __get_builtin_constructor
+
+for __func_name in __always_supported:
+    # try them all, some may not work due to the OpenSSL
+    # version not supporting that algorithm.
+    try:
+        globals()[__func_name] = __get_hash(__func_name)
+    except ValueError:
+        import logging
+        logging.exception('code for hash %s was not found.', __func_name)
+
+# Cleanup locals()
+del __always_supported, __func_name, __get_hash
+del __py_new, __hash_new, __get_openssl_constructor
diff --git a/src/setup.py b/src/setup.py
--- a/src/setup.py
+++ b/src/setup.py
@@ -3,8 +3,10 @@
 __revision__ = "$Id$"
 import sys
 import os
+import re
 
-from distutils2.core import setup
+from distutils2.core import setup, Extension
+from distutils2.compiler.ccompiler import new_compiler
 from distutils2.command.sdist import sdist
 from distutils2.command.install import install
 from distutils2 import __version__ as VERSION
@@ -31,6 +33,7 @@
 
 DEV_SUFFIX = '.dev%d' % get_tip_revision('..')
 
+
 class install_hg(install):
 
     user_options = install.user_options + [
@@ -62,10 +65,143 @@
             self.distribution.metadata.version += DEV_SUFFIX
         sdist.run(self)
 
+
+# additional paths to check, set from the command line
+SSL_INCDIR=''   # --openssl-incdir=
+SSL_LIBDIR=''   # --openssl-libdir=
+SSL_DIR=''      # --openssl-prefix=
+
+def add_dir_to_list(dirlist, dir):
+    """Add the directory 'dir' to the list 'dirlist' (at the front) if
+    'dir' actually exists and is a directory.  If 'dir' is already in
+    'dirlist' it is moved to the front."""
+    if dir is not None and os.path.isdir(dir) and dir not in dirlist:
+        if dir in dirlist:
+            dirlist.remove(dir)
+        dirlist.insert(0, dir)
+
+
+def prepare_hashlib_extensions():
+    """Decide which C extensions to build and create the appropriate
+    Extension objects to build them.  Returns a list of Extensions."""
+
+    # this CCompiler object is only used to locate include files
+    compiler = new_compiler()
+
+    # Ensure that these paths are always checked
+    if os.name == 'posix':
+        add_dir_to_list(compiler.library_dirs, '/usr/local/lib')
+        add_dir_to_list(compiler.include_dirs, '/usr/local/include')
+
+        add_dir_to_list(compiler.library_dirs, '/usr/local/ssl/lib')
+        add_dir_to_list(compiler.include_dirs, '/usr/local/ssl/include')
+
+        add_dir_to_list(compiler.library_dirs, '/usr/contrib/ssl/lib')
+        add_dir_to_list(compiler.include_dirs, '/usr/contrib/ssl/include')
+
+        add_dir_to_list(compiler.library_dirs, '/usr/lib')
+        add_dir_to_list(compiler.include_dirs, '/usr/include')
+
+    # look in command line supplied paths
+    if SSL_LIBDIR:
+        add_dir_to_list(compiler.library_dirs, SSL_LIBDIR)
+    if SSL_INCDIR:
+        add_dir_to_list(compiler.include_dirs, SSL_INCDIR)
+    if SSL_DIR:
+        if os.name == 'nt':
+            add_dir_to_list(compiler.library_dirs, os.path.join(SSL_DIR, 'out32dll'))
+            # prefer the static library
+            add_dir_to_list(compiler.library_dirs, os.path.join(SSL_DIR, 'out32'))
+        else:
+            add_dir_to_list(compiler.library_dirs, os.path.join(SSL_DIR, 'lib'))
+        add_dir_to_list(compiler.include_dirs, os.path.join(SSL_DIR, 'include'))
+
+    osNameLibsMap = {
+        'posix':  ['ssl', 'crypto'],
+        'nt':     ['libeay32',  'gdi32', 'advapi32', 'user32'],
+    }
+
+    if not osNameLibsMap.has_key(os.name):
+        print "unknown OS, please update setup.py"
+        sys.exit(1)
+
+    exts = []
+
+    ssl_inc_dirs = []
+    ssl_incs = []
+    for inc_dir in compiler.include_dirs:
+        f = os.path.join(inc_dir, 'openssl', 'ssl.h')
+        if os.path.exists(f):
+            ssl_incs.append(f)
+            ssl_inc_dirs.append(inc_dir)
+
+    ssl_lib = compiler.find_library_file(compiler.library_dirs, osNameLibsMap[os.name][0])
+
+    # find out which version of OpenSSL we have
+    openssl_ver = 0
+    openssl_ver_re = re.compile(
+        '^\s*#\s*define\s+OPENSSL_VERSION_NUMBER\s+(0x[0-9a-fA-F]+)' )
+    ssl_inc_dir = ''
+    for ssl_inc_dir in ssl_inc_dirs:
+        name = os.path.join(ssl_inc_dir, 'openssl', 'opensslv.h')
+        if os.path.isfile(name):
+            try:
+                incfile = open(name, 'r')
+                for line in incfile:
+                    m = openssl_ver_re.match(line)
+                    if m:
+                        openssl_ver = eval(m.group(1))
+                        break
+            except IOError:
+                pass
+
+        # first version found is what we'll use
+        if openssl_ver:
+            break
+
+    if (ssl_inc_dir and
+        ssl_lib is not None and
+        openssl_ver >= 0x00907000):
+
+        print 'Using OpenSSL version 0x%08x from' % openssl_ver
+        print ' Headers:\t', ssl_inc_dir
+        print ' Library:\t', ssl_lib
+
+        # The _hashlib module wraps optimized implementations
+        # of hash functions from the OpenSSL library.
+        exts.append( Extension('_hashlib', ['_hashopenssl.c'],
+                               include_dirs = [ ssl_inc_dir ],
+                               library_dirs = [ os.path.dirname(ssl_lib) ],
+                               libraries = osNameLibsMap[os.name]) )
+    else:
+        exts.append( Extension('_sha', ['shamodule.c']) )
+        exts.append( Extension('_md5',
+                        sources = ['md5module.c', 'md5.c'],
+                        depends = ['md5.h']) )
+
+    if (not ssl_lib or openssl_ver < 0x00908000):
+        # OpenSSL doesn't do these until 0.9.8 so we'll bring our own
+        exts.append( Extension('_sha256', ['sha256module.c']) )
+        exts.append( Extension('_sha512', ['sha512module.c']) )
+
+    def prependModules(filename):
+        return os.path.join('Modules', filename)
+
+    # all the C code is in the Modules subdirectory, prepend the path
+    for ext in exts:
+        ext.sources = [ prependModules(fn) for fn in ext.sources ]
+        if hasattr(ext, 'depends') and ext.depends is not None:
+            ext.depends = [ prependModules(fn) for fn in ext.depends ]
+
+    return exts
+
 setup_kwargs = {}
 if sys.version < '2.6':
     setup_kwargs['scripts'] = ['distutils2/mkpkg.py']
 
+if sys.version < '2.5':
+    setup_kwargs['ext_modules'] = prepare_hashlib_extensions()
+
 _CLASSIFIERS = """\
 Development Status :: 3 - Alpha
 Intended Audience :: Developers
@@ -97,6 +233,4 @@
                     ('Repository', 'http://hg.python.org/distutils2'),
                     ('Bug tracker', 'http://bugs.python.org')],
        **setup_kwargs
-       )
-
-
+          )

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: Support of prefer_source and
	prefer_final.
Message-ID: <E1OVLoW-0000lA-Kc@dinsdale.python.org>

tarek.ziade pushed 5e8672536566 to distutils2:

http://hg.python.org/distutils2/rev/5e8672536566
changeset:   336:5e8672536566
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Thu Jul 01 19:59:01 2010 +0200
summary:     Support of prefer_source and prefer_final.
files:       src/distutils2/pypi/simple.py

diff --git a/src/distutils2/pypi/simple.py b/src/distutils2/pypi/simple.py
--- a/src/distutils2/pypi/simple.py
+++ b/src/distutils2/pypi/simple.py
@@ -63,7 +63,8 @@
     """
 
     def __init__(self, index_url=PYPI_DEFAULT_INDEX_URL, hosts=DEFAULT_HOSTS,
-                 follow_externals=False, mirrors_url=PYPI_DEFAULT_MIRROR_URL,
+                 follow_externals=False, prefer_source=True,
+                 prefer_final=False, mirrors_url=PYPI_DEFAULT_MIRROR_URL,
                  mirrors=None, timeout=SOCKET_TIMEOUT):
         """Class constructor.
 
@@ -75,12 +76,19 @@
                       hosts.
         :param follow_externals: tell if following external links is needed or
                                  not. Default is False.
+        :param prefer_source: if there is binary and source distributions, the
+                              source prevails.
+        :param prefer_final: if the version is not mentioned, and the last
+                             version is not a "final" one (alpha, beta, etc.),
+                             pick up the last final version.
         :param mirrors_url: the url to look on for DNS records giving mirror
                             adresses.
         :param mirrors: a list of mirrors to check out if problems
                              occurs while working with the one given in "url"
         :param timeout: time in seconds to consider a url has timeouted.
         """
+        self.follow_externals = follow_externals
+
         if not index_url.endswith("/"):
             index_url += "/"
         self._index_urls = [index_url]
@@ -90,7 +98,8 @@
         self._index_urls.extend(mirrors)
         self._current_index_url = 0
         self._timeout = timeout
-        self.follow_externals = follow_externals
+        self._prefer_source = prefer_source
+        self._prefer_final = prefer_final
 
         # create a regexp to match all given hosts
         self._allowed_hosts = re.compile('|'.join(map(translate, hosts))).match
@@ -101,19 +110,7 @@
         self._processed_urls = []
         self._distributions = {}
 
-    def get(self, requirements):
-        """Browse the PyPI index to find distributions that fullfil the
-        given requirements, and return the most recent one.
-        """
-        predicate = self._get_version_predicate(requirements)
-        dists = self.find(predicate)
-
-        if len(dists) == 0:
-            raise DistributionNotFound(requirements)
-
-        return dists.get_last(predicate)
-
-    def find(self, requirements):
+    def find(self, requirements, prefer_source=None, prefer_final=None):
         """Browse the PyPI to find distributions that fullfil the given
         requirements.
 
@@ -121,8 +118,17 @@
                              version specifiers, as described in PEP345.
         :type requirements:  You can pass either a version.VersionPredicate
                              or a string.
+        :param prefer_source: if there is binary and source distributions, the
+                              source prevails.
+        :param prefer_final: if the version is not mentioned, and the last
+                             version is not a "final" one (alpha, beta, etc.),
+                             pick up the last final version.
         """
         requirements = self._get_version_predicate(requirements)
+        if prefer_source is None:
+            prefer_source = self._prefer_source
+        if prefer_final is None:
+            prefer_final = self._prefer_final
 
         # process the index for this project
         self._process_pypi_page(requirements.name)
@@ -130,12 +136,28 @@
         # filter with requirements and return the results
         if requirements.name in self._distributions:
             dists = self._distributions[requirements.name].filter(requirements)
+            dists.sort(prefer_source=prefer_source, prefer_final=prefer_final)
         else:
             dists = []
 
         return dists
 
-    def download(self, requirements, temp_path=None):
+    def get(self, requirements, *args, **kwargs):
+        """Browse the PyPI index to find distributions that fullfil the
+        given requirements, and return the most recent one.
+
+        You can specify prefer_final and prefer_source arguments here.
+        If not, the default one will be used.
+        """
+        predicate = self._get_version_predicate(requirements)
+        dists = self.find(predicate, *args, **kwargs)
+
+        if len(dists) == 0:
+            raise DistributionNotFound(requirements)
+
+        return dists.get_last(predicate)
+
+    def download(self, requirements, temp_path=None, *args, **kwargs):
         """Download the distribution, using the requirements.
 
         If more than one distribution match the requirements, use the last
@@ -146,8 +168,12 @@
         Returns the complete absolute path to the downloaded archive.
 
         :param requirements: The same as the find attribute of `find`.
+
+        You can specify prefer_final and prefer_source arguments here.
+        If not, the default one will be used.
         """
-        return self.get(requirements).download(path=temp_path)
+        return self.get(requirements, *args, **kwargs)\
+                   .download(path=temp_path)
 
     def _get_version_predicate(self, requirements):
         """Return a VersionPredicate object, from a string or an already

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: New behavior for follow external.
Message-ID: <E1OVLoW-0000kO-BF@dinsdale.python.org>

tarek.ziade pushed f28c41acac7c to distutils2:

http://hg.python.org/distutils2/rev/f28c41acac7c
changeset:   324:f28c41acac7c
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Tue Jun 29 11:38:55 2010 +0200
summary:     New behavior for follow external.
files:       src/distutils2/pypi/dist.py, src/distutils2/pypi/simple.py, src/distutils2/tests/test_pypi_dist.py, src/distutils2/tests/test_pypi_simple.py

diff --git a/src/distutils2/pypi/dist.py b/src/distutils2/pypi/dist.py
--- a/src/distutils2/pypi/dist.py
+++ b/src/distutils2/pypi/dist.py
@@ -30,11 +30,13 @@
     """
 
     @classmethod
-    def from_url(cls, url, probable_dist_name=None):
+    def from_url(cls, url, probable_dist_name=None, is_external=True):
         """Build a Distribution from a url archive (egg or zip or tgz).
 
         :param url: complete url of the distribution
         :param probable_dist_name: A probable name of the distribution. 
+        :param is_external: Tell if the url commes from an index or from 
+                            an external URL.
         """
         # if the url contains a md5 hash, get it.
         md5_hash = None
@@ -55,36 +57,79 @@
 
         name, version = split_archive_name(archive_name)
         if extension_matched is True:
-            return PyPIDistribution(name, version, url=url, md5_hash=md5_hash)
+            return PyPIDistribution(name, version, url=url, md5_hash=md5_hash, 
+                                    is_external=is_external)
 
-    def __init__(self, name, version, type=None, url=None, md5_hash=None):
-        """Create a new instance of PyPIDistribution
+    def __init__(self, name, version, type=None, url=None, md5_hash=None, 
+                 is_external=True):
+        """Create a new instance of PyPIDistribution.
+
+        :param name: the name of the distribution
+        :param version: the version of the distribution
+        :param type: the type of the dist (eg. source, bin-*, etc.)
+        :param url: URL where we found this distribution
+        :param md5_hash: the MD5 hash of the distribution. That's used to 
+                         check that the download is good enough.
+        :param is_external: we need to know if the provided url comes from an 
+                            index browsing, or from an external resource.
+
         """
         self.name = name
         self.version = version
-        self.url = url
         self.type = type
-        self.md5_hash = md5_hash
-        # set the downloaded path to Null by default. The goal here
+        # set the downloaded path to None by default. The goal here
         # is to not download distributions multiple times
         self.location = None
+        # We store urls in dict, because we need to have a bit more informations
+        # than the simple URL. It will be used later to find the good url to
+        # use.
+        # We have two _url* attributes: _url and _urls. _urls contains a list of 
+        # dict for the different urls, and _url contains the choosen url, in 
+        # order to dont make the selection process multiple times.
+        self._urls = []
+        self._url = None
+        self.add_url(url, md5_hash, is_external)
+    
+    def add_url(self, url, md5=None, is_external=True):
+        """Add a new url to the list of urls"""
+        self._urls.append({
+            'url': url,
+            'md5': md5,
+            'is_external': is_external,
+        })
+        # reset the url selection process
+        self._url = None
 
-    def download(self, url=None, path=None):
+    @property
+    def url(self):
+        """Pick up the right url for the list of urls in self.urls"""
+        # We return internal urls over externals.
+        # If there is more than one internal or external, return the first
+        # one.
+        if self._url is None:
+            if len(self._urls) > 1:
+                internals_urls = [u for u in self._urls \
+                                  if u['is_external'] == False]
+                if len(internals_urls) >= 1: 
+                    self._url = internals_urls[0]
+            if self._url is None:
+                self._url = self._urls[0]
+        return self._url
+
+    def download(self, path=None):
         """Download the distribution to a path, and return it.
 
         If the path is given in path, use this, otherwise, generates a new one
         """
-        if url is not None:
-            self.url = url
-
         if path is None:
             path = tempfile.mkdtemp()
 
         # if we do not have downloaded it yet, do it.
         if self.location is None:
-            archive_name = urlparse.urlparse(self.url)[2].split('/')[-1]
-            filename, headers = urllib.urlretrieve(
-                self.url, path + "/" + archive_name)
+            url = self.url['url']
+            archive_name = urlparse.urlparse(url)[2].split('/')[-1]
+            filename, headers = urllib.urlretrieve(url, 
+                                                   path + "/" + archive_name)
             self.location = filename
             self._check_md5(filename)
         return self.location
@@ -92,13 +137,14 @@
     def _check_md5(self, filename):
         """Check that the md5 checksum of the given file matches the one in
         self._md5_hash."""
-        if self.md5_hash is not None:
+        md5_hash = self.url['md5']
+        if md5_hash is not None:
             f = open(filename)
             hash = md5()
             hash.update(f.read())
-            if hash.hexdigest() != self.md5_hash:
+            if hash.hexdigest() != md5_hash:
                 raise MD5HashDoesNotMatch("%s instead of %s"
-                    % (hash.hexdigest(), self.md5_hash))
+                    % (hash.hexdigest(), md5_hash))
 
     def __repr__(self):
         return "<%s %s (%s)>" \
@@ -141,6 +187,11 @@
 
     Contains methods and facilities to sort and filter distributions.
     """
+    def __init__(self, list=[]):
+        # To disable the ability to pass lists on instanciation
+        super(PyPIDistributions, self).__init__()
+        for item in list:
+            self.append(item)
 
     def filter(self, predicate):
         """Filter the distributions and return a subset of distributions that
@@ -158,6 +209,20 @@
         distributions.sort()
         return distributions[-1]
 
+    def append(self, o):
+        """Append a new distribution to the list.
+        
+        If a distribution with the same name and version exists, just grab the
+        URL informations and add a new new url for the existing one.
+        """
+        similar_dists = [d for d in self if d.name == o.name and
+                         d.version == o.version]
+        if len(similar_dists) > 0:
+            dist = similar_dists[0]
+            dist.add_url(**o.url)
+        else:
+            super(PyPIDistributions, self).append(o)
+        
 
 def split_archive_name(archive_name, probable_name=None):
     """Split an archive name into two parts: name and version.
diff --git a/src/distutils2/pypi/simple.py b/src/distutils2/pypi/simple.py
--- a/src/distutils2/pypi/simple.py
+++ b/src/distutils2/pypi/simple.py
@@ -21,7 +21,7 @@
 
 # -- Constants -----------------------------------------------
 PYPI_DEFAULT_INDEX_URL = "http://pypi.python.org/simple/"
-DEFAULT_HOSTS = ("*.python.org",)
+DEFAULT_HOSTS = ("*",)  
 SOCKET_TIMEOUT = 15
 USER_AGENT = "Python-urllib/%s distutils2/%s" % (
     sys.version[:3], __distutils2_version__)
@@ -61,22 +61,26 @@
     """Provides useful tools to request the Python Package Index simple API
     """
 
-    def __init__(self, url=PYPI_DEFAULT_INDEX_URL, hosts=DEFAULT_HOSTS,
-        mirrors=[], timeout=SOCKET_TIMEOUT):
+    def __init__(self, index_url=PYPI_DEFAULT_INDEX_URL, hosts=DEFAULT_HOSTS,
+                 mirrors=[], follow_externals=False, timeout=SOCKET_TIMEOUT):
         """Class constructor.
 
         :param index_url: the url of the simple index to search on.
-        :param hosts: a list of hosts allowed to be considered as internals (to
-        differenciate with external hosts)
-        :param follow_externals: tell if follow external links is needed or not
-        Default is False
+        :param follow_externals: tell if following external links is needed or 
+                                 not. Default is False.
+        :param hosts: a list of hosts allowed to be processed while using
+                      follow_externals=True. Default behavior is to follow all 
+                      hosts.
         :param mirrors: a list of mirrors to check out if problems occurs while
-        working with the one given in "url"
+                        working with the one given in "url"
+        :param timeout: time in seconds to consider a url has timeouted.
         """
-        self._index_urls = [url]
+        self._index_urls = [index_url]
         self._index_urls.extend(mirrors)
         self._current_index_url = 0
         self._timeout = timeout
+        self.follow_externals = follow_externals
+
         # create a regexp to match all given hosts
         self._allowed_hosts = re.compile('|'.join(map(translate, hosts))).match
 
@@ -150,27 +154,38 @@
         """Switch to the next mirror (eg. point self.index_url to the next
         url.
         """
+        # Internally, iter over the _index_url iterable, if we have read all
+        # of the available indexes, raise an exception.
         if self._current_index_url < len(self._index_urls):
             self._current_index_url = self._current_index_url + 1
         else:
             raise UnableToDownload("All mirrors fails")
     
     def _is_browsable(self, url):
-        """Tell if the given URL needs to be browsed or not, according to the
-        object internal attributes.
+        """Tell if the given URL can be browsed or not.
 
         It uses the follow_externals and the hosts list to tell if the given
         url is browsable or not.
         """
-        if self._allowed_hosts(urlparse.urlparse(url)[1]):  # 1 is netloc
+        # if _index_url is contained in the given URL, we are browsing the
+        # index, and it's always "browsable".
+        # We asume here that if the url starts with "." or "..", it's browsable
+        # too. This is useful as the simple index make heavy use of relative
+        # URLS.
+        if self.index_url in url:
             return True
-        else:
-            return False
+        elif self.follow_externals is True:
+            if self._allowed_hosts(urlparse.urlparse(url)[1]):  # 1 is netloc
+                return True
+            else:
+                return False
+        return False
 
     def _is_distribution(self, link):
         """Tell if the given URL matches to a distribution name or not.
         """
         #XXX find a better way to check that links are distributions
+        # Using a regexp ?
         for ext in EXTENSIONS:
             if ext in link:
                 return True
@@ -189,15 +204,18 @@
         self._distributions[dist.name].append(dist)
         return self._distributions[dist.name]
 
-    def _process_url(self, url, project_name=None,follow_links=True, 
-        filter_rel=False):
+    def _process_url(self, url, project_name=None, follow_links=True):
         """Process an url and search for distributions packages.
 
-        :param url: the url to analyse
+        For each URL found, if it's a download, creates a PyPIdistribution 
+        object. If it's a homepage and we can follow links, process it too.
+
+        :param url: the url to process
         :param project_name: the project name we are searching for.
         :param follow_links: Do not want to follow links more than from one
-        level. This parameter tells if we want to follow the links we find (eg.
-        run recursively this method on it)
+                             level. This parameter tells if we want to follow 
+                             the links we find (eg. run recursively this 
+                             method on it)
         """
         f = self._open_url(url)
         base_url = f.url
@@ -207,10 +225,10 @@
             for link, is_download in link_matcher(f.read(), base_url):
                 if link not in self._processed_urls:
                     if self._is_distribution(link) or is_download:
+                        self._processed_urls.append(link)
                         # it's a distribution, so create a dist object
-                        self._processed_urls.append(link)
                         self._register_dist(PyPIDistribution.from_url(link,
-                            project_name))
+                            project_name, is_external=not self.index_url in url))
                     else:
                         if self._is_browsable(link) and follow_links:
                             self._process_url(link, project_name,
@@ -220,29 +238,36 @@
         """Returns the right link matcher function of the given url
         """
         if self.index_url in url:
-            return self._simple_link_finder
+            return self._simple_link_matcher
         else:
-            return self._default_link_finder
+            return self._default_link_matcher
 
-    def _simple_link_finder(self, content, base_url):
+    def _simple_link_matcher(self, content, base_url):
         """Yield all links with a rel="download" or rel="homepage".
 
-        This matches the simple index requirements for matching links
+        This matches the simple index requirements for matching links.
+        If follow_externals is set to False, dont yeld the external
+        urls.
         """
         for match in REL.finditer(content):
             tag, rel = match.groups()
             rels = map(str.strip, rel.lower().split(','))
             if 'homepage' in rels or 'download' in rels:
                 for match in HREF.finditer(tag):
-                    yield (urlparse.urljoin(base_url, 
-                        self._htmldecode(match.group(1))), "downloads" in rels)
+                    url = urlparse.urljoin(base_url,
+                                           self._htmldecode(match.group(1)))
+                    if 'download' in rels or self._is_browsable(url):
+                        # yield a list of (url, is_download)
+                        yield (urlparse.urljoin(base_url, url), 
+                               'download' in rels)
 
-    def _default_link_finder(self, content, base_url):
+    def _default_link_matcher(self, content, base_url):
         """Yield all links found on the page.
         """
         for match in HREF.finditer(content):
-            yield (urlparse.urljoin(base_url, 
-                self._htmldecode(match.group(1))), False)
+            url = urlparse.urljoin(base_url, self._htmldecode(match.group(1)))
+            if self._is_browsable(url):
+                yield (url, False)
 
     def _process_pypi_page(self, name):
         """Find and process a PyPI page for the given project name.
@@ -262,7 +287,8 @@
 
     @socket_timeout()
     def _open_url(self, url):
-        """Open a urllib2 request, handling HTTP authentication
+        """Open a urllib2 request, handling HTTP authentication.
+
         """
         try:
             scheme, netloc, path, params, query, frag = urlparse.urlparse(url)
diff --git a/src/distutils2/tests/test_pypi_dist.py b/src/distutils2/tests/test_pypi_dist.py
--- a/src/distutils2/tests/test_pypi_dist.py
+++ b/src/distutils2/tests/test_pypi_dist.py
@@ -37,20 +37,49 @@
             'foobar-1.1b2.tar.gz#md5=123123123123123': {
                 'name': 'foobar',
                 'version': '1.1b2',
-                'url': 'http://test.tld/foobar-1.1b2.tar.gz',  # without hash
-                'md5_hash': '123123123123123',
+                'url': {
+                    'url': 'http://test.tld/foobar-1.1b2.tar.gz',  # no hash
+                    'md5': '123123123123123',
+                }
             },
             'foobar-1.1-rc2.tar.gz': {  # use suggested name
                 'name': 'foobar',
                 'version': '1.1c2',
-                'url': 'http://test.tld/foobar-1.1-rc2.tar.gz',
+                'url': {
+                    'url': 'http://test.tld/foobar-1.1-rc2.tar.gz',
+                }
             }
         }
 
         for url, attributes in url_list.items():
             dist = Dist.from_url("http://test.tld/" + url)
             for attribute, value in attributes.items():
-                self.assertEqual(getattr(dist, attribute), value)
+                if isinstance(value, dict):
+                    mylist = getattr(dist, attribute)
+                    for val in value.keys():
+                        self.assertEqual(value[val], mylist[val])
+                else:
+                    self.assertEqual(getattr(dist, attribute), value)
+
+    def test_get_url(self):
+        """Test that the url property works well"""
+
+        d = Dist("foobar", "1.1", url="test_url")
+        self.assertDictEqual(d.url, {
+            "url": "test_url",
+            "is_external": True,
+            "md5": None,
+        })
+
+        # add a new url
+        d.add_url(url="internal_url", is_external=False)
+        self.assertEqual(d._url, None)
+        self.assertDictEqual(d.url, {
+            "url": "internal_url",
+            "is_external": False,
+            "md5": None,
+        })
+        self.assertEqual(2, len(d._urls))
 
     def test_comparaison(self):
         """Test that we can compare PyPIDistributions"""
@@ -127,6 +156,26 @@
         self.assertIn(dists[0], filtered)
         self.assertIn(dists[1], filtered)
 
+    def test_append(self):
+        """Test the append method of PyPIDistributions"""
+        # When adding a new item to the list, the behavior is to test if
+        # a distribution with the same name and version number already exists,
+        # and if so, to add url informations to the existing PyPIDistribution
+        # object.
+        # If no object matches, just add "normally" the object to the list.
+
+        dists = Dists([
+            Dist("FooBar", "1.1", url="external_url"),
+        ])
+        self.assertEqual(1, len(dists))
+        dists.append(Dist("FooBar", "1.1", url="internal_url",
+                     is_external=False))
+        self.assertEqual(1, len(dists))
+        self.assertEqual(2, len(dists[0]._urls))
+
+        dists.append(Dist("Foobar", "1.1.1"))
+        self.assertEqual(2, len(dists))
+
 
 def test_suite():
     suite = unittest.TestSuite()
diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -83,29 +83,6 @@
                     'http://www.famfamfam.com/">')
             index.process_index(url, page)
 
-    def test_browsable_urls(self):
-        """Test we can give PyPI index a list of allowed hosts
-
-        """
-        index = simple.SimpleIndex(hosts=["pypi.python.org", "test.org"])
-        good_urls = (
-            "http://pypi.python.org/foo/bar",
-            "http://pypi.python.org/simple/foobar",
-            "http://test.org",
-            "http://test.org/",
-            "http://test.org/simple/",
-        )
-        bad_urls = (
-            "http://python.org",
-            "http://test.tld",
-        )
-
-        for url in good_urls:
-            self.assertTrue(index._is_browsable(url))
-
-        for url in bad_urls:
-            self.assertFalse(index._is_browsable(url))
-
     @use_pypi_server("test_found_links")
     def test_found_links(self, server):
         """Browse the index, asking for a specified distribution version
@@ -128,34 +105,60 @@
         self.assertEqual(last_distribution.version, '2.0.1')
 
     def test_is_browsable(self):
+        index = simple.SimpleIndex(follow_externals=False)
+        self.assertTrue(index._is_browsable(index.index_url + "test"))
+
+        # Now, when following externals, we can have a list of hosts to trust.
+        # and don't follow other external links than the one described here.
+        index = simple.SimpleIndex(hosts=["pypi.python.org", "test.org"],
+                                   follow_externals=True)
+        good_urls = (
+            "http://pypi.python.org/foo/bar",
+            "http://pypi.python.org/simple/foobar",
+            "http://test.org",
+            "http://test.org/",
+            "http://test.org/simple/",
+        )
+        bad_urls = (
+            "http://python.org",
+            "http://test.tld",
+        )
+
+        for url in good_urls:
+            self.assertTrue(index._is_browsable(url))
+
+        for url in bad_urls:
+            self.assertFalse(index._is_browsable(url))
+
         # allow all hosts
-        index = simple.SimpleIndex(hosts=("*",))
+        index = simple.SimpleIndex(follow_externals=True, hosts=("*",))
         self.assertTrue(index._is_browsable("http://an-external.link/path"))
         self.assertTrue(index._is_browsable("pypi.test.tld/a/path"))
 
         # specify a list of hosts we want to allow
-        index = simple.SimpleIndex(hosts=("*.test.tld",))
+        index = simple.SimpleIndex(follow_externals=True, 
+                                   hosts=("*.test.tld",))
         self.assertFalse(index._is_browsable("http://an-external.link/path"))
         self.assertTrue(index._is_browsable("http://pypi.test.tld/a/path"))
 
     @use_pypi_server("with_externals")
-    def test_process_external_pages(self, server):
+    def test_restrict_hosts(self, server):
         """Include external pages
         """
         # Try to request the package index, wich contains links to "externals"
         # resources. They have to  be scanned too.
-        index = self._get_simple_index(server, hosts=("*",))
+        index = self._get_simple_index(server, follow_externals=True)
         index.get("foobar")
         self.assertIn(server.full_address + "/external/external.html",
             index._processed_urls)
 
     @use_pypi_server("with_real_externals")
-    def test_disable_external_pages(self, server):
-        """Exclude external pages if disabled
+    def test_restrict_hosts(self, server):
+        """Only use a list of allowed hosts is possible
         """
         # Test that telling the simple pyPI client to not retrieve external
         # works
-        index = self._get_simple_index(server, hosts=(server.full_address,))
+        index = self._get_simple_index(server, follow_externals=False)
         index.get("foobar")
         self.assertNotIn(server.full_address + "/external/external.html",
             index._processed_urls)
@@ -186,16 +189,14 @@
         index_url = server.full_address + '/simple/'
 
         # scan a test index
-        index = simple.SimpleIndex(index_url)
+        index = simple.SimpleIndex(index_url, follow_externals=True)
         dists = index.find("foobar")
         server.stop()
 
-        # the distribution has been found
-        self.assertTrue(dists is not [])
         # we have only one link, because links are compared without md5
         self.assertEqual(len(dists), 1)
         # the link should be from the index
-        self.assertEqual('12345678901234567', dists[0].md5_hash)
+        self.assertEqual('12345678901234567', dists[0].url['md5'])
 
     @use_pypi_server(static_filesystem_paths=["with_norel_links"],
         static_uri_paths=["simple", "external"])
@@ -205,7 +206,7 @@
         to not be processed by the package index, while processing "pages".
         """
         # process the pages
-        index = self._get_simple_index(server)
+        index = self._get_simple_index(server, follow_externals=True)
         index.find("foobar")
         # now it should have processed only pages with links rel="download" 
         # and rel="homepage"
@@ -236,9 +237,42 @@
             
             # this should not raise a timeout
             self.assertEqual(4, len(index.find("foo")))
-        except Exception, e:
+        finally:
             mirror.stop()
-            raise e
+    
+    def test_simple_link_matcher(self):
+        """Test that the simple link matcher yields the right links"""
+        index = simple.SimpleIndex(follow_externals=False)
+        
+        # Here, we define:
+        #   1. one link that must be followed, cause it's a download one
+        #   2. one link that must *not* be followed, cause the is_browsable
+        #      returns false for it.
+        #   3. one link that must be followed cause it's a homepage that is
+        #      browsable
+        self.assertTrue(index._is_browsable("%stest" % index.index_url))
+        self.assertFalse(index._is_browsable("http://dl-link2"))
+        content = """
+        <a href="http://dl-link1" rel="download">download_link1</a>
+        <a href="http://dl-link2" rel="homepage">homepage_link1</a>
+        <a href="%stest" rel="homepage">homepage_link2</a>
+        """ % index.index_url
+
+        # Test that the simple link matcher yield the good links.
+        generator = index._simple_link_matcher(content, index.index_url)
+        self.assertEqual(('http://dl-link1', True), generator.next())
+        self.assertEqual(('%stest' % index.index_url, False), 
+                         generator.next())
+        self.assertRaises(StopIteration, generator.next)
+
+        # Follow the external links is possible
+        index.follow_externals = True
+        generator = index._simple_link_matcher(content, index.index_url)
+        self.assertEqual(('http://dl-link1', True), generator.next())
+        self.assertEqual(('http://dl-link2', False), generator.next())
+        self.assertEqual(('%stest' % index.index_url, False), 
+                         generator.next())
+        self.assertRaises(StopIteration, generator.next)
 
 def test_suite():
     return unittest2.makeSuite(PyPISimpleTestCase)

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: remove distribute related python
	modules
Message-ID: <E1OVLoW-0000k3-7S@dinsdale.python.org>

tarek.ziade pushed 06f3ae643475 to distutils2:

http://hg.python.org/distutils2/rev/06f3ae643475
changeset:   319:06f3ae643475
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Tue Jun 22 16:13:18 2010 +0200
summary:     remove distribute related python modules
files:       src/distutils2/pypi/oldsimple.py, src/distutils2/pypi/pkg_resources.py

diff --git a/src/distutils2/pypi/oldsimple.py b/src/distutils2/pypi/oldsimple.py
deleted file mode 100644
--- a/src/distutils2/pypi/oldsimple.py
+++ /dev/null
@@ -1,806 +0,0 @@
-"""distutils2.pypi.package_index
-
-Implement a simple way to use the PyPI API's, by using known URIs from 
-PyPI. Works with the API defined on 
-http://peak.telecommunity.com/DevCenter/EasyInstall#package-index-api , 
-and available at http://pypi.python.org/simple/ .
-
-"""
-# XXX Make a explaination document on how this API is made.
-
-import sys
-import os.path
-import re
-import urlparse
-import urllib2
-import shutil
-import socket
-import cStringIO
-import httplib
-import logging
-import tempfile
-
-from distutils2.errors import DistutilsError
-from distutils2 import __version__ as __distutils2_version__
-
-try:
-    from hashlib import md5
-except ImportError:
-    from md5 import md5
-from fnmatch import translate
-
-# -- Constants ----------------------------------------------------------------
-
-__all__ = ['PackageIndex', 
-           'distros_for_url', 
-           'parse_bdist_wininst',
-           'interpret_distro_name',
-]
-
-_SOCKET_TIMEOUT = 15
-SOURCE_DIST = 1
-EGG_FRAGMENT = re.compile(r'^egg=([-A-Za-z0-9_.]+)$')
-HREF = re.compile("""href\\s*=\\s*['"]?([^'"> ]+)""", re.I)
-PYPI_MD5 = re.compile(
-    '<a href="([^"#]+)">([^<]+)</a>\n\s+\\(<a (?:title="MD5 hash"\n\s+)'
-    'href="[^?]+\?:action=show_md5&amp;digest=([0-9a-f]{32})">md5</a>\\)'
-)
-URL_SCHEME = re.compile('([-+.a-z0-9]{2,}):',re.I).match
-EXTENSIONS = ".tar.gz .tar.bz2 .tar .zip .tgz".split()
-
-# -- Dependencies from distribute ---------------------------------------------
-
-# XXX Remove these dependencies, refactor the package_index codebase to not 
-# depends anymore on distribute classes and internal workings.
-from pkg_resources import *
-
-# -- Base functions -----------------------------------------------------------
-
-def parse_bdist_wininst(name):
-    """Parse a bdist_win package name, and return it's package base name 
-    and version number if possible.
-
-    Return a tuple of (base,pyversion) for possible .exe names"""
-
-    lower = name.lower()
-    base, py_ver = None, None
-
-    if lower.endswith('.exe'):
-        if lower.endswith('.win32.exe'):
-            base = name[:-10]
-        elif lower.startswith('.win32-py',-16):
-            py_ver = name[-7:-4]
-            base = name[:-16]
-
-    return base,py_ver
-
-def egg_info_for_url(url):
-    """
-    """
-    scheme, server, path, parameters, query, fragment = urlparse.urlparse(url)
-    base = urllib2.unquote(path.split('/')[-1])
-    if '#' in base:
-        base, fragment = base.split('#',1)
-    return base,fragment
-
-def distros_for_url(url, metadata=None):
-    """Yield egg or source distribution objects that might be found at a URL"""
-    base, fragment = egg_info_for_url(url)
-    for dist in distros_for_location(url, base, metadata): 
-        yield dist
-    if fragment:
-        match = EGG_FRAGMENT.match(fragment)
-        if match:
-            for dist in interpret_distro_name(url, match.group(1), metadata, 
-                precedence = CHECKOUT_DIST):
-                yield dist
-
-def distros_for_location(location, basename, metadata=None):
-    """Yield egg or source distribution objects based on basename"""
-    if basename.endswith('.egg.zip'):
-        basename = basename[:-4]    # strip the .zip
-    if basename.endswith('.egg') and '-' in basename:
-        # only one, unambiguous interpretation
-        return [Distribution.from_location(location, basename, metadata)]
-
-    if basename.endswith('.exe'):
-        win_base, py_ver = parse_bdist_wininst(basename)
-        if win_base is not None:
-            return interpret_distro_name(location, win_base, metadata, py_ver, 
-                BINARY_DIST, "win32")
-
-    # Try source distro extensions (.zip, .tgz, etc.)
-    for ext in EXTENSIONS:
-        if basename.endswith(ext):
-            basename = basename[:-len(ext)]
-            return interpret_distro_name(location, basename, metadata)
-    return []  # no extension matched
-
-def distros_for_filename(filename, metadata=None):
-    """Yield possible egg or source distribution objects based on a 
-    filename.
-    """
-    return distros_for_location(normalize_path(filename), 
-        os.path.basename(filename), metadata)
-
-def interpret_distro_name(location, basename, metadata,
-    py_version=None, precedence=SOURCE_DIST, platform=None):
-    """Generate alternative interpretations of a source distro name
-
-    Note: if `location` is a filesystem filename, you should call
-    ``pkg_resources.normalize_path()`` on it before passing it to this
-    routine!
-    """
-    # Generate alternative interpretations of a source distro name
-    # Because some packages are ambiguous as to name/versions split
-    # e.g. "adns-python-1.1.0", "egenix-mx-commercial", etc.
-    # So, we generate each possible interepretation (e.g. "adns, python-1.1.0"
-    # "adns-python, 1.1.0", and "adns-python-1.1.0, no version").  In practice,
-    # the spurious interpretations should be ignored, because in the event
-    # there's also an "adns" package, the spurious "python-1.1.0" version will
-    # compare lower than any numeric version number, and is therefore unlikely
-    # to match a request for it.  It's still a potential problem, though, and
-    # in the long run PyPI and the distutils should go for "safe" names and
-    # versions in distribution archive names (sdist and bdist).
-
-    parts = basename.split('-')
-    if not py_version:
-        for i,p in enumerate(parts[2:]):
-            if len(p)==5 and p.startswith('py2.'):
-                return # It's a bdist_dumb, not an sdist -- bail out
-
-    for p in range(1,len(parts)+1):
-        yield Distribution(
-            location, metadata, '-'.join(parts[:p]), '-'.join(parts[p:]),
-            py_version=py_version, precedence = precedence,
-            platform = platform)
-
-REL = re.compile("""<([^>]*\srel\s*=\s*['"]?([^'">]+)[^>]*)>""", re.I)
-
-def find_external_links(url, page):
-    """Find rel="homepage" and rel="download" links in `page`, yielding URLs"""
-
-    for match in REL.finditer(page):
-        tag, rel = match.groups()
-        rels = map(str.strip, rel.lower().split(','))
-        if 'homepage' in rels or 'download' in rels:
-            for match in HREF.finditer(tag):
-                yield urlparse.urljoin(url, htmldecode(match.group(1)))
-
-    for tag in ("<th>Home Page", "<th>Download URL"):
-        pos = page.find(tag)
-        if pos!=-1:
-            match = HREF.search(page,pos)
-            if match:
-                yield urlparse.urljoin(url, htmldecode(match.group(1)))
-
-user_agent = "Python-urllib/%s distutils2/%s" % (
-    sys.version[:3], __distutils2_version__
-)
-
-class SimpleIndex(Environment):
-    """A distribution index that scans web pages for download URLs"""
-
-    def __init__(self, index_url="http://pypi.python.org/simple", hosts=('*',),
-        follow_externals=True, *args, **kw):
-        self._init_logging()
-        Environment.__init__(self,*args,**kw)
-        self.index_url = index_url + "/"[:not index_url.endswith('/')]
-        self.scanned_urls = {}
-        self.fetched_urls = {}
-        self.package_pages = {}
-        self.allows = re.compile('|'.join(map(translate,hosts))).match
-        self.to_scan = []
-        self.follow_externals = follow_externals
-
-    def _init_logging(self):
-        import sys
-#        logging.basicConfig(stream=sys.stdout, level=logging.DEBUG)
-
-    def process_url(self, url, retrieve=False):
-        """Evaluate a URL as a possible download, and maybe retrieve it"""
-
-        if url in self.scanned_urls and not retrieve:
-            return
-        self.scanned_urls[url] = True
-        if not URL_SCHEME(url):
-            self.process_filename(url)
-            return
-        else:
-            dists = list(distros_for_url(url))
-            if dists:
-                if not self.url_ok(url):
-                    return
-                self.debug("Found link: %s", url)
-
-        if dists or not retrieve or url in self.fetched_urls:
-            map(self.add, dists)
-            return  # don't need the actual page
-
-        if not self.url_ok(url):
-            self.fetched_urls[url] = True
-            return
-
-        self.info("Reading %s", url)
-        f = self.open_url(url, "Download error: %s -- Some packages may not be found!")
-        if f is None: return
-        self.fetched_urls[url] = self.fetched_urls[f.url] = True
-
-        if 'html' not in f.headers.get('content-type', '').lower():
-            f.close()   # not html, we can't process it
-            return
-
-        base = f.url     # handle redirects
-        page = f.read()
-        if sys.version_info >= (3,):
-            charset = f.headers.get_param('charset') or 'latin-1'
-            page = page.decode(charset, "ignore")
-        f.close()
-
-        if url.startswith(self.index_url):
-            if self.follow_externals:
-                for match in HREF.finditer(page):
-                    link = urlparse.urljoin(base, htmldecode(match.group(1)))
-                    self.process_url(link)
-        if url.startswith(self.index_url) and getattr(f,'code',None)!=404:
-            page = self.process_index(url, page)
-
-    def process_filename(self, fn, nested=False):
-        # process filenames or directories
-        if not os.path.exists(fn):
-            self.warn("Not found: %s", fn)
-            return
-
-        if os.path.isdir(fn) and not nested:
-            path = os.path.realpath(fn)
-            for item in os.listdir(path):
-                self.process_filename(os.path.join(path,item), True)
-
-        dists = distros_for_filename(fn)
-        if dists:
-            self.debug("Found: %s", fn)
-            map(self.add, dists)
-
-    def url_ok(self, url, fatal=False):
-        s = URL_SCHEME(url)
-        if (s and s.group(1).lower()=='file') or self.allows(urlparse.urlparse(url)[1]):
-            return True
-        msg = "\nLink to % s ***BLOCKED*** by --allow-hosts\n"
-        if fatal:
-            raise DistutilsError(msg % url)
-        else:
-            self.warn(msg, url)
-        return False
-
-    def scan_egg_links(self, search_path):
-        for item in search_path:
-            if os.path.isdir(item):
-                for entry in os.listdir(item):
-                    if entry.endswith('.egg-link'):
-                        self.scan_egg_link(item, entry)
-
-    def scan_egg_link(self, path, entry):
-        lines = filter(None, map(str.strip, open(os.path.join(path, entry))))
-        if len(lines)==2:
-            for dist in find_distributions(os.path.join(path, lines[0])):
-                dist.location = os.path.join(path, *lines)
-                dist.precedence = SOURCE_DIST
-                self.add(dist)
-
-    def process_index(self,url,page):
-        """Process the content of PyPI page. For each link, if it's a PyPI one,
-        browse it's page, and satinize the output.
-        """
-        def scan(link):
-            # Process a URL to see if it's for a package page
-            if link.startswith(self.index_url):
-                parts = map(
-                    urllib2.unquote, link[len(self.index_url):].split('/')
-                )
-                if len(parts)==2 and '#' not in parts[1]:
-                    # it's a package page, sanitize and index it
-                    pkg = safe_name(parts[0])
-                    ver = safe_version(parts[1])
-                    self.package_pages.setdefault(pkg.lower(),{})[link] = True
-                    return to_filename(pkg), to_filename(ver)
-            return None, None
-
-        # process an index page into the package-page index
-        for match in HREF.finditer(page):
-            try:
-                scan( urlparse.urljoin(url, htmldecode(match.group(1))) )
-            except ValueError:
-                pass
-
-        pkg, ver = scan(url)   # ensure this page is in the page index
-        if pkg:
-            # process individual package page
-            for new_url in find_external_links(url, page):
-                # Process the found URL
-                base, frag = egg_info_for_url(new_url)
-                if base.endswith('.py') and not frag:
-                    if ver:
-                        new_url+='#egg=%s-%s' % (pkg,ver)
-                    else:
-                        self.need_version_info(url)
-                self.scan_url(new_url)
-
-            return PYPI_MD5.sub(
-                lambda m: '<a href="%s#md5=%s">%s</a>' % m.group(1,3,2), page
-            )
-        else:
-            return ""   # no sense double-scanning non-package pages
-
-    def need_version_info(self, url):
-        self.scan_all(
-            "Page at %s links to .py file(s) without version info; an index "
-            "scan is required.", url
-        )
-
-    def scan_all(self, msg=None, *args):
-        if self.index_url not in self.fetched_urls:
-            if msg: self.warn(msg,*args)
-            self.info(
-                "Scanning index of all packages (this may take a while)"
-            )
-        self.scan_url(self.index_url)
-
-    def find_packages(self, requirement):
-        self.scan_url(self.index_url + requirement.unsafe_name+'/')
-
-        if not self.package_pages.get(requirement.key):
-            # Fall back to safe version of the name
-            self.scan_url(self.index_url + requirement.project_name+'/')
-
-        if not self.package_pages.get(requirement.key):
-            # We couldn't find the target package, so search the index page too
-            self.not_found_in_index(requirement)
-
-        for url in list(self.package_pages.get(requirement.key,())):
-            # scan each page that might be related to the desired package
-            self.scan_url(url)
-
-    def obtain(self, requirement, installer=None):
-        self.prescan(); self.find_packages(requirement)
-        for dist in self[requirement.key]:
-            if dist in requirement:
-                return dist
-            self.debug("%s does not match %s", requirement, dist)
-        return super(SimpleIndex, self).obtain(requirement,installer)
-
-    def check_md5(self, cs, info, filename, tfp):
-        if re.match('md5=[0-9a-f]{32}$', info):
-            self.debug("Validating md5 checksum for %s", filename)
-            if cs.hexdigest()<>info[4:]:
-                tfp.close()
-                os.unlink(filename)
-                raise DistutilsError(
-                    "MD5 validation failed for "+os.path.basename(filename)+
-                    "; possible download problem?")
-
-    def add_find_links(self, urls):
-        """Add `urls` to the list that will be prescanned for searches"""
-        for url in urls:
-            if (
-                self.to_scan is None        # if we have already "gone online"
-                or not URL_SCHEME(url)      # or it's a local file/directory
-                or url.startswith('file:')
-                or list(distros_for_url(url))   # or a direct package link
-            ):
-                # then go ahead and process it now
-                self.scan_url(url)
-            else:
-                # otherwise, defer retrieval till later
-                self.to_scan.append(url)
-
-    def prescan(self):
-        """Scan urls scheduled for prescanning (e.g. --find-links)"""
-        if self.to_scan:
-            map(self.scan_url, self.to_scan)
-        self.to_scan = None     # from now on, go ahead and process immediately
-
-    def not_found_in_index(self, requirement):
-        if self[requirement.key]:   # we've seen at least one distro
-            meth, msg = self.info, "Couldn't retrieve index page for %r"
-        else:   # no distros seen for this name, might be misspelled
-            meth, msg = (self.warn,
-                "Couldn't find index page for %r (maybe misspelled?)")
-        meth(msg, requirement.unsafe_name)
-        self.scan_all()
-
-    def download(self, spec, tmpdir=None):
-        """Locate and/or download `spec` to `tmpdir`, returning a local path
-
-        `spec` may be a ``Requirement`` object, or a string containing a URL,
-        an existing local filename, or a project/version requirement spec
-        (i.e. the string form of a ``Requirement`` object).  If it is the URL
-        of a .py file with an unambiguous ``#egg=name-version`` tag (i.e., one
-        that escapes ``-`` as ``_`` throughout), a trivial ``setup.py`` is
-        automatically created alongside the downloaded file.
-
-        If `spec` is a ``Requirement`` object or a string containing a
-        project/version requirement spec, this method returns the location of
-        a matching distribution (possibly after downloading it to `tmpdir`).
-        If `spec` is a locally existing file or directory name, it is simply
-        returned unchanged.  If `spec` is a URL, it is downloaded to a subpath
-        of `tmpdir`, and the local filename is returned.  Various errors may be
-        raised if a problem occurs during downloading.
-        """
-        if tmpdir is None:
-            tmpdir=tempfile.mkdtemp()
-        if not isinstance(spec,Requirement):
-            scheme = URL_SCHEME(spec)
-            if scheme:
-                # It's a url, download it to tmpdir
-                found = self._download_url(scheme.group(1), spec, tmpdir)
-                base, fragment = egg_info_for_url(spec)
-                if base.endswith('.py'):
-                    found = self.gen_setup(found,fragment,tmpdir)
-                return found
-            elif os.path.exists(spec):
-                # Existing file or directory, just return it
-                return spec
-            else:
-                try:
-                    spec = Requirement.parse(spec)
-                except ValueError:
-                    raise DistutilsError(
-                        "Not a URL, existing file, or requirement spec: %r" %
-                        (spec,)
-                    )
-        return getattr(self.fetch_distribution(spec, tmpdir),'location',None)
-
-    def fetch_distribution(self,
-        requirement, tmpdir, force_scan=False, source=False, develop_ok=False
-    ):
-        """Obtain a distribution suitable for fulfilling `requirement`
-
-        `requirement` must be a ``pkg_resources.Requirement`` instance.
-        If necessary, or if the `force_scan` flag is set, the requirement is
-        searched for in the (online) package index as well as the locally
-        installed packages.  If a distribution matching `requirement` is found,
-        the returned distribution's ``location`` is the value you would have
-        gotten from calling the ``download()`` method with the matching
-        distribution's URL or filename.  If no matching distribution is found,
-        ``None`` is returned.
-
-        If the `source` flag is set, only source distributions and source
-        checkout links will be considered.  Unless the `develop_ok` flag is
-        set, development and system eggs (i.e., those using the ``.egg-info``
-        format) will be ignored.
-        """
-        
-        # process a Requirement
-        self.info("Searching for %s", requirement)
-        skipped = {}
-
-        def find(req):
-            # Find a matching distribution; may be called more than once
-
-            for dist in self[req.key]:
-
-                if dist.precedence==DEVELOP_DIST and not develop_ok:
-                    if dist not in skipped:
-                        self.warn("Skipping development or system egg: %s",dist)
-                        skipped[dist] = 1
-                    continue
-
-                if dist in req and (dist.precedence<=SOURCE_DIST or not source):
-                    self.info("Best match: %s", dist)
-                    return dist.clone(
-                        location=self.download(dist.location, tmpdir)
-                    )
-
-        if force_scan:
-            self.prescan()
-            self.find_packages(requirement)
-
-        dist = find(requirement)
-        if dist is None and self.to_scan is not None:
-            self.prescan()
-            dist = find(requirement)
-
-        if dist is None and not force_scan:
-            self.find_packages(requirement)
-            dist = find(requirement)
-
-        if dist is None:
-            self.warn(
-                "No local packages or download links found for %s%s",
-                (source and "a source distribution of " or ""),
-                requirement,
-            )
-        return dist
-
-    def fetch(self, requirement, tmpdir, force_scan=False, source=False):
-        """Obtain a file suitable for fulfilling `requirement`
-
-        DEPRECATED; use the ``fetch_distribution()`` method now instead.  For
-        backward compatibility, this routine is identical but returns the
-        ``location`` of the downloaded distribution instead of a distribution
-        object.
-        """
-        dist = self.fetch_distribution(requirement,tmpdir,force_scan,source)
-        if dist is not None:
-            return dist.location
-        return None
-    
-    def gen_setup(self, filename, fragment, tmpdir):
-        match = EGG_FRAGMENT.match(fragment)
-        dists = match and [d for d in
-            interpret_distro_name(filename, match.group(1), None) if d.version
-        ] or []
-
-        if len(dists)==1:   # unambiguous ``#egg`` fragment
-            basename = os.path.basename(filename)
-
-            # Make sure the file has been downloaded to the temp dir.
-            if os.path.dirname(filename) != tmpdir:
-                dst = os.path.join(tmpdir, basename)
-                from setuptools.command.easy_install import samefile
-                if not samefile(filename, dst):
-                    shutil.copy2(filename, dst)
-                    filename=dst
-
-            file = open(os.path.join(tmpdir, 'setup.py'), 'w')
-            file.write(
-                "from setuptools import setup\n"
-                "setup(name=%r, version=%r, py_modules=[%r])\n"
-                % (
-                    dists[0].project_name, dists[0].version,
-                    os.path.splitext(basename)[0]
-                )
-            )
-            file.close()
-            return filename
-
-        elif match:
-            raise DistutilsError(
-                "Can't unambiguously interpret project/version identifier %r; "
-                "any dashes in the name or version should be escaped using "
-                "underscores. %r" % (fragment,dists)
-            )
-        else:
-            raise DistutilsError(
-                "Can't process plain .py without an '#egg=name-version'"
-                " suffix to enable automatic setup script generation."
-            )
-
-    dl_blocksize = 8192
-    def _download_to(self, url, filename):
-        self.info("Downloading %s", url)
-        # Download the file
-        fp, tfp, info = None, None, None
-        try:
-            if '#' in url:
-                url, info = url.split('#', 1)
-            fp = self.open_url(url)
-            if isinstance(fp, urllib2.HTTPError):
-                raise DistutilsError(
-                    "Can't download %s: %s %s" % (url, fp.code,fp.msg)
-                )
-            cs = md5()
-            headers = fp.info()
-            blocknum = 0
-            bs = self.dl_blocksize
-            size = -1
-            if "content-length" in headers:
-                size = int(headers["Content-Length"])
-                self.reporthook(url, filename, blocknum, bs, size)
-            tfp = open(filename,'wb')
-            while True:
-                block = fp.read(bs)
-                if block:
-                    cs.update(block)
-                    tfp.write(block)
-                    blocknum += 1
-                    self.reporthook(url, filename, blocknum, bs, size)
-                else:
-                    break
-            if info: self.check_md5(cs, info, filename, tfp)
-            return headers
-        finally:
-            if fp: fp.close()
-            if tfp: tfp.close()
-
-    def reporthook(self, url, filename, blocknum, blksize, size):
-        pass    # no-op
-
-    def open_url(self, url, warning=None):
-        if url.startswith('file:'):
-            return local_open(url)
-        try:
-            return open_with_auth(url)
-        except (ValueError, httplib.InvalidURL), v:
-            msg = ' '.join([str(arg) for arg in v.args])
-            if warning:
-                self.warn(warning, msg)
-            else:
-                raise DistutilsError('%s %s' % (url, msg))
-        except urllib2.HTTPError, v:
-            return v
-        except urllib2.URLError, v:
-            if warning:
-                self.warn(warning, v.reason)
-            else:
-                raise DistutilsError("Download error for %s: %s"
-                                     % (url, v.reason))
-        except httplib.BadStatusLine, v:
-            if warning:
-                self.warn(warning, v.line)
-            else:
-                raise DistutilsError('%s returned a bad status line. '
-                                     'The server might be down, %s' % \
-                                             (url, v.line))
-        except httplib.HTTPException, v:
-            if warning:
-                self.warn(warning, v)
-            else:
-                raise DistutilsError("Download error for %s: %s"
-                                     % (url, v))
-
-    def _download_url(self, scheme, url, tmpdir):
-        # Determine download filename
-        name = filter(None,urlparse.urlparse(url)[2].split('/'))
-        if name:
-            name = name[-1]
-            while '..' in name:
-                name = name.replace('..','.').replace('\\','_')
-        else:
-            name = "__downloaded__"    # default if URL has no path contents
-
-        if name.endswith('.egg.zip'):
-            name = name[:-4]    # strip the extra .zip before download
-
-        filename = os.path.join(tmpdir,name)
-
-        # Download the file
-        if scheme=='svn' or scheme.startswith('svn+'):
-            return self._download_svn(url, filename)
-        elif scheme=='file':
-            return urllib2.url2pathname(urlparse.urlparse(url)[2])
-        else:
-            self.url_ok(url, True)   # raises error if not allowed
-            return self._attempt_download(url, filename)
-
-    def scan_url(self, url):
-        self.process_url(url, True)
-
-    def _attempt_download(self, url, filename):
-        headers = self._download_to(url, filename)
-        if 'html' in headers.get('content-type','').lower():
-            return self._download_html(url, headers, filename)
-        else:
-            return filename
-
-    def _download_html(self, url, headers, filename):
-        file = open(filename)
-        for line in file:
-            if line.strip():
-                # Check for a subversion index page
-                if re.search(r'<title>([^- ]+ - )?Revision \d+:', line):
-                    # it's a subversion index page:
-                    file.close()
-                    os.unlink(filename)
-                    return self._download_svn(url, filename)
-                break   # not an index page
-        file.close()
-        os.unlink(filename)
-        raise DistutilsError("Unexpected HTML page found at "+url)
-
-    def _download_svn(self, url, filename):
-        url = url.split('#',1)[0]   # remove any fragment for svn's sake
-        self.info("Doing subversion checkout from %s to %s", url, filename)
-        os.system("svn checkout -q %s %s" % (url, filename))
-        return filename
-
-    def debug(self, msg, *args):
-        logging.debug(msg, *args)
-
-    def info(self, msg, *args):
-        logging.info(msg, *args)
-
-    def warn(self, msg, *args):
-        logging.warn(msg, *args)
-
-# This pattern matches a character entity reference (a decimal numeric
-# references, a hexadecimal numeric reference, or a named reference).
-entity_sub = re.compile(r'&(#(\d+|x[\da-fA-F]+)|[\w.:-]+);?').sub
-
-def uchr(c):
-    if not isinstance(c, int):
-        return c
-    if c>255: 
-        return unichr(c)
-    return chr(c)
-
-def decode_entity(match):
-    what = match.group(1)
-    if what.startswith('#x'):
-        what = int(what[2:], 16)
-    elif what.startswith('#'):
-        what = int(what[1:])
-    else:
-        from htmlentitydefs import name2codepoint
-        what = name2codepoint.get(what, match.group(0))
-    return uchr(what)
-
-def htmldecode(text):
-    """Decode HTML entities in the given text."""
-    return entity_sub(decode_entity, text)
-
-def socket_timeout(timeout=15):
-    def _socket_timeout(func):
-        def _socket_timeout(*args, **kwargs):
-            old_timeout = socket.getdefaulttimeout()
-            socket.setdefaulttimeout(timeout)
-            try:
-                return func(*args, **kwargs)
-            finally:
-                socket.setdefaulttimeout(old_timeout)
-        return _socket_timeout
-    return _socket_timeout
-
-def _open_url(self, url):
-    """Open a urllib2 request, handling HTTP authentication"""
-
-    scheme, netloc, path, params, query, frag = urlparse.urlparse(url)
-
-    if scheme in ('http', 'https'):
-        auth, host = urllib2.splituser(netloc)
-    else:
-        auth = None
-
-    if auth:
-        auth = "Basic " + urllib2.unquote(auth).encode('base64').strip()
-        new_url = urlparse.urlunparse((scheme,host,path,params,query,frag))
-        request = urllib2.Request(new_url)
-        request.add_header("Authorization", auth)
-    else:
-        request = urllib2.Request(url)
-
-    request.add_header('User-Agent', user_agent)
-    fp = urllib2.urlopen(request)
-
-    if auth:
-        # Put authentication info back into request URL if same host,
-        # so that links found on the page will work
-        s2, h2, path2, param2, query2, frag2 = urlparse.urlparse(fp.url)
-        if s2==scheme and h2==host:
-            fp.url = urlparse.urlunparse((s2,netloc,path2,param2,query2,frag2))
-
-    return fp
-
-
-# adding a timeout to avoid freezing package_index
-open_with_auth = socket_timeout(_SOCKET_TIMEOUT)(open_with_auth)
-
-def fix_sf_url(url):
-    return url      # backward compatibility
-
-def local_open(url):
-    """Read a local path, with special support for directories"""
-    scheme, server, path, param, query, frag = urlparse.urlparse(url)
-    filename = urllib2.url2pathname(path)
-    if os.path.isfile(filename):
-        return urllib2.urlopen(url)
-    elif path.endswith('/') and os.path.isdir(filename):
-        files = []
-        for f in os.listdir(filename):
-            if f=='index.html':
-                fp = open(os.path.join(filename,f),'rb')
-                body = fp.read()
-                fp.close()
-                break
-            elif os.path.isdir(os.path.join(filename,f)):
-                f+='/'
-            files.append("<a href=%r>%s</a>" % (f,f))
-        else:
-            body = ("<html><head><title>%s</title>" % url) + \
-                "</head><body>%s</body></html>" % '\n'.join(files)
-        status, message = 200, "OK"
-    else:
-        status, message, body = 404, "Path not found", "Not found"
-
-    return urllib2.HTTPError(url, status, message,
-            {'content-type':'text/html'}, cStringIO.StringIO(body))
diff --git a/src/distutils2/pypi/pkg_resources.py b/src/distutils2/pypi/pkg_resources.py
deleted file mode 100644
--- a/src/distutils2/pypi/pkg_resources.py
+++ /dev/null
@@ -1,2693 +0,0 @@
-"""Package resource API
---------------------
-
-A resource is a logical file contained within a package, or a logical
-subdirectory thereof.  The package resource API expects resource names
-to have their path parts separated with ``/``, *not* whatever the local
-path separator is.  Do not use os.path operations to manipulate resource
-names being passed into the API.
-
-The package resource API is designed to work with normal filesystem packages,
-.egg files, and unpacked .egg files.  It can also work in a limited way with
-.zip files and with custom PEP 302 loaders that support the ``get_data()``
-method.
-"""
-
-import sys, os, zipimport, time, re, imp, types
-from urlparse import urlparse, urlunparse
-
-try:
-    frozenset
-except NameError:
-    from sets import ImmutableSet as frozenset
-
-# capture these to bypass sandboxing
-from os import utime
-try:
-    from os import mkdir, rename, unlink
-    WRITE_SUPPORT = True
-except ImportError:
-    # no write support, probably under GAE
-    WRITE_SUPPORT = False
-
-from os import open as os_open
-from os.path import isdir, split
-
-# This marker is used to simplify the process that checks is the
-# setuptools package was installed by the Setuptools project
-# or by the Distribute project, in case Setuptools creates
-# a distribution with the same version.
-#
-# The bootstrapping script for instance, will check if this
-# attribute is present to decide wether to reinstall the package
-_distribute = True
-
-def _bypass_ensure_directory(name, mode=0777):
-    # Sandbox-bypassing version of ensure_directory()
-    if not WRITE_SUPPORT:
-        raise IOError('"os.mkdir" not supported on this platform.')
-    dirname, filename = split(name)
-    if dirname and filename and not isdir(dirname):
-        _bypass_ensure_directory(dirname)
-        mkdir(dirname, mode)
-
-
-
-
-
-
-
-
-def get_supported_platform():
-    """Return this platform's maximum compatible version.
-
-    distutils.util.get_platform() normally reports the minimum version
-    of Mac OS X that would be required to *use* extensions produced by
-    distutils.  But what we want when checking compatibility is to know the
-    version of Mac OS X that we are *running*.  To allow usage of packages that
-    explicitly require a newer version of Mac OS X, we must also know the
-    current version of the OS.
-
-    If this condition occurs for any other platform with a version in its
-    platform strings, this function should be extended accordingly.
-    """
-    plat = get_build_platform(); m = macosVersionString.match(plat)
-    if m is not None and sys.platform == "darwin":
-        try:
-            plat = 'macosx-%s-%s' % ('.'.join(_macosx_vers()[:2]), m.group(3))
-        except ValueError:
-            pass    # not Mac OS X
-    return plat
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-__all__ = [
-    # Basic resource access and distribution/entry point discovery
-    'require', 'run_script', 'get_provider',  'get_distribution',
-    'load_entry_point', 'get_entry_map', 'get_entry_info', 'iter_entry_points',
-    'resource_string', 'resource_stream', 'resource_filename',
-    'resource_listdir', 'resource_exists', 'resource_isdir',
-
-    # Environmental control
-    'declare_namespace', 'working_set', 'add_activation_listener',
-    'find_distributions', 'set_extraction_path', 'cleanup_resources',
-    'get_default_cache',
-
-    # Primary implementation classes
-    'Environment', 'WorkingSet', 'ResourceManager',
-    'Distribution', 'Requirement', 'EntryPoint',
-
-    # Exceptions
-    'ResolutionError','VersionConflict','DistributionNotFound','UnknownExtra',
-    'ExtractionError',
-
-    # Parsing functions and string utilities
-    'parse_requirements', 'parse_version', 'safe_name', 'safe_version',
-    'get_platform', 'compatible_platforms', 'yield_lines', 'split_sections',
-    'safe_extra', 'to_filename',
-
-    # filesystem utilities
-    'ensure_directory', 'normalize_path',
-
-    # Distribution "precedence" constants
-    'EGG_DIST', 'BINARY_DIST', 'SOURCE_DIST', 'CHECKOUT_DIST', 'DEVELOP_DIST',
-
-    # "Provider" interfaces, implementations, and registration/lookup APIs
-    'IMetadataProvider', 'IResourceProvider', 'FileMetadata',
-    'PathMetadata', 'EggMetadata', 'EmptyProvider', 'empty_provider',
-    'NullProvider', 'EggProvider', 'DefaultProvider', 'ZipProvider',
-    'register_finder', 'register_namespace_handler', 'register_loader_type',
-    'fixup_namespace_packages', 'get_importer',
-
-    # Deprecated/backward compatibility only
-    'run_main', 'AvailableDistributions',
-]
-class ResolutionError(Exception):
-    """Abstract base for dependency resolution errors"""
-    def __repr__(self):
-        return self.__class__.__name__+repr(self.args)
-
-class VersionConflict(ResolutionError):
-    """An already-installed version conflicts with the requested version"""
-
-class DistributionNotFound(ResolutionError):
-    """A requested distribution was not found"""
-
-class UnknownExtra(ResolutionError):
-    """Distribution doesn't have an "extra feature" of the given name"""
-_provider_factories = {}
-
-PY_MAJOR = sys.version[:3]
-EGG_DIST    = 3
-BINARY_DIST = 2
-SOURCE_DIST = 1
-CHECKOUT_DIST = 0
-DEVELOP_DIST = -1
-
-def register_loader_type(loader_type, provider_factory):
-    """Register `provider_factory` to make providers for `loader_type`
-
-    `loader_type` is the type or class of a PEP 302 ``module.__loader__``,
-    and `provider_factory` is a function that, passed a *module* object,
-    returns an ``IResourceProvider`` for that module.
-    """
-    _provider_factories[loader_type] = provider_factory
-
-def get_provider(moduleOrReq):
-    """Return an IResourceProvider for the named module or requirement"""
-    if isinstance(moduleOrReq,Requirement):
-        return working_set.find(moduleOrReq) or require(str(moduleOrReq))[0]
-    try:
-        module = sys.modules[moduleOrReq]
-    except KeyError:
-        __import__(moduleOrReq)
-        module = sys.modules[moduleOrReq]
-    loader = getattr(module, '__loader__', None)
-    return _find_adapter(_provider_factories, loader)(module)
-
-def _macosx_vers(_cache=[]):
-    if not _cache:
-        import platform
-        version = platform.mac_ver()[0]
-        # fallback for MacPorts
-        if version == '':
-            import plistlib
-            plist = '/System/Library/CoreServices/SystemVersion.plist'
-            if os.path.exists(plist):
-                if hasattr(plistlib, 'readPlist'):
-                    plist_content = plistlib.readPlist(plist)
-                    if 'ProductVersion' in plist_content:
-                        version = plist_content['ProductVersion']
-
-        _cache.append(version.split('.'))
-    return _cache[0]
-
-def _macosx_arch(machine):
-    return {'PowerPC':'ppc', 'Power_Macintosh':'ppc'}.get(machine,machine)
-
-def get_build_platform():
-    """Return this platform's string for platform-specific distributions
-
-    XXX Currently this is the same as ``distutils.util.get_platform()``, but it
-    needs some hacks for Linux and Mac OS X.
-    """
-    try:
-        from distutils.util import get_platform
-    except ImportError:
-        from sysconfig import get_platform
-
-    plat = get_platform()
-    if sys.platform == "darwin" and not plat.startswith('macosx-'):
-        try:
-            version = _macosx_vers()
-            machine = os.uname()[4].replace(" ", "_")
-            return "macosx-%d.%d-%s" % (int(version[0]), int(version[1]),
-                _macosx_arch(machine))
-        except ValueError:
-            # if someone is running a non-Mac darwin system, this will fall
-            # through to the default implementation
-            pass
-    return plat
-
-macosVersionString = re.compile(r"macosx-(\d+)\.(\d+)-(.*)")
-darwinVersionString = re.compile(r"darwin-(\d+)\.(\d+)\.(\d+)-(.*)")
-get_platform = get_build_platform   # XXX backward compat
-
-def compatible_platforms(provided,required):
-    """Can code for the `provided` platform run on the `required` platform?
-
-    Returns true if either platform is ``None``, or the platforms are equal.
-
-    XXX Needs compatibility checks for Linux and other unixy OSes.
-    """
-    if provided is None or required is None or provided==required:
-        return True     # easy case
-
-    # Mac OS X special cases
-    reqMac = macosVersionString.match(required)
-    if reqMac:
-        provMac = macosVersionString.match(provided)
-
-        # is this a Mac package?
-        if not provMac:
-            # this is backwards compatibility for packages built before
-            # setuptools 0.6. All packages built after this point will
-            # use the new macosx designation.
-            provDarwin = darwinVersionString.match(provided)
-            if provDarwin:
-                dversion = int(provDarwin.group(1))
-                macosversion = "%s.%s" % (reqMac.group(1), reqMac.group(2))
-                if dversion == 7 and macosversion >= "10.3" or \
-                    dversion == 8 and macosversion >= "10.4":
-
-                    #import warnings
-                    #warnings.warn("Mac eggs should be rebuilt to "
-                    #    "use the macosx designation instead of darwin.",
-                    #    category=DeprecationWarning)
-                    return True
-            return False    # egg isn't macosx or legacy darwin
-
-        # are they the same major version and machine type?
-        if provMac.group(1) != reqMac.group(1) or \
-            provMac.group(3) != reqMac.group(3):
-            return False
-
-
-
-        # is the required OS major update >= the provided one?
-        if int(provMac.group(2)) > int(reqMac.group(2)):
-            return False
-
-        return True
-
-    # XXX Linux and other platforms' special cases should go here
-    return False
-
-
-def run_script(dist_spec, script_name):
-    """Locate distribution `dist_spec` and run its `script_name` script"""
-    ns = sys._getframe(1).f_globals
-    name = ns['__name__']
-    ns.clear()
-    ns['__name__'] = name
-    require(dist_spec)[0].run_script(script_name, ns)
-
-run_main = run_script   # backward compatibility
-
-def get_distribution(dist):
-    """Return a current distribution object for a Requirement or string"""
-    if isinstance(dist,basestring): dist = Requirement.parse(dist)
-    if isinstance(dist,Requirement): dist = get_provider(dist)
-    if not isinstance(dist,Distribution):
-        raise TypeError("Expected string, Requirement, or Distribution", dist)
-    return dist
-
-def load_entry_point(dist, group, name):
-    """Return `name` entry point of `group` for `dist` or raise ImportError"""
-    return get_distribution(dist).load_entry_point(group, name)
-
-def get_entry_map(dist, group=None):
-    """Return the entry point map for `group`, or the full entry map"""
-    return get_distribution(dist).get_entry_map(group)
-
-def get_entry_info(dist, group, name):
-    """Return the EntryPoint object for `group`+`name`, or ``None``"""
-    return get_distribution(dist).get_entry_info(group, name)
-
-
-class IMetadataProvider:
-
-    def has_metadata(name):
-        """Does the package's distribution contain the named metadata?"""
-
-    def get_metadata(name):
-        """The named metadata resource as a string"""
-
-    def get_metadata_lines(name):
-        """Yield named metadata resource as list of non-blank non-comment lines
-
-       Leading and trailing whitespace is stripped from each line, and lines
-       with ``#`` as the first non-blank character are omitted."""
-
-    def metadata_isdir(name):
-        """Is the named metadata a directory?  (like ``os.path.isdir()``)"""
-
-    def metadata_listdir(name):
-        """List of metadata names in the directory (like ``os.listdir()``)"""
-
-    def run_script(script_name, namespace):
-        """Execute the named script in the supplied namespace dictionary"""
-
-
-
-
-
-
-
-
-
-
-class IResourceProvider(IMetadataProvider):
-    """An object that provides access to package resources"""
-
-    def get_resource_filename(manager, resource_name):
-        """Return a true filesystem path for `resource_name`
-
-        `manager` must be an ``IResourceManager``"""
-
-    def get_resource_stream(manager, resource_name):
-        """Return a readable file-like object for `resource_name`
-
-        `manager` must be an ``IResourceManager``"""
-
-    def get_resource_string(manager, resource_name):
-        """Return a string containing the contents of `resource_name`
-
-        `manager` must be an ``IResourceManager``"""
-
-    def has_resource(resource_name):
-        """Does the package contain the named resource?"""
-
-    def resource_isdir(resource_name):
-        """Is the named resource a directory?  (like ``os.path.isdir()``)"""
-
-    def resource_listdir(resource_name):
-        """List of resource names in the directory (like ``os.listdir()``)"""
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-class WorkingSet(object):
-    """A collection of active distributions on sys.path (or a similar list)"""
-
-    def __init__(self, entries=None):
-        """Create working set from list of path entries (default=sys.path)"""
-        self.entries = []
-        self.entry_keys = {}
-        self.by_key = {}
-        self.callbacks = []
-
-        if entries is None:
-            entries = sys.path
-
-        for entry in entries:
-            self.add_entry(entry)
-
-
-    def add_entry(self, entry):
-        """Add a path item to ``.entries``, finding any distributions on it
-
-        ``find_distributions(entry,True)`` is used to find distributions
-        corresponding to the path entry, and they are added.  `entry` is
-        always appended to ``.entries``, even if it is already present.
-        (This is because ``sys.path`` can contain the same value more than
-        once, and the ``.entries`` of the ``sys.path`` WorkingSet should always
-        equal ``sys.path``.)
-        """
-        self.entry_keys.setdefault(entry, [])
-        self.entries.append(entry)
-        for dist in find_distributions(entry, True):
-            self.add(dist, entry, False)
-
-
-    def __contains__(self,dist):
-        """True if `dist` is the active distribution for its project"""
-        return self.by_key.get(dist.key) == dist
-
-
-
-
-
-    def find(self, req):
-        """Find a distribution matching requirement `req`
-
-        If there is an active distribution for the requested project, this
-        returns it as long as it meets the version requirement specified by
-        `req`.  But, if there is an active distribution for the project and it
-        does *not* meet the `req` requirement, ``VersionConflict`` is raised.
-        If there is no active distribution for the requested project, ``None``
-        is returned.
-        """
-        dist = self.by_key.get(req.key)
-        if dist is not None and dist not in req:
-            raise VersionConflict(dist,req)     # XXX add more info
-        else:
-            return dist
-
-    def iter_entry_points(self, group, name=None):
-        """Yield entry point objects from `group` matching `name`
-
-        If `name` is None, yields all entry points in `group` from all
-        distributions in the working set, otherwise only ones matching
-        both `group` and `name` are yielded (in distribution order).
-        """
-        for dist in self:
-            entries = dist.get_entry_map(group)
-            if name is None:
-                for ep in entries.values():
-                    yield ep
-            elif name in entries:
-                yield entries[name]
-
-    def run_script(self, requires, script_name):
-        """Locate distribution for `requires` and run `script_name` script"""
-        ns = sys._getframe(1).f_globals
-        name = ns['__name__']
-        ns.clear()
-        ns['__name__'] = name
-        self.require(requires)[0].run_script(script_name, ns)
-
-
-
-    def __iter__(self):
-        """Yield distributions for non-duplicate projects in the working set
-
-        The yield order is the order in which the items' path entries were
-        added to the working set.
-        """
-        seen = {}
-        for item in self.entries:
-            for key in self.entry_keys[item]:
-                if key not in seen:
-                    seen[key]=1
-                    yield self.by_key[key]
-
-    def add(self, dist, entry=None, insert=True):
-        """Add `dist` to working set, associated with `entry`
-
-        If `entry` is unspecified, it defaults to the ``.location`` of `dist`.
-        On exit from this routine, `entry` is added to the end of the working
-        set's ``.entries`` (if it wasn't already present).
-
-        `dist` is only added to the working set if it's for a project that
-        doesn't already have a distribution in the set.  If it's added, any
-        callbacks registered with the ``subscribe()`` method will be called.
-        """
-        if insert:
-            dist.insert_on(self.entries, entry)
-
-        if entry is None:
-            entry = dist.location
-        keys = self.entry_keys.setdefault(entry,[])
-        keys2 = self.entry_keys.setdefault(dist.location,[])
-        if dist.key in self.by_key:
-            return      # ignore hidden distros
-
-        self.by_key[dist.key] = dist
-        if dist.key not in keys:
-            keys.append(dist.key)
-        if dist.key not in keys2:
-            keys2.append(dist.key)
-        self._added_new(dist)
-
-    def resolve(self, requirements, env=None, installer=None, replacement=True):
-        """List all distributions needed to (recursively) meet `requirements`
-
-        `requirements` must be a sequence of ``Requirement`` objects.  `env`,
-        if supplied, should be an ``Environment`` instance.  If
-        not supplied, it defaults to all distributions available within any
-        entry or distribution in the working set.  `installer`, if supplied,
-        will be invoked with each requirement that cannot be met by an
-        already-installed distribution; it should return a ``Distribution`` or
-        ``None``.
-        """
-
-        requirements = list(requirements)[::-1]  # set up the stack
-        processed = {}  # set of processed requirements
-        best = {}  # key -> dist
-        to_activate = []
-
-        while requirements:
-            req = requirements.pop(0)   # process dependencies breadth-first
-            if _override_setuptools(req) and replacement:
-                req = Requirement.parse('distribute')
-
-            if req in processed:
-                # Ignore cyclic or redundant dependencies
-                continue
-            dist = best.get(req.key)
-            if dist is None:
-                # Find the best distribution and add it to the map
-                dist = self.by_key.get(req.key)
-                if dist is None:
-                    if env is None:
-                        env = Environment(self.entries)
-                    dist = best[req.key] = env.best_match(req, self, installer)
-                    if dist is None:
-                        #msg = ("The '%s' distribution was not found on this "
-                        #       "system, and is required by this application.")
-                        #raise DistributionNotFound(msg % req)
-
-                        # unfortunately, zc.buildout uses a str(err)
-                        # to get the name of the distribution here..
-                        raise DistributionNotFound(req)
-                to_activate.append(dist)
-            if dist not in req:
-                # Oops, the "best" so far conflicts with a dependency
-                raise VersionConflict(dist,req) # XXX put more info here
-            requirements.extend(dist.requires(req.extras)[::-1])
-            processed[req] = True
-
-        return to_activate    # return list of distros to activate
-
-    def find_plugins(self,
-        plugin_env, full_env=None, installer=None, fallback=True
-    ):
-        """Find all activatable distributions in `plugin_env`
-
-        Example usage::
-
-            distributions, errors = working_set.find_plugins(
-                Environment(plugin_dirlist)
-            )
-            map(working_set.add, distributions)  # add plugins+libs to sys.path
-            print 'Could not load', errors        # display errors
-
-        The `plugin_env` should be an ``Environment`` instance that contains
-        only distributions that are in the project's "plugin directory" or
-        directories. The `full_env`, if supplied, should be an ``Environment``
-        contains all currently-available distributions.  If `full_env` is not
-        supplied, one is created automatically from the ``WorkingSet`` this
-        method is called on, which will typically mean that every directory on
-        ``sys.path`` will be scanned for distributions.
-
-        `installer` is a standard installer callback as used by the
-        ``resolve()`` method. The `fallback` flag indicates whether we should
-        attempt to resolve older versions of a plugin if the newest version
-        cannot be resolved.
-
-        This method returns a 2-tuple: (`distributions`, `error_info`), where
-        `distributions` is a list of the distributions found in `plugin_env`
-        that were loadable, along with any other distributions that are needed
-        to resolve their dependencies.  `error_info` is a dictionary mapping
-        unloadable plugin distributions to an exception instance describing the
-        error that occurred. Usually this will be a ``DistributionNotFound`` or
-        ``VersionConflict`` instance.
-        """
-
-        plugin_projects = list(plugin_env)
-        plugin_projects.sort()  # scan project names in alphabetic order
-
-        error_info = {}
-        distributions = {}
-
-        if full_env is None:
-            env = Environment(self.entries)
-            env += plugin_env
-        else:
-            env = full_env + plugin_env
-
-        shadow_set = self.__class__([])
-        map(shadow_set.add, self)   # put all our entries in shadow_set
-
-        for project_name in plugin_projects:
-
-            for dist in plugin_env[project_name]:
-
-                req = [dist.as_requirement()]
-
-                try:
-                    resolvees = shadow_set.resolve(req, env, installer)
-
-                except ResolutionError,v:
-                    error_info[dist] = v    # save error info
-                    if fallback:
-                        continue    # try the next older version of project
-                    else:
-                        break       # give up on this project, keep going
-
-                else:
-                    map(shadow_set.add, resolvees)
-                    distributions.update(dict.fromkeys(resolvees))
-
-                    # success, no need to try any more versions of this project
-                    break
-
-        distributions = list(distributions)
-        distributions.sort()
-
-        return distributions, error_info
-
-
-
-
-
-    def require(self, *requirements):
-        """Ensure that distributions matching `requirements` are activated
-
-        `requirements` must be a string or a (possibly-nested) sequence
-        thereof, specifying the distributions and versions required.  The
-        return value is a sequence of the distributions that needed to be
-        activated to fulfill the requirements; all relevant distributions are
-        included, even if they were already activated in this working set.
-        """
-
-        needed = self.resolve(parse_requirements(requirements))
-
-        for dist in needed:
-            self.add(dist)
-
-        return needed
-
-
-    def subscribe(self, callback):
-        """Invoke `callback` for all distributions (including existing ones)"""
-        if callback in self.callbacks:
-            return
-        self.callbacks.append(callback)
-        for dist in self:
-            callback(dist)
-
-
-    def _added_new(self, dist):
-        for callback in self.callbacks:
-            callback(dist)
-
-
-
-
-
-
-
-
-
-
-
-class Environment(object):
-    """Searchable snapshot of distributions on a search path"""
-
-    def __init__(self, search_path=None, platform=get_supported_platform(), python=PY_MAJOR):
-        """Snapshot distributions available on a search path
-
-        Any distributions found on `search_path` are added to the environment.
-        `search_path` should be a sequence of ``sys.path`` items.  If not
-        supplied, ``sys.path`` is used.
-
-        `platform` is an optional string specifying the name of the platform
-        that platform-specific distributions must be compatible with.  If
-        unspecified, it defaults to the current platform.  `python` is an
-        optional string naming the desired version of Python (e.g. ``'2.4'``);
-        it defaults to the current version.
-
-        You may explicitly set `platform` (and/or `python`) to ``None`` if you
-        wish to map *all* distributions, not just those compatible with the
-        running platform or Python version.
-        """
-        self._distmap = {}
-        self._cache = {}
-        self.platform = platform
-        self.python = python
-        self.scan(search_path)
-
-    def can_add(self, dist):
-        """Is distribution `dist` acceptable for this environment?
-
-        The distribution must match the platform and python version
-        requirements specified when this environment was created, or False
-        is returned.
-        """
-        return (self.python is None or dist.py_version is None
-            or dist.py_version==self.python) \
-           and compatible_platforms(dist.platform,self.platform)
-
-    def remove(self, dist):
-        """Remove `dist` from the environment"""
-        self._distmap[dist.key].remove(dist)
-
-    def scan(self, search_path=None):
-        """Scan `search_path` for distributions usable in this environment
-
-        Any distributions found are added to the environment.
-        `search_path` should be a sequence of ``sys.path`` items.  If not
-        supplied, ``sys.path`` is used.  Only distributions conforming to
-        the platform/python version defined at initialization are added.
-        """
-        if search_path is None:
-            search_path = sys.path
-
-        for item in search_path:
-            for dist in find_distributions(item):
-                self.add(dist)
-
-    def __getitem__(self,project_name):
-        """Return a newest-to-oldest list of distributions for `project_name`
-        """
-        try:
-            return self._cache[project_name]
-        except KeyError:
-            project_name = project_name.lower()
-            if project_name not in self._distmap:
-                return []
-
-        if project_name not in self._cache:
-            dists = self._cache[project_name] = self._distmap[project_name]
-            _sort_dists(dists)
-
-        return self._cache[project_name]
-
-    def add(self,dist):
-        """Add `dist` if we ``can_add()`` it and it isn't already added"""
-        if self.can_add(dist) and dist.has_version():
-            dists = self._distmap.setdefault(dist.key,[])
-            if dist not in dists:
-                dists.append(dist)
-                if dist.key in self._cache:
-                    _sort_dists(self._cache[dist.key])
-
-
-    def best_match(self, req, working_set, installer=None):
-        """Find distribution best matching `req` and usable on `working_set`
-
-        This calls the ``find(req)`` method of the `working_set` to see if a
-        suitable distribution is already active.  (This may raise
-        ``VersionConflict`` if an unsuitable version of the project is already
-        active in the specified `working_set`.)  If a suitable distribution
-        isn't active, this method returns the newest distribution in the
-        environment that meets the ``Requirement`` in `req`.  If no suitable
-        distribution is found, and `installer` is supplied, then the result of
-        calling the environment's ``obtain(req, installer)`` method will be
-        returned.
-        """
-        dist = working_set.find(req)
-        if dist is not None:
-            return dist
-        for dist in self[req.key]:
-            if dist in req:
-                return dist
-        return self.obtain(req, installer) # try and download/install
-
-    def obtain(self, requirement, installer=None):
-        """Obtain a distribution matching `requirement` (e.g. via download)
-
-        Obtain a distro that matches requirement (e.g. via download).  In the
-        base ``Environment`` class, this routine just returns
-        ``installer(requirement)``, unless `installer` is None, in which case
-        None is returned instead.  This method is a hook that allows subclasses
-        to attempt other ways of obtaining a distribution before falling back
-        to the `installer` argument."""
-        if installer is not None:
-            return installer(requirement)
-
-    def __iter__(self):
-        """Yield the unique project names of the available distributions"""
-        for key in self._distmap.keys():
-            if self[key]: yield key
-
-
-
-
-    def __iadd__(self, other):
-        """In-place addition of a distribution or environment"""
-        if isinstance(other,Distribution):
-            self.add(other)
-        elif isinstance(other,Environment):
-            for project in other:
-                for dist in other[project]:
-                    self.add(dist)
-        else:
-            raise TypeError("Can't add %r to environment" % (other,))
-        return self
-
-    def __add__(self, other):
-        """Add an environment or distribution to an environment"""
-        new = self.__class__([], platform=None, python=None)
-        for env in self, other:
-            new += env
-        return new
-
-
-AvailableDistributions = Environment    # XXX backward compatibility
-
-
-class ExtractionError(RuntimeError):
-    """An error occurred extracting a resource
-
-    The following attributes are available from instances of this exception:
-
-    manager
-        The resource manager that raised this exception
-
-    cache_path
-        The base directory for resource extraction
-
-    original_error
-        The exception instance that caused extraction to fail
-    """
-
-
-
-
-class ResourceManager:
-    """Manage resource extraction and packages"""
-    extraction_path = None
-
-    def __init__(self):
-        self.cached_files = {}
-
-    def resource_exists(self, package_or_requirement, resource_name):
-        """Does the named resource exist?"""
-        return get_provider(package_or_requirement).has_resource(resource_name)
-
-    def resource_isdir(self, package_or_requirement, resource_name):
-        """Is the named resource an existing directory?"""
-        return get_provider(package_or_requirement).resource_isdir(
-            resource_name
-        )
-
-    def resource_filename(self, package_or_requirement, resource_name):
-        """Return a true filesystem path for specified resource"""
-        return get_provider(package_or_requirement).get_resource_filename(
-            self, resource_name
-        )
-
-    def resource_stream(self, package_or_requirement, resource_name):
-        """Return a readable file-like object for specified resource"""
-        return get_provider(package_or_requirement).get_resource_stream(
-            self, resource_name
-        )
-
-    def resource_string(self, package_or_requirement, resource_name):
-        """Return specified resource as a string"""
-        return get_provider(package_or_requirement).get_resource_string(
-            self, resource_name
-        )
-
-    def resource_listdir(self, package_or_requirement, resource_name):
-        """List the contents of the named resource directory"""
-        return get_provider(package_or_requirement).resource_listdir(
-            resource_name
-        )
-
-    def extraction_error(self):
-        """Give an error message for problems extracting file(s)"""
-
-        old_exc = sys.exc_info()[1]
-        cache_path = self.extraction_path or get_default_cache()
-
-        err = ExtractionError("""Can't extract file(s) to egg cache
-
-The following error occurred while trying to extract file(s) to the Python egg
-cache:
-
-  %s
-
-The Python egg cache directory is currently set to:
-
-  %s
-
-Perhaps your account does not have write access to this directory?  You can
-change the cache directory by setting the PYTHON_EGG_CACHE environment
-variable to point to an accessible directory.
-"""         % (old_exc, cache_path)
-        )
-        err.manager        = self
-        err.cache_path     = cache_path
-        err.original_error = old_exc
-        raise err
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-    def get_cache_path(self, archive_name, names=()):
-        """Return absolute location in cache for `archive_name` and `names`
-
-        The parent directory of the resulting path will be created if it does
-        not already exist.  `archive_name` should be the base filename of the
-        enclosing egg (which may not be the name of the enclosing zipfile!),
-        including its ".egg" extension.  `names`, if provided, should be a
-        sequence of path name parts "under" the egg's extraction location.
-
-        This method should only be called by resource providers that need to
-        obtain an extraction location, and only for names they intend to
-        extract, as it tracks the generated names for possible cleanup later.
-        """
-        extract_path = self.extraction_path or get_default_cache()
-        target_path = os.path.join(extract_path, archive_name+'-tmp', *names)
-        try:
-            _bypass_ensure_directory(target_path)
-        except:
-            self.extraction_error()
-
-        self.cached_files[target_path] = 1
-        return target_path
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-    def postprocess(self, tempname, filename):
-        """Perform any platform-specific postprocessing of `tempname`
-
-        This is where Mac header rewrites should be done; other platforms don't
-        have anything special they should do.
-
-        Resource providers should call this method ONLY after successfully
-        extracting a compressed resource.  They must NOT call it on resources
-        that are already in the filesystem.
-
-        `tempname` is the current (temporary) name of the file, and `filename`
-        is the name it will be renamed to by the caller after this routine
-        returns.
-        """
-
-        if os.name == 'posix':
-            # Make the resource executable
-            mode = ((os.stat(tempname).st_mode) | 0555) & 07777
-            os.chmod(tempname, mode)
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-    def set_extraction_path(self, path):
-        """Set the base path where resources will be extracted to, if needed.
-
-        If you do not call this routine before any extractions take place, the
-        path defaults to the return value of ``get_default_cache()``.  (Which
-        is based on the ``PYTHON_EGG_CACHE`` environment variable, with various
-        platform-specific fallbacks.  See that routine's documentation for more
-        details.)
-
-        Resources are extracted to subdirectories of this path based upon
-        information given by the ``IResourceProvider``.  You may set this to a
-        temporary directory, but then you must call ``cleanup_resources()`` to
-        delete the extracted files when done.  There is no guarantee that
-        ``cleanup_resources()`` will be able to remove all extracted files.
-
-        (Note: you may not change the extraction path for a given resource
-        manager once resources have been extracted, unless you first call
-        ``cleanup_resources()``.)
-        """
-        if self.cached_files:
-            raise ValueError(
-                "Can't change extraction path, files already extracted"
-            )
-
-        self.extraction_path = path
-
-    def cleanup_resources(self, force=False):
-        """
-        Delete all extracted resource files and directories, returning a list
-        of the file and directory names that could not be successfully removed.
-        This function does not have any concurrency protection, so it should
-        generally only be called when the extraction path is a temporary
-        directory exclusive to a single process.  This method is not
-        automatically called; you must call it explicitly or register it as an
-        ``atexit`` function if you wish to ensure cleanup of a temporary
-        directory used for extractions.
-        """
-        # XXX
-
-
-
-def get_default_cache():
-    """Determine the default cache location
-
-    This returns the ``PYTHON_EGG_CACHE`` environment variable, if set.
-    Otherwise, on Windows, it returns a "Python-Eggs" subdirectory of the
-    "Application Data" directory.  On all other systems, it's "~/.python-eggs".
-    """
-    try:
-        return os.environ['PYTHON_EGG_CACHE']
-    except KeyError:
-        pass
-
-    if os.name!='nt':
-        return os.path.expanduser('~/.python-eggs')
-
-    app_data = 'Application Data'   # XXX this may be locale-specific!
-    app_homes = [
-        (('APPDATA',), None),       # best option, should be locale-safe
-        (('USERPROFILE',), app_data),
-        (('HOMEDRIVE','HOMEPATH'), app_data),
-        (('HOMEPATH',), app_data),
-        (('HOME',), None),
-        (('WINDIR',), app_data),    # 95/98/ME
-    ]
-
-    for keys, subdir in app_homes:
-        dirname = ''
-        for key in keys:
-            if key in os.environ:
-                dirname = os.path.join(dirname, os.environ[key])
-            else:
-                break
-        else:
-            if subdir:
-                dirname = os.path.join(dirname,subdir)
-            return os.path.join(dirname, 'Python-Eggs')
-    else:
-        raise RuntimeError(
-            "Please set the PYTHON_EGG_CACHE enviroment variable"
-        )
-
-def safe_name(name):
-    """Convert an arbitrary string to a standard distribution name
-
-    Any runs of non-alphanumeric/. characters are replaced with a single '-'.
-    """
-    return re.sub('[^A-Za-z0-9.]+', '-', name)
-
-
-def safe_version(version):
-    """Convert an arbitrary string to a standard version string
-
-    Spaces become dots, and all other non-alphanumeric characters become
-    dashes, with runs of multiple dashes condensed to a single dash.
-    """
-    version = version.replace(' ','.')
-    return re.sub('[^A-Za-z0-9.]+', '-', version)
-
-
-def safe_extra(extra):
-    """Convert an arbitrary string to a standard 'extra' name
-
-    Any runs of non-alphanumeric characters are replaced with a single '_',
-    and the result is always lowercased.
-    """
-    return re.sub('[^A-Za-z0-9.]+', '_', extra).lower()
-
-
-def to_filename(name):
-    """Convert a project or version name to its filename-escaped form
-
-    Any '-' characters are currently replaced with '_'.
-    """
-    return name.replace('-','_')
-
-
-
-
-
-
-
-
-class NullProvider:
-    """Try to implement resources and metadata for arbitrary PEP 302 loaders"""
-
-    egg_name = None
-    egg_info = None
-    loader = None
-
-    def __init__(self, module):
-        self.loader = getattr(module, '__loader__', None)
-        self.module_path = os.path.dirname(getattr(module, '__file__', ''))
-
-    def get_resource_filename(self, manager, resource_name):
-        return self._fn(self.module_path, resource_name)
-
-    def get_resource_stream(self, manager, resource_name):
-        return StringIO(self.get_resource_string(manager, resource_name))
-
-    def get_resource_string(self, manager, resource_name):
-        return self._get(self._fn(self.module_path, resource_name))
-
-    def has_resource(self, resource_name):
-        return self._has(self._fn(self.module_path, resource_name))
-
-    def has_metadata(self, name):
-        return self.egg_info and self._has(self._fn(self.egg_info,name))
-
-    if sys.version_info <= (3,):
-        def get_metadata(self, name):
-            if not self.egg_info:
-                return ""
-            return self._get(self._fn(self.egg_info,name))
-    else:
-        def get_metadata(self, name):
-            if not self.egg_info:
-                return ""
-            return self._get(self._fn(self.egg_info,name)).decode("utf-8")
-
-    def get_metadata_lines(self, name):
-        return yield_lines(self.get_metadata(name))
-
-    def resource_isdir(self,resource_name):
-        return self._isdir(self._fn(self.module_path, resource_name))
-
-    def metadata_isdir(self,name):
-        return self.egg_info and self._isdir(self._fn(self.egg_info,name))
-
-
-    def resource_listdir(self,resource_name):
-        return self._listdir(self._fn(self.module_path,resource_name))
-
-    def metadata_listdir(self,name):
-        if self.egg_info:
-            return self._listdir(self._fn(self.egg_info,name))
-        return []
-
-    def run_script(self,script_name,namespace):
-        script = 'scripts/'+script_name
-        if not self.has_metadata(script):
-            raise ResolutionError("No script named %r" % script_name)
-        script_text = self.get_metadata(script).replace('\r\n','\n')
-        script_text = script_text.replace('\r','\n')
-        script_filename = self._fn(self.egg_info,script)
-        namespace['__file__'] = script_filename
-        if os.path.exists(script_filename):
-            execfile(script_filename, namespace, namespace)
-        else:
-            from linecache import cache
-            cache[script_filename] = (
-                len(script_text), 0, script_text.split('\n'), script_filename
-            )
-            script_code = compile(script_text,script_filename,'exec')
-            exec script_code in namespace, namespace
-
-    def _has(self, path):
-        raise NotImplementedError(
-            "Can't perform this operation for unregistered loader type"
-        )
-
-    def _isdir(self, path):
-        raise NotImplementedError(
-            "Can't perform this operation for unregistered loader type"
-        )
-
-    def _listdir(self, path):
-        raise NotImplementedError(
-            "Can't perform this operation for unregistered loader type"
-        )
-
-    def _fn(self, base, resource_name):
-        if resource_name:
-            return os.path.join(base, *resource_name.split('/'))
-        return base
-
-    def _get(self, path):
-        if hasattr(self.loader, 'get_data'):
-            return self.loader.get_data(path)
-        raise NotImplementedError(
-            "Can't perform this operation for loaders without 'get_data()'"
-        )
-
-register_loader_type(object, NullProvider)
-
-
-class EggProvider(NullProvider):
-    """Provider based on a virtual filesystem"""
-
-    def __init__(self,module):
-        NullProvider.__init__(self,module)
-        self._setup_prefix()
-
-    def _setup_prefix(self):
-        # we assume here that our metadata may be nested inside a "basket"
-        # of multiple eggs; that's why we use module_path instead of .archive
-        path = self.module_path
-        old = None
-        while path!=old:
-            if path.lower().endswith('.egg'):
-                self.egg_name = os.path.basename(path)
-                self.egg_info = os.path.join(path, 'EGG-INFO')
-                self.egg_root = path
-                break
-            old = path
-            path, base = os.path.split(path)
-
-
-
-
-
-
-class DefaultProvider(EggProvider):
-    """Provides access to package resources in the filesystem"""
-
-    def _has(self, path):
-        return os.path.exists(path)
-
-    def _isdir(self,path):
-        return os.path.isdir(path)
-
-    def _listdir(self,path):
-        return os.listdir(path)
-
-    def get_resource_stream(self, manager, resource_name):
-        return open(self._fn(self.module_path, resource_name), 'rb')
-
-    def _get(self, path):
-        stream = open(path, 'rb')
-        try:
-            return stream.read()
-        finally:
-            stream.close()
-
-register_loader_type(type(None), DefaultProvider)
-
-
-class EmptyProvider(NullProvider):
-    """Provider that returns nothing for all requests"""
-
-    _isdir = _has = lambda self,path: False
-    _get          = lambda self,path: ''
-    _listdir      = lambda self,path: []
-    module_path   = None
-
-    def __init__(self):
-        pass
-
-empty_provider = EmptyProvider()
-
-
-
-
-class ZipProvider(EggProvider):
-    """Resource support for zips and eggs"""
-
-    eagers = None
-
-    def __init__(self, module):
-        EggProvider.__init__(self,module)
-        self.zipinfo = zipimport._zip_directory_cache[self.loader.archive]
-        self.zip_pre = self.loader.archive+os.sep
-
-    def _zipinfo_name(self, fspath):
-        # Convert a virtual filename (full path to file) into a zipfile subpath
-        # usable with the zipimport directory cache for our target archive
-        if fspath.startswith(self.zip_pre):
-            return fspath[len(self.zip_pre):]
-        raise AssertionError(
-            "%s is not a subpath of %s" % (fspath,self.zip_pre)
-        )
-
-    def _parts(self,zip_path):
-        # Convert a zipfile subpath into an egg-relative path part list
-        fspath = self.zip_pre+zip_path  # pseudo-fs path
-        if fspath.startswith(self.egg_root+os.sep):
-            return fspath[len(self.egg_root)+1:].split(os.sep)
-        raise AssertionError(
-            "%s is not a subpath of %s" % (fspath,self.egg_root)
-        )
-
-    def get_resource_filename(self, manager, resource_name):
-        if not self.egg_name:
-            raise NotImplementedError(
-                "resource_filename() only supported for .egg, not .zip"
-            )
-        # no need to lock for extraction, since we use temp names
-        zip_path = self._resource_to_zip(resource_name)
-        eagers = self._get_eager_resources()
-        if '/'.join(self._parts(zip_path)) in eagers:
-            for name in eagers:
-                self._extract_resource(manager, self._eager_to_zip(name))
-        return self._extract_resource(manager, zip_path)
-
-    def _extract_resource(self, manager, zip_path):
-
-        if zip_path in self._index():
-            for name in self._index()[zip_path]:
-                last = self._extract_resource(
-                    manager, os.path.join(zip_path, name)
-                )
-            return os.path.dirname(last)  # return the extracted directory name
-
-        zip_stat = self.zipinfo[zip_path]
-        t,d,size = zip_stat[5], zip_stat[6], zip_stat[3]
-        date_time = (
-            (d>>9)+1980, (d>>5)&0xF, d&0x1F,                      # ymd
-            (t&0xFFFF)>>11, (t>>5)&0x3F, (t&0x1F) * 2, 0, 0, -1   # hms, etc.
-        )
-        timestamp = time.mktime(date_time)
-
-        try:
-            if not WRITE_SUPPORT:
-                raise IOError('"os.rename" and "os.unlink" are not supported '
-                              'on this platform')
-
-            real_path = manager.get_cache_path(
-                self.egg_name, self._parts(zip_path)
-            )
-
-            if os.path.isfile(real_path):
-                stat = os.stat(real_path)
-                if stat.st_size==size and stat.st_mtime==timestamp:
-                    # size and stamp match, don't bother extracting
-                    return real_path
-
-            outf, tmpnam = _mkstemp(".$extract", dir=os.path.dirname(real_path))
-            os.write(outf, self.loader.get_data(zip_path))
-            os.close(outf)
-            utime(tmpnam, (timestamp,timestamp))
-            manager.postprocess(tmpnam, real_path)
-
-            try:
-                rename(tmpnam, real_path)
-
-            except os.error:
-                if os.path.isfile(real_path):
-                    stat = os.stat(real_path)
-
-                    if stat.st_size==size and stat.st_mtime==timestamp:
-                        # size and stamp match, somebody did it just ahead of
-                        # us, so we're done
-                        return real_path
-                    elif os.name=='nt':     # Windows, del old file and retry
-                        unlink(real_path)
-                        rename(tmpnam, real_path)
-                        return real_path
-                raise
-
-        except os.error:
-            manager.extraction_error()  # report a user-friendly error
-
-        return real_path
-
-    def _get_eager_resources(self):
-        if self.eagers is None:
-            eagers = []
-            for name in ('native_libs.txt', 'eager_resources.txt'):
-                if self.has_metadata(name):
-                    eagers.extend(self.get_metadata_lines(name))
-            self.eagers = eagers
-        return self.eagers
-
-    def _index(self):
-        try:
-            return self._dirindex
-        except AttributeError:
-            ind = {}
-            for path in self.zipinfo:
-                parts = path.split(os.sep)
-                while parts:
-                    parent = os.sep.join(parts[:-1])
-                    if parent in ind:
-                        ind[parent].append(parts[-1])
-                        break
-                    else:
-                        ind[parent] = [parts.pop()]
-            self._dirindex = ind
-            return ind
-
-    def _has(self, fspath):
-        zip_path = self._zipinfo_name(fspath)
-        return zip_path in self.zipinfo or zip_path in self._index()
-
-    def _isdir(self,fspath):
-        return self._zipinfo_name(fspath) in self._index()
-
-    def _listdir(self,fspath):
-        return list(self._index().get(self._zipinfo_name(fspath), ()))
-
-    def _eager_to_zip(self,resource_name):
-        return self._zipinfo_name(self._fn(self.egg_root,resource_name))
-
-    def _resource_to_zip(self,resource_name):
-        return self._zipinfo_name(self._fn(self.module_path,resource_name))
-
-register_loader_type(zipimport.zipimporter, ZipProvider)
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-class FileMetadata(EmptyProvider):
-    """Metadata handler for standalone PKG-INFO files
-
-    Usage::
-
-        metadata = FileMetadata("/path/to/PKG-INFO")
-
-    This provider rejects all data and metadata requests except for PKG-INFO,
-    which is treated as existing, and will be the contents of the file at
-    the provided location.
-    """
-
-    def __init__(self,path):
-        self.path = path
-
-    def has_metadata(self,name):
-        return name=='PKG-INFO'
-
-    def get_metadata(self,name):
-        if name=='PKG-INFO':
-            f = open(self.path,'rU')
-            metadata = f.read()
-            f.close()
-            return metadata
-        raise KeyError("No metadata except PKG-INFO is available")
-
-    def get_metadata_lines(self,name):
-        return yield_lines(self.get_metadata(name))
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-class PathMetadata(DefaultProvider):
-    """Metadata provider for egg directories
-
-    Usage::
-
-        # Development eggs:
-
-        egg_info = "/path/to/PackageName.egg-info"
-        base_dir = os.path.dirname(egg_info)
-        metadata = PathMetadata(base_dir, egg_info)
-        dist_name = os.path.splitext(os.path.basename(egg_info))[0]
-        dist = Distribution(basedir,project_name=dist_name,metadata=metadata)
-
-        # Unpacked egg directories:
-
-        egg_path = "/path/to/PackageName-ver-pyver-etc.egg"
-        metadata = PathMetadata(egg_path, os.path.join(egg_path,'EGG-INFO'))
-        dist = Distribution.from_filename(egg_path, metadata=metadata)
-    """
-
-    def __init__(self, path, egg_info):
-        self.module_path = path
-        self.egg_info = egg_info
-
-
-class EggMetadata(ZipProvider):
-    """Metadata provider for .egg files"""
-
-    def __init__(self, importer):
-        """Create a metadata provider from a zipimporter"""
-
-        self.zipinfo = zipimport._zip_directory_cache[importer.archive]
-        self.zip_pre = importer.archive+os.sep
-        self.loader = importer
-        if importer.prefix:
-            self.module_path = os.path.join(importer.archive, importer.prefix)
-        else:
-            self.module_path = importer.archive
-        self._setup_prefix()
-
-
-class ImpWrapper:
-    """PEP 302 Importer that wraps Python's "normal" import algorithm"""
-
-    def __init__(self, path=None):
-        self.path = path
-
-    def find_module(self, fullname, path=None):
-        subname = fullname.split(".")[-1]
-        if subname != fullname and self.path is None:
-            return None
-        if self.path is None:
-            path = None
-        else:
-            path = [self.path]
-        try:
-            file, filename, etc = imp.find_module(subname, path)
-        except ImportError:
-            return None
-        return ImpLoader(file, filename, etc)
-
-
-class ImpLoader:
-    """PEP 302 Loader that wraps Python's "normal" import algorithm"""
-
-    def __init__(self, file, filename, etc):
-        self.file = file
-        self.filename = filename
-        self.etc = etc
-
-    def load_module(self, fullname):
-        try:
-            mod = imp.load_module(fullname, self.file, self.filename, self.etc)
-        finally:
-            if self.file: self.file.close()
-        # Note: we don't set __loader__ because we want the module to look
-        # normal; i.e. this is just a wrapper for standard import machinery
-        return mod
-
-
-
-
-def get_importer(path_item):
-    """Retrieve a PEP 302 "importer" for the given path item
-
-    If there is no importer, this returns a wrapper around the builtin import
-    machinery.  The returned importer is only cached if it was created by a
-    path hook.
-    """
-    try:
-        importer = sys.path_importer_cache[path_item]
-    except KeyError:
-        for hook in sys.path_hooks:
-            try:
-                importer = hook(path_item)
-            except ImportError:
-                pass
-            else:
-                break
-        else:
-            importer = None
-
-    sys.path_importer_cache.setdefault(path_item,importer)
-    if importer is None:
-        try:
-            importer = ImpWrapper(path_item)
-        except ImportError:
-            pass
-    return importer
-
-try:
-    from pkgutil import get_importer, ImpImporter
-except ImportError:
-    pass    # Python 2.3 or 2.4, use our own implementation
-else:
-    ImpWrapper = ImpImporter    # Python 2.5, use pkgutil's implementation
-    del ImpLoader, ImpImporter
-
-
-
-
-
-
-_distribution_finders = {}
-
-def register_finder(importer_type, distribution_finder):
-    """Register `distribution_finder` to find distributions in sys.path items
-
-    `importer_type` is the type or class of a PEP 302 "Importer" (sys.path item
-    handler), and `distribution_finder` is a callable that, passed a path
-    item and the importer instance, yields ``Distribution`` instances found on
-    that path item.  See ``pkg_resources.find_on_path`` for an example."""
-    _distribution_finders[importer_type] = distribution_finder
-
-
-def find_distributions(path_item, only=False):
-    """Yield distributions accessible via `path_item`"""
-    importer = get_importer(path_item)
-    finder = _find_adapter(_distribution_finders, importer)
-    return finder(importer, path_item, only)
-
-def find_in_zip(importer, path_item, only=False):
-    metadata = EggMetadata(importer)
-    if metadata.has_metadata('PKG-INFO'):
-        yield Distribution.from_filename(path_item, metadata=metadata)
-    if only:
-        return  # don't yield nested distros
-    for subitem in metadata.resource_listdir('/'):
-        if subitem.endswith('.egg'):
-            subpath = os.path.join(path_item, subitem)
-            for dist in find_in_zip(zipimport.zipimporter(subpath), subpath):
-                yield dist
-
-register_finder(zipimport.zipimporter, find_in_zip)
-
-def StringIO(*args, **kw):
-    """Thunk to load the real StringIO on demand"""
-    global StringIO
-    try:
-        from cStringIO import StringIO
-    except ImportError:
-        from StringIO import StringIO
-    return StringIO(*args,**kw)
-
-def find_nothing(importer, path_item, only=False):
-    return ()
-register_finder(object,find_nothing)
-
-def find_on_path(importer, path_item, only=False):
-    """Yield distributions accessible on a sys.path directory"""
-    path_item = _normalize_cached(path_item)
-
-    if os.path.isdir(path_item) and os.access(path_item, os.R_OK):
-        if path_item.lower().endswith('.egg'):
-            # unpacked egg
-            yield Distribution.from_filename(
-                path_item, metadata=PathMetadata(
-                    path_item, os.path.join(path_item,'EGG-INFO')
-                )
-            )
-        else:
-            # scan for .egg and .egg-info in directory
-            for entry in os.listdir(path_item):
-                lower = entry.lower()
-                if lower.endswith('.egg-info'):
-                    fullpath = os.path.join(path_item, entry)
-                    if os.path.isdir(fullpath):
-                        # egg-info directory, allow getting metadata
-                        metadata = PathMetadata(path_item, fullpath)
-                    else:
-                        metadata = FileMetadata(fullpath)
-                    yield Distribution.from_location(
-                        path_item,entry,metadata,precedence=DEVELOP_DIST
-                    )
-                elif not only and lower.endswith('.egg'):
-                    for dist in find_distributions(os.path.join(path_item, entry)):
-                        yield dist
-                elif not only and lower.endswith('.egg-link'):
-                    for line in open(os.path.join(path_item, entry)):
-                        if not line.strip(): continue
-                        for item in find_distributions(os.path.join(path_item,line.rstrip())):
-                            yield item
-                        break
-register_finder(ImpWrapper,find_on_path)
-
-_namespace_handlers = {}
-_namespace_packages = {}
-
-def register_namespace_handler(importer_type, namespace_handler):
-    """Register `namespace_handler` to declare namespace packages
-
-    `importer_type` is the type or class of a PEP 302 "Importer" (sys.path item
-    handler), and `namespace_handler` is a callable like this::
-
-        def namespace_handler(importer,path_entry,moduleName,module):
-            # return a path_entry to use for child packages
-
-    Namespace handlers are only called if the importer object has already
-    agreed that it can handle the relevant path item, and they should only
-    return a subpath if the module __path__ does not already contain an
-    equivalent subpath.  For an example namespace handler, see
-    ``pkg_resources.file_ns_handler``.
-    """
-    _namespace_handlers[importer_type] = namespace_handler
-
-def _handle_ns(packageName, path_item):
-    """Ensure that named package includes a subpath of path_item (if needed)"""
-    importer = get_importer(path_item)
-    if importer is None:
-        return None
-    loader = importer.find_module(packageName)
-    if loader is None:
-        return None
-    module = sys.modules.get(packageName)
-    if module is None:
-        module = sys.modules[packageName] = types.ModuleType(packageName)
-        module.__path__ = []; _set_parent_ns(packageName)
-    elif not hasattr(module,'__path__'):
-        raise TypeError("Not a package:", packageName)
-    handler = _find_adapter(_namespace_handlers, importer)
-    subpath = handler(importer,path_item,packageName,module)
-    if subpath is not None:
-        path = module.__path__; path.append(subpath)
-        loader.load_module(packageName); module.__path__ = path
-    return subpath
-
-def declare_namespace(packageName):
-    """Declare that package 'packageName' is a namespace package"""
-
-    imp.acquire_lock()
-    try:
-        if packageName in _namespace_packages:
-            return
-
-        path, parent = sys.path, None
-        if '.' in packageName:
-            parent = '.'.join(packageName.split('.')[:-1])
-            declare_namespace(parent)
-            __import__(parent)
-            try:
-                path = sys.modules[parent].__path__
-            except AttributeError:
-                raise TypeError("Not a package:", parent)
-
-        # Track what packages are namespaces, so when new path items are added,
-        # they can be updated
-        _namespace_packages.setdefault(parent,[]).append(packageName)
-        _namespace_packages.setdefault(packageName,[])
-
-        for path_item in path:
-            # Ensure all the parent's path items are reflected in the child,
-            # if they apply
-            _handle_ns(packageName, path_item)
-
-    finally:
-        imp.release_lock()
-
-def fixup_namespace_packages(path_item, parent=None):
-    """Ensure that previously-declared namespace packages include path_item"""
-    imp.acquire_lock()
-    try:
-        for package in _namespace_packages.get(parent,()):
-            subpath = _handle_ns(package, path_item)
-            if subpath: fixup_namespace_packages(subpath,package)
-    finally:
-        imp.release_lock()
-
-def file_ns_handler(importer, path_item, packageName, module):
-    """Compute an ns-package subpath for a filesystem or zipfile importer"""
-
-    subpath = os.path.join(path_item, packageName.split('.')[-1])
-    normalized = _normalize_cached(subpath)
-    for item in module.__path__:
-        if _normalize_cached(item)==normalized:
-            break
-    else:
-        # Only return the path if it's not already there
-        return subpath
-
-register_namespace_handler(ImpWrapper,file_ns_handler)
-register_namespace_handler(zipimport.zipimporter,file_ns_handler)
-
-
-def null_ns_handler(importer, path_item, packageName, module):
-    return None
-
-register_namespace_handler(object,null_ns_handler)
-
-
-def normalize_path(filename):
-    """Normalize a file/dir name for comparison purposes"""
-    return os.path.normcase(os.path.realpath(filename))
-
-def _normalize_cached(filename,_cache={}):
-    try:
-        return _cache[filename]
-    except KeyError:
-        _cache[filename] = result = normalize_path(filename)
-        return result
-
-def _set_parent_ns(packageName):
-    parts = packageName.split('.')
-    name = parts.pop()
-    if parts:
-        parent = '.'.join(parts)
-        setattr(sys.modules[parent], name, sys.modules[packageName])
-
-
-def yield_lines(strs):
-    """Yield non-empty/non-comment lines of a ``basestring`` or sequence"""
-    if isinstance(strs,basestring):
-        for s in strs.splitlines():
-            s = s.strip()
-            if s and not s.startswith('#'):     # skip blank lines/comments
-                yield s
-    else:
-        for ss in strs:
-            for s in yield_lines(ss):
-                yield s
-
-LINE_END = re.compile(r"\s*(#.*)?$").match         # whitespace and comment
-CONTINUE = re.compile(r"\s*\\\s*(#.*)?$").match    # line continuation
-DISTRO   = re.compile(r"\s*((\w|[-.])+)").match    # Distribution or extra
-VERSION  = re.compile(r"\s*(<=?|>=?|==|!=)\s*((\w|[-.])+)").match  # ver. info
-COMMA    = re.compile(r"\s*,").match               # comma between items
-OBRACKET = re.compile(r"\s*\[").match
-CBRACKET = re.compile(r"\s*\]").match
-MODULE   = re.compile(r"\w+(\.\w+)*$").match
-EGG_NAME = re.compile(
-    r"(?P<name>[^-]+)"
-    r"( -(?P<ver>[^-]+) (-py(?P<pyver>[^-]+) (-(?P<plat>.+))? )? )?",
-    re.VERBOSE | re.IGNORECASE
-).match
-
-component_re = re.compile(r'(\d+ | [a-z]+ | \.| -)', re.VERBOSE)
-replace = {'pre':'c', 'preview':'c','-':'final-','rc':'c','dev':'@'}.get
-
-def _parse_version_parts(s):
-    for part in component_re.split(s):
-        part = replace(part,part)
-        if not part or part=='.':
-            continue
-        if part[:1] in '0123456789':
-            yield part.zfill(8)    # pad for numeric comparison
-        else:
-            yield '*'+part
-
-    yield '*final'  # ensure that alpha/beta/candidate are before final
-
-def parse_version(s):
-    """Convert a version string to a chronologically-sortable key
-
-    This is a rough cross between distutils' StrictVersion and LooseVersion;
-    if you give it versions that would work with StrictVersion, then it behaves
-    the same; otherwise it acts like a slightly-smarter LooseVersion. It is
-    *possible* to create pathological version coding schemes that will fool
-    this parser, but they should be very rare in practice.
-
-    The returned value will be a tuple of strings.  Numeric portions of the
-    version are padded to 8 digits so they will compare numerically, but
-    without relying on how numbers compare relative to strings.  Dots are
-    dropped, but dashes are retained.  Trailing zeros between alpha segments
-    or dashes are suppressed, so that e.g. "2.4.0" is considered the same as
-    "2.4". Alphanumeric parts are lower-cased.
-
-    The algorithm assumes that strings like "-" and any alpha string that
-    alphabetically follows "final"  represents a "patch level".  So, "2.4-1"
-    is assumed to be a branch or patch of "2.4", and therefore "2.4.1" is
-    considered newer than "2.4-1", which in turn is newer than "2.4".
-
-    Strings like "a", "b", "c", "alpha", "beta", "candidate" and so on (that
-    come before "final" alphabetically) are assumed to be pre-release versions,
-    so that the version "2.4" is considered newer than "2.4a1".
-
-    Finally, to handle miscellaneous cases, the strings "pre", "preview", and
-    "rc" are treated as if they were "c", i.e. as though they were release
-    candidates, and therefore are not as new as a version string that does not
-    contain them, and "dev" is replaced with an '@' so that it sorts lower than
-    than any other pre-release tag.
-    """
-    parts = []
-    for part in _parse_version_parts(s.lower()):
-        if part.startswith('*'):
-            if part<'*final':   # remove '-' before a prerelease tag
-                while parts and parts[-1]=='*final-': parts.pop()
-            # remove trailing zeros from each series of numeric parts
-            while parts and parts[-1]=='00000000':
-                parts.pop()
-        parts.append(part)
-    return tuple(parts)
-
-class EntryPoint(object):
-    """Object representing an advertised importable object"""
-
-    def __init__(self, name, module_name, attrs=(), extras=(), dist=None):
-        if not MODULE(module_name):
-            raise ValueError("Invalid module name", module_name)
-        self.name = name
-        self.module_name = module_name
-        self.attrs = tuple(attrs)
-        self.extras = Requirement.parse(("x[%s]" % ','.join(extras))).extras
-        self.dist = dist
-
-    def __str__(self):
-        s = "%s = %s" % (self.name, self.module_name)
-        if self.attrs:
-            s += ':' + '.'.join(self.attrs)
-        if self.extras:
-            s += ' [%s]' % ','.join(self.extras)
-        return s
-
-    def __repr__(self):
-        return "EntryPoint.parse(%r)" % str(self)
-
-    def load(self, require=True, env=None, installer=None):
-        if require: self.require(env, installer)
-        entry = __import__(self.module_name, globals(),globals(), ['__name__'])
-        for attr in self.attrs:
-            try:
-                entry = getattr(entry,attr)
-            except AttributeError:
-                raise ImportError("%r has no %r attribute" % (entry,attr))
-        return entry
-
-    def require(self, env=None, installer=None):
-        if self.extras and not self.dist:
-            raise UnknownExtra("Can't require() without a distribution", self)
-        map(working_set.add,
-            working_set.resolve(self.dist.requires(self.extras),env,installer))
-
-
-
-    #@classmethod
-    def parse(cls, src, dist=None):
-        """Parse a single entry point from string `src`
-
-        Entry point syntax follows the form::
-
-            name = some.module:some.attr [extra1,extra2]
-
-        The entry name and module name are required, but the ``:attrs`` and
-        ``[extras]`` parts are optional
-        """
-        try:
-            attrs = extras = ()
-            name,value = src.split('=',1)
-            if '[' in value:
-                value,extras = value.split('[',1)
-                req = Requirement.parse("x["+extras)
-                if req.specs: raise ValueError
-                extras = req.extras
-            if ':' in value:
-                value,attrs = value.split(':',1)
-                if not MODULE(attrs.rstrip()):
-                    raise ValueError
-                attrs = attrs.rstrip().split('.')
-        except ValueError:
-            raise ValueError(
-                "EntryPoint must be in 'name=module:attrs [extras]' format",
-                src
-            )
-        else:
-            return cls(name.strip(), value.strip(), attrs, extras, dist)
-
-    parse = classmethod(parse)
-
-
-
-
-
-
-
-
-    #@classmethod
-    def parse_group(cls, group, lines, dist=None):
-        """Parse an entry point group"""
-        if not MODULE(group):
-            raise ValueError("Invalid group name", group)
-        this = {}
-        for line in yield_lines(lines):
-            ep = cls.parse(line, dist)
-            if ep.name in this:
-                raise ValueError("Duplicate entry point", group, ep.name)
-            this[ep.name]=ep
-        return this
-
-    parse_group = classmethod(parse_group)
-
-    #@classmethod
-    def parse_map(cls, data, dist=None):
-        """Parse a map of entry point groups"""
-        if isinstance(data,dict):
-            data = data.items()
-        else:
-            data = split_sections(data)
-        maps = {}
-        for group, lines in data:
-            if group is None:
-                if not lines:
-                    continue
-                raise ValueError("Entry points must be listed in groups")
-            group = group.strip()
-            if group in maps:
-                raise ValueError("Duplicate group name", group)
-            maps[group] = cls.parse_group(group, lines, dist)
-        return maps
-
-    parse_map = classmethod(parse_map)
-
-
-def _remove_md5_fragment(location):
-    if not location:
-        return ''
-    parsed = urlparse(location)
-    if parsed[-1].startswith('md5='):
-        return urlunparse(parsed[:-1] + ('',))
-    return location
-
-
-class Distribution(object):
-    """Wrap an actual or potential sys.path entry w/metadata"""
-    def __init__(self,
-        location=None, metadata=None, project_name=None, version=None,
-        py_version=PY_MAJOR, platform=None, precedence = EGG_DIST
-    ):
-        self.project_name = safe_name(project_name or 'Unknown')
-        if version is not None:
-            self._version = safe_version(version)
-        self.py_version = py_version
-        self.platform = platform
-        self.location = location
-        self.precedence = precedence
-        self._provider = metadata or empty_provider
-
-    #@classmethod
-    def from_location(cls,location,basename,metadata=None,**kw):
-        project_name, version, py_version, platform = [None]*4
-        basename, ext = os.path.splitext(basename)
-        if ext.lower() in (".egg",".egg-info"):
-            match = EGG_NAME(basename)
-            if match:
-                project_name, version, py_version, platform = match.group(
-                    'name','ver','pyver','plat'
-                )
-        return cls(
-            location, metadata, project_name=project_name, version=version,
-            py_version=py_version, platform=platform, **kw
-        )
-    from_location = classmethod(from_location)
-
-
-    hashcmp = property(
-        lambda self: (
-            getattr(self,'parsed_version',()),
-            self.precedence,
-            self.key,
-            _remove_md5_fragment(self.location),
-            self.py_version,
-            self.platform
-        )
-    )
-    def __hash__(self): return hash(self.hashcmp)
-    def __lt__(self, other):
-        return self.hashcmp < other.hashcmp
-    def __le__(self, other):
-        return self.hashcmp <= other.hashcmp
-    def __gt__(self, other):
-        return self.hashcmp > other.hashcmp
-    def __ge__(self, other):
-        return self.hashcmp >= other.hashcmp
-    def __eq__(self, other):
-        if not isinstance(other, self.__class__):
-            # It's not a Distribution, so they are not equal
-            return False
-        return self.hashcmp == other.hashcmp
-    def __ne__(self, other):
-        return not self == other
-
-    # These properties have to be lazy so that we don't have to load any
-    # metadata until/unless it's actually needed.  (i.e., some distributions
-    # may not know their name or version without loading PKG-INFO)
-
-    #@property
-    def key(self):
-        try:
-            return self._key
-        except AttributeError:
-            self._key = key = self.project_name.lower()
-            return key
-    key = property(key)
-
-    #@property
-    def parsed_version(self):
-        try:
-            return self._parsed_version
-        except AttributeError:
-            self._parsed_version = pv = parse_version(self.version)
-            return pv
-
-    parsed_version = property(parsed_version)
-
-    #@property
-    def version(self):
-        try:
-            return self._version
-        except AttributeError:
-            for line in self._get_metadata('PKG-INFO'):
-                if line.lower().startswith('version:'):
-                    self._version = safe_version(line.split(':',1)[1].strip())
-                    return self._version
-            else:
-                raise ValueError(
-                    "Missing 'Version:' header and/or PKG-INFO file", self
-                )
-    version = property(version)
-
-
-
-
-    #@property
-    def _dep_map(self):
-        try:
-            return self.__dep_map
-        except AttributeError:
-            dm = self.__dep_map = {None: []}
-            for name in 'requires.txt', 'depends.txt':
-                for extra,reqs in split_sections(self._get_metadata(name)):
-                    if extra: extra = safe_extra(extra)
-                    dm.setdefault(extra,[]).extend(parse_requirements(reqs))
-            return dm
-    _dep_map = property(_dep_map)
-
-    def requires(self,extras=()):
-        """List of Requirements needed for this distro if `extras` are used"""
-        dm = self._dep_map
-        deps = []
-        deps.extend(dm.get(None,()))
-        for ext in extras:
-            try:
-                deps.extend(dm[safe_extra(ext)])
-            except KeyError:
-                raise UnknownExtra(
-                    "%s has no such extra feature %r" % (self, ext)
-                )
-        return deps
-
-    def _get_metadata(self,name):
-        if self.has_metadata(name):
-            for line in self.get_metadata_lines(name):
-                yield line
-
-    def activate(self,path=None):
-        """Ensure distribution is importable on `path` (default=sys.path)"""
-        if path is None: path = sys.path
-        self.insert_on(path)
-        if path is sys.path:
-            fixup_namespace_packages(self.location)
-            map(declare_namespace, self._get_metadata('namespace_packages.txt'))
-
-
-    def egg_name(self):
-        """Return what this distribution's standard .egg filename should be"""
-        filename = "%s-%s-py%s" % (
-            to_filename(self.project_name), to_filename(self.version),
-            self.py_version or PY_MAJOR
-        )
-
-        if self.platform:
-            filename += '-'+self.platform
-        return filename
-
-    def __repr__(self):
-        if self.location:
-            return "%s (%s)" % (self,self.location)
-        else:
-            return str(self)
-
-    def __str__(self):
-        try: version = getattr(self,'version',None)
-        except ValueError: version = None
-        version = version or "[unknown version]"
-        return "%s %s" % (self.project_name,version)
-
-    def __getattr__(self,attr):
-        """Delegate all unrecognized public attributes to .metadata provider"""
-        if attr.startswith('_'):
-            raise AttributeError,attr
-        return getattr(self._provider, attr)
-
-    #@classmethod
-    def from_filename(cls,filename,metadata=None, **kw):
-        return cls.from_location(
-            _normalize_cached(filename), os.path.basename(filename), metadata,
-            **kw
-        )
-    from_filename = classmethod(from_filename)
-
-    def as_requirement(self):
-        """Return a ``Requirement`` that matches this distribution exactly"""
-        return Requirement.parse('%s==%s' % (self.project_name, self.version))
-
-    def load_entry_point(self, group, name):
-        """Return the `name` entry point of `group` or raise ImportError"""
-        ep = self.get_entry_info(group,name)
-        if ep is None:
-            raise ImportError("Entry point %r not found" % ((group,name),))
-        return ep.load()
-
-    def get_entry_map(self, group=None):
-        """Return the entry point map for `group`, or the full entry map"""
-        try:
-            ep_map = self._ep_map
-        except AttributeError:
-            ep_map = self._ep_map = EntryPoint.parse_map(
-                self._get_metadata('entry_points.txt'), self
-            )
-        if group is not None:
-            return ep_map.get(group,{})
-        return ep_map
-
-    def get_entry_info(self, group, name):
-        """Return the EntryPoint object for `group`+`name`, or ``None``"""
-        return self.get_entry_map(group).get(name)
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-    def insert_on(self, path, loc = None):
-        """Insert self.location in path before its nearest parent directory"""
-
-        loc = loc or self.location
-
-        if self.project_name == 'setuptools':
-            try:
-                version = self.version
-            except ValueError:
-                version = ''
-            if '0.7' in version:
-                raise ValueError(
-                    "A 0.7-series setuptools cannot be installed "
-                    "with distribute. Found one at %s" % str(self.location))
-
-        if not loc:
-            return
-
-        if path is sys.path:
-            self.check_version_conflict()
-
-        nloc = _normalize_cached(loc)
-        bdir = os.path.dirname(nloc)
-        npath= map(_normalize_cached, path)
-
-        bp = None
-        for p, item in enumerate(npath):
-            if item==nloc:
-                break
-            elif item==bdir and self.precedence==EGG_DIST:
-                # if it's an .egg, give it precedence over its directory
-                path.insert(p, loc)
-                npath.insert(p, nloc)
-                break
-        else:
-            path.append(loc)
-            return
-
-        # p is the spot where we found or inserted loc; now remove duplicates
-        while 1:
-            try:
-                np = npath.index(nloc, p+1)
-            except ValueError:
-                break
-            else:
-                del npath[np], path[np]
-                p = np  # ha!
-
-        return
-
-
-
-    def check_version_conflict(self):
-        if self.key=='distribute':
-            return      # ignore the inevitable setuptools self-conflicts  :(
-
-        nsp = dict.fromkeys(self._get_metadata('namespace_packages.txt'))
-        loc = normalize_path(self.location)
-        for modname in self._get_metadata('top_level.txt'):
-            if (modname not in sys.modules or modname in nsp
-                or modname in _namespace_packages
-            ):
-                continue
-            if modname in ('pkg_resources', 'setuptools', 'site'):
-                continue
-            fn = getattr(sys.modules[modname], '__file__', None)
-            if fn and (normalize_path(fn).startswith(loc) or
-                       fn.startswith(self.location)):
-                continue
-            issue_warning(
-                "Module %s was already imported from %s, but %s is being added"
-                " to sys.path" % (modname, fn, self.location),
-            )
-
-    def has_version(self):
-        try:
-            self.version
-        except ValueError:
-            issue_warning("Unbuilt egg for "+repr(self))
-            return False
-        return True
-
-    def clone(self,**kw):
-        """Copy this distribution, substituting in any changed keyword args"""
-        for attr in (
-            'project_name', 'version', 'py_version', 'platform', 'location',
-            'precedence'
-        ):
-            kw.setdefault(attr, getattr(self,attr,None))
-        kw.setdefault('metadata', self._provider)
-        return self.__class__(**kw)
-
-
-
-
-    #@property
-    def extras(self):
-        return [dep for dep in self._dep_map if dep]
-    extras = property(extras)
-
-
-def issue_warning(*args,**kw):
-    level = 1
-    g = globals()
-    try:
-        # find the first stack frame that is *not* code in
-        # the pkg_resources module, to use for the warning
-        while sys._getframe(level).f_globals is g:
-            level += 1
-    except ValueError:
-        pass
-    from warnings import warn
-    warn(stacklevel = level+1, *args, **kw)
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-def parse_requirements(strs):
-    """Yield ``Requirement`` objects for each specification in `strs`
-
-    `strs` must be an instance of ``basestring``, or a (possibly-nested)
-    iterable thereof.
-    """
-    # create a steppable iterator, so we can handle \-continuations
-    lines = iter(yield_lines(strs))
-
-    def scan_list(ITEM,TERMINATOR,line,p,groups,item_name):
-
-        items = []
-
-        while not TERMINATOR(line,p):
-            if CONTINUE(line,p):
-                try:
-                    line = lines.next(); p = 0
-                except StopIteration:
-                    raise ValueError(
-                        "\\ must not appear on the last nonblank line"
-                    )
-
-            match = ITEM(line,p)
-            if not match:
-                raise ValueError("Expected "+item_name+" in",line,"at",line[p:])
-
-            items.append(match.group(*groups))
-            p = match.end()
-
-            match = COMMA(line,p)
-            if match:
-                p = match.end() # skip the comma
-            elif not TERMINATOR(line,p):
-                raise ValueError(
-                    "Expected ',' or end-of-list in",line,"at",line[p:]
-                )
-
-        match = TERMINATOR(line,p)
-        if match: p = match.end()   # skip the terminator, if any
-        return line, p, items
-
-    for line in lines:
-        match = DISTRO(line)
-        if not match:
-            raise ValueError("Missing distribution spec", line)
-        project_name = match.group(1)
-        p = match.end()
-        extras = []
-
-        match = OBRACKET(line,p)
-        if match:
-            p = match.end()
-            line, p, extras = scan_list(
-                DISTRO, CBRACKET, line, p, (1,), "'extra' name"
-            )
-
-        line, p, specs = scan_list(VERSION,LINE_END,line,p,(1,2),"version spec")
-        specs = [(op,safe_version(val)) for op,val in specs]
-        yield Requirement(project_name, specs, extras)
-
-
-def _sort_dists(dists):
-    tmp = [(dist.hashcmp,dist) for dist in dists]
-    tmp.sort()
-    dists[::-1] = [d for hc,d in tmp]
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-class Requirement:
-    def __init__(self, project_name, specs, extras):
-        """DO NOT CALL THIS UNDOCUMENTED METHOD; use Requirement.parse()!"""
-        self.unsafe_name, project_name = project_name, safe_name(project_name)
-        self.project_name, self.key = project_name, project_name.lower()
-        index = [(parse_version(v),state_machine[op],op,v) for op,v in specs]
-        index.sort()
-        self.specs = [(op,ver) for parsed,trans,op,ver in index]
-        self.index, self.extras = index, tuple(map(safe_extra,extras))
-        self.hashCmp = (
-            self.key, tuple([(op,parsed) for parsed,trans,op,ver in index]),
-            frozenset(self.extras)
-        )
-        self.__hash = hash(self.hashCmp)
-
-    def __str__(self):
-        specs = ','.join([''.join(s) for s in self.specs])
-        extras = ','.join(self.extras)
-        if extras: extras = '[%s]' % extras
-        return '%s%s%s' % (self.project_name, extras, specs)
-
-    def __eq__(self,other):
-        return isinstance(other,Requirement) and self.hashCmp==other.hashCmp
-
-    def __contains__(self,item):
-        if isinstance(item,Distribution):
-            if item.key <> self.key: return False
-            if self.index: item = item.parsed_version  # only get if we need it
-        elif isinstance(item,basestring):
-            item = parse_version(item)
-        last = None
-        compare = lambda a, b: (a > b) - (a < b) # -1, 0, 1
-        for parsed,trans,op,ver in self.index:
-            action = trans[compare(item,parsed)] # Indexing: 0, 1, -1
-            if action=='F':     return False
-            elif action=='T':   return True
-            elif action=='+':   last = True
-            elif action=='-' or last is None:   last = False
-        if last is None: last = True    # no rules encountered
-        return last
-
-
-    def __hash__(self):
-        return self.__hash
-
-    def __repr__(self): return "Requirement.parse(%r)" % str(self)
-
-    #@staticmethod
-    def parse(s, replacement=True):
-        reqs = list(parse_requirements(s))
-        if reqs:
-            if len(reqs) == 1:
-                founded_req = reqs[0]
-                # if asked for setuptools distribution
-                # and if distribute is installed, we want to give
-                # distribute instead
-                if _override_setuptools(founded_req) and replacement:
-                    distribute = list(parse_requirements('distribute'))
-                    if len(distribute) == 1:
-                        return distribute[0]
-                    return founded_req
-                else:
-                    return founded_req
-
-            raise ValueError("Expected only one requirement", s)
-        raise ValueError("No requirements found", s)
-
-    parse = staticmethod(parse)
-
-state_machine = {
-    #       =><
-    '<' :  '--T',
-    '<=':  'T-T',
-    '>' :  'F+F',
-    '>=':  'T+F',
-    '==':  'T..',
-    '!=':  'F++',
-}
-
-
-def _override_setuptools(req):
-    """Return True when distribute wants to override a setuptools dependency.
-
-    We want to override when the requirement is setuptools and the version is
-    a variant of 0.6.
-
-    """
-    if req.project_name == 'setuptools':
-        if not len(req.specs):
-            # Just setuptools: ok
-            return True
-        for comparator, version in req.specs:
-            if comparator in ['==', '>=', '>']:
-                if '0.7' in version:
-                    # We want some setuptools not from the 0.6 series.
-                    return False
-        return True
-    return False
-
-
-def _get_mro(cls):
-    """Get an mro for a type or classic class"""
-    if not isinstance(cls,type):
-        class cls(cls,object): pass
-        return cls.__mro__[1:]
-    return cls.__mro__
-
-def _find_adapter(registry, ob):
-    """Return an adapter factory for `ob` from `registry`"""
-    for t in _get_mro(getattr(ob, '__class__', type(ob))):
-        if t in registry:
-            return registry[t]
-
-
-def ensure_directory(path):
-    """Ensure that the parent directory of `path` exists"""
-    dirname = os.path.dirname(path)
-    if not os.path.isdir(dirname):
-        os.makedirs(dirname)
-
-def split_sections(s):
-    """Split a string or iterable thereof into (section,content) pairs
-
-    Each ``section`` is a stripped version of the section header ("[section]")
-    and each ``content`` is a list of stripped lines excluding blank lines and
-    comment-only lines.  If there are any such lines before the first section
-    header, they're returned in a first ``section`` of ``None``.
-    """
-    section = None
-    content = []
-    for line in yield_lines(s):
-        if line.startswith("["):
-            if line.endswith("]"):
-                if section or content:
-                    yield section, content
-                section = line[1:-1].strip()
-                content = []
-            else:
-                raise ValueError("Invalid section heading", line)
-        else:
-            content.append(line)
-
-    # wrap up last segment
-    yield section, content
-
-def _mkstemp(*args,**kw):
-    from tempfile import mkstemp
-    old_open = os.open
-    try:
-        os.open = os_open   # temporarily bypass sandboxing
-        return mkstemp(*args,**kw)
-    finally:
-        os.open = old_open  # and then put it back
-
-
-# Set up global resource manager
-_manager = ResourceManager()
-def _initialize(g):
-    for name in dir(_manager):
-        if not name.startswith('_'):
-            g[name] = getattr(_manager, name)
-_initialize(globals())
-
-# Prepare the master working set and make the ``require()`` API available
-working_set = WorkingSet()
-try:
-    # Does the main program list any requirements?
-    from __main__ import __requires__
-except ImportError:
-    pass # No: just use the default working set based on sys.path
-else:
-    # Yes: ensure the requirements are met, by prefixing sys.path if necessary
-    try:
-        working_set.require(__requires__)
-    except VersionConflict:     # try it without defaults already on sys.path
-        working_set = WorkingSet([])    # by starting with an empty path
-        for dist in working_set.resolve(
-            parse_requirements(__requires__), Environment()
-        ):
-            working_set.add(dist)
-        for entry in sys.path:  # add any missing entries from sys.path
-            if entry not in working_set.entries:
-                working_set.add_entry(entry)
-        sys.path[:] = working_set.entries   # then copy back to sys.path
-
-require = working_set.require
-iter_entry_points = working_set.iter_entry_points
-add_activation_listener = working_set.subscribe
-run_script = working_set.run_script
-run_main = run_script   # backward compatibility
-# Activate all distributions already on sys.path, and ensure that
-# all distributions added to the working set in the future (e.g. by
-# calling ``require()``) will get activated as well.
-add_activation_listener(lambda dist: dist.activate())
-working_set.entries=[]; map(working_set.add_entry,sys.path) # match order
-

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: removes some trailing spaces
Message-ID: <E1OVLoW-0000km-Fy@dinsdale.python.org>

tarek.ziade pushed be0c887b2eb1 to distutils2:

http://hg.python.org/distutils2/rev/be0c887b2eb1
changeset:   330:be0c887b2eb1
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Thu Jul 01 13:37:19 2010 +0200
summary:     removes some trailing spaces
files:       src/distutils2/pypi/dist.py, src/distutils2/pypi/simple.py, src/distutils2/tests/test_pypi_dist.py, src/distutils2/tests/test_pypi_simple.py

diff --git a/src/distutils2/pypi/dist.py b/src/distutils2/pypi/dist.py
--- a/src/distutils2/pypi/dist.py
+++ b/src/distutils2/pypi/dist.py
@@ -34,8 +34,8 @@
         """Build a Distribution from a url archive (egg or zip or tgz).
 
         :param url: complete url of the distribution
-        :param probable_dist_name: A probable name of the distribution. 
-        :param is_external: Tell if the url commes from an index or from 
+        :param probable_dist_name: A probable name of the distribution.
+        :param is_external: Tell if the url commes from an index or from
                             an external URL.
         """
         # if the url contains a md5 hash, get it.
@@ -57,10 +57,10 @@
 
         name, version = split_archive_name(archive_name)
         if extension_matched is True:
-            return PyPIDistribution(name, version, url=url, hashname="md5", 
+            return PyPIDistribution(name, version, url=url, hashname="md5",
                                     hashval=md5_hash, is_external=is_external)
 
-    def __init__(self, name, version, type=None, url=None, hashname=None, 
+    def __init__(self, name, version, type=None, url=None, hashname=None,
                  hashval=None, is_external=True):
         """Create a new instance of PyPIDistribution.
 
@@ -71,7 +71,7 @@
         :param hashname: the name of the hash we want to use. Refer to the
                          hashlib.new documentation for more information.
         :param hashval: the hash value.
-        :param is_external: we need to know if the provided url comes from an 
+        :param is_external: we need to know if the provided url comes from an
                             index browsing, or from an external resource.
 
         """
@@ -84,13 +84,13 @@
         # We store urls in dict, because we need to have a bit more informations
         # than the simple URL. It will be used later to find the good url to
         # use.
-        # We have two _url* attributes: _url and _urls. _urls contains a list of 
-        # dict for the different urls, and _url contains the choosen url, in 
+        # We have two _url* attributes: _url and _urls. _urls contains a list of
+        # dict for the different urls, and _url contains the choosen url, in
         # order to dont make the selection process multiple times.
         self._urls = []
         self._url = None
         self.add_url(url, hashname, hashval, is_external)
-    
+
     def add_url(self, url, hashname=None, hashval=None, is_external=True):
         """Add a new url to the list of urls"""
         if hashname is not None:
@@ -118,7 +118,7 @@
             if len(self._urls) > 1:
                 internals_urls = [u for u in self._urls \
                                   if u['is_external'] == False]
-                if len(internals_urls) >= 1: 
+                if len(internals_urls) >= 1:
                     self._url = internals_urls[0]
             if self._url is None:
                 self._url = self._urls[0]
@@ -136,7 +136,7 @@
         if self.location is None:
             url = self.url['url']
             archive_name = urlparse.urlparse(url)[2].split('/')[-1]
-            filename, headers = urllib.urlretrieve(url, 
+            filename, headers = urllib.urlretrieve(url,
                                                    path + "/" + archive_name)
             self.location = filename
             self._check_md5(filename)
@@ -220,7 +220,7 @@
 
     def append(self, o):
         """Append a new distribution to the list.
-        
+
         If a distribution with the same name and version exists, just grab the
         URL informations and add a new new url for the existing one.
         """
@@ -231,7 +231,7 @@
             dist.add_url(**o.url)
         else:
             super(PyPIDistributions, self).append(o)
-        
+
 
 def split_archive_name(archive_name, probable_name=None):
     """Split an archive name into two parts: name and version.
diff --git a/src/distutils2/pypi/simple.py b/src/distutils2/pypi/simple.py
--- a/src/distutils2/pypi/simple.py
+++ b/src/distutils2/pypi/simple.py
@@ -21,7 +21,7 @@
 
 # -- Constants -----------------------------------------------
 PYPI_DEFAULT_INDEX_URL = "http://pypi.python.org/simple/"
-DEFAULT_HOSTS = ("*",)  
+DEFAULT_HOSTS = ("*",)
 SOCKET_TIMEOUT = 15
 USER_AGENT = "Python-urllib/%s distutils2/%s" % (
     sys.version[:3], __distutils2_version__)
@@ -66,10 +66,10 @@
         """Class constructor.
 
         :param index_url: the url of the simple index to search on.
-        :param follow_externals: tell if following external links is needed or 
+        :param follow_externals: tell if following external links is needed or
                                  not. Default is False.
         :param hosts: a list of hosts allowed to be processed while using
-                      follow_externals=True. Default behavior is to follow all 
+                      follow_externals=True. Default behavior is to follow all
                       hosts.
         :param mirrors: a list of mirrors to check out if problems occurs while
                         working with the one given in "url"
@@ -107,8 +107,8 @@
         requirements.
 
         :param requirements: A project name and it's distribution, using
-                             version specifiers, as described in PEP345. 
-        :type requirements:  You can pass either a version.VersionPredicate 
+                             version specifiers, as described in PEP345.
+        :type requirements:  You can pass either a version.VersionPredicate
                              or a string.
         """
         requirements = self._get_version_predicate(requirements)
@@ -134,7 +134,7 @@
 
         Returns the complete absolute path to the downloaded archive.
 
-        :param requirements: The same as the find attribute of `find`. 
+        :param requirements: The same as the find attribute of `find`.
         """
         return self.get(requirements).download(path=temp_path)
 
@@ -145,10 +145,10 @@
         if isinstance(requirements, str):
             requirements = VersionPredicate(requirements)
         return requirements
-    
+
     @property
     def index_url(self):
-        return self._index_urls[self._current_index_url] 
+        return self._index_urls[self._current_index_url]
 
     def _switch_to_next_mirror(self):
         """Switch to the next mirror (eg. point self.index_url to the next
@@ -160,7 +160,7 @@
             self._current_index_url = self._current_index_url + 1
         else:
             raise UnableToDownload("All mirrors fails")
-    
+
     def _is_browsable(self, url):
         """Tell if the given URL can be browsed or not.
 
@@ -207,14 +207,14 @@
     def _process_url(self, url, project_name=None, follow_links=True):
         """Process an url and search for distributions packages.
 
-        For each URL found, if it's a download, creates a PyPIdistribution 
+        For each URL found, if it's a download, creates a PyPIdistribution
         object. If it's a homepage and we can follow links, process it too.
 
         :param url: the url to process
         :param project_name: the project name we are searching for.
         :param follow_links: Do not want to follow links more than from one
-                             level. This parameter tells if we want to follow 
-                             the links we find (eg. run recursively this 
+                             level. This parameter tells if we want to follow
+                             the links we find (eg. run recursively this
                              method on it)
         """
         f = self._open_url(url)
@@ -233,7 +233,7 @@
                         if self._is_browsable(link) and follow_links:
                             self._process_url(link, project_name,
                                 follow_links=False)
-    
+
     def _get_link_matcher(self, url):
         """Returns the right link matcher function of the given url
         """
@@ -258,7 +258,7 @@
                                            self._htmldecode(match.group(1)))
                     if 'download' in rels or self._is_browsable(url):
                         # yield a list of (url, is_download)
-                        yield (urlparse.urljoin(base_url, url), 
+                        yield (urlparse.urljoin(base_url, url),
                                'download' in rels)
 
     def _default_link_matcher(self, content, base_url):
@@ -279,7 +279,7 @@
             url = self.index_url + name + "/"
             self._process_url(url, name)
         except DownloadError:
-            # if an error occurs, try with the next index_url 
+            # if an error occurs, try with the next index_url
             # (provided by the mirrors)
             self._switch_to_next_mirror()
             self._distributions.clear()
diff --git a/src/distutils2/tests/test_pypi_dist.py b/src/distutils2/tests/test_pypi_dist.py
--- a/src/distutils2/tests/test_pypi_dist.py
+++ b/src/distutils2/tests/test_pypi_dist.py
@@ -145,7 +145,7 @@
         # should be ok
         Dist("FooBar", "0.1", hashname="md5", hashval="value")
 
-        self.assertRaises(UnsupportedHashName, Dist, "FooBar", "0.1", 
+        self.assertRaises(UnsupportedHashName, Dist, "FooBar", "0.1",
                           hashname="invalid_hashname", hashval="value")
 
 
diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -136,7 +136,7 @@
         self.assertTrue(index._is_browsable("pypi.test.tld/a/path"))
 
         # specify a list of hosts we want to allow
-        index = simple.SimpleIndex(follow_externals=True, 
+        index = simple.SimpleIndex(follow_externals=True,
                                    hosts=("*.test.tld",))
         self.assertFalse(index._is_browsable("http://an-external.link/path"))
         self.assertTrue(index._is_browsable("http://pypi.test.tld/a/path"))
@@ -209,13 +209,13 @@
         # process the pages
         index = self._get_simple_index(server, follow_externals=True)
         index.find("foobar")
-        # now it should have processed only pages with links rel="download" 
+        # now it should have processed only pages with links rel="download"
         # and rel="homepage"
-        self.assertIn("%s/simple/foobar/" % server.full_address, 
+        self.assertIn("%s/simple/foobar/" % server.full_address,
             index._processed_urls)  # it's the simple index page
-        self.assertIn("%s/external/homepage.html" % server.full_address, 
+        self.assertIn("%s/external/homepage.html" % server.full_address,
             index._processed_urls)  # the external homepage is rel="homepage"
-        self.assertNotIn("%s/external/nonrel.html" % server.full_address, 
+        self.assertNotIn("%s/external/nonrel.html" % server.full_address,
             index._processed_urls)  # this link contains no rel=*
         self.assertNotIn("%s/unrelated-0.2.tar.gz" % server.full_address,
             index._processed_urls)  # linked from simple index (no rel)
@@ -229,22 +229,22 @@
         server = PyPIServer("foo_bar_baz")
         mirror = PyPIServer("foo_bar_baz")
         mirror.start()  # we dont start the server here
-        
+
         try:
             # create the index using both servers
-            index = simple.SimpleIndex(server.full_address + "/simple/", 
+            index = simple.SimpleIndex(server.full_address + "/simple/",
                 hosts=('*',), timeout=1,  # set the timeout to 1s for the tests
                 mirrors=[mirror.full_address + "/simple/",])
-            
+
             # this should not raise a timeout
             self.assertEqual(4, len(index.find("foo")))
         finally:
             mirror.stop()
-    
+
     def test_simple_link_matcher(self):
         """Test that the simple link matcher yields the right links"""
         index = simple.SimpleIndex(follow_externals=False)
-        
+
         # Here, we define:
         #   1. one link that must be followed, cause it's a download one
         #   2. one link that must *not* be followed, cause the is_browsable
@@ -262,7 +262,7 @@
         # Test that the simple link matcher yield the good links.
         generator = index._simple_link_matcher(content, index.index_url)
         self.assertEqual(('http://dl-link1', True), generator.next())
-        self.assertEqual(('%stest' % index.index_url, False), 
+        self.assertEqual(('%stest' % index.index_url, False),
                          generator.next())
         self.assertRaises(StopIteration, generator.next)
 
@@ -271,7 +271,7 @@
         generator = index._simple_link_matcher(content, index.index_url)
         self.assertEqual(('http://dl-link1', True), generator.next())
         self.assertEqual(('http://dl-link2', False), generator.next())
-        self.assertEqual(('%stest' % index.index_url, False), 
+        self.assertEqual(('%stest' % index.index_url, False),
                          generator.next())
         self.assertRaises(StopIteration, generator.next)
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: Rename PyPIDistribution "location"
	to	"downloaded_location".
Message-ID: <E1OVLoW-0000ky-Hd@dinsdale.python.org>

tarek.ziade pushed 5b76ba38fecf to distutils2:

http://hg.python.org/distutils2/rev/5b76ba38fecf
changeset:   333:5b76ba38fecf
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Tue Jun 29 18:20:35 2010 +0200
summary:     Rename PyPIDistribution "location" to "downloaded_location".
files:       src/distutils2/pypi/dist.py

diff --git a/src/distutils2/pypi/dist.py b/src/distutils2/pypi/dist.py
--- a/src/distutils2/pypi/dist.py
+++ b/src/distutils2/pypi/dist.py
@@ -23,7 +23,7 @@
     """Represents a distribution retrieved from PyPI.
 
     This is a simple container for various attributes as name, version,
-    location, url etc.
+    downloaded_location, url etc.
 
     The PyPIDistribution class is used by the pypi.*Index class to return
     information about distributions.
@@ -79,7 +79,7 @@
         self.type = type
         # set the downloaded path to None by default. The goal here
         # is to not download distributions multiple times
-        self.location = None
+        self.downloaded_location = None
         # We store urls in dict, because we need to have a bit more informations
         # than the simple URL. It will be used later to find the good url to
         # use.
@@ -125,14 +125,14 @@
             path = tempfile.mkdtemp()
 
         # if we do not have downloaded it yet, do it.
-        if self.location is None:
+        if self.downloaded_location is None:
             url = self.url['url']
             archive_name = urlparse.urlparse(url)[2].split('/')[-1]
             filename, headers = urllib.urlretrieve(url, 
                                                    path + "/" + archive_name)
-            self.location = filename
+            self.downloaded_location = filename
             self._check_md5(filename)
-        return self.location
+        return self.downloaded_location
 
     def _check_md5(self, filename):
         """Check that the md5 checksum of the given file matches the one in

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: add support of mirrors for PyPI
	simple client (as	described in PEP 381)
Message-ID: <E1OVLoW-0000ku-H6@dinsdale.python.org>

tarek.ziade pushed c880d980acaa to distutils2:

http://hg.python.org/distutils2/rev/c880d980acaa
changeset:   332:c880d980acaa
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Tue Jun 29 18:14:40 2010 +0200
summary:     add support of mirrors for PyPI simple client (as described in PEP 381)
files:       src/distutils2/pypi/simple.py

diff --git a/src/distutils2/pypi/simple.py b/src/distutils2/pypi/simple.py
--- a/src/distutils2/pypi/simple.py
+++ b/src/distutils2/pypi/simple.py
@@ -21,6 +21,7 @@
 
 # -- Constants -----------------------------------------------
 PYPI_DEFAULT_INDEX_URL = "http://pypi.python.org/simple/"
+PYPI_DEFAULT_MIRROR_URL = "mirrors.pypi.python.org"
 DEFAULT_HOSTS = ("*",)  
 SOCKET_TIMEOUT = 15
 USER_AGENT = "Python-urllib/%s distutils2/%s" % (
@@ -62,22 +63,28 @@
     """
 
     def __init__(self, index_url=PYPI_DEFAULT_INDEX_URL, hosts=DEFAULT_HOSTS,
-                 mirrors=[], follow_externals=False, timeout=SOCKET_TIMEOUT):
+                 follow_externals=False, mirrors_url=PYPI_DEFAULT_MIRROR_URL, 
+                 mirrors=None, timeout=SOCKET_TIMEOUT):
         """Class constructor.
 
         :param index_url: the url of the simple index to search on.
-        :param follow_externals: tell if following external links is needed or 
-                                 not. Default is False.
         :param hosts: a list of hosts allowed to be processed while using
                       follow_externals=True. Default behavior is to follow all 
                       hosts.
-        :param mirrors: a list of mirrors to check out if problems occurs while
-                        working with the one given in "url"
+        :param follow_externals: tell if following external links is needed or 
+                                 not. Default is False.
+        :param mirrors_url: the url to look on for DNS records giving mirror
+                            adresses.
+        :param mirrors: a list of mirrors to check out if problems 
+                             occurs while working with the one given in "url"
         :param timeout: time in seconds to consider a url has timeouted.
         """
         if not index_url.endswith("/"):
             index_url += "/"
         self._index_urls = [index_url]
+        # if no mirrors are defined, use the method described in PEP 381.
+        if mirrors is None:
+            mirrors = socket.gethostbyname_ex(mirrors_url)[-1]
         self._index_urls.extend(mirrors)
         self._current_index_url = 0
         self._timeout = timeout

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: Fix crazy style in setup code copied
	for hashlib
Message-ID: <E1OVLoW-0000lH-Ly@dinsdale.python.org>

tarek.ziade pushed 7bef9ca33c8d to distutils2:

http://hg.python.org/distutils2/rev/7bef9ca33c8d
changeset:   338:7bef9ca33c8d
parent:      330:be0c887b2eb1
user:        ?ric Araujo <merwok at netwok.org>
date:        Thu Jul 01 19:31:07 2010 +0200
summary:     Fix crazy style in setup code copied for hashlib
files:       src/setup.py

diff --git a/src/setup.py b/src/setup.py
--- a/src/setup.py
+++ b/src/setup.py
@@ -67,14 +67,15 @@
 
 
 # additional paths to check, set from the command line
-SSL_INCDIR=''   # --openssl-incdir=
-SSL_LIBDIR=''   # --openssl-libdir=
-SSL_DIR=''      # --openssl-prefix=
+SSL_INCDIR = ''   # --openssl-incdir=
+SSL_LIBDIR = ''   # --openssl-libdir=
+SSL_DIR = ''      # --openssl-prefix=
 
 def add_dir_to_list(dirlist, dir):
     """Add the directory 'dir' to the list 'dirlist' (at the front) if
     'dir' actually exists and is a directory.  If 'dir' is already in
-    'dirlist' it is moved to the front."""
+    'dirlist' it is moved to the front.
+    """
     if dir is not None and os.path.isdir(dir) and dir not in dirlist:
         if dir in dirlist:
             dirlist.remove(dir)
@@ -83,8 +84,8 @@
 
 def prepare_hashlib_extensions():
     """Decide which C extensions to build and create the appropriate
-    Extension objects to build them.  Returns a list of Extensions."""
-
+    Extension objects to build them.  Return a list of Extensions.
+    """
     # this CCompiler object is only used to locate include files
     compiler = new_compiler()
 
@@ -116,13 +117,12 @@
             add_dir_to_list(compiler.library_dirs, os.path.join(SSL_DIR, 'lib'))
         add_dir_to_list(compiler.include_dirs, os.path.join(SSL_DIR, 'include'))
 
-    osNameLibsMap = {
-        'posix':  ['ssl', 'crypto'],
-        'nt':     ['libeay32',  'gdi32', 'advapi32', 'user32'],
-    }
+    oslibs = {'posix': ['ssl', 'crypto'],
+              'nt': ['libeay32',  'gdi32', 'advapi32', 'user32']}
 
-    if not osNameLibsMap.has_key(os.name):
-        print "unknown OS, please update setup.py"
+    if os.name not in oslibs:
+        sys.stderr.write(
+            'unknown operating system, impossible to compile _hashlib')
         sys.exit(1)
 
     exts = []
@@ -135,7 +135,7 @@
             ssl_incs.append(f)
             ssl_inc_dirs.append(inc_dir)
 
-    ssl_lib = compiler.find_library_file(compiler.library_dirs, osNameLibsMap[os.name][0])
+    ssl_lib = compiler.find_library_file(compiler.library_dirs, oslibs[os.name][0])
 
     # find out which version of OpenSSL we have
     openssl_ver = 0
@@ -159,9 +159,7 @@
         if openssl_ver:
             break
 
-    if (ssl_inc_dir and
-        ssl_lib is not None and
-        openssl_ver >= 0x00907000):
+    if (ssl_inc_dir and ssl_lib is not None and openssl_ver >= 0x00907000):
 
         print 'Using OpenSSL version 0x%08x from' % openssl_ver
         print ' Headers:\t', ssl_inc_dir
@@ -169,29 +167,29 @@
 
         # The _hashlib module wraps optimized implementations
         # of hash functions from the OpenSSL library.
-        exts.append( Extension('_hashlib', ['_hashopenssl.c'],
-                               include_dirs = [ ssl_inc_dir ],
-                               library_dirs = [ os.path.dirname(ssl_lib) ],
-                               libraries = osNameLibsMap[os.name]) )
+        exts.append(Extension('_hashlib', ['_hashopenssl.c'],
+                              include_dirs = [ssl_inc_dir],
+                              library_dirs = [os.path.dirname(ssl_lib)],
+                              libraries = oslibs[os.name]))
     else:
-        exts.append( Extension('_sha', ['shamodule.c']) )
-        exts.append( Extension('_md5',
-                        sources = ['md5module.c', 'md5.c'],
-                        depends = ['md5.h']) )
+        exts.append(Extension('_sha', ['shamodule.c']) )
+        exts.append(Extension('_md5',
+                              sources=['md5module.c', 'md5.c'],
+                              depends=['md5.h']) )
 
     if (not ssl_lib or openssl_ver < 0x00908000):
         # OpenSSL doesn't do these until 0.9.8 so we'll bring our own
-        exts.append( Extension('_sha256', ['sha256module.c']) )
-        exts.append( Extension('_sha512', ['sha512module.c']) )
+        exts.append(Extension('_sha256', ['sha256module.c']))
+        exts.append(Extension('_sha512', ['sha512module.c']))
 
-    def prependModules(filename):
+    def prepend_modules(filename):
         return os.path.join('Modules', filename)
 
     # all the C code is in the Modules subdirectory, prepend the path
     for ext in exts:
-        ext.sources = [ prependModules(fn) for fn in ext.sources ]
+        ext.sources = [prepend_modules(fn) for fn in ext.sources]
         if hasattr(ext, 'depends') and ext.depends is not None:
-            ext.depends = [ prependModules(fn) for fn in ext.depends ]
+            ext.depends = [prepend_modules(fn) for fn in ext.depends]
 
     return exts
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: PyPiDistribution raise an exception
	if hashname is	invalid.
Message-ID: <E1OVLoW-0000ki-FM@dinsdale.python.org>

tarek.ziade pushed 5a22adb09675 to distutils2:

http://hg.python.org/distutils2/rev/5a22adb09675
changeset:   329:5a22adb09675
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Thu Jul 01 12:59:32 2010 +0200
summary:     PyPiDistribution raise an exception if hashname is invalid.
files:       src/distutils2/pypi/dist.py, src/distutils2/pypi/errors.py, src/distutils2/tests/test_pypi_dist.py

diff --git a/src/distutils2/pypi/dist.py b/src/distutils2/pypi/dist.py
--- a/src/distutils2/pypi/dist.py
+++ b/src/distutils2/pypi/dist.py
@@ -11,8 +11,9 @@
     import hashlib
 except ImportError:
     from distutils2._backport import hashlib
+
 from distutils2.version import suggest_normalized_version
-from distutils2.pypi.errors import HashDoesNotMatch
+from distutils2.pypi.errors import HashDoesNotMatch, UnsupportedHashName
 
 EXTENSIONS = ".tar.gz .tar.bz2 .tar .zip .tgz .egg".split()
 MD5_HASH = re.compile(r'^.*#md5=([a-f0-9]+)$')
@@ -92,6 +93,12 @@
     
     def add_url(self, url, hashname=None, hashval=None, is_external=True):
         """Add a new url to the list of urls"""
+        if hashname is not None:
+            try:
+                hashlib.new(hashname)
+            except ValueError:
+                raise UnsupportedHashName(hashname)
+
         self._urls.append({
             'url': url,
             'hashname': hashname,
diff --git a/src/distutils2/pypi/errors.py b/src/distutils2/pypi/errors.py
--- a/src/distutils2/pypi/errors.py
+++ b/src/distutils2/pypi/errors.py
@@ -22,7 +22,11 @@
 
 
 class HashDoesNotMatch(DownloadError):
-    """Compared MD5 hashes does not match"""
+    """Compared hashes does not match"""
+
+
+class UnsupportedHashName(PyPIError):
+    """A unsupported hashname has been used"""
 
 
 class UnableToDownload(PyPIError):
diff --git a/src/distutils2/tests/test_pypi_dist.py b/src/distutils2/tests/test_pypi_dist.py
--- a/src/distutils2/tests/test_pypi_dist.py
+++ b/src/distutils2/tests/test_pypi_dist.py
@@ -7,7 +7,7 @@
 from distutils2.tests.pypi_server import use_pypi_server
 from distutils2.tests.support import unittest
 from distutils2.version import VersionPredicate
-from distutils2.pypi.errors import HashDoesNotMatch
+from distutils2.pypi.errors import HashDoesNotMatch, UnsupportedHashName
 from distutils2.pypi.dist import (PyPIDistribution as Dist,
                                   PyPIDistributions as Dists,
                                   split_archive_name)
@@ -140,6 +140,14 @@
         shutil.rmtree(path1)
         shutil.rmtree(os.path.dirname(path2))
 
+    def test_hashname(self):
+        """Invalid hashnames raises an exception on assignation"""
+        # should be ok
+        Dist("FooBar", "0.1", hashname="md5", hashval="value")
+
+        self.assertRaises(UnsupportedHashName, Dist, "FooBar", "0.1", 
+                          hashname="invalid_hashname", hashval="value")
+
 
 class TestPyPIDistributions(unittest.TestCase):
     """test the pypi.distr.PyPIDistributions class"""

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: Full support and tests for
	prefer_source and prefer_final
Message-ID: <E1OVLoW-0000lg-PI@dinsdale.python.org>

tarek.ziade pushed b0fa106d089d to distutils2:

http://hg.python.org/distutils2/rev/b0fa106d089d
changeset:   344:b0fa106d089d
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Fri Jul 02 15:54:00 2010 +0200
summary:     Full support and tests for prefer_source and prefer_final
files:       src/distutils2/pypi/dist.py, src/distutils2/pypi/simple.py, src/distutils2/tests/test_pypi_dist.py, src/distutils2/tests/test_pypi_simple.py

diff --git a/src/distutils2/pypi/dist.py b/src/distutils2/pypi/dist.py
--- a/src/distutils2/pypi/dist.py
+++ b/src/distutils2/pypi/dist.py
@@ -7,12 +7,14 @@
 import urlparse
 import urllib
 import tempfile
+from operator import attrgetter
+
 try:
     import hashlib
 except ImportError:
     from distutils2._backport import hashlib
 
-from distutils2.version import suggest_normalized_version
+from distutils2.version import suggest_normalized_version, NormalizedVersion
 from distutils2.pypi.errors import HashDoesNotMatch, UnsupportedHashName
 
 EXTENSIONS = ".tar.gz .tar.bz2 .tar .zip .tgz .egg".split()
@@ -57,26 +59,26 @@
 
         name, version = split_archive_name(archive_name)
         if extension_matched is True:
-            return PyPIDistribution(name, version, url=url, hashname="md5",
-                                    hashval=md5_hash, is_external=is_external)
+            return PyPIDistribution(name, version, url=url, url_hashname="md5",
+                                    url_hashval=md5_hash, url_is_external=is_external)
 
-    def __init__(self, name, version, type=None, url=None, hashname=None,
-                 hashval=None, is_external=True):
+    def __init__(self, name, version, type=None, url=None, url_hashname=None,
+                 url_hashval=None, url_is_external=True):
         """Create a new instance of PyPIDistribution.
 
         :param name: the name of the distribution
         :param version: the version of the distribution
         :param type: the type of the dist (eg. source, bin-*, etc.)
         :param url: URL where we found this distribution
-        :param hashname: the name of the hash we want to use. Refer to the
+        :param url_hashname: the name of the hash we want to use. Refer to the
                          hashlib.new documentation for more information.
-        :param hashval: the hash value.
-        :param is_external: we need to know if the provided url comes from an
+        :param url_hashval: the hash value.
+        :param url_is_external: we need to know if the provided url comes from an
                             index browsing, or from an external resource.
 
         """
         self.name = name
-        self.version = version
+        self.version = NormalizedVersion(version)
         self.type = type
         # set the downloaded path to None by default. The goal here
         # is to not download distributions multiple times
@@ -89,7 +91,7 @@
         # order to dont make the selection process multiple times.
         self._urls = []
         self._url = None
-        self.add_url(url, hashname, hashval, is_external)
+        self.add_url(url, url_hashname, url_hashval, url_is_external)
 
     def add_url(self, url, hashname=None, hashval=None, is_external=True):
         """Add a new url to the list of urls"""
@@ -124,6 +126,16 @@
                 self._url = self._urls[0]
         return self._url
 
+    @property
+    def is_source(self):
+        """return if the distribution is a source one or not"""
+        return self.type == 'source'
+
+    @property
+    def is_final(self):
+        """proxy to version.is_final"""
+        return self.version.is_final
+
     def download(self, path=None):
         """Download the distribution to a path, and return it.
 
@@ -156,8 +168,9 @@
                     % (hashval.hexdigest(), expected_hashval))
 
     def __repr__(self):
-        return "<%s %s (%s)>" \
-            % (self.__class__.__name__, self.name, self.version)
+        return "%s %s %s %s" \
+            % (self.__class__.__name__, self.name, self.version,
+               self.type or "")
 
     def _check_is_comparable(self, other):
         if not isinstance(other, PyPIDistribution):
@@ -210,13 +223,28 @@
             [dist for dist in self if dist.name == predicate.name and
             predicate.match(dist.version)])
 
-    def get_last(self, predicate):
+    def get_last(self, predicate, prefer_source=None, prefer_final=None):
         """Return the most up to date version, that satisfy the given
         predicate
         """
         distributions = self.filter(predicate)
-        distributions.sort()
-        return distributions[-1]
+        distributions.sort_distributions(prefer_source, prefer_final, reverse=True)
+        return distributions[0]
+
+    def get_same_name_and_version(self):
+        """Return lists of PyPIDistribution objects that refer to the same
+        name and version number. This do not consider the type (source, binary,
+        etc.)"""
+        processed = []
+        duplicates = []
+        for dist in self:
+            if (dist.name, dist.version) not in processed:
+                processed.append((dist.name, dist.version))
+                found_duplicates = [d for d in self if d.name == dist.name and
+                                    d.version == dist.version]
+                if len(found_duplicates) > 1:
+                    duplicates.append(found_duplicates)
+        return duplicates
 
     def append(self, o):
         """Append a new distribution to the list.
@@ -225,16 +253,30 @@
         URL informations and add a new new url for the existing one.
         """
         similar_dists = [d for d in self if d.name == o.name and
-                         d.version == o.version]
+                         d.version == o.version and d.type == o.type]
         if len(similar_dists) > 0:
             dist = similar_dists[0]
             dist.add_url(**o.url)
         else:
             super(PyPIDistributions, self).append(o)
 
-    def sort(self, prefer_source=None, prefer_final=None, *args, **kwargs):
-        return super(PyPIDistributions, self).sort(*args, **kwargs)
+    def sort_distributions(self, prefer_source=None, prefer_final=None,
+                           reverse=True, *args, **kwargs):
+        """order the results with the given properties"""
 
+        sort_by = []
+        if prefer_final is not None:
+            if prefer_final is True:
+                sort_by.append("is_final")
+        sort_by.append("version")
+
+        if prefer_source is not None:
+            if prefer_source is True:
+                sort_by.append("is_source")
+
+        super(PyPIDistributions, self).sort(
+            key=lambda i: [getattr(i, arg) for arg in sort_by],
+            reverse=reverse, *args, **kwargs)
 
 def split_archive_name(archive_name, probable_name=None):
     """Split an archive name into two parts: name and version.
diff --git a/src/distutils2/pypi/simple.py b/src/distutils2/pypi/simple.py
--- a/src/distutils2/pypi/simple.py
+++ b/src/distutils2/pypi/simple.py
@@ -136,7 +136,8 @@
         # filter with requirements and return the results
         if requirements.name in self._distributions:
             dists = self._distributions[requirements.name].filter(requirements)
-            dists.sort(prefer_source=prefer_source, prefer_final=prefer_final)
+            dists.sort_distributions(prefer_source=prefer_source,
+                                     prefer_final=prefer_final)
         else:
             dists = []
 
@@ -262,8 +263,9 @@
                     if self._is_distribution(link) or is_download:
                         self._processed_urls.append(link)
                         # it's a distribution, so create a dist object
-                        self._register_dist(PyPIDistribution.from_url(link,
-                            project_name, is_external=not self.index_url in url))
+                        dist = PyPIDistribution.from_url(link, project_name,
+                                    is_external=not self.index_url in url)
+                        self._register_dist(dist)
                     else:
                         if self._is_browsable(link) and follow_links:
                             self._process_url(link, project_name,
diff --git a/src/distutils2/tests/test_pypi_dist.py b/src/distutils2/tests/test_pypi_dist.py
--- a/src/distutils2/tests/test_pypi_dist.py
+++ b/src/distutils2/tests/test_pypi_dist.py
@@ -21,18 +21,18 @@
         given on construction"""
         dist = Dist("FooBar", "1.1")
         self.assertEqual("FooBar", dist.name)
-        self.assertEqual("1.1", dist.version)
+        self.assertEqual("1.1", "%s" % dist.version)
 
     def test_from_url(self):
         """Test that the Distribution object can be built from a single URL"""
         url_list = {
             'FooBar-1.1.0.tar.gz': {
                 'name': 'foobar',  # lowercase the name
-                'version': '1.1.0',
+                'version': '1.1',
             },
             'Foo-Bar-1.1.0.zip': {
                 'name': 'foo-bar',  # keep the dash
-                'version': '1.1.0',
+                'version': '1.1',
             },
             'foobar-1.1b2.tar.gz#md5=123123123123123': {
                 'name': 'foobar',
@@ -60,7 +60,10 @@
                     for val in value.keys():
                         self.assertEqual(value[val], mylist[val])
                 else:
-                    self.assertEqual(getattr(dist, attribute), value)
+                    if attribute == "version":
+                        self.assertEqual("%s" % getattr(dist, "version"), value)
+                    else:
+                        self.assertEqual(getattr(dist, attribute), value)
 
     def test_get_url(self):
         """Test that the url property works well"""
@@ -114,12 +117,12 @@
         url = "%s/simple/foobar/foobar-0.1.tar.gz" % server.full_address
         # check md5 if given
         dist = Dist("FooBar", "0.1", url=url,
-            hashname="md5", hashval="d41d8cd98f00b204e9800998ecf8427e")
+            url_hashname="md5", url_hashval="d41d8cd98f00b204e9800998ecf8427e")
         dist.download()
 
         # a wrong md5 fails
         dist2 = Dist("FooBar", "0.1", url=url,
-            hashname="md5", hashval="wrongmd5")
+            url_hashname="md5", url_hashval="wrongmd5")
         self.assertRaises(HashDoesNotMatch, dist2.download)
 
         # we can omit the md5 hash
@@ -143,10 +146,10 @@
     def test_hashname(self):
         """Invalid hashnames raises an exception on assignation"""
         # should be ok
-        Dist("FooBar", "0.1", hashname="md5", hashval="value")
+        Dist("FooBar", "0.1", url_hashname="md5", url_hashval="value")
 
         self.assertRaises(UnsupportedHashName, Dist, "FooBar", "0.1",
-                          hashname="invalid_hashname", hashval="value")
+                          url_hashname="invalid_hashname", url_hashval="value")
 
 
 class TestPyPIDistributions(unittest.TestCase):
@@ -176,30 +179,35 @@
         # If no object matches, just add "normally" the object to the list.
 
         dists = Dists([
-            Dist("FooBar", "1.1", url="external_url"),
+            Dist("FooBar", "1.1", url="external_url", type="source"),
         ])
         self.assertEqual(1, len(dists))
         dists.append(Dist("FooBar", "1.1", url="internal_url",
-                     is_external=False))
+                          url_is_external=False, type="source"))
         self.assertEqual(1, len(dists))
         self.assertEqual(2, len(dists[0]._urls))
 
-        dists.append(Dist("Foobar", "1.1.1"))
+        dists.append(Dist("Foobar", "1.1.1", type="source"))
         self.assertEqual(2, len(dists))
 
+        # when adding a distribution whith a different type, a new distribution
+        # has to be added.
+        dists.append(Dist("Foobar", "1.1.1", type="binary"))
+        self.assertEqual(3, len(dists))
+
     def test_prefer_final(self):
         """Ordering support prefer_final"""
 
         fb10 = Dist("FooBar", "1.0")  # final distribution
-        fb11a = Dist("FooBar", "1.1a")  # alpha
-        fb12a = Dist("FooBar", "1.2a")  # alpha
-        fb12b = Dist("FooBar", "1.2b")  # beta
+        fb11a = Dist("FooBar", "1.1a1")  # alpha
+        fb12a = Dist("FooBar", "1.2a1")  # alpha
+        fb12b = Dist("FooBar", "1.2b1")  # beta
         dists = Dists([fb10, fb11a, fb12a, fb12b])
 
-        dists.sort(prefer_final=True)
+        dists.sort_distributions(prefer_final=True)
         self.assertEqual(fb10, dists[0])
 
-        dists.sort(prefer_final=False)
+        dists.sort_distributions(prefer_final=False)
         self.assertEqual(fb12b, dists[0])
 
     def test_prefer_source(self):
@@ -209,16 +217,27 @@
         fb2_binary = Dist("FooBar", "2.0", type="binary")
         dists = Dists([fb_binary, fb_source])
 
-        dists.sort(prefer_source=True)
+        dists.sort_distributions(prefer_source=True)
         self.assertEqual(fb_source, dists[0])
 
-        dists.sort(prefer_source=False)
+        dists.sort_distributions(prefer_source=False)
         self.assertEqual(fb_binary, dists[0])
 
         dists.append(fb2_binary)
-        dists.sort(prefer_source=True)
+        dists.sort_distributions(prefer_source=True)
         self.assertEqual(fb2_binary, dists[0])
 
+    def test_get_same_name_and_version(self):
+        """PyPIDistributions can return a list of "duplicates"
+        """
+        fb_source = Dist("FooBar", "1.0", type="source")
+        fb_binary = Dist("FooBar", "1.0", type="binary")
+        fb2_binary = Dist("FooBar", "2.0", type="binary")
+        dists = Dists([fb_binary, fb_source, fb2_binary])
+        duplicates = dists.get_same_name_and_version()
+        self.assertTrue(1, len(duplicates))
+        self.assertIn(fb_source, duplicates[0])
+
 
 def test_suite():
     suite = unittest.TestSuite()
diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -89,9 +89,6 @@
         """Browse the index, asking for a specified distribution version
         """
         # The PyPI index contains links for version 1.0, 1.1, 2.0 and 2.0.1
-        # We query only for the version 1.1, so all distributions must be
-        # filled in the package_index (as the url has been scanned), but
-        # "get" must only return the one we want.
         index = self._get_simple_index(server)
         last_distribution = index.get("foobar")
 
@@ -103,7 +100,7 @@
         self.assertEqual(len(index._distributions["foobar"]), 4)
 
         # and returned the most recent one
-        self.assertEqual(last_distribution.version, '2.0.1')
+        self.assertEqual("%s" % last_distribution.version, '2.0.1')
 
     def test_is_browsable(self):
         index = simple.SimpleIndex(follow_externals=False)

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: Replace spurious output with logging
	(not configurable	yet)
Message-ID: <E1OVLoW-0000lM-MS@dinsdale.python.org>

tarek.ziade pushed 735e8a8d7431 to distutils2:

http://hg.python.org/distutils2/rev/735e8a8d7431
changeset:   339:735e8a8d7431
user:        ?ric Araujo <merwok at netwok.org>
date:        Thu Jul 01 19:22:58 2010 +0200
summary:     Replace spurious output with logging (not configurable yet)
files:       src/setup.py

diff --git a/src/setup.py b/src/setup.py
--- a/src/setup.py
+++ b/src/setup.py
@@ -5,11 +5,12 @@
 import os
 import re
 
+from distutils2 import __version__ as VERSION
+from distutils2 import log
 from distutils2.core import setup, Extension
 from distutils2.compiler.ccompiler import new_compiler
 from distutils2.command.sdist import sdist
 from distutils2.command.install import install
-from distutils2 import __version__ as VERSION
 from distutils2.util import find_packages
 
 f = open('README.txt')
@@ -161,9 +162,9 @@
 
     if (ssl_inc_dir and ssl_lib is not None and openssl_ver >= 0x00907000):
 
-        print 'Using OpenSSL version 0x%08x from' % openssl_ver
-        print ' Headers:\t', ssl_inc_dir
-        print ' Library:\t', ssl_lib
+        log.info('Using OpenSSL version 0x%08x from', openssl_ver)
+        log.info(' Headers:\t%s', ssl_inc_dir)
+        log.info(' Library:\t%s', ssl_lib)
 
         # The _hashlib module wraps optimized implementations
         # of hash functions from the OpenSSL library.

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: write tests for prefer_final and
	prefer_source support
Message-ID: <E1OVLoW-0000lD-LC@dinsdale.python.org>

tarek.ziade pushed f9be1a9d5314 to distutils2:

http://hg.python.org/distutils2/rev/f9be1a9d5314
changeset:   337:f9be1a9d5314
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Thu Jul 01 20:01:10 2010 +0200
summary:     write tests for prefer_final and prefer_source support
files:       src/distutils2/tests/test_pypi_dist.py

diff --git a/src/distutils2/tests/test_pypi_dist.py b/src/distutils2/tests/test_pypi_dist.py
--- a/src/distutils2/tests/test_pypi_dist.py
+++ b/src/distutils2/tests/test_pypi_dist.py
@@ -187,6 +187,38 @@
         dists.append(Dist("Foobar", "1.1.1"))
         self.assertEqual(2, len(dists))
 
+    def test_prefer_final(self):
+        """Ordering support prefer_final"""
+
+        fb10 = Dist("FooBar", "1.0")  # final distribution
+        fb11a = Dist("FooBar", "1.1a")  # alpha
+        fb12a = Dist("FooBar", "1.2a")  # alpha
+        fb12b = Dist("FooBar", "1.2b")  # beta
+        dists = Dists([fb10, fb11a, fb12a, fb12b])
+
+        dists.sort(prefer_final=True)
+        self.assertEqual(fb10, dists[0])
+
+        dists.sort(prefer_final=False)
+        self.assertEqual(fb12b, dists[0])
+
+    def test_prefer_source(self):
+        """Ordering support prefer_source"""
+        fb_source = Dist("FooBar", "1.0", type="source")
+        fb_binary = Dist("FooBar", "1.0", type="binary")
+        fb2_binary = Dist("FooBar", "2.0", type="binary")
+        dists = Dists([fb_binary, fb_source])
+
+        dists.sort(prefer_source=True)
+        self.assertEqual(fb_source, dists[0])
+
+        dists.sort(prefer_source=False)
+        self.assertEqual(fb_binary, dists[0])
+
+        dists.append(fb2_binary)
+        dists.sort(prefer_source=True)
+        self.assertEqual(fb2_binary, dists[0])
+
 
 def test_suite():
     suite = unittest.TestSuite()

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: Set the mirror list fallback to an
	empty list if problems	occurs while getting
Message-ID: <E1OVLoW-0000lk-Pr@dinsdale.python.org>

tarek.ziade pushed 880201efc140 to distutils2:

http://hg.python.org/distutils2/rev/880201efc140
changeset:   345:880201efc140
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Fri Jul 02 18:28:20 2010 +0200
summary:     Set the mirror list fallback to an empty list if problems occurs while getting
files:       src/distutils2/pypi/simple.py

diff --git a/src/distutils2/pypi/simple.py b/src/distutils2/pypi/simple.py
--- a/src/distutils2/pypi/simple.py
+++ b/src/distutils2/pypi/simple.py
@@ -94,7 +94,10 @@
         self._index_urls = [index_url]
         # if no mirrors are defined, use the method described in PEP 381.
         if mirrors is None:
-            mirrors = socket.gethostbyname_ex(mirrors_url)[-1]
+            try:
+                mirrors = socket.gethostbyname_ex(mirrors_url)[-1]
+            except socket.gaierror:
+                mirrors = []
         self._index_urls.extend(mirrors)
         self._current_index_url = 0
         self._timeout = timeout

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: add support of is_final in
	distutils2.version
Message-ID: <E1OVLoW-0000lc-Of@dinsdale.python.org>

tarek.ziade pushed df80119b94b6 to distutils2:

http://hg.python.org/distutils2/rev/df80119b94b6
changeset:   343:df80119b94b6
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Fri Jul 02 10:42:54 2010 +0200
summary:     add support of is_final in distutils2.version
files:       src/distutils2/tests/test_version.py, src/distutils2/version.py

diff --git a/src/distutils2/tests/test_version.py b/src/distutils2/tests/test_version.py
--- a/src/distutils2/tests/test_version.py
+++ b/src/distutils2/tests/test_version.py
@@ -154,6 +154,17 @@
         # XXX need to silent the micro version in this case
         #assert not VersionPredicate('Ho (<3.0,!=2.6)').match('2.6.3')
 
+    def test_is_final(self):
+        # VersionPredicate knows is a distribution is a final one or not.
+        final_versions = ('1.0', '1.0.post456')
+        other_versions = ('1.0.dev1', '1.0a2', '1.0c3')
+
+        for version in final_versions:
+            self.assertTrue(V(version).is_final)
+        for version in other_versions:
+            self.assertFalse(V(version).is_final)
+
+
 def test_suite():
     #README = os.path.join(os.path.dirname(__file__), 'README.txt')
     #suite = [doctest.DocFileSuite(README), unittest.makeSuite(VersionTestCase)]
@@ -162,4 +173,3 @@
 
 if __name__ == "__main__":
     unittest.main(defaultTest="test_suite")
-
diff --git a/src/distutils2/version.py b/src/distutils2/version.py
--- a/src/distutils2/version.py
+++ b/src/distutils2/version.py
@@ -36,6 +36,7 @@
     (?P<postdev>(\.post(?P<post>\d+))?(\.dev(?P<dev>\d+))?)?
     $''', re.VERBOSE)
 
+
 class NormalizedVersion(object):
     """A rational version.
 
@@ -61,7 +62,8 @@
         @param error_on_huge_major_num {bool} Whether to consider an
             apparent use of a year or full date as the major version number
             an error. Default True. One of the observed patterns on PyPI before
-            the introduction of `NormalizedVersion` was version numbers like this:
+            the introduction of `NormalizedVersion` was version numbers like
+            this:
                 2009.01.03
                 20040603
                 2005.01
@@ -71,6 +73,7 @@
             the possibility of using a version number like "1.0" (i.e.
             where the major number is less than that huge major number).
         """
+        self.is_final = True  # by default, consider a version as final.
         self._parse(s, error_on_huge_major_num)
 
     @classmethod
@@ -101,6 +104,7 @@
             block += self._parse_numdots(groups.get('prerelversion'), s,
                                          pad_zeros_length=1)
             parts.append(tuple(block))
+            self.is_final = False
         else:
             parts.append(_FINAL_MARKER)
 
@@ -115,6 +119,7 @@
                     postdev.append(_FINAL_MARKER[0])
             if dev is not None:
                 postdev.extend(['dev', int(dev)])
+                self.is_final = False
             parts.append(tuple(postdev))
         else:
             parts.append(_FINAL_MARKER)
@@ -204,6 +209,7 @@
     # See http://docs.python.org/reference/datamodel#object.__hash__
     __hash__ = object.__hash__
 
+
 def suggest_normalized_version(s):
     """Suggest a normalized version close to the given version string.
 
@@ -305,7 +311,6 @@
     # PyPI stats: ~21 (0.62%) better
     rs = re.sub(r"\.?(pre|preview|-c)(\d+)$", r"c\g<2>", rs)
 
-
     # Tcl/Tk uses "px" for their post release markers
     rs = re.sub(r"p(\d+)$", r".post\1", rs)
 
@@ -322,6 +327,7 @@
 _PLAIN_VERSIONS = re.compile(r"^\s*(.*)\s*$")
 _SPLIT_CMP = re.compile(r"^\s*(<=|>=|<|>|!=|==)\s*([^\s,]+)\s*$")
 
+
 def _split_predicate(predicate):
     match = _SPLIT_CMP.match(predicate)
     if match is None:
@@ -368,6 +374,7 @@
                 return False
         return True
 
+
 class _Versions(VersionPredicate):
     def __init__(self, predicate):
         predicate = predicate.strip()
@@ -379,6 +386,7 @@
         self.predicates = [_split_predicate(pred.strip())
                            for pred in predicates.split(',')]
 
+
 class _Version(VersionPredicate):
     def __init__(self, predicate):
         predicate = predicate.strip()
@@ -388,6 +396,7 @@
         self.name = None
         self.predicates = _split_predicate(match.groups()[0])
 
+
 def is_valid_predicate(predicate):
     try:
         VersionPredicate(predicate)
@@ -396,6 +405,7 @@
     else:
         return True
 
+
 def is_valid_versions(predicate):
     try:
         _Versions(predicate)
@@ -404,6 +414,7 @@
     else:
         return True
 
+
 def is_valid_version(predicate):
     try:
         _Version(predicate)
@@ -411,4 +422,3 @@
         return False
     else:
         return True
-

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: Remove evil eval
Message-ID: <E1OVLoW-0000lP-Mv@dinsdale.python.org>

tarek.ziade pushed b338176f6c00 to distutils2:

http://hg.python.org/distutils2/rev/b338176f6c00
changeset:   340:b338176f6c00
user:        ?ric Araujo <merwok at netwok.org>
date:        Thu Jul 01 19:29:19 2010 +0200
summary:     Remove evil eval
files:       src/setup.py

diff --git a/src/setup.py b/src/setup.py
--- a/src/setup.py
+++ b/src/setup.py
@@ -151,7 +151,7 @@
                 for line in incfile:
                     m = openssl_ver_re.match(line)
                     if m:
-                        openssl_ver = eval(m.group(1))
+                        openssl_ver = int(m.group(1), 16)
                         break
             except IOError:
                 pass

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: merge merwok changes
Message-ID: <E1OVLoW-0000lY-O6@dinsdale.python.org>

tarek.ziade pushed bb514146b4f3 to distutils2:

http://hg.python.org/distutils2/rev/bb514146b4f3
changeset:   342:bb514146b4f3
parent:      337:f9be1a9d5314
parent:      341:95ac66bad7cb
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Thu Jul 01 20:40:36 2010 +0200
summary:     merge merwok changes
files:       

diff --git a/src/setup.py b/src/setup.py
--- a/src/setup.py
+++ b/src/setup.py
@@ -5,11 +5,12 @@
 import os
 import re
 
+from distutils2 import __version__ as VERSION
+from distutils2 import log
 from distutils2.core import setup, Extension
 from distutils2.compiler.ccompiler import new_compiler
 from distutils2.command.sdist import sdist
 from distutils2.command.install import install
-from distutils2 import __version__ as VERSION
 from distutils2.util import find_packages
 
 f = open('README.txt')
@@ -67,14 +68,15 @@
 
 
 # additional paths to check, set from the command line
-SSL_INCDIR=''   # --openssl-incdir=
-SSL_LIBDIR=''   # --openssl-libdir=
-SSL_DIR=''      # --openssl-prefix=
+SSL_INCDIR = ''   # --openssl-incdir=
+SSL_LIBDIR = ''   # --openssl-libdir=
+SSL_DIR = ''      # --openssl-prefix=
 
 def add_dir_to_list(dirlist, dir):
     """Add the directory 'dir' to the list 'dirlist' (at the front) if
     'dir' actually exists and is a directory.  If 'dir' is already in
-    'dirlist' it is moved to the front."""
+    'dirlist' it is moved to the front.
+    """
     if dir is not None and os.path.isdir(dir) and dir not in dirlist:
         if dir in dirlist:
             dirlist.remove(dir)
@@ -83,8 +85,8 @@
 
 def prepare_hashlib_extensions():
     """Decide which C extensions to build and create the appropriate
-    Extension objects to build them.  Returns a list of Extensions."""
-
+    Extension objects to build them.  Return a list of Extensions.
+    """
     # this CCompiler object is only used to locate include files
     compiler = new_compiler()
 
@@ -116,13 +118,12 @@
             add_dir_to_list(compiler.library_dirs, os.path.join(SSL_DIR, 'lib'))
         add_dir_to_list(compiler.include_dirs, os.path.join(SSL_DIR, 'include'))
 
-    osNameLibsMap = {
-        'posix':  ['ssl', 'crypto'],
-        'nt':     ['libeay32',  'gdi32', 'advapi32', 'user32'],
-    }
+    oslibs = {'posix': ['ssl', 'crypto'],
+              'nt': ['libeay32',  'gdi32', 'advapi32', 'user32']}
 
-    if not osNameLibsMap.has_key(os.name):
-        print "unknown OS, please update setup.py"
+    if os.name not in oslibs:
+        sys.stderr.write(
+            'unknown operating system, impossible to compile _hashlib')
         sys.exit(1)
 
     exts = []
@@ -135,7 +136,7 @@
             ssl_incs.append(f)
             ssl_inc_dirs.append(inc_dir)
 
-    ssl_lib = compiler.find_library_file(compiler.library_dirs, osNameLibsMap[os.name][0])
+    ssl_lib = compiler.find_library_file(compiler.library_dirs, oslibs[os.name][0])
 
     # find out which version of OpenSSL we have
     openssl_ver = 0
@@ -150,7 +151,7 @@
                 for line in incfile:
                     m = openssl_ver_re.match(line)
                     if m:
-                        openssl_ver = eval(m.group(1))
+                        openssl_ver = int(m.group(1), 16)
                         break
             except IOError:
                 pass
@@ -159,39 +160,37 @@
         if openssl_ver:
             break
 
-    if (ssl_inc_dir and
-        ssl_lib is not None and
-        openssl_ver >= 0x00907000):
+    if (ssl_inc_dir and ssl_lib is not None and openssl_ver >= 0x00907000):
 
-        print 'Using OpenSSL version 0x%08x from' % openssl_ver
-        print ' Headers:\t', ssl_inc_dir
-        print ' Library:\t', ssl_lib
+        log.info('Using OpenSSL version 0x%08x from', openssl_ver)
+        log.info(' Headers:\t%s', ssl_inc_dir)
+        log.info(' Library:\t%s', ssl_lib)
 
         # The _hashlib module wraps optimized implementations
         # of hash functions from the OpenSSL library.
-        exts.append( Extension('_hashlib', ['_hashopenssl.c'],
-                               include_dirs = [ ssl_inc_dir ],
-                               library_dirs = [ os.path.dirname(ssl_lib) ],
-                               libraries = osNameLibsMap[os.name]) )
+        exts.append(Extension('_hashlib', ['_hashopenssl.c'],
+                              include_dirs = [ssl_inc_dir],
+                              library_dirs = [os.path.dirname(ssl_lib)],
+                              libraries = oslibs[os.name]))
     else:
-        exts.append( Extension('_sha', ['shamodule.c']) )
-        exts.append( Extension('_md5',
-                        sources = ['md5module.c', 'md5.c'],
-                        depends = ['md5.h']) )
+        exts.append(Extension('_sha', ['shamodule.c']) )
+        exts.append(Extension('_md5',
+                              sources=['md5module.c', 'md5.c'],
+                              depends=['md5.h']) )
 
     if (not ssl_lib or openssl_ver < 0x00908000):
         # OpenSSL doesn't do these until 0.9.8 so we'll bring our own
-        exts.append( Extension('_sha256', ['sha256module.c']) )
-        exts.append( Extension('_sha512', ['sha512module.c']) )
+        exts.append(Extension('_sha256', ['sha256module.c']))
+        exts.append(Extension('_sha512', ['sha512module.c']))
 
-    def prependModules(filename):
+    def prepend_modules(filename):
         return os.path.join('Modules', filename)
 
     # all the C code is in the Modules subdirectory, prepend the path
     for ext in exts:
-        ext.sources = [ prependModules(fn) for fn in ext.sources ]
+        ext.sources = [prepend_modules(fn) for fn in ext.sources]
         if hasattr(ext, 'depends') and ext.depends is not None:
-            ext.depends = [ prependModules(fn) for fn in ext.depends ]
+            ext.depends = [prepend_modules(fn) for fn in ext.depends]
 
     return exts
 
@@ -213,24 +212,23 @@
 Topic :: System :: Systems Administration
 Topic :: Utilities"""
 
-setup (name="Distutils2",
-       version=VERSION,
-       summary="Python Distribution Utilities",
-       keywords=['packaging', 'distutils'],
-       author="Tarek Ziade",
-       author_email="tarek at ziade.org",
-       home_page="http://bitbucket.org/tarek/distutils2/wiki/Home",
-       license="PSF",
-       description=README,
-       classifier=_CLASSIFIERS.split('\n'),
-       packages=find_packages(),
-       cmdclass={'sdist': sdist_hg, 'install': install_hg},
-       package_data={'distutils2._backport': ['sysconfig.cfg']},
-       project_url=[('Mailing-list',
+setup(name="Distutils2",
+      version=VERSION,
+      summary="Python Distribution Utilities",
+      keywords=['packaging', 'distutils'],
+      author="Tarek Ziade",
+      author_email="tarek at ziade.org",
+      home_page="http://bitbucket.org/tarek/distutils2/wiki/Home",
+      license="PSF",
+      description=README,
+      classifier=_CLASSIFIERS.split('\n'),
+      packages=find_packages(),
+      cmdclass={'sdist': sdist_hg, 'install': install_hg},
+      package_data={'distutils2._backport': ['sysconfig.cfg']},
+      project_url=[('Mailing-list',
                     'http://mail.python.org/mailman/listinfo/distutils-sig/'),
-                    ('Documentation',
-                     'http://packages.python.org/Distutils2'),
-                    ('Repository', 'http://hg.python.org/distutils2'),
-                    ('Bug tracker', 'http://bugs.python.org')],
-       **setup_kwargs
-          )
+                   ('Documentation',
+                    'http://packages.python.org/Distutils2'),
+                   ('Repository', 'http://hg.python.org/distutils2'),
+                   ('Bug tracker', 'http://bugs.python.org')],
+      **setup_kwargs)

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: Other whitespace fixes in setup script
Message-ID: <E1OVLoW-0000lU-NQ@dinsdale.python.org>

tarek.ziade pushed 95ac66bad7cb to distutils2:

http://hg.python.org/distutils2/rev/95ac66bad7cb
changeset:   341:95ac66bad7cb
user:        ?ric Araujo <merwok at netwok.org>
date:        Thu Jul 01 19:29:48 2010 +0200
summary:     Other whitespace fixes in setup script
files:       src/setup.py

diff --git a/src/setup.py b/src/setup.py
--- a/src/setup.py
+++ b/src/setup.py
@@ -212,24 +212,23 @@
 Topic :: System :: Systems Administration
 Topic :: Utilities"""
 
-setup (name="Distutils2",
-       version=VERSION,
-       summary="Python Distribution Utilities",
-       keywords=['packaging', 'distutils'],
-       author="Tarek Ziade",
-       author_email="tarek at ziade.org",
-       home_page="http://bitbucket.org/tarek/distutils2/wiki/Home",
-       license="PSF",
-       description=README,
-       classifier=_CLASSIFIERS.split('\n'),
-       packages=find_packages(),
-       cmdclass={'sdist': sdist_hg, 'install': install_hg},
-       package_data={'distutils2._backport': ['sysconfig.cfg']},
-       project_url=[('Mailing-list',
+setup(name="Distutils2",
+      version=VERSION,
+      summary="Python Distribution Utilities",
+      keywords=['packaging', 'distutils'],
+      author="Tarek Ziade",
+      author_email="tarek at ziade.org",
+      home_page="http://bitbucket.org/tarek/distutils2/wiki/Home",
+      license="PSF",
+      description=README,
+      classifier=_CLASSIFIERS.split('\n'),
+      packages=find_packages(),
+      cmdclass={'sdist': sdist_hg, 'install': install_hg},
+      package_data={'distutils2._backport': ['sysconfig.cfg']},
+      project_url=[('Mailing-list',
                     'http://mail.python.org/mailman/listinfo/distutils-sig/'),
-                    ('Documentation',
-                     'http://packages.python.org/Distutils2'),
-                    ('Repository', 'http://hg.python.org/distutils2'),
-                    ('Bug tracker', 'http://bugs.python.org')],
-       **setup_kwargs
-          )
+                   ('Documentation',
+                    'http://packages.python.org/Distutils2'),
+                   ('Repository', 'http://hg.python.org/distutils2'),
+                   ('Bug tracker', 'http://bugs.python.org')],
+      **setup_kwargs)

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: update tests regarding Eric's changes
Message-ID: <E1OVLoW-0000ls-Ux@dinsdale.python.org>

tarek.ziade pushed a569f25a2038 to distutils2:

http://hg.python.org/distutils2/rev/a569f25a2038
changeset:   347:a569f25a2038
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Sat Jul 03 18:23:22 2010 +0200
summary:     update tests regarding Eric's changes
files:       src/distutils2/tests/test_pypi_dist.py, src/distutils2/tests/test_pypi_simple.py

diff --git a/src/distutils2/tests/test_pypi_dist.py b/src/distutils2/tests/test_pypi_dist.py
--- a/src/distutils2/tests/test_pypi_dist.py
+++ b/src/distutils2/tests/test_pypi_dist.py
@@ -5,6 +5,7 @@
 import tempfile
 
 from distutils2.tests.pypi_server import use_pypi_server
+from distutils2.tests import support
 from distutils2.tests.support import unittest
 from distutils2.version import VersionPredicate
 from distutils2.pypi.errors import HashDoesNotMatch, UnsupportedHashName
@@ -13,7 +14,8 @@
                                   split_archive_name)
 
 
-class TestPyPIDistribution(unittest.TestCase):
+class TestPyPIDistribution(support.TempdirManager,
+                           unittest.TestCase):
     """tests the pypi.dist.PyPIDistribution class"""
 
     def test_instanciation(self):
@@ -246,4 +248,4 @@
     return suite
 
 if __name__ == '__main__':
-    unittest.main(defaultTest="test_suite")
+    run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -5,9 +5,10 @@
 import os
 import shutil
 import tempfile
-import unittest2
 import urllib2
 
+from distutils2.tests import support
+from distutils2.tests.support import unittest
 from distutils2.tests.pypi_server import use_pypi_server, PyPIServer, \
                                          PYPI_DEFAULT_STATIC_PATH
 from distutils2.pypi import simple
@@ -15,7 +16,8 @@
 from distutils2.errors import DistutilsError
 
 
-class PyPISimpleTestCase(unittest2.TestCase):
+class PyPISimpleTestCase(support.TempdirManager,
+                         unittest.TestCase):
 
     def _get_simple_index(self, server, base_url="/simple/", hosts=None,
         *args, **kwargs):
@@ -282,7 +284,7 @@
         self.assertEqual(4, len(dists))
 
 def test_suite():
-    return unittest2.makeSuite(PyPISimpleTestCase)
+    return unittest.makeSuite(PyPISimpleTestCase)
 
 if __name__ == '__main__':
-    unittest2.main(defaultTest="test_suite")
+    run_unittest(test_suite())

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:41 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:41 +0200
Subject: [Python-checkins] distutils2: make sure the .so files are built
	everytime
Message-ID: <E1OVLoX-0000lz-3B@dinsdale.python.org>

tarek.ziade pushed 38d8b09c88c9 to distutils2:

http://hg.python.org/distutils2/rev/38d8b09c88c9
changeset:   349:38d8b09c88c9
user:        Tarek Ziade <tarek at ziade.org>
date:        Sun Jul 04 11:45:00 2010 +0200
summary:     make sure the .so files are built everytime
files:       src/tests.sh

diff --git a/src/tests.sh b/src/tests.sh
--- a/src/tests.sh
+++ b/src/tests.sh
@@ -1,5 +1,7 @@
 #!/bin/sh
 echo -n "Running tests for Python 2.4... "
+rm -rf _hashlib.so
+python2.4 setup.py build_ext -i -q 2> /dev/null > /dev/null
 python2.4 -Wd runtests.py -q 2> /dev/null
 if [ $? -ne 0 ];then
     echo "Failed"
@@ -9,6 +11,8 @@
 fi
 
 echo -n "Running tests for Python 2.5... "
+rm -rf _hashlib.so
+python2.5 setup.py build_ext -i -q 2> /dev/null > /dev/null
 python2.5 -Wd runtests.py -q 2> /dev/null
 if [ $? -ne 0 ];then
     echo "Failed"

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: merge old changes
Message-ID: <E1OVLoW-0000l2-J2@dinsdale.python.org>

tarek.ziade pushed 513b81a401e9 to distutils2:

http://hg.python.org/distutils2/rev/513b81a401e9
changeset:   334:513b81a401e9
parent:      333:5b76ba38fecf
parent:      330:be0c887b2eb1
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Thu Jul 01 16:58:18 2010 +0200
summary:     merge old changes
files:       src/distutils2/pypi/dist.py, src/distutils2/pypi/simple.py, src/distutils2/tests/test_pypi_simple.py

diff --git a/src/Modules/_hashopenssl.c b/src/Modules/_hashopenssl.c
new file mode 100644
--- /dev/null
+++ b/src/Modules/_hashopenssl.c
@@ -0,0 +1,524 @@
+/* Module that wraps all OpenSSL hash algorithms */
+
+/*
+ * Copyright (C) 2005   Gregory P. Smith (greg at krypto.org)
+ * Licensed to PSF under a Contributor Agreement.
+ *
+ * Derived from a skeleton of shamodule.c containing work performed by:
+ *
+ * Andrew Kuchling (amk at amk.ca)
+ * Greg Stein (gstein at lyra.org)
+ *
+ */
+
+#define PY_SSIZE_T_CLEAN
+
+#include "Python.h"
+#include "structmember.h"
+
+#if (PY_VERSION_HEX < 0x02050000)
+#define Py_ssize_t      int
+#endif
+
+/* EVP is the preferred interface to hashing in OpenSSL */
+#include <openssl/evp.h>
+
+#define MUNCH_SIZE INT_MAX
+
+
+#ifndef HASH_OBJ_CONSTRUCTOR
+#define HASH_OBJ_CONSTRUCTOR 0
+#endif
+
+typedef struct {
+    PyObject_HEAD
+    PyObject            *name;  /* name of this hash algorithm */
+    EVP_MD_CTX          ctx;    /* OpenSSL message digest context */
+} EVPobject;
+
+
+static PyTypeObject EVPtype;
+
+
+#define DEFINE_CONSTS_FOR_NEW(Name)  \
+    static PyObject *CONST_ ## Name ## _name_obj; \
+    static EVP_MD_CTX CONST_new_ ## Name ## _ctx; \
+    static EVP_MD_CTX *CONST_new_ ## Name ## _ctx_p = NULL;
+
+DEFINE_CONSTS_FOR_NEW(md5)
+DEFINE_CONSTS_FOR_NEW(sha1)
+DEFINE_CONSTS_FOR_NEW(sha224)
+DEFINE_CONSTS_FOR_NEW(sha256)
+DEFINE_CONSTS_FOR_NEW(sha384)
+DEFINE_CONSTS_FOR_NEW(sha512)
+
+
+static EVPobject *
+newEVPobject(PyObject *name)
+{
+    EVPobject *retval = (EVPobject *)PyObject_New(EVPobject, &EVPtype);
+
+    /* save the name for .name to return */
+    if (retval != NULL) {
+        Py_INCREF(name);
+        retval->name = name;
+    }
+
+    return retval;
+}
+
+/* Internal methods for a hash object */
+
+static void
+EVP_dealloc(PyObject *ptr)
+{
+    EVP_MD_CTX_cleanup(&((EVPobject *)ptr)->ctx);
+    Py_XDECREF(((EVPobject *)ptr)->name);
+    PyObject_Del(ptr);
+}
+
+
+/* External methods for a hash object */
+
+PyDoc_STRVAR(EVP_copy__doc__, "Return a copy of the hash object.");
+
+static PyObject *
+EVP_copy(EVPobject *self, PyObject *unused)
+{
+    EVPobject *newobj;
+
+    if ( (newobj = newEVPobject(self->name))==NULL)
+        return NULL;
+
+    EVP_MD_CTX_copy(&newobj->ctx, &self->ctx);
+    return (PyObject *)newobj;
+}
+
+PyDoc_STRVAR(EVP_digest__doc__,
+"Return the digest value as a string of binary data.");
+
+static PyObject *
+EVP_digest(EVPobject *self, PyObject *unused)
+{
+    unsigned char digest[EVP_MAX_MD_SIZE];
+    EVP_MD_CTX temp_ctx;
+    PyObject *retval;
+    unsigned int digest_size;
+
+    EVP_MD_CTX_copy(&temp_ctx, &self->ctx);
+    digest_size = EVP_MD_CTX_size(&temp_ctx);
+    EVP_DigestFinal(&temp_ctx, digest, NULL);
+
+    retval = PyString_FromStringAndSize((const char *)digest, digest_size);
+    EVP_MD_CTX_cleanup(&temp_ctx);
+    return retval;
+}
+
+PyDoc_STRVAR(EVP_hexdigest__doc__,
+"Return the digest value as a string of hexadecimal digits.");
+
+static PyObject *
+EVP_hexdigest(EVPobject *self, PyObject *unused)
+{
+    unsigned char digest[EVP_MAX_MD_SIZE];
+    EVP_MD_CTX temp_ctx;
+    PyObject *retval;
+    char *hex_digest;
+    unsigned int i, j, digest_size;
+
+    /* Get the raw (binary) digest value */
+    EVP_MD_CTX_copy(&temp_ctx, &self->ctx);
+    digest_size = EVP_MD_CTX_size(&temp_ctx);
+    EVP_DigestFinal(&temp_ctx, digest, NULL);
+
+    EVP_MD_CTX_cleanup(&temp_ctx);
+
+    /* Create a new string */
+    /* NOTE: not thread safe! modifying an already created string object */
+    /* (not a problem because we hold the GIL by default) */
+    retval = PyString_FromStringAndSize(NULL, digest_size * 2);
+    if (!retval)
+	    return NULL;
+    hex_digest = PyString_AsString(retval);
+    if (!hex_digest) {
+	    Py_DECREF(retval);
+	    return NULL;
+    }
+
+    /* Make hex version of the digest */
+    for(i=j=0; i<digest_size; i++) {
+        char c;
+        c = (digest[i] >> 4) & 0xf;
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+        c = (digest[i] & 0xf);
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+    }
+    return retval;
+}
+
+PyDoc_STRVAR(EVP_update__doc__,
+"Update this hash object's state with the provided string.");
+
+static PyObject *
+EVP_update(EVPobject *self, PyObject *args)
+{
+    unsigned char *cp;
+    Py_ssize_t len;
+
+    if (!PyArg_ParseTuple(args, "s#:update", &cp, &len))
+        return NULL;
+
+    if (len > 0 && len <= MUNCH_SIZE) {
+    EVP_DigestUpdate(&self->ctx, cp, Py_SAFE_DOWNCAST(len, Py_ssize_t,
+                                                      unsigned int));
+    } else {
+        Py_ssize_t offset = 0;
+        while (len) {
+            unsigned int process = len > MUNCH_SIZE ? MUNCH_SIZE : len;
+            EVP_DigestUpdate(&self->ctx, cp + offset, process);
+            len -= process;
+            offset += process;
+        }
+    }
+    Py_INCREF(Py_None);
+    return Py_None;
+}
+
+static PyMethodDef EVP_methods[] = {
+    {"update",	  (PyCFunction)EVP_update,    METH_VARARGS, EVP_update__doc__},
+    {"digest",	  (PyCFunction)EVP_digest,    METH_NOARGS,  EVP_digest__doc__},
+    {"hexdigest", (PyCFunction)EVP_hexdigest, METH_NOARGS,  EVP_hexdigest__doc__},
+    {"copy",	  (PyCFunction)EVP_copy,      METH_NOARGS,  EVP_copy__doc__},
+    {NULL,	  NULL}		/* sentinel */
+};
+
+static PyObject *
+EVP_get_block_size(EVPobject *self, void *closure)
+{
+    return PyInt_FromLong(EVP_MD_CTX_block_size(&((EVPobject *)self)->ctx));
+}
+
+static PyObject *
+EVP_get_digest_size(EVPobject *self, void *closure)
+{
+    return PyInt_FromLong(EVP_MD_CTX_size(&((EVPobject *)self)->ctx));
+}
+
+static PyMemberDef EVP_members[] = {
+    {"name", T_OBJECT, offsetof(EVPobject, name), READONLY, PyDoc_STR("algorithm name.")},
+    {NULL}  /* Sentinel */
+};
+
+static PyGetSetDef EVP_getseters[] = {
+    {"digest_size",
+     (getter)EVP_get_digest_size, NULL,
+     NULL,
+     NULL},
+    {"block_size",
+     (getter)EVP_get_block_size, NULL,
+     NULL,
+     NULL},
+    /* the old md5 and sha modules support 'digest_size' as in PEP 247.
+     * the old sha module also supported 'digestsize'.  ugh. */
+    {"digestsize",
+     (getter)EVP_get_digest_size, NULL,
+     NULL,
+     NULL},
+    {NULL}  /* Sentinel */
+};
+
+
+static PyObject *
+EVP_repr(PyObject *self)
+{
+    char buf[100];
+    PyOS_snprintf(buf, sizeof(buf), "<%s HASH object @ %p>",
+            PyString_AsString(((EVPobject *)self)->name), self);
+    return PyString_FromString(buf);
+}
+
+#if HASH_OBJ_CONSTRUCTOR
+static int
+EVP_tp_init(EVPobject *self, PyObject *args, PyObject *kwds)
+{
+    static char *kwlist[] = {"name", "string", NULL};
+    PyObject *name_obj = NULL;
+    char *nameStr;
+    unsigned char *cp = NULL;
+    Py_ssize_t len = 0;
+    const EVP_MD *digest;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwds, "O|s#:HASH", kwlist,
+                                     &name_obj, &cp, &len)) {
+        return -1;
+    }
+
+    if (!PyArg_Parse(name_obj, "s", &nameStr)) {
+        PyErr_SetString(PyExc_TypeError, "name must be a string");
+        return -1;
+    }
+
+    digest = EVP_get_digestbyname(nameStr);
+    if (!digest) {
+        PyErr_SetString(PyExc_ValueError, "unknown hash function");
+        return -1;
+    }
+    EVP_DigestInit(&self->ctx, digest);
+
+    self->name = name_obj;
+    Py_INCREF(self->name);
+
+    if (cp && len) {
+        if (len > 0 && len <= MUNCH_SIZE) {
+        EVP_DigestUpdate(&self->ctx, cp, Py_SAFE_DOWNCAST(len, Py_ssize_t,
+                                                          unsigned int));
+        } else {
+            Py_ssize_t offset = 0;
+            while (len) {
+                unsigned int process = len > MUNCH_SIZE ? MUNCH_SIZE : len;
+                EVP_DigestUpdate(&self->ctx, cp + offset, process);
+                len -= process;
+                offset += process;
+            }
+        }
+    }
+    
+    return 0;
+}
+#endif
+
+
+PyDoc_STRVAR(hashtype_doc,
+"A hash represents the object used to calculate a checksum of a\n\
+string of information.\n\
+\n\
+Methods:\n\
+\n\
+update() -- updates the current digest with an additional string\n\
+digest() -- return the current digest value\n\
+hexdigest() -- return the current digest as a string of hexadecimal digits\n\
+copy() -- return a copy of the current hash object\n\
+\n\
+Attributes:\n\
+\n\
+name -- the hash algorithm being used by this object\n\
+digest_size -- number of bytes in this hashes output\n");
+
+static PyTypeObject EVPtype = {
+    PyObject_HEAD_INIT(NULL)
+    0,			/*ob_size*/
+    "_hashlib.HASH",    /*tp_name*/
+    sizeof(EVPobject),	/*tp_basicsize*/
+    0,			/*tp_itemsize*/
+    /* methods */
+    EVP_dealloc,	/*tp_dealloc*/
+    0,			/*tp_print*/
+    0,                  /*tp_getattr*/
+    0,                  /*tp_setattr*/
+    0,                  /*tp_compare*/
+    EVP_repr,           /*tp_repr*/
+    0,                  /*tp_as_number*/
+    0,                  /*tp_as_sequence*/
+    0,                  /*tp_as_mapping*/
+    0,                  /*tp_hash*/
+    0,                  /*tp_call*/
+    0,                  /*tp_str*/
+    0,                  /*tp_getattro*/
+    0,                  /*tp_setattro*/
+    0,                  /*tp_as_buffer*/
+    Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE, /*tp_flags*/
+    hashtype_doc,       /*tp_doc*/
+    0,                  /*tp_traverse*/
+    0,			/*tp_clear*/
+    0,			/*tp_richcompare*/
+    0,			/*tp_weaklistoffset*/
+    0,			/*tp_iter*/
+    0,			/*tp_iternext*/
+    EVP_methods,	/* tp_methods */
+    EVP_members,	/* tp_members */
+    EVP_getseters,      /* tp_getset */
+#if 1
+    0,                  /* tp_base */
+    0,                  /* tp_dict */
+    0,                  /* tp_descr_get */
+    0,                  /* tp_descr_set */
+    0,                  /* tp_dictoffset */
+#endif
+#if HASH_OBJ_CONSTRUCTOR
+    (initproc)EVP_tp_init, /* tp_init */
+#endif
+};
+
+static PyObject *
+EVPnew(PyObject *name_obj,
+       const EVP_MD *digest, const EVP_MD_CTX *initial_ctx,
+       const unsigned char *cp, Py_ssize_t len)
+{
+    EVPobject *self;
+
+    if (!digest && !initial_ctx) {
+        PyErr_SetString(PyExc_ValueError, "unsupported hash type");
+        return NULL;
+    }
+
+    if ((self = newEVPobject(name_obj)) == NULL)
+        return NULL;
+
+    if (initial_ctx) {
+        EVP_MD_CTX_copy(&self->ctx, initial_ctx);
+    } else {
+        EVP_DigestInit(&self->ctx, digest);
+    }
+
+    if (cp && len) {
+        if (len > 0 && len <= MUNCH_SIZE) {
+            EVP_DigestUpdate(&self->ctx, cp, Py_SAFE_DOWNCAST(len, Py_ssize_t,
+                                                              unsigned int));
+        } else {
+            Py_ssize_t offset = 0;
+            while (len) {
+                unsigned int process = len > MUNCH_SIZE ? MUNCH_SIZE : len;
+                EVP_DigestUpdate(&self->ctx, cp + offset, process);
+                len -= process;
+                offset += process;
+            }
+        }
+    }
+
+    return (PyObject *)self;
+}
+
+
+/* The module-level function: new() */
+
+PyDoc_STRVAR(EVP_new__doc__,
+"Return a new hash object using the named algorithm.\n\
+An optional string argument may be provided and will be\n\
+automatically hashed.\n\
+\n\
+The MD5 and SHA1 algorithms are always supported.\n");
+
+static PyObject *
+EVP_new(PyObject *self, PyObject *args, PyObject *kwdict)
+{
+    static char *kwlist[] = {"name", "string", NULL};
+    PyObject *name_obj = NULL;
+    char *name;
+    const EVP_MD *digest;
+    unsigned char *cp = NULL;
+    Py_ssize_t len = 0;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwdict, "O|s#:new", kwlist,
+                                     &name_obj, &cp, &len)) {
+        return NULL;
+    }
+
+    if (!PyArg_Parse(name_obj, "s", &name)) {
+        PyErr_SetString(PyExc_TypeError, "name must be a string");
+        return NULL;
+    }
+
+    digest = EVP_get_digestbyname(name);
+
+    return EVPnew(name_obj, digest, NULL, cp, len);
+}
+
+/*
+ *  This macro generates constructor function definitions for specific
+ *  hash algorithms.  These constructors are much faster than calling
+ *  the generic one passing it a python string and are noticably
+ *  faster than calling a python new() wrapper.  Thats important for
+ *  code that wants to make hashes of a bunch of small strings.
+ */
+#define GEN_CONSTRUCTOR(NAME)  \
+    static PyObject * \
+    EVP_new_ ## NAME (PyObject *self, PyObject *args) \
+    { \
+        unsigned char *cp = NULL; \
+        Py_ssize_t len = 0; \
+     \
+        if (!PyArg_ParseTuple(args, "|s#:" #NAME , &cp, &len)) { \
+            return NULL; \
+        } \
+     \
+        return EVPnew( \
+                CONST_ ## NAME ## _name_obj, \
+                NULL, \
+                CONST_new_ ## NAME ## _ctx_p, \
+                cp, len); \
+    }
+
+/* a PyMethodDef structure for the constructor */
+#define CONSTRUCTOR_METH_DEF(NAME)  \
+    {"openssl_" #NAME, (PyCFunction)EVP_new_ ## NAME, METH_VARARGS, \
+        PyDoc_STR("Returns a " #NAME \
+                  " hash object; optionally initialized with a string") \
+    }
+
+/* used in the init function to setup a constructor */
+#define INIT_CONSTRUCTOR_CONSTANTS(NAME)  do { \
+    CONST_ ## NAME ## _name_obj = PyString_FromString(#NAME); \
+    if (EVP_get_digestbyname(#NAME)) { \
+        CONST_new_ ## NAME ## _ctx_p = &CONST_new_ ## NAME ## _ctx; \
+        EVP_DigestInit(CONST_new_ ## NAME ## _ctx_p, EVP_get_digestbyname(#NAME)); \
+    } \
+} while (0);
+
+GEN_CONSTRUCTOR(md5)
+GEN_CONSTRUCTOR(sha1)
+GEN_CONSTRUCTOR(sha224)
+GEN_CONSTRUCTOR(sha256)
+GEN_CONSTRUCTOR(sha384)
+GEN_CONSTRUCTOR(sha512)
+
+/* List of functions exported by this module */
+
+static struct PyMethodDef EVP_functions[] = {
+    {"new", (PyCFunction)EVP_new, METH_VARARGS|METH_KEYWORDS, EVP_new__doc__},
+    CONSTRUCTOR_METH_DEF(md5),
+    CONSTRUCTOR_METH_DEF(sha1),
+    CONSTRUCTOR_METH_DEF(sha224),
+    CONSTRUCTOR_METH_DEF(sha256),
+    CONSTRUCTOR_METH_DEF(sha384),
+    CONSTRUCTOR_METH_DEF(sha512),
+    {NULL,	NULL}		 /* Sentinel */
+};
+
+
+/* Initialize this module. */
+
+PyMODINIT_FUNC
+init_hashlib(void)
+{
+    PyObject *m;
+
+    OpenSSL_add_all_digests();
+
+    /* TODO build EVP_functions openssl_* entries dynamically based
+     * on what hashes are supported rather than listing many
+     * but having some be unsupported.  Only init appropriate
+     * constants. */
+
+    EVPtype.ob_type = &PyType_Type;
+    if (PyType_Ready(&EVPtype) < 0)
+        return;
+
+    m = Py_InitModule("_hashlib", EVP_functions);
+    if (m == NULL)
+        return;
+
+#if HASH_OBJ_CONSTRUCTOR
+    Py_INCREF(&EVPtype);
+    PyModule_AddObject(m, "HASH", (PyObject *)&EVPtype);
+#endif
+
+    /* these constants are used by the convenience constructors */
+    INIT_CONSTRUCTOR_CONSTANTS(md5);
+    INIT_CONSTRUCTOR_CONSTANTS(sha1);
+    INIT_CONSTRUCTOR_CONSTANTS(sha224);
+    INIT_CONSTRUCTOR_CONSTANTS(sha256);
+    INIT_CONSTRUCTOR_CONSTANTS(sha384);
+    INIT_CONSTRUCTOR_CONSTANTS(sha512);
+}
diff --git a/src/Modules/md5.c b/src/Modules/md5.c
new file mode 100644
--- /dev/null
+++ b/src/Modules/md5.c
@@ -0,0 +1,381 @@
+/*
+  Copyright (C) 1999, 2000, 2002 Aladdin Enterprises.  All rights reserved.
+
+  This software is provided 'as-is', without any express or implied
+  warranty.  In no event will the authors be held liable for any damages
+  arising from the use of this software.
+
+  Permission is granted to anyone to use this software for any purpose,
+  including commercial applications, and to alter it and redistribute it
+  freely, subject to the following restrictions:
+
+  1. The origin of this software must not be misrepresented; you must not
+     claim that you wrote the original software. If you use this software
+     in a product, an acknowledgment in the product documentation would be
+     appreciated but is not required.
+  2. Altered source versions must be plainly marked as such, and must not be
+     misrepresented as being the original software.
+  3. This notice may not be removed or altered from any source distribution.
+
+  L. Peter Deutsch
+  ghost at aladdin.com
+
+ */
+/* $Id: md5.c,v 1.6 2002/04/13 19:20:28 lpd Exp $ */
+/*
+  Independent implementation of MD5 (RFC 1321).
+
+  This code implements the MD5 Algorithm defined in RFC 1321, whose
+  text is available at
+	http://www.ietf.org/rfc/rfc1321.txt
+  The code is derived from the text of the RFC, including the test suite
+  (section A.5) but excluding the rest of Appendix A.  It does not include
+  any code or documentation that is identified in the RFC as being
+  copyrighted.
+
+  The original and principal author of md5.c is L. Peter Deutsch
+  <ghost at aladdin.com>.  Other authors are noted in the change history
+  that follows (in reverse chronological order):
+
+  2002-04-13 lpd Clarified derivation from RFC 1321; now handles byte order
+	either statically or dynamically; added missing #include <string.h>
+	in library.
+  2002-03-11 lpd Corrected argument list for main(), and added int return
+	type, in test program and T value program.
+  2002-02-21 lpd Added missing #include <stdio.h> in test program.
+  2000-07-03 lpd Patched to eliminate warnings about "constant is
+	unsigned in ANSI C, signed in traditional"; made test program
+	self-checking.
+  1999-11-04 lpd Edited comments slightly for automatic TOC extraction.
+  1999-10-18 lpd Fixed typo in header comment (ansi2knr rather than md5).
+  1999-05-03 lpd Original version.
+ */
+
+#include "md5.h"
+#include <string.h>
+
+#undef BYTE_ORDER	/* 1 = big-endian, -1 = little-endian, 0 = unknown */
+#ifdef ARCH_IS_BIG_ENDIAN
+#  define BYTE_ORDER (ARCH_IS_BIG_ENDIAN ? 1 : -1)
+#else
+#  define BYTE_ORDER 0
+#endif
+
+#define T_MASK ((md5_word_t)~0)
+#define T1 /* 0xd76aa478 */ (T_MASK ^ 0x28955b87)
+#define T2 /* 0xe8c7b756 */ (T_MASK ^ 0x173848a9)
+#define T3    0x242070db
+#define T4 /* 0xc1bdceee */ (T_MASK ^ 0x3e423111)
+#define T5 /* 0xf57c0faf */ (T_MASK ^ 0x0a83f050)
+#define T6    0x4787c62a
+#define T7 /* 0xa8304613 */ (T_MASK ^ 0x57cfb9ec)
+#define T8 /* 0xfd469501 */ (T_MASK ^ 0x02b96afe)
+#define T9    0x698098d8
+#define T10 /* 0x8b44f7af */ (T_MASK ^ 0x74bb0850)
+#define T11 /* 0xffff5bb1 */ (T_MASK ^ 0x0000a44e)
+#define T12 /* 0x895cd7be */ (T_MASK ^ 0x76a32841)
+#define T13    0x6b901122
+#define T14 /* 0xfd987193 */ (T_MASK ^ 0x02678e6c)
+#define T15 /* 0xa679438e */ (T_MASK ^ 0x5986bc71)
+#define T16    0x49b40821
+#define T17 /* 0xf61e2562 */ (T_MASK ^ 0x09e1da9d)
+#define T18 /* 0xc040b340 */ (T_MASK ^ 0x3fbf4cbf)
+#define T19    0x265e5a51
+#define T20 /* 0xe9b6c7aa */ (T_MASK ^ 0x16493855)
+#define T21 /* 0xd62f105d */ (T_MASK ^ 0x29d0efa2)
+#define T22    0x02441453
+#define T23 /* 0xd8a1e681 */ (T_MASK ^ 0x275e197e)
+#define T24 /* 0xe7d3fbc8 */ (T_MASK ^ 0x182c0437)
+#define T25    0x21e1cde6
+#define T26 /* 0xc33707d6 */ (T_MASK ^ 0x3cc8f829)
+#define T27 /* 0xf4d50d87 */ (T_MASK ^ 0x0b2af278)
+#define T28    0x455a14ed
+#define T29 /* 0xa9e3e905 */ (T_MASK ^ 0x561c16fa)
+#define T30 /* 0xfcefa3f8 */ (T_MASK ^ 0x03105c07)
+#define T31    0x676f02d9
+#define T32 /* 0x8d2a4c8a */ (T_MASK ^ 0x72d5b375)
+#define T33 /* 0xfffa3942 */ (T_MASK ^ 0x0005c6bd)
+#define T34 /* 0x8771f681 */ (T_MASK ^ 0x788e097e)
+#define T35    0x6d9d6122
+#define T36 /* 0xfde5380c */ (T_MASK ^ 0x021ac7f3)
+#define T37 /* 0xa4beea44 */ (T_MASK ^ 0x5b4115bb)
+#define T38    0x4bdecfa9
+#define T39 /* 0xf6bb4b60 */ (T_MASK ^ 0x0944b49f)
+#define T40 /* 0xbebfbc70 */ (T_MASK ^ 0x4140438f)
+#define T41    0x289b7ec6
+#define T42 /* 0xeaa127fa */ (T_MASK ^ 0x155ed805)
+#define T43 /* 0xd4ef3085 */ (T_MASK ^ 0x2b10cf7a)
+#define T44    0x04881d05
+#define T45 /* 0xd9d4d039 */ (T_MASK ^ 0x262b2fc6)
+#define T46 /* 0xe6db99e5 */ (T_MASK ^ 0x1924661a)
+#define T47    0x1fa27cf8
+#define T48 /* 0xc4ac5665 */ (T_MASK ^ 0x3b53a99a)
+#define T49 /* 0xf4292244 */ (T_MASK ^ 0x0bd6ddbb)
+#define T50    0x432aff97
+#define T51 /* 0xab9423a7 */ (T_MASK ^ 0x546bdc58)
+#define T52 /* 0xfc93a039 */ (T_MASK ^ 0x036c5fc6)
+#define T53    0x655b59c3
+#define T54 /* 0x8f0ccc92 */ (T_MASK ^ 0x70f3336d)
+#define T55 /* 0xffeff47d */ (T_MASK ^ 0x00100b82)
+#define T56 /* 0x85845dd1 */ (T_MASK ^ 0x7a7ba22e)
+#define T57    0x6fa87e4f
+#define T58 /* 0xfe2ce6e0 */ (T_MASK ^ 0x01d3191f)
+#define T59 /* 0xa3014314 */ (T_MASK ^ 0x5cfebceb)
+#define T60    0x4e0811a1
+#define T61 /* 0xf7537e82 */ (T_MASK ^ 0x08ac817d)
+#define T62 /* 0xbd3af235 */ (T_MASK ^ 0x42c50dca)
+#define T63    0x2ad7d2bb
+#define T64 /* 0xeb86d391 */ (T_MASK ^ 0x14792c6e)
+
+
+static void
+md5_process(md5_state_t *pms, const md5_byte_t *data /*[64]*/)
+{
+    md5_word_t
+	a = pms->abcd[0], b = pms->abcd[1],
+	c = pms->abcd[2], d = pms->abcd[3];
+    md5_word_t t;
+#if BYTE_ORDER > 0
+    /* Define storage only for big-endian CPUs. */
+    md5_word_t X[16];
+#else
+    /* Define storage for little-endian or both types of CPUs. */
+    md5_word_t xbuf[16];
+    const md5_word_t *X;
+#endif
+
+    {
+#if BYTE_ORDER == 0
+	/*
+	 * Determine dynamically whether this is a big-endian or
+	 * little-endian machine, since we can use a more efficient
+	 * algorithm on the latter.
+	 */
+	static const int w = 1;
+
+	if (*((const md5_byte_t *)&w)) /* dynamic little-endian */
+#endif
+#if BYTE_ORDER <= 0		/* little-endian */
+	{
+	    /*
+	     * On little-endian machines, we can process properly aligned
+	     * data without copying it.
+	     */
+	    if (!((data - (const md5_byte_t *)0) & 3)) {
+		/* data are properly aligned */
+		X = (const md5_word_t *)data;
+	    } else {
+		/* not aligned */
+		memcpy(xbuf, data, 64);
+		X = xbuf;
+	    }
+	}
+#endif
+#if BYTE_ORDER == 0
+	else			/* dynamic big-endian */
+#endif
+#if BYTE_ORDER >= 0		/* big-endian */
+	{
+	    /*
+	     * On big-endian machines, we must arrange the bytes in the
+	     * right order.
+	     */
+	    const md5_byte_t *xp = data;
+	    int i;
+
+#  if BYTE_ORDER == 0
+	    X = xbuf;		/* (dynamic only) */
+#  else
+#    define xbuf X		/* (static only) */
+#  endif
+	    for (i = 0; i < 16; ++i, xp += 4)
+		xbuf[i] = xp[0] + (xp[1] << 8) + (xp[2] << 16) + (xp[3] << 24);
+	}
+#endif
+    }
+
+#define ROTATE_LEFT(x, n) (((x) << (n)) | ((x) >> (32 - (n))))
+
+    /* Round 1. */
+    /* Let [abcd k s i] denote the operation
+       a = b + ((a + F(b,c,d) + X[k] + T[i]) <<< s). */
+#define F(x, y, z) (((x) & (y)) | (~(x) & (z)))
+#define SET(a, b, c, d, k, s, Ti)\
+  t = a + F(b,c,d) + X[k] + Ti;\
+  a = ROTATE_LEFT(t, s) + b
+    /* Do the following 16 operations. */
+    SET(a, b, c, d,  0,  7,  T1);
+    SET(d, a, b, c,  1, 12,  T2);
+    SET(c, d, a, b,  2, 17,  T3);
+    SET(b, c, d, a,  3, 22,  T4);
+    SET(a, b, c, d,  4,  7,  T5);
+    SET(d, a, b, c,  5, 12,  T6);
+    SET(c, d, a, b,  6, 17,  T7);
+    SET(b, c, d, a,  7, 22,  T8);
+    SET(a, b, c, d,  8,  7,  T9);
+    SET(d, a, b, c,  9, 12, T10);
+    SET(c, d, a, b, 10, 17, T11);
+    SET(b, c, d, a, 11, 22, T12);
+    SET(a, b, c, d, 12,  7, T13);
+    SET(d, a, b, c, 13, 12, T14);
+    SET(c, d, a, b, 14, 17, T15);
+    SET(b, c, d, a, 15, 22, T16);
+#undef SET
+
+     /* Round 2. */
+     /* Let [abcd k s i] denote the operation
+          a = b + ((a + G(b,c,d) + X[k] + T[i]) <<< s). */
+#define G(x, y, z) (((x) & (z)) | ((y) & ~(z)))
+#define SET(a, b, c, d, k, s, Ti)\
+  t = a + G(b,c,d) + X[k] + Ti;\
+  a = ROTATE_LEFT(t, s) + b
+     /* Do the following 16 operations. */
+    SET(a, b, c, d,  1,  5, T17);
+    SET(d, a, b, c,  6,  9, T18);
+    SET(c, d, a, b, 11, 14, T19);
+    SET(b, c, d, a,  0, 20, T20);
+    SET(a, b, c, d,  5,  5, T21);
+    SET(d, a, b, c, 10,  9, T22);
+    SET(c, d, a, b, 15, 14, T23);
+    SET(b, c, d, a,  4, 20, T24);
+    SET(a, b, c, d,  9,  5, T25);
+    SET(d, a, b, c, 14,  9, T26);
+    SET(c, d, a, b,  3, 14, T27);
+    SET(b, c, d, a,  8, 20, T28);
+    SET(a, b, c, d, 13,  5, T29);
+    SET(d, a, b, c,  2,  9, T30);
+    SET(c, d, a, b,  7, 14, T31);
+    SET(b, c, d, a, 12, 20, T32);
+#undef SET
+
+     /* Round 3. */
+     /* Let [abcd k s t] denote the operation
+          a = b + ((a + H(b,c,d) + X[k] + T[i]) <<< s). */
+#define H(x, y, z) ((x) ^ (y) ^ (z))
+#define SET(a, b, c, d, k, s, Ti)\
+  t = a + H(b,c,d) + X[k] + Ti;\
+  a = ROTATE_LEFT(t, s) + b
+     /* Do the following 16 operations. */
+    SET(a, b, c, d,  5,  4, T33);
+    SET(d, a, b, c,  8, 11, T34);
+    SET(c, d, a, b, 11, 16, T35);
+    SET(b, c, d, a, 14, 23, T36);
+    SET(a, b, c, d,  1,  4, T37);
+    SET(d, a, b, c,  4, 11, T38);
+    SET(c, d, a, b,  7, 16, T39);
+    SET(b, c, d, a, 10, 23, T40);
+    SET(a, b, c, d, 13,  4, T41);
+    SET(d, a, b, c,  0, 11, T42);
+    SET(c, d, a, b,  3, 16, T43);
+    SET(b, c, d, a,  6, 23, T44);
+    SET(a, b, c, d,  9,  4, T45);
+    SET(d, a, b, c, 12, 11, T46);
+    SET(c, d, a, b, 15, 16, T47);
+    SET(b, c, d, a,  2, 23, T48);
+#undef SET
+
+     /* Round 4. */
+     /* Let [abcd k s t] denote the operation
+          a = b + ((a + I(b,c,d) + X[k] + T[i]) <<< s). */
+#define I(x, y, z) ((y) ^ ((x) | ~(z)))
+#define SET(a, b, c, d, k, s, Ti)\
+  t = a + I(b,c,d) + X[k] + Ti;\
+  a = ROTATE_LEFT(t, s) + b
+     /* Do the following 16 operations. */
+    SET(a, b, c, d,  0,  6, T49);
+    SET(d, a, b, c,  7, 10, T50);
+    SET(c, d, a, b, 14, 15, T51);
+    SET(b, c, d, a,  5, 21, T52);
+    SET(a, b, c, d, 12,  6, T53);
+    SET(d, a, b, c,  3, 10, T54);
+    SET(c, d, a, b, 10, 15, T55);
+    SET(b, c, d, a,  1, 21, T56);
+    SET(a, b, c, d,  8,  6, T57);
+    SET(d, a, b, c, 15, 10, T58);
+    SET(c, d, a, b,  6, 15, T59);
+    SET(b, c, d, a, 13, 21, T60);
+    SET(a, b, c, d,  4,  6, T61);
+    SET(d, a, b, c, 11, 10, T62);
+    SET(c, d, a, b,  2, 15, T63);
+    SET(b, c, d, a,  9, 21, T64);
+#undef SET
+
+     /* Then perform the following additions. (That is increment each
+        of the four registers by the value it had before this block
+        was started.) */
+    pms->abcd[0] += a;
+    pms->abcd[1] += b;
+    pms->abcd[2] += c;
+    pms->abcd[3] += d;
+}
+
+void
+md5_init(md5_state_t *pms)
+{
+    pms->count[0] = pms->count[1] = 0;
+    pms->abcd[0] = 0x67452301;
+    pms->abcd[1] = /*0xefcdab89*/ T_MASK ^ 0x10325476;
+    pms->abcd[2] = /*0x98badcfe*/ T_MASK ^ 0x67452301;
+    pms->abcd[3] = 0x10325476;
+}
+
+void
+md5_append(md5_state_t *pms, const md5_byte_t *data, int nbytes)
+{
+    const md5_byte_t *p = data;
+    int left = nbytes;
+    int offset = (pms->count[0] >> 3) & 63;
+    md5_word_t nbits = (md5_word_t)(nbytes << 3);
+
+    if (nbytes <= 0)
+	return;
+
+    /* Update the message length. */
+    pms->count[1] += nbytes >> 29;
+    pms->count[0] += nbits;
+    if (pms->count[0] < nbits)
+	pms->count[1]++;
+
+    /* Process an initial partial block. */
+    if (offset) {
+	int copy = (offset + nbytes > 64 ? 64 - offset : nbytes);
+
+	memcpy(pms->buf + offset, p, copy);
+	if (offset + copy < 64)
+	    return;
+	p += copy;
+	left -= copy;
+	md5_process(pms, pms->buf);
+    }
+
+    /* Process full blocks. */
+    for (; left >= 64; p += 64, left -= 64)
+	md5_process(pms, p);
+
+    /* Process a final partial block. */
+    if (left)
+	memcpy(pms->buf, p, left);
+}
+
+void
+md5_finish(md5_state_t *pms, md5_byte_t digest[16])
+{
+    static const md5_byte_t pad[64] = {
+	0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
+    };
+    md5_byte_t data[8];
+    int i;
+
+    /* Save the length before padding. */
+    for (i = 0; i < 8; ++i)
+	data[i] = (md5_byte_t)(pms->count[i >> 2] >> ((i & 3) << 3));
+    /* Pad to 56 bytes mod 64. */
+    md5_append(pms, pad, ((55 - (pms->count[0] >> 3)) & 63) + 1);
+    /* Append the length. */
+    md5_append(pms, data, 8);
+    for (i = 0; i < 16; ++i)
+	digest[i] = (md5_byte_t)(pms->abcd[i >> 2] >> ((i & 3) << 3));
+}
diff --git a/src/Modules/md5.h b/src/Modules/md5.h
new file mode 100644
--- /dev/null
+++ b/src/Modules/md5.h
@@ -0,0 +1,91 @@
+/*
+  Copyright (C) 1999, 2002 Aladdin Enterprises.  All rights reserved.
+
+  This software is provided 'as-is', without any express or implied
+  warranty.  In no event will the authors be held liable for any damages
+  arising from the use of this software.
+
+  Permission is granted to anyone to use this software for any purpose,
+  including commercial applications, and to alter it and redistribute it
+  freely, subject to the following restrictions:
+
+  1. The origin of this software must not be misrepresented; you must not
+     claim that you wrote the original software. If you use this software
+     in a product, an acknowledgment in the product documentation would be
+     appreciated but is not required.
+  2. Altered source versions must be plainly marked as such, and must not be
+     misrepresented as being the original software.
+  3. This notice may not be removed or altered from any source distribution.
+
+  L. Peter Deutsch
+  ghost at aladdin.com
+
+ */
+/* $Id: md5.h 43594 2006-04-03 16:27:50Z matthias.klose $ */
+/*
+  Independent implementation of MD5 (RFC 1321).
+
+  This code implements the MD5 Algorithm defined in RFC 1321, whose
+  text is available at
+	http://www.ietf.org/rfc/rfc1321.txt
+  The code is derived from the text of the RFC, including the test suite
+  (section A.5) but excluding the rest of Appendix A.  It does not include
+  any code or documentation that is identified in the RFC as being
+  copyrighted.
+
+  The original and principal author of md5.h is L. Peter Deutsch
+  <ghost at aladdin.com>.  Other authors are noted in the change history
+  that follows (in reverse chronological order):
+
+  2002-04-13 lpd Removed support for non-ANSI compilers; removed
+	references to Ghostscript; clarified derivation from RFC 1321;
+	now handles byte order either statically or dynamically.
+  1999-11-04 lpd Edited comments slightly for automatic TOC extraction.
+  1999-10-18 lpd Fixed typo in header comment (ansi2knr rather than md5);
+	added conditionalization for C++ compilation from Martin
+	Purschke <purschke at bnl.gov>.
+  1999-05-03 lpd Original version.
+ */
+
+#ifndef md5_INCLUDED
+#  define md5_INCLUDED
+
+/*
+ * This package supports both compile-time and run-time determination of CPU
+ * byte order.  If ARCH_IS_BIG_ENDIAN is defined as 0, the code will be
+ * compiled to run only on little-endian CPUs; if ARCH_IS_BIG_ENDIAN is
+ * defined as non-zero, the code will be compiled to run only on big-endian
+ * CPUs; if ARCH_IS_BIG_ENDIAN is not defined, the code will be compiled to
+ * run on either big- or little-endian CPUs, but will run slightly less
+ * efficiently on either one than if ARCH_IS_BIG_ENDIAN is defined.
+ */
+
+typedef unsigned char md5_byte_t; /* 8-bit byte */
+typedef unsigned int md5_word_t; /* 32-bit word */
+
+/* Define the state of the MD5 Algorithm. */
+typedef struct md5_state_s {
+    md5_word_t count[2];	/* message length in bits, lsw first */
+    md5_word_t abcd[4];		/* digest buffer */
+    md5_byte_t buf[64];		/* accumulate block */
+} md5_state_t;
+
+#ifdef __cplusplus
+extern "C" 
+{
+#endif
+
+/* Initialize the algorithm. */
+void md5_init(md5_state_t *pms);
+
+/* Append a string to the message. */
+void md5_append(md5_state_t *pms, const md5_byte_t *data, int nbytes);
+
+/* Finish the message and return the digest. */
+void md5_finish(md5_state_t *pms, md5_byte_t digest[16]);
+
+#ifdef __cplusplus
+}  /* end extern "C" */
+#endif
+
+#endif /* md5_INCLUDED */
diff --git a/src/Modules/md5module.c b/src/Modules/md5module.c
new file mode 100644
--- /dev/null
+++ b/src/Modules/md5module.c
@@ -0,0 +1,312 @@
+
+/* MD5 module */
+
+/* This module provides an interface to the RSA Data Security,
+   Inc. MD5 Message-Digest Algorithm, described in RFC 1321.
+   It requires the files md5c.c and md5.h (which are slightly changed
+   from the versions in the RFC to avoid the "global.h" file.) */
+
+
+/* MD5 objects */
+
+#include "Python.h"
+#include "structmember.h"
+#include "md5.h"
+
+typedef struct {
+	PyObject_HEAD
+        md5_state_t	md5;		/* the context holder */
+} md5object;
+
+static PyTypeObject MD5type;
+
+#define is_md5object(v)		((v)->ob_type == &MD5type)
+
+static md5object *
+newmd5object(void)
+{
+	md5object *md5p;
+
+	md5p = PyObject_New(md5object, &MD5type);
+	if (md5p == NULL)
+		return NULL;
+
+	md5_init(&md5p->md5);	/* actual initialisation */
+	return md5p;
+}
+
+
+/* MD5 methods */
+
+static void
+md5_dealloc(md5object *md5p)
+{
+	PyObject_Del(md5p);
+}
+
+
+/* MD5 methods-as-attributes */
+
+static PyObject *
+md5_update(md5object *self, PyObject *args)
+{
+	unsigned char *cp;
+	int len;
+
+	if (!PyArg_ParseTuple(args, "s#:update", &cp, &len))
+		return NULL;
+
+	md5_append(&self->md5, cp, len);
+
+	Py_INCREF(Py_None);
+	return Py_None;
+}
+
+PyDoc_STRVAR(update_doc,
+"update (arg)\n\
+\n\
+Update the md5 object with the string arg. Repeated calls are\n\
+equivalent to a single call with the concatenation of all the\n\
+arguments.");
+
+
+static PyObject *
+md5_digest(md5object *self)
+{
+ 	md5_state_t mdContext;
+	unsigned char aDigest[16];
+
+	/* make a temporary copy, and perform the final */
+	mdContext = self->md5;
+	md5_finish(&mdContext, aDigest);
+
+	return PyString_FromStringAndSize((char *)aDigest, 16);
+}
+
+PyDoc_STRVAR(digest_doc,
+"digest() -> string\n\
+\n\
+Return the digest of the strings passed to the update() method so\n\
+far. This is a 16-byte string which may contain non-ASCII characters,\n\
+including null bytes.");
+
+
+static PyObject *
+md5_hexdigest(md5object *self)
+{
+ 	md5_state_t mdContext;
+	unsigned char digest[16];
+	unsigned char hexdigest[32];
+	int i, j;
+
+	/* make a temporary copy, and perform the final */
+	mdContext = self->md5;
+	md5_finish(&mdContext, digest);
+
+	/* Make hex version of the digest */
+	for(i=j=0; i<16; i++) {
+		char c;
+		c = (digest[i] >> 4) & 0xf;
+		c = (c>9) ? c+'a'-10 : c + '0';
+		hexdigest[j++] = c;
+		c = (digest[i] & 0xf);
+		c = (c>9) ? c+'a'-10 : c + '0';
+		hexdigest[j++] = c;
+	}
+	return PyString_FromStringAndSize((char*)hexdigest, 32);
+}
+
+
+PyDoc_STRVAR(hexdigest_doc,
+"hexdigest() -> string\n\
+\n\
+Like digest(), but returns the digest as a string of hexadecimal digits.");
+
+
+static PyObject *
+md5_copy(md5object *self)
+{
+	md5object *md5p;
+
+	if ((md5p = newmd5object()) == NULL)
+		return NULL;
+
+	md5p->md5 = self->md5;
+
+	return (PyObject *)md5p;
+}
+
+PyDoc_STRVAR(copy_doc,
+"copy() -> md5 object\n\
+\n\
+Return a copy (``clone'') of the md5 object.");
+
+
+static PyMethodDef md5_methods[] = {
+	{"update",    (PyCFunction)md5_update,    METH_VARARGS, update_doc},
+	{"digest",    (PyCFunction)md5_digest,    METH_NOARGS,  digest_doc},
+	{"hexdigest", (PyCFunction)md5_hexdigest, METH_NOARGS,  hexdigest_doc},
+	{"copy",      (PyCFunction)md5_copy,      METH_NOARGS,  copy_doc},
+	{NULL, NULL}			     /* sentinel */
+};
+
+static PyObject *
+md5_get_block_size(PyObject *self, void *closure)
+{
+    return PyInt_FromLong(64);
+}
+
+static PyObject *
+md5_get_digest_size(PyObject *self, void *closure)
+{
+    return PyInt_FromLong(16);
+}
+
+static PyObject *
+md5_get_name(PyObject *self, void *closure)
+{
+    return PyString_FromStringAndSize("MD5", 3);
+}
+
+static PyGetSetDef md5_getseters[] = {
+    {"digest_size",
+     (getter)md5_get_digest_size, NULL,
+     NULL,
+     NULL},
+    {"block_size",
+     (getter)md5_get_block_size, NULL,
+     NULL,
+     NULL},
+    {"name",
+     (getter)md5_get_name, NULL,
+     NULL,
+     NULL},
+    /* the old md5 and sha modules support 'digest_size' as in PEP 247.
+     * the old sha module also supported 'digestsize'.  ugh. */
+    {"digestsize",
+     (getter)md5_get_digest_size, NULL,
+     NULL,
+     NULL},
+    {NULL}  /* Sentinel */
+};
+
+
+PyDoc_STRVAR(module_doc,
+"This module implements the interface to RSA's MD5 message digest\n\
+algorithm (see also Internet RFC 1321). Its use is quite\n\
+straightforward: use the new() to create an md5 object. You can now\n\
+feed this object with arbitrary strings using the update() method, and\n\
+at any point you can ask it for the digest (a strong kind of 128-bit\n\
+checksum, a.k.a. ``fingerprint'') of the concatenation of the strings\n\
+fed to it so far using the digest() method.\n\
+\n\
+Functions:\n\
+\n\
+new([arg]) -- return a new md5 object, initialized with arg if provided\n\
+md5([arg]) -- DEPRECATED, same as new, but for compatibility\n\
+\n\
+Special Objects:\n\
+\n\
+MD5Type -- type object for md5 objects");
+
+PyDoc_STRVAR(md5type_doc,
+"An md5 represents the object used to calculate the MD5 checksum of a\n\
+string of information.\n\
+\n\
+Methods:\n\
+\n\
+update() -- updates the current digest with an additional string\n\
+digest() -- return the current digest value\n\
+hexdigest() -- return the current digest as a string of hexadecimal digits\n\
+copy() -- return a copy of the current md5 object");
+
+static PyTypeObject MD5type = {
+	PyObject_HEAD_INIT(NULL)
+	0,			  /*ob_size*/
+	"_md5.md5",		  /*tp_name*/
+	sizeof(md5object),	  /*tp_size*/
+	0,			  /*tp_itemsize*/
+	/* methods */
+	(destructor)md5_dealloc,  /*tp_dealloc*/
+	0,			  /*tp_print*/
+	0,                        /*tp_getattr*/
+	0,			  /*tp_setattr*/
+	0,			  /*tp_compare*/
+	0,			  /*tp_repr*/
+        0,			  /*tp_as_number*/
+	0,                        /*tp_as_sequence*/
+	0,			  /*tp_as_mapping*/
+	0, 			  /*tp_hash*/
+	0,			  /*tp_call*/
+	0,			  /*tp_str*/
+	0,			  /*tp_getattro*/
+	0,			  /*tp_setattro*/
+	0,	                  /*tp_as_buffer*/
+	Py_TPFLAGS_DEFAULT,	  /*tp_flags*/
+	md5type_doc,		  /*tp_doc*/
+        0,                        /*tp_traverse*/
+        0,			  /*tp_clear*/
+        0,			  /*tp_richcompare*/
+        0,			  /*tp_weaklistoffset*/
+        0,			  /*tp_iter*/
+        0,			  /*tp_iternext*/
+        md5_methods,	          /*tp_methods*/
+        0,      	          /*tp_members*/
+        md5_getseters,            /*tp_getset*/
+};
+
+
+/* MD5 functions */
+
+static PyObject *
+MD5_new(PyObject *self, PyObject *args)
+{
+	md5object *md5p;
+	unsigned char *cp = NULL;
+	int len = 0;
+
+	if (!PyArg_ParseTuple(args, "|s#:new", &cp, &len))
+		return NULL;
+
+	if ((md5p = newmd5object()) == NULL)
+		return NULL;
+
+	if (cp)
+		md5_append(&md5p->md5, cp, len);
+
+	return (PyObject *)md5p;
+}
+
+PyDoc_STRVAR(new_doc,
+"new([arg]) -> md5 object\n\
+\n\
+Return a new md5 object. If arg is present, the method call update(arg)\n\
+is made.");
+
+
+/* List of functions exported by this module */
+
+static PyMethodDef md5_functions[] = {
+	{"new",		(PyCFunction)MD5_new, METH_VARARGS, new_doc},
+	{NULL,		NULL}	/* Sentinel */
+};
+
+
+/* Initialize this module. */
+
+PyMODINIT_FUNC
+init_md5(void)
+{
+	PyObject *m, *d;
+
+        MD5type.ob_type = &PyType_Type;
+        if (PyType_Ready(&MD5type) < 0)
+            return;
+	m = Py_InitModule3("_md5", md5_functions, module_doc);
+	if (m == NULL)
+	    return;
+	d = PyModule_GetDict(m);
+	PyDict_SetItemString(d, "MD5Type", (PyObject *)&MD5type);
+	PyModule_AddIntConstant(m, "digest_size", 16);
+	/* No need to check the error here, the caller will do that */
+}
diff --git a/src/Modules/sha256module.c b/src/Modules/sha256module.c
new file mode 100644
--- /dev/null
+++ b/src/Modules/sha256module.c
@@ -0,0 +1,701 @@
+/* SHA256 module */
+
+/* This module provides an interface to NIST's SHA-256 and SHA-224 Algorithms */
+
+/* See below for information about the original code this module was
+   based upon. Additional work performed by:
+
+   Andrew Kuchling (amk at amk.ca)
+   Greg Stein (gstein at lyra.org)
+   Trevor Perrin (trevp at trevp.net)
+
+   Copyright (C) 2005   Gregory P. Smith (greg at krypto.org)
+   Licensed to PSF under a Contributor Agreement.
+
+*/
+
+/* SHA objects */
+
+#include "Python.h"
+#include "structmember.h"
+
+
+/* Endianness testing and definitions */
+#define TestEndianness(variable) {int i=1; variable=PCT_BIG_ENDIAN;\
+	if (*((char*)&i)==1) variable=PCT_LITTLE_ENDIAN;}
+
+#define PCT_LITTLE_ENDIAN 1
+#define PCT_BIG_ENDIAN 0
+
+/* Some useful types */
+
+typedef unsigned char SHA_BYTE;
+
+#if SIZEOF_INT == 4
+typedef unsigned int SHA_INT32;	/* 32-bit integer */
+#else
+/* not defined. compilation will die. */
+#endif
+
+/* The SHA block size and message digest sizes, in bytes */
+
+#define SHA_BLOCKSIZE    64
+#define SHA_DIGESTSIZE  32
+
+/* The structure for storing SHA info */
+
+typedef struct {
+    PyObject_HEAD
+    SHA_INT32 digest[8];		/* Message digest */
+    SHA_INT32 count_lo, count_hi;	/* 64-bit bit count */
+    SHA_BYTE data[SHA_BLOCKSIZE];	/* SHA data buffer */
+    int Endianness;
+    int local;				/* unprocessed amount in data */
+    int digestsize;
+} SHAobject;
+
+/* When run on a little-endian CPU we need to perform byte reversal on an
+   array of longwords. */
+
+static void longReverse(SHA_INT32 *buffer, int byteCount, int Endianness)
+{
+    SHA_INT32 value;
+
+    if ( Endianness == PCT_BIG_ENDIAN )
+	return;
+
+    byteCount /= sizeof(*buffer);
+    while (byteCount--) {
+        value = *buffer;
+        value = ( ( value & 0xFF00FF00L ) >> 8  ) | \
+                ( ( value & 0x00FF00FFL ) << 8 );
+        *buffer++ = ( value << 16 ) | ( value >> 16 );
+    }
+}
+
+static void SHAcopy(SHAobject *src, SHAobject *dest)
+{
+    dest->Endianness = src->Endianness;
+    dest->local = src->local;
+    dest->digestsize = src->digestsize;
+    dest->count_lo = src->count_lo;
+    dest->count_hi = src->count_hi;
+    memcpy(dest->digest, src->digest, sizeof(src->digest));
+    memcpy(dest->data, src->data, sizeof(src->data));
+}
+
+
+/* ------------------------------------------------------------------------
+ *
+ * This code for the SHA-256 algorithm was noted as public domain. The
+ * original headers are pasted below.
+ *
+ * Several changes have been made to make it more compatible with the
+ * Python environment and desired interface.
+ *
+ */
+
+/* LibTomCrypt, modular cryptographic library -- Tom St Denis
+ *
+ * LibTomCrypt is a library that provides various cryptographic
+ * algorithms in a highly modular and flexible manner.
+ *
+ * The library is free for all purposes without any express
+ * gurantee it works.
+ *
+ * Tom St Denis, tomstdenis at iahu.ca, http://libtomcrypt.org
+ */
+
+
+/* SHA256 by Tom St Denis */
+
+/* Various logical functions */
+#define ROR(x, y)\
+( ((((unsigned long)(x)&0xFFFFFFFFUL)>>(unsigned long)((y)&31)) | \
+((unsigned long)(x)<<(unsigned long)(32-((y)&31)))) & 0xFFFFFFFFUL)
+#define Ch(x,y,z)       (z ^ (x & (y ^ z)))
+#define Maj(x,y,z)      (((x | y) & z) | (x & y)) 
+#define S(x, n)         ROR((x),(n))
+#define R(x, n)         (((x)&0xFFFFFFFFUL)>>(n))
+#define Sigma0(x)       (S(x, 2) ^ S(x, 13) ^ S(x, 22))
+#define Sigma1(x)       (S(x, 6) ^ S(x, 11) ^ S(x, 25))
+#define Gamma0(x)       (S(x, 7) ^ S(x, 18) ^ R(x, 3))
+#define Gamma1(x)       (S(x, 17) ^ S(x, 19) ^ R(x, 10))
+
+
+static void
+sha_transform(SHAobject *sha_info)
+{
+    int i;
+	SHA_INT32 S[8], W[64], t0, t1;
+
+    memcpy(W, sha_info->data, sizeof(sha_info->data));
+    longReverse(W, (int)sizeof(sha_info->data), sha_info->Endianness);
+
+    for (i = 16; i < 64; ++i) {
+		W[i] = Gamma1(W[i - 2]) + W[i - 7] + Gamma0(W[i - 15]) + W[i - 16];
+    }
+    for (i = 0; i < 8; ++i) {
+        S[i] = sha_info->digest[i];
+    }
+
+    /* Compress */
+#define RND(a,b,c,d,e,f,g,h,i,ki)                    \
+     t0 = h + Sigma1(e) + Ch(e, f, g) + ki + W[i];   \
+     t1 = Sigma0(a) + Maj(a, b, c);                  \
+     d += t0;                                        \
+     h  = t0 + t1;
+
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],0,0x428a2f98);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],1,0x71374491);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],2,0xb5c0fbcf);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],3,0xe9b5dba5);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],4,0x3956c25b);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],5,0x59f111f1);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],6,0x923f82a4);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],7,0xab1c5ed5);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],8,0xd807aa98);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],9,0x12835b01);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],10,0x243185be);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],11,0x550c7dc3);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],12,0x72be5d74);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],13,0x80deb1fe);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],14,0x9bdc06a7);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],15,0xc19bf174);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],16,0xe49b69c1);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],17,0xefbe4786);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],18,0x0fc19dc6);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],19,0x240ca1cc);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],20,0x2de92c6f);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],21,0x4a7484aa);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],22,0x5cb0a9dc);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],23,0x76f988da);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],24,0x983e5152);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],25,0xa831c66d);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],26,0xb00327c8);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],27,0xbf597fc7);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],28,0xc6e00bf3);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],29,0xd5a79147);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],30,0x06ca6351);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],31,0x14292967);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],32,0x27b70a85);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],33,0x2e1b2138);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],34,0x4d2c6dfc);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],35,0x53380d13);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],36,0x650a7354);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],37,0x766a0abb);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],38,0x81c2c92e);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],39,0x92722c85);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],40,0xa2bfe8a1);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],41,0xa81a664b);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],42,0xc24b8b70);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],43,0xc76c51a3);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],44,0xd192e819);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],45,0xd6990624);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],46,0xf40e3585);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],47,0x106aa070);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],48,0x19a4c116);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],49,0x1e376c08);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],50,0x2748774c);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],51,0x34b0bcb5);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],52,0x391c0cb3);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],53,0x4ed8aa4a);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],54,0x5b9cca4f);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],55,0x682e6ff3);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],56,0x748f82ee);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],57,0x78a5636f);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],58,0x84c87814);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],59,0x8cc70208);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],60,0x90befffa);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],61,0xa4506ceb);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],62,0xbef9a3f7);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],63,0xc67178f2);
+
+#undef RND     
+    
+    /* feedback */
+    for (i = 0; i < 8; i++) {
+        sha_info->digest[i] = sha_info->digest[i] + S[i];
+    }
+
+}
+
+
+
+/* initialize the SHA digest */
+
+static void
+sha_init(SHAobject *sha_info)
+{
+    TestEndianness(sha_info->Endianness)
+    sha_info->digest[0] = 0x6A09E667L;
+    sha_info->digest[1] = 0xBB67AE85L;
+    sha_info->digest[2] = 0x3C6EF372L;
+    sha_info->digest[3] = 0xA54FF53AL;
+    sha_info->digest[4] = 0x510E527FL;
+    sha_info->digest[5] = 0x9B05688CL;
+    sha_info->digest[6] = 0x1F83D9ABL;
+    sha_info->digest[7] = 0x5BE0CD19L;
+    sha_info->count_lo = 0L;
+    sha_info->count_hi = 0L;
+    sha_info->local = 0;
+    sha_info->digestsize = 32;
+}
+
+static void
+sha224_init(SHAobject *sha_info)
+{
+    TestEndianness(sha_info->Endianness)
+    sha_info->digest[0] = 0xc1059ed8L;
+    sha_info->digest[1] = 0x367cd507L;
+    sha_info->digest[2] = 0x3070dd17L;
+    sha_info->digest[3] = 0xf70e5939L;
+    sha_info->digest[4] = 0xffc00b31L;
+    sha_info->digest[5] = 0x68581511L;
+    sha_info->digest[6] = 0x64f98fa7L;
+    sha_info->digest[7] = 0xbefa4fa4L;
+    sha_info->count_lo = 0L;
+    sha_info->count_hi = 0L;
+    sha_info->local = 0;
+    sha_info->digestsize = 28;
+}
+
+
+/* update the SHA digest */
+
+static void
+sha_update(SHAobject *sha_info, SHA_BYTE *buffer, int count)
+{
+    int i;
+    SHA_INT32 clo;
+
+    clo = sha_info->count_lo + ((SHA_INT32) count << 3);
+    if (clo < sha_info->count_lo) {
+        ++sha_info->count_hi;
+    }
+    sha_info->count_lo = clo;
+    sha_info->count_hi += (SHA_INT32) count >> 29;
+    if (sha_info->local) {
+        i = SHA_BLOCKSIZE - sha_info->local;
+        if (i > count) {
+            i = count;
+        }
+        memcpy(((SHA_BYTE *) sha_info->data) + sha_info->local, buffer, i);
+        count -= i;
+        buffer += i;
+        sha_info->local += i;
+        if (sha_info->local == SHA_BLOCKSIZE) {
+            sha_transform(sha_info);
+        }
+        else {
+            return;
+        }
+    }
+    while (count >= SHA_BLOCKSIZE) {
+        memcpy(sha_info->data, buffer, SHA_BLOCKSIZE);
+        buffer += SHA_BLOCKSIZE;
+        count -= SHA_BLOCKSIZE;
+        sha_transform(sha_info);
+    }
+    memcpy(sha_info->data, buffer, count);
+    sha_info->local = count;
+}
+
+/* finish computing the SHA digest */
+
+static void
+sha_final(unsigned char digest[SHA_DIGESTSIZE], SHAobject *sha_info)
+{
+    int count;
+    SHA_INT32 lo_bit_count, hi_bit_count;
+
+    lo_bit_count = sha_info->count_lo;
+    hi_bit_count = sha_info->count_hi;
+    count = (int) ((lo_bit_count >> 3) & 0x3f);
+    ((SHA_BYTE *) sha_info->data)[count++] = 0x80;
+    if (count > SHA_BLOCKSIZE - 8) {
+	memset(((SHA_BYTE *) sha_info->data) + count, 0,
+	       SHA_BLOCKSIZE - count);
+	sha_transform(sha_info);
+	memset((SHA_BYTE *) sha_info->data, 0, SHA_BLOCKSIZE - 8);
+    }
+    else {
+	memset(((SHA_BYTE *) sha_info->data) + count, 0,
+	       SHA_BLOCKSIZE - 8 - count);
+    }
+
+    /* GJS: note that we add the hi/lo in big-endian. sha_transform will
+       swap these values into host-order. */
+    sha_info->data[56] = (hi_bit_count >> 24) & 0xff;
+    sha_info->data[57] = (hi_bit_count >> 16) & 0xff;
+    sha_info->data[58] = (hi_bit_count >>  8) & 0xff;
+    sha_info->data[59] = (hi_bit_count >>  0) & 0xff;
+    sha_info->data[60] = (lo_bit_count >> 24) & 0xff;
+    sha_info->data[61] = (lo_bit_count >> 16) & 0xff;
+    sha_info->data[62] = (lo_bit_count >>  8) & 0xff;
+    sha_info->data[63] = (lo_bit_count >>  0) & 0xff;
+    sha_transform(sha_info);
+    digest[ 0] = (unsigned char) ((sha_info->digest[0] >> 24) & 0xff);
+    digest[ 1] = (unsigned char) ((sha_info->digest[0] >> 16) & 0xff);
+    digest[ 2] = (unsigned char) ((sha_info->digest[0] >>  8) & 0xff);
+    digest[ 3] = (unsigned char) ((sha_info->digest[0]      ) & 0xff);
+    digest[ 4] = (unsigned char) ((sha_info->digest[1] >> 24) & 0xff);
+    digest[ 5] = (unsigned char) ((sha_info->digest[1] >> 16) & 0xff);
+    digest[ 6] = (unsigned char) ((sha_info->digest[1] >>  8) & 0xff);
+    digest[ 7] = (unsigned char) ((sha_info->digest[1]      ) & 0xff);
+    digest[ 8] = (unsigned char) ((sha_info->digest[2] >> 24) & 0xff);
+    digest[ 9] = (unsigned char) ((sha_info->digest[2] >> 16) & 0xff);
+    digest[10] = (unsigned char) ((sha_info->digest[2] >>  8) & 0xff);
+    digest[11] = (unsigned char) ((sha_info->digest[2]      ) & 0xff);
+    digest[12] = (unsigned char) ((sha_info->digest[3] >> 24) & 0xff);
+    digest[13] = (unsigned char) ((sha_info->digest[3] >> 16) & 0xff);
+    digest[14] = (unsigned char) ((sha_info->digest[3] >>  8) & 0xff);
+    digest[15] = (unsigned char) ((sha_info->digest[3]      ) & 0xff);
+    digest[16] = (unsigned char) ((sha_info->digest[4] >> 24) & 0xff);
+    digest[17] = (unsigned char) ((sha_info->digest[4] >> 16) & 0xff);
+    digest[18] = (unsigned char) ((sha_info->digest[4] >>  8) & 0xff);
+    digest[19] = (unsigned char) ((sha_info->digest[4]      ) & 0xff);
+    digest[20] = (unsigned char) ((sha_info->digest[5] >> 24) & 0xff);
+    digest[21] = (unsigned char) ((sha_info->digest[5] >> 16) & 0xff);
+    digest[22] = (unsigned char) ((sha_info->digest[5] >>  8) & 0xff);
+    digest[23] = (unsigned char) ((sha_info->digest[5]      ) & 0xff);
+    digest[24] = (unsigned char) ((sha_info->digest[6] >> 24) & 0xff);
+    digest[25] = (unsigned char) ((sha_info->digest[6] >> 16) & 0xff);
+    digest[26] = (unsigned char) ((sha_info->digest[6] >>  8) & 0xff);
+    digest[27] = (unsigned char) ((sha_info->digest[6]      ) & 0xff);
+    digest[28] = (unsigned char) ((sha_info->digest[7] >> 24) & 0xff);
+    digest[29] = (unsigned char) ((sha_info->digest[7] >> 16) & 0xff);
+    digest[30] = (unsigned char) ((sha_info->digest[7] >>  8) & 0xff);
+    digest[31] = (unsigned char) ((sha_info->digest[7]      ) & 0xff);
+}
+
+/*
+ * End of copied SHA code.
+ *
+ * ------------------------------------------------------------------------
+ */
+
+static PyTypeObject SHA224type;
+static PyTypeObject SHA256type;
+
+
+static SHAobject *
+newSHA224object(void)
+{
+    return (SHAobject *)PyObject_New(SHAobject, &SHA224type);
+}
+
+static SHAobject *
+newSHA256object(void)
+{
+    return (SHAobject *)PyObject_New(SHAobject, &SHA256type);
+}
+
+/* Internal methods for a hash object */
+
+static void
+SHA_dealloc(PyObject *ptr)
+{
+    PyObject_Del(ptr);
+}
+
+
+/* External methods for a hash object */
+
+PyDoc_STRVAR(SHA256_copy__doc__, "Return a copy of the hash object.");
+
+static PyObject *
+SHA256_copy(SHAobject *self, PyObject *unused)
+{
+    SHAobject *newobj;
+
+    if (((PyObject*)self)->ob_type == &SHA256type) {
+        if ( (newobj = newSHA256object())==NULL)
+            return NULL;
+    } else {
+        if ( (newobj = newSHA224object())==NULL)
+            return NULL;
+    }
+
+    SHAcopy(self, newobj);
+    return (PyObject *)newobj;
+}
+
+PyDoc_STRVAR(SHA256_digest__doc__,
+"Return the digest value as a string of binary data.");
+
+static PyObject *
+SHA256_digest(SHAobject *self, PyObject *unused)
+{
+    unsigned char digest[SHA_DIGESTSIZE];
+    SHAobject temp;
+
+    SHAcopy(self, &temp);
+    sha_final(digest, &temp);
+    return PyString_FromStringAndSize((const char *)digest, self->digestsize);
+}
+
+PyDoc_STRVAR(SHA256_hexdigest__doc__,
+"Return the digest value as a string of hexadecimal digits.");
+
+static PyObject *
+SHA256_hexdigest(SHAobject *self, PyObject *unused)
+{
+    unsigned char digest[SHA_DIGESTSIZE];
+    SHAobject temp;
+    PyObject *retval;
+    char *hex_digest;
+    int i, j;
+
+    /* Get the raw (binary) digest value */
+    SHAcopy(self, &temp);
+    sha_final(digest, &temp);
+
+    /* Create a new string */
+    retval = PyString_FromStringAndSize(NULL, self->digestsize * 2);
+    if (!retval)
+	    return NULL;
+    hex_digest = PyString_AsString(retval);
+    if (!hex_digest) {
+	    Py_DECREF(retval);
+	    return NULL;
+    }
+
+    /* Make hex version of the digest */
+    for(i=j=0; i<self->digestsize; i++) {
+        char c;
+        c = (digest[i] >> 4) & 0xf;
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+        c = (digest[i] & 0xf);
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+    }
+    return retval;
+}
+
+PyDoc_STRVAR(SHA256_update__doc__,
+"Update this hash object's state with the provided string.");
+
+static PyObject *
+SHA256_update(SHAobject *self, PyObject *args)
+{
+    unsigned char *cp;
+    int len;
+
+    if (!PyArg_ParseTuple(args, "s#:update", &cp, &len))
+        return NULL;
+
+    sha_update(self, cp, len);
+
+    Py_INCREF(Py_None);
+    return Py_None;
+}
+
+static PyMethodDef SHA_methods[] = {
+    {"copy",	  (PyCFunction)SHA256_copy,      METH_NOARGS,  SHA256_copy__doc__},
+    {"digest",	  (PyCFunction)SHA256_digest,    METH_NOARGS,  SHA256_digest__doc__},
+    {"hexdigest", (PyCFunction)SHA256_hexdigest, METH_NOARGS,  SHA256_hexdigest__doc__},
+    {"update",	  (PyCFunction)SHA256_update,    METH_VARARGS, SHA256_update__doc__},
+    {NULL,	  NULL}		/* sentinel */
+};
+
+static PyObject *
+SHA256_get_block_size(PyObject *self, void *closure)
+{
+    return PyInt_FromLong(SHA_BLOCKSIZE);
+}
+
+static PyObject *
+SHA256_get_name(PyObject *self, void *closure)
+{
+    if (((SHAobject *)self)->digestsize == 32)
+        return PyString_FromStringAndSize("SHA256", 6);
+    else
+        return PyString_FromStringAndSize("SHA224", 6);
+}
+
+static PyGetSetDef SHA_getseters[] = {
+    {"block_size",
+     (getter)SHA256_get_block_size, NULL,
+     NULL,
+     NULL},
+    {"name",
+     (getter)SHA256_get_name, NULL,
+     NULL,
+     NULL},
+    {NULL}  /* Sentinel */
+};
+
+static PyMemberDef SHA_members[] = {
+    {"digest_size", T_INT, offsetof(SHAobject, digestsize), READONLY, NULL},
+    /* the old md5 and sha modules support 'digest_size' as in PEP 247.
+     * the old sha module also supported 'digestsize'.  ugh. */
+    {"digestsize", T_INT, offsetof(SHAobject, digestsize), READONLY, NULL},
+    {NULL}  /* Sentinel */
+};
+
+static PyTypeObject SHA224type = {
+    PyObject_HEAD_INIT(NULL)
+    0,			/*ob_size*/
+    "_sha256.sha224",	/*tp_name*/
+    sizeof(SHAobject),	/*tp_size*/
+    0,			/*tp_itemsize*/
+    /* methods */
+    SHA_dealloc,	/*tp_dealloc*/
+    0,			/*tp_print*/
+    0,          	/*tp_getattr*/
+    0,                  /*tp_setattr*/
+    0,                  /*tp_compare*/
+    0,                  /*tp_repr*/
+    0,                  /*tp_as_number*/
+    0,                  /*tp_as_sequence*/
+    0,                  /*tp_as_mapping*/
+    0,                  /*tp_hash*/
+    0,                  /*tp_call*/
+    0,                  /*tp_str*/
+    0,                  /*tp_getattro*/
+    0,                  /*tp_setattro*/
+    0,                  /*tp_as_buffer*/
+    Py_TPFLAGS_DEFAULT, /*tp_flags*/
+    0,                  /*tp_doc*/
+    0,                  /*tp_traverse*/
+    0,			/*tp_clear*/
+    0,			/*tp_richcompare*/
+    0,			/*tp_weaklistoffset*/
+    0,			/*tp_iter*/
+    0,			/*tp_iternext*/
+    SHA_methods,	/* tp_methods */
+    SHA_members,	/* tp_members */
+    SHA_getseters,      /* tp_getset */
+};
+
+static PyTypeObject SHA256type = {
+    PyObject_HEAD_INIT(NULL)
+    0,			/*ob_size*/
+    "_sha256.sha256",	/*tp_name*/
+    sizeof(SHAobject),	/*tp_size*/
+    0,			/*tp_itemsize*/
+    /* methods */
+    SHA_dealloc,	/*tp_dealloc*/
+    0,			/*tp_print*/
+    0,          	/*tp_getattr*/
+    0,                  /*tp_setattr*/
+    0,                  /*tp_compare*/
+    0,                  /*tp_repr*/
+    0,                  /*tp_as_number*/
+    0,                  /*tp_as_sequence*/
+    0,                  /*tp_as_mapping*/
+    0,                  /*tp_hash*/
+    0,                  /*tp_call*/
+    0,                  /*tp_str*/
+    0,                  /*tp_getattro*/
+    0,                  /*tp_setattro*/
+    0,                  /*tp_as_buffer*/
+    Py_TPFLAGS_DEFAULT, /*tp_flags*/
+    0,                  /*tp_doc*/
+    0,                  /*tp_traverse*/
+    0,			/*tp_clear*/
+    0,			/*tp_richcompare*/
+    0,			/*tp_weaklistoffset*/
+    0,			/*tp_iter*/
+    0,			/*tp_iternext*/
+    SHA_methods,	/* tp_methods */
+    SHA_members,	/* tp_members */
+    SHA_getseters,      /* tp_getset */
+};
+
+
+/* The single module-level function: new() */
+
+PyDoc_STRVAR(SHA256_new__doc__,
+"Return a new SHA-256 hash object; optionally initialized with a string.");
+
+static PyObject *
+SHA256_new(PyObject *self, PyObject *args, PyObject *kwdict)
+{
+    static char *kwlist[] = {"string", NULL};
+    SHAobject *new;
+    unsigned char *cp = NULL;
+    int len;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwdict, "|s#:new", kwlist,
+                                     &cp, &len)) {
+        return NULL;
+    }
+
+    if ((new = newSHA256object()) == NULL)
+        return NULL;
+
+    sha_init(new);
+
+    if (PyErr_Occurred()) {
+        Py_DECREF(new);
+        return NULL;
+    }
+    if (cp)
+        sha_update(new, cp, len);
+
+    return (PyObject *)new;
+}
+
+PyDoc_STRVAR(SHA224_new__doc__,
+"Return a new SHA-224 hash object; optionally initialized with a string.");
+
+static PyObject *
+SHA224_new(PyObject *self, PyObject *args, PyObject *kwdict)
+{
+    static char *kwlist[] = {"string", NULL};
+    SHAobject *new;
+    unsigned char *cp = NULL;
+    int len;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwdict, "|s#:new", kwlist,
+                                     &cp, &len)) {
+        return NULL;
+    }
+
+    if ((new = newSHA224object()) == NULL)
+        return NULL;
+
+    sha224_init(new);
+
+    if (PyErr_Occurred()) {
+        Py_DECREF(new);
+        return NULL;
+    }
+    if (cp)
+        sha_update(new, cp, len);
+
+    return (PyObject *)new;
+}
+
+
+/* List of functions exported by this module */
+
+static struct PyMethodDef SHA_functions[] = {
+    {"sha256", (PyCFunction)SHA256_new, METH_VARARGS|METH_KEYWORDS, SHA256_new__doc__},
+    {"sha224", (PyCFunction)SHA224_new, METH_VARARGS|METH_KEYWORDS, SHA224_new__doc__},
+    {NULL,	NULL}		 /* Sentinel */
+};
+
+
+/* Initialize this module. */
+
+#define insint(n,v) { PyModule_AddIntConstant(m,n,v); }
+
+PyMODINIT_FUNC
+init_sha256(void)
+{
+    PyObject *m;
+
+    SHA224type.ob_type = &PyType_Type;
+    if (PyType_Ready(&SHA224type) < 0)
+        return;
+    SHA256type.ob_type = &PyType_Type;
+    if (PyType_Ready(&SHA256type) < 0)
+        return;
+    m = Py_InitModule("_sha256", SHA_functions);
+    if (m == NULL)
+	return;
+}
diff --git a/src/Modules/sha512module.c b/src/Modules/sha512module.c
new file mode 100644
--- /dev/null
+++ b/src/Modules/sha512module.c
@@ -0,0 +1,769 @@
+/* SHA512 module */
+
+/* This module provides an interface to NIST's SHA-512 and SHA-384 Algorithms */
+
+/* See below for information about the original code this module was
+   based upon. Additional work performed by:
+
+   Andrew Kuchling (amk at amk.ca)
+   Greg Stein (gstein at lyra.org)
+   Trevor Perrin (trevp at trevp.net)
+
+   Copyright (C) 2005   Gregory P. Smith (greg at krypto.org)
+   Licensed to PSF under a Contributor Agreement.
+
+*/
+
+/* SHA objects */
+
+#include "Python.h"
+#include "structmember.h"
+
+#ifdef PY_LONG_LONG /* If no PY_LONG_LONG, don't compile anything! */
+
+/* Endianness testing and definitions */
+#define TestEndianness(variable) {int i=1; variable=PCT_BIG_ENDIAN;\
+	if (*((char*)&i)==1) variable=PCT_LITTLE_ENDIAN;}
+
+#define PCT_LITTLE_ENDIAN 1
+#define PCT_BIG_ENDIAN 0
+
+/* Some useful types */
+
+typedef unsigned char SHA_BYTE;
+
+#if SIZEOF_INT == 4
+typedef unsigned int SHA_INT32;	/* 32-bit integer */
+typedef unsigned PY_LONG_LONG SHA_INT64;	/* 64-bit integer */
+#else
+/* not defined. compilation will die. */
+#endif
+
+/* The SHA block size and message digest sizes, in bytes */
+
+#define SHA_BLOCKSIZE   128
+#define SHA_DIGESTSIZE  64
+
+/* The structure for storing SHA info */
+
+typedef struct {
+    PyObject_HEAD
+    SHA_INT64 digest[8];		/* Message digest */
+    SHA_INT32 count_lo, count_hi;	/* 64-bit bit count */
+    SHA_BYTE data[SHA_BLOCKSIZE];	/* SHA data buffer */
+    int Endianness;
+    int local;				/* unprocessed amount in data */
+    int digestsize;
+} SHAobject;
+
+/* When run on a little-endian CPU we need to perform byte reversal on an
+   array of longwords. */
+
+static void longReverse(SHA_INT64 *buffer, int byteCount, int Endianness)
+{
+    SHA_INT64 value;
+
+    if ( Endianness == PCT_BIG_ENDIAN )
+	return;
+
+    byteCount /= sizeof(*buffer);
+    while (byteCount--) {
+        value = *buffer;
+
+		((unsigned char*)buffer)[0] = (unsigned char)(value >> 56) & 0xff;
+		((unsigned char*)buffer)[1] = (unsigned char)(value >> 48) & 0xff;
+		((unsigned char*)buffer)[2] = (unsigned char)(value >> 40) & 0xff;
+		((unsigned char*)buffer)[3] = (unsigned char)(value >> 32) & 0xff;
+		((unsigned char*)buffer)[4] = (unsigned char)(value >> 24) & 0xff;
+		((unsigned char*)buffer)[5] = (unsigned char)(value >> 16) & 0xff;
+		((unsigned char*)buffer)[6] = (unsigned char)(value >>  8) & 0xff;
+		((unsigned char*)buffer)[7] = (unsigned char)(value      ) & 0xff;
+        
+		buffer++;
+    }
+}
+
+static void SHAcopy(SHAobject *src, SHAobject *dest)
+{
+    dest->Endianness = src->Endianness;
+    dest->local = src->local;
+    dest->digestsize = src->digestsize;
+    dest->count_lo = src->count_lo;
+    dest->count_hi = src->count_hi;
+    memcpy(dest->digest, src->digest, sizeof(src->digest));
+    memcpy(dest->data, src->data, sizeof(src->data));
+}
+
+
+/* ------------------------------------------------------------------------
+ *
+ * This code for the SHA-512 algorithm was noted as public domain. The
+ * original headers are pasted below.
+ *
+ * Several changes have been made to make it more compatible with the
+ * Python environment and desired interface.
+ *
+ */
+
+/* LibTomCrypt, modular cryptographic library -- Tom St Denis
+ *
+ * LibTomCrypt is a library that provides various cryptographic
+ * algorithms in a highly modular and flexible manner.
+ *
+ * The library is free for all purposes without any express
+ * gurantee it works.
+ *
+ * Tom St Denis, tomstdenis at iahu.ca, http://libtomcrypt.org
+ */
+
+
+/* SHA512 by Tom St Denis */
+
+/* Various logical functions */
+#define ROR64(x, y) \
+    ( ((((x) & 0xFFFFFFFFFFFFFFFFULL)>>((unsigned PY_LONG_LONG)(y) & 63)) | \
+      ((x)<<((unsigned PY_LONG_LONG)(64-((y) & 63))))) & 0xFFFFFFFFFFFFFFFFULL)
+#define Ch(x,y,z)       (z ^ (x & (y ^ z)))
+#define Maj(x,y,z)      (((x | y) & z) | (x & y)) 
+#define S(x, n)         ROR64((x),(n))
+#define R(x, n)         (((x) & 0xFFFFFFFFFFFFFFFFULL) >> ((unsigned PY_LONG_LONG)n))
+#define Sigma0(x)       (S(x, 28) ^ S(x, 34) ^ S(x, 39))
+#define Sigma1(x)       (S(x, 14) ^ S(x, 18) ^ S(x, 41))
+#define Gamma0(x)       (S(x, 1) ^ S(x, 8) ^ R(x, 7))
+#define Gamma1(x)       (S(x, 19) ^ S(x, 61) ^ R(x, 6))
+
+
+static void
+sha512_transform(SHAobject *sha_info)
+{
+    int i;
+    SHA_INT64 S[8], W[80], t0, t1;
+
+    memcpy(W, sha_info->data, sizeof(sha_info->data));
+    longReverse(W, (int)sizeof(sha_info->data), sha_info->Endianness);
+
+    for (i = 16; i < 80; ++i) {
+		W[i] = Gamma1(W[i - 2]) + W[i - 7] + Gamma0(W[i - 15]) + W[i - 16];
+    }
+    for (i = 0; i < 8; ++i) {
+        S[i] = sha_info->digest[i];
+    }
+
+    /* Compress */
+#define RND(a,b,c,d,e,f,g,h,i,ki)                    \
+     t0 = h + Sigma1(e) + Ch(e, f, g) + ki + W[i];   \
+     t1 = Sigma0(a) + Maj(a, b, c);                  \
+     d += t0;                                        \
+     h  = t0 + t1;
+
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],0,0x428a2f98d728ae22ULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],1,0x7137449123ef65cdULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],2,0xb5c0fbcfec4d3b2fULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],3,0xe9b5dba58189dbbcULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],4,0x3956c25bf348b538ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],5,0x59f111f1b605d019ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],6,0x923f82a4af194f9bULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],7,0xab1c5ed5da6d8118ULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],8,0xd807aa98a3030242ULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],9,0x12835b0145706fbeULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],10,0x243185be4ee4b28cULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],11,0x550c7dc3d5ffb4e2ULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],12,0x72be5d74f27b896fULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],13,0x80deb1fe3b1696b1ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],14,0x9bdc06a725c71235ULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],15,0xc19bf174cf692694ULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],16,0xe49b69c19ef14ad2ULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],17,0xefbe4786384f25e3ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],18,0x0fc19dc68b8cd5b5ULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],19,0x240ca1cc77ac9c65ULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],20,0x2de92c6f592b0275ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],21,0x4a7484aa6ea6e483ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],22,0x5cb0a9dcbd41fbd4ULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],23,0x76f988da831153b5ULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],24,0x983e5152ee66dfabULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],25,0xa831c66d2db43210ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],26,0xb00327c898fb213fULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],27,0xbf597fc7beef0ee4ULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],28,0xc6e00bf33da88fc2ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],29,0xd5a79147930aa725ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],30,0x06ca6351e003826fULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],31,0x142929670a0e6e70ULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],32,0x27b70a8546d22ffcULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],33,0x2e1b21385c26c926ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],34,0x4d2c6dfc5ac42aedULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],35,0x53380d139d95b3dfULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],36,0x650a73548baf63deULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],37,0x766a0abb3c77b2a8ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],38,0x81c2c92e47edaee6ULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],39,0x92722c851482353bULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],40,0xa2bfe8a14cf10364ULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],41,0xa81a664bbc423001ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],42,0xc24b8b70d0f89791ULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],43,0xc76c51a30654be30ULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],44,0xd192e819d6ef5218ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],45,0xd69906245565a910ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],46,0xf40e35855771202aULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],47,0x106aa07032bbd1b8ULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],48,0x19a4c116b8d2d0c8ULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],49,0x1e376c085141ab53ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],50,0x2748774cdf8eeb99ULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],51,0x34b0bcb5e19b48a8ULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],52,0x391c0cb3c5c95a63ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],53,0x4ed8aa4ae3418acbULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],54,0x5b9cca4f7763e373ULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],55,0x682e6ff3d6b2b8a3ULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],56,0x748f82ee5defb2fcULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],57,0x78a5636f43172f60ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],58,0x84c87814a1f0ab72ULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],59,0x8cc702081a6439ecULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],60,0x90befffa23631e28ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],61,0xa4506cebde82bde9ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],62,0xbef9a3f7b2c67915ULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],63,0xc67178f2e372532bULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],64,0xca273eceea26619cULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],65,0xd186b8c721c0c207ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],66,0xeada7dd6cde0eb1eULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],67,0xf57d4f7fee6ed178ULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],68,0x06f067aa72176fbaULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],69,0x0a637dc5a2c898a6ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],70,0x113f9804bef90daeULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],71,0x1b710b35131c471bULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],72,0x28db77f523047d84ULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],73,0x32caab7b40c72493ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],74,0x3c9ebe0a15c9bebcULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],75,0x431d67c49c100d4cULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],76,0x4cc5d4becb3e42b6ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],77,0x597f299cfc657e2aULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],78,0x5fcb6fab3ad6faecULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],79,0x6c44198c4a475817ULL);
+
+#undef RND     
+    
+    /* feedback */
+    for (i = 0; i < 8; i++) {
+        sha_info->digest[i] = sha_info->digest[i] + S[i];
+    }
+
+}
+
+
+
+/* initialize the SHA digest */
+
+static void
+sha512_init(SHAobject *sha_info)
+{
+    TestEndianness(sha_info->Endianness)
+    sha_info->digest[0] = 0x6a09e667f3bcc908ULL;
+    sha_info->digest[1] = 0xbb67ae8584caa73bULL;
+    sha_info->digest[2] = 0x3c6ef372fe94f82bULL;
+    sha_info->digest[3] = 0xa54ff53a5f1d36f1ULL;
+    sha_info->digest[4] = 0x510e527fade682d1ULL;
+    sha_info->digest[5] = 0x9b05688c2b3e6c1fULL;
+    sha_info->digest[6] = 0x1f83d9abfb41bd6bULL;
+    sha_info->digest[7] = 0x5be0cd19137e2179ULL;
+    sha_info->count_lo = 0L;
+    sha_info->count_hi = 0L;
+    sha_info->local = 0;
+    sha_info->digestsize = 64;
+}
+
+static void
+sha384_init(SHAobject *sha_info)
+{
+    TestEndianness(sha_info->Endianness)
+    sha_info->digest[0] = 0xcbbb9d5dc1059ed8ULL;
+    sha_info->digest[1] = 0x629a292a367cd507ULL;
+    sha_info->digest[2] = 0x9159015a3070dd17ULL;
+    sha_info->digest[3] = 0x152fecd8f70e5939ULL;
+    sha_info->digest[4] = 0x67332667ffc00b31ULL;
+    sha_info->digest[5] = 0x8eb44a8768581511ULL;
+    sha_info->digest[6] = 0xdb0c2e0d64f98fa7ULL;
+    sha_info->digest[7] = 0x47b5481dbefa4fa4ULL;
+    sha_info->count_lo = 0L;
+    sha_info->count_hi = 0L;
+    sha_info->local = 0;
+    sha_info->digestsize = 48;
+}
+
+
+/* update the SHA digest */
+
+static void
+sha512_update(SHAobject *sha_info, SHA_BYTE *buffer, int count)
+{
+    int i;
+    SHA_INT32 clo;
+
+    clo = sha_info->count_lo + ((SHA_INT32) count << 3);
+    if (clo < sha_info->count_lo) {
+        ++sha_info->count_hi;
+    }
+    sha_info->count_lo = clo;
+    sha_info->count_hi += (SHA_INT32) count >> 29;
+    if (sha_info->local) {
+        i = SHA_BLOCKSIZE - sha_info->local;
+        if (i > count) {
+            i = count;
+        }
+        memcpy(((SHA_BYTE *) sha_info->data) + sha_info->local, buffer, i);
+        count -= i;
+        buffer += i;
+        sha_info->local += i;
+        if (sha_info->local == SHA_BLOCKSIZE) {
+            sha512_transform(sha_info);
+        }
+        else {
+            return;
+        }
+    }
+    while (count >= SHA_BLOCKSIZE) {
+        memcpy(sha_info->data, buffer, SHA_BLOCKSIZE);
+        buffer += SHA_BLOCKSIZE;
+        count -= SHA_BLOCKSIZE;
+        sha512_transform(sha_info);
+    }
+    memcpy(sha_info->data, buffer, count);
+    sha_info->local = count;
+}
+
+/* finish computing the SHA digest */
+
+static void
+sha512_final(unsigned char digest[SHA_DIGESTSIZE], SHAobject *sha_info)
+{
+    int count;
+    SHA_INT32 lo_bit_count, hi_bit_count;
+
+    lo_bit_count = sha_info->count_lo;
+    hi_bit_count = sha_info->count_hi;
+    count = (int) ((lo_bit_count >> 3) & 0x7f);
+    ((SHA_BYTE *) sha_info->data)[count++] = 0x80;
+    if (count > SHA_BLOCKSIZE - 16) {
+	memset(((SHA_BYTE *) sha_info->data) + count, 0,
+	       SHA_BLOCKSIZE - count);
+	sha512_transform(sha_info);
+	memset((SHA_BYTE *) sha_info->data, 0, SHA_BLOCKSIZE - 16);
+    }
+    else {
+	memset(((SHA_BYTE *) sha_info->data) + count, 0,
+	       SHA_BLOCKSIZE - 16 - count);
+    }
+
+    /* GJS: note that we add the hi/lo in big-endian. sha512_transform will
+       swap these values into host-order. */
+    sha_info->data[112] = 0;
+    sha_info->data[113] = 0;
+    sha_info->data[114] = 0;
+    sha_info->data[115] = 0;
+    sha_info->data[116] = 0;
+    sha_info->data[117] = 0;
+    sha_info->data[118] = 0;
+    sha_info->data[119] = 0;
+    sha_info->data[120] = (hi_bit_count >> 24) & 0xff;
+    sha_info->data[121] = (hi_bit_count >> 16) & 0xff;
+    sha_info->data[122] = (hi_bit_count >>  8) & 0xff;
+    sha_info->data[123] = (hi_bit_count >>  0) & 0xff;
+    sha_info->data[124] = (lo_bit_count >> 24) & 0xff;
+    sha_info->data[125] = (lo_bit_count >> 16) & 0xff;
+    sha_info->data[126] = (lo_bit_count >>  8) & 0xff;
+    sha_info->data[127] = (lo_bit_count >>  0) & 0xff;
+    sha512_transform(sha_info);
+    digest[ 0] = (unsigned char) ((sha_info->digest[0] >> 56) & 0xff);
+    digest[ 1] = (unsigned char) ((sha_info->digest[0] >> 48) & 0xff);
+    digest[ 2] = (unsigned char) ((sha_info->digest[0] >> 40) & 0xff);
+    digest[ 3] = (unsigned char) ((sha_info->digest[0] >> 32) & 0xff);
+    digest[ 4] = (unsigned char) ((sha_info->digest[0] >> 24) & 0xff);
+    digest[ 5] = (unsigned char) ((sha_info->digest[0] >> 16) & 0xff);
+    digest[ 6] = (unsigned char) ((sha_info->digest[0] >>  8) & 0xff);
+    digest[ 7] = (unsigned char) ((sha_info->digest[0]      ) & 0xff);
+    digest[ 8] = (unsigned char) ((sha_info->digest[1] >> 56) & 0xff);
+    digest[ 9] = (unsigned char) ((sha_info->digest[1] >> 48) & 0xff);
+    digest[10] = (unsigned char) ((sha_info->digest[1] >> 40) & 0xff);
+    digest[11] = (unsigned char) ((sha_info->digest[1] >> 32) & 0xff);
+    digest[12] = (unsigned char) ((sha_info->digest[1] >> 24) & 0xff);
+    digest[13] = (unsigned char) ((sha_info->digest[1] >> 16) & 0xff);
+    digest[14] = (unsigned char) ((sha_info->digest[1] >>  8) & 0xff);
+    digest[15] = (unsigned char) ((sha_info->digest[1]      ) & 0xff);
+    digest[16] = (unsigned char) ((sha_info->digest[2] >> 56) & 0xff);
+    digest[17] = (unsigned char) ((sha_info->digest[2] >> 48) & 0xff);
+    digest[18] = (unsigned char) ((sha_info->digest[2] >> 40) & 0xff);
+    digest[19] = (unsigned char) ((sha_info->digest[2] >> 32) & 0xff);
+    digest[20] = (unsigned char) ((sha_info->digest[2] >> 24) & 0xff);
+    digest[21] = (unsigned char) ((sha_info->digest[2] >> 16) & 0xff);
+    digest[22] = (unsigned char) ((sha_info->digest[2] >>  8) & 0xff);
+    digest[23] = (unsigned char) ((sha_info->digest[2]      ) & 0xff);
+    digest[24] = (unsigned char) ((sha_info->digest[3] >> 56) & 0xff);
+    digest[25] = (unsigned char) ((sha_info->digest[3] >> 48) & 0xff);
+    digest[26] = (unsigned char) ((sha_info->digest[3] >> 40) & 0xff);
+    digest[27] = (unsigned char) ((sha_info->digest[3] >> 32) & 0xff);
+    digest[28] = (unsigned char) ((sha_info->digest[3] >> 24) & 0xff);
+    digest[29] = (unsigned char) ((sha_info->digest[3] >> 16) & 0xff);
+    digest[30] = (unsigned char) ((sha_info->digest[3] >>  8) & 0xff);
+    digest[31] = (unsigned char) ((sha_info->digest[3]      ) & 0xff);
+    digest[32] = (unsigned char) ((sha_info->digest[4] >> 56) & 0xff);
+    digest[33] = (unsigned char) ((sha_info->digest[4] >> 48) & 0xff);
+    digest[34] = (unsigned char) ((sha_info->digest[4] >> 40) & 0xff);
+    digest[35] = (unsigned char) ((sha_info->digest[4] >> 32) & 0xff);
+    digest[36] = (unsigned char) ((sha_info->digest[4] >> 24) & 0xff);
+    digest[37] = (unsigned char) ((sha_info->digest[4] >> 16) & 0xff);
+    digest[38] = (unsigned char) ((sha_info->digest[4] >>  8) & 0xff);
+    digest[39] = (unsigned char) ((sha_info->digest[4]      ) & 0xff);
+    digest[40] = (unsigned char) ((sha_info->digest[5] >> 56) & 0xff);
+    digest[41] = (unsigned char) ((sha_info->digest[5] >> 48) & 0xff);
+    digest[42] = (unsigned char) ((sha_info->digest[5] >> 40) & 0xff);
+    digest[43] = (unsigned char) ((sha_info->digest[5] >> 32) & 0xff);
+    digest[44] = (unsigned char) ((sha_info->digest[5] >> 24) & 0xff);
+    digest[45] = (unsigned char) ((sha_info->digest[5] >> 16) & 0xff);
+    digest[46] = (unsigned char) ((sha_info->digest[5] >>  8) & 0xff);
+    digest[47] = (unsigned char) ((sha_info->digest[5]      ) & 0xff);
+    digest[48] = (unsigned char) ((sha_info->digest[6] >> 56) & 0xff);
+    digest[49] = (unsigned char) ((sha_info->digest[6] >> 48) & 0xff);
+    digest[50] = (unsigned char) ((sha_info->digest[6] >> 40) & 0xff);
+    digest[51] = (unsigned char) ((sha_info->digest[6] >> 32) & 0xff);
+    digest[52] = (unsigned char) ((sha_info->digest[6] >> 24) & 0xff);
+    digest[53] = (unsigned char) ((sha_info->digest[6] >> 16) & 0xff);
+    digest[54] = (unsigned char) ((sha_info->digest[6] >>  8) & 0xff);
+    digest[55] = (unsigned char) ((sha_info->digest[6]      ) & 0xff);
+    digest[56] = (unsigned char) ((sha_info->digest[7] >> 56) & 0xff);
+    digest[57] = (unsigned char) ((sha_info->digest[7] >> 48) & 0xff);
+    digest[58] = (unsigned char) ((sha_info->digest[7] >> 40) & 0xff);
+    digest[59] = (unsigned char) ((sha_info->digest[7] >> 32) & 0xff);
+    digest[60] = (unsigned char) ((sha_info->digest[7] >> 24) & 0xff);
+    digest[61] = (unsigned char) ((sha_info->digest[7] >> 16) & 0xff);
+    digest[62] = (unsigned char) ((sha_info->digest[7] >>  8) & 0xff);
+    digest[63] = (unsigned char) ((sha_info->digest[7]      ) & 0xff);
+}
+
+/*
+ * End of copied SHA code.
+ *
+ * ------------------------------------------------------------------------
+ */
+
+static PyTypeObject SHA384type;
+static PyTypeObject SHA512type;
+
+
+static SHAobject *
+newSHA384object(void)
+{
+    return (SHAobject *)PyObject_New(SHAobject, &SHA384type);
+}
+
+static SHAobject *
+newSHA512object(void)
+{
+    return (SHAobject *)PyObject_New(SHAobject, &SHA512type);
+}
+
+/* Internal methods for a hash object */
+
+static void
+SHA512_dealloc(PyObject *ptr)
+{
+    PyObject_Del(ptr);
+}
+
+
+/* External methods for a hash object */
+
+PyDoc_STRVAR(SHA512_copy__doc__, "Return a copy of the hash object.");
+
+static PyObject *
+SHA512_copy(SHAobject *self, PyObject *unused)
+{
+    SHAobject *newobj;
+
+    if (((PyObject*)self)->ob_type == &SHA512type) {
+        if ( (newobj = newSHA512object())==NULL)
+            return NULL;
+    } else {
+        if ( (newobj = newSHA384object())==NULL)
+            return NULL;
+    }
+
+    SHAcopy(self, newobj);
+    return (PyObject *)newobj;
+}
+
+PyDoc_STRVAR(SHA512_digest__doc__,
+"Return the digest value as a string of binary data.");
+
+static PyObject *
+SHA512_digest(SHAobject *self, PyObject *unused)
+{
+    unsigned char digest[SHA_DIGESTSIZE];
+    SHAobject temp;
+
+    SHAcopy(self, &temp);
+    sha512_final(digest, &temp);
+    return PyString_FromStringAndSize((const char *)digest, self->digestsize);
+}
+
+PyDoc_STRVAR(SHA512_hexdigest__doc__,
+"Return the digest value as a string of hexadecimal digits.");
+
+static PyObject *
+SHA512_hexdigest(SHAobject *self, PyObject *unused)
+{
+    unsigned char digest[SHA_DIGESTSIZE];
+    SHAobject temp;
+    PyObject *retval;
+    char *hex_digest;
+    int i, j;
+
+    /* Get the raw (binary) digest value */
+    SHAcopy(self, &temp);
+    sha512_final(digest, &temp);
+
+    /* Create a new string */
+    retval = PyString_FromStringAndSize(NULL, self->digestsize * 2);
+    if (!retval)
+	    return NULL;
+    hex_digest = PyString_AsString(retval);
+    if (!hex_digest) {
+	    Py_DECREF(retval);
+	    return NULL;
+    }
+
+    /* Make hex version of the digest */
+    for (i=j=0; i<self->digestsize; i++) {
+        char c;
+        c = (digest[i] >> 4) & 0xf;
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+        c = (digest[i] & 0xf);
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+    }
+    return retval;
+}
+
+PyDoc_STRVAR(SHA512_update__doc__,
+"Update this hash object's state with the provided string.");
+
+static PyObject *
+SHA512_update(SHAobject *self, PyObject *args)
+{
+    unsigned char *cp;
+    int len;
+
+    if (!PyArg_ParseTuple(args, "s#:update", &cp, &len))
+        return NULL;
+
+    sha512_update(self, cp, len);
+
+    Py_INCREF(Py_None);
+    return Py_None;
+}
+
+static PyMethodDef SHA_methods[] = {
+    {"copy",	  (PyCFunction)SHA512_copy,      METH_NOARGS, SHA512_copy__doc__},
+    {"digest",	  (PyCFunction)SHA512_digest,    METH_NOARGS, SHA512_digest__doc__},
+    {"hexdigest", (PyCFunction)SHA512_hexdigest, METH_NOARGS, SHA512_hexdigest__doc__},
+    {"update",	  (PyCFunction)SHA512_update,    METH_VARARGS, SHA512_update__doc__},
+    {NULL,	  NULL}		/* sentinel */
+};
+
+static PyObject *
+SHA512_get_block_size(PyObject *self, void *closure)
+{
+    return PyInt_FromLong(SHA_BLOCKSIZE);
+}
+
+static PyObject *
+SHA512_get_name(PyObject *self, void *closure)
+{
+    if (((SHAobject *)self)->digestsize == 64)
+        return PyString_FromStringAndSize("SHA512", 6);
+    else
+        return PyString_FromStringAndSize("SHA384", 6);
+}
+
+static PyGetSetDef SHA_getseters[] = {
+    {"block_size",
+     (getter)SHA512_get_block_size, NULL,
+     NULL,
+     NULL},
+    {"name",
+     (getter)SHA512_get_name, NULL,
+     NULL,
+     NULL},
+    {NULL}  /* Sentinel */
+};
+
+static PyMemberDef SHA_members[] = {
+    {"digest_size", T_INT, offsetof(SHAobject, digestsize), READONLY, NULL},
+    /* the old md5 and sha modules support 'digest_size' as in PEP 247.
+     * the old sha module also supported 'digestsize'.  ugh. */
+    {"digestsize", T_INT, offsetof(SHAobject, digestsize), READONLY, NULL},
+    {NULL}  /* Sentinel */
+};
+
+static PyTypeObject SHA384type = {
+    PyObject_HEAD_INIT(NULL)
+    0,			/*ob_size*/
+    "_sha512.sha384",	/*tp_name*/
+    sizeof(SHAobject),	/*tp_size*/
+    0,			/*tp_itemsize*/
+    /* methods */
+    SHA512_dealloc,	/*tp_dealloc*/
+    0,			/*tp_print*/
+    0,          	/*tp_getattr*/
+    0,                  /*tp_setattr*/
+    0,                  /*tp_compare*/
+    0,                  /*tp_repr*/
+    0,                  /*tp_as_number*/
+    0,                  /*tp_as_sequence*/
+    0,                  /*tp_as_mapping*/
+    0,                  /*tp_hash*/
+    0,                  /*tp_call*/
+    0,                  /*tp_str*/
+    0,                  /*tp_getattro*/
+    0,                  /*tp_setattro*/
+    0,                  /*tp_as_buffer*/
+    Py_TPFLAGS_DEFAULT, /*tp_flags*/
+    0,                  /*tp_doc*/
+    0,                  /*tp_traverse*/
+    0,			/*tp_clear*/
+    0,			/*tp_richcompare*/
+    0,			/*tp_weaklistoffset*/
+    0,			/*tp_iter*/
+    0,			/*tp_iternext*/
+    SHA_methods,	/* tp_methods */
+    SHA_members,	/* tp_members */
+    SHA_getseters,      /* tp_getset */
+};
+
+static PyTypeObject SHA512type = {
+    PyObject_HEAD_INIT(NULL)
+    0,			/*ob_size*/
+    "_sha512.sha512",	/*tp_name*/
+    sizeof(SHAobject),	/*tp_size*/
+    0,			/*tp_itemsize*/
+    /* methods */
+    SHA512_dealloc,	/*tp_dealloc*/
+    0,			/*tp_print*/
+    0,          	/*tp_getattr*/
+    0,                  /*tp_setattr*/
+    0,                  /*tp_compare*/
+    0,                  /*tp_repr*/
+    0,                  /*tp_as_number*/
+    0,                  /*tp_as_sequence*/
+    0,                  /*tp_as_mapping*/
+    0,                  /*tp_hash*/
+    0,                  /*tp_call*/
+    0,                  /*tp_str*/
+    0,                  /*tp_getattro*/
+    0,                  /*tp_setattro*/
+    0,                  /*tp_as_buffer*/
+    Py_TPFLAGS_DEFAULT, /*tp_flags*/
+    0,                  /*tp_doc*/
+    0,                  /*tp_traverse*/
+    0,			/*tp_clear*/
+    0,			/*tp_richcompare*/
+    0,			/*tp_weaklistoffset*/
+    0,			/*tp_iter*/
+    0,			/*tp_iternext*/
+    SHA_methods,	/* tp_methods */
+    SHA_members,	/* tp_members */
+    SHA_getseters,      /* tp_getset */
+};
+
+
+/* The single module-level function: new() */
+
+PyDoc_STRVAR(SHA512_new__doc__,
+"Return a new SHA-512 hash object; optionally initialized with a string.");
+
+static PyObject *
+SHA512_new(PyObject *self, PyObject *args, PyObject *kwdict)
+{
+    static char *kwlist[] = {"string", NULL};
+    SHAobject *new;
+    unsigned char *cp = NULL;
+    int len;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwdict, "|s#:new", kwlist,
+                                     &cp, &len)) {
+        return NULL;
+    }
+
+    if ((new = newSHA512object()) == NULL)
+        return NULL;
+
+    sha512_init(new);
+
+    if (PyErr_Occurred()) {
+        Py_DECREF(new);
+        return NULL;
+    }
+    if (cp)
+        sha512_update(new, cp, len);
+
+    return (PyObject *)new;
+}
+
+PyDoc_STRVAR(SHA384_new__doc__,
+"Return a new SHA-384 hash object; optionally initialized with a string.");
+
+static PyObject *
+SHA384_new(PyObject *self, PyObject *args, PyObject *kwdict)
+{
+    static char *kwlist[] = {"string", NULL};
+    SHAobject *new;
+    unsigned char *cp = NULL;
+    int len;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwdict, "|s#:new", kwlist,
+                                     &cp, &len)) {
+        return NULL;
+    }
+
+    if ((new = newSHA384object()) == NULL)
+        return NULL;
+
+    sha384_init(new);
+
+    if (PyErr_Occurred()) {
+        Py_DECREF(new);
+        return NULL;
+    }
+    if (cp)
+        sha512_update(new, cp, len);
+
+    return (PyObject *)new;
+}
+
+
+/* List of functions exported by this module */
+
+static struct PyMethodDef SHA_functions[] = {
+    {"sha512", (PyCFunction)SHA512_new, METH_VARARGS|METH_KEYWORDS, SHA512_new__doc__},
+    {"sha384", (PyCFunction)SHA384_new, METH_VARARGS|METH_KEYWORDS, SHA384_new__doc__},
+    {NULL,	NULL}		 /* Sentinel */
+};
+
+
+/* Initialize this module. */
+
+#define insint(n,v) { PyModule_AddIntConstant(m,n,v); }
+
+PyMODINIT_FUNC
+init_sha512(void)
+{
+    PyObject *m;
+
+    SHA384type.ob_type = &PyType_Type;
+    if (PyType_Ready(&SHA384type) < 0)
+        return;
+    SHA512type.ob_type = &PyType_Type;
+    if (PyType_Ready(&SHA512type) < 0)
+        return;
+    m = Py_InitModule("_sha512", SHA_functions);
+    if (m == NULL)
+	return;
+}
+
+#endif
diff --git a/src/Modules/shamodule.c b/src/Modules/shamodule.c
new file mode 100644
--- /dev/null
+++ b/src/Modules/shamodule.c
@@ -0,0 +1,593 @@
+/* SHA module */
+
+/* This module provides an interface to NIST's Secure Hash Algorithm */
+
+/* See below for information about the original code this module was
+   based upon. Additional work performed by:
+
+   Andrew Kuchling (amk at amk.ca)
+   Greg Stein (gstein at lyra.org)
+
+   Copyright (C) 2005   Gregory P. Smith (greg at krypto.org)
+   Licensed to PSF under a Contributor Agreement.
+
+*/
+
+/* SHA objects */
+
+#include "Python.h"
+#include "structmember.h"
+
+
+/* Endianness testing and definitions */
+#define TestEndianness(variable) {int i=1; variable=PCT_BIG_ENDIAN;\
+	if (*((char*)&i)==1) variable=PCT_LITTLE_ENDIAN;}
+
+#define PCT_LITTLE_ENDIAN 1
+#define PCT_BIG_ENDIAN 0
+
+/* Some useful types */
+
+typedef unsigned char SHA_BYTE;
+
+#if SIZEOF_INT == 4
+typedef unsigned int SHA_INT32;	/* 32-bit integer */
+#else
+/* not defined. compilation will die. */
+#endif
+
+/* The SHA block size and message digest sizes, in bytes */
+
+#define SHA_BLOCKSIZE    64
+#define SHA_DIGESTSIZE  20
+
+/* The structure for storing SHS info */
+
+typedef struct {
+    PyObject_HEAD
+    SHA_INT32 digest[5];		/* Message digest */
+    SHA_INT32 count_lo, count_hi;	/* 64-bit bit count */
+    SHA_BYTE data[SHA_BLOCKSIZE];	/* SHA data buffer */
+    int Endianness;
+    int local;				/* unprocessed amount in data */
+} SHAobject;
+
+/* When run on a little-endian CPU we need to perform byte reversal on an
+   array of longwords. */
+
+static void longReverse(SHA_INT32 *buffer, int byteCount, int Endianness)
+{
+    SHA_INT32 value;
+
+    if ( Endianness == PCT_BIG_ENDIAN )
+	return;
+
+    byteCount /= sizeof(*buffer);
+    while (byteCount--) {
+        value = *buffer;
+        value = ( ( value & 0xFF00FF00L ) >> 8  ) | \
+                ( ( value & 0x00FF00FFL ) << 8 );
+        *buffer++ = ( value << 16 ) | ( value >> 16 );
+    }
+}
+
+static void SHAcopy(SHAobject *src, SHAobject *dest)
+{
+    dest->Endianness = src->Endianness;
+    dest->local = src->local;
+    dest->count_lo = src->count_lo;
+    dest->count_hi = src->count_hi;
+    memcpy(dest->digest, src->digest, sizeof(src->digest));
+    memcpy(dest->data, src->data, sizeof(src->data));
+}
+
+
+/* ------------------------------------------------------------------------
+ *
+ * This code for the SHA algorithm was noted as public domain. The original
+ * headers are pasted below.
+ *
+ * Several changes have been made to make it more compatible with the
+ * Python environment and desired interface.
+ *
+ */
+
+/* NIST Secure Hash Algorithm */
+/* heavily modified by Uwe Hollerbach <uh at alumni.caltech edu> */
+/* from Peter C. Gutmann's implementation as found in */
+/* Applied Cryptography by Bruce Schneier */
+/* Further modifications to include the "UNRAVEL" stuff, below */
+
+/* This code is in the public domain */
+
+/* UNRAVEL should be fastest & biggest */
+/* UNROLL_LOOPS should be just as big, but slightly slower */
+/* both undefined should be smallest and slowest */
+
+#define UNRAVEL
+/* #define UNROLL_LOOPS */
+
+/* The SHA f()-functions.  The f1 and f3 functions can be optimized to
+   save one boolean operation each - thanks to Rich Schroeppel,
+   rcs at cs.arizona.edu for discovering this */
+
+/*#define f1(x,y,z)	((x & y) | (~x & z))		// Rounds  0-19 */
+#define f1(x,y,z)	(z ^ (x & (y ^ z)))		/* Rounds  0-19 */
+#define f2(x,y,z)	(x ^ y ^ z)			/* Rounds 20-39 */
+/*#define f3(x,y,z)	((x & y) | (x & z) | (y & z))	// Rounds 40-59 */
+#define f3(x,y,z)	((x & y) | (z & (x | y)))	/* Rounds 40-59 */
+#define f4(x,y,z)	(x ^ y ^ z)			/* Rounds 60-79 */
+
+/* SHA constants */
+
+#define CONST1		0x5a827999L			/* Rounds  0-19 */
+#define CONST2		0x6ed9eba1L			/* Rounds 20-39 */
+#define CONST3		0x8f1bbcdcL			/* Rounds 40-59 */
+#define CONST4		0xca62c1d6L			/* Rounds 60-79 */
+
+/* 32-bit rotate */
+
+#define R32(x,n)	((x << n) | (x >> (32 - n)))
+
+/* the generic case, for when the overall rotation is not unraveled */
+
+#define FG(n)	\
+    T = R32(A,5) + f##n(B,C,D) + E + *WP++ + CONST##n;	\
+    E = D; D = C; C = R32(B,30); B = A; A = T
+
+/* specific cases, for when the overall rotation is unraveled */
+
+#define FA(n)	\
+    T = R32(A,5) + f##n(B,C,D) + E + *WP++ + CONST##n; B = R32(B,30)
+
+#define FB(n)	\
+    E = R32(T,5) + f##n(A,B,C) + D + *WP++ + CONST##n; A = R32(A,30)
+
+#define FC(n)	\
+    D = R32(E,5) + f##n(T,A,B) + C + *WP++ + CONST##n; T = R32(T,30)
+
+#define FD(n)	\
+    C = R32(D,5) + f##n(E,T,A) + B + *WP++ + CONST##n; E = R32(E,30)
+
+#define FE(n)	\
+    B = R32(C,5) + f##n(D,E,T) + A + *WP++ + CONST##n; D = R32(D,30)
+
+#define FT(n)	\
+    A = R32(B,5) + f##n(C,D,E) + T + *WP++ + CONST##n; C = R32(C,30)
+
+/* do SHA transformation */
+
+static void
+sha_transform(SHAobject *sha_info)
+{
+    int i;
+    SHA_INT32 T, A, B, C, D, E, W[80], *WP;
+
+    memcpy(W, sha_info->data, sizeof(sha_info->data));
+    longReverse(W, (int)sizeof(sha_info->data), sha_info->Endianness);
+
+    for (i = 16; i < 80; ++i) {
+	W[i] = W[i-3] ^ W[i-8] ^ W[i-14] ^ W[i-16];
+
+	/* extra rotation fix */
+	W[i] = R32(W[i], 1);
+    }
+    A = sha_info->digest[0];
+    B = sha_info->digest[1];
+    C = sha_info->digest[2];
+    D = sha_info->digest[3];
+    E = sha_info->digest[4];
+    WP = W;
+#ifdef UNRAVEL
+    FA(1); FB(1); FC(1); FD(1); FE(1); FT(1); FA(1); FB(1); FC(1); FD(1);
+    FE(1); FT(1); FA(1); FB(1); FC(1); FD(1); FE(1); FT(1); FA(1); FB(1);
+    FC(2); FD(2); FE(2); FT(2); FA(2); FB(2); FC(2); FD(2); FE(2); FT(2);
+    FA(2); FB(2); FC(2); FD(2); FE(2); FT(2); FA(2); FB(2); FC(2); FD(2);
+    FE(3); FT(3); FA(3); FB(3); FC(3); FD(3); FE(3); FT(3); FA(3); FB(3);
+    FC(3); FD(3); FE(3); FT(3); FA(3); FB(3); FC(3); FD(3); FE(3); FT(3);
+    FA(4); FB(4); FC(4); FD(4); FE(4); FT(4); FA(4); FB(4); FC(4); FD(4);
+    FE(4); FT(4); FA(4); FB(4); FC(4); FD(4); FE(4); FT(4); FA(4); FB(4);
+    sha_info->digest[0] += E;
+    sha_info->digest[1] += T;
+    sha_info->digest[2] += A;
+    sha_info->digest[3] += B;
+    sha_info->digest[4] += C;
+#else /* !UNRAVEL */
+#ifdef UNROLL_LOOPS
+    FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1);
+    FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1);
+    FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2);
+    FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2);
+    FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3);
+    FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3);
+    FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4);
+    FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4);
+#else /* !UNROLL_LOOPS */
+    for (i =  0; i < 20; ++i) { FG(1); }
+    for (i = 20; i < 40; ++i) { FG(2); }
+    for (i = 40; i < 60; ++i) { FG(3); }
+    for (i = 60; i < 80; ++i) { FG(4); }
+#endif /* !UNROLL_LOOPS */
+    sha_info->digest[0] += A;
+    sha_info->digest[1] += B;
+    sha_info->digest[2] += C;
+    sha_info->digest[3] += D;
+    sha_info->digest[4] += E;
+#endif /* !UNRAVEL */
+}
+
+/* initialize the SHA digest */
+
+static void
+sha_init(SHAobject *sha_info)
+{
+    TestEndianness(sha_info->Endianness)
+
+    sha_info->digest[0] = 0x67452301L;
+    sha_info->digest[1] = 0xefcdab89L;
+    sha_info->digest[2] = 0x98badcfeL;
+    sha_info->digest[3] = 0x10325476L;
+    sha_info->digest[4] = 0xc3d2e1f0L;
+    sha_info->count_lo = 0L;
+    sha_info->count_hi = 0L;
+    sha_info->local = 0;
+}
+
+/* update the SHA digest */
+
+static void
+sha_update(SHAobject *sha_info, SHA_BYTE *buffer, int count)
+{
+    int i;
+    SHA_INT32 clo;
+
+    clo = sha_info->count_lo + ((SHA_INT32) count << 3);
+    if (clo < sha_info->count_lo) {
+        ++sha_info->count_hi;
+    }
+    sha_info->count_lo = clo;
+    sha_info->count_hi += (SHA_INT32) count >> 29;
+    if (sha_info->local) {
+        i = SHA_BLOCKSIZE - sha_info->local;
+        if (i > count) {
+            i = count;
+        }
+        memcpy(((SHA_BYTE *) sha_info->data) + sha_info->local, buffer, i);
+        count -= i;
+        buffer += i;
+        sha_info->local += i;
+        if (sha_info->local == SHA_BLOCKSIZE) {
+            sha_transform(sha_info);
+        }
+        else {
+            return;
+        }
+    }
+    while (count >= SHA_BLOCKSIZE) {
+        memcpy(sha_info->data, buffer, SHA_BLOCKSIZE);
+        buffer += SHA_BLOCKSIZE;
+        count -= SHA_BLOCKSIZE;
+        sha_transform(sha_info);
+    }
+    memcpy(sha_info->data, buffer, count);
+    sha_info->local = count;
+}
+
+/* finish computing the SHA digest */
+
+static void
+sha_final(unsigned char digest[20], SHAobject *sha_info)
+{
+    int count;
+    SHA_INT32 lo_bit_count, hi_bit_count;
+
+    lo_bit_count = sha_info->count_lo;
+    hi_bit_count = sha_info->count_hi;
+    count = (int) ((lo_bit_count >> 3) & 0x3f);
+    ((SHA_BYTE *) sha_info->data)[count++] = 0x80;
+    if (count > SHA_BLOCKSIZE - 8) {
+	memset(((SHA_BYTE *) sha_info->data) + count, 0,
+	       SHA_BLOCKSIZE - count);
+	sha_transform(sha_info);
+	memset((SHA_BYTE *) sha_info->data, 0, SHA_BLOCKSIZE - 8);
+    }
+    else {
+	memset(((SHA_BYTE *) sha_info->data) + count, 0,
+	       SHA_BLOCKSIZE - 8 - count);
+    }
+
+    /* GJS: note that we add the hi/lo in big-endian. sha_transform will
+       swap these values into host-order. */
+    sha_info->data[56] = (hi_bit_count >> 24) & 0xff;
+    sha_info->data[57] = (hi_bit_count >> 16) & 0xff;
+    sha_info->data[58] = (hi_bit_count >>  8) & 0xff;
+    sha_info->data[59] = (hi_bit_count >>  0) & 0xff;
+    sha_info->data[60] = (lo_bit_count >> 24) & 0xff;
+    sha_info->data[61] = (lo_bit_count >> 16) & 0xff;
+    sha_info->data[62] = (lo_bit_count >>  8) & 0xff;
+    sha_info->data[63] = (lo_bit_count >>  0) & 0xff;
+    sha_transform(sha_info);
+    digest[ 0] = (unsigned char) ((sha_info->digest[0] >> 24) & 0xff);
+    digest[ 1] = (unsigned char) ((sha_info->digest[0] >> 16) & 0xff);
+    digest[ 2] = (unsigned char) ((sha_info->digest[0] >>  8) & 0xff);
+    digest[ 3] = (unsigned char) ((sha_info->digest[0]      ) & 0xff);
+    digest[ 4] = (unsigned char) ((sha_info->digest[1] >> 24) & 0xff);
+    digest[ 5] = (unsigned char) ((sha_info->digest[1] >> 16) & 0xff);
+    digest[ 6] = (unsigned char) ((sha_info->digest[1] >>  8) & 0xff);
+    digest[ 7] = (unsigned char) ((sha_info->digest[1]      ) & 0xff);
+    digest[ 8] = (unsigned char) ((sha_info->digest[2] >> 24) & 0xff);
+    digest[ 9] = (unsigned char) ((sha_info->digest[2] >> 16) & 0xff);
+    digest[10] = (unsigned char) ((sha_info->digest[2] >>  8) & 0xff);
+    digest[11] = (unsigned char) ((sha_info->digest[2]      ) & 0xff);
+    digest[12] = (unsigned char) ((sha_info->digest[3] >> 24) & 0xff);
+    digest[13] = (unsigned char) ((sha_info->digest[3] >> 16) & 0xff);
+    digest[14] = (unsigned char) ((sha_info->digest[3] >>  8) & 0xff);
+    digest[15] = (unsigned char) ((sha_info->digest[3]      ) & 0xff);
+    digest[16] = (unsigned char) ((sha_info->digest[4] >> 24) & 0xff);
+    digest[17] = (unsigned char) ((sha_info->digest[4] >> 16) & 0xff);
+    digest[18] = (unsigned char) ((sha_info->digest[4] >>  8) & 0xff);
+    digest[19] = (unsigned char) ((sha_info->digest[4]      ) & 0xff);
+}
+
+/*
+ * End of copied SHA code.
+ *
+ * ------------------------------------------------------------------------
+ */
+
+static PyTypeObject SHAtype;
+
+
+static SHAobject *
+newSHAobject(void)
+{
+    return (SHAobject *)PyObject_New(SHAobject, &SHAtype);
+}
+
+/* Internal methods for a hashing object */
+
+static void
+SHA_dealloc(PyObject *ptr)
+{
+    PyObject_Del(ptr);
+}
+
+
+/* External methods for a hashing object */
+
+PyDoc_STRVAR(SHA_copy__doc__, "Return a copy of the hashing object.");
+
+static PyObject *
+SHA_copy(SHAobject *self, PyObject *unused)
+{
+    SHAobject *newobj;
+
+    if ( (newobj = newSHAobject())==NULL)
+        return NULL;
+
+    SHAcopy(self, newobj);
+    return (PyObject *)newobj;
+}
+
+PyDoc_STRVAR(SHA_digest__doc__,
+"Return the digest value as a string of binary data.");
+
+static PyObject *
+SHA_digest(SHAobject *self, PyObject *unused)
+{
+    unsigned char digest[SHA_DIGESTSIZE];
+    SHAobject temp;
+
+    SHAcopy(self, &temp);
+    sha_final(digest, &temp);
+    return PyString_FromStringAndSize((const char *)digest, sizeof(digest));
+}
+
+PyDoc_STRVAR(SHA_hexdigest__doc__,
+"Return the digest value as a string of hexadecimal digits.");
+
+static PyObject *
+SHA_hexdigest(SHAobject *self, PyObject *unused)
+{
+    unsigned char digest[SHA_DIGESTSIZE];
+    SHAobject temp;
+    PyObject *retval;
+    char *hex_digest;
+    int i, j;
+
+    /* Get the raw (binary) digest value */
+    SHAcopy(self, &temp);
+    sha_final(digest, &temp);
+
+    /* Create a new string */
+    retval = PyString_FromStringAndSize(NULL, sizeof(digest) * 2);
+    if (!retval)
+	    return NULL;
+    hex_digest = PyString_AsString(retval);
+    if (!hex_digest) {
+	    Py_DECREF(retval);
+	    return NULL;
+    }
+
+    /* Make hex version of the digest */
+    for(i=j=0; i<sizeof(digest); i++) {
+        char c;
+        c = (digest[i] >> 4) & 0xf;
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+        c = (digest[i] & 0xf);
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+    }
+    return retval;
+}
+
+PyDoc_STRVAR(SHA_update__doc__,
+"Update this hashing object's state with the provided string.");
+
+static PyObject *
+SHA_update(SHAobject *self, PyObject *args)
+{
+    unsigned char *cp;
+    int len;
+
+    if (!PyArg_ParseTuple(args, "s#:update", &cp, &len))
+        return NULL;
+
+    sha_update(self, cp, len);
+
+    Py_INCREF(Py_None);
+    return Py_None;
+}
+
+static PyMethodDef SHA_methods[] = {
+    {"copy",	  (PyCFunction)SHA_copy,      METH_NOARGS,  SHA_copy__doc__},
+    {"digest",	  (PyCFunction)SHA_digest,    METH_NOARGS,  SHA_digest__doc__},
+    {"hexdigest", (PyCFunction)SHA_hexdigest, METH_NOARGS,  SHA_hexdigest__doc__},
+    {"update",	  (PyCFunction)SHA_update,    METH_VARARGS, SHA_update__doc__},
+    {NULL,	  NULL}		/* sentinel */
+};
+
+static PyObject *
+SHA_get_block_size(PyObject *self, void *closure)
+{
+    return PyInt_FromLong(SHA_BLOCKSIZE);
+}
+
+static PyObject *
+SHA_get_digest_size(PyObject *self, void *closure)
+{
+    return PyInt_FromLong(SHA_DIGESTSIZE);
+}
+
+static PyObject *
+SHA_get_name(PyObject *self, void *closure)
+{
+    return PyString_FromStringAndSize("SHA1", 4);
+}
+
+static PyGetSetDef SHA_getseters[] = {
+    {"digest_size",
+     (getter)SHA_get_digest_size, NULL,
+     NULL,
+     NULL},
+    {"block_size",
+     (getter)SHA_get_block_size, NULL,
+     NULL,
+     NULL},
+    {"name",
+     (getter)SHA_get_name, NULL,
+     NULL,
+     NULL},
+    /* the old md5 and sha modules support 'digest_size' as in PEP 247.
+     * the old sha module also supported 'digestsize'.  ugh. */
+    {"digestsize",
+     (getter)SHA_get_digest_size, NULL,
+     NULL,
+     NULL},
+    {NULL}  /* Sentinel */
+};
+
+static PyTypeObject SHAtype = {
+    PyObject_HEAD_INIT(NULL)
+    0,			/*ob_size*/
+    "_sha.sha",		/*tp_name*/
+    sizeof(SHAobject),	/*tp_size*/
+    0,			/*tp_itemsize*/
+    /* methods */
+    SHA_dealloc,	/*tp_dealloc*/
+    0,			/*tp_print*/
+    0,                  /*tp_getattr*/
+    0,                  /*tp_setattr*/
+    0,                  /*tp_compare*/
+    0,                  /*tp_repr*/
+    0,                  /*tp_as_number*/
+    0,                  /*tp_as_sequence*/
+    0,                  /*tp_as_mapping*/
+    0,                  /*tp_hash*/
+    0,                  /*tp_call*/
+    0,                  /*tp_str*/
+    0,                  /*tp_getattro*/
+    0,                  /*tp_setattro*/
+    0,                  /*tp_as_buffer*/
+    Py_TPFLAGS_DEFAULT, /*tp_flags*/
+    0,                  /*tp_doc*/
+    0,                  /*tp_traverse*/
+    0,			/*tp_clear*/
+    0,			/*tp_richcompare*/
+    0,			/*tp_weaklistoffset*/
+    0,			/*tp_iter*/
+    0,			/*tp_iternext*/
+    SHA_methods,	/* tp_methods */
+    0,                  /* tp_members */
+    SHA_getseters,      /* tp_getset */
+};
+
+
+/* The single module-level function: new() */
+
+PyDoc_STRVAR(SHA_new__doc__,
+"Return a new SHA hashing object.  An optional string argument\n\
+may be provided; if present, this string will be automatically\n\
+hashed.");
+
+static PyObject *
+SHA_new(PyObject *self, PyObject *args, PyObject *kwdict)
+{
+    static char *kwlist[] = {"string", NULL};
+    SHAobject *new;
+    unsigned char *cp = NULL;
+    int len;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwdict, "|s#:new", kwlist,
+                                     &cp, &len)) {
+        return NULL;
+    }
+
+    if ((new = newSHAobject()) == NULL)
+        return NULL;
+
+    sha_init(new);
+
+    if (PyErr_Occurred()) {
+        Py_DECREF(new);
+        return NULL;
+    }
+    if (cp)
+        sha_update(new, cp, len);
+
+    return (PyObject *)new;
+}
+
+
+/* List of functions exported by this module */
+
+static struct PyMethodDef SHA_functions[] = {
+    {"new", (PyCFunction)SHA_new, METH_VARARGS|METH_KEYWORDS, SHA_new__doc__},
+    {NULL,	NULL}		 /* Sentinel */
+};
+
+
+/* Initialize this module. */
+
+#define insint(n,v) { PyModule_AddIntConstant(m,n,v); }
+
+PyMODINIT_FUNC
+init_sha(void)
+{
+    PyObject *m;
+
+    SHAtype.ob_type = &PyType_Type;
+    if (PyType_Ready(&SHAtype) < 0)
+        return;
+    m = Py_InitModule("_sha", SHA_functions);
+    if (m == NULL)
+	return;
+
+    /* Add some symbolic constants to the module */
+    insint("blocksize", 1);  /* For future use, in case some hash
+                                functions require an integral number of
+                                blocks */ 
+    insint("digestsize", 20);
+    insint("digest_size", 20);
+}
diff --git a/src/distutils2/_backport/hashlib.py b/src/distutils2/_backport/hashlib.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/hashlib.py
@@ -0,0 +1,143 @@
+# $Id$
+#
+#  Copyright (C) 2005   Gregory P. Smith (greg at krypto.org)
+#  Licensed to PSF under a Contributor Agreement.
+#
+
+__doc__ = """hashlib module - A common interface to many hash functions.
+
+new(name, string='') - returns a new hash object implementing the
+                       given hash function; initializing the hash
+                       using the given string data.
+
+Named constructor functions are also available, these are much faster
+than using new():
+
+md5(), sha1(), sha224(), sha256(), sha384(), and sha512()
+
+More algorithms may be available on your platform but the above are
+guaranteed to exist.
+
+NOTE: If you want the adler32 or crc32 hash functions they are available in
+the zlib module.
+
+Choose your hash function wisely.  Some have known collision weaknesses.
+sha384 and sha512 will be slow on 32 bit platforms.
+
+Hash objects have these methods:
+ - update(arg): Update the hash object with the string arg. Repeated calls
+                are equivalent to a single call with the concatenation of all
+                the arguments.
+ - digest():    Return the digest of the strings passed to the update() method
+                so far. This may contain non-ASCII characters, including
+                NUL bytes.
+ - hexdigest(): Like digest() except the digest is returned as a string of
+                double length, containing only hexadecimal digits.
+ - copy():      Return a copy (clone) of the hash object. This can be used to
+                efficiently compute the digests of strings that share a common
+                initial substring.
+
+For example, to obtain the digest of the string 'Nobody inspects the
+spammish repetition':
+
+    >>> import hashlib
+    >>> m = hashlib.md5()
+    >>> m.update("Nobody inspects")
+    >>> m.update(" the spammish repetition")
+    >>> m.digest()
+    '\\xbbd\\x9c\\x83\\xdd\\x1e\\xa5\\xc9\\xd9\\xde\\xc9\\xa1\\x8d\\xf0\\xff\\xe9'
+
+More condensed:
+
+    >>> hashlib.sha224("Nobody inspects the spammish repetition").hexdigest()
+    'a4337bc45a8fc544c03f52dc550cd6e1e87021bc896588bd79e901e2'
+
+"""
+
+# This tuple and __get_builtin_constructor() must be modified if a new
+# always available algorithm is added.
+__always_supported = ('md5', 'sha1', 'sha224', 'sha256', 'sha384', 'sha512')
+
+algorithms = __always_supported
+
+__all__ = __always_supported + ('new', 'algorithms')
+
+
+def __get_builtin_constructor(name):
+    if name in ('SHA1', 'sha1'):
+        import _sha
+        return _sha.new
+    elif name in ('MD5', 'md5'):
+        import _md5
+        return _md5.new
+    elif name in ('SHA256', 'sha256', 'SHA224', 'sha224'):
+        import _sha256
+        bs = name[3:]
+        if bs == '256':
+            return _sha256.sha256
+        elif bs == '224':
+            return _sha256.sha224
+    elif name in ('SHA512', 'sha512', 'SHA384', 'sha384'):
+        import _sha512
+        bs = name[3:]
+        if bs == '512':
+            return _sha512.sha512
+        elif bs == '384':
+            return _sha512.sha384
+
+    raise ValueError('unsupported hash type %s' % name)
+
+
+def __get_openssl_constructor(name):
+    try:
+        f = getattr(_hashlib, 'openssl_' + name)
+        # Allow the C module to raise ValueError.  The function will be
+        # defined but the hash not actually available thanks to OpenSSL.
+        f()
+        # Use the C function directly (very fast)
+        return f
+    except (AttributeError, ValueError):
+        return __get_builtin_constructor(name)
+
+
+def __py_new(name, string=''):
+    """new(name, string='') - Return a new hashing object using the named algorithm;
+    optionally initialized with a string.
+    """
+    return __get_builtin_constructor(name)(string)
+
+
+def __hash_new(name, string=''):
+    """new(name, string='') - Return a new hashing object using the named algorithm;
+    optionally initialized with a string.
+    """
+    try:
+        return _hashlib.new(name, string)
+    except ValueError:
+        # If the _hashlib module (OpenSSL) doesn't support the named
+        # hash, try using our builtin implementations.
+        # This allows for SHA224/256 and SHA384/512 support even though
+        # the OpenSSL library prior to 0.9.8 doesn't provide them.
+        return __get_builtin_constructor(name)(string)
+
+
+try:
+    import _hashlib
+    new = __hash_new
+    __get_hash = __get_openssl_constructor
+except ImportError:
+    new = __py_new
+    __get_hash = __get_builtin_constructor
+
+for __func_name in __always_supported:
+    # try them all, some may not work due to the OpenSSL
+    # version not supporting that algorithm.
+    try:
+        globals()[__func_name] = __get_hash(__func_name)
+    except ValueError:
+        import logging
+        logging.exception('code for hash %s was not found.', __func_name)
+
+# Cleanup locals()
+del __always_supported, __func_name, __get_hash
+del __py_new, __hash_new, __get_openssl_constructor
diff --git a/src/distutils2/_backport/tests/test_pkgutil.py b/src/distutils2/_backport/tests/test_pkgutil.py
--- a/src/distutils2/_backport/tests/test_pkgutil.py
+++ b/src/distutils2/_backport/tests/test_pkgutil.py
@@ -1,6 +1,7 @@
 # -*- coding: utf-8 -*-
 """Tests for PEP 376 pkgutil functionality"""
 import unittest2
+import unittest2.compatibility
 import sys
 import os
 import csv
diff --git a/src/distutils2/pypi/dist.py b/src/distutils2/pypi/dist.py
--- a/src/distutils2/pypi/dist.py
+++ b/src/distutils2/pypi/dist.py
@@ -8,12 +8,12 @@
 import urllib
 import tempfile
 try:
-    from hashlib import md5
+    import hashlib
 except ImportError:
-    from md5 import md5
+    from distutils2._backport import hashlib
 
 from distutils2.version import suggest_normalized_version
-from distutils2.pypi.errors import MD5HashDoesNotMatch
+from distutils2.pypi.errors import HashDoesNotMatch, UnsupportedHashName
 
 EXTENSIONS = ".tar.gz .tar.bz2 .tar .zip .tgz .egg".split()
 MD5_HASH = re.compile(r'^.*#md5=([a-f0-9]+)$')
@@ -34,8 +34,8 @@
         """Build a Distribution from a url archive (egg or zip or tgz).
 
         :param url: complete url of the distribution
-        :param probable_dist_name: A probable name of the distribution. 
-        :param is_external: Tell if the url commes from an index or from 
+        :param probable_dist_name: A probable name of the distribution.
+        :param is_external: Tell if the url commes from an index or from
                             an external URL.
         """
         # if the url contains a md5 hash, get it.
@@ -57,20 +57,21 @@
 
         name, version = split_archive_name(archive_name)
         if extension_matched is True:
-            return PyPIDistribution(name, version, url=url, md5_hash=md5_hash, 
-                                    is_external=is_external)
+            return PyPIDistribution(name, version, url=url, hashname="md5",
+                                    hashval=md5_hash, is_external=is_external)
 
-    def __init__(self, name, version, type=None, url=None, md5_hash=None, 
-                 is_external=True):
+    def __init__(self, name, version, type=None, url=None, hashname=None,
+                 hashval=None, is_external=True):
         """Create a new instance of PyPIDistribution.
 
         :param name: the name of the distribution
         :param version: the version of the distribution
         :param type: the type of the dist (eg. source, bin-*, etc.)
         :param url: URL where we found this distribution
-        :param md5_hash: the MD5 hash of the distribution. That's used to 
-                         check that the download is good enough.
-        :param is_external: we need to know if the provided url comes from an 
+        :param hashname: the name of the hash we want to use. Refer to the
+                         hashlib.new documentation for more information.
+        :param hashval: the hash value.
+        :param is_external: we need to know if the provided url comes from an
                             index browsing, or from an external resource.
 
         """
@@ -83,18 +84,25 @@
         # We store urls in dict, because we need to have a bit more informations
         # than the simple URL. It will be used later to find the good url to
         # use.
-        # We have two _url* attributes: _url and _urls. _urls contains a list of 
-        # dict for the different urls, and _url contains the choosen url, in 
+        # We have two _url* attributes: _url and _urls. _urls contains a list of
+        # dict for the different urls, and _url contains the choosen url, in
         # order to dont make the selection process multiple times.
         self._urls = []
         self._url = None
-        self.add_url(url, md5_hash, is_external)
-    
-    def add_url(self, url, md5=None, is_external=True):
+        self.add_url(url, hashname, hashval, is_external)
+
+    def add_url(self, url, hashname=None, hashval=None, is_external=True):
         """Add a new url to the list of urls"""
+        if hashname is not None:
+            try:
+                hashlib.new(hashname)
+            except ValueError:
+                raise UnsupportedHashName(hashname)
+
         self._urls.append({
             'url': url,
-            'md5': md5,
+            'hashname': hashname,
+            'hashval': hashval,
             'is_external': is_external,
         })
         # reset the url selection process
@@ -110,7 +118,7 @@
             if len(self._urls) > 1:
                 internals_urls = [u for u in self._urls \
                                   if u['is_external'] == False]
-                if len(internals_urls) >= 1: 
+                if len(internals_urls) >= 1:
                     self._url = internals_urls[0]
             if self._url is None:
                 self._url = self._urls[0]
@@ -128,7 +136,7 @@
         if self.downloaded_location is None:
             url = self.url['url']
             archive_name = urlparse.urlparse(url)[2].split('/')[-1]
-            filename, headers = urllib.urlretrieve(url, 
+            filename, headers = urllib.urlretrieve(url,
                                                    path + "/" + archive_name)
             self.downloaded_location = filename
             self._check_md5(filename)
@@ -136,15 +144,16 @@
 
     def _check_md5(self, filename):
         """Check that the md5 checksum of the given file matches the one in
-        self._md5_hash."""
-        md5_hash = self.url['md5']
-        if md5_hash is not None:
+        url param"""
+        hashname = self.url['hashname']
+        expected_hashval = self.url['hashval']
+        if not None in (expected_hashval, hashname):
             f = open(filename)
-            hash = md5()
-            hash.update(f.read())
-            if hash.hexdigest() != md5_hash:
-                raise MD5HashDoesNotMatch("%s instead of %s"
-                    % (hash.hexdigest(), md5_hash))
+            hashval = hashlib.new(hashname)
+            hashval.update(f.read())
+            if hashval.hexdigest() != expected_hashval:
+                raise HashDoesNotMatch("got %s instead of %s"
+                    % (hashval.hexdigest(), expected_hashval))
 
     def __repr__(self):
         return "<%s %s (%s)>" \
@@ -211,7 +220,7 @@
 
     def append(self, o):
         """Append a new distribution to the list.
-        
+
         If a distribution with the same name and version exists, just grab the
         URL informations and add a new new url for the existing one.
         """
@@ -222,7 +231,7 @@
             dist.add_url(**o.url)
         else:
             super(PyPIDistributions, self).append(o)
-        
+
 
 def split_archive_name(archive_name, probable_name=None):
     """Split an archive name into two parts: name and version.
diff --git a/src/distutils2/pypi/errors.py b/src/distutils2/pypi/errors.py
--- a/src/distutils2/pypi/errors.py
+++ b/src/distutils2/pypi/errors.py
@@ -21,8 +21,12 @@
     """An error has occurs while downloading"""
 
 
-class MD5HashDoesNotMatch(DownloadError):
-    """Compared MD5 hashes does not match"""
+class HashDoesNotMatch(DownloadError):
+    """Compared hashes does not match"""
+
+
+class UnsupportedHashName(PyPIError):
+    """A unsupported hashname has been used"""
 
 
 class UnableToDownload(PyPIError):
diff --git a/src/distutils2/pypi/simple.py b/src/distutils2/pypi/simple.py
--- a/src/distutils2/pypi/simple.py
+++ b/src/distutils2/pypi/simple.py
@@ -22,7 +22,7 @@
 # -- Constants -----------------------------------------------
 PYPI_DEFAULT_INDEX_URL = "http://pypi.python.org/simple/"
 PYPI_DEFAULT_MIRROR_URL = "mirrors.pypi.python.org"
-DEFAULT_HOSTS = ("*",)  
+DEFAULT_HOSTS = ("*",)
 SOCKET_TIMEOUT = 15
 USER_AGENT = "Python-urllib/%s distutils2/%s" % (
     sys.version[:3], __distutils2_version__)
@@ -63,19 +63,21 @@
     """
 
     def __init__(self, index_url=PYPI_DEFAULT_INDEX_URL, hosts=DEFAULT_HOSTS,
-                 follow_externals=False, mirrors_url=PYPI_DEFAULT_MIRROR_URL, 
+                 follow_externals=False, mirrors_url=PYPI_DEFAULT_MIRROR_URL,
                  mirrors=None, timeout=SOCKET_TIMEOUT):
         """Class constructor.
 
         :param index_url: the url of the simple index to search on.
+        :param follow_externals: tell if following external links is needed or
+                                 not. Default is False.
         :param hosts: a list of hosts allowed to be processed while using
-                      follow_externals=True. Default behavior is to follow all 
+                      follow_externals=True. Default behavior is to follow all
                       hosts.
-        :param follow_externals: tell if following external links is needed or 
+        :param follow_externals: tell if following external links is needed or
                                  not. Default is False.
         :param mirrors_url: the url to look on for DNS records giving mirror
                             adresses.
-        :param mirrors: a list of mirrors to check out if problems 
+        :param mirrors: a list of mirrors to check out if problems
                              occurs while working with the one given in "url"
         :param timeout: time in seconds to consider a url has timeouted.
         """
@@ -116,8 +118,8 @@
         requirements.
 
         :param requirements: A project name and it's distribution, using
-                             version specifiers, as described in PEP345. 
-        :type requirements:  You can pass either a version.VersionPredicate 
+                             version specifiers, as described in PEP345.
+        :type requirements:  You can pass either a version.VersionPredicate
                              or a string.
         """
         requirements = self._get_version_predicate(requirements)
@@ -143,7 +145,7 @@
 
         Returns the complete absolute path to the downloaded archive.
 
-        :param requirements: The same as the find attribute of `find`. 
+        :param requirements: The same as the find attribute of `find`.
         """
         return self.get(requirements).download(path=temp_path)
 
@@ -154,10 +156,10 @@
         if isinstance(requirements, str):
             requirements = VersionPredicate(requirements)
         return requirements
-    
+
     @property
     def index_url(self):
-        return self._index_urls[self._current_index_url] 
+        return self._index_urls[self._current_index_url]
 
     def _switch_to_next_mirror(self):
         """Switch to the next mirror (eg. point self.index_url to the next
@@ -169,7 +171,7 @@
             self._current_index_url = self._current_index_url + 1
         else:
             raise UnableToDownload("All mirrors fails")
-    
+
     def _is_browsable(self, url):
         """Tell if the given URL can be browsed or not.
 
@@ -214,14 +216,14 @@
     def _process_url(self, url, project_name=None, follow_links=True):
         """Process an url and search for distributions packages.
 
-        For each URL found, if it's a download, creates a PyPIdistribution 
+        For each URL found, if it's a download, creates a PyPIdistribution
         object. If it's a homepage and we can follow links, process it too.
 
         :param url: the url to process
         :param project_name: the project name we are searching for.
         :param follow_links: Do not want to follow links more than from one
-                             level. This parameter tells if we want to follow 
-                             the links we find (eg. run recursively this 
+                             level. This parameter tells if we want to follow
+                             the links we find (eg. run recursively this
                              method on it)
         """
         f = self._open_url(url)
@@ -240,7 +242,7 @@
                         if self._is_browsable(link) and follow_links:
                             self._process_url(link, project_name,
                                 follow_links=False)
-    
+
     def _get_link_matcher(self, url):
         """Returns the right link matcher function of the given url
         """
@@ -265,7 +267,7 @@
                                            self._htmldecode(match.group(1)))
                     if 'download' in rels or self._is_browsable(url):
                         # yield a list of (url, is_download)
-                        yield (urlparse.urljoin(base_url, url), 
+                        yield (urlparse.urljoin(base_url, url),
                                'download' in rels)
 
     def _default_link_matcher(self, content, base_url):
@@ -286,7 +288,7 @@
             url = self.index_url + name + "/"
             self._process_url(url, name)
         except DownloadError:
-            # if an error occurs, try with the next index_url 
+            # if an error occurs, try with the next index_url
             # (provided by the mirrors)
             self._switch_to_next_mirror()
             self._distributions.clear()
@@ -305,7 +307,7 @@
                 auth, host = urllib2.splituser(netloc)
             else:
                 auth = None
-            
+
             # add index.html automatically for filesystem paths
             if scheme == 'file':
                 if url.endswith('/'):
diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
--- a/src/distutils2/tests/pypi_server.py
+++ b/src/distutils2/tests/pypi_server.py
@@ -26,8 +26,10 @@
         def wrapped(*args, **kwargs):
             server = PyPIServer(*server_args, **server_kwargs)
             server.start()
-            func(server=server, *args, **kwargs)
-            server.stop()
+            try:
+                func(server=server, *args, **kwargs)
+            finally:
+                server.stop()
         return wrapped
     return wrapper
 
diff --git a/src/distutils2/tests/test_pypi_dist.py b/src/distutils2/tests/test_pypi_dist.py
--- a/src/distutils2/tests/test_pypi_dist.py
+++ b/src/distutils2/tests/test_pypi_dist.py
@@ -7,7 +7,7 @@
 from distutils2.tests.pypi_server import use_pypi_server
 from distutils2.tests.support import unittest
 from distutils2.version import VersionPredicate
-from distutils2.pypi.errors import MD5HashDoesNotMatch
+from distutils2.pypi.errors import HashDoesNotMatch, UnsupportedHashName
 from distutils2.pypi.dist import (PyPIDistribution as Dist,
                                   PyPIDistributions as Dists,
                                   split_archive_name)
@@ -39,7 +39,8 @@
                 'version': '1.1b2',
                 'url': {
                     'url': 'http://test.tld/foobar-1.1b2.tar.gz',  # no hash
-                    'md5': '123123123123123',
+                    'hashval': '123123123123123',
+                    'hashname': 'md5',
                 }
             },
             'foobar-1.1-rc2.tar.gz': {  # use suggested name
@@ -68,7 +69,8 @@
         self.assertDictEqual(d.url, {
             "url": "test_url",
             "is_external": True,
-            "md5": None,
+            "hashname": None,
+            "hashval": None,
         })
 
         # add a new url
@@ -77,7 +79,8 @@
         self.assertDictEqual(d.url, {
             "url": "internal_url",
             "is_external": False,
-            "md5": None,
+            "hashname": None,
+            "hashval": None,
         })
         self.assertEqual(2, len(d._urls))
 
@@ -111,13 +114,13 @@
         url = "%s/simple/foobar/foobar-0.1.tar.gz" % server.full_address
         # check md5 if given
         dist = Dist("FooBar", "0.1", url=url,
-            md5_hash="d41d8cd98f00b204e9800998ecf8427e")
+            hashname="md5", hashval="d41d8cd98f00b204e9800998ecf8427e")
         dist.download()
 
         # a wrong md5 fails
         dist2 = Dist("FooBar", "0.1", url=url,
-            md5_hash="wrongmd5")
-        self.assertRaises(MD5HashDoesNotMatch, dist2.download)
+            hashname="md5", hashval="wrongmd5")
+        self.assertRaises(HashDoesNotMatch, dist2.download)
 
         # we can omit the md5 hash
         dist3 = Dist("FooBar", "0.1", url=url)
@@ -137,6 +140,14 @@
         shutil.rmtree(path1)
         shutil.rmtree(os.path.dirname(path2))
 
+    def test_hashname(self):
+        """Invalid hashnames raises an exception on assignation"""
+        # should be ok
+        Dist("FooBar", "0.1", hashname="md5", hashval="value")
+
+        self.assertRaises(UnsupportedHashName, Dist, "FooBar", "0.1",
+                          hashname="invalid_hashname", hashval="value")
+
 
 class TestPyPIDistributions(unittest.TestCase):
     """test the pypi.distr.PyPIDistributions class"""
diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -137,7 +137,7 @@
         self.assertTrue(index._is_browsable("pypi.test.tld/a/path"))
 
         # specify a list of hosts we want to allow
-        index = simple.SimpleIndex(follow_externals=True, 
+        index = simple.SimpleIndex(follow_externals=True,
                                    hosts=("*.test.tld",))
         self.assertFalse(index._is_browsable("http://an-external.link/path"))
         self.assertTrue(index._is_browsable("http://pypi.test.tld/a/path"))
@@ -197,7 +197,8 @@
         # we have only one link, because links are compared without md5
         self.assertEqual(len(dists), 1)
         # the link should be from the index
-        self.assertEqual('12345678901234567', dists[0].url['md5'])
+        self.assertEqual('12345678901234567', dists[0].url['hashval'])
+        self.assertEqual('md5', dists[0].url['hashname'])
 
     @use_pypi_server(static_filesystem_paths=["with_norel_links"],
         static_uri_paths=["simple", "external"])
@@ -209,13 +210,13 @@
         # process the pages
         index = self._get_simple_index(server, follow_externals=True)
         index.find("foobar")
-        # now it should have processed only pages with links rel="download" 
+        # now it should have processed only pages with links rel="download"
         # and rel="homepage"
-        self.assertIn("%s/simple/foobar/" % server.full_address, 
+        self.assertIn("%s/simple/foobar/" % server.full_address,
             index._processed_urls)  # it's the simple index page
-        self.assertIn("%s/external/homepage.html" % server.full_address, 
+        self.assertIn("%s/external/homepage.html" % server.full_address,
             index._processed_urls)  # the external homepage is rel="homepage"
-        self.assertNotIn("%s/external/nonrel.html" % server.full_address, 
+        self.assertNotIn("%s/external/nonrel.html" % server.full_address,
             index._processed_urls)  # this link contains no rel=*
         self.assertNotIn("%s/unrelated-0.2.tar.gz" % server.full_address,
             index._processed_urls)  # linked from simple index (no rel)
@@ -229,22 +230,22 @@
         server = PyPIServer("foo_bar_baz")
         mirror = PyPIServer("foo_bar_baz")
         mirror.start()  # we dont start the server here
-        
+
         try:
             # create the index using both servers
-            index = simple.SimpleIndex(server.full_address + "/simple/", 
+            index = simple.SimpleIndex(server.full_address + "/simple/",
                 hosts=('*',), timeout=1,  # set the timeout to 1s for the tests
                 mirrors=[mirror.full_address + "/simple/",])
-            
+
             # this should not raise a timeout
             self.assertEqual(4, len(index.find("foo")))
         finally:
             mirror.stop()
-    
+
     def test_simple_link_matcher(self):
         """Test that the simple link matcher yields the right links"""
         index = simple.SimpleIndex(follow_externals=False)
-        
+
         # Here, we define:
         #   1. one link that must be followed, cause it's a download one
         #   2. one link that must *not* be followed, cause the is_browsable
@@ -262,7 +263,7 @@
         # Test that the simple link matcher yield the good links.
         generator = index._simple_link_matcher(content, index.index_url)
         self.assertEqual(('http://dl-link1', True), generator.next())
-        self.assertEqual(('%stest' % index.index_url, False), 
+        self.assertEqual(('%stest' % index.index_url, False),
                          generator.next())
         self.assertRaises(StopIteration, generator.next)
 
@@ -271,7 +272,7 @@
         generator = index._simple_link_matcher(content, index.index_url)
         self.assertEqual(('http://dl-link1', True), generator.next())
         self.assertEqual(('http://dl-link2', False), generator.next())
-        self.assertEqual(('%stest' % index.index_url, False), 
+        self.assertEqual(('%stest' % index.index_url, False),
                          generator.next())
         self.assertRaises(StopIteration, generator.next)
 
diff --git a/src/setup.py b/src/setup.py
--- a/src/setup.py
+++ b/src/setup.py
@@ -3,8 +3,10 @@
 __revision__ = "$Id$"
 import sys
 import os
+import re
 
-from distutils2.core import setup
+from distutils2.core import setup, Extension
+from distutils2.compiler.ccompiler import new_compiler
 from distutils2.command.sdist import sdist
 from distutils2.command.install import install
 from distutils2 import __version__ as VERSION
@@ -31,6 +33,7 @@
 
 DEV_SUFFIX = '.dev%d' % get_tip_revision('..')
 
+
 class install_hg(install):
 
     user_options = install.user_options + [
@@ -62,10 +65,143 @@
             self.distribution.metadata.version += DEV_SUFFIX
         sdist.run(self)
 
+
+# additional paths to check, set from the command line
+SSL_INCDIR=''   # --openssl-incdir=
+SSL_LIBDIR=''   # --openssl-libdir=
+SSL_DIR=''      # --openssl-prefix=
+
+def add_dir_to_list(dirlist, dir):
+    """Add the directory 'dir' to the list 'dirlist' (at the front) if
+    'dir' actually exists and is a directory.  If 'dir' is already in
+    'dirlist' it is moved to the front."""
+    if dir is not None and os.path.isdir(dir) and dir not in dirlist:
+        if dir in dirlist:
+            dirlist.remove(dir)
+        dirlist.insert(0, dir)
+
+
+def prepare_hashlib_extensions():
+    """Decide which C extensions to build and create the appropriate
+    Extension objects to build them.  Returns a list of Extensions."""
+
+    # this CCompiler object is only used to locate include files
+    compiler = new_compiler()
+
+    # Ensure that these paths are always checked
+    if os.name == 'posix':
+        add_dir_to_list(compiler.library_dirs, '/usr/local/lib')
+        add_dir_to_list(compiler.include_dirs, '/usr/local/include')
+
+        add_dir_to_list(compiler.library_dirs, '/usr/local/ssl/lib')
+        add_dir_to_list(compiler.include_dirs, '/usr/local/ssl/include')
+
+        add_dir_to_list(compiler.library_dirs, '/usr/contrib/ssl/lib')
+        add_dir_to_list(compiler.include_dirs, '/usr/contrib/ssl/include')
+
+        add_dir_to_list(compiler.library_dirs, '/usr/lib')
+        add_dir_to_list(compiler.include_dirs, '/usr/include')
+
+    # look in command line supplied paths
+    if SSL_LIBDIR:
+        add_dir_to_list(compiler.library_dirs, SSL_LIBDIR)
+    if SSL_INCDIR:
+        add_dir_to_list(compiler.include_dirs, SSL_INCDIR)
+    if SSL_DIR:
+        if os.name == 'nt':
+            add_dir_to_list(compiler.library_dirs, os.path.join(SSL_DIR, 'out32dll'))
+            # prefer the static library
+            add_dir_to_list(compiler.library_dirs, os.path.join(SSL_DIR, 'out32'))
+        else:
+            add_dir_to_list(compiler.library_dirs, os.path.join(SSL_DIR, 'lib'))
+        add_dir_to_list(compiler.include_dirs, os.path.join(SSL_DIR, 'include'))
+
+    osNameLibsMap = {
+        'posix':  ['ssl', 'crypto'],
+        'nt':     ['libeay32',  'gdi32', 'advapi32', 'user32'],
+    }
+
+    if not osNameLibsMap.has_key(os.name):
+        print "unknown OS, please update setup.py"
+        sys.exit(1)
+
+    exts = []
+
+    ssl_inc_dirs = []
+    ssl_incs = []
+    for inc_dir in compiler.include_dirs:
+        f = os.path.join(inc_dir, 'openssl', 'ssl.h')
+        if os.path.exists(f):
+            ssl_incs.append(f)
+            ssl_inc_dirs.append(inc_dir)
+
+    ssl_lib = compiler.find_library_file(compiler.library_dirs, osNameLibsMap[os.name][0])
+
+    # find out which version of OpenSSL we have
+    openssl_ver = 0
+    openssl_ver_re = re.compile(
+        '^\s*#\s*define\s+OPENSSL_VERSION_NUMBER\s+(0x[0-9a-fA-F]+)' )
+    ssl_inc_dir = ''
+    for ssl_inc_dir in ssl_inc_dirs:
+        name = os.path.join(ssl_inc_dir, 'openssl', 'opensslv.h')
+        if os.path.isfile(name):
+            try:
+                incfile = open(name, 'r')
+                for line in incfile:
+                    m = openssl_ver_re.match(line)
+                    if m:
+                        openssl_ver = eval(m.group(1))
+                        break
+            except IOError:
+                pass
+
+        # first version found is what we'll use
+        if openssl_ver:
+            break
+
+    if (ssl_inc_dir and
+        ssl_lib is not None and
+        openssl_ver >= 0x00907000):
+
+        print 'Using OpenSSL version 0x%08x from' % openssl_ver
+        print ' Headers:\t', ssl_inc_dir
+        print ' Library:\t', ssl_lib
+
+        # The _hashlib module wraps optimized implementations
+        # of hash functions from the OpenSSL library.
+        exts.append( Extension('_hashlib', ['_hashopenssl.c'],
+                               include_dirs = [ ssl_inc_dir ],
+                               library_dirs = [ os.path.dirname(ssl_lib) ],
+                               libraries = osNameLibsMap[os.name]) )
+    else:
+        exts.append( Extension('_sha', ['shamodule.c']) )
+        exts.append( Extension('_md5',
+                        sources = ['md5module.c', 'md5.c'],
+                        depends = ['md5.h']) )
+
+    if (not ssl_lib or openssl_ver < 0x00908000):
+        # OpenSSL doesn't do these until 0.9.8 so we'll bring our own
+        exts.append( Extension('_sha256', ['sha256module.c']) )
+        exts.append( Extension('_sha512', ['sha512module.c']) )
+
+    def prependModules(filename):
+        return os.path.join('Modules', filename)
+
+    # all the C code is in the Modules subdirectory, prepend the path
+    for ext in exts:
+        ext.sources = [ prependModules(fn) for fn in ext.sources ]
+        if hasattr(ext, 'depends') and ext.depends is not None:
+            ext.depends = [ prependModules(fn) for fn in ext.depends ]
+
+    return exts
+
 setup_kwargs = {}
 if sys.version < '2.6':
     setup_kwargs['scripts'] = ['distutils2/mkpkg.py']
 
+if sys.version < '2.5':
+    setup_kwargs['ext_modules'] = prepare_hashlib_extensions()
+
 _CLASSIFIERS = """\
 Development Status :: 3 - Alpha
 Intended Audience :: Developers
@@ -97,6 +233,4 @@
                     ('Repository', 'http://hg.python.org/distutils2'),
                     ('Bug tracker', 'http://bugs.python.org')],
        **setup_kwargs
-       )
-
-
+          )

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:41 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:41 +0200
Subject: [Python-checkins] distutils2: commented out a few tests - need to
	be fixed
Message-ID: <E1OVLoX-0000m4-46@dinsdale.python.org>

tarek.ziade pushed 8b73f08d1d8d to distutils2:

http://hg.python.org/distutils2/rev/8b73f08d1d8d
changeset:   350:8b73f08d1d8d
tag:         tip
user:        Tarek Ziade <tarek at ziade.org>
date:        Sun Jul 04 11:45:19 2010 +0200
summary:     commented out a few tests - need to be fixed
files:       src/distutils2/tests/test_pypi_simple.py

diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -213,6 +213,10 @@
         # and rel="homepage"
         self.assertIn("%s/simple/foobar/" % server.full_address,
             index._processed_urls)  # it's the simple index page
+
+        # FIXME
+        return
+
         self.assertIn("%s/external/homepage.html" % server.full_address,
             index._processed_urls)  # the external homepage is rel="homepage"
         self.assertNotIn("%s/external/nonrel.html" % server.full_address,

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:40 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:40 +0200
Subject: [Python-checkins] distutils2: merge upstream
Message-ID: <E1OVLoW-0000lo-Tz@dinsdale.python.org>

tarek.ziade pushed 6f98873a0fc9 to distutils2:

http://hg.python.org/distutils2/rev/6f98873a0fc9
changeset:   346:6f98873a0fc9
parent:      345:880201efc140
parent:      251:32559ebbec5c
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Sat Jul 03 17:16:42 2010 +0200
summary:     merge upstream
files:       .hgignore, docs/design/pep-0376.txt, docs/design/wiki.rst, docs/source/metadata.rst, src/CONTRIBUTORS.txt, src/distutils2/_backport/tests/test_pkgutil.py, src/distutils2/command/cmd.py, src/distutils2/dist.py, src/distutils2/tests/test_upload.py, src/distutils2/tests/test_version.py, src/setup.py

diff --git a/.hgignore b/.hgignore
--- a/.hgignore
+++ b/.hgignore
@@ -1,5 +1,8 @@
-.*\.pyc$
-.*\.pyo$
-^src/build
-^docs/build
-.*\.swp$
+syntax: glob
+*.py[co]
+__pycache__/
+configure.cache
+build/
+MANIFEST
+dist/
+*.swp
diff --git a/docs/design/pep-0376.txt b/docs/design/pep-0376.txt
--- a/docs/design/pep-0376.txt
+++ b/docs/design/pep-0376.txt
@@ -402,7 +402,7 @@
   ``.egg-info`` directory that contains a PKG-INFO that matches `name` 
   for the `name` metadata.
 
-  Notice that there should be at most one result. The first result founded
+  Notice that there should be at most one result. The first result found
   is returned. If the directory is not found, returns None.
 
 - ``get_file_users(path)`` -> iterator of ``Distribution`` instances.
diff --git a/docs/design/wiki.rst b/docs/design/wiki.rst
--- a/docs/design/wiki.rst
+++ b/docs/design/wiki.rst
@@ -399,7 +399,7 @@
   the folder hierarchy from the module until we find a setup.cfg? A setup.cfg is
   necessary if you use distutils2, is it not?
 
-  -> information founded in setup.cfg will be put in the *FILES* file upon
+  -> information found in setup.cfg will be put in the *FILES* file upon
   installation in the egg-info directory. 
   IOW in the unbuit-egg case, we would need to create that dir, then use 
   pkgutil APIs.
diff --git a/src/CONTRIBUTORS.txt b/src/CONTRIBUTORS.txt
--- a/src/CONTRIBUTORS.txt
+++ b/src/CONTRIBUTORS.txt
@@ -5,7 +5,7 @@
 Distutils2 is a project that was started and that is maintained by
 Tarek Ziad??, and many people are contributing to the project.
 
-If you did, please add your name below in alphabetical order !
+If you did, please add your name below in alphabetical order!
 
 Thanks to:
 
@@ -15,6 +15,8 @@
 - Nicolas Cadou
 - Josip Djolonga
 - Yannick Gringas
+- Jeremy Kloth
+- Martin von L??wis
 - Carl Meyer
 - Alexis Metaireau
 - Michael Mulich
@@ -23,5 +25,3 @@
 - Erik Rose
 - Brian Rosner
 - Alexandre Vassalotti
-- Martin von L??wis
-
diff --git a/src/DEVNOTES.txt b/src/DEVNOTES.txt
--- a/src/DEVNOTES.txt
+++ b/src/DEVNOTES.txt
@@ -3,8 +3,8 @@
 
 - Distutils2 runs from 2.4 to 3.2 (3.x not implemented yet), so
   make sure you don't use a syntax that doesn't work under
-  a specific Python version.
+  one of these Python versions.
 
 - Always run tests.sh before you push a change. This implies
-  that you have all Python versions installed.
+  that you have all Python versions installed from 2.4 to 2.6.
 
diff --git a/src/README.txt b/src/README.txt
--- a/src/README.txt
+++ b/src/README.txt
@@ -2,7 +2,7 @@
 Distutils2
 ==========
 
-Welcome to Distutils2 !
+Welcome to Distutils2!
 
 Distutils2 is the new version of Distutils. It's not backward compatible with
 Distutils but provides more features, and implement most new packaging
@@ -10,6 +10,6 @@
 
 See the documentation at http://packages.python.org/Distutils2 for more info.
 
-**Beware that Distutils2 is its in early stage and should not be used in
+**Beware that Distutils2 is in its early stage and should not be used in
 production. Its API is subject to changes**
 
diff --git a/src/distutils2/README b/src/distutils2/README
--- a/src/distutils2/README
+++ b/src/distutils2/README
@@ -1,4 +1,4 @@
-This directory contains the Distutils package.
+This directory contains the Distutils2 package.
 
 There's a full documentation available at:
 
@@ -8,6 +8,6 @@
 
     http://www.python.org/sigs/distutils-sig/
 
-WARNING: Distutils2 must remain compatible with 2.4
+WARNING: Distutils2 must remain compatible with Python 2.4
 
 $Id: README 70017 2009-02-27 12:53:34Z tarek.ziade $
diff --git a/src/distutils2/__init__.py b/src/distutils2/__init__.py
--- a/src/distutils2/__init__.py
+++ b/src/distutils2/__init__.py
@@ -1,16 +1,20 @@
 """distutils
 
-The main package for the Python Module Distribution Utilities.  Normally
-used from a setup script as
+The main package for the Python Distribution Utilities 2.  Setup
+scripts should import the setup function from distutils2.core:
 
-   from distutils.core import setup
+    from distutils2.core import setup
 
-   setup (...)
+    setup(name=..., version=..., ...)
+
+Third-party tools can use parts of Distutils2 as building blocks
+without causing the other modules to be imported:
+
+    import distutils2.version
+    import distutils2.pypi.simple
+    import distutils2.tests.pypi_server
 """
-__all__ = ['__version__', 'setup']
+__all__ = ['__version__']
 
 __revision__ = "$Id: __init__.py 78020 2010-02-06 16:37:32Z benjamin.peterson $"
 __version__ = "1.0a2"
-
-from distutils2.core import setup
-
diff --git a/src/distutils2/_backport/pkgutil.py b/src/distutils2/_backport/pkgutil.py
--- a/src/distutils2/_backport/pkgutil.py
+++ b/src/distutils2/_backport/pkgutil.py
@@ -1,4 +1,4 @@
-""" Utilities to support packages. """
+"""Utilities to support packages."""
 
 # NOTE: This module must remain compatible with Python 2.3, as it is shared
 # by setuptools for distribution with Python 2.3 and up.
@@ -179,7 +179,7 @@
 iter_importer_modules = simplegeneric(iter_importer_modules)
 
 
-class ImpImporter:
+class ImpImporter(object):
     """:pep:`302` Importer that wraps Python's "classic" import algorithm
 
     ``ImpImporter(dirname)`` produces a :pep:`302` importer that searches that
@@ -242,7 +242,7 @@
                 yield prefix + modname, ispkg
 
 
-class ImpLoader:
+class ImpLoader(object):
     """:pep:`302` Loader that wraps Python's "classic" import algorithm """
 
     code = source = None
@@ -732,6 +732,9 @@
     def __eq__(self, other):
         return isinstance(other, Distribution) and self.path == other.path
 
+    # See http://docs.python.org/reference/datamodel#object.__hash__
+    __hash__ = object.__hash__
+
 
 class EggInfoDistribution(object):
     """Created with the *path* of the ``.egg-info`` directory or file provided
@@ -847,6 +850,9 @@
         return isinstance(other, EggInfoDistribution) and \
                self.path == other.path
 
+    # See http://docs.python.org/reference/datamodel#object.__hash__
+    __hash__ = object.__hash__
+
 
 def _normalize_dist_name(name):
     """Returns a normalized name from the given *name*.
@@ -918,7 +924,7 @@
     returned if one is found that has metadata that matches *name* for the
     *name* metadata field.
 
-    This function only returns the first result founded, as no more than one
+    This function only returns the first result found, as no more than one
     value is expected. If the directory is not found, ``None`` is returned.
 
     :rtype: :class:`Distribution` or :class:`EggInfoDistribution` or None"""
@@ -976,7 +982,7 @@
     then all files and directories ending with ``.egg-info`` are considered
     as well and returns an :class:`EggInfoDistribution` instance.
 
-    This function only returns the first result founded, since no more than
+    This function only returns the first result found, since no more than
     one values are expected. If the directory is not found, returns ``None``.
 
     :parameter version: a version specifier that indicates the version
diff --git a/src/distutils2/_backport/sysconfig.cfg b/src/distutils2/_backport/sysconfig.cfg
--- a/src/distutils2/_backport/sysconfig.cfg
+++ b/src/distutils2/_backport/sysconfig.cfg
@@ -1,6 +1,6 @@
 [globals]
 # These are the useful categories that are sometimes referenced at runtime,
-# using pkgutils.open():
+# using pkgutil.open():
 config             = {confdir}/{distribution.name}         # Configuration files
 appdata            = {datadir}/{distribution.name}         # Non-writable data that is independent of architecture (images, many xml/text files)
 appdata.arch       = {libdir}/{distribution.name}          # Non-writable data that is architecture-dependent (some binary data formats)
diff --git a/src/distutils2/_backport/sysconfig.py b/src/distutils2/_backport/sysconfig.py
--- a/src/distutils2/_backport/sysconfig.py
+++ b/src/distutils2/_backport/sysconfig.py
@@ -5,7 +5,7 @@
 import os
 import re
 from os.path import pardir, abspath
-from ConfigParser import ConfigParser
+from ConfigParser import RawConfigParser
 
 _PREFIX = os.path.normpath(sys.prefix)
 _EXEC_PREFIX = os.path.normpath(sys.exec_prefix)
@@ -14,7 +14,7 @@
 # XXX _CONFIG_DIR will be set by the Makefile later
 _CONFIG_DIR = os.path.normpath(os.path.dirname(__file__))
 _CONFIG_FILE = os.path.join(_CONFIG_DIR, 'sysconfig.cfg')
-_SCHEMES = ConfigParser()
+_SCHEMES = RawConfigParser()
 _SCHEMES.read(_CONFIG_FILE)
 _VAR_REPL = re.compile(r'\{([^{]*?)\}')
 
@@ -580,10 +580,11 @@
                 # behaviour.
                 pass
             else:
-                m = re.search(
-                        r'<key>ProductUserVisibleVersion</key>\s*' +
-                        r'<string>(.*?)</string>', f.read())
-                f.close()
+                try:
+                    m = re.search(r'<key>ProductUserVisibleVersion</key>\s*'
+                                  r'<string>(.*?)</string>', f.read())
+                finally:
+                    f.close()
                 if m is not None:
                     macrelease = '.'.join(m.group(1).split('.')[:2])
                 # else: fall back to the default behaviour
diff --git a/src/distutils2/_backport/tarfile.py b/src/distutils2/_backport/tarfile.py
--- a/src/distutils2/_backport/tarfile.py
+++ b/src/distutils2/_backport/tarfile.py
@@ -370,7 +370,7 @@
 #---------------------------
 # internal stream interface
 #---------------------------
-class _LowLevelFile:
+class _LowLevelFile(object):
     """Low-level file object. Supports reading and writing.
        It is used instead of a regular file object for streaming
        access.
@@ -394,7 +394,7 @@
     def write(self, s):
         os.write(self.fd, s)
 
-class _Stream:
+class _Stream(object):
     """Class that serves as an adapter between TarFile and
        a stream-like object.  The stream-like object only
        needs to have a read() or write() method and is accessed
@@ -2003,8 +2003,10 @@
         # Append the tar header and data to the archive.
         if tarinfo.isreg():
             f = bltn_open(name, "rb")
-            self.addfile(tarinfo, f)
-            f.close()
+            try:
+                self.addfile(tarinfo, f)
+            finally:
+                f.close()
 
         elif tarinfo.isdir():
             self.addfile(tarinfo)
@@ -2214,9 +2216,11 @@
         """
         source = self.extractfile(tarinfo)
         target = bltn_open(targetpath, "wb")
-        copyfileobj(source, target)
-        source.close()
-        target.close()
+        try:
+            copyfileobj(source, target)
+        finally:
+            source.close()
+            target.close()
 
     def makeunknown(self, tarinfo, targetpath):
         """Make a file from a TarInfo object with an unknown type
@@ -2423,7 +2427,7 @@
             print >> sys.stderr, msg
 # class TarFile
 
-class TarIter:
+class TarIter(object):
     """Iterator Class.
 
        for tarinfo in TarFile(...):
@@ -2460,7 +2464,7 @@
         return tarinfo
 
 # Helper classes for sparse file support
-class _section:
+class _section(object):
     """Base class for _data and _hole.
     """
     def __init__(self, offset, size):
@@ -2507,7 +2511,7 @@
 #---------------------------------------------
 TAR_PLAIN = 0           # zipfile.ZIP_STORED
 TAR_GZIPPED = 8         # zipfile.ZIP_DEFLATED
-class TarFileCompat:
+class TarFileCompat(object):
     """TarFile class compatible with standard module zipfile's
        ZipFile class.
     """
@@ -2564,8 +2568,10 @@
        are able to handle, else return False.
     """
     try:
-        t = open(name)
-        t.close()
+        try:
+            t = open(name)
+        finally:
+            t.close()
         return True
     except TarError:
         return False
diff --git a/src/distutils2/_backport/tests/test_sysconfig.py b/src/distutils2/_backport/tests/test_sysconfig.py
--- a/src/distutils2/_backport/tests/test_sysconfig.py
+++ b/src/distutils2/_backport/tests/test_sysconfig.py
@@ -8,7 +8,7 @@
 import os
 import shutil
 from copy import copy, deepcopy
-from ConfigParser import ConfigParser
+from ConfigParser import RawConfigParser
 
 from test.test_support import run_unittest, TESTFN
 
@@ -107,7 +107,7 @@
         wanted.sort()
         scheme = scheme.items()
         scheme.sort()
-        self.assertEquals(scheme, wanted)
+        self.assertEqual(scheme, wanted)
 
     def test_get_config_vars(self):
         cvars = get_config_vars()
@@ -120,21 +120,21 @@
         sys.version = ('2.4.4 (#71, Oct 18 2006, 08:34:43) '
                        '[MSC v.1310 32 bit (Intel)]')
         sys.platform = 'win32'
-        self.assertEquals(get_platform(), 'win32')
+        self.assertEqual(get_platform(), 'win32')
 
         # windows XP, amd64
         os.name = 'nt'
         sys.version = ('2.4.4 (#71, Oct 18 2006, 08:34:43) '
                        '[MSC v.1310 32 bit (Amd64)]')
         sys.platform = 'win32'
-        self.assertEquals(get_platform(), 'win-amd64')
+        self.assertEqual(get_platform(), 'win-amd64')
 
         # windows XP, itanium
         os.name = 'nt'
         sys.version = ('2.4.4 (#71, Oct 18 2006, 08:34:43) '
                        '[MSC v.1310 32 bit (Itanium)]')
         sys.platform = 'win32'
-        self.assertEquals(get_platform(), 'win-ia64')
+        self.assertEqual(get_platform(), 'win-ia64')
 
         # macbook
         os.name = 'posix'
@@ -153,9 +153,9 @@
         maxint = sys.maxint
         try:
             sys.maxint = 2147483647
-            self.assertEquals(get_platform(), 'macosx-10.3-ppc')
+            self.assertEqual(get_platform(), 'macosx-10.3-ppc')
             sys.maxint = 9223372036854775807
-            self.assertEquals(get_platform(), 'macosx-10.3-ppc64')
+            self.assertEqual(get_platform(), 'macosx-10.3-ppc64')
         finally:
             sys.maxint = maxint
 
@@ -173,9 +173,9 @@
         maxint = sys.maxint
         try:
             sys.maxint = 2147483647
-            self.assertEquals(get_platform(), 'macosx-10.3-i386')
+            self.assertEqual(get_platform(), 'macosx-10.3-i386')
             sys.maxint = 9223372036854775807
-            self.assertEquals(get_platform(), 'macosx-10.3-x86_64')
+            self.assertEqual(get_platform(), 'macosx-10.3-x86_64')
         finally:
             sys.maxint = maxint
 
@@ -186,33 +186,33 @@
                                        '-fno-strict-aliasing -fno-common '
                                        '-dynamic -DNDEBUG -g -O3')
 
-        self.assertEquals(get_platform(), 'macosx-10.4-fat')
+        self.assertEqual(get_platform(), 'macosx-10.4-fat')
 
         get_config_vars()['CFLAGS'] = ('-arch x86_64 -arch i386 -isysroot '
                                        '/Developer/SDKs/MacOSX10.4u.sdk  '
                                        '-fno-strict-aliasing -fno-common '
                                        '-dynamic -DNDEBUG -g -O3')
 
-        self.assertEquals(get_platform(), 'macosx-10.4-intel')
+        self.assertEqual(get_platform(), 'macosx-10.4-intel')
 
         get_config_vars()['CFLAGS'] = ('-arch x86_64 -arch ppc -arch i386 -isysroot '
                                        '/Developer/SDKs/MacOSX10.4u.sdk  '
                                        '-fno-strict-aliasing -fno-common '
                                        '-dynamic -DNDEBUG -g -O3')
-        self.assertEquals(get_platform(), 'macosx-10.4-fat3')
+        self.assertEqual(get_platform(), 'macosx-10.4-fat3')
 
         get_config_vars()['CFLAGS'] = ('-arch ppc64 -arch x86_64 -arch ppc -arch i386 -isysroot '
                                        '/Developer/SDKs/MacOSX10.4u.sdk  '
                                        '-fno-strict-aliasing -fno-common '
                                        '-dynamic -DNDEBUG -g -O3')
-        self.assertEquals(get_platform(), 'macosx-10.4-universal')
+        self.assertEqual(get_platform(), 'macosx-10.4-universal')
 
         get_config_vars()['CFLAGS'] = ('-arch x86_64 -arch ppc64 -isysroot '
                                        '/Developer/SDKs/MacOSX10.4u.sdk  '
                                        '-fno-strict-aliasing -fno-common '
                                        '-dynamic -DNDEBUG -g -O3')
 
-        self.assertEquals(get_platform(), 'macosx-10.4-fat64')
+        self.assertEqual(get_platform(), 'macosx-10.4-fat64')
 
         for arch in ('ppc', 'i386', 'x86_64', 'ppc64'):
             get_config_vars()['CFLAGS'] = ('-arch %s -isysroot '
@@ -220,7 +220,7 @@
                                            '-fno-strict-aliasing -fno-common '
                                            '-dynamic -DNDEBUG -g -O3'%(arch,))
 
-            self.assertEquals(get_platform(), 'macosx-10.4-%s'%(arch,))
+            self.assertEqual(get_platform(), 'macosx-10.4-%s'%(arch,))
 
         # linux debian sarge
         os.name = 'posix'
@@ -230,7 +230,7 @@
         self._set_uname(('Linux', 'aglae', '2.6.21.1dedibox-r7',
                     '#1 Mon Apr 30 17:25:38 CEST 2007', 'i686'))
 
-        self.assertEquals(get_platform(), 'linux-i686')
+        self.assertEqual(get_platform(), 'linux-i686')
 
         # XXX more platforms to tests here
 
@@ -241,11 +241,11 @@
     def test_get_scheme_names(self):
         wanted = ('nt', 'nt_user', 'os2', 'os2_home', 'posix_home',
                   'posix_prefix', 'posix_user')
-        self.assertEquals(get_scheme_names(), wanted)
+        self.assertEqual(get_scheme_names(), wanted)
 
     def test_expand_globals(self):
 
-        config = ConfigParser()
+        config = RawConfigParser()
         config.add_section('globals')
         config.set('globals', 'foo', 'ok')
         config.add_section('posix')
@@ -254,8 +254,8 @@
 
         _expand_globals(config)
 
-        self.assertEquals(config.get('posix', 'foo'), 'ok')
-        self.assertEquals(config.get('posix', 'more'), '/etc/ok')
+        self.assertEqual(config.get('posix', 'foo'), 'ok')
+        self.assertEqual(config.get('posix', 'more'), '/etc/ok')
 
         # we might not have globals after all
         # extending again (==no more globals section)
diff --git a/src/distutils2/command/bdist.py b/src/distutils2/command/bdist.py
--- a/src/distutils2/command/bdist.py
+++ b/src/distutils2/command/bdist.py
@@ -62,7 +62,7 @@
                       'os2': 'zip'}
 
     # Establish the preferred order (for the --help-formats option).
-    format_commands = ['rpm', 'gztar', 'bztar', 'ztar', 'tar',
+    format_commands = ['gztar', 'bztar', 'ztar', 'tar',
                        'wininst', 'zip', 'msi']
 
     # And the real information.
@@ -96,7 +96,7 @@
 
         # 'bdist_base' -- parent of per-built-distribution-format
         # temporary directories (eg. we'll probably have
-        # "build/bdist.<plat>/dumb", "build/bdist.<plat>/rpm", etc.)
+        # "build/bdist.<plat>/dumb", etc.)
         if self.bdist_base is None:
             build_base = self.get_finalized_command('build').build_base
             self.bdist_base = os.path.join(build_base,
@@ -126,7 +126,7 @@
         # Reinitialize and run each command.
         for i in range(len(self.formats)):
             cmd_name = commands[i]
-            sub_cmd = self.reinitialize_command(cmd_name)
+            sub_cmd = self.get_reinitialized_command(cmd_name)
 
             # passing the owner and group names for tar archiving
             if cmd_name == 'bdist_dumb':
diff --git a/src/distutils2/command/bdist_dumb.py b/src/distutils2/command/bdist_dumb.py
--- a/src/distutils2/command/bdist_dumb.py
+++ b/src/distutils2/command/bdist_dumb.py
@@ -85,7 +85,7 @@
         if not self.skip_build:
             self.run_command('build')
 
-        install = self.reinitialize_command('install', reinit_subcommands=1)
+        install = self.get_reinitialized_command('install', reinit_subcommands=1)
         install.root = self.bdist_dir
         install.skip_build = self.skip_build
         install.warn_dir = 0
diff --git a/src/distutils2/command/bdist_msi.py b/src/distutils2/command/bdist_msi.py
--- a/src/distutils2/command/bdist_msi.py
+++ b/src/distutils2/command/bdist_msi.py
@@ -177,12 +177,12 @@
         if not self.skip_build:
             self.run_command('build')
 
-        install = self.reinitialize_command('install', reinit_subcommands=1)
+        install = self.get_reinitialized_command('install', reinit_subcommands=1)
         install.prefix = self.bdist_dir
         install.skip_build = self.skip_build
         install.warn_dir = 0
 
-        install_lib = self.reinitialize_command('install_lib')
+        install_lib = self.get_reinitialized_command('install_lib')
         # we do not want to include pyc or pyo files
         install_lib.compile = 0
         install_lib.optimize = 0
diff --git a/src/distutils2/command/bdist_wininst.py b/src/distutils2/command/bdist_wininst.py
--- a/src/distutils2/command/bdist_wininst.py
+++ b/src/distutils2/command/bdist_wininst.py
@@ -126,13 +126,13 @@
         if not self.skip_build:
             self.run_command('build')
 
-        install = self.reinitialize_command('install', reinit_subcommands=1)
+        install = self.get_reinitialized_command('install', reinit_subcommands=1)
         install.root = self.bdist_dir
         install.skip_build = self.skip_build
         install.warn_dir = 0
         install.plat_name = self.plat_name
 
-        install_lib = self.reinitialize_command('install_lib')
+        install_lib = self.get_reinitialized_command('install_lib')
         # we do not want to include pyc or pyo files
         install_lib.compile = 0
         install_lib.optimize = 0
diff --git a/src/distutils2/command/cmd.py b/src/distutils2/command/cmd.py
--- a/src/distutils2/command/cmd.py
+++ b/src/distutils2/command/cmd.py
@@ -19,7 +19,7 @@
 except ImportError:
     from distutils2._backport.shutil import make_archive
 
-class Command:
+class Command(object):
     """Abstract base class for defining command classes, the "worker bees"
     of the Distutils.  A useful analogy for command classes is to think of
     them as subroutines with local variables called "options".  The options
@@ -188,6 +188,31 @@
         """
         log.log(level, msg)
 
+    # -- External interface --------------------------------------------
+    # (called by outsiders)
+
+    def get_source_files(self):
+        """Return the list of files that are used as inputs to this command,
+        i.e. the files used to generate the output files.  The result is used
+        by the `sdist` command in determining the set of default files.
+
+        Command classes should implement this method if they operate on files
+        from the source tree.
+        """
+        return []
+
+    def get_outputs(self):
+        """Return the list of files that would be produced if this command
+        were actually run.  Not affected by the "dry-run" flag or whether
+        any other commands have been run.
+
+        Command classes should implement this method if they produce any
+        output files that get consumed by another command.  e.g., `build_ext`
+        returns the list of built extension modules, but not any temporary
+        files used in the compilation process.
+        """
+        return []
+
     # -- Option validation methods -------------------------------------
     # (these are very handy in writing the 'finalize_options()' method)
     #
@@ -307,10 +332,8 @@
         cmd_obj.ensure_finalized()
         return cmd_obj
 
-    # XXX rename to 'get_reinitialized_command()'? (should do the
-    # same in dist.py, if so)
-    def reinitialize_command(self, command, reinit_subcommands=0):
-        return self.distribution.reinitialize_command(
+    def get_reinitialized_command(self, command, reinit_subcommands=0):
+        return self.distribution.get_reinitialized_command(
             command, reinit_subcommands)
 
     def run_command(self, command):
diff --git a/src/distutils2/command/sdist.py b/src/distutils2/command/sdist.py
--- a/src/distutils2/command/sdist.py
+++ b/src/distutils2/command/sdist.py
@@ -20,7 +20,7 @@
 from distutils2.core import Command
 from distutils2 import util
 from distutils2.errors import (DistutilsPlatformError, DistutilsOptionError,
-                              DistutilsTemplateError)
+                               DistutilsTemplateError)
 from distutils2.manifest import Manifest
 from distutils2 import log
 from distutils2.util import convert_path, newer
diff --git a/src/distutils2/compiler/bcppcompiler.py b/src/distutils2/compiler/bcppcompiler.py
--- a/src/distutils2/compiler/bcppcompiler.py
+++ b/src/distutils2/compiler/bcppcompiler.py
@@ -16,7 +16,7 @@
 import os
 
 from distutils2.errors import (DistutilsExecError, CompileError, LibError,
-                              LinkError, UnknownFileError)
+                               LinkError, UnknownFileError)
 from distutils2.compiler.ccompiler import CCompiler, gen_preprocess_options
 from distutils2.file_util import write_file
 from distutils2.dep_util import newer
diff --git a/src/distutils2/compiler/ccompiler.py b/src/distutils2/compiler/ccompiler.py
--- a/src/distutils2/compiler/ccompiler.py
+++ b/src/distutils2/compiler/ccompiler.py
@@ -10,7 +10,7 @@
 import re
 
 from distutils2.errors import (CompileError, LinkError, UnknownFileError,
-                              DistutilsPlatformError, DistutilsModuleError)
+                               DistutilsPlatformError, DistutilsModuleError)
 from distutils2.spawn import spawn
 from distutils2.util import split_quoted, execute, newer_group
 from distutils2 import log
@@ -76,7 +76,7 @@
 
         compiler.shared_lib_extension = so_ext
 
-class CCompiler:
+class CCompiler(object):
     """Abstract base class to define the interface that must be implemented
     by real compiler classes.  Also has some utility methods used by
     several compiler classes.
@@ -800,14 +800,16 @@
             library_dirs = []
         fd, fname = tempfile.mkstemp(".c", funcname, text=True)
         f = os.fdopen(fd, "w")
-        for incl in includes:
-            f.write("""#include "%s"\n""" % incl)
-        f.write("""\
+        try:
+            for incl in includes:
+                f.write("""#include "%s"\n""" % incl)
+            f.write("""\
 main (int argc, char **argv) {
     %s();
 }
 """ % funcname)
-        f.close()
+        finally:
+            f.close()
         try:
             objects = self.compile([fname], include_dirs=include_dirs)
         except CompileError:
@@ -1048,7 +1050,7 @@
         module_name = "distutils2.compiler." + module_name
         __import__ (module_name)
         module = sys.modules[module_name]
-        klass = vars(module)[class_name]
+        cls = vars(module)[class_name]
     except ImportError:
         raise DistutilsModuleError, \
               "can't compile C/C++ code: unable to load module '%s'" % \
@@ -1061,7 +1063,7 @@
     # XXX The None is necessary to preserve backwards compatibility
     # with classes that expect verbose to be the first positional
     # argument.
-    return klass(None, dry_run, force)
+    return cls(None, dry_run, force)
 
 
 def gen_preprocess_options(macros, include_dirs):
diff --git a/src/distutils2/compiler/emxccompiler.py b/src/distutils2/compiler/emxccompiler.py
--- a/src/distutils2/compiler/emxccompiler.py
+++ b/src/distutils2/compiler/emxccompiler.py
@@ -25,9 +25,8 @@
 from warnings import warn
 
 from distutils2.compiler.unixccompiler import UnixCCompiler
-from distutils2.util import write_file
 from distutils2.errors import DistutilsExecError, CompileError, UnknownFileError
-from distutils2.util import get_compiler_versions
+from distutils2.util import get_compiler_versions, write_file
 
 class EMXCCompiler (UnixCCompiler):
 
@@ -273,8 +272,10 @@
         # It would probably better to read single lines to search.
         # But we do this only once, and it is fast enough
         f = open(fn)
-        s = f.read()
-        f.close()
+        try:
+            s = f.read()
+        finally:
+            f.close()
 
     except IOError, exc:
         # if we can't read this file, we cannot say it is wrong
diff --git a/src/distutils2/compiler/msvc9compiler.py b/src/distutils2/compiler/msvc9compiler.py
--- a/src/distutils2/compiler/msvc9compiler.py
+++ b/src/distutils2/compiler/msvc9compiler.py
@@ -20,7 +20,7 @@
 import re
 
 from distutils2.errors import (DistutilsExecError, DistutilsPlatformError,
-                              CompileError, LibError, LinkError)
+                               CompileError, LibError, LinkError)
 from distutils2.compiler.ccompiler import CCompiler, gen_lib_options
 from distutils2 import log
 from distutils2.util import get_platform
@@ -50,7 +50,7 @@
     'win-ia64' : 'ia64',
 }
 
-class Reg:
+class Reg(object):
     """Helper class to read values from the registry
     """
 
@@ -112,7 +112,7 @@
         return s
     convert_mbcs = staticmethod(convert_mbcs)
 
-class MacroExpander:
+class MacroExpander(object):
 
     def __init__(self, version):
         self.macros = {}
diff --git a/src/distutils2/compiler/msvccompiler.py b/src/distutils2/compiler/msvccompiler.py
--- a/src/distutils2/compiler/msvccompiler.py
+++ b/src/distutils2/compiler/msvccompiler.py
@@ -15,7 +15,7 @@
 import string
 
 from distutils2.errors import (DistutilsExecError, DistutilsPlatformError,
-                              CompileError, LibError, LinkError)
+                               CompileError, LibError, LinkError)
 from distutils2.compiler.ccompiler import CCompiler, gen_lib_options
 from distutils2 import log
 
@@ -104,7 +104,7 @@
             pass
     return s
 
-class MacroExpander:
+class MacroExpander(object):
 
     def __init__(self, version):
         self.macros = {}
diff --git a/src/distutils2/compiler/unixccompiler.py b/src/distutils2/compiler/unixccompiler.py
--- a/src/distutils2/compiler/unixccompiler.py
+++ b/src/distutils2/compiler/unixccompiler.py
@@ -19,10 +19,10 @@
 from types import StringType, NoneType
 
 from distutils2.util import newer
-from distutils2.compiler.ccompiler import \
-     CCompiler, gen_preprocess_options, gen_lib_options
-from distutils2.errors import \
-     DistutilsExecError, CompileError, LibError, LinkError
+from distutils2.compiler.ccompiler import (CCompiler, gen_preprocess_options,
+                                           gen_lib_options)
+from distutils2.errors import (DistutilsExecError, CompileError,
+                               LibError, LinkError)
 from distutils2 import log
 
 try:
diff --git a/src/distutils2/config.py b/src/distutils2/config.py
--- a/src/distutils2/config.py
+++ b/src/distutils2/config.py
@@ -4,7 +4,7 @@
 that uses .pypirc in the distutils.command package.
 """
 import os
-from ConfigParser import ConfigParser
+from ConfigParser import RawConfigParser
 
 from distutils2.command.cmd import Command
 
@@ -59,7 +59,7 @@
         if os.path.exists(rc):
             self.announce('Using PyPI login from %s' % rc)
             repository = self.repository or self.DEFAULT_REPOSITORY
-            config = ConfigParser()
+            config = RawConfigParser()
             config.read(rc)
             sections = config.sections()
             if 'distutils' in sections:
diff --git a/src/distutils2/core.py b/src/distutils2/core.py
--- a/src/distutils2/core.py
+++ b/src/distutils2/core.py
@@ -2,8 +2,9 @@
 
 The only module that needs to be imported to use the Distutils; provides
 the 'setup' function (which is to be called from the setup script).  Also
-indirectly provides the Distribution and Command classes, although they are
-really defined in distutils2.dist and distutils2.cmd.
+exports useful classes so that setup scripts can import them from here
+although they are really defined in other modules: Distribution, Command,
+PyPIRCommand, Extension, find_packages.
 """
 
 __revision__ = "$Id: core.py 77704 2010-01-23 09:23:15Z tarek.ziade $"
@@ -12,7 +13,7 @@
 import os
 
 from distutils2.errors import (DistutilsSetupError, DistutilsArgError,
-                              DistutilsError, CCompilerError)
+                               DistutilsError, CCompilerError)
 from distutils2.util import grok_environment_error
 
 # Mainly import these so setup scripts can "from distutils2.core import" them.
@@ -20,6 +21,7 @@
 from distutils2.command.cmd import Command
 from distutils2.config import PyPIRCCommand
 from distutils2.extension import Extension
+from distutils2.util import find_packages
 
 # This is a barebones help message generated displayed when the user
 # runs the setup script with no arguments at all.  More useful help
@@ -94,11 +96,7 @@
 
     # Determine the distribution class -- either caller-supplied or
     # our Distribution (see below).
-    klass = attrs.get('distclass')
-    if klass:
-        del attrs['distclass']
-    else:
-        klass = Distribution
+    distclass = attrs.pop('distclass', Distribution)
 
     if 'script_name' not in attrs:
         attrs['script_name'] = os.path.basename(sys.argv[0])
@@ -108,7 +106,7 @@
     # Create the Distribution instance, using the remaining arguments
     # (ie. everything except distclass) to initialize it
     try:
-        _setup_distribution = dist = klass(attrs)
+        _setup_distribution = dist = distclass(attrs)
     except DistutilsSetupError, msg:
         if 'name' in attrs:
             raise SystemExit, "error in %s setup command: %s" % \
diff --git a/src/distutils2/depgraph.py b/src/distutils2/depgraph.py
--- a/src/distutils2/depgraph.py
+++ b/src/distutils2/depgraph.py
@@ -13,7 +13,7 @@
     """
     Represents a dependency graph between distributions.
 
-    The depedency relationships are stored in an ``adjacency_list`` that maps
+    The dependency relationships are stored in an ``adjacency_list`` that maps
     distributions to a list of ``(other, label)`` tuples where  ``other``
     is a distribution and the edge is labelled with ``label`` (i.e. the version
     specifier, if such was provided). Also, for more efficient traversal, for
diff --git a/src/distutils2/dist.py b/src/distutils2/dist.py
--- a/src/distutils2/dist.py
+++ b/src/distutils2/dist.py
@@ -13,8 +13,10 @@
 except ImportError:
     warnings = None
 
+from ConfigParser import RawConfigParser
+
 from distutils2.errors import (DistutilsOptionError, DistutilsArgError,
-                              DistutilsModuleError, DistutilsClassError)
+                               DistutilsModuleError, DistutilsClassError)
 from distutils2.fancy_getopt import FancyGetopt, translate_longopt
 from distutils2.util import check_environ, strtobool
 from distutils2 import log
@@ -248,7 +250,7 @@
                 elif hasattr(self, key):
                     setattr(self, key, val)
                 else:
-                    msg = "Unknown distribution option: %s" % repr(key)
+                    msg = "Unknown distribution option: %r" % key
                     if warnings is not None:
                         warnings.warn(msg)
                     else:
@@ -362,14 +364,12 @@
         return files
 
     def parse_config_files(self, filenames=None):
-        from ConfigParser import ConfigParser
-
         if filenames is None:
             filenames = self.find_config_files()
 
         log.debug("Distribution.parse_config_files():")
 
-        parser = ConfigParser()
+        parser = RawConfigParser()
         for filename in filenames:
             log.debug("  reading %s" % filename)
             parser.read(filename)
@@ -383,7 +383,7 @@
                         opt = opt.replace('-', '_')
                         opt_dict[opt] = (filename, val)
 
-            # Make the ConfigParser forget everything (so we retain
+            # Make the RawConfigParser forget everything (so we retain
             # the original filenames that options come from)
             parser.__init__()
 
@@ -627,16 +627,16 @@
 
         for command in self.commands:
             if isinstance(command, type) and issubclass(command, Command):
-                klass = command
+                cls = command
             else:
-                klass = self.get_command_class(command)
-            if (hasattr(klass, 'help_options') and
-                isinstance(klass.help_options, list)):
-                parser.set_option_table(klass.user_options +
-                                        fix_help_options(klass.help_options))
+                cls = self.get_command_class(command)
+            if (hasattr(cls, 'help_options') and
+                isinstance(cls.help_options, list)):
+                parser.set_option_table(cls.user_options +
+                                        fix_help_options(cls.help_options))
             else:
-                parser.set_option_table(klass.user_options)
-            parser.print_help("Options for '%s' command:" % klass.__name__)
+                parser.set_option_table(cls.user_options)
+            parser.print_help("Options for '%s' command:" % cls.__name__)
             print('')
 
         print(gen_usage(self.script_name))
@@ -688,11 +688,11 @@
         print(header + ":")
 
         for cmd in commands:
-            klass = self.cmdclass.get(cmd)
-            if not klass:
-                klass = self.get_command_class(cmd)
+            cls = self.cmdclass.get(cmd)
+            if not cls:
+                cls = self.get_command_class(cmd)
             try:
-                description = klass.description
+                description = cls.description
             except AttributeError:
                 description = "(no description available)"
 
@@ -754,11 +754,11 @@
 
         rv = []
         for cmd in (std_commands + extra_commands):
-            klass = self.cmdclass.get(cmd)
-            if not klass:
-                klass = self.get_command_class(cmd)
+            cls = self.cmdclass.get(cmd)
+            if not cls:
+                cls = self.get_command_class(cmd)
             try:
-                description = klass.description
+                description = cls.description
             except AttributeError:
                 description = "(no description available)"
             rv.append((cmd, description))
@@ -790,13 +790,13 @@
         Raises DistutilsModuleError if the expected module could not be
         found, or if that module does not define the expected class.
         """
-        klass = self.cmdclass.get(command)
-        if klass:
-            return klass
+        cls = self.cmdclass.get(command)
+        if cls:
+            return cls
 
         for pkgname in self.get_command_packages():
             module_name = "%s.%s" % (pkgname, command)
-            klass_name = command
+            class_name = command
 
             try:
                 __import__ (module_name)
@@ -805,14 +805,14 @@
                 continue
 
             try:
-                klass = getattr(module, klass_name)
+                cls = getattr(module, class_name)
             except AttributeError:
                 raise DistutilsModuleError, \
                       "invalid command '%s' (no class '%s' in module '%s')" \
-                      % (command, klass_name, module_name)
+                      % (command, class_name, module_name)
 
-            self.cmdclass[command] = klass
-            return klass
+            self.cmdclass[command] = cls
+            return cls
 
         raise DistutilsModuleError("invalid command '%s'" % command)
 
@@ -828,8 +828,8 @@
             log.debug("Distribution.get_command_obj(): " \
                       "creating '%s' command object" % command)
 
-            klass = self.get_command_class(command)
-            cmd_obj = self.command_obj[command] = klass(self)
+            cls = self.get_command_class(command)
+            cmd_obj = self.command_obj[command] = cls(self)
             self.have_run[command] = 0
 
             # Set any options that were supplied in config files
@@ -885,7 +885,7 @@
             except ValueError, msg:
                 raise DistutilsOptionError, msg
 
-    def reinitialize_command(self, command, reinit_subcommands=0):
+    def get_reinitialized_command(self, command, reinit_subcommands=0):
         """Reinitializes a command to the state it was in when first
         returned by 'get_command_obj()': ie., initialized but not yet
         finalized.  This provides the opportunity to sneak option
@@ -920,7 +920,7 @@
 
         if reinit_subcommands:
             for sub in command.get_sub_commands():
-                self.reinitialize_command(sub, reinit_subcommands)
+                self.get_reinitialized_command(sub, reinit_subcommands)
 
         return command
 
diff --git a/src/distutils2/extension.py b/src/distutils2/extension.py
--- a/src/distutils2/extension.py
+++ b/src/distutils2/extension.py
@@ -23,7 +23,7 @@
 # import that large-ish module (indirectly, through distutils.core) in
 # order to do anything.
 
-class Extension:
+class Extension(object):
     """Just a collection of attributes that describes an extension
     module and everything needed to build it (hopefully in a portable
     way, but there are hooks that let you be as unportable as you need).
diff --git a/src/distutils2/fancy_getopt.py b/src/distutils2/fancy_getopt.py
--- a/src/distutils2/fancy_getopt.py
+++ b/src/distutils2/fancy_getopt.py
@@ -30,7 +30,7 @@
 # (for use as attributes of some object).
 longopt_xlate = string.maketrans('-', '_')
 
-class FancyGetopt:
+class FancyGetopt(object):
     """Wrapper around the standard 'getopt()' module that provides some
     handy extra functionality:
       * short and long options are tied together
@@ -473,7 +473,7 @@
     return string.translate(opt, longopt_xlate)
 
 
-class OptionDummy:
+class OptionDummy(object):
     """Dummy class just used as a place to hold command-line option
     values as instance attributes."""
 
diff --git a/src/distutils2/log.py b/src/distutils2/log.py
--- a/src/distutils2/log.py
+++ b/src/distutils2/log.py
@@ -11,14 +11,14 @@
 
 import sys
 
-class Log:
+class Log(object):
 
     def __init__(self, threshold=WARN):
         self.threshold = threshold
 
     def _log(self, level, msg, args):
         if level not in (DEBUG, INFO, WARN, ERROR, FATAL):
-            raise ValueError('%s wrong log level' % str(level))
+            raise ValueError('%s wrong log level' % level)
 
         if level >= self.threshold:
             if args:
diff --git a/src/distutils2/metadata.py b/src/distutils2/metadata.py
--- a/src/distutils2/metadata.py
+++ b/src/distutils2/metadata.py
@@ -105,7 +105,6 @@
     keys = fields.keys()
     possible_versions = ['1.0', '1.1', '1.2']
 
-
     # first let's try to see if a field is not part of one of the version
     for key in keys:
         if key not in _241_FIELDS and '1.0' in possible_versions:
@@ -128,9 +127,9 @@
         raise MetadataConflictError('You used incompatible 1.1 and 1.2 fields')
 
     # we have the choice, either 1.0, or 1.2
-    #   - 1.0 has a broken Summary field but work with all tools
+    #   - 1.0 has a broken Summary field but works with all tools
     #   - 1.1 is to avoid
-    #   - 1.2 fixes Summary but is not spreaded yet
+    #   - 1.2 fixes Summary but is not widespread yet
     if not is_1_1 and not is_1_2:
         # we couldn't find any specific marker
         if PKG_INFO_PREFERRED_VERSION in possible_versions:
@@ -185,12 +184,12 @@
 class DistributionMetadata(object):
     """Distribution meta-data class (1.0 or 1.2).
     """
-    def __init__(self, path=None, platform_dependant=False,
+    def __init__(self, path=None, platform_dependent=False,
                  execution_context=None, fileobj=None):
         self._fields = {}
         self.version = None
         self.docutils_support = _HAS_DOCUTILS
-        self.platform_dependant = platform_dependant
+        self.platform_dependent = platform_dependent
         if path is not None:
             self.read(path)
         elif fileobj is not None:
@@ -263,7 +262,7 @@
         return reporter.messages
 
     def _platform(self, value):
-        if not self.platform_dependant or ';' not in value:
+        if not self.platform_dependent or ';' not in value:
             return True, value
         value, marker = value.split(';')
         return _interpret(marker, self.execution_context), value
@@ -518,7 +517,7 @@
             raise NameError(value)
 
     def _nonsense_op(self):
-        msg = 'This operation is not supported : "%s"' % str(self)
+        msg = 'This operation is not supported : "%s"' % self
         raise SyntaxError(msg)
 
     def __call__(self):
@@ -635,4 +634,3 @@
     operations = _CHAIN(execution_context)
     tokenize(StringIO(marker).readline, operations.eat)
     return operations.result()
-
diff --git a/src/distutils2/mkpkg.py b/src/distutils2/mkpkg.py
--- a/src/distutils2/mkpkg.py
+++ b/src/distutils2/mkpkg.py
@@ -675,7 +675,7 @@
 troveDict = buildTroveDict(troveList)
 
 
-class SetupClass:
+class SetupClass(object):
     def __init__(self):
         self.config = None
         self.classifierDict = {}
@@ -717,14 +717,16 @@
 
     def inspectFile(self, path):
         fp = open(path, 'r')
-        for line in [ fp.readline() for x in range(10) ]:
-            m = re.match(r'^#!.*python((?P<major>\d)(\.\d+)?)?$', line)
-            if m:
-                if m.group('major') == '3':
-                    self.classifierDict['Programming Language :: Python :: 3'] = 1
-                else:
-                    self.classifierDict['Programming Language :: Python :: 2'] = 1
-        fp.close()
+        try:
+            for line in [ fp.readline() for x in range(10) ]:
+                m = re.match(r'^#!.*python((?P<major>\d)(\.\d+)?)?$', line)
+                if m:
+                    if m.group('major') == '3':
+                        self.classifierDict['Programming Language :: Python :: 3'] = 1
+                    else:
+                        self.classifierDict['Programming Language :: Python :: 2'] = 1
+        finally:
+            fp.close()
 
 
     def inspectDirectory(self):
@@ -885,38 +887,33 @@
         if os.path.exists('setup.py'): shutil.move('setup.py', 'setup.py.old')
 
         fp = open('setup.py', 'w')
-        fp.write('#!/usr/bin/env python\n\n')
-        fp.write('from distutils2.core import setup\n\n')
+        try:
+            fp.write('#!/usr/bin/env python\n\n')
+            fp.write('from distutils2.core import setup\n\n')
+            fp.write('setup(name=%s,\n' % repr(self.setupData['name']))
+            fp.write('      version=%s,\n' % repr(self.setupData['version']))
+            fp.write('      description=%s,\n'
+                    % repr(self.setupData['description']))
+            fp.write('      author=%s,\n' % repr(self.setupData['author']))
+            fp.write('      author_email=%s,\n'
+                    % repr(self.setupData['author_email']))
+            if self.setupData['url']:
+                fp.write('      url=%s,\n' % repr(self.setupData['url']))
+            if self.setupData['classifier']:
+                fp.write('      classifier=[\n')
+                for classifier in sorted(self.setupData['classifier'].keys()):
+                    fp.write('            %s,\n' % repr(classifier))
+                fp.write('         ],\n')
+            if self.setupData['packages']:
+                fp.write('      packages=%s,\n'
+                        % repr(self._dotted_packages(self.setupData['packages'])))
+                fp.write('      package_dir=%s,\n'
+                        % repr(self.setupData['packages']))
+            fp.write('      #scripts=[\'path/to/script\']\n')
 
-        fp.write('from sys import version\n')
-        fp.write('if version < \'2.2.3\':\n')
-        fp.write('    from distutils2.dist import DistributionMetadata\n')
-        fp.write('    DistributionMetadata.classifier = None\n')
-        fp.write('    DistributionMetadata.download_url = None\n')
-
-        fp.write('setup(name = %s,\n' % repr(self.setupData['name']))
-        fp.write('        version = %s,\n' % repr(self.setupData['version']))
-        fp.write('        description = %s,\n'
-                % repr(self.setupData['description']))
-        fp.write('        author = %s,\n' % repr(self.setupData['author']))
-        fp.write('        author_email = %s,\n'
-                % repr(self.setupData['author_email']))
-        if self.setupData['url']:
-            fp.write('        url = %s,\n' % repr(self.setupData['url']))
-        if self.setupData['classifier']:
-            fp.write('        classifier = [\n')
-            for classifier in sorted(self.setupData['classifier'].keys()):
-                fp.write('              %s,\n' % repr(classifier))
-            fp.write('           ],\n')
-        if self.setupData['packages']:
-            fp.write('        packages = %s,\n'
-                    % repr(self._dotted_packages(self.setupData['packages'])))
-            fp.write('        package_dir = %s,\n'
-                    % repr(self.setupData['packages']))
-        fp.write('        #scripts = [\'path/to/script\']\n')
-
-        fp.write('        )\n')
-        fp.close()
+            fp.write('      )\n')
+        finally:
+            fp.close()
         os.chmod('setup.py', 0755)
 
         print 'Wrote "setup.py".'
diff --git a/src/distutils2/spawn.py b/src/distutils2/spawn.py
--- a/src/distutils2/spawn.py
+++ b/src/distutils2/spawn.py
@@ -14,7 +14,7 @@
 from distutils2.errors import DistutilsPlatformError, DistutilsExecError
 from distutils2 import log
 
-def spawn(cmd, search_path=1, verbose=0, dry_run=0):
+def spawn(cmd, search_path=1, verbose=0, dry_run=0, env=None):
     """Run another program, specified as a command list 'cmd', in a new process.
 
     'cmd' is just the argument list for the new process, ie.
@@ -27,15 +27,18 @@
     must be the exact path to the executable.  If 'dry_run' is true,
     the command will not actually be run.
 
+    If 'env' is given, it's a environment dictionary used for the execution
+    environment.
+
     Raise DistutilsExecError if running the program fails in any way; just
     return on success.
     """
     if os.name == 'posix':
-        _spawn_posix(cmd, search_path, dry_run=dry_run)
+        _spawn_posix(cmd, search_path, dry_run=dry_run, env=env)
     elif os.name == 'nt':
-        _spawn_nt(cmd, search_path, dry_run=dry_run)
+        _spawn_nt(cmd, search_path, dry_run=dry_run, env=env)
     elif os.name == 'os2':
-        _spawn_os2(cmd, search_path, dry_run=dry_run)
+        _spawn_os2(cmd, search_path, dry_run=dry_run, env=env)
     else:
         raise DistutilsPlatformError, \
               "don't know how to spawn programs on platform '%s'" % os.name
@@ -56,7 +59,7 @@
             args[i] = '"%s"' % arg
     return args
 
-def _spawn_nt(cmd, search_path=1, verbose=0, dry_run=0):
+def _spawn_nt(cmd, search_path=1, verbose=0, dry_run=0, env=None):
     executable = cmd[0]
     cmd = _nt_quote_args(cmd)
     if search_path:
@@ -66,7 +69,11 @@
     if not dry_run:
         # spawn for NT requires a full path to the .exe
         try:
-            rc = os.spawnv(os.P_WAIT, executable, cmd)
+            if env is None:
+                rc = os.spawnv(os.P_WAIT, executable, cmd)
+            else:
+                rc = os.spawnve(os.P_WAIT, executable, cmd, env)
+
         except OSError, exc:
             # this seems to happen when the command isn't found
             raise DistutilsExecError, \
@@ -76,7 +83,7 @@
             raise DistutilsExecError, \
                   "command '%s' failed with exit status %d" % (cmd[0], rc)
 
-def _spawn_os2(cmd, search_path=1, verbose=0, dry_run=0):
+def _spawn_os2(cmd, search_path=1, verbose=0, dry_run=0, env=None):
     executable = cmd[0]
     if search_path:
         # either we find one or it stays the same
@@ -85,7 +92,11 @@
     if not dry_run:
         # spawnv for OS/2 EMX requires a full path to the .exe
         try:
-            rc = os.spawnv(os.P_WAIT, executable, cmd)
+            if env is None:
+                rc = os.spawnv(os.P_WAIT, executable, cmd)
+            else:
+                rc = os.spawnve(os.P_WAIT, executable, cmd, env)
+
         except OSError, exc:
             # this seems to happen when the command isn't found
             raise DistutilsExecError, \
@@ -97,16 +108,24 @@
                   "command '%s' failed with exit status %d" % (cmd[0], rc)
 
 
-def _spawn_posix(cmd, search_path=1, verbose=0, dry_run=0):
+def _spawn_posix(cmd, search_path=1, verbose=0, dry_run=0, env=None):
     log.info(' '.join(cmd))
     if dry_run:
         return
-    exec_fn = search_path and os.execvp or os.execv
+
+    if env is None:
+        exec_fn = search_path and os.execvp or os.execv
+    else:
+        exec_fn = search_path and os.execvpe or os.execve
+
     pid = os.fork()
 
     if pid == 0:  # in the child
         try:
-            exec_fn(cmd[0], cmd)
+            if env is None:
+                exec_fn(cmd[0], cmd)
+            else:
+                exec_fn(cmd[0], cmd, env)
         except OSError, e:
             sys.stderr.write("unable to execute %s: %s\n" %
                              (cmd[0], e.strerror))
diff --git a/src/distutils2/tests/__init__.py b/src/distutils2/tests/__init__.py
--- a/src/distutils2/tests/__init__.py
+++ b/src/distutils2/tests/__init__.py
@@ -45,7 +45,7 @@
     """Test failed."""
 
 
-class BasicTestRunner:
+class BasicTestRunner(object):
     def run(self, test):
         result = unittest.TestResult()
         test(result)
diff --git a/src/distutils2/tests/support.py b/src/distutils2/tests/support.py
--- a/src/distutils2/tests/support.py
+++ b/src/distutils2/tests/support.py
@@ -14,7 +14,6 @@
 
 from distutils2 import log
 from distutils2.log import DEBUG, INFO, WARN, ERROR, FATAL
-from distutils2.core import Distribution
 
 if sys.version_info >= (2, 7):
     # improved unittest package from 2.7's standard library
@@ -42,7 +41,7 @@
 
     def _log(self, level, msg, args):
         if level not in (DEBUG, INFO, WARN, ERROR, FATAL):
-            raise ValueError('%s wrong log level' % str(level))
+            raise ValueError('%s wrong log level' % level)
         self.logs.append((level, msg, args))
 
     def get_logs(self, *levels):
@@ -105,6 +104,7 @@
         It returns the package directory and the distribution
         instance.
         """
+        from distutils2.dist import Distribution
         tmp_dir = self.mkdtemp()
         pkg_dir = os.path.join(tmp_dir, pkg_name)
         os.mkdir(pkg_dir)
diff --git a/src/distutils2/tests/test_bdist.py b/src/distutils2/tests/test_bdist.py
--- a/src/distutils2/tests/test_bdist.py
+++ b/src/distutils2/tests/test_bdist.py
@@ -1,8 +1,6 @@
 """Tests for distutils.command.bdist."""
 import sys
 import os
-import tempfile
-import shutil
 
 from distutils2.tests import run_unittest
 
@@ -25,7 +23,7 @@
         cmd = bdist(dist)
         cmd.formats = ['msi']
         cmd.ensure_finalized()
-        self.assertEquals(cmd.formats, ['msi'])
+        self.assertEqual(cmd.formats, ['msi'])
 
         # what format bdist offers ?
         # XXX an explicit list in bdist is
@@ -33,9 +31,9 @@
         # we should add a registry
         formats = ['zip', 'gztar', 'bztar', 'ztar', 'tar', 'wininst', 'msi']
         formats.sort()
-        founded = cmd.format_command.keys()
-        founded.sort()
-        self.assertEquals(founded, formats)
+        found = cmd.format_command.keys()
+        found.sort()
+        self.assertEqual(found, formats)
 
 def test_suite():
     return unittest.makeSuite(BuildTestCase)
diff --git a/src/distutils2/tests/test_bdist_dumb.py b/src/distutils2/tests/test_bdist_dumb.py
--- a/src/distutils2/tests/test_bdist_dumb.py
+++ b/src/distutils2/tests/test_bdist_dumb.py
@@ -78,7 +78,7 @@
             base = base.replace(':', '-')
 
         wanted = ['%s.zip' % base]
-        self.assertEquals(dist_created, wanted)
+        self.assertEqual(dist_created, wanted)
 
         # now let's check what we have in the zip file
         # XXX to be done
@@ -87,16 +87,16 @@
         pkg_dir, dist = self.create_dist()
         os.chdir(pkg_dir)
         cmd = bdist_dumb(dist)
-        self.assertEquals(cmd.bdist_dir, None)
+        self.assertEqual(cmd.bdist_dir, None)
         cmd.finalize_options()
 
         # bdist_dir is initialized to bdist_base/dumb if not set
         base = cmd.get_finalized_command('bdist').bdist_base
-        self.assertEquals(cmd.bdist_dir, os.path.join(base, 'dumb'))
+        self.assertEqual(cmd.bdist_dir, os.path.join(base, 'dumb'))
 
         # the format is set to a default value depending on the os.name
         default = cmd.default_format[os.name]
-        self.assertEquals(cmd.format, default)
+        self.assertEqual(cmd.format, default)
 
 def test_suite():
     return unittest.makeSuite(BuildDumbTestCase)
diff --git a/src/distutils2/tests/test_build.py b/src/distutils2/tests/test_build.py
--- a/src/distutils2/tests/test_build.py
+++ b/src/distutils2/tests/test_build.py
@@ -20,11 +20,11 @@
         cmd.finalize_options()
 
         # if not specified, plat_name gets the current platform
-        self.assertEquals(cmd.plat_name, get_platform())
+        self.assertEqual(cmd.plat_name, get_platform())
 
         # build_purelib is build + lib
         wanted = os.path.join(cmd.build_base, 'lib')
-        self.assertEquals(cmd.build_purelib, wanted)
+        self.assertEqual(cmd.build_purelib, wanted)
 
         # build_platlib is 'build/lib.platform-x.x[-pydebug]'
         # examples:
@@ -34,21 +34,21 @@
             self.assertTrue(cmd.build_platlib.endswith('-pydebug'))
             plat_spec += '-pydebug'
         wanted = os.path.join(cmd.build_base, 'lib' + plat_spec)
-        self.assertEquals(cmd.build_platlib, wanted)
+        self.assertEqual(cmd.build_platlib, wanted)
 
         # by default, build_lib = build_purelib
-        self.assertEquals(cmd.build_lib, cmd.build_purelib)
+        self.assertEqual(cmd.build_lib, cmd.build_purelib)
 
         # build_temp is build/temp.<plat>
         wanted = os.path.join(cmd.build_base, 'temp' + plat_spec)
-        self.assertEquals(cmd.build_temp, wanted)
+        self.assertEqual(cmd.build_temp, wanted)
 
         # build_scripts is build/scripts-x.x
         wanted = os.path.join(cmd.build_base, 'scripts-' +  sys.version[0:3])
-        self.assertEquals(cmd.build_scripts, wanted)
+        self.assertEqual(cmd.build_scripts, wanted)
 
         # executable is os.path.normpath(sys.executable)
-        self.assertEquals(cmd.executable, os.path.normpath(sys.executable))
+        self.assertEqual(cmd.executable, os.path.normpath(sys.executable))
 
 def test_suite():
     return unittest.makeSuite(BuildTestCase)
diff --git a/src/distutils2/tests/test_build_clib.py b/src/distutils2/tests/test_build_clib.py
--- a/src/distutils2/tests/test_build_clib.py
+++ b/src/distutils2/tests/test_build_clib.py
@@ -55,14 +55,14 @@
         self.assertRaises(DistutilsSetupError, cmd.get_source_files)
 
         cmd.libraries = [('name', {'sources': ['a', 'b']})]
-        self.assertEquals(cmd.get_source_files(), ['a', 'b'])
+        self.assertEqual(cmd.get_source_files(), ['a', 'b'])
 
         cmd.libraries = [('name', {'sources': ('a', 'b')})]
-        self.assertEquals(cmd.get_source_files(), ['a', 'b'])
+        self.assertEqual(cmd.get_source_files(), ['a', 'b'])
 
         cmd.libraries = [('name', {'sources': ('a', 'b')}),
                          ('name2', {'sources': ['c', 'd']})]
-        self.assertEquals(cmd.get_source_files(), ['a', 'b', 'c', 'd'])
+        self.assertEqual(cmd.get_source_files(), ['a', 'b', 'c', 'd'])
 
     def test_build_libraries(self):
 
@@ -91,11 +91,11 @@
 
         cmd.include_dirs = 'one-dir'
         cmd.finalize_options()
-        self.assertEquals(cmd.include_dirs, ['one-dir'])
+        self.assertEqual(cmd.include_dirs, ['one-dir'])
 
         cmd.include_dirs = None
         cmd.finalize_options()
-        self.assertEquals(cmd.include_dirs, [])
+        self.assertEqual(cmd.include_dirs, [])
 
         cmd.distribution.libraries = 'WONTWORK'
         self.assertRaises(DistutilsSetupError, cmd.finalize_options)
diff --git a/src/distutils2/tests/test_build_ext.py b/src/distutils2/tests/test_build_ext.py
--- a/src/distutils2/tests/test_build_ext.py
+++ b/src/distutils2/tests/test_build_ext.py
@@ -1,6 +1,5 @@
 import sys
 import os
-import tempfile
 import shutil
 from StringIO import StringIO
 import warnings
@@ -81,11 +80,11 @@
         for attr in ('error', 'foo', 'new', 'roj'):
             self.assertTrue(hasattr(xx, attr))
 
-        self.assertEquals(xx.foo(2, 5), 7)
-        self.assertEquals(xx.foo(13,15), 28)
-        self.assertEquals(xx.new().demo(), None)
+        self.assertEqual(xx.foo(2, 5), 7)
+        self.assertEqual(xx.foo(13,15), 28)
+        self.assertEqual(xx.new().demo(), None)
         doc = 'This is a template module just for instruction.'
-        self.assertEquals(xx.__doc__, doc)
+        self.assertEqual(xx.__doc__, doc)
         self.assertTrue(isinstance(xx.Null(), xx.Null))
         self.assertTrue(isinstance(xx.Str(), xx.Str))
 
@@ -195,7 +194,7 @@
         cmd = build_ext(dist)
         cmd.libraries = 'my_lib'
         cmd.finalize_options()
-        self.assertEquals(cmd.libraries, ['my_lib'])
+        self.assertEqual(cmd.libraries, ['my_lib'])
 
         # make sure cmd.library_dirs is turned into a list
         # if it's a string
@@ -209,7 +208,7 @@
         cmd = build_ext(dist)
         cmd.rpath = os.pathsep.join(['one', 'two'])
         cmd.finalize_options()
-        self.assertEquals(cmd.rpath, ['one', 'two'])
+        self.assertEqual(cmd.rpath, ['one', 'two'])
 
         # XXX more tests to perform for win32
 
@@ -218,25 +217,25 @@
         cmd = build_ext(dist)
         cmd.define = 'one,two'
         cmd.finalize_options()
-        self.assertEquals(cmd.define, [('one', '1'), ('two', '1')])
+        self.assertEqual(cmd.define, [('one', '1'), ('two', '1')])
 
         # make sure undef is turned into a list of
         # strings if they are ','-separated strings
         cmd = build_ext(dist)
         cmd.undef = 'one,two'
         cmd.finalize_options()
-        self.assertEquals(cmd.undef, ['one', 'two'])
+        self.assertEqual(cmd.undef, ['one', 'two'])
 
         # make sure swig_opts is turned into a list
         cmd = build_ext(dist)
         cmd.swig_opts = None
         cmd.finalize_options()
-        self.assertEquals(cmd.swig_opts, [])
+        self.assertEqual(cmd.swig_opts, [])
 
         cmd = build_ext(dist)
         cmd.swig_opts = '1 2'
         cmd.finalize_options()
-        self.assertEquals(cmd.swig_opts, ['1', '2'])
+        self.assertEqual(cmd.swig_opts, ['1', '2'])
 
     def test_check_extensions_list(self):
         dist = Distribution()
@@ -272,7 +271,7 @@
         # check_extensions_list adds in ext the values passed
         # when they are in ('include_dirs', 'library_dirs', 'libraries'
         # 'extra_objects', 'extra_compile_args', 'extra_link_args')
-        self.assertEquals(ext.libraries, 'foo')
+        self.assertEqual(ext.libraries, 'foo')
         self.assertTrue(not hasattr(ext, 'some'))
 
         # 'macros' element of build info dict must be 1- or 2-tuple
@@ -282,15 +281,15 @@
 
         exts[0][1]['macros'] = [('1', '2'), ('3',)]
         cmd.check_extensions_list(exts)
-        self.assertEquals(exts[0].undef_macros, ['3'])
-        self.assertEquals(exts[0].define_macros, [('1', '2')])
+        self.assertEqual(exts[0].undef_macros, ['3'])
+        self.assertEqual(exts[0].define_macros, [('1', '2')])
 
     def test_get_source_files(self):
         modules = [Extension('foo', ['xxx'], optional=False)]
         dist = Distribution({'name': 'xx', 'ext_modules': modules})
         cmd = build_ext(dist)
         cmd.ensure_finalized()
-        self.assertEquals(cmd.get_source_files(), ['xxx'])
+        self.assertEqual(cmd.get_source_files(), ['xxx'])
 
     def test_compiler_option(self):
         # cmd.compiler is an option and
@@ -301,7 +300,7 @@
         cmd.compiler = 'unix'
         cmd.ensure_finalized()
         cmd.run()
-        self.assertEquals(cmd.compiler, 'unix')
+        self.assertEqual(cmd.compiler, 'unix')
 
     def test_get_outputs(self):
         tmp_dir = self.mkdtemp()
@@ -312,7 +311,7 @@
                              'ext_modules': [ext]})
         cmd = build_ext(dist)
         cmd.ensure_finalized()
-        self.assertEquals(len(cmd.get_outputs()), 1)
+        self.assertEqual(len(cmd.get_outputs()), 1)
 
         if os.name == "nt":
             cmd.debug = sys.executable.endswith("_d.exe")
@@ -332,19 +331,19 @@
         finally:
             os.chdir(old_wd)
         self.assertTrue(os.path.exists(so_file))
-        self.assertEquals(os.path.splitext(so_file)[-1],
+        self.assertEqual(os.path.splitext(so_file)[-1],
                           sysconfig.get_config_var('SO'))
         so_dir = os.path.dirname(so_file)
-        self.assertEquals(so_dir, other_tmp_dir)
+        self.assertEqual(so_dir, other_tmp_dir)
 
         cmd.inplace = 0
         cmd.run()
         so_file = cmd.get_outputs()[0]
         self.assertTrue(os.path.exists(so_file))
-        self.assertEquals(os.path.splitext(so_file)[-1],
+        self.assertEqual(os.path.splitext(so_file)[-1],
                           sysconfig.get_config_var('SO'))
         so_dir = os.path.dirname(so_file)
-        self.assertEquals(so_dir, cmd.build_lib)
+        self.assertEqual(so_dir, cmd.build_lib)
 
         # inplace = 0, cmd.package = 'bar'
         build_py = cmd.get_finalized_command('build_py')
@@ -352,7 +351,7 @@
         path = cmd.get_ext_fullpath('foo')
         # checking that the last directory is the build_dir
         path = os.path.split(path)[0]
-        self.assertEquals(path, cmd.build_lib)
+        self.assertEqual(path, cmd.build_lib)
 
         # inplace = 1, cmd.package = 'bar'
         cmd.inplace = 1
@@ -366,7 +365,7 @@
         # checking that the last directory is bar
         path = os.path.split(path)[0]
         lastdir = os.path.split(path)[-1]
-        self.assertEquals(lastdir, 'bar')
+        self.assertEqual(lastdir, 'bar')
 
     def test_ext_fullpath(self):
         ext = sysconfig.get_config_vars()['SO']
@@ -382,14 +381,14 @@
         curdir = os.getcwd()
         wanted = os.path.join(curdir, 'src', 'lxml', 'etree' + ext)
         path = cmd.get_ext_fullpath('lxml.etree')
-        self.assertEquals(wanted, path)
+        self.assertEqual(wanted, path)
 
         # building lxml.etree not inplace
         cmd.inplace = 0
         cmd.build_lib = os.path.join(curdir, 'tmpdir')
         wanted = os.path.join(curdir, 'tmpdir', 'lxml', 'etree' + ext)
         path = cmd.get_ext_fullpath('lxml.etree')
-        self.assertEquals(wanted, path)
+        self.assertEqual(wanted, path)
 
         # building twisted.runner.portmap not inplace
         build_py = cmd.get_finalized_command('build_py')
@@ -398,13 +397,13 @@
         path = cmd.get_ext_fullpath('twisted.runner.portmap')
         wanted = os.path.join(curdir, 'tmpdir', 'twisted', 'runner',
                               'portmap' + ext)
-        self.assertEquals(wanted, path)
+        self.assertEqual(wanted, path)
 
         # building twisted.runner.portmap inplace
         cmd.inplace = 1
         path = cmd.get_ext_fullpath('twisted.runner.portmap')
         wanted = os.path.join(curdir, 'twisted', 'runner', 'portmap' + ext)
-        self.assertEquals(wanted, path)
+        self.assertEqual(wanted, path)
 
 def test_suite():
     src = _get_source_filename()
diff --git a/src/distutils2/tests/test_build_py.py b/src/distutils2/tests/test_build_py.py
--- a/src/distutils2/tests/test_build_py.py
+++ b/src/distutils2/tests/test_build_py.py
@@ -19,11 +19,15 @@
     def test_package_data(self):
         sources = self.mkdtemp()
         f = open(os.path.join(sources, "__init__.py"), "w")
-        f.write("# Pretend this is a package.")
-        f.close()
+        try:
+            f.write("# Pretend this is a package.")
+        finally:
+            f.close()
         f = open(os.path.join(sources, "README.txt"), "w")
-        f.write("Info about this package")
-        f.close()
+        try:
+            f.write("Info about this package")
+        finally:
+            f.close()
 
         destination = self.mkdtemp()
 
diff --git a/src/distutils2/tests/test_build_scripts.py b/src/distutils2/tests/test_build_scripts.py
--- a/src/distutils2/tests/test_build_scripts.py
+++ b/src/distutils2/tests/test_build_scripts.py
@@ -74,8 +74,10 @@
 
     def write_script(self, dir, name, text):
         f = open(os.path.join(dir, name), "w")
-        f.write(text)
-        f.close()
+        try:
+            f.write(text)
+        finally:
+            f.close()
 
     def test_version_int(self):
         source = self.mkdtemp()
diff --git a/src/distutils2/tests/test_ccompiler.py b/src/distutils2/tests/test_ccompiler.py
--- a/src/distutils2/tests/test_ccompiler.py
+++ b/src/distutils2/tests/test_ccompiler.py
@@ -31,7 +31,7 @@
         opts = gen_lib_options(compiler, libdirs, runlibdirs, libs)
         wanted = ['-Llib1', '-Llib2', '-cool', '-Rrunlib1', 'found',
                   '-lname2']
-        self.assertEquals(opts, wanted)
+        self.assertEqual(opts, wanted)
 
     def test_customize_compiler(self):
 
@@ -51,7 +51,7 @@
 
         comp = compiler()
         customize_compiler(comp)
-        self.assertEquals(comp.exes['archiver'], 'my_ar -arflags')
+        self.assertEqual(comp.exes['archiver'], 'my_ar -arflags')
 
 def test_suite():
     return unittest.makeSuite(CCompilerTestCase)
diff --git a/src/distutils2/tests/test_check.py b/src/distutils2/tests/test_check.py
--- a/src/distutils2/tests/test_check.py
+++ b/src/distutils2/tests/test_check.py
@@ -37,7 +37,7 @@
                     'name': 'xxx', 'version': 'xxx'
                     }
         cmd = self._run(metadata)
-        self.assertEquals(len(cmd._warnings), 0)
+        self.assertEqual(len(cmd._warnings), 0)
 
         # now with the strict mode, we should
         # get an error if there are missing metadata
@@ -45,7 +45,7 @@
 
         # and of course, no error when all metadata are present
         cmd = self._run(metadata, strict=1)
-        self.assertEquals(len(cmd._warnings), 0)
+        self.assertEqual(len(cmd._warnings), 0)
 
     def test_check_restructuredtext(self):
         if not _HAS_DOCUTILS: # won't test without docutils
@@ -55,7 +55,7 @@
         pkg_info, dist = self.create_dist(description=broken_rest)
         cmd = check(dist)
         cmd.check_restructuredtext()
-        self.assertEquals(len(cmd._warnings), 1)
+        self.assertEqual(len(cmd._warnings), 1)
 
         # let's see if we have an error with strict=1
         metadata = {'home_page': 'xxx', 'author': 'xxx',
@@ -69,7 +69,7 @@
         # and non-broken rest
         metadata['description'] = 'title\n=====\n\ntest'
         cmd = self._run(metadata, strict=1, restructuredtext=1)
-        self.assertEquals(len(cmd._warnings), 0)
+        self.assertEqual(len(cmd._warnings), 0)
 
     def test_check_all(self):
 
diff --git a/src/distutils2/tests/test_cmd.py b/src/distutils2/tests/test_cmd.py
--- a/src/distutils2/tests/test_cmd.py
+++ b/src/distutils2/tests/test_cmd.py
@@ -43,7 +43,7 @@
 
         # making sure execute gets called properly
         def _execute(func, args, exec_msg, level):
-            self.assertEquals(exec_msg, 'generating out from in')
+            self.assertEqual(exec_msg, 'generating out from in')
         cmd.force = True
         cmd.execute = _execute
         cmd.make_file(infiles='in', outfile='out', func='func', args=())
@@ -62,7 +62,7 @@
 
         wanted = ["command options for 'MyCmd':", '  option1 = 1',
                   '  option2 = 1']
-        self.assertEquals(msgs, wanted)
+        self.assertEqual(msgs, wanted)
 
     def test_ensure_string(self):
         cmd = self.cmd
@@ -80,7 +80,7 @@
         cmd = self.cmd
         cmd.option1 = 'ok,dok'
         cmd.ensure_string_list('option1')
-        self.assertEquals(cmd.option1, ['ok', 'dok'])
+        self.assertEqual(cmd.option1, ['ok', 'dok'])
 
         cmd.option2 = ['xxx', 'www']
         cmd.ensure_string_list('option2')
diff --git a/src/distutils2/tests/test_config.py b/src/distutils2/tests/test_config.py
--- a/src/distutils2/tests/test_config.py
+++ b/src/distutils2/tests/test_config.py
@@ -1,7 +1,6 @@
 """Tests for distutils.pypirc.pypirc."""
 import sys
 import os
-import tempfile
 import shutil
 
 from distutils2.core import PyPIRCCommand
@@ -87,20 +86,20 @@
 
         config = config.items()
         config.sort()
-        waited = [('password', 'secret'), ('realm', 'pypi'),
-                  ('repository', 'http://pypi.python.org/pypi'),
-                  ('server', 'server1'), ('username', 'me')]
-        self.assertEquals(config, waited)
+        expected = [('password', 'secret'), ('realm', 'pypi'),
+                    ('repository', 'http://pypi.python.org/pypi'),
+                    ('server', 'server1'), ('username', 'me')]
+        self.assertEqual(config, expected)
 
         # old format
         self.write_file(self.rc, PYPIRC_OLD)
         config = cmd._read_pypirc()
         config = config.items()
         config.sort()
-        waited = [('password', 'secret'), ('realm', 'pypi'),
-                  ('repository', 'http://pypi.python.org/pypi'),
-                  ('server', 'server-login'), ('username', 'tarek')]
-        self.assertEquals(config, waited)
+        expected = [('password', 'secret'), ('realm', 'pypi'),
+                    ('repository', 'http://pypi.python.org/pypi'),
+                    ('server', 'server-login'), ('username', 'tarek')]
+        self.assertEqual(config, expected)
 
     def test_server_empty_registration(self):
         cmd = self._cmd(self.dist)
@@ -109,7 +108,7 @@
         cmd._store_pypirc('tarek', 'xxx')
         self.assertTrue(os.path.exists(rc))
         content = open(rc).read()
-        self.assertEquals(content, WANTED)
+        self.assertEqual(content, WANTED)
 
 def test_suite():
     return unittest.makeSuite(PyPIRCCommandTestCase)
diff --git a/src/distutils2/tests/test_config_cmd.py b/src/distutils2/tests/test_config_cmd.py
--- a/src/distutils2/tests/test_config_cmd.py
+++ b/src/distutils2/tests/test_config_cmd.py
@@ -34,7 +34,7 @@
             f.close()
 
         dump_file(this_file, 'I am the header')
-        self.assertEquals(len(self._logs), numlines+1)
+        self.assertEqual(len(self._logs), numlines+1)
 
     def test_search_cpp(self):
         if sys.platform == 'win32':
@@ -44,10 +44,10 @@
 
         # simple pattern searches
         match = cmd.search_cpp(pattern='xxx', body='// xxx')
-        self.assertEquals(match, 0)
+        self.assertEqual(match, 0)
 
         match = cmd.search_cpp(pattern='_configtest', body='// xxx')
-        self.assertEquals(match, 1)
+        self.assertEqual(match, 1)
 
     def test_finalize_options(self):
         # finalize_options does a bit of transformation
@@ -59,9 +59,9 @@
         cmd.library_dirs = 'three%sfour' % os.pathsep
         cmd.ensure_finalized()
 
-        self.assertEquals(cmd.include_dirs, ['one', 'two'])
-        self.assertEquals(cmd.libraries, ['one'])
-        self.assertEquals(cmd.library_dirs, ['three', 'four'])
+        self.assertEqual(cmd.include_dirs, ['one', 'two'])
+        self.assertEqual(cmd.libraries, ['one'])
+        self.assertEqual(cmd.library_dirs, ['three', 'four'])
 
     def test_clean(self):
         # _clean removes files
diff --git a/src/distutils2/tests/test_converter.py b/src/distutils2/tests/test_converter.py
--- a/src/distutils2/tests/test_converter.py
+++ b/src/distutils2/tests/test_converter.py
@@ -30,7 +30,7 @@
             wanted = file_.replace('before', 'after')
             wanted = _read_file(os.path.join(convdir, wanted))
             res = ref.refactor_string(original, 'setup.py')
-            self.assertEquals(str(res), wanted)
+            self.assertEqual(str(res), wanted)
 
 def test_suite():
     return unittest.makeSuite(ConverterTestCase)
diff --git a/src/distutils2/tests/test_cygwinccompiler.py b/src/distutils2/tests/test_cygwinccompiler.py
--- a/src/distutils2/tests/test_cygwinccompiler.py
+++ b/src/distutils2/tests/test_cygwinccompiler.py
@@ -44,48 +44,48 @@
         sys.version = ('2.6.1 (r261:67515, Dec  6 2008, 16:42:21) \n[GCC '
                        '4.0.1 (Apple Computer, Inc. build 5370)]')
 
-        self.assertEquals(check_config_h()[0], CONFIG_H_OK)
+        self.assertEqual(check_config_h()[0], CONFIG_H_OK)
 
         # then it tries to see if it can find "__GNUC__" in pyconfig.h
         sys.version = 'something without the *CC word'
 
         # if the file doesn't exist it returns  CONFIG_H_UNCERTAIN
-        self.assertEquals(check_config_h()[0], CONFIG_H_UNCERTAIN)
+        self.assertEqual(check_config_h()[0], CONFIG_H_UNCERTAIN)
 
         # if it exists but does not contain __GNUC__, it returns CONFIG_H_NOTOK
         self.write_file(self.python_h, 'xxx')
-        self.assertEquals(check_config_h()[0], CONFIG_H_NOTOK)
+        self.assertEqual(check_config_h()[0], CONFIG_H_NOTOK)
 
         # and CONFIG_H_OK if __GNUC__ is found
         self.write_file(self.python_h, 'xxx __GNUC__ xxx')
-        self.assertEquals(check_config_h()[0], CONFIG_H_OK)
+        self.assertEqual(check_config_h()[0], CONFIG_H_OK)
 
     def test_get_msvcr(self):
 
         # none
         sys.version  = ('2.6.1 (r261:67515, Dec  6 2008, 16:42:21) '
                         '\n[GCC 4.0.1 (Apple Computer, Inc. build 5370)]')
-        self.assertEquals(get_msvcr(), None)
+        self.assertEqual(get_msvcr(), None)
 
         # MSVC 7.0
         sys.version = ('2.5.1 (r251:54863, Apr 18 2007, 08:51:08) '
                        '[MSC v.1300 32 bits (Intel)]')
-        self.assertEquals(get_msvcr(), ['msvcr70'])
+        self.assertEqual(get_msvcr(), ['msvcr70'])
 
         # MSVC 7.1
         sys.version = ('2.5.1 (r251:54863, Apr 18 2007, 08:51:08) '
                        '[MSC v.1310 32 bits (Intel)]')
-        self.assertEquals(get_msvcr(), ['msvcr71'])
+        self.assertEqual(get_msvcr(), ['msvcr71'])
 
         # VS2005 / MSVC 8.0
         sys.version = ('2.5.1 (r251:54863, Apr 18 2007, 08:51:08) '
                        '[MSC v.1400 32 bits (Intel)]')
-        self.assertEquals(get_msvcr(), ['msvcr80'])
+        self.assertEqual(get_msvcr(), ['msvcr80'])
 
         # VS2008 / MSVC 9.0
         sys.version = ('2.5.1 (r251:54863, Apr 18 2007, 08:51:08) '
                        '[MSC v.1500 32 bits (Intel)]')
-        self.assertEquals(get_msvcr(), ['msvcr90'])
+        self.assertEqual(get_msvcr(), ['msvcr90'])
 
         # unknown
         sys.version = ('2.5.1 (r251:54863, Apr 18 2007, 08:51:08) '
diff --git a/src/distutils2/tests/test_depgraph.py b/src/distutils2/tests/test_depgraph.py
--- a/src/distutils2/tests/test_depgraph.py
+++ b/src/distutils2/tests/test_depgraph.py
@@ -27,7 +27,7 @@
         self.assertListEqual(sorted(l1), sorted(l2))
 
     def setUp(self):
-        super(unittest.TestCase, self).setUp()
+        super(DepGraphTestCase, self).setUp()
         path = os.path.join(os.path.dirname(__file__), '..', '_backport',
                             'tests', 'fake_dists')
         path = os.path.abspath(path)
@@ -177,7 +177,7 @@
         self.checkLists(matches, expected)
 
     def tearDown(self):
-        super(unittest.TestCase, self).tearDown()
+        super(DepGraphTestCase, self).tearDown()
         sys.path = self.sys_path
 
 def test_suite():
diff --git a/src/distutils2/tests/test_dist.py b/src/distutils2/tests/test_dist.py
--- a/src/distutils2/tests/test_dist.py
+++ b/src/distutils2/tests/test_dist.py
@@ -71,11 +71,11 @@
         sys.argv.append("build")
 
         __, stdout = captured_stdout(self.create_distribution, files)
-        self.assertEquals(stdout, '')
+        self.assertEqual(stdout, '')
         distutils2.dist.DEBUG = True
         try:
             __, stdout = captured_stdout(self.create_distribution, files)
-            self.assertEquals(stdout, '')
+            self.assertEqual(stdout, '')
         finally:
             distutils2.dist.DEBUG = False
 
@@ -129,13 +129,13 @@
         # Check DistributionMetadata handling of Unicode fields
         tmp_dir = self.mkdtemp()
         my_file = os.path.join(tmp_dir, 'f')
-        klass = Distribution
+        cls = Distribution
 
-        dist = klass(attrs={'author': u'Mister Caf??',
-                            'name': 'my.package',
-                            'maintainer': u'Caf?? Junior',
-                            'summary': u'Caf?? torr??fi??',
-                            'description': u'H??h??h??'})
+        dist = cls(attrs={'author': u'Mister Caf??',
+                          'name': 'my.package',
+                          'maintainer': u'Caf?? Junior',
+                          'summary': u'Caf?? torr??fi??',
+                          'description': u'H??h??h??'})
 
 
         # let's make sure the file can be written
@@ -144,11 +144,11 @@
         dist.metadata.write_file(open(my_file, 'w'))
 
         # regular ascii is of course always usable
-        dist = klass(attrs={'author': 'Mister Cafe',
-                            'name': 'my.package',
-                            'maintainer': 'Cafe Junior',
-                            'summary': 'Cafe torrefie',
-                            'description': 'Hehehe'})
+        dist = cls(attrs={'author': 'Mister Cafe',
+                          'name': 'my.package',
+                          'maintainer': 'Cafe Junior',
+                          'summary': 'Cafe torrefie',
+                          'description': 'Hehehe'})
 
         my_file2 = os.path.join(tmp_dir, 'f2')
         dist.metadata.write_file(open(my_file, 'w'))
@@ -156,7 +156,7 @@
     def test_empty_options(self):
         # an empty options dictionary should not stay in the
         # list of attributes
-        klass = Distribution
+        cls = Distribution
 
         # catching warnings
         warns = []
@@ -166,15 +166,15 @@
         old_warn = warnings.warn
         warnings.warn = _warn
         try:
-            dist = klass(attrs={'author': 'xxx',
-                                'name': 'xxx',
-                                'version': 'xxx',
-                                'url': 'xxxx',
-                                'options': {}})
+            dist = cls(attrs={'author': 'xxx',
+                              'name': 'xxx',
+                              'version': 'xxx',
+                              'url': 'xxxx',
+                              'options': {}})
         finally:
             warnings.warn = old_warn
 
-        self.assertEquals(len(warns), 0)
+        self.assertEqual(len(warns), 0)
 
     def test_finalize_options(self):
 
@@ -185,20 +185,20 @@
         dist.finalize_options()
 
         # finalize_option splits platforms and keywords
-        self.assertEquals(dist.metadata['platform'], ['one', 'two'])
-        self.assertEquals(dist.metadata['keywords'], ['one', 'two'])
+        self.assertEqual(dist.metadata['platform'], ['one', 'two'])
+        self.assertEqual(dist.metadata['keywords'], ['one', 'two'])
 
     def test_get_command_packages(self):
         dist = Distribution()
-        self.assertEquals(dist.command_packages, None)
+        self.assertEqual(dist.command_packages, None)
         cmds = dist.get_command_packages()
-        self.assertEquals(cmds, ['distutils2.command'])
-        self.assertEquals(dist.command_packages,
+        self.assertEqual(cmds, ['distutils2.command'])
+        self.assertEqual(dist.command_packages,
                           ['distutils2.command'])
 
         dist.command_packages = 'one,two'
         cmds = dist.get_command_packages()
-        self.assertEquals(cmds, ['distutils2.command', 'one', 'two'])
+        self.assertEqual(cmds, ['distutils2.command', 'one', 'two'])
 
 
     def test_announce(self):
@@ -238,7 +238,7 @@
             os.path.expanduser = old_expander
 
         # make sure --no-user-cfg disables the user cfg file
-        self.assertEquals(len(all_files)-1, len(files))
+        self.assertEqual(len(all_files)-1, len(files))
 
 
 class MetadataTestCase(support.TempdirManager, support.EnvironGuard,
@@ -340,8 +340,10 @@
         temp_dir = self.mkdtemp()
         user_filename = os.path.join(temp_dir, user_filename)
         f = open(user_filename, 'w')
-        f.write('.')
-        f.close()
+        try:
+            f.write('.')
+        finally:
+            f.close()
 
         try:
             dist = Distribution()
@@ -365,8 +367,8 @@
     def test_fix_help_options(self):
         help_tuples = [('a', 'b', 'c', 'd'), (1, 2, 3, 4)]
         fancy_options = fix_help_options(help_tuples)
-        self.assertEquals(fancy_options[0], ('a', 'b', 'c'))
-        self.assertEquals(fancy_options[1], (1, 2, 3))
+        self.assertEqual(fancy_options[0], ('a', 'b', 'c'))
+        self.assertEqual(fancy_options[1], (1, 2, 3))
 
     def test_show_help(self):
         # smoke test, just makes sure some help is displayed
@@ -412,14 +414,14 @@
         PKG_INFO.seek(0)
 
         metadata.read_file(PKG_INFO)
-        self.assertEquals(metadata['name'], "package")
-        self.assertEquals(metadata['version'], "1.0")
-        self.assertEquals(metadata['summary'], "xxx")
-        self.assertEquals(metadata['download_url'], 'http://example.com')
-        self.assertEquals(metadata['keywords'], ['one', 'two'])
-        self.assertEquals(metadata['platform'], [])
-        self.assertEquals(metadata['obsoletes'], [])
-        self.assertEquals(metadata['requires-dist'], ['foo'])
+        self.assertEqual(metadata['name'], "package")
+        self.assertEqual(metadata['version'], "1.0")
+        self.assertEqual(metadata['summary'], "xxx")
+        self.assertEqual(metadata['download_url'], 'http://example.com')
+        self.assertEqual(metadata['keywords'], ['one', 'two'])
+        self.assertEqual(metadata['platform'], [])
+        self.assertEqual(metadata['obsoletes'], [])
+        self.assertEqual(metadata['requires-dist'], ['foo'])
 
 def test_suite():
     suite = unittest.TestSuite()
diff --git a/src/distutils2/tests/test_install.py b/src/distutils2/tests/test_install.py
--- a/src/distutils2/tests/test_install.py
+++ b/src/distutils2/tests/test_install.py
@@ -139,23 +139,23 @@
 
         # two elements
         cmd.handle_extra_path()
-        self.assertEquals(cmd.extra_path, ['path', 'dirs'])
-        self.assertEquals(cmd.extra_dirs, 'dirs')
-        self.assertEquals(cmd.path_file, 'path')
+        self.assertEqual(cmd.extra_path, ['path', 'dirs'])
+        self.assertEqual(cmd.extra_dirs, 'dirs')
+        self.assertEqual(cmd.path_file, 'path')
 
         # one element
         cmd.extra_path = ['path']
         cmd.handle_extra_path()
-        self.assertEquals(cmd.extra_path, ['path'])
-        self.assertEquals(cmd.extra_dirs, 'path')
-        self.assertEquals(cmd.path_file, 'path')
+        self.assertEqual(cmd.extra_path, ['path'])
+        self.assertEqual(cmd.extra_dirs, 'path')
+        self.assertEqual(cmd.path_file, 'path')
 
         # none
         dist.extra_path = cmd.extra_path = None
         cmd.handle_extra_path()
-        self.assertEquals(cmd.extra_path, None)
-        self.assertEquals(cmd.extra_dirs, '')
-        self.assertEquals(cmd.path_file, None)
+        self.assertEqual(cmd.extra_path, None)
+        self.assertEqual(cmd.extra_dirs, '')
+        self.assertEqual(cmd.path_file, None)
 
         # three elements (no way !)
         cmd.extra_path = 'path,dirs,again'
@@ -199,7 +199,7 @@
         # line (the egg info file)
         f = open(cmd.record)
         try:
-            self.assertEquals(len(f.readlines()), 1)
+            self.assertEqual(len(f.readlines()), 1)
         finally:
             f.close()
 
diff --git a/src/distutils2/tests/test_install_data.py b/src/distutils2/tests/test_install_data.py
--- a/src/distutils2/tests/test_install_data.py
+++ b/src/distutils2/tests/test_install_data.py
@@ -27,14 +27,14 @@
         self.write_file(two, 'xxx')
 
         cmd.data_files = [one, (inst2, [two])]
-        self.assertEquals(cmd.get_inputs(), [one, (inst2, [two])])
+        self.assertEqual(cmd.get_inputs(), [one, (inst2, [two])])
 
         # let's run the command
         cmd.ensure_finalized()
         cmd.run()
 
         # let's check the result
-        self.assertEquals(len(cmd.get_outputs()), 2)
+        self.assertEqual(len(cmd.get_outputs()), 2)
         rtwo = os.path.split(two)[-1]
         self.assertTrue(os.path.exists(os.path.join(inst2, rtwo)))
         rone = os.path.split(one)[-1]
@@ -47,7 +47,7 @@
         cmd.run()
 
         # let's check the result
-        self.assertEquals(len(cmd.get_outputs()), 2)
+        self.assertEqual(len(cmd.get_outputs()), 2)
         self.assertTrue(os.path.exists(os.path.join(inst2, rtwo)))
         self.assertTrue(os.path.exists(os.path.join(inst, rone)))
         cmd.outfiles = []
@@ -65,7 +65,7 @@
         cmd.run()
 
         # let's check the result
-        self.assertEquals(len(cmd.get_outputs()), 4)
+        self.assertEqual(len(cmd.get_outputs()), 4)
         self.assertTrue(os.path.exists(os.path.join(inst2, rtwo)))
         self.assertTrue(os.path.exists(os.path.join(inst, rone)))
 
diff --git a/src/distutils2/tests/test_install_headers.py b/src/distutils2/tests/test_install_headers.py
--- a/src/distutils2/tests/test_install_headers.py
+++ b/src/distutils2/tests/test_install_headers.py
@@ -23,7 +23,7 @@
 
         pkg_dir, dist = self.create_dist(headers=headers)
         cmd = install_headers(dist)
-        self.assertEquals(cmd.get_inputs(), headers)
+        self.assertEqual(cmd.get_inputs(), headers)
 
         # let's run the command
         cmd.install_dir = os.path.join(pkg_dir, 'inst')
@@ -31,7 +31,7 @@
         cmd.run()
 
         # let's check the results
-        self.assertEquals(len(cmd.get_outputs()), 2)
+        self.assertEqual(len(cmd.get_outputs()), 2)
 
 def test_suite():
     return unittest.makeSuite(InstallHeadersTestCase)
diff --git a/src/distutils2/tests/test_install_lib.py b/src/distutils2/tests/test_install_lib.py
--- a/src/distutils2/tests/test_install_lib.py
+++ b/src/distutils2/tests/test_install_lib.py
@@ -25,8 +25,8 @@
         cmd = install_lib(dist)
 
         cmd.finalize_options()
-        self.assertEquals(cmd.compile, 1)
-        self.assertEquals(cmd.optimize, 0)
+        self.assertEqual(cmd.compile, 1)
+        self.assertEqual(cmd.optimize, 0)
 
         # optimize must be 0, 1, or 2
         cmd.optimize = 'foo'
@@ -36,7 +36,7 @@
 
         cmd.optimize = '2'
         cmd.finalize_options()
-        self.assertEquals(cmd.optimize, 2)
+        self.assertEqual(cmd.optimize, 2)
 
     @unittest.skipIf(no_bytecode, 'byte-compile not supported')
     def test_byte_compile(self):
@@ -82,7 +82,7 @@
         cmd.distribution.script_name = 'setup.py'
 
         # get_input should return 2 elements
-        self.assertEquals(len(cmd.get_inputs()), 2)
+        self.assertEqual(len(cmd.get_inputs()), 2)
 
     @unittest.skipUnless(bytecode_support, 'sys.dont_write_bytecode not supported')
     def test_dont_write_bytecode(self):
diff --git a/src/distutils2/tests/test_install_scripts.py b/src/distutils2/tests/test_install_scripts.py
--- a/src/distutils2/tests/test_install_scripts.py
+++ b/src/distutils2/tests/test_install_scripts.py
@@ -42,8 +42,10 @@
         def write_script(name, text):
             expected.append(name)
             f = open(os.path.join(source, name), "w")
-            f.write(text)
-            f.close()
+            try:
+                f.write(text)
+            finally:
+                f.close()
 
         write_script("script1.py", ("#! /usr/bin/env python2.3\n"
                                     "# bogus script w/ Python sh-bang\n"
diff --git a/src/distutils2/tests/test_manifest.py b/src/distutils2/tests/test_manifest.py
--- a/src/distutils2/tests/test_manifest.py
+++ b/src/distutils2/tests/test_manifest.py
@@ -44,7 +44,7 @@
 
         # the manifest should have been read
         # and 3 warnings issued (we ddidn't provided the files)
-        self.assertEquals(len(warns), 3)
+        self.assertEqual(len(warns), 3)
         for warn in warns:
             self.assertIn('warning: no files found matching', warn)
 
diff --git a/src/distutils2/tests/test_metadata.py b/src/distutils2/tests/test_metadata.py
--- a/src/distutils2/tests/test_metadata.py
+++ b/src/distutils2/tests/test_metadata.py
@@ -5,9 +5,9 @@
 
 from distutils2.metadata import (DistributionMetadata, _interpret,
                                  PKG_INFO_PREFERRED_VERSION)
-from distutils2.tests.support import unittest
+from distutils2.tests.support import unittest, LoggingSilencer
 
-class DistributionMetadataTestCase(unittest.TestCase):
+class DistributionMetadataTestCase(LoggingSilencer, unittest.TestCase):
 
 
     def test_interpret(self):
@@ -15,15 +15,15 @@
         version = sys.version.split()[0]
         os_name = os.name
 
-        assert _interpret("sys.platform == '%s'" % platform)
-        assert _interpret("sys.platform == '%s' or python_version == '2.4'" \
-                % platform)
-        assert _interpret("sys.platform == '%s' and "
-                          "python_full_version == '%s'"\
-                % (platform, version))
-        assert _interpret("'%s' == sys.platform" % platform)
+        self.assertTrue(_interpret("sys.platform == '%s'" % platform))
+        self.assertTrue(_interpret(
+            "sys.platform == '%s' or python_version == '2.4'" % platform))
+        self.assertTrue(_interpret(
+            "sys.platform == '%s' and python_full_version == '%s'" %
+            (platform, version)))
+        self.assertTrue(_interpret("'%s' == sys.platform" % platform))
 
-        assert _interpret('os.name == "%s"' % os_name)
+        self.assertTrue(_interpret('os.name == "%s"' % os_name))
 
         # stuff that need to raise a syntax error
         ops = ('os.name == os.name', 'os.name == 2', "'2' == '2'",
@@ -35,24 +35,25 @@
         OP = 'os.name == "%s"' % os_name
         AND = ' and '
         OR = ' or '
-        assert _interpret(OP+AND+OP)
-        assert _interpret(OP+AND+OP+AND+OP)
-        assert _interpret(OP+OR+OP)
-        assert _interpret(OP+OR+OP+OR+OP)
+        self.assertTrue(_interpret(OP + AND + OP))
+        self.assertTrue(_interpret(OP + AND + OP + AND + OP))
+        self.assertTrue(_interpret(OP + OR + OP))
+        self.assertTrue(_interpret(OP + OR + OP + OR + OP))
 
         # other operators
-        assert _interpret("os.name != 'buuuu'")
-        assert _interpret("python_version > '1.0'")
-        assert _interpret("python_version < '5.0'")
-        assert _interpret("python_version <= '5.0'")
-        assert _interpret("python_version >= '1.0'")
-        assert _interpret("'%s' in os.name" % os_name)
-        assert _interpret("'buuuu' not in os.name")
-        assert _interpret("'buuuu' not in os.name and '%s' in os.name" \
-                            % os_name)
+        self.assertTrue(_interpret("os.name != 'buuuu'"))
+        self.assertTrue(_interpret("python_version > '1.0'"))
+        self.assertTrue(_interpret("python_version < '5.0'"))
+        self.assertTrue(_interpret("python_version <= '5.0'"))
+        self.assertTrue(_interpret("python_version >= '1.0'"))
+        self.assertTrue(_interpret("'%s' in os.name" % os_name))
+        self.assertTrue(_interpret("'buuuu' not in os.name"))
+        self.assertTrue(_interpret(
+            "'buuuu' not in os.name and '%s' in os.name" % os_name))
 
         # execution context
-        assert _interpret('python_version == "0.1"', {'python_version': '0.1'})
+        self.assertTrue(_interpret('python_version == "0.1"',
+                                   {'python_version': '0.1'}))
 
     def test_metadata_read_write(self):
 
@@ -63,25 +64,28 @@
         res.seek(0)
         res = res.read()
         f = open(PKG_INFO)
-        wanted = f.read()
+        try:
+            # XXX this is not used
+            wanted = f.read()
+        finally:
+            f.close()
         self.assertTrue('Keywords: keyring,password,crypt' in res)
-        f.close()
 
     def test_metadata_markers(self):
         # see if we can be platform-aware
         PKG_INFO = os.path.join(os.path.dirname(__file__), 'PKG-INFO')
         content = open(PKG_INFO).read()
         content = content % sys.platform
-        metadata = DistributionMetadata(platform_dependant=True)
+        metadata = DistributionMetadata(platform_dependent=True)
         metadata.read_file(StringIO(content))
-        self.assertEquals(metadata['Requires-Dist'], ['bar'])
+        self.assertEqual(metadata['Requires-Dist'], ['bar'])
 
         # test with context
         context = {'sys.platform': 'okook'}
-        metadata = DistributionMetadata(platform_dependant=True,
+        metadata = DistributionMetadata(platform_dependent=True,
                                         execution_context=context)
         metadata.read_file(StringIO(content))
-        self.assertEquals(metadata['Requires-Dist'], ['foo'])
+        self.assertEqual(metadata['Requires-Dist'], ['foo'])
 
     def test_description(self):
         PKG_INFO = os.path.join(os.path.dirname(__file__), 'PKG-INFO')
@@ -93,14 +97,14 @@
         # see if we can read the description now
         DESC = os.path.join(os.path.dirname(__file__), 'LONG_DESC.txt')
         wanted = open(DESC).read()
-        self.assertEquals(wanted, metadata['Description'])
+        self.assertEqual(wanted, metadata['Description'])
 
         # save the file somewhere and make sure we can read it back
         out = StringIO()
         metadata.write_file(out)
         out.seek(0)
         metadata.read_file(out)
-        self.assertEquals(wanted, metadata['Description'])
+        self.assertEqual(wanted, metadata['Description'])
 
     def test_mapper_apis(self):
         PKG_INFO = os.path.join(os.path.dirname(__file__), 'PKG-INFO')
@@ -115,25 +119,25 @@
     def test_versions(self):
         metadata = DistributionMetadata()
         metadata['Obsoletes'] = 'ok'
-        self.assertEquals(metadata['Metadata-Version'], '1.1')
+        self.assertEqual(metadata['Metadata-Version'], '1.1')
 
         del metadata['Obsoletes']
         metadata['Obsoletes-Dist'] = 'ok'
-        self.assertEquals(metadata['Metadata-Version'], '1.2')
+        self.assertEqual(metadata['Metadata-Version'], '1.2')
 
         del metadata['Obsoletes-Dist']
         metadata['Version'] = '1'
-        self.assertEquals(metadata['Metadata-Version'], '1.0')
+        self.assertEqual(metadata['Metadata-Version'], '1.0')
 
         PKG_INFO = os.path.join(os.path.dirname(__file__),
                                 'SETUPTOOLS-PKG-INFO')
         metadata.read_file(StringIO(open(PKG_INFO).read()))
-        self.assertEquals(metadata['Metadata-Version'], '1.0')
+        self.assertEqual(metadata['Metadata-Version'], '1.0')
 
         PKG_INFO = os.path.join(os.path.dirname(__file__),
                                 'SETUPTOOLS-PKG-INFO2')
         metadata.read_file(StringIO(open(PKG_INFO).read()))
-        self.assertEquals(metadata['Metadata-Version'], '1.1')
+        self.assertEqual(metadata['Metadata-Version'], '1.1')
 
     def test_warnings(self):
         metadata = DistributionMetadata()
@@ -161,7 +165,7 @@
 
         # we should have a certain amount of warnings
         num_wanted = len(values)
-        self.assertEquals(num_wanted, res)
+        self.assertEqual(num_wanted, res)
 
     def test_multiple_predicates(self):
         metadata = DistributionMetadata()
@@ -184,29 +188,29 @@
             res = m.warns
             del m.warns
 
-        self.assertEquals(res, 0)
+        self.assertEqual(res, 0)
 
     def test_project_url(self):
         metadata = DistributionMetadata()
         metadata['Project-URL'] = [('one', 'http://ok')]
-        self.assertEquals(metadata['Project-URL'],
+        self.assertEqual(metadata['Project-URL'],
                           [('one', 'http://ok')])
-        self.assertEquals(metadata.version, '1.2')
+        self.assertEqual(metadata.version, '1.2')
 
     def test_check(self):
         metadata = DistributionMetadata()
         metadata['Version'] = 'rr'
         metadata['Requires-dist'] = ['Foo (a)']
         missing, warnings = metadata.check()
-        self.assertEquals(missing, ['Name', 'Home-page'])
-        self.assertEquals(len(warnings), 2)
+        self.assertEqual(missing, ['Name', 'Home-page'])
+        self.assertEqual(len(warnings), 2)
 
     def test_best_choice(self):
         metadata = DistributionMetadata()
         metadata['Version'] = '1.0'
-        self.assertEquals(metadata.version, PKG_INFO_PREFERRED_VERSION)
+        self.assertEqual(metadata.version, PKG_INFO_PREFERRED_VERSION)
         metadata['Classifier'] = ['ok']
-        self.assertEquals(metadata.version, '1.2')
+        self.assertEqual(metadata.version, '1.2')
 
     def test_project_urls(self):
         # project-url is a bit specific, make sure we write it
@@ -214,7 +218,7 @@
         metadata = DistributionMetadata()
         metadata['Version'] = '1.0'
         metadata['Project-Url'] = [('one', 'http://ok')]
-        self.assertEquals(metadata['Project-Url'], [('one', 'http://ok')])
+        self.assertEqual(metadata['Project-Url'], [('one', 'http://ok')])
         file_ = StringIO()
         metadata.write_file(file_)
         file_.seek(0)
@@ -224,7 +228,7 @@
         file_.seek(0)
         metadata = DistributionMetadata()
         metadata.read_file(file_)
-        self.assertEquals(metadata['Project-Url'], [('one', 'http://ok')])
+        self.assertEqual(metadata['Project-Url'], [('one', 'http://ok')])
 
 
 def test_suite():
diff --git a/src/distutils2/tests/test_msvc9compiler.py b/src/distutils2/tests/test_msvc9compiler.py
--- a/src/distutils2/tests/test_msvc9compiler.py
+++ b/src/distutils2/tests/test_msvc9compiler.py
@@ -105,7 +105,7 @@
         import _winreg
         HKCU = _winreg.HKEY_CURRENT_USER
         keys = Reg.read_keys(HKCU, 'xxxx')
-        self.assertEquals(keys, None)
+        self.assertEqual(keys, None)
 
         keys = Reg.read_keys(HKCU, r'Control Panel')
         self.assertTrue('Desktop' in keys)
@@ -116,20 +116,24 @@
         tempdir = self.mkdtemp()
         manifest = os.path.join(tempdir, 'manifest')
         f = open(manifest, 'w')
-        f.write(_MANIFEST)
-        f.close()
+        try:
+            f.write(_MANIFEST)
+        finally:
+            f.close()
 
         compiler = MSVCCompiler()
         compiler._remove_visual_c_ref(manifest)
 
         # see what we got
         f = open(manifest)
-        # removing trailing spaces
-        content = '\n'.join([line.rstrip() for line in f.readlines()])
-        f.close()
+        try:
+            # removing trailing spaces
+            content = '\n'.join([line.rstrip() for line in f.readlines()])
+        finally:
+            f.close()
 
         # makes sure the manifest was properly cleaned
-        self.assertEquals(content, _CLEANED_MANIFEST)
+        self.assertEqual(content, _CLEANED_MANIFEST)
 
 
 def test_suite():
diff --git a/src/distutils2/tests/test_register.py b/src/distutils2/tests/test_register.py
--- a/src/distutils2/tests/test_register.py
+++ b/src/distutils2/tests/test_register.py
@@ -124,7 +124,7 @@
 
         # with the content similar to WANTED_PYPIRC
         content = open(self.rc).read()
-        self.assertEquals(content, WANTED_PYPIRC)
+        self.assertEqual(content, WANTED_PYPIRC)
 
         # now let's make sure the .pypirc file generated
         # really works : we shouldn't be asked anything
@@ -141,7 +141,7 @@
         self.assertTrue(self.conn.reqs, 2)
         req1 = dict(self.conn.reqs[0].headers)
         req2 = dict(self.conn.reqs[1].headers)
-        self.assertEquals(req2['Content-length'], req1['Content-length'])
+        self.assertEqual(req2['Content-length'], req1['Content-length'])
         self.assertTrue('xxx' in self.conn.reqs[1].data)
 
     def test_password_not_in_file(self):
@@ -154,7 +154,7 @@
 
         # dist.password should be set
         # therefore used afterwards by other commands
-        self.assertEquals(cmd.distribution.password, 'password')
+        self.assertEqual(cmd.distribution.password, 'password')
 
     def test_registering(self):
         # this test runs choice 2
@@ -171,7 +171,7 @@
         self.assertTrue(self.conn.reqs, 1)
         req = self.conn.reqs[0]
         headers = dict(req.headers)
-        self.assertEquals(headers['Content-length'], '608')
+        self.assertEqual(headers['Content-length'], '608')
         self.assertTrue('tarek' in req.data)
 
     def test_password_reset(self):
@@ -189,7 +189,7 @@
         self.assertTrue(self.conn.reqs, 1)
         req = self.conn.reqs[0]
         headers = dict(req.headers)
-        self.assertEquals(headers['Content-length'], '290')
+        self.assertEqual(headers['Content-length'], '290')
         self.assertTrue('tarek' in req.data)
 
     @unittest.skipUnless(DOCUTILS_SUPPORT, 'needs docutils')
@@ -246,8 +246,8 @@
         cmd.ensure_finalized()
         cmd.distribution.metadata['Requires-Dist'] = ['lxml']
         data = cmd.build_post_data('submit')
-        self.assertEquals(data['metadata_version'], '1.2')
-        self.assertEquals(data['requires_dist'], ['lxml'])
+        self.assertEqual(data['metadata_version'], '1.2')
+        self.assertEqual(data['requires_dist'], ['lxml'])
 
 def test_suite():
     return unittest.makeSuite(RegisterTestCase)
diff --git a/src/distutils2/tests/test_sdist.py b/src/distutils2/tests/test_sdist.py
--- a/src/distutils2/tests/test_sdist.py
+++ b/src/distutils2/tests/test_sdist.py
@@ -20,7 +20,6 @@
 
 from os.path import join
 import sys
-import tempfile
 import warnings
 
 from distutils2.tests import captured_stdout
@@ -128,7 +127,7 @@
         # now let's check what we have
         dist_folder = join(self.tmp_dir, 'dist')
         files = os.listdir(dist_folder)
-        self.assertEquals(files, ['fake-1.0.zip'])
+        self.assertEqual(files, ['fake-1.0.zip'])
 
         zip_file = zipfile.ZipFile(join(dist_folder, 'fake-1.0.zip'))
         try:
@@ -137,7 +136,7 @@
             zip_file.close()
 
         # making sure everything has been pruned correctly
-        self.assertEquals(len(content), 4)
+        self.assertEqual(len(content), 4)
 
     @unittest.skipUnless(zlib, "requires zlib")
     def test_make_distribution(self):
@@ -159,7 +158,7 @@
         dist_folder = join(self.tmp_dir, 'dist')
         result = os.listdir(dist_folder)
         result.sort()
-        self.assertEquals(result,
+        self.assertEqual(result,
                           ['fake-1.0.tar', 'fake-1.0.tar.gz'] )
 
         os.remove(join(dist_folder, 'fake-1.0.tar'))
@@ -173,7 +172,7 @@
 
         result = os.listdir(dist_folder)
         result.sort()
-        self.assertEquals(result,
+        self.assertEqual(result,
                 ['fake-1.0.tar', 'fake-1.0.tar.gz'])
 
     @unittest.skipUnless(zlib, "requires zlib")
@@ -223,7 +222,7 @@
         # now let's check what we have
         dist_folder = join(self.tmp_dir, 'dist')
         files = os.listdir(dist_folder)
-        self.assertEquals(files, ['fake-1.0.zip'])
+        self.assertEqual(files, ['fake-1.0.zip'])
 
         zip_file = zipfile.ZipFile(join(dist_folder, 'fake-1.0.zip'))
         try:
@@ -232,11 +231,11 @@
             zip_file.close()
 
         # making sure everything was added
-        self.assertEquals(len(content), 11)
+        self.assertEqual(len(content), 11)
 
         # checking the MANIFEST
         manifest = open(join(self.tmp_dir, 'MANIFEST')).read()
-        self.assertEquals(manifest, MANIFEST % {'sep': os.sep})
+        self.assertEqual(manifest, MANIFEST % {'sep': os.sep})
 
     @unittest.skipUnless(zlib, "requires zlib")
     def test_metadata_check_option(self):
@@ -248,7 +247,7 @@
         cmd.ensure_finalized()
         cmd.run()
         warnings = self.get_logs(WARN)
-        self.assertEquals(len(warnings), 1)
+        self.assertEqual(len(warnings), 1)
 
         # trying with a complete set of metadata
         self.clear_logs()
@@ -260,7 +259,7 @@
         # removing manifest generated warnings
         warnings = [warn for warn in warnings if
                     not warn.endswith('-- skipping')]
-        self.assertEquals(len(warnings), 0)
+        self.assertEqual(len(warnings), 0)
 
 
     def test_show_formats(self):
@@ -270,7 +269,7 @@
         num_formats = len(get_archive_formats())
         output = [line for line in stdout.split('\n')
                   if line.strip().startswith('--formats=')]
-        self.assertEquals(len(output), num_formats)
+        self.assertEqual(len(output), num_formats)
 
     def test_finalize_options(self):
 
@@ -278,9 +277,9 @@
         cmd.finalize_options()
 
         # default options set by finalize
-        self.assertEquals(cmd.manifest, 'MANIFEST')
-        self.assertEquals(cmd.template, 'MANIFEST.in')
-        self.assertEquals(cmd.dist_dir, 'dist')
+        self.assertEqual(cmd.manifest, 'MANIFEST')
+        self.assertEqual(cmd.template, 'MANIFEST.in')
+        self.assertEqual(cmd.dist_dir, 'dist')
 
         # formats has to be a string splitable on (' ', ',') or
         # a stringlist
@@ -317,8 +316,8 @@
         archive = tarfile.open(archive_name)
         try:
             for member in archive.getmembers():
-                self.assertEquals(member.uid, 0)
-                self.assertEquals(member.gid, 0)
+                self.assertEqual(member.uid, 0)
+                self.assertEqual(member.gid, 0)
         finally:
             archive.close()
 
@@ -339,7 +338,7 @@
         # rights (see #7408)
         try:
             for member in archive.getmembers():
-                self.assertEquals(member.uid, os.getuid())
+                self.assertEqual(member.uid, os.getuid())
         finally:
             archive.close()
 
diff --git a/src/distutils2/tests/test_spawn.py b/src/distutils2/tests/test_spawn.py
--- a/src/distutils2/tests/test_spawn.py
+++ b/src/distutils2/tests/test_spawn.py
@@ -20,7 +20,7 @@
                                (['nochange', 'nospace'],
                                 ['nochange', 'nospace'])):
             res = _nt_quote_args(args)
-            self.assertEquals(res, wanted)
+            self.assertEqual(res, wanted)
 
 
     @unittest.skipUnless(os.name in ('nt', 'posix'),
diff --git a/src/distutils2/tests/test_util.py b/src/distutils2/tests/test_util.py
--- a/src/distutils2/tests/test_util.py
+++ b/src/distutils2/tests/test_util.py
@@ -100,7 +100,7 @@
             return '/'.join(path)
         os.path.join = _join
 
-        self.assertEquals(convert_path('/home/to/my/stuff'),
+        self.assertEqual(convert_path('/home/to/my/stuff'),
                           '/home/to/my/stuff')
 
         # win
@@ -112,9 +112,9 @@
         self.assertRaises(ValueError, convert_path, '/home/to/my/stuff')
         self.assertRaises(ValueError, convert_path, 'home/to/my/stuff/')
 
-        self.assertEquals(convert_path('home/to/my/stuff'),
+        self.assertEqual(convert_path('home/to/my/stuff'),
                           'home\\to\\my\\stuff')
-        self.assertEquals(convert_path('.'),
+        self.assertEqual(convert_path('.'),
                           os.curdir)
 
     def test_change_root(self):
@@ -127,9 +127,9 @@
             return '/'.join(path)
         os.path.join = _join
 
-        self.assertEquals(change_root('/root', '/old/its/here'),
+        self.assertEqual(change_root('/root', '/old/its/here'),
                           '/root/old/its/here')
-        self.assertEquals(change_root('/root', 'its/here'),
+        self.assertEqual(change_root('/root', 'its/here'),
                           '/root/its/here')
 
         # windows
@@ -146,9 +146,9 @@
             return '\\'.join(path)
         os.path.join = _join
 
-        self.assertEquals(change_root('c:\\root', 'c:\\old\\its\\here'),
+        self.assertEqual(change_root('c:\\root', 'c:\\old\\its\\here'),
                           'c:\\root\\old\\its\\here')
-        self.assertEquals(change_root('c:\\root', 'its\\here'),
+        self.assertEqual(change_root('c:\\root', 'its\\here'),
                           'c:\\root\\its\\here')
 
         # BugsBunny os (it's a great os)
@@ -159,7 +159,7 @@
         # XXX platforms to be covered: os2, mac
 
     def test_split_quoted(self):
-        self.assertEquals(split_quoted('""one"" "two" \'three\' \\four'),
+        self.assertEqual(split_quoted('""one"" "two" \'three\' \\four'),
                           ['one', 'two', 'three', 'four'])
 
     def test_strtobool(self):
@@ -177,7 +177,7 @@
         res = rfc822_escape(header)
         wanted = ('I am a%(8s)spoor%(8s)slonesome%(8s)s'
                   'header%(8s)s') % {'8s': '\n'+8*' '}
-        self.assertEquals(res, wanted)
+        self.assertEqual(res, wanted)
 
     def test_find_exe_version(self):
         # the ld version scheme under MAC OS is:
@@ -195,7 +195,7 @@
                                 ('@(#)PROGRAM:ld  PROJECT:ld64-95.2.12',
                                  '95.2.12')):
             result = _MAC_OS_X_LD_VERSION.search(output)
-            self.assertEquals(result.group(1), version)
+            self.assertEqual(result.group(1), version)
 
     def _find_executable(self, name):
         if name in self._exes:
@@ -205,43 +205,43 @@
     def test_get_compiler_versions(self):
         # get_versions calls distutils.spawn.find_executable on
         # 'gcc', 'ld' and 'dllwrap'
-        self.assertEquals(get_compiler_versions(), (None, None, None))
+        self.assertEqual(get_compiler_versions(), (None, None, None))
 
         # Let's fake we have 'gcc' and it returns '3.4.5'
         self._exes['gcc'] = 'gcc (GCC) 3.4.5 (mingw special)\nFSF'
         res = get_compiler_versions()
-        self.assertEquals(str(res[0]), '3.4.5')
+        self.assertEqual(str(res[0]), '3.4.5')
 
         # and let's see what happens when the version
         # doesn't match the regular expression
         # (\d+\.\d+(\.\d+)*)
         self._exes['gcc'] = 'very strange output'
         res = get_compiler_versions()
-        self.assertEquals(res[0], None)
+        self.assertEqual(res[0], None)
 
         # same thing for ld
         if sys.platform != 'darwin':
             self._exes['ld'] = 'GNU ld version 2.17.50 20060824'
             res = get_compiler_versions()
-            self.assertEquals(str(res[1]), '2.17.50')
+            self.assertEqual(str(res[1]), '2.17.50')
             self._exes['ld'] = '@(#)PROGRAM:ld  PROJECT:ld64-77'
             res = get_compiler_versions()
-            self.assertEquals(res[1], None)
+            self.assertEqual(res[1], None)
         else:
             self._exes['ld'] = 'GNU ld version 2.17.50 20060824'
             res = get_compiler_versions()
-            self.assertEquals(res[1], None)
+            self.assertEqual(res[1], None)
             self._exes['ld'] = '@(#)PROGRAM:ld  PROJECT:ld64-77'
             res = get_compiler_versions()
-            self.assertEquals(str(res[1]), '77')
+            self.assertEqual(str(res[1]), '77')
 
         # and dllwrap
         self._exes['dllwrap'] = 'GNU dllwrap 2.17.50 20060824\nFSF'
         res = get_compiler_versions()
-        self.assertEquals(str(res[2]), '2.17.50')
+        self.assertEqual(str(res[2]), '2.17.50')
         self._exes['dllwrap'] = 'Cheese Wrap'
         res = get_compiler_versions()
-        self.assertEquals(res[2], None)
+        self.assertEqual(res[2], None)
 
     @unittest.skipUnless(hasattr(sys, 'dont_write_bytecode'),
                           'no dont_write_bytecode support')
@@ -294,7 +294,7 @@
         self.write_file(os.path.join(pkg5, '__init__.py'))
 
         res = find_packages([root], ['pkg1.pkg2'])
-        self.assertEquals(set(res), set(['pkg1', 'pkg5', 'pkg1.pkg3', 'pkg1.pkg3.pkg6']))
+        self.assertEqual(set(res), set(['pkg1', 'pkg5', 'pkg1.pkg3', 'pkg1.pkg3.pkg6']))
 
 
 def test_suite():
diff --git a/src/distutils2/tests/test_version.py b/src/distutils2/tests/test_version.py
--- a/src/distutils2/tests/test_version.py
+++ b/src/distutils2/tests/test_version.py
@@ -25,15 +25,15 @@
     def test_basic_versions(self):
 
         for v, s in self.versions:
-            self.assertEquals(str(v), s)
+            self.assertEqual(str(v), s)
 
     def test_from_parts(self):
 
         for v, s in self.versions:
             parts = v.parts
             v2 = V.from_parts(*v.parts)
-            self.assertEquals(v, v2)
-            self.assertEquals(str(v), str(v2))
+            self.assertEqual(v, v2)
+            self.assertEqual(str(v), str(v2))
 
     def test_irrational_versions(self):
 
@@ -96,34 +96,34 @@
 
     def test_suggest_normalized_version(self):
 
-        self.assertEquals(suggest('1.0'), '1.0')
-        self.assertEquals(suggest('1.0-alpha1'), '1.0a1')
-        self.assertEquals(suggest('1.0c2'), '1.0c2')
-        self.assertEquals(suggest('walla walla washington'), None)
-        self.assertEquals(suggest('2.4c1'), '2.4c1')
+        self.assertEqual(suggest('1.0'), '1.0')
+        self.assertEqual(suggest('1.0-alpha1'), '1.0a1')
+        self.assertEqual(suggest('1.0c2'), '1.0c2')
+        self.assertEqual(suggest('walla walla washington'), None)
+        self.assertEqual(suggest('2.4c1'), '2.4c1')
 
         # from setuptools
-        self.assertEquals(suggest('0.4a1.r10'), '0.4a1.post10')
-        self.assertEquals(suggest('0.7a1dev-r66608'), '0.7a1.dev66608')
-        self.assertEquals(suggest('0.6a9.dev-r41475'), '0.6a9.dev41475')
-        self.assertEquals(suggest('2.4preview1'), '2.4c1')
-        self.assertEquals(suggest('2.4pre1') , '2.4c1')
-        self.assertEquals(suggest('2.1-rc2'), '2.1c2')
+        self.assertEqual(suggest('0.4a1.r10'), '0.4a1.post10')
+        self.assertEqual(suggest('0.7a1dev-r66608'), '0.7a1.dev66608')
+        self.assertEqual(suggest('0.6a9.dev-r41475'), '0.6a9.dev41475')
+        self.assertEqual(suggest('2.4preview1'), '2.4c1')
+        self.assertEqual(suggest('2.4pre1') , '2.4c1')
+        self.assertEqual(suggest('2.1-rc2'), '2.1c2')
 
         # from pypi
-        self.assertEquals(suggest('0.1dev'), '0.1.dev0')
-        self.assertEquals(suggest('0.1.dev'), '0.1.dev0')
+        self.assertEqual(suggest('0.1dev'), '0.1.dev0')
+        self.assertEqual(suggest('0.1.dev'), '0.1.dev0')
 
         # we want to be able to parse Twisted
         # development versions are like post releases in Twisted
-        self.assertEquals(suggest('9.0.0+r2363'), '9.0.0.post2363')
+        self.assertEqual(suggest('9.0.0+r2363'), '9.0.0.post2363')
 
         # pre-releases are using markers like "pre1"
-        self.assertEquals(suggest('9.0.0pre1'), '9.0.0c1')
+        self.assertEqual(suggest('9.0.0pre1'), '9.0.0c1')
 
         # we want to be able to parse Tcl-TK
         # they us "p1" "p2" for post releases
-        self.assertEquals(suggest('1.4p1'), '1.4.post1')
+        self.assertEqual(suggest('1.4p1'), '1.4.post1')
 
     def test_predicate(self):
         # VersionPredicate knows how to parse stuff like:
diff --git a/src/distutils2/util.py b/src/distutils2/util.py
--- a/src/distutils2/util.py
+++ b/src/distutils2/util.py
@@ -1,12 +1,12 @@
 """distutils.util
 
-Miscellaneous utility functions -- anything that doesn't fit into
-one of the other *util.py modules.
+Miscellaneous utility functions.
 """
 
 __revision__ = "$Id: util.py 77761 2010-01-26 22:46:15Z tarek.ziade $"
 
 import sys, os, string, re
+from copy import copy
 from fnmatch import fnmatchcase
 
 from distutils2.errors import (DistutilsPlatformError, DistutilsFileError,
@@ -350,12 +350,8 @@
     # "Indirect" byte-compilation: write a temporary script and then
     # run it with the appropriate flags.
     if not direct:
-        try:
-            from tempfile import mkstemp
-            (script_fd, script_name) = mkstemp(".py")
-        except ImportError:
-            from tempfile import mktemp
-            (script_fd, script_name) = None, mktemp(".py")
+        from tempfile import mkstemp
+        script_fd, script_name = mkstemp(".py")
         log.info("writing byte-compilation script '%s'", script_name)
         if not dry_run:
             if script_fd is not None:
@@ -363,43 +359,50 @@
             else:
                 script = open(script_name, "w")
 
-            script.write("""\
-from distutils.util import byte_compile
+            try:
+                script.write("""\
+from distutils2.util import byte_compile
 files = [
 """)
 
-            # XXX would be nice to write absolute filenames, just for
-            # safety's sake (script should be more robust in the face of
-            # chdir'ing before running it).  But this requires abspath'ing
-            # 'prefix' as well, and that breaks the hack in build_lib's
-            # 'byte_compile()' method that carefully tacks on a trailing
-            # slash (os.sep really) to make sure the prefix here is "just
-            # right".  This whole prefix business is rather delicate -- the
-            # problem is that it's really a directory, but I'm treating it
-            # as a dumb string, so trailing slashes and so forth matter.
+                # XXX would be nice to write absolute filenames, just for
+                # safety's sake (script should be more robust in the face of
+                # chdir'ing before running it).  But this requires abspath'ing
+                # 'prefix' as well, and that breaks the hack in build_lib's
+                # 'byte_compile()' method that carefully tacks on a trailing
+                # slash (os.sep really) to make sure the prefix here is "just
+                # right".  This whole prefix business is rather delicate -- the
+                # problem is that it's really a directory, but I'm treating it
+                # as a dumb string, so trailing slashes and so forth matter.
 
-            #py_files = map(os.path.abspath, py_files)
-            #if prefix:
-            #    prefix = os.path.abspath(prefix)
+                #py_files = map(os.path.abspath, py_files)
+                #if prefix:
+                #    prefix = os.path.abspath(prefix)
 
-            script.write(",\n".join(map(repr, py_files)) + "]\n")
-            script.write("""
+                script.write(",\n".join(map(repr, py_files)) + "]\n")
+                script.write("""
 byte_compile(files, optimize=%r, force=%r,
              prefix=%r, base_dir=%r,
              verbose=%r, dry_run=0,
              direct=1)
 """ % (optimize, force, prefix, base_dir, verbose))
 
-            script.close()
+            finally:
+                script.close()
 
         cmd = [sys.executable, script_name]
         if optimize == 1:
             cmd.insert(1, "-O")
         elif optimize == 2:
             cmd.insert(1, "-OO")
-        spawn(cmd, dry_run=dry_run)
-        execute(os.remove, (script_name,), "removing %s" % script_name,
-                dry_run=dry_run)
+
+        env = copy(os.environ)
+        env['PYTHONPATH'] = ':'.join(sys.path)
+        try:
+            spawn(cmd, dry_run=dry_run, env=env)
+        finally:
+            execute(os.remove, (script_name,), "removing %s" % script_name,
+                    dry_run=dry_run)
 
     # "Direct" byte-compilation: use the py_compile module to compile
     # right here, right now.  Note that the script generated in indirect
@@ -539,10 +542,12 @@
     """Create a file with the specified name and write 'contents' (a
     sequence of strings without line terminators) to it.
     """
-    f = open(filename, "w")
-    for line in contents:
-        f.write(line + "\n")
-    f.close()
+    try:
+        f = open(filename, "w")
+        for line in contents:
+            f.write(line + "\n")
+    finally:
+        f.close()
 
 def _is_package(path):
     """Returns True if path is a package (a dir with an __init__ file."""
diff --git a/src/runtests.py b/src/runtests.py
--- a/src/runtests.py
+++ b/src/runtests.py
@@ -1,6 +1,6 @@
 """Tests for distutils2.
 
-The tests for distutils2 are defined in the distutils2.tests package;
+The tests for distutils2 are defined in the distutils2.tests package.
 """
 import sys
 
@@ -8,8 +8,7 @@
     import distutils2.tests
     from distutils2.tests import run_unittest, reap_children, TestFailed
     from distutils2._backport.tests import test_suite as btest_suite
-    # just supporting -q right now
-    # to enable detailed/quiet output
+    # XXX just supporting -q right now to enable detailed/quiet output
     if len(sys.argv) > 1:
         verbose = sys.argv[-1] != '-q'
     else:
@@ -17,7 +16,7 @@
     try:
         try:
             run_unittest([distutils2.tests.test_suite(), btest_suite()],
-                    verbose_=verbose)
+                         verbose_=verbose)
             return 0
         except TestFailed:
             return 1
@@ -28,7 +27,7 @@
     try:
         from distutils2.tests.support import unittest
     except ImportError:
-        print('Error: You have to install unittest2')
+        sys.stderr.write('Error: You have to install unittest2')
         sys.exit(1)
 
     sys.exit(test_main())
diff --git a/src/setup.py b/src/setup.py
--- a/src/setup.py
+++ b/src/setup.py
@@ -11,6 +11,7 @@
 from distutils2.compiler.ccompiler import new_compiler
 from distutils2.command.sdist import sdist
 from distutils2.command.install import install
+from distutils2 import __version__ as VERSION
 from distutils2.util import find_packages
 
 f = open('README.txt')
diff --git a/src/tests.sh b/src/tests.sh
--- a/src/tests.sh
+++ b/src/tests.sh
@@ -1,28 +1,36 @@
 #!/bin/sh
-echo -n "Running tests for Python 2.4..."
-python2.4 runtests.py -q > /dev/null 2> /dev/null
+echo -n "Running tests for Python 2.4... "
+python2.4 -Wd runtests.py -q 2> /dev/null
 if [ $? -ne 0 ];then
     echo "Failed"
-    exit $1
+    exit 1
 else
     echo "Success"
 fi
 
-echo -n "Running tests for Python 2.5..."
-python2.5 runtests.py -q > /dev/null 2> /dev/null
+echo -n "Running tests for Python 2.5... "
+python2.5 -Wd runtests.py -q 2> /dev/null
 if [ $? -ne 0 ];then
     echo "Failed"
-    exit $1
+    exit 1
 else
     echo "Success"
 fi
 
-echo -n "Running tests for Python 2.6..."
-python2.6 runtests.py -q > /dev/null 2> /dev/null
+echo -n "Running tests for Python 2.6... "
+python2.6 -Wd -bb -3 runtests.py -q 2> /dev/null
 if [ $? -ne 0 ];then
     echo "Failed"
-    exit $1
+    exit 1
 else
     echo "Success"
 fi
 
+echo -n "Running tests for Python 2.7... "
+python2.7 -Wd -bb -3 runtests.py -q 2> /dev/null
+if [ $? -ne 0 ];then
+    echo "Failed"
+    exit 1
+else
+    echo "Success"
+fi

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 11:48:41 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 04 Jul 2010 11:48:41 +0200
Subject: [Python-checkins] distutils2: merged Alexis/Konrad branches
Message-ID: <E1OVLoX-0000lw-1a@dinsdale.python.org>

tarek.ziade pushed c9708f8ad60c to distutils2:

http://hg.python.org/distutils2/rev/c9708f8ad60c
changeset:   348:c9708f8ad60c
parent:      256:f696d39e9e86
parent:      347:a569f25a2038
user:        Tarek Ziade <tarek at ziade.org>
date:        Sun Jul 04 11:34:47 2010 +0200
summary:     merged Alexis/Konrad branches
files:       src/distutils2/dist.py

diff --git a/docs/design/pep-0376.txt b/docs/design/pep-0376.txt
--- a/docs/design/pep-0376.txt
+++ b/docs/design/pep-0376.txt
@@ -633,7 +633,7 @@
 
 Distributions installed using existing, pre-standardization formats do not have
 the necessary metadata available for the new API, and thus will be
-ignored. Third-party tools may of course to continue to support previous
+ignored. Third-party tools may of course continue to support previous
 formats in addition to the new format, in order to ease the transition.
 
 
diff --git a/docs/design/wiki.rst b/docs/design/wiki.rst
--- a/docs/design/wiki.rst
+++ b/docs/design/wiki.rst
@@ -282,7 +282,7 @@
   mailman/etc/*               = {config}                # 8
   mailman/foo/**/bar/*.cfg    = {config}/baz            # 9
   mailman/foo/**/*.cfg        = {config}/hmm            # 9, 10
-  some-new-semantic.txt       = {funky-crazy-category}  # 11
+  some-new-semantic.sns       = {funky-crazy-category}  # 11
 
 The glob definitions are relative paths that match files from the top
 of the source tree (the location of ``setup.cfg``). Forward slashes
diff --git a/docs/source/index.rst b/docs/source/index.rst
--- a/docs/source/index.rst
+++ b/docs/source/index.rst
@@ -14,6 +14,10 @@
    metadata
    pkgutil
    depgraph
+   new_commands
+   test_framework
+   pypi
+   version
 
 Indices and tables
 ==================
diff --git a/docs/source/metadata.rst b/docs/source/metadata.rst
--- a/docs/source/metadata.rst
+++ b/docs/source/metadata.rst
@@ -17,7 +17,7 @@
 Reading metadata
 ================
 
-The :class:`DistributionMetadata` class can be instantiated with the path of
+The :class:`DistributionMetadata` class can be instanciated with the path of
 the metadata file, and provides a dict-like interface to the values::
 
     >>> from distutils2.metadata import DistributionMetadata
@@ -32,14 +32,14 @@
     ["pywin32; sys.platform == 'win32'", "Sphinx"]
 
 The fields that supports environment markers can be automatically ignored if
-the object is instantiated using the ``platform_dependent`` option.
+the object is instanciated using the ``platform_dependant`` option.
 :class:`DistributionMetadata` will interpret in the case the markers and will
 automatically remove the fields that are not compliant with the running
 environment. Here's an example under Mac OS X. The win32 dependency
 we saw earlier is ignored::
 
     >>> from distutils2.metadata import DistributionMetadata
-    >>> metadata = DistributionMetadata('PKG-INFO', platform_dependent=True)
+    >>> metadata = DistributionMetadata('PKG-INFO', platform_dependant=True)
     >>> metadata['Requires-Dist']
     ['bar']
 
@@ -53,7 +53,7 @@
 
     >>> from distutils2.metadata import DistributionMetadata
     >>> context = {'sys.platform': 'win32'}
-    >>> metadata = DistributionMetadata('PKG-INFO', platform_dependent=True,
+    >>> metadata = DistributionMetadata('PKG-INFO', platform_dependant=True,
     ...                                 execution_context=context)
     ...
     >>> metadata['Requires-Dist'] = ["pywin32; sys.platform == 'win32'",
@@ -71,7 +71,7 @@
     >>> metadata.write('/to/my/PKG-INFO')
 
 The class will pick the best version for the metadata, depending on the values
-provided. If all the values provided exist in all versions, the class will
+provided. If all the values provided exists in all versions, the class will
 use :attr:`metadata.PKG_INFO_PREFERRED_VERSION`. It is set by default to 1.0.
 
 
@@ -79,7 +79,7 @@
 ==================================
 
 Some fields in :pep:`345` have to follow a version scheme in their versions
-predicate. When the scheme is violated, a warning is emitted::
+predicate. When the scheme is violated, a warning is emited::
 
     >>> from distutils2.metadata import DistributionMetadata
     >>> metadata = DistributionMetadata()
@@ -90,3 +90,6 @@
 
 
 .. TODO talk about check()
+
+
+
diff --git a/docs/source/new_commands.rst b/docs/source/new_commands.rst
new file mode 100644
--- /dev/null
+++ b/docs/source/new_commands.rst
@@ -0,0 +1,65 @@
+========
+Commands
+========
+
+Distutils2 provides a set of commands that are not present in distutils itself.
+You might recognize some of them from other projects, like Distribute or
+Setuptools.
+
+``upload_docs`` - Upload package documentation to PyPI
+======================================================
+
+PyPI now supports uploading project documentation to the dedicated URL
+http://packages.python.org/<project>/.
+
+The ``upload_docs`` command will create the necessary zip file out of a
+documentation directory and will post to the repository.
+
+Note that to upload the documentation of a project, the corresponding version
+must already be registered with PyPI, using the distutils ``register``
+command -- just like the ``upload`` command.
+
+Assuming there is an ``Example`` project with documentation in the
+subdirectory ``docs``, e.g.::
+
+  Example/
+  |-- example.py
+  |-- setup.cfg
+  |-- setup.py
+  |-- docs
+  |   |-- build
+  |   |   `-- html
+  |   |   |   |-- index.html
+  |   |   |   `-- tips_tricks.html
+  |   |-- conf.py
+  |   |-- index.txt
+  |   `-- tips_tricks.txt
+
+You can simply pass the documentation directory path to the ``upload_docs``
+command::
+
+    python setup.py upload_docs --upload-dir=docs/build/html
+
+As with any other ``setuptools`` based command, you can define useful
+defaults in the ``setup.cfg`` of your Python project, e.g.:
+
+.. code-block:: ini
+
+    [upload_docs]
+    upload-dir = docs/build/html
+
+The ``upload_docs`` command has the following options:
+
+``--upload-dir``
+    The directory to be uploaded to the repository. The default value is
+    ``docs`` in project root.
+
+``--show-response``
+    Display the full response text from server; this is useful for debugging
+    PyPI problems.
+
+``--repository=URL, -r URL``
+    The URL of the repository to upload to.  Defaults to
+    http://pypi.python.org/pypi (i.e., the main PyPI installation).
+
+
diff --git a/docs/source/pypi.rst b/docs/source/pypi.rst
new file mode 100644
--- /dev/null
+++ b/docs/source/pypi.rst
@@ -0,0 +1,195 @@
+=========================================
+Tools to query PyPI: the PyPI package
+=========================================
+
+Distutils2 comes with a module (eg. `distutils2.pypi`) which contains
+facilities to access the Python Package Index (named "pypi", and avalaible on
+the url `http://pypi.python.org`.
+
+There is two ways to retrieve data from pypi: using the *simple* API, and using
+*XML-RPC*. The first one is in fact a set of HTML pages avalaible at
+`http://pypi.python.org/simple/`, and the second one contains a set of XML-RPC
+methods. In order to reduce the overload caused by running distant methods on 
+the pypi server (by using the XML-RPC methods), the best way to retrieve 
+informations is by using the simple API, when it contains the information you 
+need.
+
+Distutils2 provides two python modules to ease the work with those two APIs:
+`distutils2.pypi.simple` and `distutils2.pypi.xmlrpc`. Both of them depends on 
+another python module: `distutils2.pypi.dist`.
+
+
+Requesting information via the "simple" API `distutils2.pypi.simple`
+====================================================================
+
+`distutils2.pypi.simple` can process the Python Package Index and return and 
+download urls of distributions, for specific versions or latests, but it also 
+can process external html pages, with the goal to find *pypi unhosted* versions 
+of python distributions.
+
+You should use `distutils2.pypi.simple` for: 
+
+    * Search distributions by name and versions.
+    * Process pypi external pages.
+    * Download distributions by name and versions.
+
+And should not be used to:
+
+    * Things that will end up in too long index processing (like "finding all
+      distributions with a specific version, no matters the name")
+
+API
+----
+
+Here is a complete overview of the APIs of the SimpleIndex class.
+
+.. autoclass:: distutils2.pypi.simple.SimpleIndex
+    :members:
+
+Usage Exemples
+---------------
+
+To help you understand how using the `SimpleIndex` class, here are some basic
+usages.
+
+Request PyPI to get a specific distribution
+++++++++++++++++++++++++++++++++++++++++++++
+
+Supposing you want to scan the PyPI index to get a list of distributions for 
+the "foobar" project. You can use the "find" method for that::
+
+    >>> from distutils2.pypi import SimpleIndex
+    >>> client = SimpleIndex()
+    >>> client.find("foobar")
+    [<PyPIDistribution "Foobar 1.1">, <PyPIDistribution "Foobar 1.2">]
+    
+Note that you also can request the client about specific versions, using version
+specifiers (described in `PEP 345 
+<http://www.python.org/dev/peps/pep-0345/#version-specifiers>`_)::
+
+    >>> client.find("foobar < 1.2")
+    [<PyPIDistribution "foobar 1.1">, ]
+
+`find` returns a list of distributions, but you also can get the last
+distribution (the more up to date) that fullfil your requirements, like this::
+    
+    >>> client.get("foobar < 1.2")
+    <PyPIDistribution "foobar 1.1">
+
+Download distributions
++++++++++++++++++++++++
+
+As it can get the urls of distributions provided by PyPI, the `SimpleIndex` 
+client also can download the distributions and put it for you in a temporary
+destination::
+
+    >>> client.download("foobar")
+    /tmp/temp_dir/foobar-1.2.tar.gz
+
+You also can specify the directory you want to download to::
+    
+    >>> client.download("foobar", "/path/to/my/dir")
+    /path/to/my/dir/foobar-1.2.tar.gz
+
+While downloading, the md5 of the archive will be checked, if not matches, it
+will try another time, then if fails again, raise `MD5HashDoesNotMatchError`.
+
+Internally, that's not the SimpleIndex which download the distributions, but the
+`PyPIDistribution` class. Please refer to this documentation for more details.
+
+Following PyPI external links
+++++++++++++++++++++++++++++++
+
+The default behavior for distutils2 is to *not* follow the links provided
+by HTML pages in the "simple index", to find distributions related
+downloads.
+
+It's possible to tell the PyPIClient to follow external links by setting the 
+`follow_externals` attribute, on instanciation or after::
+
+    >>> client = SimpleIndex(follow_externals=True)
+
+or ::
+
+    >>> client = SimpleIndex()
+    >>> client.follow_externals = True
+
+Working with external indexes, and mirrors
++++++++++++++++++++++++++++++++++++++++++++
+
+The default `SimpleIndex` behavior is to rely on the Python Package index stored
+on PyPI (http://pypi.python.org/simple).
+
+As you can need to work with a local index, or private indexes, you can specify
+it using the index_url parameter::
+
+    >>> client = SimpleIndex(index_url="file://filesystem/path/")
+
+or ::
+
+    >>> client = SimpleIndex(index_url="http://some.specific.url/")
+
+You also can specify mirrors to fallback on in case the first index_url you
+provided doesnt respond, or not correctly. The default behavior for
+`SimpleIndex` is to use the list provided by Python.org DNS records, as
+described in the :pep:`381` about mirroring infrastructure.
+
+If you don't want to rely on these, you could specify the list of mirrors you
+want to try by specifying the `mirrors` attribute. It's a simple iterable::
+
+    >>> mirrors = ["http://first.mirror","http://second.mirror"]
+    >>> client = SimpleIndex(mirrors=mirrors)
+
+
+Requesting informations via XML-RPC (`distutils2.pypi.XmlRpcIndex`)
+==========================================================================
+
+The other method to request the Python package index, is using the XML-RPC
+methods. Distutils2 provides a simple wrapper around `xmlrpclib
+<http://docs.python.org/library/xmlrpclib.html>`_, that can return you
+`PyPIDistribution` objects.
+
+::
+    >>> from distutils2.pypi import XmlRpcIndex()
+    >>> client = XmlRpcIndex()
+
+
+PyPI Distributions
+==================
+
+Both `SimpleIndex` and `XmlRpcIndex` classes works with the classes provided
+in the `pypi.dist` package.
+
+`PyPIDistribution`
+------------------
+
+`PyPIDistribution` is a simple class that defines the following attributes:
+
+:name:
+    The name of the package. `foobar` in our exemples here
+:version:
+    The version of the package
+:location:
+    If the files from the archive has been downloaded, here is the path where
+    you can find them.
+:url:
+    The url of the distribution
+
+.. autoclass:: distutils2.pypi.dist.PyPIDistribution
+    :members:
+
+`PyPIDistributions`
+-------------------
+
+The `dist` module also provides another class, to work with lists of 
+`PyPIDistribution` classes. It allow to filter results and is used as a 
+container of 
+
+.. autoclass:: distutils2.pypi.dist.PyPIDistributions
+    :members:
+
+At a higher level
+=================
+
+XXX : A description about a wraper around PyPI simple and XmlRpc Indexes
+(PyPIIndex ?) 
diff --git a/docs/source/test_framework.rst b/docs/source/test_framework.rst
new file mode 100644
--- /dev/null
+++ b/docs/source/test_framework.rst
@@ -0,0 +1,88 @@
+==============
+Test Framework
+==============
+
+When you are testing code that works with distutils, you might find these tools
+useful.
+
+``PyPIServer``
+==============
+
+PyPIServer is a class that implements an HTTP server running in a separate
+thread. All it does is record the requests for further inspection. The recorded
+data is available under ``requests`` attribute. The default
+HTTP response can be overriden with the ``default_response_status``,
+``default_response_headers`` and ``default_response_data`` attributes.
+
+By default, when accessing the server with urls beginning with `/simple/`, 
+the server also record your requests, but will look for files under 
+the `/tests/pypiserver/simple/` path.
+
+You can tell the sever to serve static files for other paths. This could be 
+accomplished by using the `static_uri_paths` parameter, as below::
+
+    server = PyPIServer(static_uri_paths=["first_path", "second_path"])
+
+You need to create the content that will be served under the 
+`/tests/pypiserver/default` path. If you want to serve content from another 
+place, you also can specify another filesystem path (wich need to be under
+`tests/pypiserver/`. This will replace the default behavior of the server, and
+it will not serve content from the `default` dir ::
+
+    server = PyPIServer(static_filesystem_paths=["path/to/your/dir"])
+
+If you just need to add some paths to the existing ones, you can do as shown, 
+keeping in mind that the server will alwas try to load paths in reverse order 
+(e.g here, try "another/super/path" then the default one) ::
+
+    server = PyPIServer(test_static_path="another/super/path")
+    server = PyPIServer("another/super/path")
+    # or 
+    server.static_filesystem_paths.append("another/super/path")
+
+As a result of what, in your tests, while you need to use the PyPIServer, in
+order to isolates the test cases, the best practice is to place the common files
+in the `default` folder, and to create a directory for each specific test case::
+
+    server = PyPIServer(static_filesystem_paths = ["default", "test_pypi_server"],
+        static_uri_paths=["simple", "external"])
+
+``PyPIServerTestCase``
+======================
+
+``PyPIServerTestCase`` is a test case class with setUp and tearDown methods that
+take care of a single PyPIServer instance attached as a ``pypi`` attribute on
+the test class. Use it as one of the base classes in your test case::
+
+  class UploadTestCase(PyPIServerTestCase):
+      def test_something(self):
+          cmd = self.prepare_command()
+          cmd.ensure_finalized()
+          cmd.repository = self.pypi.full_address
+          cmd.run()
+
+          environ, request_data = self.pypi.requests[-1]
+          self.assertEqual(request_data, EXPECTED_REQUEST_DATA)
+
+The ``use_pypi_server`` decorator
+=================================
+
+You also can use a decorator for your tests, if you do not need the same server
+instance along all you test case. So, you can specify, for each test method,
+some initialisation parameters for the server.
+
+For this, you need to add a `server` parameter to your method, like this::
+
+    class SampleTestCase(TestCase):
+        @use_pypi_server()
+        def test_somthing(self, server):
+            # your tests goes here
+
+The decorator will instanciate the server for you, and run and stop it just
+before and after your method call. You also can pass the server initializer,
+just like this::
+
+    class SampleTestCase(TestCase):
+        @use_pypi_server("test_case_name")
+        def test_something(self, server):
+            # something
diff --git a/docs/source/version.rst b/docs/source/version.rst
new file mode 100644
--- /dev/null
+++ b/docs/source/version.rst
@@ -0,0 +1,64 @@
+======================
+Working with versions
+======================
+
+Distutils2 ships with a python package capable to work with version numbers.
+It's an implementation of version specifiers `as defined in PEP 345
+<http://www.python.org/dev/peps/pep-0345/#version-specifiers>`_ about
+Metadata.
+
+`distutils2.version.NormalizedVersion`
+======================================
+
+A Normalized version corresponds to a specific version of a distribution, as
+described in the PEP 345. So, you can work with the `NormalizedVersion` like
+this::
+
+    >>> NormalizedVersion("1.2b1")
+    NormalizedVersion('1.2b1')
+
+If you try to use irrational version specifiers, an `IrrationalVersionError`
+will be raised::
+
+    >>> NormalizedVersion("irrational_version_number")
+    ...
+    IrrationalVersionError: irrational_version_number
+
+You can compare NormalizedVersion objects, like this::
+
+    >>> NormalizedVersion("1.2b1") < NormalizedVersion("1.2")
+    True
+
+NormalizedVersion is used internally by `VersionPredicate` to do his stuff.
+
+`distutils2.version.suggest_normalized_version`
+-----------------------------------------------
+
+You also can let the normalized version be suggested to you, using the
+`suggest_normalized_version` function::
+
+    >>> suggest_normalized_version('2.1-rc1') 
+    2.1c1
+
+If `suggest_normalized_version` can't actually suggest you a version, it will
+return `None`::
+
+    >>> print suggest_normalized_version('not a version')
+    None
+
+`distutils2.version.VersionPredicate`
+=====================================
+
+`VersionPredicate` knows how to parse stuff like "ProjectName (>=version)", the
+class also provides a `match` method to test if a version number is the version
+predicate::
+
+    >>> version = VersionPredicate("ProjectName (<1.2,>1.0")
+    >>> version.match("1.2.1")
+    False
+    >>> version.match("1.1.1")
+    True
+
+`is_valid_predicate`
+--------------------
+
diff --git a/src/CONTRIBUTORS.txt b/src/CONTRIBUTORS.txt
--- a/src/CONTRIBUTORS.txt
+++ b/src/CONTRIBUTORS.txt
@@ -18,6 +18,7 @@
 - Jeremy Kloth
 - Martin von L??wis
 - Carl Meyer
+- Alexis Metaireau
 - Michael Mulich
 - George Peris
 - Sean Reifschneider
diff --git a/src/Modules/_hashopenssl.c b/src/Modules/_hashopenssl.c
new file mode 100644
--- /dev/null
+++ b/src/Modules/_hashopenssl.c
@@ -0,0 +1,524 @@
+/* Module that wraps all OpenSSL hash algorithms */
+
+/*
+ * Copyright (C) 2005   Gregory P. Smith (greg at krypto.org)
+ * Licensed to PSF under a Contributor Agreement.
+ *
+ * Derived from a skeleton of shamodule.c containing work performed by:
+ *
+ * Andrew Kuchling (amk at amk.ca)
+ * Greg Stein (gstein at lyra.org)
+ *
+ */
+
+#define PY_SSIZE_T_CLEAN
+
+#include "Python.h"
+#include "structmember.h"
+
+#if (PY_VERSION_HEX < 0x02050000)
+#define Py_ssize_t      int
+#endif
+
+/* EVP is the preferred interface to hashing in OpenSSL */
+#include <openssl/evp.h>
+
+#define MUNCH_SIZE INT_MAX
+
+
+#ifndef HASH_OBJ_CONSTRUCTOR
+#define HASH_OBJ_CONSTRUCTOR 0
+#endif
+
+typedef struct {
+    PyObject_HEAD
+    PyObject            *name;  /* name of this hash algorithm */
+    EVP_MD_CTX          ctx;    /* OpenSSL message digest context */
+} EVPobject;
+
+
+static PyTypeObject EVPtype;
+
+
+#define DEFINE_CONSTS_FOR_NEW(Name)  \
+    static PyObject *CONST_ ## Name ## _name_obj; \
+    static EVP_MD_CTX CONST_new_ ## Name ## _ctx; \
+    static EVP_MD_CTX *CONST_new_ ## Name ## _ctx_p = NULL;
+
+DEFINE_CONSTS_FOR_NEW(md5)
+DEFINE_CONSTS_FOR_NEW(sha1)
+DEFINE_CONSTS_FOR_NEW(sha224)
+DEFINE_CONSTS_FOR_NEW(sha256)
+DEFINE_CONSTS_FOR_NEW(sha384)
+DEFINE_CONSTS_FOR_NEW(sha512)
+
+
+static EVPobject *
+newEVPobject(PyObject *name)
+{
+    EVPobject *retval = (EVPobject *)PyObject_New(EVPobject, &EVPtype);
+
+    /* save the name for .name to return */
+    if (retval != NULL) {
+        Py_INCREF(name);
+        retval->name = name;
+    }
+
+    return retval;
+}
+
+/* Internal methods for a hash object */
+
+static void
+EVP_dealloc(PyObject *ptr)
+{
+    EVP_MD_CTX_cleanup(&((EVPobject *)ptr)->ctx);
+    Py_XDECREF(((EVPobject *)ptr)->name);
+    PyObject_Del(ptr);
+}
+
+
+/* External methods for a hash object */
+
+PyDoc_STRVAR(EVP_copy__doc__, "Return a copy of the hash object.");
+
+static PyObject *
+EVP_copy(EVPobject *self, PyObject *unused)
+{
+    EVPobject *newobj;
+
+    if ( (newobj = newEVPobject(self->name))==NULL)
+        return NULL;
+
+    EVP_MD_CTX_copy(&newobj->ctx, &self->ctx);
+    return (PyObject *)newobj;
+}
+
+PyDoc_STRVAR(EVP_digest__doc__,
+"Return the digest value as a string of binary data.");
+
+static PyObject *
+EVP_digest(EVPobject *self, PyObject *unused)
+{
+    unsigned char digest[EVP_MAX_MD_SIZE];
+    EVP_MD_CTX temp_ctx;
+    PyObject *retval;
+    unsigned int digest_size;
+
+    EVP_MD_CTX_copy(&temp_ctx, &self->ctx);
+    digest_size = EVP_MD_CTX_size(&temp_ctx);
+    EVP_DigestFinal(&temp_ctx, digest, NULL);
+
+    retval = PyString_FromStringAndSize((const char *)digest, digest_size);
+    EVP_MD_CTX_cleanup(&temp_ctx);
+    return retval;
+}
+
+PyDoc_STRVAR(EVP_hexdigest__doc__,
+"Return the digest value as a string of hexadecimal digits.");
+
+static PyObject *
+EVP_hexdigest(EVPobject *self, PyObject *unused)
+{
+    unsigned char digest[EVP_MAX_MD_SIZE];
+    EVP_MD_CTX temp_ctx;
+    PyObject *retval;
+    char *hex_digest;
+    unsigned int i, j, digest_size;
+
+    /* Get the raw (binary) digest value */
+    EVP_MD_CTX_copy(&temp_ctx, &self->ctx);
+    digest_size = EVP_MD_CTX_size(&temp_ctx);
+    EVP_DigestFinal(&temp_ctx, digest, NULL);
+
+    EVP_MD_CTX_cleanup(&temp_ctx);
+
+    /* Create a new string */
+    /* NOTE: not thread safe! modifying an already created string object */
+    /* (not a problem because we hold the GIL by default) */
+    retval = PyString_FromStringAndSize(NULL, digest_size * 2);
+    if (!retval)
+	    return NULL;
+    hex_digest = PyString_AsString(retval);
+    if (!hex_digest) {
+	    Py_DECREF(retval);
+	    return NULL;
+    }
+
+    /* Make hex version of the digest */
+    for(i=j=0; i<digest_size; i++) {
+        char c;
+        c = (digest[i] >> 4) & 0xf;
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+        c = (digest[i] & 0xf);
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+    }
+    return retval;
+}
+
+PyDoc_STRVAR(EVP_update__doc__,
+"Update this hash object's state with the provided string.");
+
+static PyObject *
+EVP_update(EVPobject *self, PyObject *args)
+{
+    unsigned char *cp;
+    Py_ssize_t len;
+
+    if (!PyArg_ParseTuple(args, "s#:update", &cp, &len))
+        return NULL;
+
+    if (len > 0 && len <= MUNCH_SIZE) {
+    EVP_DigestUpdate(&self->ctx, cp, Py_SAFE_DOWNCAST(len, Py_ssize_t,
+                                                      unsigned int));
+    } else {
+        Py_ssize_t offset = 0;
+        while (len) {
+            unsigned int process = len > MUNCH_SIZE ? MUNCH_SIZE : len;
+            EVP_DigestUpdate(&self->ctx, cp + offset, process);
+            len -= process;
+            offset += process;
+        }
+    }
+    Py_INCREF(Py_None);
+    return Py_None;
+}
+
+static PyMethodDef EVP_methods[] = {
+    {"update",	  (PyCFunction)EVP_update,    METH_VARARGS, EVP_update__doc__},
+    {"digest",	  (PyCFunction)EVP_digest,    METH_NOARGS,  EVP_digest__doc__},
+    {"hexdigest", (PyCFunction)EVP_hexdigest, METH_NOARGS,  EVP_hexdigest__doc__},
+    {"copy",	  (PyCFunction)EVP_copy,      METH_NOARGS,  EVP_copy__doc__},
+    {NULL,	  NULL}		/* sentinel */
+};
+
+static PyObject *
+EVP_get_block_size(EVPobject *self, void *closure)
+{
+    return PyInt_FromLong(EVP_MD_CTX_block_size(&((EVPobject *)self)->ctx));
+}
+
+static PyObject *
+EVP_get_digest_size(EVPobject *self, void *closure)
+{
+    return PyInt_FromLong(EVP_MD_CTX_size(&((EVPobject *)self)->ctx));
+}
+
+static PyMemberDef EVP_members[] = {
+    {"name", T_OBJECT, offsetof(EVPobject, name), READONLY, PyDoc_STR("algorithm name.")},
+    {NULL}  /* Sentinel */
+};
+
+static PyGetSetDef EVP_getseters[] = {
+    {"digest_size",
+     (getter)EVP_get_digest_size, NULL,
+     NULL,
+     NULL},
+    {"block_size",
+     (getter)EVP_get_block_size, NULL,
+     NULL,
+     NULL},
+    /* the old md5 and sha modules support 'digest_size' as in PEP 247.
+     * the old sha module also supported 'digestsize'.  ugh. */
+    {"digestsize",
+     (getter)EVP_get_digest_size, NULL,
+     NULL,
+     NULL},
+    {NULL}  /* Sentinel */
+};
+
+
+static PyObject *
+EVP_repr(PyObject *self)
+{
+    char buf[100];
+    PyOS_snprintf(buf, sizeof(buf), "<%s HASH object @ %p>",
+            PyString_AsString(((EVPobject *)self)->name), self);
+    return PyString_FromString(buf);
+}
+
+#if HASH_OBJ_CONSTRUCTOR
+static int
+EVP_tp_init(EVPobject *self, PyObject *args, PyObject *kwds)
+{
+    static char *kwlist[] = {"name", "string", NULL};
+    PyObject *name_obj = NULL;
+    char *nameStr;
+    unsigned char *cp = NULL;
+    Py_ssize_t len = 0;
+    const EVP_MD *digest;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwds, "O|s#:HASH", kwlist,
+                                     &name_obj, &cp, &len)) {
+        return -1;
+    }
+
+    if (!PyArg_Parse(name_obj, "s", &nameStr)) {
+        PyErr_SetString(PyExc_TypeError, "name must be a string");
+        return -1;
+    }
+
+    digest = EVP_get_digestbyname(nameStr);
+    if (!digest) {
+        PyErr_SetString(PyExc_ValueError, "unknown hash function");
+        return -1;
+    }
+    EVP_DigestInit(&self->ctx, digest);
+
+    self->name = name_obj;
+    Py_INCREF(self->name);
+
+    if (cp && len) {
+        if (len > 0 && len <= MUNCH_SIZE) {
+        EVP_DigestUpdate(&self->ctx, cp, Py_SAFE_DOWNCAST(len, Py_ssize_t,
+                                                          unsigned int));
+        } else {
+            Py_ssize_t offset = 0;
+            while (len) {
+                unsigned int process = len > MUNCH_SIZE ? MUNCH_SIZE : len;
+                EVP_DigestUpdate(&self->ctx, cp + offset, process);
+                len -= process;
+                offset += process;
+            }
+        }
+    }
+    
+    return 0;
+}
+#endif
+
+
+PyDoc_STRVAR(hashtype_doc,
+"A hash represents the object used to calculate a checksum of a\n\
+string of information.\n\
+\n\
+Methods:\n\
+\n\
+update() -- updates the current digest with an additional string\n\
+digest() -- return the current digest value\n\
+hexdigest() -- return the current digest as a string of hexadecimal digits\n\
+copy() -- return a copy of the current hash object\n\
+\n\
+Attributes:\n\
+\n\
+name -- the hash algorithm being used by this object\n\
+digest_size -- number of bytes in this hashes output\n");
+
+static PyTypeObject EVPtype = {
+    PyObject_HEAD_INIT(NULL)
+    0,			/*ob_size*/
+    "_hashlib.HASH",    /*tp_name*/
+    sizeof(EVPobject),	/*tp_basicsize*/
+    0,			/*tp_itemsize*/
+    /* methods */
+    EVP_dealloc,	/*tp_dealloc*/
+    0,			/*tp_print*/
+    0,                  /*tp_getattr*/
+    0,                  /*tp_setattr*/
+    0,                  /*tp_compare*/
+    EVP_repr,           /*tp_repr*/
+    0,                  /*tp_as_number*/
+    0,                  /*tp_as_sequence*/
+    0,                  /*tp_as_mapping*/
+    0,                  /*tp_hash*/
+    0,                  /*tp_call*/
+    0,                  /*tp_str*/
+    0,                  /*tp_getattro*/
+    0,                  /*tp_setattro*/
+    0,                  /*tp_as_buffer*/
+    Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE, /*tp_flags*/
+    hashtype_doc,       /*tp_doc*/
+    0,                  /*tp_traverse*/
+    0,			/*tp_clear*/
+    0,			/*tp_richcompare*/
+    0,			/*tp_weaklistoffset*/
+    0,			/*tp_iter*/
+    0,			/*tp_iternext*/
+    EVP_methods,	/* tp_methods */
+    EVP_members,	/* tp_members */
+    EVP_getseters,      /* tp_getset */
+#if 1
+    0,                  /* tp_base */
+    0,                  /* tp_dict */
+    0,                  /* tp_descr_get */
+    0,                  /* tp_descr_set */
+    0,                  /* tp_dictoffset */
+#endif
+#if HASH_OBJ_CONSTRUCTOR
+    (initproc)EVP_tp_init, /* tp_init */
+#endif
+};
+
+static PyObject *
+EVPnew(PyObject *name_obj,
+       const EVP_MD *digest, const EVP_MD_CTX *initial_ctx,
+       const unsigned char *cp, Py_ssize_t len)
+{
+    EVPobject *self;
+
+    if (!digest && !initial_ctx) {
+        PyErr_SetString(PyExc_ValueError, "unsupported hash type");
+        return NULL;
+    }
+
+    if ((self = newEVPobject(name_obj)) == NULL)
+        return NULL;
+
+    if (initial_ctx) {
+        EVP_MD_CTX_copy(&self->ctx, initial_ctx);
+    } else {
+        EVP_DigestInit(&self->ctx, digest);
+    }
+
+    if (cp && len) {
+        if (len > 0 && len <= MUNCH_SIZE) {
+            EVP_DigestUpdate(&self->ctx, cp, Py_SAFE_DOWNCAST(len, Py_ssize_t,
+                                                              unsigned int));
+        } else {
+            Py_ssize_t offset = 0;
+            while (len) {
+                unsigned int process = len > MUNCH_SIZE ? MUNCH_SIZE : len;
+                EVP_DigestUpdate(&self->ctx, cp + offset, process);
+                len -= process;
+                offset += process;
+            }
+        }
+    }
+
+    return (PyObject *)self;
+}
+
+
+/* The module-level function: new() */
+
+PyDoc_STRVAR(EVP_new__doc__,
+"Return a new hash object using the named algorithm.\n\
+An optional string argument may be provided and will be\n\
+automatically hashed.\n\
+\n\
+The MD5 and SHA1 algorithms are always supported.\n");
+
+static PyObject *
+EVP_new(PyObject *self, PyObject *args, PyObject *kwdict)
+{
+    static char *kwlist[] = {"name", "string", NULL};
+    PyObject *name_obj = NULL;
+    char *name;
+    const EVP_MD *digest;
+    unsigned char *cp = NULL;
+    Py_ssize_t len = 0;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwdict, "O|s#:new", kwlist,
+                                     &name_obj, &cp, &len)) {
+        return NULL;
+    }
+
+    if (!PyArg_Parse(name_obj, "s", &name)) {
+        PyErr_SetString(PyExc_TypeError, "name must be a string");
+        return NULL;
+    }
+
+    digest = EVP_get_digestbyname(name);
+
+    return EVPnew(name_obj, digest, NULL, cp, len);
+}
+
+/*
+ *  This macro generates constructor function definitions for specific
+ *  hash algorithms.  These constructors are much faster than calling
+ *  the generic one passing it a python string and are noticably
+ *  faster than calling a python new() wrapper.  Thats important for
+ *  code that wants to make hashes of a bunch of small strings.
+ */
+#define GEN_CONSTRUCTOR(NAME)  \
+    static PyObject * \
+    EVP_new_ ## NAME (PyObject *self, PyObject *args) \
+    { \
+        unsigned char *cp = NULL; \
+        Py_ssize_t len = 0; \
+     \
+        if (!PyArg_ParseTuple(args, "|s#:" #NAME , &cp, &len)) { \
+            return NULL; \
+        } \
+     \
+        return EVPnew( \
+                CONST_ ## NAME ## _name_obj, \
+                NULL, \
+                CONST_new_ ## NAME ## _ctx_p, \
+                cp, len); \
+    }
+
+/* a PyMethodDef structure for the constructor */
+#define CONSTRUCTOR_METH_DEF(NAME)  \
+    {"openssl_" #NAME, (PyCFunction)EVP_new_ ## NAME, METH_VARARGS, \
+        PyDoc_STR("Returns a " #NAME \
+                  " hash object; optionally initialized with a string") \
+    }
+
+/* used in the init function to setup a constructor */
+#define INIT_CONSTRUCTOR_CONSTANTS(NAME)  do { \
+    CONST_ ## NAME ## _name_obj = PyString_FromString(#NAME); \
+    if (EVP_get_digestbyname(#NAME)) { \
+        CONST_new_ ## NAME ## _ctx_p = &CONST_new_ ## NAME ## _ctx; \
+        EVP_DigestInit(CONST_new_ ## NAME ## _ctx_p, EVP_get_digestbyname(#NAME)); \
+    } \
+} while (0);
+
+GEN_CONSTRUCTOR(md5)
+GEN_CONSTRUCTOR(sha1)
+GEN_CONSTRUCTOR(sha224)
+GEN_CONSTRUCTOR(sha256)
+GEN_CONSTRUCTOR(sha384)
+GEN_CONSTRUCTOR(sha512)
+
+/* List of functions exported by this module */
+
+static struct PyMethodDef EVP_functions[] = {
+    {"new", (PyCFunction)EVP_new, METH_VARARGS|METH_KEYWORDS, EVP_new__doc__},
+    CONSTRUCTOR_METH_DEF(md5),
+    CONSTRUCTOR_METH_DEF(sha1),
+    CONSTRUCTOR_METH_DEF(sha224),
+    CONSTRUCTOR_METH_DEF(sha256),
+    CONSTRUCTOR_METH_DEF(sha384),
+    CONSTRUCTOR_METH_DEF(sha512),
+    {NULL,	NULL}		 /* Sentinel */
+};
+
+
+/* Initialize this module. */
+
+PyMODINIT_FUNC
+init_hashlib(void)
+{
+    PyObject *m;
+
+    OpenSSL_add_all_digests();
+
+    /* TODO build EVP_functions openssl_* entries dynamically based
+     * on what hashes are supported rather than listing many
+     * but having some be unsupported.  Only init appropriate
+     * constants. */
+
+    EVPtype.ob_type = &PyType_Type;
+    if (PyType_Ready(&EVPtype) < 0)
+        return;
+
+    m = Py_InitModule("_hashlib", EVP_functions);
+    if (m == NULL)
+        return;
+
+#if HASH_OBJ_CONSTRUCTOR
+    Py_INCREF(&EVPtype);
+    PyModule_AddObject(m, "HASH", (PyObject *)&EVPtype);
+#endif
+
+    /* these constants are used by the convenience constructors */
+    INIT_CONSTRUCTOR_CONSTANTS(md5);
+    INIT_CONSTRUCTOR_CONSTANTS(sha1);
+    INIT_CONSTRUCTOR_CONSTANTS(sha224);
+    INIT_CONSTRUCTOR_CONSTANTS(sha256);
+    INIT_CONSTRUCTOR_CONSTANTS(sha384);
+    INIT_CONSTRUCTOR_CONSTANTS(sha512);
+}
diff --git a/src/Modules/md5.c b/src/Modules/md5.c
new file mode 100644
--- /dev/null
+++ b/src/Modules/md5.c
@@ -0,0 +1,381 @@
+/*
+  Copyright (C) 1999, 2000, 2002 Aladdin Enterprises.  All rights reserved.
+
+  This software is provided 'as-is', without any express or implied
+  warranty.  In no event will the authors be held liable for any damages
+  arising from the use of this software.
+
+  Permission is granted to anyone to use this software for any purpose,
+  including commercial applications, and to alter it and redistribute it
+  freely, subject to the following restrictions:
+
+  1. The origin of this software must not be misrepresented; you must not
+     claim that you wrote the original software. If you use this software
+     in a product, an acknowledgment in the product documentation would be
+     appreciated but is not required.
+  2. Altered source versions must be plainly marked as such, and must not be
+     misrepresented as being the original software.
+  3. This notice may not be removed or altered from any source distribution.
+
+  L. Peter Deutsch
+  ghost at aladdin.com
+
+ */
+/* $Id: md5.c,v 1.6 2002/04/13 19:20:28 lpd Exp $ */
+/*
+  Independent implementation of MD5 (RFC 1321).
+
+  This code implements the MD5 Algorithm defined in RFC 1321, whose
+  text is available at
+	http://www.ietf.org/rfc/rfc1321.txt
+  The code is derived from the text of the RFC, including the test suite
+  (section A.5) but excluding the rest of Appendix A.  It does not include
+  any code or documentation that is identified in the RFC as being
+  copyrighted.
+
+  The original and principal author of md5.c is L. Peter Deutsch
+  <ghost at aladdin.com>.  Other authors are noted in the change history
+  that follows (in reverse chronological order):
+
+  2002-04-13 lpd Clarified derivation from RFC 1321; now handles byte order
+	either statically or dynamically; added missing #include <string.h>
+	in library.
+  2002-03-11 lpd Corrected argument list for main(), and added int return
+	type, in test program and T value program.
+  2002-02-21 lpd Added missing #include <stdio.h> in test program.
+  2000-07-03 lpd Patched to eliminate warnings about "constant is
+	unsigned in ANSI C, signed in traditional"; made test program
+	self-checking.
+  1999-11-04 lpd Edited comments slightly for automatic TOC extraction.
+  1999-10-18 lpd Fixed typo in header comment (ansi2knr rather than md5).
+  1999-05-03 lpd Original version.
+ */
+
+#include "md5.h"
+#include <string.h>
+
+#undef BYTE_ORDER	/* 1 = big-endian, -1 = little-endian, 0 = unknown */
+#ifdef ARCH_IS_BIG_ENDIAN
+#  define BYTE_ORDER (ARCH_IS_BIG_ENDIAN ? 1 : -1)
+#else
+#  define BYTE_ORDER 0
+#endif
+
+#define T_MASK ((md5_word_t)~0)
+#define T1 /* 0xd76aa478 */ (T_MASK ^ 0x28955b87)
+#define T2 /* 0xe8c7b756 */ (T_MASK ^ 0x173848a9)
+#define T3    0x242070db
+#define T4 /* 0xc1bdceee */ (T_MASK ^ 0x3e423111)
+#define T5 /* 0xf57c0faf */ (T_MASK ^ 0x0a83f050)
+#define T6    0x4787c62a
+#define T7 /* 0xa8304613 */ (T_MASK ^ 0x57cfb9ec)
+#define T8 /* 0xfd469501 */ (T_MASK ^ 0x02b96afe)
+#define T9    0x698098d8
+#define T10 /* 0x8b44f7af */ (T_MASK ^ 0x74bb0850)
+#define T11 /* 0xffff5bb1 */ (T_MASK ^ 0x0000a44e)
+#define T12 /* 0x895cd7be */ (T_MASK ^ 0x76a32841)
+#define T13    0x6b901122
+#define T14 /* 0xfd987193 */ (T_MASK ^ 0x02678e6c)
+#define T15 /* 0xa679438e */ (T_MASK ^ 0x5986bc71)
+#define T16    0x49b40821
+#define T17 /* 0xf61e2562 */ (T_MASK ^ 0x09e1da9d)
+#define T18 /* 0xc040b340 */ (T_MASK ^ 0x3fbf4cbf)
+#define T19    0x265e5a51
+#define T20 /* 0xe9b6c7aa */ (T_MASK ^ 0x16493855)
+#define T21 /* 0xd62f105d */ (T_MASK ^ 0x29d0efa2)
+#define T22    0x02441453
+#define T23 /* 0xd8a1e681 */ (T_MASK ^ 0x275e197e)
+#define T24 /* 0xe7d3fbc8 */ (T_MASK ^ 0x182c0437)
+#define T25    0x21e1cde6
+#define T26 /* 0xc33707d6 */ (T_MASK ^ 0x3cc8f829)
+#define T27 /* 0xf4d50d87 */ (T_MASK ^ 0x0b2af278)
+#define T28    0x455a14ed
+#define T29 /* 0xa9e3e905 */ (T_MASK ^ 0x561c16fa)
+#define T30 /* 0xfcefa3f8 */ (T_MASK ^ 0x03105c07)
+#define T31    0x676f02d9
+#define T32 /* 0x8d2a4c8a */ (T_MASK ^ 0x72d5b375)
+#define T33 /* 0xfffa3942 */ (T_MASK ^ 0x0005c6bd)
+#define T34 /* 0x8771f681 */ (T_MASK ^ 0x788e097e)
+#define T35    0x6d9d6122
+#define T36 /* 0xfde5380c */ (T_MASK ^ 0x021ac7f3)
+#define T37 /* 0xa4beea44 */ (T_MASK ^ 0x5b4115bb)
+#define T38    0x4bdecfa9
+#define T39 /* 0xf6bb4b60 */ (T_MASK ^ 0x0944b49f)
+#define T40 /* 0xbebfbc70 */ (T_MASK ^ 0x4140438f)
+#define T41    0x289b7ec6
+#define T42 /* 0xeaa127fa */ (T_MASK ^ 0x155ed805)
+#define T43 /* 0xd4ef3085 */ (T_MASK ^ 0x2b10cf7a)
+#define T44    0x04881d05
+#define T45 /* 0xd9d4d039 */ (T_MASK ^ 0x262b2fc6)
+#define T46 /* 0xe6db99e5 */ (T_MASK ^ 0x1924661a)
+#define T47    0x1fa27cf8
+#define T48 /* 0xc4ac5665 */ (T_MASK ^ 0x3b53a99a)
+#define T49 /* 0xf4292244 */ (T_MASK ^ 0x0bd6ddbb)
+#define T50    0x432aff97
+#define T51 /* 0xab9423a7 */ (T_MASK ^ 0x546bdc58)
+#define T52 /* 0xfc93a039 */ (T_MASK ^ 0x036c5fc6)
+#define T53    0x655b59c3
+#define T54 /* 0x8f0ccc92 */ (T_MASK ^ 0x70f3336d)
+#define T55 /* 0xffeff47d */ (T_MASK ^ 0x00100b82)
+#define T56 /* 0x85845dd1 */ (T_MASK ^ 0x7a7ba22e)
+#define T57    0x6fa87e4f
+#define T58 /* 0xfe2ce6e0 */ (T_MASK ^ 0x01d3191f)
+#define T59 /* 0xa3014314 */ (T_MASK ^ 0x5cfebceb)
+#define T60    0x4e0811a1
+#define T61 /* 0xf7537e82 */ (T_MASK ^ 0x08ac817d)
+#define T62 /* 0xbd3af235 */ (T_MASK ^ 0x42c50dca)
+#define T63    0x2ad7d2bb
+#define T64 /* 0xeb86d391 */ (T_MASK ^ 0x14792c6e)
+
+
+static void
+md5_process(md5_state_t *pms, const md5_byte_t *data /*[64]*/)
+{
+    md5_word_t
+	a = pms->abcd[0], b = pms->abcd[1],
+	c = pms->abcd[2], d = pms->abcd[3];
+    md5_word_t t;
+#if BYTE_ORDER > 0
+    /* Define storage only for big-endian CPUs. */
+    md5_word_t X[16];
+#else
+    /* Define storage for little-endian or both types of CPUs. */
+    md5_word_t xbuf[16];
+    const md5_word_t *X;
+#endif
+
+    {
+#if BYTE_ORDER == 0
+	/*
+	 * Determine dynamically whether this is a big-endian or
+	 * little-endian machine, since we can use a more efficient
+	 * algorithm on the latter.
+	 */
+	static const int w = 1;
+
+	if (*((const md5_byte_t *)&w)) /* dynamic little-endian */
+#endif
+#if BYTE_ORDER <= 0		/* little-endian */
+	{
+	    /*
+	     * On little-endian machines, we can process properly aligned
+	     * data without copying it.
+	     */
+	    if (!((data - (const md5_byte_t *)0) & 3)) {
+		/* data are properly aligned */
+		X = (const md5_word_t *)data;
+	    } else {
+		/* not aligned */
+		memcpy(xbuf, data, 64);
+		X = xbuf;
+	    }
+	}
+#endif
+#if BYTE_ORDER == 0
+	else			/* dynamic big-endian */
+#endif
+#if BYTE_ORDER >= 0		/* big-endian */
+	{
+	    /*
+	     * On big-endian machines, we must arrange the bytes in the
+	     * right order.
+	     */
+	    const md5_byte_t *xp = data;
+	    int i;
+
+#  if BYTE_ORDER == 0
+	    X = xbuf;		/* (dynamic only) */
+#  else
+#    define xbuf X		/* (static only) */
+#  endif
+	    for (i = 0; i < 16; ++i, xp += 4)
+		xbuf[i] = xp[0] + (xp[1] << 8) + (xp[2] << 16) + (xp[3] << 24);
+	}
+#endif
+    }
+
+#define ROTATE_LEFT(x, n) (((x) << (n)) | ((x) >> (32 - (n))))
+
+    /* Round 1. */
+    /* Let [abcd k s i] denote the operation
+       a = b + ((a + F(b,c,d) + X[k] + T[i]) <<< s). */
+#define F(x, y, z) (((x) & (y)) | (~(x) & (z)))
+#define SET(a, b, c, d, k, s, Ti)\
+  t = a + F(b,c,d) + X[k] + Ti;\
+  a = ROTATE_LEFT(t, s) + b
+    /* Do the following 16 operations. */
+    SET(a, b, c, d,  0,  7,  T1);
+    SET(d, a, b, c,  1, 12,  T2);
+    SET(c, d, a, b,  2, 17,  T3);
+    SET(b, c, d, a,  3, 22,  T4);
+    SET(a, b, c, d,  4,  7,  T5);
+    SET(d, a, b, c,  5, 12,  T6);
+    SET(c, d, a, b,  6, 17,  T7);
+    SET(b, c, d, a,  7, 22,  T8);
+    SET(a, b, c, d,  8,  7,  T9);
+    SET(d, a, b, c,  9, 12, T10);
+    SET(c, d, a, b, 10, 17, T11);
+    SET(b, c, d, a, 11, 22, T12);
+    SET(a, b, c, d, 12,  7, T13);
+    SET(d, a, b, c, 13, 12, T14);
+    SET(c, d, a, b, 14, 17, T15);
+    SET(b, c, d, a, 15, 22, T16);
+#undef SET
+
+     /* Round 2. */
+     /* Let [abcd k s i] denote the operation
+          a = b + ((a + G(b,c,d) + X[k] + T[i]) <<< s). */
+#define G(x, y, z) (((x) & (z)) | ((y) & ~(z)))
+#define SET(a, b, c, d, k, s, Ti)\
+  t = a + G(b,c,d) + X[k] + Ti;\
+  a = ROTATE_LEFT(t, s) + b
+     /* Do the following 16 operations. */
+    SET(a, b, c, d,  1,  5, T17);
+    SET(d, a, b, c,  6,  9, T18);
+    SET(c, d, a, b, 11, 14, T19);
+    SET(b, c, d, a,  0, 20, T20);
+    SET(a, b, c, d,  5,  5, T21);
+    SET(d, a, b, c, 10,  9, T22);
+    SET(c, d, a, b, 15, 14, T23);
+    SET(b, c, d, a,  4, 20, T24);
+    SET(a, b, c, d,  9,  5, T25);
+    SET(d, a, b, c, 14,  9, T26);
+    SET(c, d, a, b,  3, 14, T27);
+    SET(b, c, d, a,  8, 20, T28);
+    SET(a, b, c, d, 13,  5, T29);
+    SET(d, a, b, c,  2,  9, T30);
+    SET(c, d, a, b,  7, 14, T31);
+    SET(b, c, d, a, 12, 20, T32);
+#undef SET
+
+     /* Round 3. */
+     /* Let [abcd k s t] denote the operation
+          a = b + ((a + H(b,c,d) + X[k] + T[i]) <<< s). */
+#define H(x, y, z) ((x) ^ (y) ^ (z))
+#define SET(a, b, c, d, k, s, Ti)\
+  t = a + H(b,c,d) + X[k] + Ti;\
+  a = ROTATE_LEFT(t, s) + b
+     /* Do the following 16 operations. */
+    SET(a, b, c, d,  5,  4, T33);
+    SET(d, a, b, c,  8, 11, T34);
+    SET(c, d, a, b, 11, 16, T35);
+    SET(b, c, d, a, 14, 23, T36);
+    SET(a, b, c, d,  1,  4, T37);
+    SET(d, a, b, c,  4, 11, T38);
+    SET(c, d, a, b,  7, 16, T39);
+    SET(b, c, d, a, 10, 23, T40);
+    SET(a, b, c, d, 13,  4, T41);
+    SET(d, a, b, c,  0, 11, T42);
+    SET(c, d, a, b,  3, 16, T43);
+    SET(b, c, d, a,  6, 23, T44);
+    SET(a, b, c, d,  9,  4, T45);
+    SET(d, a, b, c, 12, 11, T46);
+    SET(c, d, a, b, 15, 16, T47);
+    SET(b, c, d, a,  2, 23, T48);
+#undef SET
+
+     /* Round 4. */
+     /* Let [abcd k s t] denote the operation
+          a = b + ((a + I(b,c,d) + X[k] + T[i]) <<< s). */
+#define I(x, y, z) ((y) ^ ((x) | ~(z)))
+#define SET(a, b, c, d, k, s, Ti)\
+  t = a + I(b,c,d) + X[k] + Ti;\
+  a = ROTATE_LEFT(t, s) + b
+     /* Do the following 16 operations. */
+    SET(a, b, c, d,  0,  6, T49);
+    SET(d, a, b, c,  7, 10, T50);
+    SET(c, d, a, b, 14, 15, T51);
+    SET(b, c, d, a,  5, 21, T52);
+    SET(a, b, c, d, 12,  6, T53);
+    SET(d, a, b, c,  3, 10, T54);
+    SET(c, d, a, b, 10, 15, T55);
+    SET(b, c, d, a,  1, 21, T56);
+    SET(a, b, c, d,  8,  6, T57);
+    SET(d, a, b, c, 15, 10, T58);
+    SET(c, d, a, b,  6, 15, T59);
+    SET(b, c, d, a, 13, 21, T60);
+    SET(a, b, c, d,  4,  6, T61);
+    SET(d, a, b, c, 11, 10, T62);
+    SET(c, d, a, b,  2, 15, T63);
+    SET(b, c, d, a,  9, 21, T64);
+#undef SET
+
+     /* Then perform the following additions. (That is increment each
+        of the four registers by the value it had before this block
+        was started.) */
+    pms->abcd[0] += a;
+    pms->abcd[1] += b;
+    pms->abcd[2] += c;
+    pms->abcd[3] += d;
+}
+
+void
+md5_init(md5_state_t *pms)
+{
+    pms->count[0] = pms->count[1] = 0;
+    pms->abcd[0] = 0x67452301;
+    pms->abcd[1] = /*0xefcdab89*/ T_MASK ^ 0x10325476;
+    pms->abcd[2] = /*0x98badcfe*/ T_MASK ^ 0x67452301;
+    pms->abcd[3] = 0x10325476;
+}
+
+void
+md5_append(md5_state_t *pms, const md5_byte_t *data, int nbytes)
+{
+    const md5_byte_t *p = data;
+    int left = nbytes;
+    int offset = (pms->count[0] >> 3) & 63;
+    md5_word_t nbits = (md5_word_t)(nbytes << 3);
+
+    if (nbytes <= 0)
+	return;
+
+    /* Update the message length. */
+    pms->count[1] += nbytes >> 29;
+    pms->count[0] += nbits;
+    if (pms->count[0] < nbits)
+	pms->count[1]++;
+
+    /* Process an initial partial block. */
+    if (offset) {
+	int copy = (offset + nbytes > 64 ? 64 - offset : nbytes);
+
+	memcpy(pms->buf + offset, p, copy);
+	if (offset + copy < 64)
+	    return;
+	p += copy;
+	left -= copy;
+	md5_process(pms, pms->buf);
+    }
+
+    /* Process full blocks. */
+    for (; left >= 64; p += 64, left -= 64)
+	md5_process(pms, p);
+
+    /* Process a final partial block. */
+    if (left)
+	memcpy(pms->buf, p, left);
+}
+
+void
+md5_finish(md5_state_t *pms, md5_byte_t digest[16])
+{
+    static const md5_byte_t pad[64] = {
+	0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
+    };
+    md5_byte_t data[8];
+    int i;
+
+    /* Save the length before padding. */
+    for (i = 0; i < 8; ++i)
+	data[i] = (md5_byte_t)(pms->count[i >> 2] >> ((i & 3) << 3));
+    /* Pad to 56 bytes mod 64. */
+    md5_append(pms, pad, ((55 - (pms->count[0] >> 3)) & 63) + 1);
+    /* Append the length. */
+    md5_append(pms, data, 8);
+    for (i = 0; i < 16; ++i)
+	digest[i] = (md5_byte_t)(pms->abcd[i >> 2] >> ((i & 3) << 3));
+}
diff --git a/src/Modules/md5.h b/src/Modules/md5.h
new file mode 100644
--- /dev/null
+++ b/src/Modules/md5.h
@@ -0,0 +1,91 @@
+/*
+  Copyright (C) 1999, 2002 Aladdin Enterprises.  All rights reserved.
+
+  This software is provided 'as-is', without any express or implied
+  warranty.  In no event will the authors be held liable for any damages
+  arising from the use of this software.
+
+  Permission is granted to anyone to use this software for any purpose,
+  including commercial applications, and to alter it and redistribute it
+  freely, subject to the following restrictions:
+
+  1. The origin of this software must not be misrepresented; you must not
+     claim that you wrote the original software. If you use this software
+     in a product, an acknowledgment in the product documentation would be
+     appreciated but is not required.
+  2. Altered source versions must be plainly marked as such, and must not be
+     misrepresented as being the original software.
+  3. This notice may not be removed or altered from any source distribution.
+
+  L. Peter Deutsch
+  ghost at aladdin.com
+
+ */
+/* $Id: md5.h 43594 2006-04-03 16:27:50Z matthias.klose $ */
+/*
+  Independent implementation of MD5 (RFC 1321).
+
+  This code implements the MD5 Algorithm defined in RFC 1321, whose
+  text is available at
+	http://www.ietf.org/rfc/rfc1321.txt
+  The code is derived from the text of the RFC, including the test suite
+  (section A.5) but excluding the rest of Appendix A.  It does not include
+  any code or documentation that is identified in the RFC as being
+  copyrighted.
+
+  The original and principal author of md5.h is L. Peter Deutsch
+  <ghost at aladdin.com>.  Other authors are noted in the change history
+  that follows (in reverse chronological order):
+
+  2002-04-13 lpd Removed support for non-ANSI compilers; removed
+	references to Ghostscript; clarified derivation from RFC 1321;
+	now handles byte order either statically or dynamically.
+  1999-11-04 lpd Edited comments slightly for automatic TOC extraction.
+  1999-10-18 lpd Fixed typo in header comment (ansi2knr rather than md5);
+	added conditionalization for C++ compilation from Martin
+	Purschke <purschke at bnl.gov>.
+  1999-05-03 lpd Original version.
+ */
+
+#ifndef md5_INCLUDED
+#  define md5_INCLUDED
+
+/*
+ * This package supports both compile-time and run-time determination of CPU
+ * byte order.  If ARCH_IS_BIG_ENDIAN is defined as 0, the code will be
+ * compiled to run only on little-endian CPUs; if ARCH_IS_BIG_ENDIAN is
+ * defined as non-zero, the code will be compiled to run only on big-endian
+ * CPUs; if ARCH_IS_BIG_ENDIAN is not defined, the code will be compiled to
+ * run on either big- or little-endian CPUs, but will run slightly less
+ * efficiently on either one than if ARCH_IS_BIG_ENDIAN is defined.
+ */
+
+typedef unsigned char md5_byte_t; /* 8-bit byte */
+typedef unsigned int md5_word_t; /* 32-bit word */
+
+/* Define the state of the MD5 Algorithm. */
+typedef struct md5_state_s {
+    md5_word_t count[2];	/* message length in bits, lsw first */
+    md5_word_t abcd[4];		/* digest buffer */
+    md5_byte_t buf[64];		/* accumulate block */
+} md5_state_t;
+
+#ifdef __cplusplus
+extern "C" 
+{
+#endif
+
+/* Initialize the algorithm. */
+void md5_init(md5_state_t *pms);
+
+/* Append a string to the message. */
+void md5_append(md5_state_t *pms, const md5_byte_t *data, int nbytes);
+
+/* Finish the message and return the digest. */
+void md5_finish(md5_state_t *pms, md5_byte_t digest[16]);
+
+#ifdef __cplusplus
+}  /* end extern "C" */
+#endif
+
+#endif /* md5_INCLUDED */
diff --git a/src/Modules/md5module.c b/src/Modules/md5module.c
new file mode 100644
--- /dev/null
+++ b/src/Modules/md5module.c
@@ -0,0 +1,312 @@
+
+/* MD5 module */
+
+/* This module provides an interface to the RSA Data Security,
+   Inc. MD5 Message-Digest Algorithm, described in RFC 1321.
+   It requires the files md5c.c and md5.h (which are slightly changed
+   from the versions in the RFC to avoid the "global.h" file.) */
+
+
+/* MD5 objects */
+
+#include "Python.h"
+#include "structmember.h"
+#include "md5.h"
+
+typedef struct {
+	PyObject_HEAD
+        md5_state_t	md5;		/* the context holder */
+} md5object;
+
+static PyTypeObject MD5type;
+
+#define is_md5object(v)		((v)->ob_type == &MD5type)
+
+static md5object *
+newmd5object(void)
+{
+	md5object *md5p;
+
+	md5p = PyObject_New(md5object, &MD5type);
+	if (md5p == NULL)
+		return NULL;
+
+	md5_init(&md5p->md5);	/* actual initialisation */
+	return md5p;
+}
+
+
+/* MD5 methods */
+
+static void
+md5_dealloc(md5object *md5p)
+{
+	PyObject_Del(md5p);
+}
+
+
+/* MD5 methods-as-attributes */
+
+static PyObject *
+md5_update(md5object *self, PyObject *args)
+{
+	unsigned char *cp;
+	int len;
+
+	if (!PyArg_ParseTuple(args, "s#:update", &cp, &len))
+		return NULL;
+
+	md5_append(&self->md5, cp, len);
+
+	Py_INCREF(Py_None);
+	return Py_None;
+}
+
+PyDoc_STRVAR(update_doc,
+"update (arg)\n\
+\n\
+Update the md5 object with the string arg. Repeated calls are\n\
+equivalent to a single call with the concatenation of all the\n\
+arguments.");
+
+
+static PyObject *
+md5_digest(md5object *self)
+{
+ 	md5_state_t mdContext;
+	unsigned char aDigest[16];
+
+	/* make a temporary copy, and perform the final */
+	mdContext = self->md5;
+	md5_finish(&mdContext, aDigest);
+
+	return PyString_FromStringAndSize((char *)aDigest, 16);
+}
+
+PyDoc_STRVAR(digest_doc,
+"digest() -> string\n\
+\n\
+Return the digest of the strings passed to the update() method so\n\
+far. This is a 16-byte string which may contain non-ASCII characters,\n\
+including null bytes.");
+
+
+static PyObject *
+md5_hexdigest(md5object *self)
+{
+ 	md5_state_t mdContext;
+	unsigned char digest[16];
+	unsigned char hexdigest[32];
+	int i, j;
+
+	/* make a temporary copy, and perform the final */
+	mdContext = self->md5;
+	md5_finish(&mdContext, digest);
+
+	/* Make hex version of the digest */
+	for(i=j=0; i<16; i++) {
+		char c;
+		c = (digest[i] >> 4) & 0xf;
+		c = (c>9) ? c+'a'-10 : c + '0';
+		hexdigest[j++] = c;
+		c = (digest[i] & 0xf);
+		c = (c>9) ? c+'a'-10 : c + '0';
+		hexdigest[j++] = c;
+	}
+	return PyString_FromStringAndSize((char*)hexdigest, 32);
+}
+
+
+PyDoc_STRVAR(hexdigest_doc,
+"hexdigest() -> string\n\
+\n\
+Like digest(), but returns the digest as a string of hexadecimal digits.");
+
+
+static PyObject *
+md5_copy(md5object *self)
+{
+	md5object *md5p;
+
+	if ((md5p = newmd5object()) == NULL)
+		return NULL;
+
+	md5p->md5 = self->md5;
+
+	return (PyObject *)md5p;
+}
+
+PyDoc_STRVAR(copy_doc,
+"copy() -> md5 object\n\
+\n\
+Return a copy (``clone'') of the md5 object.");
+
+
+static PyMethodDef md5_methods[] = {
+	{"update",    (PyCFunction)md5_update,    METH_VARARGS, update_doc},
+	{"digest",    (PyCFunction)md5_digest,    METH_NOARGS,  digest_doc},
+	{"hexdigest", (PyCFunction)md5_hexdigest, METH_NOARGS,  hexdigest_doc},
+	{"copy",      (PyCFunction)md5_copy,      METH_NOARGS,  copy_doc},
+	{NULL, NULL}			     /* sentinel */
+};
+
+static PyObject *
+md5_get_block_size(PyObject *self, void *closure)
+{
+    return PyInt_FromLong(64);
+}
+
+static PyObject *
+md5_get_digest_size(PyObject *self, void *closure)
+{
+    return PyInt_FromLong(16);
+}
+
+static PyObject *
+md5_get_name(PyObject *self, void *closure)
+{
+    return PyString_FromStringAndSize("MD5", 3);
+}
+
+static PyGetSetDef md5_getseters[] = {
+    {"digest_size",
+     (getter)md5_get_digest_size, NULL,
+     NULL,
+     NULL},
+    {"block_size",
+     (getter)md5_get_block_size, NULL,
+     NULL,
+     NULL},
+    {"name",
+     (getter)md5_get_name, NULL,
+     NULL,
+     NULL},
+    /* the old md5 and sha modules support 'digest_size' as in PEP 247.
+     * the old sha module also supported 'digestsize'.  ugh. */
+    {"digestsize",
+     (getter)md5_get_digest_size, NULL,
+     NULL,
+     NULL},
+    {NULL}  /* Sentinel */
+};
+
+
+PyDoc_STRVAR(module_doc,
+"This module implements the interface to RSA's MD5 message digest\n\
+algorithm (see also Internet RFC 1321). Its use is quite\n\
+straightforward: use the new() to create an md5 object. You can now\n\
+feed this object with arbitrary strings using the update() method, and\n\
+at any point you can ask it for the digest (a strong kind of 128-bit\n\
+checksum, a.k.a. ``fingerprint'') of the concatenation of the strings\n\
+fed to it so far using the digest() method.\n\
+\n\
+Functions:\n\
+\n\
+new([arg]) -- return a new md5 object, initialized with arg if provided\n\
+md5([arg]) -- DEPRECATED, same as new, but for compatibility\n\
+\n\
+Special Objects:\n\
+\n\
+MD5Type -- type object for md5 objects");
+
+PyDoc_STRVAR(md5type_doc,
+"An md5 represents the object used to calculate the MD5 checksum of a\n\
+string of information.\n\
+\n\
+Methods:\n\
+\n\
+update() -- updates the current digest with an additional string\n\
+digest() -- return the current digest value\n\
+hexdigest() -- return the current digest as a string of hexadecimal digits\n\
+copy() -- return a copy of the current md5 object");
+
+static PyTypeObject MD5type = {
+	PyObject_HEAD_INIT(NULL)
+	0,			  /*ob_size*/
+	"_md5.md5",		  /*tp_name*/
+	sizeof(md5object),	  /*tp_size*/
+	0,			  /*tp_itemsize*/
+	/* methods */
+	(destructor)md5_dealloc,  /*tp_dealloc*/
+	0,			  /*tp_print*/
+	0,                        /*tp_getattr*/
+	0,			  /*tp_setattr*/
+	0,			  /*tp_compare*/
+	0,			  /*tp_repr*/
+        0,			  /*tp_as_number*/
+	0,                        /*tp_as_sequence*/
+	0,			  /*tp_as_mapping*/
+	0, 			  /*tp_hash*/
+	0,			  /*tp_call*/
+	0,			  /*tp_str*/
+	0,			  /*tp_getattro*/
+	0,			  /*tp_setattro*/
+	0,	                  /*tp_as_buffer*/
+	Py_TPFLAGS_DEFAULT,	  /*tp_flags*/
+	md5type_doc,		  /*tp_doc*/
+        0,                        /*tp_traverse*/
+        0,			  /*tp_clear*/
+        0,			  /*tp_richcompare*/
+        0,			  /*tp_weaklistoffset*/
+        0,			  /*tp_iter*/
+        0,			  /*tp_iternext*/
+        md5_methods,	          /*tp_methods*/
+        0,      	          /*tp_members*/
+        md5_getseters,            /*tp_getset*/
+};
+
+
+/* MD5 functions */
+
+static PyObject *
+MD5_new(PyObject *self, PyObject *args)
+{
+	md5object *md5p;
+	unsigned char *cp = NULL;
+	int len = 0;
+
+	if (!PyArg_ParseTuple(args, "|s#:new", &cp, &len))
+		return NULL;
+
+	if ((md5p = newmd5object()) == NULL)
+		return NULL;
+
+	if (cp)
+		md5_append(&md5p->md5, cp, len);
+
+	return (PyObject *)md5p;
+}
+
+PyDoc_STRVAR(new_doc,
+"new([arg]) -> md5 object\n\
+\n\
+Return a new md5 object. If arg is present, the method call update(arg)\n\
+is made.");
+
+
+/* List of functions exported by this module */
+
+static PyMethodDef md5_functions[] = {
+	{"new",		(PyCFunction)MD5_new, METH_VARARGS, new_doc},
+	{NULL,		NULL}	/* Sentinel */
+};
+
+
+/* Initialize this module. */
+
+PyMODINIT_FUNC
+init_md5(void)
+{
+	PyObject *m, *d;
+
+        MD5type.ob_type = &PyType_Type;
+        if (PyType_Ready(&MD5type) < 0)
+            return;
+	m = Py_InitModule3("_md5", md5_functions, module_doc);
+	if (m == NULL)
+	    return;
+	d = PyModule_GetDict(m);
+	PyDict_SetItemString(d, "MD5Type", (PyObject *)&MD5type);
+	PyModule_AddIntConstant(m, "digest_size", 16);
+	/* No need to check the error here, the caller will do that */
+}
diff --git a/src/Modules/sha256module.c b/src/Modules/sha256module.c
new file mode 100644
--- /dev/null
+++ b/src/Modules/sha256module.c
@@ -0,0 +1,701 @@
+/* SHA256 module */
+
+/* This module provides an interface to NIST's SHA-256 and SHA-224 Algorithms */
+
+/* See below for information about the original code this module was
+   based upon. Additional work performed by:
+
+   Andrew Kuchling (amk at amk.ca)
+   Greg Stein (gstein at lyra.org)
+   Trevor Perrin (trevp at trevp.net)
+
+   Copyright (C) 2005   Gregory P. Smith (greg at krypto.org)
+   Licensed to PSF under a Contributor Agreement.
+
+*/
+
+/* SHA objects */
+
+#include "Python.h"
+#include "structmember.h"
+
+
+/* Endianness testing and definitions */
+#define TestEndianness(variable) {int i=1; variable=PCT_BIG_ENDIAN;\
+	if (*((char*)&i)==1) variable=PCT_LITTLE_ENDIAN;}
+
+#define PCT_LITTLE_ENDIAN 1
+#define PCT_BIG_ENDIAN 0
+
+/* Some useful types */
+
+typedef unsigned char SHA_BYTE;
+
+#if SIZEOF_INT == 4
+typedef unsigned int SHA_INT32;	/* 32-bit integer */
+#else
+/* not defined. compilation will die. */
+#endif
+
+/* The SHA block size and message digest sizes, in bytes */
+
+#define SHA_BLOCKSIZE    64
+#define SHA_DIGESTSIZE  32
+
+/* The structure for storing SHA info */
+
+typedef struct {
+    PyObject_HEAD
+    SHA_INT32 digest[8];		/* Message digest */
+    SHA_INT32 count_lo, count_hi;	/* 64-bit bit count */
+    SHA_BYTE data[SHA_BLOCKSIZE];	/* SHA data buffer */
+    int Endianness;
+    int local;				/* unprocessed amount in data */
+    int digestsize;
+} SHAobject;
+
+/* When run on a little-endian CPU we need to perform byte reversal on an
+   array of longwords. */
+
+static void longReverse(SHA_INT32 *buffer, int byteCount, int Endianness)
+{
+    SHA_INT32 value;
+
+    if ( Endianness == PCT_BIG_ENDIAN )
+	return;
+
+    byteCount /= sizeof(*buffer);
+    while (byteCount--) {
+        value = *buffer;
+        value = ( ( value & 0xFF00FF00L ) >> 8  ) | \
+                ( ( value & 0x00FF00FFL ) << 8 );
+        *buffer++ = ( value << 16 ) | ( value >> 16 );
+    }
+}
+
+static void SHAcopy(SHAobject *src, SHAobject *dest)
+{
+    dest->Endianness = src->Endianness;
+    dest->local = src->local;
+    dest->digestsize = src->digestsize;
+    dest->count_lo = src->count_lo;
+    dest->count_hi = src->count_hi;
+    memcpy(dest->digest, src->digest, sizeof(src->digest));
+    memcpy(dest->data, src->data, sizeof(src->data));
+}
+
+
+/* ------------------------------------------------------------------------
+ *
+ * This code for the SHA-256 algorithm was noted as public domain. The
+ * original headers are pasted below.
+ *
+ * Several changes have been made to make it more compatible with the
+ * Python environment and desired interface.
+ *
+ */
+
+/* LibTomCrypt, modular cryptographic library -- Tom St Denis
+ *
+ * LibTomCrypt is a library that provides various cryptographic
+ * algorithms in a highly modular and flexible manner.
+ *
+ * The library is free for all purposes without any express
+ * gurantee it works.
+ *
+ * Tom St Denis, tomstdenis at iahu.ca, http://libtomcrypt.org
+ */
+
+
+/* SHA256 by Tom St Denis */
+
+/* Various logical functions */
+#define ROR(x, y)\
+( ((((unsigned long)(x)&0xFFFFFFFFUL)>>(unsigned long)((y)&31)) | \
+((unsigned long)(x)<<(unsigned long)(32-((y)&31)))) & 0xFFFFFFFFUL)
+#define Ch(x,y,z)       (z ^ (x & (y ^ z)))
+#define Maj(x,y,z)      (((x | y) & z) | (x & y)) 
+#define S(x, n)         ROR((x),(n))
+#define R(x, n)         (((x)&0xFFFFFFFFUL)>>(n))
+#define Sigma0(x)       (S(x, 2) ^ S(x, 13) ^ S(x, 22))
+#define Sigma1(x)       (S(x, 6) ^ S(x, 11) ^ S(x, 25))
+#define Gamma0(x)       (S(x, 7) ^ S(x, 18) ^ R(x, 3))
+#define Gamma1(x)       (S(x, 17) ^ S(x, 19) ^ R(x, 10))
+
+
+static void
+sha_transform(SHAobject *sha_info)
+{
+    int i;
+	SHA_INT32 S[8], W[64], t0, t1;
+
+    memcpy(W, sha_info->data, sizeof(sha_info->data));
+    longReverse(W, (int)sizeof(sha_info->data), sha_info->Endianness);
+
+    for (i = 16; i < 64; ++i) {
+		W[i] = Gamma1(W[i - 2]) + W[i - 7] + Gamma0(W[i - 15]) + W[i - 16];
+    }
+    for (i = 0; i < 8; ++i) {
+        S[i] = sha_info->digest[i];
+    }
+
+    /* Compress */
+#define RND(a,b,c,d,e,f,g,h,i,ki)                    \
+     t0 = h + Sigma1(e) + Ch(e, f, g) + ki + W[i];   \
+     t1 = Sigma0(a) + Maj(a, b, c);                  \
+     d += t0;                                        \
+     h  = t0 + t1;
+
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],0,0x428a2f98);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],1,0x71374491);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],2,0xb5c0fbcf);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],3,0xe9b5dba5);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],4,0x3956c25b);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],5,0x59f111f1);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],6,0x923f82a4);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],7,0xab1c5ed5);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],8,0xd807aa98);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],9,0x12835b01);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],10,0x243185be);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],11,0x550c7dc3);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],12,0x72be5d74);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],13,0x80deb1fe);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],14,0x9bdc06a7);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],15,0xc19bf174);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],16,0xe49b69c1);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],17,0xefbe4786);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],18,0x0fc19dc6);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],19,0x240ca1cc);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],20,0x2de92c6f);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],21,0x4a7484aa);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],22,0x5cb0a9dc);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],23,0x76f988da);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],24,0x983e5152);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],25,0xa831c66d);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],26,0xb00327c8);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],27,0xbf597fc7);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],28,0xc6e00bf3);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],29,0xd5a79147);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],30,0x06ca6351);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],31,0x14292967);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],32,0x27b70a85);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],33,0x2e1b2138);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],34,0x4d2c6dfc);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],35,0x53380d13);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],36,0x650a7354);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],37,0x766a0abb);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],38,0x81c2c92e);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],39,0x92722c85);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],40,0xa2bfe8a1);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],41,0xa81a664b);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],42,0xc24b8b70);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],43,0xc76c51a3);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],44,0xd192e819);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],45,0xd6990624);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],46,0xf40e3585);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],47,0x106aa070);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],48,0x19a4c116);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],49,0x1e376c08);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],50,0x2748774c);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],51,0x34b0bcb5);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],52,0x391c0cb3);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],53,0x4ed8aa4a);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],54,0x5b9cca4f);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],55,0x682e6ff3);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],56,0x748f82ee);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],57,0x78a5636f);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],58,0x84c87814);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],59,0x8cc70208);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],60,0x90befffa);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],61,0xa4506ceb);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],62,0xbef9a3f7);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],63,0xc67178f2);
+
+#undef RND     
+    
+    /* feedback */
+    for (i = 0; i < 8; i++) {
+        sha_info->digest[i] = sha_info->digest[i] + S[i];
+    }
+
+}
+
+
+
+/* initialize the SHA digest */
+
+static void
+sha_init(SHAobject *sha_info)
+{
+    TestEndianness(sha_info->Endianness)
+    sha_info->digest[0] = 0x6A09E667L;
+    sha_info->digest[1] = 0xBB67AE85L;
+    sha_info->digest[2] = 0x3C6EF372L;
+    sha_info->digest[3] = 0xA54FF53AL;
+    sha_info->digest[4] = 0x510E527FL;
+    sha_info->digest[5] = 0x9B05688CL;
+    sha_info->digest[6] = 0x1F83D9ABL;
+    sha_info->digest[7] = 0x5BE0CD19L;
+    sha_info->count_lo = 0L;
+    sha_info->count_hi = 0L;
+    sha_info->local = 0;
+    sha_info->digestsize = 32;
+}
+
+static void
+sha224_init(SHAobject *sha_info)
+{
+    TestEndianness(sha_info->Endianness)
+    sha_info->digest[0] = 0xc1059ed8L;
+    sha_info->digest[1] = 0x367cd507L;
+    sha_info->digest[2] = 0x3070dd17L;
+    sha_info->digest[3] = 0xf70e5939L;
+    sha_info->digest[4] = 0xffc00b31L;
+    sha_info->digest[5] = 0x68581511L;
+    sha_info->digest[6] = 0x64f98fa7L;
+    sha_info->digest[7] = 0xbefa4fa4L;
+    sha_info->count_lo = 0L;
+    sha_info->count_hi = 0L;
+    sha_info->local = 0;
+    sha_info->digestsize = 28;
+}
+
+
+/* update the SHA digest */
+
+static void
+sha_update(SHAobject *sha_info, SHA_BYTE *buffer, int count)
+{
+    int i;
+    SHA_INT32 clo;
+
+    clo = sha_info->count_lo + ((SHA_INT32) count << 3);
+    if (clo < sha_info->count_lo) {
+        ++sha_info->count_hi;
+    }
+    sha_info->count_lo = clo;
+    sha_info->count_hi += (SHA_INT32) count >> 29;
+    if (sha_info->local) {
+        i = SHA_BLOCKSIZE - sha_info->local;
+        if (i > count) {
+            i = count;
+        }
+        memcpy(((SHA_BYTE *) sha_info->data) + sha_info->local, buffer, i);
+        count -= i;
+        buffer += i;
+        sha_info->local += i;
+        if (sha_info->local == SHA_BLOCKSIZE) {
+            sha_transform(sha_info);
+        }
+        else {
+            return;
+        }
+    }
+    while (count >= SHA_BLOCKSIZE) {
+        memcpy(sha_info->data, buffer, SHA_BLOCKSIZE);
+        buffer += SHA_BLOCKSIZE;
+        count -= SHA_BLOCKSIZE;
+        sha_transform(sha_info);
+    }
+    memcpy(sha_info->data, buffer, count);
+    sha_info->local = count;
+}
+
+/* finish computing the SHA digest */
+
+static void
+sha_final(unsigned char digest[SHA_DIGESTSIZE], SHAobject *sha_info)
+{
+    int count;
+    SHA_INT32 lo_bit_count, hi_bit_count;
+
+    lo_bit_count = sha_info->count_lo;
+    hi_bit_count = sha_info->count_hi;
+    count = (int) ((lo_bit_count >> 3) & 0x3f);
+    ((SHA_BYTE *) sha_info->data)[count++] = 0x80;
+    if (count > SHA_BLOCKSIZE - 8) {
+	memset(((SHA_BYTE *) sha_info->data) + count, 0,
+	       SHA_BLOCKSIZE - count);
+	sha_transform(sha_info);
+	memset((SHA_BYTE *) sha_info->data, 0, SHA_BLOCKSIZE - 8);
+    }
+    else {
+	memset(((SHA_BYTE *) sha_info->data) + count, 0,
+	       SHA_BLOCKSIZE - 8 - count);
+    }
+
+    /* GJS: note that we add the hi/lo in big-endian. sha_transform will
+       swap these values into host-order. */
+    sha_info->data[56] = (hi_bit_count >> 24) & 0xff;
+    sha_info->data[57] = (hi_bit_count >> 16) & 0xff;
+    sha_info->data[58] = (hi_bit_count >>  8) & 0xff;
+    sha_info->data[59] = (hi_bit_count >>  0) & 0xff;
+    sha_info->data[60] = (lo_bit_count >> 24) & 0xff;
+    sha_info->data[61] = (lo_bit_count >> 16) & 0xff;
+    sha_info->data[62] = (lo_bit_count >>  8) & 0xff;
+    sha_info->data[63] = (lo_bit_count >>  0) & 0xff;
+    sha_transform(sha_info);
+    digest[ 0] = (unsigned char) ((sha_info->digest[0] >> 24) & 0xff);
+    digest[ 1] = (unsigned char) ((sha_info->digest[0] >> 16) & 0xff);
+    digest[ 2] = (unsigned char) ((sha_info->digest[0] >>  8) & 0xff);
+    digest[ 3] = (unsigned char) ((sha_info->digest[0]      ) & 0xff);
+    digest[ 4] = (unsigned char) ((sha_info->digest[1] >> 24) & 0xff);
+    digest[ 5] = (unsigned char) ((sha_info->digest[1] >> 16) & 0xff);
+    digest[ 6] = (unsigned char) ((sha_info->digest[1] >>  8) & 0xff);
+    digest[ 7] = (unsigned char) ((sha_info->digest[1]      ) & 0xff);
+    digest[ 8] = (unsigned char) ((sha_info->digest[2] >> 24) & 0xff);
+    digest[ 9] = (unsigned char) ((sha_info->digest[2] >> 16) & 0xff);
+    digest[10] = (unsigned char) ((sha_info->digest[2] >>  8) & 0xff);
+    digest[11] = (unsigned char) ((sha_info->digest[2]      ) & 0xff);
+    digest[12] = (unsigned char) ((sha_info->digest[3] >> 24) & 0xff);
+    digest[13] = (unsigned char) ((sha_info->digest[3] >> 16) & 0xff);
+    digest[14] = (unsigned char) ((sha_info->digest[3] >>  8) & 0xff);
+    digest[15] = (unsigned char) ((sha_info->digest[3]      ) & 0xff);
+    digest[16] = (unsigned char) ((sha_info->digest[4] >> 24) & 0xff);
+    digest[17] = (unsigned char) ((sha_info->digest[4] >> 16) & 0xff);
+    digest[18] = (unsigned char) ((sha_info->digest[4] >>  8) & 0xff);
+    digest[19] = (unsigned char) ((sha_info->digest[4]      ) & 0xff);
+    digest[20] = (unsigned char) ((sha_info->digest[5] >> 24) & 0xff);
+    digest[21] = (unsigned char) ((sha_info->digest[5] >> 16) & 0xff);
+    digest[22] = (unsigned char) ((sha_info->digest[5] >>  8) & 0xff);
+    digest[23] = (unsigned char) ((sha_info->digest[5]      ) & 0xff);
+    digest[24] = (unsigned char) ((sha_info->digest[6] >> 24) & 0xff);
+    digest[25] = (unsigned char) ((sha_info->digest[6] >> 16) & 0xff);
+    digest[26] = (unsigned char) ((sha_info->digest[6] >>  8) & 0xff);
+    digest[27] = (unsigned char) ((sha_info->digest[6]      ) & 0xff);
+    digest[28] = (unsigned char) ((sha_info->digest[7] >> 24) & 0xff);
+    digest[29] = (unsigned char) ((sha_info->digest[7] >> 16) & 0xff);
+    digest[30] = (unsigned char) ((sha_info->digest[7] >>  8) & 0xff);
+    digest[31] = (unsigned char) ((sha_info->digest[7]      ) & 0xff);
+}
+
+/*
+ * End of copied SHA code.
+ *
+ * ------------------------------------------------------------------------
+ */
+
+static PyTypeObject SHA224type;
+static PyTypeObject SHA256type;
+
+
+static SHAobject *
+newSHA224object(void)
+{
+    return (SHAobject *)PyObject_New(SHAobject, &SHA224type);
+}
+
+static SHAobject *
+newSHA256object(void)
+{
+    return (SHAobject *)PyObject_New(SHAobject, &SHA256type);
+}
+
+/* Internal methods for a hash object */
+
+static void
+SHA_dealloc(PyObject *ptr)
+{
+    PyObject_Del(ptr);
+}
+
+
+/* External methods for a hash object */
+
+PyDoc_STRVAR(SHA256_copy__doc__, "Return a copy of the hash object.");
+
+static PyObject *
+SHA256_copy(SHAobject *self, PyObject *unused)
+{
+    SHAobject *newobj;
+
+    if (((PyObject*)self)->ob_type == &SHA256type) {
+        if ( (newobj = newSHA256object())==NULL)
+            return NULL;
+    } else {
+        if ( (newobj = newSHA224object())==NULL)
+            return NULL;
+    }
+
+    SHAcopy(self, newobj);
+    return (PyObject *)newobj;
+}
+
+PyDoc_STRVAR(SHA256_digest__doc__,
+"Return the digest value as a string of binary data.");
+
+static PyObject *
+SHA256_digest(SHAobject *self, PyObject *unused)
+{
+    unsigned char digest[SHA_DIGESTSIZE];
+    SHAobject temp;
+
+    SHAcopy(self, &temp);
+    sha_final(digest, &temp);
+    return PyString_FromStringAndSize((const char *)digest, self->digestsize);
+}
+
+PyDoc_STRVAR(SHA256_hexdigest__doc__,
+"Return the digest value as a string of hexadecimal digits.");
+
+static PyObject *
+SHA256_hexdigest(SHAobject *self, PyObject *unused)
+{
+    unsigned char digest[SHA_DIGESTSIZE];
+    SHAobject temp;
+    PyObject *retval;
+    char *hex_digest;
+    int i, j;
+
+    /* Get the raw (binary) digest value */
+    SHAcopy(self, &temp);
+    sha_final(digest, &temp);
+
+    /* Create a new string */
+    retval = PyString_FromStringAndSize(NULL, self->digestsize * 2);
+    if (!retval)
+	    return NULL;
+    hex_digest = PyString_AsString(retval);
+    if (!hex_digest) {
+	    Py_DECREF(retval);
+	    return NULL;
+    }
+
+    /* Make hex version of the digest */
+    for(i=j=0; i<self->digestsize; i++) {
+        char c;
+        c = (digest[i] >> 4) & 0xf;
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+        c = (digest[i] & 0xf);
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+    }
+    return retval;
+}
+
+PyDoc_STRVAR(SHA256_update__doc__,
+"Update this hash object's state with the provided string.");
+
+static PyObject *
+SHA256_update(SHAobject *self, PyObject *args)
+{
+    unsigned char *cp;
+    int len;
+
+    if (!PyArg_ParseTuple(args, "s#:update", &cp, &len))
+        return NULL;
+
+    sha_update(self, cp, len);
+
+    Py_INCREF(Py_None);
+    return Py_None;
+}
+
+static PyMethodDef SHA_methods[] = {
+    {"copy",	  (PyCFunction)SHA256_copy,      METH_NOARGS,  SHA256_copy__doc__},
+    {"digest",	  (PyCFunction)SHA256_digest,    METH_NOARGS,  SHA256_digest__doc__},
+    {"hexdigest", (PyCFunction)SHA256_hexdigest, METH_NOARGS,  SHA256_hexdigest__doc__},
+    {"update",	  (PyCFunction)SHA256_update,    METH_VARARGS, SHA256_update__doc__},
+    {NULL,	  NULL}		/* sentinel */
+};
+
+static PyObject *
+SHA256_get_block_size(PyObject *self, void *closure)
+{
+    return PyInt_FromLong(SHA_BLOCKSIZE);
+}
+
+static PyObject *
+SHA256_get_name(PyObject *self, void *closure)
+{
+    if (((SHAobject *)self)->digestsize == 32)
+        return PyString_FromStringAndSize("SHA256", 6);
+    else
+        return PyString_FromStringAndSize("SHA224", 6);
+}
+
+static PyGetSetDef SHA_getseters[] = {
+    {"block_size",
+     (getter)SHA256_get_block_size, NULL,
+     NULL,
+     NULL},
+    {"name",
+     (getter)SHA256_get_name, NULL,
+     NULL,
+     NULL},
+    {NULL}  /* Sentinel */
+};
+
+static PyMemberDef SHA_members[] = {
+    {"digest_size", T_INT, offsetof(SHAobject, digestsize), READONLY, NULL},
+    /* the old md5 and sha modules support 'digest_size' as in PEP 247.
+     * the old sha module also supported 'digestsize'.  ugh. */
+    {"digestsize", T_INT, offsetof(SHAobject, digestsize), READONLY, NULL},
+    {NULL}  /* Sentinel */
+};
+
+static PyTypeObject SHA224type = {
+    PyObject_HEAD_INIT(NULL)
+    0,			/*ob_size*/
+    "_sha256.sha224",	/*tp_name*/
+    sizeof(SHAobject),	/*tp_size*/
+    0,			/*tp_itemsize*/
+    /* methods */
+    SHA_dealloc,	/*tp_dealloc*/
+    0,			/*tp_print*/
+    0,          	/*tp_getattr*/
+    0,                  /*tp_setattr*/
+    0,                  /*tp_compare*/
+    0,                  /*tp_repr*/
+    0,                  /*tp_as_number*/
+    0,                  /*tp_as_sequence*/
+    0,                  /*tp_as_mapping*/
+    0,                  /*tp_hash*/
+    0,                  /*tp_call*/
+    0,                  /*tp_str*/
+    0,                  /*tp_getattro*/
+    0,                  /*tp_setattro*/
+    0,                  /*tp_as_buffer*/
+    Py_TPFLAGS_DEFAULT, /*tp_flags*/
+    0,                  /*tp_doc*/
+    0,                  /*tp_traverse*/
+    0,			/*tp_clear*/
+    0,			/*tp_richcompare*/
+    0,			/*tp_weaklistoffset*/
+    0,			/*tp_iter*/
+    0,			/*tp_iternext*/
+    SHA_methods,	/* tp_methods */
+    SHA_members,	/* tp_members */
+    SHA_getseters,      /* tp_getset */
+};
+
+static PyTypeObject SHA256type = {
+    PyObject_HEAD_INIT(NULL)
+    0,			/*ob_size*/
+    "_sha256.sha256",	/*tp_name*/
+    sizeof(SHAobject),	/*tp_size*/
+    0,			/*tp_itemsize*/
+    /* methods */
+    SHA_dealloc,	/*tp_dealloc*/
+    0,			/*tp_print*/
+    0,          	/*tp_getattr*/
+    0,                  /*tp_setattr*/
+    0,                  /*tp_compare*/
+    0,                  /*tp_repr*/
+    0,                  /*tp_as_number*/
+    0,                  /*tp_as_sequence*/
+    0,                  /*tp_as_mapping*/
+    0,                  /*tp_hash*/
+    0,                  /*tp_call*/
+    0,                  /*tp_str*/
+    0,                  /*tp_getattro*/
+    0,                  /*tp_setattro*/
+    0,                  /*tp_as_buffer*/
+    Py_TPFLAGS_DEFAULT, /*tp_flags*/
+    0,                  /*tp_doc*/
+    0,                  /*tp_traverse*/
+    0,			/*tp_clear*/
+    0,			/*tp_richcompare*/
+    0,			/*tp_weaklistoffset*/
+    0,			/*tp_iter*/
+    0,			/*tp_iternext*/
+    SHA_methods,	/* tp_methods */
+    SHA_members,	/* tp_members */
+    SHA_getseters,      /* tp_getset */
+};
+
+
+/* The single module-level function: new() */
+
+PyDoc_STRVAR(SHA256_new__doc__,
+"Return a new SHA-256 hash object; optionally initialized with a string.");
+
+static PyObject *
+SHA256_new(PyObject *self, PyObject *args, PyObject *kwdict)
+{
+    static char *kwlist[] = {"string", NULL};
+    SHAobject *new;
+    unsigned char *cp = NULL;
+    int len;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwdict, "|s#:new", kwlist,
+                                     &cp, &len)) {
+        return NULL;
+    }
+
+    if ((new = newSHA256object()) == NULL)
+        return NULL;
+
+    sha_init(new);
+
+    if (PyErr_Occurred()) {
+        Py_DECREF(new);
+        return NULL;
+    }
+    if (cp)
+        sha_update(new, cp, len);
+
+    return (PyObject *)new;
+}
+
+PyDoc_STRVAR(SHA224_new__doc__,
+"Return a new SHA-224 hash object; optionally initialized with a string.");
+
+static PyObject *
+SHA224_new(PyObject *self, PyObject *args, PyObject *kwdict)
+{
+    static char *kwlist[] = {"string", NULL};
+    SHAobject *new;
+    unsigned char *cp = NULL;
+    int len;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwdict, "|s#:new", kwlist,
+                                     &cp, &len)) {
+        return NULL;
+    }
+
+    if ((new = newSHA224object()) == NULL)
+        return NULL;
+
+    sha224_init(new);
+
+    if (PyErr_Occurred()) {
+        Py_DECREF(new);
+        return NULL;
+    }
+    if (cp)
+        sha_update(new, cp, len);
+
+    return (PyObject *)new;
+}
+
+
+/* List of functions exported by this module */
+
+static struct PyMethodDef SHA_functions[] = {
+    {"sha256", (PyCFunction)SHA256_new, METH_VARARGS|METH_KEYWORDS, SHA256_new__doc__},
+    {"sha224", (PyCFunction)SHA224_new, METH_VARARGS|METH_KEYWORDS, SHA224_new__doc__},
+    {NULL,	NULL}		 /* Sentinel */
+};
+
+
+/* Initialize this module. */
+
+#define insint(n,v) { PyModule_AddIntConstant(m,n,v); }
+
+PyMODINIT_FUNC
+init_sha256(void)
+{
+    PyObject *m;
+
+    SHA224type.ob_type = &PyType_Type;
+    if (PyType_Ready(&SHA224type) < 0)
+        return;
+    SHA256type.ob_type = &PyType_Type;
+    if (PyType_Ready(&SHA256type) < 0)
+        return;
+    m = Py_InitModule("_sha256", SHA_functions);
+    if (m == NULL)
+	return;
+}
diff --git a/src/Modules/sha512module.c b/src/Modules/sha512module.c
new file mode 100644
--- /dev/null
+++ b/src/Modules/sha512module.c
@@ -0,0 +1,769 @@
+/* SHA512 module */
+
+/* This module provides an interface to NIST's SHA-512 and SHA-384 Algorithms */
+
+/* See below for information about the original code this module was
+   based upon. Additional work performed by:
+
+   Andrew Kuchling (amk at amk.ca)
+   Greg Stein (gstein at lyra.org)
+   Trevor Perrin (trevp at trevp.net)
+
+   Copyright (C) 2005   Gregory P. Smith (greg at krypto.org)
+   Licensed to PSF under a Contributor Agreement.
+
+*/
+
+/* SHA objects */
+
+#include "Python.h"
+#include "structmember.h"
+
+#ifdef PY_LONG_LONG /* If no PY_LONG_LONG, don't compile anything! */
+
+/* Endianness testing and definitions */
+#define TestEndianness(variable) {int i=1; variable=PCT_BIG_ENDIAN;\
+	if (*((char*)&i)==1) variable=PCT_LITTLE_ENDIAN;}
+
+#define PCT_LITTLE_ENDIAN 1
+#define PCT_BIG_ENDIAN 0
+
+/* Some useful types */
+
+typedef unsigned char SHA_BYTE;
+
+#if SIZEOF_INT == 4
+typedef unsigned int SHA_INT32;	/* 32-bit integer */
+typedef unsigned PY_LONG_LONG SHA_INT64;	/* 64-bit integer */
+#else
+/* not defined. compilation will die. */
+#endif
+
+/* The SHA block size and message digest sizes, in bytes */
+
+#define SHA_BLOCKSIZE   128
+#define SHA_DIGESTSIZE  64
+
+/* The structure for storing SHA info */
+
+typedef struct {
+    PyObject_HEAD
+    SHA_INT64 digest[8];		/* Message digest */
+    SHA_INT32 count_lo, count_hi;	/* 64-bit bit count */
+    SHA_BYTE data[SHA_BLOCKSIZE];	/* SHA data buffer */
+    int Endianness;
+    int local;				/* unprocessed amount in data */
+    int digestsize;
+} SHAobject;
+
+/* When run on a little-endian CPU we need to perform byte reversal on an
+   array of longwords. */
+
+static void longReverse(SHA_INT64 *buffer, int byteCount, int Endianness)
+{
+    SHA_INT64 value;
+
+    if ( Endianness == PCT_BIG_ENDIAN )
+	return;
+
+    byteCount /= sizeof(*buffer);
+    while (byteCount--) {
+        value = *buffer;
+
+		((unsigned char*)buffer)[0] = (unsigned char)(value >> 56) & 0xff;
+		((unsigned char*)buffer)[1] = (unsigned char)(value >> 48) & 0xff;
+		((unsigned char*)buffer)[2] = (unsigned char)(value >> 40) & 0xff;
+		((unsigned char*)buffer)[3] = (unsigned char)(value >> 32) & 0xff;
+		((unsigned char*)buffer)[4] = (unsigned char)(value >> 24) & 0xff;
+		((unsigned char*)buffer)[5] = (unsigned char)(value >> 16) & 0xff;
+		((unsigned char*)buffer)[6] = (unsigned char)(value >>  8) & 0xff;
+		((unsigned char*)buffer)[7] = (unsigned char)(value      ) & 0xff;
+        
+		buffer++;
+    }
+}
+
+static void SHAcopy(SHAobject *src, SHAobject *dest)
+{
+    dest->Endianness = src->Endianness;
+    dest->local = src->local;
+    dest->digestsize = src->digestsize;
+    dest->count_lo = src->count_lo;
+    dest->count_hi = src->count_hi;
+    memcpy(dest->digest, src->digest, sizeof(src->digest));
+    memcpy(dest->data, src->data, sizeof(src->data));
+}
+
+
+/* ------------------------------------------------------------------------
+ *
+ * This code for the SHA-512 algorithm was noted as public domain. The
+ * original headers are pasted below.
+ *
+ * Several changes have been made to make it more compatible with the
+ * Python environment and desired interface.
+ *
+ */
+
+/* LibTomCrypt, modular cryptographic library -- Tom St Denis
+ *
+ * LibTomCrypt is a library that provides various cryptographic
+ * algorithms in a highly modular and flexible manner.
+ *
+ * The library is free for all purposes without any express
+ * gurantee it works.
+ *
+ * Tom St Denis, tomstdenis at iahu.ca, http://libtomcrypt.org
+ */
+
+
+/* SHA512 by Tom St Denis */
+
+/* Various logical functions */
+#define ROR64(x, y) \
+    ( ((((x) & 0xFFFFFFFFFFFFFFFFULL)>>((unsigned PY_LONG_LONG)(y) & 63)) | \
+      ((x)<<((unsigned PY_LONG_LONG)(64-((y) & 63))))) & 0xFFFFFFFFFFFFFFFFULL)
+#define Ch(x,y,z)       (z ^ (x & (y ^ z)))
+#define Maj(x,y,z)      (((x | y) & z) | (x & y)) 
+#define S(x, n)         ROR64((x),(n))
+#define R(x, n)         (((x) & 0xFFFFFFFFFFFFFFFFULL) >> ((unsigned PY_LONG_LONG)n))
+#define Sigma0(x)       (S(x, 28) ^ S(x, 34) ^ S(x, 39))
+#define Sigma1(x)       (S(x, 14) ^ S(x, 18) ^ S(x, 41))
+#define Gamma0(x)       (S(x, 1) ^ S(x, 8) ^ R(x, 7))
+#define Gamma1(x)       (S(x, 19) ^ S(x, 61) ^ R(x, 6))
+
+
+static void
+sha512_transform(SHAobject *sha_info)
+{
+    int i;
+    SHA_INT64 S[8], W[80], t0, t1;
+
+    memcpy(W, sha_info->data, sizeof(sha_info->data));
+    longReverse(W, (int)sizeof(sha_info->data), sha_info->Endianness);
+
+    for (i = 16; i < 80; ++i) {
+		W[i] = Gamma1(W[i - 2]) + W[i - 7] + Gamma0(W[i - 15]) + W[i - 16];
+    }
+    for (i = 0; i < 8; ++i) {
+        S[i] = sha_info->digest[i];
+    }
+
+    /* Compress */
+#define RND(a,b,c,d,e,f,g,h,i,ki)                    \
+     t0 = h + Sigma1(e) + Ch(e, f, g) + ki + W[i];   \
+     t1 = Sigma0(a) + Maj(a, b, c);                  \
+     d += t0;                                        \
+     h  = t0 + t1;
+
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],0,0x428a2f98d728ae22ULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],1,0x7137449123ef65cdULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],2,0xb5c0fbcfec4d3b2fULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],3,0xe9b5dba58189dbbcULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],4,0x3956c25bf348b538ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],5,0x59f111f1b605d019ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],6,0x923f82a4af194f9bULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],7,0xab1c5ed5da6d8118ULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],8,0xd807aa98a3030242ULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],9,0x12835b0145706fbeULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],10,0x243185be4ee4b28cULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],11,0x550c7dc3d5ffb4e2ULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],12,0x72be5d74f27b896fULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],13,0x80deb1fe3b1696b1ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],14,0x9bdc06a725c71235ULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],15,0xc19bf174cf692694ULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],16,0xe49b69c19ef14ad2ULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],17,0xefbe4786384f25e3ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],18,0x0fc19dc68b8cd5b5ULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],19,0x240ca1cc77ac9c65ULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],20,0x2de92c6f592b0275ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],21,0x4a7484aa6ea6e483ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],22,0x5cb0a9dcbd41fbd4ULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],23,0x76f988da831153b5ULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],24,0x983e5152ee66dfabULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],25,0xa831c66d2db43210ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],26,0xb00327c898fb213fULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],27,0xbf597fc7beef0ee4ULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],28,0xc6e00bf33da88fc2ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],29,0xd5a79147930aa725ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],30,0x06ca6351e003826fULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],31,0x142929670a0e6e70ULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],32,0x27b70a8546d22ffcULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],33,0x2e1b21385c26c926ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],34,0x4d2c6dfc5ac42aedULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],35,0x53380d139d95b3dfULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],36,0x650a73548baf63deULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],37,0x766a0abb3c77b2a8ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],38,0x81c2c92e47edaee6ULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],39,0x92722c851482353bULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],40,0xa2bfe8a14cf10364ULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],41,0xa81a664bbc423001ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],42,0xc24b8b70d0f89791ULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],43,0xc76c51a30654be30ULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],44,0xd192e819d6ef5218ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],45,0xd69906245565a910ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],46,0xf40e35855771202aULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],47,0x106aa07032bbd1b8ULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],48,0x19a4c116b8d2d0c8ULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],49,0x1e376c085141ab53ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],50,0x2748774cdf8eeb99ULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],51,0x34b0bcb5e19b48a8ULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],52,0x391c0cb3c5c95a63ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],53,0x4ed8aa4ae3418acbULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],54,0x5b9cca4f7763e373ULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],55,0x682e6ff3d6b2b8a3ULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],56,0x748f82ee5defb2fcULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],57,0x78a5636f43172f60ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],58,0x84c87814a1f0ab72ULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],59,0x8cc702081a6439ecULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],60,0x90befffa23631e28ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],61,0xa4506cebde82bde9ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],62,0xbef9a3f7b2c67915ULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],63,0xc67178f2e372532bULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],64,0xca273eceea26619cULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],65,0xd186b8c721c0c207ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],66,0xeada7dd6cde0eb1eULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],67,0xf57d4f7fee6ed178ULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],68,0x06f067aa72176fbaULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],69,0x0a637dc5a2c898a6ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],70,0x113f9804bef90daeULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],71,0x1b710b35131c471bULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],72,0x28db77f523047d84ULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],73,0x32caab7b40c72493ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],74,0x3c9ebe0a15c9bebcULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],75,0x431d67c49c100d4cULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],76,0x4cc5d4becb3e42b6ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],77,0x597f299cfc657e2aULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],78,0x5fcb6fab3ad6faecULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],79,0x6c44198c4a475817ULL);
+
+#undef RND     
+    
+    /* feedback */
+    for (i = 0; i < 8; i++) {
+        sha_info->digest[i] = sha_info->digest[i] + S[i];
+    }
+
+}
+
+
+
+/* initialize the SHA digest */
+
+static void
+sha512_init(SHAobject *sha_info)
+{
+    TestEndianness(sha_info->Endianness)
+    sha_info->digest[0] = 0x6a09e667f3bcc908ULL;
+    sha_info->digest[1] = 0xbb67ae8584caa73bULL;
+    sha_info->digest[2] = 0x3c6ef372fe94f82bULL;
+    sha_info->digest[3] = 0xa54ff53a5f1d36f1ULL;
+    sha_info->digest[4] = 0x510e527fade682d1ULL;
+    sha_info->digest[5] = 0x9b05688c2b3e6c1fULL;
+    sha_info->digest[6] = 0x1f83d9abfb41bd6bULL;
+    sha_info->digest[7] = 0x5be0cd19137e2179ULL;
+    sha_info->count_lo = 0L;
+    sha_info->count_hi = 0L;
+    sha_info->local = 0;
+    sha_info->digestsize = 64;
+}
+
+static void
+sha384_init(SHAobject *sha_info)
+{
+    TestEndianness(sha_info->Endianness)
+    sha_info->digest[0] = 0xcbbb9d5dc1059ed8ULL;
+    sha_info->digest[1] = 0x629a292a367cd507ULL;
+    sha_info->digest[2] = 0x9159015a3070dd17ULL;
+    sha_info->digest[3] = 0x152fecd8f70e5939ULL;
+    sha_info->digest[4] = 0x67332667ffc00b31ULL;
+    sha_info->digest[5] = 0x8eb44a8768581511ULL;
+    sha_info->digest[6] = 0xdb0c2e0d64f98fa7ULL;
+    sha_info->digest[7] = 0x47b5481dbefa4fa4ULL;
+    sha_info->count_lo = 0L;
+    sha_info->count_hi = 0L;
+    sha_info->local = 0;
+    sha_info->digestsize = 48;
+}
+
+
+/* update the SHA digest */
+
+static void
+sha512_update(SHAobject *sha_info, SHA_BYTE *buffer, int count)
+{
+    int i;
+    SHA_INT32 clo;
+
+    clo = sha_info->count_lo + ((SHA_INT32) count << 3);
+    if (clo < sha_info->count_lo) {
+        ++sha_info->count_hi;
+    }
+    sha_info->count_lo = clo;
+    sha_info->count_hi += (SHA_INT32) count >> 29;
+    if (sha_info->local) {
+        i = SHA_BLOCKSIZE - sha_info->local;
+        if (i > count) {
+            i = count;
+        }
+        memcpy(((SHA_BYTE *) sha_info->data) + sha_info->local, buffer, i);
+        count -= i;
+        buffer += i;
+        sha_info->local += i;
+        if (sha_info->local == SHA_BLOCKSIZE) {
+            sha512_transform(sha_info);
+        }
+        else {
+            return;
+        }
+    }
+    while (count >= SHA_BLOCKSIZE) {
+        memcpy(sha_info->data, buffer, SHA_BLOCKSIZE);
+        buffer += SHA_BLOCKSIZE;
+        count -= SHA_BLOCKSIZE;
+        sha512_transform(sha_info);
+    }
+    memcpy(sha_info->data, buffer, count);
+    sha_info->local = count;
+}
+
+/* finish computing the SHA digest */
+
+static void
+sha512_final(unsigned char digest[SHA_DIGESTSIZE], SHAobject *sha_info)
+{
+    int count;
+    SHA_INT32 lo_bit_count, hi_bit_count;
+
+    lo_bit_count = sha_info->count_lo;
+    hi_bit_count = sha_info->count_hi;
+    count = (int) ((lo_bit_count >> 3) & 0x7f);
+    ((SHA_BYTE *) sha_info->data)[count++] = 0x80;
+    if (count > SHA_BLOCKSIZE - 16) {
+	memset(((SHA_BYTE *) sha_info->data) + count, 0,
+	       SHA_BLOCKSIZE - count);
+	sha512_transform(sha_info);
+	memset((SHA_BYTE *) sha_info->data, 0, SHA_BLOCKSIZE - 16);
+    }
+    else {
+	memset(((SHA_BYTE *) sha_info->data) + count, 0,
+	       SHA_BLOCKSIZE - 16 - count);
+    }
+
+    /* GJS: note that we add the hi/lo in big-endian. sha512_transform will
+       swap these values into host-order. */
+    sha_info->data[112] = 0;
+    sha_info->data[113] = 0;
+    sha_info->data[114] = 0;
+    sha_info->data[115] = 0;
+    sha_info->data[116] = 0;
+    sha_info->data[117] = 0;
+    sha_info->data[118] = 0;
+    sha_info->data[119] = 0;
+    sha_info->data[120] = (hi_bit_count >> 24) & 0xff;
+    sha_info->data[121] = (hi_bit_count >> 16) & 0xff;
+    sha_info->data[122] = (hi_bit_count >>  8) & 0xff;
+    sha_info->data[123] = (hi_bit_count >>  0) & 0xff;
+    sha_info->data[124] = (lo_bit_count >> 24) & 0xff;
+    sha_info->data[125] = (lo_bit_count >> 16) & 0xff;
+    sha_info->data[126] = (lo_bit_count >>  8) & 0xff;
+    sha_info->data[127] = (lo_bit_count >>  0) & 0xff;
+    sha512_transform(sha_info);
+    digest[ 0] = (unsigned char) ((sha_info->digest[0] >> 56) & 0xff);
+    digest[ 1] = (unsigned char) ((sha_info->digest[0] >> 48) & 0xff);
+    digest[ 2] = (unsigned char) ((sha_info->digest[0] >> 40) & 0xff);
+    digest[ 3] = (unsigned char) ((sha_info->digest[0] >> 32) & 0xff);
+    digest[ 4] = (unsigned char) ((sha_info->digest[0] >> 24) & 0xff);
+    digest[ 5] = (unsigned char) ((sha_info->digest[0] >> 16) & 0xff);
+    digest[ 6] = (unsigned char) ((sha_info->digest[0] >>  8) & 0xff);
+    digest[ 7] = (unsigned char) ((sha_info->digest[0]      ) & 0xff);
+    digest[ 8] = (unsigned char) ((sha_info->digest[1] >> 56) & 0xff);
+    digest[ 9] = (unsigned char) ((sha_info->digest[1] >> 48) & 0xff);
+    digest[10] = (unsigned char) ((sha_info->digest[1] >> 40) & 0xff);
+    digest[11] = (unsigned char) ((sha_info->digest[1] >> 32) & 0xff);
+    digest[12] = (unsigned char) ((sha_info->digest[1] >> 24) & 0xff);
+    digest[13] = (unsigned char) ((sha_info->digest[1] >> 16) & 0xff);
+    digest[14] = (unsigned char) ((sha_info->digest[1] >>  8) & 0xff);
+    digest[15] = (unsigned char) ((sha_info->digest[1]      ) & 0xff);
+    digest[16] = (unsigned char) ((sha_info->digest[2] >> 56) & 0xff);
+    digest[17] = (unsigned char) ((sha_info->digest[2] >> 48) & 0xff);
+    digest[18] = (unsigned char) ((sha_info->digest[2] >> 40) & 0xff);
+    digest[19] = (unsigned char) ((sha_info->digest[2] >> 32) & 0xff);
+    digest[20] = (unsigned char) ((sha_info->digest[2] >> 24) & 0xff);
+    digest[21] = (unsigned char) ((sha_info->digest[2] >> 16) & 0xff);
+    digest[22] = (unsigned char) ((sha_info->digest[2] >>  8) & 0xff);
+    digest[23] = (unsigned char) ((sha_info->digest[2]      ) & 0xff);
+    digest[24] = (unsigned char) ((sha_info->digest[3] >> 56) & 0xff);
+    digest[25] = (unsigned char) ((sha_info->digest[3] >> 48) & 0xff);
+    digest[26] = (unsigned char) ((sha_info->digest[3] >> 40) & 0xff);
+    digest[27] = (unsigned char) ((sha_info->digest[3] >> 32) & 0xff);
+    digest[28] = (unsigned char) ((sha_info->digest[3] >> 24) & 0xff);
+    digest[29] = (unsigned char) ((sha_info->digest[3] >> 16) & 0xff);
+    digest[30] = (unsigned char) ((sha_info->digest[3] >>  8) & 0xff);
+    digest[31] = (unsigned char) ((sha_info->digest[3]      ) & 0xff);
+    digest[32] = (unsigned char) ((sha_info->digest[4] >> 56) & 0xff);
+    digest[33] = (unsigned char) ((sha_info->digest[4] >> 48) & 0xff);
+    digest[34] = (unsigned char) ((sha_info->digest[4] >> 40) & 0xff);
+    digest[35] = (unsigned char) ((sha_info->digest[4] >> 32) & 0xff);
+    digest[36] = (unsigned char) ((sha_info->digest[4] >> 24) & 0xff);
+    digest[37] = (unsigned char) ((sha_info->digest[4] >> 16) & 0xff);
+    digest[38] = (unsigned char) ((sha_info->digest[4] >>  8) & 0xff);
+    digest[39] = (unsigned char) ((sha_info->digest[4]      ) & 0xff);
+    digest[40] = (unsigned char) ((sha_info->digest[5] >> 56) & 0xff);
+    digest[41] = (unsigned char) ((sha_info->digest[5] >> 48) & 0xff);
+    digest[42] = (unsigned char) ((sha_info->digest[5] >> 40) & 0xff);
+    digest[43] = (unsigned char) ((sha_info->digest[5] >> 32) & 0xff);
+    digest[44] = (unsigned char) ((sha_info->digest[5] >> 24) & 0xff);
+    digest[45] = (unsigned char) ((sha_info->digest[5] >> 16) & 0xff);
+    digest[46] = (unsigned char) ((sha_info->digest[5] >>  8) & 0xff);
+    digest[47] = (unsigned char) ((sha_info->digest[5]      ) & 0xff);
+    digest[48] = (unsigned char) ((sha_info->digest[6] >> 56) & 0xff);
+    digest[49] = (unsigned char) ((sha_info->digest[6] >> 48) & 0xff);
+    digest[50] = (unsigned char) ((sha_info->digest[6] >> 40) & 0xff);
+    digest[51] = (unsigned char) ((sha_info->digest[6] >> 32) & 0xff);
+    digest[52] = (unsigned char) ((sha_info->digest[6] >> 24) & 0xff);
+    digest[53] = (unsigned char) ((sha_info->digest[6] >> 16) & 0xff);
+    digest[54] = (unsigned char) ((sha_info->digest[6] >>  8) & 0xff);
+    digest[55] = (unsigned char) ((sha_info->digest[6]      ) & 0xff);
+    digest[56] = (unsigned char) ((sha_info->digest[7] >> 56) & 0xff);
+    digest[57] = (unsigned char) ((sha_info->digest[7] >> 48) & 0xff);
+    digest[58] = (unsigned char) ((sha_info->digest[7] >> 40) & 0xff);
+    digest[59] = (unsigned char) ((sha_info->digest[7] >> 32) & 0xff);
+    digest[60] = (unsigned char) ((sha_info->digest[7] >> 24) & 0xff);
+    digest[61] = (unsigned char) ((sha_info->digest[7] >> 16) & 0xff);
+    digest[62] = (unsigned char) ((sha_info->digest[7] >>  8) & 0xff);
+    digest[63] = (unsigned char) ((sha_info->digest[7]      ) & 0xff);
+}
+
+/*
+ * End of copied SHA code.
+ *
+ * ------------------------------------------------------------------------
+ */
+
+static PyTypeObject SHA384type;
+static PyTypeObject SHA512type;
+
+
+static SHAobject *
+newSHA384object(void)
+{
+    return (SHAobject *)PyObject_New(SHAobject, &SHA384type);
+}
+
+static SHAobject *
+newSHA512object(void)
+{
+    return (SHAobject *)PyObject_New(SHAobject, &SHA512type);
+}
+
+/* Internal methods for a hash object */
+
+static void
+SHA512_dealloc(PyObject *ptr)
+{
+    PyObject_Del(ptr);
+}
+
+
+/* External methods for a hash object */
+
+PyDoc_STRVAR(SHA512_copy__doc__, "Return a copy of the hash object.");
+
+static PyObject *
+SHA512_copy(SHAobject *self, PyObject *unused)
+{
+    SHAobject *newobj;
+
+    if (((PyObject*)self)->ob_type == &SHA512type) {
+        if ( (newobj = newSHA512object())==NULL)
+            return NULL;
+    } else {
+        if ( (newobj = newSHA384object())==NULL)
+            return NULL;
+    }
+
+    SHAcopy(self, newobj);
+    return (PyObject *)newobj;
+}
+
+PyDoc_STRVAR(SHA512_digest__doc__,
+"Return the digest value as a string of binary data.");
+
+static PyObject *
+SHA512_digest(SHAobject *self, PyObject *unused)
+{
+    unsigned char digest[SHA_DIGESTSIZE];
+    SHAobject temp;
+
+    SHAcopy(self, &temp);
+    sha512_final(digest, &temp);
+    return PyString_FromStringAndSize((const char *)digest, self->digestsize);
+}
+
+PyDoc_STRVAR(SHA512_hexdigest__doc__,
+"Return the digest value as a string of hexadecimal digits.");
+
+static PyObject *
+SHA512_hexdigest(SHAobject *self, PyObject *unused)
+{
+    unsigned char digest[SHA_DIGESTSIZE];
+    SHAobject temp;
+    PyObject *retval;
+    char *hex_digest;
+    int i, j;
+
+    /* Get the raw (binary) digest value */
+    SHAcopy(self, &temp);
+    sha512_final(digest, &temp);
+
+    /* Create a new string */
+    retval = PyString_FromStringAndSize(NULL, self->digestsize * 2);
+    if (!retval)
+	    return NULL;
+    hex_digest = PyString_AsString(retval);
+    if (!hex_digest) {
+	    Py_DECREF(retval);
+	    return NULL;
+    }
+
+    /* Make hex version of the digest */
+    for (i=j=0; i<self->digestsize; i++) {
+        char c;
+        c = (digest[i] >> 4) & 0xf;
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+        c = (digest[i] & 0xf);
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+    }
+    return retval;
+}
+
+PyDoc_STRVAR(SHA512_update__doc__,
+"Update this hash object's state with the provided string.");
+
+static PyObject *
+SHA512_update(SHAobject *self, PyObject *args)
+{
+    unsigned char *cp;
+    int len;
+
+    if (!PyArg_ParseTuple(args, "s#:update", &cp, &len))
+        return NULL;
+
+    sha512_update(self, cp, len);
+
+    Py_INCREF(Py_None);
+    return Py_None;
+}
+
+static PyMethodDef SHA_methods[] = {
+    {"copy",	  (PyCFunction)SHA512_copy,      METH_NOARGS, SHA512_copy__doc__},
+    {"digest",	  (PyCFunction)SHA512_digest,    METH_NOARGS, SHA512_digest__doc__},
+    {"hexdigest", (PyCFunction)SHA512_hexdigest, METH_NOARGS, SHA512_hexdigest__doc__},
+    {"update",	  (PyCFunction)SHA512_update,    METH_VARARGS, SHA512_update__doc__},
+    {NULL,	  NULL}		/* sentinel */
+};
+
+static PyObject *
+SHA512_get_block_size(PyObject *self, void *closure)
+{
+    return PyInt_FromLong(SHA_BLOCKSIZE);
+}
+
+static PyObject *
+SHA512_get_name(PyObject *self, void *closure)
+{
+    if (((SHAobject *)self)->digestsize == 64)
+        return PyString_FromStringAndSize("SHA512", 6);
+    else
+        return PyString_FromStringAndSize("SHA384", 6);
+}
+
+static PyGetSetDef SHA_getseters[] = {
+    {"block_size",
+     (getter)SHA512_get_block_size, NULL,
+     NULL,
+     NULL},
+    {"name",
+     (getter)SHA512_get_name, NULL,
+     NULL,
+     NULL},
+    {NULL}  /* Sentinel */
+};
+
+static PyMemberDef SHA_members[] = {
+    {"digest_size", T_INT, offsetof(SHAobject, digestsize), READONLY, NULL},
+    /* the old md5 and sha modules support 'digest_size' as in PEP 247.
+     * the old sha module also supported 'digestsize'.  ugh. */
+    {"digestsize", T_INT, offsetof(SHAobject, digestsize), READONLY, NULL},
+    {NULL}  /* Sentinel */
+};
+
+static PyTypeObject SHA384type = {
+    PyObject_HEAD_INIT(NULL)
+    0,			/*ob_size*/
+    "_sha512.sha384",	/*tp_name*/
+    sizeof(SHAobject),	/*tp_size*/
+    0,			/*tp_itemsize*/
+    /* methods */
+    SHA512_dealloc,	/*tp_dealloc*/
+    0,			/*tp_print*/
+    0,          	/*tp_getattr*/
+    0,                  /*tp_setattr*/
+    0,                  /*tp_compare*/
+    0,                  /*tp_repr*/
+    0,                  /*tp_as_number*/
+    0,                  /*tp_as_sequence*/
+    0,                  /*tp_as_mapping*/
+    0,                  /*tp_hash*/
+    0,                  /*tp_call*/
+    0,                  /*tp_str*/
+    0,                  /*tp_getattro*/
+    0,                  /*tp_setattro*/
+    0,                  /*tp_as_buffer*/
+    Py_TPFLAGS_DEFAULT, /*tp_flags*/
+    0,                  /*tp_doc*/
+    0,                  /*tp_traverse*/
+    0,			/*tp_clear*/
+    0,			/*tp_richcompare*/
+    0,			/*tp_weaklistoffset*/
+    0,			/*tp_iter*/
+    0,			/*tp_iternext*/
+    SHA_methods,	/* tp_methods */
+    SHA_members,	/* tp_members */
+    SHA_getseters,      /* tp_getset */
+};
+
+static PyTypeObject SHA512type = {
+    PyObject_HEAD_INIT(NULL)
+    0,			/*ob_size*/
+    "_sha512.sha512",	/*tp_name*/
+    sizeof(SHAobject),	/*tp_size*/
+    0,			/*tp_itemsize*/
+    /* methods */
+    SHA512_dealloc,	/*tp_dealloc*/
+    0,			/*tp_print*/
+    0,          	/*tp_getattr*/
+    0,                  /*tp_setattr*/
+    0,                  /*tp_compare*/
+    0,                  /*tp_repr*/
+    0,                  /*tp_as_number*/
+    0,                  /*tp_as_sequence*/
+    0,                  /*tp_as_mapping*/
+    0,                  /*tp_hash*/
+    0,                  /*tp_call*/
+    0,                  /*tp_str*/
+    0,                  /*tp_getattro*/
+    0,                  /*tp_setattro*/
+    0,                  /*tp_as_buffer*/
+    Py_TPFLAGS_DEFAULT, /*tp_flags*/
+    0,                  /*tp_doc*/
+    0,                  /*tp_traverse*/
+    0,			/*tp_clear*/
+    0,			/*tp_richcompare*/
+    0,			/*tp_weaklistoffset*/
+    0,			/*tp_iter*/
+    0,			/*tp_iternext*/
+    SHA_methods,	/* tp_methods */
+    SHA_members,	/* tp_members */
+    SHA_getseters,      /* tp_getset */
+};
+
+
+/* The single module-level function: new() */
+
+PyDoc_STRVAR(SHA512_new__doc__,
+"Return a new SHA-512 hash object; optionally initialized with a string.");
+
+static PyObject *
+SHA512_new(PyObject *self, PyObject *args, PyObject *kwdict)
+{
+    static char *kwlist[] = {"string", NULL};
+    SHAobject *new;
+    unsigned char *cp = NULL;
+    int len;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwdict, "|s#:new", kwlist,
+                                     &cp, &len)) {
+        return NULL;
+    }
+
+    if ((new = newSHA512object()) == NULL)
+        return NULL;
+
+    sha512_init(new);
+
+    if (PyErr_Occurred()) {
+        Py_DECREF(new);
+        return NULL;
+    }
+    if (cp)
+        sha512_update(new, cp, len);
+
+    return (PyObject *)new;
+}
+
+PyDoc_STRVAR(SHA384_new__doc__,
+"Return a new SHA-384 hash object; optionally initialized with a string.");
+
+static PyObject *
+SHA384_new(PyObject *self, PyObject *args, PyObject *kwdict)
+{
+    static char *kwlist[] = {"string", NULL};
+    SHAobject *new;
+    unsigned char *cp = NULL;
+    int len;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwdict, "|s#:new", kwlist,
+                                     &cp, &len)) {
+        return NULL;
+    }
+
+    if ((new = newSHA384object()) == NULL)
+        return NULL;
+
+    sha384_init(new);
+
+    if (PyErr_Occurred()) {
+        Py_DECREF(new);
+        return NULL;
+    }
+    if (cp)
+        sha512_update(new, cp, len);
+
+    return (PyObject *)new;
+}
+
+
+/* List of functions exported by this module */
+
+static struct PyMethodDef SHA_functions[] = {
+    {"sha512", (PyCFunction)SHA512_new, METH_VARARGS|METH_KEYWORDS, SHA512_new__doc__},
+    {"sha384", (PyCFunction)SHA384_new, METH_VARARGS|METH_KEYWORDS, SHA384_new__doc__},
+    {NULL,	NULL}		 /* Sentinel */
+};
+
+
+/* Initialize this module. */
+
+#define insint(n,v) { PyModule_AddIntConstant(m,n,v); }
+
+PyMODINIT_FUNC
+init_sha512(void)
+{
+    PyObject *m;
+
+    SHA384type.ob_type = &PyType_Type;
+    if (PyType_Ready(&SHA384type) < 0)
+        return;
+    SHA512type.ob_type = &PyType_Type;
+    if (PyType_Ready(&SHA512type) < 0)
+        return;
+    m = Py_InitModule("_sha512", SHA_functions);
+    if (m == NULL)
+	return;
+}
+
+#endif
diff --git a/src/Modules/shamodule.c b/src/Modules/shamodule.c
new file mode 100644
--- /dev/null
+++ b/src/Modules/shamodule.c
@@ -0,0 +1,593 @@
+/* SHA module */
+
+/* This module provides an interface to NIST's Secure Hash Algorithm */
+
+/* See below for information about the original code this module was
+   based upon. Additional work performed by:
+
+   Andrew Kuchling (amk at amk.ca)
+   Greg Stein (gstein at lyra.org)
+
+   Copyright (C) 2005   Gregory P. Smith (greg at krypto.org)
+   Licensed to PSF under a Contributor Agreement.
+
+*/
+
+/* SHA objects */
+
+#include "Python.h"
+#include "structmember.h"
+
+
+/* Endianness testing and definitions */
+#define TestEndianness(variable) {int i=1; variable=PCT_BIG_ENDIAN;\
+	if (*((char*)&i)==1) variable=PCT_LITTLE_ENDIAN;}
+
+#define PCT_LITTLE_ENDIAN 1
+#define PCT_BIG_ENDIAN 0
+
+/* Some useful types */
+
+typedef unsigned char SHA_BYTE;
+
+#if SIZEOF_INT == 4
+typedef unsigned int SHA_INT32;	/* 32-bit integer */
+#else
+/* not defined. compilation will die. */
+#endif
+
+/* The SHA block size and message digest sizes, in bytes */
+
+#define SHA_BLOCKSIZE    64
+#define SHA_DIGESTSIZE  20
+
+/* The structure for storing SHS info */
+
+typedef struct {
+    PyObject_HEAD
+    SHA_INT32 digest[5];		/* Message digest */
+    SHA_INT32 count_lo, count_hi;	/* 64-bit bit count */
+    SHA_BYTE data[SHA_BLOCKSIZE];	/* SHA data buffer */
+    int Endianness;
+    int local;				/* unprocessed amount in data */
+} SHAobject;
+
+/* When run on a little-endian CPU we need to perform byte reversal on an
+   array of longwords. */
+
+static void longReverse(SHA_INT32 *buffer, int byteCount, int Endianness)
+{
+    SHA_INT32 value;
+
+    if ( Endianness == PCT_BIG_ENDIAN )
+	return;
+
+    byteCount /= sizeof(*buffer);
+    while (byteCount--) {
+        value = *buffer;
+        value = ( ( value & 0xFF00FF00L ) >> 8  ) | \
+                ( ( value & 0x00FF00FFL ) << 8 );
+        *buffer++ = ( value << 16 ) | ( value >> 16 );
+    }
+}
+
+static void SHAcopy(SHAobject *src, SHAobject *dest)
+{
+    dest->Endianness = src->Endianness;
+    dest->local = src->local;
+    dest->count_lo = src->count_lo;
+    dest->count_hi = src->count_hi;
+    memcpy(dest->digest, src->digest, sizeof(src->digest));
+    memcpy(dest->data, src->data, sizeof(src->data));
+}
+
+
+/* ------------------------------------------------------------------------
+ *
+ * This code for the SHA algorithm was noted as public domain. The original
+ * headers are pasted below.
+ *
+ * Several changes have been made to make it more compatible with the
+ * Python environment and desired interface.
+ *
+ */
+
+/* NIST Secure Hash Algorithm */
+/* heavily modified by Uwe Hollerbach <uh at alumni.caltech edu> */
+/* from Peter C. Gutmann's implementation as found in */
+/* Applied Cryptography by Bruce Schneier */
+/* Further modifications to include the "UNRAVEL" stuff, below */
+
+/* This code is in the public domain */
+
+/* UNRAVEL should be fastest & biggest */
+/* UNROLL_LOOPS should be just as big, but slightly slower */
+/* both undefined should be smallest and slowest */
+
+#define UNRAVEL
+/* #define UNROLL_LOOPS */
+
+/* The SHA f()-functions.  The f1 and f3 functions can be optimized to
+   save one boolean operation each - thanks to Rich Schroeppel,
+   rcs at cs.arizona.edu for discovering this */
+
+/*#define f1(x,y,z)	((x & y) | (~x & z))		// Rounds  0-19 */
+#define f1(x,y,z)	(z ^ (x & (y ^ z)))		/* Rounds  0-19 */
+#define f2(x,y,z)	(x ^ y ^ z)			/* Rounds 20-39 */
+/*#define f3(x,y,z)	((x & y) | (x & z) | (y & z))	// Rounds 40-59 */
+#define f3(x,y,z)	((x & y) | (z & (x | y)))	/* Rounds 40-59 */
+#define f4(x,y,z)	(x ^ y ^ z)			/* Rounds 60-79 */
+
+/* SHA constants */
+
+#define CONST1		0x5a827999L			/* Rounds  0-19 */
+#define CONST2		0x6ed9eba1L			/* Rounds 20-39 */
+#define CONST3		0x8f1bbcdcL			/* Rounds 40-59 */
+#define CONST4		0xca62c1d6L			/* Rounds 60-79 */
+
+/* 32-bit rotate */
+
+#define R32(x,n)	((x << n) | (x >> (32 - n)))
+
+/* the generic case, for when the overall rotation is not unraveled */
+
+#define FG(n)	\
+    T = R32(A,5) + f##n(B,C,D) + E + *WP++ + CONST##n;	\
+    E = D; D = C; C = R32(B,30); B = A; A = T
+
+/* specific cases, for when the overall rotation is unraveled */
+
+#define FA(n)	\
+    T = R32(A,5) + f##n(B,C,D) + E + *WP++ + CONST##n; B = R32(B,30)
+
+#define FB(n)	\
+    E = R32(T,5) + f##n(A,B,C) + D + *WP++ + CONST##n; A = R32(A,30)
+
+#define FC(n)	\
+    D = R32(E,5) + f##n(T,A,B) + C + *WP++ + CONST##n; T = R32(T,30)
+
+#define FD(n)	\
+    C = R32(D,5) + f##n(E,T,A) + B + *WP++ + CONST##n; E = R32(E,30)
+
+#define FE(n)	\
+    B = R32(C,5) + f##n(D,E,T) + A + *WP++ + CONST##n; D = R32(D,30)
+
+#define FT(n)	\
+    A = R32(B,5) + f##n(C,D,E) + T + *WP++ + CONST##n; C = R32(C,30)
+
+/* do SHA transformation */
+
+static void
+sha_transform(SHAobject *sha_info)
+{
+    int i;
+    SHA_INT32 T, A, B, C, D, E, W[80], *WP;
+
+    memcpy(W, sha_info->data, sizeof(sha_info->data));
+    longReverse(W, (int)sizeof(sha_info->data), sha_info->Endianness);
+
+    for (i = 16; i < 80; ++i) {
+	W[i] = W[i-3] ^ W[i-8] ^ W[i-14] ^ W[i-16];
+
+	/* extra rotation fix */
+	W[i] = R32(W[i], 1);
+    }
+    A = sha_info->digest[0];
+    B = sha_info->digest[1];
+    C = sha_info->digest[2];
+    D = sha_info->digest[3];
+    E = sha_info->digest[4];
+    WP = W;
+#ifdef UNRAVEL
+    FA(1); FB(1); FC(1); FD(1); FE(1); FT(1); FA(1); FB(1); FC(1); FD(1);
+    FE(1); FT(1); FA(1); FB(1); FC(1); FD(1); FE(1); FT(1); FA(1); FB(1);
+    FC(2); FD(2); FE(2); FT(2); FA(2); FB(2); FC(2); FD(2); FE(2); FT(2);
+    FA(2); FB(2); FC(2); FD(2); FE(2); FT(2); FA(2); FB(2); FC(2); FD(2);
+    FE(3); FT(3); FA(3); FB(3); FC(3); FD(3); FE(3); FT(3); FA(3); FB(3);
+    FC(3); FD(3); FE(3); FT(3); FA(3); FB(3); FC(3); FD(3); FE(3); FT(3);
+    FA(4); FB(4); FC(4); FD(4); FE(4); FT(4); FA(4); FB(4); FC(4); FD(4);
+    FE(4); FT(4); FA(4); FB(4); FC(4); FD(4); FE(4); FT(4); FA(4); FB(4);
+    sha_info->digest[0] += E;
+    sha_info->digest[1] += T;
+    sha_info->digest[2] += A;
+    sha_info->digest[3] += B;
+    sha_info->digest[4] += C;
+#else /* !UNRAVEL */
+#ifdef UNROLL_LOOPS
+    FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1);
+    FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1);
+    FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2);
+    FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2);
+    FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3);
+    FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3);
+    FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4);
+    FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4);
+#else /* !UNROLL_LOOPS */
+    for (i =  0; i < 20; ++i) { FG(1); }
+    for (i = 20; i < 40; ++i) { FG(2); }
+    for (i = 40; i < 60; ++i) { FG(3); }
+    for (i = 60; i < 80; ++i) { FG(4); }
+#endif /* !UNROLL_LOOPS */
+    sha_info->digest[0] += A;
+    sha_info->digest[1] += B;
+    sha_info->digest[2] += C;
+    sha_info->digest[3] += D;
+    sha_info->digest[4] += E;
+#endif /* !UNRAVEL */
+}
+
+/* initialize the SHA digest */
+
+static void
+sha_init(SHAobject *sha_info)
+{
+    TestEndianness(sha_info->Endianness)
+
+    sha_info->digest[0] = 0x67452301L;
+    sha_info->digest[1] = 0xefcdab89L;
+    sha_info->digest[2] = 0x98badcfeL;
+    sha_info->digest[3] = 0x10325476L;
+    sha_info->digest[4] = 0xc3d2e1f0L;
+    sha_info->count_lo = 0L;
+    sha_info->count_hi = 0L;
+    sha_info->local = 0;
+}
+
+/* update the SHA digest */
+
+static void
+sha_update(SHAobject *sha_info, SHA_BYTE *buffer, int count)
+{
+    int i;
+    SHA_INT32 clo;
+
+    clo = sha_info->count_lo + ((SHA_INT32) count << 3);
+    if (clo < sha_info->count_lo) {
+        ++sha_info->count_hi;
+    }
+    sha_info->count_lo = clo;
+    sha_info->count_hi += (SHA_INT32) count >> 29;
+    if (sha_info->local) {
+        i = SHA_BLOCKSIZE - sha_info->local;
+        if (i > count) {
+            i = count;
+        }
+        memcpy(((SHA_BYTE *) sha_info->data) + sha_info->local, buffer, i);
+        count -= i;
+        buffer += i;
+        sha_info->local += i;
+        if (sha_info->local == SHA_BLOCKSIZE) {
+            sha_transform(sha_info);
+        }
+        else {
+            return;
+        }
+    }
+    while (count >= SHA_BLOCKSIZE) {
+        memcpy(sha_info->data, buffer, SHA_BLOCKSIZE);
+        buffer += SHA_BLOCKSIZE;
+        count -= SHA_BLOCKSIZE;
+        sha_transform(sha_info);
+    }
+    memcpy(sha_info->data, buffer, count);
+    sha_info->local = count;
+}
+
+/* finish computing the SHA digest */
+
+static void
+sha_final(unsigned char digest[20], SHAobject *sha_info)
+{
+    int count;
+    SHA_INT32 lo_bit_count, hi_bit_count;
+
+    lo_bit_count = sha_info->count_lo;
+    hi_bit_count = sha_info->count_hi;
+    count = (int) ((lo_bit_count >> 3) & 0x3f);
+    ((SHA_BYTE *) sha_info->data)[count++] = 0x80;
+    if (count > SHA_BLOCKSIZE - 8) {
+	memset(((SHA_BYTE *) sha_info->data) + count, 0,
+	       SHA_BLOCKSIZE - count);
+	sha_transform(sha_info);
+	memset((SHA_BYTE *) sha_info->data, 0, SHA_BLOCKSIZE - 8);
+    }
+    else {
+	memset(((SHA_BYTE *) sha_info->data) + count, 0,
+	       SHA_BLOCKSIZE - 8 - count);
+    }
+
+    /* GJS: note that we add the hi/lo in big-endian. sha_transform will
+       swap these values into host-order. */
+    sha_info->data[56] = (hi_bit_count >> 24) & 0xff;
+    sha_info->data[57] = (hi_bit_count >> 16) & 0xff;
+    sha_info->data[58] = (hi_bit_count >>  8) & 0xff;
+    sha_info->data[59] = (hi_bit_count >>  0) & 0xff;
+    sha_info->data[60] = (lo_bit_count >> 24) & 0xff;
+    sha_info->data[61] = (lo_bit_count >> 16) & 0xff;
+    sha_info->data[62] = (lo_bit_count >>  8) & 0xff;
+    sha_info->data[63] = (lo_bit_count >>  0) & 0xff;
+    sha_transform(sha_info);
+    digest[ 0] = (unsigned char) ((sha_info->digest[0] >> 24) & 0xff);
+    digest[ 1] = (unsigned char) ((sha_info->digest[0] >> 16) & 0xff);
+    digest[ 2] = (unsigned char) ((sha_info->digest[0] >>  8) & 0xff);
+    digest[ 3] = (unsigned char) ((sha_info->digest[0]      ) & 0xff);
+    digest[ 4] = (unsigned char) ((sha_info->digest[1] >> 24) & 0xff);
+    digest[ 5] = (unsigned char) ((sha_info->digest[1] >> 16) & 0xff);
+    digest[ 6] = (unsigned char) ((sha_info->digest[1] >>  8) & 0xff);
+    digest[ 7] = (unsigned char) ((sha_info->digest[1]      ) & 0xff);
+    digest[ 8] = (unsigned char) ((sha_info->digest[2] >> 24) & 0xff);
+    digest[ 9] = (unsigned char) ((sha_info->digest[2] >> 16) & 0xff);
+    digest[10] = (unsigned char) ((sha_info->digest[2] >>  8) & 0xff);
+    digest[11] = (unsigned char) ((sha_info->digest[2]      ) & 0xff);
+    digest[12] = (unsigned char) ((sha_info->digest[3] >> 24) & 0xff);
+    digest[13] = (unsigned char) ((sha_info->digest[3] >> 16) & 0xff);
+    digest[14] = (unsigned char) ((sha_info->digest[3] >>  8) & 0xff);
+    digest[15] = (unsigned char) ((sha_info->digest[3]      ) & 0xff);
+    digest[16] = (unsigned char) ((sha_info->digest[4] >> 24) & 0xff);
+    digest[17] = (unsigned char) ((sha_info->digest[4] >> 16) & 0xff);
+    digest[18] = (unsigned char) ((sha_info->digest[4] >>  8) & 0xff);
+    digest[19] = (unsigned char) ((sha_info->digest[4]      ) & 0xff);
+}
+
+/*
+ * End of copied SHA code.
+ *
+ * ------------------------------------------------------------------------
+ */
+
+static PyTypeObject SHAtype;
+
+
+static SHAobject *
+newSHAobject(void)
+{
+    return (SHAobject *)PyObject_New(SHAobject, &SHAtype);
+}
+
+/* Internal methods for a hashing object */
+
+static void
+SHA_dealloc(PyObject *ptr)
+{
+    PyObject_Del(ptr);
+}
+
+
+/* External methods for a hashing object */
+
+PyDoc_STRVAR(SHA_copy__doc__, "Return a copy of the hashing object.");
+
+static PyObject *
+SHA_copy(SHAobject *self, PyObject *unused)
+{
+    SHAobject *newobj;
+
+    if ( (newobj = newSHAobject())==NULL)
+        return NULL;
+
+    SHAcopy(self, newobj);
+    return (PyObject *)newobj;
+}
+
+PyDoc_STRVAR(SHA_digest__doc__,
+"Return the digest value as a string of binary data.");
+
+static PyObject *
+SHA_digest(SHAobject *self, PyObject *unused)
+{
+    unsigned char digest[SHA_DIGESTSIZE];
+    SHAobject temp;
+
+    SHAcopy(self, &temp);
+    sha_final(digest, &temp);
+    return PyString_FromStringAndSize((const char *)digest, sizeof(digest));
+}
+
+PyDoc_STRVAR(SHA_hexdigest__doc__,
+"Return the digest value as a string of hexadecimal digits.");
+
+static PyObject *
+SHA_hexdigest(SHAobject *self, PyObject *unused)
+{
+    unsigned char digest[SHA_DIGESTSIZE];
+    SHAobject temp;
+    PyObject *retval;
+    char *hex_digest;
+    int i, j;
+
+    /* Get the raw (binary) digest value */
+    SHAcopy(self, &temp);
+    sha_final(digest, &temp);
+
+    /* Create a new string */
+    retval = PyString_FromStringAndSize(NULL, sizeof(digest) * 2);
+    if (!retval)
+	    return NULL;
+    hex_digest = PyString_AsString(retval);
+    if (!hex_digest) {
+	    Py_DECREF(retval);
+	    return NULL;
+    }
+
+    /* Make hex version of the digest */
+    for(i=j=0; i<sizeof(digest); i++) {
+        char c;
+        c = (digest[i] >> 4) & 0xf;
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+        c = (digest[i] & 0xf);
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+    }
+    return retval;
+}
+
+PyDoc_STRVAR(SHA_update__doc__,
+"Update this hashing object's state with the provided string.");
+
+static PyObject *
+SHA_update(SHAobject *self, PyObject *args)
+{
+    unsigned char *cp;
+    int len;
+
+    if (!PyArg_ParseTuple(args, "s#:update", &cp, &len))
+        return NULL;
+
+    sha_update(self, cp, len);
+
+    Py_INCREF(Py_None);
+    return Py_None;
+}
+
+static PyMethodDef SHA_methods[] = {
+    {"copy",	  (PyCFunction)SHA_copy,      METH_NOARGS,  SHA_copy__doc__},
+    {"digest",	  (PyCFunction)SHA_digest,    METH_NOARGS,  SHA_digest__doc__},
+    {"hexdigest", (PyCFunction)SHA_hexdigest, METH_NOARGS,  SHA_hexdigest__doc__},
+    {"update",	  (PyCFunction)SHA_update,    METH_VARARGS, SHA_update__doc__},
+    {NULL,	  NULL}		/* sentinel */
+};
+
+static PyObject *
+SHA_get_block_size(PyObject *self, void *closure)
+{
+    return PyInt_FromLong(SHA_BLOCKSIZE);
+}
+
+static PyObject *
+SHA_get_digest_size(PyObject *self, void *closure)
+{
+    return PyInt_FromLong(SHA_DIGESTSIZE);
+}
+
+static PyObject *
+SHA_get_name(PyObject *self, void *closure)
+{
+    return PyString_FromStringAndSize("SHA1", 4);
+}
+
+static PyGetSetDef SHA_getseters[] = {
+    {"digest_size",
+     (getter)SHA_get_digest_size, NULL,
+     NULL,
+     NULL},
+    {"block_size",
+     (getter)SHA_get_block_size, NULL,
+     NULL,
+     NULL},
+    {"name",
+     (getter)SHA_get_name, NULL,
+     NULL,
+     NULL},
+    /* the old md5 and sha modules support 'digest_size' as in PEP 247.
+     * the old sha module also supported 'digestsize'.  ugh. */
+    {"digestsize",
+     (getter)SHA_get_digest_size, NULL,
+     NULL,
+     NULL},
+    {NULL}  /* Sentinel */
+};
+
+static PyTypeObject SHAtype = {
+    PyObject_HEAD_INIT(NULL)
+    0,			/*ob_size*/
+    "_sha.sha",		/*tp_name*/
+    sizeof(SHAobject),	/*tp_size*/
+    0,			/*tp_itemsize*/
+    /* methods */
+    SHA_dealloc,	/*tp_dealloc*/
+    0,			/*tp_print*/
+    0,                  /*tp_getattr*/
+    0,                  /*tp_setattr*/
+    0,                  /*tp_compare*/
+    0,                  /*tp_repr*/
+    0,                  /*tp_as_number*/
+    0,                  /*tp_as_sequence*/
+    0,                  /*tp_as_mapping*/
+    0,                  /*tp_hash*/
+    0,                  /*tp_call*/
+    0,                  /*tp_str*/
+    0,                  /*tp_getattro*/
+    0,                  /*tp_setattro*/
+    0,                  /*tp_as_buffer*/
+    Py_TPFLAGS_DEFAULT, /*tp_flags*/
+    0,                  /*tp_doc*/
+    0,                  /*tp_traverse*/
+    0,			/*tp_clear*/
+    0,			/*tp_richcompare*/
+    0,			/*tp_weaklistoffset*/
+    0,			/*tp_iter*/
+    0,			/*tp_iternext*/
+    SHA_methods,	/* tp_methods */
+    0,                  /* tp_members */
+    SHA_getseters,      /* tp_getset */
+};
+
+
+/* The single module-level function: new() */
+
+PyDoc_STRVAR(SHA_new__doc__,
+"Return a new SHA hashing object.  An optional string argument\n\
+may be provided; if present, this string will be automatically\n\
+hashed.");
+
+static PyObject *
+SHA_new(PyObject *self, PyObject *args, PyObject *kwdict)
+{
+    static char *kwlist[] = {"string", NULL};
+    SHAobject *new;
+    unsigned char *cp = NULL;
+    int len;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwdict, "|s#:new", kwlist,
+                                     &cp, &len)) {
+        return NULL;
+    }
+
+    if ((new = newSHAobject()) == NULL)
+        return NULL;
+
+    sha_init(new);
+
+    if (PyErr_Occurred()) {
+        Py_DECREF(new);
+        return NULL;
+    }
+    if (cp)
+        sha_update(new, cp, len);
+
+    return (PyObject *)new;
+}
+
+
+/* List of functions exported by this module */
+
+static struct PyMethodDef SHA_functions[] = {
+    {"new", (PyCFunction)SHA_new, METH_VARARGS|METH_KEYWORDS, SHA_new__doc__},
+    {NULL,	NULL}		 /* Sentinel */
+};
+
+
+/* Initialize this module. */
+
+#define insint(n,v) { PyModule_AddIntConstant(m,n,v); }
+
+PyMODINIT_FUNC
+init_sha(void)
+{
+    PyObject *m;
+
+    SHAtype.ob_type = &PyType_Type;
+    if (PyType_Ready(&SHAtype) < 0)
+        return;
+    m = Py_InitModule("_sha", SHA_functions);
+    if (m == NULL)
+	return;
+
+    /* Add some symbolic constants to the module */
+    insint("blocksize", 1);  /* For future use, in case some hash
+                                functions require an integral number of
+                                blocks */ 
+    insint("digestsize", 20);
+    insint("digest_size", 20);
+}
diff --git a/src/distutils2/_backport/hashlib.py b/src/distutils2/_backport/hashlib.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/hashlib.py
@@ -0,0 +1,143 @@
+# $Id$
+#
+#  Copyright (C) 2005   Gregory P. Smith (greg at krypto.org)
+#  Licensed to PSF under a Contributor Agreement.
+#
+
+__doc__ = """hashlib module - A common interface to many hash functions.
+
+new(name, string='') - returns a new hash object implementing the
+                       given hash function; initializing the hash
+                       using the given string data.
+
+Named constructor functions are also available, these are much faster
+than using new():
+
+md5(), sha1(), sha224(), sha256(), sha384(), and sha512()
+
+More algorithms may be available on your platform but the above are
+guaranteed to exist.
+
+NOTE: If you want the adler32 or crc32 hash functions they are available in
+the zlib module.
+
+Choose your hash function wisely.  Some have known collision weaknesses.
+sha384 and sha512 will be slow on 32 bit platforms.
+
+Hash objects have these methods:
+ - update(arg): Update the hash object with the string arg. Repeated calls
+                are equivalent to a single call with the concatenation of all
+                the arguments.
+ - digest():    Return the digest of the strings passed to the update() method
+                so far. This may contain non-ASCII characters, including
+                NUL bytes.
+ - hexdigest(): Like digest() except the digest is returned as a string of
+                double length, containing only hexadecimal digits.
+ - copy():      Return a copy (clone) of the hash object. This can be used to
+                efficiently compute the digests of strings that share a common
+                initial substring.
+
+For example, to obtain the digest of the string 'Nobody inspects the
+spammish repetition':
+
+    >>> import hashlib
+    >>> m = hashlib.md5()
+    >>> m.update("Nobody inspects")
+    >>> m.update(" the spammish repetition")
+    >>> m.digest()
+    '\\xbbd\\x9c\\x83\\xdd\\x1e\\xa5\\xc9\\xd9\\xde\\xc9\\xa1\\x8d\\xf0\\xff\\xe9'
+
+More condensed:
+
+    >>> hashlib.sha224("Nobody inspects the spammish repetition").hexdigest()
+    'a4337bc45a8fc544c03f52dc550cd6e1e87021bc896588bd79e901e2'
+
+"""
+
+# This tuple and __get_builtin_constructor() must be modified if a new
+# always available algorithm is added.
+__always_supported = ('md5', 'sha1', 'sha224', 'sha256', 'sha384', 'sha512')
+
+algorithms = __always_supported
+
+__all__ = __always_supported + ('new', 'algorithms')
+
+
+def __get_builtin_constructor(name):
+    if name in ('SHA1', 'sha1'):
+        import _sha
+        return _sha.new
+    elif name in ('MD5', 'md5'):
+        import _md5
+        return _md5.new
+    elif name in ('SHA256', 'sha256', 'SHA224', 'sha224'):
+        import _sha256
+        bs = name[3:]
+        if bs == '256':
+            return _sha256.sha256
+        elif bs == '224':
+            return _sha256.sha224
+    elif name in ('SHA512', 'sha512', 'SHA384', 'sha384'):
+        import _sha512
+        bs = name[3:]
+        if bs == '512':
+            return _sha512.sha512
+        elif bs == '384':
+            return _sha512.sha384
+
+    raise ValueError('unsupported hash type %s' % name)
+
+
+def __get_openssl_constructor(name):
+    try:
+        f = getattr(_hashlib, 'openssl_' + name)
+        # Allow the C module to raise ValueError.  The function will be
+        # defined but the hash not actually available thanks to OpenSSL.
+        f()
+        # Use the C function directly (very fast)
+        return f
+    except (AttributeError, ValueError):
+        return __get_builtin_constructor(name)
+
+
+def __py_new(name, string=''):
+    """new(name, string='') - Return a new hashing object using the named algorithm;
+    optionally initialized with a string.
+    """
+    return __get_builtin_constructor(name)(string)
+
+
+def __hash_new(name, string=''):
+    """new(name, string='') - Return a new hashing object using the named algorithm;
+    optionally initialized with a string.
+    """
+    try:
+        return _hashlib.new(name, string)
+    except ValueError:
+        # If the _hashlib module (OpenSSL) doesn't support the named
+        # hash, try using our builtin implementations.
+        # This allows for SHA224/256 and SHA384/512 support even though
+        # the OpenSSL library prior to 0.9.8 doesn't provide them.
+        return __get_builtin_constructor(name)(string)
+
+
+try:
+    import _hashlib
+    new = __hash_new
+    __get_hash = __get_openssl_constructor
+except ImportError:
+    new = __py_new
+    __get_hash = __get_builtin_constructor
+
+for __func_name in __always_supported:
+    # try them all, some may not work due to the OpenSSL
+    # version not supporting that algorithm.
+    try:
+        globals()[__func_name] = __get_hash(__func_name)
+    except ValueError:
+        import logging
+        logging.exception('code for hash %s was not found.', __func_name)
+
+# Cleanup locals()
+del __always_supported, __func_name, __get_hash
+del __py_new, __hash_new, __get_openssl_constructor
diff --git a/src/distutils2/_backport/tests/test_pkgutil.py b/src/distutils2/_backport/tests/test_pkgutil.py
--- a/src/distutils2/_backport/tests/test_pkgutil.py
+++ b/src/distutils2/_backport/tests/test_pkgutil.py
@@ -1,5 +1,7 @@
 # -*- coding: utf-8 -*-
 """Tests for PEP 376 pkgutil functionality"""
+import unittest2
+import unittest2.compatibility
 import sys
 import os
 import csv
@@ -13,7 +15,6 @@
     from md5 import md5
 
 from test.test_support import run_unittest, TESTFN
-from distutils2.tests.support import unittest
 
 from distutils2._backport import pkgutil
 
@@ -23,15 +24,13 @@
 # TODO Add a test for absolute pathed RECORD items (e.g. /etc/myapp/config.ini)
 
 # Adapted from Python 2.7's trunk
-class TestPkgUtilData(unittest.TestCase):
+class TestPkgUtilData(unittest2.TestCase):
 
     def setUp(self):
-        super(TestPkgUtilData, self).setUp()
         self.dirname = tempfile.mkdtemp()
         sys.path.insert(0, self.dirname)
 
     def tearDown(self):
-        super(TestPkgUtilData, self).tearDown()
         del sys.path[0]
         shutil.rmtree(self.dirname)
 
@@ -46,22 +45,15 @@
         os.mkdir(package_dir)
         # Empty init.py
         f = open(os.path.join(package_dir, '__init__.py'), "wb")
-        try:
-            pass
-        finally:
-            f.close()
+        f.close()
         # Resource files, res.txt, sub/res.txt
         f = open(os.path.join(package_dir, 'res.txt'), "wb")
-        try:
-            f.write(RESOURCE_DATA)
-        finally:
-            f.close()
+        f.write(RESOURCE_DATA)
+        f.close()
         os.mkdir(os.path.join(package_dir, 'sub'))
         f = open(os.path.join(package_dir, 'sub', 'res.txt'), "wb")
-        try:
-            f.write(RESOURCE_DATA)
-        finally:
-            f.close()
+        f.write(RESOURCE_DATA)
+        f.close()
 
         # Check we can read the resources
         res1 = pkgutil.get_data(pkg, 'res.txt')
@@ -81,14 +73,13 @@
         # Make a package with some resources
         zip_file = os.path.join(self.dirname, zip)
         z = zipfile.ZipFile(zip_file, 'w')
-        try:
-            # Empty init.py
-            z.writestr(pkg + '/__init__.py', "")
-            # Resource files, res.txt, sub/res.txt
-            z.writestr(pkg + '/res.txt', RESOURCE_DATA)
-            z.writestr(pkg + '/sub/res.txt', RESOURCE_DATA)
-        finally:
-            z.close()
+
+        # Empty init.py
+        z.writestr(pkg + '/__init__.py', "")
+        # Resource files, res.txt, sub/res.txt
+        z.writestr(pkg + '/res.txt', RESOURCE_DATA)
+        z.writestr(pkg + '/sub/res.txt', RESOURCE_DATA)
+        z.close()
 
         # Check we can read the resources
         sys.path.insert(0, zip_file)
@@ -101,7 +92,7 @@
         del sys.modules[pkg]
 
 # Adapted from Python 2.7's trunk
-class TestPkgUtilPEP302(unittest.TestCase):
+class TestPkgUtilPEP302(unittest2.TestCase):
 
     class MyTestLoader(object):
         def load_module(self, fullname):
@@ -123,12 +114,10 @@
             return TestPkgUtilPEP302.MyTestLoader()
 
     def setUp(self):
-        super(TestPkgUtilPEP302, self).setUp()
         sys.meta_path.insert(0, self.MyTestImporter())
 
     def tearDown(self):
         del sys.meta_path[0]
-        super(TestPkgUtilPEP302, self).tearDown()
 
     def test_getdata_pep302(self):
         # Use a dummy importer/loader
@@ -146,11 +135,10 @@
         del sys.modules['foo']
 
 
-class TestPkgUtilDistribution(unittest.TestCase):
+class TestPkgUtilDistribution(unittest2.TestCase):
     """Tests the pkgutil.Distribution class"""
 
     def setUp(self):
-        super(TestPkgUtilDistribution, self).setUp()
         self.fake_dists_path = os.path.abspath(
             os.path.join(os.path.dirname(__file__), 'fake_dists'))
 
@@ -199,7 +187,6 @@
         for distinfo_dir in self.distinfo_dirs:
             record_file = os.path.join(distinfo_dir, 'RECORD')
             open(record_file, 'w').close()
-        super(TestPkgUtilDistribution, self).tearDown()
 
     def test_instantiation(self):
         """Test the Distribution class's instantiation provides us with usable
@@ -302,11 +289,10 @@
         self.assertEqual(sorted(found), sorted(distinfo_record_paths))
 
 
-class TestPkgUtilPEP376(unittest.TestCase):
+class TestPkgUtilPEP376(unittest2.TestCase):
     """Tests for the new functionality added in PEP 376."""
 
     def setUp(self):
-        super(TestPkgUtilPEP376, self).setUp()
         # Setup the path environment with our fake distributions
         current_path = os.path.abspath(os.path.dirname(__file__))
         self.sys_path = sys.path[:]
@@ -315,7 +301,6 @@
 
     def tearDown(self):
         sys.path[:] = self.sys_path
-        super(TestPkgUtilPEP376, self).tearDown()
 
     def test_distinfo_dirname(self):
         """Given a name and a version, we expect the distinfo_dirname function
@@ -543,8 +528,8 @@
 
 
 def test_suite():
-    suite = unittest.TestSuite()
-    testcase_loader = unittest.loader.defaultTestLoader.loadTestsFromTestCase
+    suite = unittest2.TestSuite()
+    testcase_loader = unittest2.loader.defaultTestLoader.loadTestsFromTestCase
     suite.addTest(testcase_loader(TestPkgUtilData))
     suite.addTest(testcase_loader(TestPkgUtilDistribution))
     suite.addTest(testcase_loader(TestPkgUtilPEP302))
diff --git a/src/distutils2/command/cmd.py b/src/distutils2/command/cmd.py
--- a/src/distutils2/command/cmd.py
+++ b/src/distutils2/command/cmd.py
@@ -57,8 +57,7 @@
     def __init__(self, dist):
         """Create and initialize a new Command object.  Most importantly,
         invokes the 'initialize_options()' method, which is the real
-        initializer and depends on the actual command being
-        instantiated.
+        initializer and depends on the actual command being instantiated.
         """
         # late import because of mutual dependence between these classes
         from distutils2.dist import Distribution
diff --git a/src/distutils2/command/upload.py b/src/distutils2/command/upload.py
--- a/src/distutils2/command/upload.py
+++ b/src/distutils2/command/upload.py
@@ -160,7 +160,7 @@
         # send the data
         try:
             result = urlopen(request)
-            status = result.getcode()
+            status = result.code
             reason = result.msg
         except socket.error, e:
             self.announce(str(e), log.ERROR)
diff --git a/src/distutils2/command/upload_docs.py b/src/distutils2/command/upload_docs.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/command/upload_docs.py
@@ -0,0 +1,135 @@
+import base64, httplib, os.path, socket, tempfile, urlparse, zipfile
+from cStringIO import StringIO
+from distutils2 import log
+from distutils2.command.upload import upload
+from distutils2.core import PyPIRCCommand
+from distutils2.errors import DistutilsFileError
+
+def zip_dir(directory):
+    """Compresses recursively contents of directory into a StringIO object"""
+    destination = StringIO()
+    zip_file = zipfile.ZipFile(destination, "w")
+    for root, dirs, files in os.walk(directory):
+        for name in files:
+            full = os.path.join(root, name)
+            relative = root[len(directory):].lstrip(os.path.sep)
+            dest = os.path.join(relative, name)
+            zip_file.write(full, dest)
+    zip_file.close()
+    return destination
+
+# grabbed from
+#    http://code.activestate.com/recipes/146306-http-client-to-post-using-multipartform-data/
+def encode_multipart(fields, files, boundary=None):
+    """
+    fields is a sequence of (name, value) elements for regular form fields.
+    files is a sequence of (name, filename, value) elements for data to be uploaded as files
+    Return (content_type, body) ready for httplib.HTTP instance
+    """
+    if boundary is None:
+        boundary = '--------------GHSKFJDLGDS7543FJKLFHRE75642756743254'
+    l = []
+    for (key, value) in fields:
+        l.extend([
+            '--' + boundary,
+            'Content-Disposition: form-data; name="%s"' % key,
+            '',
+            value])
+    for (key, filename, value) in files:
+        l.extend([
+            '--' + boundary,
+            'Content-Disposition: form-data; name="%s"; filename="%s"' % (key, filename),
+            '',
+            value])
+    l.append('--' + boundary + '--')
+    l.append('')
+    body =  '\r\n'.join(l)
+    content_type = 'multipart/form-data; boundary=%s' % boundary
+    return content_type, body
+
+class upload_docs(PyPIRCCommand):
+
+    user_options = [
+        ('repository=', 'r', "url of repository [default: %s]" % upload.DEFAULT_REPOSITORY),
+        ('show-response', None, 'display full response text from server'),
+        ('upload-dir=', None, 'directory to upload'),
+        ]
+
+    def initialize_options(self):
+        PyPIRCCommand.initialize_options(self)
+        self.upload_dir = "build/docs"
+
+    def finalize_options(self):
+        PyPIRCCommand.finalize_options(self)
+        if self.upload_dir == None:
+            build = self.get_finalized_command('build')
+            self.upload_dir = os.path.join(build.build_base, "docs")
+        self.announce('Using upload directory %s' % self.upload_dir)
+        self.verify_upload_dir(self.upload_dir)
+        config = self._read_pypirc()
+        if config != {}:
+            self.username = config['username']
+            self.password = config['password']
+            self.repository = config['repository']
+            self.realm = config['realm']
+
+    def verify_upload_dir(self, upload_dir):
+        self.ensure_dirname('upload_dir')
+        index_location = os.path.join(upload_dir, "index.html")
+        if not os.path.exists(index_location):
+            mesg = "No 'index.html found in docs directory (%s)"
+            raise DistutilsFileError(mesg % upload_dir)
+
+    def run(self):
+        tmp_dir = tempfile.mkdtemp()
+        name = self.distribution.metadata['Name']
+        zip_file = zip_dir(self.upload_dir)
+
+        fields = {':action': 'doc_upload', 'name': name}.items()
+        files = [('content', name, zip_file.getvalue())]
+        content_type, body = encode_multipart(fields, files)
+
+        credentials = self.username + ':' + self.password
+        auth = "Basic " + base64.encodestring(credentials).strip()
+
+        self.announce("Submitting documentation to %s" % (self.repository),
+                      log.INFO)
+
+        schema, netloc, url, params, query, fragments = \
+            urlparse.urlparse(self.repository)
+        if schema == "http":
+            conn = httplib.HTTPConnection(netloc)
+        elif schema == "https":
+            conn = httplib.HTTPSConnection(netloc)
+        else:
+            raise AssertionError("unsupported schema "+schema)
+
+        try:
+            conn.connect()
+            conn.putrequest("POST", url)
+            conn.putheader('Content-type', content_type)
+            conn.putheader('Content-length', str(len(body)))
+            conn.putheader('Authorization', auth)
+            conn.endheaders()
+            conn.send(body)
+        except socket.error, e:
+            self.announce(str(e), log.ERROR)
+            return
+
+        r = conn.getresponse()
+
+        if r.status == 200:
+            self.announce('Server response (%s): %s' % (r.status, r.reason),
+                          log.INFO)
+        elif r.status == 301:
+            location = r.getheader('Location')
+            if location is None:
+                location = 'http://packages.python.org/%s/' % meta.get_name()
+            self.announce('Upload successful. Visit %s' % location,
+                          log.INFO)
+        else:
+            self.announce('Upload failed (%s): %s' % (r.status, r.reason),
+                          log.ERROR)
+
+        if self.show_response:
+            print "\n".join(['-'*75, r.read(), '-'*75])
diff --git a/src/distutils2/dist.py b/src/distutils2/dist.py
--- a/src/distutils2/dist.py
+++ b/src/distutils2/dist.py
@@ -150,8 +150,8 @@
         # information here (and enough command-line options) that it's
         # worth it.  Also delegate 'get_XXX()' methods to the 'metadata'
         # object in a sneaky and underhanded (but efficient!) way.
+        self.metadata = DistributionMetadata()
 
-        self.metadata = DistributionMetadata()
         #for basename in self.metadata._METHOD_BASENAMES:
         #    method_name = "get_" + basename
         #    setattr(self, method_name, getattr(self.metadata, method_name))
diff --git a/src/distutils2/pypi/__init__.py b/src/distutils2/pypi/__init__.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/pypi/__init__.py
@@ -0,0 +1,8 @@
+"""distutils2.pypi
+
+Package containing ways to interact with the PyPI APIs.
+""" 
+
+__all__ = ['simple',
+           'dist',
+]
diff --git a/src/distutils2/pypi/dist.py b/src/distutils2/pypi/dist.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/pypi/dist.py
@@ -0,0 +1,315 @@
+"""distutils2.pypi.dist
+
+Provides the PyPIDistribution class thats represents a distribution retrieved
+on PyPI.
+"""
+import re
+import urlparse
+import urllib
+import tempfile
+from operator import attrgetter
+
+try:
+    import hashlib
+except ImportError:
+    from distutils2._backport import hashlib
+
+from distutils2.version import suggest_normalized_version, NormalizedVersion
+from distutils2.pypi.errors import HashDoesNotMatch, UnsupportedHashName
+
+EXTENSIONS = ".tar.gz .tar.bz2 .tar .zip .tgz .egg".split()
+MD5_HASH = re.compile(r'^.*#md5=([a-f0-9]+)$')
+
+
+class PyPIDistribution(object):
+    """Represents a distribution retrieved from PyPI.
+
+    This is a simple container for various attributes as name, version,
+    downloaded_location, url etc.
+
+    The PyPIDistribution class is used by the pypi.*Index class to return
+    information about distributions.
+    """
+
+    @classmethod
+    def from_url(cls, url, probable_dist_name=None, is_external=True):
+        """Build a Distribution from a url archive (egg or zip or tgz).
+
+        :param url: complete url of the distribution
+        :param probable_dist_name: A probable name of the distribution.
+        :param is_external: Tell if the url commes from an index or from
+                            an external URL.
+        """
+        # if the url contains a md5 hash, get it.
+        md5_hash = None
+        match = MD5_HASH.match(url)
+        if match is not None:
+            md5_hash = match.group(1)
+            # remove the hash
+            url = url.replace("#md5=%s" % md5_hash, "")
+
+        # parse the archive name to find dist name and version
+        archive_name = urlparse.urlparse(url)[2].split('/')[-1]
+        extension_matched = False
+        # remove the extension from the name
+        for ext in EXTENSIONS:
+            if archive_name.endswith(ext):
+                archive_name = archive_name[:-len(ext)]
+                extension_matched = True
+
+        name, version = split_archive_name(archive_name)
+        if extension_matched is True:
+            return PyPIDistribution(name, version, url=url, url_hashname="md5",
+                                    url_hashval=md5_hash, url_is_external=is_external)
+
+    def __init__(self, name, version, type=None, url=None, url_hashname=None,
+                 url_hashval=None, url_is_external=True):
+        """Create a new instance of PyPIDistribution.
+
+        :param name: the name of the distribution
+        :param version: the version of the distribution
+        :param type: the type of the dist (eg. source, bin-*, etc.)
+        :param url: URL where we found this distribution
+        :param url_hashname: the name of the hash we want to use. Refer to the
+                         hashlib.new documentation for more information.
+        :param url_hashval: the hash value.
+        :param url_is_external: we need to know if the provided url comes from an
+                            index browsing, or from an external resource.
+
+        """
+        self.name = name
+        self.version = NormalizedVersion(version)
+        self.type = type
+        # set the downloaded path to None by default. The goal here
+        # is to not download distributions multiple times
+        self.downloaded_location = None
+        # We store urls in dict, because we need to have a bit more informations
+        # than the simple URL. It will be used later to find the good url to
+        # use.
+        # We have two _url* attributes: _url and _urls. _urls contains a list of
+        # dict for the different urls, and _url contains the choosen url, in
+        # order to dont make the selection process multiple times.
+        self._urls = []
+        self._url = None
+        self.add_url(url, url_hashname, url_hashval, url_is_external)
+
+    def add_url(self, url, hashname=None, hashval=None, is_external=True):
+        """Add a new url to the list of urls"""
+        if hashname is not None:
+            try:
+                hashlib.new(hashname)
+            except ValueError:
+                raise UnsupportedHashName(hashname)
+
+        self._urls.append({
+            'url': url,
+            'hashname': hashname,
+            'hashval': hashval,
+            'is_external': is_external,
+        })
+        # reset the url selection process
+        self._url = None
+
+    @property
+    def url(self):
+        """Pick up the right url for the list of urls in self.urls"""
+        # We return internal urls over externals.
+        # If there is more than one internal or external, return the first
+        # one.
+        if self._url is None:
+            if len(self._urls) > 1:
+                internals_urls = [u for u in self._urls \
+                                  if u['is_external'] == False]
+                if len(internals_urls) >= 1:
+                    self._url = internals_urls[0]
+            if self._url is None:
+                self._url = self._urls[0]
+        return self._url
+
+    @property
+    def is_source(self):
+        """return if the distribution is a source one or not"""
+        return self.type == 'source'
+
+    @property
+    def is_final(self):
+        """proxy to version.is_final"""
+        return self.version.is_final
+
+    def download(self, path=None):
+        """Download the distribution to a path, and return it.
+
+        If the path is given in path, use this, otherwise, generates a new one
+        """
+        if path is None:
+            path = tempfile.mkdtemp()
+
+        # if we do not have downloaded it yet, do it.
+        if self.downloaded_location is None:
+            url = self.url['url']
+            archive_name = urlparse.urlparse(url)[2].split('/')[-1]
+            filename, headers = urllib.urlretrieve(url,
+                                                   path + "/" + archive_name)
+            self.downloaded_location = filename
+            self._check_md5(filename)
+        return self.downloaded_location
+
+    def _check_md5(self, filename):
+        """Check that the md5 checksum of the given file matches the one in
+        url param"""
+        hashname = self.url['hashname']
+        expected_hashval = self.url['hashval']
+        if not None in (expected_hashval, hashname):
+            f = open(filename)
+            hashval = hashlib.new(hashname)
+            hashval.update(f.read())
+            if hashval.hexdigest() != expected_hashval:
+                raise HashDoesNotMatch("got %s instead of %s"
+                    % (hashval.hexdigest(), expected_hashval))
+
+    def __repr__(self):
+        return "%s %s %s %s" \
+            % (self.__class__.__name__, self.name, self.version,
+               self.type or "")
+
+    def _check_is_comparable(self, other):
+        if not isinstance(other, PyPIDistribution):
+            raise TypeError("cannot compare %s and %s"
+                % (type(self).__name__, type(other).__name__))
+        elif self.name != other.name:
+            raise TypeError("cannot compare %s and %s"
+                % (self.name, other.name))
+
+    def __eq__(self, other):
+        self._check_is_comparable(other)
+        return self.version == other.version
+
+    def __lt__(self, other):
+        self._check_is_comparable(other)
+        return self.version < other.version
+
+    def __ne__(self, other):
+        return not self.__eq__(other)
+
+    def __gt__(self, other):
+        return not (self.__lt__(other) or self.__eq__(other))
+
+    def __le__(self, other):
+        return self.__eq__(other) or self.__lt__(other)
+
+    def __ge__(self, other):
+        return self.__eq__(other) or self.__gt__(other)
+
+    # See http://docs.python.org/reference/datamodel#object.__hash__
+    __hash__ = object.__hash__
+
+
+class PyPIDistributions(list):
+    """A container of PyPIDistribution objects.
+
+    Contains methods and facilities to sort and filter distributions.
+    """
+    def __init__(self, list=[]):
+        # To disable the ability to pass lists on instanciation
+        super(PyPIDistributions, self).__init__()
+        for item in list:
+            self.append(item)
+
+    def filter(self, predicate):
+        """Filter the distributions and return a subset of distributions that
+        match the given predicate
+        """
+        return PyPIDistributions(
+            [dist for dist in self if dist.name == predicate.name and
+            predicate.match(dist.version)])
+
+    def get_last(self, predicate, prefer_source=None, prefer_final=None):
+        """Return the most up to date version, that satisfy the given
+        predicate
+        """
+        distributions = self.filter(predicate)
+        distributions.sort_distributions(prefer_source, prefer_final, reverse=True)
+        return distributions[0]
+
+    def get_same_name_and_version(self):
+        """Return lists of PyPIDistribution objects that refer to the same
+        name and version number. This do not consider the type (source, binary,
+        etc.)"""
+        processed = []
+        duplicates = []
+        for dist in self:
+            if (dist.name, dist.version) not in processed:
+                processed.append((dist.name, dist.version))
+                found_duplicates = [d for d in self if d.name == dist.name and
+                                    d.version == dist.version]
+                if len(found_duplicates) > 1:
+                    duplicates.append(found_duplicates)
+        return duplicates
+
+    def append(self, o):
+        """Append a new distribution to the list.
+
+        If a distribution with the same name and version exists, just grab the
+        URL informations and add a new new url for the existing one.
+        """
+        similar_dists = [d for d in self if d.name == o.name and
+                         d.version == o.version and d.type == o.type]
+        if len(similar_dists) > 0:
+            dist = similar_dists[0]
+            dist.add_url(**o.url)
+        else:
+            super(PyPIDistributions, self).append(o)
+
+    def sort_distributions(self, prefer_source=None, prefer_final=None,
+                           reverse=True, *args, **kwargs):
+        """order the results with the given properties"""
+
+        sort_by = []
+        if prefer_final is not None:
+            if prefer_final is True:
+                sort_by.append("is_final")
+        sort_by.append("version")
+
+        if prefer_source is not None:
+            if prefer_source is True:
+                sort_by.append("is_source")
+
+        super(PyPIDistributions, self).sort(
+            key=lambda i: [getattr(i, arg) for arg in sort_by],
+            reverse=reverse, *args, **kwargs)
+
+def split_archive_name(archive_name, probable_name=None):
+    """Split an archive name into two parts: name and version.
+
+    Return the tuple (name, version)
+    """
+    # Try to determine wich part is the name and wich is the version using the
+    # "-" separator. Take the larger part to be the version number then reduce
+    # if this not works.
+    def eager_split(str, maxsplit=2):
+        # split using the "-" separator
+        splits = str.rsplit("-", maxsplit)
+        name = splits[0]
+        version = "-".join(splits[1:])
+        if version.startswith("-"):
+            version = version[1:]
+        if suggest_normalized_version(version) is None and maxsplit >= 0:
+            # we dont get a good version number: recurse !
+            return eager_split(str, maxsplit - 1)
+        else:
+            return (name, version)
+    if probable_name is not None:
+        probable_name = probable_name.lower()
+    name = None
+    if probable_name is not None and probable_name in archive_name:
+        # we get the name from probable_name, if given.
+        name = probable_name
+        version = archive_name.lstrip(name)
+    else:
+        name, version = eager_split(archive_name)
+
+    version = suggest_normalized_version(version)
+    if version != "" and name != "":
+        return (name.lower(), version)
+    else:
+        raise CantParseArchiveName(archive_name)
diff --git a/src/distutils2/pypi/errors.py b/src/distutils2/pypi/errors.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/pypi/errors.py
@@ -0,0 +1,33 @@
+"""distutils2.pypi.errors
+
+All errors and exceptions raised by PyPiIndex classes.
+"""
+from distutils2.errors import DistutilsError
+
+
+class PyPIError(DistutilsError):
+    """The base class for errors of the pypi python package."""
+
+
+class DistributionNotFound(PyPIError):
+    """No distribution match the given requirements."""
+
+
+class CantParseArchiveName(PyPIError):
+    """An archive name can't be parsed to find distribution name and version"""
+
+
+class DownloadError(PyPIError):
+    """An error has occurs while downloading"""
+
+
+class HashDoesNotMatch(DownloadError):
+    """Compared hashes does not match"""
+
+
+class UnsupportedHashName(PyPIError):
+    """A unsupported hashname has been used"""
+
+
+class UnableToDownload(PyPIError):
+    """All mirrors have been tried, without success"""
diff --git a/src/distutils2/pypi/simple.py b/src/distutils2/pypi/simple.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/pypi/simple.py
@@ -0,0 +1,395 @@
+"""pypi.simple
+
+Contains the class "SimpleIndex", a simple spider to find and retrieve
+distributions on the Python Package Index, using it's "simple" API,
+avalaible at http://pypi.python.org/simple/
+"""
+from fnmatch import translate
+import urlparse
+import sys
+import re
+import urllib2
+import httplib
+import socket
+
+from distutils2.version import VersionPredicate
+from distutils2.pypi.dist import PyPIDistribution, PyPIDistributions, \
+    EXTENSIONS
+from distutils2.pypi.errors import PyPIError, DistributionNotFound, \
+    DownloadError, UnableToDownload
+from distutils2 import __version__ as __distutils2_version__
+
+# -- Constants -----------------------------------------------
+PYPI_DEFAULT_INDEX_URL = "http://pypi.python.org/simple/"
+PYPI_DEFAULT_MIRROR_URL = "mirrors.pypi.python.org"
+DEFAULT_HOSTS = ("*",)
+SOCKET_TIMEOUT = 15
+USER_AGENT = "Python-urllib/%s distutils2/%s" % (
+    sys.version[:3], __distutils2_version__)
+
+# -- Regexps -------------------------------------------------
+EGG_FRAGMENT = re.compile(r'^egg=([-A-Za-z0-9_.]+)$')
+HREF = re.compile("""href\\s*=\\s*['"]?([^'"> ]+)""", re.I)
+PYPI_MD5 = re.compile(
+    '<a href="([^"#]+)">([^<]+)</a>\n\s+\\(<a (?:title="MD5 hash"\n\s+)'
+    'href="[^?]+\?:action=show_md5&amp;digest=([0-9a-f]{32})">md5</a>\\)')
+URL_SCHEME = re.compile('([-+.a-z0-9]{2,}):', re.I).match
+
+# This pattern matches a character entity reference (a decimal numeric
+# references, a hexadecimal numeric reference, or a named reference).
+ENTITY_SUB = re.compile(r'&(#(\d+|x[\da-fA-F]+)|[\w.:-]+);?').sub
+REL = re.compile("""<([^>]*\srel\s*=\s*['"]?([^'">]+)[^>]*)>""", re.I)
+
+
+def socket_timeout(timeout=SOCKET_TIMEOUT):
+    """Decorator to add a socket timeout when requesting pages on PyPI.
+    """
+    def _socket_timeout(func):
+        def _socket_timeout(self, *args, **kwargs):
+            old_timeout = socket.getdefaulttimeout()
+            if hasattr(self, "_timeout"):
+                timeout = self._timeout
+            socket.setdefaulttimeout(timeout)
+            try:
+                return func(self, *args, **kwargs)
+            finally:
+                socket.setdefaulttimeout(old_timeout)
+        return _socket_timeout
+    return _socket_timeout
+
+
+class SimpleIndex(object):
+    """Provides useful tools to request the Python Package Index simple API
+    """
+
+    def __init__(self, index_url=PYPI_DEFAULT_INDEX_URL, hosts=DEFAULT_HOSTS,
+                 follow_externals=False, prefer_source=True,
+                 prefer_final=False, mirrors_url=PYPI_DEFAULT_MIRROR_URL,
+                 mirrors=None, timeout=SOCKET_TIMEOUT):
+        """Class constructor.
+
+        :param index_url: the url of the simple index to search on.
+        :param follow_externals: tell if following external links is needed or
+                                 not. Default is False.
+        :param hosts: a list of hosts allowed to be processed while using
+                      follow_externals=True. Default behavior is to follow all
+                      hosts.
+        :param follow_externals: tell if following external links is needed or
+                                 not. Default is False.
+        :param prefer_source: if there is binary and source distributions, the
+                              source prevails.
+        :param prefer_final: if the version is not mentioned, and the last
+                             version is not a "final" one (alpha, beta, etc.),
+                             pick up the last final version.
+        :param mirrors_url: the url to look on for DNS records giving mirror
+                            adresses.
+        :param mirrors: a list of mirrors to check out if problems
+                             occurs while working with the one given in "url"
+        :param timeout: time in seconds to consider a url has timeouted.
+        """
+        self.follow_externals = follow_externals
+
+        if not index_url.endswith("/"):
+            index_url += "/"
+        self._index_urls = [index_url]
+        # if no mirrors are defined, use the method described in PEP 381.
+        if mirrors is None:
+            try:
+                mirrors = socket.gethostbyname_ex(mirrors_url)[-1]
+            except socket.gaierror:
+                mirrors = []
+        self._index_urls.extend(mirrors)
+        self._current_index_url = 0
+        self._timeout = timeout
+        self._prefer_source = prefer_source
+        self._prefer_final = prefer_final
+
+        # create a regexp to match all given hosts
+        self._allowed_hosts = re.compile('|'.join(map(translate, hosts))).match
+
+        # we keep an index of pages we have processed, in order to avoid
+        # scanning them multple time (eg. if there is multiple pages pointing
+        # on one)
+        self._processed_urls = []
+        self._distributions = {}
+
+    def find(self, requirements, prefer_source=None, prefer_final=None):
+        """Browse the PyPI to find distributions that fullfil the given
+        requirements.
+
+        :param requirements: A project name and it's distribution, using
+                             version specifiers, as described in PEP345.
+        :type requirements:  You can pass either a version.VersionPredicate
+                             or a string.
+        :param prefer_source: if there is binary and source distributions, the
+                              source prevails.
+        :param prefer_final: if the version is not mentioned, and the last
+                             version is not a "final" one (alpha, beta, etc.),
+                             pick up the last final version.
+        """
+        requirements = self._get_version_predicate(requirements)
+        if prefer_source is None:
+            prefer_source = self._prefer_source
+        if prefer_final is None:
+            prefer_final = self._prefer_final
+
+        # process the index for this project
+        self._process_pypi_page(requirements.name)
+
+        # filter with requirements and return the results
+        if requirements.name in self._distributions:
+            dists = self._distributions[requirements.name].filter(requirements)
+            dists.sort_distributions(prefer_source=prefer_source,
+                                     prefer_final=prefer_final)
+        else:
+            dists = []
+
+        return dists
+
+    def get(self, requirements, *args, **kwargs):
+        """Browse the PyPI index to find distributions that fullfil the
+        given requirements, and return the most recent one.
+
+        You can specify prefer_final and prefer_source arguments here.
+        If not, the default one will be used.
+        """
+        predicate = self._get_version_predicate(requirements)
+        dists = self.find(predicate, *args, **kwargs)
+
+        if len(dists) == 0:
+            raise DistributionNotFound(requirements)
+
+        return dists.get_last(predicate)
+
+    def download(self, requirements, temp_path=None, *args, **kwargs):
+        """Download the distribution, using the requirements.
+
+        If more than one distribution match the requirements, use the last
+        version.
+        Download the distribution, and put it in the temp_path. If no temp_path
+        is given, creates and return one.
+
+        Returns the complete absolute path to the downloaded archive.
+
+        :param requirements: The same as the find attribute of `find`.
+
+        You can specify prefer_final and prefer_source arguments here.
+        If not, the default one will be used.
+        """
+        return self.get(requirements, *args, **kwargs)\
+                   .download(path=temp_path)
+
+    def _get_version_predicate(self, requirements):
+        """Return a VersionPredicate object, from a string or an already
+        existing object.
+        """
+        if isinstance(requirements, str):
+            requirements = VersionPredicate(requirements)
+        return requirements
+
+    @property
+    def index_url(self):
+        return self._index_urls[self._current_index_url]
+
+    def _switch_to_next_mirror(self):
+        """Switch to the next mirror (eg. point self.index_url to the next
+        url.
+        """
+        # Internally, iter over the _index_url iterable, if we have read all
+        # of the available indexes, raise an exception.
+        if self._current_index_url < len(self._index_urls):
+            self._current_index_url = self._current_index_url + 1
+        else:
+            raise UnableToDownload("All mirrors fails")
+
+    def _is_browsable(self, url):
+        """Tell if the given URL can be browsed or not.
+
+        It uses the follow_externals and the hosts list to tell if the given
+        url is browsable or not.
+        """
+        # if _index_url is contained in the given URL, we are browsing the
+        # index, and it's always "browsable".
+        # local files are always considered browable resources
+        if self.index_url in url or urlparse.urlparse(url)[0] == "file":
+            return True
+        elif self.follow_externals is True:
+            if self._allowed_hosts(urlparse.urlparse(url)[1]):  # 1 is netloc
+                return True
+            else:
+                return False
+        return False
+
+    def _is_distribution(self, link):
+        """Tell if the given URL matches to a distribution name or not.
+        """
+        #XXX find a better way to check that links are distributions
+        # Using a regexp ?
+        for ext in EXTENSIONS:
+            if ext in link:
+                return True
+        return False
+
+    def _register_dist(self, dist):
+        """Register a distribution as a part of fetched distributions for
+        SimpleIndex.
+
+        Return the PyPIDistributions object for the specified project name
+        """
+        # Internally, check if a entry exists with the project name, if not,
+        # create a new one, and if exists, add the dist to the pool.
+        if not dist.name in self._distributions:
+            self._distributions[dist.name] = PyPIDistributions()
+        self._distributions[dist.name].append(dist)
+        return self._distributions[dist.name]
+
+    def _process_url(self, url, project_name=None, follow_links=True):
+        """Process an url and search for distributions packages.
+
+        For each URL found, if it's a download, creates a PyPIdistribution
+        object. If it's a homepage and we can follow links, process it too.
+
+        :param url: the url to process
+        :param project_name: the project name we are searching for.
+        :param follow_links: Do not want to follow links more than from one
+                             level. This parameter tells if we want to follow
+                             the links we find (eg. run recursively this
+                             method on it)
+        """
+        f = self._open_url(url)
+        base_url = f.url
+        if url not in self._processed_urls:
+            self._processed_urls.append(url)
+            link_matcher = self._get_link_matcher(url)
+            for link, is_download in link_matcher(f.read(), base_url):
+                if link not in self._processed_urls:
+                    if self._is_distribution(link) or is_download:
+                        self._processed_urls.append(link)
+                        # it's a distribution, so create a dist object
+                        dist = PyPIDistribution.from_url(link, project_name,
+                                    is_external=not self.index_url in url)
+                        self._register_dist(dist)
+                    else:
+                        if self._is_browsable(link) and follow_links:
+                            self._process_url(link, project_name,
+                                follow_links=False)
+
+    def _get_link_matcher(self, url):
+        """Returns the right link matcher function of the given url
+        """
+        if self.index_url in url:
+            return self._simple_link_matcher
+        else:
+            return self._default_link_matcher
+
+    def _simple_link_matcher(self, content, base_url):
+        """Yield all links with a rel="download" or rel="homepage".
+
+        This matches the simple index requirements for matching links.
+        If follow_externals is set to False, dont yeld the external
+        urls.
+        """
+        for match in REL.finditer(content):
+            tag, rel = match.groups()
+            rels = map(str.strip, rel.lower().split(','))
+            if 'homepage' in rels or 'download' in rels:
+                for match in HREF.finditer(tag):
+                    url = urlparse.urljoin(base_url,
+                                           self._htmldecode(match.group(1)))
+                    if 'download' in rels or self._is_browsable(url):
+                        # yield a list of (url, is_download)
+                        yield (urlparse.urljoin(base_url, url),
+                               'download' in rels)
+
+    def _default_link_matcher(self, content, base_url):
+        """Yield all links found on the page.
+        """
+        for match in HREF.finditer(content):
+            url = urlparse.urljoin(base_url, self._htmldecode(match.group(1)))
+            if self._is_browsable(url):
+                yield (url, False)
+
+    def _process_pypi_page(self, name):
+        """Find and process a PyPI page for the given project name.
+
+        :param name: the name of the project to find the page
+        """
+        try:
+            # Browse and index the content of the given PyPI page.
+            url = self.index_url + name + "/"
+            self._process_url(url, name)
+        except DownloadError:
+            # if an error occurs, try with the next index_url
+            # (provided by the mirrors)
+            self._switch_to_next_mirror()
+            self._distributions.clear()
+            self._process_pypi_page(name)
+
+    @socket_timeout()
+    def _open_url(self, url):
+        """Open a urllib2 request, handling HTTP authentication, and local
+        files support.
+
+        """
+        try:
+            scheme, netloc, path, params, query, frag = urlparse.urlparse(url)
+
+            if scheme in ('http', 'https'):
+                auth, host = urllib2.splituser(netloc)
+            else:
+                auth = None
+
+            # add index.html automatically for filesystem paths
+            if scheme == 'file':
+                if url.endswith('/'):
+                    url += "index.html"
+
+            if auth:
+                auth = "Basic " + \
+                    urllib2.unquote(auth).encode('base64').strip()
+                new_url = urlparse.urlunparse((
+                    scheme, host, path, params, query, frag))
+                request = urllib2.Request(new_url)
+                request.add_header("Authorization", auth)
+            else:
+                request = urllib2.Request(url)
+            request.add_header('User-Agent', USER_AGENT)
+            fp = urllib2.urlopen(request)
+
+            if auth:
+                # Put authentication info back into request URL if same host,
+                # so that links found on the page will work
+                s2, h2, path2, param2, query2, frag2 = \
+                    urlparse.urlparse(fp.url)
+                if s2 == scheme and h2 == host:
+                    fp.url = urlparse.urlunparse(
+                        (s2, netloc, path2, param2, query2, frag2))
+
+            return fp
+        except (ValueError, httplib.InvalidURL), v:
+            msg = ' '.join([str(arg) for arg in v.args])
+            raise PyPIError('%s %s' % (url, msg))
+        except urllib2.HTTPError, v:
+            return v
+        except urllib2.URLError, v:
+            raise DownloadError("Download error for %s: %s" % (url, v.reason))
+        except httplib.BadStatusLine, v:
+            raise DownloadError('%s returned a bad status line. '
+                'The server might be down, %s' % (url, v.line))
+        except httplib.HTTPException, v:
+            raise DownloadError("Download error for %s: %s" % (url, v))
+
+    def _decode_entity(self, match):
+        what = match.group(1)
+        if what.startswith('#x'):
+            what = int(what[2:], 16)
+        elif what.startswith('#'):
+            what = int(what[1:])
+        else:
+            from htmlentitydefs import name2codepoint
+            what = name2codepoint.get(what, match.group(0))
+        return unichr(what)
+
+    def _htmldecode(self, text):
+        """Decode HTML entities in the given text."""
+        return ENTITY_SUB(self._decode_entity, text)
diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypi_server.py
@@ -0,0 +1,195 @@
+"""Mocked PyPI Server implementation, to use in tests.
+
+This module also provides a simple test case to extend if you need to use
+the PyPIServer all along your test case. Be sure to read the documentation 
+before any use.
+"""
+
+import Queue
+import threading
+import time
+import urllib2
+from BaseHTTPServer import HTTPServer
+from SimpleHTTPServer import SimpleHTTPRequestHandler
+import os.path
+import select
+
+from distutils2.tests.support import unittest
+
+PYPI_DEFAULT_STATIC_PATH = os.path.dirname(os.path.abspath(__file__)) + "/pypiserver"
+
+def use_pypi_server(*server_args, **server_kwargs):
+    """Decorator to make use of the PyPIServer for test methods, 
+    just when needed, and not for the entire duration of the testcase.
+    """
+    def wrapper(func):
+        def wrapped(*args, **kwargs):
+            server = PyPIServer(*server_args, **server_kwargs)
+            server.start()
+            try:
+                func(server=server, *args, **kwargs)
+            finally:
+                server.stop()
+        return wrapped
+    return wrapper
+
+class PyPIServerTestCase(unittest.TestCase):
+
+    def setUp(self):
+        super(PyPIServerTestCase, self).setUp()
+        self.pypi = PyPIServer()
+        self.pypi.start()
+
+    def tearDown(self):
+        super(PyPIServerTestCase, self).tearDown()
+        self.pypi.stop()
+
+class PyPIServer(threading.Thread):
+    """PyPI Mocked server.
+    Provides a mocked version of the PyPI API's, to ease tests.
+
+    Support serving static content and serving previously given text.
+    """
+
+    def __init__(self, test_static_path=None,
+                 static_filesystem_paths=["default"], static_uri_paths=["simple"]):
+        """Initialize the server.
+
+        static_uri_paths and static_base_path are parameters used to provides
+        respectively the http_paths to serve statically, and where to find the
+        matching files on the filesystem.
+        """
+        threading.Thread.__init__(self)
+        self._run = True
+        self.httpd = HTTPServer(('', 0), PyPIRequestHandler)
+        self.httpd.RequestHandlerClass.log_request = lambda *_: None
+        self.httpd.RequestHandlerClass.pypi_server = self
+        self.address = (self.httpd.server_name, self.httpd.server_port)
+        self.request_queue = Queue.Queue()
+        self._requests = []
+        self.default_response_status = 200
+        self.default_response_headers = [('Content-type', 'text/plain')]
+        self.default_response_data = "hello"
+        
+        # initialize static paths / filesystems
+        self.static_uri_paths = static_uri_paths
+        if test_static_path is not None:
+            static_filesystem_paths.append(test_static_path)
+        self.static_filesystem_paths = [PYPI_DEFAULT_STATIC_PATH + "/" + path
+            for path in static_filesystem_paths]
+
+    def run(self):
+        # loop because we can't stop it otherwise, for python < 2.6
+        while self._run:
+            r, w, e = select.select([self.httpd], [], [], 0.5)
+            if r:
+                self.httpd.handle_request()
+
+    def stop(self):
+        """self shutdown is not supported for python < 2.6"""
+        self._run = False
+
+    def get_next_response(self):
+        return (self.default_response_status,
+                self.default_response_headers,
+                self.default_response_data)
+
+    @property
+    def requests(self):
+        """Use this property to get all requests that have been made
+        to the server
+        """
+        while True:
+            try:
+                self._requests.append(self.request_queue.get_nowait())
+            except Queue.Empty:
+                break
+        return self._requests
+
+    @property
+    def full_address(self):
+        return "http://%s:%s" % self.address
+
+
+class PyPIRequestHandler(SimpleHTTPRequestHandler):
+    # we need to access the pypi server while serving the content
+    pypi_server = None
+
+    def do_POST(self):
+        return self.serve_request()
+    def do_GET(self):
+        return self.serve_request()
+    def do_DELETE(self):
+        return self.serve_request()
+    def do_PUT(self):
+        return self.serve_request()
+
+    def serve_request(self):
+        """Serve the content.
+
+        Also record the requests to be accessed later. If trying to access an
+        url matching a static uri, serve static content, otherwise serve
+        what is provided by the `get_next_response` method.
+        """
+        # record the request. Read the input only on PUT or POST requests
+        if self.command in ("PUT", "POST"):
+            if 'content-length' in self.headers.dict:
+                request_data = self.rfile.read(
+                    int(self.headers['content-length']))
+            else:
+                request_data = self.rfile.read()
+        elif self.command in ("GET", "DELETE"):
+            request_data = ''
+
+        self.pypi_server.request_queue.put((self, request_data))
+
+        # serve the content from local disc if we request an URL beginning
+        # by a pattern defined in `static_paths`
+        url_parts = self.path.split("/")
+        if (len(url_parts) > 1 and 
+                url_parts[1] in self.pypi_server.static_uri_paths):
+            data = None
+            # always take the last first.
+            fs_paths = []
+            fs_paths.extend(self.pypi_server.static_filesystem_paths)
+            fs_paths.reverse()
+            relative_path = self.path
+            for fs_path in fs_paths:
+                try:
+                    if self.path.endswith("/"):
+                        relative_path += "index.html"
+                    file = open(fs_path + relative_path)
+                    data = file.read()
+                    if relative_path.endswith('.tar.gz'):
+                        headers=[('Content-type', 'application/x-gtar')]
+                    else:
+                        headers=[('Content-type', 'text/html')]
+                    self.make_response(data, headers=headers)
+                except IOError:
+                    pass
+
+            if data is None:
+                self.make_response("Not found", 404)
+
+        # otherwise serve the content from get_next_response
+        else:
+            # send back a response
+            status, headers, data = self.pypi_server.get_next_response()
+            self.make_response(data, status, headers)
+
+    def make_response(self, data, status=200,
+                      headers=[('Content-type', 'text/html')]):
+        """Send the response to the HTTP client"""
+        if not isinstance(status, int):
+            try:
+                status = int(status)
+            except ValueError:
+                # we probably got something like YYY Codename. 
+                # Just get the first 3 digits
+                status = int(status[:3])
+
+        self.send_response(status)
+        for header, value in headers:
+            self.send_header(header, value)
+        self.end_headers()
+        self.wfile.write(data)
diff --git a/src/distutils2/tests/pypiserver/downloads_with_md5/simple/badmd5/badmd5-0.1.tar.gz b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/badmd5/badmd5-0.1.tar.gz
new file mode 100644
diff --git a/src/distutils2/tests/pypiserver/downloads_with_md5/simple/badmd5/index.html b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/badmd5/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/badmd5/index.html
@@ -0,0 +1,3 @@
+<html><body>
+<a href="badmd5-0.1.tar.gz#md5=3e3d86693d6564c807272b11b3069dfe" rel="download">badmd5-0.1.tar.gz</a><br/>
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/downloads_with_md5/simple/foobar/foobar-0.1.tar.gz b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/foobar/foobar-0.1.tar.gz
new file mode 100644
diff --git a/src/distutils2/tests/pypiserver/downloads_with_md5/simple/foobar/index.html b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/foobar/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/foobar/index.html
@@ -0,0 +1,3 @@
+<html><body>
+<a href="foobar-0.1.tar.gz#md5=d41d8cd98f00b204e9800998ecf8427e" rel="download">foobar-0.1.tar.gz</a><br/>
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/downloads_with_md5/simple/index.html b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/index.html
@@ -0,0 +1,2 @@
+<a href="foobar/">foobar/</a> 
+<a href="badmd5/">badmd5/</a> 
diff --git a/src/distutils2/tests/pypiserver/foo_bar_baz/simple/bar/index.html b/src/distutils2/tests/pypiserver/foo_bar_baz/simple/bar/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/foo_bar_baz/simple/bar/index.html
@@ -0,0 +1,6 @@
+<html><head><title>Links for bar</title></head><body><h1>Links for bar</h1>
+<a rel="download" href="../../packages/source/F/bar/bar-1.0.tar.gz">bar-1.0.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/bar/bar-1.0.1.tar.gz">bar-1.0.1.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/bar/bar-2.0.tar.gz">bar-2.0.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/bar/bar-2.0.1.tar.gz">bar-2.0.1.tar.gz</a><br/> 
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/foo_bar_baz/simple/baz/index.html b/src/distutils2/tests/pypiserver/foo_bar_baz/simple/baz/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/foo_bar_baz/simple/baz/index.html
@@ -0,0 +1,6 @@
+<html><head><title>Links for baz</title></head><body><h1>Links for baz</h1>
+<a rel="download" href="../../packages/source/F/baz/baz-1.0.tar.gz">baz-1.0.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/baz/baz-1.0.1.tar.gz">baz-1.0.1.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/baz/baz-2.0.tar.gz">baz-2.0.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/baz/baz-2.0.1.tar.gz">baz-2.0.1.tar.gz</a><br/> 
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/foo_bar_baz/simple/foo/index.html b/src/distutils2/tests/pypiserver/foo_bar_baz/simple/foo/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/foo_bar_baz/simple/foo/index.html
@@ -0,0 +1,6 @@
+<html><head><title>Links for foo</title></head><body><h1>Links for foo</h1>
+<a rel="download" href="../../packages/source/F/foo/foo-1.0.tar.gz">foo-1.0.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/foo/foo-1.0.1.tar.gz">foo-1.0.1.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/foo/foo-2.0.tar.gz">foo-2.0.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/foo/foo-2.0.1.tar.gz">foo-2.0.1.tar.gz</a><br/> 
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/foo_bar_baz/simple/index.html b/src/distutils2/tests/pypiserver/foo_bar_baz/simple/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/foo_bar_baz/simple/index.html
@@ -0,0 +1,3 @@
+<a href="foo/">foo/</a> 
+<a href="bar/">bar/</a> 
+<a href="baz/">baz/</a> 
diff --git a/src/distutils2/tests/pypiserver/test_found_links/simple/foobar/index.html b/src/distutils2/tests/pypiserver/test_found_links/simple/foobar/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/test_found_links/simple/foobar/index.html
@@ -0,0 +1,6 @@
+<html><head><title>Links for Foobar</title></head><body><h1>Links for Foobar</h1>
+<a rel="download" href="../../packages/source/F/Foobar/Foobar-1.0.tar.gz#md5=98fa833fdabcdd78d00245aead66c174">Foobar-1.0.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/Foobar/Foobar-1.0.1.tar.gz#md5=2351efb20f6b7b5d9ce80fa4cb1bd9ca">Foobar-1.0.1.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/Foobar/Foobar-2.0.tar.gz#md5=98fa833fdabcdd78d00245aead66c274">Foobar-2.0.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/Foobar/Foobar-2.0.1.tar.gz#md5=2352efb20f6b7b5d9ce80fa4cb2bd9ca">Foobar-2.0.1.tar.gz</a><br/> 
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/test_found_links/simple/index.html b/src/distutils2/tests/pypiserver/test_found_links/simple/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/test_found_links/simple/index.html
@@ -0,0 +1,1 @@
+<a href="foobar/">foobar/</a> 
diff --git a/src/distutils2/tests/pypiserver/test_pypi_server/external/index.html b/src/distutils2/tests/pypiserver/test_pypi_server/external/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/test_pypi_server/external/index.html
@@ -0,0 +1,1 @@
+index.html from external server
diff --git a/src/distutils2/tests/pypiserver/test_pypi_server/simple/index.html b/src/distutils2/tests/pypiserver/test_pypi_server/simple/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/test_pypi_server/simple/index.html
@@ -0,0 +1,1 @@
+Yeah
diff --git a/src/distutils2/tests/pypiserver/with_externals/external/external.html b/src/distutils2/tests/pypiserver/with_externals/external/external.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/with_externals/external/external.html
@@ -0,0 +1,3 @@
+<html><body>
+<a href="/foobar-0.1.tar.gz#md5=1__bad_md5___">bad old link</a>
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/with_externals/simple/foobar/index.html b/src/distutils2/tests/pypiserver/with_externals/simple/foobar/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/with_externals/simple/foobar/index.html
@@ -0,0 +1,4 @@
+<html><body>
+<a rel ="download" href="/foobar-0.1.tar.gz#md5=12345678901234567">foobar-0.1.tar.gz</a><br/>
+<a href="../../external/external.html" rel="homepage">external homepage</a><br/>
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/with_externals/simple/index.html b/src/distutils2/tests/pypiserver/with_externals/simple/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/with_externals/simple/index.html
@@ -0,0 +1,1 @@
+<a href="foobar/">foobar/</a> 
diff --git a/src/distutils2/tests/pypiserver/with_norel_links/external/homepage.html b/src/distutils2/tests/pypiserver/with_norel_links/external/homepage.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/with_norel_links/external/homepage.html
@@ -0,0 +1,7 @@
+<html>
+<body>
+<p>a rel=homepage HTML page</p>
+<a href="/foobar-2.0.tar.gz">foobar 2.0</a>
+</body>
+</html>
+
diff --git a/src/distutils2/tests/pypiserver/with_norel_links/external/nonrel.html b/src/distutils2/tests/pypiserver/with_norel_links/external/nonrel.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/with_norel_links/external/nonrel.html
@@ -0,0 +1,1 @@
+A page linked without rel="download" or rel="homepage" link.
diff --git a/src/distutils2/tests/pypiserver/with_norel_links/simple/foobar/index.html b/src/distutils2/tests/pypiserver/with_norel_links/simple/foobar/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/with_norel_links/simple/foobar/index.html
@@ -0,0 +1,6 @@
+<html><body>
+<a rel="download" href="/foobar-0.1.tar.gz" rel="download">foobar-0.1.tar.gz</a><br/>
+<a href="../../external/homepage.html" rel="homepage">external homepage</a><br/>
+<a href="../../external/nonrel.html">unrelated link</a><br/>
+<a href="/unrelated-0.2.tar.gz">unrelated download</a></br/>
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/with_norel_links/simple/index.html b/src/distutils2/tests/pypiserver/with_norel_links/simple/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/with_norel_links/simple/index.html
@@ -0,0 +1,1 @@
+<a href="foobar/">foobar/</a> 
diff --git a/src/distutils2/tests/pypiserver/with_real_externals/simple/foobar/index.html b/src/distutils2/tests/pypiserver/with_real_externals/simple/foobar/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/with_real_externals/simple/foobar/index.html
@@ -0,0 +1,4 @@
+<html><body>
+<a rel="download" href="/foobar-0.1.tar.gz#md5=0_correct_md5">foobar-0.1.tar.gz</a><br/>
+<a href="http://a-really-external-website/external/external.html" rel="homepage">external homepage</a><br/>
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/with_real_externals/simple/index.html b/src/distutils2/tests/pypiserver/with_real_externals/simple/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/with_real_externals/simple/index.html
@@ -0,0 +1,1 @@
+<a href="foobar/">foobar/</a> 
diff --git a/src/distutils2/tests/test_pypi_dist.py b/src/distutils2/tests/test_pypi_dist.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/test_pypi_dist.py
@@ -0,0 +1,251 @@
+"""Tests for the distutils2.pypi.dist module."""
+
+import os
+import shutil
+import tempfile
+
+from distutils2.tests.pypi_server import use_pypi_server
+from distutils2.tests import support
+from distutils2.tests.support import unittest
+from distutils2.version import VersionPredicate
+from distutils2.pypi.errors import HashDoesNotMatch, UnsupportedHashName
+from distutils2.pypi.dist import (PyPIDistribution as Dist,
+                                  PyPIDistributions as Dists,
+                                  split_archive_name)
+
+
+class TestPyPIDistribution(support.TempdirManager,
+                           unittest.TestCase):
+    """tests the pypi.dist.PyPIDistribution class"""
+
+    def test_instanciation(self):
+        """Test the Distribution class provides us the good attributes when
+        given on construction"""
+        dist = Dist("FooBar", "1.1")
+        self.assertEqual("FooBar", dist.name)
+        self.assertEqual("1.1", "%s" % dist.version)
+
+    def test_from_url(self):
+        """Test that the Distribution object can be built from a single URL"""
+        url_list = {
+            'FooBar-1.1.0.tar.gz': {
+                'name': 'foobar',  # lowercase the name
+                'version': '1.1',
+            },
+            'Foo-Bar-1.1.0.zip': {
+                'name': 'foo-bar',  # keep the dash
+                'version': '1.1',
+            },
+            'foobar-1.1b2.tar.gz#md5=123123123123123': {
+                'name': 'foobar',
+                'version': '1.1b2',
+                'url': {
+                    'url': 'http://test.tld/foobar-1.1b2.tar.gz',  # no hash
+                    'hashval': '123123123123123',
+                    'hashname': 'md5',
+                }
+            },
+            'foobar-1.1-rc2.tar.gz': {  # use suggested name
+                'name': 'foobar',
+                'version': '1.1c2',
+                'url': {
+                    'url': 'http://test.tld/foobar-1.1-rc2.tar.gz',
+                }
+            }
+        }
+
+        for url, attributes in url_list.items():
+            dist = Dist.from_url("http://test.tld/" + url)
+            for attribute, value in attributes.items():
+                if isinstance(value, dict):
+                    mylist = getattr(dist, attribute)
+                    for val in value.keys():
+                        self.assertEqual(value[val], mylist[val])
+                else:
+                    if attribute == "version":
+                        self.assertEqual("%s" % getattr(dist, "version"), value)
+                    else:
+                        self.assertEqual(getattr(dist, attribute), value)
+
+    def test_get_url(self):
+        """Test that the url property works well"""
+
+        d = Dist("foobar", "1.1", url="test_url")
+        self.assertDictEqual(d.url, {
+            "url": "test_url",
+            "is_external": True,
+            "hashname": None,
+            "hashval": None,
+        })
+
+        # add a new url
+        d.add_url(url="internal_url", is_external=False)
+        self.assertEqual(d._url, None)
+        self.assertDictEqual(d.url, {
+            "url": "internal_url",
+            "is_external": False,
+            "hashname": None,
+            "hashval": None,
+        })
+        self.assertEqual(2, len(d._urls))
+
+    def test_comparaison(self):
+        """Test that we can compare PyPIDistributions"""
+        foo1 = Dist("foo", "1.0")
+        foo2 = Dist("foo", "2.0")
+        bar = Dist("bar", "2.0")
+        # assert we use the version to compare
+        self.assertTrue(foo1 < foo2)
+        self.assertFalse(foo1 > foo2)
+        self.assertFalse(foo1 == foo2)
+
+        # assert we can't compare dists with different names
+        self.assertRaises(TypeError, foo1.__eq__, bar)
+
+    def test_split_archive_name(self):
+        """Test we can split the archive names"""
+        names = {
+            'foo-bar-baz-1.0-rc2': ('foo-bar-baz', '1.0c2'),
+            'foo-bar-baz-1.0': ('foo-bar-baz', '1.0'),
+            'foobarbaz-1.0': ('foobarbaz', '1.0'),
+        }
+        for name, results in names.items():
+            self.assertEqual(results, split_archive_name(name))
+
+    @use_pypi_server("downloads_with_md5")
+    def test_download(self, server):
+        """Download is possible, and the md5 is checked if given"""
+
+        url = "%s/simple/foobar/foobar-0.1.tar.gz" % server.full_address
+        # check md5 if given
+        dist = Dist("FooBar", "0.1", url=url,
+            url_hashname="md5", url_hashval="d41d8cd98f00b204e9800998ecf8427e")
+        dist.download()
+
+        # a wrong md5 fails
+        dist2 = Dist("FooBar", "0.1", url=url,
+            url_hashname="md5", url_hashval="wrongmd5")
+        self.assertRaises(HashDoesNotMatch, dist2.download)
+
+        # we can omit the md5 hash
+        dist3 = Dist("FooBar", "0.1", url=url)
+        dist3.download()
+
+        # and specify a temporary location
+        # for an already downloaded dist
+        path1 = tempfile.mkdtemp()
+        dist3.download(path=path1)
+        # and for a new one
+        path2_base = tempfile.mkdtemp()
+        dist4 = Dist("FooBar", "0.1", url=url)
+        path2 = dist4.download(path=path2_base)
+        self.assertTrue(path2_base in path2)
+
+        # remove the temp folders
+        shutil.rmtree(path1)
+        shutil.rmtree(os.path.dirname(path2))
+
+    def test_hashname(self):
+        """Invalid hashnames raises an exception on assignation"""
+        # should be ok
+        Dist("FooBar", "0.1", url_hashname="md5", url_hashval="value")
+
+        self.assertRaises(UnsupportedHashName, Dist, "FooBar", "0.1",
+                          url_hashname="invalid_hashname", url_hashval="value")
+
+
+class TestPyPIDistributions(unittest.TestCase):
+    """test the pypi.distr.PyPIDistributions class"""
+
+    def test_filter(self):
+        """Test we filter the distributions the right way, using version
+        predicate match method"""
+        dists = Dists((
+            Dist("FooBar", "1.1"),
+            Dist("FooBar", "1.1.1"),
+            Dist("FooBar", "1.2"),
+            Dist("FooBar", "1.2.1"),
+        ))
+        filtered = dists.filter(VersionPredicate("FooBar (<1.2)"))
+        self.assertNotIn(dists[2], filtered)
+        self.assertNotIn(dists[3], filtered)
+        self.assertIn(dists[0], filtered)
+        self.assertIn(dists[1], filtered)
+
+    def test_append(self):
+        """Test the append method of PyPIDistributions"""
+        # When adding a new item to the list, the behavior is to test if
+        # a distribution with the same name and version number already exists,
+        # and if so, to add url informations to the existing PyPIDistribution
+        # object.
+        # If no object matches, just add "normally" the object to the list.
+
+        dists = Dists([
+            Dist("FooBar", "1.1", url="external_url", type="source"),
+        ])
+        self.assertEqual(1, len(dists))
+        dists.append(Dist("FooBar", "1.1", url="internal_url",
+                          url_is_external=False, type="source"))
+        self.assertEqual(1, len(dists))
+        self.assertEqual(2, len(dists[0]._urls))
+
+        dists.append(Dist("Foobar", "1.1.1", type="source"))
+        self.assertEqual(2, len(dists))
+
+        # when adding a distribution whith a different type, a new distribution
+        # has to be added.
+        dists.append(Dist("Foobar", "1.1.1", type="binary"))
+        self.assertEqual(3, len(dists))
+
+    def test_prefer_final(self):
+        """Ordering support prefer_final"""
+
+        fb10 = Dist("FooBar", "1.0")  # final distribution
+        fb11a = Dist("FooBar", "1.1a1")  # alpha
+        fb12a = Dist("FooBar", "1.2a1")  # alpha
+        fb12b = Dist("FooBar", "1.2b1")  # beta
+        dists = Dists([fb10, fb11a, fb12a, fb12b])
+
+        dists.sort_distributions(prefer_final=True)
+        self.assertEqual(fb10, dists[0])
+
+        dists.sort_distributions(prefer_final=False)
+        self.assertEqual(fb12b, dists[0])
+
+    def test_prefer_source(self):
+        """Ordering support prefer_source"""
+        fb_source = Dist("FooBar", "1.0", type="source")
+        fb_binary = Dist("FooBar", "1.0", type="binary")
+        fb2_binary = Dist("FooBar", "2.0", type="binary")
+        dists = Dists([fb_binary, fb_source])
+
+        dists.sort_distributions(prefer_source=True)
+        self.assertEqual(fb_source, dists[0])
+
+        dists.sort_distributions(prefer_source=False)
+        self.assertEqual(fb_binary, dists[0])
+
+        dists.append(fb2_binary)
+        dists.sort_distributions(prefer_source=True)
+        self.assertEqual(fb2_binary, dists[0])
+
+    def test_get_same_name_and_version(self):
+        """PyPIDistributions can return a list of "duplicates"
+        """
+        fb_source = Dist("FooBar", "1.0", type="source")
+        fb_binary = Dist("FooBar", "1.0", type="binary")
+        fb2_binary = Dist("FooBar", "2.0", type="binary")
+        dists = Dists([fb_binary, fb_source, fb2_binary])
+        duplicates = dists.get_same_name_and_version()
+        self.assertTrue(1, len(duplicates))
+        self.assertIn(fb_source, duplicates[0])
+
+
+def test_suite():
+    suite = unittest.TestSuite()
+    suite.addTest(unittest.makeSuite(TestPyPIDistribution))
+    suite.addTest(unittest.makeSuite(TestPyPIDistributions))
+    return suite
+
+if __name__ == '__main__':
+    run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_pypi_server.py b/src/distutils2/tests/test_pypi_server.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/test_pypi_server.py
@@ -0,0 +1,70 @@
+"""Tests for distutils.command.bdist."""
+import urllib
+import urllib2
+import os.path
+
+from distutils2.tests.pypi_server import PyPIServer, PYPI_DEFAULT_STATIC_PATH
+from distutils2.tests.support import unittest
+
+
+class PyPIServerTest(unittest.TestCase):
+
+    def test_records_requests(self):
+        """We expect that PyPIServer can log our requests"""
+        server = PyPIServer()
+        server.start()
+        self.assertEqual(len(server.requests), 0)
+
+        data = "Rock Around The Bunker"
+        headers = {"X-test-header": "Mister Iceberg"}
+
+        request = urllib2.Request(server.full_address, data, headers)
+        urllib2.urlopen(request)
+        self.assertEqual(len(server.requests), 1)
+        handler, request_data = server.requests[-1]
+        self.assertIn("Rock Around The Bunker", request_data)
+        self.assertIn("x-test-header", handler.headers.dict)
+        self.assertEqual(handler.headers.dict["x-test-header"],
+                         "Mister Iceberg")
+        server.stop()
+
+    def test_serve_static_content(self):
+        """PYPI Mocked server can serve static content from disk.
+        """
+
+        def uses_local_files_for(server, url_path):
+            """Test that files are served statically (eg. the output from the
+            server is the same than the one made by a simple file read.
+            """
+            url = server.full_address + url_path
+            request = urllib2.Request(url)
+            response = urllib2.urlopen(request)
+            file = open(PYPI_DEFAULT_STATIC_PATH + "/test_pypi_server" +
+               url_path)
+            return response.read() == file.read()
+
+        server = PyPIServer(static_uri_paths=["simple", "external"],
+            static_filesystem_paths=["test_pypi_server"])
+        server.start()
+
+        # the file does not exists on the disc, so it might not be served
+        url = server.full_address + "/simple/unexisting_page"
+        request = urllib2.Request(url)
+        try:
+            urllib2.urlopen(request)
+        except urllib2.HTTPError,e:
+            self.assertEqual(e.code, 404)
+
+        # now try serving a content that do exists
+        self.assertTrue(uses_local_files_for(server, "/simple/index.html"))
+
+        # and another one in another root path
+        self.assertTrue(uses_local_files_for(server, "/external/index.html"))
+        server.stop()
+
+
+def test_suite():
+    return unittest.makeSuite(PyPIServerTest)
+
+if __name__ == '__main__':
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -0,0 +1,290 @@
+"""Tests for the pypi.simple module.
+
+"""
+import sys
+import os
+import shutil
+import tempfile
+import urllib2
+
+from distutils2.tests import support
+from distutils2.tests.support import unittest
+from distutils2.tests.pypi_server import use_pypi_server, PyPIServer, \
+                                         PYPI_DEFAULT_STATIC_PATH
+from distutils2.pypi import simple
+
+from distutils2.errors import DistutilsError
+
+
+class PyPISimpleTestCase(support.TempdirManager,
+                         unittest.TestCase):
+
+    def _get_simple_index(self, server, base_url="/simple/", hosts=None,
+        *args, **kwargs):
+        """Build and return a SimpleSimpleIndex instance, with the test server
+        urls
+        """
+        if hosts is None:
+            hosts = (server.full_address.strip("http://"),)
+        kwargs['hosts'] = hosts
+        return simple.SimpleIndex(server.full_address + base_url, *args,
+            **kwargs)
+
+    def test_bad_urls(self):
+        index = simple.SimpleIndex()
+        url = 'http://127.0.0.1:0/nonesuch/test_simple'
+        try:
+            v = index._open_url(url)
+        except Exception, v:
+            self.assertTrue(url in str(v))
+        else:
+            self.assertTrue(isinstance(v, urllib2.HTTPError))
+
+        # issue 16
+        # easy_install inquant.contentmirror.plone breaks because of a typo
+        # in its home URL
+        index = simple.SimpleIndex(
+            hosts=('www.example.com',))
+
+        url = 'url:%20https://svn.plone.org/svn/collective/inquant.contentmirror.plone/trunk'
+        try:
+            v = index._open_url(url)
+        except Exception, v:
+            self.assertTrue(url in str(v))
+        else:
+            self.assertTrue(isinstance(v, urllib2.HTTPError))
+
+        def _urlopen(*args):
+            import httplib
+            raise httplib.BadStatusLine('line')
+
+        old_urlopen = urllib2.urlopen
+        urllib2.urlopen = _urlopen
+        url = 'http://example.com'
+        try:
+            try:
+                v = index._open_url(url)
+            except Exception, v:
+                self.assertTrue('line' in str(v))
+            else:
+                raise AssertionError('Should have raise here!')
+        finally:
+            urllib2.urlopen = old_urlopen
+
+        # issue 20
+        url = 'http://http://svn.pythonpaste.org/Paste/wphp/trunk'
+        try:
+            index._open_url(url)
+        except Exception, v:
+            self.assertTrue('nonnumeric port' in str(v))
+
+        # issue #160
+        if sys.version_info[0] == 2 and sys.version_info[1] == 7:
+            # this should not fail
+            url = 'http://example.com'
+            page = ('<a href="http://www.famfamfam.com]('
+                    'http://www.famfamfam.com/">')
+            index.process_index(url, page)
+
+    @use_pypi_server("test_found_links")
+    def test_found_links(self, server):
+        """Browse the index, asking for a specified distribution version
+        """
+        # The PyPI index contains links for version 1.0, 1.1, 2.0 and 2.0.1
+        index = self._get_simple_index(server)
+        last_distribution = index.get("foobar")
+
+        # we have scanned the index page
+        self.assertIn(server.full_address + "/simple/foobar/",
+            index._processed_urls)
+
+        # we have found 4 distributions in this page
+        self.assertEqual(len(index._distributions["foobar"]), 4)
+
+        # and returned the most recent one
+        self.assertEqual("%s" % last_distribution.version, '2.0.1')
+
+    def test_is_browsable(self):
+        index = simple.SimpleIndex(follow_externals=False)
+        self.assertTrue(index._is_browsable(index.index_url + "test"))
+
+        # Now, when following externals, we can have a list of hosts to trust.
+        # and don't follow other external links than the one described here.
+        index = simple.SimpleIndex(hosts=["pypi.python.org", "test.org"],
+                                   follow_externals=True)
+        good_urls = (
+            "http://pypi.python.org/foo/bar",
+            "http://pypi.python.org/simple/foobar",
+            "http://test.org",
+            "http://test.org/",
+            "http://test.org/simple/",
+        )
+        bad_urls = (
+            "http://python.org",
+            "http://test.tld",
+        )
+
+        for url in good_urls:
+            self.assertTrue(index._is_browsable(url))
+
+        for url in bad_urls:
+            self.assertFalse(index._is_browsable(url))
+
+        # allow all hosts
+        index = simple.SimpleIndex(follow_externals=True, hosts=("*",))
+        self.assertTrue(index._is_browsable("http://an-external.link/path"))
+        self.assertTrue(index._is_browsable("pypi.test.tld/a/path"))
+
+        # specify a list of hosts we want to allow
+        index = simple.SimpleIndex(follow_externals=True,
+                                   hosts=("*.test.tld",))
+        self.assertFalse(index._is_browsable("http://an-external.link/path"))
+        self.assertTrue(index._is_browsable("http://pypi.test.tld/a/path"))
+
+    @use_pypi_server("with_externals")
+    def test_restrict_hosts(self, server):
+        """Include external pages
+        """
+        # Try to request the package index, wich contains links to "externals"
+        # resources. They have to  be scanned too.
+        index = self._get_simple_index(server, follow_externals=True)
+        index.get("foobar")
+        self.assertIn(server.full_address + "/external/external.html",
+            index._processed_urls)
+
+    @use_pypi_server("with_real_externals")
+    def test_restrict_hosts(self, server):
+        """Only use a list of allowed hosts is possible
+        """
+        # Test that telling the simple pyPI client to not retrieve external
+        # works
+        index = self._get_simple_index(server, follow_externals=False)
+        index.get("foobar")
+        self.assertNotIn(server.full_address + "/external/external.html",
+            index._processed_urls)
+
+    @use_pypi_server("with_egg_files")
+    def test_scan_egg_files(self, server):
+        """Assert that egg files are indexed as well"""
+        pass
+
+    @use_pypi_server(static_filesystem_paths=["with_externals"],
+        static_uri_paths=["simple", "external"])
+    def test_links_priority(self, server):
+        """
+        Download links from the pypi simple index should be used before
+        external download links.
+        http://bitbucket.org/tarek/distribute/issue/163/md5-validation-error
+
+        Usecase :
+        - someone uploads a package on pypi, a md5 is generated
+        - someone manually coindexes this link (with the md5 in the url) onto
+          an external page accessible from the package page.
+        - someone reuploads the package (with a different md5)
+        - while easy_installing, an MD5 error occurs because the external link
+          is used
+        -> The index should use the link from pypi, not the external one.
+        """
+        # start an index server
+        index_url = server.full_address + '/simple/'
+
+        # scan a test index
+        index = simple.SimpleIndex(index_url, follow_externals=True)
+        dists = index.find("foobar")
+        server.stop()
+
+        # we have only one link, because links are compared without md5
+        self.assertEqual(len(dists), 1)
+        # the link should be from the index
+        self.assertEqual('12345678901234567', dists[0].url['hashval'])
+        self.assertEqual('md5', dists[0].url['hashname'])
+
+    @use_pypi_server(static_filesystem_paths=["with_norel_links"],
+        static_uri_paths=["simple", "external"])
+    def test_not_scan_all_links(self, server):
+        """Do not follow all index page links.
+        The links not tagged with rel="download" and rel="homepage" have
+        to not be processed by the package index, while processing "pages".
+        """
+        # process the pages
+        index = self._get_simple_index(server, follow_externals=True)
+        index.find("foobar")
+        # now it should have processed only pages with links rel="download"
+        # and rel="homepage"
+        self.assertIn("%s/simple/foobar/" % server.full_address,
+            index._processed_urls)  # it's the simple index page
+        self.assertIn("%s/external/homepage.html" % server.full_address,
+            index._processed_urls)  # the external homepage is rel="homepage"
+        self.assertNotIn("%s/external/nonrel.html" % server.full_address,
+            index._processed_urls)  # this link contains no rel=*
+        self.assertNotIn("%s/unrelated-0.2.tar.gz" % server.full_address,
+            index._processed_urls)  # linked from simple index (no rel)
+        self.assertIn("%s/foobar-0.1.tar.gz" % server.full_address,
+            index._processed_urls)  # linked from simple index (rel)
+        self.assertIn("%s/foobar-2.0.tar.gz" % server.full_address,
+            index._processed_urls)  # linked from external homepage (rel)
+
+    def test_uses_mirrors(self):
+        """When the main repository seems down, try using the given mirrors"""
+        server = PyPIServer("foo_bar_baz")
+        mirror = PyPIServer("foo_bar_baz")
+        mirror.start()  # we dont start the server here
+
+        try:
+            # create the index using both servers
+            index = simple.SimpleIndex(server.full_address + "/simple/",
+                hosts=('*',), timeout=1,  # set the timeout to 1s for the tests
+                mirrors=[mirror.full_address + "/simple/",])
+
+            # this should not raise a timeout
+            self.assertEqual(4, len(index.find("foo")))
+        finally:
+            mirror.stop()
+
+    def test_simple_link_matcher(self):
+        """Test that the simple link matcher yields the right links"""
+        index = simple.SimpleIndex(follow_externals=False)
+
+        # Here, we define:
+        #   1. one link that must be followed, cause it's a download one
+        #   2. one link that must *not* be followed, cause the is_browsable
+        #      returns false for it.
+        #   3. one link that must be followed cause it's a homepage that is
+        #      browsable
+        self.assertTrue(index._is_browsable("%stest" % index.index_url))
+        self.assertFalse(index._is_browsable("http://dl-link2"))
+        content = """
+        <a href="http://dl-link1" rel="download">download_link1</a>
+        <a href="http://dl-link2" rel="homepage">homepage_link1</a>
+        <a href="%stest" rel="homepage">homepage_link2</a>
+        """ % index.index_url
+
+        # Test that the simple link matcher yield the good links.
+        generator = index._simple_link_matcher(content, index.index_url)
+        self.assertEqual(('http://dl-link1', True), generator.next())
+        self.assertEqual(('%stest' % index.index_url, False),
+                         generator.next())
+        self.assertRaises(StopIteration, generator.next)
+
+        # Follow the external links is possible
+        index.follow_externals = True
+        generator = index._simple_link_matcher(content, index.index_url)
+        self.assertEqual(('http://dl-link1', True), generator.next())
+        self.assertEqual(('http://dl-link2', False), generator.next())
+        self.assertEqual(('%stest' % index.index_url, False),
+                         generator.next())
+        self.assertRaises(StopIteration, generator.next)
+
+    def test_browse_local_files(self):
+        """Test that we can browse local files"""
+        index_path = os.sep.join(["file://" + PYPI_DEFAULT_STATIC_PATH,
+                                  "test_found_links", "simple"])
+        index = simple.SimpleIndex(index_path)
+        dists = index.find("foobar")
+        self.assertEqual(4, len(dists))
+
+def test_suite():
+    return unittest.makeSuite(PyPISimpleTestCase)
+
+if __name__ == '__main__':
+    run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_upload.py b/src/distutils2/tests/test_upload.py
--- a/src/distutils2/tests/test_upload.py
+++ b/src/distutils2/tests/test_upload.py
@@ -1,34 +1,15 @@
 """Tests for distutils.command.upload."""
 # -*- encoding: utf8 -*-
-import sys
-import os
+import os, sys
 
-from distutils2.command import upload as upload_mod
 from distutils2.command.upload import upload
 from distutils2.core import Distribution
 
+from distutils2.tests.pypi_server import PyPIServer, PyPIServerTestCase
 from distutils2.tests import support
 from distutils2.tests.support import unittest
 from distutils2.tests.test_config import PYPIRC, PyPIRCCommandTestCase
 
-PYPIRC_LONG_PASSWORD = """\
-[distutils]
-
-index-servers =
-    server1
-    server2
-
-[server1]
-username:me
-password:aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
-
-[server2]
-username:meagain
-password: secret
-realm:acme
-repository:http://another.pypi/
-"""
-
 
 PYPIRC_NOPASSWORD = """\
 [distutils]
@@ -40,47 +21,18 @@
 username:me
 """
 
-class FakeOpen(object):
-
-    def __init__(self, url):
-        self.url = url
-        if not isinstance(url, str):
-            self.req = url
-        else:
-            self.req = None
-        self.msg = 'OK'
-
-    def getcode(self):
-        return 200
-
-
-class uploadTestCase(PyPIRCCommandTestCase):
-
-    def setUp(self):
-        super(uploadTestCase, self).setUp()
-        self.old_open = upload_mod.urlopen
-        upload_mod.urlopen = self._urlopen
-        self.last_open = None
-
-    def tearDown(self):
-        upload_mod.urlopen = self.old_open
-        super(uploadTestCase, self).tearDown()
-
-    def _urlopen(self, url):
-        self.last_open = FakeOpen(url)
-        return self.last_open
+class UploadTestCase(PyPIServerTestCase, PyPIRCCommandTestCase):
 
     def test_finalize_options(self):
-
         # new format
         self.write_file(self.rc, PYPIRC)
         dist = Distribution()
         cmd = upload(dist)
         cmd.finalize_options()
-        for attr, expected in (('username', 'me'), ('password', 'secret'),
-                               ('realm', 'pypi'),
-                               ('repository', 'http://pypi.python.org/pypi')):
-            self.assertEqual(getattr(cmd, attr), expected)
+        for attr, waited in (('username', 'me'), ('password', 'secret'),
+                             ('realm', 'pypi'),
+                             ('repository', 'http://pypi.python.org/pypi')):
+            self.assertEqual(getattr(cmd, attr), waited)
 
     def test_saved_password(self):
         # file with no password
@@ -89,7 +41,7 @@
         # make sure it passes
         dist = Distribution()
         cmd = upload(dist)
-        cmd.finalize_options()
+        cmd.ensure_finalized()
         self.assertEqual(cmd.password, None)
 
         # make sure we get it as well, if another command
@@ -100,33 +52,31 @@
         self.assertEqual(cmd.password, 'xxx')
 
     def test_upload(self):
-        tmp = self.mkdtemp()
-        path = os.path.join(tmp, 'xxx')
+        path = os.path.join(self.tmp_dir, 'xxx')
         self.write_file(path)
         command, pyversion, filename = 'xxx', '2.6', path
         dist_files = [(command, pyversion, filename)]
-        self.write_file(self.rc, PYPIRC_LONG_PASSWORD)
 
         # lets run it
         pkg_dir, dist = self.create_dist(dist_files=dist_files, author=u'd??d??')
         cmd = upload(dist)
         cmd.ensure_finalized()
+        cmd.repository = self.pypi.full_address
         cmd.run()
 
         # what did we send ?
-        self.assertIn('d??d??', self.last_open.req.data)
-        headers = dict(self.last_open.req.headers)
-        self.assertTrue(int(headers['Content-length']) < 2000)
-        self.assertTrue(headers['Content-type'].startswith('multipart/form-data'))
-        self.assertEqual(self.last_open.req.get_method(), 'POST')
-        self.assertEqual(self.last_open.req.get_full_url(),
-                          'http://pypi.python.org/pypi')
-        self.assertTrue('xxx' in self.last_open.req.data)
-        auth = self.last_open.req.headers['Authorization']
-        self.assertFalse('\n' in auth)
+        handler, request_data = self.pypi.requests[-1]
+        headers = handler.headers.dict
+        self.assertIn('d??d??', request_data)
+        self.assertIn('xxx', request_data)
+        self.assertEqual(int(headers['content-length']), len(request_data))
+        self.assertTrue(int(headers['content-length']) < 2000)
+        self.assertTrue(headers['content-type'].startswith('multipart/form-data'))
+        self.assertEqual(handler.command, 'POST')
+        self.assertNotIn('\n', headers['authorization'])
 
 def test_suite():
-    return unittest.makeSuite(uploadTestCase)
+    return unittest.makeSuite(UploadTestCase)
 
 if __name__ == "__main__":
     unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_upload_docs.py b/src/distutils2/tests/test_upload_docs.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/test_upload_docs.py
@@ -0,0 +1,198 @@
+"""Tests for distutils.command.upload_docs."""
+# -*- encoding: utf8 -*-
+import httplib, os, os.path, shutil, sys, tempfile, zipfile
+from cStringIO import StringIO
+
+from distutils2.command import upload_docs as upload_docs_mod
+from distutils2.command.upload_docs import (upload_docs, zip_dir,
+                                    encode_multipart)
+from distutils2.core import Distribution
+
+from distutils2.errors import DistutilsFileError, DistutilsOptionError
+
+from distutils2.tests import support
+from distutils2.tests.pypi_server import PyPIServer, PyPIServerTestCase
+from distutils2.tests.test_config import PyPIRCCommandTestCase
+from distutils2.tests.support import unittest
+
+
+EXPECTED_MULTIPART_OUTPUT = "\r\n".join([
+'---x',
+'Content-Disposition: form-data; name="a"',
+'',
+'b',
+'---x',
+'Content-Disposition: form-data; name="c"',
+'',
+'d',
+'---x',
+'Content-Disposition: form-data; name="e"; filename="f"',
+'',
+'g',
+'---x',
+'Content-Disposition: form-data; name="h"; filename="i"',
+'',
+'j',
+'---x--',
+'',
+])
+
+PYPIRC = """\
+[distutils]
+index-servers = server1
+
+[server1]
+repository = %s
+username = real_slim_shady
+password = long_island
+"""
+
+class UploadDocsTestCase(PyPIServerTestCase, PyPIRCCommandTestCase):
+
+    def setUp(self):
+        super(UploadDocsTestCase, self).setUp()
+        self.dist = Distribution()
+        self.dist.metadata['Name'] = "distr-name"
+        self.cmd = upload_docs(self.dist)
+
+    def test_default_uploaddir(self):
+        sandbox = tempfile.mkdtemp()
+        previous = os.getcwd()
+        os.chdir(sandbox)
+        try:
+            os.mkdir("build")
+            self.prepare_sample_dir("build")
+            self.cmd.ensure_finalized()
+            self.assertEqual(self.cmd.upload_dir, os.path.join("build", "docs"))
+        finally:
+            os.chdir(previous)
+
+    def prepare_sample_dir(self, sample_dir=None):
+        if sample_dir is None:
+            sample_dir = tempfile.mkdtemp()
+        os.mkdir(os.path.join(sample_dir, "docs"))
+        self.write_file(os.path.join(sample_dir, "docs", "index.html"), "Ce mortel ennui")
+        self.write_file(os.path.join(sample_dir, "index.html"), "Oh la la")
+        return sample_dir
+
+    def test_zip_dir(self):
+        source_dir = self.prepare_sample_dir()
+        compressed = zip_dir(source_dir)
+
+        zip_f = zipfile.ZipFile(compressed)
+        self.assertEqual(zip_f.namelist(), ['index.html', 'docs/index.html'])
+
+    def test_encode_multipart(self):
+        fields = [("a", "b"), ("c", "d")]
+        files = [("e", "f", "g"), ("h", "i", "j")]
+        content_type, body = encode_multipart(fields, files, "-x")
+        self.assertEqual(content_type, "multipart/form-data; boundary=-x")
+        self.assertEqual(body, EXPECTED_MULTIPART_OUTPUT)
+
+    def prepare_command(self):
+        self.cmd.upload_dir = self.prepare_sample_dir()
+        self.cmd.ensure_finalized()
+        self.cmd.repository = self.pypi.full_address
+        self.cmd.username = "username"
+        self.cmd.password = "password"
+
+    def test_upload(self):
+        self.prepare_command()
+        self.cmd.run()
+
+        self.assertEqual(len(self.pypi.requests), 1)
+        handler, request_data = self.pypi.requests[-1]
+        self.assertIn("content", request_data)
+        self.assertIn("Basic", handler.headers.dict['authorization'])
+        self.assertTrue(handler.headers.dict['content-type']
+            .startswith('multipart/form-data;'))
+
+        action, name, content =\
+            request_data.split("----------------GHSKFJDLGDS7543FJKLFHRE75642756743254")[1:4]
+
+        # check that we picked the right chunks
+        self.assertIn('name=":action"', action)
+        self.assertIn('name="name"', name)
+        self.assertIn('name="content"', content)
+
+        # check their contents
+        self.assertIn("doc_upload", action)
+        self.assertIn("distr-name", name)
+        self.assertIn("docs/index.html", content)
+        self.assertIn("Ce mortel ennui", content)
+
+    def test_https_connection(self):
+        https_called = False
+        orig_https = upload_docs_mod.httplib.HTTPSConnection
+        def https_conn_wrapper(*args):
+            https_called = True
+            return upload_docs_mod.httplib.HTTPConnection(*args) # the testing server is http
+        upload_docs_mod.httplib.HTTPSConnection = https_conn_wrapper
+        try:
+            self.prepare_command()
+            self.cmd.run()
+            self.assertFalse(https_called)
+
+            self.cmd.repository = self.cmd.repository.replace("http", "https")
+            self.cmd.run()
+            self.assertFalse(https_called)
+        finally:
+            upload_docs_mod.httplib.HTTPSConnection = orig_https
+
+    def test_handling_response(self):
+        calls = []
+        def aggr(*args):
+            calls.append(args)
+        self.pypi.default_response_status = '403 Forbidden'
+        self.prepare_command()
+        self.cmd.announce = aggr
+        self.cmd.run()
+        message, _ = calls[-1]
+        self.assertIn('Upload failed (403): Forbidden', message)
+
+        calls = []
+        self.pypi.default_response_status = '301 Moved Permanently'
+        self.pypi.default_response_headers.append(("Location", "brand_new_location"))
+        self.cmd.run()
+        message, _ = calls[-1]
+        self.assertIn('brand_new_location', message)
+
+    def test_reads_pypirc_data(self):
+        self.write_file(self.rc, PYPIRC % self.pypi.full_address)
+        self.cmd.repository = self.pypi.full_address
+        self.cmd.upload_dir = self.prepare_sample_dir()
+        self.cmd.ensure_finalized()
+        self.assertEqual(self.cmd.username, "real_slim_shady")
+        self.assertEqual(self.cmd.password, "long_island")
+
+    def test_checks_index_html_presence(self):
+        self.cmd.upload_dir = self.prepare_sample_dir()
+        os.remove(os.path.join(self.cmd.upload_dir, "index.html"))
+        self.assertRaises(DistutilsFileError, self.cmd.ensure_finalized)
+
+    def test_checks_upload_dir(self):
+        self.cmd.upload_dir = self.prepare_sample_dir()
+        shutil.rmtree(os.path.join(self.cmd.upload_dir))
+        self.assertRaises(DistutilsOptionError, self.cmd.ensure_finalized)
+
+    def test_show_response(self):
+        orig_stdout = sys.stdout
+        write_args = []
+        class MockStdIn(object):
+            def write(self, arg):
+                write_args.append(arg)
+        sys.stdout = MockStdIn()
+        try:
+            self.prepare_command()
+            self.cmd.show_response = True
+            self.cmd.run()
+        finally:
+            sys.stdout = orig_stdout
+        self.assertTrue(write_args[0], "should report the response")
+        self.assertIn(self.pypi.default_response_data + "\n", write_args[0])
+
+def test_suite():
+    return unittest.makeSuite(UploadDocsTestCase)
+
+if __name__ == "__main__":
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_version.py b/src/distutils2/tests/test_version.py
--- a/src/distutils2/tests/test_version.py
+++ b/src/distutils2/tests/test_version.py
@@ -154,6 +154,17 @@
         # XXX need to silent the micro version in this case
         #assert not VersionPredicate('Ho (<3.0,!=2.6)').match('2.6.3')
 
+    def test_is_final(self):
+        # VersionPredicate knows is a distribution is a final one or not.
+        final_versions = ('1.0', '1.0.post456')
+        other_versions = ('1.0.dev1', '1.0a2', '1.0c3')
+
+        for version in final_versions:
+            self.assertTrue(V(version).is_final)
+        for version in other_versions:
+            self.assertFalse(V(version).is_final)
+
+
 def test_suite():
     #README = os.path.join(os.path.dirname(__file__), 'README.txt')
     #suite = [doctest.DocFileSuite(README), unittest.makeSuite(VersionTestCase)]
@@ -162,4 +173,3 @@
 
 if __name__ == "__main__":
     unittest.main(defaultTest="test_suite")
-
diff --git a/src/distutils2/version.py b/src/distutils2/version.py
--- a/src/distutils2/version.py
+++ b/src/distutils2/version.py
@@ -36,6 +36,7 @@
     (?P<postdev>(\.post(?P<post>\d+))?(\.dev(?P<dev>\d+))?)?
     $''', re.VERBOSE)
 
+
 class NormalizedVersion(object):
     """A rational version.
 
@@ -61,7 +62,8 @@
         @param error_on_huge_major_num {bool} Whether to consider an
             apparent use of a year or full date as the major version number
             an error. Default True. One of the observed patterns on PyPI before
-            the introduction of `NormalizedVersion` was version numbers like this:
+            the introduction of `NormalizedVersion` was version numbers like
+            this:
                 2009.01.03
                 20040603
                 2005.01
@@ -71,6 +73,7 @@
             the possibility of using a version number like "1.0" (i.e.
             where the major number is less than that huge major number).
         """
+        self.is_final = True  # by default, consider a version as final.
         self._parse(s, error_on_huge_major_num)
 
     @classmethod
@@ -101,6 +104,7 @@
             block += self._parse_numdots(groups.get('prerelversion'), s,
                                          pad_zeros_length=1)
             parts.append(tuple(block))
+            self.is_final = False
         else:
             parts.append(_FINAL_MARKER)
 
@@ -115,6 +119,7 @@
                     postdev.append(_FINAL_MARKER[0])
             if dev is not None:
                 postdev.extend(['dev', int(dev)])
+                self.is_final = False
             parts.append(tuple(postdev))
         else:
             parts.append(_FINAL_MARKER)
@@ -204,6 +209,7 @@
     # See http://docs.python.org/reference/datamodel#object.__hash__
     __hash__ = object.__hash__
 
+
 def suggest_normalized_version(s):
     """Suggest a normalized version close to the given version string.
 
@@ -215,7 +221,7 @@
     on observation of versions currently in use on PyPI. Given a dump of
     those version during PyCon 2009, 4287 of them:
     - 2312 (53.93%) match NormalizedVersion without change
-    - with the automatic suggestion
+      with the automatic suggestion
     - 3474 (81.04%) match when using this suggestion method
 
     @param s {str} An irrational version string.
@@ -305,7 +311,6 @@
     # PyPI stats: ~21 (0.62%) better
     rs = re.sub(r"\.?(pre|preview|-c)(\d+)$", r"c\g<2>", rs)
 
-
     # Tcl/Tk uses "px" for their post release markers
     rs = re.sub(r"p(\d+)$", r".post\1", rs)
 
@@ -322,6 +327,7 @@
 _PLAIN_VERSIONS = re.compile(r"^\s*(.*)\s*$")
 _SPLIT_CMP = re.compile(r"^\s*(<=|>=|<|>|!=|==)\s*([^\s,]+)\s*$")
 
+
 def _split_predicate(predicate):
     match = _SPLIT_CMP.match(predicate)
     if match is None:
@@ -368,6 +374,7 @@
                 return False
         return True
 
+
 class _Versions(VersionPredicate):
     def __init__(self, predicate):
         predicate = predicate.strip()
@@ -379,6 +386,7 @@
         self.predicates = [_split_predicate(pred.strip())
                            for pred in predicates.split(',')]
 
+
 class _Version(VersionPredicate):
     def __init__(self, predicate):
         predicate = predicate.strip()
@@ -388,6 +396,7 @@
         self.name = None
         self.predicates = _split_predicate(match.groups()[0])
 
+
 def is_valid_predicate(predicate):
     try:
         VersionPredicate(predicate)
@@ -396,6 +405,7 @@
     else:
         return True
 
+
 def is_valid_versions(predicate):
     try:
         _Versions(predicate)
@@ -404,6 +414,7 @@
     else:
         return True
 
+
 def is_valid_version(predicate):
     try:
         _Version(predicate)
@@ -411,4 +422,3 @@
         return False
     else:
         return True
-
diff --git a/src/setup.py b/src/setup.py
--- a/src/setup.py
+++ b/src/setup.py
@@ -3,8 +3,12 @@
 __revision__ = "$Id$"
 import sys
 import os
+import re
 
-from distutils2.core import setup
+from distutils2 import __version__ as VERSION
+from distutils2 import log
+from distutils2.core import setup, Extension
+from distutils2.compiler.ccompiler import new_compiler
 from distutils2.command.sdist import sdist
 from distutils2.command.install import install
 from distutils2 import __version__ as VERSION
@@ -31,6 +35,7 @@
 
 DEV_SUFFIX = '.dev%d' % get_tip_revision('..')
 
+
 class install_hg(install):
 
     user_options = install.user_options + [
@@ -62,10 +67,141 @@
             self.distribution.metadata.version += DEV_SUFFIX
         sdist.run(self)
 
+
+# additional paths to check, set from the command line
+SSL_INCDIR = ''   # --openssl-incdir=
+SSL_LIBDIR = ''   # --openssl-libdir=
+SSL_DIR = ''      # --openssl-prefix=
+
+def add_dir_to_list(dirlist, dir):
+    """Add the directory 'dir' to the list 'dirlist' (at the front) if
+    'dir' actually exists and is a directory.  If 'dir' is already in
+    'dirlist' it is moved to the front.
+    """
+    if dir is not None and os.path.isdir(dir) and dir not in dirlist:
+        if dir in dirlist:
+            dirlist.remove(dir)
+        dirlist.insert(0, dir)
+
+
+def prepare_hashlib_extensions():
+    """Decide which C extensions to build and create the appropriate
+    Extension objects to build them.  Return a list of Extensions.
+    """
+    # this CCompiler object is only used to locate include files
+    compiler = new_compiler()
+
+    # Ensure that these paths are always checked
+    if os.name == 'posix':
+        add_dir_to_list(compiler.library_dirs, '/usr/local/lib')
+        add_dir_to_list(compiler.include_dirs, '/usr/local/include')
+
+        add_dir_to_list(compiler.library_dirs, '/usr/local/ssl/lib')
+        add_dir_to_list(compiler.include_dirs, '/usr/local/ssl/include')
+
+        add_dir_to_list(compiler.library_dirs, '/usr/contrib/ssl/lib')
+        add_dir_to_list(compiler.include_dirs, '/usr/contrib/ssl/include')
+
+        add_dir_to_list(compiler.library_dirs, '/usr/lib')
+        add_dir_to_list(compiler.include_dirs, '/usr/include')
+
+    # look in command line supplied paths
+    if SSL_LIBDIR:
+        add_dir_to_list(compiler.library_dirs, SSL_LIBDIR)
+    if SSL_INCDIR:
+        add_dir_to_list(compiler.include_dirs, SSL_INCDIR)
+    if SSL_DIR:
+        if os.name == 'nt':
+            add_dir_to_list(compiler.library_dirs, os.path.join(SSL_DIR, 'out32dll'))
+            # prefer the static library
+            add_dir_to_list(compiler.library_dirs, os.path.join(SSL_DIR, 'out32'))
+        else:
+            add_dir_to_list(compiler.library_dirs, os.path.join(SSL_DIR, 'lib'))
+        add_dir_to_list(compiler.include_dirs, os.path.join(SSL_DIR, 'include'))
+
+    oslibs = {'posix': ['ssl', 'crypto'],
+              'nt': ['libeay32',  'gdi32', 'advapi32', 'user32']}
+
+    if os.name not in oslibs:
+        sys.stderr.write(
+            'unknown operating system, impossible to compile _hashlib')
+        sys.exit(1)
+
+    exts = []
+
+    ssl_inc_dirs = []
+    ssl_incs = []
+    for inc_dir in compiler.include_dirs:
+        f = os.path.join(inc_dir, 'openssl', 'ssl.h')
+        if os.path.exists(f):
+            ssl_incs.append(f)
+            ssl_inc_dirs.append(inc_dir)
+
+    ssl_lib = compiler.find_library_file(compiler.library_dirs, oslibs[os.name][0])
+
+    # find out which version of OpenSSL we have
+    openssl_ver = 0
+    openssl_ver_re = re.compile(
+        '^\s*#\s*define\s+OPENSSL_VERSION_NUMBER\s+(0x[0-9a-fA-F]+)' )
+    ssl_inc_dir = ''
+    for ssl_inc_dir in ssl_inc_dirs:
+        name = os.path.join(ssl_inc_dir, 'openssl', 'opensslv.h')
+        if os.path.isfile(name):
+            try:
+                incfile = open(name, 'r')
+                for line in incfile:
+                    m = openssl_ver_re.match(line)
+                    if m:
+                        openssl_ver = int(m.group(1), 16)
+                        break
+            except IOError:
+                pass
+
+        # first version found is what we'll use
+        if openssl_ver:
+            break
+
+    if (ssl_inc_dir and ssl_lib is not None and openssl_ver >= 0x00907000):
+
+        log.info('Using OpenSSL version 0x%08x from', openssl_ver)
+        log.info(' Headers:\t%s', ssl_inc_dir)
+        log.info(' Library:\t%s', ssl_lib)
+
+        # The _hashlib module wraps optimized implementations
+        # of hash functions from the OpenSSL library.
+        exts.append(Extension('_hashlib', ['_hashopenssl.c'],
+                              include_dirs = [ssl_inc_dir],
+                              library_dirs = [os.path.dirname(ssl_lib)],
+                              libraries = oslibs[os.name]))
+    else:
+        exts.append(Extension('_sha', ['shamodule.c']) )
+        exts.append(Extension('_md5',
+                              sources=['md5module.c', 'md5.c'],
+                              depends=['md5.h']) )
+
+    if (not ssl_lib or openssl_ver < 0x00908000):
+        # OpenSSL doesn't do these until 0.9.8 so we'll bring our own
+        exts.append(Extension('_sha256', ['sha256module.c']))
+        exts.append(Extension('_sha512', ['sha512module.c']))
+
+    def prepend_modules(filename):
+        return os.path.join('Modules', filename)
+
+    # all the C code is in the Modules subdirectory, prepend the path
+    for ext in exts:
+        ext.sources = [prepend_modules(fn) for fn in ext.sources]
+        if hasattr(ext, 'depends') and ext.depends is not None:
+            ext.depends = [prepend_modules(fn) for fn in ext.depends]
+
+    return exts
+
 setup_kwargs = {}
 if sys.version < '2.6':
     setup_kwargs['scripts'] = ['distutils2/mkpkg.py']
 
+if sys.version < '2.5':
+    setup_kwargs['ext_modules'] = prepare_hashlib_extensions()
+
 _CLASSIFIERS = """\
 Development Status :: 3 - Alpha
 Intended Audience :: Developers
@@ -77,26 +213,23 @@
 Topic :: System :: Systems Administration
 Topic :: Utilities"""
 
-setup (name="Distutils2",
-       version=VERSION,
-       summary="Python Distribution Utilities",
-       keywords=['packaging', 'distutils'],
-       author="Tarek Ziade",
-       author_email="tarek at ziade.org",
-       home_page="http://bitbucket.org/tarek/distutils2/wiki/Home",
-       license="PSF",
-       description=README,
-       classifier=_CLASSIFIERS.split('\n'),
-       packages=find_packages(),
-       cmdclass={'sdist': sdist_hg, 'install': install_hg},
-       package_data={'distutils2._backport': ['sysconfig.cfg']},
-       project_url=[('Mailing list',
+setup(name="Distutils2",
+      version=VERSION,
+      summary="Python Distribution Utilities",
+      keywords=['packaging', 'distutils'],
+      author="Tarek Ziade",
+      author_email="tarek at ziade.org",
+      home_page="http://bitbucket.org/tarek/distutils2/wiki/Home",
+      license="PSF",
+      description=README,
+      classifier=_CLASSIFIERS.split('\n'),
+      packages=find_packages(),
+      cmdclass={'sdist': sdist_hg, 'install': install_hg},
+      package_data={'distutils2._backport': ['sysconfig.cfg']},
+      project_url=[('Mailing-list',
                     'http://mail.python.org/mailman/listinfo/distutils-sig/'),
-                    ('Documentation',
-                     'http://packages.python.org/Distutils2'),
-                    ('Repository', 'http://hg.python.org/distutils2'),
-                    ('Bug tracker', 'http://bugs.python.org')],
-       **setup_kwargs
-       )
-
-
+                   ('Documentation',
+                    'http://packages.python.org/Distutils2'),
+                   ('Repository', 'http://hg.python.org/distutils2'),
+                   ('Bug tracker', 'http://bugs.python.org')],
+      **setup_kwargs)

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul  4 12:15:11 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun,  4 Jul 2010 12:15:11 +0200 (CEST)
Subject: [Python-checkins] r82527 - python/branches/py3k/Modules/audioop.c
Message-ID: <20100704101511.621C9EE9AA@mail.python.org>

Author: mark.dickinson
Date: Sun Jul  4 12:15:11 2010
New Revision: 82527

Log:
Fix refleak in Modules/audioop.c.

Modified:
   python/branches/py3k/Modules/audioop.c

Modified: python/branches/py3k/Modules/audioop.c
==============================================================================
--- python/branches/py3k/Modules/audioop.c	(original)
+++ python/branches/py3k/Modules/audioop.c	Sun Jul  4 12:15:11 2010
@@ -807,10 +807,13 @@
         return 0;
     cp = pcp.buf;
     len = pcp.len;
-    if (!audioop_check_parameters(len, size))
+    if (!audioop_check_parameters(len, size)) {
+        PyBuffer_Release(&pcp);
         return NULL;
+    }
     if (((len / size) & 1) != 0) {
         PyErr_SetString(AudioopError, "not a whole number of frames");
+        PyBuffer_Release(&pcp);
         return NULL;
     }
 
@@ -824,8 +827,10 @@
     }
 
     rv = PyBytes_FromStringAndSize(NULL, len/2);
-    if ( rv == 0 )
+    if ( rv == 0 ) {
+        PyBuffer_Release(&pcp);
         return 0;
+    }
     ncp = (signed char *)PyBytes_AsString(rv);
 
 

From python-checkins at python.org  Sun Jul  4 12:15:59 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun,  4 Jul 2010 12:15:59 +0200 (CEST)
Subject: [Python-checkins] r82528 - in python/branches/release31-maint:
	Modules/audioop.c
Message-ID: <20100704101559.285D2EE9B2@mail.python.org>

Author: mark.dickinson
Date: Sun Jul  4 12:15:59 2010
New Revision: 82528

Log:
Merged revisions 82527 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82527 | mark.dickinson | 2010-07-04 11:15:11 +0100 (Sun, 04 Jul 2010) | 1 line
  
  Fix refleak in Modules/audioop.c.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Modules/audioop.c

Modified: python/branches/release31-maint/Modules/audioop.c
==============================================================================
--- python/branches/release31-maint/Modules/audioop.c	(original)
+++ python/branches/release31-maint/Modules/audioop.c	Sun Jul  4 12:15:59 2010
@@ -804,10 +804,13 @@
         return 0;
     cp = pcp.buf;
     len = pcp.len;
-    if (!audioop_check_parameters(len, size))
+    if (!audioop_check_parameters(len, size)) {
+        PyBuffer_Release(&pcp);
         return NULL;
+    }
     if (((len / size) & 1) != 0) {
         PyErr_SetString(AudioopError, "not a whole number of frames");
+        PyBuffer_Release(&pcp);
         return NULL;
     }
 
@@ -821,8 +824,10 @@
     }
 
     rv = PyBytes_FromStringAndSize(NULL, len/2);
-    if ( rv == 0 )
+    if ( rv == 0 ) {
+        PyBuffer_Release(&pcp);
         return 0;
+    }
     ncp = (signed char *)PyBytes_AsString(rv);
 
 

From python-checkins at python.org  Sun Jul  4 16:24:41 2010
From: python-checkins at python.org (florent.xicluna)
Date: Sun,  4 Jul 2010 16:24:41 +0200 (CEST)
Subject: [Python-checkins] r82529 -
	python/branches/release27-maint/Lib/test/test_coercion.py
Message-ID: <20100704142441.02936EE98C@mail.python.org>

Author: florent.xicluna
Date: Sun Jul  4 16:24:40 2010
New Revision: 82529

Log:
Issue #9145: Fix a regression due to r79539

Modified:
   python/branches/release27-maint/Lib/test/test_coercion.py

Modified: python/branches/release27-maint/Lib/test/test_coercion.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_coercion.py	(original)
+++ python/branches/release27-maint/Lib/test/test_coercion.py	Sun Jul  4 16:24:40 2010
@@ -2,6 +2,7 @@
 import unittest
 from test.test_support import run_unittest, TestFailed, check_warnings
 
+
 # Fake a number that implements numeric methods through __coerce__
 class CoerceNumber:
     def __init__(self, arg):
@@ -222,6 +223,11 @@
             infix_results[key] = res
 
 
+with check_warnings(("classic (int|long) division", DeprecationWarning),
+                    quiet=True):
+    process_infix_results()
+    # now infix_results has two lists of results for every pairing.
+
 prefix_binops = [ 'divmod' ]
 prefix_results = [
     [(1,0), (1L,0L), (0.0,2.0), ((1+0j),0j), TE, TE, TE, TE, (1,0)],
@@ -336,9 +342,6 @@
                          DeprecationWarning),
                         ("classic (int|long) division", DeprecationWarning),
                         quiet=True):
-        process_infix_results()
-        # now infix_results has two lists of results for every pairing.
-
         run_unittest(CoercionTest)
 
 if __name__ == "__main__":

From python-checkins at python.org  Sun Jul  4 18:11:41 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sun,  4 Jul 2010 18:11:41 +0200 (CEST)
Subject: [Python-checkins] r82530 - in sandbox/trunk/2to3/lib2to3:
	fixes/fix_itertools_imports.py tests/test_fixers.py
Message-ID: <20100704161141.4D439EE982@mail.python.org>

Author: benjamin.peterson
Date: Sun Jul  4 18:11:41 2010
New Revision: 82530

Log:
simplify ignore star imports from itertools #8892

Modified:
   sandbox/trunk/2to3/lib2to3/fixes/fix_itertools_imports.py
   sandbox/trunk/2to3/lib2to3/tests/test_fixers.py

Modified: sandbox/trunk/2to3/lib2to3/fixes/fix_itertools_imports.py
==============================================================================
--- sandbox/trunk/2to3/lib2to3/fixes/fix_itertools_imports.py	(original)
+++ sandbox/trunk/2to3/lib2to3/fixes/fix_itertools_imports.py	Sun Jul  4 18:11:41 2010
@@ -20,6 +20,9 @@
             if child.type == token.NAME:
                 member = child.value
                 name_node = child
+            elif child.type == token.STAR:
+                # Just leave the import as is.
+                return
             else:
                 assert child.type == syms.import_as_name
                 name_node = child.children[0]

Modified: sandbox/trunk/2to3/lib2to3/tests/test_fixers.py
==============================================================================
--- sandbox/trunk/2to3/lib2to3/tests/test_fixers.py	(original)
+++ sandbox/trunk/2to3/lib2to3/tests/test_fixers.py	Sun Jul  4 18:11:41 2010
@@ -3670,6 +3670,10 @@
         a = "from itertools import bar, filterfalse, foo"
         self.check(b, a)
 
+    def test_import_star(self):
+        s = "from itertools import *"
+        self.unchanged(s)
+
 
     def test_unchanged(self):
         s = "from itertools import foo"

From python-checkins at python.org  Sun Jul  4 18:13:21 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sun,  4 Jul 2010 18:13:21 +0200 (CEST)
Subject: [Python-checkins] r82531 -
	sandbox/trunk/2to3/lib2to3/fixes/fix_itertools_imports.py
Message-ID: <20100704161321.36C68EE9D7@mail.python.org>

Author: benjamin.peterson
Date: Sun Jul  4 18:13:20 2010
New Revision: 82531

Log:
wrap with parenthesis not \

Modified:
   sandbox/trunk/2to3/lib2to3/fixes/fix_itertools_imports.py

Modified: sandbox/trunk/2to3/lib2to3/fixes/fix_itertools_imports.py
==============================================================================
--- sandbox/trunk/2to3/lib2to3/fixes/fix_itertools_imports.py	(original)
+++ sandbox/trunk/2to3/lib2to3/fixes/fix_itertools_imports.py	Sun Jul  4 18:13:20 2010
@@ -47,8 +47,8 @@
             children[-1].remove()
 
         # If there are no imports left, just get rid of the entire statement
-        if not (imports.children or getattr(imports, 'value', None)) or \
-                imports.parent is None:
+        if (not (imports.children or getattr(imports, 'value', None)) or
+            imports.parent is None):
             p = node.prefix
             node = BlankLine()
             node.prefix = p

From python-checkins at python.org  Sun Jul  4 18:23:54 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun,  4 Jul 2010 18:23:54 +0200 (CEST)
Subject: [Python-checkins] r82532 - in python/branches/release27-maint:
	Lib/test/test_parser.py Misc/NEWS Modules/parsermodule.c
Message-ID: <20100704162354.D4518EE9DB@mail.python.org>

Author: mark.dickinson
Date: Sun Jul  4 18:23:54 2010
New Revision: 82532

Log:
Issue #9128:  Validate class decorator syntax correctly in parser module.


Modified:
   python/branches/release27-maint/Lib/test/test_parser.py
   python/branches/release27-maint/Misc/NEWS
   python/branches/release27-maint/Modules/parsermodule.c

Modified: python/branches/release27-maint/Lib/test/test_parser.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_parser.py	(original)
+++ python/branches/release27-maint/Lib/test/test_parser.py	Sun Jul  4 18:23:54 2010
@@ -180,6 +180,14 @@
 
     def test_class_defs(self):
         self.check_suite("class foo():pass")
+        self.check_suite("@class_decorator\n"
+                         "class foo():pass")
+        self.check_suite("@class_decorator(arg)\n"
+                         "class foo():pass")
+        self.check_suite("@decorator1\n"
+                         "@decorator2\n"
+                         "class foo():pass")
+
 
     def test_import_from_statement(self):
         self.check_suite("from sys.path import *")

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Sun Jul  4 18:23:54 2010
@@ -16,6 +16,7 @@
 Library
 -------
 
+- Issue #9128: Fix validation of class decorators in parser module.
 
 Extension Modules
 -----------------

Modified: python/branches/release27-maint/Modules/parsermodule.c
==============================================================================
--- python/branches/release27-maint/Modules/parsermodule.c	(original)
+++ python/branches/release27-maint/Modules/parsermodule.c	Sun Jul  4 18:23:54 2010
@@ -2682,14 +2682,15 @@
 static int
 validate_decorated(node *tree)
 {
-  int nch = NCH(tree);
-  int ok = (validate_ntype(tree, decorated)
-            && (nch == 2)
-            && validate_decorators(RCHILD(tree, -2))
-            && (validate_funcdef(RCHILD(tree, -1))
-                || validate_class(RCHILD(tree, -1)))
-            );
-  return ok;
+    int nch = NCH(tree);
+    int ok = (validate_ntype(tree, decorated)
+              && (nch == 2)
+              && validate_decorators(RCHILD(tree, -2)));
+    if (TYPE(RCHILD(tree, -1)) == funcdef)
+        ok = ok && validate_funcdef(RCHILD(tree, -1));
+    else
+        ok = ok && validate_class(RCHILD(tree, -1));
+    return ok;
 }
 
 static int

From python-checkins at python.org  Sun Jul  4 18:28:09 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Sun,  4 Jul 2010 18:28:09 +0200 (CEST)
Subject: [Python-checkins] r82533 -
	python/branches/py3k/Modules/datetimemodule.c
Message-ID: <20100704162809.02961EE9D8@mail.python.org>

Author: alexander.belopolsky
Date: Sun Jul  4 18:28:08 2010
New Revision: 82533

Log:
Issue #9152: Removed dead code in datetime module

Modified:
   python/branches/py3k/Modules/datetimemodule.c

Modified: python/branches/py3k/Modules/datetimemodule.c
==============================================================================
--- python/branches/py3k/Modules/datetimemodule.c	(original)
+++ python/branches/py3k/Modules/datetimemodule.c	Sun Jul  4 18:28:08 2010
@@ -1023,11 +1023,6 @@
             Py_DECREF(result);
             result = NULL;
         }
-        else if (!PyUnicode_Check(result)) {
-            PyObject *temp = PyUnicode_FromObject(result);
-            Py_DECREF(result);
-            result = temp;
-        }
     }
     return result;
 }

From python-checkins at python.org  Sun Jul  4 18:28:57 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun,  4 Jul 2010 18:28:57 +0200 (CEST)
Subject: [Python-checkins] r82534 - in python/branches/release26-maint:
	Lib/test/test_parser.py Misc/NEWS Modules/parsermodule.c
Message-ID: <20100704162857.3FB6BEE9E1@mail.python.org>

Author: mark.dickinson
Date: Sun Jul  4 18:28:57 2010
New Revision: 82534

Log:
Issue #9128: Fix validation of class decorators in parser module.


Modified:
   python/branches/release26-maint/Lib/test/test_parser.py
   python/branches/release26-maint/Misc/NEWS
   python/branches/release26-maint/Modules/parsermodule.c

Modified: python/branches/release26-maint/Lib/test/test_parser.py
==============================================================================
--- python/branches/release26-maint/Lib/test/test_parser.py	(original)
+++ python/branches/release26-maint/Lib/test/test_parser.py	Sun Jul  4 18:28:57 2010
@@ -156,6 +156,14 @@
 
     def test_class_defs(self):
         self.check_suite("class foo():pass")
+        self.check_suite("@class_decorator\n"
+                         "class foo():pass")
+        self.check_suite("@class_decorator(arg)\n"
+                         "class foo():pass")
+        self.check_suite("@decorator1\n"
+                         "@decorator2\n"
+                         "class foo():pass")
+
 
     def test_import_from_statement(self):
         self.check_suite("from sys.path import *")

Modified: python/branches/release26-maint/Misc/NEWS
==============================================================================
--- python/branches/release26-maint/Misc/NEWS	(original)
+++ python/branches/release26-maint/Misc/NEWS	Sun Jul  4 18:28:57 2010
@@ -81,6 +81,8 @@
 Library
 -------
 
+- Issue #9128: Fix validation of class decorators in parser module.
+
 - Issue #7673: Fix security vulnerability (CVE-2010-2089) in the audioop
   module, ensure that the input string length is a multiple of the frame size
 

Modified: python/branches/release26-maint/Modules/parsermodule.c
==============================================================================
--- python/branches/release26-maint/Modules/parsermodule.c	(original)
+++ python/branches/release26-maint/Modules/parsermodule.c	Sun Jul  4 18:28:57 2010
@@ -2679,14 +2679,15 @@
 static int
 validate_decorated(node *tree)
 {
-  int nch = NCH(tree);
-  int ok = (validate_ntype(tree, decorated)
-            && (nch == 2)
-            && validate_decorators(RCHILD(tree, -2))
-            && (validate_funcdef(RCHILD(tree, -1))
-                || validate_class(RCHILD(tree, -1)))
-            );
-  return ok;
+    int nch = NCH(tree);
+    int ok = (validate_ntype(tree, decorated)
+              && (nch == 2)
+              && validate_decorators(RCHILD(tree, -2)));
+    if (TYPE(RCHILD(tree, -1)) == funcdef)
+        ok = ok && validate_funcdef(RCHILD(tree, -1));
+    else
+        ok = ok && validate_class(RCHILD(tree, -1));
+    return ok;
 }
 
 static int

From python-checkins at python.org  Sun Jul  4 18:33:07 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sun,  4 Jul 2010 18:33:07 +0200 (CEST)
Subject: [Python-checkins] r82535 - python/branches/py3k/Lib/lib2to3
Message-ID: <20100704163307.4A145F68C3@mail.python.org>

Author: benjamin.peterson
Date: Sun Jul  4 18:33:07 2010
New Revision: 82535

Log:
Removed merge tracking for "svnmerge" for 
svn+ssh://pythondev at svn.python.org/sandbox/trunk/2to3/lib2to3


Modified:
   python/branches/py3k/Lib/lib2to3/   (props changed)

From python-checkins at python.org  Sun Jul  4 18:33:25 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Sun,  4 Jul 2010 18:33:25 +0200 (CEST)
Subject: [Python-checkins] r82536 - in python/branches/release31-maint:
	Modules/datetimemodule.c
Message-ID: <20100704163325.A1BA1EE9B6@mail.python.org>

Author: alexander.belopolsky
Date: Sun Jul  4 18:33:25 2010
New Revision: 82536

Log:
Merged revisions 82533 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82533 | alexander.belopolsky | 2010-07-04 12:28:08 -0400 (Sun, 04 Jul 2010) | 1 line
  
  Issue #9152: Removed dead code in datetime module
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Modules/datetimemodule.c

Modified: python/branches/release31-maint/Modules/datetimemodule.c
==============================================================================
--- python/branches/release31-maint/Modules/datetimemodule.c	(original)
+++ python/branches/release31-maint/Modules/datetimemodule.c	Sun Jul  4 18:33:25 2010
@@ -957,11 +957,6 @@
             Py_DECREF(result);
             result = NULL;
         }
-        else if (!PyUnicode_Check(result)) {
-            PyObject *temp = PyUnicode_FromObject(result);
-            Py_DECREF(result);
-            result = temp;
-        }
     }
     return result;
 }

From python-checkins at python.org  Sun Jul  4 18:37:31 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun,  4 Jul 2010 18:37:31 +0200 (CEST)
Subject: [Python-checkins] r82537 - in python/branches/py3k:
	Lib/test/test_parser.py Misc/NEWS Modules/parsermodule.c
Message-ID: <20100704163731.9EF2DEE9D8@mail.python.org>

Author: mark.dickinson
Date: Sun Jul  4 18:37:31 2010
New Revision: 82537

Log:
Issue #9128: Fix validation of class decorators in parser module.


Modified:
   python/branches/py3k/Lib/test/test_parser.py
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/Modules/parsermodule.c

Modified: python/branches/py3k/Lib/test/test_parser.py
==============================================================================
--- python/branches/py3k/Lib/test/test_parser.py	(original)
+++ python/branches/py3k/Lib/test/test_parser.py	Sun Jul  4 18:37:31 2010
@@ -149,6 +149,13 @@
     def test_class_defs(self):
         self.check_suite("class foo():pass")
         self.check_suite("class foo(object):pass")
+        self.check_suite("@class_decorator\n"
+                         "class foo():pass")
+        self.check_suite("@class_decorator(arg)\n"
+                         "class foo():pass")
+        self.check_suite("@decorator1\n"
+                         "@decorator2\n"
+                         "class foo():pass")
 
     def test_import_from_statement(self):
         self.check_suite("from sys.path import *")

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sun Jul  4 18:37:31 2010
@@ -468,6 +468,8 @@
 Library
 -------
 
+- Issue #9128: Fix validation of class decorators in parser module.
+
 - Issue #9094: python -m pickletools will now disassemble pickle files
   listed in the command line arguments.  See output of python -m
   pickletools -h for more details.

Modified: python/branches/py3k/Modules/parsermodule.c
==============================================================================
--- python/branches/py3k/Modules/parsermodule.c	(original)
+++ python/branches/py3k/Modules/parsermodule.c	Sun Jul  4 18:37:31 2010
@@ -2509,14 +2509,15 @@
 static int
 validate_decorated(node *tree)
 {
-  int nch = NCH(tree);
-  int ok = (validate_ntype(tree, decorated)
-            && (nch == 2)
-            && validate_decorators(RCHILD(tree, -2))
-            && (validate_funcdef(RCHILD(tree, -1))
-                || validate_class(RCHILD(tree, -1)))
-            );
-  return ok;
+    int nch = NCH(tree);
+    int ok = (validate_ntype(tree, decorated)
+              && (nch == 2)
+              && validate_decorators(RCHILD(tree, -2)));
+    if (TYPE(RCHILD(tree, -1)) == funcdef)
+        ok = ok && validate_funcdef(RCHILD(tree, -1));
+    else
+        ok = ok && validate_class(RCHILD(tree, -1));
+    return ok;
 }
 
 static int

From python-checkins at python.org  Sun Jul  4 18:37:39 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sun,  4 Jul 2010 18:37:39 +0200 (CEST)
Subject: [Python-checkins] r82538 - python/branches/py3k/Lib/lib2to3
Message-ID: <20100704163739.8514EEE9ED@mail.python.org>

Author: benjamin.peterson
Date: Sun Jul  4 18:37:39 2010
New Revision: 82538

Log:
Initialized merge tracking via "svnmerge" with revisions "0-80937" from 
svn+ssh://pythondev at svn.python.org/sandbox/trunk/2to3


Modified:
   python/branches/py3k/Lib/lib2to3/   (props changed)

From python-checkins at python.org  Sun Jul  4 18:39:03 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun,  4 Jul 2010 18:39:03 +0200 (CEST)
Subject: [Python-checkins] r82539 - in python/branches/release31-maint:
	Lib/test/test_parser.py Misc/NEWS Modules/parsermodule.c
Message-ID: <20100704163903.645CDEE9BC@mail.python.org>

Author: mark.dickinson
Date: Sun Jul  4 18:39:03 2010
New Revision: 82539

Log:
Merged revisions 82537 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82537 | mark.dickinson | 2010-07-04 17:37:31 +0100 (Sun, 04 Jul 2010) | 2 lines
  
  Issue #9128: Fix validation of class decorators in parser module.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/test/test_parser.py
   python/branches/release31-maint/Misc/NEWS
   python/branches/release31-maint/Modules/parsermodule.c

Modified: python/branches/release31-maint/Lib/test/test_parser.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_parser.py	(original)
+++ python/branches/release31-maint/Lib/test/test_parser.py	Sun Jul  4 18:39:03 2010
@@ -150,6 +150,13 @@
     def test_class_defs(self):
         self.check_suite("class foo():pass")
         self.check_suite("class foo(object):pass")
+        self.check_suite("@class_decorator\n"
+                         "class foo():pass")
+        self.check_suite("@class_decorator(arg)\n"
+                         "class foo():pass")
+        self.check_suite("@decorator1\n"
+                         "@decorator2\n"
+                         "class foo():pass")
 
     def test_import_from_statement(self):
         self.check_suite("from sys.path import *")

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Sun Jul  4 18:39:03 2010
@@ -75,6 +75,8 @@
 Library
 -------
 
+- Issue #9128: Fix validation of class decorators in parser module.
+
 - Issue #5468: urlencode to handle bytes type and other encodings in its query
   parameter. Patch by Dan Mahn.
 

Modified: python/branches/release31-maint/Modules/parsermodule.c
==============================================================================
--- python/branches/release31-maint/Modules/parsermodule.c	(original)
+++ python/branches/release31-maint/Modules/parsermodule.c	Sun Jul  4 18:39:03 2010
@@ -2507,14 +2507,15 @@
 static int
 validate_decorated(node *tree)
 {
-  int nch = NCH(tree);
-  int ok = (validate_ntype(tree, decorated)
-            && (nch == 2)
-            && validate_decorators(RCHILD(tree, -2))
-            && (validate_funcdef(RCHILD(tree, -1))
-                || validate_class(RCHILD(tree, -1)))
-            );
-  return ok;
+    int nch = NCH(tree);
+    int ok = (validate_ntype(tree, decorated)
+              && (nch == 2)
+              && validate_decorators(RCHILD(tree, -2)));
+    if (TYPE(RCHILD(tree, -1)) == funcdef)
+        ok = ok && validate_funcdef(RCHILD(tree, -1));
+    else
+        ok = ok && validate_class(RCHILD(tree, -1));
+    return ok;
 }
 
 static int

From python-checkins at python.org  Sun Jul  4 18:40:27 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sun,  4 Jul 2010 18:40:27 +0200 (CEST)
Subject: [Python-checkins] r82540 - python/branches/py3k/Lib/lib2to3
Message-ID: <20100704164027.BC154EE9D8@mail.python.org>

Author: benjamin.peterson
Date: Sun Jul  4 18:40:27 2010
New Revision: 82540

Log:
Removed merge tracking for "svnmerge" for 
svn+ssh://pythondev at svn.python.org/sandbox/trunk/2to3


Modified:
   python/branches/py3k/Lib/lib2to3/   (props changed)

From python-checkins at python.org  Sun Jul  4 18:41:33 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sun,  4 Jul 2010 18:41:33 +0200 (CEST)
Subject: [Python-checkins] r82541 - python/branches/py3k/Lib/lib2to3
Message-ID: <20100704164133.8D2F6EE9D1@mail.python.org>

Author: benjamin.peterson
Date: Sun Jul  4 18:41:33 2010
New Revision: 82541

Log:
Initialized merge tracking via "svnmerge" with revisions "0-80937" from 
svn+ssh://pythondev at svn.python.org/sandbox/trunk/2to3/lib2to3


Modified:
   python/branches/py3k/Lib/lib2to3/   (props changed)

From python-checkins at python.org  Sun Jul  4 18:44:15 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sun,  4 Jul 2010 18:44:15 +0200 (CEST)
Subject: [Python-checkins] r82542 - in python/branches/py3k/Lib/lib2to3:
	fixes/fix_itertools_imports.py tests/test_fixers.py
Message-ID: <20100704164415.A1696EE9D1@mail.python.org>

Author: benjamin.peterson
Date: Sun Jul  4 18:44:15 2010
New Revision: 82542

Log:
Merged revisions 81478,82530-82531 via svnmerge from 
svn+ssh://pythondev at svn.python.org/sandbox/trunk/2to3/lib2to3

........
  r81478 | benjamin.peterson | 2010-05-22 13:47:39 -0500 (Sat, 22 May 2010) | 1 line
  
  ensure doctests have some future_features
........
  r82530 | benjamin.peterson | 2010-07-04 11:11:41 -0500 (Sun, 04 Jul 2010) | 1 line
  
  simplify ignore star imports from itertools #8892
........
  r82531 | benjamin.peterson | 2010-07-04 11:13:20 -0500 (Sun, 04 Jul 2010) | 1 line
  
  wrap with parenthesis not \
........


Modified:
   python/branches/py3k/Lib/lib2to3/   (props changed)
   python/branches/py3k/Lib/lib2to3/fixes/fix_itertools_imports.py
   python/branches/py3k/Lib/lib2to3/tests/test_fixers.py

Modified: python/branches/py3k/Lib/lib2to3/fixes/fix_itertools_imports.py
==============================================================================
--- python/branches/py3k/Lib/lib2to3/fixes/fix_itertools_imports.py	(original)
+++ python/branches/py3k/Lib/lib2to3/fixes/fix_itertools_imports.py	Sun Jul  4 18:44:15 2010
@@ -20,6 +20,9 @@
             if child.type == token.NAME:
                 member = child.value
                 name_node = child
+            elif child.type == token.STAR:
+                # Just leave the import as is.
+                return
             else:
                 assert child.type == syms.import_as_name
                 name_node = child.children[0]
@@ -44,8 +47,8 @@
             children[-1].remove()
 
         # If there are no imports left, just get rid of the entire statement
-        if not (imports.children or getattr(imports, 'value', None)) or \
-                imports.parent is None:
+        if (not (imports.children or getattr(imports, 'value', None)) or
+            imports.parent is None):
             p = node.prefix
             node = BlankLine()
             node.prefix = p

Modified: python/branches/py3k/Lib/lib2to3/tests/test_fixers.py
==============================================================================
--- python/branches/py3k/Lib/lib2to3/tests/test_fixers.py	(original)
+++ python/branches/py3k/Lib/lib2to3/tests/test_fixers.py	Sun Jul  4 18:44:15 2010
@@ -3670,6 +3670,10 @@
         a = "from itertools import bar, filterfalse, foo"
         self.check(b, a)
 
+    def test_import_star(self):
+        s = "from itertools import *"
+        self.unchanged(s)
+
 
     def test_unchanged(self):
         s = "from itertools import foo"

From python-checkins at python.org  Sun Jul  4 18:47:57 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun,  4 Jul 2010 18:47:57 +0200 (CEST)
Subject: [Python-checkins] r82543 - in python/branches/release27-maint:
	Lib/test/test_parser.py Misc/NEWS Modules/parsermodule.c
Message-ID: <20100704164757.35808F68CB@mail.python.org>

Author: mark.dickinson
Date: Sun Jul  4 18:47:56 2010
New Revision: 82543

Log:
Issue #9130: Fix validation of relative imports in parser module.


Modified:
   python/branches/release27-maint/Lib/test/test_parser.py
   python/branches/release27-maint/Misc/NEWS
   python/branches/release27-maint/Modules/parsermodule.c

Modified: python/branches/release27-maint/Lib/test/test_parser.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_parser.py	(original)
+++ python/branches/release27-maint/Lib/test/test_parser.py	Sun Jul  4 18:47:56 2010
@@ -221,6 +221,12 @@
         self.check_suite("import sys as system, math")
         self.check_suite("import sys, math as my_math")
 
+    def test_relative_imports(self):
+        self.check_suite("from . import name")
+        self.check_suite("from .. import name")
+        self.check_suite("from .pkg import name")
+        self.check_suite("from ..pkg import name")
+
     def test_pep263(self):
         self.check_suite("# -*- coding: iso-8859-1 -*-\n"
                          "pass\n")
@@ -519,6 +525,20 @@
                 (0, ''))
         self.check_bad_tree(tree, "malformed global ast")
 
+    def test_missing_import_source(self):
+        # from import a
+        tree = \
+            (257,
+             (267,
+              (268,
+               (269,
+                (281,
+                 (283, (1, 'from'), (1, 'import'),
+                  (286, (284, (1, 'fred')))))),
+               (4, ''))),
+             (4, ''), (0, ''))
+        self.check_bad_tree(tree, "from import a")
+
 
 class CompileTestCase(unittest.TestCase):
 

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Sun Jul  4 18:47:56 2010
@@ -16,6 +16,8 @@
 Library
 -------
 
+- Issue #9130: Fix validation of relative imports in parser module.
+
 - Issue #9128: Fix validation of class decorators in parser module.
 
 Extension Modules

Modified: python/branches/release27-maint/Modules/parsermodule.c
==============================================================================
--- python/branches/release27-maint/Modules/parsermodule.c	(original)
+++ python/branches/release27-maint/Modules/parsermodule.c	Sun Jul  4 18:47:56 2010
@@ -1886,8 +1886,8 @@
         return i-1;
 }
 
-/* 'from' ('.'* dotted_name | '.') 'import' ('*' | '(' import_as_names ')' |
- *     import_as_names
+/* import_from: ('from' ('.'* dotted_name | '.'+)
+ *               'import' ('*' | '(' import_as_names ')' | import_as_names))
  */
 static int
 validate_import_from(node *tree)
@@ -1897,7 +1897,8 @@
         int havename = (TYPE(CHILD(tree, ndots + 1)) == dotted_name);
         int offset = ndots + havename;
         int res = validate_ntype(tree, import_from)
-                && (nch >= 4 + ndots)
+                && (offset >= 1)
+                && (nch >= 3 + offset)
                 && validate_name(CHILD(tree, 0), "from")
                 && (!havename || validate_dotted_name(CHILD(tree, ndots + 1)))
                 && validate_name(CHILD(tree, offset + 1), "import");

From python-checkins at python.org  Sun Jul  4 18:49:46 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sun,  4 Jul 2010 18:49:46 +0200 (CEST)
Subject: [Python-checkins] r82544 - in python/branches/release31-maint:
	Lib/lib2to3/fixes/fix_itertools_imports.py
	Lib/lib2to3/tests/test_fixers.py
Message-ID: <20100704164946.50025EE9F1@mail.python.org>

Author: benjamin.peterson
Date: Sun Jul  4 18:49:46 2010
New Revision: 82544

Log:
Merged revisions 82542 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

................
  r82542 | benjamin.peterson | 2010-07-04 11:44:15 -0500 (Sun, 04 Jul 2010) | 17 lines
  
  Merged revisions 81478,82530-82531 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/sandbox/trunk/2to3/lib2to3
  
  ........
    r81478 | benjamin.peterson | 2010-05-22 13:47:39 -0500 (Sat, 22 May 2010) | 1 line
    
    ensure doctests have some future_features
  ........
    r82530 | benjamin.peterson | 2010-07-04 11:11:41 -0500 (Sun, 04 Jul 2010) | 1 line
    
    simplify ignore star imports from itertools #8892
  ........
    r82531 | benjamin.peterson | 2010-07-04 11:13:20 -0500 (Sun, 04 Jul 2010) | 1 line
    
    wrap with parenthesis not \
  ........
................


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/lib2to3/fixes/fix_itertools_imports.py
   python/branches/release31-maint/Lib/lib2to3/tests/test_fixers.py

Modified: python/branches/release31-maint/Lib/lib2to3/fixes/fix_itertools_imports.py
==============================================================================
--- python/branches/release31-maint/Lib/lib2to3/fixes/fix_itertools_imports.py	(original)
+++ python/branches/release31-maint/Lib/lib2to3/fixes/fix_itertools_imports.py	Sun Jul  4 18:49:46 2010
@@ -20,6 +20,9 @@
             if child.type == token.NAME:
                 member = child.value
                 name_node = child
+            elif child.type == token.STAR:
+                # Just leave the import as is.
+                return
             else:
                 assert child.type == syms.import_as_name
                 name_node = child.children[0]
@@ -44,8 +47,8 @@
             children[-1].remove()
 
         # If there are no imports left, just get rid of the entire statement
-        if not (imports.children or getattr(imports, 'value', None)) or \
-                imports.parent is None:
+        if (not (imports.children or getattr(imports, 'value', None)) or
+            imports.parent is None):
             p = node.prefix
             node = BlankLine()
             node.prefix = p

Modified: python/branches/release31-maint/Lib/lib2to3/tests/test_fixers.py
==============================================================================
--- python/branches/release31-maint/Lib/lib2to3/tests/test_fixers.py	(original)
+++ python/branches/release31-maint/Lib/lib2to3/tests/test_fixers.py	Sun Jul  4 18:49:46 2010
@@ -3670,6 +3670,10 @@
         a = "from itertools import bar, filterfalse, foo"
         self.check(b, a)
 
+    def test_import_star(self):
+        s = "from itertools import *"
+        self.unchanged(s)
+
 
     def test_unchanged(self):
         s = "from itertools import foo"

From python-checkins at python.org  Sun Jul  4 18:53:16 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sun,  4 Jul 2010 18:53:16 +0200 (CEST)
Subject: [Python-checkins] r82545 - in python/branches/release27-maint:
	Lib/lib2to3/fixes/fix_itertools_imports.py
	Lib/lib2to3/tests/test_fixers.py
Message-ID: <20100704165316.86906EE9E0@mail.python.org>

Author: benjamin.peterson
Date: Sun Jul  4 18:53:16 2010
New Revision: 82545

Log:
Merged revisions 82542 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

................
  r82542 | benjamin.peterson | 2010-07-04 11:44:15 -0500 (Sun, 04 Jul 2010) | 17 lines
  
  Merged revisions 81478,82530-82531 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/sandbox/trunk/2to3/lib2to3
  
  ........
    r81478 | benjamin.peterson | 2010-05-22 13:47:39 -0500 (Sat, 22 May 2010) | 1 line
    
    ensure doctests have some future_features
  ........
    r82530 | benjamin.peterson | 2010-07-04 11:11:41 -0500 (Sun, 04 Jul 2010) | 1 line
    
    simplify ignore star imports from itertools #8892
  ........
    r82531 | benjamin.peterson | 2010-07-04 11:13:20 -0500 (Sun, 04 Jul 2010) | 1 line
    
    wrap with parenthesis not \
  ........
................


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/lib2to3/fixes/fix_itertools_imports.py
   python/branches/release27-maint/Lib/lib2to3/tests/test_fixers.py

Modified: python/branches/release27-maint/Lib/lib2to3/fixes/fix_itertools_imports.py
==============================================================================
--- python/branches/release27-maint/Lib/lib2to3/fixes/fix_itertools_imports.py	(original)
+++ python/branches/release27-maint/Lib/lib2to3/fixes/fix_itertools_imports.py	Sun Jul  4 18:53:16 2010
@@ -20,6 +20,9 @@
             if child.type == token.NAME:
                 member = child.value
                 name_node = child
+            elif child.type == token.STAR:
+                # Just leave the import as is.
+                return
             else:
                 assert child.type == syms.import_as_name
                 name_node = child.children[0]
@@ -44,8 +47,8 @@
             children[-1].remove()
 
         # If there are no imports left, just get rid of the entire statement
-        if not (imports.children or getattr(imports, 'value', None)) or \
-                imports.parent is None:
+        if (not (imports.children or getattr(imports, 'value', None)) or
+            imports.parent is None):
             p = node.prefix
             node = BlankLine()
             node.prefix = p

Modified: python/branches/release27-maint/Lib/lib2to3/tests/test_fixers.py
==============================================================================
--- python/branches/release27-maint/Lib/lib2to3/tests/test_fixers.py	(original)
+++ python/branches/release27-maint/Lib/lib2to3/tests/test_fixers.py	Sun Jul  4 18:53:16 2010
@@ -3670,6 +3670,10 @@
         a = "from itertools import bar, filterfalse, foo"
         self.check(b, a)
 
+    def test_import_star(self):
+        s = "from itertools import *"
+        self.unchanged(s)
+
 
     def test_unchanged(self):
         s = "from itertools import foo"

From python-checkins at python.org  Sun Jul  4 18:59:04 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun,  4 Jul 2010 18:59:04 +0200 (CEST)
Subject: [Python-checkins] r82546 - in python/branches/release26-maint:
	Lib/test/test_parser.py Misc/NEWS Modules/parsermodule.c
Message-ID: <20100704165904.9473EEE9F9@mail.python.org>

Author: mark.dickinson
Date: Sun Jul  4 18:59:04 2010
New Revision: 82546

Log:
Issue #9130: Fix validation of relative imports in parser module.


Modified:
   python/branches/release26-maint/Lib/test/test_parser.py
   python/branches/release26-maint/Misc/NEWS
   python/branches/release26-maint/Modules/parsermodule.c

Modified: python/branches/release26-maint/Lib/test/test_parser.py
==============================================================================
--- python/branches/release26-maint/Lib/test/test_parser.py	(original)
+++ python/branches/release26-maint/Lib/test/test_parser.py	Sun Jul  4 18:59:04 2010
@@ -197,6 +197,12 @@
         self.check_suite("import sys as system, math")
         self.check_suite("import sys, math as my_math")
 
+    def test_relative_imports(self):
+        self.check_suite("from . import name")
+        self.check_suite("from .. import name")
+        self.check_suite("from .pkg import name")
+        self.check_suite("from ..pkg import name")
+
     def test_pep263(self):
         self.check_suite("# -*- coding: iso-8859-1 -*-\n"
                          "pass\n")
@@ -494,6 +500,20 @@
                 (0, ''))
         self.check_bad_tree(tree, "malformed global ast")
 
+    def test_missing_import_source(self):
+        # from import a
+        tree = \
+            (257,
+             (267,
+              (268,
+               (269,
+                (281,
+                 (283, (1, 'from'), (1, 'import'),
+                  (286, (284, (1, 'fred')))))),
+               (4, ''))),
+             (4, ''), (0, ''))
+        self.check_bad_tree(tree, "from import a")
+
 
 class CompileTestCase(unittest.TestCase):
 

Modified: python/branches/release26-maint/Misc/NEWS
==============================================================================
--- python/branches/release26-maint/Misc/NEWS	(original)
+++ python/branches/release26-maint/Misc/NEWS	Sun Jul  4 18:59:04 2010
@@ -81,6 +81,8 @@
 Library
 -------
 
+- Issue #9130: Fix validation of relative imports in parser module.
+
 - Issue #9128: Fix validation of class decorators in parser module.
 
 - Issue #7673: Fix security vulnerability (CVE-2010-2089) in the audioop

Modified: python/branches/release26-maint/Modules/parsermodule.c
==============================================================================
--- python/branches/release26-maint/Modules/parsermodule.c	(original)
+++ python/branches/release26-maint/Modules/parsermodule.c	Sun Jul  4 18:59:04 2010
@@ -1886,8 +1886,8 @@
         return i-1;
 }
 
-/* 'from' ('.'* dotted_name | '.') 'import' ('*' | '(' import_as_names ')' |
- *     import_as_names
+/* import_from: ('from' ('.'* dotted_name | '.'+)
+ *               'import' ('*' | '(' import_as_names ')' | import_as_names))
  */
 static int
 validate_import_from(node *tree)
@@ -1897,7 +1897,8 @@
         int havename = (TYPE(CHILD(tree, ndots + 1)) == dotted_name);
         int offset = ndots + havename;
         int res = validate_ntype(tree, import_from)
-                && (nch >= 4 + ndots)
+                && (offset >= 1)
+                && (nch >= 3 + offset)
                 && validate_name(CHILD(tree, 0), "from")
                 && (!havename || validate_dotted_name(CHILD(tree, ndots + 1)))
                 && validate_name(CHILD(tree, offset + 1), "import");

From python-checkins at python.org  Sun Jul  4 19:00:20 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Sun,  4 Jul 2010 19:00:20 +0200 (CEST)
Subject: [Python-checkins] r82547 - python/branches/py3k/Lib/pydoc.py
Message-ID: <20100704170020.8CCDFEE9FA@mail.python.org>

Author: alexander.belopolsky
Date: Sun Jul  4 19:00:20 2010
New Revision: 82547

Log:
Issue #9118: help(None) will now return NoneType doc instead of
starting interactive help.


Modified:
   python/branches/py3k/Lib/pydoc.py

Modified: python/branches/py3k/Lib/pydoc.py
==============================================================================
--- python/branches/py3k/Lib/pydoc.py	(original)
+++ python/branches/py3k/Lib/pydoc.py	Sun Jul  4 19:00:20 2010
@@ -1705,8 +1705,9 @@
             return ''
         return '<pydoc.Helper instance>'
 
-    def __call__(self, request=None):
-        if request is not None:
+    _GoInteractive = object()
+    def __call__(self, request=_GoInteractive):
+        if request is not self._GoInteractive:
             self.help(request)
         else:
             self.intro()

From python-checkins at python.org  Sun Jul  4 19:28:33 2010
From: python-checkins at python.org (georg.brandl)
Date: Sun,  4 Jul 2010 19:28:33 +0200 (CEST)
Subject: [Python-checkins] r82548 -
	python/branches/release27-maint/Doc/library/functions.rst
Message-ID: <20100704172833.47D1BEE9C7@mail.python.org>

Author: georg.brandl
Date: Sun Jul  4 19:28:33 2010
New Revision: 82548

Log:
#8472: fix misleading reference to ifilterfalse() in filter() docs.

Modified:
   python/branches/release27-maint/Doc/library/functions.rst

Modified: python/branches/release27-maint/Doc/library/functions.rst
==============================================================================
--- python/branches/release27-maint/Doc/library/functions.rst	(original)
+++ python/branches/release27-maint/Doc/library/functions.rst	Sun Jul  4 19:28:33 2010
@@ -402,8 +402,9 @@
    iterable if function(item)]`` if function is not ``None`` and ``[item for item
    in iterable if item]`` if function is ``None``.
 
-   See :func:`itertools.ifilterfalse` for the complementary function that returns
-   elements of *iterable* for which *function* returns false.
+   See :func:`itertools.ifilter` and :func:`itertools.ifilterfalse` for iterator
+   versions of this function, including a variation that filters for elements
+   where the *function* returns false.
 
 
 .. function:: float([x])

From python-checkins at python.org  Sun Jul  4 19:33:33 2010
From: python-checkins at python.org (georg.brandl)
Date: Sun,  4 Jul 2010 19:33:33 +0200 (CEST)
Subject: [Python-checkins] r82549 - in python/branches/release31-maint:
	Doc/glossary.rst Doc/library/binascii.rst
	Doc/library/reprlib.rst Doc/library/stdtypes.rst
Message-ID: <20100704173333.EF4C8EE982@mail.python.org>

Author: georg.brandl
Date: Sun Jul  4 19:33:33 2010
New Revision: 82549

Log:
Merged revisions 77316 via svnmerge from 
svn+ssh://svn.python.org/python/branches/py3k

........
  r77316 | georg.brandl | 2010-01-05 11:22:04 +0100 (Di, 05 Jan 2010) | 1 line
  
  Assorted doc fixes by Florent.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Doc/glossary.rst
   python/branches/release31-maint/Doc/library/binascii.rst
   python/branches/release31-maint/Doc/library/reprlib.rst
   python/branches/release31-maint/Doc/library/stdtypes.rst

Modified: python/branches/release31-maint/Doc/glossary.rst
==============================================================================
--- python/branches/release31-maint/Doc/glossary.rst	(original)
+++ python/branches/release31-maint/Doc/glossary.rst	Sun Jul  4 19:33:33 2010
@@ -355,7 +355,7 @@
 
    list comprehension
       A compact way to process all or part of the elements in a sequence and
-      return a list with the results.  ``result = ["0x%02x" % x for x in
+      return a list with the results.  ``result = ['{:#04x}'.format(x) for x in
       range(256) if x % 2 == 0]`` generates a list of strings containing
       even hex numbers (0x..) in the range from 0 to 255. The :keyword:`if`
       clause is optional.  If omitted, all elements in ``range(256)`` are
@@ -473,7 +473,7 @@
       object drops to zero, it is deallocated.  Reference counting is
       generally not visible to Python code, but it is a key element of the
       :term:`CPython` implementation.  The :mod:`sys` module defines a
-      :func:`getrefcount` function that programmers can call to return the
+      :func:`~sys.getrefcount` function that programmers can call to return the
       reference count for a particular object.
 
    __slots__

Modified: python/branches/release31-maint/Doc/library/binascii.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/binascii.rst	(original)
+++ python/branches/release31-maint/Doc/library/binascii.rst	Sun Jul  4 19:33:33 2010
@@ -109,11 +109,11 @@
    use as a checksum algorithm, it is not suitable for use as a general hash
    algorithm.  Use as follows::
 
-      print(binascii.crc32("hello world"))
+      print(binascii.crc32(b"hello world"))
       # Or, in two pieces:
-      crc = binascii.crc32("hello")
-      crc = binascii.crc32(" world", crc) & 0xffffffff
-      print('crc32 = 0x%08x' % crc)
+      crc = binascii.crc32(b"hello")
+      crc = binascii.crc32(b" world", crc) & 0xffffffff
+      print('crc32 = {:#010x}'.format(crc))
 
 .. note::
    To generate the same numeric value across all Python versions and

Modified: python/branches/release31-maint/Doc/library/reprlib.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/reprlib.rst	(original)
+++ python/branches/release31-maint/Doc/library/reprlib.rst	Sun Jul  4 19:33:33 2010
@@ -130,5 +130,5 @@
                return repr(obj)
 
    aRepr = MyRepr()
-   print aRepr.repr(sys.stdin)          # prints '<stdin>'
+   print(aRepr.repr(sys.stdin))         # prints '<stdin>'
 

Modified: python/branches/release31-maint/Doc/library/stdtypes.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/stdtypes.rst	(original)
+++ python/branches/release31-maint/Doc/library/stdtypes.rst	Sun Jul  4 19:33:33 2010
@@ -1903,7 +1903,7 @@
    .. describe:: iter(d)
 
       Return an iterator over the keys of the dictionary.  This is a shortcut
-      for :meth:`iterkeys`.
+      for ``iter(d.keys())``.
 
    .. method:: clear()
 

From python-checkins at python.org  Sun Jul  4 19:38:32 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Sun,  4 Jul 2010 19:38:32 +0200 (CEST)
Subject: [Python-checkins] r82550 - python/branches/py3k/Demo/classes/Rev.py
Message-ID: <20100704173832.E5788EE982@mail.python.org>

Author: alexander.belopolsky
Date: Sun Jul  4 19:38:32 2010
New Revision: 82550

Log:
Fixed doctests

Modified:
   python/branches/py3k/Demo/classes/Rev.py

Modified: python/branches/py3k/Demo/classes/Rev.py
==============================================================================
--- python/branches/py3k/Demo/classes/Rev.py	(original)
+++ python/branches/py3k/Demo/classes/Rev.py	Sun Jul  4 19:38:32 2010
@@ -4,8 +4,7 @@
 
 It works on mutable or inmutable sequences.
 
->>> chars = list(Rev('Hello World!'))
->>> print ''.join(chars)
+>>> Rev('Hello World!')
 !dlroW olleH
 
 The .forw is so you can use anonymous sequences in __init__, and still
@@ -16,16 +15,16 @@
 confusion. Maybe it should be change to copy input sequence to break
 the connection completely ? )
 
->>> nnn = range(3)
+>>> nnn = list(range(3))
 >>> rnn = Rev(nnn)
->>> for n in rnn: print n
+>>> for n in rnn: n
 ...
 2
 1
 0
 >>> for n in range(4, 6): nnn.append(n)   # update nnn
 ...
->>> for n in rnn: print n     # prints reversed updated values
+>>> for n in rnn: n     # prints reversed updated values
 ...
 5
 4
@@ -35,7 +34,7 @@
 >>> nnn = nnn[1:-1]
 >>> nnn
 [1, 2, 4]
->>> for n in rnn: print n     # prints reversed values of old nnn
+>>> for n in rnn: n     # prints reversed values of old nnn
 ...
 5
 4
@@ -45,16 +44,15 @@
 
 #
 >>> WH = Rev('Hello World!')
->>> print WH.forw, WH.back
+>>> print(WH.forw, WH.back)
 Hello World! !dlroW olleH
->>> nnn = Rev(range(1, 10))
->>> print nnn.forw
+>>> nnn = Rev(list(range(1, 10)))
+>>> print(nnn.forw)
 [1, 2, 3, 4, 5, 6, 7, 8, 9]
->>> print nnn.back
+>>> print(nnn.back)
 [9, 8, 7, 6, 5, 4, 3, 2, 1]
 
->>> rrr = Rev(nnn)
->>> rrr
+>>> Rev(nnn)
 <1, 2, 3, 4, 5, 6, 7, 8, 9>
 
 '''

From python-checkins at python.org  Sun Jul  4 19:47:30 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Sun,  4 Jul 2010 19:47:30 +0200 (CEST)
Subject: [Python-checkins] r82551 -
	python/branches/py3k/Demo/classes/Range.py
Message-ID: <20100704174730.D74E9EE982@mail.python.org>

Author: alexander.belopolsky
Date: Sun Jul  4 19:47:30 2010
New Revision: 82551

Log:
Fixed the test

Modified:
   python/branches/py3k/Demo/classes/Range.py

Modified: python/branches/py3k/Demo/classes/Range.py
==============================================================================
--- python/branches/py3k/Demo/classes/Range.py	(original)
+++ python/branches/py3k/Demo/classes/Range.py	Sun Jul  4 19:47:30 2010
@@ -66,7 +66,7 @@
 def test():
     import time, builtins
     #Just a quick sanity check
-    correct_result = builtins.range(5, 100, 3)
+    correct_result = list(builtins.range(5, 100, 3))
     oldrange_result = list(oldrange(5, 100, 3))
     genrange_result = list(genrange(5, 100, 3))
     if genrange_result != correct_result or oldrange_result != correct_result:

From python-checkins at python.org  Sun Jul  4 20:11:51 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun,  4 Jul 2010 20:11:51 +0200 (CEST)
Subject: [Python-checkins] r82552 - in python/branches/py3k:
	Lib/test/test_parser.py Misc/NEWS Modules/parsermodule.c
Message-ID: <20100704181151.6D751EE9A2@mail.python.org>

Author: mark.dickinson
Date: Sun Jul  4 20:11:51 2010
New Revision: 82552

Log:
Issue #9130: Fix validation of relative imports in parser module.


Modified:
   python/branches/py3k/Lib/test/test_parser.py
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/Modules/parsermodule.c

Modified: python/branches/py3k/Lib/test/test_parser.py
==============================================================================
--- python/branches/py3k/Lib/test/test_parser.py	(original)
+++ python/branches/py3k/Lib/test/test_parser.py	Sun Jul  4 20:11:51 2010
@@ -189,6 +189,12 @@
         self.check_suite("import sys as system, math")
         self.check_suite("import sys, math as my_math")
 
+    def test_relative_imports(self):
+        self.check_suite("from . import name")
+        self.check_suite("from .. import name")
+        self.check_suite("from .pkg import name")
+        self.check_suite("from ..pkg import name")
+
     def test_pep263(self):
         self.check_suite("# -*- coding: iso-8859-1 -*-\n"
                          "pass\n")
@@ -464,6 +470,20 @@
                 (0, ''))
         self.check_bad_tree(tree, "malformed global ast")
 
+    def test_missing_import_source(self):
+        # from import a
+        tree = \
+            (257,
+             (267,
+              (268,
+               (269,
+                (281,
+                 (283, (1, 'from'), (1, 'import'),
+                  (286, (284, (1, 'fred')))))),
+               (4, ''))),
+             (4, ''), (0, ''))
+        self.check_bad_tree(tree, "from import a")
+
 
 class CompileTestCase(unittest.TestCase):
 

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sun Jul  4 20:11:51 2010
@@ -468,6 +468,8 @@
 Library
 -------
 
+- Issue #9130: Fix validation of relative imports in parser module.
+
 - Issue #9128: Fix validation of class decorators in parser module.
 
 - Issue #9094: python -m pickletools will now disassemble pickle files

Modified: python/branches/py3k/Modules/parsermodule.c
==============================================================================
--- python/branches/py3k/Modules/parsermodule.c	(original)
+++ python/branches/py3k/Modules/parsermodule.c	Sun Jul  4 20:11:51 2010
@@ -1767,8 +1767,8 @@
         return i-1;
 }
 
-/* 'from' ('.'* dotted_name | '.') 'import' ('*' | '(' import_as_names ')' |
- *     import_as_names
+/* import_from: ('from' ('.'* dotted_name | '.'+)
+ *               'import' ('*' | '(' import_as_names ')' | import_as_names))
  */
 static int
 validate_import_from(node *tree)
@@ -1778,7 +1778,8 @@
         int havename = (TYPE(CHILD(tree, ndots + 1)) == dotted_name);
         int offset = ndots + havename;
         int res = validate_ntype(tree, import_from)
-                && (nch >= 4 + ndots)
+                && (offset >= 1)
+                && (nch >= 3 + offset)
                 && validate_name(CHILD(tree, 0), "from")
                 && (!havename || validate_dotted_name(CHILD(tree, ndots + 1)))
                 && validate_name(CHILD(tree, offset + 1), "import");

From python-checkins at python.org  Sun Jul  4 20:15:26 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun,  4 Jul 2010 20:15:26 +0200 (CEST)
Subject: [Python-checkins] r82553 -
	python/branches/py3k/Lib/test/test_parser.py
Message-ID: <20100704181526.697FBEE9A9@mail.python.org>

Author: mark.dickinson
Date: Sun Jul  4 20:15:26 2010
New Revision: 82553

Log:
Fix symbol numbers in test_parser test.

Modified:
   python/branches/py3k/Lib/test/test_parser.py

Modified: python/branches/py3k/Lib/test/test_parser.py
==============================================================================
--- python/branches/py3k/Lib/test/test_parser.py	(original)
+++ python/branches/py3k/Lib/test/test_parser.py	Sun Jul  4 20:15:26 2010
@@ -471,18 +471,18 @@
         self.check_bad_tree(tree, "malformed global ast")
 
     def test_missing_import_source(self):
-        # from import a
+        # from import fred
         tree = \
             (257,
-             (267,
-              (268,
-               (269,
-                (281,
-                 (283, (1, 'from'), (1, 'import'),
-                  (286, (284, (1, 'fred')))))),
+             (268,
+              (269,
+               (270,
+                (282,
+                 (284, (1, 'from'), (1, 'import'),
+                  (287, (285, (1, 'fred')))))),
                (4, ''))),
              (4, ''), (0, ''))
-        self.check_bad_tree(tree, "from import a")
+        self.check_bad_tree(tree, "from import fred")
 
 
 class CompileTestCase(unittest.TestCase):

From python-checkins at python.org  Sun Jul  4 20:16:43 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun,  4 Jul 2010 20:16:43 +0200 (CEST)
Subject: [Python-checkins] r82554 - in python/branches/release31-maint:
	Lib/test/test_parser.py Misc/NEWS Modules/parsermodule.c
Message-ID: <20100704181643.A48E1EE9A9@mail.python.org>

Author: mark.dickinson
Date: Sun Jul  4 20:16:43 2010
New Revision: 82554

Log:
Merged revisions 82552-82553 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82552 | mark.dickinson | 2010-07-04 19:11:51 +0100 (Sun, 04 Jul 2010) | 2 lines
  
  Issue #9130: Fix validation of relative imports in parser module.
........
  r82553 | mark.dickinson | 2010-07-04 19:15:26 +0100 (Sun, 04 Jul 2010) | 1 line
  
  Fix symbol numbers in test_parser test.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/test/test_parser.py
   python/branches/release31-maint/Misc/NEWS
   python/branches/release31-maint/Modules/parsermodule.c

Modified: python/branches/release31-maint/Lib/test/test_parser.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_parser.py	(original)
+++ python/branches/release31-maint/Lib/test/test_parser.py	Sun Jul  4 20:16:43 2010
@@ -190,6 +190,12 @@
         self.check_suite("import sys as system, math")
         self.check_suite("import sys, math as my_math")
 
+    def test_relative_imports(self):
+        self.check_suite("from . import name")
+        self.check_suite("from .. import name")
+        self.check_suite("from .pkg import name")
+        self.check_suite("from ..pkg import name")
+
     def test_pep263(self):
         self.check_suite("# -*- coding: iso-8859-1 -*-\n"
                          "pass\n")
@@ -459,6 +465,20 @@
                 (0, ''))
         self.check_bad_tree(tree, "malformed global ast")
 
+    def test_missing_import_source(self):
+        # from import fred
+        tree = \
+            (257,
+             (268,
+              (269,
+               (270,
+                (282,
+                 (284, (1, 'from'), (1, 'import'),
+                  (287, (285, (1, 'fred')))))),
+               (4, ''))),
+             (4, ''), (0, ''))
+        self.check_bad_tree(tree, "from import fred")
+
 
 class CompileTestCase(unittest.TestCase):
 

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Sun Jul  4 20:16:43 2010
@@ -75,6 +75,8 @@
 Library
 -------
 
+- Issue #9130: Fix validation of relative imports in parser module.
+
 - Issue #9128: Fix validation of class decorators in parser module.
 
 - Issue #5468: urlencode to handle bytes type and other encodings in its query

Modified: python/branches/release31-maint/Modules/parsermodule.c
==============================================================================
--- python/branches/release31-maint/Modules/parsermodule.c	(original)
+++ python/branches/release31-maint/Modules/parsermodule.c	Sun Jul  4 20:16:43 2010
@@ -1763,8 +1763,8 @@
         return i-1;
 }
 
-/* 'from' ('.'* dotted_name | '.') 'import' ('*' | '(' import_as_names ')' |
- *     import_as_names
+/* import_from: ('from' ('.'* dotted_name | '.'+)
+ *               'import' ('*' | '(' import_as_names ')' | import_as_names))
  */
 static int
 validate_import_from(node *tree)
@@ -1774,7 +1774,8 @@
         int havename = (TYPE(CHILD(tree, ndots + 1)) == dotted_name);
         int offset = ndots + havename;
         int res = validate_ntype(tree, import_from)
-                && (nch >= 4 + ndots)
+                && (offset >= 1)
+                && (nch >= 3 + offset)
                 && validate_name(CHILD(tree, 0), "from")
                 && (!havename || validate_dotted_name(CHILD(tree, ndots + 1)))
                 && validate_name(CHILD(tree, offset + 1), "import");

From python-checkins at python.org  Sun Jul  4 20:38:57 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun,  4 Jul 2010 20:38:57 +0200 (CEST)
Subject: [Python-checkins] r82555 - in python/branches/py3k:
	Lib/test/test_parser.py Modules/parsermodule.c
Message-ID: <20100704183857.F33DDEE988@mail.python.org>

Author: mark.dickinson
Date: Sun Jul  4 20:38:57 2010
New Revision: 82555

Log:
Issue #9130: Validate ellipsis tokens in relative imports.


Modified:
   python/branches/py3k/Lib/test/test_parser.py
   python/branches/py3k/Modules/parsermodule.c

Modified: python/branches/py3k/Lib/test/test_parser.py
==============================================================================
--- python/branches/py3k/Lib/test/test_parser.py	(original)
+++ python/branches/py3k/Lib/test/test_parser.py	Sun Jul  4 20:38:57 2010
@@ -192,8 +192,14 @@
     def test_relative_imports(self):
         self.check_suite("from . import name")
         self.check_suite("from .. import name")
+        # check all the way up to '....', since '...' is tokenized
+        # differently from '.' (it's an ellipsis token).
+        self.check_suite("from ... import name")
+        self.check_suite("from .... import name")
         self.check_suite("from .pkg import name")
         self.check_suite("from ..pkg import name")
+        self.check_suite("from ...pkg import name")
+        self.check_suite("from ....pkg import name")
 
     def test_pep263(self):
         self.check_suite("# -*- coding: iso-8859-1 -*-\n"

Modified: python/branches/py3k/Modules/parsermodule.c
==============================================================================
--- python/branches/py3k/Modules/parsermodule.c	(original)
+++ python/branches/py3k/Modules/parsermodule.c	Sun Jul  4 20:38:57 2010
@@ -1754,17 +1754,17 @@
                 && validate_dotted_as_names(CHILD(tree, 1)));
 }
 
-/* Helper function to count the number of leading dots in
+/* Helper function to count the number of leading dots (or ellipsis tokens) in
  * 'from ...module import name'
  */
 static int
 count_from_dots(node *tree)
 {
-        int i;
-        for (i = 1; i < NCH(tree); i++)
-                if (TYPE(CHILD(tree, i)) != DOT)
-                        break;
-        return i-1;
+    int i;
+    for (i = 1; i < NCH(tree); i++)
+        if (TYPE(CHILD(tree, i)) != DOT && TYPE(CHILD(tree, i)) != ELLIPSIS)
+            break;
+    return i - 1;
 }
 
 /* import_from: ('from' ('.'* dotted_name | '.'+)

From python-checkins at python.org  Sun Jul  4 20:39:48 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun,  4 Jul 2010 20:39:48 +0200 (CEST)
Subject: [Python-checkins] r82556 - in python/branches/release31-maint:
	Lib/test/test_parser.py Modules/parsermodule.c
Message-ID: <20100704183948.D7802EE988@mail.python.org>

Author: mark.dickinson
Date: Sun Jul  4 20:39:48 2010
New Revision: 82556

Log:
Merged revisions 82555 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82555 | mark.dickinson | 2010-07-04 19:38:57 +0100 (Sun, 04 Jul 2010) | 2 lines
  
  Issue #9130: Validate ellipsis tokens in relative imports.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/test/test_parser.py
   python/branches/release31-maint/Modules/parsermodule.c

Modified: python/branches/release31-maint/Lib/test/test_parser.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_parser.py	(original)
+++ python/branches/release31-maint/Lib/test/test_parser.py	Sun Jul  4 20:39:48 2010
@@ -193,8 +193,14 @@
     def test_relative_imports(self):
         self.check_suite("from . import name")
         self.check_suite("from .. import name")
+        # check all the way up to '....', since '...' is tokenized
+        # differently from '.' (it's an ellipsis token).
+        self.check_suite("from ... import name")
+        self.check_suite("from .... import name")
         self.check_suite("from .pkg import name")
         self.check_suite("from ..pkg import name")
+        self.check_suite("from ...pkg import name")
+        self.check_suite("from ....pkg import name")
 
     def test_pep263(self):
         self.check_suite("# -*- coding: iso-8859-1 -*-\n"

Modified: python/branches/release31-maint/Modules/parsermodule.c
==============================================================================
--- python/branches/release31-maint/Modules/parsermodule.c	(original)
+++ python/branches/release31-maint/Modules/parsermodule.c	Sun Jul  4 20:39:48 2010
@@ -1750,17 +1750,17 @@
                 && validate_dotted_as_names(CHILD(tree, 1)));
 }
 
-/* Helper function to count the number of leading dots in
+/* Helper function to count the number of leading dots (or ellipsis tokens) in
  * 'from ...module import name'
  */
 static int
 count_from_dots(node *tree)
 {
-        int i;
-        for (i = 1; i < NCH(tree); i++)
-                if (TYPE(CHILD(tree, i)) != DOT)
-                        break;
-        return i-1;
+    int i;
+    for (i = 1; i < NCH(tree); i++)
+        if (TYPE(CHILD(tree, i)) != DOT && TYPE(CHILD(tree, i)) != ELLIPSIS)
+            break;
+    return i - 1;
 }
 
 /* import_from: ('from' ('.'* dotted_name | '.'+)

From python-checkins at python.org  Sun Jul  4 20:49:18 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun,  4 Jul 2010 20:49:18 +0200 (CEST)
Subject: [Python-checkins] r82557 -
	python/branches/py3k/Demo/parser/test_parser.py
Message-ID: <20100704184918.E59ECEFDF@mail.python.org>

Author: mark.dickinson
Date: Sun Jul  4 20:49:18 2010
New Revision: 82557

Log:
Make Demo/parser/test_parser.py run.

Modified:
   python/branches/py3k/Demo/parser/test_parser.py

Modified: python/branches/py3k/Demo/parser/test_parser.py
==============================================================================
--- python/branches/py3k/Demo/parser/test_parser.py	(original)
+++ python/branches/py3k/Demo/parser/test_parser.py	Sun Jul  4 20:49:18 2010
@@ -11,19 +11,19 @@
     global _numFailed
     print('----', fileName, end=' ')
     try:
-        ast = parser.suite(t)
-        tup = parser.ast2tuple(ast)
-        # this discards the first AST; a huge memory savings when running
+        st = parser.suite(t)
+        tup = parser.st2tuple(st)
+        # this discards the first ST; a huge memory savings when running
         # against a large source file like Tkinter.py.
-        ast = None
-        new = parser.tuple2ast(tup)
+        st = None
+        new = parser.tuple2st(tup)
     except parser.ParserError as err:
         print()
         print('parser module raised exception on input file', fileName + ':')
         traceback.print_exc()
         _numFailed = _numFailed + 1
     else:
-        if tup != parser.ast2tuple(new):
+        if tup != parser.st2tuple(new):
             print()
             print('parser module failed on input file', fileName)
             _numFailed = _numFailed + 1

From python-checkins at python.org  Sun Jul  4 21:23:49 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun,  4 Jul 2010 21:23:49 +0200 (CEST)
Subject: [Python-checkins] r82558 - python/branches/py3k/Lib/heapq.py
Message-ID: <20100704192349.35283EE9A9@mail.python.org>

Author: mark.dickinson
Date: Sun Jul  4 21:23:49 2010
New Revision: 82558

Log:
Remove coding cookie from heapq.py.

Modified:
   python/branches/py3k/Lib/heapq.py

Modified: python/branches/py3k/Lib/heapq.py
==============================================================================
--- python/branches/py3k/Lib/heapq.py	(original)
+++ python/branches/py3k/Lib/heapq.py	Sun Jul  4 21:23:49 2010
@@ -1,5 +1,3 @@
-# -*- coding: latin-1 -*-
-
 """Heap queue algorithm (a.k.a. priority queue).
 
 Heaps are arrays for which a[k] <= a[2*k+1] and a[k] <= a[2*k+2] for
@@ -34,7 +32,7 @@
 
 __about__ = """Heap queues
 
-[explanation by Fran?ois Pinard]
+[explanation by Fran??ois Pinard]
 
 Heaps are arrays for which a[k] <= a[2*k+1] and a[k] <= a[2*k+2] for
 all k, counting elements from 0.  For the sake of comparison,

From python-checkins at python.org  Sun Jul  4 21:27:12 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun,  4 Jul 2010 21:27:12 +0200 (CEST)
Subject: [Python-checkins] r82559 - python/branches/py3k/Lib/shlex.py
Message-ID: <20100704192712.D1B3AEE9A9@mail.python.org>

Author: mark.dickinson
Date: Sun Jul  4 21:27:12 2010
New Revision: 82559

Log:
Re-encode shlex.py in UTF-8, and remove coding cookie.

Modified:
   python/branches/py3k/Lib/shlex.py

Modified: python/branches/py3k/Lib/shlex.py
==============================================================================
--- python/branches/py3k/Lib/shlex.py	(original)
+++ python/branches/py3k/Lib/shlex.py	Sun Jul  4 21:27:12 2010
@@ -1,4 +1,3 @@
-# -*- coding: iso-8859-1 -*-
 """A lexical analyzer class for simple shell-like syntaxes."""
 
 # Module and documentation by Eric S. Raymond, 21 Dec 1998
@@ -35,8 +34,8 @@
         self.wordchars = ('abcdfeghijklmnopqrstuvwxyz'
                           'ABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789_')
         if self.posix:
-            self.wordchars += ('????????????????????????????????'
-                               '??????????????????????????????')
+            self.wordchars += ('????????????????????????????????????????????????????????????????'
+                               '????????????????????????????????????????????????????????????')
         self.whitespace = ' \t\r\n'
         self.whitespace_split = False
         self.quotes = '\'"'

From python-checkins at python.org  Sun Jul  4 22:07:09 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun,  4 Jul 2010 22:07:09 +0200 (CEST)
Subject: [Python-checkins] r82560 - python/branches/py3k/Lib/shlex.py
Message-ID: <20100704200709.C420BEEA09@mail.python.org>

Author: mark.dickinson
Date: Sun Jul  4 22:07:09 2010
New Revision: 82560

Log:
Revert -r82559;  it's not clear that this is the right thing to do, and the change obscures the original intentions.

Modified:
   python/branches/py3k/Lib/shlex.py

Modified: python/branches/py3k/Lib/shlex.py
==============================================================================
--- python/branches/py3k/Lib/shlex.py	(original)
+++ python/branches/py3k/Lib/shlex.py	Sun Jul  4 22:07:09 2010
@@ -1,3 +1,4 @@
+# -*- coding: iso-8859-1 -*-
 """A lexical analyzer class for simple shell-like syntaxes."""
 
 # Module and documentation by Eric S. Raymond, 21 Dec 1998
@@ -34,8 +35,8 @@
         self.wordchars = ('abcdfeghijklmnopqrstuvwxyz'
                           'ABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789_')
         if self.posix:
-            self.wordchars += ('????????????????????????????????????????????????????????????????'
-                               '????????????????????????????????????????????????????????????')
+            self.wordchars += ('????????????????????????????????'
+                               '??????????????????????????????')
         self.whitespace = ' \t\r\n'
         self.whitespace_split = False
         self.quotes = '\'"'

From python-checkins at python.org  Sun Jul  4 22:47:00 2010
From: python-checkins at python.org (andrew.kuchling)
Date: Sun,  4 Jul 2010 22:47:00 +0200 (CEST)
Subject: [Python-checkins] r82561 - python/branches/py3k/Tools/README
Message-ID: <20100704204700.3A3E0F68E8@mail.python.org>

Author: andrew.kuchling
Date: Sun Jul  4 22:47:00 2010
New Revision: 82561

Log:
Add brief explanation; remove no-longer-working link

Modified:
   python/branches/py3k/Tools/README

Modified: python/branches/py3k/Tools/README
==============================================================================
--- python/branches/py3k/Tools/README	(original)
+++ python/branches/py3k/Tools/README	Sun Jul  4 22:47:00 2010
@@ -1,9 +1,7 @@
 This directory contains a number of Python programs that are useful
 while building or extending Python.
 
-faqwiz		FAQ Wizard.
-		See http://www.python.org/cgi-bin/faqw.py
-		for a live example.
+faqwiz		FAQ Wizard: a CGI script for a user-editable FAQ.
 
 freeze		Create a stand-alone executable from a Python program.
 

From python-checkins at python.org  Sun Jul  4 22:52:18 2010
From: python-checkins at python.org (brett.cannon)
Date: Sun,  4 Jul 2010 22:52:18 +0200 (CEST)
Subject: [Python-checkins] r82562 - in sandbox/trunk/pep362: README setup.py
	tests/test_pep362.py
Message-ID: <20100704205218.0EEF4F68C9@mail.python.org>

Author: brett.cannon
Date: Sun Jul  4 22:52:17 2010
New Revision: 82562

Log:
Update to run under Python 2.7 and 3.1 without any warnings.

Modified:
   sandbox/trunk/pep362/README
   sandbox/trunk/pep362/setup.py
   sandbox/trunk/pep362/tests/test_pep362.py

Modified: sandbox/trunk/pep362/README
==============================================================================
--- sandbox/trunk/pep362/README	(original)
+++ sandbox/trunk/pep362/README	Sun Jul  4 22:52:17 2010
@@ -38,6 +38,11 @@
 
 The version number will get bumped as open issues in the PEP are closed.
 
+0.6.2
+-----
+
+* Move test suite over to assert* methods.
+
 0.6
 ---
 

Modified: sandbox/trunk/pep362/setup.py
==============================================================================
--- sandbox/trunk/pep362/setup.py	(original)
+++ sandbox/trunk/pep362/setup.py	Sun Jul  4 22:52:17 2010
@@ -3,7 +3,7 @@
 setup(
         # Package metadata.
         name='pep362',
-        version='0.6.1',
+        version='0.6.2',
         description='Implementation of PEP 362 (Function Signature objects)',
         author='Brett Cannon',
         author_email='brett at python.org',
@@ -13,8 +13,12 @@
         packages=['tests'],
         data_files=['README'],
         classifiers=[
+            'Programming Language :: Python :: 2',
             'Programming Language :: Python :: 2.5',
             'Programming Language :: Python :: 2.6',
+            'Programming Language :: Python :: 2.7',
+            'Programming Language :: Python :: 3',
             'Programming Language :: Python :: 3.0',
+            'Programming Language :: Python :: 3.1',
         ]
     )

Modified: sandbox/trunk/pep362/tests/test_pep362.py
==============================================================================
--- sandbox/trunk/pep362/tests/test_pep362.py	(original)
+++ sandbox/trunk/pep362/tests/test_pep362.py	Sun Jul  4 22:52:17 2010
@@ -29,40 +29,40 @@
         # Must test both using a string and a tuple of strings.
         name = "test"
         param = pep362.Parameter(name, 0)
-        self.failUnlessEqual(param.name, name)
+        self.assertEqual(param.name, name)
         name = ('a', ('b',))
         param = pep362.Parameter(name, 0)
-        self.failUnlessEqual(param.name, name)
+        self.assertEqual(param.name, name)
 
     def test_position(self):
         # Test the 'position' attribute.
         pos = 42
         param = pep362.Parameter("_", pos)
-        self.failUnlessEqual(param.position, pos)
+        self.assertEqual(param.position, pos)
 
     def test_default_values(self):
         # Testing that 'default_value' is not set is handled in the testing of
         # that attribute.
         default_value = 42
         param = pep362.Parameter('_', 0, True, default_value)
-        self.failUnlessEqual(param.default_value, default_value)
+        self.assertEqual(param.default_value, default_value)
         param = pep362.Parameter('_', 0, False)
-        self.failUnless(not hasattr(param, 'default_value'))
+        self.assertTrue(not hasattr(param, 'default_value'))
 
     def test_keyword_only(self):
         # Setting the value for keyword_only should create an attribute.
         for value in (True, False):
             param = pep362.Parameter('_', 0, keyword_only=value)
-            self.failUnlessEqual(param.keyword_only, value)
+            self.assertEqual(param.keyword_only, value)
 
     def test_annotations(self):
         # If has_annotation is False then 'annotation' should not exist.
         param = pep362.Parameter('_', 0, has_annotation=False)
-        self.failUnless(not hasattr(param, 'annotation'))
+        self.assertTrue(not hasattr(param, 'annotation'))
         annotation = 42
         param = pep362.Parameter('_', 0, has_annotation=True,
                                     annotation=annotation)
-        self.failUnlessEqual(param.annotation, annotation)
+        self.assertEqual(param.annotation, annotation)
 
 
 class SignatureObjectTests(unittest.TestCase):
@@ -71,152 +71,152 @@
         # __getitem__() should return the Parameter object for the name
         # parameter.
         sig = pep362.Signature(pep362_fodder.default_args)
-        self.failUnless(sig['a'])
+        self.assertTrue(sig['a'])
         param = sig['a']
-        self.failUnlessEqual(param.name, 'a')
+        self.assertTrue(param.name, 'a')
 
     def test_iter(self):
         # The iterator should return all Parameter objects in the proper order.
         sig = pep362.Signature(pep362_fodder.default_args)
         params = list(sig)
-        self.failUnlessEqual(len(params), 1)
-        self.failUnlessEqual(params[0].name, 'a')
+        self.assertEqual(len(params), 1)
+        self.assertEqual(params[0].name, 'a')
 
     def test_no_args(self):
         # Test a function with no arguments.
         sig = pep362.Signature(pep362_fodder.no_args)
-        self.failUnlessEqual('no_args', sig.name)
-        self.failUnless(not sig.var_args)
-        self.failUnless(not sig.var_kw_args)
-        self.failUnlessEqual(0, len(list(sig)))
+        self.assertEqual('no_args', sig.name)
+        self.assertTrue(not sig.var_args)
+        self.assertTrue(not sig.var_kw_args)
+        self.assertEqual(0, len(list(sig)))
 
     def test_var_args(self):
         # Test the var_args attribute.
         sig = pep362.Signature(pep362_fodder.var_args)
-        self.failUnlessEqual('args', sig.var_args)
-        self.failUnlessEqual(0, len(list(sig)))
+        self.assertEqual('args', sig.var_args)
+        self.assertEqual(0, len(list(sig)))
         sig = pep362.Signature(pep362_fodder.no_args)
-        self.failUnlessEqual('', sig.var_args)
+        self.assertEqual('', sig.var_args)
 
     def test_var_kw_args(self):
         # Test the var_kw_args attribute and annotations.
         sig = pep362.Signature(pep362_fodder.var_kw_args)
-        self.failUnlessEqual('var_kw_args', sig.name)
-        self.failUnlessEqual('kwargs', sig.var_kw_args)
-        self.failUnlessEqual(0, len(list(sig)))
+        self.assertEqual('var_kw_args', sig.name)
+        self.assertEqual('kwargs', sig.var_kw_args)
+        self.assertEqual(0, len(list(sig)))
         sig = pep362.Signature(pep362_fodder.no_args)
-        self.failUnlessEqual('', sig.var_kw_args)
+        self.assertEqual('', sig.var_kw_args)
 
     def test_parameter_positional(self):
         # A function with positional arguments should work.
         sig = pep362.Signature(pep362_fodder.no_default_args)
-        self.failUnlessEqual('no_default_args', sig.name)
+        self.assertEqual('no_default_args', sig.name)
         param = sig['a']
-        self.failUnlessEqual('a', param.name)
-        self.failUnlessEqual(0, param.position)
-        self.failUnless(not hasattr(param, 'default_value'))
+        self.assertEqual('a', param.name)
+        self.assertEqual(0, param.position)
+        self.assertTrue(not hasattr(param, 'default_value'))
 
     def test_parameter_default(self):
         # Default parameters for a function should work.
         sig = pep362.Signature(pep362_fodder.default_args)
-        self.failUnlessEqual('default_args', sig.name)
+        self.assertEqual('default_args', sig.name)
         param = sig['a']
-        self.failUnlessEqual('a', param.name)
-        self.failUnlessEqual(0, param.position)
-        self.failUnlessEqual(42, param.default_value)
+        self.assertEqual('a', param.name)
+        self.assertEqual(0, param.position)
+        self.assertEqual(42, param.default_value)
 
     @version_specific(2)
     def test_parameter_tuple(self):
         # A function with a tuple as a parameter should work.
         sig = pep362.Signature(pep362_py2_fodder.tuple_args)
-        self.failUnlessEqual('tuple_args', sig.name)
+        self.assertEqual('tuple_args', sig.name)
         param = list(sig)[0]
-        self.failUnless(isinstance(param.name, tuple))
-        self.failUnlessEqual(('a', ('b',)), param.name)
-        self.failUnlessEqual(0, param.position)
-        self.failUnless(not hasattr(param, 'default_value'))
+        self.assertTrue(isinstance(param.name, tuple))
+        self.assertEqual(('a', ('b',)), param.name)
+        self.assertEqual(0, param.position)
+        self.assertTrue(not hasattr(param, 'default_value'))
 
     @version_specific(2)
     def test_parameter_tuple_default(self):
         # A default argument for a tuple parameter needs to work.
         sig = pep362.Signature(pep362_py2_fodder.default_tuple_args)
-        self.failUnlessEqual('default_tuple_args', sig.name)
+        self.assertEqual('default_tuple_args', sig.name)
         param = list(sig)[0]
-        self.failUnlessEqual(('a', ('b',)), param.name)
-        self.failUnlessEqual(0, param.position)
-        self.failUnlessEqual((1, (2,)), param.default_value)
+        self.assertEqual(('a', ('b',)), param.name)
+        self.assertEqual(0, param.position)
+        self.assertEqual((1, (2,)), param.default_value)
 
     @version_specific(3)
     def test_keyword_only(self):
         # Is a function containing keyword-only parameters handled properly?
         sig = pep362.Signature(pep362_py3k_fodder.keyword_only)
         param = sig['a']
-        self.failUnlessEqual(param.name, 'a')
-        self.failUnless(param.keyword_only)
-        self.failUnlessEqual(param.position, 0)
+        self.assertEqual(param.name, 'a')
+        self.assertTrue(param.keyword_only)
+        self.assertEqual(param.position, 0)
 
     @version_specific(3)
     def test_keyword_only_default(self):
         # Default arguments can work for keyword-only parameters.
         sig = pep362.Signature(pep362_py3k_fodder.keyword_only_default)
         param = sig['a']
-        self.failUnlessEqual(param.name, 'a')
-        self.failUnless(param.keyword_only)
-        self.failUnlessEqual(param.position, 0)
-        self.failUnlessEqual(param.default_value, 42)
+        self.assertEqual(param.name, 'a')
+        self.assertTrue(param.keyword_only)
+        self.assertEqual(param.position, 0)
+        self.assertEqual(param.default_value, 42)
 
     @version_specific(3)
     def test_annotations(self):
         # Make sure the proper annotation is found.
         sig = pep362.Signature(pep362_py3k_fodder.arg_annotation)
         param = sig['a']
-        self.failUnlessEqual(param.name, 'a')
-        self.failUnlessEqual(param.annotation, int)
+        self.assertEqual(param.name, 'a')
+        self.assertEqual(param.annotation, int)
 
     @version_specific(3)
     def test_annotations_default(self):
         # Annotations with a default value should work.
         sig = pep362.Signature(pep362_py3k_fodder.arg_annotation_default)
         param = sig['a']
-        self.failUnlessEqual(param.name, 'a')
-        self.failUnlessEqual(param.annotation, int)
-        self.failUnlessEqual(param.default_value, 42)
+        self.assertEqual(param.name, 'a')
+        self.assertEqual(param.annotation, int)
+        self.assertEqual(param.default_value, 42)
 
     @version_specific(3)
     def test_annotation_keyword_only(self):
         # Keyword-only parameters can have an annotation.
         sig = pep362.Signature(pep362_py3k_fodder.arg_annotation_keyword_only)
         param = sig['a']
-        self.failUnlessEqual(param.name, 'a')
-        self.failUnlessEqual(param.annotation, int)
-        self.failUnless(param.keyword_only)
+        self.assertEqual(param.name, 'a')
+        self.assertEqual(param.annotation, int)
+        self.assertTrue(param.keyword_only)
 
     @version_specific(3)
     def test_return_annotation(self):
         # The return value annotation.
         sig = pep362.Signature(pep362_py3k_fodder.return_annotation)
-        self.failUnlessEqual(sig.return_annotation, int)
+        self.assertEqual(sig.return_annotation, int)
 
     @version_specific(3)
     def test_var_annotations(self):
         # Annotation on variable arguments (*args & **kwargs).
         sig = pep362.Signature(pep362_py3k_fodder.arg_annotation_var)
-        self.failUnlessEqual(sig.var_annotations[sig.var_args], int)
-        self.failUnlessEqual(sig.var_annotations[sig.var_kw_args], str)
+        self.assertEqual(sig.var_annotations[sig.var_args], int)
+        self.assertEqual(sig.var_annotations[sig.var_kw_args], str)
 
     def test_signature(self):
         def fresh_func():
             pass
-        self.failUnless(not hasattr(fresh_func, '__signature__'))
+        self.assertTrue(not hasattr(fresh_func, '__signature__'))
         sig = pep362.signature(fresh_func)
-        self.failUnlessEqual(sig, fresh_func.__signature__)
+        self.assertEqual(sig, fresh_func.__signature__)
         sig2 = pep362.signature(fresh_func)
-        self.failUnlessEqual(sig, sig2)
+        self.assertEqual(sig, sig2)
         class FreshClass(object):
             def fresh_method(self):
                 pass
         sig = pep362.signature(FreshClass.fresh_method)
-        self.failUnlessEqual(sig, FreshClass.fresh_method.__signature__)
+        self.assertEqual(sig, FreshClass.fresh_method.__signature__)
 
 
 class SignatureBindTests(unittest.TestCase):
@@ -226,65 +226,65 @@
     def test_no_parameters(self):
         sig = pep362.Signature(pep362_fodder.no_args)
         binding = sig.bind()
-        self.failUnlessEqual({}, binding)
-        self.failUnlessRaises(pep362.BindError, sig.bind, 42)
-        self.failUnlessRaises(pep362.BindError, sig.bind, a=0)
+        self.assertEqual({}, binding)
+        self.assertRaises(pep362.BindError, sig.bind, 42)
+        self.assertRaises(pep362.BindError, sig.bind, a=0)
 
     def test_var_parameters(self):
         sig = pep362.Signature(pep362_fodder.var_args)
         binding = sig.bind(0, 1, 2)
-        self.failUnlessEqual({'args':(0, 1, 2)}, binding)
+        self.assertEqual({'args':(0, 1, 2)}, binding)
         binding = sig.bind()
-        self.failUnlessEqual({'args':tuple()}, binding)
-        self.failUnlessRaises(pep362.BindError, sig.bind, a=0)
+        self.assertEqual({'args':tuple()}, binding)
+        self.assertRaises(pep362.BindError, sig.bind, a=0)
 
     def test_var_kw_parameters(self):
         sig = pep362.Signature(pep362_fodder.var_kw_args)
         binding = sig.bind(a=0)
-        self.failUnlessEqual({'kwargs':{'a':0}}, binding)
+        self.assertEqual({'kwargs':{'a':0}}, binding)
         binding = sig.bind()
-        self.failUnlessEqual({'kwargs':{}}, binding)
-        self.failUnlessRaises(pep362.BindError, sig.bind, 42)
+        self.assertEqual({'kwargs':{}}, binding)
+        self.assertRaises(pep362.BindError, sig.bind, 42)
 
     def test_positional_parameters(self):
         sig = pep362.Signature(pep362_fodder.no_default_args)
         binding = sig.bind(42)
-        self.failUnlessEqual({'a':42}, binding)
+        self.assertEqual({'a':42}, binding)
         binding = sig.bind(a=42)
-        self.failUnlessEqual({'a':42}, binding)
-        self.failUnlessRaises(pep362.BindError, sig.bind)
-        self.failUnlessRaises(pep362.BindError, sig.bind, 0, 1)
-        self.failUnlessRaises(pep362.BindError, sig.bind, b=0)
+        self.assertEqual({'a':42}, binding)
+        self.assertRaises(pep362.BindError, sig.bind)
+        self.assertRaises(pep362.BindError, sig.bind, 0, 1)
+        self.assertRaises(pep362.BindError, sig.bind, b=0)
 
     def test_keyword_parameters(self):
         sig = pep362.Signature(pep362_fodder.default_args)
         binding = sig.bind()
-        self.failUnlessEqual({'a':42}, binding)
+        self.assertEqual({'a':42}, binding)
         binding = sig.bind(0)
-        self.failUnlessEqual({'a':0}, binding)
+        self.assertEqual({'a':0}, binding)
         binding = sig.bind(a=0)
-        self.failUnlessEqual({'a':0}, binding)
-        self.failUnlessRaises(pep362.BindError, sig.bind, 0, 1)
-        self.failUnlessRaises(pep362.BindError, sig.bind, a=0, b=1)
-        self.failUnlessRaises(pep362.BindError, sig.bind, b=1)
+        self.assertEqual({'a':0}, binding)
+        self.assertRaises(pep362.BindError, sig.bind, 0, 1)
+        self.assertRaises(pep362.BindError, sig.bind, a=0, b=1)
+        self.assertRaises(pep362.BindError, sig.bind, b=1)
 
     @version_specific(2)
     def test_tuple_parameter(self):
         sig = pep362.Signature(pep362_py2_fodder.tuple_args)
         arg = (1, ((2,),))
         binding = sig.bind(arg)
-        self.failUnlessEqual({'a':1, 'b':(2,)}, binding)
-        self.failUnlessRaises(pep362.BindError, sig.bind, (1,2,3))
-        self.failUnlessRaises(pep362.BindError, sig.bind, (1, 2))
+        self.assertEqual({'a':1, 'b':(2,)}, binding)
+        self.assertRaises(pep362.BindError, sig.bind, (1,2,3))
+        self.assertRaises(pep362.BindError, sig.bind, (1, 2))
 
     @version_specific(2)
     def test_default_tuple_parameter(self):
         sig = pep362.Signature(pep362_py2_fodder.default_tuple_args)
         binding = sig.bind()
-        self.failUnlessEqual({'a':1, 'b':2}, binding)
+        self.assertEqual({'a':1, 'b':2}, binding)
         arg = (0, (1,))
         binding = sig.bind(arg)
-        self.failUnlessEqual({'a':0, 'b':1}, binding)
+        self.assertEqual({'a':0, 'b':1}, binding)
 
     @version_specific(2)
     def test_py2_all_args(self):
@@ -295,41 +295,41 @@
                     (('b', ('c',)), 1, False, None),
                     ('d', 2, True, 0),
                     (('e', ('f',)), 3, True, (4, (5,))))
-        self.failUnlessEqual(len(list(sig)), len(expect))
+        self.assertEqual(len(list(sig)), len(expect))
         for param, check in zip(list(sig), expect):
             name, pos, has_default, default_value = check
-            self.failUnlessEqual(param.name, name)
-            self.failUnlessEqual(param.position, pos)
+            self.assertEqual(param.name, name)
+            self.assertEqual(param.position, pos)
             if has_default:
-                self.failUnlessEqual(param.default_value, default_value)
+                self.assertEqual(param.default_value, default_value)
             else:
-                self.failUnless(not hasattr(param, 'default_value'))
-            self.failUnless(not param.keyword_only)
-            self.failUnless(not hasattr(param, 'annotation'))
-        self.failUnlessEqual(sig.var_args, 'g')
-        self.failUnlessEqual(sig.var_kw_args, 'h')
-        self.failUnlessEqual(len(sig.var_annotations), 0)
+                self.assertTrue(not hasattr(param, 'default_value'))
+            self.assertTrue(not param.keyword_only)
+            self.assertTrue(not hasattr(param, 'annotation'))
+        self.assertEqual(sig.var_args, 'g')
+        self.assertEqual(sig.var_kw_args, 'h')
+        self.assertEqual(len(sig.var_annotations), 0)
         binding = sig.bind(0, (1, (2,)), d=3, i=7)
         expected = {'a':0, 'b':1, 'c':2, 'd':3, 'e':4, 'f':5, 'g':tuple(),
                     'h':{'i':7}}
-        self.failUnlessEqual(expected, binding)
+        self.assertEqual(expected, binding)
 
     @version_specific(3)
     def test_keyword_only(self):
         sig = pep362.Signature(pep362_py3k_fodder.keyword_only)
         binding = sig.bind(a=42)
-        self.failUnlessEqual(binding, {'a':42})
-        self.failUnlessRaises(pep362.BindError, sig.bind)
-        self.failUnlessRaises(pep362.BindError, sig.bind, 42)
+        self.assertEqual(binding, {'a':42})
+        self.assertRaises(pep362.BindError, sig.bind)
+        self.assertRaises(pep362.BindError, sig.bind, 42)
 
     @version_specific(3)
     def test_keyword_only_default(self):
         sig = pep362.Signature(pep362_py3k_fodder.keyword_only_default)
         binding = sig.bind()
-        self.failUnlessEqual(binding, {'a':42})
+        self.assertEqual(binding, {'a':42})
         binding = sig.bind(a=1)
-        self.failUnlessEqual(binding, {'a':1})
-        self.failUnlessRaises(pep362.BindError, sig.bind, 1)
+        self.assertEqual(binding, {'a':1})
+        self.assertRaises(pep362.BindError, sig.bind, 1)
 
     @version_specific(3)
     def test_all_py3k_args(self):
@@ -340,41 +340,41 @@
                     ('d', 1, False, True, 0, False, None),
                     ('g', 2, True, False, None, True, int),
                     ('h', 3, True, True, 8, True, int))
-        self.failUnlessEqual(len(list(sig)), len(expected),
+        self.assertEqual(len(list(sig)), len(expected),
                 "len(%r) != len(%r)" % ([param.name
                                             for param in sig],
                                         [expect[0] for expect in expected]))
         for param, check in zip(sig, expected):
             name, pos, kw_only, has_default, default, has_anno, anno = check
-            self.failUnlessEqual(param.name, name)
-            self.failUnlessEqual(param.position, pos)
+            self.assertEqual(param.name, name)
+            self.assertEqual(param.position, pos)
             if kw_only:
-                self.failUnless(param.keyword_only)
+                self.assertTrue(param.keyword_only)
             else:
-                self.failUnless(not param.keyword_only)
+                self.assertTrue(not param.keyword_only)
             if has_default:
-                self.failUnlessEqual(param.default_value, default)
+                self.assertEqual(param.default_value, default)
             else:
-                self.failUnless(not hasattr(param, 'default_value'))
+                self.assertTrue(not hasattr(param, 'default_value'))
             if has_anno:
-                self.failUnlessEqual(param.annotation, anno)
+                self.assertEqual(param.annotation, anno)
             else:
-                self.failUnless(not hasattr(param, 'annotation'))
-        self.failUnlessEqual(sig.var_args, 'args')
-        self.failUnless(sig.var_args in sig.var_annotations)
-        self.failUnlessEqual(sig.var_annotations[sig.var_args], int)
-        self.failUnlessEqual(sig.var_kw_args, 'kwargs')
-        self.failUnless(sig.var_kw_args in sig.var_annotations)
-        self.failUnlessEqual(sig.var_annotations[sig.var_kw_args], int)
-        self.failUnlessEqual(sig.return_annotation, int)
+                self.assertTrue(not hasattr(param, 'annotation'))
+        self.assertEqual(sig.var_args, 'args')
+        self.assertTrue(sig.var_args in sig.var_annotations)
+        self.assertEqual(sig.var_annotations[sig.var_args], int)
+        self.assertEqual(sig.var_kw_args, 'kwargs')
+        self.assertTrue(sig.var_kw_args in sig.var_annotations)
+        self.assertEqual(sig.var_annotations[sig.var_kw_args], int)
+        self.assertEqual(sig.return_annotation, int)
         binding = sig.bind(0, 3, 6, g=7, i=9)
         expected = {'a':0, 'd':3, 'g':7, 'h':8, 'args':(6,), 'kwargs':{'i':9}}
-        self.failUnlessEqual(binding, expected)
+        self.assertEqual(binding, expected)
 
     def test_too_many_arguments(self):
         # Only one argument should pair up with a parameter.
         sig = pep362.Signature(pep362_fodder.no_default_args)
-        self.failUnlessRaises(pep362.BindError, sig.bind, 1, a=1)
+        self.assertRaises(pep362.BindError, sig.bind, 1, a=1)
 
 
 def test_main():

From python-checkins at python.org  Sun Jul  4 23:03:54 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sun,  4 Jul 2010 23:03:54 +0200 (CEST)
Subject: [Python-checkins] r82563 -
	python/branches/release27-maint/Include/patchlevel.h
Message-ID: <20100704210354.199D8EDE8@mail.python.org>

Author: benjamin.peterson
Date: Sun Jul  4 23:03:53 2010
New Revision: 82563

Log:
technically not at 2.7.1 yet

Modified:
   python/branches/release27-maint/Include/patchlevel.h

Modified: python/branches/release27-maint/Include/patchlevel.h
==============================================================================
--- python/branches/release27-maint/Include/patchlevel.h	(original)
+++ python/branches/release27-maint/Include/patchlevel.h	Sun Jul  4 23:03:53 2010
@@ -22,12 +22,12 @@
 /*--start constants--*/
 #define PY_MAJOR_VERSION	2
 #define PY_MINOR_VERSION	7
-#define PY_MICRO_VERSION	1
-#define PY_RELEASE_LEVEL	PY_RELEASE_LEVEL_ALPHA
+#define PY_MICRO_VERSION	0
+#define PY_RELEASE_LEVEL	PY_RELEASE_LEVEL_FINAL
 #define PY_RELEASE_SERIAL	0
 
 /* Version as a string */
-#define PY_VERSION      	"2.7.1a0"
+#define PY_VERSION      	"2.7.0+"
 /*--end constants--*/
 
 /* Subversion Revision number of this file (not of the repository) */

From python-checkins at python.org  Sun Jul  4 23:09:50 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Sun,  4 Jul 2010 23:09:50 +0200 (CEST)
Subject: [Python-checkins] r82564 - peps/trunk/pep-0381.txt
Message-ID: <20100704210950.A5EB3EE9A9@mail.python.org>

Author: martin.v.loewis
Date: Sun Jul  4 23:09:50 2010
New Revision: 82564

Log:
Make myself coauthor.


Modified:
   peps/trunk/pep-0381.txt

Modified: peps/trunk/pep-0381.txt
==============================================================================
--- peps/trunk/pep-0381.txt	(original)
+++ peps/trunk/pep-0381.txt	Sun Jul  4 23:09:50 2010
@@ -2,7 +2,7 @@
 Title: Mirroring infrastructure for PyPI
 Version: $Revision$
 Last-Modified: $Date$
-Author: Tarek Ziad? <tarek at ziade.org>
+Author: Tarek Ziad? <tarek at ziade.org>, Martin v. L?wis <martin at v.loewis.de>
 Status: Draft
 Type: Standards Track
 Content-Type: text/x-rst

From python-checkins at python.org  Sun Jul  4 23:13:29 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Sun,  4 Jul 2010 23:13:29 +0200 (CEST)
Subject: [Python-checkins] r82565 - peps/trunk/pep-0381.txt
Message-ID: <20100704211329.3B475EEA03@mail.python.org>

Author: martin.v.loewis
Date: Sun Jul  4 23:13:29 2010
New Revision: 82565

Log:
Document name assignments to mirrors.


Modified:
   peps/trunk/pep-0381.txt

Modified: peps/trunk/pep-0381.txt
==============================================================================
--- peps/trunk/pep-0381.txt	(original)
+++ peps/trunk/pep-0381.txt	Sun Jul  4 23:13:29 2010
@@ -46,16 +46,17 @@
 added in a mirror list in the PyPI application after it
 has been checked to be compliant with the mirroring rules.
 
-The mirror list is handled by a DNS entry for this hostname:
+The mirror list is provided as a list of host names of the
+form
 
-    mirrors.pypi.python.org
+    X.pypi.python.org
 
-When a mirror is added into the DNS, it becomes an official
-IP for `mirrors.pypi.python.org`, and requests will be sent
-to the given IP. Therefore the mirror maintainer should not
-change the IP provided. If the IP has to change for any reason,
-the mirror maintainer has to send a mail to catalog-SIG at least
-one week before the change so the DNS entry can be changed on time.
+The values of X are the sequence a,b,c,...,aa,ab,...
+a.pypi.python.org is the master server; the mirrors start
+with b. A CNAME record last.pypi.python.org points to the
+last host name. Mirror operators should use a static address,
+and report planned changes to that address in advance to
+catalog-sig.
 
 The new mirror also appears at `http://pypi.python.org/mirrors`
 which is a human-readable page that gives the list of mirrors.

From python-checkins at python.org  Sun Jul  4 23:33:54 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Sun,  4 Jul 2010 23:33:54 +0200 (CEST)
Subject: [Python-checkins] r82566 - peps/trunk/pep-0381.txt
Message-ID: <20100704213354.69ED4EE988@mail.python.org>

Author: martin.v.loewis
Date: Sun Jul  4 23:33:54 2010
New Revision: 82566

Log:
Describe the mirroring protocol. Drop URLs that don't
need to be mirrored.


Modified:
   peps/trunk/pep-0381.txt

Modified: peps/trunk/pep-0381.txt
==============================================================================
--- peps/trunk/pep-0381.txt	(original)
+++ peps/trunk/pep-0381.txt	Sun Jul  4 23:33:54 2010
@@ -87,14 +87,12 @@
 Special pages a mirror needs to provide
 =======================================
 
-A mirror is a strict copy of PyPI, so it provides the same structure
+A mirror is a subset copy of PyPI, so it provides the same structure
 by copying it.
 
-- pypi: html version of the package index
 - simple: rest version of the package index
 - packages: packages, stored by Python version, and letters
-- stats : statistics on downloads
-- XXX
+- serversig: signatures for the simple pages
 
 It also needs to provide two specific elements:
 
@@ -166,9 +164,19 @@
 The mirroring protocol
 ::::::::::::::::::::::
 
-XXX Need to describe the protocol here.
+Mirrors must reduce the amount of data transfered between the central
+server and the mirror. To achieve that, they MUST use the changelog()
+PyPI XML-RPC call, and only refetch the packages that have been
+changed since the last time. For each package P, they MUST copy
+documents /simple/P/ and /serversig/P. If a package is deleted on the
+central server, they MUST delete the package and all associated files.
+To detect modification of package files, they MAY cache the file's
+ETag, and MAY request skipping it using the If-none-match header.
 
-The z3c.pypimirror package [#zcpkg]_ provides an application that
+Each mirroring tool MUST identify itself using a descripte User-agent
+header.
+
+The pep381client package [#pep381client]_ provides an application that
 respects this protocol to browse PyPI.
 
 User-agent request header 
@@ -190,13 +198,13 @@
 :::::::::::::::::::::::::::::::::::::::::
 
 Clients that are browsing PyPI should be able to use alternative
-mirrors, by getting the list of the mirrors using `mirrors.pypi.python.org`.
+mirrors, by getting the list of the mirrors using `last.pypi.python.org`.
 
 Code example::
 
     >>> import socket
-    >>> socket.gethostbyname_ex('mirrors.pypi.python.org')[-1]
-    ['82.94.164.163', '88.191.64.248']
+    >>> socket.gethostbyname_ex('last.pypi.python.org')[0]
+    'h.pypi.python.org'
 
 The clients so far that could use this mechanism:
 
@@ -268,8 +276,8 @@
 .. [#pep305]
    http://www.python.org/dev/peps/pep-0305/#id19
 
-.. [#zcpkg]
-   http://pypi.python.org/pypi/z3c.pypimirror
+.. [#pep381client]
+   http://pypi.python.org/pypi/pep381client
 
 .. [#iso8601]
    http://en.wikipedia.org/wiki/ISO_8601
@@ -290,10 +298,10 @@
    http://www.chrisarndt.de/projects/eggbasket
 
 
-Aknowledgments
-==============
+Acknowledgments
+===============
 
-Martin von Loewis, Georg Brandl.
+Georg Brandl.
 
 
 Copyright

From python-checkins at python.org  Mon Jul  5 00:05:34 2010
From: python-checkins at python.org (brett.cannon)
Date: Mon,  5 Jul 2010 00:05:34 +0200 (CEST)
Subject: [Python-checkins] r82567 -
	python/branches/py3k/Tools/scripts/patchcheck.py
Message-ID: <20100704220534.C2635EE9E2@mail.python.org>

Author: brett.cannon
Date: Mon Jul  5 00:05:34 2010
New Revision: 82567

Log:
Do some basic refactoring. Along the way also list what files had their
whitespace cleaned up and do the proper plurality for "N file(s)" based on N.

Refactoring closes issue8912. Thanks to ?ric Araujo for the patch.


Modified:
   python/branches/py3k/Tools/scripts/patchcheck.py

Modified: python/branches/py3k/Tools/scripts/patchcheck.py
==============================================================================
--- python/branches/py3k/Tools/scripts/patchcheck.py	(original)
+++ python/branches/py3k/Tools/scripts/patchcheck.py	Mon Jul  5 00:05:34 2010
@@ -5,6 +5,10 @@
 import reindent
 
 
+def n_files_str(count):
+    """Return 'N file(s)' with the proper plurality on 'file'."""
+    return "{} file{}".format(count, "s" if count != 1 else "")
+
 def status(message, modal=False, info=None):
     """Decorator to output status info to stdout."""
     def decorated_fxn(fxn):
@@ -26,7 +30,7 @@
     return decorated_fxn
 
 @status("Getting the list of files that have been added/changed",
-            info=lambda x: "%s files" % len(x))
+            info=lambda x: n_files_str(len(x)))
 def changed_files():
     """Run ``svn status`` and return a set of files that have been
     changed/added."""
@@ -44,20 +48,30 @@
             files.add(path)
     return files
 
- at status("Fixing whitespace", info=lambda x: "%s files" % x)
+def report_modified_files(file_paths):
+    count = len(file_paths)
+    if count == 0:
+        return n_files_str(count)
+    else:
+        lines = ["{}:".format(n_files_str(count))]
+        for path in file_paths:
+            lines.append("  {}".format(path))
+        return "\n".join(lines)
+
+ at status("Fixing whitespace", info=report_modified_files)
 def normalize_whitespace(file_paths):
     """Make sure that the whitespace for .py files have been normalized."""
     reindent.makebackup = False  # No need to create backups.
-    result = list(map(reindent.check, (x for x in file_paths if x.endswith('.py'))))
-    return sum(result)
+    fixed = []
+    for path in (x for x in file_paths if x.endswith('.py')):
+        if reindent.check(path):
+            fixed.append(path)
+    return fixed
 
 @status("Docs modified", modal=True)
 def docs_modified(file_paths):
-    """Report if any files in the Docs directory."""
-    for path in file_paths:
-        if path.startswith("Doc"):
-            return True
-    return False
+    """Report if any file in the Doc directory has been changed."""
+    return bool(file_paths)
 
 @status("Misc/ACKS updated", modal=True)
 def credit_given(file_paths):
@@ -72,14 +86,18 @@
 
 def main():
     file_paths = changed_files()
-    # PEP 7/8 verification.
-    normalize_whitespace(file_paths)
+    python_files = [fn for fn in file_paths if fn.endswith('.py')]
+    c_files = [fn for fn in file_paths if fn.endswith(('.c', '.h'))]
+    docs = [fn for fn in file_paths if fn.startswith('Doc')]
+    special_files = {'Misc/ACKS', 'Misc/NEWS'} & set(file_paths)
+    # PEP 8 whitespace rules enforcement.
+    normalize_whitespace(python_files)
     # Docs updated.
-    docs_modified(file_paths)
+    docs_modified(docs)
     # Misc/ACKS changed.
-    credit_given(file_paths)
+    credit_given(special_files)
     # Misc/NEWS changed.
-    reported_news(file_paths)
+    reported_news(special_files)
 
     # Test suite run and passed.
     print()

From solipsis at pitrou.net  Mon Jul  5 01:24:08 2010
From: solipsis at pitrou.net (solipsis at pitrou.net)
Date: Mon,  5 Jul 2010 01:24:08 +0200 (CEST)
Subject: [Python-checkins] Daily py3k reference leaks (r82561): sum=0
Message-ID: <20100704232408.DCC4A1770A@ns6635.ovh.net>

py3k results for svn r82561 (hg cset a7135c36f6d7)
--------------------------------------------------



Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/antoine/py3k/refleaks/reflogGiOYa7', '-x']

From python-checkins at python.org  Mon Jul  5 04:33:34 2010
From: python-checkins at python.org (ezio.melotti)
Date: Mon,  5 Jul 2010 04:33:34 +0200 (CEST)
Subject: [Python-checkins] r82568 -
	tracker/instances/python-dev/html/page.html
Message-ID: <20100705023334.CE11BEE9A6@mail.python.org>

Author: ezio.melotti
Date: Mon Jul  5 04:33:34 2010
New Revision: 82568

Log:
Remember the search text in the input box when loading the page with the results.

Modified:
   tracker/instances/python-dev/html/page.html

Modified: tracker/instances/python-dev/html/page.html
==============================================================================
--- tracker/instances/python-dev/html/page.html	(original)
+++ tracker/instances/python-dev/html/page.html	Mon Jul  5 04:33:34 2010
@@ -42,7 +42,8 @@
        <input type="hidden" name="@sort" value="-activity" />
        <input type="hidden" name="@filter" value="status"/>
        <input type="hidden" name="@action" value="searchid"/>
-       <input class="input-text" id="search-text" name="@search_text" size="10" />
+       <input tal:attributes="value python:request.form.getvalue('@search_text') or nothing;"
+              class="input-text" id="search-text" name="@search_text" size="10" />
        <input type="submit" id="submit" value="search" name="submit"
               class="input-button"/>
        <input type="radio" name="status" tal:attributes="value status_notresolved;

From ezio.melotti at gmail.com  Mon Jul  5 04:45:43 2010
From: ezio.melotti at gmail.com (Ezio Melotti)
Date: Mon, 05 Jul 2010 05:45:43 +0300
Subject: [Python-checkins] r82562 - in sandbox/trunk/pep362: README
 setup.py tests/test_pep362.py
In-Reply-To: <20100704205218.0EEF4F68C9@mail.python.org>
References: <20100704205218.0EEF4F68C9@mail.python.org>
Message-ID: <4C314757.2060505@gmail.com>

On 04/07/2010 23.52, brett.cannon wrote:
> Author: brett.cannon
> Date: Sun Jul  4 22:52:17 2010
> New Revision: 82562
>
> Log:
> Update to run under Python 2.7 and 3.1 without any warnings.
>
> Modified:
>     sandbox/trunk/pep362/README
>     sandbox/trunk/pep362/setup.py
>     sandbox/trunk/pep362/tests/test_pep362.py
>
> Modified: sandbox/trunk/pep362/README
> ==============================================================================
> --- sandbox/trunk/pep362/README	(original)
> +++ sandbox/trunk/pep362/README	Sun Jul  4 22:52:17 2010
> @@ -38,6 +38,11 @@
>
>   The version number will get bumped as open issues in the PEP are closed.
>
> +0.6.2
> +-----
> +
> +* Move test suite over to assert* methods.
> +
>   0.6
>   ---
>
>
> Modified: sandbox/trunk/pep362/setup.py
> ==============================================================================
> --- sandbox/trunk/pep362/setup.py	(original)
> +++ sandbox/trunk/pep362/setup.py	Sun Jul  4 22:52:17 2010
> @@ -3,7 +3,7 @@
> [...]
>       @version_specific(2)
>       def test_parameter_tuple(self):
>           # A function with a tuple as a parameter should work.
>           sig = pep362.Signature(pep362_py2_fodder.tuple_args)
> -        self.failUnlessEqual('tuple_args', sig.name)
> +        self.assertEqual('tuple_args', sig.name)
>           param = list(sig)[0]
> -        self.failUnless(isinstance(param.name, tuple))
> -        self.failUnlessEqual(('a', ('b',)), param.name)
> -        self.failUnlessEqual(0, param.position)
> -        self.failUnless(not hasattr(param, 'default_value'))
> +        self.assertTrue(isinstance(param.name, tuple))
This could be an assertIsInstance (unless you need to support Python <=2.6).

> +        self.assertEqual(('a', ('b',)), param.name)
> +        self.assertEqual(0, param.position)
> +        self.assertTrue(not hasattr(param, 'default_value'))
This and several others could use assertFalse, instead of assertTrue(not 
...)

>
> [...]

Best Regards,
Ezio Melotti

From python-checkins at python.org  Mon Jul  5 07:31:22 2010
From: python-checkins at python.org (ezio.melotti)
Date: Mon,  5 Jul 2010 07:31:22 +0200 (CEST)
Subject: [Python-checkins] r82569 -
	tracker/instances/python-dev/html/style.css
Message-ID: <20100705053122.6F4D3EE9AC@mail.python.org>

Author: ezio.melotti
Date: Mon Jul  5 07:31:22 2010
New Revision: 82569

Log:
Reduce indentation in the left menu to leave more space for the queries and be aligned with the rest. Also remove duplicate rule and unnecessary units.

Modified:
   tracker/instances/python-dev/html/style.css

Modified: tracker/instances/python-dev/html/style.css
==============================================================================
--- tracker/instances/python-dev/html/style.css	(original)
+++ tracker/instances/python-dev/html/style.css	Mon Jul  5 07:31:22 2010
@@ -21,12 +21,6 @@
   margin: 0px;
 }
 
-#menu ul.level-three li {
-    margin-left: 0px;
-    border: thin solid red;
-}
-
-
 @media print 
 {
   .index-controls { display: none;}
@@ -73,8 +67,8 @@
   display: block;
   border: 0;
   padding: 0.1em;
-  margin: 0 3em 0px 1.8em;
-  padding-left:1em;
+  margin: 0 3em 0 1em;
+  padding-left: 1em;
   color: #5E72A5;
   background-image: none;
   width: 10em !important;
@@ -83,10 +77,13 @@
   font-size: 95%;
 }
 
-#menu ul.level-three li
-{
-    margin: 0px;
-    padding: 0px;
+#menu ul.level-three li a {
+  margin-left: 1em;
+}
+
+#menu ul.level-three li {
+  margin: 0;
+  padding: 0;
 }
 
 form {

From python-checkins at python.org  Mon Jul  5 13:41:43 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Mon,  5 Jul 2010 13:41:43 +0200 (CEST)
Subject: [Python-checkins] r82570 -
	python/branches/py3k/Doc/library/stdtypes.rst
Message-ID: <20100705114143.25BBFEE99A@mail.python.org>

Author: senthil.kumaran
Date: Mon Jul  5 13:41:42 2010
New Revision: 82570

Log:
Fix: Issue9091 Minor documentation clarification.



Modified:
   python/branches/py3k/Doc/library/stdtypes.rst

Modified: python/branches/py3k/Doc/library/stdtypes.rst
==============================================================================
--- python/branches/py3k/Doc/library/stdtypes.rst	(original)
+++ python/branches/py3k/Doc/library/stdtypes.rst	Mon Jul  5 13:41:42 2010
@@ -966,7 +966,8 @@
 
 .. method:: str.capitalize()
 
-   Return a copy of the string with only its first character capitalized.
+   Return a copy of the string with its first character capitalized and the
+   rest lowered.
 
 
 .. method:: str.center(width[, fillchar])

From python-checkins at python.org  Mon Jul  5 13:44:34 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Mon,  5 Jul 2010 13:44:34 +0200 (CEST)
Subject: [Python-checkins] r82571 - in python/branches/release27-maint:
	Doc/library/stdtypes.rst
Message-ID: <20100705114434.41A90EE99A@mail.python.org>

Author: senthil.kumaran
Date: Mon Jul  5 13:44:34 2010
New Revision: 82571

Log:
Merged revisions 82570 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82570 | senthil.kumaran | 2010-07-05 17:11:42 +0530 (Mon, 05 Jul 2010) | 3 lines
  
  Fix: Issue9091 Minor documentation clarification.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Doc/library/stdtypes.rst

Modified: python/branches/release27-maint/Doc/library/stdtypes.rst
==============================================================================
--- python/branches/release27-maint/Doc/library/stdtypes.rst	(original)
+++ python/branches/release27-maint/Doc/library/stdtypes.rst	Mon Jul  5 13:44:34 2010
@@ -830,7 +830,8 @@
 
 .. method:: str.capitalize()
 
-   Return a copy of the string with only its first character capitalized.
+   Return a copy of the string with its first character capitalized and the
+   rest lowered.
 
    For 8-bit strings, this method is locale-dependent.
 

From python-checkins at python.org  Mon Jul  5 13:44:49 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Mon,  5 Jul 2010 13:44:49 +0200 (CEST)
Subject: [Python-checkins] r82572 - in python/branches/release31-maint:
	Doc/library/stdtypes.rst
Message-ID: <20100705114449.317D5EE9D9@mail.python.org>

Author: senthil.kumaran
Date: Mon Jul  5 13:44:49 2010
New Revision: 82572

Log:
Merged revisions 82570 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82570 | senthil.kumaran | 2010-07-05 17:11:42 +0530 (Mon, 05 Jul 2010) | 3 lines
  
  Fix: Issue9091 Minor documentation clarification.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Doc/library/stdtypes.rst

Modified: python/branches/release31-maint/Doc/library/stdtypes.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/stdtypes.rst	(original)
+++ python/branches/release31-maint/Doc/library/stdtypes.rst	Mon Jul  5 13:44:49 2010
@@ -800,7 +800,8 @@
 
 .. method:: str.capitalize()
 
-   Return a copy of the string with only its first character capitalized.
+   Return a copy of the string with its first character capitalized and the
+   rest lowered.
 
 
 .. method:: str.center(width[, fillchar])

From python-checkins at python.org  Mon Jul  5 14:00:56 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Mon,  5 Jul 2010 14:00:56 +0200 (CEST)
Subject: [Python-checkins] r82573 - in python/branches/py3k/Objects:
	bytes_methods.c unicodeobject.c
Message-ID: <20100705120056.9D7C9EE996@mail.python.org>

Author: senthil.kumaran
Date: Mon Jul  5 14:00:56 2010
New Revision: 82573

Log:
Fix the docstrings of the capitalize method.



Modified:
   python/branches/py3k/Objects/bytes_methods.c
   python/branches/py3k/Objects/unicodeobject.c

Modified: python/branches/py3k/Objects/bytes_methods.c
==============================================================================
--- python/branches/py3k/Objects/bytes_methods.c	(original)
+++ python/branches/py3k/Objects/bytes_methods.c	Mon Jul  5 14:00:56 2010
@@ -332,7 +332,8 @@
 PyDoc_STRVAR_shared(_Py_capitalize__doc__,
 "B.capitalize() -> copy of B\n\
 \n\
-Return a copy of B with only its first character capitalized (ASCII).");
+Return a copy of B with only its first character capitalized (ASCII)\n\
+and the rest lower-cased.");
 
 void
 _Py_bytes_capitalize(char *result, char *s, Py_ssize_t len)

Modified: python/branches/py3k/Objects/unicodeobject.c
==============================================================================
--- python/branches/py3k/Objects/unicodeobject.c	(original)
+++ python/branches/py3k/Objects/unicodeobject.c	Mon Jul  5 14:00:56 2010
@@ -6705,7 +6705,7 @@
              "S.capitalize() -> str\n\
 \n\
 Return a capitalized version of S, i.e. make the first character\n\
-have upper case.");
+have upper case and the rest lower case.");
 
 static PyObject*
 unicode_capitalize(PyUnicodeObject *self)

From python-checkins at python.org  Mon Jul  5 14:04:08 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Mon,  5 Jul 2010 14:04:08 +0200 (CEST)
Subject: [Python-checkins] r82574 - in python/branches/release27-maint:
	Objects/bytes_methods.c Objects/unicodeobject.c
Message-ID: <20100705120408.1C854EE99A@mail.python.org>

Author: senthil.kumaran
Date: Mon Jul  5 14:04:07 2010
New Revision: 82574

Log:
Merged revisions 82573 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82573 | senthil.kumaran | 2010-07-05 17:30:56 +0530 (Mon, 05 Jul 2010) | 3 lines
  
  Fix the docstrings of the capitalize method.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Objects/bytes_methods.c
   python/branches/release27-maint/Objects/unicodeobject.c

Modified: python/branches/release27-maint/Objects/bytes_methods.c
==============================================================================
--- python/branches/release27-maint/Objects/bytes_methods.c	(original)
+++ python/branches/release27-maint/Objects/bytes_methods.c	Mon Jul  5 14:04:07 2010
@@ -332,7 +332,8 @@
 PyDoc_STRVAR_shared(_Py_capitalize__doc__,
 "B.capitalize() -> copy of B\n\
 \n\
-Return a copy of B with only its first character capitalized (ASCII).");
+Return a copy of B with only its first character capitalized (ASCII)\n\
+and the rest lower-cased.");
 
 void
 _Py_bytes_capitalize(char *result, char *s, Py_ssize_t len)

Modified: python/branches/release27-maint/Objects/unicodeobject.c
==============================================================================
--- python/branches/release27-maint/Objects/unicodeobject.c	(original)
+++ python/branches/release27-maint/Objects/unicodeobject.c	Mon Jul  5 14:04:07 2010
@@ -5938,7 +5938,7 @@
              "S.capitalize() -> unicode\n\
 \n\
 Return a capitalized version of S, i.e. make the first character\n\
-have upper case.");
+have upper case and the rest lower case.");
 
 static PyObject*
 unicode_capitalize(PyUnicodeObject *self)

From python-checkins at python.org  Mon Jul  5 14:04:24 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Mon,  5 Jul 2010 14:04:24 +0200 (CEST)
Subject: [Python-checkins] r82575 - in python/branches/release31-maint:
	Objects/bytes_methods.c Objects/unicodeobject.c
Message-ID: <20100705120424.2673FF4B4@mail.python.org>

Author: senthil.kumaran
Date: Mon Jul  5 14:04:23 2010
New Revision: 82575

Log:
Merged revisions 82573 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82573 | senthil.kumaran | 2010-07-05 17:30:56 +0530 (Mon, 05 Jul 2010) | 3 lines
  
  Fix the docstrings of the capitalize method.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Objects/bytes_methods.c
   python/branches/release31-maint/Objects/unicodeobject.c

Modified: python/branches/release31-maint/Objects/bytes_methods.c
==============================================================================
--- python/branches/release31-maint/Objects/bytes_methods.c	(original)
+++ python/branches/release31-maint/Objects/bytes_methods.c	Mon Jul  5 14:04:23 2010
@@ -332,7 +332,8 @@
 PyDoc_STRVAR_shared(_Py_capitalize__doc__,
 "B.capitalize() -> copy of B\n\
 \n\
-Return a copy of B with only its first character capitalized (ASCII).");
+Return a copy of B with only its first character capitalized (ASCII)\n\
+and the rest lower-cased.");
 
 void
 _Py_bytes_capitalize(char *result, char *s, Py_ssize_t len)

Modified: python/branches/release31-maint/Objects/unicodeobject.c
==============================================================================
--- python/branches/release31-maint/Objects/unicodeobject.c	(original)
+++ python/branches/release31-maint/Objects/unicodeobject.c	Mon Jul  5 14:04:23 2010
@@ -6766,7 +6766,7 @@
              "S.capitalize() -> str\n\
 \n\
 Return a capitalized version of S, i.e. make the first character\n\
-have upper case.");
+have upper case and the rest lower case.");
 
 static PyObject*
 unicode_capitalize(PyUnicodeObject *self)

From python-checkins at python.org  Mon Jul  5 16:59:41 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Mon,  5 Jul 2010 16:59:41 +0200 (CEST)
Subject: [Python-checkins] r82576 - python/branches/py3k/Lib/test/test_imp.py
Message-ID: <20100705145941.1F407EBB7@mail.python.org>

Author: benjamin.peterson
Date: Mon Jul  5 16:59:40 2010
New Revision: 82576

Log:
pydoc still has a silly encoding

Modified:
   python/branches/py3k/Lib/test/test_imp.py

Modified: python/branches/py3k/Lib/test/test_imp.py
==============================================================================
--- python/branches/py3k/Lib/test/test_imp.py	(original)
+++ python/branches/py3k/Lib/test/test_imp.py	Mon Jul  5 16:59:40 2010
@@ -44,7 +44,7 @@
 class ImportTests(unittest.TestCase):
 
     def test_find_module_encoding(self):
-        fd = imp.find_module("heapq")[0]
+        fd = imp.find_module("pydoc")[0]
         self.assertEqual(fd.encoding, "iso-8859-1")
 
     def test_issue1267(self):

From python-checkins at python.org  Mon Jul  5 17:01:22 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Mon,  5 Jul 2010 17:01:22 +0200 (CEST)
Subject: [Python-checkins] r82577 - python/branches/py3k/Objects/typeobject.c
Message-ID: <20100705150122.A246AF061@mail.python.org>

Author: benjamin.peterson
Date: Mon Jul  5 17:01:22 2010
New Revision: 82577

Log:
cleanup basicsize logic #3268

Modified:
   python/branches/py3k/Objects/typeobject.c

Modified: python/branches/py3k/Objects/typeobject.c
==============================================================================
--- python/branches/py3k/Objects/typeobject.c	(original)
+++ python/branches/py3k/Objects/typeobject.c	Mon Jul  5 17:01:22 2010
@@ -3476,11 +3476,8 @@
 static void
 inherit_special(PyTypeObject *type, PyTypeObject *base)
 {
-    Py_ssize_t oldsize, newsize;
 
     /* Copying basicsize is connected to the GC flags */
-    oldsize = base->tp_basicsize;
-    newsize = type->tp_basicsize ? type->tp_basicsize : oldsize;
     if (!(type->tp_flags & Py_TPFLAGS_HAVE_GC) &&
         (base->tp_flags & Py_TPFLAGS_HAVE_GC) &&
         (!type->tp_traverse && !type->tp_clear)) {
@@ -3507,7 +3504,8 @@
                 type->tp_new = base->tp_new;
         }
     }
-    type->tp_basicsize = newsize;
+    if (type->tp_basicsize == 0)
+        type->tp_basicsize = base->tp_basicsize;
 
     /* Copy other non-function slots */
 

From python-checkins at python.org  Mon Jul  5 17:05:34 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Mon,  5 Jul 2010 17:05:34 +0200 (CEST)
Subject: [Python-checkins] r82578 -
	python/branches/py3k/Lib/test/test_datetime.py
Message-ID: <20100705150534.05EABF134@mail.python.org>

Author: alexander.belopolsky
Date: Mon Jul  5 17:05:33 2010
New Revision: 82578

Log:
Added more tests for utctimetuple()

Modified:
   python/branches/py3k/Lib/test/test_datetime.py

Modified: python/branches/py3k/Lib/test/test_datetime.py
==============================================================================
--- python/branches/py3k/Lib/test/test_datetime.py	(original)
+++ python/branches/py3k/Lib/test/test_datetime.py	Mon Jul  5 17:05:33 2010
@@ -3008,7 +3008,7 @@
 
     def test_utctimetuple(self):
         class DST(tzinfo):
-            def __init__(self, dstvalue):
+            def __init__(self, dstvalue=0):
                 if isinstance(dstvalue, int):
                     dstvalue = timedelta(minutes=dstvalue)
                 self.dstvalue = dstvalue
@@ -3043,6 +3043,26 @@
             # is never in effect for a UTC time.
             self.assertEqual(0, t.tm_isdst)
 
+        # For naive datetime, utctimetuple == timetuple except for isdst
+        d = cls(1, 2, 3, 10, 20, 30, 40)
+        t = d.utctimetuple()
+        self.assertEqual(t[:-1], d.timetuple()[:-1])
+        self.assertEqual(0, t.tm_isdst)
+        # Same if utcoffset is None
+        class NOFS(DST):
+            def utcoffset(self, dt):
+                return None
+        d = cls(1, 2, 3, 10, 20, 30, 40, tzinfo=NOFS())
+        t = d.utctimetuple()
+        self.assertEqual(t[:-1], d.timetuple()[:-1])
+        self.assertEqual(0, t.tm_isdst)
+        # Check that bad tzinfo is detected
+        class BOFS(DST):
+            def utcoffset(self, dt):
+                return "EST"
+        d = cls(1, 2, 3, 10, 20, 30, 40, tzinfo=BOFS())
+        self.assertRaises(TypeError, d.utctimetuple)
+
         # Check that utctimetuple() is the same as
         # astimezone(utc).timetuple()
         d = cls(2010, 11, 13, 14, 15, 16, 171819)

From python-checkins at python.org  Mon Jul  5 17:26:36 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Mon,  5 Jul 2010 17:26:36 +0200 (CEST)
Subject: [Python-checkins] r82579 - in python/branches/release31-maint:
	Lib/test/test_datetime.py
Message-ID: <20100705152636.89BE1EC35@mail.python.org>

Author: alexander.belopolsky
Date: Mon Jul  5 17:26:36 2010
New Revision: 82579

Log:
Merged revisions 82578 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82578 | alexander.belopolsky | 2010-07-05 11:05:33 -0400 (Mon, 05 Jul 2010) | 1 line
  
  Added more tests for utctimetuple()
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/test/test_datetime.py

Modified: python/branches/release31-maint/Lib/test/test_datetime.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_datetime.py	(original)
+++ python/branches/release31-maint/Lib/test/test_datetime.py	Mon Jul  5 17:26:36 2010
@@ -2738,7 +2738,7 @@
 
     def test_utctimetuple(self):
         class DST(tzinfo):
-            def __init__(self, dstvalue):
+            def __init__(self, dstvalue=0):
                 if isinstance(dstvalue, int):
                     dstvalue = timedelta(minutes=dstvalue)
                 self.dstvalue = dstvalue
@@ -2772,6 +2772,25 @@
             self.assertEqual(d.toordinal() - date(1, 1, 1).toordinal() + 1,
                              t.tm_yday)
             self.assertEqual(0, t.tm_isdst)
+        # For naive datetime, utctimetuple == timetuple except for isdst
+        d = cls(1, 2, 3, 10, 20, 30, 40)
+        t = d.utctimetuple()
+        self.assertEqual(t[:-1], d.timetuple()[:-1])
+        self.assertEqual(0, t.tm_isdst)
+        # Same if utcoffset is None
+        class NOFS(DST):
+            def utcoffset(self, dt):
+                return None
+        d = cls(1, 2, 3, 10, 20, 30, 40, tzinfo=NOFS())
+        t = d.utctimetuple()
+        self.assertEqual(t[:-1], d.timetuple()[:-1])
+        self.assertEqual(0, t.tm_isdst)
+        # Check that bad tzinfo is detected
+        class BOFS(DST):
+            def utcoffset(self, dt):
+                return "EST"
+        d = cls(1, 2, 3, 10, 20, 30, 40, tzinfo=BOFS())
+        self.assertRaises(TypeError, d.utctimetuple)
 
         # At the edges, UTC adjustment can normalize into years out-of-range
         # for a datetime object.  Ensure that a correct timetuple is

From python-checkins at python.org  Mon Jul  5 18:06:06 2010
From: python-checkins at python.org (r.david.murray)
Date: Mon,  5 Jul 2010 18:06:06 +0200 (CEST)
Subject: [Python-checkins] r82581 - in python/branches/release27-maint:
	Lib/optparse.py Misc/NEWS
Message-ID: <20100705160606.14CE1F151@mail.python.org>

Author: r.david.murray
Date: Mon Jul  5 18:06:05 2010
New Revision: 82581

Log:
#9161: Fix regression in acceptance of unicode strings in add_option calls.

The original change in 2.6 was made during rc1 changeover, and did not
get ported to 2.7.  The original change may not even have been
intentional, but if so it doesn't seem to have caused any problems.


Modified:
   python/branches/release27-maint/Lib/optparse.py
   python/branches/release27-maint/Misc/NEWS

Modified: python/branches/release27-maint/Lib/optparse.py
==============================================================================
--- python/branches/release27-maint/Lib/optparse.py	(original)
+++ python/branches/release27-maint/Lib/optparse.py	Mon Jul  5 18:06:05 2010
@@ -1008,7 +1008,7 @@
         """add_option(Option)
            add_option(opt_str, ..., kwarg=val, ...)
         """
-        if type(args[0]) is types.StringType:
+        if type(args[0]) in types.StringTypes:
             option = self.option_class(*args, **kwargs)
         elif len(args) == 1 and not kwargs:
             option = args[0]

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Mon Jul  5 18:06:05 2010
@@ -16,6 +16,9 @@
 Library
 -------
 
+- Issue #9161: Fix regression in optparse's acceptance of unicode
+  strings in add_option calls.
+
 - Issue #9130: Fix validation of relative imports in parser module.
 
 - Issue #9128: Fix validation of class decorators in parser module.

From python-checkins at python.org  Mon Jul  5 18:50:01 2010
From: python-checkins at python.org (georg.brandl)
Date: Mon,  5 Jul 2010 18:50:01 +0200 (CEST)
Subject: [Python-checkins] r82584 - peps/trunk/pep-0392.txt
Message-ID: <20100705165001.CA467EDD9@mail.python.org>

Author: georg.brandl
Date: Mon Jul  5 18:50:01 2010
New Revision: 82584

Log:
Giving back the time machine.

Modified:
   peps/trunk/pep-0392.txt

Modified: peps/trunk/pep-0392.txt
==============================================================================
--- peps/trunk/pep-0392.txt	(original)
+++ peps/trunk/pep-0392.txt	Mon Jul  5 18:50:01 2010
@@ -43,8 +43,8 @@
 
 - 3.2 beta 2: November 27, 2010
 - 3.2 candidate 1: December 11, 2010
-- 3.2 candidate 2: January 1, 2010
-- 3.2 final: January 15, 2010
+- 3.2 candidate 2: January 1, 2011
+- 3.2 final: January 15, 2011
 
 .. don't forget to update final date above as well
 

From python-checkins at python.org  Mon Jul  5 19:11:05 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Mon,  5 Jul 2010 19:11:05 +0200 (CEST)
Subject: [Python-checkins] r82585 -
	python/branches/py3k/Lib/test/test_math.py
Message-ID: <20100705171105.485DDCA1C@mail.python.org>

Author: benjamin.peterson
Date: Mon Jul  5 19:11:05 2010
New Revision: 82585

Log:
looking up on the type is correct, so this isn't an XXX

Modified:
   python/branches/py3k/Lib/test/test_math.py

Modified: python/branches/py3k/Lib/test/test_math.py
==============================================================================
--- python/branches/py3k/Lib/test/test_math.py	(original)
+++ python/branches/py3k/Lib/test/test_math.py	Mon Jul  5 19:11:05 2010
@@ -923,13 +923,6 @@
         self.assertRaises(TypeError, math.trunc, 1, 2)
         self.assertRaises(TypeError, math.trunc, TestNoTrunc())
 
-        # XXX Doesn't work because the method is looked up on
-        #     the type only.
-        #t = TestNoTrunc()
-        #t.__trunc__ = lambda *args: args
-        #self.assertEquals((), math.trunc(t))
-        #self.assertRaises(TypeError, math.trunc, t, 0)
-
     def testIsnan(self):
         self.assertTrue(math.isnan(float("nan")))
         self.assertTrue(math.isnan(float("inf")* 0.))

From python-checkins at python.org  Mon Jul  5 19:13:21 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Mon,  5 Jul 2010 19:13:21 +0200 (CEST)
Subject: [Python-checkins] r82586 -
	python/branches/release27-maint/Lib/test/test_math.py
Message-ID: <20100705171321.B6B67CA1C@mail.python.org>

Author: benjamin.peterson
Date: Mon Jul  5 19:13:21 2010
New Revision: 82586

Log:
be more generous to implementations that have implemented correctly

Modified:
   python/branches/release27-maint/Lib/test/test_math.py

Modified: python/branches/release27-maint/Lib/test/test_math.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_math.py	(original)
+++ python/branches/release27-maint/Lib/test/test_math.py	Mon Jul  5 19:13:21 2010
@@ -876,8 +876,8 @@
 
         self.assertRaises(TypeError, math.trunc)
         self.assertRaises(TypeError, math.trunc, 1, 2)
-        # XXX: This is not ideal, but see the comment in math_trunc().
-        self.assertRaises(AttributeError, math.trunc, TestNoTrunc())
+        self.assertRaises((AttributeError, TypeError), math.trunc,
+                          TestNoTrunc())
 
         t = TestNoTrunc()
         t.__trunc__ = lambda *args: args

From python-checkins at python.org  Mon Jul  5 19:48:38 2010
From: python-checkins at python.org (georg.brandl)
Date: Mon,  5 Jul 2010 19:48:38 +0200 (CEST)
Subject: [Python-checkins] r82587 - python/branches/py3k/Doc/whatsnew/2.7.rst
Message-ID: <20100705174838.986FEEDD9@mail.python.org>

Author: georg.brandl
Date: Mon Jul  5 19:48:38 2010
New Revision: 82587

Log:
#9166: NotImplemented is not an exception.

Modified:
   python/branches/py3k/Doc/whatsnew/2.7.rst

Modified: python/branches/py3k/Doc/whatsnew/2.7.rst
==============================================================================
--- python/branches/py3k/Doc/whatsnew/2.7.rst	(original)
+++ python/branches/py3k/Doc/whatsnew/2.7.rst	Mon Jul  5 19:48:38 2010
@@ -1072,7 +1072,7 @@
   (Added by Raymond Hettinger; :issue:`1818`.)
 
   Finally, the :class:`~collections.Mapping` abstract base class now
-  raises a :exc:`NotImplemented` exception if a mapping is compared to
+  returns :const:`NotImplemented` if a mapping is compared to
   another type that isn't a :class:`Mapping`.
   (Fixed by Daniel Stutzbach; :issue:`8729`.)
 

From python-checkins at python.org  Mon Jul  5 19:50:56 2010
From: python-checkins at python.org (georg.brandl)
Date: Mon,  5 Jul 2010 19:50:56 +0200 (CEST)
Subject: [Python-checkins] r82588 - in python/branches/release27-maint:
	Doc/whatsnew/2.7.rst
Message-ID: <20100705175056.1EB50EC40@mail.python.org>

Author: georg.brandl
Date: Mon Jul  5 19:50:55 2010
New Revision: 82588

Log:
Merged revisions 82587 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82587 | georg.brandl | 2010-07-05 19:48:38 +0200 (Mo, 05 Jul 2010) | 1 line
  
  #9166: NotImplemented is not an exception.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Doc/whatsnew/2.7.rst

Modified: python/branches/release27-maint/Doc/whatsnew/2.7.rst
==============================================================================
--- python/branches/release27-maint/Doc/whatsnew/2.7.rst	(original)
+++ python/branches/release27-maint/Doc/whatsnew/2.7.rst	Mon Jul  5 19:50:55 2010
@@ -1073,7 +1073,7 @@
   (Added by Raymond Hettinger; :issue:`1818`.)
 
   Finally, the :class:`~collections.Mapping` abstract base class now
-  raises a :exc:`NotImplemented` exception if a mapping is compared to
+  returns :const:`NotImplemented` if a mapping is compared to
   another type that isn't a :class:`Mapping`.
   (Fixed by Daniel Stutzbach; :issue:`8729`.)
 

From python-checkins at python.org  Mon Jul  5 19:57:31 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Mon,  5 Jul 2010 19:57:31 +0200 (CEST)
Subject: [Python-checkins] r82589 -
	python/branches/py3k/Lib/test/test_datetime.py
Message-ID: <20100705175731.774D6F197@mail.python.org>

Author: alexander.belopolsky
Date: Mon Jul  5 19:57:31 2010
New Revision: 82589

Log:
Added two more test cases for datetime

Modified:
   python/branches/py3k/Lib/test/test_datetime.py

Modified: python/branches/py3k/Lib/test/test_datetime.py
==============================================================================
--- python/branches/py3k/Lib/test/test_datetime.py	(original)
+++ python/branches/py3k/Lib/test/test_datetime.py	Mon Jul  5 19:57:31 2010
@@ -881,6 +881,7 @@
     def test_computations(self):
         a = self.theclass(2002, 1, 31)
         b = self.theclass(1956, 1, 31)
+        c = self.theclass(2001,2,1)
 
         diff = a-b
         self.assertEqual(diff.days, 46*365 + len(range(1956, 2002, 4)))
@@ -906,6 +907,7 @@
         self.assertEqual(a - (a + day), -day)
         self.assertEqual(a - (a - week), week)
         self.assertEqual(a - (a - day), day)
+        self.assertEqual(c - (c - day), day)
 
         # Add/sub ints or floats should be illegal
         for i in 1, 1.0:
@@ -1347,6 +1349,10 @@
         for month_byte in b'9', b'\0', b'\r', b'\xff':
             self.assertRaises(TypeError, self.theclass,
                                          base[:2] + month_byte + base[3:])
+        # Good bytes, but bad tzinfo:
+        self.assertRaises(TypeError, self.theclass,
+                          bytes([1] * len(base)), 'EST')
+
         for ord_byte in range(1, 13):
             # This shouldn't blow up because of the month byte alone.  If
             # the implementation changes to do more-careful checking, it may

From brett at python.org  Mon Jul  5 21:22:19 2010
From: brett at python.org (Brett Cannon)
Date: Mon, 5 Jul 2010 12:22:19 -0700
Subject: [Python-checkins] r82562 - in sandbox/trunk/pep362: README
	setup.py tests/test_pep362.py
In-Reply-To: <4C314757.2060505@gmail.com>
References: <20100704205218.0EEF4F68C9@mail.python.org>
	<4C314757.2060505@gmail.com>
Message-ID: <AANLkTimJQAmHcB0or0CuTsRG6XHoophzS3tAt1tLOosp@mail.gmail.com>

I decided to do a find-and-replace rather than trying to "fix" the
method calls with a more proper one. It was just faster and easier.

On Sun, Jul 4, 2010 at 19:45, Ezio Melotti <ezio.melotti at gmail.com> wrote:
> On 04/07/2010 23.52, brett.cannon wrote:
>>
>> Author: brett.cannon
>> Date: Sun Jul ?4 22:52:17 2010
>> New Revision: 82562
>>
>> Log:
>> Update to run under Python 2.7 and 3.1 without any warnings.
>>
>> Modified:
>> ? ?sandbox/trunk/pep362/README
>> ? ?sandbox/trunk/pep362/setup.py
>> ? ?sandbox/trunk/pep362/tests/test_pep362.py
>>
>> Modified: sandbox/trunk/pep362/README
>>
>> ==============================================================================
>> --- sandbox/trunk/pep362/README (original)
>> +++ sandbox/trunk/pep362/README Sun Jul ?4 22:52:17 2010
>> @@ -38,6 +38,11 @@
>>
>> ?The version number will get bumped as open issues in the PEP are closed.
>>
>> +0.6.2
>> +-----
>> +
>> +* Move test suite over to assert* methods.
>> +
>> ?0.6
>> ?---
>>
>>
>> Modified: sandbox/trunk/pep362/setup.py
>>
>> ==============================================================================
>> --- sandbox/trunk/pep362/setup.py ? ? ? (original)
>> +++ sandbox/trunk/pep362/setup.py ? ? ? Sun Jul ?4 22:52:17 2010
>> @@ -3,7 +3,7 @@
>> [...]
>> ? ? ?@version_specific(2)
>> ? ? ?def test_parameter_tuple(self):
>> ? ? ? ? ?# A function with a tuple as a parameter should work.
>> ? ? ? ? ?sig = pep362.Signature(pep362_py2_fodder.tuple_args)
>> - ? ? ? ?self.failUnlessEqual('tuple_args', sig.name)
>> + ? ? ? ?self.assertEqual('tuple_args', sig.name)
>> ? ? ? ? ?param = list(sig)[0]
>> - ? ? ? ?self.failUnless(isinstance(param.name, tuple))
>> - ? ? ? ?self.failUnlessEqual(('a', ('b',)), param.name)
>> - ? ? ? ?self.failUnlessEqual(0, param.position)
>> - ? ? ? ?self.failUnless(not hasattr(param, 'default_value'))
>> + ? ? ? ?self.assertTrue(isinstance(param.name, tuple))
>
> This could be an assertIsInstance (unless you need to support Python <=2.6).
>
>> + ? ? ? ?self.assertEqual(('a', ('b',)), param.name)
>> + ? ? ? ?self.assertEqual(0, param.position)
>> + ? ? ? ?self.assertTrue(not hasattr(param, 'default_value'))
>
> This and several others could use assertFalse, instead of assertTrue(not
> ...)
>
>>
>> [...]
>
> Best Regards,
> Ezio Melotti
> _______________________________________________
> Python-checkins mailing list
> Python-checkins at python.org
> http://mail.python.org/mailman/listinfo/python-checkins
>

From python-checkins at python.org  Mon Jul  5 22:04:55 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Mon,  5 Jul 2010 22:04:55 +0200 (CEST)
Subject: [Python-checkins] r82591 - python/branches/py3k/Grammar/Grammar
Message-ID: <20100705200455.163ABF4D4@mail.python.org>

Author: benjamin.peterson
Date: Mon Jul  5 22:04:54 2010
New Revision: 82591

Log:
untabify

Modified:
   python/branches/py3k/Grammar/Grammar

Modified: python/branches/py3k/Grammar/Grammar
==============================================================================
--- python/branches/py3k/Grammar/Grammar	(original)
+++ python/branches/py3k/Grammar/Grammar	Mon Jul  5 22:04:54 2010
@@ -11,9 +11,9 @@
 # "How to Change Python's Grammar"
 
 # Start symbols for the grammar:
-#	single_input is a single interactive statement;
-#	file_input is a module or sequence of commands read from an input file;
-#	eval_input is the input for the eval() and input() functions.
+#       single_input is a single interactive statement;
+#       file_input is a module or sequence of commands read from an input file;
+#       eval_input is the input for the eval() and input() functions.
 # NB: compound_stmt in single_input is followed by extra NEWLINE!
 single_input: NEWLINE | simple_stmt | compound_stmt NEWLINE
 file_input: (NEWLINE | stmt)* ENDMARKER
@@ -71,9 +71,9 @@
 for_stmt: 'for' exprlist 'in' testlist ':' suite ['else' ':' suite]
 try_stmt: ('try' ':' suite
            ((except_clause ':' suite)+
-	    ['else' ':' suite]
-	    ['finally' ':' suite] |
-	   'finally' ':' suite))
+            ['else' ':' suite]
+            ['finally' ':' suite] |
+           'finally' ':' suite))
 with_stmt: 'with' with_item (',' with_item)*  ':' suite
 with_item: test ['as' expr]
 # NB compile.c makes sure that the default except clause is last

From python-checkins at python.org  Mon Jul  5 22:05:43 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Mon,  5 Jul 2010 22:05:43 +0200 (CEST)
Subject: [Python-checkins] r82592 -
	python/branches/release27-maint/Grammar/Grammar
Message-ID: <20100705200543.89E15F41E@mail.python.org>

Author: benjamin.peterson
Date: Mon Jul  5 22:05:43 2010
New Revision: 82592

Log:
untabify

Modified:
   python/branches/release27-maint/Grammar/Grammar

Modified: python/branches/release27-maint/Grammar/Grammar
==============================================================================
--- python/branches/release27-maint/Grammar/Grammar	(original)
+++ python/branches/release27-maint/Grammar/Grammar	Mon Jul  5 22:05:43 2010
@@ -11,9 +11,9 @@
 # "How to Change Python's Grammar"
 
 # Start symbols for the grammar:
-#	single_input is a single interactive statement;
-#	file_input is a module or sequence of commands read from an input file;
-#	eval_input is the input for the eval() and input() functions.
+#       single_input is a single interactive statement;
+#       file_input is a module or sequence of commands read from an input file;
+#       eval_input is the input for the eval() and input() functions.
 # NB: compound_stmt in single_input is followed by extra NEWLINE!
 single_input: NEWLINE | simple_stmt | compound_stmt NEWLINE
 file_input: (NEWLINE | stmt)* ENDMARKER
@@ -68,9 +68,9 @@
 for_stmt: 'for' exprlist 'in' testlist ':' suite ['else' ':' suite]
 try_stmt: ('try' ':' suite
            ((except_clause ':' suite)+
-	    ['else' ':' suite]
-	    ['finally' ':' suite] |
-	   'finally' ':' suite))
+            ['else' ':' suite]
+            ['finally' ':' suite] |
+           'finally' ':' suite))
 with_stmt: 'with' with_item (',' with_item)*  ':' suite
 with_item: test ['as' expr]
 # NB compile.c makes sure that the default except clause is last

From python-checkins at python.org  Mon Jul  5 22:13:06 2010
From: python-checkins at python.org (georg.brandl)
Date: Mon,  5 Jul 2010 22:13:06 +0200 (CEST)
Subject: [Python-checkins] r82593 - python/branches/py3k/Demo/classes/README
Message-ID: <20100705201306.C40A8F4D4@mail.python.org>

Author: georg.brandl
Date: Mon Jul  5 22:13:06 2010
New Revision: 82593

Log:
Remove Dbm.

Modified:
   python/branches/py3k/Demo/classes/README

Modified: python/branches/py3k/Demo/classes/README
==============================================================================
--- python/branches/py3k/Demo/classes/README	(original)
+++ python/branches/py3k/Demo/classes/README	Mon Jul  5 22:13:06 2010
@@ -2,7 +2,6 @@
 
 Complex.py	Complex numbers
 Dates.py	Date manipulation package by Tim Peters
-Dbm.py		Wrapper around built-in dbm, supporting	arbitrary values
 Range.py	Example of a generator: re-implement built-in range()
 Rev.py		Yield the reverse of a sequence
 Vec.py		A simple vector class

From python-checkins at python.org  Mon Jul  5 22:13:42 2010
From: python-checkins at python.org (georg.brandl)
Date: Mon,  5 Jul 2010 22:13:42 +0200 (CEST)
Subject: [Python-checkins] r82594 - python/branches/py3k/Demo/classes/Vec.py
Message-ID: <20100705201342.0ABB3E795@mail.python.org>

Author: georg.brandl
Date: Mon Jul  5 22:13:41 2010
New Revision: 82594

Log:
Update Vec class constructor, remove indirection via function, use operator module.

Modified:
   python/branches/py3k/Demo/classes/Vec.py

Modified: python/branches/py3k/Demo/classes/Vec.py
==============================================================================
--- python/branches/py3k/Demo/classes/Vec.py	(original)
+++ python/branches/py3k/Demo/classes/Vec.py	Mon Jul  5 22:13:41 2010
@@ -1,8 +1,6 @@
 # A simple vector class
 
-
-def vec(*v):
-    return Vec(*v)
+import operator
 
 
 class Vec:
@@ -10,14 +8,16 @@
     def __init__(self, *v):
         self.v = list(v)
 
-    def fromlist(self, v):
+    @classmethod
+    def fromlist(cls, v):
         if not isinstance(v, list):
             raise TypeError
-        self.v = v[:]
-        return self
+        inst = cls()
+        inst.v = v
+        return inst
 
     def __repr__(self):
-        return 'vec(' + repr(self.v)[1:-1] + ')'
+        return 'Vec(' + repr(self.v)[1:-1] + ')'
 
     def __len__(self):
         return len(self.v)
@@ -27,24 +27,24 @@
 
     def __add__(self, other):
         # Element-wise addition
-        v = list(map(lambda x, y: x+y, self, other))
-        return Vec().fromlist(v)
+        v = list(map(operator.add, self, other))
+        return Vec.fromlist(v)
 
     def __sub__(self, other):
         # Element-wise subtraction
-        v = list(map(lambda x, y: x-y, self, other))
-        return Vec().fromlist(v)
+        v = list(map(operator.sub, self, other))
+        return Vec.fromlist(v)
 
     def __mul__(self, scalar):
         # Multiply by scalar
         v = [x*scalar for x in self.v]
-        return Vec().fromlist(v)
+        return Vec.fromlist(v)
 
 
 
 def test():
-    a = vec(1, 2, 3)
-    b = vec(3, 2, 1)
+    a = Vec(1, 2, 3)
+    b = Vec(3, 2, 1)
     print(a)
     print(b)
     print(a+b)

From python-checkins at python.org  Mon Jul  5 22:14:26 2010
From: python-checkins at python.org (mark.dickinson)
Date: Mon,  5 Jul 2010 22:14:26 +0200 (CEST)
Subject: [Python-checkins] r82595 - python/branches/py3k/Modules/_math.c
Message-ID: <20100705201426.90D11D319@mail.python.org>

Author: mark.dickinson
Date: Mon Jul  5 22:14:26 2010
New Revision: 82595

Log:
Post-detabification cleanup.

Modified:
   python/branches/py3k/Modules/_math.c

Modified: python/branches/py3k/Modules/_math.c
==============================================================================
--- python/branches/py3k/Modules/_math.c	(original)
+++ python/branches/py3k/Modules/_math.c	Mon Jul  5 22:14:26 2010
@@ -56,13 +56,13 @@
         if (Py_IS_INFINITY(x)) {
             return x+x;
         } else {
-            return log(x)+ln2;                  /* acosh(huge)=log(2x) */
+            return log(x)+ln2;          /* acosh(huge)=log(2x) */
         }
     }
     else if (x == 1.) {
-        return 0.0;                             /* acosh(1) = 0 */
+        return 0.0;                     /* acosh(1) = 0 */
     }
-    else if (x > 2.) {                          /* 2 < x < 2**28 */
+    else if (x > 2.) {                  /* 2 < x < 2**28 */
         double t = x*x;
         return log(2.0*x - 1.0 / (x + sqrt(t - 1.0)));
     }
@@ -94,7 +94,7 @@
         return x+x;
     }
     if (absx < two_pow_m28) {           /* |x| < 2**-28 */
-        return x;               /* return x inexact except 0 */
+        return x;                       /* return x inexact except 0 */
     }
     if (absx > two_pow_p28) {           /* |x| > 2**28 */
         w = log(absx)+ln2;
@@ -114,9 +114,9 @@
  * Method :
  *    1.Reduced x to positive by atanh(-x) = -atanh(x)
  *    2.For x>=0.5
- *                1           2x                          x
- *      atanh(x) = --- * log(1 + -------) = 0.5 * log1p(2 * --------)
- *                2          1 - x                    1 - x
+ *                  1              2x                          x
+ *      atanh(x) = --- * log(1 + -------) = 0.5 * log1p(2 * -------)
+ *                  2             1 - x                      1 - x
  *
  *      For x<0.5
  *      atanh(x) = 0.5*log1p(2x+2x*x/(1-x))
@@ -194,7 +194,7 @@
     /* For x small, we use the following approach.  Let y be the nearest float
        to 1+x, then
 
-      1+x = y * (1 - (y-1-x)/y)
+         1+x = y * (1 - (y-1-x)/y)
 
        so log(1+x) = log(y) + log(1-(y-1-x)/y).  Since (y-1-x)/y is tiny, the
        second term is well approximated by (y-1-x)/y.  If abs(x) >=

From python-checkins at python.org  Mon Jul  5 22:16:40 2010
From: python-checkins at python.org (mark.dickinson)
Date: Mon,  5 Jul 2010 22:16:40 +0200 (CEST)
Subject: [Python-checkins] r82596 - in python/branches/release27-maint:
	Modules/_math.c
Message-ID: <20100705201640.3FF96C81F@mail.python.org>

Author: mark.dickinson
Date: Mon Jul  5 22:16:40 2010
New Revision: 82596

Log:
Merged revisions 82595 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82595 | mark.dickinson | 2010-07-05 21:14:26 +0100 (Mon, 05 Jul 2010) | 1 line
  
  Post-detabification cleanup.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Modules/_math.c

Modified: python/branches/release27-maint/Modules/_math.c
==============================================================================
--- python/branches/release27-maint/Modules/_math.c	(original)
+++ python/branches/release27-maint/Modules/_math.c	Mon Jul  5 22:16:40 2010
@@ -56,13 +56,13 @@
         if (Py_IS_INFINITY(x)) {
             return x+x;
         } else {
-            return log(x)+ln2;                  /* acosh(huge)=log(2x) */
+            return log(x)+ln2;          /* acosh(huge)=log(2x) */
         }
     }
     else if (x == 1.) {
-        return 0.0;                             /* acosh(1) = 0 */
+        return 0.0;                     /* acosh(1) = 0 */
     }
-    else if (x > 2.) {                          /* 2 < x < 2**28 */
+    else if (x > 2.) {                  /* 2 < x < 2**28 */
         double t = x*x;
         return log(2.0*x - 1.0 / (x + sqrt(t - 1.0)));
     }
@@ -94,7 +94,7 @@
         return x+x;
     }
     if (absx < two_pow_m28) {           /* |x| < 2**-28 */
-        return x;               /* return x inexact except 0 */
+        return x;                       /* return x inexact except 0 */
     }
     if (absx > two_pow_p28) {           /* |x| > 2**28 */
         w = log(absx)+ln2;
@@ -114,9 +114,9 @@
  * Method :
  *    1.Reduced x to positive by atanh(-x) = -atanh(x)
  *    2.For x>=0.5
- *                1           2x                          x
- *      atanh(x) = --- * log(1 + -------) = 0.5 * log1p(2 * --------)
- *                2          1 - x                    1 - x
+ *                  1              2x                          x
+ *      atanh(x) = --- * log(1 + -------) = 0.5 * log1p(2 * -------)
+ *                  2             1 - x                      1 - x
  *
  *      For x<0.5
  *      atanh(x) = 0.5*log1p(2x+2x*x/(1-x))
@@ -194,7 +194,7 @@
     /* For x small, we use the following approach.  Let y be the nearest float
        to 1+x, then
 
-      1+x = y * (1 - (y-1-x)/y)
+         1+x = y * (1 - (y-1-x)/y)
 
        so log(1+x) = log(y) + log(1-(y-1-x)/y).  Since (y-1-x)/y is tiny, the
        second term is well approximated by (y-1-x)/y.  If abs(x) >=

From alexander.belopolsky at gmail.com  Mon Jul  5 22:42:16 2010
From: alexander.belopolsky at gmail.com (Alexander Belopolsky)
Date: Mon, 5 Jul 2010 16:42:16 -0400
Subject: [Python-checkins] r82594 -
	python/branches/py3k/Demo/classes/Vec.py
In-Reply-To: <20100705201342.0ABB3E795@mail.python.org>
References: <20100705201342.0ABB3E795@mail.python.org>
Message-ID: <AANLkTimnDaAC2psR9PqeuGpYA-ZBBc8zwEKc0SXps_3x@mail.gmail.com>

On Mon, Jul 5, 2010 at 4:13 PM, georg.brandl <python-checkins at python.org> wrote:
..
> ? ? ? ? # Element-wise addition
> - ? ? ? ?v = list(map(lambda x, y: x+y, self, other))
> - ? ? ? ?return Vec().fromlist(v)
> + ? ? ? ?v = list(map(operator.add, self, other))
> + ? ? ? ?return Vec.fromlist(v)

I wonder if it would be more fitting in a 3.x demo to compute v as

v = [x + y for x, y in zip(self.v, other.v)]

or provide __iter__ and do simply

v = [x + y for x, y in zip(self, other)]

I know this suggestion may be premature before it is decided what
stays in Demo and what does not, but +import operator has caught my
eye.

From g.brandl at gmx.net  Mon Jul  5 23:12:22 2010
From: g.brandl at gmx.net (Georg Brandl)
Date: Mon, 05 Jul 2010 23:12:22 +0200
Subject: [Python-checkins] r82570 -
	python/branches/py3k/Doc/library/stdtypes.rst
In-Reply-To: <20100705114143.25BBFEE99A@mail.python.org>
References: <20100705114143.25BBFEE99A@mail.python.org>
Message-ID: <i0thpi$8ng$1@dough.gmane.org>

Am 05.07.2010 13:41, schrieb senthil.kumaran:
> Author: senthil.kumaran
> Date: Mon Jul  5 13:41:42 2010
> New Revision: 82570
> 
> Log:
> Fix: Issue9091 Minor documentation clarification.
> 
> 
> 
> Modified:
>    python/branches/py3k/Doc/library/stdtypes.rst
> 
> Modified: python/branches/py3k/Doc/library/stdtypes.rst
> ==============================================================================
> --- python/branches/py3k/Doc/library/stdtypes.rst	(original)
> +++ python/branches/py3k/Doc/library/stdtypes.rst	Mon Jul  5 13:41:42 2010
> @@ -966,7 +966,8 @@
>  
>  .. method:: str.capitalize()
>  
> -   Return a copy of the string with only its first character capitalized.
> +   Return a copy of the string with its first character capitalized and the
> +   rest lowered.

That should be "lowercased" IINM.

Georg


From g.brandl at gmx.net  Mon Jul  5 23:14:49 2010
From: g.brandl at gmx.net (Georg Brandl)
Date: Mon, 05 Jul 2010 23:14:49 +0200
Subject: [Python-checkins] r82594 -
	python/branches/py3k/Demo/classes/Vec.py
In-Reply-To: <AANLkTimnDaAC2psR9PqeuGpYA-ZBBc8zwEKc0SXps_3x@mail.gmail.com>
References: <20100705201342.0ABB3E795@mail.python.org>
	<AANLkTimnDaAC2psR9PqeuGpYA-ZBBc8zwEKc0SXps_3x@mail.gmail.com>
Message-ID: <i0thu5$8ng$2@dough.gmane.org>

Am 05.07.2010 22:42, schrieb Alexander Belopolsky:
> On Mon, Jul 5, 2010 at 4:13 PM, georg.brandl <python-checkins at python.org> wrote:
> ..
>>         # Element-wise addition
>> -        v = list(map(lambda x, y: x+y, self, other))
>> -        return Vec().fromlist(v)
>> +        v = list(map(operator.add, self, other))
>> +        return Vec.fromlist(v)
> 
> I wonder if it would be more fitting in a 3.x demo to compute v as
> 
> v = [x + y for x, y in zip(self.v, other.v)]
> 
> or provide __iter__ and do simply
> 
> v = [x + y for x, y in zip(self, other)]
> 
> I know this suggestion may be premature before it is decided what
> stays in Demo and what does not, but +import operator has caught my
> eye.

Well, why not.  Go ahead.

Georg


From python-checkins at python.org  Mon Jul  5 23:36:21 2010
From: python-checkins at python.org (victor.stinner)
Date: Mon,  5 Jul 2010 23:36:21 +0200 (CEST)
Subject: [Python-checkins] r82597 - python/branches/py3k/Doc/c-api/arg.rst
Message-ID: <20100705213621.9047BEDB2@mail.python.org>

Author: victor.stinner
Date: Mon Jul  5 23:36:21 2010
New Revision: 82597

Log:
#9158: Fix y* format of PyArg_Parse*() functions documentation


Modified:
   python/branches/py3k/Doc/c-api/arg.rst

Modified: python/branches/py3k/Doc/c-api/arg.rst
==============================================================================
--- python/branches/py3k/Doc/c-api/arg.rst	(original)
+++ python/branches/py3k/Doc/c-api/arg.rst	Mon Jul  5 23:36:21 2010
@@ -100,7 +100,7 @@
    contain embedded NUL bytes; if it does, a :exc:`TypeError`
    exception is raised.
 
-``y*`` (:class:`bytes`, :class:`bytearray` or buffer compatible object) [Py_buffer \*]
+``y*`` (:class:`bytes`, :class:`bytearray` or buffer compatible object) [Py_buffer]
    This variant on ``s*`` doesn't accept Unicode objects, only objects
    supporting the buffer protocol.  **This is the recommended way to accept
    binary data.**

From python-checkins at python.org  Mon Jul  5 23:38:37 2010
From: python-checkins at python.org (victor.stinner)
Date: Mon,  5 Jul 2010 23:38:37 +0200 (CEST)
Subject: [Python-checkins] r82598 - in python/branches/release31-maint:
	Doc/c-api/arg.rst
Message-ID: <20100705213837.C98C4F68D1@mail.python.org>

Author: victor.stinner
Date: Mon Jul  5 23:38:37 2010
New Revision: 82598

Log:
Merged revisions 82597 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82597 | victor.stinner | 2010-07-05 23:36:21 +0200 (lun., 05 juil. 2010) | 2 lines
  
  #9158: Fix y* format of PyArg_Parse*() functions documentation
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Doc/c-api/arg.rst

Modified: python/branches/release31-maint/Doc/c-api/arg.rst
==============================================================================
--- python/branches/release31-maint/Doc/c-api/arg.rst	(original)
+++ python/branches/release31-maint/Doc/c-api/arg.rst	Mon Jul  5 23:38:37 2010
@@ -100,7 +100,7 @@
    contain embedded NUL bytes; if it does, a :exc:`TypeError`
    exception is raised.
 
-``y*`` (:class:`bytes`, :class:`bytearray` or buffer compatible object) [Py_buffer \*]
+``y*`` (:class:`bytes`, :class:`bytearray` or buffer compatible object) [Py_buffer]
    This variant on ``s*`` doesn't accept Unicode objects, only objects
    supporting the buffer protocol.  **This is the recommended way to accept
    binary data.**

From python-checkins at python.org  Mon Jul  5 23:44:06 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Mon,  5 Jul 2010 23:44:06 +0200 (CEST)
Subject: [Python-checkins] r82599 - python/branches/py3k/Demo/classes/Vec.py
Message-ID: <20100705214406.07607EDB2@mail.python.org>

Author: alexander.belopolsky
Date: Mon Jul  5 23:44:05 2010
New Revision: 82599

Log:
"Modernized" the demo a little.

Modified:
   python/branches/py3k/Demo/classes/Vec.py

Modified: python/branches/py3k/Demo/classes/Vec.py
==============================================================================
--- python/branches/py3k/Demo/classes/Vec.py	(original)
+++ python/branches/py3k/Demo/classes/Vec.py	Mon Jul  5 23:44:05 2010
@@ -1,10 +1,27 @@
-# A simple vector class
+class Vec:
+    """ A simple vector class
 
-import operator
+    Instances of the Vec class  can be constructed from numbers
 
+    >>> a = Vec(1, 2, 3)
+    >>> b = Vec(3, 2, 1)
 
-class Vec:
+    added
+    >>> a + b
+    Vec(4, 4, 4)
+
+    subtracted
+    >>> a - b
+    Vec(-2, 0, 2)
+
+    and multiplied by a scalar on the left
+    >>> 3.0 * a
+    Vec(3.0, 6.0, 9.0)
 
+    or on the right
+    >>> a * 3.0
+    Vec(3.0, 6.0, 9.0)
+    """
     def __init__(self, *v):
         self.v = list(v)
 
@@ -17,7 +34,8 @@
         return inst
 
     def __repr__(self):
-        return 'Vec(' + repr(self.v)[1:-1] + ')'
+        args = ', '.join(repr(x) for x in self.v)
+        return 'Vec({})'.format(args)
 
     def __len__(self):
         return len(self.v)
@@ -27,28 +45,24 @@
 
     def __add__(self, other):
         # Element-wise addition
-        v = list(map(operator.add, self, other))
+        v = [x + y for x, y in zip(self.v, other.v)]
         return Vec.fromlist(v)
 
     def __sub__(self, other):
         # Element-wise subtraction
-        v = list(map(operator.sub, self, other))
+        v = [x - y for x, y in zip(self.v, other.v)]
         return Vec.fromlist(v)
 
     def __mul__(self, scalar):
         # Multiply by scalar
-        v = [x*scalar for x in self.v]
+        v = [x * scalar for x in self.v]
         return Vec.fromlist(v)
 
+    __rmul__ = __mul__
 
 
 def test():
-    a = Vec(1, 2, 3)
-    b = Vec(3, 2, 1)
-    print(a)
-    print(b)
-    print(a+b)
-    print(a-b)
-    print(a*3.0)
+    import doctest
+    doctest.testmod()
 
 test()

From python-checkins at python.org  Tue Jul  6 00:11:16 2010
From: python-checkins at python.org (brett.cannon)
Date: Tue,  6 Jul 2010 00:11:16 +0200 (CEST)
Subject: [Python-checkins] r82600 -
	python/branches/py3k/Lib/test/test_xmlrpc_net.py
Message-ID: <20100705221116.E7A75D2D2@mail.python.org>

Author: brett.cannon
Date: Tue Jul  6 00:11:16 2010
New Revision: 82600

Log:
Fix test_xmlrpc_net to no longer fail since there are no more buildbots for trunk.

Modified:
   python/branches/py3k/Lib/test/test_xmlrpc_net.py

Modified: python/branches/py3k/Lib/test/test_xmlrpc_net.py
==============================================================================
--- python/branches/py3k/Lib/test/test_xmlrpc_net.py	(original)
+++ python/branches/py3k/Lib/test/test_xmlrpc_net.py	Tue Jul  6 00:11:16 2010
@@ -48,7 +48,7 @@
         # Perform a minimal sanity check on the result, just to be sure
         # the request means what we think it means.
         self.assertIsInstance(builders, collections.Sequence)
-        self.assertTrue([x for x in builders if "trunk" in x], builders)
+        self.assertTrue([x for x in builders if "3.x" in x], builders)
 
 
 def test_main():

From eric at trueblade.com  Tue Jul  6 01:12:31 2010
From: eric at trueblade.com (Eric Smith)
Date: Mon, 05 Jul 2010 19:12:31 -0400
Subject: [Python-checkins] r82570
	-	python/branches/py3k/Doc/library/stdtypes.rst
In-Reply-To: <i0thpi$8ng$1@dough.gmane.org>
References: <20100705114143.25BBFEE99A@mail.python.org>
	<i0thpi$8ng$1@dough.gmane.org>
Message-ID: <4C3266DF.7050907@trueblade.com>

On 7/5/10 5:12 PM, Georg Brandl wrote:
> Am 05.07.2010 13:41, schrieb senthil.kumaran:
>> Author: senthil.kumaran
>> Date: Mon Jul  5 13:41:42 2010
>> New Revision: 82570
>>
>> Log:
>> Fix: Issue9091 Minor documentation clarification.
>>
>>
>>
>> Modified:
>>     python/branches/py3k/Doc/library/stdtypes.rst
>>
>> Modified: python/branches/py3k/Doc/library/stdtypes.rst
>> ==============================================================================
>> --- python/branches/py3k/Doc/library/stdtypes.rst	(original)
>> +++ python/branches/py3k/Doc/library/stdtypes.rst	Mon Jul  5 13:41:42 2010
>> @@ -966,7 +966,8 @@
>>
>>   .. method:: str.capitalize()
>>
>> -   Return a copy of the string with only its first character capitalized.
>> +   Return a copy of the string with its first character capitalized and the
>> +   rest lowered.
>
> That should be "lowercased" IINM.

Yes, that would be a better word.

-- 
Eric.

From solipsis at pitrou.net  Tue Jul  6 01:25:03 2010
From: solipsis at pitrou.net (solipsis at pitrou.net)
Date: Tue,  6 Jul 2010 01:25:03 +0200 (CEST)
Subject: [Python-checkins] Daily py3k reference leaks (r82595): sum=0
Message-ID: <20100705232503.4021C17721@ns6635.ovh.net>

py3k results for svn r82595 (hg cset 43bb2c640f17)
--------------------------------------------------



Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/antoine/py3k/refleaks/reflogTSIzMe', '-x']

From python-checkins at python.org  Tue Jul  6 04:08:36 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Tue,  6 Jul 2010 04:08:36 +0200 (CEST)
Subject: [Python-checkins] r82601 -
	python/branches/py3k/Doc/library/stdtypes.rst
Message-ID: <20100706020836.E47A6F0BA@mail.python.org>

Author: senthil.kumaran
Date: Tue Jul  6 04:08:36 2010
New Revision: 82601

Log:
Change 'lowered' to 'lowercased'



Modified:
   python/branches/py3k/Doc/library/stdtypes.rst

Modified: python/branches/py3k/Doc/library/stdtypes.rst
==============================================================================
--- python/branches/py3k/Doc/library/stdtypes.rst	(original)
+++ python/branches/py3k/Doc/library/stdtypes.rst	Tue Jul  6 04:08:36 2010
@@ -967,7 +967,7 @@
 .. method:: str.capitalize()
 
    Return a copy of the string with its first character capitalized and the
-   rest lowered.
+   rest lowercased.
 
 
 .. method:: str.center(width[, fillchar])

From python-checkins at python.org  Tue Jul  6 05:03:34 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Tue,  6 Jul 2010 05:03:34 +0200 (CEST)
Subject: [Python-checkins] r82602 - in python/branches/release31-maint:
	Doc/library/stdtypes.rst
Message-ID: <20100706030334.4C4EFED8F@mail.python.org>

Author: senthil.kumaran
Date: Tue Jul  6 05:03:34 2010
New Revision: 82602

Log:
Merged revisions 82601 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82601 | senthil.kumaran | 2010-07-06 07:38:36 +0530 (Tue, 06 Jul 2010) | 3 lines
  
  Change 'lowered' to 'lowercased'
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Doc/library/stdtypes.rst

Modified: python/branches/release31-maint/Doc/library/stdtypes.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/stdtypes.rst	(original)
+++ python/branches/release31-maint/Doc/library/stdtypes.rst	Tue Jul  6 05:03:34 2010
@@ -801,7 +801,7 @@
 .. method:: str.capitalize()
 
    Return a copy of the string with its first character capitalized and the
-   rest lowered.
+   rest lowercased.
 
 
 .. method:: str.center(width[, fillchar])

From python-checkins at python.org  Tue Jul  6 05:06:53 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Tue,  6 Jul 2010 05:06:53 +0200 (CEST)
Subject: [Python-checkins] r82603 - in python/branches/release27-maint:
	Doc/library/stdtypes.rst
Message-ID: <20100706030653.33F60F074@mail.python.org>

Author: senthil.kumaran
Date: Tue Jul  6 05:06:53 2010
New Revision: 82603

Log:
Merged revisions 82601 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82601 | senthil.kumaran | 2010-07-06 07:38:36 +0530 (Tue, 06 Jul 2010) | 3 lines
  
  Change 'lowered' to 'lowercased'
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Doc/library/stdtypes.rst

Modified: python/branches/release27-maint/Doc/library/stdtypes.rst
==============================================================================
--- python/branches/release27-maint/Doc/library/stdtypes.rst	(original)
+++ python/branches/release27-maint/Doc/library/stdtypes.rst	Tue Jul  6 05:06:53 2010
@@ -831,7 +831,7 @@
 .. method:: str.capitalize()
 
    Return a copy of the string with its first character capitalized and the
-   rest lowered.
+   rest lowercased.
 
    For 8-bit strings, this method is locale-dependent.
 

From orsenthil at gmail.com  Tue Jul  6 05:07:53 2010
From: orsenthil at gmail.com (Senthil Kumaran)
Date: Tue, 6 Jul 2010 08:37:53 +0530
Subject: [Python-checkins] r82570 -
 python/branches/py3k/Doc/library/stdtypes.rst
In-Reply-To: <4C3266DF.7050907@trueblade.com>
References: <20100705114143.25BBFEE99A@mail.python.org>
	<i0thpi$8ng$1@dough.gmane.org> <4C3266DF.7050907@trueblade.com>
Message-ID: <20100706030753.GD1583@remy>

On Mon, Jul 05, 2010 at 07:12:31PM -0400, Eric Smith wrote:
> >>+   Return a copy of the string with its first character capitalized and the
> >>+   rest lowered.
> >
> >That should be "lowercased" IINM.
> 
> Yes, that would be a better word.
> 

I changed that one. Thanks!

-- 
Senthil

A arte ? uma magia que liberta a mentira de ser verdadeira.
                -- Theodor Adorno, fil?sofo alem?o

From python-checkins at python.org  Tue Jul  6 12:53:30 2010
From: python-checkins at python.org (matthias.klose)
Date: Tue,  6 Jul 2010 12:53:30 +0200 (CEST)
Subject: [Python-checkins] r82604 - python/branches/py3k/Python/sysmodule.c
Message-ID: <20100706105330.B3BD0EEA07@mail.python.org>

Author: matthias.klose
Date: Tue Jul  6 12:53:30 2010
New Revision: 82604

Log:
- sysmodule.c (get_hash_info): Define as static function.


Modified:
   python/branches/py3k/Python/sysmodule.c

Modified: python/branches/py3k/Python/sysmodule.c
==============================================================================
--- python/branches/py3k/Python/sysmodule.c	(original)
+++ python/branches/py3k/Python/sysmodule.c	Tue Jul  6 12:53:30 2010
@@ -595,7 +595,7 @@
     5,
 };
 
-PyObject *
+static PyObject *
 get_hash_info(void)
 {
     PyObject *hash_info;

From python-checkins at python.org  Tue Jul  6 15:54:53 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Tue,  6 Jul 2010 15:54:53 +0200 (CEST)
Subject: [Python-checkins] r82605 - in sandbox/branches/py3k-datetime:
	datetime.py test_datetime.py
Message-ID: <20100706135453.5D3C5EE98E@mail.python.org>

Author: alexander.belopolsky
Date: Tue Jul  6 15:54:47 2010
New Revision: 82605

Log:
Added type check for better match with C implementation

Modified:
   sandbox/branches/py3k-datetime/datetime.py
   sandbox/branches/py3k-datetime/test_datetime.py

Modified: sandbox/branches/py3k-datetime/datetime.py
==============================================================================
--- sandbox/branches/py3k-datetime/datetime.py	(original)
+++ sandbox/branches/py3k-datetime/datetime.py	Tue Jul  6 15:54:47 2010
@@ -665,7 +665,8 @@
 
         year, month, day (required, base 1)
         """
-        if isinstance(year, bytes) and len(year) == 4:
+        if (isinstance(year, bytes) and len(year) == 4 and
+            1 <= year[2] <= 12 and month is None):  # Month is sane
             # Pickle support
             self = object.__new__(cls)
             self.__setstate(year)
@@ -978,6 +979,8 @@
         else:
             return (self.__class__, args, state)
 
+_tzinfo_class = tzinfo
+
 class time:
     """Time with time zone.
 
@@ -1272,7 +1275,10 @@
         (self._hour, self._minute, self._second,
          us1, us2, us3) = string
         self._microsecond = (((us1 << 8) | us2) << 8) | us3
-        self._tzinfo = tzinfo
+        if tzinfo is None or isinstance(tzinfo, _tzinfo_class):
+            self._tzinfo = tzinfo
+        else:
+            raise TypeError("bad tzinfo state arg %r" % tzinfo)
 
     def __reduce__(self):
         return (time, self._getstate())
@@ -1741,7 +1747,10 @@
          self._minute, self._second, us1, us2, us3) = string
         self._year = yhi * 256 + ylo
         self._microsecond = (((us1 << 8) | us2) << 8) | us3
-        self._tzinfo = tzinfo
+        if tzinfo is None or isinstance(tzinfo, _tzinfo_class):
+            self._tzinfo = tzinfo
+        else:
+            raise TypeError("bad tzinfo state arg %r" % tzinfo)
 
     def __reduce__(self):
         return (self.__class__, self._getstate())

Modified: sandbox/branches/py3k-datetime/test_datetime.py
==============================================================================
--- sandbox/branches/py3k-datetime/test_datetime.py	(original)
+++ sandbox/branches/py3k-datetime/test_datetime.py	Tue Jul  6 15:54:47 2010
@@ -69,34 +69,11 @@
         return self.__name
     def dst(self, dt):
         return self.__dstoffset
-    def __eq__(self, other):
-        return (self.__offset == other.__offset and
-                self.__name == other.__name and
-                self.__dstoffset == other.__dstoffset)
-    def __getinitargs__(self):
-        return (self.__offset,
-                self.__name,
-                self.__dstoffset)
 
 class PicklableFixedOffset(FixedOffset):
 
     def __init__(self, offset=None, name=None, dstoffset=None):
         FixedOffset.__init__(self, offset, name, dstoffset)
-        
-
-class PicklableFixedOffsetWithState(PicklableFixedOffset):
-
-    def __init__(self, offset=None, name=None, dstoffset=None,
-                 state=None):
-        FixedOffset.__init__(self, offset, name, dstoffset)
-        self.__state = state
-    def __getstate__(self):
-        return self.__state
-    def __setstate__(self, state):
-        self.__state = state
-    def __eq__(self, other):
-        return (FixedOffset.__eq__(self, other) and
-                self.__state == other.__state)
 
 class TestTZInfo(unittest.TestCase):
 
@@ -148,7 +125,6 @@
         offset = timedelta(minutes=-300)
         for otype, args in [
             (PicklableFixedOffset, (offset, 'cookie')),
-            (PicklableFixedOffsetWithState, (offset, 'cookie', 'state')),
             (timezone, (offset,)),
             (timezone, (offset, "EST"))]:
             orig = otype(*args)
@@ -164,7 +140,6 @@
                 self.assertIs(type(derived), otype)
                 self.assertEqual(derived.utcoffset(None), offset)
                 self.assertEqual(derived.tzname(None), oname)
-                self.assertEqual(derived, orig)
 
 class TestTimeZone(unittest.TestCase):
 
@@ -261,33 +236,6 @@
                              t.replace(tzinfo=tz).dst())
 
 #############################################################################
-# Base clase for testing pickling state of timedelta, time, date and
-# datetime instances.
-
-class ReduceStructure:
-    initargs = 1, 1, 1
-    statetype = (bytes, tzinfo)
-    def test_reduce(self):
-        obj = self.theclass(*self.initargs)
-        factory, state = obj.__reduce__()[:2]
-        self.assertIs(factory, self.theclass)
-        for v, t in zip(state, self.statetype):
-            self.assertIsInstance(v, t)
-
-    def test_construct_from_state(self):
-        obj = self.theclass(*self.initargs)
-        factory, state = obj.__reduce__()[:2]
-        initbytes = state[0]
-        # This test should be overriden in tests for classes that
-        # don't produce their state as bytes.  Curently, timedelta
-        # and timezone
-        assert isinstance(initbytes, bytes)
-        obj = factory(initbytes)
-        self.assertIs(type(obj), self.theclass)
-        self.assertRaises(TypeError, factory, initbytes[:-1]) 
-        self.assertRaises(TypeError, factory, initbytes + b'x') 
-
-#############################################################################
 # Base clase for testing a particular aspect of timedelta, time, date and
 # datetime comparisons.
 
@@ -324,10 +272,10 @@
 #############################################################################
 # timedelta tests
 
-class TestTimeDelta(HarmlessMixedComparison, ReduceStructure, unittest.TestCase):
+class TestTimeDelta(HarmlessMixedComparison, unittest.TestCase):
 
     theclass = timedelta
-    statetype = (int, int, int)
+
     def test_constructor(self):
         eq = self.assertEqual
         td = timedelta
@@ -353,9 +301,6 @@
         eq(td(seconds=0.001), td(milliseconds=1))
         eq(td(milliseconds=0.001), td(microseconds=1))
 
-    def test_construct_from_state(self):
-        pass
-
     def test_computations(self):
         eq = self.assertEqual
         td = timedelta
@@ -788,12 +733,11 @@
 class SubclassDate(date):
     sub_var = 1
 
-class TestDate(HarmlessMixedComparison, ReduceStructure, unittest.TestCase):
+class TestDate(HarmlessMixedComparison, unittest.TestCase):
     # Tests here should pass for both dates and datetimes, except for a
     # few tests that TestDateTime overrides.
 
     theclass = date
-    statetype = (bytes,)
 
     def test_basic_attributes(self):
         dt = self.theclass(2002, 3, 1)
@@ -937,6 +881,7 @@
     def test_computations(self):
         a = self.theclass(2002, 1, 31)
         b = self.theclass(1956, 1, 31)
+        c = self.theclass(2001,2,1)
 
         diff = a-b
         self.assertEqual(diff.days, 46*365 + len(range(1956, 2002, 4)))
@@ -962,6 +907,7 @@
         self.assertEqual(a - (a + day), -day)
         self.assertEqual(a - (a - week), week)
         self.assertEqual(a - (a - day), day)
+        self.assertEqual(c - (c - day), day)
 
         # Add/sub ints or floats should be illegal
         for i in 1, 1.0:
@@ -1403,6 +1349,10 @@
         for month_byte in b'9', b'\0', b'\r', b'\xff':
             self.assertRaises(TypeError, self.theclass,
                                          base[:2] + month_byte + base[3:])
+        # Good bytes, but bad tzinfo:
+        self.assertRaises(TypeError, self.theclass,
+                          bytes([1] * len(base)), 'EST')
+
         for ord_byte in range(1, 13):
             # This shouldn't blow up because of the month byte alone.  If
             # the implementation changes to do more-careful checking, it may
@@ -1418,7 +1368,6 @@
 class TestDateTime(TestDate):
 
     theclass = datetime
-    statetype = (bytes, tzinfo)
 
     def test_basic_attributes(self):
         dt = self.theclass(2002, 3, 1, 12, 0)
@@ -1767,8 +1716,6 @@
         # than one microsecond smaller than an integer.
         self.assertEqual(self.theclass.fromtimestamp(0.9999999),
                          self.theclass.fromtimestamp(1))
-        self.assertEqual(self.theclass.utcfromtimestamp(0.9999999),
-                         self.theclass.utcfromtimestamp(1))
 
     def test_insane_fromtimestamp(self):
         # It's possible that some platform maps time_t to double,
@@ -1996,7 +1943,7 @@
 class SubclassTime(time):
     sub_var = 1
 
-class TestTime(HarmlessMixedComparison, ReduceStructure, unittest.TestCase):
+class TestTime(HarmlessMixedComparison, unittest.TestCase):
 
     theclass = time
 
@@ -3067,7 +3014,7 @@
 
     def test_utctimetuple(self):
         class DST(tzinfo):
-            def __init__(self, dstvalue):
+            def __init__(self, dstvalue=0):
                 if isinstance(dstvalue, int):
                     dstvalue = timedelta(minutes=dstvalue)
                 self.dstvalue = dstvalue
@@ -3102,6 +3049,26 @@
             # is never in effect for a UTC time.
             self.assertEqual(0, t.tm_isdst)
 
+        # For naive datetime, utctimetuple == timetuple except for isdst
+        d = cls(1, 2, 3, 10, 20, 30, 40)
+        t = d.utctimetuple()
+        self.assertEqual(t[:-1], d.timetuple()[:-1])
+        self.assertEqual(0, t.tm_isdst)
+        # Same if utcoffset is None
+        class NOFS(DST):
+            def utcoffset(self, dt):
+                return None
+        d = cls(1, 2, 3, 10, 20, 30, 40, tzinfo=NOFS())
+        t = d.utctimetuple()
+        self.assertEqual(t[:-1], d.timetuple()[:-1])
+        self.assertEqual(0, t.tm_isdst)
+        # Check that bad tzinfo is detected
+        class BOFS(DST):
+            def utcoffset(self, dt):
+                return "EST"
+        d = cls(1, 2, 3, 10, 20, 30, 40, tzinfo=BOFS())
+        self.assertRaises(TypeError, d.utctimetuple)
+
         # Check that utctimetuple() is the same as
         # astimezone(utc).timetuple()
         d = cls(2010, 11, 13, 14, 15, 16, 171819)

From python-checkins at python.org  Tue Jul  6 16:55:57 2010
From: python-checkins at python.org (mark.dickinson)
Date: Tue,  6 Jul 2010 16:55:57 +0200 (CEST)
Subject: [Python-checkins] r82606 - python/branches/py3k-issue4970
Message-ID: <20100706145557.5EE5EEE98E@mail.python.org>

Author: mark.dickinson
Date: Tue Jul  6 16:55:57 2010
New Revision: 82606

Log:
Delete branch that was used to fix issue 4970 (signal 32 on gentoo buildbot).


Removed:
   python/branches/py3k-issue4970/

From python-checkins at python.org  Tue Jul  6 17:00:41 2010
From: python-checkins at python.org (mark.dickinson)
Date: Tue,  6 Jul 2010 17:00:41 +0200 (CEST)
Subject: [Python-checkins] r82607 - python/branches/py3k/Modules/_math.c
Message-ID: <20100706150041.12A07EEA0C@mail.python.org>

Author: mark.dickinson
Date: Tue Jul  6 17:00:40 2010
New Revision: 82607

Log:
Indentation and PEP 7 fixes.

Modified:
   python/branches/py3k/Modules/_math.c

Modified: python/branches/py3k/Modules/_math.c
==============================================================================
--- python/branches/py3k/Modules/_math.c	(original)
+++ python/branches/py3k/Modules/_math.c	Tue Jul  6 17:00:40 2010
@@ -55,7 +55,8 @@
     else if (x >= two_pow_p28) {        /* x > 2**28 */
         if (Py_IS_INFINITY(x)) {
             return x+x;
-        } else {
+        }
+        else {
             return log(x)+ln2;          /* acosh(huge)=log(2x) */
         }
     }
@@ -173,15 +174,15 @@
     */
 
     if (fabs(x) < 0.7) {
-    double u;
-    u = exp(x);
-    if (u == 1.0)
-        return x;
-    else
-        return (u - 1.0) * x / log(u);
+        double u;
+        u = exp(x);
+        if (u == 1.0)
+            return x;
+        else
+            return (u - 1.0) * x / log(u);
     }
     else
-    return exp(x) - 1.0;
+        return exp(x) - 1.0;
 }
 
 /* log1p(x) = log(1+x).  The log1p function is designed to avoid the
@@ -213,17 +214,19 @@
 
     double y;
     if (fabs(x) < DBL_EPSILON/2.) {
-    return x;
-    } else if (-0.5 <= x && x <= 1.) {
-    /* WARNING: it's possible than an overeager compiler
-       will incorrectly optimize the following two lines
-       to the equivalent of "return log(1.+x)". If this
-       happens, then results from log1p will be inaccurate
-       for small x. */
-    y = 1.+x;
-    return log(y)-((y-1.)-x)/y;
-    } else {
-    /* NaNs and infinities should end up here */
-    return log(1.+x);
+        return x;
+    }
+    else if (-0.5 <= x && x <= 1.) {
+        /* WARNING: it's possible than an overeager compiler
+           will incorrectly optimize the following two lines
+           to the equivalent of "return log(1.+x)". If this
+           happens, then results from log1p will be inaccurate
+           for small x. */
+        y = 1.+x;
+        return log(y)-((y-1.)-x)/y;
+    }
+    else {
+        /* NaNs and infinities should end up here */
+        return log(1.+x);
     }
 }

From python-checkins at python.org  Tue Jul  6 17:03:42 2010
From: python-checkins at python.org (mark.dickinson)
Date: Tue,  6 Jul 2010 17:03:42 +0200 (CEST)
Subject: [Python-checkins] r82608 - in python/branches/release27-maint:
	Modules/_math.c
Message-ID: <20100706150342.35800EE99A@mail.python.org>

Author: mark.dickinson
Date: Tue Jul  6 17:03:42 2010
New Revision: 82608

Log:
Merged revisions 82607 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82607 | mark.dickinson | 2010-07-06 16:00:40 +0100 (Tue, 06 Jul 2010) | 1 line
  
  Indentation and PEP 7 fixes.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Modules/_math.c

Modified: python/branches/release27-maint/Modules/_math.c
==============================================================================
--- python/branches/release27-maint/Modules/_math.c	(original)
+++ python/branches/release27-maint/Modules/_math.c	Tue Jul  6 17:03:42 2010
@@ -55,7 +55,8 @@
     else if (x >= two_pow_p28) {        /* x > 2**28 */
         if (Py_IS_INFINITY(x)) {
             return x+x;
-        } else {
+        }
+        else {
             return log(x)+ln2;          /* acosh(huge)=log(2x) */
         }
     }
@@ -173,15 +174,15 @@
     */
 
     if (fabs(x) < 0.7) {
-    double u;
-    u = exp(x);
-    if (u == 1.0)
-        return x;
-    else
-        return (u - 1.0) * x / log(u);
+        double u;
+        u = exp(x);
+        if (u == 1.0)
+            return x;
+        else
+            return (u - 1.0) * x / log(u);
     }
     else
-    return exp(x) - 1.0;
+        return exp(x) - 1.0;
 }
 
 /* log1p(x) = log(1+x).  The log1p function is designed to avoid the
@@ -213,17 +214,19 @@
 
     double y;
     if (fabs(x) < DBL_EPSILON/2.) {
-    return x;
-    } else if (-0.5 <= x && x <= 1.) {
-    /* WARNING: it's possible than an overeager compiler
-       will incorrectly optimize the following two lines
-       to the equivalent of "return log(1.+x)". If this
-       happens, then results from log1p will be inaccurate
-       for small x. */
-    y = 1.+x;
-    return log(y)-((y-1.)-x)/y;
-    } else {
-    /* NaNs and infinities should end up here */
-    return log(1.+x);
+        return x;
+    }
+    else if (-0.5 <= x && x <= 1.) {
+        /* WARNING: it's possible than an overeager compiler
+           will incorrectly optimize the following two lines
+           to the equivalent of "return log(1.+x)". If this
+           happens, then results from log1p will be inaccurate
+           for small x. */
+        y = 1.+x;
+        return log(y)-((y-1.)-x)/y;
+    }
+    else {
+        /* NaNs and infinities should end up here */
+        return log(1.+x);
     }
 }

From python-checkins at python.org  Tue Jul  6 17:08:55 2010
From: python-checkins at python.org (vinay.sajip)
Date: Tue,  6 Jul 2010 17:08:55 +0200 (CEST)
Subject: [Python-checkins] r82609 - in python/branches:
	py3k/Doc/library/logging.rst release26-maint/Doc/library/logging.rst
	release27-maint/Doc/library/logging.rst
Message-ID: <20100706150855.F3889EE992@mail.python.org>

Author: vinay.sajip
Date: Tue Jul  6 17:08:55 2010
New Revision: 82609

Log:
Minor improvements to logging documentation.

Modified:
   python/branches/py3k/Doc/library/logging.rst
   python/branches/release26-maint/Doc/library/logging.rst
   python/branches/release27-maint/Doc/library/logging.rst

Modified: python/branches/py3k/Doc/library/logging.rst
==============================================================================
--- python/branches/py3k/Doc/library/logging.rst	(original)
+++ python/branches/py3k/Doc/library/logging.rst	Tue Jul  6 17:08:55 2010
@@ -246,7 +246,7 @@
   methods listed above, but this is how to log at custom log levels.
 
 :func:`getLogger` returns a reference to a logger instance with the specified
-if it is provided, or ``root`` if not.  The names are period-separated
+name if it is provided, or ``root`` if not.  The names are period-separated
 hierarchical structures.  Multiple calls to :func:`getLogger` with the same name
 will return a reference to the same logger object.  Loggers that are further
 down in the hierarchical list are children of loggers higher up in the list.
@@ -1572,6 +1572,11 @@
       69 myapp.area2     WARNING  Jail zesty vixen who grabbed pay from quack.
       69 myapp.area2     ERROR    The five boxing wizards jump quickly.
 
+Note that there are some security issues with pickle in some scenarios. If
+these affect you, you can use an alternative serialization scheme by overriding
+the :meth:`makePickle` method and implementing your alternative there, as
+well as adapting the above script to use your alternative serialization.
+
 Using arbitrary objects as messages
 -----------------------------------
 

Modified: python/branches/release26-maint/Doc/library/logging.rst
==============================================================================
--- python/branches/release26-maint/Doc/library/logging.rst	(original)
+++ python/branches/release26-maint/Doc/library/logging.rst	Tue Jul  6 17:08:55 2010
@@ -55,10 +55,12 @@
 
 Most applications are probably going to want to log to a file, so let's start
 with that case. Using the :func:`basicConfig` function, we can set up the
-default handler so that debug messages are written to a file::
+default handler so that debug messages are written to a file (in the example,
+we assume that you have the appropriate permissions to create a file called
+*example.log* in the current directory)::
 
    import logging
-   LOG_FILENAME = '/tmp/logging_example.out'
+   LOG_FILENAME = 'example.log'
    logging.basicConfig(filename=LOG_FILENAME,level=logging.DEBUG)
 
    logging.debug('This message should go to the log file')
@@ -77,7 +79,7 @@
    import logging
    import logging.handlers
 
-   LOG_FILENAME = '/tmp/logging_rotatingfile_example.out'
+   LOG_FILENAME = 'logging_rotatingfile_example.out'
 
    # Set up a specific logger with our desired output level
    my_logger = logging.getLogger('MyLogger')
@@ -102,14 +104,14 @@
 The result should be 6 separate files, each with part of the log history for the
 application::
 
-   /tmp/logging_rotatingfile_example.out
-   /tmp/logging_rotatingfile_example.out.1
-   /tmp/logging_rotatingfile_example.out.2
-   /tmp/logging_rotatingfile_example.out.3
-   /tmp/logging_rotatingfile_example.out.4
-   /tmp/logging_rotatingfile_example.out.5
+   logging_rotatingfile_example.out
+   logging_rotatingfile_example.out.1
+   logging_rotatingfile_example.out.2
+   logging_rotatingfile_example.out.3
+   logging_rotatingfile_example.out.4
+   logging_rotatingfile_example.out.5
 
-The most current file is always :file:`/tmp/logging_rotatingfile_example.out`,
+The most current file is always :file:`logging_rotatingfile_example.out`,
 and each time it reaches the size limit it is renamed with the suffix
 ``.1``. Each of the existing backup files is renamed to increment the suffix
 (``.1`` becomes ``.2``, etc.)  and the ``.6`` file is erased.
@@ -246,16 +248,16 @@
   methods listed above, but this is how to log at custom log levels.
 
 :func:`getLogger` returns a reference to a logger instance with the specified
-if it it is provided, or ``root`` if not.  The names are period-separated
+name if it is provided, or ``root`` if not.  The names are period-separated
 hierarchical structures.  Multiple calls to :func:`getLogger` with the same name
 will return a reference to the same logger object.  Loggers that are further
 down in the hierarchical list are children of loggers higher up in the list.
 For example, given a logger with a name of ``foo``, loggers with names of
-``foo.bar``, ``foo.bar.baz``, and ``foo.bam`` are all children of ``foo``.
-Child loggers propagate messages up to their parent loggers.  Because of this,
-it is unnecessary to define and configure all the loggers an application uses.
-It is sufficient to configure a top-level logger and create child loggers as
-needed.
+``foo.bar``, ``foo.bar.baz``, and ``foo.bam`` are all descendants of ``foo``.
+Child loggers propagate messages up to the handlers associated with their
+ancestor loggers.  Because of this, it is unnecessary to define and configure
+handlers for all the loggers an application uses. It is sufficient to
+configure handlers for a top-level logger and create child loggers as needed.
 
 
 Handlers
@@ -283,15 +285,16 @@
   are there two :func:`setLevel` methods?  The level set in the logger
   determines which severity of messages it will pass to its handlers.  The level
   set in each handler determines which messages that handler will send on.
-  :func:`setFormatter` selects a Formatter object for this handler to use.
+
+* :func:`setFormatter` selects a Formatter object for this handler to use.
 
 * :func:`addFilter` and :func:`removeFilter` respectively configure and
   deconfigure filter objects on handlers.
 
-Application code should not directly instantiate and use handlers.  Instead, the
-:class:`Handler` class is a base class that defines the interface that all
-Handlers should have and establishes some default behavior that child classes
-can use (or override).
+Application code should not directly instantiate and use instances of
+:class:`Handler`.  Instead, the :class:`Handler` class is a base class that
+defines the interface that all handlers should have and establishes some
+default behavior that child classes can use (or override).
 
 
 Formatters
@@ -510,7 +513,9 @@
 can have zero, one or more handlers associated with it (via the
 :meth:`addHandler` method of :class:`Logger`). In addition to any handlers
 directly associated with a logger, *all handlers associated with all ancestors
-of the logger* are called to dispatch the message.
+of the logger* are called to dispatch the message (unless the *propagate* flag
+for a logger is set to a false value, at which point the passing to ancestor
+handlers stops).
 
 Just as for loggers, handlers can have levels associated with them. A handler's
 level acts as a filter in the same way as a logger's level does. If a handler
@@ -712,7 +717,11 @@
 
    Provides an overriding level *lvl* for all loggers which takes precedence over
    the logger's own level. When the need arises to temporarily throttle logging
-   output down across the whole application, this function can be useful.
+   output down across the whole application, this function can be useful. Its
+   effect is to disable all logging calls of severity *lvl* and below, so that
+   if you call it with a value of INFO, then all INFO and DEBUG events would be
+   discarded, whereas those of severity WARNING and above would be processed
+   according to the logger's effective level.
 
 
 .. function:: addLevelName(lvl, levelName)
@@ -748,12 +757,12 @@
 
    Does basic configuration for the logging system by creating a
    :class:`StreamHandler` with a default :class:`Formatter` and adding it to the
-   root logger. The function does nothing if any handlers have been defined for
-   the root logger. The functions :func:`debug`, :func:`info`, :func:`warning`,
+   root logger. The functions :func:`debug`, :func:`info`, :func:`warning`,
    :func:`error` and :func:`critical` will call :func:`basicConfig` automatically
    if no handlers are defined for the root logger.
 
-   This function does nothing if the root logger already has handlers configured.
+   This function does nothing if the root logger already has handlers
+   configured for it.
 
    .. versionchanged:: 2.4
       Formerly, :func:`basicConfig` did not take any keyword arguments.
@@ -827,8 +836,8 @@
 .. attribute:: Logger.propagate
 
    If this evaluates to false, logging messages are not passed by this logger or by
-   child loggers to higher level (ancestor) loggers. The constructor sets this
-   attribute to 1.
+   its child loggers to the handlers of higher level (ancestor) loggers. The
+   constructor sets this attribute to 1.
 
 
 .. method:: Logger.setLevel(lvl)
@@ -1043,14 +1052,14 @@
 
    logging.basicConfig(level=logging.DEBUG,
                        format='%(asctime)s %(levelname)s %(message)s',
-                       filename='/tmp/myapp.log',
+                       filename='myapp.log',
                        filemode='w')
    logging.debug('A debug message')
    logging.info('Some information')
    logging.warning('A shot across the bows')
 
 The :meth:`basicConfig` method is used to change the configuration defaults,
-which results in output (written to ``/tmp/myapp.log``) which should look
+which results in output (written to ``myapp.log``) which should look
 something like the following::
 
    2004-07-02 13:00:08,743 DEBUG A debug message
@@ -1499,6 +1508,11 @@
       69 myapp.area2     WARNING  Jail zesty vixen who grabbed pay from quack.
       69 myapp.area2     ERROR    The five boxing wizards jump quickly.
 
+Note that there are some security issues with pickle in some scenarios. If
+these affect you, you can use an alternative serialization scheme by overriding
+the :meth:`makePickle` method and implementing your alternative there, as
+well as adapting the above script to use your alternative serialization.
+
 Using arbitrary objects as messages
 -----------------------------------
 
@@ -1659,14 +1673,14 @@
 StreamHandler
 ^^^^^^^^^^^^^
 
-.. module:: logging.handlers
-
 The :class:`StreamHandler` class, located in the core :mod:`logging` package,
 sends logging output to streams such as *sys.stdout*, *sys.stderr* or any
 file-like object (or, more precisely, any object which supports :meth:`write`
 and :meth:`flush` methods).
 
 
+.. currentmodule:: logging
+
 .. class:: StreamHandler([strm])
 
    Returns a new instance of the :class:`StreamHandler` class. If *strm* is
@@ -1850,6 +1864,11 @@
    The extensions are date-and-time based, using the strftime format
    ``%Y-%m-%d_%H-%M-%S`` or a leading portion thereof, depending on the
    rollover interval.
+
+   When computing the next rollover time for the first time (when the handler
+   is created), the last modification time of an existing log file, or else
+   the current time, is used to compute when the next rotation will occur.
+
    If the *utc* argument is true, times in UTC will be used; otherwise
    local time is used.
 
@@ -1922,6 +1941,11 @@
       Pickles the record's attribute dictionary in binary format with a length
       prefix, and returns it ready for transmission across the socket.
 
+      Note that pickles aren't completely secure. If you are concerned about
+      security, you may want to override this method to implement a more secure
+      mechanism. For example, you can sign pickles using HMAC and then verify
+      them on the receiving end, or alternatively you can disable unpickling of
+      global objects on the receiving end.
 
    .. method:: send(packet)
 
@@ -2002,6 +2026,85 @@
       or integers - if strings are passed, internal mapping dictionaries are
       used to convert them to integers.
 
+      The symbolic ``LOG_`` values are defined in :class:`SysLogHandler` and
+      mirror the values defined in the ``sys/syslog.h`` header file.
+
+      **Priorities**
+
+      +--------------------------+---------------+
+      | Name (string)            | Symbolic value|
+      +==========================+===============+
+      | ``alert``                | LOG_ALERT     |
+      +--------------------------+---------------+
+      | ``crit`` or ``critical`` | LOG_CRIT      |
+      +--------------------------+---------------+
+      | ``debug``                | LOG_DEBUG     |
+      +--------------------------+---------------+
+      | ``emerg`` or ``panic``   | LOG_EMERG     |
+      +--------------------------+---------------+
+      | ``err`` or ``error``     | LOG_ERR       |
+      +--------------------------+---------------+
+      | ``info``                 | LOG_INFO      |
+      +--------------------------+---------------+
+      | ``notice``               | LOG_NOTICE    |
+      +--------------------------+---------------+
+      | ``warn`` or ``warning``  | LOG_WARNING   |
+      +--------------------------+---------------+
+
+      **Facilities**
+
+      +---------------+---------------+
+      | Name (string) | Symbolic value|
+      +===============+===============+
+      | ``auth``      | LOG_AUTH      |
+      +---------------+---------------+
+      | ``authpriv``  | LOG_AUTHPRIV  |
+      +---------------+---------------+
+      | ``cron``      | LOG_CRON      |
+      +---------------+---------------+
+      | ``daemon``    | LOG_DAEMON    |
+      +---------------+---------------+
+      | ``ftp``       | LOG_FTP       |
+      +---------------+---------------+
+      | ``kern``      | LOG_KERN      |
+      +---------------+---------------+
+      | ``lpr``       | LOG_LPR       |
+      +---------------+---------------+
+      | ``mail``      | LOG_MAIL      |
+      +---------------+---------------+
+      | ``news``      | LOG_NEWS      |
+      +---------------+---------------+
+      | ``syslog``    | LOG_SYSLOG    |
+      +---------------+---------------+
+      | ``user``      | LOG_USER      |
+      +---------------+---------------+
+      | ``uucp``      | LOG_UUCP      |
+      +---------------+---------------+
+      | ``local0``    | LOG_LOCAL0    |
+      +---------------+---------------+
+      | ``local1``    | LOG_LOCAL1    |
+      +---------------+---------------+
+      | ``local2``    | LOG_LOCAL2    |
+      +---------------+---------------+
+      | ``local3``    | LOG_LOCAL3    |
+      +---------------+---------------+
+      | ``local4``    | LOG_LOCAL4    |
+      +---------------+---------------+
+      | ``local5``    | LOG_LOCAL5    |
+      +---------------+---------------+
+      | ``local6``    | LOG_LOCAL6    |
+      +---------------+---------------+
+      | ``local7``    | LOG_LOCAL7    |
+      +---------------+---------------+
+
+   .. method:: mapPriority(levelname)
+
+      Maps a logging level name to a syslog priority name.
+      You may need to override this if you are using custom levels, or
+      if the default algorithm is not suitable for your needs. The
+      default algorithm maps ``DEBUG``, ``INFO``, ``WARNING``, ``ERROR`` and
+      ``CRITICAL`` to the equivalent syslog names, and all other level
+      names to "warning".
 
 .. _nt-eventlog-handler:
 
@@ -2481,17 +2584,17 @@
 ^^^^^^^^^^^^^^^^^^^^^^^^^
 
 The configuration file format understood by :func:`fileConfig` is based on
-ConfigParser functionality. The file must contain sections called ``[loggers]``,
-``[handlers]`` and ``[formatters]`` which identify by name the entities of each
-type which are defined in the file. For each such entity, there is a separate
-section which identified how that entity is configured. Thus, for a logger named
-``log01`` in the ``[loggers]`` section, the relevant configuration details are
-held in a section ``[logger_log01]``. Similarly, a handler called ``hand01`` in
-the ``[handlers]`` section will have its configuration held in a section called
-``[handler_hand01]``, while a formatter called ``form01`` in the
-``[formatters]`` section will have its configuration specified in a section
-called ``[formatter_form01]``. The root logger configuration must be specified
-in a section called ``[logger_root]``.
+:mod:`ConfigParser` functionality. The file must contain sections called
+``[loggers]``, ``[handlers]`` and ``[formatters]`` which identify by name the
+entities of each type which are defined in the file. For each such entity,
+there is a separate section which identifies how that entity is configured.
+Thus, for a logger named ``log01`` in the ``[loggers]`` section, the relevant
+configuration details are held in a section ``[logger_log01]``. Similarly, a
+handler called ``hand01`` in the ``[handlers]`` section will have its
+configuration held in a section called ``[handler_hand01]``, while a formatter
+called ``form01`` in the ``[formatters]`` section will have its configuration
+specified in a section called ``[formatter_form01]``. The root logger
+configuration must be specified in a section called ``[logger_root]``.
 
 Examples of these sections in the file are given below. ::
 

Modified: python/branches/release27-maint/Doc/library/logging.rst
==============================================================================
--- python/branches/release27-maint/Doc/library/logging.rst	(original)
+++ python/branches/release27-maint/Doc/library/logging.rst	Tue Jul  6 17:08:55 2010
@@ -248,7 +248,7 @@
   methods listed above, but this is how to log at custom log levels.
 
 :func:`getLogger` returns a reference to a logger instance with the specified
-if it is provided, or ``root`` if not.  The names are period-separated
+name if it is provided, or ``root`` if not.  The names are period-separated
 hierarchical structures.  Multiple calls to :func:`getLogger` with the same name
 will return a reference to the same logger object.  Loggers that are further
 down in the hierarchical list are children of loggers higher up in the list.
@@ -1590,6 +1590,11 @@
       69 myapp.area2     WARNING  Jail zesty vixen who grabbed pay from quack.
       69 myapp.area2     ERROR    The five boxing wizards jump quickly.
 
+Note that there are some security issues with pickle in some scenarios. If
+these affect you, you can use an alternative serialization scheme by overriding
+the :meth:`makePickle` method and implementing your alternative there, as
+well as adapting the above script to use your alternative serialization.
+
 Using arbitrary objects as messages
 -----------------------------------
 

From python-checkins at python.org  Tue Jul  6 17:11:44 2010
From: python-checkins at python.org (mark.dickinson)
Date: Tue,  6 Jul 2010 17:11:44 +0200 (CEST)
Subject: [Python-checkins] r82610 -
	python/branches/py3k/Objects/floatobject.c
Message-ID: <20100706151144.AE1A4EE992@mail.python.org>

Author: mark.dickinson
Date: Tue Jul  6 17:11:44 2010
New Revision: 82610

Log:
Style nit.

Modified:
   python/branches/py3k/Objects/floatobject.c

Modified: python/branches/py3k/Objects/floatobject.c
==============================================================================
--- python/branches/py3k/Objects/floatobject.c	(original)
+++ python/branches/py3k/Objects/floatobject.c	Tue Jul  6 17:11:44 2010
@@ -1422,7 +1422,7 @@
                     round_up = 1;
                     break;
                 }
-        if (round_up == 1) {
+        if (round_up) {
             x += 2*half_eps;
             if (top_exp == DBL_MAX_EXP &&
                 x == ldexp((double)(2*half_eps), DBL_MANT_DIG))

From python-checkins at python.org  Tue Jul  6 17:39:15 2010
From: python-checkins at python.org (mark.dickinson)
Date: Tue,  6 Jul 2010 17:39:15 +0200 (CEST)
Subject: [Python-checkins] r82611 - python/branches/py3k-dtoa
Message-ID: <20100706153915.109B0EE9B1@mail.python.org>

Author: mark.dickinson
Date: Tue Jul  6 17:39:14 2010
New Revision: 82611

Log:
Initialized merge tracking via "svnmerge" with revisions "1-82024" from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k


Modified:
   python/branches/py3k-dtoa/   (props changed)

From python-checkins at python.org  Tue Jul  6 17:55:29 2010
From: python-checkins at python.org (mark.dickinson)
Date: Tue,  6 Jul 2010 17:55:29 +0200 (CEST)
Subject: [Python-checkins] r82612 - in python/branches/py3k-dtoa:
	Demo/classes/Complex.py Demo/classes/Dates.py
	Demo/classes/Dbm.py Demo/classes/README Demo/classes/Range.py
	Demo/classes/Rev.py Demo/classes/Vec.py Demo/embed/Makefile
	Demo/embed/demo.c Demo/md5test/md5driver.py
	Demo/parser/test_parser.py Demo/parser/test_unparse.py
	Demo/parser/unparse.py Doc/README.txt Doc/c-api/arg.rst
	Doc/c-api/datetime.rst Doc/c-api/init.rst Doc/c-api/intro.rst
	Doc/c-api/number.rst Doc/c-api/unicode.rst
	Doc/data/refcounts.dat Doc/distutils/builtdist.rst
	Doc/documenting/building.rst Doc/documenting/index.rst
	Doc/documenting/markup.rst Doc/documenting/style.rst
	Doc/faq/gui.rst Doc/glossary.rst Doc/howto/functional.rst
	Doc/howto/unicode.rst Doc/library/2to3.rst
	Doc/library/calendar.rst Doc/library/codecs.rst
	Doc/library/contextlib.rst Doc/library/datetime.rst
	Doc/library/dis.rst Doc/library/http.server.rst
	Doc/library/importlib.rst Doc/library/logging.rst
	Doc/library/os.path.rst Doc/library/os.rst
	Doc/library/pdb.rst Doc/library/re.rst Doc/library/socket.rst
	Doc/library/ssl.rst Doc/library/stdtypes.rst
	Doc/library/string.rst Doc/library/struct.rst
	Doc/library/sys.rst Doc/library/sysconfig.rst
	Doc/library/unittest.rst Doc/library/urllib.parse.rst
	Doc/library/urllib.request.rst Doc/library/warnings.rst
	Doc/reference/compound_stmts.rst Doc/reference/datamodel.rst
	Doc/reference/introduction.rst Doc/reference/lexical_analysis.rst
	Doc/tools/sphinxext/pyspecific.py Doc/tutorial/classes.rst
	Doc/tutorial/controlflow.rst Doc/tutorial/introduction.rst
	Doc/using/mac.rst Doc/whatsnew/2.7.rst Doc/whatsnew/3.0.rst
	Doc/whatsnew/3.2.rst Grammar/Grammar Include/longobject.h
	Include/pyctype.h Include/traceback.h Include/unicodeobject.h
	Lib/_strptime.py Lib/contextlib.py Lib/ctypes/__init__.py
	Lib/ctypes/test/test_callbacks.py
	Lib/ctypes/test/test_win32.py Lib/dis.py
	Lib/distutils/msvc9compiler.py Lib/distutils/unixccompiler.py
	Lib/encodings/aliases.py Lib/encodings/cp858.py Lib/heapq.py
	Lib/http/cookiejar.py Lib/idlelib/textView.py
	Lib/importlib/__init__.py Lib/importlib/_bootstrap.py
	Lib/importlib/abc.py
	Lib/importlib/test/extension/test_case_sensitivity.py
	Lib/importlib/test/extension/test_finder.py
	Lib/importlib/test/extension/test_loader.py
	Lib/importlib/test/extension/test_path_hook.py
	Lib/importlib/test/import_/test_path.py
	Lib/importlib/test/regrtest.py
	Lib/importlib/test/source/test_abc_loader.py
	Lib/importlib/test/source/test_case_sensitivity.py
	Lib/importlib/test/source/test_file_loader.py
	Lib/importlib/test/source/test_finder.py
	Lib/importlib/test/source/test_path_hook.py
	Lib/importlib/test/source/test_source_encoding.py
	Lib/importlib/test/test_abc.py Lib/importlib/test/test_api.py
	Lib/importlib/test/test_util.py Lib/inspect.py Lib/lib2to3
	Lib/lib2to3/fixes/fix_itertools_imports.py
	Lib/lib2to3/tests/test_fixers.py Lib/logging/config.py
	Lib/macpath.py Lib/ntpath.py Lib/os2emxpath.py Lib/pdb.doc
	Lib/pdb.py Lib/pickletools.py Lib/posixpath.py Lib/pydoc.py
	Lib/site.py Lib/smtpd.py Lib/subprocess.py
	Lib/test/test_audioop.py Lib/test/test_codeccallbacks.py
	Lib/test/test_codecs.py Lib/test/test_complex.py
	Lib/test/test_contextlib.py Lib/test/test_datetime.py
	Lib/test/test_descr.py Lib/test/test_dict.py
	Lib/test/test_dis.py Lib/test/test_enumerate.py
	Lib/test/test_exceptions.py Lib/test/test_extcall.py
	Lib/test/test_float.py Lib/test/test_genericpath.py
	Lib/test/test_getargs2.py Lib/test/test_http_cookiejar.py
	Lib/test/test_imp.py Lib/test/test_import.py
	Lib/test/test_inspect.py Lib/test/test_iter.py
	Lib/test/test_keywordonlyarg.py Lib/test/test_logging.py
	Lib/test/test_math.py Lib/test/test_optparse.py
	Lib/test/test_os.py Lib/test/test_parser.py
	Lib/test/test_property.py Lib/test/test_pyexpat.py
	Lib/test/test_runpy.py Lib/test/test_strtod.py
	Lib/test/test_subprocess.py Lib/test/test_syntax.py
	Lib/test/test_tokenize.py Lib/test/test_tuple.py
	Lib/test/test_unicode.py Lib/test/test_urllib.py
	Lib/test/test_xmlrpc_net.py Lib/urllib/parse.py
	Lib/urllib/request.py Lib/uuid.py Lib/warnings.py Misc/ACKS
	Misc/NEWS Misc/Vim/vimrc Modules/_ctypes/callproc.c
	Modules/_ctypes/libffi/fficonfig.py.in
	Modules/_ctypes/libffi_msvc/LICENSE
	Modules/_ctypes/libffi_msvc/README
	Modules/_ctypes/libffi_msvc/README.ctypes
	Modules/_ctypes/libffi_msvc/ffi.c Modules/_ctypes/libffi_msvc/ffi.h
	Modules/_ctypes/libffi_msvc/ffi_common.h
	Modules/_ctypes/libffi_msvc/fficonfig.h
	Modules/_ctypes/libffi_msvc/ffitarget.h
	Modules/_ctypes/libffi_msvc/prep_cif.c
	Modules/_ctypes/libffi_msvc/types.c
	Modules/_ctypes/libffi_msvc/win32.c
	Modules/_ctypes/libffi_msvc/win64.asm Modules/_math.c
	Modules/_scproxy.c Modules/_ssl.c Modules/_testcapimodule.c
	Modules/_time.c Modules/_time.h Modules/audioop.c
	Modules/datetimemodule.c Modules/mathmodule.c
	Modules/parsermodule.c Modules/posixmodule.c
	Modules/selectmodule.c Modules/timemodule.c
	Modules/xxmodule.c Objects/abstract.c Objects/bytes_methods.c
	Objects/descrobject.c Objects/enumobject.c
	Objects/exceptions.c Objects/floatobject.c
	Objects/frameobject.c Objects/typeobject.c
	Objects/unicodeobject.c PC/VS8.0/pythoncore.vcproj
	PCbuild/_ctypes.vcproj PCbuild/pythoncore.vcproj
	Parser/Python.asdl Parser/asdl_c.py Python/Python-ast.c
	Python/_warnings.c Python/ast.c Python/ceval.c
	Python/compile.c Python/getargs.c Python/import.c
	Python/pythonrun.c Python/sysmodule.c Python/traceback.c
	README Tools/README Tools/scripts Tools/scripts/patchcheck.py
	Tools/webchecker configure setup.py
Message-ID: <20100706155529.F3BC4EEA13@mail.python.org>

Author: mark.dickinson
Date: Tue Jul  6 17:55:27 2010
New Revision: 82612

Log:
Merged revisions 82028,82034-82035,82037-82038,82041,82043,82045,82049,82051,82053,82055,82057,82059,82061,82068,82070-82071,82073,82076,82084,82089-82090,82113-82115,82119,82125,82128,82131,82138,82146,82148,82152,82156,82159,82162-82163,82166,82168,82171,82180-82181,82184-82186,82190,82193-82194,82200,82202,82204,82206,82208,82211,82214,82220,82223,82226,82228,82230-82232,82234,82238,82244,82247,82252,82255,82259,82262-82263,82269-82270,82273,82278,82281,82283,82286,82288-82302,82304-82316,82318,82320-82321,82326,82328,82331,82333,82336,82342-82343,82346-82347,82349,82351,82354,82357-82360,82365,82370-82371,82373,82375-82377,82381,82383-82384,82386-82387,82391-82394,82402,82405,82407,82410,82413,82416-82417,82422,82430,82434,82437-82438,82440,82442-82443,82448,82452,82456,82462,82464-82465,82467-82468,82471-82472,82478,82480-82481,82484-82485,82487-82488,82495,82507,82510,82514-82524,82526-82527,82533,82535,82537-82538,82540-82542,82547,82550-82553,82555,82557-82561,82567,82570,82573,82576-82578,82585,82587,82589,82591,82593-82595,82597,82599-82601,82604,82607,82609-82610 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

................
  r82028 | senthil.kumaran | 2010-06-16 18:44:57 +0100 (Wed, 16 Jun 2010) | 9 lines
  
  Merged revisions 82026 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82026 | senthil.kumaran | 2010-06-16 23:07:32 +0530 (Wed, 16 Jun 2010) | 3 lines
    
    Addressing RDM's review comments on the doc change.
  ........
................
  r82034 | alexander.belopolsky | 2010-06-16 23:38:15 +0100 (Wed, 16 Jun 2010) | 6 lines
  
  Issue #9012: "Separate compilation of time and datetime modules."
  Segregated code shared between time and datetime modules into
  Modules/_time.c.  Added a new header file, Modules/_time.h, which
  will be used instead of Include/timefuncs.h for declarations shared
  between time and datetime modules.
................
  r82035 | victor.stinner | 2010-06-17 00:05:06 +0100 (Thu, 17 Jun 2010) | 4 lines
  
  Issue #9012: Add _time.c and _time.h to the Visual Studio project file
  
  Add these new files, added by r82034, to the pythoncore project.
................
  r82037 | victor.stinner | 2010-06-17 00:33:54 +0100 (Thu, 17 Jun 2010) | 4 lines
  
  Issue #850997: mbcs encoding (Windows only) handles errors argument: strict
  mode raises unicode errors. The encoder only supports "strict" and "replace"
  error handlers, the decoder only supports "strict" and "ignore" error handlers.
................
  r82038 | victor.stinner | 2010-06-17 00:48:49 +0100 (Thu, 17 Jun 2010) | 2 lines
  
  Issue #850997: Oops, I forgot the author of the patch: Mark Hammond
................
  r82041 | r.david.murray | 2010-06-17 03:04:29 +0100 (Thu, 17 Jun 2010) | 16 lines
  
  Merged revisions 82039 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82039 | r.david.murray | 2010-06-16 21:36:52 -0400 (Wed, 16 Jun 2010) | 10 lines
    
    #8720: fix inspect regression by teaching getsourcefile about linecache.
    
    The fix for issue 4050 caused a regression:  before that fix, source
    lines in the linecache would eventually be found by inspect.  After the
    fix inspect reports an error earlier, and the source isn't found.
    The fix for the fix is to have getsourcefile look in the linecache for
    the file and return the psuedo-filename if the source is there, just as
    it already returns it if there is a PEP 302 loader.
  ........
................
  r82043 | mark.dickinson | 2010-06-17 13:33:22 +0100 (Thu, 17 Jun 2010) | 6 lines
  
  Issue #9011: Remove buggy and unnecessary ST->AST compilation code
  dealing with unary minus applied to a constant.  The removed code was
  mutating the ST, causing a second compilation to fail.  (The peephole
  optimizer already takes care of optimizing this case, so there's no
  lost optimization opportunity here.)
................
  r82045 | r.david.murray | 2010-06-17 14:23:18 +0100 (Thu, 17 Jun 2010) | 5 lines
  
  Don't use os.normcase when the result we are expecting is None.
  
  This worked fine on linux but fails on Windows.  That may or may
  not be a but in normcase.
................
  r82049 | senthil.kumaran | 2010-06-17 17:48:06 +0100 (Thu, 17 Jun 2010) | 9 lines
  
  Merged revisions 82047 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82047 | senthil.kumaran | 2010-06-17 22:08:34 +0530 (Thu, 17 Jun 2010) | 3 lines
    
    Fix Issue4452 - Incorrect docstring of os.setpgrp
  ........
................
  r82051 | mark.dickinson | 2010-06-17 19:24:52 +0100 (Thu, 17 Jun 2010) | 2 lines
  
  Add note about changes to the `round` function between 2.x and 3.x.
................
  r82053 | alexander.belopolsky | 2010-06-17 19:30:34 +0100 (Thu, 17 Jun 2010) | 2 lines
  
  Issue #6641: The datetime.strptime method now supports the %z directive.
................
  r82055 | barry.warsaw | 2010-06-17 19:38:20 +0100 (Thu, 17 Jun 2010) | 2 lines
  
  Typo repair.
................
  r82057 | victor.stinner | 2010-06-17 22:43:33 +0100 (Thu, 17 Jun 2010) | 2 lines
  
  Issue #8203: Fix IDLE Credits dialog: view_file() uses its encoding argument.
................
  r82059 | victor.stinner | 2010-06-18 00:08:50 +0100 (Fri, 18 Jun 2010) | 5 lines
  
  Issue #6543: Write the traceback in the terminal encoding instead of utf-8.
  Fix the encoding of the modules filename.
  
  Reindent also traceback.h, just because I hate tabs :-)
................
  r82061 | victor.stinner | 2010-06-18 00:17:37 +0100 (Fri, 18 Jun 2010) | 2 lines
  
  Issue #6543: Mention the author of the patch, Amaury Forgeot d'Arc
................
  r82068 | senthil.kumaran | 2010-06-18 16:08:18 +0100 (Fri, 18 Jun 2010) | 3 lines
  
  Fix Issue1368368 - prompt_user_passwd() in FancyURLopener masks 401 Unauthorized error page
................
  r82070 | alexander.belopolsky | 2010-06-18 17:22:00 +0100 (Fri, 18 Jun 2010) | 1 line
  
  Fixed a typo in a comment.
................
  r82071 | alexander.belopolsky | 2010-06-18 17:57:49 +0100 (Fri, 18 Jun 2010) | 1 line
  
  Added a new line at the end of the file.
................
  r82073 | alexander.belopolsky | 2010-06-18 19:44:37 +0100 (Fri, 18 Jun 2010) | 4 lines
  
  Issue #6641: Original commit for this issue, r82053, introduced a
  regression making datetime subclass' strptime return datetime rather
  than subclass instances.  Fixed this bug and a few typos.
................
  r82076 | jean-paul.calderone | 2010-06-18 21:03:54 +0100 (Fri, 18 Jun 2010) | 19 lines
  
  Merged revisions 82075 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82075 | jean-paul.calderone | 2010-06-18 16:00:17 -0400 (Fri, 18 Jun 2010) | 12 lines
    
    Revert r60115
    
    This revision introduced quoting for strings containing | based
    on a misunderstanding of the commonly used quoting rules used
    on Windows.
    
    | is interpreted by cmd.exe, not by the MS C runtime argv initializer.
    It only needs to be quoted if it is part of an argument passed through
    cmd.exe.
    
    See issue1300, issue7839, and issue8972.
  ........
................
  r82084 | victor.stinner | 2010-06-19 00:59:45 +0100 (Sat, 19 Jun 2010) | 6 lines
  
  Issue #8939: Improve arg.rst
  
   * Add :ctype: to Py_BEGIN_ALLOW_THREADS and int
   * "s" and "s#" formats of Py_BuildValue(): specify that the Python object type
     is str in the description
................
  r82089 | jean-paul.calderone | 2010-06-19 20:54:48 +0100 (Sat, 19 Jun 2010) | 1 line
  
  merge forward from the python 2.x branch
................
  r82090 | jean-paul.calderone | 2010-06-19 20:58:37 +0100 (Sat, 19 Jun 2010) | 1 line
  
  Revert r82089.  Commit was intended for a branch.
................
  r82113 | benjamin.peterson | 2010-06-20 16:25:14 +0100 (Sun, 20 Jun 2010) | 8 lines
  
  Blocked revisions 82112 via svnmerge
  
  ........
    r82112 | benjamin.peterson | 2010-06-20 10:12:04 -0500 (Sun, 20 Jun 2010) | 1 line
    
    update emacs section
  ........
................
  r82114 | benjamin.peterson | 2010-06-20 16:27:07 +0100 (Sun, 20 Jun 2010) | 1 line
  
  update release schedule url
................
  r82115 | benjamin.peterson | 2010-06-20 16:27:42 +0100 (Sun, 20 Jun 2010) | 1 line
  
  add copyright years
................
  r82119 | mark.dickinson | 2010-06-20 21:01:04 +0100 (Sun, 20 Jun 2010) | 9 lines
  
  Merged revisions 82117 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82117 | mark.dickinson | 2010-06-20 19:50:19 +0100 (Sun, 20 Jun 2010) | 1 line
    
    Merge test_strtod and test_float string-to-float conversion tests.
  ........
................
  r82125 | brett.cannon | 2010-06-21 03:49:35 +0100 (Mon, 21 Jun 2010) | 2 lines
  
  Move over to assertIs.
................
  r82128 | alexander.belopolsky | 2010-06-21 16:21:14 +0100 (Mon, 21 Jun 2010) | 2 lines
  
  Issue #9005: Prevent utctimetuple() from producing year 0 or year 10,000.
................
  r82131 | benjamin.peterson | 2010-06-21 16:37:16 +0100 (Mon, 21 Jun 2010) | 9 lines
  
  Merged revisions 82130 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82130 | benjamin.peterson | 2010-06-21 10:27:46 -0500 (Mon, 21 Jun 2010) | 1 line
    
    fix finding visual studio 2008 on 64 bit #8854
  ........
................
  r82138 | thomas.heller | 2010-06-21 17:00:31 +0100 (Mon, 21 Jun 2010) | 13 lines
  
  Merged revisions 82126-82127 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82126 | thomas.heller | 2010-06-21 16:00:24 +0200 (Mo, 21 Jun 2010) | 1 line
    
    Fix #8959 by reverting revision 80761.
  ........
    r82127 | thomas.heller | 2010-06-21 17:01:18 +0200 (Mo, 21 Jun 2010) | 2 lines
    
    Add tests for problems reported in issue 8959.
  ........
................
  r82146 | senthil.kumaran | 2010-06-22 03:42:52 +0100 (Tue, 22 Jun 2010) | 3 lines
  
  Minor docs issue.
................
  r82148 | senthil.kumaran | 2010-06-22 03:57:23 +0100 (Tue, 22 Jun 2010) | 3 lines
  
  Minor comment formatting.
................
  r82152 | alexander.belopolsky | 2010-06-22 15:07:33 +0100 (Tue, 22 Jun 2010) | 1 line
  
  Added more test cases
................
  r82156 | benjamin.peterson | 2010-06-22 19:11:34 +0100 (Tue, 22 Jun 2010) | 10 lines
  
  Blocked revisions 82155 via svnmerge
  
  ........
    r82155 | benjamin.peterson | 2010-06-22 13:09:02 -0500 (Tue, 22 Jun 2010) | 4 lines
    
    keep UserDict an old-style class
    
    Be generous in abc.py to allow this.
  ........
................
  r82159 | benjamin.peterson | 2010-06-22 20:21:52 +0100 (Tue, 22 Jun 2010) | 9 lines
  
  Merged revisions 82157 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82157 | benjamin.peterson | 2010-06-22 14:16:37 -0500 (Tue, 22 Jun 2010) | 1 line
    
    remove INT_MAX assertions; they can fail with large Py_ssize_t #9058
  ........
................
  r82162 | benjamin.peterson | 2010-06-22 20:52:02 +0100 (Tue, 22 Jun 2010) | 8 lines
  
  Blocked revisions 82161 via svnmerge
  
  ........
    r82161 | benjamin.peterson | 2010-06-22 14:49:47 -0500 (Tue, 22 Jun 2010) | 1 line
    
    bump revision
  ........
................
  r82163 | benjamin.peterson | 2010-06-22 21:02:39 +0100 (Tue, 22 Jun 2010) | 9 lines
  
  Merged revisions 82160 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82160 | benjamin.peterson | 2010-06-22 14:45:51 -0500 (Tue, 22 Jun 2010) | 1 line
    
    spacing nit; this isn't C
  ........
................
  r82166 | benjamin.peterson | 2010-06-22 21:29:32 +0100 (Tue, 22 Jun 2010) | 9 lines
  
  Merged revisions 82165 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82165 | benjamin.peterson | 2010-06-22 15:26:20 -0500 (Tue, 22 Jun 2010) | 1 line
    
    must force gc here
  ........
................
  r82168 | benjamin.peterson | 2010-06-22 21:34:34 +0100 (Tue, 22 Jun 2010) | 9 lines
  
  Merged revisions 82167 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82167 | benjamin.peterson | 2010-06-22 15:32:02 -0500 (Tue, 22 Jun 2010) | 1 line
    
    mark ref counting as impl detail
  ........
................
  r82171 | antoine.pitrou | 2010-06-22 22:49:39 +0100 (Tue, 22 Jun 2010) | 10 lines
  
  Merged revisions 82169 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82169 | antoine.pitrou | 2010-06-22 23:42:05 +0200 (mar., 22 juin 2010) | 4 lines
    
    Fix misindents in compile.c (for Benjamin).
    Of course, whoever used the wrong indentation rules needs to be spanked.
  ........
................
  r82180 | stefan.krah | 2010-06-23 19:35:33 +0100 (Wed, 23 Jun 2010) | 8 lines
  
  Blocked revisions 82177 via svnmerge
  
  ........
    r82177 | stefan.krah | 2010-06-23 20:12:09 +0200 (Wed, 23 Jun 2010) | 3 lines
    
    Issue #8930: Remaining indentation fixes after the Grand Unified Indenting.
  ........
................
  r82181 | stefan.krah | 2010-06-23 19:42:39 +0100 (Wed, 23 Jun 2010) | 3 lines
  
  Issue #8930: Remaining indentation fixes after the Grand Unified Indenting.
................
  r82184 | alexander.belopolsky | 2010-06-23 22:40:15 +0100 (Wed, 23 Jun 2010) | 1 line
  
  Issue #9051: Instances of timezone class can now be pickled.
................
  r82185 | alexander.belopolsky | 2010-06-23 23:29:48 +0100 (Wed, 23 Jun 2010) | 1 line
  
  Test future pickle protocols.  Thanks Antoine Pitrou for suggestion.
................
  r82186 | alexander.belopolsky | 2010-06-23 23:58:49 +0100 (Wed, 23 Jun 2010) | 1 line
  
  Added more test cases
................
  r82190 | benjamin.peterson | 2010-06-24 01:17:03 +0100 (Thu, 24 Jun 2010) | 9 lines
  
  Merged revisions 82189 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82189 | benjamin.peterson | 2010-06-23 19:12:40 -0500 (Wed, 23 Jun 2010) | 1 line
    
    prevent assignment to set literals
  ........
................
  r82193 | stefan.krah | 2010-06-24 11:22:10 +0100 (Thu, 24 Jun 2010) | 10 lines
  
  Blocked revisions 82191 via svnmerge
  
  ........
    r82191 | stefan.krah | 2010-06-24 11:33:05 +0200 (Thu, 24 Jun 2010) | 5 lines
    
    Issue #9020: The Py_IS* macros from pyctype.h should generally only be
    used with signed/unsigned char arguments. For integer arguments, EOF
    has to be handled separately.
  ........
................
  r82194 | stefan.krah | 2010-06-24 11:25:08 +0100 (Thu, 24 Jun 2010) | 3 lines
  
  Add specification for the Py_IS* macros in pyctype.h.
................
  r82200 | victor.stinner | 2010-06-24 23:08:25 +0100 (Thu, 24 Jun 2010) | 3 lines
  
  Issue #8949: "z" format of PyArg_Parse*() functions doesn't accept bytes
  objects, as described in the documentation.
................
  r82202 | victor.stinner | 2010-06-24 23:31:12 +0100 (Thu, 24 Jun 2010) | 2 lines
  
  PyArg_Parse*() functions: factorize code for s/z and u/Z formats
................
  r82204 | antoine.pitrou | 2010-06-24 23:34:04 +0100 (Thu, 24 Jun 2010) | 5 lines
  
  Issue #8682: The ssl module now temporary increments the reference count of
  a socket object got through `PyWeakref_GetObject`, so as to avoid possible
  deallocation while the object is still being used.
................
  r82206 | victor.stinner | 2010-06-24 23:57:10 +0100 (Thu, 24 Jun 2010) | 2 lines
  
  getbuffer(): release the buffer on error (if the buffer is not contiguous)
................
  r82208 | victor.stinner | 2010-06-25 01:02:38 +0100 (Fri, 25 Jun 2010) | 3 lines
  
  Issue #8850: Remove "w" and "w#" formats from PyArg_Parse*() functions, use
  "w*" format instead. Add tests for "w*" format.
................
  r82211 | antoine.pitrou | 2010-06-25 01:07:34 +0100 (Fri, 25 Jun 2010) | 10 lines
  
  Merged revisions 82210 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82210 | antoine.pitrou | 2010-06-25 02:03:21 +0200 (ven., 25 juin 2010) | 4 lines
    
    Issue #9075: In the ssl module, remove the setting of a `debug` flag
    on an OpenSSL structure.
  ........
................
  r82214 | ezio.melotti | 2010-06-25 11:56:11 +0100 (Fri, 25 Jun 2010) | 1 line
  
  #9018: os.path.normcase() now raises a TypeError if the argument is not str or bytes.
................
  r82220 | benjamin.peterson | 2010-06-25 20:30:21 +0100 (Fri, 25 Jun 2010) | 1 line
  
  only take into account positional arguments count in related error messages
................
  r82223 | mark.dickinson | 2010-06-25 21:22:24 +0100 (Fri, 25 Jun 2010) | 9 lines
  
  Merged revisions 82221 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82221 | mark.dickinson | 2010-06-25 21:19:48 +0100 (Fri, 25 Jun 2010) | 1 line
    
    Fix indentation of Python code example in C comment.
  ........
................
  r82226 | benjamin.peterson | 2010-06-25 21:37:22 +0100 (Fri, 25 Jun 2010) | 9 lines
  
  Merged revisions 82225 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82225 | benjamin.peterson | 2010-06-25 15:34:01 -0500 (Fri, 25 Jun 2010) | 1 line
    
    mark implementation detail as such
  ........
................
  r82228 | benjamin.peterson | 2010-06-25 22:24:10 +0100 (Fri, 25 Jun 2010) | 8 lines
  
  Blocked revisions 82227 via svnmerge
  
  ........
    r82227 | benjamin.peterson | 2010-06-25 16:19:04 -0500 (Fri, 25 Jun 2010) | 1 line
    
    this must be a typo
  ........
................
  r82230 | benjamin.peterson | 2010-06-26 00:02:11 +0100 (Sat, 26 Jun 2010) | 9 lines
  
  Merged revisions 82229 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82229 | benjamin.peterson | 2010-06-25 17:58:47 -0500 (Fri, 25 Jun 2010) | 1 line
    
    mark implementation detail as such
  ........
................
  r82231 | benjamin.peterson | 2010-06-26 00:24:35 +0100 (Sat, 26 Jun 2010) | 1 line
  
  fix typos
................
  r82232 | benjamin.peterson | 2010-06-26 00:37:15 +0100 (Sat, 26 Jun 2010) | 1 line
  
  remove old, outdated tool
................
  r82234 | r.david.murray | 2010-06-26 01:17:12 +0100 (Sat, 26 Jun 2010) | 9 lines
  
  Merged revisions 82233 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82233 | r.david.murray | 2010-06-25 20:06:44 -0400 (Fri, 25 Jun 2010) | 2 lines
    
    #4640: add a test to optparse that proves issue is invalid.
  ........
................
  r82238 | alexander.belopolsky | 2010-06-26 03:15:07 +0100 (Sat, 26 Jun 2010) | 11 lines
  
  Merged revisions 82236 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82236 | alexander.belopolsky | 2010-06-25 22:05:19 -0400 (Fri, 25 Jun 2010) | 4 lines
    
    Issue #9024: Fixed mark-up for PyDateTime_IMPORT macro and added that
    it is typically used in module initialization function. Thanks Tim
    Golden for the patch.
  ........
................
  r82244 | r.david.murray | 2010-06-26 04:34:33 +0100 (Sat, 26 Jun 2010) | 9 lines
  
  Merged revisions 82242 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82242 | r.david.murray | 2010-06-25 23:27:32 -0400 (Fri, 25 Jun 2010) | 2 lines
    
    Fix indentation in recently added test.
  ........
................
  r82247 | r.david.murray | 2010-06-26 19:44:14 +0100 (Sat, 26 Jun 2010) | 14 lines
  
  Blocked revisions 82246 via svnmerge
  
  ........
    r82246 | r.david.murray | 2010-06-26 14:39:50 -0400 (Sat, 26 Jun 2010) | 8 lines
    
    #9085: email versions have gotten out of sync, 2.7 is actually 4.0.3.
    
    In 2.5 the email version was bumped to 4.0.2 just before release but
    after the maintenance branch was set up.  The bump was not backported
    to trunk, and 2.6 went out the door with a version number of 4.0.1.
    I bumped this to 4.0.2 because of a behavior change, but that makes
    it look the same as 2.5.  So bump it again to 4.0.3.
  ........
................
  r82252 | ezio.melotti | 2010-06-26 19:50:39 +0100 (Sat, 26 Jun 2010) | 9 lines
  
  Merged revisions 82248 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82248 | ezio.melotti | 2010-06-26 21:44:42 +0300 (Sat, 26 Jun 2010) | 1 line
    
    Fix extra space.
  ........
................
  r82255 | alexander.belopolsky | 2010-06-26 19:57:02 +0100 (Sat, 26 Jun 2010) | 9 lines
  
  Merged revisions 82251 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82251 | alexander.belopolsky | 2010-06-26 14:49:10 -0400 (Sat, 26 Jun 2010) | 1 line
    
    Issue 9024: Added :cdata: tag to PyDateTimeAPI
  ........
................
  r82259 | brett.cannon | 2010-06-26 23:29:06 +0100 (Sat, 26 Jun 2010) | 6 lines
  
  Having CFLAGS come before OPT prevents silencing certain classes of warnings
  from the fact that OPT contains -Wall be default. This is annoying when
  compilers like clang have thorough debugging information about things that
  Python does extensively (e.g. -Wunused-value for unused return values caused by
  a macro use).
................
  r82262 | georg.brandl | 2010-06-27 11:17:12 +0100 (Sun, 27 Jun 2010) | 1 line
  
  #9078: fix some Unicode C API descriptions, in comments and docs.
................
  r82263 | georg.brandl | 2010-06-27 11:37:48 +0100 (Sun, 27 Jun 2010) | 1 line
  
  #9064: accept number of frames for "up" and "down" commands in pdb.
................
  r82269 | georg.brandl | 2010-06-27 11:59:19 +0100 (Sun, 27 Jun 2010) | 1 line
  
  Wording fix.
................
  r82270 | georg.brandl | 2010-06-27 12:24:03 +0100 (Sun, 27 Jun 2010) | 1 line
  
  Untabify.
................
  r82273 | ronald.oussoren | 2010-06-27 13:37:46 +0100 (Sun, 27 Jun 2010) | 15 lines
  
  Merged revisions 82272 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82272 | ronald.oussoren | 2010-06-27 14:36:16 +0200 (Sun, 27 Jun 2010) | 8 lines
    
    Two small fixes for the support for SDKs on MacOSX:
    
    1) The code that checks if an path should be located in the SDK
       explicitly excludes /usr/local. This fixes issue9046
    
    2) The SDK variant for filtering "db_dirs_to_check" in setup.py
       was not doing anything because of a missing assignment.
  ........
................
  r82278 | ronald.oussoren | 2010-06-27 13:45:47 +0100 (Sun, 27 Jun 2010) | 12 lines
  
  Merged revisions 82276 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82276 | ronald.oussoren | 2010-06-27 14:43:47 +0200 (Sun, 27 Jun 2010) | 5 lines
    
    The uuid library on OSX 10.5 seems to contain the same bug as the one
    on OSX 10.6, therefore don't use it there either.
    
    This fixes issue8621.
  ........
................
  r82281 | ronald.oussoren | 2010-06-27 13:51:31 +0100 (Sun, 27 Jun 2010) | 19 lines
  
  Merged revisions 82150 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82150 | ronald.oussoren | 2010-06-22 11:32:22 +0200 (Tue, 22 Jun 2010) | 12 lines
    
    The code in _scproxy (a mac specific helper module to 
    detect proxy settings) had the wrong logic for detecting
    if the checkbox 'Exclude simple hostnames' is checked. This
    checkin fixes that.
    
    As a result the test failure 'Issue8455' goes away on systems
    where the checkbox is not checked.
    
    I'm carefully avoiding saying that is fixes that issue,
    test_urllib2_localnet assumes that system proxy settings are
    empty (not just on OSX, see Issue8455 for details).
  ........
................
  r82283 | georg.brandl | 2010-06-27 14:28:06 +0100 (Sun, 27 Jun 2010) | 8 lines
  
  Blocked revisions 82271 via svnmerge
  
  ........
    r82271 | georg.brandl | 2010-06-27 13:24:15 +0200 (So, 27 Jun 2010) | 1 line
    
    Untabify.
  ........
................
  r82286 | ronald.oussoren | 2010-06-27 15:26:30 +0100 (Sun, 27 Jun 2010) | 11 lines
  
  Merged revisions 82284 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82284 | ronald.oussoren | 2010-06-27 15:59:39 +0200 (Sun, 27 Jun 2010) | 4 lines
    
    Fix for Issue8883: without this patch test_urllib will fail
    when there is a bare IP address in the "Bypass proxy settings
    for these Hosts & Domains" list on MacOSX.
  ........
................
  r82288 | mark.dickinson | 2010-06-27 18:36:54 +0100 (Sun, 27 Jun 2010) | 1 line
  
  Issue #9089: PyNumber_Int is no more.  Remove it from the docs.
................
  r82289 | mark.dickinson | 2010-06-27 19:18:01 +0100 (Sun, 27 Jun 2010) | 1 line
  
  Issue #9089: Remove intobject.h from MSVC project files.
................
  r82290 | mark.dickinson | 2010-06-27 19:19:09 +0100 (Sun, 27 Jun 2010) | 1 line
  
  Issue #9089: Remove references to intobject.c and intobject.h from comments.
................
  r82291 | brett.cannon | 2010-06-27 21:40:18 +0100 (Sun, 27 Jun 2010) | 1 line
  
  Revert r82259; still no good way to override instead of replace OPT, but changing order alters pydebug optimization flags.
................
  r82292 | benjamin.peterson | 2010-06-27 21:50:37 +0100 (Sun, 27 Jun 2010) | 451 lines
  
  Blocked revisions 79620,79624-79625,79633-79637,79639-79640,79650-79651,79654,79673,79693,79695,79697-79698,79704,79706-79707,79710,79712-79714,79718-79719,79722,79730,79744,79750,79755,79760,79769,79808,79811,79890,80062,80065,80082,80086,80171,80189,80242-80243,80374,80411-80412,80445,80579-80581,80700,80737,80745,80752,80794,80813-80814,80823,80963,80982,81012-81013,81030,81035,81088,81140-81141,81149-81151,81275,81484,81502,81517,81561,81705,81847,81919,81964,82056,82136-82137,82142,82149 via svnmerge
  
  ........
    r79620 | brian.curtin | 2010-04-02 16:18:14 -0500 (Fri, 02 Apr 2010) | 9 lines
    
    Implement #7347. Add CreateKeyEx, DeleteKeyEx, and update _winreg tests.
    
    *ReflectionKey functions used to not be documented or tested, but they are
    now sufficiently documented and tested on platforms where they apply. 
    Additionally, fixed a bug in QueryReflectionKey which was returning an
    incorrect value.
    
    All tests pass from XP through Windows 7, on 32 and 64-bit platforms.
  ........
    r79624 | ezio.melotti | 2010-04-02 16:43:10 -0500 (Fri, 02 Apr 2010) | 1 line
    
    Fix test_compiler.py that was using unittest.__file__ to find Lib/ (unittest is now a package).
  ........
    r79625 | brian.curtin | 2010-04-02 16:51:37 -0500 (Fri, 02 Apr 2010) | 2 lines
    
    Add a line about #7347 to Misc\News
  ........
    r79633 | brian.curtin | 2010-04-02 18:26:06 -0500 (Fri, 02 Apr 2010) | 14 lines
    
    Implement #1220212. Add os.kill support for Windows.
    
    os.kill takes one of two newly added signals, CTRL_C_EVENT and 
    CTRL_BREAK_EVENT, or any integer value. The events are a special case
    which work with subprocess console applications which implement a 
    special console control handler. Any other value but those two will 
    cause os.kill to use TerminateProcess, outright killing the process.
    
    This change adds win_console_handler.py, which is a script to implement 
    SetConsoleCtrlHandler and applicable handler function, using ctypes.
    
    subprocess also gets another attribute which is a necessary flag to 
    creationflags in Popen in order to send the CTRL events.
  ........
    r79634 | brian.curtin | 2010-04-02 18:31:28 -0500 (Fri, 02 Apr 2010) | 2 lines
    
    Add note about #1220212 (os.kill on Windows)
  ........
    r79635 | benjamin.peterson | 2010-04-02 18:31:40 -0500 (Fri, 02 Apr 2010) | 1 line
    
    set svn:eol-style on new file
  ........
    r79636 | benjamin.peterson | 2010-04-02 18:59:41 -0500 (Fri, 02 Apr 2010) | 4 lines
    
    always check _PyString_Resize for error
    
    also normalize how this error is checked
  ........
    r79637 | raymond.hettinger | 2010-04-02 19:39:26 -0500 (Fri, 02 Apr 2010) | 1 line
    
    Clear cyclical references in list based OrderedDict.
  ........
    r79639 | benjamin.peterson | 2010-04-02 19:57:33 -0500 (Fri, 02 Apr 2010) | 1 line
    
    more _PyString_Resize error checking
  ........
    r79640 | brian.curtin | 2010-04-02 19:59:32 -0500 (Fri, 02 Apr 2010) | 3 lines
    
    Fix assertRaises usage on reflection functions which should raise
    NotImplementedError on Windows XP and below.
  ........
    r79650 | raymond.hettinger | 2010-04-02 22:14:28 -0500 (Fri, 02 Apr 2010) | 1 line
    
    Improve clear() method.  Keeps key/value refcnts >= 1 until final dict.clear() so that decrefs to zero won't trigger arbitrary code .  Also runs a bit faster.
  ........
    r79651 | raymond.hettinger | 2010-04-03 02:57:09 -0500 (Sat, 03 Apr 2010) | 1 line
    
    Factor-out constant expressions
  ........
    r79654 | victor.stinner | 2010-04-03 03:40:16 -0500 (Sat, 03 Apr 2010) | 5 lines
    
    Issue #8227: Fix C API documentation, argument parsing
    
     * 'z', 'z#', 'z*' does also accept Unicode
     * unify types name: replace "string or Unicode objet" by "string or Unicode"
  ........
    r79673 | brian.curtin | 2010-04-03 08:58:39 -0500 (Sat, 03 Apr 2010) | 2 lines
    
    Add missing return statement in an error condition.
  ........
    r79693 | benjamin.peterson | 2010-04-03 10:38:38 -0500 (Sat, 03 Apr 2010) | 1 line
    
    wrap
  ........
    r79695 | benjamin.peterson | 2010-04-03 10:40:29 -0500 (Sat, 03 Apr 2010) | 1 line
    
    remove unneeded argument
  ........
    r79697 | benjamin.peterson | 2010-04-03 10:44:56 -0500 (Sat, 03 Apr 2010) | 1 line
    
    silence PyCObject warnings in bsddb
  ........
    r79698 | benjamin.peterson | 2010-04-03 10:45:59 -0500 (Sat, 03 Apr 2010) | 1 line
    
    spelling
  ........
    r79704 | benjamin.peterson | 2010-04-03 10:58:15 -0500 (Sat, 03 Apr 2010) | 1 line
    
    remove deprecation warnings silence attempting
  ........
    r79706 | benjamin.peterson | 2010-04-03 11:06:42 -0500 (Sat, 03 Apr 2010) | 1 line
    
    stop CObject deprecation warnings in test___all__
  ........
    r79707 | mark.dickinson | 2010-04-03 11:41:20 -0500 (Sat, 03 Apr 2010) | 5 lines
    
    Ensure 'module removed' warning messages contain the word 'module' or 'package'.
    
    This should fix the test_py3kwarn failure on OS X.  test_support.import_module
    also requires this.
  ........
    r79710 | mark.dickinson | 2010-04-03 11:54:02 -0500 (Sat, 03 Apr 2010) | 1 line
    
    Replace backquotes with repr(), to silence a SyntaxWarning.
  ........
    r79712 | raymond.hettinger | 2010-04-03 12:10:05 -0500 (Sat, 03 Apr 2010) | 1 line
    
    Silence a compiler warning.
  ........
    r79713 | raymond.hettinger | 2010-04-03 13:10:37 -0500 (Sat, 03 Apr 2010) | 1 line
    
    Add count() method to collections.deque().
  ........
    r79714 | mark.dickinson | 2010-04-03 13:17:54 -0500 (Sat, 03 Apr 2010) | 1 line
    
    Silence DeprecationWarnings from uses of has_key and <> in plat-mac.
  ........
    r79718 | antoine.pitrou | 2010-04-03 16:42:18 -0500 (Sat, 03 Apr 2010) | 3 lines
    
    Remove useless (?) import from r79706
  ........
    r79719 | benjamin.peterson | 2010-04-03 16:50:40 -0500 (Sat, 03 Apr 2010) | 1 line
    
    import bsddb more robustly
  ........
    r79722 | raymond.hettinger | 2010-04-03 17:34:15 -0500 (Sat, 03 Apr 2010) | 1 line
    
    Expand test coverage for deque.count().
  ........
    r79730 | raymond.hettinger | 2010-04-03 20:24:59 -0500 (Sat, 03 Apr 2010) | 1 line
    
    Issue 5479:  Add functools.total_ordering class decorator.
  ........
    r79744 | raymond.hettinger | 2010-04-04 02:33:46 -0500 (Sun, 04 Apr 2010) | 1 line
    
    Documentation nit
  ........
    r79750 | raymond.hettinger | 2010-04-04 13:34:45 -0500 (Sun, 04 Apr 2010) | 1 line
    
    Add functools.CmpToKey()
  ........
    r79755 | raymond.hettinger | 2010-04-04 16:45:01 -0500 (Sun, 04 Apr 2010) | 6 lines
    
    Add tests for cmp_to_key.
    Adopt PEP 8 compliant function name.
    Factor-out existing uses cmp_to_key.
    Update documentation to use internal pointers instead of external resource.
  ........
    r79760 | raymond.hettinger | 2010-04-04 17:24:03 -0500 (Sun, 04 Apr 2010) | 1 line
    
    Add tests for functools.total_ordering.
  ........
    r79769 | benjamin.peterson | 2010-04-04 18:23:22 -0500 (Sun, 04 Apr 2010) | 1 line
    
    fix dis on new style classes #8310
  ........
    r79808 | raymond.hettinger | 2010-04-05 13:53:43 -0500 (Mon, 05 Apr 2010) | 1 line
    
    Classes that override __eq__ also need to define __hash__.
  ........
    r79811 | brian.curtin | 2010-04-05 14:04:23 -0500 (Mon, 05 Apr 2010) | 7 lines
    
    Fix a failing test on an apparently slow Windows buildbot.
    
    On slower Windows machines, waiting 0.1 seconds can sometimes not be 
    enough for a subprocess to start and be ready to accept signals, causing 
    the test to fail. One buildbot is also choking on input()/EOFError so 
    that was changed to not depend on input.
  ........
    r79890 | mark.dickinson | 2010-04-07 05:18:27 -0500 (Wed, 07 Apr 2010) | 4 lines
    
    Use some more interesting test values for (unsigned) long long ctypes tests,
    in the hope of getting more information about the test_ctypes failures
    on Sparc (see issue #8314).
  ........
    r80062 | r.david.murray | 2010-04-13 15:57:40 -0500 (Tue, 13 Apr 2010) | 2 lines
    
    Issue #5277: Fix quote counting when parsing RFC 2231 encoded parameters.
  ........
    r80065 | brian.curtin | 2010-04-13 21:24:24 -0500 (Tue, 13 Apr 2010) | 6 lines
    
    Add a small sleep to let a subprocess start before terminating it.
    
    David Bolen's buildbot isn't know for it's speed, and it seems that we
    may have been trying to kill the subprocess before it was fully 
    initialized. I ran with this change on the bot itself and it seemed to work.
  ........
    r80082 | georg.brandl | 2010-04-14 16:36:49 -0500 (Wed, 14 Apr 2010) | 1 line
    
    #8370: fix module name in backported doc addition.
  ........
    r80086 | brian.curtin | 2010-04-14 19:40:40 -0500 (Wed, 14 Apr 2010) | 9 lines
    
    Fix os.kill tests to be more robust and work with slower machines.
    
    Rather than depending on some sleep value, start up an interpreter
    as a subprocess and communicate with it. Because subprocess pipes 
    can't be read from until EOF and I want to read from them before that,
    use ctypes to peek by using PeekNamedPipe. Once the subprocess has 
    written the message, then it is ready to roll and accept signals.
    After that, kill it.
  ........
    r80171 | antoine.pitrou | 2010-04-18 06:16:24 -0500 (Sun, 18 Apr 2010) | 3 lines
    
    Fix bootstrap after r80166
  ........
    r80189 | victor.stinner | 2010-04-18 13:22:25 -0500 (Sun, 18 Apr 2010) | 1 line
    
    Revert r80166 (and r80171), restore Lib/platform.py. subprocess cannot be used in platform.py
  ........
    r80242 | ronald.oussoren | 2010-04-20 03:53:12 -0500 (Tue, 20 Apr 2010) | 8 lines
    
    The PythonLauncher change is needed due
    to changes in how the BASECFLAGS and CFLAGS
    variables get filled by configure.
    
    The Mac/Makefile.in change ensures that 
    pythonw gets build with the rigth deployment
    targets.
  ........
    r80243 | ronald.oussoren | 2010-04-20 03:54:48 -0500 (Tue, 20 Apr 2010) | 3 lines
    
    This patch fixes the handling of a weak-linked
    variable and should fix issue #8095.
  ........
    r80374 | antoine.pitrou | 2010-04-22 12:44:41 -0500 (Thu, 22 Apr 2010) | 3 lines
    
    Temporarily display OpenSSL version in verbose run -- trying to diagnose failure on Neal's buildbot.
  ........
    r80411 | florent.xicluna | 2010-04-23 12:59:10 -0500 (Fri, 23 Apr 2010) | 2 lines
    
    Remove ImportWarnings filters.  They become obsolete after r79310, issue #8205.
  ........
    r80412 | florent.xicluna | 2010-04-23 13:10:12 -0500 (Fri, 23 Apr 2010) | 2 lines
    
    Fix the "regrtest -s" switch.
  ........
    r80445 | brian.curtin | 2010-04-24 12:10:22 -0500 (Sat, 24 Apr 2010) | 2 lines
    
    Fix #5774. Some _winreg functions are documented as taking kwargs but don't.
  ........
    r80579 | nick.coghlan | 2010-04-28 09:34:30 -0500 (Wed, 28 Apr 2010) | 1 line
    
    Also mention patch submitter's name in NEWS, not just in the commit message
  ........
    r80580 | nick.coghlan | 2010-04-28 09:51:08 -0500 (Wed, 28 Apr 2010) | 1 line
    
    Issue 8202: when using the -m command line switch, sys.argv[0] is now '-m' instead of '-c' while searching for the module to be executed
  ........
    r80581 | nick.coghlan | 2010-04-28 09:53:59 -0500 (Wed, 28 Apr 2010) | 1 line
    
    Add version changed note for -m tinkering with sys.argv[0] during the search process
  ........
    r80700 | ronald.oussoren | 2010-05-02 04:55:57 -0500 (Sun, 02 May 2010) | 3 lines
    
    Small update to r80698 to ensure that webbrowser.open
    uses the default browser.
  ........
    r80737 | brett.cannon | 2010-05-03 18:57:15 -0500 (Mon, 03 May 2010) | 4 lines
    
    Fix two potential uninitialization errors and an unneeded assignment.
    
    Found using Clang's static analyzer.
  ........
    r80745 | brett.cannon | 2010-05-03 20:04:53 -0500 (Mon, 03 May 2010) | 5 lines
    
    Mention the code clean-up thanks to Clang's static analyzer in Modules.
    
    Was not applied to modules that will not compile under OS X, dbmmodule.c,
    getaddrinfo.c, and getnameinfo.c.
  ........
    r80752 | victor.stinner | 2010-05-04 06:35:36 -0500 (Tue, 04 May 2010) | 4 lines
    
    _pyio: Fix TextIOWrapper constructor: os has no device_encoding() function
    
    _io module doesn't call this function which was introduced in Python3.
  ........
    r80794 | barry.warsaw | 2010-05-05 11:17:22 -0500 (Wed, 05 May 2010) | 1 line
    
    NEWS
  ........
    r80813 | brett.cannon | 2010-05-05 15:20:19 -0500 (Wed, 05 May 2010) | 4 lines
    
    Remove three unneeded variable assignments.
    
    Found using Clang's static analyzer.
  ........
    r80814 | brett.cannon | 2010-05-05 15:24:30 -0500 (Wed, 05 May 2010) | 1 line
    
    Partially revert the over-reaching r80813.
  ........
    r80823 | brett.cannon | 2010-05-05 15:54:53 -0500 (Wed, 05 May 2010) | 1 line
    
    Mention how Clang's static anaylzer was run over Objects/ and Python/.
  ........
    r80963 | ronald.oussoren | 2010-05-08 03:44:37 -0500 (Sat, 08 May 2010) | 4 lines
    
    Fix for issue #7724: make it possible to build using
    the OSX 10.4u SDK on MacOSX 10.6 by honoring the specified
    SDK when looking for files.
  ........
    r80982 | antoine.pitrou | 2010-05-08 10:23:57 -0500 (Sat, 08 May 2010) | 3 lines
    
    Revert r80963 - it broke compilation everywhere
  ........
    r81012 | gregory.p.smith | 2010-05-08 18:38:49 -0500 (Sat, 08 May 2010) | 2 lines
    
    Fixes [issue7245] Better Ctrl-C support in pdb.
  ........
    r81013 | gregory.p.smith | 2010-05-08 20:20:20 -0500 (Sat, 08 May 2010) | 5 lines
    
    Revert r81012.  buildbot problems and its questionable of me to even
    add this to trunk while we're on the way to 2.7rc1.
    
    When fixed this can go into py3k first.  Sorry.
  ........
    r81030 | antoine.pitrou | 2010-05-09 10:15:11 -0500 (Sun, 09 May 2010) | 3 lines
    
    Fixup indentation of PC/_msi.c
  ........
    r81035 | antoine.pitrou | 2010-05-09 13:32:48 -0500 (Sun, 09 May 2010) | 3 lines
    
    Remove spurious newlines, and add version number.
  ........
    r81088 | brian.curtin | 2010-05-11 14:13:13 -0500 (Tue, 11 May 2010) | 13 lines
    
    #8575 - Update and reorganize some _winreg contents.
    
    I've removed the hopeful note about a future higher-level module since
    it's been in there for quite a long time and nothing of the sort has 
    come up. There are a few places where markup was added to cross-reference
    other sections, and many of the external links have been removed and now
    point to newly created sections containing previously undocumented
    information.
    
    The Value Types section was created and it's contents were taken from
    a function-specific area, since it applies to more than just that
    function. It fits in better with the other newly documented constants.
  ........
    r81140 | florent.xicluna | 2010-05-13 12:05:29 -0500 (Thu, 13 May 2010) | 3 lines
    
    Add sensible information about the OS X platform to diagnose issue #8423:
    test_pep277 fails on "x86 Tiger" buildbot but not on "PPC Tiger".
  ........
    r81141 | florent.xicluna | 2010-05-13 13:16:06 -0500 (Thu, 13 May 2010) | 2 lines
    
    Revert the additional OS X information (r81140).  Keep the endianness information.
  ........
    r81149 | florent.xicluna | 2010-05-13 16:40:01 -0500 (Thu, 13 May 2010) | 2 lines
    
    Better test skipping, with message in the log.
  ........
    r81150 | florent.xicluna | 2010-05-13 16:41:05 -0500 (Thu, 13 May 2010) | 2 lines
    
    Improve test feedback to troubleshoot issue #8423 on OS X.
  ........
    r81151 | florent.xicluna | 2010-05-13 18:46:48 -0500 (Thu, 13 May 2010) | 2 lines
    
    Revert changeset r81150 which helped diagnose issue #8423 on some OS X buildbot.
  ........
    r81275 | antoine.pitrou | 2010-05-17 14:56:59 -0500 (Mon, 17 May 2010) | 4 lines
    
    Issue #7079: Fix a possible crash when closing a file object while using
    it from another thread.  Patch by Daniel Stutzbach.
  ........
    r81484 | georg.brandl | 2010-05-22 16:26:21 -0500 (Sat, 22 May 2010) | 1 line
    
    NEWS rewrap and punctuation consistency.
  ........
    r81502 | georg.brandl | 2010-05-24 16:37:54 -0500 (Mon, 24 May 2010) | 1 line
    
    #8616: update module name
  ........
    r81517 | brian.curtin | 2010-05-25 10:06:15 -0500 (Tue, 25 May 2010) | 5 lines
    
    Fix #2810 - handle the case where some registry calls return
    ERROR_MORE_DATA, requiring another call to get the remaining data.
    
    Patch by Daniel Stutzbach
  ........
    r81561 | alexander.belopolsky | 2010-05-26 15:57:04 -0500 (Wed, 26 May 2010) | 1 line
    
    Added Andrej Krpic. (Thanks for issue #7879 patch.)
  ........
    r81705 | r.david.murray | 2010-06-04 14:51:06 -0500 (Fri, 04 Jun 2010) | 9 lines
    
    #4487: have Charset check with codecs for possible aliases.
    
    Previously, unexpected results occurred when email was passed, for example,
    'utf8' as a charset name, since email would accept it but would *not* use
    the 'utf-8' codec for it, even though Python itself recognises that as
    an alias for utf-8.  Now Charset checks with codecs for aliases as well
    as its own internal table.  Issue 8898 has been opened to change this
    further in py3k so that all aliasing is routed through the codecs module.
  ........
    r81847 | brian.curtin | 2010-06-08 16:15:06 -0500 (Tue, 08 Jun 2010) | 3 lines
    
    Move a variable declration outside of a loop to match what was
    done in r81843 for py3k.
  ........
    r81919 | vinay.sajip | 2010-06-11 17:56:50 -0500 (Fri, 11 Jun 2010) | 1 line
    
    Issue #8924: logging: Improved error handling for Unicode in exception text.
  ........
    r81964 | nick.coghlan | 2010-06-13 01:50:39 -0500 (Sun, 13 Jun 2010) | 1 line
    
    Revert r80580 due to some unintended side effects. See issue #8202 for details.
  ........
    r82056 | florent.xicluna | 2010-06-17 15:30:56 -0500 (Thu, 17 Jun 2010) | 2 lines
    
    Add few words about test.test_genericpath.CommonTest
  ........
    r82136 | benjamin.peterson | 2010-06-21 10:56:45 -0500 (Mon, 21 Jun 2010) | 1 line
    
    update pydoc-topics
  ........
    r82137 | benjamin.peterson | 2010-06-21 10:57:57 -0500 (Mon, 21 Jun 2010) | 1 line
    
    bump verson to 2.7rc2
  ........
    r82142 | benjamin.peterson | 2010-06-21 12:28:30 -0500 (Mon, 21 Jun 2010) | 1 line
    
    towards final release
  ........
    r82149 | ronald.oussoren | 2010-06-22 04:18:28 -0500 (Tue, 22 Jun 2010) | 11 lines
    
    Fix for issue8446:
    
    * Don't import 'ic' in webbrowser, that module is no longer used
    * Remove 'MacOS' from the list of modules that should emit a Py3kWarning on import.
      This is needed because one of the earlier tests triggers and import of this
      extension, and that causes a failure in test_py3kwarn (running test_py3kwarn 
      separately worked fine)
    
    With these changes 'make tests' no longer says that test_py3kwarn fails.
  ........
................
  r82293 | benjamin.peterson | 2010-06-27 21:54:28 +0100 (Sun, 27 Jun 2010) | 13 lines
  
  Merged revisions 80712,81651 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r80712 | vinay.sajip | 2010-05-03 10:11:53 -0500 (Mon, 03 May 2010) | 1 line
    
    Issue #8576: logging updated to remove usage of find_unused_port().
  ........
    r81651 | vinay.sajip | 2010-06-02 05:05:31 -0500 (Wed, 02 Jun 2010) | 1 line
    
    Logging: improved error reporting for BaseConfigurator.resolve().
  ........
................
  r82294 | benjamin.peterson | 2010-06-27 22:05:30 +0100 (Sun, 27 Jun 2010) | 174 lines
  
  Blocked revisions 80731-80736,80739,80741-80744,80746-80747,80807-80812,80815-80822,80843,80874 via svnmerge
  
  ........
    r80731 | brett.cannon | 2010-05-03 18:41:23 -0500 (Mon, 03 May 2010) | 4 lines
    
    Remove unneeded variable initialization.
    
    Found using Clang's static analyzer.
  ........
    r80732 | brett.cannon | 2010-05-03 18:41:51 -0500 (Mon, 03 May 2010) | 4 lines
    
    Remove an unused variable.
    
    Found using Clang's static analyzer.
  ........
    r80733 | brett.cannon | 2010-05-03 18:42:40 -0500 (Mon, 03 May 2010) | 3 lines
    
    Clean up whitespace and remove unneeded variable initialization as found by
    Clang.
  ........
    r80734 | brett.cannon | 2010-05-03 18:43:49 -0500 (Mon, 03 May 2010) | 4 lines
    
    Remove unneeded variable mutation and initializations.
    
    Found using Clang's static analyzer.
  ........
    r80735 | brett.cannon | 2010-05-03 18:44:54 -0500 (Mon, 03 May 2010) | 4 lines
    
    Remove unneeded variable assignment.
    
    Found using Clang's static analyzer.
  ........
    r80736 | brett.cannon | 2010-05-03 18:51:28 -0500 (Mon, 03 May 2010) | 4 lines
    
    Remove unused variables and a variable initialization.
    
    Found using Clang's static analyzer.
  ........
    r80739 | brett.cannon | 2010-05-03 19:36:00 -0500 (Mon, 03 May 2010) | 4 lines
    
    Prevent a possible NULL de-reference and an unneeded variable assignment.
    
    Found using Clang's static analyzer.
  ........
    r80741 | brett.cannon | 2010-05-03 19:48:11 -0500 (Mon, 03 May 2010) | 4 lines
    
    Fix a Py_DECREF to a Py_XDECREF.
    
    Found using Clang's static analyzer.
  ........
    r80742 | brett.cannon | 2010-05-03 19:52:41 -0500 (Mon, 03 May 2010) | 4 lines
    
    Strip out extraneous whitespace, cast a some `const char *` to `void *` when
    passed to free() and make a `char *` to a `const char *` as found by Clang's
    static analyzer.
  ........
    r80743 | brett.cannon | 2010-05-03 19:57:44 -0500 (Mon, 03 May 2010) | 4 lines
    
    Remove an unneeded variable assignment.
    
    Found using Clang's static analyzer.
  ........
    r80744 | brett.cannon | 2010-05-03 20:01:00 -0500 (Mon, 03 May 2010) | 2 lines
    
    Fix some whitespace.
  ........
    r80746 | brett.cannon | 2010-05-03 20:16:51 -0500 (Mon, 03 May 2010) | 4 lines
    
    Remove an unneeded variable and assignment.
    
    Found using Clang's static analyzer.
  ........
    r80747 | brett.cannon | 2010-05-03 20:23:36 -0500 (Mon, 03 May 2010) | 4 lines
    
    Pull a NULL pointer check up to cover more cases in the function.
    
    Found using Clang's static analyzer.
  ........
    r80807 | brett.cannon | 2010-05-05 15:14:27 -0500 (Wed, 05 May 2010) | 4 lines
    
    Remove an unnecessary variable.
    
    Found using Clang's static analyzer.
  ........
    r80808 | brett.cannon | 2010-05-05 15:15:14 -0500 (Wed, 05 May 2010) | 2 lines
    
    Remove extraneous whitespace.
  ........
    r80809 | brett.cannon | 2010-05-05 15:16:09 -0500 (Wed, 05 May 2010) | 4 lines
    
    Remove an unneeded variable increment.
    
    Found using Clang's static analyzer.
  ........
    r80810 | brett.cannon | 2010-05-05 15:16:50 -0500 (Wed, 05 May 2010) | 4 lines
    
    Remove an unneeded variable.
    
    Found using Clang's static analyzer.
  ........
    r80811 | brett.cannon | 2010-05-05 15:18:23 -0500 (Wed, 05 May 2010) | 4 lines
    
    Remove an unneeded variable assignment.
    
    Found using Clang's static analyzer.
  ........
    r80812 | brett.cannon | 2010-05-05 15:19:26 -0500 (Wed, 05 May 2010) | 4 lines
    
    Remove an unneeded assignment.
    
    Found using Clang's static analyzer.
  ........
    r80815 | brett.cannon | 2010-05-05 15:25:47 -0500 (Wed, 05 May 2010) | 1 line
    
    Fix whitespace.
  ........
    r80816 | brett.cannon | 2010-05-05 15:26:58 -0500 (Wed, 05 May 2010) | 4 lines
    
    Remove an unneeded variable and fix a little whitespace.
    
    Found using Clang's static analyzer.
  ........
    r80817 | brett.cannon | 2010-05-05 15:27:49 -0500 (Wed, 05 May 2010) | 4 lines
    
    Remove an unneeded variable increment.
    
    Found using Clang's static analyzer.
  ........
    r80818 | brett.cannon | 2010-05-05 15:30:30 -0500 (Wed, 05 May 2010) | 8 lines
    
    Remove two unneeded branches to an 'if' statement by applying De Morgan's Law
    and creating a single 'if' statement along with a NULL default value for a
    variable.
    
    Also clean up a bunch of whitespace.
    
    Found using Clang's static analyzer.
  ........
    r80819 | brett.cannon | 2010-05-05 15:35:24 -0500 (Wed, 05 May 2010) | 4 lines
    
    Remove an unnecessary variable assignment.
    
    Found using Clang's static analyzer.
  ........
    r80820 | brett.cannon | 2010-05-05 15:38:52 -0500 (Wed, 05 May 2010) | 4 lines
    
    Change to a Py_XDECREF and fix some whitespace.
    
    Found using Clang's static analyzer.
  ........
    r80821 | brett.cannon | 2010-05-05 15:50:03 -0500 (Wed, 05 May 2010) | 4 lines
    
    Explicitly add stdio.h and string.h to make strtod.c work standalone.
    
    Found using Clang's static analyzer.
  ........
    r80822 | brett.cannon | 2010-05-05 15:53:20 -0500 (Wed, 05 May 2010) | 4 lines
    
    Remove an unneeded variable assignment.
    
    Found using Clang's static analyzer.
  ........
    r80843 | brett.cannon | 2010-05-05 18:11:08 -0500 (Wed, 05 May 2010) | 1 line
    
    Revert a change where an expression is not needed now, but could be in the future.
  ........
    r80874 | brett.cannon | 2010-05-06 12:56:36 -0500 (Thu, 06 May 2010) | 5 lines
    
    Add back in a line that was unneeded which advanced a pointer, but commented
    out as it is currently unneeded.
    
    This effectively adds back in the line removed in r80809 as a comment.
  ........
................
  r82295 | benjamin.peterson | 2010-06-27 22:15:09 +0100 (Sun, 27 Jun 2010) | 16 lines
  
  Blocked revisions 79925-79926,80397 via svnmerge
  
  ........
    r79925 | nick.coghlan | 2010-04-10 09:24:36 -0500 (Sat, 10 Apr 2010) | 1 line
    
    Try to turn some buildbots green by allowing test_multiprocessing to pass even if it hits the sys.exc_clear code in the threading module, and improve the test coverage by making the ctypes dependencies a bit more granular (two of the cited ctypes objects don't exist on my system)
  ........
    r79926 | nick.coghlan | 2010-04-10 10:01:54 -0500 (Sat, 10 Apr 2010) | 1 line
    
    Fix typo in comment
  ........
    r80397 | sean.reifschneider | 2010-04-23 03:38:24 -0500 (Fri, 23 Apr 2010) | 2 lines
    
    Small comment documentation change to clarify "ident" selection.
  ........
................
  r82296 | benjamin.peterson | 2010-06-27 22:22:34 +0100 (Sun, 27 Jun 2010) | 12 lines
  
  Blocked revisions 80649,80710 via svnmerge
  
  ........
    r80649 | tarek.ziade | 2010-04-30 07:15:12 -0500 (Fri, 30 Apr 2010) | 1 line
    
    Fixed #8577. distutils.sysconfig.get_python_inc() now differenciates buildir and srcdir
  ........
    r80710 | brett.cannon | 2010-05-02 16:51:23 -0500 (Sun, 02 May 2010) | 1 line
    
    Update the Vim syntax highlight file.
  ........
................
  r82297 | benjamin.peterson | 2010-06-27 22:27:49 +0100 (Sun, 27 Jun 2010) | 16 lines
  
  Blocked revisions 80787,80793 via svnmerge
  
  ........
    r80787 | ronald.oussoren | 2010-05-05 10:32:39 -0500 (Wed, 05 May 2010) | 5 lines
    
    Force exit using os._exit instead of sys.exit,
    this makes sure that the child does not continue
    testing.
  ........
    r80793 | barry.warsaw | 2010-05-05 11:15:09 -0500 (Wed, 05 May 2010) | 5 lines
    
    Bug 7755: audiotest.au is arguably copyrighted material, but definitely makes
    Debian unhappy.  The actual contents of the audio clip are unimportant, so
    replace it with something that we know is okay.  Guido likes woodpeckers.
  ........
................
  r82298 | benjamin.peterson | 2010-06-27 22:45:24 +0100 (Sun, 27 Jun 2010) | 37 lines
  
  Merged revisions 81465-81466,81468,81679,81735,81760,81868,82183 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r81465 | georg.brandl | 2010-05-22 06:29:19 -0500 (Sat, 22 May 2010) | 2 lines
    
    Issue #3924: Ignore cookies with invalid "version" field in cookielib.
  ........
    r81466 | georg.brandl | 2010-05-22 06:31:16 -0500 (Sat, 22 May 2010) | 1 line
    
    Underscore the name of an internal utility function.
  ........
    r81468 | georg.brandl | 2010-05-22 06:43:25 -0500 (Sat, 22 May 2010) | 1 line
    
    #8635: document enumerate() start parameter in docstring.
  ........
    r81679 | benjamin.peterson | 2010-06-03 16:21:03 -0500 (Thu, 03 Jun 2010) | 1 line
    
    use a set for membership testing
  ........
    r81735 | michael.foord | 2010-06-05 06:46:59 -0500 (Sat, 05 Jun 2010) | 1 line
    
    Extract error message truncating into a method (unittest.TestCase._truncateMessage).
  ........
    r81760 | michael.foord | 2010-06-05 14:38:42 -0500 (Sat, 05 Jun 2010) | 1 line
    
    Issue 8302. SkipTest exception is setUpClass or setUpModule is now reported as a skip rather than an error.
  ........
    r81868 | benjamin.peterson | 2010-06-09 14:45:04 -0500 (Wed, 09 Jun 2010) | 1 line
    
    fix code formatting
  ........
    r82183 | benjamin.peterson | 2010-06-23 15:29:26 -0500 (Wed, 23 Jun 2010) | 1 line
    
    cpython only gc tests
  ........
................
  r82299 | benjamin.peterson | 2010-06-27 22:48:35 +0100 (Sun, 27 Jun 2010) | 13 lines
  
  Merged revisions 81881-81882 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r81881 | andrew.kuchling | 2010-06-10 19:16:08 -0500 (Thu, 10 Jun 2010) | 1 line
    
    #5753: update demo.c to use PySys_SetArgvEx(), and add a comment
  ........
    r81882 | andrew.kuchling | 2010-06-10 19:23:01 -0500 (Thu, 10 Jun 2010) | 1 line
    
    #5753: Suggest PySys_SetArgvEx() instead of PySys_SetArgv()
  ........
................
  r82300 | brett.cannon | 2010-06-27 22:49:22 +0100 (Sun, 27 Jun 2010) | 4 lines
  
  Add a link to PEP 3147 from the importlib docs.
  
  Closes issue 8667. Thanks Ashley Sands for the patch.
................
  r82301 | benjamin.peterson | 2010-06-27 23:32:30 +0100 (Sun, 27 Jun 2010) | 303 lines
  
  Merged revisions 80605-80609,80642-80646,80651-80652,80674,80684-80686,80748,80852,80854,80870,80872-80873,80907,80915-80916,80951-80952,80976-80977,80985,81038-81040,81042,81053,81070,81104-81105,81114,81125,81245,81285,81402,81463,81516,81562-81563,81567,81593,81635,81680-81681,81684,81801,81888,81931-81933,81939-81942,81963,81984,81991,82120,82188,82264-82267 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r80605 | andrew.kuchling | 2010-04-28 19:22:16 -0500 (Wed, 28 Apr 2010) | 1 line
    
    Add various items
  ........
    r80606 | andrew.kuchling | 2010-04-28 20:44:30 -0500 (Wed, 28 Apr 2010) | 6 lines
    
    Fix doubled 'the'.
    Markup fixes to use :exc:, :option: in a few places.
      (Glitch: unittest.main's -c ends up a link to the Python
      interpreter's -c option.  Should we skip using :option: for that
      switch, or disable the auto-linking somehow?)
  ........
    r80607 | andrew.kuchling | 2010-04-28 20:45:41 -0500 (Wed, 28 Apr 2010) | 1 line
    
    Add various unittest items
  ........
    r80608 | benjamin.peterson | 2010-04-28 22:18:05 -0500 (Wed, 28 Apr 2010) | 1 line
    
    update pypy description
  ........
    r80609 | benjamin.peterson | 2010-04-28 22:30:59 -0500 (Wed, 28 Apr 2010) | 1 line
    
    update pypy url
  ........
    r80642 | andrew.kuchling | 2010-04-29 19:49:09 -0500 (Thu, 29 Apr 2010) | 1 line
    
    Always add space after RFC; reword paragraph
  ........
    r80643 | andrew.kuchling | 2010-04-29 19:52:31 -0500 (Thu, 29 Apr 2010) | 6 lines
    
    Reword paragraph to make its meaning clearer.
    
    Antoine Pitrou: is my version of the paragraph still correct?
    
    R. David Murray: is this more understandable than the previous version?
  ........
    r80644 | andrew.kuchling | 2010-04-29 20:02:15 -0500 (Thu, 29 Apr 2010) | 1 line
    
    Fix typos
  ........
    r80645 | andrew.kuchling | 2010-04-29 20:32:47 -0500 (Thu, 29 Apr 2010) | 1 line
    
    Markup fix; clarify by adding 'in that order'
  ........
    r80646 | andrew.kuchling | 2010-04-29 20:33:40 -0500 (Thu, 29 Apr 2010) | 1 line
    
    Add various items; rearrange unittest section a bit
  ........
    r80651 | andrew.kuchling | 2010-04-30 08:46:55 -0500 (Fri, 30 Apr 2010) | 1 line
    
    Minor grammar re-wording
  ........
    r80652 | andrew.kuchling | 2010-04-30 08:47:34 -0500 (Fri, 30 Apr 2010) | 1 line
    
    Add item
  ........
    r80674 | andrew.kuchling | 2010-04-30 20:19:16 -0500 (Fri, 30 Apr 2010) | 1 line
    
    Add various items
  ........
    r80684 | andrew.kuchling | 2010-05-01 07:05:52 -0500 (Sat, 01 May 2010) | 1 line
    
    Minor grammar fix
  ........
    r80685 | andrew.kuchling | 2010-05-01 07:06:51 -0500 (Sat, 01 May 2010) | 1 line
    
    Describe memoryview
  ........
    r80686 | antoine.pitrou | 2010-05-01 07:16:39 -0500 (Sat, 01 May 2010) | 4 lines
    
    Fix attribution. Travis didn't do much and he did a bad work.
    (yes, this is a sensitive subject, sorry)
  ........
    r80748 | andrew.kuchling | 2010-05-03 20:24:22 -0500 (Mon, 03 May 2010) | 1 line
    
    Add some more items; the urlparse change is added twice
  ........
    r80852 | andrew.kuchling | 2010-05-05 20:09:47 -0500 (Wed, 05 May 2010) | 1 line
    
    Reword paragraph; fix filename, which should be pyconfig.h
  ........
    r80854 | andrew.kuchling | 2010-05-05 20:10:56 -0500 (Wed, 05 May 2010) | 1 line
    
    Add various items
  ........
    r80870 | andrew.kuchling | 2010-05-06 09:14:09 -0500 (Thu, 06 May 2010) | 1 line
    
    Describe ElementTree 1.3; rearrange new-module sections; describe dict views as sets; small edits and items
  ........
    r80872 | andrew.kuchling | 2010-05-06 12:21:59 -0500 (Thu, 06 May 2010) | 1 line
    
    Add 2 items; record ideas for two initial sections; clarify wording
  ........
    r80873 | andrew.kuchling | 2010-05-06 12:27:57 -0500 (Thu, 06 May 2010) | 1 line
    
    Change section title; point to unittest2
  ........
    r80907 | andrew.kuchling | 2010-05-06 20:45:14 -0500 (Thu, 06 May 2010) | 1 line
    
    Add a new section on the development plan; add an item
  ........
    r80915 | antoine.pitrou | 2010-05-07 05:15:51 -0500 (Fri, 07 May 2010) | 3 lines
    
    Fix some markup and a class name. Also, wrap a long line.
  ........
    r80916 | andrew.kuchling | 2010-05-07 06:30:47 -0500 (Fri, 07 May 2010) | 1 line
    
    Re-word text
  ........
    r80951 | andrew.kuchling | 2010-05-07 20:15:26 -0500 (Fri, 07 May 2010) | 1 line
    
    Add two items
  ........
    r80952 | andrew.kuchling | 2010-05-07 20:35:55 -0500 (Fri, 07 May 2010) | 1 line
    
    Get accents correct
  ........
    r80976 | andrew.kuchling | 2010-05-08 08:28:03 -0500 (Sat, 08 May 2010) | 1 line
    
    Add logging.dictConfig example; give up on writing a Ttk example
  ........
    r80977 | andrew.kuchling | 2010-05-08 08:29:46 -0500 (Sat, 08 May 2010) | 1 line
    
    Markup fixes
  ........
    r80985 | andrew.kuchling | 2010-05-08 10:39:46 -0500 (Sat, 08 May 2010) | 7 lines
    
    Write summary of the 2.7 release; rewrite the future section some more;
    mention PYTHONWARNINGS env. var; tweak some examples for readability.
    
    And with this commit, the "What's New" is done... except for a
    complete read-through to polish the text, and fixing any reported errors,
    but those tasks can easily wait until after beta2.
  ........
    r81038 | benjamin.peterson | 2010-05-09 16:09:40 -0500 (Sun, 09 May 2010) | 1 line
    
    finish clause
  ........
    r81039 | andrew.kuchling | 2010-05-10 09:18:27 -0500 (Mon, 10 May 2010) | 1 line
    
    Markup fix; re-word a sentence
  ........
    r81040 | andrew.kuchling | 2010-05-10 09:20:12 -0500 (Mon, 10 May 2010) | 1 line
    
    Use title case
  ........
    r81042 | andrew.kuchling | 2010-05-10 10:03:35 -0500 (Mon, 10 May 2010) | 1 line
    
    Link to unittest2 article
  ........
    r81053 | florent.xicluna | 2010-05-10 14:59:22 -0500 (Mon, 10 May 2010) | 2 lines
    
    Add a link on maketrans().
  ........
    r81070 | andrew.kuchling | 2010-05-10 18:13:41 -0500 (Mon, 10 May 2010) | 1 line
    
    Fix typo
  ........
    r81104 | andrew.kuchling | 2010-05-11 19:38:44 -0500 (Tue, 11 May 2010) | 1 line
    
    Revision pass: lots of edits, typo fixes, rearrangements
  ........
    r81105 | andrew.kuchling | 2010-05-11 19:40:47 -0500 (Tue, 11 May 2010) | 1 line
    
    Let's call this done
  ........
    r81114 | andrew.kuchling | 2010-05-12 08:56:07 -0500 (Wed, 12 May 2010) | 1 line
    
    Grammar fix
  ........
    r81125 | andrew.kuchling | 2010-05-12 13:56:48 -0500 (Wed, 12 May 2010) | 1 line
    
    #8696: add documentation for logging.config.dictConfig (PEP 391)
  ........
    r81245 | andrew.kuchling | 2010-05-16 18:31:16 -0500 (Sun, 16 May 2010) | 1 line
    
    Add cross-reference to later section
  ........
    r81285 | vinay.sajip | 2010-05-18 03:16:27 -0500 (Tue, 18 May 2010) | 1 line
    
    Fixed minor typo in ReST markup.
  ........
    r81402 | vinay.sajip | 2010-05-21 12:41:34 -0500 (Fri, 21 May 2010) | 1 line
    
    Updated logging documentation with more dictConfig information.
  ........
    r81463 | georg.brandl | 2010-05-22 03:17:23 -0500 (Sat, 22 May 2010) | 1 line
    
    #8785: less confusing description of regex.find*.
  ........
    r81516 | andrew.kuchling | 2010-05-25 08:34:08 -0500 (Tue, 25 May 2010) | 1 line
    
    Add three items
  ........
    r81562 | andrew.kuchling | 2010-05-27 08:22:53 -0500 (Thu, 27 May 2010) | 1 line
    
    Rewrite wxWidgets section
  ........
    r81563 | andrew.kuchling | 2010-05-27 08:30:09 -0500 (Thu, 27 May 2010) | 1 line
    
    Remove top-level 'General Questions' section, pushing up the questions it contains
  ........
    r81567 | andrew.kuchling | 2010-05-27 16:29:59 -0500 (Thu, 27 May 2010) | 1 line
    
    Add item
  ........
    r81593 | georg.brandl | 2010-05-29 03:46:18 -0500 (Sat, 29 May 2010) | 1 line
    
    #8616: add new turtle demo "nim".
  ........
    r81635 | georg.brandl | 2010-06-01 02:25:23 -0500 (Tue, 01 Jun 2010) | 1 line
    
    Put docs for RegexObject.search() before RegexObject.match() to mirror re.search() and re.match() order.
  ........
    r81680 | vinay.sajip | 2010-06-03 17:34:42 -0500 (Thu, 03 Jun 2010) | 1 line
    
    Issue #8890: Documentation changed to avoid reference to temporary files.
  ........
    r81681 | sean.reifschneider | 2010-06-03 20:51:26 -0500 (Thu, 03 Jun 2010) | 2 lines
    
    Issue8810: Clearing up docstring for tzinfo.utcoffset.
  ........
    r81684 | vinay.sajip | 2010-06-04 08:41:02 -0500 (Fri, 04 Jun 2010) | 1 line
    
    Issue #8890: Documentation changed to avoid reference to temporary files - other cases covered.
  ........
    r81801 | andrew.kuchling | 2010-06-07 08:38:40 -0500 (Mon, 07 Jun 2010) | 1 line
    
    #8875: Remove duplicated paragraph
  ........
    r81888 | andrew.kuchling | 2010-06-10 20:54:58 -0500 (Thu, 10 Jun 2010) | 1 line
    
    Add a few more items
  ........
    r81931 | georg.brandl | 2010-06-12 01:26:54 -0500 (Sat, 12 Jun 2010) | 1 line
    
    Fix punctuation.
  ........
    r81932 | georg.brandl | 2010-06-12 01:28:58 -0500 (Sat, 12 Jun 2010) | 1 line
    
    Document that an existing directory raises in mkdir().
  ........
    r81933 | georg.brandl | 2010-06-12 01:45:33 -0500 (Sat, 12 Jun 2010) | 1 line
    
    Update version in README.
  ........
    r81939 | georg.brandl | 2010-06-12 04:45:01 -0500 (Sat, 12 Jun 2010) | 1 line
    
    Use newer toctree syntax.
  ........
    r81940 | georg.brandl | 2010-06-12 04:45:28 -0500 (Sat, 12 Jun 2010) | 1 line
    
    Add document on how to build.
  ........
    r81941 | georg.brandl | 2010-06-12 04:45:58 -0500 (Sat, 12 Jun 2010) | 1 line
    
    Fix gratuitous indentation.
  ........
    r81942 | georg.brandl | 2010-06-12 04:46:03 -0500 (Sat, 12 Jun 2010) | 1 line
    
    Update README.
  ........
    r81963 | andrew.kuchling | 2010-06-12 15:00:55 -0500 (Sat, 12 Jun 2010) | 1 line
    
    Grammar fix
  ........
    r81984 | georg.brandl | 2010-06-14 10:58:39 -0500 (Mon, 14 Jun 2010) | 1 line
    
    #8993: fix reference.
  ........
    r81991 | andrew.kuchling | 2010-06-14 19:38:58 -0500 (Mon, 14 Jun 2010) | 1 line
    
    Add another bunch of items
  ........
    r82120 | andrew.kuchling | 2010-06-20 16:45:45 -0500 (Sun, 20 Jun 2010) | 1 line
    
    Note that Python 3.x isn't covered; add forward ref. for UTF-8; note error in 2.5 and up
  ........
    r82188 | benjamin.peterson | 2010-06-23 19:02:46 -0500 (Wed, 23 Jun 2010) | 1 line
    
    remove reverted changed
  ........
    r82264 | georg.brandl | 2010-06-27 05:47:47 -0500 (Sun, 27 Jun 2010) | 1 line
    
    Confusing punctuation.
  ........
    r82265 | georg.brandl | 2010-06-27 05:49:23 -0500 (Sun, 27 Jun 2010) | 1 line
    
    Use designated syntax for optional grammar element.
  ........
    r82266 | georg.brandl | 2010-06-27 05:51:44 -0500 (Sun, 27 Jun 2010) | 1 line
    
    Fix URL.
  ........
    r82267 | georg.brandl | 2010-06-27 05:55:38 -0500 (Sun, 27 Jun 2010) | 1 line
    
    Two typos.
  ........
................
  r82302 | benjamin.peterson | 2010-06-27 23:37:28 +0100 (Sun, 27 Jun 2010) | 15 lines
  
  Merged revisions 81380 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r81380 | brett.cannon | 2010-05-20 13:37:55 -0500 (Thu, 20 May 2010) | 8 lines
    
    Turned out that if you used explicit relative import syntax
    (e.g. from .os import sep) and it failed, import would still try the implicit
    relative import semantics of an absolute import (from os import sep). That's
    not right, so when level is negative, only do explicit relative import
    semantics.
    
    Fixes issue #7902. Thanks to Meador Inge for the patch.
  ........
................
  r82304 | benjamin.peterson | 2010-06-27 23:41:29 +0100 (Sun, 27 Jun 2010) | 13 lines
  
  Merged revisions 81499,81506 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r81499 | georg.brandl | 2010-05-24 16:29:07 -0500 (Mon, 24 May 2010) | 1 line
    
    #8016: add the CP858 codec (approved by Benjamin).  (Also add CP720 to the tests, it was missing there.)
  ........
    r81506 | benjamin.peterson | 2010-05-24 17:04:53 -0500 (Mon, 24 May 2010) | 1 line
    
    set svn:eol-style
  ........
................
  r82305 | benjamin.peterson | 2010-06-27 23:44:51 +0100 (Sun, 27 Jun 2010) | 9 lines
  
  Merged revisions 82303 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82303 | benjamin.peterson | 2010-06-27 17:40:26 -0500 (Sun, 27 Jun 2010) | 1 line
    
    mark tracking tests as implementation details
  ........
................
  r82306 | benjamin.peterson | 2010-06-28 00:16:51 +0100 (Mon, 28 Jun 2010) | 9 lines
  
  Blocked revisions 81700 via svnmerge
  
  ........
    r81700 | martin.v.loewis | 2010-06-04 14:15:32 -0500 (Fri, 04 Jun 2010) | 2 lines
    
    Display installer warning that Windows 2000 won't 
    be supported in future releases.
  ........
................
  r82307 | benjamin.peterson | 2010-06-28 00:21:13 +0100 (Mon, 28 Jun 2010) | 9 lines
  
  Merged revisions 81880 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r81880 | andrew.kuchling | 2010-06-10 19:14:34 -0500 (Thu, 10 Jun 2010) | 1 line
    
    Edit comments for current Python; bump version number of Python
  ........
................
  r82308 | benjamin.peterson | 2010-06-28 00:26:04 +0100 (Mon, 28 Jun 2010) | 9 lines
  
  Recorded merge of revisions 80866 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r80866 | andrew.kuchling | 2010-05-06 08:03:39 -0500 (Thu, 06 May 2010) | 1 line
    
    Use anonymous hyperlinks
  ........
................
  r82309 | benjamin.peterson | 2010-06-28 00:31:03 +0100 (Mon, 28 Jun 2010) | 8 lines
  
  Blocked revisions 80396 via svnmerge
  
  ........
    r80396 | sean.reifschneider | 2010-04-23 03:31:55 -0500 (Fri, 23 Apr 2010) | 2 lines
    
    issue8451: Making syslog module use sys.argv[0] for "ident".
  ........
................
  r82310 | benjamin.peterson | 2010-06-28 00:34:03 +0100 (Mon, 28 Jun 2010) | 9 lines
  
  Merged revisions 81004 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r81004 | mark.dickinson | 2010-05-08 14:13:21 -0500 (Sat, 08 May 2010) | 1 line
    
    Fix configure bug that was misreporting availability of MacOS X 10.5 SDK as available on Linux but not on OS X.
  ........
................
  r82311 | benjamin.peterson | 2010-06-28 00:49:45 +0100 (Mon, 28 Jun 2010) | 24 lines
  
  Merged revisions 79894-79895,80098,80120 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r79894 | jeroen.ruigrok | 2010-04-07 09:33:37 -0500 (Wed, 07 Apr 2010) | 5 lines
    
    FreeBSD is not a sysv platform, so use freebsd.S instead of sysv.S (as is
    also the case in FreeBSD's port of libffi).
    
    Reviewed by: dickinsm
  ........
    r79895 | jeroen.ruigrok | 2010-04-07 11:34:08 -0500 (Wed, 07 Apr 2010) | 2 lines
    
    Document the libffi FreeBSD fix.
  ........
    r80098 | benjamin.peterson | 2010-04-15 16:42:16 -0500 (Thu, 15 Apr 2010) | 1 line
    
    add space
  ........
    r80120 | antoine.pitrou | 2010-04-16 16:34:02 -0500 (Fri, 16 Apr 2010) | 3 lines
    
    Ignore jinja2
  ........
................
  r82312 | benjamin.peterson | 2010-06-28 00:53:17 +0100 (Mon, 28 Jun 2010) | 8 lines
  
  Blocked revisions 79596 via svnmerge
  
  ........
    r79596 | eric.smith | 2010-04-02 07:30:56 -0500 (Fri, 02 Apr 2010) | 1 line
    
    Issue 7994: Make object.__format__ with a non-empty format string a PendingDecprecationWarning. Still need to remove uses of this from various tests.
  ........
................
  r82313 | brett.cannon | 2010-06-28 00:57:46 +0100 (Mon, 28 Jun 2010) | 29 lines
  
  Implement importlib.abc.SourceLoader and deprecate PyLoader and PyPycLoader.
  
  SourceLoader is a simplification of both PyLoader and PyPycLoader. If one only
  wants to use source, then they need to only implement get_data and
  get_filename. To also use bytecode -- sourceless loading is not supported --
  then two abstract methods -- path_mtime and set_data -- need to be implemented.
  Compared to PyLoader and PyPycLoader, there are less abstract methods
  introduced and bytecode files become an optimization controlled by the ABC and
  hidden from the user (this need came about as PEP 3147 showed that not treating
  bytecode as an optimization can cause problems for compatibility).
  
  PyLoader is deprecated in favor of SourceLoader. To be compatible from Python
  3.1 onwards, a subclass need only use simple methods for source_path and
  is_package. Otherwise conditional subclassing based on whether Python 3.1 or
  Python 3.2 is being is the only change. The documentation and docstring for
  PyLoader explain what is exactly needed.
  
  PyPycLoader is deprecated also in favor of SourceLoader. Because PEP 3147
  shifted bytecode path details so much, there is no foolproof way to provide
  backwards-compatibility with SourceLoader. Because of this the class is simply
  deprecated and users should move to SourceLoader (and optionally PyLoader for
  Python 3.1). This does lead to a loss of support for sourceless loading
  unfortunately.
  
  At some point before Python 3.2 is released, SourceLoader will be moved over to
  importlib._bootstrap so that the core code of importlib relies on the new code
  instead of the old PyPycLoader code. This commit is being done now so that
  there is no issue in having the API in Python 3.1a1.
................
  r82314 | benjamin.peterson | 2010-06-28 01:01:59 +0100 (Mon, 28 Jun 2010) | 32 lines
  
  Merged revisions 77402,77505,77510 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r77402 | brett.cannon | 2010-01-09 20:56:19 -0600 (Sat, 09 Jan 2010) | 12 lines
    
    DeprecationWarning is now silent by default.
    
    This was originally suggested by Guido, discussed on the stdlib-sig mailing
    list, and given the OK by Guido directly to me. What this change essentially
    means is that Python has taken a policy of silencing warnings that are only
    of interest to developers by default. This should prevent users from seeing
    warnings which are triggered by an application being run against a new
    interpreter before the app developer has a chance to update their code.
    
    Closes issue #7319. Thanks to Antoine Pitrou, Ezio Melotti, and Brian Curtin
    for helping with the issue.
  ........
    r77505 | brett.cannon | 2010-01-14 14:00:28 -0600 (Thu, 14 Jan 2010) | 7 lines
    
    The silencing of DeprecationWarning was not taking -3 into consideration. Since
    Py3K warnings are DeprecationWarning by default this was causing -3 to
    essentially be a no-op. Now DeprecationWarning is only silenced if -3 is not
    used.
    
    Closes issue #7700. Thanks Ezio Melotti and Florent Xicluna for patch help.
  ........
    r77510 | brett.cannon | 2010-01-14 19:31:45 -0600 (Thu, 14 Jan 2010) | 1 line
    
    Remove C++/C99-style comments.
  ........
................
  r82315 | benjamin.peterson | 2010-06-28 01:16:12 +0100 (Mon, 28 Jun 2010) | 1 line
  
  fix versionchanged
................
  r82316 | benjamin.peterson | 2010-06-28 01:24:13 +0100 (Mon, 28 Jun 2010) | 1 line
  
  fix test with more obviously incorrect bytecode
................
  r82318 | benjamin.peterson | 2010-06-28 04:07:10 +0100 (Mon, 28 Jun 2010) | 9 lines
  
  Merged revisions 82317 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82317 | benjamin.peterson | 2010-06-27 21:58:25 -0500 (Sun, 27 Jun 2010) | 1 line
    
    remove unused last argument to property_copy
  ........
................
  r82320 | brett.cannon | 2010-06-28 05:57:24 +0100 (Mon, 28 Jun 2010) | 11 lines
  
  Move importlib.abc.SourceLoader to _bootstrap.
  
  Required updating code relying on other modules to switch to _bootstrap's
  unique module requirements. This led to the realization that
  get_code was being too liberal in its exception catching when calling set_data
  by blindly grabbing IOError. Shifted the responsibility of safely ignoring
  writes to a read-only path to set_data.
  
  Importlib is still not relying on SourceLoader yet; requires creating a
  SourcelessLoader and updating the source finder.
................
  r82321 | brett.cannon | 2010-06-28 06:46:25 +0100 (Mon, 28 Jun 2010) | 2 lines
  
  Add an inheritance test for importlib.abc.SourceLoader.
................
  r82326 | senthil.kumaran | 2010-06-28 15:08:00 +0100 (Mon, 28 Jun 2010) | 9 lines
  
  Merged revisions 82324 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82324 | senthil.kumaran | 2010-06-28 19:26:46 +0530 (Mon, 28 Jun 2010) | 3 lines
    
    Fix Issue8653 - Docstring for urlunsplit function.
  ........
................
  r82328 | alexander.belopolsky | 2010-06-28 15:36:55 +0100 (Mon, 28 Jun 2010) | 1 line
  
  Added missing white space
................
  r82331 | benjamin.peterson | 2010-06-28 16:39:55 +0100 (Mon, 28 Jun 2010) | 9 lines
  
  Merged revisions 82330 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82330 | benjamin.peterson | 2010-06-28 10:36:40 -0500 (Mon, 28 Jun 2010) | 1 line
    
    testcapi tests are definitely cpython only
  ........
................
  r82333 | benjamin.peterson | 2010-06-28 16:43:25 +0100 (Mon, 28 Jun 2010) | 9 lines
  
  Merged revisions 82332 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82332 | benjamin.peterson | 2010-06-28 10:41:06 -0500 (Mon, 28 Jun 2010) | 1 line
    
    fix skipping condition
  ........
................
  r82336 | senthil.kumaran | 2010-06-28 18:31:40 +0100 (Mon, 28 Jun 2010) | 9 lines
  
  Merged revisions 82334 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82334 | senthil.kumaran | 2010-06-28 22:37:40 +0530 (Mon, 28 Jun 2010) | 3 lines
    
    Documentation correction. urllib2.urlopen.info returns a mimetools.Message instance not HTTPMessage object.
  ........
................
  r82342 | benjamin.peterson | 2010-06-28 20:43:42 +0100 (Mon, 28 Jun 2010) | 1 line
  
  update error message
................
  r82343 | mark.dickinson | 2010-06-28 20:44:20 +0100 (Mon, 28 Jun 2010) | 23 lines
  
  Merged revisions 82338,82340-82341 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82338 | mark.dickinson | 2010-06-28 20:31:41 +0100 (Mon, 28 Jun 2010) | 9 lines
    
    Fix some shallow bugs in Demo/parser/unparse.py, and add tests:
    
      - insert commas between entries in del statement
      - left and right shifts were represented as >> and << (respectively); reverse
      - unindent properly after for: else:  or while: else:
      - add parens around the result of an unary operation
      - add parens around negative numbers, to avoid turning (-1)**2 into -1**2.
  ........
    r82340 | mark.dickinson | 2010-06-28 20:34:15 +0100 (Mon, 28 Jun 2010) | 1 line
    
    Fix typo in test_unparse.py.
  ........
    r82341 | mark.dickinson | 2010-06-28 20:38:19 +0100 (Mon, 28 Jun 2010) | 1 line
    
    Set svn:eol-style on test_unparse.py.
  ........
................
  r82346 | mark.dickinson | 2010-06-28 21:09:18 +0100 (Mon, 28 Jun 2010) | 9 lines
  
  Merged revisions 82345 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82345 | mark.dickinson | 2010-06-28 20:54:19 +0100 (Mon, 28 Jun 2010) | 1 line
    
    unparse.py:  fix mispaced parentheses in chained comparisons
  ........
................
  r82347 | mark.dickinson | 2010-06-28 22:14:17 +0100 (Mon, 28 Jun 2010) | 18 lines
  
  Update Demo/parser/unparse.py to current Python 3.x syntax.  Additions:
   - relative imports
   - keyword-only arguments
   - function annotations
   - class decorators
   - raise ... from ...
   - except ... as ...
   - nonlocal
   - bytes literals
   - set literals
   - set comprehensions
   - dict comprehensions
  Removals:
   - print statement.
  
  Some of this should be backported to 2.x.
................
  r82349 | mark.dickinson | 2010-06-28 22:39:51 +0100 (Mon, 28 Jun 2010) | 1 line
  
  unparse.py: Typo fix.
................
  r82351 | brian.curtin | 2010-06-29 01:14:28 +0100 (Tue, 29 Jun 2010) | 3 lines
  
  Update md5driver.py for 3.x.
  Changed an import, replaced md5.new() with md5(), and added an encode where needed.
................
  r82354 | benjamin.peterson | 2010-06-29 02:27:35 +0100 (Tue, 29 Jun 2010) | 1 line
  
  rephrase and remove uneeded statement
................
  r82357 | mark.dickinson | 2010-06-29 08:38:37 +0100 (Tue, 29 Jun 2010) | 9 lines
  
  Merged revisions 82356 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82356 | mark.dickinson | 2010-06-29 08:37:25 +0100 (Tue, 29 Jun 2010) | 1 line
    
    Spelling.
  ........
................
  r82358 | mark.dickinson | 2010-06-29 08:48:23 +0100 (Tue, 29 Jun 2010) | 1 line
  
  unparse.py: respect coding cookie in input files
................
  r82359 | mark.dickinson | 2010-06-29 09:52:36 +0100 (Tue, 29 Jun 2010) | 1 line
  
  Add parentheses around numeric literals, to avoid turning 3 .bit_length() into 3.bit_length().
................
  r82360 | mark.dickinson | 2010-06-29 11:01:48 +0100 (Tue, 29 Jun 2010) | 8 lines
  
  More unparse.py fixes:
  
   - parenthesize lambdas, to avoid turning (lambda : int)() into lambda: int()
   - unparse an infinite float literals in the AST as an overflowing finite value
  
  unparse.py now successfully round-trips on all valid Lib/*.py and Lib/test/*.py files.
................
  r82365 | mark.dickinson | 2010-06-29 15:08:23 +0100 (Tue, 29 Jun 2010) | 1 line
  
  test_unparse.py: Do roundtrip testing for all Python files in Lib and Lib/test.
................
  r82370 | vinay.sajip | 2010-06-29 16:31:15 +0100 (Tue, 29 Jun 2010) | 1 line
  
  Added information about pickle security and SocketHandler, and some cross-reference targets.
................
  r82371 | benjamin.peterson | 2010-06-29 16:32:34 +0100 (Tue, 29 Jun 2010) | 8 lines
  
  Blocked revisions 82368 via svnmerge
  
  ........
    r82368 | benjamin.peterson | 2010-06-29 10:18:02 -0500 (Tue, 29 Jun 2010) | 1 line
    
    group cStringIO docs under its factory function
  ........
................
  r82373 | benjamin.peterson | 2010-06-29 16:57:57 +0100 (Tue, 29 Jun 2010) | 1 line
  
  note that byte strings can be raw, too #9114
................
  r82375 | brett.cannon | 2010-06-29 19:26:11 +0100 (Tue, 29 Jun 2010) | 1 line
  
  Make a sentence a little less awkward.
................
  r82376 | benjamin.peterson | 2010-06-29 19:36:39 +0100 (Tue, 29 Jun 2010) | 1 line
  
  update for nonlocal keyword
................
  r82377 | mark.dickinson | 2010-06-29 19:38:59 +0100 (Tue, 29 Jun 2010) | 1 line
  
  Fix typo in unparsing of a class definition.
................
  r82381 | mark.dickinson | 2010-06-29 21:10:42 +0100 (Tue, 29 Jun 2010) | 9 lines
  
  Merged revisions 82379 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82379 | mark.dickinson | 2010-06-29 21:09:12 +0100 (Tue, 29 Jun 2010) | 1 line
    
    Issue #1789:  clarify that the 'size' column in struct docs refers to standard size.
  ........
................
  r82383 | mark.dickinson | 2010-06-30 09:32:11 +0100 (Wed, 30 Jun 2010) | 1 line
  
  Collapse else: if: ... into elif: 
................
  r82384 | mark.dickinson | 2010-06-30 09:46:53 +0100 (Wed, 30 Jun 2010) | 1 line
  
  Output try-except-finally statements where appropriate.
................
  r82386 | mark.dickinson | 2010-06-30 10:05:47 +0100 (Wed, 30 Jun 2010) | 1 line
  
  Better solution for attribute access on integer literals.
................
  r82387 | mark.dickinson | 2010-06-30 10:14:43 +0100 (Wed, 30 Jun 2010) | 1 line
  
  Remove backtick syntax for repr.
................
  r82391 | mark.dickinson | 2010-06-30 12:13:36 +0100 (Wed, 30 Jun 2010) | 1 line
  
  Issue #9011:  Tests for Python 3.2's treatment of negated imaginary literals.
................
  r82392 | mark.dickinson | 2010-06-30 12:14:30 +0100 (Wed, 30 Jun 2010) | 1 line
  
  Revert accidental extra changes included in r82391.
................
  r82393 | mark.dickinson | 2010-06-30 12:45:53 +0100 (Wed, 30 Jun 2010) | 1 line
  
  Unparse infinite imaginary literals correctly.  Add some more numeric tests.
................
  r82394 | michael.foord | 2010-06-30 13:17:50 +0100 (Wed, 30 Jun 2010) | 1 line
  
  Issue 9110. Adding ContextDecorator to contextlib. This enables the creation of APIs that act as decorators as well as context managers. contextlib.contextmanager changed to use ContextDecorator.
................
  r82402 | mark.dickinson | 2010-06-30 17:36:12 +0100 (Wed, 30 Jun 2010) | 12 lines
  
  Blocked revisions 82399-82400 via svnmerge
  
  ........
    r82399 | mark.dickinson | 2010-06-30 15:19:56 +0100 (Wed, 30 Jun 2010) | 1 line
    
    Update Demo/parser directory; backport unparse fixes from py3k.
  ........
    r82400 | mark.dickinson | 2010-06-30 17:27:57 +0100 (Wed, 30 Jun 2010) | 2 lines
    
    Issue #9125:  Update parser module for "except ... as ..." syntax.
  ........
................
  r82405 | benjamin.peterson | 2010-06-30 18:39:45 +0100 (Wed, 30 Jun 2010) | 9 lines
  
  Merged revisions 82403 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82403 | benjamin.peterson | 2010-06-30 12:11:08 -0500 (Wed, 30 Jun 2010) | 1 line
    
    mark test depending on ref counting
  ........
................
  r82407 | giampaolo.rodola | 2010-06-30 18:47:39 +0100 (Wed, 30 Jun 2010) | 9 lines
  
  Merged revisions 82404 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82404 | giampaolo.rodola | 2010-06-30 19:38:28 +0200 (mer, 30 giu 2010) | 1 line
    
    fix issue #6589: cleanup asyncore.socket_map if smtpd.SMTPServer constructor raises an exception
  ........
................
  r82410 | benjamin.peterson | 2010-06-30 19:44:04 +0100 (Wed, 30 Jun 2010) | 9 lines
  
  Merged revisions 82409 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82409 | benjamin.peterson | 2010-06-30 13:41:08 -0500 (Wed, 30 Jun 2010) | 1 line
    
    an AttributeError is perfectly acceptable here
  ........
................
  r82413 | ezio.melotti | 2010-07-01 08:32:02 +0100 (Thu, 01 Jul 2010) | 13 lines
  
  Update PyUnicode_DecodeUTF8 from RFC 2279 to RFC 3629.
  
  1) #8271: when a byte sequence is invalid, only the start byte and all the
     valid continuation bytes are now replaced by U+FFFD, instead of replacing
     the number of bytes specified by the start byte.
     See http://www.unicode.org/versions/Unicode5.2.0/ch03.pdf (pages 94-95);
  2) 5- and 6-bytes-long UTF-8 sequences are now considered invalid (no changes
     in behavior);
  3) Change the error messages "unexpected code byte" to "invalid start byte"
     and "invalid data" to "invalid continuation byte";
  4) Add an extensive set of tests in test_unicode;
  5) Fix test_codeccallbacks because it was failing after this change.
................
  r82416 | benjamin.peterson | 2010-07-01 16:09:14 +0100 (Thu, 01 Jul 2010) | 8 lines
  
  Blocked revisions 82415 via svnmerge
  
  ........
    r82415 | benjamin.peterson | 2010-07-01 10:07:15 -0500 (Thu, 01 Jul 2010) | 1 line
    
    remove docs about delegating to special methods; it does no such thing
  ........
................
  r82417 | benjamin.peterson | 2010-07-01 16:16:55 +0100 (Thu, 01 Jul 2010) | 1 line
  
  correctly lookup __trunc__ and __floor__
................
  r82422 | benjamin.peterson | 2010-07-01 18:49:01 +0100 (Thu, 01 Jul 2010) | 9 lines
  
  Merged revisions 82420 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82420 | benjamin.peterson | 2010-07-01 12:45:52 -0500 (Thu, 01 Jul 2010) | 1 line
    
    fix fixer name
  ........
................
  r82430 | benjamin.peterson | 2010-07-02 00:39:26 +0100 (Fri, 02 Jul 2010) | 8 lines
  
  Blocked revisions 82429 via svnmerge
  
  ........
    r82429 | benjamin.peterson | 2010-07-01 18:35:37 -0500 (Thu, 01 Jul 2010) | 1 line
    
    trunc does use the special method, though
  ........
................
  r82434 | georg.brandl | 2010-07-02 08:41:51 +0100 (Fri, 02 Jul 2010) | 9 lines
  
  Merged revisions 82433 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82433 | georg.brandl | 2010-07-02 09:33:50 +0200 (Fr, 02 Jul 2010) | 1 line
    
    Grammar and markup fixes.
  ........
................
  r82437 | benjamin.peterson | 2010-07-02 14:35:17 +0100 (Fri, 02 Jul 2010) | 1 line
  
  account for different ref counting semantics of _PyObject_LookupSpecial
................
  r82438 | benjamin.peterson | 2010-07-02 14:46:42 +0100 (Fri, 02 Jul 2010) | 1 line
  
  fix lookup of __ceil__
................
  r82440 | nick.coghlan | 2010-07-02 16:57:50 +0100 (Fri, 02 Jul 2010) | 9 lines
  
  Merged revisions 82439 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82439 | nick.coghlan | 2010-07-03 01:50:14 +1000 (Sat, 03 Jul 2010) | 1 line
    
    Fix a couple of minor nits in What's New before the release goes out
  ........
................
  r82442 | mark.dickinson | 2010-07-02 17:05:15 +0100 (Fri, 02 Jul 2010) | 1 line
  
  Style/consistency nit: make math_floor and math_ceil code look the same.
................
  r82443 | nick.coghlan | 2010-07-02 17:14:05 +0100 (Fri, 02 Jul 2010) | 9 lines
  
  Merged revisions 82441 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82441 | nick.coghlan | 2010-07-03 02:01:53 +1000 (Sat, 03 Jul 2010) | 1 line
    
    Issue 8202 actually caused an unintended behavioural change and was reverted in 2.7 - remove reference from What's New
  ........
................
  r82448 | benjamin.peterson | 2010-07-02 20:45:07 +0100 (Fri, 02 Jul 2010) | 9 lines
  
  Merged revisions 82447 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82447 | benjamin.peterson | 2010-07-02 14:41:39 -0500 (Fri, 02 Jul 2010) | 1 line
    
    add space
  ........
................
  r82452 | mark.dickinson | 2010-07-02 21:26:07 +0100 (Fri, 02 Jul 2010) | 9 lines
  
  Merged revisions 82446 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82446 | mark.dickinson | 2010-07-02 19:06:52 +0100 (Fri, 02 Jul 2010) | 1 line
    
    Clarify sys.float_info documentation.
  ........
................
  r82456 | brett.cannon | 2010-07-02 23:03:00 +0100 (Fri, 02 Jul 2010) | 7 lines
  
  Some people have mistaken Python bytecode as being stable and unchanging. In
  reality it's simply an implementation detail for CPython. This point is now
  clearly documented in both the docs for dis and the glossary.
  
  Closes issue #7829. Thanks to Terry Reedy for some initial suggestions on
  wording.
................
  r82462 | ezio.melotti | 2010-07-03 00:18:51 +0100 (Sat, 03 Jul 2010) | 17 lines
  
  Merged revisions 82455,82457,82459 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82455 | eric.smith | 2010-07-03 00:44:16 +0300 (Sat, 03 Jul 2010) | 1 line
    
    Moved period outside paren, where it belongs.
  ........
    r82457 | ezio.melotti | 2010-07-03 01:17:29 +0300 (Sat, 03 Jul 2010) | 1 line
    
    #9139: Add examples for str.format().
  ........
    r82459 | ezio.melotti | 2010-07-03 01:50:39 +0300 (Sat, 03 Jul 2010) | 1 line
    
    #9139: the thousands separator is new in 2.7.  Also add a missing variable in the example.
  ........
................
  r82464 | benjamin.peterson | 2010-07-03 00:25:44 +0100 (Sat, 03 Jul 2010) | 9 lines
  
  Merged revisions 82461 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82461 | benjamin.peterson | 2010-07-02 18:05:27 -0500 (Fri, 02 Jul 2010) | 1 line
    
    don't require the presence of __getformat__ or __setformat__; use requires_IEEE_754 globally
  ........
................
  r82465 | brett.cannon | 2010-07-03 02:32:48 +0100 (Sat, 03 Jul 2010) | 3 lines
  
  Make test_import a little bit more robust for cleaning up after itself in the
  face of a failure.
................
  r82467 | alexander.belopolsky | 2010-07-03 04:35:27 +0100 (Sat, 03 Jul 2010) | 9 lines
  
  Merged revisions 82466 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82466 | alexander.belopolsky | 2010-07-02 23:27:12 -0400 (Fri, 02 Jul 2010) | 1 line
    
    Revert r81681 (issue 8810).
  ........
................
  r82468 | ezio.melotti | 2010-07-03 05:52:19 +0100 (Sat, 03 Jul 2010) | 1 line
  
  Update comment about surrogates.
................
  r82471 | nick.coghlan | 2010-07-03 08:36:51 +0100 (Sat, 03 Jul 2010) | 1 line
  
  Issue 6507: accept source strings directly in dis.dis(). Original patch by Daniel Urban
................
  r82472 | nick.coghlan | 2010-07-03 08:38:28 +0100 (Sat, 03 Jul 2010) | 1 line
  
  Issue 6507: missing patch submitter from NEWS entry in previous checkin
................
  r82478 | mark.dickinson | 2010-07-03 10:17:16 +0100 (Sat, 03 Jul 2010) | 9 lines
  
  Merged revisions 82476 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82476 | mark.dickinson | 2010-07-03 10:15:09 +0100 (Sat, 03 Jul 2010) | 1 line
    
    Fix typo in sys.float_info docs.
  ........
................
  r82480 | georg.brandl | 2010-07-03 11:21:50 +0100 (Sat, 03 Jul 2010) | 1 line
  
  Wrap and use the correct directive.
................
  r82481 | georg.brandl | 2010-07-03 11:22:10 +0100 (Sat, 03 Jul 2010) | 1 line
  
  Use the right role.
................
  r82484 | georg.brandl | 2010-07-03 11:26:17 +0100 (Sat, 03 Jul 2010) | 9 lines
  
  Recorded merge of revisions 82474 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82474 | georg.brandl | 2010-07-03 10:40:13 +0200 (Sa, 03 Jul 2010) | 1 line
    
    Fix role name.
  ........
................
  r82485 | georg.brandl | 2010-07-03 11:26:54 +0100 (Sat, 03 Jul 2010) | 9 lines
  
  Merged revisions 82483 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82483 | georg.brandl | 2010-07-03 12:25:54 +0200 (Sa, 03 Jul 2010) | 1 line
    
    Add link to bytecode docs.
  ........
................
  r82487 | georg.brandl | 2010-07-03 11:33:26 +0100 (Sat, 03 Jul 2010) | 1 line
  
  Fix markup.
................
  r82488 | georg.brandl | 2010-07-03 11:41:33 +0100 (Sat, 03 Jul 2010) | 1 line
  
  Remove the need for a "()" empty argument list after opcodes.
................
  r82495 | victor.stinner | 2010-07-03 14:44:22 +0100 (Sat, 03 Jul 2010) | 10 lines
  
  Merged revisions 82492 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82492 | victor.stinner | 2010-07-03 15:36:19 +0200 (sam., 03 juil. 2010) | 3 lines
    
    Issue #7673: Fix security vulnerability (CVE-2010-2089) in the audioop module,
    ensure that the input string length is a multiple of the frame size
  ........
................
  r82507 | benjamin.peterson | 2010-07-03 16:09:20 +0100 (Sat, 03 Jul 2010) | 3 lines
  
  Removed merge tracking for "svnmerge" for 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k/python/trunk
................
  r82510 | senthil.kumaran | 2010-07-03 18:48:22 +0100 (Sat, 03 Jul 2010) | 4 lines
  
  Fix Issue5468 - urlencode to handle bytes and other alternate encodings.
  (Extensive tests provided). Patch by Dan Mahn.
................
  r82514 | alexander.belopolsky | 2010-07-03 21:35:53 +0100 (Sat, 03 Jul 2010) | 3 lines
  
  Issue #9094: Make python -m pickletools disassemble pickles given in
  the command line.
................
  r82515 | mark.dickinson | 2010-07-03 22:00:51 +0100 (Sat, 03 Jul 2010) | 1 line
  
  Remove nonexistent tools from Tools README.
................
  r82516 | brett.cannon | 2010-07-03 22:37:28 +0100 (Sat, 03 Jul 2010) | 2 lines
  
  Ignore __pycache__.
................
  r82517 | alexander.belopolsky | 2010-07-03 22:42:47 +0100 (Sat, 03 Jul 2010) | 3 lines
  
  Issue #9151: Demo/classes/Dates.py does not work in 3.x
  Made minimal changes to make included test pass.
................
  r82518 | brett.cannon | 2010-07-03 22:48:25 +0100 (Sat, 03 Jul 2010) | 34 lines
  
  Make importlib.abc.SourceLoader the primary mechanism for importlib.
  
  This required moving the class from importlib/abc.py into
  importlib/_bootstrap.py and jiggering some code to work better with the class.
  This included changing how the file finder worked to better meet import
  semantics. This also led to fixing importlib to handle the empty string from
  sys.path as import currently does (and making me wish we didn't support that
  instead just required people to insert '.' instead to represent cwd).
  
  It also required making the new set_data abstractmethod create
  any needed subdirectories implicitly thanks to __pycache__ (it was either this
  or grow the SourceLoader ABC to gain an 'exists' method and either a mkdir
  method or have set_data with no data arg mean to create a directory).
  
  Lastly, as an optimization the file loaders cache the file path where the
  finder found something to use for loading (this is thanks to having a
  sourceless loader separate from the source loader to simplify the code and
  cut out stat calls).
  Unfortunately test_runpy assumed a loader would always work for a module, even
  if you changed from underneath it what it was expected to work with. By simply
  dropping the previous loader in test_runpy so the proper loader can be returned
  by the finder fixed the failure.
  
  At this point importlib deviates from import on two points:
  
  1. The exception raised when trying to import a file is different (import does
  an explicit file check to print a special message, importlib just says the path
  cannot be imported as if it was just some module name).
  
  2. the co_filename on a code object is not being set to where bytecode was
  actually loaded from instead of where the marshalled code object originally
  came from (a solution for this has already been agreed upon on python-dev but has
  not been implemented yet; issue8611).
................
  r82519 | brett.cannon | 2010-07-03 22:50:52 +0100 (Sat, 03 Jul 2010) | 1 line
  
  Trailing whitespace is bad for .rst files.
................
  r82520 | brett.cannon | 2010-07-03 23:03:16 +0100 (Sat, 03 Jul 2010) | 1 line
  
  Fix a spelling mistake in a comment.
................
  r82521 | alexander.belopolsky | 2010-07-03 23:05:41 +0100 (Sat, 03 Jul 2010) | 5 lines
  
  Issue #9151: Demo/classes/Dates.py does not work in 3.x Converted
  descriptive comment into a docstring.  Cast attributes to int in
  __init__.  Use __new__ instead of deleting attributes to
  "uninitialize".
................
  r82522 | brett.cannon | 2010-07-03 23:18:47 +0100 (Sat, 03 Jul 2010) | 34 lines
  
  Make importlib.abc.SourceLoader the primary mechanism for importlib.
  
  This required moving the class from importlib/abc.py into
  importlib/_bootstrap.py and jiggering some code to work better with the class.
  This included changing how the file finder worked to better meet import
  semantics. This also led to fixing importlib to handle the empty string from
  sys.path as import currently does (and making me wish we didn't support that
  instead just required people to insert '.' instead to represent cwd).
  
  It also required making the new set_data abstractmethod create
  any needed subdirectories implicitly thanks to __pycache__ (it was either this
  or grow the SourceLoader ABC to gain an 'exists' method and either a mkdir
  method or have set_data with no data arg mean to create a directory).
  
  Lastly, as an optimization the file loaders cache the file path where the
  finder found something to use for loading (this is thanks to having a
  sourceless loader separate from the source loader to simplify the code and
  cut out stat calls).
  Unfortunately test_runpy assumed a loader would always work for a module, even
  if you changed from underneath it what it was expected to work with. By simply
  dropping the previous loader in test_runpy so the proper loader can be returned
  by the finder fixed the failure.
  
  At this point importlib deviates from import on two points:
  
  1. The exception raised when trying to import a file is different (import does
  an explicit file check to print a special message, importlib just says the path
  cannot be imported as if it was just some module name).
  
  2. the co_filename on a code object is not being set to where bytecode was
  actually loaded from instead of where the marshalled code object originally
  came from (a solution for this has already been agreed upon on python-dev but has
  not been implemented yet; issue8611).
................
  r82523 | brett.cannon | 2010-07-03 23:32:41 +0100 (Sat, 03 Jul 2010) | 34 lines
  
  Make importlib.abc.SourceLoader the primary mechanism for importlib.
  
  This required moving the class from importlib/abc.py into
  importlib/_bootstrap.py and jiggering some code to work better with the class.
  This included changing how the file finder worked to better meet import
  semantics. This also led to fixing importlib to handle the empty string from
  sys.path as import currently does (and making me wish we didn't support that
  instead just required people to insert '.' instead to represent cwd).
  
  It also required making the new set_data abstractmethod create
  any needed subdirectories implicitly thanks to __pycache__ (it was either this
  or grow the SourceLoader ABC to gain an 'exists' method and either a mkdir
  method or have set_data with no data arg mean to create a directory).
  
  Lastly, as an optimization the file loaders cache the file path where the
  finder found something to use for loading (this is thanks to having a
  sourceless loader separate from the source loader to simplify the code and
  cut out stat calls).
  Unfortunately test_runpy assumed a loader would always work for a module, even
  if you changed from underneath it what it was expected to work with. By simply
  dropping the previous loader in test_runpy so the proper loader can be returned
  by the finder fixed the failure.
  
  At this point importlib deviates from import on two points:
  
  1. The exception raised when trying to import a file is different (import does
  an explicit file check to print a special message, importlib just says the path
  cannot be imported as if it was just some module name).
  
  2. the co_filename on a code object is not being set to where bytecode was
  actually loaded from instead of where the marshalled code object originally
  came from (a solution for this has already been agreed upon on python-dev but has
  not been implemented yet; issue8611).
................
  r82524 | alexander.belopolsky | 2010-07-03 23:36:06 +0100 (Sat, 03 Jul 2010) | 1 line
  
  Made minimal modifications to pass included tests
................
  r82526 | georg.brandl | 2010-07-04 08:25:18 +0100 (Sun, 04 Jul 2010) | 1 line
  
  Remove old and unsafe Dbm demo class.
................
  r82527 | mark.dickinson | 2010-07-04 11:15:11 +0100 (Sun, 04 Jul 2010) | 1 line
  
  Fix refleak in Modules/audioop.c.
................
  r82533 | alexander.belopolsky | 2010-07-04 17:28:08 +0100 (Sun, 04 Jul 2010) | 1 line
  
  Issue #9152: Removed dead code in datetime module
................
  r82535 | benjamin.peterson | 2010-07-04 17:33:07 +0100 (Sun, 04 Jul 2010) | 3 lines
  
  Removed merge tracking for "svnmerge" for 
  svn+ssh://pythondev at svn.python.org/sandbox/trunk/2to3/lib2to3
................
  r82537 | mark.dickinson | 2010-07-04 17:37:31 +0100 (Sun, 04 Jul 2010) | 2 lines
  
  Issue #9128: Fix validation of class decorators in parser module.
................
  r82538 | benjamin.peterson | 2010-07-04 17:37:39 +0100 (Sun, 04 Jul 2010) | 3 lines
  
  Initialized merge tracking via "svnmerge" with revisions "0-80937" from 
  svn+ssh://pythondev at svn.python.org/sandbox/trunk/2to3
................
  r82540 | benjamin.peterson | 2010-07-04 17:40:27 +0100 (Sun, 04 Jul 2010) | 3 lines
  
  Removed merge tracking for "svnmerge" for 
  svn+ssh://pythondev at svn.python.org/sandbox/trunk/2to3
................
  r82541 | benjamin.peterson | 2010-07-04 17:41:33 +0100 (Sun, 04 Jul 2010) | 3 lines
  
  Initialized merge tracking via "svnmerge" with revisions "0-80937" from 
  svn+ssh://pythondev at svn.python.org/sandbox/trunk/2to3/lib2to3
................
  r82542 | benjamin.peterson | 2010-07-04 17:44:15 +0100 (Sun, 04 Jul 2010) | 17 lines
  
  Merged revisions 81478,82530-82531 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/sandbox/trunk/2to3/lib2to3
  
  ........
    r81478 | benjamin.peterson | 2010-05-22 13:47:39 -0500 (Sat, 22 May 2010) | 1 line
    
    ensure doctests have some future_features
  ........
    r82530 | benjamin.peterson | 2010-07-04 11:11:41 -0500 (Sun, 04 Jul 2010) | 1 line
    
    simplify ignore star imports from itertools #8892
  ........
    r82531 | benjamin.peterson | 2010-07-04 11:13:20 -0500 (Sun, 04 Jul 2010) | 1 line
    
    wrap with parenthesis not \
  ........
................
  r82547 | alexander.belopolsky | 2010-07-04 18:00:20 +0100 (Sun, 04 Jul 2010) | 3 lines
  
  Issue #9118: help(None) will now return NoneType doc instead of
  starting interactive help.
................
  r82550 | alexander.belopolsky | 2010-07-04 18:38:32 +0100 (Sun, 04 Jul 2010) | 1 line
  
  Fixed doctests
................
  r82551 | alexander.belopolsky | 2010-07-04 18:47:30 +0100 (Sun, 04 Jul 2010) | 1 line
  
  Fixed the test
................
  r82552 | mark.dickinson | 2010-07-04 19:11:51 +0100 (Sun, 04 Jul 2010) | 2 lines
  
  Issue #9130: Fix validation of relative imports in parser module.
................
  r82553 | mark.dickinson | 2010-07-04 19:15:26 +0100 (Sun, 04 Jul 2010) | 1 line
  
  Fix symbol numbers in test_parser test.
................
  r82555 | mark.dickinson | 2010-07-04 19:38:57 +0100 (Sun, 04 Jul 2010) | 2 lines
  
  Issue #9130: Validate ellipsis tokens in relative imports.
................
  r82557 | mark.dickinson | 2010-07-04 19:49:18 +0100 (Sun, 04 Jul 2010) | 1 line
  
  Make Demo/parser/test_parser.py run.
................
  r82558 | mark.dickinson | 2010-07-04 20:23:49 +0100 (Sun, 04 Jul 2010) | 1 line
  
  Remove coding cookie from heapq.py.
................
  r82559 | mark.dickinson | 2010-07-04 20:27:12 +0100 (Sun, 04 Jul 2010) | 1 line
  
  Re-encode shlex.py in UTF-8, and remove coding cookie.
................
  r82560 | mark.dickinson | 2010-07-04 21:07:09 +0100 (Sun, 04 Jul 2010) | 1 line
  
  Revert -r82559;  it's not clear that this is the right thing to do, and the change obscures the original intentions.
................
  r82561 | andrew.kuchling | 2010-07-04 21:47:00 +0100 (Sun, 04 Jul 2010) | 1 line
  
  Add brief explanation; remove no-longer-working link
................
  r82567 | brett.cannon | 2010-07-04 23:05:34 +0100 (Sun, 04 Jul 2010) | 5 lines
  
  Do some basic refactoring. Along the way also list what files had their
  whitespace cleaned up and do the proper plurality for "N file(s)" based on N.
  
  Refactoring closes issue8912. Thanks to ??ric Araujo for the patch.
................
  r82570 | senthil.kumaran | 2010-07-05 12:41:42 +0100 (Mon, 05 Jul 2010) | 3 lines
  
  Fix: Issue9091 Minor documentation clarification.
................
  r82573 | senthil.kumaran | 2010-07-05 13:00:56 +0100 (Mon, 05 Jul 2010) | 3 lines
  
  Fix the docstrings of the capitalize method.
................
  r82576 | benjamin.peterson | 2010-07-05 15:59:40 +0100 (Mon, 05 Jul 2010) | 1 line
  
  pydoc still has a silly encoding
................
  r82577 | benjamin.peterson | 2010-07-05 16:01:22 +0100 (Mon, 05 Jul 2010) | 1 line
  
  cleanup basicsize logic #3268
................
  r82578 | alexander.belopolsky | 2010-07-05 16:05:33 +0100 (Mon, 05 Jul 2010) | 1 line
  
  Added more tests for utctimetuple()
................
  r82585 | benjamin.peterson | 2010-07-05 18:11:05 +0100 (Mon, 05 Jul 2010) | 1 line
  
  looking up on the type is correct, so this isn't an XXX
................
  r82587 | georg.brandl | 2010-07-05 18:48:38 +0100 (Mon, 05 Jul 2010) | 1 line
  
  #9166: NotImplemented is not an exception.
................
  r82589 | alexander.belopolsky | 2010-07-05 18:57:31 +0100 (Mon, 05 Jul 2010) | 1 line
  
  Added two more test cases for datetime
................
  r82591 | benjamin.peterson | 2010-07-05 21:04:54 +0100 (Mon, 05 Jul 2010) | 1 line
  
  untabify
................
  r82593 | georg.brandl | 2010-07-05 21:13:06 +0100 (Mon, 05 Jul 2010) | 1 line
  
  Remove Dbm.
................
  r82594 | georg.brandl | 2010-07-05 21:13:41 +0100 (Mon, 05 Jul 2010) | 1 line
  
  Update Vec class constructor, remove indirection via function, use operator module.
................
  r82595 | mark.dickinson | 2010-07-05 21:14:26 +0100 (Mon, 05 Jul 2010) | 1 line
  
  Post-detabification cleanup.
................
  r82597 | victor.stinner | 2010-07-05 22:36:21 +0100 (Mon, 05 Jul 2010) | 2 lines
  
  #9158: Fix y* format of PyArg_Parse*() functions documentation
................
  r82599 | alexander.belopolsky | 2010-07-05 22:44:05 +0100 (Mon, 05 Jul 2010) | 1 line
  
  "Modernized" the demo a little.
................
  r82600 | brett.cannon | 2010-07-05 23:11:16 +0100 (Mon, 05 Jul 2010) | 1 line
  
  Fix test_xmlrpc_net to no longer fail since there are no more buildbots for trunk.
................
  r82601 | senthil.kumaran | 2010-07-06 03:08:36 +0100 (Tue, 06 Jul 2010) | 3 lines
  
  Change 'lowered' to 'lowercased'
................
  r82604 | matthias.klose | 2010-07-06 11:53:30 +0100 (Tue, 06 Jul 2010) | 2 lines
  
  - sysmodule.c (get_hash_info): Define as static function.
................
  r82607 | mark.dickinson | 2010-07-06 16:00:40 +0100 (Tue, 06 Jul 2010) | 1 line
  
  Indentation and PEP 7 fixes.
................
  r82609 | vinay.sajip | 2010-07-06 16:08:55 +0100 (Tue, 06 Jul 2010) | 1 line
  
  Minor improvements to logging documentation.
................
  r82610 | mark.dickinson | 2010-07-06 16:11:44 +0100 (Tue, 06 Jul 2010) | 1 line
  
  Style nit.
................


Added:
   python/branches/py3k-dtoa/Demo/parser/test_unparse.py
      - copied unchanged from r82610, /python/branches/py3k/Demo/parser/test_unparse.py
   python/branches/py3k-dtoa/Doc/documenting/building.rst
      - copied unchanged from r82610, /python/branches/py3k/Doc/documenting/building.rst
   python/branches/py3k-dtoa/Lib/encodings/cp858.py
      - copied unchanged from r82610, /python/branches/py3k/Lib/encodings/cp858.py
   python/branches/py3k-dtoa/Modules/_ctypes/libffi_msvc/LICENSE
      - copied unchanged from r82610, /python/branches/py3k/Modules/_ctypes/libffi_msvc/LICENSE
   python/branches/py3k-dtoa/Modules/_ctypes/libffi_msvc/README
      - copied unchanged from r82610, /python/branches/py3k/Modules/_ctypes/libffi_msvc/README
   python/branches/py3k-dtoa/Modules/_ctypes/libffi_msvc/README.ctypes
      - copied unchanged from r82610, /python/branches/py3k/Modules/_ctypes/libffi_msvc/README.ctypes
   python/branches/py3k-dtoa/Modules/_ctypes/libffi_msvc/ffi.c
      - copied unchanged from r82610, /python/branches/py3k/Modules/_ctypes/libffi_msvc/ffi.c
   python/branches/py3k-dtoa/Modules/_ctypes/libffi_msvc/ffi_common.h
      - copied unchanged from r82610, /python/branches/py3k/Modules/_ctypes/libffi_msvc/ffi_common.h
   python/branches/py3k-dtoa/Modules/_ctypes/libffi_msvc/ffitarget.h
      - copied unchanged from r82610, /python/branches/py3k/Modules/_ctypes/libffi_msvc/ffitarget.h
   python/branches/py3k-dtoa/Modules/_ctypes/libffi_msvc/prep_cif.c
      - copied unchanged from r82610, /python/branches/py3k/Modules/_ctypes/libffi_msvc/prep_cif.c
   python/branches/py3k-dtoa/Modules/_ctypes/libffi_msvc/types.c
      - copied unchanged from r82610, /python/branches/py3k/Modules/_ctypes/libffi_msvc/types.c
   python/branches/py3k-dtoa/Modules/_ctypes/libffi_msvc/win32.c
      - copied unchanged from r82610, /python/branches/py3k/Modules/_ctypes/libffi_msvc/win32.c
   python/branches/py3k-dtoa/Modules/_ctypes/libffi_msvc/win64.asm
      - copied unchanged from r82610, /python/branches/py3k/Modules/_ctypes/libffi_msvc/win64.asm
   python/branches/py3k-dtoa/Modules/_time.c
      - copied unchanged from r82610, /python/branches/py3k/Modules/_time.c
   python/branches/py3k-dtoa/Modules/_time.h
      - copied unchanged from r82610, /python/branches/py3k/Modules/_time.h
Removed:
   python/branches/py3k-dtoa/Demo/classes/Dbm.py
   python/branches/py3k-dtoa/Tools/webchecker/
Modified:
   python/branches/py3k-dtoa/   (props changed)
   python/branches/py3k-dtoa/Demo/classes/Complex.py
   python/branches/py3k-dtoa/Demo/classes/Dates.py
   python/branches/py3k-dtoa/Demo/classes/README
   python/branches/py3k-dtoa/Demo/classes/Range.py
   python/branches/py3k-dtoa/Demo/classes/Rev.py
   python/branches/py3k-dtoa/Demo/classes/Vec.py
   python/branches/py3k-dtoa/Demo/embed/Makefile
   python/branches/py3k-dtoa/Demo/embed/demo.c
   python/branches/py3k-dtoa/Demo/md5test/md5driver.py
   python/branches/py3k-dtoa/Demo/parser/test_parser.py
   python/branches/py3k-dtoa/Demo/parser/unparse.py
   python/branches/py3k-dtoa/Doc/README.txt
   python/branches/py3k-dtoa/Doc/c-api/arg.rst
   python/branches/py3k-dtoa/Doc/c-api/datetime.rst
   python/branches/py3k-dtoa/Doc/c-api/init.rst
   python/branches/py3k-dtoa/Doc/c-api/intro.rst
   python/branches/py3k-dtoa/Doc/c-api/number.rst
   python/branches/py3k-dtoa/Doc/c-api/unicode.rst
   python/branches/py3k-dtoa/Doc/data/refcounts.dat
   python/branches/py3k-dtoa/Doc/distutils/builtdist.rst
   python/branches/py3k-dtoa/Doc/documenting/index.rst
   python/branches/py3k-dtoa/Doc/documenting/markup.rst
   python/branches/py3k-dtoa/Doc/documenting/style.rst
   python/branches/py3k-dtoa/Doc/faq/gui.rst
   python/branches/py3k-dtoa/Doc/glossary.rst
   python/branches/py3k-dtoa/Doc/howto/functional.rst
   python/branches/py3k-dtoa/Doc/howto/unicode.rst
   python/branches/py3k-dtoa/Doc/library/2to3.rst
   python/branches/py3k-dtoa/Doc/library/calendar.rst
   python/branches/py3k-dtoa/Doc/library/codecs.rst
   python/branches/py3k-dtoa/Doc/library/contextlib.rst
   python/branches/py3k-dtoa/Doc/library/datetime.rst
   python/branches/py3k-dtoa/Doc/library/dis.rst
   python/branches/py3k-dtoa/Doc/library/http.server.rst
   python/branches/py3k-dtoa/Doc/library/importlib.rst
   python/branches/py3k-dtoa/Doc/library/logging.rst
   python/branches/py3k-dtoa/Doc/library/os.path.rst
   python/branches/py3k-dtoa/Doc/library/os.rst
   python/branches/py3k-dtoa/Doc/library/pdb.rst
   python/branches/py3k-dtoa/Doc/library/re.rst
   python/branches/py3k-dtoa/Doc/library/socket.rst
   python/branches/py3k-dtoa/Doc/library/ssl.rst
   python/branches/py3k-dtoa/Doc/library/stdtypes.rst
   python/branches/py3k-dtoa/Doc/library/string.rst
   python/branches/py3k-dtoa/Doc/library/struct.rst
   python/branches/py3k-dtoa/Doc/library/sys.rst
   python/branches/py3k-dtoa/Doc/library/sysconfig.rst
   python/branches/py3k-dtoa/Doc/library/unittest.rst
   python/branches/py3k-dtoa/Doc/library/urllib.parse.rst
   python/branches/py3k-dtoa/Doc/library/urllib.request.rst
   python/branches/py3k-dtoa/Doc/library/warnings.rst
   python/branches/py3k-dtoa/Doc/reference/compound_stmts.rst
   python/branches/py3k-dtoa/Doc/reference/datamodel.rst
   python/branches/py3k-dtoa/Doc/reference/introduction.rst
   python/branches/py3k-dtoa/Doc/reference/lexical_analysis.rst
   python/branches/py3k-dtoa/Doc/tools/sphinxext/pyspecific.py
   python/branches/py3k-dtoa/Doc/tutorial/classes.rst
   python/branches/py3k-dtoa/Doc/tutorial/controlflow.rst
   python/branches/py3k-dtoa/Doc/tutorial/introduction.rst
   python/branches/py3k-dtoa/Doc/using/mac.rst
   python/branches/py3k-dtoa/Doc/whatsnew/2.7.rst
   python/branches/py3k-dtoa/Doc/whatsnew/3.0.rst
   python/branches/py3k-dtoa/Doc/whatsnew/3.2.rst
   python/branches/py3k-dtoa/Grammar/Grammar
   python/branches/py3k-dtoa/Include/longobject.h
   python/branches/py3k-dtoa/Include/pyctype.h
   python/branches/py3k-dtoa/Include/traceback.h
   python/branches/py3k-dtoa/Include/unicodeobject.h
   python/branches/py3k-dtoa/Lib/_strptime.py
   python/branches/py3k-dtoa/Lib/contextlib.py
   python/branches/py3k-dtoa/Lib/ctypes/__init__.py
   python/branches/py3k-dtoa/Lib/ctypes/test/test_callbacks.py
   python/branches/py3k-dtoa/Lib/ctypes/test/test_win32.py
   python/branches/py3k-dtoa/Lib/dis.py
   python/branches/py3k-dtoa/Lib/distutils/msvc9compiler.py
   python/branches/py3k-dtoa/Lib/distutils/unixccompiler.py
   python/branches/py3k-dtoa/Lib/encodings/aliases.py
   python/branches/py3k-dtoa/Lib/heapq.py
   python/branches/py3k-dtoa/Lib/http/cookiejar.py
   python/branches/py3k-dtoa/Lib/idlelib/textView.py
   python/branches/py3k-dtoa/Lib/importlib/__init__.py
   python/branches/py3k-dtoa/Lib/importlib/_bootstrap.py
   python/branches/py3k-dtoa/Lib/importlib/abc.py
   python/branches/py3k-dtoa/Lib/importlib/test/extension/test_case_sensitivity.py
   python/branches/py3k-dtoa/Lib/importlib/test/extension/test_finder.py
   python/branches/py3k-dtoa/Lib/importlib/test/extension/test_loader.py
   python/branches/py3k-dtoa/Lib/importlib/test/extension/test_path_hook.py
   python/branches/py3k-dtoa/Lib/importlib/test/import_/test_path.py
   python/branches/py3k-dtoa/Lib/importlib/test/regrtest.py
   python/branches/py3k-dtoa/Lib/importlib/test/source/test_abc_loader.py
   python/branches/py3k-dtoa/Lib/importlib/test/source/test_case_sensitivity.py
   python/branches/py3k-dtoa/Lib/importlib/test/source/test_file_loader.py
   python/branches/py3k-dtoa/Lib/importlib/test/source/test_finder.py
   python/branches/py3k-dtoa/Lib/importlib/test/source/test_path_hook.py
   python/branches/py3k-dtoa/Lib/importlib/test/source/test_source_encoding.py
   python/branches/py3k-dtoa/Lib/importlib/test/test_abc.py
   python/branches/py3k-dtoa/Lib/importlib/test/test_api.py
   python/branches/py3k-dtoa/Lib/importlib/test/test_util.py
   python/branches/py3k-dtoa/Lib/inspect.py
   python/branches/py3k-dtoa/Lib/lib2to3/   (props changed)
   python/branches/py3k-dtoa/Lib/lib2to3/fixes/fix_itertools_imports.py
   python/branches/py3k-dtoa/Lib/lib2to3/tests/test_fixers.py
   python/branches/py3k-dtoa/Lib/logging/config.py
   python/branches/py3k-dtoa/Lib/macpath.py
   python/branches/py3k-dtoa/Lib/ntpath.py
   python/branches/py3k-dtoa/Lib/os2emxpath.py
   python/branches/py3k-dtoa/Lib/pdb.doc
   python/branches/py3k-dtoa/Lib/pdb.py
   python/branches/py3k-dtoa/Lib/pickletools.py
   python/branches/py3k-dtoa/Lib/posixpath.py
   python/branches/py3k-dtoa/Lib/pydoc.py
   python/branches/py3k-dtoa/Lib/site.py
   python/branches/py3k-dtoa/Lib/smtpd.py
   python/branches/py3k-dtoa/Lib/subprocess.py
   python/branches/py3k-dtoa/Lib/test/test_audioop.py
   python/branches/py3k-dtoa/Lib/test/test_codeccallbacks.py
   python/branches/py3k-dtoa/Lib/test/test_codecs.py
   python/branches/py3k-dtoa/Lib/test/test_complex.py
   python/branches/py3k-dtoa/Lib/test/test_contextlib.py
   python/branches/py3k-dtoa/Lib/test/test_datetime.py
   python/branches/py3k-dtoa/Lib/test/test_descr.py
   python/branches/py3k-dtoa/Lib/test/test_dict.py
   python/branches/py3k-dtoa/Lib/test/test_dis.py
   python/branches/py3k-dtoa/Lib/test/test_enumerate.py
   python/branches/py3k-dtoa/Lib/test/test_exceptions.py
   python/branches/py3k-dtoa/Lib/test/test_extcall.py
   python/branches/py3k-dtoa/Lib/test/test_float.py
   python/branches/py3k-dtoa/Lib/test/test_genericpath.py
   python/branches/py3k-dtoa/Lib/test/test_getargs2.py
   python/branches/py3k-dtoa/Lib/test/test_http_cookiejar.py
   python/branches/py3k-dtoa/Lib/test/test_imp.py
   python/branches/py3k-dtoa/Lib/test/test_import.py
   python/branches/py3k-dtoa/Lib/test/test_inspect.py
   python/branches/py3k-dtoa/Lib/test/test_iter.py
   python/branches/py3k-dtoa/Lib/test/test_keywordonlyarg.py
   python/branches/py3k-dtoa/Lib/test/test_logging.py
   python/branches/py3k-dtoa/Lib/test/test_math.py
   python/branches/py3k-dtoa/Lib/test/test_optparse.py
   python/branches/py3k-dtoa/Lib/test/test_os.py
   python/branches/py3k-dtoa/Lib/test/test_parser.py
   python/branches/py3k-dtoa/Lib/test/test_property.py
   python/branches/py3k-dtoa/Lib/test/test_pyexpat.py
   python/branches/py3k-dtoa/Lib/test/test_runpy.py
   python/branches/py3k-dtoa/Lib/test/test_strtod.py
   python/branches/py3k-dtoa/Lib/test/test_subprocess.py
   python/branches/py3k-dtoa/Lib/test/test_syntax.py
   python/branches/py3k-dtoa/Lib/test/test_tokenize.py
   python/branches/py3k-dtoa/Lib/test/test_tuple.py
   python/branches/py3k-dtoa/Lib/test/test_unicode.py
   python/branches/py3k-dtoa/Lib/test/test_urllib.py
   python/branches/py3k-dtoa/Lib/test/test_xmlrpc_net.py
   python/branches/py3k-dtoa/Lib/urllib/parse.py
   python/branches/py3k-dtoa/Lib/urllib/request.py
   python/branches/py3k-dtoa/Lib/uuid.py
   python/branches/py3k-dtoa/Lib/warnings.py
   python/branches/py3k-dtoa/Misc/ACKS
   python/branches/py3k-dtoa/Misc/NEWS
   python/branches/py3k-dtoa/Misc/Vim/vimrc
   python/branches/py3k-dtoa/Modules/_ctypes/callproc.c
   python/branches/py3k-dtoa/Modules/_ctypes/libffi/fficonfig.py.in
   python/branches/py3k-dtoa/Modules/_ctypes/libffi_msvc/ffi.h
   python/branches/py3k-dtoa/Modules/_ctypes/libffi_msvc/fficonfig.h
   python/branches/py3k-dtoa/Modules/_math.c
   python/branches/py3k-dtoa/Modules/_scproxy.c
   python/branches/py3k-dtoa/Modules/_ssl.c
   python/branches/py3k-dtoa/Modules/_testcapimodule.c
   python/branches/py3k-dtoa/Modules/audioop.c
   python/branches/py3k-dtoa/Modules/datetimemodule.c
   python/branches/py3k-dtoa/Modules/mathmodule.c
   python/branches/py3k-dtoa/Modules/parsermodule.c
   python/branches/py3k-dtoa/Modules/posixmodule.c
   python/branches/py3k-dtoa/Modules/selectmodule.c
   python/branches/py3k-dtoa/Modules/timemodule.c
   python/branches/py3k-dtoa/Modules/xxmodule.c
   python/branches/py3k-dtoa/Objects/abstract.c
   python/branches/py3k-dtoa/Objects/bytes_methods.c
   python/branches/py3k-dtoa/Objects/descrobject.c
   python/branches/py3k-dtoa/Objects/enumobject.c
   python/branches/py3k-dtoa/Objects/exceptions.c
   python/branches/py3k-dtoa/Objects/floatobject.c
   python/branches/py3k-dtoa/Objects/frameobject.c
   python/branches/py3k-dtoa/Objects/typeobject.c
   python/branches/py3k-dtoa/Objects/unicodeobject.c
   python/branches/py3k-dtoa/PC/VS8.0/pythoncore.vcproj
   python/branches/py3k-dtoa/PCbuild/_ctypes.vcproj
   python/branches/py3k-dtoa/PCbuild/pythoncore.vcproj
   python/branches/py3k-dtoa/Parser/Python.asdl
   python/branches/py3k-dtoa/Parser/asdl_c.py
   python/branches/py3k-dtoa/Python/Python-ast.c
   python/branches/py3k-dtoa/Python/_warnings.c
   python/branches/py3k-dtoa/Python/ast.c
   python/branches/py3k-dtoa/Python/ceval.c
   python/branches/py3k-dtoa/Python/compile.c
   python/branches/py3k-dtoa/Python/getargs.c
   python/branches/py3k-dtoa/Python/import.c
   python/branches/py3k-dtoa/Python/pythonrun.c
   python/branches/py3k-dtoa/Python/sysmodule.c
   python/branches/py3k-dtoa/Python/traceback.c
   python/branches/py3k-dtoa/README
   python/branches/py3k-dtoa/Tools/README
   python/branches/py3k-dtoa/Tools/scripts/   (props changed)
   python/branches/py3k-dtoa/Tools/scripts/patchcheck.py
   python/branches/py3k-dtoa/configure
   python/branches/py3k-dtoa/setup.py

Modified: python/branches/py3k-dtoa/Demo/classes/Complex.py
==============================================================================
--- python/branches/py3k-dtoa/Demo/classes/Complex.py	(original)
+++ python/branches/py3k-dtoa/Demo/classes/Complex.py	Tue Jul  6 17:55:27 2010
@@ -151,13 +151,9 @@
             raise ValueError("can't convert Complex with nonzero im to float")
         return float(self.re)
 
-    def __cmp__(self, other):
+    def __eq__(self, other):
         other = ToComplex(other)
-        return cmp((self.re, self.im), (other.re, other.im))
-
-    def __rcmp__(self, other):
-        other = ToComplex(other)
-        return cmp(other, self)
+        return (self.re, self.im) == (other.re, other.im)
 
     def __bool__(self):
         return not (self.re == self.im == 0)
@@ -190,14 +186,14 @@
 
     __rmul__ = __mul__
 
-    def __div__(self, other):
+    def __truediv__(self, other):
         other = ToComplex(other)
         d = float(other.re*other.re + other.im*other.im)
         if not d: raise ZeroDivisionError('Complex division')
         return Complex((self.re*other.re + self.im*other.im) / d,
                        (self.im*other.re - self.re*other.im) / d)
 
-    def __rdiv__(self, other):
+    def __rtruediv__(self, other):
         other = ToComplex(other)
         return other / self
 
@@ -226,8 +222,9 @@
     print('       ', a, 'and', b, end=' ')
     try:
         result = eval(expr)
-    except:
-        result = sys.exc_info()[0]
+    except Exception as e:
+        print('!!\t!!\t!! error: {}'.format(e))
+        return
     print('->', result)
     if isinstance(result, str) or isinstance(value, str):
         ok = (result == value)
@@ -295,13 +292,6 @@
                     (Complex(1), Complex(0,10), 1),
                     (2, Complex(4,0), 16),
             ],
-            'cmp(a,b)': [
-                    (1, 10, -1),
-                    (1, Complex(0,10), 1),
-                    (Complex(0,10), 1, -1),
-                    (Complex(0,10), Complex(1), -1),
-                    (Complex(1), Complex(0,10), 1),
-            ],
     }
     for expr in sorted(testsuite):
         print(expr + ':')

Modified: python/branches/py3k-dtoa/Demo/classes/Dates.py
==============================================================================
--- python/branches/py3k-dtoa/Demo/classes/Dates.py	(original)
+++ python/branches/py3k-dtoa/Demo/classes/Dates.py	Tue Jul  6 17:55:27 2010
@@ -1,45 +1,47 @@
-# Class Date supplies date objects that support date arithmetic.
-#
-# Date(month,day,year) returns a Date object.  An instance prints as,
-# e.g., 'Mon 16 Aug 1993'.
-#
-# Addition, subtraction, comparison operators, min, max, and sorting
-# all work as expected for date objects:  int+date or date+int returns
-# the date `int' days from `date'; date+date raises an exception;
-# date-int returns the date `int' days before `date'; date2-date1 returns
-# an integer, the number of days from date1 to date2; int-date raises an
-# exception; date1 < date2 is true iff date1 occurs before date2 (&
-# similarly for other comparisons); min(date1,date2) is the earlier of
-# the two dates and max(date1,date2) the later; and date objects can be
-# used as dictionary keys.
-#
-# Date objects support one visible method, date.weekday().  This returns
-# the day of the week the date falls on, as a string.
-#
-# Date objects also have 4 read-only data attributes:
-#   .month  in 1..12
-#   .day    in 1..31
-#   .year   int or long int
-#   .ord    the ordinal of the date relative to an arbitrary staring point
-#
-# The Dates module also supplies function today(), which returns the
-# current date as a date object.
-#
-# Those entranced by calendar trivia will be disappointed, as no attempt
-# has been made to accommodate the Julian (etc) system.  On the other
-# hand, at least this package knows that 2000 is a leap year but 2100
-# isn't, and works fine for years with a hundred decimal digits <wink>.
-
-# Tim Peters   tim at ksr.com
-# not speaking for Kendall Square Research Corp
+"""
+Class Date supplies date objects that support date arithmetic.
 
-# Adapted to Python 1.1 (where some hacks to overcome coercion are unnecessary)
-# by Guido van Rossum
+Date(month,day,year) returns a Date object.  An instance prints as,
+e.g., 'Mon 16 Aug 1993'.
 
-# Note that as of Python 2.3, a datetime module is included in the stardard
-# library.
+Addition, subtraction, comparison operators, min, max, and sorting
+all work as expected for date objects:  int+date or date+int returns
+the date `int' days from `date'; date+date raises an exception;
+date-int returns the date `int' days before `date'; date2-date1 returns
+an integer, the number of days from date1 to date2; int-date raises an
+exception; date1 < date2 is true iff date1 occurs before date2 (&
+similarly for other comparisons); min(date1,date2) is the earlier of
+the two dates and max(date1,date2) the later; and date objects can be
+used as dictionary keys.
+
+Date objects support one visible method, date.weekday().  This returns
+the day of the week the date falls on, as a string.
+
+Date objects also have 4 read-only data attributes:
+  .month  in 1..12
+  .day    in 1..31
+  .year   int or long int
+  .ord    the ordinal of the date relative to an arbitrary staring point
+
+The Dates module also supplies function today(), which returns the
+current date as a date object.
+
+Those entranced by calendar trivia will be disappointed, as no attempt
+has been made to accommodate the Julian (etc) system.  On the other
+hand, at least this package knows that 2000 is a leap year but 2100
+isn't, and works fine for years with a hundred decimal digits <wink>.
+
+Tim Peters   tim at ksr.com
+not speaking for Kendall Square Research Corp
+
+Adapted to Python 1.1 (where some hacks to overcome coercion are unnecessary)
+by Guido van Rossum
+
+Note that as of Python 2.3, a datetime module is included in the stardard
+library.
+"""
 
-# vi:set tabsize=8:
+import functools
 
 _MONTH_NAMES = [ 'January', 'February', 'March', 'April', 'May',
                  'June', 'July', 'August', 'September', 'October',
@@ -57,8 +59,6 @@
     dbm = dbm + dim
 del dbm, dim
 
-_INT_TYPES = type(1), type(1)
-
 def _is_leap(year):           # 1 if leap year, else 0
     if year % 4 != 0: return 0
     if year % 400 == 0: return 1
@@ -85,11 +85,12 @@
 _DI400Y = _days_before_year(400)      # number of days in 400 years
 
 def _num2date(n):             # return date with ordinal n
-    if type(n) not in _INT_TYPES:
+    if not isinstance(n, int):
         raise TypeError('argument must be integer: %r' % type(n))
 
-    ans = Date(1,1,1)   # arguments irrelevant; just getting a Date obj
-    del ans.ord, ans.month, ans.day, ans.year # un-initialize it
+    # Get uninitialized Date object.  This is necesary because once
+    # attributes are set, they cannot be changed.
+    ans = Date.__new__(Date)
     ans.ord = n
 
     n400 = (n-1)//_DI400Y                # # of 400-year blocks preceding
@@ -99,11 +100,7 @@
     if dby >= n:
         more = more - 1
         dby = dby - _days_in_year(more)
-    year, n = year + more, int(n - dby)
-
-    try: year = int(year)               # chop to int, if it fits
-    except (ValueError, OverflowError): pass
-
+    year, n = year + more, n - dby
     month = min(n//29 + 1, 12)
     dbm = _days_before_month(month, year)
     if dbm >= n:
@@ -114,9 +111,9 @@
     return ans
 
 def _num2day(n):      # return weekday name of day with ordinal n
-    return _DAY_NAMES[ int(n % 7) ]
-
+    return _DAY_NAMES[n % 7]
 
+ at functools.total_ordering
 class Date:
     def __init__(self, month, day, year):
         if not 1 <= month <= 12:
@@ -124,7 +121,7 @@
         dim = _days_in_month(month, year)
         if not 1 <= day <= dim:
             raise ValueError('day must be in 1..%r: %r' % (dim, day))
-        self.month, self.day, self.year = month, day, year
+        self.month, self.day, self.year = map(int, (month, day, year))
         self.ord = _date2num(self)
 
     # don't allow setting existing attributes
@@ -133,8 +130,11 @@
             raise AttributeError('read-only attribute ' + name)
         self.__dict__[name] = value
 
-    def __cmp__(self, other):
-        return cmp(self.ord, other.ord)
+    def __eq__(self, other):
+        return self.ord == other.ord
+
+    def __lt__(self, other):
+        return self.ord < other.ord
 
     # define a hash function so dates can be used as dictionary keys
     def __hash__(self):
@@ -150,14 +150,14 @@
 
     # Python 1.1 coerces neither int+date nor date+int
     def __add__(self, n):
-        if type(n) not in _INT_TYPES:
+        if not isinstance(n, int):
             raise TypeError('can\'t add %r to date' % type(n))
         return _num2date(self.ord + n)
     __radd__ = __add__ # handle int+date
 
     # Python 1.1 coerces neither date-int nor date-date
     def __sub__(self, other):
-        if type(other) in _INT_TYPES:           # date-int
+        if isinstance(other, int):           # date-int
             return _num2date(self.ord - other)
         else:
             return self.ord - other.ord         # date-date

Deleted: python/branches/py3k-dtoa/Demo/classes/Dbm.py
==============================================================================
--- python/branches/py3k-dtoa/Demo/classes/Dbm.py	Tue Jul  6 17:55:27 2010
+++ (empty file)
@@ -1,66 +0,0 @@
-# A wrapper around the (optional) built-in class dbm, supporting keys
-# and values of almost any type instead of just string.
-# (Actually, this works only for keys and values that can be read back
-# correctly after being converted to a string.)
-
-
-class Dbm:
-
-    def __init__(self, filename, mode, perm):
-        import dbm.ndbm
-        self.db = dbm.ndbm.open(filename, mode, perm)
-
-    def __repr__(self):
-        s = ''
-        for key in self.keys():
-            t = repr(key) + ': ' + repr(self[key])
-            if s: t = ', ' + t
-            s = s + t
-        return '{' + s + '}'
-
-    def __len__(self):
-        return len(self.db)
-
-    def __getitem__(self, key):
-        return eval(self.db[repr(key)])
-
-    def __setitem__(self, key, value):
-        self.db[repr(key)] = repr(value)
-
-    def __delitem__(self, key):
-        del self.db[repr(key)]
-
-    def keys(self):
-        res = []
-        for key in self.db.keys():
-            res.append(eval(key))
-        return res
-
-    def has_key(self, key):
-        return repr(key) in self.db
-
-
-def test():
-    d = Dbm('@dbm', 'rw', 0o600)
-    print(d)
-    while 1:
-        try:
-            key = eval(input('key: '))
-            if key in d:
-                value = d[key]
-                print('currently:', value)
-            value = eval(input('value: '))
-            if value is None:
-                del d[key]
-            else:
-                d[key] = value
-        except KeyboardInterrupt:
-            print('')
-            print(d)
-        except EOFError:
-            print('[eof]')
-            break
-    print(d)
-
-
-test()

Modified: python/branches/py3k-dtoa/Demo/classes/README
==============================================================================
--- python/branches/py3k-dtoa/Demo/classes/README	(original)
+++ python/branches/py3k-dtoa/Demo/classes/README	Tue Jul  6 17:55:27 2010
@@ -2,7 +2,6 @@
 
 Complex.py	Complex numbers
 Dates.py	Date manipulation package by Tim Peters
-Dbm.py		Wrapper around built-in dbm, supporting	arbitrary values
 Range.py	Example of a generator: re-implement built-in range()
 Rev.py		Yield the reverse of a sequence
 Vec.py		A simple vector class

Modified: python/branches/py3k-dtoa/Demo/classes/Range.py
==============================================================================
--- python/branches/py3k-dtoa/Demo/classes/Range.py	(original)
+++ python/branches/py3k-dtoa/Demo/classes/Range.py	Tue Jul  6 17:55:27 2010
@@ -66,7 +66,7 @@
 def test():
     import time, builtins
     #Just a quick sanity check
-    correct_result = builtins.range(5, 100, 3)
+    correct_result = list(builtins.range(5, 100, 3))
     oldrange_result = list(oldrange(5, 100, 3))
     genrange_result = list(genrange(5, 100, 3))
     if genrange_result != correct_result or oldrange_result != correct_result:

Modified: python/branches/py3k-dtoa/Demo/classes/Rev.py
==============================================================================
--- python/branches/py3k-dtoa/Demo/classes/Rev.py	(original)
+++ python/branches/py3k-dtoa/Demo/classes/Rev.py	Tue Jul  6 17:55:27 2010
@@ -4,8 +4,7 @@
 
 It works on mutable or inmutable sequences.
 
->>> chars = list(Rev('Hello World!'))
->>> print ''.join(chars)
+>>> Rev('Hello World!')
 !dlroW olleH
 
 The .forw is so you can use anonymous sequences in __init__, and still
@@ -16,16 +15,16 @@
 confusion. Maybe it should be change to copy input sequence to break
 the connection completely ? )
 
->>> nnn = range(3)
+>>> nnn = list(range(3))
 >>> rnn = Rev(nnn)
->>> for n in rnn: print n
+>>> for n in rnn: n
 ...
 2
 1
 0
 >>> for n in range(4, 6): nnn.append(n)   # update nnn
 ...
->>> for n in rnn: print n     # prints reversed updated values
+>>> for n in rnn: n     # prints reversed updated values
 ...
 5
 4
@@ -35,7 +34,7 @@
 >>> nnn = nnn[1:-1]
 >>> nnn
 [1, 2, 4]
->>> for n in rnn: print n     # prints reversed values of old nnn
+>>> for n in rnn: n     # prints reversed values of old nnn
 ...
 5
 4
@@ -45,16 +44,15 @@
 
 #
 >>> WH = Rev('Hello World!')
->>> print WH.forw, WH.back
+>>> print(WH.forw, WH.back)
 Hello World! !dlroW olleH
->>> nnn = Rev(range(1, 10))
->>> print nnn.forw
+>>> nnn = Rev(list(range(1, 10)))
+>>> print(nnn.forw)
 [1, 2, 3, 4, 5, 6, 7, 8, 9]
->>> print nnn.back
+>>> print(nnn.back)
 [9, 8, 7, 6, 5, 4, 3, 2, 1]
 
->>> rrr = Rev(nnn)
->>> rrr
+>>> Rev(nnn)
 <1, 2, 3, 4, 5, 6, 7, 8, 9>
 
 '''

Modified: python/branches/py3k-dtoa/Demo/classes/Vec.py
==============================================================================
--- python/branches/py3k-dtoa/Demo/classes/Vec.py	(original)
+++ python/branches/py3k-dtoa/Demo/classes/Vec.py	Tue Jul  6 17:55:27 2010
@@ -1,23 +1,41 @@
-# A simple vector class
+class Vec:
+    """ A simple vector class
 
+    Instances of the Vec class  can be constructed from numbers
 
-def vec(*v):
-    return Vec(*v)
+    >>> a = Vec(1, 2, 3)
+    >>> b = Vec(3, 2, 1)
 
+    added
+    >>> a + b
+    Vec(4, 4, 4)
 
-class Vec:
+    subtracted
+    >>> a - b
+    Vec(-2, 0, 2)
+
+    and multiplied by a scalar on the left
+    >>> 3.0 * a
+    Vec(3.0, 6.0, 9.0)
 
+    or on the right
+    >>> a * 3.0
+    Vec(3.0, 6.0, 9.0)
+    """
     def __init__(self, *v):
         self.v = list(v)
 
-    def fromlist(self, v):
+    @classmethod
+    def fromlist(cls, v):
         if not isinstance(v, list):
             raise TypeError
-        self.v = v[:]
-        return self
+        inst = cls()
+        inst.v = v
+        return inst
 
     def __repr__(self):
-        return 'vec(' + repr(self.v)[1:-1] + ')'
+        args = ', '.join(repr(x) for x in self.v)
+        return 'Vec({})'.format(args)
 
     def __len__(self):
         return len(self.v)
@@ -27,28 +45,24 @@
 
     def __add__(self, other):
         # Element-wise addition
-        v = list(map(lambda x, y: x+y, self, other))
-        return Vec().fromlist(v)
+        v = [x + y for x, y in zip(self.v, other.v)]
+        return Vec.fromlist(v)
 
     def __sub__(self, other):
         # Element-wise subtraction
-        v = list(map(lambda x, y: x-y, self, other))
-        return Vec().fromlist(v)
+        v = [x - y for x, y in zip(self.v, other.v)]
+        return Vec.fromlist(v)
 
     def __mul__(self, scalar):
         # Multiply by scalar
-        v = [x*scalar for x in self.v]
-        return Vec().fromlist(v)
+        v = [x * scalar for x in self.v]
+        return Vec.fromlist(v)
 
+    __rmul__ = __mul__
 
 
 def test():
-    a = vec(1, 2, 3)
-    b = vec(3, 2, 1)
-    print(a)
-    print(b)
-    print(a+b)
-    print(a-b)
-    print(a*3.0)
+    import doctest
+    doctest.testmod()
 
 test()

Modified: python/branches/py3k-dtoa/Demo/embed/Makefile
==============================================================================
--- python/branches/py3k-dtoa/Demo/embed/Makefile	(original)
+++ python/branches/py3k-dtoa/Demo/embed/Makefile	Tue Jul  6 17:55:27 2010
@@ -1,5 +1,5 @@
 # Makefile for embedded Python use demo.
-# (This version tailored for my Red Hat Linux 6.1 setup;
+# (This version originally written on Red Hat Linux 6.1;
 # edit lines marked with XXX.)
 
 # XXX The compiler you are using
@@ -10,7 +10,7 @@
 srcdir=		../..
 
 # Python version
-VERSION=	3.0
+VERSION=	3.2
 
 # Compiler flags
 OPT=		-g
@@ -21,7 +21,7 @@
 # The Python library
 LIBPYTHON=	$(blddir)/libpython$(VERSION).a
 
-# XXX edit LIBS (in particular) to match $(blddir)/Modules/Makefile
+# XXX edit LIBS (in particular) to match $(blddir)/Makefile
 LIBS=		-lnsl -ldl -lreadline -ltermcap -lieee -lpthread -lutil
 LDFLAGS=	-Xlinker -export-dynamic
 SYSLIBS=	-lm

Modified: python/branches/py3k-dtoa/Demo/embed/demo.c
==============================================================================
--- python/branches/py3k-dtoa/Demo/embed/demo.c	(original)
+++ python/branches/py3k-dtoa/Demo/embed/demo.c	Tue Jul  6 17:55:27 2010
@@ -20,10 +20,19 @@
     Py_Initialize();
 
     /* Define sys.argv.  It is up to the application if you
-       want this; you can also let it undefined (since the Python
+       want this; you can also leave it undefined (since the Python
        code is generally not a main program it has no business
-       touching sys.argv...) */
-    PySys_SetArgv(2, args);
+       touching sys.argv...) 
+
+       If the third argument is true, sys.path is modified to include
+       either the directory containing the script named by argv[0], or
+       the current working directory.  This can be risky; if you run
+       an application embedding Python in a directory controlled by
+       someone else, attackers could put a Trojan-horse module in the
+       directory (say, a file named os.py) that your application would
+       then import and run.
+    */
+    PySys_SetArgvEx(argc, argv, 0);
 
     /* Do some application specific code */
     printf("Hello, brave new world\n\n");

Modified: python/branches/py3k-dtoa/Demo/md5test/md5driver.py
==============================================================================
--- python/branches/py3k-dtoa/Demo/md5test/md5driver.py	(original)
+++ python/branches/py3k-dtoa/Demo/md5test/md5driver.py	Tue Jul  6 17:55:27 2010
@@ -1,11 +1,10 @@
+from hashlib import md5
 import string
-import md5
 from sys import argv
 
 def MDPrint(str):
     outstr = ''
-    for i in str:
-        o = ord(i)
+    for o in str:
         outstr = (outstr
                   + string.hexdigits[(o >> 4) & 0xF]
                   + string.hexdigits[o & 0xF])
@@ -42,7 +41,7 @@
     print('MD5 time trial. Processing', TEST_BYTES, 'characters...')
     t1 = time()
 
-    mdContext = md5.new()
+    mdContext = md5()
 
     for i in range(TEST_BLOCKS):
         mdContext.update(data)
@@ -57,13 +56,13 @@
 
 
 def MDString(str):
-    MDPrint(md5.new(str).digest())
+    MDPrint(md5(str.encode("utf-8")).digest())
     print('"' + str + '"')
 
 
 def MDFile(filename):
     f = open(filename, 'rb')
-    mdContext = md5.new()
+    mdContext = md5()
 
     while 1:
         data = f.read(1024)
@@ -78,7 +77,7 @@
 import sys
 
 def MDFilter():
-    mdContext = md5.new()
+    mdContext = md5()
 
     while 1:
         data = sys.stdin.read(16)

Modified: python/branches/py3k-dtoa/Demo/parser/test_parser.py
==============================================================================
--- python/branches/py3k-dtoa/Demo/parser/test_parser.py	(original)
+++ python/branches/py3k-dtoa/Demo/parser/test_parser.py	Tue Jul  6 17:55:27 2010
@@ -11,19 +11,19 @@
     global _numFailed
     print('----', fileName, end=' ')
     try:
-        ast = parser.suite(t)
-        tup = parser.ast2tuple(ast)
-        # this discards the first AST; a huge memory savings when running
+        st = parser.suite(t)
+        tup = parser.st2tuple(st)
+        # this discards the first ST; a huge memory savings when running
         # against a large source file like Tkinter.py.
-        ast = None
-        new = parser.tuple2ast(tup)
+        st = None
+        new = parser.tuple2st(tup)
     except parser.ParserError as err:
         print()
         print('parser module raised exception on input file', fileName + ':')
         traceback.print_exc()
         _numFailed = _numFailed + 1
     else:
-        if tup != parser.ast2tuple(new):
+        if tup != parser.st2tuple(new):
             print()
             print('parser module failed on input file', fileName)
             _numFailed = _numFailed + 1

Modified: python/branches/py3k-dtoa/Demo/parser/unparse.py
==============================================================================
--- python/branches/py3k-dtoa/Demo/parser/unparse.py	(original)
+++ python/branches/py3k-dtoa/Demo/parser/unparse.py	Tue Jul  6 17:55:27 2010
@@ -1,9 +1,15 @@
 "Usage: unparse.py <path to source file>"
 import sys
-import _ast
+import math
+import ast
+import tokenize
 import io
 import os
 
+# Large float and imaginary literals get turned into infinities in the AST.
+# We unparse those infinities to INFSTR.
+INFSTR = "1e" + repr(sys.float_info.max_10_exp + 1)
+
 def interleave(inter, f, seq):
     """Call f on each item in seq, calling inter() in between.
     """
@@ -20,7 +26,7 @@
 class Unparser:
     """Methods in this class recursively traverse an AST and
     output source code for the abstract syntax; original formatting
-    is disregarged. """
+    is disregarded. """
 
     def __init__(self, tree, file = sys.stdout):
         """Unparser(tree, file=sys.stdout) -> None.
@@ -80,10 +86,11 @@
 
     def _ImportFrom(self, t):
         self.fill("from ")
-        self.write(t.module)
+        self.write("." * t.level)
+        if t.module:
+            self.write(t.module)
         self.write(" import ")
         interleave(lambda: self.write(", "), self.dispatch, t.names)
-        # XXX(jpe) what is level for?
 
     def _Assign(self, t):
         self.fill()
@@ -115,7 +122,7 @@
 
     def _Delete(self, t):
         self.fill("del ")
-        self.dispatch(t.targets)
+        interleave(lambda: self.write(", "), self.dispatch, t.targets)
 
     def _Assert(self, t):
         self.fill("assert ")
@@ -124,24 +131,14 @@
             self.write(", ")
             self.dispatch(t.msg)
 
-    def _Print(self, t):
-        self.fill("print ")
-        do_comma = False
-        if t.dest:
-            self.write(">>")
-            self.dispatch(t.dest)
-            do_comma = True
-        for e in t.values:
-            if do_comma:self.write(", ")
-            else:do_comma=True
-            self.dispatch(e)
-        if not t.nl:
-            self.write(",")
-
     def _Global(self, t):
         self.fill("global ")
         interleave(lambda: self.write(", "), self.write, t.names)
 
+    def _Nonlocal(self, t):
+        self.fill("nonlocal ")
+        interleave(lambda: self.write(", "), self.write, t.names)
+
     def _Yield(self, t):
         self.write("(")
         self.write("yield")
@@ -151,15 +148,15 @@
         self.write(")")
 
     def _Raise(self, t):
-        self.fill('raise ')
-        if t.type:
-            self.dispatch(t.type)
-        if t.inst:
-            self.write(", ")
-            self.dispatch(t.inst)
-        if t.tback:
-            self.write(", ")
-            self.dispatch(t.tback)
+        self.fill("raise")
+        if not t.exc:
+            assert not t.cause
+            return
+        self.write(" ")
+        self.dispatch(t.exc)
+        if t.cause:
+            self.write(" from ")
+            self.dispatch(t.cause)
 
     def _TryExcept(self, t):
         self.fill("try")
@@ -176,10 +173,14 @@
             self.leave()
 
     def _TryFinally(self, t):
-        self.fill("try")
-        self.enter()
-        self.dispatch(t.body)
-        self.leave()
+        if len(t.body) == 1 and isinstance(t.body[0], ast.TryExcept):
+            # try-except-finally
+            self.dispatch(t.body)
+        else:
+            self.fill("try")
+            self.enter()
+            self.dispatch(t.body)
+            self.leave()
 
         self.fill("finally")
         self.enter()
@@ -192,21 +193,40 @@
             self.write(" ")
             self.dispatch(t.type)
         if t.name:
-            self.write(", ")
-            self.dispatch(t.name)
+            self.write(" as ")
+            self.write(t.name)
         self.enter()
         self.dispatch(t.body)
         self.leave()
 
     def _ClassDef(self, t):
         self.write("\n")
+        for deco in t.decorator_list:
+            self.fill("@")
+            self.dispatch(deco)
         self.fill("class "+t.name)
-        if t.bases:
-            self.write("(")
-            for a in t.bases:
-                self.dispatch(a)
-                self.write(", ")
-            self.write(")")
+        self.write("(")
+        comma = False
+        for e in t.bases:
+            if comma: self.write(", ")
+            else: comma = True
+            self.dispatch(e)
+        for e in t.keywords:
+            if comma: self.write(", ")
+            else: comma = True
+            self.dispatch(e)
+        if t.starargs:
+            if comma: self.write(", ")
+            else: comma = True
+            self.write("*")
+            self.dispatch(t.starargs)
+        if t.kwargs:
+            if comma: self.write(", ")
+            else: comma = True
+            self.write("**")
+            self.dispatch(t.kwargs)
+        self.write(")")
+
         self.enter()
         self.dispatch(t.body)
         self.leave()
@@ -219,6 +239,9 @@
         self.fill("def "+t.name + "(")
         self.dispatch(t.args)
         self.write(")")
+        if t.returns:
+            self.write(" -> ")
+            self.dispatch(t.returns)
         self.enter()
         self.dispatch(t.body)
         self.leave()
@@ -235,15 +258,24 @@
             self.fill("else")
             self.enter()
             self.dispatch(t.orelse)
-            self.leave
+            self.leave()
 
     def _If(self, t):
         self.fill("if ")
         self.dispatch(t.test)
         self.enter()
-        # XXX elif?
         self.dispatch(t.body)
         self.leave()
+        # collapse nested ifs into equivalent elifs.
+        while (t.orelse and len(t.orelse) == 1 and
+               isinstance(t.orelse[0], ast.If)):
+            t = t.orelse[0]
+            self.fill("elif ")
+            self.dispatch(t.test)
+            self.enter()
+            self.dispatch(t.body)
+            self.leave()
+        # final else
         if t.orelse:
             self.fill("else")
             self.enter()
@@ -260,7 +292,7 @@
             self.fill("else")
             self.enter()
             self.dispatch(t.orelse)
-            self.leave
+            self.leave()
 
     def _With(self, t):
         self.fill("with ")
@@ -273,19 +305,18 @@
         self.leave()
 
     # expr
+    def _Bytes(self, t):
+        self.write(repr(t.s))
+
     def _Str(self, tree):
         self.write(repr(tree.s))
 
     def _Name(self, t):
         self.write(t.id)
 
-    def _Repr(self, t):
-        self.write("`")
-        self.dispatch(t.value)
-        self.write("`")
-
     def _Num(self, t):
-        self.write(repr(t.n))
+        # Substitute overflowing decimal literal for AST infinities.
+        self.write(repr(t.n).replace("inf", INFSTR))
 
     def _List(self, t):
         self.write("[")
@@ -306,6 +337,22 @@
             self.dispatch(gen)
         self.write(")")
 
+    def _SetComp(self, t):
+        self.write("{")
+        self.dispatch(t.elt)
+        for gen in t.generators:
+            self.dispatch(gen)
+        self.write("}")
+
+    def _DictComp(self, t):
+        self.write("{")
+        self.dispatch(t.key)
+        self.write(": ")
+        self.dispatch(t.value)
+        for gen in t.generators:
+            self.dispatch(gen)
+        self.write("}")
+
     def _comprehension(self, t):
         self.write(" for ")
         self.dispatch(t.target)
@@ -324,14 +371,20 @@
         self.dispatch(t.orelse)
         self.write(")")
 
+    def _Set(self, t):
+        assert(t.elts) # should be at least one element
+        self.write("{")
+        interleave(lambda: self.write(", "), self.dispatch, t.elts)
+        self.write("}")
+
     def _Dict(self, t):
         self.write("{")
-        def writem(xxx_todo_changeme):
-            (k, v) = xxx_todo_changeme
+        def write_pair(pair):
+            (k, v) = pair
             self.dispatch(k)
             self.write(": ")
             self.dispatch(v)
-        interleave(lambda: self.write(", "), writem, zip(t.keys, t.values))
+        interleave(lambda: self.write(", "), write_pair, zip(t.keys, t.values))
         self.write("}")
 
     def _Tuple(self, t):
@@ -346,13 +399,14 @@
 
     unop = {"Invert":"~", "Not": "not", "UAdd":"+", "USub":"-"}
     def _UnaryOp(self, t):
-        self.write(self.unop[t.op.__class__.__name__])
         self.write("(")
+        self.write(self.unop[t.op.__class__.__name__])
+        self.write(" ")
         self.dispatch(t.operand)
         self.write(")")
 
     binop = { "Add":"+", "Sub":"-", "Mult":"*", "Div":"/", "Mod":"%",
-                    "LShift":">>", "RShift":"<<", "BitOr":"|", "BitXor":"^", "BitAnd":"&",
+                    "LShift":"<<", "RShift":">>", "BitOr":"|", "BitXor":"^", "BitAnd":"&",
                     "FloorDiv":"//", "Pow": "**"}
     def _BinOp(self, t):
         self.write("(")
@@ -369,9 +423,9 @@
         for o, e in zip(t.ops, t.comparators):
             self.write(" " + self.cmpops[o.__class__.__name__] + " ")
             self.dispatch(e)
-            self.write(")")
+        self.write(")")
 
-    boolops = {_ast.And: 'and', _ast.Or: 'or'}
+    boolops = {ast.And: 'and', ast.Or: 'or'}
     def _BoolOp(self, t):
         self.write("(")
         s = " %s " % self.boolops[t.op.__class__]
@@ -380,6 +434,11 @@
 
     def _Attribute(self,t):
         self.dispatch(t.value)
+        # Special case: 3.__abs__() is a syntax error, so if t.value
+        # is an integer literal then we need to either parenthesize
+        # it or add an extra space to get 3 .__abs__().
+        if isinstance(t.value, ast.Num) and isinstance(t.value.n, int):
+            self.write(" ")
         self.write(".")
         self.write(t.attr)
 
@@ -433,28 +492,55 @@
     def _ExtSlice(self, t):
         interleave(lambda: self.write(', '), self.dispatch, t.dims)
 
+    # argument
+    def _arg(self, t):
+        self.write(t.arg)
+        if t.annotation:
+            self.write(": ")
+            self.dispatch(t.annotation)
+
     # others
     def _arguments(self, t):
         first = True
-        nonDef = len(t.args)-len(t.defaults)
-        for a in t.args[0:nonDef]:
+        # normal arguments
+        defaults = [None] * (len(t.args) - len(t.defaults)) + t.defaults
+        for a, d in zip(t.args, defaults):
             if first:first = False
             else: self.write(", ")
             self.dispatch(a)
-        for a,d in zip(t.args[nonDef:], t.defaults):
-            if first:first = False
-            else: self.write(", ")
-            self.dispatch(a),
-            self.write("=")
-            self.dispatch(d)
-        if t.vararg:
+            if d:
+                self.write("=")
+                self.dispatch(d)
+
+        # varargs, or bare '*' if no varargs but keyword-only arguments present
+        if t.vararg or t.kwonlyargs:
             if first:first = False
             else: self.write(", ")
-            self.write("*"+t.vararg)
+            self.write("*")
+            if t.vararg:
+                self.write(t.vararg)
+                if t.varargannotation:
+                    self.write(": ")
+                    self.dispatch(t.varargannotation)
+
+        # keyword-only arguments
+        if t.kwonlyargs:
+            for a, d in zip(t.kwonlyargs, t.kw_defaults):
+                if first:first = False
+                else: self.write(", ")
+                self.dispatch(a),
+                if d:
+                    self.write("=")
+                    self.dispatch(d)
+
+        # kwargs
         if t.kwarg:
             if first:first = False
             else: self.write(", ")
             self.write("**"+t.kwarg)
+            if t.kwargannotation:
+                self.write(": ")
+                self.dispatch(t.kwargannotation)
 
     def _keyword(self, t):
         self.write(t.arg)
@@ -462,10 +548,12 @@
         self.dispatch(t.value)
 
     def _Lambda(self, t):
+        self.write("(")
         self.write("lambda ")
         self.dispatch(t.args)
         self.write(": ")
         self.dispatch(t.body)
+        self.write(")")
 
     def _alias(self, t):
         self.write(t.name)
@@ -473,8 +561,11 @@
             self.write(" as "+t.asname)
 
 def roundtrip(filename, output=sys.stdout):
-    source = open(filename).read()
-    tree = compile(source, filename, "exec", _ast.PyCF_ONLY_AST)
+    with open(filename, "rb") as pyfile:
+        encoding = tokenize.detect_encoding(pyfile.readline)[0]
+    with open(filename, "r", encoding=encoding) as pyfile:
+        source = pyfile.read()
+    tree = compile(source, filename, "exec", ast.PyCF_ONLY_AST)
     Unparser(tree, output)
 
 

Modified: python/branches/py3k-dtoa/Doc/README.txt
==============================================================================
--- python/branches/py3k-dtoa/Doc/README.txt	(original)
+++ python/branches/py3k-dtoa/Doc/README.txt	Tue Jul  6 17:55:27 2010
@@ -14,12 +14,11 @@
 Building the docs
 =================
 
-You need to install Python 2.4 or higher (but Python 3.0 is not supported yet);
-the toolset used to build the docs are written in Python.  The toolset used
-to build the documentation is called *Sphinx*, it is not included in this
-tree, but maintained separately in the Python Subversion repository.  Also
-needed are Jinja, a templating engine (included in Sphinx as a Subversion
-external), and optionally Pygments, a code highlighter.
+You need to have Python 2.4 or higher installed; the toolset used to build the
+docs is written in Python.  It is called *Sphinx*, it is not included in this
+tree, but maintained separately.  Also needed are the docutils, supplying the
+base markup that Sphinx uses, Jinja, a templating engine, and optionally
+Pygments, a code highlighter.
 
 
 Using make
@@ -47,29 +46,29 @@
    convert them into a single Compiled HTML (.chm) file -- these are popular
    under Microsoft Windows, but very handy on every platform.
 
-   To create the CHM file, you need to run the Microsoft HTML Help Workshop
-   over the generated project (.hhp) file.
+   To create the CHM file, you need to run the Microsoft HTML Help Workshop over
+   the generated project (.hhp) file.
 
- * "latex", which builds LaTeX source files that can be run with "pdflatex"
-   to produce PDF documents.
+ * "latex", which builds LaTeX source files as input to "pdflatex" to produce
+   PDF documents.
 
  * "text", which builds a plain text file for each source file.
 
  * "linkcheck", which checks all external references to see whether they are
-   broken, redirected or malformed, and outputs this information to stdout
-   as well as a plain-text (.txt) file.
+   broken, redirected or malformed, and outputs this information to stdout as
+   well as a plain-text (.txt) file.
 
  * "changes", which builds an overview over all versionadded/versionchanged/
    deprecated items in the current version. This is meant as a help for the
    writer of the "What's New" document.
 
- * "coverage", which builds a coverage overview for standard library modules
-   and C API.
+ * "coverage", which builds a coverage overview for standard library modules and
+   C API.
 
- * "pydoc-topics", which builds a Python module containing a dictionary
-   with plain text documentation for the labels defined in
-   `tools/sphinxext/pyspecific.py` -- pydoc needs these to show topic
-   and keyword help.
+ * "pydoc-topics", which builds a Python module containing a dictionary with
+   plain text documentation for the labels defined in
+   `tools/sphinxext/pyspecific.py` -- pydoc needs these to show topic and
+   keyword help.
 
 A "make update" updates the Subversion checkouts in `tools/`.
 
@@ -95,7 +94,7 @@
 
 or by installing it from PyPI.
 
-You can optionally also install Pygments, either as a checkout via :: 
+You can optionally also install Pygments, either as a checkout via ::
 
    svn co http://svn.python.org/projects/external/Pygments-1.3.1/pygments tools/pygments
 

Modified: python/branches/py3k-dtoa/Doc/c-api/arg.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/c-api/arg.rst	(original)
+++ python/branches/py3k-dtoa/Doc/c-api/arg.rst	Tue Jul  6 17:55:27 2010
@@ -36,7 +36,7 @@
 
 However, when a :ctype:`Py_buffer` structure gets filled, the underlying
 buffer is locked so that the caller can subsequently use the buffer even
-inside a ``Py_BEGIN_ALLOW_THREADS`` block without the risk of mutable data
+inside a :ctype:`Py_BEGIN_ALLOW_THREADS` block without the risk of mutable data
 being resized or destroyed.  As a result, **you have to call**
 :cfunc:`PyBuffer_Release` after you have finished processing the data (or
 in any early abort case).
@@ -48,9 +48,9 @@
    the length argument (int or :ctype:`Py_ssize_t`) is controlled by
    defining the macro :cmacro:`PY_SSIZE_T_CLEAN` before including
    :file:`Python.h`.  If the macro was defined, length is a
-   :ctype:`Py_ssize_t` rather than an int.  This behavior will change
+   :ctype:`Py_ssize_t` rather than an :ctype:`int`. This behavior will change
    in a future Python version to only support :ctype:`Py_ssize_t` and
-   drop int support.  It is best to always define :cmacro:`PY_SSIZE_T_CLEAN`.
+   drop :ctype:`int` support. It is best to always define :cmacro:`PY_SSIZE_T_CLEAN`.
 
 
 ``s`` (:class:`str`) [const char \*]
@@ -100,7 +100,7 @@
    contain embedded NUL bytes; if it does, a :exc:`TypeError`
    exception is raised.
 
-``y*`` (:class:`bytes`, :class:`bytearray` or buffer compatible object) [Py_buffer \*]
+``y*`` (:class:`bytes`, :class:`bytearray` or buffer compatible object) [Py_buffer]
    This variant on ``s*`` doesn't accept Unicode objects, only objects
    supporting the buffer protocol.  **This is the recommended way to accept
    binary data.**
@@ -150,21 +150,11 @@
    any conversion.  Raises :exc:`TypeError` if the object is not a Unicode
    object.  The C variable may also be declared as :ctype:`PyObject\*`.
 
-``w`` (:class:`bytearray` or read-write character buffer) [char \*]
-   Similar to ``y``, but accepts any object which implements the read-write buffer
-   interface.  The caller must determine the length of the buffer by other means,
-   or use ``w#`` instead.  Only single-segment buffer objects are accepted;
-   :exc:`TypeError` is raised for all others.
-
 ``w*`` (:class:`bytearray` or read-write byte-oriented buffer) [Py_buffer]
-   This is to ``w`` what ``y*`` is to ``y``.
-
-``w#`` (:class:`bytearray` or read-write character buffer) [char \*, int]
-   Like ``y#``, but accepts any object which implements the read-write buffer
-   interface.  The :ctype:`char \*` variable is set to point to the first byte
-   of the buffer, and the :ctype:`int` is set to the length of the buffer.
-   Only single-segment buffer objects are accepted; :exc:`TypeError` is raised
-   for all others.
+   This format accepts any object which implements the read-write buffer
+   interface. It fills a :ctype:`Py_buffer` structure provided by the caller.
+   The buffer may contain embedded null bytes. The caller have to call
+   :cfunc:`PyBuffer_Release` when it is done with the buffer.
 
 ``es`` (:class:`str`) [const char \*encoding, char \*\*buffer]
    This variant on ``s`` is used for encoding Unicode into a character buffer.
@@ -492,11 +482,11 @@
    strings a tad more readable.
 
    ``s`` (:class:`str` or ``None``) [char \*]
-      Convert a null-terminated C string to a Python object using ``'utf-8'``
+      Convert a null-terminated C string to a Python :class:`str` object using ``'utf-8'``
       encoding. If the C string pointer is *NULL*, ``None`` is used.
 
    ``s#`` (:class:`str` or ``None``) [char \*, int]
-      Convert a C string and its length to a Python object using ``'utf-8'``
+      Convert a C string and its length to a Python :class:`str` object using ``'utf-8'``
       encoding. If the C string pointer is *NULL*, the length is ignored and
       ``None`` is returned.
 

Modified: python/branches/py3k-dtoa/Doc/c-api/datetime.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/c-api/datetime.rst	(original)
+++ python/branches/py3k-dtoa/Doc/c-api/datetime.rst	Tue Jul  6 17:55:27 2010
@@ -8,9 +8,10 @@
 Various date and time objects are supplied by the :mod:`datetime` module.
 Before using any of these functions, the header file :file:`datetime.h` must be
 included in your source (note that this is not included by :file:`Python.h`),
-and the macro :cfunc:`PyDateTime_IMPORT` must be invoked.  The macro puts a
-pointer to a C structure into a static variable,  ``PyDateTimeAPI``, that is
-used by the following macros.
+and the macro :cmacro:`PyDateTime_IMPORT` must be invoked, usually as part of
+the module initialisation function.  The macro puts a pointer to a C structure
+into a static variable, :cdata:`PyDateTimeAPI`, that is used by the following
+macros.
 
 Type-check macros:
 

Modified: python/branches/py3k-dtoa/Doc/c-api/init.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/c-api/init.rst	(original)
+++ python/branches/py3k-dtoa/Doc/c-api/init.rst	Tue Jul  6 17:55:27 2010
@@ -385,7 +385,7 @@
 
 .. cfunction:: void PySys_SetArgv(int argc, wchar_t **argv)
 
-   This function works like :cfunc:`PySys_SetArgv` with *updatepath* set to 1.
+   This function works like :cfunc:`PySys_SetArgvEx` with *updatepath* set to 1.
 
 
 .. cfunction:: void Py_SetPythonHome(wchar_t *home)

Modified: python/branches/py3k-dtoa/Doc/c-api/intro.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/c-api/intro.rst	(original)
+++ python/branches/py3k-dtoa/Doc/c-api/intro.rst	Tue Jul  6 17:55:27 2010
@@ -519,12 +519,12 @@
 :mod:`builtins`, :mod:`__main__`, :mod:`sys`, and :mod:`exceptions`.  It also
 initializes the module search path (``sys.path``).
 
-.. index:: single: PySys_SetArgv()
+.. index:: single: PySys_SetArgvEx()
 
 :cfunc:`Py_Initialize` does not set the "script argument list"  (``sys.argv``).
-If this variable is needed by Python code that  will be executed later, it must
-be set explicitly with a call to  ``PySys_SetArgv(argc, argv)`` subsequent to
-the call to :cfunc:`Py_Initialize`.
+If this variable is needed by Python code that will be executed later, it must
+be set explicitly with a call to  ``PySys_SetArgvEx(argc, argv, updatepath)``
+after the call to :cfunc:`Py_Initialize`.
 
 On most systems (in particular, on Unix and Windows, although the details are
 slightly different), :cfunc:`Py_Initialize` calculates the module search path

Modified: python/branches/py3k-dtoa/Doc/c-api/number.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/c-api/number.rst	(original)
+++ python/branches/py3k-dtoa/Doc/c-api/number.rst	Tue Jul  6 17:55:27 2010
@@ -215,18 +215,6 @@
    the Python statement ``o1 |= o2``.
 
 
-.. cfunction:: PyObject* PyNumber_Int(PyObject *o)
-
-   Returns the *o* converted to an integer object on success, or *NULL* on
-   failure.  This is the equivalent of the Python expression ``int(o)``.
-
-   .. note::
-
-     This function is defined in the transitional :file:`intobject.h`
-     header file.  It will be removed completely in Python 3.1.  Use
-     the :cfunc:`PyNumber_Long` function instead.
-
-
 .. cfunction:: PyObject* PyNumber_Long(PyObject *o)
 
    .. index:: builtin: int

Modified: python/branches/py3k-dtoa/Doc/c-api/unicode.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/c-api/unicode.rst	(original)
+++ python/branches/py3k-dtoa/Doc/c-api/unicode.rst	Tue Jul  6 17:55:27 2010
@@ -346,10 +346,10 @@
    Coerce an encoded object *obj* to an Unicode object and return a reference with
    incremented refcount.
 
-   String and other char buffer compatible objects are decoded according to the
-   given encoding and using the error handling defined by errors.  Both can be
-   *NULL* to have the interface use the default values (see the next section for
-   details).
+   :class:`bytes`, :class:`bytearray` and other char buffer compatible objects
+   are decoded according to the given encoding and using the error handling
+   defined by errors. Both can be *NULL* to have the interface use the default
+   values (see the next section for details).
 
    All other objects, including Unicode objects, cause a :exc:`TypeError` to be
    set.

Modified: python/branches/py3k-dtoa/Doc/data/refcounts.dat
==============================================================================
--- python/branches/py3k-dtoa/Doc/data/refcounts.dat	(original)
+++ python/branches/py3k-dtoa/Doc/data/refcounts.dat	Tue Jul  6 17:55:27 2010
@@ -846,9 +846,6 @@
 PyNumber_InPlaceXor:PyObject*:v:0:
 PyNumber_InPlaceXor:PyObject*:w:0:
 
-PyNumber_Int:PyObject*::+1:
-PyNumber_Int:PyObject*:o:0:
-
 PyNumber_Invert:PyObject*::+1:
 PyNumber_Invert:PyObject*:o:0:
 

Modified: python/branches/py3k-dtoa/Doc/distutils/builtdist.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/distutils/builtdist.rst	(original)
+++ python/branches/py3k-dtoa/Doc/distutils/builtdist.rst	Tue Jul  6 17:55:27 2010
@@ -322,7 +322,7 @@
 option.
 
 By default the installer will display the cool "Python Powered" logo when it is
-run, but you can also supply your own 152x161 bitmap which must be a Windows
+run, but you can also supply your own 152x261 bitmap which must be a Windows
 :file:`.bmp` file with the :option:`--bitmap` option.
 
 The installer will also display a large title on the desktop background window
@@ -375,7 +375,7 @@
 The Postinstallation script
 ---------------------------
 
-Starting with Python 2.3, a postinstallation script can be specified which the
+Starting with Python 2.3, a postinstallation script can be specified with the
 :option:`--install-script` option.  The basename of the script must be
 specified, and the script filename must also be listed in the scripts argument
 to the setup function.

Modified: python/branches/py3k-dtoa/Doc/documenting/index.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/documenting/index.rst	(original)
+++ python/branches/py3k-dtoa/Doc/documenting/index.rst	Tue Jul  6 17:55:27 2010
@@ -10,9 +10,9 @@
 `reStructuredText`_, developed by the `docutils`_ project, amended by custom
 directives and using a toolset named `Sphinx`_ to postprocess the HTML output.
 
-This document describes the style guide for our documentation, the custom
-reStructuredText markup introduced to support Python documentation and how it
-should be used, as well as the Sphinx build system.
+This document describes the style guide for our documentation as well as the
+custom reStructuredText markup introduced by Sphinx to support Python
+documentation and how it should be used.
 
 .. _reStructuredText: http://docutils.sf.net/rst.html
 .. _docutils: http://docutils.sf.net/
@@ -35,3 +35,4 @@
    rest.rst
    markup.rst
    fromlatex.rst
+   building.rst

Modified: python/branches/py3k-dtoa/Doc/documenting/markup.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/documenting/markup.rst	(original)
+++ python/branches/py3k-dtoa/Doc/documenting/markup.rst	Tue Jul  6 17:55:27 2010
@@ -698,10 +698,10 @@
       .. toctree::
          :maxdepth: 2
 
-         intro.rst
-         strings.rst
-         datatypes.rst
-         numeric.rst
+         intro
+         strings
+         datatypes
+         numeric
          (many more files listed here)
 
    This accomplishes two things:
@@ -709,8 +709,8 @@
    * Tables of contents from all those files are inserted, with a maximum depth
      of two, that means one nested heading.  ``toctree`` directives in those
      files are also taken into account.
-   * Sphinx knows that the relative order of the files ``intro.rst``,
-     ``strings.rst`` and so forth, and it knows that they are children of the
+   * Sphinx knows that the relative order of the files ``intro``,
+     ``strings`` and so forth, and it knows that they are children of the
      shown file, the library index.  From this information it generates "next
      chapter", "previous chapter" and "parent chapter" links.
 

Modified: python/branches/py3k-dtoa/Doc/documenting/style.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/documenting/style.rst	(original)
+++ python/branches/py3k-dtoa/Doc/documenting/style.rst	Tue Jul  6 17:55:27 2010
@@ -7,7 +7,7 @@
 wherever possible. This particular style guide was selected mostly because it
 seems reasonable and is easy to get online.
 
-Topics which are not covered in the Apple's style guide will be discussed in
+Topics which are not covered in Apple's style guide will be discussed in
 this document.
 
 All reST files use an indentation of 3 spaces.  The maximum line length is 80

Modified: python/branches/py3k-dtoa/Doc/faq/gui.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/faq/gui.rst	(original)
+++ python/branches/py3k-dtoa/Doc/faq/gui.rst	Tue Jul  6 17:55:27 2010
@@ -13,14 +13,14 @@
 =====================
 
 What platform-independent GUI toolkits exist for Python?
---------------------------------------------------------
+========================================================
 
 Depending on what platform(s) you are aiming at, there are several.
 
 .. XXX check links
 
 Tkinter
-'''''''
+-------
 
 Standard builds of Python include an object-oriented interface to the Tcl/Tk
 widget set, called Tkinter.  This is probably the easiest to install and use.
@@ -29,22 +29,27 @@
 Unix platforms.
 
 wxWidgets
-'''''''''
+---------
 
-wxWidgets is a GUI class library written in C++ that's a portable
-interface to various platform-specific libraries, and that has a
-Python interface called `wxPython <http://www.wxpython.org>`__.
-
-wxWidgets preserves the look and feel of the
-underlying graphics toolkit, and has a large set of widgets and
-collection of GDI classes.  See `the wxWidgets page
-<http://www.wxwidgets.org>`_ for more details.
+wxWidgets (http://www.wxwidgets.org) is a free, portable GUI class
+library written in C++ that provides a native look and feel on a
+number of platforms, with Windows, MacOS X, GTK, X11, all listed as
+current stable targets.  Language bindings are available for a number
+of languages including Python, Perl, Ruby, etc.
+
+wxPython (http://www.wxpython.org) is the Python binding for
+wxwidgets.  While it often lags slightly behind the official wxWidgets
+releases, it also offers a number of features via pure Python
+extensions that are not available in other language bindings.  There
+is an active wxPython user and developer community.
+
+Both wxWidgets and wxPython are free, open source, software with
+permissive licences that allow their use in commercial products as
+well as in freeware or shareware.
 
-wxWidgets supports Windows and MacOS; on Unix variants,
-it supports both GTk+ and Motif toolkits.
 
 Qt
-'''
+---
 
 There are bindings available for the Qt toolkit (`PyQt
 <http://www.riverbankcomputing.co.uk/software/pyqt/>`_) and for KDE (`PyKDE <http://www.riverbankcomputing.co.uk/software/pykde/intro>`__).  If
@@ -55,13 +60,13 @@
 <http://www.trolltech.com>`_.
 
 Gtk+
-''''
+----
 
 PyGtk bindings for the `Gtk+ toolkit <http://www.gtk.org>`_ have been
 implemented by James Henstridge; see <http://www.pygtk.org>.
 
 FLTK
-''''
+----
 
 Python bindings for `the FLTK toolkit <http://www.fltk.org>`_, a simple yet
 powerful and mature cross-platform windowing system, are available from `the
@@ -69,7 +74,7 @@
 
 
 FOX
-'''
+----
 
 A wrapper for `the FOX toolkit <http://www.fox-toolkit.org/>`_ called `FXpy
 <http://fxpy.sourceforge.net/>`_ is available.  FOX supports both Unix variants
@@ -77,13 +82,13 @@
 
 
 OpenGL
-''''''
+------
 
 For OpenGL bindings, see `PyOpenGL <http://pyopengl.sourceforge.net>`_.
 
 
 What platform-specific GUI toolkits exist for Python?
------------------------------------------------------
+========================================================
 
 `The Mac port <http://python.org/download/mac>`_ by Jack Jansen has a rich and
 ever-growing set of modules that support the native Mac toolbox calls.  The port

Modified: python/branches/py3k-dtoa/Doc/glossary.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/glossary.rst	(original)
+++ python/branches/py3k-dtoa/Doc/glossary.rst	Tue Jul  6 17:55:27 2010
@@ -57,11 +57,17 @@
 
    bytecode
       Python source code is compiled into bytecode, the internal representation
-      of a Python program in the interpreter.  The bytecode is also cached in
-      ``.pyc`` and ``.pyo`` files so that executing the same file is faster the
-      second time (recompilation from source to bytecode can be avoided).  This
-      "intermediate language" is said to run on a :term:`virtual machine`
-      that executes the machine code corresponding to each bytecode.
+      of a Python program in the CPython interpreter.  The bytecode is also
+      cached in ``.pyc`` and ``.pyo`` files so that executing the same file is
+      faster the second time (recompilation from source to bytecode can be
+      avoided).  This "intermediate language" is said to run on a
+      :term:`virtual machine` that executes the machine code corresponding to
+      each bytecode. Do note that bytecodes are not expected to work between
+      different Python virtual machines, nor to be stable between Python
+      releases.
+
+      A list of bytecode instructions can be found in the documentation for
+      :ref:`the dis module <bytecodes>`.
 
    class
       A template for creating user-defined objects. Class definitions
@@ -424,10 +430,11 @@
    nested scope
       The ability to refer to a variable in an enclosing definition.  For
       instance, a function defined inside another function can refer to
-      variables in the outer function.  Note that nested scopes work only for
-      reference and not for assignment which will always write to the innermost
-      scope.  In contrast, local variables both read and write in the innermost
-      scope.  Likewise, global variables read and write to the global namespace.
+      variables in the outer function.  Note that nested scopes by default work
+      only for reference and not for assignment.  Local variables both read and
+      write in the innermost scope.  Likewise, global variables read and write
+      to the global namespace.  The :keyword:`nonlocal` allows writing to outer
+      scopes.
 
    new-style class
       Old name for the flavor of classes now used for all class objects.  In

Modified: python/branches/py3k-dtoa/Doc/howto/functional.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/howto/functional.rst	(original)
+++ python/branches/py3k-dtoa/Doc/howto/functional.rst	Tue Jul  6 17:55:27 2010
@@ -5,9 +5,6 @@
 :Author: A. M. Kuchling
 :Release: 0.31
 
-(This is a first draft.  Please send comments/error reports/suggestions to
-amk at amk.ca.)
-
 In this document, we'll take a tour of Python's features suitable for
 implementing programs in a functional style.  After an introduction to the
 concepts of functional programming, we'll look at language features such as

Modified: python/branches/py3k-dtoa/Doc/howto/unicode.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/howto/unicode.rst	(original)
+++ python/branches/py3k-dtoa/Doc/howto/unicode.rst	Tue Jul  6 17:55:27 2010
@@ -4,10 +4,12 @@
   Unicode HOWTO
 *****************
 
-:Release: 1.1
+:Release: 1.11
 
-This HOWTO discusses Python's support for Unicode, and explains various problems
-that people commonly encounter when trying to work with Unicode.
+This HOWTO discusses Python 2.x's support for Unicode, and explains
+various problems that people commonly encounter when trying to work
+with Unicode.  (This HOWTO has not yet been updated to cover the 3.x
+versions of Python.)
 
 
 Introduction to Unicode
@@ -146,8 +148,9 @@
 4. Many Internet standards are defined in terms of textual data, and can't
    handle content with embedded zero bytes.
 
-Generally people don't use this encoding, instead choosing other encodings that
-are more efficient and convenient.
+Generally people don't use this encoding, instead choosing other
+encodings that are more efficient and convenient.  UTF-8 is probably
+the most commonly supported encoding; it will be discussed below.
 
 Encodings don't have to handle every possible Unicode character, and most
 encodings don't.  The rules for converting a Unicode string into the ASCII
@@ -223,8 +226,8 @@
 <http://en.wikipedia.org/wiki/UTF-8>, for example.
 
 
-Python's Unicode Support
-========================
+Python 2.x's Unicode Support
+============================
 
 Now that you've learned the rudiments of Unicode, we can look at Python's
 Unicode features.
@@ -266,8 +269,8 @@
     >>> b'\x80abc'.decode("utf-8", "ignore")
     'abc'
 
-Encodings are specified as strings containing the encoding's name.  Python comes
-with roughly 100 different encodings; see the Python Library Reference at
+Encodings are specified as strings containing the encoding's name.  Python 3.2
+comes with roughly 100 different encodings; see the Python Library Reference at
 :ref:`standard-encodings` for a list.  Some encodings have multiple names; for
 example, 'latin-1', 'iso_8859_1' and '8859' are all synonyms for the same
 encoding.
@@ -626,7 +629,10 @@
 
 Version 1.1: Feb-Nov 2008.  Updates the document with respect to Python 3 changes.
 
+Version 1.11: posted June 20 2010.  Notes that Python 3.x is not covered,
+and that the HOWTO only covers 2.x.
 
+.. comment Describe Python 3.x support (new section? new document?)
 .. comment Additional topic: building Python w/ UCS2 or UCS4 support
 .. comment Describe use of codecs.StreamRecoder and StreamReaderWriter
 

Modified: python/branches/py3k-dtoa/Doc/library/2to3.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/library/2to3.rst	(original)
+++ python/branches/py3k-dtoa/Doc/library/2to3.rst	Tue Jul  6 17:55:27 2010
@@ -276,7 +276,7 @@
 
    Converts the :keyword:`print` statement to the :func:`print` function.
 
-.. 2to3fixer:: raises
+.. 2to3fixer:: raise
 
    Converts ``raise E, V`` to ``raise E(V)``, and ``raise E, V, T`` to ``raise
    E(V).with_traceback(T)``.  If ``E`` is a tuple, the translation will be

Modified: python/branches/py3k-dtoa/Doc/library/calendar.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/library/calendar.rst	(original)
+++ python/branches/py3k-dtoa/Doc/library/calendar.rst	Tue Jul  6 17:55:27 2010
@@ -15,7 +15,7 @@
 are given as integers. For related
 functionality, see also the :mod:`datetime` and :mod:`time` modules.
 
-Most of these functions and classses rely on the :mod:`datetime` module which
+Most of these functions and classes rely on the :mod:`datetime` module which
 uses an idealized calendar, the current Gregorian calendar indefinitely extended
 in both directions.  This matches the definition of the "proleptic Gregorian"
 calendar in Dershowitz and Reingold's book "Calendrical Calculations", where

Modified: python/branches/py3k-dtoa/Doc/library/codecs.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/library/codecs.rst	(original)
+++ python/branches/py3k-dtoa/Doc/library/codecs.rst	Tue Jul  6 17:55:27 2010
@@ -953,6 +953,8 @@
 +-----------------+--------------------------------+--------------------------------+
 | cp857           | 857, IBM857                    | Turkish                        |
 +-----------------+--------------------------------+--------------------------------+
+| cp858           | 858, IBM858                    | Western Europe                 |
++-----------------+--------------------------------+--------------------------------+
 | cp860           | 860, IBM860                    | Portuguese                     |
 +-----------------+--------------------------------+--------------------------------+
 | cp861           | 861, CP-IS, IBM861             | Icelandic                      |
@@ -1223,6 +1225,23 @@
    Convert a label to Unicode, as specified in :rfc:`3490`.
 
 
+:mod:`encodings.mbcs` --- Windows ANSI codepage
+-----------------------------------------------
+
+.. module:: encodings.mbcs
+   :synopsis: Windows ANSI codepage
+
+Encode operand according to the ANSI codepage (CP_ACP). This codec only
+supports ``'strict'`` and ``'replace'`` error handlers to encode, and
+``'strict'`` and ``'ignore'`` error handlers to decode.
+
+Availability: Windows only.
+
+.. versionchanged:: 3.2
+   Before 3.2, the *errors* argument was ignored; ``'replace'`` was always used
+   to encode, and ``'ignore'`` to decode.
+
+
 :mod:`encodings.utf_8_sig` --- UTF-8 codec with BOM signature
 -------------------------------------------------------------
 

Modified: python/branches/py3k-dtoa/Doc/library/contextlib.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/library/contextlib.rst	(original)
+++ python/branches/py3k-dtoa/Doc/library/contextlib.rst	Tue Jul  6 17:55:27 2010
@@ -51,6 +51,11 @@
    the exception has been handled, and execution will resume with the statement
    immediately following the :keyword:`with` statement.
 
+   contextmanager uses :class:`ContextDecorator` so the context managers it
+   creates can be used as decorators as well as in :keyword:`with` statements.
+
+   .. versionchanged:: 3.2
+      Use of :class:`ContextDecorator`.
 
 .. function:: closing(thing)
 
@@ -79,6 +84,58 @@
    ``page.close()`` will be called when the :keyword:`with` block is exited.
 
 
+.. class:: ContextDecorator()
+
+   A base class that enables a context manager to also be used as a decorator.
+
+   Context managers inheriting from ``ContextDecorator`` have to implement
+   ``__enter__`` and ``__exit__`` as normal. ``__exit__`` retains its optional
+   exception handling even when used as a decorator.
+
+   Example::
+
+      from contextlib import ContextDecorator
+
+      class mycontext(ContextDecorator):
+         def __enter__(self):
+            print('Starting')
+            return self
+
+         def __exit__(self, *exc):
+            print('Finishing')
+            return False
+
+      >>> @mycontext()
+      ... def function():
+      ...    print('The bit in the middle')
+      ...
+      >>> function()
+      Starting
+      The bit in the middle
+      Finishing
+
+      >>> with mycontext():
+      ...    print('The bit in the middle')
+      ...
+      Starting
+      The bit in the middle
+      Finishing
+
+   Existing context managers that already have a base class can be extended by
+   using ``ContextDecorator`` as a mixin class::
+
+      from contextlib import ContextDecorator
+
+      class mycontext(ContextBaseClass, ContextDecorator):
+         def __enter__(self):
+            return self
+
+         def __exit__(self, *exc):
+            return False
+
+   .. versionadded:: 3.2
+
+
 .. seealso::
 
    :pep:`0343` - The "with" statement

Modified: python/branches/py3k-dtoa/Doc/library/datetime.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/library/datetime.rst	(original)
+++ python/branches/py3k-dtoa/Doc/library/datetime.rst	Tue Jul  6 17:55:27 2010
@@ -974,10 +974,10 @@
    ``d.dst()`` returns.  DST is never in effect for a UTC time.
 
    If *d* is aware, *d* is normalized to UTC time, by subtracting
-   ``d.utcoffset()``, and a :class:`time.struct_time` for the normalized time is
-   returned.  :attr:`tm_isdst` is forced to 0. Note that the result's
-   :attr:`tm_year` member may be :const:`MINYEAR`\ -1 or :const:`MAXYEAR`\ +1, if
-   *d*.year was ``MINYEAR`` or ``MAXYEAR`` and UTC adjustment spills over a year
+   ``d.utcoffset()``, and a :class:`time.struct_time` for the
+   normalized time is returned.  :attr:`tm_isdst` is forced to 0. Note
+   that an :exc:`OverflowError` may be raised if *d*.year was
+   ``MINYEAR`` or ``MAXYEAR`` and UTC adjustment spills over a year
    boundary.
 
 
@@ -1760,3 +1760,10 @@
 (5)
    For example, if :meth:`utcoffset` returns ``timedelta(hours=-3, minutes=-30)``,
    ``%z`` is replaced with the string ``'-0330'``.
+
+.. versionadded:: 3.2
+
+   When the ``%z`` directive is provided to the :meth:`strptime`
+   method, an aware :class:`datetime` object will be produced.  The
+   ``tzinfo`` of the result will be set to a :class:`timezone`
+   instance.

Modified: python/branches/py3k-dtoa/Doc/library/dis.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/library/dis.rst	(original)
+++ python/branches/py3k-dtoa/Doc/library/dis.rst	Tue Jul  6 17:55:27 2010
@@ -5,12 +5,19 @@
    :synopsis: Disassembler for Python bytecode.
 
 
-The :mod:`dis` module supports the analysis of Python :term:`bytecode` by
-disassembling it.  Since there is no Python assembler, this module defines the
-Python assembly language.  The Python bytecode which this module takes as an
+The :mod:`dis` module supports the analysis of CPython :term:`bytecode` by
+disassembling it. The CPython bytecode which this module takes as an
 input is defined in the file :file:`Include/opcode.h` and used by the compiler
 and the interpreter.
 
+.. impl-detail::
+
+   Bytecode is an implementation detail of the CPython interpreter!  No
+   guarantees are made that bytecode will not be added, removed, or changed
+   between versions of Python.  Use of this module should not be considered to
+   work across Python VMs or Python releases.
+
+
 Example: Given the function :func:`myfunc`::
 
    def myfunc(alist):
@@ -31,23 +38,27 @@
 
 .. function:: dis(x=None)
 
-   Disassemble the *x* object. *x* can denote either a module, a
-   class, a method, a function, or a code object.   For a module, it disassembles
-   all functions.  For a class, it disassembles all methods.  For a single code
-   sequence, it prints one line per bytecode instruction.  If no object is
-   provided, it disassembles the last traceback.
+   Disassemble the *x* object.  *x* can denote either a module, a
+   class, a method, a function, a code object, a string of source code or a
+   byte sequence of raw bytecode.  For a module, it disassembles all
+   functions.  For a class, it disassembles all methods.  For a code object
+   or sequence of raw bytecode, it prints one line per bytecode instruction.
+   Strings are first compiled to code objects with the :func:`compile`
+   built-in function before being disassembled.  If no object is provided,
+   this function disassembles the last traceback.
 
 
 .. function:: distb(tb=None)
 
-   Disassembles the top-of-stack function of a traceback, using the last traceback
-   if none was passed.  The instruction causing the exception is indicated.
+   Disassemble the top-of-stack function of a traceback, using the last
+   traceback if none was passed.  The instruction causing the exception is
+   indicated.
 
 
 .. function:: disassemble(code, lasti=-1)
               disco(code, lasti=-1)
 
-   Disassembles a code object, indicating the last instruction if *lasti* was
+   Disassemble a code object, indicating the last instruction if *lasti* was
    provided.  The output is divided in the following columns:
 
    #. the line number, for the first instruction of each line
@@ -133,225 +144,233 @@
 The Python compiler currently generates the following bytecode instructions.
 
 
-.. opcode:: STOP_CODE ()
+**General instructions**
+
+.. opcode:: STOP_CODE
 
    Indicates end-of-code to the compiler, not used by the interpreter.
 
 
-.. opcode:: NOP ()
+.. opcode:: NOP
 
    Do nothing code.  Used as a placeholder by the bytecode optimizer.
 
 
-.. opcode:: POP_TOP ()
+.. opcode:: POP_TOP
 
    Removes the top-of-stack (TOS) item.
 
 
-.. opcode:: ROT_TWO ()
+.. opcode:: ROT_TWO
 
    Swaps the two top-most stack items.
 
 
-.. opcode:: ROT_THREE ()
+.. opcode:: ROT_THREE
 
    Lifts second and third stack item one position up, moves top down to position
    three.
 
 
-.. opcode:: ROT_FOUR ()
+.. opcode:: ROT_FOUR
 
    Lifts second, third and forth stack item one position up, moves top down to
    position four.
 
 
-.. opcode:: DUP_TOP ()
+.. opcode:: DUP_TOP
 
    Duplicates the reference on top of the stack.
 
-Unary Operations take the top of the stack, apply the operation, and push the
-result back on the stack.
 
+**Unary operations**
+
+Unary operations take the top of the stack, apply the operation, and push the
+result back on the stack.
 
-.. opcode:: UNARY_POSITIVE ()
+.. opcode:: UNARY_POSITIVE
 
    Implements ``TOS = +TOS``.
 
 
-.. opcode:: UNARY_NEGATIVE ()
+.. opcode:: UNARY_NEGATIVE
 
    Implements ``TOS = -TOS``.
 
 
-.. opcode:: UNARY_NOT ()
+.. opcode:: UNARY_NOT
 
    Implements ``TOS = not TOS``.
 
 
-.. opcode:: UNARY_INVERT ()
+.. opcode:: UNARY_INVERT
 
    Implements ``TOS = ~TOS``.
 
 
-.. opcode:: GET_ITER ()
+.. opcode:: GET_ITER
 
    Implements ``TOS = iter(TOS)``.
 
+
+**Binary operations**
+
 Binary operations remove the top of the stack (TOS) and the second top-most
 stack item (TOS1) from the stack.  They perform the operation, and put the
 result back on the stack.
 
-
-.. opcode:: BINARY_POWER ()
+.. opcode:: BINARY_POWER
 
    Implements ``TOS = TOS1 ** TOS``.
 
 
-.. opcode:: BINARY_MULTIPLY ()
+.. opcode:: BINARY_MULTIPLY
 
    Implements ``TOS = TOS1 * TOS``.
 
 
-.. opcode:: BINARY_FLOOR_DIVIDE ()
+.. opcode:: BINARY_FLOOR_DIVIDE
 
    Implements ``TOS = TOS1 // TOS``.
 
 
-.. opcode:: BINARY_TRUE_DIVIDE ()
+.. opcode:: BINARY_TRUE_DIVIDE
 
    Implements ``TOS = TOS1 / TOS``.
 
 
-.. opcode:: BINARY_MODULO ()
+.. opcode:: BINARY_MODULO
 
    Implements ``TOS = TOS1 % TOS``.
 
 
-.. opcode:: BINARY_ADD ()
+.. opcode:: BINARY_ADD
 
    Implements ``TOS = TOS1 + TOS``.
 
 
-.. opcode:: BINARY_SUBTRACT ()
+.. opcode:: BINARY_SUBTRACT
 
    Implements ``TOS = TOS1 - TOS``.
 
 
-.. opcode:: BINARY_SUBSCR ()
+.. opcode:: BINARY_SUBSCR
 
    Implements ``TOS = TOS1[TOS]``.
 
 
-.. opcode:: BINARY_LSHIFT ()
+.. opcode:: BINARY_LSHIFT
 
    Implements ``TOS = TOS1 << TOS``.
 
 
-.. opcode:: BINARY_RSHIFT ()
+.. opcode:: BINARY_RSHIFT
 
    Implements ``TOS = TOS1 >> TOS``.
 
 
-.. opcode:: BINARY_AND ()
+.. opcode:: BINARY_AND
 
    Implements ``TOS = TOS1 & TOS``.
 
 
-.. opcode:: BINARY_XOR ()
+.. opcode:: BINARY_XOR
 
    Implements ``TOS = TOS1 ^ TOS``.
 
 
-.. opcode:: BINARY_OR ()
+.. opcode:: BINARY_OR
 
    Implements ``TOS = TOS1 | TOS``.
 
+
+**In-place operations**
+
 In-place operations are like binary operations, in that they remove TOS and
 TOS1, and push the result back on the stack, but the operation is done in-place
 when TOS1 supports it, and the resulting TOS may be (but does not have to be)
 the original TOS1.
 
-
-.. opcode:: INPLACE_POWER ()
+.. opcode:: INPLACE_POWER
 
    Implements in-place ``TOS = TOS1 ** TOS``.
 
 
-.. opcode:: INPLACE_MULTIPLY ()
+.. opcode:: INPLACE_MULTIPLY
 
    Implements in-place ``TOS = TOS1 * TOS``.
 
 
-.. opcode:: INPLACE_FLOOR_DIVIDE ()
+.. opcode:: INPLACE_FLOOR_DIVIDE
 
    Implements in-place ``TOS = TOS1 // TOS``.
 
 
-.. opcode:: INPLACE_TRUE_DIVIDE ()
+.. opcode:: INPLACE_TRUE_DIVIDE
 
    Implements in-place ``TOS = TOS1 / TOS``.
 
 
-.. opcode:: INPLACE_MODULO ()
+.. opcode:: INPLACE_MODULO
 
    Implements in-place ``TOS = TOS1 % TOS``.
 
 
-.. opcode:: INPLACE_ADD ()
+.. opcode:: INPLACE_ADD
 
    Implements in-place ``TOS = TOS1 + TOS``.
 
 
-.. opcode:: INPLACE_SUBTRACT ()
+.. opcode:: INPLACE_SUBTRACT
 
    Implements in-place ``TOS = TOS1 - TOS``.
 
 
-.. opcode:: INPLACE_LSHIFT ()
+.. opcode:: INPLACE_LSHIFT
 
    Implements in-place ``TOS = TOS1 << TOS``.
 
 
-.. opcode:: INPLACE_RSHIFT ()
+.. opcode:: INPLACE_RSHIFT
 
    Implements in-place ``TOS = TOS1 >> TOS``.
 
 
-.. opcode:: INPLACE_AND ()
+.. opcode:: INPLACE_AND
 
    Implements in-place ``TOS = TOS1 & TOS``.
 
 
-.. opcode:: INPLACE_XOR ()
+.. opcode:: INPLACE_XOR
 
    Implements in-place ``TOS = TOS1 ^ TOS``.
 
 
-.. opcode:: INPLACE_OR ()
+.. opcode:: INPLACE_OR
 
    Implements in-place ``TOS = TOS1 | TOS``.
 
 
-.. opcode:: STORE_SUBSCR ()
+.. opcode:: STORE_SUBSCR
 
    Implements ``TOS1[TOS] = TOS2``.
 
 
-.. opcode:: DELETE_SUBSCR ()
+.. opcode:: DELETE_SUBSCR
 
    Implements ``del TOS1[TOS]``.
 
-Miscellaneous opcodes.
 
+**Miscellaneous opcodes**
 
-.. opcode:: PRINT_EXPR ()
+.. opcode:: PRINT_EXPR
 
    Implements the expression statement for the interactive mode.  TOS is removed
    from the stack and printed.  In non-interactive mode, an expression statement is
    terminated with ``POP_STACK``.
 
 
-.. opcode:: BREAK_LOOP ()
+.. opcode:: BREAK_LOOP
 
    Terminates a loop due to a :keyword:`break` statement.
 
@@ -377,36 +396,35 @@
    Calls ``dict.setitem(TOS1[-i], TOS, TOS1)``.  Used to implement dict
    comprehensions.
 
-
 For all of the SET_ADD, LIST_APPEND and MAP_ADD instructions, while the
 added value or key/value pair is popped off, the container object remains on
 the stack so that it is available for further iterations of the loop.
 
 
-.. opcode:: RETURN_VALUE ()
+.. opcode:: RETURN_VALUE
 
    Returns with TOS to the caller of the function.
 
 
-.. opcode:: YIELD_VALUE ()
+.. opcode:: YIELD_VALUE
 
    Pops ``TOS`` and yields it from a :term:`generator`.
 
 
-.. opcode:: IMPORT_STAR ()
+.. opcode:: IMPORT_STAR
 
    Loads all symbols not starting with ``'_'`` directly from the module TOS to the
    local namespace. The module is popped after loading all names. This opcode
    implements ``from module import *``.
 
 
-.. opcode:: POP_BLOCK ()
+.. opcode:: POP_BLOCK
 
    Removes one block from the block stack.  Per frame, there is a  stack of blocks,
    denoting nested loops, try statements, and such.
 
 
-.. opcode:: POP_EXCEPT ()
+.. opcode:: POP_EXCEPT
 
    Removes one block from the block stack. The popped block must be an exception
    handler block, as implicitly created when entering an except handler.
@@ -414,14 +432,14 @@
    last three popped values are used to restore the exception state.
 
 
-.. opcode:: END_FINALLY ()
+.. opcode:: END_FINALLY
 
    Terminates a :keyword:`finally` clause.  The interpreter recalls whether the
    exception has to be re-raised, or whether the function returns, and continues
    with the outer-next block.
 
 
-.. opcode:: LOAD_BUILD_CLASS ()
+.. opcode:: LOAD_BUILD_CLASS
 
    Pushes :func:`builtins.__build_class__` onto the stack.  It is later called
    by ``CALL_FUNCTION`` to construct a class.
@@ -439,7 +457,7 @@
    :opcode:`UNPACK_SEQUENCE`).
 
 
-.. opcode:: WITH_CLEANUP ()
+.. opcode:: WITH_CLEANUP
 
    Cleans up the stack when a :keyword:`with` statement block exits.  TOS is
    the context manager's :meth:`__exit__` bound method. Below TOS are 1--3
@@ -649,7 +667,7 @@
    Pushes a try block from a try-except clause onto the block stack. *delta* points
    to the finally block.
 
-.. opcode:: STORE_MAP ()
+.. opcode:: STORE_MAP
 
    Store a key and value pair in a dictionary.  Pops the key and value while leaving
    the dictionary on the stack.
@@ -765,7 +783,7 @@
    variable-arguments tuple, followed by explicit keyword and positional arguments.
 
 
-.. opcode:: HAVE_ARGUMENT ()
+.. opcode:: HAVE_ARGUMENT
 
    This is not really an opcode.  It identifies the dividing line between opcodes
    which don't take arguments ``< HAVE_ARGUMENT`` and those which do ``>=

Modified: python/branches/py3k-dtoa/Doc/library/http.server.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/library/http.server.rst	(original)
+++ python/branches/py3k-dtoa/Doc/library/http.server.rst	Tue Jul  6 17:55:27 2010
@@ -284,28 +284,30 @@
       For example usage, see the implementation of the :func:`test` function
       invocation in the :mod:`http.server` module.
 
-The :class:`SimpleHTTPRequestHandler` class can be invoked the following manner
-with the :mod:`http.server` to create a very basic webserver serving files
-relative to the current directory.::
 
-        import http.server
-        import socketserver
+The :class:`SimpleHTTPRequestHandler` class can be used in the following
+manner in order to create a very basic webserver serving files relative to
+the current directory. ::
 
-        PORT = 8000
+   import http.server
+   import socketserver
 
-        Handler = http.server.SimpleHTTPRequestHandler
+   PORT = 8000
 
-        httpd = socketserver.TCPServer(("", PORT), Handler)
+   Handler = http.server.SimpleHTTPRequestHandler
 
-        print("serving at port", PORT)
-        httpd.serve_forever()
+   httpd = socketserver.TCPServer(("", PORT), Handler)
 
-:mod:`http.server` can also be invoked directly using the ``-m`` switch of
-interpreter a with ``port number`` argument which interfaces
-:class:`SimpleHTTPRequestHandler` by default.::
+   print("serving at port", PORT)
+   httpd.serve_forever()
+
+:mod:`http.server` can also be invoked directly using the :option:`-m`
+switch of the interpreter a with ``port number`` argument.  Similar to
+the previous example, this serves files relative to the current directory. ::
 
         python -m http.server 8000
 
+
 .. class:: CGIHTTPRequestHandler(request, client_address, server)
 
    This class is used to serve either files or output of CGI scripts from the

Modified: python/branches/py3k-dtoa/Doc/library/importlib.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/library/importlib.rst	(original)
+++ python/branches/py3k-dtoa/Doc/library/importlib.rst	Tue Jul  6 17:55:27 2010
@@ -18,12 +18,12 @@
 :func:`__import__` function) in Python source code. This provides an
 implementation of :keyword:`import` which is portable to any Python
 interpreter. This also provides a reference implementation which is easier to
-comprehend than one in a programming language other than Python.
+comprehend than one implemented in a programming language other than Python.
 
-Two, the components to implement :keyword:`import` can be exposed in this
+Two, the components to implement :keyword:`import` are exposed in this
 package, making it easier for users to create their own custom objects (known
 generically as an :term:`importer`) to participate in the import process.
-Details on providing custom importers can be found in :pep:`302`.
+Details on custom importers can be found in :pep:`302`.
 
 .. seealso::
 
@@ -36,8 +36,7 @@
         in :data:`sys.modules`).
 
     The :func:`.__import__` function
-        The built-in function for which the :keyword:`import` statement is
-        syntactic sugar.
+        The :keyword:`import` statement is syntactic sugar for this function.
 
     :pep:`235`
         Import on Case-Insensitive Platforms
@@ -46,7 +45,7 @@
         Defining Python Source Code Encodings
 
     :pep:`302`
-        New Import Hooks.
+        New Import Hooks
 
     :pep:`328`
         Imports: Multi-Line and Absolute/Relative
@@ -57,14 +56,16 @@
     :pep:`3120`
         Using UTF-8 as the Default Source Encoding
 
+    :pep:`3147`
+        PYC Repository Directories
+
 
 Functions
 ---------
 
 .. function:: __import__(name, globals={}, locals={}, fromlist=list(), level=0)
 
-    An implementation of the built-in :func:`__import__` function. See the
-    built-in function's documentation for usage instructions.
+    An implementation of the built-in :func:`__import__` function.
 
 .. function:: import_module(name, package=None)
 
@@ -210,22 +211,112 @@
 
     .. method:: get_filename(fullname)
 
-        An abstract method that is to return the value for :attr:`__file__` for
+        An abstract method that is to return the value of :attr:`__file__` for
         the specified module. If no path is available, :exc:`ImportError` is
         raised.
 
+        If source code is available, then the method should return the path to
+        the source file, regardless of whether a bytecode was used to load the
+        module.
+
+
+.. class:: SourceLoader
+
+    An abstract base class for implementing source (and optionally bytecode)
+    file loading. The class inherits from both :class:`ResourceLoader` and
+    :class:`ExecutionLoader`, requiring the implementation of:
+
+    * :meth:`ResourceLoader.get_data`
+    * :meth:`ExecutionLoader.get_filename`
+          Implement to only return the path to the source file; sourceless
+          loading is not supported.
+
+    The abstract methods defined by this class are to add optional bytecode
+    file support. Not implementing these optional methods causes the loader to
+    only work with source code. Implementing the methods allows the loader to
+    work with source *and* bytecode files; it does not allow for *sourceless*
+    loading where only bytecode is provided.  Bytecode files are an
+    optimization to speed up loading by removing the parsing step of Python's
+    compiler, and so no bytecode-specific API is exposed.
+
+    .. method:: path_mtime(self, path)
+
+        Optional abstract method which returns the modification time for the
+        specified path.
+
+    .. method:: set_data(self, path, data)
+
+        Optional abstract method which writes the specified bytes to a file
+        path. Any intermediate directories which do not exist are to be created
+        automatically.
+
+        When writing to the path fails because the path is read-only
+        (:attr:`errno.EACCES`), do not propagate the exception.
+
+    .. method:: get_code(self, fullname)
+
+        Concrete implementation of :meth:`InspectLoader.get_code`.
+
+    .. method:: load_module(self, fullname)
+
+        Concrete implementation of :meth:`Loader.load_module`.
+
+    .. method:: get_source(self, fullname)
+
+        Concrete implementation of :meth:`InspectLoader.get_source`.
+
+    .. method:: is_package(self, fullname)
+
+        Concrete implementation of :meth:`InspectLoader.is_package`. A module
+        is determined to be a package if its file path is a file named
+        ``__init__`` when the file extension is removed.
+
 
 .. class:: PyLoader
 
     An abstract base class inheriting from
-    :class:`importlib.abc.ExecutionLoader` and
-    :class:`importlib.abc.ResourceLoader` designed to ease the loading of
+    :class:`ExecutionLoader` and
+    :class:`ResourceLoader` designed to ease the loading of
     Python source modules (bytecode is not handled; see
-    :class:`importlib.abc.PyPycLoader` for a source/bytecode ABC). A subclass
+    :class:`SourceLoader` for a source/bytecode ABC). A subclass
     implementing this ABC will only need to worry about exposing how the source
     code is stored; all other details for loading Python source code will be
     handled by the concrete implementations of key methods.
 
+    .. deprecated:: 3.2
+        This class has been deprecated in favor of :class:`SourceLoader` and is
+        slated for removal in Python 3.4. See below for how to create a
+        subclass that is compatbile with Python 3.1 onwards.
+
+    If compatibility with Python 3.1 is required, then use the following idiom
+    to implement a subclass that will work with Python 3.1 onwards (make sure
+    to implement :meth:`ExecutionLoader.get_filename`)::
+
+        try:
+            from importlib.abc import SourceLoader
+        except ImportError:
+            from importlib.abc import PyLoader as SourceLoader
+
+
+        class CustomLoader(SourceLoader):
+            def get_filename(self, fullname):
+                """Return the path to the source file."""
+                # Implement ...
+
+            def source_path(self, fullname):
+                """Implement source_path in terms of get_filename."""
+                try:
+                    return self.get_filename(fullname)
+                except ImportError:
+                    return None
+
+            def is_package(self, fullname):
+                """Implement is_package by looking for an __init__ file
+                name as returned by get_filename."""
+                filename = os.path.basename(self.get_filename(fullname))
+                return os.path.splitext(filename)[0] == '__init__'
+
+
     .. method:: source_path(fullname)
 
         An abstract method that returns the path to the source code for a
@@ -267,10 +358,18 @@
 
 .. class:: PyPycLoader
 
-    An abstract base class inheriting from :class:`importlib.abc.PyLoader`.
+    An abstract base class inheriting from :class:`PyLoader`.
     This ABC is meant to help in creating loaders that support both Python
     source and bytecode.
 
+    .. deprecated:: 3.2
+        This class has been deprecated in favor of :class:`SourceLoader` and to
+        properly support :pep:`3147`. If compatibility is required with
+        Python 3.1, implement both :class:`SourceLoader` and :class:`PyLoader`;
+        instructions on how to do so are included in the documentation for
+        :class:`PyLoader`. Do note that this solution will not support
+        sourceless/bytecode-only loading; only source *and* bytecode loading.
+
     .. method:: source_mtime(fullname)
 
         An abstract method which returns the modification time for the source
@@ -289,8 +388,8 @@
     .. method:: get_filename(fullname)
 
         A concrete implementation of
-        :meth:`importlib.abc.ExecutionLoader.get_filename` that relies on
-        :meth:`importlib.abc.PyLoader.source_path` and :meth:`bytecode_path`.
+        :meth:`ExecutionLoader.get_filename` that relies on
+        :meth:`PyLoader.source_path` and :meth:`bytecode_path`.
         If :meth:`source_path` returns a path, then that value is returned.
         Else if :meth:`bytecode_path` returns a path, that path will be
         returned. If a path is not available from both methods,
@@ -417,100 +516,3 @@
     attribute to be used at the global level of the module during
     initialization.
 
-
-Example
--------
-
-Below is an example meta path importer that uses a dict for back-end storage
-for source code. While not an optimal solution -- manipulations of
-:attr:`__path__` on packages does not influence import -- it does illustrate
-what little is required to implement an importer.
-
-.. testcode::
-
-    """An importer where source is stored in a dict."""
-    from importlib import abc
-
-
-    class DictImporter(abc.Finder, abc.PyLoader):
-
-        """A meta path importer that stores source code in a dict.
-
-        The keys are the module names -- packages must end in ``.__init__``.
-        The values must be something that can be passed to 'bytes'.
-
-        """
-
-        def __init__(self, memory):
-            """Store the dict."""
-            self.memory = memory
-
-        def contains(self, name):
-            """See if a module or package is in the dict."""
-            if name in self.memory:
-                return name
-            package_name = '{}.__init__'.format(name)
-            if  package_name in self.memory:
-                return package_name
-            return False
-
-        __contains__ = contains  # Convenience.
-
-        def find_module(self, fullname, path=None):
-            """Find the module in the dict."""
-            if fullname in self:
-                return self
-            return None
-
-        def source_path(self, fullname):
-            """Return the module name if the module is in the dict."""
-            if not fullname in self:
-                raise ImportError
-            return fullname
-
-        def get_data(self, path):
-            """Return the bytes for the source.
-
-            The value found in the dict is passed through 'bytes' before being
-            returned.
-
-            """
-            name = self.contains(path)
-            if not name:
-                raise IOError
-            return bytes(self.memory[name])
-
-        def is_package(self, fullname):
-            """Tell if module is a package based on whether the dict contains the
-            name with ``.__init__`` appended to it."""
-            if fullname not in self:
-                raise ImportError
-            if fullname in self.memory:
-                return False
-            # If name is in this importer but not as it is then it must end in
-            # ``__init__``.
-            else:
-                return True
-
-.. testcode::
-    :hide:
-
-    import importlib
-    import sys
-
-
-    # Build the dict; keys of name, value of __package__.
-    names = {'_top_level': '', '_pkg.__init__': '_pkg', '_pkg.mod': '_pkg'}
-    source = {name: "name = {!r}".format(name).encode() for name in names}
-
-    # Register the meta path importer.
-    importer = DictImporter(source)
-    sys.meta_path.append(importer)
-
-    # Sanity check.
-    for name in names:
-        module = importlib.import_module(name)
-        assert module.__name__ == name
-        assert getattr(module, 'name') == name
-        assert module.__loader__ is importer
-        assert module.__package__ == names[name]

Modified: python/branches/py3k-dtoa/Doc/library/logging.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/library/logging.rst	(original)
+++ python/branches/py3k-dtoa/Doc/library/logging.rst	Tue Jul  6 17:55:27 2010
@@ -53,10 +53,12 @@
 
 Most applications are probably going to want to log to a file, so let's start
 with that case. Using the :func:`basicConfig` function, we can set up the
-default handler so that debug messages are written to a file::
+default handler so that debug messages are written to a file (in the example,
+we assume that you have the appropriate permissions to create a file called
+*example.log* in the current directory)::
 
    import logging
-   LOG_FILENAME = '/tmp/logging_example.out'
+   LOG_FILENAME = 'example.log'
    logging.basicConfig(filename=LOG_FILENAME,level=logging.DEBUG)
 
    logging.debug('This message should go to the log file')
@@ -75,7 +77,7 @@
    import logging
    import logging.handlers
 
-   LOG_FILENAME = '/tmp/logging_rotatingfile_example.out'
+   LOG_FILENAME = 'logging_rotatingfile_example.out'
 
    # Set up a specific logger with our desired output level
    my_logger = logging.getLogger('MyLogger')
@@ -100,14 +102,14 @@
 The result should be 6 separate files, each with part of the log history for the
 application::
 
-   /tmp/logging_rotatingfile_example.out
-   /tmp/logging_rotatingfile_example.out.1
-   /tmp/logging_rotatingfile_example.out.2
-   /tmp/logging_rotatingfile_example.out.3
-   /tmp/logging_rotatingfile_example.out.4
-   /tmp/logging_rotatingfile_example.out.5
+   logging_rotatingfile_example.out
+   logging_rotatingfile_example.out.1
+   logging_rotatingfile_example.out.2
+   logging_rotatingfile_example.out.3
+   logging_rotatingfile_example.out.4
+   logging_rotatingfile_example.out.5
 
-The most current file is always :file:`/tmp/logging_rotatingfile_example.out`,
+The most current file is always :file:`logging_rotatingfile_example.out`,
 and each time it reaches the size limit it is renamed with the suffix
 ``.1``. Each of the existing backup files is renamed to increment the suffix
 (``.1`` becomes ``.2``, etc.)  and the ``.6`` file is erased.
@@ -244,7 +246,7 @@
   methods listed above, but this is how to log at custom log levels.
 
 :func:`getLogger` returns a reference to a logger instance with the specified
-if it is provided, or ``root`` if not.  The names are period-separated
+name if it is provided, or ``root`` if not.  The names are period-separated
 hierarchical structures.  Multiple calls to :func:`getLogger` with the same name
 will return a reference to the same logger object.  Loggers that are further
 down in the hierarchical list are children of loggers higher up in the list.
@@ -321,24 +323,34 @@
 Configuring Logging
 ^^^^^^^^^^^^^^^^^^^
 
-Programmers can configure logging either by creating loggers, handlers, and
-formatters explicitly in a main module with the configuration methods listed
-above (using Python code), or by creating a logging config file.  The following
-code is an example of configuring a very simple logger, a console handler, and a
-simple formatter in a Python module::
+Programmers can configure logging in three ways:
+
+1. Creating loggers, handlers, and formatters explicitly using Python
+   code that calls the configuration methods listed above.
+2. Creating a logging config file and reading it using the :func:`fileConfig`
+   function.
+3. Creating a dictionary of configuration information and passing it
+   to the :func:`dictConfig` function.
+
+The following example configures a very simple logger, a console
+handler, and a simple formatter using Python code::
 
     import logging
 
     # create logger
     logger = logging.getLogger("simple_example")
     logger.setLevel(logging.DEBUG)
+
     # create console handler and set level to debug
     ch = logging.StreamHandler()
     ch.setLevel(logging.DEBUG)
+
     # create formatter
     formatter = logging.Formatter("%(asctime)s - %(name)s - %(levelname)s - %(message)s")
+
     # add formatter to ch
     ch.setFormatter(formatter)
+
     # add ch to logger
     logger.addHandler(ch)
 
@@ -428,6 +440,45 @@
 class defined in package `mypackage` and module `mymodule`, where `mypackage`
 is available on the Python import path).
 
+.. versionchanged:: 3.2
+
+In Python 3.2, a new means of configuring logging has been introduced, using
+dictionaries to hold configuration information. This provides a superset of the
+functionality of the config-file-based approach outlined above, and is the
+recommended configuration method for new applications and deployments. Because
+a Python dictionary is used to hold configuration information, and since you
+can populate that dictionary using different means, you have more options for
+configuration. For example, you can use a configuration file in JSON format,
+or, if you have access to YAML processing functionality, a file in YAML
+format, to populate the configuration dictionary. Or, of course, you can
+construct the dictionary in Python code, receive it in pickled form over a
+socket, or use whatever approach makes sense for your application.
+
+Here's an example of the same configuration as above, in YAML format for
+the new dictionary-based approach::
+
+    version: 1
+    formatters:
+      simple:
+        format: format=%(asctime)s - %(name)s - %(levelname)s - %(message)s
+    handlers:
+      console:
+        class: logging.StreamHandler
+        level: DEBUG
+        formatter: simple
+        stream: ext://sys.stdout
+    loggers:
+      simpleExample:
+        level: DEBUG
+        handlers: [console]
+        propagate: no
+    root:
+        level: DEBUG
+        handlers: [console]
+
+For more information about logging using a dictionary, see
+:ref:`logging-config-api`.
+
 .. _library-config:
 
 Configuring Logging for a Library
@@ -1065,14 +1116,14 @@
 
    logging.basicConfig(level=logging.DEBUG,
                        format='%(asctime)s %(levelname)s %(message)s',
-                       filename='/tmp/myapp.log',
+                       filename='myapp.log',
                        filemode='w')
    logging.debug('A debug message')
    logging.info('Some information')
    logging.warning('A shot across the bows')
 
 The :meth:`basicConfig` method is used to change the configuration defaults,
-which results in output (written to ``/tmp/myapp.log``) which should look
+which results in output (written to ``myapp.log``) which should look
 something like the following::
 
    2004-07-02 13:00:08,743 DEBUG A debug message
@@ -1364,6 +1415,8 @@
    2008-01-18 14:49:54,033 d.e.f WARNING  IP: 127.0.0.1       User: jim      A message at WARNING level with 2 parameters
 
 
+.. _multiple-processes:
+
 Logging to a single file from multiple processes
 ------------------------------------------------
 
@@ -1519,6 +1572,11 @@
       69 myapp.area2     WARNING  Jail zesty vixen who grabbed pay from quack.
       69 myapp.area2     ERROR    The five boxing wizards jump quickly.
 
+Note that there are some security issues with pickle in some scenarios. If
+these affect you, you can use an alternative serialization scheme by overriding
+the :meth:`makePickle` method and implementing your alternative there, as
+well as adapting the above script to use your alternative serialization.
+
 Using arbitrary objects as messages
 -----------------------------------
 
@@ -1674,6 +1732,8 @@
    :exc:`NotImplementedError`.
 
 
+.. _stream-handler:
+
 StreamHandler
 ^^^^^^^^^^^^^
 
@@ -1707,6 +1767,8 @@
       no output, so an explicit :meth:`flush` call may be needed at times.
 
 
+.. _file-handler:
+
 FileHandler
 ^^^^^^^^^^^
 
@@ -1715,7 +1777,7 @@
 :class:`StreamHandler`.
 
 
-.. class:: FileHandler(filename, mode='a', encoding=None, delay=0)
+.. class:: FileHandler(filename, mode='a', encoding=None, delay=False)
 
    Returns a new instance of the :class:`FileHandler` class. The specified file is
    opened and used as the stream for logging. If *mode* is not specified,
@@ -1733,6 +1795,7 @@
 
       Outputs the record to the file.
 
+.. _null-handler:
 
 NullHandler
 ^^^^^^^^^^^
@@ -1756,6 +1819,8 @@
 See :ref:`library-config` for more information on how to use
 :class:`NullHandler`.
 
+.. _watched-file-handler:
+
 WatchedFileHandler
 ^^^^^^^^^^^^^^^^^^
 
@@ -1794,6 +1859,7 @@
       changed.  If it has, the existing stream is flushed and closed and the
       file opened again, before outputting the record to the file.
 
+.. _rotating-file-handler:
 
 RotatingFileHandler
 ^^^^^^^^^^^^^^^^^^^
@@ -1834,6 +1900,7 @@
       Outputs the record to the file, catering for rollover as described
       previously.
 
+.. _timed-rotating-file-handler:
 
 TimedRotatingFileHandler
 ^^^^^^^^^^^^^^^^^^^^^^^^
@@ -1843,7 +1910,7 @@
 timed intervals.
 
 
-.. class:: TimedRotatingFileHandler(filename, when='h', interval=1, backupCount=0, encoding=None, delay=0, utc=False)
+.. class:: TimedRotatingFileHandler(filename, when='h', interval=1, backupCount=0, encoding=None, delay=False, utc=False)
 
    Returns a new instance of the :class:`TimedRotatingFileHandler` class. The
    specified file is opened and used as the stream for logging. On rotating it also
@@ -1886,6 +1953,9 @@
    one is deleted. The deletion logic uses the interval to determine which
    files to delete, so changing the interval may leave old files lying around.
 
+   If *delay* is true, then file opening is deferred until the first call to
+   :meth:`emit`.
+
 
    .. method:: doRollover()
 
@@ -1897,6 +1967,8 @@
       Outputs the record to the file, catering for rollover as described above.
 
 
+.. _socket-handler:
+
 SocketHandler
 ^^^^^^^^^^^^^
 
@@ -1943,6 +2015,11 @@
       Pickles the record's attribute dictionary in binary format with a length
       prefix, and returns it ready for transmission across the socket.
 
+      Note that pickles aren't completely secure. If you are concerned about
+      security, you may want to override this method to implement a more secure
+      mechanism. For example, you can sign pickles using HMAC and then verify
+      them on the receiving end, or alternatively you can disable unpickling of
+      global objects on the receiving end.
 
    .. method:: send(packet)
 
@@ -1950,6 +2027,8 @@
       partial sends which can happen when the network is busy.
 
 
+.. _datagram-handler:
+
 DatagramHandler
 ^^^^^^^^^^^^^^^
 
@@ -1983,6 +2062,8 @@
       Send a pickled string to a socket.
 
 
+.. _syslog-handler:
+
 SysLogHandler
 ^^^^^^^^^^^^^
 
@@ -2172,6 +2253,7 @@
       lookup to get the message ID. This version returns 1, which is the base
       message ID in :file:`win32service.pyd`.
 
+.. _smtp-handler:
 
 SMTPHandler
 ^^^^^^^^^^^
@@ -2200,6 +2282,7 @@
       If you want to specify a subject line which is record-dependent, override
       this method.
 
+.. _memory-handler:
 
 MemoryHandler
 ^^^^^^^^^^^^^
@@ -2270,6 +2353,8 @@
       Checks for buffer full or a record at the *flushLevel* or higher.
 
 
+.. _http-handler:
+
 HTTPHandler
 ^^^^^^^^^^^
 
@@ -2412,6 +2497,7 @@
       just uses :func:`traceback.print_exception`. The resulting string is
       returned.
 
+.. _filter:
 
 Filter Objects
 --------------
@@ -2437,6 +2523,7 @@
       yes. If deemed appropriate, the record may be modified in-place by this
       method.
 
+.. _log-record:
 
 LogRecord Objects
 -----------------
@@ -2468,6 +2555,7 @@
       Returns the message for this :class:`LogRecord` instance after merging any
       user-supplied arguments with the message.
 
+.. _logger-adapter:
 
 LoggerAdapter Objects
 ---------------------
@@ -2553,19 +2641,57 @@
 in :mod:`logging` itself) and defining handlers which are declared either in
 :mod:`logging` or :mod:`logging.handlers`.
 
+.. function:: dictConfig(config)
+
+    Takes the logging configuration from a dictionary.  The contents of
+    this dictionary are described in :ref:`logging-config-dictschema`
+    below.
+
+    If an error is encountered during configuration, this function will
+    raise a :exc:`ValueError`, :exc:`TypeError`, :exc:`AttributeError`
+    or :exc:`ImportError` with a suitably descriptive message.  The
+    following is a (possibly incomplete) list of conditions which will
+    raise an error:
+
+    * A ``level`` which is not a string or which is a string not
+      corresponding to an actual logging level.
+    * A ``propagate`` value which is not a boolean.
+    * An id which does not have a corresponding destination.
+    * A non-existent handler id found during an incremental call.
+    * An invalid logger name.
+    * Inability to resolve to an internal or external object.
+
+    Parsing is performed by the :class:`DictConfigurator` class, whose
+    constructor is passed the dictionary used for configuration, and
+    has a :meth:`configure` method.  The :mod:`logging.config` module
+    has a callable attribute :attr:`dictConfigClass`
+    which is initially set to :class:`DictConfigurator`.
+    You can replace the value of :attr:`dictConfigClass` with a
+    suitable implementation of your own.
+
+    :func:`dictConfig` calls :attr:`dictConfigClass` passing
+    the specified dictionary, and then calls the :meth:`configure` method on
+    the returned object to put the configuration into effect::
+
+          def dictConfig(config):
+              dictConfigClass(config).configure()
+
+    For example, a subclass of :class:`DictConfigurator` could call
+    ``DictConfigurator.__init__()`` in its own :meth:`__init__()`, then
+    set up custom prefixes which would be usable in the subsequent
+    :meth:`configure` call. :attr:`dictConfigClass` would be bound to
+    this new subclass, and then :func:`dictConfig` could be called exactly as
+    in the default, uncustomized state.
 
-.. function:: fileConfig(fname, defaults=None, disable_existing_loggers=True)
+.. function:: fileConfig(fname[, defaults])
 
    Reads the logging configuration from a :mod:`configparser`\-format file named
    *fname*. This function can be called several times from an application,
-   allowing an end user the ability to select from various pre-canned
+   allowing an end user to select from various pre-canned
    configurations (if the developer provides a mechanism to present the choices
    and load the chosen configuration). Defaults to be passed to the ConfigParser
    can be specified in the *defaults* argument.
 
-   If *disable_existing_loggers* is true, any existing loggers that are not
-   children of named loggers will be disabled.
-
 
 .. function:: listen(port=DEFAULT_LOGGING_CONFIG_PORT)
 
@@ -2589,6 +2715,402 @@
    :func:`listen`.
 
 
+.. _logging-config-dictschema:
+
+Configuration dictionary schema
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Describing a logging configuration requires listing the various
+objects to create and the connections between them; for example, you
+may create a handler named "console" and then say that the logger
+named "startup" will send its messages to the "console" handler.
+These objects aren't limited to those provided by the :mod:`logging`
+module because you might write your own formatter or handler class.
+The parameters to these classes may also need to include external
+objects such as ``sys.stderr``.  The syntax for describing these
+objects and connections is defined in :ref:`logging-config-dict-connections`
+below.
+
+Dictionary Schema Details
+"""""""""""""""""""""""""
+
+The dictionary passed to :func:`dictConfig` must contain the following
+keys:
+
+* `version` - to be set to an integer value representing the schema
+  version.  The only valid value at present is 1, but having this key
+  allows the schema to evolve while still preserving backwards
+  compatibility.
+
+All other keys are optional, but if present they will be interpreted
+as described below.  In all cases below where a 'configuring dict' is
+mentioned, it will be checked for the special ``'()'`` key to see if a
+custom instantiation is required.  If so, the mechanism described in
+:ref:`logging-config-dict-userdef` below is used to create an instance;
+otherwise, the context is used to determine what to instantiate.
+
+* `formatters` - the corresponding value will be a dict in which each
+  key is a formatter id and each value is a dict describing how to
+  configure the corresponding Formatter instance.
+
+  The configuring dict is searched for keys ``format`` and ``datefmt``
+  (with defaults of ``None``) and these are used to construct a
+  :class:`logging.Formatter` instance.
+
+* `filters` - the corresponding value will be a dict in which each key
+  is a filter id and each value is a dict describing how to configure
+  the corresponding Filter instance.
+
+  The configuring dict is searched for the key ``name`` (defaulting to the
+  empty string) and this is used to construct a :class:`logging.Filter`
+  instance.
+
+* `handlers` - the corresponding value will be a dict in which each
+  key is a handler id and each value is a dict describing how to
+  configure the corresponding Handler instance.
+
+  The configuring dict is searched for the following keys:
+
+  * ``class`` (mandatory).  This is the fully qualified name of the
+    handler class.
+
+  * ``level`` (optional).  The level of the handler.
+
+  * ``formatter`` (optional).  The id of the formatter for this
+    handler.
+
+  * ``filters`` (optional).  A list of ids of the filters for this
+    handler.
+
+  All *other* keys are passed through as keyword arguments to the
+  handler's constructor.  For example, given the snippet::
+
+      handlers:
+        console:
+          class : logging.StreamHandler
+          formatter: brief
+          level   : INFO
+          filters: [allow_foo]
+          stream  : ext://sys.stdout
+        file:
+          class : logging.handlers.RotatingFileHandler
+          formatter: precise
+          filename: logconfig.log
+          maxBytes: 1024
+          backupCount: 3
+
+  the handler with id ``console`` is instantiated as a
+  :class:`logging.StreamHandler`, using ``sys.stdout`` as the underlying
+  stream.  The handler with id ``file`` is instantiated as a
+  :class:`logging.handlers.RotatingFileHandler` with the keyword arguments
+  ``filename='logconfig.log', maxBytes=1024, backupCount=3``.
+
+* `loggers` - the corresponding value will be a dict in which each key
+  is a logger name and each value is a dict describing how to
+  configure the corresponding Logger instance.
+
+  The configuring dict is searched for the following keys:
+
+  * ``level`` (optional).  The level of the logger.
+
+  * ``propagate`` (optional).  The propagation setting of the logger.
+
+  * ``filters`` (optional).  A list of ids of the filters for this
+    logger.
+
+  * ``handlers`` (optional).  A list of ids of the handlers for this
+    logger.
+
+  The specified loggers will be configured according to the level,
+  propagation, filters and handlers specified.
+
+* `root` - this will be the configuration for the root logger.
+  Processing of the configuration will be as for any logger, except
+  that the ``propagate`` setting will not be applicable.
+
+* `incremental` - whether the configuration is to be interpreted as
+  incremental to the existing configuration.  This value defaults to
+  ``False``, which means that the specified configuration replaces the
+  existing configuration with the same semantics as used by the
+  existing :func:`fileConfig` API.
+
+  If the specified value is ``True``, the configuration is processed
+  as described in the section on :ref:`logging-config-dict-incremental`.
+
+* `disable_existing_loggers` - whether any existing loggers are to be
+  disabled. This setting mirrors the parameter of the same name in
+  :func:`fileConfig`. If absent, this parameter defaults to ``True``.
+  This value is ignored if `incremental` is ``True``.
+
+.. _logging-config-dict-incremental:
+
+Incremental Configuration
+"""""""""""""""""""""""""
+
+It is difficult to provide complete flexibility for incremental
+configuration.  For example, because objects such as filters
+and formatters are anonymous, once a configuration is set up, it is
+not possible to refer to such anonymous objects when augmenting a
+configuration.
+
+Furthermore, there is not a compelling case for arbitrarily altering
+the object graph of loggers, handlers, filters, formatters at
+run-time, once a configuration is set up; the verbosity of loggers and
+handlers can be controlled just by setting levels (and, in the case of
+loggers, propagation flags).  Changing the object graph arbitrarily in
+a safe way is problematic in a multi-threaded environment; while not
+impossible, the benefits are not worth the complexity it adds to the
+implementation.
+
+Thus, when the ``incremental`` key of a configuration dict is present
+and is ``True``, the system will completely ignore any ``formatters`` and
+``filters`` entries, and process only the ``level``
+settings in the ``handlers`` entries, and the ``level`` and
+``propagate`` settings in the ``loggers`` and ``root`` entries.
+
+Using a value in the configuration dict lets configurations to be sent
+over the wire as pickled dicts to a socket listener. Thus, the logging
+verbosity of a long-running application can be altered over time with
+no need to stop and restart the application.
+
+.. _logging-config-dict-connections:
+
+Object connections
+""""""""""""""""""
+
+The schema describes a set of logging objects - loggers,
+handlers, formatters, filters - which are connected to each other in
+an object graph.  Thus, the schema needs to represent connections
+between the objects.  For example, say that, once configured, a
+particular logger has attached to it a particular handler.  For the
+purposes of this discussion, we can say that the logger represents the
+source, and the handler the destination, of a connection between the
+two.  Of course in the configured objects this is represented by the
+logger holding a reference to the handler.  In the configuration dict,
+this is done by giving each destination object an id which identifies
+it unambiguously, and then using the id in the source object's
+configuration to indicate that a connection exists between the source
+and the destination object with that id.
+
+So, for example, consider the following YAML snippet::
+
+    formatters:
+      brief:
+        # configuration for formatter with id 'brief' goes here
+      precise:
+        # configuration for formatter with id 'precise' goes here
+    handlers:
+      h1: #This is an id
+       # configuration of handler with id 'h1' goes here
+       formatter: brief
+      h2: #This is another id
+       # configuration of handler with id 'h2' goes here
+       formatter: precise
+    loggers:
+      foo.bar.baz:
+        # other configuration for logger 'foo.bar.baz'
+        handlers: [h1, h2]
+
+(Note: YAML used here because it's a little more readable than the
+equivalent Python source form for the dictionary.)
+
+The ids for loggers are the logger names which would be used
+programmatically to obtain a reference to those loggers, e.g.
+``foo.bar.baz``.  The ids for Formatters and Filters can be any string
+value (such as ``brief``, ``precise`` above) and they are transient,
+in that they are only meaningful for processing the configuration
+dictionary and used to determine connections between objects, and are
+not persisted anywhere when the configuration call is complete.
+
+The above snippet indicates that logger named ``foo.bar.baz`` should
+have two handlers attached to it, which are described by the handler
+ids ``h1`` and ``h2``. The formatter for ``h1`` is that described by id
+``brief``, and the formatter for ``h2`` is that described by id
+``precise``.
+
+
+.. _logging-config-dict-userdef:
+
+User-defined objects
+""""""""""""""""""""
+
+The schema supports user-defined objects for handlers, filters and
+formatters.  (Loggers do not need to have different types for
+different instances, so there is no support in this configuration
+schema for user-defined logger classes.)
+
+Objects to be configured are described by dictionaries
+which detail their configuration.  In some places, the logging system
+will be able to infer from the context how an object is to be
+instantiated, but when a user-defined object is to be instantiated,
+the system will not know how to do this.  In order to provide complete
+flexibility for user-defined object instantiation, the user needs
+to provide a 'factory' - a callable which is called with a
+configuration dictionary and which returns the instantiated object.
+This is signalled by an absolute import path to the factory being
+made available under the special key ``'()'``.  Here's a concrete
+example::
+
+    formatters:
+      brief:
+        format: '%(message)s'
+      default:
+        format: '%(asctime)s %(levelname)-8s %(name)-15s %(message)s'
+        datefmt: '%Y-%m-%d %H:%M:%S'
+      custom:
+          (): my.package.customFormatterFactory
+          bar: baz
+          spam: 99.9
+          answer: 42
+
+The above YAML snippet defines three formatters.  The first, with id
+``brief``, is a standard :class:`logging.Formatter` instance with the
+specified format string.  The second, with id ``default``, has a
+longer format and also defines the time format explicitly, and will
+result in a :class:`logging.Formatter` initialized with those two format
+strings.  Shown in Python source form, the ``brief`` and ``default``
+formatters have configuration sub-dictionaries::
+
+    {
+      'format' : '%(message)s'
+    }
+
+and::
+
+    {
+      'format' : '%(asctime)s %(levelname)-8s %(name)-15s %(message)s',
+      'datefmt' : '%Y-%m-%d %H:%M:%S'
+    }
+
+respectively, and as these dictionaries do not contain the special key
+``'()'``, the instantiation is inferred from the context: as a result,
+standard :class:`logging.Formatter` instances are created.  The
+configuration sub-dictionary for the third formatter, with id
+``custom``, is::
+
+  {
+    '()' : 'my.package.customFormatterFactory',
+    'bar' : 'baz',
+    'spam' : 99.9,
+    'answer' : 42
+  }
+
+and this contains the special key ``'()'``, which means that
+user-defined instantiation is wanted.  In this case, the specified
+factory callable will be used. If it is an actual callable it will be
+used directly - otherwise, if you specify a string (as in the example)
+the actual callable will be located using normal import mechanisms.
+The callable will be called with the **remaining** items in the
+configuration sub-dictionary as keyword arguments.  In the above
+example, the formatter with id ``custom`` will be assumed to be
+returned by the call::
+
+    my.package.customFormatterFactory(bar='baz', spam=99.9, answer=42)
+
+The key ``'()'`` has been used as the special key because it is not a
+valid keyword parameter name, and so will not clash with the names of
+the keyword arguments used in the call.  The ``'()'`` also serves as a
+mnemonic that the corresponding value is a callable.
+
+
+.. _logging-config-dict-externalobj:
+
+Access to external objects
+""""""""""""""""""""""""""
+
+There are times where a configuration needs to refer to objects
+external to the configuration, for example ``sys.stderr``.  If the
+configuration dict is constructed using Python code, this is
+straightforward, but a problem arises when the configuration is
+provided via a text file (e.g. JSON, YAML).  In a text file, there is
+no standard way to distinguish ``sys.stderr`` from the literal string
+``'sys.stderr'``.  To facilitate this distinction, the configuration
+system looks for certain special prefixes in string values and
+treat them specially.  For example, if the literal string
+``'ext://sys.stderr'`` is provided as a value in the configuration,
+then the ``ext://`` will be stripped off and the remainder of the
+value processed using normal import mechanisms.
+
+The handling of such prefixes is done in a way analogous to protocol
+handling: there is a generic mechanism to look for prefixes which
+match the regular expression ``^(?P<prefix>[a-z]+)://(?P<suffix>.*)$``
+whereby, if the ``prefix`` is recognised, the ``suffix`` is processed
+in a prefix-dependent manner and the result of the processing replaces
+the string value.  If the prefix is not recognised, then the string
+value will be left as-is.
+
+
+.. _logging-config-dict-internalobj:
+
+Access to internal objects
+""""""""""""""""""""""""""
+
+As well as external objects, there is sometimes also a need to refer
+to objects in the configuration.  This will be done implicitly by the
+configuration system for things that it knows about.  For example, the
+string value ``'DEBUG'`` for a ``level`` in a logger or handler will
+automatically be converted to the value ``logging.DEBUG``, and the
+``handlers``, ``filters`` and ``formatter`` entries will take an
+object id and resolve to the appropriate destination object.
+
+However, a more generic mechanism is needed for user-defined
+objects which are not known to the :mod:`logging` module.  For
+example, consider :class:`logging.handlers.MemoryHandler`, which takes
+a ``target`` argument which is another handler to delegate to. Since
+the system already knows about this class, then in the configuration,
+the given ``target`` just needs to be the object id of the relevant
+target handler, and the system will resolve to the handler from the
+id.  If, however, a user defines a ``my.package.MyHandler`` which has
+an ``alternate`` handler, the configuration system would not know that
+the ``alternate`` referred to a handler.  To cater for this, a generic
+resolution system allows the user to specify::
+
+    handlers:
+      file:
+        # configuration of file handler goes here
+
+      custom:
+        (): my.package.MyHandler
+        alternate: cfg://handlers.file
+
+The literal string ``'cfg://handlers.file'`` will be resolved in an
+analogous way to strings with the ``ext://`` prefix, but looking
+in the configuration itself rather than the import namespace.  The
+mechanism allows access by dot or by index, in a similar way to
+that provided by ``str.format``.  Thus, given the following snippet::
+
+    handlers:
+      email:
+        class: logging.handlers.SMTPHandler
+        mailhost: localhost
+        fromaddr: my_app at domain.tld
+        toaddrs:
+          - support_team at domain.tld
+          - dev_team at domain.tld
+        subject: Houston, we have a problem.
+
+in the configuration, the string ``'cfg://handlers'`` would resolve to
+the dict with key ``handlers``, the string ``'cfg://handlers.email``
+would resolve to the dict with key ``email`` in the ``handlers`` dict,
+and so on.  The string ``'cfg://handlers.email.toaddrs[1]`` would
+resolve to ``'dev_team.domain.tld'`` and the string
+``'cfg://handlers.email.toaddrs[0]'`` would resolve to the value
+``'support_team at domain.tld'``. The ``subject`` value could be accessed
+using either ``'cfg://handlers.email.subject'`` or, equivalently,
+``'cfg://handlers.email[subject]'``.  The latter form only needs to be
+used if the key contains spaces or non-alphanumeric characters.  If an
+index value consists only of decimal digits, access will be attempted
+using the corresponding integer value, falling back to the string
+value if needed.
+
+Given a string ``cfg://handlers.myhandler.mykey.123``, this will
+resolve to ``config_dict['handlers']['myhandler']['mykey']['123']``.
+If the string is specified as ``cfg://handlers.myhandler.mykey[123]``,
+the system will attempt to retrieve the value from
+``config_dict['handlers']['myhandler']['mykey'][123]``, and fall back
+to ``config_dict['handlers']['myhandler']['mykey']['123']`` if that
+fails.
+
 .. _logging-config-fileformat:
 
 Configuration file format

Modified: python/branches/py3k-dtoa/Doc/library/os.path.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/library/os.path.rst	(original)
+++ python/branches/py3k-dtoa/Doc/library/os.path.rst	Tue Jul  6 17:55:27 2010
@@ -201,6 +201,7 @@
    Normalize the case of a pathname.  On Unix and Mac OS X, this returns the
    path unchanged; on case-insensitive filesystems, it converts the path to
    lowercase.  On Windows, it also converts forward slashes to backward slashes.
+   Raise a TypeError if the type of *path* is not ``str`` or ``bytes``.
 
 
 .. function:: normpath(path)

Modified: python/branches/py3k-dtoa/Doc/library/os.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/library/os.rst	(original)
+++ python/branches/py3k-dtoa/Doc/library/os.rst	Tue Jul  6 17:55:27 2010
@@ -1115,7 +1115,8 @@
 
    Create a directory named *path* with numeric mode *mode*. The default *mode*
    is ``0o777`` (octal).  On some systems, *mode* is ignored.  Where it is used,
-   the current umask value is first masked out.
+   the current umask value is first masked out.  If the directory already
+   exists, :exc:`OSError` is raised.
 
    It is also possible to create temporary directories; see the
    :mod:`tempfile` module's :func:`tempfile.mkdtemp` function.

Modified: python/branches/py3k-dtoa/Doc/library/pdb.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/library/pdb.rst	(original)
+++ python/branches/py3k-dtoa/Doc/library/pdb.rst	Tue Jul  6 17:55:27 2010
@@ -210,11 +210,13 @@
    Print a stack trace, with the most recent frame at the bottom.  An arrow
    indicates the current frame, which determines the context of most commands.
 
-d(own)
-   Move the current frame one level down in the stack trace (to a newer frame).
-
-u(p)
-   Move the current frame one level up in the stack trace (to an older frame).
+d(own) [*count*]
+   Move the current frame *count* (default one) levels down in the stack trace
+   (to a newer frame).
+
+u(p) [*count*]
+   Move the current frame *count* (default one) levels up in the stack trace
+   (to an older frame).
 
 b(reak) [[*filename*:]\ *lineno* | *function*\ [, *condition*]]
    With a *lineno* argument, set a break there in the current file.  With a

Modified: python/branches/py3k-dtoa/Doc/library/re.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/library/re.rst	(original)
+++ python/branches/py3k-dtoa/Doc/library/re.rst	Tue Jul  6 17:55:27 2010
@@ -709,18 +709,12 @@
 
    The :class:`RegexObject` class supports the following methods and attributes:
 
+   .. method:: RegexObject.search(string[, pos[, endpos]])
 
-   .. method:: RegexObject.match(string[, pos[, endpos]])
-
-      If zero or more characters at the beginning of *string* match this regular
-      expression, return a corresponding :class:`MatchObject` instance.  Return
-      ``None`` if the string does not match the pattern; note that this is different
-      from a zero-length match.
-
-      .. note::
-
-         If you want to locate a match anywhere in *string*, use
-         :meth:`~RegexObject.search` instead.
+      Scan through *string* looking for a location where this regular expression
+      produces a match, and return a corresponding :class:`MatchObject` instance.
+      Return ``None`` if no position in the string matches the pattern; note that this
+      is different from finding a zero-length match at some point in the string.
 
       The optional second parameter *pos* gives an index in the string where the
       search is to start; it defaults to ``0``.  This is not completely equivalent to
@@ -732,24 +726,34 @@
       will be as if the string is *endpos* characters long, so only the characters
       from *pos* to ``endpos - 1`` will be searched for a match.  If *endpos* is less
       than *pos*, no match will be found, otherwise, if *rx* is a compiled regular
-      expression object, ``rx.match(string, 0, 50)`` is equivalent to
-      ``rx.match(string[:50], 0)``.
+      expression object, ``rx.search(string, 0, 50)`` is equivalent to
+      ``rx.search(string[:50], 0)``.
 
-         >>> pattern = re.compile("o")
-         >>> pattern.match("dog")      # No match as "o" is not at the start of "dog."
-         >>> pattern.match("dog", 1)   # Match as "o" is the 2nd character of "dog".
-         <_sre.SRE_Match object at ...>
+      >>> pattern = re.compile("d")
+      >>> pattern.search("dog")     # Match at index 0
+      <_sre.SRE_Match object at ...>
+      >>> pattern.search("dog", 1)  # No match; search doesn't include the "d"
 
 
-   .. method:: RegexObject.search(string[, pos[, endpos]])
+   .. method:: RegexObject.match(string[, pos[, endpos]])
 
-      Scan through *string* looking for a location where this regular expression
-      produces a match, and return a corresponding :class:`MatchObject` instance.
-      Return ``None`` if no position in the string matches the pattern; note that this
-      is different from finding a zero-length match at some point in the string.
+      If zero or more characters at the *beginning* of *string* match this regular
+      expression, return a corresponding :class:`MatchObject` instance.  Return
+      ``None`` if the string does not match the pattern; note that this is different
+      from a zero-length match.
 
       The optional *pos* and *endpos* parameters have the same meaning as for the
-      :meth:`~RegexObject.match` method.
+      :meth:`~RegexObject.search` method.
+
+      .. note::
+
+         If you want to locate a match anywhere in *string*, use
+         :meth:`~RegexObject.search` instead.
+
+      >>> pattern = re.compile("o")
+      >>> pattern.match("dog")      # No match as "o" is not at the start of "dog".
+      >>> pattern.match("dog", 1)   # Match as "o" is the 2nd character of "dog".
+      <_sre.SRE_Match object at ...>
 
 
    .. method:: RegexObject.split(string[, maxsplit=0])
@@ -759,12 +763,16 @@
 
    .. method:: RegexObject.findall(string[, pos[, endpos]])
 
-      Identical to the :func:`findall` function, using the compiled pattern.
+      Similar to the :func:`findall` function, using the compiled pattern, but
+      also accepts optional *pos* and *endpos* parameters that limit the search
+      region like for :meth:`match`.
 
 
    .. method:: RegexObject.finditer(string[, pos[, endpos]])
 
-      Identical to the :func:`finditer` function, using the compiled pattern.
+      Similar to the :func:`finditer` function, using the compiled pattern, but
+      also accepts optional *pos* and *endpos* parameters that limit the search
+      region like for :meth:`match`.
 
 
    .. method:: RegexObject.sub(repl, string[, count=0])

Modified: python/branches/py3k-dtoa/Doc/library/socket.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/library/socket.rst	(original)
+++ python/branches/py3k-dtoa/Doc/library/socket.rst	Tue Jul  6 17:55:27 2010
@@ -71,18 +71,21 @@
 tuple, and the fields depend on the address type. The general tuple form is
 ``(addr_type, v1, v2, v3 [, scope])``, where:
 
-   - *addr_type* is one of TIPC_ADDR_NAMESEQ, TIPC_ADDR_NAME, or
-     TIPC_ADDR_ID.
-   - *scope* is one of TIPC_ZONE_SCOPE, TIPC_CLUSTER_SCOPE, and
-     TIPC_NODE_SCOPE.
-   - If *addr_type* is TIPC_ADDR_NAME, then *v1* is the server type, *v2* is
-     the port identifier, and *v3* should be 0.
+- *addr_type* is one of TIPC_ADDR_NAMESEQ, TIPC_ADDR_NAME, or
+  TIPC_ADDR_ID.
+- *scope* is one of TIPC_ZONE_SCOPE, TIPC_CLUSTER_SCOPE, and
+  TIPC_NODE_SCOPE.
+- If *addr_type* is TIPC_ADDR_NAME, then *v1* is the server type, *v2* is
+  the port identifier, and *v3* should be 0.
 
-     If *addr_type* is TIPC_ADDR_NAMESEQ, then *v1* is the server type, *v2*
-     is the lower port number, and *v3* is the upper port number.
+  If *addr_type* is TIPC_ADDR_NAMESEQ, then *v1* is the server type, *v2*
+  is the lower port number, and *v3* is the upper port number.
 
-     If *addr_type* is TIPC_ADDR_ID, then *v1* is the node, *v2* is the
-     reference, and *v3* should be set to 0.
+  If *addr_type* is TIPC_ADDR_ID, then *v1* is the node, *v2* is the
+  reference, and *v3* should be set to 0.
+
+  If *addr_type* is TIPC_ADDR_ID, then *v1* is the node, *v2* is the
+  reference, and *v3* should be set to 0.
 
 
 All errors raise exceptions.  The normal exceptions for invalid argument types
@@ -690,7 +693,7 @@
 
    Set a timeout on blocking socket operations.  The *value* argument can be a
    nonnegative float expressing seconds, or ``None``. If a float is given,
-   subsequent socket operations will raise an :exc:`timeout` exception if the
+   subsequent socket operations will raise a :exc:`timeout` exception if the
    timeout period *value* has elapsed before the operation has completed.  Setting
    a timeout of ``None`` disables timeouts on socket operations.
    ``s.settimeout(0.0)`` is equivalent to ``s.setblocking(0)``;

Modified: python/branches/py3k-dtoa/Doc/library/ssl.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/library/ssl.rst	(original)
+++ python/branches/py3k-dtoa/Doc/library/ssl.rst	Tue Jul  6 17:55:27 2010
@@ -110,13 +110,16 @@
 
    .. note::
 
-      This information varies depending on the version of OpenSSL.
-      For instance, in some older versions of OpenSSL (such as 0.9.7l on
-      OS X 10.4), an SSLv2 client could not connect to an SSLv23 server.
-      Conversely, starting from 1.0.0, an SSLv23 client will actually
-      try the SSLv3 protocol unless you explicitly enable SSLv2 ciphers.
+      Which connections succeed will vary depending on the version of
+      OpenSSL.  For instance, in some older versions of OpenSSL (such
+      as 0.9.7l on OS X 10.4), an SSLv2 client could not connect to an
+      SSLv23 server.  Another example: beginning with OpenSSL 1.0.0,
+      an SSLv23 client will not actually attempt SSLv2 connections
+      unless you explicitly enable SSLv2 ciphers; for example, you
+      might specify ``"ALL"`` or ``"SSLv2"`` as the *ciphers* parameter
+      to enable them.
 
-   The parameter ``ciphers`` sets the available ciphers for this SSL object.
+   The *ciphers* parameter sets the available ciphers for this SSL object.
    It should be a string in the `OpenSSL cipher list format
    <http://www.openssl.org/docs/apps/ciphers.html#CIPHER_LIST_FORMAT>`_.
 

Modified: python/branches/py3k-dtoa/Doc/library/stdtypes.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/library/stdtypes.rst	(original)
+++ python/branches/py3k-dtoa/Doc/library/stdtypes.rst	Tue Jul  6 17:55:27 2010
@@ -452,7 +452,7 @@
     Equivalent to::
 
         def bit_length(self):
-            s = bin(x)          # binary representation:  bin(-37) --> '-0b100101'
+            s = bin(self)       # binary representation:  bin(-37) --> '-0b100101'
             s = s.lstrip('-0b') # remove leading zeros and minus sign
             return len(s)       # len('100101') --> 6
 
@@ -966,7 +966,8 @@
 
 .. method:: str.capitalize()
 
-   Return a copy of the string with only its first character capitalized.
+   Return a copy of the string with its first character capitalized and the
+   rest lowercased.
 
 
 .. method:: str.center(width[, fillchar])
@@ -1329,6 +1330,10 @@
    You can use :meth:`str.maketrans` to create a translation map from
    character-to-character mappings in different formats.
 
+   You can use the :func:`~string.maketrans` helper function in the :mod:`string`
+   module to create a translation table. For string objects, set the *table*
+   argument to ``None`` for translations that only delete characters:
+
    .. note::
 
       An even more flexible approach is to create a custom character mapping
@@ -2295,7 +2300,7 @@
       File "<stdin>", line 1, in <module>
       ValueError: cannot modify size of memoryview object
 
-   Notice how the size of the memoryview object can not be changed.
+   Notice how the size of the memoryview object cannot be changed.
 
 
    :class:`memoryview` has two methods:

Modified: python/branches/py3k-dtoa/Doc/library/string.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/library/string.rst	(original)
+++ python/branches/py3k-dtoa/Doc/library/string.rst	Tue Jul  6 17:55:27 2010
@@ -174,7 +174,7 @@
    .. method:: convert_field(value, conversion)
 
       Converts the value (returned by :meth:`get_field`) given a conversion type
-      (as in the tuple returned by the :meth:`parse` method.)  The default
+      (as in the tuple returned by the :meth:`parse` method).  The default
       version understands 'r' (repr) and 's' (str) conversion types.
 
 
@@ -185,7 +185,7 @@
 
 The :meth:`str.format` method and the :class:`Formatter` class share the same
 syntax for format strings (although in the case of :class:`Formatter`,
-subclasses can define their own format string syntax.)
+subclasses can define their own format string syntax).
 
 Format strings contain "replacement fields" surrounded by curly braces ``{}``.
 Anything that is not contained in braces is considered literal text, which is
@@ -197,7 +197,7 @@
    .. productionlist:: sf
       replacement_field: "{" [`field_name`] ["!" `conversion`] [":" `format_spec`] "}"
       field_name: arg_name ("." `attribute_name` | "[" `element_index` "]")*
-      arg_name: (`identifier` | `integer`)?
+      arg_name: [`identifier` | `integer`]
       attribute_name: `identifier`
       element_index: `integer` | `index_string`
       index_string: <any source character except "]"> +
@@ -211,6 +211,8 @@
 preceded by an exclamation point ``'!'``, and a *format_spec*, which is preceded
 by a colon ``':'``.  These specify a non-default format for the replacement value.
 
+See also the :ref:`formatspec` section.
+
 The *field_name* itself begins with an *arg_name* that is either either a number or a
 keyword.  If it's a number, it refers to a positional argument, and if it's a keyword,
 it refers to a named keyword argument.  If the numerical arg_names in a format string
@@ -221,6 +223,10 @@
 attribute using :func:`getattr`, while an expression of the form ``'[index]'``
 does an index lookup using :func:`__getitem__`.
 
+.. versionchanged:: 3.1
+   The positional argument specifiers can be omitted, so ``'{} {}'`` is
+   equivalent to ``'{0} {1}'``.
+
 Some simple format string examples::
 
    "First, thou shalt count to {0}" # References first positional argument
@@ -261,26 +267,7 @@
 format_spec are substituted before the *format_spec* string is interpreted.
 This allows the formatting of a value to be dynamically specified.
 
-For example, suppose you wanted to have a replacement field whose field width is
-determined by another variable::
-
-   "A man with two {0:{1}}".format("noses", 10)
-
-This would first evaluate the inner replacement field, making the format string
-effectively::
-
-   "A man with two {0:10}"
-
-Then the outer replacement field would be evaluated, producing::
-
-   "noses     "
-
-Which is substituted into the string, yielding::
-
-   "A man with two noses     "
-
-(The extra space is because we specified a field width of 10, and because left
-alignment is the default for strings.)
+See the :ref:`formatexamples` section for some examples.
 
 
 .. _formatspec:
@@ -290,7 +277,7 @@
 
 "Format specifications" are used within replacement fields contained within a
 format string to define how individual values are presented (see
-:ref:`formatstrings`.)  They can also be passed directly to the built-in
+:ref:`formatstrings`).  They can also be passed directly to the built-in
 :func:`format` function.  Each formattable type may define how the format
 specification is to be interpreted.
 
@@ -324,7 +311,7 @@
    | Option  | Meaning                                                  |
    +=========+==========================================================+
    | ``'<'`` | Forces the field to be left-aligned within the available |
-   |         | space (This is the default.)                             |
+   |         | space (this is the default).                             |
    +---------+----------------------------------------------------------+
    | ``'>'`` | Forces the field to be right-aligned within the          |
    |         | available space.                                         |
@@ -366,6 +353,9 @@
 For a locale aware separator, use the ``'n'`` integer presentation type
 instead.
 
+.. versionchanged:: 3.1
+   Added the ``','`` option (see also :pep:`378`).
+
 *width* is a decimal integer defining the minimum field width.  If not
 specified, then the field width will be determined by the content.
 
@@ -484,6 +474,147 @@
    +---------+----------------------------------------------------------+
 
 
+.. _formatexamples:
+
+Format examples
+^^^^^^^^^^^^^^^
+
+This section contains examples of the new format syntax and comparison with
+the old ``%``-formatting.
+
+In most of the cases the syntax is similar to the old ``%``-formatting, with the
+addition of the ``{}`` and with ``:`` used instead of ``%``.
+For example, ``'%03.2f'`` can be translated to ``'{:03.2f}'``.
+
+The new format syntax also supports new and different options, shown in the
+follow examples.
+
+Accessing arguments by position::
+
+   >>> '{0}, {1}, {2}'.format('a', 'b', 'c')
+   'a, b, c'
+   >>> '{}, {}, {}'.format('a', 'b', 'c')  # 3.1+ only
+   'a, b, c'
+   >>> '{2}, {1}, {0}'.format('a', 'b', 'c')
+   'c, b, a'
+   >>> '{2}, {1}, {0}'.format(*'abc')      # unpacking argument sequence
+   'c, b, a'
+   >>> '{0}{1}{0}'.format('abra', 'cad')   # arguments' indices can be repeated
+   'abracadabra'
+
+Accessing arguments by name::
+
+   >>> 'Coordinates: {latitude}, {longitude}'.format(latitude='37.24N', longitude='-115.81W')
+   'Coordinates: 37.24N, -115.81W'
+   >>> coord = {'latitude': '37.24N', 'longitude': '-115.81W'}
+   >>> 'Coordinates: {latitude}, {longitude}'.format(**coord)
+   'Coordinates: 37.24N, -115.81W'
+
+Accessing arguments' attributes::
+
+   >>> c = 3-5j
+   >>> ('The complex number {0} is formed from the real part {0.real} '
+   ...  'and the imaginary part {0.imag}.').format(c)
+   'The complex number (3-5j) is formed from the real part 3.0 and the imaginary part -5.0.'
+   >>> class Point:
+   ...     def __init__(self, x, y):
+   ...         self.x, self.y = x, y
+   ...     def __str__(self):
+   ...         return 'Point({self.x}, {self.y})'.format(self=self)
+   ...
+   >>> str(Point(4, 2))
+   'Point(4, 2)'
+
+Accessing arguments' items::
+
+   >>> coord = (3, 5)
+   >>> 'X: {0[0]};  Y: {0[1]}'.format(coord)
+   'X: 3;  Y: 5'
+
+Replacing ``%s`` and ``%r``::
+
+   >>> "repr() shows quotes: {!r}; str() doesn't: {!s}".format('test1', 'test2')
+   "repr() shows quotes: 'test1'; str() doesn't: test2"
+
+Aligning the text and specifying a width::
+
+   >>> '{:<30}'.format('left aligned')
+   'left aligned                  '
+   >>> '{:>30}'.format('right aligned')
+   '                 right aligned'
+   >>> '{:^30}'.format('centered')
+   '           centered           '
+   >>> '{:*^30}'.format('centered')  # use '*' as a fill char
+   '***********centered***********'
+
+Replacing ``%+f``, ``%-f``, and ``% f`` and specifying a sign::
+
+   >>> '{:+f}; {:+f}'.format(3.14, -3.14)  # show it always
+   '+3.140000; -3.140000'
+   >>> '{: f}; {: f}'.format(3.14, -3.14)  # show a space for positive numbers
+   ' 3.140000; -3.140000'
+   >>> '{:-f}; {:-f}'.format(3.14, -3.14)  # show only the minus -- same as '{:f}; {:f}'
+   '3.140000; -3.140000'
+
+Replacing ``%x`` and ``%o`` and converting the value to different bases::
+
+   >>> # format also supports binary numbers
+   >>> "int: {0:d};  hex: {0:x};  oct: {0:o};  bin: {0:b}".format(42)
+   'int: 42;  hex: 2a;  oct: 52;  bin: 101010'
+   >>> # with 0x, 0o, or 0b as prefix:
+   >>> "int: {0:d};  hex: {0:#x};  oct: {0:#o};  bin: {0:#b}".format(42)
+   'int: 42;  hex: 0x2a;  oct: 0o52;  bin: 0b101010'
+
+Using the comma as a thousands separator::
+
+   >>> '{:,}'.format(1234567890)
+   '1,234,567,890'
+
+Expressing a percentage::
+
+   >>> points = 19
+   >>> total = 22
+   >>> 'Correct answers: {:.2%}.'.format(points/total)
+   'Correct answers: 86.36%'
+
+Using type-specific formatting::
+
+   >>> import datetime
+   >>> d = datetime.datetime(2010, 7, 4, 12, 15, 58)
+   >>> '{:%Y-%m-%d %H:%M:%S}'.format(d)
+   '2010-07-04 12:15:58'
+
+Nesting arguments and more complex examples::
+
+   >>> for align, text in zip('<^>', ['left', 'center', 'right']):
+   ...     '{0:{align}{fill}16}'.format(text, fill=align, align=align)
+   ...
+   'left<<<<<<<<<<<<'
+   '^^^^^center^^^^^'
+   '>>>>>>>>>>>right'
+   >>>
+   >>> octets = [192, 168, 0, 1]
+   >>> '{:02X}{:02X}{:02X}{:02X}'.format(*octets)
+   'C0A80001'
+   >>> int(_, 16)
+   3232235521
+   >>>
+   >>> width = 5
+   >>> for num in range(5,12):
+   ...     for base in 'dXob':
+   ...         print('{0:{width}{base}}'.format(num, base=base, width=width), end=' ')
+   ...     print()
+   ...
+       5     5     5   101
+       6     6     6   110
+       7     7     7   111
+       8     8    10  1000
+       9     9    11  1001
+      10     A    12  1010
+      11     B    13  1011
+
+
+
 .. _template-strings:
 
 Template strings

Modified: python/branches/py3k-dtoa/Doc/library/struct.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/library/struct.rst	(original)
+++ python/branches/py3k-dtoa/Doc/library/struct.rst	Tue Jul  6 17:55:27 2010
@@ -151,7 +151,11 @@
 ^^^^^^^^^^^^^^^^^
 
 Format characters have the following meaning; the conversion between C and
-Python values should be obvious given their types:
+Python values should be obvious given their types.  The 'Standard size' column
+refers to the size of the packed value in bytes when using standard size; that
+is, when the format string starts with one of ``'<'``, ``'>'``, ``'!'`` or
+``'='``.  When using native size, the size of the packed value is
+platform-dependent.
 
 +--------+-------------------------+--------------------+----------------+------------+
 | Format | C Type                  | Python type        | Standard size  | Notes      |

Modified: python/branches/py3k-dtoa/Doc/library/sys.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/library/sys.rst	(original)
+++ python/branches/py3k-dtoa/Doc/library/sys.rst	Tue Jul  6 17:55:27 2010
@@ -223,44 +223,65 @@
 .. data:: float_info
 
    A structseq holding information about the float type. It contains low level
-   information about the precision and internal representation. Please study
-   your system's :file:`float.h` for more information.
-
-   +---------------------+--------------------------------------------------+
-   | attribute           |  explanation                                     |
-   +=====================+==================================================+
-   | :const:`epsilon`    | Difference between 1 and the next representable  |
-   |                     | floating point number                            |
-   +---------------------+--------------------------------------------------+
-   | :const:`dig`        | digits (see :file:`float.h`)                     |
-   +---------------------+--------------------------------------------------+
-   | :const:`mant_dig`   | mantissa digits (see :file:`float.h`)            |
-   +---------------------+--------------------------------------------------+
-   | :const:`max`        | maximum representable finite float               |
-   +---------------------+--------------------------------------------------+
-   | :const:`max_exp`    | maximum int e such that radix**(e-1) is in the   |
-   |                     | range of finite representable floats             |
-   +---------------------+--------------------------------------------------+
-   | :const:`max_10_exp` | maximum int e such that 10**e is in the          |
-   |                     | range of finite representable floats             |
-   +---------------------+--------------------------------------------------+
-   | :const:`min`        | Minimum positive normalizer float                |
-   +---------------------+--------------------------------------------------+
-   | :const:`min_exp`    | minimum int e such that radix**(e-1) is a        |
-   |                     | normalized float                                 |
-   +---------------------+--------------------------------------------------+
-   | :const:`min_10_exp` | minimum int e such that 10**e is a normalized    |
-   |                     | float                                            |
-   +---------------------+--------------------------------------------------+
-   | :const:`radix`      | radix of exponent                                |
-   +---------------------+--------------------------------------------------+
-   | :const:`rounds`     | addition rounds (see :file:`float.h`)            |
-   +---------------------+--------------------------------------------------+
-
-   .. note::
-
-      The information in the table is simplified.
+   information about the precision and internal representation.  The values
+   correspond to the various floating-point constants defined in the standard
+   header file :file:`float.h` for the 'C' programming language; see section
+   5.2.4.2.2 of the 1999 ISO/IEC C standard [C99]_, 'Characteristics of
+   floating types', for details.
+
+   +---------------------+----------------+--------------------------------------------------+
+   | attribute           | float.h macro  | explanation                                      |
+   +=====================+================+==================================================+
+   | :const:`epsilon`    | DBL_EPSILON    | difference between 1 and the least value greater |
+   |                     |                | than 1 that is representable as a float          |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`dig`        | DBL_DIG        | maximum number of decimal digits that can be     |
+   |                     |                | faithfully represented in a float;  see below    |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`mant_dig`   | DBL_MANT_DIG   | float precision: the number of base-``radix``    |
+   |                     |                | digits in the significand of a float             |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`max`        | DBL_MAX        | maximum representable finite float               |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`max_exp`    | DBL_MAX_EXP    | maximum integer e such that ``radix**(e-1)`` is  |
+   |                     |                | a representable finite float                     |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`max_10_exp` | DBL_MAX_10_EXP | maximum integer e such that ``10**e`` is in the  |
+   |                     |                | range of representable finite floats             |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`min`        | DBL_MIN        | minimum positive normalized float                |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`min_exp`    | DBL_MIN_EXP    | minimum integer e such that ``radix**(e-1)`` is  |
+   |                     |                | a normalized float                               |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`min_10_exp` | DBL_MIN_10_EXP | minimum integer e such that ``10**e`` is a       |
+   |                     |                | normalized float                                 |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`radix`      | FLT_RADIX      | radix of exponent representation                 |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`rounds`     | FLT_ROUNDS     | constant representing rounding mode              |
+   |                     |                | used for arithmetic operations                   |
+   +---------------------+----------------+--------------------------------------------------+
+
+   The attribute :attr:`sys.float_info.dig` needs further explanation.  If
+   ``s`` is any string representing a decimal number with at most
+   :attr:`sys.float_info.dig` significant digits, then converting ``s`` to a
+   float and back again will recover a string representing the same decimal
+   value::
 
+      >>> import sys
+      >>> sys.float_info.dig
+      15
+      >>> s = '3.14159265358979'    # decimal string with 15 significant digits
+      >>> format(float(s), '.15g')  # convert to float and back -> same value
+      '3.14159265358979'
+
+   But for strings with more than :attr:`sys.float_info.dig` significant digits,
+   this isn't always true::
+
+      >>> s = '9876543211234567'    # 16 significant digits is too many!
+      >>> format(float(s), '.16g')  # conversion changes value
+      '9876543211234568'
 
 .. data:: float_repr_style
 
@@ -957,3 +978,8 @@
    first three characters of :const:`version`.  It is provided in the :mod:`sys`
    module for informational purposes; modifying this value has no effect on the
    registry keys used by Python. Availability: Windows.
+
+.. rubric:: Citations
+
+.. [C99] ISO/IEC 9899:1999.  "Programming languages -- C."  A public draft of this standard is available at http://www.open-std.org/jtc1/sc22/wg14/www/docs/n1256.pdf .
+

Modified: python/branches/py3k-dtoa/Doc/library/sysconfig.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/library/sysconfig.rst	(original)
+++ python/branches/py3k-dtoa/Doc/library/sysconfig.rst	Tue Jul  6 17:55:27 2010
@@ -16,9 +16,9 @@
 Configuration variables
 -----------------------
 
-A Python distribution contains a :file:`Makefile` file and a :file:`python.h`
-that are necessary to build the Python binary itself, but also any C extension
-created in a third party project and compiled using :mod:`distutils`.
+A Python distribution contains a :file:`Makefile` and a :file:`pyconfig.h`
+header file that are necessary to build both the Python binary itself and
+third-party C extensions compiled using :mod:`distutils`.
 
 :mod:`sysconfig` puts all variables found in these files in a dictionary that
 can be accessed using :func:`get_config_vars` or :func:`get_config_var`.

Modified: python/branches/py3k-dtoa/Doc/library/unittest.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/library/unittest.rst	(original)
+++ python/branches/py3k-dtoa/Doc/library/unittest.rst	Tue Jul  6 17:55:27 2010
@@ -220,26 +220,26 @@
 
 unittest supports three command options.
 
-* -f / --failfast
+* :option:`-b` / :option:`--buffer`
 
-  Stop the test run on the first error or failure.
+  The standard output and standard error streams are buffered during the test
+  run. Output during a passing test is discarded. Output is echoed normally
+  on test fail or error and is added to the failure messages.
 
-* -c / --catch
+* :option:`-c` / :option:`--catch`
 
-  Control-c during the test run waits for the current test to end and then
-  reports all the results so far. A second control-c raises the normal
-  ``KeyboardInterrupt`` exception.
+  Control-C during the test run waits for the current test to end and then
+  reports all the results so far. A second control-C raises the normal
+  :exc:`KeyboardInterrupt` exception.
 
   See `Signal Handling`_ for the functions that provide this functionality.
 
-* -b / --buffer
+* :option:`-f` / :option:`--failfast`
 
-  The standard out and standard error streams are buffered during the test
-  run. Output during a passing test is discarded. Output is echoed normally
-  on test fail or error and is added to the failure messages.
+  Stop the test run on the first error or failure.
 
 .. versionadded:: 3.2
-   The command line options ``-c``, ``-b`` and ``-f`` where added.
+   The command line options ``-c``, ``-b`` and ``-f`` were added.
 
 The command line can also be used for test discovery, for running all of the
 tests in a project or just a subset.
@@ -270,8 +270,9 @@
    -t directory     Top level directory of project (default to
                     start directory)
 
-The -s, -p, & -t options can be passsed in as positional arguments. The
-following two command lines are equivalent::
+The :option:`-s`, :option:`-p`, and :option:`-t` options can be passed in
+as positional arguments in that order. The following two command lines
+are equivalent::
 
    python -m unittest discover -s project_directory -p '*_test.py'
    python -m unittest discover project_directory '*_test.py'
@@ -829,6 +830,11 @@
       compare equal, the test will fail with the explanation given by *msg*, or
       :const:`None`.
 
+      If *delta* is supplied instead of *places* then the difference
+      between *first* and *second* must be less than *delta*.
+
+      Supplying both *delta* and *places* raises a ``TypeError``.
+
       .. versionchanged:: 3.2
          Objects that compare equal are automatically almost equal.
          Added the ``delta`` keyword argument.
@@ -849,7 +855,7 @@
       compare equal, the test will fail with the explanation given by *msg*, or
       :const:`None`.
 
-      If *delta* is supplied instead of *places* then the the difference
+      If *delta* is supplied instead of *places* then the difference
       between *first* and *second* must be more than *delta*.
 
       Supplying both *delta* and *places* raises a ``TypeError``.
@@ -1902,12 +1908,12 @@
 Signal Handling
 ---------------
 
-The -c / --catch command line option to unittest, along with the ``catchbreak``
+The :option:`-c`/:option:`--catch` command line option to unittest, along with the ``catchbreak``
 parameter to :func:`unittest.main()`, provide more friendly handling of
-control-c during a test run. With catch break behavior enabled control-c will
+control-C during a test run. With catch break behavior enabled control-C will
 allow the currently running test to complete, and the test run will then end
 and report all the results so far. A second control-c will raise a
-``KeyboardInterrupt`` in the usual way.
+:exc:`KeyboardInterrupt` in the usual way.
 
 The control-c handling signal handler attempts to remain compatible with code or
 tests that install their own :const:`signal.SIGINT` handler. If the ``unittest``

Modified: python/branches/py3k-dtoa/Doc/library/urllib.parse.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/library/urllib.parse.rst	(original)
+++ python/branches/py3k-dtoa/Doc/library/urllib.parse.rst	Tue Jul  6 17:55:27 2010
@@ -310,23 +310,29 @@
    ``b'a&\xef'``.
 
 
-.. function:: urlencode(query, doseq=False)
+.. function:: urlencode(query, doseq=False, safe='', encoding=None, errors=None)
 
-   Convert a mapping object or a sequence of two-element tuples to a
-   "url-encoded" string, suitable to pass to :func:`urlopen` above as the
-   optional *data* argument.  This is useful to pass a dictionary of form
-   fields to a ``POST`` request.  The resulting string is a series of
-   ``key=value`` pairs separated by ``'&'`` characters, where both *key* and
-   *value* are quoted using :func:`quote_plus` above. When a sequence of
-   two-element tuples is used as the *query* argument, the first element of
-   each tuple is a key and the second is a value. The value element in itself
-   can be a sequence and in that case, if the optional parameter *doseq* is
-   evaluates to *True*, individual ``key=value`` pairs separated by ``'&'`` are
-   generated for each element of the value sequence for the key.  The order of
-   parameters in the encoded string will match the order of parameter tuples in
-   the sequence. This module provides the functions :func:`parse_qs` and
-   :func:`parse_qsl` which are used to parse query strings into Python data
-   structures.
+   Convert a mapping object or a sequence of two-element tuples, which may
+   either be a :class:`str` or a :class:`bytes`,  to a "url-encoded" string,
+   suitable to pass to :func:`urlopen` above as the optional *data* argument.
+   This is useful to pass a dictionary of form fields to a ``POST`` request.
+   The resulting string is a series of ``key=value`` pairs separated by ``'&'``
+   characters, where both *key* and *value* are quoted using :func:`quote_plus`
+   above. When a sequence of two-element tuples is used as the *query*
+   argument, the first element of each tuple is a key and the second is a
+   value. The value element in itself can be a sequence and in that case, if
+   the optional parameter *doseq* is evaluates to *True*, individual
+   ``key=value`` pairs separated by ``'&'`` are generated for each element of
+   the value sequence for the key.  The order of parameters in the encoded
+   string will match the order of parameter tuples in the sequence. This module
+   provides the functions :func:`parse_qs` and :func:`parse_qsl` which are used
+   to parse query strings into Python data structures.
+
+   When *query* parameter is a :class:`str`, the *safe*, *encoding* and *error*
+   parameters are sent the :func:`quote_plus` for encoding.
+
+   .. versionchanged:: 3.2
+      query paramater supports bytes and string.
 
 
 .. seealso::

Modified: python/branches/py3k-dtoa/Doc/library/urllib.request.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/library/urllib.request.rst	(original)
+++ python/branches/py3k-dtoa/Doc/library/urllib.request.rst	Tue Jul  6 17:55:27 2010
@@ -40,8 +40,8 @@
      commonly used to determine if a redirect was followed
 
    * :meth:`info` --- return the meta-information of the page, such as headers,
-     in the form of an :class:`http.client.HTTPMessage` instance (see `Quick
-     Reference to HTTP Headers <http://www.cs.tut.fi/~jkorpela/http.html>`_)
+     in the form of an :func:`email.message_from_string` instance (see
+     `Quick Reference to HTTP Headers <http://www.cs.tut.fi/~jkorpela/http.html>`_)
 
    Raises :exc:`URLError` on errors.
 

Modified: python/branches/py3k-dtoa/Doc/library/warnings.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/library/warnings.rst	(original)
+++ python/branches/py3k-dtoa/Doc/library/warnings.rst	Tue Jul  6 17:55:27 2010
@@ -57,7 +57,7 @@
 | :exc:`UserWarning`               | The default category for :func:`warn`.        |
 +----------------------------------+-----------------------------------------------+
 | :exc:`DeprecationWarning`        | Base category for warnings about deprecated   |
-|                                  | features.                                     |
+|                                  | features (ignored by default).                |
 +----------------------------------+-----------------------------------------------+
 | :exc:`SyntaxWarning`             | Base category for warnings about dubious      |
 |                                  | syntactic features.                           |
@@ -91,6 +91,9 @@
 standard warning categories.  A warning category must always be a subclass of
 the :exc:`Warning` class.
 
+.. versionchanged:: 3.2
+   :exc:`DeprecationWarning` is ignored by default.
+
 
 .. _warning-filter:
 
@@ -150,14 +153,6 @@
 :mod:`warnings` module parses these when it is first imported (invalid options
 are ignored, after printing a message to ``sys.stderr``).
 
-The warnings that are ignored by default may be enabled by passing :option:`-Wd`
-to the interpreter. This enables default handling for all warnings, including
-those that are normally ignored by default. This is particular useful for
-enabling ImportWarning when debugging problems importing a developed package.
-ImportWarning can also be enabled explicitly in Python code using::
-
-   warnings.simplefilter('default', ImportWarning)
-
 
 .. _warning-suppress:
 
@@ -233,6 +228,37 @@
 entries from the warnings list before each new operation).
 
 
+Updating Code For New Versions of Python
+----------------------------------------
+
+Warnings that are only of interest to the developer are ignored by default. As
+such you should make sure to test your code with typically ignored warnings
+made visible. You can do this from the command-line by passing :option:`-Wd`
+to the interpreter (this is shorthand for :option:`-W default`).  This enables
+default handling for all warnings, including those that are ignored by default.
+To change what action is taken for encountered warnings you simply change what
+argument is passed to :option:`-W`, e.g. :option:`-W error`. See the
+:option:`-W` flag for more details on what is possible.
+
+To programmatically do the same as :option:`-Wd`, use::
+
+  warnings.simplefilter('default')
+
+Make sure to execute this code as soon as possible. This prevents the
+registering of what warnings have been raised from unexpectedly influencing how
+future warnings are treated.
+
+Having certain warnings ignored by default is done to prevent a user from
+seeing warnings that are only of interest to the developer. As you do not
+necessarily have control over what interpreter a user uses to run their code,
+it is possible that a new version of Python will be released between your
+release cycles.  The new interpreter release could trigger new warnings in your
+code that were not there in an older interpreter, e.g.
+:exc:`DeprecationWarning` for a module that you are using. While you as a
+developer want to be notified that your code is using a deprecated module, to a
+user this information is essentially noise and provides no benefit to them.
+
+
 .. _warning-functions:
 
 Available Functions

Modified: python/branches/py3k-dtoa/Doc/reference/compound_stmts.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/reference/compound_stmts.rst	(original)
+++ python/branches/py3k-dtoa/Doc/reference/compound_stmts.rst	Tue Jul  6 17:55:27 2010
@@ -272,14 +272,12 @@
        try:
            foo
        finally:
-           N = None
            del N
 
-That means that you have to assign the exception to a different name if you want
-to be able to refer to it after the except clause.  The reason for this is that
-with the traceback attached to them, exceptions will form a reference cycle with
-the stack frame, keeping all locals in that frame alive until the next garbage
-collection occurs.
+This means the exception must be assigned to a different name to be able to
+refer to it after the except clause.  Exceptions are cleared because with the
+traceback attached to them, they form a reference cycle with the stack frame,
+keeping all locals in that frame alive until the next garbage collection occurs.
 
 .. index::
    module: sys

Modified: python/branches/py3k-dtoa/Doc/reference/datamodel.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/reference/datamodel.rst	(original)
+++ python/branches/py3k-dtoa/Doc/reference/datamodel.rst	Tue Jul  6 17:55:27 2010
@@ -1595,7 +1595,7 @@
 
 In particular, the metaclass :class:`abc.ABCMeta` implements these methods in
 order to allow the addition of Abstract Base Classes (ABCs) as "virtual base
-classes" to any class or type (including built-in types), and including to other
+classes" to any class or type (including built-in types), including other
 ABCs.
 
 .. method:: class.__instancecheck__(self, instance)
@@ -1614,7 +1614,7 @@
 
 Note that these methods are looked up on the type (metaclass) of a class.  They
 cannot be defined as class methods in the actual class.  This is consistent with
-the lookup of special methods that are called on instances, only that in this
+the lookup of special methods that are called on instances, only in this
 case the instance is itself a class.
 
 .. seealso::

Modified: python/branches/py3k-dtoa/Doc/reference/introduction.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/reference/introduction.rst	(original)
+++ python/branches/py3k-dtoa/Doc/reference/introduction.rst	Tue Jul  6 17:55:27 2010
@@ -69,12 +69,12 @@
    more information, see `the IronPython website <http://www.ironpython.com/>`_.
 
 PyPy
-   An implementation of Python written in Python; even the bytecode interpreter is
-   written in Python.  This is executed using CPython as the underlying
-   interpreter.  One of the goals of the project is to encourage experimentation
-   with the language itself by making it easier to modify the interpreter (since it
-   is written in Python).  Additional information is available on `the PyPy
-   project's home page <http://codespeak.net/pypy/>`_.
+   An implementation of Python written completely in Python. It supports several
+   advanced features not found in other implementations like stackless support
+   and a Just in Time compiler. One of the goals of the project is to encourage
+   experimentation with the language itself by making it easier to modify the
+   interpreter (since it is written in Python).  Additional information is
+   available on `the PyPy project's home page <http://pypy.org/>`_.
 
 Each of these implementations varies in some way from the language as documented
 in this manual, or introduces specific information beyond what's covered in the

Modified: python/branches/py3k-dtoa/Doc/reference/lexical_analysis.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/reference/lexical_analysis.rst	(original)
+++ python/branches/py3k-dtoa/Doc/reference/lexical_analysis.rst	Tue Jul  6 17:55:27 2010
@@ -407,7 +407,7 @@
 
 .. productionlist::
    bytesliteral: `bytesprefix`(`shortbytes` | `longbytes`)
-   bytesprefix: "b" | "B"
+   bytesprefix: "b" | "B" | "br" | "Br" | "bR" | "BR"
    shortbytes: "'" `shortbytesitem`* "'" | '"' `shortbytesitem`* '"'
    longbytes: "'''" `longbytesitem`* "'''" | '"""' `longbytesitem`* '"""'
    shortbytesitem: `shortbyteschar` | `bytesescapeseq`
@@ -431,16 +431,16 @@
 characters that otherwise have a special meaning, such as newline, backslash
 itself, or the quote character.
 
-String literals may optionally be prefixed with a letter ``'r'`` or ``'R'``;
-such strings are called :dfn:`raw strings` and treat backslashes as literal
-characters.  As a result, ``'\U'`` and ``'\u'`` escapes in raw strings are not
-treated specially.
-
 Bytes literals are always prefixed with ``'b'`` or ``'B'``; they produce an
 instance of the :class:`bytes` type instead of the :class:`str` type.  They
 may only contain ASCII characters; bytes with a numeric value of 128 or greater
 must be expressed with escapes.
 
+Both string and bytes literals may optionally be prefixed with a letter ``'r'``
+or ``'R'``; such strings are called :dfn:`raw strings` and treat backslashes as
+literal characters.  As a result, in string literals, ``'\U'`` and ``'\u'``
+escapes in raw strings are not treated specially.
+
 In triple-quoted strings, unescaped newlines and quotes are allowed (and are
 retained), except that three unescaped quotes in a row terminate the string.  (A
 "quote" is the character used to open the string, i.e. either ``'`` or ``"``.)
@@ -545,9 +545,9 @@
 String literal concatenation
 ----------------------------
 
-Multiple adjacent string literals (delimited by whitespace), possibly using
-different quoting conventions, are allowed, and their meaning is the same as
-their concatenation.  Thus, ``"hello" 'world'`` is equivalent to
+Multiple adjacent string or bytes literals (delimited by whitespace), possibly
+using different quoting conventions, are allowed, and their meaning is the same
+as their concatenation.  Thus, ``"hello" 'world'`` is equivalent to
 ``"helloworld"``.  This feature can be used to reduce the number of backslashes
 needed, to split long strings conveniently across long lines, or even to add
 comments to parts of strings, for example::

Modified: python/branches/py3k-dtoa/Doc/tools/sphinxext/pyspecific.py
==============================================================================
--- python/branches/py3k-dtoa/Doc/tools/sphinxext/pyspecific.py	(original)
+++ python/branches/py3k-dtoa/Doc/tools/sphinxext/pyspecific.py	Tue Jul  6 17:55:27 2010
@@ -5,7 +5,7 @@
 
     Sphinx extension with Python doc-specific markup.
 
-    :copyright: 2008, 2009 by Georg Brandl.
+    :copyright: 2008, 2009, 2010 by Georg Brandl.
     :license: Python license.
 """
 
@@ -149,7 +149,7 @@
 import re
 from sphinx import addnodes
 
-opcode_sig_re = re.compile(r'(\w+(?:\+\d)?)\s*\((.*)\)')
+opcode_sig_re = re.compile(r'(\w+(?:\+\d)?)(?:\s*\((.*)\))?')
 
 def parse_opcode_signature(env, sig, signode):
     """Transform an opcode signature into RST nodes."""
@@ -158,9 +158,10 @@
         raise ValueError
     opname, arglist = m.groups()
     signode += addnodes.desc_name(opname, opname)
-    paramlist = addnodes.desc_parameterlist()
-    signode += paramlist
-    paramlist += addnodes.desc_parameter(arglist, arglist)
+    if arglist is not None:
+        paramlist = addnodes.desc_parameterlist()
+        signode += paramlist
+        paramlist += addnodes.desc_parameter(arglist, arglist)
     return opname.strip()
 
 

Modified: python/branches/py3k-dtoa/Doc/tutorial/classes.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/tutorial/classes.rst	(original)
+++ python/branches/py3k-dtoa/Doc/tutorial/classes.rst	Tue Jul  6 17:55:27 2010
@@ -580,7 +580,7 @@
 =================
 
 "Private" instance variables that cannot be accessed except from inside an
-object, don't exist in Python.  However, there is a convention that is followed
+object don't exist in Python.  However, there is a convention that is followed
 by most Python code: a name prefixed with an underscore (e.g. ``_spam``) should
 be treated as a non-public part of the API (whether it is a function, a method
 or a data member).  It should be considered an implementation detail and subject

Modified: python/branches/py3k-dtoa/Doc/tutorial/controlflow.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/tutorial/controlflow.rst	(original)
+++ python/branches/py3k-dtoa/Doc/tutorial/controlflow.rst	Tue Jul  6 17:55:27 2010
@@ -87,7 +87,6 @@
 If you do need to iterate over a sequence of numbers, the built-in function
 :func:`range` comes in handy.  It generates arithmetic progressions::
 
-
     >>> for i in range(5):
     ...     print(i)
     ...
@@ -97,9 +96,7 @@
     3
     4
 
-
-
-The given end point is never part of the generated list; ``range(10)`` generates
+The given end point is never part of the generated sequence; ``range(10)`` generates
 10 values, the legal indices for items of a sequence of length 10.  It
 is possible to let the range start at another number, or to specify a different
 increment (even negative; sometimes this is called the 'step')::

Modified: python/branches/py3k-dtoa/Doc/tutorial/introduction.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/tutorial/introduction.rst	(original)
+++ python/branches/py3k-dtoa/Doc/tutorial/introduction.rst	Tue Jul  6 17:55:27 2010
@@ -208,7 +208,7 @@
 
    print(hello)
 
-Note that newlines still need to be embedded in the string using ``\n``; the
+Note that newlines still need to be embedded in the string using ``\n`` -- the
 newline following the trailing backslash is discarded.  This example would print
 the following:
 

Modified: python/branches/py3k-dtoa/Doc/using/mac.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/using/mac.rst	(original)
+++ python/branches/py3k-dtoa/Doc/using/mac.rst	Tue Jul  6 17:55:27 2010
@@ -114,8 +114,8 @@
 =======
 
 MacPython ships with the standard IDLE development environment. A good
-introduction to using IDLE can be found at http://hkn.eecs.berkeley.edu/
-dyoo/python/idle_intro/index.html.
+introduction to using IDLE can be found at
+http://hkn.eecs.berkeley.edu/~dyoo/python/idle_intro/index.html.
 
 
 .. _mac-package-manager:

Modified: python/branches/py3k-dtoa/Doc/whatsnew/2.7.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/whatsnew/2.7.rst	(original)
+++ python/branches/py3k-dtoa/Doc/whatsnew/2.7.rst	Tue Jul  6 17:55:27 2010
@@ -6,12 +6,11 @@
 :Release: |release|
 :Date: |today|
 
-.. Fix accents on Kristjan Valur Jonsson, Fuerstenau
-
-.. Big jobs: ElementTree 1.3, pep 391, sysconfig
-..  unittest test discovery
 ..  hyperlink all the methods & functions.
 
+.. T_STRING_INPLACE not described in main docs
+.. "Format String Syntax" in string.rst could use many more examples.
+
 .. $Id$
    Rules for maintenance:
 
@@ -54,20 +53,77 @@
    when researching a change.
 
 This article explains the new features in Python 2.7.  The final
-release of 2.7 is currently scheduled for June 2010; the detailed
+release of 2.7 is currently scheduled for July 2010; the detailed
 schedule is described in :pep:`373`.
 
-Python 2.7 is planned to be the last major release in the 2.x series.
-Though more major releases have not been absolutely ruled out, the
-Python maintainers are planning to focus more on Python 3.x.  Despite
-that, it's likely that the 2.7 release will have a longer period of
-maintenance compared to earlier 2.x versions.
-
-.. Compare with previous release in 2 - 3 sentences here.
-   add hyperlink when the documentation becomes available online.
+Numeric handling has been improved in many ways, for both
+floating-point numbers and for the :class:`Decimal` class.  There are
+some useful additions to the standard library, such as a greatly
+enhanced :mod:`unittest` module, the :mod:`argparse` module for
+parsing command-line options, convenient ordered-dictionary and
+:class:`Counter` classes in the :mod:`collections` module, and many
+other improvements.
+
+Python 2.7 is planned to be the last of the 2.x releases, so we worked
+on making it a good release for the long term.  To help with porting
+to Python 3, several new features from the Python 3.x series have been
+included in 2.7.
+
+This article doesn't attempt to provide a complete specification of
+the new features, but instead provides a convenient overview.  For
+full details, you should refer to the documentation for Python 2.7 at
+http://docs.python.org. If you want to understand the rationale for
+the design and implementation, refer to the PEP for a particular new
+feature or the issue on http://bugs.python.org in which a change was
+discussed.  Whenever possible, "What's New in Python" links to the
+bug/patch item for each change.
 
 .. _whatsnew27-python31:
 
+The Future for Python 2.x
+=========================
+
+Python 2.7 is intended to be the last major release in the 2.x series.
+The Python maintainers are planning to focus their future efforts on
+the Python 3.x series.
+
+This means that 2.7 will remain in place for a long time, running
+production systems that have not been ported to Python 3.x.
+Two consequences of the long-term significance of 2.7 are:
+
+* It's very likely the 2.7 release will have a longer period of
+  maintenance compared to earlier 2.x versions.  Python 2.7 will
+  continue to be maintained while the transition to 3.x continues, and
+  the developers are planning to support Python 2.7 with bug-fix
+  releases beyond the typical two years.
+
+* A policy decision was made to silence warnings only of interest to
+  developers.  :exc:`DeprecationWarning` and its
+  descendants are now ignored unless otherwise requested, preventing
+  users from seeing warnings triggered by an application.  This change
+  was also made in the branch that will become Python 3.2. (Discussed
+  on stdlib-sig and carried out in :issue:`7319`.)
+
+  In previous releases, :exc:`DeprecationWarning` messages were
+  enabled by default, providing Python developers with a clear
+  indication of where their code may break in a future major version
+  of Python.
+
+  However, there are increasingly many users of Python-based
+  applications who are not directly involved in the development of
+  those applications.  :exc:`DeprecationWarning` messages are
+  irrelevant to such users, making them worry about an application
+  that's actually working correctly and burdening application developers
+  with responding to these concerns.
+
+  You can re-enable display of :exc:`DeprecationWarning` messages by
+  running Python with the :option:`-Wdefault` (short form:
+  :option:`-Wd`) switch, or by setting the :envvar:`PYTHONWARNINGS`
+  environment variable to ``"default"`` (or ``"d"``) before running
+  Python.  Python code can also re-enable them
+  by calling ``warnings.simplefilter('default')``.
+
+
 Python 3.1 Features
 =======================
 
@@ -78,25 +134,31 @@
 
 A partial list of 3.1 features that were backported to 2.7:
 
-* A version of the :mod:`io` library, rewritten in C for performance.
+* The syntax for set literals (``{1,2,3}`` is a mutable set).
+* Dictionary and set comprehensions (``{ i: i*2 for i in range(3)}``).
+* Multiple context managers in a single :keyword:`with` statement.
+* A new version of the :mod:`io` library, rewritten in C for performance.
 * The ordered-dictionary type described in :ref:`pep-0372`.
-* The new format specifier described in :ref:`pep-0378`.
+* The new ``","`` format specifier described in :ref:`pep-0378`.
 * The :class:`memoryview` object.
-* A small subset of the :mod:`importlib` module `described below <#importlib-section>`__.
+* A small subset of the :mod:`importlib` module,
+  `described below <#importlib-section>`__.
 * Float-to-string and string-to-float conversions now round their
-  results more correctly.  And :func:`repr` of a floating-point
+  results more correctly, and :func:`repr` of a floating-point
   number *x* returns a result that's guaranteed to round back to the
   same number when converted back to a string.
+* The :ctype:`PyCapsule` type, used to provide a C API for extension modules.
 * The :cfunc:`PyLong_AsLongAndOverflow` C API function.
 
-One porting change: the :option:`-3` switch now automatically
-enables the :option:`-Qwarn` switch that causes warnings
-about using classic division with integers and long integers.
-
 Other new Python3-mode warnings include:
 
 * :func:`operator.isCallable` and :func:`operator.sequenceIncludes`,
-  which are not supported in 3.x.
+  which are not supported in 3.x, now trigger warnings.
+* The :option:`-3` switch now automatically
+  enables the :option:`-Qwarn` switch that causes warnings
+  about using classic division with integers and long integers.
+
+
 
 .. ========================================================================
 .. Large, PEP-level features and changes should be described here.
@@ -104,22 +166,22 @@
 
 .. _pep-0372:
 
-PEP 372: Adding an ordered dictionary to collections
+PEP 372: Adding an Ordered Dictionary to collections
 ====================================================
 
 Regular Python dictionaries iterate over key/value pairs in arbitrary order.
 Over the years, a number of authors have written alternative implementations
 that remember the order that the keys were originally inserted.  Based on
-the experiences from those implementations, a new
-:class:`~collections.OrderedDict` class has been introduced in the
-:mod:`collections` module.
+the experiences from those implementations, 2.7 introduces a new
+:class:`~collections.OrderedDict` class in the :mod:`collections` module.
 
-The :class:`~collections.OrderedDict` API is substantially the same as regular
-dictionaries but will iterate over keys and values in a guaranteed order
+The :class:`~collections.OrderedDict` API provides the same interface as regular
+dictionaries but iterates over keys and values in a guaranteed order
 depending on when a key was first inserted::
 
     >>> from collections import OrderedDict
-    >>> d = OrderedDict([('first', 1), ('second', 2),
+    >>> d = OrderedDict([('first', 1),
+    ...                  ('second', 2),
     ...                  ('third', 3)])
     >>> d.items()
     [('first', 1), ('second', 2), ('third', 3)]
@@ -156,9 +218,11 @@
 Comparing two ordered dictionaries checks both the keys and values,
 and requires that the insertion order was the same::
 
-    >>> od1 = OrderedDict([('first', 1), ('second', 2),
+    >>> od1 = OrderedDict([('first', 1),
+    ...                    ('second', 2),
     ...                    ('third', 3)])
-    >>> od2 = OrderedDict([('third', 3), ('first', 1),
+    >>> od2 = OrderedDict([('third', 3),
+    ...                    ('first', 1),
     ...                    ('second', 2)])
     >>> od1 == od2
     False
@@ -176,17 +240,12 @@
 deletion doesn't have to traverse the entire linked list and therefore
 remains O(1).
 
-.. XXX check O(1)-ness with Raymond
-..     Also check if the 'somenamedtuple' in the collection module should
-..     be replaced/removed in order to use
-..     :meth:`~collections.namedtuple._asdict()` (see below)
-
 The standard library now supports use of ordered dictionaries in several
 modules.
 
-* The :mod:`ConfigParser` module uses them by default, letting
-  configuration files be read, modified, and then written back in their original
-  order.
+* The :mod:`ConfigParser` module uses them by default, meaning that
+  configuration files can now read, modified, and then written back
+  in their original order.
 
 * The :meth:`~collections.somenamedtuple._asdict()` method for
   :func:`collections.namedtuple` now returns an ordered dictionary with the
@@ -210,7 +269,7 @@
 =================================================
 
 To make program output more readable, it can be useful to add
-separators to large numbers and render them as
+separators to large numbers, rendering them as
 18,446,744,073,709,551,616 instead of 18446744073709551616.
 
 The fully general solution for doing this is the :mod:`locale` module,
@@ -237,8 +296,6 @@
 comma-formatting mechanism isn't as general as the :mod:`locale`
 module, but it's easier to use.
 
-.. XXX "Format String Syntax" in string.rst could use many more examples.
-
 .. seealso::
 
    :pep:`378` - Format Specifier for Thousands Separator
@@ -248,13 +305,13 @@
 ======================================================
 
 The :mod:`argparse` module for parsing command-line arguments was
-added, intended as a more powerful replacement for the
+added as a more powerful replacement for the
 :mod:`optparse` module.
 
 This means Python now supports three different modules for parsing
 command-line arguments: :mod:`getopt`, :mod:`optparse`, and
 :mod:`argparse`.  The :mod:`getopt` module closely resembles the C
-:cfunc:`getopt` function, so it remains useful if you're writing a
+library's :cfunc:`getopt` function, so it remains useful if you're writing a
 Python prototype that will eventually be rewritten in C.
 :mod:`optparse` becomes redundant, but there are no plans to remove it
 because there are many scripts still using it, and there's no
@@ -306,23 +363,28 @@
       -o FILE     direct output to FILE instead of stdout
       -C NUM      display NUM lines of added context
 
-Similarly to :mod:`optparse`, the command-line switches and arguments
+As with :mod:`optparse`, the command-line switches and arguments
 are returned as an object with attributes named by the *dest* parameters::
 
     -> ./python.exe argparse-example.py -v
-    {'output': None, 'is_verbose': True, 'context': 0, 'inputs': []}
+    {'output': None,
+     'is_verbose': True,
+     'context': 0,
+     'inputs': []}
 
     -> ./python.exe argparse-example.py -v -o /tmp/output -C 4 file1 file2
-    {'output': '/tmp/output', 'is_verbose': True, 'context': 4,
+    {'output': '/tmp/output',
+     'is_verbose': True,
+     'context': 4,
      'inputs': ['file1', 'file2']}
 
 :mod:`argparse` has much fancier validation than :mod:`optparse`; you
 can specify an exact number of arguments as an integer, 0 or more
 arguments by passing ``'*'``, 1 or more by passing ``'+'``, or an
 optional argument with ``'?'``.  A top-level parser can contain
-sub-parsers, so you can define subcommands that have different sets of
+sub-parsers to define subcommands that have different sets of
 switches, as in ``svn commit``, ``svn checkout``, etc.  You can
-specify an argument type as :class:`~argparse.FileType`, which will
+specify an argument's type as :class:`~argparse.FileType`, which will
 automatically open files for you and understands that ``'-'`` means
 standard input or output.
 
@@ -331,6 +393,8 @@
    `argparse module documentation <http://docs.python.org/dev/library/argparse.html>`__
 
    `Upgrading optparse code to use argparse <http://docs.python.org/dev/library/argparse.html#upgrading-optparse-code>`__
+     Part of the Python documentation, describing how to convert
+     code that uses :mod:`optparse`.
 
    :pep:`389` - argparse - New Command Line Parsing Module
      PEP written and implemented by Steven Bethard.
@@ -338,34 +402,88 @@
 PEP 391: Dictionary-Based Configuration For Logging
 ====================================================
 
-.. not documented in library reference yet.
+.. XXX not documented in library reference yet; add link here once it's added.
 
-The :mod:`logging` module is very flexible; an application can define
+The :mod:`logging` module is very flexible; applications can define
 a tree of logging subsystems, and each logger in this tree can filter
 out certain messages, format them differently, and direct messages to
 a varying number of handlers.
 
 All this flexibility can require a lot of configuration.  You can
 write Python statements to create objects and set their properties,
-but a complex set-up would require verbose but boring code.
+but a complex set-up requires verbose but boring code.
 :mod:`logging` also supports a :func:`~logging.config.fileConfig`
 function that parses a file, but the file format doesn't support
 configuring filters, and it's messier to generate programmatically.
 
 Python 2.7 adds a :func:`~logging.config.dictConfig` function that
-uses a dictionary, and there are many ways to produce a dictionary
-from different sources.  You can construct one with code, of course.
-Python's standard library now includes a JSON parser, so you could
-parse a file containing JSON, or you could use a YAML parsing library
-if one is installed.
+uses a dictionary to configure logging.  There are many ways to
+produce a dictionary from different sources: construct one with code;
+parse a file containing JSON; or use a YAML parsing library if one is
+installed.
+
+The following example configures two loggers, the root logger and a
+logger named "network".   Messages sent to the root logger will be
+sent to the system log using the syslog protocol, and messages
+to the "network" logger will be written to a :file:`network.log` file
+that will be rotated once the log reaches 1Mb.
+
+::
+
+    import logging
+    import logging.config
 
-XXX describe an example.
+    configdict = {
+     'version': 1,    # Configuration schema in use; must be 1 for now
+     'formatters': {
+         'standard': {
+             'format': ('%(asctime)s %(name)-15s '
+                        '%(levelname)-8s %(message)s')}},
+
+     'handlers': {'netlog': {'backupCount': 10,
+                         'class': 'logging.handlers.RotatingFileHandler',
+                         'filename': '/logs/network.log',
+                         'formatter': 'standard',
+                         'level': 'INFO',
+                         'maxBytes': 1024*1024},
+                  'syslog': {'class': 'logging.handlers.SysLogHandler',
+                             'formatter': 'standard',
+                             'level': 'ERROR'}},
+
+     # Specify all the subordinate loggers
+     'loggers': {
+                 'network': {
+                             'handlers': ['netlog']
+                 }
+     },
+     # Specify properties of the root logger
+     'root': {
+              'handlers': ['syslog']
+     },
+    }
+
+    # Set up configuration
+    logging.config.dictConfig(configdict)
+
+    # As an example, log two error messages
+    logger = logging.getLogger('/')
+    logger.error('Database not found')
 
-Two smaller enhancements to the logging module are:
+    netlogger = logging.getLogger('network')
+    netlogger.error('Connection failed')
+
+Three smaller enhancements to the :mod:`logging` module, all
+implemented by Vinay Sajip, are:
 
 .. rev79293
 
-* :class:`Logger` instances gained a :meth:`getChild` that retrieves a
+* The :class:`~logging.handlers.SysLogHandler` class now supports
+  syslogging over TCP.  The constructor has a *socktype* parameter
+  giving the type of socket to use, either :const:`socket.SOCK_DGRAM`
+  for UDP or :const:`socket.SOCK_STREAM` for TCP.  The default
+  protocol remains UDP.
+
+* :class:`Logger` instances gained a :meth:`getChild` method that retrieves a
   descendant logger using a relative path.  For example,
   once you retrieve a logger by doing ``log = getLogger('app')``,
   calling ``log.getChild('network.listen')`` is equivalent to
@@ -387,12 +505,10 @@
 are different in Python 3.x.  They return an object called a :dfn:`view`
 instead of a fully materialized list.
 
-.. Views can be iterated over, but they also behave like sets.  XXX not working.
-
 It's not possible to change the return values of :meth:`keys`,
 :meth:`values`, and :meth:`items` in Python 2.7 because too much code
 would break.  Instead the 3.x versions were added under the new names
-of :meth:`viewkeys`, :meth:`viewvalues`, and :meth:`viewitems`.
+:meth:`viewkeys`, :meth:`viewvalues`, and :meth:`viewitems`.
 
 ::
 
@@ -402,6 +518,17 @@
     >>> d.viewkeys()
     dict_keys([0, 130, 10, 140, 20, 150, 30, ..., 250])
 
+Views can be iterated over, but the key and item views also behave
+like sets.  The ``&`` operator performs intersection, and ``|``
+performs a union::
+
+    >>> d1 = dict((i*10, chr(65+i)) for i in range(26))
+    >>> d2 = dict((i**.5, i) for i in range(1000))
+    >>> d1.viewkeys() & d2.viewkeys()
+    set([0.0, 10.0, 20.0, 30.0])
+    >>> d1.viewkeys() | range(0, 30)
+    set([0, 1, 130, 3, 4, 5, 6, ..., 120, 250])
+
 The view keeps track of the dictionary and its contents change as the
 dictionary is modified::
 
@@ -433,6 +560,58 @@
      Backported to 2.7 by Alexandre Vassalotti; :issue:`1967`.
 
 
+PEP 3137: The memoryview Object
+====================================================
+
+The :class:`memoryview` object provides a view of another object's
+memory content that matches the :class:`bytes` type's interface.
+
+    >>> import string
+    >>> m = memoryview(string.letters)
+    >>> m
+    <memory at 0x37f850>
+    >>> len(m)           # Returns length of underlying object
+    52
+    >>> m[0], m[25], m[26]   # Indexing returns one byte
+    ('a', 'z', 'A')
+    >>> m2 = m[0:26]         # Slicing returns another memoryview
+    >>> m2
+    <memory at 0x37f080>
+
+The content of the view can be converted to a string of bytes or
+a list of integers:
+
+    >>> m2.tobytes()
+    'abcdefghijklmnopqrstuvwxyz'
+    >>> m2.tolist()
+    [97, 98, 99, 100, 101, 102, 103, ... 121, 122]
+    >>>
+
+:class:`memoryview` objects allow modifying the underlying object if
+it's a mutable object.
+
+    >>> m2[0] = 75
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    TypeError: cannot modify read-only memory
+    >>> b = bytearray(string.letters)  # Creating a mutable object
+    >>> b
+    bytearray(b'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ')
+    >>> mb = memoryview(b)
+    >>> mb[0] = '*'         # Assign to view, changing the bytearray.
+    >>> b[0:5]              # The bytearray has been changed.
+    bytearray(b'*bcde')
+    >>>
+
+.. seealso::
+
+   :pep:`3137` - Immutable Bytes and Mutable Buffer
+     PEP written by Guido van Rossum.
+     Implemented by Travis Oliphant, Antoine Pitrou and others.
+     Backported to 2.7 by Antoine Pitrou; :issue:`2396`.
+
+
+
 Other Language Changes
 ======================
 
@@ -458,9 +637,9 @@
   3.x, generalizing list/generator comprehensions to use
   the literal syntax for sets and dictionaries.
 
-    >>> {x:x*x for x in range(6)}
+    >>> {x: x*x for x in range(6)}
     {0: 0, 1: 1, 2: 4, 3: 9, 4: 16, 5: 25}
-    >>> {'a'*x for x in range(6)}
+    >>> {('a'*x) for x in range(6)}
     set(['', 'a', 'aa', 'aaa', 'aaaa', 'aaaaa'])
 
   Backported by Alexandre Vassalotti; :issue:`2333`.
@@ -490,8 +669,8 @@
   in many different places: :func:`str` on
   floats and complex numbers; the :class:`float` and :class:`complex`
   constructors;
-  numeric formatting; serialization and
-  deserialization of floats and complex numbers using the
+  numeric formatting; serializing and
+  deserializing floats and complex numbers using the
   :mod:`marshal`, :mod:`pickle`
   and :mod:`json` modules;
   parsing of float and imaginary literals in Python code;
@@ -506,7 +685,7 @@
   .. maybe add an example?
 
   The rounding library responsible for this improvement works on
-  Windows, and on Unix platforms using the gcc, icc, or suncc
+  Windows and on Unix platforms using the gcc, icc, or suncc
   compilers.  There may be a small number of platforms where correct
   operation of this code cannot be guaranteed, so the code is not
   used on such systems.  You can find out which code is being used
@@ -516,6 +695,37 @@
   Implemented by Eric Smith and Mark Dickinson, using David Gay's
   :file:`dtoa.c` library; :issue:`7117`.
 
+* Conversions from long integers and regular integers to floating
+  point now round differently, returning the floating-point number
+  closest to the number.  This doesn't matter for small integers that
+  can be converted exactly, but for large numbers that will
+  unavoidably lose precision, Python 2.7 now approximates more
+  closely.  For example, Python 2.6 computed the following::
+
+    >>> n = 295147905179352891391
+    >>> float(n)
+    2.9514790517935283e+20
+    >>> n - long(float(n))
+    65535L
+
+  Python 2.7's floating-point result is larger, but much closer to the
+  true value::
+
+    >>> n = 295147905179352891391
+    >>> float(n)
+    2.9514790517935289e+20
+    >>> n - long(float(n))
+    -1L
+
+  (Implemented by Mark Dickinson; :issue:`3166`.)
+
+  Integer division is also more accurate in its rounding behaviours.  (Also
+  implemented by Mark Dickinson; :issue:`1811`.)
+
+* Implicit coercion for complex numbers has been removed; the interpreter
+  will no longer ever attempt to call a :meth:`__coerce__` method on complex
+  objects.  (Removed by Meador Inge and Mark Dickinson; :issue:`5211`.)
+
 * The :meth:`str.format` method now supports automatic numbering of the replacement
   fields.  This makes using :meth:`str.format` more closely resemble using
   ``%s`` formatting::
@@ -543,6 +753,16 @@
   so it will now produce 'INF' and 'NAN'.
   (Contributed by Eric Smith; :issue:`3382`.)
 
+  A low-level change: the :meth:`object.__format__` method now triggers
+  a :exc:`PendingDeprecationWarning` if it's passed a format string,
+  because the :meth:`__format__` method for :class:`object` converts
+  the object to a string representation and formats that.  Previously
+  the method silently applied the format string to the string
+  representation, but that could hide mistakes in Python code.  If
+  you're supplying formatting information such as an alignment or
+  precision, presumably you're expecting the formatting to be applied
+  in some object-specific way.  (Fixed by Eric Smith; :issue:`7994`.)
+
 * The :func:`int` and :func:`long` types gained a ``bit_length``
   method that returns the number of bits necessary to represent
   its argument in binary::
@@ -560,32 +780,11 @@
 
   (Contributed by Fredrik Johansson and Victor Stinner; :issue:`3439`.)
 
-* Conversions from long integers and regular integers to floating
-  point now round differently, returning the floating-point number
-  closest to the number.  This doesn't matter for small integers that
-  can be converted exactly, but for large numbers that will
-  unavoidably lose precision, Python 2.7 now approximates more
-  closely.  For example, Python 2.6 computed the following::
-
-    >>> n = 295147905179352891391
-    >>> float(n)
-    2.9514790517935283e+20
-    >>> n - long(float(n))
-    65535L
-
-  Python 2.7's floating-point result is larger, but much closer to the
-  true value::
-
-    >>> n = 295147905179352891391
-    >>> float(n)
-    2.9514790517935289e+20
-    >>> n - long(float(n))
-    -1L
-
-  (Implemented by Mark Dickinson; :issue:`3166`.)
-
-  Integer division is also more accurate in its rounding behaviours.  (Also
-  implemented by Mark Dickinson; :issue:`1811`.)
+* The :keyword:`import` statement will no longer try a relative import
+  if an absolute import (e.g. ``from .os import sep``) fails.  This
+  fixes a bug, but could possibly break certain :keyword:`import`
+  statements that were only working by accident.  (Fixed by Meador Inge;
+  :issue:`7902`.)
 
 * It's now possible for a subclass of the built-in :class:`unicode` type
   to override the :meth:`__unicode__` method.  (Implemented by
@@ -603,9 +802,15 @@
   (Contributed by Amaury Forgeot d'Arc, after a suggestion by
   George Sakkis; :issue:`5982`.)
 
-* A new encoding named "cp720", used primarily for Arabic text, is now
-  supported.  (Contributed by Alexander Belchenko and Amaury Forgeot
-  d'Arc; :issue:`1616979`.)
+* When a restricted set of attributes were set using ``__slots__``,
+  deleting an unset attribute would not raise :exc:`AttributeError`
+  as you would expect.  Fixed by Benjamin Peterson; :issue:`7604`.)
+
+* Two new encodings are now supported: "cp720", used primarily for
+  Arabic text; and "cp858", a variant of CP 850 that adds the euro
+  symbol.  (CP720 contributed by Alexander Belchenko and Amaury
+  Forgeot d'Arc in :issue:`1616979`; CP858 contributed by Tim Hatch in
+  :issue:`8016`.)
 
 * The :class:`file` object will now set the :attr:`filename` attribute
   on the :exc:`IOError` exception when trying to open a directory
@@ -615,7 +820,7 @@
   (fixed by Stefan Krah; :issue:`5677`).
 
 * The Python tokenizer now translates line endings itself, so the
-  :func:`compile` built-in function can now accept code using any
+  :func:`compile` built-in function now accepts code using any
   line-ending convention.  Additionally, it no longer requires that the
   code end in a newline.
 
@@ -648,13 +853,12 @@
 For example, the following setting will print warnings every time
 they occur, but turn warnings from the :mod:`Cookie` module into an
 error.  (The exact syntax for setting an environment variable varies
-across operating systems and shells, so it may be different for you.)
+across operating systems and shells.)
 
 ::
 
   export PYTHONWARNINGS=all,error:::Cookie:0
 
-
 .. ======================================================================
 
 
@@ -678,7 +882,7 @@
   any of them.  This would previously take quadratic
   time for garbage collection, but now the number of full garbage collections
   is reduced as the number of objects on the heap grows.
-  The new logic is to only perform a full garbage collection pass when
+  The new logic only performs a full garbage collection pass when
   the middle generation has been collected 10 times and when the
   number of survivor objects from the middle generation exceeds 10% of
   the number of objects in the oldest generation.  (Suggested by Martin
@@ -788,11 +992,11 @@
   The new version features better Python 3.x compatibility, various bug fixes,
   and adds several new BerkeleyDB flags and methods.
   (Updated by Jes??s Cea Avi??n; :issue:`8156`.  The pybsddb
-  changelog can be browsed at http://hg.jcea.es/pybsddb/file/tip/ChangeLog.)
+  changelog can be read at http://hg.jcea.es/pybsddb/file/tip/ChangeLog.)
 
 * The :mod:`bz2` module's :class:`~bz2.BZ2File` now supports the context
-  management protocol, so you can write ``with bz2.BZ2File(...) as f: ...``.
-  (Contributed by Hagen Fuerstenau; :issue:`3860`.)
+  management protocol, so you can write ``with bz2.BZ2File(...) as f:``.
+  (Contributed by Hagen F??rstenau; :issue:`3860`.)
 
 * New class: the :class:`~collections.Counter` class in the :mod:`collections`
   module is useful for tallying data.  :class:`~collections.Counter` instances
@@ -816,7 +1020,7 @@
      >>> c['z']
      0
 
-  There are three additional :class:`~collections.Counter` methods:
+  There are three additional :class:`~collections.Counter` methods.
   :meth:`~collections.Counter.most_common` returns the N most common
   elements and their counts.  :meth:`~collections.Counter.elements`
   returns an iterator over the contained elements, repeating each
@@ -843,12 +1047,20 @@
 
   .. revision 79660
 
-  The new :class:`~collections.OrderedDict` class is described in the earlier
+  New class: :class:`~collections.OrderedDict` is described in the earlier
   section :ref:`pep-0372`.
 
+  New method: The :class:`~collections.deque` data type now has a
+  :meth:`~collections.deque.count` method that returns the number of
+  contained elements equal to the supplied argument *x*, and a
+  :meth:`~collections.deque.reverse` method that reverses the elements
+  of the deque in-place.  :class:`deque` also exposes its maximum
+  length as the read-only :attr:`~collections.deque.maxlen` attribute.
+  (Both features added by Raymond Hettinger.)
+
   The :class:`~collections.namedtuple` class now has an optional *rename* parameter.
   If *rename* is true, field names that are invalid because they've
-  been repeated or that aren't legal Python identifiers will be
+  been repeated or aren't legal Python identifiers will be
   renamed to legal names that are derived from the field's
   position within the list of fields:
 
@@ -859,13 +1071,43 @@
 
   (Added by Raymond Hettinger; :issue:`1818`.)
 
-  The :class:`~collections.deque` data type now has a
-  :meth:`~collections.deque.count` method that returns the number of
-  contained elements equal to the supplied argument *x*, and a
-  :meth:`~collections.deque.reverse` method that reverses the elements
-  of the deque in-place.  :class:`deque` also exposes its maximum
-  length as the read-only :attr:`~collections.deque.maxlen` attribute.
-  (Both features added by Raymond Hettinger.)
+  Finally, the :class:`~collections.Mapping` abstract base class now
+  returns :const:`NotImplemented` if a mapping is compared to
+  another type that isn't a :class:`Mapping`.
+  (Fixed by Daniel Stutzbach; :issue:`8729`.)
+
+* Constructors for the parsing classes in the :mod:`ConfigParser` module now
+  take a *allow_no_value* parameter, defaulting to false; if true,
+  options without values will be allowed.  For example::
+
+    >>> import ConfigParser, StringIO
+    >>> sample_config = """
+    ... [mysqld]
+    ... user = mysql
+    ... pid-file = /var/run/mysqld/mysqld.pid
+    ... skip-bdb
+    ... """
+    >>> config = ConfigParser.RawConfigParser(allow_no_value=True)
+    >>> config.readfp(StringIO.StringIO(sample_config))
+    >>> config.get('mysqld', 'user')
+    'mysql'
+    >>> print config.get('mysqld', 'skip-bdb')
+    None
+    >>> print config.get('mysqld', 'unknown')
+    Traceback (most recent call last):
+      ...
+    NoOptionError: No option 'unknown' in section: 'mysqld'
+
+  (Contributed by Mats Kindahl; :issue:`7005`.)
+
+* Deprecated function: :func:`contextlib.nested`, which allows
+  handling more than one context manager with a single :keyword:`with`
+  statement, has been deprecated, because the :keyword:`with` statement
+  now supports multiple context managers.
+
+* The :mod:`cookielib` module now ignores cookies that have an invalid
+  version field, one that doesn't contain an integer value.  (Fixed by
+  John J. Lee; :issue:`3924`.)
 
 * The :mod:`copy` module's :func:`~copy.deepcopy` function will now
   correctly copy bound instance methods.  (Implemented by
@@ -885,7 +1127,7 @@
 * New method: the :class:`~decimal.Decimal` class gained a
   :meth:`~decimal.Decimal.from_float` class method that performs an exact
   conversion of a floating-point number to a :class:`~decimal.Decimal`.
-  Note that this is an **exact** conversion that strives for the
+  This exact conversion strives for the
   closest decimal approximation to the floating-point representation's value;
   the resulting decimal value will therefore still include the inaccuracy,
   if any.
@@ -893,27 +1135,57 @@
   ``Decimal('0.1000000000000000055511151231257827021181583404541015625')``.
   (Implemented by Raymond Hettinger; :issue:`4796`.)
 
-  Most of the methods of the :class:`~decimal.Context` class now accept integers
-  as well as :class:`~decimal.Decimal` instances; the only exceptions are the
-  :meth:`~decimal.Context.canonical` and :meth:`~decimal.Context.is_canonical`
-  methods.  (Patch by Juan Jos?? Conti; :issue:`7633`.)
+  Comparing instances of :class:`Decimal` with floating-point
+  numbers now produces sensible results based on the numeric values
+  of the operands.  Previously such comparisons would fall back to
+  Python's default rules for comparing objects, which produced arbitrary
+  results based on their type.  Note that you still cannot combine
+  :class:`Decimal` and floating-point in other operations such as addition,
+  since you should be explicitly choosing how to convert between float and
+  :class:`Decimal`.
+  (Fixed by Mark Dickinson; :issue:`2531`.)
 
   The constructor for :class:`~decimal.Decimal` now accepts
   floating-point numbers (added by Raymond Hettinger; :issue:`8257`)
   and non-European Unicode characters such as Arabic-Indic digits
   (contributed by Mark Dickinson; :issue:`6595`).
 
+  Most of the methods of the :class:`~decimal.Context` class now accept integers
+  as well as :class:`~decimal.Decimal` instances; the only exceptions are the
+  :meth:`~decimal.Context.canonical` and :meth:`~decimal.Context.is_canonical`
+  methods.  (Patch by Juan Jos?? Conti; :issue:`7633`.)
+
   When using :class:`~decimal.Decimal` instances with a string's
   :meth:`~str.format` method, the default alignment was previously
-  left-alignment.  This has been changed to right-alignment, which seems
+  left-alignment.  This has been changed to right-alignment, which is
   more sensible for numeric types.  (Changed by Mark Dickinson; :issue:`6857`.)
 
+  Comparisons involving a signaling NaN value (or ``sNAN``) now signal
+  :const:`InvalidOperation` instead of silently returning a true or
+  false value depending on the comparison operator.  Quiet NaN values
+  (or ``NaN``) are now hashable.  (Fixed by Mark Dickinson;
+  :issue:`7279`.)
+
 * The :mod:`difflib` module now produces output that is more
-  compatible with modern :command:`diff`/:command:`patch` tools thanks
-  to two changes: 1) the header giving the filename now uses a tab
-  character instead of spaces as a separator, and 2) the date format
-  used is now ISO-8601 style, ``2005-01-26 23:30:50``.  (Fixed by
-  Anatoly Techtonik; :issue:`7585`.)
+  compatible with modern :command:`diff`/:command:`patch` tools
+  through one small change, using a tab character instead of spaces as
+  a separator in the header giving the filename.  (Fixed by Anatoly
+  Techtonik; :issue:`7585`.)
+
+* The Distutils ``sdist`` command now always regenerates the
+  :file:`MANIFEST` file, since even if the :file:`MANIFEST.in` or
+  :file:`setup.py` files haven't been modified, the user might have
+  created some new files that should be included.
+  (Fixed by Tarek Ziad??; :issue:`8688`.)
+
+* The :mod:`doctest` module's :const:`IGNORE_EXCEPTION_DETAIL` flag
+  will now ignore the name of the module containing the exception
+  being tested.  (Patch by Lennart Regebro; :issue:`7490`.)
+
+* The :mod:`email` module's :class:`~email.message.Message` class will
+  now accept a Unicode-valued payload, automatically converting the
+  payload to the encoding specified by :attr:`output_charset`.
+  (Added by R. David Murray; :issue:`1368247`.)
 
 * The :class:`~fractions.Fraction` class now accepts a single float or
   :class:`~decimal.Decimal` instance, or two rational numbers, as
@@ -921,17 +1193,18 @@
   rationals added in :issue:`5812`, and float/decimal in
   :issue:`8294`.)
 
-  An oversight was fixed, making the :class:`Fraction` match the other
-  numeric types; ordering comparisons (``<``, ``<=``, ``>``, ``>=``) between
+  Ordering comparisons (``<``, ``<=``, ``>``, ``>=``) between
   fractions and complex numbers now raise a :exc:`TypeError`.
+  This fixes an oversight, making the :class:`Fraction` match the other
+  numeric types.
 
   .. revision 79455
 
-* New class: a new :class:`~ftplib.FTP_TLS` class in
+* New class: :class:`~ftplib.FTP_TLS` in
   the :mod:`ftplib` module provides secure FTP
   connections using TLS encapsulation of authentication as well as
   subsequent control and data transfers.
-  (Contributed by Giampaolo Rodola', :issue:`2054`.)
+  (Contributed by Giampaolo Rodola; :issue:`2054`.)
 
   The :meth:`~ftplib.FTP.storbinary` method for binary uploads can now restart
   uploads thanks to an added *rest* parameter (patch by Pablo Mouzo;
@@ -957,8 +1230,8 @@
   otherwise. (Contributed by Antoine Pitrou; :issue:`4688`.)
 
 * The :mod:`gzip` module's :class:`~gzip.GzipFile` now supports the context
-  management protocol, so you can write ``with gzip.GzipFile(...) as f: ...``
-  (contributed by Hagen Fuerstenau; :issue:`3860`), and it now implements
+  management protocol, so you can write ``with gzip.GzipFile(...) as f:``
+  (contributed by Hagen F??rstenau; :issue:`3860`), and it now implements
   the :class:`io.BufferedIOBase` ABC, so you can wrap it with
   :class:`io.BufferedReader` for faster processing
   (contributed by Nir Aides; :issue:`7471`).
@@ -973,18 +1246,25 @@
 * New attribute: the :mod:`hashlib` module now has an :attr:`~hashlib.hashlib.algorithms`
   attribute containing a tuple naming the supported algorithms.
   In Python 2.7, ``hashlib.algorithms`` contains
-  ``('md5', 'sha1', 'sha224', 'sha256', 'sha384', 'sha512')``
+  ``('md5', 'sha1', 'sha224', 'sha256', 'sha384', 'sha512')``.
   (Contributed by Carl Chenet; :issue:`7418`.)
 
 * The default :class:`~httplib.HTTPResponse` class used by the :mod:`httplib` module now
   supports buffering, resulting in much faster reading of HTTP responses.
-  (Contributed by Kristjan Valur Jonsson; :issue:`4879`.)
+  (Contributed by Kristj??n Valur J??nsson; :issue:`4879`.)
 
   The :class:`~httplib.HTTPConnection` and :class:`~httplib.HTTPSConnection` classes
   now support a *source_address* parameter, a ``(host, port)`` 2-tuple
   giving the source address that will be used for the connection.
   (Contributed by Eldon Ziegler; :issue:`3972`.)
 
+* The :mod:`ihooks` module now supports relative imports.  Note that
+  :mod:`ihooks` is an older module for customizing imports,
+  superseded by the :mod:`imputil` module added in Python 2.0.
+  (Relative import support added by Neil Schemenauer.)
+
+  .. revision 75423
+
 * The :mod:`imaplib` module now supports IPv6 addresses.
   (Contributed by Derek Morr; :issue:`1655`.)
 
@@ -997,9 +1277,9 @@
     >>> def f(a, b=1, *pos, **named):
     ...     pass
     >>> getcallargs(f, 1, 2, 3)
-    {'a': 1, 'named': {}, 'b': 2, 'pos': (3,)}
+    {'a': 1, 'b': 2, 'pos': (3,), 'named': {}}
     >>> getcallargs(f, a=2, x=4)
-    {'a': 2, 'named': {'x': 4}, 'b': 1, 'pos': ()}
+    {'a': 2, 'b': 1, 'pos': (), 'named': {'x': 4}}
     >>> getcallargs(f)
     Traceback (most recent call last):
     ...
@@ -1050,8 +1330,8 @@
   floats or :class:`~decimal.Decimal` instances.  (Implemented by Raymond
   Hettinger; :issue:`5032`.)
 
-  :func:`itertools.combinations` and :func:`itertools.product` were
-  previously raising :exc:`ValueError` for values of *r* larger than
+  :func:`itertools.combinations` and :func:`itertools.product`
+  previously raised :exc:`ValueError` for values of *r* larger than
   the input iterable.  This was deemed a specification error, so they
   now return an empty iterator.  (Fixed by Raymond Hettinger; :issue:`4816`.)
 
@@ -1065,6 +1345,12 @@
   with any object literal that decodes to a list of pairs.
   (Contributed by Raymond Hettinger; :issue:`5381`.)
 
+* The :mod:`mailbox` module's :class:`Maildir` class now records the
+  timestamp on the directories it reads, and only re-reads them if the
+  modification time has subsequently changed.  This improves
+  performance by avoiding unneeded directory scans.  (Fixed by
+  A.M. Kuchling and Antoine Pitrou; :issue:`1607951`, :issue:`6896`.)
+
 * New functions: the :mod:`math` module gained
   :func:`~math.erf` and :func:`~math.erfc` for the error function and the complementary error function,
   :func:`~math.expm1` which computes ``e**x - 1`` with more precision than
@@ -1095,7 +1381,8 @@
   real, effective, and saved GIDs and UIDs;
   :func:`~os.setresgid` and :func:`~os.setresuid`, which set
   real, effective, and saved GIDs and UIDs to new values;
-  :func:`~os.initgroups`.  (GID/UID functions
+  :func:`~os.initgroups`, which initialize the group access list
+  for the current process.  (GID/UID functions
   contributed by Travis H.; :issue:`6508`.  Support for initgroups added
   by Jean-Paul Calderone; :issue:`7333`.)
 
@@ -1117,6 +1404,20 @@
   now accept an optional *flags* argument, for consistency with the
   other functions in the module.  (Added by Gregory P. Smith.)
 
+* New function: :func:`~runpy.run_path` in the :mod:`runpy` module
+  will execute the code at a provided *path* argument.  *path* can be
+  the path of a Python source file (:file:`example.py`), a compiled
+  bytecode file (:file:`example.pyc`), a directory
+  (:file:`./package/`), or a zip archive (:file:`example.zip`).  If a
+  directory or zip path is provided, it will be added to the front of
+  ``sys.path`` and the module :mod:`__main__` will be imported.  It's
+  expected that the directory or zip contains a :file:`__main__.py`;
+  if it doesn't, some other :file:`__main__.py` might be imported from
+  a location later in ``sys.path``.  This makes more of the machinery
+  of :mod:`runpy` available to scripts that want to mimic the way
+  Python's command line processes an explicit path name.
+  (Added by Nick Coghlan; :issue:`6816`.)
+
 * New function: in the :mod:`shutil` module, :func:`~shutil.make_archive`
   takes a filename, archive type (zip or tar-format), and a directory
   path, and creates an archive containing the directory's contents.
@@ -1128,12 +1429,17 @@
   named pipes like a regular file by opening them for reading, and
   this would block indefinitely.  (Fixed by Antoine Pitrou; :issue:`3002`.)
 
+* The :mod:`signal` module no longer re-installs the signal handler
+  unless this is truly necessary, which fixes a bug that could make it
+  impossible to catch the EINTR signal robustly.  (Fixed by
+  Charles-Francois Natali; :issue:`8354`.)
+
 * New functions: in the :mod:`site` module, three new functions
   return various site- and user-specific paths.
   :func:`~site.getsitepackages` returns a list containing all
-  global site-packages directories, and
+  global site-packages directories,
   :func:`~site.getusersitepackages` returns the path of the user's
-  site-packages directory.
+  site-packages directory, and
   :func:`~site.getuserbase` returns the value of the :envvar:`USER_BASE`
   environment variable, giving the path to a directory that can be used
   to store data.
@@ -1144,20 +1450,7 @@
   catch and swallow the :exc:`KeyboardInterrupt` exception.  (Fixed by
   Victor Stinner; :issue:`3137`.)
 
-* The :mod:`socket` module's :class:`~ssl.SSL` objects now support the
-  buffer API, which fixed a test suite failure (fix by Antoine Pitrou;
-  :issue:`7133`).  :class:`SSL` objects also now automatically set
-  OpenSSL's :cmacro:`SSL_MODE_AUTO_RETRY`, which will prevent an error
-  code being returned from :meth:`recv` operations that trigger an SSL
-  renegotiation (fix by Antoine Pitrou; :issue:`8222`).
-
-  The version of OpenSSL being used is now available as the module
-  attributes :attr:`OPENSSL_VERSION` (a string),
-  :attr:`OPENSSL_VERSION_INFO` (a 5-tuple), and
-  :attr:`OPENSSL_VERSION_NUMBER` (an integer).  (Added by Antoine
-  Pitrou; :issue:`8321`.)
-
-  The :func:`~socket.create_connection` function
+* The :func:`~socket.create_connection` function
   gained a *source_address* parameter, a ``(host, port)`` 2-tuple
   giving the source address that will be used for the connection.
   (Contributed by Eldon Ziegler; :issue:`3972`.)
@@ -1168,11 +1461,16 @@
   Antoine Pitrou; :issue:`8104`.)
 
 * The :mod:`SocketServer` module's :class:`~SocketServer.TCPServer` class now
-  has a :attr:`~SocketServer.TCPServer.disable_nagle_algorithm` class attribute.
-  The default value is False; if overridden to be True,
+  supports socket timeouts and disabling the Nagle algorithm.
+  The :attr:`~SocketServer.TCPServer.disable_nagle_algorithm` class attribute
+  defaults to False; if overridden to be True,
   new request connections will have the TCP_NODELAY option set to
   prevent buffering many small sends into a single TCP packet.
-  (Contributed by Kristjan Valur Jonsson; :issue:`6192`.)
+  The :attr:`~SocketServer.TCPServer.timeout` class attribute can hold
+  a timeout in seconds that will be applied to the request socket; if
+  no request is received within that time, :meth:`handle_timeout`
+  will be called and :meth:`handle_request` will return.
+  (Contributed by Kristj??n Valur J??nsson; :issue:`6192` and :issue:`6267`.)
 
 * Updated module: the :mod:`sqlite3` module has been updated to
   version 2.6.0 of the `pysqlite package <http://code.google.com/p/pysqlite/>`__. Version 2.6.0 includes a number of bugfixes, and adds
@@ -1181,6 +1479,32 @@
   and then call :meth:`~sqlite3.Connection.load_extension` to load a particular shared library.
   (Updated by Gerhard H??ring.)
 
+* The :mod:`ssl` module's :class:`ssl.SSLSocket` objects now support the
+  buffer API, which fixed a test suite failure (fix by Antoine Pitrou;
+  :issue:`7133`) and automatically set
+  OpenSSL's :cmacro:`SSL_MODE_AUTO_RETRY`, which will prevent an error
+  code being returned from :meth:`recv` operations that trigger an SSL
+  renegotiation (fix by Antoine Pitrou; :issue:`8222`).
+
+  The :func:`ssl.wrap_socket` constructor function now takes a
+  *ciphers* argument that's a string listing the encryption algorithms
+  to be allowed; the format of the string is described
+  `in the OpenSSL documentation
+  <http://www.openssl.org/docs/apps/ciphers.html#CIPHER_LIST_FORMAT>`__.
+  (Added by Antoine Pitrou; :issue:`8322`.)
+
+  Another change makes the extension load all of OpenSSL's ciphers and
+  digest algorithms so that they're all available.  Some SSL
+  certificates couldn't be verified, reporting an "unknown algorithm"
+  error.  (Reported by Beda Kosata, and fixed by Antoine Pitrou;
+  :issue:`8484`.)
+
+  The version of OpenSSL being used is now available as the module
+  attributes :data:`ssl.OPENSSL_VERSION` (a string),
+  :data:`ssl.OPENSSL_VERSION_INFO` (a 5-tuple), and
+  :data:`ssl.OPENSSL_VERSION_NUMBER` (an integer).  (Added by Antoine
+  Pitrou; :issue:`8321`.)
+
 * The :mod:`struct` module will no longer silently ignore overflow
   errors when a value is too large for a particular integer format
   code (one of ``bBhHiIlLqQ``); it now always raises a
@@ -1216,6 +1540,10 @@
   false for ones that are implicitly global.
   (Contributed by Jeremy Hylton.)
 
+* The :mod:`syslog` module will now use the value of ``sys.argv[0]`` as the
+  identifier instead of the previous default value of ``'python'``.
+  (Changed by Sean Reifschneider; :issue:`8451`.)
+
 * The ``sys.version_info`` value is now a named tuple, with attributes
   named :attr:`major`, :attr:`minor`, :attr:`micro`,
   :attr:`releaselevel`, and :attr:`serial`.  (Contributed by Ross
@@ -1237,7 +1565,7 @@
 
   :mod:`tarfile` now supports filtering the :class:`~tarfile.TarInfo`
   objects being added to a tar file.  When you call :meth:`~tarfile.TarFile.add`,
-  instance, you may supply an optional *filter* argument
+  you may supply an optional *filter* argument
   that's a callable.  The *filter* callable will be passed the
   :class:`~tarfile.TarInfo` for every file being added, and can modify and return it.
   If the callable returns ``None``, the file will be excluded from the
@@ -1262,8 +1590,39 @@
   and has been updated to version 5.2.0 (updated by
   Florent Xicluna; :issue:`8024`).
 
-* The :class:`~UserDict.UserDict` class is now a new-style class.  (Changed by
-  Benjamin Peterson.)
+* The :mod:`urlparse` module's :func:`~urlparse.urlsplit` now handles
+  unknown URL schemes in a fashion compliant with :rfc:`3986`: if the
+  URL is of the form ``"<something>://..."``, the text before the
+  ``://`` is treated as the scheme, even if it's a made-up scheme that
+  the module doesn't know about.  This change may break code that
+  worked around the old behaviour.  For example, Python 2.6.4 or 2.5
+  will return the following:
+
+    >>> import urlparse
+    >>> urlparse.urlsplit('invented://host/filename?query')
+    ('invented', '', '//host/filename?query', '', '')
+
+  Python 2.7 (and Python 2.6.5) will return:
+
+    >>> import urlparse
+    >>> urlparse.urlsplit('invented://host/filename?query')
+    ('invented', 'host', '/filename?query', '', '')
+
+  (Python 2.7 actually produces slightly different output, since it
+  returns a named tuple instead of a standard tuple.)
+
+  The :mod:`urlparse` module also supports IPv6 literal addresses as defined by
+  :rfc:`2732` (contributed by Senthil Kumaran; :issue:`2987`). ::
+
+    >>> urlparse.urlparse('http://[1080::8:800:200C:417A]/foo')
+    ParseResult(scheme='http', netloc='[1080::8:800:200C:417A]',
+                path='/foo', params='', query='', fragment='')
+
+* New class: the :class:`~weakref.WeakSet` class in the :mod:`weakref`
+  module is a set that only holds weak references to its elements; elements
+  will be removed once there are no references pointing to them.
+  (Originally implemented in Python 3.x by Raymond Hettinger, and backported
+  to 2.7 by Michael Foord.)
 
 * The ElementTree library, :mod:`xml.etree`, no longer escapes
   ampersands and angle brackets when outputting an XML processing
@@ -1271,13 +1630,22 @@
   or comment (which looks like ``<!-- comment -->``).
   (Patch by Neil Muller; :issue:`2746`.)
 
+* The XML-RPC client and server, provided by the :mod:`xmlrpclib` and
+  :mod:`SimpleXMLRPCServer` modules, have improved performance by
+  supporting HTTP/1.1 keep-alive and by optionally using gzip encoding
+  to compress the XML being exchanged.  The gzip compression is
+  controlled by the :attr:`encode_threshold` attribute of
+  :class:`SimpleXMLRPCRequestHandler`, which contains a size in bytes;
+  responses larger than this will be compressed.
+  (Contributed by Kristj??n Valur J??nsson; :issue:`6267`.)
+
 * The :mod:`zipfile` module's :class:`~zipfile.ZipFile` now supports the context
-  management protocol, so you can write ``with zipfile.ZipFile(...) as f: ...``.
+  management protocol, so you can write ``with zipfile.ZipFile(...) as f:``.
   (Contributed by Brian Curtin; :issue:`5511`.)
 
-  :mod:`zipfile` now supports archiving empty directories and
+  :mod:`zipfile` now also supports archiving empty directories and
   extracts them correctly.  (Fixed by Kuba Wieczorek; :issue:`4710`.)
-  Reading files out of an archive is now faster, and interleaving
+  Reading files out of an archive is faster, and interleaving
   :meth:`~zipfile.ZipFile.read` and :meth:`~zipfile.ZipFile.readline` now works correctly.
   (Contributed by Nir Aides; :issue:`7610`.)
 
@@ -1291,36 +1659,157 @@
   :issue:`6003`.)
 
 
+.. ======================================================================
+.. whole new modules get described in subsections here
+
+
+.. _importlib-section:
+
+New module: importlib
+------------------------------
+
+Python 3.1 includes the :mod:`importlib` package, a re-implementation
+of the logic underlying Python's :keyword:`import` statement.
+:mod:`importlib` is useful for implementors of Python interpreters and
+to users who wish to write new importers that can participate in the
+import process.  Python 2.7 doesn't contain the complete
+:mod:`importlib` package, but instead has a tiny subset that contains
+a single function, :func:`~importlib.import_module`.
+
+``import_module(name, package=None)`` imports a module.  *name* is
+a string containing the module or package's name.  It's possible to do
+relative imports by providing a string that begins with a ``.``
+character, such as ``..utils.errors``.  For relative imports, the
+*package* argument must be provided and is the name of the package that
+will be used as the anchor for
+the relative import.  :func:`~importlib.import_module` both inserts the imported
+module into ``sys.modules`` and returns the module object.
+
+Here are some examples::
+
+    >>> from importlib import import_module
+    >>> anydbm = import_module('anydbm')  # Standard absolute import
+    >>> anydbm
+    <module 'anydbm' from '/p/python/Lib/anydbm.py'>
+    >>> # Relative import
+    >>> file_util = import_module('..file_util', 'distutils.command')
+    >>> file_util
+    <module 'distutils.file_util' from '/python/Lib/distutils/file_util.pyc'>
+
+:mod:`importlib` was implemented by Brett Cannon and introduced in
+Python 3.1.
+
+
 New module: sysconfig
 ---------------------------------
 
-XXX A new :mod:`sysconfig` module has been extracted from
-:mod:`distutils` and put in the standard library.
+The :mod:`sysconfig` module has been pulled out of the Distutils
+package, becoming a new top-level module in its own right.
+:mod:`sysconfig` provides functions for getting information about
+Python's build process: compiler switches, installation paths, the
+platform name, and whether Python is running from its source
+directory.
+
+Some of the functions in the module are:
+
+* :func:`~sysconfig.get_config_var` returns variables from Python's
+  Makefile and the :file:`pyconfig.h` file.
+* :func:`~sysconfig.get_config_vars` returns a dictionary containing
+  all of the configuration variables.
+* :func:`~sysconfig.getpath` returns the configured path for
+  a particular type of module: the standard library,
+  site-specific modules, platform-specific modules, etc.
+* :func:`~sysconfig.is_python_build` returns true if you're running a
+  binary from a Python source tree, and false otherwise.
+
+Consult the :mod:`sysconfig` documentation for more details and for
+a complete list of functions.
+
+The Distutils package and :mod:`sysconfig` are now maintained by Tarek
+Ziad??, who has also started a Distutils2 package (source repository at
+http://hg.python.org/distutils2/) for developing a next-generation
+version of Distutils.
 
-The :mod:`sysconfig` module provides access to Python's configuration
-information like the list of installation paths and the configuration
-variables relevant for the current platform. (contributed by Tarek)
 
-Updated module: ElementTree 1.3
----------------------------------
+ttk: Themed Widgets for Tk
+--------------------------
+
+Tcl/Tk 8.5 includes a set of themed widgets that re-implement basic Tk
+widgets but have a more customizable appearance and can therefore more
+closely resemble the native platform's widgets.  This widget
+set was originally called Tile, but was renamed to Ttk (for "themed Tk")
+on being added to Tcl/Tck release 8.5.
 
-XXX write this.
+To learn more, read the :mod:`ttk` module documentation.  You may also
+wish to read the Tcl/Tk manual page describing the
+Ttk theme engine, available at
+http://www.tcl.tk/man/tcl8.5/TkCmd/ttk_intro.htm. Some
+screenshots of the Python/Ttk code in use are at
+http://code.google.com/p/python-ttk/wiki/Screenshots.
 
-.. ======================================================================
-.. whole new modules get described in subsections here
+The :mod:`ttk` module was written by Guilherme Polo and added in
+:issue:`2983`.  An alternate version called ``Tile.py``, written by
+Martin Franklin and maintained by Kevin Walzer, was proposed for
+inclusion in :issue:`2618`, but the authors argued that Guilherme
+Polo's work was more comprehensive.
 
 
-Unit Testing Enhancements
+.. _unittest-section:
+
+Updated module: unittest
 ---------------------------------
 
-The :mod:`unittest` module was enhanced in several ways.
-The progress messages now shows 'x' for expected failures
+The :mod:`unittest` module was greatly enhanced; many
+new features were added.  Most of these features were implemented
+by Michael Foord, unless otherwise noted.  The enhanced version of
+the module is downloadable separately for use with Python versions 2.4 to 2.6,
+packaged as the :mod:`unittest2` package, from
+http://pypi.python.org/pypi/unittest2.
+
+When used from the command line, the module can automatically discover
+tests.  It's not as fancy as `py.test <http://pytest.org>`__ or
+`nose <http://code.google.com/p/python-nose/>`__, but provides a simple way
+to run tests kept within a set of package directories.  For example,
+the following command will search the :file:`test/` subdirectory for
+any importable test files named ``test*.py``::
+
+   python -m unittest discover -s test
+
+Consult the :mod:`unittest` module documentation for more details.
+(Developed in :issue:`6001`.)
+
+The :func:`main` function supports some other new options:
+
+* :option:`-b` or :option:`--buffer` will buffer the standard output
+  and standard error streams during each test.  If the test passes,
+  any resulting output will be discarded; on failure, the buffered
+  output will be displayed.
+
+* :option:`-c` or :option:`--catch` will cause the control-C interrupt
+  to be handled more gracefully.  Instead of interrupting the test
+  process immediately, the currently running test will be completed
+  and then the partial results up to the interruption will be reported.
+  If you're impatient, a second press of control-C will cause an immediate
+  interruption.
+
+  This control-C handler tries to avoid causing problems when the code
+  being tested or the tests being run have defined a signal handler of
+  their own, by noticing that a signal handler was already set and
+  calling it.  If this doesn't work for you, there's a
+  :func:`removeHandler` decorator that can be used to mark tests that
+  should have the control-C handling disabled.
+
+* :option:`-f` or :option:`--failfast` makes
+  test execution stop immediately when a test fails instead of
+  continuing to execute further tests.  (Suggested by Cliff Dyer and
+  implemented by Michael Foord; :issue:`8074`.)
+
+The progress messages now show 'x' for expected failures
 and 'u' for unexpected successes when run in verbose mode.
 (Contributed by Benjamin Peterson.)
-Test cases can raise the :exc:`~unittest.SkipTest` exception to skip a test.
-(:issue:`1034053`.)
 
-.. XXX describe test discovery (Contributed by Michael Foord; :issue:`6001`.)
+Test cases can raise the :exc:`~unittest.SkipTest` exception to skip a
+test (:issue:`1034053`).
 
 The error messages for :meth:`~unittest.TestCase.assertEqual`,
 :meth:`~unittest.TestCase.assertTrue`, and :meth:`~unittest.TestCase.assertFalse`
@@ -1330,7 +1819,7 @@
 provide will be printed for failures.  (Added by Michael Foord; :issue:`5663`.)
 
 The :meth:`~unittest.TestCase.assertRaises` method now
-return a context handler when called without providing a callable
+returns a context handler when called without providing a callable
 object to run.  For example, you can write this::
 
   with self.assertRaises(KeyError):
@@ -1350,7 +1839,7 @@
 
 The methods :meth:`~unittest.TestCase.addCleanup` and
 :meth:`~unittest.TestCase.doCleanups` were added.
-:meth:`~unittest.TestCase.addCleanup` allows you to add cleanup functions that
+:meth:`~unittest.TestCase.addCleanup` lets you add cleanup functions that
 will be called unconditionally (after :meth:`~unittest.TestCase.setUp` if
 :meth:`~unittest.TestCase.setUp` fails, otherwise after :meth:`~unittest.TestCase.tearDown`). This allows
 for much simpler resource allocation and deallocation during tests
@@ -1382,10 +1871,10 @@
   differences in the two strings.  This comparison is now used by
   default when Unicode strings are compared with :meth:`~unittest.TestCase.assertEqual`.
 
-* :meth:`~unittest.TestCase.assertRegexpMatches` checks whether its first argument is a
-  string matching a regular expression provided as its second argument.
-
-  .. XXX add assertNotRegexpMatches see issue 8038
+* :meth:`~unittest.TestCase.assertRegexpMatches` and
+  :meth:`~unittest.TestCase.assertNotRegexpMatches` checks whether the
+  first argument is a string matching or not matching the regular
+  expression provided as the second argument (:issue:`8038`).
 
 * :meth:`~unittest.TestCase.assertRaisesRegexp` checks whether a particular exception
   is raised, and then also checks that the string representation of
@@ -1414,9 +1903,10 @@
   all of the key/value pairs in *first* are found in *second*.
 
 * :meth:`~unittest.TestCase.assertAlmostEqual` and :meth:`~unittest.TestCase.assertNotAlmostEqual` test
-  whether *first* and *second* are approximately equal by computing
-  their difference, rounding the result to an optionally-specified number
-  of *places* (the default is 7), and comparing to zero.
+  whether *first* and *second* are approximately equal.  This method
+  can either round their difference to an optionally-specified number
+  of *places* (the default is 7) and compare it to zero, or require
+  the difference to be smaller than a supplied *delta* value.
 
 * :meth:`~unittest.TestLoader.loadTestsFromName` properly honors the
   :attr:`~unittest.TestLoader.suiteClass` attribute of
@@ -1428,20 +1918,13 @@
   objects being compared are of the specified type.  This function
   should compare the two objects and raise an exception if they don't
   match; it's a good idea for the function to provide additional
-  information about why the two objects are matching, much as the new
+  information about why the two objects aren't matching, much as the new
   sequence comparison methods do.
 
 :func:`unittest.main` now takes an optional ``exit`` argument.  If
-False, :func:`~unittest.main` doesn't call :func:`sys.exit`, allowing it to be
-used from the interactive interpreter. (Contributed by J. Pablo
-Fern??ndez; :issue:`3379`.)
-
-A new command-line switch, :option:`-f` or :option:`--failfast`, makes
-test execution stop immediately when a test fails instead of
-continuing to execute further tests.  (Suggested by Cliff Dyer and
-implemented by Michael Foord; :issue:`8074`.)
-
-.. XXX document the other new switches
+False, :func:`~unittest.main` doesn't call :func:`sys.exit`, allowing
+:func:`main` to be used from the interactive interpreter.
+(Contributed by J. Pablo Fern??ndez; :issue:`3379`.)
 
 :class:`~unittest.TestResult` has new :meth:`~unittest.TestResult.startTestRun` and
 :meth:`~unittest.TestResult.stopTestRun` methods that are called immediately before
@@ -1450,70 +1933,98 @@
 With all these changes, the :file:`unittest.py` was becoming awkwardly
 large, so the module was turned into a package and the code split into
 several files (by Benjamin Peterson).  This doesn't affect how the
-module is imported.
-
-
-.. _importlib-section:
-
-importlib: Importing Modules
-------------------------------
+module is imported or used.
 
-Python 3.1 includes the :mod:`importlib` package, a re-implementation
-of the logic underlying Python's :keyword:`import` statement.
-:mod:`importlib` is useful for implementors of Python interpreters and
-to users who wish to write new importers that can participate in the
-import process.  Python 2.7 doesn't contain the complete
-:mod:`importlib` package, but instead has a tiny subset that contains
-a single function, :func:`~importlib.import_module`.
-
-``import_module(name, package=None)`` imports a module.  *name* is
-a string containing the module or package's name.  It's possible to do
-relative imports by providing a string that begins with a ``.``
-character, such as ``..utils.errors``.  For relative imports, the
-*package* argument must be provided and is the name of the package that
-will be used as the anchor for
-the relative import.  :func:`~importlib.import_module` both inserts the imported
-module into ``sys.modules`` and returns the module object.
-
-Here are some examples::
-
-    >>> from importlib import import_module
-    >>> anydbm = import_module('anydbm')  # Standard absolute import
-    >>> anydbm
-    <module 'anydbm' from '/p/python/Lib/anydbm.py'>
-    >>> # Relative import
-    >>> sysconfig = import_module('..sysconfig', 'distutils.command')
-    >>> sysconfig
-    <module 'distutils.sysconfig' from '/p/python/Lib/distutils/sysconfig.pyc'>
-
-:mod:`importlib` was implemented by Brett Cannon and introduced in
-Python 3.1.
-
-
-ttk: Themed Widgets for Tk
---------------------------
-
-Tcl/Tk 8.5 includes a set of themed widgets that re-implement basic Tk
-widgets but have a more customizable appearance and can therefore more
-closely resemble the native platform's widgets.  This widget
-set was originally called Tile, but was renamed to Ttk (for "themed Tk")
-on being added to Tcl/Tck release 8.5.
+.. seealso::
 
-XXX write a brief discussion and an example here.
+  http://www.voidspace.org.uk/python/articles/unittest2.shtml
+    Describes the new features, how to use them, and the
+    rationale for various design decisions.  (By Michael Foord.)
 
-The :mod:`ttk` module was written by Guilherme Polo and added in
-:issue:`2983`.  An alternate version called ``Tile.py``, written by
-Martin Franklin and maintained by Kevin Walzer, was proposed for
-inclusion in :issue:`2618`, but the authors argued that Guilherme
-Polo's work was more comprehensive.
+.. _elementtree-section:
 
+Updated module: ElementTree 1.3
+---------------------------------
 
-Deprecations and Removals
-=========================
+The version of the ElementTree library included with Python was updated to
+version 1.3.  Some of the new features are:
 
-* :func:`contextlib.nested`, which allows handling more than one context manager
-  with one :keyword:`with` statement, has been deprecated; :keyword:`with`
-  supports multiple context managers syntactically now.
+* The various parsing functions now take a *parser* keyword argument
+  giving an :class:`XMLParser` instance that will
+  be used.  This makes it possible to override the file's internal encoding::
+
+    p = ET.XMLParser(encoding='utf-8')
+    t = ET.XML("""<root/>""", parser=p)
+
+  Errors in parsing XML now raise a :exc:`ParseError` exception, whose
+  instances have a :attr:`position` attribute
+  containing a (*line*, *column*) tuple giving the location of the problem.
+
+* ElementTree's code for converting trees to a string has been
+  significantly reworked, making it roughly twice as fast in many
+  cases.  The :class:`ElementTree` :meth:`write` and :class:`Element`
+  :meth:`write` methods now have a *method* parameter that can be
+  "xml" (the default), "html", or "text".  HTML mode will output empty
+  elements as ``<empty></empty>`` instead of ``<empty/>``, and text
+  mode will skip over elements and only output the text chunks.  If
+  you set the :attr:`tag` attribute of an element to ``None`` but
+  leave its children in place, the element will be omitted when the
+  tree is written out, so you don't need to do more extensive rearrangement
+  to remove a single element.
+
+  Namespace handling has also been improved.  All ``xmlns:<whatever>``
+  declarations are now output on the root element, not scattered throughout
+  the resulting XML.  You can set the default namespace for a tree
+  by setting the :attr:`default_namespace` attribute and can
+  register new prefixes with :meth:`register_namespace`.  In XML mode,
+  you can use the true/false *xml_declaration* parameter to suppress the
+  XML declaration.
+
+* New :class:`Element` method: :meth:`extend` appends the items from a
+  sequence to the element's children.  Elements themselves behave like
+  sequences, so it's easy to move children from one element to
+  another::
+
+    from xml.etree import ElementTree as ET
+
+    t = ET.XML("""<list>
+      <item>1</item> <item>2</item>  <item>3</item>
+    </list>""")
+    new = ET.XML('<root/>')
+    new.extend(t)
+
+    # Outputs <root><item>1</item>...</root>
+    print ET.tostring(new)
+
+* New :class:`Element` method: :meth:`iter` yields the children of the
+  element as a generator.  It's also possible to write ``for child in
+  elem:`` to loop over an element's children.  The existing method
+  :meth:`getiterator` is now deprecated, as is :meth:`getchildren`
+  which constructs and returns a list of children.
+
+* New :class:`Element` method: :meth:`itertext` yields all chunks of
+  text that are descendants of the element.  For example::
+
+    t = ET.XML("""<list>
+      <item>1</item> <item>2</item>  <item>3</item>
+    </list>""")
+
+    # Outputs ['\n  ', '1', ' ', '2', '  ', '3', '\n']
+    print list(t.itertext())
+
+* Deprecated: using an element as a Boolean (i.e., ``if elem:``) would
+  return true if the element had any children, or false if there were
+  no children.  This behaviour is confusing -- ``None`` is false, but
+  so is a childless element? -- so it will now trigger a
+  :exc:`FutureWarning`.  In your code, you should be explicit: write
+  ``len(elem) != 0`` if you're interested in the number of children,
+  or ``elem is not None``.
+
+Fredrik Lundh develops ElementTree and produced the 1.3 version;
+you can read his article describing 1.3 at
+http://effbot.org/zone/elementtree-13-intro.htm.
+Florent Xicluna updated the version included with
+Python, after discussions on python-dev and in :issue:`6472`.)
 
 .. ======================================================================
 
@@ -1528,9 +2039,9 @@
   <http://sourceware.org/gdb/current/onlinedocs/gdb/Python.html>`__.
   When you begin debugging an executable program P, GDB will look for
   a file named ``P-gdb.py`` and automatically read it.  Dave Malcolm
-  contributed a :file:`python-gdb.py` that adds a number of useful
-  commands when debugging Python itself.  For example, there are
-  ``py-up`` and ``py-down`` that go up or down one Python stack frame,
+  contributed a :file:`python-gdb.py` that adds a number of
+  commands useful when debugging Python itself.  For example,
+  ``py-up`` and ``py-down`` go up or down one Python stack frame,
   which usually corresponds to several C stack frames.  ``py-print``
   prints the value of a Python variable, and ``py-bt`` prints the
   Python stack trace.  (Added as a result of :issue:`8032`.)
@@ -1543,11 +2054,11 @@
 * :cfunc:`Py_AddPendingCall` is now thread-safe, letting any
   worker thread submit notifications to the main Python thread.  This
   is particularly useful for asynchronous IO operations.
-  (Contributed by Kristjan Valur Jonsson; :issue:`4293`.)
+  (Contributed by Kristj??n Valur J??nsson; :issue:`4293`.)
 
 * New function: :cfunc:`PyCode_NewEmpty` creates an empty code object;
   only the filename, function name, and first line number are required.
-  This is useful to extension modules that are attempting to
+  This is useful for extension modules that are attempting to
   construct a more useful traceback stack.  Previously such
   extensions needed to call :cfunc:`PyCode_New`, which had many
   more arguments.  (Added by Jeffrey Yasskin.)
@@ -1555,7 +2066,7 @@
 * New function: :cfunc:`PyErr_NewExceptionWithDoc` creates a new
   exception class, just as the existing :cfunc:`PyErr_NewException` does,
   but takes an extra ``char *`` argument containing the docstring for the
-  new exception class.  (Added by the 'lekma' user on the Python bug tracker;
+  new exception class.  (Added by 'lekma' on the Python bug tracker;
   :issue:`7033`.)
 
 * New function: :cfunc:`PyFrame_GetLineNumber` takes a frame object
@@ -1576,6 +2087,28 @@
   :cfunc:`PyOS_ascii_strtod` and :cfunc:`PyOS_ascii_atof` functions
   are now deprecated.
 
+* New function: :cfunc:`PySys_SetArgvEx` sets the value of
+  ``sys.argv`` and can optionally update ``sys.path`` to include the
+  directory containing the script named by ``sys.argv[0]`` depending
+  on the value of an *updatepath* parameter.
+
+  This function was added to close a security hole for applications
+  that embed Python.  The old function, :cfunc:`PySys_SetArgv`, would
+  always update ``sys.path``, and sometimes it would add the current
+  directory.  This meant that, if you ran an application embedding
+  Python in a directory controlled by someone else, attackers could
+  put a Trojan-horse module in the directory (say, a file named
+  :file:`os.py`) that your application would then import and run.
+
+  If you maintain a C/C++ application that embeds Python, check
+  whether you're calling :cfunc:`PySys_SetArgv` and carefully consider
+  whether the application should be using :cfunc:`PySys_SetArgvEx`
+  with *updatepath* set to false.
+
+  Security issue reported as `CVE-2008-5983
+  <http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2008-5983>`_;
+  discussed in :issue:`5753`, and fixed by Antoine Pitrou.
+
 * New macros: the Python header files now define the following macros:
   :cmacro:`Py_ISALNUM`,
   :cmacro:`Py_ISALPHA`,
@@ -1594,9 +2127,14 @@
 
   .. XXX these macros don't seem to be described in the c-api docs.
 
+* Removed function: :cmacro:`PyEval_CallObject` is now only available
+  as a macro.  A function version was being kept around to preserve
+  ABI linking compatibility, but that was in 1997; it can certainly be
+  deleted by now.  (Removed by Antoine Pitrou; :issue:`8276`.)
+
 * New format codes: the :cfunc:`PyFormat_FromString`,
-  :cfunc:`PyFormat_FromStringV`, and :cfunc:`PyErr_Format` now
-  accepts ``%lld`` and ``%llu`` format codes for displaying values of
+  :cfunc:`PyFormat_FromStringV`, and :cfunc:`PyErr_Format` functions now
+  accept ``%lld`` and ``%llu`` format codes for displaying
   C's :ctype:`long long` types.
   (Contributed by Mark Dickinson; :issue:`7228`.)
 
@@ -1610,7 +2148,7 @@
   ever release the lock, since the other threads weren't replicated,
   and the child process would no longer be able to perform imports.
 
-  Python 2.7 now acquires the import lock before performing an
+  Python 2.7 acquires the import lock before performing an
   :func:`os.fork`, and will also clean up any locks created using the
   :mod:`threading` module.  C extension modules that have internal
   locks, or that call :cfunc:`fork()` themselves, will not benefit
@@ -1623,6 +2161,12 @@
   being raised when an interpreter shuts down.
   (Patch by Adam Olsen; :issue:`1722344`.)
 
+* When using the :ctype:`PyMemberDef` structure to define attributes
+  of a type, Python will no longer let you try to delete or set a
+  :const:`T_STRING_INPLACE` attribute.
+
+  .. rev 79644
+
 * Global symbols defined by the :mod:`ctypes` module are now prefixed
   with ``Py``, or with ``_ctypes``.  (Implemented by Thomas
   Heller; :issue:`3102`.)
@@ -1631,15 +2175,15 @@
   building the :mod:`pyexpat` module to use the system Expat library.
   (Contributed by Arfrever Frehtes Taifersar Arahesis; :issue:`7609`.)
 
-* New configure option: compiling Python with the
+* New configure option: the
   :option:`--with-valgrind` option will now disable the pymalloc
   allocator, which is difficult for the Valgrind memory-error detector
   to analyze correctly.
   Valgrind will therefore be better at detecting memory leaks and
   overruns. (Contributed by James Henstridge; :issue:`2422`.)
 
-* New configure option: you can now supply no arguments to
-  :option:`--with-dbmliborder=` in order to build none of the various
+* New configure option: you can now supply an empty string to
+  :option:`--with-dbmliborder=` in order to disable all of the various
   DBM modules.  (Added by Arfrever Frehtes Taifersar Arahesis;
   :issue:`6491`.)
 
@@ -1660,6 +2204,52 @@
   Arfrever Frehtes Taifersar Arahesis; :issue:`6094`.)
 
 
+.. _whatsnew27-capsules:
+
+Capsules
+-------------------
+
+Python 3.1 adds a new C datatype, :ctype:`PyCapsule`, for providing a
+C API to an extension module.  A capsule is essentially the holder of
+a C ``void *`` pointer, and is made available as a module attribute; for
+example, the :mod:`socket` module's API is exposed as ``socket.CAPI``,
+and :mod:`unicodedata` exposes ``ucnhash_CAPI``.  Other extensions
+can import the module, access its dictionary to get the capsule
+object, and then get the ``void *`` pointer, which will usually point
+to an array of pointers to the module's various API functions.
+
+There is an existing data type already used for this,
+:ctype:`PyCObject`, but it doesn't provide type safety.  Evil code
+written in pure Python could cause a segmentation fault by taking a
+:ctype:`PyCObject` from module A and somehow substituting it for the
+:ctype:`PyCObject` in module B.   Capsules know their own name,
+and getting the pointer requires providing the name::
+
+   void *vtable;
+
+   if (!PyCapsule_IsValid(capsule, "mymodule.CAPI") {
+           PyErr_SetString(PyExc_ValueError, "argument type invalid");
+           return NULL;
+   }
+
+   vtable = PyCapsule_GetPointer(capsule, "mymodule.CAPI");
+
+You are assured that ``vtable`` points to whatever you're expecting.
+If a different capsule was passed in, :cfunc:`PyCapsule_IsValid` would
+detect the mismatched name and return false.  Refer to
+:ref:`using-capsules` for more information on using these objects.
+
+Python 2.7 now uses capsules internally to provide various
+extension-module APIs, but the :cfunc:`PyCObject_AsVoidPtr` was
+modified to handle capsules, preserving compile-time compatibility
+with the :ctype:`CObject` interface.  Use of
+:cfunc:`PyCObject_AsVoidPtr` will signal a
+:exc:`PendingDeprecationWarning`, which is silent by default.
+
+Implemented in Python 3.1 and backported to 2.7 by Larry Hastings;
+discussed in :issue:`5630`.
+
+
 .. ======================================================================
 
 Port-Specific Changes: Windows
@@ -1672,16 +2262,24 @@
   and :data:`LIBRARIES_ASSEMBLY_NAME_PREFIX`.
   (Contributed by David Cournapeau; :issue:`4365`.)
 
+* The :mod:`_winreg` module for accessing the registry now implements
+  the :func:`CreateKeyEx` and :func:`DeleteKeyEx` functions, extended
+  versions of previously-supported functions that take several extra
+  arguments.  The :func:`DisableReflectionKey`,
+  :func:`EnableReflectionKey`, and :func:`QueryReflectionKey` were also
+  tested and documented.
+  (Implemented by Brian Curtin: :issue:`7347`.)
+
 * The new :cfunc:`_beginthreadex` API is used to start threads, and
   the native thread-local storage functions are now used.
-  (Contributed by Kristjan Valur Jonsson; :issue:`3582`.)
+  (Contributed by Kristj??n Valur J??nsson; :issue:`3582`.)
 
 * The :func:`os.kill` function now works on Windows.  The signal value
   can be the constants :const:`CTRL_C_EVENT`,
-  :const:`CTRL_BREAK_EVENT`, or any integer.  The Control-C and
-  Control-Break keystroke events can be sent to subprocesses; any
-  other value will use the :cfunc:`TerminateProcess` API.
-  (Contributed by Miki Tebeka; :issue:`1220212`.)
+  :const:`CTRL_BREAK_EVENT`, or any integer.  The first two constants
+  will send Control-C and Control-Break keystroke events to
+  subprocesses; any other value will use the :cfunc:`TerminateProcess`
+  API.  (Contributed by Miki Tebeka; :issue:`1220212`.)
 
 * The :func:`os.listdir` function now correctly fails
   for an empty path.  (Fixed by Hirokazu Yamamoto; :issue:`5913`.)
@@ -1700,18 +2298,29 @@
   installation and a user-installed copy of the same version.
   (Changed by Ronald Oussoren; :issue:`4865`.)
 
+Port-Specific Changes: FreeBSD
+-----------------------------------
+
+* FreeBSD 7.1's :const:`SO_SETFIB` constant, used with
+  :func:`~socket.getsockopt`/:func:`~socket.setsockopt` to select an
+  alternate routing table, is now available in the :mod:`socket`
+  module.  (Added by Kyle VanderBeek; :issue:`8235`.)
 
 Other Changes and Fixes
 =======================
 
 * Two benchmark scripts, :file:`iobench` and :file:`ccbench`, were
   added to the :file:`Tools` directory.  :file:`iobench` measures the
-  speed of built-in file I/O objects (as returned by :func:`open`)
+  speed of the built-in file I/O objects returned by :func:`open`
   while performing various operations, and :file:`ccbench` is a
   concurrency benchmark that tries to measure computing throughput,
   thread switching latency, and IO processing bandwidth when
   performing several tasks using a varying number of threads.
 
+* The :file:`Tools/i18n/msgfmt.py` script now understands plural
+  forms in :file:`.po` files.  (Fixed by Martin von L??wis;
+  :issue:`5464`.)
+
 * When importing a module from a :file:`.pyc` or :file:`.pyo` file
   with an existing :file:`.py` counterpart, the :attr:`co_filename`
   attributes of the resulting code objects are overwritten when the
@@ -1747,14 +2356,13 @@
 This section lists previously described changes and other bugfixes
 that may require changes to your code:
 
-* When using :class:`Decimal` instances with a string's
-  :meth:`format` method, the default alignment was previously
-  left-alignment.  This has been changed to right-alignment, which might
-  change the output of your programs.
-  (Changed by Mark Dickinson; :issue:`6857`.)
+* The :func:`range` function processes its arguments more
+  consistently; it will now call :meth:`__int__` on non-float,
+  non-integer arguments that are supplied to it.  (Fixed by Alexander
+  Belopolsky; :issue:`1533`.)
 
-  Another :meth:`format`-related change: the default precision used
-  for floating-point and complex numbers was changed from 6 decimal
+* The string :meth:`format` method changed the default precision used
+  for floating-point and complex numbers from 6 decimal
   places to 12, which matches the precision used by :func:`str`.
   (Changed by Eric Smith; :issue:`5920`.)
 
@@ -1764,18 +2372,79 @@
   affects new-style classes (derived from :class:`object`) and C extension
   types.  (:issue:`6101`.)
 
-* The :meth:`readline` method of :class:`StringIO` objects now does
-  nothing when a negative length is requested, as other file-like
-  objects do.  (:issue:`7348`).
+* Due to a bug in Python 2.6, the *exc_value* parameter to
+  :meth:`__exit__` methods was often the string representation of the
+  exception, not an instance.  This was fixed in 2.7, so *exc_value*
+  will be an instance as expected.  (Fixed by Florent Xicluna;
+  :issue:`7853`.)
+
+* When a restricted set of attributes were set using ``__slots__``,
+  deleting an unset attribute would not raise :exc:`AttributeError`
+  as you would expect.  Fixed by Benjamin Peterson; :issue:`7604`.)
 
 In the standard library:
 
+* Operations with :class:`datetime` instances that resulted in a year
+  falling outside the supported range didn't always raise
+  :exc:`OverflowError`.  Such errors are now checked more carefully
+  and will now raise the exception. (Reported by Mark Leander, patch
+  by Anand B. Pillai and Alexander Belopolsky; :issue:`7150`.)
+
+* When using :class:`Decimal` instances with a string's
+  :meth:`format` method, the default alignment was previously
+  left-alignment.  This has been changed to right-alignment, which might
+  change the output of your programs.
+  (Changed by Mark Dickinson; :issue:`6857`.)
+
+  Comparisons involving a signaling NaN value (or ``sNAN``) now signal
+  :const:`InvalidOperation` instead of silently returning a true or
+  false value depending on the comparison operator.  Quiet NaN values
+  (or ``NaN``) are now hashable.  (Fixed by Mark Dickinson;
+  :issue:`7279`.)
+
 * The ElementTree library, :mod:`xml.etree`, no longer escapes
   ampersands and angle brackets when outputting an XML processing
   instruction (which looks like `<?xml-stylesheet href="#style1"?>`)
   or comment (which looks like `<!-- comment -->`).
   (Patch by Neil Muller; :issue:`2746`.)
 
+* The :meth:`readline` method of :class:`StringIO` objects now does
+  nothing when a negative length is requested, as other file-like
+  objects do.  (:issue:`7348`).
+
+* The :mod:`syslog` module will now use the value of ``sys.argv[0]`` as the
+  identifier instead of the previous default value of ``'python'``.
+  (Changed by Sean Reifschneider; :issue:`8451`.)
+
+* The :mod:`tarfile` module's default error handling has changed, to
+  no longer suppress fatal errors.  The default error level was previously 0,
+  which meant that errors would only result in a message being written to the
+  debug log, but because the debug log is not activated by default,
+  these errors go unnoticed.  The default error level is now 1,
+  which raises an exception if there's an error.
+  (Changed by Lars Gust??bel; :issue:`7357`.)
+
+* The :mod:`urlparse` module's :func:`~urlparse.urlsplit` now handles
+  unknown URL schemes in a fashion compliant with :rfc:`3986`: if the
+  URL is of the form ``"<something>://..."``, the text before the
+  ``://`` is treated as the scheme, even if it's a made-up scheme that
+  the module doesn't know about.  This change may break code that
+  worked around the old behaviour.  For example, Python 2.6.4 or 2.5
+  will return the following:
+
+    >>> import urlparse
+    >>> urlparse.urlsplit('invented://host/filename?query')
+    ('invented', '', '//host/filename?query', '', '')
+
+  Python 2.7 (and Python 2.6.5) will return:
+
+    >>> import urlparse
+    >>> urlparse.urlsplit('invented://host/filename?query')
+    ('invented', 'host', '/filename?query', '', '')
+
+  (Python 2.7 actually produces slightly different output, since it
+  returns a named tuple instead of a standard tuple.)
+
 For C extensions:
 
 * C extensions that use integer format codes with the ``PyArg_Parse*``
@@ -1786,6 +2455,14 @@
   :cfunc:`PyOS_ascii_strtod` and :cfunc:`PyOS_ascii_atof` functions,
   which are now deprecated.
 
+For applications that embed Python:
+
+* The :cfunc:`PySys_SetArgvEx` function was added, letting
+  applications close a security hole when the existing
+  :cfunc:`PySys_SetArgv` function was used.  Check whether you're
+  calling :cfunc:`PySys_SetArgv` and carefully consider whether the
+  application should be using :cfunc:`PySys_SetArgvEx` with
+  *updatepath* set to false.
 
 .. ======================================================================
 
@@ -1797,5 +2474,6 @@
 
 The author would like to thank the following people for offering
 suggestions, corrections and assistance with various drafts of this
-article: Ryan Lovett, R. David Murray, Hugh Secker-Walker.
+article: Nick Coghlan, Philip Jenvey, Ryan Lovett, R. David Murray,
+Hugh Secker-Walker.
 

Modified: python/branches/py3k-dtoa/Doc/whatsnew/3.0.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/whatsnew/3.0.rst	(original)
+++ python/branches/py3k-dtoa/Doc/whatsnew/3.0.rst	Tue Jul  6 17:55:27 2010
@@ -798,6 +798,15 @@
 * A new built-in function :func:`next` was added to call the
   :meth:`__next__` method on an object.
 
+* The :func:`round` function rounding strategy and return type have
+  changed.  Exact halfway cases are now rounded to the nearest even
+  result instead of away from zero.  (For example, ``round(2.5)`` now
+  returns ``2`` rather than ``3``.)  :func:`round(x[, n])` now
+  delegates to ``x.__round__([n])`` instead of always returning a
+  float.  It generally returns an integer when called with a single
+  argument and a value of the same type as ``x`` when called with two
+  arguments.
+
 * Moved :func:`intern` to :func:`sys.intern`.
 
 * Removed: :func:`apply`.  Instead of ``apply(f, args)`` use

Modified: python/branches/py3k-dtoa/Doc/whatsnew/3.2.rst
==============================================================================
--- python/branches/py3k-dtoa/Doc/whatsnew/3.2.rst	(original)
+++ python/branches/py3k-dtoa/Doc/whatsnew/3.2.rst	Tue Jul  6 17:55:27 2010
@@ -173,7 +173,8 @@
 
 * bytearray objects cannot be used anymore as filenames: convert them to bytes
 
-* "t#" format of PyArg_Parse*() functions has been removed: use "s#" or "s*"
-  instead
+* PyArg_Parse*() functions:
+
+  * "t#" format has been removed: use "s#" or "s*" instead
+  * "w" and "w#" formats has been removed: use "w*" instead
 
-* Stub

Modified: python/branches/py3k-dtoa/Grammar/Grammar
==============================================================================
--- python/branches/py3k-dtoa/Grammar/Grammar	(original)
+++ python/branches/py3k-dtoa/Grammar/Grammar	Tue Jul  6 17:55:27 2010
@@ -11,9 +11,9 @@
 # "How to Change Python's Grammar"
 
 # Start symbols for the grammar:
-#	single_input is a single interactive statement;
-#	file_input is a module or sequence of commands read from an input file;
-#	eval_input is the input for the eval() and input() functions.
+#       single_input is a single interactive statement;
+#       file_input is a module or sequence of commands read from an input file;
+#       eval_input is the input for the eval() and input() functions.
 # NB: compound_stmt in single_input is followed by extra NEWLINE!
 single_input: NEWLINE | simple_stmt | compound_stmt NEWLINE
 file_input: (NEWLINE | stmt)* ENDMARKER
@@ -71,9 +71,9 @@
 for_stmt: 'for' exprlist 'in' testlist ':' suite ['else' ':' suite]
 try_stmt: ('try' ':' suite
            ((except_clause ':' suite)+
-	    ['else' ':' suite]
-	    ['finally' ':' suite] |
-	   'finally' ':' suite))
+            ['else' ':' suite]
+            ['finally' ':' suite] |
+           'finally' ':' suite))
 with_stmt: 'with' with_item (',' with_item)*  ':' suite
 with_item: test ['as' expr]
 # NB compile.c makes sure that the default except clause is last

Modified: python/branches/py3k-dtoa/Include/longobject.h
==============================================================================
--- python/branches/py3k-dtoa/Include/longobject.h	(original)
+++ python/branches/py3k-dtoa/Include/longobject.h	Tue Jul  6 17:55:27 2010
@@ -58,7 +58,7 @@
 #error "sizeof(pid_t) is neither sizeof(int), sizeof(long) or sizeof(long long)"
 #endif /* SIZEOF_PID_T */
 
-/* For use by intobject.c only */
+/* Used by Python/mystrtoul.c. */
 PyAPI_DATA(unsigned char) _PyLong_DigitValue[256];
 
 /* _PyLong_Frexp returns a double x and an exponent e such that the

Modified: python/branches/py3k-dtoa/Include/pyctype.h
==============================================================================
--- python/branches/py3k-dtoa/Include/pyctype.h	(original)
+++ python/branches/py3k-dtoa/Include/pyctype.h	Tue Jul  6 17:55:27 2010
@@ -11,6 +11,9 @@
 
 extern const unsigned int _Py_ctype_table[256];
 
+/* Unlike their C counterparts, the following macros are not meant to
+ * handle an int with any of the values [EOF, 0-UCHAR_MAX]. The argument
+ * must be a signed/unsigned char. */
 #define Py_ISLOWER(c)  (_Py_ctype_table[Py_CHARMASK(c)] & PY_CTF_LOWER)
 #define Py_ISUPPER(c)  (_Py_ctype_table[Py_CHARMASK(c)] & PY_CTF_UPPER)
 #define Py_ISALPHA(c)  (_Py_ctype_table[Py_CHARMASK(c)] & PY_CTF_ALPHA)

Modified: python/branches/py3k-dtoa/Include/traceback.h
==============================================================================
--- python/branches/py3k-dtoa/Include/traceback.h	(original)
+++ python/branches/py3k-dtoa/Include/traceback.h	Tue Jul  6 17:55:27 2010
@@ -10,16 +10,16 @@
 /* Traceback interface */
 
 typedef struct _traceback {
-	PyObject_HEAD
-	struct _traceback *tb_next;
-	struct _frame *tb_frame;
-	int tb_lasti;
-	int tb_lineno;
+    PyObject_HEAD
+    struct _traceback *tb_next;
+    struct _frame *tb_frame;
+    int tb_lasti;
+    int tb_lineno;
 } PyTracebackObject;
 
 PyAPI_FUNC(int) PyTraceBack_Here(struct _frame *);
 PyAPI_FUNC(int) PyTraceBack_Print(PyObject *, PyObject *);
-PyAPI_FUNC(int) _Py_DisplaySourceLine(PyObject *, const char *, int, int);
+PyAPI_FUNC(int) _Py_DisplaySourceLine(PyObject *, PyObject *, int, int);
 
 /* Reveal traceback type so we can typecheck traceback objects */
 PyAPI_DATA(PyTypeObject) PyTraceBack_Type;

Modified: python/branches/py3k-dtoa/Include/unicodeobject.h
==============================================================================
--- python/branches/py3k-dtoa/Include/unicodeobject.h	(original)
+++ python/branches/py3k-dtoa/Include/unicodeobject.h	Tue Jul  6 17:55:27 2010
@@ -498,14 +498,14 @@
     Py_ssize_t size             /* size of buffer */
     );
 
-/* Similar to PyUnicode_FromUnicode(), but u points to Latin-1 encoded bytes */
+/* Similar to PyUnicode_FromUnicode(), but u points to UTF-8 encoded bytes */
 PyAPI_FUNC(PyObject*) PyUnicode_FromStringAndSize(
     const char *u,        /* char buffer */
     Py_ssize_t size       /* size of buffer */
     );
 
 /* Similar to PyUnicode_FromUnicode(), but u points to null-terminated
-   Latin-1 encoded bytes */
+   UTF-8 encoded bytes */
 PyAPI_FUNC(PyObject*) PyUnicode_FromString(
     const char *u        /* string */
     );
@@ -550,7 +550,7 @@
 
    Coercion is done in the following way:
 
-   1. String and other char buffer compatible objects are decoded
+   1. bytes, bytearray and other char buffer compatible objects are decoded
       under the assumptions that they contain data using the current
       default encoding. Decoding is done in "strict" mode.
 
@@ -574,7 +574,7 @@
    Unicode objects are passed back as-is (subclasses are converted to
    true Unicode objects), all other objects are delegated to
    PyUnicode_FromEncodedObject(obj, NULL, "strict") which results in
-   using the default encoding as basis for decoding the object.
+   using UTF-8 encoding as basis for decoding the object.
 
    The API returns NULL in case of an error. The caller is responsible
    for decref'ing the returned objects.
@@ -606,7 +606,7 @@
 
 #ifdef HAVE_WCHAR_H
 
-/* Create a Unicode Object from the whcar_t buffer w of the given
+/* Create a Unicode Object from the wchar_t buffer w of the given
    size.
 
    The buffer is copied into the new object. */
@@ -665,7 +665,7 @@
    parameters encoding and errors have the same semantics as the ones
    of the builtin unicode() API.
 
-   Setting encoding to NULL causes the default encoding to be used.
+   Setting encoding to NULL causes the default encoding (UTF-8) to be used.
 
    Error handling is set by errors which may also be set to NULL
    meaning to use the default handling defined for the codec. Default

Modified: python/branches/py3k-dtoa/Lib/_strptime.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/_strptime.py	(original)
+++ python/branches/py3k-dtoa/Lib/_strptime.py	Tue Jul  6 17:55:27 2010
@@ -16,7 +16,9 @@
 from re import compile as re_compile
 from re import IGNORECASE, ASCII
 from re import escape as re_escape
-from datetime import date as datetime_date
+from datetime import (date as datetime_date,
+                      timedelta as datetime_timedelta,
+                      timezone as datetime_timezone)
 try:
     from _thread import allocate_lock as _thread_allocate_lock
 except:
@@ -204,6 +206,7 @@
             #XXX: Does 'Y' need to worry about having less or more than
             #     4 digits?
             'Y': r"(?P<Y>\d\d\d\d)",
+            'z': r"(?P<z>[+-]\d\d[0-5]\d)",
             'A': self.__seqToRE(self.locale_time.f_weekday, 'A'),
             'a': self.__seqToRE(self.locale_time.a_weekday, 'a'),
             'B': self.__seqToRE(self.locale_time.f_month[1:], 'B'),
@@ -293,7 +296,9 @@
 
 
 def _strptime(data_string, format="%a %b %d %H:%M:%S %Y"):
-    """Return a time struct based on the input string and the format string."""
+    """Return a 2-tuple consisting of a time struct and an int containing
+    the number of microseconds based on the input string and the
+    format string."""
 
     for index, arg in enumerate([data_string, format]):
         if not isinstance(arg, str):
@@ -333,10 +338,12 @@
     if len(data_string) != found.end():
         raise ValueError("unconverted data remains: %s" %
                           data_string[found.end():])
+
     year = 1900
     month = day = 1
     hour = minute = second = fraction = 0
     tz = -1
+    tzoffset = None
     # Default to -1 to signify that values not known; not critical to have,
     # though
     week_of_year = -1
@@ -417,6 +424,11 @@
             else:
                 # W starts week on Monday.
                 week_of_year_start = 0
+        elif group_key == 'z':
+            z = found_dict['z']
+            tzoffset = int(z[1:3]) * 60 + int(z[3:5])
+            if z.startswith("-"):
+                tzoffset = -tzoffset
         elif group_key == 'Z':
             # Since -1 is default value only need to worry about setting tz if
             # it can be something other than -1.
@@ -453,9 +465,35 @@
         day = datetime_result.day
     if weekday == -1:
         weekday = datetime_date(year, month, day).weekday()
-    return (time.struct_time((year, month, day,
-                              hour, minute, second,
-                              weekday, julian, tz)), fraction)
+    # Add timezone info
+    tzname = found_dict.get("Z")
+    if tzoffset is not None:
+        gmtoff = tzoffset * 60
+    else:
+        gmtoff = None
+
+    return (year, month, day,
+            hour, minute, second,
+            weekday, julian, tz, gmtoff, tzname), fraction
 
 def _strptime_time(data_string, format="%a %b %d %H:%M:%S %Y"):
-    return _strptime(data_string, format)[0]
+    """Return a time struct based on the input string and the
+    format string."""
+    tt = _strptime(data_string, format)[0]
+    return time.struct_time(tt[:9])
+
+def _strptime_datetime(class_, data_string, format="%a %b %d %H:%M:%S %Y"):
+    """Return a class_ instance based on the input string and the
+    format string."""
+    tt, fraction = _strptime(data_string, format)
+    gmtoff, tzname = tt[-2:]
+    args = tt[:6] + (fraction,)
+    if gmtoff is not None:
+        tzdelta = datetime_timedelta(seconds=gmtoff)
+        if tzname:
+            tz = datetime_timezone(tzdelta, tzname)
+        else:
+            tz = datetime_timezone(tzdelta)
+        args += (tz,)
+
+    return class_(*args)

Modified: python/branches/py3k-dtoa/Lib/contextlib.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/contextlib.py	(original)
+++ python/branches/py3k-dtoa/Lib/contextlib.py	Tue Jul  6 17:55:27 2010
@@ -4,9 +4,20 @@
 from functools import wraps
 from warnings import warn
 
-__all__ = ["contextmanager", "closing"]
+__all__ = ["contextmanager", "closing", "ContextDecorator"]
 
-class GeneratorContextManager(object):
+
+class ContextDecorator(object):
+    "A base class or mixin that enables context managers to work as decorators."
+    def __call__(self, func):
+        @wraps(func)
+        def inner(*args, **kwds):
+            with self:
+                return func(*args, **kwds)
+        return inner
+
+
+class GeneratorContextManager(ContextDecorator):
     """Helper for @contextmanager decorator."""
 
     def __init__(self, gen):

Modified: python/branches/py3k-dtoa/Lib/ctypes/__init__.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/ctypes/__init__.py	(original)
+++ python/branches/py3k-dtoa/Lib/ctypes/__init__.py	Tue Jul  6 17:55:27 2010
@@ -265,7 +265,7 @@
     pass
 else:
     if _os.name in ("nt", "ce"):
-        set_conversion_mode("mbcs", "ignore")
+        set_conversion_mode("mbcs", "strict")
     else:
         set_conversion_mode("ascii", "strict")
 

Modified: python/branches/py3k-dtoa/Lib/ctypes/test/test_callbacks.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/ctypes/test/test_callbacks.py	(original)
+++ python/branches/py3k-dtoa/Lib/ctypes/test/test_callbacks.py	Tue Jul  6 17:55:27 2010
@@ -166,6 +166,41 @@
 
         self.assertTrue(diff < 0.01, "%s not less than 0.01" % diff)
 
+    def test_issue_8959_a(self):
+        from ctypes.util import find_library
+        libc_path = find_library("c")
+        if not libc_path:
+            return # cannot test
+        libc = CDLL(libc_path)
+
+        @CFUNCTYPE(c_int, POINTER(c_int), POINTER(c_int))
+        def cmp_func(a, b):
+            return a[0] - b[0]
+
+        array = (c_int * 5)(5, 1, 99, 7, 33)
+
+        libc.qsort(array, len(array), sizeof(c_int), cmp_func)
+        self.assertEqual(array[:], [1, 5, 7, 33, 99])
+
+    try:
+        WINFUNCTYPE
+    except NameError:
+        pass
+    else:
+        def test_issue_8959_b(self):
+            from ctypes.wintypes import BOOL, HWND, LPARAM
+            global windowCount
+            windowCount = 0
+
+            @WINFUNCTYPE(BOOL, HWND, LPARAM)
+            def EnumWindowsCallbackFunc(hwnd, lParam):
+                global windowCount
+                windowCount += 1
+                return True #Allow windows to keep enumerating
+
+            windll.user32.EnumWindows(EnumWindowsCallbackFunc, 0)
+            self.assertFalse(windowCount == 0)
+
 ################################################################
 
 if __name__ == '__main__':

Modified: python/branches/py3k-dtoa/Lib/ctypes/test/test_win32.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/ctypes/test/test_win32.py	(original)
+++ python/branches/py3k-dtoa/Lib/ctypes/test/test_win32.py	Tue Jul  6 17:55:27 2010
@@ -6,6 +6,32 @@
 
 import _ctypes_test
 
+if sys.platform == "win32" and sizeof(c_void_p) == sizeof(c_int):
+    # Only windows 32-bit has different calling conventions.
+
+    class WindowsTestCase(unittest.TestCase):
+        def test_callconv_1(self):
+            # Testing stdcall function
+
+            IsWindow = windll.user32.IsWindow
+            # ValueError: Procedure probably called with not enough arguments (4 bytes missing)
+            self.assertRaises(ValueError, IsWindow)
+
+            # This one should succeeed...
+            self.assertEqual(0, IsWindow(0))
+
+            # ValueError: Procedure probably called with too many arguments (8 bytes in excess)
+            self.assertRaises(ValueError, IsWindow, 0, 0, 0)
+
+        def test_callconv_2(self):
+            # Calling stdcall function as cdecl
+
+            IsWindow = cdll.user32.IsWindow
+
+            # ValueError: Procedure called with not enough arguments (4 bytes missing)
+            # or wrong calling convention
+            self.assertRaises(ValueError, IsWindow, None)
+
 if sys.platform == "win32":
     class FunctionCallTestCase(unittest.TestCase):
 

Modified: python/branches/py3k-dtoa/Lib/dis.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/dis.py	(original)
+++ python/branches/py3k-dtoa/Lib/dis.py	Tue Jul  6 17:55:27 2010
@@ -12,6 +12,22 @@
 
 _have_code = (types.MethodType, types.FunctionType, types.CodeType, type)
 
+def _try_compile(source, name):
+    """Attempts to compile the given source, first as an expression and
+       then as a statement if the first approach fails.
+
+       Utility function to accept strings in functions that otherwise
+       expect code objects
+    """
+    # ncoghlan: currently only used by dis(), but plan to add an
+    # equivalent for show_code() as well (but one that returns a
+    # string rather than printing directly to the console)
+    try:
+        c = compile(source, name, 'eval')
+    except SyntaxError:
+        c = compile(source, name, 'exec')
+    return c
+
 def dis(x=None):
     """Disassemble classes, methods, functions, or code.
 
@@ -38,7 +54,9 @@
     elif hasattr(x, 'co_code'):
         disassemble(x)
     elif isinstance(x, (bytes, bytearray)):
-        disassemble_string(x)
+        _disassemble_bytes(x)
+    elif isinstance(x, str):
+        _disassemble_str(x)
     else:
         raise TypeError("don't know how to disassemble %s objects" %
                         type(x).__name__)
@@ -157,7 +175,7 @@
                 print('(' + free[oparg] + ')', end=' ')
         print()
 
-def disassemble_string(code, lasti=-1, varnames=None, names=None,
+def _disassemble_bytes(code, lasti=-1, varnames=None, names=None,
                        constants=None):
     labels = findlabels(code)
     n = len(code)
@@ -196,6 +214,10 @@
                 print('(' + cmp_op[oparg] + ')', end=' ')
         print()
 
+def _disassemble_str(source):
+    """Compile the source string, then disassemble the code object."""
+    disassemble(_try_compile(source, '<dis>'))
+
 disco = disassemble                     # XXX For backwards compatibility
 
 def findlabels(code):

Modified: python/branches/py3k-dtoa/Lib/distutils/msvc9compiler.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/distutils/msvc9compiler.py	(original)
+++ python/branches/py3k-dtoa/Lib/distutils/msvc9compiler.py	Tue Jul  6 17:55:27 2010
@@ -37,9 +37,18 @@
          winreg.HKEY_LOCAL_MACHINE,
          winreg.HKEY_CLASSES_ROOT)
 
-VS_BASE = r"Software\Microsoft\VisualStudio\%0.1f"
-WINSDK_BASE = r"Software\Microsoft\Microsoft SDKs\Windows"
-NET_BASE = r"Software\Microsoft\.NETFramework"
+NATIVE_WIN64 = (sys.platform == 'win32' and sys.maxsize > 2**32)
+if NATIVE_WIN64:
+    # Visual C++ is a 32-bit application, so we need to look in
+    # the corresponding registry branch, if we're running a
+    # 64-bit Python on Win64
+    VS_BASE = r"Software\Wow6432Node\Microsoft\VisualStudio\%0.1f"
+    WINSDK_BASE = r"Software\Wow6432Node\Microsoft\Microsoft SDKs\Windows"
+    NET_BASE = r"Software\Wow6432Node\Microsoft\.NETFramework"
+else:
+    VS_BASE = r"Software\Microsoft\VisualStudio\%0.1f"
+    WINSDK_BASE = r"Software\Microsoft\Microsoft SDKs\Windows"
+    NET_BASE = r"Software\Microsoft\.NETFramework"
 
 # A map keyed by get_platform() return values to values accepted by
 # 'vcvarsall.bat'.  Note a cross-compile may combine these (eg, 'x86_amd64' is

Modified: python/branches/py3k-dtoa/Lib/distutils/unixccompiler.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/distutils/unixccompiler.py	(original)
+++ python/branches/py3k-dtoa/Lib/distutils/unixccompiler.py	Tue Jul  6 17:55:27 2010
@@ -340,7 +340,9 @@
             static = os.path.join(dir, static_f)
 
             if sys.platform == 'darwin' and (
-                    dir.startswith('/System/') or dir.startswith('/usr/')):
+                dir.startswith('/System/') or (
+                dir.startswith('/usr/') and not dir.startswith('/usr/local/'))):
+
                 shared = os.path.join(sysroot, dir[1:], shared_f)
                 dylib = os.path.join(sysroot, dir[1:], dylib_f)
                 static = os.path.join(sysroot, dir[1:], static_f)

Modified: python/branches/py3k-dtoa/Lib/encodings/aliases.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/encodings/aliases.py	(original)
+++ python/branches/py3k-dtoa/Lib/encodings/aliases.py	Tue Jul  6 17:55:27 2010
@@ -146,6 +146,11 @@
     'csibm857'           : 'cp857',
     'ibm857'             : 'cp857',
 
+    # cp858 codec
+    '858'                : 'cp858',
+    'csibm858'           : 'cp858',
+    'ibm858'             : 'cp858',
+
     # cp860 codec
     '860'                : 'cp860',
     'csibm860'           : 'cp860',

Modified: python/branches/py3k-dtoa/Lib/heapq.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/heapq.py	(original)
+++ python/branches/py3k-dtoa/Lib/heapq.py	Tue Jul  6 17:55:27 2010
@@ -1,5 +1,3 @@
-# -*- coding: latin-1 -*-
-
 """Heap queue algorithm (a.k.a. priority queue).
 
 Heaps are arrays for which a[k] <= a[2*k+1] and a[k] <= a[2*k+2] for
@@ -34,7 +32,7 @@
 
 __about__ = """Heap queues
 
-[explanation by Fran?ois Pinard]
+[explanation by Fran??ois Pinard]
 
 Heaps are arrays for which a[k] <= a[2*k+1] and a[k] <= a[2*k+2] for
 all k, counting elements from 0.  For the sake of comparison,

Modified: python/branches/py3k-dtoa/Lib/http/cookiejar.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/http/cookiejar.py	(original)
+++ python/branches/py3k-dtoa/Lib/http/cookiejar.py	Tue Jul  6 17:55:27 2010
@@ -436,6 +436,13 @@
         if attr: headers.append("; ".join(attr))
     return ", ".join(headers)
 
+def strip_quotes(text):
+    if text.startswith('"'):
+        text = text[1:]
+    if text.endswith('"'):
+        text = text[:-1]
+    return text
+
 def parse_ns_headers(ns_headers):
     """Ad-hoc parser for Netscape protocol cookie-attributes.
 
@@ -453,7 +460,7 @@
     """
     known_attrs = ("expires", "domain", "path", "secure",
                    # RFC 2109 attrs (may turn up in Netscape cookies, too)
-                   "port", "max-age")
+                   "version", "port", "max-age")
 
     result = []
     for ns_header in ns_headers:
@@ -473,12 +480,11 @@
                     k = lc
                 if k == "version":
                     # This is an RFC 2109 cookie.
+                    v = strip_quotes(v)
                     version_set = True
                 if k == "expires":
                     # convert expires date to seconds since epoch
-                    if v.startswith('"'): v = v[1:]
-                    if v.endswith('"'): v = v[:-1]
-                    v = http2time(v)  # None if invalid
+                    v = http2time(strip_quotes(v))  # None if invalid
             pairs.append((k, v))
 
         if pairs:
@@ -1446,7 +1452,11 @@
 
         # set the easy defaults
         version = standard.get("version", None)
-        if version is not None: version = int(version)
+        if version is not None:
+            try:
+                version = int(version)
+            except ValueError:
+                return None  # invalid version, ignore cookie
         secure = standard.get("secure", False)
         # (discard is also set if expires is Absent)
         discard = standard.get("discard", False)

Modified: python/branches/py3k-dtoa/Lib/idlelib/textView.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/idlelib/textView.py	(original)
+++ python/branches/py3k-dtoa/Lib/idlelib/textView.py	Tue Jul  6 17:55:27 2010
@@ -62,11 +62,7 @@
 
 def view_file(parent, title, filename, encoding=None):
     try:
-        if encoding:
-            import codecs
-            textFile = codecs.open(filename, 'r')
-        else:
-            textFile = open(filename, 'r')
+        textFile = open(filename, 'r', encoding=encoding)
     except IOError:
         import tkinter.messagebox as tkMessageBox
         tkMessageBox.showerror(title='File Load Error',

Modified: python/branches/py3k-dtoa/Lib/importlib/__init__.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/importlib/__init__.py	(original)
+++ python/branches/py3k-dtoa/Lib/importlib/__init__.py	Tue Jul  6 17:55:27 2010
@@ -36,7 +36,7 @@
     """
     if 'PYTHONCASEOK' in os.environ:
         return True
-    elif check in os.listdir(directory):
+    elif check in os.listdir(directory if directory else os.getcwd()):
         return True
     return False
 

Modified: python/branches/py3k-dtoa/Lib/importlib/_bootstrap.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/importlib/_bootstrap.py	(original)
+++ python/branches/py3k-dtoa/Lib/importlib/_bootstrap.py	Tue Jul  6 17:55:27 2010
@@ -22,7 +22,7 @@
 def _path_join(*args):
     """Replacement for os.path.join."""
     return path_sep.join(x[:-len(path_sep)] if x.endswith(path_sep) else x
-                            for x in args)
+                            for x in args if x)
 
 
 def _path_exists(path):
@@ -53,6 +53,8 @@
 # XXX Could also expose Modules/getpath.c:isdir()
 def _path_isdir(path):
     """Replacement for os.path.isdir."""
+    if not path:
+        path = _os.getcwd()
     return _path_is_mode_type(path, 0o040000)
 
 
@@ -99,6 +101,8 @@
     new.__dict__.update(old.__dict__)
 
 
+code_type = type(_wrap.__code__)
+
 # Finder/loader utility code ##################################################
 
 def set_package(fxn):
@@ -138,7 +142,7 @@
     the second argument.
 
     """
-    def decorated(self, fullname):
+    def decorated(self, fullname, *args, **kwargs):
         module = sys.modules.get(fullname)
         is_reload = bool(module)
         if not is_reload:
@@ -148,7 +152,7 @@
             module = imp.new_module(fullname)
             sys.modules[fullname] = module
         try:
-            return fxn(self, module)
+            return fxn(self, module, *args, **kwargs)
         except:
             if not is_reload:
                 del sys.modules[fullname]
@@ -301,258 +305,227 @@
         return imp.is_frozen_package(fullname)
 
 
-class PyLoader:
+class _LoaderBasics:
 
-    """Loader base class for Python source code.
+    """Base class of common code needed by both SourceLoader and
+    _SourcelessFileLoader."""
 
-    Subclasses need to implement the methods:
+    def is_package(self, fullname):
+        """Concrete implementation of InspectLoader.is_package by checking if
+        the path returned by get_filename has a filename of '__init__.py'."""
+        filename = self.get_filename(fullname).rpartition(path_sep)[2]
+        return filename.rsplit('.', 1)[0] == '__init__'
+
+    def _bytes_from_bytecode(self, fullname, data, source_mtime):
+        """Return the marshalled bytes from bytecode, verifying the magic
+        number and timestamp alon the way.
 
-    - source_path
-    - get_data
-    - is_package
+        If source_mtime is None then skip the timestamp check.
 
-    """
+        """
+        magic = data[:4]
+        raw_timestamp = data[4:8]
+        if len(magic) != 4 or magic != imp.get_magic():
+            raise ImportError("bad magic number in {}".format(fullname))
+        elif len(raw_timestamp) != 4:
+            raise EOFError("bad timestamp in {}".format(fullname))
+        elif source_mtime is not None:
+            if marshal._r_long(raw_timestamp) != source_mtime:
+                raise ImportError("bytecode is stale for {}".format(fullname))
+        # Can't return the code object as errors from marshal loading need to
+        # propagate even when source is available.
+        return data[8:]
 
     @module_for_loader
-    def load_module(self, module):
-        """Initialize the module."""
+    def _load_module(self, module, *, sourceless=False):
+        """Helper for load_module able to handle either source or sourceless
+        loading."""
         name = module.__name__
-        code_object = self.get_code(module.__name__)
+        code_object = self.get_code(name)
         module.__file__ = self.get_filename(name)
+        if not sourceless:
+            module.__cached__ = imp.cache_from_source(module.__file__)
+        else:
+            module.__cached__ = module.__file__
+        module.__package__ = name
         if self.is_package(name):
-            module.__path__  = [module.__file__.rsplit(path_sep, 1)[0]]
-        module.__package__ = module.__name__
-        if not hasattr(module, '__path__'):
+            module.__path__ = [module.__file__.rsplit(path_sep, 1)[0]]
+        else:
             module.__package__ = module.__package__.rpartition('.')[0]
         module.__loader__ = self
         exec(code_object, module.__dict__)
         return module
 
-    def get_filename(self, fullname):
-        """Return the path to the source file, else raise ImportError."""
-        path = self.source_path(fullname)
-        if path is not None:
-            return path
-        else:
-            raise ImportError("no source path available for "
-                                "{0!r}".format(fullname))
 
-    def get_code(self, fullname):
-        """Get a code object from source."""
-        source_path = self.source_path(fullname)
-        if source_path is None:
-            message = "a source path must exist to load {0}".format(fullname)
-            raise ImportError(message)
-        source = self.get_data(source_path)
-        return compile(source, source_path, 'exec', dont_inherit=True)
+class SourceLoader(_LoaderBasics):
 
-    # Never use in implementing import! Imports code within the method.
-    def get_source(self, fullname):
-        """Return the source code for a module.
+    def path_mtime(self, path:str) -> int:
+        """Optional method that returns the modification time for the specified
+        path.
 
-        self.source_path() and self.get_data() are used to implement this
-        method.
+        Implementing this method allows the loader to read bytecode files.
 
         """
-        path = self.source_path(fullname)
-        if path is None:
-            return None
-        try:
-            source_bytes = self.get_data(path)
-        except IOError:
-            return ImportError("source not available through get_data()")
-        import io
-        import tokenize
-        encoding = tokenize.detect_encoding(io.BytesIO(source_bytes).readline)
-        return source_bytes.decode(encoding[0])
+        raise NotImplementedError
 
+    def set_data(self, path:str, data:bytes) -> None:
+        """Optional method which writes data to a file path.
 
-class PyPycLoader(PyLoader):
+        Implementing this method allows for the writing of bytecode files.
 
-    """Loader base class for Python source and bytecode.
+        """
+        raise NotImplementedError
 
-    Requires implementing the methods needed for PyLoader as well as
-    source_mtime, bytecode_path, and write_bytecode.
 
-    """
-
-    def get_filename(self, fullname):
-        """Return the source or bytecode file path."""
-        path = self.source_path(fullname)
-        if path is not None:
-            return path
-        path = self.bytecode_path(fullname)
-        if path is not None:
-            return path
-        raise ImportError("no source or bytecode path available for "
-                            "{0!r}".format(fullname))
+    def get_source(self, fullname):
+        """Concrete implementation of InspectLoader.get_source."""
+        import tokenize
+        path = self.get_filename(fullname)
+        try:
+            source_bytes = self.get_data(path)
+        except IOError:
+            raise ImportError("source not available through get_data()")
+        encoding = tokenize.detect_encoding(_io.BytesIO(source_bytes).readline)
+        newline_decoder = _io.IncrementalNewlineDecoder(None, True)
+        return newline_decoder.decode(source_bytes.decode(encoding[0]))
 
     def get_code(self, fullname):
-        """Get a code object from source or bytecode."""
-        # XXX Care enough to make sure this call does not happen if the magic
-        #     number is bad?
-        source_timestamp = self.source_mtime(fullname)
-        # Try to use bytecode if it is available.
-        bytecode_path = self.bytecode_path(fullname)
-        if bytecode_path:
-            data = self.get_data(bytecode_path)
+        """Concrete implementation of InspectLoader.get_code.
+
+        Reading of bytecode requires path_mtime to be implemented. To write
+        bytecode, set_data must also be implemented.
+
+        """
+        source_path = self.get_filename(fullname)
+        bytecode_path = imp.cache_from_source(source_path)
+        source_mtime = None
+        if bytecode_path is not None:
             try:
-                magic = data[:4]
-                if len(magic) < 4:
-                    raise ImportError("bad magic number in {}".format(fullname))
-                raw_timestamp = data[4:8]
-                if len(raw_timestamp) < 4:
-                    raise EOFError("bad timestamp in {}".format(fullname))
-                pyc_timestamp = marshal._r_long(raw_timestamp)
-                bytecode = data[8:]
-                # Verify that the magic number is valid.
-                if imp.get_magic() != magic:
-                    raise ImportError("bad magic number in {}".format(fullname))
-                # Verify that the bytecode is not stale (only matters when
-                # there is source to fall back on.
-                if source_timestamp:
-                    if pyc_timestamp < source_timestamp:
-                        raise ImportError("bytecode is stale")
-            except (ImportError, EOFError):
-                # If source is available give it a shot.
-                if source_timestamp is not None:
+                source_mtime = self.path_mtime(source_path)
+            except NotImplementedError:
+                pass
+            else:
+                try:
+                    data = self.get_data(bytecode_path)
+                except IOError:
                     pass
                 else:
-                    raise
-            else:
-                # Bytecode seems fine, so try to use it.
-                # XXX If the bytecode is ill-formed, would it be beneficial to
-                #     try for using source if available and issue a warning?
-                return marshal.loads(bytecode)
-        elif source_timestamp is None:
-            raise ImportError("no source or bytecode available to create code "
-                                "object for {0!r}".format(fullname))
-        # Use the source.
-        code_object = super().get_code(fullname)
-        # Generate bytecode and write it out.
-        if not sys.dont_write_bytecode:
+                    try:
+                        bytes_data = self._bytes_from_bytecode(fullname, data,
+                                                               source_mtime)
+                    except (ImportError, EOFError):
+                        pass
+                    else:
+                        found = marshal.loads(bytes_data)
+                        if isinstance(found, code_type):
+                            return found
+                        else:
+                            msg = "Non-code object in {}"
+                            raise ImportError(msg.format(bytecode_path))
+        source_bytes = self.get_data(source_path)
+        code_object = compile(source_bytes, source_path, 'exec',
+                                dont_inherit=True)
+        if (not sys.dont_write_bytecode and bytecode_path is not None and
+                source_mtime is not None):
+            # If e.g. Jython ever implements imp.cache_from_source to have
+            # their own cached file format, this block of code will most likely
+            # throw an exception.
             data = bytearray(imp.get_magic())
-            data.extend(marshal._w_long(source_timestamp))
+            data.extend(marshal._w_long(source_mtime))
             data.extend(marshal.dumps(code_object))
-            self.write_bytecode(fullname, data)
+            try:
+                self.set_data(bytecode_path, data)
+            except NotImplementedError:
+                pass
         return code_object
 
+    def load_module(self, fullname):
+        """Concrete implementation of Loader.load_module.
 
-class _PyFileLoader(PyLoader):
+        Requires ExecutionLoader.get_filename and ResourceLoader.get_data to be
+        implemented to load source code. Use of bytecode is dictated by whether
+        get_code uses/writes bytecode.
 
-    """Load a Python source file."""
+        """
+        return self._load_module(fullname)
 
-    def __init__(self, name, path, is_pkg):
-        self._name = name
-        self._is_pkg = is_pkg
-        # Figure out the base path based on whether it was source or bytecode
-        # that was found.
-        try:
-            self._base_path = _path_without_ext(path, imp.PY_SOURCE)
-        except ValueError:
-            self._base_path = _path_without_ext(path, imp.PY_COMPILED)
-
-    def _find_path(self, ext_type):
-        """Find a path from the base path and the specified extension type that
-        exists, returning None if one is not found."""
-        for suffix in _suffix_list(ext_type):
-            path = self._base_path + suffix
-            if _path_exists(path):
-                return path
-        else:
-            return None
+
+class _FileLoader:
+
+    """Base file loader class which implements the loader protocol methods that
+    require file system usage."""
+
+    def __init__(self, fullname, path):
+        """Cache the module name and the path to the file found by the
+        finder."""
+        self._name = fullname
+        self._path = path
 
     @_check_name
-    def source_path(self, fullname):
-        """Return the path to an existing source file for the module, or None
-        if one cannot be found."""
-        # Not a property so that it is easy to override.
-        return self._find_path(imp.PY_SOURCE)
+    def get_filename(self, fullname):
+        """Return the path to the source file as found by the finder."""
+        return self._path
 
     def get_data(self, path):
         """Return the data from path as raw bytes."""
-        return _io.FileIO(path, 'r').read()  # Assuming bytes.
+        with _closing(_io.FileIO(path, 'r')) as file:
+            return file.read()
 
-    @_check_name
-    def is_package(self, fullname):
-        """Return a boolean based on whether the module is a package.
 
-        Raises ImportError (like get_source) if the loader cannot handle the
-        package.
+class _SourceFileLoader(_FileLoader, SourceLoader):
 
-        """
-        return self._is_pkg
+    """Concrete implementation of SourceLoader using the file system."""
 
+    def path_mtime(self, path):
+        """Return the modification time for the path."""
+        return int(_os.stat(path).st_mtime)
 
-class _PyPycFileLoader(PyPycLoader, _PyFileLoader):
+    def set_data(self, path, data):
+        """Write bytes data to a file."""
+        try:
+            with _closing(_io.FileIO(path, 'wb')) as file:
+                file.write(data)
+        except IOError as exc:
+            if exc.errno == errno.ENOENT:
+                directory, _, filename = path.rpartition(path_sep)
+                sub_directories = []
+                while not _path_isdir(directory):
+                    directory, _, sub_dir = directory.rpartition(path_sep)
+                    sub_directories.append(sub_dir)
+                    for part in reversed(sub_directories):
+                        directory = _path_join(directory, part)
+                        try:
+                            _os.mkdir(directory)
+                        except IOError as exc:
+                            if exc.errno != errno.EACCES:
+                                raise
+                            else:
+                                return
+                return self.set_data(path, data)
+            elif exc.errno != errno.EACCES:
+                raise
 
-    """Load a module from a source or bytecode file."""
 
-    def _find_path(self, ext_type):
-        """Return PEP 3147 path if ext_type is PY_COMPILED, otherwise
-        super()._find_path() is called."""
-        if ext_type == imp.PY_COMPILED:
-            # We don't really care what the extension on self._base_path is,
-            # as long as it has exactly one dot.
-            source_path = self._base_path + '.py'
-            pycache_path = imp.cache_from_source(source_path)
-            legacy_path = self._base_path + '.pyc'
-            # The rule is: if the source file exists, then Python always uses
-            # the __pycache__/foo.<tag>.pyc file.  If the source file does not
-            # exist, then Python uses the legacy path.
-            pyc_path = (pycache_path
-                        if _path_exists(source_path)
-                        else legacy_path)
-            return (pyc_path if _path_exists(pyc_path) else None)
-        return super()._find_path(ext_type)
+class _SourcelessFileLoader(_FileLoader, _LoaderBasics):
 
-    @_check_name
-    def source_mtime(self, name):
-        """Return the modification time of the source for the specified
-        module."""
-        source_path = self.source_path(name)
-        if not source_path:
-            return None
-        return int(_os.stat(source_path).st_mtime)
+    """Loader which handles sourceless file imports."""
 
-    @_check_name
-    def bytecode_path(self, fullname):
-        """Return the path to a bytecode file, or None if one does not
-        exist."""
-        # Not a property for easy overriding.
-        return self._find_path(imp.PY_COMPILED)
-
-    @_check_name
-    def write_bytecode(self, name, data):
-        """Write out 'data' for the specified module, returning a boolean
-        signifying if the write-out actually occurred.
+    def load_module(self, fullname):
+        return self._load_module(fullname, sourceless=True)
 
-        Raises ImportError (just like get_source) if the specified module
-        cannot be handled by the loader.
+    def get_code(self, fullname):
+        path = self.get_filename(fullname)
+        data = self.get_data(path)
+        bytes_data = self._bytes_from_bytecode(fullname, data, None)
+        found = marshal.loads(bytes_data)
+        if isinstance(found, code_type):
+            return found
+        else:
+            raise ImportError("Non-code object in {}".format(path))
 
-        """
-        bytecode_path = self.bytecode_path(name)
-        if not bytecode_path:
-            source_path = self.source_path(name)
-            bytecode_path = imp.cache_from_source(source_path)
-            # Ensure that the __pycache__ directory exists.  We can't use
-            # os.path.dirname() here.
-            dirname, sep, basename = bytecode_path.rpartition(path_sep)
-            try:
-                _os.mkdir(dirname)
-            except OSError as error:
-                if error.errno != errno.EEXIST:
-                    raise
-        try:
-            # Assuming bytes.
-            with _closing(_io.FileIO(bytecode_path, 'w')) as bytecode_file:
-                bytecode_file.write(data)
-                return True
-        except IOError as exc:
-            if exc.errno == errno.EACCES:
-                return False
-            else:
-                raise
+    def get_source(self, fullname):
+        """Return None as there is no source code."""
+        return None
 
 
 class _ExtensionFileLoader:
@@ -563,7 +536,7 @@
 
     """
 
-    def __init__(self, name, path, is_pkg):
+    def __init__(self, name, path):
         """Initialize the loader.
 
         If is_pkg is True then an exception is raised as extension modules
@@ -572,8 +545,6 @@
         """
         self._name = name
         self._path = path
-        if is_pkg:
-            raise ValueError("extension modules cannot be packages")
 
     @_check_name
     @set_package
@@ -671,147 +642,88 @@
             return None
 
 
-class _ChainedFinder:
-
-    """Finder that sequentially calls other finders."""
-
-    def __init__(self, *finders):
-        self._finders = finders
-
-    def find_module(self, fullname, path=None):
-        for finder in self._finders:
-            result = finder.find_module(fullname, path)
-            if result:
-                return result
-        else:
-            return None
-
-
 class _FileFinder:
 
-    """Base class for file finders.
-
-    Subclasses are expected to define the following attributes:
-
-        * _suffixes
-            Sequence of file suffixes whose order will be followed.
+    """File-based finder.
 
-        * _possible_package
-            True if importer should check for packages.
-
-        * _loader
-            A callable that takes the module name, a file path, and whether
-            the path points to a package and returns a loader for the module
-            found at that path.
+    Constructor takes a list of objects detailing what file extensions their
+    loader supports along with whether it can be used for a package.
 
     """
 
-    def __init__(self, path_entry):
-        """Initialize an importer for the passed-in sys.path entry (which is
-        assumed to have already been verified as an existing directory).
-
-        Can be used as an entry on sys.path_hook.
+    def __init__(self, path, *details):
+        """Initialize with finder details."""
+        packages = []
+        modules = []
+        for detail in details:
+            modules.extend((suffix, detail.loader) for suffix in detail.suffixes)
+            if detail.supports_packages:
+                packages.extend((suffix, detail.loader)
+                                for suffix in detail.suffixes)
+        self.packages = packages
+        self.modules = modules
+        self.path = path
 
-        """
-        absolute_path = _path_absolute(path_entry)
-        if not _path_isdir(absolute_path):
-            raise ImportError("only directories are supported")
-        self._path_entry = absolute_path
-
-    def find_module(self, fullname, path=None):
+    def find_module(self, fullname):
+        """Try to find a loader for the specified module."""
         tail_module = fullname.rpartition('.')[2]
-        package_directory = None
-        if self._possible_package:
-            for ext in self._suffixes:
-                package_directory = _path_join(self._path_entry, tail_module)
-                init_filename = '__init__' + ext
-                package_init = _path_join(package_directory, init_filename)
-                if (_path_isfile(package_init) and
-                        _case_ok(self._path_entry, tail_module) and
-                        _case_ok(package_directory, init_filename)):
-                    return self._loader(fullname, package_init, True)
-        for ext in self._suffixes:
-            file_name = tail_module + ext
-            file_path = _path_join(self._path_entry, file_name)
-            if (_path_isfile(file_path) and
-                    _case_ok(self._path_entry, file_name)):
-                return self._loader(fullname, file_path, False)
-        else:
-            # Raise a warning if it matches a directory w/o an __init__ file.
-            if (package_directory is not None and
-                    _path_isdir(package_directory) and
-                    _case_ok(self._path_entry, tail_module)):
-                _warnings.warn("Not importing directory %s: missing __init__"
-                                % package_directory, ImportWarning)
-            return None
-
-
-class _PyFileFinder(_FileFinder):
-
-    """Importer for source/bytecode files."""
-
-    _possible_package = True
-    _loader = _PyFileLoader
-
-    def __init__(self, path_entry):
-        # Lack of imp during class creation means _suffixes is set here.
-        # Make sure that Python source files are listed first!  Needed for an
-        # optimization by the loader.
-        self._suffixes = _suffix_list(imp.PY_SOURCE)
-        super().__init__(path_entry)
-
-
-class _PyPycFileFinder(_PyFileFinder):
+        base_path = _path_join(self.path, tail_module)
+        if _path_isdir(base_path) and _case_ok(self.path, tail_module):
+            for suffix, loader in self.packages:
+                init_filename = '__init__' + suffix
+                full_path = _path_join(base_path, init_filename)
+                if (_path_isfile(full_path) and
+                        _case_ok(base_path, init_filename)):
+                    return loader(fullname, full_path)
+            else:
+                msg = "Not importing directory {}: missing __init__"
+                _warnings.warn(msg.format(base_path), ImportWarning)
+        for suffix, loader in self.modules:
+            mod_filename = tail_module + suffix
+            full_path = _path_join(self.path, mod_filename)
+            if _path_isfile(full_path) and _case_ok(self.path, mod_filename):
+                return loader(fullname, full_path)
+        return None
 
-    """Finder for source and bytecode files."""
+class _SourceFinderDetails:
 
-    _loader = _PyPycFileLoader
+    loader = _SourceFileLoader
+    supports_packages = True
 
-    def __init__(self, path_entry):
-        super().__init__(path_entry)
-        self._suffixes += _suffix_list(imp.PY_COMPILED)
+    def __init__(self):
+        self.suffixes = _suffix_list(imp.PY_SOURCE)
 
+class _SourcelessFinderDetails:
 
+    loader = _SourcelessFileLoader
+    supports_packages = True
 
+    def __init__(self):
+        self.suffixes = _suffix_list(imp.PY_COMPILED)
 
-class _ExtensionFileFinder(_FileFinder):
 
-    """Importer for extension files."""
+class _ExtensionFinderDetails:
 
-    _possible_package = False
-    _loader = _ExtensionFileLoader
+    loader = _ExtensionFileLoader
+    supports_packages = False
 
-    def __init__(self, path_entry):
-        # Assigning to _suffixes here instead of at the class level because
-        # imp is not imported at the time of class creation.
-        self._suffixes = _suffix_list(imp.C_EXTENSION)
-        super().__init__(path_entry)
+    def __init__(self):
+        self.suffixes = _suffix_list(imp.C_EXTENSION)
 
 
 # Import itself ###############################################################
 
-def _chained_path_hook(*path_hooks):
-    """Create a closure which sequentially checks path hooks to see which ones
-    (if any) can work with a path."""
-    def path_hook(entry):
-        """Check to see if 'entry' matches any of the enclosed path hooks."""
-        finders = []
-        for hook in path_hooks:
-            try:
-                finder = hook(entry)
-            except ImportError:
-                continue
-            else:
-                finders.append(finder)
-        if not finders:
-            raise ImportError("no finder found")
-        else:
-            return _ChainedFinder(*finders)
-
-    return path_hook
+def _file_path_hook(path):
+    """If the path is a directory, return a file-based finder."""
+    if _path_isdir(path):
+        return _FileFinder(path, _ExtensionFinderDetails(),
+                           _SourceFinderDetails(),
+                           _SourcelessFinderDetails())
+    else:
+        raise ImportError("only directories are supported")
 
 
-_DEFAULT_PATH_HOOK = _chained_path_hook(_ExtensionFileFinder, _PyPycFileFinder)
+_DEFAULT_PATH_HOOK = _file_path_hook
 
 class _DefaultPathFinder(PathFinder):
 

Modified: python/branches/py3k-dtoa/Lib/importlib/abc.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/importlib/abc.py	(original)
+++ python/branches/py3k-dtoa/Lib/importlib/abc.py	Tue Jul  6 17:55:27 2010
@@ -1,8 +1,16 @@
 """Abstract base classes related to import."""
 from . import _bootstrap
 from . import machinery
+from . import util
 import abc
+import imp
+import io
+import marshal
+import os.path
+import sys
+import tokenize
 import types
+import warnings
 
 
 class Loader(metaclass=abc.ABCMeta):
@@ -58,19 +66,19 @@
     def is_package(self, fullname:str) -> bool:
         """Abstract method which when implemented should return whether the
         module is a package."""
-        return NotImplementedError
+        raise NotImplementedError
 
     @abc.abstractmethod
     def get_code(self, fullname:str) -> types.CodeType:
         """Abstract method which when implemented should return the code object
         for the module"""
-        return NotImplementedError
+        raise NotImplementedError
 
     @abc.abstractmethod
     def get_source(self, fullname:str) -> str:
         """Abstract method which should return the source code for the
         module."""
-        return NotImplementedError
+        raise NotImplementedError
 
 InspectLoader.register(machinery.BuiltinImporter)
 InspectLoader.register(machinery.FrozenImporter)
@@ -92,33 +100,170 @@
         raise NotImplementedError
 
 
-class PyLoader(_bootstrap.PyLoader, ResourceLoader, ExecutionLoader):
+class SourceLoader(_bootstrap.SourceLoader, ResourceLoader, ExecutionLoader):
 
-    """Abstract base class to assist in loading source code by requiring only
-    back-end storage methods to be implemented.
+    """Abstract base class for loading source code (and optionally any
+    corresponding bytecode).
+
+    To support loading from source code, the abstractmethods inherited from
+    ResourceLoader and ExecutionLoader need to be implemented. To also support
+    loading from bytecode, the optional methods specified directly by this ABC
+    is required.
+
+    Inherited abstractmethods not implemented in this ABC:
+
+        * ResourceLoader.get_data
+        * ExecutionLoader.get_filename
+
+    """
 
-    The methods get_code, get_source, and load_module are implemented for the
-    user.
+
+class PyLoader(SourceLoader):
+
+    """Implement the deprecated PyLoader ABC in terms of SourceLoader.
+
+    This class has been deprecated! It is slated for removal in Python 3.4.
+    If compatibility with Python 3.1 is not needed then implement the
+    SourceLoader ABC instead of this class. If Python 3.1 compatibility is
+    needed, then use the following idiom to have a single class that is
+    compatible with Python 3.1 onwards::
+
+        try:
+            from importlib.abc import SourceLoader
+        except ImportError:
+            from importlib.abc import PyLoader as SourceLoader
+
+
+        class CustomLoader(SourceLoader):
+            def get_filename(self, fullname):
+                # Implement ...
+
+            def source_path(self, fullname):
+                '''Implement source_path in terms of get_filename.'''
+                try:
+                    return self.get_filename(fullname)
+                except ImportError:
+                    return None
+
+            def is_package(self, fullname):
+                filename = os.path.basename(self.get_filename(fullname))
+                return os.path.splitext(filename)[0] == '__init__'
 
     """
 
     @abc.abstractmethod
+    def is_package(self, fullname):
+        raise NotImplementedError
+
+    @abc.abstractmethod
     def source_path(self, fullname:str) -> object:
         """Abstract method which when implemented should return the path to the
-        sourced code for the module."""
+        source code for the module."""
         raise NotImplementedError
 
+    def get_filename(self, fullname):
+        """Implement get_filename in terms of source_path.
+
+        As get_filename should only return a source file path there is no
+        chance of the path not existing but loading still being possible, so
+        ImportError should propagate instead of being turned into returning
+        None.
+
+        """
+        warnings.warn("importlib.abc.PyLoader is deprecated and is "
+                            "slated for removal in Python 3.4; "
+                            "use SourceLoader instead. "
+                            "See the importlib documentation on how to be "
+                            "compatible with Python 3.1 onwards.",
+                        PendingDeprecationWarning)
+        path = self.source_path(fullname)
+        if path is None:
+            raise ImportError
+        else:
+            return path
+
 
-class PyPycLoader(_bootstrap.PyPycLoader, PyLoader):
+class PyPycLoader(PyLoader):
 
     """Abstract base class to assist in loading source and bytecode by
     requiring only back-end storage methods to be implemented.
 
+    This class has been deprecated! Removal is slated for Python 3.4. Implement
+    the SourceLoader ABC instead. If Python 3.1 compatibility is needed, see
+    PyLoader.
+
     The methods get_code, get_source, and load_module are implemented for the
     user.
 
     """
 
+    def get_filename(self, fullname):
+        """Return the source or bytecode file path."""
+        path = self.source_path(fullname)
+        if path is not None:
+            return path
+        path = self.bytecode_path(fullname)
+        if path is not None:
+            return path
+        raise ImportError("no source or bytecode path available for "
+                            "{0!r}".format(fullname))
+
+    def get_code(self, fullname):
+        """Get a code object from source or bytecode."""
+        warnings.warn("importlib.abc.PyPycLoader is deprecated and slated for "
+                            "removal in Python 3.4; use SourceLoader instead. "
+                            "If Python 3.1 compatibility is required, see the "
+                            "latest documentation for PyLoader.",
+                        PendingDeprecationWarning)
+        source_timestamp = self.source_mtime(fullname)
+        # Try to use bytecode if it is available.
+        bytecode_path = self.bytecode_path(fullname)
+        if bytecode_path:
+            data = self.get_data(bytecode_path)
+            try:
+                magic = data[:4]
+                if len(magic) < 4:
+                    raise ImportError("bad magic number in {}".format(fullname))
+                raw_timestamp = data[4:8]
+                if len(raw_timestamp) < 4:
+                    raise EOFError("bad timestamp in {}".format(fullname))
+                pyc_timestamp = marshal._r_long(raw_timestamp)
+                bytecode = data[8:]
+                # Verify that the magic number is valid.
+                if imp.get_magic() != magic:
+                    raise ImportError("bad magic number in {}".format(fullname))
+                # Verify that the bytecode is not stale (only matters when
+                # there is source to fall back on.
+                if source_timestamp:
+                    if pyc_timestamp < source_timestamp:
+                        raise ImportError("bytecode is stale")
+            except (ImportError, EOFError):
+                # If source is available give it a shot.
+                if source_timestamp is not None:
+                    pass
+                else:
+                    raise
+            else:
+                # Bytecode seems fine, so try to use it.
+                return marshal.loads(bytecode)
+        elif source_timestamp is None:
+            raise ImportError("no source or bytecode available to create code "
+                                "object for {0!r}".format(fullname))
+        # Use the source.
+        source_path = self.source_path(fullname)
+        if source_path is None:
+            message = "a source path must exist to load {0}".format(fullname)
+            raise ImportError(message)
+        source = self.get_data(source_path)
+        code_object = compile(source, source_path, 'exec', dont_inherit=True)
+        # Generate bytecode and write it out.
+        if not sys.dont_write_bytecode:
+            data = bytearray(imp.get_magic())
+            data.extend(marshal._w_long(source_timestamp))
+            data.extend(marshal.dumps(code_object))
+            self.write_bytecode(fullname, data)
+        return code_object
+
     @abc.abstractmethod
     def source_mtime(self, fullname:str) -> int:
         """Abstract method which when implemented should return the

Modified: python/branches/py3k-dtoa/Lib/importlib/test/extension/test_case_sensitivity.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/importlib/test/extension/test_case_sensitivity.py	(original)
+++ python/branches/py3k-dtoa/Lib/importlib/test/extension/test_case_sensitivity.py	Tue Jul  6 17:55:27 2010
@@ -13,7 +13,8 @@
         good_name = ext_util.NAME
         bad_name = good_name.upper()
         assert good_name != bad_name
-        finder = _bootstrap._ExtensionFileFinder(ext_util.PATH)
+        finder = _bootstrap._FileFinder(ext_util.PATH,
+                                        _bootstrap._ExtensionFinderDetails())
         return finder.find_module(bad_name)
 
     def test_case_sensitive(self):

Modified: python/branches/py3k-dtoa/Lib/importlib/test/extension/test_finder.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/importlib/test/extension/test_finder.py	(original)
+++ python/branches/py3k-dtoa/Lib/importlib/test/extension/test_finder.py	Tue Jul  6 17:55:27 2010
@@ -9,7 +9,8 @@
     """Test the finder for extension modules."""
 
     def find_module(self, fullname):
-        importer = _bootstrap._ExtensionFileFinder(util.PATH)
+        importer = _bootstrap._FileFinder(util.PATH,
+                                          _bootstrap._ExtensionFinderDetails())
         return importer.find_module(fullname)
 
     def test_module(self):

Modified: python/branches/py3k-dtoa/Lib/importlib/test/extension/test_loader.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/importlib/test/extension/test_loader.py	(original)
+++ python/branches/py3k-dtoa/Lib/importlib/test/extension/test_loader.py	Tue Jul  6 17:55:27 2010
@@ -13,7 +13,7 @@
 
     def load_module(self, fullname):
         loader = _bootstrap._ExtensionFileLoader(ext_util.NAME,
-                                                ext_util.FILEPATH, False)
+                                                ext_util.FILEPATH)
         return loader.load_module(fullname)
 
     def test_module(self):

Modified: python/branches/py3k-dtoa/Lib/importlib/test/extension/test_path_hook.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/importlib/test/extension/test_path_hook.py	(original)
+++ python/branches/py3k-dtoa/Lib/importlib/test/extension/test_path_hook.py	Tue Jul  6 17:55:27 2010
@@ -14,7 +14,7 @@
     # XXX Should it only work for directories containing an extension module?
 
     def hook(self, entry):
-        return _bootstrap._ExtensionFileFinder(entry)
+        return _bootstrap._file_path_hook(entry)
 
     def test_success(self):
         # Path hook should handle a directory where a known extension module

Modified: python/branches/py3k-dtoa/Lib/importlib/test/import_/test_path.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/importlib/test/import_/test_path.py	(original)
+++ python/branches/py3k-dtoa/Lib/importlib/test/import_/test_path.py	Tue Jul  6 17:55:27 2010
@@ -5,6 +5,7 @@
 import imp
 import os
 import sys
+import tempfile
 from test import support
 from types import MethodType
 import unittest
@@ -80,23 +81,28 @@
 
     def test_implicit_hooks(self):
         # Test that the implicit path hooks are used.
-        existing_path = os.path.dirname(support.TESTFN)
         bad_path = '<path>'
         module = '<module>'
         assert not os.path.exists(bad_path)
-        with util.import_state():
-            nothing = _bootstrap._DefaultPathFinder.find_module(module,
-                                                           path=[existing_path])
-            self.assertTrue(nothing is None)
-            self.assertTrue(existing_path in sys.path_importer_cache)
-            self.assertTrue(not isinstance(sys.path_importer_cache[existing_path],
-                                        imp.NullImporter))
-            nothing = _bootstrap._DefaultPathFinder.find_module(module,
-                                                                path=[bad_path])
-            self.assertTrue(nothing is None)
-            self.assertTrue(bad_path in sys.path_importer_cache)
-            self.assertTrue(isinstance(sys.path_importer_cache[bad_path],
-                                    imp.NullImporter))
+        existing_path = tempfile.mkdtemp()
+        try:
+            with util.import_state():
+                nothing = _bootstrap._DefaultPathFinder.find_module(module,
+                                                        path=[existing_path])
+                self.assertTrue(nothing is None)
+                self.assertTrue(existing_path in sys.path_importer_cache)
+                result = isinstance(sys.path_importer_cache[existing_path],
+                                    imp.NullImporter)
+                self.assertFalse(result)
+                nothing = _bootstrap._DefaultPathFinder.find_module(module,
+                                                            path=[bad_path])
+                self.assertTrue(nothing is None)
+                self.assertTrue(bad_path in sys.path_importer_cache)
+                self.assertTrue(isinstance(sys.path_importer_cache[bad_path],
+                                           imp.NullImporter))
+        finally:
+            os.rmdir(existing_path)
+
 
     def test_path_importer_cache_has_None(self):
         # Test that the default hook is used when sys.path_importer_cache

Modified: python/branches/py3k-dtoa/Lib/importlib/test/regrtest.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/importlib/test/regrtest.py	(original)
+++ python/branches/py3k-dtoa/Lib/importlib/test/regrtest.py	Tue Jul  6 17:55:27 2010
@@ -6,9 +6,11 @@
 this script.
 
 XXX FAILING
-    test_import
-        execution bit
+  * test_import
+    - test_incorrect_code_name
         file name differing between __file__ and co_filename (r68360 on trunk)
+    - test_import_by_filename
+        exception for trying to import by file name does not match
 
 """
 import importlib

Modified: python/branches/py3k-dtoa/Lib/importlib/test/source/test_abc_loader.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/importlib/test/source/test_abc_loader.py	(original)
+++ python/branches/py3k-dtoa/Lib/importlib/test/source/test_abc_loader.py	Tue Jul  6 17:55:27 2010
@@ -1,14 +1,68 @@
 import importlib
 from importlib import abc
+
 from .. import abc as testing_abc
 from .. import util
 from . import util as source_util
+
 import imp
+import inspect
+import io
 import marshal
 import os
 import sys
 import types
 import unittest
+import warnings
+
+
+class SourceOnlyLoaderMock(abc.SourceLoader):
+
+    # Globals that should be defined for all modules.
+    source = (b"_ = '::'.join([__name__, __file__, __cached__, __package__, "
+              b"repr(__loader__)])")
+
+    def __init__(self, path):
+        self.path = path
+
+    def get_data(self, path):
+        assert self.path == path
+        return self.source
+
+    def get_filename(self, fullname):
+        return self.path
+
+
+class SourceLoaderMock(SourceOnlyLoaderMock):
+
+    source_mtime = 1
+
+    def __init__(self, path, magic=imp.get_magic()):
+        super().__init__(path)
+        self.bytecode_path = imp.cache_from_source(self.path)
+        data = bytearray(magic)
+        data.extend(marshal._w_long(self.source_mtime))
+        code_object = compile(self.source, self.path, 'exec',
+                                dont_inherit=True)
+        data.extend(marshal.dumps(code_object))
+        self.bytecode = bytes(data)
+        self.written = {}
+
+    def get_data(self, path):
+        if path == self.path:
+            return super().get_data(path)
+        elif path == self.bytecode_path:
+            return self.bytecode
+        else:
+            raise IOError
+
+    def path_mtime(self, path):
+        assert path == self.path
+        return self.source_mtime
+
+    def set_data(self, path, data):
+        self.written[path] = bytes(data)
+        return path == self.bytecode_path
 
 
 class PyLoaderMock(abc.PyLoader):
@@ -33,17 +87,42 @@
         return self.source
 
     def is_package(self, name):
+        filename = os.path.basename(self.get_filename(name))
+        return os.path.splitext(filename)[0] == '__init__'
+
+    def source_path(self, name):
         try:
-            return '__init__' in self.module_paths[name]
+            return self.module_paths[name]
         except KeyError:
             raise ImportError
 
-    def source_path(self, name):
+    def get_filename(self, name):
+        """Silence deprecation warning."""
+        with warnings.catch_warnings(record=True) as w:
+            warnings.simplefilter("always")
+            path = super().get_filename(name)
+            assert len(w) == 1
+            assert issubclass(w[0].category, PendingDeprecationWarning)
+            return path
+
+
+class PyLoaderCompatMock(PyLoaderMock):
+
+    """Mock that matches what is suggested to have a loader that is compatible
+    from Python 3.1 onwards."""
+
+    def get_filename(self, fullname):
         try:
-            return self.module_paths[name]
+            return self.module_paths[fullname]
         except KeyError:
             raise ImportError
 
+    def source_path(self, fullname):
+        try:
+            return self.get_filename(fullname)
+        except ImportError:
+            return None
+
 
 class PyPycLoaderMock(abc.PyPycLoader, PyLoaderMock):
 
@@ -114,6 +193,13 @@
         except TypeError:
             return '__init__' in self.bytecode_to_path[name]
 
+    def get_code(self, name):
+        with warnings.catch_warnings(record=True) as w:
+            warnings.simplefilter("always")
+            code_object = super().get_code(name)
+            assert len(w) == 1
+            assert issubclass(w[0].category, PendingDeprecationWarning)
+            return code_object
 
 class PyLoaderTests(testing_abc.LoaderTests):
 
@@ -200,6 +286,14 @@
         return mock
 
 
+class PyLoaderCompatTests(PyLoaderTests):
+
+    """Test that the suggested code to make a loader that is compatible from
+    Python 3.1 forward works."""
+
+    mocker = PyLoaderCompatMock
+
+
 class PyLoaderInterfaceTests(unittest.TestCase):
 
     """Tests for importlib.abc.PyLoader to make sure that when source_path()
@@ -234,30 +328,6 @@
             mock.get_filename(name)
 
 
-class PyLoaderGetSourceTests(unittest.TestCase):
-
-    """Tests for importlib.abc.PyLoader.get_source()."""
-
-    def test_default_encoding(self):
-        # Should have no problems with UTF-8 text.
-        name = 'mod'
-        mock = PyLoaderMock({name: os.path.join('path', 'to', 'mod')})
-        source = 'x = "??"'
-        mock.source = source.encode('utf-8')
-        returned_source = mock.get_source(name)
-        self.assertEqual(returned_source, source)
-
-    def test_decoded_source(self):
-        # Decoding should work.
-        name = 'mod'
-        mock = PyLoaderMock({name: os.path.join('path', 'to', 'mod')})
-        source = "# coding: Latin-1\nx='??'"
-        assert source.encode('latin-1') != source.encode('utf-8')
-        mock.source = source.encode('latin-1')
-        returned_source = mock.get_source(name)
-        self.assertEqual(returned_source, source)
-
-
 class PyPycLoaderTests(PyLoaderTests):
 
     """Tests for importlib.abc.PyPycLoader."""
@@ -413,7 +483,7 @@
     def test_bad_bytecode(self):
         # Malformed code object bytecode should lead to a ValueError.
         name = 'mod'
-        bc = {name: {'path': os.path.join('path', 'to', 'mod'), 'bc': b'XXX'}}
+        bc = {name: {'path': os.path.join('path', 'to', 'mod'), 'bc': b'1234'}}
         mock = PyPycLoaderMock({name: None}, bc)
         with util.uncache(name), self.assertRaises(ValueError):
             mock.load_module(name)
@@ -465,12 +535,341 @@
             mock.load_module(name)
 
 
+class SourceLoaderTestHarness(unittest.TestCase):
+
+    def setUp(self, *, is_package=True, **kwargs):
+        self.package = 'pkg'
+        if is_package:
+            self.path = os.path.join(self.package, '__init__.py')
+            self.name = self.package
+        else:
+            module_name = 'mod'
+            self.path = os.path.join(self.package, '.'.join(['mod', 'py']))
+            self.name = '.'.join([self.package, module_name])
+        self.cached = imp.cache_from_source(self.path)
+        self.loader = self.loader_mock(self.path, **kwargs)
+
+    def verify_module(self, module):
+        self.assertEqual(module.__name__, self.name)
+        self.assertEqual(module.__file__, self.path)
+        self.assertEqual(module.__cached__, self.cached)
+        self.assertEqual(module.__package__, self.package)
+        self.assertEqual(module.__loader__, self.loader)
+        values = module._.split('::')
+        self.assertEqual(values[0], self.name)
+        self.assertEqual(values[1], self.path)
+        self.assertEqual(values[2], self.cached)
+        self.assertEqual(values[3], self.package)
+        self.assertEqual(values[4], repr(self.loader))
+
+    def verify_code(self, code_object):
+        module = imp.new_module(self.name)
+        module.__file__ = self.path
+        module.__cached__ = self.cached
+        module.__package__ = self.package
+        module.__loader__ = self.loader
+        module.__path__ = []
+        exec(code_object, module.__dict__)
+        self.verify_module(module)
+
+
+class SourceOnlyLoaderTests(SourceLoaderTestHarness):
+
+    """Test importlib.abc.SourceLoader for source-only loading.
+
+    Reload testing is subsumed by the tests for
+    importlib.util.module_for_loader.
+
+    """
+
+    loader_mock = SourceOnlyLoaderMock
+
+    def test_get_source(self):
+        # Verify the source code is returned as a string.
+        # If an IOError is raised by get_data then raise ImportError.
+        expected_source = self.loader.source.decode('utf-8')
+        self.assertEqual(self.loader.get_source(self.name), expected_source)
+        def raise_IOError(path):
+            raise IOError
+        self.loader.get_data = raise_IOError
+        with self.assertRaises(ImportError):
+            self.loader.get_source(self.name)
+
+    def test_is_package(self):
+        # Properly detect when loading a package.
+        self.setUp(is_package=True)
+        self.assertTrue(self.loader.is_package(self.name))
+        self.setUp(is_package=False)
+        self.assertFalse(self.loader.is_package(self.name))
+
+    def test_get_code(self):
+        # Verify the code object is created.
+        code_object = self.loader.get_code(self.name)
+        self.verify_code(code_object)
+
+    def test_load_module(self):
+        # Loading a module should set __name__, __loader__, __package__,
+        # __path__ (for packages), __file__, and __cached__.
+        # The module should also be put into sys.modules.
+        with util.uncache(self.name):
+            module = self.loader.load_module(self.name)
+            self.verify_module(module)
+            self.assertEqual(module.__path__, [os.path.dirname(self.path)])
+            self.assertTrue(self.name in sys.modules)
+
+    def test_package_settings(self):
+        # __package__ needs to be set, while __path__ is set on if the module
+        # is a package.
+        # Testing the values for a package are covered by test_load_module.
+        self.setUp(is_package=False)
+        with util.uncache(self.name):
+            module = self.loader.load_module(self.name)
+            self.verify_module(module)
+            self.assertTrue(not hasattr(module, '__path__'))
+
+    def test_get_source_encoding(self):
+        # Source is considered encoded in UTF-8 by default unless otherwise
+        # specified by an encoding line.
+        source = "_ = '??'"
+        self.loader.source = source.encode('utf-8')
+        returned_source = self.loader.get_source(self.name)
+        self.assertEqual(returned_source, source)
+        source = "# coding: latin-1\n_ = ??"
+        self.loader.source = source.encode('latin-1')
+        returned_source = self.loader.get_source(self.name)
+        self.assertEqual(returned_source, source)
+
+
+ at unittest.skipIf(sys.dont_write_bytecode, "sys.dont_write_bytecode is true")
+class SourceLoaderBytecodeTests(SourceLoaderTestHarness):
+
+    """Test importlib.abc.SourceLoader's use of bytecode.
+
+    Source-only testing handled by SourceOnlyLoaderTests.
+
+    """
+
+    loader_mock = SourceLoaderMock
+
+    def verify_code(self, code_object, *, bytecode_written=False):
+        super().verify_code(code_object)
+        if bytecode_written:
+            self.assertIn(self.cached, self.loader.written)
+            data = bytearray(imp.get_magic())
+            data.extend(marshal._w_long(self.loader.source_mtime))
+            data.extend(marshal.dumps(code_object))
+            self.assertEqual(self.loader.written[self.cached], bytes(data))
+
+    def test_code_with_everything(self):
+        # When everything should work.
+        code_object = self.loader.get_code(self.name)
+        self.verify_code(code_object)
+
+    def test_no_bytecode(self):
+        # If no bytecode exists then move on to the source.
+        self.loader.bytecode_path = "<does not exist>"
+        # Sanity check
+        with self.assertRaises(IOError):
+            bytecode_path = imp.cache_from_source(self.path)
+            self.loader.get_data(bytecode_path)
+        code_object = self.loader.get_code(self.name)
+        self.verify_code(code_object, bytecode_written=True)
+
+    def test_code_bad_timestamp(self):
+        # Bytecode is only used when the timestamp matches the source EXACTLY.
+        for source_mtime in (0, 2):
+            assert source_mtime != self.loader.source_mtime
+            original = self.loader.source_mtime
+            self.loader.source_mtime = source_mtime
+            # If bytecode is used then EOFError would be raised by marshal.
+            self.loader.bytecode = self.loader.bytecode[8:]
+            code_object = self.loader.get_code(self.name)
+            self.verify_code(code_object, bytecode_written=True)
+            self.loader.source_mtime = original
+
+    def test_code_bad_magic(self):
+        # Skip over bytecode with a bad magic number.
+        self.setUp(magic=b'0000')
+        # If bytecode is used then EOFError would be raised by marshal.
+        self.loader.bytecode = self.loader.bytecode[8:]
+        code_object = self.loader.get_code(self.name)
+        self.verify_code(code_object, bytecode_written=True)
+
+    def test_dont_write_bytecode(self):
+        # Bytecode is not written if sys.dont_write_bytecode is true.
+        # Can assume it is false already thanks to the skipIf class decorator.
+        try:
+            sys.dont_write_bytecode = True
+            self.loader.bytecode_path = "<does not exist>"
+            code_object = self.loader.get_code(self.name)
+            self.assertNotIn(self.cached, self.loader.written)
+        finally:
+            sys.dont_write_bytecode = False
+
+    def test_no_set_data(self):
+        # If set_data is not defined, one can still read bytecode.
+        self.setUp(magic=b'0000')
+        original_set_data = self.loader.__class__.set_data
+        try:
+            del self.loader.__class__.set_data
+            code_object = self.loader.get_code(self.name)
+            self.verify_code(code_object)
+        finally:
+            self.loader.__class__.set_data = original_set_data
+
+    def test_set_data_raises_exceptions(self):
+        # Raising NotImplementedError or IOError is okay for set_data.
+        def raise_exception(exc):
+            def closure(*args, **kwargs):
+                raise exc
+            return closure
+
+        self.setUp(magic=b'0000')
+        self.loader.set_data = raise_exception(NotImplementedError)
+        code_object = self.loader.get_code(self.name)
+        self.verify_code(code_object)
+
+
+class SourceLoaderGetSourceTests(unittest.TestCase):
+
+    """Tests for importlib.abc.SourceLoader.get_source()."""
+
+    def test_default_encoding(self):
+        # Should have no problems with UTF-8 text.
+        name = 'mod'
+        mock = SourceOnlyLoaderMock('mod.file')
+        source = 'x = "??"'
+        mock.source = source.encode('utf-8')
+        returned_source = mock.get_source(name)
+        self.assertEqual(returned_source, source)
+
+    def test_decoded_source(self):
+        # Decoding should work.
+        name = 'mod'
+        mock = SourceOnlyLoaderMock("mod.file")
+        source = "# coding: Latin-1\nx='??'"
+        assert source.encode('latin-1') != source.encode('utf-8')
+        mock.source = source.encode('latin-1')
+        returned_source = mock.get_source(name)
+        self.assertEqual(returned_source, source)
+
+    def test_universal_newlines(self):
+        # PEP 302 says universal newlines should be used.
+        name = 'mod'
+        mock = SourceOnlyLoaderMock('mod.file')
+        source = "x = 42\r\ny = -13\r\n"
+        mock.source = source.encode('utf-8')
+        expect = io.IncrementalNewlineDecoder(None, True).decode(source)
+        self.assertEqual(mock.get_source(name), expect)
+
+class AbstractMethodImplTests(unittest.TestCase):
+
+    """Test the concrete abstractmethod implementations."""
+
+    class Loader(abc.Loader):
+        def load_module(self, fullname):
+            super().load_module(fullname)
+
+    class Finder(abc.Finder):
+        def find_module(self, _):
+            super().find_module(_)
+
+    class ResourceLoader(Loader, abc.ResourceLoader):
+        def get_data(self, _):
+            super().get_data(_)
+
+    class InspectLoader(Loader, abc.InspectLoader):
+        def is_package(self, _):
+            super().is_package(_)
+
+        def get_code(self, _):
+            super().get_code(_)
+
+        def get_source(self, _):
+            super().get_source(_)
+
+    class ExecutionLoader(InspectLoader, abc.ExecutionLoader):
+        def get_filename(self, _):
+            super().get_filename(_)
+
+    class SourceLoader(ResourceLoader, ExecutionLoader, abc.SourceLoader):
+        pass
+
+    class PyLoader(ResourceLoader, InspectLoader, abc.PyLoader):
+        def source_path(self, _):
+            super().source_path(_)
+
+    class PyPycLoader(PyLoader, abc.PyPycLoader):
+        def bytecode_path(self, _):
+            super().bytecode_path(_)
+
+        def source_mtime(self, _):
+            super().source_mtime(_)
+
+        def write_bytecode(self, _, _2):
+            super().write_bytecode(_, _2)
+
+    def raises_NotImplementedError(self, ins, *args):
+        for method_name in args:
+            method = getattr(ins, method_name)
+            arg_count = len(inspect.getfullargspec(method)[0]) - 1
+            args = [''] * arg_count
+            try:
+                method(*args)
+            except NotImplementedError:
+                pass
+            else:
+                msg = "{}.{} did not raise NotImplementedError"
+                self.fail(msg.format(ins.__class__.__name__, method_name))
+
+    def test_Loader(self):
+        self.raises_NotImplementedError(self.Loader(), 'load_module')
+
+    # XXX misplaced; should be somewhere else
+    def test_Finder(self):
+        self.raises_NotImplementedError(self.Finder(), 'find_module')
+
+    def test_ResourceLoader(self):
+        self.raises_NotImplementedError(self.ResourceLoader(), 'load_module',
+                                        'get_data')
+
+    def test_InspectLoader(self):
+        self.raises_NotImplementedError(self.InspectLoader(), 'load_module',
+                                        'is_package', 'get_code', 'get_source')
+
+    def test_ExecutionLoader(self):
+        self.raises_NotImplementedError(self.ExecutionLoader(), 'load_module',
+                                        'is_package', 'get_code', 'get_source',
+                                        'get_filename')
+
+    def test_SourceLoader(self):
+        ins = self.SourceLoader()
+        # Required abstractmethods.
+        self.raises_NotImplementedError(ins, 'get_filename', 'get_data')
+        # Optional abstractmethods.
+        self.raises_NotImplementedError(ins,'path_mtime', 'set_data')
+
+    def test_PyLoader(self):
+        self.raises_NotImplementedError(self.PyLoader(), 'source_path',
+                                        'get_data', 'is_package')
+
+    def test_PyPycLoader(self):
+        self.raises_NotImplementedError(self.PyPycLoader(), 'source_path',
+                                        'source_mtime', 'bytecode_path',
+                                        'write_bytecode')
+
+
 def test_main():
     from test.support import run_unittest
-    run_unittest(PyLoaderTests, PyLoaderInterfaceTests, PyLoaderGetSourceTests,
+    run_unittest(PyLoaderTests, PyLoaderCompatTests,
+                    PyLoaderInterfaceTests,
                     PyPycLoaderTests, PyPycLoaderInterfaceTests,
                     SkipWritingBytecodeTests, RegeneratedBytecodeTests,
-                    BadBytecodeFailureTests, MissingPathsTests)
+                    BadBytecodeFailureTests, MissingPathsTests,
+                    SourceOnlyLoaderTests,
+                    SourceLoaderBytecodeTests,
+                    SourceLoaderGetSourceTests,
+                    AbstractMethodImplTests)
 
 
 if __name__ == '__main__':

Modified: python/branches/py3k-dtoa/Lib/importlib/test/source/test_case_sensitivity.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/importlib/test/source/test_case_sensitivity.py	(original)
+++ python/branches/py3k-dtoa/Lib/importlib/test/source/test_case_sensitivity.py	Tue Jul  6 17:55:27 2010
@@ -19,7 +19,9 @@
     assert name != name.lower()
 
     def find(self, path):
-        finder = _bootstrap._PyPycFileFinder(path)
+        finder = _bootstrap._FileFinder(path,
+                                        _bootstrap._SourceFinderDetails(),
+                                        _bootstrap._SourcelessFinderDetails())
         return finder.find_module(self.name)
 
     def sensitivity_test(self):
@@ -27,7 +29,7 @@
         sensitive_pkg = 'sensitive.{0}'.format(self.name)
         insensitive_pkg = 'insensitive.{0}'.format(self.name.lower())
         context = source_util.create_modules(insensitive_pkg, sensitive_pkg)
-        with  context as mapping:
+        with context as mapping:
             sensitive_path = os.path.join(mapping['.root'], 'sensitive')
             insensitive_path = os.path.join(mapping['.root'], 'insensitive')
             return self.find(sensitive_path), self.find(insensitive_path)
@@ -37,7 +39,7 @@
             env.unset('PYTHONCASEOK')
             sensitive, insensitive = self.sensitivity_test()
             self.assertTrue(hasattr(sensitive, 'load_module'))
-            self.assertIn(self.name, sensitive._base_path)
+            self.assertIn(self.name, sensitive.get_filename(self.name))
             self.assertIsNone(insensitive)
 
     def test_insensitive(self):
@@ -45,9 +47,9 @@
             env.set('PYTHONCASEOK', '1')
             sensitive, insensitive = self.sensitivity_test()
             self.assertTrue(hasattr(sensitive, 'load_module'))
-            self.assertIn(self.name, sensitive._base_path)
+            self.assertIn(self.name, sensitive.get_filename(self.name))
             self.assertTrue(hasattr(insensitive, 'load_module'))
-            self.assertIn(self.name, insensitive._base_path)
+            self.assertIn(self.name, insensitive.get_filename(self.name))
 
 
 def test_main():

Modified: python/branches/py3k-dtoa/Lib/importlib/test/source/test_file_loader.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/importlib/test/source/test_file_loader.py	(original)
+++ python/branches/py3k-dtoa/Lib/importlib/test/source/test_file_loader.py	Tue Jul  6 17:55:27 2010
@@ -1,9 +1,11 @@
 import importlib
 from importlib import _bootstrap
 from .. import abc
+from .. import util
 from . import util as source_util
 
 import imp
+import marshal
 import os
 import py_compile
 import stat
@@ -23,8 +25,7 @@
     # [basic]
     def test_module(self):
         with source_util.create_modules('_temp') as mapping:
-            loader = _bootstrap._PyPycFileLoader('_temp', mapping['_temp'],
-                                                    False)
+            loader = _bootstrap._SourceFileLoader('_temp', mapping['_temp'])
             module = loader.load_module('_temp')
             self.assertTrue('_temp' in sys.modules)
             check = {'__name__': '_temp', '__file__': mapping['_temp'],
@@ -34,9 +35,8 @@
 
     def test_package(self):
         with source_util.create_modules('_pkg.__init__') as mapping:
-            loader = _bootstrap._PyPycFileLoader('_pkg',
-                                                 mapping['_pkg.__init__'],
-                                                 True)
+            loader = _bootstrap._SourceFileLoader('_pkg',
+                                                 mapping['_pkg.__init__'])
             module = loader.load_module('_pkg')
             self.assertTrue('_pkg' in sys.modules)
             check = {'__name__': '_pkg', '__file__': mapping['_pkg.__init__'],
@@ -48,8 +48,8 @@
 
     def test_lacking_parent(self):
         with source_util.create_modules('_pkg.__init__', '_pkg.mod')as mapping:
-            loader = _bootstrap._PyPycFileLoader('_pkg.mod',
-                                                    mapping['_pkg.mod'], False)
+            loader = _bootstrap._SourceFileLoader('_pkg.mod',
+                                                    mapping['_pkg.mod'])
             module = loader.load_module('_pkg.mod')
             self.assertTrue('_pkg.mod' in sys.modules)
             check = {'__name__': '_pkg.mod', '__file__': mapping['_pkg.mod'],
@@ -63,8 +63,7 @@
 
     def test_module_reuse(self):
         with source_util.create_modules('_temp') as mapping:
-            loader = _bootstrap._PyPycFileLoader('_temp', mapping['_temp'],
-                                                    False)
+            loader = _bootstrap._SourceFileLoader('_temp', mapping['_temp'])
             module = loader.load_module('_temp')
             module_id = id(module)
             module_dict_id = id(module.__dict__)
@@ -74,7 +73,7 @@
             # everything that has happened above can be too fast;
             # force an mtime on the source that is guaranteed to be different
             # than the original mtime.
-            loader.source_mtime = self.fake_mtime(loader.source_mtime)
+            loader.path_mtime = self.fake_mtime(loader.path_mtime)
             module = loader.load_module('_temp')
             self.assertTrue('testing_var' in module.__dict__,
                          "'testing_var' not in "
@@ -94,8 +93,7 @@
                 setattr(orig_module, attr, value)
             with open(mapping[name], 'w') as file:
                 file.write('+++ bad syntax +++')
-            loader = _bootstrap._PyPycFileLoader('_temp', mapping['_temp'],
-                                                    False)
+            loader = _bootstrap._SourceFileLoader('_temp', mapping['_temp'])
             with self.assertRaises(SyntaxError):
                 loader.load_module(name)
             for attr in attributes:
@@ -106,17 +104,32 @@
         with source_util.create_modules('_temp') as mapping:
             with open(mapping['_temp'], 'w') as file:
                 file.write('=')
-            loader = _bootstrap._PyPycFileLoader('_temp', mapping['_temp'],
-                                                    False)
+            loader = _bootstrap._SourceFileLoader('_temp', mapping['_temp'])
             with self.assertRaises(SyntaxError):
                 loader.load_module('_temp')
             self.assertTrue('_temp' not in sys.modules)
 
+    def test_file_from_empty_string_dir(self):
+        # Loading a module found from an empty string entry on sys.path should
+        # not only work, but keep all attributes relative.
+        with open('_temp.py', 'w') as file:
+            file.write("# test file for importlib")
+        try:
+            with util.uncache('_temp'):
+                loader = _bootstrap._SourceFileLoader('_temp', '_temp.py')
+                mod = loader.load_module('_temp')
+                self.assertEqual('_temp.py', mod.__file__)
+                self.assertEqual(imp.cache_from_source('_temp.py'),
+                                 mod.__cached__)
+
+        finally:
+            os.unlink('_temp.py')
+
 
 class BadBytecodeTest(unittest.TestCase):
 
     def import_(self, file, module_name):
-        loader = _bootstrap._PyPycFileLoader(module_name, file, False)
+        loader = self.loader(module_name, file)
         module = loader.load_module(module_name)
         self.assertTrue(module_name in sys.modules)
 
@@ -129,101 +142,156 @@
         except KeyError:
             pass
         py_compile.compile(mapping[name])
-        bytecode_path = imp.cache_from_source(mapping[name])
-        with open(bytecode_path, 'rb') as file:
-            bc = file.read()
-        new_bc = manipulator(bc)
-        with open(bytecode_path, 'wb') as file:
-            if new_bc:
-                file.write(new_bc)
-        if del_source:
+        if not del_source:
+            bytecode_path = imp.cache_from_source(mapping[name])
+        else:
             os.unlink(mapping[name])
-            make_legacy_pyc(mapping[name])
+            bytecode_path = make_legacy_pyc(mapping[name])
+        if manipulator:
+            with open(bytecode_path, 'rb') as file:
+                bc = file.read()
+                new_bc = manipulator(bc)
+            with open(bytecode_path, 'wb') as file:
+                if new_bc is not None:
+                    file.write(new_bc)
         return bytecode_path
 
-    @source_util.writes_bytecode_files
-    def test_empty_file(self):
-        # When a .pyc is empty, regenerate it if possible, else raise
-        # ImportError.
+    def _test_empty_file(self, test, *, del_source=False):
         with source_util.create_modules('_temp') as mapping:
             bc_path = self.manipulate_bytecode('_temp', mapping,
-                                                lambda bc: None)
-            self.import_(mapping['_temp'], '_temp')
-            with open(bc_path, 'rb') as file:
-                self.assertGreater(len(file.read()), 8)
-            self.manipulate_bytecode('_temp', mapping, lambda bc: None,
-                                        del_source=True)
-            with self.assertRaises(ImportError):
-                self.import_(mapping['_temp'], '_temp')
+                                                lambda bc: b'',
+                                                del_source=del_source)
+            test('_temp', mapping, bc_path)
 
     @source_util.writes_bytecode_files
-    def test_partial_magic(self):
+    def _test_partial_magic(self, test, *, del_source=False):
         # When their are less than 4 bytes to a .pyc, regenerate it if
         # possible, else raise ImportError.
         with source_util.create_modules('_temp') as mapping:
             bc_path = self.manipulate_bytecode('_temp', mapping,
-                                                lambda bc: bc[:3])
-            self.import_(mapping['_temp'], '_temp')
-            with open(bc_path, 'rb') as file:
-                self.assertGreater(len(file.read()), 8)
-            self.manipulate_bytecode('_temp', mapping, lambda bc: bc[:3],
-                                        del_source=True)
+                                                lambda bc: bc[:3],
+                                                del_source=del_source)
+            test('_temp', mapping, bc_path)
+
+    def _test_magic_only(self, test, *, del_source=False):
+        with source_util.create_modules('_temp') as mapping:
+            bc_path = self.manipulate_bytecode('_temp', mapping,
+                                                lambda bc: bc[:4],
+                                                del_source=del_source)
+            test('_temp', mapping, bc_path)
+
+    def _test_partial_timestamp(self, test, *, del_source=False):
+        with source_util.create_modules('_temp') as mapping:
+            bc_path = self.manipulate_bytecode('_temp', mapping,
+                                                lambda bc: bc[:7],
+                                                del_source=del_source)
+            test('_temp', mapping, bc_path)
+
+    def _test_no_marshal(self, *, del_source=False):
+        with source_util.create_modules('_temp') as mapping:
+            bc_path = self.manipulate_bytecode('_temp', mapping,
+                                                lambda bc: bc[:8],
+                                                del_source=del_source)
+            file_path = mapping['_temp'] if not del_source else bc_path
+            with self.assertRaises(EOFError):
+                self.import_(file_path, '_temp')
+
+    def _test_non_code_marshal(self, *, del_source=False):
+        with source_util.create_modules('_temp') as mapping:
+            bytecode_path = self.manipulate_bytecode('_temp', mapping,
+                                    lambda bc: bc[:8] + marshal.dumps(b'abcd'),
+                                    del_source=del_source)
+            file_path = mapping['_temp'] if not del_source else bytecode_path
             with self.assertRaises(ImportError):
-                self.import_(mapping['_temp'], '_temp')
+                self.import_(file_path, '_temp')
+
+    def _test_bad_marshal(self, *, del_source=False):
+        with source_util.create_modules('_temp') as mapping:
+            bytecode_path = self.manipulate_bytecode('_temp', mapping,
+                                                lambda bc: bc[:8] + b'<test>',
+                                                del_source=del_source)
+            file_path = mapping['_temp'] if not del_source else bytecode_path
+            with self.assertRaises(ValueError):
+                self.import_(file_path, '_temp')
+
+    def _test_bad_magic(self, test, *, del_source=False):
+        with source_util.create_modules('_temp') as mapping:
+            bc_path = self.manipulate_bytecode('_temp', mapping,
+                                    lambda bc: b'\x00\x00\x00\x00' + bc[4:])
+            test('_temp', mapping, bc_path)
+
+
+class SourceLoaderBadBytecodeTest(BadBytecodeTest):
+
+    loader = _bootstrap._SourceFileLoader
+
+    @source_util.writes_bytecode_files
+    def test_empty_file(self):
+        # When a .pyc is empty, regenerate it if possible, else raise
+        # ImportError.
+        def test(name, mapping, bytecode_path):
+            self.import_(mapping[name], name)
+            with open(bytecode_path, 'rb') as file:
+                self.assertGreater(len(file.read()), 8)
+
+        self._test_empty_file(test)
+
+    def test_partial_magic(self):
+        def test(name, mapping, bytecode_path):
+            self.import_(mapping[name], name)
+            with open(bytecode_path, 'rb') as file:
+                self.assertGreater(len(file.read()), 8)
+
+        self._test_partial_magic(test)
 
     @source_util.writes_bytecode_files
     def test_magic_only(self):
         # When there is only the magic number, regenerate the .pyc if possible,
         # else raise EOFError.
-        with source_util.create_modules('_temp') as mapping:
-            bc_path = self.manipulate_bytecode('_temp', mapping,
-                                                lambda bc: bc[:4])
-            self.import_(mapping['_temp'], '_temp')
-            with open(bc_path, 'rb') as file:
+        def test(name, mapping, bytecode_path):
+            self.import_(mapping[name], name)
+            with open(bytecode_path, 'rb') as file:
                 self.assertGreater(len(file.read()), 8)
-            self.manipulate_bytecode('_temp', mapping, lambda bc: bc[:4],
-                                        del_source=True)
-            with self.assertRaises(EOFError):
-                self.import_(mapping['_temp'], '_temp')
+
+    @source_util.writes_bytecode_files
+    def test_bad_magic(self):
+        # When the magic number is different, the bytecode should be
+        # regenerated.
+        def test(name, mapping, bytecode_path):
+            self.import_(mapping[name], name)
+            with open(bytecode_path, 'rb') as bytecode_file:
+                self.assertEqual(bytecode_file.read(4), imp.get_magic())
+
+        self._test_bad_magic(test)
 
     @source_util.writes_bytecode_files
     def test_partial_timestamp(self):
         # When the timestamp is partial, regenerate the .pyc, else
         # raise EOFError.
-        with source_util.create_modules('_temp') as mapping:
-            bc_path = self.manipulate_bytecode('_temp', mapping,
-                                                lambda bc: bc[:7])
-            self.import_(mapping['_temp'], '_temp')
+        def test(name, mapping, bc_path):
+            self.import_(mapping[name], name)
             with open(bc_path, 'rb') as file:
                 self.assertGreater(len(file.read()), 8)
-            self.manipulate_bytecode('_temp', mapping, lambda bc: bc[:7],
-                                        del_source=True)
-            with self.assertRaises(EOFError):
-                self.import_(mapping['_temp'], '_temp')
 
     @source_util.writes_bytecode_files
     def test_no_marshal(self):
         # When there is only the magic number and timestamp, raise EOFError.
-        with source_util.create_modules('_temp') as mapping:
-            bc_path = self.manipulate_bytecode('_temp', mapping,
-                                                lambda bc: bc[:8])
-            with self.assertRaises(EOFError):
-                self.import_(mapping['_temp'], '_temp')
+        self._test_no_marshal()
 
     @source_util.writes_bytecode_files
-    def test_bad_magic(self):
-        # When the magic number is different, the bytecode should be
-        # regenerated.
-        with source_util.create_modules('_temp') as mapping:
-            bc_path = self.manipulate_bytecode('_temp', mapping,
-                                    lambda bc: b'\x00\x00\x00\x00' + bc[4:])
-            self.import_(mapping['_temp'], '_temp')
-            with open(bc_path, 'rb') as bytecode_file:
-                self.assertEqual(bytecode_file.read(4), imp.get_magic())
+    def test_non_code_marshal(self):
+        self._test_non_code_marshal()
+        # XXX ImportError when sourceless
+
+    # [bad marshal]
+    @source_util.writes_bytecode_files
+    def test_bad_marshal(self):
+        # Bad marshal data should raise a ValueError.
+        self._test_bad_marshal()
 
     # [bad timestamp]
     @source_util.writes_bytecode_files
-    def test_bad_bytecode(self):
+    def test_old_timestamp(self):
         # When the timestamp is older than the source, bytecode should be
         # regenerated.
         zeros = b'\x00\x00\x00\x00'
@@ -240,23 +308,6 @@
                 bytecode_file.seek(4)
                 self.assertEqual(bytecode_file.read(4), source_timestamp)
 
-    # [bad marshal]
-    @source_util.writes_bytecode_files
-    def test_bad_marshal(self):
-        # Bad marshal data should raise a ValueError.
-        with source_util.create_modules('_temp') as mapping:
-            bytecode_path = imp.cache_from_source(mapping['_temp'])
-            source_mtime = os.path.getmtime(mapping['_temp'])
-            source_timestamp = importlib._w_long(source_mtime)
-            source_util.ensure_bytecode_path(bytecode_path)
-            with open(bytecode_path, 'wb') as bytecode_file:
-                bytecode_file.write(imp.get_magic())
-                bytecode_file.write(source_timestamp)
-                bytecode_file.write(b'AAAA')
-            with self.assertRaises(ValueError):
-                self.import_(mapping['_temp'], '_temp')
-            self.assertTrue('_temp' not in sys.modules)
-
     # [bytecode read-only]
     @source_util.writes_bytecode_files
     def test_read_only_bytecode(self):
@@ -279,9 +330,57 @@
                 os.chmod(bytecode_path, stat.S_IWUSR)
 
 
+class SourcelessLoaderBadBytecodeTest(BadBytecodeTest):
+
+    loader = _bootstrap._SourcelessFileLoader
+
+    def test_empty_file(self):
+        def test(name, mapping, bytecode_path):
+            with self.assertRaises(ImportError):
+                self.import_(bytecode_path, name)
+
+        self._test_empty_file(test, del_source=True)
+
+    def test_partial_magic(self):
+        def test(name, mapping, bytecode_path):
+            with self.assertRaises(ImportError):
+                self.import_(bytecode_path, name)
+        self._test_partial_magic(test, del_source=True)
+
+    def test_magic_only(self):
+        def test(name, mapping, bytecode_path):
+            with self.assertRaises(EOFError):
+                self.import_(bytecode_path, name)
+
+        self._test_magic_only(test, del_source=True)
+
+    def test_bad_magic(self):
+        def test(name, mapping, bytecode_path):
+            with self.assertRaises(ImportError):
+                self.import_(bytecode_path, name)
+
+        self._test_bad_magic(test, del_source=True)
+
+    def test_partial_timestamp(self):
+        def test(name, mapping, bytecode_path):
+            with self.assertRaises(EOFError):
+                self.import_(bytecode_path, name)
+
+        self._test_partial_timestamp(test, del_source=True)
+
+    def test_no_marshal(self):
+        self._test_no_marshal(del_source=True)
+
+    def test_non_code_marshal(self):
+        self._test_non_code_marshal(del_source=True)
+
+
 def test_main():
     from test.support import run_unittest
-    run_unittest(SimpleTest, BadBytecodeTest)
+    run_unittest(SimpleTest,
+                 SourceLoaderBadBytecodeTest,
+                 SourcelessLoaderBadBytecodeTest
+                 )
 
 
 if __name__ == '__main__':

Modified: python/branches/py3k-dtoa/Lib/importlib/test/source/test_finder.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/importlib/test/source/test_finder.py	(original)
+++ python/branches/py3k-dtoa/Lib/importlib/test/source/test_finder.py	Tue Jul  6 17:55:27 2010
@@ -34,7 +34,9 @@
     """
 
     def import_(self, root, module):
-        finder = _bootstrap._PyPycFileFinder(root)
+        finder = _bootstrap._FileFinder(root,
+                                        _bootstrap._SourceFinderDetails(),
+                                        _bootstrap._SourcelessFinderDetails())
         return finder.find_module(module)
 
     def run_test(self, test, create=None, *, compile_=None, unlink=None):
@@ -113,10 +115,9 @@
 
     # [package over modules]
     def test_package_over_module(self):
-        # XXX This is not a blackbox test!
         name = '_temp'
         loader = self.run_test(name, {'{0}.__init__'.format(name), name})
-        self.assertTrue('__init__' in loader._base_path)
+        self.assertTrue('__init__' in loader.get_filename(name))
 
 
     def test_failure(self):
@@ -131,6 +132,17 @@
             with self.assertRaises(ImportWarning):
                 self.run_test('pkg', {'pkg.__init__'}, unlink={'pkg.__init__'})
 
+    def test_empty_string_for_dir(self):
+        # The empty string from sys.path means to search in the cwd.
+        finder = _bootstrap._FileFinder('', _bootstrap._SourceFinderDetails())
+        with open('mod.py', 'w') as file:
+            file.write("# test file for importlib")
+        try:
+            loader = finder.find_module('mod')
+            self.assertTrue(hasattr(loader, 'load_module'))
+        finally:
+            os.unlink('mod.py')
+
 
 def test_main():
     from test.support import run_unittest

Modified: python/branches/py3k-dtoa/Lib/importlib/test/source/test_path_hook.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/importlib/test/source/test_path_hook.py	(original)
+++ python/branches/py3k-dtoa/Lib/importlib/test/source/test_path_hook.py	Tue Jul  6 17:55:27 2010
@@ -8,11 +8,14 @@
     """Test the path hook for source."""
 
     def test_success(self):
-        # XXX Only work on existing directories?
         with source_util.create_modules('dummy') as mapping:
-            self.assertTrue(hasattr(_bootstrap._FileFinder(mapping['.root']),
+            self.assertTrue(hasattr(_bootstrap._file_path_hook(mapping['.root']),
                                  'find_module'))
 
+    def test_empty_string(self):
+        # The empty string represents the cwd.
+        self.assertTrue(hasattr(_bootstrap._file_path_hook(''), 'find_module'))
+
 
 def test_main():
     from test.support import run_unittest

Modified: python/branches/py3k-dtoa/Lib/importlib/test/source/test_source_encoding.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/importlib/test/source/test_source_encoding.py	(original)
+++ python/branches/py3k-dtoa/Lib/importlib/test/source/test_source_encoding.py	Tue Jul  6 17:55:27 2010
@@ -35,8 +35,8 @@
         with source_util.create_modules(self.module_name) as mapping:
             with open(mapping[self.module_name], 'wb') as file:
                 file.write(source)
-            loader = _bootstrap._PyPycFileLoader(self.module_name,
-                                       mapping[self.module_name], False)
+            loader = _bootstrap._SourceFileLoader(self.module_name,
+                                       mapping[self.module_name])
             return loader.load_module(self.module_name)
 
     def create_source(self, encoding):
@@ -97,8 +97,8 @@
         with source_util.create_modules(module_name) as mapping:
             with open(mapping[module_name], 'wb') as file:
                 file.write(source)
-            loader = _bootstrap._PyPycFileLoader(module_name,
-                                                 mapping[module_name], False)
+            loader = _bootstrap._SourceFileLoader(module_name,
+                                                 mapping[module_name])
             return loader.load_module(module_name)
 
     # [cr]

Modified: python/branches/py3k-dtoa/Lib/importlib/test/test_abc.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/importlib/test/test_abc.py	(original)
+++ python/branches/py3k-dtoa/Lib/importlib/test/test_abc.py	Tue Jul  6 17:55:27 2010
@@ -59,6 +59,11 @@
     subclasses = [abc.PyLoader]
 
 
+class SourceLoader(InheritanceTests, unittest.TestCase):
+
+    superclasses = [abc.ResourceLoader, abc.ExecutionLoader]
+
+
 class PyLoader(InheritanceTests, unittest.TestCase):
 
     superclasses = [abc.Loader, abc.ResourceLoader, abc.ExecutionLoader]

Modified: python/branches/py3k-dtoa/Lib/importlib/test/test_api.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/importlib/test/test_api.py	(original)
+++ python/branches/py3k-dtoa/Lib/importlib/test/test_api.py	Tue Jul  6 17:55:27 2010
@@ -27,7 +27,7 @@
                 self.assertEqual(module.__name__, name)
 
     def test_shallow_relative_package_import(self):
-        # Test importing a module from a package through a relatve import.
+        # Test importing a module from a package through a relative import.
         pkg_name = 'pkg'
         pkg_long_name = '{0}.__init__'.format(pkg_name)
         module_name = 'mod'

Modified: python/branches/py3k-dtoa/Lib/importlib/test/test_util.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/importlib/test/test_util.py	(original)
+++ python/branches/py3k-dtoa/Lib/importlib/test/test_util.py	Tue Jul  6 17:55:27 2010
@@ -40,7 +40,7 @@
         with test_util.uncache(name):
             sys.modules[name] = module
             returned_module = self.return_module(name)
-            self.assertTrue(sys.modules[name] is returned_module)
+            self.assertIs(returned_module, sys.modules[name])
 
     def test_new_module_failure(self):
         # Test that a module is removed from sys.modules if added but an
@@ -57,7 +57,7 @@
         with test_util.uncache(name):
             sys.modules[name] = module
             self.raise_exception(name)
-            self.assertTrue(sys.modules[name] is module)
+            self.assertIs(module, sys.modules[name])
 
 
 class SetPackageTests(unittest.TestCase):

Modified: python/branches/py3k-dtoa/Lib/inspect.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/inspect.py	(original)
+++ python/branches/py3k-dtoa/Lib/inspect.py	Tue Jul  6 17:55:27 2010
@@ -437,7 +437,9 @@
     if info: return info[0]
 
 def getsourcefile(object):
-    """Return the Python source file an object was defined in, if it exists."""
+    """Return the filename that can be used to locate an object's source.
+    Return None if no way can be identified to get the source.
+    """
     filename = getfile(object)
     if filename[-4:].lower() in ('.pyc', '.pyo'):
         filename = filename[:-4] + '.py'
@@ -450,6 +452,9 @@
     # only return a non-existent filename if the module has a PEP 302 loader
     if hasattr(getmodule(object, filename), '__loader__'):
         return filename
+    # or it is in the linecache
+    if filename in linecache.cache:
+        return filename
 
 def getabsfile(object, _filename=None):
     """Return an absolute path to the source or compiled file for an object.
@@ -954,7 +959,7 @@
         else:
             arg2value[varargs] = ()
     elif 0 < num_args < num_pos:
-        raise TypeError('%s() takes %s %d %s (%d given)' % (
+        raise TypeError('%s() takes %s %d positional %s (%d given)' % (
             f_name, 'at most' if defaults else 'exactly', num_args,
             'arguments' if num_args > 1 else 'argument', num_total))
     elif num_args == 0 and num_total:

Modified: python/branches/py3k-dtoa/Lib/lib2to3/fixes/fix_itertools_imports.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/lib2to3/fixes/fix_itertools_imports.py	(original)
+++ python/branches/py3k-dtoa/Lib/lib2to3/fixes/fix_itertools_imports.py	Tue Jul  6 17:55:27 2010
@@ -20,6 +20,9 @@
             if child.type == token.NAME:
                 member = child.value
                 name_node = child
+            elif child.type == token.STAR:
+                # Just leave the import as is.
+                return
             else:
                 assert child.type == syms.import_as_name
                 name_node = child.children[0]
@@ -44,8 +47,8 @@
             children[-1].remove()
 
         # If there are no imports left, just get rid of the entire statement
-        if not (imports.children or getattr(imports, 'value', None)) or \
-                imports.parent is None:
+        if (not (imports.children or getattr(imports, 'value', None)) or
+            imports.parent is None):
             p = node.prefix
             node = BlankLine()
             node.prefix = p

Modified: python/branches/py3k-dtoa/Lib/lib2to3/tests/test_fixers.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/lib2to3/tests/test_fixers.py	(original)
+++ python/branches/py3k-dtoa/Lib/lib2to3/tests/test_fixers.py	Tue Jul  6 17:55:27 2010
@@ -3670,6 +3670,10 @@
         a = "from itertools import bar, filterfalse, foo"
         self.check(b, a)
 
+    def test_import_star(self):
+        s = "from itertools import *"
+        self.unchanged(s)
+
 
     def test_unchanged(self):
         s = "from itertools import foo"

Modified: python/branches/py3k-dtoa/Lib/logging/config.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/logging/config.py	(original)
+++ python/branches/py3k-dtoa/Lib/logging/config.py	Tue Jul  6 17:55:27 2010
@@ -386,15 +386,21 @@
         """
         name = s.split('.')
         used = name.pop(0)
-        found = self.importer(used)
-        for frag in name:
-            used += '.' + frag
-            try:
-                found = getattr(found, frag)
-            except AttributeError:
-                self.importer(used)
-                found = getattr(found, frag)
-        return found
+        try:
+            found = self.importer(used)
+            for frag in name:
+                used += '.' + frag
+                try:
+                    found = getattr(found, frag)
+                except AttributeError:
+                    self.importer(used)
+                    found = getattr(found, frag)
+            return found
+        except ImportError:
+            e, tb = sys.exc_info()[1:]
+            v = ValueError('Cannot resolve %r: %s' % (s, e))
+            v.__cause__, v.__traceback__ = e, tb
+            raise v
 
     def ext_convert(self, value):
         """Default converter for the ext:// protocol."""
@@ -873,6 +879,8 @@
         def run(self):
             server = self.rcvr(port=self.port, handler=self.hdlr,
                                ready=self.ready)
+            if self.port == 0:
+                self.port = server.server_address[1]
             self.ready.set()
             global _listener
             logging._acquireLock()

Modified: python/branches/py3k-dtoa/Lib/macpath.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/macpath.py	(original)
+++ python/branches/py3k-dtoa/Lib/macpath.py	Tue Jul  6 17:55:27 2010
@@ -32,6 +32,9 @@
 # Normalize the case of a pathname.  Dummy in Posix, but <s>.lower() here.
 
 def normcase(path):
+    if not isinstance(path, (bytes, str)):
+        raise TypeError("normcase() argument must be str or bytes, "
+                        "not '{}'".format(path.__class__.__name__))
     return path.lower()
 
 

Modified: python/branches/py3k-dtoa/Lib/ntpath.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/ntpath.py	(original)
+++ python/branches/py3k-dtoa/Lib/ntpath.py	Tue Jul  6 17:55:27 2010
@@ -78,6 +78,9 @@
     """Normalize case of pathname.
 
     Makes all characters lowercase and all slashes into backslashes."""
+    if not isinstance(s, (bytes, str)):
+        raise TypeError("normcase() argument must be str or bytes, "
+                        "not '{}'".format(s.__class__.__name__))
     return s.replace(_get_altsep(s), _get_sep(s)).lower()
 
 

Modified: python/branches/py3k-dtoa/Lib/os2emxpath.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/os2emxpath.py	(original)
+++ python/branches/py3k-dtoa/Lib/os2emxpath.py	Tue Jul  6 17:55:27 2010
@@ -36,6 +36,9 @@
     """Normalize case of pathname.
 
     Makes all characters lowercase and all altseps into seps."""
+    if not isinstance(s, (bytes, str)):
+        raise TypeError("normcase() argument must be str or bytes, "
+                        "not '{}'".format(s.__class__.__name__))
     return s.replace('\\', '/').lower()
 
 

Modified: python/branches/py3k-dtoa/Lib/pdb.doc
==============================================================================
--- python/branches/py3k-dtoa/Lib/pdb.doc	(original)
+++ python/branches/py3k-dtoa/Lib/pdb.doc	Tue Jul  6 17:55:27 2010
@@ -68,13 +68,13 @@
         An arrow indicates the "current frame", which determines the
         context of most commands.
 
-d(own)
-        Move the current frame one level down in the stack trace
-        (to a newer frame).
-
-u(p)
-        Move the current frame one level up in the stack trace
-        (to an older frame).
+d(own) [ count ]
+        Move the current frame count (default one) levels down in the
+        stack trace (to a newer frame).
+
+u(p) [ count ]
+        Move the current frame count (default one) levels up in the
+        stack trace (to an older frame).
 
 b(reak) [ ([filename:]lineno | function) [, condition] ]
         With a filename:line number argument, set a break there.  If
@@ -138,8 +138,8 @@
 run [args...]
         Restart the debugged python program. If a string is supplied it is
         splitted with "shlex", and the result is used as the new sys.argv.
-	History, breakpoints, actions and debugger options are preserved.
-	"restart" is an alias for "run".
+        History, breakpoints, actions and debugger options are preserved.
+        "restart" is an alias for "run".
 
 c(ont(inue))
         Continue execution, only stop when a breakpoint is encountered.

Modified: python/branches/py3k-dtoa/Lib/pdb.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/pdb.py	(original)
+++ python/branches/py3k-dtoa/Lib/pdb.py	Tue Jul  6 17:55:27 2010
@@ -618,26 +618,44 @@
     do_w = do_where
     do_bt = do_where
 
+    def _select_frame(self, number):
+        assert 0 <= number < len(self.stack)
+        self.curindex = number
+        self.curframe = self.stack[self.curindex][0]
+        self.curframe_locals = self.curframe.f_locals
+        self.print_stack_entry(self.stack[self.curindex])
+        self.lineno = None
+
     def do_up(self, arg):
         if self.curindex == 0:
             print('*** Oldest frame', file=self.stdout)
+            return
+        try:
+            count = int(arg or 1)
+        except ValueError:
+            print('*** Invalid frame count (%s)' % arg, file=self.stdout)
+            return
+        if count < 0:
+            newframe = 0
         else:
-            self.curindex = self.curindex - 1
-            self.curframe = self.stack[self.curindex][0]
-            self.curframe_locals = self.curframe.f_locals
-            self.print_stack_entry(self.stack[self.curindex])
-            self.lineno = None
+            newframe = max(0, self.curindex - count)
+        self._select_frame(newframe)
     do_u = do_up
 
     def do_down(self, arg):
         if self.curindex + 1 == len(self.stack):
             print('*** Newest frame', file=self.stdout)
+            return
+        try:
+            count = int(arg or 1)
+        except ValueError:
+            print('*** Invalid frame count (%s)' % arg, file=self.stdout)
+            return
+        if count < 0:
+            newframe = len(self.stack) - 1
         else:
-            self.curindex = self.curindex + 1
-            self.curframe = self.stack[self.curindex][0]
-            self.curframe_locals = self.curframe.f_locals
-            self.print_stack_entry(self.stack[self.curindex])
-            self.lineno = None
+            newframe = min(len(self.stack) - 1, self.curindex + count)
+        self._select_frame(newframe)
     do_d = do_down
 
     def do_until(self, arg):

Modified: python/branches/py3k-dtoa/Lib/pickletools.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/pickletools.py	(original)
+++ python/branches/py3k-dtoa/Lib/pickletools.py	Tue Jul  6 17:55:27 2010
@@ -2330,4 +2330,43 @@
     return doctest.testmod()
 
 if __name__ == "__main__":
-    _test()
+    import sys, argparse
+    parser = argparse.ArgumentParser(
+        description='disassemble one or more pickle files')
+    parser.add_argument(
+        'pickle_file', type=argparse.FileType('br'),
+        nargs='*', help='the pickle file')
+    parser.add_argument(
+        '-o', '--output', default=sys.stdout, type=argparse.FileType('w'),
+        help='the file where the output should be written')
+    parser.add_argument(
+        '-m', '--memo', action='store_true',
+        help='preserve memo between disassemblies')
+    parser.add_argument(
+        '-l', '--indentlevel', default=4, type=int,
+        help='the number of blanks by which to indent a new MARK level')
+    parser.add_argument(
+        '-p', '--preamble', default="==> {name} <==",
+        help='if more than one pickle file is specified, print this before'
+        ' each disassembly')
+    parser.add_argument(
+        '-t', '--test', action='store_true',
+        help='run self-test suite')
+    parser.add_argument(
+        '-v', action='store_true',
+        help='run verbosely; only affects self-test run')
+    args = parser.parse_args()
+    if args.test:
+        _test()
+    else:
+        if not args.pickle_file:
+            parser.print_help()
+        elif len(args.pickle_file) == 1:
+            dis(args.pickle_file[0], args.output,
+                indentlevel=args.indentlevel)
+        else:
+            memo = {} if args.memo else None
+            for f in args.pickle_file:
+                preamble = args.preamble.format(name=f.name)
+                args.output.write(preamble + '\n')
+                dis(f, args.output, memo, args.indentlevel)

Modified: python/branches/py3k-dtoa/Lib/posixpath.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/posixpath.py	(original)
+++ python/branches/py3k-dtoa/Lib/posixpath.py	Tue Jul  6 17:55:27 2010
@@ -49,6 +49,9 @@
 def normcase(s):
     """Normalize case of pathname.  Has no effect under Posix"""
     # TODO: on Mac OS X, this should really return s.lower().
+    if not isinstance(s, (bytes, str)):
+        raise TypeError("normcase() argument must be str or bytes, "
+                        "not '{}'".format(s.__class__.__name__))
     return s
 
 

Modified: python/branches/py3k-dtoa/Lib/pydoc.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/pydoc.py	(original)
+++ python/branches/py3k-dtoa/Lib/pydoc.py	Tue Jul  6 17:55:27 2010
@@ -1705,8 +1705,9 @@
             return ''
         return '<pydoc.Helper instance>'
 
-    def __call__(self, request=None):
-        if request is not None:
+    _GoInteractive = object()
+    def __call__(self, request=_GoInteractive):
+        if request is not self._GoInteractive:
             self.help(request)
         else:
             self.intro()

Modified: python/branches/py3k-dtoa/Lib/site.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/site.py	(original)
+++ python/branches/py3k-dtoa/Lib/site.py	Tue Jul  6 17:55:27 2010
@@ -273,12 +273,12 @@
     environment, and will return a list of full paths.
     """
     sitepackages = []
-    seen = []
+    seen = set()
 
     for prefix in PREFIXES:
         if not prefix or prefix in seen:
             continue
-        seen.append(prefix)
+        seen.add(prefix)
 
         if sys.platform in ('os2emx', 'riscos'):
             sitepackages.append(os.path.join(prefix, "Lib", "site-packages"))

Modified: python/branches/py3k-dtoa/Lib/smtpd.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/smtpd.py	(original)
+++ python/branches/py3k-dtoa/Lib/smtpd.py	Tue Jul  6 17:55:27 2010
@@ -274,14 +274,19 @@
         self._localaddr = localaddr
         self._remoteaddr = remoteaddr
         asyncore.dispatcher.__init__(self)
-        self.create_socket(socket.AF_INET, socket.SOCK_STREAM)
-        # try to re-use a server port if possible
-        self.set_reuse_addr()
-        self.bind(localaddr)
-        self.listen(5)
-        print('%s started at %s\n\tLocal addr: %s\n\tRemote addr:%s' % (
-            self.__class__.__name__, time.ctime(time.time()),
-            localaddr, remoteaddr), file=DEBUGSTREAM)
+        try:
+            self.create_socket(socket.AF_INET, socket.SOCK_STREAM)
+            # try to re-use a server port if possible
+            self.set_reuse_addr()
+            self.bind(localaddr)
+            self.listen(5)
+        except:
+            self.close()
+            raise
+        else:
+            print('%s started at %s\n\tLocal addr: %s\n\tRemote addr:%s' % (
+                self.__class__.__name__, time.ctime(time.time()),
+                localaddr, remoteaddr), file=DEBUGSTREAM)
 
     def handle_accept(self):
         conn, addr = self.accept()

Modified: python/branches/py3k-dtoa/Lib/subprocess.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/subprocess.py	(original)
+++ python/branches/py3k-dtoa/Lib/subprocess.py	Tue Jul  6 17:55:27 2010
@@ -500,8 +500,8 @@
 
     2) A string surrounded by double quotation marks is
        interpreted as a single argument, regardless of white space
-       or pipe characters contained within.  A quoted string can be
-       embedded in an argument.
+       contained within.  A quoted string can be embedded in an
+       argument.
 
     3) A double quotation mark preceded by a backslash is
        interpreted as a literal double quotation mark.
@@ -529,7 +529,7 @@
         if result:
             result.append(' ')
 
-        needquote = (" " in arg) or ("\t" in arg) or ("|" in arg) or not arg
+        needquote = (" " in arg) or ("\t" in arg) or not arg
         if needquote:
             result.append('"')
 

Modified: python/branches/py3k-dtoa/Lib/test/test_audioop.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_audioop.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_audioop.py	Tue Jul  6 17:55:27 2010
@@ -20,6 +20,12 @@
 
 data = [gendata1(), gendata2(), gendata4()]
 
+INVALID_DATA = [
+    ('abc', 0),
+    ('abc', 2),
+    ('abc', 4),
+]
+
 
 class TestAudioop(unittest.TestCase):
 
@@ -168,6 +174,33 @@
         self.assertRaises(audioop.error,
             audioop.findmax, ''.join(chr(x) for x in range(256)), -2392392)
 
+    def test_issue7673(self):
+        state = None
+        for data, size in INVALID_DATA:
+            size2 = size
+            self.assertRaises(audioop.error, audioop.getsample, data, size, 0)
+            self.assertRaises(audioop.error, audioop.max, data, size)
+            self.assertRaises(audioop.error, audioop.minmax, data, size)
+            self.assertRaises(audioop.error, audioop.avg, data, size)
+            self.assertRaises(audioop.error, audioop.rms, data, size)
+            self.assertRaises(audioop.error, audioop.avgpp, data, size)
+            self.assertRaises(audioop.error, audioop.maxpp, data, size)
+            self.assertRaises(audioop.error, audioop.cross, data, size)
+            self.assertRaises(audioop.error, audioop.mul, data, size, 1.0)
+            self.assertRaises(audioop.error, audioop.tomono, data, size, 0.5, 0.5)
+            self.assertRaises(audioop.error, audioop.tostereo, data, size, 0.5, 0.5)
+            self.assertRaises(audioop.error, audioop.add, data, data, size)
+            self.assertRaises(audioop.error, audioop.bias, data, size, 0)
+            self.assertRaises(audioop.error, audioop.reverse, data, size)
+            self.assertRaises(audioop.error, audioop.lin2lin, data, size, size2)
+            self.assertRaises(audioop.error, audioop.ratecv, data, size, 1, 1, 1, state)
+            self.assertRaises(audioop.error, audioop.lin2ulaw, data, size)
+            self.assertRaises(audioop.error, audioop.ulaw2lin, data, size)
+            self.assertRaises(audioop.error, audioop.lin2alaw, data, size)
+            self.assertRaises(audioop.error, audioop.alaw2lin, data, size)
+            self.assertRaises(audioop.error, audioop.lin2adpcm, data, size, state)
+            self.assertRaises(audioop.error, audioop.adpcm2lin, data, size, state)
+
 def test_main():
     run_unittest(TestAudioop)
 

Modified: python/branches/py3k-dtoa/Lib/test/test_codeccallbacks.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_codeccallbacks.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_codeccallbacks.py	Tue Jul  6 17:55:27 2010
@@ -153,28 +153,30 @@
             sout += bytes("\\U%08x" % sys.maxunicode, "ascii")
         self.assertEqual(sin.encode("iso-8859-15", "backslashreplace"), sout)
 
-    def test_decoderelaxedutf8(self):
-        # This is the test for a decoding callback handler,
-        # that relaxes the UTF-8 minimal encoding restriction.
-        # A null byte that is encoded as "\xc0\x80" will be
-        # decoded as a null byte. All other illegal sequences
-        # will be handled strictly.
+    def test_decoding_callbacks(self):
+        # This is a test for a decoding callback handler
+        # that allows the decoding of the invalid sequence
+        # "\xc0\x80" and returns "\x00" instead of raising an error.
+        # All other illegal sequences will be handled strictly.
         def relaxedutf8(exc):
             if not isinstance(exc, UnicodeDecodeError):
                 raise TypeError("don't know how to handle %r" % exc)
-            if exc.object[exc.start:exc.end].startswith(b"\xc0\x80"):
+            if exc.object[exc.start:exc.start+2] == b"\xc0\x80":
                 return ("\x00", exc.start+2) # retry after two bytes
             else:
                 raise exc
 
-        codecs.register_error(
-            "test.relaxedutf8", relaxedutf8)
+        codecs.register_error("test.relaxedutf8", relaxedutf8)
 
+        # all the "\xc0\x80" will be decoded to "\x00"
         sin = b"a\x00b\xc0\x80c\xc3\xbc\xc0\x80\xc0\x80"
         sout = "a\x00b\x00c\xfc\x00\x00"
         self.assertEqual(sin.decode("utf-8", "test.relaxedutf8"), sout)
+
+        # "\xc0\x81" is not valid and a UnicodeDecodeError will be raised
         sin = b"\xc0\x80\xc0\x81"
-        self.assertRaises(UnicodeError, sin.decode, "utf-8", "test.relaxedutf8")
+        self.assertRaises(UnicodeDecodeError, sin.decode,
+                          "utf-8", "test.relaxedutf8")
 
     def test_charmapencode(self):
         # For charmap encodings the replacement string will be

Modified: python/branches/py3k-dtoa/Lib/test/test_codecs.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_codecs.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_codecs.py	Tue Jul  6 17:55:27 2010
@@ -1263,6 +1263,7 @@
     "cp424",
     "cp437",
     "cp500",
+    "cp720",
     "cp737",
     "cp775",
     "cp850",
@@ -1270,6 +1271,7 @@
     "cp855",
     "cp856",
     "cp857",
+    "cp858",
     "cp860",
     "cp861",
     "cp862",
@@ -1358,11 +1360,6 @@
     "idna",
 ]
 
-# The following encodings only support "strict" mode
-only_strict_mode = [
-    "idna",
-]
-
 class BasicUnicodeTest(unittest.TestCase, MixInCheckStateHandling):
     def test_basics(self):
         s = "abc123" # all codecs should be able to encode these
@@ -1437,7 +1434,7 @@
                     result = "".join(codecs.iterdecode(codecs.iterencode("", encoding), encoding))
                     self.assertEqual(result, "")
 
-                if encoding not in only_strict_mode:
+                if encoding not in ("idna", "mbcs"):
                     # check incremental decoder/encoder with errors argument
                     try:
                         encoder = codecs.getincrementalencoder(encoding)("ignore")

Modified: python/branches/py3k-dtoa/Lib/test/test_complex.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_complex.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_complex.py	Tue Jul  6 17:55:27 2010
@@ -437,6 +437,23 @@
 
     @unittest.skipUnless(float.__getformat__("double").startswith("IEEE"),
                          "test requires IEEE 754 doubles")
+    def test_negated_imaginary_literal(self):
+        z0 = -0j
+        z1 = -7j
+        z2 = -1e1000j
+        # Note: In versions of Python < 3.2, a negated imaginary literal
+        # accidentally ended up with real part 0.0 instead of -0.0, thanks to a
+        # modification during CST -> AST translation (see issue #9011).  That's
+        # fixed in Python 3.2.
+        self.assertFloatsAreIdentical(z0.real, -0.0)
+        self.assertFloatsAreIdentical(z0.imag, -0.0)
+        self.assertFloatsAreIdentical(z1.real, -0.0)
+        self.assertFloatsAreIdentical(z1.imag, -7.0)
+        self.assertFloatsAreIdentical(z2.real, -0.0)
+        self.assertFloatsAreIdentical(z2.imag, -INF)
+
+    @unittest.skipUnless(float.__getformat__("double").startswith("IEEE"),
+                         "test requires IEEE 754 doubles")
     def test_overflow(self):
         self.assertEqual(complex("1e500"), complex(INF, 0.0))
         self.assertEqual(complex("-1e500j"), complex(0.0, -INF))

Modified: python/branches/py3k-dtoa/Lib/test/test_contextlib.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_contextlib.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_contextlib.py	Tue Jul  6 17:55:27 2010
@@ -202,6 +202,169 @@
                 return True
         self.boilerPlate(lock, locked)
 
+
+class mycontext(ContextDecorator):
+    started = False
+    exc = None
+    catch = False
+
+    def __enter__(self):
+        self.started = True
+        return self
+
+    def __exit__(self, *exc):
+        self.exc = exc
+        return self.catch
+
+
+class TestContextDecorator(unittest.TestCase):
+
+    def test_contextdecorator(self):
+        context = mycontext()
+        with context as result:
+            self.assertIs(result, context)
+            self.assertTrue(context.started)
+
+        self.assertEqual(context.exc, (None, None, None))
+
+
+    def test_contextdecorator_with_exception(self):
+        context = mycontext()
+
+        with self.assertRaisesRegexp(NameError, 'foo'):
+            with context:
+                raise NameError('foo')
+        self.assertIsNotNone(context.exc)
+        self.assertIs(context.exc[0], NameError)
+
+        context = mycontext()
+        context.catch = True
+        with context:
+            raise NameError('foo')
+        self.assertIsNotNone(context.exc)
+        self.assertIs(context.exc[0], NameError)
+
+
+    def test_decorator(self):
+        context = mycontext()
+
+        @context
+        def test():
+            self.assertIsNone(context.exc)
+            self.assertTrue(context.started)
+        test()
+        self.assertEqual(context.exc, (None, None, None))
+
+
+    def test_decorator_with_exception(self):
+        context = mycontext()
+
+        @context
+        def test():
+            self.assertIsNone(context.exc)
+            self.assertTrue(context.started)
+            raise NameError('foo')
+
+        with self.assertRaisesRegexp(NameError, 'foo'):
+            test()
+        self.assertIsNotNone(context.exc)
+        self.assertIs(context.exc[0], NameError)
+
+
+    def test_decorating_method(self):
+        context = mycontext()
+
+        class Test(object):
+
+            @context
+            def method(self, a, b, c=None):
+                self.a = a
+                self.b = b
+                self.c = c
+
+        # these tests are for argument passing when used as a decorator
+        test = Test()
+        test.method(1, 2)
+        self.assertEqual(test.a, 1)
+        self.assertEqual(test.b, 2)
+        self.assertEqual(test.c, None)
+
+        test = Test()
+        test.method('a', 'b', 'c')
+        self.assertEqual(test.a, 'a')
+        self.assertEqual(test.b, 'b')
+        self.assertEqual(test.c, 'c')
+
+        test = Test()
+        test.method(a=1, b=2)
+        self.assertEqual(test.a, 1)
+        self.assertEqual(test.b, 2)
+
+
+    def test_typo_enter(self):
+        class mycontext(ContextDecorator):
+            def __unter__(self):
+                pass
+            def __exit__(self, *exc):
+                pass
+
+        with self.assertRaises(AttributeError):
+            with mycontext():
+                pass
+
+
+    def test_typo_exit(self):
+        class mycontext(ContextDecorator):
+            def __enter__(self):
+                pass
+            def __uxit__(self, *exc):
+                pass
+
+        with self.assertRaises(AttributeError):
+            with mycontext():
+                pass
+
+
+    def test_contextdecorator_as_mixin(self):
+        class somecontext(object):
+            started = False
+            exc = None
+
+            def __enter__(self):
+                self.started = True
+                return self
+
+            def __exit__(self, *exc):
+                self.exc = exc
+
+        class mycontext(somecontext, ContextDecorator):
+            pass
+
+        context = mycontext()
+        @context
+        def test():
+            self.assertIsNone(context.exc)
+            self.assertTrue(context.started)
+        test()
+        self.assertEqual(context.exc, (None, None, None))
+
+
+    def test_contextmanager_as_decorator(self):
+        state = []
+        @contextmanager
+        def woohoo(y):
+            state.append(y)
+            yield
+            state.append(999)
+
+        @woohoo(1)
+        def test(x):
+            self.assertEqual(state, [1])
+            state.append(x)
+        test('something')
+        self.assertEqual(state, [1, 'something', 999])
+
+
 # This is needed to make the test actually run under regrtest.py!
 def test_main():
     support.run_unittest(__name__)

Modified: python/branches/py3k-dtoa/Lib/test/test_datetime.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_datetime.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_datetime.py	Tue Jul  6 17:55:27 2010
@@ -17,9 +17,11 @@
 from datetime import time
 from datetime import timezone
 from datetime import date, datetime
+import time as _time
 
-pickle_choices = [(pickle, pickle, proto) for proto in range(3)]
-assert len(pickle_choices) == 3
+pickle_choices = [(pickle, pickle, proto)
+                  for proto in range(pickle.HIGHEST_PROTOCOL + 1)]
+assert len(pickle_choices) == pickle.HIGHEST_PROTOCOL + 1
 
 # An arbitrary collection of objects of non-datetime types, for testing
 # mixed-type comparisons.
@@ -121,18 +123,23 @@
     def test_pickling_subclass(self):
         # Make sure we can pickle/unpickle an instance of a subclass.
         offset = timedelta(minutes=-300)
-        orig = PicklableFixedOffset(offset, 'cookie')
-        self.assertIsInstance(orig, tzinfo)
-        self.assertTrue(type(orig) is PicklableFixedOffset)
-        self.assertEqual(orig.utcoffset(None), offset)
-        self.assertEqual(orig.tzname(None), 'cookie')
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertIsInstance(derived, tzinfo)
-            self.assertTrue(type(derived) is PicklableFixedOffset)
-            self.assertEqual(derived.utcoffset(None), offset)
-            self.assertEqual(derived.tzname(None), 'cookie')
+        for otype, args in [
+            (PicklableFixedOffset, (offset, 'cookie')),
+            (timezone, (offset,)),
+            (timezone, (offset, "EST"))]:
+            orig = otype(*args)
+            oname = orig.tzname(None)
+            self.assertIsInstance(orig, tzinfo)
+            self.assertIs(type(orig), otype)
+            self.assertEqual(orig.utcoffset(None), offset)
+            self.assertEqual(orig.tzname(None), oname)
+            for pickler, unpickler, proto in pickle_choices:
+                green = pickler.dumps(orig, proto)
+                derived = unpickler.loads(green)
+                self.assertIsInstance(derived, tzinfo)
+                self.assertIs(type(derived), otype)
+                self.assertEqual(derived.utcoffset(None), offset)
+                self.assertEqual(derived.tzname(None), oname)
 
 class TestTimeZone(unittest.TestCase):
 
@@ -165,6 +172,7 @@
         with self.assertRaises(TypeError): timezone(42)
         with self.assertRaises(TypeError): timezone(ZERO, None)
         with self.assertRaises(TypeError): timezone(ZERO, 42)
+        with self.assertRaises(TypeError): timezone(ZERO, 'ABC', 'extra')
 
     def test_inheritance(self):
         self.assertIsInstance(timezone.utc, tzinfo)
@@ -200,6 +208,8 @@
     def test_fromutc(self):
         with self.assertRaises(ValueError):
             timezone.utc.fromutc(self.DT)
+        with self.assertRaises(TypeError):
+            timezone.utc.fromutc('not datetime')
         for tz in [self.EST, self.ACDT, Eastern]:
             utctime = self.DT.replace(tzinfo=tz)
             local = tz.fromutc(utctime)
@@ -300,6 +310,7 @@
         c = td(0, 0, 1000) # One millisecond
         eq(a+b+c, td(7, 60, 1000))
         eq(a-b, td(6, 24*3600 - 60))
+        eq(b.__rsub__(a), td(6, 24*3600 - 60))
         eq(-a, td(-7))
         eq(+a, td(7))
         eq(-b, td(-1, 24*3600 - 60))
@@ -377,6 +388,7 @@
         self.assertRaises(ZeroDivisionError, lambda: a // zero)
         self.assertRaises(ZeroDivisionError, lambda: a / zero)
         self.assertRaises(ZeroDivisionError, lambda: a / 0.0)
+        self.assertRaises(TypeError, lambda: a / '')
 
     @requires_IEEE_754
     def test_disallowed_special(self):
@@ -502,10 +514,20 @@
                    microseconds=999999)),
            "999999999 days, 23:59:59.999999")
 
+    def test_repr(self):
+        name = 'datetime.' + self.theclass.__name__
+        self.assertEqual(repr(self.theclass(1)),
+                         "%s(1)" % name)
+        self.assertEqual(repr(self.theclass(10, 2)),
+                         "%s(10, 2)" % name)
+        self.assertEqual(repr(self.theclass(-10, 2, 400000)),
+                         "%s(-10, 2, 400000)" % name)
+
     def test_roundtrip(self):
         for td in (timedelta(days=999999999, hours=23, minutes=59,
                              seconds=59, microseconds=999999),
                    timedelta(days=-999999999),
+                   timedelta(days=-999999999, seconds=1),
                    timedelta(days=1, seconds=2, microseconds=3)):
 
             # Verify td -> string -> td identity.
@@ -859,6 +881,7 @@
     def test_computations(self):
         a = self.theclass(2002, 1, 31)
         b = self.theclass(1956, 1, 31)
+        c = self.theclass(2001,2,1)
 
         diff = a-b
         self.assertEqual(diff.days, 46*365 + len(range(1956, 2002, 4)))
@@ -884,6 +907,7 @@
         self.assertEqual(a - (a + day), -day)
         self.assertEqual(a - (a - week), week)
         self.assertEqual(a - (a - day), day)
+        self.assertEqual(c - (c - day), day)
 
         # Add/sub ints or floats should be illegal
         for i in 1, 1.0:
@@ -1099,8 +1123,13 @@
             self.assertEqual(b.__format__(fmt), 'B')
 
     def test_resolution_info(self):
-        self.assertIsInstance(self.theclass.min, self.theclass)
-        self.assertIsInstance(self.theclass.max, self.theclass)
+        # XXX: Should min and max respect subclassing?
+        if issubclass(self.theclass, datetime):
+            expected_class = datetime
+        else:
+            expected_class = date
+        self.assertIsInstance(self.theclass.min, expected_class)
+        self.assertIsInstance(self.theclass.max, expected_class)
         self.assertIsInstance(self.theclass.resolution, timedelta)
         self.assertTrue(self.theclass.max > self.theclass.min)
 
@@ -1320,6 +1349,10 @@
         for month_byte in b'9', b'\0', b'\r', b'\xff':
             self.assertRaises(TypeError, self.theclass,
                                          base[:2] + month_byte + base[3:])
+        # Good bytes, but bad tzinfo:
+        self.assertRaises(TypeError, self.theclass,
+                          bytes([1] * len(base)), 'EST')
+
         for ord_byte in range(1, 13):
             # This shouldn't blow up because of the month byte alone.  If
             # the implementation changes to do more-careful checking, it may
@@ -1731,10 +1764,42 @@
 
         string = '2004-12-01 13:02:47.197'
         format = '%Y-%m-%d %H:%M:%S.%f'
-        result, frac = _strptime._strptime(string, format)
-        expected = self.theclass(*(result[0:6]+(frac,)))
+        expected = _strptime._strptime_datetime(self.theclass, string, format)
         got = self.theclass.strptime(string, format)
         self.assertEqual(expected, got)
+        self.assertIs(type(expected), self.theclass)
+        self.assertIs(type(got), self.theclass)
+
+        strptime = self.theclass.strptime
+        self.assertEqual(strptime("+0002", "%z").utcoffset(), 2 * MINUTE)
+        self.assertEqual(strptime("-0002", "%z").utcoffset(), -2 * MINUTE)
+        # Only local timezone and UTC are supported
+        for tzseconds, tzname in ((0, 'UTC'), (0, 'GMT'),
+                                 (-_time.timezone, _time.tzname[0])):
+            if tzseconds < 0:
+                sign = '-'
+                seconds = -tzseconds
+            else:
+                sign ='+'
+                seconds = tzseconds
+            hours, minutes = divmod(seconds//60, 60)
+            dtstr = "{}{:02d}{:02d} {}".format(sign, hours, minutes, tzname)
+            dt = strptime(dtstr, "%z %Z")
+            self.assertEqual(dt.utcoffset(), timedelta(seconds=tzseconds))
+            self.assertEqual(dt.tzname(), tzname)
+        # Can produce inconsistent datetime
+        dtstr, fmt = "+1234 UTC", "%z %Z"
+        dt = strptime(dtstr, fmt)
+        self.assertEqual(dt.utcoffset(), 12 * HOUR + 34 * MINUTE)
+        self.assertEqual(dt.tzname(), 'UTC')
+        # yet will roundtrip
+        self.assertEqual(dt.strftime(fmt), dtstr)
+
+        # Produce naive datetime if no %z is provided
+        self.assertEqual(strptime("UTC", "%Z").tzinfo, None)
+
+        with self.assertRaises(ValueError): strptime("-2400", "%z")
+        with self.assertRaises(ValueError): strptime("-000", "%z")
 
     def test_more_timetuple(self):
         # This tests fields beyond those tested by the TestDate.test_timetuple.
@@ -1789,6 +1854,8 @@
         self.assertRaises(TypeError, combine, t, d) # args reversed
         self.assertRaises(TypeError, combine, d, t, 1) # too many args
         self.assertRaises(TypeError, combine, "date", "time") # wrong types
+        self.assertRaises(TypeError, combine, d, "time") # wrong type
+        self.assertRaises(TypeError, combine, "date", t) # wrong type
 
     def test_replace(self):
         cls = self.theclass
@@ -1831,6 +1898,8 @@
             def dst(self, dt): return timedelta(0)
         bog = Bogus()
         self.assertRaises(ValueError, dt.astimezone, bog)   # naive
+        self.assertRaises(ValueError,
+                          dt.replace(tzinfo=bog).astimezone, f)
 
         class AlsoBogus(tzinfo):
             def utcoffset(self, dt): return timedelta(0)
@@ -1865,6 +1934,12 @@
         self.assertEqual(dt2.newmeth(-7), dt1.year + dt1.month +
                                           dt1.second - 7)
 
+class TestSubclassDateTime(TestDateTime):
+    theclass = SubclassDatetime
+    # Override tests not designed for subclass
+    def test_roundtrip(self):
+        pass
+
 class SubclassTime(time):
     sub_var = 1
 
@@ -2801,6 +2876,11 @@
         maxdiff = max - min
         self.assertEqual(maxdiff, self.theclass.max - self.theclass.min +
                                   timedelta(minutes=2*1439))
+        # Different tzinfo, but the same offset
+        tza = timezone(HOUR, 'A')
+        tzb = timezone(HOUR, 'B')
+        delta = min.replace(tzinfo=tza) - max.replace(tzinfo=tzb)
+        self.assertEqual(delta, self.theclass.min - self.theclass.max)
 
     def test_tzinfo_now(self):
         meth = self.theclass.now
@@ -2934,7 +3014,7 @@
 
     def test_utctimetuple(self):
         class DST(tzinfo):
-            def __init__(self, dstvalue):
+            def __init__(self, dstvalue=0):
                 if isinstance(dstvalue, int):
                     dstvalue = timedelta(minutes=dstvalue)
                 self.dstvalue = dstvalue
@@ -2953,8 +3033,6 @@
             def utcoffset(self, dt):
                 return self.uofs
 
-        # Ensure tm_isdst is 0 regardless of what dst() says:  DST is never
-        # in effect for a UTC time.
         for dstvalue in -33, 33, 0, None:
             d = cls(1, 2, 3, 10, 20, 30, 40, tzinfo=UOFS(-53, dstvalue))
             t = d.utctimetuple()
@@ -2967,34 +3045,52 @@
             self.assertEqual(d.weekday(), t.tm_wday)
             self.assertEqual(d.toordinal() - date(1, 1, 1).toordinal() + 1,
                              t.tm_yday)
+            # Ensure tm_isdst is 0 regardless of what dst() says: DST
+            # is never in effect for a UTC time.
             self.assertEqual(0, t.tm_isdst)
 
-        # At the edges, UTC adjustment can normalize into years out-of-range
-        # for a datetime object.  Ensure that a correct timetuple is
-        # created anyway.
+        # For naive datetime, utctimetuple == timetuple except for isdst
+        d = cls(1, 2, 3, 10, 20, 30, 40)
+        t = d.utctimetuple()
+        self.assertEqual(t[:-1], d.timetuple()[:-1])
+        self.assertEqual(0, t.tm_isdst)
+        # Same if utcoffset is None
+        class NOFS(DST):
+            def utcoffset(self, dt):
+                return None
+        d = cls(1, 2, 3, 10, 20, 30, 40, tzinfo=NOFS())
+        t = d.utctimetuple()
+        self.assertEqual(t[:-1], d.timetuple()[:-1])
+        self.assertEqual(0, t.tm_isdst)
+        # Check that bad tzinfo is detected
+        class BOFS(DST):
+            def utcoffset(self, dt):
+                return "EST"
+        d = cls(1, 2, 3, 10, 20, 30, 40, tzinfo=BOFS())
+        self.assertRaises(TypeError, d.utctimetuple)
+
+        # Check that utctimetuple() is the same as
+        # astimezone(utc).timetuple()
+        d = cls(2010, 11, 13, 14, 15, 16, 171819)
+        for tz in [timezone.min, timezone.utc, timezone.max]:
+            dtz = d.replace(tzinfo=tz)
+            self.assertEqual(dtz.utctimetuple()[:-1],
+                             dtz.astimezone(timezone.utc).timetuple()[:-1])
+        # At the edges, UTC adjustment can produce years out-of-range
+        # for a datetime object.  Ensure that an OverflowError is
+        # raised.
         tiny = cls(MINYEAR, 1, 1, 0, 0, 37, tzinfo=UOFS(1439))
         # That goes back 1 minute less than a full day.
-        t = tiny.utctimetuple()
-        self.assertEqual(t.tm_year, MINYEAR-1)
-        self.assertEqual(t.tm_mon, 12)
-        self.assertEqual(t.tm_mday, 31)
-        self.assertEqual(t.tm_hour, 0)
-        self.assertEqual(t.tm_min, 1)
-        self.assertEqual(t.tm_sec, 37)
-        self.assertEqual(t.tm_yday, 366)    # "year 0" is a leap year
-        self.assertEqual(t.tm_isdst, 0)
+        self.assertRaises(OverflowError, tiny.utctimetuple)
 
         huge = cls(MAXYEAR, 12, 31, 23, 59, 37, 999999, tzinfo=UOFS(-1439))
         # That goes forward 1 minute less than a full day.
-        t = huge.utctimetuple()
-        self.assertEqual(t.tm_year, MAXYEAR+1)
-        self.assertEqual(t.tm_mon, 1)
-        self.assertEqual(t.tm_mday, 1)
-        self.assertEqual(t.tm_hour, 23)
-        self.assertEqual(t.tm_min, 58)
-        self.assertEqual(t.tm_sec, 37)
-        self.assertEqual(t.tm_yday, 1)
-        self.assertEqual(t.tm_isdst, 0)
+        self.assertRaises(OverflowError, huge.utctimetuple)
+        # More overflow cases
+        tiny = cls.min.replace(tzinfo=timezone(MINUTE))
+        self.assertRaises(OverflowError, tiny.utctimetuple)
+        huge = cls.max.replace(tzinfo=timezone(-MINUTE))
+        self.assertRaises(OverflowError, huge.utctimetuple)
 
     def test_tzinfo_isoformat(self):
         zero = FixedOffset(0, "+00:00")
@@ -3196,6 +3292,7 @@
     return dt
 
 ZERO = timedelta(0)
+MINUTE = timedelta(minutes=1)
 HOUR = timedelta(hours=1)
 DAY = timedelta(days=1)
 # In the US, DST starts at 2am (standard time) on the first Sunday in April.
@@ -3439,6 +3536,19 @@
             def dst(self, dt): return None
         self.assertRaises(ValueError, now.astimezone, notok())
 
+        # Sometimes blow up. In the following, tzinfo.dst()
+        # implementation may return None or not None depending on
+        # whether DST is assumed to be in effect.  In this situation,
+        # a ValueError should be raised by astimezone().
+        class tricky_notok(ok):
+            def dst(self, dt):
+                if dt.year == 2000:
+                    return None
+                else:
+                    return 10*HOUR
+        dt = self.theclass(2001, 1, 1).replace(tzinfo=utc_real)
+        self.assertRaises(ValueError, dt.astimezone, tricky_notok())
+
     def test_fromutc(self):
         self.assertRaises(TypeError, Eastern.fromutc)   # not enough args
         now = datetime.utcnow().replace(tzinfo=utc_real)

Modified: python/branches/py3k-dtoa/Lib/test/test_descr.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_descr.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_descr.py	Tue Jul  6 17:55:27 2010
@@ -1,6 +1,7 @@
 import builtins
 import sys
 import types
+import math
 import unittest
 
 from copy import deepcopy
@@ -1578,6 +1579,9 @@
             ("__exit__", run_context, swallow, set(), {"__enter__" : iden}),
             ("__complex__", complex, complex_num, set(), {}),
             ("__format__", format, format_impl, set(), {}),
+            ("__floor__", math.floor, zero, set(), {}),
+            ("__trunc__", math.trunc, zero, set(), {}),
+            ("__ceil__", math.ceil, zero, set(), {}),
             ]
 
         class Checker(object):

Modified: python/branches/py3k-dtoa/Lib/test/test_dict.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_dict.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_dict.py	Tue Jul  6 17:55:27 2010
@@ -664,6 +664,7 @@
         gc.collect()
         self.assertTrue(gc.is_tracked(t), t)
 
+    @support.cpython_only
     def test_track_literals(self):
         # Test GC-optimization of dict literals
         x, y, z, w = 1.5, "a", (1, None), []
@@ -681,6 +682,7 @@
         self._tracked({1: {}})
         self._tracked({1: set()})
 
+    @support.cpython_only
     def test_track_dynamic(self):
         # Test GC-optimization of dynamically-created dicts
         class MyObject(object):
@@ -744,6 +746,7 @@
         d.update([(x, y), (z, w)])
         self._tracked(d)
 
+    @support.cpython_only
     def test_track_subtypes(self):
         # Dict subtypes are always tracked
         class MyDict(dict):

Modified: python/branches/py3k-dtoa/Lib/test/test_dis.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_dis.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_dis.py	Tue Jul  6 17:55:27 2010
@@ -96,6 +96,46 @@
 
 """
 
+expr_str = "x + 1"
+
+dis_expr_str = """\
+  1           0 LOAD_NAME                0 (x)
+              3 LOAD_CONST               0 (1)
+              6 BINARY_ADD
+              7 RETURN_VALUE
+"""
+
+simple_stmt_str = "x = x + 1"
+
+dis_simple_stmt_str = """\
+  1           0 LOAD_NAME                0 (x)
+              3 LOAD_CONST               0 (1)
+              6 BINARY_ADD
+              7 STORE_NAME               0 (x)
+             10 LOAD_CONST               1 (None)
+             13 RETURN_VALUE
+"""
+
+compound_stmt_str = """\
+x = 0
+while 1:
+    x += 1"""
+# Trailing newline has been deliberately omitted
+
+dis_compound_stmt_str = """\
+  1           0 LOAD_CONST               0 (0)
+              3 STORE_NAME               0 (x)
+
+  2           6 SETUP_LOOP              13 (to 22)
+
+  3     >>    9 LOAD_NAME                0 (x)
+             12 LOAD_CONST               1 (1)
+             15 INPLACE_ADD
+             16 STORE_NAME               0 (x)
+             19 JUMP_ABSOLUTE            9
+        >>   22 LOAD_CONST               2 (None)
+             25 RETURN_VALUE
+"""
 
 class DisTests(unittest.TestCase):
     def do_disassembly_test(self, func, expected):
@@ -166,6 +206,11 @@
         from test import dis_module
         self.do_disassembly_test(dis_module, dis_module_expected_results)
 
+    def test_disassemble_str(self):
+        self.do_disassembly_test(expr_str, dis_expr_str)
+        self.do_disassembly_test(simple_stmt_str, dis_simple_stmt_str)
+        self.do_disassembly_test(compound_stmt_str, dis_compound_stmt_str)
+
 def test_main():
     run_unittest(DisTests)
 

Modified: python/branches/py3k-dtoa/Lib/test/test_enumerate.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_enumerate.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_enumerate.py	Tue Jul  6 17:55:27 2010
@@ -103,6 +103,7 @@
         self.assertRaises(TypeError, self.enum, 'abc', 'a') # wrong type
         self.assertRaises(TypeError, self.enum, 'abc', 2, 3) # too many arguments
 
+    @support.cpython_only
     def test_tuple_reuse(self):
         # Tests an implementation detail where tuple is reused
         # whenever nothing else holds a reference to it
@@ -145,6 +146,7 @@
         x = range(1)
         self.assertEqual(type(reversed(x)), type(iter(x)))
 
+    @support.cpython_only
     def test_len(self):
         # This is an implementation detail, not an interface requirement
         from test.test_iterlen import len

Modified: python/branches/py3k-dtoa/Lib/test/test_exceptions.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_exceptions.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_exceptions.py	Tue Jul  6 17:55:27 2010
@@ -7,7 +7,7 @@
 import weakref
 
 from test.support import (TESTFN, unlink, run_unittest, captured_output,
-                          gc_collect)
+                          gc_collect, cpython_only)
 
 # XXX This is not really enough, each *operation* should be tested!
 
@@ -137,6 +137,7 @@
         ckmsg(s, "'continue' not properly in loop")
         ckmsg("continue\n", "'continue' not properly in loop")
 
+    @cpython_only
     def testSettingException(self):
         # test that setting an exception at the C level works even if the
         # exception object can't be constructed.
@@ -669,6 +670,7 @@
         tb2 = raiseMemError()
         self.assertEqual(tb1, tb2)
 
+    @cpython_only
     def test_exception_with_doc(self):
         import _testcapi
         doc2 = "This is a test docstring."

Modified: python/branches/py3k-dtoa/Lib/test/test_extcall.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_extcall.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_extcall.py	Tue Jul  6 17:55:27 2010
@@ -279,13 +279,13 @@
     >>> f(6, a=4, *(1, 2, 3))
     Traceback (most recent call last):
       ...
-    TypeError: f() takes exactly 1 argument (5 given)
+    TypeError: f() takes exactly 1 positional argument (5 given)
     >>> def f(a, *, kw):
     ...    pass
     >>> f(6, 4, kw=4)
     Traceback (most recent call last):
       ...
-    TypeError: f() takes exactly 2 arguments (3 given)
+    TypeError: f() takes exactly 1 positional argument (3 given)
 """
 
 import sys

Modified: python/branches/py3k-dtoa/Lib/test/test_float.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_float.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_float.py	Tue Jul  6 17:55:27 2010
@@ -7,13 +7,17 @@
 from math import isinf, isnan, copysign, ldexp
 import operator
 import random, fractions
-import re
 
 INF = float("inf")
 NAN = float("nan")
 
+have_getformat = hasattr(float, "__getformat__")
+requires_getformat = unittest.skipUnless(have_getformat,
+                                         "requires __getformat__")
+requires_setformat = unittest.skipUnless(hasattr(float, "__setformat__"),
+                                         "requires __setformat__")
 # decorator for skipping tests on non-IEEE 754 platforms
-requires_IEEE_754 = unittest.skipUnless(
+requires_IEEE_754 = unittest.skipUnless(have_getformat and
     float.__getformat__("double").startswith("IEEE"),
     "test requires IEEE 754 doubles")
 
@@ -21,74 +25,6 @@
 test_dir = os.path.dirname(__file__) or os.curdir
 format_testfile = os.path.join(test_dir, 'formatfloat_testcases.txt')
 
-finite_decimal_parser = re.compile(r"""    # A numeric string consists of:
-    (?P<sign>[-+])?          # an optional sign, followed by
-    (?=\d|\.\d)              # a number with at least one digit
-    (?P<int>\d*)             # having a (possibly empty) integer part
-    (?:\.(?P<frac>\d*))?     # followed by an optional fractional part
-    (?:E(?P<exp>[-+]?\d+))?  # and an optional exponent
-    \Z
-""", re.VERBOSE | re.IGNORECASE | re.UNICODE).match
-
-# Pure Python version of correctly rounded string->float conversion.
-# Avoids any use of floating-point by returning the result as a hex string.
-def strtod(s, mant_dig=53, min_exp = -1021, max_exp = 1024):
-    """Convert a finite decimal string to a hex string representing an
-    IEEE 754 binary64 float.  Return 'inf' or '-inf' on overflow.
-    This function makes no use of floating-point arithmetic at any
-    stage."""
-
-    # parse string into a pair of integers 'a' and 'b' such that
-    # abs(decimal value) = a/b, and a boolean 'negative'.
-    m = finite_decimal_parser(s)
-    if m is None:
-        raise ValueError('invalid numeric string')
-    fraction = m.group('frac') or ''
-    intpart = int(m.group('int') + fraction)
-    exp = int(m.group('exp') or '0') - len(fraction)
-    negative = m.group('sign') == '-'
-    a, b = intpart*10**max(exp, 0), 10**max(0, -exp)
-
-    # quick return for zeros
-    if not a:
-        return '-0x0.0p+0' if negative else '0x0.0p+0'
-
-    # compute exponent e for result; may be one too small in the case
-    # that the rounded value of a/b lies in a different binade from a/b
-    d = a.bit_length() - b.bit_length()
-    d += (a >> d if d >= 0 else a << -d) >= b
-    e = max(d, min_exp) - mant_dig
-
-    # approximate a/b by number of the form q * 2**e; adjust e if necessary
-    a, b = a << max(-e, 0), b << max(e, 0)
-    q, r = divmod(a, b)
-    if 2*r > b or 2*r == b and q & 1:
-        q += 1
-        if q.bit_length() == mant_dig+1:
-            q //= 2
-            e += 1
-
-    # double check that (q, e) has the right form
-    assert q.bit_length() <= mant_dig and e >= min_exp - mant_dig
-    assert q.bit_length() == mant_dig or e == min_exp - mant_dig
-
-    # check for overflow and underflow
-    if e + q.bit_length() > max_exp:
-        return '-inf' if negative else 'inf'
-    if not q:
-        return '-0x0.0p+0' if negative else '0x0.0p+0'
-
-    # for hex representation, shift so # bits after point is a multiple of 4
-    hexdigs = 1 + (mant_dig-2)//4
-    shift = 3 - (mant_dig-2)%4
-    q, e = q << shift, e - shift
-    return '{}0x{:x}.{:0{}x}p{:+d}'.format(
-        '-' if negative else '',
-        q // 16**hexdigs,
-        q % 16**hexdigs,
-        hexdigs,
-        e + 4*hexdigs)
-
 class GeneralFloatCases(unittest.TestCase):
 
     def test_float(self):
@@ -446,6 +382,7 @@
             #self.assertTrue(0.0 > pow_op(-2.0, -1047) > -1e-315)
 
 
+ at requires_setformat
 class FormatFunctionsTestCase(unittest.TestCase):
 
     def setUp(self):
@@ -496,6 +433,7 @@
 # on non-IEEE platforms, attempting to unpack a bit pattern
 # representing an infinity or a NaN should raise an exception.
 
+ at requires_setformat
 class UnknownFormatTestCase(unittest.TestCase):
     def setUp(self):
         self.save_formats = {'double':float.__getformat__('double'),
@@ -528,36 +466,25 @@
 # let's also try to guarantee that -0.0 and 0.0 don't get confused.
 
 class IEEEFormatTestCase(unittest.TestCase):
-    if float.__getformat__("double").startswith("IEEE"):
-        def test_double_specials_do_unpack(self):
-            for fmt, data in [('>d', BE_DOUBLE_INF),
-                              ('>d', BE_DOUBLE_NAN),
-                              ('<d', LE_DOUBLE_INF),
-                              ('<d', LE_DOUBLE_NAN)]:
-                struct.unpack(fmt, data)
-
-    if float.__getformat__("float").startswith("IEEE"):
-        def test_float_specials_do_unpack(self):
-            for fmt, data in [('>f', BE_FLOAT_INF),
-                              ('>f', BE_FLOAT_NAN),
-                              ('<f', LE_FLOAT_INF),
-                              ('<f', LE_FLOAT_NAN)]:
-                struct.unpack(fmt, data)
-
-    if float.__getformat__("double").startswith("IEEE"):
-        def test_negative_zero(self):
-            def pos_pos():
-                return 0.0, math.atan2(0.0, -1)
-            def pos_neg():
-                return 0.0, math.atan2(-0.0, -1)
-            def neg_pos():
-                return -0.0, math.atan2(0.0, -1)
-            def neg_neg():
-                return -0.0, math.atan2(-0.0, -1)
-            self.assertEquals(pos_pos(), neg_pos())
-            self.assertEquals(pos_neg(), neg_neg())
+
+    @requires_IEEE_754
+    def test_double_specials_do_unpack(self):
+        for fmt, data in [('>d', BE_DOUBLE_INF),
+                          ('>d', BE_DOUBLE_NAN),
+                          ('<d', LE_DOUBLE_INF),
+                          ('<d', LE_DOUBLE_NAN)]:
+            struct.unpack(fmt, data)
+
+    @requires_IEEE_754
+    def test_float_specials_do_unpack(self):
+        for fmt, data in [('>f', BE_FLOAT_INF),
+                          ('>f', BE_FLOAT_NAN),
+                          ('<f', LE_FLOAT_INF),
+                          ('<f', LE_FLOAT_NAN)]:
+            struct.unpack(fmt, data)
 
 class FormatTestCase(unittest.TestCase):
+
     def test_format(self):
         # these should be rewritten to use both format(x, spec) and
         # x.__format__(spec)
@@ -611,8 +538,7 @@
         self.assertEqual(format(INF, 'f'), 'inf')
         self.assertEqual(format(INF, 'F'), 'INF')
 
-    @unittest.skipUnless(float.__getformat__("double").startswith("IEEE"),
-                         "test requires IEEE 754 doubles")
+    @requires_IEEE_754
     def test_format_testfile(self):
         for line in open(format_testfile):
             if line.startswith('--'):
@@ -692,9 +618,10 @@
             self.assertEqual(s, repr(float(s)))
             self.assertEqual(negs, repr(float(negs)))
 
+
+ at requires_IEEE_754
 class RoundTestCase(unittest.TestCase):
-    @unittest.skipUnless(float.__getformat__("double").startswith("IEEE"),
-                         "test requires IEEE 754 doubles")
+
     def test_inf_nan(self):
         self.assertRaises(OverflowError, round, INF)
         self.assertRaises(OverflowError, round, -INF)
@@ -704,8 +631,6 @@
         self.assertRaises(TypeError, round, NAN, "ceci n'est pas un integer")
         self.assertRaises(TypeError, round, -0.0, 1j)
 
-    @unittest.skipUnless(float.__getformat__("double").startswith("IEEE"),
-                         "test requires IEEE 754 doubles")
     def test_large_n(self):
         for n in [324, 325, 400, 2**31-1, 2**31, 2**32, 2**100]:
             self.assertEqual(round(123.456, n), 123.456)
@@ -718,8 +643,6 @@
         self.assertEqual(round(1e150, 309), 1e150)
         self.assertEqual(round(1.4e-315, 315), 1e-315)
 
-    @unittest.skipUnless(float.__getformat__("double").startswith("IEEE"),
-                         "test requires IEEE 754 doubles")
     def test_small_n(self):
         for n in [-308, -309, -400, 1-2**31, -2**31, -2**31-1, -2**100]:
             self.assertEqual(round(123.456, n), 0.0)
@@ -727,8 +650,6 @@
             self.assertEqual(round(1e300, n), 0.0)
             self.assertEqual(round(1e-320, n), 0.0)
 
-    @unittest.skipUnless(float.__getformat__("double").startswith("IEEE"),
-                         "test requires IEEE 754 doubles")
     def test_overflow(self):
         self.assertRaises(OverflowError, round, 1.6e308, -308)
         self.assertRaises(OverflowError, round, -1.7e308, -308)
@@ -776,9 +697,6 @@
             self.assertEqual(float(format(x, '.2f')), round(x, 2))
             self.assertEqual(float(format(x, '.3f')), round(x, 3))
 
-
-    @unittest.skipUnless(float.__getformat__("double").startswith("IEEE"),
-                         "test requires IEEE 754 doubles")
     def test_format_specials(self):
         # Test formatting of nans and infs.
 
@@ -1334,38 +1252,6 @@
             else:
                 self.identical(x, fromHex(toHex(x)))
 
-class StrtodTestCase(unittest.TestCase):
-    def check_string(self, s):
-        expected = strtod(s)
-        try:
-            fs = float(s)
-        except OverflowError:
-            got = '-inf' if s[0] == '-' else 'inf'
-        else:
-            got = fs.hex()
-        self.assertEqual(expected, got,
-                         "Incorrectly rounded str->float conversion for "
-                         "{}: expected {}, got {}".format(s, expected, got))
-
-    @unittest.skipUnless(getattr(sys, 'float_repr_style', '') == 'short',
-                         "applies only when using short float repr style")
-    def test_bug7632(self):
-        # check a few particular values that gave incorrectly rounded
-        # results with previous versions of dtoa.c
-        test_strings = [
-            '94393431193180696942841837085033647913224148539854e-358',
-            '12579816049008305546974391768996369464963024663104e-357',
-            '17489628565202117263145367596028389348922981857013e-357',
-            '18487398785991994634182916638542680759613590482273e-357',
-            '32002864200581033134358724675198044527469366773928e-358',
-            '73608278998966969345824653500136787876436005957953e-358',
-            '64774478836417299491718435234611299336288082136054e-358',
-            '13704940134126574534878641876947980878824688451169e-357',
-            '46697445774047060960624497964425416610480524760471e-358',
-            ]
-        for s in test_strings:
-            self.check_string(s)
-
 
 def test_main():
     support.run_unittest(
@@ -1378,7 +1264,6 @@
         RoundTestCase,
         InfNanTest,
         HexFloatTestCase,
-        StrtodTestCase,
         )
 
 if __name__ == '__main__':

Modified: python/branches/py3k-dtoa/Lib/test/test_genericpath.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_genericpath.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_genericpath.py	Tue Jul  6 17:55:27 2010
@@ -194,14 +194,18 @@
     ]
 
     def test_normcase(self):
-        # Check that normcase() is idempotent
-        p = "FoO/./BaR"
-        p = self.pathmodule.normcase(p)
-        self.assertEqual(p, self.pathmodule.normcase(p))
-
-        p = b"FoO/./BaR"
-        p = self.pathmodule.normcase(p)
-        self.assertEqual(p, self.pathmodule.normcase(p))
+        normcase = self.pathmodule.normcase
+        # check that normcase() is idempotent
+        for p in ["FoO/./BaR", b"FoO/./BaR"]:
+            p = normcase(p)
+            self.assertEqual(p, normcase(p))
+
+        self.assertEqual(normcase(''), '')
+        self.assertEqual(normcase(b''), b'')
+
+        # check that normcase raises a TypeError for invalid types
+        for path in (None, True, 0, 2.5, [], bytearray(b''), {'o','o'}):
+            self.assertRaises(TypeError, normcase, path)
 
     def test_splitdrive(self):
         # splitdrive for non-NT paths

Modified: python/branches/py3k-dtoa/Lib/test/test_getargs2.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_getargs2.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_getargs2.py	Tue Jul  6 17:55:27 2010
@@ -325,7 +325,7 @@
         from _testcapi import getargs_z
         self.assertEqual(getargs_z('abc\xe9'), b'abc\xc3\xa9')
         self.assertRaises(TypeError, getargs_z, 'nul:\0')
-        self.assertEqual(getargs_z(b'bytes'), b'bytes')
+        self.assertRaises(TypeError, getargs_z, b'bytes')
         self.assertRaises(TypeError, getargs_z, bytearray(b'bytearray'))
         self.assertRaises(TypeError, getargs_z, memoryview(b'memoryview'))
         self.assertIsNone(getargs_z(None))
@@ -375,6 +375,16 @@
         self.assertRaises(TypeError, getargs_y_hash, memoryview(b'memoryview'))
         self.assertRaises(TypeError, getargs_y_hash, None)
 
+    def test_w_star(self):
+        # getargs_w_star() modifies first and last byte
+        from _testcapi import getargs_w_star
+        self.assertRaises(TypeError, getargs_w_star, 'abc\xe9')
+        self.assertRaises(TypeError, getargs_w_star, b'bytes')
+        self.assertRaises(TypeError, getargs_w_star, b'nul:\0')
+        self.assertEqual(getargs_w_star(bytearray(b'bytearray')), b'[ytearra]')
+        self.assertEqual(getargs_w_star(memoryview(b'memoryview')), b'[emoryvie]')
+        self.assertRaises(TypeError, getargs_w_star, None)
+
 
 class Unicode_TestCase(unittest.TestCase):
     def test_u(self):

Modified: python/branches/py3k-dtoa/Lib/test/test_http_cookiejar.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_http_cookiejar.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_http_cookiejar.py	Tue Jul  6 17:55:27 2010
@@ -97,6 +97,7 @@
 
 
 class HeaderTests(TestCase):
+
     def test_parse_ns_headers(self):
         # quotes should be stripped
         expected = [[('foo', 'bar'), ('expires', 2209069412), ('version', '0')]]
@@ -106,6 +107,16 @@
             ]:
             self.assertEquals(parse_ns_headers([hdr]), expected)
 
+    def test_parse_ns_headers_version(self):
+
+        # quotes should be stripped
+        expected = [[('foo', 'bar'), ('version', '1')]]
+        for hdr in [
+            'foo=bar; version="1"',
+            'foo=bar; Version="1"',
+            ]:
+            self.assertEquals(parse_ns_headers([hdr]), expected)
+
     def test_parse_ns_headers_special_names(self):
         # names such as 'expires' are not special in first name=value pair
         # of Set-Cookie: header
@@ -1020,6 +1031,8 @@
             ["Set-Cookie2: a=foo; path=/; Version=1; domain"],
             # bad max-age
             ["Set-Cookie: b=foo; max-age=oops"],
+            # bad version
+            ["Set-Cookie: b=foo; version=spam"],
             ]:
             c = cookiejar_from_cookie_headers(headers)
             # these bad cookies shouldn't be set

Modified: python/branches/py3k-dtoa/Lib/test/test_imp.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_imp.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_imp.py	Tue Jul  6 17:55:27 2010
@@ -44,7 +44,7 @@
 class ImportTests(unittest.TestCase):
 
     def test_find_module_encoding(self):
-        fd = imp.find_module("heapq")[0]
+        fd = imp.find_module("pydoc")[0]
         self.assertEqual(fd.encoding, "iso-8859-1")
 
     def test_issue1267(self):

Modified: python/branches/py3k-dtoa/Lib/test/test_import.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_import.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_import.py	Tue Jul  6 17:55:27 2010
@@ -147,22 +147,24 @@
         filename = module + '.py'
 
         # Create a file with a list of 65000 elements.
-        with open(filename, 'w+') as f:
+        with open(filename, 'w') as f:
             f.write('d = [\n')
             for i in range(65000):
                 f.write('"",\n')
             f.write(']')
 
-        # Compile & remove .py file; we only need .pyc (or .pyo).
-        # Bytecode must be relocated from the PEP 3147 bytecode-only location.
-        py_compile.compile(filename)
-        unlink(filename)
-        make_legacy_pyc(filename)
+        try:
+            # Compile & remove .py file; we only need .pyc (or .pyo).
+            # Bytecode must be relocated from the PEP 3147 bytecode-only location.
+            py_compile.compile(filename)
+        finally:
+            unlink(filename)
 
         # Need to be able to load from current dir.
         sys.path.append('')
 
         try:
+            make_legacy_pyc(filename)
             # This used to crash.
             exec('import ' + module)
         finally:
@@ -468,6 +470,17 @@
         ns = dict(__package__=object())
         self.assertRaises(ValueError, check_relative)
 
+    def test_absolute_import_without_future(self):
+        # If absolute import syntax is used, then do not try to perform
+        # a relative import in the face of failure.
+        # Issue #7902.
+        try:
+            from .os import sep
+        except ImportError:
+            pass
+        else:
+            self.fail("explicit relative import triggered an "
+                      "implicit relative import")
 
 class OverridingImportBuiltinTests(unittest.TestCase):
     def test_override_builtin(self):

Modified: python/branches/py3k-dtoa/Lib/test/test_inspect.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_inspect.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_inspect.py	Tue Jul  6 17:55:27 2010
@@ -3,6 +3,7 @@
 import types
 import unittest
 import inspect
+import linecache
 import datetime
 import collections
 from os.path import normcase
@@ -275,6 +276,11 @@
     def test_getsourcefile(self):
         self.assertEqual(normcase(inspect.getsourcefile(mod.spam)), modfile)
         self.assertEqual(normcase(inspect.getsourcefile(git.abuse)), modfile)
+        fn = "_non_existing_filename_used_for_sourcefile_test.py"
+        co = compile("None", fn, "exec")
+        self.assertEqual(inspect.getsourcefile(co), None)
+        linecache.cache[co.co_filename] = (1, None, "None", co.co_filename)
+        self.assertEqual(normcase(inspect.getsourcefile(co)), fn)
 
     def test_getfile(self):
         self.assertEqual(inspect.getfile(mod.StupidGit), mod.__file__)
@@ -373,6 +379,15 @@
         self.assertRaises(IOError, inspect.getsource, unicodedata)
         self.assertRaises(IOError, inspect.findsource, unicodedata)
 
+    def test_findsource_code_in_linecache(self):
+        lines = ["x=1"]
+        co = compile(lines[0], "_dynamically_created_file", "exec")
+        self.assertRaises(IOError, inspect.findsource, co)
+        self.assertRaises(IOError, inspect.getsource, co)
+        linecache.cache[co.co_filename] = (1, None, lines, co.co_filename)
+        self.assertEquals(inspect.findsource(co), (lines,0))
+        self.assertEquals(inspect.getsource(co), lines[0])
+
 # Helper for testing classify_class_attrs.
 def attrs_wo_objs(cls):
     return [t[:3] for t in inspect.classify_class_attrs(cls)]

Modified: python/branches/py3k-dtoa/Lib/test/test_iter.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_iter.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_iter.py	Tue Jul  6 17:55:27 2010
@@ -1,7 +1,7 @@
 # Test iterators.
 
 import unittest
-from test.support import run_unittest, TESTFN, unlink
+from test.support import run_unittest, TESTFN, unlink, cpython_only
 
 # Test result of triple loop (too big to inline)
 TRIPLETS = [(0, 0, 0), (0, 0, 1), (0, 0, 2),
@@ -761,8 +761,9 @@
         (a, b), (c,) = IteratingSequenceClass(2), {42: 24}
         self.assertEqual((a, b, c), (0, 1, 42))
 
-        # Test reference count behavior
 
+    @cpython_only
+    def test_ref_counting_behavior(self):
         class C(object):
             count = 0
             def __new__(cls):

Modified: python/branches/py3k-dtoa/Lib/test/test_keywordonlyarg.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_keywordonlyarg.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_keywordonlyarg.py	Tue Jul  6 17:55:27 2010
@@ -73,6 +73,14 @@
         fundef3 += "lastarg):\n  pass\n"
         compile(fundef3, "<test>", "single")
 
+    def testTooManyPositionalErrorMessage(self):
+        def f(a, b=None, *, c=None):
+            pass
+        with self.assertRaises(TypeError) as exc:
+            f(1, 2, 3)
+        expected = "f() takes at most 2 positional arguments (3 given)"
+        self.assertEqual(str(exc.exception), expected)
+
     def testSyntaxErrorForFunctionCall(self):
         self.assertRaisesSyntaxError("f(p, k=1, p2)")
         self.assertRaisesSyntaxError("f(p, k1=50, *(1,2), k1=100)")

Modified: python/branches/py3k-dtoa/Lib/test/test_logging.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_logging.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_logging.py	Tue Jul  6 17:55:27 2010
@@ -1670,6 +1670,8 @@
         t = logging.config.listen(port)
         t.start()
         t.ready.wait()
+        # Now get the port allocated
+        port = t.port
         t.ready.clear()
         try:
             sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)

Modified: python/branches/py3k-dtoa/Lib/test/test_math.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_math.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_math.py	Tue Jul  6 17:55:27 2010
@@ -923,13 +923,6 @@
         self.assertRaises(TypeError, math.trunc, 1, 2)
         self.assertRaises(TypeError, math.trunc, TestNoTrunc())
 
-        # XXX Doesn't work because the method is looked up on
-        #     the type only.
-        #t = TestNoTrunc()
-        #t.__trunc__ = lambda *args: args
-        #self.assertEquals((), math.trunc(t))
-        #self.assertRaises(TypeError, math.trunc, t, 0)
-
     def testIsnan(self):
         self.assertTrue(math.isnan(float("nan")))
         self.assertTrue(math.isnan(float("inf")* 0.))

Modified: python/branches/py3k-dtoa/Lib/test/test_optparse.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_optparse.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_optparse.py	Tue Jul  6 17:55:27 2010
@@ -754,6 +754,11 @@
                            {'a': "-b3", 'boo': None, 'foo': None},
                            [])
 
+    def test_combined_single_invalid_option(self):
+        self.parser.add_option("-t", action="store_true")
+        self.assertParseFail(["-test"],
+                             "no such option: -e")
+
 class TestBool(BaseTest):
     def setUp(self):
         options = [make_option("-v",

Modified: python/branches/py3k-dtoa/Lib/test/test_os.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_os.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_os.py	Tue Jul  6 17:55:27 2010
@@ -58,6 +58,7 @@
         os.closerange(first, first + 2)
         self.assertRaises(OSError, os.write, first, b"a")
 
+    @support.cpython_only
     def test_rename(self):
         path = support.TESTFN
         old = sys.getrefcount(path)

Modified: python/branches/py3k-dtoa/Lib/test/test_parser.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_parser.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_parser.py	Tue Jul  6 17:55:27 2010
@@ -149,6 +149,13 @@
     def test_class_defs(self):
         self.check_suite("class foo():pass")
         self.check_suite("class foo(object):pass")
+        self.check_suite("@class_decorator\n"
+                         "class foo():pass")
+        self.check_suite("@class_decorator(arg)\n"
+                         "class foo():pass")
+        self.check_suite("@decorator1\n"
+                         "@decorator2\n"
+                         "class foo():pass")
 
     def test_import_from_statement(self):
         self.check_suite("from sys.path import *")
@@ -182,6 +189,18 @@
         self.check_suite("import sys as system, math")
         self.check_suite("import sys, math as my_math")
 
+    def test_relative_imports(self):
+        self.check_suite("from . import name")
+        self.check_suite("from .. import name")
+        # check all the way up to '....', since '...' is tokenized
+        # differently from '.' (it's an ellipsis token).
+        self.check_suite("from ... import name")
+        self.check_suite("from .... import name")
+        self.check_suite("from .pkg import name")
+        self.check_suite("from ..pkg import name")
+        self.check_suite("from ...pkg import name")
+        self.check_suite("from ....pkg import name")
+
     def test_pep263(self):
         self.check_suite("# -*- coding: iso-8859-1 -*-\n"
                          "pass\n")
@@ -457,6 +476,20 @@
                 (0, ''))
         self.check_bad_tree(tree, "malformed global ast")
 
+    def test_missing_import_source(self):
+        # from import fred
+        tree = \
+            (257,
+             (268,
+              (269,
+               (270,
+                (282,
+                 (284, (1, 'from'), (1, 'import'),
+                  (287, (285, (1, 'fred')))))),
+               (4, ''))),
+             (4, ''), (0, ''))
+        self.check_bad_tree(tree, "from import fred")
+
 
 class CompileTestCase(unittest.TestCase):
 
@@ -484,8 +517,18 @@
         st = parser.suite('a = "\\u1"')
         self.assertRaises(SyntaxError, parser.compilest, st)
 
+    def test_issue_9011(self):
+        # Issue 9011: compilation of an unary minus expression changed
+        # the meaning of the ST, so that a second compilation produced
+        # incorrect results.
+        st = parser.expr('-3')
+        code1 = parser.compilest(st)
+        self.assertEqual(eval(code1), -3)
+        code2 = parser.compilest(st)
+        self.assertEqual(eval(code2), -3)
+
 class ParserStackLimitTestCase(unittest.TestCase):
-    """try to push the parser to/over it's limits.
+    """try to push the parser to/over its limits.
     see http://bugs.python.org/issue1881 for a discussion
     """
     def _nested_expression(self, level):

Modified: python/branches/py3k-dtoa/Lib/test/test_property.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_property.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_property.py	Tue Jul  6 17:55:27 2010
@@ -118,7 +118,7 @@
         self.assertEqual(base.__class__.spam.__doc__, "spam spam spam")
         self.assertEqual(sub.__class__.spam.__doc__, "spam spam spam")
 
-    @unittest.skipIf(sys.flags.optimize >= 1,
+    @unittest.skipIf(sys.flags.optimize >= 2,
                      "Docstrings are omitted with -O2 and above")
     def test_property_getter_doc_override(self):
         newgettersub = PropertySubNewGetter()

Modified: python/branches/py3k-dtoa/Lib/test/test_pyexpat.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_pyexpat.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_pyexpat.py	Tue Jul  6 17:55:27 2010
@@ -184,7 +184,7 @@
             self.fail()
         except TypeError as e:
             self.assertEquals(str(e),
-                'ParserCreate() argument 2 must be string or None, not int')
+                'ParserCreate() argument 2 must be str or None, not int')
 
         try:
             expat.ParserCreate(namespace_separator='too long')

Modified: python/branches/py3k-dtoa/Lib/test/test_runpy.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_runpy.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_runpy.py	Tue Jul  6 17:55:27 2010
@@ -6,7 +6,7 @@
 import re
 import tempfile
 import py_compile
-from test.support import forget, make_legacy_pyc, run_unittest, verbose
+from test.support import forget, make_legacy_pyc, run_unittest, unload, verbose
 from test.script_helper import (
     make_pkg, make_script, make_zip_pkg, make_zip_script, temp_dir)
 
@@ -174,6 +174,7 @@
             __import__(mod_name)
             os.remove(mod_fname)
             make_legacy_pyc(mod_fname)
+            unload(mod_name)  # In case loader caches paths
             if verbose: print("Running from compiled:", mod_name)
             d2 = run_module(mod_name) # Read from bytecode
             self.assertIn("x", d2)
@@ -197,6 +198,7 @@
             __import__(mod_name)
             os.remove(mod_fname)
             make_legacy_pyc(mod_fname)
+            unload(mod_name)  # In case loader caches paths
             if verbose: print("Running from compiled:", pkg_name)
             d2 = run_module(pkg_name) # Read from bytecode
             self.assertIn("x", d2)
@@ -252,6 +254,7 @@
             __import__(mod_name)
             os.remove(mod_fname)
             make_legacy_pyc(mod_fname)
+            unload(mod_name)  # In case the loader caches paths
             if verbose: print("Running from compiled:", mod_name)
             d2 = run_module(mod_name, run_name=run_name) # Read from bytecode
             self.assertIn("__package__", d2)
@@ -405,7 +408,11 @@
 
 
 def test_main():
-    run_unittest(RunModuleCodeTest, RunModuleTest, RunPathTest)
+    run_unittest(
+                 RunModuleCodeTest,
+                 RunModuleTest,
+                 RunPathTest
+                 )
 
 if __name__ == "__main__":
     test_main()

Modified: python/branches/py3k-dtoa/Lib/test/test_strtod.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_strtod.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_strtod.py	Tue Jul  6 17:55:27 2010
@@ -23,6 +23,8 @@
     \Z
 """, re.VERBOSE | re.IGNORECASE).match
 
+# Pure Python version of correctly rounded string->float conversion.
+# Avoids any use of floating-point by returning the result as a hex string.
 def strtod(s, mant_dig=53, min_exp = -1021, max_exp = 1024):
     """Convert a finite decimal string to a hex string representing an
     IEEE 754 binary64 float.  Return 'inf' or '-inf' on overflow.
@@ -330,6 +332,10 @@
             '18487398785991994634182916638542680759613590482273e-357',
             '32002864200581033134358724675198044527469366773928e-358',
             '94393431193180696942841837085033647913224148539854e-358',
+            '73608278998966969345824653500136787876436005957953e-358',
+            '64774478836417299491718435234611299336288082136054e-358',
+            '13704940134126574534878641876947980878824688451169e-357',
+            '46697445774047060960624497964425416610480524760471e-358',
             # failing case for bug introduced by METD in r77451 (attempted
             # fix for issue 7632, bug 2), and fixed in r77482.
             '28639097178261763178489759107321392745108491825303e-311',

Modified: python/branches/py3k-dtoa/Lib/test/test_subprocess.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_subprocess.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_subprocess.py	Tue Jul  6 17:55:27 2010
@@ -478,8 +478,6 @@
                          '"a\\\\b\\ c" d e')
         self.assertEqual(subprocess.list2cmdline(['ab', '']),
                          'ab ""')
-        self.assertEqual(subprocess.list2cmdline(['echo', 'foo|bar']),
-                         'echo "foo|bar"')
 
 
     def test_poll(self):

Modified: python/branches/py3k-dtoa/Lib/test/test_syntax.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_syntax.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_syntax.py	Tue Jul  6 17:55:27 2010
@@ -478,10 +478,12 @@
 
 >>> del ()
 Traceback (most recent call last):
-   ...
-  File "<doctest test.test_syntax[50]>", line 1
 SyntaxError: can't delete ()
 
+>>> {1, 2, 3} = 42
+Traceback (most recent call last):
+SyntaxError: can't assign to literal
+
 """
 
 import re

Modified: python/branches/py3k-dtoa/Lib/test/test_tokenize.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_tokenize.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_tokenize.py	Tue Jul  6 17:55:27 2010
@@ -42,7 +42,7 @@
         ...
     IndentationError: unindent does not match any outer indentation level
 
-There are some standard formattig practises that are easy to get right.
+There are some standard formatting practices that are easy to get right.
 
     >>> roundtrip("if x == 1:\\n"
     ...           "    print(x)\\n")

Modified: python/branches/py3k-dtoa/Lib/test/test_tuple.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_tuple.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_tuple.py	Tue Jul  6 17:55:27 2010
@@ -96,6 +96,7 @@
         gc.collect()
         self.assertTrue(gc.is_tracked(t), t)
 
+    @support.cpython_only
     def test_track_literals(self):
         # Test GC-optimization of tuple literals
         x, y, z = 1.5, "a", []
@@ -136,16 +137,19 @@
         self._tracked(tp(tuple([obj]) for obj in [x, y, z]))
         self._tracked(tuple(tp([obj]) for obj in [x, y, z]))
 
+    @support.cpython_only
     def test_track_dynamic(self):
         # Test GC-optimization of dynamically constructed tuples.
         self.check_track_dynamic(tuple, False)
 
+    @support.cpython_only
     def test_track_subtypes(self):
         # Tuple subtypes must always be tracked
         class MyTuple(tuple):
             pass
         self.check_track_dynamic(MyTuple, True)
 
+    @support.cpython_only
     def test_bug7466(self):
         # Trying to untrack an unfinished tuple could crash Python
         self._not_tracked(tuple(gc.collect() for i in range(101)))

Modified: python/branches/py3k-dtoa/Lib/test/test_unicode.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_unicode.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_unicode.py	Tue Jul  6 17:55:27 2010
@@ -942,6 +942,159 @@
         # * strict decoding testing for all of the
         #   UTF8_ERROR cases in PyUnicode_DecodeUTF8
 
+    def test_utf8_decode_valid_sequences(self):
+        sequences = [
+            # single byte
+            (b'\x00', '\x00'), (b'a', 'a'), (b'\x7f', '\x7f'),
+            # 2 bytes
+            (b'\xc2\x80', '\x80'), (b'\xdf\xbf', '\u07ff'),
+            # 3 bytes
+            (b'\xe0\xa0\x80', '\u0800'), (b'\xed\x9f\xbf', '\ud7ff'),
+            (b'\xee\x80\x80', '\uE000'), (b'\xef\xbf\xbf', '\uffff'),
+            # 4 bytes
+            (b'\xF0\x90\x80\x80', '\U00010000'),
+            (b'\xf4\x8f\xbf\xbf', '\U0010FFFF')
+        ]
+        for seq, res in sequences:
+            self.assertEqual(seq.decode('utf-8'), res)
+
+
+    def test_utf8_decode_invalid_sequences(self):
+        # continuation bytes in a sequence of 2, 3, or 4 bytes
+        continuation_bytes = [bytes([x]) for x in range(0x80, 0xC0)]
+        # start bytes of a 2-byte sequence equivalent to codepoints < 0x7F
+        invalid_2B_seq_start_bytes = [bytes([x]) for x in range(0xC0, 0xC2)]
+        # start bytes of a 4-byte sequence equivalent to codepoints > 0x10FFFF
+        invalid_4B_seq_start_bytes = [bytes([x]) for x in range(0xF5, 0xF8)]
+        invalid_start_bytes = (
+            continuation_bytes + invalid_2B_seq_start_bytes +
+            invalid_4B_seq_start_bytes + [bytes([x]) for x in range(0xF7, 0x100)]
+        )
+
+        for byte in invalid_start_bytes:
+            self.assertRaises(UnicodeDecodeError, byte.decode, 'utf-8')
+
+        for sb in invalid_2B_seq_start_bytes:
+            for cb in continuation_bytes:
+                self.assertRaises(UnicodeDecodeError, (sb+cb).decode, 'utf-8')
+
+        for sb in invalid_4B_seq_start_bytes:
+            for cb1 in continuation_bytes[:3]:
+                for cb3 in continuation_bytes[:3]:
+                    self.assertRaises(UnicodeDecodeError,
+                                      (sb+cb1+b'\x80'+cb3).decode, 'utf-8')
+
+        for cb in [bytes([x]) for x in range(0x80, 0xA0)]:
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xE0'+cb+b'\x80').decode, 'utf-8')
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xE0'+cb+b'\xBF').decode, 'utf-8')
+        # surrogates
+        for cb in [bytes([x]) for x in range(0xA0, 0xC0)]:
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xED'+cb+b'\x80').decode, 'utf-8')
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xED'+cb+b'\xBF').decode, 'utf-8')
+        for cb in [bytes([x]) for x in range(0x80, 0x90)]:
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xF0'+cb+b'\x80\x80').decode, 'utf-8')
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xF0'+cb+b'\xBF\xBF').decode, 'utf-8')
+        for cb in [bytes([x]) for x in range(0x90, 0xC0)]:
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xF4'+cb+b'\x80\x80').decode, 'utf-8')
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xF4'+cb+b'\xBF\xBF').decode, 'utf-8')
+
+    def test_issue8271(self):
+        # Issue #8271: during the decoding of an invalid UTF-8 byte sequence,
+        # only the start byte and the continuation byte(s) are now considered
+        # invalid, instead of the number of bytes specified by the start byte.
+        # See http://www.unicode.org/versions/Unicode5.2.0/ch03.pdf (page 95,
+        # table 3-8, Row 2) for more information about the algorithm used.
+        FFFD = '\ufffd'
+        sequences = [
+            # invalid start bytes
+            (b'\x80', FFFD), # continuation byte
+            (b'\x80\x80', FFFD*2), # 2 continuation bytes
+            (b'\xc0', FFFD),
+            (b'\xc0\xc0', FFFD*2),
+            (b'\xc1', FFFD),
+            (b'\xc1\xc0', FFFD*2),
+            (b'\xc0\xc1', FFFD*2),
+            # with start byte of a 2-byte sequence
+            (b'\xc2', FFFD), # only the start byte
+            (b'\xc2\xc2', FFFD*2), # 2 start bytes
+            (b'\xc2\xc2\xc2', FFFD*3), # 2 start bytes
+            (b'\xc2\x41', FFFD+'A'), # invalid continuation byte
+            # with start byte of a 3-byte sequence
+            (b'\xe1', FFFD), # only the start byte
+            (b'\xe1\xe1', FFFD*2), # 2 start bytes
+            (b'\xe1\xe1\xe1', FFFD*3), # 3 start bytes
+            (b'\xe1\xe1\xe1\xe1', FFFD*4), # 4 start bytes
+            (b'\xe1\x80', FFFD), # only 1 continuation byte
+            (b'\xe1\x41', FFFD+'A'), # invalid continuation byte
+            (b'\xe1\x41\x80', FFFD+'A'+FFFD), # invalid cb followed by valid cb
+            (b'\xe1\x41\x41', FFFD+'AA'), # 2 invalid continuation bytes
+            (b'\xe1\x80\x41', FFFD+'A'), # only 1 valid continuation byte
+            (b'\xe1\x80\xe1\x41', FFFD*2+'A'), # 1 valid and the other invalid
+            (b'\xe1\x41\xe1\x80', FFFD+'A'+FFFD), # 1 invalid and the other valid
+            # with start byte of a 4-byte sequence
+            (b'\xf1', FFFD), # only the start byte
+            (b'\xf1\xf1', FFFD*2), # 2 start bytes
+            (b'\xf1\xf1\xf1', FFFD*3), # 3 start bytes
+            (b'\xf1\xf1\xf1\xf1', FFFD*4), # 4 start bytes
+            (b'\xf1\xf1\xf1\xf1\xf1', FFFD*5), # 5 start bytes
+            (b'\xf1\x80', FFFD), # only 1 continuation bytes
+            (b'\xf1\x80\x80', FFFD), # only 2 continuation bytes
+            (b'\xf1\x80\x41', FFFD+'A'), # 1 valid cb and 1 invalid
+            (b'\xf1\x80\x41\x41', FFFD+'AA'), # 1 valid cb and 1 invalid
+            (b'\xf1\x80\x80\x41', FFFD+'A'), # 2 valid cb and 1 invalid
+            (b'\xf1\x41\x80', FFFD+'A'+FFFD), # 1 invalid cv and 1 valid
+            (b'\xf1\x41\x80\x80', FFFD+'A'+FFFD*2), # 1 invalid cb and 2 invalid
+            (b'\xf1\x41\x80\x41', FFFD+'A'+FFFD+'A'), # 2 invalid cb and 1 invalid
+            (b'\xf1\x41\x41\x80', FFFD+'AA'+FFFD), # 1 valid cb and 1 invalid
+            (b'\xf1\x41\xf1\x80', FFFD+'A'+FFFD),
+            (b'\xf1\x41\x80\xf1', FFFD+'A'+FFFD*2),
+            (b'\xf1\xf1\x80\x41', FFFD*2+'A'),
+            (b'\xf1\x41\xf1\xf1', FFFD+'A'+FFFD*2),
+            # with invalid start byte of a 4-byte sequence (rfc2279)
+            (b'\xf5', FFFD), # only the start byte
+            (b'\xf5\xf5', FFFD*2), # 2 start bytes
+            (b'\xf5\x80', FFFD*2), # only 1 continuation byte
+            (b'\xf5\x80\x80', FFFD*3), # only 2 continuation byte
+            (b'\xf5\x80\x80\x80', FFFD*4), # 3 continuation bytes
+            (b'\xf5\x80\x41', FFFD*2+'A'), #  1 valid cb and 1 invalid
+            (b'\xf5\x80\x41\xf5', FFFD*2+'A'+FFFD),
+            (b'\xf5\x41\x80\x80\x41', FFFD+'A'+FFFD*2+'A'),
+            # with invalid start byte of a 5-byte sequence (rfc2279)
+            (b'\xf8', FFFD), # only the start byte
+            (b'\xf8\xf8', FFFD*2), # 2 start bytes
+            (b'\xf8\x80', FFFD*2), # only one continuation byte
+            (b'\xf8\x80\x41', FFFD*2 + 'A'), # 1 valid cb and 1 invalid
+            (b'\xf8\x80\x80\x80\x80', FFFD*5), # invalid 5 bytes seq with 5 bytes
+            # with invalid start byte of a 6-byte sequence (rfc2279)
+            (b'\xfc', FFFD), # only the start byte
+            (b'\xfc\xfc', FFFD*2), # 2 start bytes
+            (b'\xfc\x80\x80', FFFD*3), # only 2 continuation bytes
+            (b'\xfc\x80\x80\x80\x80\x80', FFFD*6), # 6 continuation bytes
+            # invalid start byte
+            (b'\xfe', FFFD),
+            (b'\xfe\x80\x80', FFFD*3),
+            # other sequences
+            (b'\xf1\x80\x41\x42\x43', '\ufffd\x41\x42\x43'),
+            (b'\xf1\x80\xff\x42\x43', '\ufffd\ufffd\x42\x43'),
+            (b'\xf1\x80\xc2\x81\x43', '\ufffd\x81\x43'),
+            (b'\x61\xF1\x80\x80\xE1\x80\xC2\x62\x80\x63\x80\xBF\x64',
+             '\x61\uFFFD\uFFFD\uFFFD\x62\uFFFD\x63\uFFFD\uFFFD\x64'),
+        ]
+        for n, (seq, res) in enumerate(sequences):
+            self.assertRaises(UnicodeDecodeError, seq.decode, 'utf-8', 'strict')
+            self.assertEqual(seq.decode('utf-8', 'replace'), res)
+            self.assertEqual((seq+b'b').decode('utf-8', 'replace'), res+'b')
+            self.assertEqual(seq.decode('utf-8', 'ignore'),
+                             res.replace('\uFFFD', ''))
+
     def test_codecs_idna(self):
         # Test whether trailing dot is preserved
         self.assertEqual("www.python.org.".encode("idna"), b"www.python.org.")
@@ -1039,8 +1192,8 @@
         s = bytes(range(128))
         for encoding in (
             'cp037', 'cp1026',
-            'cp437', 'cp500', 'cp737', 'cp775', 'cp850',
-            'cp852', 'cp855', 'cp860', 'cp861', 'cp862',
+            'cp437', 'cp500', 'cp720', 'cp737', 'cp775', 'cp850',
+            'cp852', 'cp855', 'cp858', 'cp860', 'cp861', 'cp862',
             'cp863', 'cp865', 'cp866',
             'iso8859_10', 'iso8859_13', 'iso8859_14', 'iso8859_15',
             'iso8859_2', 'iso8859_3', 'iso8859_4', 'iso8859_5', 'iso8859_6',
@@ -1067,8 +1220,8 @@
         s = bytes(range(128, 256))
         for encoding in (
             'cp037', 'cp1026',
-            'cp437', 'cp500', 'cp737', 'cp775', 'cp850',
-            'cp852', 'cp855', 'cp860', 'cp861', 'cp862',
+            'cp437', 'cp500', 'cp720', 'cp737', 'cp775', 'cp850',
+            'cp852', 'cp855', 'cp858', 'cp860', 'cp861', 'cp862',
             'cp863', 'cp865', 'cp866',
             'iso8859_10', 'iso8859_13', 'iso8859_14', 'iso8859_15',
             'iso8859_2', 'iso8859_4', 'iso8859_5',

Modified: python/branches/py3k-dtoa/Lib/test/test_urllib.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_urllib.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_urllib.py	Tue Jul  6 17:55:27 2010
@@ -795,6 +795,116 @@
         self.assertEqual("a=a&a=b",
                          urllib.parse.urlencode({"a": {"a": 1, "b": 1}}, True))
 
+    def test_urlencode_encoding(self):
+        # ASCII encoding. Expect %3F with errors="replace'
+        given = (('\u00a0', '\u00c1'),)
+        expect = '%3F=%3F'
+        result = urllib.parse.urlencode(given, encoding="ASCII", errors="replace")
+        self.assertEqual(expect, result)
+
+        # Default is UTF-8 encoding.
+        given = (('\u00a0', '\u00c1'),)
+        expect = '%C2%A0=%C3%81'
+        result = urllib.parse.urlencode(given)
+        self.assertEqual(expect, result)
+
+        # Latin-1 encoding.
+        given = (('\u00a0', '\u00c1'),)
+        expect = '%A0=%C1'
+        result = urllib.parse.urlencode(given, encoding="latin-1")
+        self.assertEqual(expect, result)
+
+    def test_urlencode_encoding_doseq(self):
+        # ASCII Encoding. Expect %3F with errors="replace'
+        given = (('\u00a0', '\u00c1'),)
+        expect = '%3F=%3F'
+        result = urllib.parse.urlencode(given, doseq=True,
+                                        encoding="ASCII", errors="replace")
+        self.assertEqual(expect, result)
+
+        # ASCII Encoding. On a sequence of values.
+        given = (("\u00a0", (1, "\u00c1")),)
+        expect = '%3F=1&%3F=%3F'
+        result = urllib.parse.urlencode(given, True,
+                                        encoding="ASCII", errors="replace")
+        self.assertEqual(expect, result)
+
+        # Utf-8
+        given = (("\u00a0", "\u00c1"),)
+        expect = '%C2%A0=%C3%81'
+        result = urllib.parse.urlencode(given, True)
+        self.assertEqual(expect, result)
+
+        given = (("\u00a0", (42, "\u00c1")),)
+        expect = '%C2%A0=42&%C2%A0=%C3%81'
+        result = urllib.parse.urlencode(given, True)
+        self.assertEqual(expect, result)
+
+        # latin-1
+        given = (("\u00a0", "\u00c1"),)
+        expect = '%A0=%C1'
+        result = urllib.parse.urlencode(given, True, encoding="latin-1")
+        self.assertEqual(expect, result)
+
+        given = (("\u00a0", (42, "\u00c1")),)
+        expect = '%A0=42&%A0=%C1'
+        result = urllib.parse.urlencode(given, True, encoding="latin-1")
+        self.assertEqual(expect, result)
+
+    def test_urlencode_bytes(self):
+        given = ((b'\xa0\x24', b'\xc1\x24'),)
+        expect = '%A0%24=%C1%24'
+        result = urllib.parse.urlencode(given)
+        self.assertEqual(expect, result)
+        result = urllib.parse.urlencode(given, True)
+        self.assertEqual(expect, result)
+
+        # Sequence of values
+        given = ((b'\xa0\x24', (42, b'\xc1\x24')),)
+        expect = '%A0%24=42&%A0%24=%C1%24'
+        result = urllib.parse.urlencode(given, True)
+        self.assertEqual(expect, result)
+
+    def test_urlencode_encoding_safe_parameter(self):
+
+        # Send '$' (\x24) as safe character
+        # Default utf-8 encoding
+
+        given = ((b'\xa0\x24', b'\xc1\x24'),)
+        result = urllib.parse.urlencode(given, safe=":$")
+        expect = '%A0$=%C1$'
+        self.assertEqual(expect, result)
+
+        given = ((b'\xa0\x24', b'\xc1\x24'),)
+        result = urllib.parse.urlencode(given, doseq=True, safe=":$")
+        expect = '%A0$=%C1$'
+        self.assertEqual(expect, result)
+
+        # Safe parameter in sequence
+        given = ((b'\xa0\x24', (b'\xc1\x24', 0xd, 42)),)
+        expect = '%A0$=%C1$&%A0$=13&%A0$=42'
+        result = urllib.parse.urlencode(given, True, safe=":$")
+        self.assertEqual(expect, result)
+
+        # Test all above in latin-1 encoding
+
+        given = ((b'\xa0\x24', b'\xc1\x24'),)
+        result = urllib.parse.urlencode(given, safe=":$",
+                                        encoding="latin-1")
+        expect = '%A0$=%C1$'
+        self.assertEqual(expect, result)
+
+        given = ((b'\xa0\x24', b'\xc1\x24'),)
+        expect = '%A0$=%C1$'
+        result = urllib.parse.urlencode(given, doseq=True, safe=":$",
+                                        encoding="latin-1")
+
+        given = ((b'\xa0\x24', (b'\xc1\x24', 0xd, 42)),)
+        expect = '%A0$=%C1$&%A0$=13&%A0$=42'
+        result = urllib.parse.urlencode(given, True, safe=":$",
+                                        encoding="latin-1")
+        self.assertEqual(expect, result)
+
 class Pathname_Tests(unittest.TestCase):
     """Test pathname2url() and url2pathname()"""
 

Modified: python/branches/py3k-dtoa/Lib/test/test_xmlrpc_net.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_xmlrpc_net.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_xmlrpc_net.py	Tue Jul  6 17:55:27 2010
@@ -48,7 +48,7 @@
         # Perform a minimal sanity check on the result, just to be sure
         # the request means what we think it means.
         self.assertIsInstance(builders, collections.Sequence)
-        self.assertTrue([x for x in builders if "trunk" in x], builders)
+        self.assertTrue([x for x in builders if "3.x" in x], builders)
 
 
 def test_main():

Modified: python/branches/py3k-dtoa/Lib/urllib/parse.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/urllib/parse.py	(original)
+++ python/branches/py3k-dtoa/Lib/urllib/parse.py	Tue Jul  6 17:55:27 2010
@@ -8,21 +8,22 @@
 RFC 2732 : "Format for Literal IPv6 Addresses in URL's by R.Hinden, B.Carpenter
 and L.Masinter, December 1999.
 
-RFC2396:  "Uniform Resource Identifiers (URI)": Generic Syntax by T.
+RFC 2396:  "Uniform Resource Identifiers (URI)": Generic Syntax by T.
 Berners-Lee, R. Fielding, and L. Masinter, August 1998.
 
-RFC2368: "The mailto URL scheme", by P.Hoffman , L Masinter, J. Zwinski, July 1998.
+RFC 2368: "The mailto URL scheme", by P.Hoffman , L Masinter, J. Zwinski, July 1998.
 
 RFC 1808: "Relative Uniform Resource Locators", by R. Fielding, UC Irvine, June
 1995.
 
-RFC1738: "Uniform Resource Locators (URL)" by T. Berners-Lee, L. Masinter, M.
+RFC 1738: "Uniform Resource Locators (URL)" by T. Berners-Lee, L. Masinter, M.
 McCahill, December 1994
 
-RFC 3986 is considered the current standard and any changes to urlparse module
-should conform to this.  urlparse module is not entirely compliant with this.
-The defacto scenarios of parsing are considered sometimes and for backward
-compatiblity purposes, older RFC uses of parsing are retained. The testcases in
+RFC 3986 is considered the current standard and any future changes to
+urlparse module should conform with it.  The urlparse module is
+currently not entirely compliant with this RFC due to defacto
+scenarios for parsing, and for backward compatibility purposes, some
+parsing quirks from older RFCs are retained. The testcases in
 test_urlparse.py provides a good indicator of parsing behavior.
 """
 
@@ -220,6 +221,11 @@
     return urlunsplit((scheme, netloc, url, query, fragment))
 
 def urlunsplit(components):
+    """Combine the elements of a tuple as returned by urlsplit() into a
+    complete URL as a string. The data argument can be any five-item iterable.
+    This may result in a slightly different, but equivalent URL, if the URL that
+    was parsed originally had unnecessary delimiters (for example, a ? with an
+    empty query; the RFC states that these are equivalent)."""
     scheme, netloc, url, query, fragment = components
     if netloc or (scheme and scheme in uses_netloc and url[:2] != '//'):
         if url and url[:1] != '/': url = '/' + url
@@ -553,7 +559,7 @@
         _safe_quoters[safe] = quoter = Quoter(safe).__getitem__
     return ''.join([quoter(char) for char in bs])
 
-def urlencode(query, doseq=False):
+def urlencode(query, doseq=False, safe='', encoding=None, errors=None):
     """Encode a sequence of two-element tuples or dictionary into a URL query string.
 
     If any values in the query arg are sequences and doseq is true, each
@@ -562,6 +568,10 @@
     If the query arg is a sequence of two-element tuples, the order of the
     parameters in the output will match the order of parameters in the
     input.
+
+    The query arg may be either a string or a bytes type. When query arg is a
+    string, the safe, encoding and error parameters are sent the quote_plus for
+    encoding.
     """
 
     if hasattr(query, "items"):
@@ -586,14 +596,28 @@
     l = []
     if not doseq:
         for k, v in query:
-            k = quote_plus(str(k))
-            v = quote_plus(str(v))
+            if isinstance(k, bytes):
+                k = quote_plus(k, safe)
+            else:
+                k = quote_plus(str(k), safe, encoding, errors)
+
+            if isinstance(v, bytes):
+                v = quote_plus(v, safe)
+            else:
+                v = quote_plus(str(v), safe, encoding, errors)
             l.append(k + '=' + v)
     else:
         for k, v in query:
-            k = quote_plus(str(k))
-            if isinstance(v, str):
-                v = quote_plus(v)
+            if isinstance(k, bytes):
+                k = quote_plus(k, safe)
+            else:
+                k = quote_plus(str(k), safe, encoding, errors)
+
+            if isinstance(v, bytes):
+                v = quote_plus(v, safe)
+                l.append(k + '=' + v)
+            elif isinstance(v, str):
+                v = quote_plus(v, safe, encoding, errors)
                 l.append(k + '=' + v)
             else:
                 try:
@@ -601,12 +625,16 @@
                     x = len(v)
                 except TypeError:
                     # not a sequence
-                    v = quote_plus(str(v))
+                    v = quote_plus(str(v), safe, encoding, errors)
                     l.append(k + '=' + v)
                 else:
                     # loop over the sequence
                     for elt in v:
-                        l.append(k + '=' + quote_plus(str(elt)))
+                        if isinstance(elt, bytes):
+                            elt = quote_plus(elt, safe)
+                        else:
+                            elt = quote_plus(str(elt), safe, encoding, errors)
+                        l.append(k + '=' + elt)
     return '&'.join(l)
 
 # Utilities to parse URLs (most of these return None for missing parts):

Modified: python/branches/py3k-dtoa/Lib/urllib/request.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/urllib/request.py	(original)
+++ python/branches/py3k-dtoa/Lib/urllib/request.py	Tue Jul  6 17:55:27 2010
@@ -1866,7 +1866,8 @@
         else:
             return self.http_error_default(url, fp, errcode, errmsg, headers)
 
-    def http_error_401(self, url, fp, errcode, errmsg, headers, data=None):
+    def http_error_401(self, url, fp, errcode, errmsg, headers, data=None,
+            retry=False):
         """Error 401 -- authentication required.
         This function supports Basic authentication only."""
         if not 'www-authenticate' in headers:
@@ -1882,13 +1883,17 @@
         if scheme.lower() != 'basic':
             URLopener.http_error_default(self, url, fp,
                                          errcode, errmsg, headers)
+        if not retry:
+            URLopener.http_error_default(self, url, fp, errcode, errmsg,
+                    headers)
         name = 'retry_' + self.type + '_basic_auth'
         if data is None:
             return getattr(self,name)(url, realm)
         else:
             return getattr(self,name)(url, realm, data)
 
-    def http_error_407(self, url, fp, errcode, errmsg, headers, data=None):
+    def http_error_407(self, url, fp, errcode, errmsg, headers, data=None,
+            retry=False):
         """Error 407 -- proxy authentication required.
         This function supports Basic authentication only."""
         if not 'proxy-authenticate' in headers:
@@ -1904,6 +1909,9 @@
         if scheme.lower() != 'basic':
             URLopener.http_error_default(self, url, fp,
                                          errcode, errmsg, headers)
+        if not retry:
+            URLopener.http_error_default(self, url, fp, errcode, errmsg,
+                    headers)
         name = 'retry_proxy_' + self.type + '_basic_auth'
         if data is None:
             return getattr(self,name)(url, realm)
@@ -2196,8 +2204,13 @@
                         continue
 
                 base = ip2num(m.group(1))
-                mask = int(m.group(2)[1:])
-                mask = 32 - mask
+                mask = m.group(2)
+                if mask is None:
+                    mask = 8 * (m.group(1).count('.') + 1)
+
+                else:
+                    mask = int(mask[1:])
+                    mask = 32 - mask
 
                 if (hostIP >> mask) == (base >> mask):
                     return True

Modified: python/branches/py3k-dtoa/Lib/uuid.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/uuid.py	(original)
+++ python/branches/py3k-dtoa/Lib/uuid.py	Tue Jul  6 17:55:27 2010
@@ -427,17 +427,17 @@
         if hasattr(lib, 'uuid_generate_time'):
             _uuid_generate_time = lib.uuid_generate_time
 
-    # The uuid_generate_* functions are broken on MacOS X 10.6, as noted
+    # The uuid_generate_* functions are broken on MacOS X 10.5, as noted
     # in issue #8621 the function generates the same sequence of values
     # in the parent process and all children created using fork (unless
     # those children use exec as well).
     #
-    # Assume that the uuid_generate functions are broken from 10.6 onward,
+    # Assume that the uuid_generate functions are broken from 10.5 onward,
     # the test can be adjusted when a later version is fixed.
     import sys
     if sys.platform == 'darwin':
         import os
-        if int(os.uname()[2].split('.')[0]) >= 10:
+        if int(os.uname()[2].split('.')[0]) >= 9:
             _uuid_generate_random = _uuid_generate_time = None
 
     # On Windows prior to 2000, UuidCreate gives a UUID containing the

Modified: python/branches/py3k-dtoa/Lib/warnings.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/warnings.py	(original)
+++ python/branches/py3k-dtoa/Lib/warnings.py	Tue Jul  6 17:55:27 2010
@@ -371,8 +371,10 @@
 # Module initialization
 _processoptions(sys.warnoptions)
 if not _warnings_defaults:
-    simplefilter("ignore", category=PendingDeprecationWarning, append=1)
-    simplefilter("ignore", category=ImportWarning, append=1)
+    silence = [ImportWarning, PendingDeprecationWarning]
+    silence.append(DeprecationWarning)
+    for cls in silence:
+        simplefilter("ignore", category=cls)
     bytes_warning = sys.flags.bytes_warning
     if bytes_warning > 1:
         bytes_action = "error"

Modified: python/branches/py3k-dtoa/Misc/ACKS
==============================================================================
--- python/branches/py3k-dtoa/Misc/ACKS	(original)
+++ python/branches/py3k-dtoa/Misc/ACKS	Tue Jul  6 17:55:27 2010
@@ -797,6 +797,7 @@
 Eren T??rkay
 Lionel Ulmer
 Roger Upole
+Daniel Urban
 Michael Urman
 Hector Urtubia
 Andi Vajda

Modified: python/branches/py3k-dtoa/Misc/NEWS
==============================================================================
--- python/branches/py3k-dtoa/Misc/NEWS	(original)
+++ python/branches/py3k-dtoa/Misc/NEWS	Tue Jul  6 17:55:27 2010
@@ -12,6 +12,40 @@
 Core and Builtins
 -----------------
 
+- Issue #8271: during the decoding of an invalid UTF-8 byte sequence, only the
+  start byte and the continuation byte(s) are now considered invalid, instead
+  of the number of bytes specified by the start byte.
+  E.g.: '\xf1\x80AB'.decode('utf-8', 'replace') now returns u'\ufffdAB' and
+  replaces with U+FFFD only the start byte ('\xf1') and the continuation byte
+  ('\x80') even if '\xf1' is the start byte of a 4-bytes sequence.
+  Previous versions returned a single u'\ufffd'.
+
+- Issue #9011: A negated imaginary literal (e.g., "-7j") now has real
+  part -0.0 rather than 0.0.  So "-7j" is now exactly equivalent to
+  "-(7j)".
+
+- Be more specific in error messages about positional arguments.
+
+- Issue #8949: "z" format of PyArg_Parse*() functions doesn't accept bytes
+  objects, as described in the documentation.
+
+- Issue #6543: Write the traceback in the terminal encoding instead of utf-8.
+  Fix the encoding of the modules filename. Patch written by Amaury Forgeot
+  d'Arc.
+
+- Issue #9011: Remove buggy and unnecessary (in 3.x) ST->AST
+  compilation code dealing with unary minus applied to a constant.
+  The removed code was mutating the ST, causing a second compilation
+  to fail.
+
+- Issue #850997: mbcs encoding (Windows only) handles errors argument: strict
+  mode raises unicode errors. The encoder only supports "strict" and "replace"
+  error handlers, the decoder only supports "strict" and "ignore" error
+  handlers. Patch written by Mark Hammond.
+
+- Issue #8850: Remove "w" and "w#" formats from PyArg_Parse*() functions, use
+  "w*" format instead. Add tests for "w*" format.
+
 - Issue #8592: PyArg_Parse*() functions raise a TypeError for "y", "u" and "Z"
   formats if the string contains a null byte/character. Write unit tests for
   string formats.
@@ -23,6 +57,8 @@
 - Issue #8969: On Windows, use mbcs codec in strict mode to encode and decode
   filenames and enable os.fsencode().
 
+- Issue #9058: Remove assertions about INT_MAX in UnicodeDecodeError.
+
 - Issue #8941: decoding big endian UTF-32 data in UCS-2 builds could crash
   the interpreter with characters outside the Basic Multilingual Plane
   (higher than 0x10000).
@@ -432,6 +468,55 @@
 Library
 -------
 
+- Issue #9130: Fix validation of relative imports in parser module.
+
+- Issue #9128: Fix validation of class decorators in parser module.
+
+- Issue #9094: python -m pickletools will now disassemble pickle files
+  listed in the command line arguments.  See output of python -m
+  pickletools -h for more details.
+
+- Issue #5468: urlencode to handle bytes type and other encodings in its query
+  parameter. Patch by Dan Mahn.
+
+- Issue #7673: Fix security vulnerability (CVE-2010-2089) in the audioop
+  module, ensure that the input string length is a multiple of the frame size
+
+- Issue #6507: Accept source strings in dis.dis(). Original patch by
+  Daniel Urban.
+
+- Issue #7829: Clearly document that the dis module is exposing an
+  implementation detail that is not stable between Python VMs or releases.
+
+- Issue #6589: cleanup asyncore.socket_map in case smtpd.SMTPServer constructor
+  raises an exception.
+
+- Issue #9110: Addition of ContextDecorator to contextlib, for creating APIs
+  that act as both context managers and decorators. contextmanager changes
+  to use ContextDecorator.
+
+- Implement importlib.abc.SourceLoader and deprecate PyLoader and PyPycLoader
+  for removal in Python 3.4.
+
+- Issue #9064: pdb's "up" and "down" commands now accept an optional argument.
+
+- Issue #9018: os.path.normcase() now raises a TypeError if the argument is
+  not ``str`` or ``bytes``.
+
+- Issue #9075: In the ssl module, remove the setting of a ``debug`` flag
+  on an OpenSSL structure.
+
+- Issue #8682: The ssl module now temporary increments the reference count of
+  a socket object got through ``PyWeakref_GetObject``, so as to avoid possible
+  deallocation while the object is still being used.
+
+- Issue #1368368: FancyURLOpener class changed to throw an Exception on wrong
+  password instead of presenting an interactive prompt. Older behavior can be
+  obtained by passing retry=True to http_error_xxx methods of FancyURLOpener.
+
+- Issue #8720: fix regression caused by fix for #4050 by making getsourcefile
+  smart enough to find source files in the linecache.
+
 - Issue #5610: feedparser no longer eats extra characters at the end of
   a body part if the body part ends with a \r\n.
 
@@ -1309,6 +1394,22 @@
 Extension Modules
 -----------------
 
+- In the math module, correctly lookup __trunc__, __ceil__, and __floor__ as
+  special methods.
+
+- Issue #9005: Prevent utctimetuple() from producing year 0 or year
+  10,000.  Prior to this change, timezone adjustment in utctimetuple()
+  could produce tm_year value of 0 or 10,000.  Now an OverflowError is
+  raised in these edge cases.
+
+- Issue #6641: The ``datetime.strptime`` method now supports the
+  ``%z`` directive.  When the ``%z`` directive is present in the
+  format string, an aware ``datetime`` object is returned with
+  ``tzinfo`` bound to a ``datetime.timezone`` instance constructed
+  from the parsed offset.  If both ``%z`` and ``%Z`` are present, the
+  data in ``%Z`` field is used for timezone name, but ``%Z`` data
+  without ``%z`` is discarded.
+
 - Issue #5094: The ``datetime`` module now has a simple concrete class
   implementing ``datetime.tzinfo`` interface.  Instances of the new
   class, ``datetime.timezone``, return fixed name and UTC offset from
@@ -1407,6 +1508,8 @@
 Build
 -----
 
+- Issue #8854: Fix finding Visual Studio 2008 on Windows x64.
+
 - Issue #1759169, #8864: Drop _XOPEN_SOURCE on Solaris, define it for
   multiprocessing only.
 
@@ -2026,6 +2129,8 @@
 Library
 -------
 
+- Issue #8203: Fix IDLE Credits dialog: view_file() uses its encoding argument.
+
 - Issue #5311: bdist_msi can now build packages that do not depend on a
   specific Python version.
 

Modified: python/branches/py3k-dtoa/Misc/Vim/vimrc
==============================================================================
--- python/branches/py3k-dtoa/Misc/Vim/vimrc	(original)
+++ python/branches/py3k-dtoa/Misc/Vim/vimrc	Tue Jul  6 17:55:27 2010
@@ -43,7 +43,7 @@
 " Display tabs at the beginning of a line in Python mode as bad.
 au BufRead,BufNewFile *.py,*.pyw match BadWhitespace /^\t\+/
 " Make trailing whitespace be flagged as bad.
-au BufRead,BufNewFile *.py,*.pyw,*.c,*.h match BadWhitespace /\s\+$/
+au BufRead,BufNewFile *.py,*.pyw,*.c,*.h,*.rst match BadWhitespace /\s\+$/
 
 " Wrap text after a certain number of characters
 " Python: 79 

Modified: python/branches/py3k-dtoa/Modules/_ctypes/callproc.c
==============================================================================
--- python/branches/py3k-dtoa/Modules/_ctypes/callproc.c	(original)
+++ python/branches/py3k-dtoa/Modules/_ctypes/callproc.c	Tue Jul  6 17:55:27 2010
@@ -761,6 +761,7 @@
     ffi_cif cif;
     int cc;
 #ifdef MS_WIN32
+    int delta;
 #ifndef DONT_USE_SEH
     DWORD dwExceptionCode = 0;
     EXCEPTION_RECORD record;
@@ -811,8 +812,9 @@
 #ifndef DONT_USE_SEH
     __try {
 #endif
+        delta =
 #endif
-        ffi_call(&cif, (void *)pProc, resmem, avalues);
+                ffi_call(&cif, (void *)pProc, resmem, avalues);
 #ifdef MS_WIN32
 #ifndef DONT_USE_SEH
     }
@@ -844,6 +846,35 @@
         return -1;
     }
 #endif
+#ifdef MS_WIN64
+    if (delta != 0) {
+        PyErr_Format(PyExc_RuntimeError,
+                 "ffi_call failed with code %d",
+                 delta);
+        return -1;
+    }
+#else
+    if (delta < 0) {
+        if (flags & FUNCFLAG_CDECL)
+            PyErr_Format(PyExc_ValueError,
+                     "Procedure called with not enough "
+                     "arguments (%d bytes missing) "
+                     "or wrong calling convention",
+                     -delta);
+        else
+            PyErr_Format(PyExc_ValueError,
+                     "Procedure probably called with not enough "
+                     "arguments (%d bytes missing)",
+                     -delta);
+        return -1;
+    } else if (delta > 0) {
+        PyErr_Format(PyExc_ValueError,
+                 "Procedure probably called with too many "
+                 "arguments (%d bytes in excess)",
+                 delta);
+        return -1;
+    }
+#endif
 #endif
     if ((flags & FUNCFLAG_PYTHONAPI) && PyErr_Occurred())
         return -1;
@@ -1116,7 +1147,11 @@
     }
     for (i = 0; i < argcount; ++i) {
         atypes[i] = args[i].ffi_type;
-        if (atypes[i]->type == FFI_TYPE_STRUCT)
+        if (atypes[i]->type == FFI_TYPE_STRUCT 
+#ifdef _WIN64
+            && atypes[i]->size <= sizeof(void *)
+#endif
+            )
             avalues[i] = (void *)args[i].value.p;
         else
             avalues[i] = (void *)&args[i].value;

Modified: python/branches/py3k-dtoa/Modules/_ctypes/libffi/fficonfig.py.in
==============================================================================
--- python/branches/py3k-dtoa/Modules/_ctypes/libffi/fficonfig.py.in	(original)
+++ python/branches/py3k-dtoa/Modules/_ctypes/libffi/fficonfig.py.in	Tue Jul  6 17:55:27 2010
@@ -6,7 +6,7 @@
     'MIPS_IRIX': ['src/mips/ffi.c', 'src/mips/o32.S', 'src/mips/n32.S'],
     'MIPS_LINUX': ['src/mips/ffi.c', 'src/mips/o32.S'],
     'X86': ['src/x86/ffi.c', 'src/x86/sysv.S'],
-    'X86_FREEBSD': ['src/x86/ffi.c', 'src/x86/sysv.S'],
+    'X86_FREEBSD': ['src/x86/ffi.c', 'src/x86/freebsd.S'],
     'X86_WIN32': ['src/x86/ffi.c', 'src/x86/win32.S'],
     'SPARC': ['src/sparc/ffi.c', 'src/sparc/v8.S', 'src/sparc/v9.S'],
     'ALPHA': ['src/alpha/ffi.c', 'src/alpha/osf.S'],

Modified: python/branches/py3k-dtoa/Modules/_ctypes/libffi_msvc/ffi.h
==============================================================================
--- python/branches/py3k-dtoa/Modules/_ctypes/libffi_msvc/ffi.h	(original)
+++ python/branches/py3k-dtoa/Modules/_ctypes/libffi_msvc/ffi.h	Tue Jul  6 17:55:27 2010
@@ -1,5 +1,5 @@
 /* -----------------------------------------------------------------*-C-*-
-   libffi @VERSION@ - Copyright (c) 1996-2003, 2007, 2008  Red Hat, Inc.
+   libffi 2.00-beta - Copyright (c) 1996-2003  Red Hat, Inc.
 
    Permission is hereby granted, free of charge, to any person obtaining
    a copy of this software and associated documentation files (the
@@ -12,14 +12,13 @@
    The above copyright notice and this permission notice shall be included
    in all copies or substantial portions of the Software.
 
-   THE SOFTWARE IS PROVIDED ``AS IS'', WITHOUT WARRANTY OF ANY KIND,
-   EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
-   MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
-   NONINFRINGEMENT.  IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
-   HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
-   WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
-   OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
-   DEALINGS IN THE SOFTWARE.
+   THE SOFTWARE IS PROVIDED ``AS IS'', WITHOUT WARRANTY OF ANY KIND, EXPRESS
+   OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+   MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
+   IN NO EVENT SHALL CYGNUS SOLUTIONS BE LIABLE FOR ANY CLAIM, DAMAGES OR
+   OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+   ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+   OTHER DEALINGS IN THE SOFTWARE.
 
    ----------------------------------------------------------------------- */
 
@@ -57,7 +56,7 @@
 #endif
 
 /* Specify which architecture libffi is configured for. */
-/* #define @TARGET@ */
+//XXX #define X86
 
 /* ---- System configuration information --------------------------------- */
 
@@ -65,10 +64,6 @@
 
 #ifndef LIBFFI_ASM
 
-#ifdef _MSC_VER
-#define __attribute__(X)
-#endif
-
 #include <stddef.h>
 #include <limits.h>
 
@@ -84,21 +79,12 @@
 #  ifdef __GNUC__
 #   define FFI_LONG_LONG_MAX __LONG_LONG_MAX__
 #  endif
+#  ifdef _MSC_VER
+#   define FFI_LONG_LONG_MAX _I64_MAX
+#  endif
 # endif
 #endif
 
-/* The closure code assumes that this works on pointers, i.e. a size_t	*/
-/* can hold a pointer.							*/
-
-typedef struct _ffi_type
-{
-  size_t size;
-  unsigned short alignment;
-  unsigned short type;
-  struct _ffi_type **elements;
-} ffi_type;
-
-#ifndef LIBFFI_HIDE_BASIC_TYPES
 #if SCHAR_MAX == 127
 # define ffi_type_uchar                ffi_type_uint8
 # define ffi_type_schar                ffi_type_sint8
@@ -129,23 +115,26 @@
  #error "int size not supported"
 #endif
 
+#define ffi_type_ulong         ffi_type_uint64
+#define ffi_type_slong         ffi_type_sint64
 #if LONG_MAX == 2147483647
 # if FFI_LONG_LONG_MAX != 9223372036854775807
- #error "no 64-bit data type supported"
+  #error "no 64-bit data type supported"
 # endif
 #elif LONG_MAX != 9223372036854775807
  #error "long size not supported"
 #endif
 
-#if LONG_MAX == 2147483647
-# define ffi_type_ulong        ffi_type_uint32
-# define ffi_type_slong        ffi_type_sint32
-#elif LONG_MAX == 9223372036854775807
-# define ffi_type_ulong        ffi_type_uint64
-# define ffi_type_slong        ffi_type_sint64
-#else
- #error "long size not supported"
-#endif
+/* The closure code assumes that this works on pointers, i.e. a size_t	*/
+/* can hold a pointer.							*/
+
+typedef struct _ffi_type
+{
+  size_t size;
+  unsigned short alignment;
+  unsigned short type;
+  /*@null@*/ struct _ffi_type **elements;
+} ffi_type;
 
 /* These are defined in types.c */
 extern ffi_type ffi_type_void;
@@ -159,19 +148,14 @@
 extern ffi_type ffi_type_sint64;
 extern ffi_type ffi_type_float;
 extern ffi_type ffi_type_double;
+extern ffi_type ffi_type_longdouble;
 extern ffi_type ffi_type_pointer;
 
-#if HAVE_LONG_DOUBLE
-extern ffi_type ffi_type_longdouble;
-#else
-#define ffi_type_longdouble ffi_type_double
-#endif
-#endif /* LIBFFI_HIDE_BASIC_TYPES */
 
 typedef enum {
   FFI_OK = 0,
   FFI_BAD_TYPEDEF,
-  FFI_BAD_ABI
+  FFI_BAD_ABI 
 } ffi_status;
 
 typedef unsigned FFI_TYPE;
@@ -179,8 +163,8 @@
 typedef struct {
   ffi_abi abi;
   unsigned nargs;
-  ffi_type **arg_types;
-  ffi_type *rtype;
+  /*@dependent@*/ ffi_type **arg_types;
+  /*@dependent@*/ ffi_type *rtype;
   unsigned bytes;
   unsigned flags;
 #ifdef FFI_EXTRA_CIF_FIELDS
@@ -190,16 +174,10 @@
 
 /* ---- Definitions for the raw API -------------------------------------- */
 
-#ifndef FFI_SIZEOF_ARG
-# if LONG_MAX == 2147483647
-#  define FFI_SIZEOF_ARG        4
-# elif LONG_MAX == 9223372036854775807
-#  define FFI_SIZEOF_ARG        8
-# endif
-#endif
-
-#ifndef FFI_SIZEOF_JAVA_RAW
-#  define FFI_SIZEOF_JAVA_RAW FFI_SIZEOF_ARG
+#ifdef _WIN64
+#define FFI_SIZEOF_ARG 8
+#else
+#define FFI_SIZEOF_ARG 4
 #endif
 
 typedef union {
@@ -210,25 +188,10 @@
   void*     ptr;
 } ffi_raw;
 
-#if FFI_SIZEOF_JAVA_RAW == 4 && FFI_SIZEOF_ARG == 8
-/* This is a special case for mips64/n32 ABI (and perhaps others) where
-   sizeof(void *) is 4 and FFI_SIZEOF_ARG is 8.  */
-typedef union {
-  signed int	sint;
-  unsigned int	uint;
-  float		flt;
-  char		data[FFI_SIZEOF_JAVA_RAW];
-  void*		ptr;
-} ffi_java_raw;
-#else
-typedef ffi_raw ffi_java_raw;
-#endif
-
-
-void ffi_raw_call (ffi_cif *cif,
-		   void (*fn)(void),
-		   void *rvalue,
-		   ffi_raw *avalue);
+void ffi_raw_call (/*@dependent@*/ ffi_cif *cif, 
+		   void (*fn)(), 
+		   /*@out@*/ void *rvalue, 
+		   /*@dependent@*/ ffi_raw *avalue);
 
 void ffi_ptrarray_to_raw (ffi_cif *cif, void **args, ffi_raw *raw);
 void ffi_raw_to_ptrarray (ffi_cif *cif, ffi_raw *raw, void **args);
@@ -238,35 +201,25 @@
 /* packing, even on 64-bit machines.  I.e. on 64-bit machines		*/
 /* longs and doubles are followed by an empty 64-bit word.		*/
 
-void ffi_java_raw_call (ffi_cif *cif,
-			void (*fn)(void),
-			void *rvalue,
-			ffi_java_raw *avalue);
+void ffi_java_raw_call (/*@dependent@*/ ffi_cif *cif, 
+		        void (*fn)(), 
+		        /*@out@*/ void *rvalue, 
+		        /*@dependent@*/ ffi_raw *avalue);
 
-void ffi_java_ptrarray_to_raw (ffi_cif *cif, void **args, ffi_java_raw *raw);
-void ffi_java_raw_to_ptrarray (ffi_cif *cif, ffi_java_raw *raw, void **args);
+void ffi_java_ptrarray_to_raw (ffi_cif *cif, void **args, ffi_raw *raw);
+void ffi_java_raw_to_ptrarray (ffi_cif *cif, ffi_raw *raw, void **args);
 size_t ffi_java_raw_size (ffi_cif *cif);
 
 /* ---- Definitions for closures ----------------------------------------- */
 
 #if FFI_CLOSURES
 
-#ifdef _MSC_VER
-__declspec(align(8))
-#endif
 typedef struct {
   char tramp[FFI_TRAMPOLINE_SIZE];
   ffi_cif   *cif;
   void     (*fun)(ffi_cif*,void*,void**,void*);
   void      *user_data;
-#ifdef __GNUC__
-} ffi_closure __attribute__((aligned (8)));
-#else
 } ffi_closure;
-#endif
-
-void *ffi_closure_alloc (size_t size, void **code);
-void ffi_closure_free (void *);
 
 ffi_status
 ffi_prep_closure (ffi_closure*,
@@ -274,13 +227,6 @@
 		  void (*fun)(ffi_cif*,void*,void**,void*),
 		  void *user_data);
 
-ffi_status
-ffi_prep_closure_loc (ffi_closure*,
-		      ffi_cif *,
-		      void (*fun)(ffi_cif*,void*,void**,void*),
-		      void *user_data,
-		      void*codeloc);
-
 typedef struct {
   char tramp[FFI_TRAMPOLINE_SIZE];
 
@@ -302,27 +248,6 @@
 
 } ffi_raw_closure;
 
-typedef struct {
-  char tramp[FFI_TRAMPOLINE_SIZE];
-
-  ffi_cif   *cif;
-
-#if !FFI_NATIVE_RAW_API
-
-  /* if this is enabled, then a raw closure has the same layout 
-     as a regular closure.  We use this to install an intermediate 
-     handler to do the transaltion, void** -> ffi_raw*. */
-
-  void     (*translate_args)(ffi_cif*,void*,void**,void*);
-  void      *this_closure;
-
-#endif
-
-  void     (*fun)(ffi_cif*,void*,ffi_java_raw*,void*);
-  void      *user_data;
-
-} ffi_java_raw_closure;
-
 ffi_status
 ffi_prep_raw_closure (ffi_raw_closure*,
 		      ffi_cif *cif,
@@ -330,42 +255,29 @@
 		      void *user_data);
 
 ffi_status
-ffi_prep_raw_closure_loc (ffi_raw_closure*,
-			  ffi_cif *cif,
-			  void (*fun)(ffi_cif*,void*,ffi_raw*,void*),
-			  void *user_data,
-			  void *codeloc);
-
-ffi_status
-ffi_prep_java_raw_closure (ffi_java_raw_closure*,
+ffi_prep_java_raw_closure (ffi_raw_closure*,
 		           ffi_cif *cif,
-		           void (*fun)(ffi_cif*,void*,ffi_java_raw*,void*),
+		           void (*fun)(ffi_cif*,void*,ffi_raw*,void*),
 		           void *user_data);
 
-ffi_status
-ffi_prep_java_raw_closure_loc (ffi_java_raw_closure*,
-			       ffi_cif *cif,
-			       void (*fun)(ffi_cif*,void*,ffi_java_raw*,void*),
-			       void *user_data,
-			       void *codeloc);
-
 #endif /* FFI_CLOSURES */
 
 /* ---- Public interface definition -------------------------------------- */
 
-ffi_status ffi_prep_cif(ffi_cif *cif,
+ffi_status ffi_prep_cif(/*@out@*/ /*@partial@*/ ffi_cif *cif, 
 			ffi_abi abi,
-			unsigned int nargs,
-			ffi_type *rtype,
-			ffi_type **atypes);
-
-void ffi_call(ffi_cif *cif,
-	      void (*fn)(void),
-	      void *rvalue,
-	      void **avalue);
+			unsigned int nargs, 
+			/*@dependent@*/ /*@out@*/ /*@partial@*/ ffi_type *rtype, 
+			/*@dependent@*/ ffi_type **atypes);
+
+int
+ffi_call(/*@dependent@*/ ffi_cif *cif, 
+	 void (*fn)(), 
+	 /*@out@*/ void *rvalue, 
+	 /*@dependent@*/ void **avalue);
 
 /* Useful for eliminating compiler warnings */
-#define FFI_FN(f) ((void (*)(void))f)
+#define FFI_FN(f) ((void (*)())f)
 
 /* ---- Definitions shared with assembly code ---------------------------- */
 
@@ -376,7 +288,7 @@
 #define FFI_TYPE_INT        1
 #define FFI_TYPE_FLOAT      2    
 #define FFI_TYPE_DOUBLE     3
-#if HAVE_LONG_DOUBLE
+#if 1
 #define FFI_TYPE_LONGDOUBLE 4
 #else
 #define FFI_TYPE_LONGDOUBLE FFI_TYPE_DOUBLE
@@ -400,3 +312,4 @@
 #endif
 
 #endif
+

Modified: python/branches/py3k-dtoa/Modules/_ctypes/libffi_msvc/fficonfig.h
==============================================================================
--- python/branches/py3k-dtoa/Modules/_ctypes/libffi_msvc/fficonfig.h	(original)
+++ python/branches/py3k-dtoa/Modules/_ctypes/libffi_msvc/fficonfig.h	Tue Jul  6 17:55:27 2010
@@ -1,186 +1,96 @@
 /* fficonfig.h.  Originally created by configure, now hand_maintained for MSVC. */
 
-/* fficonfig.h.in.  Generated from configure.ac by autoheader.  */
+/* fficonfig.h.  Generated automatically by configure.  */
+/* fficonfig.h.in.  Generated automatically from configure.in by autoheader.  */
 
-/* Define if building universal (internal helper macro) */
-/* #undef AC_APPLE_UNIVERSAL_BUILD */
+/* Define this for MSVC, but not for mingw32! */
+#ifdef _MSC_VER
+#define __attribute__(x) /* */
+#endif
+#define alloca _alloca
 
-/* Define to one of `_getb67', `GETB67', `getb67' for Cray-2 and Cray-YMP
-   systems. This function is required for `alloca.c' support on those systems.
-   */
-/* #undef CRAY_STACKSEG_END */
+/*----------------------------------------------------------------*/
 
-/* Define to 1 if using `alloca.c'. */
+/* Define if using alloca.c.  */
 /* #undef C_ALLOCA */
 
-/* Define to the flags needed for the .section .eh_frame directive. */
-/* #undef EH_FRAME_FLAGS */
-
-/* Define this if you want extra debugging. */
-/* #undef FFI_DEBUG */
-
-/* Cannot use malloc on this target, so, we revert to alternative means */
-/* #undef FFI_MMAP_EXEC_WRIT */
-
-/* Define this is you do not want support for the raw API. */
-#define FFI_NO_RAW_API 1
-
-/* Define this is you do not want support for aggregate types. */
-/* #undef FFI_NO_STRUCTS */
+/* Define to one of _getb67, GETB67, getb67 for Cray-2 and Cray-YMP systems.
+   This function is required for alloca.c support on those systems.  */
+/* #undef CRAY_STACKSEG_END */
 
-/* Define to 1 if you have `alloca', as a function or macro. */
+/* Define if you have alloca, as a function or macro.  */
 #define HAVE_ALLOCA 1
 
-/* Define to 1 if you have <alloca.h> and it should be used (not on Ultrix).
-   */
-/* #undef HAVE_ALLOCA_H */
-
-/* Define if your assembler supports .cfi_* directives. */
-/* #undef HAVE_AS_CFI_PSEUDO_OP */
+/* Define if you have <alloca.h> and it should be used (not on Ultrix).  */
+/* #define HAVE_ALLOCA_H 1 */
 
-/* Define if your assembler supports .register. */
-/* #undef HAVE_AS_REGISTER_PSEUDO_OP */
-
-/* Define if your assembler and linker support unaligned PC relative relocs.
-   */
-/* #undef HAVE_AS_SPARC_UA_PCREL */
-
-/* Define if your assembler supports PC relative relocs. */
-/* #undef HAVE_AS_X86_PCREL */
-
-/* Define to 1 if you have the <dlfcn.h> header file. */
-/* #undef HAVE_DLFCN_H */
-
-/* Define if __attribute__((visibility("hidden"))) is supported. */
-/* #undef HAVE_HIDDEN_VISIBILITY_ATTRIBUTE */
-
-/* Define to 1 if you have the <inttypes.h> header file. */
-/* #undef HAVE_INTTYPES_H */
+/* If using the C implementation of alloca, define if you know the
+   direction of stack growth for your system; otherwise it will be
+   automatically deduced at run-time.
+ STACK_DIRECTION > 0 => grows toward higher addresses
+ STACK_DIRECTION < 0 => grows toward lower addresses
+ STACK_DIRECTION = 0 => direction of growth unknown
+ */
+/* #undef STACK_DIRECTION */
 
-/* Define if you have the long double type and it is bigger than a double */
-/* #undef HAVE_LONG_DOUBLE */
+/* Define if you have the ANSI C header files.  */
+#define STDC_HEADERS 1
 
-/* Define to 1 if you have the `memcpy' function. */
+/* Define if you have the memcpy function.  */
 #define HAVE_MEMCPY 1
 
-/* Define to 1 if you have the <memory.h> header file. */
-/* #undef HAVE_MEMORY_H */
-
-/* Define to 1 if you have the `mmap' function. */
-/* #undef HAVE_MMAP */
-
-/* Define if mmap with MAP_ANON(YMOUS) works. */
-/* #undef HAVE_MMAP_ANON */
-
-/* Define if mmap of /dev/zero works. */
-/* #undef HAVE_MMAP_DEV_ZERO */
-
 /* Define if read-only mmap of a plain file works. */
-/* #undef HAVE_MMAP_FILE */
-
-/* Define if .eh_frame sections should be read-only. */
-/* #undef HAVE_RO_EH_FRAME */
-
-/* Define to 1 if you have the <stdint.h> header file. */
-#undef HAVE_STDINT_H
-
-/* Define to 1 if you have the <stdlib.h> header file. */
-#undef HAVE_STDLIB_H
-
-/* Define to 1 if you have the <strings.h> header file. */
-#undef HAVE_STRINGS_H
-
-/* Define to 1 if you have the <string.h> header file. */
-#undef HAVE_STRING_H
+//#define HAVE_MMAP_FILE 1
 
-/* Define to 1 if you have the <sys/mman.h> header file. */
-/* #undef HAVE_SYS_MMAN_H */
-
-/* Define to 1 if you have the <sys/stat.h> header file. */
-/* #undef HAVE_SYS_STAT_H */
+/* Define if mmap of /dev/zero works. */
+//#define HAVE_MMAP_DEV_ZERO 1
 
-/* Define to 1 if you have the <sys/types.h> header file. */
-/* #undef HAVE_SYS_TYPES_H */
+/* Define if mmap with MAP_ANON(YMOUS) works. */
+//#define HAVE_MMAP_ANON 1
 
-/* Define to 1 if you have the <unistd.h> header file. */
-/* #undef HAVE_UNISTD_H */
+/* The number of bytes in type double */
+#define SIZEOF_DOUBLE 8
 
-/* Define to the sub-directory in which libtool stores uninstalled libraries.
-   */
-/* #undef LT_OBJDIR */
+/* The number of bytes in type long double */
+#define SIZEOF_LONG_DOUBLE 12
 
-/* Define to 1 if your C compiler doesn't accept -c and -o together. */
-/* #undef NO_MINUS_C_MINUS_O */
+/* Define if you have the long double type and it is bigger than a double */
+#define HAVE_LONG_DOUBLE 1
 
-/* Name of package */
-/* #undef PACKAGE */
+/* whether byteorder is bigendian */
+/* #undef WORDS_BIGENDIAN */
 
-/* Define to the address where bug reports for this package should be sent. */
-/* #undef PACKAGE_BUGREPORT */
+/* Define if the host machine stores words of multi-word integers in
+   big-endian order. */
+/* #undef HOST_WORDS_BIG_ENDIAN */
 
-/* Define to the full name of this package. */
-/* #undef PACKAGE_NAME */
+/* 1234 = LIL_ENDIAN, 4321 = BIGENDIAN */
+#define BYTEORDER 1234
 
-/* Define to the full name and version of this package. */
-/* #undef PACKAGE_STRING */
+/* Define if your assembler and linker support unaligned PC relative relocs. */
+/* #undef HAVE_AS_SPARC_UA_PCREL */
 
-/* Define to the one symbol short name of this package. */
-/* #undef PACKAGE_TARNAME */
+/* Define if your assembler supports .register. */
+/* #undef HAVE_AS_REGISTER_PSEUDO_OP */
 
-/* Define to the home page for this package. */
-/* #undef PACKAGE_URL */
+/* Define if .eh_frame sections should be read-only. */
+/* #undef HAVE_RO_EH_FRAME */
 
-/* Define to the version of this package. */
-/* #undef PACKAGE_VERSION */
+/* Define to the flags needed for the .section .eh_frame directive. */
+/* #define EH_FRAME_FLAGS "aw" */
 
-/* The size of `double', as computed by sizeof. */
-#define SIZEOF_DOUBLE 8
+/* Define to the flags needed for the .section .eh_frame directive. */
+/* #define EH_FRAME_FLAGS "aw" */
 
-/* The size of `long double', as computed by sizeof. */
-#undef SIZEOF_LONG_DOUBLE
+/* Define this if you want extra debugging. */
+/* #undef FFI_DEBUG */
 
-/* If using the C implementation of alloca, define if you know the
-   direction of stack growth for your system; otherwise it will be
-   automatically deduced at runtime.
-	STACK_DIRECTION > 0 => grows toward higher addresses
-	STACK_DIRECTION < 0 => grows toward lower addresses
-	STACK_DIRECTION = 0 => direction of growth unknown */
-/* #undef STACK_DIRECTION */
+/* Define this is you do not want support for aggregate types. */
+/* #undef FFI_NO_STRUCTS */
 
-/* Define to 1 if you have the ANSI C header files. */
-#define STDC_HEADERS 1
+/* Define this is you do not want support for the raw API. */
+/* #undef FFI_NO_RAW_API */
 
-/* Define this if you are using Purify and want to suppress spurious messages.
-   */
+/* Define this if you are using Purify and want to suppress spurious messages. */
 /* #undef USING_PURIFY */
 
-/* Version number of package */
-/* #undef VERSION */
-
-/* Define WORDS_BIGENDIAN to 1 if your processor stores words with the most
-   significant byte first (like Motorola and SPARC, unlike Intel). */
-#if defined AC_APPLE_UNIVERSAL_BUILD
-# if defined __BIG_ENDIAN__
-#  define WORDS_BIGENDIAN 1
-# endif
-#else
-# ifndef WORDS_BIGENDIAN
-#  undef WORDS_BIGENDIAN
-# endif
-#endif
-
-
-#ifdef HAVE_HIDDEN_VISIBILITY_ATTRIBUTE
-#ifdef LIBFFI_ASM
-#define FFI_HIDDEN(name) .hidden name
-#else
-#define FFI_HIDDEN __attribute__ ((visibility ("hidden")))
-#endif
-#else
-#ifdef LIBFFI_ASM
-#define FFI_HIDDEN(name)
-#else
-#define FFI_HIDDEN
-#endif
-#endif
-

Modified: python/branches/py3k-dtoa/Modules/_math.c
==============================================================================
--- python/branches/py3k-dtoa/Modules/_math.c	(original)
+++ python/branches/py3k-dtoa/Modules/_math.c	Tue Jul  6 17:55:27 2010
@@ -55,14 +55,15 @@
     else if (x >= two_pow_p28) {        /* x > 2**28 */
         if (Py_IS_INFINITY(x)) {
             return x+x;
-        } else {
-            return log(x)+ln2;                  /* acosh(huge)=log(2x) */
+        }
+        else {
+            return log(x)+ln2;          /* acosh(huge)=log(2x) */
         }
     }
     else if (x == 1.) {
-        return 0.0;                             /* acosh(1) = 0 */
+        return 0.0;                     /* acosh(1) = 0 */
     }
-    else if (x > 2.) {                          /* 2 < x < 2**28 */
+    else if (x > 2.) {                  /* 2 < x < 2**28 */
         double t = x*x;
         return log(2.0*x - 1.0 / (x + sqrt(t - 1.0)));
     }
@@ -94,7 +95,7 @@
         return x+x;
     }
     if (absx < two_pow_m28) {           /* |x| < 2**-28 */
-        return x;               /* return x inexact except 0 */
+        return x;                       /* return x inexact except 0 */
     }
     if (absx > two_pow_p28) {           /* |x| > 2**28 */
         w = log(absx)+ln2;
@@ -114,9 +115,9 @@
  * Method :
  *    1.Reduced x to positive by atanh(-x) = -atanh(x)
  *    2.For x>=0.5
- *                1           2x                          x
- *      atanh(x) = --- * log(1 + -------) = 0.5 * log1p(2 * --------)
- *                2          1 - x                    1 - x
+ *                  1              2x                          x
+ *      atanh(x) = --- * log(1 + -------) = 0.5 * log1p(2 * -------)
+ *                  2             1 - x                      1 - x
  *
  *      For x<0.5
  *      atanh(x) = 0.5*log1p(2x+2x*x/(1-x))
@@ -173,15 +174,15 @@
     */
 
     if (fabs(x) < 0.7) {
-    double u;
-    u = exp(x);
-    if (u == 1.0)
-        return x;
-    else
-        return (u - 1.0) * x / log(u);
+        double u;
+        u = exp(x);
+        if (u == 1.0)
+            return x;
+        else
+            return (u - 1.0) * x / log(u);
     }
     else
-    return exp(x) - 1.0;
+        return exp(x) - 1.0;
 }
 
 /* log1p(x) = log(1+x).  The log1p function is designed to avoid the
@@ -194,7 +195,7 @@
     /* For x small, we use the following approach.  Let y be the nearest float
        to 1+x, then
 
-      1+x = y * (1 - (y-1-x)/y)
+         1+x = y * (1 - (y-1-x)/y)
 
        so log(1+x) = log(y) + log(1-(y-1-x)/y).  Since (y-1-x)/y is tiny, the
        second term is well approximated by (y-1-x)/y.  If abs(x) >=
@@ -213,17 +214,19 @@
 
     double y;
     if (fabs(x) < DBL_EPSILON/2.) {
-    return x;
-    } else if (-0.5 <= x && x <= 1.) {
-    /* WARNING: it's possible than an overeager compiler
-       will incorrectly optimize the following two lines
-       to the equivalent of "return log(1.+x)". If this
-       happens, then results from log1p will be inaccurate
-       for small x. */
-    y = 1.+x;
-    return log(y)-((y-1.)-x)/y;
-    } else {
-    /* NaNs and infinities should end up here */
-    return log(1.+x);
+        return x;
+    }
+    else if (-0.5 <= x && x <= 1.) {
+        /* WARNING: it's possible than an overeager compiler
+           will incorrectly optimize the following two lines
+           to the equivalent of "return log(1.+x)". If this
+           happens, then results from log1p will be inaccurate
+           for small x. */
+        y = 1.+x;
+        return log(y)-((y-1.)-x)/y;
+    }
+    else {
+        /* NaNs and infinities should end up here */
+        return log(1.+x);
     }
 }

Modified: python/branches/py3k-dtoa/Modules/_scproxy.c
==============================================================================
--- python/branches/py3k-dtoa/Modules/_scproxy.c	(original)
+++ python/branches/py3k-dtoa/Modules/_scproxy.c	Tue Jul  6 17:55:27 2010
@@ -75,7 +75,7 @@
         aNum = CFDictionaryGetValue(proxyDict,
             kSCPropNetProxiesExcludeSimpleHostnames);
         if (aNum == NULL) {
-            v = PyBool_FromLong(1);
+            v = PyBool_FromLong(0);
         } else {
             v = PyBool_FromLong(cfnum_to_int32(aNum));
         }

Modified: python/branches/py3k-dtoa/Modules/_ssl.c
==============================================================================
--- python/branches/py3k-dtoa/Modules/_ssl.c	(original)
+++ python/branches/py3k-dtoa/Modules/_ssl.c	Tue Jul  6 17:55:27 2010
@@ -212,8 +212,11 @@
                     errstr = "EOF occurred in violation of protocol";
                 } else if (ret == -1) {
                     /* underlying BIO reported an I/O error */
+                    Py_INCREF(s);
                     ERR_clear_error();
-                    return s->errorhandler();
+                    v = s->errorhandler();
+                    Py_DECREF(s);
+                    return v;
                 } else { /* possible? */
                     p = PY_SSL_ERROR_SYSCALL;
                     errstr = "Some I/O error occurred";
@@ -334,6 +337,7 @@
                      PY_SSL_ERROR_NO_SOCKET, __FILE__, __LINE__);
         return NULL;
     }
+    Py_INCREF(sock);
 
     /* just in case the blocking state of the socket has been changed */
     nonblocking = (sock->sock_timeout >= 0.0);
@@ -348,9 +352,8 @@
         ret = SSL_do_handshake(self->ssl);
         err = SSL_get_error(self->ssl, ret);
         PySSL_END_ALLOW_THREADS
-        if(PyErr_CheckSignals()) {
-            return NULL;
-        }
+        if (PyErr_CheckSignals())
+            goto error;
         if (err == SSL_ERROR_WANT_READ) {
             sockstate = check_socket_and_wait_for_timeout(sock, 0);
         } else if (err == SSL_ERROR_WANT_WRITE) {
@@ -361,22 +364,22 @@
         if (sockstate == SOCKET_HAS_TIMED_OUT) {
             PyErr_SetString(PySSLErrorObject,
                             ERRSTR("The handshake operation timed out"));
-            return NULL;
+            goto error;
         } else if (sockstate == SOCKET_HAS_BEEN_CLOSED) {
             PyErr_SetString(PySSLErrorObject,
                             ERRSTR("Underlying socket has been closed."));
-            return NULL;
+            goto error;
         } else if (sockstate == SOCKET_TOO_LARGE_FOR_SELECT) {
             PyErr_SetString(PySSLErrorObject,
                             ERRSTR("Underlying socket too large for select()."));
-            return NULL;
+            goto error;
         } else if (sockstate == SOCKET_IS_NONBLOCKING) {
             break;
         }
     } while (err == SSL_ERROR_WANT_READ || err == SSL_ERROR_WANT_WRITE);
+    Py_DECREF(sock);
     if (ret < 1)
         return PySSL_SetError(self, ret, __FILE__, __LINE__);
-    self->ssl->debug = 1;
 
     if (self->peer_cert)
         X509_free (self->peer_cert);
@@ -386,6 +389,10 @@
 
     Py_INCREF(Py_None);
     return Py_None;
+
+error:
+    Py_DECREF(sock);
+    return NULL;
 }
 
 static PyObject *
@@ -1051,9 +1058,12 @@
                      PY_SSL_ERROR_NO_SOCKET, __FILE__, __LINE__);
         return NULL;
     }
+    Py_INCREF(sock);
 
-    if (!PyArg_ParseTuple(args, "y*:write", &buf))
+    if (!PyArg_ParseTuple(args, "y*:write", &buf)) {
+        Py_DECREF(sock);
         return NULL;
+    }
 
     /* just in case the blocking state of the socket has been changed */
     nonblocking = (sock->sock_timeout >= 0.0);
@@ -1103,6 +1113,7 @@
         }
     } while (err == SSL_ERROR_WANT_READ || err == SSL_ERROR_WANT_WRITE);
 
+    Py_DECREF(sock);
     PyBuffer_Release(&buf);
     if (len > 0)
         return PyLong_FromLong(len);
@@ -1110,6 +1121,7 @@
         return PySSL_SetError(self, len, __FILE__, __LINE__);
 
 error:
+    Py_DECREF(sock);
     PyBuffer_Release(&buf);
     return NULL;
 }
@@ -1159,21 +1171,23 @@
                      PY_SSL_ERROR_NO_SOCKET, __FILE__, __LINE__);
         return NULL;
     }
+    Py_INCREF(sock);
 
     if (!PyArg_ParseTuple(args, "|Oi:read", &dest, &count))
-        return NULL;
+        goto error;
+
     if ((dest == NULL) || (dest == Py_None)) {
         if (!(dest = PyByteArray_FromStringAndSize((char *) 0, len)))
-            return NULL;
+            goto error;
         mem = PyByteArray_AS_STRING(dest);
     } else if (PyLong_Check(dest)) {
         len = PyLong_AS_LONG(dest);
         if (!(dest = PyByteArray_FromStringAndSize((char *) 0, len)))
-            return NULL;
+            goto error;
         mem = PyByteArray_AS_STRING(dest);
     } else {
         if (PyObject_GetBuffer(dest, &buf, PyBUF_CONTIG) < 0)
-            return NULL;
+            goto error;
         mem = buf.buf;
         len = buf.len;
         if ((count > 0) && (count <= len))
@@ -1240,6 +1254,7 @@
         goto error;
     }
   done:
+    Py_DECREF(sock);
     if (!buf_passed) {
         PyObject *res = PyBytes_FromStringAndSize(mem, count);
         Py_DECREF(dest);
@@ -1249,8 +1264,9 @@
         return PyLong_FromLong(count);
     }
   error:
+    Py_DECREF(sock);
     if (!buf_passed) {
-        Py_DECREF(dest);
+        Py_XDECREF(dest);
     } else {
         PyBuffer_Release(&buf);
     }
@@ -1275,6 +1291,7 @@
                      PY_SSL_ERROR_NO_SOCKET, __FILE__, __LINE__);
         return NULL;
     }
+    Py_INCREF(sock);
 
     /* Just in case the blocking state of the socket has been changed */
     nonblocking = (sock->sock_timeout >= 0.0);
@@ -1324,24 +1341,29 @@
             else
                 PyErr_SetString(PySSLErrorObject,
                                 "The write operation timed out");
-            return NULL;
+            goto error;
         }
         else if (sockstate == SOCKET_TOO_LARGE_FOR_SELECT) {
             PyErr_SetString(PySSLErrorObject,
                             "Underlying socket too large for select().");
-            return NULL;
+            goto error;
         }
         else if (sockstate != SOCKET_OPERATION_OK)
             /* Retain the SSL error code */
             break;
     }
 
-    if (err < 0)
+    if (err < 0) {
+        Py_DECREF(sock);
         return PySSL_SetError(self, err, __FILE__, __LINE__);
-    else {
-        Py_INCREF(sock);
-        return (PyObject *) sock;
     }
+    else
+        /* It's already INCREF'ed */
+        return (PyObject *) sock;
+
+error:
+    Py_DECREF(sock);
+    return NULL;
 }
 
 PyDoc_STRVAR(PySSL_SSLshutdown_doc,

Modified: python/branches/py3k-dtoa/Modules/_testcapimodule.c
==============================================================================
--- python/branches/py3k-dtoa/Modules/_testcapimodule.c	(original)
+++ python/branches/py3k-dtoa/Modules/_testcapimodule.c	Tue Jul  6 17:55:27 2010
@@ -1386,6 +1386,28 @@
 }
 
 static PyObject *
+getargs_w_star(PyObject *self, PyObject *args)
+{
+    Py_buffer buffer;
+    PyObject *result;
+    char *str;
+
+    if (!PyArg_ParseTuple(args, "w*:getargs_w_star", &buffer))
+        return NULL;
+
+    if (2 <= buffer.len) {
+        str = buffer.buf;
+        str[0] = '[';
+        str[buffer.len-1] = ']';
+    }
+
+    result = PyBytes_FromStringAndSize(buffer.buf, buffer.len);
+    PyBuffer_Release(&buffer);
+    return result;
+}
+
+
+static PyObject *
 test_empty_argparse(PyObject *self)
 {
     /* Test that formats can begin with '|'. See issue #4720. */
@@ -2227,6 +2249,7 @@
     {"getargs_u_hash",          getargs_u_hash,                  METH_VARARGS},
     {"getargs_Z",               getargs_Z,                       METH_VARARGS},
     {"getargs_Z_hash",          getargs_Z_hash,                  METH_VARARGS},
+    {"getargs_w_star",          getargs_w_star,                  METH_VARARGS},
     {"codec_incrementalencoder",
      (PyCFunction)codec_incrementalencoder,      METH_VARARGS},
     {"codec_incrementaldecoder",

Modified: python/branches/py3k-dtoa/Modules/audioop.c
==============================================================================
--- python/branches/py3k-dtoa/Modules/audioop.c	(original)
+++ python/branches/py3k-dtoa/Modules/audioop.c	Tue Jul  6 17:55:27 2010
@@ -297,6 +297,29 @@
 
 static PyObject *AudioopError;
 
+static int
+audioop_check_size(int size)
+{
+    if (size != 1 && size != 2 && size != 4) {
+        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
+        return 0;
+    }
+    else
+        return 1;
+}
+
+static int
+audioop_check_parameters(int len, int size)
+{
+    if (!audioop_check_size(size))
+        return 0;
+    if (len % size != 0) {
+        PyErr_SetString(AudioopError, "not a whole number of frames");
+        return 0;
+    }
+    return 1;
+}
+
 static PyObject *
 audioop_getsample(PyObject *self, PyObject *args)
 {
@@ -306,10 +329,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#in:getsample", &cp, &len, &size, &i) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     if ( i < 0 || i >= len/size ) {
         PyErr_SetString(AudioopError, "Index out of range");
         return 0;
@@ -330,10 +351,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:max", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     for ( i=0; i<len; i+= size) {
         if ( size == 1 )      val = (int)*CHARP(cp, i);
         else if ( size == 2 ) val = (int)*SHORTP(cp, i);
@@ -354,10 +373,8 @@
 
     if (!PyArg_ParseTuple(args, "s#i:minmax", &cp, &len, &size))
         return NULL;
-    if (size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
+    if (!audioop_check_parameters(len, size))
         return NULL;
-    }
     for (i = 0; i < len; i += size) {
         if (size == 1) val = (int) *CHARP(cp, i);
         else if (size == 2) val = (int) *SHORTP(cp, i);
@@ -378,10 +395,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:avg", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     for ( i=0; i<len; i+= size) {
         if ( size == 1 )      val = (int)*CHARP(cp, i);
         else if ( size == 2 ) val = (int)*SHORTP(cp, i);
@@ -405,10 +420,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:rms", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     for ( i=0; i<len; i+= size) {
         if ( size == 1 )      val = (int)*CHARP(cp, i);
         else if ( size == 2 ) val = (int)*SHORTP(cp, i);
@@ -616,10 +629,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:avgpp", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     /* Compute first delta value ahead. Also automatically makes us
     ** skip the first extreme value
     */
@@ -673,10 +684,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:maxpp", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     /* Compute first delta value ahead. Also automatically makes us
     ** skip the first extreme value
     */
@@ -725,10 +734,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:cross", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     ncross = -1;
     prevval = 17; /* Anything <> 0,1 */
     for ( i=0; i<len; i+= size) {
@@ -753,6 +760,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#id:mul", &cp, &len, &size, &factor ) )
         return 0;
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     if ( size == 1 ) maxval = (double) 0x7f;
     else if ( size == 2 ) maxval = (double) 0x7fff;
@@ -798,6 +807,15 @@
         return 0;
     cp = pcp.buf;
     len = pcp.len;
+    if (!audioop_check_parameters(len, size)) {
+        PyBuffer_Release(&pcp);
+        return NULL;
+    }
+    if (((len / size) & 1) != 0) {
+        PyErr_SetString(AudioopError, "not a whole number of frames");
+        PyBuffer_Release(&pcp);
+        return NULL;
+    }
 
     if ( size == 1 ) maxval = (double) 0x7f;
     else if ( size == 2 ) maxval = (double) 0x7fff;
@@ -809,8 +827,10 @@
     }
 
     rv = PyBytes_FromStringAndSize(NULL, len/2);
-    if ( rv == 0 )
+    if ( rv == 0 ) {
+        PyBuffer_Release(&pcp);
         return 0;
+    }
     ncp = (signed char *)PyBytes_AsString(rv);
 
 
@@ -845,6 +865,8 @@
     if ( !PyArg_ParseTuple(args, "s#idd:tostereo",
                            &cp, &len, &size, &fac1, &fac2 ) )
         return 0;
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     if ( size == 1 ) maxval = (double) 0x7f;
     else if ( size == 2 ) maxval = (double) 0x7fff;
@@ -903,7 +925,8 @@
     if ( !PyArg_ParseTuple(args, "s#s#i:add",
                       &cp1, &len1, &cp2, &len2, &size ) )
         return 0;
-
+    if (!audioop_check_parameters(len1, size))
+        return NULL;
     if ( len1 != len2 ) {
         PyErr_SetString(AudioopError, "Lengths should be the same");
         return 0;
@@ -958,10 +981,8 @@
                       &cp, &len, &size , &bias) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     rv = PyBytes_FromStringAndSize(NULL, len);
     if ( rv == 0 )
@@ -994,10 +1015,8 @@
                       &cp, &len, &size) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     rv = PyBytes_FromStringAndSize(NULL, len);
     if ( rv == 0 )
@@ -1031,11 +1050,10 @@
                       &cp, &len, &size, &size2) )
         return 0;
 
-    if ( (size != 1 && size != 2 && size != 4) ||
-         (size2 != 1 && size2 != 2 && size2 != 4)) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
+    if (!audioop_check_size(size2))
+        return NULL;
 
     if (len/size > PY_SSIZE_T_MAX/size2) {
         PyErr_SetString(PyExc_MemoryError,
@@ -1086,10 +1104,8 @@
                           &nchannels, &inrate, &outrate, &state,
                           &weightA, &weightB))
         return NULL;
-    if (size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
+    if (!audioop_check_size(size))
         return NULL;
-    }
     if (nchannels < 1) {
         PyErr_SetString(AudioopError, "# of channels should be >= 1");
         return NULL;
@@ -1265,10 +1281,8 @@
                            &cp, &len, &size) )
         return 0 ;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     rv = PyBytes_FromStringAndSize(NULL, len/size);
     if ( rv == 0 )
@@ -1299,10 +1313,8 @@
                            &cp, &len, &size) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     if (len > PY_SSIZE_T_MAX/size) {
         PyErr_SetString(PyExc_MemoryError,
@@ -1338,10 +1350,8 @@
                            &cp, &len, &size) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     rv = PyBytes_FromStringAndSize(NULL, len/size);
     if ( rv == 0 )
@@ -1372,10 +1382,8 @@
                            &cp, &len, &size) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     if (len > PY_SSIZE_T_MAX/size) {
         PyErr_SetString(PyExc_MemoryError,
@@ -1413,11 +1421,8 @@
                            &cp, &len, &size, &state) )
         return 0;
 
-
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     str = PyBytes_FromStringAndSize(NULL, len/(size*2));
     if ( str == 0 )
@@ -1522,10 +1527,8 @@
                            &cp, &len, &size, &state) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     /* Decode state, should have (value, step) */
     if ( state == Py_None ) {

Modified: python/branches/py3k-dtoa/Modules/datetimemodule.c
==============================================================================
--- python/branches/py3k-dtoa/Modules/datetimemodule.c	(original)
+++ python/branches/py3k-dtoa/Modules/datetimemodule.c	Tue Jul  6 17:55:27 2010
@@ -1023,11 +1023,6 @@
             Py_DECREF(result);
             result = NULL;
         }
-        else if (!PyUnicode_Check(result)) {
-            PyObject *temp = PyUnicode_FromObject(result);
-            Py_DECREF(result);
-            result = temp;
-        }
     }
     return result;
 }
@@ -3469,6 +3464,14 @@
     return add_datetime_timedelta(dt, (PyDateTime_Delta *)self->offset, 1);
 }
 
+static PyObject *
+timezone_getinitargs(PyDateTime_TimeZone *self)
+{
+    if (self->name == NULL)
+        return Py_BuildValue("(O)", self->offset);
+    return Py_BuildValue("(OO)", self->offset, self->name);
+}
+
 static PyMethodDef timezone_methods[] = {
     {"tzname", (PyCFunction)timezone_tzname, METH_O,
      PyDoc_STR("If name is specified when timezone is created, returns the name."
@@ -3483,6 +3486,9 @@
     {"fromutc", (PyCFunction)timezone_fromutc, METH_O,
      PyDoc_STR("datetime in UTC -> datetime in local time.")},
 
+    {"__getinitargs__", (PyCFunction)timezone_getinitargs, METH_NOARGS,
+     PyDoc_STR("pickle support")},
+
     {NULL, NULL}
 };
 
@@ -4362,82 +4368,23 @@
     return result;
 }
 
-/* Return new datetime from time.strptime(). */
+/* Return new datetime from _strptime.strptime_datetime(). */
 static PyObject *
 datetime_strptime(PyObject *cls, PyObject *args)
 {
     static PyObject *module = NULL;
-    PyObject *result = NULL, *obj, *st = NULL, *frac = NULL;
     const Py_UNICODE *string, *format;
 
     if (!PyArg_ParseTuple(args, "uu:strptime", &string, &format))
         return NULL;
 
-    if (module == NULL &&
-        (module = PyImport_ImportModuleNoBlock("_strptime")) == NULL)
-        return NULL;
-
-    /* _strptime._strptime returns a two-element tuple.  The first
-       element is a time.struct_time object.  The second is the
-       microseconds (which are not defined for time.struct_time). */
-    obj = PyObject_CallMethod(module, "_strptime", "uu", string, format);
-    if (obj != NULL) {
-        int i, good_timetuple = 1;
-        long int ia[7];
-        if (PySequence_Check(obj) && PySequence_Size(obj) == 2) {
-            st = PySequence_GetItem(obj, 0);
-            frac = PySequence_GetItem(obj, 1);
-            if (st == NULL || frac == NULL)
-                good_timetuple = 0;
-            /* copy y/m/d/h/m/s values out of the
-               time.struct_time */
-            if (good_timetuple &&
-                PySequence_Check(st) &&
-                PySequence_Size(st) >= 6) {
-                for (i=0; i < 6; i++) {
-                    PyObject *p = PySequence_GetItem(st, i);
-                    if (p == NULL) {
-                        good_timetuple = 0;
-                        break;
-                    }
-                    if (PyLong_Check(p))
-                        ia[i] = PyLong_AsLong(p);
-                    else
-                        good_timetuple = 0;
-                    Py_DECREF(p);
-                }
-/*                              if (PyLong_CheckExact(p)) {
-                                        ia[i] = PyLong_AsLongAndOverflow(p, &overflow);
-                                        if (overflow)
-                                                good_timetuple = 0;
-                                }
-                                else
-                                        good_timetuple = 0;
-                                Py_DECREF(p);
-*/                      }
-                        else
-                                good_timetuple = 0;
-                        /* follow that up with a little dose of microseconds */
-            if (PyLong_Check(frac))
-                ia[6] = PyLong_AsLong(frac);
-            else
-                good_timetuple = 0;
-        }
-        else
-            good_timetuple = 0;
-        if (good_timetuple)
-            result = PyObject_CallFunction(cls, "iiiiiii",
-                                           ia[0], ia[1], ia[2],
-                                           ia[3], ia[4], ia[5],
-                                           ia[6]);
-        else
-            PyErr_SetString(PyExc_ValueError,
-                "unexpected value from _strptime._strptime");
+    if (module == NULL) {
+        module = PyImport_ImportModuleNoBlock("_strptime");
+        if (module == NULL)
+            return NULL;
     }
-    Py_XDECREF(obj);
-    Py_XDECREF(st);
-    Py_XDECREF(frac);
-    return result;
+    return PyObject_CallMethod(module, "_strptime_datetime", "Ouu",
+                               cls, string, format);
 }
 
 /* Return new datetime from date/datetime and time arguments. */
@@ -4991,15 +4938,9 @@
 
         mm -= offset;
         stat = normalize_datetime(&y, &m, &d, &hh, &mm, &ss, &us);
-        if (stat < 0) {
-            /* At the edges, it's possible we overflowed
-             * beyond MINYEAR or MAXYEAR.
-             */
-            if (PyErr_ExceptionMatches(PyExc_OverflowError))
-                PyErr_Clear();
-            else
-                return NULL;
-        }
+        /* OverflowError may be raised in the edge cases. */
+        if (stat < 0)
+            return NULL;
     }
     return build_struct_time(y, m, d, hh, mm, ss, 0);
 }

Modified: python/branches/py3k-dtoa/Modules/mathmodule.c
==============================================================================
--- python/branches/py3k-dtoa/Modules/mathmodule.c	(original)
+++ python/branches/py3k-dtoa/Modules/mathmodule.c	Tue Jul  6 17:55:27 2010
@@ -841,19 +841,17 @@
 
 static PyObject * math_ceil(PyObject *self, PyObject *number) {
     static PyObject *ceil_str = NULL;
-    PyObject *method;
+    PyObject *method, *result;
 
-    if (ceil_str == NULL) {
-        ceil_str = PyUnicode_InternFromString("__ceil__");
-        if (ceil_str == NULL)
+    method = _PyObject_LookupSpecial(number, "__ceil__", &ceil_str);
+    if (method == NULL) {
+        if (PyErr_Occurred())
             return NULL;
-    }
-
-    method = _PyType_Lookup(Py_TYPE(number), ceil_str);
-    if (method == NULL)
         return math_1_to_int(number, ceil, 0);
-    else
-        return PyObject_CallFunction(method, "O", number);
+    }
+    result = PyObject_CallFunctionObjArgs(method, NULL);
+    Py_DECREF(method);
+    return result;
 }
 
 PyDoc_STRVAR(math_ceil_doc,
@@ -881,19 +879,17 @@
 
 static PyObject * math_floor(PyObject *self, PyObject *number) {
     static PyObject *floor_str = NULL;
-    PyObject *method;
+    PyObject *method, *result;
 
-    if (floor_str == NULL) {
-        floor_str = PyUnicode_InternFromString("__floor__");
-        if (floor_str == NULL)
+    method = _PyObject_LookupSpecial(number, "__floor__", &floor_str);
+    if (method == NULL) {
+        if (PyErr_Occurred())
             return NULL;
-    }
-
-    method = _PyType_Lookup(Py_TYPE(number), floor_str);
-    if (method == NULL)
         return math_1_to_int(number, floor, 0);
-    else
-        return PyObject_CallFunction(method, "O", number);
+    }
+    result = PyObject_CallFunctionObjArgs(method, NULL);
+    Py_DECREF(method);
+    return result;
 }
 
 PyDoc_STRVAR(math_floor_doc,
@@ -989,17 +985,17 @@
    def msum(iterable):
        partials = []  # sorted, non-overlapping partial sums
        for x in iterable:
-       i = 0
-       for y in partials:
-           if abs(x) < abs(y):
-           x, y = y, x
-           hi = x + y
-           lo = y - (hi - x)
-           if lo:
-           partials[i] = lo
-           i += 1
-           x = hi
-       partials[i:] = [x]
+           i = 0
+           for y in partials:
+               if abs(x) < abs(y):
+                   x, y = y, x
+               hi = x + y
+               lo = y - (hi - x)
+               if lo:
+                   partials[i] = lo
+                   i += 1
+               x = hi
+           partials[i:] = [x]
        return sum_exact(partials)
 
    Rounded x+y stored in hi with the roundoff stored in lo.  Together hi+lo
@@ -1420,27 +1416,24 @@
 math_trunc(PyObject *self, PyObject *number)
 {
     static PyObject *trunc_str = NULL;
-    PyObject *trunc;
+    PyObject *trunc, *result;
 
     if (Py_TYPE(number)->tp_dict == NULL) {
         if (PyType_Ready(Py_TYPE(number)) < 0)
             return NULL;
     }
 
-    if (trunc_str == NULL) {
-        trunc_str = PyUnicode_InternFromString("__trunc__");
-        if (trunc_str == NULL)
-            return NULL;
-    }
-
-    trunc = _PyType_Lookup(Py_TYPE(number), trunc_str);
+    trunc = _PyObject_LookupSpecial(number, "__trunc__", &trunc_str);
     if (trunc == NULL) {
-        PyErr_Format(PyExc_TypeError,
-                     "type %.100s doesn't define __trunc__ method",
-                     Py_TYPE(number)->tp_name);
+        if (!PyErr_Occurred())
+            PyErr_Format(PyExc_TypeError,
+                         "type %.100s doesn't define __trunc__ method",
+                         Py_TYPE(number)->tp_name);
         return NULL;
     }
-    return PyObject_CallFunctionObjArgs(trunc, number, NULL);
+    result = PyObject_CallFunctionObjArgs(trunc, NULL);
+    Py_DECREF(trunc);
+    return result;
 }
 
 PyDoc_STRVAR(math_trunc_doc,

Modified: python/branches/py3k-dtoa/Modules/parsermodule.c
==============================================================================
--- python/branches/py3k-dtoa/Modules/parsermodule.c	(original)
+++ python/branches/py3k-dtoa/Modules/parsermodule.c	Tue Jul  6 17:55:27 2010
@@ -1754,21 +1754,21 @@
                 && validate_dotted_as_names(CHILD(tree, 1)));
 }
 
-/* Helper function to count the number of leading dots in
+/* Helper function to count the number of leading dots (or ellipsis tokens) in
  * 'from ...module import name'
  */
 static int
 count_from_dots(node *tree)
 {
-        int i;
-        for (i = 1; i < NCH(tree); i++)
-                if (TYPE(CHILD(tree, i)) != DOT)
-                        break;
-        return i-1;
+    int i;
+    for (i = 1; i < NCH(tree); i++)
+        if (TYPE(CHILD(tree, i)) != DOT && TYPE(CHILD(tree, i)) != ELLIPSIS)
+            break;
+    return i - 1;
 }
 
-/* 'from' ('.'* dotted_name | '.') 'import' ('*' | '(' import_as_names ')' |
- *     import_as_names
+/* import_from: ('from' ('.'* dotted_name | '.'+)
+ *               'import' ('*' | '(' import_as_names ')' | import_as_names))
  */
 static int
 validate_import_from(node *tree)
@@ -1778,7 +1778,8 @@
         int havename = (TYPE(CHILD(tree, ndots + 1)) == dotted_name);
         int offset = ndots + havename;
         int res = validate_ntype(tree, import_from)
-                && (nch >= 4 + ndots)
+                && (offset >= 1)
+                && (nch >= 3 + offset)
                 && validate_name(CHILD(tree, 0), "from")
                 && (!havename || validate_dotted_name(CHILD(tree, ndots + 1)))
                 && validate_name(CHILD(tree, offset + 1), "import");
@@ -2509,14 +2510,15 @@
 static int
 validate_decorated(node *tree)
 {
-  int nch = NCH(tree);
-  int ok = (validate_ntype(tree, decorated)
-            && (nch == 2)
-            && validate_decorators(RCHILD(tree, -2))
-            && (validate_funcdef(RCHILD(tree, -1))
-                || validate_class(RCHILD(tree, -1)))
-            );
-  return ok;
+    int nch = NCH(tree);
+    int ok = (validate_ntype(tree, decorated)
+              && (nch == 2)
+              && validate_decorators(RCHILD(tree, -2)));
+    if (TYPE(RCHILD(tree, -1)) == funcdef)
+        ok = ok && validate_funcdef(RCHILD(tree, -1));
+    else
+        ok = ok && validate_class(RCHILD(tree, -1));
+    return ok;
 }
 
 static int

Modified: python/branches/py3k-dtoa/Modules/posixmodule.c
==============================================================================
--- python/branches/py3k-dtoa/Modules/posixmodule.c	(original)
+++ python/branches/py3k-dtoa/Modules/posixmodule.c	Tue Jul  6 17:55:27 2010
@@ -3925,7 +3925,7 @@
 #ifdef HAVE_SETPGRP
 PyDoc_STRVAR(posix_setpgrp__doc__,
 "setpgrp()\n\n\
-Make this process a session leader.");
+Make this process the process group leader.");
 
 static PyObject *
 posix_setpgrp(PyObject *self, PyObject *noargs)

Modified: python/branches/py3k-dtoa/Modules/selectmodule.c
==============================================================================
--- python/branches/py3k-dtoa/Modules/selectmodule.c	(original)
+++ python/branches/py3k-dtoa/Modules/selectmodule.c	Tue Jul  6 17:55:27 2010
@@ -91,7 +91,7 @@
     fd2obj[0].obj = (PyObject*)0;            /* set list to zero size */
     FD_ZERO(set);
 
-    fast_seq=PySequence_Fast(seq, "arguments 1-3 must be sequences");
+    fast_seq = PySequence_Fast(seq, "arguments 1-3 must be sequences");
     if (!fast_seq)
         return -1;
 

Modified: python/branches/py3k-dtoa/Modules/timemodule.c
==============================================================================
--- python/branches/py3k-dtoa/Modules/timemodule.c	(original)
+++ python/branches/py3k-dtoa/Modules/timemodule.c	Tue Jul  6 17:55:27 2010
@@ -90,32 +90,6 @@
 /* For Y2K check */
 static PyObject *moddict;
 
-/* Exposed in timefuncs.h. */
-time_t
-_PyTime_DoubleToTimet(double x)
-{
-    time_t result;
-    double diff;
-
-    result = (time_t)x;
-    /* How much info did we lose?  time_t may be an integral or
-     * floating type, and we don't know which.  If it's integral,
-     * we don't know whether C truncates, rounds, returns the floor,
-     * etc.  If we lost a second or more, the C rounding is
-     * unreasonable, or the input just doesn't fit in a time_t;
-     * call it an error regardless.  Note that the original cast to
-     * time_t can cause a C error too, but nothing we can do to
-     * worm around that.
-     */
-    diff = x - (double)result;
-    if (diff <= -1.0 || diff >= 1.0) {
-        PyErr_SetString(PyExc_ValueError,
-                        "timestamp out of range for platform time_t");
-        result = (time_t)-1;
-    }
-    return result;
-}
-
 static PyObject *
 time_time(PyObject *self, PyObject *unused)
 {

Modified: python/branches/py3k-dtoa/Modules/xxmodule.c
==============================================================================
--- python/branches/py3k-dtoa/Modules/xxmodule.c	(original)
+++ python/branches/py3k-dtoa/Modules/xxmodule.c	Tue Jul  6 17:55:27 2010
@@ -10,7 +10,7 @@
    your own types of attributes instead.  Maybe you want to name your
    local variables other than 'self'.  If your object type is needed in
    other files, you'll have to create a file "foobarobject.h"; see
-   intobject.h for an example. */
+   floatobject.h for an example. */
 
 /* Xxo objects */
 

Modified: python/branches/py3k-dtoa/Objects/abstract.c
==============================================================================
--- python/branches/py3k-dtoa/Objects/abstract.c	(original)
+++ python/branches/py3k-dtoa/Objects/abstract.c	Tue Jul  6 17:55:27 2010
@@ -2522,8 +2522,7 @@
     }
     else {
         if (!check_class(cls,
-            "isinstance() arg 2 must be a class, type,"
-            " or tuple of classes and types"))
+            "isinstance() arg 2 must be a type or tuple of types"))
             return -1;
         icls = PyObject_GetAttr(inst, __class__);
         if (icls == NULL) {

Modified: python/branches/py3k-dtoa/Objects/bytes_methods.c
==============================================================================
--- python/branches/py3k-dtoa/Objects/bytes_methods.c	(original)
+++ python/branches/py3k-dtoa/Objects/bytes_methods.c	Tue Jul  6 17:55:27 2010
@@ -332,7 +332,8 @@
 PyDoc_STRVAR_shared(_Py_capitalize__doc__,
 "B.capitalize() -> copy of B\n\
 \n\
-Return a copy of B with only its first character capitalized (ASCII).");
+Return a copy of B with only its first character capitalized (ASCII)\n\
+and the rest lower-cased.");
 
 void
 _Py_bytes_capitalize(char *result, char *s, Py_ssize_t len)

Modified: python/branches/py3k-dtoa/Objects/descrobject.c
==============================================================================
--- python/branches/py3k-dtoa/Objects/descrobject.c	(original)
+++ python/branches/py3k-dtoa/Objects/descrobject.c	Tue Jul  6 17:55:27 2010
@@ -1115,7 +1115,7 @@
 } propertyobject;
 
 static PyObject * property_copy(PyObject *, PyObject *, PyObject *,
-                                  PyObject *, PyObject *);
+                                  PyObject *);
 
 static PyMemberDef property_members[] = {
     {"fget", T_OBJECT, offsetof(propertyobject, prop_get), READONLY},
@@ -1132,7 +1132,7 @@
 static PyObject *
 property_getter(PyObject *self, PyObject *getter)
 {
-    return property_copy(self, getter, NULL, NULL, NULL);
+    return property_copy(self, getter, NULL, NULL);
 }
 
 
@@ -1142,7 +1142,7 @@
 static PyObject *
 property_setter(PyObject *self, PyObject *setter)
 {
-    return property_copy(self, NULL, setter, NULL, NULL);
+    return property_copy(self, NULL, setter, NULL);
 }
 
 
@@ -1152,7 +1152,7 @@
 static PyObject *
 property_deleter(PyObject *self, PyObject *deleter)
 {
-    return property_copy(self, NULL, NULL, deleter, NULL);
+    return property_copy(self, NULL, NULL, deleter);
 }
 
 
@@ -1221,11 +1221,10 @@
 }
 
 static PyObject *
-property_copy(PyObject *old, PyObject *get, PyObject *set, PyObject *del,
-                PyObject *doc)
+property_copy(PyObject *old, PyObject *get, PyObject *set, PyObject *del)
 {
     propertyobject *pold = (propertyobject *)old;
-    PyObject *new, *type;
+    PyObject *new, *type, *doc;
 
     type = PyObject_Type(old);
     if (type == NULL)
@@ -1243,15 +1242,12 @@
         Py_XDECREF(del);
         del = pold->prop_del ? pold->prop_del : Py_None;
     }
-    if (doc == NULL || doc == Py_None) {
-        Py_XDECREF(doc);
-        if (pold->getter_doc && get != Py_None) {
-            /* make _init use __doc__ from getter */
-            doc = Py_None;
-        }
-        else {
-            doc = pold->prop_doc ? pold->prop_doc : Py_None;
-        }
+    if (pold->getter_doc && get != Py_None) {
+        /* make _init use __doc__ from getter */
+        doc = Py_None;
+    }
+    else {
+        doc = pold->prop_doc ? pold->prop_doc : Py_None;
     }
 
     new =  PyObject_CallFunction(type, "OOOO", get, set, del, doc);

Modified: python/branches/py3k-dtoa/Objects/enumobject.c
==============================================================================
--- python/branches/py3k-dtoa/Objects/enumobject.c	(original)
+++ python/branches/py3k-dtoa/Objects/enumobject.c	Tue Jul  6 17:55:27 2010
@@ -159,12 +159,13 @@
 }
 
 PyDoc_STRVAR(enum_doc,
-"enumerate(iterable) -> iterator for index, value of iterable\n"
+"enumerate(iterable[, start]) -> iterator for index, value of iterable\n"
 "\n"
 "Return an enumerate object.  iterable must be another object that supports\n"
 "iteration.  The enumerate object yields pairs containing a count (from\n"
-"zero) and a value yielded by the iterable argument.  enumerate is useful\n"
-"for obtaining an indexed list: (0, seq[0]), (1, seq[1]), (2, seq[2]), ...");
+"start, which defaults to zero) and a value yielded by the iterable argument.\n"
+"enumerate is useful for obtaining an indexed list:\n"
+"    (0, seq[0]), (1, seq[1]), (2, seq[2]), ...");
 
 PyTypeObject PyEnum_Type = {
     PyVarObject_HEAD_INIT(&PyType_Type, 0)

Modified: python/branches/py3k-dtoa/Objects/exceptions.c
==============================================================================
--- python/branches/py3k-dtoa/Objects/exceptions.c	(original)
+++ python/branches/py3k-dtoa/Objects/exceptions.c	Tue Jul  6 17:55:27 2010
@@ -1622,9 +1622,6 @@
     const char *encoding, const char *object, Py_ssize_t length,
     Py_ssize_t start, Py_ssize_t end, const char *reason)
 {
-    assert(length < INT_MAX);
-    assert(start < INT_MAX);
-    assert(end < INT_MAX);
     return PyObject_CallFunction(PyExc_UnicodeDecodeError, "sy#nns",
                                  encoding, object, length, start, end, reason);
 }

Modified: python/branches/py3k-dtoa/Objects/floatobject.c
==============================================================================
--- python/branches/py3k-dtoa/Objects/floatobject.c	(original)
+++ python/branches/py3k-dtoa/Objects/floatobject.c	Tue Jul  6 17:55:27 2010
@@ -21,7 +21,22 @@
 extern int finite(double);
 #endif
 
-/* Special free list -- see comments for same code in intobject.c. */
+/* Special free list
+
+   Since some Python programs can spend much of their time allocating
+   and deallocating floats, these operations should be very fast.
+   Therefore we use a dedicated allocation scheme with a much lower
+   overhead (in space and time) than straight malloc(): a simple
+   dedicated free list, filled when necessary with memory from malloc().
+
+   block_list is a singly-linked list of all PyFloatBlocks ever allocated,
+   linked via their next members.  PyFloatBlocks are never returned to the
+   system before shutdown (PyFloat_Fini).
+
+   free_list is a singly-linked list of available PyFloatObjects, linked
+   via abuse of their ob_type members.
+*/
+
 #define BLOCK_SIZE      1000    /* 1K less typical malloc overhead */
 #define BHEAD_SIZE      8       /* Enough for a 64-bit pointer */
 #define N_FLOATOBJECTS  ((BLOCK_SIZE - BHEAD_SIZE) / sizeof(PyFloatObject))
@@ -1157,7 +1172,7 @@
         return float_str((PyFloatObject *)v);
 
     if (x == 0.0) {
-        if(copysign(1.0, x) == -1.0)
+        if (copysign(1.0, x) == -1.0)
             return PyUnicode_FromString("-0x0.0p+0");
         else
             return PyUnicode_FromString("0x0.0p+0");
@@ -1407,7 +1422,7 @@
                     round_up = 1;
                     break;
                 }
-        if (round_up == 1) {
+        if (round_up) {
             x += 2*half_eps;
             if (top_exp == DBL_MAX_EXP &&
                 x == ldexp((double)(2*half_eps), DBL_MANT_DIG))

Modified: python/branches/py3k-dtoa/Objects/frameobject.c
==============================================================================
--- python/branches/py3k-dtoa/Objects/frameobject.c	(original)
+++ python/branches/py3k-dtoa/Objects/frameobject.c	Tue Jul  6 17:55:27 2010
@@ -391,7 +391,7 @@
        the local variables in f_localsplus are NULL.
 
    2. We also maintain a separate free list of stack frames (just like
-   integers are allocated in a special way -- see intobject.c).  When
+   floats are allocated in a special way -- see floatobject.c).  When
    a stack frame is on the free list, only the following members have
    a meaning:
     ob_type             == &Frametype

Modified: python/branches/py3k-dtoa/Objects/typeobject.c
==============================================================================
--- python/branches/py3k-dtoa/Objects/typeobject.c	(original)
+++ python/branches/py3k-dtoa/Objects/typeobject.c	Tue Jul  6 17:55:27 2010
@@ -3476,11 +3476,8 @@
 static void
 inherit_special(PyTypeObject *type, PyTypeObject *base)
 {
-    Py_ssize_t oldsize, newsize;
 
     /* Copying basicsize is connected to the GC flags */
-    oldsize = base->tp_basicsize;
-    newsize = type->tp_basicsize ? type->tp_basicsize : oldsize;
     if (!(type->tp_flags & Py_TPFLAGS_HAVE_GC) &&
         (base->tp_flags & Py_TPFLAGS_HAVE_GC) &&
         (!type->tp_traverse && !type->tp_clear)) {
@@ -3507,7 +3504,8 @@
                 type->tp_new = base->tp_new;
         }
     }
-    type->tp_basicsize = newsize;
+    if (type->tp_basicsize == 0)
+        type->tp_basicsize = base->tp_basicsize;
 
     /* Copy other non-function slots */
 

Modified: python/branches/py3k-dtoa/Objects/unicodeobject.c
==============================================================================
--- python/branches/py3k-dtoa/Objects/unicodeobject.c	(original)
+++ python/branches/py3k-dtoa/Objects/unicodeobject.c	Tue Jul  6 17:55:27 2010
@@ -1273,7 +1273,7 @@
            case of a TypeError. */
         if (PyErr_ExceptionMatches(PyExc_TypeError))
             PyErr_Format(PyExc_TypeError,
-                         "coercing to str: need string or buffer, "
+                         "coercing to str: need bytes, bytearray or char buffer, "
                          "%.80s found",
                          Py_TYPE(obj)->tp_name);
         goto onError;
@@ -1767,6 +1767,33 @@
     return 0;
 }
 
+/* create or adjust a UnicodeDecodeError */
+static void
+make_decode_exception(PyObject **exceptionObject,
+                      const char *encoding,
+                      const char *input, Py_ssize_t length,
+                      Py_ssize_t startpos, Py_ssize_t endpos,
+                      const char *reason)
+{
+    if (*exceptionObject == NULL) {
+        *exceptionObject = PyUnicodeDecodeError_Create(
+            encoding, input, length, startpos, endpos, reason);
+    }
+    else {
+        if (PyUnicodeDecodeError_SetStart(*exceptionObject, startpos))
+            goto onError;
+        if (PyUnicodeDecodeError_SetEnd(*exceptionObject, endpos))
+            goto onError;
+        if (PyUnicodeDecodeError_SetReason(*exceptionObject, reason))
+            goto onError;
+    }
+    return;
+
+onError:
+    Py_DECREF(*exceptionObject);
+    *exceptionObject = NULL;
+}
+
 /* error handling callback helper:
    build arguments, call the callback and check the arguments,
    if no exception occurred, copy the replacement to the output
@@ -1800,20 +1827,13 @@
             goto onError;
     }
 
-    if (*exceptionObject == NULL) {
-        *exceptionObject = PyUnicodeDecodeError_Create(
-            encoding, *input, *inend-*input, *startinpos, *endinpos, reason);
-        if (*exceptionObject == NULL)
-            goto onError;
-    }
-    else {
-        if (PyUnicodeDecodeError_SetStart(*exceptionObject, *startinpos))
-            goto onError;
-        if (PyUnicodeDecodeError_SetEnd(*exceptionObject, *endinpos))
-            goto onError;
-        if (PyUnicodeDecodeError_SetReason(*exceptionObject, reason))
-            goto onError;
-    }
+    make_decode_exception(exceptionObject,
+        encoding,
+        *input, *inend - *input,
+        *startinpos, *endinpos,
+        reason);
+    if (*exceptionObject == NULL)
+        goto onError;
 
     restuple = PyObject_CallFunctionObjArgs(*errorHandler, *exceptionObject, NULL);
     if (restuple == NULL)
@@ -2265,24 +2285,24 @@
 
 static
 char utf8_code_length[256] = {
-    /* Map UTF-8 encoded prefix byte to sequence length.  zero means
-       illegal prefix.  see RFC 2279 for details */
-    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
-    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+    /* Map UTF-8 encoded prefix byte to sequence length.  Zero means
+       illegal prefix.  See RFC 3629 for details */
+    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 00-0F */
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
-    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 70-7F */
+    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 80-8F */
     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
-    2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
-    3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
-    4, 4, 4, 4, 4, 4, 4, 4, 5, 5, 5, 5, 6, 6, 0, 0
+    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* B0-BF */
+    0, 0, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, /* C0-C1 + C2-CF */
+    2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, /* D0-DF */
+    3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, /* E0-EF */
+    4, 4, 4, 4, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0  /* F0-F4 + F5-FF */
 };
 
 PyObject *PyUnicode_DecodeUTF8(const char *s,
@@ -2312,6 +2332,7 @@
 {
     const char *starts = s;
     int n;
+    int k;
     Py_ssize_t startinpos;
     Py_ssize_t endinpos;
     Py_ssize_t outpos;
@@ -2395,7 +2416,9 @@
             else {
                 errmsg = "unexpected end of data";
                 startinpos = s-starts;
-                endinpos = size;
+                endinpos = startinpos+1;
+                for (k=1; (k < size-startinpos) && ((s[k]&0xC0) == 0x80); k++)
+                    endinpos++;
                 goto utf8Error;
             }
         }
@@ -2403,7 +2426,7 @@
         switch (n) {
 
         case 0:
-            errmsg = "unexpected code byte";
+            errmsg = "invalid start byte";
             startinpos = s-starts;
             endinpos = startinpos+1;
             goto utf8Error;
@@ -2416,63 +2439,67 @@
 
         case 2:
             if ((s[1] & 0xc0) != 0x80) {
-                errmsg = "invalid data";
+                errmsg = "invalid continuation byte";
                 startinpos = s-starts;
-                endinpos = startinpos+2;
+                endinpos = startinpos + 1;
                 goto utf8Error;
             }
             ch = ((s[0] & 0x1f) << 6) + (s[1] & 0x3f);
-            if (ch < 0x80) {
-                startinpos = s-starts;
-                endinpos = startinpos+2;
-                errmsg = "illegal encoding";
-                goto utf8Error;
-            }
-            else
-                *p++ = (Py_UNICODE)ch;
+            assert ((ch > 0x007F) && (ch <= 0x07FF));
+            *p++ = (Py_UNICODE)ch;
             break;
 
         case 3:
+            /* Decoding UTF-8 sequences in range \xed\xa0\x80-\xed\xbf\xbf
+               will result in surrogates in range d800-dfff. Surrogates are
+               not valid UTF-8 so they are rejected.
+               See http://www.unicode.org/versions/Unicode5.2.0/ch03.pdf
+               (table 3-7) and http://www.rfc-editor.org/rfc/rfc3629.txt */
             if ((s[1] & 0xc0) != 0x80 ||
-                (s[2] & 0xc0) != 0x80) {
-                errmsg = "invalid data";
+                (s[2] & 0xc0) != 0x80 ||
+                ((unsigned char)s[0] == 0xE0 &&
+                 (unsigned char)s[1] < 0xA0) ||
+                ((unsigned char)s[0] == 0xED &&
+                 (unsigned char)s[1] > 0x9F)) {
+                errmsg = "invalid continuation byte";
                 startinpos = s-starts;
-                endinpos = startinpos+3;
+                endinpos = startinpos + 1;
+
+                /* if s[1] first two bits are 1 and 0, then the invalid
+                   continuation byte is s[2], so increment endinpos by 1,
+                   if not, s[1] is invalid and endinpos doesn't need to
+                   be incremented. */
+                if ((s[1] & 0xC0) == 0x80)
+                    endinpos++;
                 goto utf8Error;
             }
             ch = ((s[0] & 0x0f) << 12) + ((s[1] & 0x3f) << 6) + (s[2] & 0x3f);
-            if (ch < 0x0800 || (ch >= 0xd800 && ch <= 0xDFFF)) {
-                errmsg = "illegal encoding";
-                startinpos = s-starts;
-                endinpos = startinpos+3;
-                goto utf8Error;
-            }
-            else
-                *p++ = (Py_UNICODE)ch;
+            assert ((ch > 0x07FF) && (ch <= 0xFFFF));
+            *p++ = (Py_UNICODE)ch;
             break;
 
         case 4:
             if ((s[1] & 0xc0) != 0x80 ||
                 (s[2] & 0xc0) != 0x80 ||
-                (s[3] & 0xc0) != 0x80) {
-                errmsg = "invalid data";
+                (s[3] & 0xc0) != 0x80 ||
+                ((unsigned char)s[0] == 0xF0 &&
+                 (unsigned char)s[1] < 0x90) ||
+                ((unsigned char)s[0] == 0xF4 &&
+                 (unsigned char)s[1] > 0x8F)) {
+                errmsg = "invalid continuation byte";
                 startinpos = s-starts;
-                endinpos = startinpos+4;
+                endinpos = startinpos + 1;
+                if ((s[1] & 0xC0) == 0x80) {
+                    endinpos++;
+                    if ((s[2] & 0xC0) == 0x80)
+                        endinpos++;
+                }
                 goto utf8Error;
             }
             ch = ((s[0] & 0x7) << 18) + ((s[1] & 0x3f) << 12) +
-                ((s[2] & 0x3f) << 6) + (s[3] & 0x3f);
-            /* validate and convert to UTF-16 */
-            if ((ch < 0x10000)        /* minimum value allowed for 4
-                                         byte encoding */
-                || (ch > 0x10ffff))   /* maximum value allowed for
-                                         UTF-16 */
-            {
-                errmsg = "illegal encoding";
-                startinpos = s-starts;
-                endinpos = startinpos+4;
-                goto utf8Error;
-            }
+                 ((s[2] & 0x3f) << 6) + (s[3] & 0x3f);
+            assert ((ch > 0xFFFF) && (ch <= 0x10ffff));
+
 #ifdef Py_UNICODE_WIDE
             *p++ = (Py_UNICODE)ch;
 #else
@@ -2488,13 +2515,6 @@
             *p++ = (Py_UNICODE)(0xDC00 + (ch & 0x03FF));
 #endif
             break;
-
-        default:
-            /* Other sizes are only needed for UCS-4 */
-            errmsg = "unsupported Unicode code range";
-            startinpos = s-starts;
-            endinpos = startinpos+n;
-            goto utf8Error;
         }
         s += n;
         continue;
@@ -4552,32 +4572,46 @@
 static int decode_mbcs(PyUnicodeObject **v,
                        const char *s, /* MBCS string */
                        int size, /* sizeof MBCS string */
-                       int final)
+                       int final,
+                       const char *errors)
 {
     Py_UNICODE *p;
-    Py_ssize_t n = 0;
-    int usize = 0;
+    Py_ssize_t n;
+    DWORD usize;
+    DWORD flags;
 
     assert(size >= 0);
 
+    /* check and handle 'errors' arg */
+    if (errors==NULL || strcmp(errors, "strict")==0)
+        flags = MB_ERR_INVALID_CHARS;
+    else if (strcmp(errors, "ignore")==0)
+        flags = 0;
+    else {
+        PyErr_Format(PyExc_ValueError,
+                     "mbcs encoding does not support errors='%s'",
+                     errors);
+        return -1;
+    }
+
     /* Skip trailing lead-byte unless 'final' is set */
     if (!final && size >= 1 && is_dbcs_lead_byte(s, size - 1))
         --size;
 
     /* First get the size of the result */
     if (size > 0) {
-        usize = MultiByteToWideChar(CP_ACP, 0, s, size, NULL, 0);
-        if (usize == 0) {
-            PyErr_SetFromWindowsErrWithFilename(0, NULL);
-            return -1;
-        }
-    }
+        usize = MultiByteToWideChar(CP_ACP, flags, s, size, NULL, 0);
+        if (usize==0)
+            goto mbcs_decode_error;
+    } else
+        usize = 0;
 
     if (*v == NULL) {
         /* Create unicode object */
         *v = _PyUnicode_New(usize);
         if (*v == NULL)
             return -1;
+        n = 0;
     }
     else {
         /* Extend unicode object */
@@ -4587,15 +4621,35 @@
     }
 
     /* Do the conversion */
-    if (size > 0) {
+    if (usize > 0) {
         p = PyUnicode_AS_UNICODE(*v) + n;
-        if (0 == MultiByteToWideChar(CP_ACP, 0, s, size, p, usize)) {
-            PyErr_SetFromWindowsErrWithFilename(0, NULL);
-            return -1;
+        if (0 == MultiByteToWideChar(CP_ACP, flags, s, size, p, usize)) {
+            goto mbcs_decode_error;
         }
     }
-
     return size;
+
+mbcs_decode_error:
+    /* If the last error was ERROR_NO_UNICODE_TRANSLATION, then
+       we raise a UnicodeDecodeError - else it is a 'generic'
+       windows error
+     */
+    if (GetLastError()==ERROR_NO_UNICODE_TRANSLATION) {
+        /* Ideally, we should get reason from FormatMessage - this
+           is the Windows 2000 English version of the message
+        */
+        PyObject *exc = NULL;
+        const char *reason = "No mapping for the Unicode character exists "
+                             "in the target multi-byte code page.";
+        make_decode_exception(&exc, "mbcs", s, size, 0, 0, reason);
+        if (exc != NULL) {
+            PyCodec_StrictErrors(exc);
+            Py_DECREF(exc);
+        }
+    } else {
+        PyErr_SetFromWindowsErrWithFilename(0, NULL);
+    }
+    return -1;
 }
 
 PyObject *PyUnicode_DecodeMBCSStateful(const char *s,
@@ -4612,10 +4666,10 @@
 #ifdef NEED_RETRY
   retry:
     if (size > INT_MAX)
-        done = decode_mbcs(&v, s, INT_MAX, 0);
+        done = decode_mbcs(&v, s, INT_MAX, 0, errors);
     else
 #endif
-        done = decode_mbcs(&v, s, (int)size, !consumed);
+        done = decode_mbcs(&v, s, (int)size, !consumed, errors);
 
     if (done < 0) {
         Py_XDECREF(v);
@@ -4649,20 +4703,45 @@
  */
 static int encode_mbcs(PyObject **repr,
                        const Py_UNICODE *p, /* unicode */
-                       int size) /* size of unicode */
+                       int size, /* size of unicode */
+                       const char* errors)
 {
-    int mbcssize = 0;
-    Py_ssize_t n = 0;
+    BOOL usedDefaultChar = FALSE;
+    BOOL *pusedDefaultChar;
+    int mbcssize;
+    Py_ssize_t n;
+    PyObject *exc = NULL;
+    DWORD flags;
 
     assert(size >= 0);
 
+    /* check and handle 'errors' arg */
+    if (errors==NULL || strcmp(errors, "strict")==0) {
+        flags = WC_NO_BEST_FIT_CHARS;
+        pusedDefaultChar = &usedDefaultChar;
+    } else if (strcmp(errors, "replace")==0) {
+        flags = 0;
+        pusedDefaultChar = NULL;
+    } else {
+         PyErr_Format(PyExc_ValueError,
+                      "mbcs encoding does not support errors='%s'",
+                      errors);
+         return -1;
+    }
+
     /* First get the size of the result */
     if (size > 0) {
-        mbcssize = WideCharToMultiByte(CP_ACP, 0, p, size, NULL, 0, NULL, NULL);
+        mbcssize = WideCharToMultiByte(CP_ACP, flags, p, size, NULL, 0,
+                                       NULL, pusedDefaultChar);
         if (mbcssize == 0) {
             PyErr_SetFromWindowsErrWithFilename(0, NULL);
             return -1;
         }
+        /* If we used a default char, then we failed! */
+        if (pusedDefaultChar && *pusedDefaultChar)
+            goto mbcs_encode_error;
+    } else {
+        mbcssize = 0;
     }
 
     if (*repr == NULL) {
@@ -4670,6 +4749,7 @@
         *repr = PyBytes_FromStringAndSize(NULL, mbcssize);
         if (*repr == NULL)
             return -1;
+        n = 0;
     }
     else {
         /* Extend string object */
@@ -4681,13 +4761,20 @@
     /* Do the conversion */
     if (size > 0) {
         char *s = PyBytes_AS_STRING(*repr) + n;
-        if (0 == WideCharToMultiByte(CP_ACP, 0, p, size, s, mbcssize, NULL, NULL)) {
+        if (0 == WideCharToMultiByte(CP_ACP, flags, p, size, s, mbcssize,
+                                     NULL, pusedDefaultChar)) {
             PyErr_SetFromWindowsErrWithFilename(0, NULL);
             return -1;
         }
+        if (pusedDefaultChar && *pusedDefaultChar)
+            goto mbcs_encode_error;
     }
-
     return 0;
+
+mbcs_encode_error:
+    raise_encode_exception(&exc, "mbcs", p, size, 0, 0, "invalid character");
+    Py_XDECREF(exc);
+    return -1;
 }
 
 PyObject *PyUnicode_EncodeMBCS(const Py_UNICODE *p,
@@ -4700,10 +4787,10 @@
 #ifdef NEED_RETRY
   retry:
     if (size > INT_MAX)
-        ret = encode_mbcs(&repr, p, INT_MAX);
+        ret = encode_mbcs(&repr, p, INT_MAX, errors);
     else
 #endif
-        ret = encode_mbcs(&repr, p, (int)size);
+        ret = encode_mbcs(&repr, p, (int)size, errors);
 
     if (ret < 0) {
         Py_XDECREF(repr);
@@ -6618,7 +6705,7 @@
              "S.capitalize() -> str\n\
 \n\
 Return a capitalized version of S, i.e. make the first character\n\
-have upper case.");
+have upper case and the rest lower case.");
 
 static PyObject*
 unicode_capitalize(PyUnicodeObject *self)
@@ -7890,7 +7977,7 @@
 }
 
 PyDoc_STRVAR(replace__doc__,
-             "S.replace (old, new[, count]) -> str\n\
+             "S.replace(old, new[, count]) -> str\n\
 \n\
 Return a copy of S with all occurrences of substring\n\
 old replaced by new.  If the optional argument count is\n\

Modified: python/branches/py3k-dtoa/PC/VS8.0/pythoncore.vcproj
==============================================================================
--- python/branches/py3k-dtoa/PC/VS8.0/pythoncore.vcproj	(original)
+++ python/branches/py3k-dtoa/PC/VS8.0/pythoncore.vcproj	Tue Jul  6 17:55:27 2010
@@ -759,10 +759,6 @@
 				>
 			</File>
 			<File
-				RelativePath="..\..\Include\intobject.h"
-				>
-			</File>
-			<File
 				RelativePath="..\..\Include\intrcheck.h"
 				>
 			</File>

Modified: python/branches/py3k-dtoa/PCbuild/_ctypes.vcproj
==============================================================================
--- python/branches/py3k-dtoa/PCbuild/_ctypes.vcproj	(original)
+++ python/branches/py3k-dtoa/PCbuild/_ctypes.vcproj	Tue Jul  6 17:55:27 2010
@@ -1,7 +1,7 @@
 <?xml version="1.0" encoding="Windows-1252"?>
 <VisualStudioProject
 	ProjectType="Visual C++"
-	Version="9,00"
+	Version="9.00"
 	Name="_ctypes"
 	ProjectGUID="{0E9791DB-593A-465F-98BC-681011311618}"
 	RootNamespace="_ctypes"
@@ -42,8 +42,7 @@
 			/>
 			<Tool
 				Name="VCCLCompilerTool"
-				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc;..\Modules\_ctypes\libffi\include;..\Modules\_ctypes\libffi\src\x86"
-				PreprocessorDefinitions="X86_WIN32"
+				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc"
 			/>
 			<Tool
 				Name="VCManagedResourceCompilerTool"
@@ -104,8 +103,7 @@
 			/>
 			<Tool
 				Name="VCCLCompilerTool"
-				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc;..\Modules\_ctypes\libffi\include;..\Modules\_ctypes\libffi\src\x86"
-				PreprocessorDefinitions="X86_WIN64"
+				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc"
 			/>
 			<Tool
 				Name="VCManagedResourceCompilerTool"
@@ -166,8 +164,7 @@
 			/>
 			<Tool
 				Name="VCCLCompilerTool"
-				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc;..\Modules\_ctypes\libffi\include;..\Modules\_ctypes\libffi\src\x86"
-				PreprocessorDefinitions="X86_WIN32"
+				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc"
 			/>
 			<Tool
 				Name="VCManagedResourceCompilerTool"
@@ -231,8 +228,7 @@
 			/>
 			<Tool
 				Name="VCCLCompilerTool"
-				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc;..\Modules\_ctypes\libffi\include;..\Modules\_ctypes\libffi\src\x86"
-				PreprocessorDefinitions="X86_WIN64"
+				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc"
 			/>
 			<Tool
 				Name="VCManagedResourceCompilerTool"
@@ -295,8 +291,7 @@
 			/>
 			<Tool
 				Name="VCCLCompilerTool"
-				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc;..\Modules\_ctypes\libffi\include;..\Modules\_ctypes\libffi\src\x86"
-				PreprocessorDefinitions="X86_WIN32"
+				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc"
 			/>
 			<Tool
 				Name="VCManagedResourceCompilerTool"
@@ -360,8 +355,7 @@
 			/>
 			<Tool
 				Name="VCCLCompilerTool"
-				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc;..\Modules\_ctypes\libffi\include;..\Modules\_ctypes\libffi\src\x86"
-				PreprocessorDefinitions="X86_WIN64"
+				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc"
 			/>
 			<Tool
 				Name="VCManagedResourceCompilerTool"
@@ -425,8 +419,7 @@
 			/>
 			<Tool
 				Name="VCCLCompilerTool"
-				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc;..\Modules\_ctypes\libffi\include;..\Modules\_ctypes\libffi\src\x86"
-				PreprocessorDefinitions="X86_WIN32"
+				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc"
 			/>
 			<Tool
 				Name="VCManagedResourceCompilerTool"
@@ -490,8 +483,7 @@
 			/>
 			<Tool
 				Name="VCCLCompilerTool"
-				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc;..\Modules\_ctypes\libffi\include;..\Modules\_ctypes\libffi\src\x86"
-				PreprocessorDefinitions="X86_WIN64"
+				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc"
 			/>
 			<Tool
 				Name="VCManagedResourceCompilerTool"
@@ -551,7 +543,7 @@
 				>
 			</File>
 			<File
-				RelativePath="..\Modules\_ctypes\libffi\include\ffi_common.h"
+				RelativePath="..\Modules\_ctypes\libffi_msvc\ffi_common.h"
 				>
 			</File>
 			<File
@@ -559,7 +551,7 @@
 				>
 			</File>
 			<File
-				RelativePath="..\Modules\_ctypes\libffi\src\x86\ffitarget.h"
+				RelativePath="..\Modules\_ctypes\libffi_msvc\ffitarget.h"
 				>
 			</File>
 		</Filter>
@@ -583,23 +575,15 @@
 				>
 			</File>
 			<File
-				RelativePath="..\Modules\_ctypes\libffi\src\x86\ffi.c"
+				RelativePath="..\Modules\_ctypes\libffi_msvc\ffi.c"
 				>
-				<FileConfiguration
-					Name="Debug|x64"
-					>
-					<Tool
-						Name="VCCLCompilerTool"
-						PreprocessorDefinitions="X86_WIN64"
-					/>
-				</FileConfiguration>
 			</File>
 			<File
 				RelativePath="..\Modules\_ctypes\malloc_closure.c"
 				>
 			</File>
 			<File
-				RelativePath="..\Modules\_ctypes\libffi\src\prep_cif.c"
+				RelativePath="..\Modules\_ctypes\libffi_msvc\prep_cif.c"
 				>
 			</File>
 			<File
@@ -607,32 +591,14 @@
 				>
 			</File>
 			<File
-				RelativePath="..\Modules\_ctypes\libffi\src\x86\win32.S"
+				RelativePath="..\Modules\_ctypes\libffi_msvc\win32.c"
 				>
 				<FileConfiguration
-					Name="Debug|Win32"
-					>
-					<Tool
-						Name="VCCustomBuildTool"
-						CommandLine="cl /nologo /EP /I ..\Modules\_ctypes\libffi_msvc /I ..\Modules\_ctypes\libffi\src\x86 $(InputPath) &gt; $(IntDir)$(InputName).asm&#x0D;&#x0A;ml /nologo /Zi /c /Fo $(IntDir)$(InputName).obj $(IntDir)$(InputName).asm&#x0D;&#x0A;"
-						Outputs="$(IntDir)$(InputName).obj"
-					/>
-				</FileConfiguration>
-				<FileConfiguration
 					Name="Debug|x64"
 					ExcludedFromBuild="true"
 					>
 					<Tool
-						Name="VCCustomBuildTool"
-					/>
-				</FileConfiguration>
-				<FileConfiguration
-					Name="Release|Win32"
-					>
-					<Tool
-						Name="VCCustomBuildTool"
-						CommandLine="cl /nologo /EP /I ..\Modules\_ctypes\libffi_msvc /I ..\Modules\_ctypes\libffi\src\x86 $(InputPath) &gt; $(IntDir)$(InputName).asm&#x0D;&#x0A;ml /nologo /c /Fo $(IntDir)$(InputName).obj $(IntDir)$(InputName).asm&#x0D;&#x0A;"
-						Outputs="$(IntDir)$(InputName).obj"
+						Name="VCCLCompilerTool"
 					/>
 				</FileConfiguration>
 				<FileConfiguration
@@ -640,16 +606,7 @@
 					ExcludedFromBuild="true"
 					>
 					<Tool
-						Name="VCCustomBuildTool"
-					/>
-				</FileConfiguration>
-				<FileConfiguration
-					Name="PGInstrument|Win32"
-					>
-					<Tool
-						Name="VCCustomBuildTool"
-						CommandLine="cl /nologo /EP /I ..\Modules\_ctypes\libffi_msvc /I ..\Modules\_ctypes\libffi\src\x86 $(InputPath) &gt; $(IntDir)$(InputName).asm&#x0D;&#x0A;ml /nologo /Zi /c /Fo $(IntDir)$(InputName).obj $(IntDir)$(InputName).asm&#x0D;&#x0A;"
-						Outputs="$(IntDir)$(InputName).obj"
+						Name="VCCLCompilerTool"
 					/>
 				</FileConfiguration>
 				<FileConfiguration
@@ -657,16 +614,7 @@
 					ExcludedFromBuild="true"
 					>
 					<Tool
-						Name="VCCustomBuildTool"
-					/>
-				</FileConfiguration>
-				<FileConfiguration
-					Name="PGUpdate|Win32"
-					>
-					<Tool
-						Name="VCCustomBuildTool"
-						CommandLine="cl /nologo /EP /I ..\Modules\_ctypes\libffi_msvc /I ..\Modules\_ctypes\libffi\src\x86 $(InputPath) &gt; $(IntDir)$(InputName).asm&#x0D;&#x0A;ml /nologo /Zi /c /Fo $(IntDir)$(InputName).obj $(IntDir)$(InputName).asm&#x0D;&#x0A;"
-						Outputs="$(IntDir)$(InputName).obj"
+						Name="VCCLCompilerTool"
 					/>
 				</FileConfiguration>
 				<FileConfiguration
@@ -674,12 +622,12 @@
 					ExcludedFromBuild="true"
 					>
 					<Tool
-						Name="VCCustomBuildTool"
+						Name="VCCLCompilerTool"
 					/>
 				</FileConfiguration>
 			</File>
 			<File
-				RelativePath="..\Modules\_ctypes\libffi\src\x86\win64.S"
+				RelativePath="..\Modules\_ctypes\libffi_msvc\win64.asm"
 				>
 				<FileConfiguration
 					Name="Debug|Win32"
@@ -694,8 +642,8 @@
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="cl /nologo /EP /I ..\Modules\_ctypes\libffi_msvc /I ..\Modules\_ctypes\libffi\src\x86 $(InputPath) &gt; $(IntDir)$(InputName).asm&#x0D;&#x0A;ml64 /Zi /nologo /c /Fo $(IntDir)$(InputName).obj $(IntDir)$(InputName).asm&#x0D;&#x0A;"
-						Outputs="$(IntDir)$(InputName).obj"
+						CommandLine="ml64 /nologo /c /Zi /Fo &quot;$(IntDir)\win64.obj&quot; &quot;$(InputPath)&quot;&#x0D;&#x0A;"
+						Outputs="$(IntDir)\win64.obj"
 					/>
 				</FileConfiguration>
 				<FileConfiguration
@@ -711,8 +659,8 @@
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="cl /nologo /EP /I ..\Modules\_ctypes\libffi_msvc /I ..\Modules\_ctypes\libffi\src\x86 $(InputPath) &gt; $(IntDir)$(InputName).asm&#x0D;&#x0A;ml64 /nologo /c /Fo $(IntDir)$(InputName).obj $(IntDir)$(InputName).asm&#x0D;&#x0A;"
-						Outputs="$(IntDir)$(InputName).obj"
+						CommandLine="ml64 /nologo /c /Fo &quot;$(IntDir)\win64.obj&quot; &quot;$(InputPath)&quot;&#x0D;&#x0A;"
+						Outputs="$(IntDir)\win64.obj"
 					/>
 				</FileConfiguration>
 				<FileConfiguration
@@ -728,8 +676,8 @@
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="cl /nologo /EP /I ..\Modules\_ctypes\libffi_msvc /I ..\Modules\_ctypes\libffi\src\x86 $(InputPath) &gt; $(IntDir)$(InputName).asm&#x0D;&#x0A;ml64 /nologo /c /Fo $(IntDir)$(InputName).obj $(IntDir)$(InputName).asm&#x0D;&#x0A;"
-						Outputs="$(IntDir)$(InputName).obj"
+						CommandLine="ml64 /nologo /c /Fo &quot;$(IntDir)\win64.obj&quot; &quot;$(InputPath)&quot;&#x0D;&#x0A;"
+						Outputs="$(IntDir)\win64.obj"
 					/>
 				</FileConfiguration>
 				<FileConfiguration
@@ -745,8 +693,8 @@
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="cl /nologo /EP /I ..\Modules\_ctypes\libffi_msvc /I ..\Modules\_ctypes\libffi\src\x86 $(InputPath) &gt; $(IntDir)$(InputName).asm&#x0D;&#x0A;ml64 /nologo /c /Fo $(IntDir)$(InputName).obj $(IntDir)$(InputName).asm&#x0D;&#x0A;"
-						Outputs="$(IntDir)$(InputName).obj"
+						CommandLine="ml64 /nologo /c /Fo &quot;$(IntDir)\win64.obj&quot; &quot;$(InputPath)&quot;&#x0D;&#x0A;"
+						Outputs="$(IntDir)\win64.obj"
 					/>
 				</FileConfiguration>
 			</File>

Modified: python/branches/py3k-dtoa/PCbuild/pythoncore.vcproj
==============================================================================
--- python/branches/py3k-dtoa/PCbuild/pythoncore.vcproj	(original)
+++ python/branches/py3k-dtoa/PCbuild/pythoncore.vcproj	Tue Jul  6 17:55:27 2010
@@ -755,10 +755,6 @@
 				>
 			</File>
 			<File
-				RelativePath="..\Include\intobject.h"
-				>
-			</File>
-			<File
 				RelativePath="..\Include\intrcheck.h"
 				>
 			</File>
@@ -1040,6 +1036,14 @@
 				>
 			</File>
 			<File
+				RelativePath="..\Modules\_time.c"
+				>
+			</File>
+			<File
+				RelativePath="..\Modules\_time.h"
+				>
+			</File>
+			<File
 				RelativePath="..\Modules\_weakref.c"
 				>
 			</File>

Modified: python/branches/py3k-dtoa/Parser/Python.asdl
==============================================================================
--- python/branches/py3k-dtoa/Parser/Python.asdl	(original)
+++ python/branches/py3k-dtoa/Parser/Python.asdl	Tue Jul  6 17:55:27 2010
@@ -17,7 +17,7 @@
 			 expr? starargs,
 			 expr? kwargs,
 			 stmt* body,
-			 expr *decorator_list)
+			 expr* decorator_list)
 	      | Return(expr? value)
 
 	      | Delete(expr* targets)

Modified: python/branches/py3k-dtoa/Parser/asdl_c.py
==============================================================================
--- python/branches/py3k-dtoa/Parser/asdl_c.py	(original)
+++ python/branches/py3k-dtoa/Parser/asdl_c.py	Tue Jul  6 17:55:27 2010
@@ -732,8 +732,9 @@
 {
     int i, result;
     PyObject *s, *l = PyTuple_New(num_fields);
-    if (!l) return 0;
-    for(i = 0; i < num_fields; i++) {
+    if (!l)
+        return 0;
+    for (i = 0; i < num_fields; i++) {
         s = PyUnicode_FromString(attrs[i]);
         if (!s) {
             Py_DECREF(l);

Modified: python/branches/py3k-dtoa/Python/Python-ast.c
==============================================================================
--- python/branches/py3k-dtoa/Python/Python-ast.c	(original)
+++ python/branches/py3k-dtoa/Python/Python-ast.c	Tue Jul  6 17:55:27 2010
@@ -2,7 +2,7 @@
 
 
 /*
-   __version__ 73626.
+   __version__ 82163.
 
    This module must be committed separately after each AST grammar change;
    The __version__ number is set to the revision number of the commit
@@ -537,8 +537,9 @@
 {
     int i, result;
     PyObject *s, *l = PyTuple_New(num_fields);
-    if (!l) return 0;
-    for(i = 0; i < num_fields; i++) {
+    if (!l)
+        return 0;
+    for (i = 0; i < num_fields; i++) {
         s = PyUnicode_FromString(attrs[i]);
         if (!s) {
             Py_DECREF(l);
@@ -6773,7 +6774,7 @@
             NULL;
         if (PyModule_AddIntConstant(m, "PyCF_ONLY_AST", PyCF_ONLY_AST) < 0)
                 return NULL;
-        if (PyModule_AddStringConstant(m, "__version__", "73626") < 0)
+        if (PyModule_AddStringConstant(m, "__version__", "82163") < 0)
                 return NULL;
         if (PyDict_SetItemString(d, "mod", (PyObject*)mod_type) < 0) return
             NULL;

Modified: python/branches/py3k-dtoa/Python/_warnings.c
==============================================================================
--- python/branches/py3k-dtoa/Python/_warnings.c	(original)
+++ python/branches/py3k-dtoa/Python/_warnings.c	Tue Jul  6 17:55:27 2010
@@ -251,7 +251,7 @@
 
     name = PyObject_GetAttrString(category, "__name__");
     if (name == NULL)  /* XXX Can an object lack a '__name__' attribute? */
-            return;
+        return;
 
     f_stderr = PySys_GetObject("stderr");
     if (f_stderr == NULL) {
@@ -282,8 +282,7 @@
         PyFile_WriteString("\n", f_stderr);
     }
     else
-        if (_Py_DisplaySourceLine(f_stderr, _PyUnicode_AsString(filename),
-                              lineno, 2) < 0)
+        if (_Py_DisplaySourceLine(f_stderr, filename, lineno, 2) < 0)
                 return;
     PyErr_Clear();
 }
@@ -847,28 +846,35 @@
 static PyObject *
 init_filters(void)
 {
-    PyObject *filters = PyList_New(3);
+    /* Don't silence DeprecationWarning if -3 was used. */
+    PyObject *filters = PyList_New(4);
+    unsigned int pos = 0;  /* Post-incremented in each use. */
+    unsigned int x;
     const char *bytes_action;
+
     if (filters == NULL)
         return NULL;
 
-    PyList_SET_ITEM(filters, 0,
+    PyList_SET_ITEM(filters, pos++,
+                    create_filter(PyExc_DeprecationWarning, "ignore"));
+    PyList_SET_ITEM(filters, pos++,
                     create_filter(PyExc_PendingDeprecationWarning, "ignore"));
-    PyList_SET_ITEM(filters, 1, create_filter(PyExc_ImportWarning, "ignore"));
+    PyList_SET_ITEM(filters, pos++,
+                    create_filter(PyExc_ImportWarning, "ignore"));
     if (Py_BytesWarningFlag > 1)
         bytes_action = "error";
     else if (Py_BytesWarningFlag)
         bytes_action = "default";
     else
         bytes_action = "ignore";
-    PyList_SET_ITEM(filters, 2, create_filter(PyExc_BytesWarning,
+    PyList_SET_ITEM(filters, pos++, create_filter(PyExc_BytesWarning,
                     bytes_action));
 
-    if (PyList_GET_ITEM(filters, 0) == NULL ||
-        PyList_GET_ITEM(filters, 1) == NULL ||
-        PyList_GET_ITEM(filters, 2) == NULL) {
-        Py_DECREF(filters);
-        return NULL;
+    for (x = 0; x < pos; x += 1) {
+        if (PyList_GET_ITEM(filters, x) == NULL) {
+            Py_DECREF(filters);
+            return NULL;
+        }
     }
 
     return filters;

Modified: python/branches/py3k-dtoa/Python/ast.c
==============================================================================
--- python/branches/py3k-dtoa/Python/ast.c	(original)
+++ python/branches/py3k-dtoa/Python/ast.c	Tue Jul  6 17:55:27 2010
@@ -1678,34 +1678,8 @@
 static expr_ty
 ast_for_factor(struct compiling *c, const node *n)
 {
-    node *pfactor, *ppower, *patom, *pnum;
     expr_ty expression;
 
-    /* If the unary - operator is applied to a constant, don't generate
-       a UNARY_NEGATIVE opcode.  Just store the approriate value as a
-       constant.  The peephole optimizer already does something like
-       this but it doesn't handle the case where the constant is
-       (sys.maxint - 1).  In that case, we want a PyIntObject, not a
-       PyLongObject.
-    */
-    if (TYPE(CHILD(n, 0)) == MINUS
-        && NCH(n) == 2
-        && TYPE((pfactor = CHILD(n, 1))) == factor
-        && NCH(pfactor) == 1
-        && TYPE((ppower = CHILD(pfactor, 0))) == power
-        && NCH(ppower) == 1
-        && TYPE((patom = CHILD(ppower, 0))) == atom
-        && TYPE((pnum = CHILD(patom, 0))) == NUMBER) {
-        char *s = PyObject_MALLOC(strlen(STR(pnum)) + 2);
-        if (s == NULL)
-            return NULL;
-        s[0] = '-';
-        strcpy(s + 1, STR(pnum));
-        PyObject_FREE(STR(pnum));
-        STR(pnum) = s;
-        return ast_for_atom(c, patom);
-    }
-
     expression = ast_for_expr(c, CHILD(n, 1));
     if (!expression)
         return NULL;

Modified: python/branches/py3k-dtoa/Python/ceval.c
==============================================================================
--- python/branches/py3k-dtoa/Python/ceval.c	(original)
+++ python/branches/py3k-dtoa/Python/ceval.c	Tue Jul  6 17:55:27 2010
@@ -87,7 +87,7 @@
     inst = inst1 - inst0 - intr;
     loop = loop1 - loop0 - intr;
     fprintf(stderr, "opcode=%03d t=%d inst=%06lld loop=%06lld\n",
-        opcode, ticked, inst, loop);
+            opcode, ticked, inst, loop);
 }
 
 #endif
@@ -126,10 +126,10 @@
 static int call_trace(Py_tracefunc, PyObject *, PyFrameObject *,
                       int, PyObject *);
 static int call_trace_protected(Py_tracefunc, PyObject *,
-                                 PyFrameObject *, int, PyObject *);
+                                PyFrameObject *, int, PyObject *);
 static void call_exc_trace(Py_tracefunc, PyObject *, PyFrameObject *);
 static int maybe_call_line_trace(Py_tracefunc, PyObject *,
-                                  PyFrameObject *, int *, int *, int *);
+                                 PyFrameObject *, int *, int *, int *);
 
 static PyObject * cmp_outcome(int, PyObject *, PyObject *);
 static PyObject * import_from(PyObject *, PyObject *);
@@ -718,14 +718,14 @@
 
 /* Status code for main loop (reason for stack unwind) */
 enum why_code {
-                WHY_NOT =       0x0001, /* No error */
-        WHY_EXCEPTION = 0x0002,         /* Exception occurred */
-        WHY_RERAISE =           0x0004, /* Exception re-raised by 'finally' */
-        WHY_RETURN =            0x0008, /* 'return' statement */
-        WHY_BREAK =             0x0010, /* 'break' statement */
-        WHY_CONTINUE =          0x0020, /* 'continue' statement */
-        WHY_YIELD =             0x0040, /* 'yield' operator */
-        WHY_SILENCED = 0x0080 /* Exception silenced by 'with' */
+        WHY_NOT =       0x0001, /* No error */
+        WHY_EXCEPTION = 0x0002, /* Exception occurred */
+        WHY_RERAISE =   0x0004, /* Exception re-raised by 'finally' */
+        WHY_RETURN =    0x0008, /* 'return' statement */
+        WHY_BREAK =     0x0010, /* 'break' statement */
+        WHY_CONTINUE =  0x0020, /* 'continue' statement */
+        WHY_YIELD =     0x0040, /* 'yield' operator */
+        WHY_SILENCED =  0x0080  /* Exception silenced by 'with' */
 };
 
 static enum why_code do_raise(PyObject *, PyObject *);
@@ -1005,38 +1005,38 @@
 
 /* The stack can grow at most MAXINT deep, as co_nlocals and
    co_stacksize are ints. */
-#define STACK_LEVEL()   ((int)(stack_pointer - f->f_valuestack))
-#define EMPTY()         (STACK_LEVEL() == 0)
-#define TOP()           (stack_pointer[-1])
-#define SECOND()        (stack_pointer[-2])
-#define THIRD()         (stack_pointer[-3])
-#define FOURTH()        (stack_pointer[-4])
-#define PEEK(n)         (stack_pointer[-(n)])
-#define SET_TOP(v)      (stack_pointer[-1] = (v))
-#define SET_SECOND(v)   (stack_pointer[-2] = (v))
-#define SET_THIRD(v)    (stack_pointer[-3] = (v))
-#define SET_FOURTH(v)   (stack_pointer[-4] = (v))
-#define SET_VALUE(n, v) (stack_pointer[-(n)] = (v))
-#define BASIC_STACKADJ(n)       (stack_pointer += n)
-#define BASIC_PUSH(v)   (*stack_pointer++ = (v))
-#define BASIC_POP()     (*--stack_pointer)
+#define STACK_LEVEL()     ((int)(stack_pointer - f->f_valuestack))
+#define EMPTY()           (STACK_LEVEL() == 0)
+#define TOP()             (stack_pointer[-1])
+#define SECOND()          (stack_pointer[-2])
+#define THIRD()           (stack_pointer[-3])
+#define FOURTH()          (stack_pointer[-4])
+#define PEEK(n)           (stack_pointer[-(n)])
+#define SET_TOP(v)        (stack_pointer[-1] = (v))
+#define SET_SECOND(v)     (stack_pointer[-2] = (v))
+#define SET_THIRD(v)      (stack_pointer[-3] = (v))
+#define SET_FOURTH(v)     (stack_pointer[-4] = (v))
+#define SET_VALUE(n, v)   (stack_pointer[-(n)] = (v))
+#define BASIC_STACKADJ(n) (stack_pointer += n)
+#define BASIC_PUSH(v)     (*stack_pointer++ = (v))
+#define BASIC_POP()       (*--stack_pointer)
 
 #ifdef LLTRACE
 #define PUSH(v)         { (void)(BASIC_PUSH(v), \
-                   lltrace && prtrace(TOP(), "push")); \
-                   assert(STACK_LEVEL() <= co->co_stacksize); }
+                          lltrace && prtrace(TOP(), "push")); \
+                          assert(STACK_LEVEL() <= co->co_stacksize); }
 #define POP()           ((void)(lltrace && prtrace(TOP(), "pop")), \
-             BASIC_POP())
+                         BASIC_POP())
 #define STACKADJ(n)     { (void)(BASIC_STACKADJ(n), \
-                   lltrace && prtrace(TOP(), "stackadj")); \
-                   assert(STACK_LEVEL() <= co->co_stacksize); }
+                          lltrace && prtrace(TOP(), "stackadj")); \
+                          assert(STACK_LEVEL() <= co->co_stacksize); }
 #define EXT_POP(STACK_POINTER) ((void)(lltrace && \
-                prtrace((STACK_POINTER)[-1], "ext_pop")), \
-                *--(STACK_POINTER))
+                                prtrace((STACK_POINTER)[-1], "ext_pop")), \
+                                *--(STACK_POINTER))
 #else
-#define PUSH(v)         BASIC_PUSH(v)
-#define POP()           BASIC_POP()
-#define STACKADJ(n)     BASIC_STACKADJ(n)
+#define PUSH(v)                BASIC_PUSH(v)
+#define POP()                  BASIC_POP()
+#define STACKADJ(n)            BASIC_STACKADJ(n)
 #define EXT_POP(STACK_POINTER) (*--(STACK_POINTER))
 #endif
 
@@ -1051,8 +1051,8 @@
    accessed by other code (e.g. a __del__ method or gc.collect()) and the
    variable would be pointing to already-freed memory. */
 #define SETLOCAL(i, value)      do { PyObject *tmp = GETLOCAL(i); \
-                     GETLOCAL(i) = value; \
-                     Py_XDECREF(tmp); } while (0)
+                                     GETLOCAL(i) = value; \
+                                     Py_XDECREF(tmp); } while (0)
 
 
 #define UNWIND_BLOCK(b) \
@@ -1318,7 +1318,7 @@
             it doesn't have to be remembered across a full loop */
         if (HAS_ARG(opcode))
             oparg = NEXTARG();
-      dispatch_opcode:
+    dispatch_opcode:
 #ifdef DYNAMIC_EXECUTION_PROFILE
 #ifdef DXPAIRS
         dxpairs[lastopcode][opcode]++;
@@ -2174,8 +2174,8 @@
                 break;
             if (oparg < PyTuple_GET_SIZE(co->co_cellvars)) {
                 v = PyTuple_GET_ITEM(co->co_cellvars,
-                                       oparg);
-                   format_exc_check_arg(
+                                     oparg);
+                format_exc_check_arg(
                        PyExc_UnboundLocalError,
                        UNBOUNDLOCAL_ERROR_MSG,
                        v);
@@ -2695,7 +2695,7 @@
             func = *pfunc;
 
             if (PyMethod_Check(func)
-            && PyMethod_GET_SELF(func) != NULL) {
+                && PyMethod_GET_SELF(func) != NULL) {
                 PyObject *self = PyMethod_GET_SELF(func);
                 Py_INCREF(self);
                 func = PyMethod_GET_FUNCTION(func);
@@ -3100,11 +3100,11 @@
             if (!(co->co_flags & CO_VARARGS)) {
                 PyErr_Format(PyExc_TypeError,
                     "%U() takes %s %d "
-                    "argument%s (%d given)",
+                    "positional argument%s (%d given)",
                     co->co_name,
                     defcount ? "at most" : "exactly",
-                    total_args,
-                    total_args == 1 ? "" : "s",
+                    co->co_argcount,
+                    co->co_argcount == 1 ? "" : "s",
                     argcount + kwcount);
                 goto fail;
             }

Modified: python/branches/py3k-dtoa/Python/compile.c
==============================================================================
--- python/branches/py3k-dtoa/Python/compile.c	(original)
+++ python/branches/py3k-dtoa/Python/compile.c	Tue Jul  6 17:55:27 2010
@@ -968,7 +968,7 @@
 {
     int arg = compiler_add_o(c, dict, o);
     if (arg < 0)
-    return 0;
+        return 0;
     return compiler_addop_i(c, opcode, arg);
 }
 
@@ -979,11 +979,11 @@
     int arg;
     PyObject *mangled = _Py_Mangle(c->u->u_private, o);
     if (!mangled)
-    return 0;
+        return 0;
     arg = compiler_add_o(c, dict, mangled);
     Py_DECREF(mangled);
     if (arg < 0)
-    return 0;
+        return 0;
     return compiler_addop_i(c, opcode, arg);
 }
 
@@ -1134,7 +1134,7 @@
 compiler_isdocstring(stmt_ty s)
 {
     if (s->kind != Expr_kind)
-    return 0;
+        return 0;
     return s->v.Expr.value->kind == Str_kind;
 }
 
@@ -1240,11 +1240,11 @@
     PyObject *k, *v;
     k = PyTuple_Pack(2, name, name->ob_type);
     if (k == NULL)
-    return -1;
+        return -1;
     v = PyDict_GetItem(dict, k);
     Py_DECREF(k);
     if (v == NULL)
-    return -1;
+        return -1;
     return PyLong_AS_LONG(v);
 }
 
@@ -3073,7 +3073,7 @@
     block = compiler_new_block(c);
     finally = compiler_new_block(c);
     if (!block || !finally)
-    return 0;
+        return 0;
 
     /* Evaluate EXPR */
     VISIT(c, expr, s->v.With.context_expr);
@@ -3082,15 +3082,15 @@
     /* SETUP_WITH pushes a finally block. */
     compiler_use_next_block(c, block);
     if (!compiler_push_fblock(c, FINALLY_TRY, block)) {
-    return 0;
+        return 0;
     }
 
     if (s->v.With.optional_vars) {
-    VISIT(c, expr, s->v.With.optional_vars);
+        VISIT(c, expr, s->v.With.optional_vars);
     }
     else {
     /* Discard result from context.__enter__() */
-    ADDOP(c, POP_TOP);
+        ADDOP(c, POP_TOP);
     }
 
     /* BLOCK code */
@@ -3103,7 +3103,7 @@
     ADDOP_O(c, LOAD_CONST, Py_None, consts);
     compiler_use_next_block(c, finally);
     if (!compiler_push_fblock(c, FINALLY_END, finally))
-    return 0;
+        return 0;
 
     /* Finally block starts; context.__exit__ is on the stack under
        the exception or return information. Just issue our magic
@@ -3942,7 +3942,7 @@
     freevars = dict_keys_inorder(c->u->u_freevars, PyTuple_Size(cellvars));
     if (!freevars)
         goto error;
-    filename = PyUnicode_DecodeFSDefault(c->c_filename);
+    filename = PyUnicode_FromString(c->c_filename);
     if (!filename)
         goto error;
 

Modified: python/branches/py3k-dtoa/Python/getargs.c
==============================================================================
--- python/branches/py3k-dtoa/Python/getargs.c	(original)
+++ python/branches/py3k-dtoa/Python/getargs.c	Tue Jul  6 17:55:27 2010
@@ -853,70 +853,6 @@
     /* XXX WAAAAH!  's', 'y', 'z', 'u', 'Z', 'e', 'w' codes all
        need to be cleaned up! */
 
-    case 's': {/* text string */
-        if (*format == '*') {
-            Py_buffer *p = (Py_buffer *)va_arg(*p_va, Py_buffer *);
-
-            if (PyUnicode_Check(arg)) {
-                uarg = UNICODE_DEFAULT_ENCODING(arg);
-                if (uarg == NULL)
-                    return converterr(CONV_UNICODE,
-                                      arg, msgbuf, bufsize);
-                PyBuffer_FillInfo(p, arg,
-                                  PyBytes_AS_STRING(uarg), PyBytes_GET_SIZE(uarg),
-                                  1, 0);
-            }
-            else { /* any buffer-like object */
-                char *buf;
-                if (getbuffer(arg, p, &buf) < 0)
-                    return converterr(buf, arg, msgbuf, bufsize);
-            }
-            if (addcleanup(p, freelist, cleanup_buffer)) {
-                return converterr(
-                    "(cleanup problem)",
-                    arg, msgbuf, bufsize);
-            }
-            format++;
-        } else if (*format == '#') {
-            void **p = (void **)va_arg(*p_va, char **);
-            FETCH_SIZE;
-
-            if (PyUnicode_Check(arg)) {
-                uarg = UNICODE_DEFAULT_ENCODING(arg);
-                if (uarg == NULL)
-                    return converterr(CONV_UNICODE,
-                                      arg, msgbuf, bufsize);
-                *p = PyBytes_AS_STRING(uarg);
-                STORE_SIZE(PyBytes_GET_SIZE(uarg));
-            }
-            else { /* any buffer-like object */
-                /* XXX Really? */
-                char *buf;
-                Py_ssize_t count = convertbuffer(arg, p, &buf);
-                if (count < 0)
-                    return converterr(buf, arg, msgbuf, bufsize);
-                STORE_SIZE(count);
-            }
-            format++;
-        } else {
-            char **p = va_arg(*p_va, char **);
-
-            if (PyUnicode_Check(arg)) {
-                uarg = UNICODE_DEFAULT_ENCODING(arg);
-                if (uarg == NULL)
-                    return converterr(CONV_UNICODE,
-                                      arg, msgbuf, bufsize);
-                *p = PyBytes_AS_STRING(uarg);
-            }
-            else
-                return converterr("string", arg, msgbuf, bufsize);
-            if ((Py_ssize_t) strlen(*p) != PyBytes_GET_SIZE(uarg))
-                return converterr("string without null bytes",
-                                  arg, msgbuf, bufsize);
-        }
-        break;
-    }
-
     case 'y': {/* any buffer-like object, but not PyUnicode */
         void **p = (void **)va_arg(*p_va, char **);
         char *buf;
@@ -948,11 +884,14 @@
         break;
     }
 
-    case 'z': {/* like 's' or 's#', but None is okay, stored as NULL */
+    case 's': /* text string */
+    case 'z': /* text string or None */
+    {
         if (*format == '*') {
+            /* "s*" or "z*" */
             Py_buffer *p = (Py_buffer *)va_arg(*p_va, Py_buffer *);
 
-            if (arg == Py_None)
+            if (c == 'z' && arg == Py_None)
                 PyBuffer_FillInfo(p, NULL, NULL, 0, 1, 0);
             else if (PyUnicode_Check(arg)) {
                 uarg = UNICODE_DEFAULT_ENCODING(arg);
@@ -975,11 +914,12 @@
             }
             format++;
         } else if (*format == '#') { /* any buffer-like object */
+            /* "s#" or "z#" */
             void **p = (void **)va_arg(*p_va, char **);
             FETCH_SIZE;
 
-            if (arg == Py_None) {
-                *p = 0;
+            if (c == 'z' && arg == Py_None) {
+                *p = NULL;
                 STORE_SIZE(0);
             }
             else if (PyUnicode_Check(arg)) {
@@ -1000,16 +940,12 @@
             }
             format++;
         } else {
+            /* "s" or "z" */
             char **p = va_arg(*p_va, char **);
             uarg = NULL;
 
-            if (arg == Py_None)
-                *p = 0;
-            else if (PyBytes_Check(arg)) {
-                /* Enable null byte check below */
-                uarg = arg;
-                *p = PyBytes_AS_STRING(arg);
-            }
+            if (c == 'z' && arg == Py_None)
+                *p = NULL;
             else if (PyUnicode_Check(arg)) {
                 uarg = UNICODE_DEFAULT_ENCODING(arg);
                 if (uarg == NULL)
@@ -1018,24 +954,28 @@
                 *p = PyBytes_AS_STRING(uarg);
             }
             else
-                return converterr("string or None",
+                return converterr(c == 'z' ? "str or None" : "str",
                                   arg, msgbuf, bufsize);
             if (*p != NULL && uarg != NULL &&
                 (Py_ssize_t) strlen(*p) != PyBytes_GET_SIZE(uarg))
                 return converterr(
-                    "string without null bytes or None",
+                    c == 'z' ? "str without null bytes or None"
+                             : "str without null bytes",
                     arg, msgbuf, bufsize);
         }
         break;
     }
 
-    case 'Z': {/* unicode, may be NULL (None) */
+    case 'u': /* raw unicode buffer (Py_UNICODE *) */
+    case 'Z': /* raw unicode buffer or None */
+    {
         if (*format == '#') { /* any buffer-like object */
+            /* "s#" or "Z#" */
             Py_UNICODE **p = va_arg(*p_va, Py_UNICODE **);
             FETCH_SIZE;
 
-            if (arg == Py_None) {
-                *p = 0;
+            if (c == 'Z' && arg == Py_None) {
+                *p = NULL;
                 STORE_SIZE(0);
             }
             else if (PyUnicode_Check(arg)) {
@@ -1046,10 +986,11 @@
                 return converterr("str or None", arg, msgbuf, bufsize);
             format++;
         } else {
+            /* "s" or "Z" */
             Py_UNICODE **p = va_arg(*p_va, Py_UNICODE **);
 
-            if (arg == Py_None)
-                *p = 0;
+            if (c == 'Z' && arg == Py_None)
+                *p = NULL;
             else if (PyUnicode_Check(arg)) {
                 *p = PyUnicode_AS_UNICODE(arg);
                 if (Py_UNICODE_strlen(*p) != PyUnicode_GET_SIZE(arg))
@@ -1057,7 +998,8 @@
                         "str without null character or None",
                         arg, msgbuf, bufsize);
             } else
-                return converterr("str or None", arg, msgbuf, bufsize);
+                return converterr(c == 'Z' ? "str or None" : "str",
+                                  arg, msgbuf, bufsize);
         }
         break;
     }
@@ -1227,24 +1169,6 @@
         break;
     }
 
-    case 'u': {/* raw unicode buffer (Py_UNICODE *) */
-        Py_UNICODE **p = va_arg(*p_va, Py_UNICODE **);
-        if (!PyUnicode_Check(arg))
-            return converterr("str", arg, msgbuf, bufsize);
-        *p = PyUnicode_AS_UNICODE(arg);
-        if (*format == '#') { /* store pointer and size */
-            FETCH_SIZE;
-            STORE_SIZE(PyUnicode_GET_SIZE(arg));
-            format++;
-        } else {
-            if (Py_UNICODE_strlen(*p) != PyUnicode_GET_SIZE(arg))
-                return converterr(
-                    "str without null character",
-                    arg, msgbuf, bufsize);
-        }
-        break;
-    }
-
     case 'S': { /* PyBytes object */
         PyObject **p = va_arg(*p_va, PyObject **);
         if (PyBytes_Check(arg))
@@ -1307,58 +1231,28 @@
     }
 
 
-    case 'w': { /* memory buffer, read-write access */
+    case 'w': { /* "w*": memory buffer, read-write access */
         void **p = va_arg(*p_va, void **);
-        PyBufferProcs *pb = arg->ob_type->tp_as_buffer;
-        Py_ssize_t count;
-        int temp=-1;
-        Py_buffer view;
 
-        if (pb && pb->bf_releasebuffer && *format != '*')
-            /* Buffer must be released, yet caller does not use
-               the Py_buffer protocol. */
-            return converterr("pinned buffer", arg, msgbuf, bufsize);
-
-
-        if (pb && pb->bf_getbuffer && *format == '*') {
-            /* Caller is interested in Py_buffer, and the object
-               supports it directly. */
-            format++;
-            if (PyObject_GetBuffer(arg, (Py_buffer*)p, PyBUF_WRITABLE) < 0) {
-                PyErr_Clear();
-                return converterr("read-write buffer", arg, msgbuf, bufsize);
-            }
-            if (addcleanup(p, freelist, cleanup_buffer)) {
-                return converterr(
-                    "(cleanup problem)",
-                    arg, msgbuf, bufsize);
-            }
-            if (!PyBuffer_IsContiguous((Py_buffer*)p, 'C'))
-                return converterr("contiguous buffer", arg, msgbuf, bufsize);
-            break;
-        }
+        if (*format != '*')
+            return converterr(
+                "invalid use of 'w' format character",
+                arg, msgbuf, bufsize);
+        format++;
 
-        /* Here we have processed w*, only w and w# remain. */
-        if (pb == NULL ||
-            pb->bf_getbuffer == NULL ||
-            ((temp = PyObject_GetBuffer(arg, &view,
-                                        PyBUF_SIMPLE)) != 0) ||
-            view.readonly == 1) {
-            if (temp==0) {
-                PyBuffer_Release(&view);
-            }
-            return converterr("single-segment read-write buffer",
-                              arg, msgbuf, bufsize);
+        /* Caller is interested in Py_buffer, and the object
+           supports it directly. */
+        if (PyObject_GetBuffer(arg, (Py_buffer*)p, PyBUF_WRITABLE) < 0) {
+            PyErr_Clear();
+            return converterr("read-write buffer", arg, msgbuf, bufsize);
         }
-
-        if ((count = view.len) < 0)
-            return converterr("(unspecified)", arg, msgbuf, bufsize);
-        *p = view.buf;
-        if (*format == '#') {
-            FETCH_SIZE;
-            STORE_SIZE(count);
-            format++;
+        if (addcleanup(p, freelist, cleanup_buffer)) {
+            return converterr(
+                "(cleanup problem)",
+                arg, msgbuf, bufsize);
         }
+        if (!PyBuffer_IsContiguous((Py_buffer*)p, 'C'))
+            return converterr("contiguous buffer", arg, msgbuf, bufsize);
         break;
     }
 
@@ -1416,6 +1310,7 @@
         return -1;
     }
     if (!PyBuffer_IsContiguous(view, 'C')) {
+        PyBuffer_Release(view);
         *errmsg = "contiguous buffer";
         return -1;
     }

Modified: python/branches/py3k-dtoa/Python/import.c
==============================================================================
--- python/branches/py3k-dtoa/Python/import.c	(original)
+++ python/branches/py3k-dtoa/Python/import.c	Tue Jul  6 17:55:27 2010
@@ -558,10 +558,10 @@
    copy can be retrieved from there by calling
    _PyImport_FindExtension().
 
-   Modules which do support multiple multiple initialization set
-   their m_size field to a non-negative number (indicating the size
-   of the module-specific state). They are still recorded in the
-   extensions dictionary, to avoid loading shared libraries twice.
+   Modules which do support multiple initialization set their m_size
+   field to a non-negative number (indicating the size of the
+   module-specific state). They are still recorded in the extensions
+   dictionary, to avoid loading shared libraries twice.
 */
 
 int
@@ -2385,7 +2385,8 @@
     if (parent == NULL)
         return NULL;
 
-    head = load_next(parent, Py_None, &name, buf, &buflen);
+    head = load_next(parent, level < 0 ? Py_None : parent, &name, buf,
+                        &buflen);
     if (head == NULL)
         return NULL;
 

Modified: python/branches/py3k-dtoa/Python/pythonrun.c
==============================================================================
--- python/branches/py3k-dtoa/Python/pythonrun.c	(original)
+++ python/branches/py3k-dtoa/Python/pythonrun.c	Tue Jul  6 17:55:27 2010
@@ -1190,7 +1190,7 @@
     d = PyModule_GetDict(m);
     if (PyDict_GetItemString(d, "__file__") == NULL) {
         PyObject *f;
-        f = PyUnicode_DecodeFSDefault(filename);
+        f = PyUnicode_FromString(filename);
         if (f == NULL)
             return -1;
         if (PyDict_SetItemString(d, "__file__", f) < 0) {

Modified: python/branches/py3k-dtoa/Python/sysmodule.c
==============================================================================
--- python/branches/py3k-dtoa/Python/sysmodule.c	(original)
+++ python/branches/py3k-dtoa/Python/sysmodule.c	Tue Jul  6 17:55:27 2010
@@ -595,7 +595,7 @@
     5,
 };
 
-PyObject *
+static PyObject *
 get_hash_info(void)
 {
     PyObject *hash_info;

Modified: python/branches/py3k-dtoa/Python/traceback.c
==============================================================================
--- python/branches/py3k-dtoa/Python/traceback.c	(original)
+++ python/branches/py3k-dtoa/Python/traceback.c	Tue Jul  6 17:55:27 2010
@@ -133,33 +133,38 @@
     return 0;
 }
 
-static int
-_Py_FindSourceFile(const char* filename, char* namebuf, size_t namelen, int open_flags)
+static PyObject *
+_Py_FindSourceFile(PyObject *filename, char* namebuf, size_t namelen, PyObject *io)
 {
-    int i;
-    int fd = -1;
+    Py_ssize_t i;
+    PyObject *binary;
     PyObject *v;
-    Py_ssize_t _npath;
-    int npath;
+    Py_ssize_t npath;
     size_t taillen;
     PyObject *syspath;
     const char* path;
     const char* tail;
+    const char* filepath;
     Py_ssize_t len;
 
+    filepath = _PyUnicode_AsString(filename);
+    if (filepath == NULL) {
+        PyErr_Clear();
+        return NULL;
+    }
+
     /* Search tail of filename in sys.path before giving up */
-    tail = strrchr(filename, SEP);
+    tail = strrchr(filepath, SEP);
     if (tail == NULL)
-        tail = filename;
+        tail = filepath;
     else
         tail++;
     taillen = strlen(tail);
 
     syspath = PySys_GetObject("path");
     if (syspath == NULL || !PyList_Check(syspath))
-        return -1;
-    _npath = PyList_Size(syspath);
-    npath = Py_SAFE_DOWNCAST(_npath, Py_ssize_t, int);
+        return NULL;
+    npath = PyList_Size(syspath);
 
     for (i = 0; i < npath; i++) {
         v = PyList_GetItem(syspath, i);
@@ -170,6 +175,10 @@
         if (!PyUnicode_Check(v))
             continue;
         path = _PyUnicode_AsStringAndSize(v, &len);
+        if (path == NULL) {
+            PyErr_Clear();
+            continue;
+        }
         if (len + 1 + (Py_ssize_t)taillen >= (Py_ssize_t)namelen - 1)
             continue; /* Too long */
         strcpy(namebuf, path);
@@ -178,31 +187,27 @@
         if (len > 0 && namebuf[len-1] != SEP)
             namebuf[len++] = SEP;
         strcpy(namebuf+len, tail);
-        Py_BEGIN_ALLOW_THREADS
-        fd = open(namebuf, open_flags);
-        Py_END_ALLOW_THREADS
-        if (0 <= fd) {
-            return fd;
-        }
+
+        binary = PyObject_CallMethod(io, "open", "ss", namebuf, "rb");
+        if (binary != NULL)
+            return binary;
+        PyErr_Clear();
     }
-    return -1;
+    return NULL;
 }
 
 int
-_Py_DisplaySourceLine(PyObject *f, const char *filename, int lineno, int indent)
+_Py_DisplaySourceLine(PyObject *f, PyObject *filename, int lineno, int indent)
 {
     int err = 0;
     int fd;
     int i;
     char *found_encoding;
     char *encoding;
+    PyObject *io;
+    PyObject *binary;
     PyObject *fob = NULL;
     PyObject *lineobj = NULL;
-#ifdef O_BINARY
-    const int open_flags = O_RDONLY | O_BINARY;   /* necessary for Windows */
-#else
-    const int open_flags = O_RDONLY;
-#endif
     char buf[MAXPATHLEN+1];
     Py_UNICODE *u, *p;
     Py_ssize_t len;
@@ -210,27 +215,32 @@
     /* open the file */
     if (filename == NULL)
         return 0;
-    Py_BEGIN_ALLOW_THREADS
-    fd = open(filename, open_flags);
-    Py_END_ALLOW_THREADS
-    if (fd < 0) {
-        fd = _Py_FindSourceFile(filename, buf, sizeof(buf), open_flags);
-        if (fd < 0)
+
+    io = PyImport_ImportModuleNoBlock("io");
+    if (io == NULL)
+        return -1;
+    binary = PyObject_CallMethod(io, "open", "Os", filename, "rb");
+
+    if (binary == NULL) {
+        binary = _Py_FindSourceFile(filename, buf, sizeof(buf), io);
+        if (binary == NULL) {
+            Py_DECREF(io);
             return 0;
-        filename = buf;
+        }
     }
 
     /* use the right encoding to decode the file as unicode */
+    fd = PyObject_AsFileDescriptor(binary);
     found_encoding = PyTokenizer_FindEncoding(fd);
-    encoding = (found_encoding != NULL) ? found_encoding :
-        (char*)PyUnicode_GetDefaultEncoding();
+    encoding = (found_encoding != NULL) ? found_encoding : "utf-8";
     lseek(fd, 0, 0); /* Reset position */
-    fob = PyFile_FromFd(fd, (char*)filename, "r", -1, (char*)encoding,
-        NULL, NULL, 1);
+    fob = PyObject_CallMethod(io, "TextIOWrapper", "Os", binary, encoding);
+    Py_DECREF(io);
+    Py_DECREF(binary);
     PyMem_FREE(found_encoding);
+
     if (fob == NULL) {
         PyErr_Clear();
-        close(fd);
         return 0;
     }
 
@@ -287,17 +297,19 @@
 }
 
 static int
-tb_displayline(PyObject *f, const char *filename, int lineno, const char *name)
+tb_displayline(PyObject *f, PyObject *filename, int lineno, PyObject *name)
 {
-    int err = 0;
-    char linebuf[2000];
+    int err;
+    PyObject *line;
 
     if (filename == NULL || name == NULL)
         return -1;
-    /* This is needed by Emacs' compile command */
-#define FMT "  File \"%.500s\", line %d, in %.500s\n"
-    PyOS_snprintf(linebuf, sizeof(linebuf), FMT, filename, lineno, name);
-    err = PyFile_WriteString(linebuf, f);
+    line = PyUnicode_FromFormat("  File \"%U\", line %d, in %U\n",
+                                filename, lineno, name);
+    if (line == NULL)
+        return -1;
+    err = PyFile_WriteObject(line, f, Py_PRINT_RAW);
+    Py_DECREF(line);
     if (err != 0)
         return err;
     return _Py_DisplaySourceLine(f, filename, lineno, 4);
@@ -316,10 +328,9 @@
     while (tb != NULL && err == 0) {
         if (depth <= limit) {
             err = tb_displayline(f,
-                _PyUnicode_AsString(
-                    tb->tb_frame->f_code->co_filename),
-                tb->tb_lineno,
-                _PyUnicode_AsString(tb->tb_frame->f_code->co_name));
+                                 tb->tb_frame->f_code->co_filename,
+                                 tb->tb_lineno,
+                                 tb->tb_frame->f_code->co_name);
         }
         depth--;
         tb = tb->tb_next;

Modified: python/branches/py3k-dtoa/README
==============================================================================
--- python/branches/py3k-dtoa/README	(original)
+++ python/branches/py3k-dtoa/README	Tue Jul  6 17:55:27 2010
@@ -172,13 +172,13 @@
 Release Schedule
 ----------------
 
-See PEP 375 for release details: http://www.python.org/dev/peps/pep-0375/
+See PEP 392 for release details: http://www.python.org/dev/peps/pep-0392/
 
 
 Copyright and License Information
 ---------------------------------
 
-Copyright (c) 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008
+Copyright (c) 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010
 Python Software Foundation.
 All rights reserved.
 

Modified: python/branches/py3k-dtoa/Tools/README
==============================================================================
--- python/branches/py3k-dtoa/Tools/README	(original)
+++ python/branches/py3k-dtoa/Tools/README	Tue Jul  6 17:55:27 2010
@@ -1,18 +1,7 @@
 This directory contains a number of Python programs that are useful
 while building or extending Python.
 
-audiopy		Audiopy is a program to control the Solaris audio
-		device, allowing you to choose both the input and
-		output devices, and to set the output volume, that can
-		be run either as a command-line script, or as a
-		Tkinter application.
-
-compiler	Tools used to maintain the compiler package in the
-		standard library.
-
-faqwiz		FAQ Wizard.
-		See http://www.python.org/cgi-bin/faqw.py
-		for a live example.
+faqwiz		FAQ Wizard: a CGI script for a user-editable FAQ.
 
 freeze		Create a stand-alone executable from a Python program.
 
@@ -36,8 +25,6 @@
 versioncheck	A tool to automate checking whether you have the latest
 		version of a package (by Jack Jansen).
 
-webchecker	A link checker for web sites.
-
 world		Script to take a list of Internet addresses and print
 		out where in the world those addresses originate from,
 		based on the top-level domain country code found in

Modified: python/branches/py3k-dtoa/Tools/scripts/patchcheck.py
==============================================================================
--- python/branches/py3k-dtoa/Tools/scripts/patchcheck.py	(original)
+++ python/branches/py3k-dtoa/Tools/scripts/patchcheck.py	Tue Jul  6 17:55:27 2010
@@ -5,6 +5,10 @@
 import reindent
 
 
+def n_files_str(count):
+    """Return 'N file(s)' with the proper plurality on 'file'."""
+    return "{} file{}".format(count, "s" if count != 1 else "")
+
 def status(message, modal=False, info=None):
     """Decorator to output status info to stdout."""
     def decorated_fxn(fxn):
@@ -26,7 +30,7 @@
     return decorated_fxn
 
 @status("Getting the list of files that have been added/changed",
-            info=lambda x: "%s files" % len(x))
+            info=lambda x: n_files_str(len(x)))
 def changed_files():
     """Run ``svn status`` and return a set of files that have been
     changed/added."""
@@ -44,20 +48,30 @@
             files.add(path)
     return files
 
- at status("Fixing whitespace", info=lambda x: "%s files" % x)
+def report_modified_files(file_paths):
+    count = len(file_paths)
+    if count == 0:
+        return n_files_str(count)
+    else:
+        lines = ["{}:".format(n_files_str(count))]
+        for path in file_paths:
+            lines.append("  {}".format(path))
+        return "\n".join(lines)
+
+ at status("Fixing whitespace", info=report_modified_files)
 def normalize_whitespace(file_paths):
     """Make sure that the whitespace for .py files have been normalized."""
     reindent.makebackup = False  # No need to create backups.
-    result = list(map(reindent.check, (x for x in file_paths if x.endswith('.py'))))
-    return sum(result)
+    fixed = []
+    for path in (x for x in file_paths if x.endswith('.py')):
+        if reindent.check(path):
+            fixed.append(path)
+    return fixed
 
 @status("Docs modified", modal=True)
 def docs_modified(file_paths):
-    """Report if any files in the Docs directory."""
-    for path in file_paths:
-        if path.startswith("Doc"):
-            return True
-    return False
+    """Report if any file in the Doc directory has been changed."""
+    return bool(file_paths)
 
 @status("Misc/ACKS updated", modal=True)
 def credit_given(file_paths):
@@ -72,14 +86,18 @@
 
 def main():
     file_paths = changed_files()
-    # PEP 7/8 verification.
-    normalize_whitespace(file_paths)
+    python_files = [fn for fn in file_paths if fn.endswith('.py')]
+    c_files = [fn for fn in file_paths if fn.endswith(('.c', '.h'))]
+    docs = [fn for fn in file_paths if fn.startswith('Doc')]
+    special_files = {'Misc/ACKS', 'Misc/NEWS'} & set(file_paths)
+    # PEP 8 whitespace rules enforcement.
+    normalize_whitespace(python_files)
     # Docs updated.
-    docs_modified(file_paths)
+    docs_modified(docs)
     # Misc/ACKS changed.
-    credit_given(file_paths)
+    credit_given(special_files)
     # Misc/NEWS changed.
-    reported_news(file_paths)
+    reported_news(special_files)
 
     # Test suite run and passed.
     print()

Modified: python/branches/py3k-dtoa/configure
==============================================================================
--- python/branches/py3k-dtoa/configure	(original)
+++ python/branches/py3k-dtoa/configure	Tue Jul  6 17:55:27 2010
@@ -1,5 +1,5 @@
 #! /bin/sh
-# From configure.in Revision: 81078 .
+# From configure.in Revision: 82090 .
 # Guess values for system-dependent variables and create Makefiles.
 # Generated by GNU Autoconf 2.65 for python 3.2.
 #

Modified: python/branches/py3k-dtoa/setup.py
==============================================================================
--- python/branches/py3k-dtoa/setup.py	(original)
+++ python/branches/py3k-dtoa/setup.py	Tue Jul  6 17:55:27 2010
@@ -46,7 +46,7 @@
     """
     Returns True if 'path' can be located in an OSX SDK
     """
-    return path.startswith('/usr/') or path.startswith('/System/')
+    return (path.startswith('/usr/') and not path.startswith('/usr/local')) or path.startswith('/System/')
 
 def find_file(filename, std_dirs, paths):
     """Searches for the directory where a given file is located,
@@ -450,9 +450,9 @@
                                depends=['_math.h'],
                                libraries=math_libs) )
         # time operations and variables
-        exts.append( Extension('time', ['timemodule.c'],
+        exts.append( Extension('time', ['timemodule.c', '_time.c'],
                                libraries=math_libs) )
-        exts.append( Extension('datetime', ['datetimemodule.c', 'timemodule.c'],
+        exts.append( Extension('datetime', ['datetimemodule.c', '_time.c'],
                                libraries=math_libs) )
         # fast iterator tools implemented in C
         exts.append( Extension("itertools", ["itertoolsmodule.c"]) )
@@ -862,6 +862,7 @@
                         else:
                             if os.path.isdir(dn):
                                 tmp.append(dn)
+                    db_dirs_to_check = tmp
 
                     db_dirs_to_check = tmp
 

From python-checkins at python.org  Tue Jul  6 20:31:05 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Tue,  6 Jul 2010 20:31:05 +0200 (CEST)
Subject: [Python-checkins] r82613 - python/branches/release31-maint
Message-ID: <20100706183105.D5C6D126452@mail.python.org>

Author: alexander.belopolsky
Date: Tue Jul  6 20:31:05 2010
New Revision: 82613

Log:
Blocked revisions 82514 via svnmerge

........
  r82514 | alexander.belopolsky | 2010-07-03 16:35:53 -0400 (Sat, 03 Jul 2010) | 3 lines
  
  Issue #9094: Make python -m pickletools disassemble pickles given in
  the command line.
........


Modified:
   python/branches/release31-maint/   (props changed)

From python-checkins at python.org  Tue Jul  6 20:52:42 2010
From: python-checkins at python.org (mark.dickinson)
Date: Tue,  6 Jul 2010 20:52:42 +0200 (CEST)
Subject: [Python-checkins] r82614 - in python/branches/py3k-dtoa:
	Lib/test/test_float.py Objects/floatobject.c configure
	configure.in pyconfig.h.in
Message-ID: <20100706185242.4A6C612646F@mail.python.org>

Author: mark.dickinson
Date: Tue Jul  6 20:52:42 2010
New Revision: 82614

Log:
Add float.__setround__ and float.__getround__ for rounding mode control (for test purposes only).

Modified:
   python/branches/py3k-dtoa/Lib/test/test_float.py
   python/branches/py3k-dtoa/Objects/floatobject.c
   python/branches/py3k-dtoa/configure
   python/branches/py3k-dtoa/configure.in
   python/branches/py3k-dtoa/pyconfig.h.in

Modified: python/branches/py3k-dtoa/Lib/test/test_float.py
==============================================================================
--- python/branches/py3k-dtoa/Lib/test/test_float.py	(original)
+++ python/branches/py3k-dtoa/Lib/test/test_float.py	Tue Jul  6 20:52:42 2010
@@ -16,6 +16,11 @@
                                          "requires __getformat__")
 requires_setformat = unittest.skipUnless(hasattr(float, "__setformat__"),
                                          "requires __setformat__")
+
+have_setround = hasattr(float, "__setround__")
+requires_setround = unittest.skipUnless(have_setround,
+                                        "requires __setround__")
+
 # decorator for skipping tests on non-IEEE 754 platforms
 requires_IEEE_754 = unittest.skipUnless(have_getformat and
     float.__getformat__("double").startswith("IEEE"),
@@ -382,6 +387,17 @@
             #self.assertTrue(0.0 > pow_op(-2.0, -1047) > -1e-315)
 
 
+ at requires_setround
+class RoundingModeTestCase(unittest.TestCase):
+    def test_roundtrip(self):
+        # just check that we can use __setround__ and __getround__
+        # to set and get rounding mode
+        modes = "tonearest", "upward", "downward", "towardzero"
+        for mode in modes:
+            float.__setround__(mode)
+            self.assertEqual(float.__getround__(), mode)
+
+
 @requires_setformat
 class FormatFunctionsTestCase(unittest.TestCase):
 
@@ -1264,6 +1280,7 @@
         RoundTestCase,
         InfNanTest,
         HexFloatTestCase,
+        RoundingModeTestCase,
         )
 
 if __name__ == '__main__':

Modified: python/branches/py3k-dtoa/Objects/floatobject.c
==============================================================================
--- python/branches/py3k-dtoa/Objects/floatobject.c	(original)
+++ python/branches/py3k-dtoa/Objects/floatobject.c	Tue Jul  6 20:52:42 2010
@@ -9,6 +9,10 @@
 
 #include <ctype.h>
 #include <float.h>
+#if defined(HAVE_FESETROUND) && defined(HAVE_FEGETROUND) && \
+    defined(HAVE_FENV_H)
+#include <fenv.h>
+#endif
 
 #undef MAX
 #undef MIN
@@ -1745,6 +1749,77 @@
 "Overrides the automatic determination of C-level floating point type.\n"
 "This affects how floats are converted to and from binary strings.");
 
+#if defined(HAVE_FESETROUND) && defined(HAVE_FEGETROUND) && \
+    defined(HAVE_FENV_H)
+
+static PyObject *
+float_setround(PyTypeObject *v, PyObject* arg)
+{
+    int mode, result;
+
+    if (!PyUnicode_Check(arg)) {
+        PyErr_SetString(PyExc_TypeError,
+            "mode should be a string");
+        return NULL;
+    }
+
+    if (!PyUnicode_CompareWithASCIIString(arg, "tonearest"))
+        mode = FE_TONEAREST;
+    else if (!PyUnicode_CompareWithASCIIString(arg, "downward"))
+        mode = FE_DOWNWARD;
+    else if (!PyUnicode_CompareWithASCIIString(arg, "upward"))
+        mode = FE_UPWARD;
+    else if (!PyUnicode_CompareWithASCIIString(arg, "towardzero"))
+        mode = FE_TOWARDZERO;
+    else {
+        PyErr_SetString(PyExc_ValueError,
+                        "mode should be one of 'tonearest', 'downward', "
+                        "'upward', or 'towardzero'");
+        return NULL;
+    }
+    result = fesetround(mode);
+    if (result) {
+        PyErr_SetString(PyExc_RuntimeError,
+                        "Failed to set rounding mode");
+        return NULL;
+    }
+    Py_RETURN_NONE;
+}
+
+PyDoc_STRVAR(float_setround_doc,
+"float.__setround__(mode) -> None\n"
+"\n"
+"You probably don't want to use this function.  It exists mainly to be\n"
+"used in Python's test suite.\n");
+
+static PyObject *
+float_getround(PyTypeObject *v)
+{
+    int i;
+    i = fegetround();
+
+    switch(i) {
+    case FE_TONEAREST:
+        return PyUnicode_FromString("tonearest");
+    case FE_DOWNWARD:
+        return PyUnicode_FromString("downward");
+    case FE_UPWARD:
+        return PyUnicode_FromString("upward");
+    case FE_TOWARDZERO:
+        return PyUnicode_FromString("towardzero");
+    default:
+        return PyUnicode_FromString("unknown");
+    }
+}
+
+PyDoc_STRVAR(float_getround_doc,
+"float.__getround__() -> str\n"
+"\n"
+"You probably don't want to use this function.  It exists mainly to be\n"
+"used in Python's test suite.\n");
+
+#endif  /* defined(HAVE_FESETROUND) && defined(HAVE_FEGETROUND) && ... */
+
 static PyObject *
 float_getzero(PyObject *v, void *closure)
 {
@@ -1798,6 +1873,13 @@
      METH_O|METH_CLASS,                 float_getformat_doc},
     {"__setformat__",           (PyCFunction)float_setformat,
      METH_VARARGS|METH_CLASS,           float_setformat_doc},
+#if defined(HAVE_FESETROUND) && defined(HAVE_FEGETROUND) && \
+    defined(HAVE_FENV_H)
+    {"__setround__", (PyCFunction)float_setround,
+     METH_O|METH_CLASS,           float_setround_doc},
+    {"__getround__", (PyCFunction)float_getround,
+     METH_NOARGS|METH_CLASS,           float_getround_doc},
+#endif
     {"__format__",          (PyCFunction)float__format__,
      METH_VARARGS,                  float__format__doc},
     {NULL,              NULL}           /* sentinel */

Modified: python/branches/py3k-dtoa/configure
==============================================================================
--- python/branches/py3k-dtoa/configure	(original)
+++ python/branches/py3k-dtoa/configure	Tue Jul  6 20:52:42 2010
@@ -1,5 +1,5 @@
 #! /bin/sh
-# From configure.in Revision: 82090 .
+# From configure.in Revision: 81583 .
 # Guess values for system-dependent variables and create Makefiles.
 # Generated by GNU Autoconf 2.65 for python 3.2.
 #
@@ -1929,11 +1929,11 @@
        cat confdefs.h - <<_ACEOF >conftest.$ac_ext
 /* end confdefs.h.  */
 $ac_includes_default
+	     enum { N = $2 / 2 - 1 };
 int
 main ()
 {
-static int test_array [1 - 2 * !(enum { N = $2 / 2 - 1 };
-	     0 < ($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1))];
+static int test_array [1 - 2 * !(0 < ($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1))];
 test_array [0] = 0
 
   ;
@@ -1944,11 +1944,11 @@
   cat confdefs.h - <<_ACEOF >conftest.$ac_ext
 /* end confdefs.h.  */
 $ac_includes_default
+	        enum { N = $2 / 2 - 1 };
 int
 main ()
 {
-static int test_array [1 - 2 * !(enum { N = $2 / 2 - 1 };
-		($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1)
+static int test_array [1 - 2 * !(($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1)
 		 < ($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 2))];
 test_array [0] = 0
 
@@ -5995,7 +5995,7 @@
 fi
 
 for ac_header in asm/types.h conio.h curses.h direct.h dlfcn.h errno.h \
-fcntl.h grp.h \
+fcntl.h fenv.h grp.h \
 ieeefp.h io.h langinfo.h libintl.h ncurses.h poll.h process.h pthread.h \
 shadow.h signal.h stdint.h stropts.h termios.h thread.h \
 unistd.h utime.h \
@@ -11818,6 +11818,19 @@
 fi
 done
 
+for ac_func in fesetround fegetround
+do :
+  as_ac_var=`$as_echo "ac_cv_func_$ac_func" | $as_tr_sh`
+ac_fn_c_check_func "$LINENO" "$ac_func" "$as_ac_var"
+eval as_val=\$$as_ac_var
+   if test "x$as_val" = x""yes; then :
+  cat >>confdefs.h <<_ACEOF
+#define `$as_echo "HAVE_$ac_func" | $as_tr_cpp` 1
+_ACEOF
+
+fi
+done
+
 ac_fn_c_check_decl "$LINENO" "isinf" "ac_cv_have_decl_isinf" "#include <math.h>
 "
 if test "x$ac_cv_have_decl_isinf" = x""yes; then :

Modified: python/branches/py3k-dtoa/configure.in
==============================================================================
--- python/branches/py3k-dtoa/configure.in	(original)
+++ python/branches/py3k-dtoa/configure.in	Tue Jul  6 20:52:42 2010
@@ -1299,7 +1299,7 @@
 # checks for header files
 AC_HEADER_STDC
 AC_CHECK_HEADERS(asm/types.h conio.h curses.h direct.h dlfcn.h errno.h \
-fcntl.h grp.h \
+fcntl.h fenv.h grp.h \
 ieeefp.h io.h langinfo.h libintl.h ncurses.h poll.h process.h pthread.h \
 shadow.h signal.h stdint.h stropts.h termios.h thread.h \
 unistd.h utime.h \
@@ -3433,6 +3433,7 @@
 
 AC_CHECK_FUNCS([acosh asinh atanh copysign erf erfc expm1 finite gamma])
 AC_CHECK_FUNCS([hypot lgamma log1p round tgamma])
+AC_CHECK_FUNCS([fesetround fegetround])
 AC_CHECK_DECLS([isinf, isnan, isfinite], [], [], [[#include <math.h>]])
 
 LIBS=$LIBS_SAVE

Modified: python/branches/py3k-dtoa/pyconfig.h.in
==============================================================================
--- python/branches/py3k-dtoa/pyconfig.h.in	(original)
+++ python/branches/py3k-dtoa/pyconfig.h.in	Tue Jul  6 20:52:42 2010
@@ -208,6 +208,15 @@
 /* Define if you have the 'fdatasync' function. */
 #undef HAVE_FDATASYNC
 
+/* Define to 1 if you have the `fegetround' function. */
+#undef HAVE_FEGETROUND
+
+/* Define to 1 if you have the <fenv.h> header file. */
+#undef HAVE_FENV_H
+
+/* Define to 1 if you have the `fesetround' function. */
+#undef HAVE_FESETROUND
+
 /* Define to 1 if you have the `finite' function. */
 #undef HAVE_FINITE
 

From python-checkins at python.org  Wed Jul  7 00:58:50 2010
From: python-checkins at python.org (georg.brandl)
Date: Wed,  7 Jul 2010 00:58:50 +0200 (CEST)
Subject: [Python-checkins] r82615 -
	python/branches/py3k/Doc/using/cmdline.rst
Message-ID: <20100706225850.71E8EEEA86@mail.python.org>

Author: georg.brandl
Date: Wed Jul  7 00:58:50 2010
New Revision: 82615

Log:
Fix typo.

Modified:
   python/branches/py3k/Doc/using/cmdline.rst

Modified: python/branches/py3k/Doc/using/cmdline.rst
==============================================================================
--- python/branches/py3k/Doc/using/cmdline.rst	(original)
+++ python/branches/py3k/Doc/using/cmdline.rst	Wed Jul  7 00:58:50 2010
@@ -61,7 +61,7 @@
 
 .. cmdoption:: -c <command>
 
-   Execute the Python code in *command*.  *command* can be one ore more
+   Execute the Python code in *command*.  *command* can be one or more
    statements separated by newlines, with significant leading whitespace as in
    normal module code.
 

From python-checkins at python.org  Wed Jul  7 01:18:01 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Wed,  7 Jul 2010 01:18:01 +0200 (CEST)
Subject: [Python-checkins] r82616 - in sandbox/branches/py3k-datetime:
	datetime.py test_datetime.py
Message-ID: <20100706231801.CBAA8EEA22@mail.python.org>

Author: alexander.belopolsky
Date: Wed Jul  7 01:18:01 2010
New Revision: 82616

Log:
Issue #9000: datetime.timezone objects now have eval-friendly repr.

Modified:
   sandbox/branches/py3k-datetime/datetime.py
   sandbox/branches/py3k-datetime/test_datetime.py

Modified: sandbox/branches/py3k-datetime/datetime.py
==============================================================================
--- sandbox/branches/py3k-datetime/datetime.py	(original)
+++ sandbox/branches/py3k-datetime/datetime.py	Wed Jul  7 01:18:01 2010
@@ -705,7 +705,16 @@
     # Conversions to string
 
     def __repr__(self):
-        "Convert to formal string, for repr()."
+        """Convert to formal string, for repr().
+
+        >>> dt = datetime(2010, 1, 1)
+        >>> repr(dt)
+        'datetime.datetime(2010, 1, 1, 0, 0)'
+
+        >>> dt = datetime(2010, 1, 1, tzinfo=timezone.utc)
+        >>> repr(dt)
+        'datetime.datetime(2010, 1, 1, 0, 0, tzinfo=datetime.timezone.utc)'
+        """
         return "%s(%d, %d, %d)" % ('datetime.' + self.__class__.__name__,
                                    self._year,
                                    self._month,
@@ -1524,7 +1533,7 @@
         return s
 
     def __repr__(self):
-        "Convert to formal string, for repr()."
+        """Convert to formal string, for repr()."""
         L = [self._year, self._month, self._day, # These are never zero
              self._hour, self._minute, self._second, self._microsecond]
         if L[-1] == 0:
@@ -1813,6 +1822,24 @@
     def __hash__(self):
         return hash(self._offset)
     
+    def __repr__(self):
+        """Convert to formal string, for repr().
+
+        >>> tz = timezone.utc
+        >>> repr(tz)
+        'datetime.timezone.utc'
+        >>> tz = timezone(timedelta(hours=-5), 'EST')
+        >>> repr(tz)
+        "datetime.timezone(datetime.timedelta(-1, 68400), 'EST')"
+        """
+        if self is self.utc:
+            return 'datetime.timezone.utc'
+        if self._name is None:
+            return "%s(%r)" % ('datetime.' + self.__class__.__name__,
+                               self._offset)
+        return "%s(%r, %r)" % ('datetime.' + self.__class__.__name__,
+                               self._offset, self._name)
+
     def __str__(self):
         return self.tzname(None)
 

Modified: sandbox/branches/py3k-datetime/test_datetime.py
==============================================================================
--- sandbox/branches/py3k-datetime/test_datetime.py	(original)
+++ sandbox/branches/py3k-datetime/test_datetime.py	Wed Jul  7 01:18:01 2010
@@ -153,6 +153,15 @@
                    timezone.min, timezone.max]:
             self.assertEqual(str(tz), tz.tzname(None))
 
+    def test_repr(self):
+        import datetime
+        for tz in [self.ACDT, self.EST, timezone.utc,
+                   timezone.min, timezone.max]:
+            # test round-trip
+            tzrep = repr(tz)
+            self.assertEqual(tz, eval(tzrep))
+
+
     def test_class_members(self):
         limit = timedelta(hours=23, minutes=59)
         self.assertEqual(timezone.utc.utcoffset(None), ZERO)

From python-checkins at python.org  Wed Jul  7 01:19:45 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Wed,  7 Jul 2010 01:19:45 +0200 (CEST)
Subject: [Python-checkins] r82617 - in python/branches/py3k:
	Lib/test/test_datetime.py Misc/NEWS Modules/datetimemodule.c
Message-ID: <20100706231945.A96EAEEA90@mail.python.org>

Author: alexander.belopolsky
Date: Wed Jul  7 01:19:45 2010
New Revision: 82617

Log:
Issue #9000: datetime.timezone objects now have eval-friendly repr.

Modified:
   python/branches/py3k/Lib/test/test_datetime.py
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/Modules/datetimemodule.c

Modified: python/branches/py3k/Lib/test/test_datetime.py
==============================================================================
--- python/branches/py3k/Lib/test/test_datetime.py	(original)
+++ python/branches/py3k/Lib/test/test_datetime.py	Wed Jul  7 01:19:45 2010
@@ -153,6 +153,15 @@
                    timezone.min, timezone.max]:
             self.assertEqual(str(tz), tz.tzname(None))
 
+    def test_repr(self):
+        import datetime
+        for tz in [self.ACDT, self.EST, timezone.utc,
+                   timezone.min, timezone.max]:
+            # test round-trip
+            tzrep = repr(tz)
+            self.assertEqual(tz, eval(tzrep))
+
+
     def test_class_members(self):
         limit = timedelta(hours=23, minutes=59)
         self.assertEqual(timezone.utc.utcoffset(None), ZERO)

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Wed Jul  7 01:19:45 2010
@@ -1394,6 +1394,8 @@
 Extension Modules
 -----------------
 
+- Issue #9000: datetime.timezone objects now have eval-friendly repr.
+
 - In the math module, correctly lookup __trunc__, __ceil__, and __floor__ as
   special methods.
 

Modified: python/branches/py3k/Modules/datetimemodule.c
==============================================================================
--- python/branches/py3k/Modules/datetimemodule.c	(original)
+++ python/branches/py3k/Modules/datetimemodule.c	Wed Jul  7 01:19:45 2010
@@ -780,6 +780,8 @@
     PyObject *name;
 } PyDateTime_TimeZone;
 
+PyObject *PyDateTime_TimeZone_UTC;
+
 /* Create new timezone instance checking offset range.  This
    function does not check the name argument.  Caller must assure
    that offset is a timedelta instance and name is either NULL
@@ -3380,6 +3382,24 @@
 }
 
 static PyObject *
+timezone_repr(PyDateTime_TimeZone *self)
+{
+    /* Note that although timezone is not subclassable, it is convenient
+       to use Py_TYPE(self)->tp_name here. */
+    const char *type_name = Py_TYPE(self)->tp_name;
+
+    if (((PyObject *)self) == PyDateTime_TimeZone_UTC)
+        return PyUnicode_FromFormat("%s.utc", type_name);
+
+    if (self->name == NULL)
+        return PyUnicode_FromFormat("%s(%R)", type_name, self->offset);
+
+    return PyUnicode_FromFormat("%s(%R, %R)", type_name, self->offset,
+                                self->name);
+}
+
+
+static PyObject *
 timezone_str(PyDateTime_TimeZone *self)
 {
     char buf[10];
@@ -3505,7 +3525,7 @@
     0,                                /* tp_getattr */
     0,                                /* tp_setattr */
     0,                                /* tp_reserved */
-    0,                                /* tp_repr */
+    (reprfunc)timezone_repr,          /* tp_repr */
     0,                                /* tp_as_number */
     0,                                /* tp_as_sequence */
     0,                                /* tp_as_mapping */
@@ -5260,7 +5280,7 @@
     Py_DECREF(delta);
     if (x == NULL || PyDict_SetItemString(d, "utc", x) < 0)
         return NULL;
-    Py_DECREF(x);
+    PyDateTime_TimeZone_UTC = x;
 
     delta = new_delta(-1, 60, 0, 1); /* -23:59 */
     if (delta == NULL)

From solipsis at pitrou.net  Wed Jul  7 01:23:58 2010
From: solipsis at pitrou.net (solipsis at pitrou.net)
Date: Wed,  7 Jul 2010 01:23:58 +0200 (CEST)
Subject: [Python-checkins] Daily py3k reference leaks (r82610): sum=0
Message-ID: <20100706232358.3C4571770A@ns6635.ovh.net>

py3k results for svn r82610 (hg cset c1eeefdcb39b)
--------------------------------------------------



Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/antoine/py3k/refleaks/reflognbauQS', '-x']

From python-checkins at python.org  Wed Jul  7 01:26:35 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Wed,  7 Jul 2010 01:26:35 +0200 (CEST)
Subject: [Python-checkins] r82618 - python/branches/release31-maint
Message-ID: <20100706232635.9643DC9C1@mail.python.org>

Author: alexander.belopolsky
Date: Wed Jul  7 01:26:35 2010
New Revision: 82618

Log:
Blocked revisions 82617 via svnmerge

........
  r82617 | alexander.belopolsky | 2010-07-06 19:19:45 -0400 (Tue, 06 Jul 2010) | 1 line
  
  Issue #9000: datetime.timezone objects now have eval-friendly repr.
........


Modified:
   python/branches/release31-maint/   (props changed)

From python-checkins at python.org  Wed Jul  7 01:44:27 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Wed,  7 Jul 2010 01:44:27 +0200 (CEST)
Subject: [Python-checkins] r82619 -
	sandbox/branches/py3k-datetime/datetime.py
Message-ID: <20100706234427.09F3FEEA01@mail.python.org>

Author: alexander.belopolsky
Date: Wed Jul  7 01:44:26 2010
New Revision: 82619

Log:
Use custom sentinel instead of varargs in timezone()

Modified:
   sandbox/branches/py3k-datetime/datetime.py

Modified: sandbox/branches/py3k-datetime/datetime.py
==============================================================================
--- sandbox/branches/py3k-datetime/datetime.py	(original)
+++ sandbox/branches/py3k-datetime/datetime.py	Wed Jul  7 01:44:26 2010
@@ -1784,18 +1784,13 @@
 class timezone(tzinfo):
     __slots__ = '_offset', '_name'
 
-    def __init__(self, offset, *args):
-        # Reproduce C behavior
-        n = len(args)
-        if n > 1:
-            raise TypeError("timezone() takes at most 2 arguments (%d given)"
-                            % n)
-        if n == 0:
+    # Sentinel value to disallow None
+    _NoneGiven = object()
+    def __init__(self, offset, name=_NoneGiven):
+        if name is self._NoneGiven:
             name = None
-        else:
-            name = args[0]
-            if not isinstance(name, str):
-                raise TypeError("name must be a string")
+        elif not isinstance(name, str):
+            raise TypeError("name must be a string")
         if isinstance(offset, timedelta):
             if self._minoffset <= offset <= self._maxoffset:
                 if (offset.microseconds != 0 or

From python-checkins at python.org  Wed Jul  7 13:03:50 2010
From: python-checkins at python.org (matthias.klose)
Date: Wed,  7 Jul 2010 13:03:50 +0200 (CEST)
Subject: [Python-checkins] r82620 -
	python/branches/release27-maint/Doc/whatsnew/2.7.rst
Message-ID: <20100707110350.A73D8EE9C4@mail.python.org>

Author: matthias.klose
Date: Wed Jul  7 13:03:50 2010
New Revision: 82620

Log:
Remove whatsnew item about --with-computed-gotos.


Modified:
   python/branches/release27-maint/Doc/whatsnew/2.7.rst

Modified: python/branches/release27-maint/Doc/whatsnew/2.7.rst
==============================================================================
--- python/branches/release27-maint/Doc/whatsnew/2.7.rst	(original)
+++ python/branches/release27-maint/Doc/whatsnew/2.7.rst	Wed Jul  7 13:03:50 2010
@@ -868,12 +868,6 @@
 
 Several performance enhancements have been added:
 
-.. * A new :program:`configure` option, :option:`--with-computed-gotos`,
-   compiles the main bytecode interpreter loop using a new dispatch
-   mechanism that gives speedups of up to 20%, depending on the system
-   and benchmark.  The new mechanism is only supported on certain
-   compilers, such as gcc, SunPro, and icc.
-
 * A new opcode was added to perform the initial setup for
   :keyword:`with` statements, looking up the :meth:`__enter__` and
   :meth:`__exit__` methods.  (Contributed by Benjamin Peterson.)

From python-checkins at python.org  Wed Jul  7 13:06:57 2010
From: python-checkins at python.org (matthias.klose)
Date: Wed,  7 Jul 2010 13:06:57 +0200 (CEST)
Subject: [Python-checkins] r82621 -
	python/branches/release31-maint/Doc/whatsnew/2.7.rst
Message-ID: <20100707110657.E1EBDEEA6A@mail.python.org>

Author: matthias.klose
Date: Wed Jul  7 13:06:57 2010
New Revision: 82621

Log:
Remove whatsnew 2.7 item about --with-computed-gotos.


Modified:
   python/branches/release31-maint/Doc/whatsnew/2.7.rst

Modified: python/branches/release31-maint/Doc/whatsnew/2.7.rst
==============================================================================
--- python/branches/release31-maint/Doc/whatsnew/2.7.rst	(original)
+++ python/branches/release31-maint/Doc/whatsnew/2.7.rst	Wed Jul  7 13:06:57 2010
@@ -158,12 +158,6 @@
 
 Several performance enhancements have been added:
 
-.. * A new :program:`configure` option, :option:`--with-computed-gotos`,
-   compiles the main bytecode interpreter loop using a new dispatch
-   mechanism that gives speedups of up to 20%, depending on the system
-   and benchmark.  The new mechanism is only supported on certain
-   compilers, such as gcc, SunPro, and icc.
-
 * The garbage collector now performs better when many objects are
   being allocated without deallocating any.  A full garbage collection
   pass is only performed when the middle generation has been collected

From ncoghlan at gmail.com  Wed Jul  7 15:46:06 2010
From: ncoghlan at gmail.com (Nick Coghlan)
Date: Wed, 7 Jul 2010 23:46:06 +1000
Subject: [Python-checkins] r82619 -
	sandbox/branches/py3k-datetime/datetime.py
In-Reply-To: <20100706234427.09F3FEEA01@mail.python.org>
References: <20100706234427.09F3FEEA01@mail.python.org>
Message-ID: <AANLkTill738flNZvTsLTO90yyOw4ELFvLgr3DnN9NH5_@mail.gmail.com>

On Wed, Jul 7, 2010 at 9:44 AM, alexander.belopolsky
<python-checkins at python.org> wrote:
> + ? ?# Sentinel value to disallow None
> + ? ?_NoneGiven = object()
> + ? ?def __init__(self, offset, name=_NoneGiven):
> + ? ? ? ?if name is self._NoneGiven:
> ? ? ? ? ? ? name = None

I suggest calling your sentinel value here something more like
"_NothingGiven", "_NotSupplied" or "_Omitted". The current name sounds
like it is meant to mean "None was given", when that is clearly not
the case.

Cheers,
Nick.

-- 
Nick Coghlan   |   ncoghlan at gmail.com   |   Brisbane, Australia

From python-checkins at python.org  Wed Jul  7 16:18:42 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Wed,  7 Jul 2010 16:18:42 +0200 (CEST)
Subject: [Python-checkins] r82622 -
	sandbox/branches/py3k-datetime/datetime.py
Message-ID: <20100707141842.57C4DEE98A@mail.python.org>

Author: alexander.belopolsky
Date: Wed Jul  7 16:18:42 2010
New Revision: 82622

Log:
Renamed the sentinel from  _NoneGiven to _Omitted.  Thanks Nick Coghlan.

Modified:
   sandbox/branches/py3k-datetime/datetime.py

Modified: sandbox/branches/py3k-datetime/datetime.py
==============================================================================
--- sandbox/branches/py3k-datetime/datetime.py	(original)
+++ sandbox/branches/py3k-datetime/datetime.py	Wed Jul  7 16:18:42 2010
@@ -1785,9 +1785,9 @@
     __slots__ = '_offset', '_name'
 
     # Sentinel value to disallow None
-    _NoneGiven = object()
-    def __init__(self, offset, name=_NoneGiven):
-        if name is self._NoneGiven:
+    _Omitted = object()
+    def __init__(self, offset, name=_Omitted):
+        if name is self._Omitted:
             name = None
         elif not isinstance(name, str):
             raise TypeError("name must be a string")

From python-checkins at python.org  Wed Jul  7 17:51:17 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Wed,  7 Jul 2010 17:51:17 +0200 (CEST)
Subject: [Python-checkins] r82623 -
	python/branches/release27-maint/Lib/test/test_math.py
Message-ID: <20100707155117.D5417EEB03@mail.python.org>

Author: benjamin.peterson
Date: Wed Jul  7 17:51:17 2010
New Revision: 82623

Log:
these tests are the result of cpython's incorrect implementation

Modified:
   python/branches/release27-maint/Lib/test/test_math.py

Modified: python/branches/release27-maint/Lib/test/test_math.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_math.py	(original)
+++ python/branches/release27-maint/Lib/test/test_math.py	Wed Jul  7 17:51:17 2010
@@ -879,11 +879,6 @@
         self.assertRaises((AttributeError, TypeError), math.trunc,
                           TestNoTrunc())
 
-        t = TestNoTrunc()
-        t.__trunc__ = lambda *args: args
-        self.assertEquals((), math.trunc(t))
-        self.assertRaises(TypeError, math.trunc, t, 0)
-
     def testIsnan(self):
         self.assertTrue(math.isnan(float("nan")))
         self.assertTrue(math.isnan(float("inf")* 0.))

From python-checkins at python.org  Wed Jul  7 18:10:44 2010
From: python-checkins at python.org (mark.dickinson)
Date: Wed,  7 Jul 2010 18:10:44 +0200 (CEST)
Subject: [Python-checkins] r82624 - python/branches/py3k/Grammar/Grammar
Message-ID: <20100707161044.AD7FEEEAF6@mail.python.org>

Author: mark.dickinson
Date: Wed Jul  7 18:10:44 2010
New Revision: 82624

Log:
Kill gratuitous space.

Modified:
   python/branches/py3k/Grammar/Grammar

Modified: python/branches/py3k/Grammar/Grammar
==============================================================================
--- python/branches/py3k/Grammar/Grammar	(original)
+++ python/branches/py3k/Grammar/Grammar	Wed Jul  7 18:10:44 2010
@@ -29,7 +29,7 @@
                 | tfpdef ['=' test] (',' tfpdef ['=' test])* [','])
 tfpdef: NAME [':' test]
 varargslist: ((vfpdef ['=' test] ',')*
-              ('*' [vfpdef] (',' vfpdef ['=' test])*  [',' '**' vfpdef] | '**' vfpdef)
+              ('*' [vfpdef] (',' vfpdef ['=' test])* [',' '**' vfpdef] | '**' vfpdef)
               | vfpdef ['=' test] (',' vfpdef ['=' test])* [','])
 vfpdef: NAME
 

From python-checkins at python.org  Wed Jul  7 18:17:31 2010
From: python-checkins at python.org (mark.dickinson)
Date: Wed,  7 Jul 2010 18:17:31 +0200 (CEST)
Subject: [Python-checkins] r82625 - python/branches/py3k/Modules/mathmodule.c
Message-ID: <20100707161731.65B33EEB0F@mail.python.org>

Author: mark.dickinson
Date: Wed Jul  7 18:17:31 2010
New Revision: 82625

Log:
Minor refactoring in lgamma code, for clarity.

Modified:
   python/branches/py3k/Modules/mathmodule.c

Modified: python/branches/py3k/Modules/mathmodule.c
==============================================================================
--- python/branches/py3k/Modules/mathmodule.c	(original)
+++ python/branches/py3k/Modules/mathmodule.c	Wed Jul  7 18:17:31 2010
@@ -69,6 +69,7 @@
 
 static const double pi = 3.141592653589793238462643383279502884197;
 static const double sqrtpi = 1.772453850905516027298167483341145182798;
+static const double logpi = 1.144729885849400174143427351353058711647;
 
 static double
 sinpi(double x)
@@ -356,20 +357,15 @@
     if (absx < 1e-20)
         return -log(absx);
 
-    /* Lanczos' formula */
-    if (x > 0.0) {
-        /* we could save a fraction of a ulp in accuracy by having a
-           second set of numerator coefficients for lanczos_sum that
-           absorbed the exp(-lanczos_g) term, and throwing out the
-           lanczos_g subtraction below; it's probably not worth it. */
-        r = log(lanczos_sum(x)) - lanczos_g +
-            (x-0.5)*(log(x+lanczos_g-0.5)-1);
-    }
-    else {
-        r = log(pi) - log(fabs(sinpi(absx))) - log(absx) -
-            (log(lanczos_sum(absx)) - lanczos_g +
-             (absx-0.5)*(log(absx+lanczos_g-0.5)-1));
-    }
+    /* Lanczos' formula.  We could save a fraction of a ulp in accuracy by
+       having a second set of numerator coefficients for lanczos_sum that
+       absorbed the exp(-lanczos_g) term, and throwing out the lanczos_g
+       subtraction below; it's probably not worth it. */
+    r = log(lanczos_sum(absx)) - lanczos_g;
+    r += (absx - 0.5) * (log(absx + lanczos_g - 0.5) - 1);
+    if (x < 0.0)
+        /* Use reflection formula to get value for negative x. */
+        r = logpi - log(fabs(sinpi(absx))) - log(absx) - r;
     if (Py_IS_INFINITY(r))
         errno = ERANGE;
     return r;

From python-checkins at python.org  Wed Jul  7 18:21:29 2010
From: python-checkins at python.org (mark.dickinson)
Date: Wed,  7 Jul 2010 18:21:29 +0200 (CEST)
Subject: [Python-checkins] r82626 - in python/branches/py3k:
	Lib/test/math_testcases.txt Lib/test/test_math.py Misc/NEWS
	Modules/mathmodule.c
Message-ID: <20100707162129.96D7EEE993@mail.python.org>

Author: mark.dickinson
Date: Wed Jul  7 18:21:29 2010
New Revision: 82626

Log:
Issue #9186:  log1p(-1.0) should raise ValueError, not OverflowError.

Modified:
   python/branches/py3k/Lib/test/math_testcases.txt
   python/branches/py3k/Lib/test/test_math.py
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/Modules/mathmodule.c

Modified: python/branches/py3k/Lib/test/math_testcases.txt
==============================================================================
--- python/branches/py3k/Lib/test/math_testcases.txt	(original)
+++ python/branches/py3k/Lib/test/math_testcases.txt	Wed Jul  7 18:21:29 2010
@@ -371,6 +371,78 @@
 
 
 -----------------------------------------------------------
+-- log1p: log(1 + x), without precision loss for small x --
+-----------------------------------------------------------
+
+-- special values
+log1p0000 log1p 0.0 -> 0.0
+log1p0001 log1p -0.0 -> -0.0
+log1p0002 log1p inf -> inf
+log1p0003 log1p -inf -> nan             invalid
+log1p0004 log1p nan -> nan
+
+-- singularity at -1.0
+log1p0010 log1p -1.0 -> -inf            divide-by-zero
+log1p0011 log1p -0.9999999999999999 -> -36.736800569677101
+
+-- finite values < 1.0 are invalid
+log1p0020 log1p -1.0000000000000002 -> nan invalid
+log1p0021 log1p -1.1 -> nan invalid
+log1p0022 log1p -2.0 -> nan invalid
+log1p0023 log1p -1e300 -> nan invalid
+
+-- tiny x: log1p(x) ~ x
+log1p0110 log1p 5e-324 -> 5e-324
+log1p0111 log1p 1e-320 -> 1e-320
+log1p0112 log1p 1e-300 -> 1e-300
+log1p0113 log1p 1e-150 -> 1e-150
+log1p0114 log1p 1e-20 -> 1e-20
+
+log1p0120 log1p -5e-324 -> -5e-324
+log1p0121 log1p -1e-320 -> -1e-320
+log1p0122 log1p -1e-300 -> -1e-300
+log1p0123 log1p -1e-150 -> -1e-150
+log1p0124 log1p -1e-20 -> -1e-20
+
+-- some (mostly) random small and moderate-sized values
+log1p0200 log1p -0.89156889782277482 -> -2.2216403106762863
+log1p0201 log1p -0.23858496047770464 -> -0.27257668276980057
+log1p0202 log1p -0.011641726191307515 -> -0.011710021654495657
+log1p0203 log1p -0.0090126398571693817 -> -0.0090534993825007650
+log1p0204 log1p -0.00023442805985712781 -> -0.00023445554240995693
+log1p0205 log1p -1.5672870980936349e-5 -> -1.5672993801662046e-5
+log1p0206 log1p -7.9650013274825295e-6 -> -7.9650330482740401e-6
+log1p0207 log1p -2.5202948343227410e-7 -> -2.5202951519170971e-7
+log1p0208 log1p -8.2446372820745855e-11 -> -8.2446372824144559e-11
+log1p0209 log1p -8.1663670046490789e-12 -> -8.1663670046824230e-12
+log1p0210 log1p 7.0351735084656292e-18 -> 7.0351735084656292e-18
+log1p0211 log1p 5.2732161907375226e-12 -> 5.2732161907236188e-12
+log1p0212 log1p 1.0000000000000000e-10 -> 9.9999999995000007e-11
+log1p0213 log1p 2.1401273266000197e-9 -> 2.1401273243099470e-9
+log1p0214 log1p 1.2668914653979560e-8 -> 1.2668914573728861e-8
+log1p0215 log1p 1.6250007816299069e-6 -> 1.6249994613175672e-6
+log1p0216 log1p 8.3740495645839399e-6 -> 8.3740145024266269e-6
+log1p0217 log1p 3.0000000000000001e-5 -> 2.9999550008999799e-5
+log1p0218 log1p 0.0070000000000000001 -> 0.0069756137364252423
+log1p0219 log1p 0.013026235315053002 -> 0.012942123564008787
+log1p0220 log1p 0.013497160797236184 -> 0.013406885521915038
+log1p0221 log1p 0.027625599078135284 -> 0.027250897463483054
+log1p0222 log1p 0.14179687245544870 -> 0.13260322540908789
+
+-- large values
+log1p0300 log1p 1.7976931348623157e+308 -> 709.78271289338397
+log1p0301 log1p 1.0000000000000001e+300 -> 690.77552789821368
+log1p0302 log1p 1.0000000000000001e+70 -> 161.18095650958321
+log1p0303 log1p 10000000000.000000 -> 23.025850930040455
+
+-- other values transferred from testLog1p in test_math
+log1p0400 log1p -0.63212055882855767 -> -1.0000000000000000
+log1p0401 log1p 1.7182818284590451 -> 1.0000000000000000
+log1p0402 log1p 1.0000000000000000 -> 0.69314718055994529
+log1p0403 log1p 1.2379400392853803e+27 -> 62.383246250395075
+
+
+-----------------------------------------------------------
 -- expm1: exp(x) - 1, without precision loss for small x --
 -----------------------------------------------------------
 

Modified: python/branches/py3k/Lib/test/test_math.py
==============================================================================
--- python/branches/py3k/Lib/test/test_math.py	(original)
+++ python/branches/py3k/Lib/test/test_math.py	Wed Jul  7 18:21:29 2010
@@ -647,15 +647,7 @@
 
     def testLog1p(self):
         self.assertRaises(TypeError, math.log1p)
-        self.ftest('log1p(1/e -1)', math.log1p(1/math.e-1), -1)
-        self.ftest('log1p(0)', math.log1p(0), 0)
-        self.ftest('log1p(e-1)', math.log1p(math.e-1), 1)
-        self.ftest('log1p(1)', math.log1p(1), math.log(2))
-        self.assertEquals(math.log1p(INF), INF)
-        self.assertRaises(ValueError, math.log1p, NINF)
-        self.assertTrue(math.isnan(math.log1p(NAN)))
         n= 2**90
-        self.assertAlmostEquals(math.log1p(n), 62.383246250395075)
         self.assertAlmostEquals(math.log1p(n), math.log1p(float(n)))
 
     def testLog10(self):

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Wed Jul  7 18:21:29 2010
@@ -468,6 +468,8 @@
 Library
 -------
 
+- Issue #9186: Fix math.log1p(-1.0) to raise ValueError, not OverflowError.
+
 - Issue #9130: Fix validation of relative imports in parser module.
 
 - Issue #9128: Fix validation of class decorators in parser module.

Modified: python/branches/py3k/Modules/mathmodule.c
==============================================================================
--- python/branches/py3k/Modules/mathmodule.c	(original)
+++ python/branches/py3k/Modules/mathmodule.c	Wed Jul  7 18:21:29 2010
@@ -896,7 +896,7 @@
       "gamma(x)\n\nGamma function at x.")
 FUNC1A(lgamma, m_lgamma,
       "lgamma(x)\n\nNatural logarithm of absolute value of Gamma function at x.")
-FUNC1(log1p, m_log1p, 1,
+FUNC1(log1p, m_log1p, 0,
       "log1p(x)\n\nReturn the natural logarithm of 1+x (base e).\n"
       "The result is computed in a way which is accurate for x near zero.")
 FUNC1(sin, sin, 0,

From python-checkins at python.org  Wed Jul  7 19:59:55 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Wed,  7 Jul 2010 19:59:55 +0200 (CEST)
Subject: [Python-checkins] r82627 -
	sandbox/branches/py3k-datetime/datetime.py
Message-ID: <20100707175955.7E101EE981@mail.python.org>

Author: alexander.belopolsky
Date: Wed Jul  7 19:59:55 2010
New Revision: 82627

Log:
Whitespace cleanup

Modified:
   sandbox/branches/py3k-datetime/datetime.py

Modified: sandbox/branches/py3k-datetime/datetime.py
==============================================================================
--- sandbox/branches/py3k-datetime/datetime.py	(original)
+++ sandbox/branches/py3k-datetime/datetime.py	Wed Jul  7 19:59:55 2010
@@ -13,7 +13,7 @@
 Sources for time zone and DST data: http://www.twinsun.com/tz/tz-link.htm
 
 This was originally copied from the sandbox of the CPython CVS repository.
-Thanks to Tim Peters for suggesting using it. 
+Thanks to Tim Peters for suggesting using it.
 """
 
 import time as _time
@@ -457,7 +457,7 @@
 
     def total_seconds(self):
         """Total seconds in the duration."""
-        return ((self.days * 86400 + self.seconds)*10**6 + 
+        return ((self.days * 86400 + self.seconds)*10**6 +
                 self.microseconds) / 10**6
 
     # Read-only field accessors
@@ -564,7 +564,7 @@
                           other._to_microseconds())
             return q, timedelta(0, 0, r)
         return NotImplemented
-        
+
     # Comparisons of timedelta objects with other.
 
     def __eq__(self, other):
@@ -1044,7 +1044,7 @@
         return self._hour
 
     @property
-    def minute(self): 
+    def minute(self):
         """minute (0-59)"""
         return self._minute
 
@@ -1052,7 +1052,7 @@
     def second(self):
         """second (0-59)"""
         return self._second
-    
+
     @property
     def microsecond(self):
         """microsecond (0-999999)"""
@@ -1332,7 +1332,7 @@
         return self._hour
 
     @property
-    def minute(self): 
+    def minute(self):
         """minute (0-59)"""
         return self._minute
 
@@ -1340,7 +1340,7 @@
     def second(self):
         """second (0-59)"""
         return self._second
-    
+
     @property
     def microsecond(self):
         """microsecond (0-999999)"""
@@ -1539,7 +1539,7 @@
         if L[-1] == 0:
             del L[-1]
         if L[-1] == 0:
-            del L[-1]            
+            del L[-1]
         s = ", ".join(map(str, L))
         s = "%s(%s)" % ('datetime.' + self.__class__.__name__, s)
         if self._tzinfo is not None:
@@ -1813,10 +1813,10 @@
 
     def __eq__(self, other):
         return self._offset == other._offset
-    
+
     def __hash__(self):
         return hash(self._offset)
-    
+
     def __repr__(self):
         """Convert to formal string, for repr().
 
@@ -1857,7 +1857,7 @@
             return None
         raise TypeError("dst() argument must be a datetime instance"
                         " or None")
-    
+
     def fromutc(self, dt):
         if isinstance(dt, datetime):
             if dt.tzinfo is not self:
@@ -2082,4 +2082,3 @@
 perverse time zone returns a negative dst()).  So a breaking case must be
 pretty bizarre, and a tzinfo subclass can override fromutc() if it is.
 """
-

From python-checkins at python.org  Wed Jul  7 20:44:05 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Wed,  7 Jul 2010 20:44:05 +0200 (CEST)
Subject: [Python-checkins] r82628 -
	python/branches/py3k/Lib/test/test_struct.py
Message-ID: <20100707184405.5FB81EEB10@mail.python.org>

Author: benjamin.peterson
Date: Wed Jul  7 20:44:05 2010
New Revision: 82628

Log:
this needn't be in the loop

Modified:
   python/branches/py3k/Lib/test/test_struct.py

Modified: python/branches/py3k/Lib/test/test_struct.py
==============================================================================
--- python/branches/py3k/Lib/test/test_struct.py	(original)
+++ python/branches/py3k/Lib/test/test_struct.py	Wed Jul  7 20:44:05 2010
@@ -503,8 +503,8 @@
                 self.assertFalse(prefix, msg='encoded bool is not one byte: %r'
                                              %packed)
 
-            for c in [b'\x01', b'\x7f', b'\xff', b'\x0f', b'\xf0']:
-                self.assertTrue(struct.unpack('>?', c)[0])
+        for c in [b'\x01', b'\x7f', b'\xff', b'\x0f', b'\xf0']:
+            self.assertTrue(struct.unpack('>?', c)[0])
 
     def test_count_overflow(self):
         hugecount = '{}b'.format(sys.maxsize+1)

From python-checkins at python.org  Wed Jul  7 20:51:44 2010
From: python-checkins at python.org (georg.brandl)
Date: Wed,  7 Jul 2010 20:51:44 +0200 (CEST)
Subject: [Python-checkins] r82629 -
	python/branches/py3k/Doc/distutils/sourcedist.rst
Message-ID: <20100707185144.21FA9EE9EA@mail.python.org>

Author: georg.brandl
Date: Wed Jul  7 20:51:43 2010
New Revision: 82629

Log:
Make comment out of an awkward note.

Modified:
   python/branches/py3k/Doc/distutils/sourcedist.rst

Modified: python/branches/py3k/Doc/distutils/sourcedist.rst
==============================================================================
--- python/branches/py3k/Doc/distutils/sourcedist.rst	(original)
+++ python/branches/py3k/Doc/distutils/sourcedist.rst	Wed Jul  7 20:51:43 2010
@@ -76,10 +76,10 @@
   :option:`packages` options
 
 * all C source files mentioned in the :option:`ext_modules` or
-  :option:`libraries` options (
+  :option:`libraries` options
 
-  **\*\*** getting C library sources currently broken---no
-  :meth:`get_source_files` method in :file:`build_clib.py`! **\*\***)
+  .. XXX Getting C library sources is currently broken -- no
+     :meth:`get_source_files` method in :file:`build_clib.py`!
 
 * scripts identified by the :option:`scripts` option
   See :ref:`distutils-installing-scripts`.

From python-checkins at python.org  Wed Jul  7 20:54:59 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Wed,  7 Jul 2010 20:54:59 +0200 (CEST)
Subject: [Python-checkins] r82630 - in python/branches/py3k:
	Lib/test/test_struct.py Modules/_struct.c
Message-ID: <20100707185459.B97A6EEAFE@mail.python.org>

Author: benjamin.peterson
Date: Wed Jul  7 20:54:59 2010
New Revision: 82630

Log:
don't ignore exceptions from PyObject_IsTrue

Modified:
   python/branches/py3k/Lib/test/test_struct.py
   python/branches/py3k/Modules/_struct.c

Modified: python/branches/py3k/Lib/test/test_struct.py
==============================================================================
--- python/branches/py3k/Lib/test/test_struct.py	(original)
+++ python/branches/py3k/Lib/test/test_struct.py	Wed Jul  7 20:54:59 2010
@@ -475,6 +475,9 @@
             self.assertEqual(value, 0x12345678)
 
     def test_bool(self):
+        class ExplodingBool(object):
+            def __bool__(self):
+                raise IOError
         for prefix in tuple("<>!=")+('',):
             false = (), [], [], '', 0
             true = [1], 'test', 5, -1, 0xffffffff+1, 0xffffffff/2
@@ -503,6 +506,9 @@
                 self.assertFalse(prefix, msg='encoded bool is not one byte: %r'
                                              %packed)
 
+            self.assertRaises(IOError, struct.pack, prefix + '?',
+                              ExplodingBool())
+
         for c in [b'\x01', b'\x7f', b'\xff', b'\x0f', b'\xf0']:
             self.assertTrue(struct.unpack('>?', c)[0])
 

Modified: python/branches/py3k/Modules/_struct.c
==============================================================================
--- python/branches/py3k/Modules/_struct.c	(original)
+++ python/branches/py3k/Modules/_struct.c	Wed Jul  7 20:54:59 2010
@@ -591,9 +591,13 @@
 static int
 np_bool(char *p, PyObject *v, const formatdef *f)
 {
-    BOOL_TYPE y;
+    int y;
+    BOOL_TYPE x;
     y = PyObject_IsTrue(v);
-    memcpy(p, (char *)&y, sizeof y);
+    if (y < 0)
+        return -1;
+    x = y;
+    memcpy(p, (char *)&x, sizeof x);
     return 0;
 }
 
@@ -865,6 +869,8 @@
 {
     char y;
     y = PyObject_IsTrue(v);
+    if (y < 0)
+        return -1;
     memcpy(p, (char *)&y, sizeof y);
     return 0;
 }

From python-checkins at python.org  Wed Jul  7 21:03:36 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Wed,  7 Jul 2010 21:03:36 +0200 (CEST)
Subject: [Python-checkins] r82631 - in python/branches/release27-maint:
	Lib/test/test_struct.py Modules/_struct.c
Message-ID: <20100707190336.8AC3FC988@mail.python.org>

Author: benjamin.peterson
Date: Wed Jul  7 21:03:36 2010
New Revision: 82631

Log:
Merged revisions 82628,82630 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82628 | benjamin.peterson | 2010-07-07 13:44:05 -0500 (Wed, 07 Jul 2010) | 1 line
  
  this needn't be in the loop
........
  r82630 | benjamin.peterson | 2010-07-07 13:54:59 -0500 (Wed, 07 Jul 2010) | 1 line
  
  don't ignore exceptions from PyObject_IsTrue
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/test/test_struct.py
   python/branches/release27-maint/Modules/_struct.c

Modified: python/branches/release27-maint/Lib/test/test_struct.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_struct.py	(original)
+++ python/branches/release27-maint/Lib/test/test_struct.py	Wed Jul  7 21:03:36 2010
@@ -491,6 +491,9 @@
             self.test_unpack_from(cls=buffer)
 
     def test_bool(self):
+        class ExplodingBool(object):
+            def __nonzero__(self):
+                raise IOError
         for prefix in tuple("<>!=")+('',):
             false = (), [], [], '', 0
             true = [1], 'test', 5, -1, 0xffffffffL+1, 0xffffffff//2
@@ -519,8 +522,11 @@
                 self.assertFalse(prefix, msg='encoded bool is not one byte: %r'
                                              %packed)
 
-            for c in '\x01\x7f\xff\x0f\xf0':
-                self.assertTrue(struct.unpack('>?', c)[0])
+            self.assertRaises(IOError, struct.pack, prefix + '?',
+                              ExplodingBool())
+
+        for c in [b'\x01', b'\x7f', b'\xff', b'\x0f', b'\xf0']:
+            self.assertTrue(struct.unpack('>?', c)[0])
 
     @unittest.skipUnless(IS32BIT, "Specific to 32bit machines")
     def test_crasher(self):

Modified: python/branches/release27-maint/Modules/_struct.c
==============================================================================
--- python/branches/release27-maint/Modules/_struct.c	(original)
+++ python/branches/release27-maint/Modules/_struct.c	Wed Jul  7 21:03:36 2010
@@ -636,9 +636,13 @@
 static int
 np_bool(char *p, PyObject *v, const formatdef *f)
 {
-    BOOL_TYPE y;
+    int y;
+    BOOL_TYPE x;
     y = PyObject_IsTrue(v);
-    memcpy(p, (char *)&y, sizeof y);
+    if (y < 0)
+        return -1;
+    x = y;
+    memcpy(p, (char *)&x, sizeof x);
     return 0;
 }
 
@@ -910,6 +914,8 @@
 {
     char y;
     y = PyObject_IsTrue(v);
+    if (y < 0)
+        return -1;
     memcpy(p, (char *)&y, sizeof y);
     return 0;
 }

From python-checkins at python.org  Wed Jul  7 21:04:36 2010
From: python-checkins at python.org (georg.brandl)
Date: Wed,  7 Jul 2010 21:04:36 +0200 (CEST)
Subject: [Python-checkins] r82632 - in python/branches/py3k/Doc/distutils:
	apiref.rst builtdist.rst setupscript.rst
Message-ID: <20100707190436.C43A6EE993@mail.python.org>

Author: georg.brandl
Date: Wed Jul  7 21:04:36 2010
New Revision: 82632

Log:
Turn more notes into comments.

Modified:
   python/branches/py3k/Doc/distutils/apiref.rst
   python/branches/py3k/Doc/distutils/builtdist.rst
   python/branches/py3k/Doc/distutils/setupscript.rst

Modified: python/branches/py3k/Doc/distutils/apiref.rst
==============================================================================
--- python/branches/py3k/Doc/distutils/apiref.rst	(original)
+++ python/branches/py3k/Doc/distutils/apiref.rst	Wed Jul  7 21:04:36 2010
@@ -995,7 +995,7 @@
    errors are ignored (apart from being reported to ``sys.stdout`` if *verbose* is
    true).
 
-**\*\*** Some of this could be replaced with the shutil module? **\*\***
+.. XXX Some of this could be replaced with the shutil module?
 
 
 :mod:`distutils.file_util` --- Single file operations
@@ -1311,8 +1311,7 @@
   the "negative alias" of :option:`--verbose`, then :option:`--quiet` on the
   command line sets *verbose* to false.
 
-**\*\*** Should be replaced with :mod:`optik` (which is also now known as
-:mod:`optparse` in Python 2.3 and later). **\*\***
+.. XXX Should be replaced with :mod:`optparse`.
 
 
 .. function:: fancy_getopt(options, negative_opt, object, args)

Modified: python/branches/py3k/Doc/distutils/builtdist.rst
==============================================================================
--- python/branches/py3k/Doc/distutils/builtdist.rst	(original)
+++ python/branches/py3k/Doc/distutils/builtdist.rst	Wed Jul  7 21:04:36 2010
@@ -146,8 +146,8 @@
 Creating dumb built distributions
 =================================
 
-**\*\*** Need to document absolute vs. prefix-relative packages here, but first
-I have to implement it! **\*\***
+.. XXX Need to document absolute vs. prefix-relative packages here, but first
+       I have to implement it!
 
 
 .. _creating-rpms:

Modified: python/branches/py3k/Doc/distutils/setupscript.rst
==============================================================================
--- python/branches/py3k/Doc/distutils/setupscript.rst	(original)
+++ python/branches/py3k/Doc/distutils/setupscript.rst	Wed Jul  7 21:04:36 2010
@@ -207,7 +207,7 @@
 SWIG on the interface file and compile the resulting C/C++ file into your
 extension.
 
-**\*\*** SWIG support is rough around the edges and largely untested! **\*\***
+.. XXX SWIG support is rough around the edges and largely untested!
 
 This warning notwithstanding, options to SWIG can be currently passed like
 this::
@@ -326,7 +326,7 @@
 (Again, this sort of non-portable construct should be avoided if you intend to
 distribute your code.)
 
-**\*\*** Should mention clib libraries here or somewhere else! **\*\***
+.. XXX Should mention clib libraries here or somewhere else!
 
 
 Other options

From python-checkins at python.org  Wed Jul  7 21:04:49 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Wed,  7 Jul 2010 21:04:49 +0200 (CEST)
Subject: [Python-checkins] r82633 - python/branches/py3k/Misc/NEWS
Message-ID: <20100707190449.126D3EEB0F@mail.python.org>

Author: benjamin.peterson
Date: Wed Jul  7 21:04:48 2010
New Revision: 82633

Log:
add NEWS

Modified:
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Wed Jul  7 21:04:48 2010
@@ -1396,6 +1396,9 @@
 Extension Modules
 -----------------
 
+- In struct.pack, correctly propogate exceptions from computing the truth of an
+  object in the '?' format.
+
 - Issue #9000: datetime.timezone objects now have eval-friendly repr.
 
 - In the math module, correctly lookup __trunc__, __ceil__, and __floor__ as

From python-checkins at python.org  Wed Jul  7 21:05:35 2010
From: python-checkins at python.org (georg.brandl)
Date: Wed,  7 Jul 2010 21:05:35 +0200 (CEST)
Subject: [Python-checkins] r82634 - in python/branches/release27-maint:
	Doc/distutils/apiref.rst Doc/distutils/builtdist.rst
	Doc/distutils/setupscript.rst Doc/distutils/sourcedist.rst
Message-ID: <20100707190535.D6722EEB3D@mail.python.org>

Author: georg.brandl
Date: Wed Jul  7 21:05:35 2010
New Revision: 82634

Log:
Merged revisions 82629,82632 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82629 | georg.brandl | 2010-07-07 20:51:43 +0200 (Mi, 07 Jul 2010) | 1 line
  
  Make comment out of an awkward note.
........
  r82632 | georg.brandl | 2010-07-07 21:04:36 +0200 (Mi, 07 Jul 2010) | 1 line
  
  Turn more notes into comments.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Doc/distutils/apiref.rst
   python/branches/release27-maint/Doc/distutils/builtdist.rst
   python/branches/release27-maint/Doc/distutils/setupscript.rst
   python/branches/release27-maint/Doc/distutils/sourcedist.rst

Modified: python/branches/release27-maint/Doc/distutils/apiref.rst
==============================================================================
--- python/branches/release27-maint/Doc/distutils/apiref.rst	(original)
+++ python/branches/release27-maint/Doc/distutils/apiref.rst	Wed Jul  7 21:05:35 2010
@@ -995,7 +995,7 @@
    errors are ignored (apart from being reported to ``sys.stdout`` if *verbose* is
    true).
 
-**\*\*** Some of this could be replaced with the shutil module? **\*\***
+.. XXX Some of this could be replaced with the shutil module?
 
 
 :mod:`distutils.file_util` --- Single file operations
@@ -1311,8 +1311,7 @@
   the "negative alias" of :option:`--verbose`, then :option:`--quiet` on the
   command line sets *verbose* to false.
 
-**\*\*** Should be replaced with :mod:`optik` (which is also now known as
-:mod:`optparse` in Python 2.3 and later). **\*\***
+.. XXX Should be replaced with :mod:`optparse`.
 
 
 .. function:: fancy_getopt(options, negative_opt, object, args)

Modified: python/branches/release27-maint/Doc/distutils/builtdist.rst
==============================================================================
--- python/branches/release27-maint/Doc/distutils/builtdist.rst	(original)
+++ python/branches/release27-maint/Doc/distutils/builtdist.rst	Wed Jul  7 21:05:35 2010
@@ -146,8 +146,8 @@
 Creating dumb built distributions
 =================================
 
-**\*\*** Need to document absolute vs. prefix-relative packages here, but first
-I have to implement it! **\*\***
+.. XXX Need to document absolute vs. prefix-relative packages here, but first
+       I have to implement it!
 
 
 .. _creating-rpms:

Modified: python/branches/release27-maint/Doc/distutils/setupscript.rst
==============================================================================
--- python/branches/release27-maint/Doc/distutils/setupscript.rst	(original)
+++ python/branches/release27-maint/Doc/distutils/setupscript.rst	Wed Jul  7 21:05:35 2010
@@ -207,7 +207,7 @@
 SWIG on the interface file and compile the resulting C/C++ file into your
 extension.
 
-**\*\*** SWIG support is rough around the edges and largely untested! **\*\***
+.. XXX SWIG support is rough around the edges and largely untested!
 
 This warning notwithstanding, options to SWIG can be currently passed like
 this::
@@ -326,7 +326,7 @@
 (Again, this sort of non-portable construct should be avoided if you intend to
 distribute your code.)
 
-**\*\*** Should mention clib libraries here or somewhere else! **\*\***
+.. XXX Should mention clib libraries here or somewhere else!
 
 
 Other options

Modified: python/branches/release27-maint/Doc/distutils/sourcedist.rst
==============================================================================
--- python/branches/release27-maint/Doc/distutils/sourcedist.rst	(original)
+++ python/branches/release27-maint/Doc/distutils/sourcedist.rst	Wed Jul  7 21:05:35 2010
@@ -76,10 +76,10 @@
   :option:`packages` options
 
 * all C source files mentioned in the :option:`ext_modules` or
-  :option:`libraries` options (
+  :option:`libraries` options
 
-  **\*\*** getting C library sources currently broken---no
-  :meth:`get_source_files` method in :file:`build_clib.py`! **\*\***)
+  .. XXX Getting C library sources is currently broken -- no
+     :meth:`get_source_files` method in :file:`build_clib.py`!
 
 * scripts identified by the :option:`scripts` option
   See :ref:`distutils-installing-scripts`.

From python-checkins at python.org  Wed Jul  7 21:09:12 2010
From: python-checkins at python.org (georg.brandl)
Date: Wed,  7 Jul 2010 21:09:12 +0200 (CEST)
Subject: [Python-checkins] r82635 - in python/branches/release27-maint:
	Doc/using/cmdline.rst
Message-ID: <20100707190912.63C8AEE9C4@mail.python.org>

Author: georg.brandl
Date: Wed Jul  7 21:09:12 2010
New Revision: 82635

Log:
Merged revisions 82615 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82615 | georg.brandl | 2010-07-07 00:58:50 +0200 (Mi, 07 Jul 2010) | 1 line
  
  Fix typo.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Doc/using/cmdline.rst

Modified: python/branches/release27-maint/Doc/using/cmdline.rst
==============================================================================
--- python/branches/release27-maint/Doc/using/cmdline.rst	(original)
+++ python/branches/release27-maint/Doc/using/cmdline.rst	Wed Jul  7 21:09:12 2010
@@ -61,7 +61,7 @@
 
 .. cmdoption:: -c <command>
 
-   Execute the Python code in *command*.  *command* can be one ore more
+   Execute the Python code in *command*.  *command* can be one or more
    statements separated by newlines, with significant leading whitespace as in
    normal module code.
 

From python-checkins at python.org  Wed Jul  7 22:54:01 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Wed,  7 Jul 2010 22:54:01 +0200 (CEST)
Subject: [Python-checkins] r82636 - in python/branches/py3k:
	Include/structseq.h Misc/NEWS Objects/structseq.c
Message-ID: <20100707205401.A8198F68FA@mail.python.org>

Author: benjamin.peterson
Date: Wed Jul  7 22:54:01 2010
New Revision: 82636

Log:
make struct sequences subclass tuple; kill lots of code

This fixes #8413.


Modified:
   python/branches/py3k/Include/structseq.h
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/Objects/structseq.c

Modified: python/branches/py3k/Include/structseq.h
==============================================================================
--- python/branches/py3k/Include/structseq.h	(original)
+++ python/branches/py3k/Include/structseq.h	Wed Jul  7 22:54:01 2010
@@ -26,17 +26,12 @@
 
 PyAPI_FUNC(PyObject *) PyStructSequence_New(PyTypeObject* type);
 
-typedef struct {
-    PyObject_VAR_HEAD
-    PyObject *ob_item[1];
-} PyStructSequence;
+typedef PyTupleObject PyStructSequence;
 
 /* Macro, *only* to be used to fill in brand new objects */
-#define PyStructSequence_SET_ITEM(op, i, v) \
-    (((PyStructSequence *)(op))->ob_item[i] = v)
+#define PyStructSequence_SET_ITEM(op, i, v) PyTuple_SET_ITEM(op, i, v)
 
-#define PyStructSequence_GET_ITEM(op, i) \
-    (((PyStructSequence *)(op))->ob_item[i])
+#define PyStructSequence_GET_ITEM(op, i) PyTuple_GET_ITEM(op, i)
 
 
 #ifdef __cplusplus

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Wed Jul  7 22:54:01 2010
@@ -12,6 +12,8 @@
 Core and Builtins
 -----------------
 
+- Issue #8413: structsequence now subclasses tuple.
+
 - Issue #8271: during the decoding of an invalid UTF-8 byte sequence, only the
   start byte and the continuation byte(s) are now considered invalid, instead
   of the number of bytes specified by the start byte.

Modified: python/branches/py3k/Objects/structseq.c
==============================================================================
--- python/branches/py3k/Objects/structseq.c	(original)
+++ python/branches/py3k/Objects/structseq.c	Wed Jul  7 22:54:01 2010
@@ -30,113 +30,15 @@
 PyStructSequence_New(PyTypeObject *type)
 {
     PyStructSequence *obj;
+    Py_ssize_t size = REAL_SIZE_TP(type), i;
 
-    obj = PyObject_New(PyStructSequence, type);
+    obj = PyObject_GC_NewVar(PyStructSequence, type, size);
     if (obj == NULL)
         return NULL;
-    Py_SIZE(obj) = VISIBLE_SIZE_TP(type);
+    for (i = 0; i < size; i++)
+        obj->ob_item[i] = NULL;
 
-    return (PyObject*) obj;
-}
-
-static void
-structseq_dealloc(PyStructSequence *obj)
-{
-    Py_ssize_t i, size;
-
-    size = REAL_SIZE(obj);
-    for (i = 0; i < size; ++i) {
-        Py_XDECREF(obj->ob_item[i]);
-    }
-    PyObject_Del(obj);
-}
-
-static Py_ssize_t
-structseq_length(PyStructSequence *obj)
-{
-    return VISIBLE_SIZE(obj);
-}
-
-static PyObject*
-structseq_item(PyStructSequence *obj, Py_ssize_t i)
-{
-    if (i < 0 || i >= VISIBLE_SIZE(obj)) {
-        PyErr_SetString(PyExc_IndexError, "tuple index out of range");
-        return NULL;
-    }
-    Py_INCREF(obj->ob_item[i]);
-    return obj->ob_item[i];
-}
-
-static PyObject*
-structseq_slice(PyStructSequence *obj, Py_ssize_t low, Py_ssize_t high)
-{
-    PyTupleObject *np;
-    Py_ssize_t i;
-
-    if (low < 0)
-        low = 0;
-    if (high > VISIBLE_SIZE(obj))
-        high = VISIBLE_SIZE(obj);
-    if (high < low)
-        high = low;
-    np = (PyTupleObject *)PyTuple_New(high-low);
-    if (np == NULL)
-        return NULL;
-    for(i = low; i < high; ++i) {
-        PyObject *v = obj->ob_item[i];
-        Py_INCREF(v);
-        PyTuple_SET_ITEM(np, i-low, v);
-    }
-    return (PyObject *) np;
-}
-
-static PyObject *
-structseq_subscript(PyStructSequence *self, PyObject *item)
-{
-    if (PyIndex_Check(item)) {
-        Py_ssize_t i = PyNumber_AsSsize_t(item, PyExc_IndexError);
-        if (i == -1 && PyErr_Occurred())
-            return NULL;
-
-        if (i < 0)
-            i += VISIBLE_SIZE(self);
-
-        if (i < 0 || i >= VISIBLE_SIZE(self)) {
-            PyErr_SetString(PyExc_IndexError,
-                "tuple index out of range");
-            return NULL;
-        }
-        Py_INCREF(self->ob_item[i]);
-        return self->ob_item[i];
-    }
-    else if (PySlice_Check(item)) {
-        Py_ssize_t start, stop, step, slicelen, cur, i;
-        PyObject *result;
-
-        if (PySlice_GetIndicesEx((PySliceObject *)item,
-                                 VISIBLE_SIZE(self), &start, &stop,
-                                 &step, &slicelen) < 0) {
-            return NULL;
-        }
-        if (slicelen <= 0)
-            return PyTuple_New(0);
-        result = PyTuple_New(slicelen);
-        if (result == NULL)
-            return NULL;
-        for (cur = start, i = 0; i < slicelen;
-             cur += step, i++) {
-            PyObject *v = self->ob_item[cur];
-            Py_INCREF(v);
-            PyTuple_SET_ITEM(result, i, v);
-        }
-        return result;
-    }
-    else {
-        PyErr_SetString(PyExc_TypeError,
-                        "structseq index must be integer");
-        return NULL;
-    }
+    return (PyObject*)obj;
 }
 
 static PyObject *
@@ -223,11 +125,6 @@
     return (PyObject*) res;
 }
 
-static PyObject *
-make_tuple(PyStructSequence *obj)
-{
-    return structseq_slice(obj, 0, VISIBLE_SIZE(obj));
-}
 
 static PyObject *
 structseq_repr(PyStructSequence *obj)
@@ -236,7 +133,6 @@
 #define REPR_BUFFER_SIZE 512
 #define TYPE_MAXSIZE 100
 
-    PyObject *tup;
     PyTypeObject *typ = Py_TYPE(obj);
     int i, removelast = 0;
     Py_ssize_t len;
@@ -246,10 +142,6 @@
     /* pointer to end of writeable buffer; safes space for "...)\0" */
     endofbuf= &buf[REPR_BUFFER_SIZE-5];
 
-    if ((tup = make_tuple(obj)) == NULL) {
-        return NULL;
-    }
-
     /* "typename(", limited to  TYPE_MAXSIZE */
     len = strlen(typ->tp_name) > TYPE_MAXSIZE ? TYPE_MAXSIZE :
                             strlen(typ->tp_name);
@@ -262,19 +154,14 @@
         char *cname, *crepr;
 
         cname = typ->tp_members[i].name;
-
-        val = PyTuple_GetItem(tup, i);
-        if (cname == NULL || val == NULL) {
+        if (cname == NULL)
             return NULL;
-        }
+        val = PyStructSequence_GET_ITEM(obj, i);
         repr = PyObject_Repr(val);
-        if (repr == NULL) {
-            Py_DECREF(tup);
+        if (repr == NULL)
             return NULL;
-        }
         crepr = _PyUnicode_AsString(repr);
         if (crepr == NULL) {
-            Py_DECREF(tup);
             Py_DECREF(repr);
             return NULL;
         }
@@ -300,7 +187,6 @@
             break;
         }
     }
-    Py_DECREF(tup);
     if (removelast) {
         /* overwrite last ", " */
         pbuf-=2;
@@ -312,62 +198,6 @@
 }
 
 static PyObject *
-structseq_concat(PyStructSequence *obj, PyObject *b)
-{
-    PyObject *tup, *result;
-    tup = make_tuple(obj);
-    result = PySequence_Concat(tup, b);
-    Py_DECREF(tup);
-    return result;
-}
-
-static PyObject *
-structseq_repeat(PyStructSequence *obj, Py_ssize_t n)
-{
-    PyObject *tup, *result;
-    tup = make_tuple(obj);
-    result = PySequence_Repeat(tup, n);
-    Py_DECREF(tup);
-    return result;
-}
-
-static int
-structseq_contains(PyStructSequence *obj, PyObject *o)
-{
-    PyObject *tup;
-    int result;
-    tup = make_tuple(obj);
-    if (!tup)
-        return -1;
-    result = PySequence_Contains(tup, o);
-    Py_DECREF(tup);
-    return result;
-}
-
-static long
-structseq_hash(PyObject *obj)
-{
-    PyObject *tup;
-    long result;
-    tup = make_tuple((PyStructSequence*) obj);
-    if (!tup)
-        return -1;
-    result = PyObject_Hash(tup);
-    Py_DECREF(tup);
-    return result;
-}
-
-static PyObject *
-structseq_richcompare(PyObject *obj, PyObject *o2, int op)
-{
-    PyObject *tup, *result;
-    tup = make_tuple((PyStructSequence*) obj);
-    result = PyObject_RichCompare(tup, o2, op);
-    Py_DECREF(tup);
-    return result;
-}
-
-static PyObject *
 structseq_reduce(PyStructSequence* self)
 {
     PyObject* tup;
@@ -409,53 +239,36 @@
     return result;
 }
 
-static PySequenceMethods structseq_as_sequence = {
-    (lenfunc)structseq_length,
-    (binaryfunc)structseq_concat,           /* sq_concat */
-    (ssizeargfunc)structseq_repeat,         /* sq_repeat */
-    (ssizeargfunc)structseq_item,               /* sq_item */
-    0,                                          /* sq_slice */
-    0,                                          /* sq_ass_item */
-    0,                                          /* sq_ass_slice */
-    (objobjproc)structseq_contains,             /* sq_contains */
-};
-
-static PyMappingMethods structseq_as_mapping = {
-    (lenfunc)structseq_length,
-    (binaryfunc)structseq_subscript,
-};
-
 static PyMethodDef structseq_methods[] = {
-    {"__reduce__", (PyCFunction)structseq_reduce,
-     METH_NOARGS, NULL},
+    {"__reduce__", (PyCFunction)structseq_reduce, METH_NOARGS, NULL},
     {NULL, NULL}
 };
 
 static PyTypeObject _struct_sequence_template = {
     PyVarObject_HEAD_INIT(&PyType_Type, 0)
     NULL,                                       /* tp_name */
-    0,                                          /* tp_basicsize */
-    0,                                          /* tp_itemsize */
-    (destructor)structseq_dealloc,              /* tp_dealloc */
+    sizeof(PyStructSequence) - sizeof(PyObject *), /* tp_basicsize */
+    sizeof(PyObject *),                         /* tp_itemsize */
+    0,                                          /* tp_dealloc */
     0,                                          /* tp_print */
     0,                                          /* tp_getattr */
     0,                                          /* tp_setattr */
     0,                                          /* tp_reserved */
     (reprfunc)structseq_repr,                   /* tp_repr */
     0,                                          /* tp_as_number */
-    &structseq_as_sequence,                     /* tp_as_sequence */
-    &structseq_as_mapping,                      /* tp_as_mapping */
-    structseq_hash,                             /* tp_hash */
+    0,                                          /* tp_as_sequence */
+    0,                                          /* tp_as_mapping */
+    0,                                          /* tp_hash */
     0,                                          /* tp_call */
     0,                                          /* tp_str */
     0,                                          /* tp_getattro */
     0,                                          /* tp_setattro */
     0,                                          /* tp_as_buffer */
-    Py_TPFLAGS_DEFAULT,                     /* tp_flags */
+    Py_TPFLAGS_DEFAULT,                         /* tp_flags */
     NULL,                                       /* tp_doc */
     0,                                          /* tp_traverse */
     0,                                          /* tp_clear */
-    structseq_richcompare,                      /* tp_richcompare */
+    0,                                          /* tp_richcompare */
     0,                                          /* tp_weaklistoffset */
     0,                                          /* tp_iter */
     0,                                          /* tp_iternext */
@@ -494,11 +307,9 @@
     n_members = i;
 
     memcpy(type, &_struct_sequence_template, sizeof(PyTypeObject));
+    type->tp_base = &PyTuple_Type;
     type->tp_name = desc->name;
     type->tp_doc = desc->doc;
-    type->tp_basicsize = sizeof(PyStructSequence)+
-        sizeof(PyObject*)*(n_members-1);
-    type->tp_itemsize = 0;
 
     members = PyMem_NEW(PyMemberDef, n_members-n_unnamed_members+1);
     if (members == NULL)

From python-checkins at python.org  Thu Jul  8 00:45:06 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Thu,  8 Jul 2010 00:45:06 +0200 (CEST)
Subject: [Python-checkins] r82637 -
	python/branches/py3k/Lib/test/test_struct.py
Message-ID: <20100707224506.A24B4EEB68@mail.python.org>

Author: benjamin.peterson
Date: Thu Jul  8 00:45:06 2010
New Revision: 82637

Log:
ValueError in this case is also acceptable

Modified:
   python/branches/py3k/Lib/test/test_struct.py

Modified: python/branches/py3k/Lib/test/test_struct.py
==============================================================================
--- python/branches/py3k/Lib/test/test_struct.py	(original)
+++ python/branches/py3k/Lib/test/test_struct.py	Thu Jul  8 00:45:06 2010
@@ -438,8 +438,10 @@
 
         # Go beyond boundaries.
         small_buf = array.array('b', b' '*10)
-        self.assertRaises(struct.error, s.pack_into, small_buf, 0, test_string)
-        self.assertRaises(struct.error, s.pack_into, small_buf, 2, test_string)
+        self.assertRaises((ValueError, struct.error), s.pack_into, small_buf, 0,
+                          test_string)
+        self.assertRaises((ValueError, struct.error), s.pack_into, small_buf, 2,
+                          test_string)
 
         # Test bogus offset (issue 3694)
         sb = small_buf
@@ -463,8 +465,10 @@
 
         # Go beyond boundaries.
         small_buf = array.array('b', b' '*10)
-        self.assertRaises(struct.error, pack_into, small_buf, 0, test_string)
-        self.assertRaises(struct.error, pack_into, small_buf, 2, test_string)
+        self.assertRaises((ValueError, struct.error), pack_into, small_buf, 0,
+                          test_string)
+        self.assertRaises((ValueError, struct.error), pack_into, small_buf, 2,
+                          test_string)
 
     def test_unpack_with_buffer(self):
         # SF bug 1563759: struct.unpack doens't support buffer protocol objects

From python-checkins at python.org  Thu Jul  8 00:46:01 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Thu,  8 Jul 2010 00:46:01 +0200 (CEST)
Subject: [Python-checkins] r82638 - in python/branches/release31-maint:
	Lib/test/test_struct.py Modules/_struct.c
Message-ID: <20100707224601.13701EEBDD@mail.python.org>

Author: benjamin.peterson
Date: Thu Jul  8 00:46:00 2010
New Revision: 82638

Log:
Merged revisions 82628,82630 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82628 | benjamin.peterson | 2010-07-07 13:44:05 -0500 (Wed, 07 Jul 2010) | 1 line
  
  this needn't be in the loop
........
  r82630 | benjamin.peterson | 2010-07-07 13:54:59 -0500 (Wed, 07 Jul 2010) | 1 line
  
  don't ignore exceptions from PyObject_IsTrue
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/test/test_struct.py
   python/branches/release31-maint/Modules/_struct.c

Modified: python/branches/release31-maint/Lib/test/test_struct.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_struct.py	(original)
+++ python/branches/release31-maint/Lib/test/test_struct.py	Thu Jul  8 00:46:00 2010
@@ -480,6 +480,9 @@
             self.assertEqual(value, 0x12345678)
 
     def test_bool(self):
+        class ExplodingBool(object):
+            def __bool__(self):
+                raise IOError
         for prefix in tuple("<>!=")+('',):
             false = (), [], [], '', 0
             true = [1], 'test', 5, -1, 0xffffffff+1, 0xffffffff/2
@@ -508,8 +511,11 @@
                 self.assertFalse(prefix, msg='encoded bool is not one byte: %r'
                                              %packed)
 
-            for c in [b'\x01', b'\x7f', b'\xff', b'\x0f', b'\xf0']:
-                self.assertTrue(struct.unpack('>?', c)[0])
+            self.assertRaises(IOError, struct.pack, prefix + '?',
+                              ExplodingBool())
+
+        for c in [b'\x01', b'\x7f', b'\xff', b'\x0f', b'\xf0']:
+            self.assertTrue(struct.unpack('>?', c)[0])
 
     def test_count_overflow(self):
         hugecount = '{}b'.format(sys.maxsize+1)

Modified: python/branches/release31-maint/Modules/_struct.c
==============================================================================
--- python/branches/release31-maint/Modules/_struct.c	(original)
+++ python/branches/release31-maint/Modules/_struct.c	Thu Jul  8 00:46:00 2010
@@ -580,9 +580,13 @@
 static int
 np_bool(char *p, PyObject *v, const formatdef *f)
 {
-    BOOL_TYPE y;
+    int y;
+    BOOL_TYPE x;
     y = PyObject_IsTrue(v);
-    memcpy(p, (char *)&y, sizeof y);
+    if (y < 0)
+        return -1;
+    x = y;
+    memcpy(p, (char *)&x, sizeof x);
     return 0;
 }
 
@@ -854,6 +858,8 @@
 {
     char y;
     y = PyObject_IsTrue(v);
+    if (y < 0)
+        return -1;
     memcpy(p, (char *)&y, sizeof y);
     return 0;
 }

From python-checkins at python.org  Thu Jul  8 00:50:58 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Thu,  8 Jul 2010 00:50:58 +0200 (CEST)
Subject: [Python-checkins] r82639 - in python/branches/release31-maint:
	Lib/test/test_struct.py
Message-ID: <20100707225058.9C669EEBEE@mail.python.org>

Author: benjamin.peterson
Date: Thu Jul  8 00:50:58 2010
New Revision: 82639

Log:
Merged revisions 82637 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82637 | benjamin.peterson | 2010-07-07 17:45:06 -0500 (Wed, 07 Jul 2010) | 1 line
  
  ValueError in this case is also acceptable
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/test/test_struct.py

Modified: python/branches/release31-maint/Lib/test/test_struct.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_struct.py	(original)
+++ python/branches/release31-maint/Lib/test/test_struct.py	Thu Jul  8 00:50:58 2010
@@ -443,8 +443,10 @@
 
         # Go beyond boundaries.
         small_buf = array.array('b', b' '*10)
-        self.assertRaises(struct.error, s.pack_into, small_buf, 0, test_string)
-        self.assertRaises(struct.error, s.pack_into, small_buf, 2, test_string)
+        self.assertRaises((ValueError, struct.error), s.pack_into, small_buf, 0,
+                          test_string)
+        self.assertRaises((ValueError, struct.error), s.pack_into, small_buf, 2,
+                          test_string)
 
         # Test bogus offset (issue 3694)
         sb = small_buf
@@ -468,8 +470,10 @@
 
         # Go beyond boundaries.
         small_buf = array.array('b', b' '*10)
-        self.assertRaises(struct.error, pack_into, small_buf, 0, test_string)
-        self.assertRaises(struct.error, pack_into, small_buf, 2, test_string)
+        self.assertRaises((ValueError, struct.error), pack_into, small_buf, 0,
+                          test_string)
+        self.assertRaises((ValueError, struct.error), pack_into, small_buf, 2,
+                          test_string)
 
     def test_unpack_with_buffer(self):
         # SF bug 1563759: struct.unpack doens't support buffer protocol objects

From solipsis at pitrou.net  Thu Jul  8 01:24:45 2010
From: solipsis at pitrou.net (solipsis at pitrou.net)
Date: Thu,  8 Jul 2010 01:24:45 +0200 (CEST)
Subject: [Python-checkins] Daily py3k reference leaks (r82636): sum=0
Message-ID: <20100707232445.726061770A@ns6635.ovh.net>

py3k results for svn r82636 (hg cset b673e8062e16)
--------------------------------------------------



Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/antoine/py3k/refleaks/reflog9-BDj5', '-x']

From python-checkins at python.org  Thu Jul  8 01:26:57 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Thu,  8 Jul 2010 01:26:57 +0200 (CEST)
Subject: [Python-checkins] r82640 - in python/branches/release27-maint:
	Lib/test/test_struct.py
Message-ID: <20100707232657.D3A22EEC07@mail.python.org>

Author: benjamin.peterson
Date: Thu Jul  8 01:26:57 2010
New Revision: 82640

Log:
Merged revisions 82637 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82637 | benjamin.peterson | 2010-07-07 17:45:06 -0500 (Wed, 07 Jul 2010) | 1 line
  
  ValueError in this case is also acceptable
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/test/test_struct.py

Modified: python/branches/release27-maint/Lib/test/test_struct.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_struct.py	(original)
+++ python/branches/release27-maint/Lib/test/test_struct.py	Thu Jul  8 01:26:57 2010
@@ -450,8 +450,10 @@
 
         # Go beyond boundaries.
         small_buf = array.array('c', ' '*10)
-        self.assertRaises(struct.error, s.pack_into, small_buf, 0, test_string)
-        self.assertRaises(struct.error, s.pack_into, small_buf, 2, test_string)
+        self.assertRaises((ValueError, struct.error), s.pack_into, small_buf, 0,
+                          test_string)
+        self.assertRaises((ValueError, struct.error), s.pack_into, small_buf, 2,
+                          test_string)
 
         # Test bogus offset (issue 3694)
         sb = small_buf
@@ -475,8 +477,10 @@
 
         # Go beyond boundaries.
         small_buf = array.array('c', ' '*10)
-        self.assertRaises(struct.error, pack_into, small_buf, 0, test_string)
-        self.assertRaises(struct.error, pack_into, small_buf, 2, test_string)
+        self.assertRaises((ValueError, struct.error), pack_into, small_buf, 0,
+                          test_string)
+        self.assertRaises((ValueError, struct.error), pack_into, small_buf, 2,
+                          test_string)
 
     def test_unpack_with_buffer(self):
         with check_py3k_warnings(("buffer.. not supported in 3.x",

From python-checkins at python.org  Thu Jul  8 01:46:38 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Thu,  8 Jul 2010 01:46:38 +0200 (CEST)
Subject: [Python-checkins] r82641 -
	sandbox/branches/py3k-datetime/datetime.py
Message-ID: <20100707234638.98E58EEC09@mail.python.org>

Author: alexander.belopolsky
Date: Thu Jul  8 01:46:38 2010
New Revision: 82641

Log:
Issue #5288: Eliminated round-trips between timdelta and int offsets

Modified:
   sandbox/branches/py3k-datetime/datetime.py

Modified: sandbox/branches/py3k-datetime/datetime.py
==============================================================================
--- sandbox/branches/py3k-datetime/datetime.py	(original)
+++ sandbox/branches/py3k-datetime/datetime.py	Thu Jul  8 01:46:38 2010
@@ -176,7 +176,7 @@
     if year < 1900:
         raise ValueError("year=%d is before 1900; the datetime strftime() "
                          "methods require year >= 1900" % year)
-    # Don't call _utcoffset() or tzname() unless actually needed.
+    # Don't call utcoffset() or tzname() unless actually needed.
     freplace = None # the string to use for %f
     zreplace = None # the string to use for %z
     Zreplace = None # the string to use for %Z
@@ -200,14 +200,16 @@
                 elif ch == 'z':
                     if zreplace is None:
                         zreplace = ""
-                        if hasattr(object, "_utcoffset"):
-                            offset = object._utcoffset()
+                        if hasattr(object, "utcoffset"):
+                            offset = object.utcoffset()
                             if offset is not None:
                                 sign = '+'
-                                if offset < 0:
+                                if offset.days < 0:
                                     offset = -offset
                                     sign = '-'
-                                h, m = divmod(offset, 60)
+                                h, m = divmod(offset, timedelta(hours=1))
+                                assert not m % timedelta(0, 60), "whole minute"
+                                m //= timedelta(0, 60)
                                 zreplace = '%c%02d%02d' % (sign, h, m)
                     assert '%' not in zreplace
                     newformat.append(zreplace)
@@ -250,23 +252,17 @@
 def _check_utc_offset(name, offset):
     assert name in ("utcoffset", "dst")
     if offset is None:
-        return None
+        return
     if not isinstance(offset, timedelta):
         raise TypeError("tzinfo.%s() must return None "
                         "or timedelta, not '%s'" % (name, type(offset)))
-    days = offset.days
-    if days < -1 or days > 0:
-        offset = 1440  # trigger out-of-range
-    else:
-        seconds = days * 86400 + offset.seconds
-        minutes, seconds = divmod(seconds, 60)
-        if seconds or offset.microseconds:
-            raise ValueError("tzinfo.%s() must return a whole number "
-                             "of minutes" % name)
-        offset = minutes
-    if -1440 < offset < 1440:
-        return offset
-    raise ValueError("%s()=%d, must be in -1439..1439" % (name, offset))
+    if offset % timedelta(0, 60) or offset.microseconds:
+        raise ValueError("tzinfo.%s() must return a whole number "
+                         "of minutes, got %s" % (name, offset))
+    if not -timedelta(1) < offset < timedelta(1):
+        raise ValueError("%s()=%s, must be must be strictly between"
+                         " -timedelta(hours=24) and timedelta(hours=24)"
+                         % (name, offset))
 
 def _check_date_fields(year, month, day):
     if not isinstance(year, int):
@@ -1112,8 +1108,8 @@
         if mytz is ottz:
             base_compare = True
         else:
-            myoff = self._utcoffset()
-            otoff = other._utcoffset()
+            myoff = self.utcoffset()
+            otoff = other.utcoffset()
             base_compare = myoff == otoff
 
         if base_compare:
@@ -1123,17 +1119,20 @@
                         other._microsecond))
         if myoff is None or otoff is None:
             raise TypeError("cannot compare naive and aware times")
-        myhhmm = self._hour * 60 + self._minute - myoff
-        othhmm = other._hour * 60 + other._minute - otoff
+        myhhmm = self._hour * 60 + self._minute - myoff//timedelta(0, 60)
+        othhmm = other._hour * 60 + other._minute - otoff//timedelta(0, 60)
         return _cmp((myhhmm, self._second, self._microsecond),
                     (othhmm, other._second, other._microsecond))
 
     def __hash__(self):
         """Hash."""
-        tzoff = self._utcoffset()
+        tzoff = self.utcoffset()
         if not tzoff: # zero or None
             return hash(self._getstate()[0])
-        h, m = divmod(self.hour * 60 + self.minute - tzoff, 60)
+        h, m = divmod(timedelta(hours=self.hour, minutes=self.minute) - tzoff,
+                      timedelta(hours=1))
+        assert not m % timedelta(0, 60), "whole minute"
+        m //= timedelta(0, 60)
         if 0 <= h < 24:
             return hash(time(h, m, self.second, self.microsecond))
         return hash((h, m, self.second, self.microsecond))
@@ -1142,14 +1141,16 @@
 
     def _tzstr(self, sep=":"):
         """Return formatted timezone offset (+xx:xx) or None."""
-        off = self._utcoffset()
+        off = self.utcoffset()
         if off is not None:
-            if off < 0:
+            if off.days < 0:
                 sign = "-"
                 off = -off
             else:
                 sign = "+"
-            hh, mm = divmod(off, 60)
+            hh, mm = divmod(off, timedelta(hours=1))
+            assert not mm % timedelta(0, 60), "whole minute"
+            mm //= timedelta(0, 60)
             assert 0 <= hh < 24
             off = "%s%02d%s%02d" % (sign, hh, sep, mm)
         return off
@@ -1205,16 +1206,10 @@
     def utcoffset(self):
         """Return the timezone offset in minutes east of UTC (negative west of
         UTC)."""
-        offset = _call_tzinfo_method(self._tzinfo, "utcoffset", None)
-        offset = _check_utc_offset("utcoffset", offset)
-        if offset is not None:
-            offset = timedelta(minutes=offset)
-        return offset
-
-    # Return an integer (or None) instead of a timedelta (or None).
-    def _utcoffset(self):
-        offset = _call_tzinfo_method(self._tzinfo, "utcoffset", None)
-        offset = _check_utc_offset("utcoffset", offset)
+        if self._tzinfo is None:
+            return None
+        offset = self._tzinfo.utcoffset(None)
+        _check_utc_offset("utcoffset", offset)
         return offset
 
     def tzname(self):
@@ -1224,7 +1219,9 @@
         it mean anything in particular. For example, "GMT", "UTC", "-500",
         "-5:00", "EDT", "US/Eastern", "America/New York" are all valid replies.
         """
-        name = _call_tzinfo_method(self._tzinfo, "tzname", None)
+        if self._tzinfo is None:
+            return None
+        name = self._tzinfo.tzname(None)
         _check_tzname(name)
         return name
 
@@ -1237,10 +1234,10 @@
         need to consult dst() unless you're interested in displaying the DST
         info.
         """
-        offset = _call_tzinfo_method(self._tzinfo, "dst", None)
-        offset = _check_utc_offset("dst", offset)
-        if offset is not None:
-            offset = timedelta(minutes=offset)
+        if self._tzinfo is None:
+            return None
+        offset = self._tzinfo.dst(None)
+        _check_utc_offset("dst", offset)
         return offset
 
     def replace(self, hour=None, minute=None, second=None, microsecond=None,
@@ -1263,8 +1260,8 @@
     def __bool__(self):
         if self.second or self.microsecond:
             return True
-        offset = self._utcoffset() or 0
-        return self.hour * 60 + self.minute - offset != 0
+        offset = self.utcoffset() or timedelta(0)
+        return timedelta(hours=self.hour, minutes=self.minute) != offset
 
     # Pickle support.
 
@@ -1417,11 +1414,13 @@
 
     def timetuple(self):
         "Return local time tuple compatible with time.localtime()."
-        dst = self._dst()
+        dst = self.dst()
         if dst is None:
             dst = -1
         elif dst:
             dst = 1
+        else:
+            dst = 0
         return _build_struct_time(self.year, self.month, self.day,
                                   self.hour, self.minute, self.second,
                                   dst)
@@ -1521,14 +1520,16 @@
                                   sep) +
                 _format_time(self._hour, self._minute, self._second,
                              self._microsecond))
-        off = self._utcoffset()
+        off = self.utcoffset()
         if off is not None:
-            if off < 0:
+            if off.days < 0:
                 sign = "-"
                 off = -off
             else:
                 sign = "+"
-            hh, mm = divmod(off, 60)
+            hh, mm = divmod(off, timedelta(hours=1))
+            assert not mm % timedelta(0, 60), "whole minute"
+            mm //= timedelta(0, 60)
             s += "%s%02d:%02d" % (sign, hh, mm)
         return s
 
@@ -1560,16 +1561,10 @@
     def utcoffset(self):
         """Return the timezone offset in minutes east of UTC (negative west of
         UTC)."""
-        offset = _call_tzinfo_method(self._tzinfo, "utcoffset", self)
-        offset = _check_utc_offset("utcoffset", offset)
-        if offset is not None:
-            offset = timedelta(minutes=offset)
-        return offset
-
-    # Return an integer (or None) instead of a timedelta (or None).
-    def _utcoffset(self):
-        offset = _call_tzinfo_method(self._tzinfo, "utcoffset", self)
-        offset = _check_utc_offset("utcoffset", offset)
+        if self._tzinfo is None:
+            return None
+        offset = self._tzinfo.utcoffset(self)
+        _check_utc_offset("utcoffset", offset)
         return offset
 
     def tzname(self):
@@ -1592,16 +1587,10 @@
         need to consult dst() unless you're interested in displaying the DST
         info.
         """
-        offset = _call_tzinfo_method(self._tzinfo, "dst", self)
-        offset = _check_utc_offset("dst", offset)
-        if offset is not None:
-            offset = timedelta(minutes=offset)
-        return offset
-
-    # Return an integer (or None) instead of a timedelta (or None).1573
-    def _dst(self):
-        offset = _call_tzinfo_method(self._tzinfo, "dst", self)
-        offset = _check_utc_offset("dst", offset)
+        if self._tzinfo is None:
+            return None
+        offset = self._tzinfo.dst(self)
+        _check_utc_offset("dst", offset)
         return offset
 
     # Comparisons of datetime objects with other.
@@ -1664,9 +1653,9 @@
             base_compare = True
         else:
             if mytz is not None:
-                myoff = self._utcoffset()
+                myoff = self.utcoffset()
             if ottz is not None:
-                otoff = other._utcoffset()
+                otoff = other.utcoffset()
             base_compare = myoff == otoff
 
         if base_compare:
@@ -1721,21 +1710,21 @@
                          self._microsecond - other._microsecond)
         if self._tzinfo is other._tzinfo:
             return base
-        myoff = self._utcoffset()
-        otoff = other._utcoffset()
+        myoff = self.utcoffset()
+        otoff = other.utcoffset()
         if myoff == otoff:
             return base
         if myoff is None or otoff is None:
             raise TypeError("cannot mix naive and timezone-aware time")
-        return base + timedelta(minutes = otoff-myoff)
+        return base + otoff - myoff
 
     def __hash__(self):
-        tzoff = self._utcoffset()
+        tzoff = self.utcoffset()
         if tzoff is None:
             return hash(self._getstate()[0])
         days = _ymd2ord(self.year, self.month, self.day)
-        seconds = self.hour * 3600 + (self.minute - tzoff) * 60 + self.second
-        return hash(timedelta(days, seconds, self.microsecond))
+        seconds = self.hour * 3600 + self.minute * 60 + self.second
+        return hash(timedelta(days, seconds, self.microsecond) - tzoff)
 
     # Pickle support.
 

From python-checkins at python.org  Thu Jul  8 01:56:38 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Thu,  8 Jul 2010 01:56:38 +0200 (CEST)
Subject: [Python-checkins] r82642 -
	python/branches/py3k/Modules/datetimemodule.c
Message-ID: <20100707235638.DA6D2EEC1A@mail.python.org>

Author: alexander.belopolsky
Date: Thu Jul  8 01:56:38 2010
New Revision: 82642

Log:
Issue #5288: Eliminated round-trips between timdelta and int offsets

Modified:
   python/branches/py3k/Modules/datetimemodule.c

Modified: python/branches/py3k/Modules/datetimemodule.c
==============================================================================
--- python/branches/py3k/Modules/datetimemodule.c	(original)
+++ python/branches/py3k/Modules/datetimemodule.c	Thu Jul  8 01:56:38 2010
@@ -88,8 +88,11 @@
 /* p is a pointer to a time or a datetime object; HASTZINFO(p) returns
  * p->hastzinfo.
  */
-#define HASTZINFO(p)            (((_PyDateTime_BaseTZInfo *)(p))->hastzinfo)
-
+#define HASTZINFO(p) (((_PyDateTime_BaseTZInfo *)(p))->hastzinfo)
+#define GET_TIME_TZINFO(p) (HASTZINFO(p) ? \
+                            ((PyDateTime_Time *)(p))->tzinfo : Py_None)
+#define GET_DT_TZINFO(p) (HASTZINFO(p) ? \
+                          ((PyDateTime_DateTime *)(p))->tzinfo : Py_None)
 /* M is a char or int claiming to be a valid month.  The macro is equivalent
  * to the two-sided Python test
  *      1 <= M <= 12
@@ -839,25 +842,6 @@
     return -1;
 }
 
-/* Return tzinfo.methname(tzinfoarg), without any checking of results.
- * If tzinfo is None, returns None.
- */
-static PyObject *
-call_tzinfo_method(PyObject *tzinfo, char *methname, PyObject *tzinfoarg)
-{
-    PyObject *result;
-
-    assert(tzinfo && methname && tzinfoarg);
-    assert(check_tzinfo_subclass(tzinfo) >= 0);
-    if (tzinfo == Py_None) {
-        result = Py_None;
-        Py_INCREF(result);
-    }
-    else
-        result = PyObject_CallMethod(tzinfo, methname, "O", tzinfoarg);
-    return result;
-}
-
 /* If self has a tzinfo member, return a BORROWED reference to it.  Else
  * return NULL, which is NOT AN ERROR.  There are no error returns here,
  * and the caller must not decref the result.
@@ -875,69 +859,53 @@
     return tzinfo;
 }
 
-/* Call getattr(tzinfo, name)(tzinfoarg), and extract an int from the
- * result.  tzinfo must be an instance of the tzinfo class.  If the method
- * returns None, this returns 0 and sets *none to 1.  If the method doesn't
- * return None or timedelta, TypeError is raised and this returns -1.  If it
- * returnsa timedelta and the value is out of range or isn't a whole number
- * of minutes, ValueError is raised and this returns -1.
- * Else *none is set to 0 and the integer method result is returned.
+/* Call getattr(tzinfo, name)(tzinfoarg), and check the result.  tzinfo must
+ * be an instance of the tzinfo class.  If the method returns None, this
+ * returns None.  If the method doesn't return None or timedelta, TypeError is
+ * raised and this returns NULL.  If it returns a timedelta and the value is
+ * out of range or isn't a whole number of minutes, ValueError is raised and
+ * this returns NULL.  Else result is returned.
  */
-static int
-call_utc_tzinfo_method(PyObject *tzinfo, char *name, PyObject *tzinfoarg,
-                       int *none)
+static PyObject *
+call_tzinfo_method(PyObject *tzinfo, char *name, PyObject *tzinfoarg)
 {
-    PyObject *u;
-    int result = -1;
+    PyObject *offset;
 
     assert(tzinfo != NULL);
-    assert(PyTZInfo_Check(tzinfo));
+    assert(PyTZInfo_Check(tzinfo) || tzinfo == Py_None);
     assert(tzinfoarg != NULL);
 
-    *none = 0;
-    u = call_tzinfo_method(tzinfo, name, tzinfoarg);
-    if (u == NULL)
-        return -1;
-
-    else if (u == Py_None) {
-        result = 0;
-        *none = 1;
-    }
-    else if (PyDelta_Check(u)) {
-        const int days = GET_TD_DAYS(u);
-        if (days < -1 || days > 0)
-            result = 24*60;             /* trigger ValueError below */
-        else {
-            /* next line can't overflow because we know days
-             * is -1 or 0 now
-             */
-            int ss = days * 24 * 3600 + GET_TD_SECONDS(u);
-            result = divmod(ss, 60, &ss);
-            if (ss || GET_TD_MICROSECONDS(u)) {
-                PyErr_Format(PyExc_ValueError,
-                             "tzinfo.%s() must return a "
-                             "whole number of minutes",
-                             name);
-                result = -1;
-            }
+    if (tzinfo == Py_None)
+        Py_RETURN_NONE;
+    offset = PyObject_CallMethod(tzinfo, name, "O", tzinfoarg);
+    if (offset == Py_None || offset == NULL)
+        return offset;
+    if (PyDelta_Check(offset)) {
+        if (GET_TD_MICROSECONDS(offset) != 0 || GET_TD_SECONDS(offset) % 60 != 0) {
+            Py_DECREF(offset);
+            PyErr_Format(PyExc_ValueError, "offset must be a timedelta"
+                         " representing a whole number of minutes");
+            return NULL;
+        }
+        if ((GET_TD_DAYS(offset) == -1 && GET_TD_SECONDS(offset) == 0) ||
+            GET_TD_DAYS(offset) < -1 || GET_TD_DAYS(offset) >= 1) {
+            Py_DECREF(offset);
+            PyErr_Format(PyExc_ValueError, "offset must be a timedelta"
+                         " strictly between -timedelta(hours=24) and"
+                         " timedelta(hours=24).");
+            return NULL;
         }
     }
     else {
+        Py_DECREF(offset);
         PyErr_Format(PyExc_TypeError,
                      "tzinfo.%s() must return None or "
-                     "timedelta, not '%s'",
-                     name, Py_TYPE(u)->tp_name);
+                     "timedelta, not '%.200s'",
+                     name, Py_TYPE(offset)->tp_name);
+        return NULL;
     }
 
-    Py_DECREF(u);
-    if (result < -1439 || result > 1439) {
-        PyErr_Format(PyExc_ValueError,
-                     "tzinfo.%s() returned %d; must be in "
-                     "-1439 .. 1439",
-                     name, result);
-        result = -1;
-    }
-    return result;
+    return offset;
 }
 
 /* Call tzinfo.utcoffset(tzinfoarg), and extract an integer from the
@@ -948,37 +916,10 @@
  * # of minutes), ValueError is raised and this returns -1.  Else *none is
  * set to 0 and the offset is returned (as int # of minutes east of UTC).
  */
-static int
-call_utcoffset(PyObject *tzinfo, PyObject *tzinfoarg, int *none)
-{
-    return call_utc_tzinfo_method(tzinfo, "utcoffset", tzinfoarg, none);
-}
-
-/* Call tzinfo.name(tzinfoarg), and return the offset as a timedelta or None.
- */
 static PyObject *
-offset_as_timedelta(PyObject *tzinfo, char *name, PyObject *tzinfoarg) {
-    PyObject *result;
-
-    assert(tzinfo && name && tzinfoarg);
-    if (tzinfo == Py_None) {
-        result = Py_None;
-        Py_INCREF(result);
-    }
-    else {
-        int none;
-        int offset = call_utc_tzinfo_method(tzinfo, name, tzinfoarg,
-                                            &none);
-        if (offset < 0 && PyErr_Occurred())
-            return NULL;
-        if (none) {
-            result = Py_None;
-            Py_INCREF(result);
-        }
-        else
-            result = new_delta(0, offset * 60, 0, 1);
-    }
-    return result;
+call_utcoffset(PyObject *tzinfo, PyObject *tzinfoarg)
+{
+    return call_tzinfo_method(tzinfo, "utcoffset", tzinfoarg);
 }
 
 /* Call tzinfo.dst(tzinfoarg), and extract an integer from the
@@ -989,10 +930,10 @@
  * ValueError is raised and this returns -1.  Else *none is set to 0 and
  * the offset is returned (as an int # of minutes east of UTC).
  */
-static int
-call_dst(PyObject *tzinfo, PyObject *tzinfoarg, int *none)
+static PyObject *
+call_dst(PyObject *tzinfo, PyObject *tzinfoarg)
 {
-    return call_utc_tzinfo_method(tzinfo, "dst", tzinfoarg, none);
+    return call_tzinfo_method(tzinfo, "dst", tzinfoarg);
 }
 
 /* Call tzinfo.tzname(tzinfoarg), and return the result.  tzinfo must be
@@ -1010,102 +951,23 @@
     assert(check_tzinfo_subclass(tzinfo) >= 0);
     assert(tzinfoarg != NULL);
 
-    if (tzinfo == Py_None) {
-        result = Py_None;
-        Py_INCREF(result);
-    }
-    else
-        result = PyObject_CallMethod(tzinfo, "tzname", "O", tzinfoarg);
+    if (tzinfo == Py_None)
+        Py_RETURN_NONE;
 
-    if (result != NULL && result != Py_None) {
-        if (!PyUnicode_Check(result)) {
-            PyErr_Format(PyExc_TypeError, "tzinfo.tzname() must "
-                         "return None or a string, not '%s'",
-                         Py_TYPE(result)->tp_name);
-            Py_DECREF(result);
-            result = NULL;
-        }
-    }
-    return result;
-}
+    result = PyObject_CallMethod(tzinfo, "tzname", "O", tzinfoarg);
 
-typedef enum {
-              /* an exception has been set; the caller should pass it on */
-          OFFSET_ERROR,
-
-          /* type isn't date, datetime, or time subclass */
-          OFFSET_UNKNOWN,
-
-          /* date,
-           * datetime with !hastzinfo
-           * datetime with None tzinfo,
-           * datetime where utcoffset() returns None
-           * time with !hastzinfo
-           * time with None tzinfo,
-           * time where utcoffset() returns None
-           */
-          OFFSET_NAIVE,
-
-          /* time or datetime where utcoffset() doesn't return None */
-          OFFSET_AWARE
-} naivety;
-
-/* Classify an object as to whether it's naive or offset-aware.  See
- * the "naivety" typedef for details.  If the type is aware, *offset is set
- * to minutes east of UTC (as returned by the tzinfo.utcoffset() method).
- * If the type is offset-naive (or unknown, or error), *offset is set to 0.
- * tzinfoarg is the argument to pass to the tzinfo.utcoffset() method.
- */
-static naivety
-classify_utcoffset(PyObject *op, PyObject *tzinfoarg, int *offset)
-{
-    int none;
-    PyObject *tzinfo;
+    if (result == NULL || result == Py_None)
+        return result;
 
-    assert(tzinfoarg != NULL);
-    *offset = 0;
-    tzinfo = get_tzinfo_member(op);     /* NULL means no tzinfo, not error */
-    if (tzinfo == Py_None)
-        return OFFSET_NAIVE;
-    if (tzinfo == NULL) {
-        /* note that a datetime passes the PyDate_Check test */
-        return (PyTime_Check(op) || PyDate_Check(op)) ?
-               OFFSET_NAIVE : OFFSET_UNKNOWN;
-    }
-    *offset = call_utcoffset(tzinfo, tzinfoarg, &none);
-    if (*offset == -1 && PyErr_Occurred())
-        return OFFSET_ERROR;
-    return none ? OFFSET_NAIVE : OFFSET_AWARE;
-}
-
-/* Classify two objects as to whether they're naive or offset-aware.
- * This isn't quite the same as calling classify_utcoffset() twice:  for
- * binary operations (comparison and subtraction), we generally want to
- * ignore the tzinfo members if they're identical.  This is by design,
- * so that results match "naive" expectations when mixing objects from a
- * single timezone.  So in that case, this sets both offsets to 0 and
- * both naiveties to OFFSET_NAIVE.
- * The function returns 0 if everything's OK, and -1 on error.
- */
-static int
-classify_two_utcoffsets(PyObject *o1, int *offset1, naivety *n1,
-                        PyObject *tzinfoarg1,
-                        PyObject *o2, int *offset2, naivety *n2,
-                        PyObject *tzinfoarg2)
-{
-    if (get_tzinfo_member(o1) == get_tzinfo_member(o2)) {
-        *offset1 = *offset2 = 0;
-        *n1 = *n2 = OFFSET_NAIVE;
-    }
-    else {
-        *n1 = classify_utcoffset(o1, tzinfoarg1, offset1);
-        if (*n1 == OFFSET_ERROR)
-            return -1;
-        *n2 = classify_utcoffset(o2, tzinfoarg2, offset2);
-        if (*n2 == OFFSET_ERROR)
-            return -1;
+    if (!PyUnicode_Check(result)) {
+        PyErr_Format(PyExc_TypeError, "tzinfo.tzname() must "
+                     "return None or a string, not '%s'",
+                     Py_TYPE(result)->tp_name);
+        Py_DECREF(result);
+        result = NULL;
     }
-    return 0;
+
+    return result;
 }
 
 /* repr is like "someclass(arg1, arg2)".  If tzinfo isn't None,
@@ -1157,6 +1019,8 @@
                                 GET_YEAR(date));
 }
 
+static PyObject *delta_negative(PyDateTime_Delta *self);
+
 /* Add an hours & minutes UTC offset string to buf.  buf has no more than
  * buflen bytes remaining.  The UTC offset is gotten by calling
  * tzinfo.uctoffset(tzinfoarg).  If that returns None, \0 is stored into
@@ -1171,28 +1035,41 @@
 format_utcoffset(char *buf, size_t buflen, const char *sep,
                 PyObject *tzinfo, PyObject *tzinfoarg)
 {
-    int offset;
-    int hours;
-    int minutes;
+    PyObject *offset;
+    int hours, minutes, seconds;
     char sign;
-    int none;
 
     assert(buflen >= 1);
 
-    offset = call_utcoffset(tzinfo, tzinfoarg, &none);
-    if (offset == -1 && PyErr_Occurred())
+    offset = call_utcoffset(tzinfo, tzinfoarg);
+    if (offset == NULL)
         return -1;
-    if (none) {
+    if (offset == Py_None) {
+        Py_DECREF(offset);
         *buf = '\0';
         return 0;
     }
-    sign = '+';
-    if (offset < 0) {
+    /* Offset is normalized, so it is negative if days < 0 */
+    if (GET_TD_DAYS(offset) < 0) {
+        PyObject *temp = offset;
         sign = '-';
-        offset = - offset;
+        offset = delta_negative((PyDateTime_Delta *)offset);
+        Py_DECREF(temp);
+        if (offset == NULL)
+            return -1;
     }
-    hours = divmod(offset, 60, &minutes);
+    else {
+        sign = '+';
+    }
+    /* Offset is not negative here. */
+    seconds = GET_TD_SECONDS(offset);
+    Py_DECREF(offset);
+    minutes = divmod(seconds, 60, &seconds);
+    hours = divmod(minutes, 60, &minutes);
+    assert(seconds == 0);
+    /* XXX ignore sub-minute data, curently not allowed. */
     PyOS_snprintf(buf, buflen, "%c%02d%s%02d", sign, hours, sep, minutes);
+
     return 0;
 }
 
@@ -1434,7 +1311,7 @@
         format = PyUnicode_FromString(PyBytes_AS_STRING(newfmt));
         if (format != NULL) {
             result = PyObject_CallMethod(time, "strftime", "OO",
-                                         format, timetuple);
+                                         format, timetuple, NULL);
             Py_DECREF(format);
         }
         Py_DECREF(time);
@@ -1937,17 +1814,24 @@
     return result;
 }
 
+static int
+delta_cmp(PyObject *self, PyObject *other)
+{
+    int diff = GET_TD_DAYS(self) - GET_TD_DAYS(other);
+    if (diff == 0) {
+        diff = GET_TD_SECONDS(self) - GET_TD_SECONDS(other);
+        if (diff == 0)
+            diff = GET_TD_MICROSECONDS(self) -
+                GET_TD_MICROSECONDS(other);
+    }
+    return diff;
+}
+
 static PyObject *
 delta_richcompare(PyObject *self, PyObject *other, int op)
 {
     if (PyDelta_Check(other)) {
-        int diff = GET_TD_DAYS(self) - GET_TD_DAYS(other);
-        if (diff == 0) {
-            diff = GET_TD_SECONDS(self) - GET_TD_SECONDS(other);
-            if (diff == 0)
-                diff = GET_TD_MICROSECONDS(self) -
-                       GET_TD_MICROSECONDS(other);
-        }
+        int diff = delta_cmp(self, other);
         return diff_to_bool(diff, op);
     }
     else {
@@ -3119,76 +3003,73 @@
     return tzinfo_nogo("dst");
 }
 
+
+static PyObject *add_datetime_timedelta(PyDateTime_DateTime *date,
+                                        PyDateTime_Delta *delta,
+                                        int factor);
+static PyObject *datetime_utcoffset(PyObject *self, PyObject *);
+static PyObject *datetime_dst(PyObject *self, PyObject *);
+
 static PyObject *
-tzinfo_fromutc(PyDateTime_TZInfo *self, PyDateTime_DateTime *dt)
+tzinfo_fromutc(PyDateTime_TZInfo *self, PyObject *dt)
 {
-    int y, m, d, hh, mm, ss, us;
-
-    PyObject *result;
-    int off, dst;
-    int none;
-    int delta;
+    PyObject *result = NULL;
+    PyObject *off = NULL, *dst = NULL;
+    PyDateTime_Delta *delta = NULL;
 
-    if (! PyDateTime_Check(dt)) {
+    if (!PyDateTime_Check(dt)) {
         PyErr_SetString(PyExc_TypeError,
                         "fromutc: argument must be a datetime");
         return NULL;
     }
-    if (! HASTZINFO(dt) || dt->tzinfo != (PyObject *)self) {
+    if (GET_DT_TZINFO(dt) != (PyObject *)self) {
         PyErr_SetString(PyExc_ValueError, "fromutc: dt.tzinfo "
                         "is not self");
         return NULL;
     }
 
-    off = call_utcoffset(dt->tzinfo, (PyObject *)dt, &none);
-    if (off == -1 && PyErr_Occurred())
+    off = datetime_utcoffset(dt, NULL);
+    if (off == NULL)
         return NULL;
-    if (none) {
+    if (off == Py_None) {
         PyErr_SetString(PyExc_ValueError, "fromutc: non-None "
                         "utcoffset() result required");
-        return NULL;
+        goto Fail;
     }
 
-    dst = call_dst(dt->tzinfo, (PyObject *)dt, &none);
-    if (dst == -1 && PyErr_Occurred())
-        return NULL;
-    if (none) {
+    dst = datetime_dst(dt, NULL);
+    if (dst == NULL)
+        goto Fail;
+    if (dst == Py_None) {
         PyErr_SetString(PyExc_ValueError, "fromutc: non-None "
                         "dst() result required");
-        return NULL;
+        goto Fail;
     }
 
-    y = GET_YEAR(dt);
-    m = GET_MONTH(dt);
-    d = GET_DAY(dt);
-    hh = DATE_GET_HOUR(dt);
-    mm = DATE_GET_MINUTE(dt);
-    ss = DATE_GET_SECOND(dt);
-    us = DATE_GET_MICROSECOND(dt);
-
-    delta = off - dst;
-    mm += delta;
-    if ((mm < 0 || mm >= 60) &&
-        normalize_datetime(&y, &m, &d, &hh, &mm, &ss, &us) < 0)
-        return NULL;
-    result = new_datetime(y, m, d, hh, mm, ss, us, dt->tzinfo);
+    delta = (PyDateTime_Delta *)delta_subtract(off, dst);
+    if (delta == NULL)
+        goto Fail;
+    result = add_datetime_timedelta((PyDateTime_DateTime *)dt, delta, 1);
     if (result == NULL)
-        return result;
-
-    dst = call_dst(dt->tzinfo, result, &none);
-    if (dst == -1 && PyErr_Occurred())
         goto Fail;
-    if (none)
-        goto Inconsistent;
-    if (dst == 0)
-        return result;
 
-    mm += dst;
-    if ((mm < 0 || mm >= 60) &&
-        normalize_datetime(&y, &m, &d, &hh, &mm, &ss, &us) < 0)
+    Py_DECREF(dst);
+    dst = call_dst(GET_DT_TZINFO(dt), result);
+    if (dst == NULL)
         goto Fail;
-    Py_DECREF(result);
-    result = new_datetime(y, m, d, hh, mm, ss, us, dt->tzinfo);
+    if (dst == Py_None)
+        goto Inconsistent;
+    if (delta_bool(delta) != 0) {
+        PyObject *temp = result;
+        result = add_datetime_timedelta((PyDateTime_DateTime *)result,
+                                        (PyDateTime_Delta *)dst, 1);
+        Py_DECREF(temp);
+        if (result == NULL)
+            goto Fail;
+    }
+    Py_DECREF(delta);
+    Py_DECREF(dst);
+    Py_DECREF(off);
     return result;
 
 Inconsistent:
@@ -3197,7 +3078,10 @@
 
     /* fall thru to failure */
 Fail:
-    Py_DECREF(result);
+    Py_XDECREF(off);
+    Py_XDECREF(dst);
+    Py_XDECREF(delta);
+    Py_XDECREF(result);
     return NULL;
 }
 
@@ -3464,18 +3348,14 @@
 }
 
 static PyObject *
-add_datetime_timedelta(PyDateTime_DateTime *date, PyDateTime_Delta *delta,
-                       int factor);
-
-static PyObject *
 timezone_fromutc(PyDateTime_TimeZone *self, PyDateTime_DateTime *dt)
 {
-    if (! PyDateTime_Check(dt)) {
+    if (!PyDateTime_Check(dt)) {
         PyErr_SetString(PyExc_TypeError,
                         "fromutc: argument must be a datetime");
         return NULL;
     }
-    if (! HASTZINFO(dt) || dt->tzinfo != (PyObject *)self) {
+    if (!HASTZINFO(dt) || dt->tzinfo != (PyObject *)self) {
         PyErr_SetString(PyExc_ValueError, "fromutc: dt.tzinfo "
                         "is not self");
         return NULL;
@@ -3689,21 +3569,18 @@
 
 /* These are all METH_NOARGS, so don't need to check the arglist. */
 static PyObject *
-time_utcoffset(PyDateTime_Time *self, PyObject *unused) {
-    return offset_as_timedelta(HASTZINFO(self) ? self->tzinfo : Py_None,
-                               "utcoffset", Py_None);
+time_utcoffset(PyObject *self, PyObject *unused) {
+    return call_utcoffset(GET_TIME_TZINFO(self), Py_None);
 }
 
 static PyObject *
-time_dst(PyDateTime_Time *self, PyObject *unused) {
-    return offset_as_timedelta(HASTZINFO(self) ? self->tzinfo : Py_None,
-                               "dst", Py_None);
+time_dst(PyObject *self, PyObject *unused) {
+    return call_dst(GET_TIME_TZINFO(self), Py_None);
 }
 
 static PyObject *
 time_tzname(PyDateTime_Time *self, PyObject *unused) {
-    return call_tzname(HASTZINFO(self) ? self->tzinfo : Py_None,
-                       Py_None);
+    return call_tzname(GET_TIME_TZINFO(self), Py_None);
 }
 
 /*
@@ -3758,7 +3635,7 @@
                                       TIME_GET_MINUTE(self),
                                       TIME_GET_SECOND(self));
 
-    if (result == NULL || ! HASTZINFO(self) || self->tzinfo == Py_None)
+    if (result == NULL || !HASTZINFO(self) || self->tzinfo == Py_None)
         return result;
 
     /* We need to append the UTC offset. */
@@ -3809,98 +3686,108 @@
 static PyObject *
 time_richcompare(PyObject *self, PyObject *other, int op)
 {
+    PyObject *result = NULL;
+    PyObject *offset1, *offset2;
     int diff;
-    naivety n1, n2;
-    int offset1, offset2;
 
     if (! PyTime_Check(other)) {
         Py_INCREF(Py_NotImplemented);
         return Py_NotImplemented;
     }
-    if (classify_two_utcoffsets(self, &offset1, &n1, Py_None,
-                                other, &offset2, &n2, Py_None) < 0)
+
+    if (GET_TIME_TZINFO(self) == GET_TIME_TZINFO(other)) {
+        diff = memcmp(((PyDateTime_Time *)self)->data,
+                      ((PyDateTime_Time *)other)->data,
+                      _PyDateTime_TIME_DATASIZE);
+        return diff_to_bool(diff, op);
+    }
+    offset1 = time_utcoffset(self, NULL);
+    if (offset1 == NULL)
         return NULL;
-    assert(n1 != OFFSET_UNKNOWN && n2 != OFFSET_UNKNOWN);
+    offset2 = time_utcoffset(other, NULL);
+    if (offset2 == NULL)
+        goto done;
     /* If they're both naive, or both aware and have the same offsets,
      * we get off cheap.  Note that if they're both naive, offset1 ==
-     * offset2 == 0 at this point.
+     * offset2 == Py_None at this point.
      */
-    if (n1 == n2 && offset1 == offset2) {
+    if ((offset1 == offset2) ||
+        (PyDelta_Check(offset1) && PyDelta_Check(offset2) &&
+         delta_cmp(offset1, offset2) == 0)) {
         diff = memcmp(((PyDateTime_Time *)self)->data,
                       ((PyDateTime_Time *)other)->data,
                       _PyDateTime_TIME_DATASIZE);
-        return diff_to_bool(diff, op);
+        result = diff_to_bool(diff, op);
     }
-
-    if (n1 == OFFSET_AWARE && n2 == OFFSET_AWARE) {
-        assert(offset1 != offset2);             /* else last "if" handled it */
-        /* Convert everything except microseconds to seconds.  These
-         * can't overflow (no more than the # of seconds in 2 days).
-         */
-        offset1 = TIME_GET_HOUR(self) * 3600 +
-                  (TIME_GET_MINUTE(self) - offset1) * 60 +
-                  TIME_GET_SECOND(self);
-        offset2 = TIME_GET_HOUR(other) * 3600 +
-                  (TIME_GET_MINUTE(other) - offset2) * 60 +
-                  TIME_GET_SECOND(other);
-        diff = offset1 - offset2;
+    /* The hard case: both aware with different UTC offsets */
+    else if (offset1 != Py_None && offset2 != Py_None) {
+        int offsecs1, offsecs2;
+        assert(offset1 != offset2); /* else last "if" handled it */
+        offsecs1 = TIME_GET_HOUR(self) * 3600 +
+                   TIME_GET_MINUTE(self) * 60 +
+                   TIME_GET_SECOND(self) -
+                   GET_TD_DAYS(offset1) * 86400 -
+                   GET_TD_SECONDS(offset1);
+        offsecs2 = TIME_GET_HOUR(other) * 3600 +
+                   TIME_GET_MINUTE(other) * 60 +
+                   TIME_GET_SECOND(other) -
+                   GET_TD_DAYS(offset2) * 86400 -
+                   GET_TD_SECONDS(offset2);
+        diff = offsecs1 - offsecs2;
         if (diff == 0)
             diff = TIME_GET_MICROSECOND(self) -
                    TIME_GET_MICROSECOND(other);
-        return diff_to_bool(diff, op);
+        result = diff_to_bool(diff, op);
     }
-
-    assert(n1 != n2);
-    PyErr_SetString(PyExc_TypeError,
-                    "can't compare offset-naive and "
-                    "offset-aware times");
-    return NULL;
+    else {
+        PyErr_SetString(PyExc_TypeError,
+                        "can't compare offset-naive and "
+                        "offset-aware times");
+    }
+ done:
+    Py_DECREF(offset1);
+    Py_XDECREF(offset2);
+    return result;
 }
 
 static long
 time_hash(PyDateTime_Time *self)
 {
     if (self->hashcode == -1) {
-        naivety n;
-        int offset;
-        PyObject *temp;
-
-        n = classify_utcoffset((PyObject *)self, Py_None, &offset);
-        assert(n != OFFSET_UNKNOWN);
-        if (n == OFFSET_ERROR)
+        PyObject *offset;
+
+        offset = time_utcoffset((PyObject *)self, NULL);
+
+        if (offset == NULL)
             return -1;
 
         /* Reduce this to a hash of another object. */
-        if (offset == 0) {
+        if (offset == Py_None)
             self->hashcode = generic_hash(
                 (unsigned char *)self->data, _PyDateTime_TIME_DATASIZE);
-            return self->hashcode;
-        }
         else {
-            int hour;
-            int minute;
-
-            assert(n == OFFSET_AWARE);
+            PyObject *temp1, *temp2;
+            int seconds, microseconds;
             assert(HASTZINFO(self));
-            hour = divmod(TIME_GET_HOUR(self) * 60 +
-                            TIME_GET_MINUTE(self) - offset,
-                          60,
-                          &minute);
-            if (0 <= hour && hour < 24)
-                temp = new_time(hour, minute,
-                                TIME_GET_SECOND(self),
-                                TIME_GET_MICROSECOND(self),
-                                Py_None);
-            else
-                temp = Py_BuildValue("iiii",
-                           hour, minute,
-                           TIME_GET_SECOND(self),
-                           TIME_GET_MICROSECOND(self));
-        }
-        if (temp != NULL) {
-            self->hashcode = PyObject_Hash(temp);
-            Py_DECREF(temp);
+            seconds = TIME_GET_HOUR(self) * 3600 +
+                      TIME_GET_MINUTE(self) * 60 +
+                      TIME_GET_SECOND(self);
+            microseconds = TIME_GET_MICROSECOND(self);
+            temp1 = new_delta(0, seconds, microseconds, 1);
+            if (temp1 == NULL) {
+                Py_DECREF(offset);
+                return -1;
+            }
+            temp2 = delta_subtract(temp1, offset);
+            Py_DECREF(temp1);
+            if (temp2 == NULL) {
+                Py_DECREF(offset);
+                return -1;
+            }
+            self->hashcode = PyObject_Hash(temp2);
+            Py_DECREF(temp2);
         }
+        Py_DECREF(offset);
     }
     return self->hashcode;
 }
@@ -3929,10 +3816,10 @@
 }
 
 static int
-time_bool(PyDateTime_Time *self)
+time_bool(PyObject *self)
 {
-    int offset;
-    int none;
+    PyObject *offset, *tzinfo;
+    int offsecs = 0;
 
     if (TIME_GET_SECOND(self) || TIME_GET_MICROSECOND(self)) {
         /* Since utcoffset is in whole minutes, nothing can
@@ -3940,13 +3827,15 @@
          */
         return 1;
     }
-    offset = 0;
-    if (HASTZINFO(self) && self->tzinfo != Py_None) {
-        offset = call_utcoffset(self->tzinfo, Py_None, &none);
-        if (offset == -1 && PyErr_Occurred())
+    tzinfo = GET_TIME_TZINFO(self);
+    if (tzinfo != Py_None) {
+        offset = call_utcoffset(tzinfo, Py_None);
+        if (offset == NULL)
             return -1;
+        offsecs = GET_TD_DAYS(offset)*86400 + GET_TD_SECONDS(offset);
+        Py_DECREF(offset);
     }
-    return (TIME_GET_MINUTE(self) - offset + TIME_GET_HOUR(self)*60) != 0;
+    return (TIME_GET_MINUTE(self)*60 - offsecs + TIME_GET_HOUR(self)*3600) != 0;
 }
 
 /* Pickle support, a simple use of __reduce__. */
@@ -4455,21 +4344,18 @@
 
 /* These are all METH_NOARGS, so don't need to check the arglist. */
 static PyObject *
-datetime_utcoffset(PyDateTime_DateTime *self, PyObject *unused) {
-    return offset_as_timedelta(HASTZINFO(self) ? self->tzinfo : Py_None,
-                               "utcoffset", (PyObject *)self);
+datetime_utcoffset(PyObject *self, PyObject *unused) {
+    return call_utcoffset(GET_DT_TZINFO(self), self);
 }
 
 static PyObject *
-datetime_dst(PyDateTime_DateTime *self, PyObject *unused) {
-    return offset_as_timedelta(HASTZINFO(self) ? self->tzinfo : Py_None,
-                               "dst", (PyObject *)self);
+datetime_dst(PyObject *self, PyObject *unused) {
+    return call_dst(GET_DT_TZINFO(self), self);
 }
 
 static PyObject *
-datetime_tzname(PyDateTime_DateTime *self, PyObject *unused) {
-    return call_tzname(HASTZINFO(self) ? self->tzinfo : Py_None,
-                       (PyObject *)self);
+datetime_tzname(PyObject *self, PyObject *unused) {
+    return call_tzname(GET_DT_TZINFO(self), self);
 }
 
 /*
@@ -4536,21 +4422,43 @@
         /* datetime - ??? */
         if (PyDateTime_Check(right)) {
             /* datetime - datetime */
-            naivety n1, n2;
-            int offset1, offset2;
+            PyObject *offset1, *offset2, *offdiff = NULL;
             int delta_d, delta_s, delta_us;
 
-            if (classify_two_utcoffsets(left, &offset1, &n1, left,
-                                        right, &offset2, &n2,
-                                        right) < 0)
-                return NULL;
-            assert(n1 != OFFSET_UNKNOWN && n2 != OFFSET_UNKNOWN);
-            if (n1 != n2) {
-                PyErr_SetString(PyExc_TypeError,
-                    "can't subtract offset-naive and "
-                    "offset-aware datetimes");
-                return NULL;
+            if (GET_DT_TZINFO(left) == GET_DT_TZINFO(right)) {
+                offset2 = offset1 = Py_None;
+                Py_INCREF(offset1);
+                Py_INCREF(offset2);
             }
+            else {
+                offset1 = datetime_utcoffset(left, NULL);
+                if (offset1 == NULL)
+                    return NULL;
+                offset2 = datetime_utcoffset(right, NULL);
+                if (offset2 == NULL) {
+                    Py_DECREF(offset1);
+                    return NULL;
+                }
+                if ((offset1 != Py_None) != (offset2 != Py_None)) {
+                    PyErr_SetString(PyExc_TypeError,
+                                    "can't subtract offset-naive and "
+                                    "offset-aware datetimes");
+                    Py_DECREF(offset1);
+                    Py_DECREF(offset2);
+                    return NULL;
+                }
+            }
+            if ((offset1 != offset2) &&
+                delta_cmp(offset1, offset2) != 0) {
+                offdiff = delta_subtract(offset1, offset2);
+                if (offdiff == NULL) {
+                    Py_DECREF(offset1);
+                    Py_DECREF(offset2);
+                    return NULL;
+                }
+            }
+            Py_DECREF(offset1);
+            Py_DECREF(offset2);
             delta_d = ymd_to_ord(GET_YEAR(left),
                                  GET_MONTH(left),
                                  GET_DAY(left)) -
@@ -4569,11 +4477,13 @@
                        DATE_GET_SECOND(right));
             delta_us = DATE_GET_MICROSECOND(left) -
                        DATE_GET_MICROSECOND(right);
-            /* (left - offset1) - (right - offset2) =
-             * (left - right) + (offset2 - offset1)
-             */
-            delta_s += (offset2 - offset1) * 60;
             result = new_delta(delta_d, delta_s, delta_us, 1);
+            if (offdiff != NULL) {
+                PyObject *temp = result;
+                result = delta_subtract(result, offdiff);
+                Py_DECREF(temp);
+                Py_DECREF(offdiff);
+            }
         }
         else if (PyDelta_Check(right)) {
             /* datetime - delta */
@@ -4683,9 +4593,9 @@
 static PyObject *
 datetime_richcompare(PyObject *self, PyObject *other, int op)
 {
+    PyObject *result = NULL;
+    PyObject *offset1, *offset2;
     int diff;
-    naivety n1, n2;
-    int offset1, offset2;
 
     if (! PyDateTime_Check(other)) {
         if (PyDate_Check(other)) {
@@ -4706,85 +4616,99 @@
         return Py_NotImplemented;
     }
 
-    if (classify_two_utcoffsets(self, &offset1, &n1, self,
-                                other, &offset2, &n2, other) < 0)
+    if (GET_DT_TZINFO(self) == GET_DT_TZINFO(other)) {
+        diff = memcmp(((PyDateTime_DateTime *)self)->data,
+                      ((PyDateTime_DateTime *)other)->data,
+                      _PyDateTime_DATETIME_DATASIZE);
+        return diff_to_bool(diff, op);
+    }
+    offset1 = datetime_utcoffset(self, NULL);
+    if (offset1 == NULL)
         return NULL;
-    assert(n1 != OFFSET_UNKNOWN && n2 != OFFSET_UNKNOWN);
+    offset2 = datetime_utcoffset(other, NULL);
+    if (offset2 == NULL)
+        goto done;
     /* If they're both naive, or both aware and have the same offsets,
      * we get off cheap.  Note that if they're both naive, offset1 ==
-     * offset2 == 0 at this point.
+     * offset2 == Py_None at this point.
      */
-    if (n1 == n2 && offset1 == offset2) {
+    if ((offset1 == offset2) ||
+        (PyDelta_Check(offset1) && PyDelta_Check(offset2) &&
+         delta_cmp(offset1, offset2) == 0)) {
         diff = memcmp(((PyDateTime_DateTime *)self)->data,
                       ((PyDateTime_DateTime *)other)->data,
                       _PyDateTime_DATETIME_DATASIZE);
-        return diff_to_bool(diff, op);
+        result = diff_to_bool(diff, op);
     }
-
-    if (n1 == OFFSET_AWARE && n2 == OFFSET_AWARE) {
+    else if (offset1 != Py_None && offset2 != Py_None) {
         PyDateTime_Delta *delta;
 
-        assert(offset1 != offset2);             /* else last "if" handled it */
+        assert(offset1 != offset2); /* else last "if" handled it */
         delta = (PyDateTime_Delta *)datetime_subtract((PyObject *)self,
                                                        other);
         if (delta == NULL)
-            return NULL;
+            goto done;
         diff = GET_TD_DAYS(delta);
         if (diff == 0)
             diff = GET_TD_SECONDS(delta) |
                    GET_TD_MICROSECONDS(delta);
         Py_DECREF(delta);
-        return diff_to_bool(diff, op);
+        result = diff_to_bool(diff, op);
     }
-
-    assert(n1 != n2);
-    PyErr_SetString(PyExc_TypeError,
-                    "can't compare offset-naive and "
-                    "offset-aware datetimes");
-    return NULL;
+    else {
+        PyErr_SetString(PyExc_TypeError,
+                        "can't compare offset-naive and "
+                        "offset-aware datetimes");
+    }
+ done:
+    Py_DECREF(offset1);
+    Py_XDECREF(offset2);
+    return result;
 }
 
 static long
 datetime_hash(PyDateTime_DateTime *self)
 {
     if (self->hashcode == -1) {
-        naivety n;
-        int offset;
-        PyObject *temp;
-
-        n = classify_utcoffset((PyObject *)self, (PyObject *)self,
-                               &offset);
-        assert(n != OFFSET_UNKNOWN);
-        if (n == OFFSET_ERROR)
+        PyObject *offset;
+
+        offset = datetime_utcoffset((PyObject *)self, NULL);
+
+        if (offset == NULL)
             return -1;
 
         /* Reduce this to a hash of another object. */
-        if (n == OFFSET_NAIVE) {
+        if (offset == Py_None)
             self->hashcode = generic_hash(
                 (unsigned char *)self->data, _PyDateTime_DATETIME_DATASIZE);
-            return self->hashcode;
-        }
         else {
-            int days;
-            int seconds;
+            PyObject *temp1, *temp2;
+            int days, seconds;
 
-            assert(n == OFFSET_AWARE);
             assert(HASTZINFO(self));
             days = ymd_to_ord(GET_YEAR(self),
                               GET_MONTH(self),
                               GET_DAY(self));
             seconds = DATE_GET_HOUR(self) * 3600 +
-                      (DATE_GET_MINUTE(self) - offset) * 60 +
+                      DATE_GET_MINUTE(self) * 60 +
                       DATE_GET_SECOND(self);
-            temp = new_delta(days,
-                             seconds,
-                             DATE_GET_MICROSECOND(self),
-                             1);
-        }
-        if (temp != NULL) {
-            self->hashcode = PyObject_Hash(temp);
-            Py_DECREF(temp);
+            temp1 = new_delta(days, seconds,
+                              DATE_GET_MICROSECOND(self),
+                              1);
+            if (temp1 == NULL) {
+                Py_DECREF(offset);
+                return -1;
+            }
+            temp2 = delta_subtract(temp1, offset);
+            Py_DECREF(temp1);
+            if (temp2 == NULL) {
+                Py_DECREF(offset);
+                return -1;
+            }
+            self->hashcode = PyObject_Hash(temp2);
+            Py_DECREF(temp2);
         }
+        Py_DECREF(offset);
     }
     return self->hashcode;
 }
@@ -4819,10 +4743,9 @@
 static PyObject *
 datetime_astimezone(PyDateTime_DateTime *self, PyObject *args, PyObject *kw)
 {
-    int y, m, d, hh, mm, ss, us;
     PyObject *result;
-    int offset, none;
-
+    PyObject *offset;
+    PyObject *temp;
     PyObject *tzinfo;
     static char *keywords[] = {"tz", NULL};
 
@@ -4840,39 +4763,35 @@
     }
 
     /* Convert self to UTC. */
-    offset = call_utcoffset(self->tzinfo, (PyObject *)self, &none);
-    if (offset == -1 && PyErr_Occurred())
+    offset = datetime_utcoffset((PyObject *)self, NULL);
+    if (offset == NULL)
         return NULL;
-    if (none)
-        goto NeedAware;
+    if (offset == Py_None) {
+        Py_DECREF(offset);
+      NeedAware:
+        PyErr_SetString(PyExc_ValueError, "astimezone() cannot be applied to "
+                        "a naive datetime");
+        return NULL;
+    }
 
-    y = GET_YEAR(self);
-    m = GET_MONTH(self);
-    d = GET_DAY(self);
-    hh = DATE_GET_HOUR(self);
-    mm = DATE_GET_MINUTE(self);
-    ss = DATE_GET_SECOND(self);
-    us = DATE_GET_MICROSECOND(self);
-
-    mm -= offset;
-    if ((mm < 0 || mm >= 60) &&
-        normalize_datetime(&y, &m, &d, &hh, &mm, &ss, &us) < 0)
+    /* result = self - offset */
+    result = add_datetime_timedelta(self,
+                (PyDateTime_Delta *)offset, -1);
+    Py_DECREF(offset);
+    if (result == NULL)
         return NULL;
 
     /* Attach new tzinfo and let fromutc() do the rest. */
-    result = new_datetime(y, m, d, hh, mm, ss, us, tzinfo);
-    if (result != NULL) {
-        PyObject *temp = result;
+    temp = ((PyDateTime_DateTime *)result)->tzinfo;
+    ((PyDateTime_DateTime *)result)->tzinfo = tzinfo;
+    Py_INCREF(tzinfo);
+    Py_DECREF(temp);
 
-        result = PyObject_CallMethod(tzinfo, "fromutc", "O", temp);
-        Py_DECREF(temp);
-    }
-    return result;
+    temp = result;
+    result = PyObject_CallMethod(tzinfo, "fromutc", "O", temp);
+    Py_DECREF(temp);
 
-NeedAware:
-    PyErr_SetString(PyExc_ValueError, "astimezone() cannot be applied to "
-                                      "a naive datetime");
-    return NULL;
+    return result;
 }
 
 static PyObject *
@@ -4881,17 +4800,15 @@
     int dstflag = -1;
 
     if (HASTZINFO(self) && self->tzinfo != Py_None) {
-        int none;
+        PyObject * dst;
 
-        dstflag = call_dst(self->tzinfo, (PyObject *)self, &none);
-        if (dstflag == -1 && PyErr_Occurred())
+        dst = call_dst(self->tzinfo, (PyObject *)self);
+        if (dst == NULL)
             return NULL;
 
-        if (none)
-            dstflag = -1;
-        else if (dstflag != 0)
-            dstflag = 1;
-
+        if (dst != Py_None)
+            dstflag = delta_bool((PyDateTime_Delta *)dst);
+        Py_DECREF(dst);
     }
     return build_struct_time(GET_YEAR(self),
                              GET_MONTH(self),
@@ -4927,41 +4844,47 @@
                     DATE_GET_MINUTE(self),
                     DATE_GET_SECOND(self),
                     DATE_GET_MICROSECOND(self),
-                    HASTZINFO(self) ? self->tzinfo : Py_None);
+                    GET_DT_TZINFO(self));
 }
 
 static PyObject *
 datetime_utctimetuple(PyDateTime_DateTime *self)
 {
-    int y = GET_YEAR(self);
-    int m = GET_MONTH(self);
-    int d = GET_DAY(self);
-    int hh = DATE_GET_HOUR(self);
-    int mm = DATE_GET_MINUTE(self);
-    int ss = DATE_GET_SECOND(self);
-    int us = 0;         /* microseconds are ignored in a timetuple */
-    int offset = 0;
-
-    if (HASTZINFO(self) && self->tzinfo != Py_None) {
-        int none;
+    int y, m, d, hh, mm, ss;
+    PyObject *tzinfo;
+    PyDateTime_DateTime *utcself;
 
-        offset = call_utcoffset(self->tzinfo, (PyObject *)self, &none);
-        if (offset == -1 && PyErr_Occurred())
-            return NULL;
+    tzinfo = GET_DT_TZINFO(self);
+    if (tzinfo == Py_None) {
+        utcself = self;
+        Py_INCREF(utcself);
     }
-    /* Even if offset is 0, don't call timetuple() -- tm_isdst should be
-     * 0 in a UTC timetuple regardless of what dst() says.
-     */
-    if (offset) {
-        /* Subtract offset minutes & normalize. */
-        int stat;
-
-        mm -= offset;
-        stat = normalize_datetime(&y, &m, &d, &hh, &mm, &ss, &us);
-        /* OverflowError may be raised in the edge cases. */
-        if (stat < 0)
+    else {
+        PyObject *offset;
+        offset = call_utcoffset(tzinfo, (PyObject *)self);
+        if (offset == NULL)
             return NULL;
+        if (offset == Py_None) {
+            Py_DECREF(offset);
+            utcself = self;
+            Py_INCREF(utcself);
+        }
+        else {
+            utcself = (PyDateTime_DateTime *)add_datetime_timedelta(self,
+                                                (PyDateTime_Delta *)offset, -1);
+            Py_DECREF(offset);
+            if (utcself == NULL)
+                return NULL;
+        }
     }
+    y = GET_YEAR(utcself);
+    m = GET_MONTH(utcself);
+    d = GET_DAY(utcself);
+    hh = DATE_GET_HOUR(utcself);
+    mm = DATE_GET_MINUTE(utcself);
+    ss = DATE_GET_SECOND(utcself);
+
+    Py_DECREF(utcself);
     return build_struct_time(y, m, d, hh, mm, ss, 0);
 }
 

From python-checkins at python.org  Thu Jul  8 02:13:56 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Thu,  8 Jul 2010 02:13:56 +0200 (CEST)
Subject: [Python-checkins] r82643 -
	sandbox/branches/py3k-datetime/datetime.py
Message-ID: <20100708001356.90D75EEC05@mail.python.org>

Author: alexander.belopolsky
Date: Thu Jul  8 02:13:56 2010
New Revision: 82643

Log:
Use named argument in timedelta.

Modified:
   sandbox/branches/py3k-datetime/datetime.py

Modified: sandbox/branches/py3k-datetime/datetime.py
==============================================================================
--- sandbox/branches/py3k-datetime/datetime.py	(original)
+++ sandbox/branches/py3k-datetime/datetime.py	Thu Jul  8 02:13:56 2010
@@ -208,8 +208,8 @@
                                     offset = -offset
                                     sign = '-'
                                 h, m = divmod(offset, timedelta(hours=1))
-                                assert not m % timedelta(0, 60), "whole minute"
-                                m //= timedelta(0, 60)
+                                assert not m % timedelta(minutes=1), "whole minute"
+                                m //= timedelta(minutes=1)
                                 zreplace = '%c%02d%02d' % (sign, h, m)
                     assert '%' not in zreplace
                     newformat.append(zreplace)
@@ -256,7 +256,7 @@
     if not isinstance(offset, timedelta):
         raise TypeError("tzinfo.%s() must return None "
                         "or timedelta, not '%s'" % (name, type(offset)))
-    if offset % timedelta(0, 60) or offset.microseconds:
+    if offset % timedelta(minutes=1) or offset.microseconds:
         raise ValueError("tzinfo.%s() must return a whole number "
                          "of minutes, got %s" % (name, offset))
     if not -timedelta(1) < offset < timedelta(1):
@@ -1119,8 +1119,8 @@
                         other._microsecond))
         if myoff is None or otoff is None:
             raise TypeError("cannot compare naive and aware times")
-        myhhmm = self._hour * 60 + self._minute - myoff//timedelta(0, 60)
-        othhmm = other._hour * 60 + other._minute - otoff//timedelta(0, 60)
+        myhhmm = self._hour * 60 + self._minute - myoff//timedelta(minutes=1)
+        othhmm = other._hour * 60 + other._minute - otoff//timedelta(minutes=1)
         return _cmp((myhhmm, self._second, self._microsecond),
                     (othhmm, other._second, other._microsecond))
 
@@ -1131,8 +1131,8 @@
             return hash(self._getstate()[0])
         h, m = divmod(timedelta(hours=self.hour, minutes=self.minute) - tzoff,
                       timedelta(hours=1))
-        assert not m % timedelta(0, 60), "whole minute"
-        m //= timedelta(0, 60)
+        assert not m % timedelta(minutes=1), "whole minute"
+        m //= timedelta(minutes=1)
         if 0 <= h < 24:
             return hash(time(h, m, self.second, self.microsecond))
         return hash((h, m, self.second, self.microsecond))
@@ -1149,8 +1149,8 @@
             else:
                 sign = "+"
             hh, mm = divmod(off, timedelta(hours=1))
-            assert not mm % timedelta(0, 60), "whole minute"
-            mm //= timedelta(0, 60)
+            assert not mm % timedelta(minutes=1), "whole minute"
+            mm //= timedelta(minutes=1)
             assert 0 <= hh < 24
             off = "%s%02d%s%02d" % (sign, hh, sep, mm)
         return off
@@ -1528,8 +1528,8 @@
             else:
                 sign = "+"
             hh, mm = divmod(off, timedelta(hours=1))
-            assert not mm % timedelta(0, 60), "whole minute"
-            mm //= timedelta(0, 60)
+            assert not mm % timedelta(minutes=1), "whole minute"
+            mm //= timedelta(minutes=1)
             s += "%s%02d:%02d" % (sign, hh, mm)
         return s
 

From nnorwitz at gmail.com  Thu Jul  8 12:21:35 2010
From: nnorwitz at gmail.com (Neal Norwitz)
Date: Thu, 8 Jul 2010 06:21:35 -0400
Subject: [Python-checkins] Python Regression Test Failures refleak (1)
Message-ID: <20100708102135.GA16716@kbk-i386-bb.psfb.org>

More important issues:
----------------------
test_bz2 leaked [-82, 0, 83] references, sum=1

Less important issues:
----------------------

From python-checkins at python.org  Thu Jul  8 17:03:02 2010
From: python-checkins at python.org (ezio.melotti)
Date: Thu,  8 Jul 2010 17:03:02 +0200 (CEST)
Subject: [Python-checkins] r82644 - python/branches/py3k/Modules/mathmodule.c
Message-ID: <20100708150302.3593BEED3B@mail.python.org>

Author: ezio.melotti
Date: Thu Jul  8 17:03:02 2010
New Revision: 82644

Log:
Fix typo in a comment in mathmodule.c.

Modified:
   python/branches/py3k/Modules/mathmodule.c

Modified: python/branches/py3k/Modules/mathmodule.c
==============================================================================
--- python/branches/py3k/Modules/mathmodule.c	(original)
+++ python/branches/py3k/Modules/mathmodule.c	Thu Jul  8 17:03:02 2010
@@ -1170,7 +1170,7 @@
  * http://www.luschny.de/math/factorial/binarysplitfact.html
  *
  * Faster algorithms exist, but they're more complicated and depend on
- * a fast prime factoriazation algorithm.
+ * a fast prime factorization algorithm.
  *
  * Notes on the algorithm
  * ----------------------

From python-checkins at python.org  Thu Jul  8 17:08:14 2010
From: python-checkins at python.org (ezio.melotti)
Date: Thu,  8 Jul 2010 17:08:14 +0200 (CEST)
Subject: [Python-checkins] r82645 - python/branches/release31-maint
Message-ID: <20100708150814.DE317EED17@mail.python.org>

Author: ezio.melotti
Date: Thu Jul  8 17:08:14 2010
New Revision: 82645

Log:
Blocked revisions 82644 via svnmerge

........
  r82644 | ezio.melotti | 2010-07-08 18:03:02 +0300 (Thu, 08 Jul 2010) | 1 line
  
  Fix typo in a comment in mathmodule.c.
........


Modified:
   python/branches/release31-maint/   (props changed)

From python-checkins at python.org  Thu Jul  8 19:23:40 2010
From: python-checkins at python.org (mark.dickinson)
Date: Thu,  8 Jul 2010 19:23:40 +0200 (CEST)
Subject: [Python-checkins] r82646 -
	python/branches/py3k/Lib/test/test_decimal.py
Message-ID: <20100708172340.89B91EEDB3@mail.python.org>

Author: mark.dickinson
Date: Thu Jul  8 19:23:40 2010
New Revision: 82646

Log:
In test_decimal, convert heuristic for skipping tests into an explicit skiplist.

Modified:
   python/branches/py3k/Lib/test/test_decimal.py

Modified: python/branches/py3k/Lib/test/test_decimal.py
==============================================================================
--- python/branches/py3k/Lib/test/test_decimal.py	(original)
+++ python/branches/py3k/Lib/test/test_decimal.py	Thu Jul  8 19:23:40 2010
@@ -78,10 +78,41 @@
 
 # list of individual .decTest test ids that correspond to tests that
 # we're skipping for one reason or another.
-skipped_test_ids = [
-    'scbx164',  # skipping apparently implementation-specific scaleb
-    'scbx165',  # tests, pending clarification of scaleb rules.
-]
+skipped_test_ids = set([
+    # Skip implementation-specific scaleb tests.
+    'scbx164',
+    'scbx165',
+
+    # For some operations (currently exp, ln, log10, power), the decNumber
+    # reference implementation imposes additional restrictions on the context
+    # and operands.  These restrictions are not part of the specification;
+    # however, the effect of these restrictions does show up in some of the
+    # testcases.  We skip testcases that violate these restrictions, since
+    # Decimal behaves differently from decNumber for these testcases so these
+    # testcases would otherwise fail.
+    'expx901',
+    'expx902',
+    'expx903',
+    'expx905',
+    'lnx901',
+    'lnx902',
+    'lnx903',
+    'lnx905',
+    'logx901',
+    'logx902',
+    'logx903',
+    'logx905',
+    'powx1183',
+    'powx1184',
+    'powx4001',
+    'powx4002',
+    'powx4003',
+    'powx4005',
+    'powx4008',
+    'powx4010',
+    'powx4012',
+    'powx4014',
+    ])
 
 # Make sure it actually raises errors when not expected and caught in flags
 # Slower, since it runs some things several times.
@@ -172,27 +203,6 @@
     'same_quantum',
     )
 
-# For some operations (currently exp, ln, log10, power), the decNumber
-# reference implementation imposes additional restrictions on the
-# context and operands.  These restrictions are not part of the
-# specification; however, the effect of these restrictions does show
-# up in some of the testcases.  We skip testcases that violate these
-# restrictions, since Decimal behaves differently from decNumber for
-# these testcases so these testcases would otherwise fail.
-
-decNumberRestricted = ('power', 'ln', 'log10', 'exp')
-DEC_MAX_MATH = 999999
-def outside_decNumber_bounds(v, context):
-    if (context.prec > DEC_MAX_MATH or
-        context.Emax > DEC_MAX_MATH or
-        -context.Emin > DEC_MAX_MATH):
-        return True
-    if not v._is_special and v and (
-        v.adjusted() > DEC_MAX_MATH or
-        v.adjusted() < 1-2*DEC_MAX_MATH):
-        return True
-    return False
-
 class DecimalTest(unittest.TestCase):
     """Class which tests the Decimal class against the test cases.
 
@@ -330,22 +340,6 @@
 
         ans = FixQuotes(ans)
 
-        # skip tests that are related to bounds imposed in the decNumber
-        # reference implementation
-        if fname in decNumberRestricted:
-            if fname == 'power':
-                if not (vals[1]._isinteger() and
-                        -1999999997 <= vals[1] <= 999999999):
-                    if outside_decNumber_bounds(vals[0], self.context) or \
-                            outside_decNumber_bounds(vals[1], self.context):
-                        #print "Skipping test %s" % s
-                        return
-            else:
-                if outside_decNumber_bounds(vals[0], self.context):
-                    #print "Skipping test %s" % s
-                    return
-
-
         if EXTENDEDERRORTEST and fname not in ('to_sci_string', 'to_eng_string'):
             for error in theirexceptions:
                 self.context.traps[error] = 1

From python-checkins at python.org  Thu Jul  8 20:51:30 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Thu,  8 Jul 2010 20:51:30 +0200 (CEST)
Subject: [Python-checkins] r82647 - in python/branches/py3k:
	Lib/test/test_gdb.py Misc/NEWS
Message-ID: <20100708185130.670BCF061@mail.python.org>

Author: antoine.pitrou
Date: Thu Jul  8 20:51:30 2010
New Revision: 82647

Log:
Issue #8605: Skip test_gdb if Python is compiled with optimizations.



Modified:
   python/branches/py3k/Lib/test/test_gdb.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/test/test_gdb.py
==============================================================================
--- python/branches/py3k/Lib/test/test_gdb.py	(original)
+++ python/branches/py3k/Lib/test/test_gdb.py	Thu Jul  8 20:51:30 2010
@@ -9,6 +9,7 @@
 import sys
 import unittest
 import locale
+import sysconfig
 
 from test.support import run_unittest, findfile
 
@@ -664,6 +665,15 @@
                                     r".*\na = 1\nb = 2\nc = 3\n.*")
 
 def test_main():
+    cflags = sysconfig.get_config_vars()['PY_CFLAGS']
+    final_opt = ""
+    for opt in cflags.split():
+        if opt.startswith('-O'):
+            final_opt = opt
+    if final_opt and final_opt != '-O0':
+        raise unittest.SkipTest("Python was built with compiler optimizations, "
+                                "tests can't reliably succeed")
+
     run_unittest(PrettyPrintTests,
                  PyListTests,
                  StackNavigationTests,

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Thu Jul  8 20:51:30 2010
@@ -1614,6 +1614,8 @@
 Tests
 -----
 
+- Issue #8605: Skip test_gdb if Python is compiled with optimizations.
+
 - Issue #7449: Skip test_socketserver if threading support is disabled
 
 - Issue #8672: Add a zlib test ensuring that an incomplete stream can be

From python-checkins at python.org  Thu Jul  8 20:54:04 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Thu,  8 Jul 2010 20:54:04 +0200 (CEST)
Subject: [Python-checkins] r82648 - in python/branches/release27-maint:
	Lib/test/test_gdb.py Misc/NEWS
Message-ID: <20100708185404.46DFDF061@mail.python.org>

Author: antoine.pitrou
Date: Thu Jul  8 20:54:04 2010
New Revision: 82648

Log:
Merged revisions 82647 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82647 | antoine.pitrou | 2010-07-08 20:51:30 +0200 (jeu., 08 juil. 2010) | 3 lines
  
  Issue #8605: Skip test_gdb if Python is compiled with optimizations.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/test/test_gdb.py
   python/branches/release27-maint/Misc/NEWS

Modified: python/branches/release27-maint/Lib/test/test_gdb.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_gdb.py	(original)
+++ python/branches/release27-maint/Lib/test/test_gdb.py	Thu Jul  8 20:54:04 2010
@@ -8,6 +8,7 @@
 import subprocess
 import sys
 import unittest
+import sysconfig
 
 from test.test_support import run_unittest, findfile
 
@@ -678,6 +679,15 @@
                                     r".*\na = 1\nb = 2\nc = 3\n.*")
 
 def test_main():
+    cflags = sysconfig.get_config_vars()['PY_CFLAGS']
+    final_opt = ""
+    for opt in cflags.split():
+        if opt.startswith('-O'):
+            final_opt = opt
+    if final_opt and final_opt != '-O0':
+        raise unittest.SkipTest("Python was built with compiler optimizations, "
+                                "tests can't reliably succeed")
+
     run_unittest(PrettyPrintTests,
                  PyListTests,
                  StackNavigationTests,

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Thu Jul  8 20:54:04 2010
@@ -26,6 +26,11 @@
 Extension Modules
 -----------------
 
+Tests
+-----
+
+- Issue #8605: Skip test_gdb if Python is compiled with optimizations.
+
 
 What's New in Python 2.7?
 =========================

From python-checkins at python.org  Thu Jul  8 21:03:34 2010
From: python-checkins at python.org (mark.dickinson)
Date: Thu,  8 Jul 2010 21:03:34 +0200 (CEST)
Subject: [Python-checkins] r82649 - in python/branches/py3k: Lib/decimal.py
	Lib/test/decimaltestdata/extra.decTest Misc/NEWS
Message-ID: <20100708190334.72A1DEE9BA@mail.python.org>

Author: mark.dickinson
Date: Thu Jul  8 21:03:34 2010
New Revision: 82649

Log:
Fix a performance issue in Decimal.pow.  Thanks Stefan Krah for finding this.

Modified:
   python/branches/py3k/Lib/decimal.py
   python/branches/py3k/Lib/test/decimaltestdata/extra.decTest
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/decimal.py
==============================================================================
--- python/branches/py3k/Lib/decimal.py	(original)
+++ python/branches/py3k/Lib/decimal.py	Thu Jul  8 21:03:34 2010
@@ -2047,12 +2047,14 @@
         # case where xc == 1: result is 10**(xe*y), with xe*y
         # required to be an integer
         if xc == 1:
-            if ye >= 0:
-                exponent = xe*yc*10**ye
-            else:
-                exponent, remainder = divmod(xe*yc, 10**-ye)
-                if remainder:
-                    return None
+            xe *= yc
+            # result is now 10**(xe * 10**ye);  xe * 10**ye must be integral
+            while xe % 10 == 0:
+                xe //= 10
+                ye += 1
+            if ye < 0:
+                return None
+            exponent = xe * 10**ye
             if y.sign == 1:
                 exponent = -exponent
             # if other is a nonnegative integer, use ideal exponent

Modified: python/branches/py3k/Lib/test/decimaltestdata/extra.decTest
==============================================================================
--- python/branches/py3k/Lib/test/decimaltestdata/extra.decTest	(original)
+++ python/branches/py3k/Lib/test/decimaltestdata/extra.decTest	Thu Jul  8 21:03:34 2010
@@ -213,7 +213,20 @@
 extr1659 shift 1234567 4 -> 0
 extr1660 shift 1234567 5 -> NaN Invalid_operation
 
+-- Cases where the power function was impossibly slow to determine that the
+-- result is inexact.  Thanks Stefan Krah for identifying this problem.
+precision: 16
+maxExponent: 999999999
+minExponent: -999999999
+extr1700 power 10 1e-999999999 -> 1.000000000000000 Inexact Rounded
+extr1701 power 100.0 -557.71e-742888888 -> 1.000000000000000 Inexact Rounded
+extr1702 power 10 1e-100 -> 1.000000000000000 Inexact Rounded
 
+-- A couple of interesting exact cases for power.  Note that the specification
+-- requires these to be reported as Inexact.
+extr1710 power 1e375 56e-3 -> 1.000000000000000E+21 Inexact Rounded
+extr1711 power 10000 0.75 -> 1000.000000000000 Inexact Rounded
+extr1712 power 1e-24 0.875 -> 1.000000000000000E-21 Inexact Rounded
 
 -- Tests for the is_* boolean operations
 precision: 9

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Thu Jul  8 21:03:34 2010
@@ -470,6 +470,10 @@
 Library
 -------
 
+- Fix extreme speed issue in Decimal.pow when the base is an exact
+  power of 10 and the exponent is tiny (for example,
+  Decimal(10) ** Decimal('1e-999999999')).
+
 - Issue #9186: Fix math.log1p(-1.0) to raise ValueError, not OverflowError.
 
 - Issue #9130: Fix validation of relative imports in parser module.

From python-checkins at python.org  Thu Jul  8 21:09:16 2010
From: python-checkins at python.org (mark.dickinson)
Date: Thu,  8 Jul 2010 21:09:16 +0200 (CEST)
Subject: [Python-checkins] r82650 - python/branches/py3k/Lib/decimal.py
Message-ID: <20100708190916.F05BAEE9A3@mail.python.org>

Author: mark.dickinson
Date: Thu Jul  8 21:09:16 2010
New Revision: 82650

Log:
Fix misplaced exactness check that was causing unnecessary work in Decimal.__pow__.

Modified:
   python/branches/py3k/Lib/decimal.py

Modified: python/branches/py3k/Lib/decimal.py
==============================================================================
--- python/branches/py3k/Lib/decimal.py	(original)
+++ python/branches/py3k/Lib/decimal.py	Thu Jul  8 21:09:16 2010
@@ -2327,9 +2327,10 @@
         # try for an exact result with precision +1
         if ans is None:
             ans = self._power_exact(other, context.prec + 1)
-            if ans is not None and result_sign == 1:
-                ans = _dec_from_triple(1, ans._int, ans._exp)
-            exact = True
+            if ans is not None:
+                if result_sign == 1:
+                    ans = _dec_from_triple(1, ans._int, ans._exp)
+                exact = True
 
         # usual case: inexact result, x**y computed directly as exp(y*log(x))
         if ans is None:

From alexander.belopolsky at gmail.com  Thu Jul  8 21:12:16 2010
From: alexander.belopolsky at gmail.com (Alexander Belopolsky)
Date: Thu, 8 Jul 2010 15:12:16 -0400
Subject: [Python-checkins] r82646 -
	python/branches/py3k/Lib/test/test_decimal.py
In-Reply-To: <20100708172340.89B91EEDB3@mail.python.org>
References: <20100708172340.89B91EEDB3@mail.python.org>
Message-ID: <AANLkTinQcGtcLoBfnAFTUlmJRCtI5g1SeVtRS0GUU62z@mail.gmail.com>

On Thu, Jul 8, 2010 at 1:23 PM, mark.dickinson
<python-checkins at python.org> wrote:
..
> +skipped_test_ids = set([ ..

Any reason for not using set literal here?

From dickinsm at gmail.com  Thu Jul  8 21:15:46 2010
From: dickinsm at gmail.com (Mark Dickinson)
Date: Thu, 8 Jul 2010 20:15:46 +0100
Subject: [Python-checkins] r82646 -
	python/branches/py3k/Lib/test/test_decimal.py
In-Reply-To: <AANLkTinQcGtcLoBfnAFTUlmJRCtI5g1SeVtRS0GUU62z@mail.gmail.com>
References: <20100708172340.89B91EEDB3@mail.python.org>
	<AANLkTinQcGtcLoBfnAFTUlmJRCtI5g1SeVtRS0GUU62z@mail.gmail.com>
Message-ID: <AANLkTiljL-59PLG6kYO_bH32SB1tgsBD1bW8xFcg9S7D@mail.gmail.com>

On Thu, Jul 8, 2010 at 8:12 PM, Alexander Belopolsky
<alexander.belopolsky at gmail.com> wrote:
> On Thu, Jul 8, 2010 at 1:23 PM, mark.dickinson
> <python-checkins at python.org> wrote:
> ..
>> +skipped_test_ids = set([ ..
>
> Any reason for not using set literal here?

Yes;  I'm planning to backport these fixes to 2.7, and it's been
agreed that the 2.x version of decimal.py should remain compatible
with Python 2.5.  It's not 100% clear that that should also applies to
test_decimal.py, but there's no point gratuitously breaking
compatibility with Python 2.5 there either.

I'm also considering backporting to 2.6, where set literals aren't available. :)

Mark

From dickinsm at gmail.com  Thu Jul  8 21:19:26 2010
From: dickinsm at gmail.com (Mark Dickinson)
Date: Thu, 8 Jul 2010 20:19:26 +0100
Subject: [Python-checkins] r82646 -
	python/branches/py3k/Lib/test/test_decimal.py
In-Reply-To: <AANLkTiljL-59PLG6kYO_bH32SB1tgsBD1bW8xFcg9S7D@mail.gmail.com>
References: <20100708172340.89B91EEDB3@mail.python.org>
	<AANLkTinQcGtcLoBfnAFTUlmJRCtI5g1SeVtRS0GUU62z@mail.gmail.com>
	<AANLkTiljL-59PLG6kYO_bH32SB1tgsBD1bW8xFcg9S7D@mail.gmail.com>
Message-ID: <AANLkTikZuTNaXBXhMR8iPXwn_rlUdReoG3I50vnvjWJL@mail.gmail.com>

On Thu, Jul 8, 2010 at 8:15 PM, Mark Dickinson <dickinsm at gmail.com> wrote:
> with Python 2.5. ?It's not 100% clear that that should also applies to
s/applies/apply/

> I'm also considering backporting to 2.6, where set literals aren't available. :)

I just realised that this particular changeset looks more like a
cleanup than a bugfix, so it's probably not clear to others why
backporting is appropriate;  my justification for the backport is that
this changeset is needed to backport r82649, which *is* a bug.  (The
testcases for r82649 won't run without this change, since they get
killed by the heuristic that this changeset removes.)

Mark

From python-checkins at python.org  Thu Jul  8 21:21:59 2010
From: python-checkins at python.org (mark.dickinson)
Date: Thu,  8 Jul 2010 21:21:59 +0200 (CEST)
Subject: [Python-checkins] r82651 - in python/branches/release31-maint:
	Lib/decimal.py Lib/test/decimaltestdata/extra.decTest
	Lib/test/test_decimal.py Misc/NEWS
Message-ID: <20100708192159.ABF04EEB85@mail.python.org>

Author: mark.dickinson
Date: Thu Jul  8 21:21:59 2010
New Revision: 82651

Log:
Merged revisions 82646,82649-82650 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82646 | mark.dickinson | 2010-07-08 18:23:40 +0100 (Thu, 08 Jul 2010) | 1 line
  
  In test_decimal, convert heuristic for skipping tests into an explicit skiplist.
........
  r82649 | mark.dickinson | 2010-07-08 20:03:34 +0100 (Thu, 08 Jul 2010) | 1 line
  
  Fix a performance issue in Decimal.pow.  Thanks Stefan Krah for finding this.
........
  r82650 | mark.dickinson | 2010-07-08 20:09:16 +0100 (Thu, 08 Jul 2010) | 1 line
  
  Fix misplaced exactness check that was causing unnecessary work in Decimal.__pow__.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/decimal.py
   python/branches/release31-maint/Lib/test/decimaltestdata/extra.decTest
   python/branches/release31-maint/Lib/test/test_decimal.py
   python/branches/release31-maint/Misc/NEWS

Modified: python/branches/release31-maint/Lib/decimal.py
==============================================================================
--- python/branches/release31-maint/Lib/decimal.py	(original)
+++ python/branches/release31-maint/Lib/decimal.py	Thu Jul  8 21:21:59 2010
@@ -2044,12 +2044,14 @@
         # case where xc == 1: result is 10**(xe*y), with xe*y
         # required to be an integer
         if xc == 1:
-            if ye >= 0:
-                exponent = xe*yc*10**ye
-            else:
-                exponent, remainder = divmod(xe*yc, 10**-ye)
-                if remainder:
-                    return None
+            xe *= yc
+            # result is now 10**(xe * 10**ye);  xe * 10**ye must be integral
+            while xe % 10 == 0:
+                xe //= 10
+                ye += 1
+            if ye < 0:
+                return None
+            exponent = xe * 10**ye
             if y.sign == 1:
                 exponent = -exponent
             # if other is a nonnegative integer, use ideal exponent
@@ -2322,9 +2324,10 @@
         # try for an exact result with precision +1
         if ans is None:
             ans = self._power_exact(other, context.prec + 1)
-            if ans is not None and result_sign == 1:
-                ans = _dec_from_triple(1, ans._int, ans._exp)
-            exact = True
+            if ans is not None:
+                if result_sign == 1:
+                    ans = _dec_from_triple(1, ans._int, ans._exp)
+                exact = True
 
         # usual case: inexact result, x**y computed directly as exp(y*log(x))
         if ans is None:

Modified: python/branches/release31-maint/Lib/test/decimaltestdata/extra.decTest
==============================================================================
--- python/branches/release31-maint/Lib/test/decimaltestdata/extra.decTest	(original)
+++ python/branches/release31-maint/Lib/test/decimaltestdata/extra.decTest	Thu Jul  8 21:21:59 2010
@@ -213,7 +213,20 @@
 extr1659 shift 1234567 4 -> 0
 extr1660 shift 1234567 5 -> NaN Invalid_operation
 
+-- Cases where the power function was impossibly slow to determine that the
+-- result is inexact.  Thanks Stefan Krah for identifying this problem.
+precision: 16
+maxExponent: 999999999
+minExponent: -999999999
+extr1700 power 10 1e-999999999 -> 1.000000000000000 Inexact Rounded
+extr1701 power 100.0 -557.71e-742888888 -> 1.000000000000000 Inexact Rounded
+extr1702 power 10 1e-100 -> 1.000000000000000 Inexact Rounded
 
+-- A couple of interesting exact cases for power.  Note that the specification
+-- requires these to be reported as Inexact.
+extr1710 power 1e375 56e-3 -> 1.000000000000000E+21 Inexact Rounded
+extr1711 power 10000 0.75 -> 1000.000000000000 Inexact Rounded
+extr1712 power 1e-24 0.875 -> 1.000000000000000E-21 Inexact Rounded
 
 -- Tests for the is_* boolean operations
 precision: 9

Modified: python/branches/release31-maint/Lib/test/test_decimal.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_decimal.py	(original)
+++ python/branches/release31-maint/Lib/test/test_decimal.py	Thu Jul  8 21:21:59 2010
@@ -71,10 +71,41 @@
 
 # list of individual .decTest test ids that correspond to tests that
 # we're skipping for one reason or another.
-skipped_test_ids = [
-    'scbx164',  # skipping apparently implementation-specific scaleb
-    'scbx165',  # tests, pending clarification of scaleb rules.
-]
+skipped_test_ids = set([
+    # Skip implementation-specific scaleb tests.
+    'scbx164',
+    'scbx165',
+
+    # For some operations (currently exp, ln, log10, power), the decNumber
+    # reference implementation imposes additional restrictions on the context
+    # and operands.  These restrictions are not part of the specification;
+    # however, the effect of these restrictions does show up in some of the
+    # testcases.  We skip testcases that violate these restrictions, since
+    # Decimal behaves differently from decNumber for these testcases so these
+    # testcases would otherwise fail.
+    'expx901',
+    'expx902',
+    'expx903',
+    'expx905',
+    'lnx901',
+    'lnx902',
+    'lnx903',
+    'lnx905',
+    'logx901',
+    'logx902',
+    'logx903',
+    'logx905',
+    'powx1183',
+    'powx1184',
+    'powx4001',
+    'powx4002',
+    'powx4003',
+    'powx4005',
+    'powx4008',
+    'powx4010',
+    'powx4012',
+    'powx4014',
+    ])
 
 # Make sure it actually raises errors when not expected and caught in flags
 # Slower, since it runs some things several times.
@@ -165,27 +196,6 @@
     'same_quantum',
     )
 
-# For some operations (currently exp, ln, log10, power), the decNumber
-# reference implementation imposes additional restrictions on the
-# context and operands.  These restrictions are not part of the
-# specification; however, the effect of these restrictions does show
-# up in some of the testcases.  We skip testcases that violate these
-# restrictions, since Decimal behaves differently from decNumber for
-# these testcases so these testcases would otherwise fail.
-
-decNumberRestricted = ('power', 'ln', 'log10', 'exp')
-DEC_MAX_MATH = 999999
-def outside_decNumber_bounds(v, context):
-    if (context.prec > DEC_MAX_MATH or
-        context.Emax > DEC_MAX_MATH or
-        -context.Emin > DEC_MAX_MATH):
-        return True
-    if not v._is_special and v and (
-        v.adjusted() > DEC_MAX_MATH or
-        v.adjusted() < 1-2*DEC_MAX_MATH):
-        return True
-    return False
-
 class DecimalTest(unittest.TestCase):
     """Class which tests the Decimal class against the test cases.
 
@@ -323,22 +333,6 @@
 
         ans = FixQuotes(ans)
 
-        # skip tests that are related to bounds imposed in the decNumber
-        # reference implementation
-        if fname in decNumberRestricted:
-            if fname == 'power':
-                if not (vals[1]._isinteger() and
-                        -1999999997 <= vals[1] <= 999999999):
-                    if outside_decNumber_bounds(vals[0], self.context) or \
-                            outside_decNumber_bounds(vals[1], self.context):
-                        #print "Skipping test %s" % s
-                        return
-            else:
-                if outside_decNumber_bounds(vals[0], self.context):
-                    #print "Skipping test %s" % s
-                    return
-
-
         if EXTENDEDERRORTEST and fname not in ('to_sci_string', 'to_eng_string'):
             for error in theirexceptions:
                 self.context.traps[error] = 1

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Thu Jul  8 21:21:59 2010
@@ -75,6 +75,10 @@
 Library
 -------
 
+- Fix extreme speed issue in Decimal.pow when the base is an exact
+  power of 10 and the exponent is tiny (for example,
+  Decimal(10) ** Decimal('1e-999999999')).
+
 - Issue #9130: Fix validation of relative imports in parser module.
 
 - Issue #9128: Fix validation of class decorators in parser module.

From python-checkins at python.org  Thu Jul  8 21:24:40 2010
From: python-checkins at python.org (mark.dickinson)
Date: Thu,  8 Jul 2010 21:24:40 +0200 (CEST)
Subject: [Python-checkins] r82652 - in python/branches/release27-maint:
	Lib/decimal.py Lib/test/decimaltestdata/extra.decTest
	Lib/test/test_decimal.py Misc/NEWS
Message-ID: <20100708192440.DF5F3EE982@mail.python.org>

Author: mark.dickinson
Date: Thu Jul  8 21:24:40 2010
New Revision: 82652

Log:
Merged revisions 82646,82649-82650 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82646 | mark.dickinson | 2010-07-08 18:23:40 +0100 (Thu, 08 Jul 2010) | 1 line
  
  In test_decimal, convert heuristic for skipping tests into an explicit skiplist.
........
  r82649 | mark.dickinson | 2010-07-08 20:03:34 +0100 (Thu, 08 Jul 2010) | 1 line
  
  Fix a performance issue in Decimal.pow.  Thanks Stefan Krah for finding this.
........
  r82650 | mark.dickinson | 2010-07-08 20:09:16 +0100 (Thu, 08 Jul 2010) | 1 line
  
  Fix misplaced exactness check that was causing unnecessary work in Decimal.__pow__.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/decimal.py
   python/branches/release27-maint/Lib/test/decimaltestdata/extra.decTest
   python/branches/release27-maint/Lib/test/test_decimal.py
   python/branches/release27-maint/Misc/NEWS

Modified: python/branches/release27-maint/Lib/decimal.py
==============================================================================
--- python/branches/release27-maint/Lib/decimal.py	(original)
+++ python/branches/release27-maint/Lib/decimal.py	Thu Jul  8 21:24:40 2010
@@ -1990,12 +1990,14 @@
         # case where xc == 1: result is 10**(xe*y), with xe*y
         # required to be an integer
         if xc == 1:
-            if ye >= 0:
-                exponent = xe*yc*10**ye
-            else:
-                exponent, remainder = divmod(xe*yc, 10**-ye)
-                if remainder:
-                    return None
+            xe *= yc
+            # result is now 10**(xe * 10**ye);  xe * 10**ye must be integral
+            while xe % 10 == 0:
+                xe //= 10
+                ye += 1
+            if ye < 0:
+                return None
+            exponent = xe * 10**ye
             if y.sign == 1:
                 exponent = -exponent
             # if other is a nonnegative integer, use ideal exponent
@@ -2268,9 +2270,10 @@
         # try for an exact result with precision +1
         if ans is None:
             ans = self._power_exact(other, context.prec + 1)
-            if ans is not None and result_sign == 1:
-                ans = _dec_from_triple(1, ans._int, ans._exp)
-            exact = True
+            if ans is not None:
+                if result_sign == 1:
+                    ans = _dec_from_triple(1, ans._int, ans._exp)
+                exact = True
 
         # usual case: inexact result, x**y computed directly as exp(y*log(x))
         if ans is None:

Modified: python/branches/release27-maint/Lib/test/decimaltestdata/extra.decTest
==============================================================================
--- python/branches/release27-maint/Lib/test/decimaltestdata/extra.decTest	(original)
+++ python/branches/release27-maint/Lib/test/decimaltestdata/extra.decTest	Thu Jul  8 21:24:40 2010
@@ -213,7 +213,20 @@
 extr1659 shift 1234567 4 -> 0
 extr1660 shift 1234567 5 -> NaN Invalid_operation
 
+-- Cases where the power function was impossibly slow to determine that the
+-- result is inexact.  Thanks Stefan Krah for identifying this problem.
+precision: 16
+maxExponent: 999999999
+minExponent: -999999999
+extr1700 power 10 1e-999999999 -> 1.000000000000000 Inexact Rounded
+extr1701 power 100.0 -557.71e-742888888 -> 1.000000000000000 Inexact Rounded
+extr1702 power 10 1e-100 -> 1.000000000000000 Inexact Rounded
 
+-- A couple of interesting exact cases for power.  Note that the specification
+-- requires these to be reported as Inexact.
+extr1710 power 1e375 56e-3 -> 1.000000000000000E+21 Inexact Rounded
+extr1711 power 10000 0.75 -> 1000.000000000000 Inexact Rounded
+extr1712 power 1e-24 0.875 -> 1.000000000000000E-21 Inexact Rounded
 
 -- Tests for the is_* boolean operations
 precision: 9

Modified: python/branches/release27-maint/Lib/test/test_decimal.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_decimal.py	(original)
+++ python/branches/release27-maint/Lib/test/test_decimal.py	Thu Jul  8 21:24:40 2010
@@ -77,10 +77,41 @@
 
 # list of individual .decTest test ids that correspond to tests that
 # we're skipping for one reason or another.
-skipped_test_ids = [
-    'scbx164',  # skipping apparently implementation-specific scaleb
-    'scbx165',  # tests, pending clarification of scaleb rules.
-]
+skipped_test_ids = set([
+    # Skip implementation-specific scaleb tests.
+    'scbx164',
+    'scbx165',
+
+    # For some operations (currently exp, ln, log10, power), the decNumber
+    # reference implementation imposes additional restrictions on the context
+    # and operands.  These restrictions are not part of the specification;
+    # however, the effect of these restrictions does show up in some of the
+    # testcases.  We skip testcases that violate these restrictions, since
+    # Decimal behaves differently from decNumber for these testcases so these
+    # testcases would otherwise fail.
+    'expx901',
+    'expx902',
+    'expx903',
+    'expx905',
+    'lnx901',
+    'lnx902',
+    'lnx903',
+    'lnx905',
+    'logx901',
+    'logx902',
+    'logx903',
+    'logx905',
+    'powx1183',
+    'powx1184',
+    'powx4001',
+    'powx4002',
+    'powx4003',
+    'powx4005',
+    'powx4008',
+    'powx4010',
+    'powx4012',
+    'powx4014',
+    ])
 
 # Make sure it actually raises errors when not expected and caught in flags
 # Slower, since it runs some things several times.
@@ -171,27 +202,6 @@
     'same_quantum',
     )
 
-# For some operations (currently exp, ln, log10, power), the decNumber
-# reference implementation imposes additional restrictions on the
-# context and operands.  These restrictions are not part of the
-# specification; however, the effect of these restrictions does show
-# up in some of the testcases.  We skip testcases that violate these
-# restrictions, since Decimal behaves differently from decNumber for
-# these testcases so these testcases would otherwise fail.
-
-decNumberRestricted = ('power', 'ln', 'log10', 'exp')
-DEC_MAX_MATH = 999999
-def outside_decNumber_bounds(v, context):
-    if (context.prec > DEC_MAX_MATH or
-        context.Emax > DEC_MAX_MATH or
-        -context.Emin > DEC_MAX_MATH):
-        return True
-    if not v._is_special and v and (
-        v.adjusted() > DEC_MAX_MATH or
-        v.adjusted() < 1-2*DEC_MAX_MATH):
-        return True
-    return False
-
 class DecimalTest(unittest.TestCase):
     """Class which tests the Decimal class against the test cases.
 
@@ -329,22 +339,6 @@
 
         ans = FixQuotes(ans)
 
-        # skip tests that are related to bounds imposed in the decNumber
-        # reference implementation
-        if fname in decNumberRestricted:
-            if fname == 'power':
-                if not (vals[1]._isinteger() and
-                        -1999999997 <= vals[1] <= 999999999):
-                    if outside_decNumber_bounds(vals[0], self.context) or \
-                            outside_decNumber_bounds(vals[1], self.context):
-                        #print "Skipping test %s" % s
-                        return
-            else:
-                if outside_decNumber_bounds(vals[0], self.context):
-                    #print "Skipping test %s" % s
-                    return
-
-
         if EXTENDEDERRORTEST and fname not in ('to_sci_string', 'to_eng_string'):
             for error in theirexceptions:
                 self.context.traps[error] = 1

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Thu Jul  8 21:24:40 2010
@@ -515,6 +515,10 @@
 - Issue #8329: Don't return the same lists from select.select when no fds are
   changed.
 
+- Fix extreme speed issue in Decimal.pow when the base is an exact
+  power of 10 and the exponent is tiny (for example,
+  Decimal(10) ** Decimal('1e-999999999')).
+
 - Issue #8259: ``1L << (2**31)`` no longer produces an 'outrageous shift error'
   on 64-bit machines.  The shift count for either left or right shift is
   permitted to be up to sys.maxsize.

From python-checkins at python.org  Thu Jul  8 21:27:24 2010
From: python-checkins at python.org (mark.dickinson)
Date: Thu,  8 Jul 2010 21:27:24 +0200 (CEST)
Subject: [Python-checkins] r82653 - in python/branches/release26-maint:
	Lib/decimal.py Lib/test/decimaltestdata/extra.decTest
	Lib/test/test_decimal.py Misc/NEWS
Message-ID: <20100708192724.3E568EEAF1@mail.python.org>

Author: mark.dickinson
Date: Thu Jul  8 21:27:24 2010
New Revision: 82653

Log:
Fix Decimal speed issue;  backport of r82652 from release27-maint.

Modified:
   python/branches/release26-maint/Lib/decimal.py
   python/branches/release26-maint/Lib/test/decimaltestdata/extra.decTest
   python/branches/release26-maint/Lib/test/test_decimal.py
   python/branches/release26-maint/Misc/NEWS

Modified: python/branches/release26-maint/Lib/decimal.py
==============================================================================
--- python/branches/release26-maint/Lib/decimal.py	(original)
+++ python/branches/release26-maint/Lib/decimal.py	Thu Jul  8 21:27:24 2010
@@ -1918,12 +1918,14 @@
         # case where xc == 1: result is 10**(xe*y), with xe*y
         # required to be an integer
         if xc == 1:
-            if ye >= 0:
-                exponent = xe*yc*10**ye
-            else:
-                exponent, remainder = divmod(xe*yc, 10**-ye)
-                if remainder:
-                    return None
+            xe *= yc
+            # result is now 10**(xe * 10**ye);  xe * 10**ye must be integral
+            while xe % 10 == 0:
+                xe //= 10
+                ye += 1
+            if ye < 0:
+                return None
+            exponent = xe * 10**ye
             if y.sign == 1:
                 exponent = -exponent
             # if other is a nonnegative integer, use ideal exponent
@@ -2196,9 +2198,10 @@
         # try for an exact result with precision +1
         if ans is None:
             ans = self._power_exact(other, context.prec + 1)
-            if ans is not None and result_sign == 1:
-                ans = _dec_from_triple(1, ans._int, ans._exp)
-            exact = True
+            if ans is not None:
+                if result_sign == 1:
+                    ans = _dec_from_triple(1, ans._int, ans._exp)
+                exact = True
 
         # usual case: inexact result, x**y computed directly as exp(y*log(x))
         if ans is None:

Modified: python/branches/release26-maint/Lib/test/decimaltestdata/extra.decTest
==============================================================================
--- python/branches/release26-maint/Lib/test/decimaltestdata/extra.decTest	(original)
+++ python/branches/release26-maint/Lib/test/decimaltestdata/extra.decTest	Thu Jul  8 21:27:24 2010
@@ -213,7 +213,20 @@
 extr1659 shift 1234567 4 -> 0
 extr1660 shift 1234567 5 -> NaN Invalid_operation
 
+-- Cases where the power function was impossibly slow to determine that the
+-- result is inexact.  Thanks Stefan Krah for identifying this problem.
+precision: 16
+maxExponent: 999999999
+minExponent: -999999999
+extr1700 power 10 1e-999999999 -> 1.000000000000000 Inexact Rounded
+extr1701 power 100.0 -557.71e-742888888 -> 1.000000000000000 Inexact Rounded
+extr1702 power 10 1e-100 -> 1.000000000000000 Inexact Rounded
 
+-- A couple of interesting exact cases for power.  Note that the specification
+-- requires these to be reported as Inexact.
+extr1710 power 1e375 56e-3 -> 1.000000000000000E+21 Inexact Rounded
+extr1711 power 10000 0.75 -> 1000.000000000000 Inexact Rounded
+extr1712 power 1e-24 0.875 -> 1.000000000000000E-21 Inexact Rounded
 
 -- Tests for the is_* boolean operations
 precision: 9

Modified: python/branches/release26-maint/Lib/test/test_decimal.py
==============================================================================
--- python/branches/release26-maint/Lib/test/test_decimal.py	(original)
+++ python/branches/release26-maint/Lib/test/test_decimal.py	Thu Jul  8 21:27:24 2010
@@ -72,10 +72,41 @@
 
 # list of individual .decTest test ids that correspond to tests that
 # we're skipping for one reason or another.
-skipped_test_ids = [
-    'scbx164',  # skipping apparently implementation-specific scaleb
-    'scbx165',  # tests, pending clarification of scaleb rules.
-]
+skipped_test_ids = set([
+    # Skip implementation-specific scaleb tests.
+    'scbx164',
+    'scbx165',
+
+    # For some operations (currently exp, ln, log10, power), the decNumber
+    # reference implementation imposes additional restrictions on the context
+    # and operands.  These restrictions are not part of the specification;
+    # however, the effect of these restrictions does show up in some of the
+    # testcases.  We skip testcases that violate these restrictions, since
+    # Decimal behaves differently from decNumber for these testcases so these
+    # testcases would otherwise fail.
+    'expx901',
+    'expx902',
+    'expx903',
+    'expx905',
+    'lnx901',
+    'lnx902',
+    'lnx903',
+    'lnx905',
+    'logx901',
+    'logx902',
+    'logx903',
+    'logx905',
+    'powx1183',
+    'powx1184',
+    'powx4001',
+    'powx4002',
+    'powx4003',
+    'powx4005',
+    'powx4008',
+    'powx4010',
+    'powx4012',
+    'powx4014',
+    ])
 
 # Make sure it actually raises errors when not expected and caught in flags
 # Slower, since it runs some things several times.
@@ -166,27 +197,6 @@
     'same_quantum',
     )
 
-# For some operations (currently exp, ln, log10, power), the decNumber
-# reference implementation imposes additional restrictions on the
-# context and operands.  These restrictions are not part of the
-# specification; however, the effect of these restrictions does show
-# up in some of the testcases.  We skip testcases that violate these
-# restrictions, since Decimal behaves differently from decNumber for
-# these testcases so these testcases would otherwise fail.
-
-decNumberRestricted = ('power', 'ln', 'log10', 'exp')
-DEC_MAX_MATH = 999999
-def outside_decNumber_bounds(v, context):
-    if (context.prec > DEC_MAX_MATH or
-        context.Emax > DEC_MAX_MATH or
-        -context.Emin > DEC_MAX_MATH):
-        return True
-    if not v._is_special and v and (
-        v.adjusted() > DEC_MAX_MATH or
-        v.adjusted() < 1-2*DEC_MAX_MATH):
-        return True
-    return False
-
 class DecimalTest(unittest.TestCase):
     """Class which tests the Decimal class against the test cases.
 
@@ -324,22 +334,6 @@
 
         ans = FixQuotes(ans)
 
-        # skip tests that are related to bounds imposed in the decNumber
-        # reference implementation
-        if fname in decNumberRestricted:
-            if fname == 'power':
-                if not (vals[1]._isinteger() and
-                        -1999999997 <= vals[1] <= 999999999):
-                    if outside_decNumber_bounds(vals[0], self.context) or \
-                            outside_decNumber_bounds(vals[1], self.context):
-                        #print "Skipping test %s" % s
-                        return
-            else:
-                if outside_decNumber_bounds(vals[0], self.context):
-                    #print "Skipping test %s" % s
-                    return
-
-
         if EXTENDEDERRORTEST and fname not in ('to_sci_string', 'to_eng_string'):
             for error in theirexceptions:
                 self.context.traps[error] = 1

Modified: python/branches/release26-maint/Misc/NEWS
==============================================================================
--- python/branches/release26-maint/Misc/NEWS	(original)
+++ python/branches/release26-maint/Misc/NEWS	Thu Jul  8 21:27:24 2010
@@ -81,6 +81,10 @@
 Library
 -------
 
+- Fix extreme speed issue in Decimal.pow when the base is an exact
+  power of 10 and the exponent is tiny (for example,
+  Decimal(10) ** Decimal('1e-999999999')).
+
 - Issue #9130: Fix validation of relative imports in parser module.
 
 - Issue #9128: Fix validation of class decorators in parser module.

From nnorwitz at gmail.com  Thu Jul  8 22:58:57 2010
From: nnorwitz at gmail.com (Neal Norwitz)
Date: Thu, 8 Jul 2010 16:58:57 -0400
Subject: [Python-checkins] Python Regression Test Failures opt (1)
Message-ID: <20100708205857.GA3289@kbk-i386-bb.psfb.org>

346 tests OK.
1 test failed:
    test_subprocess
37 tests skipped:
    test_aepack test_al test_applesingle test_bsddb185 test_bsddb3
    test_cd test_cl test_curses test_epoll test_gdb test_gl
    test_imgfile test_ioctl test_kqueue test_linuxaudiodev test_macos
    test_macostools test_multiprocessing test_ossaudiodev test_pep277
    test_py3kwarn test_scriptpackages test_smtpnet test_socketserver
    test_startfile test_sunaudiodev test_tcl test_timeout test_tk
    test_ttk_guionly test_ttk_textonly test_unicode_file
    test_urllib2net test_urllibnet test_winreg test_winsound
    test_zipfile64
7 skips unexpected on linux2:
    test_epoll test_gdb test_ioctl test_multiprocessing test_tk
    test_ttk_guionly test_ttk_textonly

== CPython 2.7 (trunk:82653M, Jul 8 2010, 16:04:57) [GCC 3.3.4 20040623 (Gentoo Linux 3.3.4-r1, ssp-3.3.2-2, pie-8.7.6)]
==   Linux-2.6.9-gentoo-r1-i686-AMD_Athlon-tm-_XP_3000+-with-gentoo-1.4.16 little-endian
==   /tmp/test_python_21001
test_grammar
test_opcodes
test_dict
test_builtin
test_exceptions
test_types
test_unittest
test_doctest
test_doctest2
test_MimeWriter
test_SimpleHTTPServer
test_StringIO
test___all__
test___future__
test__locale
test_abc
test_abstract_numbers
test_aepack
test_aepack skipped -- No module named aetypes
test_aifc
test_al
test_al skipped -- No module named al
test_anydbm
test_applesingle
test_applesingle skipped -- No module named MacOS
test_argparse
test_array
test_ascii_formatd
test_ast
test_asynchat
test_asyncore
test_atexit
test_audioop
test_augassign
test_base64
test_bastion
test_bigaddrspace
test_bigmem
test_binascii
test_binhex
test_binop
test_bisect
test_bool
test_bsddb
test_bsddb185
test_bsddb185 skipped -- No module named bsddb185
test_bsddb3
test_bsddb3 skipped -- Use of the `bsddb' resource not enabled
test_buffer
test_bufio
test_bytes
test_bz2
test_calendar
test_call
test_capi
test_cd
test_cd skipped -- No module named cd
test_cfgparser
test_cgi
test_charmapcodec
test_cl
test_cl skipped -- No module named cl
test_class
test_cmath
test_cmd
test_cmd_line
test_cmd_line_script
test_code
test_codeccallbacks
test_codecencodings_cn
test_codecencodings_hk
test_codecencodings_jp
test_codecencodings_kr
test_codecencodings_tw
test_codecmaps_cn
test_codecmaps_hk
test_codecmaps_jp
test_codecmaps_kr
test_codecmaps_tw
test_codecs
test_codeop
test_coding
test_coercion
test_collections
test_colorsys
test_commands
test_compare
test_compile
test_compileall
test_compiler
test_complex
test_complex_args
test_contains
test_contextlib
test_cookie
test_cookielib
test_copy
test_copy_reg
test_cpickle
test_cprofile
test_crypt
test_csv
test_ctypes
test_curses
test_curses skipped -- Use of the `curses' resource not enabled
test_datetime
test_dbm
test_decimal
test_decorators
test_defaultdict
test_deque
test_descr
test_descrtut
test_dictcomps
test_dictviews
test_difflib
test_dircache
test_dis
test_distutils
[19646 refs]
[19646 refs]
[19646 refs]
[19646 refs]
[19643 refs]
test_dl
test_docxmlrpc
test_dumbdbm
test_dummy_thread
test_dummy_threading
test_email
test_email_codecs
test_email_renamed
test_enumerate
test_eof
test_epoll
test_epoll skipped -- kernel doesn't support epoll()
test_errno
test_exception_variations
test_extcall
test_fcntl
test_file
test_file2k
test_filecmp
test_fileinput
test_fileio
test_float
test_fnmatch
test_fork1
test_format
test_fpformat
test_fractions
test_frozen
test_ftplib
test_funcattrs
test_functools
test_future
test_future3
test_future4
test_future5
test_future_builtins
test_gc
test_gdb
test_gdb skipped -- gdb versions before 7.0 didn't support python embedding Saw:
GNU gdb 6.2.1
Copyright 2004 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB.  Type "show warranty" for details.
This GDB was configured as "i386-pc-linux-gnu".

test_gdbm
test_generators
test_genericpath
test_genexps
test_getargs
test_getargs2
test_getopt
test_gettext
test_gl
test_gl skipped -- No module named gl
test_glob
test_global
test_grp
test_gzip
test_hash
test_hashlib
test_heapq
test_hmac
test_hotshot
test_htmllib
test_htmlparser
test_httplib
test_httpservers
[15648 refs]
[15648 refs]
[15648 refs]
[25315 refs]
test_imageop
test_imaplib
test_imgfile
test_imgfile skipped -- No module named imgfile
test_imp
test_import
test_importhooks
test_importlib
test_index
test_inspect
test_int
test_int_literal
test_io
test_ioctl
test_ioctl skipped -- Unable to open /dev/tty
test_isinstance
test_iter
test_iterlen
test_itertools
test_json
test_kqueue
test_kqueue skipped -- test works only on BSD
test_largefile
test_lib2to3
test_linecache
test_linuxaudiodev
test_linuxaudiodev skipped -- Use of the `audio' resource not enabled
test_list
test_locale
test_logging
test_long
test_long_future
test_longexp
test_macos
test_macos skipped -- No module named MacOS
test_macostools
test_macostools skipped -- No module named MacOS
test_macpath
test_mailbox
test_marshal
test_math
test_md5
test_memoryio
test_memoryview
test_mhlib
test_mimetools
test_mimetypes
test_minidom
test_mmap
test_module
test_modulefinder
test_multibytecodec
test_multibytecodec_support
test_multifile
test_multiprocessing
test_multiprocessing skipped -- This platform lacks a functioning sem_open implementation, therefore, the required synchronization primitives needed will not function, see issue 3770.
test_mutants
test_mutex
test_netrc
test_new
test_nis
test_normalization
test_ntpath
test_old_mailbox
test_openpty
test_operator
test_optparse
test_os
[15648 refs]
[15648 refs]
test_ossaudiodev
test_ossaudiodev skipped -- Use of the `audio' resource not enabled
test_parser
Expecting 's_push: parser stack overflow' in next line
s_push: parser stack overflow
test_pdb
test_peepholer
test_pep247
test_pep263
test_pep277
test_pep277 skipped -- only NT+ and systems with Unicode-friendly filesystem encoding
test_pep292
test_pep352
test_pickle
test_pickletools
test_pipes
test_pkg
test_pkgimport
test_pkgutil
test_platform
[17044 refs]
[17044 refs]
test_plistlib
test_poll
test_popen
[15653 refs]
[15653 refs]
[15653 refs]
test_popen2
test_poplib
test_posix
test_posixpath
test_pow
test_pprint
test_print
test_profile
test_profilehooks
test_property
test_pstats
test_pty
test_pwd
test_py3kwarn
test_py3kwarn skipped -- test.test_py3kwarn must be run with the -3 flag
test_pyclbr
test_pydoc
[20862 refs]
[20862 refs]
[20862 refs]
[20862 refs]
[20862 refs]
[20861 refs]
[20861 refs]
test_pyexpat
test_queue
test_quopri
[18479 refs]
[18479 refs]
test_random
test_re
test_readline
test_repr
test_resource
test_rfc822
test_richcmp
test_robotparser
test_runpy
test_sax
test_scope
test_scriptpackages
test_scriptpackages skipped -- No module named aetools
test_select
test_set
test_setcomps
test_sets
test_sgmllib
test_sha
test_shelve
test_shlex
test_shutil
test_signal
test_site
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
test_slice
test_smtplib
test_smtpnet
test_smtpnet skipped -- Use of the `network' resource not enabled
test_socket
test_socketserver
test_socketserver skipped -- Use of the `network' resource not enabled
test_softspace
test_sort
test_sqlite
test_ssl
test_startfile
test_startfile skipped -- module os has no attribute startfile
test_str
test_strftime
test_string
test_stringprep
test_strop
test_strptime
test_strtod
test_struct
test_structmembers
test_structseq
test_subprocess
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15863 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
.
[15648 refs]
[15648 refs]
    this bit of output is from a test of stdout in a different process ...
[15648 refs]
[15648 refs]
[15863 refs]
[15648 refs]
[15648 refs]
[15648 refs]
p.send_signal(2,) succeeded after 2 attempts
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15863 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
.
[15648 refs]
[15648 refs]
    this bit of output is from a test of stdout in a different process ...
[15648 refs]
[15648 refs]
[15863 refs]
test test_subprocess failed -- multiple errors occurred; run in verbose mode for details
Re-running test 'test_subprocess' in verbose mode
test_call_kwargs (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_call_seq (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_check_call_nonzero (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_check_call_zero (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_check_output (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_check_output_nonzero (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_check_output_stderr (test.test_subprocess.ProcessTestCase) ... ERROR
test_check_output_stdout_arg (test.test_subprocess.ProcessTestCase) ... ERROR
test_communicate (test.test_subprocess.ProcessTestCase) ... ERROR
test_communicate_pipe_buf (test.test_subprocess.ProcessTestCase) ... ERROR
test_communicate_pipe_fd_leak (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_communicate_returns (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_communicate_stderr (test.test_subprocess.ProcessTestCase) ... ERROR
test_communicate_stdin (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_communicate_stdout (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_cwd (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_env (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_executable_with_cwd (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_executable_without_cwd (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_invalid_bufsize (test.test_subprocess.ProcessTestCase) ... ERROR
test_leaking_fds_on_error (test.test_subprocess.ProcessTestCase) ... ERROR
test_list2cmdline (test.test_subprocess.ProcessTestCase) ... ERROR
test_no_leaking (test.test_subprocess.ProcessTestCase) ... ERROR
test_poll (test.test_subprocess.ProcessTestCase) ... [15863 refs]
ERROR
test_stderr_filedes (test.test_subprocess.ProcessTestCase) ... ERROR
test_stderr_fileobj (test.test_subprocess.ProcessTestCase) ... ERROR
test_stderr_none (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_stderr_pipe (test.test_subprocess.ProcessTestCase) ... ERROR
test_stdin_filedes (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_stdin_fileobj (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_stdin_none (test.test_subprocess.ProcessTestCase) ... ERROR
test_stdin_pipe (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_stdout_filedes (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_stdout_filedes_of_stdout (test.test_subprocess.ProcessTestCase) ... .
[15648 refs]
ERROR
test_stdout_fileobj (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_stdout_none (test.test_subprocess.ProcessTestCase) ...     this bit of output is from a test of stdout in a different process ...
ERROR
test_stdout_pipe (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_stdout_stderr_file (test.test_subprocess.ProcessTestCase) ... ERROR
test_stdout_stderr_pipe (test.test_subprocess.ProcessTestCase) ... ERROR
test_universal_newlines (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_universal_newlines_communicate (test.test_subprocess.ProcessTestCase) ... ERROR
test_wait (test.test_subprocess.ProcessTestCase) ... [15863 refs]
ERROR
test_writes_before_communicate (test.test_subprocess.ProcessTestCase) ... ERROR
test_args_string (test.test_subprocess.POSIXProcessTestCase) ... [15648 refs]
ERROR
test_call_string (test.test_subprocess.POSIXProcessTestCase) ... [15648 refs]
ERROR
test_exceptions (test.test_subprocess.POSIXProcessTestCase) ... ERROR
test_invalid_args (test.test_subprocess.POSIXProcessTestCase) ... ERROR
test_kill (test.test_subprocess.POSIXProcessTestCase) ... ERROR
test_preexec (test.test_subprocess.POSIXProcessTestCase) ... [15648 refs]
ERROR
test_run_abort (test.test_subprocess.POSIXProcessTestCase) ... ERROR
test_send_signal (test.test_subprocess.POSIXProcessTestCase) ... ERROR
test_shell_sequence (test.test_subprocess.POSIXProcessTestCase) ... ERROR
test_shell_string (test.test_subprocess.POSIXProcessTestCase) ... ERROR
test_terminate (test.test_subprocess.POSIXProcessTestCase) ... ERROR
test_call_string (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_close_fds (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_creationflags (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_invalid_args (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_kill (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_send_signal (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_shell_sequence (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_shell_string (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_startupinfo (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_terminate (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_call_kwargs (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_call_seq (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_check_call_nonzero (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_check_call_zero (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_check_output (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_check_output_nonzero (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_check_output_stderr (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_check_output_stdout_arg (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_communicate (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_communicate_pipe_buf (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_communicate_pipe_fd_leak (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_communicate_returns (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_communicate_stderr (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_communicate_stdin (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_communicate_stdout (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_cwd (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_env (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_executable_with_cwd (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_executable_without_cwd (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_invalid_bufsize (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_leaking_fds_on_error (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_list2cmdline (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_no_leaking (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_poll (test.test_subprocess.ProcessTestCaseNoPoll) ... [15863 refs]
ERROR
test_stderr_filedes (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_stderr_fileobj (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_stderr_none (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_stderr_pipe (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_stdin_filedes (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_stdin_fileobj (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_stdin_none (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_stdin_pipe (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_stdout_filedes (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_stdout_filedes_of_stdout (test.test_subprocess.ProcessTestCaseNoPoll) ... .
[15648 refs]
ERROR
test_stdout_fileobj (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_stdout_none (test.test_subprocess.ProcessTestCaseNoPoll) ...     this bit of output is from a test of stdout in a different process ...
ERROR
test_stdout_pipe (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_stdout_stderr_file (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_stdout_stderr_pipe (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_universal_newlines (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_universal_newlines_communicate (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_wait (test.test_subprocess.ProcessTestCaseNoPoll) ... [15863 refs]
ERROR
test_writes_before_communicate (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_eintr_retry_call (test.test_subprocess.HelperFunctionTests) ... ok

======================================================================
ERROR: test_call_kwargs (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_call_seq (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_call_nonzero (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_call_zero (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_output (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_output_nonzero (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_output_stderr (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_output_stdout_arg (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_pipe_buf (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_pipe_fd_leak (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_returns (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_stderr (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_stdin (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_stdout (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_cwd (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_env (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_executable_with_cwd (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_executable_without_cwd (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_invalid_bufsize (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_leaking_fds_on_error (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_list2cmdline (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_no_leaking (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_poll (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stderr_filedes (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stderr_fileobj (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stderr_none (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stderr_pipe (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdin_filedes (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdin_fileobj (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdin_none (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdin_pipe (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_filedes (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_filedes_of_stdout (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_fileobj (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_none (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_pipe (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_stderr_file (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_stderr_pipe (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_universal_newlines (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_universal_newlines_communicate (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_wait (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_writes_before_communicate (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_args_string (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_call_string (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_exceptions (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_invalid_args (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_kill (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_preexec (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_run_abort (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_send_signal (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_shell_sequence (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_shell_string (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_terminate (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_call_kwargs (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_call_seq (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_call_nonzero (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_call_zero (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_output (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_output_nonzero (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_output_stderr (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_output_stdout_arg (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_pipe_buf (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_pipe_fd_leak (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_returns (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_stderr (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_stdin (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_stdout (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_cwd (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_env (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_executable_with_cwd (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_executable_without_cwd (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_invalid_bufsize (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_leaking_fds_on_error (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_list2cmdline (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_no_leaking (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_poll (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stderr_filedes (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stderr_fileobj (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stderr_none (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stderr_pipe (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdin_filedes (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdin_fileobj (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdin_none (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdin_pipe (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_filedes (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_filedes_of_stdout (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_fileobj (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_none (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_pipe (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_stderr_file (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_stderr_pipe (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_universal_newlines (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_universal_newlines_communicate (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_wait (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_writes_before_communicate (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

----------------------------------------------------------------------
Ran 108 tests in 200.898s

FAILED (errors=97, skipped=10)
test test_subprocess failed -- multiple errors occurred
test_sunaudiodev
test_sunaudiodev skipped -- No module named sunaudiodev
test_sundry
test_symtable
test_syntax
test_sys
[15648 refs]
[15648 refs]
[15648 refs]
[15877 refs]
[15671 refs]
test_sysconfig
[15648 refs]
[15648 refs]
test_tarfile
test_tcl
test_tcl skipped -- No module named _tkinter
test_telnetlib
test_tempfile
[15648 refs]
test_textwrap
test_thread
test_threaded_import
test_threadedtempfile
test_threading
[18948 refs]
[20223 refs]
[20035 refs]
[20035 refs]
[20035 refs]
[20035 refs]
test_threading_local
test_threadsignals
test_time
test_timeout
test_timeout skipped -- Use of the `network' resource not enabled
test_tk
test_tk skipped -- No module named _tkinter
test_tokenize
test_trace
test_traceback
test_transformer
test_ttk_guionly
test_ttk_guionly skipped -- No module named _tkinter
test_ttk_textonly
test_ttk_textonly skipped -- No module named _tkinter
test_tuple
test_typechecks
test_ucn
test_unary
test_undocumented_details
test_unicode
test_unicode_file
test_unicode_file skipped -- No Unicode filesystem semantics on this platform.
test_unicodedata
test_univnewlines
test_univnewlines2k
test_unpack
test_urllib
test_urllib2
test_urllib2_localnet
test_urllib2net
test_urllib2net skipped -- Use of the `network' resource not enabled
test_urllibnet
test_urllibnet skipped -- Use of the `network' resource not enabled
test_urlparse
test_userdict
test_userlist
test_userstring
test_uu
test_uuid
test_wait3
test_wait4
test_warnings
[15679 refs]
[15679 refs]
[15672 refs]
[15679 refs]
[15679 refs]
[15672 refs]
test_wave
test_weakref
test_weakset
test_whichdb
test_winreg
test_winreg skipped -- No module named _winreg
test_winsound
test_winsound skipped -- No module named winsound
test_with
test_wsgiref
test_xdrlib
test_xml_etree
test_xml_etree_c
test_xmllib
test_xmlrpc
test_xpickle
test_xpickle -- skipping backwards compat tests.
Use 'regrtest.py -u xpickle' to run them.
test_xrange
test_zipfile
test_zipfile64
test_zipfile64 skipped -- test requires loads of disk-space bytes and a long time to run
test_zipimport
test_zipimport_support
test_zlib
346 tests OK.
1 test failed:
    test_subprocess
37 tests skipped:
    test_aepack test_al test_applesingle test_bsddb185 test_bsddb3
    test_cd test_cl test_curses test_epoll test_gdb test_gl
    test_imgfile test_ioctl test_kqueue test_linuxaudiodev test_macos
    test_macostools test_multiprocessing test_ossaudiodev test_pep277
    test_py3kwarn test_scriptpackages test_smtpnet test_socketserver
    test_startfile test_sunaudiodev test_tcl test_timeout test_tk
    test_ttk_guionly test_ttk_textonly test_unicode_file
    test_urllib2net test_urllibnet test_winreg test_winsound
    test_zipfile64
7 skips unexpected on linux2:
    test_epoll test_gdb test_ioctl test_multiprocessing test_tk
    test_ttk_guionly test_ttk_textonly
[965838 refs]

From python-checkins at python.org  Thu Jul  8 23:15:36 2010
From: python-checkins at python.org (mark.dickinson)
Date: Thu,  8 Jul 2010 23:15:36 +0200 (CEST)
Subject: [Python-checkins] r82654 - in python/branches/py3k: Lib/decimal.py
	Misc/NEWS
Message-ID: <20100708211536.EC26DEEBB8@mail.python.org>

Author: mark.dickinson
Date: Thu Jul  8 23:15:36 2010
New Revision: 82654

Log:
Issue #9136: Profiling Decimal gave 'dictionary changed size during iteration'.
Remove the use of locals() that caused this error.


Modified:
   python/branches/py3k/Lib/decimal.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/decimal.py
==============================================================================
--- python/branches/py3k/Lib/decimal.py	(original)
+++ python/branches/py3k/Lib/decimal.py	Thu Jul  8 23:15:36 2010
@@ -3813,20 +3813,38 @@
                  Emin=None, Emax=None,
                  capitals=None, clamp=None,
                  _ignored_flags=None):
-        if flags is None:
-            flags = []
+        # Set defaults; for everything except flags and _ignored_flags,
+        # inherit from DefaultContext.
+        try:
+            dc = DefaultContext
+        except NameError:
+            pass
+
+        self.prec = prec if prec is not None else dc.prec
+        self.rounding = rounding if rounding is not None else dc.rounding
+        self.Emin = Emin if Emin is not None else dc.Emin
+        self.Emax = Emax if Emax is not None else dc.Emax
+        self.capitals = capitals if capitals is not None else dc.capitals
+        self.clamp = clamp if clamp is not None else dc.clamp
+
         if _ignored_flags is None:
-            _ignored_flags = []
-        if not isinstance(flags, dict):
-            flags = dict([(s, int(s in flags)) for s in _signals])
-        if traps is not None and not isinstance(traps, dict):
-            traps = dict([(s, int(s in traps)) for s in _signals])
-        for name, val in locals().items():
-            if val is None:
-                setattr(self, name, _copy.copy(getattr(DefaultContext, name)))
-            else:
-                setattr(self, name, val)
-        del self.self
+            self._ignored_flags = []
+        else:
+            self._ignored_flags = _ignored_flags
+
+        if traps is None:
+            self.traps = dc.traps.copy()
+        elif not isinstance(traps, dict):
+            self.traps = dict((s, int(s in traps)) for s in _signals)
+        else:
+            self.traps = traps
+
+        if flags is None:
+            self.flags = dict.fromkeys(_signals, 0)
+        elif not isinstance(flags, dict):
+            self.flags = dict((s, int(s in flags)) for s in _signals)
+        else:
+            self.flags = flags
 
     def __repr__(self):
         """Show the current context."""

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Thu Jul  8 23:15:36 2010
@@ -470,6 +470,10 @@
 Library
 -------
 
+- Issue #9136: Fix 'dictionary changed size during iteration'
+  RuntimeError produced when profiling the decimal module.  This was
+  due to a dangerous iteration over 'locals()' in Context.__init__.
+
 - Fix extreme speed issue in Decimal.pow when the base is an exact
   power of 10 and the exponent is tiny (for example,
   Decimal(10) ** Decimal('1e-999999999')).

From python-checkins at python.org  Thu Jul  8 23:18:21 2010
From: python-checkins at python.org (mark.dickinson)
Date: Thu,  8 Jul 2010 23:18:21 +0200 (CEST)
Subject: [Python-checkins] r82655 - in python/branches/release31-maint:
	Lib/decimal.py Misc/NEWS
Message-ID: <20100708211821.9E579EEAF6@mail.python.org>

Author: mark.dickinson
Date: Thu Jul  8 23:18:21 2010
New Revision: 82655

Log:
Merged revisions 82654 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82654 | mark.dickinson | 2010-07-08 22:15:36 +0100 (Thu, 08 Jul 2010) | 3 lines
  
  Issue #9136: Profiling Decimal gave 'dictionary changed size during iteration'.
  Remove the use of locals() that caused this error.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/decimal.py
   python/branches/release31-maint/Misc/NEWS

Modified: python/branches/release31-maint/Lib/decimal.py
==============================================================================
--- python/branches/release31-maint/Lib/decimal.py	(original)
+++ python/branches/release31-maint/Lib/decimal.py	Thu Jul  8 23:18:21 2010
@@ -3809,20 +3809,38 @@
                  Emin=None, Emax=None,
                  capitals=None, _clamp=0,
                  _ignored_flags=None):
-        if flags is None:
-            flags = []
+        # Set defaults; for everything except flags and _ignored_flags,
+        # inherit from DefaultContext.
+        try:
+            dc = DefaultContext
+        except NameError:
+            pass
+
+        self.prec = prec if prec is not None else dc.prec
+        self.rounding = rounding if rounding is not None else dc.rounding
+        self.Emin = Emin if Emin is not None else dc.Emin
+        self.Emax = Emax if Emax is not None else dc.Emax
+        self.capitals = capitals if capitals is not None else dc.capitals
+        self._clamp = _clamp if _clamp is not None else dc._clamp
+
         if _ignored_flags is None:
-            _ignored_flags = []
-        if not isinstance(flags, dict):
-            flags = dict([(s, int(s in flags)) for s in _signals])
-        if traps is not None and not isinstance(traps, dict):
-            traps = dict([(s, int(s in traps)) for s in _signals])
-        for name, val in locals().items():
-            if val is None:
-                setattr(self, name, _copy.copy(getattr(DefaultContext, name)))
-            else:
-                setattr(self, name, val)
-        del self.self
+            self._ignored_flags = []
+        else:
+            self._ignored_flags = _ignored_flags
+
+        if traps is None:
+            self.traps = dc.traps.copy()
+        elif not isinstance(traps, dict):
+            self.traps = dict((s, int(s in traps)) for s in _signals)
+        else:
+            self.traps = traps
+
+        if flags is None:
+            self.flags = dict.fromkeys(_signals, 0)
+        elif not isinstance(flags, dict):
+            self.flags = dict((s, int(s in flags)) for s in _signals)
+        else:
+            self.flags = flags
 
     def __repr__(self):
         """Show the current context."""

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Thu Jul  8 23:18:21 2010
@@ -75,6 +75,10 @@
 Library
 -------
 
+- Issue #9136: Fix 'dictionary changed size during iteration'
+  RuntimeError produced when profiling the decimal module.  This was
+  due to a dangerous iteration over 'locals()' in Context.__init__.
+
 - Fix extreme speed issue in Decimal.pow when the base is an exact
   power of 10 and the exponent is tiny (for example,
   Decimal(10) ** Decimal('1e-999999999')).

From python-checkins at python.org  Thu Jul  8 23:22:55 2010
From: python-checkins at python.org (mark.dickinson)
Date: Thu,  8 Jul 2010 23:22:55 +0200 (CEST)
Subject: [Python-checkins] r82656 - in python/branches/release27-maint:
	Lib/decimal.py Misc/NEWS
Message-ID: <20100708212255.34EAAEEC27@mail.python.org>

Author: mark.dickinson
Date: Thu Jul  8 23:22:54 2010
New Revision: 82656

Log:
Merged revisions 82654 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82654 | mark.dickinson | 2010-07-08 22:15:36 +0100 (Thu, 08 Jul 2010) | 3 lines
  
  Issue #9136: Profiling Decimal gave 'dictionary changed size during iteration'.
  Remove the use of locals() that caused this error.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/decimal.py
   python/branches/release27-maint/Misc/NEWS

Modified: python/branches/release27-maint/Lib/decimal.py
==============================================================================
--- python/branches/release27-maint/Lib/decimal.py	(original)
+++ python/branches/release27-maint/Lib/decimal.py	Thu Jul  8 23:22:54 2010
@@ -3756,22 +3756,38 @@
                  Emin=None, Emax=None,
                  capitals=None, _clamp=0,
                  _ignored_flags=None):
-        if flags is None:
-            flags = []
+        # Set defaults; for everything except flags and _ignored_flags,
+        # inherit from DefaultContext.
+        try:
+            dc = DefaultContext
+        except NameError:
+            pass
+
+        self.prec = prec if prec is not None else dc.prec
+        self.rounding = rounding if rounding is not None else dc.rounding
+        self.Emin = Emin if Emin is not None else dc.Emin
+        self.Emax = Emax if Emax is not None else dc.Emax
+        self.capitals = capitals if capitals is not None else dc.capitals
+        self._clamp = _clamp if _clamp is not None else dc._clamp
+
         if _ignored_flags is None:
-            _ignored_flags = []
-        if not isinstance(flags, dict):
-            flags = dict([(s, int(s in flags)) for s in _signals])
-            del s
-        if traps is not None and not isinstance(traps, dict):
-            traps = dict([(s, int(s in traps)) for s in _signals])
-            del s
-        for name, val in locals().items():
-            if val is None:
-                setattr(self, name, _copy.copy(getattr(DefaultContext, name)))
-            else:
-                setattr(self, name, val)
-        del self.self
+            self._ignored_flags = []
+        else:
+            self._ignored_flags = _ignored_flags
+
+        if traps is None:
+            self.traps = dc.traps.copy()
+        elif not isinstance(traps, dict):
+            self.traps = dict((s, int(s in traps)) for s in _signals)
+        else:
+            self.traps = traps
+
+        if flags is None:
+            self.flags = dict.fromkeys(_signals, 0)
+        elif not isinstance(flags, dict):
+            self.flags = dict((s, int(s in flags)) for s in _signals)
+        else:
+            self.flags = flags
 
     def __repr__(self):
         """Show the current context."""

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Thu Jul  8 23:22:54 2010
@@ -510,6 +510,10 @@
 Core and Builtins
 -----------------
 
+- Issue #9136: Fix 'dictionary changed size during iteration'
+  RuntimeError produced when profiling the decimal module.  This was
+  due to a dangerous iteration over 'locals()' in Context.__init__.
+
 - Issue #7301: Add environment variable $PYTHONWARNINGS.
 
 - Issue #8329: Don't return the same lists from select.select when no fds are

From python-checkins at python.org  Thu Jul  8 23:27:06 2010
From: python-checkins at python.org (mark.dickinson)
Date: Thu,  8 Jul 2010 23:27:06 +0200 (CEST)
Subject: [Python-checkins] r82657 - in python/branches/release26-maint:
	Lib/decimal.py Misc/NEWS
Message-ID: <20100708212706.06B55EEC0E@mail.python.org>

Author: mark.dickinson
Date: Thu Jul  8 23:27:05 2010
New Revision: 82657

Log:
Issue #9136: Fix 'dictionary changed size during iteration'
RuntimeError produced when profiling the decimal module.  This was
due to a dangerous iteration over 'locals()' in Context.__init__.
(Backport of r82656 from release27-maint.)



Modified:
   python/branches/release26-maint/Lib/decimal.py
   python/branches/release26-maint/Misc/NEWS

Modified: python/branches/release26-maint/Lib/decimal.py
==============================================================================
--- python/branches/release26-maint/Lib/decimal.py	(original)
+++ python/branches/release26-maint/Lib/decimal.py	Thu Jul  8 23:27:05 2010
@@ -3690,22 +3690,38 @@
                  Emin=None, Emax=None,
                  capitals=None, _clamp=0,
                  _ignored_flags=None):
-        if flags is None:
-            flags = []
+        # Set defaults; for everything except flags and _ignored_flags,
+        # inherit from DefaultContext.
+        try:
+            dc = DefaultContext
+        except NameError:
+            pass
+
+        self.prec = prec if prec is not None else dc.prec
+        self.rounding = rounding if rounding is not None else dc.rounding
+        self.Emin = Emin if Emin is not None else dc.Emin
+        self.Emax = Emax if Emax is not None else dc.Emax
+        self.capitals = capitals if capitals is not None else dc.capitals
+        self._clamp = _clamp if _clamp is not None else dc._clamp
+
         if _ignored_flags is None:
-            _ignored_flags = []
-        if not isinstance(flags, dict):
-            flags = dict([(s, int(s in flags)) for s in _signals])
-            del s
-        if traps is not None and not isinstance(traps, dict):
-            traps = dict([(s, int(s in traps)) for s in _signals])
-            del s
-        for name, val in locals().items():
-            if val is None:
-                setattr(self, name, _copy.copy(getattr(DefaultContext, name)))
-            else:
-                setattr(self, name, val)
-        del self.self
+            self._ignored_flags = []
+        else:
+            self._ignored_flags = _ignored_flags
+
+        if traps is None:
+            self.traps = dc.traps.copy()
+        elif not isinstance(traps, dict):
+            self.traps = dict((s, int(s in traps)) for s in _signals)
+        else:
+            self.traps = traps
+
+        if flags is None:
+            self.flags = dict.fromkeys(_signals, 0)
+        elif not isinstance(flags, dict):
+            self.flags = dict((s, int(s in flags)) for s in _signals)
+        else:
+            self.flags = flags
 
     def __repr__(self):
         """Show the current context."""

Modified: python/branches/release26-maint/Misc/NEWS
==============================================================================
--- python/branches/release26-maint/Misc/NEWS	(original)
+++ python/branches/release26-maint/Misc/NEWS	Thu Jul  8 23:27:05 2010
@@ -81,6 +81,10 @@
 Library
 -------
 
+- Issue #9136: Fix 'dictionary changed size during iteration'
+  RuntimeError produced when profiling the decimal module.  This was
+  due to a dangerous iteration over 'locals()' in Context.__init__.
+
 - Fix extreme speed issue in Decimal.pow when the base is an exact
   power of 10 and the exponent is tiny (for example,
   Decimal(10) ** Decimal('1e-999999999')).

From python-checkins at python.org  Thu Jul  8 23:28:32 2010
From: python-checkins at python.org (mark.dickinson)
Date: Thu,  8 Jul 2010 23:28:32 +0200 (CEST)
Subject: [Python-checkins] r82658 - python/branches/release27-maint/Misc/NEWS
Message-ID: <20100708212832.07547EEC55@mail.python.org>

Author: mark.dickinson
Date: Thu Jul  8 23:28:31 2010
New Revision: 82658

Log:
Move Misc/NEWS entries to the right place.

Modified:
   python/branches/release27-maint/Misc/NEWS

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Thu Jul  8 23:28:31 2010
@@ -16,6 +16,14 @@
 Library
 -------
 
+- Issue #9136: Fix 'dictionary changed size during iteration'
+  RuntimeError produced when profiling the decimal module.  This was
+  due to a dangerous iteration over 'locals()' in Context.__init__.
+
+- Fix extreme speed issue in Decimal.pow when the base is an exact
+  power of 10 and the exponent is tiny (for example,
+  Decimal(10) ** Decimal('1e-999999999')).
+
 - Issue #9161: Fix regression in optparse's acceptance of unicode
   strings in add_option calls.
 
@@ -510,19 +518,11 @@
 Core and Builtins
 -----------------
 
-- Issue #9136: Fix 'dictionary changed size during iteration'
-  RuntimeError produced when profiling the decimal module.  This was
-  due to a dangerous iteration over 'locals()' in Context.__init__.
-
 - Issue #7301: Add environment variable $PYTHONWARNINGS.
 
 - Issue #8329: Don't return the same lists from select.select when no fds are
   changed.
 
-- Fix extreme speed issue in Decimal.pow when the base is an exact
-  power of 10 and the exponent is tiny (for example,
-  Decimal(10) ** Decimal('1e-999999999')).
-
 - Issue #8259: ``1L << (2**31)`` no longer produces an 'outrageous shift error'
   on 64-bit machines.  The shift count for either left or right shift is
   permitted to be up to sys.maxsize.

From python-checkins at python.org  Thu Jul  8 23:39:09 2010
From: python-checkins at python.org (brian.curtin)
Date: Thu,  8 Jul 2010 23:39:09 +0200 (CEST)
Subject: [Python-checkins] r82659 - in python/branches/py3k:
	Doc/library/os.path.rst Doc/library/os.rst Lib/ntpath.py
	Lib/tarfile.py Lib/test/support.py
	Lib/test/symlink_support.py Lib/test/test_glob.py
	Lib/test/test_httpservers.py Lib/test/test_os.py
	Lib/test/test_platform.py Lib/test/test_posixpath.py
	Lib/test/test_shutil.py Lib/test/test_sys.py
	Lib/test/test_sysconfig.py Lib/test/test_tarfile.py Misc/ACKS
	Misc/NEWS Modules/posixmodule.c
Message-ID: <20100708213909.4B34DD267@mail.python.org>

Author: brian.curtin
Date: Thu Jul  8 23:39:08 2010
New Revision: 82659

Log:
Implement #1578269. Patch by Jason R. Coombs.

Added Windows support for os.symlink when run on Windows 6.0 or greater,
aka Vista. Previous Windows versions will raise NotImplementedError
when trying to symlink.

Includes numerous test updates and additions to test_os, including
a symlink_support module because of the fact that privilege escalation
is required in order to run the tests to ensure that the user is able
to create symlinks. By default, accounts do not have the required
privilege, so the escalation code will have to be exposed later (or
documented on how to do so). I'll be following up with that work next.

Note that the tests use ctypes, which was agreed on during the PyCon
language summit.



Added:
   python/branches/py3k/Lib/test/symlink_support.py   (contents, props changed)
Modified:
   python/branches/py3k/Doc/library/os.path.rst
   python/branches/py3k/Doc/library/os.rst
   python/branches/py3k/Lib/ntpath.py
   python/branches/py3k/Lib/tarfile.py
   python/branches/py3k/Lib/test/support.py
   python/branches/py3k/Lib/test/test_glob.py
   python/branches/py3k/Lib/test/test_httpservers.py
   python/branches/py3k/Lib/test/test_os.py
   python/branches/py3k/Lib/test/test_platform.py
   python/branches/py3k/Lib/test/test_posixpath.py
   python/branches/py3k/Lib/test/test_shutil.py
   python/branches/py3k/Lib/test/test_sys.py
   python/branches/py3k/Lib/test/test_sysconfig.py
   python/branches/py3k/Lib/test/test_tarfile.py
   python/branches/py3k/Misc/ACKS
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/Modules/posixmodule.c

Modified: python/branches/py3k/Doc/library/os.path.rst
==============================================================================
--- python/branches/py3k/Doc/library/os.path.rst	(original)
+++ python/branches/py3k/Doc/library/os.path.rst	Thu Jul  8 23:39:08 2010
@@ -231,11 +231,15 @@
 
 .. function:: samefile(path1, path2)
 
-   Return ``True`` if both pathname arguments refer to the same file or directory
-   (as indicated by device number and i-node number). Raise an exception if a
-   :func:`os.stat` call on either pathname fails.
+   Return ``True`` if both pathname arguments refer to the same file or directory.
+   On Unix, this is determined by the device number and i-node number and raises an
+   exception if a :func:`os.stat` call on either pathname fails.
+
+   On Windows, two files are the same if they resolve to the same final path
+   name using the Windows API call GetFinalPathNameByHandle and this function
+   raises an exception if handles cannot be obtained to either file.
 
-   Availability: Unix.
+   Availability: Windows, Unix.
 
 
 .. function:: sameopenfile(fp1, fp2)

Modified: python/branches/py3k/Doc/library/os.rst
==============================================================================
--- python/branches/py3k/Doc/library/os.rst	(original)
+++ python/branches/py3k/Doc/library/os.rst	Thu Jul  8 23:39:08 2010
@@ -1065,7 +1065,7 @@
 
    Like :func:`stat`, but do not follow symbolic links.  This is an alias for
    :func:`stat` on platforms that do not support symbolic links, such as
-   Windows.
+   Windows prior to 6.0 (Vista).
 
 
 .. function:: mkfifo(path[, mode])
@@ -1181,7 +1181,7 @@
    and the call may raise an UnicodeDecodeError. If the *path* is a bytes
    object, the result will be a bytes object.
 
-   Availability: Unix.
+   Availability: Unix, Windows.
 
 
 .. function:: remove(path)
@@ -1341,9 +1341,25 @@
 
 .. function:: symlink(source, link_name)
 
-   Create a symbolic link pointing to *source* named *link_name*.
+   Create a symbolic link pointing to *source* named *link_name*.  On Windows,
+   symlink version takes an additional, optional parameter,
+   *target_is_directory*, which defaults to False.
+
+   symlink(source, link_name, target_is_directory=False)
+
+   On Windows, a symlink represents a file or a directory, and does not
+   morph to the target dynamically.  For this reason, when creating a
+   symlink on Windows, if the target is not already present, the symlink
+   will default to being a file symlink.  If *target_is_directory* is set to
+   True, the symlink will be created as a directory symlink.  This
+   parameter is ignored if the target exists (and the symlink is created
+   with the same type as the target).
+
+   Symbolic link support was introduced in Windows 6.0 (Vista). *symlink*
+   will raise a NotImplementedError on Windows versions earlier than 6.0. The
+   SeCreateSymbolicLinkPrivilege is required in order to create symlinks.
 
-   Availability: Unix.
+   Availability:  Unix, Windows 6.0.
 
 
 .. function:: unlink(path)

Modified: python/branches/py3k/Lib/ntpath.py
==============================================================================
--- python/branches/py3k/Lib/ntpath.py	(original)
+++ python/branches/py3k/Lib/ntpath.py	Thu Jul  8 23:39:08 2010
@@ -16,7 +16,8 @@
            "getatime","getctime", "islink","exists","lexists","isdir","isfile",
            "ismount", "expanduser","expandvars","normpath","abspath",
            "splitunc","curdir","pardir","sep","pathsep","defpath","altsep",
-           "extsep","devnull","realpath","supports_unicode_filenames","relpath"]
+           "extsep","devnull","realpath","supports_unicode_filenames","relpath",
+           "samefile",]
 
 # strings representing various path-related bits and pieces
 # These are primarily for export; internally, they are hardcoded.
@@ -309,16 +310,28 @@
     return split(p)[0]
 
 # Is a path a symbolic link?
-# This will always return false on systems where posix.lstat doesn't exist.
+# This will always return false on systems where os.lstat doesn't exist.
 
 def islink(path):
-    """Test for symbolic link.
-    On WindowsNT/95 and OS/2 always returns false
+    """Test whether a path is a symbolic link.
+    This will always return false for Windows prior to 6.0
+    and for OS/2.
     """
-    return False
-
-# alias exists to lexists
-lexists = exists
+    try:
+        st = os.lstat(path)
+    except (os.error, AttributeError):
+        return False
+    return stat.S_ISLNK(st.st_mode)
+
+# Being true for dangling symbolic links is also useful.
+
+def lexists(path):
+    """Test whether a path exists.  Returns True for broken symbolic links"""
+    try:
+        st = os.lstat(path)
+    except (os.error, WindowsError):
+        return False
+    return True
 
 # Is a path a mount point?  Either a root (with or without drive letter)
 # or an UNC path with at most a / or \ after the mount point.
@@ -612,3 +625,17 @@
     if not rel_list:
         return _get_dot(path)
     return join(*rel_list)
+
+
+# determine if two files are in fact the same file
+def samefile(f1, f2):
+    "Test whether two pathnames reference the same actual file"
+    try:
+        from nt import _getfinalpathname
+        return _getfinalpathname(f1) == _getfinalpathname(f2)
+    except (NotImplementedError, ImportError):
+        # On Windows XP and earlier, two files are the same if their
+        #  absolute pathnames are the same.
+        # Also, on other operating systems, fake this method with a
+        #  Windows-XP approximation.
+        return abspath(f1) == abspath(f2)

Modified: python/branches/py3k/Lib/tarfile.py
==============================================================================
--- python/branches/py3k/Lib/tarfile.py	(original)
+++ python/branches/py3k/Lib/tarfile.py	Thu Jul  8 23:39:08 2010
@@ -2273,7 +2273,7 @@
           (platform limitation), we try to make a copy of the referenced file
           instead of a link.
         """
-        if hasattr(os, "symlink") and hasattr(os, "link"):
+        try:
             # For systems that support symbolic and hard links.
             if tarinfo.issym():
                 os.symlink(tarinfo.linkname, targetpath)
@@ -2282,7 +2282,15 @@
                 if os.path.exists(tarinfo._link_target):
                     os.link(tarinfo._link_target, targetpath)
                 else:
-                    self._extract_member(self._find_link_target(tarinfo), targetpath)
+                    self._extract_mem
+        except (AttributeError, NotImplementedError, WindowsError):
+            # AttributeError if no os.symlink
+            # NotImplementedError if on Windows XP
+            # WindowsError (1314) if the required privilege is not held by the client
+            if tarinfo.issym():
+                linkpath = os.path.join(os.path.dirname(tarinfo.name),tarinfo.linkname)
+            else:
+                linkpath = tarinfo.linkname
         else:
             try:
                 self._extract_member(self._find_link_target(tarinfo), targetpath)

Modified: python/branches/py3k/Lib/test/support.py
==============================================================================
--- python/branches/py3k/Lib/test/support.py	(original)
+++ python/branches/py3k/Lib/test/support.py	Thu Jul  8 23:39:08 2010
@@ -17,6 +17,7 @@
 import importlib
 import collections
 import re
+import subprocess
 import imp
 import time
 try:
@@ -38,8 +39,7 @@
     "set_memlimit", "bigmemtest", "bigaddrspacetest", "BasicTestRunner",
     "run_unittest", "run_doctest", "threading_setup", "threading_cleanup",
     "reap_children", "cpython_only", "check_impl_detail", "get_attribute",
-    "swap_item", "swap_attr",
-    ]
+    "swap_item", "swap_attr", "can_symlink", "skip_unless_symlink"]
 
 
 class Error(Exception):
@@ -1169,6 +1169,27 @@
             except:
                 break
 
+try:
+    from .symlink_support import enable_symlink_privilege
+except:
+    enable_symlink_privilege = lambda: True
+
+def can_symlink():
+    """It's no longer sufficient to test for the presence of symlink in the
+    os module - on Windows XP and earlier, os.symlink exists but a
+    NotImplementedError is thrown.
+    """
+    has_symlink = hasattr(os, 'symlink')
+    is_old_windows = sys.platform == "win32" and sys.getwindowsversion().major < 6
+    has_privilege = False if is_old_windows else enable_symlink_privilege()
+    return has_symlink and (not is_old_windows) and has_privilege
+
+def skip_unless_symlink(test):
+    """Skip decorator for tests that require functional symlink"""
+    selector = can_symlink()
+    msg = "Requires functional symlink implementation"
+    return [unittest.skip(msg)(test), test][selector]
+
 @contextlib.contextmanager
 def swap_attr(obj, attr, new_val):
     """Temporary swap out an attribute with a new object.

Added: python/branches/py3k/Lib/test/symlink_support.py
==============================================================================
--- (empty file)
+++ python/branches/py3k/Lib/test/symlink_support.py	Thu Jul  8 23:39:08 2010
@@ -0,0 +1,203 @@
+"""
+A module built to test if the current process has the privilege to
+create symlinks on Windows.
+"""
+
+# allow script to run natively under python 2.6+
+from __future__ import print_function
+
+import ctypes
+from ctypes import wintypes
+
+GetCurrentProcess = ctypes.windll.kernel32.GetCurrentProcess
+GetCurrentProcess.restype = wintypes.HANDLE
+OpenProcessToken = ctypes.windll.advapi32.OpenProcessToken
+OpenProcessToken.argtypes = (wintypes.HANDLE, wintypes.DWORD, ctypes.POINTER(wintypes.HANDLE))
+OpenProcessToken.restype = wintypes.BOOL
+
+class LUID(ctypes.Structure):
+    _fields_ = [
+            ('low_part', wintypes.DWORD),
+            ('high_part', wintypes.LONG),
+            ]
+
+    def __eq__(self, other):
+        return (
+                self.high_part == other.high_part and
+                self.low_part == other.low_part
+                )
+
+    def __ne__(self, other):
+        return not (self==other)
+
+LookupPrivilegeValue = ctypes.windll.advapi32.LookupPrivilegeValueW
+LookupPrivilegeValue.argtypes = (
+        wintypes.LPWSTR, # system name
+        wintypes.LPWSTR, # name
+        ctypes.POINTER(LUID),
+        )
+LookupPrivilegeValue.restype = wintypes.BOOL
+
+class TOKEN_INFORMATION_CLASS:
+    TokenUser = 1
+    TokenGroups = 2
+    TokenPrivileges = 3
+    # ... see http://msdn.microsoft.com/en-us/library/aa379626%28VS.85%29.aspx
+
+SE_PRIVILEGE_ENABLED_BY_DEFAULT = (0x00000001)
+SE_PRIVILEGE_ENABLED            = (0x00000002)
+SE_PRIVILEGE_REMOVED            = (0x00000004)
+SE_PRIVILEGE_USED_FOR_ACCESS    = (0x80000000)
+
+class LUID_AND_ATTRIBUTES(ctypes.Structure):
+    _fields_ = [
+            ('LUID', LUID),
+            ('attributes', wintypes.DWORD),
+            ]
+
+    def is_enabled(self):
+        return bool(self.attributes & SE_PRIVILEGE_ENABLED)
+
+    def enable(self):
+        self.attributes |= SE_PRIVILEGE_ENABLED
+
+    def get_name(self):
+        size = wintypes.DWORD(10240)
+        buf = ctypes.create_unicode_buffer(size.value)
+        res = LookupPrivilegeName(None, self.LUID, buf, size)
+        if res == 0:
+            raise RuntimeError
+        return buf[:size.value]
+
+    def __str__(self):
+        name = self.name
+        fmt = ['{name}', '{name} (enabled)'][self.is_enabled()]
+        return fmt.format(**vars())
+
+LookupPrivilegeName = ctypes.windll.advapi32.LookupPrivilegeNameW
+LookupPrivilegeName.argtypes = (
+        wintypes.LPWSTR, # lpSystemName
+        ctypes.POINTER(LUID), # lpLuid
+        wintypes.LPWSTR, # lpName
+        ctypes.POINTER(wintypes.DWORD), #cchName
+        )
+LookupPrivilegeName.restype = wintypes.BOOL
+
+class TOKEN_PRIVILEGES(ctypes.Structure):
+    _fields_ = [
+            ('count', wintypes.DWORD),
+            ('privileges', LUID_AND_ATTRIBUTES*0),
+            ]
+
+    def get_array(self):
+        array_type = LUID_AND_ATTRIBUTES*self.count
+        privileges = ctypes.cast(self.privileges, ctypes.POINTER(array_type)).contents
+        return privileges
+
+    def __iter__(self):
+        return iter(self.get_array())
+
+PTOKEN_PRIVILEGES = ctypes.POINTER(TOKEN_PRIVILEGES)
+
+GetTokenInformation = ctypes.windll.advapi32.GetTokenInformation
+GetTokenInformation.argtypes = [
+        wintypes.HANDLE, # TokenHandle
+        ctypes.c_uint, # TOKEN_INFORMATION_CLASS value
+        ctypes.c_void_p, # TokenInformation
+        wintypes.DWORD, # TokenInformationLength
+        ctypes.POINTER(wintypes.DWORD), # ReturnLength
+        ]
+GetTokenInformation.restype = wintypes.BOOL
+
+# http://msdn.microsoft.com/en-us/library/aa375202%28VS.85%29.aspx
+AdjustTokenPrivileges = ctypes.windll.advapi32.AdjustTokenPrivileges
+AdjustTokenPrivileges.restype = wintypes.BOOL
+AdjustTokenPrivileges.argtypes = [
+        wintypes.HANDLE,                # TokenHandle
+        wintypes.BOOL,                  # DisableAllPrivileges
+        PTOKEN_PRIVILEGES,              # NewState (optional)
+        wintypes.DWORD,                 # BufferLength of PreviousState
+        PTOKEN_PRIVILEGES,              # PreviousState (out, optional)
+        ctypes.POINTER(wintypes.DWORD), # ReturnLength
+        ]
+
+def get_process_token():
+    "Get the current process token"
+    token = wintypes.HANDLE()
+    TOKEN_ALL_ACCESS = 0xf01ff
+    res = OpenProcessToken(GetCurrentProcess(), TOKEN_ALL_ACCESS, token)
+    if not res > 0:
+        raise RuntimeError("Couldn't get process token")
+    return token
+
+def get_symlink_luid():
+    "Get the LUID for the SeCreateSymbolicLinkPrivilege"
+    symlink_luid = LUID()
+    res = LookupPrivilegeValue(None, "SeCreateSymbolicLinkPrivilege", symlink_luid)
+    if not res > 0:
+        raise RuntimeError("Couldn't lookup privilege value")
+    return symlink_luid
+
+def get_privilege_information():
+    "Get all privileges associated with the current process."
+    # first call with zero length to determine what size buffer we need
+
+    return_length = wintypes.DWORD()
+    params = [
+            get_process_token(),
+            TOKEN_INFORMATION_CLASS.TokenPrivileges,
+            None,
+            0,
+            return_length,
+            ]
+
+    res = GetTokenInformation(*params)
+
+    # assume we now have the necessary length in return_length
+
+    buffer = ctypes.create_string_buffer(return_length.value)
+    params[2] = buffer
+    params[3] = return_length.value
+
+    res = GetTokenInformation(*params)
+    assert res > 0, "Error in second GetTokenInformation (%d)" % res
+
+    privileges = ctypes.cast(buffer, ctypes.POINTER(TOKEN_PRIVILEGES)).contents
+    return privileges
+
+def report_privilege_information():
+    "Report all privilege information assigned to the current process."
+    privileges = get_privilege_information()
+    print("found {0} privileges".format(privileges.count))
+    tuple(map(print, privileges))
+
+def enable_symlink_privilege():
+    """
+    Try to assign the symlink privilege to the current process token.
+    Return True if the assignment is successful.
+    """
+    # create a space in memory for a TOKEN_PRIVILEGES structure
+    #  with one element
+    size = ctypes.sizeof(TOKEN_PRIVILEGES)
+    size += ctypes.sizeof(LUID_AND_ATTRIBUTES)
+    buffer = ctypes.create_string_buffer(size)
+    tp = ctypes.cast(buffer, ctypes.POINTER(TOKEN_PRIVILEGES)).contents
+    tp.count = 1
+    tp.get_array()[0].enable()
+    tp.get_array()[0].LUID = get_symlink_luid()
+    token = get_process_token()
+    res = AdjustTokenPrivileges(token, False, tp, 0, None, None)
+    if res == 0:
+        raise RuntimeError("Error in AdjustTokenPrivileges")
+
+    ERROR_NOT_ALL_ASSIGNED = 1300
+    return ctypes.windll.kernel32.GetLastError() != ERROR_NOT_ALL_ASSIGNED
+
+def main():
+    assigned = enable_symlink_privilege()
+    msg = ['failure', 'success'][assigned]
+
+    print("Symlink privilege assignment completed with {0}".format(msg))
+
+if __name__ == '__main__':
+    main()

Modified: python/branches/py3k/Lib/test/test_glob.py
==============================================================================
--- python/branches/py3k/Lib/test/test_glob.py	(original)
+++ python/branches/py3k/Lib/test/test_glob.py	Thu Jul  8 23:39:08 2010
@@ -1,5 +1,5 @@
 import unittest
-from test.support import run_unittest, TESTFN
+from test.support import run_unittest, TESTFN, skip_unless_symlink, can_symlink
 import glob
 import os
 import shutil
@@ -25,7 +25,7 @@
         self.mktemp('ZZZ')
         self.mktemp('a', 'bcd', 'EF')
         self.mktemp('a', 'bcd', 'efg', 'ha')
-        if hasattr(os, 'symlink'):
+        if can_symlink():
             os.symlink(self.norm('broken'), self.norm('sym1'))
             os.symlink(self.norm('broken'), self.norm('sym2'))
 
@@ -98,12 +98,12 @@
         # either of these results are reasonable
         self.assertIn(res[0], [self.tempdir, self.tempdir + os.sep])
 
+    @skip_unless_symlink
     def test_glob_broken_symlinks(self):
-        if hasattr(os, 'symlink'):
-            eq = self.assertSequencesEqual_noorder
-            eq(self.glob('sym*'), [self.norm('sym1'), self.norm('sym2')])
-            eq(self.glob('sym1'), [self.norm('sym1')])
-            eq(self.glob('sym2'), [self.norm('sym2')])
+        eq = self.assertSequencesEqual_noorder
+        eq(self.glob('sym*'), [self.norm('sym1'), self.norm('sym2')])
+        eq(self.glob('sym1'), [self.norm('sym1')])
+        eq(self.glob('sym2'), [self.norm('sym2')])
 
 
 def test_main():

Modified: python/branches/py3k/Lib/test/test_httpservers.py
==============================================================================
--- python/branches/py3k/Lib/test/test_httpservers.py	(original)
+++ python/branches/py3k/Lib/test/test_httpservers.py	Thu Jul  8 23:39:08 2010
@@ -299,7 +299,7 @@
 
         # The shebang line should be pure ASCII: use symlink if possible.
         # See issue #7668.
-        if hasattr(os, 'symlink'):
+        if support.can_symlink():
             self.pythonexe = os.path.join(self.parent_dir, 'python')
             os.symlink(sys.executable, self.pythonexe)
         else:

Modified: python/branches/py3k/Lib/test/test_os.py
==============================================================================
--- python/branches/py3k/Lib/test/test_os.py	(original)
+++ python/branches/py3k/Lib/test/test_os.py	Thu Jul  8 23:39:08 2010
@@ -520,7 +520,7 @@
             f = open(path, "w")
             f.write("I'm " + path + " and proud of it.  Blame test_os.\n")
             f.close()
-        if hasattr(os, "symlink"):
+        if support.can_symlink():
             os.symlink(os.path.abspath(t2_path), link_path)
             sub2_tree = (sub2_path, ["link"], ["tmp3"])
         else:
@@ -564,7 +564,7 @@
         self.assertEqual(all[flipped + 1], (sub1_path, ["SUB11"], ["tmp2"]))
         self.assertEqual(all[2 - 2 * flipped], sub2_tree)
 
-        if hasattr(os, "symlink"):
+        if support.can_symlink():
             # Walk, following symlinks.
             for root, dirs, files in os.walk(walk_path, followlinks=True):
                 if root == link_path:
@@ -1033,6 +1033,83 @@
         self._kill_with_event(signal.CTRL_BREAK_EVENT, "CTRL_BREAK_EVENT")
 
 
+def skipUnlessWindows6(test):
+    if hasattr(sys, 'getwindowsversion') and sys.getwindowsversion().major >= 6:
+        return test
+    return unittest.skip("Requires Windows Vista or later")(test)
+
+ at unittest.skipUnless(sys.platform == "win32", "Win32 specific tests")
+ at support.skip_unless_symlink
+class Win32SymlinkTests(unittest.TestCase):
+    filelink = 'filelinktest'
+    filelink_target = os.path.abspath(__file__)
+    dirlink = 'dirlinktest'
+    dirlink_target = os.path.dirname(filelink_target)
+    missing_link = 'missing link'
+
+    def setUp(self):
+        assert os.path.exists(self.dirlink_target)
+        assert os.path.exists(self.filelink_target)
+        assert not os.path.exists(self.dirlink)
+        assert not os.path.exists(self.filelink)
+        assert not os.path.exists(self.missing_link)
+
+    def tearDown(self):
+        if os.path.exists(self.filelink):
+            os.remove(self.filelink)
+        if os.path.exists(self.dirlink):
+            os.rmdir(self.dirlink)
+        if os.path.lexists(self.missing_link):
+            os.remove(self.missing_link)
+
+    def test_directory_link(self):
+        os.symlink(self.dirlink_target, self.dirlink)
+        self.assertTrue(os.path.exists(self.dirlink))
+        self.assertTrue(os.path.isdir(self.dirlink))
+        self.assertTrue(os.path.islink(self.dirlink))
+        self.check_stat(self.dirlink, self.dirlink_target)
+
+    def test_file_link(self):
+        os.symlink(self.filelink_target, self.filelink)
+        self.assertTrue(os.path.exists(self.filelink))
+        self.assertTrue(os.path.isfile(self.filelink))
+        self.assertTrue(os.path.islink(self.filelink))
+        self.check_stat(self.filelink, self.filelink_target)
+
+    def _create_missing_dir_link(self):
+        'Create a "directory" link to a non-existent target'
+        linkname = self.missing_link
+        if os.path.lexists(linkname):
+            os.remove(linkname)
+        target = r'c:\\target does not exist.29r3c740'
+        assert not os.path.exists(target)
+        target_is_dir = True
+        os.symlink(target, linkname, target_is_dir)
+
+    def test_remove_directory_link_to_missing_target(self):
+        self._create_missing_dir_link()
+        # For compatibility with Unix, os.remove will check the
+        #  directory status and call RemoveDirectory if the symlink
+        #  was created with target_is_dir==True.
+        os.remove(self.missing_link)
+
+    @unittest.skip("currently fails; consider for improvement")
+    def test_isdir_on_directory_link_to_missing_target(self):
+        self._create_missing_dir_link()
+        # consider having isdir return true for directory links
+        self.assertTrue(os.path.isdir(self.missing_link))
+
+    @unittest.skip("currently fails; consider for improvement")
+    def test_rmdir_on_directory_link_to_missing_target(self):
+        self._create_missing_dir_link()
+        # consider allowing rmdir to remove directory links
+        os.rmdir(self.missing_link)
+
+    def check_stat(self, link, target):
+        self.assertEqual(os.stat(link), os.stat(target))
+        self.assertNotEqual(os.lstat(link), os.stat(link))
+
+
 class MiscTests(unittest.TestCase):
 
     @unittest.skipIf(os.name == "nt", "POSIX specific test")
@@ -1056,6 +1133,7 @@
         PosixUidGidTests,
         Pep383Tests,
         Win32KillTests,
+        Win32SymlinkTests,
         MiscTests,
     )
 

Modified: python/branches/py3k/Lib/test/test_platform.py
==============================================================================
--- python/branches/py3k/Lib/test/test_platform.py	(original)
+++ python/branches/py3k/Lib/test/test_platform.py	Thu Jul  8 23:39:08 2010
@@ -10,20 +10,26 @@
     def test_architecture(self):
         res = platform.architecture()
 
-    if hasattr(os, "symlink"):
-        def test_architecture_via_symlink(self): # issue3762
-            def get(python):
-                cmd = [python, '-c',
-                    'import platform; print(platform.architecture())']
-                p = subprocess.Popen(cmd, stdout=subprocess.PIPE)
-                return p.communicate()
-            real = os.path.realpath(sys.executable)
-            link = os.path.abspath(support.TESTFN)
-            os.symlink(real, link)
-            try:
-                self.assertEqual(get(real), get(link))
-            finally:
-                os.remove(link)
+    @support.skip_unless_symlink
+    def test_architecture_via_symlink(self): # issue3762
+        # On Windows, the EXE needs to know where pythonXY.dll is at so we have
+        # to add the directory to the path.
+        if sys.platform == "win32":
+            os.environ["Path"] = "{};{}".format(os.path.dirname(sys.executable),
+                                                os.environ["Path"])
+
+        def get(python):
+            cmd = [python, '-c',
+                'import platform; print(platform.architecture())']
+            p = subprocess.Popen(cmd, stdout=subprocess.PIPE)
+            return p.communicate()
+        real = os.path.realpath(sys.executable)
+        link = os.path.abspath(support.TESTFN)
+        os.symlink(real, link)
+        try:
+            self.assertEqual(get(real), get(link))
+        finally:
+            os.remove(link)
 
     def test_platform(self):
         for aliased in (False, True):

Modified: python/branches/py3k/Lib/test/test_posixpath.py
==============================================================================
--- python/branches/py3k/Lib/test/test_posixpath.py	(original)
+++ python/branches/py3k/Lib/test/test_posixpath.py	Thu Jul  8 23:39:08 2010
@@ -1,7 +1,9 @@
 import unittest
 from test import support, test_genericpath
 
-import posixpath, os
+import posixpath
+import os
+import sys
 from posixpath import realpath, abspath, dirname, basename
 
 # An absolute path to a temporary filename for testing. We can't rely on TESTFN
@@ -9,6 +11,16 @@
 
 ABSTFN = abspath(support.TESTFN)
 
+def skip_if_ABSTFN_contains_backslash(test):
+    """
+    On Windows, posixpath.abspath still returns paths with backslashes
+    instead of posix forward slashes. If this is the case, several tests
+    fail, so skip them.
+    """
+    found_backslash = '\\' in ABSTFN
+    msg = "ABSTFN is not a posix path - tests fail"
+    return [test, unittest.skip(msg)(test)][found_backslash]
+
 def safe_rmdir(dirname):
     try:
         os.rmdir(dirname)
@@ -143,7 +155,7 @@
             f.write(b"foo")
             f.close()
             self.assertIs(posixpath.islink(support.TESTFN + "1"), False)
-            if hasattr(os, "symlink"):
+            if support.can_symlink():
                 os.symlink(support.TESTFN + "1", support.TESTFN + "2")
                 self.assertIs(posixpath.islink(support.TESTFN + "2"), True)
                 os.remove(support.TESTFN + "1")
@@ -154,85 +166,59 @@
             if not f.close():
                 f.close()
 
+    @staticmethod
+    def _create_file(filename):
+        with open(filename, 'wb') as f:
+            f.write(b'foo')
+
     def test_samefile(self):
-        f = open(support.TESTFN + "1", "wb")
-        try:
-            f.write(b"foo")
-            f.close()
-            self.assertIs(
-                posixpath.samefile(
-                    support.TESTFN + "1",
-                    support.TESTFN + "1"
-                ),
-                True
-            )
-            # If we don't have links, assume that os.stat doesn't return resonable
-            # inode information and thus, that samefile() doesn't work
-            if hasattr(os, "symlink"):
-                os.symlink(
-                    support.TESTFN + "1",
-                    support.TESTFN + "2"
-                )
-                self.assertIs(
-                    posixpath.samefile(
-                        support.TESTFN + "1",
-                        support.TESTFN + "2"
-                    ),
-                    True
-                )
-                os.remove(support.TESTFN + "2")
-                f = open(support.TESTFN + "2", "wb")
-                f.write(b"bar")
-                f.close()
-                self.assertIs(
-                    posixpath.samefile(
-                        support.TESTFN + "1",
-                        support.TESTFN + "2"
-                    ),
-                    False
-                )
-        finally:
-            if not f.close():
-                f.close()
+        test_fn = support.TESTFN + "1"
+        self._create_file(test_fn)
+        self.assertTrue(posixpath.samefile(test_fn, test_fn))
+        self.assertRaises(TypeError, posixpath.samefile)
+
+    @unittest.skipIf(
+        sys.platform.startswith('win'),
+        "posixpath.samefile does not work on links in Windows")
+    @support.skip_unless_symlink
+    def test_samefile_on_links(self):
+        test_fn1 = support.TESTFN + "1"
+        test_fn2 = support.TESTFN + "2"
+        self._create_file(test_fn1)
+
+        os.symlink(test_fn1, test_fn2)
+        self.assertTrue(posixpath.samefile(test_fn1, test_fn2))
+        os.remove(test_fn2)
+
+        self._create_file(test_fn2)
+        self.assertFalse(posixpath.samefile(test_fn1, test_fn2))
+
 
     def test_samestat(self):
-        f = open(support.TESTFN + "1", "wb")
-        try:
-            f.write(b"foo")
-            f.close()
-            self.assertIs(
-                posixpath.samestat(
-                    os.stat(support.TESTFN + "1"),
-                    os.stat(support.TESTFN + "1")
-                ),
-                True
-            )
-            # If we don't have links, assume that os.stat() doesn't return resonable
-            # inode information and thus, that samefile() doesn't work
-            if hasattr(os, "symlink"):
-                if hasattr(os, "symlink"):
-                    os.symlink(support.TESTFN + "1", support.TESTFN + "2")
-                    self.assertIs(
-                        posixpath.samestat(
-                            os.stat(support.TESTFN + "1"),
-                            os.stat(support.TESTFN + "2")
-                        ),
-                        True
-                    )
-                    os.remove(support.TESTFN + "2")
-                f = open(support.TESTFN + "2", "wb")
-                f.write(b"bar")
-                f.close()
-                self.assertIs(
-                    posixpath.samestat(
-                        os.stat(support.TESTFN + "1"),
-                        os.stat(support.TESTFN + "2")
-                    ),
-                    False
-                )
-        finally:
-            if not f.close():
-                f.close()
+        test_fn = support.TESTFN + "1"
+        self._create_file(test_fn)
+        test_fns = [test_fn]*2
+        stats = map(os.stat, test_fns)
+        self.assertTrue(posixpath.samestat(*stats))
+
+    @unittest.skipIf(
+        sys.platform.startswith('win'),
+        "posixpath.samestat does not work on links in Windows")
+    @support.skip_unless_symlink
+    def test_samestat_on_links(self):
+        test_fn1 = support.TESTFN + "1"
+        test_fn2 = support.TESTFN + "2"
+        test_fns = (test_fn1, test_fn2)
+        os.symlink(*test_fns)
+        stats = map(os.stat, test_fns)
+        self.assertTrue(posixpath.samestat(*stats))
+        os.remove(test_fn2)
+
+        self._create_file(test_fn2)
+        stats = map(os.stat, test_fns)
+        self.assertFalse(posixpath.samestat(*stats))
+
+        self.assertRaises(TypeError, posixpath.samestat)
 
     def test_ismount(self):
         self.assertIs(posixpath.ismount("/"), True)
@@ -286,103 +272,112 @@
         self.assertEqual(posixpath.normpath(b"///..//./foo/.//bar"),
                          b"/foo/bar")
 
-    if hasattr(os, "symlink"):
-        def test_realpath_basic(self):
-            # Basic operation.
-            try:
-                os.symlink(ABSTFN+"1", ABSTFN)
-                self.assertEqual(realpath(ABSTFN), ABSTFN+"1")
-            finally:
-                support.unlink(ABSTFN)
-
-        def test_realpath_symlink_loops(self):
-            # Bug #930024, return the path unchanged if we get into an infinite
-            # symlink loop.
-            try:
-                old_path = abspath('.')
-                os.symlink(ABSTFN, ABSTFN)
-                self.assertEqual(realpath(ABSTFN), ABSTFN)
-
-                os.symlink(ABSTFN+"1", ABSTFN+"2")
-                os.symlink(ABSTFN+"2", ABSTFN+"1")
-                self.assertEqual(realpath(ABSTFN+"1"), ABSTFN+"1")
-                self.assertEqual(realpath(ABSTFN+"2"), ABSTFN+"2")
-
-                # Test using relative path as well.
-                os.chdir(dirname(ABSTFN))
-                self.assertEqual(realpath(basename(ABSTFN)), ABSTFN)
-            finally:
-                os.chdir(old_path)
-                support.unlink(ABSTFN)
-                support.unlink(ABSTFN+"1")
-                support.unlink(ABSTFN+"2")
-
-        def test_realpath_resolve_parents(self):
-            # We also need to resolve any symlinks in the parents of a relative
-            # path passed to realpath. E.g.: current working directory is
-            # /usr/doc with 'doc' being a symlink to /usr/share/doc. We call
-            # realpath("a"). This should return /usr/share/doc/a/.
-            try:
-                old_path = abspath('.')
-                os.mkdir(ABSTFN)
-                os.mkdir(ABSTFN + "/y")
-                os.symlink(ABSTFN + "/y", ABSTFN + "/k")
-
-                os.chdir(ABSTFN + "/k")
-                self.assertEqual(realpath("a"), ABSTFN + "/y/a")
-            finally:
-                os.chdir(old_path)
-                support.unlink(ABSTFN + "/k")
-                safe_rmdir(ABSTFN + "/y")
-                safe_rmdir(ABSTFN)
-
-        def test_realpath_resolve_before_normalizing(self):
-            # Bug #990669: Symbolic links should be resolved before we
-            # normalize the path. E.g.: if we have directories 'a', 'k' and 'y'
-            # in the following hierarchy:
-            # a/k/y
-            #
-            # and a symbolic link 'link-y' pointing to 'y' in directory 'a',
-            # then realpath("link-y/..") should return 'k', not 'a'.
-            try:
-                old_path = abspath('.')
-                os.mkdir(ABSTFN)
-                os.mkdir(ABSTFN + "/k")
-                os.mkdir(ABSTFN + "/k/y")
-                os.symlink(ABSTFN + "/k/y", ABSTFN + "/link-y")
-
-                # Absolute path.
-                self.assertEqual(realpath(ABSTFN + "/link-y/.."), ABSTFN + "/k")
-                # Relative path.
-                os.chdir(dirname(ABSTFN))
-                self.assertEqual(realpath(basename(ABSTFN) + "/link-y/.."),
-                                 ABSTFN + "/k")
-            finally:
-                os.chdir(old_path)
-                support.unlink(ABSTFN + "/link-y")
-                safe_rmdir(ABSTFN + "/k/y")
-                safe_rmdir(ABSTFN + "/k")
-                safe_rmdir(ABSTFN)
-
-        def test_realpath_resolve_first(self):
-            # Bug #1213894: The first component of the path, if not absolute,
-            # must be resolved too.
-
-            try:
-                old_path = abspath('.')
-                os.mkdir(ABSTFN)
-                os.mkdir(ABSTFN + "/k")
-                os.symlink(ABSTFN, ABSTFN + "link")
-                os.chdir(dirname(ABSTFN))
-
-                base = basename(ABSTFN)
-                self.assertEqual(realpath(base + "link"), ABSTFN)
-                self.assertEqual(realpath(base + "link/k"), ABSTFN + "/k")
-            finally:
-                os.chdir(old_path)
-                support.unlink(ABSTFN + "link")
-                safe_rmdir(ABSTFN + "/k")
-                safe_rmdir(ABSTFN)
+    @support.skip_unless_symlink
+    @skip_if_ABSTFN_contains_backslash
+    def test_realpath_basic(self):
+        # Basic operation.
+        try:
+            os.symlink(ABSTFN+"1", ABSTFN)
+            self.assertEqual(realpath(ABSTFN), ABSTFN+"1")
+        finally:
+            support.unlink(ABSTFN)
+
+    @support.skip_unless_symlink
+    @skip_if_ABSTFN_contains_backslash
+    def test_realpath_symlink_loops(self):
+        # Bug #930024, return the path unchanged if we get into an infinite
+        # symlink loop.
+        try:
+            old_path = abspath('.')
+            os.symlink(ABSTFN, ABSTFN)
+            self.assertEqual(realpath(ABSTFN), ABSTFN)
+
+            os.symlink(ABSTFN+"1", ABSTFN+"2")
+            os.symlink(ABSTFN+"2", ABSTFN+"1")
+            self.assertEqual(realpath(ABSTFN+"1"), ABSTFN+"1")
+            self.assertEqual(realpath(ABSTFN+"2"), ABSTFN+"2")
+
+            # Test using relative path as well.
+            os.chdir(dirname(ABSTFN))
+            self.assertEqual(realpath(basename(ABSTFN)), ABSTFN)
+        finally:
+            os.chdir(old_path)
+            support.unlink(ABSTFN)
+            support.unlink(ABSTFN+"1")
+            support.unlink(ABSTFN+"2")
+
+    @support.skip_unless_symlink
+    @skip_if_ABSTFN_contains_backslash
+    def test_realpath_resolve_parents(self):
+        # We also need to resolve any symlinks in the parents of a relative
+        # path passed to realpath. E.g.: current working directory is
+        # /usr/doc with 'doc' being a symlink to /usr/share/doc. We call
+        # realpath("a"). This should return /usr/share/doc/a/.
+        try:
+            old_path = abspath('.')
+            os.mkdir(ABSTFN)
+            os.mkdir(ABSTFN + "/y")
+            os.symlink(ABSTFN + "/y", ABSTFN + "/k")
+
+            os.chdir(ABSTFN + "/k")
+            self.assertEqual(realpath("a"), ABSTFN + "/y/a")
+        finally:
+            os.chdir(old_path)
+            support.unlink(ABSTFN + "/k")
+            safe_rmdir(ABSTFN + "/y")
+            safe_rmdir(ABSTFN)
+
+    @support.skip_unless_symlink
+    @skip_if_ABSTFN_contains_backslash
+    def test_realpath_resolve_before_normalizing(self):
+        # Bug #990669: Symbolic links should be resolved before we
+        # normalize the path. E.g.: if we have directories 'a', 'k' and 'y'
+        # in the following hierarchy:
+        # a/k/y
+        #
+        # and a symbolic link 'link-y' pointing to 'y' in directory 'a',
+        # then realpath("link-y/..") should return 'k', not 'a'.
+        try:
+            old_path = abspath('.')
+            os.mkdir(ABSTFN)
+            os.mkdir(ABSTFN + "/k")
+            os.mkdir(ABSTFN + "/k/y")
+            os.symlink(ABSTFN + "/k/y", ABSTFN + "/link-y")
+
+            # Absolute path.
+            self.assertEqual(realpath(ABSTFN + "/link-y/.."), ABSTFN + "/k")
+            # Relative path.
+            os.chdir(dirname(ABSTFN))
+            self.assertEqual(realpath(basename(ABSTFN) + "/link-y/.."),
+                             ABSTFN + "/k")
+        finally:
+            os.chdir(old_path)
+            support.unlink(ABSTFN + "/link-y")
+            safe_rmdir(ABSTFN + "/k/y")
+            safe_rmdir(ABSTFN + "/k")
+            safe_rmdir(ABSTFN)
+
+    @support.skip_unless_symlink
+    @skip_if_ABSTFN_contains_backslash
+    def test_realpath_resolve_first(self):
+        # Bug #1213894: The first component of the path, if not absolute,
+        # must be resolved too.
+
+        try:
+            old_path = abspath('.')
+            os.mkdir(ABSTFN)
+            os.mkdir(ABSTFN + "/k")
+            os.symlink(ABSTFN, ABSTFN + "link")
+            os.chdir(dirname(ABSTFN))
+
+            base = basename(ABSTFN)
+            self.assertEqual(realpath(base + "link"), ABSTFN)
+            self.assertEqual(realpath(base + "link/k"), ABSTFN + "/k")
+        finally:
+            os.chdir(old_path)
+            support.unlink(ABSTFN + "link")
+            safe_rmdir(ABSTFN + "/k")
+            safe_rmdir(ABSTFN)
 
     def test_relpath(self):
         (real_getcwd, os.getcwd) = (os.getcwd, lambda: r"/home/user/bar")

Modified: python/branches/py3k/Lib/test/test_shutil.py
==============================================================================
--- python/branches/py3k/Lib/test/test_shutil.py	(original)
+++ python/branches/py3k/Lib/test/test_shutil.py	Thu Jul  8 23:39:08 2010
@@ -271,7 +271,7 @@
             shutil.rmtree(src_dir)
             shutil.rmtree(os.path.dirname(dst_dir))
 
-    @unittest.skipUnless(hasattr(os, 'symlink'), 'requires os.symlink')
+    @support.skip_unless_symlink
     def test_dont_copy_file_onto_link_to_itself(self):
         # bug 851123.
         os.mkdir(TESTFN)
@@ -282,10 +282,11 @@
             f.write('cheddar')
             f.close()
 
-            os.link(src, dst)
-            self.assertRaises(shutil.Error, shutil.copyfile, src, dst)
-            self.assertEqual(open(src,'r').read(), 'cheddar')
-            os.remove(dst)
+            if hasattr(os, "link"):
+                os.link(src, dst)
+                self.assertRaises(shutil.Error, shutil.copyfile, src, dst)
+                self.assertEqual(open(src,'r').read(), 'cheddar')
+                os.remove(dst)
 
             # Using `src` here would mean we end up with a symlink pointing
             # to TESTFN/TESTFN/cheese, while it should point at
@@ -300,30 +301,30 @@
             except OSError:
                 pass
 
-        @unittest.skipUnless(hasattr(os, 'symlink'), 'requires os.symlink')
-        def test_rmtree_on_symlink(self):
-            # bug 1669.
-            os.mkdir(TESTFN)
+    @support.skip_unless_symlink
+    def test_rmtree_on_symlink(self):
+        # bug 1669.
+        os.mkdir(TESTFN)
+        try:
+            src = os.path.join(TESTFN, 'cheese')
+            dst = os.path.join(TESTFN, 'shop')
+            os.mkdir(src)
+            os.symlink(src, dst)
+            self.assertRaises(OSError, shutil.rmtree, dst)
+        finally:
+            shutil.rmtree(TESTFN, ignore_errors=True)
+
+    if hasattr(os, "mkfifo"):
+        # Issue #3002: copyfile and copytree block indefinitely on named pipes
+        def test_copyfile_named_pipe(self):
+            os.mkfifo(TESTFN)
             try:
-                src = os.path.join(TESTFN, 'cheese')
-                dst = os.path.join(TESTFN, 'shop')
-                os.mkdir(src)
-                os.symlink(src, dst)
-                self.assertRaises(OSError, shutil.rmtree, dst)
+                self.assertRaises(shutil.SpecialFileError,
+                                  shutil.copyfile, TESTFN, TESTFN2)
+                self.assertRaises(shutil.SpecialFileError,
+                                  shutil.copyfile, __file__, TESTFN)
             finally:
-                shutil.rmtree(TESTFN, ignore_errors=True)
-
-    @unittest.skipUnless(hasattr(os, 'mkfifo'), 'requires os.mkfifo')
-    # Issue #3002: copyfile and copytree block indefinitely on named pipes
-    def test_copyfile_named_pipe(self):
-        os.mkfifo(TESTFN)
-        try:
-            self.assertRaises(shutil.SpecialFileError,
-                                shutil.copyfile, TESTFN, TESTFN2)
-            self.assertRaises(shutil.SpecialFileError,
-                                shutil.copyfile, __file__, TESTFN)
-        finally:
-            os.remove(TESTFN)
+                os.remove(TESTFN)
 
     @unittest.skipUnless(hasattr(os, 'mkfifo'), 'requires os.mkfifo')
     def test_copytree_named_pipe(self):
@@ -361,7 +362,7 @@
         shutil.copytree(src_dir, dst_dir, copy_function=_copy)
         self.assertEquals(len(copied), 2)
 
-    @unittest.skipUnless(hasattr(os, 'symlink'), 'requires os.symlink')
+    @support.skip_unless_symlink
     def test_copytree_dangling_symlinks(self):
 
         # a dangling symlink raises an error at the end

Modified: python/branches/py3k/Lib/test/test_sys.py
==============================================================================
--- python/branches/py3k/Lib/test/test_sys.py	(original)
+++ python/branches/py3k/Lib/test/test_sys.py	Thu Jul  8 23:39:08 2010
@@ -263,7 +263,7 @@
         # Raise SkipTest if sys doesn't have getwindowsversion attribute
         test.support.get_attribute(sys, "getwindowsversion")
         v = sys.getwindowsversion()
-        self.assertEqual(len(v), 5)
+        self.assertEqual(len(v), 9)
         self.assertIsInstance(v[0], int)
         self.assertIsInstance(v[1], int)
         self.assertIsInstance(v[2], int)

Modified: python/branches/py3k/Lib/test/test_sysconfig.py
==============================================================================
--- python/branches/py3k/Lib/test/test_sysconfig.py	(original)
+++ python/branches/py3k/Lib/test/test_sysconfig.py	Thu Jul  8 23:39:08 2010
@@ -12,7 +12,7 @@
 from copy import copy, deepcopy
 
 from test.support import (run_unittest, TESTFN, unlink, get_attribute,
-                          captured_stdout)
+                          captured_stdout, skip_unless_symlink)
 
 import sysconfig
 from sysconfig import (get_paths, get_platform, get_config_vars,
@@ -239,17 +239,23 @@
                   'posix_home', 'posix_prefix', 'posix_user')
         self.assertEquals(get_scheme_names(), wanted)
 
+    @skip_unless_symlink
     def test_symlink(self):
+        # On Windows, the EXE needs to know where pythonXY.dll is at so we have
+        # to add the directory to the path.
+        if sys.platform == "win32":
+            os.environ["Path"] = "{};{}".format(os.path.dirname(sys.executable),
+                                                os.environ["Path"])
+
         # Issue 7880
-        symlink = get_attribute(os, "symlink")
         def get(python):
             cmd = [python, '-c',
                    'import sysconfig; print(sysconfig.get_platform())']
-            p = subprocess.Popen(cmd, stdout=subprocess.PIPE)
+            p = subprocess.Popen(cmd, stdout=subprocess.PIPE, env=os.environ)
             return p.communicate()
         real = os.path.realpath(sys.executable)
         link = os.path.abspath(TESTFN)
-        symlink(real, link)
+        os.symlink(real, link)
         try:
             self.assertEqual(get(real), get(link))
         finally:

Modified: python/branches/py3k/Lib/test/test_tarfile.py
==============================================================================
--- python/branches/py3k/Lib/test/test_tarfile.py	(original)
+++ python/branches/py3k/Lib/test/test_tarfile.py	Thu Jul  8 23:39:08 2010
@@ -291,6 +291,8 @@
         self.assertTrue(self.tar.getmembers()[-1].name == "misc/eof",
                 "could not find all members")
 
+    @unittest.skipUnless(hasattr(os, "link"), "Missing hardlink implementation")
+    @support.skip_unless_symlink
     def test_extract_hardlink(self):
         # Test hardlink extraction (e.g. bug #857297).
         tar = tarfile.open(tarname, errorlevel=1, encoding="iso8859-1")
@@ -695,16 +697,16 @@
                 os.remove(target)
                 os.remove(link)
 
+    @support.skip_unless_symlink
     def test_symlink_size(self):
-        if hasattr(os, "symlink"):
-            path = os.path.join(TEMPDIR, "symlink")
-            os.symlink("link_target", path)
-            try:
-                tar = tarfile.open(tmpname, self.mode)
-                tarinfo = tar.gettarinfo(path)
-                self.assertEqual(tarinfo.size, 0)
-            finally:
-                os.remove(path)
+        path = os.path.join(TEMPDIR, "symlink")
+        os.symlink("link_target", path)
+        try:
+            tar = tarfile.open(tmpname, self.mode)
+            tarinfo = tar.gettarinfo(path)
+            self.assertEqual(tarinfo.size, 0)
+        finally:
+            os.remove(path)
 
     def test_add_self(self):
         # Test for #1257255.
@@ -1408,15 +1410,24 @@
         data = open(os.path.join(TEMPDIR, name), "rb").read()
         self.assertEqual(md5sum(data), md5_regtype)
 
+    # When 8879 gets fixed, this will need to change. Currently on Windows
+    # we have os.path.islink but no os.link, so these tests fail without the
+    # following skip until link is completed.
+    @unittest.skipIf(hasattr(os.path, "islink"),
+                     "Skip emulation - has os.path.islink but not os.link")
     def test_hardlink_extraction1(self):
         self._test_link_extraction("ustar/lnktype")
 
+    @unittest.skipIf(hasattr(os.path, "islink"),
+                     "Skip emulation - has os.path.islink but not os.link")
     def test_hardlink_extraction2(self):
         self._test_link_extraction("./ustar/linktest2/lnktype")
 
+    @unittest.skipIf(hasattr(os, "symlink"), "Skip emulation if symlink exists")
     def test_symlink_extraction1(self):
         self._test_link_extraction("ustar/symtype")
 
+    @unittest.skipIf(hasattr(os, "symlink"), "Skip emulation if symlink exists")
     def test_symlink_extraction2(self):
         self._test_link_extraction("./ustar/linktest2/symtype")
 

Modified: python/branches/py3k/Misc/ACKS
==============================================================================
--- python/branches/py3k/Misc/ACKS	(original)
+++ python/branches/py3k/Misc/ACKS	Thu Jul  8 23:39:08 2010
@@ -154,6 +154,7 @@
 Jeffery Collins
 Robert Collins
 Paul Colomiets
+Jason R. Coombs
 Geremy Condra
 Juan Jos? Conti
 Matt Conway

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Thu Jul  8 23:39:08 2010
@@ -1406,6 +1406,9 @@
 Extension Modules
 -----------------
 
+- Issue #1578269: Implement os.symlink for Windows 6.0+. Patch by
+  Jason R. Coombs
+
 - In struct.pack, correctly propogate exceptions from computing the truth of an
   object in the '?' format.
 

Modified: python/branches/py3k/Modules/posixmodule.c
==============================================================================
--- python/branches/py3k/Modules/posixmodule.c	(original)
+++ python/branches/py3k/Modules/posixmodule.c	Thu Jul  8 23:39:08 2010
@@ -571,7 +571,7 @@
 
 #ifdef MS_WINDOWS
 static PyObject *
-win32_error(char* function, char* filename)
+win32_error(char* function, const char* filename)
 {
     /* XXX We should pass the function name along in the future.
        (winreg.c also wants to pass the function name.)
@@ -978,12 +978,28 @@
     return TRUE;
 }
 
-static int
-win32_stat(const char* path, struct win32_stat *result)
+/* About the following functions: win32_lstat, win32_lstat_w, win32_stat,
+   win32_stat_w
+
+   In Posix, stat automatically traverses symlinks and returns the stat
+   structure for the target.  In Windows, the equivalent GetFileAttributes by
+   default does not traverse symlinks and instead returns attributes for
+   the symlink.
+
+   Therefore, win32_lstat will get the attributes traditionally, and
+   win32_stat will first explicitly resolve the symlink target and then will
+   call win32_lstat on that result.
+
+   The _w represent Unicode equivalents of the aformentioned ANSI functions. */
+
+static int 
+win32_lstat(const char* path, struct win32_stat *result)
 {
     WIN32_FILE_ATTRIBUTE_DATA info;
     int code;
     char *dot;
+    WIN32_FIND_DATAA find_data;
+    HANDLE find_data_handle;
     if (!GetFileAttributesExA(path, GetFileExInfoStandard, &info)) {
         if (GetLastError() != ERROR_SHARING_VIOLATION) {
             /* Protocol violation: we explicitly clear errno, instead of
@@ -1000,27 +1016,44 @@
             }
         }
     }
+
     code = attribute_data_to_stat(&info, result);
     if (code != 0)
         return code;
+
+    /* Get WIN32_FIND_DATA structure for the path to determine if
+       it is a symlink */
+    if(info.dwFileAttributes & FILE_ATTRIBUTE_REPARSE_POINT) {
+        find_data_handle = FindFirstFileA(path, &find_data);
+        if(find_data_handle != INVALID_HANDLE_VALUE) {
+            if(find_data.dwReserved0 == IO_REPARSE_TAG_SYMLINK) {
+                /* first clear the S_IFMT bits */
+                result->st_mode ^= (result->st_mode & 0170000);
+                /* now set the bits that make this a symlink */
+                result->st_mode |= 0120000;
+            }
+            FindClose(find_data_handle);
+        }
+    }
+
     /* Set S_IFEXEC if it is an .exe, .bat, ... */
     dot = strrchr(path, '.');
     if (dot) {
-        if (stricmp(dot, ".bat") == 0 ||
-        stricmp(dot, ".cmd") == 0 ||
-        stricmp(dot, ".exe") == 0 ||
-        stricmp(dot, ".com") == 0)
+        if (stricmp(dot, ".bat") == 0 || stricmp(dot, ".cmd") == 0 ||
+            stricmp(dot, ".exe") == 0 || stricmp(dot, ".com") == 0)
             result->st_mode |= 0111;
     }
     return code;
 }
 
 static int
-win32_wstat(const wchar_t* path, struct win32_stat *result)
+win32_lstat_w(const wchar_t* path, struct win32_stat *result)
 {
     int code;
     const wchar_t *dot;
     WIN32_FILE_ATTRIBUTE_DATA info;
+    WIN32_FIND_DATAW find_data;
+    HANDLE find_data_handle;
     if (!GetFileAttributesExW(path, GetFileExInfoStandard, &info)) {
         if (GetLastError() != ERROR_SHARING_VIOLATION) {
             /* Protocol violation: we explicitly clear errno, instead of
@@ -1028,8 +1061,8 @@
             errno = 0;
             return -1;
         } else {
-            /* Could not get attributes on open file. Fall back to
-               reading the directory. */
+            /* Could not get attributes on open file. Fall back to reading
+            the directory. */
             if (!attributes_from_dir_w(path, &info)) {
                 /* Very strange. This should not fail now */
                 errno = 0;
@@ -1040,18 +1073,198 @@
     code = attribute_data_to_stat(&info, result);
     if (code < 0)
         return code;
+
+    /* Get WIN32_FIND_DATA structure for the path to determine if
+       it is a symlink */
+    if(info.dwFileAttributes & FILE_ATTRIBUTE_REPARSE_POINT) {
+        find_data_handle = FindFirstFileW(path, &find_data);
+        if(find_data_handle != INVALID_HANDLE_VALUE) {
+            if(find_data.dwReserved0 == IO_REPARSE_TAG_SYMLINK) {
+                /* first clear the S_IFMT bits */
+                result->st_mode ^= (result->st_mode & 0170000);
+                /* now set the bits that make this a symlink */
+                result->st_mode |= 0120000;
+            }
+            FindClose(find_data_handle);
+        }
+    }
+
     /* Set IFEXEC if it is an .exe, .bat, ... */
     dot = wcsrchr(path, '.');
     if (dot) {
-        if (_wcsicmp(dot, L".bat") == 0 ||
-            _wcsicmp(dot, L".cmd") == 0 ||
-            _wcsicmp(dot, L".exe") == 0 ||
-            _wcsicmp(dot, L".com") == 0)
+        if (_wcsicmp(dot, L".bat") == 0 || _wcsicmp(dot, L".cmd") == 0 ||
+            _wcsicmp(dot, L".exe") == 0 || _wcsicmp(dot, L".com") == 0)
             result->st_mode |= 0111;
     }
     return code;
 }
 
+/* Grab GetFinalPathNameByHandle dynamically from kernel32 */
+static int has_GetFinalPathNameByHandle = 0;
+static DWORD (CALLBACK *Py_GetFinalPathNameByHandleA)(HANDLE, LPSTR, DWORD,
+                                                      DWORD);
+static DWORD (CALLBACK *Py_GetFinalPathNameByHandleW)(HANDLE, LPWSTR, DWORD,
+                                                      DWORD);
+static int
+check_GetFinalPathNameByHandle()
+{
+    HINSTANCE hKernel32;
+    /* only recheck */
+    if (!has_GetFinalPathNameByHandle)
+    {
+        hKernel32 = GetModuleHandle("KERNEL32");
+        *(FARPROC*)&Py_GetFinalPathNameByHandleA = GetProcAddress(hKernel32,
+                                                "GetFinalPathNameByHandleA");
+        *(FARPROC*)&Py_GetFinalPathNameByHandleW = GetProcAddress(hKernel32,
+                                                "GetFinalPathNameByHandleW");
+        has_GetFinalPathNameByHandle = Py_GetFinalPathNameByHandleA && Py_GetFinalPathNameByHandleW;
+    }
+    return has_GetFinalPathNameByHandle;
+}
+
+static int
+win32_stat(const char* path, struct win32_stat *result)
+{
+    /* Traverse the symlink to the target using
+    GetFinalPathNameByHandle()
+    */
+    int code;
+    HANDLE hFile;
+    int buf_size;
+    char *target_path;
+    int result_length;
+    WIN32_FILE_ATTRIBUTE_DATA info;
+    
+    if(!check_GetFinalPathNameByHandle()) {
+        /* if the OS doesn't have GetFinalPathNameByHandle, it doesn't
+           have symlinks, so just fall back to the traditional behavior
+           found in lstat. */
+        return win32_lstat(path, result);
+    }
+
+    hFile = CreateFileA(
+        path,
+        0, /* desired access */
+        0, /* share mode */
+        NULL, /* security attributes */
+        OPEN_EXISTING,
+        /* FILE_FLAG_BACKUP_SEMANTICS is required to open a directory */
+        FILE_FLAG_BACKUP_SEMANTICS,
+        NULL);
+    
+    if(hFile == INVALID_HANDLE_VALUE) {
+        /* Either the target doesn't exist, or we don't have access to
+           get a handle to it. If the former, we need to return an error.
+           If the latter, we can use attributes_from_dir. */
+        if (GetLastError() != ERROR_SHARING_VIOLATION) {
+            /* Protocol violation: we explicitly clear errno, instead of
+               setting it to a POSIX error. Callers should use GetLastError. */
+            errno = 0;
+            return -1;
+        } else {
+            /* Could not get attributes on open file. Fall back to
+               reading the directory. */
+            if (!attributes_from_dir(path, &info)) {
+                /* Very strange. This should not fail now */
+                errno = 0;
+                return -1;
+            }
+        }
+        code = attribute_data_to_stat(&info, result);
+    }
+    
+    buf_size = Py_GetFinalPathNameByHandleA(hFile, 0, 0, VOLUME_NAME_DOS);
+    if(!buf_size) return -1;
+    target_path = (char *)malloc((buf_size+1)*sizeof(char));
+    result_length = Py_GetFinalPathNameByHandleA(hFile, target_path,
+                                                 buf_size, VOLUME_NAME_DOS);
+    
+    if(!result_length)
+        return -1;
+
+    if(!CloseHandle(hFile))
+        return -1;
+
+    target_path[result_length] = 0;
+    code = win32_lstat(target_path, result);
+    free(target_path);
+    
+    return code;
+}
+
+static int 
+win32_stat_w(const wchar_t* path, struct win32_stat *result)
+{
+    /* Traverse the symlink to the target using GetFinalPathNameByHandle() */
+    int code;
+    HANDLE hFile;
+    int buf_size;
+    wchar_t *target_path;
+    int result_length;
+    WIN32_FILE_ATTRIBUTE_DATA info;
+    
+    if(!check_GetFinalPathNameByHandle()) {
+        /* If the OS doesn't have GetFinalPathNameByHandle, it doesn't have
+           symlinks, so just fall back to the traditional behavior found
+           in lstat. */
+        return win32_lstat_w(path, result);
+    }
+
+    hFile = CreateFileW(
+        path,
+        0, /* desired access */
+        0, /* share mode */
+        NULL, /* security attributes */
+        OPEN_EXISTING,
+        /* FILE_FLAG_BACKUP_SEMANTICS is required to open a directory */
+        FILE_ATTRIBUTE_NORMAL|FILE_FLAG_BACKUP_SEMANTICS,
+        NULL);
+    
+    if(hFile == INVALID_HANDLE_VALUE) {
+        /* Either the target doesn't exist, or we don't have access to
+           get a handle to it. If the former, we need to return an error.
+           If the latter, we can use attributes_from_dir. */
+        if (GetLastError() != ERROR_SHARING_VIOLATION) {
+            /* Protocol violation: we explicitly clear errno, instead of
+               setting it to a POSIX error. Callers should use GetLastError. */
+            errno = 0;
+            return -1;
+        } else {
+            /* Could not get attributes on open file. Fall back to
+               reading the directory. */
+            if (!attributes_from_dir_w(path, &info)) {
+                /* Very strange. This should not fail now */
+                errno = 0;
+                return -1;
+            }
+        }
+        code = attribute_data_to_stat(&info, result);
+    }
+    else {
+        /* We have a good handle to the target, use it to determine the target
+           path name (then we'll call lstat on it). */
+        buf_size = Py_GetFinalPathNameByHandleW(hFile, 0, 0, VOLUME_NAME_DOS);
+        if(!buf_size)
+            return -1;
+
+        target_path = (wchar_t *)malloc((buf_size+1)*sizeof(wchar_t));
+        result_length = Py_GetFinalPathNameByHandleW(hFile, target_path,
+                                                buf_size, VOLUME_NAME_DOS);
+        
+        if(!result_length)
+            return -1;
+
+        if(!CloseHandle(hFile))
+            return -1;
+
+        target_path[result_length] = 0;
+        code = win32_lstat_w(target_path, result);
+        free(target_path);
+    }
+    
+    return code;
+}
+
 static int
 win32_fstat(int file_number, struct win32_stat *result)
 {
@@ -2443,6 +2656,69 @@
     }
     return PyBytes_FromString(outbuf);
 } /* end of posix__getfullpathname */
+
+/* A helper function for samepath on windows */
+static PyObject *
+posix__getfinalpathname(PyObject *self, PyObject *args)
+{
+    HANDLE hFile;
+    int buf_size;
+    wchar_t *target_path;
+    int result_length;
+    PyObject *result;
+    wchar_t *path;
+    
+    if (!PyArg_ParseTuple(args, "u|:_getfullpathname", &path)) {
+        return NULL;
+    }
+
+    if(!check_GetFinalPathNameByHandle()) {
+        /* If the OS doesn't have GetFinalPathNameByHandle, return a
+           NotImplementedError. */
+        return PyErr_Format(PyExc_NotImplementedError,
+            "GetFinalPathNameByHandle not available on this platform");
+    }
+
+    hFile = CreateFileW(
+        path,
+        0, /* desired access */
+        0, /* share mode */
+        NULL, /* security attributes */
+        OPEN_EXISTING,
+        /* FILE_FLAG_BACKUP_SEMANTICS is required to open a directory */
+        FILE_FLAG_BACKUP_SEMANTICS,
+        NULL);
+    
+    if(hFile == INVALID_HANDLE_VALUE) {
+        return win32_error_unicode("GetFinalPathNamyByHandle", path);
+        return PyErr_Format(PyExc_RuntimeError, "Could not get a handle to file.");
+    }
+
+    /* We have a good handle to the target, use it to determine the
+       target path name. */
+    buf_size = Py_GetFinalPathNameByHandleW(hFile, 0, 0, VOLUME_NAME_NT);
+
+    if(!buf_size)
+        return win32_error_unicode("GetFinalPathNameByHandle", path);
+
+    target_path = (wchar_t *)malloc((buf_size+1)*sizeof(wchar_t));
+    if(!target_path)
+        return PyErr_NoMemory();
+
+    result_length = Py_GetFinalPathNameByHandleW(hFile, target_path,
+                                                 buf_size, VOLUME_NAME_DOS);
+    if(!result_length)
+        return win32_error_unicode("GetFinalPathNamyByHandle", path);
+
+    if(!CloseHandle(hFile))
+        return win32_error_unicode("GetFinalPathNameByHandle", path);
+
+    target_path[result_length] = 0;
+    result = PyUnicode_FromUnicode(target_path, result_length);
+    free(target_path);
+    return result;
+
+} /* end of posix__getfinalpathname */
 #endif /* MS_WINDOWS */
 
 PyDoc_STRVAR(posix_mkdir__doc__,
@@ -2623,7 +2899,7 @@
 posix_stat(PyObject *self, PyObject *args)
 {
 #ifdef MS_WINDOWS
-    return posix_do_stat(self, args, "O&:stat", STAT, "U:stat", win32_wstat);
+    return posix_do_stat(self, args, "O&:stat", STAT, "U:stat", win32_stat_w);
 #else
     return posix_do_stat(self, args, "O&:stat", STAT, NULL, NULL);
 #endif
@@ -2681,6 +2957,41 @@
     return PyLong_FromLong((long)i);
 }
 
+#ifdef MS_WINDOWS
+
+/* override the default DeleteFileW behavior so that directory
+symlinks can be removed with this function, the same as with
+Unix symlinks */
+BOOL WINAPI Py_DeleteFileW(LPCWSTR lpFileName)
+{
+    WIN32_FILE_ATTRIBUTE_DATA info;
+    WIN32_FIND_DATAW find_data;
+    HANDLE find_data_handle;
+    int is_directory = 0;
+    int is_link = 0;
+
+    if (GetFileAttributesExW(lpFileName, GetFileExInfoStandard, &info)) {
+        is_directory = info.dwFileAttributes & FILE_ATTRIBUTE_DIRECTORY;
+        
+        /* Get WIN32_FIND_DATA structure for the path to determine if
+           it is a symlink */
+        if(is_directory &&
+           info.dwFileAttributes & FILE_ATTRIBUTE_REPARSE_POINT) {
+            find_data_handle = FindFirstFileW(lpFileName, &find_data);
+
+            if(find_data_handle != INVALID_HANDLE_VALUE) {
+                is_link = find_data.dwReserved0 == IO_REPARSE_TAG_SYMLINK;
+                FindClose(find_data_handle);
+            }
+        }
+    }
+
+    if (is_directory && is_link)
+        return RemoveDirectoryW(lpFileName);
+
+    return DeleteFileW(lpFileName);
+}
+#endif /* MS_WINDOWS */
 
 PyDoc_STRVAR(posix_unlink__doc__,
 "unlink(path)\n\n\
@@ -2694,7 +3005,7 @@
 posix_unlink(PyObject *self, PyObject *args)
 {
 #ifdef MS_WINDOWS
-    return win32_1str(args, "remove", "y:remove", DeleteFileA, "U:remove", DeleteFileW);
+    return win32_1str(args, "remove", "y:remove", DeleteFileA, "U:remove", Py_DeleteFileW);
 #else
     return posix_1str(args, "O&:remove", unlink);
 #endif
@@ -4544,7 +4855,8 @@
     return posix_do_stat(self, args, "O&:lstat", lstat, NULL, NULL);
 #else /* !HAVE_LSTAT */
 #ifdef MS_WINDOWS
-    return posix_do_stat(self, args, "O&:lstat", STAT, "U:lstat", win32_wstat);
+    return posix_do_stat(self, args, "O&:lstat", STAT, "U:lstat",
+                         win32_lstat_w);
 #else
     return posix_do_stat(self, args, "O&:lstat", STAT, NULL, NULL);
 #endif
@@ -4609,6 +4921,193 @@
 }
 #endif /* HAVE_SYMLINK */
 
+#if !defined(HAVE_READLINK) && defined(MS_WINDOWS)
+
+PyDoc_STRVAR(win_readlink__doc__,
+"readlink(path) -> path\n\n\
+Return a string representing the path to which the symbolic link points.");
+
+/* The following structure was copied from
+   http://msdn.microsoft.com/en-us/library/ms791514.aspx as the required
+   include doesn't seem to be present in the Windows SDK (at least as included
+   with Visual Studio Express). */
+typedef struct _REPARSE_DATA_BUFFER {
+    ULONG ReparseTag;
+    USHORT ReparseDataLength;
+    USHORT Reserved;
+    union {
+        struct {
+            USHORT SubstituteNameOffset;
+            USHORT SubstituteNameLength;
+            USHORT PrintNameOffset;
+            USHORT PrintNameLength;
+            ULONG Flags;
+            WCHAR PathBuffer[1];
+        } SymbolicLinkReparseBuffer;
+
+        struct {
+            USHORT SubstituteNameOffset;
+            USHORT  SubstituteNameLength;
+            USHORT  PrintNameOffset;
+            USHORT  PrintNameLength;
+            WCHAR  PathBuffer[1];
+        } MountPointReparseBuffer;
+
+        struct {
+            UCHAR  DataBuffer[1];
+        } GenericReparseBuffer;
+    };
+} REPARSE_DATA_BUFFER, *PREPARSE_DATA_BUFFER;
+
+#define REPARSE_DATA_BUFFER_HEADER_SIZE  FIELD_OFFSET(REPARSE_DATA_BUFFER, GenericReparseBuffer)
+
+#define MAXIMUM_REPARSE_DATA_BUFFER_SIZE  ( 16 * 1024 )
+
+/* Windows readlink implementation */
+static PyObject *
+win_readlink(PyObject *self, PyObject *args)
+{
+    wchar_t *path;
+    DWORD n_bytes_returned;
+    DWORD io_result;
+    PyObject *result;
+    HANDLE reparse_point_handle;
+
+    char target_buffer[MAXIMUM_REPARSE_DATA_BUFFER_SIZE];
+    REPARSE_DATA_BUFFER *rdb = (REPARSE_DATA_BUFFER *)target_buffer;
+    wchar_t *print_name;
+
+    if (!PyArg_ParseTuple(args,
+                  "u:readlink",
+                  &path))
+        return NULL;
+
+    /* First get a handle to the reparse point */
+    Py_BEGIN_ALLOW_THREADS
+    reparse_point_handle = CreateFileW(
+        path,
+        0,
+        0,
+        0,
+        OPEN_EXISTING,
+        FILE_FLAG_OPEN_REPARSE_POINT|FILE_FLAG_BACKUP_SEMANTICS,
+        0);
+    Py_END_ALLOW_THREADS
+    
+    if (reparse_point_handle==INVALID_HANDLE_VALUE)
+    {
+        return win32_error_unicode("readlink", path);
+    }
+    
+    Py_BEGIN_ALLOW_THREADS
+    /* New call DeviceIoControl to read the reparse point */
+    io_result = DeviceIoControl(
+        reparse_point_handle,
+        FSCTL_GET_REPARSE_POINT,
+        0, 0, /* in buffer */
+        target_buffer, sizeof(target_buffer),
+        &n_bytes_returned,
+        0 /* we're not using OVERLAPPED_IO */
+        );
+    CloseHandle(reparse_point_handle);
+    Py_END_ALLOW_THREADS
+
+    if (io_result==0)
+    {
+        return win32_error_unicode("readlink", path);
+    }
+
+    if (rdb->ReparseTag != IO_REPARSE_TAG_SYMLINK)
+    {
+        PyErr_SetString(PyExc_ValueError,
+                "not a symbolic link");
+        return NULL;
+    }
+    print_name = rdb->SymbolicLinkReparseBuffer.PathBuffer + rdb->SymbolicLinkReparseBuffer.PrintNameOffset;
+    result = PyUnicode_FromWideChar(print_name, rdb->SymbolicLinkReparseBuffer.PrintNameLength/2);
+    return result;
+}
+
+#endif /* !defined(HAVE_READLINK) && defined(MS_WINDOWS) */
+
+#if !defined(HAVE_SYMLINK) && defined(MS_WINDOWS)
+
+/* Grab CreateSymbolicLinkW dynamically from kernel32 */
+static int has_CreateSymbolicLinkW = 0;
+static DWORD (CALLBACK *Py_CreateSymbolicLinkW)(LPWSTR, LPWSTR, DWORD);
+static int
+check_CreateSymbolicLinkW()
+{
+    HINSTANCE hKernel32;
+    /* only recheck */
+    if (has_CreateSymbolicLinkW)
+        return has_CreateSymbolicLinkW;
+    hKernel32 = GetModuleHandle("KERNEL32");
+    *(FARPROC*)&Py_CreateSymbolicLinkW = GetProcAddress(hKernel32, "CreateSymbolicLinkW");
+    if (Py_CreateSymbolicLinkW)
+        has_CreateSymbolicLinkW = 1;
+    return has_CreateSymbolicLinkW;
+}
+
+PyDoc_STRVAR(win_symlink__doc__,
+"symlink(src, dst, target_is_directory=False)\n\n\
+Create a symbolic link pointing to src named dst.\n\
+target_is_directory is required if the target is to be interpreted as\n\
+a directory.\n\
+This function requires Windows 6.0 or greater, and raises a\n\
+NotImplementedError otherwise.");
+
+static PyObject *
+win_symlink(PyObject *self, PyObject *args, PyObject *kwargs)
+{
+    static char *kwlist[] = {"src", "dest", "target_is_directory", NULL};
+    PyObject *src, *dest;
+    int target_is_directory = 0;
+    DWORD res;
+    WIN32_FILE_ATTRIBUTE_DATA src_info;
+    
+    if (!check_CreateSymbolicLinkW())
+    {
+        /* raise NotImplementedError */
+        return PyErr_Format(PyExc_NotImplementedError,
+            "CreateSymbolicLinkW not found");
+    }
+    if (!PyArg_ParseTupleAndKeywords(args, kwargs, "OO|i:symlink",
+        kwlist, &src, &dest, &target_is_directory))
+        return NULL;
+    if (!convert_to_unicode(&src)) { return NULL; }
+    if (!convert_to_unicode(&dest)) {
+        Py_DECREF(src);
+        return NULL;
+    }
+    
+    /* if src is a directory, ensure target_is_directory==1 */
+    if(
+        GetFileAttributesExW(
+            PyUnicode_AsUnicode(src), GetFileExInfoStandard, &src_info
+        ))
+    {
+        target_is_directory = target_is_directory ||
+            (src_info.dwFileAttributes & FILE_ATTRIBUTE_DIRECTORY);
+    }
+
+    Py_BEGIN_ALLOW_THREADS
+    res = Py_CreateSymbolicLinkW(
+        PyUnicode_AsUnicode(dest),
+        PyUnicode_AsUnicode(src),
+        target_is_directory);
+    Py_END_ALLOW_THREADS
+    Py_DECREF(src);
+    Py_DECREF(dest);
+    if (!res)
+    {
+        return win32_error_unicode("symlink", PyUnicode_AsUnicode(src));
+    }
+    
+    Py_INCREF(Py_None);
+    return Py_None;
+}
+#endif /* !defined(HAVE_SYMLINK) && defined(MS_WINDOWS) */
 
 #ifdef HAVE_TIMES
 #if defined(PYCC_VACPP) && defined(PYOS_OS2)
@@ -7076,13 +7575,19 @@
 #ifdef HAVE_READLINK
     {"readlink",        posix_readlink, METH_VARARGS, posix_readlink__doc__},
 #endif /* HAVE_READLINK */
-    {"rename",          posix_rename, METH_VARARGS, posix_rename__doc__},
-    {"rmdir",           posix_rmdir, METH_VARARGS, posix_rmdir__doc__},
-    {"stat",            posix_stat, METH_VARARGS, posix_stat__doc__},
+#if !defined(HAVE_READLINK) && defined(MS_WINDOWS)
+    {"readlink",	win_readlink, METH_VARARGS, win_readlink__doc__},
+#endif /* !defined(HAVE_READLINK) && defined(MS_WINDOWS) */
+    {"rename",	posix_rename, METH_VARARGS, posix_rename__doc__},
+    {"rmdir",	posix_rmdir, METH_VARARGS, posix_rmdir__doc__},
+    {"stat",	posix_stat, METH_VARARGS, posix_stat__doc__},
     {"stat_float_times", stat_float_times, METH_VARARGS, stat_float_times__doc__},
 #ifdef HAVE_SYMLINK
     {"symlink",         posix_symlink, METH_VARARGS, posix_symlink__doc__},
 #endif /* HAVE_SYMLINK */
+#if !defined(HAVE_SYMLINK) && defined(MS_WINDOWS)
+    {"symlink", (PyCFunction)win_symlink, METH_VARARGS | METH_KEYWORDS, win_symlink__doc__},
+#endif /* !defined(HAVE_SYMLINK) && defined(MS_WINDOWS) */
 #ifdef HAVE_SYSTEM
     {"system",          posix_system, METH_VARARGS, posix_system__doc__},
 #endif
@@ -7307,6 +7812,7 @@
     {"abort",           posix_abort, METH_NOARGS, posix_abort__doc__},
 #ifdef MS_WINDOWS
     {"_getfullpathname",        posix__getfullpathname, METH_VARARGS, NULL},
+    {"_getfinalpathname",       posix__getfinalpathname, METH_VARARGS, NULL},
 #endif
 #ifdef HAVE_GETLOADAVG
     {"getloadavg",      posix_getloadavg, METH_NOARGS, posix_getloadavg__doc__},

From python-checkins at python.org  Fri Jul  9 00:13:25 2010
From: python-checkins at python.org (brian.curtin)
Date: Fri,  9 Jul 2010 00:13:25 +0200 (CEST)
Subject: [Python-checkins] r82661 - python/branches/py3k/Lib/test/test_sys.py
Message-ID: <20100708221325.AD9DBEEC46@mail.python.org>

Author: brian.curtin
Date: Fri Jul  9 00:13:25 2010
New Revision: 82661

Log:
Undo inadvertant checkin.


Modified:
   python/branches/py3k/Lib/test/test_sys.py

Modified: python/branches/py3k/Lib/test/test_sys.py
==============================================================================
--- python/branches/py3k/Lib/test/test_sys.py	(original)
+++ python/branches/py3k/Lib/test/test_sys.py	Fri Jul  9 00:13:25 2010
@@ -263,7 +263,7 @@
         # Raise SkipTest if sys doesn't have getwindowsversion attribute
         test.support.get_attribute(sys, "getwindowsversion")
         v = sys.getwindowsversion()
-        self.assertEqual(len(v), 9)
+        self.assertEqual(len(v), 5)
         self.assertIsInstance(v[0], int)
         self.assertIsInstance(v[1], int)
         self.assertIsInstance(v[2], int)

From python-checkins at python.org  Fri Jul  9 00:16:05 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Fri,  9 Jul 2010 00:16:05 +0200 (CEST)
Subject: [Python-checkins] r82662 -
	python/branches/py3k/Lib/test/test_structseq.py
Message-ID: <20100708221605.89A3BEEBA0@mail.python.org>

Author: benjamin.peterson
Date: Fri Jul  9 00:16:05 2010
New Revision: 82662

Log:
assert tuple inheritance

Modified:
   python/branches/py3k/Lib/test/test_structseq.py

Modified: python/branches/py3k/Lib/test/test_structseq.py
==============================================================================
--- python/branches/py3k/Lib/test/test_structseq.py	(original)
+++ python/branches/py3k/Lib/test/test_structseq.py	Fri Jul  9 00:16:05 2010
@@ -7,6 +7,7 @@
 
     def test_tuple(self):
         t = time.gmtime()
+        assert isinstance(t, tuple)
         astuple = tuple(t)
         self.assertEqual(len(t), len(astuple))
         self.assertEqual(t, astuple)

From python-checkins at python.org  Fri Jul  9 00:33:03 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Fri,  9 Jul 2010 00:33:03 +0200 (CEST)
Subject: [Python-checkins] r82663 - in python/branches/py3k:
	Lib/test/test_structseq.py Objects/structseq.c
Message-ID: <20100708223303.E17E7EEAF6@mail.python.org>

Author: benjamin.peterson
Date: Fri Jul  9 00:33:03 2010
New Revision: 82663

Log:
fix repr of complicated structseqs #9206

Modified:
   python/branches/py3k/Lib/test/test_structseq.py
   python/branches/py3k/Objects/structseq.c

Modified: python/branches/py3k/Lib/test/test_structseq.py
==============================================================================
--- python/branches/py3k/Lib/test/test_structseq.py	(original)
+++ python/branches/py3k/Lib/test/test_structseq.py	Fri Jul  9 00:33:03 2010
@@ -1,7 +1,8 @@
+import os
+import time
 import unittest
 from test import support
 
-import time
 
 class StructSeqTest(unittest.TestCase):
 
@@ -34,6 +35,13 @@
         self.assertEqual(repr(t),
             "time.struct_time(tm_year=1970, tm_mon=1, tm_mday=1, tm_hour=0, "
             "tm_min=0, tm_sec=0, tm_wday=3, tm_yday=1, tm_isdst=0)")
+        # os.stat() gives a complicated struct sequence.
+        st = os.stat(__file__)
+        rep = repr(st)
+        self.assertTrue(rep.startswith("posix.stat_result"))
+        self.assertIn("st_mode=", rep)
+        self.assertIn("st_ino=", rep)
+        self.assertIn("st_dev=", rep)
 
     def test_concat(self):
         t1 = time.gmtime()

Modified: python/branches/py3k/Objects/structseq.c
==============================================================================
--- python/branches/py3k/Objects/structseq.c	(original)
+++ python/branches/py3k/Objects/structseq.c	Fri Jul  9 00:33:03 2010
@@ -35,12 +35,27 @@
     obj = PyObject_GC_NewVar(PyStructSequence, type, size);
     if (obj == NULL)
         return NULL;
+    /* Hack the size of the variable object, so invisible fields don't appear
+     to Python code. */
+    Py_SIZE(obj) = VISIBLE_SIZE_TP(type);
     for (i = 0; i < size; i++)
         obj->ob_item[i] = NULL;
 
     return (PyObject*)obj;
 }
 
+static void
+structseq_dealloc(PyStructSequence *obj)
+{
+    Py_ssize_t i, size;
+    
+    size = REAL_SIZE(obj);
+    for (i = 0; i < size; ++i) {
+        Py_XDECREF(obj->ob_item[i]);
+    }
+    PyObject_GC_Del(obj);
+}
+
 static PyObject *
 structseq_new(PyTypeObject *type, PyObject *args, PyObject *kwds)
 {
@@ -154,8 +169,11 @@
         char *cname, *crepr;
 
         cname = typ->tp_members[i].name;
-        if (cname == NULL)
+        if (cname == NULL) {
+            PyErr_Format(PyExc_SystemError, "In structseq_repr(), member %d name is NULL"
+                         " for type %.500s", i, typ->tp_name);
             return NULL;
+        }
         val = PyStructSequence_GET_ITEM(obj, i);
         repr = PyObject_Repr(val);
         if (repr == NULL)
@@ -249,7 +267,7 @@
     NULL,                                       /* tp_name */
     sizeof(PyStructSequence) - sizeof(PyObject *), /* tp_basicsize */
     sizeof(PyObject *),                         /* tp_itemsize */
-    0,                                          /* tp_dealloc */
+    (destructor)structseq_dealloc,              /* tp_dealloc */
     0,                                          /* tp_print */
     0,                                          /* tp_getattr */
     0,                                          /* tp_setattr */

From python-checkins at python.org  Fri Jul  9 01:13:57 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:13:57 +0200 (CEST)
Subject: [Python-checkins] r82664 - python/branches/import_unicode
Message-ID: <20100708231357.24E90EE987@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:13:57 2010
New Revision: 82664

Log:
create import_unicode branch, based on py3k


Added:
   python/branches/import_unicode/
      - copied from r82663, /python/branches/py3k/

From solipsis at pitrou.net  Fri Jul  9 01:25:51 2010
From: solipsis at pitrou.net (solipsis at pitrou.net)
Date: Fri,  9 Jul 2010 01:25:51 +0200 (CEST)
Subject: [Python-checkins] Daily py3k reference leaks (r82650): sum=0
Message-ID: <20100708232551.CA56E1775C@ns6635.ovh.net>

py3k results for svn r82650 (hg cset 6f69451a1dd0)
--------------------------------------------------



Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/antoine/py3k/refleaks/reflogoVdiKd', '-x']

From python-checkins at python.org  Fri Jul  9 01:32:12 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:12 +0200 (CEST)
Subject: [Python-checkins] r82665 - in python/branches/import_unicode:
	Include/fileobject.h Objects/fileobject.c Python/import.c
Message-ID: <20100708233212.CE5C7EE995@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:12 2010
New Revision: 82665

Log:
Create _PyFile_FromFdUnicode() private function

Modified:
   python/branches/import_unicode/Include/fileobject.h
   python/branches/import_unicode/Objects/fileobject.c
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Include/fileobject.h
==============================================================================
--- python/branches/import_unicode/Include/fileobject.h	(original)
+++ python/branches/import_unicode/Include/fileobject.h	Fri Jul  9 01:32:12 2010
@@ -8,8 +8,24 @@
 
 #define PY_STDIOTEXTMODE "b"
 
-PyAPI_FUNC(PyObject *) PyFile_FromFd(int, char *, char *, int, char *, char *,
-				     char *, int);
+PyAPI_FUNC(PyObject *) PyFile_FromFd(
+    int fd,
+    char *name,
+    char *mode,
+    int buffering,
+    char *encoding,
+    char *errors,
+    char *newline,
+    int closefd);
+PyAPI_FUNC(PyObject *) _PyFile_FromFdUnicode(
+    int fd,
+    PyObject *name,
+    char *mode,
+    int buffering,
+    char *encoding,
+    char *errors,
+    char *newline,
+    int closefd);
 PyAPI_FUNC(PyObject *) PyFile_GetLine(PyObject *, int);
 PyAPI_FUNC(int) PyFile_WriteObject(PyObject *, PyObject *, int);
 PyAPI_FUNC(int) PyFile_WriteString(const char *, PyObject *);

Modified: python/branches/import_unicode/Objects/fileobject.c
==============================================================================
--- python/branches/import_unicode/Objects/fileobject.c	(original)
+++ python/branches/import_unicode/Objects/fileobject.c	Fri Jul  9 01:32:12 2010
@@ -26,10 +26,10 @@
 /* External C interface */
 
 PyObject *
-PyFile_FromFd(int fd, char *name, char *mode, int buffering, char *encoding,
-              char *errors, char *newline, int closefd)
+_PyFile_FromFdUnicode(int fd, PyObject *name, char *mode, int buffering,
+                      char *encoding, char *errors, char *newline, int closefd)
 {
-    PyObject *io, *stream, *nameobj = NULL;
+    PyObject *io, *stream;
 
     io = PyImport_ImportModule("io");
     if (io == NULL)
@@ -41,18 +41,30 @@
     if (stream == NULL)
         return NULL;
     if (name != NULL) {
-        nameobj = PyUnicode_DecodeFSDefault(name);
-        if (nameobj == NULL)
+        if (PyObject_SetAttrString(stream, "name", name) < 0)
             PyErr_Clear();
-        else {
-            if (PyObject_SetAttrString(stream, "name", nameobj) < 0)
-                PyErr_Clear();
-            Py_DECREF(nameobj);
-        }
     }
     return stream;
 }
 
+
+PyObject *
+PyFile_FromFd(int fd, char *name, char *mode, int buffering, char *encoding,
+              char *errors, char *newline, int closefd)
+{
+    PyObject *nameobj, *file;
+    if (name != NULL) {
+        nameobj = PyUnicode_DecodeFSDefault(name);
+        if (nameobj == NULL)
+            PyErr_Clear();
+    } else
+        nameobj = NULL;
+    file = _PyFile_FromFdUnicode(fd, nameobj, mode, buffering,
+                                 encoding, errors, newline, closefd);
+    Py_XDECREF(nameobj);
+    return file;
+}
+
 PyObject *
 PyFile_GetLine(PyObject *f, int n)
 {

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:12 2010
@@ -3121,6 +3121,9 @@
     fdp = find_module(NULL, name, path, pathname, MAXPATHLEN+1, &fp, NULL);
     if (fdp == NULL)
         return NULL;
+    pathobj = PyUnicode_DecodeFSDefault(pathname);
+    if (pathobj == NULL)
+        return NULL;
     if (fp != NULL) {
         fd = fileno(fp);
         if (fd != -1)
@@ -3139,8 +3142,8 @@
             encoding = (found_encoding != NULL) ? found_encoding :
                    (char*)PyUnicode_GetDefaultEncoding();
         }
-        fob = PyFile_FromFd(fd, pathname, fdp->mode, -1,
-                            (char*)encoding, NULL, NULL, 1);
+        fob = _PyFile_FromFdUnicode(fd, pathobj, fdp->mode, -1,
+                                    encoding, NULL, NULL, 1);
         if (fob == NULL) {
             close(fd);
             PyMem_FREE(found_encoding);
@@ -3151,9 +3154,9 @@
         fob = Py_None;
         Py_INCREF(fob);
     }
-    pathobj = PyUnicode_DecodeFSDefault(pathname);
     ret = Py_BuildValue("NN(ssi)",
-                  fob, pathobj, fdp->suffix, fdp->mode, fdp->type);
+                        fob, pathobj,
+                        fdp->suffix, fdp->mode, fdp->type);
     PyMem_FREE(found_encoding);
 
     return ret;

From python-checkins at python.org  Fri Jul  9 01:32:14 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:14 +0200 (CEST)
Subject: [Python-checkins] r82666 -
	python/branches/import_unicode/Python/pythonrun.c
Message-ID: <20100708233214.4C449EE9BC@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:14 2010
New Revision: 82666

Log:
initialize GIL earlier in Py_InitializeEx()

Modified:
   python/branches/import_unicode/Python/pythonrun.c

Modified: python/branches/import_unicode/Python/pythonrun.c
==============================================================================
--- python/branches/import_unicode/Python/pythonrun.c	(original)
+++ python/branches/import_unicode/Python/pythonrun.c	Fri Jul  9 01:32:14 2010
@@ -206,6 +206,11 @@
         Py_FatalError("Py_Initialize: can't make first thread");
     (void) PyThreadState_Swap(tstate);
 
+    /* auto-thread-state API, if available */
+#ifdef WITH_THREAD
+    _PyGILState_Init(interp, tstate);
+#endif /* WITH_THREAD */
+
     _Py_ReadyTypes();
 
     if (!_PyFrame_Init())
@@ -286,11 +291,6 @@
         Py_FatalError(
             "Py_Initialize: can't initialize sys standard streams");
 
-    /* auto-thread-state API, if available */
-#ifdef WITH_THREAD
-    _PyGILState_Init(interp, tstate);
-#endif /* WITH_THREAD */
-
     if (!Py_NoSiteFlag)
         initsite(); /* Module site */
 }

From python-checkins at python.org  Fri Jul  9 01:32:15 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:15 +0200 (CEST)
Subject: [Python-checkins] r82667 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233215.AA280EEB20@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:15 2010
New Revision: 82667

Log:
update_compiled_module() pathname is an object

not a char*

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:15 2010
@@ -1218,24 +1218,17 @@
 }
 
 static int
-update_compiled_module(PyCodeObject *co, char *pathname)
+update_compiled_module(PyCodeObject *co, PyObject *newname)
 {
-    PyObject *oldname, *newname;
+    PyObject *oldname;
 
-    newname = PyUnicode_DecodeFSDefault(pathname);
-    if (newname == NULL)
-        return -1;
-
-    if (!PyUnicode_Compare(co->co_filename, newname)) {
-        Py_DECREF(newname);
+    if (!PyUnicode_Compare(co->co_filename, newname))
         return 0;
-    }
 
     oldname = co->co_filename;
     Py_INCREF(oldname);
     update_code_filenames(co, oldname, newname);
     Py_DECREF(oldname);
-    Py_DECREF(newname);
     return 1;
 }
 
@@ -1252,6 +1245,7 @@
     char *cpathname;
     PyCodeObject *co;
     PyObject *m;
+    PyObject *pathobj;
 
     if (fstat(fileno(fp), &st) != 0) {
         PyErr_Format(PyExc_RuntimeError,
@@ -1278,8 +1272,12 @@
         fclose(fpc);
         if (co == NULL)
             return NULL;
-        if (update_compiled_module(co, pathname) < 0)
+        pathobj = PyUnicode_DecodeFSDefault(pathname);
+        if (pathobj == NULL)
+            return NULL;
+        if (update_compiled_module(co, pathobj) < 0)
             return NULL;
+        Py_DECREF(pathobj);
         if (Py_VerboseFlag)
             PySys_WriteStderr("import %s # precompiled from %s\n",
                 name, cpathname);

From python-checkins at python.org  Fri Jul  9 01:32:17 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:17 +0200 (CEST)
Subject: [Python-checkins] r82668 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233217.1B1E9EEB49@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:16 2010
New Revision: 82668

Log:
load_source_module(): path is now an object

instead of a char*

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:16 2010
@@ -1237,15 +1237,18 @@
    byte-compiled file, use that instead. */
 
 static PyObject *
-load_source_module(char *name, char *pathname, FILE *fp)
+load_source_module(char *name, PyObject *pathobj, FILE *fp)
 {
     struct stat st;
     FILE *fpc;
     char buf[MAXPATHLEN+1];
+    char *pathname;
     char *cpathname;
     PyCodeObject *co;
     PyObject *m;
-    PyObject *pathobj;
+
+    /* FIXME: don't use _PyUnicode_AsString */
+    pathname = _PyUnicode_AsString(pathobj);
 
     if (fstat(fileno(fp), &st) != 0) {
         PyErr_Format(PyExc_RuntimeError,
@@ -1266,21 +1269,21 @@
 #endif
     cpathname = make_compiled_pathname(
         pathname, buf, (size_t)MAXPATHLEN + 1, !Py_OptimizeFlag);
-    if (cpathname != NULL &&
-        (fpc = check_compiled_module(pathname, st.st_mtime, cpathname))) {
+    if (cpathname != NULL)
+        fpc = check_compiled_module(pathname, st.st_mtime, cpathname);
+    else
+        fpc = NULL;
+    if (fpc) {
         co = read_compiled_module(cpathname, fpc);
         fclose(fpc);
         if (co == NULL)
             return NULL;
-        pathobj = PyUnicode_DecodeFSDefault(pathname);
-        if (pathobj == NULL)
-            return NULL;
         if (update_compiled_module(co, pathobj) < 0)
             return NULL;
-        Py_DECREF(pathobj);
         if (Py_VerboseFlag)
             PySys_WriteStderr("import %s # precompiled from %s\n",
                 name, cpathname);
+        pathobj = NULL;
         pathname = cpathname;
     }
     else {
@@ -2035,9 +2038,14 @@
 
     switch (type) {
 
-    case PY_SOURCE:
-        m = load_source_module(name, pathname, fp);
+    case PY_SOURCE: {
+        PyObject *pathobj = PyUnicode_DecodeFSDefault(pathname);
+        if (pathobj == NULL)
+            return NULL;
+        m = load_source_module(name, pathobj, fp);
+        Py_DECREF(pathobj);
         break;
+    }
 
     case PY_COMPILED:
         m = load_compiled_module(name, pathname, fp);
@@ -3343,22 +3351,18 @@
 imp_load_source(PyObject *self, PyObject *args)
 {
     char *name;
-    char *pathname;
+    PyObject *pathname;
     PyObject *fob = NULL;
     PyObject *m;
     FILE *fp;
-    if (!PyArg_ParseTuple(args, "ses|O:load_source",
-                          &name,
-                          Py_FileSystemDefaultEncoding, &pathname,
-                          &fob))
+    if (!PyArg_ParseTuple(args, "sU|O:load_source",
+                          &name, &pathname, &fob))
         return NULL;
-    fp = get_file(pathname, fob, "r");
-    if (fp == NULL) {
-        PyMem_Free(pathname);
+                  /* FIXME: don't use _PyUnicode_AsString */
+    fp = get_file(_PyUnicode_AsString(pathname), fob, "r");
+    if (fp == NULL)
         return NULL;
-    }
     m = load_source_module(name, pathname, fp);
-    PyMem_Free(pathname);
     fclose(fp);
     return m;
 }

From python-checkins at python.org  Fri Jul  9 01:32:18 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:18 +0200 (CEST)
Subject: [Python-checkins] r82669 - in python/branches/import_unicode:
	Include/import.h Python/import.c Python/importdl.c Python/importdl.h
Message-ID: <20100708233218.AC151EEB35@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:18 2010
New Revision: 82669

Log:
_PyImport_LoadDynamicModule() path is an object

... instead of a char*

 * Create _PyImport_FindExtensionUnicode() and
   _PyImport_FixupExtensionUnicode() functions

Modified:
   python/branches/import_unicode/Include/import.h
   python/branches/import_unicode/Python/import.c
   python/branches/import_unicode/Python/importdl.c
   python/branches/import_unicode/Python/importdl.h

Modified: python/branches/import_unicode/Include/import.h
==============================================================================
--- python/branches/import_unicode/Include/import.h	(original)
+++ python/branches/import_unicode/Include/import.h	Fri Jul  9 01:32:18 2010
@@ -44,7 +44,9 @@
 PyAPI_FUNC(void) _PyImport_ReInitLock(void);
 
 PyAPI_FUNC(PyObject *)_PyImport_FindExtension(char *, char *);
+PyAPI_FUNC(PyObject *)_PyImport_FindExtensionUnicode(char *, PyObject *);
 PyAPI_FUNC(int)_PyImport_FixupExtension(PyObject*, char *, char *);
+PyAPI_FUNC(int)_PyImport_FixupExtensionUnicode(PyObject*, char *, PyObject *);
 
 struct _inittab {
     char *name;

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:18 2010
@@ -565,7 +565,7 @@
 */
 
 int
-_PyImport_FixupExtension(PyObject *mod, char *name, char *filename)
+_PyImport_FixupExtensionUnicode(PyObject *mod, char *name, PyObject *filename)
 {
     PyObject *modules, *dict;
     struct PyModuleDef *def;
@@ -605,18 +605,32 @@
         if (def->m_base.m_copy == NULL)
             return -1;
     }
-    PyDict_SetItemString(extensions, filename, (PyObject*)def);
+    PyDict_SetItem(extensions, filename, (PyObject*)def);
     return 0;
 }
 
+int
+_PyImport_FixupExtension(PyObject *mod, char *name, char *filename)
+{
+    PyObject *fileobj;
+    int result;
+    fileobj = PyUnicode_FromString(filename);
+    if (fileobj == NULL)
+        return -1;
+    result = _PyImport_FixupExtensionUnicode(mod, name, fileobj);
+    Py_DECREF(fileobj);
+    return result;
+}
+
+
 PyObject *
-_PyImport_FindExtension(char *name, char *filename)
+_PyImport_FindExtensionUnicode(char *name, PyObject *filename)
 {
     PyObject *mod, *mdict;
     PyModuleDef* def;
     if (extensions == NULL)
         return NULL;
-    def = (PyModuleDef*)PyDict_GetItemString(extensions, filename);
+    def = (PyModuleDef*)PyDict_GetItem(extensions, filename);
     if (def == NULL)
         return NULL;
     if (def->m_size == -1) {
@@ -648,9 +662,20 @@
     }
     if (Py_VerboseFlag)
         PySys_WriteStderr("import %s # previously loaded (%s)\n",
-                          name, filename);
+                          name,
+                          /* FIXME: don't use _PyUnicode_AsString */
+                          _PyUnicode_AsString(filename));
     return mod;
+}
 
+PyObject *
+_PyImport_FindExtension(char *name, char *filename)
+{
+    PyObject *fileobj, *mod;
+    fileobj = PyUnicode_DecodeFSDefault(filename);
+    mod = _PyImport_FindExtensionUnicode(name, fileobj);
+    Py_DECREF(fileobj);
+    return mod;
 }
 
 
@@ -2052,9 +2077,14 @@
         break;
 
 #ifdef HAVE_DYNAMIC_LOADING
-    case C_EXTENSION:
-        m = _PyImport_LoadDynamicModule(name, pathname, fp);
+    case C_EXTENSION: {
+        PyObject *pathobj = PyUnicode_DecodeFSDefault(pathname);
+        if (pathobj == NULL)
+            return NULL;
+        m = _PyImport_LoadDynamicModule(name, pathobj, fp);
+        Py_DECREF(pathobj);
         break;
+    }
 #endif
 
     case PKG_DIRECTORY:
@@ -2123,9 +2153,13 @@
 init_builtin(char *name)
 {
     struct _inittab *p;
+    PyObject *path;
 
-    if (_PyImport_FindExtension(name, name) != NULL)
+    path = PyUnicode_FromString(name);
+    if (_PyImport_FindExtensionUnicode(name, path) != NULL) {
+        Py_DECREF(path);
         return 1;
+    }
 
     for (p = PyImport_Inittab; p->name != NULL; p++) {
         PyObject *mod;
@@ -2134,21 +2168,28 @@
                 PyErr_Format(PyExc_ImportError,
                     "Cannot re-init internal module %.200s",
                     name);
+                Py_DECREF(path);
                 return -1;
             }
             if (Py_VerboseFlag)
                 PySys_WriteStderr("import %s # builtin\n", name);
             mod = (*p->initfunc)();
-            if (mod == 0)
+            if (mod == 0) {
+                Py_DECREF(path);
                 return -1;
-            if (_PyImport_FixupExtension(mod, name, name) < 0)
+            }
+            if (_PyImport_FixupExtensionUnicode(mod, name, path) < 0) {
+                Py_DECREF(path);
                 return -1;
+            }
+            Py_DECREF(path);
             /* FixupExtension has put the module into sys.modules,
                so we can release our own reference. */
             Py_DECREF(mod);
             return 1;
         }
     }
+    Py_DECREF(path);
     return 0;
 }
 
@@ -3322,24 +3363,22 @@
 imp_load_dynamic(PyObject *self, PyObject *args)
 {
     char *name;
-    char *pathname;
+    PyObject *pathname;
     PyObject *fob = NULL;
     PyObject *m;
     FILE *fp = NULL;
-    if (!PyArg_ParseTuple(args, "ses|O:load_dynamic",
+    if (!PyArg_ParseTuple(args, "sU|O:load_dynamic",
                           &name,
-                          Py_FileSystemDefaultEncoding, &pathname,
+                          &pathname,
                           &fob))
         return NULL;
     if (fob) {
-        fp = get_file(pathname, fob, "r");
-        if (fp == NULL) {
-            PyMem_Free(pathname);
+                      /* FIXME: don't use _PyUnicode_AsString */
+        fp = get_file(_PyUnicode_AsString(pathname), fob, "r");
+        if (fp == NULL)
             return NULL;
-        }
     }
     m = _PyImport_LoadDynamicModule(name, pathname, fp);
-    PyMem_Free(pathname);
     if (fp)
         fclose(fp);
     return m;

Modified: python/branches/import_unicode/Python/importdl.c
==============================================================================
--- python/branches/import_unicode/Python/importdl.c	(original)
+++ python/branches/import_unicode/Python/importdl.c	Fri Jul  9 01:32:18 2010
@@ -19,16 +19,19 @@
 
 
 PyObject *
-_PyImport_LoadDynamicModule(char *name, char *pathname, FILE *fp)
+_PyImport_LoadDynamicModule(char *name, PyObject *path, FILE *fp)
 {
     PyObject *m;
-    PyObject *path;
+    char *pathname;
     char *lastdot, *shortname, *packagecontext, *oldcontext;
     dl_funcptr p0;
     PyObject* (*p)(void);
     struct PyModuleDef *def;
 
-    if ((m = _PyImport_FindExtension(name, pathname)) != NULL) {
+    /* FIXME: don't use _PyUnicode_AsString */
+    pathname = _PyUnicode_AsString(path);
+
+    if ((m = _PyImport_FindExtensionUnicode(name, path)) != NULL) {
         Py_INCREF(m);
         return m;
     }
@@ -72,11 +75,13 @@
     def->m_base.m_init = p;
 
     /* Remember the filename as the __file__ attribute */
-    path = PyUnicode_DecodeFSDefault(pathname);
-    if (PyModule_AddObject(m, "__file__", path) < 0)
+    Py_INCREF(path);
+    if (PyModule_AddObject(m, "__file__", path) < 0) {
         PyErr_Clear(); /* Not important enough to report */
+        Py_DECREF(path);
+    }
 
-    if (_PyImport_FixupExtension(m, name, pathname) < 0)
+    if (_PyImport_FixupExtensionUnicode(m, name, path) < 0)
         return NULL;
     if (Py_VerboseFlag)
         PySys_WriteStderr(

Modified: python/branches/import_unicode/Python/importdl.h
==============================================================================
--- python/branches/import_unicode/Python/importdl.h	(original)
+++ python/branches/import_unicode/Python/importdl.h	Fri Jul  9 01:32:18 2010
@@ -28,7 +28,7 @@
 extern struct filedescr * _PyImport_Filetab;
 extern const struct filedescr _PyImport_DynLoadFiletab[];
 
-extern PyObject *_PyImport_LoadDynamicModule(char *name, char *pathname,
+extern PyObject *_PyImport_LoadDynamicModule(char *name, PyObject *pathname,
                                              FILE *);
 
 /* Max length of module suffix searched for -- accommodates "module.slb" */

From python-checkins at python.org  Fri Jul  9 01:32:20 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:20 +0200 (CEST)
Subject: [Python-checkins] r82670 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233220.25D31EEAF0@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:20 2010
New Revision: 82670

Log:
load_package() path is an object

instead of a char*

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:20 2010
@@ -1381,7 +1381,7 @@
    REFERENCE COUNT */
 
 static PyObject *
-load_package(char *name, char *pathname)
+load_package(char *name, PyObject *pathobj)
 {
     PyObject *m, *d;
     PyObject *file = NULL;
@@ -1390,6 +1390,10 @@
     char buf[MAXPATHLEN+1];
     FILE *fp = NULL;
     struct filedescr *fdp;
+    char *pathname;
+
+    /* FIXME: don't use _PyUnicode_AsString */
+    pathname = _PyUnicode_AsString(pathobj);
 
     m = PyImport_AddModule(name);
     if (m == NULL)
@@ -2087,9 +2091,14 @@
     }
 #endif
 
-    case PKG_DIRECTORY:
-        m = load_package(name, pathname);
+    case PKG_DIRECTORY: {
+        PyObject *pathobj = PyUnicode_DecodeFSDefault(pathname);
+        if (pathobj == NULL)
+            return NULL;
+        m = load_package(name, pathobj);
+        Py_DECREF(pathobj);
         break;
+    }
 
     case C_BUILTIN:
     case PY_FROZEN:
@@ -3455,14 +3464,11 @@
 imp_load_package(PyObject *self, PyObject *args)
 {
     char *name;
-    char *pathname;
-    PyObject * ret;
-    if (!PyArg_ParseTuple(args, "ses:load_package",
-                          &name, Py_FileSystemDefaultEncoding, &pathname))
+    PyObject *pathname;
+    if (!PyArg_ParseTuple(args, "sU:load_package",
+                          &name, &pathname))
         return NULL;
-    ret = load_package(name, pathname);
-    PyMem_Free(pathname);
-    return ret;
+    return load_package(name, pathname);
 }
 
 static PyObject *

From python-checkins at python.org  Fri Jul  9 01:32:21 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:21 +0200 (CEST)
Subject: [Python-checkins] r82671 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233221.9BF52EEAF3@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:21 2010
New Revision: 82671

Log:
get_sourcefile(): path argument is an object

instead of char*

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:21 2010
@@ -718,7 +718,7 @@
                       "sys.modules failed");
 }
 
-static PyObject * get_sourcefile(char *file);
+static PyObject * get_sourcefile(PyObject *file);
 static char *make_source_pathname(char *pathname, char *buf);
 static char *make_compiled_pathname(char *pathname, char *buf, size_t buflen,
                                     int debug);
@@ -768,7 +768,10 @@
     /* Remember the filename as the __file__ attribute */
     v = NULL;
     if (pathname != NULL) {
-        v = get_sourcefile(pathname);
+        PyObject *pathobj = PyUnicode_DecodeFSDefault(pathname);
+        /* FIXME: check pathobj != NULL */
+        v = get_sourcefile(pathobj);
+        Py_DECREF(pathobj);
         if (v == NULL)
             PyErr_Clear();
     }
@@ -1335,12 +1338,20 @@
  * Returns the path to the py file if available, else the given path
  */
 static PyObject *
-get_sourcefile(char *file)
+get_sourcefile(PyObject *fileobj)
 {
+    /* FIXME: use Py_UNICODE* instead of char* */
     char py[MAXPATHLEN + 1];
     Py_ssize_t len;
-    PyObject *u;
     struct stat statbuf;
+    char *file;
+
+    if (!fileobj) {
+        Py_RETURN_NONE;
+    }
+
+    /* FIXME: don't use _PyUnicode_AsString */
+    file = _PyUnicode_AsString(fileobj);
 
     if (!file || !*file) {
         Py_RETURN_NONE;
@@ -1349,7 +1360,8 @@
     len = strlen(file);
     /* match '*.py?' */
     if (len > MAXPATHLEN || PyOS_strnicmp(&file[len-4], ".py", 3) != 0) {
-        return PyUnicode_DecodeFSDefault(file);
+        Py_INCREF(fileobj);
+        return fileobj;
     }
 
     /* Start by trying to turn PEP 3147 path into source path.  If that
@@ -1363,12 +1375,12 @@
 
     if (stat(py, &statbuf) == 0 &&
         S_ISREG(statbuf.st_mode)) {
-        u = PyUnicode_DecodeFSDefault(py);
+        return PyUnicode_DecodeFSDefault(py);
     }
     else {
-        u = PyUnicode_DecodeFSDefault(file);
+        Py_INCREF(fileobj);
+        return fileobj;
     }
-    return u;
 }
 
 /* Forward */
@@ -1402,7 +1414,7 @@
         PySys_WriteStderr("import %s # directory %s\n",
             name, pathname);
     d = PyModule_GetDict(m);
-    file = get_sourcefile(pathname);
+    file = get_sourcefile(pathobj);
     if (file == NULL)
         goto error;
     path = Py_BuildValue("[O]", file);

From python-checkins at python.org  Fri Jul  9 01:32:23 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:23 +0200 (CEST)
Subject: [Python-checkins] r82672 -
	python/branches/import_unicode/Include/import.h
Message-ID: <20100708233223.021C8EEBA6@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:22 2010
New Revision: 82672

Log:
replace tabs by spaces

Modified:
   python/branches/import_unicode/Include/import.h

Modified: python/branches/import_unicode/Include/import.h
==============================================================================
--- python/branches/import_unicode/Include/import.h	(original)
+++ python/branches/import_unicode/Include/import.h	Fri Jul  9 01:32:22 2010
@@ -11,18 +11,21 @@
 PyAPI_FUNC(const char *) PyImport_GetMagicTag(void);
 PyAPI_FUNC(PyObject *) PyImport_ExecCodeModule(char *name, PyObject *co);
 PyAPI_FUNC(PyObject *) PyImport_ExecCodeModuleEx(
-	char *name, PyObject *co, char *pathname);
+    char *name, PyObject *co, char *pathname);
 PyAPI_FUNC(PyObject *) PyImport_ExecCodeModuleWithPathnames(
-	char *name, PyObject *co, char *pathname, char *cpathname);
+    char *name,
+    PyObject *co,
+    char *pathname,
+    char *cpathname);
 PyAPI_FUNC(PyObject *) PyImport_GetModuleDict(void);
 PyAPI_FUNC(PyObject *) PyImport_AddModule(const char *name);
 PyAPI_FUNC(PyObject *) PyImport_ImportModule(const char *name);
 PyAPI_FUNC(PyObject *) PyImport_ImportModuleNoBlock(const char *);
 PyAPI_FUNC(PyObject *) PyImport_ImportModuleLevel(char *name,
-	PyObject *globals, PyObject *locals, PyObject *fromlist, int level);
+    PyObject *globals, PyObject *locals, PyObject *fromlist, int level);
 
 #define PyImport_ImportModuleEx(n, g, l, f) \
-	PyImport_ImportModuleLevel(n, g, l, f, -1)
+    PyImport_ImportModuleLevel(n, g, l, f, -1)
 
 PyAPI_FUNC(PyObject *) PyImport_GetImporter(PyObject *path);
 PyAPI_FUNC(PyObject *) PyImport_Import(PyObject *name);
@@ -39,7 +42,7 @@
 #endif
 
 PyAPI_FUNC(struct filedescr *) _PyImport_FindModule(
-	const char *, PyObject *, char *, size_t, FILE **, PyObject **);
+    const char *, PyObject *, char *, size_t, FILE **, PyObject **);
 PyAPI_FUNC(int) _PyImport_IsScript(struct filedescr *);
 PyAPI_FUNC(void) _PyImport_ReInitLock(void);
 

From python-checkins at python.org  Fri Jul  9 01:32:24 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:24 +0200 (CEST)
Subject: [Python-checkins] r82673 - in python/branches/import_unicode:
	Include/import.h Python/import.c
Message-ID: <20100708233224.8A02BEEB83@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:24 2010
New Revision: 82673

Log:
Create PyImport_ExecCodeModuleWithUnicodePathnames()

I love long function names!

Modified:
   python/branches/import_unicode/Include/import.h
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Include/import.h
==============================================================================
--- python/branches/import_unicode/Include/import.h	(original)
+++ python/branches/import_unicode/Include/import.h	Fri Jul  9 01:32:24 2010
@@ -17,6 +17,11 @@
     PyObject *co,
     char *pathname,
     char *cpathname);
+PyAPI_FUNC(PyObject *) PyImport_ExecCodeModuleWithUnicodePathnames(
+    char *name,
+    PyObject *co,
+    PyObject *pathname,
+    PyObject *cpathname);
 PyAPI_FUNC(PyObject *) PyImport_GetModuleDict(void);
 PyAPI_FUNC(PyObject *) PyImport_AddModule(const char *name);
 PyAPI_FUNC(PyObject *) PyImport_ImportModule(const char *name);

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:24 2010
@@ -736,8 +736,8 @@
 PyObject *
 PyImport_ExecCodeModule(char *name, PyObject *co)
 {
-    return PyImport_ExecCodeModuleWithPathnames(
-        name, co, (char *)NULL, (char *)NULL);
+    return PyImport_ExecCodeModuleWithUnicodePathnames(
+        name, co, NULL, NULL);
 }
 
 PyObject *
@@ -748,8 +748,8 @@
 }
 
 PyObject *
-PyImport_ExecCodeModuleWithPathnames(char *name, PyObject *co, char *pathname,
-                                     char *cpathname)
+PyImport_ExecCodeModuleWithUnicodePathnames(char *name, PyObject *co,
+                                            PyObject *pathobj, PyObject *cpathobj)
 {
     PyObject *modules = PyImport_GetModuleDict();
     PyObject *m, *d, *v;
@@ -767,11 +767,8 @@
     }
     /* Remember the filename as the __file__ attribute */
     v = NULL;
-    if (pathname != NULL) {
-        PyObject *pathobj = PyUnicode_DecodeFSDefault(pathname);
-        /* FIXME: check pathobj != NULL */
+    if (pathobj != NULL) {
         v = get_sourcefile(pathobj);
-        Py_DECREF(pathobj);
         if (v == NULL)
             PyErr_Clear();
     }
@@ -784,18 +781,12 @@
     Py_DECREF(v);
 
     /* Remember the pyc path name as the __cached__ attribute. */
-    if (cpathname == NULL) {
-        v = Py_None;
-        Py_INCREF(v);
-    }
-    else if ((v = PyUnicode_FromString(cpathname)) == NULL) {
-        PyErr_Clear(); /* Not important enough to report */
+    if (cpathobj != NULL)
+        v = cpathobj;
+    else
         v = Py_None;
-        Py_INCREF(v);
-    }
     if (PyDict_SetItemString(d, "__cached__", v) != 0)
         PyErr_Clear(); /* Not important enough to report */
-    Py_DECREF(v);
 
     v = PyEval_EvalCode((PyCodeObject *)co, d, d);
     if (v == NULL)
@@ -818,6 +809,29 @@
     return NULL;
 }
 
+PyObject *
+PyImport_ExecCodeModuleWithPathnames(char *name, PyObject *co, char *pathname,
+                                     char *cpathname)
+{
+    PyObject *m, *pathobj, *cpathobj;
+    if (pathname != NULL) {
+        pathobj = PyUnicode_DecodeFSDefault(pathname);
+        if (pathobj == NULL)
+            return NULL;
+    } else
+        pathobj = NULL;
+    if (cpathname != NULL) {
+        cpathobj = PyUnicode_DecodeFSDefault(cpathname);
+        if (cpathobj == NULL)
+            PyErr_Clear(); /* Not important enough to report */
+    } else
+        cpathobj = NULL;
+    m = PyImport_ExecCodeModuleWithUnicodePathnames(name, co, pathobj, cpathobj);
+    Py_XDECREF(pathobj);
+    Py_XDECREF(cpathobj);
+    return m;
+}
+
 
 /* Like strrchr(string, '/') but searches for the rightmost of either SEP
    or ALTSEP, if the latter is defined.
@@ -1068,6 +1082,7 @@
     long magic;
     PyCodeObject *co;
     PyObject *m;
+    PyObject *cpathobj;
 
     magic = PyMarshal_ReadLongFromFile(fp);
     if (magic != pyc_magic) {
@@ -1082,8 +1097,10 @@
     if (Py_VerboseFlag)
         PySys_WriteStderr("import %s # precompiled from %s\n",
             name, cpathname);
-    m = PyImport_ExecCodeModuleWithPathnames(
-        name, (PyObject *)co, cpathname, cpathname);
+    cpathobj = PyUnicode_DecodeFSDefault(cpathname);
+    m = PyImport_ExecCodeModuleWithUnicodePathnames(
+        name, (PyObject *)co, cpathobj, cpathobj);
+    Py_DECREF(cpathobj);
     Py_DECREF(co);
 
     return m;

From python-checkins at python.org  Fri Jul  9 01:32:26 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:26 +0200 (CEST)
Subject: [Python-checkins] r82674 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233226.01B70EEB39@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:25 2010
New Revision: 82674

Log:
load_source_module() uses PyImport_ExecCodeModuleWithUnicodePathnames

instead of PyImport_ExecCodeModuleWithPathnames

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:25 2010
@@ -859,6 +859,7 @@
    for the compiled file, or NULL if there's no space in the buffer.
    Doesn't set an exception. */
 
+/* FIXME: use Py_UNICODE*, not char* */
 static char *
 make_compiled_pathname(char *pathname, char *buf, size_t buflen, int debug)
 {
@@ -1289,6 +1290,7 @@
     char buf[MAXPATHLEN+1];
     char *pathname;
     char *cpathname;
+    PyObject *cpathobj;
     PyCodeObject *co;
     PyObject *m;
 
@@ -1313,28 +1315,40 @@
     }
 #endif
     cpathname = make_compiled_pathname(
-        pathname, buf, (size_t)MAXPATHLEN + 1, !Py_OptimizeFlag);
-    if (cpathname != NULL)
+        pathname, buf, sizeof(buf), !Py_OptimizeFlag);
+    if (cpathname != NULL) {
+        cpathobj = PyUnicode_DecodeFSDefault(cpathname);
+        if (cpathobj == NULL)
+            return NULL;
         fpc = check_compiled_module(pathname, st.st_mtime, cpathname);
-    else
+    }
+    else {
+        cpathobj = NULL;
         fpc = NULL;
+    }
     if (fpc) {
         co = read_compiled_module(cpathname, fpc);
         fclose(fpc);
-        if (co == NULL)
+        if (co == NULL) {
+            Py_XDECREF(cpathobj);
             return NULL;
-        if (update_compiled_module(co, pathobj) < 0)
+        }
+        if (update_compiled_module(co, pathobj) < 0) {
+            Py_XDECREF(cpathobj);
             return NULL;
+        }
         if (Py_VerboseFlag)
             PySys_WriteStderr("import %s # precompiled from %s\n",
                 name, cpathname);
-        pathobj = NULL;
-        pathname = cpathname;
+        m = PyImport_ExecCodeModuleWithUnicodePathnames(
+            name, (PyObject *)co, cpathobj, cpathobj);
     }
     else {
         co = parse_source_module(pathname, fp);
-        if (co == NULL)
+        if (co == NULL) {
+            Py_XDECREF(cpathobj);
             return NULL;
+        }
         if (Py_VerboseFlag)
             PySys_WriteStderr("import %s # from %s\n",
                 name, pathname);
@@ -1343,9 +1357,10 @@
             if (ro == NULL || !PyObject_IsTrue(ro))
                 write_compiled_module(co, cpathname, &st);
         }
+        m = PyImport_ExecCodeModuleWithUnicodePathnames(
+            name, (PyObject *)co, pathobj, cpathobj);
     }
-    m = PyImport_ExecCodeModuleWithPathnames(
-        name, (PyObject *)co, pathname, cpathname);
+    Py_XDECREF(cpathobj);
     Py_DECREF(co);
 
     return m;
@@ -3539,7 +3554,7 @@
         if ((debug = PyObject_IsTrue(debug_override)) < 0)
             return NULL;
 
-    cpathname = make_compiled_pathname(pathname, buf, MAXPATHLEN+1, debug);
+    cpathname = make_compiled_pathname(pathname, buf, sizeof(buf), debug);
     PyMem_Free(pathname);
 
     if (cpathname == NULL) {

From python-checkins at python.org  Fri Jul  9 01:32:27 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:27 +0200 (CEST)
Subject: [Python-checkins] r82675 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233227.68104EEB79@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:27 2010
New Revision: 82675

Log:
load_compiled_module(): use object for the path

instead of char*

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:27 2010
@@ -1078,12 +1078,15 @@
    module object WITH INCREMENTED REFERENCE COUNT */
 
 static PyObject *
-load_compiled_module(char *name, char *cpathname, FILE *fp)
+load_compiled_module(char *name, PyObject *cpathobj, FILE *fp)
 {
     long magic;
     PyCodeObject *co;
     PyObject *m;
-    PyObject *cpathobj;
+    char *cpathname;
+
+    /* FIXME: don't use _PyUnicode_AsString */
+    cpathname = _PyUnicode_AsString(cpathobj);
 
     magic = PyMarshal_ReadLongFromFile(fp);
     if (magic != pyc_magic) {
@@ -1098,10 +1101,8 @@
     if (Py_VerboseFlag)
         PySys_WriteStderr("import %s # precompiled from %s\n",
             name, cpathname);
-    cpathobj = PyUnicode_DecodeFSDefault(cpathname);
     m = PyImport_ExecCodeModuleWithUnicodePathnames(
         name, (PyObject *)co, cpathobj, cpathobj);
-    Py_DECREF(cpathobj);
     Py_DECREF(co);
 
     return m;
@@ -2120,9 +2121,14 @@
         break;
     }
 
-    case PY_COMPILED:
-        m = load_compiled_module(name, pathname, fp);
+    case PY_COMPILED: {
+        PyObject *pathobj = PyUnicode_DecodeFSDefault(pathname);
+        if (pathobj == NULL)
+            return NULL;
+        m = load_compiled_module(name, pathobj, fp);
+        Py_DECREF(pathobj);
         break;
+    }
 
 #ifdef HAVE_DYNAMIC_LOADING
     case C_EXTENSION: {
@@ -3390,23 +3396,20 @@
 imp_load_compiled(PyObject *self, PyObject *args)
 {
     char *name;
-    char *pathname;
+    PyObject *pathname;
     PyObject *fob = NULL;
     PyObject *m;
     FILE *fp;
-    if (!PyArg_ParseTuple(args, "ses|O:load_compiled",
-                          &name,
-                          Py_FileSystemDefaultEncoding, &pathname,
-                          &fob))
+    if (!PyArg_ParseTuple(args, "sU|O:load_compiled",
+                          &name, &pathname, &fob))
         return NULL;
-    fp = get_file(pathname, fob, "rb");
+                  /* FIXME: don't use _PyUnicode_AsString */
+    fp = get_file(_PyUnicode_AsString(pathname), fob, "rb");
     if (fp == NULL) {
-        PyMem_Free(pathname);
         return NULL;
     }
     m = load_compiled_module(name, pathname, fp);
     fclose(fp);
-    PyMem_Free(pathname);
     return m;
 }
 

From python-checkins at python.org  Fri Jul  9 01:32:28 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:28 +0200 (CEST)
Subject: [Python-checkins] r82676 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233228.CD6FCEEBA0@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:28 2010
New Revision: 82676

Log:
get_file() path is now an object

instead of char*

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:28 2010
@@ -3363,12 +3363,17 @@
 }
 
 static FILE *
-get_file(char *pathname, PyObject *fob, char *mode)
+get_file(PyObject *pathobj, PyObject *fob, char *mode)
 {
     FILE *fp;
+    char *pathname;
     if (mode[0] == 'U')
         mode = "r" PY_STDIOTEXTMODE;
     if (fob == NULL) {
+        /* FIXME: don't use _PyUnicode_AsString */
+        pathname = _PyUnicode_AsString(pathobj);
+        if (pathname == NULL)
+            return NULL;
         fp = fopen(pathname, mode);
     }
     else {
@@ -3403,8 +3408,7 @@
     if (!PyArg_ParseTuple(args, "sU|O:load_compiled",
                           &name, &pathname, &fob))
         return NULL;
-                  /* FIXME: don't use _PyUnicode_AsString */
-    fp = get_file(_PyUnicode_AsString(pathname), fob, "rb");
+    fp = get_file(pathname, fob, "rb");
     if (fp == NULL) {
         return NULL;
     }
@@ -3429,8 +3433,7 @@
                           &fob))
         return NULL;
     if (fob) {
-                      /* FIXME: don't use _PyUnicode_AsString */
-        fp = get_file(_PyUnicode_AsString(pathname), fob, "r");
+        fp = get_file(pathname, fob, "r");
         if (fp == NULL)
             return NULL;
     }
@@ -3453,8 +3456,7 @@
     if (!PyArg_ParseTuple(args, "sU|O:load_source",
                           &name, &pathname, &fob))
         return NULL;
-                  /* FIXME: don't use _PyUnicode_AsString */
-    fp = get_file(_PyUnicode_AsString(pathname), fob, "r");
+    fp = get_file(pathname, fob, "r");
     if (fp == NULL)
         return NULL;
     m = load_source_module(name, pathname, fp);

From python-checkins at python.org  Fri Jul  9 01:32:30 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:30 +0200 (CEST)
Subject: [Python-checkins] r82677 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233230.384F9EEB83@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:30 2010
New Revision: 82677

Log:
create load_builtin() static function

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:30 2010
@@ -2088,15 +2088,51 @@
 
 static int init_builtin(char *); /* Forward */
 
+static PyObject*
+load_builtin(char *name, char *pathname, int type)
+{
+    int err;
+    PyObject *m;
+    PyObject *modules;
+    
+    if (pathname != NULL && pathname[0] != '\0')
+        name = pathname;
+    if (type == C_BUILTIN)
+        err = init_builtin(name);
+    else
+        err = PyImport_ImportFrozenModule(name);
+    if (err < 0)
+        return NULL;
+    if (err == 0) {
+        PyErr_Format(PyExc_ImportError,
+                     "Purported %s module %.200s not found",
+                     type == C_BUILTIN ?
+                                "builtin" : "frozen",
+                     name);
+        return NULL;
+    }
+    modules = PyImport_GetModuleDict();
+    m = PyDict_GetItemString(modules, name);
+    if (m == NULL) {
+        PyErr_Format(
+            PyExc_ImportError,
+            "%s module %.200s not properly initialized",
+            type == C_BUILTIN ?
+                "builtin" : "frozen",
+            name);
+        return NULL;
+    }
+    Py_INCREF(m);
+    return m;
+}
+
 /* Load an external module using the default search path and return
    its module object WITH INCREMENTED REFERENCE COUNT */
 
 static PyObject *
 load_module(char *name, FILE *fp, char *pathname, int type, PyObject *loader)
 {
-    PyObject *modules;
     PyObject *m;
-    int err;
 
     /* First check that there's an open file (if we need one)  */
     switch (type) {
@@ -2152,34 +2188,7 @@
 
     case C_BUILTIN:
     case PY_FROZEN:
-        if (pathname != NULL && pathname[0] != '\0')
-            name = pathname;
-        if (type == C_BUILTIN)
-            err = init_builtin(name);
-        else
-            err = PyImport_ImportFrozenModule(name);
-        if (err < 0)
-            return NULL;
-        if (err == 0) {
-            PyErr_Format(PyExc_ImportError,
-                         "Purported %s module %.200s not found",
-                         type == C_BUILTIN ?
-                                    "builtin" : "frozen",
-                         name);
-            return NULL;
-        }
-        modules = PyImport_GetModuleDict();
-        m = PyDict_GetItemString(modules, name);
-        if (m == NULL) {
-            PyErr_Format(
-                PyExc_ImportError,
-                "%s module %.200s not properly initialized",
-                type == C_BUILTIN ?
-                    "builtin" : "frozen",
-                name);
-            return NULL;
-        }
-        Py_INCREF(m);
+        m = load_builtin(name, pathname, type);
         break;
 
     case IMP_HOOK: {

From python-checkins at python.org  Fri Jul  9 01:32:31 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:31 +0200 (CEST)
Subject: [Python-checkins] r82678 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233231.B74E6EEB20@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:31 2010
New Revision: 82678

Log:
simplify load_source()

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:31 2010
@@ -2094,7 +2094,7 @@
     int err;
     PyObject *m;
     PyObject *modules;
-    
+
     if (pathname != NULL && pathname[0] != '\0')
         name = pathname;
     if (type == C_BUILTIN)
@@ -2133,6 +2133,7 @@
 load_module(char *name, FILE *fp, char *pathname, int type, PyObject *loader)
 {
     PyObject *m;
+    PyObject *pathobj;
 
     /* First check that there's an open file (if we need one)  */
     switch (type) {
@@ -2146,45 +2147,29 @@
         }
     }
 
+    pathobj = PyUnicode_DecodeFSDefault(pathname);
+    if (pathobj == NULL)
+        return NULL;
+
     switch (type) {
 
-    case PY_SOURCE: {
-        PyObject *pathobj = PyUnicode_DecodeFSDefault(pathname);
-        if (pathobj == NULL)
-            return NULL;
+    case PY_SOURCE:
         m = load_source_module(name, pathobj, fp);
-        Py_DECREF(pathobj);
         break;
-    }
 
-    case PY_COMPILED: {
-        PyObject *pathobj = PyUnicode_DecodeFSDefault(pathname);
-        if (pathobj == NULL)
-            return NULL;
+    case PY_COMPILED:
         m = load_compiled_module(name, pathobj, fp);
-        Py_DECREF(pathobj);
         break;
-    }
 
 #ifdef HAVE_DYNAMIC_LOADING
-    case C_EXTENSION: {
-        PyObject *pathobj = PyUnicode_DecodeFSDefault(pathname);
-        if (pathobj == NULL)
-            return NULL;
+    case C_EXTENSION:
         m = _PyImport_LoadDynamicModule(name, pathobj, fp);
-        Py_DECREF(pathobj);
         break;
-    }
 #endif
 
-    case PKG_DIRECTORY: {
-        PyObject *pathobj = PyUnicode_DecodeFSDefault(pathname);
-        if (pathobj == NULL)
-            return NULL;
+    case PKG_DIRECTORY:
         m = load_package(name, pathobj);
-        Py_DECREF(pathobj);
         break;
-    }
 
     case C_BUILTIN:
     case PY_FROZEN:
@@ -2195,6 +2180,7 @@
         if (loader == NULL) {
             PyErr_SetString(PyExc_ImportError,
                             "import hook without loader");
+            Py_DECREF(pathobj);
             return NULL;
         }
         m = PyObject_CallMethod(loader, "load_module", "s", name);
@@ -2208,6 +2194,7 @@
         m = NULL;
 
     }
+    Py_DECREF(pathobj);
 
     return m;
 }

From python-checkins at python.org  Fri Jul  9 01:32:33 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:33 +0200 (CEST)
Subject: [Python-checkins] r82679 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233233.2B534EEB04@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:33 2010
New Revision: 82679

Log:
load_builtin() path is an object

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:33 2010
@@ -2089,14 +2089,21 @@
 static int init_builtin(char *); /* Forward */
 
 static PyObject*
-load_builtin(char *name, char *pathname, int type)
+load_builtin(char *name, PyObject *pathobj, int type)
 {
     int err;
     PyObject *m;
     PyObject *modules;
+    char *pathname;
 
-    if (pathname != NULL && pathname[0] != '\0')
-        name = pathname;
+    if (pathobj != NULL) {
+        /* FIXME: don't use _PyUnicode_AsString */
+        pathname = _PyUnicode_AsString(pathobj);
+        if (pathname == NULL)
+            return NULL;
+        if (pathname[0] != '\0')
+            name = pathname;
+    }
     if (type == C_BUILTIN)
         err = init_builtin(name);
     else
@@ -2173,7 +2180,7 @@
 
     case C_BUILTIN:
     case PY_FROZEN:
-        m = load_builtin(name, pathname, type);
+        m = load_builtin(name, pathobj, type);
         break;
 
     case IMP_HOOK: {

From python-checkins at python.org  Fri Jul  9 01:32:34 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:34 +0200 (CEST)
Subject: [Python-checkins] r82680 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233234.D176FEEB79@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:34 2010
New Revision: 82680

Log:
load_module() path is an object

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:34 2010
@@ -1417,7 +1417,7 @@
 }
 
 /* Forward */
-static PyObject *load_module(char *, FILE *, char *, int, PyObject *);
+static PyObject *load_module(char *, FILE *, PyObject *, int, PyObject *);
 static struct filedescr *find_module(char *, char *, PyObject *,
                                      char *, size_t, FILE **, PyObject **);
 static struct _frozen * find_frozen(char *);
@@ -1436,6 +1436,7 @@
     FILE *fp = NULL;
     struct filedescr *fdp;
     char *pathname;
+    PyObject *bufobj;
 
     /* FIXME: don't use _PyUnicode_AsString */
     pathname = _PyUnicode_AsString(pathobj);
@@ -1469,7 +1470,11 @@
             m = NULL;
         goto cleanup;
     }
-    m = load_module(name, fp, buf, fdp->type, NULL);
+    bufobj = PyUnicode_DecodeFSDefault(buf);
+    if (bufobj != NULL)
+        m = load_module(name, fp, bufobj, fdp->type, NULL);
+    else
+        m = NULL;
     if (fp != NULL)
         fclose(fp);
     goto cleanup;
@@ -2137,10 +2142,9 @@
    its module object WITH INCREMENTED REFERENCE COUNT */
 
 static PyObject *
-load_module(char *name, FILE *fp, char *pathname, int type, PyObject *loader)
+load_module(char *name, FILE *fp, PyObject *pathobj, int type, PyObject *loader)
 {
     PyObject *m;
-    PyObject *pathobj;
 
     /* First check that there's an open file (if we need one)  */
     switch (type) {
@@ -2154,10 +2158,6 @@
         }
     }
 
-    pathobj = PyUnicode_DecodeFSDefault(pathname);
-    if (pathobj == NULL)
-        return NULL;
-
     switch (type) {
 
     case PY_SOURCE:
@@ -2201,7 +2201,6 @@
         m = NULL;
 
     }
-    Py_DECREF(pathobj);
 
     return m;
 }
@@ -2929,6 +2928,7 @@
 {
     PyObject *modules = PyImport_GetModuleDict();
     PyObject *m = NULL;
+    PyObject *pathobj;
 
     /* Require:
        if mod == None: subname == fullname
@@ -2966,7 +2966,11 @@
             Py_INCREF(Py_None);
             return Py_None;
         }
-        m = load_module(fullname, fp, buf, fdp->type, loader);
+        pathobj = PyUnicode_DecodeFSDefault(buf);
+        if (pathobj != NULL)
+            m = load_module(fullname, fp, pathobj, fdp->type, loader);
+        else
+            m = NULL;
         Py_XDECREF(loader);
         if (fp)
             fclose(fp);
@@ -2994,6 +2998,7 @@
     char buf[MAXPATHLEN+1];
     struct filedescr *fdp;
     FILE *fp = NULL;
+    PyObject *pathobj;
     PyObject *newm;
 
     if (modules_reloading == NULL) {
@@ -3061,7 +3066,11 @@
         return NULL;
     }
 
-    newm = load_module(name, fp, buf, fdp->type, loader);
+    pathobj = PyUnicode_DecodeFSDefault(buf);
+    if (pathobj != NULL)
+        newm = load_module(name, fp, pathobj, fdp->type, loader);
+    else
+        newm = NULL;
     Py_XDECREF(loader);
 
     if (fp)
@@ -3472,16 +3481,16 @@
 {
     char *name;
     PyObject *fob;
-    char *pathname;
+    PyObject *pathname;
     PyObject * ret;
     char *suffix; /* Unused */
     char *mode;
     int type;
     FILE *fp;
 
-    if (!PyArg_ParseTuple(args, "sOes(ssi):load_module",
+    if (!PyArg_ParseTuple(args, "sOU(ssi):load_module",
                           &name, &fob,
-                          Py_FileSystemDefaultEncoding, &pathname,
+                          &pathname,
                           &suffix, &mode, &type))
         return NULL;
     if (*mode) {
@@ -3492,7 +3501,6 @@
         if (!(*mode == 'r' || *mode == 'U') || strchr(mode, '+')) {
             PyErr_Format(PyExc_ValueError,
                          "invalid file open mode %.200s", mode);
-            PyMem_Free(pathname);
             return NULL;
         }
     }
@@ -3500,13 +3508,10 @@
         fp = NULL;
     else {
         fp = get_file(NULL, fob, mode);
-        if (fp == NULL) {
-            PyMem_Free(pathname);
+        if (fp == NULL)
             return NULL;
-        }
     }
     ret = load_module(name, fp, pathname, type, NULL);
-    PyMem_Free(pathname);
     if (fp)
         fclose(fp);
     return ret;

From python-checkins at python.org  Fri Jul  9 01:32:36 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:36 +0200 (CEST)
Subject: [Python-checkins] r82681 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233236.61414EEAF0@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:36 2010
New Revision: 82681

Log:
change make_compiled_pathname() API

but not the code

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:36 2010
@@ -720,8 +720,8 @@
 
 static PyObject * get_sourcefile(PyObject *file);
 static char *make_source_pathname(char *pathname, char *buf);
-static char *make_compiled_pathname(char *pathname, char *buf, size_t buflen,
-                                    int debug);
+static PyObject *make_compiled_pathname(PyObject *pathname, char *buf,
+                                        size_t buflen, int debug);
 
 /* Execute a code object in a module and return the module object
  * WITH INCREMENTED REFERENCE COUNT.  If an error occurs, name is
@@ -861,7 +861,7 @@
 
 /* FIXME: use Py_UNICODE*, not char* */
 static char *
-make_compiled_pathname(char *pathname, char *buf, size_t buflen, int debug)
+_make_compiled_pathname(char *pathname, char *buf, size_t buflen, int debug)
 {
     /* foo.py -> __pycache__/foo.<tag>.pyc */
     size_t len = strlen(pathname);
@@ -963,6 +963,21 @@
     return buf;
 }
 
+static PyObject*
+make_compiled_pathname(PyObject *pathobj, char *buf, size_t buflen, int debug)
+{
+    char *pathname, *cpathname;
+
+    /* FIXME: don't use _PyUnicode_AsString */
+    pathname = _PyUnicode_AsString(pathobj);
+
+    cpathname = _make_compiled_pathname(pathname, buf, buflen, debug);
+    if (cpathname != NULL)
+        return PyUnicode_DecodeFSDefault(cpathname);
+    else
+        return NULL;
+}
+
 
 /* Given a pathname to a Python byte compiled file, return the path to the
    source file, if the path matches the PEP 3147 format.  This does not check
@@ -1315,16 +1330,21 @@
         return NULL;
     }
 #endif
-    cpathname = make_compiled_pathname(
-        pathname, buf, sizeof(buf), !Py_OptimizeFlag);
-    if (cpathname != NULL) {
-        cpathobj = PyUnicode_DecodeFSDefault(cpathname);
-        if (cpathobj == NULL)
+    cpathobj = make_compiled_pathname(
+        pathobj, buf, sizeof(buf), !Py_OptimizeFlag);
+    if (cpathobj != NULL) {
+        /* FIXME: don't use _PyUnicode_AsString */
+        cpathname = _PyUnicode_AsString(cpathobj);
+        if (cpathobj == NULL) {
+            Py_DECREF(cpathobj);
             return NULL;
+        }
         fpc = check_compiled_module(pathname, st.st_mtime, cpathname);
     }
     else {
-        cpathobj = NULL;
+        if (PyErr_Occurred())
+            return NULL;
+        cpathname = NULL;
         fpc = NULL;
     }
     if (fpc) {
@@ -3554,13 +3574,14 @@
     static char *kwlist[] = {"path", "debug_override", NULL};
 
     char buf[MAXPATHLEN+1];
-    char *pathname, *cpathname;
+    PyObject *pathname;
     PyObject *debug_override = Py_None;
     int debug = !Py_OptimizeFlag;
+    PyObject *cpathname;
 
     if (!PyArg_ParseTupleAndKeywords(
-                args, kws, "es|O", kwlist,
-                Py_FileSystemDefaultEncoding, &pathname, &debug_override))
+                args, kws, "U|O", kwlist,
+                &pathname, &debug_override))
         return NULL;
 
     if (debug_override != Py_None)
@@ -3568,13 +3589,11 @@
             return NULL;
 
     cpathname = make_compiled_pathname(pathname, buf, sizeof(buf), debug);
-    PyMem_Free(pathname);
-
     if (cpathname == NULL) {
         PyErr_Format(PyExc_SystemError, "path buffer too short");
         return NULL;
     }
-    return PyUnicode_FromString(buf);
+    return cpathname;
 }
 
 PyDoc_STRVAR(doc_cache_from_source,

From python-checkins at python.org  Fri Jul  9 01:32:37 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:37 +0200 (CEST)
Subject: [Python-checkins] r82682 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233237.D199FEEAF3@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:37 2010
New Revision: 82682

Log:
change make_source_pathname(): use objects

and not char*

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:37 2010
@@ -719,7 +719,7 @@
 }
 
 static PyObject * get_sourcefile(PyObject *file);
-static char *make_source_pathname(char *pathname, char *buf);
+static PyObject *make_source_pathname(PyObject *pathname, char *buf);
 static PyObject *make_compiled_pathname(PyObject *pathname, char *buf,
                                         size_t buflen, int debug);
 
@@ -986,7 +986,7 @@
    the resulting path will always be shorter. */
 
 static char *
-make_source_pathname(char *pathname, char *buf)
+_make_source_pathname(char *pathname, char *buf)
 {
     /* __pycache__/foo.<tag>.pyc -> foo.py */
     size_t i, j;
@@ -1032,6 +1032,21 @@
     return buf;
 }
 
+static PyObject*
+make_source_pathname(PyObject *pathobj, char *buf)
+{
+    char *pathname, *cpathname;
+
+    /* FIXME: don't use _PyUnicode_AsString */
+    pathname = _PyUnicode_AsString(pathobj);
+
+    cpathname = _make_source_pathname(pathname, buf);
+    if (cpathname != NULL)
+        return PyUnicode_DecodeFSDefault(cpathname);
+    else
+        return NULL;
+}
+
 /* Given a pathname for a Python source file, its time of last
    modification, and a pathname for a compiled file, check whether the
    compiled file represents the same version of the source.  If so,
@@ -1394,19 +1409,20 @@
 get_sourcefile(PyObject *fileobj)
 {
     /* FIXME: use Py_UNICODE* instead of char* */
-    char py[MAXPATHLEN + 1];
+    char buf[MAXPATHLEN + 1];
+    char *py;
+    PyObject *pyobj;
     Py_ssize_t len;
     struct stat statbuf;
     char *file;
 
-    if (!fileobj) {
+    if (fileobj == NULL) {
         Py_RETURN_NONE;
     }
 
     /* FIXME: don't use _PyUnicode_AsString */
     file = _PyUnicode_AsString(fileobj);
-
-    if (!file || !*file) {
+    if (file == NULL || !*file) {
         Py_RETURN_NONE;
     }
 
@@ -1421,16 +1437,25 @@
      * fails, just chop off the trailing character, i.e. legacy pyc path
      * to py.
      */
-    if (make_source_pathname(file, py) == NULL) {
-        strncpy(py, file, len-1);
-        py[len-1] = '\0';
+    pyobj = make_source_pathname(fileobj, buf);
+    if (pyobj == NULL) {
+        if (PyErr_Occurred())
+            return NULL;
+        pyobj = PyUnicode_FromStringAndSize(file, len - 1);
+        if (pyobj == NULL)
+            return NULL;
     }
 
+    /* FIXME: don't use _PyUnicode_AsString */
+    py = _PyUnicode_AsString(pyobj);
+    if (py == NULL)
+        return NULL;
     if (stat(py, &statbuf) == 0 &&
         S_ISREG(statbuf.st_mode)) {
-        return PyUnicode_DecodeFSDefault(py);
+        return pyobj;
     }
     else {
+        Py_DECREF(pyobj);
         Py_INCREF(fileobj);
         return fileobj;
     }
@@ -3611,22 +3636,23 @@
 {
     static char *kwlist[] = {"path", NULL};
 
-    char *pathname;
+    PyObject *pathname;
     char buf[MAXPATHLEN+1];
+    PyObject *result;
 
     if (!PyArg_ParseTupleAndKeywords(
-                args, kws, "es", kwlist,
-                Py_FileSystemDefaultEncoding, &pathname))
+                args, kws, "U", kwlist,
+                &pathname))
         return NULL;
 
-    if (make_source_pathname(pathname, buf) == NULL) {
-        PyErr_Format(PyExc_ValueError, "Not a PEP 3147 pyc path: %s",
-                     pathname);
-        PyMem_Free(pathname);
+    result = make_source_pathname(pathname, buf);
+    if (result == NULL) {
+        if (!PyErr_Occurred())
+            PyErr_Format(PyExc_ValueError, "Not a PEP 3147 pyc path: %U",
+                         pathname);
         return NULL;
     }
-    PyMem_Free(pathname);
-    return PyUnicode_FromString(buf);
+    return result;
 }
 
 PyDoc_STRVAR(doc_source_from_cache,

From python-checkins at python.org  Fri Jul  9 01:32:39 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:39 +0200 (CEST)
Subject: [Python-checkins] r82683 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233239.82BD2EEBCA@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:39 2010
New Revision: 82683

Log:
get_sourcefile() uses Py_UNICODE* for file

Move buf into make_source_pathname()

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:39 2010
@@ -719,7 +719,7 @@
 }
 
 static PyObject * get_sourcefile(PyObject *file);
-static PyObject *make_source_pathname(PyObject *pathname, char *buf);
+static PyObject *make_source_pathname(PyObject *pathname);
 static PyObject *make_compiled_pathname(PyObject *pathname, char *buf,
                                         size_t buflen, int debug);
 
@@ -1033,8 +1033,10 @@
 }
 
 static PyObject*
-make_source_pathname(PyObject *pathobj, char *buf)
+make_source_pathname(PyObject *pathobj)
 {
+    /* FIXME: use Py_UNICODE* instead of char* */
+    char buf[MAXPATHLEN + 1];
     char *pathname, *cpathname;
 
     /* FIXME: don't use _PyUnicode_AsString */
@@ -1408,27 +1410,26 @@
 static PyObject *
 get_sourcefile(PyObject *fileobj)
 {
-    /* FIXME: use Py_UNICODE* instead of char* */
-    char buf[MAXPATHLEN + 1];
     char *py;
     PyObject *pyobj;
+    Py_UNICODE *file;
     Py_ssize_t len;
     struct stat statbuf;
-    char *file;
 
     if (fileobj == NULL) {
         Py_RETURN_NONE;
     }
 
-    /* FIXME: don't use _PyUnicode_AsString */
-    file = _PyUnicode_AsString(fileobj);
-    if (file == NULL || !*file) {
+    len = PyUnicode_GET_SIZE(fileobj);
+    if (len == 0)
         Py_RETURN_NONE;
-    }
+    file = PyUnicode_AS_UNICODE(fileobj);
 
-    len = strlen(file);
     /* match '*.py?' */
-    if (len > MAXPATHLEN || PyOS_strnicmp(&file[len-4], ".py", 3) != 0) {
+    if (4 <= len
+        && file[len-3] == '.'
+        && (file[len-2] == 'p' || file[len-2] == 'P')
+        && (file[len-1] == 'y' || file[len-1] == 'Y')) {
         Py_INCREF(fileobj);
         return fileobj;
     }
@@ -1437,11 +1438,11 @@
      * fails, just chop off the trailing character, i.e. legacy pyc path
      * to py.
      */
-    pyobj = make_source_pathname(fileobj, buf);
+    pyobj = make_source_pathname(fileobj);
     if (pyobj == NULL) {
         if (PyErr_Occurred())
             return NULL;
-        pyobj = PyUnicode_FromStringAndSize(file, len - 1);
+        pyobj = PyUnicode_FromUnicode(file, len - 1);
         if (pyobj == NULL)
             return NULL;
     }
@@ -3637,7 +3638,6 @@
     static char *kwlist[] = {"path", NULL};
 
     PyObject *pathname;
-    char buf[MAXPATHLEN+1];
     PyObject *result;
 
     if (!PyArg_ParseTupleAndKeywords(
@@ -3645,7 +3645,7 @@
                 &pathname))
         return NULL;
 
-    result = make_source_pathname(pathname, buf);
+    result = make_source_pathname(pathname);
     if (result == NULL) {
         if (!PyErr_Occurred())
             PyErr_Format(PyExc_ValueError, "Not a PEP 3147 pyc path: %U",

From python-checkins at python.org  Fri Jul  9 01:32:41 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:41 +0200 (CEST)
Subject: [Python-checkins] r82684 - in python/branches/import_unicode:
	Include/sysmodule.h Python/sysmodule.c
Message-ID: <20100708233241.0F420EEBC1@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:40 2010
New Revision: 82684

Log:
Create PySys_FormatStderr() function

Similar to PySys_WriteStderr() but use PyUnicode_FromFormatV() instead
of PyOS_snprintf()

Modified:
   python/branches/import_unicode/Include/sysmodule.h
   python/branches/import_unicode/Python/sysmodule.c

Modified: python/branches/import_unicode/Include/sysmodule.h
==============================================================================
--- python/branches/import_unicode/Include/sysmodule.h	(original)
+++ python/branches/import_unicode/Include/sysmodule.h	Fri Jul  9 01:32:40 2010
@@ -14,9 +14,10 @@
 PyAPI_FUNC(void) PySys_SetPath(const wchar_t *);
 
 PyAPI_FUNC(void) PySys_WriteStdout(const char *format, ...)
-			Py_GCC_ATTRIBUTE((format(printf, 1, 2)));
+                 Py_GCC_ATTRIBUTE((format(printf, 1, 2)));
 PyAPI_FUNC(void) PySys_WriteStderr(const char *format, ...)
-			Py_GCC_ATTRIBUTE((format(printf, 1, 2)));
+                 Py_GCC_ATTRIBUTE((format(printf, 1, 2)));
+PyAPI_FUNC(void) PySys_FormatStderr(const char *format, ...);
 
 PyAPI_DATA(PyObject *) _PySys_TraceFunc, *_PySys_ProfileFunc;
 

Modified: python/branches/import_unicode/Python/sysmodule.c
==============================================================================
--- python/branches/import_unicode/Python/sysmodule.c	(original)
+++ python/branches/import_unicode/Python/sysmodule.c	Fri Jul  9 01:32:40 2010
@@ -1834,18 +1834,14 @@
    PyErr_CheckSignals(): avoid the call to PyObject_Str(). */
 
 static int
-sys_pyfile_write(const char *text, PyObject *file)
+sys_pyfile_write_unicode(PyObject *unicode, PyObject *file)
 {
-    PyObject *unicode = NULL, *writer = NULL, *args = NULL, *result = NULL;
+    PyObject *writer = NULL, *args = NULL, *result = NULL;
     int err;
 
     if (file == NULL)
         return -1;
 
-    unicode = PyUnicode_FromString(text);
-    if (unicode == NULL)
-        goto error;
-
     writer = PyObject_GetAttrString(file, "write");
     if (writer == NULL)
         goto error;
@@ -1865,13 +1861,29 @@
 error:
     err = -1;
 finally:
-    Py_XDECREF(unicode);
     Py_XDECREF(writer);
     Py_XDECREF(args);
     Py_XDECREF(result);
     return err;
 }
 
+static int
+sys_pyfile_write(const char *text, PyObject *file)
+{
+    PyObject *unicode = NULL;
+    int err;
+
+    if (file == NULL)
+        return -1;
+
+    unicode = PyUnicode_FromString(text);
+    if (unicode == NULL)
+        return -1;
+
+    err = sys_pyfile_write_unicode(unicode, file);
+    Py_DECREF(unicode);
+    return err;
+}
 
 /* APIs to write to sys.stdout or sys.stderr using a printf-like interface.
    Adapted from code submitted by Just van Rossum.
@@ -1918,10 +1930,28 @@
     }
     if (written < 0 || (size_t)written >= sizeof(buffer)) {
         const char *truncated = "... truncated";
-        if (sys_pyfile_write(truncated, file) != 0) {
-            PyErr_Clear();
+        if (sys_pyfile_write(truncated, file) != 0)
             fputs(truncated, fp);
-        }
+    }
+    PyErr_Restore(error_type, error_value, error_traceback);
+}
+
+static void
+myformat(char *name, FILE *fp, const char *format, va_list va)
+{
+    PyObject *file;
+    PyObject *error_type, *error_value, *error_traceback;
+    PyObject *unicode;
+    char *str;
+
+    PyErr_Fetch(&error_type, &error_value, &error_traceback);
+    file = PySys_GetObject(name);
+    unicode = PyUnicode_FromFormatV(format, va);
+    if (sys_pyfile_write_unicode(unicode, file) != 0) {
+        PyErr_Clear();
+        str = _PyUnicode_AsString(unicode);
+        if (str != NULL)
+            fputs(str, fp);
     }
     PyErr_Restore(error_type, error_value, error_traceback);
 }
@@ -1945,3 +1975,13 @@
     mywrite("stderr", stderr, format, va);
     va_end(va);
 }
+
+void
+PySys_FormatStderr(const char *format, ...)
+{
+    va_list va;
+
+    va_start(va, format);
+    myformat("stderr", stderr, format, va);
+    va_end(va);
+}

From python-checkins at python.org  Fri Jul  9 01:32:42 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:42 +0200 (CEST)
Subject: [Python-checkins] r82685 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233242.7BA43EEB62@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:42 2010
New Revision: 82685

Log:
move the buffer into make_compiled_pathname()

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:42 2010
@@ -720,8 +720,7 @@
 
 static PyObject * get_sourcefile(PyObject *file);
 static PyObject *make_source_pathname(PyObject *pathname);
-static PyObject *make_compiled_pathname(PyObject *pathname, char *buf,
-                                        size_t buflen, int debug);
+static PyObject *make_compiled_pathname(PyObject *pathname, int debug);
 
 /* Execute a code object in a module and return the module object
  * WITH INCREMENTED REFERENCE COUNT.  If an error occurs, name is
@@ -964,14 +963,15 @@
 }
 
 static PyObject*
-make_compiled_pathname(PyObject *pathobj, char *buf, size_t buflen, int debug)
+make_compiled_pathname(PyObject *pathobj, int debug)
 {
+    char buf[MAXPATHLEN+1];
     char *pathname, *cpathname;
 
     /* FIXME: don't use _PyUnicode_AsString */
     pathname = _PyUnicode_AsString(pathobj);
 
-    cpathname = _make_compiled_pathname(pathname, buf, buflen, debug);
+    cpathname = _make_compiled_pathname(pathname, buf, sizeof(buf), debug);
     if (cpathname != NULL)
         return PyUnicode_DecodeFSDefault(cpathname);
     else
@@ -1347,8 +1347,7 @@
         return NULL;
     }
 #endif
-    cpathobj = make_compiled_pathname(
-        pathobj, buf, sizeof(buf), !Py_OptimizeFlag);
+    cpathobj = make_compiled_pathname(pathobj, !Py_OptimizeFlag);
     if (cpathobj != NULL) {
         /* FIXME: don't use _PyUnicode_AsString */
         cpathname = _PyUnicode_AsString(cpathobj);
@@ -3598,8 +3597,6 @@
 imp_cache_from_source(PyObject *self, PyObject *args, PyObject *kws)
 {
     static char *kwlist[] = {"path", "debug_override", NULL};
-
-    char buf[MAXPATHLEN+1];
     PyObject *pathname;
     PyObject *debug_override = Py_None;
     int debug = !Py_OptimizeFlag;
@@ -3614,7 +3611,7 @@
         if ((debug = PyObject_IsTrue(debug_override)) < 0)
             return NULL;
 
-    cpathname = make_compiled_pathname(pathname, buf, sizeof(buf), debug);
+    cpathname = make_compiled_pathname(pathname, debug);
     if (cpathname == NULL) {
         PyErr_Format(PyExc_SystemError, "path buffer too short");
         return NULL;

From python-checkins at python.org  Fri Jul  9 01:32:43 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:43 +0200 (CEST)
Subject: [Python-checkins] r82686 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233243.E0F03EEBC1@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:43 2010
New Revision: 82686

Log:
use PySys_FormatStderr()

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:43 2010
@@ -661,10 +661,8 @@
         return NULL;
     }
     if (Py_VerboseFlag)
-        PySys_WriteStderr("import %s # previously loaded (%s)\n",
-                          name,
-                          /* FIXME: don't use _PyUnicode_AsString */
-                          _PyUnicode_AsString(filename));
+        PySys_FormatStderr("import %s # previously loaded (%U)\n",
+                           name, filename);
     return mod;
 }
 
@@ -1131,8 +1129,8 @@
     if (co == NULL)
         return NULL;
     if (Py_VerboseFlag)
-        PySys_WriteStderr("import %s # precompiled from %s\n",
-            name, cpathname);
+        PySys_FormatStderr("import %s # precompiled from %U\n",
+                           name, cpathobj);
     m = PyImport_ExecCodeModuleWithUnicodePathnames(
         name, (PyObject *)co, cpathobj, cpathobj);
     Py_DECREF(co);
@@ -1320,7 +1318,6 @@
 {
     struct stat st;
     FILE *fpc;
-    char buf[MAXPATHLEN+1];
     char *pathname;
     char *cpathname;
     PyObject *cpathobj;
@@ -1375,8 +1372,8 @@
             return NULL;
         }
         if (Py_VerboseFlag)
-            PySys_WriteStderr("import %s # precompiled from %s\n",
-                name, cpathname);
+            PySys_FormatStderr("import %s # precompiled from %U\n",
+                               name, cpathobj);
         m = PyImport_ExecCodeModuleWithUnicodePathnames(
             name, (PyObject *)co, cpathobj, cpathobj);
     }
@@ -1387,8 +1384,7 @@
             return NULL;
         }
         if (Py_VerboseFlag)
-            PySys_WriteStderr("import %s # from %s\n",
-                name, pathname);
+            PySys_FormatStderr("import %s # from %U\n", name, pathobj);
         if (cpathname) {
             PyObject *ro = PySys_GetObject("dont_write_bytecode");
             if (ro == NULL || !PyObject_IsTrue(ro))
@@ -1490,8 +1486,7 @@
     if (m == NULL)
         return NULL;
     if (Py_VerboseFlag)
-        PySys_WriteStderr("import %s # directory %s\n",
-            name, pathname);
+        PySys_FormatStderr("import %s # directory %U\n", name, pathobj);
     d = PyModule_GetDict(m);
     file = get_sourcefile(pathobj);
     if (file == NULL)

From python-checkins at python.org  Fri Jul  9 01:32:45 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:45 +0200 (CEST)
Subject: [Python-checkins] r82687 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233245.58D0FEEB79@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:45 2010
New Revision: 82687

Log:
use object in error message

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:45 2010
@@ -1329,8 +1329,8 @@
 
     if (fstat(fileno(fp), &st) != 0) {
         PyErr_Format(PyExc_RuntimeError,
-                     "unable to get file status from '%s'",
-                     pathname);
+                     "unable to get file status from '%U'",
+                     pathobj);
         return NULL;
     }
 #if SIZEOF_TIME_T > 4

From python-checkins at python.org  Fri Jul  9 01:32:46 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:46 +0200 (CEST)
Subject: [Python-checkins] r82688 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233246.BB2B4EEB30@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:46 2010
New Revision: 82688

Log:
parse_source_module() uses object

not char*

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:46 2010
@@ -1141,15 +1141,22 @@
 /* Parse a source file and return the corresponding code object */
 
 static PyCodeObject *
-parse_source_module(const char *pathname, FILE *fp)
+parse_source_module(PyObject *pathobj, FILE *fp)
 {
     PyCodeObject *co = NULL;
     mod_ty mod;
     PyCompilerFlags flags;
+    char *pathname;
+
     PyArena *arena = PyArena_New();
     if (arena == NULL)
         return NULL;
 
+    /* FIXME: use PyUnicode_EncodeFSDefault() */
+    pathname = _PyUnicode_AsString(pathobj);
+    if (pathname == NULL)
+        return NULL;
+
     flags.cf_flags = 0;
     mod = PyParser_ASTFromFile(fp, pathname, NULL,
                                Py_file_input, 0, 0, &flags,
@@ -1324,8 +1331,10 @@
     PyCodeObject *co;
     PyObject *m;
 
-    /* FIXME: don't use _PyUnicode_AsString */
+    /* FIXME: use PyUnicode_EncodeFSDefault() */
     pathname = _PyUnicode_AsString(pathobj);
+    if (pathname == NULL)
+        return NULL;
 
     if (fstat(fileno(fp), &st) != 0) {
         PyErr_Format(PyExc_RuntimeError,
@@ -1378,7 +1387,7 @@
             name, (PyObject *)co, cpathobj, cpathobj);
     }
     else {
-        co = parse_source_module(pathname, fp);
+        co = parse_source_module(pathobj, fp);
         if (co == NULL) {
             Py_XDECREF(cpathobj);
             return NULL;
@@ -1442,7 +1451,7 @@
             return NULL;
     }
 
-    /* FIXME: don't use _PyUnicode_AsString */
+    /* FIXME: use PyUnicode_EncodeFSDefault() */
     py = _PyUnicode_AsString(pyobj);
     if (py == NULL)
         return NULL;
@@ -1476,12 +1485,8 @@
     char buf[MAXPATHLEN+1];
     FILE *fp = NULL;
     struct filedescr *fdp;
-    char *pathname;
     PyObject *bufobj;
 
-    /* FIXME: don't use _PyUnicode_AsString */
-    pathname = _PyUnicode_AsString(pathobj);
-
     m = PyImport_AddModule(name);
     if (m == NULL)
         return NULL;

From python-checkins at python.org  Fri Jul  9 01:32:48 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:48 +0200 (CEST)
Subject: [Python-checkins] r82689 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233248.39D72EE9BE@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:48 2010
New Revision: 82689

Log:
check_compiled_module() and read_compiled_module() use objects

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:48 2010
@@ -1055,11 +1055,17 @@
    Doesn't set an exception. */
 
 static FILE *
-check_compiled_module(char *pathname, time_t mtime, char *cpathname)
+check_compiled_module(PyObject *pathobj, time_t mtime, PyObject *cpathobj)
 {
     FILE *fp;
     long magic;
     long pyc_mtime;
+    char *cpathname;
+
+    /* FIXME: use PyUnicode_EncodeFSDefault() */
+    cpathname = _PyUnicode_AsString(cpathobj);
+    if (cpathname == NULL)
+        return NULL;
 
     fp = fopen(cpathname, "rb");
     if (fp == NULL)
@@ -1067,19 +1073,19 @@
     magic = PyMarshal_ReadLongFromFile(fp);
     if (magic != pyc_magic) {
         if (Py_VerboseFlag)
-            PySys_WriteStderr("# %s has bad magic\n", cpathname);
+            PySys_FormatStderr("# %U has bad magic\n", cpathobj);
         fclose(fp);
         return NULL;
     }
     pyc_mtime = PyMarshal_ReadLongFromFile(fp);
     if (pyc_mtime != mtime) {
         if (Py_VerboseFlag)
-            PySys_WriteStderr("# %s has bad mtime\n", cpathname);
+            PySys_FormatStderr("# %s has bad mtime\n", cpathobj);
         fclose(fp);
         return NULL;
     }
     if (Py_VerboseFlag)
-        PySys_WriteStderr("# %s matches %s\n", cpathname, pathname);
+        PySys_FormatStderr("# %U matches %U\n", cpathobj, pathobj);
     return fp;
 }
 
@@ -1087,7 +1093,7 @@
 /* Read a code object from a file and check it for validity */
 
 static PyCodeObject *
-read_compiled_module(char *cpathname, FILE *fp)
+read_compiled_module(PyObject *cpathobj, FILE *fp)
 {
     PyObject *co;
 
@@ -1096,7 +1102,7 @@
         return NULL;
     if (!PyCode_Check(co)) {
         PyErr_Format(PyExc_ImportError,
-                     "Non-code object in %.200s", cpathname);
+                     "Non-code object in %U", cpathobj);
         Py_DECREF(co);
         return NULL;
     }
@@ -1113,19 +1119,15 @@
     long magic;
     PyCodeObject *co;
     PyObject *m;
-    char *cpathname;
-
-    /* FIXME: don't use _PyUnicode_AsString */
-    cpathname = _PyUnicode_AsString(cpathobj);
 
     magic = PyMarshal_ReadLongFromFile(fp);
     if (magic != pyc_magic) {
         PyErr_Format(PyExc_ImportError,
-                     "Bad magic number in %.200s", cpathname);
+                     "Bad magic number in %U", cpathobj);
         return NULL;
     }
     (void) PyMarshal_ReadLongFromFile(fp);
-    co = read_compiled_module(cpathname, fp);
+    co = read_compiled_module(cpathobj, fp);
     if (co == NULL)
         return NULL;
     if (Py_VerboseFlag)
@@ -1325,17 +1327,11 @@
 {
     struct stat st;
     FILE *fpc;
-    char *pathname;
     char *cpathname;
     PyObject *cpathobj;
     PyCodeObject *co;
     PyObject *m;
 
-    /* FIXME: use PyUnicode_EncodeFSDefault() */
-    pathname = _PyUnicode_AsString(pathobj);
-    if (pathname == NULL)
-        return NULL;
-
     if (fstat(fileno(fp), &st) != 0) {
         PyErr_Format(PyExc_RuntimeError,
                      "unable to get file status from '%U'",
@@ -1361,7 +1357,7 @@
             Py_DECREF(cpathobj);
             return NULL;
         }
-        fpc = check_compiled_module(pathname, st.st_mtime, cpathname);
+        fpc = check_compiled_module(pathobj, st.st_mtime, cpathobj);
     }
     else {
         if (PyErr_Occurred())
@@ -1370,7 +1366,7 @@
         fpc = NULL;
     }
     if (fpc) {
-        co = read_compiled_module(cpathname, fpc);
+        co = read_compiled_module(cpathobj, fpc);
         fclose(fpc);
         if (co == NULL) {
             Py_XDECREF(cpathobj);

From python-checkins at python.org  Fri Jul  9 01:32:49 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:49 +0200 (CEST)
Subject: [Python-checkins] r82690 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233249.9DD0CEEB49@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:49 2010
New Revision: 82690

Log:
write_compiled_module() uses an object for the path

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:49 2010
@@ -1211,7 +1211,8 @@
    remove the file. */
 
 static void
-write_compiled_module(PyCodeObject *co, char *cpathname, struct stat *srcstat)
+write_compiled_module(PyCodeObject *co, PyObject *cpathobj,
+                      struct stat *srcstat)
 {
     FILE *fp;
     char *dirpath;
@@ -1228,6 +1229,14 @@
                       S_IWUSR | S_IWGRP | S_IWOTH);
 #endif
     int saved;
+    char *cpathname;
+
+    /* FIXME: use PyUnicode_EncodeFSDefault() */
+    cpathname = _PyUnicode_AsString(cpathobj);
+    if (cpathname == NULL) {
+        PyErr_Clear();
+        return;
+    }
 
     /* Ensure that the __pycache__ directory exists. */
     dirpath = rightmost_sep(cpathname);
@@ -1327,7 +1336,6 @@
 {
     struct stat st;
     FILE *fpc;
-    char *cpathname;
     PyObject *cpathobj;
     PyCodeObject *co;
     PyObject *m;
@@ -1350,19 +1358,11 @@
     }
 #endif
     cpathobj = make_compiled_pathname(pathobj, !Py_OptimizeFlag);
-    if (cpathobj != NULL) {
-        /* FIXME: don't use _PyUnicode_AsString */
-        cpathname = _PyUnicode_AsString(cpathobj);
-        if (cpathobj == NULL) {
-            Py_DECREF(cpathobj);
-            return NULL;
-        }
+    if (cpathobj != NULL)
         fpc = check_compiled_module(pathobj, st.st_mtime, cpathobj);
-    }
     else {
         if (PyErr_Occurred())
             return NULL;
-        cpathname = NULL;
         fpc = NULL;
     }
     if (fpc) {
@@ -1390,10 +1390,10 @@
         }
         if (Py_VerboseFlag)
             PySys_FormatStderr("import %s # from %U\n", name, pathobj);
-        if (cpathname) {
+        if (cpathobj != NULL) {
             PyObject *ro = PySys_GetObject("dont_write_bytecode");
             if (ro == NULL || !PyObject_IsTrue(ro))
-                write_compiled_module(co, cpathname, &st);
+                write_compiled_module(co, cpathobj, &st);
         }
         m = PyImport_ExecCodeModuleWithUnicodePathnames(
             name, (PyObject *)co, pathobj, cpathobj);

From python-checkins at python.org  Fri Jul  9 01:32:51 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:51 +0200 (CEST)
Subject: [Python-checkins] r82691 - in python/branches/import_unicode:
	Include/import.h Python/import.c
Message-ID: <20100708233251.257D7EEB20@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:50 2010
New Revision: 82691

Log:
Remove unused private function _PyImport_FindModule()

Modified:
   python/branches/import_unicode/Include/import.h
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Include/import.h
==============================================================================
--- python/branches/import_unicode/Include/import.h	(original)
+++ python/branches/import_unicode/Include/import.h	Fri Jul  9 01:32:50 2010
@@ -46,8 +46,7 @@
 #define _PyImport_ReleaseLock() 1
 #endif
 
-PyAPI_FUNC(struct filedescr *) _PyImport_FindModule(
-    const char *, PyObject *, char *, size_t, FILE **, PyObject **);
+struct filedescr;
 PyAPI_FUNC(int) _PyImport_IsScript(struct filedescr *);
 PyAPI_FUNC(void) _PyImport_ReInitLock(void);
 

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:50 2010
@@ -983,6 +983,7 @@
    3147 style, NULL is returned.  buf must be at least as big as pathname;
    the resulting path will always be shorter. */
 
+/* FIXME: use Py_UNICODE* instead of char* */
 static char *
 _make_source_pathname(char *pathname, char *buf)
 {
@@ -1914,14 +1915,6 @@
 /* Helpers for main.c
  *  Find the source file corresponding to a named module
  */
-struct filedescr *
-_PyImport_FindModule(const char *name, PyObject *path, char *buf,
-            size_t buflen, FILE **p_fp, PyObject **p_loader)
-{
-    return find_module((char *) name, (char *) name, path,
-                       buf, buflen, p_fp, p_loader);
-}
-
 PyAPI_FUNC(int) _PyImport_IsScript(struct filedescr * fd)
 {
     return fd->type == PY_SOURCE || fd->type == PY_COMPILED;

From python-checkins at python.org  Fri Jul  9 01:32:52 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:52 +0200 (CEST)
Subject: [Python-checkins] r82692 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233252.9737DEEB83@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:52 2010
New Revision: 82692

Log:
change find_module() API: path is stored as an object

fix also refleaks in functions calling find_module()

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:52 2010
@@ -1466,7 +1466,7 @@
 /* Forward */
 static PyObject *load_module(char *, FILE *, PyObject *, int, PyObject *);
 static struct filedescr *find_module(char *, char *, PyObject *,
-                                     char *, size_t, FILE **, PyObject **);
+                                     PyObject **, FILE **, PyObject **);
 static struct _frozen * find_frozen(char *);
 
 /* Load a package and return its module object WITH INCREMENTED
@@ -1479,7 +1479,6 @@
     PyObject *file = NULL;
     PyObject *path = NULL;
     int err;
-    char buf[MAXPATHLEN+1];
     FILE *fp = NULL;
     struct filedescr *fdp;
     PyObject *bufobj;
@@ -1501,8 +1500,7 @@
         err = PyDict_SetItemString(d, "__path__", path);
     if (err != 0)
         goto error;
-    buf[0] = '\0';
-    fdp = find_module(name, "__init__", path, buf, sizeof(buf), &fp, NULL);
+    fdp = find_module(name, "__init__", path, &bufobj, &fp, NULL);
     if (fdp == NULL) {
         if (PyErr_ExceptionMatches(PyExc_ImportError)) {
             PyErr_Clear();
@@ -1512,11 +1510,7 @@
             m = NULL;
         goto cleanup;
     }
-    bufobj = PyUnicode_DecodeFSDefault(buf);
-    if (bufobj != NULL)
-        m = load_module(name, fp, bufobj, fdp->type, NULL);
-    else
-        m = NULL;
+    m = load_module(name, fp, bufobj, fdp->type, NULL);
     if (fp != NULL)
         fclose(fp);
     goto cleanup;
@@ -1640,7 +1634,7 @@
 static struct filedescr importhookdescr = {"", "", IMP_HOOK};
 
 static struct filedescr *
-find_module(char *fullname, char *subname, PyObject *path, char *buf,
+_find_module(char *fullname, char *subname, PyObject *path, char *buf,
             size_t buflen, FILE **p_fp, PyObject **p_loader)
 {
     Py_ssize_t i, npath;
@@ -1912,6 +1906,28 @@
     return fdp;
 }
 
+static struct filedescr *
+find_module(char *fullname, char *subname, PyObject *path, PyObject **result,
+            FILE **p_fp, PyObject **p_loader)
+{
+    char buf[MAXPATHLEN+1];
+    struct filedescr *fd;
+    buf[0] = '\0';
+    fd = _find_module(fullname, subname, path,
+                      buf, sizeof(buf),
+                      p_fp, p_loader);
+    if (fd != NULL) {
+        *result = PyUnicode_DecodeFSDefault(buf);
+        if (*result == NULL) {
+            if (*p_fp != NULL)
+                fclose(*p_fp);
+            return NULL;
+        }
+    } else
+        *result = NULL;
+    return fd;
+}
+
 /* Helpers for main.c
  *  Find the source file corresponding to a named module
  */
@@ -2974,7 +2990,6 @@
     }
     else {
         PyObject *path, *loader = NULL;
-        char buf[MAXPATHLEN+1];
         struct filedescr *fdp;
         FILE *fp = NULL;
 
@@ -2989,8 +3004,7 @@
             }
         }
 
-        buf[0] = '\0';
-        fdp = find_module(fullname, subname, path, buf, MAXPATHLEN+1,
+        fdp = find_module(fullname, subname, path, &pathobj,
                           &fp, &loader);
         Py_XDECREF(path);
         if (fdp == NULL) {
@@ -3000,11 +3014,8 @@
             Py_INCREF(Py_None);
             return Py_None;
         }
-        pathobj = PyUnicode_DecodeFSDefault(buf);
-        if (pathobj != NULL)
-            m = load_module(fullname, fp, pathobj, fdp->type, loader);
-        else
-            m = NULL;
+        m = load_module(fullname, fp, pathobj, fdp->type, loader);
+        Py_DECREF(pathobj);
         Py_XDECREF(loader);
         if (fp)
             fclose(fp);
@@ -3029,7 +3040,6 @@
     PyObject *modules = PyImport_GetModuleDict();
     PyObject *path = NULL, *loader = NULL, *existing_m = NULL;
     char *name, *subname;
-    char buf[MAXPATHLEN+1];
     struct filedescr *fdp;
     FILE *fp = NULL;
     PyObject *pathobj;
@@ -3090,8 +3100,7 @@
         if (path == NULL)
             PyErr_Clear();
     }
-    buf[0] = '\0';
-    fdp = find_module(name, subname, path, buf, MAXPATHLEN+1, &fp, &loader);
+    fdp = find_module(name, subname, path, &pathobj, &fp, &loader);
     Py_XDECREF(path);
 
     if (fdp == NULL) {
@@ -3100,11 +3109,8 @@
         return NULL;
     }
 
-    pathobj = PyUnicode_DecodeFSDefault(buf);
-    if (pathobj != NULL)
-        newm = load_module(name, fp, pathobj, fdp->type, loader);
-    else
-        newm = NULL;
+    newm = load_module(name, fp, pathobj, fdp->type, loader);
+    Py_DECREF(pathobj);
     Py_XDECREF(loader);
 
     if (fp)
@@ -3261,21 +3267,16 @@
     PyObject *fob, *ret;
     PyObject *pathobj;
     struct filedescr *fdp;
-    char pathname[MAXPATHLEN+1];
     FILE *fp = NULL;
     int fd = -1;
     char *found_encoding = NULL;
     char *encoding = NULL;
 
-    pathname[0] = '\0';
     if (path == Py_None)
         path = NULL;
-    fdp = find_module(NULL, name, path, pathname, MAXPATHLEN+1, &fp, NULL);
+    fdp = find_module(NULL, name, path, &pathobj, &fp, NULL);
     if (fdp == NULL)
         return NULL;
-    pathobj = PyUnicode_DecodeFSDefault(pathname);
-    if (pathobj == NULL)
-        return NULL;
     if (fp != NULL) {
         fd = fileno(fp);
         if (fd != -1)
@@ -3289,8 +3290,10 @@
                memory. */
             found_encoding = PyTokenizer_FindEncoding(fd);
             lseek(fd, 0, 0); /* Reset position */
-            if (found_encoding == NULL && PyErr_Occurred())
+            if (found_encoding == NULL && PyErr_Occurred()) {
+                Py_DECREF(pathobj);
                 return NULL;
+            }
             encoding = (found_encoding != NULL) ? found_encoding :
                    (char*)PyUnicode_GetDefaultEncoding();
         }
@@ -3298,6 +3301,7 @@
                                     encoding, NULL, NULL, 1);
         if (fob == NULL) {
             close(fd);
+            Py_DECREF(pathobj);
             PyMem_FREE(found_encoding);
             return NULL;
         }

From python-checkins at python.org  Fri Jul  9 01:32:54 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:54 +0200 (CEST)
Subject: [Python-checkins] r82693 - in python/branches/import_unicode:
	Include/warnings.h Objects/moduleobject.c
	Objects/typeobject.c Objects/unicodeobject.c
	Python/_warnings.c Python/import.c
Message-ID: <20100708233254.65116EEB20@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:54 2010
New Revision: 82693

Log:
Create PyErr_WarnUnicode() and PyErr_WarnFormat()

Modified:
   python/branches/import_unicode/Include/warnings.h
   python/branches/import_unicode/Objects/moduleobject.c
   python/branches/import_unicode/Objects/typeobject.c
   python/branches/import_unicode/Objects/unicodeobject.c
   python/branches/import_unicode/Python/_warnings.c
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Include/warnings.h
==============================================================================
--- python/branches/import_unicode/Include/warnings.h	(original)
+++ python/branches/import_unicode/Include/warnings.h	Fri Jul  9 01:32:54 2010
@@ -7,6 +7,8 @@
 PyAPI_FUNC(PyObject*) _PyWarnings_Init(void);
 
 PyAPI_FUNC(int) PyErr_WarnEx(PyObject *, const char *, Py_ssize_t);
+PyAPI_FUNC(int) PyErr_WarnUnicode(PyObject *, PyObject *, Py_ssize_t);
+PyAPI_FUNC(int) PyErr_WarnFormat(PyObject *, Py_ssize_t, const char *, ...);
 PyAPI_FUNC(int) PyErr_WarnExplicit(PyObject *, const char *, const char *, int,
                                     const char *, PyObject *);
 

Modified: python/branches/import_unicode/Objects/moduleobject.c
==============================================================================
--- python/branches/import_unicode/Objects/moduleobject.c	(original)
+++ python/branches/import_unicode/Objects/moduleobject.c	Fri Jul  9 01:32:54 2010
@@ -56,10 +56,6 @@
     return NULL;
 }
 
-static char api_version_warning[] =
-"Python C API version mismatch for module %.100s:\
- This Python has API version %d, module %.100s has version %d.";
-
 PyObject *
 PyModule_Create2(struct PyModuleDef* module, int module_api_version)
 {
@@ -79,12 +75,13 @@
     }
     name = module->m_name;
     if (module_api_version != PYTHON_API_VERSION) {
-        char message[512];
-        PyOS_snprintf(message, sizeof(message),
-                      api_version_warning, name,
-                      PYTHON_API_VERSION, name,
-                      module_api_version);
-        if (PyErr_WarnEx(PyExc_RuntimeWarning, message, 1))
+        int err;
+        err = PyErr_WarnFormat(PyExc_RuntimeWarning, 1,
+            "Python C API version mismatch for module %.100s: "
+            "This Python has API version %d, module %.100s has version %d.",
+             name,
+             PYTHON_API_VERSION, name, module_api_version);
+        if (err)
             return NULL;
     }
     /* Make sure name is fully qualified.

Modified: python/branches/import_unicode/Objects/typeobject.c
==============================================================================
--- python/branches/import_unicode/Objects/typeobject.c	(original)
+++ python/branches/import_unicode/Objects/typeobject.c	Fri Jul  9 01:32:54 2010
@@ -3892,13 +3892,11 @@
        tp_reserved) but not tp_richcompare. */
     if (type->tp_reserved && !type->tp_richcompare) {
         int error;
-        char msg[240];
-        PyOS_snprintf(msg, sizeof(msg),
-                      "Type %.100s defines tp_reserved (formerly "
-                      "tp_compare) but not tp_richcompare. "
-                      "Comparisons may not behave as intended.",
-                      type->tp_name);
-        error = PyErr_WarnEx(PyExc_DeprecationWarning, msg, 1);
+        error = PyErr_WarnFormat(PyExc_DeprecationWarning, 1,
+            "Type %.100s defines tp_reserved (formerly "
+            "tp_compare) but not tp_richcompare. "
+            "Comparisons may not behave as intended.",
+            type->tp_name);
         if (error == -1)
             goto error;
     }

Modified: python/branches/import_unicode/Objects/unicodeobject.c
==============================================================================
--- python/branches/import_unicode/Objects/unicodeobject.c	(original)
+++ python/branches/import_unicode/Objects/unicodeobject.c	Fri Jul  9 01:32:54 2010
@@ -1558,12 +1558,13 @@
 
     /* If the codec returns a buffer, raise a warning and convert to bytes */
     if (PyByteArray_Check(v)) {
-        char msg[100];
+        int error;
         PyObject *b;
-        PyOS_snprintf(msg, sizeof(msg),
-                      "encoder %s returned buffer instead of bytes",
-                      encoding);
-        if (PyErr_WarnEx(PyExc_RuntimeWarning, msg, 1) < 0) {
+
+        error = PyErr_WarnFormat(PyExc_RuntimeWarning, 1,
+            "encoder %s returned buffer instead of bytes",
+            encoding);
+        if (error) {
             Py_DECREF(v);
             return NULL;
         }

Modified: python/branches/import_unicode/Python/_warnings.c
==============================================================================
--- python/branches/import_unicode/Python/_warnings.c	(original)
+++ python/branches/import_unicode/Python/_warnings.c	Fri Jul  9 01:32:54 2010
@@ -710,21 +710,18 @@
                                 registry, NULL);
 }
 
-
 /* Function to issue a warning message; may raise an exception. */
+
 int
-PyErr_WarnEx(PyObject *category, const char *text, Py_ssize_t stack_level)
+PyErr_WarnUnicode(PyObject *category, PyObject *message,
+                  Py_ssize_t stack_level)
 {
     PyObject *res;
-    PyObject *message = PyUnicode_FromString(text);
-    if (message == NULL)
-        return -1;
 
     if (category == NULL)
         category = PyExc_RuntimeWarning;
 
     res = do_warn(message, category, stack_level);
-    Py_DECREF(message);
     if (res == NULL)
         return -1;
     Py_DECREF(res);
@@ -732,6 +729,42 @@
     return 0;
 }
 
+int
+PyErr_WarnFormat(PyObject *category, Py_ssize_t stack_level,
+                 const char *format, ...)
+{
+    int ret;
+    PyObject *unicode;
+    va_list vargs;
+
+#ifdef HAVE_STDARG_PROTOTYPES
+    va_start(vargs, format);
+#else
+    va_start(vargs);
+#endif
+    unicode = PyUnicode_FromFormatV(format, vargs);
+    if (unicode != NULL) {
+        ret = PyErr_WarnUnicode(category, unicode, stack_level);
+        Py_DECREF(unicode);
+    }
+    else
+        ret = -1;
+    va_end(vargs);
+    return ret;
+}
+
+int
+PyErr_WarnEx(PyObject *category, const char *text, Py_ssize_t stack_level)
+{
+    int ret;
+    PyObject *message = PyUnicode_FromString(text);
+    if (message == NULL)
+        return -1;
+    ret = PyErr_WarnUnicode(category, message, stack_level);
+    Py_DECREF(message);
+    return ret;
+}
+
 /* PyErr_Warn is only for backwards compatability and will be removed.
    Use PyErr_WarnEx instead. */
 

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:54 2010
@@ -2756,19 +2756,16 @@
     parent = PyDict_GetItemString(modules, buf);
     if (parent == NULL) {
         if (orig_level < 1) {
-            PyObject *err_msg = PyBytes_FromFormat(
+            int err;
+            err = PyErr_WarnFormat(PyExc_RuntimeWarning, 1,
                 "Parent module '%.200s' not found "
-                "while handling absolute import", buf);
-            if (err_msg == NULL) {
-                return NULL;
-            }
-            if (!PyErr_WarnEx(PyExc_RuntimeWarning,
-                              PyBytes_AsString(err_msg), 1)) {
+                "while handling absolute import",
+                buf);
+            if (!err) {
                 *buf = '\0';
                 *p_buflen = 0;
                 parent = Py_None;
             }
-            Py_DECREF(err_msg);
         } else {
             PyErr_Format(PyExc_SystemError,
                 "Parent module '%.200s' not loaded, "

From python-checkins at python.org  Fri Jul  9 01:32:55 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:55 +0200 (CEST)
Subject: [Python-checkins] r82694 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233255.E651CEEB39@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:55 2010
New Revision: 82694

Log:
use PyErr_WarnFormat()

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:55 2010
@@ -1819,14 +1819,12 @@
                 return &fd_package;
             }
             else {
-                char warnstr[MAXPATHLEN+80];
-                sprintf(warnstr, "Not importing directory "
-                    "'%.*s': missing __init__.py",
-                    MAXPATHLEN, buf);
-                if (PyErr_WarnEx(PyExc_ImportWarning,
-                                 warnstr, 1)) {
+                int warnerr;
+                warnerr = PyErr_WarnFormat(PyExc_ImportWarning, 1,
+                    "Not importing directory '%s': missing __init__.py",
+                    buf);
+                if (warnerr)
                     return NULL;
-                }
             }
         }
 #endif
@@ -1866,7 +1864,7 @@
 #endif /* PYOS_OS2 */
             strcpy(buf+len, fdp->suffix);
             if (Py_VerboseFlag > 1)
-                PySys_WriteStderr("# trying %s\n", buf);
+                PySys_FormatStderr("# trying %s\n", buf);
             filemode = fdp->mode;
             if (filemode[0] == 'U')
                 filemode = "r" PY_STDIOTEXTMODE;

From python-checkins at python.org  Fri Jul  9 01:32:57 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:57 +0200 (CEST)
Subject: [Python-checkins] r82695 - in python/branches/import_unicode:
	Doc/library/sys.rst Python/import.c
Message-ID: <20100708233257.8D264EEB39@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:57 2010
New Revision: 82695

Log:
_find_module() fills directly path argument

 * find_module(): rename "path" argument to "search_path"
 * find_module(): result can be NULL
 * reject bytes path
 * fix refleak

Modified:
   python/branches/import_unicode/Doc/library/sys.rst
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Doc/library/sys.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/sys.rst	(original)
+++ python/branches/import_unicode/Doc/library/sys.rst	Fri Jul  9 01:32:57 2010
@@ -599,9 +599,9 @@
 
    .. index:: triple: module; search; path
 
-   A list of strings that specifies the search path for modules. Initialized from
-   the environment variable :envvar:`PYTHONPATH`, plus an installation-dependent
-   default.
+   A list of unicode strings that specifies the search path for modules.
+   Initialized from the environment variable :envvar:`PYTHONPATH`, plus an
+   installation-dependent default.
 
    As initialized upon program startup, the first item of this list, ``path[0]``,
    is the directory containing the script that was used to invoke the Python

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:57 2010
@@ -1477,7 +1477,7 @@
 {
     PyObject *m, *d;
     PyObject *file = NULL;
-    PyObject *path = NULL;
+    PyObject *search_path = NULL;
     int err;
     FILE *fp = NULL;
     struct filedescr *fdp;
@@ -1492,15 +1492,15 @@
     file = get_sourcefile(pathobj);
     if (file == NULL)
         goto error;
-    path = Py_BuildValue("[O]", file);
-    if (path == NULL)
+    search_path = Py_BuildValue("[O]", file);
+    if (search_path == NULL)
         goto error;
     err = PyDict_SetItemString(d, "__file__", file);
     if (err == 0)
-        err = PyDict_SetItemString(d, "__path__", path);
+        err = PyDict_SetItemString(d, "__path__", search_path);
     if (err != 0)
         goto error;
-    fdp = find_module(name, "__init__", path, &bufobj, &fp, NULL);
+    fdp = find_module(name, "__init__", search_path, &bufobj, &fp, NULL);
     if (fdp == NULL) {
         if (PyErr_ExceptionMatches(PyExc_ImportError)) {
             PyErr_Clear();
@@ -1511,6 +1511,7 @@
         goto cleanup;
     }
     m = load_module(name, fp, bufobj, fdp->type, NULL);
+    Py_XDECREF(bufobj);
     if (fp != NULL)
         fclose(fp);
     goto cleanup;
@@ -1518,7 +1519,7 @@
   error:
     m = NULL;
   cleanup:
-    Py_XDECREF(path);
+    Py_XDECREF(search_path);
     Py_XDECREF(file);
     return m;
 }
@@ -1634,8 +1635,9 @@
 static struct filedescr importhookdescr = {"", "", IMP_HOOK};
 
 static struct filedescr *
-_find_module(char *fullname, char *subname, PyObject *path, char *buf,
-            size_t buflen, FILE **p_fp, PyObject **p_loader)
+_find_module(char *fullname, char *subname, PyObject *search_path,
+            char *buf, size_t buflen, PyObject **path,
+            FILE **p_fp, PyObject **p_loader)
 {
     Py_ssize_t i, npath;
     size_t len, namelen;
@@ -1653,6 +1655,8 @@
     size_t saved_namelen;
     char *saved_buf = NULL;
 #endif
+
+    *path = NULL;
     if (p_loader != NULL)
         *p_loader = NULL;
 
@@ -1681,8 +1685,8 @@
             PyObject *hook = PyList_GetItem(meta_path, i);
             loader = PyObject_CallMethod(hook, "find_module",
                                          "sO", fullname,
-                                         path != NULL ?
-                                         path : Py_None);
+                                         search_path != NULL ?
+                                         search_path : Py_None);
             if (loader == NULL) {
                 Py_DECREF(meta_path);
                 return NULL;  /* true error */
@@ -1699,26 +1703,35 @@
     }
 
     if (find_frozen(fullname) != NULL) {
-        strcpy(buf, fullname);
+        *path = PyUnicode_DecodeFSDefault(fullname);
+        if (*path == NULL)
+            return NULL;
         return &fd_frozen;
     }
 
-    if (path == NULL) {
+    if (search_path == NULL) {
         if (is_builtin(name)) {
-            strcpy(buf, name);
+            *path = PyUnicode_DecodeFSDefault(name);
+            if (*path == NULL)
+                return NULL;
             return &fd_builtin;
         }
 #ifdef MS_COREDLL
         fp = PyWin_FindRegisteredModule(name, &fdp, buf, buflen);
         if (fp != NULL) {
+            *path = PyUnicode_DecodeFSDefault(buf);
+            if (*path == NULL) {
+                fclose(fp);
+                return NULL;
+            }
             *p_fp = fp;
             return fdp;
         }
 #endif
-        path = PySys_GetObject("path");
+        search_path = PySys_GetObject("path");
     }
 
-    if (path == NULL || !PyList_Check(path)) {
+    if (search_path == NULL || !PyList_Check(search_path)) {
         PyErr_SetString(PyExc_ImportError,
                         "sys.path must be a list of directory names");
         return NULL;
@@ -1739,10 +1752,10 @@
         return NULL;
     }
 
-    npath = PyList_Size(path);
+    npath = PyList_Size(search_path);
     namelen = strlen(name);
     for (i = 0; i < npath; i++) {
-        PyObject *v = PyList_GetItem(path, i);
+        PyObject *v = PyList_GetItem(search_path, i);
         PyObject *origv = v;
         const char *base;
         Py_ssize_t size;
@@ -1753,10 +1766,8 @@
             if (v == NULL)
                 return NULL;
         }
-        else if (!PyBytes_Check(v))
-            continue;
         else
-            Py_INCREF(v);
+            continue;
 
         base = PyBytes_AS_STRING(v);
         size = PyBytes_GET_SIZE(v);
@@ -1816,6 +1827,9 @@
             S_ISDIR(statbuf.st_mode) &&         /* it's a directory */
             case_ok(buf, len, namelen, name)) { /* case matches */
             if (find_init_module(buf)) { /* and has __init__.py */
+                *path = PyUnicode_DecodeFSDefault(buf);
+                if (*path == NULL)
+                    return NULL;
                 return &fd_package;
             }
             else {
@@ -1900,29 +1914,25 @@
                      "No module named %.200s", name);
         return NULL;
     }
+    *path = PyUnicode_DecodeFSDefault(buf);
+    if (*path == NULL) {
+        fclose(fp);
+        return NULL;
+    }
     *p_fp = fp;
     return fdp;
 }
 
 static struct filedescr *
-find_module(char *fullname, char *subname, PyObject *path, PyObject **result,
+find_module(char *fullname, char *subname, PyObject *search_path, PyObject **path,
             FILE **p_fp, PyObject **p_loader)
 {
     char buf[MAXPATHLEN+1];
     struct filedescr *fd;
     buf[0] = '\0';
-    fd = _find_module(fullname, subname, path,
-                      buf, sizeof(buf),
+    fd = _find_module(fullname, subname, search_path,
+                      buf, sizeof(buf), path,
                       p_fp, p_loader);
-    if (fd != NULL) {
-        *result = PyUnicode_DecodeFSDefault(buf);
-        if (*result == NULL) {
-            if (*p_fp != NULL)
-                fclose(*p_fp);
-            return NULL;
-        }
-    } else
-        *result = NULL;
     return fd;
 }
 
@@ -2206,6 +2216,11 @@
         }
     }
 
+    if (pathobj == NULL && type != IMP_HOOK) {
+        PyErr_BadInternalCall();
+        return NULL;
+    }
+
     switch (type) {
 
     case PY_SOURCE:
@@ -2235,7 +2250,7 @@
         if (loader == NULL) {
             PyErr_SetString(PyExc_ImportError,
                             "import hook without loader");
-            Py_DECREF(pathobj);
+            Py_XDECREF(pathobj);
             return NULL;
         }
         m = PyObject_CallMethod(loader, "load_module", "s", name);
@@ -2984,24 +2999,24 @@
         Py_INCREF(m);
     }
     else {
-        PyObject *path, *loader = NULL;
+        PyObject *search_path, *loader = NULL;
         struct filedescr *fdp;
         FILE *fp = NULL;
 
         if (mod == Py_None)
-            path = NULL;
+            search_path = NULL;
         else {
-            path = PyObject_GetAttrString(mod, "__path__");
-            if (path == NULL) {
+            search_path = PyObject_GetAttrString(mod, "__path__");
+            if (search_path == NULL) {
                 PyErr_Clear();
                 Py_INCREF(Py_None);
                 return Py_None;
             }
         }
 
-        fdp = find_module(fullname, subname, path, &pathobj,
+        fdp = find_module(fullname, subname, search_path, &pathobj,
                           &fp, &loader);
-        Py_XDECREF(path);
+        Py_XDECREF(search_path);
         if (fdp == NULL) {
             if (!PyErr_ExceptionMatches(PyExc_ImportError))
                 return NULL;
@@ -3010,7 +3025,7 @@
             return Py_None;
         }
         m = load_module(fullname, fp, pathobj, fdp->type, loader);
-        Py_DECREF(pathobj);
+        Py_XDECREF(pathobj);
         Py_XDECREF(loader);
         if (fp)
             fclose(fp);
@@ -3033,7 +3048,7 @@
     PyInterpreterState *interp = PyThreadState_Get()->interp;
     PyObject *modules_reloading = interp->modules_reloading;
     PyObject *modules = PyImport_GetModuleDict();
-    PyObject *path = NULL, *loader = NULL, *existing_m = NULL;
+    PyObject *search_path = NULL, *loader = NULL, *existing_m = NULL;
     char *name, *subname;
     struct filedescr *fdp;
     FILE *fp = NULL;
@@ -3091,12 +3106,12 @@
         }
         Py_DECREF(parentname);
         subname++;
-        path = PyObject_GetAttrString(parent, "__path__");
-        if (path == NULL)
+        search_path = PyObject_GetAttrString(parent, "__path__");
+        if (search_path == NULL)
             PyErr_Clear();
     }
-    fdp = find_module(name, subname, path, &pathobj, &fp, &loader);
-    Py_XDECREF(path);
+    fdp = find_module(name, subname, search_path, &pathobj, &fp, &loader);
+    Py_XDECREF(search_path);
 
     if (fdp == NULL) {
         Py_XDECREF(loader);
@@ -3105,7 +3120,7 @@
     }
 
     newm = load_module(name, fp, pathobj, fdp->type, loader);
-    Py_DECREF(pathobj);
+    Py_XDECREF(pathobj);
     Py_XDECREF(loader);
 
     if (fp)
@@ -3256,7 +3271,7 @@
 }
 
 static PyObject *
-call_find_module(char *name, PyObject *path)
+call_find_module(char *name, PyObject *search_path)
 {
     extern int fclose(FILE *);
     PyObject *fob, *ret;
@@ -3267,9 +3282,9 @@
     char *found_encoding = NULL;
     char *encoding = NULL;
 
-    if (path == Py_None)
-        path = NULL;
-    fdp = find_module(NULL, name, path, &pathobj, &fp, NULL);
+    if (search_path == Py_None)
+        search_path = NULL;
+    fdp = find_module(NULL, name, search_path, &pathobj, &fp, NULL);
     if (fdp == NULL)
         return NULL;
     if (fp != NULL) {
@@ -3286,7 +3301,7 @@
             found_encoding = PyTokenizer_FindEncoding(fd);
             lseek(fd, 0, 0); /* Reset position */
             if (found_encoding == NULL && PyErr_Occurred()) {
-                Py_DECREF(pathobj);
+                Py_XDECREF(pathobj);
                 return NULL;
             }
             encoding = (found_encoding != NULL) ? found_encoding :
@@ -3296,7 +3311,7 @@
                                     encoding, NULL, NULL, 1);
         if (fob == NULL) {
             close(fd);
-            Py_DECREF(pathobj);
+            Py_XDECREF(pathobj);
             PyMem_FREE(found_encoding);
             return NULL;
         }
@@ -3306,7 +3321,7 @@
         Py_INCREF(fob);
     }
     ret = Py_BuildValue("NN(ssi)",
-                        fob, pathobj,
+                        fob, (pathobj != NULL) ? pathobj : Py_None,
                         fdp->suffix, fdp->mode, fdp->type);
     PyMem_FREE(found_encoding);
 

From python-checkins at python.org  Fri Jul  9 01:32:59 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:32:59 +0200 (CEST)
Subject: [Python-checkins] r82696 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233259.029EBEEBCA@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:32:58 2010
New Revision: 82696

Log:
PyWin_FindRegisteredModule() uses its own buffer

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:32:58 2010
@@ -1710,6 +1710,10 @@
     }
 
     if (search_path == NULL) {
+#ifdef MS_COREDLL
+        /* FIXME: use buf buffer */
+        char bbuf[MAXPATHLEN+1];
+#endif
         if (is_builtin(name)) {
             *path = PyUnicode_DecodeFSDefault(name);
             if (*path == NULL)
@@ -1717,9 +1721,9 @@
             return &fd_builtin;
         }
 #ifdef MS_COREDLL
-        fp = PyWin_FindRegisteredModule(name, &fdp, buf, buflen);
+        fp = PyWin_FindRegisteredModule(name, &fdp, bbuf, sizeof(bbuf));
         if (fp != NULL) {
-            *path = PyUnicode_DecodeFSDefault(buf);
+            *path = PyUnicode_DecodeFSDefault(bbuf);
             if (*path == NULL) {
                 fclose(fp);
                 return NULL;

From python-checkins at python.org  Fri Jul  9 01:33:00 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:00 +0200 (CEST)
Subject: [Python-checkins] r82697 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233300.6A8F2EEB39@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:00 2010
New Revision: 82697

Log:
create unicode object

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:33:00 2010
@@ -1655,6 +1655,7 @@
     size_t saved_namelen;
     char *saved_buf = NULL;
 #endif
+    PyObject *unicode;
 
     *path = NULL;
     if (p_loader != NULL)
@@ -1827,24 +1828,28 @@
         /* Check for package import (buf holds a directory name,
            and there's an __init__ module in that directory */
 #ifdef HAVE_STAT
+        unicode = PyUnicode_DecodeFSDefault(buf);
+        if (unicode == NULL)
+            return NULL;
         if (stat(buf, &statbuf) == 0 &&         /* it exists */
             S_ISDIR(statbuf.st_mode) &&         /* it's a directory */
             case_ok(buf, len, namelen, name)) { /* case matches */
             if (find_init_module(buf)) { /* and has __init__.py */
-                *path = PyUnicode_DecodeFSDefault(buf);
-                if (*path == NULL)
-                    return NULL;
+                *path = unicode;
                 return &fd_package;
             }
             else {
                 int warnerr;
                 warnerr = PyErr_WarnFormat(PyExc_ImportWarning, 1,
-                    "Not importing directory '%s': missing __init__.py",
-                    buf);
-                if (warnerr)
+                    "Not importing directory '%U': missing __init__.py",
+                    unicode);
+                if (warnerr) {
+                    Py_DECREF(unicode);
                     return NULL;
+                }
             }
         }
+        Py_DECREF(unicode);
 #endif
 #if defined(PYOS_OS2)
         /* take a snapshot of the module spec for restoration

From python-checkins at python.org  Fri Jul  9 01:33:01 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:01 +0200 (CEST)
Subject: [Python-checkins] r82698 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233301.E44C7EEB62@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:01 2010
New Revision: 82698

Log:
create stat_unicode()

NullImporter does now support unicode path

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:33:01 2010
@@ -1632,6 +1632,7 @@
 
 static int case_ok(char *, Py_ssize_t, Py_ssize_t, char *);
 static int find_init_module(char *); /* Forward */
+static int stat_unicode(PyObject *unicode, struct stat *statbuf);
 static struct filedescr importhookdescr = {"", "", IMP_HOOK};
 
 static struct filedescr *
@@ -1831,7 +1832,7 @@
         unicode = PyUnicode_DecodeFSDefault(buf);
         if (unicode == NULL)
             return NULL;
-        if (stat(buf, &statbuf) == 0 &&         /* it exists */
+        if (stat_unicode(unicode, &statbuf) == 0 &&         /* it exists */
             S_ISDIR(statbuf.st_mode) &&         /* it's a directory */
             case_ok(buf, len, namelen, name)) { /* case matches */
             if (find_init_module(buf)) { /* and has __init__.py */
@@ -2109,6 +2110,15 @@
 
 
 #ifdef HAVE_STAT
+static int
+stat_unicode(PyObject *unicode, struct stat *statbuf)
+{
+    char *pathstr = _PyUnicode_AsString(unicode);
+    if (pathstr == NULL)
+        return 1;
+    return stat(pathstr, statbuf);
+}
+
 /* Helper to look for __init__.py or __init__.py[co] in potential package */
 static int
 find_init_module(char *buf)
@@ -3774,19 +3784,16 @@
 static int
 NullImporter_init(NullImporter *self, PyObject *args, PyObject *kwds)
 {
-    char *path;
-    Py_ssize_t pathlen;
+    PyObject *pathobj;
 
     if (!_PyArg_NoKeywords("NullImporter()", kwds))
         return -1;
 
-    if (!PyArg_ParseTuple(args, "es:NullImporter",
-                          Py_FileSystemDefaultEncoding, &path))
+    if (!PyArg_ParseTuple(args, "U:NullImporter",
+                          &pathobj))
         return -1;
 
-    pathlen = strlen(path);
-    if (pathlen == 0) {
-        PyMem_Free(path);
+    if (PyUnicode_GET_SIZE(pathobj) == 0) {
         PyErr_SetString(PyExc_ImportError, "empty pathname");
         return -1;
     } else {
@@ -3794,8 +3801,7 @@
         struct stat statbuf;
         int rv;
 
-        rv = stat(path, &statbuf);
-        PyMem_Free(path);
+        rv = stat_unicode(pathobj, &statbuf);
         if (rv == 0) {
             /* it exists */
             if (S_ISDIR(statbuf.st_mode)) {
@@ -3807,12 +3813,14 @@
         }
 #else /* MS_WINDOWS */
         DWORD rv;
+        char *path;
+        /* FIXME: use PyUnicode_AsWideChar() and GetFileAttributesW() */
+        path = _PyUnicode_AsString(pathobj);
         /* see issue1293 and issue3677:
          * stat() on Windows doesn't recognise paths like
          * "e:\\shared\\" and "\\\\whiterab-c2znlh\\shared" as dirs.
          */
         rv = GetFileAttributesA(path);
-        PyMem_Free(path);
         if (rv != INVALID_FILE_ATTRIBUTES) {
             /* it exists */
             if (rv & FILE_ATTRIBUTE_DIRECTORY) {

From python-checkins at python.org  Fri Jul  9 01:33:03 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:03 +0200 (CEST)
Subject: [Python-checkins] r82699 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233303.65C70D6C1@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:03 2010
New Revision: 82699

Log:
create fopen_unicode()

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:33:03 2010
@@ -136,6 +136,13 @@
     {0, 0}
 };
 
+/* Forward */
+static FILE* fopen_unicode(PyObject *unicode, const char *mode);
+#ifdef HAVE_STAT
+static int stat_unicode(PyObject *unicode, struct stat *statbuf);
+static int find_init_module(char *);
+#endif
+
 
 /* Initialize things */
 
@@ -1061,14 +1068,8 @@
     FILE *fp;
     long magic;
     long pyc_mtime;
-    char *cpathname;
-
-    /* FIXME: use PyUnicode_EncodeFSDefault() */
-    cpathname = _PyUnicode_AsString(cpathobj);
-    if (cpathname == NULL)
-        return NULL;
 
-    fp = fopen(cpathname, "rb");
+    fp = fopen_unicode(cpathobj, "rb");
     if (fp == NULL)
         return NULL;
     magic = PyMarshal_ReadLongFromFile(fp);
@@ -1411,7 +1412,6 @@
 static PyObject *
 get_sourcefile(PyObject *fileobj)
 {
-    char *py;
     PyObject *pyobj;
     Py_UNICODE *file;
     Py_ssize_t len;
@@ -1448,11 +1448,7 @@
             return NULL;
     }
 
-    /* FIXME: use PyUnicode_EncodeFSDefault() */
-    py = _PyUnicode_AsString(pyobj);
-    if (py == NULL)
-        return NULL;
-    if (stat(py, &statbuf) == 0 &&
+    if (stat_unicode(pyobj, &statbuf) == 0 &&
         S_ISREG(statbuf.st_mode)) {
         return pyobj;
     }
@@ -1631,8 +1627,6 @@
 #endif
 
 static int case_ok(char *, Py_ssize_t, Py_ssize_t, char *);
-static int find_init_module(char *); /* Forward */
-static int stat_unicode(PyObject *unicode, struct stat *statbuf);
 static struct filedescr importhookdescr = {"", "", IMP_HOOK};
 
 static struct filedescr *
@@ -1887,20 +1881,23 @@
             }
 #endif /* PYOS_OS2 */
             strcpy(buf+len, fdp->suffix);
+            unicode = PyUnicode_DecodeFSDefault(buf);
             if (Py_VerboseFlag > 1)
-                PySys_FormatStderr("# trying %s\n", buf);
+                PySys_FormatStderr("# trying %U\n", unicode);
             filemode = fdp->mode;
             if (filemode[0] == 'U')
                 filemode = "r" PY_STDIOTEXTMODE;
-            fp = fopen(buf, filemode);
+            fp = fopen_unicode(unicode, filemode);
             if (fp != NULL) {
-                if (case_ok(buf, len, namelen, name))
+                if (case_ok(buf, len, namelen, name)) {
+                    Py_DECREF(unicode);
                     break;
-                else {                   /* continue search */
+                } else {                   /* continue search */
                     fclose(fp);
                     fp = NULL;
                 }
             }
+            Py_DECREF(unicode);
 #if defined(PYOS_OS2)
             /* restore the saved snapshot */
             strcpy(buf, saved_buf);
@@ -2108,11 +2105,22 @@
 #endif
 }
 
+static FILE*
+fopen_unicode(PyObject *unicode, const char *mode)
+{
+    /* FIXME: use PyUnicode_EncodeFSDefault() */
+    char *pathstr = _PyUnicode_AsString(unicode);
+    if (pathstr == NULL)
+        return NULL;
+    return fopen(pathstr, mode);
+}
+
 
 #ifdef HAVE_STAT
 static int
 stat_unicode(PyObject *unicode, struct stat *statbuf)
 {
+    /* FIXME: use PyUnicode_EncodeFSDefault() */
     char *pathstr = _PyUnicode_AsString(unicode);
     if (pathstr == NULL)
         return 1;
@@ -2127,6 +2135,7 @@
     size_t i = save_len;
     char *pname;  /* pointer to start of __init__ */
     struct stat statbuf;
+    PyObject *unicode;
 
 /*      For calling case_ok(buf, len, namelen, name):
  *      /a/b/c/d/e/f/g/h/i/j/k/some_long_module_name.py\0
@@ -2141,26 +2150,32 @@
     buf[i++] = SEP;
     pname = buf + i;
     strcpy(pname, "__init__.py");
-    if (stat(buf, &statbuf) == 0) {
+    unicode = PyUnicode_DecodeFSDefault(buf);
+    if (stat_unicode(unicode, &statbuf) == 0) {
         if (case_ok(buf,
                     save_len + 9,               /* len("/__init__") */
                 8,                              /* len("__init__") */
                 pname)) {
             buf[save_len] = '\0';
+            Py_DECREF(unicode);
             return 1;
         }
     }
+    Py_DECREF(unicode);
     i += strlen(pname);
     strcpy(buf+i, Py_OptimizeFlag ? "o" : "c");
-    if (stat(buf, &statbuf) == 0) {
+    unicode = PyUnicode_DecodeFSDefault(buf);
+    if (stat_unicode(unicode, &statbuf) == 0) {
         if (case_ok(buf,
                     save_len + 9,               /* len("/__init__") */
                 8,                              /* len("__init__") */
                 pname)) {
             buf[save_len] = '\0';
+            Py_DECREF(unicode);
             return 1;
         }
     }
+    Py_DECREF(unicode);
     buf[save_len] = '\0';
     return 0;
 }
@@ -3445,15 +3460,10 @@
 get_file(PyObject *pathobj, PyObject *fob, char *mode)
 {
     FILE *fp;
-    char *pathname;
     if (mode[0] == 'U')
         mode = "r" PY_STDIOTEXTMODE;
     if (fob == NULL) {
-        /* FIXME: don't use _PyUnicode_AsString */
-        pathname = _PyUnicode_AsString(pathobj);
-        if (pathname == NULL)
-            return NULL;
-        fp = fopen(pathname, mode);
+        fp = fopen_unicode(pathobj, mode);
     }
     else {
         int fd = PyObject_AsFileDescriptor(fob);

From python-checkins at python.org  Fri Jul  9 01:33:04 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:04 +0200 (CEST)
Subject: [Python-checkins] r82700 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233304.DBF8BEEB20@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:04 2010
New Revision: 82700

Log:
find_init_module() uses unicode

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:33:04 2010
@@ -140,7 +140,7 @@
 static FILE* fopen_unicode(PyObject *unicode, const char *mode);
 #ifdef HAVE_STAT
 static int stat_unicode(PyObject *unicode, struct stat *statbuf);
-static int find_init_module(char *);
+static int find_init_module(PyObject *);
 #endif
 
 
@@ -1829,7 +1829,7 @@
         if (stat_unicode(unicode, &statbuf) == 0 &&         /* it exists */
             S_ISDIR(statbuf.st_mode) &&         /* it's a directory */
             case_ok(buf, len, namelen, name)) { /* case matches */
-            if (find_init_module(buf)) { /* and has __init__.py */
+            if (find_init_module(unicode)) { /* and has __init__.py */
                 *path = unicode;
                 return &fd_package;
             }
@@ -2129,54 +2129,37 @@
 
 /* Helper to look for __init__.py or __init__.py[co] in potential package */
 static int
-find_init_module(char *buf)
+find_init_module(PyObject *bufobj)
 {
-    const size_t save_len = strlen(buf);
-    size_t i = save_len;
-    char *pname;  /* pointer to start of __init__ */
     struct stat statbuf;
     PyObject *unicode;
+    char *_buf;
 
-/*      For calling case_ok(buf, len, namelen, name):
- *      /a/b/c/d/e/f/g/h/i/j/k/some_long_module_name.py\0
- *      ^                      ^                   ^    ^
- *      |--------------------- buf ---------------------|
- *      |------------------- len ------------------|
- *                             |------ name -------|
- *                             |----- namelen -----|
- */
-    if (save_len + 13 >= MAXPATHLEN)
-        return 0;
-    buf[i++] = SEP;
-    pname = buf + i;
-    strcpy(pname, "__init__.py");
-    unicode = PyUnicode_DecodeFSDefault(buf);
+    unicode = PyUnicode_FromFormat("%U%c__init__.py", bufobj, SEP);
     if (stat_unicode(unicode, &statbuf) == 0) {
-        if (case_ok(buf,
-                    save_len + 9,               /* len("/__init__") */
-                8,                              /* len("__init__") */
-                pname)) {
-            buf[save_len] = '\0';
+        /* FIXME: don't use _PyUnicode_AsString */
+        _buf = _PyUnicode_AsString(unicode);
+        if (case_ok(_buf,
+                    strlen(_buf) - 3,   /* length without .py suffix */
+                    8, "__init__")) {
             Py_DECREF(unicode);
             return 1;
         }
     }
     Py_DECREF(unicode);
-    i += strlen(pname);
-    strcpy(buf+i, Py_OptimizeFlag ? "o" : "c");
-    unicode = PyUnicode_DecodeFSDefault(buf);
+
+    unicode = PyUnicode_FromFormat("%U%c__init__.py%c", bufobj, SEP, Py_OptimizeFlag ? "o" : "c");
     if (stat_unicode(unicode, &statbuf) == 0) {
-        if (case_ok(buf,
-                    save_len + 9,               /* len("/__init__") */
-                8,                              /* len("__init__") */
-                pname)) {
-            buf[save_len] = '\0';
+        /* FIXME: don't use _PyUnicode_AsString */
+        _buf = _PyUnicode_AsString(unicode);
+        if (case_ok(_buf,
+                    strlen(_buf) - 3,   /* length without .pyc/.pyo suffix */
+                    8, "__init__")) {
             Py_DECREF(unicode);
             return 1;
         }
     }
     Py_DECREF(unicode);
-    buf[save_len] = '\0';
     return 0;
 }
 

From python-checkins at python.org  Fri Jul  9 01:33:06 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:06 +0200 (CEST)
Subject: [Python-checkins] r82701 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233306.51DB1EEB62@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:06 2010
New Revision: 82701

Log:
case_ok() now accepts an object

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:33:06 2010
@@ -1626,7 +1626,7 @@
                                         char *, Py_ssize_t);
 #endif
 
-static int case_ok(char *, Py_ssize_t, Py_ssize_t, char *);
+static int case_ok(PyObject *, Py_ssize_t, Py_ssize_t, char *);
 static struct filedescr importhookdescr = {"", "", IMP_HOOK};
 
 static struct filedescr *
@@ -1828,7 +1828,7 @@
             return NULL;
         if (stat_unicode(unicode, &statbuf) == 0 &&         /* it exists */
             S_ISDIR(statbuf.st_mode) &&         /* it's a directory */
-            case_ok(buf, len, namelen, name)) { /* case matches */
+            case_ok(unicode, 0, namelen, name)) { /* case matches */
             if (find_init_module(unicode)) { /* and has __init__.py */
                 *path = unicode;
                 return &fd_package;
@@ -1889,7 +1889,7 @@
                 filemode = "r" PY_STDIOTEXTMODE;
             fp = fopen_unicode(unicode, filemode);
             if (fp != NULL) {
-                if (case_ok(buf, len, namelen, name)) {
+                if (case_ok(unicode, strlen(fdp->suffix), namelen, name)) {
                     Py_DECREF(unicode);
                     break;
                 } else {                   /* continue search */
@@ -1996,8 +1996,12 @@
 #endif
 
 static int
-case_ok(char *buf, Py_ssize_t len, Py_ssize_t namelen, char *name)
+case_ok(PyObject *bufobj, Py_ssize_t lendelta, Py_ssize_t namelen, char *name)
 {
+    /* FIXME: don't use _PyUnicode_AsString */
+    char *buf = _PyUnicode_AsString(bufobj);
+    Py_ssize_t len = strlen(buf) - lendelta;
+
 /* Pick a platform-specific implementation; the sequence of #if's here should
  * match the sequence just above.
  */
@@ -2133,14 +2137,11 @@
 {
     struct stat statbuf;
     PyObject *unicode;
-    char *_buf;
 
     unicode = PyUnicode_FromFormat("%U%c__init__.py", bufobj, SEP);
     if (stat_unicode(unicode, &statbuf) == 0) {
-        /* FIXME: don't use _PyUnicode_AsString */
-        _buf = _PyUnicode_AsString(unicode);
-        if (case_ok(_buf,
-                    strlen(_buf) - 3,   /* length without .py suffix */
+        if (case_ok(unicode,
+                    3,   /* ignore ".py" suffix */
                     8, "__init__")) {
             Py_DECREF(unicode);
             return 1;
@@ -2150,10 +2151,8 @@
 
     unicode = PyUnicode_FromFormat("%U%c__init__.py%c", bufobj, SEP, Py_OptimizeFlag ? "o" : "c");
     if (stat_unicode(unicode, &statbuf) == 0) {
-        /* FIXME: don't use _PyUnicode_AsString */
-        _buf = _PyUnicode_AsString(unicode);
-        if (case_ok(_buf,
-                    strlen(_buf) - 3,   /* length without .pyc/.pyo suffix */
+        if (case_ok(unicode,
+                    4,   /* ignore ".pyc" / ".pyo" suffix */
                     8, "__init__")) {
             Py_DECREF(unicode);
             return 1;

From python-checkins at python.org  Fri Jul  9 01:33:07 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:07 +0200 (CEST)
Subject: [Python-checkins] r82702 - in python/branches/import_unicode:
	PC/import_nt.c Python/import.c
Message-ID: <20100708233307.C8DCAEEBCF@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:07 2010
New Revision: 82702

Log:
Move buffer into PyWin_FindRegisteredModule()

The function now returns the path as a PyUnicodeObject*

Modified:
   python/branches/import_unicode/PC/import_nt.c
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/PC/import_nt.c
==============================================================================
--- python/branches/import_unicode/PC/import_nt.c	(original)
+++ python/branches/import_unicode/PC/import_nt.c	Fri Jul  9 01:33:07 2010
@@ -17,9 +17,11 @@
 
 FILE *PyWin_FindRegisteredModule(const char *moduleName,
                                  struct filedescr **ppFileDesc,
-                                 char *pathBuf,
-                                 Py_ssize_t pathLen)
+                                 PyObject **pathobj)
 {
+    /* FIXME: use wchar_t* type */
+    char pathBuf[MAXPATHLEN+1];
+    Py_ssize_t pathLen = sizeof(pathBuf);
     char *moduleKey;
     const char keyPrefix[] = "Software\\Python\\PythonCore\\";
     const char keySuffix[] = "\\Modules\\";
@@ -45,6 +47,9 @@
                      sizeof(keySuffix) +
                      strlen(moduleName) +
                      sizeof(debugString) - 1;
+
+    *pathobj = NULL;
+
     /* alloca == no free required, but memory only local to fn,
      * also no heap fragmentation!
      */
@@ -80,7 +85,14 @@
     if (fdp->suffix == NULL)
         return NULL;
     fp = fopen(pathBuf, fdp->mode);
-    if (fp != NULL)
+    if (fp == NULL)
+        return NULL;
+    *pathobj = PyUnicode_DecodeFSDefault(pathBuf);
+    if (*pathobj != NULL)
         *ppFileDesc = fdp;
+    else {
+        fclose(fp);
+        fp = NULL;
+    }
     return fp;
 }

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:33:07 2010
@@ -1623,7 +1623,7 @@
 
 #ifdef MS_COREDLL
 extern FILE *PyWin_FindRegisteredModule(const char *, struct filedescr **,
-                                        char *, Py_ssize_t);
+                                        PyObject **);
 #endif
 
 static int case_ok(PyObject *, Py_ssize_t, Py_ssize_t, char *);
@@ -1707,8 +1707,7 @@
 
     if (search_path == NULL) {
 #ifdef MS_COREDLL
-        /* FIXME: use buf buffer */
-        char bbuf[MAXPATHLEN+1];
+        PyObject *winpath;
 #endif
         if (is_builtin(name)) {
             *path = PyUnicode_DecodeFSDefault(name);
@@ -1717,16 +1716,14 @@
             return &fd_builtin;
         }
 #ifdef MS_COREDLL
-        fp = PyWin_FindRegisteredModule(name, &fdp, bbuf, sizeof(bbuf));
+        fp = PyWin_FindRegisteredModule(name, &fdp, &winpath);
         if (fp != NULL) {
-            *path = PyUnicode_DecodeFSDefault(bbuf);
-            if (*path == NULL) {
-                fclose(fp);
-                return NULL;
-            }
             *p_fp = fp;
+            *path = winpath;
             return fdp;
         }
+        else if (PyErr_Occurred())
+            return NULL;
 #endif
         search_path = PySys_GetObject("path");
     }

From python-checkins at python.org  Fri Jul  9 01:33:09 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:09 +0200 (CEST)
Subject: [Python-checkins] r82703 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233309.3E990EEB83@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:09 2010
New Revision: 82703

Log:
fix compiler warning (unused variable)

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:33:09 2010
@@ -1995,9 +1995,12 @@
 static int
 case_ok(PyObject *bufobj, Py_ssize_t lendelta, Py_ssize_t namelen, char *name)
 {
+#if defined(MS_WINDOWS) || defined(DJGPP) || (defined(__MACH__) && defined(__APPLE__) \
+    || defined(__CYGWIN__)) && defined(HAVE_DIRENT_H) || defined(PYOS_OS2)
     /* FIXME: don't use _PyUnicode_AsString */
     char *buf = _PyUnicode_AsString(bufobj);
     Py_ssize_t len = strlen(buf) - lendelta;
+#endif
 
 /* Pick a platform-specific implementation; the sequence of #if's here should
  * match the sequence just above.

From python-checkins at python.org  Fri Jul  9 01:33:10 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:10 +0200 (CEST)
Subject: [Python-checkins] r82704 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233310.B5E12EEBC1@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:10 2010
New Revision: 82704

Log:
only use unicode in find_module()

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:33:10 2010
@@ -1630,12 +1630,12 @@
 static struct filedescr importhookdescr = {"", "", IMP_HOOK};
 
 static struct filedescr *
-_find_module(char *fullname, char *subname, PyObject *search_path,
-            char *buf, size_t buflen, PyObject **path,
+find_module(char *fullname, char *subname, PyObject *search_path,
+            PyObject **path,
             FILE **p_fp, PyObject **p_loader)
 {
     Py_ssize_t i, npath;
-    size_t len, namelen;
+    size_t namelen;
     struct filedescr *fdp = NULL;
     char *filemode;
     FILE *fp = NULL;
@@ -1646,11 +1646,11 @@
     static struct filedescr fd_package = {"", "", PKG_DIRECTORY};
     char name[MAXPATHLEN+1];
 #if defined(PYOS_OS2)
-    size_t saved_len;
+    PyObject *unicode_saved;
     size_t saved_namelen;
-    char *saved_buf = NULL;
 #endif
     PyObject *unicode;
+    PyObject *unicode_without_suffix;
 
     *path = NULL;
     if (p_loader != NULL)
@@ -1752,40 +1752,25 @@
     npath = PyList_Size(search_path);
     namelen = strlen(name);
     for (i = 0; i < npath; i++) {
-        PyObject *v = PyList_GetItem(search_path, i);
-        PyObject *origv = v;
-        const char *base;
-        Py_ssize_t size;
-        if (!v)
-            return NULL;
-        if (PyUnicode_Check(v)) {
-            v = PyUnicode_EncodeFSDefault(v);
-            if (v == NULL)
-                return NULL;
-        }
-        else
-            continue;
+        PyObject *v;
+        Py_UNICODE *unicode_buf;
+        Py_ssize_t unicode_len;
 
-        base = PyBytes_AS_STRING(v);
-        size = PyBytes_GET_SIZE(v);
-        len = size;
-        if (len + 2 + namelen + MAXSUFFIXSIZE >= buflen) {
-            Py_DECREF(v);
-            continue; /* Too long */
-        }
-        strcpy(buf, base);
-        Py_DECREF(v);
+        v = PyList_GetItem(search_path, i);
+        if (v == NULL)
+            return NULL;
+        if (!PyUnicode_Check(v))
+            continue;
 
-        if (strlen(buf) != len) {
-            continue; /* v contains '\0' */
-        }
+        unicode_buf = PyUnicode_AS_UNICODE(v);
+        unicode_len = PyUnicode_GET_SIZE(v);
 
         /* sys.path_hooks import hook */
         if (p_loader != NULL) {
             PyObject *importer;
 
             importer = get_path_importer(path_importer_cache,
-                                         path_hooks, origv);
+                                         path_hooks, v);
             if (importer == NULL) {
                 return NULL;
             }
@@ -1808,21 +1793,20 @@
         }
         /* no hook was found, use builtin import */
 
-        if (len > 0 && buf[len-1] != SEP
+        if (unicode_len > 0 && unicode_buf[unicode_len-1] != SEP
 #ifdef ALTSEP
-            && buf[len-1] != ALTSEP
+            && unicode_buf[unicode_len-1] != ALTSEP
 #endif
             )
-            buf[len++] = SEP;
-        strcpy(buf+len, name);
-        len += namelen;
+            unicode = PyUnicode_FromFormat("%U%c%s", v, SEP, name);
+        else
+            unicode = PyUnicode_FromFormat("%U%s", v, name);
+        if (unicode == NULL)
+            return NULL;
 
         /* Check for package import (buf holds a directory name,
            and there's an __init__ module in that directory */
 #ifdef HAVE_STAT
-        unicode = PyUnicode_DecodeFSDefault(buf);
-        if (unicode == NULL)
-            return NULL;
         if (stat_unicode(unicode, &statbuf) == 0 &&         /* it exists */
             S_ISDIR(statbuf.st_mode) &&         /* it's a directory */
             case_ok(unicode, 0, namelen, name)) { /* case matches */
@@ -1841,16 +1825,18 @@
                 }
             }
         }
-        Py_DECREF(unicode);
 #endif
+        Py_INCREF(unicode);
+        unicode_without_suffix = unicode;
 #if defined(PYOS_OS2)
         /* take a snapshot of the module spec for restoration
          * after the 8 character DLL hackery
          */
-        saved_buf = strdup(buf);
-        saved_len = len;
+        Py_INCREF(unicode_without_suffix);
+        saved_unicode = unicode_without_suffix;
         saved_namelen = namelen;
 #endif /* PYOS_OS2 */
+        Py_CLEAR(unicode);
         for (fdp = _PyImport_Filetab; fdp->suffix != NULL; fdp++) {
 #if defined(PYOS_OS2) && defined(HAVE_DYNAMIC_LOADING)
             /* OS/2 limits DLLs to 8 character names (w/o
@@ -1870,15 +1856,18 @@
                         scan++;
                 }
                 if (scan->suffix != NULL) {
+                    PyObject* truncated;
                     /* yes, so truncate the name */
                     namelen = 8;
-                    len -= strlen(subname) - namelen;
-                    buf[len] = '\0';
+                    truncated = PyUnicode_FromUnicode(
+                        PyUnicode_AS_UNICODE(unicode_without_suffix);
+                        PyUnicode_GET_SIZE(unicode_without_suffix) - (strlen(subname) - namelen));
+                    Py_DECREF(unicode_without_suffix);
+                    unicode_without_suffix = truncated;
                 }
             }
 #endif /* PYOS_OS2 */
-            strcpy(buf+len, fdp->suffix);
-            unicode = PyUnicode_DecodeFSDefault(buf);
+            unicode = PyUnicode_FromFormat("%U%s", unicode_without_suffix, fdp->suffix);
             if (Py_VerboseFlag > 1)
                 PySys_FormatStderr("# trying %U\n", unicode);
             filemode = fdp->mode;
@@ -1887,59 +1876,39 @@
             fp = fopen_unicode(unicode, filemode);
             if (fp != NULL) {
                 if (case_ok(unicode, strlen(fdp->suffix), namelen, name)) {
-                    Py_DECREF(unicode);
                     break;
                 } else {                   /* continue search */
                     fclose(fp);
                     fp = NULL;
                 }
             }
-            Py_DECREF(unicode);
 #if defined(PYOS_OS2)
             /* restore the saved snapshot */
-            strcpy(buf, saved_buf);
-            len = saved_len;
+            Py_DECREF(unicode_without_suffix);
+            Py_INCREF(saved_unicode);
+            unicode_without_suffix = saved_unicode;
             namelen = saved_namelen;
 #endif
         }
+        Py_CLEAR(unicode_without_suffix);
 #if defined(PYOS_OS2)
-        /* don't need/want the module name snapshot anymore */
-        if (saved_buf)
-        {
-            free(saved_buf);
-            saved_buf = NULL;
-        }
+        Py_DECREF(saved_unicode);
 #endif
         if (fp != NULL)
             break;
+        Py_DECREF(unicode);
     }
+    Py_XDECREF(unicode_without_suffix);
     if (fp == NULL) {
         PyErr_Format(PyExc_ImportError,
                      "No module named %.200s", name);
         return NULL;
     }
-    *path = PyUnicode_DecodeFSDefault(buf);
-    if (*path == NULL) {
-        fclose(fp);
-        return NULL;
-    }
+    *path = unicode;
     *p_fp = fp;
     return fdp;
 }
 
-static struct filedescr *
-find_module(char *fullname, char *subname, PyObject *search_path, PyObject **path,
-            FILE **p_fp, PyObject **p_loader)
-{
-    char buf[MAXPATHLEN+1];
-    struct filedescr *fd;
-    buf[0] = '\0';
-    fd = _find_module(fullname, subname, search_path,
-                      buf, sizeof(buf), path,
-                      p_fp, p_loader);
-    return fd;
-}
-
 /* Helpers for main.c
  *  Find the source file corresponding to a named module
  */

From python-checkins at python.org  Fri Jul  9 01:33:12 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:12 +0200 (CEST)
Subject: [Python-checkins] r82705 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233312.3F77FF1C8@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:12 2010
New Revision: 82705

Log:
remove useless name buffer in find_module()

add const to fullname and name arguments of find_module()

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:33:12 2010
@@ -1461,9 +1461,9 @@
 
 /* Forward */
 static PyObject *load_module(char *, FILE *, PyObject *, int, PyObject *);
-static struct filedescr *find_module(char *, char *, PyObject *,
+static struct filedescr *find_module(const char *, const char *, PyObject *,
                                      PyObject **, FILE **, PyObject **);
-static struct _frozen * find_frozen(char *);
+static struct _frozen * find_frozen(const char *);
 
 /* Load a package and return its module object WITH INCREMENTED
    REFERENCE COUNT */
@@ -1524,7 +1524,7 @@
 /* Helper to test for built-in module */
 
 static int
-is_builtin(char *name)
+is_builtin(const char *name)
 {
     int i;
     for (i = 0; PyImport_Inittab[i].name != NULL; i++) {
@@ -1626,11 +1626,11 @@
                                         PyObject **);
 #endif
 
-static int case_ok(PyObject *, Py_ssize_t, Py_ssize_t, char *);
+static int case_ok(PyObject *, Py_ssize_t, Py_ssize_t, const char *);
 static struct filedescr importhookdescr = {"", "", IMP_HOOK};
 
 static struct filedescr *
-find_module(char *fullname, char *subname, PyObject *search_path,
+find_module(const char *fullname, const char *name, PyObject *search_path,
             PyObject **path,
             FILE **p_fp, PyObject **p_loader)
 {
@@ -1644,7 +1644,6 @@
     static struct filedescr fd_frozen = {"", "", PY_FROZEN};
     static struct filedescr fd_builtin = {"", "", C_BUILTIN};
     static struct filedescr fd_package = {"", "", PKG_DIRECTORY};
-    char name[MAXPATHLEN+1];
 #if defined(PYOS_OS2)
     PyObject *unicode_saved;
     size_t saved_namelen;
@@ -1656,12 +1655,11 @@
     if (p_loader != NULL)
         *p_loader = NULL;
 
-    if (strlen(subname) > MAXPATHLEN) {
+    if (strlen(name) > MAXPATHLEN) {
         PyErr_SetString(PyExc_OverflowError,
                         "module name is too long");
         return NULL;
     }
-    strcpy(name, subname);
 
     /* sys.meta_path import hook */
     if (p_loader != NULL) {
@@ -1845,7 +1843,7 @@
              * dynamically loaded module we're going to try,
              * truncate the name before trying
              */
-            if (strlen(subname) > 8) {
+            if (strlen(name) > 8) {
                 /* is this an attempt to load a C extension? */
                 const struct filedescr *scan;
                 scan = _PyImport_DynLoadFiletab;
@@ -1861,7 +1859,7 @@
                     namelen = 8;
                     truncated = PyUnicode_FromUnicode(
                         PyUnicode_AS_UNICODE(unicode_without_suffix);
-                        PyUnicode_GET_SIZE(unicode_without_suffix) - (strlen(subname) - namelen));
+                        PyUnicode_GET_SIZE(unicode_without_suffix) - (strlen(name) - namelen));
                     Py_DECREF(unicode_without_suffix);
                     unicode_without_suffix = truncated;
                 }
@@ -1962,7 +1960,7 @@
 #endif
 
 static int
-case_ok(PyObject *bufobj, Py_ssize_t lendelta, Py_ssize_t namelen, char *name)
+case_ok(PyObject *bufobj, Py_ssize_t lendelta, Py_ssize_t namelen, const char *name)
 {
 #if defined(MS_WINDOWS) || defined(DJGPP) || (defined(__MACH__) && defined(__APPLE__) \
     || defined(__CYGWIN__)) && defined(HAVE_DIRENT_H) || defined(PYOS_OS2)
@@ -2306,7 +2304,7 @@
 /* Frozen modules */
 
 static struct _frozen *
-find_frozen(char *name)
+find_frozen(const char *name)
 {
     struct _frozen *p;
 
@@ -2685,6 +2683,7 @@
             int error;
 
             modname_str = _PyUnicode_AsStringAndSize(modname, &len);
+            /* FIXME: don't use _PyUnicode_AsStringAndSize or check that modname_str is not NULL */
             if (len > MAXPATHLEN) {
                 PyErr_SetString(PyExc_ValueError,
                                 "Module name too long");

From python-checkins at python.org  Fri Jul  9 01:33:13 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:13 +0200 (CEST)
Subject: [Python-checkins] r82706 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233313.A5101EEBCA@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:13 2010
New Revision: 82706

Log:
add FIXME

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:33:13 2010
@@ -970,6 +970,7 @@
 static PyObject*
 make_compiled_pathname(PyObject *pathobj, int debug)
 {
+    /* FIXME: use Py_UNICODE* instead of char* */
     char buf[MAXPATHLEN+1];
     char *pathname, *cpathname;
 
@@ -2521,6 +2522,7 @@
 import_module_level(char *name, PyObject *globals, PyObject *locals,
                     PyObject *fromlist, int level)
 {
+    /* FIXME: use unicode */
     char buf[MAXPATHLEN+1];
     Py_ssize_t buflen = 0;
     PyObject *parent, *head, *next, *tail;

From python-checkins at python.org  Fri Jul  9 01:33:15 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:15 +0200 (CEST)
Subject: [Python-checkins] r82707 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233315.2CEFAEEB79@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:14 2010
New Revision: 82707

Log:
use PyUnicode_EncodeFSDefault() in fopen_unicode() and stat_unicode()

Cleanup case_ok()

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:33:14 2010
@@ -1963,30 +1963,27 @@
 static int
 case_ok(PyObject *bufobj, Py_ssize_t lendelta, Py_ssize_t namelen, const char *name)
 {
-#if defined(MS_WINDOWS) || defined(DJGPP) || (defined(__MACH__) && defined(__APPLE__) \
-    || defined(__CYGWIN__)) && defined(HAVE_DIRENT_H) || defined(PYOS_OS2)
-    /* FIXME: don't use _PyUnicode_AsString */
-    char *buf = _PyUnicode_AsString(bufobj);
-    Py_ssize_t len = strlen(buf) - lendelta;
-#endif
-
 /* Pick a platform-specific implementation; the sequence of #if's here should
  * match the sequence just above.
  */
 
 /* MS_WINDOWS */
 #if defined(MS_WINDOWS)
+    char *buf;
     WIN32_FIND_DATA data;
     HANDLE h;
 
     if (Py_GETENV("PYTHONCASEOK") != NULL)
         return 1;
 
+    /* FIXME: use PyUnicode_AsWideChar() and FindFirstFileW() */
+    buf = _PyUnicode_AsString(bufobj);
+
     h = FindFirstFile(buf, &data);
     if (h == INVALID_HANDLE_VALUE) {
         PyErr_Format(PyExc_NameError,
-          "Can't find file for module %.100s\n(filename %.300s)",
-          name, buf);
+          "Can't find file for module %.100s\n(filename %U)",
+          name, bufobj);
         return 0;
     }
     FindClose(h);
@@ -1994,31 +1991,42 @@
 
 /* DJGPP */
 #elif defined(DJGPP)
+    char *buf;
     struct ffblk ffblk;
     int done;
 
     if (Py_GETENV("PYTHONCASEOK") != NULL)
         return 1;
 
+    /* FIXME: use PyUnicode_EncodeFSDefault() */
+    buf = _PyUnicode_AsString(bufobj);
+
     done = findfirst(buf, &ffblk, FA_ARCH|FA_RDONLY|FA_HIDDEN|FA_DIREC);
     if (done) {
         PyErr_Format(PyExc_NameError,
-          "Can't find file for module %.100s\n(filename %.300s)",
-          name, buf);
+          "Can't find file for module %.100s\n(filename %U)",
+          name, bufobj);
         return 0;
     }
     return strncmp(ffblk.ff_name, name, namelen) == 0;
 
 /* new-fangled macintosh (macosx) or Cygwin */
 #elif (defined(__MACH__) && defined(__APPLE__) || defined(__CYGWIN__)) && defined(HAVE_DIRENT_H)
+    char *buf;
+    Py_ssize_t len;
     DIR *dirp;
     struct dirent *dp;
     char dirname[MAXPATHLEN + 1];
-    const int dirlen = len - namelen - 1; /* don't want trailing SEP */
+    int dirlen;
 
     if (Py_GETENV("PYTHONCASEOK") != NULL)
         return 1;
 
+    /* FIXME: use PyUnicode_EncodeFSDefault() */
+    buf = _PyUnicode_AsString(bufobj);
+    len = strlen(buf) - lendelta;
+    dirlen = len - namelen - 1; /* don't want trailing SEP */
+
     /* Copy the dir component into dirname; substitute "." if empty */
     if (dirlen <= 0) {
         dirname[0] = '.';
@@ -2052,6 +2060,7 @@
 
 /* OS/2 */
 #elif defined(PYOS_OS2)
+    char *buf;
     HDIR hdir = 1;
     ULONG srchcnt = 1;
     FILEFINDBUF3 ffbuf;
@@ -2060,6 +2069,9 @@
     if (Py_GETENV("PYTHONCASEOK") != NULL)
         return 1;
 
+    /* FIXME: use PyUnicode_EncodeFSDefault() */
+    buf = _PyUnicode_AsString(bufobj);
+
     rc = DosFindFirst(buf,
                       &hdir,
                       FILE_READONLY | FILE_HIDDEN | FILE_SYSTEM | FILE_DIRECTORY,
@@ -2080,11 +2092,13 @@
 static FILE*
 fopen_unicode(PyObject *unicode, const char *mode)
 {
-    /* FIXME: use PyUnicode_EncodeFSDefault() */
-    char *pathstr = _PyUnicode_AsString(unicode);
-    if (pathstr == NULL)
-        return NULL;
-    return fopen(pathstr, mode);
+    FILE *f;
+    PyObject *bytes = PyUnicode_EncodeFSDefault(unicode);
+    if (bytes == NULL)
+        return NULL;
+    f = fopen(PyBytes_AS_STRING(bytes), mode);
+    Py_DECREF(bytes);
+    return f;
 }
 
 
@@ -2092,11 +2106,13 @@
 static int
 stat_unicode(PyObject *unicode, struct stat *statbuf)
 {
-    /* FIXME: use PyUnicode_EncodeFSDefault() */
-    char *pathstr = _PyUnicode_AsString(unicode);
-    if (pathstr == NULL)
-        return 1;
-    return stat(pathstr, statbuf);
+    int ret;
+    PyObject *bytes = PyUnicode_EncodeFSDefault(unicode);
+    if (bytes == NULL)
+        return -1;
+    ret = stat(PyBytes_AS_STRING(bytes), statbuf);
+    Py_DECREF(bytes);
+    return ret;
 }
 
 /* Helper to look for __init__.py or __init__.py[co] in potential package */

From python-checkins at python.org  Fri Jul  9 01:33:16 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:16 +0200 (CEST)
Subject: [Python-checkins] r82708 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233316.99D43EEBA0@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:16 2010
New Revision: 82708

Log:
write_compiled_module() uses PyUnicode_EncodeFSDefault()

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:33:16 2010
@@ -1232,14 +1232,15 @@
                       S_IWUSR | S_IWGRP | S_IWOTH);
 #endif
     int saved;
+    PyObject *cpathbytes;
     char *cpathname;
 
-    /* FIXME: use PyUnicode_EncodeFSDefault() */
-    cpathname = _PyUnicode_AsString(cpathobj);
-    if (cpathname == NULL) {
+    cpathbytes = PyUnicode_EncodeFSDefault(cpathobj);
+    if (cpathbytes == NULL) {
         PyErr_Clear();
         return;
     }
+    cpathname = PyBytes_AS_STRING(cpathbytes);
 
     /* Ensure that the __pycache__ directory exists. */
     dirpath = rightmost_sep(cpathname);
@@ -1248,6 +1249,7 @@
             PySys_WriteStderr(
                 "# no %s path found %s\n",
                 CACHEDIR, cpathname);
+        Py_DECREF(cpathbytes);
         return;
     }
     saved = *dirpath;
@@ -1258,6 +1260,7 @@
         if (Py_VerboseFlag)
             PySys_WriteStderr(
                 "# cannot create cache dir %s\n", cpathname);
+        Py_DECREF(cpathbytes);
         return;
     }
     *dirpath = saved;
@@ -1267,6 +1270,7 @@
         if (Py_VerboseFlag)
             PySys_WriteStderr(
                 "# can't create %s\n", cpathname);
+        Py_DECREF(cpathbytes);
         return;
     }
     PyMarshal_WriteLongToFile(pyc_magic, fp, Py_MARSHAL_VERSION);
@@ -1279,6 +1283,7 @@
         /* Don't keep partial file */
         fclose(fp);
         (void) unlink(cpathname);
+        Py_DECREF(cpathbytes);
         return;
     }
     /* Now write the true mtime */
@@ -1289,6 +1294,7 @@
     fclose(fp);
     if (Py_VerboseFlag)
         PySys_WriteStderr("# wrote %s\n", cpathname);
+    Py_DECREF(cpathbytes);
 }
 
 static void

From python-checkins at python.org  Fri Jul  9 01:33:18 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:18 +0200 (CEST)
Subject: [Python-checkins] r82709 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233318.139C2EE9BE@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:17 2010
New Revision: 82709

Log:
parse_source_module() uses PyUnicode_EncodeFSDefault()

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:33:17 2010
@@ -1151,16 +1151,17 @@
     PyCodeObject *co = NULL;
     mod_ty mod;
     PyCompilerFlags flags;
+    PyObject *pathbytes;
     char *pathname;
 
     PyArena *arena = PyArena_New();
     if (arena == NULL)
         return NULL;
 
-    /* FIXME: use PyUnicode_EncodeFSDefault() */
-    pathname = _PyUnicode_AsString(pathobj);
-    if (pathname == NULL)
+    pathbytes = PyUnicode_EncodeFSDefault(pathobj);
+    if (pathbytes == NULL)
         return NULL;
+    pathname = PyBytes_AS_STRING(pathbytes);
 
     flags.cf_flags = 0;
     mod = PyParser_ASTFromFile(fp, pathname, NULL,
@@ -1169,6 +1170,7 @@
     if (mod) {
         co = PyAST_Compile(mod, pathname, NULL, arena);
     }
+    Py_DECREF(pathbytes);
     PyArena_Free(arena);
     return co;
 }

From python-checkins at python.org  Fri Jul  9 01:33:19 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:19 +0200 (CEST)
Subject: [Python-checkins] r82710 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233319.7AF9DEEB83@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:19 2010
New Revision: 82710

Log:
initialize unicode_without_suffix to NULL

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:33:19 2010
@@ -1658,7 +1658,7 @@
     size_t saved_namelen;
 #endif
     PyObject *unicode;
-    PyObject *unicode_without_suffix;
+    PyObject *unicode_without_suffix = NULL;
 
     *path = NULL;
     if (p_loader != NULL)

From python-checkins at python.org  Fri Jul  9 01:33:21 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:21 +0200 (CEST)
Subject: [Python-checkins] r82711 - in python/branches/import_unicode:
	Include/Python.h Modules/main.c
Message-ID: <20100708233321.2F196EEBA0@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:21 2010
New Revision: 82711

Log:
create _Py_wchar2char()

Use it in _wfopen() (implementation for non Windows OS)

Modified:
   python/branches/import_unicode/Include/Python.h
   python/branches/import_unicode/Modules/main.c

Modified: python/branches/import_unicode/Include/Python.h
==============================================================================
--- python/branches/import_unicode/Include/Python.h	(original)
+++ python/branches/import_unicode/Include/Python.h	Fri Jul  9 01:33:21 2010
@@ -128,8 +128,10 @@
 /* _Py_Mangle is defined in compile.c */
 PyAPI_FUNC(PyObject*) _Py_Mangle(PyObject *p, PyObject *name);
 
-/* _Py_char2wchar lives in main.c */
+/* _Py_char2wchar and _Py_wchar2char lives in main.c */
 PyAPI_FUNC(wchar_t *) _Py_char2wchar(char *);
+PyAPI_FUNC(char*) _Py_wchar2char(const wchar_t *text);
+
 #ifdef __cplusplus
 }
 #endif

Modified: python/branches/import_unicode/Modules/main.c
==============================================================================
--- python/branches/import_unicode/Modules/main.c	(original)
+++ python/branches/import_unicode/Modules/main.c	Fri Jul  9 01:33:21 2010
@@ -105,20 +105,21 @@
 static FILE*
 _wfopen(const wchar_t *path, const wchar_t *mode)
 {
-    char cpath[PATH_MAX];
+    FILE *f;
+    char *cpath;
     char cmode[10];
     size_t r;
-    r = wcstombs(cpath, path, PATH_MAX);
-    if (r == (size_t)-1 || r >= PATH_MAX) {
-        errno = EINVAL;
+    cpath = _Py_wchar2char(path);
+    if (cpath == NULL)
         return NULL;
-    }
     r = wcstombs(cmode, mode, 10);
     if (r == (size_t)-1 || r >= 10) {
         errno = EINVAL;
         return NULL;
     }
-    return fopen(cpath, cmode);
+    f = fopen(cpath, cmode);
+    PyMem_Free(cpath);
+    return f;
 }
 #endif
 
@@ -711,6 +712,43 @@
     *argv = orig_argv;
 }
 
+char*
+_Py_wchar2char(const wchar_t *text)
+{
+    char *result, *bytes;
+    size_t i, len, converted, size;
+    wchar_t c;
+
+    len = wcslen(text);
+    /* FIXME: use better heuristic for the buffer size */
+    size = len * 10 + 1; /* +1 for the nul byte at the end */
+    result = PyMem_Malloc(size);
+    if (result == NULL)
+        return NULL;
+
+    bytes = result;
+    for (i=0; i < len; i++) {
+        c = text[i];
+        if (c >= 0xd800 && c <= 0xdfff) {
+            /* Surrogate character */
+            *bytes++ = c - 0xdc00;
+            size--;
+            continue;
+        } else {
+            wchar_t buf[2];
+            buf[0] = c;
+            buf[1] = 0;
+            converted = wcstombs(bytes, buf, size);
+            if (converted == (size_t)-1 || converted == 0) {
+                PyMem_Free(result);
+                return NULL;
+            }
+            bytes += converted;
+            size -= converted;
+        }
+    }
+    return result;
+}
 
 wchar_t*
 _Py_char2wchar(char* arg)

From python-checkins at python.org  Fri Jul  9 01:33:22 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:22 +0200 (CEST)
Subject: [Python-checkins] r82712 -
	python/branches/import_unicode/Lib/test/test_import.py
Message-ID: <20100708233322.DFC53EEBCA@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:22 2010
New Revision: 82712

Log:
Fix test_import for non ASCII path

Skip test_import_by_filename is the path is not encodable to the file
system encoding

Modified:
   python/branches/import_unicode/Lib/test/test_import.py

Modified: python/branches/import_unicode/Lib/test/test_import.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_import.py	(original)
+++ python/branches/import_unicode/Lib/test/test_import.py	Fri Jul  9 01:33:22 2010
@@ -289,6 +289,13 @@
 
     def test_import_by_filename(self):
         path = os.path.abspath(TESTFN)
+        encoding = sys.getfilesystemencoding()
+        try:
+            path.encode(encoding)
+        except UnicodeEncodeError:
+            self.skipTest('path is not encodable to {}'.format(encoding))
+        else:
+            pass
         with self.assertRaises(ImportError) as c:
             __import__(path)
         self.assertEqual("Import by filename is not supported.",

From python-checkins at python.org  Fri Jul  9 01:33:24 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:24 +0200 (CEST)
Subject: [Python-checkins] r82713 -
	python/branches/import_unicode/Python/_warnings.c
Message-ID: <20100708233324.66D12EEAF0@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:24 2010
New Revision: 82713

Log:
warnings: use unicode for the filename

Modified:
   python/branches/import_unicode/Python/_warnings.c

Modified: python/branches/import_unicode/Python/_warnings.c
==============================================================================
--- python/branches/import_unicode/Python/_warnings.c	(original)
+++ python/branches/import_unicode/Python/_warnings.c	Fri Jul  9 01:33:24 2010
@@ -498,23 +498,21 @@
     *filename = PyDict_GetItemString(globals, "__file__");
     if (*filename != NULL) {
         Py_ssize_t len = PyUnicode_GetSize(*filename);
-        const char *file_str = _PyUnicode_AsString(*filename);
-            if (file_str == NULL || (len < 0 && PyErr_Occurred()))
-            goto handle_error;
+        Py_UNICODE *unicode = PyUnicode_AS_UNICODE(*filename);
 
         /* if filename.lower().endswith((".pyc", ".pyo")): */
         if (len >= 4 &&
-            file_str[len-4] == '.' &&
-            tolower(file_str[len-3]) == 'p' &&
-            tolower(file_str[len-2]) == 'y' &&
-            (tolower(file_str[len-1]) == 'c' ||
-                tolower(file_str[len-1]) == 'o'))
+            unicode[len-4] == '.' &&
+            Py_UNICODE_TOLOWER(unicode[len-3]) == 'p' &&
+            Py_UNICODE_TOLOWER(unicode[len-2]) == 'y' &&
+            (Py_UNICODE_TOLOWER(unicode[len-1]) == 'c' ||
+                Py_UNICODE_TOLOWER(unicode[len-1]) == 'o'))
         {
-            *filename = PyUnicode_FromStringAndSize(file_str, len-1);
-                if (*filename == NULL)
-                        goto handle_error;
-            }
-            else
+            *filename = PyUnicode_FromUnicode(unicode, len-1);
+            if (*filename == NULL)
+                goto handle_error;
+        }
+        else
             Py_INCREF(*filename);
     }
     else {

From python-checkins at python.org  Fri Jul  9 01:33:25 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:25 +0200 (CEST)
Subject: [Python-checkins] r82714 -
	python/branches/import_unicode/Python/importdl.c
Message-ID: <20100708233325.D65B4EE9A2@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:25 2010
New Revision: 82714

Log:
importdl: use PyUnicode_EncodeFSDefault()

Instead of the evil _PyUnicode_AsString()

Modified:
   python/branches/import_unicode/Python/importdl.c

Modified: python/branches/import_unicode/Python/importdl.c
==============================================================================
--- python/branches/import_unicode/Python/importdl.c	(original)
+++ python/branches/import_unicode/Python/importdl.c	Fri Jul  9 01:33:25 2010
@@ -22,15 +22,12 @@
 _PyImport_LoadDynamicModule(char *name, PyObject *path, FILE *fp)
 {
     PyObject *m;
-    char *pathname;
+    PyObject *pathbytes;
     char *lastdot, *shortname, *packagecontext, *oldcontext;
     dl_funcptr p0;
     PyObject* (*p)(void);
     struct PyModuleDef *def;
 
-    /* FIXME: don't use _PyUnicode_AsString */
-    pathname = _PyUnicode_AsString(path);
-
     if ((m = _PyImport_FindExtensionUnicode(name, path)) != NULL) {
         Py_INCREF(m);
         return m;
@@ -45,7 +42,12 @@
         shortname = lastdot+1;
     }
 
-    p0 = _PyImport_GetDynLoadFunc(name, shortname, pathname, fp);
+    /* FIXME: pass path, not pathname, at least to the Windows implementation */
+    pathbytes = PyUnicode_EncodeFSDefault(path);
+    if (pathbytes == NULL)
+        return NULL;
+    p0 = _PyImport_GetDynLoadFunc(name, shortname, PyBytes_AsString(pathbytes), fp);
+    Py_DECREF(pathbytes);
     p = (PyObject*(*)(void))p0;
     if (PyErr_Occurred())
         return NULL;
@@ -84,9 +86,9 @@
     if (_PyImport_FixupExtensionUnicode(m, name, path) < 0)
         return NULL;
     if (Py_VerboseFlag)
-        PySys_WriteStderr(
-            "import %s # dynamically loaded from %s\n",
-            name, pathname);
+        PySys_FormatStderr(
+            "import %s # dynamically loaded from %U\n",
+            name, path);
     return m;
 }
 

From python-checkins at python.org  Fri Jul  9 01:33:27 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:27 +0200 (CEST)
Subject: [Python-checkins] r82715 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100708233327.4A6EBEE995@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:27 2010
New Revision: 82715

Log:
Use PyUnicode_EncodeFSDefault() in make_xxx_pathname()

instead of _PyUnicode_AsString()

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul  9 01:33:27 2010
@@ -972,12 +972,18 @@
 {
     /* FIXME: use Py_UNICODE* instead of char* */
     char buf[MAXPATHLEN+1];
+    PyObject *pathbytes;
     char *pathname, *cpathname;
 
-    /* FIXME: don't use _PyUnicode_AsString */
-    pathname = _PyUnicode_AsString(pathobj);
+    pathbytes = PyUnicode_EncodeFSDefault(pathobj);
+    if (pathbytes == NULL)
+        return NULL;
+
+    pathname = strdup(PyBytes_AsString(pathbytes));
+    Py_DECREF(pathbytes);
 
     cpathname = _make_compiled_pathname(pathname, buf, sizeof(buf), debug);
+    free(pathname);
     if (cpathname != NULL)
         return PyUnicode_DecodeFSDefault(cpathname);
     else
@@ -1044,12 +1050,20 @@
 {
     /* FIXME: use Py_UNICODE* instead of char* */
     char buf[MAXPATHLEN + 1];
+    PyObject *pathbytes;
     char *pathname, *cpathname;
 
-    /* FIXME: don't use _PyUnicode_AsString */
-    pathname = _PyUnicode_AsString(pathobj);
+    pathbytes = PyUnicode_EncodeFSDefault(pathobj);
+    if (pathbytes == NULL)
+        return NULL;
+
+    pathname = strdup(PyBytes_AsString(pathbytes));
+    Py_DECREF(pathbytes);
+    if (pathname == NULL)
+        return NULL;
 
     cpathname = _make_source_pathname(pathname, buf);
+    free(pathname);
     if (cpathname != NULL)
         return PyUnicode_DecodeFSDefault(cpathname);
     else

From python-checkins at python.org  Fri Jul  9 01:33:28 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:28 +0200 (CEST)
Subject: [Python-checkins] r82716 -
	python/branches/import_unicode/Python/ceval.c
Message-ID: <20100708233328.AC29EEEBA6@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:28 2010
New Revision: 82716

Log:
ceval: remove filename debug variable

_PyUnicode_AsString() may raise an exception

Modified:
   python/branches/import_unicode/Python/ceval.c

Modified: python/branches/import_unicode/Python/ceval.c
==============================================================================
--- python/branches/import_unicode/Python/ceval.c	(original)
+++ python/branches/import_unicode/Python/ceval.c	Fri Jul  9 01:33:28 2010
@@ -796,10 +796,6 @@
     unsigned char *first_instr;
     PyObject *names;
     PyObject *consts;
-#if defined(Py_DEBUG) || defined(LLTRACE)
-    /* Make it easier to find out where we are with a debugger */
-    char *filename;
-#endif
 
 /* Computed GOTOs, or
        the-optimization-commonly-but-improperly-known-as-"threaded code"
@@ -1199,9 +1195,6 @@
 #ifdef LLTRACE
     lltrace = PyDict_GetItemString(f->f_globals, "__lltrace__") != NULL;
 #endif
-#if defined(Py_DEBUG) || defined(LLTRACE)
-    filename = _PyUnicode_AsString(co->co_filename);
-#endif
 
     why = WHY_NOT;
     err = 0;

From python-checkins at python.org  Fri Jul  9 01:33:30 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:30 +0200 (CEST)
Subject: [Python-checkins] r82717 -
	python/branches/import_unicode/Parser/tokenizer.c
Message-ID: <20100708233330.14872EEB2A@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:29 2010
New Revision: 82717

Log:
tokenizer uses unicode filename

Modified:
   python/branches/import_unicode/Parser/tokenizer.c

Modified: python/branches/import_unicode/Parser/tokenizer.c
==============================================================================
--- python/branches/import_unicode/Parser/tokenizer.c	(original)
+++ python/branches/import_unicode/Parser/tokenizer.c	Fri Jul  9 01:33:29 2010
@@ -467,10 +467,14 @@
     if (io == NULL)
         goto cleanup;
 
-    if (tok->filename)
-        stream = PyObject_CallMethod(io, "open", "ssis",
-                                     tok->filename, "r", -1, enc);
-    else
+    if (tok->filename) {
+        PyObject *filename = PyUnicode_DecodeFSDefault(tok->filename);
+        if (filename == NULL)
+            goto cleanup;
+        stream = PyObject_CallMethod(io, "open", "Osis",
+                                     filename, "r", -1, enc);
+        Py_DECREF(filename);
+    } else
         stream = PyObject_CallMethod(io, "open", "isisOOO",
                         fileno(tok->fp), "r", -1, enc, Py_None, Py_None, Py_False);
     if (stream == NULL)

From python-checkins at python.org  Fri Jul  9 01:33:31 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:31 +0200 (CEST)
Subject: [Python-checkins] r82718 -
	python/branches/import_unicode/Modules/zipimport.c
Message-ID: <20100708233331.95A43EEB35@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:31 2010
New Revision: 82718

Log:
zipimporter uses PyUnicode_FSConverter()

to use the file system encoding, not utf-8

Modified:
   python/branches/import_unicode/Modules/zipimport.c

Modified: python/branches/import_unicode/Modules/zipimport.c
==============================================================================
--- python/branches/import_unicode/Modules/zipimport.c	(original)
+++ python/branches/import_unicode/Modules/zipimport.c	Fri Jul  9 01:33:31 2010
@@ -60,26 +60,31 @@
 static int
 zipimporter_init(ZipImporter *self, PyObject *args, PyObject *kwds)
 {
+    PyObject *pathbytes;
     char *path, *p, *prefix, buf[MAXPATHLEN+2];
     size_t len;
 
     if (!_PyArg_NoKeywords("zipimporter()", kwds))
         return -1;
 
-    if (!PyArg_ParseTuple(args, "s:zipimporter", &path))
+    if (!PyArg_ParseTuple(args, "O&:zipimporter", PyUnicode_FSConverter, &pathbytes))
         return -1;
 
-    len = strlen(path);
+    len = PyBytes_GET_SIZE(pathbytes);
     if (len == 0) {
         PyErr_SetString(ZipImportError, "archive path is empty");
+        Py_DECREF(pathbytes);
         return -1;
     }
+    path = PyBytes_AsString(pathbytes);
     if (len >= MAXPATHLEN) {
         PyErr_SetString(ZipImportError,
                         "archive path too long");
+        Py_DECREF(pathbytes);
         return -1;
     }
     strcpy(buf, path);
+    Py_DECREF(pathbytes);
 
 #ifdef ALTSEP
     for (p = buf; *p; p++) {

From python-checkins at python.org  Fri Jul  9 01:33:33 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:33 +0200 (CEST)
Subject: [Python-checkins] r82719 -
	python/branches/import_unicode/Modules/getpath.c
Message-ID: <20100708233333.2DC22EEAF6@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:32 2010
New Revision: 82719

Log:
fix calculate_path() to use surrogates

Modified:
   python/branches/import_unicode/Modules/getpath.c

Modified: python/branches/import_unicode/Modules/getpath.c
==============================================================================
--- python/branches/import_unicode/Modules/getpath.c	(original)
+++ python/branches/import_unicode/Modules/getpath.c	Fri Jul  9 01:33:32 2010
@@ -139,58 +139,58 @@
 static int
 _wstat(const wchar_t* path, struct stat *buf)
 {
-    char fname[PATH_MAX];
-    size_t res = wcstombs(fname, path, sizeof(fname));
-    if (res == (size_t)-1) {
-        errno = EINVAL;
+    int ret;
+    char *fname;
+    fname = _Py_wchar2char(path);
+    if (fname == NULL)
         return -1;
-    }
-    return stat(fname, buf);
+    ret = stat(fname, buf);
+    PyMem_Free(fname);
+    return ret;
 }
 #endif
 
-#ifndef MS_WINDOWS
 static wchar_t*
-_wgetcwd(wchar_t *buf, size_t size)
+_Py_wgetcwd(wchar_t *buf, size_t size)
 {
+#ifdef MS_WINDOWS
+    wchar_t buffer[MAXPATHLEN + 1];
+    wchar_t *ret, *result;
+    size_t len;
+    ret = _wgetcwd(buffer, MAXPATHLEN);
+    if (ret == NULL)
+        return NULL;
+    len = wcslen(buffer);
+    result = PyMem_Malloc(len + 1);
+    if (result == NULL)
+        return NULL;
+    wcscpy(result, buffer);
+    return result;
+#else
     char fname[PATH_MAX];
     if (getcwd(fname, PATH_MAX) == NULL)
         return NULL;
-    if (mbstowcs(buf, fname, size) >= size) {
-        errno = ERANGE;
-        return NULL;
-    }
-    return buf;
-}
+    return _Py_char2wchar(fname);
 #endif
+}
 
 #ifdef HAVE_READLINK
-int
-_Py_wreadlink(const wchar_t *path, wchar_t *buf, size_t bufsiz)
+wchar_t*
+_Py_wreadlink(const wchar_t *path)
 {
     char cbuf[PATH_MAX];
     char cpath[PATH_MAX];
     int res;
     size_t r1 = wcstombs(cpath, path, PATH_MAX);
-    if (r1 == (size_t)-1 || r1 >= PATH_MAX) {
-        errno = EINVAL;
-        return -1;
-    }
+    if (r1 == (size_t)-1 || r1 >= PATH_MAX)
+        return NULL;
     res = (int)readlink(cpath, cbuf, PATH_MAX);
     if (res == -1)
-        return -1;
-    if (res == PATH_MAX) {
-        errno = EINVAL;
-        return -1;
-    }
+        return NULL;
+    if (res == PATH_MAX)
+        return NULL;
     cbuf[res] = '\0'; /* buf will be null terminated */
-    r1 = mbstowcs(buf, cbuf, bufsiz);
-    if (r1 == -1) {
-        errno = EINVAL;
-        return -1;
-    }
-    return (int)r1;
-
+    return _Py_char2wchar(cbuf);
 }
 #endif
 
@@ -295,11 +295,19 @@
     if (p[0] == SEP)
         wcscpy(path, p);
     else {
-        _wgetcwd(path, MAXPATHLEN);
+        wchar_t *cwd;
+        cwd = _Py_wgetcwd(path, MAXPATHLEN);
+        if (cwd == NULL)
+            return /* FIXME: return an error */; 
+        if (wcslen(cwd) >= MAXPATHLEN)
+            return /* FIXME: return an error */; 
+        wcscpy(path, cwd);
+        PyMem_Free(cwd);
         if (p[0] == '.' && p[1] == SEP)
             p += 2;
         joinpath(path, p);
     }
+    return /* FIXME: return ok */;
 }
 
 /* absolutize() requires that path be allocated at least MAXPATHLEN+1 bytes. */
@@ -563,19 +571,20 @@
 
 #if HAVE_READLINK
     {
-        wchar_t tmpbuffer[MAXPATHLEN+1];
-        int linklen = _Py_wreadlink(progpath, tmpbuffer, MAXPATHLEN);
-        while (linklen != -1) {
-            if (tmpbuffer[0] == SEP)
-                /* tmpbuffer should never be longer than MAXPATHLEN,
+        wchar_t *link;
+        link = _Py_wreadlink(progpath);
+        while (link != NULL) {
+            if (link[0] == SEP)
+                /* link should never be longer than MAXPATHLEN,
                    but extra check does not hurt */
-                wcsncpy(argv0_path, tmpbuffer, MAXPATHLEN);
+                wcsncpy(argv0_path, link, MAXPATHLEN);
             else {
                 /* Interpret relative to progpath */
                 reduce(argv0_path);
-                joinpath(argv0_path, tmpbuffer);
+                joinpath(argv0_path, link);
             }
-            linklen = _Py_wreadlink(argv0_path, tmpbuffer, MAXPATHLEN);
+            PyMem_Free(link);
+            link = _Py_wreadlink(argv0_path);
         }
     }
 #endif /* HAVE_READLINK */

From python-checkins at python.org  Fri Jul  9 01:33:34 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:34 +0200 (CEST)
Subject: [Python-checkins] r82720 -
	python/branches/import_unicode/Python/bltinmodule.c
Message-ID: <20100708233334.C4409EE9CB@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:34 2010
New Revision: 82720

Log:
compile() uses PyUnicode_FSConverter

Modified:
   python/branches/import_unicode/Python/bltinmodule.c

Modified: python/branches/import_unicode/Python/bltinmodule.c
==============================================================================
--- python/branches/import_unicode/Python/bltinmodule.c	(original)
+++ python/branches/import_unicode/Python/bltinmodule.c	Fri Jul  9 01:33:34 2010
@@ -542,7 +542,9 @@
 static PyObject *
 builtin_compile(PyObject *self, PyObject *args, PyObject *kwds)
 {
+    PyObject *code;
     char *str;
+    PyObject *fileobj;
     char *filename;
     char *startstr;
     int mode = -1;
@@ -555,11 +557,13 @@
                              "dont_inherit", NULL};
     int start[] = {Py_file_input, Py_eval_input, Py_single_input};
 
-    if (!PyArg_ParseTupleAndKeywords(args, kwds, "Oss|ii:compile",
-                                     kwlist, &cmd, &filename, &startstr,
+    if (!PyArg_ParseTupleAndKeywords(args, kwds, "OO&s|ii:compile", kwlist,
+                                     &cmd, PyUnicode_FSConverter, &fileobj, &startstr,
                                      &supplied_flags, &dont_inherit))
         return NULL;
 
+    filename = PyBytes_AsString(fileobj);
+
     cf.cf_flags = supplied_flags | PyCF_SOURCE_IS_UTF8;
 
     if (supplied_flags &
@@ -567,6 +571,7 @@
     {
         PyErr_SetString(PyExc_ValueError,
                         "compile(): unrecognised flags");
+        Py_DECREF(fileobj);
         return NULL;
     }
     /* XXX Warn if (supplied_flags & PyCF_MASK_OBSOLETE) != 0? */
@@ -584,12 +589,15 @@
     else {
         PyErr_SetString(PyExc_ValueError,
                         "compile() arg 3 must be 'exec', 'eval' or 'single'");
+        Py_DECREF(fileobj);
         return NULL;
     }
 
     is_ast = PyAST_Check(cmd);
-    if (is_ast == -1)
+    if (is_ast == -1) {
+        Py_DECREF(fileobj);
         return NULL;
+    }
     if (is_ast) {
         PyObject *result;
         if (supplied_flags & PyCF_ONLY_AST) {
@@ -604,20 +612,26 @@
             mod = PyAST_obj2mod(cmd, arena, mode);
             if (mod == NULL) {
                 PyArena_Free(arena);
+                Py_DECREF(fileobj);
                 return NULL;
             }
             result = (PyObject*)PyAST_Compile(mod, filename,
                                               &cf, arena);
             PyArena_Free(arena);
         }
+        Py_DECREF(fileobj);
         return result;
     }
 
     str = source_as_string(cmd, "compile", "string, bytes, AST or code", &cf);
-    if (str == NULL)
+    if (str == NULL) {
+        Py_DECREF(fileobj);
         return NULL;
+    }
 
-    return Py_CompileStringFlags(str, filename, start[mode], &cf);
+    code = Py_CompileStringFlags(str, filename, start[mode], &cf);
+    Py_DECREF(fileobj);
+    return code;
 }
 
 PyDoc_STRVAR(compile_doc,

From python-checkins at python.org  Fri Jul  9 01:33:36 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:36 +0200 (CEST)
Subject: [Python-checkins] r82721 - in python/branches/import_unicode:
	Include/moduleobject.h Objects/moduleobject.c
Message-ID: <20100708233336.7D425EEBC3@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:36 2010
New Revision: 82721

Log:
Create PyModule_GetFilenameObject() (public) function

Modified:
   python/branches/import_unicode/Include/moduleobject.h
   python/branches/import_unicode/Objects/moduleobject.c

Modified: python/branches/import_unicode/Include/moduleobject.h
==============================================================================
--- python/branches/import_unicode/Include/moduleobject.h	(original)
+++ python/branches/import_unicode/Include/moduleobject.h	Fri Jul  9 01:33:36 2010
@@ -16,6 +16,7 @@
 PyAPI_FUNC(PyObject *) PyModule_GetDict(PyObject *);
 PyAPI_FUNC(const char *) PyModule_GetName(PyObject *);
 PyAPI_FUNC(const char *) PyModule_GetFilename(PyObject *);
+PyAPI_FUNC(PyObject *) PyModule_GetFilenameObject(PyObject *);
 PyAPI_FUNC(void) _PyModule_Clear(PyObject *);
 PyAPI_FUNC(struct PyModuleDef*) PyModule_GetDef(PyObject*);
 PyAPI_FUNC(void*) PyModule_GetState(PyObject*);

Modified: python/branches/import_unicode/Objects/moduleobject.c
==============================================================================
--- python/branches/import_unicode/Objects/moduleobject.c	(original)
+++ python/branches/import_unicode/Objects/moduleobject.c	Fri Jul  9 01:33:36 2010
@@ -188,8 +188,8 @@
     return _PyUnicode_AsString(nameobj);
 }
 
-static PyObject*
-module_getfilename(PyObject *m)
+PyObject*
+PyModule_GetFilenameObject(PyObject *m)
 {
     PyObject *d;
     PyObject *fileobj;
@@ -212,7 +212,7 @@
 PyModule_GetFilename(PyObject *m)
 {
     PyObject *fileobj;
-    fileobj = module_getfilename(m);
+    fileobj = PyModule_GetFilenameObject(m);
     if (fileobj == NULL)
         return NULL;
     return _PyUnicode_AsString(fileobj);
@@ -353,7 +353,7 @@
         PyErr_Clear();
         name = "?";
     }
-    filename = module_getfilename((PyObject *)m);
+    filename = PyModule_GetFilenameObject((PyObject *)m);
     if (filename == NULL) {
         PyErr_Clear();
         return PyUnicode_FromFormat("<module '%s' (built-in)>", name);

From python-checkins at python.org  Fri Jul  9 01:33:38 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri,  9 Jul 2010 01:33:38 +0200 (CEST)
Subject: [Python-checkins] r82722 -
	python/branches/import_unicode/Python/pythonrun.c
Message-ID: <20100708233338.1C48CEEBC3@mail.python.org>

Author: victor.stinner
Date: Fri Jul  9 01:33:37 2010
New Revision: 82722

Log:
initfsencoding() reencodes sys.path and sys.modules filenames

Modified:
   python/branches/import_unicode/Python/pythonrun.c

Modified: python/branches/import_unicode/Python/pythonrun.c
==============================================================================
--- python/branches/import_unicode/Python/pythonrun.c	(original)
+++ python/branches/import_unicode/Python/pythonrun.c	Fri Jul  9 01:33:37 2010
@@ -696,6 +696,118 @@
     }
 }
 
+static PyObject*
+reencode_filename(PyObject *module, PyObject *file, const char *new_encoding)
+{
+    PyObject *file_bytes = NULL;
+    PyObject *new_file = NULL;
+
+    file_bytes = PyUnicode_AsEncodedString(
+            file,
+            Py_FileSystemDefaultEncoding,
+            "surrogateescape");
+    if (file_bytes == NULL)
+        return NULL;
+    new_file = PyUnicode_Decode(
+            PyBytes_AsString(file_bytes),
+            PyBytes_GET_SIZE(file_bytes),
+            new_encoding,
+            "surrogateescape");
+    Py_DECREF(file_bytes);
+    return new_file;
+}
+
+static int
+reencode_module_path(PyObject *module, PyObject *path, const char *new_encoding)
+{
+    PyObject *filename;
+    PyObject *new_filename;
+    Py_ssize_t i, size;
+
+    size = PyList_Size(path);
+    for (i=0; i<size; i++) {
+        filename = PyList_GetItem(path, i);
+        if (filename == NULL)
+            return 1;
+
+        new_filename = reencode_filename(module, filename, new_encoding);
+        if (new_filename == NULL)
+            return 1;
+        if (PyList_SetItem(path, i, new_filename))
+            return 1;
+    }
+    return 0;
+}
+
+static int
+reencode_modules_path(const char *new_encoding)
+{
+    PyInterpreterState *interp = PyThreadState_GET()->interp;
+    PyObject *modules = interp->modules;
+    PyObject *values;
+    PyObject *iter = NULL;
+    PyObject *module = NULL;
+    PyObject *module_dict = NULL;
+    PyObject *file;
+    PyObject *path;
+    int ret = 1;
+
+    /* FIXME: Re-encode PySys_GetObject("path_importer_cache") keys? */
+    /* FIXME: Re-encode co_filename of all code objects! */
+
+    if (strcmp(new_encoding, PyUnicode_GetDefaultEncoding()) == 0)
+        return 0;
+
+
+    values = PyObject_CallMethod(modules, "values", "");
+    if (values == NULL)
+        return 1;
+
+    iter = PyObject_GetIter(values);
+    Py_DECREF(values);
+    if (iter == NULL)
+        return 1;
+
+    for (module = PyIter_Next(iter); module != NULL; module = PyIter_Next(iter)) {
+        file = PyModule_GetFilenameObject(module);
+        if (file != NULL) {
+            PyObject *new_file = reencode_filename(module, file, new_encoding);
+            if (new_file == NULL)
+                goto error;
+            if (PyObject_SetAttrString(module, "__file__", new_file)) {
+                Py_DECREF(new_file);
+                goto error;
+            }
+            Py_DECREF(new_file);
+        } else {
+            PyErr_Clear();
+        }
+
+        module_dict = PyModule_GetDict(module);
+        if (module_dict == NULL)
+            goto error;
+        path = PyDict_GetItemString(module_dict, "__path__");
+        if (path != NULL) {
+            if (reencode_module_path(module, path, new_encoding))
+                goto error;
+        } else {
+            PyErr_Clear();
+        }
+
+        Py_CLEAR(module);
+    }
+
+    ret = 0;
+    goto finally;
+
+error:
+    ret = 1;
+finally:
+    Py_XDECREF(iter);
+    Py_XDECREF(module);
+    return ret;
+}
+
 static void
 initfsencoding(void)
 {
@@ -711,6 +823,10 @@
            stdin and stdout if these are terminals.  */
         codeset = get_codeset();
         if (codeset != NULL) {
+            if (reencode_modules_path(codeset))
+                Py_FatalError(
+                        "Py_Initialize: can't reencode paths");
+
             Py_FileSystemDefaultEncoding = codeset;
             Py_HasFileSystemDefaultEncoding = 0;
             return;

From victor.stinner at haypocalc.com  Fri Jul  9 02:13:58 2010
From: victor.stinner at haypocalc.com (Victor Stinner)
Date: Fri, 9 Jul 2010 02:13:58 +0200
Subject: [Python-checkins] r82666 -
	python/branches/import_unicode/Python/pythonrun.c
In-Reply-To: <20100708233214.4C449EE9BC@mail.python.org>
References: <20100708233214.4C449EE9BC@mail.python.org>
Message-ID: <201007090213.58471.victor.stinner@haypocalc.com>

Le vendredi 09 juillet 2010 01:32:14, victor.stinner a ?crit :
> Author: victor.stinner
> Date: Fri Jul  9 01:32:14 2010
> New Revision: 82666
> 
> Log:
> initialize GIL earlier in Py_InitializeEx()

I should maybe reopen issue #8063 (or open a new one): it's really annoying 
when gdb crashes while I'm debugging the import machinery (or anything else 
while Python is starting).

-- 
Victor Stinner
http://www.haypocalc.com/

From victor.stinner at haypocalc.com  Fri Jul  9 02:19:09 2010
From: victor.stinner at haypocalc.com (Victor Stinner)
Date: Fri, 9 Jul 2010 02:19:09 +0200
Subject: [Python-checkins] r82695 - in python/branches/import_unicode:
	Doc/library/sys.rst Python/import.c
In-Reply-To: <20100708233257.8D264EEB39@mail.python.org>
References: <20100708233257.8D264EEB39@mail.python.org>
Message-ID: <201007090219.09573.victor.stinner@haypocalc.com>

> New Revision: 82695
> 
> Log:
> _find_module() fills directly path argument
> 
>  * find_module(): rename "path" argument to "search_path"
>  * find_module(): result can be NULL
>  * reject bytes path
>  * fix refleak

Reject bytes path is temporary. I think that I can restore support of bytes 
using PyUnicode_DecodeFSDefault().

-- 
Victor Stinner
http://www.haypocalc.com/

From victor.stinner at haypocalc.com  Fri Jul  9 02:22:49 2010
From: victor.stinner at haypocalc.com (Victor Stinner)
Date: Fri, 9 Jul 2010 02:22:49 +0200
Subject: [Python-checkins] r82702 - in python/branches/import_unicode:
	PC/import_nt.c Python/import.c
In-Reply-To: <20100708233307.C8DCAEEBCF@mail.python.org>
References: <20100708233307.C8DCAEEBCF@mail.python.org>
Message-ID: <201007090222.49344.victor.stinner@haypocalc.com>

Le vendredi 09 juillet 2010 01:33:07, victor.stinner a ?crit :
> Author: victor.stinner
> Date: Fri Jul  9 01:33:07 2010
> New Revision: 82702
> 
> Log:
> Move buffer into PyWin_FindRegisteredModule()
> 
> The function now returns the path as a PyUnicodeObject*

Should I consider PyWin_FindRegisteredModule() as a public function? Because 
it's not a good idea to change public functions API :-)

It's not listed in any header file and import.c uses it with "extern ...".

-- 
Victor Stinner
http://www.haypocalc.com/

From victor.stinner at haypocalc.com  Fri Jul  9 02:26:13 2010
From: victor.stinner at haypocalc.com (Victor Stinner)
Date: Fri, 9 Jul 2010 02:26:13 +0200
Subject: [Python-checkins] r82711 - in python/branches/import_unicode:
	Include/Python.h Modules/main.c
In-Reply-To: <20100708233321.2F196EEBA0@mail.python.org>
References: <20100708233321.2F196EEBA0@mail.python.org>
Message-ID: <201007090226.13309.victor.stinner@haypocalc.com>

Le vendredi 09 juillet 2010 01:33:21, victor.stinner a ?crit :
> Author: victor.stinner
> Date: Fri Jul  9 01:33:21 2010
> New Revision: 82711
> 
> Log:
> create _Py_wchar2char()

The implementation is very inefficient, it should be optimized.

I don't know if it works for 16 and 32 bits wchar_t, and narrow and wide 
Python builds.

-- 
Victor Stinner
http://www.haypocalc.com/

From victor.stinner at haypocalc.com  Fri Jul  9 02:29:17 2010
From: victor.stinner at haypocalc.com (Victor Stinner)
Date: Fri, 9 Jul 2010 02:29:17 +0200
Subject: [Python-checkins] r82716 -
	python/branches/import_unicode/Python/ceval.c
In-Reply-To: <20100708233328.AC29EEEBA6@mail.python.org>
References: <20100708233328.AC29EEEBA6@mail.python.org>
Message-ID: <201007090229.17476.victor.stinner@haypocalc.com>

Le vendredi 09 juillet 2010 01:33:28, victor.stinner a ?crit :
> Author: victor.stinner
> Date: Fri Jul  9 01:33:28 2010
> New Revision: 82716
> 
> Log:
> ceval: remove filename debug variable
> 
> _PyUnicode_AsString() may raise an exception

Another solution is to store and then restore the exception (eg. as does 
mywrite()) and set filename to NULL on error.

-- 
Victor Stinner
http://www.haypocalc.com/

From nnorwitz at gmail.com  Fri Jul  9 02:56:33 2010
From: nnorwitz at gmail.com (Neal Norwitz)
Date: Thu, 8 Jul 2010 20:56:33 -0400
Subject: [Python-checkins] Python Regression Test Failures refleak (1)
Message-ID: <20100709005633.GA15572@kbk-i386-bb.psfb.org>

More important issues:
----------------------
test_format leaked [0, 60, 0] references, sum=60

Less important issues:
----------------------

From python-checkins at python.org  Fri Jul  9 03:58:26 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Fri,  9 Jul 2010 03:58:26 +0200 (CEST)
Subject: [Python-checkins] r82723 -
	python/branches/py3k/Lib/test/test_structseq.py
Message-ID: <20100709015826.8797FEE988@mail.python.org>

Author: benjamin.peterson
Date: Fri Jul  9 03:58:26 2010
New Revision: 82723

Log:
use assert method

Modified:
   python/branches/py3k/Lib/test/test_structseq.py

Modified: python/branches/py3k/Lib/test/test_structseq.py
==============================================================================
--- python/branches/py3k/Lib/test/test_structseq.py	(original)
+++ python/branches/py3k/Lib/test/test_structseq.py	Fri Jul  9 03:58:26 2010
@@ -8,7 +8,7 @@
 
     def test_tuple(self):
         t = time.gmtime()
-        assert isinstance(t, tuple)
+        self.assertIsInstance(t, tuple)
         astuple = tuple(t)
         self.assertEqual(len(t), len(astuple))
         self.assertEqual(t, astuple)

From g.brandl at gmx.net  Fri Jul  9 09:09:43 2010
From: g.brandl at gmx.net (Georg Brandl)
Date: Fri, 09 Jul 2010 09:09:43 +0200
Subject: [Python-checkins] r82659 - in python/branches/py3k:
 Doc/library/os.path.rst Doc/library/os.rst Lib/ntpath.py Lib/tarfile.py
 Lib/test/support.py Lib/test/symlink_support.py Lib/test/test_glob.py
 Lib/test/test_httpservers.py Lib/test/test_os.py Lib/test/test_platform.py
 Lib/test/test_posixpath.py Lib/test/test_shutil.py Lib/test/test_sys.py
 Lib/test/test_sysconfig.py Lib/test/test_tarfile.py Misc/ACKS Misc/NEWS
 Modules/posixmodule.c
In-Reply-To: <20100708213909.4B34DD267@mail.python.org>
References: <20100708213909.4B34DD267@mail.python.org>
Message-ID: <i16i39$v05$1@dough.gmane.org>

Am 08.07.2010 23:39, schrieb brian.curtin:
> Author: brian.curtin
> Date: Thu Jul  8 23:39:08 2010
> New Revision: 82659
> 
> Log:
> Implement #1578269. Patch by Jason R. Coombs.
> 
> Added Windows support for os.symlink when run on Windows 6.0 or greater,
> aka Vista. Previous Windows versions will raise NotImplementedError
> when trying to symlink.
> 
> Includes numerous test updates and additions to test_os, including
> a symlink_support module because of the fact that privilege escalation
> is required in order to run the tests to ensure that the user is able
> to create symlinks. By default, accounts do not have the required
> privilege, so the escalation code will have to be exposed later (or
> documented on how to do so). I'll be following up with that work next.
> 
> Note that the tests use ctypes, which was agreed on during the PyCon
> language summit.
> 
> 
> 
> Added:
>    python/branches/py3k/Lib/test/symlink_support.py   (contents, props changed)
> Modified:
>    python/branches/py3k/Doc/library/os.path.rst
>    python/branches/py3k/Doc/library/os.rst
>    python/branches/py3k/Lib/ntpath.py
>    python/branches/py3k/Lib/tarfile.py
>    python/branches/py3k/Lib/test/support.py
>    python/branches/py3k/Lib/test/test_glob.py
>    python/branches/py3k/Lib/test/test_httpservers.py
>    python/branches/py3k/Lib/test/test_os.py
>    python/branches/py3k/Lib/test/test_platform.py
>    python/branches/py3k/Lib/test/test_posixpath.py
>    python/branches/py3k/Lib/test/test_shutil.py
>    python/branches/py3k/Lib/test/test_sys.py
>    python/branches/py3k/Lib/test/test_sysconfig.py
>    python/branches/py3k/Lib/test/test_tarfile.py
>    python/branches/py3k/Misc/ACKS
>    python/branches/py3k/Misc/NEWS
>    python/branches/py3k/Modules/posixmodule.c
> 
> Modified: python/branches/py3k/Doc/library/os.path.rst
> ==============================================================================
> --- python/branches/py3k/Doc/library/os.path.rst	(original)
> +++ python/branches/py3k/Doc/library/os.path.rst	Thu Jul  8 23:39:08 2010
> @@ -231,11 +231,15 @@
>  
>  .. function:: samefile(path1, path2)
>  
> -   Return ``True`` if both pathname arguments refer to the same file or directory
> -   (as indicated by device number and i-node number). Raise an exception if a
> -   :func:`os.stat` call on either pathname fails.
> +   Return ``True`` if both pathname arguments refer to the same file or directory.
> +   On Unix, this is determined by the device number and i-node number and raises an
> +   exception if a :func:`os.stat` call on either pathname fails.
> +
> +   On Windows, two files are the same if they resolve to the same final path
> +   name using the Windows API call GetFinalPathNameByHandle and this function
> +   raises an exception if handles cannot be obtained to either file.
>  
> -   Availability: Unix.
> +   Availability: Windows, Unix.

There is a ".. versionchanged" missing here (and in the other places where
Windows availability was added).

> Modified: python/branches/py3k/Lib/tarfile.py
> ==============================================================================
> --- python/branches/py3k/Lib/tarfile.py	(original)
> +++ python/branches/py3k/Lib/tarfile.py	Thu Jul  8 23:39:08 2010
> @@ -2273,7 +2273,7 @@
>            (platform limitation), we try to make a copy of the referenced file
>            instead of a link.
>          """
> -        if hasattr(os, "symlink") and hasattr(os, "link"):
> +        try:
>              # For systems that support symbolic and hard links.
>              if tarinfo.issym():
>                  os.symlink(tarinfo.linkname, targetpath)
> @@ -2282,7 +2282,15 @@
>                  if os.path.exists(tarinfo._link_target):
>                      os.link(tarinfo._link_target, targetpath)
>                  else:
> -                    self._extract_member(self._find_link_target(tarinfo), targetpath)
> +                    self._extract_mem
> +        except (AttributeError, NotImplementedError, WindowsError):
> +            # AttributeError if no os.symlink
> +            # NotImplementedError if on Windows XP
> +            # WindowsError (1314) if the required privilege is not held by the client
> +            if tarinfo.issym():
> +                linkpath = os.path.join(os.path.dirname(tarinfo.name),tarinfo.linkname)
> +            else:
> +                linkpath = tarinfo.linkname
>          else:
>              try:
>                  self._extract_member(self._find_link_target(tarinfo), targetpath)

Please observe PEP 8 (no overlong lines, space after comma).  (I also see
overlong lines in other files in the diff.)

> Modified: python/branches/py3k/Misc/ACKS
> ==============================================================================
> --- python/branches/py3k/Misc/ACKS	(original)
> +++ python/branches/py3k/Misc/ACKS	Thu Jul  8 23:39:08 2010
> @@ -154,6 +154,7 @@
>  Jeffery Collins
>  Robert Collins
>  Paul Colomiets
> +Jason R. Coombs
>  Geremy Condra
>  Juan Jos? Conti
>  Matt Conway

This doesn't seem to be the correct ordering :)

Georg

-- 
Thus spake the Lord: Thou shalt indent with four spaces. No more, no less.
Four shall be the number of spaces thou shalt indent, and the number of thy
indenting shall be four. Eight shalt thou not indent, nor either indent thou
two, excepting that thou then proceed to four. Tabs are right out.


From python-checkins at python.org  Fri Jul  9 09:33:15 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri,  9 Jul 2010 09:33:15 +0200 (CEST)
Subject: [Python-checkins] r82724 -
	python/branches/py3k/Doc/tools/sphinxext/indexsidebar.html
Message-ID: <20100709073315.AD349EE985@mail.python.org>

Author: georg.brandl
Date: Fri Jul  9 09:33:15 2010
New Revision: 82724

Log:
2.7 is now stable.

Modified:
   python/branches/py3k/Doc/tools/sphinxext/indexsidebar.html

Modified: python/branches/py3k/Doc/tools/sphinxext/indexsidebar.html
==============================================================================
--- python/branches/py3k/Doc/tools/sphinxext/indexsidebar.html	(original)
+++ python/branches/py3k/Doc/tools/sphinxext/indexsidebar.html	Fri Jul  9 09:33:15 2010
@@ -2,8 +2,7 @@
             <p><a href="{{ pathto('download') }}">Download these documents</a></p>
 	    <h3>Docs for other versions</h3>
 	    <ul>
-	      <li><a href="http://docs.python.org/2.6/">Python 2.6 (stable)</a></li>
-	      <li><a href="http://docs.python.org/2.7/">Python 2.7 (in development)</a></li>
+	      <li><a href="http://docs.python.org/2.7/">Python 2.7 (stable)</a></li>
 	      <li><a href="http://docs.python.org/3.1/">Python 3.1 (stable)</a></li>
               <li><a href="http://www.python.org/doc/versions/">Old versions</a></li>
             </ul>

From python-checkins at python.org  Fri Jul  9 09:38:13 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri,  9 Jul 2010 09:38:13 +0200 (CEST)
Subject: [Python-checkins] r82725 - python/branches/release27-maint
Message-ID: <20100709073813.45386EE99D@mail.python.org>

Author: georg.brandl
Date: Fri Jul  9 09:38:13 2010
New Revision: 82725

Log:
Recorded merge of revisions 82724 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82724 | georg.brandl | 2010-07-09 09:33:15 +0200 (Fr, 09 Jul 2010) | 1 line
  
  2.7 is now stable.
........


Modified:
   python/branches/release27-maint/   (props changed)

From python-checkins at python.org  Fri Jul  9 09:42:13 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri,  9 Jul 2010 09:42:13 +0200 (CEST)
Subject: [Python-checkins] r82726 -
	python/branches/release26-maint/Doc/tools/sphinxext/indexsidebar.html
Message-ID: <20100709074213.8EF53EE9A5@mail.python.org>

Author: georg.brandl
Date: Fri Jul  9 09:42:13 2010
New Revision: 82726

Log:
2.7 is stable now.

Modified:
   python/branches/release26-maint/Doc/tools/sphinxext/indexsidebar.html

Modified: python/branches/release26-maint/Doc/tools/sphinxext/indexsidebar.html
==============================================================================
--- python/branches/release26-maint/Doc/tools/sphinxext/indexsidebar.html	(original)
+++ python/branches/release26-maint/Doc/tools/sphinxext/indexsidebar.html	Fri Jul  9 09:42:13 2010
@@ -2,7 +2,7 @@
             <p><a href="{{ pathto('download') }}">Download these documents</a></p>
 	    <h3>Docs for other versions</h3>
 	    <ul>
-	      <li><a href="http://docs.python.org/2.7/">Python 2.7 (in development)</a></li>
+	      <li><a href="http://docs.python.org/2.7/">Python 2.7 (stable)</a></li>
 	      <li><a href="http://docs.python.org/3.1/">Python 3.1 (stable)</a></li>
 	      <li><a href="http://docs.python.org/dev/py3k/">Python 3.2 (in development)</a></li>
               <li><a href="http://www.python.org/doc/versions/">Old versions</a></li>

From python-checkins at python.org  Fri Jul  9 09:46:10 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri,  9 Jul 2010 09:46:10 +0200 (CEST)
Subject: [Python-checkins] r82727 - python/branches/release26-maint
Message-ID: <20100709074610.97579EE9A5@mail.python.org>

Author: georg.brandl
Date: Fri Jul  9 09:46:10 2010
New Revision: 82727

Log:
Initialized merge tracking via "svnmerge" with revisions "1-82726" from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint


Modified:
   python/branches/release26-maint/   (props changed)

From python-checkins at python.org  Fri Jul  9 09:48:05 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri,  9 Jul 2010 09:48:05 +0200 (CEST)
Subject: [Python-checkins] r82728 - python/branches/release26-maint
Message-ID: <20100709074805.7BF89EE9A5@mail.python.org>

Author: georg.brandl
Date: Fri Jul  9 09:48:05 2010
New Revision: 82728

Log:
Recorded merge of revisions 82725 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

................
  r82725 | georg.brandl | 2010-07-09 09:38:13 +0200 (Fr, 09 Jul 2010) | 9 lines
  
  Recorded merge of revisions 82724 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k
  
  ........
    r82724 | georg.brandl | 2010-07-09 09:33:15 +0200 (Fr, 09 Jul 2010) | 1 line
    
    2.7 is now stable.
  ........
................


Modified:
   python/branches/release26-maint/   (props changed)

From python-checkins at python.org  Fri Jul  9 09:51:44 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri,  9 Jul 2010 09:51:44 +0200 (CEST)
Subject: [Python-checkins] r82729 - in python/branches/release26-maint:
	Doc/distutils/apiref.rst Doc/distutils/builtdist.rst
	Doc/distutils/setupscript.rst Doc/distutils/sourcedist.rst
	Doc/library/functions.rst Doc/using/cmdline.rst
Message-ID: <20100709075144.1986DD618@mail.python.org>

Author: georg.brandl
Date: Fri Jul  9 09:51:43 2010
New Revision: 82729

Log:
Merged revisions 82548,82634-82635 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

................
  r82548 | georg.brandl | 2010-07-04 19:28:33 +0200 (So, 04 Jul 2010) | 1 line
  
  #8472: fix misleading reference to ifilterfalse() in filter() docs.
................
  r82634 | georg.brandl | 2010-07-07 21:05:35 +0200 (Mi, 07 Jul 2010) | 13 lines
  
  Merged revisions 82629,82632 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k
  
  ........
    r82629 | georg.brandl | 2010-07-07 20:51:43 +0200 (Mi, 07 Jul 2010) | 1 line
    
    Make comment out of an awkward note.
  ........
    r82632 | georg.brandl | 2010-07-07 21:04:36 +0200 (Mi, 07 Jul 2010) | 1 line
    
    Turn more notes into comments.
  ........
................
  r82635 | georg.brandl | 2010-07-07 21:09:12 +0200 (Mi, 07 Jul 2010) | 9 lines
  
  Merged revisions 82615 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k
  
  ........
    r82615 | georg.brandl | 2010-07-07 00:58:50 +0200 (Mi, 07 Jul 2010) | 1 line
    
    Fix typo.
  ........
................


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Doc/distutils/apiref.rst
   python/branches/release26-maint/Doc/distutils/builtdist.rst
   python/branches/release26-maint/Doc/distutils/setupscript.rst
   python/branches/release26-maint/Doc/distutils/sourcedist.rst
   python/branches/release26-maint/Doc/library/functions.rst
   python/branches/release26-maint/Doc/using/cmdline.rst

Modified: python/branches/release26-maint/Doc/distutils/apiref.rst
==============================================================================
--- python/branches/release26-maint/Doc/distutils/apiref.rst	(original)
+++ python/branches/release26-maint/Doc/distutils/apiref.rst	Fri Jul  9 09:51:43 2010
@@ -1012,7 +1012,7 @@
    errors are ignored (apart from being reported to ``sys.stdout`` if *verbose* is
    true).
 
-**\*\*** Some of this could be replaced with the shutil module? **\*\***
+.. XXX Some of this could be replaced with the shutil module?
 
 
 :mod:`distutils.file_util` --- Single file operations
@@ -1328,8 +1328,7 @@
   the "negative alias" of :option:`--verbose`, then :option:`--quiet` on the
   command line sets *verbose* to false.
 
-**\*\*** Should be replaced with :mod:`optik` (which is also now known as
-:mod:`optparse` in Python 2.3 and later). **\*\***
+.. XXX Should be replaced with :mod:`optparse`.
 
 
 .. function:: fancy_getopt(options, negative_opt, object, args)

Modified: python/branches/release26-maint/Doc/distutils/builtdist.rst
==============================================================================
--- python/branches/release26-maint/Doc/distutils/builtdist.rst	(original)
+++ python/branches/release26-maint/Doc/distutils/builtdist.rst	Fri Jul  9 09:51:43 2010
@@ -143,8 +143,8 @@
 Creating dumb built distributions
 =================================
 
-**\*\*** Need to document absolute vs. prefix-relative packages here, but first
-I have to implement it! **\*\***
+.. XXX Need to document absolute vs. prefix-relative packages here, but first
+       I have to implement it!
 
 
 .. _creating-rpms:

Modified: python/branches/release26-maint/Doc/distutils/setupscript.rst
==============================================================================
--- python/branches/release26-maint/Doc/distutils/setupscript.rst	(original)
+++ python/branches/release26-maint/Doc/distutils/setupscript.rst	Fri Jul  9 09:51:43 2010
@@ -207,7 +207,7 @@
 SWIG on the interface file and compile the resulting C/C++ file into your
 extension.
 
-**\*\*** SWIG support is rough around the edges and largely untested! **\*\***
+.. XXX SWIG support is rough around the edges and largely untested!
 
 This warning notwithstanding, options to SWIG can be currently passed like
 this::
@@ -326,7 +326,7 @@
 (Again, this sort of non-portable construct should be avoided if you intend to
 distribute your code.)
 
-**\*\*** Should mention clib libraries here or somewhere else! **\*\***
+.. XXX Should mention clib libraries here or somewhere else!
 
 
 Other options

Modified: python/branches/release26-maint/Doc/distutils/sourcedist.rst
==============================================================================
--- python/branches/release26-maint/Doc/distutils/sourcedist.rst	(original)
+++ python/branches/release26-maint/Doc/distutils/sourcedist.rst	Fri Jul  9 09:51:43 2010
@@ -68,10 +68,10 @@
   :option:`packages` options
 
 * all C source files mentioned in the :option:`ext_modules` or
-  :option:`libraries` options (
+  :option:`libraries` options
 
-  **\*\*** getting C library sources currently broken---no
-  :meth:`get_source_files` method in :file:`build_clib.py`! **\*\***)
+  .. XXX Getting C library sources is currently broken -- no
+     :meth:`get_source_files` method in :file:`build_clib.py`!
 
 * scripts identified by the :option:`scripts` option
 

Modified: python/branches/release26-maint/Doc/library/functions.rst
==============================================================================
--- python/branches/release26-maint/Doc/library/functions.rst	(original)
+++ python/branches/release26-maint/Doc/library/functions.rst	Fri Jul  9 09:51:43 2010
@@ -399,8 +399,9 @@
    iterable if function(item)]`` if function is not ``None`` and ``[item for item
    in iterable if item]`` if function is ``None``.
 
-   See :func:`itertools.ifilterfalse` for the complementary function that returns
-   elements of *iterable* for which *function* returns false.
+   See :func:`itertools.ifilter` and :func:`itertools.ifilterfalse` for iterator
+   versions of this function, including a variation that filters for elements
+   where the *function* returns false.
 
 
 .. function:: float([x])

Modified: python/branches/release26-maint/Doc/using/cmdline.rst
==============================================================================
--- python/branches/release26-maint/Doc/using/cmdline.rst	(original)
+++ python/branches/release26-maint/Doc/using/cmdline.rst	Fri Jul  9 09:51:43 2010
@@ -58,7 +58,7 @@
 
 .. cmdoption:: -c <command>
 
-   Execute the Python code in *command*.  *command* can be one ore more
+   Execute the Python code in *command*.  *command* can be one or more
    statements separated by newlines, with significant leading whitespace as in
    normal module code.
 

From nnorwitz at gmail.com  Fri Jul  9 10:34:58 2010
From: nnorwitz at gmail.com (Neal Norwitz)
Date: Fri, 9 Jul 2010 04:34:58 -0400
Subject: [Python-checkins] Python Regression Test Failures basics (1)
Message-ID: <20100709083458.GA27283@kbk-i386-bb.psfb.org>

346 tests OK.
1 test failed:
    test_subprocess
37 tests skipped:
    test_aepack test_al test_applesingle test_bsddb185 test_bsddb3
    test_cd test_cl test_curses test_epoll test_gdb test_gl
    test_imgfile test_ioctl test_kqueue test_linuxaudiodev test_macos
    test_macostools test_multiprocessing test_ossaudiodev test_pep277
    test_py3kwarn test_scriptpackages test_smtpnet test_socketserver
    test_startfile test_sunaudiodev test_tcl test_timeout test_tk
    test_ttk_guionly test_ttk_textonly test_unicode_file
    test_urllib2net test_urllibnet test_winreg test_winsound
    test_zipfile64
7 skips unexpected on linux2:
    test_epoll test_gdb test_ioctl test_multiprocessing test_tk
    test_ttk_guionly test_ttk_textonly

== CPython 2.7 (trunk:82729M, Jul 9 2010, 04:02:46) [GCC 3.3.4 20040623 (Gentoo Linux 3.3.4-r1, ssp-3.3.2-2, pie-8.7.6)]
==   Linux-2.6.9-gentoo-r1-i686-AMD_Athlon-tm-_XP_3000+-with-gentoo-1.4.16 little-endian
==   /tmp/test_python_3958
test_grammar
test_opcodes
test_dict
test_builtin
test_exceptions
test_types
test_unittest
test_doctest
test_doctest2
test_MimeWriter
test_SimpleHTTPServer
test_StringIO
test___all__
test___future__
test__locale
test_abc
test_abstract_numbers
test_aepack
test_aepack skipped -- No module named aetypes
test_aifc
test_al
test_al skipped -- No module named al
test_anydbm
test_applesingle
test_applesingle skipped -- No module named MacOS
test_argparse
test_array
test_ascii_formatd
test_ast
test_asynchat
test_asyncore
test_atexit
test_audioop
test_augassign
test_base64
test_bastion
test_bigaddrspace
test_bigmem
test_binascii
test_binhex
test_binop
test_bisect
test_bool
test_bsddb
test_bsddb185
test_bsddb185 skipped -- No module named bsddb185
test_bsddb3
test_bsddb3 skipped -- Use of the `bsddb' resource not enabled
test_buffer
test_bufio
test_bytes
test_bz2
test_calendar
test_call
test_capi
test_cd
test_cd skipped -- No module named cd
test_cfgparser
test_cgi
test_charmapcodec
test_cl
test_cl skipped -- No module named cl
test_class
test_cmath
test_cmd
test_cmd_line
test_cmd_line_script
test_code
test_codeccallbacks
test_codecencodings_cn
test_codecencodings_hk
test_codecencodings_jp
test_codecencodings_kr
test_codecencodings_tw
test_codecmaps_cn
	fetching http://source.icu-project.org/repos/icu/data/trunk/charset/data/xml/gb-18030-2000.xml ...
	fetching http://people.freebsd.org/~perky/i18n/EUC-CN.TXT ...
	fetching http://www.unicode.org/Public/MAPPINGS/VENDORS/MICSFT/WINDOWS/CP936.TXT ...
test_codecmaps_hk
	fetching http://people.freebsd.org/~perky/i18n/BIG5HKSCS-2004.TXT ...
test_codecmaps_jp
	fetching http://www.unicode.org/Public/MAPPINGS/VENDORS/MICSFT/WINDOWS/CP932.TXT ...
	fetching http://people.freebsd.org/~perky/i18n/EUC-JISX0213.TXT ...
	fetching http://people.freebsd.org/~perky/i18n/EUC-JP.TXT ...
	fetching http://www.unicode.org/Public/MAPPINGS/OBSOLETE/EASTASIA/JIS/SHIFTJIS.TXT ...
	fetching http://people.freebsd.org/~perky/i18n/SHIFT_JISX0213.TXT ...
test_codecmaps_kr
	fetching http://www.unicode.org/Public/MAPPINGS/VENDORS/MICSFT/WINDOWS/CP949.TXT ...
	fetching http://people.freebsd.org/~perky/i18n/EUC-KR.TXT ...
	fetching http://www.unicode.org/Public/MAPPINGS/OBSOLETE/EASTASIA/KSC/JOHAB.TXT ...
test_codecmaps_tw
	fetching http://www.unicode.org/Public/MAPPINGS/OBSOLETE/EASTASIA/OTHER/BIG5.TXT ...
	fetching http://www.unicode.org/Public/MAPPINGS/VENDORS/MICSFT/WINDOWS/CP950.TXT ...
test_codecs
test_codeop
test_coding
test_coercion
test_collections
test_colorsys
test_commands
test_compare
test_compile
test_compileall
test_compiler
test_complex
test_complex_args
test_contains
test_contextlib
test_cookie
test_cookielib
test_copy
test_copy_reg
test_cpickle
test_cprofile
test_crypt
test_csv
test_ctypes
test_curses
test_curses skipped -- Use of the `curses' resource not enabled
test_datetime
test_dbm
test_decimal
test_decorators
test_defaultdict
test_deque
test_descr
test_descrtut
test_dictcomps
test_dictviews
test_difflib
test_dircache
test_dis
test_distutils
[19643 refs]
test_dl
test_docxmlrpc
test_dumbdbm
test_dummy_thread
test_dummy_threading
test_email
test_email_codecs
test_email_renamed
test_enumerate
test_eof
test_epoll
test_epoll skipped -- kernel doesn't support epoll()
test_errno
test_exception_variations
test_extcall
test_fcntl
test_file
test_file2k
test_filecmp
test_fileinput
test_fileio
test_float
test_fnmatch
test_fork1
test_format
test_fpformat
test_fractions
test_frozen
test_ftplib
test_funcattrs
test_functools
test_future
test_future3
test_future4
test_future5
test_future_builtins
test_gc
test_gdb
test_gdb skipped -- gdb versions before 7.0 didn't support python embedding Saw:
GNU gdb 6.2.1
Copyright 2004 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB.  Type "show warranty" for details.
This GDB was configured as "i386-pc-linux-gnu".

test_gdbm
test_generators
test_genericpath
test_genexps
test_getargs
test_getargs2
test_getopt
test_gettext
test_gl
test_gl skipped -- No module named gl
test_glob
test_global
test_grp
test_gzip
test_hash
test_hashlib
test_heapq
test_hmac
test_hotshot
test_htmllib
test_htmlparser
test_httplib
test_httpservers
[15648 refs]
[15648 refs]
[15648 refs]
[25315 refs]
test_imageop
test_imaplib
test_imgfile
test_imgfile skipped -- No module named imgfile
test_imp
test_import
test_importhooks
test_importlib
test_index
test_inspect
test_int
test_int_literal
test_io
test_ioctl
test_ioctl skipped -- Unable to open /dev/tty
test_isinstance
test_iter
test_iterlen
test_itertools
test_json
test_kqueue
test_kqueue skipped -- test works only on BSD
test_largefile
test_lib2to3
test_linecache
test_linuxaudiodev
test_linuxaudiodev skipped -- Use of the `audio' resource not enabled
test_list
test_locale
test_logging
test_long
test_long_future
test_longexp
test_macos
test_macos skipped -- No module named MacOS
test_macostools
test_macostools skipped -- No module named MacOS
test_macpath
test_mailbox
test_marshal
test_math
test_md5
test_memoryio
test_memoryview
test_mhlib
test_mimetools
test_mimetypes
test_minidom
test_mmap
test_module
test_modulefinder
test_multibytecodec
test_multibytecodec_support
test_multifile
test_multiprocessing
test_multiprocessing skipped -- This platform lacks a functioning sem_open implementation, therefore, the required synchronization primitives needed will not function, see issue 3770.
test_mutants
test_mutex
test_netrc
test_new
test_nis
test_normalization
	fetching http://www.unicode.org/Public/5.2.0/ucd/NormalizationTest.txt ...
test_ntpath
test_old_mailbox
test_openpty
test_operator
test_optparse
test_os
[15648 refs]
[15648 refs]
test_ossaudiodev
test_ossaudiodev skipped -- Use of the `audio' resource not enabled
test_parser
Expecting 's_push: parser stack overflow' in next line
s_push: parser stack overflow
test_pdb
test_peepholer
test_pep247
test_pep263
test_pep277
test_pep277 skipped -- only NT+ and systems with Unicode-friendly filesystem encoding
test_pep292
test_pep352
test_pickle
test_pickletools
test_pipes
test_pkg
test_pkgimport
test_pkgutil
test_platform
[17044 refs]
[17044 refs]
test_plistlib
test_poll
test_popen
[15653 refs]
[15653 refs]
[15653 refs]
test_popen2
test_poplib
test_posix
test_posixpath
test_pow
test_pprint
test_print
test_profile
test_profilehooks
test_property
test_pstats
test_pty
test_pwd
test_py3kwarn
test_py3kwarn skipped -- test.test_py3kwarn must be run with the -3 flag
test_pyclbr
test_pydoc
[20862 refs]
[20862 refs]
[20862 refs]
[20862 refs]
[20862 refs]
[20861 refs]
[20861 refs]
test_pyexpat
test_queue
test_quopri
[18479 refs]
[18479 refs]
test_random
test_re
test_readline
test_repr
test_resource
test_rfc822
test_richcmp
test_robotparser
test_runpy
test_sax
test_scope
test_scriptpackages
test_scriptpackages skipped -- No module named aetools
test_select
test_set
test_setcomps
test_sets
test_sgmllib
test_sha
test_shelve
test_shlex
test_shutil
test_signal
test_site
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
test_slice
test_smtplib
test_smtpnet
test_smtpnet skipped -- Use of the `network' resource not enabled
test_socket
test_socketserver
test_socketserver skipped -- Use of the `network' resource not enabled
test_softspace
test_sort
test_sqlite
test_ssl
test_startfile
test_startfile skipped -- module os has no attribute startfile
test_str
test_strftime
test_string
test_stringprep
test_strop
test_strptime
test_strtod
test_struct
test_structmembers
test_structseq
test_subprocess
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15863 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
.
[15648 refs]
[15648 refs]
    this bit of output is from a test of stdout in a different process ...
[15648 refs]
[15648 refs]
[15863 refs]
[15648 refs]
[15648 refs]
p.kill() succeeded after 2 attempts
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15863 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
.
[15648 refs]
[15648 refs]
    this bit of output is from a test of stdout in a different process ...
[15648 refs]
[15648 refs]
[15863 refs]
test test_subprocess failed -- multiple errors occurred; run in verbose mode for details
Re-running test 'test_subprocess' in verbose mode
test_call_kwargs (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_call_seq (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_check_call_nonzero (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_check_call_zero (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_check_output (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_check_output_nonzero (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_check_output_stderr (test.test_subprocess.ProcessTestCase) ... ERROR
test_check_output_stdout_arg (test.test_subprocess.ProcessTestCase) ... ERROR
test_communicate (test.test_subprocess.ProcessTestCase) ... ERROR
test_communicate_pipe_buf (test.test_subprocess.ProcessTestCase) ... ERROR
test_communicate_pipe_fd_leak (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_communicate_returns (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_communicate_stderr (test.test_subprocess.ProcessTestCase) ... ERROR
test_communicate_stdin (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_communicate_stdout (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_cwd (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_env (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_executable_with_cwd (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_executable_without_cwd (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_invalid_bufsize (test.test_subprocess.ProcessTestCase) ... ERROR
test_leaking_fds_on_error (test.test_subprocess.ProcessTestCase) ... ERROR
test_list2cmdline (test.test_subprocess.ProcessTestCase) ... ERROR
test_no_leaking (test.test_subprocess.ProcessTestCase) ... ERROR
test_poll (test.test_subprocess.ProcessTestCase) ... [15863 refs]
ERROR
test_stderr_filedes (test.test_subprocess.ProcessTestCase) ... ERROR
test_stderr_fileobj (test.test_subprocess.ProcessTestCase) ... ERROR
test_stderr_none (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_stderr_pipe (test.test_subprocess.ProcessTestCase) ... ERROR
test_stdin_filedes (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_stdin_fileobj (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_stdin_none (test.test_subprocess.ProcessTestCase) ... ERROR
test_stdin_pipe (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_stdout_filedes (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_stdout_filedes_of_stdout (test.test_subprocess.ProcessTestCase) ... .
[15648 refs]
ERROR
test_stdout_fileobj (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_stdout_none (test.test_subprocess.ProcessTestCase) ...     this bit of output is from a test of stdout in a different process ...
ERROR
test_stdout_pipe (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_stdout_stderr_file (test.test_subprocess.ProcessTestCase) ... ERROR
test_stdout_stderr_pipe (test.test_subprocess.ProcessTestCase) ... ERROR
test_universal_newlines (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_universal_newlines_communicate (test.test_subprocess.ProcessTestCase) ... ERROR
test_wait (test.test_subprocess.ProcessTestCase) ... [15863 refs]
ERROR
test_writes_before_communicate (test.test_subprocess.ProcessTestCase) ... ERROR
test_args_string (test.test_subprocess.POSIXProcessTestCase) ... [15648 refs]
ERROR
test_call_string (test.test_subprocess.POSIXProcessTestCase) ... [15648 refs]
ERROR
test_exceptions (test.test_subprocess.POSIXProcessTestCase) ... ERROR
test_invalid_args (test.test_subprocess.POSIXProcessTestCase) ... ERROR
test_kill (test.test_subprocess.POSIXProcessTestCase) ... ERROR
test_preexec (test.test_subprocess.POSIXProcessTestCase) ... [15648 refs]
ERROR
test_run_abort (test.test_subprocess.POSIXProcessTestCase) ... ERROR
test_send_signal (test.test_subprocess.POSIXProcessTestCase) ... ERROR
test_shell_sequence (test.test_subprocess.POSIXProcessTestCase) ... ERROR
test_shell_string (test.test_subprocess.POSIXProcessTestCase) ... ERROR
test_terminate (test.test_subprocess.POSIXProcessTestCase) ... ERROR
test_call_string (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_close_fds (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_creationflags (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_invalid_args (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_kill (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_send_signal (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_shell_sequence (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_shell_string (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_startupinfo (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_terminate (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_call_kwargs (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_call_seq (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_check_call_nonzero (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_check_call_zero (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_check_output (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_check_output_nonzero (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_check_output_stderr (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_check_output_stdout_arg (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_communicate (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_communicate_pipe_buf (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_communicate_pipe_fd_leak (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_communicate_returns (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_communicate_stderr (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_communicate_stdin (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_communicate_stdout (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_cwd (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_env (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_executable_with_cwd (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_executable_without_cwd (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_invalid_bufsize (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_leaking_fds_on_error (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_list2cmdline (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_no_leaking (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_poll (test.test_subprocess.ProcessTestCaseNoPoll) ... [15863 refs]
ERROR
test_stderr_filedes (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_stderr_fileobj (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_stderr_none (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_stderr_pipe (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_stdin_filedes (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_stdin_fileobj (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_stdin_none (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_stdin_pipe (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_stdout_filedes (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_stdout_filedes_of_stdout (test.test_subprocess.ProcessTestCaseNoPoll) ... .
[15648 refs]
ERROR
test_stdout_fileobj (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_stdout_none (test.test_subprocess.ProcessTestCaseNoPoll) ...     this bit of output is from a test of stdout in a different process ...
ERROR
test_stdout_pipe (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_stdout_stderr_file (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_stdout_stderr_pipe (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_universal_newlines (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_universal_newlines_communicate (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_wait (test.test_subprocess.ProcessTestCaseNoPoll) ... [15863 refs]
ERROR
test_writes_before_communicate (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_eintr_retry_call (test.test_subprocess.HelperFunctionTests) ... ok

======================================================================
ERROR: test_call_kwargs (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_call_seq (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_call_nonzero (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_call_zero (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_output (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_output_nonzero (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_output_stderr (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_output_stdout_arg (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_pipe_buf (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_pipe_fd_leak (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_returns (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_stderr (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_stdin (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_stdout (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_cwd (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_env (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_executable_with_cwd (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_executable_without_cwd (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_invalid_bufsize (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_leaking_fds_on_error (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_list2cmdline (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_no_leaking (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_poll (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stderr_filedes (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stderr_fileobj (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stderr_none (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stderr_pipe (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdin_filedes (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdin_fileobj (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdin_none (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdin_pipe (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_filedes (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_filedes_of_stdout (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_fileobj (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_none (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_pipe (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_stderr_file (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_stderr_pipe (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_universal_newlines (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_universal_newlines_communicate (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_wait (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_writes_before_communicate (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_args_string (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_call_string (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_exceptions (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_invalid_args (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_kill (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_preexec (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_run_abort (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_send_signal (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_shell_sequence (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_shell_string (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_terminate (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_call_kwargs (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_call_seq (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_call_nonzero (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_call_zero (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_output (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_output_nonzero (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_output_stderr (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_output_stdout_arg (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_pipe_buf (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_pipe_fd_leak (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_returns (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_stderr (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_stdin (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_stdout (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_cwd (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_env (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_executable_with_cwd (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_executable_without_cwd (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_invalid_bufsize (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_leaking_fds_on_error (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_list2cmdline (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_no_leaking (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_poll (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stderr_filedes (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stderr_fileobj (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stderr_none (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stderr_pipe (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdin_filedes (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdin_fileobj (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdin_none (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdin_pipe (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_filedes (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_filedes_of_stdout (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_fileobj (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_none (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_pipe (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_stderr_file (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_stderr_pipe (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_universal_newlines (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_universal_newlines_communicate (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_wait (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_writes_before_communicate (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

----------------------------------------------------------------------
Ran 108 tests in 111.160s

FAILED (errors=97, skipped=10)
test test_subprocess failed -- multiple errors occurred
test_sunaudiodev
test_sunaudiodev skipped -- No module named sunaudiodev
test_sundry
test_symtable
test_syntax
test_sys
[15648 refs]
[15648 refs]
[15648 refs]
[15877 refs]
[15671 refs]
test_sysconfig
[15648 refs]
[15648 refs]
test_tarfile
test_tcl
test_tcl skipped -- No module named _tkinter
test_telnetlib
test_tempfile
[15648 refs]
test_textwrap
test_thread
test_threaded_import
test_threadedtempfile
test_threading
[18948 refs]
[20223 refs]
[20035 refs]
[20035 refs]
[20035 refs]
[20035 refs]
test_threading_local
test_threadsignals
test_time
test_timeout
test_timeout skipped -- Use of the `network' resource not enabled
test_tk
test_tk skipped -- No module named _tkinter
test_tokenize
test_trace
test_traceback
test_transformer
test_ttk_guionly
test_ttk_guionly skipped -- No module named _tkinter
test_ttk_textonly
test_ttk_textonly skipped -- No module named _tkinter
test_tuple
test_typechecks
test_ucn
test_unary
test_undocumented_details
test_unicode
test_unicode_file
test_unicode_file skipped -- No Unicode filesystem semantics on this platform.
test_unicodedata
test_univnewlines
test_univnewlines2k
test_unpack
test_urllib
test_urllib2
test_urllib2_localnet
test_urllib2net
test_urllib2net skipped -- Use of the `network' resource not enabled
test_urllibnet
test_urllibnet skipped -- Use of the `network' resource not enabled
test_urlparse
test_userdict
test_userlist
test_userstring
test_uu
test_uuid
test_wait3
test_wait4
test_warnings
[15679 refs]
[15679 refs]
[15672 refs]
[15679 refs]
[15679 refs]
[15672 refs]
test_wave
test_weakref
test_weakset
test_whichdb
test_winreg
test_winreg skipped -- No module named _winreg
test_winsound
test_winsound skipped -- No module named winsound
test_with
test_wsgiref
test_xdrlib
test_xml_etree
test_xml_etree_c
test_xmllib
test_xmlrpc
test_xpickle
test_xpickle -- skipping backwards compat tests.
Use 'regrtest.py -u xpickle' to run them.
test_xrange
test_zipfile
test_zipfile64
test_zipfile64 skipped -- test requires loads of disk-space bytes and a long time to run
test_zipimport
test_zipimport_support
test_zlib
346 tests OK.
1 test failed:
    test_subprocess
37 tests skipped:
    test_aepack test_al test_applesingle test_bsddb185 test_bsddb3
    test_cd test_cl test_curses test_epoll test_gdb test_gl
    test_imgfile test_ioctl test_kqueue test_linuxaudiodev test_macos
    test_macostools test_multiprocessing test_ossaudiodev test_pep277
    test_py3kwarn test_scriptpackages test_smtpnet test_socketserver
    test_startfile test_sunaudiodev test_tcl test_timeout test_tk
    test_ttk_guionly test_ttk_textonly test_unicode_file
    test_urllib2net test_urllibnet test_winreg test_winsound
    test_zipfile64
7 skips unexpected on linux2:
    test_epoll test_gdb test_ioctl test_multiprocessing test_tk
    test_ttk_guionly test_ttk_textonly
[966950 refs]

From ncoghlan at gmail.com  Fri Jul  9 12:26:56 2010
From: ncoghlan at gmail.com (Nick Coghlan)
Date: Fri, 9 Jul 2010 20:26:56 +1000
Subject: [Python-checkins] Python Regression Test Failures basics (1)
In-Reply-To: <20100709083458.GA27283@kbk-i386-bb.psfb.org>
References: <20100709083458.GA27283@kbk-i386-bb.psfb.org>
Message-ID: <AANLkTilSMxqLkUjjrDCljtq70n76PATJoOVCsJnulj0z@mail.gmail.com>

On Fri, Jul 9, 2010 at 6:34 PM, Neal Norwitz <nnorwitz at gmail.com> wrote:
> == CPython 2.7 (trunk:82729M, Jul 9 2010, 04:02:46) [GCC 3.3.4 20040623 (Gentoo Linux 3.3.4-r1, ssp-3.3.2-2, pie-8.7.6)]
> == ? Linux-2.6.9-gentoo-r1-i686-AMD_Athlon-tm-_XP_3000+-with-gentoo-1.4.16 little-endian

Neal, these scripts need to be updated to either pull from py3k or
from the 2.7 maintenance branch. They aren't going to see much action
monitoring the trunk from now on :)

Cheers,
Nick.

-- 
Nick Coghlan?? |?? ncoghlan at gmail.com?? |?? Brisbane, Australia

From python-checkins at python.org  Fri Jul  9 14:23:21 2010
From: python-checkins at python.org (r.david.murray)
Date: Fri,  9 Jul 2010 14:23:21 +0200 (CEST)
Subject: [Python-checkins] r82730 - in python/branches/py3k: Lib/fnmatch.py
	Lib/test/test_fnmatch.py Misc/ACKS Misc/NEWS
Message-ID: <20100709122321.BCC55EE9EF@mail.python.org>

Author: r.david.murray
Date: Fri Jul  9 14:23:21 2010
New Revision: 82730

Log:
7846: limit fnmatch pattern cache to _MAXCACHE=100 entries.

Patch by Andrew Clegg.


Modified:
   python/branches/py3k/Lib/fnmatch.py
   python/branches/py3k/Lib/test/test_fnmatch.py
   python/branches/py3k/Misc/ACKS
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/fnmatch.py
==============================================================================
--- python/branches/py3k/Lib/fnmatch.py	(original)
+++ python/branches/py3k/Lib/fnmatch.py	Fri Jul  9 14:23:21 2010
@@ -16,6 +16,7 @@
 
 _cache = {}  # Maps text patterns to compiled regexen.
 _cacheb = {}  # Ditto for bytes patterns.
+_MAXCACHE = 100 # Maximum size of caches
 
 def fnmatch(name, pat):
     """Test whether FILENAME matches PATTERN.
@@ -48,6 +49,8 @@
             res = bytes(res_str, 'ISO-8859-1')
         else:
             res = translate(pat)
+        if len(cache) >= _MAXCACHE:
+            cache.clear()
         cache[pat] = regex = re.compile(res)
     return regex.match
 

Modified: python/branches/py3k/Lib/test/test_fnmatch.py
==============================================================================
--- python/branches/py3k/Lib/test/test_fnmatch.py	(original)
+++ python/branches/py3k/Lib/test/test_fnmatch.py	Fri Jul  9 14:23:21 2010
@@ -3,7 +3,7 @@
 from test import support
 import unittest
 
-from fnmatch import fnmatch, fnmatchcase
+from fnmatch import fnmatch, fnmatchcase, _MAXCACHE, _cache, _cacheb
 
 
 class FnmatchTestCase(unittest.TestCase):
@@ -60,6 +60,21 @@
         self.check_match(b'test\xff', b'te*\xff')
         self.check_match(b'foo\nbar', b'foo*')
 
+    def test_cache_clearing(self):
+        # check that caches do not grow too large
+        # http://bugs.python.org/issue7846
+
+        # string pattern cache
+        for i in range(_MAXCACHE + 1):
+            fnmatch('foo', '?' * i)
+
+        self.assertLessEqual(len(_cache), _MAXCACHE)
+
+        # bytes pattern cache
+        for i in range(_MAXCACHE + 1):
+            fnmatch(b'foo', b'?' * i)
+        self.assertLessEqual(len(_cacheb), _MAXCACHE)
+
 
 def test_main():
     support.run_unittest(FnmatchTestCase)

Modified: python/branches/py3k/Misc/ACKS
==============================================================================
--- python/branches/py3k/Misc/ACKS	(original)
+++ python/branches/py3k/Misc/ACKS	Fri Jul  9 14:23:21 2010
@@ -146,6 +146,7 @@
 David Cinege
 Mike Clarkson
 Brad Clements
+Andrew Clegg
 Steve Clift
 Nick Coghlan
 Josh Cogliati

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Fri Jul  9 14:23:21 2010
@@ -470,6 +470,8 @@
 Library
 -------
 
+- Issue #7646: The fnmatch pattern cache no longer grows without bound.
+
 - Issue #9136: Fix 'dictionary changed size during iteration'
   RuntimeError produced when profiling the decimal module.  This was
   due to a dangerous iteration over 'locals()' in Context.__init__.

From ncoghlan at gmail.com  Fri Jul  9 14:51:27 2010
From: ncoghlan at gmail.com (Nick Coghlan)
Date: Fri, 9 Jul 2010 22:51:27 +1000
Subject: [Python-checkins] r82702 - in python/branches/import_unicode:
	PC/import_nt.c Python/import.c
In-Reply-To: <201007090222.49344.victor.stinner@haypocalc.com>
References: <20100708233307.C8DCAEEBCF@mail.python.org>
	<201007090222.49344.victor.stinner@haypocalc.com>
Message-ID: <AANLkTinI7k_QT6dtI7LN-y0Dd4LKZ1HCTea5dNzcWi9b@mail.gmail.com>

On Fri, Jul 9, 2010 at 10:22 AM, Victor Stinner
<victor.stinner at haypocalc.com> wrote:
> Le vendredi 09 juillet 2010 01:33:07, victor.stinner a ?crit :
>> Author: victor.stinner
>> Date: Fri Jul ?9 01:33:07 2010
>> New Revision: 82702
>>
>> Log:
>> Move buffer into PyWin_FindRegisteredModule()
>>
>> The function now returns the path as a PyUnicodeObject*
>
> Should I consider PyWin_FindRegisteredModule() as a public function? Because
> it's not a good idea to change public functions API :-)
>
> It's not listed in any header file and import.c uses it with "extern ...".

You should be OK. Probably worth adding the missing underscore and
mentioning doing so on python-dev though - if anyone screams, you can
then add back the old (no-underscore) API easily enough.

Cheers,
Nick.

-- 
Nick Coghlan?? |?? ncoghlan at gmail.com?? |?? Brisbane, Australia

From python-checkins at python.org  Fri Jul  9 15:14:03 2010
From: python-checkins at python.org (r.david.murray)
Date: Fri,  9 Jul 2010 15:14:03 +0200 (CEST)
Subject: [Python-checkins] r82731 - python/branches/py3k/Misc/ACKS
Message-ID: <20100709131403.EE51AEED2F@mail.python.org>

Author: r.david.murray
Date: Fri Jul  9 15:14:03 2010
New Revision: 82731

Log:
Fix sort order mistake in Misc/ACKS.


Modified:
   python/branches/py3k/Misc/ACKS

Modified: python/branches/py3k/Misc/ACKS
==============================================================================
--- python/branches/py3k/Misc/ACKS	(original)
+++ python/branches/py3k/Misc/ACKS	Fri Jul  9 15:14:03 2010
@@ -145,8 +145,8 @@
 Vadim Chugunov
 David Cinege
 Mike Clarkson
-Brad Clements
 Andrew Clegg
+Brad Clements
 Steve Clift
 Nick Coghlan
 Josh Cogliati

From python-checkins at python.org  Fri Jul  9 15:16:00 2010
From: python-checkins at python.org (r.david.murray)
Date: Fri,  9 Jul 2010 15:16:00 +0200 (CEST)
Subject: [Python-checkins] r82732 - in python/branches/release27-maint:
	Lib/fnmatch.py Lib/test/test_fnmatch.py Misc/ACKS
Message-ID: <20100709131600.67E66EED2D@mail.python.org>

Author: r.david.murray
Date: Fri Jul  9 15:16:00 2010
New Revision: 82732

Log:
Merged revisions 82730 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

'merge' in the sense that this is the same fix by the same author.

........
  r82730 | r.david.murray | 2010-07-09 08:23:21 -0400 (Fri, 09 Jul 2010) | 4 lines
  
  7846: limit fnmatch pattern cache to _MAXCACHE=100 entries.
  
  Patch by Andrew Clegg.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/fnmatch.py
   python/branches/release27-maint/Lib/test/test_fnmatch.py
   python/branches/release27-maint/Misc/ACKS

Modified: python/branches/release27-maint/Lib/fnmatch.py
==============================================================================
--- python/branches/release27-maint/Lib/fnmatch.py	(original)
+++ python/branches/release27-maint/Lib/fnmatch.py	Fri Jul  9 15:16:00 2010
@@ -15,6 +15,7 @@
 __all__ = ["filter", "fnmatch","fnmatchcase","translate"]
 
 _cache = {}
+_MAXCACHE = 100
 
 def fnmatch(name, pat):
     """Test whether FILENAME matches PATTERN.
@@ -44,6 +45,8 @@
     pat=os.path.normcase(pat)
     if not pat in _cache:
         res = translate(pat)
+        if len(_cache) >= _MAXCACHE:
+            _cache.clear()
         _cache[pat] = re.compile(res)
     match=_cache[pat].match
     if os.path is posixpath:
@@ -66,6 +69,8 @@
 
     if not pat in _cache:
         res = translate(pat)
+        if len(_cache) >= _MAXCACHE:
+            _cache.clear()
         _cache[pat] = re.compile(res)
     return _cache[pat].match(name) is not None
 

Modified: python/branches/release27-maint/Lib/test/test_fnmatch.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_fnmatch.py	(original)
+++ python/branches/release27-maint/Lib/test/test_fnmatch.py	Fri Jul  9 15:16:00 2010
@@ -3,7 +3,7 @@
 from test import test_support
 import unittest
 
-from fnmatch import fnmatch, fnmatchcase
+from fnmatch import fnmatch, fnmatchcase, _MAXCACHE, _cache
 
 
 class FnmatchTestCase(unittest.TestCase):
@@ -49,6 +49,15 @@
         check('AbC', 'abc', 0, fnmatchcase)
         check('abc', 'AbC', 0, fnmatchcase)
 
+    def test_cache_clearing(self):
+        # check that caches do not grow too large
+        # http://bugs.python.org/issue7846
+
+        # string pattern cache
+        for i in range(_MAXCACHE + 1):
+            fnmatch('foo', '?' * i)
+
+        self.assertLessEqual(len(_cache), _MAXCACHE)
 
 def test_main():
     test_support.run_unittest(FnmatchTestCase)

Modified: python/branches/release27-maint/Misc/ACKS
==============================================================================
--- python/branches/release27-maint/Misc/ACKS	(original)
+++ python/branches/release27-maint/Misc/ACKS	Fri Jul  9 15:16:00 2010
@@ -144,6 +144,7 @@
 David Cinege
 Mike Clarkson
 Brad Clements
+Andrew Clegg
 Steve Clift
 Nick Coghlan
 Josh Cogliati

From python-checkins at python.org  Fri Jul  9 15:16:27 2010
From: python-checkins at python.org (r.david.murray)
Date: Fri,  9 Jul 2010 15:16:27 +0200 (CEST)
Subject: [Python-checkins] r82733 - in python/branches/release31-maint:
	Lib/fnmatch.py Lib/test/test_fnmatch.py Misc/ACKS Misc/NEWS
Message-ID: <20100709131627.14523C399@mail.python.org>

Author: r.david.murray
Date: Fri Jul  9 15:16:26 2010
New Revision: 82733

Log:
Merged revisions 82730-82731 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82730 | r.david.murray | 2010-07-09 08:23:21 -0400 (Fri, 09 Jul 2010) | 4 lines
  
  7846: limit fnmatch pattern cache to _MAXCACHE=100 entries.
  
  Patch by Andrew Clegg.
........
  r82731 | r.david.murray | 2010-07-09 09:14:03 -0400 (Fri, 09 Jul 2010) | 2 lines
  
  Fix sort order mistake in Misc/ACKS.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/fnmatch.py
   python/branches/release31-maint/Lib/test/test_fnmatch.py
   python/branches/release31-maint/Misc/ACKS
   python/branches/release31-maint/Misc/NEWS

Modified: python/branches/release31-maint/Lib/fnmatch.py
==============================================================================
--- python/branches/release31-maint/Lib/fnmatch.py	(original)
+++ python/branches/release31-maint/Lib/fnmatch.py	Fri Jul  9 15:16:26 2010
@@ -16,6 +16,7 @@
 
 _cache = {}  # Maps text patterns to compiled regexen.
 _cacheb = {}  # Ditto for bytes patterns.
+_MAXCACHE = 100 # Maximum size of caches
 
 def fnmatch(name, pat):
     """Test whether FILENAME matches PATTERN.
@@ -48,6 +49,8 @@
             res = bytes(res_str, 'ISO-8859-1')
         else:
             res = translate(pat)
+        if len(cache) >= _MAXCACHE:
+            cache.clear()
         cache[pat] = regex = re.compile(res)
     return regex.match
 

Modified: python/branches/release31-maint/Lib/test/test_fnmatch.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_fnmatch.py	(original)
+++ python/branches/release31-maint/Lib/test/test_fnmatch.py	Fri Jul  9 15:16:26 2010
@@ -3,7 +3,7 @@
 from test import support
 import unittest
 
-from fnmatch import fnmatch, fnmatchcase
+from fnmatch import fnmatch, fnmatchcase, _MAXCACHE, _cache, _cacheb
 
 
 class FnmatchTestCase(unittest.TestCase):
@@ -55,6 +55,21 @@
         self.check_match(b'test\xff', b'te*\xff')
         self.check_match(b'foo\nbar', b'foo*')
 
+    def test_cache_clearing(self):
+        # check that caches do not grow too large
+        # http://bugs.python.org/issue7846
+
+        # string pattern cache
+        for i in range(_MAXCACHE + 1):
+            fnmatch('foo', '?' * i)
+
+        self.assertLessEqual(len(_cache), _MAXCACHE)
+
+        # bytes pattern cache
+        for i in range(_MAXCACHE + 1):
+            fnmatch(b'foo', b'?' * i)
+        self.assertLessEqual(len(_cacheb), _MAXCACHE)
+
 
 def test_main():
     support.run_unittest(FnmatchTestCase)

Modified: python/branches/release31-maint/Misc/ACKS
==============================================================================
--- python/branches/release31-maint/Misc/ACKS	(original)
+++ python/branches/release31-maint/Misc/ACKS	Fri Jul  9 15:16:26 2010
@@ -137,6 +137,7 @@
 Vadim Chugunov
 David Cinege
 Mike Clarkson
+Andrew Clegg
 Brad Clements
 Steve Clift
 Nick Coghlan

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Fri Jul  9 15:16:26 2010
@@ -75,6 +75,8 @@
 Library
 -------
 
+- Issue #7646: The fnmatch pattern cache no longer grows without bound.
+
 - Issue #9136: Fix 'dictionary changed size during iteration'
   RuntimeError produced when profiling the decimal module.  This was
   due to a dangerous iteration over 'locals()' in Context.__init__.

From python-checkins at python.org  Fri Jul  9 15:17:45 2010
From: python-checkins at python.org (r.david.murray)
Date: Fri,  9 Jul 2010 15:17:45 +0200 (CEST)
Subject: [Python-checkins] r82734 - in python/branches/release27-maint:
	Misc/ACKS
Message-ID: <20100709131745.6849DEED45@mail.python.org>

Author: r.david.murray
Date: Fri Jul  9 15:17:45 2010
New Revision: 82734

Log:
Merged revisions 82731 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82731 | r.david.murray | 2010-07-09 09:14:03 -0400 (Fri, 09 Jul 2010) | 2 lines
  
  Fix sort order mistake in Misc/ACKS.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Misc/ACKS

Modified: python/branches/release27-maint/Misc/ACKS
==============================================================================
--- python/branches/release27-maint/Misc/ACKS	(original)
+++ python/branches/release27-maint/Misc/ACKS	Fri Jul  9 15:17:45 2010
@@ -143,8 +143,8 @@
 Vadim Chugunov
 David Cinege
 Mike Clarkson
-Brad Clements
 Andrew Clegg
+Brad Clements
 Steve Clift
 Nick Coghlan
 Josh Cogliati

From python-checkins at python.org  Fri Jul  9 15:20:40 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Fri,  9 Jul 2010 15:20:40 +0200 (CEST)
Subject: [Python-checkins] r82735 -
	python/branches/py3k/Lib/test/test_struct.py
Message-ID: <20100709132040.C3BD6EECB6@mail.python.org>

Author: benjamin.peterson
Date: Fri Jul  9 15:20:40 2010
New Revision: 82735

Log:
OverflowError is fine

Modified:
   python/branches/py3k/Lib/test/test_struct.py

Modified: python/branches/py3k/Lib/test/test_struct.py
==============================================================================
--- python/branches/py3k/Lib/test/test_struct.py	(original)
+++ python/branches/py3k/Lib/test/test_struct.py	Fri Jul  9 15:20:40 2010
@@ -234,7 +234,8 @@
                                                                  b'\x01' + got)
                 else:
                     # x is out of range -- verify pack realizes that.
-                    self.assertRaises(struct.error, pack, format, x)
+                    self.assertRaises((OverflowError, struct.error), pack,
+                                      format, x)
 
             def run(self):
                 from random import randrange

From python-checkins at python.org  Fri Jul  9 15:21:35 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Fri,  9 Jul 2010 15:21:35 +0200 (CEST)
Subject: [Python-checkins] r82736 - in python/branches/release27-maint:
	Lib/test/test_struct.py
Message-ID: <20100709132135.6C3F1EED22@mail.python.org>

Author: benjamin.peterson
Date: Fri Jul  9 15:21:35 2010
New Revision: 82736

Log:
Merged revisions 82735 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82735 | benjamin.peterson | 2010-07-09 08:20:40 -0500 (Fri, 09 Jul 2010) | 1 line
  
  OverflowError is fine
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/test/test_struct.py

Modified: python/branches/release27-maint/Lib/test/test_struct.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_struct.py	(original)
+++ python/branches/release27-maint/Lib/test/test_struct.py	Fri Jul  9 15:21:35 2010
@@ -243,7 +243,8 @@
                                                                  '\x01' + got)
                 else:
                     # x is out of range -- verify pack realizes that.
-                    self.assertRaises(struct.error, pack, format, x)
+                    self.assertRaises((OverflowError, struct.error), pack,
+                                      format, x)
 
             def run(self):
                 from random import randrange

From python-checkins at python.org  Fri Jul  9 15:22:07 2010
From: python-checkins at python.org (brian.curtin)
Date: Fri,  9 Jul 2010 15:22:07 +0200 (CEST)
Subject: [Python-checkins] r82737 - python/branches/py3k/Misc/ACKS
Message-ID: <20100709132207.C2D3CEECB6@mail.python.org>

Author: brian.curtin
Date: Fri Jul  9 15:22:07 2010
New Revision: 82737

Log:
Reorder Jason "two O's" Coombs


Modified:
   python/branches/py3k/Misc/ACKS

Modified: python/branches/py3k/Misc/ACKS
==============================================================================
--- python/branches/py3k/Misc/ACKS	(original)
+++ python/branches/py3k/Misc/ACKS	Fri Jul  9 15:22:07 2010
@@ -155,11 +155,11 @@
 Jeffery Collins
 Robert Collins
 Paul Colomiets
-Jason R. Coombs
 Geremy Condra
 Juan Jos? Conti
 Matt Conway
 David M. Cooke
+Jason R. Coombs
 Greg Copeland
 Aldo Cortesi
 David Costanzo

From python-checkins at python.org  Fri Jul  9 15:25:57 2010
From: python-checkins at python.org (r.david.murray)
Date: Fri,  9 Jul 2010 15:25:57 +0200 (CEST)
Subject: [Python-checkins] r82738 - python/branches/release27-maint/Misc/NEWS
Message-ID: <20100709132557.176B8EE9A0@mail.python.org>

Author: r.david.murray
Date: Fri Jul  9 15:25:56 2010
New Revision: 82738

Log:
Add missing NEWS entry for 7646.


Modified:
   python/branches/release27-maint/Misc/NEWS

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Fri Jul  9 15:25:56 2010
@@ -16,6 +16,8 @@
 Library
 -------
 
+- Issue #7646: The fnmatch pattern cache no longer grows without bound.
+
 - Issue #9136: Fix 'dictionary changed size during iteration'
   RuntimeError produced when profiling the decimal module.  This was
   due to a dangerous iteration over 'locals()' in Context.__init__.

From python-checkins at python.org  Fri Jul  9 15:28:42 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Fri,  9 Jul 2010 15:28:42 +0200 (CEST)
Subject: [Python-checkins] r82739 -
	python/branches/py3k/Lib/test/test_struct.py
Message-ID: <20100709132842.C3178EED02@mail.python.org>

Author: benjamin.peterson
Date: Fri Jul  9 15:28:42 2010
New Revision: 82739

Log:
allow more exceptions

Modified:
   python/branches/py3k/Lib/test/test_struct.py

Modified: python/branches/py3k/Lib/test/test_struct.py
==============================================================================
--- python/branches/py3k/Lib/test/test_struct.py	(original)
+++ python/branches/py3k/Lib/test/test_struct.py	Fri Jul  9 15:28:42 2010
@@ -47,8 +47,8 @@
 
         self.assertRaises(struct.error, struct.pack, 'iii', 3)
         self.assertRaises(struct.error, struct.pack, 'i', 3, 3, 3)
-        self.assertRaises(struct.error, struct.pack, 'i', 'foo')
-        self.assertRaises(struct.error, struct.pack, 'P', 'foo')
+        self.assertRaises((TypeError, struct.error), struct.pack, 'i', 'foo')
+        self.assertRaises((TypeError, struct.error), struct.pack, 'P', 'foo')
         self.assertRaises(struct.error, struct.unpack, 'd', b'flap')
         s = struct.pack('ii', 1, 2)
         self.assertRaises(struct.error, struct.unpack, 'iii', s)
@@ -446,7 +446,7 @@
 
         # Test bogus offset (issue 3694)
         sb = small_buf
-        self.assertRaises(TypeError, struct.pack_into, b'', sb, None)
+        self.assertRaises((TypeError, struct.error), struct.pack_into, b'', sb, None)
 
     def test_pack_into_fn(self):
         test_string = b'Reykjavik rocks, eow!'

From python-checkins at python.org  Fri Jul  9 15:29:33 2010
From: python-checkins at python.org (r.david.murray)
Date: Fri,  9 Jul 2010 15:29:33 +0200 (CEST)
Subject: [Python-checkins] r82740 - in python/branches/release26-maint:
	Lib/fnmatch.py Lib/test/test_fnmatch.py Misc/ACKS Misc/NEWS
Message-ID: <20100709132933.D0E1FEEC95@mail.python.org>

Author: r.david.murray
Date: Fri Jul  9 15:29:33 2010
New Revision: 82740

Log:
Merged revisions 82732,82734,82738 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

................
  r82732 | r.david.murray | 2010-07-09 09:16:00 -0400 (Fri, 09 Jul 2010) | 13 lines
  
  Merged revisions 82730 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k
  
  'merge' in the sense that this is the same fix by the same author.
  
  ........
    r82730 | r.david.murray | 2010-07-09 08:23:21 -0400 (Fri, 09 Jul 2010) | 4 lines
    
    7846: limit fnmatch pattern cache to _MAXCACHE=100 entries.
    
    Patch by Andrew Clegg.
  ........
................
  r82734 | r.david.murray | 2010-07-09 09:17:45 -0400 (Fri, 09 Jul 2010) | 9 lines
  
  Merged revisions 82731 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k
  
  ........
    r82731 | r.david.murray | 2010-07-09 09:14:03 -0400 (Fri, 09 Jul 2010) | 2 lines
    
    Fix sort order mistake in Misc/ACKS.
  ........
................
  r82738 | r.david.murray | 2010-07-09 09:25:56 -0400 (Fri, 09 Jul 2010) | 2 lines
  
  Add missing NEWS entry for 7646.
................


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Lib/fnmatch.py
   python/branches/release26-maint/Lib/test/test_fnmatch.py
   python/branches/release26-maint/Misc/ACKS
   python/branches/release26-maint/Misc/NEWS

Modified: python/branches/release26-maint/Lib/fnmatch.py
==============================================================================
--- python/branches/release26-maint/Lib/fnmatch.py	(original)
+++ python/branches/release26-maint/Lib/fnmatch.py	Fri Jul  9 15:29:33 2010
@@ -15,6 +15,7 @@
 __all__ = ["filter", "fnmatch","fnmatchcase","translate"]
 
 _cache = {}
+_MAXCACHE = 100
 
 def fnmatch(name, pat):
     """Test whether FILENAME matches PATTERN.
@@ -44,6 +45,8 @@
     pat=os.path.normcase(pat)
     if not pat in _cache:
         res = translate(pat)
+        if len(_cache) >= _MAXCACHE:
+            _cache.clear()
         _cache[pat] = re.compile(res)
     match=_cache[pat].match
     if os.path is posixpath:
@@ -66,6 +69,8 @@
 
     if not pat in _cache:
         res = translate(pat)
+        if len(_cache) >= _MAXCACHE:
+            _cache.clear()
         _cache[pat] = re.compile(res)
     return _cache[pat].match(name) is not None
 

Modified: python/branches/release26-maint/Lib/test/test_fnmatch.py
==============================================================================
--- python/branches/release26-maint/Lib/test/test_fnmatch.py	(original)
+++ python/branches/release26-maint/Lib/test/test_fnmatch.py	Fri Jul  9 15:29:33 2010
@@ -3,7 +3,7 @@
 from test import test_support
 import unittest
 
-from fnmatch import fnmatch, fnmatchcase
+from fnmatch import fnmatch, fnmatchcase, _MAXCACHE, _cache
 
 
 class FnmatchTestCase(unittest.TestCase):
@@ -49,6 +49,15 @@
         check('AbC', 'abc', 0, fnmatchcase)
         check('abc', 'AbC', 0, fnmatchcase)
 
+    def test_cache_clearing(self):
+        # check that caches do not grow too large
+        # http://bugs.python.org/issue7846
+
+        # string pattern cache
+        for i in range(_MAXCACHE + 1):
+            fnmatch('foo', '?' * i)
+
+        self.assertTrue(len(_cache) <= _MAXCACHE)
 
 def test_main():
     test_support.run_unittest(FnmatchTestCase)

Modified: python/branches/release26-maint/Misc/ACKS
==============================================================================
--- python/branches/release26-maint/Misc/ACKS	(original)
+++ python/branches/release26-maint/Misc/ACKS	Fri Jul  9 15:29:33 2010
@@ -132,6 +132,7 @@
 Vadim Chugunov
 David Cinege
 Mike Clarkson
+Andrew Clegg
 Brad Clements
 Steve Clift
 Nick Coghlan

Modified: python/branches/release26-maint/Misc/NEWS
==============================================================================
--- python/branches/release26-maint/Misc/NEWS	(original)
+++ python/branches/release26-maint/Misc/NEWS	Fri Jul  9 15:29:33 2010
@@ -81,6 +81,8 @@
 Library
 -------
 
+- Issue #7646: The fnmatch pattern cache no longer grows without bound.
+
 - Issue #9136: Fix 'dictionary changed size during iteration'
   RuntimeError produced when profiling the decimal module.  This was
   due to a dangerous iteration over 'locals()' in Context.__init__.

From python-checkins at python.org  Fri Jul  9 15:31:11 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Fri,  9 Jul 2010 15:31:11 +0200 (CEST)
Subject: [Python-checkins] r82741 -
	python/branches/py3k/Lib/test/test_struct.py
Message-ID: <20100709133111.AE70EEED02@mail.python.org>

Author: benjamin.peterson
Date: Fri Jul  9 15:31:11 2010
New Revision: 82741

Log:
wrap

Modified:
   python/branches/py3k/Lib/test/test_struct.py

Modified: python/branches/py3k/Lib/test/test_struct.py
==============================================================================
--- python/branches/py3k/Lib/test/test_struct.py	(original)
+++ python/branches/py3k/Lib/test/test_struct.py	Fri Jul  9 15:31:11 2010
@@ -446,7 +446,8 @@
 
         # Test bogus offset (issue 3694)
         sb = small_buf
-        self.assertRaises((TypeError, struct.error), struct.pack_into, b'', sb, None)
+        self.assertRaises((TypeError, struct.error), struct.pack_into, b'', sb,
+                          None)
 
     def test_pack_into_fn(self):
         test_string = b'Reykjavik rocks, eow!'

From python-checkins at python.org  Fri Jul  9 15:33:03 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Fri,  9 Jul 2010 15:33:03 +0200 (CEST)
Subject: [Python-checkins] r82742 - in python/branches/release27-maint:
	Lib/test/test_struct.py
Message-ID: <20100709133303.A4629EED49@mail.python.org>

Author: benjamin.peterson
Date: Fri Jul  9 15:33:03 2010
New Revision: 82742

Log:
Merged revisions 82739,82741 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82739 | benjamin.peterson | 2010-07-09 08:28:42 -0500 (Fri, 09 Jul 2010) | 1 line
  
  allow more exceptions
........
  r82741 | benjamin.peterson | 2010-07-09 08:31:11 -0500 (Fri, 09 Jul 2010) | 1 line
  
  wrap
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/test/test_struct.py

Modified: python/branches/release27-maint/Lib/test/test_struct.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_struct.py	(original)
+++ python/branches/release27-maint/Lib/test/test_struct.py	Fri Jul  9 15:33:03 2010
@@ -62,8 +62,8 @@
 
         self.assertRaises(struct.error, struct.pack, 'iii', 3)
         self.assertRaises(struct.error, struct.pack, 'i', 3, 3, 3)
-        self.assertRaises(struct.error, struct.pack, 'i', 'foo')
-        self.assertRaises(struct.error, struct.pack, 'P', 'foo')
+        self.assertRaises((TypeError, struct.error), struct.pack, 'i', 'foo')
+        self.assertRaises((TypeError, struct.error), struct.pack, 'P', 'foo')
         self.assertRaises(struct.error, struct.unpack, 'd', 'flap')
         s = struct.pack('ii', 1, 2)
         self.assertRaises(struct.error, struct.unpack, 'iii', s)
@@ -458,7 +458,8 @@
 
         # Test bogus offset (issue 3694)
         sb = small_buf
-        self.assertRaises(TypeError, struct.pack_into, b'1', sb, None)
+        self.assertRaises((TypeError, struct.error), struct.pack_into, b'', sb,
+                          None)
 
     def test_pack_into_fn(self):
         test_string = 'Reykjavik rocks, eow!'

From python-checkins at python.org  Fri Jul  9 15:54:27 2010
From: python-checkins at python.org (brian.curtin)
Date: Fri,  9 Jul 2010 15:54:27 +0200 (CEST)
Subject: [Python-checkins] r82743 - in python/branches/py3k/Lib: tarfile.py
	test/test_posixpath.py
Message-ID: <20100709135427.7483EEED6A@mail.python.org>

Author: brian.curtin
Date: Fri Jul  9 15:54:27 2010
New Revision: 82743

Log:
Fix the breakage of Lib/tarfile.py on non-Windows platforms due to 
using WindowsError in a try/except. Only add WindowsError to the list of
exceptions to catch when we are actually running on Windows.

Additionally, add a call that was left out in test_posixpath.

Thanks Amaury, Antoine, and Jason.


Modified:
   python/branches/py3k/Lib/tarfile.py
   python/branches/py3k/Lib/test/test_posixpath.py

Modified: python/branches/py3k/Lib/tarfile.py
==============================================================================
--- python/branches/py3k/Lib/tarfile.py	(original)
+++ python/branches/py3k/Lib/tarfile.py	Fri Jul  9 15:54:27 2010
@@ -55,6 +55,15 @@
 except ImportError:
     grp = pwd = None
 
+# os.symlink on Windows prior to 6.0 raises NotImplementedError
+symlink_exception = (AttributeError, NotImplementedError)
+try:
+    # WindowsError (1314) will be raised if the caller does not hold the
+    # SeCreateSymbolicLinkPrivilege privilege
+    symlink_exception += (WindowsError,)
+except NameError:
+    pass
+
 # from tarfile import *
 __all__ = ["TarFile", "TarInfo", "is_tarfile", "TarError"]
 
@@ -2283,17 +2292,16 @@
                     os.link(tarinfo._link_target, targetpath)
                 else:
                     self._extract_mem
-        except (AttributeError, NotImplementedError, WindowsError):
-            # AttributeError if no os.symlink
-            # NotImplementedError if on Windows XP
-            # WindowsError (1314) if the required privilege is not held by the client
+        except symlink_exception:
             if tarinfo.issym():
-                linkpath = os.path.join(os.path.dirname(tarinfo.name),tarinfo.linkname)
+                linkpath = os.path.join(os.path.dirname(tarinfo.name),
+                                        tarinfo.linkname)
             else:
                 linkpath = tarinfo.linkname
         else:
             try:
-                self._extract_member(self._find_link_target(tarinfo), targetpath)
+                self._extract_member(self._find_link_target(tarinfo),
+                                     targetpath)
             except KeyError:
                 raise ExtractError("unable to resolve link inside archive")
 

Modified: python/branches/py3k/Lib/test/test_posixpath.py
==============================================================================
--- python/branches/py3k/Lib/test/test_posixpath.py	(original)
+++ python/branches/py3k/Lib/test/test_posixpath.py	Fri Jul  9 15:54:27 2010
@@ -208,6 +208,7 @@
     def test_samestat_on_links(self):
         test_fn1 = support.TESTFN + "1"
         test_fn2 = support.TESTFN + "2"
+        self._create_file(test_fn1)
         test_fns = (test_fn1, test_fn2)
         os.symlink(*test_fns)
         stats = map(os.stat, test_fns)

From python-checkins at python.org  Fri Jul  9 17:15:09 2010
From: python-checkins at python.org (brian.curtin)
Date: Fri,  9 Jul 2010 17:15:09 +0200 (CEST)
Subject: [Python-checkins] r82744 - in python/branches/py3k/Doc/library:
	os.path.rst os.rst
Message-ID: <20100709151509.733AEEED2C@mail.python.org>

Author: brian.curtin
Date: Fri Jul  9 17:15:09 2010
New Revision: 82744

Log:
Adding versionchanged to the various Windows os.symlink additions, along
with a few minor touchups.


Modified:
   python/branches/py3k/Doc/library/os.path.rst
   python/branches/py3k/Doc/library/os.rst

Modified: python/branches/py3k/Doc/library/os.path.rst
==============================================================================
--- python/branches/py3k/Doc/library/os.path.rst	(original)
+++ python/branches/py3k/Doc/library/os.path.rst	Fri Jul  9 17:15:09 2010
@@ -236,11 +236,13 @@
    exception if a :func:`os.stat` call on either pathname fails.
 
    On Windows, two files are the same if they resolve to the same final path
-   name using the Windows API call GetFinalPathNameByHandle and this function
+   name using the Windows API call GetFinalPathNameByHandle. This function
    raises an exception if handles cannot be obtained to either file.
 
    Availability: Windows, Unix.
 
+   .. versionchanged:: 3.2 Added Windows support
+
 
 .. function:: sameopenfile(fp1, fp2)
 

Modified: python/branches/py3k/Doc/library/os.rst
==============================================================================
--- python/branches/py3k/Doc/library/os.rst	(original)
+++ python/branches/py3k/Doc/library/os.rst	Fri Jul  9 17:15:09 2010
@@ -1064,8 +1064,9 @@
 .. function:: lstat(path)
 
    Like :func:`stat`, but do not follow symbolic links.  This is an alias for
-   :func:`stat` on platforms that do not support symbolic links, such as
-   Windows prior to 6.0 (Vista).
+   :func:`stat` on platforms that do not support symbolic links.
+
+   .. versionchanged:: 3.2 Added support for Windows 6.0 (Vista)
 
 
 .. function:: mkfifo(path[, mode])
@@ -1181,7 +1182,9 @@
    and the call may raise an UnicodeDecodeError. If the *path* is a bytes
    object, the result will be a bytes object.
 
-   Availability: Unix, Windows.
+   Availability: Unix, Windows
+
+   .. versionchanged:: 3.2 Added support for Windows 6.0 (Vista)
 
 
 .. function:: remove(path)
@@ -1341,8 +1344,9 @@
 
 .. function:: symlink(source, link_name)
 
-   Create a symbolic link pointing to *source* named *link_name*.  On Windows,
-   symlink version takes an additional, optional parameter,
+   Create a symbolic link pointing to *source* named *link_name*.
+
+   On Windows, symlink version takes an additional, optional parameter,
    *target_is_directory*, which defaults to False.
 
    symlink(source, link_name, target_is_directory=False)
@@ -1359,7 +1363,9 @@
    will raise a NotImplementedError on Windows versions earlier than 6.0. The
    SeCreateSymbolicLinkPrivilege is required in order to create symlinks.
 
-   Availability:  Unix, Windows 6.0.
+   Availability:  Unix, Windows
+
+   .. versionchanged:: 3.2 Added support for Windows 6.0 (Vista)
 
 
 .. function:: unlink(path)

From python-checkins at python.org  Fri Jul  9 17:58:59 2010
From: python-checkins at python.org (brian.curtin)
Date: Fri,  9 Jul 2010 17:58:59 +0200 (CEST)
Subject: [Python-checkins] r82745 - in python/branches/py3k:
	Lib/test/symlink_support.py Lib/test/test_os.py
	Lib/test/test_platform.py Lib/test/test_sysconfig.py
	Lib/test/test_tarfile.py Modules/posixmodule.c
Message-ID: <20100709155859.E06E2EEDB2@mail.python.org>

Author: brian.curtin
Date: Fri Jul  9 17:58:59 2010
New Revision: 82745

Log:
Re-flow several long lines from #1578269.


Modified:
   python/branches/py3k/Lib/test/symlink_support.py
   python/branches/py3k/Lib/test/test_os.py
   python/branches/py3k/Lib/test/test_platform.py
   python/branches/py3k/Lib/test/test_sysconfig.py
   python/branches/py3k/Lib/test/test_tarfile.py
   python/branches/py3k/Modules/posixmodule.c

Modified: python/branches/py3k/Lib/test/symlink_support.py
==============================================================================
--- python/branches/py3k/Lib/test/symlink_support.py	(original)
+++ python/branches/py3k/Lib/test/symlink_support.py	Fri Jul  9 17:58:59 2010
@@ -12,7 +12,8 @@
 GetCurrentProcess = ctypes.windll.kernel32.GetCurrentProcess
 GetCurrentProcess.restype = wintypes.HANDLE
 OpenProcessToken = ctypes.windll.advapi32.OpenProcessToken
-OpenProcessToken.argtypes = (wintypes.HANDLE, wintypes.DWORD, ctypes.POINTER(wintypes.HANDLE))
+OpenProcessToken.argtypes = (wintypes.HANDLE, wintypes.DWORD,
+                             ctypes.POINTER(wintypes.HANDLE))
 OpenProcessToken.restype = wintypes.BOOL
 
 class LUID(ctypes.Structure):
@@ -91,7 +92,8 @@
 
     def get_array(self):
         array_type = LUID_AND_ATTRIBUTES*self.count
-        privileges = ctypes.cast(self.privileges, ctypes.POINTER(array_type)).contents
+        privileges = ctypes.cast(self.privileges,
+                                 ctypes.POINTER(array_type)).contents
         return privileges
 
     def __iter__(self):
@@ -133,7 +135,8 @@
 def get_symlink_luid():
     "Get the LUID for the SeCreateSymbolicLinkPrivilege"
     symlink_luid = LUID()
-    res = LookupPrivilegeValue(None, "SeCreateSymbolicLinkPrivilege", symlink_luid)
+    res = LookupPrivilegeValue(None, "SeCreateSymbolicLinkPrivilege",
+                               symlink_luid)
     if not res > 0:
         raise RuntimeError("Couldn't lookup privilege value")
     return symlink_luid

Modified: python/branches/py3k/Lib/test/test_os.py
==============================================================================
--- python/branches/py3k/Lib/test/test_os.py	(original)
+++ python/branches/py3k/Lib/test/test_os.py	Fri Jul  9 17:58:59 2010
@@ -1034,7 +1034,8 @@
 
 
 def skipUnlessWindows6(test):
-    if hasattr(sys, 'getwindowsversion') and sys.getwindowsversion().major >= 6:
+    if (hasattr(sys, 'getwindowsversion')
+        and sys.getwindowsversion().major >= 6):
         return test
     return unittest.skip("Requires Windows Vista or later")(test)
 

Modified: python/branches/py3k/Lib/test/test_platform.py
==============================================================================
--- python/branches/py3k/Lib/test/test_platform.py	(original)
+++ python/branches/py3k/Lib/test/test_platform.py	Fri Jul  9 17:58:59 2010
@@ -15,8 +15,8 @@
         # On Windows, the EXE needs to know where pythonXY.dll is at so we have
         # to add the directory to the path.
         if sys.platform == "win32":
-            os.environ["Path"] = "{};{}".format(os.path.dirname(sys.executable),
-                                                os.environ["Path"])
+            os.environ["Path"] = "{};{}".format(
+                os.path.dirname(sys.executable), os.environ["Path"])
 
         def get(python):
             cmd = [python, '-c',

Modified: python/branches/py3k/Lib/test/test_sysconfig.py
==============================================================================
--- python/branches/py3k/Lib/test/test_sysconfig.py	(original)
+++ python/branches/py3k/Lib/test/test_sysconfig.py	Fri Jul  9 17:58:59 2010
@@ -244,8 +244,8 @@
         # On Windows, the EXE needs to know where pythonXY.dll is at so we have
         # to add the directory to the path.
         if sys.platform == "win32":
-            os.environ["Path"] = "{};{}".format(os.path.dirname(sys.executable),
-                                                os.environ["Path"])
+            os.environ["Path"] = "{};{}".format(
+                os.path.dirname(sys.executable), os.environ["Path"])
 
         # Issue 7880
         def get(python):

Modified: python/branches/py3k/Lib/test/test_tarfile.py
==============================================================================
--- python/branches/py3k/Lib/test/test_tarfile.py	(original)
+++ python/branches/py3k/Lib/test/test_tarfile.py	Fri Jul  9 17:58:59 2010
@@ -291,7 +291,8 @@
         self.assertTrue(self.tar.getmembers()[-1].name == "misc/eof",
                 "could not find all members")
 
-    @unittest.skipUnless(hasattr(os, "link"), "Missing hardlink implementation")
+    @unittest.skipUnless(hasattr(os, "link"),
+                         "Missing hardlink implementation")
     @support.skip_unless_symlink
     def test_extract_hardlink(self):
         # Test hardlink extraction (e.g. bug #857297).
@@ -1423,11 +1424,13 @@
     def test_hardlink_extraction2(self):
         self._test_link_extraction("./ustar/linktest2/lnktype")
 
-    @unittest.skipIf(hasattr(os, "symlink"), "Skip emulation if symlink exists")
+    @unittest.skipIf(hasattr(os, "symlink"),
+                     "Skip emulation if symlink exists")
     def test_symlink_extraction1(self):
         self._test_link_extraction("ustar/symtype")
 
-    @unittest.skipIf(hasattr(os, "symlink"), "Skip emulation if symlink exists")
+    @unittest.skipIf(hasattr(os, "symlink"),
+                     "Skip emulation if symlink exists")
     def test_symlink_extraction2(self):
         self._test_link_extraction("./ustar/linktest2/symtype")
 

Modified: python/branches/py3k/Modules/posixmodule.c
==============================================================================
--- python/branches/py3k/Modules/posixmodule.c	(original)
+++ python/branches/py3k/Modules/posixmodule.c	Fri Jul  9 17:58:59 2010
@@ -1117,7 +1117,8 @@
                                                 "GetFinalPathNameByHandleA");
         *(FARPROC*)&Py_GetFinalPathNameByHandleW = GetProcAddress(hKernel32,
                                                 "GetFinalPathNameByHandleW");
-        has_GetFinalPathNameByHandle = Py_GetFinalPathNameByHandleA && Py_GetFinalPathNameByHandleW;
+        has_GetFinalPathNameByHandle = Py_GetFinalPathNameByHandleA &&
+                                       Py_GetFinalPathNameByHandleW;
     }
     return has_GetFinalPathNameByHandle;
 }
@@ -1310,9 +1311,12 @@
     /* similar to stat() */
     result->st_mode = attributes_to_mode(info.dwFileAttributes);
     result->st_size = (((__int64)info.nFileSizeHigh)<<32) + info.nFileSizeLow;
-    FILE_TIME_to_time_t_nsec(&info.ftCreationTime, &result->st_ctime, &result->st_ctime_nsec);
-    FILE_TIME_to_time_t_nsec(&info.ftLastWriteTime, &result->st_mtime, &result->st_mtime_nsec);
-    FILE_TIME_to_time_t_nsec(&info.ftLastAccessTime, &result->st_atime, &result->st_atime_nsec);
+    FILE_TIME_to_time_t_nsec(&info.ftCreationTime, &result->st_ctime,
+                             &result->st_ctime_nsec);
+    FILE_TIME_to_time_t_nsec(&info.ftLastWriteTime, &result->st_mtime,
+                             &result->st_mtime_nsec);
+    FILE_TIME_to_time_t_nsec(&info.ftLastAccessTime, &result->st_atime,
+                             &result->st_atime_nsec);
     /* specific to fstat() */
     result->st_nlink = info.nNumberOfLinks;
     result->st_ino = (((__int64)info.nFileIndexHigh)<<32) + info.nFileIndexLow;
@@ -2691,7 +2695,8 @@
     
     if(hFile == INVALID_HANDLE_VALUE) {
         return win32_error_unicode("GetFinalPathNamyByHandle", path);
-        return PyErr_Format(PyExc_RuntimeError, "Could not get a handle to file.");
+        return PyErr_Format(PyExc_RuntimeError,
+                            "Could not get a handle to file.");
     }
 
     /* We have a good handle to the target, use it to determine the
@@ -3005,7 +3010,8 @@
 posix_unlink(PyObject *self, PyObject *args)
 {
 #ifdef MS_WINDOWS
-    return win32_1str(args, "remove", "y:remove", DeleteFileA, "U:remove", Py_DeleteFileW);
+    return win32_1str(args, "remove", "y:remove", DeleteFileA,
+                      "U:remove", Py_DeleteFileW);
 #else
     return posix_1str(args, "O&:remove", unlink);
 #endif
@@ -4959,7 +4965,8 @@
     };
 } REPARSE_DATA_BUFFER, *PREPARSE_DATA_BUFFER;
 
-#define REPARSE_DATA_BUFFER_HEADER_SIZE  FIELD_OFFSET(REPARSE_DATA_BUFFER, GenericReparseBuffer)
+#define REPARSE_DATA_BUFFER_HEADER_SIZE  FIELD_OFFSET(REPARSE_DATA_BUFFER,\
+                                                      GenericReparseBuffer)
 
 #define MAXIMUM_REPARSE_DATA_BUFFER_SIZE  ( 16 * 1024 )
 
@@ -5023,8 +5030,11 @@
                 "not a symbolic link");
         return NULL;
     }
-    print_name = rdb->SymbolicLinkReparseBuffer.PathBuffer + rdb->SymbolicLinkReparseBuffer.PrintNameOffset;
-    result = PyUnicode_FromWideChar(print_name, rdb->SymbolicLinkReparseBuffer.PrintNameLength/2);
+    print_name = rdb->SymbolicLinkReparseBuffer.PathBuffer +
+                 rdb->SymbolicLinkReparseBuffer.PrintNameOffset;
+
+    result = PyUnicode_FromWideChar(print_name,
+                    rdb->SymbolicLinkReparseBuffer.PrintNameLength/2);
     return result;
 }
 
@@ -5043,7 +5053,8 @@
     if (has_CreateSymbolicLinkW)
         return has_CreateSymbolicLinkW;
     hKernel32 = GetModuleHandle("KERNEL32");
-    *(FARPROC*)&Py_CreateSymbolicLinkW = GetProcAddress(hKernel32, "CreateSymbolicLinkW");
+    *(FARPROC*)&Py_CreateSymbolicLinkW = GetProcAddress(hKernel32,
+                                                        "CreateSymbolicLinkW");
     if (Py_CreateSymbolicLinkW)
         has_CreateSymbolicLinkW = 1;
     return has_CreateSymbolicLinkW;
@@ -7586,7 +7597,8 @@
     {"symlink",         posix_symlink, METH_VARARGS, posix_symlink__doc__},
 #endif /* HAVE_SYMLINK */
 #if !defined(HAVE_SYMLINK) && defined(MS_WINDOWS)
-    {"symlink", (PyCFunction)win_symlink, METH_VARARGS | METH_KEYWORDS, win_symlink__doc__},
+    {"symlink", (PyCFunction)win_symlink, METH_VARARGS | METH_KEYWORDS,
+                win_symlink__doc__},
 #endif /* !defined(HAVE_SYMLINK) && defined(MS_WINDOWS) */
 #ifdef HAVE_SYSTEM
     {"system",          posix_system, METH_VARARGS, posix_system__doc__},

From brian.curtin at gmail.com  Fri Jul  9 18:02:10 2010
From: brian.curtin at gmail.com (Brian Curtin)
Date: Fri, 9 Jul 2010 11:02:10 -0500
Subject: [Python-checkins] r82659 - in python/branches/py3k:
	Doc/library/os.path.rst Doc/library/os.rst Lib/ntpath.py
	Lib/tarfile.py Lib/test/support.py
	Lib/test/symlink_support.py Lib/test/test_glob.py
	Lib/test/test_httpservers.py Lib/test/test_os.py Lib
Message-ID: <AANLkTik4khBczBc2ppoS_7oOBubd9IX0m8nnu9u7gfP6@mail.gmail.com>

On Fri, Jul 9, 2010 at 02:09, Georg Brandl <g.brandl at gmx.net> wrote:

> Am 08.07.2010 23:39, schrieb brian.curtin:
> > Author: brian.curtin
> > Date: Thu Jul  8 23:39:08 2010
> > New Revision: 82659
> >
> > Log:
> > Implement #1578269. Patch by Jason R. Coombs.
> >
> > Added Windows support for os.symlink when run on Windows 6.0 or greater,
> > aka Vista. Previous Windows versions will raise NotImplementedError
> > when trying to symlink.
> >
> > Includes numerous test updates and additions to test_os, including
> > a symlink_support module because of the fact that privilege escalation
> > is required in order to run the tests to ensure that the user is able
> > to create symlinks. By default, accounts do not have the required
> > privilege, so the escalation code will have to be exposed later (or
> > documented on how to do so). I'll be following up with that work next.
> >
> > Note that the tests use ctypes, which was agreed on during the PyCon
> > language summit.
> >
> >
> >
> > Added:
> >    python/branches/py3k/Lib/test/symlink_support.py   (contents, props
> changed)
> > Modified:
> >    python/branches/py3k/Doc/library/os.path.rst
> >    python/branches/py3k/Doc/library/os.rst
> >    python/branches/py3k/Lib/ntpath.py
> >    python/branches/py3k/Lib/tarfile.py
> >    python/branches/py3k/Lib/test/support.py
> >    python/branches/py3k/Lib/test/test_glob.py
> >    python/branches/py3k/Lib/test/test_httpservers.py
> >    python/branches/py3k/Lib/test/test_os.py
> >    python/branches/py3k/Lib/test/test_platform.py
> >    python/branches/py3k/Lib/test/test_posixpath.py
> >    python/branches/py3k/Lib/test/test_shutil.py
> >    python/branches/py3k/Lib/test/test_sys.py
> >    python/branches/py3k/Lib/test/test_sysconfig.py
> >    python/branches/py3k/Lib/test/test_tarfile.py
> >    python/branches/py3k/Misc/ACKS
> >    python/branches/py3k/Misc/NEWS
> >    python/branches/py3k/Modules/posixmodule.c
> >
> > Modified: python/branches/py3k/Doc/library/os.path.rst
> >
> ==============================================================================
> > --- python/branches/py3k/Doc/library/os.path.rst      (original)
> > +++ python/branches/py3k/Doc/library/os.path.rst      Thu Jul  8 23:39:08
> 2010
> > @@ -231,11 +231,15 @@
> >
> >  .. function:: samefile(path1, path2)
> >
> > -   Return ``True`` if both pathname arguments refer to the same file or
> directory
> > -   (as indicated by device number and i-node number). Raise an exception
> if a
> > -   :func:`os.stat` call on either pathname fails.
> > +   Return ``True`` if both pathname arguments refer to the same file or
> directory.
> > +   On Unix, this is determined by the device number and i-node number
> and raises an
> > +   exception if a :func:`os.stat` call on either pathname fails.
> > +
> > +   On Windows, two files are the same if they resolve to the same final
> path
> > +   name using the Windows API call GetFinalPathNameByHandle and this
> function
> > +   raises an exception if handles cannot be obtained to either file.
> >
> > -   Availability: Unix.
> > +   Availability: Windows, Unix.
>
> There is a ".. versionchanged" missing here (and in the other places where
> Windows availability was added).
>
> > Modified: python/branches/py3k/Lib/tarfile.py
> >
> ==============================================================================
> > --- python/branches/py3k/Lib/tarfile.py       (original)
> > +++ python/branches/py3k/Lib/tarfile.py       Thu Jul  8 23:39:08 2010
> > @@ -2273,7 +2273,7 @@
> >            (platform limitation), we try to make a copy of the referenced
> file
> >            instead of a link.
> >          """
> > -        if hasattr(os, "symlink") and hasattr(os, "link"):
> > +        try:
> >              # For systems that support symbolic and hard links.
> >              if tarinfo.issym():
> >                  os.symlink(tarinfo.linkname, targetpath)
> > @@ -2282,7 +2282,15 @@
> >                  if os.path.exists(tarinfo._link_target):
> >                      os.link(tarinfo._link_target, targetpath)
> >                  else:
> > -
>  self._extract_member(self._find_link_target(tarinfo), targetpath)
> > +                    self._extract_mem
> > +        except (AttributeError, NotImplementedError, WindowsError):
> > +            # AttributeError if no os.symlink
> > +            # NotImplementedError if on Windows XP
> > +            # WindowsError (1314) if the required privilege is not held
> by the client
> > +            if tarinfo.issym():
> > +                linkpath = os.path.join(os.path.dirname(tarinfo.name
> ),tarinfo.linkname)
> > +            else:
> > +                linkpath = tarinfo.linkname
> >          else:
> >              try:
> >                  self._extract_member(self._find_link_target(tarinfo),
> targetpath)
>
> Please observe PEP 8 (no overlong lines, space after comma).  (I also see
> overlong lines in other files in the diff.)
>
> > Modified: python/branches/py3k/Misc/ACKS
> >
> ==============================================================================
> > --- python/branches/py3k/Misc/ACKS    (original)
> > +++ python/branches/py3k/Misc/ACKS    Thu Jul  8 23:39:08 2010
> > @@ -154,6 +154,7 @@
> >  Jeffery Collins
> >  Robert Collins
> >  Paul Colomiets
> > +Jason R. Coombs
> >  Geremy Condra
> >  Juan Jos? Conti
> >  Matt Conway
>
> This doesn't seem to be the correct ordering :)
>
> Georg


Changed the docs in r82744, fixed the various line length overflows in
r82745, and re-ordered Jason's name in r82737.

Thanks for taking a look.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-checkins/attachments/20100709/1c38fa5e/attachment.html>

From python-checkins at python.org  Fri Jul  9 18:30:58 2010
From: python-checkins at python.org (jeffrey.yasskin)
Date: Fri,  9 Jul 2010 18:30:58 +0200 (CEST)
Subject: [Python-checkins] r82746 - in python/branches/py3k:
	Lib/sysconfig.py Makefile.pre.in Misc/NEWS Modules/makesetup
	configure configure.in
Message-ID: <20100709163058.9A02BEED94@mail.python.org>

Author: jeffrey.yasskin
Date: Fri Jul  9 18:30:58 2010
New Revision: 82746

Log:
Issue #9189: Allow users to set $CFLAGS, $CPPFLAGS, and $LDFLAGS when running
configure to append to Python's default values for those variables, and
similarly allow users to set $XXFLAGS on the make command line to append to the
values set by configure.

In the makefile, this renames the variables that used to be $XXFLAGS to
$PY_XXFLAGS, and renames the old $PY_CFLAGS to $PY_CORE_CFLAGS.  To compensate,
sysconfig now aliases $XXFLAGS=$PY_XXFLAGS so that scripts using it keep
working.  I see that as the right interface, not a backward-compatibility hack,
since these are logically the $XXFLAGS variables; we just use a different name
in the makefile to deal with make's semantics.


Modified:
   python/branches/py3k/Lib/sysconfig.py
   python/branches/py3k/Makefile.pre.in
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/Modules/makesetup
   python/branches/py3k/configure
   python/branches/py3k/configure.in

Modified: python/branches/py3k/Lib/sysconfig.py
==============================================================================
--- python/branches/py3k/Lib/sysconfig.py	(original)
+++ python/branches/py3k/Lib/sysconfig.py	Fri Jul  9 18:30:58 2010
@@ -259,6 +259,11 @@
                 # bogus variable reference; just drop it since we can't deal
                 variables.remove(name)
 
+    # Add in CFLAGS, LDFLAGS, and CPPFLAGS, which are named with a
+    # prefix in the Makefile.
+    for var in ('CFLAGS', 'LDFLAGS', 'CPPFLAGS'):
+        done[var] = done['PY_' + var]
+
     # save the results in the global dictionary
     vars.update(done)
     return vars

Modified: python/branches/py3k/Makefile.pre.in
==============================================================================
--- python/branches/py3k/Makefile.pre.in	(original)
+++ python/branches/py3k/Makefile.pre.in	Fri Jul  9 18:30:58 2010
@@ -59,12 +59,18 @@
 # Compiler options
 OPT=		@OPT@
 BASECFLAGS=	@BASECFLAGS@
-CFLAGS=		$(BASECFLAGS) @CFLAGS@ $(OPT) $(EXTRA_CFLAGS)
+CONFIGURE_CFLAGS=	@CFLAGS@
+CONFIGURE_CPPFLAGS=	@CPPFLAGS@
+CONFIGURE_LDFLAGS=	@LDFLAGS@
+# Avoid assigning CFLAGS, LDFLAGS, etc. so users can use them on the
+# command line to append to these values without stomping the pre-set
+# values.
+PY_CFLAGS=	$(BASECFLAGS) $(OPT) $(CONFIGURE_CFLAGS) $(CFLAGS) $(EXTRA_CFLAGS)
 # Both CPPFLAGS and LDFLAGS need to contain the shell's value for setup.py to
 # be able to build extension modules using the directories specified in the
 # environment variables
-CPPFLAGS=	-I. -IInclude -I$(srcdir)/Include @CPPFLAGS@
-LDFLAGS=	@LDFLAGS@
+PY_CPPFLAGS=	-I. -IInclude -I$(srcdir)/Include $(CONFIGURE_CPPFLAGS) $(CPPFLAGS)
+PY_LDFLAGS=	$(CONFIGURE_LDFLAGS) $(LDFLAGS)
 LDLAST=		@LDLAST@
 SGI_ABI=	@SGI_ABI@
 CCSHARED=	@CCSHARED@
@@ -73,7 +79,7 @@
 # Extra C flags added for building the interpreter object files.
 CFLAGSFORSHARED=@CFLAGSFORSHARED@
 # C flags used for building the interpreter object files
-PY_CFLAGS=	$(CFLAGS) $(CPPFLAGS) $(CFLAGSFORSHARED) -DPy_BUILD_CORE
+PY_CORE_CFLAGS=	$(PY_CFLAGS) $(PY_CPPFLAGS) $(CFLAGSFORSHARED) -DPy_BUILD_CORE
 
 
 # Machine-dependent subdirectories
@@ -411,7 +417,7 @@
 
 # Build the interpreter
 $(BUILDPYTHON):	Modules/python.o $(LIBRARY) $(LDLIBRARY)
-		$(LINKCC) $(LDFLAGS) $(LINKFORSHARED) -o $@ \
+		$(LINKCC) $(PY_LDFLAGS) $(LINKFORSHARED) -o $@ \
 			Modules/python.o \
 			$(BLDLIBRARY) $(LIBS) $(MODLIBS) $(SYSLIBS) $(LDLAST)
 
@@ -422,8 +428,8 @@
 # Build the shared modules
 sharedmods: $(BUILDPYTHON)
 	@case $$MAKEFLAGS in \
-	*s*) $(RUNSHARED) CC='$(CC)' LDSHARED='$(BLDSHARED)' LDFLAGS='$(LDFLAGS)' OPT='$(OPT)' ./$(BUILDPYTHON) -E $(srcdir)/setup.py -q build;; \
-	*) $(RUNSHARED) CC='$(CC)' LDSHARED='$(BLDSHARED)' LDFLAGS='$(LDFLAGS)' OPT='$(OPT)' ./$(BUILDPYTHON) -E $(srcdir)/setup.py build;; \
+	*s*) $(RUNSHARED) CC='$(CC)' LDSHARED='$(BLDSHARED)' LDFLAGS='$(PY_LDFLAGS)' OPT='$(OPT)' ./$(BUILDPYTHON) -E $(srcdir)/setup.py -q build;; \
+	*) $(RUNSHARED) CC='$(CC)' LDSHARED='$(BLDSHARED)' LDFLAGS='$(PY_LDFLAGS)' OPT='$(OPT)' ./$(BUILDPYTHON) -E $(srcdir)/setup.py build;; \
 	esac
 
 # Build static library
@@ -440,18 +446,18 @@
 
 libpython$(VERSION).so: $(LIBRARY_OBJS)
 	if test $(INSTSONAME) != $(LDLIBRARY); then \
-		$(LDSHARED) $(LDFLAGS) -Wl,-h$(INSTSONAME) -o $(INSTSONAME) $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
+		$(LDSHARED) $(PY_LDFLAGS) -Wl,-h$(INSTSONAME) -o $(INSTSONAME) $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
 		$(LN) -f $(INSTSONAME) $@; \
 	else \
-		$(LDSHARED) $(LDFLAGS) -o $@ $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
+		$(LDSHARED) $(PY_LDFLAGS) -o $@ $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
 	fi
 
 libpython$(VERSION).dylib: $(LIBRARY_OBJS)
-	 $(CC) -dynamiclib -Wl,-single_module $(LDFLAGS) -undefined dynamic_lookup -Wl,-install_name,$(prefix)/lib/libpython$(VERSION).dylib -Wl,-compatibility_version,$(VERSION) -Wl,-current_version,$(VERSION) -o $@ $(LIBRARY_OBJS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
+	 $(CC) -dynamiclib -Wl,-single_module $(PY_LDFLAGS) -undefined dynamic_lookup -Wl,-install_name,$(prefix)/lib/libpython$(VERSION).dylib -Wl,-compatibility_version,$(VERSION) -Wl,-current_version,$(VERSION) -o $@ $(LIBRARY_OBJS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
 		 
 
 libpython$(VERSION).sl: $(LIBRARY_OBJS)
-	$(LDSHARED) $(LDFLAGS) -o $@ $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST)
+	$(LDSHARED) $(PY_LDFLAGS) -o $@ $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST)
 
 # Copy up the gdb python hooks into a position where they can be automatically
 # loaded by gdb during Lib/test/test_gdb.py
@@ -497,7 +503,7 @@
 # for a shared core library; otherwise, this rule is a noop.
 $(DLLLIBRARY) libpython$(VERSION).dll.a: $(LIBRARY_OBJS)
 	if test -n "$(DLLLIBRARY)"; then \
-		$(LDSHARED) $(LDFLAGS) -Wl,--out-implib=$@ -o $(DLLLIBRARY) $^ \
+		$(LDSHARED) $(PY_LDFLAGS) -Wl,--out-implib=$@ -o $(DLLLIBRARY) $^ \
 			$(LIBS) $(MODLIBS) $(SYSLIBS) $(LDLAST); \
 	else true; \
 	fi
@@ -541,10 +547,10 @@
 		$(SIGNAL_OBJS) \
 		$(MODOBJS) \
 		$(srcdir)/Modules/getbuildinfo.c
-	$(CC) -c $(PY_CFLAGS) -DSVNVERSION="\"`LC_ALL=C $(SVNVERSION)`\"" -o $@ $(srcdir)/Modules/getbuildinfo.c
+	$(CC) -c $(PY_CORE_CFLAGS) -DSVNVERSION="\"`LC_ALL=C $(SVNVERSION)`\"" -o $@ $(srcdir)/Modules/getbuildinfo.c
 
 Modules/getpath.o: $(srcdir)/Modules/getpath.c Makefile
-	$(CC) -c $(PY_CFLAGS) -DPYTHONPATH='"$(PYTHONPATH)"' \
+	$(CC) -c $(PY_CORE_CFLAGS) -DPYTHONPATH='"$(PYTHONPATH)"' \
 		-DPREFIX='"$(prefix)"' \
 		-DEXEC_PREFIX='"$(exec_prefix)"' \
 		-DVERSION='"$(VERSION)"' \
@@ -552,7 +558,7 @@
 		-o $@ $(srcdir)/Modules/getpath.c
 
 Modules/python.o: $(srcdir)/Modules/python.c
-	$(MAINCC) -c $(PY_CFLAGS) -o $@ $(srcdir)/Modules/python.c
+	$(MAINCC) -c $(PY_CORE_CFLAGS) -o $@ $(srcdir)/Modules/python.c
 
 $(IO_OBJS): $(IO_H)
 
@@ -561,7 +567,7 @@
 		-$(PGEN) $(GRAMMAR_INPUT) $(GRAMMAR_H) $(GRAMMAR_C)
 
 $(PGEN):	$(PGENOBJS)
-		$(CC) $(OPT) $(LDFLAGS) $(PGENOBJS) $(LIBS) -o $(PGEN)
+		$(CC) $(OPT) $(PY_LDFLAGS) $(PGENOBJS) $(LIBS) -o $(PGEN)
 
 Parser/grammar.o:	$(srcdir)/Parser/grammar.c \
 				$(srcdir)/Include/token.h \
@@ -581,10 +587,10 @@
 Python/compile.o Python/symtable.o Python/ast.o: $(GRAMMAR_H) $(AST_H)
 
 Python/getplatform.o: $(srcdir)/Python/getplatform.c
-		$(CC) -c $(PY_CFLAGS) -DPLATFORM='"$(MACHDEP)"' -o $@ $(srcdir)/Python/getplatform.c
+		$(CC) -c $(PY_CORE_CFLAGS) -DPLATFORM='"$(MACHDEP)"' -o $@ $(srcdir)/Python/getplatform.c
 
 Python/importdl.o: $(srcdir)/Python/importdl.c
-		$(CC) -c $(PY_CFLAGS) -I$(DLINCLDIR) -o $@ $(srcdir)/Python/importdl.c
+		$(CC) -c $(PY_CORE_CFLAGS) -I$(DLINCLDIR) -o $@ $(srcdir)/Python/importdl.c
 
 Objects/unicodectype.o:	$(srcdir)/Objects/unicodectype.c \
 				$(srcdir)/Objects/unicodetype_db.h
@@ -1130,7 +1136,7 @@
 
 # Some make's put the object file in the current directory
 .c.o:
-	$(CC) -c $(PY_CFLAGS) -o $@ $<
+	$(CC) -c $(PY_CORE_CFLAGS) -o $@ $<
 
 # Run reindent on the library
 reindent:

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Fri Jul  9 18:30:58 2010
@@ -1530,6 +1530,12 @@
 Build
 -----
 
+- Issue #9189: Make a user-specified CFLAGS, CPPFLAGS, or LDFLAGS
+  setting override the configure and makefile defaults, without
+  deleting options the user didn't intend to override.  Developers
+  should no longer need to specify OPT or EXTRA_CFLAGS, although those
+  variables are still present for backward-compatibility.
+
 - Issue #8854: Fix finding Visual Studio 2008 on Windows x64.
 
 - Issue #1759169, #8864: Drop _XOPEN_SOURCE on Solaris, define it for

Modified: python/branches/py3k/Modules/makesetup
==============================================================================
--- python/branches/py3k/Modules/makesetup	(original)
+++ python/branches/py3k/Modules/makesetup	Fri Jul  9 18:30:58 2010
@@ -219,7 +219,7 @@
 			case $doconfig in
 			no)	cc="$cc \$(CCSHARED) \$(CFLAGS) \$(CPPFLAGS)";;
 			*)
-				cc="$cc \$(PY_CFLAGS)";;
+				cc="$cc \$(PY_CORE_CFLAGS)";;
 			esac
 			rule="$obj: $src; $cc $cpps -c $src -o $obj"
 			echo "$rule" >>$rulesf

Modified: python/branches/py3k/configure
==============================================================================
--- python/branches/py3k/configure	(original)
+++ python/branches/py3k/configure	Fri Jul  9 18:30:58 2010
@@ -1929,11 +1929,11 @@
        cat confdefs.h - <<_ACEOF >conftest.$ac_ext
 /* end confdefs.h.  */
 $ac_includes_default
+	     enum { N = $2 / 2 - 1 };
 int
 main ()
 {
-static int test_array [1 - 2 * !(enum { N = $2 / 2 - 1 };
-	     0 < ($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1))];
+static int test_array [1 - 2 * !(0 < ($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1))];
 test_array [0] = 0
 
   ;
@@ -1944,11 +1944,11 @@
   cat confdefs.h - <<_ACEOF >conftest.$ac_ext
 /* end confdefs.h.  */
 $ac_includes_default
+	        enum { N = $2 / 2 - 1 };
 int
 main ()
 {
-static int test_array [1 - 2 * !(enum { N = $2 / 2 - 1 };
-		($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1)
+static int test_array [1 - 2 * !(($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1)
 		 < ($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 2))];
 test_array [0] = 0
 
@@ -3158,9 +3158,12 @@
 (it is also a good idea to do 'make clean' before compiling)" "$LINENO" 5
 fi
 
-# If the user set CFLAGS, use this instead of the automatically
-# determined setting
-preset_cflags="$CFLAGS"
+# Don't let AC_PROG_CC set the default CFLAGS. It normally sets -g -O2
+# when the compiler supports them, but we don't always want -O2, and
+# we set -g later.
+if test -z "$CFLAGS"; then
+        CFLAGS=
+fi
 ac_ext=c
 ac_cpp='$CPP $CPPFLAGS'
 ac_compile='$CC -c $CFLAGS $CPPFLAGS conftest.$ac_ext >&5'
@@ -3952,10 +3955,6 @@
 ac_link='$CC -o conftest$ac_exeext $CFLAGS $CPPFLAGS $LDFLAGS conftest.$ac_ext $LIBS >&5'
 ac_compiler_gnu=$ac_cv_c_compiler_gnu
 
-if test ! -z "$preset_cflags"
-then
-	CFLAGS=$preset_cflags
-fi
 
 
 

Modified: python/branches/py3k/configure.in
==============================================================================
--- python/branches/py3k/configure.in	(original)
+++ python/branches/py3k/configure.in	Fri Jul  9 18:30:58 2010
@@ -461,14 +461,13 @@
 (it is also a good idea to do 'make clean' before compiling)])
 fi
 
-# If the user set CFLAGS, use this instead of the automatically
-# determined setting
-preset_cflags="$CFLAGS"
-AC_PROG_CC
-if test ! -z "$preset_cflags"
-then
-	CFLAGS=$preset_cflags
+# Don't let AC_PROG_CC set the default CFLAGS. It normally sets -g -O2
+# when the compiler supports them, but we don't always want -O2, and
+# we set -g later.
+if test -z "$CFLAGS"; then
+        CFLAGS=
 fi
+AC_PROG_CC
 
 AC_SUBST(CXX)
 AC_SUBST(MAINCC)

From python-checkins at python.org  Fri Jul  9 20:15:28 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Fri,  9 Jul 2010 20:15:28 +0200 (CEST)
Subject: [Python-checkins] r82747 -
	python/branches/release27-maint/Lib/test/test_struct.py
Message-ID: <20100709181528.B2F00EB74@mail.python.org>

Author: benjamin.peterson
Date: Fri Jul  9 20:15:28 2010
New Revision: 82747

Log:
this makes checking for warnings less error prone

Modified:
   python/branches/release27-maint/Lib/test/test_struct.py

Modified: python/branches/release27-maint/Lib/test/test_struct.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_struct.py	(original)
+++ python/branches/release27-maint/Lib/test/test_struct.py	Fri Jul  9 20:15:28 2010
@@ -317,9 +317,8 @@
                                   randrange)
                 with check_warnings(("integer argument expected, "
                                      "got non-integer", DeprecationWarning)):
-                    self.assertRaises((TypeError, struct.error),
-                                      struct.pack, self.format,
-                                      3+42j)
+                    with self.assertRaises((TypeError, struct.error)):
+                        struct.pack(self.format, 3+42j)
 
                 # an attempt to convert a non-integer (with an
                 # implicit conversion via __int__) should succeed,

From python-checkins at python.org  Fri Jul  9 20:28:25 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Fri,  9 Jul 2010 20:28:25 +0200 (CEST)
Subject: [Python-checkins] r82748 - python/branches/py3k/Lib/socket.py
Message-ID: <20100709182825.34F83F2F6@mail.python.org>

Author: benjamin.peterson
Date: Fri Jul  9 20:28:25 2010
New Revision: 82748

Log:
remove SocketIO.__del__; close() can handle it

Modified:
   python/branches/py3k/Lib/socket.py

Modified: python/branches/py3k/Lib/socket.py
==============================================================================
--- python/branches/py3k/Lib/socket.py	(original)
+++ python/branches/py3k/Lib/socket.py	Fri Jul  9 20:28:25 2010
@@ -252,10 +252,6 @@
         self._sock._decref_socketios()
         self._sock = None
 
-    def __del__(self):
-        if not self.closed:
-            self._sock._decref_socketios()
-
 
 def getfqdn(name=''):
     """Get fully qualified domain name from name.

From python-checkins at python.org  Fri Jul  9 21:25:48 2010
From: python-checkins at python.org (mark.dickinson)
Date: Fri,  9 Jul 2010 21:25:48 +0200 (CEST)
Subject: [Python-checkins] r82749 -
	python/branches/py3k/Objects/bytearrayobject.c
Message-ID: <20100709192548.D66AFEE989@mail.python.org>

Author: mark.dickinson
Date: Fri Jul  9 21:25:48 2010
New Revision: 82749

Log:
Silence gcc warning. (In function 'bytearray_init': warning: 'value' may be used uninitialized in this function).

Modified:
   python/branches/py3k/Objects/bytearrayobject.c

Modified: python/branches/py3k/Objects/bytearrayobject.c
==============================================================================
--- python/branches/py3k/Objects/bytearrayobject.c	(original)
+++ python/branches/py3k/Objects/bytearrayobject.c	Fri Jul  9 21:25:48 2010
@@ -33,6 +33,7 @@
         PyObject *index = PyNumber_Index(arg);
         if (index == NULL) {
             PyErr_Format(PyExc_TypeError, "an integer is required");
+            *value = -1;
             return 0;
         }
         face_value = PyLong_AsLong(index);
@@ -42,6 +43,7 @@
     if (face_value < 0 || face_value >= 256) {
         /* this includes the OverflowError in case the long is too large */
         PyErr_SetString(PyExc_ValueError, "byte must be in range(0, 256)");
+        *value = -1;
         return 0;
     }
 

From python-checkins at python.org  Fri Jul  9 21:37:00 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Fri,  9 Jul 2010 21:37:00 +0200 (CEST)
Subject: [Python-checkins] r82750 -
	python/branches/py3k/Lib/test/test_structseq.py
Message-ID: <20100709193700.AD80FEE9A6@mail.python.org>

Author: benjamin.peterson
Date: Fri Jul  9 21:37:00 2010
New Revision: 82750

Log:
be more robust across platforms

Modified:
   python/branches/py3k/Lib/test/test_structseq.py

Modified: python/branches/py3k/Lib/test/test_structseq.py
==============================================================================
--- python/branches/py3k/Lib/test/test_structseq.py	(original)
+++ python/branches/py3k/Lib/test/test_structseq.py	Fri Jul  9 21:37:00 2010
@@ -38,7 +38,7 @@
         # os.stat() gives a complicated struct sequence.
         st = os.stat(__file__)
         rep = repr(st)
-        self.assertTrue(rep.startswith("posix.stat_result"))
+        self.assertTrue(rep.startswith(os.name + ".stat_result"))
         self.assertIn("st_mode=", rep)
         self.assertIn("st_ino=", rep)
         self.assertIn("st_dev=", rep)

From python-checkins at python.org  Fri Jul  9 21:39:56 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Fri,  9 Jul 2010 21:39:56 +0200 (CEST)
Subject: [Python-checkins] r82751 -
	python/branches/release31-maint/Lib/test/test_xmlrpc_net.py
Message-ID: <20100709193956.69CD2EE984@mail.python.org>

Author: benjamin.peterson
Date: Fri Jul  9 21:39:56 2010
New Revision: 82751

Log:
3.x is the new trunk

Modified:
   python/branches/release31-maint/Lib/test/test_xmlrpc_net.py

Modified: python/branches/release31-maint/Lib/test/test_xmlrpc_net.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_xmlrpc_net.py	(original)
+++ python/branches/release31-maint/Lib/test/test_xmlrpc_net.py	Fri Jul  9 21:39:56 2010
@@ -48,7 +48,7 @@
         # Perform a minimal sanity check on the result, just to be sure
         # the request means what we think it means.
         self.assertTrue(isinstance(builders, collections.Sequence), type(builders))
-        self.assertTrue([x for x in builders if "trunk" in x], builders)
+        self.assertTrue([x for x in builders if "3.x" in x], builders)
 
 
 def test_main():

From python-checkins at python.org  Fri Jul  9 21:43:26 2010
From: python-checkins at python.org (mark.dickinson)
Date: Fri,  9 Jul 2010 21:43:26 +0200 (CEST)
Subject: [Python-checkins] r82752 -
	python/branches/release31-maint/Python/pystrtod.c
Message-ID: <20100709194326.5F511C997@mail.python.org>

Author: mark.dickinson
Date: Fri Jul  9 21:43:26 2010
New Revision: 82752

Log:
Temporary checkin to try to diagnose i386 Ubuntu buildbot failure.

Modified:
   python/branches/release31-maint/Python/pystrtod.c

Modified: python/branches/release31-maint/Python/pystrtod.c
==============================================================================
--- python/branches/release31-maint/Python/pystrtod.c	(original)
+++ python/branches/release31-maint/Python/pystrtod.c	Fri Jul  9 21:43:26 2010
@@ -1003,7 +1003,7 @@
                something starting with a digit, an 'I',  or 'N' */
             strncpy(p, "ERR", 3);
             p += 3;
-            assert(0);
+            /* assert(0); */
         }
         goto exit;
     }

From benjamin at python.org  Fri Jul  9 21:43:48 2010
From: benjamin at python.org (Benjamin Peterson)
Date: Fri, 9 Jul 2010 14:43:48 -0500
Subject: [Python-checkins] r82746 - in python/branches/py3k:
	Lib/sysconfig.py Makefile.pre.in Misc/NEWS Modules/makesetup
	configure configure.in
In-Reply-To: <20100709163058.9A02BEED94@mail.python.org>
References: <20100709163058.9A02BEED94@mail.python.org>
Message-ID: <AANLkTininO50I5fHw0K7JZXTzvRTcYwVD7w3h5pIihzs@mail.gmail.com>

This broke Windows:
http://python.org/dev/buildbot/stable/builders/amd64%20gentoo%203.x/builds/1352/steps/test/logs/stdio

2010/7/9 jeffrey.yasskin <python-checkins at python.org>:
> Author: jeffrey.yasskin
> Date: Fri Jul ?9 18:30:58 2010
> New Revision: 82746
>
> Log:
> Issue #9189: Allow users to set $CFLAGS, $CPPFLAGS, and $LDFLAGS when running
> configure to append to Python's default values for those variables, and
> similarly allow users to set $XXFLAGS on the make command line to append to the
> values set by configure.
>
> In the makefile, this renames the variables that used to be $XXFLAGS to
> $PY_XXFLAGS, and renames the old $PY_CFLAGS to $PY_CORE_CFLAGS. ?To compensate,
> sysconfig now aliases $XXFLAGS=$PY_XXFLAGS so that scripts using it keep
> working. ?I see that as the right interface, not a backward-compatibility hack,
> since these are logically the $XXFLAGS variables; we just use a different name
> in the makefile to deal with make's semantics.
>
>
> Modified:
> ? python/branches/py3k/Lib/sysconfig.py
> ? python/branches/py3k/Makefile.pre.in
> ? python/branches/py3k/Misc/NEWS
> ? python/branches/py3k/Modules/makesetup
> ? python/branches/py3k/configure
> ? python/branches/py3k/configure.in
>
> Modified: python/branches/py3k/Lib/sysconfig.py
> ==============================================================================
> --- python/branches/py3k/Lib/sysconfig.py ? ? ? (original)
> +++ python/branches/py3k/Lib/sysconfig.py ? ? ? Fri Jul ?9 18:30:58 2010
> @@ -259,6 +259,11 @@
> ? ? ? ? ? ? ? ? # bogus variable reference; just drop it since we can't deal
> ? ? ? ? ? ? ? ? variables.remove(name)
>
> + ? ?# Add in CFLAGS, LDFLAGS, and CPPFLAGS, which are named with a
> + ? ?# prefix in the Makefile.
> + ? ?for var in ('CFLAGS', 'LDFLAGS', 'CPPFLAGS'):
> + ? ? ? ?done[var] = done['PY_' + var]
> +
> ? ? # save the results in the global dictionary
> ? ? vars.update(done)
> ? ? return vars
>
> Modified: python/branches/py3k/Makefile.pre.in
> ==============================================================================
> --- python/branches/py3k/Makefile.pre.in ? ? ? ?(original)
> +++ python/branches/py3k/Makefile.pre.in ? ? ? ?Fri Jul ?9 18:30:58 2010
> @@ -59,12 +59,18 @@
> ?# Compiler options
> ?OPT= ? ? ? ? ? @OPT@
> ?BASECFLAGS= ? ?@BASECFLAGS@
> -CFLAGS= ? ? ? ? ? ? ? ?$(BASECFLAGS) @CFLAGS@ $(OPT) $(EXTRA_CFLAGS)
> +CONFIGURE_CFLAGS= ? ? ?@CFLAGS@
> +CONFIGURE_CPPFLAGS= ? ?@CPPFLAGS@
> +CONFIGURE_LDFLAGS= ? ? @LDFLAGS@
> +# Avoid assigning CFLAGS, LDFLAGS, etc. so users can use them on the
> +# command line to append to these values without stomping the pre-set
> +# values.
> +PY_CFLAGS= ? ? $(BASECFLAGS) $(OPT) $(CONFIGURE_CFLAGS) $(CFLAGS) $(EXTRA_CFLAGS)
> ?# Both CPPFLAGS and LDFLAGS need to contain the shell's value for setup.py to
> ?# be able to build extension modules using the directories specified in the
> ?# environment variables
> -CPPFLAGS= ? ? ?-I. -IInclude -I$(srcdir)/Include @CPPFLAGS@
> -LDFLAGS= ? ? ? @LDFLAGS@
> +PY_CPPFLAGS= ? -I. -IInclude -I$(srcdir)/Include $(CONFIGURE_CPPFLAGS) $(CPPFLAGS)
> +PY_LDFLAGS= ? ?$(CONFIGURE_LDFLAGS) $(LDFLAGS)
> ?LDLAST= ? ? ? ? ? ? ? ?@LDLAST@
> ?SGI_ABI= ? ? ? @SGI_ABI@
> ?CCSHARED= ? ? ?@CCSHARED@
> @@ -73,7 +79,7 @@
> ?# Extra C flags added for building the interpreter object files.
> ?CFLAGSFORSHARED=@CFLAGSFORSHARED@
> ?# C flags used for building the interpreter object files
> -PY_CFLAGS= ? ? $(CFLAGS) $(CPPFLAGS) $(CFLAGSFORSHARED) -DPy_BUILD_CORE
> +PY_CORE_CFLAGS= ? ? ? ?$(PY_CFLAGS) $(PY_CPPFLAGS) $(CFLAGSFORSHARED) -DPy_BUILD_CORE
>
>
> ?# Machine-dependent subdirectories
> @@ -411,7 +417,7 @@
>
> ?# Build the interpreter
> ?$(BUILDPYTHON): ? ? ? ?Modules/python.o $(LIBRARY) $(LDLIBRARY)
> - ? ? ? ? ? ? ? $(LINKCC) $(LDFLAGS) $(LINKFORSHARED) -o $@ \
> + ? ? ? ? ? ? ? $(LINKCC) $(PY_LDFLAGS) $(LINKFORSHARED) -o $@ \
> ? ? ? ? ? ? ? ? ? ? ? ?Modules/python.o \
> ? ? ? ? ? ? ? ? ? ? ? ?$(BLDLIBRARY) $(LIBS) $(MODLIBS) $(SYSLIBS) $(LDLAST)
>
> @@ -422,8 +428,8 @@
> ?# Build the shared modules
> ?sharedmods: $(BUILDPYTHON)
> ? ? ? ?@case $$MAKEFLAGS in \
> - ? ? ? *s*) $(RUNSHARED) CC='$(CC)' LDSHARED='$(BLDSHARED)' LDFLAGS='$(LDFLAGS)' OPT='$(OPT)' ./$(BUILDPYTHON) -E $(srcdir)/setup.py -q build;; \
> - ? ? ? *) $(RUNSHARED) CC='$(CC)' LDSHARED='$(BLDSHARED)' LDFLAGS='$(LDFLAGS)' OPT='$(OPT)' ./$(BUILDPYTHON) -E $(srcdir)/setup.py build;; \
> + ? ? ? *s*) $(RUNSHARED) CC='$(CC)' LDSHARED='$(BLDSHARED)' LDFLAGS='$(PY_LDFLAGS)' OPT='$(OPT)' ./$(BUILDPYTHON) -E $(srcdir)/setup.py -q build;; \
> + ? ? ? *) $(RUNSHARED) CC='$(CC)' LDSHARED='$(BLDSHARED)' LDFLAGS='$(PY_LDFLAGS)' OPT='$(OPT)' ./$(BUILDPYTHON) -E $(srcdir)/setup.py build;; \
> ? ? ? ?esac
>
> ?# Build static library
> @@ -440,18 +446,18 @@
>
> ?libpython$(VERSION).so: $(LIBRARY_OBJS)
> ? ? ? ?if test $(INSTSONAME) != $(LDLIBRARY); then \
> - ? ? ? ? ? ? ? $(LDSHARED) $(LDFLAGS) -Wl,-h$(INSTSONAME) -o $(INSTSONAME) $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
> + ? ? ? ? ? ? ? $(LDSHARED) $(PY_LDFLAGS) -Wl,-h$(INSTSONAME) -o $(INSTSONAME) $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
> ? ? ? ? ? ? ? ?$(LN) -f $(INSTSONAME) $@; \
> ? ? ? ?else \
> - ? ? ? ? ? ? ? $(LDSHARED) $(LDFLAGS) -o $@ $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
> + ? ? ? ? ? ? ? $(LDSHARED) $(PY_LDFLAGS) -o $@ $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
> ? ? ? ?fi
>
> ?libpython$(VERSION).dylib: $(LIBRARY_OBJS)
> - ? ? ? ?$(CC) -dynamiclib -Wl,-single_module $(LDFLAGS) -undefined dynamic_lookup -Wl,-install_name,$(prefix)/lib/libpython$(VERSION).dylib -Wl,-compatibility_version,$(VERSION) -Wl,-current_version,$(VERSION) -o $@ $(LIBRARY_OBJS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
> + ? ? ? ?$(CC) -dynamiclib -Wl,-single_module $(PY_LDFLAGS) -undefined dynamic_lookup -Wl,-install_name,$(prefix)/lib/libpython$(VERSION).dylib -Wl,-compatibility_version,$(VERSION) -Wl,-current_version,$(VERSION) -o $@ $(LIBRARY_OBJS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
>
>
> ?libpython$(VERSION).sl: $(LIBRARY_OBJS)
> - ? ? ? $(LDSHARED) $(LDFLAGS) -o $@ $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST)
> + ? ? ? $(LDSHARED) $(PY_LDFLAGS) -o $@ $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST)
>
> ?# Copy up the gdb python hooks into a position where they can be automatically
> ?# loaded by gdb during Lib/test/test_gdb.py
> @@ -497,7 +503,7 @@
> ?# for a shared core library; otherwise, this rule is a noop.
> ?$(DLLLIBRARY) libpython$(VERSION).dll.a: $(LIBRARY_OBJS)
> ? ? ? ?if test -n "$(DLLLIBRARY)"; then \
> - ? ? ? ? ? ? ? $(LDSHARED) $(LDFLAGS) -Wl,--out-implib=$@ -o $(DLLLIBRARY) $^ \
> + ? ? ? ? ? ? ? $(LDSHARED) $(PY_LDFLAGS) -Wl,--out-implib=$@ -o $(DLLLIBRARY) $^ \
> ? ? ? ? ? ? ? ? ? ? ? ?$(LIBS) $(MODLIBS) $(SYSLIBS) $(LDLAST); \
> ? ? ? ?else true; \
> ? ? ? ?fi
> @@ -541,10 +547,10 @@
> ? ? ? ? ? ? ? ?$(SIGNAL_OBJS) \
> ? ? ? ? ? ? ? ?$(MODOBJS) \
> ? ? ? ? ? ? ? ?$(srcdir)/Modules/getbuildinfo.c
> - ? ? ? $(CC) -c $(PY_CFLAGS) -DSVNVERSION="\"`LC_ALL=C $(SVNVERSION)`\"" -o $@ $(srcdir)/Modules/getbuildinfo.c
> + ? ? ? $(CC) -c $(PY_CORE_CFLAGS) -DSVNVERSION="\"`LC_ALL=C $(SVNVERSION)`\"" -o $@ $(srcdir)/Modules/getbuildinfo.c
>
> ?Modules/getpath.o: $(srcdir)/Modules/getpath.c Makefile
> - ? ? ? $(CC) -c $(PY_CFLAGS) -DPYTHONPATH='"$(PYTHONPATH)"' \
> + ? ? ? $(CC) -c $(PY_CORE_CFLAGS) -DPYTHONPATH='"$(PYTHONPATH)"' \
> ? ? ? ? ? ? ? ?-DPREFIX='"$(prefix)"' \
> ? ? ? ? ? ? ? ?-DEXEC_PREFIX='"$(exec_prefix)"' \
> ? ? ? ? ? ? ? ?-DVERSION='"$(VERSION)"' \
> @@ -552,7 +558,7 @@
> ? ? ? ? ? ? ? ?-o $@ $(srcdir)/Modules/getpath.c
>
> ?Modules/python.o: $(srcdir)/Modules/python.c
> - ? ? ? $(MAINCC) -c $(PY_CFLAGS) -o $@ $(srcdir)/Modules/python.c
> + ? ? ? $(MAINCC) -c $(PY_CORE_CFLAGS) -o $@ $(srcdir)/Modules/python.c
>
> ?$(IO_OBJS): $(IO_H)
>
> @@ -561,7 +567,7 @@
> ? ? ? ? ? ? ? ?-$(PGEN) $(GRAMMAR_INPUT) $(GRAMMAR_H) $(GRAMMAR_C)
>
> ?$(PGEN): ? ? ? $(PGENOBJS)
> - ? ? ? ? ? ? ? $(CC) $(OPT) $(LDFLAGS) $(PGENOBJS) $(LIBS) -o $(PGEN)
> + ? ? ? ? ? ? ? $(CC) $(OPT) $(PY_LDFLAGS) $(PGENOBJS) $(LIBS) -o $(PGEN)
>
> ?Parser/grammar.o: ? ? ?$(srcdir)/Parser/grammar.c \
> ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?$(srcdir)/Include/token.h \
> @@ -581,10 +587,10 @@
> ?Python/compile.o Python/symtable.o Python/ast.o: $(GRAMMAR_H) $(AST_H)
>
> ?Python/getplatform.o: $(srcdir)/Python/getplatform.c
> - ? ? ? ? ? ? ? $(CC) -c $(PY_CFLAGS) -DPLATFORM='"$(MACHDEP)"' -o $@ $(srcdir)/Python/getplatform.c
> + ? ? ? ? ? ? ? $(CC) -c $(PY_CORE_CFLAGS) -DPLATFORM='"$(MACHDEP)"' -o $@ $(srcdir)/Python/getplatform.c
>
> ?Python/importdl.o: $(srcdir)/Python/importdl.c
> - ? ? ? ? ? ? ? $(CC) -c $(PY_CFLAGS) -I$(DLINCLDIR) -o $@ $(srcdir)/Python/importdl.c
> + ? ? ? ? ? ? ? $(CC) -c $(PY_CORE_CFLAGS) -I$(DLINCLDIR) -o $@ $(srcdir)/Python/importdl.c
>
> ?Objects/unicodectype.o: ? ? ? ?$(srcdir)/Objects/unicodectype.c \
> ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?$(srcdir)/Objects/unicodetype_db.h
> @@ -1130,7 +1136,7 @@
>
> ?# Some make's put the object file in the current directory
> ?.c.o:
> - ? ? ? $(CC) -c $(PY_CFLAGS) -o $@ $<
> + ? ? ? $(CC) -c $(PY_CORE_CFLAGS) -o $@ $<
>
> ?# Run reindent on the library
> ?reindent:
>
> Modified: python/branches/py3k/Misc/NEWS
> ==============================================================================
> --- python/branches/py3k/Misc/NEWS ? ? ?(original)
> +++ python/branches/py3k/Misc/NEWS ? ? ?Fri Jul ?9 18:30:58 2010
> @@ -1530,6 +1530,12 @@
> ?Build
> ?-----
>
> +- Issue #9189: Make a user-specified CFLAGS, CPPFLAGS, or LDFLAGS
> + ?setting override the configure and makefile defaults, without
> + ?deleting options the user didn't intend to override. ?Developers
> + ?should no longer need to specify OPT or EXTRA_CFLAGS, although those
> + ?variables are still present for backward-compatibility.
> +
> ?- Issue #8854: Fix finding Visual Studio 2008 on Windows x64.
>
> ?- Issue #1759169, #8864: Drop _XOPEN_SOURCE on Solaris, define it for
>
> Modified: python/branches/py3k/Modules/makesetup
> ==============================================================================
> --- python/branches/py3k/Modules/makesetup ? ? ?(original)
> +++ python/branches/py3k/Modules/makesetup ? ? ?Fri Jul ?9 18:30:58 2010
> @@ -219,7 +219,7 @@
> ? ? ? ? ? ? ? ? ? ? ? ?case $doconfig in
> ? ? ? ? ? ? ? ? ? ? ? ?no) ? ? cc="$cc \$(CCSHARED) \$(CFLAGS) \$(CPPFLAGS)";;
> ? ? ? ? ? ? ? ? ? ? ? ?*)
> - ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? cc="$cc \$(PY_CFLAGS)";;
> + ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? cc="$cc \$(PY_CORE_CFLAGS)";;
> ? ? ? ? ? ? ? ? ? ? ? ?esac
> ? ? ? ? ? ? ? ? ? ? ? ?rule="$obj: $src; $cc $cpps -c $src -o $obj"
> ? ? ? ? ? ? ? ? ? ? ? ?echo "$rule" >>$rulesf
>
> Modified: python/branches/py3k/configure
> ==============================================================================
> --- python/branches/py3k/configure ? ? ?(original)
> +++ python/branches/py3k/configure ? ? ?Fri Jul ?9 18:30:58 2010
> @@ -1929,11 +1929,11 @@
> ? ? ? ?cat confdefs.h - <<_ACEOF >conftest.$ac_ext
> ?/* end confdefs.h. ?*/
> ?$ac_includes_default
> + ? ? ? ? ? ?enum { N = $2 / 2 - 1 };
> ?int
> ?main ()
> ?{
> -static int test_array [1 - 2 * !(enum { N = $2 / 2 - 1 };
> - ? ? ? ? ? ?0 < ($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1))];
> +static int test_array [1 - 2 * !(0 < ($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1))];
> ?test_array [0] = 0
>
> ? ;
> @@ -1944,11 +1944,11 @@
> ? cat confdefs.h - <<_ACEOF >conftest.$ac_ext
> ?/* end confdefs.h. ?*/
> ?$ac_includes_default
> + ? ? ? ? ? ? ? enum { N = $2 / 2 - 1 };
> ?int
> ?main ()
> ?{
> -static int test_array [1 - 2 * !(enum { N = $2 / 2 - 1 };
> - ? ? ? ? ? ? ? ($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1)
> +static int test_array [1 - 2 * !(($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1)
> ? ? ? ? ? ? ? ? < ($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 2))];
> ?test_array [0] = 0
>
> @@ -3158,9 +3158,12 @@
> ?(it is also a good idea to do 'make clean' before compiling)" "$LINENO" 5
> ?fi
>
> -# If the user set CFLAGS, use this instead of the automatically
> -# determined setting
> -preset_cflags="$CFLAGS"
> +# Don't let AC_PROG_CC set the default CFLAGS. It normally sets -g -O2
> +# when the compiler supports them, but we don't always want -O2, and
> +# we set -g later.
> +if test -z "$CFLAGS"; then
> + ? ? ? ?CFLAGS=
> +fi
> ?ac_ext=c
> ?ac_cpp='$CPP $CPPFLAGS'
> ?ac_compile='$CC -c $CFLAGS $CPPFLAGS conftest.$ac_ext >&5'
> @@ -3952,10 +3955,6 @@
> ?ac_link='$CC -o conftest$ac_exeext $CFLAGS $CPPFLAGS $LDFLAGS conftest.$ac_ext $LIBS >&5'
> ?ac_compiler_gnu=$ac_cv_c_compiler_gnu
>
> -if test ! -z "$preset_cflags"
> -then
> - ? ? ? CFLAGS=$preset_cflags
> -fi
>
>
>
>
> Modified: python/branches/py3k/configure.in
> ==============================================================================
> --- python/branches/py3k/configure.in ? (original)
> +++ python/branches/py3k/configure.in ? Fri Jul ?9 18:30:58 2010
> @@ -461,14 +461,13 @@
> ?(it is also a good idea to do 'make clean' before compiling)])
> ?fi
>
> -# If the user set CFLAGS, use this instead of the automatically
> -# determined setting
> -preset_cflags="$CFLAGS"
> -AC_PROG_CC
> -if test ! -z "$preset_cflags"
> -then
> - ? ? ? CFLAGS=$preset_cflags
> +# Don't let AC_PROG_CC set the default CFLAGS. It normally sets -g -O2
> +# when the compiler supports them, but we don't always want -O2, and
> +# we set -g later.
> +if test -z "$CFLAGS"; then
> + ? ? ? ?CFLAGS=
> ?fi
> +AC_PROG_CC
>
> ?AC_SUBST(CXX)
> ?AC_SUBST(MAINCC)
> _______________________________________________
> Python-checkins mailing list
> Python-checkins at python.org
> http://mail.python.org/mailman/listinfo/python-checkins
>



-- 
Regards,
Benjamin

From python-checkins at python.org  Fri Jul  9 21:55:05 2010
From: python-checkins at python.org (jeffrey.yasskin)
Date: Fri,  9 Jul 2010 21:55:05 +0200 (CEST)
Subject: [Python-checkins] r82753 - python/branches/py3k/Lib/sysconfig.py
Message-ID: <20100709195505.831B2F463@mail.python.org>

Author: jeffrey.yasskin
Date: Fri Jul  9 21:55:05 2010
New Revision: 82753

Log:
Oops. Fix distutils tests that r82746 broke.


Modified:
   python/branches/py3k/Lib/sysconfig.py

Modified: python/branches/py3k/Lib/sysconfig.py
==============================================================================
--- python/branches/py3k/Lib/sysconfig.py	(original)
+++ python/branches/py3k/Lib/sysconfig.py	Fri Jul  9 21:55:05 2010
@@ -262,7 +262,9 @@
     # Add in CFLAGS, LDFLAGS, and CPPFLAGS, which are named with a
     # prefix in the Makefile.
     for var in ('CFLAGS', 'LDFLAGS', 'CPPFLAGS'):
-        done[var] = done['PY_' + var]
+        makefile_value = done.get('PY_' + var)
+        if makefile_value is not None:
+            done[var] = makefile_value
 
     # save the results in the global dictionary
     vars.update(done)

From python-checkins at python.org  Fri Jul  9 22:10:37 2010
From: python-checkins at python.org (mark.dickinson)
Date: Fri,  9 Jul 2010 22:10:37 +0200 (CEST)
Subject: [Python-checkins] r82754 -
	python/branches/release31-maint/Python/pystrtod.c
Message-ID: <20100709201037.A7D71EE9F4@mail.python.org>

Author: mark.dickinson
Date: Fri Jul  9 22:10:37 2010
New Revision: 82754

Log:
Another temporary checkin to diagnose test_float failure on 3.1 Ubuntu buildbot.

Modified:
   python/branches/release31-maint/Python/pystrtod.c

Modified: python/branches/release31-maint/Python/pystrtod.c
==============================================================================
--- python/branches/release31-maint/Python/pystrtod.c	(original)
+++ python/branches/release31-maint/Python/pystrtod.c	Fri Jul  9 22:10:37 2010
@@ -1001,9 +1001,17 @@
         else {
             /* shouldn't get here: Gay's code should always return
                something starting with a digit, an 'I',  or 'N' */
-            strncpy(p, "ERR", 3);
-            p += 3;
-            /* assert(0); */
+            printf("Unexpected failure in format_float_short. "
+                   "Arguments: d = %.17g, format_code = %d, "
+                   "mode = %d, precision = %ld\n",
+                   d, format_code, mode, precision);
+            printf("digits == %.100s\n", digits);
+            PyErr_Format(PyExc_RuntimeError,
+                         "Unexpected failure in format_float_short. "
+                         "Arguments: d = %.17g, format_code = %d, "
+                         "mode = %d, precision = %ld\n",
+                         d, format_code, mode, precision);
+            return NULL;
         }
         goto exit;
     }

From jyasskin at gmail.com  Fri Jul  9 23:10:26 2010
From: jyasskin at gmail.com (Jeffrey Yasskin)
Date: Fri, 9 Jul 2010 14:10:26 -0700
Subject: [Python-checkins] r82746 - in python/branches/py3k:
	Lib/sysconfig.py Makefile.pre.in Misc/NEWS Modules/makesetup
	configure configure.in
In-Reply-To: <AANLkTininO50I5fHw0K7JZXTzvRTcYwVD7w3h5pIihzs@mail.gmail.com>
References: <20100709163058.9A02BEED94@mail.python.org>
	<AANLkTininO50I5fHw0K7JZXTzvRTcYwVD7w3h5pIihzs@mail.gmail.com>
Message-ID: <AANLkTinmEsb-6xzpyMeHwfW0dwjW8s8aKD889GBgIQBE@mail.gmail.com>

Fixed by 82753:
http://python.org/dev/buildbot/stable/builders/amd64%20gentoo%203.x/builds/1354/steps/test/logs/stdio
Sorry about that.

On Fri, Jul 9, 2010 at 12:43 PM, Benjamin Peterson <benjamin at python.org> wrote:
> This broke Windows:
> http://python.org/dev/buildbot/stable/builders/amd64%20gentoo%203.x/builds/1352/steps/test/logs/stdio
>
> 2010/7/9 jeffrey.yasskin <python-checkins at python.org>:
>> Author: jeffrey.yasskin
>> Date: Fri Jul ?9 18:30:58 2010
>> New Revision: 82746
>>
>> Log:
>> Issue #9189: Allow users to set $CFLAGS, $CPPFLAGS, and $LDFLAGS when running
>> configure to append to Python's default values for those variables, and
>> similarly allow users to set $XXFLAGS on the make command line to append to the
>> values set by configure.
>>
>> In the makefile, this renames the variables that used to be $XXFLAGS to
>> $PY_XXFLAGS, and renames the old $PY_CFLAGS to $PY_CORE_CFLAGS. ?To compensate,
>> sysconfig now aliases $XXFLAGS=$PY_XXFLAGS so that scripts using it keep
>> working. ?I see that as the right interface, not a backward-compatibility hack,
>> since these are logically the $XXFLAGS variables; we just use a different name
>> in the makefile to deal with make's semantics.
>>
>>
>> Modified:
>> ? python/branches/py3k/Lib/sysconfig.py
>> ? python/branches/py3k/Makefile.pre.in
>> ? python/branches/py3k/Misc/NEWS
>> ? python/branches/py3k/Modules/makesetup
>> ? python/branches/py3k/configure
>> ? python/branches/py3k/configure.in
>>
>> Modified: python/branches/py3k/Lib/sysconfig.py
>> ==============================================================================
>> --- python/branches/py3k/Lib/sysconfig.py ? ? ? (original)
>> +++ python/branches/py3k/Lib/sysconfig.py ? ? ? Fri Jul ?9 18:30:58 2010
>> @@ -259,6 +259,11 @@
>> ? ? ? ? ? ? ? ? # bogus variable reference; just drop it since we can't deal
>> ? ? ? ? ? ? ? ? variables.remove(name)
>>
>> + ? ?# Add in CFLAGS, LDFLAGS, and CPPFLAGS, which are named with a
>> + ? ?# prefix in the Makefile.
>> + ? ?for var in ('CFLAGS', 'LDFLAGS', 'CPPFLAGS'):
>> + ? ? ? ?done[var] = done['PY_' + var]
>> +
>> ? ? # save the results in the global dictionary
>> ? ? vars.update(done)
>> ? ? return vars
>>
>> Modified: python/branches/py3k/Makefile.pre.in
>> ==============================================================================
>> --- python/branches/py3k/Makefile.pre.in ? ? ? ?(original)
>> +++ python/branches/py3k/Makefile.pre.in ? ? ? ?Fri Jul ?9 18:30:58 2010
>> @@ -59,12 +59,18 @@
>> ?# Compiler options
>> ?OPT= ? ? ? ? ? @OPT@
>> ?BASECFLAGS= ? ?@BASECFLAGS@
>> -CFLAGS= ? ? ? ? ? ? ? ?$(BASECFLAGS) @CFLAGS@ $(OPT) $(EXTRA_CFLAGS)
>> +CONFIGURE_CFLAGS= ? ? ?@CFLAGS@
>> +CONFIGURE_CPPFLAGS= ? ?@CPPFLAGS@
>> +CONFIGURE_LDFLAGS= ? ? @LDFLAGS@
>> +# Avoid assigning CFLAGS, LDFLAGS, etc. so users can use them on the
>> +# command line to append to these values without stomping the pre-set
>> +# values.
>> +PY_CFLAGS= ? ? $(BASECFLAGS) $(OPT) $(CONFIGURE_CFLAGS) $(CFLAGS) $(EXTRA_CFLAGS)
>> ?# Both CPPFLAGS and LDFLAGS need to contain the shell's value for setup.py to
>> ?# be able to build extension modules using the directories specified in the
>> ?# environment variables
>> -CPPFLAGS= ? ? ?-I. -IInclude -I$(srcdir)/Include @CPPFLAGS@
>> -LDFLAGS= ? ? ? @LDFLAGS@
>> +PY_CPPFLAGS= ? -I. -IInclude -I$(srcdir)/Include $(CONFIGURE_CPPFLAGS) $(CPPFLAGS)
>> +PY_LDFLAGS= ? ?$(CONFIGURE_LDFLAGS) $(LDFLAGS)
>> ?LDLAST= ? ? ? ? ? ? ? ?@LDLAST@
>> ?SGI_ABI= ? ? ? @SGI_ABI@
>> ?CCSHARED= ? ? ?@CCSHARED@
>> @@ -73,7 +79,7 @@
>> ?# Extra C flags added for building the interpreter object files.
>> ?CFLAGSFORSHARED=@CFLAGSFORSHARED@
>> ?# C flags used for building the interpreter object files
>> -PY_CFLAGS= ? ? $(CFLAGS) $(CPPFLAGS) $(CFLAGSFORSHARED) -DPy_BUILD_CORE
>> +PY_CORE_CFLAGS= ? ? ? ?$(PY_CFLAGS) $(PY_CPPFLAGS) $(CFLAGSFORSHARED) -DPy_BUILD_CORE
>>
>>
>> ?# Machine-dependent subdirectories
>> @@ -411,7 +417,7 @@
>>
>> ?# Build the interpreter
>> ?$(BUILDPYTHON): ? ? ? ?Modules/python.o $(LIBRARY) $(LDLIBRARY)
>> - ? ? ? ? ? ? ? $(LINKCC) $(LDFLAGS) $(LINKFORSHARED) -o $@ \
>> + ? ? ? ? ? ? ? $(LINKCC) $(PY_LDFLAGS) $(LINKFORSHARED) -o $@ \
>> ? ? ? ? ? ? ? ? ? ? ? ?Modules/python.o \
>> ? ? ? ? ? ? ? ? ? ? ? ?$(BLDLIBRARY) $(LIBS) $(MODLIBS) $(SYSLIBS) $(LDLAST)
>>
>> @@ -422,8 +428,8 @@
>> ?# Build the shared modules
>> ?sharedmods: $(BUILDPYTHON)
>> ? ? ? ?@case $$MAKEFLAGS in \
>> - ? ? ? *s*) $(RUNSHARED) CC='$(CC)' LDSHARED='$(BLDSHARED)' LDFLAGS='$(LDFLAGS)' OPT='$(OPT)' ./$(BUILDPYTHON) -E $(srcdir)/setup.py -q build;; \
>> - ? ? ? *) $(RUNSHARED) CC='$(CC)' LDSHARED='$(BLDSHARED)' LDFLAGS='$(LDFLAGS)' OPT='$(OPT)' ./$(BUILDPYTHON) -E $(srcdir)/setup.py build;; \
>> + ? ? ? *s*) $(RUNSHARED) CC='$(CC)' LDSHARED='$(BLDSHARED)' LDFLAGS='$(PY_LDFLAGS)' OPT='$(OPT)' ./$(BUILDPYTHON) -E $(srcdir)/setup.py -q build;; \
>> + ? ? ? *) $(RUNSHARED) CC='$(CC)' LDSHARED='$(BLDSHARED)' LDFLAGS='$(PY_LDFLAGS)' OPT='$(OPT)' ./$(BUILDPYTHON) -E $(srcdir)/setup.py build;; \
>> ? ? ? ?esac
>>
>> ?# Build static library
>> @@ -440,18 +446,18 @@
>>
>> ?libpython$(VERSION).so: $(LIBRARY_OBJS)
>> ? ? ? ?if test $(INSTSONAME) != $(LDLIBRARY); then \
>> - ? ? ? ? ? ? ? $(LDSHARED) $(LDFLAGS) -Wl,-h$(INSTSONAME) -o $(INSTSONAME) $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
>> + ? ? ? ? ? ? ? $(LDSHARED) $(PY_LDFLAGS) -Wl,-h$(INSTSONAME) -o $(INSTSONAME) $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
>> ? ? ? ? ? ? ? ?$(LN) -f $(INSTSONAME) $@; \
>> ? ? ? ?else \
>> - ? ? ? ? ? ? ? $(LDSHARED) $(LDFLAGS) -o $@ $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
>> + ? ? ? ? ? ? ? $(LDSHARED) $(PY_LDFLAGS) -o $@ $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
>> ? ? ? ?fi
>>
>> ?libpython$(VERSION).dylib: $(LIBRARY_OBJS)
>> - ? ? ? ?$(CC) -dynamiclib -Wl,-single_module $(LDFLAGS) -undefined dynamic_lookup -Wl,-install_name,$(prefix)/lib/libpython$(VERSION).dylib -Wl,-compatibility_version,$(VERSION) -Wl,-current_version,$(VERSION) -o $@ $(LIBRARY_OBJS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
>> + ? ? ? ?$(CC) -dynamiclib -Wl,-single_module $(PY_LDFLAGS) -undefined dynamic_lookup -Wl,-install_name,$(prefix)/lib/libpython$(VERSION).dylib -Wl,-compatibility_version,$(VERSION) -Wl,-current_version,$(VERSION) -o $@ $(LIBRARY_OBJS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
>>
>>
>> ?libpython$(VERSION).sl: $(LIBRARY_OBJS)
>> - ? ? ? $(LDSHARED) $(LDFLAGS) -o $@ $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST)
>> + ? ? ? $(LDSHARED) $(PY_LDFLAGS) -o $@ $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST)
>>
>> ?# Copy up the gdb python hooks into a position where they can be automatically
>> ?# loaded by gdb during Lib/test/test_gdb.py
>> @@ -497,7 +503,7 @@
>> ?# for a shared core library; otherwise, this rule is a noop.
>> ?$(DLLLIBRARY) libpython$(VERSION).dll.a: $(LIBRARY_OBJS)
>> ? ? ? ?if test -n "$(DLLLIBRARY)"; then \
>> - ? ? ? ? ? ? ? $(LDSHARED) $(LDFLAGS) -Wl,--out-implib=$@ -o $(DLLLIBRARY) $^ \
>> + ? ? ? ? ? ? ? $(LDSHARED) $(PY_LDFLAGS) -Wl,--out-implib=$@ -o $(DLLLIBRARY) $^ \
>> ? ? ? ? ? ? ? ? ? ? ? ?$(LIBS) $(MODLIBS) $(SYSLIBS) $(LDLAST); \
>> ? ? ? ?else true; \
>> ? ? ? ?fi
>> @@ -541,10 +547,10 @@
>> ? ? ? ? ? ? ? ?$(SIGNAL_OBJS) \
>> ? ? ? ? ? ? ? ?$(MODOBJS) \
>> ? ? ? ? ? ? ? ?$(srcdir)/Modules/getbuildinfo.c
>> - ? ? ? $(CC) -c $(PY_CFLAGS) -DSVNVERSION="\"`LC_ALL=C $(SVNVERSION)`\"" -o $@ $(srcdir)/Modules/getbuildinfo.c
>> + ? ? ? $(CC) -c $(PY_CORE_CFLAGS) -DSVNVERSION="\"`LC_ALL=C $(SVNVERSION)`\"" -o $@ $(srcdir)/Modules/getbuildinfo.c
>>
>> ?Modules/getpath.o: $(srcdir)/Modules/getpath.c Makefile
>> - ? ? ? $(CC) -c $(PY_CFLAGS) -DPYTHONPATH='"$(PYTHONPATH)"' \
>> + ? ? ? $(CC) -c $(PY_CORE_CFLAGS) -DPYTHONPATH='"$(PYTHONPATH)"' \
>> ? ? ? ? ? ? ? ?-DPREFIX='"$(prefix)"' \
>> ? ? ? ? ? ? ? ?-DEXEC_PREFIX='"$(exec_prefix)"' \
>> ? ? ? ? ? ? ? ?-DVERSION='"$(VERSION)"' \
>> @@ -552,7 +558,7 @@
>> ? ? ? ? ? ? ? ?-o $@ $(srcdir)/Modules/getpath.c
>>
>> ?Modules/python.o: $(srcdir)/Modules/python.c
>> - ? ? ? $(MAINCC) -c $(PY_CFLAGS) -o $@ $(srcdir)/Modules/python.c
>> + ? ? ? $(MAINCC) -c $(PY_CORE_CFLAGS) -o $@ $(srcdir)/Modules/python.c
>>
>> ?$(IO_OBJS): $(IO_H)
>>
>> @@ -561,7 +567,7 @@
>> ? ? ? ? ? ? ? ?-$(PGEN) $(GRAMMAR_INPUT) $(GRAMMAR_H) $(GRAMMAR_C)
>>
>> ?$(PGEN): ? ? ? $(PGENOBJS)
>> - ? ? ? ? ? ? ? $(CC) $(OPT) $(LDFLAGS) $(PGENOBJS) $(LIBS) -o $(PGEN)
>> + ? ? ? ? ? ? ? $(CC) $(OPT) $(PY_LDFLAGS) $(PGENOBJS) $(LIBS) -o $(PGEN)
>>
>> ?Parser/grammar.o: ? ? ?$(srcdir)/Parser/grammar.c \
>> ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?$(srcdir)/Include/token.h \
>> @@ -581,10 +587,10 @@
>> ?Python/compile.o Python/symtable.o Python/ast.o: $(GRAMMAR_H) $(AST_H)
>>
>> ?Python/getplatform.o: $(srcdir)/Python/getplatform.c
>> - ? ? ? ? ? ? ? $(CC) -c $(PY_CFLAGS) -DPLATFORM='"$(MACHDEP)"' -o $@ $(srcdir)/Python/getplatform.c
>> + ? ? ? ? ? ? ? $(CC) -c $(PY_CORE_CFLAGS) -DPLATFORM='"$(MACHDEP)"' -o $@ $(srcdir)/Python/getplatform.c
>>
>> ?Python/importdl.o: $(srcdir)/Python/importdl.c
>> - ? ? ? ? ? ? ? $(CC) -c $(PY_CFLAGS) -I$(DLINCLDIR) -o $@ $(srcdir)/Python/importdl.c
>> + ? ? ? ? ? ? ? $(CC) -c $(PY_CORE_CFLAGS) -I$(DLINCLDIR) -o $@ $(srcdir)/Python/importdl.c
>>
>> ?Objects/unicodectype.o: ? ? ? ?$(srcdir)/Objects/unicodectype.c \
>> ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?$(srcdir)/Objects/unicodetype_db.h
>> @@ -1130,7 +1136,7 @@
>>
>> ?# Some make's put the object file in the current directory
>> ?.c.o:
>> - ? ? ? $(CC) -c $(PY_CFLAGS) -o $@ $<
>> + ? ? ? $(CC) -c $(PY_CORE_CFLAGS) -o $@ $<
>>
>> ?# Run reindent on the library
>> ?reindent:
>>
>> Modified: python/branches/py3k/Misc/NEWS
>> ==============================================================================
>> --- python/branches/py3k/Misc/NEWS ? ? ?(original)
>> +++ python/branches/py3k/Misc/NEWS ? ? ?Fri Jul ?9 18:30:58 2010
>> @@ -1530,6 +1530,12 @@
>> ?Build
>> ?-----
>>
>> +- Issue #9189: Make a user-specified CFLAGS, CPPFLAGS, or LDFLAGS
>> + ?setting override the configure and makefile defaults, without
>> + ?deleting options the user didn't intend to override. ?Developers
>> + ?should no longer need to specify OPT or EXTRA_CFLAGS, although those
>> + ?variables are still present for backward-compatibility.
>> +
>> ?- Issue #8854: Fix finding Visual Studio 2008 on Windows x64.
>>
>> ?- Issue #1759169, #8864: Drop _XOPEN_SOURCE on Solaris, define it for
>>
>> Modified: python/branches/py3k/Modules/makesetup
>> ==============================================================================
>> --- python/branches/py3k/Modules/makesetup ? ? ?(original)
>> +++ python/branches/py3k/Modules/makesetup ? ? ?Fri Jul ?9 18:30:58 2010
>> @@ -219,7 +219,7 @@
>> ? ? ? ? ? ? ? ? ? ? ? ?case $doconfig in
>> ? ? ? ? ? ? ? ? ? ? ? ?no) ? ? cc="$cc \$(CCSHARED) \$(CFLAGS) \$(CPPFLAGS)";;
>> ? ? ? ? ? ? ? ? ? ? ? ?*)
>> - ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? cc="$cc \$(PY_CFLAGS)";;
>> + ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? cc="$cc \$(PY_CORE_CFLAGS)";;
>> ? ? ? ? ? ? ? ? ? ? ? ?esac
>> ? ? ? ? ? ? ? ? ? ? ? ?rule="$obj: $src; $cc $cpps -c $src -o $obj"
>> ? ? ? ? ? ? ? ? ? ? ? ?echo "$rule" >>$rulesf
>>
>> Modified: python/branches/py3k/configure
>> ==============================================================================
>> --- python/branches/py3k/configure ? ? ?(original)
>> +++ python/branches/py3k/configure ? ? ?Fri Jul ?9 18:30:58 2010
>> @@ -1929,11 +1929,11 @@
>> ? ? ? ?cat confdefs.h - <<_ACEOF >conftest.$ac_ext
>> ?/* end confdefs.h. ?*/
>> ?$ac_includes_default
>> + ? ? ? ? ? ?enum { N = $2 / 2 - 1 };
>> ?int
>> ?main ()
>> ?{
>> -static int test_array [1 - 2 * !(enum { N = $2 / 2 - 1 };
>> - ? ? ? ? ? ?0 < ($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1))];
>> +static int test_array [1 - 2 * !(0 < ($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1))];
>> ?test_array [0] = 0
>>
>> ? ;
>> @@ -1944,11 +1944,11 @@
>> ? cat confdefs.h - <<_ACEOF >conftest.$ac_ext
>> ?/* end confdefs.h. ?*/
>> ?$ac_includes_default
>> + ? ? ? ? ? ? ? enum { N = $2 / 2 - 1 };
>> ?int
>> ?main ()
>> ?{
>> -static int test_array [1 - 2 * !(enum { N = $2 / 2 - 1 };
>> - ? ? ? ? ? ? ? ($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1)
>> +static int test_array [1 - 2 * !(($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1)
>> ? ? ? ? ? ? ? ? < ($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 2))];
>> ?test_array [0] = 0
>>
>> @@ -3158,9 +3158,12 @@
>> ?(it is also a good idea to do 'make clean' before compiling)" "$LINENO" 5
>> ?fi
>>
>> -# If the user set CFLAGS, use this instead of the automatically
>> -# determined setting
>> -preset_cflags="$CFLAGS"
>> +# Don't let AC_PROG_CC set the default CFLAGS. It normally sets -g -O2
>> +# when the compiler supports them, but we don't always want -O2, and
>> +# we set -g later.
>> +if test -z "$CFLAGS"; then
>> + ? ? ? ?CFLAGS=
>> +fi
>> ?ac_ext=c
>> ?ac_cpp='$CPP $CPPFLAGS'
>> ?ac_compile='$CC -c $CFLAGS $CPPFLAGS conftest.$ac_ext >&5'
>> @@ -3952,10 +3955,6 @@
>> ?ac_link='$CC -o conftest$ac_exeext $CFLAGS $CPPFLAGS $LDFLAGS conftest.$ac_ext $LIBS >&5'
>> ?ac_compiler_gnu=$ac_cv_c_compiler_gnu
>>
>> -if test ! -z "$preset_cflags"
>> -then
>> - ? ? ? CFLAGS=$preset_cflags
>> -fi
>>
>>
>>
>>
>> Modified: python/branches/py3k/configure.in
>> ==============================================================================
>> --- python/branches/py3k/configure.in ? (original)
>> +++ python/branches/py3k/configure.in ? Fri Jul ?9 18:30:58 2010
>> @@ -461,14 +461,13 @@
>> ?(it is also a good idea to do 'make clean' before compiling)])
>> ?fi
>>
>> -# If the user set CFLAGS, use this instead of the automatically
>> -# determined setting
>> -preset_cflags="$CFLAGS"
>> -AC_PROG_CC
>> -if test ! -z "$preset_cflags"
>> -then
>> - ? ? ? CFLAGS=$preset_cflags
>> +# Don't let AC_PROG_CC set the default CFLAGS. It normally sets -g -O2
>> +# when the compiler supports them, but we don't always want -O2, and
>> +# we set -g later.
>> +if test -z "$CFLAGS"; then
>> + ? ? ? ?CFLAGS=
>> ?fi
>> +AC_PROG_CC
>>
>> ?AC_SUBST(CXX)
>> ?AC_SUBST(MAINCC)
>> _______________________________________________
>> Python-checkins mailing list
>> Python-checkins at python.org
>> http://mail.python.org/mailman/listinfo/python-checkins
>>
>
>
>
> --
> Regards,
> Benjamin
>

From python-checkins at python.org  Sat Jul 10 00:12:52 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sat, 10 Jul 2010 00:12:52 +0200 (CEST)
Subject: [Python-checkins] r82755 -
	python/branches/release31-maint/Python/pystrtod.c
Message-ID: <20100709221252.C5725EE999@mail.python.org>

Author: mark.dickinson
Date: Sat Jul 10 00:12:52 2010
New Revision: 82755

Log:
Revert temporary debugging commits (r82752, r82754).

Modified:
   python/branches/release31-maint/Python/pystrtod.c

Modified: python/branches/release31-maint/Python/pystrtod.c
==============================================================================
--- python/branches/release31-maint/Python/pystrtod.c	(original)
+++ python/branches/release31-maint/Python/pystrtod.c	Sat Jul 10 00:12:52 2010
@@ -1001,17 +1001,9 @@
         else {
             /* shouldn't get here: Gay's code should always return
                something starting with a digit, an 'I',  or 'N' */
-            printf("Unexpected failure in format_float_short. "
-                   "Arguments: d = %.17g, format_code = %d, "
-                   "mode = %d, precision = %ld\n",
-                   d, format_code, mode, precision);
-            printf("digits == %.100s\n", digits);
-            PyErr_Format(PyExc_RuntimeError,
-                         "Unexpected failure in format_float_short. "
-                         "Arguments: d = %.17g, format_code = %d, "
-                         "mode = %d, precision = %ld\n",
-                         d, format_code, mode, precision);
-            return NULL;
+            strncpy(p, "ERR", 3);
+            p += 3;
+            assert(0);
         }
         goto exit;
     }

From python-checkins at python.org  Sat Jul 10 00:15:07 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sat, 10 Jul 2010 00:15:07 +0200 (CEST)
Subject: [Python-checkins] r82756 - in python/branches/release31-maint:
	configure configure.in
Message-ID: <20100709221507.657B0EE9B5@mail.python.org>

Author: mark.dickinson
Date: Sat Jul 10 00:15:07 2010
New Revision: 82756

Log:
Temporary debug commit: include gcc version in configure output.  If this doesn't help with reproducing test_float failures then I'll create a separate branch to work on this issue.

Modified:
   python/branches/release31-maint/configure
   python/branches/release31-maint/configure.in

Modified: python/branches/release31-maint/configure
==============================================================================
--- python/branches/release31-maint/configure	(original)
+++ python/branches/release31-maint/configure	Sat Jul 10 00:15:07 2010
@@ -1,5 +1,5 @@
 #! /bin/sh
-# From configure.in Revision: 80702 .
+# From configure.in Revision: 82141 .
 # Guess values for system-dependent variables and create Makefiles.
 # Generated by GNU Autoconf 2.65 for python 3.1.
 #
@@ -1923,11 +1923,11 @@
        cat confdefs.h - <<_ACEOF >conftest.$ac_ext
 /* end confdefs.h.  */
 $ac_includes_default
+	     enum { N = $2 / 2 - 1 };
 int
 main ()
 {
-static int test_array [1 - 2 * !(enum { N = $2 / 2 - 1 };
-	     0 < ($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1))];
+static int test_array [1 - 2 * !(0 < ($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1))];
 test_array [0] = 0
 
   ;
@@ -1938,11 +1938,11 @@
   cat confdefs.h - <<_ACEOF >conftest.$ac_ext
 /* end confdefs.h.  */
 $ac_includes_default
+	        enum { N = $2 / 2 - 1 };
 int
 main ()
 {
-static int test_array [1 - 2 * !(enum { N = $2 / 2 - 1 };
-		($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1)
+static int test_array [1 - 2 * !(($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1)
 		 < ($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 2))];
 test_array [0] = 0
 
@@ -4126,6 +4126,15 @@
   " >&2;}
 fi
 
+# If using GCC, give result of 'gcc -v' in configure output.
+if test "$GCC" = yes; then
+{ $as_echo "$as_me:${as_lineno-$LINENO}: checking gcc version" >&5
+$as_echo_n "checking gcc version... " >&6; }
+gcc_version=`$CC -v`
+{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $gcc_version" >&5
+$as_echo "$gcc_version" >&6; }
+fi
+
 
 # checks for UNIX variants that set C preprocessor variables
 

Modified: python/branches/release31-maint/configure.in
==============================================================================
--- python/branches/release31-maint/configure.in	(original)
+++ python/branches/release31-maint/configure.in	Sat Jul 10 00:15:07 2010
@@ -504,6 +504,13 @@
   ])
 fi
 
+# If using GCC, give result of 'gcc -v' in configure output.
+if test "$GCC" = yes; then
+AC_MSG_CHECKING(gcc version)
+gcc_version=`$CC -v`
+AC_MSG_RESULT($gcc_version)
+fi
+
 
 # checks for UNIX variants that set C preprocessor variables
 AC_AIX

From solipsis at pitrou.net  Sat Jul 10 01:24:16 2010
From: solipsis at pitrou.net (solipsis at pitrou.net)
Date: Sat, 10 Jul 2010 01:24:16 +0200 (CEST)
Subject: [Python-checkins] Daily py3k reference leaks (r82753): sum=-252
Message-ID: <20100709232416.C4B081770A@ns6635.ovh.net>

py3k results for svn r82753 (hg cset 9ee888b8715b)
--------------------------------------------------

test_fnmatch leaked [114, 114, -480] references, sum=-252


Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/antoine/py3k/refleaks/reflog6AkyYS', '-x']

From nnorwitz at gmail.com  Sat Jul 10 03:04:05 2010
From: nnorwitz at gmail.com (Neal Norwitz)
Date: Fri, 9 Jul 2010 18:04:05 -0700
Subject: [Python-checkins] Python Regression Test Failures basics (1)
In-Reply-To: <AANLkTilSMxqLkUjjrDCljtq70n76PATJoOVCsJnulj0z@mail.gmail.com>
References: <20100709083458.GA27283@kbk-i386-bb.psfb.org>
	<AANLkTilSMxqLkUjjrDCljtq70n76PATJoOVCsJnulj0z@mail.gmail.com>
Message-ID: <AANLkTilifiQ2JaL-O0wq-HoEXcEQRoCkRZexs36LujIH@mail.gmail.com>

On Fri, Jul 9, 2010 at 3:26 AM, Nick Coghlan <ncoghlan at gmail.com> wrote:
> On Fri, Jul 9, 2010 at 6:34 PM, Neal Norwitz <nnorwitz at gmail.com> wrote:
>> == CPython 2.7 (trunk:82729M, Jul 9 2010, 04:02:46) [GCC 3.3.4 20040623 (Gentoo Linux 3.3.4-r1, ssp-3.3.2-2, pie-8.7.6)]
>> == ? Linux-2.6.9-gentoo-r1-i686-AMD_Athlon-tm-_XP_3000+-with-gentoo-1.4.16 little-endian
>
> Neal, these scripts need to be updated to either pull from py3k or
> from the 2.7 maintenance branch. They aren't going to see much action
> monitoring the trunk from now on :)

Good point.  Thanks for point this out.  I'm not sure if I have access
to this machine any more.  Kurt, can you fix this or make sure I have
access?  IIRC, there was a firewall that was opened for my IP address.
 I'm sure my address has changed since then.

I won't get a chance to look at this for over a week.  I'll be on
vacation for a bit.  I can look at this when I return.  Nick, feel
free to ping me in a few weeks if this isn't resolved.

n

From python-checkins at python.org  Sat Jul 10 10:58:37 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 10 Jul 2010 10:58:37 +0200 (CEST)
Subject: [Python-checkins] r82757 - in python/branches/py3k/Doc/library:
	os.path.rst os.rst
Message-ID: <20100710085837.3A460EE98E@mail.python.org>

Author: georg.brandl
Date: Sat Jul 10 10:58:37 2010
New Revision: 82757

Log:
Fix markup.

Modified:
   python/branches/py3k/Doc/library/os.path.rst
   python/branches/py3k/Doc/library/os.rst

Modified: python/branches/py3k/Doc/library/os.path.rst
==============================================================================
--- python/branches/py3k/Doc/library/os.path.rst	(original)
+++ python/branches/py3k/Doc/library/os.path.rst	Sat Jul 10 10:58:37 2010
@@ -241,7 +241,8 @@
 
    Availability: Windows, Unix.
 
-   .. versionchanged:: 3.2 Added Windows support
+   .. versionchanged:: 3.2
+      Added Windows support.
 
 
 .. function:: sameopenfile(fp1, fp2)

Modified: python/branches/py3k/Doc/library/os.rst
==============================================================================
--- python/branches/py3k/Doc/library/os.rst	(original)
+++ python/branches/py3k/Doc/library/os.rst	Sat Jul 10 10:58:37 2010
@@ -1066,7 +1066,8 @@
    Like :func:`stat`, but do not follow symbolic links.  This is an alias for
    :func:`stat` on platforms that do not support symbolic links.
 
-   .. versionchanged:: 3.2 Added support for Windows 6.0 (Vista)
+   .. versionchanged:: 3.2
+      Added support for Windows 6.0 (Vista) symbolic links.
 
 
 .. function:: mkfifo(path[, mode])
@@ -1184,7 +1185,8 @@
 
    Availability: Unix, Windows
 
-   .. versionchanged:: 3.2 Added support for Windows 6.0 (Vista)
+   .. versionchanged:: 3.2
+      Added support for Windows 6.0 (Vista) symbolic links.
 
 
 .. function:: remove(path)
@@ -1365,7 +1367,8 @@
 
    Availability:  Unix, Windows
 
-   .. versionchanged:: 3.2 Added support for Windows 6.0 (Vista)
+   .. versionchanged:: 3.2
+      Added support for Windows 6.0 (Vista) symbolic links.
 
 
 .. function:: unlink(path)

From python-checkins at python.org  Sat Jul 10 12:23:40 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 10 Jul 2010 12:23:40 +0200 (CEST)
Subject: [Python-checkins] r82758 -
	python/branches/py3k/Doc/library/struct.rst
Message-ID: <20100710102340.40216EE98E@mail.python.org>

Author: georg.brandl
Date: Sat Jul 10 12:23:40 2010
New Revision: 82758

Log:
Emphasize role of count for Pascal string.

Modified:
   python/branches/py3k/Doc/library/struct.rst

Modified: python/branches/py3k/Doc/library/struct.rst
==============================================================================
--- python/branches/py3k/Doc/library/struct.rst	(original)
+++ python/branches/py3k/Doc/library/struct.rst	Sat Jul 10 12:23:40 2010
@@ -259,18 +259,16 @@
    In 3.0, some of the integer formats wrapped out-of-range values and
    raised :exc:`DeprecationWarning` instead of :exc:`struct.error`.
 
-
 The ``'p'`` format character encodes a "Pascal string", meaning a short
-variable-length string stored in a fixed number of bytes. The count is the total
-number of bytes stored.  The first byte stored is the length of the string, or
-255, whichever is smaller.  The bytes of the string follow.  If the string
-passed in to :func:`pack` is too long (longer than the count minus 1), only the
-leading count-1 bytes of the string are stored.  If the string is shorter than
-count-1, it is padded with null bytes so that exactly count bytes in all are
-used.  Note that for :func:`unpack`, the ``'p'`` format character consumes count
-bytes, but that the string returned can never contain more than 255 bytes.
-
-
+variable-length string stored in a *fixed number of bytes*, given by the count.
+The first byte stored is the length of the string, or 255, whichever is
+smaller.  The bytes of the string follow.  If the string passed in to
+:func:`pack` is too long (longer than the count minus 1), only the leading
+``count-1`` bytes of the string are stored.  If the string is shorter than
+``count-1``, it is padded with null bytes so that exactly count bytes in all
+are used.  Note that for :func:`unpack`, the ``'p'`` format character consumes
+``count`` bytes, but that the string returned can never contain more than 255
+bytes.
 
 For the ``'?'`` format character, the return value is either :const:`True` or
 :const:`False`. When packing, the truth value of the argument object is used.

From python-checkins at python.org  Sat Jul 10 12:32:36 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 10 Jul 2010 12:32:36 +0200 (CEST)
Subject: [Python-checkins] r82759 - in python/branches/py3k:
	Lib/test/test_unpack.py Python/ceval.c
Message-ID: <20100710103236.4435DEE981@mail.python.org>

Author: georg.brandl
Date: Sat Jul 10 12:32:36 2010
New Revision: 82759

Log:
#3071: tell how many values were expected when unpacking too many.

Modified:
   python/branches/py3k/Lib/test/test_unpack.py
   python/branches/py3k/Python/ceval.c

Modified: python/branches/py3k/Lib/test/test_unpack.py
==============================================================================
--- python/branches/py3k/Lib/test/test_unpack.py	(original)
+++ python/branches/py3k/Lib/test/test_unpack.py	Sat Jul 10 12:32:36 2010
@@ -62,14 +62,14 @@
     >>> a, b = t
     Traceback (most recent call last):
       ...
-    ValueError: too many values to unpack
+    ValueError: too many values to unpack (expected 2)
 
 Unpacking tuple of wrong size
 
     >>> a, b = l
     Traceback (most recent call last):
       ...
-    ValueError: too many values to unpack
+    ValueError: too many values to unpack (expected 2)
 
 Unpacking sequence too short
 
@@ -83,7 +83,7 @@
     >>> a, b = Seq()
     Traceback (most recent call last):
       ...
-    ValueError: too many values to unpack
+    ValueError: too many values to unpack (expected 2)
 
 Unpacking a sequence where the test for too long raises a different kind of
 error

Modified: python/branches/py3k/Python/ceval.c
==============================================================================
--- python/branches/py3k/Python/ceval.c	(original)
+++ python/branches/py3k/Python/ceval.c	Sat Jul 10 12:32:36 2010
@@ -3464,7 +3464,8 @@
             return 1;
         }
         Py_DECREF(w);
-        PyErr_SetString(PyExc_ValueError, "too many values to unpack");
+        PyErr_Format(PyExc_ValueError, "too many values to unpack "
+                     "(expected %d)", argcnt);
         goto Error;
     }
 

From python-checkins at python.org  Sat Jul 10 12:39:57 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 10 Jul 2010 12:39:57 +0200 (CEST)
Subject: [Python-checkins] r82760 - python/branches/py3k/Doc/glossary.rst
Message-ID: <20100710103957.8C91CEEA63@mail.python.org>

Author: georg.brandl
Date: Sat Jul 10 12:39:57 2010
New Revision: 82760

Log:
#3214: improve description of duck-typing in glossary.

Modified:
   python/branches/py3k/Doc/glossary.rst

Modified: python/branches/py3k/Doc/glossary.rst
==============================================================================
--- python/branches/py3k/Doc/glossary.rst	(original)
+++ python/branches/py3k/Doc/glossary.rst	Sat Jul 10 12:39:57 2010
@@ -154,9 +154,9 @@
       object.
 
    duck-typing
-      A pythonic programming style which determines an object's type by inspection
-      of its method or attribute signature rather than by explicit relationship
-      to some type object ("If it looks like a duck and quacks like a duck, it
+      A programming style which does not look at an object's type to determine
+      if it has the right interface; instead, the method or attribute is simply
+      called or used ("If it looks like a duck and quacks like a duck, it
       must be a duck.")  By emphasizing interfaces rather than specific types,
       well-designed code improves its flexibility by allowing polymorphic
       substitution.  Duck-typing avoids tests using :func:`type` or

From python-checkins at python.org  Sat Jul 10 13:40:13 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 10 Jul 2010 13:40:13 +0200 (CEST)
Subject: [Python-checkins] r82761 -
	python/branches/py3k/Lib/xml/dom/expatbuilder.py
Message-ID: <20100710114013.2F076EEA50@mail.python.org>

Author: georg.brandl
Date: Sat Jul 10 13:40:13 2010
New Revision: 82761

Log:
#1434090: properly append child in expatbuilder doctype handler.

Modified:
   python/branches/py3k/Lib/xml/dom/expatbuilder.py

Modified: python/branches/py3k/Lib/xml/dom/expatbuilder.py
==============================================================================
--- python/branches/py3k/Lib/xml/dom/expatbuilder.py	(original)
+++ python/branches/py3k/Lib/xml/dom/expatbuilder.py	Sat Jul 10 13:40:13 2010
@@ -242,7 +242,7 @@
         doctype = self.document.implementation.createDocumentType(
             doctypeName, publicId, systemId)
         doctype.ownerDocument = self.document
-        self.document.childNodes.append(doctype)
+        _append_child(self.document, doctype)
         self.document.doctype = doctype
         if self._filter and self._filter.acceptNode(doctype) == FILTER_REJECT:
             self.document.doctype = None

From python-checkins at python.org  Sat Jul 10 13:51:07 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 10 Jul 2010 13:51:07 +0200 (CEST)
Subject: [Python-checkins] r82762 -
	python/branches/py3k/Doc/library/multiprocessing.rst
Message-ID: <20100710115107.1C636EE985@mail.python.org>

Author: georg.brandl
Date: Sat Jul 10 13:51:06 2010
New Revision: 82762

Log:
#8338: fix outdated class name.

Modified:
   python/branches/py3k/Doc/library/multiprocessing.rst

Modified: python/branches/py3k/Doc/library/multiprocessing.rst
==============================================================================
--- python/branches/py3k/Doc/library/multiprocessing.rst	(original)
+++ python/branches/py3k/Doc/library/multiprocessing.rst	Sat Jul 10 13:51:06 2010
@@ -2228,8 +2228,8 @@
 
 
 An example of how a pool of worker processes can each run a
-:class:`SimpleHTTPServer.HttpServer` instance while sharing a single listening
-socket.
+:class:`~http.server.SimpleHTTPRequestHandler` instance while sharing a single
+listening socket.
 
 .. literalinclude:: ../includes/mp_webserver.py
 

From python-checkins at python.org  Sat Jul 10 14:01:35 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 10 Jul 2010 14:01:35 +0200 (CEST)
Subject: [Python-checkins] r82763 -
	python/branches/py3k/Doc/library/sqlite3.rst
Message-ID: <20100710120135.11C8CEE985@mail.python.org>

Author: georg.brandl
Date: Sat Jul 10 14:01:34 2010
New Revision: 82763

Log:
#8456: fix signature of sqlite3.connect().

Modified:
   python/branches/py3k/Doc/library/sqlite3.rst

Modified: python/branches/py3k/Doc/library/sqlite3.rst
==============================================================================
--- python/branches/py3k/Doc/library/sqlite3.rst	(original)
+++ python/branches/py3k/Doc/library/sqlite3.rst	Sat Jul 10 14:01:34 2010
@@ -136,7 +136,7 @@
    first blank for the column name: the column name would simply be "x".
 
 
-.. function:: connect(database[, timeout, isolation_level, detect_types, factory])
+.. function:: connect(database[, timeout, detect_types, isolation_level, check_same_thread, factory, cached_statements])
 
    Opens a connection to the SQLite database file *database*. You can use
    ``":memory:"`` to open a database connection to a database that resides in RAM

From python-checkins at python.org  Sat Jul 10 14:20:38 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 10 Jul 2010 14:20:38 +0200 (CEST)
Subject: [Python-checkins] r82764 -
	python/branches/py3k/Doc/library/doctest.rst
Message-ID: <20100710122038.C4E8AEE981@mail.python.org>

Author: georg.brandl
Date: Sat Jul 10 14:20:38 2010
New Revision: 82764

Log:
#8564: update docs on integrating doctest/unittest with unittest(2) test discovery.

Modified:
   python/branches/py3k/Doc/library/doctest.rst

Modified: python/branches/py3k/Doc/library/doctest.rst
==============================================================================
--- python/branches/py3k/Doc/library/doctest.rst	(original)
+++ python/branches/py3k/Doc/library/doctest.rst	Sat Jul 10 14:20:38 2010
@@ -913,18 +913,16 @@
 As your collection of doctest'ed modules grows, you'll want a way to run all
 their doctests systematically.  :mod:`doctest` provides two functions that can
 be used to create :mod:`unittest` test suites from modules and text files
-containing doctests.  These test suites can then be run using :mod:`unittest`
-test runners::
+containing doctests.  To integrate with :mod:`unittest` test discovery, include
+a :func:`load_tests` function in your test module::
 
    import unittest
    import doctest
-   import my_module_with_doctests, and_another
+   import my_module_with_doctests
 
-   suite = unittest.TestSuite()
-   for mod in my_module_with_doctests, and_another:
-       suite.addTest(doctest.DocTestSuite(mod))
-   runner = unittest.TextTestRunner()
-   runner.run(suite)
+   def load_tests(loader, tests, ignore):
+       tests.addTests(doctest.DocTestSuite(my_module_with_doctests))
+       return test
 
 There are two main functions for creating :class:`unittest.TestSuite` instances
 from text files and modules with doctests:

From python-checkins at python.org  Sat Jul 10 15:51:42 2010
From: python-checkins at python.org (michael.foord)
Date: Sat, 10 Jul 2010 15:51:42 +0200 (CEST)
Subject: [Python-checkins] r82765 - in
	python/branches/release27-maint/Lib/unittest: case.py
	test/test_case.py
Message-ID: <20100710135142.97F3AEEA1A@mail.python.org>

Author: michael.foord
Date: Sat Jul 10 15:51:42 2010
New Revision: 82765

Log:
Fix error message for comparing single line strings in unittest.TestCase.assertEqual.
Issue 9174

Modified:
   python/branches/release27-maint/Lib/unittest/case.py
   python/branches/release27-maint/Lib/unittest/test/test_case.py

Modified: python/branches/release27-maint/Lib/unittest/case.py
==============================================================================
--- python/branches/release27-maint/Lib/unittest/case.py	(original)
+++ python/branches/release27-maint/Lib/unittest/case.py	Sat Jul 10 15:51:42 2010
@@ -895,9 +895,14 @@
                 'Second argument is not a string'))
 
         if first != second:
-            standardMsg = '%s != %s' % (safe_repr(first, True), safe_repr(second, True))
-            diff = '\n' + ''.join(difflib.ndiff(first.splitlines(True),
-                                                       second.splitlines(True)))
+            firstlines = first.splitlines(True)
+            secondlines = second.splitlines(True)
+            if len(firstlines) == 1 and first.strip('\r\n') == first:
+                firstlines = [first + '\n']
+                secondlines = [second + '\n']
+            standardMsg = '%s != %s' % (safe_repr(first, True),
+                                        safe_repr(second, True))
+            diff = '\n' + ''.join(difflib.ndiff(firstlines, secondlines))
             standardMsg = self._truncateMessage(standardMsg, diff)
             self.fail(self._formatMessage(msg, standardMsg))
 

Modified: python/branches/release27-maint/Lib/unittest/test/test_case.py
==============================================================================
--- python/branches/release27-maint/Lib/unittest/test/test_case.py	(original)
+++ python/branches/release27-maint/Lib/unittest/test/test_case.py	Sat Jul 10 15:51:42 2010
@@ -874,6 +874,21 @@
                 # unicode strings - so we can't use it for this check
                 self.assertTrue(sample_text_error == error)
 
+    def testAsertEqualSingleLine(self):
+        sample_text = u"laden swallows fly slowly"
+        revised_sample_text = u"unladen swallows fly quickly"
+        sample_text_error = """\
+- laden swallows fly slowly
+?                    ^^^^
++ unladen swallows fly quickly
+? ++                   ^^^^^
+"""
+        try:
+            self.assertEqual(sample_text, revised_sample_text)
+        except self.failureException as e:
+            error = str(e).split('\n', 1)[1]
+            self.assertTrue(sample_text_error == error)
+
     def testAssertIsNone(self):
         self.assertIsNone(None)
         self.assertRaises(self.failureException, self.assertIsNone, False)

From python-checkins at python.org  Sat Jul 10 15:52:13 2010
From: python-checkins at python.org (r.david.murray)
Date: Sat, 10 Jul 2010 15:52:13 +0200 (CEST)
Subject: [Python-checkins] r82766 - in python/branches/py3k:
	Doc/library/fnmatch.rst Lib/fnmatch.py Lib/test/test_fnmatch.py
Message-ID: <20100710135213.890E4EEA51@mail.python.org>

Author: r.david.murray
Date: Sat Jul 10 15:52:13 2010
New Revision: 82766

Log:
Fix 'refleak' introduced by fnmatch cache purge tests.

This introduces a 'purge' function for the fnmatch module analogous
to the 'purge' function in the re module.


Modified:
   python/branches/py3k/Doc/library/fnmatch.rst
   python/branches/py3k/Lib/fnmatch.py
   python/branches/py3k/Lib/test/test_fnmatch.py

Modified: python/branches/py3k/Doc/library/fnmatch.rst
==============================================================================
--- python/branches/py3k/Doc/library/fnmatch.rst	(original)
+++ python/branches/py3k/Doc/library/fnmatch.rst	Sat Jul 10 15:52:13 2010
@@ -82,6 +82,13 @@
       <_sre.SRE_Match object at 0x...>
 
 
+.. function:: purge()
+
+   Clear the internal pattern cache.
+
+   .. versionadded:: 3.2
+
+
 .. seealso::
 
    Module :mod:`glob`

Modified: python/branches/py3k/Lib/fnmatch.py
==============================================================================
--- python/branches/py3k/Lib/fnmatch.py	(original)
+++ python/branches/py3k/Lib/fnmatch.py	Sat Jul 10 15:52:13 2010
@@ -12,12 +12,17 @@
 
 import re
 
-__all__ = ["filter", "fnmatch","fnmatchcase","translate"]
+__all__ = ["filter", "fnmatch", "fnmatchcase", "purge", "translate"]
 
 _cache = {}  # Maps text patterns to compiled regexen.
 _cacheb = {}  # Ditto for bytes patterns.
 _MAXCACHE = 100 # Maximum size of caches
 
+def purge():
+    """Clear the pattern cache"""
+    _cache.clear()
+    _cacheb.clear()
+
 def fnmatch(name, pat):
     """Test whether FILENAME matches PATTERN.
 

Modified: python/branches/py3k/Lib/test/test_fnmatch.py
==============================================================================
--- python/branches/py3k/Lib/test/test_fnmatch.py	(original)
+++ python/branches/py3k/Lib/test/test_fnmatch.py	Sat Jul 10 15:52:13 2010
@@ -3,10 +3,14 @@
 from test import support
 import unittest
 
-from fnmatch import fnmatch, fnmatchcase, _MAXCACHE, _cache, _cacheb
+from fnmatch import fnmatch, fnmatchcase, _MAXCACHE, _cache, _cacheb, purge
 
 
 class FnmatchTestCase(unittest.TestCase):
+
+    def tearDown(self):
+        purge()
+
     def check_match(self, filename, pattern, should_match=1, fn=fnmatch):
         if should_match:
             self.assertTrue(fn(filename, pattern),

From python-checkins at python.org  Sat Jul 10 15:52:22 2010
From: python-checkins at python.org (michael.foord)
Date: Sat, 10 Jul 2010 15:52:22 +0200 (CEST)
Subject: [Python-checkins] r82767 - in python/branches/py3k/Lib/unittest:
	case.py test/test_case.py
Message-ID: <20100710135222.97BDCEEA3A@mail.python.org>

Author: michael.foord
Date: Sat Jul 10 15:52:22 2010
New Revision: 82767

Log:
Fix error message for comparing single line strings in unittest.TestCase.assertEqual.
Issue 9174.

Modified:
   python/branches/py3k/Lib/unittest/case.py
   python/branches/py3k/Lib/unittest/test/test_case.py

Modified: python/branches/py3k/Lib/unittest/case.py
==============================================================================
--- python/branches/py3k/Lib/unittest/case.py	(original)
+++ python/branches/py3k/Lib/unittest/case.py	Sat Jul 10 15:52:22 2010
@@ -940,9 +940,14 @@
                 'Second argument is not a string'))
 
         if first != second:
-            standardMsg = '%s != %s' % (safe_repr(first, True), safe_repr(second, True))
-            diff = '\n' + ''.join(difflib.ndiff(first.splitlines(True),
-                                                       second.splitlines(True)))
+            firstlines = first.splitlines(True)
+            secondlines = second.splitlines(True)
+            if len(firstlines) == 1 and first.strip('\r\n') == first:
+                firstlines = [first + '\n']
+                secondlines = [second + '\n']
+            standardMsg = '%s != %s' % (safe_repr(first, True),
+                                        safe_repr(second, True))
+            diff = '\n' + ''.join(difflib.ndiff(firstlines, secondlines))
             standardMsg = self._truncateMessage(standardMsg, diff)
             self.fail(self._formatMessage(msg, standardMsg))
 

Modified: python/branches/py3k/Lib/unittest/test/test_case.py
==============================================================================
--- python/branches/py3k/Lib/unittest/test/test_case.py	(original)
+++ python/branches/py3k/Lib/unittest/test/test_case.py	Sat Jul 10 15:52:22 2010
@@ -841,6 +841,21 @@
             # so can't use assertEqual either. Just use assertTrue.
             self.assertTrue(sample_text_error == error)
 
+    def testAsertEqualSingleLine(self):
+        sample_text = "laden swallows fly slowly"
+        revised_sample_text = "unladen swallows fly quickly"
+        sample_text_error = """\
+- laden swallows fly slowly
+?                    ^^^^
++ unladen swallows fly quickly
+? ++                   ^^^^^
+"""
+        try:
+            self.assertEqual(sample_text, revised_sample_text)
+        except self.failureException as e:
+            error = str(e).split('\n', 1)[1]
+            self.assertTrue(sample_text_error == error)
+
     def testAssertIsNone(self):
         self.assertIsNone(None)
         self.assertRaises(self.failureException, self.assertIsNone, False)

From python-checkins at python.org  Sat Jul 10 16:00:10 2010
From: python-checkins at python.org (r.david.murray)
Date: Sat, 10 Jul 2010 16:00:10 +0200 (CEST)
Subject: [Python-checkins] r82768 - in python/branches/release31-maint:
	Lib/fnmatch.py Lib/test/test_fnmatch.py
Message-ID: <20100710140010.E071AEEA51@mail.python.org>

Author: r.david.murray
Date: Sat Jul 10 16:00:10 2010
New Revision: 82768

Log:
Merged revisions 82766 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

Since 'purge' is an API change, modified it to be _purge for 3.1
and deleted the doc update.

........
  r82766 | r.david.murray | 2010-07-10 09:52:13 -0400 (Sat, 10 Jul 2010) | 5 lines
  
  Fix 'refleak' introduced by fnmatch cache purge tests.
  
  This introduces a 'purge' function for the fnmatch module analogous
  to the 'purge' function in the re module.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/fnmatch.py
   python/branches/release31-maint/Lib/test/test_fnmatch.py

Modified: python/branches/release31-maint/Lib/fnmatch.py
==============================================================================
--- python/branches/release31-maint/Lib/fnmatch.py	(original)
+++ python/branches/release31-maint/Lib/fnmatch.py	Sat Jul 10 16:00:10 2010
@@ -12,12 +12,17 @@
 
 import re
 
-__all__ = ["filter", "fnmatch","fnmatchcase","translate"]
+__all__ = ["filter", "fnmatch", "fnmatchcase", "translate"]
 
 _cache = {}  # Maps text patterns to compiled regexen.
 _cacheb = {}  # Ditto for bytes patterns.
 _MAXCACHE = 100 # Maximum size of caches
 
+def _purge():
+    """Clear the pattern cache"""
+    _cache.clear()
+    _cacheb.clear()
+
 def fnmatch(name, pat):
     """Test whether FILENAME matches PATTERN.
 

Modified: python/branches/release31-maint/Lib/test/test_fnmatch.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_fnmatch.py	(original)
+++ python/branches/release31-maint/Lib/test/test_fnmatch.py	Sat Jul 10 16:00:10 2010
@@ -3,10 +3,14 @@
 from test import support
 import unittest
 
-from fnmatch import fnmatch, fnmatchcase, _MAXCACHE, _cache, _cacheb
+from fnmatch import fnmatch, fnmatchcase, _MAXCACHE, _cache, _cacheb, _purge
 
 
 class FnmatchTestCase(unittest.TestCase):
+
+    def tearDown(self):
+        _purge()
+
     def check_match(self, filename, pattern, should_match=1):
         if should_match:
             self.assertTrue(fnmatch(filename, pattern),

From python-checkins at python.org  Sat Jul 10 16:06:51 2010
From: python-checkins at python.org (r.david.murray)
Date: Sat, 10 Jul 2010 16:06:51 +0200 (CEST)
Subject: [Python-checkins] r82769 - in python/branches/release27-maint:
	Lib/fnmatch.py Lib/test/test_fnmatch.py
Message-ID: <20100710140651.52727EE9CF@mail.python.org>

Author: r.david.murray
Date: Sat Jul 10 16:06:51 2010
New Revision: 82769

Log:
Merged revisions 82766 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

Since 'purge' is an API change, modified it to be _purge for 2.x
and deleted the doc update.

........
  r82766 | r.david.murray | 2010-07-10 09:52:13 -0400 (Sat, 10 Jul 2010) | 5 lines
  
  Fix 'refleak' introduced by fnmatch cache purge tests.
  
  This introduces a 'purge' function for the fnmatch module analogous
  to the 'purge' function in the re module.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/fnmatch.py
   python/branches/release27-maint/Lib/test/test_fnmatch.py

Modified: python/branches/release27-maint/Lib/fnmatch.py
==============================================================================
--- python/branches/release27-maint/Lib/fnmatch.py	(original)
+++ python/branches/release27-maint/Lib/fnmatch.py	Sat Jul 10 16:06:51 2010
@@ -12,11 +12,15 @@
 
 import re
 
-__all__ = ["filter", "fnmatch","fnmatchcase","translate"]
+__all__ = ["filter", "fnmatch", "fnmatchcase", "translate"]
 
 _cache = {}
 _MAXCACHE = 100
 
+def _purge():
+    """Clear the pattern cache"""
+    _cache.clear()
+
 def fnmatch(name, pat):
     """Test whether FILENAME matches PATTERN.
 

Modified: python/branches/release27-maint/Lib/test/test_fnmatch.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_fnmatch.py	(original)
+++ python/branches/release27-maint/Lib/test/test_fnmatch.py	Sat Jul 10 16:06:51 2010
@@ -4,9 +4,14 @@
 import unittest
 
 from fnmatch import fnmatch, fnmatchcase, _MAXCACHE, _cache
+from fnmatch import fnmatch, fnmatchcase, _MAXCACHE, _cache, _purge
 
 
 class FnmatchTestCase(unittest.TestCase):
+
+    def tearDown(self):
+        _purge()
+
     def check_match(self, filename, pattern, should_match=1, fn=fnmatch):
         if should_match:
             self.assertTrue(fn(filename, pattern),

From python-checkins at python.org  Sat Jul 10 16:11:51 2010
From: python-checkins at python.org (r.david.murray)
Date: Sat, 10 Jul 2010 16:11:51 +0200 (CEST)
Subject: [Python-checkins] r82770 - in python/branches/release26-maint:
	Lib/fnmatch.py Lib/test/test_fnmatch.py
Message-ID: <20100710141151.2F468EE9C7@mail.python.org>

Author: r.david.murray
Date: Sat Jul 10 16:11:51 2010
New Revision: 82770

Log:
Merged revisions 82769 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

................
  r82769 | r.david.murray | 2010-07-10 10:06:51 -0400 (Sat, 10 Jul 2010) | 15 lines
  
  Merged revisions 82766 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k
  
  Since 'purge' is an API change, modified it to be _purge for 2.x
  and deleted the doc update.
  
  ........
    r82766 | r.david.murray | 2010-07-10 09:52:13 -0400 (Sat, 10 Jul 2010) | 5 lines
    
    Fix 'refleak' introduced by fnmatch cache purge tests.
    
    This introduces a 'purge' function for the fnmatch module analogous
    to the 'purge' function in the re module.
  ........
................


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Lib/fnmatch.py
   python/branches/release26-maint/Lib/test/test_fnmatch.py

Modified: python/branches/release26-maint/Lib/fnmatch.py
==============================================================================
--- python/branches/release26-maint/Lib/fnmatch.py	(original)
+++ python/branches/release26-maint/Lib/fnmatch.py	Sat Jul 10 16:11:51 2010
@@ -12,11 +12,15 @@
 
 import re
 
-__all__ = ["filter", "fnmatch","fnmatchcase","translate"]
+__all__ = ["filter", "fnmatch", "fnmatchcase", "translate"]
 
 _cache = {}
 _MAXCACHE = 100
 
+def _purge():
+    """Clear the pattern cache"""
+    _cache.clear()
+
 def fnmatch(name, pat):
     """Test whether FILENAME matches PATTERN.
 

Modified: python/branches/release26-maint/Lib/test/test_fnmatch.py
==============================================================================
--- python/branches/release26-maint/Lib/test/test_fnmatch.py	(original)
+++ python/branches/release26-maint/Lib/test/test_fnmatch.py	Sat Jul 10 16:11:51 2010
@@ -4,9 +4,14 @@
 import unittest
 
 from fnmatch import fnmatch, fnmatchcase, _MAXCACHE, _cache
+from fnmatch import fnmatch, fnmatchcase, _MAXCACHE, _cache, _purge
 
 
 class FnmatchTestCase(unittest.TestCase):
+
+    def tearDown(self):
+        _purge()
+
     def check_match(self, filename, pattern, should_match=1, fn=fnmatch):
         if should_match:
             self.assertTrue(fn(filename, pattern),

From python-checkins at python.org  Sat Jul 10 16:23:36 2010
From: python-checkins at python.org (r.david.murray)
Date: Sat, 10 Jul 2010 16:23:36 +0200 (CEST)
Subject: [Python-checkins] r82771 - python/branches/py3k/Doc/library/re.rst
Message-ID: <20100710142336.5AEA2EE981@mail.python.org>

Author: r.david.murray
Date: Sat Jul 10 16:23:36 2010
New Revision: 82771

Log:
Add missing docs for re module 'purge' function.


Modified:
   python/branches/py3k/Doc/library/re.rst

Modified: python/branches/py3k/Doc/library/re.rst
==============================================================================
--- python/branches/py3k/Doc/library/re.rst	(original)
+++ python/branches/py3k/Doc/library/re.rst	Sat Jul 10 16:23:36 2010
@@ -692,6 +692,11 @@
    metacharacters in it.
 
 
+.. function:: purge()
+
+   Clear the regular expression cache.
+
+
 .. exception:: error
 
    Exception raised when a string passed to one of the functions here is not a

From python-checkins at python.org  Sat Jul 10 16:24:46 2010
From: python-checkins at python.org (r.david.murray)
Date: Sat, 10 Jul 2010 16:24:46 +0200 (CEST)
Subject: [Python-checkins] r82772 - in python/branches/release31-maint:
	Doc/library/re.rst
Message-ID: <20100710142446.D8135EE981@mail.python.org>

Author: r.david.murray
Date: Sat Jul 10 16:24:46 2010
New Revision: 82772

Log:
Merged revisions 82771 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82771 | r.david.murray | 2010-07-10 10:23:36 -0400 (Sat, 10 Jul 2010) | 2 lines
  
  Add missing docs for re module 'purge' function.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Doc/library/re.rst

Modified: python/branches/release31-maint/Doc/library/re.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/re.rst	(original)
+++ python/branches/release31-maint/Doc/library/re.rst	Sat Jul 10 16:24:46 2010
@@ -692,6 +692,11 @@
    metacharacters in it.
 
 
+.. function:: purge()
+
+   Clear the regular expression cache.
+
+
 .. exception:: error
 
    Exception raised when a string passed to one of the functions here is not a

From python-checkins at python.org  Sat Jul 10 16:25:18 2010
From: python-checkins at python.org (r.david.murray)
Date: Sat, 10 Jul 2010 16:25:18 +0200 (CEST)
Subject: [Python-checkins] r82773 - in python/branches/release27-maint:
	Doc/library/re.rst
Message-ID: <20100710142518.3C2B0EE981@mail.python.org>

Author: r.david.murray
Date: Sat Jul 10 16:25:18 2010
New Revision: 82773

Log:
Merged revisions 82771 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82771 | r.david.murray | 2010-07-10 10:23:36 -0400 (Sat, 10 Jul 2010) | 2 lines
  
  Add missing docs for re module 'purge' function.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Doc/library/re.rst

Modified: python/branches/release27-maint/Doc/library/re.rst
==============================================================================
--- python/branches/release27-maint/Doc/library/re.rst	(original)
+++ python/branches/release27-maint/Doc/library/re.rst	Sat Jul 10 16:25:18 2010
@@ -672,6 +672,11 @@
    metacharacters in it.
 
 
+.. function:: purge()
+
+   Clear the regular expression cache.
+
+
 .. exception:: error
 
    Exception raised when a string passed to one of the functions here is not a

From python-checkins at python.org  Sat Jul 10 16:27:36 2010
From: python-checkins at python.org (r.david.murray)
Date: Sat, 10 Jul 2010 16:27:36 +0200 (CEST)
Subject: [Python-checkins] r82774 - in python/branches/release26-maint:
	Doc/library/re.rst
Message-ID: <20100710142736.0CFD0EE981@mail.python.org>

Author: r.david.murray
Date: Sat Jul 10 16:27:35 2010
New Revision: 82774

Log:
Merged revisions 82773 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

................
  r82773 | r.david.murray | 2010-07-10 10:25:18 -0400 (Sat, 10 Jul 2010) | 9 lines
  
  Merged revisions 82771 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k
  
  ........
    r82771 | r.david.murray | 2010-07-10 10:23:36 -0400 (Sat, 10 Jul 2010) | 2 lines
    
    Add missing docs for re module 'purge' function.
  ........
................


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Doc/library/re.rst

Modified: python/branches/release26-maint/Doc/library/re.rst
==============================================================================
--- python/branches/release26-maint/Doc/library/re.rst	(original)
+++ python/branches/release26-maint/Doc/library/re.rst	Sat Jul 10 16:27:35 2010
@@ -660,6 +660,11 @@
    metacharacters in it.
 
 
+.. function:: purge()
+
+   Clear the regular expression cache.
+
+
 .. exception:: error
 
    Exception raised when a string passed to one of the functions here is not a

From python-checkins at python.org  Sat Jul 10 16:34:33 2010
From: python-checkins at python.org (michael.foord)
Date: Sat, 10 Jul 2010 16:34:33 +0200 (CEST)
Subject: [Python-checkins] r82775 - python/branches/release27-maint
Message-ID: <20100710143433.DBA47EE981@mail.python.org>

Author: michael.foord
Date: Sat Jul 10 16:34:33 2010
New Revision: 82775

Log:
Blocked revisions 82767 via svnmerge

........
  r82767 | michael.foord | 2010-07-10 14:52:22 +0100 (Sat, 10 Jul 2010) | 2 lines
  
  Fix error message for comparing single line strings in unittest.TestCase.assertEqual.
  Issue 9174.
........


Modified:
   python/branches/release27-maint/   (props changed)

From python-checkins at python.org  Sat Jul 10 16:42:30 2010
From: python-checkins at python.org (michael.foord)
Date: Sat, 10 Jul 2010 16:42:30 +0200 (CEST)
Subject: [Python-checkins] r82776 - python/branches/release31-maint
Message-ID: <20100710144230.B97EDEE981@mail.python.org>

Author: michael.foord
Date: Sat Jul 10 16:42:30 2010
New Revision: 82776

Log:
Blocked revisions 82767 via svnmerge

........
  r82767 | michael.foord | 2010-07-10 14:52:22 +0100 (Sat, 10 Jul 2010) | 2 lines
  
  Fix error message for comparing single line strings in unittest.TestCase.assertEqual.
  Issue 9174.
........


Modified:
   python/branches/release31-maint/   (props changed)

From python-checkins at python.org  Sat Jul 10 17:14:45 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sat, 10 Jul 2010 17:14:45 +0200 (CEST)
Subject: [Python-checkins] r82777 -
	python/branches/py3k/Lib/test/test_struct.py
Message-ID: <20100710151445.9A5ABEE9C7@mail.python.org>

Author: benjamin.peterson
Date: Sat Jul 10 17:14:45 2010
New Revision: 82777

Log:
ValueError is eventually what we want to move to, I suppose

Modified:
   python/branches/py3k/Lib/test/test_struct.py

Modified: python/branches/py3k/Lib/test/test_struct.py
==============================================================================
--- python/branches/py3k/Lib/test/test_struct.py	(original)
+++ python/branches/py3k/Lib/test/test_struct.py	Sat Jul 10 17:14:45 2010
@@ -234,8 +234,8 @@
                                                                  b'\x01' + got)
                 else:
                     # x is out of range -- verify pack realizes that.
-                    self.assertRaises((OverflowError, struct.error), pack,
-                                      format, x)
+                    self.assertRaises((OverflowError, ValueError, struct.error),
+                                      pack, format, x)
 
             def run(self):
                 from random import randrange

From python-checkins at python.org  Sat Jul 10 17:17:09 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sat, 10 Jul 2010 17:17:09 +0200 (CEST)
Subject: [Python-checkins] r82778 - in python/branches/release27-maint:
	Lib/test/test_struct.py
Message-ID: <20100710151709.14A34EE9C5@mail.python.org>

Author: benjamin.peterson
Date: Sat Jul 10 17:17:08 2010
New Revision: 82778

Log:
Merged revisions 82777 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82777 | benjamin.peterson | 2010-07-10 10:14:45 -0500 (Sat, 10 Jul 2010) | 1 line
  
  ValueError is eventually what we want to move to, I suppose
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/test/test_struct.py

Modified: python/branches/release27-maint/Lib/test/test_struct.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_struct.py	(original)
+++ python/branches/release27-maint/Lib/test/test_struct.py	Sat Jul 10 17:17:08 2010
@@ -243,8 +243,8 @@
                                                                  '\x01' + got)
                 else:
                     # x is out of range -- verify pack realizes that.
-                    self.assertRaises((OverflowError, struct.error), pack,
-                                      format, x)
+                    self.assertRaises((OverflowError, ValueError, struct.error),
+                                      pack, format, x)
 
             def run(self):
                 from random import randrange

From python-checkins at python.org  Sat Jul 10 21:45:08 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sat, 10 Jul 2010 21:45:08 +0200 (CEST)
Subject: [Python-checkins] r82779 - sandbox/trunk/2to3/lib2to3/refactor.py
Message-ID: <20100710194508.81EAAEE9D3@mail.python.org>

Author: benjamin.peterson
Date: Sat Jul 10 21:45:08 2010
New Revision: 82779

Log:
typo in attribute name #9217

Modified:
   sandbox/trunk/2to3/lib2to3/refactor.py

Modified: sandbox/trunk/2to3/lib2to3/refactor.py
==============================================================================
--- sandbox/trunk/2to3/lib2to3/refactor.py	(original)
+++ sandbox/trunk/2to3/lib2to3/refactor.py	Sat Jul 10 21:45:08 2010
@@ -517,7 +517,7 @@
         try:
             tree = self.parse_block(block, lineno, indent)
         except Exception, err:
-            if self.log.isEnabledFor(logging.DEBUG):
+            if self.logger.isEnabledFor(logging.DEBUG):
                 for line in block:
                     self.log_debug("Source: %s", line.rstrip(u"\n"))
             self.log_error("Can't parse docstring in %s line %s: %s: %s",

From solipsis at pitrou.net  Sun Jul 11 01:24:03 2010
From: solipsis at pitrou.net (solipsis at pitrou.net)
Date: Sun, 11 Jul 2010 01:24:03 +0200 (CEST)
Subject: [Python-checkins] Daily py3k reference leaks (r82777): sum=0
Message-ID: <20100710232403.6BF8F17721@ns6635.ovh.net>

py3k results for svn r82777 (hg cset 704f6d4cc57b)
--------------------------------------------------



Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/antoine/py3k/refleaks/reflogVIEOMR', '-x']

From python-checkins at python.org  Sun Jul 11 05:12:43 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Sun, 11 Jul 2010 05:12:43 +0200 (CEST)
Subject: [Python-checkins] r82780 - in python/branches/py3k/Lib:
	test/test_urllib2.py urllib/request.py
Message-ID: <20100711031243.5D3BEEE98A@mail.python.org>

Author: senthil.kumaran
Date: Sun Jul 11 05:12:43 2010
New Revision: 82780

Log:
Stricter verification for file based url scheme and reliance on ftp protocol.



Modified:
   python/branches/py3k/Lib/test/test_urllib2.py
   python/branches/py3k/Lib/urllib/request.py

Modified: python/branches/py3k/Lib/test/test_urllib2.py
==============================================================================
--- python/branches/py3k/Lib/test/test_urllib2.py	(original)
+++ python/branches/py3k/Lib/test/test_urllib2.py	Sun Jul 11 05:12:43 2010
@@ -731,6 +731,8 @@
             ("file://ftp.example.com///foo.txt", False),
 # XXXX bug: fails with OSError, should be URLError
             ("file://ftp.example.com/foo.txt", False),
+            ("file://somehost//foo/something.txt", True),
+            ("file://localhost//foo/something.txt", False),
             ]:
             req = Request(url)
             try:
@@ -741,6 +743,7 @@
             else:
                 self.assertTrue(o.req is req)
                 self.assertEqual(req.type, "ftp")
+            self.assertEqual(req.type is "ftp", ftp)
 
     def test_http(self):
 

Modified: python/branches/py3k/Lib/urllib/request.py
==============================================================================
--- python/branches/py3k/Lib/urllib/request.py	(original)
+++ python/branches/py3k/Lib/urllib/request.py	Sun Jul 11 05:12:43 2010
@@ -1188,7 +1188,8 @@
     # Use local file or FTP depending on form of URL
     def file_open(self, req):
         url = req.selector
-        if url[:2] == '//' and url[2:3] != '/':
+        if url[:2] == '//' and url[2:3] != '/' and (req.host and
+                req.host != 'localhost'):
             req.type = 'ftp'
             return self.parent.open(req)
         else:

From python-checkins at python.org  Sun Jul 11 05:15:25 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Sun, 11 Jul 2010 05:15:25 +0200 (CEST)
Subject: [Python-checkins] r82781 - in python/branches/release31-maint:
	Lib/test/test_urllib2.py Lib/urllib/request.py
Message-ID: <20100711031525.CCFA0EE98A@mail.python.org>

Author: senthil.kumaran
Date: Sun Jul 11 05:15:25 2010
New Revision: 82781

Log:
Merged revisions 82780 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82780 | senthil.kumaran | 2010-07-11 08:42:43 +0530 (Sun, 11 Jul 2010) | 3 lines
  
  Stricter verification for file based url scheme and reliance on ftp protocol.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/test/test_urllib2.py
   python/branches/release31-maint/Lib/urllib/request.py

Modified: python/branches/release31-maint/Lib/test/test_urllib2.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_urllib2.py	(original)
+++ python/branches/release31-maint/Lib/test/test_urllib2.py	Sun Jul 11 05:15:25 2010
@@ -736,6 +736,8 @@
             ("file://ftp.example.com///foo.txt", False),
 # XXXX bug: fails with OSError, should be URLError
             ("file://ftp.example.com/foo.txt", False),
+            ("file://somehost//foo/something.txt", True),
+            ("file://localhost//foo/something.txt", False),
             ]:
             req = Request(url)
             try:
@@ -746,6 +748,7 @@
             else:
                 self.assertTrue(o.req is req)
                 self.assertEqual(req.type, "ftp")
+            self.assertEqual(req.type is "ftp", ftp)
 
     def test_http(self):
 

Modified: python/branches/release31-maint/Lib/urllib/request.py
==============================================================================
--- python/branches/release31-maint/Lib/urllib/request.py	(original)
+++ python/branches/release31-maint/Lib/urllib/request.py	Sun Jul 11 05:15:25 2010
@@ -1188,7 +1188,8 @@
     # Use local file or FTP depending on form of URL
     def file_open(self, req):
         url = req.selector
-        if url[:2] == '//' and url[2:3] != '/':
+        if url[:2] == '//' and url[2:3] != '/' and (req.host and
+                req.host != 'localhost'):
             req.type = 'ftp'
             return self.parent.open(req)
         else:

From python-checkins at python.org  Sun Jul 11 05:18:51 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Sun, 11 Jul 2010 05:18:51 +0200 (CEST)
Subject: [Python-checkins] r82782 - in python/branches/release27-maint:
	Lib/test/test_urllib2.py Lib/urllib2.py
Message-ID: <20100711031851.6AED3EE98A@mail.python.org>

Author: senthil.kumaran
Date: Sun Jul 11 05:18:51 2010
New Revision: 82782

Log:
Merged revisions 82780 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82780 | senthil.kumaran | 2010-07-11 08:42:43 +0530 (Sun, 11 Jul 2010) | 3 lines
  
  Stricter verification for file based url scheme and reliance on ftp protocol.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/test/test_urllib2.py
   python/branches/release27-maint/Lib/urllib2.py

Modified: python/branches/release27-maint/Lib/test/test_urllib2.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_urllib2.py	(original)
+++ python/branches/release27-maint/Lib/test/test_urllib2.py	Sun Jul 11 05:18:51 2010
@@ -726,6 +726,8 @@
             ("file://ftp.example.com///foo.txt", False),
 # XXXX bug: fails with OSError, should be URLError
             ("file://ftp.example.com/foo.txt", False),
+            ("file://somehost//foo/something.txt", True),
+            ("file://localhost//foo/something.txt", False),
             ]:
             req = Request(url)
             try:
@@ -736,6 +738,7 @@
             else:
                 self.assertTrue(o.req is req)
                 self.assertEqual(req.type, "ftp")
+            self.assertEqual(req.type is "ftp", ftp)
 
     def test_http(self):
 

Modified: python/branches/release27-maint/Lib/urllib2.py
==============================================================================
--- python/branches/release27-maint/Lib/urllib2.py	(original)
+++ python/branches/release27-maint/Lib/urllib2.py	Sun Jul 11 05:18:51 2010
@@ -1262,7 +1262,8 @@
     # Use local file or FTP depending on form of URL
     def file_open(self, req):
         url = req.get_selector()
-        if url[:2] == '//' and url[2:3] != '/':
+        if url[:2] == '//' and url[2:3] != '/' and (req.host and
+                req.host != 'localhost'):
             req.type = 'ftp'
             return self.parent.open(req)
         else:

From python-checkins at python.org  Sun Jul 11 05:33:38 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Sun, 11 Jul 2010 05:33:38 +0200 (CEST)
Subject: [Python-checkins] r82783 - in python/branches/release26-maint:
	Lib/test/test_urllib2.py Lib/urllib2.py
Message-ID: <20100711033338.A2A3AEE981@mail.python.org>

Author: senthil.kumaran
Date: Sun Jul 11 05:33:38 2010
New Revision: 82783

Log:
Merged revisions 82782 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

................
  r82782 | senthil.kumaran | 2010-07-11 08:48:51 +0530 (Sun, 11 Jul 2010) | 9 lines
  
  Merged revisions 82780 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k
  
  ........
    r82780 | senthil.kumaran | 2010-07-11 08:42:43 +0530 (Sun, 11 Jul 2010) | 3 lines
    
    Stricter verification for file based url scheme and reliance on ftp protocol.
  ........
................


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Lib/test/test_urllib2.py
   python/branches/release26-maint/Lib/urllib2.py

Modified: python/branches/release26-maint/Lib/test/test_urllib2.py
==============================================================================
--- python/branches/release26-maint/Lib/test/test_urllib2.py	(original)
+++ python/branches/release26-maint/Lib/test/test_urllib2.py	Sun Jul 11 05:33:38 2010
@@ -728,6 +728,8 @@
             ("file://ftp.example.com///foo.txt", False),
 # XXXX bug: fails with OSError, should be URLError
             ("file://ftp.example.com/foo.txt", False),
+            ("file://somehost//foo/something.txt", True),
+            ("file://localhost//foo/something.txt", False),
             ]:
             req = Request(url)
             try:
@@ -738,6 +740,7 @@
             else:
                 self.assert_(o.req is req)
                 self.assertEqual(req.type, "ftp")
+            self.assertEqual(req.type is "ftp", ftp)
 
     def test_http(self):
 

Modified: python/branches/release26-maint/Lib/urllib2.py
==============================================================================
--- python/branches/release26-maint/Lib/urllib2.py	(original)
+++ python/branches/release26-maint/Lib/urllib2.py	Sun Jul 11 05:33:38 2010
@@ -1259,7 +1259,8 @@
     # Use local file or FTP depending on form of URL
     def file_open(self, req):
         url = req.get_selector()
-        if url[:2] == '//' and url[2:3] != '/':
+        if url[:2] == '//' and url[2:3] != '/' and (req.host and
+                req.host != 'localhost'):
             req.type = 'ftp'
             return self.parent.open(req)
         else:

From python-checkins at python.org  Sun Jul 11 05:36:35 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sun, 11 Jul 2010 05:36:35 +0200 (CEST)
Subject: [Python-checkins] r82784 -
	python/branches/py3k/Objects/stringlib/formatter.h
Message-ID: <20100711033635.83D95EE981@mail.python.org>

Author: benjamin.peterson
Date: Sun Jul 11 05:36:35 2010
New Revision: 82784

Log:
remove unneeded error check

Modified:
   python/branches/py3k/Objects/stringlib/formatter.h

Modified: python/branches/py3k/Objects/stringlib/formatter.h
==============================================================================
--- python/branches/py3k/Objects/stringlib/formatter.h	(original)
+++ python/branches/py3k/Objects/stringlib/formatter.h	Sun Jul 11 05:36:35 2010
@@ -776,14 +776,6 @@
             goto done;
         }
 
-        /* Error to specify a comma. */
-        if (format->thousands_separators) {
-            PyErr_SetString(PyExc_ValueError,
-                            "Thousands separators not allowed with integer"
-                            " format specifier 'c'");
-            goto done;
-        }
-
         /* taken from unicodeobject.c formatchar() */
         /* Integer input truncated to a character */
 /* XXX: won't work for int */

From python-checkins at python.org  Sun Jul 11 07:01:52 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Sun, 11 Jul 2010 07:01:52 +0200 (CEST)
Subject: [Python-checkins] r82785 -
	python/branches/py3k/Doc/library/http.cookiejar.rst
Message-ID: <20100711050152.84184EE9D3@mail.python.org>

Author: senthil.kumaran
Date: Sun Jul 11 07:01:52 2010
New Revision: 82785

Log:
Issue2007: Removing reference to MSIE Cookie handling by mechanize. Suggested by John J Lee.



Modified:
   python/branches/py3k/Doc/library/http.cookiejar.rst

Modified: python/branches/py3k/Doc/library/http.cookiejar.rst
==============================================================================
--- python/branches/py3k/Doc/library/http.cookiejar.rst	(original)
+++ python/branches/py3k/Doc/library/http.cookiejar.rst	Sun Jul 11 07:01:52 2010
@@ -108,10 +108,6 @@
       :mod:`http.cookiejar` and :mod:`http.cookies` modules do not depend on each
       other.
 
-   http://wwwsearch.sourceforge.net/mechanize/
-      Extensions to this module, including a class for reading Microsoft Internet
-      Explorer cookies on Windows.
-
    http://wp.netscape.com/newsref/std/cookie_spec.html
       The specification of the original Netscape cookie protocol.  Though this is
       still the dominant protocol, the 'Netscape cookie protocol' implemented by all
@@ -292,11 +288,8 @@
 FileCookieJar subclasses and co-operation with web browsers
 -----------------------------------------------------------
 
-The following :class:`CookieJar` subclasses are provided for reading and writing
-.  Further :class:`CookieJar` subclasses, including one that reads Microsoft
-Internet Explorer cookies, are available at
-http://wwwsearch.sourceforge.net/mechanize/.
-
+The following :class:`CookieJar` subclasses are provided for reading and
+writing .
 
 .. class:: MozillaCookieJar(filename, delayload=None, policy=None)
 

From python-checkins at python.org  Sun Jul 11 07:06:37 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Sun, 11 Jul 2010 07:06:37 +0200 (CEST)
Subject: [Python-checkins] r82786 - in python/branches/release31-maint:
	Doc/library/http.cookiejar.rst
Message-ID: <20100711050637.6C49DEEA4F@mail.python.org>

Author: senthil.kumaran
Date: Sun Jul 11 07:06:37 2010
New Revision: 82786

Log:
Merged revisions 82785 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82785 | senthil.kumaran | 2010-07-11 10:31:52 +0530 (Sun, 11 Jul 2010) | 3 lines
  
  Issue2007: Removing reference to MSIE Cookie handling by mechanize. Suggested by John J Lee.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Doc/library/http.cookiejar.rst

Modified: python/branches/release31-maint/Doc/library/http.cookiejar.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/http.cookiejar.rst	(original)
+++ python/branches/release31-maint/Doc/library/http.cookiejar.rst	Sun Jul 11 07:06:37 2010
@@ -108,10 +108,6 @@
       :mod:`http.cookiejar` and :mod:`http.cookies` modules do not depend on each
       other.
 
-   http://wwwsearch.sourceforge.net/mechanize/
-      Extensions to this module, including a class for reading Microsoft Internet
-      Explorer cookies on Windows.
-
    http://wp.netscape.com/newsref/std/cookie_spec.html
       The specification of the original Netscape cookie protocol.  Though this is
       still the dominant protocol, the 'Netscape cookie protocol' implemented by all
@@ -292,11 +288,8 @@
 FileCookieJar subclasses and co-operation with web browsers
 -----------------------------------------------------------
 
-The following :class:`CookieJar` subclasses are provided for reading and writing
-.  Further :class:`CookieJar` subclasses, including one that reads Microsoft
-Internet Explorer cookies, are available at
-http://wwwsearch.sourceforge.net/mechanize/.
-
+The following :class:`CookieJar` subclasses are provided for reading and
+writing .
 
 .. class:: MozillaCookieJar(filename, delayload=None, policy=None)
 

From python-checkins at python.org  Sun Jul 11 07:11:10 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Sun, 11 Jul 2010 07:11:10 +0200 (CEST)
Subject: [Python-checkins] r82787 - in python/branches/release27-maint:
	Doc/library/cookielib.rst
Message-ID: <20100711051110.D0E6EEEA4F@mail.python.org>

Author: senthil.kumaran
Date: Sun Jul 11 07:11:10 2010
New Revision: 82787

Log:
Merged revisions 82785 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82785 | senthil.kumaran | 2010-07-11 10:31:52 +0530 (Sun, 11 Jul 2010) | 3 lines
  
  Issue2007: Removing reference to MSIE Cookie handling by mechanize. Suggested by John J Lee.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Doc/library/cookielib.rst

Modified: python/branches/release27-maint/Doc/library/cookielib.rst
==============================================================================
--- python/branches/release27-maint/Doc/library/cookielib.rst	(original)
+++ python/branches/release27-maint/Doc/library/cookielib.rst	Sun Jul 11 07:11:10 2010
@@ -121,10 +121,6 @@
       HTTP cookie classes, principally useful for server-side code.  The
       :mod:`cookielib` and :mod:`Cookie` modules do not depend on each other.
 
-   http://wwwsearch.sourceforge.net/mechanize/
-      Extensions to this module, including a class for reading Microsoft Internet
-      Explorer cookies on Windows.
-
    http://wp.netscape.com/newsref/std/cookie_spec.html
       The specification of the original Netscape cookie protocol.  Though this is
       still the dominant protocol, the 'Netscape cookie protocol' implemented by all
@@ -310,10 +306,8 @@
 FileCookieJar subclasses and co-operation with web browsers
 -----------------------------------------------------------
 
-The following :class:`CookieJar` subclasses are provided for reading and writing
-.  Further :class:`CookieJar` subclasses, including one that reads Microsoft
-Internet Explorer cookies, are available at
-http://wwwsearch.sourceforge.net/mechanize/ .
+The following :class:`CookieJar` subclasses are provided for reading and
+writing .
 
 .. class:: MozillaCookieJar(filename, delayload=None, policy=None)
 

From python-checkins at python.org  Sun Jul 11 07:21:35 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Sun, 11 Jul 2010 07:21:35 +0200 (CEST)
Subject: [Python-checkins] r82788 - in python/branches/release26-maint:
	Doc/library/cookielib.rst
Message-ID: <20100711052135.60EE6EE98A@mail.python.org>

Author: senthil.kumaran
Date: Sun Jul 11 07:21:35 2010
New Revision: 82788

Log:
Merged revisions 82787 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

................
  r82787 | senthil.kumaran | 2010-07-11 10:41:10 +0530 (Sun, 11 Jul 2010) | 9 lines
  
  Merged revisions 82785 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k
  
  ........
    r82785 | senthil.kumaran | 2010-07-11 10:31:52 +0530 (Sun, 11 Jul 2010) | 3 lines
    
    Issue2007: Removing reference to MSIE Cookie handling by mechanize. Suggested by John J Lee.
  ........
................


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Doc/library/cookielib.rst

Modified: python/branches/release26-maint/Doc/library/cookielib.rst
==============================================================================
--- python/branches/release26-maint/Doc/library/cookielib.rst	(original)
+++ python/branches/release26-maint/Doc/library/cookielib.rst	Sun Jul 11 07:21:35 2010
@@ -121,10 +121,6 @@
       HTTP cookie classes, principally useful for server-side code.  The
       :mod:`cookielib` and :mod:`Cookie` modules do not depend on each other.
 
-   http://wwwsearch.sourceforge.net/mechanize/
-      Extensions to this module, including a class for reading Microsoft Internet
-      Explorer cookies on Windows.
-
    http://wp.netscape.com/newsref/std/cookie_spec.html
       The specification of the original Netscape cookie protocol.  Though this is
       still the dominant protocol, the 'Netscape cookie protocol' implemented by all
@@ -310,10 +306,8 @@
 FileCookieJar subclasses and co-operation with web browsers
 -----------------------------------------------------------
 
-The following :class:`CookieJar` subclasses are provided for reading and writing
-.  Further :class:`CookieJar` subclasses, including one that reads Microsoft
-Internet Explorer cookies, are available at
-http://wwwsearch.sourceforge.net/mechanize/ .
+The following :class:`CookieJar` subclasses are provided for reading and
+writing .
 
 .. class:: MozillaCookieJar(filename, delayload=None, policy=None)
 

From python-checkins at python.org  Sun Jul 11 10:33:16 2010
From: python-checkins at python.org (georg.brandl)
Date: Sun, 11 Jul 2010 10:33:16 +0200 (CEST)
Subject: [Python-checkins] r82789 - python/branches/py3k/Doc/Makefile
Message-ID: <20100711083316.E305FEEACE@mail.python.org>

Author: georg.brandl
Date: Sun Jul 11 10:33:16 2010
New Revision: 82789

Log:
Silence makeindex.

Modified:
   python/branches/py3k/Doc/Makefile

Modified: python/branches/py3k/Doc/Makefile
==============================================================================
--- python/branches/py3k/Doc/Makefile	(original)
+++ python/branches/py3k/Doc/Makefile	Sun Jul 11 10:33:16 2010
@@ -144,6 +144,7 @@
 	# archive the A4 latex
 	-rm -r build/latex
 	make latex PAPER=a4
+	-sed -i 's/makeindex/makeindex -q/' build/latex/Makefile
 	(cd build/latex; make clean && make all-pdf && make FMT=pdf zip bz2)
 	cp build/latex/docs-pdf.zip dist/python-$(DISTVERSION)-docs-pdf-a4.zip
 	cp build/latex/docs-pdf.tar.bz2 dist/python-$(DISTVERSION)-docs-pdf-a4.tar.bz2
@@ -151,6 +152,7 @@
 	# archive the letter latex
 	rm -r build/latex
 	make latex PAPER=letter
+	-sed -i 's/makeindex/makeindex -q/' build/latex/Makefile
 	(cd build/latex; make clean && make all-pdf && make FMT=pdf zip bz2)
 	cp build/latex/docs-pdf.zip dist/python-$(DISTVERSION)-docs-pdf-letter.zip
 	cp build/latex/docs-pdf.tar.bz2 dist/python-$(DISTVERSION)-docs-pdf-letter.tar.bz2

From python-checkins at python.org  Sun Jul 11 10:36:20 2010
From: python-checkins at python.org (georg.brandl)
Date: Sun, 11 Jul 2010 10:36:20 +0200 (CEST)
Subject: [Python-checkins] r82790 - python/branches/py3k/Doc/glossary.rst
Message-ID: <20100711083620.414EEEEA50@mail.python.org>

Author: georg.brandl
Date: Sun Jul 11 10:36:20 2010
New Revision: 82790

Log:
#3214 followup: add link to ABC entry.

Modified:
   python/branches/py3k/Doc/glossary.rst

Modified: python/branches/py3k/Doc/glossary.rst
==============================================================================
--- python/branches/py3k/Doc/glossary.rst	(original)
+++ python/branches/py3k/Doc/glossary.rst	Sun Jul 11 10:36:20 2010
@@ -160,9 +160,9 @@
       must be a duck.")  By emphasizing interfaces rather than specific types,
       well-designed code improves its flexibility by allowing polymorphic
       substitution.  Duck-typing avoids tests using :func:`type` or
-      :func:`isinstance`. (Note, however, that duck-typing can be complemented
-      with abstract base classes.) Instead, it typically employs :func:`hasattr`
-      tests or :term:`EAFP` programming.
+      :func:`isinstance`.  (Note, however, that duck-typing can be complemented
+      with :term:`abstract base class`\ es.)  Instead, it typically employs
+      :func:`hasattr` tests or :term:`EAFP` programming.
 
    EAFP
       Easier to ask for forgiveness than permission.  This common Python coding

From python-checkins at python.org  Sun Jul 11 10:52:52 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Sun, 11 Jul 2010 10:52:52 +0200 (CEST)
Subject: [Python-checkins] r82791 - in python/branches/release27-maint:
	Lib/distutils/util.py Lib/sysconfig.py Misc/NEWS
Message-ID: <20100711085252.8E120EEADA@mail.python.org>

Author: ronald.oussoren
Date: Sun Jul 11 10:52:52 2010
New Revision: 82791

Log:
Fix for issue #9164: with this patch sysconfig and distuls don't break
when duplicate '-arch foo' flags end up in CFLAGS (which may happen when
building a universal build using macports)


Modified:
   python/branches/release27-maint/Lib/distutils/util.py
   python/branches/release27-maint/Lib/sysconfig.py
   python/branches/release27-maint/Misc/NEWS

Modified: python/branches/release27-maint/Lib/distutils/util.py
==============================================================================
--- python/branches/release27-maint/Lib/distutils/util.py	(original)
+++ python/branches/release27-maint/Lib/distutils/util.py	Sun Jul 11 10:52:52 2010
@@ -144,8 +144,7 @@
                 cflags = get_config_vars().get('CFLAGS')
 
                 archs = re.findall('-arch\s+(\S+)', cflags)
-                archs.sort()
-                archs = tuple(archs)
+                archs = tuple(sorted(set(archs)))
 
                 if len(archs) == 1:
                     machine = archs[0]

Modified: python/branches/release27-maint/Lib/sysconfig.py
==============================================================================
--- python/branches/release27-maint/Lib/sysconfig.py	(original)
+++ python/branches/release27-maint/Lib/sysconfig.py	Sun Jul 11 10:52:52 2010
@@ -645,8 +645,7 @@
                 cflags = get_config_vars().get('CFLAGS')
 
                 archs = re.findall('-arch\s+(\S+)', cflags)
-                archs.sort()
-                archs = tuple(archs)
+                archs = tuple(sorted(set(archs)))
 
                 if len(archs) == 1:
                     machine = archs[0]

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Sun Jul 11 10:52:52 2010
@@ -33,6 +33,8 @@
 
 - Issue #9128: Fix validation of class decorators in parser module.
 
+- Issue #9164: Ensure sysconfig handles dupblice archs while building on OSX
+
 Extension Modules
 -----------------
 

From python-checkins at python.org  Sun Jul 11 10:56:18 2010
From: python-checkins at python.org (georg.brandl)
Date: Sun, 11 Jul 2010 10:56:18 +0200 (CEST)
Subject: [Python-checkins] r82792 - in python/branches/release31-maint:
	Doc/glossary.rst
Message-ID: <20100711085618.26241EEAC9@mail.python.org>

Author: georg.brandl
Date: Sun Jul 11 10:56:18 2010
New Revision: 82792

Log:
Merged revisions 82790 via svnmerge from 
svn+ssh://svn.python.org/python/branches/py3k

........
  r82790 | georg.brandl | 2010-07-11 10:36:20 +0200 (So, 11 Jul 2010) | 1 line
  
  #3214 followup: add link to ABC entry.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Doc/glossary.rst

Modified: python/branches/release31-maint/Doc/glossary.rst
==============================================================================
--- python/branches/release31-maint/Doc/glossary.rst	(original)
+++ python/branches/release31-maint/Doc/glossary.rst	Sun Jul 11 10:56:18 2010
@@ -154,9 +154,9 @@
       must be a duck.")  By emphasizing interfaces rather than specific types,
       well-designed code improves its flexibility by allowing polymorphic
       substitution.  Duck-typing avoids tests using :func:`type` or
-      :func:`isinstance`. (Note, however, that duck-typing can be complemented
-      with abstract base classes.) Instead, it typically employs :func:`hasattr`
-      tests or :term:`EAFP` programming.
+      :func:`isinstance`.  (Note, however, that duck-typing can be complemented
+      with :term:`abstract base class`\ es.)  Instead, it typically employs
+      :func:`hasattr` tests or :term:`EAFP` programming.
 
    EAFP
       Easier to ask for forgiveness than permission.  This common Python coding

From python-checkins at python.org  Sun Jul 11 10:56:18 2010
From: python-checkins at python.org (georg.brandl)
Date: Sun, 11 Jul 2010 10:56:18 +0200 (CEST)
Subject: [Python-checkins] r82793 - in python/branches/release27-maint:
	Doc/glossary.rst
Message-ID: <20100711085618.4E721EEAD3@mail.python.org>

Author: georg.brandl
Date: Sun Jul 11 10:56:18 2010
New Revision: 82793

Log:
Merged revisions 82790 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82790 | georg.brandl | 2010-07-11 10:36:20 +0200 (So, 11 Jul 2010) | 1 line
  
  #3214 followup: add link to ABC entry.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Doc/glossary.rst

Modified: python/branches/release27-maint/Doc/glossary.rst
==============================================================================
--- python/branches/release27-maint/Doc/glossary.rst	(original)
+++ python/branches/release27-maint/Doc/glossary.rst	Sun Jul 11 10:56:18 2010
@@ -163,9 +163,9 @@
       must be a duck.")  By emphasizing interfaces rather than specific types,
       well-designed code improves its flexibility by allowing polymorphic
       substitution.  Duck-typing avoids tests using :func:`type` or
-      :func:`isinstance`. (Note, however, that duck-typing can be complemented
-      with abstract base classes.) Instead, it typically employs :func:`hasattr`
-      tests or :term:`EAFP` programming.
+      :func:`isinstance`.  (Note, however, that duck-typing can be complemented
+      with :term:`abstract base class`\ es.)  Instead, it typically employs
+      :func:`hasattr` tests or :term:`EAFP` programming.
 
    EAFP
       Easier to ask for forgiveness than permission.  This common Python coding

From python-checkins at python.org  Sun Jul 11 10:57:05 2010
From: python-checkins at python.org (georg.brandl)
Date: Sun, 11 Jul 2010 10:57:05 +0200 (CEST)
Subject: [Python-checkins] r82794 - in python/branches/release27-maint:
	Doc/Makefile
Message-ID: <20100711085705.5EF19D28E@mail.python.org>

Author: georg.brandl
Date: Sun Jul 11 10:57:05 2010
New Revision: 82794

Log:
Merged revisions 82789 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82789 | georg.brandl | 2010-07-11 10:33:16 +0200 (So, 11 Jul 2010) | 1 line
  
  Silence makeindex.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Doc/Makefile

Modified: python/branches/release27-maint/Doc/Makefile
==============================================================================
--- python/branches/release27-maint/Doc/Makefile	(original)
+++ python/branches/release27-maint/Doc/Makefile	Sun Jul 11 10:57:05 2010
@@ -144,6 +144,7 @@
 	# archive the A4 latex
 	-rm -r build/latex
 	make latex PAPER=a4
+	-sed -i 's/makeindex/makeindex -q/' build/latex/Makefile
 	(cd build/latex; make clean && make all-pdf && make FMT=pdf zip bz2)
 	cp build/latex/docs-pdf.zip dist/python-$(DISTVERSION)-docs-pdf-a4.zip
 	cp build/latex/docs-pdf.tar.bz2 dist/python-$(DISTVERSION)-docs-pdf-a4.tar.bz2
@@ -151,6 +152,7 @@
 	# archive the letter latex
 	rm -r build/latex
 	make latex PAPER=letter
+	-sed -i 's/makeindex/makeindex -q/' build/latex/Makefile
 	(cd build/latex; make clean && make all-pdf && make FMT=pdf zip bz2)
 	cp build/latex/docs-pdf.zip dist/python-$(DISTVERSION)-docs-pdf-letter.zip
 	cp build/latex/docs-pdf.tar.bz2 dist/python-$(DISTVERSION)-docs-pdf-letter.tar.bz2

From python-checkins at python.org  Sun Jul 11 10:57:06 2010
From: python-checkins at python.org (georg.brandl)
Date: Sun, 11 Jul 2010 10:57:06 +0200 (CEST)
Subject: [Python-checkins] r82795 - in python/branches/release31-maint:
	Doc/Makefile
Message-ID: <20100711085706.DF81FEEA83@mail.python.org>

Author: georg.brandl
Date: Sun Jul 11 10:57:06 2010
New Revision: 82795

Log:
Merged revisions 82789 via svnmerge from 
svn+ssh://svn.python.org/python/branches/py3k

........
  r82789 | georg.brandl | 2010-07-11 10:33:16 +0200 (So, 11 Jul 2010) | 1 line
  
  Silence makeindex.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Doc/Makefile

Modified: python/branches/release31-maint/Doc/Makefile
==============================================================================
--- python/branches/release31-maint/Doc/Makefile	(original)
+++ python/branches/release31-maint/Doc/Makefile	Sun Jul 11 10:57:06 2010
@@ -143,6 +143,7 @@
 	# archive the A4 latex
 	-rm -r build/latex
 	make latex PAPER=a4
+	-sed -i 's/makeindex/makeindex -q/' build/latex/Makefile
 	(cd build/latex; make clean && make all-pdf && make FMT=pdf zip bz2)
 	cp build/latex/docs-pdf.zip dist/python-$(DISTVERSION)-docs-pdf-a4.zip
 	cp build/latex/docs-pdf.tar.bz2 dist/python-$(DISTVERSION)-docs-pdf-a4.tar.bz2
@@ -150,6 +151,7 @@
 	# archive the letter latex
 	rm -r build/latex
 	make latex PAPER=letter
+	-sed -i 's/makeindex/makeindex -q/' build/latex/Makefile
 	(cd build/latex; make clean && make all-pdf && make FMT=pdf zip bz2)
 	cp build/latex/docs-pdf.zip dist/python-$(DISTVERSION)-docs-pdf-letter.zip
 	cp build/latex/docs-pdf.tar.bz2 dist/python-$(DISTVERSION)-docs-pdf-letter.tar.bz2

From python-checkins at python.org  Sun Jul 11 11:05:07 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Sun, 11 Jul 2010 11:05:07 +0200 (CEST)
Subject: [Python-checkins] r82796 - in python/branches/py3k:
	Lib/sysconfig.py Misc/NEWS
Message-ID: <20100711090507.A16BFEEAD2@mail.python.org>

Author: ronald.oussoren
Date: Sun Jul 11 11:05:07 2010
New Revision: 82796

Log:
Fix for issue 9164


Modified:
   python/branches/py3k/Lib/sysconfig.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/sysconfig.py
==============================================================================
--- python/branches/py3k/Lib/sysconfig.py	(original)
+++ python/branches/py3k/Lib/sysconfig.py	Sun Jul 11 11:05:07 2010
@@ -660,8 +660,7 @@
                 cflags = get_config_vars().get('CFLAGS')
 
                 archs = re.findall('-arch\s+(\S+)', cflags)
-                archs.sort()
-                archs = tuple(archs)
+                archs = tuple(sorted(set(archs)))
 
                 if len(archs) == 1:
                     machine = archs[0]

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sun Jul 11 11:05:07 2010
@@ -470,6 +470,8 @@
 Library
 -------
 
+- Issue #9164: Ensure that sysconfig handles duplicate -arch flags in CFLAGS
+
 - Issue #7646: The fnmatch pattern cache no longer grows without bound.
 
 - Issue #9136: Fix 'dictionary changed size during iteration'

From python-checkins at python.org  Sun Jul 11 11:08:11 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Sun, 11 Jul 2010 11:08:11 +0200 (CEST)
Subject: [Python-checkins] r82797 - in python/branches/release26-maint:
	Lib/distutils/util.py Misc/NEWS
Message-ID: <20100711090811.A7856EE98A@mail.python.org>

Author: ronald.oussoren
Date: Sun Jul 11 11:08:11 2010
New Revision: 82797

Log:
Merged revisions 82791 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

........
  r82791 | ronald.oussoren | 2010-07-11 10:52:52 +0200 (Sun, 11 Jul 2010) | 4 lines
  
  Fix for issue #9164: with this patch sysconfig and distuls don't break
  when duplicate '-arch foo' flags end up in CFLAGS (which may happen when
  building a universal build using macports)
........


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Lib/distutils/util.py
   python/branches/release26-maint/Misc/NEWS

Modified: python/branches/release26-maint/Lib/distutils/util.py
==============================================================================
--- python/branches/release26-maint/Lib/distutils/util.py	(original)
+++ python/branches/release26-maint/Lib/distutils/util.py	Sun Jul 11 11:08:11 2010
@@ -144,8 +144,7 @@
                 cflags = get_config_vars().get('CFLAGS')
 
                 archs = re.findall('-arch\s+(\S+)', cflags)
-                archs.sort()
-                archs = tuple(archs)
+                archs = tuple(sorted(set(archs)))
 
                 if len(archs) == 1:
                     machine = archs[0]

Modified: python/branches/release26-maint/Misc/NEWS
==============================================================================
--- python/branches/release26-maint/Misc/NEWS	(original)
+++ python/branches/release26-maint/Misc/NEWS	Sun Jul 11 11:08:11 2010
@@ -81,6 +81,8 @@
 Library
 -------
 
+- Issue #9164: Ensure sysconfig handles dupblice archs while building on OSX
+
 - Issue #7646: The fnmatch pattern cache no longer grows without bound.
 
 - Issue #9136: Fix 'dictionary changed size during iteration'

From python-checkins at python.org  Sun Jul 11 11:23:12 2010
From: python-checkins at python.org (georg.brandl)
Date: Sun, 11 Jul 2010 11:23:12 +0200 (CEST)
Subject: [Python-checkins] r82798 -
	python/branches/py3k/Doc/library/socket.rst
Message-ID: <20100711092312.11D83C9DF@mail.python.org>

Author: georg.brandl
Date: Sun Jul 11 11:23:11 2010
New Revision: 82798

Log:
#6774: explain shutdown() behavior varying with platform.

Modified:
   python/branches/py3k/Doc/library/socket.rst

Modified: python/branches/py3k/Doc/library/socket.rst
==============================================================================
--- python/branches/py3k/Doc/library/socket.rst	(original)
+++ python/branches/py3k/Doc/library/socket.rst	Sun Jul 11 11:23:11 2010
@@ -748,7 +748,9 @@
    Shut down one or both halves of the connection.  If *how* is :const:`SHUT_RD`,
    further receives are disallowed.  If *how* is :const:`SHUT_WR`, further sends
    are disallowed.  If *how* is :const:`SHUT_RDWR`, further sends and receives are
-   disallowed.
+   disallowed.  Depending on the platform, shutting down one half of the connection
+   can also close the opposite half (e.g. on Mac OS X, ``shutdown(SHUT_WR)`` does
+   not allow further reads on the other end of the connection).
 
 Note that there are no methods :meth:`read` or :meth:`write`; use
 :meth:`~socket.recv` and :meth:`~socket.send` without *flags* argument instead.

From python-checkins at python.org  Sun Jul 11 11:26:57 2010
From: python-checkins at python.org (georg.brandl)
Date: Sun, 11 Jul 2010 11:26:57 +0200 (CEST)
Subject: [Python-checkins] r82799 - python/branches/py3k/Doc/whatsnew/2.7.rst
Message-ID: <20100711092657.D2439EEAB1@mail.python.org>

Author: georg.brandl
Date: Sun Jul 11 11:26:57 2010
New Revision: 82799

Log:
Fix typo.

Modified:
   python/branches/py3k/Doc/whatsnew/2.7.rst

Modified: python/branches/py3k/Doc/whatsnew/2.7.rst
==============================================================================
--- python/branches/py3k/Doc/whatsnew/2.7.rst	(original)
+++ python/branches/py3k/Doc/whatsnew/2.7.rst	Sun Jul 11 11:26:57 2010
@@ -146,7 +146,7 @@
 * Float-to-string and string-to-float conversions now round their
   results more correctly, and :func:`repr` of a floating-point
   number *x* returns a result that's guaranteed to round back to the
-  same number when converted back to a string.
+  same number when converted back to a float.
 * The :ctype:`PyCapsule` type, used to provide a C API for extension modules.
 * The :cfunc:`PyLong_AsLongAndOverflow` C API function.
 

From python-checkins at python.org  Sun Jul 11 11:29:09 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Sun, 11 Jul 2010 11:29:09 +0200 (CEST)
Subject: [Python-checkins] r82800 -
	python/branches/release31-maint/Lib/distutils/util.py
Message-ID: <20100711092909.40C8EEEAB8@mail.python.org>

Author: ronald.oussoren
Date: Sun Jul 11 11:29:09 2010
New Revision: 82800

Log:
Fix for issue 9164


Modified:
   python/branches/release31-maint/Lib/distutils/util.py

Modified: python/branches/release31-maint/Lib/distutils/util.py
==============================================================================
--- python/branches/release31-maint/Lib/distutils/util.py	(original)
+++ python/branches/release31-maint/Lib/distutils/util.py	Sun Jul 11 11:29:09 2010
@@ -143,8 +143,7 @@
                 cflags = get_config_vars().get('CFLAGS')
 
                 archs = re.findall('-arch\s+(\S+)', cflags)
-                archs.sort()
-                archs = tuple(archs)
+                archs = tuple(sorted(set(archs)))
 
                 if len(archs) == 1:
                     machine = archs[0]

From python-checkins at python.org  Sun Jul 11 11:33:39 2010
From: python-checkins at python.org (georg.brandl)
Date: Sun, 11 Jul 2010 11:33:39 +0200 (CEST)
Subject: [Python-checkins] r82801 - in python/branches/py3k:
	Doc/library/functions.rst Modules/_io/_iomodule.c
Message-ID: <20100711093339.7435FEEAC5@mail.python.org>

Author: georg.brandl
Date: Sun Jul 11 11:33:39 2010
New Revision: 82801

Log:
#9184: fix default value for "buffering" param of open().

Modified:
   python/branches/py3k/Doc/library/functions.rst
   python/branches/py3k/Modules/_io/_iomodule.c

Modified: python/branches/py3k/Doc/library/functions.rst
==============================================================================
--- python/branches/py3k/Doc/library/functions.rst	(original)
+++ python/branches/py3k/Doc/library/functions.rst	Sun Jul 11 11:33:39 2010
@@ -676,7 +676,7 @@
    :meth:`__index__` method that returns an integer.
 
 
-.. function:: open(file, mode='r', buffering=None, encoding=None, errors=None, newline=None, closefd=True)
+.. function:: open(file, mode='r', buffering=-1, encoding=None, errors=None, newline=None, closefd=True)
 
    Open *file* and return a corresponding stream.  If the file cannot be opened,
    an :exc:`IOError` is raised.

Modified: python/branches/py3k/Modules/_io/_iomodule.c
==============================================================================
--- python/branches/py3k/Modules/_io/_iomodule.c	(original)
+++ python/branches/py3k/Modules/_io/_iomodule.c	Sun Jul 11 11:33:39 2010
@@ -176,7 +176,7 @@
  * The main open() function
  */
 PyDoc_STRVAR(open_doc,
-"open(file, mode='r', buffering=None, encoding=None,\n"
+"open(file, mode='r', buffering=-1, encoding=None,\n"
 "     errors=None, newline=None, closefd=True) -> file object\n"
 "\n"
 "Open file and return a stream.  Raise IOError upon failure.\n"

From python-checkins at python.org  Sun Jul 11 11:34:16 2010
From: python-checkins at python.org (georg.brandl)
Date: Sun, 11 Jul 2010 11:34:16 +0200 (CEST)
Subject: [Python-checkins] r82802 - in python/branches/release27-maint:
	Doc/whatsnew/2.7.rst
Message-ID: <20100711093416.7FA53C9D2@mail.python.org>

Author: georg.brandl
Date: Sun Jul 11 11:34:16 2010
New Revision: 82802

Log:
Merged revisions 82799 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82799 | georg.brandl | 2010-07-11 11:26:57 +0200 (So, 11 Jul 2010) | 1 line
  
  Fix typo.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Doc/whatsnew/2.7.rst

Modified: python/branches/release27-maint/Doc/whatsnew/2.7.rst
==============================================================================
--- python/branches/release27-maint/Doc/whatsnew/2.7.rst	(original)
+++ python/branches/release27-maint/Doc/whatsnew/2.7.rst	Sun Jul 11 11:34:16 2010
@@ -144,7 +144,7 @@
 * Float-to-string and string-to-float conversions now round their
   results more correctly, and :func:`repr` of a floating-point
   number *x* returns a result that's guaranteed to round back to the
-  same number when converted back to a string.
+  same number when converted back to a float.
 * The :ctype:`PyCapsule` type, used to provide a C API for extension modules.
 * The :cfunc:`PyLong_AsLongAndOverflow` C API function.
 

From python-checkins at python.org  Sun Jul 11 11:36:31 2010
From: python-checkins at python.org (georg.brandl)
Date: Sun, 11 Jul 2010 11:36:31 +0200 (CEST)
Subject: [Python-checkins] r82803 - python/branches/release27-maint
Message-ID: <20100711093631.BC5BEEEA4F@mail.python.org>

Author: georg.brandl
Date: Sun Jul 11 11:36:31 2010
New Revision: 82803

Log:
Blocked revisions 82801 via svnmerge

........
  r82801 | georg.brandl | 2010-07-11 11:33:39 +0200 (So, 11 Jul 2010) | 1 line
  
  #9184: fix default value for "buffering" param of open().
........


Modified:
   python/branches/release27-maint/   (props changed)

From python-checkins at python.org  Sun Jul 11 11:41:21 2010
From: python-checkins at python.org (georg.brandl)
Date: Sun, 11 Jul 2010 11:41:21 +0200 (CEST)
Subject: [Python-checkins] r82804 - in python/branches/py3k:
	Doc/library/ast.rst Lib/ast.py Misc/NEWS
Message-ID: <20100711094121.8E4C5EEA4F@mail.python.org>

Author: georg.brandl
Date: Sun Jul 11 11:41:21 2010
New Revision: 82804

Log:
Allow set literals in literal_eval().

Modified:
   python/branches/py3k/Doc/library/ast.rst
   python/branches/py3k/Lib/ast.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Doc/library/ast.rst
==============================================================================
--- python/branches/py3k/Doc/library/ast.rst	(original)
+++ python/branches/py3k/Doc/library/ast.rst	Sun Jul 11 11:41:21 2010
@@ -119,12 +119,15 @@
 
    Safely evaluate an expression node or a string containing a Python
    expression.  The string or node provided may only consist of the following
-   Python literal structures: strings, numbers, tuples, lists, dicts, booleans,
-   and ``None``.
+   Python literal structures: strings, numbers, tuples, lists, dicts, sets,
+   booleans, and ``None``.
 
    This can be used for safely evaluating strings containing Python expressions
    from untrusted sources without the need to parse the values oneself.
 
+   .. versionchanged:: 3.2
+      Now allows set literals.
+
 
 .. function:: get_docstring(node, clean=True)
 

Modified: python/branches/py3k/Lib/ast.py
==============================================================================
--- python/branches/py3k/Lib/ast.py	(original)
+++ python/branches/py3k/Lib/ast.py	Sun Jul 11 11:41:21 2010
@@ -58,6 +58,8 @@
             return tuple(map(_convert, node.elts))
         elif isinstance(node, List):
             return list(map(_convert, node.elts))
+        elif isinstance(node, Set):
+            return set(map(_convert, node.elts))
         elif isinstance(node, Dict):
             return dict((_convert(k), _convert(v)) for k, v
                         in zip(node.keys, node.values))

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sun Jul 11 11:41:21 2010
@@ -470,7 +470,9 @@
 Library
 -------
 
-- Issue #9164: Ensure that sysconfig handles duplicate -arch flags in CFLAGS
+- ``ast.literal_eval()`` now allows set literals.
+
+- Issue #9164: Ensure that sysconfig handles duplicate -arch flags in CFLAGS.
 
 - Issue #7646: The fnmatch pattern cache no longer grows without bound.
 

From python-checkins at python.org  Sun Jul 11 11:42:10 2010
From: python-checkins at python.org (georg.brandl)
Date: Sun, 11 Jul 2010 11:42:10 +0200 (CEST)
Subject: [Python-checkins] r82805 -
	python/branches/py3k/Doc/library/functions.rst
Message-ID: <20100711094210.3A222EEA4F@mail.python.org>

Author: georg.brandl
Date: Sun Jul 11 11:42:10 2010
New Revision: 82805

Log:
#7935: cross-reference to ast.literal_eval() from eval() docs.

Modified:
   python/branches/py3k/Doc/library/functions.rst

Modified: python/branches/py3k/Doc/library/functions.rst
==============================================================================
--- python/branches/py3k/Doc/library/functions.rst	(original)
+++ python/branches/py3k/Doc/library/functions.rst	Sun Jul 11 11:42:10 2010
@@ -335,6 +335,9 @@
    returns the current global and local dictionary, respectively, which may be
    useful to pass around for use by :func:`eval` or :func:`exec`.
 
+   See :func:`ast.literal_eval` for a function that can safely evaluate strings
+   with expressions containing only literals.
+
 
 .. function:: exec(object[, globals[, locals]])
 

From python-checkins at python.org  Sun Jul 11 12:22:44 2010
From: python-checkins at python.org (georg.brandl)
Date: Sun, 11 Jul 2010 12:22:44 +0200 (CEST)
Subject: [Python-checkins] r82806 - in python/branches/py3k/Doc/distutils:
	apiref.rst extending.rst
Message-ID: <20100711102244.738CEEEA83@mail.python.org>

Author: georg.brandl
Date: Sun Jul 11 12:22:44 2010
New Revision: 82806

Log:
#9223: link to Command class reference, and move Command interface docs nearer to class docs.

Modified:
   python/branches/py3k/Doc/distutils/apiref.rst
   python/branches/py3k/Doc/distutils/extending.rst

Modified: python/branches/py3k/Doc/distutils/apiref.rst
==============================================================================
--- python/branches/py3k/Doc/distutils/apiref.rst	(original)
+++ python/branches/py3k/Doc/distutils/apiref.rst	Sun Jul 11 12:22:44 2010
@@ -147,11 +147,11 @@
 In addition, the :mod:`distutils.core` module exposed a number of  classes that
 live elsewhere.
 
-* :class:`Extension` from :mod:`distutils.extension`
+* :class:`~distutils.extension.Extension` from :mod:`distutils.extension`
 
-* :class:`Command` from :mod:`distutils.cmd`
+* :class:`~distutils.cmd.Command` from :mod:`distutils.cmd`
 
-* :class:`Distribution` from :mod:`distutils.dist`
+* :class:`~distutils.dist.Distribution` from :mod:`distutils.dist`
 
 A short description of each of these follows, but see the relevant module for
 the full reference.
@@ -1679,8 +1679,8 @@
 ===================================================================
 
 .. module:: distutils.cmd
-   :synopsis: This module provides the abstract base class Command. This class is subclassed
-              by the modules in the distutils.command  subpackage.
+   :synopsis: This module provides the abstract base class Command. This class
+              is subclassed by the modules in the distutils.command subpackage.
 
 
 This module supplies the abstract base class :class:`Command`.
@@ -1690,20 +1690,84 @@
 
    Abstract base class for defining command classes, the "worker bees" of the
    Distutils.  A useful analogy for command classes is to think of them as
-   subroutines with local variables called *options*.  The options are declared in
-   :meth:`initialize_options` and defined (given their final values) in
-   :meth:`finalize_options`, both of which must be defined by every command class.
-   The distinction between the two is necessary because option values might come
-   from the outside world (command line, config file, ...), and any options
-   dependent on other options must be computed after these outside influences have
-   been processed --- hence :meth:`finalize_options`.  The body of the subroutine,
-   where it does all its work based on the values of its options, is the
-   :meth:`run` method, which must also be implemented by every command class.
+   subroutines with local variables called *options*.  The options are declared
+   in :meth:`initialize_options` and defined (given their final values) in
+   :meth:`finalize_options`, both of which must be defined by every command
+   class.  The distinction between the two is necessary because option values
+   might come from the outside world (command line, config file, ...), and any
+   options dependent on other options must be computed after these outside
+   influences have been processed --- hence :meth:`finalize_options`.  The body
+   of the subroutine, where it does all its work based on the values of its
+   options, is the :meth:`run` method, which must also be implemented by every
+   command class.
 
-   The class constructor takes a single argument *dist*, a  :class:`Distribution`
+   The class constructor takes a single argument *dist*, a :class:`Distribution`
    instance.
 
 
+Creating a new Distutils command
+================================
+
+This section outlines the steps to create a new Distutils command.
+
+A new command lives in a module in the :mod:`distutils.command` package. There
+is a sample template in that directory called :file:`command_template`.  Copy
+this file to a new module with the same name as the new command you're
+implementing.  This module should implement a class with the same name as the
+module (and the command).  So, for instance, to create the command
+``peel_banana`` (so that users can run ``setup.py peel_banana``), you'd copy
+:file:`command_template` to :file:`distutils/command/peel_banana.py`, then edit
+it so that it's implementing the class :class:`peel_banana`, a subclass of
+:class:`distutils.cmd.Command`.
+
+Subclasses of :class:`Command` must define the following methods.
+
+.. method:: Command.initialize_options()
+
+   Set default values for all the options that this command supports.  Note that
+   these defaults may be overridden by other commands, by the setup script, by
+   config files, or by the command-line.  Thus, this is not the place to code
+   dependencies between options; generally, :meth:`initialize_options`
+   implementations are just a bunch of ``self.foo = None`` assignments.
+
+
+.. method:: Command.finalize_options()
+
+   Set final values for all the options that this command supports. This is
+   always called as late as possible, ie.  after any option assignments from the
+   command-line or from other commands have been done.  Thus, this is the place
+   to to code option dependencies: if *foo* depends on *bar*, then it is safe to
+   set *foo* from *bar* as long as *foo* still has the same value it was
+   assigned in :meth:`initialize_options`.
+
+
+.. method:: Command.run()
+
+   A command's raison d'etre: carry out the action it exists to perform, controlled
+   by the options initialized in :meth:`initialize_options`, customized by other
+   commands, the setup script, the command-line, and config files, and finalized in
+   :meth:`finalize_options`.  All terminal output and filesystem interaction should
+   be done by :meth:`run`.
+
+
+.. attribute:: Command.sub_commands
+
+   *sub_commands* formalizes the notion of a "family" of commands,
+   e.g. ``install`` as the parent with sub-commands ``install_lib``,
+   ``install_headers``, etc.  The parent of a family of commands defines
+   *sub_commands* as a class attribute; it's a list of 2-tuples ``(command_name,
+   predicate)``, with *command_name* a string and *predicate* a function, a
+   string or ``None``.  *predicate* is a method of the parent command that
+   determines whether the corresponding command is applicable in the current
+   situation.  (E.g. we ``install_headers`` is only applicable if we have any C
+   header files to install.)  If *predicate* is ``None``, that command is always
+   applicable.
+
+   *sub_commands* is usually defined at the *end* of a class, because
+   predicates can be methods of the class, so they must already have been
+   defined.  The canonical example is the :command:`install` command.
+
+
 :mod:`distutils.command` --- Individual Distutils commands
 ==========================================================
 
@@ -1942,63 +2006,3 @@
 This is described in more detail in :pep:`301`.
 
 .. % todo
-
-
-Creating a new Distutils command
-================================
-
-This section outlines the steps to create a new Distutils command.
-
-A new command lives in a module in the :mod:`distutils.command` package. There
-is a sample template in that directory called  :file:`command_template`. Copy
-this file to a new module with the same name as the new command you're
-implementing. This module should implement a class with the same name as the
-module (and the command). So, for instance, to create the command
-``peel_banana`` (so that users can run ``setup.py peel_banana``), you'd copy
-:file:`command_template`  to :file:`distutils/command/peel_banana.py`, then edit
-it so that it's implementing the class :class:`peel_banana`, a subclass of
-:class:`distutils.cmd.Command`.
-
-Subclasses of :class:`Command` must define the following methods.
-
-
-.. method:: Command.initialize_options()
-
-   Set default values for all the options that this command supports.  Note that
-   these defaults may be overridden by other commands, by the setup script, by
-   config files, or by the command-line.  Thus, this is not the place to code
-   dependencies between options; generally, :meth:`initialize_options`
-   implementations are just a bunch of ``self.foo = None`` assignments.
-
-
-.. method:: Command.finalize_options()
-
-   Set final values for all the options that this command supports. This is
-   always called as late as possible, ie.  after any option assignments from the
-   command-line or from other commands have been done.  Thus, this is the place
-   to to code option dependencies: if *foo* depends on *bar*, then it is safe to
-   set *foo* from *bar* as long as *foo* still has the same value it was
-   assigned in :meth:`initialize_options`.
-
-
-.. method:: Command.run()
-
-   A command's raison d'etre: carry out the action it exists to perform, controlled
-   by the options initialized in :meth:`initialize_options`, customized by other
-   commands, the setup script, the command-line, and config files, and finalized in
-   :meth:`finalize_options`.  All terminal output and filesystem interaction should
-   be done by :meth:`run`.
-
-*sub_commands* formalizes the notion of a "family" of commands, eg. ``install``
-as the parent with sub-commands ``install_lib``, ``install_headers``, etc.  The
-parent of a family of commands defines *sub_commands* as a class attribute; it's
-a list of 2-tuples ``(command_name, predicate)``, with *command_name* a string
-and *predicate* a function, a string or None. *predicate* is a method of
-the parent command that determines whether the corresponding command is
-applicable in the current situation.  (Eg. we ``install_headers`` is only
-applicable if we have any C header files to install.)  If *predicate* is None,
-that command is always applicable.
-
-*sub_commands* is usually defined at the \*end\* of a class, because predicates
-can be methods of the class, so they must already have been defined.  The
-canonical example is the :command:`install` command.

Modified: python/branches/py3k/Doc/distutils/extending.rst
==============================================================================
--- python/branches/py3k/Doc/distutils/extending.rst	(original)
+++ python/branches/py3k/Doc/distutils/extending.rst	Sun Jul 11 12:22:44 2010
@@ -15,8 +15,8 @@
 should be copied into packages in addition to :file:`.py` files as a
 convenience.
 
-Most distutils command implementations are subclasses of the :class:`Command`
-class from :mod:`distutils.cmd`.  New commands may directly inherit from
+Most distutils command implementations are subclasses of the
+:class:`distutils.cmd.Command` class.  New commands may directly inherit from
 :class:`Command`, while replacements often derive from :class:`Command`
 indirectly, directly subclassing the command they are replacing.  Commands are
 required to derive from :class:`Command`.

From python-checkins at python.org  Sun Jul 11 12:29:37 2010
From: python-checkins at python.org (georg.brandl)
Date: Sun, 11 Jul 2010 12:29:37 +0200 (CEST)
Subject: [Python-checkins] r82807 - in python/branches/release27-maint:
	Doc/distutils/extending.rst
Message-ID: <20100711102937.6EC24EE9D3@mail.python.org>

Author: georg.brandl
Date: Sun Jul 11 12:29:37 2010
New Revision: 82807

Log:
Merged revisions 82806 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82806 | georg.brandl | 2010-07-11 12:22:44 +0200 (So, 11 Jul 2010) | 1 line
  
  #9223: link to Command class reference, and move Command interface docs nearer to class docs.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Doc/distutils/extending.rst

Modified: python/branches/release27-maint/Doc/distutils/extending.rst
==============================================================================
--- python/branches/release27-maint/Doc/distutils/extending.rst	(original)
+++ python/branches/release27-maint/Doc/distutils/extending.rst	Sun Jul 11 12:29:37 2010
@@ -15,8 +15,8 @@
 should be copied into packages in addition to :file:`.py` files as a
 convenience.
 
-Most distutils command implementations are subclasses of the :class:`Command`
-class from :mod:`distutils.cmd`.  New commands may directly inherit from
+Most distutils command implementations are subclasses of the
+:class:`distutils.cmd.Command` class.  New commands may directly inherit from
 :class:`Command`, while replacements often derive from :class:`Command`
 indirectly, directly subclassing the command they are replacing.  Commands are
 required to derive from :class:`Command`.

From python-checkins at python.org  Sun Jul 11 12:41:08 2010
From: python-checkins at python.org (georg.brandl)
Date: Sun, 11 Jul 2010 12:41:08 +0200 (CEST)
Subject: [Python-checkins] r82808 - in python/branches/release31-maint:
	Doc/README.txt Doc/c-api/init.rst Doc/distutils/builtdist.rst
	Doc/documenting/building.rst Doc/documenting/index.rst
	Doc/documenting/markup.rst Doc/documenting/style.rst
	Doc/faq/gui.rst Doc/howto/functional.rst
	Doc/howto/unicode.rst Doc/library/calendar.rst
	Doc/library/logging.rst Doc/library/os.rst Doc/library/re.rst
	Doc/library/socket.rst Doc/library/stdtypes.rst
	Doc/library/string.rst Doc/library/unittest.rst
	Doc/reference/datamodel.rst Doc/reference/introduction.rst
	Doc/tutorial/classes.rst Doc/tutorial/introduction.rst
	Doc/using/mac.rst Doc/whatsnew/2.7.rst Lib/urllib/parse.py
	Modules/datetimemodule.c
Message-ID: <20100711104108.218DEEE9D3@mail.python.org>

Author: georg.brandl
Date: Sun Jul 11 12:41:07 2010
New Revision: 82808

Log:
Merged revisions 82301 via svnmerge from 
svn+ssh://svn.python.org/python/branches/py3k

................
  r82301 | benjamin.peterson | 2010-06-28 00:32:30 +0200 (Mo, 28 Jun 2010) | 303 lines
  
  Merged revisions 80605-80609,80642-80646,80651-80652,80674,80684-80686,80748,80852,80854,80870,80872-80873,80907,80915-80916,80951-80952,80976-80977,80985,81038-81040,81042,81053,81070,81104-81105,81114,81125,81245,81285,81402,81463,81516,81562-81563,81567,81593,81635,81680-81681,81684,81801,81888,81931-81933,81939-81942,81963,81984,81991,82120,82188,82264-82267 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r80605 | andrew.kuchling | 2010-04-28 19:22:16 -0500 (Wed, 28 Apr 2010) | 1 line
    
    Add various items
  ........
    r80606 | andrew.kuchling | 2010-04-28 20:44:30 -0500 (Wed, 28 Apr 2010) | 6 lines
    
    Fix doubled 'the'.
    Markup fixes to use :exc:, :option: in a few places.
      (Glitch: unittest.main's -c ends up a link to the Python
      interpreter's -c option.  Should we skip using :option: for that
      switch, or disable the auto-linking somehow?)
  ........
    r80607 | andrew.kuchling | 2010-04-28 20:45:41 -0500 (Wed, 28 Apr 2010) | 1 line
    
    Add various unittest items
  ........
    r80608 | benjamin.peterson | 2010-04-28 22:18:05 -0500 (Wed, 28 Apr 2010) | 1 line
    
    update pypy description
  ........
    r80609 | benjamin.peterson | 2010-04-28 22:30:59 -0500 (Wed, 28 Apr 2010) | 1 line
    
    update pypy url
  ........
    r80642 | andrew.kuchling | 2010-04-29 19:49:09 -0500 (Thu, 29 Apr 2010) | 1 line
    
    Always add space after RFC; reword paragraph
  ........
    r80643 | andrew.kuchling | 2010-04-29 19:52:31 -0500 (Thu, 29 Apr 2010) | 6 lines
    
    Reword paragraph to make its meaning clearer.
    
    Antoine Pitrou: is my version of the paragraph still correct?
    
    R. David Murray: is this more understandable than the previous version?
  ........
    r80644 | andrew.kuchling | 2010-04-29 20:02:15 -0500 (Thu, 29 Apr 2010) | 1 line
    
    Fix typos
  ........
    r80645 | andrew.kuchling | 2010-04-29 20:32:47 -0500 (Thu, 29 Apr 2010) | 1 line
    
    Markup fix; clarify by adding 'in that order'
  ........
    r80646 | andrew.kuchling | 2010-04-29 20:33:40 -0500 (Thu, 29 Apr 2010) | 1 line
    
    Add various items; rearrange unittest section a bit
  ........
    r80651 | andrew.kuchling | 2010-04-30 08:46:55 -0500 (Fri, 30 Apr 2010) | 1 line
    
    Minor grammar re-wording
  ........
    r80652 | andrew.kuchling | 2010-04-30 08:47:34 -0500 (Fri, 30 Apr 2010) | 1 line
    
    Add item
  ........
    r80674 | andrew.kuchling | 2010-04-30 20:19:16 -0500 (Fri, 30 Apr 2010) | 1 line
    
    Add various items
  ........
    r80684 | andrew.kuchling | 2010-05-01 07:05:52 -0500 (Sat, 01 May 2010) | 1 line
    
    Minor grammar fix
  ........
    r80685 | andrew.kuchling | 2010-05-01 07:06:51 -0500 (Sat, 01 May 2010) | 1 line
    
    Describe memoryview
  ........
    r80686 | antoine.pitrou | 2010-05-01 07:16:39 -0500 (Sat, 01 May 2010) | 4 lines
    
    Fix attribution. Travis didn't do much and he did a bad work.
    (yes, this is a sensitive subject, sorry)
  ........
    r80748 | andrew.kuchling | 2010-05-03 20:24:22 -0500 (Mon, 03 May 2010) | 1 line
    
    Add some more items; the urlparse change is added twice
  ........
    r80852 | andrew.kuchling | 2010-05-05 20:09:47 -0500 (Wed, 05 May 2010) | 1 line
    
    Reword paragraph; fix filename, which should be pyconfig.h
  ........
    r80854 | andrew.kuchling | 2010-05-05 20:10:56 -0500 (Wed, 05 May 2010) | 1 line
    
    Add various items
  ........
    r80870 | andrew.kuchling | 2010-05-06 09:14:09 -0500 (Thu, 06 May 2010) | 1 line
    
    Describe ElementTree 1.3; rearrange new-module sections; describe dict views as sets; small edits and items
  ........
    r80872 | andrew.kuchling | 2010-05-06 12:21:59 -0500 (Thu, 06 May 2010) | 1 line
    
    Add 2 items; record ideas for two initial sections; clarify wording
  ........
    r80873 | andrew.kuchling | 2010-05-06 12:27:57 -0500 (Thu, 06 May 2010) | 1 line
    
    Change section title; point to unittest2
  ........
    r80907 | andrew.kuchling | 2010-05-06 20:45:14 -0500 (Thu, 06 May 2010) | 1 line
    
    Add a new section on the development plan; add an item
  ........
    r80915 | antoine.pitrou | 2010-05-07 05:15:51 -0500 (Fri, 07 May 2010) | 3 lines
    
    Fix some markup and a class name. Also, wrap a long line.
  ........
    r80916 | andrew.kuchling | 2010-05-07 06:30:47 -0500 (Fri, 07 May 2010) | 1 line
    
    Re-word text
  ........
    r80951 | andrew.kuchling | 2010-05-07 20:15:26 -0500 (Fri, 07 May 2010) | 1 line
    
    Add two items
  ........
    r80952 | andrew.kuchling | 2010-05-07 20:35:55 -0500 (Fri, 07 May 2010) | 1 line
    
    Get accents correct
  ........
    r80976 | andrew.kuchling | 2010-05-08 08:28:03 -0500 (Sat, 08 May 2010) | 1 line
    
    Add logging.dictConfig example; give up on writing a Ttk example
  ........
    r80977 | andrew.kuchling | 2010-05-08 08:29:46 -0500 (Sat, 08 May 2010) | 1 line
    
    Markup fixes
  ........
    r80985 | andrew.kuchling | 2010-05-08 10:39:46 -0500 (Sat, 08 May 2010) | 7 lines
    
    Write summary of the 2.7 release; rewrite the future section some more;
    mention PYTHONWARNINGS env. var; tweak some examples for readability.
    
    And with this commit, the "What's New" is done... except for a
    complete read-through to polish the text, and fixing any reported errors,
    but those tasks can easily wait until after beta2.
  ........
    r81038 | benjamin.peterson | 2010-05-09 16:09:40 -0500 (Sun, 09 May 2010) | 1 line
    
    finish clause
  ........
    r81039 | andrew.kuchling | 2010-05-10 09:18:27 -0500 (Mon, 10 May 2010) | 1 line
    
    Markup fix; re-word a sentence
  ........
    r81040 | andrew.kuchling | 2010-05-10 09:20:12 -0500 (Mon, 10 May 2010) | 1 line
    
    Use title case
  ........
    r81042 | andrew.kuchling | 2010-05-10 10:03:35 -0500 (Mon, 10 May 2010) | 1 line
    
    Link to unittest2 article
  ........
    r81053 | florent.xicluna | 2010-05-10 14:59:22 -0500 (Mon, 10 May 2010) | 2 lines
    
    Add a link on maketrans().
  ........
    r81070 | andrew.kuchling | 2010-05-10 18:13:41 -0500 (Mon, 10 May 2010) | 1 line
    
    Fix typo
  ........
    r81104 | andrew.kuchling | 2010-05-11 19:38:44 -0500 (Tue, 11 May 2010) | 1 line
    
    Revision pass: lots of edits, typo fixes, rearrangements
  ........
    r81105 | andrew.kuchling | 2010-05-11 19:40:47 -0500 (Tue, 11 May 2010) | 1 line
    
    Let's call this done
  ........
    r81114 | andrew.kuchling | 2010-05-12 08:56:07 -0500 (Wed, 12 May 2010) | 1 line
    
    Grammar fix
  ........
    r81125 | andrew.kuchling | 2010-05-12 13:56:48 -0500 (Wed, 12 May 2010) | 1 line
    
    #8696: add documentation for logging.config.dictConfig (PEP 391)
  ........
    r81245 | andrew.kuchling | 2010-05-16 18:31:16 -0500 (Sun, 16 May 2010) | 1 line
    
    Add cross-reference to later section
  ........
    r81285 | vinay.sajip | 2010-05-18 03:16:27 -0500 (Tue, 18 May 2010) | 1 line
    
    Fixed minor typo in ReST markup.
  ........
    r81402 | vinay.sajip | 2010-05-21 12:41:34 -0500 (Fri, 21 May 2010) | 1 line
    
    Updated logging documentation with more dictConfig information.
  ........
    r81463 | georg.brandl | 2010-05-22 03:17:23 -0500 (Sat, 22 May 2010) | 1 line
    
    #8785: less confusing description of regex.find*.
  ........
    r81516 | andrew.kuchling | 2010-05-25 08:34:08 -0500 (Tue, 25 May 2010) | 1 line
    
    Add three items
  ........
    r81562 | andrew.kuchling | 2010-05-27 08:22:53 -0500 (Thu, 27 May 2010) | 1 line
    
    Rewrite wxWidgets section
  ........
    r81563 | andrew.kuchling | 2010-05-27 08:30:09 -0500 (Thu, 27 May 2010) | 1 line
    
    Remove top-level 'General Questions' section, pushing up the questions it contains
  ........
    r81567 | andrew.kuchling | 2010-05-27 16:29:59 -0500 (Thu, 27 May 2010) | 1 line
    
    Add item
  ........
    r81593 | georg.brandl | 2010-05-29 03:46:18 -0500 (Sat, 29 May 2010) | 1 line
    
    #8616: add new turtle demo "nim".
  ........
    r81635 | georg.brandl | 2010-06-01 02:25:23 -0500 (Tue, 01 Jun 2010) | 1 line
    
    Put docs for RegexObject.search() before RegexObject.match() to mirror re.search() and re.match() order.
  ........
    r81680 | vinay.sajip | 2010-06-03 17:34:42 -0500 (Thu, 03 Jun 2010) | 1 line
    
    Issue #8890: Documentation changed to avoid reference to temporary files.
  ........
    r81681 | sean.reifschneider | 2010-06-03 20:51:26 -0500 (Thu, 03 Jun 2010) | 2 lines
    
    Issue8810: Clearing up docstring for tzinfo.utcoffset.
  ........
    r81684 | vinay.sajip | 2010-06-04 08:41:02 -0500 (Fri, 04 Jun 2010) | 1 line
    
    Issue #8890: Documentation changed to avoid reference to temporary files - other cases covered.
  ........
    r81801 | andrew.kuchling | 2010-06-07 08:38:40 -0500 (Mon, 07 Jun 2010) | 1 line
    
    #8875: Remove duplicated paragraph
  ........
    r81888 | andrew.kuchling | 2010-06-10 20:54:58 -0500 (Thu, 10 Jun 2010) | 1 line
    
    Add a few more items
  ........
    r81931 | georg.brandl | 2010-06-12 01:26:54 -0500 (Sat, 12 Jun 2010) | 1 line
    
    Fix punctuation.
  ........
    r81932 | georg.brandl | 2010-06-12 01:28:58 -0500 (Sat, 12 Jun 2010) | 1 line
    
    Document that an existing directory raises in mkdir().
  ........
    r81933 | georg.brandl | 2010-06-12 01:45:33 -0500 (Sat, 12 Jun 2010) | 1 line
    
    Update version in README.
  ........
    r81939 | georg.brandl | 2010-06-12 04:45:01 -0500 (Sat, 12 Jun 2010) | 1 line
    
    Use newer toctree syntax.
  ........
    r81940 | georg.brandl | 2010-06-12 04:45:28 -0500 (Sat, 12 Jun 2010) | 1 line
    
    Add document on how to build.
  ........
    r81941 | georg.brandl | 2010-06-12 04:45:58 -0500 (Sat, 12 Jun 2010) | 1 line
    
    Fix gratuitous indentation.
  ........
    r81942 | georg.brandl | 2010-06-12 04:46:03 -0500 (Sat, 12 Jun 2010) | 1 line
    
    Update README.
  ........
    r81963 | andrew.kuchling | 2010-06-12 15:00:55 -0500 (Sat, 12 Jun 2010) | 1 line
    
    Grammar fix
  ........
    r81984 | georg.brandl | 2010-06-14 10:58:39 -0500 (Mon, 14 Jun 2010) | 1 line
    
    #8993: fix reference.
  ........
    r81991 | andrew.kuchling | 2010-06-14 19:38:58 -0500 (Mon, 14 Jun 2010) | 1 line
    
    Add another bunch of items
  ........
    r82120 | andrew.kuchling | 2010-06-20 16:45:45 -0500 (Sun, 20 Jun 2010) | 1 line
    
    Note that Python 3.x isn't covered; add forward ref. for UTF-8; note error in 2.5 and up
  ........
    r82188 | benjamin.peterson | 2010-06-23 19:02:46 -0500 (Wed, 23 Jun 2010) | 1 line
    
    remove reverted changed
  ........
    r82264 | georg.brandl | 2010-06-27 05:47:47 -0500 (Sun, 27 Jun 2010) | 1 line
    
    Confusing punctuation.
  ........
    r82265 | georg.brandl | 2010-06-27 05:49:23 -0500 (Sun, 27 Jun 2010) | 1 line
    
    Use designated syntax for optional grammar element.
  ........
    r82266 | georg.brandl | 2010-06-27 05:51:44 -0500 (Sun, 27 Jun 2010) | 1 line
    
    Fix URL.
  ........
    r82267 | georg.brandl | 2010-06-27 05:55:38 -0500 (Sun, 27 Jun 2010) | 1 line
    
    Two typos.
  ........
................


Added:
   python/branches/release31-maint/Doc/documenting/building.rst
      - copied unchanged from r82301, /python/branches/py3k/Doc/documenting/building.rst
Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Doc/README.txt
   python/branches/release31-maint/Doc/c-api/init.rst
   python/branches/release31-maint/Doc/distutils/builtdist.rst
   python/branches/release31-maint/Doc/documenting/index.rst
   python/branches/release31-maint/Doc/documenting/markup.rst
   python/branches/release31-maint/Doc/documenting/style.rst
   python/branches/release31-maint/Doc/faq/gui.rst
   python/branches/release31-maint/Doc/howto/functional.rst
   python/branches/release31-maint/Doc/howto/unicode.rst
   python/branches/release31-maint/Doc/library/calendar.rst
   python/branches/release31-maint/Doc/library/logging.rst
   python/branches/release31-maint/Doc/library/os.rst
   python/branches/release31-maint/Doc/library/re.rst
   python/branches/release31-maint/Doc/library/socket.rst
   python/branches/release31-maint/Doc/library/stdtypes.rst
   python/branches/release31-maint/Doc/library/string.rst
   python/branches/release31-maint/Doc/library/unittest.rst
   python/branches/release31-maint/Doc/reference/datamodel.rst
   python/branches/release31-maint/Doc/reference/introduction.rst
   python/branches/release31-maint/Doc/tutorial/classes.rst
   python/branches/release31-maint/Doc/tutorial/introduction.rst
   python/branches/release31-maint/Doc/using/mac.rst
   python/branches/release31-maint/Doc/whatsnew/2.7.rst
   python/branches/release31-maint/Lib/urllib/parse.py
   python/branches/release31-maint/Modules/datetimemodule.c

Modified: python/branches/release31-maint/Doc/README.txt
==============================================================================
--- python/branches/release31-maint/Doc/README.txt	(original)
+++ python/branches/release31-maint/Doc/README.txt	Sun Jul 11 12:41:07 2010
@@ -14,12 +14,11 @@
 Building the docs
 =================
 
-You need to install Python 2.4 or higher (but Python 3.0 is not supported yet);
-the toolset used to build the docs are written in Python.  The toolset used
-to build the documentation is called *Sphinx*, it is not included in this
-tree, but maintained separately in the Python Subversion repository.  Also
-needed are Jinja, a templating engine (included in Sphinx as a Subversion
-external), and optionally Pygments, a code highlighter.
+You need to have Python 2.4 or higher installed; the toolset used to build the
+docs is written in Python.  It is called *Sphinx*, it is not included in this
+tree, but maintained separately.  Also needed are the docutils, supplying the
+base markup that Sphinx uses, Jinja, a templating engine, and optionally
+Pygments, a code highlighter.
 
 
 Using make
@@ -47,29 +46,29 @@
    convert them into a single Compiled HTML (.chm) file -- these are popular
    under Microsoft Windows, but very handy on every platform.
 
-   To create the CHM file, you need to run the Microsoft HTML Help Workshop
-   over the generated project (.hhp) file.
+   To create the CHM file, you need to run the Microsoft HTML Help Workshop over
+   the generated project (.hhp) file.
 
- * "latex", which builds LaTeX source files that can be run with "pdflatex"
-   to produce PDF documents.
+ * "latex", which builds LaTeX source files as input to "pdflatex" to produce
+   PDF documents.
 
  * "text", which builds a plain text file for each source file.
 
  * "linkcheck", which checks all external references to see whether they are
-   broken, redirected or malformed, and outputs this information to stdout
-   as well as a plain-text (.txt) file.
+   broken, redirected or malformed, and outputs this information to stdout as
+   well as a plain-text (.txt) file.
 
  * "changes", which builds an overview over all versionadded/versionchanged/
    deprecated items in the current version. This is meant as a help for the
    writer of the "What's New" document.
 
- * "coverage", which builds a coverage overview for standard library modules
-   and C API.
+ * "coverage", which builds a coverage overview for standard library modules and
+   C API.
 
- * "pydoc-topics", which builds a Python module containing a dictionary
-   with plain text documentation for the labels defined in
-   `tools/sphinxext/pyspecific.py` -- pydoc needs these to show topic
-   and keyword help.
+ * "pydoc-topics", which builds a Python module containing a dictionary with
+   plain text documentation for the labels defined in
+   `tools/sphinxext/pyspecific.py` -- pydoc needs these to show topic and
+   keyword help.
 
 A "make update" updates the Subversion checkouts in `tools/`.
 
@@ -95,7 +94,7 @@
 
 or by installing it from PyPI.
 
-You can optionally also install Pygments, either as a checkout via :: 
+You can optionally also install Pygments, either as a checkout via ::
 
    svn co http://svn.python.org/projects/external/Pygments-1.3.1/pygments tools/pygments
 

Modified: python/branches/release31-maint/Doc/c-api/init.rst
==============================================================================
--- python/branches/release31-maint/Doc/c-api/init.rst	(original)
+++ python/branches/release31-maint/Doc/c-api/init.rst	Sun Jul 11 12:41:07 2010
@@ -392,7 +392,7 @@
 
 .. cfunction:: void PySys_SetArgv(int argc, wchar_t **argv)
 
-   This function works like :cfunc:`PySys_SetArgv` with *updatepath* set to 1.
+   This function works like :cfunc:`PySys_SetArgvEx` with *updatepath* set to 1.
 
 
 .. cfunction:: void Py_SetPythonHome(wchar_t *home)

Modified: python/branches/release31-maint/Doc/distutils/builtdist.rst
==============================================================================
--- python/branches/release31-maint/Doc/distutils/builtdist.rst	(original)
+++ python/branches/release31-maint/Doc/distutils/builtdist.rst	Sun Jul 11 12:41:07 2010
@@ -321,7 +321,7 @@
 option.
 
 By default the installer will display the cool "Python Powered" logo when it is
-run, but you can also supply your own 152x161 bitmap which must be a Windows
+run, but you can also supply your own 152x261 bitmap which must be a Windows
 :file:`.bmp` file with the :option:`--bitmap` option.
 
 The installer will also display a large title on the desktop background window
@@ -374,7 +374,7 @@
 The Postinstallation script
 ---------------------------
 
-Starting with Python 2.3, a postinstallation script can be specified which the
+Starting with Python 2.3, a postinstallation script can be specified with the
 :option:`--install-script` option.  The basename of the script must be
 specified, and the script filename must also be listed in the scripts argument
 to the setup function.

Modified: python/branches/release31-maint/Doc/documenting/index.rst
==============================================================================
--- python/branches/release31-maint/Doc/documenting/index.rst	(original)
+++ python/branches/release31-maint/Doc/documenting/index.rst	Sun Jul 11 12:41:07 2010
@@ -10,9 +10,9 @@
 `reStructuredText`_, developed by the `docutils`_ project, amended by custom
 directives and using a toolset named `Sphinx`_ to postprocess the HTML output.
 
-This document describes the style guide for our documentation, the custom
-reStructuredText markup introduced to support Python documentation and how it
-should be used, as well as the Sphinx build system.
+This document describes the style guide for our documentation as well as the
+custom reStructuredText markup introduced by Sphinx to support Python
+documentation and how it should be used.
 
 .. _reStructuredText: http://docutils.sf.net/rst.html
 .. _docutils: http://docutils.sf.net/
@@ -35,3 +35,4 @@
    rest.rst
    markup.rst
    fromlatex.rst
+   building.rst

Modified: python/branches/release31-maint/Doc/documenting/markup.rst
==============================================================================
--- python/branches/release31-maint/Doc/documenting/markup.rst	(original)
+++ python/branches/release31-maint/Doc/documenting/markup.rst	Sun Jul 11 12:41:07 2010
@@ -698,10 +698,10 @@
       .. toctree::
          :maxdepth: 2
 
-         intro.rst
-         strings.rst
-         datatypes.rst
-         numeric.rst
+         intro
+         strings
+         datatypes
+         numeric
          (many more files listed here)
 
    This accomplishes two things:
@@ -709,8 +709,8 @@
    * Tables of contents from all those files are inserted, with a maximum depth
      of two, that means one nested heading.  ``toctree`` directives in those
      files are also taken into account.
-   * Sphinx knows that the relative order of the files ``intro.rst``,
-     ``strings.rst`` and so forth, and it knows that they are children of the
+   * Sphinx knows that the relative order of the files ``intro``,
+     ``strings`` and so forth, and it knows that they are children of the
      shown file, the library index.  From this information it generates "next
      chapter", "previous chapter" and "parent chapter" links.
 

Modified: python/branches/release31-maint/Doc/documenting/style.rst
==============================================================================
--- python/branches/release31-maint/Doc/documenting/style.rst	(original)
+++ python/branches/release31-maint/Doc/documenting/style.rst	Sun Jul 11 12:41:07 2010
@@ -7,7 +7,7 @@
 wherever possible. This particular style guide was selected mostly because it
 seems reasonable and is easy to get online.
 
-Topics which are not covered in the Apple's style guide will be discussed in
+Topics which are not covered in Apple's style guide will be discussed in
 this document.
 
 All reST files use an indentation of 3 spaces.  The maximum line length is 80

Modified: python/branches/release31-maint/Doc/faq/gui.rst
==============================================================================
--- python/branches/release31-maint/Doc/faq/gui.rst	(original)
+++ python/branches/release31-maint/Doc/faq/gui.rst	Sun Jul 11 12:41:07 2010
@@ -10,14 +10,14 @@
 =====================
 
 What platform-independent GUI toolkits exist for Python?
---------------------------------------------------------
+========================================================
 
 Depending on what platform(s) you are aiming at, there are several.
 
 .. XXX check links
 
 Tkinter
-'''''''
+-------
 
 Standard builds of Python include an object-oriented interface to the Tcl/Tk
 widget set, called Tkinter.  This is probably the easiest to install and use.
@@ -26,23 +26,26 @@
 Unix platforms.
 
 wxWindows
-'''''''''
+---------
 
-wxWindows is a portable GUI class library written in C++ that's a portable
-interface to various platform-specific libraries; wxWidgets is a Python
-interface to wxWindows.  wxWindows supports Windows and MacOS; on Unix variants,
-it supports both GTk+ and Motif toolkits.  wxWindows preserves the look and feel
-of the underlying graphics toolkit, and there is quite a rich widget set and
-collection of GDI classes.  See `the wxWindows page <http://www.wxwindows.org>`_
-for more details.
-
-`wxWidgets <http://wxwidgets.org>`_ is an extension module that wraps many of
-the wxWindows C++ classes, and is quickly gaining popularity amongst Python
-developers.  You can get wxWidgets as part of the source or CVS distribution of
-wxWindows, or directly from its home page.
+wxWidgets (http://www.wxwidgets.org) is a free, portable GUI class
+library written in C++ that provides a native look and feel on a
+number of platforms, with Windows, MacOS X, GTK, X11, all listed as
+current stable targets.  Language bindings are available for a number
+of languages including Python, Perl, Ruby, etc.
+
+wxPython (http://www.wxpython.org) is the Python binding for
+wxwidgets.  While it often lags slightly behind the official wxWidgets
+releases, it also offers a number of features via pure Python
+extensions that are not available in other language bindings.  There
+is an active wxPython user and developer community.
+
+Both wxWidgets and wxPython are free, open source, software with
+permissive licences that allow their use in commercial products as
+well as in freeware or shareware.
 
 Qt
-'''
+---
 
 There are bindings available for the Qt toolkit (`PyQt
 <http://www.riverbankcomputing.co.uk/software/pyqt/>`_) and for KDE (PyKDE).  If
@@ -53,13 +56,13 @@
 <http://www.trolltech.com>`_.
 
 Gtk+
-''''
+----
 
 PyGtk bindings for the `Gtk+ toolkit <http://www.gtk.org>`_ have been
 implemented by by James Henstridge; see ftp://ftp.gtk.org/pub/gtk/python/.
 
 FLTK
-''''
+----
 
 Python bindings for `the FLTK toolkit <http://www.fltk.org>`_, a simple yet
 powerful and mature cross-platform windowing system, are available from `the
@@ -67,7 +70,7 @@
 
 
 FOX
-'''
+----
 
 A wrapper for `the FOX toolkit <http://www.fox-toolkit.org/>`_ called `FXpy
 <http://fxpy.sourceforge.net/>`_ is available.  FOX supports both Unix variants
@@ -75,13 +78,13 @@
 
 
 OpenGL
-''''''
+------
 
 For OpenGL bindings, see `PyOpenGL <http://pyopengl.sourceforge.net>`_.
 
 
 What platform-specific GUI toolkits exist for Python?
------------------------------------------------------
+========================================================
 
 `The Mac port <http://python.org/download/mac>`_ by Jack Jansen has a rich and
 ever-growing set of modules that support the native Mac toolbox calls.  The port

Modified: python/branches/release31-maint/Doc/howto/functional.rst
==============================================================================
--- python/branches/release31-maint/Doc/howto/functional.rst	(original)
+++ python/branches/release31-maint/Doc/howto/functional.rst	Sun Jul 11 12:41:07 2010
@@ -5,9 +5,6 @@
 :Author: A. M. Kuchling
 :Release: 0.31
 
-(This is a first draft.  Please send comments/error reports/suggestions to
-amk at amk.ca.)
-
 In this document, we'll take a tour of Python's features suitable for
 implementing programs in a functional style.  After an introduction to the
 concepts of functional programming, we'll look at language features such as

Modified: python/branches/release31-maint/Doc/howto/unicode.rst
==============================================================================
--- python/branches/release31-maint/Doc/howto/unicode.rst	(original)
+++ python/branches/release31-maint/Doc/howto/unicode.rst	Sun Jul 11 12:41:07 2010
@@ -4,10 +4,12 @@
   Unicode HOWTO
 *****************
 
-:Release: 1.1
+:Release: 1.11
 
-This HOWTO discusses Python's support for Unicode, and explains various problems
-that people commonly encounter when trying to work with Unicode.
+This HOWTO discusses Python 2.x's support for Unicode, and explains
+various problems that people commonly encounter when trying to work
+with Unicode.  (This HOWTO has not yet been updated to cover the 3.x
+versions of Python.)
 
 
 Introduction to Unicode
@@ -146,8 +148,9 @@
 4. Many Internet standards are defined in terms of textual data, and can't
    handle content with embedded zero bytes.
 
-Generally people don't use this encoding, instead choosing other encodings that
-are more efficient and convenient.
+Generally people don't use this encoding, instead choosing other
+encodings that are more efficient and convenient.  UTF-8 is probably
+the most commonly supported encoding; it will be discussed below.
 
 Encodings don't have to handle every possible Unicode character, and most
 encodings don't.  The rules for converting a Unicode string into the ASCII
@@ -223,8 +226,8 @@
 <http://en.wikipedia.org/wiki/UTF-8>, for example.
 
 
-Python's Unicode Support
-========================
+Python 2.x's Unicode Support
+============================
 
 Now that you've learned the rudiments of Unicode, we can look at Python's
 Unicode features.
@@ -266,8 +269,8 @@
     >>> b'\x80abc'.decode("utf-8", "ignore")
     'abc'
 
-Encodings are specified as strings containing the encoding's name.  Python comes
-with roughly 100 different encodings; see the Python Library Reference at
+Encodings are specified as strings containing the encoding's name.  Python 3.2
+comes with roughly 100 different encodings; see the Python Library Reference at
 :ref:`standard-encodings` for a list.  Some encodings have multiple names; for
 example, 'latin-1', 'iso_8859_1' and '8859' are all synonyms for the same
 encoding.
@@ -626,7 +629,10 @@
 
 Version 1.1: Feb-Nov 2008.  Updates the document with respect to Python 3 changes.
 
+Version 1.11: posted June 20 2010.  Notes that Python 3.x is not covered,
+and that the HOWTO only covers 2.x.
 
+.. comment Describe Python 3.x support (new section? new document?)
 .. comment Additional topic: building Python w/ UCS2 or UCS4 support
 .. comment Describe use of codecs.StreamRecoder and StreamReaderWriter
 

Modified: python/branches/release31-maint/Doc/library/calendar.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/calendar.rst	(original)
+++ python/branches/release31-maint/Doc/library/calendar.rst	Sun Jul 11 12:41:07 2010
@@ -15,7 +15,7 @@
 are given as integers. For related
 functionality, see also the :mod:`datetime` and :mod:`time` modules.
 
-Most of these functions and classses rely on the :mod:`datetime` module which
+Most of these functions and classes rely on the :mod:`datetime` module which
 uses an idealized calendar, the current Gregorian calendar indefinitely extended
 in both directions.  This matches the definition of the "proleptic Gregorian"
 calendar in Dershowitz and Reingold's book "Calendrical Calculations", where

Modified: python/branches/release31-maint/Doc/library/logging.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/logging.rst	(original)
+++ python/branches/release31-maint/Doc/library/logging.rst	Sun Jul 11 12:41:07 2010
@@ -53,10 +53,12 @@
 
 Most applications are probably going to want to log to a file, so let's start
 with that case. Using the :func:`basicConfig` function, we can set up the
-default handler so that debug messages are written to a file::
+default handler so that debug messages are written to a file (in the example,
+we assume that you have the appropriate permissions to create a file called
+*example.log* in the current directory)::
 
    import logging
-   LOG_FILENAME = '/tmp/logging_example.out'
+   LOG_FILENAME = 'example.log'
    logging.basicConfig(filename=LOG_FILENAME,level=logging.DEBUG)
 
    logging.debug('This message should go to the log file')
@@ -75,7 +77,7 @@
    import logging
    import logging.handlers
 
-   LOG_FILENAME = '/tmp/logging_rotatingfile_example.out'
+   LOG_FILENAME = 'logging_rotatingfile_example.out'
 
    # Set up a specific logger with our desired output level
    my_logger = logging.getLogger('MyLogger')
@@ -100,14 +102,14 @@
 The result should be 6 separate files, each with part of the log history for the
 application::
 
-   /tmp/logging_rotatingfile_example.out
-   /tmp/logging_rotatingfile_example.out.1
-   /tmp/logging_rotatingfile_example.out.2
-   /tmp/logging_rotatingfile_example.out.3
-   /tmp/logging_rotatingfile_example.out.4
-   /tmp/logging_rotatingfile_example.out.5
+   logging_rotatingfile_example.out
+   logging_rotatingfile_example.out.1
+   logging_rotatingfile_example.out.2
+   logging_rotatingfile_example.out.3
+   logging_rotatingfile_example.out.4
+   logging_rotatingfile_example.out.5
 
-The most current file is always :file:`/tmp/logging_rotatingfile_example.out`,
+The most current file is always :file:`logging_rotatingfile_example.out`,
 and each time it reaches the size limit it is renamed with the suffix
 ``.1``. Each of the existing backup files is renamed to increment the suffix
 (``.1`` becomes ``.2``, etc.)  and the ``.6`` file is erased.
@@ -1039,14 +1041,14 @@
 
    logging.basicConfig(level=logging.DEBUG,
                        format='%(asctime)s %(levelname)s %(message)s',
-                       filename='/tmp/myapp.log',
+                       filename='myapp.log',
                        filemode='w')
    logging.debug('A debug message')
    logging.info('Some information')
    logging.warning('A shot across the bows')
 
 The :meth:`basicConfig` method is used to change the configuration defaults,
-which results in output (written to ``/tmp/myapp.log``) which should look
+which results in output (written to ``myapp.log``) which should look
 something like the following::
 
    2004-07-02 13:00:08,743 DEBUG A debug message

Modified: python/branches/release31-maint/Doc/library/os.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/os.rst	(original)
+++ python/branches/release31-maint/Doc/library/os.rst	Sun Jul 11 12:41:07 2010
@@ -1003,7 +1003,8 @@
 
    Create a directory named *path* with numeric mode *mode*. The default *mode*
    is ``0o777`` (octal).  On some systems, *mode* is ignored.  Where it is used,
-   the current umask value is first masked out.
+   the current umask value is first masked out.  If the directory already
+   exists, :exc:`OSError` is raised.
 
    It is also possible to create temporary directories; see the
    :mod:`tempfile` module's :func:`tempfile.mkdtemp` function.

Modified: python/branches/release31-maint/Doc/library/re.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/re.rst	(original)
+++ python/branches/release31-maint/Doc/library/re.rst	Sun Jul 11 12:41:07 2010
@@ -714,18 +714,12 @@
 
    The :class:`RegexObject` class supports the following methods and attributes:
 
+   .. method:: RegexObject.search(string[, pos[, endpos]])
 
-   .. method:: RegexObject.match(string[, pos[, endpos]])
-
-      If zero or more characters at the beginning of *string* match this regular
-      expression, return a corresponding :class:`MatchObject` instance.  Return
-      ``None`` if the string does not match the pattern; note that this is different
-      from a zero-length match.
-
-      .. note::
-
-         If you want to locate a match anywhere in *string*, use
-         :meth:`~RegexObject.search` instead.
+      Scan through *string* looking for a location where this regular expression
+      produces a match, and return a corresponding :class:`MatchObject` instance.
+      Return ``None`` if no position in the string matches the pattern; note that this
+      is different from finding a zero-length match at some point in the string.
 
       The optional second parameter *pos* gives an index in the string where the
       search is to start; it defaults to ``0``.  This is not completely equivalent to
@@ -737,24 +731,34 @@
       will be as if the string is *endpos* characters long, so only the characters
       from *pos* to ``endpos - 1`` will be searched for a match.  If *endpos* is less
       than *pos*, no match will be found, otherwise, if *rx* is a compiled regular
-      expression object, ``rx.match(string, 0, 50)`` is equivalent to
-      ``rx.match(string[:50], 0)``.
+      expression object, ``rx.search(string, 0, 50)`` is equivalent to
+      ``rx.search(string[:50], 0)``.
 
-         >>> pattern = re.compile("o")
-         >>> pattern.match("dog")      # No match as "o" is not at the start of "dog."
-         >>> pattern.match("dog", 1)   # Match as "o" is the 2nd character of "dog".
-         <_sre.SRE_Match object at ...>
+      >>> pattern = re.compile("d")
+      >>> pattern.search("dog")     # Match at index 0
+      <_sre.SRE_Match object at ...>
+      >>> pattern.search("dog", 1)  # No match; search doesn't include the "d"
 
 
-   .. method:: RegexObject.search(string[, pos[, endpos]])
+   .. method:: RegexObject.match(string[, pos[, endpos]])
 
-      Scan through *string* looking for a location where this regular expression
-      produces a match, and return a corresponding :class:`MatchObject` instance.
-      Return ``None`` if no position in the string matches the pattern; note that this
-      is different from finding a zero-length match at some point in the string.
+      If zero or more characters at the *beginning* of *string* match this regular
+      expression, return a corresponding :class:`MatchObject` instance.  Return
+      ``None`` if the string does not match the pattern; note that this is different
+      from a zero-length match.
 
       The optional *pos* and *endpos* parameters have the same meaning as for the
-      :meth:`~RegexObject.match` method.
+      :meth:`~RegexObject.search` method.
+
+      .. note::
+
+         If you want to locate a match anywhere in *string*, use
+         :meth:`~RegexObject.search` instead.
+
+      >>> pattern = re.compile("o")
+      >>> pattern.match("dog")      # No match as "o" is not at the start of "dog".
+      >>> pattern.match("dog", 1)   # Match as "o" is the 2nd character of "dog".
+      <_sre.SRE_Match object at ...>
 
 
    .. method:: RegexObject.split(string[, maxsplit=0])
@@ -764,12 +768,16 @@
 
    .. method:: RegexObject.findall(string[, pos[, endpos]])
 
-      Identical to the :func:`findall` function, using the compiled pattern.
+      Similar to the :func:`findall` function, using the compiled pattern, but
+      also accepts optional *pos* and *endpos* parameters that limit the search
+      region like for :meth:`match`.
 
 
    .. method:: RegexObject.finditer(string[, pos[, endpos]])
 
-      Identical to the :func:`finditer` function, using the compiled pattern.
+      Similar to the :func:`finditer` function, using the compiled pattern, but
+      also accepts optional *pos* and *endpos* parameters that limit the search
+      region like for :meth:`match`.
 
 
    .. method:: RegexObject.sub(repl, string[, count=0])

Modified: python/branches/release31-maint/Doc/library/socket.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/socket.rst	(original)
+++ python/branches/release31-maint/Doc/library/socket.rst	Sun Jul 11 12:41:07 2010
@@ -72,18 +72,21 @@
 tuple, and the fields depend on the address type. The general tuple form is
 ``(addr_type, v1, v2, v3 [, scope])``, where:
 
-   - *addr_type* is one of TIPC_ADDR_NAMESEQ, TIPC_ADDR_NAME, or
-     TIPC_ADDR_ID.
-   - *scope* is one of TIPC_ZONE_SCOPE, TIPC_CLUSTER_SCOPE, and
-     TIPC_NODE_SCOPE.
-   - If *addr_type* is TIPC_ADDR_NAME, then *v1* is the server type, *v2* is
-     the port identifier, and *v3* should be 0.
+- *addr_type* is one of TIPC_ADDR_NAMESEQ, TIPC_ADDR_NAME, or
+  TIPC_ADDR_ID.
+- *scope* is one of TIPC_ZONE_SCOPE, TIPC_CLUSTER_SCOPE, and
+  TIPC_NODE_SCOPE.
+- If *addr_type* is TIPC_ADDR_NAME, then *v1* is the server type, *v2* is
+  the port identifier, and *v3* should be 0.
 
-     If *addr_type* is TIPC_ADDR_NAMESEQ, then *v1* is the server type, *v2*
-     is the lower port number, and *v3* is the upper port number.
+  If *addr_type* is TIPC_ADDR_NAMESEQ, then *v1* is the server type, *v2*
+  is the lower port number, and *v3* is the upper port number.
 
-     If *addr_type* is TIPC_ADDR_ID, then *v1* is the node, *v2* is the
-     reference, and *v3* should be set to 0.
+  If *addr_type* is TIPC_ADDR_ID, then *v1* is the node, *v2* is the
+  reference, and *v3* should be set to 0.
+
+  If *addr_type* is TIPC_ADDR_ID, then *v1* is the node, *v2* is the
+  reference, and *v3* should be set to 0.
 
 
 All errors raise exceptions.  The normal exceptions for invalid argument types
@@ -684,7 +687,7 @@
 
    Set a timeout on blocking socket operations.  The *value* argument can be a
    nonnegative float expressing seconds, or ``None``. If a float is given,
-   subsequent socket operations will raise an :exc:`timeout` exception if the
+   subsequent socket operations will raise a :exc:`timeout` exception if the
    timeout period *value* has elapsed before the operation has completed.  Setting
    a timeout of ``None`` disables timeouts on socket operations.
    ``s.settimeout(0.0)`` is equivalent to ``s.setblocking(0)``;

Modified: python/branches/release31-maint/Doc/library/stdtypes.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/stdtypes.rst	(original)
+++ python/branches/release31-maint/Doc/library/stdtypes.rst	Sun Jul 11 12:41:07 2010
@@ -1162,6 +1162,10 @@
    You can use :meth:`str.maketrans` to create a translation map from
    character-to-character mappings in different formats.
 
+   You can use the :func:`~string.maketrans` helper function in the :mod:`string`
+   module to create a translation table. For string objects, set the *table*
+   argument to ``None`` for translations that only delete characters:
+
    .. note::
 
       An even more flexible approach is to create a custom character mapping
@@ -2118,7 +2122,7 @@
       File "<stdin>", line 1, in <module>
       ValueError: cannot modify size of memoryview object
 
-   Notice how the size of the memoryview object can not be changed.
+   Notice how the size of the memoryview object cannot be changed.
 
 
    :class:`memoryview` has two methods:

Modified: python/branches/release31-maint/Doc/library/string.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/string.rst	(original)
+++ python/branches/release31-maint/Doc/library/string.rst	Sun Jul 11 12:41:07 2010
@@ -197,7 +197,7 @@
    .. productionlist:: sf
       replacement_field: "{" [`field_name`] ["!" `conversion`] [":" `format_spec`] "}"
       field_name: arg_name ("." `attribute_name` | "[" `element_index` "]")*
-      arg_name: (`identifier` | `integer`)?
+      arg_name: [`identifier` | `integer`]
       attribute_name: `identifier`
       element_index: `integer` | `index_string`
       index_string: <any source character except "]"> +

Modified: python/branches/release31-maint/Doc/library/unittest.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/unittest.rst	(original)
+++ python/branches/release31-maint/Doc/library/unittest.rst	Sun Jul 11 12:41:07 2010
@@ -1413,5 +1413,5 @@
    Calling ``main`` actually returns an instance of the ``TestProgram`` class.
    This stores the result of the tests run as the ``result`` attribute.
 
-   .. versionchanged:: 3.2
+   .. versionchanged:: 3.1
       The ``exit`` parameter was added.

Modified: python/branches/release31-maint/Doc/reference/datamodel.rst
==============================================================================
--- python/branches/release31-maint/Doc/reference/datamodel.rst	(original)
+++ python/branches/release31-maint/Doc/reference/datamodel.rst	Sun Jul 11 12:41:07 2010
@@ -1596,7 +1596,7 @@
 
 In particular, the metaclass :class:`abc.ABCMeta` implements these methods in
 order to allow the addition of Abstract Base Classes (ABCs) as "virtual base
-classes" to any class or type (including built-in types), and including to other
+classes" to any class or type (including built-in types), including other
 ABCs.
 
 .. method:: class.__instancecheck__(self, instance)
@@ -1615,7 +1615,7 @@
 
 Note that these methods are looked up on the type (metaclass) of a class.  They
 cannot be defined as class methods in the actual class.  This is consistent with
-the lookup of special methods that are called on instances, only that in this
+the lookup of special methods that are called on instances, only in this
 case the instance is itself a class.
 
 .. seealso::

Modified: python/branches/release31-maint/Doc/reference/introduction.rst
==============================================================================
--- python/branches/release31-maint/Doc/reference/introduction.rst	(original)
+++ python/branches/release31-maint/Doc/reference/introduction.rst	Sun Jul 11 12:41:07 2010
@@ -69,12 +69,12 @@
    more information, see `the IronPython website <http://www.ironpython.com/>`_.
 
 PyPy
-   An implementation of Python written in Python; even the bytecode interpreter is
-   written in Python.  This is executed using CPython as the underlying
-   interpreter.  One of the goals of the project is to encourage experimentation
-   with the language itself by making it easier to modify the interpreter (since it
-   is written in Python).  Additional information is available on `the PyPy
-   project's home page <http://codespeak.net/pypy/>`_.
+   An implementation of Python written completely in Python. It supports several
+   advanced features not found in other implementations like stackless support
+   and a Just in Time compiler. One of the goals of the project is to encourage
+   experimentation with the language itself by making it easier to modify the
+   interpreter (since it is written in Python).  Additional information is
+   available on `the PyPy project's home page <http://pypy.org/>`_.
 
 Each of these implementations varies in some way from the language as documented
 in this manual, or introduces specific information beyond what's covered in the

Modified: python/branches/release31-maint/Doc/tutorial/classes.rst
==============================================================================
--- python/branches/release31-maint/Doc/tutorial/classes.rst	(original)
+++ python/branches/release31-maint/Doc/tutorial/classes.rst	Sun Jul 11 12:41:07 2010
@@ -580,7 +580,7 @@
 =================
 
 "Private" instance variables that cannot be accessed except from inside an
-object, don't exist in Python.  However, there is a convention that is followed
+object don't exist in Python.  However, there is a convention that is followed
 by most Python code: a name prefixed with an underscore (e.g. ``_spam``) should
 be treated as a non-public part of the API (whether it is a function, a method
 or a data member).  It should be considered an implementation detail and subject

Modified: python/branches/release31-maint/Doc/tutorial/introduction.rst
==============================================================================
--- python/branches/release31-maint/Doc/tutorial/introduction.rst	(original)
+++ python/branches/release31-maint/Doc/tutorial/introduction.rst	Sun Jul 11 12:41:07 2010
@@ -208,7 +208,7 @@
 
    print(hello)
 
-Note that newlines still need to be embedded in the string using ``\n``; the
+Note that newlines still need to be embedded in the string using ``\n`` -- the
 newline following the trailing backslash is discarded.  This example would print
 the following:
 

Modified: python/branches/release31-maint/Doc/using/mac.rst
==============================================================================
--- python/branches/release31-maint/Doc/using/mac.rst	(original)
+++ python/branches/release31-maint/Doc/using/mac.rst	Sun Jul 11 12:41:07 2010
@@ -114,8 +114,8 @@
 =======
 
 MacPython ships with the standard IDLE development environment. A good
-introduction to using IDLE can be found at http://hkn.eecs.berkeley.edu/
-dyoo/python/idle_intro/index.html.
+introduction to using IDLE can be found at
+http://hkn.eecs.berkeley.edu/~dyoo/python/idle_intro/index.html.
 
 
 .. _mac-package-manager:

Modified: python/branches/release31-maint/Doc/whatsnew/2.7.rst
==============================================================================
--- python/branches/release31-maint/Doc/whatsnew/2.7.rst	(original)
+++ python/branches/release31-maint/Doc/whatsnew/2.7.rst	Sun Jul 11 12:41:07 2010
@@ -6,7 +6,10 @@
 :Release: |release|
 :Date: |today|
 
-.. Fix accents on Kristjan Valur Jonsson, Fuerstenau, Tarek Ziade.
+..  hyperlink all the methods & functions.
+
+.. T_STRING_INPLACE not described in main docs
+.. "Format String Syntax" in string.rst could use many more examples.
 
 .. $Id$
    Rules for maintenance:
@@ -37,7 +40,7 @@
    * You can comment out your additions if you like, but it's not
    necessary (especially when a final release is some months away).
 
-   * Credit the author of a patch or bugfix.   Just the name is
+   * Credit the author of a patch or bugfix.  Just the name is
    sufficient; the e-mail address isn't necessary.
 
    * It's helpful to add the bug/patch number in a parenthetical comment.
@@ -49,38 +52,564 @@
    This saves the maintainer some effort going through the SVN logs
    when researching a change.
 
-This article explains the new features in Python 2.7.
-No release schedule has been decided yet for 2.7.
+This article explains the new features in Python 2.7.  The final
+release of 2.7 is currently scheduled for July 2010; the detailed
+schedule is described in :pep:`373`.
+
+Numeric handling has been improved in many ways, for both
+floating-point numbers and for the :class:`Decimal` class.  There are
+some useful additions to the standard library, such as a greatly
+enhanced :mod:`unittest` module, the :mod:`argparse` module for
+parsing command-line options, convenient ordered-dictionary and
+:class:`Counter` classes in the :mod:`collections` module, and many
+other improvements.
+
+Python 2.7 is planned to be the last of the 2.x releases, so we worked
+on making it a good release for the long term.  To help with porting
+to Python 3, several new features from the Python 3.x series have been
+included in 2.7.
+
+This article doesn't attempt to provide a complete specification of
+the new features, but instead provides a convenient overview.  For
+full details, you should refer to the documentation for Python 2.7 at
+http://docs.python.org. If you want to understand the rationale for
+the design and implementation, refer to the PEP for a particular new
+feature or the issue on http://bugs.python.org in which a change was
+discussed.  Whenever possible, "What's New in Python" links to the
+bug/patch item for each change.
+
+.. _whatsnew27-python31:
+
+The Future for Python 2.x
+=========================
+
+Python 2.7 is intended to be the last major release in the 2.x series.
+The Python maintainers are planning to focus their future efforts on
+the Python 3.x series.
+
+This means that 2.7 will remain in place for a long time, running
+production systems that have not been ported to Python 3.x.
+Two consequences of the long-term significance of 2.7 are:
+
+* It's very likely the 2.7 release will have a longer period of
+  maintenance compared to earlier 2.x versions.  Python 2.7 will
+  continue to be maintained while the transition to 3.x continues, and
+  the developers are planning to support Python 2.7 with bug-fix
+  releases beyond the typical two years.
+
+* A policy decision was made to silence warnings only of interest to
+  developers.  :exc:`DeprecationWarning` and its
+  descendants are now ignored unless otherwise requested, preventing
+  users from seeing warnings triggered by an application.  This change
+  was also made in the branch that will become Python 3.2. (Discussed
+  on stdlib-sig and carried out in :issue:`7319`.)
+
+  In previous releases, :exc:`DeprecationWarning` messages were
+  enabled by default, providing Python developers with a clear
+  indication of where their code may break in a future major version
+  of Python.
+
+  However, there are increasingly many users of Python-based
+  applications who are not directly involved in the development of
+  those applications.  :exc:`DeprecationWarning` messages are
+  irrelevant to such users, making them worry about an application
+  that's actually working correctly and burdening application developers
+  with responding to these concerns.
+
+  You can re-enable display of :exc:`DeprecationWarning` messages by
+  running Python with the :option:`-Wdefault` (short form:
+  :option:`-Wd`) switch, or by setting the :envvar:`PYTHONWARNINGS`
+  environment variable to ``"default"`` (or ``"d"``) before running
+  Python.  Python code can also re-enable them
+  by calling ``warnings.simplefilter('default')``.
 
-.. Compare with previous release in 2 - 3 sentences here.
-   add hyperlink when the documentation becomes available online.
 
-Python 3.1
-================
+Python 3.1 Features
+=======================
 
 Much as Python 2.6 incorporated features from Python 3.0,
-version 2.7 is influenced by features from 3.1.
+version 2.7 incorporates some of the new features
+in Python 3.1.  The 2.x series continues to provide tools
+for migrating to the 3.x series.
+
+A partial list of 3.1 features that were backported to 2.7:
+
+* The syntax for set literals (``{1,2,3}`` is a mutable set).
+* Dictionary and set comprehensions (``{ i: i*2 for i in range(3)}``).
+* Multiple context managers in a single :keyword:`with` statement.
+* A new version of the :mod:`io` library, rewritten in C for performance.
+* The ordered-dictionary type described in :ref:`pep-0372`.
+* The new ``","`` format specifier described in :ref:`pep-0378`.
+* The :class:`memoryview` object.
+* A small subset of the :mod:`importlib` module,
+  `described below <#importlib-section>`__.
+* Float-to-string and string-to-float conversions now round their
+  results more correctly, and :func:`repr` of a floating-point
+  number *x* returns a result that's guaranteed to round back to the
+  same number when converted back to a float.
+* The :ctype:`PyCapsule` type, used to provide a C API for extension modules.
+* The :cfunc:`PyLong_AsLongAndOverflow` C API function.
+
+Other new Python3-mode warnings include:
+
+* :func:`operator.isCallable` and :func:`operator.sequenceIncludes`,
+  which are not supported in 3.x, now trigger warnings.
+* The :option:`-3` switch now automatically
+  enables the :option:`-Qwarn` switch that causes warnings
+  about using classic division with integers and long integers.
 
-XXX mention importlib; anything else?
 
-One porting change: the :option:`-3` switch now automatically
-enables the :option:`-Qwarn` switch that causes warnings
-about using classic division with integers and long integers.
 
 .. ========================================================================
 .. Large, PEP-level features and changes should be described here.
 .. ========================================================================
 
-PEP 372: Adding an ordered dictionary to collections
+.. _pep-0372:
+
+PEP 372: Adding an Ordered Dictionary to collections
+====================================================
+
+Regular Python dictionaries iterate over key/value pairs in arbitrary order.
+Over the years, a number of authors have written alternative implementations
+that remember the order that the keys were originally inserted.  Based on
+the experiences from those implementations, 2.7 introduces a new
+:class:`~collections.OrderedDict` class in the :mod:`collections` module.
+
+The :class:`~collections.OrderedDict` API provides the same interface as regular
+dictionaries but iterates over keys and values in a guaranteed order
+depending on when a key was first inserted::
+
+    >>> from collections import OrderedDict
+    >>> d = OrderedDict([('first', 1),
+    ...                  ('second', 2),
+    ...                  ('third', 3)])
+    >>> d.items()
+    [('first', 1), ('second', 2), ('third', 3)]
+
+If a new entry overwrites an existing entry, the original insertion
+position is left unchanged::
+
+    >>> d['second'] = 4
+    >>> d.items()
+    [('first', 1), ('second', 4), ('third', 3)]
+
+Deleting an entry and reinserting it will move it to the end::
+
+    >>> del d['second']
+    >>> d['second'] = 5
+    >>> d.items()
+    [('first', 1), ('third', 3), ('second', 5)]
+
+The :meth:`~collections.OrderedDict.popitem` method has an optional *last*
+argument that defaults to True.  If *last* is True, the most recently
+added key is returned and removed; if it's False, the
+oldest key is selected::
+
+    >>> od = OrderedDict([(x,0) for x in range(20)])
+    >>> od.popitem()
+    (19, 0)
+    >>> od.popitem()
+    (18, 0)
+    >>> od.popitem(last=False)
+    (0, 0)
+    >>> od.popitem(last=False)
+    (1, 0)
+
+Comparing two ordered dictionaries checks both the keys and values,
+and requires that the insertion order was the same::
+
+    >>> od1 = OrderedDict([('first', 1),
+    ...                    ('second', 2),
+    ...                    ('third', 3)])
+    >>> od2 = OrderedDict([('third', 3),
+    ...                    ('first', 1),
+    ...                    ('second', 2)])
+    >>> od1 == od2
+    False
+    >>> # Move 'third' key to the end
+    >>> del od2['third']; od2['third'] = 3
+    >>> od1 == od2
+    True
+
+Comparing an :class:`~collections.OrderedDict` with a regular dictionary
+ignores the insertion order and just compares the keys and values.
+
+How does the :class:`~collections.OrderedDict` work?  It maintains a
+doubly-linked list of keys, appending new keys to the list as they're inserted.
+A secondary dictionary maps keys to their corresponding list node, so
+deletion doesn't have to traverse the entire linked list and therefore
+remains O(1).
+
+The standard library now supports use of ordered dictionaries in several
+modules.
+
+* The :mod:`ConfigParser` module uses them by default, meaning that
+  configuration files can now read, modified, and then written back
+  in their original order.
+
+* The :meth:`~collections.somenamedtuple._asdict()` method for
+  :func:`collections.namedtuple` now returns an ordered dictionary with the
+  values appearing in the same order as the underlying tuple indices.
+
+* The :mod:`json` module's :class:`~json.JSONDecoder` class
+  constructor was extended with an *object_pairs_hook* parameter to
+  allow :class:`OrderedDict` instances to be built by the decoder.
+  Support was also added for third-party tools like
+  `PyYAML <http://pyyaml.org/>`_.
+
+.. seealso::
+
+   :pep:`372` - Adding an ordered dictionary to collections
+     PEP written by Armin Ronacher and Raymond Hettinger;
+     implemented by Raymond Hettinger.
+
+.. _pep-0378:
+
+PEP 378: Format Specifier for Thousands Separator
+=================================================
+
+To make program output more readable, it can be useful to add
+separators to large numbers, rendering them as
+18,446,744,073,709,551,616 instead of 18446744073709551616.
+
+The fully general solution for doing this is the :mod:`locale` module,
+which can use different separators ("," in North America, "." in
+Europe) and different grouping sizes, but :mod:`locale` is complicated
+to use and unsuitable for multi-threaded applications where different
+threads are producing output for different locales.
+
+Therefore, a simple comma-grouping mechanism has been added to the
+mini-language used by the :meth:`str.format` method.  When
+formatting a floating-point number, simply include a comma between the
+width and the precision::
+
+   >>> '{:20,.2f}'.format(18446744073709551616.0)
+   '18,446,744,073,709,551,616.00'
+
+When formatting an integer, include the comma after the width:
+
+   >>> '{:20,d}'.format(18446744073709551616)
+   '18,446,744,073,709,551,616'
+
+This mechanism is not adaptable at all; commas are always used as the
+separator and the grouping is always into three-digit groups.  The
+comma-formatting mechanism isn't as general as the :mod:`locale`
+module, but it's easier to use.
+
+.. seealso::
+
+   :pep:`378` - Format Specifier for Thousands Separator
+     PEP written by Raymond Hettinger; implemented by Eric Smith.
+
+PEP 389: The argparse Module for Parsing Command Lines
+======================================================
+
+The :mod:`argparse` module for parsing command-line arguments was
+added as a more powerful replacement for the
+:mod:`optparse` module.
+
+This means Python now supports three different modules for parsing
+command-line arguments: :mod:`getopt`, :mod:`optparse`, and
+:mod:`argparse`.  The :mod:`getopt` module closely resembles the C
+library's :cfunc:`getopt` function, so it remains useful if you're writing a
+Python prototype that will eventually be rewritten in C.
+:mod:`optparse` becomes redundant, but there are no plans to remove it
+because there are many scripts still using it, and there's no
+automated way to update these scripts.  (Making the :mod:`argparse`
+API consistent with :mod:`optparse`'s interface was discussed but
+rejected as too messy and difficult.)
+
+In short, if you're writing a new script and don't need to worry
+about compatibility with earlier versions of Python, use
+:mod:`argparse` instead of :mod:`optparse`.
+
+Here's an example::
+
+    import argparse
+
+    parser = argparse.ArgumentParser(description='Command-line example.')
+
+    # Add optional switches
+    parser.add_argument('-v', action='store_true', dest='is_verbose',
+                        help='produce verbose output')
+    parser.add_argument('-o', action='store', dest='output',
+                        metavar='FILE',
+                        help='direct output to FILE instead of stdout')
+    parser.add_argument('-C', action='store', type=int, dest='context',
+                        metavar='NUM', default=0,
+                        help='display NUM lines of added context')
+
+    # Allow any number of additional arguments.
+    parser.add_argument(nargs='*', action='store', dest='inputs',
+                        help='input filenames (default is stdin)')
+
+    args = parser.parse_args()
+    print args.__dict__
+
+Unless you override it, :option:`-h` and :option:`--help` switches
+are automatically added, and produce neatly formatted output::
+
+    -> ./python.exe argparse-example.py --help
+    usage: argparse-example.py [-h] [-v] [-o FILE] [-C NUM] [inputs [inputs ...]]
+
+    Command-line example.
+
+    positional arguments:
+      inputs      input filenames (default is stdin)
+
+    optional arguments:
+      -h, --help  show this help message and exit
+      -v          produce verbose output
+      -o FILE     direct output to FILE instead of stdout
+      -C NUM      display NUM lines of added context
+
+As with :mod:`optparse`, the command-line switches and arguments
+are returned as an object with attributes named by the *dest* parameters::
+
+    -> ./python.exe argparse-example.py -v
+    {'output': None,
+     'is_verbose': True,
+     'context': 0,
+     'inputs': []}
+
+    -> ./python.exe argparse-example.py -v -o /tmp/output -C 4 file1 file2
+    {'output': '/tmp/output',
+     'is_verbose': True,
+     'context': 4,
+     'inputs': ['file1', 'file2']}
+
+:mod:`argparse` has much fancier validation than :mod:`optparse`; you
+can specify an exact number of arguments as an integer, 0 or more
+arguments by passing ``'*'``, 1 or more by passing ``'+'``, or an
+optional argument with ``'?'``.  A top-level parser can contain
+sub-parsers to define subcommands that have different sets of
+switches, as in ``svn commit``, ``svn checkout``, etc.  You can
+specify an argument's type as :class:`~argparse.FileType`, which will
+automatically open files for you and understands that ``'-'`` means
+standard input or output.
+
+.. seealso::
+
+   `argparse module documentation <http://docs.python.org/dev/library/argparse.html>`__
+
+   `Upgrading optparse code to use argparse <http://docs.python.org/dev/library/argparse.html#upgrading-optparse-code>`__
+     Part of the Python documentation, describing how to convert
+     code that uses :mod:`optparse`.
+
+   :pep:`389` - argparse - New Command Line Parsing Module
+     PEP written and implemented by Steven Bethard.
+
+PEP 391: Dictionary-Based Configuration For Logging
 ====================================================
 
-XXX write this
+.. XXX not documented in library reference yet; add link here once it's added.
+
+The :mod:`logging` module is very flexible; applications can define
+a tree of logging subsystems, and each logger in this tree can filter
+out certain messages, format them differently, and direct messages to
+a varying number of handlers.
+
+All this flexibility can require a lot of configuration.  You can
+write Python statements to create objects and set their properties,
+but a complex set-up requires verbose but boring code.
+:mod:`logging` also supports a :func:`~logging.config.fileConfig`
+function that parses a file, but the file format doesn't support
+configuring filters, and it's messier to generate programmatically.
+
+Python 2.7 adds a :func:`~logging.config.dictConfig` function that
+uses a dictionary to configure logging.  There are many ways to
+produce a dictionary from different sources: construct one with code;
+parse a file containing JSON; or use a YAML parsing library if one is
+installed.
+
+The following example configures two loggers, the root logger and a
+logger named "network".   Messages sent to the root logger will be
+sent to the system log using the syslog protocol, and messages
+to the "network" logger will be written to a :file:`network.log` file
+that will be rotated once the log reaches 1Mb.
+
+::
+
+    import logging
+    import logging.config
+
+    configdict = {
+     'version': 1,    # Configuration schema in use; must be 1 for now
+     'formatters': {
+         'standard': {
+             'format': ('%(asctime)s %(name)-15s '
+                        '%(levelname)-8s %(message)s')}},
+
+     'handlers': {'netlog': {'backupCount': 10,
+                         'class': 'logging.handlers.RotatingFileHandler',
+                         'filename': '/logs/network.log',
+                         'formatter': 'standard',
+                         'level': 'INFO',
+                         'maxBytes': 1024*1024},
+                  'syslog': {'class': 'logging.handlers.SysLogHandler',
+                             'formatter': 'standard',
+                             'level': 'ERROR'}},
+
+     # Specify all the subordinate loggers
+     'loggers': {
+                 'network': {
+                             'handlers': ['netlog']
+                 }
+     },
+     # Specify properties of the root logger
+     'root': {
+              'handlers': ['syslog']
+     },
+    }
+
+    # Set up configuration
+    logging.config.dictConfig(configdict)
+
+    # As an example, log two error messages
+    logger = logging.getLogger('/')
+    logger.error('Database not found')
+
+    netlogger = logging.getLogger('network')
+    netlogger.error('Connection failed')
+
+Three smaller enhancements to the :mod:`logging` module, all
+implemented by Vinay Sajip, are:
+
+.. rev79293
+
+* The :class:`~logging.handlers.SysLogHandler` class now supports
+  syslogging over TCP.  The constructor has a *socktype* parameter
+  giving the type of socket to use, either :const:`socket.SOCK_DGRAM`
+  for UDP or :const:`socket.SOCK_STREAM` for TCP.  The default
+  protocol remains UDP.
+
+* :class:`Logger` instances gained a :meth:`getChild` method that retrieves a
+  descendant logger using a relative path.  For example,
+  once you retrieve a logger by doing ``log = getLogger('app')``,
+  calling ``log.getChild('network.listen')`` is equivalent to
+  ``getLogger('app.network.listen')``.
+
+* The :class:`LoggerAdapter` class gained a :meth:`isEnabledFor` method
+  that takes a *level* and returns whether the underlying logger would
+  process a message of that level of importance.
+
+.. seealso::
+
+   :pep:`391` - Dictionary-Based Configuration For Logging
+     PEP written and implemented by Vinay Sajip.
+
+PEP 3106: Dictionary Views
+====================================================
+
+The dictionary methods :meth:`keys`, :meth:`values`, and :meth:`items`
+are different in Python 3.x.  They return an object called a :dfn:`view`
+instead of a fully materialized list.
+
+It's not possible to change the return values of :meth:`keys`,
+:meth:`values`, and :meth:`items` in Python 2.7 because too much code
+would break.  Instead the 3.x versions were added under the new names
+:meth:`viewkeys`, :meth:`viewvalues`, and :meth:`viewitems`.
+
+::
+
+    >>> d = dict((i*10, chr(65+i)) for i in range(26))
+    >>> d
+    {0: 'A', 130: 'N', 10: 'B', 140: 'O', 20: ..., 250: 'Z'}
+    >>> d.viewkeys()
+    dict_keys([0, 130, 10, 140, 20, 150, 30, ..., 250])
+
+Views can be iterated over, but the key and item views also behave
+like sets.  The ``&`` operator performs intersection, and ``|``
+performs a union::
+
+    >>> d1 = dict((i*10, chr(65+i)) for i in range(26))
+    >>> d2 = dict((i**.5, i) for i in range(1000))
+    >>> d1.viewkeys() & d2.viewkeys()
+    set([0.0, 10.0, 20.0, 30.0])
+    >>> d1.viewkeys() | range(0, 30)
+    set([0, 1, 130, 3, 4, 5, 6, ..., 120, 250])
+
+The view keeps track of the dictionary and its contents change as the
+dictionary is modified::
+
+    >>> vk = d.viewkeys()
+    >>> vk
+    dict_keys([0, 130, 10, ..., 250])
+    >>> d[260] = '&'
+    >>> vk
+    dict_keys([0, 130, 260, 10, ..., 250])
+
+However, note that you can't add or remove keys while you're iterating
+over the view::
+
+    >>> for k in vk:
+    ...     d[k*2] = k
+    ...
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    RuntimeError: dictionary changed size during iteration
+
+You can use the view methods in Python 2.x code, and the 2to3
+converter will change them to the standard :meth:`keys`,
+:meth:`values`, and :meth:`items` methods.
+
+.. seealso::
+
+   :pep:`3106` - Revamping dict.keys(), .values() and .items()
+     PEP written by Guido van Rossum.
+     Backported to 2.7 by Alexandre Vassalotti; :issue:`1967`.
+
+
+PEP 3137: The memoryview Object
+====================================================
+
+The :class:`memoryview` object provides a view of another object's
+memory content that matches the :class:`bytes` type's interface.
+
+    >>> import string
+    >>> m = memoryview(string.letters)
+    >>> m
+    <memory at 0x37f850>
+    >>> len(m)           # Returns length of underlying object
+    52
+    >>> m[0], m[25], m[26]   # Indexing returns one byte
+    ('a', 'z', 'A')
+    >>> m2 = m[0:26]         # Slicing returns another memoryview
+    >>> m2
+    <memory at 0x37f080>
+
+The content of the view can be converted to a string of bytes or
+a list of integers:
+
+    >>> m2.tobytes()
+    'abcdefghijklmnopqrstuvwxyz'
+    >>> m2.tolist()
+    [97, 98, 99, 100, 101, 102, 103, ... 121, 122]
+    >>>
+
+:class:`memoryview` objects allow modifying the underlying object if
+it's a mutable object.
+
+    >>> m2[0] = 75
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    TypeError: cannot modify read-only memory
+    >>> b = bytearray(string.letters)  # Creating a mutable object
+    >>> b
+    bytearray(b'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ')
+    >>> mb = memoryview(b)
+    >>> mb[0] = '*'         # Assign to view, changing the bytearray.
+    >>> b[0:5]              # The bytearray has been changed.
+    bytearray(b'*bcde')
+    >>>
+
+.. seealso::
+
+   :pep:`3137` - Immutable Bytes and Mutable Buffer
+     PEP written by Guido van Rossum.
+     Implemented by Travis Oliphant, Antoine Pitrou and others.
+     Backported to 2.7 by Antoine Pitrou; :issue:`2396`.
 
-Several modules will now use :class:`OrderedDict` by default.  The
-:mod:`ConfigParser` module uses :class:`OrderedDict` for the list
-of sections and the options within a section.
-The :meth:`namedtuple._asdict` method returns an :class:`OrderedDict`
-as well.
 
 
 Other Language Changes
@@ -88,7 +617,116 @@
 
 Some smaller changes made to the core Python language are:
 
-* :meth:`str.format` method now supports automatic numbering of the replacement
+* The syntax for set literals has been backported from Python 3.x.
+  Curly brackets are used to surround the contents of the resulting
+  mutable set; set literals are
+  distinguished from dictionaries by not containing colons and values.
+  ``{}`` continues to represent an empty dictionary; use
+  ``set()`` for an empty set.
+
+    >>> {1,2,3,4,5}
+    set([1, 2, 3, 4, 5])
+    >>> set() # empty set
+    set([])
+    >>> {}    # empty dict
+    {}
+
+  Backported by Alexandre Vassalotti; :issue:`2335`.
+
+* Dictionary and set comprehensions are another feature backported from
+  3.x, generalizing list/generator comprehensions to use
+  the literal syntax for sets and dictionaries.
+
+    >>> {x: x*x for x in range(6)}
+    {0: 0, 1: 1, 2: 4, 3: 9, 4: 16, 5: 25}
+    >>> {('a'*x) for x in range(6)}
+    set(['', 'a', 'aa', 'aaa', 'aaaa', 'aaaaa'])
+
+  Backported by Alexandre Vassalotti; :issue:`2333`.
+
+* The :keyword:`with` statement can now use multiple context managers
+  in one statement.  Context managers are processed from left to right
+  and each one is treated as beginning a new :keyword:`with` statement.
+  This means that::
+
+   with A() as a, B() as b:
+       ... suite of statements ...
+
+  is equivalent to::
+
+   with A() as a:
+       with B() as b:
+           ... suite of statements ...
+
+  The :func:`contextlib.nested` function provides a very similar
+  function, so it's no longer necessary and has been deprecated.
+
+  (Proposed in http://codereview.appspot.com/53094; implemented by
+  Georg Brandl.)
+
+* Conversions between floating-point numbers and strings are
+  now correctly rounded on most platforms.  These conversions occur
+  in many different places: :func:`str` on
+  floats and complex numbers; the :class:`float` and :class:`complex`
+  constructors;
+  numeric formatting; serializing and
+  deserializing floats and complex numbers using the
+  :mod:`marshal`, :mod:`pickle`
+  and :mod:`json` modules;
+  parsing of float and imaginary literals in Python code;
+  and :class:`~decimal.Decimal`-to-float conversion.
+
+  Related to this, the :func:`repr` of a floating-point number *x*
+  now returns a result based on the shortest decimal string that's
+  guaranteed to round back to *x* under correct rounding (with
+  round-half-to-even rounding mode).  Previously it gave a string
+  based on rounding x to 17 decimal digits.
+
+  .. maybe add an example?
+
+  The rounding library responsible for this improvement works on
+  Windows and on Unix platforms using the gcc, icc, or suncc
+  compilers.  There may be a small number of platforms where correct
+  operation of this code cannot be guaranteed, so the code is not
+  used on such systems.  You can find out which code is being used
+  by checking :data:`sys.float_repr_style`,  which will be ``short``
+  if the new code is in use and ``legacy`` if it isn't.
+
+  Implemented by Eric Smith and Mark Dickinson, using David Gay's
+  :file:`dtoa.c` library; :issue:`7117`.
+
+* Conversions from long integers and regular integers to floating
+  point now round differently, returning the floating-point number
+  closest to the number.  This doesn't matter for small integers that
+  can be converted exactly, but for large numbers that will
+  unavoidably lose precision, Python 2.7 now approximates more
+  closely.  For example, Python 2.6 computed the following::
+
+    >>> n = 295147905179352891391
+    >>> float(n)
+    2.9514790517935283e+20
+    >>> n - long(float(n))
+    65535L
+
+  Python 2.7's floating-point result is larger, but much closer to the
+  true value::
+
+    >>> n = 295147905179352891391
+    >>> float(n)
+    2.9514790517935289e+20
+    >>> n - long(float(n))
+    -1L
+
+  (Implemented by Mark Dickinson; :issue:`3166`.)
+
+  Integer division is also more accurate in its rounding behaviours.  (Also
+  implemented by Mark Dickinson; :issue:`1811`.)
+
+* Implicit coercion for complex numbers has been removed; the interpreter
+  will no longer ever attempt to call a :meth:`__coerce__` method on complex
+  objects.  (Removed by Meador Inge and Mark Dickinson; :issue:`5211`.)
+
+* The :meth:`str.format` method now supports automatic numbering of the replacement
   fields.  This makes using :meth:`str.format` more closely resemble using
   ``%s`` formatting::
 
@@ -102,14 +740,35 @@
   specifier will use the next argument, and so on.  You can't mix auto-numbering
   and explicit numbering -- either number all of your specifier fields or none
   of them -- but you can mix auto-numbering and named fields, as in the second
-  example above.  (Contributed by Eric Smith; :issue`5237`.)
+  example above.  (Contributed by Eric Smith; :issue:`5237`.)
+
+  Complex numbers now correctly support usage with :func:`format`,
+  and default to being right-aligned.
+  Specifying a precision or comma-separation applies to both the real
+  and imaginary parts of the number, but a specified field width and
+  alignment is applied to the whole of the resulting ``1.5+3j``
+  output.  (Contributed by Eric Smith; :issue:`1588` and :issue:`7988`.)
+
+  The 'F' format code now always formats its output using uppercase characters,
+  so it will now produce 'INF' and 'NAN'.
+  (Contributed by Eric Smith; :issue:`3382`.)
+
+  A low-level change: the :meth:`object.__format__` method now triggers
+  a :exc:`PendingDeprecationWarning` if it's passed a format string,
+  because the :meth:`__format__` method for :class:`object` converts
+  the object to a string representation and formats that.  Previously
+  the method silently applied the format string to the string
+  representation, but that could hide mistakes in Python code.  If
+  you're supplying formatting information such as an alignment or
+  precision, presumably you're expecting the formatting to be applied
+  in some object-specific way.  (Fixed by Eric Smith; :issue:`7994`.)
 
 * The :func:`int` and :func:`long` types gained a ``bit_length``
   method that returns the number of bits necessary to represent
   its argument in binary::
 
       >>> n = 37
-      >>> bin(37)
+      >>> bin(n)
       '0b100101'
       >>> n.bit_length()
       6
@@ -121,34 +780,84 @@
 
   (Contributed by Fredrik Johansson and Victor Stinner; :issue:`3439`.)
 
-* Conversions from long integers and regular integers to floating
-  point now round differently, returning the floating-point number
-  closest to the number.  This doesn't matter for small integers that
-  can be converted exactly, but for large numbers that will
-  unavoidably lose precision, Python 2.7 will now approximate more
-  closely.  For example, Python 2.6 computed the following::
+* The :keyword:`import` statement will no longer try a relative import
+  if an absolute import (e.g. ``from .os import sep``) fails.  This
+  fixes a bug, but could possibly break certain :keyword:`import`
+  statements that were only working by accident.  (Fixed by Meador Inge;
+  :issue:`7902`.)
+
+* It's now possible for a subclass of the built-in :class:`unicode` type
+  to override the :meth:`__unicode__` method.  (Implemented by
+  Victor Stinner; :issue:`1583863`.)
 
-    >>> n = 295147905179352891391
-    >>> float(n)
-    2.9514790517935283e+20
-    >>> n - long(float(n))
-    65535L
+* The :class:`bytearray` type's :meth:`~bytearray.translate` method now accepts
+  ``None`` as its first argument.  (Fixed by Georg Brandl;
+  :issue:`4759`.)
 
-  Python 2.7's floating-point result is larger, but much closer to the
-  true value::
+  .. bytearray doesn't seem to be documented
 
-    >>> n = 295147905179352891391
-    >>> float(n)
-    2.9514790517935289e+20
-    >>> n-long(float(n)
-    ... )
-    -1L
+* When using ``@classmethod`` and ``@staticmethod`` to wrap
+  methods as class or static methods, the wrapper object now
+  exposes the wrapped function as their :attr:`__func__` attribute.
+  (Contributed by Amaury Forgeot d'Arc, after a suggestion by
+  George Sakkis; :issue:`5982`.)
+
+* When a restricted set of attributes were set using ``__slots__``,
+  deleting an unset attribute would not raise :exc:`AttributeError`
+  as you would expect.  Fixed by Benjamin Peterson; :issue:`7604`.)
+
+* Two new encodings are now supported: "cp720", used primarily for
+  Arabic text; and "cp858", a variant of CP 850 that adds the euro
+  symbol.  (CP720 contributed by Alexander Belchenko and Amaury
+  Forgeot d'Arc in :issue:`1616979`; CP858 contributed by Tim Hatch in
+  :issue:`8016`.)
+
+* The :class:`file` object will now set the :attr:`filename` attribute
+  on the :exc:`IOError` exception when trying to open a directory
+  on POSIX platforms (noted by Jan Kaliszewski; :issue:`4764`), and
+  now explicitly checks for and forbids writing to read-only file objects
+  instead of trusting the C library to catch and report the error
+  (fixed by Stefan Krah; :issue:`5677`).
+
+* The Python tokenizer now translates line endings itself, so the
+  :func:`compile` built-in function now accepts code using any
+  line-ending convention.  Additionally, it no longer requires that the
+  code end in a newline.
+
+* Extra parentheses in function definitions are illegal in Python 3.x,
+  meaning that you get a syntax error from ``def f((x)): pass``.  In
+  Python3-warning mode, Python 2.7 will now warn about this odd usage.
+  (Noted by James Lingard; :issue:`7362`.)
+
+* It's now possible to create weak references to old-style class
+  objects.  New-style classes were always weak-referenceable.  (Fixed
+  by Antoine Pitrou; :issue:`8268`.)
+
+* When a module object is garbage-collected, the module's dictionary is
+  now only cleared if no one else is holding a reference to the
+  dictionary (:issue:`7140`).
 
-  (Implemented by Mark Dickinson; :issue:`3166`.)
+.. ======================================================================
 
-* The :class:`bytearray` type's :meth:`translate` method will
-  now accept ``None`` as its first argument.  (Fixed by Georg Brandl;
-  :issue:`4759`.)
+.. _new-27-interpreter:
+
+Interpreter Changes
+-------------------------------
+
+A new environment variable, :envvar:`PYTHONWARNINGS`,
+allows controlling warnings.  It should be set to a string
+containing warning settings, equivalent to those
+used with the :option:`-W` switch, separated by commas.
+(Contributed by Brian Curtin; :issue:`7301`.)
+
+For example, the following setting will print warnings every time
+they occur, but turn warnings from the :mod:`Cookie` module into an
+error.  (The exact syntax for setting an environment variable varies
+across operating systems and shells.)
+
+::
+
+  export PYTHONWARNINGS=all,error:::Cookie:0
 
 .. ======================================================================
 
@@ -158,16 +867,26 @@
 
 Several performance enhancements have been added:
 
-* The garbage collector now performs better when many objects are
-  being allocated without deallocating any.  A full garbage collection
-  pass is only performed when the middle generation has been collected
-  10 times and when the number of survivor objects from the middle
-  generation exceeds 10% of the number of objects in the oldest
-  generation.  The second condition was added to reduce the number
-  of full garbage collections as the number of objects on the heap grows,
-  avoiding quadratic performance when allocating very many objects.
-  (Suggested by Martin von Loewis and implemented by Antoine Pitrou;
-  :issue:`4074`.)
+.. * A new :program:`configure` option, :option:`--with-computed-gotos`,
+   compiles the main bytecode interpreter loop using a new dispatch
+   mechanism that gives speedups of up to 20%, depending on the system
+   and benchmark.  The new mechanism is only supported on certain
+   compilers, such as gcc, SunPro, and icc.
+
+* A new opcode was added to perform the initial setup for
+  :keyword:`with` statements, looking up the :meth:`__enter__` and
+  :meth:`__exit__` methods.  (Contributed by Benjamin Peterson.)
+
+* The garbage collector now performs better for one common usage
+  pattern: when many objects are being allocated without deallocating
+  any of them.  This would previously take quadratic
+  time for garbage collection, but now the number of full garbage collections
+  is reduced as the number of objects on the heap grows.
+  The new logic only performs a full garbage collection pass when
+  the middle generation has been collected 10 times and when the
+  number of survivor objects from the middle generation exceeds 10% of
+  the number of objects in the oldest generation.  (Suggested by Martin
+  von L?wis and implemented by Antoine Pitrou; :issue:`4074`.)
 
 * The garbage collector tries to avoid tracking simple containers
   which can't be part of a cycle. In Python 2.7, this is now true for
@@ -178,7 +897,7 @@
   considered and traversed by the collector.
   (Contributed by Antoine Pitrou; :issue:`4688`.)
 
-* Integers are now stored internally either in base 2**15 or in base
+* Long integers are now stored internally either in base 2**15 or in base
   2**30, the base being determined at build time.  Previously, they
   were always stored in base 2**15.  Using base 2**30 gives
   significant performance improvements on 64-bit machines, but
@@ -189,7 +908,7 @@
 
   Apart from the performance improvements this change should be
   invisible to end users, with one exception: for testing and
-  debugging purposes there's a new structseq ``sys.long_info`` that
+  debugging purposes there's a new structseq :data:`sys.long_info` that
   provides information about the internal format, giving the number of
   bits per digit and the size in bytes of the C type used to store
   each digit::
@@ -210,6 +929,8 @@
   Various benchmarks show speedups of between 50% and 150% for long
   integer divisions and modulo operations.
   (Contributed by Mark Dickinson; :issue:`5512`.)
+  Bitwise operations are also significantly faster (initial patch by
+  Gregory Smith; :issue:`1087418`).
 
 * The implementation of ``%`` checks for the left-side operand being
   a Python string and special-cases it; this results in a 1-3%
@@ -221,10 +942,31 @@
   faster bytecode.  (Patch by Antoine Pitrou, back-ported to 2.7
   by Jeffrey Yasskin; :issue:`4715`.)
 
+* Converting an integer or long integer to a decimal string was made
+  faster by special-casing base 10 instead of using a generalized
+  conversion function that supports arbitrary bases.
+  (Patch by Gawain Bolton; :issue:`6713`.)
+
+* The :meth:`split`, :meth:`replace`, :meth:`rindex`,
+  :meth:`rpartition`, and :meth:`rsplit` methods of string-like types
+  (strings, Unicode strings, and :class:`bytearray` objects) now use a
+  fast reverse-search algorithm instead of a character-by-character
+  scan.  This is sometimes faster by a factor of 10.  (Added by
+  Florent Xicluna; :issue:`7462` and :issue:`7622`.)
+
+* The :mod:`pickle` and :mod:`cPickle` modules now automatically
+  intern the strings used for attribute names, reducing memory usage
+  of the objects resulting from unpickling.  (Contributed by Jake
+  McGuire; :issue:`5084`.)
+
+* The :mod:`cPickle` module now special-cases dictionaries,
+  nearly halving the time required to pickle them.
+  (Contributed by Collin Winter; :issue:`5670`.)
+
 .. ======================================================================
 
-New, Improved, and Deprecated Modules
-=====================================
+New and Improved Modules
+========================
 
 As in every release, Python's standard library received a number of
 enhancements and bug fixes.  Here's a partial list of the most notable
@@ -232,13 +974,33 @@
 :file:`Misc/NEWS` file in the source tree for a more complete list of
 changes, or look through the Subversion logs for all the details.
 
-* The :mod:`bz2` module's :class:`BZ2File` now supports the context
-  management protocol, so you can write ``with bz2.BZ2File(...) as f: ...``.
-  (Contributed by Hagen Fuerstenau; :issue:`3860`.)
-
-* New class: the :class:`Counter` class in the :mod:`collections` module is
-  useful for tallying data.  :class:`Counter` instances behave mostly
-  like dictionaries but return zero for missing keys instead of
+* The :mod:`bdb` module's base debugging class :class:`~bdb.Bdb`
+  gained a feature for skipping modules.  The constructor
+  now takes an iterable containing glob-style patterns such as
+  ``django.*``; the debugger will not step into stack frames
+  from a module that matches one of these patterns.
+  (Contributed by Maru Newby after a suggestion by
+  Senthil Kumaran; :issue:`5142`.)
+
+* The :mod:`binascii` module now supports the buffer API, so it can be
+  used with :class:`memoryview` instances and other similar buffer objects.
+  (Backported from 3.x by Florent Xicluna; :issue:`7703`.)
+
+* Updated module: the :mod:`bsddb` module has been updated from 4.7.2devel9
+  to version 4.8.4 of
+  `the pybsddb package <http://www.jcea.es/programacion/pybsddb.htm>`__.
+  The new version features better Python 3.x compatibility, various bug fixes,
+  and adds several new BerkeleyDB flags and methods.
+  (Updated by Jes?s Cea Avi?n; :issue:`8156`.  The pybsddb
+  changelog can be read at http://hg.jcea.es/pybsddb/file/tip/ChangeLog.)
+
+* The :mod:`bz2` module's :class:`~bz2.BZ2File` now supports the context
+  management protocol, so you can write ``with bz2.BZ2File(...) as f:``.
+  (Contributed by Hagen F?rstenau; :issue:`3860`.)
+
+* New class: the :class:`~collections.Counter` class in the :mod:`collections`
+  module is useful for tallying data.  :class:`~collections.Counter` instances
+  behave mostly like dictionaries but return zero for missing keys instead of
   raising a :exc:`KeyError`:
 
   .. doctest::
@@ -258,10 +1020,15 @@
      >>> c['z']
      0
 
-  There are two additional :class:`Counter` methods: :meth:`most_common`
-  returns the N most common elements and their counts, and :meth:`elements`
-  returns an iterator over the contained element, repeating each element
-  as many times as its count::
+  There are three additional :class:`~collections.Counter` methods.
+  :meth:`~collections.Counter.most_common` returns the N most common
+  elements and their counts.  :meth:`~collections.Counter.elements`
+  returns an iterator over the contained elements, repeating each
+  element as many times as its count.
+  :meth:`~collections.Counter.subtract` takes an iterable and
+  subtracts one for each element instead of adding; if the argument is
+  a dictionary or another :class:`Counter`, the counts are
+  subtracted. ::
 
     >>> c.most_common(5)
     [(' ', 6), ('e', 5), ('s', 3), ('a', 2), ('i', 2)]
@@ -270,12 +1037,30 @@
        'e', 'e', 'e', 'e', 'e', 'g', 'f', 'i', 'i',
        'h', 'h', 'm', 'l', 'l', 'o', 'n', 'p', 's',
        's', 's', 'r', 't', 't', 'x'
+    >>> c['e']
+    5
+    >>> c.subtract('very heavy on the letter e')
+    >>> c['e']    # Count is now lower
+    -1
 
   Contributed by Raymond Hettinger; :issue:`1696199`.
 
-  The :class:`namedtuple` class now has an optional *rename* parameter.
+  .. revision 79660
+
+  New class: :class:`~collections.OrderedDict` is described in the earlier
+  section :ref:`pep-0372`.
+
+  New method: The :class:`~collections.deque` data type now has a
+  :meth:`~collections.deque.count` method that returns the number of
+  contained elements equal to the supplied argument *x*, and a
+  :meth:`~collections.deque.reverse` method that reverses the elements
+  of the deque in-place.  :class:`deque` also exposes its maximum
+  length as the read-only :attr:`~collections.deque.maxlen` attribute.
+  (Both features added by Raymond Hettinger.)
+
+  The :class:`~collections.namedtuple` class now has an optional *rename* parameter.
   If *rename* is true, field names that are invalid because they've
-  been repeated or that aren't legal Python identifiers will be
+  been repeated or aren't legal Python identifiers will be
   renamed to legal names that are derived from the field's
   position within the list of fields:
 
@@ -286,30 +1071,63 @@
 
   (Added by Raymond Hettinger; :issue:`1818`.)
 
-  The :class:`deque` data type now exposes its maximum length as the
-  read-only :attr:`maxlen` attribute.  (Added by Raymond Hettinger.)
+  Finally, the :class:`~collections.Mapping` abstract base class now
+  returns :const:`NotImplemented` if a mapping is compared to
+  another type that isn't a :class:`Mapping`.
+  (Fixed by Daniel Stutzbach; :issue:`8729`.)
+
+* Constructors for the parsing classes in the :mod:`ConfigParser` module now
+  take a *allow_no_value* parameter, defaulting to false; if true,
+  options without values will be allowed.  For example::
+
+    >>> import ConfigParser, StringIO
+    >>> sample_config = """
+    ... [mysqld]
+    ... user = mysql
+    ... pid-file = /var/run/mysqld/mysqld.pid
+    ... skip-bdb
+    ... """
+    >>> config = ConfigParser.RawConfigParser(allow_no_value=True)
+    >>> config.readfp(StringIO.StringIO(sample_config))
+    >>> config.get('mysqld', 'user')
+    'mysql'
+    >>> print config.get('mysqld', 'skip-bdb')
+    None
+    >>> print config.get('mysqld', 'unknown')
+    Traceback (most recent call last):
+      ...
+    NoOptionError: No option 'unknown' in section: 'mysqld'
+
+  (Contributed by Mats Kindahl; :issue:`7005`.)
 
-* In Distutils, :func:`distutils.sdist.add_defaults` now uses
-  *package_dir* and *data_files* to create the MANIFEST file.
-  :mod:`distutils.sysconfig` will now read the :envvar:`AR`
-  environment variable.
-
-  It is no longer mandatory to store clear-text passwords in the
-  :file:`.pypirc` file when registering and uploading packages to PyPI. As long
-  as the username is present in that file, the :mod:`distutils` package will
-  prompt for the password if not present.  (Added by Tarek Ziade,
-  based on an initial contribution by Nathan Van Gheem; :issue:`4394`.)
-
-  A Distutils setup can now specify that a C extension is optional by
-  setting the *optional* option setting to true.  If this optional is
-  supplied, failure to build the extension will not abort the build
-  process, but instead simply not install the failing extension.
-  (Contributed by Georg Brandl; :issue:`5583`.)
-
-* New method: the :class:`Decimal` class gained a
-  :meth:`from_float` class method that performs an exact conversion
-  of a floating-point number to a :class:`Decimal`.
-  Note that this is an **exact** conversion that strives for the
+* Deprecated function: :func:`contextlib.nested`, which allows
+  handling more than one context manager with a single :keyword:`with`
+  statement, has been deprecated, because the :keyword:`with` statement
+  now supports multiple context managers.
+
+* The :mod:`cookielib` module now ignores cookies that have an invalid
+  version field, one that doesn't contain an integer value.  (Fixed by
+  John J. Lee; :issue:`3924`.)
+
+* The :mod:`copy` module's :func:`~copy.deepcopy` function will now
+  correctly copy bound instance methods.  (Implemented by
+  Robert Collins; :issue:`1515`.)
+
+* The :mod:`ctypes` module now always converts ``None`` to a C NULL
+  pointer for arguments declared as pointers.  (Changed by Thomas
+  Heller; :issue:`4606`.)  The underlying `libffi library
+  <http://sourceware.org/libffi/>`__ has been updated to version
+  3.0.9, containing various fixes for different platforms.  (Updated
+  by Matthias Klose; :issue:`8142`.)
+
+* New method: the :mod:`datetime` module's :class:`~datetime.timedelta` class
+  gained a :meth:`~datetime.timedelta.total_seconds` method that returns the
+  number of seconds in the duration.  (Contributed by Brian Quinlan; :issue:`5788`.)
+
+* New method: the :class:`~decimal.Decimal` class gained a
+  :meth:`~decimal.Decimal.from_float` class method that performs an exact
+  conversion of a floating-point number to a :class:`~decimal.Decimal`.
+  This exact conversion strives for the
   closest decimal approximation to the floating-point representation's value;
   the resulting decimal value will therefore still include the inaccuracy,
   if any.
@@ -317,35 +1135,186 @@
   ``Decimal('0.1000000000000000055511151231257827021181583404541015625')``.
   (Implemented by Raymond Hettinger; :issue:`4796`.)
 
-* The :class:`Fraction` class will now accept two rational numbers
-  as arguments to its constructor.
-  (Implemented by Mark Dickinson; :issue:`5812`.)
+  Comparing instances of :class:`Decimal` with floating-point
+  numbers now produces sensible results based on the numeric values
+  of the operands.  Previously such comparisons would fall back to
+  Python's default rules for comparing objects, which produced arbitrary
+  results based on their type.  Note that you still cannot combine
+  :class:`Decimal` and floating-point in other operations such as addition,
+  since you should be explicitly choosing how to convert between float and
+  :class:`Decimal`.
+  (Fixed by Mark Dickinson; :issue:`2531`.)
+
+  The constructor for :class:`~decimal.Decimal` now accepts
+  floating-point numbers (added by Raymond Hettinger; :issue:`8257`)
+  and non-European Unicode characters such as Arabic-Indic digits
+  (contributed by Mark Dickinson; :issue:`6595`).
+
+  Most of the methods of the :class:`~decimal.Context` class now accept integers
+  as well as :class:`~decimal.Decimal` instances; the only exceptions are the
+  :meth:`~decimal.Context.canonical` and :meth:`~decimal.Context.is_canonical`
+  methods.  (Patch by Juan Jos? Conti; :issue:`7633`.)
+
+  When using :class:`~decimal.Decimal` instances with a string's
+  :meth:`~str.format` method, the default alignment was previously
+  left-alignment.  This has been changed to right-alignment, which is
+  more sensible for numeric types.  (Changed by Mark Dickinson; :issue:`6857`.)
+
+  Comparisons involving a signaling NaN value (or ``sNAN``) now signal
+  :const:`InvalidOperation` instead of silently returning a true or
+  false value depending on the comparison operator.  Quiet NaN values
+  (or ``NaN``) are now hashable.  (Fixed by Mark Dickinson;
+  :issue:`7279`.)
+
+* The :mod:`difflib` module now produces output that is more
+  compatible with modern :command:`diff`/:command:`patch` tools
+  through one small change, using a tab character instead of spaces as
+  a separator in the header giving the filename.  (Fixed by Anatoly
+  Techtonik; :issue:`7585`.)
+
+* The Distutils ``sdist`` command now always regenerates the
+  :file:`MANIFEST` file, since even if the :file:`MANIFEST.in` or
+  :file:`setup.py` files haven't been modified, the user might have
+  created some new files that should be included.
+  (Fixed by Tarek Ziad?; :issue:`8688`.)
+
+* The :mod:`doctest` module's :const:`IGNORE_EXCEPTION_DETAIL` flag
+  will now ignore the name of the module containing the exception
+  being tested.  (Patch by Lennart Regebro; :issue:`7490`.)
+
+* The :mod:`email` module's :class:`~email.message.Message` class will
+  now accept a Unicode-valued payload, automatically converting the
+  payload to the encoding specified by :attr:`output_charset`.
+  (Added by R. David Murray; :issue:`1368247`.)
+
+* The :class:`~fractions.Fraction` class now accepts a single float or
+  :class:`~decimal.Decimal` instance, or two rational numbers, as
+  arguments to its constructor.  (Implemented by Mark Dickinson;
+  rationals added in :issue:`5812`, and float/decimal in
+  :issue:`8294`.)
+
+  Ordering comparisons (``<``, ``<=``, ``>``, ``>=``) between
+  fractions and complex numbers now raise a :exc:`TypeError`.
+  This fixes an oversight, making the :class:`Fraction` match the other
+  numeric types.
+
+  .. revision 79455
+
+* New class: :class:`~ftplib.FTP_TLS` in
+  the :mod:`ftplib` module provides secure FTP
+  connections using TLS encapsulation of authentication as well as
+  subsequent control and data transfers.
+  (Contributed by Giampaolo Rodola; :issue:`2054`.)
+
+  The :meth:`~ftplib.FTP.storbinary` method for binary uploads can now restart
+  uploads thanks to an added *rest* parameter (patch by Pablo Mouzo;
+  :issue:`6845`.)
+
+* New class decorator: :func:`total_ordering` in the :mod:`functools`
+  module takes a class that defines an :meth:`__eq__` method and one of
+  :meth:`__lt__`, :meth:`__le__`, :meth:`__gt__`, or :meth:`__ge__`,
+  and generates the missing comparison methods.  Since the
+  :meth:`__cmp__` method is being deprecated in Python 3.x,
+  this decorator makes it easier to define ordered classes.
+  (Added by Raymond Hettinger; :issue:`5479`.)
+
+  New function: :func:`cmp_to_key` will take an old-style comparison
+  function that expects two arguments and return a new callable that
+  can be used as the *key* parameter to functions such as
+  :func:`sorted`, :func:`min` and :func:`max`, etc.  The primary
+  intended use is to help with making code compatible with Python 3.x.
+  (Added by Raymond Hettinger.)
 
-* New function: the :mod:`gc` module's :func:`is_tracked` returns
+* New function: the :mod:`gc` module's :func:`~gc.is_tracked` returns
   true if a given instance is tracked by the garbage collector, false
   otherwise. (Contributed by Antoine Pitrou; :issue:`4688`.)
 
-* The :mod:`gzip` module's :class:`GzipFile` now supports the context
-  management protocol, so you can write ``with gzip.GzipFile(...) as f: ...``.
-  (Contributed by Hagen Fuerstenau; :issue:`3860`.)
-  It's now possible to override the modification time
+* The :mod:`gzip` module's :class:`~gzip.GzipFile` now supports the context
+  management protocol, so you can write ``with gzip.GzipFile(...) as f:``
+  (contributed by Hagen F?rstenau; :issue:`3860`), and it now implements
+  the :class:`io.BufferedIOBase` ABC, so you can wrap it with
+  :class:`io.BufferedReader` for faster processing
+  (contributed by Nir Aides; :issue:`7471`).
+  It's also now possible to override the modification time
   recorded in a gzipped file by providing an optional timestamp to
   the constructor.  (Contributed by Jacques Frechet; :issue:`4272`.)
 
-* The :class:`io.FileIO` class now raises an :exc:`OSError` when passed
+  Files in gzip format can be padded with trailing zero bytes; the
+  :mod:`gzip` module will now consume these trailing bytes.  (Fixed by
+  Tadek Pietraszek and Brian Curtin; :issue:`2846`.)
+
+* New attribute: the :mod:`hashlib` module now has an :attr:`~hashlib.hashlib.algorithms`
+  attribute containing a tuple naming the supported algorithms.
+  In Python 2.7, ``hashlib.algorithms`` contains
+  ``('md5', 'sha1', 'sha224', 'sha256', 'sha384', 'sha512')``.
+  (Contributed by Carl Chenet; :issue:`7418`.)
+
+* The default :class:`~httplib.HTTPResponse` class used by the :mod:`httplib` module now
+  supports buffering, resulting in much faster reading of HTTP responses.
+  (Contributed by Kristj?n Valur J?nsson; :issue:`4879`.)
+
+  The :class:`~httplib.HTTPConnection` and :class:`~httplib.HTTPSConnection` classes
+  now support a *source_address* parameter, a ``(host, port)`` 2-tuple
+  giving the source address that will be used for the connection.
+  (Contributed by Eldon Ziegler; :issue:`3972`.)
+
+* The :mod:`ihooks` module now supports relative imports.  Note that
+  :mod:`ihooks` is an older module for customizing imports,
+  superseded by the :mod:`imputil` module added in Python 2.0.
+  (Relative import support added by Neil Schemenauer.)
+
+  .. revision 75423
+
+* The :mod:`imaplib` module now supports IPv6 addresses.
+  (Contributed by Derek Morr; :issue:`1655`.)
+
+* New function: the :mod:`inspect` module's :func:`~inspect.getcallargs`
+  takes a callable and its positional and keyword arguments,
+  and figures out which of the callable's parameters will receive each argument,
+  returning a dictionary mapping argument names to their values.  For example::
+
+    >>> from inspect import getcallargs
+    >>> def f(a, b=1, *pos, **named):
+    ...     pass
+    >>> getcallargs(f, 1, 2, 3)
+    {'a': 1, 'b': 2, 'pos': (3,), 'named': {}}
+    >>> getcallargs(f, a=2, x=4)
+    {'a': 2, 'b': 1, 'pos': (), 'named': {'x': 4}}
+    >>> getcallargs(f)
+    Traceback (most recent call last):
+    ...
+    TypeError: f() takes at least 1 argument (0 given)
+
+  Contributed by George Sakkis; :issue:`3135`.
+
+* Updated module: The :mod:`io` library has been upgraded to the version shipped with
+  Python 3.1.  For 3.1, the I/O library was entirely rewritten in C
+  and is 2 to 20 times faster depending on the task being performed.  The
+  original Python version was renamed to the :mod:`_pyio` module.
+
+  One minor resulting change: the :class:`io.TextIOBase` class now
+  has an :attr:`errors` attribute giving the error setting
+  used for encoding and decoding errors (one of ``'strict'``, ``'replace'``,
+  ``'ignore'``).
+
+  The :class:`io.FileIO` class now raises an :exc:`OSError` when passed
   an invalid file descriptor.  (Implemented by Benjamin Peterson;
-  :issue:`4991`.)
+  :issue:`4991`.)  The :meth:`~io.IOBase.truncate` method now preserves the
+  file position; previously it would change the file position to the
+  end of the new file.  (Fixed by Pascal Chambon; :issue:`6939`.)
 
-* New function: ``itertools.compress(*data*, *selectors*)`` takes two
+* New function: ``itertools.compress(data, selectors)`` takes two
   iterators.  Elements of *data* are returned if the corresponding
   value in *selectors* is true::
 
     itertools.compress('ABCDEF', [1,0,1,0,1,1]) =>
       A, C, E, F
 
-  New function: ``itertools.combinations_with_replacement(*iter*, *r*)``
+  .. maybe here is better to use >>> list(itertools.compress(...)) instead
+
+  New function: ``itertools.combinations_with_replacement(iter, r)``
   returns all the possible *r*-length combinations of elements from the
-  iterable *iter*.  Unlike :func:`combinations`, individual elements
+  iterable *iter*.  Unlike :func:`~itertools.combinations`, individual elements
   can be repeated in the generated combinations::
 
     itertools.combinations_with_replacement('abc', 2) =>
@@ -355,45 +1324,200 @@
   Note that elements are treated as unique depending on their position
   in the input, not their actual values.
 
-  The :class:`itertools.count` function now has a *step* argument that
-  allows incrementing by values other than 1.  :func:`count` also
+  The :func:`itertools.count` function now has a *step* argument that
+  allows incrementing by values other than 1.  :func:`~itertools.count` also
   now allows keyword arguments, and using non-integer values such as
-  floats or :class:`Decimal` instances.  (Implemented by Raymond
+  floats or :class:`~decimal.Decimal` instances.  (Implemented by Raymond
   Hettinger; :issue:`5032`.)
 
-  :func:`itertools.combinations` and :func:`itertools.product` were
-  previously raising :exc:`ValueError` for values of *r* larger than
+  :func:`itertools.combinations` and :func:`itertools.product`
+  previously raised :exc:`ValueError` for values of *r* larger than
   the input iterable.  This was deemed a specification error, so they
   now return an empty iterator.  (Fixed by Raymond Hettinger; :issue:`4816`.)
 
-* The :mod:`json` module was upgraded to version 2.0.9 of the
+* Updated module: The :mod:`json` module was upgraded to version 2.0.9 of the
   simplejson package, which includes a C extension that makes
   encoding and decoding faster.
   (Contributed by Bob Ippolito; :issue:`4136`.)
 
-  To support the new :class:`OrderedDict` type, :func:`json.load`
+  To support the new :class:`collections.OrderedDict` type, :func:`json.load`
   now has an optional *object_pairs_hook* parameter that will be called
   with any object literal that decodes to a list of pairs.
   (Contributed by Raymond Hettinger; :issue:`5381`.)
 
+* The :mod:`mailbox` module's :class:`Maildir` class now records the
+  timestamp on the directories it reads, and only re-reads them if the
+  modification time has subsequently changed.  This improves
+  performance by avoiding unneeded directory scans.  (Fixed by
+  A.M. Kuchling and Antoine Pitrou; :issue:`1607951`, :issue:`6896`.)
+
+* New functions: the :mod:`math` module gained
+  :func:`~math.erf` and :func:`~math.erfc` for the error function and the complementary error function,
+  :func:`~math.expm1` which computes ``e**x - 1`` with more precision than
+  using :func:`~math.exp` and subtracting 1,
+  :func:`~math.gamma` for the Gamma function, and
+  :func:`~math.lgamma` for the natural log of the Gamma function.
+  (Contributed by Mark Dickinson and nirinA raseliarison; :issue:`3366`.)
+
 * The :mod:`multiprocessing` module's :class:`Manager*` classes
   can now be passed a callable that will be called whenever
   a subprocess is started, along with a set of arguments that will be
   passed to the callable.
   (Contributed by lekma; :issue:`5585`.)
 
+  The :class:`~multiprocessing.Pool` class, which controls a pool of worker processes,
+  now has an optional *maxtasksperchild* parameter.  Worker processes
+  will perform the specified number of tasks and then exit, causing the
+  :class:`~multiprocessing.Pool` to start a new worker.  This is useful if tasks may leak
+  memory or other resources, or if some tasks will cause the worker to
+  become very large.
+  (Contributed by Charles Cazabon; :issue:`6963`.)
+
+* The :mod:`nntplib` module now supports IPv6 addresses.
+  (Contributed by Derek Morr; :issue:`1664`.)
+
+* New functions: the :mod:`os` module wraps the following POSIX system
+  calls: :func:`~os.getresgid` and :func:`~os.getresuid`, which return the
+  real, effective, and saved GIDs and UIDs;
+  :func:`~os.setresgid` and :func:`~os.setresuid`, which set
+  real, effective, and saved GIDs and UIDs to new values;
+  :func:`~os.initgroups`, which initialize the group access list
+  for the current process.  (GID/UID functions
+  contributed by Travis H.; :issue:`6508`.  Support for initgroups added
+  by Jean-Paul Calderone; :issue:`7333`.)
+
+  The :func:`os.fork` function now re-initializes the import lock in
+  the child process; this fixes problems on Solaris when :func:`~os.fork`
+  is called from a thread.  (Fixed by Zsolt Cserna; :issue:`7242`.)
+
+* In the :mod:`os.path` module, the :func:`~os.path.normpath` and
+  :func:`~os.path.abspath` functions now preserve Unicode; if their input path
+  is a Unicode string, the return value is also a Unicode string.
+  (:meth:`~os.path.normpath` fixed by Matt Giuca in :issue:`5827`;
+  :meth:`~os.path.abspath` fixed by Ezio Melotti in :issue:`3426`.)
+
 * The :mod:`pydoc` module now has help for the various symbols that Python
   uses.  You can now do ``help('<<')`` or ``help('@')``, for example.
   (Contributed by David Laban; :issue:`4739`.)
 
-* The :mod:`re` module's :func:`split`, :func:`sub`, and :func:`subn`
+* The :mod:`re` module's :func:`~re.split`, :func:`~re.sub`, and :func:`~re.subn`
   now accept an optional *flags* argument, for consistency with the
   other functions in the module.  (Added by Gregory P. Smith.)
 
+* New function: :func:`~runpy.run_path` in the :mod:`runpy` module
+  will execute the code at a provided *path* argument.  *path* can be
+  the path of a Python source file (:file:`example.py`), a compiled
+  bytecode file (:file:`example.pyc`), a directory
+  (:file:`./package/`), or a zip archive (:file:`example.zip`).  If a
+  directory or zip path is provided, it will be added to the front of
+  ``sys.path`` and the module :mod:`__main__` will be imported.  It's
+  expected that the directory or zip contains a :file:`__main__.py`;
+  if it doesn't, some other :file:`__main__.py` might be imported from
+  a location later in ``sys.path``.  This makes more of the machinery
+  of :mod:`runpy` available to scripts that want to mimic the way
+  Python's command line processes an explicit path name.
+  (Added by Nick Coghlan; :issue:`6816`.)
+
+* New function: in the :mod:`shutil` module, :func:`~shutil.make_archive`
+  takes a filename, archive type (zip or tar-format), and a directory
+  path, and creates an archive containing the directory's contents.
+  (Added by Tarek Ziad?.)
+
+  :mod:`shutil`'s :func:`~shutil.copyfile` and :func:`~shutil.copytree`
+  functions now raise a :exc:`~shutil.SpecialFileError` exception when
+  asked to copy a named pipe.  Previously the code would treat
+  named pipes like a regular file by opening them for reading, and
+  this would block indefinitely.  (Fixed by Antoine Pitrou; :issue:`3002`.)
+
+* The :mod:`signal` module no longer re-installs the signal handler
+  unless this is truly necessary, which fixes a bug that could make it
+  impossible to catch the EINTR signal robustly.  (Fixed by
+  Charles-Francois Natali; :issue:`8354`.)
+
+* New functions: in the :mod:`site` module, three new functions
+  return various site- and user-specific paths.
+  :func:`~site.getsitepackages` returns a list containing all
+  global site-packages directories,
+  :func:`~site.getusersitepackages` returns the path of the user's
+  site-packages directory, and
+  :func:`~site.getuserbase` returns the value of the :envvar:`USER_BASE`
+  environment variable, giving the path to a directory that can be used
+  to store data.
+  (Contributed by Tarek Ziad?; :issue:`6693`.)
+
+  The :mod:`site` module now reports exceptions occurring
+  when the :mod:`sitecustomize` module is imported, and will no longer
+  catch and swallow the :exc:`KeyboardInterrupt` exception.  (Fixed by
+  Victor Stinner; :issue:`3137`.)
+
+* The :func:`~socket.create_connection` function
+  gained a *source_address* parameter, a ``(host, port)`` 2-tuple
+  giving the source address that will be used for the connection.
+  (Contributed by Eldon Ziegler; :issue:`3972`.)
+
+  The :meth:`~socket.socket.recv_into` and :meth:`~socket.socket.recvfrom_into`
+  methods will now write into objects that support the buffer API, most usefully
+  the :class:`bytearray` and :class:`memoryview` objects.  (Implemented by
+  Antoine Pitrou; :issue:`8104`.)
+
+* The :mod:`SocketServer` module's :class:`~SocketServer.TCPServer` class now
+  supports socket timeouts and disabling the Nagle algorithm.
+  The :attr:`~SocketServer.TCPServer.disable_nagle_algorithm` class attribute
+  defaults to False; if overridden to be True,
+  new request connections will have the TCP_NODELAY option set to
+  prevent buffering many small sends into a single TCP packet.
+  The :attr:`~SocketServer.TCPServer.timeout` class attribute can hold
+  a timeout in seconds that will be applied to the request socket; if
+  no request is received within that time, :meth:`handle_timeout`
+  will be called and :meth:`handle_request` will return.
+  (Contributed by Kristj?n Valur J?nsson; :issue:`6192` and :issue:`6267`.)
+
+* Updated module: the :mod:`sqlite3` module has been updated to
+  version 2.6.0 of the `pysqlite package <http://code.google.com/p/pysqlite/>`__. Version 2.6.0 includes a number of bugfixes, and adds
+  the ability to load SQLite extensions from shared libraries.
+  Call the ``enable_load_extension(True)`` method to enable extensions,
+  and then call :meth:`~sqlite3.Connection.load_extension` to load a particular shared library.
+  (Updated by Gerhard H?ring.)
+
+* The :mod:`ssl` module's :class:`ssl.SSLSocket` objects now support the
+  buffer API, which fixed a test suite failure (fix by Antoine Pitrou;
+  :issue:`7133`) and automatically set
+  OpenSSL's :cmacro:`SSL_MODE_AUTO_RETRY`, which will prevent an error
+  code being returned from :meth:`recv` operations that trigger an SSL
+  renegotiation (fix by Antoine Pitrou; :issue:`8222`).
+
+  The :func:`ssl.wrap_socket` constructor function now takes a
+  *ciphers* argument that's a string listing the encryption algorithms
+  to be allowed; the format of the string is described
+  `in the OpenSSL documentation
+  <http://www.openssl.org/docs/apps/ciphers.html#CIPHER_LIST_FORMAT>`__.
+  (Added by Antoine Pitrou; :issue:`8322`.)
+
+  Another change makes the extension load all of OpenSSL's ciphers and
+  digest algorithms so that they're all available.  Some SSL
+  certificates couldn't be verified, reporting an "unknown algorithm"
+  error.  (Reported by Beda Kosata, and fixed by Antoine Pitrou;
+  :issue:`8484`.)
+
+  The version of OpenSSL being used is now available as the module
+  attributes :data:`ssl.OPENSSL_VERSION` (a string),
+  :data:`ssl.OPENSSL_VERSION_INFO` (a 5-tuple), and
+  :data:`ssl.OPENSSL_VERSION_NUMBER` (an integer).  (Added by Antoine
+  Pitrou; :issue:`8321`.)
+
+* The :mod:`struct` module will no longer silently ignore overflow
+  errors when a value is too large for a particular integer format
+  code (one of ``bBhHiIlLqQ``); it now always raises a
+  :exc:`struct.error` exception.  (Changed by Mark Dickinson;
+  :issue:`1523`.)  The :func:`~struct.pack` function will also
+  attempt to use :meth:`__index__` to convert and pack non-integers
+  before trying the :meth:`__int__` method or reporting an error.
+  (Changed by Mark Dickinson; :issue:`8300`.)
+
 * New function: the :mod:`subprocess` module's
-  :func:`check_output` runs a command with a specified set of arguments
+  :func:`~subprocess.check_output` runs a command with a specified set of arguments
   and returns the command's output as a string when the command runs without
-  error, or raises a :exc:`CalledProcessError` exception otherwise.
+  error, or raises a :exc:`~subprocess.CalledProcessError` exception otherwise.
 
   ::
 
@@ -407,149 +1531,158 @@
 
   (Contributed by Gregory P. Smith.)
 
-* New function: :func:`is_declared_global` in the :mod:`symtable` module
+  The :mod:`subprocess` module will now retry its internal system calls
+  on receiving an :const:`EINTR` signal.  (Reported by several people; final
+  patch by Gregory P. Smith in :issue:`1068268`.)
+
+* New function: :func:`~symtable.is_declared_global` in the :mod:`symtable` module
   returns true for variables that are explicitly declared to be global,
   false for ones that are implicitly global.
   (Contributed by Jeremy Hylton.)
 
-* The ``sys.version_info`` value is now a named tuple, with attributes
-  named ``major``, ``minor``, ``micro``, ``releaselevel``, and ``serial``.
-  (Contributed by Ross Light; :issue:`4285`.)
+* The :mod:`syslog` module will now use the value of ``sys.argv[0]`` as the
+  identifier instead of the previous default value of ``'python'``.
+  (Changed by Sean Reifschneider; :issue:`8451`.)
 
-* The :mod:`threading` module's :meth:`Event.wait` method now returns
-  the internal flag on exit.  This means the method will usually
-  return true because :meth:`wait` is supposed to block until the
+* The ``sys.version_info`` value is now a named tuple, with attributes
+  named :attr:`major`, :attr:`minor`, :attr:`micro`,
+  :attr:`releaselevel`, and :attr:`serial`.  (Contributed by Ross
+  Light; :issue:`4285`.)
+
+  :func:`sys.getwindowsversion` also returns a named tuple,
+  with attributes named :attr:`major`, :attr:`minor`, :attr:`build`,
+  :attr:`platform`, :attr:`service_pack`, :attr:`service_pack_major`,
+  :attr:`service_pack_minor`, :attr:`suite_mask`, and
+  :attr:`product_type`.  (Contributed by Brian Curtin; :issue:`7766`.)
+
+* The :mod:`tarfile` module's default error handling has changed, to
+  no longer suppress fatal errors.  The default error level was previously 0,
+  which meant that errors would only result in a message being written to the
+  debug log, but because the debug log is not activated by default,
+  these errors go unnoticed.  The default error level is now 1,
+  which raises an exception if there's an error.
+  (Changed by Lars Gust?bel; :issue:`7357`.)
+
+  :mod:`tarfile` now supports filtering the :class:`~tarfile.TarInfo`
+  objects being added to a tar file.  When you call :meth:`~tarfile.TarFile.add`,
+  you may supply an optional *filter* argument
+  that's a callable.  The *filter* callable will be passed the
+  :class:`~tarfile.TarInfo` for every file being added, and can modify and return it.
+  If the callable returns ``None``, the file will be excluded from the
+  resulting archive.  This is more powerful than the existing
+  *exclude* argument, which has therefore been deprecated.
+  (Added by Lars Gust?bel; :issue:`6856`.)
+  The :class:`~tarfile.TarFile` class also now supports the context manager protocol.
+  (Added by Lars Gust?bel; :issue:`7232`.)
+
+* The :meth:`~threading.Event.wait` method of the :class:`threading.Event` class
+  now returns the internal flag on exit.  This means the method will usually
+  return true because :meth:`~threading.Event.wait` is supposed to block until the
   internal flag becomes true.  The return value will only be false if
   a timeout was provided and the operation timed out.
-  (Contributed by XXX; :issue:`1674032`.)
+  (Contributed by Tim Lesher; :issue:`1674032`.)
+
+* The Unicode database provided by the :mod:`unicodedata` module is
+  now used internally to determine which characters are numeric,
+  whitespace, or represent line breaks.  The database also
+  includes information from the :file:`Unihan.txt` data file (patch
+  by Anders Chrigstr?m and Amaury Forgeot d'Arc; :issue:`1571184`)
+  and has been updated to version 5.2.0 (updated by
+  Florent Xicluna; :issue:`8024`).
+
+* The :mod:`urlparse` module's :func:`~urlparse.urlsplit` now handles
+  unknown URL schemes in a fashion compliant with :rfc:`3986`: if the
+  URL is of the form ``"<something>://..."``, the text before the
+  ``://`` is treated as the scheme, even if it's a made-up scheme that
+  the module doesn't know about.  This change may break code that
+  worked around the old behaviour.  For example, Python 2.6.4 or 2.5
+  will return the following:
+
+    >>> import urlparse
+    >>> urlparse.urlsplit('invented://host/filename?query')
+    ('invented', '', '//host/filename?query', '', '')
+
+  Python 2.7 (and Python 2.6.5) will return:
+
+    >>> import urlparse
+    >>> urlparse.urlsplit('invented://host/filename?query')
+    ('invented', 'host', '/filename?query', '', '')
+
+  (Python 2.7 actually produces slightly different output, since it
+  returns a named tuple instead of a standard tuple.)
+
+  The :mod:`urlparse` module also supports IPv6 literal addresses as defined by
+  :rfc:`2732` (contributed by Senthil Kumaran; :issue:`2987`). ::
+
+    >>> urlparse.urlparse('http://[1080::8:800:200C:417A]/foo')
+    ParseResult(scheme='http', netloc='[1080::8:800:200C:417A]',
+                path='/foo', params='', query='', fragment='')
+
+* New class: the :class:`~weakref.WeakSet` class in the :mod:`weakref`
+  module is a set that only holds weak references to its elements; elements
+  will be removed once there are no references pointing to them.
+  (Originally implemented in Python 3.x by Raymond Hettinger, and backported
+  to 2.7 by Michael Foord.)
+
+* The ElementTree library, :mod:`xml.etree`, no longer escapes
+  ampersands and angle brackets when outputting an XML processing
+  instruction (which looks like ``<?xml-stylesheet href="#style1"?>``)
+  or comment (which looks like ``<!-- comment -->``).
+  (Patch by Neil Muller; :issue:`2746`.)
+
+* The XML-RPC client and server, provided by the :mod:`xmlrpclib` and
+  :mod:`SimpleXMLRPCServer` modules, have improved performance by
+  supporting HTTP/1.1 keep-alive and by optionally using gzip encoding
+  to compress the XML being exchanged.  The gzip compression is
+  controlled by the :attr:`encode_threshold` attribute of
+  :class:`SimpleXMLRPCRequestHandler`, which contains a size in bytes;
+  responses larger than this will be compressed.
+  (Contributed by Kristj?n Valur J?nsson; :issue:`6267`.)
+
+* The :mod:`zipfile` module's :class:`~zipfile.ZipFile` now supports the context
+  management protocol, so you can write ``with zipfile.ZipFile(...) as f:``.
+  (Contributed by Brian Curtin; :issue:`5511`.)
 
-* The :mod:`unittest` module was enhanced in several ways.
-  The progress messages will now show 'x' for expected failures
-  and 'u' for unexpected successes when run in verbose mode.
-  (Contributed by Benjamin Peterson.)
-  Test cases can raise the :exc:`SkipTest` exception to skip a test.
-  (:issue:`1034053`.)
-
-  The error messages for :meth:`assertEqual`,
-  :meth:`assertTrue`, and :meth:`assertFalse`
-  failures now provide more information.  If you set the
-  :attr:`longMessage` attribute of your :class:`TestCase` classes to
-  true, both the standard error message and any additional message you
-  provide will be printed for failures.  (Added by Michael Foord; :issue:`5663`.)
-
-  The :meth:`assertRaises` and :meth:`failUnlessRaises` methods now
-  return a context handler when called without providing a callable
-  object to run.  For example, you can write this::
-
-    with self.assertRaises(KeyError):
-        raise ValueError
-
-  (Implemented by Antoine Pitrou; :issue:`4444`.)
-
-  The methods :meth:`addCleanup` and :meth:`doCleanups` were added.
-  :meth:`addCleanup` allows you to add cleanup functions that
-  will be called unconditionally (after :meth:`setUp` if
-  :meth:`setUp` fails, otherwise after :meth:`tearDown`). This allows
-  for much simpler resource allocation and deallocation during tests.
-  :issue:`5679`
-
-  A number of new methods were added that provide more specialized
-  tests.  Many of these methods were written by Google engineers
-  for use in their test suites; Gregory P. Smith, Michael Foord, and
-  GvR worked on merging them into Python's version of :mod:`unittest`.
-
-  * :meth:`assertIsNone` and :meth:`assertIsNotNone` take one
-    expression and verify that the result is or is not ``None``.
-
-  * :meth:`assertIs` and :meth:`assertIsNot` take two values and check
-    whether the two values evaluate to the same object or not.
-    (Added by Michael Foord; :issue:`2578`.)
-
-  * :meth:`assertGreater`, :meth:`assertGreaterEqual`,
-    :meth:`assertLess`, and :meth:`assertLessEqual` compare
-    two quantities.
-
-  * :meth:`assertMultiLineEqual` compares two strings, and if they're
-    not equal, displays a helpful comparison that highlights the
-    differences in the two strings.
-
-  * :meth:`assertRegexpMatches` checks whether its first argument is a
-    string matching a regular expression provided as its second argument.
-
-  * :meth:`assertRaisesRegexp` checks whether a particular exception
-    is raised, and then also checks that the string representation of
-    the exception matches the provided regular expression.
-
-  * :meth:`assertIn` and :meth:`assertNotIn` tests whether
-    *first* is or is not in  *second*.
-
-  * :meth:`assertSameElements` tests whether two provided sequences
-    contain the same elements.
-
-  * :meth:`assertSetEqual` compares whether two sets are equal, and
-    only reports the differences between the sets in case of error.
-
-  * Similarly, :meth:`assertListEqual` and :meth:`assertTupleEqual`
-    compare the specified types and explain the differences.
-    More generally, :meth:`assertSequenceEqual` compares two sequences
-    and can optionally check whether both sequences are of a
-    particular type.
-
-  * :meth:`assertDictEqual` compares two dictionaries and reports the
-    differences.  :meth:`assertDictContainsSubset` checks whether
-    all of the key/value pairs in *first* are found in *second*.
-
-  * :meth:`assertAlmostEqual` and :meth:`assertNotAlmostEqual` short-circuit
-    (automatically pass or fail without checking decimal places) if the objects
-    are equal.
-
-  * :meth:`loadTestsFromName` properly honors the ``suiteClass`` attribute of
-    the :class:`TestLoader`. (Fixed by Mark Roddy; :issue:`6866`.)
-
-  * A new hook, :meth:`addTypeEqualityFunc` takes a type object and a
-    function.  The :meth:`assertEqual` method will use the function
-    when both of the objects being compared are of the specified type.
-    This function should compare the two objects and raise an
-    exception if they don't match; it's a good idea for the function
-    to provide additional information about why the two objects are
-    matching, much as the new sequence comparison methods do.
-
-  :func:`unittest.main` now takes an optional ``exit`` argument.
-  If False ``main`` doesn't call :func:`sys.exit` allowing it to
-  be used from the interactive interpreter. :issue:`3379`.
-
-  :class:`TestResult` has new :meth:`startTestRun` and
-  :meth:`stopTestRun` methods; called immediately before
-  and after a test run. :issue:`5728` by Robert Collins.
+  :mod:`zipfile` now also supports archiving empty directories and
+  extracts them correctly.  (Fixed by Kuba Wieczorek; :issue:`4710`.)
+  Reading files out of an archive is faster, and interleaving
+  :meth:`~zipfile.ZipFile.read` and :meth:`~zipfile.ZipFile.readline` now works correctly.
+  (Contributed by Nir Aides; :issue:`7610`.)
 
-* The :func:`is_zipfile` function in the :mod:`zipfile` module will now
-  accept a file object, in addition to the path names accepted in earlier
+  The :func:`~zipfile.is_zipfile` function now
+  accepts a file object, in addition to the path names accepted in earlier
   versions.  (Contributed by Gabriel Genellina; :issue:`4756`.)
 
-  :mod:`zipfile` now supports archiving empty directories and
-  extracts them correctly.  (Fixed by Kuba Wieczorek; :issue:`4710`.)
+  The :meth:`~zipfile.ZipFile.writestr` method now has an optional *compress_type* parameter
+  that lets you override the default compression method specified in the
+  :class:`~zipfile.ZipFile` constructor.  (Contributed by Ronald Oussoren;
+  :issue:`6003`.)
+
 
 .. ======================================================================
 .. whole new modules get described in subsections here
 
-importlib: Importing Modules
+
+.. _importlib-section:
+
+New module: importlib
 ------------------------------
 
 Python 3.1 includes the :mod:`importlib` package, a re-implementation
 of the logic underlying Python's :keyword:`import` statement.
 :mod:`importlib` is useful for implementors of Python interpreters and
-to user who wish to write new importers that can participate in the
+to users who wish to write new importers that can participate in the
 import process.  Python 2.7 doesn't contain the complete
 :mod:`importlib` package, but instead has a tiny subset that contains
-a single function, :func:`import_module`.
+a single function, :func:`~importlib.import_module`.
 
-``import_module(*name*, *package*=None)`` imports a module.  *name* is
+``import_module(name, package=None)`` imports a module.  *name* is
 a string containing the module or package's name.  It's possible to do
 relative imports by providing a string that begins with a ``.``
 character, such as ``..utils.errors``.  For relative imports, the
 *package* argument must be provided and is the name of the package that
 will be used as the anchor for
-the relative import.  :func:`import_module` both inserts the imported
+the relative import.  :func:`~importlib.import_module` both inserts the imported
 module into ``sys.modules`` and returns the module object.
 
 Here are some examples::
@@ -559,14 +1692,45 @@
     >>> anydbm
     <module 'anydbm' from '/p/python/Lib/anydbm.py'>
     >>> # Relative import
-    >>> sysconfig = import_module('..sysconfig', 'distutils.command')
-    >>> sysconfig
-    <module 'distutils.sysconfig' from '/p/python/Lib/distutils/sysconfig.pyc'>
+    >>> file_util = import_module('..file_util', 'distutils.command')
+    >>> file_util
+    <module 'distutils.file_util' from '/python/Lib/distutils/file_util.pyc'>
 
 :mod:`importlib` was implemented by Brett Cannon and introduced in
 Python 3.1.
 
 
+New module: sysconfig
+---------------------------------
+
+The :mod:`sysconfig` module has been pulled out of the Distutils
+package, becoming a new top-level module in its own right.
+:mod:`sysconfig` provides functions for getting information about
+Python's build process: compiler switches, installation paths, the
+platform name, and whether Python is running from its source
+directory.
+
+Some of the functions in the module are:
+
+* :func:`~sysconfig.get_config_var` returns variables from Python's
+  Makefile and the :file:`pyconfig.h` file.
+* :func:`~sysconfig.get_config_vars` returns a dictionary containing
+  all of the configuration variables.
+* :func:`~sysconfig.getpath` returns the configured path for
+  a particular type of module: the standard library,
+  site-specific modules, platform-specific modules, etc.
+* :func:`~sysconfig.is_python_build` returns true if you're running a
+  binary from a Python source tree, and false otherwise.
+
+Consult the :mod:`sysconfig` documentation for more details and for
+a complete list of functions.
+
+The Distutils package and :mod:`sysconfig` are now maintained by Tarek
+Ziad?, who has also started a Distutils2 package (source repository at
+http://hg.python.org/distutils2/) for developing a next-generation
+version of Distutils.
+
+
 ttk: Themed Widgets for Tk
 --------------------------
 
@@ -576,7 +1740,12 @@
 set was originally called Tile, but was renamed to Ttk (for "themed Tk")
 on being added to Tcl/Tck release 8.5.
 
-XXX write a brief discussion and an example here.
+To learn more, read the :mod:`ttk` module documentation.  You may also
+wish to read the Tcl/Tk manual page describing the
+Ttk theme engine, available at
+http://www.tcl.tk/man/tcl8.5/TkCmd/ttk_intro.htm. Some
+screenshots of the Python/Ttk code in use are at
+http://code.google.com/p/python-ttk/wiki/Screenshots.
 
 The :mod:`ttk` module was written by Guilherme Polo and added in
 :issue:`2983`.  An alternate version called ``Tile.py``, written by
@@ -584,6 +1753,279 @@
 inclusion in :issue:`2618`, but the authors argued that Guilherme
 Polo's work was more comprehensive.
 
+
+.. _unittest-section:
+
+Updated module: unittest
+---------------------------------
+
+The :mod:`unittest` module was greatly enhanced; many
+new features were added.  Most of these features were implemented
+by Michael Foord, unless otherwise noted.  The enhanced version of
+the module is downloadable separately for use with Python versions 2.4 to 2.6,
+packaged as the :mod:`unittest2` package, from
+http://pypi.python.org/pypi/unittest2.
+
+When used from the command line, the module can automatically discover
+tests.  It's not as fancy as `py.test <http://pytest.org>`__ or
+`nose <http://code.google.com/p/python-nose/>`__, but provides a simple way
+to run tests kept within a set of package directories.  For example,
+the following command will search the :file:`test/` subdirectory for
+any importable test files named ``test*.py``::
+
+   python -m unittest discover -s test
+
+Consult the :mod:`unittest` module documentation for more details.
+(Developed in :issue:`6001`.)
+
+The :func:`main` function supports some other new options:
+
+* :option:`-b` or :option:`--buffer` will buffer the standard output
+  and standard error streams during each test.  If the test passes,
+  any resulting output will be discarded; on failure, the buffered
+  output will be displayed.
+
+* :option:`-c` or :option:`--catch` will cause the control-C interrupt
+  to be handled more gracefully.  Instead of interrupting the test
+  process immediately, the currently running test will be completed
+  and then the partial results up to the interruption will be reported.
+  If you're impatient, a second press of control-C will cause an immediate
+  interruption.
+
+  This control-C handler tries to avoid causing problems when the code
+  being tested or the tests being run have defined a signal handler of
+  their own, by noticing that a signal handler was already set and
+  calling it.  If this doesn't work for you, there's a
+  :func:`removeHandler` decorator that can be used to mark tests that
+  should have the control-C handling disabled.
+
+* :option:`-f` or :option:`--failfast` makes
+  test execution stop immediately when a test fails instead of
+  continuing to execute further tests.  (Suggested by Cliff Dyer and
+  implemented by Michael Foord; :issue:`8074`.)
+
+The progress messages now show 'x' for expected failures
+and 'u' for unexpected successes when run in verbose mode.
+(Contributed by Benjamin Peterson.)
+
+Test cases can raise the :exc:`~unittest.SkipTest` exception to skip a
+test (:issue:`1034053`).
+
+The error messages for :meth:`~unittest.TestCase.assertEqual`,
+:meth:`~unittest.TestCase.assertTrue`, and :meth:`~unittest.TestCase.assertFalse`
+failures now provide more information.  If you set the
+:attr:`~unittest.TestCase.longMessage` attribute of your :class:`~unittest.TestCase` classes to
+True, both the standard error message and any additional message you
+provide will be printed for failures.  (Added by Michael Foord; :issue:`5663`.)
+
+The :meth:`~unittest.TestCase.assertRaises` method now
+returns a context handler when called without providing a callable
+object to run.  For example, you can write this::
+
+  with self.assertRaises(KeyError):
+      {}['foo']
+
+(Implemented by Antoine Pitrou; :issue:`4444`.)
+
+.. rev 78774
+
+Module- and class-level setup and teardown fixtures are now supported.
+Modules can contain :func:`~unittest.setUpModule` and :func:`~unittest.tearDownModule`
+functions.  Classes can have :meth:`~unittest.TestCase.setUpClass` and
+:meth:`~unittest.TestCase.tearDownClass` methods that must be defined as class methods
+(using ``@classmethod`` or equivalent).  These functions and
+methods are invoked when the test runner switches to a test case in a
+different module or class.
+
+The methods :meth:`~unittest.TestCase.addCleanup` and
+:meth:`~unittest.TestCase.doCleanups` were added.
+:meth:`~unittest.TestCase.addCleanup` lets you add cleanup functions that
+will be called unconditionally (after :meth:`~unittest.TestCase.setUp` if
+:meth:`~unittest.TestCase.setUp` fails, otherwise after :meth:`~unittest.TestCase.tearDown`). This allows
+for much simpler resource allocation and deallocation during tests
+(:issue:`5679`).
+
+A number of new methods were added that provide more specialized
+tests.  Many of these methods were written by Google engineers
+for use in their test suites; Gregory P. Smith, Michael Foord, and
+GvR worked on merging them into Python's version of :mod:`unittest`.
+
+* :meth:`~unittest.TestCase.assertIsNone` and :meth:`~unittest.TestCase.assertIsNotNone` take one
+  expression and verify that the result is or is not ``None``.
+
+* :meth:`~unittest.TestCase.assertIs` and :meth:`~unittest.TestCase.assertIsNot`
+  take two values and check whether the two values evaluate to the same object or not.
+  (Added by Michael Foord; :issue:`2578`.)
+
+* :meth:`~unittest.TestCase.assertIsInstance` and
+  :meth:`~unittest.TestCase.assertNotIsInstance` check whether
+  the resulting object is an instance of a particular class, or of
+  one of a tuple of classes.  (Added by Georg Brandl; :issue:`7031`.)
+
+* :meth:`~unittest.TestCase.assertGreater`, :meth:`~unittest.TestCase.assertGreaterEqual`,
+  :meth:`~unittest.TestCase.assertLess`, and :meth:`~unittest.TestCase.assertLessEqual` compare
+  two quantities.
+
+* :meth:`~unittest.TestCase.assertMultiLineEqual` compares two strings, and if they're
+  not equal, displays a helpful comparison that highlights the
+  differences in the two strings.  This comparison is now used by
+  default when Unicode strings are compared with :meth:`~unittest.TestCase.assertEqual`.
+
+* :meth:`~unittest.TestCase.assertRegexpMatches` and
+  :meth:`~unittest.TestCase.assertNotRegexpMatches` checks whether the
+  first argument is a string matching or not matching the regular
+  expression provided as the second argument (:issue:`8038`).
+
+* :meth:`~unittest.TestCase.assertRaisesRegexp` checks whether a particular exception
+  is raised, and then also checks that the string representation of
+  the exception matches the provided regular expression.
+
+* :meth:`~unittest.TestCase.assertIn` and :meth:`~unittest.TestCase.assertNotIn`
+  tests whether *first* is or is not in  *second*.
+
+* :meth:`~unittest.TestCase.assertItemsEqual` tests whether two provided sequences
+  contain the same elements.
+
+* :meth:`~unittest.TestCase.assertSetEqual` compares whether two sets are equal, and
+  only reports the differences between the sets in case of error.
+
+* Similarly, :meth:`~unittest.TestCase.assertListEqual` and :meth:`~unittest.TestCase.assertTupleEqual`
+  compare the specified types and explain any differences without necessarily
+  printing their full values; these methods are now used by default
+  when comparing lists and tuples using :meth:`~unittest.TestCase.assertEqual`.
+  More generally, :meth:`~unittest.TestCase.assertSequenceEqual` compares two sequences
+  and can optionally check whether both sequences are of a
+  particular type.
+
+* :meth:`~unittest.TestCase.assertDictEqual` compares two dictionaries and reports the
+  differences; it's now used by default when you compare two dictionaries
+  using :meth:`~unittest.TestCase.assertEqual`.  :meth:`~unittest.TestCase.assertDictContainsSubset` checks whether
+  all of the key/value pairs in *first* are found in *second*.
+
+* :meth:`~unittest.TestCase.assertAlmostEqual` and :meth:`~unittest.TestCase.assertNotAlmostEqual` test
+  whether *first* and *second* are approximately equal.  This method
+  can either round their difference to an optionally-specified number
+  of *places* (the default is 7) and compare it to zero, or require
+  the difference to be smaller than a supplied *delta* value.
+
+* :meth:`~unittest.TestLoader.loadTestsFromName` properly honors the
+  :attr:`~unittest.TestLoader.suiteClass` attribute of
+  the :class:`~unittest.TestLoader`. (Fixed by Mark Roddy; :issue:`6866`.)
+
+* A new hook lets you extend the :meth:`~unittest.TestCase.assertEqual` method to handle
+  new data types.  The :meth:`~unittest.TestCase.addTypeEqualityFunc` method takes a type
+  object and a function. The function will be used when both of the
+  objects being compared are of the specified type.  This function
+  should compare the two objects and raise an exception if they don't
+  match; it's a good idea for the function to provide additional
+  information about why the two objects aren't matching, much as the new
+  sequence comparison methods do.
+
+:func:`unittest.main` now takes an optional ``exit`` argument.  If
+False, :func:`~unittest.main` doesn't call :func:`sys.exit`, allowing
+:func:`main` to be used from the interactive interpreter.
+(Contributed by J. Pablo Fern?ndez; :issue:`3379`.)
+
+:class:`~unittest.TestResult` has new :meth:`~unittest.TestResult.startTestRun` and
+:meth:`~unittest.TestResult.stopTestRun` methods that are called immediately before
+and after a test run.  (Contributed by Robert Collins; :issue:`5728`.)
+
+With all these changes, the :file:`unittest.py` was becoming awkwardly
+large, so the module was turned into a package and the code split into
+several files (by Benjamin Peterson).  This doesn't affect how the
+module is imported or used.
+
+.. seealso::
+
+  http://www.voidspace.org.uk/python/articles/unittest2.shtml
+    Describes the new features, how to use them, and the
+    rationale for various design decisions.  (By Michael Foord.)
+
+.. _elementtree-section:
+
+Updated module: ElementTree 1.3
+---------------------------------
+
+The version of the ElementTree library included with Python was updated to
+version 1.3.  Some of the new features are:
+
+* The various parsing functions now take a *parser* keyword argument
+  giving an :class:`XMLParser` instance that will
+  be used.  This makes it possible to override the file's internal encoding::
+
+    p = ET.XMLParser(encoding='utf-8')
+    t = ET.XML("""<root/>""", parser=p)
+
+  Errors in parsing XML now raise a :exc:`ParseError` exception, whose
+  instances have a :attr:`position` attribute
+  containing a (*line*, *column*) tuple giving the location of the problem.
+
+* ElementTree's code for converting trees to a string has been
+  significantly reworked, making it roughly twice as fast in many
+  cases.  The :class:`ElementTree` :meth:`write` and :class:`Element`
+  :meth:`write` methods now have a *method* parameter that can be
+  "xml" (the default), "html", or "text".  HTML mode will output empty
+  elements as ``<empty></empty>`` instead of ``<empty/>``, and text
+  mode will skip over elements and only output the text chunks.  If
+  you set the :attr:`tag` attribute of an element to ``None`` but
+  leave its children in place, the element will be omitted when the
+  tree is written out, so you don't need to do more extensive rearrangement
+  to remove a single element.
+
+  Namespace handling has also been improved.  All ``xmlns:<whatever>``
+  declarations are now output on the root element, not scattered throughout
+  the resulting XML.  You can set the default namespace for a tree
+  by setting the :attr:`default_namespace` attribute and can
+  register new prefixes with :meth:`register_namespace`.  In XML mode,
+  you can use the true/false *xml_declaration* parameter to suppress the
+  XML declaration.
+
+* New :class:`Element` method: :meth:`extend` appends the items from a
+  sequence to the element's children.  Elements themselves behave like
+  sequences, so it's easy to move children from one element to
+  another::
+
+    from xml.etree import ElementTree as ET
+
+    t = ET.XML("""<list>
+      <item>1</item> <item>2</item>  <item>3</item>
+    </list>""")
+    new = ET.XML('<root/>')
+    new.extend(t)
+
+    # Outputs <root><item>1</item>...</root>
+    print ET.tostring(new)
+
+* New :class:`Element` method: :meth:`iter` yields the children of the
+  element as a generator.  It's also possible to write ``for child in
+  elem:`` to loop over an element's children.  The existing method
+  :meth:`getiterator` is now deprecated, as is :meth:`getchildren`
+  which constructs and returns a list of children.
+
+* New :class:`Element` method: :meth:`itertext` yields all chunks of
+  text that are descendants of the element.  For example::
+
+    t = ET.XML("""<list>
+      <item>1</item> <item>2</item>  <item>3</item>
+    </list>""")
+
+    # Outputs ['\n  ', '1', ' ', '2', '  ', '3', '\n']
+    print list(t.itertext())
+
+* Deprecated: using an element as a Boolean (i.e., ``if elem:``) would
+  return true if the element had any children, or false if there were
+  no children.  This behaviour is confusing -- ``None`` is false, but
+  so is a childless element? -- so it will now trigger a
+  :exc:`FutureWarning`.  In your code, you should be explicit: write
+  ``len(elem) != 0`` if you're interested in the number of children,
+  or ``elem is not None``.
+
+Fredrik Lundh develops ElementTree and produced the 1.3 version;
+you can read his article describing 1.3 at
+http://effbot.org/zone/elementtree-13-intro.htm.
+Florent Xicluna updated the version included with
+Python, after discussions on python-dev and in :issue:`6472`.)
+
 .. ======================================================================
 
 
@@ -592,26 +2034,222 @@
 
 Changes to Python's build process and to the C API include:
 
+* The latest release of the GNU Debugger, GDB 7, can be `scripted
+  using Python
+  <http://sourceware.org/gdb/current/onlinedocs/gdb/Python.html>`__.
+  When you begin debugging an executable program P, GDB will look for
+  a file named ``P-gdb.py`` and automatically read it.  Dave Malcolm
+  contributed a :file:`python-gdb.py` that adds a number of
+  commands useful when debugging Python itself.  For example,
+  ``py-up`` and ``py-down`` go up or down one Python stack frame,
+  which usually corresponds to several C stack frames.  ``py-print``
+  prints the value of a Python variable, and ``py-bt`` prints the
+  Python stack trace.  (Added as a result of :issue:`8032`.)
+
 * If you use the :file:`.gdbinit` file provided with Python,
-  the "pyo" macro in the 2.7 version will now work when the thread being
-  debugged doesn't hold the GIL; the macro will now acquire it before printing.
+  the "pyo" macro in the 2.7 version now works correctly when the thread being
+  debugged doesn't hold the GIL; the macro now acquires it before printing.
   (Contributed by Victor Stinner; :issue:`3632`.)
 
 * :cfunc:`Py_AddPendingCall` is now thread-safe, letting any
   worker thread submit notifications to the main Python thread.  This
   is particularly useful for asynchronous IO operations.
-  (Contributed by Kristjan Valur Jonsson; :issue:`4293`.)
+  (Contributed by Kristj?n Valur J?nsson; :issue:`4293`.)
+
+* New function: :cfunc:`PyCode_NewEmpty` creates an empty code object;
+  only the filename, function name, and first line number are required.
+  This is useful for extension modules that are attempting to
+  construct a more useful traceback stack.  Previously such
+  extensions needed to call :cfunc:`PyCode_New`, which had many
+  more arguments.  (Added by Jeffrey Yasskin.)
+
+* New function: :cfunc:`PyErr_NewExceptionWithDoc` creates a new
+  exception class, just as the existing :cfunc:`PyErr_NewException` does,
+  but takes an extra ``char *`` argument containing the docstring for the
+  new exception class.  (Added by 'lekma' on the Python bug tracker;
+  :issue:`7033`.)
+
+* New function: :cfunc:`PyFrame_GetLineNumber` takes a frame object
+  and returns the line number that the frame is currently executing.
+  Previously code would need to get the index of the bytecode
+  instruction currently executing, and then look up the line number
+  corresponding to that address.  (Added by Jeffrey Yasskin.)
+
+* New functions: :cfunc:`PyLong_AsLongAndOverflow` and
+  :cfunc:`PyLong_AsLongLongAndOverflow`  approximates a Python long
+  integer as a C :ctype:`long` or :ctype:`long long`.
+  If the number is too large to fit into
+  the output type, an *overflow* flag is set and returned to the caller.
+  (Contributed by Case Van Horsen; :issue:`7528` and :issue:`7767`.)
+
+* New function: stemming from the rewrite of string-to-float conversion,
+  a new :cfunc:`PyOS_string_to_double` function was added.  The old
+  :cfunc:`PyOS_ascii_strtod` and :cfunc:`PyOS_ascii_atof` functions
+  are now deprecated.
+
+* New function: :cfunc:`PySys_SetArgvEx` sets the value of
+  ``sys.argv`` and can optionally update ``sys.path`` to include the
+  directory containing the script named by ``sys.argv[0]`` depending
+  on the value of an *updatepath* parameter.
+
+  This function was added to close a security hole for applications
+  that embed Python.  The old function, :cfunc:`PySys_SetArgv`, would
+  always update ``sys.path``, and sometimes it would add the current
+  directory.  This meant that, if you ran an application embedding
+  Python in a directory controlled by someone else, attackers could
+  put a Trojan-horse module in the directory (say, a file named
+  :file:`os.py`) that your application would then import and run.
+
+  If you maintain a C/C++ application that embeds Python, check
+  whether you're calling :cfunc:`PySys_SetArgv` and carefully consider
+  whether the application should be using :cfunc:`PySys_SetArgvEx`
+  with *updatepath* set to false.
+
+  Security issue reported as `CVE-2008-5983
+  <http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2008-5983>`_;
+  discussed in :issue:`5753`, and fixed by Antoine Pitrou.
+
+* New macros: the Python header files now define the following macros:
+  :cmacro:`Py_ISALNUM`,
+  :cmacro:`Py_ISALPHA`,
+  :cmacro:`Py_ISDIGIT`,
+  :cmacro:`Py_ISLOWER`,
+  :cmacro:`Py_ISSPACE`,
+  :cmacro:`Py_ISUPPER`,
+  :cmacro:`Py_ISXDIGIT`,
+  and :cmacro:`Py_TOLOWER`, :cmacro:`Py_TOUPPER`.
+  All of these functions are analogous to the C
+  standard macros for classifying characters, but ignore the current
+  locale setting, because in
+  several places Python needs to analyze characters in a
+  locale-independent way.  (Added by Eric Smith;
+  :issue:`5793`.)
+
+  .. XXX these macros don't seem to be described in the c-api docs.
+
+* Removed function: :cmacro:`PyEval_CallObject` is now only available
+  as a macro.  A function version was being kept around to preserve
+  ABI linking compatibility, but that was in 1997; it can certainly be
+  deleted by now.  (Removed by Antoine Pitrou; :issue:`8276`.)
+
+* New format codes: the :cfunc:`PyFormat_FromString`,
+  :cfunc:`PyFormat_FromStringV`, and :cfunc:`PyErr_Format` functions now
+  accept ``%lld`` and ``%llu`` format codes for displaying
+  C's :ctype:`long long` types.
+  (Contributed by Mark Dickinson; :issue:`7228`.)
+
+* The complicated interaction between threads and process forking has
+  been changed.  Previously, the child process created by
+  :func:`os.fork` might fail because the child is created with only a
+  single thread running, the thread performing the :func:`os.fork`.
+  If other threads were holding a lock, such as Python's import lock,
+  when the fork was performed, the lock would still be marked as
+  "held" in the new process.  But in the child process nothing would
+  ever release the lock, since the other threads weren't replicated,
+  and the child process would no longer be able to perform imports.
+
+  Python 2.7 acquires the import lock before performing an
+  :func:`os.fork`, and will also clean up any locks created using the
+  :mod:`threading` module.  C extension modules that have internal
+  locks, or that call :cfunc:`fork()` themselves, will not benefit
+  from this clean-up.
+
+  (Fixed by Thomas Wouters; :issue:`1590864`.)
+
+* The :cfunc:`Py_Finalize` function now calls the internal
+  :func:`threading._shutdown` function; this prevents some exceptions from
+  being raised when an interpreter shuts down.
+  (Patch by Adam Olsen; :issue:`1722344`.)
+
+* When using the :ctype:`PyMemberDef` structure to define attributes
+  of a type, Python will no longer let you try to delete or set a
+  :const:`T_STRING_INPLACE` attribute.
+
+  .. rev 79644
 
 * Global symbols defined by the :mod:`ctypes` module are now prefixed
-  with ``Py`, or with ``_ctypes``.  (Implemented by Thomas
+  with ``Py``, or with ``_ctypes``.  (Implemented by Thomas
   Heller; :issue:`3102`.)
 
+* New configure option: the :option:`--with-system-expat` switch allows
+  building the :mod:`pyexpat` module to use the system Expat library.
+  (Contributed by Arfrever Frehtes Taifersar Arahesis; :issue:`7609`.)
+
+* New configure option: the
+  :option:`--with-valgrind` option will now disable the pymalloc
+  allocator, which is difficult for the Valgrind memory-error detector
+  to analyze correctly.
+  Valgrind will therefore be better at detecting memory leaks and
+  overruns. (Contributed by James Henstridge; :issue:`2422`.)
+
+* New configure option: you can now supply an empty string to
+  :option:`--with-dbmliborder=` in order to disable all of the various
+  DBM modules.  (Added by Arfrever Frehtes Taifersar Arahesis;
+  :issue:`6491`.)
+
 * The :program:`configure` script now checks for floating-point rounding bugs
   on certain 32-bit Intel chips and defines a :cmacro:`X87_DOUBLE_ROUNDING`
   preprocessor definition.  No code currently uses this definition,
   but it's available if anyone wishes to use it.
   (Added by Mark Dickinson; :issue:`2937`.)
 
+  :program:`configure` also now sets a :envvar:`LDCXXSHARED` Makefile
+  variable for supporting C++ linking.  (Contributed by Arfrever
+  Frehtes Taifersar Arahesis; :issue:`1222585`.)
+
+* The build process now creates the necessary files for pkg-config
+  support.  (Contributed by Clinton Roy; :issue:`3585`.)
+
+* The build process now supports Subversion 1.7.  (Contributed by
+  Arfrever Frehtes Taifersar Arahesis; :issue:`6094`.)
+
+
+.. _whatsnew27-capsules:
+
+Capsules
+-------------------
+
+Python 3.1 adds a new C datatype, :ctype:`PyCapsule`, for providing a
+C API to an extension module.  A capsule is essentially the holder of
+a C ``void *`` pointer, and is made available as a module attribute; for
+example, the :mod:`socket` module's API is exposed as ``socket.CAPI``,
+and :mod:`unicodedata` exposes ``ucnhash_CAPI``.  Other extensions
+can import the module, access its dictionary to get the capsule
+object, and then get the ``void *`` pointer, which will usually point
+to an array of pointers to the module's various API functions.
+
+There is an existing data type already used for this,
+:ctype:`PyCObject`, but it doesn't provide type safety.  Evil code
+written in pure Python could cause a segmentation fault by taking a
+:ctype:`PyCObject` from module A and somehow substituting it for the
+:ctype:`PyCObject` in module B.   Capsules know their own name,
+and getting the pointer requires providing the name::
+
+   void *vtable;
+
+   if (!PyCapsule_IsValid(capsule, "mymodule.CAPI") {
+           PyErr_SetString(PyExc_ValueError, "argument type invalid");
+           return NULL;
+   }
+
+   vtable = PyCapsule_GetPointer(capsule, "mymodule.CAPI");
+
+You are assured that ``vtable`` points to whatever you're expecting.
+If a different capsule was passed in, :cfunc:`PyCapsule_IsValid` would
+detect the mismatched name and return false.  Refer to
+:ref:`using-capsules` for more information on using these objects.
+
+Python 2.7 now uses capsules internally to provide various
+extension-module APIs, but the :cfunc:`PyCObject_AsVoidPtr` was
+modified to handle capsules, preserving compile-time compatibility
+with the :ctype:`CObject` interface.  Use of
+:cfunc:`PyCObject_AsVoidPtr` will signal a
+:exc:`PendingDeprecationWarning`, which is silent by default.
+
+Implemented in Python 3.1 and backported to 2.7 by Larry Hastings;
+discussed in :issue:`5630`.
+
+
 .. ======================================================================
 
 Port-Specific Changes: Windows
@@ -624,24 +2262,65 @@
   and :data:`LIBRARIES_ASSEMBLY_NAME_PREFIX`.
   (Contributed by David Cournapeau; :issue:`4365`.)
 
+* The :mod:`_winreg` module for accessing the registry now implements
+  the :func:`CreateKeyEx` and :func:`DeleteKeyEx` functions, extended
+  versions of previously-supported functions that take several extra
+  arguments.  The :func:`DisableReflectionKey`,
+  :func:`EnableReflectionKey`, and :func:`QueryReflectionKey` were also
+  tested and documented.
+  (Implemented by Brian Curtin: :issue:`7347`.)
+
 * The new :cfunc:`_beginthreadex` API is used to start threads, and
   the native thread-local storage functions are now used.
-  (Contributed by Kristjan Valur Jonsson; :issue:`3582`.)
+  (Contributed by Kristj?n Valur J?nsson; :issue:`3582`.)
+
+* The :func:`os.kill` function now works on Windows.  The signal value
+  can be the constants :const:`CTRL_C_EVENT`,
+  :const:`CTRL_BREAK_EVENT`, or any integer.  The first two constants
+  will send Control-C and Control-Break keystroke events to
+  subprocesses; any other value will use the :cfunc:`TerminateProcess`
+  API.  (Contributed by Miki Tebeka; :issue:`1220212`.)
+
+* The :func:`os.listdir` function now correctly fails
+  for an empty path.  (Fixed by Hirokazu Yamamoto; :issue:`5913`.)
+
+* The :mod:`mimelib` module will now read the MIME database from
+  the Windows registry when initializing.
+  (Patch by Gabriel Genellina; :issue:`4969`.)
 
 .. ======================================================================
 
 Port-Specific Changes: Mac OS X
 -----------------------------------
 
-* The ``/Library/Python/2.7/site-packages`` is now appended to
+* The path ``/Library/Python/2.7/site-packages`` is now appended to
   ``sys.path``, in order to share added packages between the system
   installation and a user-installed copy of the same version.
   (Changed by Ronald Oussoren; :issue:`4865`.)
 
+Port-Specific Changes: FreeBSD
+-----------------------------------
+
+* FreeBSD 7.1's :const:`SO_SETFIB` constant, used with
+  :func:`~socket.getsockopt`/:func:`~socket.setsockopt` to select an
+  alternate routing table, is now available in the :mod:`socket`
+  module.  (Added by Kyle VanderBeek; :issue:`8235`.)
 
 Other Changes and Fixes
 =======================
 
+* Two benchmark scripts, :file:`iobench` and :file:`ccbench`, were
+  added to the :file:`Tools` directory.  :file:`iobench` measures the
+  speed of the built-in file I/O objects returned by :func:`open`
+  while performing various operations, and :file:`ccbench` is a
+  concurrency benchmark that tries to measure computing throughput,
+  thread switching latency, and IO processing bandwidth when
+  performing several tasks using a varying number of threads.
+
+* The :file:`Tools/i18n/msgfmt.py` script now understands plural
+  forms in :file:`.po` files.  (Fixed by Martin von L?wis;
+  :issue:`5464`.)
+
 * When importing a module from a :file:`.pyc` or :file:`.pyo` file
   with an existing :file:`.py` counterpart, the :attr:`co_filename`
   attributes of the resulting code objects are overwritten when the
@@ -652,9 +2331,22 @@
 * The :file:`regrtest.py` script now takes a :option:`--randseed=`
   switch that takes an integer that will be used as the random seed
   for the :option:`-r` option that executes tests in random order.
-  The :option:`-r` option also now reports the seed that was used
+  The :option:`-r` option also reports the seed that was used
   (Added by Collin Winter.)
 
+* Another :file:`regrtest.py` switch is :option:`-j`, which
+  takes an integer specifying how many tests run in parallel. This
+  allows reducing the total runtime on multi-core machines.
+  This option is compatible with several other options, including the
+  :option:`-R` switch which is known to produce long runtimes.
+  (Added by Antoine Pitrou, :issue:`6152`.)  This can also be used
+  with a new :option:`-F` switch that runs selected tests in a loop
+  until they fail.  (Added by Antoine Pitrou; :issue:`7312`.)
+
+* When executed as a script, the :file:`py_compile.py` module now
+  accepts ``'-'`` as an argument, which will read standard input for
+  the list of filenames to be compiled.  (Contributed by Piotr
+  O?arowski; :issue:`8233`.)
 
 .. ======================================================================
 
@@ -664,12 +2356,114 @@
 This section lists previously described changes and other bugfixes
 that may require changes to your code:
 
+* The :func:`range` function processes its arguments more
+  consistently; it will now call :meth:`__int__` on non-float,
+  non-integer arguments that are supplied to it.  (Fixed by Alexander
+  Belopolsky; :issue:`1533`.)
+
+* The string :meth:`format` method changed the default precision used
+  for floating-point and complex numbers from 6 decimal
+  places to 12, which matches the precision used by :func:`str`.
+  (Changed by Eric Smith; :issue:`5920`.)
+
 * Because of an optimization for the :keyword:`with` statement, the special
   methods :meth:`__enter__` and :meth:`__exit__` must belong to the object's
   type, and cannot be directly attached to the object's instance.  This
   affects new-style classes (derived from :class:`object`) and C extension
   types.  (:issue:`6101`.)
 
+* Due to a bug in Python 2.6, the *exc_value* parameter to
+  :meth:`__exit__` methods was often the string representation of the
+  exception, not an instance.  This was fixed in 2.7, so *exc_value*
+  will be an instance as expected.  (Fixed by Florent Xicluna;
+  :issue:`7853`.)
+
+* When a restricted set of attributes were set using ``__slots__``,
+  deleting an unset attribute would not raise :exc:`AttributeError`
+  as you would expect.  Fixed by Benjamin Peterson; :issue:`7604`.)
+
+In the standard library:
+
+* Operations with :class:`datetime` instances that resulted in a year
+  falling outside the supported range didn't always raise
+  :exc:`OverflowError`.  Such errors are now checked more carefully
+  and will now raise the exception. (Reported by Mark Leander, patch
+  by Anand B. Pillai and Alexander Belopolsky; :issue:`7150`.)
+
+* When using :class:`Decimal` instances with a string's
+  :meth:`format` method, the default alignment was previously
+  left-alignment.  This has been changed to right-alignment, which might
+  change the output of your programs.
+  (Changed by Mark Dickinson; :issue:`6857`.)
+
+  Comparisons involving a signaling NaN value (or ``sNAN``) now signal
+  :const:`InvalidOperation` instead of silently returning a true or
+  false value depending on the comparison operator.  Quiet NaN values
+  (or ``NaN``) are now hashable.  (Fixed by Mark Dickinson;
+  :issue:`7279`.)
+
+* The ElementTree library, :mod:`xml.etree`, no longer escapes
+  ampersands and angle brackets when outputting an XML processing
+  instruction (which looks like `<?xml-stylesheet href="#style1"?>`)
+  or comment (which looks like `<!-- comment -->`).
+  (Patch by Neil Muller; :issue:`2746`.)
+
+* The :meth:`readline` method of :class:`StringIO` objects now does
+  nothing when a negative length is requested, as other file-like
+  objects do.  (:issue:`7348`).
+
+* The :mod:`syslog` module will now use the value of ``sys.argv[0]`` as the
+  identifier instead of the previous default value of ``'python'``.
+  (Changed by Sean Reifschneider; :issue:`8451`.)
+
+* The :mod:`tarfile` module's default error handling has changed, to
+  no longer suppress fatal errors.  The default error level was previously 0,
+  which meant that errors would only result in a message being written to the
+  debug log, but because the debug log is not activated by default,
+  these errors go unnoticed.  The default error level is now 1,
+  which raises an exception if there's an error.
+  (Changed by Lars Gust?bel; :issue:`7357`.)
+
+* The :mod:`urlparse` module's :func:`~urlparse.urlsplit` now handles
+  unknown URL schemes in a fashion compliant with :rfc:`3986`: if the
+  URL is of the form ``"<something>://..."``, the text before the
+  ``://`` is treated as the scheme, even if it's a made-up scheme that
+  the module doesn't know about.  This change may break code that
+  worked around the old behaviour.  For example, Python 2.6.4 or 2.5
+  will return the following:
+
+    >>> import urlparse
+    >>> urlparse.urlsplit('invented://host/filename?query')
+    ('invented', '', '//host/filename?query', '', '')
+
+  Python 2.7 (and Python 2.6.5) will return:
+
+    >>> import urlparse
+    >>> urlparse.urlsplit('invented://host/filename?query')
+    ('invented', 'host', '/filename?query', '', '')
+
+  (Python 2.7 actually produces slightly different output, since it
+  returns a named tuple instead of a standard tuple.)
+
+For C extensions:
+
+* C extensions that use integer format codes with the ``PyArg_Parse*``
+  family of functions will now raise a :exc:`TypeError` exception
+  instead of triggering a :exc:`DeprecationWarning` (:issue:`5080`).
+
+* Use the new :cfunc:`PyOS_string_to_double` function instead of the old
+  :cfunc:`PyOS_ascii_strtod` and :cfunc:`PyOS_ascii_atof` functions,
+  which are now deprecated.
+
+For applications that embed Python:
+
+* The :cfunc:`PySys_SetArgvEx` function was added, letting
+  applications close a security hole when the existing
+  :cfunc:`PySys_SetArgv` function was used.  Check whether you're
+  calling :cfunc:`PySys_SetArgv` and carefully consider whether the
+  application should be using :cfunc:`PySys_SetArgvEx` with
+  *updatepath* set to false.
+
 .. ======================================================================
 
 
@@ -680,5 +2474,6 @@
 
 The author would like to thank the following people for offering
 suggestions, corrections and assistance with various drafts of this
-article: no one yet.
+article: Nick Coghlan, Philip Jenvey, Ryan Lovett, R. David Murray,
+Hugh Secker-Walker.
 

Modified: python/branches/release31-maint/Lib/urllib/parse.py
==============================================================================
--- python/branches/release31-maint/Lib/urllib/parse.py	(original)
+++ python/branches/release31-maint/Lib/urllib/parse.py	Sun Jul 11 12:41:07 2010
@@ -5,21 +5,22 @@
 RFC 3986 (STD66): "Uniform Resource Identifiers" by T. Berners-Lee, R. Fielding
 and L.  Masinter, January 2005.
 
-RFC2396:  "Uniform Resource Identifiers (URI)": Generic Syntax by T.
+RFC 2396:  "Uniform Resource Identifiers (URI)": Generic Syntax by T.
 Berners-Lee, R. Fielding, and L. Masinter, August 1998.
 
-RFC2368: "The mailto URL scheme", by P.Hoffman , L Masinter, J. Zwinski, July 1998.
+RFC 2368: "The mailto URL scheme", by P.Hoffman , L Masinter, J. Zwinski, July 1998.
 
 RFC 1808: "Relative Uniform Resource Locators", by R. Fielding, UC Irvine, June
 1995.
 
-RFC1738: "Uniform Resource Locators (URL)" by T. Berners-Lee, L. Masinter, M.
+RFC 1738: "Uniform Resource Locators (URL)" by T. Berners-Lee, L. Masinter, M.
 McCahill, December 1994
 
-RFC 3986 is considered the current standard and any changes to urlparse module
-should conform to this.  urlparse module is not entirely compliant with this.
-The defacto scenarios of parsing are considered sometimes and for backward
-compatiblity purposes, older RFC uses of parsing are retained. The testcases in
+RFC 3986 is considered the current standard and any future changes to
+urlparse module should conform with it.  The urlparse module is
+currently not entirely compliant with this RFC due to defacto
+scenarios for parsing, and for backward compatibility purposes, some
+parsing quirks from older RFCs are retained. The testcases in
 test_urlparse.py provides a good indicator of parsing behavior.
 """
 

Modified: python/branches/release31-maint/Modules/datetimemodule.c
==============================================================================
--- python/branches/release31-maint/Modules/datetimemodule.c	(original)
+++ python/branches/release31-maint/Modules/datetimemodule.c	Sun Jul 11 12:41:07 2010
@@ -2941,7 +2941,8 @@
      PyDoc_STR("datetime -> DST offset in minutes east of UTC.")},
 
     {"fromutc",         (PyCFunction)tzinfo_fromutc,            METH_O,
-     PyDoc_STR("datetime in UTC -> datetime in local time.")},
+     PyDoc_STR("datetime -> timedelta showing offset from UTC, negative "
+           "values indicating West of UTC")},
 
     {"__reduce__",  (PyCFunction)tzinfo_reduce,             METH_NOARGS,
      PyDoc_STR("-> (cls, state)")},

From python-checkins at python.org  Sun Jul 11 12:41:45 2010
From: python-checkins at python.org (georg.brandl)
Date: Sun, 11 Jul 2010 12:41:45 +0200 (CEST)
Subject: [Python-checkins] r82809 - in
	python/branches/release31-maint/Doc/whatsnew: 2.7.rst index.rst
Message-ID: <20100711104145.1572BEE9D3@mail.python.org>

Author: georg.brandl
Date: Sun Jul 11 12:41:44 2010
New Revision: 82809

Log:
2.7 came after 3.1, so its whatsnew document will confuse users who expect 2.7 features in 3.1.

Removed:
   python/branches/release31-maint/Doc/whatsnew/2.7.rst
Modified:
   python/branches/release31-maint/Doc/whatsnew/index.rst

Deleted: python/branches/release31-maint/Doc/whatsnew/2.7.rst
==============================================================================
--- python/branches/release31-maint/Doc/whatsnew/2.7.rst	Sun Jul 11 12:41:44 2010
+++ (empty file)
@@ -1,2479 +0,0 @@
-****************************
-  What's New in Python 2.7
-****************************
-
-:Author: A.M. Kuchling (amk at amk.ca)
-:Release: |release|
-:Date: |today|
-
-..  hyperlink all the methods & functions.
-
-.. T_STRING_INPLACE not described in main docs
-.. "Format String Syntax" in string.rst could use many more examples.
-
-.. $Id$
-   Rules for maintenance:
-
-   * Anyone can add text to this document.  Do not spend very much time
-   on the wording of your changes, because your text will probably
-   get rewritten to some degree.
-
-   * The maintainer will go through Misc/NEWS periodically and add
-   changes; it's therefore more important to add your changes to
-   Misc/NEWS than to this file.
-
-   * This is not a complete list of every single change; completeness
-   is the purpose of Misc/NEWS.  Some changes I consider too small
-   or esoteric to include.  If such a change is added to the text,
-   I'll just remove it.  (This is another reason you shouldn't spend
-   too much time on writing your addition.)
-
-   * If you want to draw your new text to the attention of the
-   maintainer, add 'XXX' to the beginning of the paragraph or
-   section.
-
-   * It's OK to just add a fragmentary note about a change.  For
-   example: "XXX Describe the transmogrify() function added to the
-   socket module."  The maintainer will research the change and
-   write the necessary text.
-
-   * You can comment out your additions if you like, but it's not
-   necessary (especially when a final release is some months away).
-
-   * Credit the author of a patch or bugfix.  Just the name is
-   sufficient; the e-mail address isn't necessary.
-
-   * It's helpful to add the bug/patch number in a parenthetical comment.
-
-   XXX Describe the transmogrify() function added to the socket
-   module.
-   (Contributed by P.Y. Developer; :issue:`12345`.)
-
-   This saves the maintainer some effort going through the SVN logs
-   when researching a change.
-
-This article explains the new features in Python 2.7.  The final
-release of 2.7 is currently scheduled for July 2010; the detailed
-schedule is described in :pep:`373`.
-
-Numeric handling has been improved in many ways, for both
-floating-point numbers and for the :class:`Decimal` class.  There are
-some useful additions to the standard library, such as a greatly
-enhanced :mod:`unittest` module, the :mod:`argparse` module for
-parsing command-line options, convenient ordered-dictionary and
-:class:`Counter` classes in the :mod:`collections` module, and many
-other improvements.
-
-Python 2.7 is planned to be the last of the 2.x releases, so we worked
-on making it a good release for the long term.  To help with porting
-to Python 3, several new features from the Python 3.x series have been
-included in 2.7.
-
-This article doesn't attempt to provide a complete specification of
-the new features, but instead provides a convenient overview.  For
-full details, you should refer to the documentation for Python 2.7 at
-http://docs.python.org. If you want to understand the rationale for
-the design and implementation, refer to the PEP for a particular new
-feature or the issue on http://bugs.python.org in which a change was
-discussed.  Whenever possible, "What's New in Python" links to the
-bug/patch item for each change.
-
-.. _whatsnew27-python31:
-
-The Future for Python 2.x
-=========================
-
-Python 2.7 is intended to be the last major release in the 2.x series.
-The Python maintainers are planning to focus their future efforts on
-the Python 3.x series.
-
-This means that 2.7 will remain in place for a long time, running
-production systems that have not been ported to Python 3.x.
-Two consequences of the long-term significance of 2.7 are:
-
-* It's very likely the 2.7 release will have a longer period of
-  maintenance compared to earlier 2.x versions.  Python 2.7 will
-  continue to be maintained while the transition to 3.x continues, and
-  the developers are planning to support Python 2.7 with bug-fix
-  releases beyond the typical two years.
-
-* A policy decision was made to silence warnings only of interest to
-  developers.  :exc:`DeprecationWarning` and its
-  descendants are now ignored unless otherwise requested, preventing
-  users from seeing warnings triggered by an application.  This change
-  was also made in the branch that will become Python 3.2. (Discussed
-  on stdlib-sig and carried out in :issue:`7319`.)
-
-  In previous releases, :exc:`DeprecationWarning` messages were
-  enabled by default, providing Python developers with a clear
-  indication of where their code may break in a future major version
-  of Python.
-
-  However, there are increasingly many users of Python-based
-  applications who are not directly involved in the development of
-  those applications.  :exc:`DeprecationWarning` messages are
-  irrelevant to such users, making them worry about an application
-  that's actually working correctly and burdening application developers
-  with responding to these concerns.
-
-  You can re-enable display of :exc:`DeprecationWarning` messages by
-  running Python with the :option:`-Wdefault` (short form:
-  :option:`-Wd`) switch, or by setting the :envvar:`PYTHONWARNINGS`
-  environment variable to ``"default"`` (or ``"d"``) before running
-  Python.  Python code can also re-enable them
-  by calling ``warnings.simplefilter('default')``.
-
-
-Python 3.1 Features
-=======================
-
-Much as Python 2.6 incorporated features from Python 3.0,
-version 2.7 incorporates some of the new features
-in Python 3.1.  The 2.x series continues to provide tools
-for migrating to the 3.x series.
-
-A partial list of 3.1 features that were backported to 2.7:
-
-* The syntax for set literals (``{1,2,3}`` is a mutable set).
-* Dictionary and set comprehensions (``{ i: i*2 for i in range(3)}``).
-* Multiple context managers in a single :keyword:`with` statement.
-* A new version of the :mod:`io` library, rewritten in C for performance.
-* The ordered-dictionary type described in :ref:`pep-0372`.
-* The new ``","`` format specifier described in :ref:`pep-0378`.
-* The :class:`memoryview` object.
-* A small subset of the :mod:`importlib` module,
-  `described below <#importlib-section>`__.
-* Float-to-string and string-to-float conversions now round their
-  results more correctly, and :func:`repr` of a floating-point
-  number *x* returns a result that's guaranteed to round back to the
-  same number when converted back to a float.
-* The :ctype:`PyCapsule` type, used to provide a C API for extension modules.
-* The :cfunc:`PyLong_AsLongAndOverflow` C API function.
-
-Other new Python3-mode warnings include:
-
-* :func:`operator.isCallable` and :func:`operator.sequenceIncludes`,
-  which are not supported in 3.x, now trigger warnings.
-* The :option:`-3` switch now automatically
-  enables the :option:`-Qwarn` switch that causes warnings
-  about using classic division with integers and long integers.
-
-
-
-.. ========================================================================
-.. Large, PEP-level features and changes should be described here.
-.. ========================================================================
-
-.. _pep-0372:
-
-PEP 372: Adding an Ordered Dictionary to collections
-====================================================
-
-Regular Python dictionaries iterate over key/value pairs in arbitrary order.
-Over the years, a number of authors have written alternative implementations
-that remember the order that the keys were originally inserted.  Based on
-the experiences from those implementations, 2.7 introduces a new
-:class:`~collections.OrderedDict` class in the :mod:`collections` module.
-
-The :class:`~collections.OrderedDict` API provides the same interface as regular
-dictionaries but iterates over keys and values in a guaranteed order
-depending on when a key was first inserted::
-
-    >>> from collections import OrderedDict
-    >>> d = OrderedDict([('first', 1),
-    ...                  ('second', 2),
-    ...                  ('third', 3)])
-    >>> d.items()
-    [('first', 1), ('second', 2), ('third', 3)]
-
-If a new entry overwrites an existing entry, the original insertion
-position is left unchanged::
-
-    >>> d['second'] = 4
-    >>> d.items()
-    [('first', 1), ('second', 4), ('third', 3)]
-
-Deleting an entry and reinserting it will move it to the end::
-
-    >>> del d['second']
-    >>> d['second'] = 5
-    >>> d.items()
-    [('first', 1), ('third', 3), ('second', 5)]
-
-The :meth:`~collections.OrderedDict.popitem` method has an optional *last*
-argument that defaults to True.  If *last* is True, the most recently
-added key is returned and removed; if it's False, the
-oldest key is selected::
-
-    >>> od = OrderedDict([(x,0) for x in range(20)])
-    >>> od.popitem()
-    (19, 0)
-    >>> od.popitem()
-    (18, 0)
-    >>> od.popitem(last=False)
-    (0, 0)
-    >>> od.popitem(last=False)
-    (1, 0)
-
-Comparing two ordered dictionaries checks both the keys and values,
-and requires that the insertion order was the same::
-
-    >>> od1 = OrderedDict([('first', 1),
-    ...                    ('second', 2),
-    ...                    ('third', 3)])
-    >>> od2 = OrderedDict([('third', 3),
-    ...                    ('first', 1),
-    ...                    ('second', 2)])
-    >>> od1 == od2
-    False
-    >>> # Move 'third' key to the end
-    >>> del od2['third']; od2['third'] = 3
-    >>> od1 == od2
-    True
-
-Comparing an :class:`~collections.OrderedDict` with a regular dictionary
-ignores the insertion order and just compares the keys and values.
-
-How does the :class:`~collections.OrderedDict` work?  It maintains a
-doubly-linked list of keys, appending new keys to the list as they're inserted.
-A secondary dictionary maps keys to their corresponding list node, so
-deletion doesn't have to traverse the entire linked list and therefore
-remains O(1).
-
-The standard library now supports use of ordered dictionaries in several
-modules.
-
-* The :mod:`ConfigParser` module uses them by default, meaning that
-  configuration files can now read, modified, and then written back
-  in their original order.
-
-* The :meth:`~collections.somenamedtuple._asdict()` method for
-  :func:`collections.namedtuple` now returns an ordered dictionary with the
-  values appearing in the same order as the underlying tuple indices.
-
-* The :mod:`json` module's :class:`~json.JSONDecoder` class
-  constructor was extended with an *object_pairs_hook* parameter to
-  allow :class:`OrderedDict` instances to be built by the decoder.
-  Support was also added for third-party tools like
-  `PyYAML <http://pyyaml.org/>`_.
-
-.. seealso::
-
-   :pep:`372` - Adding an ordered dictionary to collections
-     PEP written by Armin Ronacher and Raymond Hettinger;
-     implemented by Raymond Hettinger.
-
-.. _pep-0378:
-
-PEP 378: Format Specifier for Thousands Separator
-=================================================
-
-To make program output more readable, it can be useful to add
-separators to large numbers, rendering them as
-18,446,744,073,709,551,616 instead of 18446744073709551616.
-
-The fully general solution for doing this is the :mod:`locale` module,
-which can use different separators ("," in North America, "." in
-Europe) and different grouping sizes, but :mod:`locale` is complicated
-to use and unsuitable for multi-threaded applications where different
-threads are producing output for different locales.
-
-Therefore, a simple comma-grouping mechanism has been added to the
-mini-language used by the :meth:`str.format` method.  When
-formatting a floating-point number, simply include a comma between the
-width and the precision::
-
-   >>> '{:20,.2f}'.format(18446744073709551616.0)
-   '18,446,744,073,709,551,616.00'
-
-When formatting an integer, include the comma after the width:
-
-   >>> '{:20,d}'.format(18446744073709551616)
-   '18,446,744,073,709,551,616'
-
-This mechanism is not adaptable at all; commas are always used as the
-separator and the grouping is always into three-digit groups.  The
-comma-formatting mechanism isn't as general as the :mod:`locale`
-module, but it's easier to use.
-
-.. seealso::
-
-   :pep:`378` - Format Specifier for Thousands Separator
-     PEP written by Raymond Hettinger; implemented by Eric Smith.
-
-PEP 389: The argparse Module for Parsing Command Lines
-======================================================
-
-The :mod:`argparse` module for parsing command-line arguments was
-added as a more powerful replacement for the
-:mod:`optparse` module.
-
-This means Python now supports three different modules for parsing
-command-line arguments: :mod:`getopt`, :mod:`optparse`, and
-:mod:`argparse`.  The :mod:`getopt` module closely resembles the C
-library's :cfunc:`getopt` function, so it remains useful if you're writing a
-Python prototype that will eventually be rewritten in C.
-:mod:`optparse` becomes redundant, but there are no plans to remove it
-because there are many scripts still using it, and there's no
-automated way to update these scripts.  (Making the :mod:`argparse`
-API consistent with :mod:`optparse`'s interface was discussed but
-rejected as too messy and difficult.)
-
-In short, if you're writing a new script and don't need to worry
-about compatibility with earlier versions of Python, use
-:mod:`argparse` instead of :mod:`optparse`.
-
-Here's an example::
-
-    import argparse
-
-    parser = argparse.ArgumentParser(description='Command-line example.')
-
-    # Add optional switches
-    parser.add_argument('-v', action='store_true', dest='is_verbose',
-                        help='produce verbose output')
-    parser.add_argument('-o', action='store', dest='output',
-                        metavar='FILE',
-                        help='direct output to FILE instead of stdout')
-    parser.add_argument('-C', action='store', type=int, dest='context',
-                        metavar='NUM', default=0,
-                        help='display NUM lines of added context')
-
-    # Allow any number of additional arguments.
-    parser.add_argument(nargs='*', action='store', dest='inputs',
-                        help='input filenames (default is stdin)')
-
-    args = parser.parse_args()
-    print args.__dict__
-
-Unless you override it, :option:`-h` and :option:`--help` switches
-are automatically added, and produce neatly formatted output::
-
-    -> ./python.exe argparse-example.py --help
-    usage: argparse-example.py [-h] [-v] [-o FILE] [-C NUM] [inputs [inputs ...]]
-
-    Command-line example.
-
-    positional arguments:
-      inputs      input filenames (default is stdin)
-
-    optional arguments:
-      -h, --help  show this help message and exit
-      -v          produce verbose output
-      -o FILE     direct output to FILE instead of stdout
-      -C NUM      display NUM lines of added context
-
-As with :mod:`optparse`, the command-line switches and arguments
-are returned as an object with attributes named by the *dest* parameters::
-
-    -> ./python.exe argparse-example.py -v
-    {'output': None,
-     'is_verbose': True,
-     'context': 0,
-     'inputs': []}
-
-    -> ./python.exe argparse-example.py -v -o /tmp/output -C 4 file1 file2
-    {'output': '/tmp/output',
-     'is_verbose': True,
-     'context': 4,
-     'inputs': ['file1', 'file2']}
-
-:mod:`argparse` has much fancier validation than :mod:`optparse`; you
-can specify an exact number of arguments as an integer, 0 or more
-arguments by passing ``'*'``, 1 or more by passing ``'+'``, or an
-optional argument with ``'?'``.  A top-level parser can contain
-sub-parsers to define subcommands that have different sets of
-switches, as in ``svn commit``, ``svn checkout``, etc.  You can
-specify an argument's type as :class:`~argparse.FileType`, which will
-automatically open files for you and understands that ``'-'`` means
-standard input or output.
-
-.. seealso::
-
-   `argparse module documentation <http://docs.python.org/dev/library/argparse.html>`__
-
-   `Upgrading optparse code to use argparse <http://docs.python.org/dev/library/argparse.html#upgrading-optparse-code>`__
-     Part of the Python documentation, describing how to convert
-     code that uses :mod:`optparse`.
-
-   :pep:`389` - argparse - New Command Line Parsing Module
-     PEP written and implemented by Steven Bethard.
-
-PEP 391: Dictionary-Based Configuration For Logging
-====================================================
-
-.. XXX not documented in library reference yet; add link here once it's added.
-
-The :mod:`logging` module is very flexible; applications can define
-a tree of logging subsystems, and each logger in this tree can filter
-out certain messages, format them differently, and direct messages to
-a varying number of handlers.
-
-All this flexibility can require a lot of configuration.  You can
-write Python statements to create objects and set their properties,
-but a complex set-up requires verbose but boring code.
-:mod:`logging` also supports a :func:`~logging.config.fileConfig`
-function that parses a file, but the file format doesn't support
-configuring filters, and it's messier to generate programmatically.
-
-Python 2.7 adds a :func:`~logging.config.dictConfig` function that
-uses a dictionary to configure logging.  There are many ways to
-produce a dictionary from different sources: construct one with code;
-parse a file containing JSON; or use a YAML parsing library if one is
-installed.
-
-The following example configures two loggers, the root logger and a
-logger named "network".   Messages sent to the root logger will be
-sent to the system log using the syslog protocol, and messages
-to the "network" logger will be written to a :file:`network.log` file
-that will be rotated once the log reaches 1Mb.
-
-::
-
-    import logging
-    import logging.config
-
-    configdict = {
-     'version': 1,    # Configuration schema in use; must be 1 for now
-     'formatters': {
-         'standard': {
-             'format': ('%(asctime)s %(name)-15s '
-                        '%(levelname)-8s %(message)s')}},
-
-     'handlers': {'netlog': {'backupCount': 10,
-                         'class': 'logging.handlers.RotatingFileHandler',
-                         'filename': '/logs/network.log',
-                         'formatter': 'standard',
-                         'level': 'INFO',
-                         'maxBytes': 1024*1024},
-                  'syslog': {'class': 'logging.handlers.SysLogHandler',
-                             'formatter': 'standard',
-                             'level': 'ERROR'}},
-
-     # Specify all the subordinate loggers
-     'loggers': {
-                 'network': {
-                             'handlers': ['netlog']
-                 }
-     },
-     # Specify properties of the root logger
-     'root': {
-              'handlers': ['syslog']
-     },
-    }
-
-    # Set up configuration
-    logging.config.dictConfig(configdict)
-
-    # As an example, log two error messages
-    logger = logging.getLogger('/')
-    logger.error('Database not found')
-
-    netlogger = logging.getLogger('network')
-    netlogger.error('Connection failed')
-
-Three smaller enhancements to the :mod:`logging` module, all
-implemented by Vinay Sajip, are:
-
-.. rev79293
-
-* The :class:`~logging.handlers.SysLogHandler` class now supports
-  syslogging over TCP.  The constructor has a *socktype* parameter
-  giving the type of socket to use, either :const:`socket.SOCK_DGRAM`
-  for UDP or :const:`socket.SOCK_STREAM` for TCP.  The default
-  protocol remains UDP.
-
-* :class:`Logger` instances gained a :meth:`getChild` method that retrieves a
-  descendant logger using a relative path.  For example,
-  once you retrieve a logger by doing ``log = getLogger('app')``,
-  calling ``log.getChild('network.listen')`` is equivalent to
-  ``getLogger('app.network.listen')``.
-
-* The :class:`LoggerAdapter` class gained a :meth:`isEnabledFor` method
-  that takes a *level* and returns whether the underlying logger would
-  process a message of that level of importance.
-
-.. seealso::
-
-   :pep:`391` - Dictionary-Based Configuration For Logging
-     PEP written and implemented by Vinay Sajip.
-
-PEP 3106: Dictionary Views
-====================================================
-
-The dictionary methods :meth:`keys`, :meth:`values`, and :meth:`items`
-are different in Python 3.x.  They return an object called a :dfn:`view`
-instead of a fully materialized list.
-
-It's not possible to change the return values of :meth:`keys`,
-:meth:`values`, and :meth:`items` in Python 2.7 because too much code
-would break.  Instead the 3.x versions were added under the new names
-:meth:`viewkeys`, :meth:`viewvalues`, and :meth:`viewitems`.
-
-::
-
-    >>> d = dict((i*10, chr(65+i)) for i in range(26))
-    >>> d
-    {0: 'A', 130: 'N', 10: 'B', 140: 'O', 20: ..., 250: 'Z'}
-    >>> d.viewkeys()
-    dict_keys([0, 130, 10, 140, 20, 150, 30, ..., 250])
-
-Views can be iterated over, but the key and item views also behave
-like sets.  The ``&`` operator performs intersection, and ``|``
-performs a union::
-
-    >>> d1 = dict((i*10, chr(65+i)) for i in range(26))
-    >>> d2 = dict((i**.5, i) for i in range(1000))
-    >>> d1.viewkeys() & d2.viewkeys()
-    set([0.0, 10.0, 20.0, 30.0])
-    >>> d1.viewkeys() | range(0, 30)
-    set([0, 1, 130, 3, 4, 5, 6, ..., 120, 250])
-
-The view keeps track of the dictionary and its contents change as the
-dictionary is modified::
-
-    >>> vk = d.viewkeys()
-    >>> vk
-    dict_keys([0, 130, 10, ..., 250])
-    >>> d[260] = '&'
-    >>> vk
-    dict_keys([0, 130, 260, 10, ..., 250])
-
-However, note that you can't add or remove keys while you're iterating
-over the view::
-
-    >>> for k in vk:
-    ...     d[k*2] = k
-    ...
-    Traceback (most recent call last):
-      File "<stdin>", line 1, in <module>
-    RuntimeError: dictionary changed size during iteration
-
-You can use the view methods in Python 2.x code, and the 2to3
-converter will change them to the standard :meth:`keys`,
-:meth:`values`, and :meth:`items` methods.
-
-.. seealso::
-
-   :pep:`3106` - Revamping dict.keys(), .values() and .items()
-     PEP written by Guido van Rossum.
-     Backported to 2.7 by Alexandre Vassalotti; :issue:`1967`.
-
-
-PEP 3137: The memoryview Object
-====================================================
-
-The :class:`memoryview` object provides a view of another object's
-memory content that matches the :class:`bytes` type's interface.
-
-    >>> import string
-    >>> m = memoryview(string.letters)
-    >>> m
-    <memory at 0x37f850>
-    >>> len(m)           # Returns length of underlying object
-    52
-    >>> m[0], m[25], m[26]   # Indexing returns one byte
-    ('a', 'z', 'A')
-    >>> m2 = m[0:26]         # Slicing returns another memoryview
-    >>> m2
-    <memory at 0x37f080>
-
-The content of the view can be converted to a string of bytes or
-a list of integers:
-
-    >>> m2.tobytes()
-    'abcdefghijklmnopqrstuvwxyz'
-    >>> m2.tolist()
-    [97, 98, 99, 100, 101, 102, 103, ... 121, 122]
-    >>>
-
-:class:`memoryview` objects allow modifying the underlying object if
-it's a mutable object.
-
-    >>> m2[0] = 75
-    Traceback (most recent call last):
-      File "<stdin>", line 1, in <module>
-    TypeError: cannot modify read-only memory
-    >>> b = bytearray(string.letters)  # Creating a mutable object
-    >>> b
-    bytearray(b'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ')
-    >>> mb = memoryview(b)
-    >>> mb[0] = '*'         # Assign to view, changing the bytearray.
-    >>> b[0:5]              # The bytearray has been changed.
-    bytearray(b'*bcde')
-    >>>
-
-.. seealso::
-
-   :pep:`3137` - Immutable Bytes and Mutable Buffer
-     PEP written by Guido van Rossum.
-     Implemented by Travis Oliphant, Antoine Pitrou and others.
-     Backported to 2.7 by Antoine Pitrou; :issue:`2396`.
-
-
-
-Other Language Changes
-======================
-
-Some smaller changes made to the core Python language are:
-
-* The syntax for set literals has been backported from Python 3.x.
-  Curly brackets are used to surround the contents of the resulting
-  mutable set; set literals are
-  distinguished from dictionaries by not containing colons and values.
-  ``{}`` continues to represent an empty dictionary; use
-  ``set()`` for an empty set.
-
-    >>> {1,2,3,4,5}
-    set([1, 2, 3, 4, 5])
-    >>> set() # empty set
-    set([])
-    >>> {}    # empty dict
-    {}
-
-  Backported by Alexandre Vassalotti; :issue:`2335`.
-
-* Dictionary and set comprehensions are another feature backported from
-  3.x, generalizing list/generator comprehensions to use
-  the literal syntax for sets and dictionaries.
-
-    >>> {x: x*x for x in range(6)}
-    {0: 0, 1: 1, 2: 4, 3: 9, 4: 16, 5: 25}
-    >>> {('a'*x) for x in range(6)}
-    set(['', 'a', 'aa', 'aaa', 'aaaa', 'aaaaa'])
-
-  Backported by Alexandre Vassalotti; :issue:`2333`.
-
-* The :keyword:`with` statement can now use multiple context managers
-  in one statement.  Context managers are processed from left to right
-  and each one is treated as beginning a new :keyword:`with` statement.
-  This means that::
-
-   with A() as a, B() as b:
-       ... suite of statements ...
-
-  is equivalent to::
-
-   with A() as a:
-       with B() as b:
-           ... suite of statements ...
-
-  The :func:`contextlib.nested` function provides a very similar
-  function, so it's no longer necessary and has been deprecated.
-
-  (Proposed in http://codereview.appspot.com/53094; implemented by
-  Georg Brandl.)
-
-* Conversions between floating-point numbers and strings are
-  now correctly rounded on most platforms.  These conversions occur
-  in many different places: :func:`str` on
-  floats and complex numbers; the :class:`float` and :class:`complex`
-  constructors;
-  numeric formatting; serializing and
-  deserializing floats and complex numbers using the
-  :mod:`marshal`, :mod:`pickle`
-  and :mod:`json` modules;
-  parsing of float and imaginary literals in Python code;
-  and :class:`~decimal.Decimal`-to-float conversion.
-
-  Related to this, the :func:`repr` of a floating-point number *x*
-  now returns a result based on the shortest decimal string that's
-  guaranteed to round back to *x* under correct rounding (with
-  round-half-to-even rounding mode).  Previously it gave a string
-  based on rounding x to 17 decimal digits.
-
-  .. maybe add an example?
-
-  The rounding library responsible for this improvement works on
-  Windows and on Unix platforms using the gcc, icc, or suncc
-  compilers.  There may be a small number of platforms where correct
-  operation of this code cannot be guaranteed, so the code is not
-  used on such systems.  You can find out which code is being used
-  by checking :data:`sys.float_repr_style`,  which will be ``short``
-  if the new code is in use and ``legacy`` if it isn't.
-
-  Implemented by Eric Smith and Mark Dickinson, using David Gay's
-  :file:`dtoa.c` library; :issue:`7117`.
-
-* Conversions from long integers and regular integers to floating
-  point now round differently, returning the floating-point number
-  closest to the number.  This doesn't matter for small integers that
-  can be converted exactly, but for large numbers that will
-  unavoidably lose precision, Python 2.7 now approximates more
-  closely.  For example, Python 2.6 computed the following::
-
-    >>> n = 295147905179352891391
-    >>> float(n)
-    2.9514790517935283e+20
-    >>> n - long(float(n))
-    65535L
-
-  Python 2.7's floating-point result is larger, but much closer to the
-  true value::
-
-    >>> n = 295147905179352891391
-    >>> float(n)
-    2.9514790517935289e+20
-    >>> n - long(float(n))
-    -1L
-
-  (Implemented by Mark Dickinson; :issue:`3166`.)
-
-  Integer division is also more accurate in its rounding behaviours.  (Also
-  implemented by Mark Dickinson; :issue:`1811`.)
-
-* Implicit coercion for complex numbers has been removed; the interpreter
-  will no longer ever attempt to call a :meth:`__coerce__` method on complex
-  objects.  (Removed by Meador Inge and Mark Dickinson; :issue:`5211`.)
-
-* The :meth:`str.format` method now supports automatic numbering of the replacement
-  fields.  This makes using :meth:`str.format` more closely resemble using
-  ``%s`` formatting::
-
-    >>> '{}:{}:{}'.format(2009, 04, 'Sunday')
-    '2009:4:Sunday'
-    >>> '{}:{}:{day}'.format(2009, 4, day='Sunday')
-    '2009:4:Sunday'
-
-  The auto-numbering takes the fields from left to right, so the first ``{...}``
-  specifier will use the first argument to :meth:`str.format`, the next
-  specifier will use the next argument, and so on.  You can't mix auto-numbering
-  and explicit numbering -- either number all of your specifier fields or none
-  of them -- but you can mix auto-numbering and named fields, as in the second
-  example above.  (Contributed by Eric Smith; :issue:`5237`.)
-
-  Complex numbers now correctly support usage with :func:`format`,
-  and default to being right-aligned.
-  Specifying a precision or comma-separation applies to both the real
-  and imaginary parts of the number, but a specified field width and
-  alignment is applied to the whole of the resulting ``1.5+3j``
-  output.  (Contributed by Eric Smith; :issue:`1588` and :issue:`7988`.)
-
-  The 'F' format code now always formats its output using uppercase characters,
-  so it will now produce 'INF' and 'NAN'.
-  (Contributed by Eric Smith; :issue:`3382`.)
-
-  A low-level change: the :meth:`object.__format__` method now triggers
-  a :exc:`PendingDeprecationWarning` if it's passed a format string,
-  because the :meth:`__format__` method for :class:`object` converts
-  the object to a string representation and formats that.  Previously
-  the method silently applied the format string to the string
-  representation, but that could hide mistakes in Python code.  If
-  you're supplying formatting information such as an alignment or
-  precision, presumably you're expecting the formatting to be applied
-  in some object-specific way.  (Fixed by Eric Smith; :issue:`7994`.)
-
-* The :func:`int` and :func:`long` types gained a ``bit_length``
-  method that returns the number of bits necessary to represent
-  its argument in binary::
-
-      >>> n = 37
-      >>> bin(n)
-      '0b100101'
-      >>> n.bit_length()
-      6
-      >>> n = 2**123-1
-      >>> n.bit_length()
-      123
-      >>> (n+1).bit_length()
-      124
-
-  (Contributed by Fredrik Johansson and Victor Stinner; :issue:`3439`.)
-
-* The :keyword:`import` statement will no longer try a relative import
-  if an absolute import (e.g. ``from .os import sep``) fails.  This
-  fixes a bug, but could possibly break certain :keyword:`import`
-  statements that were only working by accident.  (Fixed by Meador Inge;
-  :issue:`7902`.)
-
-* It's now possible for a subclass of the built-in :class:`unicode` type
-  to override the :meth:`__unicode__` method.  (Implemented by
-  Victor Stinner; :issue:`1583863`.)
-
-* The :class:`bytearray` type's :meth:`~bytearray.translate` method now accepts
-  ``None`` as its first argument.  (Fixed by Georg Brandl;
-  :issue:`4759`.)
-
-  .. bytearray doesn't seem to be documented
-
-* When using ``@classmethod`` and ``@staticmethod`` to wrap
-  methods as class or static methods, the wrapper object now
-  exposes the wrapped function as their :attr:`__func__` attribute.
-  (Contributed by Amaury Forgeot d'Arc, after a suggestion by
-  George Sakkis; :issue:`5982`.)
-
-* When a restricted set of attributes were set using ``__slots__``,
-  deleting an unset attribute would not raise :exc:`AttributeError`
-  as you would expect.  Fixed by Benjamin Peterson; :issue:`7604`.)
-
-* Two new encodings are now supported: "cp720", used primarily for
-  Arabic text; and "cp858", a variant of CP 850 that adds the euro
-  symbol.  (CP720 contributed by Alexander Belchenko and Amaury
-  Forgeot d'Arc in :issue:`1616979`; CP858 contributed by Tim Hatch in
-  :issue:`8016`.)
-
-* The :class:`file` object will now set the :attr:`filename` attribute
-  on the :exc:`IOError` exception when trying to open a directory
-  on POSIX platforms (noted by Jan Kaliszewski; :issue:`4764`), and
-  now explicitly checks for and forbids writing to read-only file objects
-  instead of trusting the C library to catch and report the error
-  (fixed by Stefan Krah; :issue:`5677`).
-
-* The Python tokenizer now translates line endings itself, so the
-  :func:`compile` built-in function now accepts code using any
-  line-ending convention.  Additionally, it no longer requires that the
-  code end in a newline.
-
-* Extra parentheses in function definitions are illegal in Python 3.x,
-  meaning that you get a syntax error from ``def f((x)): pass``.  In
-  Python3-warning mode, Python 2.7 will now warn about this odd usage.
-  (Noted by James Lingard; :issue:`7362`.)
-
-* It's now possible to create weak references to old-style class
-  objects.  New-style classes were always weak-referenceable.  (Fixed
-  by Antoine Pitrou; :issue:`8268`.)
-
-* When a module object is garbage-collected, the module's dictionary is
-  now only cleared if no one else is holding a reference to the
-  dictionary (:issue:`7140`).
-
-.. ======================================================================
-
-.. _new-27-interpreter:
-
-Interpreter Changes
--------------------------------
-
-A new environment variable, :envvar:`PYTHONWARNINGS`,
-allows controlling warnings.  It should be set to a string
-containing warning settings, equivalent to those
-used with the :option:`-W` switch, separated by commas.
-(Contributed by Brian Curtin; :issue:`7301`.)
-
-For example, the following setting will print warnings every time
-they occur, but turn warnings from the :mod:`Cookie` module into an
-error.  (The exact syntax for setting an environment variable varies
-across operating systems and shells.)
-
-::
-
-  export PYTHONWARNINGS=all,error:::Cookie:0
-
-.. ======================================================================
-
-
-Optimizations
--------------
-
-Several performance enhancements have been added:
-
-.. * A new :program:`configure` option, :option:`--with-computed-gotos`,
-   compiles the main bytecode interpreter loop using a new dispatch
-   mechanism that gives speedups of up to 20%, depending on the system
-   and benchmark.  The new mechanism is only supported on certain
-   compilers, such as gcc, SunPro, and icc.
-
-* A new opcode was added to perform the initial setup for
-  :keyword:`with` statements, looking up the :meth:`__enter__` and
-  :meth:`__exit__` methods.  (Contributed by Benjamin Peterson.)
-
-* The garbage collector now performs better for one common usage
-  pattern: when many objects are being allocated without deallocating
-  any of them.  This would previously take quadratic
-  time for garbage collection, but now the number of full garbage collections
-  is reduced as the number of objects on the heap grows.
-  The new logic only performs a full garbage collection pass when
-  the middle generation has been collected 10 times and when the
-  number of survivor objects from the middle generation exceeds 10% of
-  the number of objects in the oldest generation.  (Suggested by Martin
-  von L?wis and implemented by Antoine Pitrou; :issue:`4074`.)
-
-* The garbage collector tries to avoid tracking simple containers
-  which can't be part of a cycle. In Python 2.7, this is now true for
-  tuples and dicts containing atomic types (such as ints, strings,
-  etc.). Transitively, a dict containing tuples of atomic types won't
-  be tracked either. This helps reduce the cost of each
-  garbage collection by decreasing the number of objects to be
-  considered and traversed by the collector.
-  (Contributed by Antoine Pitrou; :issue:`4688`.)
-
-* Long integers are now stored internally either in base 2**15 or in base
-  2**30, the base being determined at build time.  Previously, they
-  were always stored in base 2**15.  Using base 2**30 gives
-  significant performance improvements on 64-bit machines, but
-  benchmark results on 32-bit machines have been mixed.  Therefore,
-  the default is to use base 2**30 on 64-bit machines and base 2**15
-  on 32-bit machines; on Unix, there's a new configure option
-  :option:`--enable-big-digits` that can be used to override this default.
-
-  Apart from the performance improvements this change should be
-  invisible to end users, with one exception: for testing and
-  debugging purposes there's a new structseq :data:`sys.long_info` that
-  provides information about the internal format, giving the number of
-  bits per digit and the size in bytes of the C type used to store
-  each digit::
-
-     >>> import sys
-     >>> sys.long_info
-     sys.long_info(bits_per_digit=30, sizeof_digit=4)
-
-  (Contributed by Mark Dickinson; :issue:`4258`.)
-
-  Another set of changes made long objects a few bytes smaller: 2 bytes
-  smaller on 32-bit systems and 6 bytes on 64-bit.
-  (Contributed by Mark Dickinson; :issue:`5260`.)
-
-* The division algorithm for long integers has been made faster
-  by tightening the inner loop, doing shifts instead of multiplications,
-  and fixing an unnecessary extra iteration.
-  Various benchmarks show speedups of between 50% and 150% for long
-  integer divisions and modulo operations.
-  (Contributed by Mark Dickinson; :issue:`5512`.)
-  Bitwise operations are also significantly faster (initial patch by
-  Gregory Smith; :issue:`1087418`).
-
-* The implementation of ``%`` checks for the left-side operand being
-  a Python string and special-cases it; this results in a 1-3%
-  performance increase for applications that frequently use ``%``
-  with strings, such as templating libraries.
-  (Implemented by Collin Winter; :issue:`5176`.)
-
-* List comprehensions with an ``if`` condition are compiled into
-  faster bytecode.  (Patch by Antoine Pitrou, back-ported to 2.7
-  by Jeffrey Yasskin; :issue:`4715`.)
-
-* Converting an integer or long integer to a decimal string was made
-  faster by special-casing base 10 instead of using a generalized
-  conversion function that supports arbitrary bases.
-  (Patch by Gawain Bolton; :issue:`6713`.)
-
-* The :meth:`split`, :meth:`replace`, :meth:`rindex`,
-  :meth:`rpartition`, and :meth:`rsplit` methods of string-like types
-  (strings, Unicode strings, and :class:`bytearray` objects) now use a
-  fast reverse-search algorithm instead of a character-by-character
-  scan.  This is sometimes faster by a factor of 10.  (Added by
-  Florent Xicluna; :issue:`7462` and :issue:`7622`.)
-
-* The :mod:`pickle` and :mod:`cPickle` modules now automatically
-  intern the strings used for attribute names, reducing memory usage
-  of the objects resulting from unpickling.  (Contributed by Jake
-  McGuire; :issue:`5084`.)
-
-* The :mod:`cPickle` module now special-cases dictionaries,
-  nearly halving the time required to pickle them.
-  (Contributed by Collin Winter; :issue:`5670`.)
-
-.. ======================================================================
-
-New and Improved Modules
-========================
-
-As in every release, Python's standard library received a number of
-enhancements and bug fixes.  Here's a partial list of the most notable
-changes, sorted alphabetically by module name. Consult the
-:file:`Misc/NEWS` file in the source tree for a more complete list of
-changes, or look through the Subversion logs for all the details.
-
-* The :mod:`bdb` module's base debugging class :class:`~bdb.Bdb`
-  gained a feature for skipping modules.  The constructor
-  now takes an iterable containing glob-style patterns such as
-  ``django.*``; the debugger will not step into stack frames
-  from a module that matches one of these patterns.
-  (Contributed by Maru Newby after a suggestion by
-  Senthil Kumaran; :issue:`5142`.)
-
-* The :mod:`binascii` module now supports the buffer API, so it can be
-  used with :class:`memoryview` instances and other similar buffer objects.
-  (Backported from 3.x by Florent Xicluna; :issue:`7703`.)
-
-* Updated module: the :mod:`bsddb` module has been updated from 4.7.2devel9
-  to version 4.8.4 of
-  `the pybsddb package <http://www.jcea.es/programacion/pybsddb.htm>`__.
-  The new version features better Python 3.x compatibility, various bug fixes,
-  and adds several new BerkeleyDB flags and methods.
-  (Updated by Jes?s Cea Avi?n; :issue:`8156`.  The pybsddb
-  changelog can be read at http://hg.jcea.es/pybsddb/file/tip/ChangeLog.)
-
-* The :mod:`bz2` module's :class:`~bz2.BZ2File` now supports the context
-  management protocol, so you can write ``with bz2.BZ2File(...) as f:``.
-  (Contributed by Hagen F?rstenau; :issue:`3860`.)
-
-* New class: the :class:`~collections.Counter` class in the :mod:`collections`
-  module is useful for tallying data.  :class:`~collections.Counter` instances
-  behave mostly like dictionaries but return zero for missing keys instead of
-  raising a :exc:`KeyError`:
-
-  .. doctest::
-     :options: +NORMALIZE_WHITESPACE
-
-     >>> from collections import Counter
-     >>> c = Counter()
-     >>> for letter in 'here is a sample of english text':
-     ...   c[letter] += 1
-     ...
-     >>> c
-     Counter({' ': 6, 'e': 5, 's': 3, 'a': 2, 'i': 2, 'h': 2,
-     'l': 2, 't': 2, 'g': 1, 'f': 1, 'm': 1, 'o': 1, 'n': 1,
-     'p': 1, 'r': 1, 'x': 1})
-     >>> c['e']
-     5
-     >>> c['z']
-     0
-
-  There are three additional :class:`~collections.Counter` methods.
-  :meth:`~collections.Counter.most_common` returns the N most common
-  elements and their counts.  :meth:`~collections.Counter.elements`
-  returns an iterator over the contained elements, repeating each
-  element as many times as its count.
-  :meth:`~collections.Counter.subtract` takes an iterable and
-  subtracts one for each element instead of adding; if the argument is
-  a dictionary or another :class:`Counter`, the counts are
-  subtracted. ::
-
-    >>> c.most_common(5)
-    [(' ', 6), ('e', 5), ('s', 3), ('a', 2), ('i', 2)]
-    >>> c.elements() ->
-       'a', 'a', ' ', ' ', ' ', ' ', ' ', ' ',
-       'e', 'e', 'e', 'e', 'e', 'g', 'f', 'i', 'i',
-       'h', 'h', 'm', 'l', 'l', 'o', 'n', 'p', 's',
-       's', 's', 'r', 't', 't', 'x'
-    >>> c['e']
-    5
-    >>> c.subtract('very heavy on the letter e')
-    >>> c['e']    # Count is now lower
-    -1
-
-  Contributed by Raymond Hettinger; :issue:`1696199`.
-
-  .. revision 79660
-
-  New class: :class:`~collections.OrderedDict` is described in the earlier
-  section :ref:`pep-0372`.
-
-  New method: The :class:`~collections.deque` data type now has a
-  :meth:`~collections.deque.count` method that returns the number of
-  contained elements equal to the supplied argument *x*, and a
-  :meth:`~collections.deque.reverse` method that reverses the elements
-  of the deque in-place.  :class:`deque` also exposes its maximum
-  length as the read-only :attr:`~collections.deque.maxlen` attribute.
-  (Both features added by Raymond Hettinger.)
-
-  The :class:`~collections.namedtuple` class now has an optional *rename* parameter.
-  If *rename* is true, field names that are invalid because they've
-  been repeated or aren't legal Python identifiers will be
-  renamed to legal names that are derived from the field's
-  position within the list of fields:
-
-     >>> from collections import namedtuple
-     >>> T = namedtuple('T', ['field1', '$illegal', 'for', 'field2'], rename=True)
-     >>> T._fields
-     ('field1', '_1', '_2', 'field2')
-
-  (Added by Raymond Hettinger; :issue:`1818`.)
-
-  Finally, the :class:`~collections.Mapping` abstract base class now
-  returns :const:`NotImplemented` if a mapping is compared to
-  another type that isn't a :class:`Mapping`.
-  (Fixed by Daniel Stutzbach; :issue:`8729`.)
-
-* Constructors for the parsing classes in the :mod:`ConfigParser` module now
-  take a *allow_no_value* parameter, defaulting to false; if true,
-  options without values will be allowed.  For example::
-
-    >>> import ConfigParser, StringIO
-    >>> sample_config = """
-    ... [mysqld]
-    ... user = mysql
-    ... pid-file = /var/run/mysqld/mysqld.pid
-    ... skip-bdb
-    ... """
-    >>> config = ConfigParser.RawConfigParser(allow_no_value=True)
-    >>> config.readfp(StringIO.StringIO(sample_config))
-    >>> config.get('mysqld', 'user')
-    'mysql'
-    >>> print config.get('mysqld', 'skip-bdb')
-    None
-    >>> print config.get('mysqld', 'unknown')
-    Traceback (most recent call last):
-      ...
-    NoOptionError: No option 'unknown' in section: 'mysqld'
-
-  (Contributed by Mats Kindahl; :issue:`7005`.)
-
-* Deprecated function: :func:`contextlib.nested`, which allows
-  handling more than one context manager with a single :keyword:`with`
-  statement, has been deprecated, because the :keyword:`with` statement
-  now supports multiple context managers.
-
-* The :mod:`cookielib` module now ignores cookies that have an invalid
-  version field, one that doesn't contain an integer value.  (Fixed by
-  John J. Lee; :issue:`3924`.)
-
-* The :mod:`copy` module's :func:`~copy.deepcopy` function will now
-  correctly copy bound instance methods.  (Implemented by
-  Robert Collins; :issue:`1515`.)
-
-* The :mod:`ctypes` module now always converts ``None`` to a C NULL
-  pointer for arguments declared as pointers.  (Changed by Thomas
-  Heller; :issue:`4606`.)  The underlying `libffi library
-  <http://sourceware.org/libffi/>`__ has been updated to version
-  3.0.9, containing various fixes for different platforms.  (Updated
-  by Matthias Klose; :issue:`8142`.)
-
-* New method: the :mod:`datetime` module's :class:`~datetime.timedelta` class
-  gained a :meth:`~datetime.timedelta.total_seconds` method that returns the
-  number of seconds in the duration.  (Contributed by Brian Quinlan; :issue:`5788`.)
-
-* New method: the :class:`~decimal.Decimal` class gained a
-  :meth:`~decimal.Decimal.from_float` class method that performs an exact
-  conversion of a floating-point number to a :class:`~decimal.Decimal`.
-  This exact conversion strives for the
-  closest decimal approximation to the floating-point representation's value;
-  the resulting decimal value will therefore still include the inaccuracy,
-  if any.
-  For example, ``Decimal.from_float(0.1)`` returns
-  ``Decimal('0.1000000000000000055511151231257827021181583404541015625')``.
-  (Implemented by Raymond Hettinger; :issue:`4796`.)
-
-  Comparing instances of :class:`Decimal` with floating-point
-  numbers now produces sensible results based on the numeric values
-  of the operands.  Previously such comparisons would fall back to
-  Python's default rules for comparing objects, which produced arbitrary
-  results based on their type.  Note that you still cannot combine
-  :class:`Decimal` and floating-point in other operations such as addition,
-  since you should be explicitly choosing how to convert between float and
-  :class:`Decimal`.
-  (Fixed by Mark Dickinson; :issue:`2531`.)
-
-  The constructor for :class:`~decimal.Decimal` now accepts
-  floating-point numbers (added by Raymond Hettinger; :issue:`8257`)
-  and non-European Unicode characters such as Arabic-Indic digits
-  (contributed by Mark Dickinson; :issue:`6595`).
-
-  Most of the methods of the :class:`~decimal.Context` class now accept integers
-  as well as :class:`~decimal.Decimal` instances; the only exceptions are the
-  :meth:`~decimal.Context.canonical` and :meth:`~decimal.Context.is_canonical`
-  methods.  (Patch by Juan Jos? Conti; :issue:`7633`.)
-
-  When using :class:`~decimal.Decimal` instances with a string's
-  :meth:`~str.format` method, the default alignment was previously
-  left-alignment.  This has been changed to right-alignment, which is
-  more sensible for numeric types.  (Changed by Mark Dickinson; :issue:`6857`.)
-
-  Comparisons involving a signaling NaN value (or ``sNAN``) now signal
-  :const:`InvalidOperation` instead of silently returning a true or
-  false value depending on the comparison operator.  Quiet NaN values
-  (or ``NaN``) are now hashable.  (Fixed by Mark Dickinson;
-  :issue:`7279`.)
-
-* The :mod:`difflib` module now produces output that is more
-  compatible with modern :command:`diff`/:command:`patch` tools
-  through one small change, using a tab character instead of spaces as
-  a separator in the header giving the filename.  (Fixed by Anatoly
-  Techtonik; :issue:`7585`.)
-
-* The Distutils ``sdist`` command now always regenerates the
-  :file:`MANIFEST` file, since even if the :file:`MANIFEST.in` or
-  :file:`setup.py` files haven't been modified, the user might have
-  created some new files that should be included.
-  (Fixed by Tarek Ziad?; :issue:`8688`.)
-
-* The :mod:`doctest` module's :const:`IGNORE_EXCEPTION_DETAIL` flag
-  will now ignore the name of the module containing the exception
-  being tested.  (Patch by Lennart Regebro; :issue:`7490`.)
-
-* The :mod:`email` module's :class:`~email.message.Message` class will
-  now accept a Unicode-valued payload, automatically converting the
-  payload to the encoding specified by :attr:`output_charset`.
-  (Added by R. David Murray; :issue:`1368247`.)
-
-* The :class:`~fractions.Fraction` class now accepts a single float or
-  :class:`~decimal.Decimal` instance, or two rational numbers, as
-  arguments to its constructor.  (Implemented by Mark Dickinson;
-  rationals added in :issue:`5812`, and float/decimal in
-  :issue:`8294`.)
-
-  Ordering comparisons (``<``, ``<=``, ``>``, ``>=``) between
-  fractions and complex numbers now raise a :exc:`TypeError`.
-  This fixes an oversight, making the :class:`Fraction` match the other
-  numeric types.
-
-  .. revision 79455
-
-* New class: :class:`~ftplib.FTP_TLS` in
-  the :mod:`ftplib` module provides secure FTP
-  connections using TLS encapsulation of authentication as well as
-  subsequent control and data transfers.
-  (Contributed by Giampaolo Rodola; :issue:`2054`.)
-
-  The :meth:`~ftplib.FTP.storbinary` method for binary uploads can now restart
-  uploads thanks to an added *rest* parameter (patch by Pablo Mouzo;
-  :issue:`6845`.)
-
-* New class decorator: :func:`total_ordering` in the :mod:`functools`
-  module takes a class that defines an :meth:`__eq__` method and one of
-  :meth:`__lt__`, :meth:`__le__`, :meth:`__gt__`, or :meth:`__ge__`,
-  and generates the missing comparison methods.  Since the
-  :meth:`__cmp__` method is being deprecated in Python 3.x,
-  this decorator makes it easier to define ordered classes.
-  (Added by Raymond Hettinger; :issue:`5479`.)
-
-  New function: :func:`cmp_to_key` will take an old-style comparison
-  function that expects two arguments and return a new callable that
-  can be used as the *key* parameter to functions such as
-  :func:`sorted`, :func:`min` and :func:`max`, etc.  The primary
-  intended use is to help with making code compatible with Python 3.x.
-  (Added by Raymond Hettinger.)
-
-* New function: the :mod:`gc` module's :func:`~gc.is_tracked` returns
-  true if a given instance is tracked by the garbage collector, false
-  otherwise. (Contributed by Antoine Pitrou; :issue:`4688`.)
-
-* The :mod:`gzip` module's :class:`~gzip.GzipFile` now supports the context
-  management protocol, so you can write ``with gzip.GzipFile(...) as f:``
-  (contributed by Hagen F?rstenau; :issue:`3860`), and it now implements
-  the :class:`io.BufferedIOBase` ABC, so you can wrap it with
-  :class:`io.BufferedReader` for faster processing
-  (contributed by Nir Aides; :issue:`7471`).
-  It's also now possible to override the modification time
-  recorded in a gzipped file by providing an optional timestamp to
-  the constructor.  (Contributed by Jacques Frechet; :issue:`4272`.)
-
-  Files in gzip format can be padded with trailing zero bytes; the
-  :mod:`gzip` module will now consume these trailing bytes.  (Fixed by
-  Tadek Pietraszek and Brian Curtin; :issue:`2846`.)
-
-* New attribute: the :mod:`hashlib` module now has an :attr:`~hashlib.hashlib.algorithms`
-  attribute containing a tuple naming the supported algorithms.
-  In Python 2.7, ``hashlib.algorithms`` contains
-  ``('md5', 'sha1', 'sha224', 'sha256', 'sha384', 'sha512')``.
-  (Contributed by Carl Chenet; :issue:`7418`.)
-
-* The default :class:`~httplib.HTTPResponse` class used by the :mod:`httplib` module now
-  supports buffering, resulting in much faster reading of HTTP responses.
-  (Contributed by Kristj?n Valur J?nsson; :issue:`4879`.)
-
-  The :class:`~httplib.HTTPConnection` and :class:`~httplib.HTTPSConnection` classes
-  now support a *source_address* parameter, a ``(host, port)`` 2-tuple
-  giving the source address that will be used for the connection.
-  (Contributed by Eldon Ziegler; :issue:`3972`.)
-
-* The :mod:`ihooks` module now supports relative imports.  Note that
-  :mod:`ihooks` is an older module for customizing imports,
-  superseded by the :mod:`imputil` module added in Python 2.0.
-  (Relative import support added by Neil Schemenauer.)
-
-  .. revision 75423
-
-* The :mod:`imaplib` module now supports IPv6 addresses.
-  (Contributed by Derek Morr; :issue:`1655`.)
-
-* New function: the :mod:`inspect` module's :func:`~inspect.getcallargs`
-  takes a callable and its positional and keyword arguments,
-  and figures out which of the callable's parameters will receive each argument,
-  returning a dictionary mapping argument names to their values.  For example::
-
-    >>> from inspect import getcallargs
-    >>> def f(a, b=1, *pos, **named):
-    ...     pass
-    >>> getcallargs(f, 1, 2, 3)
-    {'a': 1, 'b': 2, 'pos': (3,), 'named': {}}
-    >>> getcallargs(f, a=2, x=4)
-    {'a': 2, 'b': 1, 'pos': (), 'named': {'x': 4}}
-    >>> getcallargs(f)
-    Traceback (most recent call last):
-    ...
-    TypeError: f() takes at least 1 argument (0 given)
-
-  Contributed by George Sakkis; :issue:`3135`.
-
-* Updated module: The :mod:`io` library has been upgraded to the version shipped with
-  Python 3.1.  For 3.1, the I/O library was entirely rewritten in C
-  and is 2 to 20 times faster depending on the task being performed.  The
-  original Python version was renamed to the :mod:`_pyio` module.
-
-  One minor resulting change: the :class:`io.TextIOBase` class now
-  has an :attr:`errors` attribute giving the error setting
-  used for encoding and decoding errors (one of ``'strict'``, ``'replace'``,
-  ``'ignore'``).
-
-  The :class:`io.FileIO` class now raises an :exc:`OSError` when passed
-  an invalid file descriptor.  (Implemented by Benjamin Peterson;
-  :issue:`4991`.)  The :meth:`~io.IOBase.truncate` method now preserves the
-  file position; previously it would change the file position to the
-  end of the new file.  (Fixed by Pascal Chambon; :issue:`6939`.)
-
-* New function: ``itertools.compress(data, selectors)`` takes two
-  iterators.  Elements of *data* are returned if the corresponding
-  value in *selectors* is true::
-
-    itertools.compress('ABCDEF', [1,0,1,0,1,1]) =>
-      A, C, E, F
-
-  .. maybe here is better to use >>> list(itertools.compress(...)) instead
-
-  New function: ``itertools.combinations_with_replacement(iter, r)``
-  returns all the possible *r*-length combinations of elements from the
-  iterable *iter*.  Unlike :func:`~itertools.combinations`, individual elements
-  can be repeated in the generated combinations::
-
-    itertools.combinations_with_replacement('abc', 2) =>
-      ('a', 'a'), ('a', 'b'), ('a', 'c'),
-      ('b', 'b'), ('b', 'c'), ('c', 'c')
-
-  Note that elements are treated as unique depending on their position
-  in the input, not their actual values.
-
-  The :func:`itertools.count` function now has a *step* argument that
-  allows incrementing by values other than 1.  :func:`~itertools.count` also
-  now allows keyword arguments, and using non-integer values such as
-  floats or :class:`~decimal.Decimal` instances.  (Implemented by Raymond
-  Hettinger; :issue:`5032`.)
-
-  :func:`itertools.combinations` and :func:`itertools.product`
-  previously raised :exc:`ValueError` for values of *r* larger than
-  the input iterable.  This was deemed a specification error, so they
-  now return an empty iterator.  (Fixed by Raymond Hettinger; :issue:`4816`.)
-
-* Updated module: The :mod:`json` module was upgraded to version 2.0.9 of the
-  simplejson package, which includes a C extension that makes
-  encoding and decoding faster.
-  (Contributed by Bob Ippolito; :issue:`4136`.)
-
-  To support the new :class:`collections.OrderedDict` type, :func:`json.load`
-  now has an optional *object_pairs_hook* parameter that will be called
-  with any object literal that decodes to a list of pairs.
-  (Contributed by Raymond Hettinger; :issue:`5381`.)
-
-* The :mod:`mailbox` module's :class:`Maildir` class now records the
-  timestamp on the directories it reads, and only re-reads them if the
-  modification time has subsequently changed.  This improves
-  performance by avoiding unneeded directory scans.  (Fixed by
-  A.M. Kuchling and Antoine Pitrou; :issue:`1607951`, :issue:`6896`.)
-
-* New functions: the :mod:`math` module gained
-  :func:`~math.erf` and :func:`~math.erfc` for the error function and the complementary error function,
-  :func:`~math.expm1` which computes ``e**x - 1`` with more precision than
-  using :func:`~math.exp` and subtracting 1,
-  :func:`~math.gamma` for the Gamma function, and
-  :func:`~math.lgamma` for the natural log of the Gamma function.
-  (Contributed by Mark Dickinson and nirinA raseliarison; :issue:`3366`.)
-
-* The :mod:`multiprocessing` module's :class:`Manager*` classes
-  can now be passed a callable that will be called whenever
-  a subprocess is started, along with a set of arguments that will be
-  passed to the callable.
-  (Contributed by lekma; :issue:`5585`.)
-
-  The :class:`~multiprocessing.Pool` class, which controls a pool of worker processes,
-  now has an optional *maxtasksperchild* parameter.  Worker processes
-  will perform the specified number of tasks and then exit, causing the
-  :class:`~multiprocessing.Pool` to start a new worker.  This is useful if tasks may leak
-  memory or other resources, or if some tasks will cause the worker to
-  become very large.
-  (Contributed by Charles Cazabon; :issue:`6963`.)
-
-* The :mod:`nntplib` module now supports IPv6 addresses.
-  (Contributed by Derek Morr; :issue:`1664`.)
-
-* New functions: the :mod:`os` module wraps the following POSIX system
-  calls: :func:`~os.getresgid` and :func:`~os.getresuid`, which return the
-  real, effective, and saved GIDs and UIDs;
-  :func:`~os.setresgid` and :func:`~os.setresuid`, which set
-  real, effective, and saved GIDs and UIDs to new values;
-  :func:`~os.initgroups`, which initialize the group access list
-  for the current process.  (GID/UID functions
-  contributed by Travis H.; :issue:`6508`.  Support for initgroups added
-  by Jean-Paul Calderone; :issue:`7333`.)
-
-  The :func:`os.fork` function now re-initializes the import lock in
-  the child process; this fixes problems on Solaris when :func:`~os.fork`
-  is called from a thread.  (Fixed by Zsolt Cserna; :issue:`7242`.)
-
-* In the :mod:`os.path` module, the :func:`~os.path.normpath` and
-  :func:`~os.path.abspath` functions now preserve Unicode; if their input path
-  is a Unicode string, the return value is also a Unicode string.
-  (:meth:`~os.path.normpath` fixed by Matt Giuca in :issue:`5827`;
-  :meth:`~os.path.abspath` fixed by Ezio Melotti in :issue:`3426`.)
-
-* The :mod:`pydoc` module now has help for the various symbols that Python
-  uses.  You can now do ``help('<<')`` or ``help('@')``, for example.
-  (Contributed by David Laban; :issue:`4739`.)
-
-* The :mod:`re` module's :func:`~re.split`, :func:`~re.sub`, and :func:`~re.subn`
-  now accept an optional *flags* argument, for consistency with the
-  other functions in the module.  (Added by Gregory P. Smith.)
-
-* New function: :func:`~runpy.run_path` in the :mod:`runpy` module
-  will execute the code at a provided *path* argument.  *path* can be
-  the path of a Python source file (:file:`example.py`), a compiled
-  bytecode file (:file:`example.pyc`), a directory
-  (:file:`./package/`), or a zip archive (:file:`example.zip`).  If a
-  directory or zip path is provided, it will be added to the front of
-  ``sys.path`` and the module :mod:`__main__` will be imported.  It's
-  expected that the directory or zip contains a :file:`__main__.py`;
-  if it doesn't, some other :file:`__main__.py` might be imported from
-  a location later in ``sys.path``.  This makes more of the machinery
-  of :mod:`runpy` available to scripts that want to mimic the way
-  Python's command line processes an explicit path name.
-  (Added by Nick Coghlan; :issue:`6816`.)
-
-* New function: in the :mod:`shutil` module, :func:`~shutil.make_archive`
-  takes a filename, archive type (zip or tar-format), and a directory
-  path, and creates an archive containing the directory's contents.
-  (Added by Tarek Ziad?.)
-
-  :mod:`shutil`'s :func:`~shutil.copyfile` and :func:`~shutil.copytree`
-  functions now raise a :exc:`~shutil.SpecialFileError` exception when
-  asked to copy a named pipe.  Previously the code would treat
-  named pipes like a regular file by opening them for reading, and
-  this would block indefinitely.  (Fixed by Antoine Pitrou; :issue:`3002`.)
-
-* The :mod:`signal` module no longer re-installs the signal handler
-  unless this is truly necessary, which fixes a bug that could make it
-  impossible to catch the EINTR signal robustly.  (Fixed by
-  Charles-Francois Natali; :issue:`8354`.)
-
-* New functions: in the :mod:`site` module, three new functions
-  return various site- and user-specific paths.
-  :func:`~site.getsitepackages` returns a list containing all
-  global site-packages directories,
-  :func:`~site.getusersitepackages` returns the path of the user's
-  site-packages directory, and
-  :func:`~site.getuserbase` returns the value of the :envvar:`USER_BASE`
-  environment variable, giving the path to a directory that can be used
-  to store data.
-  (Contributed by Tarek Ziad?; :issue:`6693`.)
-
-  The :mod:`site` module now reports exceptions occurring
-  when the :mod:`sitecustomize` module is imported, and will no longer
-  catch and swallow the :exc:`KeyboardInterrupt` exception.  (Fixed by
-  Victor Stinner; :issue:`3137`.)
-
-* The :func:`~socket.create_connection` function
-  gained a *source_address* parameter, a ``(host, port)`` 2-tuple
-  giving the source address that will be used for the connection.
-  (Contributed by Eldon Ziegler; :issue:`3972`.)
-
-  The :meth:`~socket.socket.recv_into` and :meth:`~socket.socket.recvfrom_into`
-  methods will now write into objects that support the buffer API, most usefully
-  the :class:`bytearray` and :class:`memoryview` objects.  (Implemented by
-  Antoine Pitrou; :issue:`8104`.)
-
-* The :mod:`SocketServer` module's :class:`~SocketServer.TCPServer` class now
-  supports socket timeouts and disabling the Nagle algorithm.
-  The :attr:`~SocketServer.TCPServer.disable_nagle_algorithm` class attribute
-  defaults to False; if overridden to be True,
-  new request connections will have the TCP_NODELAY option set to
-  prevent buffering many small sends into a single TCP packet.
-  The :attr:`~SocketServer.TCPServer.timeout` class attribute can hold
-  a timeout in seconds that will be applied to the request socket; if
-  no request is received within that time, :meth:`handle_timeout`
-  will be called and :meth:`handle_request` will return.
-  (Contributed by Kristj?n Valur J?nsson; :issue:`6192` and :issue:`6267`.)
-
-* Updated module: the :mod:`sqlite3` module has been updated to
-  version 2.6.0 of the `pysqlite package <http://code.google.com/p/pysqlite/>`__. Version 2.6.0 includes a number of bugfixes, and adds
-  the ability to load SQLite extensions from shared libraries.
-  Call the ``enable_load_extension(True)`` method to enable extensions,
-  and then call :meth:`~sqlite3.Connection.load_extension` to load a particular shared library.
-  (Updated by Gerhard H?ring.)
-
-* The :mod:`ssl` module's :class:`ssl.SSLSocket` objects now support the
-  buffer API, which fixed a test suite failure (fix by Antoine Pitrou;
-  :issue:`7133`) and automatically set
-  OpenSSL's :cmacro:`SSL_MODE_AUTO_RETRY`, which will prevent an error
-  code being returned from :meth:`recv` operations that trigger an SSL
-  renegotiation (fix by Antoine Pitrou; :issue:`8222`).
-
-  The :func:`ssl.wrap_socket` constructor function now takes a
-  *ciphers* argument that's a string listing the encryption algorithms
-  to be allowed; the format of the string is described
-  `in the OpenSSL documentation
-  <http://www.openssl.org/docs/apps/ciphers.html#CIPHER_LIST_FORMAT>`__.
-  (Added by Antoine Pitrou; :issue:`8322`.)
-
-  Another change makes the extension load all of OpenSSL's ciphers and
-  digest algorithms so that they're all available.  Some SSL
-  certificates couldn't be verified, reporting an "unknown algorithm"
-  error.  (Reported by Beda Kosata, and fixed by Antoine Pitrou;
-  :issue:`8484`.)
-
-  The version of OpenSSL being used is now available as the module
-  attributes :data:`ssl.OPENSSL_VERSION` (a string),
-  :data:`ssl.OPENSSL_VERSION_INFO` (a 5-tuple), and
-  :data:`ssl.OPENSSL_VERSION_NUMBER` (an integer).  (Added by Antoine
-  Pitrou; :issue:`8321`.)
-
-* The :mod:`struct` module will no longer silently ignore overflow
-  errors when a value is too large for a particular integer format
-  code (one of ``bBhHiIlLqQ``); it now always raises a
-  :exc:`struct.error` exception.  (Changed by Mark Dickinson;
-  :issue:`1523`.)  The :func:`~struct.pack` function will also
-  attempt to use :meth:`__index__` to convert and pack non-integers
-  before trying the :meth:`__int__` method or reporting an error.
-  (Changed by Mark Dickinson; :issue:`8300`.)
-
-* New function: the :mod:`subprocess` module's
-  :func:`~subprocess.check_output` runs a command with a specified set of arguments
-  and returns the command's output as a string when the command runs without
-  error, or raises a :exc:`~subprocess.CalledProcessError` exception otherwise.
-
-  ::
-
-    >>> subprocess.check_output(['df', '-h', '.'])
-    'Filesystem     Size   Used  Avail Capacity  Mounted on\n
-    /dev/disk0s2    52G    49G   3.0G    94%    /\n'
-
-    >>> subprocess.check_output(['df', '-h', '/bogus'])
-      ...
-    subprocess.CalledProcessError: Command '['df', '-h', '/bogus']' returned non-zero exit status 1
-
-  (Contributed by Gregory P. Smith.)
-
-  The :mod:`subprocess` module will now retry its internal system calls
-  on receiving an :const:`EINTR` signal.  (Reported by several people; final
-  patch by Gregory P. Smith in :issue:`1068268`.)
-
-* New function: :func:`~symtable.is_declared_global` in the :mod:`symtable` module
-  returns true for variables that are explicitly declared to be global,
-  false for ones that are implicitly global.
-  (Contributed by Jeremy Hylton.)
-
-* The :mod:`syslog` module will now use the value of ``sys.argv[0]`` as the
-  identifier instead of the previous default value of ``'python'``.
-  (Changed by Sean Reifschneider; :issue:`8451`.)
-
-* The ``sys.version_info`` value is now a named tuple, with attributes
-  named :attr:`major`, :attr:`minor`, :attr:`micro`,
-  :attr:`releaselevel`, and :attr:`serial`.  (Contributed by Ross
-  Light; :issue:`4285`.)
-
-  :func:`sys.getwindowsversion` also returns a named tuple,
-  with attributes named :attr:`major`, :attr:`minor`, :attr:`build`,
-  :attr:`platform`, :attr:`service_pack`, :attr:`service_pack_major`,
-  :attr:`service_pack_minor`, :attr:`suite_mask`, and
-  :attr:`product_type`.  (Contributed by Brian Curtin; :issue:`7766`.)
-
-* The :mod:`tarfile` module's default error handling has changed, to
-  no longer suppress fatal errors.  The default error level was previously 0,
-  which meant that errors would only result in a message being written to the
-  debug log, but because the debug log is not activated by default,
-  these errors go unnoticed.  The default error level is now 1,
-  which raises an exception if there's an error.
-  (Changed by Lars Gust?bel; :issue:`7357`.)
-
-  :mod:`tarfile` now supports filtering the :class:`~tarfile.TarInfo`
-  objects being added to a tar file.  When you call :meth:`~tarfile.TarFile.add`,
-  you may supply an optional *filter* argument
-  that's a callable.  The *filter* callable will be passed the
-  :class:`~tarfile.TarInfo` for every file being added, and can modify and return it.
-  If the callable returns ``None``, the file will be excluded from the
-  resulting archive.  This is more powerful than the existing
-  *exclude* argument, which has therefore been deprecated.
-  (Added by Lars Gust?bel; :issue:`6856`.)
-  The :class:`~tarfile.TarFile` class also now supports the context manager protocol.
-  (Added by Lars Gust?bel; :issue:`7232`.)
-
-* The :meth:`~threading.Event.wait` method of the :class:`threading.Event` class
-  now returns the internal flag on exit.  This means the method will usually
-  return true because :meth:`~threading.Event.wait` is supposed to block until the
-  internal flag becomes true.  The return value will only be false if
-  a timeout was provided and the operation timed out.
-  (Contributed by Tim Lesher; :issue:`1674032`.)
-
-* The Unicode database provided by the :mod:`unicodedata` module is
-  now used internally to determine which characters are numeric,
-  whitespace, or represent line breaks.  The database also
-  includes information from the :file:`Unihan.txt` data file (patch
-  by Anders Chrigstr?m and Amaury Forgeot d'Arc; :issue:`1571184`)
-  and has been updated to version 5.2.0 (updated by
-  Florent Xicluna; :issue:`8024`).
-
-* The :mod:`urlparse` module's :func:`~urlparse.urlsplit` now handles
-  unknown URL schemes in a fashion compliant with :rfc:`3986`: if the
-  URL is of the form ``"<something>://..."``, the text before the
-  ``://`` is treated as the scheme, even if it's a made-up scheme that
-  the module doesn't know about.  This change may break code that
-  worked around the old behaviour.  For example, Python 2.6.4 or 2.5
-  will return the following:
-
-    >>> import urlparse
-    >>> urlparse.urlsplit('invented://host/filename?query')
-    ('invented', '', '//host/filename?query', '', '')
-
-  Python 2.7 (and Python 2.6.5) will return:
-
-    >>> import urlparse
-    >>> urlparse.urlsplit('invented://host/filename?query')
-    ('invented', 'host', '/filename?query', '', '')
-
-  (Python 2.7 actually produces slightly different output, since it
-  returns a named tuple instead of a standard tuple.)
-
-  The :mod:`urlparse` module also supports IPv6 literal addresses as defined by
-  :rfc:`2732` (contributed by Senthil Kumaran; :issue:`2987`). ::
-
-    >>> urlparse.urlparse('http://[1080::8:800:200C:417A]/foo')
-    ParseResult(scheme='http', netloc='[1080::8:800:200C:417A]',
-                path='/foo', params='', query='', fragment='')
-
-* New class: the :class:`~weakref.WeakSet` class in the :mod:`weakref`
-  module is a set that only holds weak references to its elements; elements
-  will be removed once there are no references pointing to them.
-  (Originally implemented in Python 3.x by Raymond Hettinger, and backported
-  to 2.7 by Michael Foord.)
-
-* The ElementTree library, :mod:`xml.etree`, no longer escapes
-  ampersands and angle brackets when outputting an XML processing
-  instruction (which looks like ``<?xml-stylesheet href="#style1"?>``)
-  or comment (which looks like ``<!-- comment -->``).
-  (Patch by Neil Muller; :issue:`2746`.)
-
-* The XML-RPC client and server, provided by the :mod:`xmlrpclib` and
-  :mod:`SimpleXMLRPCServer` modules, have improved performance by
-  supporting HTTP/1.1 keep-alive and by optionally using gzip encoding
-  to compress the XML being exchanged.  The gzip compression is
-  controlled by the :attr:`encode_threshold` attribute of
-  :class:`SimpleXMLRPCRequestHandler`, which contains a size in bytes;
-  responses larger than this will be compressed.
-  (Contributed by Kristj?n Valur J?nsson; :issue:`6267`.)
-
-* The :mod:`zipfile` module's :class:`~zipfile.ZipFile` now supports the context
-  management protocol, so you can write ``with zipfile.ZipFile(...) as f:``.
-  (Contributed by Brian Curtin; :issue:`5511`.)
-
-  :mod:`zipfile` now also supports archiving empty directories and
-  extracts them correctly.  (Fixed by Kuba Wieczorek; :issue:`4710`.)
-  Reading files out of an archive is faster, and interleaving
-  :meth:`~zipfile.ZipFile.read` and :meth:`~zipfile.ZipFile.readline` now works correctly.
-  (Contributed by Nir Aides; :issue:`7610`.)
-
-  The :func:`~zipfile.is_zipfile` function now
-  accepts a file object, in addition to the path names accepted in earlier
-  versions.  (Contributed by Gabriel Genellina; :issue:`4756`.)
-
-  The :meth:`~zipfile.ZipFile.writestr` method now has an optional *compress_type* parameter
-  that lets you override the default compression method specified in the
-  :class:`~zipfile.ZipFile` constructor.  (Contributed by Ronald Oussoren;
-  :issue:`6003`.)
-
-
-.. ======================================================================
-.. whole new modules get described in subsections here
-
-
-.. _importlib-section:
-
-New module: importlib
-------------------------------
-
-Python 3.1 includes the :mod:`importlib` package, a re-implementation
-of the logic underlying Python's :keyword:`import` statement.
-:mod:`importlib` is useful for implementors of Python interpreters and
-to users who wish to write new importers that can participate in the
-import process.  Python 2.7 doesn't contain the complete
-:mod:`importlib` package, but instead has a tiny subset that contains
-a single function, :func:`~importlib.import_module`.
-
-``import_module(name, package=None)`` imports a module.  *name* is
-a string containing the module or package's name.  It's possible to do
-relative imports by providing a string that begins with a ``.``
-character, such as ``..utils.errors``.  For relative imports, the
-*package* argument must be provided and is the name of the package that
-will be used as the anchor for
-the relative import.  :func:`~importlib.import_module` both inserts the imported
-module into ``sys.modules`` and returns the module object.
-
-Here are some examples::
-
-    >>> from importlib import import_module
-    >>> anydbm = import_module('anydbm')  # Standard absolute import
-    >>> anydbm
-    <module 'anydbm' from '/p/python/Lib/anydbm.py'>
-    >>> # Relative import
-    >>> file_util = import_module('..file_util', 'distutils.command')
-    >>> file_util
-    <module 'distutils.file_util' from '/python/Lib/distutils/file_util.pyc'>
-
-:mod:`importlib` was implemented by Brett Cannon and introduced in
-Python 3.1.
-
-
-New module: sysconfig
----------------------------------
-
-The :mod:`sysconfig` module has been pulled out of the Distutils
-package, becoming a new top-level module in its own right.
-:mod:`sysconfig` provides functions for getting information about
-Python's build process: compiler switches, installation paths, the
-platform name, and whether Python is running from its source
-directory.
-
-Some of the functions in the module are:
-
-* :func:`~sysconfig.get_config_var` returns variables from Python's
-  Makefile and the :file:`pyconfig.h` file.
-* :func:`~sysconfig.get_config_vars` returns a dictionary containing
-  all of the configuration variables.
-* :func:`~sysconfig.getpath` returns the configured path for
-  a particular type of module: the standard library,
-  site-specific modules, platform-specific modules, etc.
-* :func:`~sysconfig.is_python_build` returns true if you're running a
-  binary from a Python source tree, and false otherwise.
-
-Consult the :mod:`sysconfig` documentation for more details and for
-a complete list of functions.
-
-The Distutils package and :mod:`sysconfig` are now maintained by Tarek
-Ziad?, who has also started a Distutils2 package (source repository at
-http://hg.python.org/distutils2/) for developing a next-generation
-version of Distutils.
-
-
-ttk: Themed Widgets for Tk
---------------------------
-
-Tcl/Tk 8.5 includes a set of themed widgets that re-implement basic Tk
-widgets but have a more customizable appearance and can therefore more
-closely resemble the native platform's widgets.  This widget
-set was originally called Tile, but was renamed to Ttk (for "themed Tk")
-on being added to Tcl/Tck release 8.5.
-
-To learn more, read the :mod:`ttk` module documentation.  You may also
-wish to read the Tcl/Tk manual page describing the
-Ttk theme engine, available at
-http://www.tcl.tk/man/tcl8.5/TkCmd/ttk_intro.htm. Some
-screenshots of the Python/Ttk code in use are at
-http://code.google.com/p/python-ttk/wiki/Screenshots.
-
-The :mod:`ttk` module was written by Guilherme Polo and added in
-:issue:`2983`.  An alternate version called ``Tile.py``, written by
-Martin Franklin and maintained by Kevin Walzer, was proposed for
-inclusion in :issue:`2618`, but the authors argued that Guilherme
-Polo's work was more comprehensive.
-
-
-.. _unittest-section:
-
-Updated module: unittest
----------------------------------
-
-The :mod:`unittest` module was greatly enhanced; many
-new features were added.  Most of these features were implemented
-by Michael Foord, unless otherwise noted.  The enhanced version of
-the module is downloadable separately for use with Python versions 2.4 to 2.6,
-packaged as the :mod:`unittest2` package, from
-http://pypi.python.org/pypi/unittest2.
-
-When used from the command line, the module can automatically discover
-tests.  It's not as fancy as `py.test <http://pytest.org>`__ or
-`nose <http://code.google.com/p/python-nose/>`__, but provides a simple way
-to run tests kept within a set of package directories.  For example,
-the following command will search the :file:`test/` subdirectory for
-any importable test files named ``test*.py``::
-
-   python -m unittest discover -s test
-
-Consult the :mod:`unittest` module documentation for more details.
-(Developed in :issue:`6001`.)
-
-The :func:`main` function supports some other new options:
-
-* :option:`-b` or :option:`--buffer` will buffer the standard output
-  and standard error streams during each test.  If the test passes,
-  any resulting output will be discarded; on failure, the buffered
-  output will be displayed.
-
-* :option:`-c` or :option:`--catch` will cause the control-C interrupt
-  to be handled more gracefully.  Instead of interrupting the test
-  process immediately, the currently running test will be completed
-  and then the partial results up to the interruption will be reported.
-  If you're impatient, a second press of control-C will cause an immediate
-  interruption.
-
-  This control-C handler tries to avoid causing problems when the code
-  being tested or the tests being run have defined a signal handler of
-  their own, by noticing that a signal handler was already set and
-  calling it.  If this doesn't work for you, there's a
-  :func:`removeHandler` decorator that can be used to mark tests that
-  should have the control-C handling disabled.
-
-* :option:`-f` or :option:`--failfast` makes
-  test execution stop immediately when a test fails instead of
-  continuing to execute further tests.  (Suggested by Cliff Dyer and
-  implemented by Michael Foord; :issue:`8074`.)
-
-The progress messages now show 'x' for expected failures
-and 'u' for unexpected successes when run in verbose mode.
-(Contributed by Benjamin Peterson.)
-
-Test cases can raise the :exc:`~unittest.SkipTest` exception to skip a
-test (:issue:`1034053`).
-
-The error messages for :meth:`~unittest.TestCase.assertEqual`,
-:meth:`~unittest.TestCase.assertTrue`, and :meth:`~unittest.TestCase.assertFalse`
-failures now provide more information.  If you set the
-:attr:`~unittest.TestCase.longMessage` attribute of your :class:`~unittest.TestCase` classes to
-True, both the standard error message and any additional message you
-provide will be printed for failures.  (Added by Michael Foord; :issue:`5663`.)
-
-The :meth:`~unittest.TestCase.assertRaises` method now
-returns a context handler when called without providing a callable
-object to run.  For example, you can write this::
-
-  with self.assertRaises(KeyError):
-      {}['foo']
-
-(Implemented by Antoine Pitrou; :issue:`4444`.)
-
-.. rev 78774
-
-Module- and class-level setup and teardown fixtures are now supported.
-Modules can contain :func:`~unittest.setUpModule` and :func:`~unittest.tearDownModule`
-functions.  Classes can have :meth:`~unittest.TestCase.setUpClass` and
-:meth:`~unittest.TestCase.tearDownClass` methods that must be defined as class methods
-(using ``@classmethod`` or equivalent).  These functions and
-methods are invoked when the test runner switches to a test case in a
-different module or class.
-
-The methods :meth:`~unittest.TestCase.addCleanup` and
-:meth:`~unittest.TestCase.doCleanups` were added.
-:meth:`~unittest.TestCase.addCleanup` lets you add cleanup functions that
-will be called unconditionally (after :meth:`~unittest.TestCase.setUp` if
-:meth:`~unittest.TestCase.setUp` fails, otherwise after :meth:`~unittest.TestCase.tearDown`). This allows
-for much simpler resource allocation and deallocation during tests
-(:issue:`5679`).
-
-A number of new methods were added that provide more specialized
-tests.  Many of these methods were written by Google engineers
-for use in their test suites; Gregory P. Smith, Michael Foord, and
-GvR worked on merging them into Python's version of :mod:`unittest`.
-
-* :meth:`~unittest.TestCase.assertIsNone` and :meth:`~unittest.TestCase.assertIsNotNone` take one
-  expression and verify that the result is or is not ``None``.
-
-* :meth:`~unittest.TestCase.assertIs` and :meth:`~unittest.TestCase.assertIsNot`
-  take two values and check whether the two values evaluate to the same object or not.
-  (Added by Michael Foord; :issue:`2578`.)
-
-* :meth:`~unittest.TestCase.assertIsInstance` and
-  :meth:`~unittest.TestCase.assertNotIsInstance` check whether
-  the resulting object is an instance of a particular class, or of
-  one of a tuple of classes.  (Added by Georg Brandl; :issue:`7031`.)
-
-* :meth:`~unittest.TestCase.assertGreater`, :meth:`~unittest.TestCase.assertGreaterEqual`,
-  :meth:`~unittest.TestCase.assertLess`, and :meth:`~unittest.TestCase.assertLessEqual` compare
-  two quantities.
-
-* :meth:`~unittest.TestCase.assertMultiLineEqual` compares two strings, and if they're
-  not equal, displays a helpful comparison that highlights the
-  differences in the two strings.  This comparison is now used by
-  default when Unicode strings are compared with :meth:`~unittest.TestCase.assertEqual`.
-
-* :meth:`~unittest.TestCase.assertRegexpMatches` and
-  :meth:`~unittest.TestCase.assertNotRegexpMatches` checks whether the
-  first argument is a string matching or not matching the regular
-  expression provided as the second argument (:issue:`8038`).
-
-* :meth:`~unittest.TestCase.assertRaisesRegexp` checks whether a particular exception
-  is raised, and then also checks that the string representation of
-  the exception matches the provided regular expression.
-
-* :meth:`~unittest.TestCase.assertIn` and :meth:`~unittest.TestCase.assertNotIn`
-  tests whether *first* is or is not in  *second*.
-
-* :meth:`~unittest.TestCase.assertItemsEqual` tests whether two provided sequences
-  contain the same elements.
-
-* :meth:`~unittest.TestCase.assertSetEqual` compares whether two sets are equal, and
-  only reports the differences between the sets in case of error.
-
-* Similarly, :meth:`~unittest.TestCase.assertListEqual` and :meth:`~unittest.TestCase.assertTupleEqual`
-  compare the specified types and explain any differences without necessarily
-  printing their full values; these methods are now used by default
-  when comparing lists and tuples using :meth:`~unittest.TestCase.assertEqual`.
-  More generally, :meth:`~unittest.TestCase.assertSequenceEqual` compares two sequences
-  and can optionally check whether both sequences are of a
-  particular type.
-
-* :meth:`~unittest.TestCase.assertDictEqual` compares two dictionaries and reports the
-  differences; it's now used by default when you compare two dictionaries
-  using :meth:`~unittest.TestCase.assertEqual`.  :meth:`~unittest.TestCase.assertDictContainsSubset` checks whether
-  all of the key/value pairs in *first* are found in *second*.
-
-* :meth:`~unittest.TestCase.assertAlmostEqual` and :meth:`~unittest.TestCase.assertNotAlmostEqual` test
-  whether *first* and *second* are approximately equal.  This method
-  can either round their difference to an optionally-specified number
-  of *places* (the default is 7) and compare it to zero, or require
-  the difference to be smaller than a supplied *delta* value.
-
-* :meth:`~unittest.TestLoader.loadTestsFromName` properly honors the
-  :attr:`~unittest.TestLoader.suiteClass` attribute of
-  the :class:`~unittest.TestLoader`. (Fixed by Mark Roddy; :issue:`6866`.)
-
-* A new hook lets you extend the :meth:`~unittest.TestCase.assertEqual` method to handle
-  new data types.  The :meth:`~unittest.TestCase.addTypeEqualityFunc` method takes a type
-  object and a function. The function will be used when both of the
-  objects being compared are of the specified type.  This function
-  should compare the two objects and raise an exception if they don't
-  match; it's a good idea for the function to provide additional
-  information about why the two objects aren't matching, much as the new
-  sequence comparison methods do.
-
-:func:`unittest.main` now takes an optional ``exit`` argument.  If
-False, :func:`~unittest.main` doesn't call :func:`sys.exit`, allowing
-:func:`main` to be used from the interactive interpreter.
-(Contributed by J. Pablo Fern?ndez; :issue:`3379`.)
-
-:class:`~unittest.TestResult` has new :meth:`~unittest.TestResult.startTestRun` and
-:meth:`~unittest.TestResult.stopTestRun` methods that are called immediately before
-and after a test run.  (Contributed by Robert Collins; :issue:`5728`.)
-
-With all these changes, the :file:`unittest.py` was becoming awkwardly
-large, so the module was turned into a package and the code split into
-several files (by Benjamin Peterson).  This doesn't affect how the
-module is imported or used.
-
-.. seealso::
-
-  http://www.voidspace.org.uk/python/articles/unittest2.shtml
-    Describes the new features, how to use them, and the
-    rationale for various design decisions.  (By Michael Foord.)
-
-.. _elementtree-section:
-
-Updated module: ElementTree 1.3
----------------------------------
-
-The version of the ElementTree library included with Python was updated to
-version 1.3.  Some of the new features are:
-
-* The various parsing functions now take a *parser* keyword argument
-  giving an :class:`XMLParser` instance that will
-  be used.  This makes it possible to override the file's internal encoding::
-
-    p = ET.XMLParser(encoding='utf-8')
-    t = ET.XML("""<root/>""", parser=p)
-
-  Errors in parsing XML now raise a :exc:`ParseError` exception, whose
-  instances have a :attr:`position` attribute
-  containing a (*line*, *column*) tuple giving the location of the problem.
-
-* ElementTree's code for converting trees to a string has been
-  significantly reworked, making it roughly twice as fast in many
-  cases.  The :class:`ElementTree` :meth:`write` and :class:`Element`
-  :meth:`write` methods now have a *method* parameter that can be
-  "xml" (the default), "html", or "text".  HTML mode will output empty
-  elements as ``<empty></empty>`` instead of ``<empty/>``, and text
-  mode will skip over elements and only output the text chunks.  If
-  you set the :attr:`tag` attribute of an element to ``None`` but
-  leave its children in place, the element will be omitted when the
-  tree is written out, so you don't need to do more extensive rearrangement
-  to remove a single element.
-
-  Namespace handling has also been improved.  All ``xmlns:<whatever>``
-  declarations are now output on the root element, not scattered throughout
-  the resulting XML.  You can set the default namespace for a tree
-  by setting the :attr:`default_namespace` attribute and can
-  register new prefixes with :meth:`register_namespace`.  In XML mode,
-  you can use the true/false *xml_declaration* parameter to suppress the
-  XML declaration.
-
-* New :class:`Element` method: :meth:`extend` appends the items from a
-  sequence to the element's children.  Elements themselves behave like
-  sequences, so it's easy to move children from one element to
-  another::
-
-    from xml.etree import ElementTree as ET
-
-    t = ET.XML("""<list>
-      <item>1</item> <item>2</item>  <item>3</item>
-    </list>""")
-    new = ET.XML('<root/>')
-    new.extend(t)
-
-    # Outputs <root><item>1</item>...</root>
-    print ET.tostring(new)
-
-* New :class:`Element` method: :meth:`iter` yields the children of the
-  element as a generator.  It's also possible to write ``for child in
-  elem:`` to loop over an element's children.  The existing method
-  :meth:`getiterator` is now deprecated, as is :meth:`getchildren`
-  which constructs and returns a list of children.
-
-* New :class:`Element` method: :meth:`itertext` yields all chunks of
-  text that are descendants of the element.  For example::
-
-    t = ET.XML("""<list>
-      <item>1</item> <item>2</item>  <item>3</item>
-    </list>""")
-
-    # Outputs ['\n  ', '1', ' ', '2', '  ', '3', '\n']
-    print list(t.itertext())
-
-* Deprecated: using an element as a Boolean (i.e., ``if elem:``) would
-  return true if the element had any children, or false if there were
-  no children.  This behaviour is confusing -- ``None`` is false, but
-  so is a childless element? -- so it will now trigger a
-  :exc:`FutureWarning`.  In your code, you should be explicit: write
-  ``len(elem) != 0`` if you're interested in the number of children,
-  or ``elem is not None``.
-
-Fredrik Lundh develops ElementTree and produced the 1.3 version;
-you can read his article describing 1.3 at
-http://effbot.org/zone/elementtree-13-intro.htm.
-Florent Xicluna updated the version included with
-Python, after discussions on python-dev and in :issue:`6472`.)
-
-.. ======================================================================
-
-
-Build and C API Changes
-=======================
-
-Changes to Python's build process and to the C API include:
-
-* The latest release of the GNU Debugger, GDB 7, can be `scripted
-  using Python
-  <http://sourceware.org/gdb/current/onlinedocs/gdb/Python.html>`__.
-  When you begin debugging an executable program P, GDB will look for
-  a file named ``P-gdb.py`` and automatically read it.  Dave Malcolm
-  contributed a :file:`python-gdb.py` that adds a number of
-  commands useful when debugging Python itself.  For example,
-  ``py-up`` and ``py-down`` go up or down one Python stack frame,
-  which usually corresponds to several C stack frames.  ``py-print``
-  prints the value of a Python variable, and ``py-bt`` prints the
-  Python stack trace.  (Added as a result of :issue:`8032`.)
-
-* If you use the :file:`.gdbinit` file provided with Python,
-  the "pyo" macro in the 2.7 version now works correctly when the thread being
-  debugged doesn't hold the GIL; the macro now acquires it before printing.
-  (Contributed by Victor Stinner; :issue:`3632`.)
-
-* :cfunc:`Py_AddPendingCall` is now thread-safe, letting any
-  worker thread submit notifications to the main Python thread.  This
-  is particularly useful for asynchronous IO operations.
-  (Contributed by Kristj?n Valur J?nsson; :issue:`4293`.)
-
-* New function: :cfunc:`PyCode_NewEmpty` creates an empty code object;
-  only the filename, function name, and first line number are required.
-  This is useful for extension modules that are attempting to
-  construct a more useful traceback stack.  Previously such
-  extensions needed to call :cfunc:`PyCode_New`, which had many
-  more arguments.  (Added by Jeffrey Yasskin.)
-
-* New function: :cfunc:`PyErr_NewExceptionWithDoc` creates a new
-  exception class, just as the existing :cfunc:`PyErr_NewException` does,
-  but takes an extra ``char *`` argument containing the docstring for the
-  new exception class.  (Added by 'lekma' on the Python bug tracker;
-  :issue:`7033`.)
-
-* New function: :cfunc:`PyFrame_GetLineNumber` takes a frame object
-  and returns the line number that the frame is currently executing.
-  Previously code would need to get the index of the bytecode
-  instruction currently executing, and then look up the line number
-  corresponding to that address.  (Added by Jeffrey Yasskin.)
-
-* New functions: :cfunc:`PyLong_AsLongAndOverflow` and
-  :cfunc:`PyLong_AsLongLongAndOverflow`  approximates a Python long
-  integer as a C :ctype:`long` or :ctype:`long long`.
-  If the number is too large to fit into
-  the output type, an *overflow* flag is set and returned to the caller.
-  (Contributed by Case Van Horsen; :issue:`7528` and :issue:`7767`.)
-
-* New function: stemming from the rewrite of string-to-float conversion,
-  a new :cfunc:`PyOS_string_to_double` function was added.  The old
-  :cfunc:`PyOS_ascii_strtod` and :cfunc:`PyOS_ascii_atof` functions
-  are now deprecated.
-
-* New function: :cfunc:`PySys_SetArgvEx` sets the value of
-  ``sys.argv`` and can optionally update ``sys.path`` to include the
-  directory containing the script named by ``sys.argv[0]`` depending
-  on the value of an *updatepath* parameter.
-
-  This function was added to close a security hole for applications
-  that embed Python.  The old function, :cfunc:`PySys_SetArgv`, would
-  always update ``sys.path``, and sometimes it would add the current
-  directory.  This meant that, if you ran an application embedding
-  Python in a directory controlled by someone else, attackers could
-  put a Trojan-horse module in the directory (say, a file named
-  :file:`os.py`) that your application would then import and run.
-
-  If you maintain a C/C++ application that embeds Python, check
-  whether you're calling :cfunc:`PySys_SetArgv` and carefully consider
-  whether the application should be using :cfunc:`PySys_SetArgvEx`
-  with *updatepath* set to false.
-
-  Security issue reported as `CVE-2008-5983
-  <http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2008-5983>`_;
-  discussed in :issue:`5753`, and fixed by Antoine Pitrou.
-
-* New macros: the Python header files now define the following macros:
-  :cmacro:`Py_ISALNUM`,
-  :cmacro:`Py_ISALPHA`,
-  :cmacro:`Py_ISDIGIT`,
-  :cmacro:`Py_ISLOWER`,
-  :cmacro:`Py_ISSPACE`,
-  :cmacro:`Py_ISUPPER`,
-  :cmacro:`Py_ISXDIGIT`,
-  and :cmacro:`Py_TOLOWER`, :cmacro:`Py_TOUPPER`.
-  All of these functions are analogous to the C
-  standard macros for classifying characters, but ignore the current
-  locale setting, because in
-  several places Python needs to analyze characters in a
-  locale-independent way.  (Added by Eric Smith;
-  :issue:`5793`.)
-
-  .. XXX these macros don't seem to be described in the c-api docs.
-
-* Removed function: :cmacro:`PyEval_CallObject` is now only available
-  as a macro.  A function version was being kept around to preserve
-  ABI linking compatibility, but that was in 1997; it can certainly be
-  deleted by now.  (Removed by Antoine Pitrou; :issue:`8276`.)
-
-* New format codes: the :cfunc:`PyFormat_FromString`,
-  :cfunc:`PyFormat_FromStringV`, and :cfunc:`PyErr_Format` functions now
-  accept ``%lld`` and ``%llu`` format codes for displaying
-  C's :ctype:`long long` types.
-  (Contributed by Mark Dickinson; :issue:`7228`.)
-
-* The complicated interaction between threads and process forking has
-  been changed.  Previously, the child process created by
-  :func:`os.fork` might fail because the child is created with only a
-  single thread running, the thread performing the :func:`os.fork`.
-  If other threads were holding a lock, such as Python's import lock,
-  when the fork was performed, the lock would still be marked as
-  "held" in the new process.  But in the child process nothing would
-  ever release the lock, since the other threads weren't replicated,
-  and the child process would no longer be able to perform imports.
-
-  Python 2.7 acquires the import lock before performing an
-  :func:`os.fork`, and will also clean up any locks created using the
-  :mod:`threading` module.  C extension modules that have internal
-  locks, or that call :cfunc:`fork()` themselves, will not benefit
-  from this clean-up.
-
-  (Fixed by Thomas Wouters; :issue:`1590864`.)
-
-* The :cfunc:`Py_Finalize` function now calls the internal
-  :func:`threading._shutdown` function; this prevents some exceptions from
-  being raised when an interpreter shuts down.
-  (Patch by Adam Olsen; :issue:`1722344`.)
-
-* When using the :ctype:`PyMemberDef` structure to define attributes
-  of a type, Python will no longer let you try to delete or set a
-  :const:`T_STRING_INPLACE` attribute.
-
-  .. rev 79644
-
-* Global symbols defined by the :mod:`ctypes` module are now prefixed
-  with ``Py``, or with ``_ctypes``.  (Implemented by Thomas
-  Heller; :issue:`3102`.)
-
-* New configure option: the :option:`--with-system-expat` switch allows
-  building the :mod:`pyexpat` module to use the system Expat library.
-  (Contributed by Arfrever Frehtes Taifersar Arahesis; :issue:`7609`.)
-
-* New configure option: the
-  :option:`--with-valgrind` option will now disable the pymalloc
-  allocator, which is difficult for the Valgrind memory-error detector
-  to analyze correctly.
-  Valgrind will therefore be better at detecting memory leaks and
-  overruns. (Contributed by James Henstridge; :issue:`2422`.)
-
-* New configure option: you can now supply an empty string to
-  :option:`--with-dbmliborder=` in order to disable all of the various
-  DBM modules.  (Added by Arfrever Frehtes Taifersar Arahesis;
-  :issue:`6491`.)
-
-* The :program:`configure` script now checks for floating-point rounding bugs
-  on certain 32-bit Intel chips and defines a :cmacro:`X87_DOUBLE_ROUNDING`
-  preprocessor definition.  No code currently uses this definition,
-  but it's available if anyone wishes to use it.
-  (Added by Mark Dickinson; :issue:`2937`.)
-
-  :program:`configure` also now sets a :envvar:`LDCXXSHARED` Makefile
-  variable for supporting C++ linking.  (Contributed by Arfrever
-  Frehtes Taifersar Arahesis; :issue:`1222585`.)
-
-* The build process now creates the necessary files for pkg-config
-  support.  (Contributed by Clinton Roy; :issue:`3585`.)
-
-* The build process now supports Subversion 1.7.  (Contributed by
-  Arfrever Frehtes Taifersar Arahesis; :issue:`6094`.)
-
-
-.. _whatsnew27-capsules:
-
-Capsules
--------------------
-
-Python 3.1 adds a new C datatype, :ctype:`PyCapsule`, for providing a
-C API to an extension module.  A capsule is essentially the holder of
-a C ``void *`` pointer, and is made available as a module attribute; for
-example, the :mod:`socket` module's API is exposed as ``socket.CAPI``,
-and :mod:`unicodedata` exposes ``ucnhash_CAPI``.  Other extensions
-can import the module, access its dictionary to get the capsule
-object, and then get the ``void *`` pointer, which will usually point
-to an array of pointers to the module's various API functions.
-
-There is an existing data type already used for this,
-:ctype:`PyCObject`, but it doesn't provide type safety.  Evil code
-written in pure Python could cause a segmentation fault by taking a
-:ctype:`PyCObject` from module A and somehow substituting it for the
-:ctype:`PyCObject` in module B.   Capsules know their own name,
-and getting the pointer requires providing the name::
-
-   void *vtable;
-
-   if (!PyCapsule_IsValid(capsule, "mymodule.CAPI") {
-           PyErr_SetString(PyExc_ValueError, "argument type invalid");
-           return NULL;
-   }
-
-   vtable = PyCapsule_GetPointer(capsule, "mymodule.CAPI");
-
-You are assured that ``vtable`` points to whatever you're expecting.
-If a different capsule was passed in, :cfunc:`PyCapsule_IsValid` would
-detect the mismatched name and return false.  Refer to
-:ref:`using-capsules` for more information on using these objects.
-
-Python 2.7 now uses capsules internally to provide various
-extension-module APIs, but the :cfunc:`PyCObject_AsVoidPtr` was
-modified to handle capsules, preserving compile-time compatibility
-with the :ctype:`CObject` interface.  Use of
-:cfunc:`PyCObject_AsVoidPtr` will signal a
-:exc:`PendingDeprecationWarning`, which is silent by default.
-
-Implemented in Python 3.1 and backported to 2.7 by Larry Hastings;
-discussed in :issue:`5630`.
-
-
-.. ======================================================================
-
-Port-Specific Changes: Windows
------------------------------------
-
-* The :mod:`msvcrt` module now contains some constants from
-  the :file:`crtassem.h` header file:
-  :data:`CRT_ASSEMBLY_VERSION`,
-  :data:`VC_ASSEMBLY_PUBLICKEYTOKEN`,
-  and :data:`LIBRARIES_ASSEMBLY_NAME_PREFIX`.
-  (Contributed by David Cournapeau; :issue:`4365`.)
-
-* The :mod:`_winreg` module for accessing the registry now implements
-  the :func:`CreateKeyEx` and :func:`DeleteKeyEx` functions, extended
-  versions of previously-supported functions that take several extra
-  arguments.  The :func:`DisableReflectionKey`,
-  :func:`EnableReflectionKey`, and :func:`QueryReflectionKey` were also
-  tested and documented.
-  (Implemented by Brian Curtin: :issue:`7347`.)
-
-* The new :cfunc:`_beginthreadex` API is used to start threads, and
-  the native thread-local storage functions are now used.
-  (Contributed by Kristj?n Valur J?nsson; :issue:`3582`.)
-
-* The :func:`os.kill` function now works on Windows.  The signal value
-  can be the constants :const:`CTRL_C_EVENT`,
-  :const:`CTRL_BREAK_EVENT`, or any integer.  The first two constants
-  will send Control-C and Control-Break keystroke events to
-  subprocesses; any other value will use the :cfunc:`TerminateProcess`
-  API.  (Contributed by Miki Tebeka; :issue:`1220212`.)
-
-* The :func:`os.listdir` function now correctly fails
-  for an empty path.  (Fixed by Hirokazu Yamamoto; :issue:`5913`.)
-
-* The :mod:`mimelib` module will now read the MIME database from
-  the Windows registry when initializing.
-  (Patch by Gabriel Genellina; :issue:`4969`.)
-
-.. ======================================================================
-
-Port-Specific Changes: Mac OS X
------------------------------------
-
-* The path ``/Library/Python/2.7/site-packages`` is now appended to
-  ``sys.path``, in order to share added packages between the system
-  installation and a user-installed copy of the same version.
-  (Changed by Ronald Oussoren; :issue:`4865`.)
-
-Port-Specific Changes: FreeBSD
------------------------------------
-
-* FreeBSD 7.1's :const:`SO_SETFIB` constant, used with
-  :func:`~socket.getsockopt`/:func:`~socket.setsockopt` to select an
-  alternate routing table, is now available in the :mod:`socket`
-  module.  (Added by Kyle VanderBeek; :issue:`8235`.)
-
-Other Changes and Fixes
-=======================
-
-* Two benchmark scripts, :file:`iobench` and :file:`ccbench`, were
-  added to the :file:`Tools` directory.  :file:`iobench` measures the
-  speed of the built-in file I/O objects returned by :func:`open`
-  while performing various operations, and :file:`ccbench` is a
-  concurrency benchmark that tries to measure computing throughput,
-  thread switching latency, and IO processing bandwidth when
-  performing several tasks using a varying number of threads.
-
-* The :file:`Tools/i18n/msgfmt.py` script now understands plural
-  forms in :file:`.po` files.  (Fixed by Martin von L?wis;
-  :issue:`5464`.)
-
-* When importing a module from a :file:`.pyc` or :file:`.pyo` file
-  with an existing :file:`.py` counterpart, the :attr:`co_filename`
-  attributes of the resulting code objects are overwritten when the
-  original filename is obsolete.  This can happen if the file has been
-  renamed, moved, or is accessed through different paths.  (Patch by
-  Ziga Seilnacht and Jean-Paul Calderone; :issue:`1180193`.)
-
-* The :file:`regrtest.py` script now takes a :option:`--randseed=`
-  switch that takes an integer that will be used as the random seed
-  for the :option:`-r` option that executes tests in random order.
-  The :option:`-r` option also reports the seed that was used
-  (Added by Collin Winter.)
-
-* Another :file:`regrtest.py` switch is :option:`-j`, which
-  takes an integer specifying how many tests run in parallel. This
-  allows reducing the total runtime on multi-core machines.
-  This option is compatible with several other options, including the
-  :option:`-R` switch which is known to produce long runtimes.
-  (Added by Antoine Pitrou, :issue:`6152`.)  This can also be used
-  with a new :option:`-F` switch that runs selected tests in a loop
-  until they fail.  (Added by Antoine Pitrou; :issue:`7312`.)
-
-* When executed as a script, the :file:`py_compile.py` module now
-  accepts ``'-'`` as an argument, which will read standard input for
-  the list of filenames to be compiled.  (Contributed by Piotr
-  O?arowski; :issue:`8233`.)
-
-.. ======================================================================
-
-Porting to Python 2.7
-=====================
-
-This section lists previously described changes and other bugfixes
-that may require changes to your code:
-
-* The :func:`range` function processes its arguments more
-  consistently; it will now call :meth:`__int__` on non-float,
-  non-integer arguments that are supplied to it.  (Fixed by Alexander
-  Belopolsky; :issue:`1533`.)
-
-* The string :meth:`format` method changed the default precision used
-  for floating-point and complex numbers from 6 decimal
-  places to 12, which matches the precision used by :func:`str`.
-  (Changed by Eric Smith; :issue:`5920`.)
-
-* Because of an optimization for the :keyword:`with` statement, the special
-  methods :meth:`__enter__` and :meth:`__exit__` must belong to the object's
-  type, and cannot be directly attached to the object's instance.  This
-  affects new-style classes (derived from :class:`object`) and C extension
-  types.  (:issue:`6101`.)
-
-* Due to a bug in Python 2.6, the *exc_value* parameter to
-  :meth:`__exit__` methods was often the string representation of the
-  exception, not an instance.  This was fixed in 2.7, so *exc_value*
-  will be an instance as expected.  (Fixed by Florent Xicluna;
-  :issue:`7853`.)
-
-* When a restricted set of attributes were set using ``__slots__``,
-  deleting an unset attribute would not raise :exc:`AttributeError`
-  as you would expect.  Fixed by Benjamin Peterson; :issue:`7604`.)
-
-In the standard library:
-
-* Operations with :class:`datetime` instances that resulted in a year
-  falling outside the supported range didn't always raise
-  :exc:`OverflowError`.  Such errors are now checked more carefully
-  and will now raise the exception. (Reported by Mark Leander, patch
-  by Anand B. Pillai and Alexander Belopolsky; :issue:`7150`.)
-
-* When using :class:`Decimal` instances with a string's
-  :meth:`format` method, the default alignment was previously
-  left-alignment.  This has been changed to right-alignment, which might
-  change the output of your programs.
-  (Changed by Mark Dickinson; :issue:`6857`.)
-
-  Comparisons involving a signaling NaN value (or ``sNAN``) now signal
-  :const:`InvalidOperation` instead of silently returning a true or
-  false value depending on the comparison operator.  Quiet NaN values
-  (or ``NaN``) are now hashable.  (Fixed by Mark Dickinson;
-  :issue:`7279`.)
-
-* The ElementTree library, :mod:`xml.etree`, no longer escapes
-  ampersands and angle brackets when outputting an XML processing
-  instruction (which looks like `<?xml-stylesheet href="#style1"?>`)
-  or comment (which looks like `<!-- comment -->`).
-  (Patch by Neil Muller; :issue:`2746`.)
-
-* The :meth:`readline` method of :class:`StringIO` objects now does
-  nothing when a negative length is requested, as other file-like
-  objects do.  (:issue:`7348`).
-
-* The :mod:`syslog` module will now use the value of ``sys.argv[0]`` as the
-  identifier instead of the previous default value of ``'python'``.
-  (Changed by Sean Reifschneider; :issue:`8451`.)
-
-* The :mod:`tarfile` module's default error handling has changed, to
-  no longer suppress fatal errors.  The default error level was previously 0,
-  which meant that errors would only result in a message being written to the
-  debug log, but because the debug log is not activated by default,
-  these errors go unnoticed.  The default error level is now 1,
-  which raises an exception if there's an error.
-  (Changed by Lars Gust?bel; :issue:`7357`.)
-
-* The :mod:`urlparse` module's :func:`~urlparse.urlsplit` now handles
-  unknown URL schemes in a fashion compliant with :rfc:`3986`: if the
-  URL is of the form ``"<something>://..."``, the text before the
-  ``://`` is treated as the scheme, even if it's a made-up scheme that
-  the module doesn't know about.  This change may break code that
-  worked around the old behaviour.  For example, Python 2.6.4 or 2.5
-  will return the following:
-
-    >>> import urlparse
-    >>> urlparse.urlsplit('invented://host/filename?query')
-    ('invented', '', '//host/filename?query', '', '')
-
-  Python 2.7 (and Python 2.6.5) will return:
-
-    >>> import urlparse
-    >>> urlparse.urlsplit('invented://host/filename?query')
-    ('invented', 'host', '/filename?query', '', '')
-
-  (Python 2.7 actually produces slightly different output, since it
-  returns a named tuple instead of a standard tuple.)
-
-For C extensions:
-
-* C extensions that use integer format codes with the ``PyArg_Parse*``
-  family of functions will now raise a :exc:`TypeError` exception
-  instead of triggering a :exc:`DeprecationWarning` (:issue:`5080`).
-
-* Use the new :cfunc:`PyOS_string_to_double` function instead of the old
-  :cfunc:`PyOS_ascii_strtod` and :cfunc:`PyOS_ascii_atof` functions,
-  which are now deprecated.
-
-For applications that embed Python:
-
-* The :cfunc:`PySys_SetArgvEx` function was added, letting
-  applications close a security hole when the existing
-  :cfunc:`PySys_SetArgv` function was used.  Check whether you're
-  calling :cfunc:`PySys_SetArgv` and carefully consider whether the
-  application should be using :cfunc:`PySys_SetArgvEx` with
-  *updatepath* set to false.
-
-.. ======================================================================
-
-
-.. _acks27:
-
-Acknowledgements
-================
-
-The author would like to thank the following people for offering
-suggestions, corrections and assistance with various drafts of this
-article: Nick Coghlan, Philip Jenvey, Ryan Lovett, R. David Murray,
-Hugh Secker-Walker.
-

Modified: python/branches/release31-maint/Doc/whatsnew/index.rst
==============================================================================
--- python/branches/release31-maint/Doc/whatsnew/index.rst	(original)
+++ python/branches/release31-maint/Doc/whatsnew/index.rst	Sun Jul 11 12:41:44 2010
@@ -13,7 +13,6 @@
 
    3.1.rst
    3.0.rst
-   2.7.rst
    2.6.rst
    2.5.rst
    2.4.rst

From python-checkins at python.org  Sun Jul 11 12:53:38 2010
From: python-checkins at python.org (victor.stinner)
Date: Sun, 11 Jul 2010 12:53:38 +0200 (CEST)
Subject: [Python-checkins] r82810 - in python/branches/import_unicode:
	Objects/codeobject.c Objects/object.c Python/pythonrun.c
Message-ID: <20100711105338.DCF1DEE9D3@mail.python.org>

Author: victor.stinner
Date: Sun Jul 11 12:53:38 2010
New Revision: 82810

Log:
Register code objects in a list

And convert code objects filename when setting the file system encoding.

Modified:
   python/branches/import_unicode/Objects/codeobject.c
   python/branches/import_unicode/Objects/object.c
   python/branches/import_unicode/Python/pythonrun.c

Modified: python/branches/import_unicode/Objects/codeobject.c
==============================================================================
--- python/branches/import_unicode/Objects/codeobject.c	(original)
+++ python/branches/import_unicode/Objects/codeobject.c	Sun Jul 11 12:53:38 2010
@@ -5,6 +5,8 @@
 #define NAME_CHARS \
     "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ_abcdefghijklmnopqrstuvwxyz"
 
+PyObject *_Py_code_object_list = NULL;
+
 /* all_name_chars(s): true iff all chars in s are valid NAME_CHARS */
 
 static int
@@ -42,6 +44,14 @@
 }
 
 
+void _PyCode_ClearList(void)
+{
+    if (_Py_code_object_list == NULL)
+        return;
+    Py_DECREF(_Py_code_object_list);
+    _Py_code_object_list = NULL;
+}
+
 PyCodeObject *
 PyCode_New(int argcount, int kwonlyargcount,
            int nlocals, int stacksize, int flags,
@@ -109,6 +119,13 @@
         co->co_lnotab = lnotab;
         co->co_zombieframe = NULL;
         co->co_weakreflist = NULL;
+
+        if (_Py_code_object_list != NULL) {
+            if (PyList_Append(_Py_code_object_list, (PyObject*)co)) {
+                Py_DECREF(co);
+                return NULL;
+            }
+        }
     }
     return co;
 }

Modified: python/branches/import_unicode/Objects/object.c
==============================================================================
--- python/branches/import_unicode/Objects/object.c	(original)
+++ python/branches/import_unicode/Objects/object.c	Sun Jul 11 12:53:38 2010
@@ -9,6 +9,8 @@
 extern "C" {
 #endif
 
+extern PyObject *_Py_code_object_list;
+
 #ifdef Py_REF_DEBUG
 Py_ssize_t _Py_RefTotal;
 
@@ -1598,6 +1600,10 @@
     if (PyType_Ready(&PyCode_Type) < 0)
         Py_FatalError("Can't initialize code type");
 
+    _Py_code_object_list = PyList_New(0);
+    if (_Py_code_object_list == NULL)
+        Py_FatalError("Can't initialize code type");
+
     if (PyType_Ready(&PyFrame_Type) < 0)
         Py_FatalError("Can't initialize frame type");
 

Modified: python/branches/import_unicode/Python/pythonrun.c
==============================================================================
--- python/branches/import_unicode/Python/pythonrun.c	(original)
+++ python/branches/import_unicode/Python/pythonrun.c	Sun Jul 11 12:53:38 2010
@@ -55,6 +55,8 @@
 
 extern grammar _PyParser_Grammar; /* From graminit.c */
 
+extern PyObject *_Py_code_object_list; /* From codeobject.c */
+
 /* Forward */
 static void initmain(void);
 static void initfsencoding(void);
@@ -74,6 +76,7 @@
 extern void _PyUnicode_Fini(void);
 extern int _PyLong_Init(void);
 extern void PyLong_Fini(void);
+extern void _PyCode_ClearList(void);
 
 #ifdef WITH_THREAD
 extern void _PyGILState_Init(PyInterpreterState *, PyThreadState *);
@@ -697,28 +700,27 @@
 }
 
 static PyObject*
-reencode_filename(PyObject *module, PyObject *file, const char *new_encoding)
+reencode_filename(PyObject *file, const char *new_encoding,
+                  const char *errors)
 {
     PyObject *file_bytes = NULL;
     PyObject *new_file = NULL;
 
-    file_bytes = PyUnicode_AsEncodedString(
-            file,
-            Py_FileSystemDefaultEncoding,
-            "surrogateescape");
+    file_bytes = PyUnicode_EncodeFSDefault(file);
     if (file_bytes == NULL)
         return NULL;
     new_file = PyUnicode_Decode(
-            PyBytes_AsString(file_bytes),
-            PyBytes_GET_SIZE(file_bytes),
-            new_encoding,
-            "surrogateescape");
+        PyBytes_AsString(file_bytes),
+        PyBytes_GET_SIZE(file_bytes),
+        new_encoding,
+        "surrogateescape");
     Py_DECREF(file_bytes);
     return new_file;
 }
 
 static int
-reencode_module_path(PyObject *module, PyObject *path, const char *new_encoding)
+reencode_module_path(PyObject *module, PyObject *path,
+                     const char *new_encoding, const char *errors)
 {
     PyObject *filename;
     PyObject *new_filename;
@@ -730,7 +732,7 @@
         if (filename == NULL)
             return 1;
 
-        new_filename = reencode_filename(module, filename, new_encoding);
+        new_filename = reencode_filename(filename, new_encoding, errors);
         if (new_filename == NULL)
             return 1;
         if (PyList_SetItem(path, i, new_filename))
@@ -750,14 +752,23 @@
     PyObject *module_dict = NULL;
     PyObject *file;
     PyObject *path;
+    Py_ssize_t i, len;
     int ret = 1;
+    char *errors;
+    PyObject *new_file;
+    PyCodeObject *co;
 
     /* FIXME: Re-encode PySys_GetObject("path_importer_cache") keys? */
-    /* FIXME: Re-encode co_filename of all code objects! */
 
-    if (strcmp(new_encoding, PyUnicode_GetDefaultEncoding()) == 0)
+    /* PyUnicode_DecodeFSDefault and PyUnicode_EncodeFSDefault use utf-8
+     * if Py_FileSystemDefaultEncoding is not set */
+    if (strcmp(new_encoding, "utf-8") == 0)
         return 0;
 
+    if (strcmp(new_encoding, "mbcs") != 0)
+        errors = "surrogateescape";
+    else
+        errors = NULL;
 
     values = PyObject_CallMethod(modules, "values", "");
     if (values == NULL)
@@ -771,7 +782,7 @@
     for (module = PyIter_Next(iter); module != NULL; module = PyIter_Next(iter)) {
         file = PyModule_GetFilenameObject(module);
         if (file != NULL) {
-            PyObject *new_file = reencode_filename(module, file, new_encoding);
+            new_file = reencode_filename(file, new_encoding, errors);
             if (new_file == NULL)
                 goto error;
             if (PyObject_SetAttrString(module, "__file__", new_file)) {
@@ -788,7 +799,7 @@
             goto error;
         path = PyDict_GetItemString(module_dict, "__path__");
         if (path != NULL) {
-            if (reencode_module_path(module, path, new_encoding))
+            if (reencode_module_path(module, path, new_encoding, errors))
                 goto error;
         } else {
             PyErr_Clear();
@@ -797,6 +808,18 @@
         Py_CLEAR(module);
     }
 
+    len = Py_SIZE(_Py_code_object_list);
+    for (i=0; i < len; i++) {
+        co = (PyCodeObject *)PyList_GET_ITEM(_Py_code_object_list, i);
+
+        new_file = reencode_filename(co->co_filename, new_encoding, errors);
+        if (new_file == NULL)
+            return 1;
+        Py_DECREF(co->co_filename);
+        co->co_filename = new_file;
+    }
+    _PyCode_ClearList();
+
     ret = 0;
     goto finally;
 
@@ -829,6 +852,7 @@
 
             Py_FileSystemDefaultEncoding = codeset;
             Py_HasFileSystemDefaultEncoding = 0;
+            _PyCode_ClearList();
             return;
         }
 
@@ -841,6 +865,8 @@
     }
 #endif
 
+    _PyCode_ClearList();
+
     /* the encoding is mbcs, utf-8 or ascii */
     codec = _PyCodec_Lookup(Py_FileSystemDefaultEncoding);
     if (!codec) {

From python-checkins at python.org  Sun Jul 11 12:53:40 2010
From: python-checkins at python.org (victor.stinner)
Date: Sun, 11 Jul 2010 12:53:40 +0200 (CEST)
Subject: [Python-checkins] r82811 -
	python/branches/import_unicode/Python/compile.c
Message-ID: <20100711105340.61046EEAB2@mail.python.org>

Author: victor.stinner
Date: Sun Jul 11 12:53:40 2010
New Revision: 82811

Log:
makenode() uses PyUnicode_DecodeFSDefault()

Modified:
   python/branches/import_unicode/Python/compile.c

Modified: python/branches/import_unicode/Python/compile.c
==============================================================================
--- python/branches/import_unicode/Python/compile.c	(original)
+++ python/branches/import_unicode/Python/compile.c	Sun Jul 11 12:53:40 2010
@@ -3942,7 +3942,7 @@
     freevars = dict_keys_inorder(c->u->u_freevars, PyTuple_Size(cellvars));
     if (!freevars)
         goto error;
-    filename = PyUnicode_FromString(c->c_filename);
+    filename = PyUnicode_DecodeFSDefault(c->c_filename);
     if (!filename)
         goto error;
 

From python-checkins at python.org  Sun Jul 11 13:23:30 2010
From: python-checkins at python.org (victor.stinner)
Date: Sun, 11 Jul 2010 13:23:30 +0200 (CEST)
Subject: [Python-checkins] r82812 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100711112330.E4CD4EEA4F@mail.python.org>

Author: victor.stinner
Date: Sun Jul 11 13:23:30 2010
New Revision: 82812

Log:
fopen_unicode(), stat_unicode(): clear error

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Sun Jul 11 13:23:30 2010
@@ -2116,8 +2116,11 @@
 {
     FILE *f;
     PyObject *bytes = PyUnicode_EncodeFSDefault(unicode);
-    if (bytes == NULL)
+    if (bytes == NULL) {
+        /* FIXME: don't clear error? */
+        PyErr_Clear();
         return NULL;
+    }
     f = fopen(PyBytes_AS_STRING(bytes), mode);
     Py_DECREF(bytes);
     return f;
@@ -2130,8 +2133,11 @@
 {
     int ret;
     PyObject *bytes = PyUnicode_EncodeFSDefault(unicode);
-    if (bytes == NULL)
+    if (bytes == NULL) {
+        /* FIXME: don't clear error? */
+        PyErr_Clear();
         return -1;
+    }
     ret = stat(PyBytes_AS_STRING(bytes), statbuf);
     Py_DECREF(bytes);
     return ret;

From python-checkins at python.org  Sun Jul 11 13:34:07 2010
From: python-checkins at python.org (victor.stinner)
Date: Sun, 11 Jul 2010 13:34:07 +0200 (CEST)
Subject: [Python-checkins] r82813 -
	python/branches/import_unicode/Modules/getpath.c
Message-ID: <20100711113407.0C364EEA77@mail.python.org>

Author: victor.stinner
Date: Sun Jul 11 13:34:06 2010
New Revision: 82813

Log:
_Py_wreadlink() uses _Py_wchar2char()

Remove also trailing spaces

Modified:
   python/branches/import_unicode/Modules/getpath.c

Modified: python/branches/import_unicode/Modules/getpath.c
==============================================================================
--- python/branches/import_unicode/Modules/getpath.c	(original)
+++ python/branches/import_unicode/Modules/getpath.c	Sun Jul 11 13:34:06 2010
@@ -179,12 +179,13 @@
 _Py_wreadlink(const wchar_t *path)
 {
     char cbuf[PATH_MAX];
-    char cpath[PATH_MAX];
+    char *cpath;
     int res;
-    size_t r1 = wcstombs(cpath, path, PATH_MAX);
-    if (r1 == (size_t)-1 || r1 >= PATH_MAX)
+    cpath = _Py_wchar2char(path);
+    if (cpath == NULL)
         return NULL;
     res = (int)readlink(cpath, cbuf, PATH_MAX);
+    PyMem_Free(cpath);
     if (res == -1)
         return NULL;
     if (res == PATH_MAX)
@@ -298,9 +299,9 @@
         wchar_t *cwd;
         cwd = _Py_wgetcwd(path, MAXPATHLEN);
         if (cwd == NULL)
-            return /* FIXME: return an error */; 
+            return /* FIXME: return an error */;
         if (wcslen(cwd) >= MAXPATHLEN)
-            return /* FIXME: return an error */; 
+            return /* FIXME: return an error */;
         wcscpy(path, cwd);
         PyMem_Free(cwd);
         if (p[0] == '.' && p[1] == SEP)

From python-checkins at python.org  Sun Jul 11 14:12:00 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Sun, 11 Jul 2010 14:12:00 +0200 (CEST)
Subject: [Python-checkins] r82814 - in python/branches/py3k: Misc/NEWS
	Objects/memoryobject.c
Message-ID: <20100711121200.314D8EE9D3@mail.python.org>

Author: antoine.pitrou
Date: Sun Jul 11 14:12:00 2010
New Revision: 82814

Log:
Issue #7616: Fix copying of overlapping memoryview slices with the Intel
compiler.



Modified:
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/Objects/memoryobject.c

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sun Jul 11 14:12:00 2010
@@ -12,6 +12,9 @@
 Core and Builtins
 -----------------
 
+- Issue #7616: Fix copying of overlapping memoryview slices with the Intel
+  compiler.
+
 - Issue #8413: structsequence now subclasses tuple.
 
 - Issue #8271: during the decoding of an invalid UTF-8 byte sequence, only the

Modified: python/branches/py3k/Objects/memoryobject.c
==============================================================================
--- python/branches/py3k/Objects/memoryobject.c	(original)
+++ python/branches/py3k/Objects/memoryobject.c	Sun Jul 11 14:12:00 2010
@@ -624,7 +624,7 @@
 static int
 memory_ass_sub(PyMemoryViewObject *self, PyObject *key, PyObject *value)
 {
-    Py_ssize_t start, len, bytelen, i;
+    Py_ssize_t start, len, bytelen;
     Py_buffer srcview;
     Py_buffer *view = &(self->view);
     char *srcbuf, *destbuf;
@@ -694,16 +694,8 @@
     if (destbuf + bytelen < srcbuf || srcbuf + bytelen < destbuf)
         /* No overlapping */
         memcpy(destbuf, srcbuf, bytelen);
-    else if (destbuf < srcbuf) {
-        /* Copy in ascending order */
-        for (i = 0; i < bytelen; i++)
-            destbuf[i] = srcbuf[i];
-    }
-    else {
-        /* Copy in descencing order */
-        for (i = bytelen - 1; i >= 0; i--)
-            destbuf[i] = srcbuf[i];
-    }
+    else
+        memmove(destbuf, srcbuf, bytelen);
 
     PyBuffer_Release(&srcview);
     return 0;

From python-checkins at python.org  Sun Jul 11 14:14:05 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Sun, 11 Jul 2010 14:14:05 +0200 (CEST)
Subject: [Python-checkins] r82815 - in python/branches/release27-maint:
	Misc/NEWS Objects/memoryobject.c
Message-ID: <20100711121405.5CA51EE9D3@mail.python.org>

Author: antoine.pitrou
Date: Sun Jul 11 14:14:05 2010
New Revision: 82815

Log:
Merged revisions 82814 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82814 | antoine.pitrou | 2010-07-11 14:12:00 +0200 (dim., 11 juil. 2010) | 4 lines
  
  Issue #7616: Fix copying of overlapping memoryview slices with the Intel
  compiler.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Misc/NEWS
   python/branches/release27-maint/Objects/memoryobject.c

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Sun Jul 11 14:14:05 2010
@@ -12,6 +12,8 @@
 Core and Builtins
 -----------------
 
+- Issue #7616: Fix copying of overlapping memoryview slices with the Intel
+  compiler.
 
 Library
 -------

Modified: python/branches/release27-maint/Objects/memoryobject.c
==============================================================================
--- python/branches/release27-maint/Objects/memoryobject.c	(original)
+++ python/branches/release27-maint/Objects/memoryobject.c	Sun Jul 11 14:14:05 2010
@@ -632,7 +632,7 @@
 static int
 memory_ass_sub(PyMemoryViewObject *self, PyObject *key, PyObject *value)
 {
-    Py_ssize_t start, len, bytelen, i;
+    Py_ssize_t start, len, bytelen;
     Py_buffer srcview;
     Py_buffer *view = &(self->view);
     char *srcbuf, *destbuf;
@@ -702,16 +702,8 @@
     if (destbuf + bytelen < srcbuf || srcbuf + bytelen < destbuf)
         /* No overlapping */
         memcpy(destbuf, srcbuf, bytelen);
-    else if (destbuf < srcbuf) {
-        /* Copy in ascending order */
-        for (i = 0; i < bytelen; i++)
-            destbuf[i] = srcbuf[i];
-    }
-    else {
-        /* Copy in descencing order */
-        for (i = bytelen - 1; i >= 0; i--)
-            destbuf[i] = srcbuf[i];
-    }
+    else
+        memmove(destbuf, srcbuf, bytelen);
 
     PyBuffer_Release(&srcview);
     return 0;

From python-checkins at python.org  Sun Jul 11 14:15:03 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Sun, 11 Jul 2010 14:15:03 +0200 (CEST)
Subject: [Python-checkins] r82816 - in python/branches/release31-maint:
	Misc/NEWS Objects/memoryobject.c
Message-ID: <20100711121503.E55A2EE9D3@mail.python.org>

Author: antoine.pitrou
Date: Sun Jul 11 14:15:03 2010
New Revision: 82816

Log:
Merged revisions 82814 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82814 | antoine.pitrou | 2010-07-11 14:12:00 +0200 (dim., 11 juil. 2010) | 4 lines
  
  Issue #7616: Fix copying of overlapping memoryview slices with the Intel
  compiler.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Misc/NEWS
   python/branches/release31-maint/Objects/memoryobject.c

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Sun Jul 11 14:15:03 2010
@@ -12,6 +12,9 @@
 Core and Builtins
 -----------------
 
+- Issue #7616: Fix copying of overlapping memoryview slices with the Intel
+  compiler.
+
 - Issue #8271: during the decoding of an invalid UTF-8 byte sequence, only the
   start byte and the continuation byte(s) are now considered invalid, instead
   of the number of bytes specified by the start byte.

Modified: python/branches/release31-maint/Objects/memoryobject.c
==============================================================================
--- python/branches/release31-maint/Objects/memoryobject.c	(original)
+++ python/branches/release31-maint/Objects/memoryobject.c	Sun Jul 11 14:15:03 2010
@@ -624,7 +624,7 @@
 static int
 memory_ass_sub(PyMemoryViewObject *self, PyObject *key, PyObject *value)
 {
-    Py_ssize_t start, len, bytelen, i;
+    Py_ssize_t start, len, bytelen;
     Py_buffer srcview;
     Py_buffer *view = &(self->view);
     char *srcbuf, *destbuf;
@@ -694,16 +694,8 @@
     if (destbuf + bytelen < srcbuf || srcbuf + bytelen < destbuf)
         /* No overlapping */
         memcpy(destbuf, srcbuf, bytelen);
-    else if (destbuf < srcbuf) {
-        /* Copy in ascending order */
-        for (i = 0; i < bytelen; i++)
-            destbuf[i] = srcbuf[i];
-    }
-    else {
-        /* Copy in descencing order */
-        for (i = bytelen - 1; i >= 0; i--)
-            destbuf[i] = srcbuf[i];
-    }
+    else
+        memmove(destbuf, srcbuf, bytelen);
 
     PyBuffer_Release(&srcview);
     return 0;

From python-checkins at python.org  Sun Jul 11 14:59:25 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sun, 11 Jul 2010 14:59:25 +0200 (CEST)
Subject: [Python-checkins] r82817 - python/branches/py3k/Lib/test/test_ast.py
Message-ID: <20100711125925.147D4EEA57@mail.python.org>

Author: benjamin.peterson
Date: Sun Jul 11 14:59:24 2010
New Revision: 82817

Log:
test for set literals

Modified:
   python/branches/py3k/Lib/test/test_ast.py

Modified: python/branches/py3k/Lib/test/test_ast.py
==============================================================================
--- python/branches/py3k/Lib/test/test_ast.py	(original)
+++ python/branches/py3k/Lib/test/test_ast.py	Sun Jul 11 14:59:24 2010
@@ -285,6 +285,7 @@
         self.assertEqual(ast.literal_eval('[1, 2, 3]'), [1, 2, 3])
         self.assertEqual(ast.literal_eval('{"foo": 42}'), {"foo": 42})
         self.assertEqual(ast.literal_eval('(True, False, None)'), (True, False, None))
+        self.assertEqual(ast.literal_eval('{1, 2, 3}'), {1, 2, 3})
         self.assertRaises(ValueError, ast.literal_eval, 'foo()')
 
     def test_literal_eval_issue4907(self):

From python-checkins at python.org  Sun Jul 11 19:38:24 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun, 11 Jul 2010 19:38:24 +0200 (CEST)
Subject: [Python-checkins] r82818 - in python/branches/py3k:
	Doc/library/cmath.rst Doc/library/math.rst
	Lib/test/test_cmath.py Lib/test/test_math.py Misc/NEWS
	Modules/cmathmodule.c Modules/mathmodule.c
Message-ID: <20100711173824.46DB9D9AE@mail.python.org>

Author: mark.dickinson
Date: Sun Jul 11 19:38:24 2010
New Revision: 82818

Log:
Issue #9165: Add math.isfinite and cmath.isfinite.


Modified:
   python/branches/py3k/Doc/library/cmath.rst
   python/branches/py3k/Doc/library/math.rst
   python/branches/py3k/Lib/test/test_cmath.py
   python/branches/py3k/Lib/test/test_math.py
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/Modules/cmathmodule.c
   python/branches/py3k/Modules/mathmodule.c

Modified: python/branches/py3k/Doc/library/cmath.rst
==============================================================================
--- python/branches/py3k/Doc/library/cmath.rst	(original)
+++ python/branches/py3k/Doc/library/cmath.rst	Sun Jul 11 19:38:24 2010
@@ -187,6 +187,12 @@
 Classification functions
 ------------------------
 
+.. function:: isfinite(x)
+
+   Return ``True`` if both the real and imaginary parts of *x* are finite,
+   and ``False`` otherwise.
+
+
 .. function:: isinf(x)
 
    Return *True* if the real or the imaginary part of x is positive

Modified: python/branches/py3k/Doc/library/math.rst
==============================================================================
--- python/branches/py3k/Doc/library/math.rst	(original)
+++ python/branches/py3k/Doc/library/math.rst	Sun Jul 11 19:38:24 2010
@@ -97,6 +97,12 @@
    <http://code.activestate.com/recipes/393090/>`_\.
 
 
+.. function:: isfinite(x)
+
+   Return ``True`` if *x* is neither an infinity nor a NaN, and
+   ``False`` otherwise.  (Note that ``0.0`` *is* considered finite.)
+
+
 .. function:: isinf(x)
 
    Check if the float *x* is positive or negative infinity.

Modified: python/branches/py3k/Lib/test/test_cmath.py
==============================================================================
--- python/branches/py3k/Lib/test/test_cmath.py	(original)
+++ python/branches/py3k/Lib/test/test_cmath.py	Sun Jul 11 19:38:24 2010
@@ -442,6 +442,15 @@
         self.assertCEqual(rect(1, pi/2), (0, 1.))
         self.assertCEqual(rect(1, -pi/2), (0, -1.))
 
+    def test_isfinite(self):
+        real_vals = [float('-inf'), -2.3, -0.0,
+                     0.0, 2.3, float('inf'), float('nan')]
+        for x in real_vals:
+            for y in real_vals:
+                z = complex(x, y)
+                self.assertEquals(cmath.isfinite(z),
+                                  math.isfinite(x) and math.isfinite(y))
+
     def test_isnan(self):
         self.assertFalse(cmath.isnan(1))
         self.assertFalse(cmath.isnan(1j))

Modified: python/branches/py3k/Lib/test/test_math.py
==============================================================================
--- python/branches/py3k/Lib/test/test_math.py	(original)
+++ python/branches/py3k/Lib/test/test_math.py	Sun Jul 11 19:38:24 2010
@@ -915,6 +915,15 @@
         self.assertRaises(TypeError, math.trunc, 1, 2)
         self.assertRaises(TypeError, math.trunc, TestNoTrunc())
 
+    def testIsfinite(self):
+        self.assertTrue(math.isfinite(0.0))
+        self.assertTrue(math.isfinite(-0.0))
+        self.assertTrue(math.isfinite(1.0))
+        self.assertTrue(math.isfinite(-1.0))
+        self.assertFalse(math.isfinite(float("nan")))
+        self.assertFalse(math.isfinite(float("inf")))
+        self.assertFalse(math.isfinite(float("-inf")))
+
     def testIsnan(self):
         self.assertTrue(math.isnan(float("nan")))
         self.assertTrue(math.isnan(float("inf")* 0.))

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sun Jul 11 19:38:24 2010
@@ -1415,6 +1415,9 @@
 Extension Modules
 -----------------
 
+- Issue #9165: Add new functions math.isfinite and cmath.isfinite, to
+  accompany existing isinf and isnan functions.
+
 - Issue #1578269: Implement os.symlink for Windows 6.0+. Patch by
   Jason R. Coombs
 

Modified: python/branches/py3k/Modules/cmathmodule.c
==============================================================================
--- python/branches/py3k/Modules/cmathmodule.c	(original)
+++ python/branches/py3k/Modules/cmathmodule.c	Sun Jul 11 19:38:24 2010
@@ -1024,6 +1024,19 @@
 Convert from polar coordinates to rectangular coordinates.");
 
 static PyObject *
+cmath_isfinite(PyObject *self, PyObject *args)
+{
+    Py_complex z;
+    if (!PyArg_ParseTuple(args, "D:isfinite", &z))
+        return NULL;
+    return PyBool_FromLong(Py_IS_FINITE(z.real) && Py_IS_FINITE(z.imag));
+}
+
+PyDoc_STRVAR(cmath_isfinite_doc,
+"isfinite(z) -> bool\n\
+Return True if both the real and imaginary parts of z are finite, else False.");
+
+static PyObject *
 cmath_isnan(PyObject *self, PyObject *args)
 {
     Py_complex z;
@@ -1065,6 +1078,7 @@
     {"cos",    cmath_cos,   METH_VARARGS, c_cos_doc},
     {"cosh",   cmath_cosh,  METH_VARARGS, c_cosh_doc},
     {"exp",    cmath_exp,   METH_VARARGS, c_exp_doc},
+    {"isfinite", cmath_isfinite, METH_VARARGS, cmath_isfinite_doc},
     {"isinf",  cmath_isinf, METH_VARARGS, cmath_isinf_doc},
     {"isnan",  cmath_isnan, METH_VARARGS, cmath_isnan_doc},
     {"log",    cmath_log,   METH_VARARGS, cmath_log_doc},

Modified: python/branches/py3k/Modules/mathmodule.c
==============================================================================
--- python/branches/py3k/Modules/mathmodule.c	(original)
+++ python/branches/py3k/Modules/mathmodule.c	Sun Jul 11 19:38:24 2010
@@ -1818,6 +1818,19 @@
 Convert angle x from degrees to radians.");
 
 static PyObject *
+math_isfinite(PyObject *self, PyObject *arg)
+{
+    double x = PyFloat_AsDouble(arg);
+    if (x == -1.0 && PyErr_Occurred())
+        return NULL;
+    return PyBool_FromLong((long)Py_IS_FINITE(x));
+}
+
+PyDoc_STRVAR(math_isfinite_doc,
+"isfinite(x) -> bool\n\n\
+Check if float x is finite (not an infinity or NaN).");
+
+static PyObject *
 math_isnan(PyObject *self, PyObject *arg)
 {
     double x = PyFloat_AsDouble(arg);
@@ -1868,6 +1881,7 @@
     {"fsum",            math_fsum,      METH_O,         math_fsum_doc},
     {"gamma",           math_gamma,     METH_O,         math_gamma_doc},
     {"hypot",           math_hypot,     METH_VARARGS,   math_hypot_doc},
+    {"isfinite",        math_isfinite,  METH_O,         math_isfinite_doc},
     {"isinf",           math_isinf,     METH_O,         math_isinf_doc},
     {"isnan",           math_isnan,     METH_O,         math_isnan_doc},
     {"ldexp",           math_ldexp,     METH_VARARGS,   math_ldexp_doc},

From python-checkins at python.org  Sun Jul 11 19:39:46 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Sun, 11 Jul 2010 19:39:46 +0200 (CEST)
Subject: [Python-checkins] r82819 - in python/branches/py3k: Misc/NEWS
	Tools/i18n/msgfmt.py
Message-ID: <20100711173946.71045EEB1A@mail.python.org>

Author: martin.v.loewis
Date: Sun Jul 11 19:39:46 2010
New Revision: 82819

Log:
Issue #8974: fix print calls in msgfmt.py.


Modified:
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/Tools/i18n/msgfmt.py

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sun Jul 11 19:39:46 2010
@@ -1773,7 +1773,7 @@
 Tools/Demos
 -----------
 
-- Issue #5464: Implement plural forms in msgfmt.py.
+- Issue #5464, #8974: Implement plural forms in msgfmt.py.
 
 - iobench (a file I/O benchmark) and ccbench (a concurrency benchmark) were
   added to the `Tools/` directory.  They were previously living in the

Modified: python/branches/py3k/Tools/i18n/msgfmt.py
==============================================================================
--- python/branches/py3k/Tools/i18n/msgfmt.py	(original)
+++ python/branches/py3k/Tools/i18n/msgfmt.py	Sun Jul 11 19:39:46 2010
@@ -165,7 +165,7 @@
             section = STR
             if l.startswith('msgstr['):
                 if not is_plural:
-                    print(sys.stderr, 'plural without msgid_plural on %s:%d' % (infile, lno),
+                    print('plural without msgid_plural on %s:%d' % (infile, lno),
                           file=sys.stderr)
                     sys.exit(1)
                 l = l.split(']', 1)[1]
@@ -173,7 +173,7 @@
                     msgstr += b'\0' # Separator of the various plural forms
             else:
                 if is_plural:
-                    print(sys.stderr, 'indexed msgstr required for plural on  %s:%d' % (infile, lno),
+                    print('indexed msgstr required for plural on  %s:%d' % (infile, lno),
                           file=sys.stderr)
                     sys.exit(1)
                 l = l[6:]

From python-checkins at python.org  Sun Jul 11 20:13:41 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun, 11 Jul 2010 20:13:41 +0200 (CEST)
Subject: [Python-checkins] r82820 - python/branches/py3k/Modules/mathmodule.c
Message-ID: <20100711181341.3A7C0EEADB@mail.python.org>

Author: mark.dickinson
Date: Sun Jul 11 20:13:41 2010
New Revision: 82820

Log:
Improve docstrings for isnan, isinf and isfinite.

Modified:
   python/branches/py3k/Modules/mathmodule.c

Modified: python/branches/py3k/Modules/mathmodule.c
==============================================================================
--- python/branches/py3k/Modules/mathmodule.c	(original)
+++ python/branches/py3k/Modules/mathmodule.c	Sun Jul 11 20:13:41 2010
@@ -1828,7 +1828,7 @@
 
 PyDoc_STRVAR(math_isfinite_doc,
 "isfinite(x) -> bool\n\n\
-Check if float x is finite (not an infinity or NaN).");
+Return True if x is neither an infinity nor a NaN, and False otherwise.");
 
 static PyObject *
 math_isnan(PyObject *self, PyObject *arg)
@@ -1841,7 +1841,7 @@
 
 PyDoc_STRVAR(math_isnan_doc,
 "isnan(x) -> bool\n\n\
-Check if float x is not a number (NaN).");
+Return True if x is a NaN (not a number), and False otherwise.");
 
 static PyObject *
 math_isinf(PyObject *self, PyObject *arg)
@@ -1854,7 +1854,7 @@
 
 PyDoc_STRVAR(math_isinf_doc,
 "isinf(x) -> bool\n\n\
-Check if float x is infinite (positive or negative).");
+Return True if x is a positive or negative infinity, and False otherwise.");
 
 static PyMethodDef math_methods[] = {
     {"acos",            math_acos,      METH_O,         math_acos_doc},

From python-checkins at python.org  Sun Jul 11 20:53:06 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun, 11 Jul 2010 20:53:06 +0200 (CEST)
Subject: [Python-checkins] r82821 - in python/branches/py3k: Lib/_abcoll.py
	Lib/test/test_collections.py Misc/NEWS
Message-ID: <20100711185306.99CFFCA14@mail.python.org>

Author: mark.dickinson
Date: Sun Jul 11 20:53:06 2010
New Revision: 82821

Log:
Issue #9137: Fix issue in MutableMapping.update, which incorrectly
treated keyword arguments called 'self' or 'other' specially.


Modified:
   python/branches/py3k/Lib/_abcoll.py
   python/branches/py3k/Lib/test/test_collections.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/_abcoll.py
==============================================================================
--- python/branches/py3k/Lib/_abcoll.py	(original)
+++ python/branches/py3k/Lib/_abcoll.py	Sun Jul 11 20:53:06 2010
@@ -480,7 +480,15 @@
         except KeyError:
             pass
 
-    def update(self, other=(), **kwds):
+    def update(*args, **kwds):
+        if len(args) > 2:
+            raise TypeError("update() takes at most 2 positional "
+                            "arguments ({} given)".format(len(args)))
+        elif not args:
+            raise TypeError("update() takes at least 1 argument (0 given)")
+        self = args[0]
+        other = args[1] if len(args) >= 2 else ()
+
         if isinstance(other, Mapping):
             for key in other:
                 self[key] = other[key]

Modified: python/branches/py3k/Lib/test/test_collections.py
==============================================================================
--- python/branches/py3k/Lib/test/test_collections.py	(original)
+++ python/branches/py3k/Lib/test/test_collections.py	Sun Jul 11 20:53:06 2010
@@ -758,6 +758,19 @@
         od.update([('a', 1), ('b', 2), ('c', 9), ('d', 4)], c=3, e=5)
         self.assertEqual(list(od.items()), pairs)                                   # mixed input
 
+        # Issue 9137: Named argument called 'other' or 'self'
+        # shouldn't be treated specially.
+        od = OrderedDict()
+        od.update(self=23)
+        self.assertEqual(list(od.items()), [('self', 23)])
+        od = OrderedDict()
+        od.update(other={})
+        self.assertEqual(list(od.items()), [('other', {})])
+        od = OrderedDict()
+        od.update(red=5, blue=6, other=7, self=8)
+        self.assertEqual(sorted(list(od.items())),
+                         [('blue', 6), ('other', 7), ('red', 5), ('self', 8)])
+
         # Make sure that direct calls to update do not clear previous contents
         # add that updates items are not moved to the end
         d = OrderedDict([('a', 1), ('b', 2), ('c', 3), ('d', 44), ('e', 55)])

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sun Jul 11 20:53:06 2010
@@ -473,6 +473,9 @@
 Library
 -------
 
+- Issue #9137: Fix issue in MutableMapping.update, which incorrectly
+  treated keyword arguments called 'self' or 'other' specially.
+
 - ``ast.literal_eval()`` now allows set literals.
 
 - Issue #9164: Ensure that sysconfig handles duplicate -arch flags in CFLAGS.

From python-checkins at python.org  Sun Jul 11 21:12:10 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun, 11 Jul 2010 21:12:10 +0200 (CEST)
Subject: [Python-checkins] r82822 -
	python/branches/py3k/Lib/test/test_cmath.py
Message-ID: <20100711191210.955F4EEB42@mail.python.org>

Author: mark.dickinson
Date: Sun Jul 11 21:12:10 2010
New Revision: 82822

Log:
assertEquals -> assertEqual

Modified:
   python/branches/py3k/Lib/test/test_cmath.py

Modified: python/branches/py3k/Lib/test/test_cmath.py
==============================================================================
--- python/branches/py3k/Lib/test/test_cmath.py	(original)
+++ python/branches/py3k/Lib/test/test_cmath.py	Sun Jul 11 21:12:10 2010
@@ -448,7 +448,7 @@
         for x in real_vals:
             for y in real_vals:
                 z = complex(x, y)
-                self.assertEquals(cmath.isfinite(z),
+                self.assertEqual(cmath.isfinite(z),
                                   math.isfinite(x) and math.isfinite(y))
 
     def test_isnan(self):

From python-checkins at python.org  Sun Jul 11 21:17:28 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun, 11 Jul 2010 21:17:28 +0200 (CEST)
Subject: [Python-checkins] r82823 - in python/branches/release27-maint:
	Lib/_abcoll.py Lib/test/test_collections.py Misc/NEWS
Message-ID: <20100711191728.EB5C6EEB03@mail.python.org>

Author: mark.dickinson
Date: Sun Jul 11 21:17:28 2010
New Revision: 82823

Log:
Merged revisions 82821 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82821 | mark.dickinson | 2010-07-11 19:53:06 +0100 (Sun, 11 Jul 2010) | 3 lines
  
  Issue #9137: Fix issue in MutableMapping.update, which incorrectly
  treated keyword arguments called 'self' or 'other' specially.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/_abcoll.py
   python/branches/release27-maint/Lib/test/test_collections.py
   python/branches/release27-maint/Misc/NEWS

Modified: python/branches/release27-maint/Lib/_abcoll.py
==============================================================================
--- python/branches/release27-maint/Lib/_abcoll.py	(original)
+++ python/branches/release27-maint/Lib/_abcoll.py	Sun Jul 11 21:17:28 2010
@@ -466,7 +466,15 @@
         except KeyError:
             pass
 
-    def update(self, other=(), **kwds):
+    def update(*args, **kwds):
+        if len(args) > 2:
+            raise TypeError("update() takes at most 2 positional "
+                            "arguments ({} given)".format(len(args)))
+        elif not args:
+            raise TypeError("update() takes at least 1 argument (0 given)")
+        self = args[0]
+        other = args[1] if len(args) >= 2 else ()
+
         if isinstance(other, Mapping):
             for key in other:
                 self[key] = other[key]

Modified: python/branches/release27-maint/Lib/test/test_collections.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_collections.py	(original)
+++ python/branches/release27-maint/Lib/test/test_collections.py	Sun Jul 11 21:17:28 2010
@@ -766,6 +766,19 @@
         od.update([('a', 1), ('b', 2), ('c', 9), ('d', 4)], c=3, e=5)
         self.assertEqual(list(od.items()), pairs)                                   # mixed input
 
+        # Issue 9137: Named argument called 'other' or 'self'
+        # shouldn't be treated specially.
+        od = OrderedDict()
+        od.update(self=23)
+        self.assertEqual(list(od.items()), [('self', 23)])
+        od = OrderedDict()
+        od.update(other={})
+        self.assertEqual(list(od.items()), [('other', {})])
+        od = OrderedDict()
+        od.update(red=5, blue=6, other=7, self=8)
+        self.assertEqual(sorted(list(od.items())),
+                         [('blue', 6), ('other', 7), ('red', 5), ('self', 8)])
+
         # Make sure that direct calls to update do not clear previous contents
         # add that updates items are not moved to the end
         d = OrderedDict([('a', 1), ('b', 2), ('c', 3), ('d', 44), ('e', 55)])

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Sun Jul 11 21:17:28 2010
@@ -1005,6 +1005,9 @@
 Library
 -------
 
+- Issue #9137: Fix issue in MutableMapping.update, which incorrectly
+  treated keyword arguments called 'self' or 'other' specially.
+
 - Issue #7835: shelve should no longer produce mysterious warnings during
   interpreter shutdown.
 

From python-checkins at python.org  Sun Jul 11 21:27:07 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun, 11 Jul 2010 21:27:07 +0200 (CEST)
Subject: [Python-checkins] r82824 - in python/branches/release31-maint:
	Lib/_abcoll.py Lib/test/test_collections.py Misc/NEWS
Message-ID: <20100711192707.19674EEB0C@mail.python.org>

Author: mark.dickinson
Date: Sun Jul 11 21:27:06 2010
New Revision: 82824

Log:
Merged revisions 82821 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82821 | mark.dickinson | 2010-07-11 19:53:06 +0100 (Sun, 11 Jul 2010) | 3 lines
  
  Issue #9137: Fix issue in MutableMapping.update, which incorrectly
  treated keyword arguments called 'self' or 'other' specially.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/_abcoll.py
   python/branches/release31-maint/Lib/test/test_collections.py
   python/branches/release31-maint/Misc/NEWS

Modified: python/branches/release31-maint/Lib/_abcoll.py
==============================================================================
--- python/branches/release31-maint/Lib/_abcoll.py	(original)
+++ python/branches/release31-maint/Lib/_abcoll.py	Sun Jul 11 21:27:06 2010
@@ -480,7 +480,15 @@
         except KeyError:
             pass
 
-    def update(self, other=(), **kwds):
+    def update(*args, **kwds):
+        if len(args) > 2:
+            raise TypeError("update() takes at most 2 positional "
+                            "arguments ({} given)".format(len(args)))
+        elif not args:
+            raise TypeError("update() takes at least 1 argument (0 given)")
+        self = args[0]
+        other = args[1] if len(args) >= 2 else ()
+
         if isinstance(other, Mapping):
             for key in other:
                 self[key] = other[key]

Modified: python/branches/release31-maint/Lib/test/test_collections.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_collections.py	(original)
+++ python/branches/release31-maint/Lib/test/test_collections.py	Sun Jul 11 21:27:06 2010
@@ -725,6 +725,19 @@
         od.update([('a', 1), ('b', 2), ('c', 9), ('d', 4)], c=3, e=5)
         self.assertEqual(list(od.items()), pairs)                                   # mixed input
 
+        # Issue 9137: Named argument called 'other' or 'self'
+        # shouldn't be treated specially.
+        od = OrderedDict()
+        od.update(self=23)
+        self.assertEqual(list(od.items()), [('self', 23)])
+        od = OrderedDict()
+        od.update(other={})
+        self.assertEqual(list(od.items()), [('other', {})])
+        od = OrderedDict()
+        od.update(red=5, blue=6, other=7, self=8)
+        self.assertEqual(sorted(list(od.items())),
+                         [('blue', 6), ('other', 7), ('red', 5), ('self', 8)])
+
         # Make sure that direct calls to update do not clear previous contents
         # add that updates items are not moved to the end
         d = OrderedDict([('a', 1), ('b', 2), ('c', 3), ('d', 44), ('e', 55)])

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Sun Jul 11 21:27:06 2010
@@ -78,6 +78,9 @@
 Library
 -------
 
+- Issue #9137: Fix issue in MutableMapping.update, which incorrectly
+  treated keyword arguments called 'self' or 'other' specially.
+
 - Issue #7646: The fnmatch pattern cache no longer grows without bound.
 
 - Issue #9136: Fix 'dictionary changed size during iteration'

From python-checkins at python.org  Sun Jul 11 21:47:37 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun, 11 Jul 2010 21:47:37 +0200 (CEST)
Subject: [Python-checkins] r82825 - in python/branches/py3k/Doc/library:
	cmath.rst math.rst
Message-ID: <20100711194737.C79FBEEB0C@mail.python.org>

Author: mark.dickinson
Date: Sun Jul 11 21:47:37 2010
New Revision: 82825

Log:
Include versionadded in (c)math.isfinite docs (thanks Ezio Melotti).  Other minor doc cleanups.

Modified:
   python/branches/py3k/Doc/library/cmath.rst
   python/branches/py3k/Doc/library/math.rst

Modified: python/branches/py3k/Doc/library/cmath.rst
==============================================================================
--- python/branches/py3k/Doc/library/cmath.rst	(original)
+++ python/branches/py3k/Doc/library/cmath.rst	Sun Jul 11 21:47:37 2010
@@ -189,19 +189,22 @@
 
 .. function:: isfinite(x)
 
-   Return ``True`` if both the real and imaginary parts of *x* are finite,
-   and ``False`` otherwise.
+   Return ``True`` if both the real and imaginary parts of *x* are finite, and
+   ``False`` otherwise.
+
+   .. versionadded:: 3.2
 
 
 .. function:: isinf(x)
 
-   Return *True* if the real or the imaginary part of x is positive
-   or negative infinity.
+   Return ``True`` if either the real or the imaginary part of *x* is an
+   infinity, and ``False`` otherwise.
 
 
 .. function:: isnan(x)
 
-   Return *True* if the real or imaginary part of x is not a number (NaN).
+   Return ``True`` if either the real or the imaginary part of *x* is a NaN,
+   and ``False`` otherwise.
 
 
 Constants

Modified: python/branches/py3k/Doc/library/math.rst
==============================================================================
--- python/branches/py3k/Doc/library/math.rst	(original)
+++ python/branches/py3k/Doc/library/math.rst	Sun Jul 11 21:47:37 2010
@@ -102,16 +102,18 @@
    Return ``True`` if *x* is neither an infinity nor a NaN, and
    ``False`` otherwise.  (Note that ``0.0`` *is* considered finite.)
 
+   .. versionadded:: 3.2
+
 
 .. function:: isinf(x)
 
-   Check if the float *x* is positive or negative infinity.
+   Return ``True`` if *x* is a positive or negative infinity, and
+   ``False`` otherwise.
 
 
 .. function:: isnan(x)
 
-   Check if the float *x* is a NaN (not a number).  For more information
-   on NaNs, see the IEEE 754 standards.
+   Return ``True`` if *x* is a NaN (not a number), and ``False`` otherwise.
 
 
 .. function:: ldexp(x, i)

From python-checkins at python.org  Sun Jul 11 22:20:42 2010
From: python-checkins at python.org (victor.stinner)
Date: Sun, 11 Jul 2010 22:20:42 +0200 (CEST)
Subject: [Python-checkins] r82826 -
	python/branches/import_unicode/Lib/test/test_sax.py
Message-ID: <20100711202042.0D0B5EE9F2@mail.python.org>

Author: victor.stinner
Date: Sun Jul 11 22:20:41 2010
New Revision: 82826

Log:
Skip test_sax if filename is not encodable to utf-8

Modified:
   python/branches/import_unicode/Lib/test/test_sax.py

Modified: python/branches/import_unicode/Lib/test/test_sax.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_sax.py	(original)
+++ python/branches/import_unicode/Lib/test/test_sax.py	Sun Jul 11 22:20:41 2010
@@ -18,6 +18,11 @@
 
 TEST_XMLFILE = findfile("test.xml", subdir="xmltestdata")
 TEST_XMLFILE_OUT = findfile("test.xml.out", subdir="xmltestdata")
+try:
+    TEST_XMLFILE.encode("utf8")
+    TEST_XMLFILE_OUT.encode("utf8")
+except UnicodeEncodeError:
+    raise unittest.SkipTest("filename is not encodable to utf8")
 
 ns_uri = "http://www.python.org/xml-ns/saxtest/"
 

From python-checkins at python.org  Sun Jul 11 22:20:43 2010
From: python-checkins at python.org (victor.stinner)
Date: Sun, 11 Jul 2010 22:20:43 +0200 (CEST)
Subject: [Python-checkins] r82827 -
	python/branches/import_unicode/Modules/getpath.c
Message-ID: <20100711202043.A1F91EEB56@mail.python.org>

Author: victor.stinner
Date: Sun Jul 11 22:20:43 2010
New Revision: 82827

Log:
getpath.c: reindent and add FIXME

Modified:
   python/branches/import_unicode/Modules/getpath.c

Modified: python/branches/import_unicode/Modules/getpath.c
==============================================================================
--- python/branches/import_unicode/Modules/getpath.c	(original)
+++ python/branches/import_unicode/Modules/getpath.c	Sun Jul 11 22:20:43 2010
@@ -466,25 +466,26 @@
 #else
     unsigned long nsexeclength = MAXPATHLEN;
 #endif
-        char execpath[MAXPATHLEN+1];
+    char execpath[MAXPATHLEN+1];
 #endif
 
     if (_path) {
-            size_t r = mbstowcs(wpath, _path, MAXPATHLEN+1);
-            path = wpath;
-            if (r == (size_t)-1 || r > MAXPATHLEN) {
-                    /* Could not convert PATH, or it's too long. */
-                    path = NULL;
-            }
+        /* FIXME: use _Py_char2wchar() */
+        size_t r = mbstowcs(wpath, _path, MAXPATHLEN+1);
+        path = wpath;
+        if (r == (size_t)-1 || r > MAXPATHLEN) {
+                /* Could not convert PATH, or it's too long. */
+                path = NULL;
+        }
     }
 
-        /* If there is no slash in the argv0 path, then we have to
-         * assume python is on the user's $PATH, since there's no
-         * other way to find a directory to start the search from.  If
-         * $PATH isn't exported, you lose.
-         */
-        if (wcschr(prog, SEP))
-                wcsncpy(progpath, prog, MAXPATHLEN);
+    /* If there is no slash in the argv0 path, then we have to
+     * assume python is on the user's $PATH, since there's no
+     * other way to find a directory to start the search from.  If
+     * $PATH isn't exported, you lose.
+     */
+    if (wcschr(prog, SEP))
+        wcsncpy(progpath, prog, MAXPATHLEN);
 #ifdef __APPLE__
      /* On Mac OS X, if a script uses an interpreter of the form
       * "#!/opt/python2.3/bin/python", the kernel only passes "python"
@@ -496,52 +497,53 @@
       * will fail if a relative path was used. but in that case,
       * absolutize() should help us out below
       */
-        else if(0 == _NSGetExecutablePath(execpath, &nsexeclength) && execpath[0] == SEP) {
-                size_t r = mbstowcs(progpath, execpath, MAXPATHLEN+1);
-                if (r == (size_t)-1 || r > MAXPATHLEN) {
-                        /* Could not convert execpath, or it's too long. */
-                        progpath[0] = '\0';
-                }
+    else if(0 == _NSGetExecutablePath(execpath, &nsexeclength) && execpath[0] == SEP) {
+        /* FIXME: use _Py_char2wchar() */
+        size_t r = mbstowcs(progpath, execpath, MAXPATHLEN+1);
+        if (r == (size_t)-1 || r > MAXPATHLEN) {
+            /* Could not convert execpath, or it's too long. */
+            progpath[0] = '\0';
         }
+    }
 #endif /* __APPLE__ */
-        else if (path) {
-                while (1) {
-                        wchar_t *delim = wcschr(path, DELIM);
-
-                        if (delim) {
-                                size_t len = delim - path;
-                                if (len > MAXPATHLEN)
-                                        len = MAXPATHLEN;
-                                wcsncpy(progpath, path, len);
-                                *(progpath + len) = '\0';
-                        }
-                        else
-                                wcsncpy(progpath, path, MAXPATHLEN);
-
-                        joinpath(progpath, prog);
-                        if (isxfile(progpath))
-                                break;
-
-                        if (!delim) {
-                                progpath[0] = L'\0';
-                                break;
-                        }
-                        path = delim + 1;
-                }
+    else if (path) {
+        while (1) {
+            wchar_t *delim = wcschr(path, DELIM);
+
+            if (delim) {
+                size_t len = delim - path;
+                if (len > MAXPATHLEN)
+                    len = MAXPATHLEN;
+                wcsncpy(progpath, path, len);
+                *(progpath + len) = '\0';
+            }
+            else
+                wcsncpy(progpath, path, MAXPATHLEN);
+
+            joinpath(progpath, prog);
+            if (isxfile(progpath))
+                break;
+
+            if (!delim) {
+                progpath[0] = L'\0';
+                break;
+            }
+            path = delim + 1;
         }
-        else
-                progpath[0] = '\0';
-        if (progpath[0] != SEP && progpath[0] != '\0')
-                absolutize(progpath);
-        wcsncpy(argv0_path, progpath, MAXPATHLEN);
-        argv0_path[MAXPATHLEN] = '\0';
+    }
+    else
+        progpath[0] = '\0';
+    if (progpath[0] != SEP && progpath[0] != '\0')
+        absolutize(progpath);
+    wcsncpy(argv0_path, progpath, MAXPATHLEN);
+    argv0_path[MAXPATHLEN] = '\0';
 
 #ifdef WITH_NEXT_FRAMEWORK
-        /* On Mac OS X we have a special case if we're running from a framework.
-        ** This is because the python home should be set relative to the library,
-        ** which is in the framework, not relative to the executable, which may
-        ** be outside of the framework. Except when we're in the build directory...
-        */
+    /* On Mac OS X we have a special case if we're running from a framework.
+     ** This is because the python home should be set relative to the library,
+     ** which is in the framework, not relative to the executable, which may
+     ** be outside of the framework. Except when we're in the build directory...
+     */
     pythonModule = NSModuleForSymbol(NSLookupAndBindSymbol("_Py_Initialize"));
     /* Use dylib functions to find out where the framework was loaded from */
     buf = (wchar_t *)NSLibraryNameForModule(pythonModule);
@@ -636,6 +638,7 @@
     bufsz = 0;
 
     if (_rtpypath) {
+        /* FIXME: use _Py_char2wchar() */
         size_t s = mbstowcs(rtpypath, _rtpypath, sizeof(rtpypath)/sizeof(wchar_t));
         if (s == (size_t)-1 || s >=sizeof(rtpypath))
             /* XXX deal with errors more gracefully */

From python-checkins at python.org  Mon Jul 12 01:06:06 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Mon, 12 Jul 2010 01:06:06 +0200 (CEST)
Subject: [Python-checkins] r82828 - in python/branches/py3k: Lib/ast.py
	Lib/test/test_ast.py Misc/NEWS
Message-ID: <20100711230606.EFC29EEB32@mail.python.org>

Author: benjamin.peterson
Date: Mon Jul 12 01:06:06 2010
New Revision: 82828

Log:
allow byte literals

Modified:
   python/branches/py3k/Lib/ast.py
   python/branches/py3k/Lib/test/test_ast.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/ast.py
==============================================================================
--- python/branches/py3k/Lib/ast.py	(original)
+++ python/branches/py3k/Lib/ast.py	Mon Jul 12 01:06:06 2010
@@ -50,7 +50,7 @@
     if isinstance(node_or_string, Expression):
         node_or_string = node_or_string.body
     def _convert(node):
-        if isinstance(node, Str):
+        if isinstance(node, (Str, Bytes)):
             return node.s
         elif isinstance(node, Num):
             return node.n

Modified: python/branches/py3k/Lib/test/test_ast.py
==============================================================================
--- python/branches/py3k/Lib/test/test_ast.py	(original)
+++ python/branches/py3k/Lib/test/test_ast.py	Mon Jul 12 01:06:06 2010
@@ -286,6 +286,7 @@
         self.assertEqual(ast.literal_eval('{"foo": 42}'), {"foo": 42})
         self.assertEqual(ast.literal_eval('(True, False, None)'), (True, False, None))
         self.assertEqual(ast.literal_eval('{1, 2, 3}'), {1, 2, 3})
+        self.assertEqual(ast.literal_eval('b"hi"'), b"hi")
         self.assertRaises(ValueError, ast.literal_eval, 'foo()')
 
     def test_literal_eval_issue4907(self):

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Mon Jul 12 01:06:06 2010
@@ -473,6 +473,8 @@
 Library
 -------
 
+- ``ast.literal_eval()`` now allows byte literals.
+
 - Issue #9137: Fix issue in MutableMapping.update, which incorrectly
   treated keyword arguments called 'self' or 'other' specially.
 

From python-checkins at python.org  Mon Jul 12 01:10:08 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Mon, 12 Jul 2010 01:10:08 +0200 (CEST)
Subject: [Python-checkins] r82829 - in python/branches/release31-maint:
	Lib/ast.py Lib/test/test_ast.py Misc/NEWS
Message-ID: <20100711231008.9B6FFEE99C@mail.python.org>

Author: benjamin.peterson
Date: Mon Jul 12 01:10:08 2010
New Revision: 82829

Log:
Merged revisions 82828 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82828 | benjamin.peterson | 2010-07-11 18:06:06 -0500 (Sun, 11 Jul 2010) | 1 line
  
  allow byte literals
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/ast.py
   python/branches/release31-maint/Lib/test/test_ast.py
   python/branches/release31-maint/Misc/NEWS

Modified: python/branches/release31-maint/Lib/ast.py
==============================================================================
--- python/branches/release31-maint/Lib/ast.py	(original)
+++ python/branches/release31-maint/Lib/ast.py	Mon Jul 12 01:10:08 2010
@@ -50,7 +50,7 @@
     if isinstance(node_or_string, Expression):
         node_or_string = node_or_string.body
     def _convert(node):
-        if isinstance(node, Str):
+        if isinstance(node, (Str, Bytes)):
             return node.s
         elif isinstance(node, Num):
             return node.n

Modified: python/branches/release31-maint/Lib/test/test_ast.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_ast.py	(original)
+++ python/branches/release31-maint/Lib/test/test_ast.py	Mon Jul 12 01:10:08 2010
@@ -271,6 +271,7 @@
         self.assertEqual(ast.literal_eval('[1, 2, 3]'), [1, 2, 3])
         self.assertEqual(ast.literal_eval('{"foo": 42}'), {"foo": 42})
         self.assertEqual(ast.literal_eval('(True, False, None)'), (True, False, None))
+        self.assertEqual(ast.literal_eval('b"hi"'), b"hi")
         self.assertRaises(ValueError, ast.literal_eval, 'foo()')
 
     def test_literal_eval_issue4907(self):

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Mon Jul 12 01:10:08 2010
@@ -78,6 +78,8 @@
 Library
 -------
 
+- ``ast.literal_eval()`` now allows byte literals.
+
 - Issue #9137: Fix issue in MutableMapping.update, which incorrectly
   treated keyword arguments called 'self' or 'other' specially.
 

From solipsis at pitrou.net  Mon Jul 12 01:24:07 2010
From: solipsis at pitrou.net (solipsis at pitrou.net)
Date: Mon, 12 Jul 2010 01:24:07 +0200 (CEST)
Subject: [Python-checkins] Daily py3k reference leaks (r82825): sum=0
Message-ID: <20100711232407.5F16E17721@ns6635.ovh.net>

py3k results for svn r82825 (hg cset 2ee063eef922)
--------------------------------------------------



Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/antoine/py3k/refleaks/reflogOvtf4w', '-x']

From python-checkins at python.org  Mon Jul 12 01:32:11 2010
From: python-checkins at python.org (brian.curtin)
Date: Mon, 12 Jul 2010 01:32:11 +0200 (CEST)
Subject: [Python-checkins] r82830 - python/branches/py3k/Misc/maintainers.rst
Message-ID: <20100711233211.64D5AEE98F@mail.python.org>

Author: brian.curtin
Date: Mon Jul 12 01:32:11 2010
New Revision: 82830

Log:
Put my name down for winreg.


Modified:
   python/branches/py3k/Misc/maintainers.rst

Modified: python/branches/py3k/Misc/maintainers.rst
==============================================================================
--- python/branches/py3k/Misc/maintainers.rst	(original)
+++ python/branches/py3k/Misc/maintainers.rst	Mon Jul 12 01:32:11 2010
@@ -226,7 +226,7 @@
 wave
 weakref             fdrake
 webbrowser          georg.brandl
-winreg
+winreg              brian.curtin
 winsound            effbot (inactive)
 wsgiref             pje
 xdrlib

From python-checkins at python.org  Mon Jul 12 08:03:18 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Mon, 12 Jul 2010 08:03:18 +0200 (CEST)
Subject: [Python-checkins] r82831 - python/branches/py3k/Misc/maintainers.rst
Message-ID: <20100712060318.46EA8EEB09@mail.python.org>

Author: martin.v.loewis
Date: Mon Jul 12 08:03:18 2010
New Revision: 82831

Log:
Remove myself.


Modified:
   python/branches/py3k/Misc/maintainers.rst

Modified: python/branches/py3k/Misc/maintainers.rst
==============================================================================
--- python/branches/py3k/Misc/maintainers.rst	(original)
+++ python/branches/py3k/Misc/maintainers.rst	Mon Jul 12 08:03:18 2010
@@ -89,7 +89,7 @@
 doctest             tim_one (inactive)
 dummy_threading     brett.cannon
 email               barry, r.david.murray
-encodings           lemburg, loewis
+encodings           lemburg
 errno
 exceptions
 fcntl
@@ -104,7 +104,7 @@
 gc                  pitrou
 getopt
 getpass
-gettext             loewis
+gettext
 glob
 grp
 gzip
@@ -125,7 +125,7 @@
 keyword
 lib2to3             benjamin.peterson
 linecache
-locale              loewis, lemburg
+locale              lemburg
 logging             vsajip
 macpath
 mailbox             andrew.kuchling
@@ -135,7 +135,7 @@
 mimetypes
 mmap
 modulefinder        theller, jvr
-msilib              loewis
+msilib
 msvcrt
 multiprocessing     jnoller
 netrc
@@ -144,7 +144,7 @@
 numbers
 operator
 optparse            aronacher
-os                  loewis
+os
 ossaudiodev
 parser
 pdb
@@ -217,7 +217,7 @@
 tty
 turtle              gregorlingl
 types
-unicodedata         loewis, lemburg, ezio.melotti
+unicodedata         lemburg, ezio.melotti
 unittest            michael.foord
 urllib              orsenthil
 uu
@@ -230,9 +230,9 @@
 winsound            effbot (inactive)
 wsgiref             pje
 xdrlib
-xml                 loewis
+xml
 xml.etree           effbot (inactive)
-xmlrpc              loewis
+xmlrpc
 zipfile
 zipimport
 zlib
@@ -277,13 +277,13 @@
 i18n                lemburg
 import machinery    brett.cannon, ncoghlan
 io                  pitrou, benjamin.peterson
-locale              lemburg, loewis
+locale              lemburg
 mathematics         mark.dickinson, eric.smith, lemburg
 memory management   tim_one, lemburg
 networking          giampaolo.rodola
 packaging           tarek, lemburg
 py3 transition      benjamin.peterson
-release management  tarek, lemburg, benjamin.peterson, barry, loewis,
+release management  tarek, lemburg, benjamin.peterson, barry,
                     gvanrossum, anthonybaxter
 str.format          eric.smith
 time and dates      lemburg

From python-checkins at python.org  Mon Jul 12 11:00:30 2010
From: python-checkins at python.org (georg.brandl)
Date: Mon, 12 Jul 2010 11:00:30 +0200 (CEST)
Subject: [Python-checkins] r82832 -
	python/branches/py3k/Doc/faq/extending.rst
Message-ID: <20100712090030.01ABDEE9A9@mail.python.org>

Author: georg.brandl
Date: Mon Jul 12 11:00:29 2010
New Revision: 82832

Log:
Take care of duplicate target name warnings.

Modified:
   python/branches/py3k/Doc/faq/extending.rst

Modified: python/branches/py3k/Doc/faq/extending.rst
==============================================================================
--- python/branches/py3k/Doc/faq/extending.rst	(original)
+++ python/branches/py3k/Doc/faq/extending.rst	Mon Jul 12 11:00:29 2010
@@ -29,6 +29,8 @@
 C++ objects with constructors are probably not a good idea.
 
 
+.. _c-wrapper-software:
+
 Writing C is hard; are there any alternatives?
 ----------------------------------------------
 
@@ -200,11 +202,7 @@
 whole lot of difference between C and C++ -- so the strategy of building a new
 Python type around a C structure (pointer) type will also work for C++ objects.
 
-For C++ libraries, you can look at `SIP
-<http://www.riverbankcomputing.co.uk/software/sip/>`_, `CXX
-<http://cxx.sourceforge.net/>`_, `Boost
-<http://www.boost.org/libs/python/doc/index.html>`_, `Weave
-<http://www.scipy.org/Weave>`_ or `SWIG <http://www.swig.org>`_
+For C++ libraries, see :ref:`c-wrapper-software`.
 
 
 I added a module using the Setup file and the make fails; why?

From python-checkins at python.org  Mon Jul 12 11:01:14 2010
From: python-checkins at python.org (georg.brandl)
Date: Mon, 12 Jul 2010 11:01:14 +0200 (CEST)
Subject: [Python-checkins] r82833 -
	python/branches/release31-maint/Doc/glossary.rst
Message-ID: <20100712090114.41834EBEB@mail.python.org>

Author: georg.brandl
Date: Mon Jul 12 11:01:14 2010
New Revision: 82833

Log:
Fix glossary markup.

Modified:
   python/branches/release31-maint/Doc/glossary.rst

Modified: python/branches/release31-maint/Doc/glossary.rst
==============================================================================
--- python/branches/release31-maint/Doc/glossary.rst	(original)
+++ python/branches/release31-maint/Doc/glossary.rst	Mon Jul 12 11:01:14 2010
@@ -420,7 +420,6 @@
       modules, respectively.
 
    nested scope
-
       The ability to refer to a variable in an enclosing definition.  For
       instance, a function defined inside another function can refer to
       variables in the outer function.  Note that nested scopes by default work

From python-checkins at python.org  Mon Jul 12 11:06:13 2010
From: python-checkins at python.org (georg.brandl)
Date: Mon, 12 Jul 2010 11:06:13 +0200 (CEST)
Subject: [Python-checkins] r82834 - python/branches/py3k/Doc/library/re.rst
Message-ID: <20100712090613.9E473EE9D1@mail.python.org>

Author: georg.brandl
Date: Mon Jul 12 11:06:13 2010
New Revision: 82834

Log:
Use raw string literals for regexes containing backlash.

Modified:
   python/branches/py3k/Doc/library/re.rst

Modified: python/branches/py3k/Doc/library/re.rst
==============================================================================
--- python/branches/py3k/Doc/library/re.rst	(original)
+++ python/branches/py3k/Doc/library/re.rst	Mon Jul 12 11:06:13 2010
@@ -1230,9 +1230,9 @@
    ...   random.shuffle(inner_word)
    ...   return m.group(1) + "".join(inner_word) + m.group(3)
    >>> text = "Professor Abdolmalek, please report your absences promptly."
-   >>> re.sub("(\w)(\w+)(\w)", repl, text)
+   >>> re.sub(r"(\w)(\w+)(\w)", repl, text)
    'Poefsrosr Aealmlobdk, pslaee reorpt your abnseces plmrptoy.'
-   >>> re.sub("(\w)(\w+)(\w)", repl, text)
+   >>> re.sub(r"(\w)(\w+)(\w)", repl, text)
    'Pofsroser Aodlambelk, plasee reoprt yuor asnebces potlmrpy.'
 
 

From python-checkins at python.org  Mon Jul 12 11:37:41 2010
From: python-checkins at python.org (mark.dickinson)
Date: Mon, 12 Jul 2010 11:37:41 +0200 (CEST)
Subject: [Python-checkins] r82835 -
	python/branches/py3k/Doc/tutorial/classes.rst
Message-ID: <20100712093741.15D67EE9BE@mail.python.org>

Author: mark.dickinson
Date: Mon Jul 12 11:37:40 2010
New Revision: 82835

Log:
Remove mention of execfile from the tutorial.

Modified:
   python/branches/py3k/Doc/tutorial/classes.rst

Modified: python/branches/py3k/Doc/tutorial/classes.rst
==============================================================================
--- python/branches/py3k/Doc/tutorial/classes.rst	(original)
+++ python/branches/py3k/Doc/tutorial/classes.rst	Mon Jul 12 11:37:40 2010
@@ -599,12 +599,12 @@
 possible to access or modify a variable that is considered private.  This can
 even be useful in special circumstances, such as in the debugger.
 
-Notice that code passed to ``exec()``, ``eval()`` or ``execfile()`` does not
-consider the classname of the invoking  class to be the current class; this is
-similar to the effect of the  ``global`` statement, the effect of which is
-likewise restricted to  code that is byte-compiled together.  The same
-restriction applies to ``getattr()``, ``setattr()`` and ``delattr()``, as well
-as when referencing ``__dict__`` directly.
+Notice that code passed to ``exec()`` or ``eval()`` does not consider the
+classname of the invoking class to be the current class; this is similar to the
+effect of the ``global`` statement, the effect of which is likewise restricted
+to code that is byte-compiled together.  The same restriction applies to
+``getattr()``, ``setattr()`` and ``delattr()``, as well as when referencing
+``__dict__`` directly.
 
 
 .. _tut-odds:

From python-checkins at python.org  Mon Jul 12 11:39:11 2010
From: python-checkins at python.org (mark.dickinson)
Date: Mon, 12 Jul 2010 11:39:11 +0200 (CEST)
Subject: [Python-checkins] r82836 - in python/branches/release31-maint:
	Doc/tutorial/classes.rst
Message-ID: <20100712093911.9ADB5EE9BE@mail.python.org>

Author: mark.dickinson
Date: Mon Jul 12 11:39:08 2010
New Revision: 82836

Log:
Merged revisions 82835 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82835 | mark.dickinson | 2010-07-12 10:37:40 +0100 (Mon, 12 Jul 2010) | 1 line
  
  Remove mention of execfile from the tutorial.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Doc/tutorial/classes.rst

Modified: python/branches/release31-maint/Doc/tutorial/classes.rst
==============================================================================
--- python/branches/release31-maint/Doc/tutorial/classes.rst	(original)
+++ python/branches/release31-maint/Doc/tutorial/classes.rst	Mon Jul 12 11:39:08 2010
@@ -599,12 +599,12 @@
 possible to access or modify a variable that is considered private.  This can
 even be useful in special circumstances, such as in the debugger.
 
-Notice that code passed to ``exec()``, ``eval()`` or ``execfile()`` does not
-consider the classname of the invoking  class to be the current class; this is
-similar to the effect of the  ``global`` statement, the effect of which is
-likewise restricted to  code that is byte-compiled together.  The same
-restriction applies to ``getattr()``, ``setattr()`` and ``delattr()``, as well
-as when referencing ``__dict__`` directly.
+Notice that code passed to ``exec()`` or ``eval()`` does not consider the
+classname of the invoking class to be the current class; this is similar to the
+effect of the ``global`` statement, the effect of which is likewise restricted
+to code that is byte-compiled together.  The same restriction applies to
+``getattr()``, ``setattr()`` and ``delattr()``, as well as when referencing
+``__dict__`` directly.
 
 
 .. _tut-odds:

From python-checkins at python.org  Mon Jul 12 16:14:18 2010
From: python-checkins at python.org (mark.dickinson)
Date: Mon, 12 Jul 2010 16:14:18 +0200 (CEST)
Subject: [Python-checkins] r82837 - python/branches/py3k/Grammar/Grammar
Message-ID: <20100712141418.BF8FFEE996@mail.python.org>

Author: mark.dickinson
Date: Mon Jul 12 16:14:18 2010
New Revision: 82837

Log:
Issue #2009: refactor varargslist and typedargslist productions to make them more friendly for third-party parsers.

Modified:
   python/branches/py3k/Grammar/Grammar

Modified: python/branches/py3k/Grammar/Grammar
==============================================================================
--- python/branches/py3k/Grammar/Grammar	(original)
+++ python/branches/py3k/Grammar/Grammar	Mon Jul 12 16:14:18 2010
@@ -24,13 +24,13 @@
 decorated: decorators (classdef | funcdef)
 funcdef: 'def' NAME parameters ['->' test] ':' suite
 parameters: '(' [typedargslist] ')'
-typedargslist: ((tfpdef ['=' test] ',')*
-                ('*' [tfpdef] (',' tfpdef ['=' test])* [',' '**' tfpdef] | '**' tfpdef)
-                | tfpdef ['=' test] (',' tfpdef ['=' test])* [','])
+typedargslist: (tfpdef ['=' test] (',' tfpdef ['=' test])* [','
+       ['*' [tfpdef] (',' tfpdef ['=' test])* [',' '**' tfpdef] | '**' tfpdef]]
+     |  '*' [tfpdef] (',' tfpdef ['=' test])* [',' '**' tfpdef] | '**' tfpdef)
 tfpdef: NAME [':' test]
-varargslist: ((vfpdef ['=' test] ',')*
-              ('*' [vfpdef] (',' vfpdef ['=' test])* [',' '**' vfpdef] | '**' vfpdef)
-              | vfpdef ['=' test] (',' vfpdef ['=' test])* [','])
+varargslist: (vfpdef ['=' test] (',' vfpdef ['=' test])* [','
+       ['*' [vfpdef] (',' vfpdef ['=' test])* [',' '**' vfpdef] | '**' vfpdef]]
+     |  '*' [vfpdef] (',' vfpdef ['=' test])* [',' '**' vfpdef] | '**' vfpdef)
 vfpdef: NAME
 
 stmt: simple_stmt | compound_stmt

From python-checkins at python.org  Mon Jul 12 16:18:21 2010
From: python-checkins at python.org (mark.dickinson)
Date: Mon, 12 Jul 2010 16:18:21 +0200 (CEST)
Subject: [Python-checkins] r82838 - python/branches/py3k/Python/graminit.c
Message-ID: <20100712141821.ED507EE9D3@mail.python.org>

Author: mark.dickinson
Date: Mon Jul 12 16:18:21 2010
New Revision: 82838

Log:
Regenerate Python/graminit.c.

Modified:
   python/branches/py3k/Python/graminit.c

Modified: python/branches/py3k/Python/graminit.c
==============================================================================
--- python/branches/py3k/Python/graminit.c	(original)
+++ python/branches/py3k/Python/graminit.c	Mon Jul 12 16:18:21 2010
@@ -180,8 +180,8 @@
     {24, 9},
 };
 static arc arcs_8_5[4] = {
-    {28, 1},
-    {31, 2},
+    {28, 10},
+    {31, 11},
     {32, 3},
     {0, 5},
 };
@@ -190,7 +190,7 @@
     {0, 6},
 };
 static arc arcs_8_7[2] = {
-    {28, 10},
+    {28, 12},
     {32, 3},
 };
 static arc arcs_8_8[1] = {
@@ -201,14 +201,40 @@
     {0, 9},
 };
 static arc arcs_8_10[3] = {
-    {30, 7},
-    {29, 11},
+    {30, 5},
+    {29, 4},
     {0, 10},
 };
-static arc arcs_8_11[1] = {
+static arc arcs_8_11[3] = {
+    {28, 13},
+    {30, 14},
+    {0, 11},
+};
+static arc arcs_8_12[3] = {
+    {30, 7},
+    {29, 15},
+    {0, 12},
+};
+static arc arcs_8_13[2] = {
+    {30, 14},
+    {0, 13},
+};
+static arc arcs_8_14[2] = {
+    {28, 16},
+    {32, 3},
+};
+static arc arcs_8_15[1] = {
     {24, 6},
 };
-static state states_8[12] = {
+static arc arcs_8_16[3] = {
+    {30, 14},
+    {29, 17},
+    {0, 16},
+};
+static arc arcs_8_17[1] = {
+    {24, 13},
+};
+static state states_8[18] = {
     {3, arcs_8_0},
     {3, arcs_8_1},
     {3, arcs_8_2},
@@ -220,7 +246,13 @@
     {1, arcs_8_8},
     {2, arcs_8_9},
     {3, arcs_8_10},
-    {1, arcs_8_11},
+    {3, arcs_8_11},
+    {3, arcs_8_12},
+    {2, arcs_8_13},
+    {2, arcs_8_14},
+    {1, arcs_8_15},
+    {3, arcs_8_16},
+    {1, arcs_8_17},
 };
 static arc arcs_9_0[1] = {
     {21, 1},
@@ -263,8 +295,8 @@
     {24, 9},
 };
 static arc arcs_10_5[4] = {
-    {34, 1},
-    {31, 2},
+    {34, 10},
+    {31, 11},
     {32, 3},
     {0, 5},
 };
@@ -273,7 +305,7 @@
     {0, 6},
 };
 static arc arcs_10_7[2] = {
-    {34, 10},
+    {34, 12},
     {32, 3},
 };
 static arc arcs_10_8[1] = {
@@ -284,14 +316,40 @@
     {0, 9},
 };
 static arc arcs_10_10[3] = {
-    {30, 7},
-    {29, 11},
+    {30, 5},
+    {29, 4},
     {0, 10},
 };
-static arc arcs_10_11[1] = {
+static arc arcs_10_11[3] = {
+    {34, 13},
+    {30, 14},
+    {0, 11},
+};
+static arc arcs_10_12[3] = {
+    {30, 7},
+    {29, 15},
+    {0, 12},
+};
+static arc arcs_10_13[2] = {
+    {30, 14},
+    {0, 13},
+};
+static arc arcs_10_14[2] = {
+    {34, 16},
+    {32, 3},
+};
+static arc arcs_10_15[1] = {
     {24, 6},
 };
-static state states_10[12] = {
+static arc arcs_10_16[3] = {
+    {30, 14},
+    {29, 17},
+    {0, 16},
+};
+static arc arcs_10_17[1] = {
+    {24, 13},
+};
+static state states_10[18] = {
     {3, arcs_10_0},
     {3, arcs_10_1},
     {3, arcs_10_2},
@@ -303,7 +361,13 @@
     {1, arcs_10_8},
     {2, arcs_10_9},
     {3, arcs_10_10},
-    {1, arcs_10_11},
+    {3, arcs_10_11},
+    {3, arcs_10_12},
+    {2, arcs_10_13},
+    {2, arcs_10_14},
+    {1, arcs_10_15},
+    {3, arcs_10_16},
+    {1, arcs_10_17},
 };
 static arc arcs_11_0[1] = {
     {21, 1},
@@ -1755,11 +1819,11 @@
      "\000\000\020\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000"},
     {263, "parameters", 0, 4, states_7,
      "\000\040\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000"},
-    {264, "typedargslist", 0, 12, states_8,
+    {264, "typedargslist", 0, 18, states_8,
      "\000\000\040\200\001\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000"},
     {265, "tfpdef", 0, 4, states_9,
      "\000\000\040\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000"},
-    {266, "varargslist", 0, 12, states_10,
+    {266, "varargslist", 0, 18, states_10,
      "\000\000\040\200\001\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000"},
     {267, "vfpdef", 0, 2, states_11,
      "\000\000\040\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000"},

From python-checkins at python.org  Mon Jul 12 21:49:41 2010
From: python-checkins at python.org (ezio.melotti)
Date: Mon, 12 Jul 2010 21:49:41 +0200 (CEST)
Subject: [Python-checkins] r82839 -
	python/branches/py3k/Lib/distutils/tests/test_sdist.py
Message-ID: <20100712194941.CF3B9EE9F1@mail.python.org>

Author: ezio.melotti
Date: Mon Jul 12 21:49:41 2010
New Revision: 82839

Log:
#6026: skip test_get_file_list when zlib is not available.

Modified:
   python/branches/py3k/Lib/distutils/tests/test_sdist.py

Modified: python/branches/py3k/Lib/distutils/tests/test_sdist.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_sdist.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/test_sdist.py	Mon Jul 12 21:49:41 2010
@@ -346,6 +346,7 @@
         finally:
             archive.close()
 
+    @unittest.skipUnless(zlib, "requires zlib")
     def test_get_file_list(self):
         # make sure MANIFEST is recalculated
         dist, cmd = self.get_cmd()

From python-checkins at python.org  Mon Jul 12 21:52:15 2010
From: python-checkins at python.org (ezio.melotti)
Date: Mon, 12 Jul 2010 21:52:15 +0200 (CEST)
Subject: [Python-checkins] r82840 - in python/branches/release27-maint:
	Lib/distutils/tests/test_sdist.py
Message-ID: <20100712195215.C035DEEBA5@mail.python.org>

Author: ezio.melotti
Date: Mon Jul 12 21:52:15 2010
New Revision: 82840

Log:
Merged revisions 82839 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82839 | ezio.melotti | 2010-07-12 22:49:41 +0300 (Mon, 12 Jul 2010) | 1 line
  
  #6026: skip test_get_file_list when zlib is not available.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/distutils/tests/test_sdist.py

Modified: python/branches/release27-maint/Lib/distutils/tests/test_sdist.py
==============================================================================
--- python/branches/release27-maint/Lib/distutils/tests/test_sdist.py	(original)
+++ python/branches/release27-maint/Lib/distutils/tests/test_sdist.py	Mon Jul 12 21:52:15 2010
@@ -346,6 +346,7 @@
         finally:
             archive.close()
 
+    @unittest.skipUnless(zlib, "requires zlib")
     def test_get_file_list(self):
         # make sure MANIFEST is recalculated
         dist, cmd = self.get_cmd()

From python-checkins at python.org  Mon Jul 12 22:00:40 2010
From: python-checkins at python.org (ezio.melotti)
Date: Mon, 12 Jul 2010 22:00:40 +0200 (CEST)
Subject: [Python-checkins] r82841 - in python/branches/release31-maint:
	Lib/distutils/tests/test_sdist.py
Message-ID: <20100712200040.22876EEC41@mail.python.org>

Author: ezio.melotti
Date: Mon Jul 12 22:00:39 2010
New Revision: 82841

Log:
Merged revisions 82839 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82839 | ezio.melotti | 2010-07-12 22:49:41 +0300 (Mon, 12 Jul 2010) | 1 line
  
  #6026: skip test_get_file_list when zlib is not available.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/distutils/tests/test_sdist.py

Modified: python/branches/release31-maint/Lib/distutils/tests/test_sdist.py
==============================================================================
--- python/branches/release31-maint/Lib/distutils/tests/test_sdist.py	(original)
+++ python/branches/release31-maint/Lib/distutils/tests/test_sdist.py	Mon Jul 12 22:00:39 2010
@@ -277,6 +277,7 @@
         self.assertRaises(DistutilsOptionError, cmd.finalize_options)
 
 
+    @unittest.skipUnless(zlib, "requires zlib")
     def test_get_file_list(self):
         # make sure MANIFEST is recalculated
         dist, cmd = self.get_cmd()

From python-checkins at python.org  Mon Jul 12 22:01:53 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Mon, 12 Jul 2010 22:01:53 +0200 (CEST)
Subject: [Python-checkins] r82842 -
	python/branches/py3k/Doc/library/stdtypes.rst
Message-ID: <20100712200153.0D3EDEEA63@mail.python.org>

Author: antoine.pitrou
Date: Mon Jul 12 22:01:52 2010
New Revision: 82842

Log:
Fix definition of len() and indexing for memoryview objects (part of #7696).



Modified:
   python/branches/py3k/Doc/library/stdtypes.rst

Modified: python/branches/py3k/Doc/library/stdtypes.rst
==============================================================================
--- python/branches/py3k/Doc/library/stdtypes.rst	(original)
+++ python/branches/py3k/Doc/library/stdtypes.rst	Mon Jul 12 22:01:52 2010
@@ -2264,10 +2264,19 @@
    buffer protocol.  Builtin objects that support the buffer protocol include
    :class:`bytes` and :class:`bytearray`.
 
-   ``len(view)`` returns the total number of bytes in the memoryview, *view*.
+   A :class:`memoryview` has the notion of an *element*, which is the
+   atomic memory unit handled by the originating object *obj*.  For many
+   simple types such as :class:`bytes` and :class:`bytearray`, an element
+   is a single byte, but other types such as :class:`array.array` may have
+   bigger elements.
+
+   ``len(view)`` returns the total number of elements in the memoryview,
+   *view*.  The :class:`~memoryview.itemsize` attribute will give you the
+   number of bytes in a single element.
 
    A :class:`memoryview` supports slicing to expose its data.  Taking a single
-   index will return a single byte.  Full slicing will result in a subview::
+   index will return a single element as a :class:`bytes` object.  Full
+   slicing will result in a subview::
 
       >>> v = memoryview(b'abcefg')
       >>> v[1]
@@ -2278,11 +2287,8 @@
       <memory at 0x77ab28>
       >>> bytes(v[1:4])
       b'bce'
-      >>> v[3:-1]
-      <memory at 0x744f18>
-      >>> bytes(v[4:-1])
 
-   If the object the memory view is over supports changing its data, the
+   If the object the memoryview is over supports changing its data, the
    memoryview supports slice assignment::
 
       >>> data = bytearray(b'abcefg')
@@ -2302,12 +2308,18 @@
 
    Notice how the size of the memoryview object cannot be changed.
 
-
    :class:`memoryview` has two methods:
 
    .. method:: tobytes()
 
-      Return the data in the buffer as a bytestring.
+      Return the data in the buffer as a bytestring.  This is equivalent to
+      calling the :class:`bytes` constructor on the memoryview. ::
+
+         >>> m = memoryview(b"abc")
+         >>> m.tobytes()
+         b'abc'
+         >>> bytes(m)
+         b'abc'
 
    .. method:: tolist()
 
@@ -2325,7 +2337,15 @@
 
    .. attribute:: itemsize
 
-      The size in bytes of each element of the memoryview.
+      The size in bytes of each element of the memoryview::
+
+         >>> m = memoryview(array.array('H', [1,2,3]))
+         >>> m.itemsize
+         2
+         >>> m[0]
+         b'\x01\x00'
+         >>> len(m[0]) == m.itemsize
+         True
 
    .. attribute:: shape
 

From python-checkins at python.org  Mon Jul 12 22:03:24 2010
From: python-checkins at python.org (mark.dickinson)
Date: Mon, 12 Jul 2010 22:03:24 +0200 (CEST)
Subject: [Python-checkins] r82843 -
	python/branches/py3k/Lib/test/test_struct.py
Message-ID: <20100712200324.D87F2EEC2C@mail.python.org>

Author: mark.dickinson
Date: Mon Jul 12 22:03:24 2010
New Revision: 82843

Log:
Yield more information on failure in test_struct boolean test.

Modified:
   python/branches/py3k/Lib/test/test_struct.py

Modified: python/branches/py3k/Lib/test/test_struct.py
==============================================================================
--- python/branches/py3k/Lib/test/test_struct.py	(original)
+++ python/branches/py3k/Lib/test/test_struct.py	Mon Jul 12 22:03:24 2010
@@ -512,8 +512,13 @@
                 self.assertFalse(prefix, msg='encoded bool is not one byte: %r'
                                              %packed)
 
-            self.assertRaises(IOError, struct.pack, prefix + '?',
-                              ExplodingBool())
+            try:
+                struct.pack(prefix + '?', ExplodingBool())
+            except IOError:
+                pass
+            else:
+                self.fail("Expected IOError: struct.pack(%r, "
+                          "ExplodingBool())" % (prefix + '?'))
 
         for c in [b'\x01', b'\x7f', b'\xff', b'\x0f', b'\xf0']:
             self.assertTrue(struct.unpack('>?', c)[0])

From python-checkins at python.org  Mon Jul 12 22:04:04 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Mon, 12 Jul 2010 22:04:04 +0200 (CEST)
Subject: [Python-checkins] r82844 - in python/branches/release31-maint:
	Doc/library/stdtypes.rst
Message-ID: <20100712200404.867C0EEC35@mail.python.org>

Author: antoine.pitrou
Date: Mon Jul 12 22:04:01 2010
New Revision: 82844

Log:
Merged revisions 82842 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82842 | antoine.pitrou | 2010-07-12 22:01:52 +0200 (lun., 12 juil. 2010) | 3 lines
  
  Fix definition of len() and indexing for memoryview objects (part of #7696).
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Doc/library/stdtypes.rst

Modified: python/branches/release31-maint/Doc/library/stdtypes.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/stdtypes.rst	(original)
+++ python/branches/release31-maint/Doc/library/stdtypes.rst	Mon Jul 12 22:04:01 2010
@@ -2086,10 +2086,19 @@
    buffer protocol.  Builtin objects that support the buffer protocol include
    :class:`bytes` and :class:`bytearray`.
 
-   ``len(view)`` returns the total number of bytes in the memoryview, *view*.
+   A :class:`memoryview` has the notion of an *element*, which is the
+   atomic memory unit handled by the originating object *obj*.  For many
+   simple types such as :class:`bytes` and :class:`bytearray`, an element
+   is a single byte, but other types such as :class:`array.array` may have
+   bigger elements.
+
+   ``len(view)`` returns the total number of elements in the memoryview,
+   *view*.  The :class:`~memoryview.itemsize` attribute will give you the
+   number of bytes in a single element.
 
    A :class:`memoryview` supports slicing to expose its data.  Taking a single
-   index will return a single byte.  Full slicing will result in a subview::
+   index will return a single element as a :class:`bytes` object.  Full
+   slicing will result in a subview::
 
       >>> v = memoryview(b'abcefg')
       >>> v[1]
@@ -2100,11 +2109,8 @@
       <memory at 0x77ab28>
       >>> bytes(v[1:4])
       b'bce'
-      >>> v[3:-1]
-      <memory at 0x744f18>
-      >>> bytes(v[4:-1])
 
-   If the object the memory view is over supports changing its data, the
+   If the object the memoryview is over supports changing its data, the
    memoryview supports slice assignment::
 
       >>> data = bytearray(b'abcefg')
@@ -2124,12 +2130,18 @@
 
    Notice how the size of the memoryview object cannot be changed.
 
-
    :class:`memoryview` has two methods:
 
    .. method:: tobytes()
 
-      Return the data in the buffer as a bytestring.
+      Return the data in the buffer as a bytestring.  This is equivalent to
+      calling the :class:`bytes` constructor on the memoryview. ::
+
+         >>> m = memoryview(b"abc")
+         >>> m.tobytes()
+         b'abc'
+         >>> bytes(m)
+         b'abc'
 
    .. method:: tolist()
 
@@ -2147,7 +2159,15 @@
 
    .. attribute:: itemsize
 
-      The size in bytes of each element of the memoryview.
+      The size in bytes of each element of the memoryview::
+
+         >>> m = memoryview(array.array('H', [1,2,3]))
+         >>> m.itemsize
+         2
+         >>> m[0]
+         b'\x01\x00'
+         >>> len(m[0]) == m.itemsize
+         True
 
    .. attribute:: shape
 

From python-checkins at python.org  Mon Jul 12 22:11:53 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Mon, 12 Jul 2010 22:11:53 +0200 (CEST)
Subject: [Python-checkins] r82845 - in python/branches/release27-maint:
	Doc/library/stdtypes.rst
Message-ID: <20100712201153.21E7AEE988@mail.python.org>

Author: antoine.pitrou
Date: Mon Jul 12 22:11:52 2010
New Revision: 82845

Log:
Merged revisions 82842 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82842 | antoine.pitrou | 2010-07-12 22:01:52 +0200 (lun., 12 juil. 2010) | 3 lines
  
  Fix definition of len() and indexing for memoryview objects (part of #7696).
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Doc/library/stdtypes.rst

Modified: python/branches/release27-maint/Doc/library/stdtypes.rst
==============================================================================
--- python/branches/release27-maint/Doc/library/stdtypes.rst	(original)
+++ python/branches/release27-maint/Doc/library/stdtypes.rst	Mon Jul 12 22:11:52 2010
@@ -2554,10 +2554,18 @@
    buffer protocol.  Builtin objects that support the buffer protocol include
    :class:`str` and :class:`bytearray` (but not :class:`unicode`).
 
-   ``len(view)`` returns the total number of bytes in the memoryview, *view*.
+   A :class:`memoryview` has the notion of an *element*, which is the
+   atomic memory unit handled by the originating object *obj*.  For many
+   simple types such as :class:`str` and :class:`bytearray`, an element
+   is a single byte, but other third-party types may expose larger elements.
+
+   ``len(view)`` returns the total number of elements in the memoryview,
+   *view*.  The :class:`~memoryview.itemsize` attribute will give you the
+   number of bytes in a single element.
 
    A :class:`memoryview` supports slicing to expose its data.  Taking a single
-   index will return a single byte.  Full slicing will result in a subview::
+   index will return a single element as a :class:`str` object.  Full
+   slicing will result in a subview::
 
       >>> v = memoryview('abcefg')
       >>> v[1]
@@ -2568,12 +2576,8 @@
       <memory at 0x77ab28>
       >>> str(v[1:4])
       'bce'
-      >>> v[3:-1]
-      <memory at 0x744f18>
-      >>> str(v[4:-1])
-      'f'
 
-   If the object the memory view is over supports changing its data, the
+   If the object the memoryview is over supports changing its data, the
    memoryview supports slice assignment::
 
       >>> data = bytearray('abcefg')
@@ -2593,13 +2597,16 @@
 
    Notice how the size of the memoryview object cannot be changed.
 
-
    :class:`memoryview` has two methods:
 
    .. method:: tobytes()
 
       Return the data in the buffer as a bytestring (an object of class
-      :class:`str`).
+      :class:`str`). ::
+
+         >>> m = memoryview("abc")
+         >>> m.tobytes()
+         'abc'
 
    .. method:: tolist()
 

From python-checkins at python.org  Mon Jul 12 23:23:20 2010
From: python-checkins at python.org (ezio.melotti)
Date: Mon, 12 Jul 2010 23:23:20 +0200 (CEST)
Subject: [Python-checkins] r82846 - in python/branches/release31-maint:
	Lib/distutils/tests/test_sdist.py
Message-ID: <20100712212320.87801EEC7E@mail.python.org>

Author: ezio.melotti
Date: Mon Jul 12 23:23:20 2010
New Revision: 82846

Log:
Revert r82841.

Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/distutils/tests/test_sdist.py

Modified: python/branches/release31-maint/Lib/distutils/tests/test_sdist.py
==============================================================================
--- python/branches/release31-maint/Lib/distutils/tests/test_sdist.py	(original)
+++ python/branches/release31-maint/Lib/distutils/tests/test_sdist.py	Mon Jul 12 23:23:20 2010
@@ -277,7 +277,6 @@
         self.assertRaises(DistutilsOptionError, cmd.finalize_options)
 
 
-    @unittest.skipUnless(zlib, "requires zlib")
     def test_get_file_list(self):
         # make sure MANIFEST is recalculated
         dist, cmd = self.get_cmd()

From python-checkins at python.org  Mon Jul 12 23:24:41 2010
From: python-checkins at python.org (ezio.melotti)
Date: Mon, 12 Jul 2010 23:24:41 +0200 (CEST)
Subject: [Python-checkins] r82847 - python/branches/release31-maint
Message-ID: <20100712212441.E3653EEC4B@mail.python.org>

Author: ezio.melotti
Date: Mon Jul 12 23:24:41 2010
New Revision: 82847

Log:
Blocked revisions 82839 via svnmerge

........
  r82839 | ezio.melotti | 2010-07-12 22:49:41 +0300 (Mon, 12 Jul 2010) | 1 line
  
  #6026: skip test_get_file_list when zlib is not available.
........


Modified:
   python/branches/release31-maint/   (props changed)

From solipsis at pitrou.net  Tue Jul 13 01:24:39 2010
From: solipsis at pitrou.net (solipsis at pitrou.net)
Date: Tue, 13 Jul 2010 01:24:39 +0200 (CEST)
Subject: [Python-checkins] Daily py3k reference leaks (r82843): sum=0
Message-ID: <20100712232439.A813E1771C@ns6635.ovh.net>

py3k results for svn r82843 (hg cset 7196c55ccaaa)
--------------------------------------------------



Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/antoine/py3k/refleaks/reflogmTUz4J', '-x']

From nnorwitz at gmail.com  Tue Jul 13 02:26:12 2010
From: nnorwitz at gmail.com (Neal Norwitz)
Date: Mon, 12 Jul 2010 20:26:12 -0400
Subject: [Python-checkins] Python Regression Test Failures refleak (1)
Message-ID: <20100713002612.GA29244@kbk-i386-bb.psfb.org>

More important issues:
----------------------
test_bz2 leaked [-77, 0, 0] references, sum=-77

Less important issues:
----------------------
test_threadsignals leaked [0, -8, 8] references, sum=0

From python-checkins at python.org  Tue Jul 13 08:38:10 2010
From: python-checkins at python.org (georg.brandl)
Date: Tue, 13 Jul 2010 08:38:10 +0200 (CEST)
Subject: [Python-checkins] r82848 - python/branches/py3k/Doc/library/ast.rst
Message-ID: <20100713063810.B0E67EEB8F@mail.python.org>

Author: georg.brandl
Date: Tue Jul 13 08:38:10 2010
New Revision: 82848

Log:
Add bytes in literal_eval doc.

Modified:
   python/branches/py3k/Doc/library/ast.rst

Modified: python/branches/py3k/Doc/library/ast.rst
==============================================================================
--- python/branches/py3k/Doc/library/ast.rst	(original)
+++ python/branches/py3k/Doc/library/ast.rst	Tue Jul 13 08:38:10 2010
@@ -119,14 +119,14 @@
 
    Safely evaluate an expression node or a string containing a Python
    expression.  The string or node provided may only consist of the following
-   Python literal structures: strings, numbers, tuples, lists, dicts, sets,
-   booleans, and ``None``.
+   Python literal structures: strings, bytes, numbers, tuples, lists, dicts,
+   sets, booleans, and ``None``.
 
    This can be used for safely evaluating strings containing Python expressions
    from untrusted sources without the need to parse the values oneself.
 
    .. versionchanged:: 3.2
-      Now allows set literals.
+      Now allows bytes and set literals.
 
 
 .. function:: get_docstring(node, clean=True)

From python-checkins at python.org  Tue Jul 13 16:47:02 2010
From: python-checkins at python.org (jeroen.ruigrok)
Date: Tue, 13 Jul 2010 16:47:02 +0200 (CEST)
Subject: [Python-checkins] r82849 - python/branches/py3k/Doc/library/os.rst
Message-ID: <20100713144702.27009EE98D@mail.python.org>

Author: jeroen.ruigrok
Date: Tue Jul 13 16:47:01 2010
New Revision: 82849

Log:
Fix documentation typo: wprite() -> write().


Modified:
   python/branches/py3k/Doc/library/os.rst

Modified: python/branches/py3k/Doc/library/os.rst
==============================================================================
--- python/branches/py3k/Doc/library/os.rst	(original)
+++ python/branches/py3k/Doc/library/os.rst	Tue Jul 13 16:47:01 2010
@@ -723,7 +723,7 @@
 
       This function is intended for low-level I/O.  For normal usage, use the
       built-in function :func:`open`, which returns a "file object" with
-      :meth:`~file.read` and :meth:`~file.wprite` methods (and many more).  To
+      :meth:`~file.read` and :meth:`~file.write` methods (and many more).  To
       wrap a file descriptor in a "file object", use :func:`fdopen`.
 
 

From python-checkins at python.org  Tue Jul 13 16:50:17 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Tue, 13 Jul 2010 16:50:17 +0200 (CEST)
Subject: [Python-checkins] r82850 - python/branches/py3k/Lib/test/support.py
Message-ID: <20100713145017.13683EEA30@mail.python.org>

Author: alexander.belopolsky
Date: Tue Jul 13 16:50:16 2010
New Revision: 82850

Log:
Set sys.modules[name] to None instead of 0 to block module import.

Modified:
   python/branches/py3k/Lib/test/support.py

Modified: python/branches/py3k/Lib/test/support.py
==============================================================================
--- python/branches/py3k/Lib/test/support.py	(original)
+++ python/branches/py3k/Lib/test/support.py	Tue Jul 13 16:50:16 2010
@@ -109,7 +109,7 @@
         orig_modules[name] = sys.modules[name]
     except KeyError:
         saved = False
-    sys.modules[name] = 0
+    sys.modules[name] = None
     return saved
 
 

From python-checkins at python.org  Tue Jul 13 16:55:04 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Tue, 13 Jul 2010 16:55:04 +0200 (CEST)
Subject: [Python-checkins] r82851 - in python/branches/release31-maint:
	Lib/test/support.py
Message-ID: <20100713145504.D0CDFEE9F0@mail.python.org>

Author: alexander.belopolsky
Date: Tue Jul 13 16:55:04 2010
New Revision: 82851

Log:
Merged revisions 82850 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82850 | alexander.belopolsky | 2010-07-13 10:50:16 -0400 (Tue, 13 Jul 2010) | 1 line
  
  Set sys.modules[name] to None instead of 0 to block module import.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/test/support.py

Modified: python/branches/release31-maint/Lib/test/support.py
==============================================================================
--- python/branches/release31-maint/Lib/test/support.py	(original)
+++ python/branches/release31-maint/Lib/test/support.py	Tue Jul 13 16:55:04 2010
@@ -98,7 +98,7 @@
         orig_modules[name] = sys.modules[name]
     except KeyError:
         saved = False
-    sys.modules[name] = 0
+    sys.modules[name] = None
     return saved
 
 

From python-checkins at python.org  Tue Jul 13 17:08:30 2010
From: python-checkins at python.org (jeroen.ruigrok)
Date: Tue, 13 Jul 2010 17:08:30 +0200 (CEST)
Subject: [Python-checkins] r82852 - in python/branches/release27-maint:
	Doc/library/os.rst
Message-ID: <20100713150830.802F8CA6A@mail.python.org>

Author: jeroen.ruigrok
Date: Tue Jul 13 17:08:30 2010
New Revision: 82852

Log:
Merged revisions 82849 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82849 | jeroen.ruigrok | 2010-07-13 16:47:01 +0200 (di, 13 jul 2010) | 2 lines
  
  Fix documentation typo: wprite() -> write().
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Doc/library/os.rst

Modified: python/branches/release27-maint/Doc/library/os.rst
==============================================================================
--- python/branches/release27-maint/Doc/library/os.rst	(original)
+++ python/branches/release27-maint/Doc/library/os.rst	Tue Jul 13 17:08:30 2010
@@ -756,7 +756,7 @@
 
       This function is intended for low-level I/O.  For normal usage, use the
       built-in function :func:`open`, which returns a "file object" with
-      :meth:`~file.read` and :meth:`~file.wprite` methods (and many more).  To
+      :meth:`~file.read` and :meth:`~file.write` methods (and many more).  To
       wrap a file descriptor in a "file object", use :func:`fdopen`.
 
 

From ezio.melotti at gmail.com  Tue Jul 13 19:17:55 2010
From: ezio.melotti at gmail.com (Ezio Melotti)
Date: Tue, 13 Jul 2010 20:17:55 +0300
Subject: [Python-checkins] r82848 -
	python/branches/py3k/Doc/library/ast.rst
In-Reply-To: <20100713063810.B0E67EEB8F@mail.python.org>
References: <20100713063810.B0E67EEB8F@mail.python.org>
Message-ID: <4C3C9FC3.6050604@gmail.com>

  On 13/07/2010 9.38, georg.brandl wrote:
> Author: georg.brandl
> Date: Tue Jul 13 08:38:10 2010
> New Revision: 82848
>
> Log:
> Add bytes in literal_eval doc.
>
> Modified:
>     python/branches/py3k/Doc/library/ast.rst
>
> Modified: python/branches/py3k/Doc/library/ast.rst
> ==============================================================================
> --- python/branches/py3k/Doc/library/ast.rst	(original)
> +++ python/branches/py3k/Doc/library/ast.rst	Tue Jul 13 08:38:10 2010
> @@ -119,14 +119,14 @@
>
>      Safely evaluate an expression node or a string containing a Python
>      expression.  The string or node provided may only consist of the following
> -   Python literal structures: strings, numbers, tuples, lists, dicts, sets,
> -   booleans, and ``None``.
> +   Python literal structures: strings, bytes, numbers, tuples, lists, dicts,
> +   sets, booleans, and ``None``.
>
>      This can be used for safely evaluating strings containing Python expressions
>      from untrusted sources without the need to parse the values oneself.
>
>      .. versionchanged:: 3.2
> -      Now allows set literals.
> +      Now allows bytes and set literals.
>
>
>   .. function:: get_docstring(node, clean=True)
> _______________________________________________
> Python-checkins mailing list
> Python-checkins at python.org
> http://mail.python.org/mailman/listinfo/python-checkins
>
This has been fixed in 3.1 too in r82829. As I said on IRC I consider 
this (and the missing set literals on 2.7/3.1) as a bug, so I would:
   1) remove the versionchanged from 3.2;
   2) add bytes literal in the doc for 3.1 (the fix has been backported 
already);
   3) backport and document set literals to 3.1 and 2.7 too;
   4) possibly update the doc for byte literals in 2.7 too (it just says 
'strings' but it supports u"" and b"" too).

(and if it can't be consider as a bug, the fix should probably be 
reverted from 3.1)

From python-checkins at python.org  Tue Jul 13 21:17:08 2010
From: python-checkins at python.org (stefan.krah)
Date: Tue, 13 Jul 2010 21:17:08 +0200 (CEST)
Subject: [Python-checkins] r82853 - in python/branches/release27-maint:
	Lib/test/test_posix.py Modules/posixmodule.c
Message-ID: <20100713191708.95D38EEA42@mail.python.org>

Author: stefan.krah
Date: Tue Jul 13 21:17:08 2010
New Revision: 82853

Log:
Issue #9185: On Solaris and OpenBSD, posix_getcwd() could loop indefinitely
if the path length exceeded PATH_MAX.



Modified:
   python/branches/release27-maint/Lib/test/test_posix.py
   python/branches/release27-maint/Modules/posixmodule.c

Modified: python/branches/release27-maint/Lib/test/test_posix.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_posix.py	(original)
+++ python/branches/release27-maint/Lib/test/test_posix.py	Tue Jul 13 21:17:08 2010
@@ -10,6 +10,7 @@
 import os
 import pwd
 import shutil
+import sys
 import unittest
 import warnings
 
@@ -345,8 +346,13 @@
                     os.chdir(dirname)
                     try:
                         os.getcwd()
-                        if current_path_length < 1027:
+                        if current_path_length < 4099:
                             _create_and_do_getcwd(dirname, current_path_length + len(dirname) + 1)
+                    except OSError as e:
+                        expected_errno = errno.ENAMETOOLONG
+                        if 'sunos' in sys.platform or 'openbsd' in sys.platform:
+                            expected_errno = errno.ERANGE # Issue 9185
+                        self.assertEqual(e.errno, expected_errno)
                     finally:
                         os.chdir('..')
                         os.rmdir(dirname)

Modified: python/branches/release27-maint/Modules/posixmodule.c
==============================================================================
--- python/branches/release27-maint/Modules/posixmodule.c	(original)
+++ python/branches/release27-maint/Modules/posixmodule.c	Tue Jul 13 21:17:08 2010
@@ -1957,6 +1957,28 @@
 "getcwd() -> path\n\n\
 Return a string representing the current working directory.");
 
+#if (defined(__sun) && defined(__SVR4)) || defined(__OpenBSD__)
+/* Issue 9185: getcwd() returns NULL/ERANGE indefinitely. */
+static PyObject *
+posix_getcwd(PyObject *self, PyObject *noargs)
+{
+    char buf[PATH_MAX+2];
+    char *res;
+
+    Py_BEGIN_ALLOW_THREADS
+#if defined(PYOS_OS2) && defined(PYCC_GCC)
+        res = _getcwd2(buf, sizeof buf);
+#else
+        res = getcwd(buf, sizeof buf);
+#endif
+    Py_END_ALLOW_THREADS
+
+    if (res == NULL)
+        return posix_error();
+
+    return PyString_FromString(buf);
+}
+#else
 static PyObject *
 posix_getcwd(PyObject *self, PyObject *noargs)
 {
@@ -1993,6 +2015,7 @@
 
     return dynamic_return;
 }
+#endif /* getcwd() NULL/ERANGE workaround. */
 
 #ifdef Py_USING_UNICODE
 PyDoc_STRVAR(posix_getcwdu__doc__,

From python-checkins at python.org  Tue Jul 13 21:40:00 2010
From: python-checkins at python.org (stefan.krah)
Date: Tue, 13 Jul 2010 21:40:00 +0200 (CEST)
Subject: [Python-checkins] r82854 -
	python/branches/release27-maint/Modules/posixmodule.c
Message-ID: <20100713194000.E48A6EE984@mail.python.org>

Author: stefan.krah
Date: Tue Jul 13 21:40:00 2010
New Revision: 82854

Log:
Remove PYOS_OS2 special cases from the Solaris/OpenBSD section.



Modified:
   python/branches/release27-maint/Modules/posixmodule.c

Modified: python/branches/release27-maint/Modules/posixmodule.c
==============================================================================
--- python/branches/release27-maint/Modules/posixmodule.c	(original)
+++ python/branches/release27-maint/Modules/posixmodule.c	Tue Jul 13 21:40:00 2010
@@ -1966,11 +1966,7 @@
     char *res;
 
     Py_BEGIN_ALLOW_THREADS
-#if defined(PYOS_OS2) && defined(PYCC_GCC)
-        res = _getcwd2(buf, sizeof buf);
-#else
-        res = getcwd(buf, sizeof buf);
-#endif
+    res = getcwd(buf, sizeof buf);
     Py_END_ALLOW_THREADS
 
     if (res == NULL)

From brett at python.org  Tue Jul 13 22:25:29 2010
From: brett at python.org (Brett Cannon)
Date: Tue, 13 Jul 2010 13:25:29 -0700
Subject: [Python-checkins] r82850 -
	python/branches/py3k/Lib/test/support.py
In-Reply-To: <20100713145017.13683EEA30@mail.python.org>
References: <20100713145017.13683EEA30@mail.python.org>
Message-ID: <AANLkTimWvMkbUsuQGEKGJkR9rotsLUOT-Aa7StxVvejV@mail.gmail.com>

Can I ask why you did this?

And just for historical background, the reason this was originally 0 instead
of None is that before absolute imports an entry in None was used to force
an absolute import instead of a relative import in a package. This doesn't
come into play in Python 3, though.

On Tue, Jul 13, 2010 at 07:50, alexander.belopolsky <
python-checkins at python.org> wrote:

> Author: alexander.belopolsky
> Date: Tue Jul 13 16:50:16 2010
> New Revision: 82850
>
> Log:
> Set sys.modules[name] to None instead of 0 to block module import.
>
> Modified:
>   python/branches/py3k/Lib/test/support.py
>
> Modified: python/branches/py3k/Lib/test/support.py
>
> ==============================================================================
> --- python/branches/py3k/Lib/test/support.py    (original)
> +++ python/branches/py3k/Lib/test/support.py    Tue Jul 13 16:50:16 2010
> @@ -109,7 +109,7 @@
>         orig_modules[name] = sys.modules[name]
>     except KeyError:
>         saved = False
> -    sys.modules[name] = 0
> +    sys.modules[name] = None
>     return saved
>
>
> _______________________________________________
> Python-checkins mailing list
> Python-checkins at python.org
> http://mail.python.org/mailman/listinfo/python-checkins
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-checkins/attachments/20100713/fd76faad/attachment.html>

From alexander.belopolsky at gmail.com  Tue Jul 13 22:39:46 2010
From: alexander.belopolsky at gmail.com (Alexander Belopolsky)
Date: Tue, 13 Jul 2010 16:39:46 -0400
Subject: [Python-checkins] r82850 -
	python/branches/py3k/Lib/test/support.py
In-Reply-To: <AANLkTimWvMkbUsuQGEKGJkR9rotsLUOT-Aa7StxVvejV@mail.gmail.com>
References: <20100713145017.13683EEA30@mail.python.org>
	<AANLkTimWvMkbUsuQGEKGJkR9rotsLUOT-Aa7StxVvejV@mail.gmail.com>
Message-ID: <AANLkTimzpYXQHQUnAkOYGammE1izJ2HB-8C5wLIASTpi@mail.gmail.com>

On Tue, Jul 13, 2010 at 4:25 PM, Brett Cannon <brett at python.org> wrote:
> Can I ask why you did this?

0 does not block import in py3k:

>>> import sys
>>> sys.modules['time'] = 0
>>> import time
>>> time
0

I ran into this problem while working on datetime.py tests.  I asked
on python-dev and Nick said it was an oversight.

http://mail.python.org/pipermail/python-dev/2010-July/101883.html


> And just for historical background, the reason this was originally 0 instead
> of None is that before absolute imports an entry in None was used to force
> an absolute import instead of a relative import in a package. This doesn't
> come into play in Python 3, though.

I was not planning to backport to 2.x.  Is there a problem with this
change in 3.x?

From brett at python.org  Tue Jul 13 22:55:06 2010
From: brett at python.org (Brett Cannon)
Date: Tue, 13 Jul 2010 13:55:06 -0700
Subject: [Python-checkins] r82850 -
	python/branches/py3k/Lib/test/support.py
In-Reply-To: <AANLkTimzpYXQHQUnAkOYGammE1izJ2HB-8C5wLIASTpi@mail.gmail.com>
References: <20100713145017.13683EEA30@mail.python.org>
	<AANLkTimWvMkbUsuQGEKGJkR9rotsLUOT-Aa7StxVvejV@mail.gmail.com> 
	<AANLkTimzpYXQHQUnAkOYGammE1izJ2HB-8C5wLIASTpi@mail.gmail.com>
Message-ID: <AANLkTilc57W_73nUC3TgX_XnJhTKyPnAXw29RrEEKCvj@mail.gmail.com>

On Tue, Jul 13, 2010 at 13:39, Alexander Belopolsky <
alexander.belopolsky at gmail.com> wrote:

> On Tue, Jul 13, 2010 at 4:25 PM, Brett Cannon <brett at python.org> wrote:
> > Can I ask why you did this?
>
> 0 does not block import in py3k:
>
> >>> import sys
> >>> sys.modules['time'] = 0
> >>> import time
> >>> time
> 0
>
> I ran into this problem while working on datetime.py tests.  I asked
> on python-dev and Nick said it was an oversight.
>
> http://mail.python.org/pipermail/python-dev/2010-July/101883.html
>
>
> > And just for historical background, the reason this was originally 0
> instead
> > of None is that before absolute imports an entry in None was used to
> force
> > an absolute import instead of a relative import in a package. This
> doesn't
> > come into play in Python 3, though.
>
> I was not planning to backport to 2.x.  Is there a problem with this
> change in 3.x?
>

I replied on python-dev to the related email to this change, but the short
answer is: no. I was just curious as to why you changed it.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-checkins/attachments/20100713/b075f3b1/attachment.html>

From python-checkins at python.org  Tue Jul 13 23:27:38 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Tue, 13 Jul 2010 23:27:38 +0200 (CEST)
Subject: [Python-checkins] r82855 - in sandbox/trunk/2to3/lib2to3:
	fixes/fix_itertools_imports.py tests/test_fixers.py
Message-ID: <20100713212738.6A56AEEA89@mail.python.org>

Author: benjamin.peterson
Date: Tue Jul 13 23:27:38 2010
New Revision: 82855

Log:
remove more extraneous commas #9245

Modified:
   sandbox/trunk/2to3/lib2to3/fixes/fix_itertools_imports.py
   sandbox/trunk/2to3/lib2to3/tests/test_fixers.py

Modified: sandbox/trunk/2to3/lib2to3/fixes/fix_itertools_imports.py
==============================================================================
--- sandbox/trunk/2to3/lib2to3/fixes/fix_itertools_imports.py	(original)
+++ sandbox/trunk/2to3/lib2to3/fixes/fix_itertools_imports.py	Tue Jul 13 23:27:38 2010
@@ -43,8 +43,8 @@
             else:
                 remove_comma ^= True
 
-        if children[-1].type == token.COMMA:
-            children[-1].remove()
+        while children and children[-1].type == token.COMMA:
+            children.pop().remove()
 
         # If there are no imports left, just get rid of the entire statement
         if (not (imports.children or getattr(imports, 'value', None)) or

Modified: sandbox/trunk/2to3/lib2to3/tests/test_fixers.py
==============================================================================
--- sandbox/trunk/2to3/lib2to3/tests/test_fixers.py	(original)
+++ sandbox/trunk/2to3/lib2to3/tests/test_fixers.py	Tue Jul 13 23:27:38 2010
@@ -3623,6 +3623,10 @@
         a = "from itertools import bar, foo"
         self.check(b, a)
 
+        b = "from itertools import chain, imap, izip"
+        a = "from itertools import chain"
+        self.check(b, a)
+
     def test_comments(self):
         b = "#foo\nfrom itertools import imap, izip"
         a = "#foo\n"

From g.brandl at gmx.net  Tue Jul 13 23:31:21 2010
From: g.brandl at gmx.net (Georg Brandl)
Date: Tue, 13 Jul 2010 23:31:21 +0200
Subject: [Python-checkins] r82848 -
	python/branches/py3k/Doc/library/ast.rst
In-Reply-To: <4C3C9FC3.6050604@gmail.com>
References: <20100713063810.B0E67EEB8F@mail.python.org>
	<4C3C9FC3.6050604@gmail.com>
Message-ID: <i1ilt5$qrq$1@dough.gmane.org>

Am 13.07.2010 19:17, schrieb Ezio Melotti:
>   On 13/07/2010 9.38, georg.brandl wrote:
>> Author: georg.brandl
>> Date: Tue Jul 13 08:38:10 2010
>> New Revision: 82848
>>
>> Log:
>> Add bytes in literal_eval doc.
>>
>> Modified:
>>     python/branches/py3k/Doc/library/ast.rst
>>
>> Modified: python/branches/py3k/Doc/library/ast.rst
>> ==============================================================================
>> --- python/branches/py3k/Doc/library/ast.rst	(original)
>> +++ python/branches/py3k/Doc/library/ast.rst	Tue Jul 13 08:38:10 2010
>> @@ -119,14 +119,14 @@
>>
>>      Safely evaluate an expression node or a string containing a Python
>>      expression.  The string or node provided may only consist of the following
>> -   Python literal structures: strings, numbers, tuples, lists, dicts, sets,
>> -   booleans, and ``None``.
>> +   Python literal structures: strings, bytes, numbers, tuples, lists, dicts,
>> +   sets, booleans, and ``None``.
>>
>>      This can be used for safely evaluating strings containing Python expressions
>>      from untrusted sources without the need to parse the values oneself.
>>
>>      .. versionchanged:: 3.2
>> -      Now allows set literals.
>> +      Now allows bytes and set literals.
>>
>>
>>   .. function:: get_docstring(node, clean=True)
>> _______________________________________________
>> Python-checkins mailing list
>> Python-checkins at python.org
>> http://mail.python.org/mailman/listinfo/python-checkins
>>
> This has been fixed in 3.1 too in r82829. As I said on IRC I consider 
> this (and the missing set literals on 2.7/3.1) as a bug, so I would:
>    1) remove the versionchanged from 3.2;
>    2) add bytes literal in the doc for 3.1 (the fix has been backported 
> already);
>    3) backport and document set literals to 3.1 and 2.7 too;
>    4) possibly update the doc for byte literals in 2.7 too (it just says 
> 'strings' but it supports u"" and b"" too).
> 
> (and if it can't be consider as a bug, the fix should probably be 
> reverted from 3.1)

And as I said on IRC :), I don't consider this a bug fix, so I won't do anything
about backporting or reverting in 3.1 or 2.7 -- this is up to Benjamin.

Georg


From benjamin at python.org  Tue Jul 13 23:52:38 2010
From: benjamin at python.org (Benjamin Peterson)
Date: Tue, 13 Jul 2010 16:52:38 -0500
Subject: [Python-checkins] r82848 -
	python/branches/py3k/Doc/library/ast.rst
In-Reply-To: <i1ilt5$qrq$1@dough.gmane.org>
References: <20100713063810.B0E67EEB8F@mail.python.org>
	<4C3C9FC3.6050604@gmail.com> <i1ilt5$qrq$1@dough.gmane.org>
Message-ID: <AANLkTintaK0xZq2IBkDivNATvZNtahUX52jHBu52pHL2@mail.gmail.com>

2010/7/13 Georg Brandl <g.brandl at gmx.net>:
> Am 13.07.2010 19:17, schrieb Ezio Melotti:
>> ? On 13/07/2010 9.38, georg.brandl wrote:
>>> Author: georg.brandl
>>> Date: Tue Jul 13 08:38:10 2010
>>> New Revision: 82848
>>>
>>> Log:
>>> Add bytes in literal_eval doc.
>>>
>>> Modified:
>>> ? ? python/branches/py3k/Doc/library/ast.rst
>>>
>>> Modified: python/branches/py3k/Doc/library/ast.rst
>>> ==============================================================================
>>> --- python/branches/py3k/Doc/library/ast.rst (original)
>>> +++ python/branches/py3k/Doc/library/ast.rst Tue Jul 13 08:38:10 2010
>>> @@ -119,14 +119,14 @@
>>>
>>> ? ? ?Safely evaluate an expression node or a string containing a Python
>>> ? ? ?expression. ?The string or node provided may only consist of the following
>>> - ? Python literal structures: strings, numbers, tuples, lists, dicts, sets,
>>> - ? booleans, and ``None``.
>>> + ? Python literal structures: strings, bytes, numbers, tuples, lists, dicts,
>>> + ? sets, booleans, and ``None``.
>>>
>>> ? ? ?This can be used for safely evaluating strings containing Python expressions
>>> ? ? ?from untrusted sources without the need to parse the values oneself.
>>>
>>> ? ? ?.. versionchanged:: 3.2
>>> - ? ? ?Now allows set literals.
>>> + ? ? ?Now allows bytes and set literals.
>>>
>>>
>>> ? .. function:: get_docstring(node, clean=True)
>>> _______________________________________________
>>> Python-checkins mailing list
>>> Python-checkins at python.org
>>> http://mail.python.org/mailman/listinfo/python-checkins
>>>
>> This has been fixed in 3.1 too in r82829. As I said on IRC I consider
>> this (and the missing set literals on 2.7/3.1) as a bug, so I would:
>> ? ?1) remove the versionchanged from 3.2;
>> ? ?2) add bytes literal in the doc for 3.1 (the fix has been backported
>> already);
>> ? ?3) backport and document set literals to 3.1 and 2.7 too;
>> ? ?4) possibly update the doc for byte literals in 2.7 too (it just says
>> 'strings' but it supports u"" and b"" too).
>>
>> (and if it can't be consider as a bug, the fix should probably be
>> reverted from 3.1)
>
> And as I said on IRC :), I don't consider this a bug fix, so I won't do anything
> about backporting or reverting in 3.1 or 2.7 -- this is up to Benjamin.

And I'm with Georg on this.



-- 
Regards,
Benjamin

From python-checkins at python.org  Wed Jul 14 01:04:56 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 14 Jul 2010 01:04:56 +0200 (CEST)
Subject: [Python-checkins] r82856 - in python/branches/py3k: Lib/sndhdr.py
	Lib/test/sndhdrdata Lib/test/sndhdrdata/README
	Lib/test/sndhdrdata/sndhdr.8svx Lib/test/sndhdrdata/sndhdr.aifc
	Lib/test/sndhdrdata/sndhdr.aiff Lib/test/sndhdrdata/sndhdr.au
	Lib/test/sndhdrdata/sndhdr.hcom Lib/test/sndhdrdata/sndhdr.sndt
	Lib/test/sndhdrdata/sndhdr.voc Lib/test/sndhdrdata/sndhdr.wav
	Lib/test/test_sndhdr.py Misc/ACKS Misc/NEWS
Message-ID: <20100713230456.CF23FEE999@mail.python.org>

Author: victor.stinner
Date: Wed Jul 14 01:04:56 2010
New Revision: 82856

Log:
Issue #9243: Fix sndhdr module and add unit tests, contributed by James Lee.


Added:
   python/branches/py3k/Lib/test/sndhdrdata/
   python/branches/py3k/Lib/test/sndhdrdata/README
   python/branches/py3k/Lib/test/sndhdrdata/sndhdr.8svx   (contents, props changed)
   python/branches/py3k/Lib/test/sndhdrdata/sndhdr.aifc   (contents, props changed)
   python/branches/py3k/Lib/test/sndhdrdata/sndhdr.aiff   (contents, props changed)
   python/branches/py3k/Lib/test/sndhdrdata/sndhdr.au   (contents, props changed)
   python/branches/py3k/Lib/test/sndhdrdata/sndhdr.hcom   (contents, props changed)
   python/branches/py3k/Lib/test/sndhdrdata/sndhdr.sndt   (contents, props changed)
   python/branches/py3k/Lib/test/sndhdrdata/sndhdr.voc   (contents, props changed)
   python/branches/py3k/Lib/test/sndhdrdata/sndhdr.wav   (contents, props changed)
   python/branches/py3k/Lib/test/test_sndhdr.py
Modified:
   python/branches/py3k/Lib/sndhdr.py
   python/branches/py3k/Misc/ACKS
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/sndhdr.py
==============================================================================
--- python/branches/py3k/Lib/sndhdr.py	(original)
+++ python/branches/py3k/Lib/sndhdr.py	Wed Jul 14 01:04:56 2010
@@ -57,12 +57,12 @@
 
 def test_aifc(h, f):
     import aifc
-    if h.startswith(b'FORM'):
+    if not h.startswith(b'FORM'):
         return None
     if h[8:12] == b'AIFC':
         fmt = 'aifc'
     elif h[8:12] == b'AIFF':
-        fmt = b'aiff'
+        fmt = 'aiff'
     else:
         return None
     f.seek(0)
@@ -123,7 +123,7 @@
 
 
 def test_voc(h, f):
-    if h.startswith(b'Creative Voice File\032'):
+    if not h.startswith(b'Creative Voice File\032'):
         return None
     sbseek = get_short_le(h[20:22])
     rate = 0
@@ -150,7 +150,7 @@
 
 
 def test_8svx(h, f):
-    if h.startswith(b'FORM') or h[8:12] != b'8SVX':
+    if not h.startswith(b'FORM') or h[8:12] != b'8SVX':
         return None
     # Should decode it to get #channels -- assume always 1
     return '8svx', 0, 1, 0, 8

Added: python/branches/py3k/Lib/test/sndhdrdata/README
==============================================================================
--- (empty file)
+++ python/branches/py3k/Lib/test/sndhdrdata/README	Wed Jul 14 01:04:56 2010
@@ -0,0 +1,12 @@
+Sound file samples used by Lib/test/test_sndhdr.py and generated using the
+following commands:
+
+   dd if=/dev/zero of=sndhdr.raw bs=20 count=1
+   sox -s -2 -c 2 -r 44100 sndhdr.raw sndhdr.<format>
+
+Sound file samples used by Lib/test/test_sndhdr.py and generated using the
+following commands:
+
+   dd if=/dev/zero of=sndhdr.raw bs=20 count=1
+   sox -s -2 -c 2 -r 44100 sndhdr.raw sndhdr.<format>
+

Added: python/branches/py3k/Lib/test/sndhdrdata/sndhdr.8svx
==============================================================================
Binary file. No diff available.

Added: python/branches/py3k/Lib/test/sndhdrdata/sndhdr.aifc
==============================================================================
Binary file. No diff available.

Added: python/branches/py3k/Lib/test/sndhdrdata/sndhdr.aiff
==============================================================================
Binary file. No diff available.

Added: python/branches/py3k/Lib/test/sndhdrdata/sndhdr.au
==============================================================================
Binary file. No diff available.

Added: python/branches/py3k/Lib/test/sndhdrdata/sndhdr.hcom
==============================================================================
Binary file. No diff available.

Added: python/branches/py3k/Lib/test/sndhdrdata/sndhdr.sndt
==============================================================================
Binary file. No diff available.

Added: python/branches/py3k/Lib/test/sndhdrdata/sndhdr.voc
==============================================================================
Binary file. No diff available.

Added: python/branches/py3k/Lib/test/sndhdrdata/sndhdr.wav
==============================================================================
Binary file. No diff available.

Added: python/branches/py3k/Lib/test/test_sndhdr.py
==============================================================================
--- (empty file)
+++ python/branches/py3k/Lib/test/test_sndhdr.py	Wed Jul 14 01:04:56 2010
@@ -0,0 +1,47 @@
+import sndhdr
+import unittest
+from test.support import findfile
+
+class TestFormats(unittest.TestCase):
+    def test_data(self):
+        for filename, expected in (
+            ('sndhdr.8svx', ('8svx', 0, 1, 0, 8)),
+            ('sndhdr.aifc', ('aifc', 44100, 2, 5, 16)),
+            ('sndhdr.aiff', ('aiff', 44100, 2, 5, 16)),
+            ('sndhdr.au', ('au', 44100, 2, 5.0, 16)),
+            ('sndhdr.hcom', ('hcom', 22050.0, 1, -1, 8)),
+            ('sndhdr.sndt', ('sndt', 44100, 1, 5, 8)),
+            ('sndhdr.voc', ('voc', 0, 1, -1, 8)),
+            ('sndhdr.wav', ('wav', 44100, 2, -1, 16)),
+        ):
+            filename = findfile(filename, subdir="sndhdrdata")
+            what = sndhdr.what(filename)
+            self.assertNotEqual(what, None, filename)
+            self.assertSequenceEqual(what, expected)
+
+if __name__ == '__main__':
+    unittest.main()
+
+import sndhdr
+import unittest
+from test.support import findfile
+
+class TestFormats(unittest.TestCase):
+    def test_data(self):
+        for filename, expected in (
+            ('sndhdr.8svx', ('8svx', 0, 1, 0, 8)),
+            ('sndhdr.aifc', ('aifc', 44100, 2, 5, 16)),
+            ('sndhdr.aiff', ('aiff', 44100, 2, 5, 16)),
+            ('sndhdr.au', ('au', 44100, 2, 5.0, 16)),
+            ('sndhdr.hcom', ('hcom', 22050.0, 1, -1, 8)),
+            ('sndhdr.sndt', ('sndt', 44100, 1, 5, 8)),
+            ('sndhdr.voc', ('voc', 0, 1, -1, 8)),
+            ('sndhdr.wav', ('wav', 44100, 2, -1, 16)),
+        ):
+            filename = findfile(filename, subdir="sndhdrdata")
+            what = sndhdr.what(filename)
+            self.assertNotEqual(what, None, filename)
+            self.assertSequenceEqual(what, expected)
+
+if __name__ == '__main__':
+    unittest.main()

Modified: python/branches/py3k/Misc/ACKS
==============================================================================
--- python/branches/py3k/Misc/ACKS	(original)
+++ python/branches/py3k/Misc/ACKS	Wed Jul 14 01:04:56 2010
@@ -454,6 +454,7 @@
 Simon Law
 Chris Lawrence
 Brian Leair
+James Lee
 John J. Lee
 Inyeol Lee
 Thomas Lee

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Wed Jul 14 01:04:56 2010
@@ -473,6 +473,8 @@
 Library
 -------
 
+- Issue #9243: Fix sndhdr module and add unit tests, contributed by James Lee.
+
 - ``ast.literal_eval()`` now allows byte literals.
 
 - Issue #9137: Fix issue in MutableMapping.update, which incorrectly

From python-checkins at python.org  Wed Jul 14 01:08:01 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 14 Jul 2010 01:08:01 +0200 (CEST)
Subject: [Python-checkins] r82857 -
	python/branches/py3k/Lib/test/test_sndhdr.py
Message-ID: <20100713230801.AE162EE999@mail.python.org>

Author: victor.stinner
Date: Wed Jul 14 01:08:01 2010
New Revision: 82857

Log:
Woops, test_sndhdr.py contains the same code twice: fix it


Modified:
   python/branches/py3k/Lib/test/test_sndhdr.py

Modified: python/branches/py3k/Lib/test/test_sndhdr.py
==============================================================================
--- python/branches/py3k/Lib/test/test_sndhdr.py	(original)
+++ python/branches/py3k/Lib/test/test_sndhdr.py	Wed Jul 14 01:08:01 2010
@@ -21,27 +21,3 @@
 
 if __name__ == '__main__':
     unittest.main()
-
-import sndhdr
-import unittest
-from test.support import findfile
-
-class TestFormats(unittest.TestCase):
-    def test_data(self):
-        for filename, expected in (
-            ('sndhdr.8svx', ('8svx', 0, 1, 0, 8)),
-            ('sndhdr.aifc', ('aifc', 44100, 2, 5, 16)),
-            ('sndhdr.aiff', ('aiff', 44100, 2, 5, 16)),
-            ('sndhdr.au', ('au', 44100, 2, 5.0, 16)),
-            ('sndhdr.hcom', ('hcom', 22050.0, 1, -1, 8)),
-            ('sndhdr.sndt', ('sndt', 44100, 1, 5, 8)),
-            ('sndhdr.voc', ('voc', 0, 1, -1, 8)),
-            ('sndhdr.wav', ('wav', 44100, 2, -1, 16)),
-        ):
-            filename = findfile(filename, subdir="sndhdrdata")
-            what = sndhdr.what(filename)
-            self.assertNotEqual(what, None, filename)
-            self.assertSequenceEqual(what, expected)
-
-if __name__ == '__main__':
-    unittest.main()

From python-checkins at python.org  Wed Jul 14 01:14:47 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 14 Jul 2010 01:14:47 +0200 (CEST)
Subject: [Python-checkins] r82858 -
	python/branches/release31-maint/Lib/test/support.py
Message-ID: <20100713231447.1FE76D3E9@mail.python.org>

Author: victor.stinner
Date: Wed Jul 14 01:14:46 2010
New Revision: 82858

Log:
Partial port of r78919 from trunk: add subdir argument to the findfile helper


Modified:
   python/branches/release31-maint/Lib/test/support.py

Modified: python/branches/release31-maint/Lib/test/support.py
==============================================================================
--- python/branches/release31-maint/Lib/test/support.py	(original)
+++ python/branches/release31-maint/Lib/test/support.py	Wed Jul 14 01:14:46 2010
@@ -387,12 +387,14 @@
     unlink(TESTFN)
 del fp
 
-def findfile(file, here=__file__):
+def findfile(file, here=__file__, subdir=None):
     """Try to find a file on sys.path and the working directory.  If it is not
     found the argument passed to the function is returned (this does not
     necessarily signal failure; could still be the legitimate path)."""
     if os.path.isabs(file):
         return file
+    if subdir is not None:
+        file = os.path.join(subdir, file)
     path = sys.path
     path = [os.path.dirname(here)] + path
     for dn in path:

From python-checkins at python.org  Wed Jul 14 01:19:20 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 14 Jul 2010 01:19:20 +0200 (CEST)
Subject: [Python-checkins] r82859 - in python/branches/release31-maint:
	Lib/sndhdr.py Lib/test/sndhdrdata Misc/ACKS Misc/NEWS
Message-ID: <20100713231920.D14B2EE9C5@mail.python.org>

Author: victor.stinner
Date: Wed Jul 14 01:19:20 2010
New Revision: 82859

Log:
Merged revisions 82856-82857 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82856 | victor.stinner | 2010-07-14 01:04:56 +0200 (mer., 14 juil. 2010) | 2 lines
  
  Issue #9243: Fix sndhdr module and add unit tests, contributed by James Lee.
........
  r82857 | victor.stinner | 2010-07-14 01:08:01 +0200 (mer., 14 juil. 2010) | 2 lines
  
  Woops, test_sndhdr.py contains the same code twice: fix it
........


Added:
   python/branches/release31-maint/Lib/test/sndhdrdata/
      - copied from r82857, /python/branches/py3k/Lib/test/sndhdrdata/
Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/sndhdr.py
   python/branches/release31-maint/Misc/ACKS
   python/branches/release31-maint/Misc/NEWS

Modified: python/branches/release31-maint/Lib/sndhdr.py
==============================================================================
--- python/branches/release31-maint/Lib/sndhdr.py	(original)
+++ python/branches/release31-maint/Lib/sndhdr.py	Wed Jul 14 01:19:20 2010
@@ -57,12 +57,12 @@
 
 def test_aifc(h, f):
     import aifc
-    if h.startswith(b'FORM'):
+    if not h.startswith(b'FORM'):
         return None
     if h[8:12] == b'AIFC':
         fmt = 'aifc'
     elif h[8:12] == b'AIFF':
-        fmt = b'aiff'
+        fmt = 'aiff'
     else:
         return None
     f.seek(0)
@@ -123,7 +123,7 @@
 
 
 def test_voc(h, f):
-    if h.startswith(b'Creative Voice File\032'):
+    if not h.startswith(b'Creative Voice File\032'):
         return None
     sbseek = get_short_le(h[20:22])
     rate = 0
@@ -150,7 +150,7 @@
 
 
 def test_8svx(h, f):
-    if h.startswith(b'FORM') or h[8:12] != b'8SVX':
+    if not h.startswith(b'FORM') or h[8:12] != b'8SVX':
         return None
     # Should decode it to get #channels -- assume always 1
     return '8svx', 0, 1, 0, 8

Modified: python/branches/release31-maint/Misc/ACKS
==============================================================================
--- python/branches/release31-maint/Misc/ACKS	(original)
+++ python/branches/release31-maint/Misc/ACKS	Wed Jul 14 01:19:20 2010
@@ -433,6 +433,7 @@
 Simon Law
 Chris Lawrence
 Brian Leair
+James Lee
 John J. Lee
 Inyeol Lee
 Thomas Lee

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Wed Jul 14 01:19:20 2010
@@ -78,6 +78,8 @@
 Library
 -------
 
+- Issue #9243: Fix sndhdr module and add unit tests, contributed by James Lee.
+
 - ``ast.literal_eval()`` now allows byte literals.
 
 - Issue #9137: Fix issue in MutableMapping.update, which incorrectly

From solipsis at pitrou.net  Wed Jul 14 01:24:11 2010
From: solipsis at pitrou.net (solipsis at pitrou.net)
Date: Wed, 14 Jul 2010 01:24:11 +0200 (CEST)
Subject: [Python-checkins] Daily py3k reference leaks (r82850): sum=0
Message-ID: <20100713232411.EE1A91771C@ns6635.ovh.net>

py3k results for svn r82850 (hg cset cef9a50dd6a2)
--------------------------------------------------



Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/antoine/py3k/refleaks/reflog64BywH', '-x']

From python-checkins at python.org  Wed Jul 14 01:31:11 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 14 Jul 2010 01:31:11 +0200 (CEST)
Subject: [Python-checkins] r82860 -
	python/branches/release31-maint/Lib/test/test_sndhdr.py
Message-ID: <20100713233111.7C148EEAB4@mail.python.org>

Author: victor.stinner
Date: Wed Jul 14 01:31:11 2010
New Revision: 82860

Log:
Add missing file from the previous commit (r82859): test_sndhdr.py


Added:
   python/branches/release31-maint/Lib/test/test_sndhdr.py

Added: python/branches/release31-maint/Lib/test/test_sndhdr.py
==============================================================================
--- (empty file)
+++ python/branches/release31-maint/Lib/test/test_sndhdr.py	Wed Jul 14 01:31:11 2010
@@ -0,0 +1,23 @@
+import sndhdr
+import unittest
+from test.support import findfile
+
+class TestFormats(unittest.TestCase):
+    def test_data(self):
+        for filename, expected in (
+            ('sndhdr.8svx', ('8svx', 0, 1, 0, 8)),
+            ('sndhdr.aifc', ('aifc', 44100, 2, 5, 16)),
+            ('sndhdr.aiff', ('aiff', 44100, 2, 5, 16)),
+            ('sndhdr.au', ('au', 44100, 2, 5.0, 16)),
+            ('sndhdr.hcom', ('hcom', 22050.0, 1, -1, 8)),
+            ('sndhdr.sndt', ('sndt', 44100, 1, 5, 8)),
+            ('sndhdr.voc', ('voc', 0, 1, -1, 8)),
+            ('sndhdr.wav', ('wav', 44100, 2, -1, 16)),
+        ):
+            filename = findfile(filename, subdir="sndhdrdata")
+            what = sndhdr.what(filename)
+            self.assertNotEqual(what, None, filename)
+            self.assertSequenceEqual(what, expected)
+
+if __name__ == '__main__':
+    unittest.main()

From python-checkins at python.org  Wed Jul 14 01:47:44 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 14 Jul 2010 01:47:44 +0200 (CEST)
Subject: [Python-checkins] r82861 -
	python/branches/import_unicode/Modules/main.c
Message-ID: <20100713234744.404C0EEA08@mail.python.org>

Author: victor.stinner
Date: Wed Jul 14 01:47:44 2010
New Revision: 82861

Log:
Use PyUnicode_EncodeFSDefault() in Py_Main()

To support surrogates in the command line.

Modified:
   python/branches/import_unicode/Modules/main.c

Modified: python/branches/import_unicode/Modules/main.c
==============================================================================
--- python/branches/import_unicode/Modules/main.c	(original)
+++ python/branches/import_unicode/Modules/main.c	Wed Jul 14 01:47:44 2010
@@ -254,6 +254,31 @@
     }
 }
 
+static int 
+run_command(wchar_t *command, PyCompilerFlags *cf)
+{
+    int sts;
+    char *commandStr;
+    PyObject *commandObj = NULL;
+    PyObject *command_bytes = NULL;
+
+    commandObj = PyUnicode_FromWideChar(command, -1);
+    if (commandObj == NULL)
+        goto error;
+    command_bytes = PyUnicode_EncodeFSDefault(commandObj);
+    Py_DECREF(commandObj);
+    if (command_bytes == NULL)
+        goto error;
+    commandStr = PyBytes_AsString(command_bytes);
+    sts = PyRun_SimpleStringFlags(commandStr, cf) != 0;
+    Py_XDECREF(command_bytes);
+    return sts;
+
+error:
+    PyErr_Print();
+    return 1;
+}
+
 
 /* Main program */
 
@@ -565,22 +590,8 @@
     }
 
     if (command) {
-        char *commandStr;
-        PyObject *commandObj = PyUnicode_FromWideChar(
-            command, wcslen(command));
+        sts = run_command(command, &cf);
         free(command);
-        if (commandObj != NULL)
-            commandStr = _PyUnicode_AsString(commandObj);
-        else
-            commandStr = NULL;
-        if (commandStr != NULL) {
-            sts = PyRun_SimpleStringFlags(commandStr, &cf) != 0;
-            Py_DECREF(commandObj);
-        }
-        else {
-            PyErr_Print();
-            sts = 1;
-        }
     } else if (module) {
         sts = RunModule(module, 1);
     }
@@ -639,12 +650,15 @@
 
         if (sts==-1) {
             PyObject *filenameObj = NULL;
+            PyObject *filename_bytes = NULL;
             char *p_cfilename = "<stdin>";
             if (filename) {
                 filenameObj = PyUnicode_FromWideChar(
                     filename, wcslen(filename));
-                if (filenameObj != NULL)
-                    p_cfilename = _PyUnicode_AsString(filenameObj);
+                if (filenameObj != NULL) {
+                    filename_bytes = PyUnicode_EncodeFSDefault(filenameObj);
+                    p_cfilename = PyBytes_AsString(filename_bytes);
+                }
                 else
                     p_cfilename = "<decoding error>";
             }
@@ -658,6 +672,7 @@
                     p_cfilename,
                     filename != NULL, &cf) != 0;
             }
+            Py_XDECREF(filename_bytes);
             Py_XDECREF(filenameObj);
         }
 

From python-checkins at python.org  Wed Jul 14 01:47:47 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 14 Jul 2010 01:47:47 +0200 (CEST)
Subject: [Python-checkins] r82862 -
	python/branches/import_unicode/Lib/test/test_xml_etree.py
Message-ID: <20100713234747.5440BEEA89@mail.python.org>

Author: victor.stinner
Date: Wed Jul 14 01:47:47 2010
New Revision: 82862

Log:
Skip test_xml_etree if the filename is not encodable to utf-8

Modified:
   python/branches/import_unicode/Lib/test/test_xml_etree.py

Modified: python/branches/import_unicode/Lib/test/test_xml_etree.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_xml_etree.py	(original)
+++ python/branches/import_unicode/Lib/test/test_xml_etree.py	Wed Jul 14 01:47:47 2010
@@ -13,6 +13,7 @@
 
 import sys
 import cgi
+import unittest
 
 from test import support
 from test.support import findfile
@@ -20,6 +21,11 @@
 from xml.etree import ElementTree as ET
 
 SIMPLE_XMLFILE = findfile("simple.xml", subdir="xmltestdata")
+try:
+    SIMPLE_XMLFILE.encode("utf8")
+except UnicodeEncodeError:
+    # ignore all tests if the current working directory is undecodable
+    raise unittest.SkipTest("filename is not encodable to utf8")
 SIMPLE_NS_XMLFILE = findfile("simple-ns.xml", subdir="xmltestdata")
 
 SAMPLE_XML = """\

From python-checkins at python.org  Wed Jul 14 01:47:50 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 14 Jul 2010 01:47:50 +0200 (CEST)
Subject: [Python-checkins] r82863 - in
	python/branches/import_unicode/Lib/test: test_urllib.py
	test_urllib2.py
Message-ID: <20100713234750.5193CEEAD8@mail.python.org>

Author: victor.stinner
Date: Wed Jul 14 01:47:50 2010
New Revision: 82863

Log:
Skip test_urllib and test_urllib2 if the filename is not encodable to utf-8

Modified:
   python/branches/import_unicode/Lib/test/test_urllib.py
   python/branches/import_unicode/Lib/test/test_urllib2.py

Modified: python/branches/import_unicode/Lib/test/test_urllib.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_urllib.py	(original)
+++ python/branches/import_unicode/Lib/test/test_urllib.py	Wed Jul 14 01:47:50 2010
@@ -221,8 +221,12 @@
             except: pass
 
     def constructLocalFileUrl(self, filePath):
-        return "file://%s" % urllib.request.pathname2url(
-            os.path.abspath(filePath))
+        filePath = os.path.abspath(filePath)
+        try:
+            filePath.encode("utf8")
+        except UnicodeEncodeError:
+            raise unittest.SkipTest("filePath is not encodable to utf8")
+        return "file://%s" % urllib.request.pathname2url(filePath)
 
     def createNewTempFile(self, data=b""):
         """Creates a new temporary file containing the specified data,

Modified: python/branches/import_unicode/Lib/test/test_urllib2.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_urllib2.py	(original)
+++ python/branches/import_unicode/Lib/test/test_urllib2.py	Wed Jul 14 01:47:50 2010
@@ -597,6 +597,10 @@
 
 
 def sanepathname2url(path):
+    try:
+        path.encode("utf8")
+    except UnicodeEncodeError:
+        raise unittest.SkipTest("path is not encodable to utf8")
     urlpath = urllib.request.pathname2url(path)
     if os.name == "nt" and urlpath.startswith("///"):
         urlpath = urlpath[2:]
@@ -1151,6 +1155,7 @@
         self.assertEqual(len(http_handler.requests), 1)
         self.assertFalse(http_handler.requests[0].has_header(auth_header))
 
+
 class MiscTests(unittest.TestCase):
 
     def test_build_opener(self):

From python-checkins at python.org  Wed Jul 14 01:47:52 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 14 Jul 2010 01:47:52 +0200 (CEST)
Subject: [Python-checkins] r82864 -
	python/branches/import_unicode/Modules/getpath.c
Message-ID: <20100713234752.851ECEEA94@mail.python.org>

Author: victor.stinner
Date: Wed Jul 14 01:47:52 2010
New Revision: 82864

Log:
calculate_path() uses _Py_char2wchar()

Modified:
   python/branches/import_unicode/Modules/getpath.c

Modified: python/branches/import_unicode/Modules/getpath.c
==============================================================================
--- python/branches/import_unicode/Modules/getpath.c	(original)
+++ python/branches/import_unicode/Modules/getpath.c	Wed Jul 14 01:47:52 2010
@@ -447,8 +447,7 @@
     wchar_t rtpypath[MAXPATHLEN+1];
     wchar_t *home = Py_GetPythonHome();
     char *_path = getenv("PATH");
-    wchar_t wpath[MAXPATHLEN+1];
-    wchar_t *path = NULL;
+    wchar_t *path = NULL, *free_path = NULL;
     wchar_t *prog = Py_GetProgramName();
     wchar_t argv0_path[MAXPATHLEN+1];
     wchar_t zip_path[MAXPATHLEN+1];
@@ -470,13 +469,8 @@
 #endif
 
     if (_path) {
-        /* FIXME: use _Py_char2wchar() */
-        size_t r = mbstowcs(wpath, _path, MAXPATHLEN+1);
-        path = wpath;
-        if (r == (size_t)-1 || r > MAXPATHLEN) {
-                /* Could not convert PATH, or it's too long. */
-                path = NULL;
-        }
+        path = _Py_char2wchar(_path);
+        free_path = path;
     }
 
     /* If there is no slash in the argv0 path, then we have to
@@ -537,6 +531,8 @@
         absolutize(progpath);
     wcsncpy(argv0_path, progpath, MAXPATHLEN);
     argv0_path[MAXPATHLEN] = '\0';
+    if (free_path != NULL)
+        PyMem_Free(free_path);
 
 #ifdef WITH_NEXT_FRAMEWORK
     /* On Mac OS X we have a special case if we're running from a framework.

From python-checkins at python.org  Wed Jul 14 01:47:54 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 14 Jul 2010 01:47:54 +0200 (CEST)
Subject: [Python-checkins] r82865 -
	python/branches/import_unicode/Lib/platform.py
Message-ID: <20100713234754.F2A9BEEA35@mail.python.org>

Author: victor.stinner
Date: Wed Jul 14 01:47:54 2010
New Revision: 82865

Log:
_syscmd_file() uses bytes type and subprocess module

Modified:
   python/branches/import_unicode/Lib/platform.py

Modified: python/branches/import_unicode/Lib/platform.py
==============================================================================
--- python/branches/import_unicode/Lib/platform.py	(original)
+++ python/branches/import_unicode/Lib/platform.py	Wed Jul 14 01:47:54 2010
@@ -946,7 +946,7 @@
     else:
         return output
 
-def _syscmd_file(target,default=''):
+def _syscmd_file(target, default=b''):
 
     """ Interface to the system's file command.
 
@@ -956,20 +956,22 @@
         case the command should fail.
 
     """
+    import subprocess
     if sys.platform in ('dos','win32','win16','os2'):
         # XXX Others too ?
         return default
-    target = _follow_symlinks(target).replace('"', '\\"')
-    try:
-        f = os.popen('file "%s" 2> %s' % (target, DEV_NULL))
-    except (AttributeError,os.error):
-        return default
-    output = f.read().strip()
-    rc = f.close()
-    if not output or rc:
+    target = _follow_symlinks(target)
+    proc = subprocess.Popen(
+        ('file', target),
+        stdout=subprocess.PIPE,
+        stderr=subprocess.PIPE)
+    stdout, stderr = proc.communicate()
+    if proc.returncode:
         return default
-    else:
-        return output
+    output = stdout.strip()
+    if not output:
+        output = default
+    return output
 
 ### Information about the used architecture
 
@@ -981,9 +983,9 @@
     'dos': ('','MSDOS'),
 }
 
-_architecture_split = re.compile(r'[\s,]').split
+_architecture_split = re.compile(br'[\s,]').split
 
-def architecture(executable=sys.executable,bits='',linkage=''):
+def architecture(executable=sys.executable,bits='', linkage=''):
 
     """ Queries the given executable (defaults to the Python interpreter
         binary) for various architecture information.
@@ -1017,9 +1019,9 @@
 
     # Get data from the 'file' system command
     if executable:
-        output = _syscmd_file(executable, '')
+        output = _syscmd_file(executable, b'')
     else:
-        output = ''
+        output = b''
 
     if not output and \
        executable == sys.executable:
@@ -1036,31 +1038,31 @@
     # Split the output into a list of strings omitting the filename
     fileout = _architecture_split(output)[1:]
 
-    if 'executable' not in fileout:
+    if b'executable' not in fileout:
         # Format not supported
         return bits,linkage
 
     # Bits
-    if '32-bit' in fileout:
+    if b'32-bit' in fileout:
         bits = '32bit'
-    elif 'N32' in fileout:
+    elif b'N32' in fileout:
         # On Irix only
         bits = 'n32bit'
-    elif '64-bit' in fileout:
+    elif b'64-bit' in fileout:
         bits = '64bit'
 
     # Linkage
-    if 'ELF' in fileout:
+    if b'ELF' in fileout:
         linkage = 'ELF'
-    elif 'PE' in fileout:
+    elif b'PE' in fileout:
         # E.g. Windows uses this format
-        if 'Windows' in fileout:
+        if b'Windows' in fileout:
             linkage = 'WindowsPE'
         else:
             linkage = 'PE'
-    elif 'COFF' in fileout:
+    elif b'COFF' in fileout:
         linkage = 'COFF'
-    elif 'MS-DOS' in fileout:
+    elif b'MS-DOS' in fileout:
         linkage = 'MSDOS'
     else:
         # XXX the A.OUT format also falls under this class...

From python-checkins at python.org  Wed Jul 14 01:47:57 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 14 Jul 2010 01:47:57 +0200 (CEST)
Subject: [Python-checkins] r82866 -
	python/branches/import_unicode/Lib/test/test_subprocess.py
Message-ID: <20100713234757.788DEEEACD@mail.python.org>

Author: victor.stinner
Date: Wed Jul 14 01:47:57 2010
New Revision: 82866

Log:
fix test_subprocess for non-ascii path but ascii locale

 * write shell script using errors="surrogateescape" (PEP 383)
 * skip test_bytes_program() if python is not installed

Modified:
   python/branches/import_unicode/Lib/test/test_subprocess.py

Modified: python/branches/import_unicode/Lib/test/test_subprocess.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_subprocess.py	(original)
+++ python/branches/import_unicode/Lib/test/test_subprocess.py	Wed Jul 14 01:47:57 2010
@@ -698,7 +698,7 @@
         # args is a string
         fd, fname = mkstemp()
         # reopen in text mode
-        with open(fd, "w") as fobj:
+        with open(fd, "w", errors="surrogateescape") as fobj:
             fobj.write("#!/bin/sh\n")
             fobj.write("exec '%s' -c 'import sys; sys.exit(47)'\n" %
                        sys.executable)
@@ -741,7 +741,7 @@
         # call() function with string argument on UNIX
         fd, fname = mkstemp()
         # reopen in text mode
-        with open(fd, "w") as fobj:
+        with open(fd, "w", errors="surrogateescape") as fobj:
             fobj.write("#!/bin/sh\n")
             fobj.write("exec '%s' -c 'import sys; sys.exit(47)'\n" %
                        sys.executable)
@@ -834,15 +834,15 @@
             stdout = stdout.rstrip(b'\n\r')
             self.assertEquals(stdout.decode('ascii'), repr(value))
 
+    def test_absolute_bytes_program(self):
+        exitcode = subprocess.call([os.fsencode(sys.executable), "-c", "pass"])
+        self.assertEquals(exitcode, 0)
+
+    @unittest.skipIf(sysconfig.is_python_build(), "need an installed Python")
     def test_bytes_program(self):
-        abs_program = os.fsencode(sys.executable)
         path, program = os.path.split(sys.executable)
         program = os.fsencode(program)
 
-        # absolute bytes path
-        exitcode = subprocess.call([abs_program, "-c", "pass"])
-        self.assertEquals(exitcode, 0)
-
         # bytes program, unicode PATH
         env = os.environ.copy()
         env["PATH"] = path

From python-checkins at python.org  Wed Jul 14 01:47:59 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 14 Jul 2010 01:47:59 +0200 (CEST)
Subject: [Python-checkins] r82867 -
	python/branches/import_unicode/Lib/distutils/file_util.py
Message-ID: <20100713234759.BAFE2EEADB@mail.python.org>

Author: victor.stinner
Date: Wed Jul 14 01:47:59 2010
New Revision: 82867

Log:
write_file() uses utf-8 with surrogateescape

Modified:
   python/branches/import_unicode/Lib/distutils/file_util.py

Modified: python/branches/import_unicode/Lib/distutils/file_util.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/file_util.py	(original)
+++ python/branches/import_unicode/Lib/distutils/file_util.py	Wed Jul 14 01:47:59 2010
@@ -225,7 +225,7 @@
     """Create a file with the specified name and write 'contents' (a
     sequence of strings without line terminators) to it.
     """
-    f = open(filename, "w")
+    f = open(filename, "w", encoding="utf-8", errors="surrogateescape")
     for line in contents:
         f.write(line + "\n")
     f.close()

From python-checkins at python.org  Wed Jul 14 01:48:02 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 14 Jul 2010 01:48:02 +0200 (CEST)
Subject: [Python-checkins] r82868 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100713234802.2FF0DEEAA9@mail.python.org>

Author: victor.stinner
Date: Wed Jul 14 01:48:01 2010
New Revision: 82868

Log:
Use PyUnicode_EncodeFSDefault() in case_ok()

except on Windows

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Wed Jul 14 01:48:01 2010
@@ -1938,18 +1938,18 @@
     return fd->type == PY_SOURCE || fd->type == PY_COMPILED;
 }
 
-/* case_ok(char* buf, Py_ssize_t len, Py_ssize_t namelen, char* name)
+/* case_ok(PyObject* fullpath, Py_ssize_t lendelta, Py_ssize_t namelen, const char* name)
  * The arguments here are tricky, best shown by example:
  *    /a/b/c/d/e/f/g/h/i/j/k/some_long_module_name.py\0
  *    ^                      ^                   ^    ^
- *    |--------------------- buf ---------------------|
+ *    |--------------------- fullpath ----------------|
  *    |------------------- len ------------------|
  *                           |------ name -------|
  *                           |----- namelen -----|
- * buf is the full path, but len only counts up to (& exclusive of) the
+ * fullpath is the full path, but len only counts up to (& exclusive of) the
  * extension.  name is the module name, also exclusive of extension.
  *
- * We've already done a successful stat() or fopen() on buf, so know that
+ * We've already done a successful stat() or fopen() on fullpath, so know that
  * there's some match, possibly case-insensitive.
  *
  * case_ok() is to return 1 if there's a case-sensitive match for
@@ -1983,7 +1983,7 @@
 #endif
 
 static int
-case_ok(PyObject *bufobj, Py_ssize_t lendelta, Py_ssize_t namelen, const char *name)
+case_ok(PyObject *fullpath_obj, Py_ssize_t lendelta, Py_ssize_t namelen, const char *name)
 {
 /* Pick a platform-specific implementation; the sequence of #if's here should
  * match the sequence just above.
@@ -1991,7 +1991,7 @@
 
 /* MS_WINDOWS */
 #if defined(MS_WINDOWS)
-    char *buf;
+    char *fullpath;
     WIN32_FIND_DATA data;
     HANDLE h;
 
@@ -1999,13 +1999,13 @@
         return 1;
 
     /* FIXME: use PyUnicode_AsWideChar() and FindFirstFileW() */
-    buf = _PyUnicode_AsString(bufobj);
+    fullpath = _PyUnicode_AsString(fullpath_obj);
 
-    h = FindFirstFile(buf, &data);
+    h = FindFirstFile(fullpath, &data);
     if (h == INVALID_HANDLE_VALUE) {
         PyErr_Format(PyExc_NameError,
           "Can't find file for module %.100s\n(filename %U)",
-          name, bufobj);
+          name, fullpath_obj);
         return 0;
     }
     FindClose(h);
@@ -2013,28 +2013,31 @@
 
 /* DJGPP */
 #elif defined(DJGPP)
-    char *buf;
     struct ffblk ffblk;
     int done;
+    PyObject *fullpath_bytes;
 
     if (Py_GETENV("PYTHONCASEOK") != NULL)
         return 1;
 
-    /* FIXME: use PyUnicode_EncodeFSDefault() */
-    buf = _PyUnicode_AsString(bufobj);
+    fullpath_bytes = PyUnicode_EncodeFSDefault(unicode);
+    if (fullpath_bytes == NULL)
+        return 0;
 
-    done = findfirst(buf, &ffblk, FA_ARCH|FA_RDONLY|FA_HIDDEN|FA_DIREC);
+    done = findfirst(PyBytes_AS_STRING(fullpath_bytes), 
+                     &ffblk, FA_ARCH|FA_RDONLY|FA_HIDDEN|FA_DIREC);
+    Py_DECREF(fullpath_bytes);
     if (done) {
         PyErr_Format(PyExc_NameError,
           "Can't find file for module %.100s\n(filename %U)",
-          name, bufobj);
+          name, fullpath_obj);
         return 0;
     }
     return strncmp(ffblk.ff_name, name, namelen) == 0;
 
 /* new-fangled macintosh (macosx) or Cygwin */
 #elif (defined(__MACH__) && defined(__APPLE__) || defined(__CYGWIN__)) && defined(HAVE_DIRENT_H)
-    char *buf;
+    char *fullpath;
     Py_ssize_t len;
     DIR *dirp;
     struct dirent *dp;
@@ -2044,9 +2047,11 @@
     if (Py_GETENV("PYTHONCASEOK") != NULL)
         return 1;
 
-    /* FIXME: use PyUnicode_EncodeFSDefault() */
-    buf = _PyUnicode_AsString(bufobj);
-    len = strlen(buf) - lendelta;
+    fullpath_bytes = PyUnicode_EncodeFSDefault(unicode);
+    if (fullpath_bytes == NULL)
+        return 0;
+    fullpath = PyBytes_AS_STRING(fullpath_bytes);
+    len = PyBytes_GET_SIZE(fullpath_bytes) - lendelta;
     dirlen = len - namelen - 1; /* don't want trailing SEP */
 
     /* Copy the dir component into dirname; substitute "." if empty */
@@ -2056,13 +2061,13 @@
     }
     else {
         assert(dirlen <= MAXPATHLEN);
-        memcpy(dirname, buf, dirlen);
+        memcpy(dirname, fullpath, dirlen);
         dirname[dirlen] = '\0';
     }
     /* Open the directory and search the entries for an exact match. */
     dirp = opendir(dirname);
     if (dirp) {
-        char *nameWithExt = buf + len - namelen;
+        char *nameWithExt = fullpath + len - namelen;
         while ((dp = readdir(dirp)) != NULL) {
             const int thislen =
 #ifdef _DIRENT_HAVE_D_NAMELEN
@@ -2073,16 +2078,17 @@
             if (thislen >= namelen &&
                 strcmp(dp->d_name, nameWithExt) == 0) {
                 (void)closedir(dirp);
+                Py_DECREF(fullpath_bytes);
                 return 1; /* Found */
             }
         }
         (void)closedir(dirp);
     }
+    Py_DECREF(fullpath_bytes);
     return 0 ; /* Not found */
 
 /* OS/2 */
 #elif defined(PYOS_OS2)
-    char *buf;
     HDIR hdir = 1;
     ULONG srchcnt = 1;
     FILEFINDBUF3 ffbuf;
@@ -2091,15 +2097,17 @@
     if (Py_GETENV("PYTHONCASEOK") != NULL)
         return 1;
 
-    /* FIXME: use PyUnicode_EncodeFSDefault() */
-    buf = _PyUnicode_AsString(bufobj);
+    fullpath_bytes = PyUnicode_EncodeFSDefault(unicode);
+    if (fullpath_bytes == NULL)
+        return 0;
 
-    rc = DosFindFirst(buf,
+    rc = DosFindFirst(PyBytes_AS_STRING(fullpath_bytes),
                       &hdir,
                       FILE_READONLY | FILE_HIDDEN | FILE_SYSTEM | FILE_DIRECTORY,
                       &ffbuf, sizeof(ffbuf),
                       &srchcnt,
                       FIL_STANDARD);
+    Py_DECREF(fullpath_bytes);
     if (rc != NO_ERROR)
         return 0;
     return strncmp(ffbuf.achName, name, namelen) == 0;

From python-checkins at python.org  Wed Jul 14 01:59:58 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 14 Jul 2010 01:59:58 +0200 (CEST)
Subject: [Python-checkins] r82869 -
	python/branches/import_unicode/Lib/test/test_sys.py
Message-ID: <20100713235958.CC340EBB7@mail.python.org>

Author: victor.stinner
Date: Wed Jul 14 01:59:58 2010
New Revision: 82869

Log:
update test_sys according to last changes

Modified:
   python/branches/import_unicode/Lib/test/test_sys.py

Modified: python/branches/import_unicode/Lib/test/test_sys.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_sys.py	(original)
+++ python/branches/import_unicode/Lib/test/test_sys.py	Wed Jul 14 01:59:58 2010
@@ -510,8 +510,9 @@
         p = subprocess.Popen([sys.executable, "-c", code], stderr=subprocess.PIPE)
         stdout, stderr = p.communicate()
         self.assertEqual(p.returncode, 1)
-        self.assert_(b"UnicodeEncodeError:" in stderr,
-            "%r not in %s" % (b"UniodeEncodeError:", ascii(stderr)))
+        pattern = b"SyntaxError: (unicode error) 'utf8' codec can't decode byte 0xff"
+        self.assert_(pattern in stderr,
+                     "%r not in %s" % (pattern, ascii(stderr)))
 
     def test_sys_flags(self):
         self.assertTrue(sys.flags)

From python-checkins at python.org  Wed Jul 14 02:05:00 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Wed, 14 Jul 2010 02:05:00 +0200 (CEST)
Subject: [Python-checkins] r82870 - in python/branches/release27-maint:
	Lib/test/test_support.py
Message-ID: <20100714000500.3C1D4EE9EE@mail.python.org>

Author: alexander.belopolsky
Date: Wed Jul 14 02:05:00 2010
New Revision: 82870

Log:
Merged revisions 82850 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82850 | alexander.belopolsky | 2010-07-13 10:50:16 -0400 (Tue, 13 Jul 2010) | 1 line
  
  Set sys.modules[name] to None instead of 0 to block module import.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/test/test_support.py

Modified: python/branches/release27-maint/Lib/test/test_support.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_support.py	(original)
+++ python/branches/release27-maint/Lib/test/test_support.py	Wed Jul 14 02:05:00 2010
@@ -105,7 +105,7 @@
         orig_modules[name] = sys.modules[name]
     except KeyError:
         saved = False
-    sys.modules[name] = 0
+    sys.modules[name] = None
     return saved
 
 

From ezio.melotti at gmail.com  Wed Jul 14 09:30:07 2010
From: ezio.melotti at gmail.com (Ezio Melotti)
Date: Wed, 14 Jul 2010 10:30:07 +0300
Subject: [Python-checkins] r82856 - in python/branches/py3k:
 Lib/sndhdr.py Lib/test/sndhdrdata
 Lib/test/sndhdrdata/README	Lib/test/sndhdrdata/sndhdr.8svx
 Lib/test/sndhdrdata/sndhdr.aifc	Lib/test/sndhdrdata/sndhdr.aiff
 Lib/test/sndhdrdata/sndhdr.au Lib/test/sndhdrdata/sndhdr.hcom
 Lib/test/sndhdrdata/sndhdr.sndt	Lib/test/sndhdrdata/sndhdr.voc
 Lib/test/sndhdrdata/sndhdr.wav	Lib/test/test_sndhdr.py Misc/ACKS Misc/NEWS
In-Reply-To: <20100713230456.CF23FEE999@mail.python.org>
References: <20100713230456.CF23FEE999@mail.python.org>
Message-ID: <4C3D677F.8040203@gmail.com>

  Hi Victor,

On 14/07/2010 2.04, victor.stinner wrote:
> Author: victor.stinner
> Date: Wed Jul 14 01:04:56 2010
> New Revision: 82856
>
> Log:
> Issue #9243: Fix sndhdr module and add unit tests, contributed by James Lee.
>
>
> Added:
>     python/branches/py3k/Lib/test/sndhdrdata/
>     python/branches/py3k/Lib/test/sndhdrdata/README
>     python/branches/py3k/Lib/test/sndhdrdata/sndhdr.8svx   (contents, props changed)
>     python/branches/py3k/Lib/test/sndhdrdata/sndhdr.aifc   (contents, props changed)
>     python/branches/py3k/Lib/test/sndhdrdata/sndhdr.aiff   (contents, props changed)
>     python/branches/py3k/Lib/test/sndhdrdata/sndhdr.au   (contents, props changed)
>     python/branches/py3k/Lib/test/sndhdrdata/sndhdr.hcom   (contents, props changed)
>     python/branches/py3k/Lib/test/sndhdrdata/sndhdr.sndt   (contents, props changed)
>     python/branches/py3k/Lib/test/sndhdrdata/sndhdr.voc   (contents, props changed)
>     python/branches/py3k/Lib/test/sndhdrdata/sndhdr.wav   (contents, props changed)
>     python/branches/py3k/Lib/test/test_sndhdr.py
> Modified:
>     python/branches/py3k/Lib/sndhdr.py
>     python/branches/py3k/Misc/ACKS
>     python/branches/py3k/Misc/NEWS
>
> Modified: python/branches/py3k/Lib/sndhdr.py
> ==============================================================================
> --- python/branches/py3k/Lib/sndhdr.py	(original)
> +++ python/branches/py3k/Lib/sndhdr.py	Wed Jul 14 01:04:56 2010
> @@ -57,12 +57,12 @@
>
>   def test_aifc(h, f):
>       import aifc
> -    if h.startswith(b'FORM'):
> +    if not h.startswith(b'FORM'):
>           return None
>       if h[8:12] == b'AIFC':
>           fmt = 'aifc'
>       elif h[8:12] == b'AIFF':
> -        fmt = b'aiff'
> +        fmt = 'aiff'
>       else:
>           return None
>       f.seek(0)
> @@ -123,7 +123,7 @@
>
>
>   def test_voc(h, f):
> -    if h.startswith(b'Creative Voice File\032'):
> +    if not h.startswith(b'Creative Voice File\032'):
>           return None
>       sbseek = get_short_le(h[20:22])
>       rate = 0
> @@ -150,7 +150,7 @@
>
>
>   def test_8svx(h, f):
> -    if h.startswith(b'FORM') or h[8:12] != b'8SVX':
> +    if not h.startswith(b'FORM') or h[8:12] != b'8SVX':
>           return None
>       # Should decode it to get #channels -- assume always 1
>       return '8svx', 0, 1, 0, 8
>
> Added: python/branches/py3k/Lib/test/sndhdrdata/README
> ==============================================================================
> --- (empty file)
> +++ python/branches/py3k/Lib/test/sndhdrdata/README	Wed Jul 14 01:04:56 2010
> @@ -0,0 +1,12 @@
> +Sound file samples used by Lib/test/test_sndhdr.py and generated using the
> +following commands:
> +
> +   dd if=/dev/zero of=sndhdr.raw bs=20 count=1
> +   sox -s -2 -c 2 -r 44100 sndhdr.raw sndhdr.<format>
> +
> +Sound file samples used by Lib/test/test_sndhdr.py and generated using the
> +following commands:
> +
> +   dd if=/dev/zero of=sndhdr.raw bs=20 count=1
> +   sox -s -2 -c 2 -r 44100 sndhdr.raw sndhdr.<format>
> +
>
> Added: python/branches/py3k/Lib/test/sndhdrdata/sndhdr.8svx
> ==============================================================================
> Binary file. No diff available.
>
> Added: python/branches/py3k/Lib/test/sndhdrdata/sndhdr.aifc
> ==============================================================================
> Binary file. No diff available.
>
> Added: python/branches/py3k/Lib/test/sndhdrdata/sndhdr.aiff
> ==============================================================================
> Binary file. No diff available.
>
> Added: python/branches/py3k/Lib/test/sndhdrdata/sndhdr.au
> ==============================================================================
> Binary file. No diff available.
>
> Added: python/branches/py3k/Lib/test/sndhdrdata/sndhdr.hcom
> ==============================================================================
> Binary file. No diff available.
>
> Added: python/branches/py3k/Lib/test/sndhdrdata/sndhdr.sndt
> ==============================================================================
> Binary file. No diff available.
>
> Added: python/branches/py3k/Lib/test/sndhdrdata/sndhdr.voc
> ==============================================================================
> Binary file. No diff available.
>
> Added: python/branches/py3k/Lib/test/sndhdrdata/sndhdr.wav
> ==============================================================================
> Binary file. No diff available.
>
> Added: python/branches/py3k/Lib/test/test_sndhdr.py
> ==============================================================================
> --- (empty file)
> +++ python/branches/py3k/Lib/test/test_sndhdr.py	Wed Jul 14 01:04:56 2010
> @@ -0,0 +1,47 @@
> +import sndhdr
> +import unittest
> +from test.support import findfile
> +
> +class TestFormats(unittest.TestCase):
> +    def test_data(self):
> +        for filename, expected in (
> +            ('sndhdr.8svx', ('8svx', 0, 1, 0, 8)),
> +            ('sndhdr.aifc', ('aifc', 44100, 2, 5, 16)),
> +            ('sndhdr.aiff', ('aiff', 44100, 2, 5, 16)),
> +            ('sndhdr.au', ('au', 44100, 2, 5.0, 16)),
> +            ('sndhdr.hcom', ('hcom', 22050.0, 1, -1, 8)),
> +            ('sndhdr.sndt', ('sndt', 44100, 1, 5, 8)),
> +            ('sndhdr.voc', ('voc', 0, 1, -1, 8)),
> +            ('sndhdr.wav', ('wav', 44100, 2, -1, 16)),
> +        ):
> +            filename = findfile(filename, subdir="sndhdrdata")
> +            what = sndhdr.what(filename)
> +            self.assertNotEqual(what, None, filename)
> +            self.assertSequenceEqual(what, expected)
> +
> +if __name__ == '__main__':
> +    unittest.main()
> +
> +import sndhdr
> +import unittest
> +from test.support import findfile
> +
> +class TestFormats(unittest.TestCase):
> +    def test_data(self):
> +        for filename, expected in (
> +            ('sndhdr.8svx', ('8svx', 0, 1, 0, 8)),
> +            ('sndhdr.aifc', ('aifc', 44100, 2, 5, 16)),
> +            ('sndhdr.aiff', ('aiff', 44100, 2, 5, 16)),
> +            ('sndhdr.au', ('au', 44100, 2, 5.0, 16)),
> +            ('sndhdr.hcom', ('hcom', 22050.0, 1, -1, 8)),
> +            ('sndhdr.sndt', ('sndt', 44100, 1, 5, 8)),
> +            ('sndhdr.voc', ('voc', 0, 1, -1, 8)),
> +            ('sndhdr.wav', ('wav', 44100, 2, -1, 16)),
> +        ):
> +            filename = findfile(filename, subdir="sndhdrdata")
> +            what = sndhdr.what(filename)
> +            self.assertNotEqual(what, None, filename)
> +            self.assertSequenceEqual(what, expected)
> +
> +if __name__ == '__main__':
> +    unittest.main()

Why this is here twice?

> Modified: python/branches/py3k/Misc/ACKS
> ==============================================================================
> --- python/branches/py3k/Misc/ACKS	(original)
> +++ python/branches/py3k/Misc/ACKS	Wed Jul 14 01:04:56 2010
> @@ -454,6 +454,7 @@
>   Simon Law
>   Chris Lawrence
>   Brian Leair
> +James Lee
>   John J. Lee
>   Inyeol Lee
>   Thomas Lee
>
> Modified: python/branches/py3k/Misc/NEWS
> ==============================================================================
> --- python/branches/py3k/Misc/NEWS	(original)
> +++ python/branches/py3k/Misc/NEWS	Wed Jul 14 01:04:56 2010
> @@ -473,6 +473,8 @@
>   Library
>   -------
>
> +- Issue #9243: Fix sndhdr module and add unit tests, contributed by James Lee.
> +
>   - ``ast.literal_eval()`` now allows byte literals.
>
>   - Issue #9137: Fix issue in MutableMapping.update, which incorrectly
> _______________________________________________
> Python-checkins mailing list
> Python-checkins at python.org
> http://mail.python.org/mailman/listinfo/python-checkins
>


From ezio.melotti at gmail.com  Wed Jul 14 09:31:58 2010
From: ezio.melotti at gmail.com (Ezio Melotti)
Date: Wed, 14 Jul 2010 10:31:58 +0300
Subject: [Python-checkins] r82856 - in python/branches/py3k:
 Lib/sndhdr.py Lib/test/sndhdrdata
 Lib/test/sndhdrdata/README	Lib/test/sndhdrdata/sndhdr.8svx
 Lib/test/sndhdrdata/sndhdr.aifc	Lib/test/sndhdrdata/sndhdr.aiff
 Lib/test/sndhdrdata/sndhdr.au Lib/test/sndhdrdata/sndhdr.hcom
 Lib/test/sndhdrdata/sndhdr.sndt	Lib/test/sndhdrdata/sndhdr.voc
 Lib/test/sndhdrdata/sndhdr.wav	Lib/test/test_sndhdr.py Misc/ACKS Misc/NEWS
In-Reply-To: <20100713230456.CF23FEE999@mail.python.org>
References: <20100713230456.CF23FEE999@mail.python.org>
Message-ID: <4C3D67EE.80808@gmail.com>

  Apparently you have fixed it already in the next commit, so just 
ignore my last message.

From python-checkins at python.org  Wed Jul 14 10:00:23 2010
From: python-checkins at python.org (georg.brandl)
Date: Wed, 14 Jul 2010 10:00:23 +0200 (CEST)
Subject: [Python-checkins] r82871 -
	python/branches/py3k/Doc/library/select.rst
Message-ID: <20100714080023.37E04EEAE6@mail.python.org>

Author: georg.brandl
Date: Wed Jul 14 10:00:22 2010
New Revision: 82871

Log:
#9258: fix copy-paste errors.

Modified:
   python/branches/py3k/Doc/library/select.rst

Modified: python/branches/py3k/Doc/library/select.rst
==============================================================================
--- python/branches/py3k/Doc/library/select.rst	(original)
+++ python/branches/py3k/Doc/library/select.rst	Wed Jul 14 10:00:22 2010
@@ -41,14 +41,14 @@
 
 .. function:: kqueue()
 
-   (Only supported on BSD.)  Returns a kernel queue object object; see section
+   (Only supported on BSD.)  Returns a kernel queue object; see section
    :ref:`kqueue-objects` below for the methods supported by kqueue objects.
 
 
 .. function:: kevent(ident, filter=KQ_FILTER_READ, flags=KQ_EV_ADD, fflags=0, data=0, udata=0)
 
-   (Only supported on BSD.)  Returns a kernel event object object; see section
-   :ref:`kevent-objects` below for the methods supported by kqueue objects.
+   (Only supported on BSD.)  Returns a kernel event object; see section
+   :ref:`kevent-objects` below for the methods supported by kevent objects.
 
 
 .. function:: select(rlist, wlist, xlist[, timeout])

From g.brandl at gmx.net  Wed Jul 14 10:46:40 2010
From: g.brandl at gmx.net (Georg Brandl)
Date: Wed, 14 Jul 2010 10:46:40 +0200
Subject: [Python-checkins] r82848 -
	python/branches/py3k/Doc/library/ast.rst
In-Reply-To: <AANLkTintaK0xZq2IBkDivNATvZNtahUX52jHBu52pHL2@mail.gmail.com>
References: <20100713063810.B0E67EEB8F@mail.python.org>	<4C3C9FC3.6050604@gmail.com>
	<i1ilt5$qrq$1@dough.gmane.org>
	<AANLkTintaK0xZq2IBkDivNATvZNtahUX52jHBu52pHL2@mail.gmail.com>
Message-ID: <i1jtfg$4v1$1@dough.gmane.org>

Am 13.07.2010 23:52, schrieb Benjamin Peterson:
> 2010/7/13 Georg Brandl <g.brandl at gmx.net>:
>> Am 13.07.2010 19:17, schrieb Ezio Melotti:
>>>   On 13/07/2010 9.38, georg.brandl wrote:
>>>> Author: georg.brandl
>>>> Date: Tue Jul 13 08:38:10 2010
>>>> New Revision: 82848
>>>>
>>>> Log:
>>>> Add bytes in literal_eval doc.
>>>>
>>>> Modified:
>>>>     python/branches/py3k/Doc/library/ast.rst
>>>>
>>>> Modified: python/branches/py3k/Doc/library/ast.rst
>>>> ==============================================================================
>>>> --- python/branches/py3k/Doc/library/ast.rst (original)
>>>> +++ python/branches/py3k/Doc/library/ast.rst Tue Jul 13 08:38:10 2010
>>>> @@ -119,14 +119,14 @@
>>>>
>>>>      Safely evaluate an expression node or a string containing a Python
>>>>      expression.  The string or node provided may only consist of the following
>>>> -   Python literal structures: strings, numbers, tuples, lists, dicts, sets,
>>>> -   booleans, and ``None``.
>>>> +   Python literal structures: strings, bytes, numbers, tuples, lists, dicts,
>>>> +   sets, booleans, and ``None``.
>>>>
>>>>      This can be used for safely evaluating strings containing Python expressions
>>>>      from untrusted sources without the need to parse the values oneself.
>>>>
>>>>      .. versionchanged:: 3.2
>>>> -      Now allows set literals.
>>>> +      Now allows bytes and set literals.
>>>>
>>>>
>>>>   .. function:: get_docstring(node, clean=True)
>>>> _______________________________________________
>>>> Python-checkins mailing list
>>>> Python-checkins at python.org
>>>> http://mail.python.org/mailman/listinfo/python-checkins
>>>>
>>> This has been fixed in 3.1 too in r82829. As I said on IRC I consider
>>> this (and the missing set literals on 2.7/3.1) as a bug, so I would:
>>>    1) remove the versionchanged from 3.2;
>>>    2) add bytes literal in the doc for 3.1 (the fix has been backported
>>> already);
>>>    3) backport and document set literals to 3.1 and 2.7 too;
>>>    4) possibly update the doc for byte literals in 2.7 too (it just says
>>> 'strings' but it supports u"" and b"" too).
>>>
>>> (and if it can't be consider as a bug, the fix should probably be
>>> reverted from 3.1)
>>
>> And as I said on IRC :), I don't consider this a bug fix, so I won't do anything
>> about backporting or reverting in 3.1 or 2.7 -- this is up to Benjamin.
> 
> And I'm with Georg on this.

Will you revert the bytes change in 3.1 then?

Georg


From python-checkins at python.org  Wed Jul 14 10:53:18 2010
From: python-checkins at python.org (georg.brandl)
Date: Wed, 14 Jul 2010 10:53:18 +0200 (CEST)
Subject: [Python-checkins] r82872 -
	python/branches/py3k/Doc/distutils/apiref.rst
Message-ID: <20100714085318.E3055EBCA@mail.python.org>

Author: georg.brandl
Date: Wed Jul 14 10:53:18 2010
New Revision: 82872

Log:
Remove XXX from text.

Modified:
   python/branches/py3k/Doc/distutils/apiref.rst

Modified: python/branches/py3k/Doc/distutils/apiref.rst
==============================================================================
--- python/branches/py3k/Doc/distutils/apiref.rst	(original)
+++ python/branches/py3k/Doc/distutils/apiref.rst	Wed Jul 14 10:53:18 2010
@@ -21,7 +21,9 @@
 .. function:: setup(arguments)
 
    The basic do-everything function that does most everything you could ever ask
-   for from a Distutils method. See XXXXX
+   for from a Distutils method.
+
+   .. See XXXXX
 
    The setup function takes a large number of arguments. These are laid out in the
    following table.
@@ -1743,11 +1745,11 @@
 
 .. method:: Command.run()
 
-   A command's raison d'etre: carry out the action it exists to perform, controlled
-   by the options initialized in :meth:`initialize_options`, customized by other
-   commands, the setup script, the command-line, and config files, and finalized in
-   :meth:`finalize_options`.  All terminal output and filesystem interaction should
-   be done by :meth:`run`.
+   A command's raison d'etre: carry out the action it exists to perform,
+   controlled by the options initialized in :meth:`initialize_options`,
+   customized by other commands, the setup script, the command-line, and config
+   files, and finalized in :meth:`finalize_options`.  All terminal output and
+   filesystem interaction should be done by :meth:`run`.
 
 
 .. attribute:: Command.sub_commands

From python-checkins at python.org  Wed Jul 14 10:53:37 2010
From: python-checkins at python.org (georg.brandl)
Date: Wed, 14 Jul 2010 10:53:37 +0200 (CEST)
Subject: [Python-checkins] r82873 -
	python/branches/py3k/Tools/gdb/libpython.py
Message-ID: <20100714085337.01071EECF@mail.python.org>

Author: georg.brandl
Date: Wed Jul 14 10:53:36 2010
New Revision: 82873

Log:
Remove unused code that would raise a NameError.

Modified:
   python/branches/py3k/Tools/gdb/libpython.py

Modified: python/branches/py3k/Tools/gdb/libpython.py
==============================================================================
--- python/branches/py3k/Tools/gdb/libpython.py	(original)
+++ python/branches/py3k/Tools/gdb/libpython.py	Wed Jul 14 10:53:36 2010
@@ -55,7 +55,6 @@
 
 Py_TPFLAGS_HEAPTYPE = (1L << 9)
 
-Py_TPFLAGS_INT_SUBCLASS      = (1L << 23)
 Py_TPFLAGS_LONG_SUBCLASS     = (1L << 24)
 Py_TPFLAGS_LIST_SUBCLASS     = (1L << 25)
 Py_TPFLAGS_TUPLE_SUBCLASS    = (1L << 26)
@@ -312,8 +311,6 @@
         if tp_flags & Py_TPFLAGS_HEAPTYPE:
             return HeapTypeObjectPtr
 
-        if tp_flags & Py_TPFLAGS_INT_SUBCLASS:
-            return PyIntObjectPtr
         if tp_flags & Py_TPFLAGS_LONG_SUBCLASS:
             return PyLongObjectPtr
         if tp_flags & Py_TPFLAGS_LIST_SUBCLASS:

From python-checkins at python.org  Wed Jul 14 10:54:40 2010
From: python-checkins at python.org (georg.brandl)
Date: Wed, 14 Jul 2010 10:54:40 +0200 (CEST)
Subject: [Python-checkins] r82874 -
	python/branches/py3k/Tools/gdb/libpython.py
Message-ID: <20100714085440.525D3EE4D@mail.python.org>

Author: georg.brandl
Date: Wed Jul 14 10:54:40 2010
New Revision: 82874

Log:
#9235: fix missing import of sys.

Modified:
   python/branches/py3k/Tools/gdb/libpython.py

Modified: python/branches/py3k/Tools/gdb/libpython.py
==============================================================================
--- python/branches/py3k/Tools/gdb/libpython.py	(original)
+++ python/branches/py3k/Tools/gdb/libpython.py	Wed Jul 14 10:54:40 2010
@@ -43,6 +43,7 @@
 from __future__ import with_statement
 import gdb
 import locale
+import sys
 
 # Look up the gdb.Type for some standard types:
 _type_char_ptr = gdb.lookup_type('char').pointer() # char*

From python-checkins at python.org  Wed Jul 14 10:55:03 2010
From: python-checkins at python.org (georg.brandl)
Date: Wed, 14 Jul 2010 10:55:03 +0200 (CEST)
Subject: [Python-checkins] r82875 - python/branches/release27-maint
Message-ID: <20100714085503.BE414EECF@mail.python.org>

Author: georg.brandl
Date: Wed Jul 14 10:55:03 2010
New Revision: 82875

Log:
Blocked revisions 82873 via svnmerge

........
  r82873 | georg.brandl | 2010-07-14 10:53:36 +0200 (Mi, 14 Jul 2010) | 1 line
  
  Remove unused code that would raise a NameError.
........


Modified:
   python/branches/release27-maint/   (props changed)

From python-checkins at python.org  Wed Jul 14 10:55:55 2010
From: python-checkins at python.org (georg.brandl)
Date: Wed, 14 Jul 2010 10:55:55 +0200 (CEST)
Subject: [Python-checkins] r82876 - in python/branches/release27-maint:
	Doc/distutils/apiref.rst Tools/gdb/libpython.py
Message-ID: <20100714085555.680F8EF1A@mail.python.org>

Author: georg.brandl
Date: Wed Jul 14 10:55:55 2010
New Revision: 82876

Log:
Merged revisions 82872,82874 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82872 | georg.brandl | 2010-07-14 10:53:18 +0200 (Mi, 14 Jul 2010) | 1 line
  
  Remove XXX from text.
........
  r82874 | georg.brandl | 2010-07-14 10:54:40 +0200 (Mi, 14 Jul 2010) | 1 line
  
  #9235: fix missing import of sys.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Doc/distutils/apiref.rst
   python/branches/release27-maint/Tools/gdb/libpython.py

Modified: python/branches/release27-maint/Doc/distutils/apiref.rst
==============================================================================
--- python/branches/release27-maint/Doc/distutils/apiref.rst	(original)
+++ python/branches/release27-maint/Doc/distutils/apiref.rst	Wed Jul 14 10:55:55 2010
@@ -21,7 +21,7 @@
 .. function:: setup(arguments)
 
    The basic do-everything function that does most everything you could ever ask
-   for from a Distutils method. See XXXXX
+   for from a Distutils method.
 
    The setup function takes a large number of arguments. These are laid out in the
    following table.

Modified: python/branches/release27-maint/Tools/gdb/libpython.py
==============================================================================
--- python/branches/release27-maint/Tools/gdb/libpython.py	(original)
+++ python/branches/release27-maint/Tools/gdb/libpython.py	Wed Jul 14 10:55:55 2010
@@ -41,6 +41,7 @@
 '''
 from __future__ import with_statement
 import gdb
+import sys
 
 # Look up the gdb.Type for some standard types:
 _type_char_ptr = gdb.lookup_type('char').pointer() # char*

From python-checkins at python.org  Wed Jul 14 12:06:07 2010
From: python-checkins at python.org (stefan.krah)
Date: Wed, 14 Jul 2010 12:06:07 +0200 (CEST)
Subject: [Python-checkins] r82877 -
	python/branches/release27-maint/Doc/library/subprocess.rst
Message-ID: <20100714100607.D243BF2F6@mail.python.org>

Author: stefan.krah
Date: Wed Jul 14 12:06:07 2010
New Revision: 82877

Log:
High byte is the exit status.



Modified:
   python/branches/release27-maint/Doc/library/subprocess.rst

Modified: python/branches/release27-maint/Doc/library/subprocess.rst
==============================================================================
--- python/branches/release27-maint/Doc/library/subprocess.rst	(original)
+++ python/branches/release27-maint/Doc/library/subprocess.rst	Wed Jul 14 12:06:07 2010
@@ -551,7 +551,7 @@
    pipe = os.popen("cmd", 'w')
    ...
    rc = pipe.close()
-   if rc is not None and rc % 256:
+   if rc is not None and rc >> 8:
        print "There were some errors"
    ==>
    process = Popen("cmd", 'w', shell=True, stdin=PIPE)

From python-checkins at python.org  Wed Jul 14 12:13:49 2010
From: python-checkins at python.org (stefan.krah)
Date: Wed, 14 Jul 2010 12:13:49 +0200 (CEST)
Subject: [Python-checkins] r82878 - in python/branches/release26-maint:
	Doc/library/subprocess.rst
Message-ID: <20100714101349.8425EEE9C7@mail.python.org>

Author: stefan.krah
Date: Wed Jul 14 12:13:49 2010
New Revision: 82878

Log:
Merged revisions 82877 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

........
  r82877 | stefan.krah | 2010-07-14 12:06:07 +0200 (Wed, 14 Jul 2010) | 3 lines
  
  High byte is the exit status.
........


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Doc/library/subprocess.rst

Modified: python/branches/release26-maint/Doc/library/subprocess.rst
==============================================================================
--- python/branches/release26-maint/Doc/library/subprocess.rst	(original)
+++ python/branches/release26-maint/Doc/library/subprocess.rst	Wed Jul 14 12:13:49 2010
@@ -513,7 +513,7 @@
    pipe = os.popen("cmd", 'w')
    ...
    rc = pipe.close()
-   if rc != None and rc % 256:
+   if rc is not None and rc >> 8:
        print "There were some errors"
    ==>
    process = Popen("cmd", 'w', shell=True, stdin=PIPE)

From python-checkins at python.org  Wed Jul 14 12:16:11 2010
From: python-checkins at python.org (stefan.krah)
Date: Wed, 14 Jul 2010 12:16:11 +0200 (CEST)
Subject: [Python-checkins] r82879 -
	python/branches/py3k/Doc/library/subprocess.rst
Message-ID: <20100714101611.7EF23EE9C7@mail.python.org>

Author: stefan.krah
Date: Wed Jul 14 12:16:11 2010
New Revision: 82879

Log:
High byte is the exit status.



Modified:
   python/branches/py3k/Doc/library/subprocess.rst

Modified: python/branches/py3k/Doc/library/subprocess.rst
==============================================================================
--- python/branches/py3k/Doc/library/subprocess.rst	(original)
+++ python/branches/py3k/Doc/library/subprocess.rst	Wed Jul 14 12:16:11 2010
@@ -575,7 +575,7 @@
    pipe = os.popen(cmd, 'w')
    ...
    rc = pipe.close()
-   if rc is not None and rc % 256:
+   if rc is not None and rc >> 8:
        print("There were some errors")
    ==>
    process = Popen(cmd, 'w', stdin=PIPE)

From python-checkins at python.org  Wed Jul 14 12:19:40 2010
From: python-checkins at python.org (stefan.krah)
Date: Wed, 14 Jul 2010 12:19:40 +0200 (CEST)
Subject: [Python-checkins] r82880 - in python/branches/release31-maint:
	Doc/library/subprocess.rst
Message-ID: <20100714101940.699E9EF64@mail.python.org>

Author: stefan.krah
Date: Wed Jul 14 12:19:40 2010
New Revision: 82880

Log:
Merged revisions 82879 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82879 | stefan.krah | 2010-07-14 12:16:11 +0200 (Wed, 14 Jul 2010) | 3 lines
  
  High byte is the exit status.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Doc/library/subprocess.rst

Modified: python/branches/release31-maint/Doc/library/subprocess.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/subprocess.rst	(original)
+++ python/branches/release31-maint/Doc/library/subprocess.rst	Wed Jul 14 12:19:40 2010
@@ -540,7 +540,7 @@
    pipe = os.popen(cmd, 'w')
    ...
    rc = pipe.close()
-   if  rc != None and rc % 256:
+   if rc is not None and rc >> 8:
        print("There were some errors")
    ==>
    process = Popen(cmd, 'w', stdin=PIPE)

From python-checkins at python.org  Wed Jul 14 12:21:23 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Wed, 14 Jul 2010 12:21:23 +0200 (CEST)
Subject: [Python-checkins] r82881 - in python/branches/py3k/Lib:
	test/test_urlparse.py urllib/parse.py
Message-ID: <20100714102123.33373EF64@mail.python.org>

Author: senthil.kumaran
Date: Wed Jul 14 12:21:22 2010
New Revision: 82881

Log:
Fix Issue5842 - Moving the tests out of urllib.parse module



Modified:
   python/branches/py3k/Lib/test/test_urlparse.py
   python/branches/py3k/Lib/urllib/parse.py

Modified: python/branches/py3k/Lib/test/test_urlparse.py
==============================================================================
--- python/branches/py3k/Lib/test/test_urlparse.py	(original)
+++ python/branches/py3k/Lib/test/test_urlparse.py	Wed Jul 14 12:21:22 2010
@@ -7,6 +7,7 @@
 RFC1808_BASE = "http://a/b/c/d;p?q#f"
 RFC2396_BASE = "http://a/b/c/d;p?q"
 RFC3986_BASE = 'http://a/b/c/d;p?q'
+SIMPLE_BASE  = 'http://a/b/c/d'
 
 # A list of test cases.  Each test case is a a two-tuple that contains
 # a string with the query and a dictionary with the expected result.
@@ -295,6 +296,37 @@
         #self.checkJoin(RFC3986_BASE, 'http:g','http:g') # strict parser
         self.checkJoin(RFC3986_BASE, 'http:g','http://a/b/c/g') #relaxed parser
 
+    def test_urljoins(self):
+        self.checkJoin(SIMPLE_BASE, 'g:h','g:h')
+        self.checkJoin(SIMPLE_BASE, 'http:g','http://a/b/c/g')
+        self.checkJoin(SIMPLE_BASE, 'http:','http://a/b/c/d')
+        self.checkJoin(SIMPLE_BASE, 'g','http://a/b/c/g')
+        self.checkJoin(SIMPLE_BASE, './g','http://a/b/c/g')
+        self.checkJoin(SIMPLE_BASE, 'g/','http://a/b/c/g/')
+        self.checkJoin(SIMPLE_BASE, '/g','http://a/g')
+        self.checkJoin(SIMPLE_BASE, '//g','http://g')
+        self.checkJoin(SIMPLE_BASE, '?y','http://a/b/c/d?y')
+        self.checkJoin(SIMPLE_BASE, 'g?y','http://a/b/c/g?y')
+        self.checkJoin(SIMPLE_BASE, 'g?y/./x','http://a/b/c/g?y/./x')
+        self.checkJoin(SIMPLE_BASE, '.','http://a/b/c/')
+        self.checkJoin(SIMPLE_BASE, './','http://a/b/c/')
+        self.checkJoin(SIMPLE_BASE, '..','http://a/b/')
+        self.checkJoin(SIMPLE_BASE, '../','http://a/b/')
+        self.checkJoin(SIMPLE_BASE, '../g','http://a/b/g')
+        self.checkJoin(SIMPLE_BASE, '../..','http://a/')
+        self.checkJoin(SIMPLE_BASE, '../../g','http://a/g')
+        self.checkJoin(SIMPLE_BASE, '../../../g','http://a/../g')
+        self.checkJoin(SIMPLE_BASE, './../g','http://a/b/g')
+        self.checkJoin(SIMPLE_BASE, './g/.','http://a/b/c/g/')
+        self.checkJoin(SIMPLE_BASE, '/./g','http://a/./g')
+        self.checkJoin(SIMPLE_BASE, 'g/./h','http://a/b/c/g/h')
+        self.checkJoin(SIMPLE_BASE, 'g/../h','http://a/b/c/h')
+        self.checkJoin(SIMPLE_BASE, 'http:g','http://a/b/c/g')
+        self.checkJoin(SIMPLE_BASE, 'http:','http://a/b/c/d')
+        self.checkJoin(SIMPLE_BASE, 'http:?y','http://a/b/c/d?y')
+        self.checkJoin(SIMPLE_BASE, 'http:g?y','http://a/b/c/g?y')
+        self.checkJoin(SIMPLE_BASE, 'http:g?y/./x','http://a/b/c/g?y/./x')
+
     def test_RFC2732(self):
         for url, hostname, port in [
             ('http://Test.python.org:5432/foo/', 'test.python.org', 5432),

Modified: python/branches/py3k/Lib/urllib/parse.py
==============================================================================
--- python/branches/py3k/Lib/urllib/parse.py	(original)
+++ python/branches/py3k/Lib/urllib/parse.py	Wed Jul 14 12:21:22 2010
@@ -799,67 +799,3 @@
     match = _valueprog.match(attr)
     if match: return match.group(1, 2)
     return attr, None
-
-test_input = """
-      http://a/b/c/d
-
-      g:h        = <URL:g:h>
-      http:g     = <URL:http://a/b/c/g>
-      http:      = <URL:http://a/b/c/d>
-      g          = <URL:http://a/b/c/g>
-      ./g        = <URL:http://a/b/c/g>
-      g/         = <URL:http://a/b/c/g/>
-      /g         = <URL:http://a/g>
-      //g        = <URL:http://g>
-      ?y         = <URL:http://a/b/c/d?y>
-      g?y        = <URL:http://a/b/c/g?y>
-      g?y/./x    = <URL:http://a/b/c/g?y/./x>
-      .          = <URL:http://a/b/c/>
-      ./         = <URL:http://a/b/c/>
-      ..         = <URL:http://a/b/>
-      ../        = <URL:http://a/b/>
-      ../g       = <URL:http://a/b/g>
-      ../..      = <URL:http://a/>
-      ../../g    = <URL:http://a/g>
-      ../../../g = <URL:http://a/../g>
-      ./../g     = <URL:http://a/b/g>
-      ./g/.      = <URL:http://a/b/c/g/>
-      /./g       = <URL:http://a/./g>
-      g/./h      = <URL:http://a/b/c/g/h>
-      g/../h     = <URL:http://a/b/c/h>
-      http:g     = <URL:http://a/b/c/g>
-      http:      = <URL:http://a/b/c/d>
-      http:?y         = <URL:http://a/b/c/d?y>
-      http:g?y        = <URL:http://a/b/c/g?y>
-      http:g?y/./x    = <URL:http://a/b/c/g?y/./x>
-"""
-
-def test():
-    base = ''
-    if sys.argv[1:]:
-        fn = sys.argv[1]
-        if fn == '-':
-            fp = sys.stdin
-        else:
-            fp = open(fn)
-    else:
-        from io import StringIO
-        fp = StringIO(test_input)
-    for line in fp:
-        words = line.split()
-        if not words:
-            continue
-        url = words[0]
-        parts = urlparse(url)
-        print('%-10s : %s' % (url, parts))
-        abs = urljoin(base, url)
-        if not base:
-            base = abs
-        wrapped = '<URL:%s>' % abs
-        print('%-10s = %s' % (url, wrapped))
-        if len(words) == 3 and words[1] == '=':
-            if wrapped != words[2]:
-                print('EXPECTED', words[2], '!!!!!!!!!!')
-
-if __name__ == '__main__':
-    test()

From python-checkins at python.org  Wed Jul 14 12:28:38 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Wed, 14 Jul 2010 12:28:38 +0200 (CEST)
Subject: [Python-checkins] r82882 - in python/branches/release31-maint:
	Lib/test/test_urlparse.py Lib/urllib/parse.py
Message-ID: <20100714102838.7335FEEA72@mail.python.org>

Author: senthil.kumaran
Date: Wed Jul 14 12:28:38 2010
New Revision: 82882

Log:
Merged revisions 82881 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82881 | senthil.kumaran | 2010-07-14 15:51:22 +0530 (Wed, 14 Jul 2010) | 3 lines
  
  Fix Issue5842 - Moving the tests out of urllib.parse module
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/test/test_urlparse.py
   python/branches/release31-maint/Lib/urllib/parse.py

Modified: python/branches/release31-maint/Lib/test/test_urlparse.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_urlparse.py	(original)
+++ python/branches/release31-maint/Lib/test/test_urlparse.py	Wed Jul 14 12:28:38 2010
@@ -7,6 +7,7 @@
 RFC1808_BASE = "http://a/b/c/d;p?q#f"
 RFC2396_BASE = "http://a/b/c/d;p?q"
 RFC3986_BASE = 'http://a/b/c/d;p?q'
+SIMPLE_BASE  = 'http://a/b/c/d'
 
 # A list of test cases.  Each test case is a a two-tuple that contains
 # a string with the query and a dictionary with the expected result.
@@ -292,6 +293,37 @@
         #self.checkJoin(RFC3986_BASE, 'http:g','http:g') # strict parser
         self.checkJoin(RFC3986_BASE, 'http:g','http://a/b/c/g') #relaxed parser
 
+    def test_urljoins(self):
+        self.checkJoin(SIMPLE_BASE, 'g:h','g:h')
+        self.checkJoin(SIMPLE_BASE, 'http:g','http://a/b/c/g')
+        self.checkJoin(SIMPLE_BASE, 'http:','http://a/b/c/d')
+        self.checkJoin(SIMPLE_BASE, 'g','http://a/b/c/g')
+        self.checkJoin(SIMPLE_BASE, './g','http://a/b/c/g')
+        self.checkJoin(SIMPLE_BASE, 'g/','http://a/b/c/g/')
+        self.checkJoin(SIMPLE_BASE, '/g','http://a/g')
+        self.checkJoin(SIMPLE_BASE, '//g','http://g')
+        self.checkJoin(SIMPLE_BASE, '?y','http://a/b/c/d?y')
+        self.checkJoin(SIMPLE_BASE, 'g?y','http://a/b/c/g?y')
+        self.checkJoin(SIMPLE_BASE, 'g?y/./x','http://a/b/c/g?y/./x')
+        self.checkJoin(SIMPLE_BASE, '.','http://a/b/c/')
+        self.checkJoin(SIMPLE_BASE, './','http://a/b/c/')
+        self.checkJoin(SIMPLE_BASE, '..','http://a/b/')
+        self.checkJoin(SIMPLE_BASE, '../','http://a/b/')
+        self.checkJoin(SIMPLE_BASE, '../g','http://a/b/g')
+        self.checkJoin(SIMPLE_BASE, '../..','http://a/')
+        self.checkJoin(SIMPLE_BASE, '../../g','http://a/g')
+        self.checkJoin(SIMPLE_BASE, '../../../g','http://a/../g')
+        self.checkJoin(SIMPLE_BASE, './../g','http://a/b/g')
+        self.checkJoin(SIMPLE_BASE, './g/.','http://a/b/c/g/')
+        self.checkJoin(SIMPLE_BASE, '/./g','http://a/./g')
+        self.checkJoin(SIMPLE_BASE, 'g/./h','http://a/b/c/g/h')
+        self.checkJoin(SIMPLE_BASE, 'g/../h','http://a/b/c/h')
+        self.checkJoin(SIMPLE_BASE, 'http:g','http://a/b/c/g')
+        self.checkJoin(SIMPLE_BASE, 'http:','http://a/b/c/d')
+        self.checkJoin(SIMPLE_BASE, 'http:?y','http://a/b/c/d?y')
+        self.checkJoin(SIMPLE_BASE, 'http:g?y','http://a/b/c/g?y')
+        self.checkJoin(SIMPLE_BASE, 'http:g?y/./x','http://a/b/c/g?y/./x')
+
     def test_urldefrag(self):
         for url, defrag, frag in [
             ('http://python.org#frag', 'http://python.org', 'frag'),

Modified: python/branches/release31-maint/Lib/urllib/parse.py
==============================================================================
--- python/branches/release31-maint/Lib/urllib/parse.py	(original)
+++ python/branches/release31-maint/Lib/urllib/parse.py	Wed Jul 14 12:28:38 2010
@@ -774,67 +774,3 @@
     match = _valueprog.match(attr)
     if match: return match.group(1, 2)
     return attr, None
-
-test_input = """
-      http://a/b/c/d
-
-      g:h        = <URL:g:h>
-      http:g     = <URL:http://a/b/c/g>
-      http:      = <URL:http://a/b/c/d>
-      g          = <URL:http://a/b/c/g>
-      ./g        = <URL:http://a/b/c/g>
-      g/         = <URL:http://a/b/c/g/>
-      /g         = <URL:http://a/g>
-      //g        = <URL:http://g>
-      ?y         = <URL:http://a/b/c/d?y>
-      g?y        = <URL:http://a/b/c/g?y>
-      g?y/./x    = <URL:http://a/b/c/g?y/./x>
-      .          = <URL:http://a/b/c/>
-      ./         = <URL:http://a/b/c/>
-      ..         = <URL:http://a/b/>
-      ../        = <URL:http://a/b/>
-      ../g       = <URL:http://a/b/g>
-      ../..      = <URL:http://a/>
-      ../../g    = <URL:http://a/g>
-      ../../../g = <URL:http://a/../g>
-      ./../g     = <URL:http://a/b/g>
-      ./g/.      = <URL:http://a/b/c/g/>
-      /./g       = <URL:http://a/./g>
-      g/./h      = <URL:http://a/b/c/g/h>
-      g/../h     = <URL:http://a/b/c/h>
-      http:g     = <URL:http://a/b/c/g>
-      http:      = <URL:http://a/b/c/d>
-      http:?y         = <URL:http://a/b/c/d?y>
-      http:g?y        = <URL:http://a/b/c/g?y>
-      http:g?y/./x    = <URL:http://a/b/c/g?y/./x>
-"""
-
-def test():
-    base = ''
-    if sys.argv[1:]:
-        fn = sys.argv[1]
-        if fn == '-':
-            fp = sys.stdin
-        else:
-            fp = open(fn)
-    else:
-        from io import StringIO
-        fp = StringIO(test_input)
-    for line in fp:
-        words = line.split()
-        if not words:
-            continue
-        url = words[0]
-        parts = urlparse(url)
-        print('%-10s : %s' % (url, parts))
-        abs = urljoin(base, url)
-        if not base:
-            base = abs
-        wrapped = '<URL:%s>' % abs
-        print('%-10s = %s' % (url, wrapped))
-        if len(words) == 3 and words[1] == '=':
-            if wrapped != words[2]:
-                print('EXPECTED', words[2], '!!!!!!!!!!')
-
-if __name__ == '__main__':
-    test()

From python-checkins at python.org  Wed Jul 14 12:39:35 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Wed, 14 Jul 2010 12:39:35 +0200 (CEST)
Subject: [Python-checkins] r82883 - in python/branches/release27-maint:
	Lib/test/test_urlparse.py Lib/urlparse.py
Message-ID: <20100714103935.8951CEE9C7@mail.python.org>

Author: senthil.kumaran
Date: Wed Jul 14 12:39:35 2010
New Revision: 82883

Log:
Merged revisions 82881 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82881 | senthil.kumaran | 2010-07-14 15:51:22 +0530 (Wed, 14 Jul 2010) | 3 lines
  
  Fix Issue5842 - Moving the tests out of urllib.parse module
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/test/test_urlparse.py
   python/branches/release27-maint/Lib/urlparse.py

Modified: python/branches/release27-maint/Lib/test/test_urlparse.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_urlparse.py	(original)
+++ python/branches/release27-maint/Lib/test/test_urlparse.py	Wed Jul 14 12:39:35 2010
@@ -7,6 +7,7 @@
 RFC1808_BASE = "http://a/b/c/d;p?q#f"
 RFC2396_BASE = "http://a/b/c/d;p?q"
 RFC3986_BASE = 'http://a/b/c/d;p?q'
+SIMPLE_BASE  = 'http://a/b/c/d'
 
 # A list of test cases.  Each test case is a a two-tuple that contains
 # a string with the query and a dictionary with the expected result.
@@ -294,6 +295,37 @@
         #self.checkJoin(RFC3986_BASE, 'http:g','http:g') # strict parser
         self.checkJoin(RFC3986_BASE, 'http:g','http://a/b/c/g') # relaxed parser
 
+    def test_urljoins(self):
+        self.checkJoin(SIMPLE_BASE, 'g:h','g:h')
+        self.checkJoin(SIMPLE_BASE, 'http:g','http://a/b/c/g')
+        self.checkJoin(SIMPLE_BASE, 'http:','http://a/b/c/d')
+        self.checkJoin(SIMPLE_BASE, 'g','http://a/b/c/g')
+        self.checkJoin(SIMPLE_BASE, './g','http://a/b/c/g')
+        self.checkJoin(SIMPLE_BASE, 'g/','http://a/b/c/g/')
+        self.checkJoin(SIMPLE_BASE, '/g','http://a/g')
+        self.checkJoin(SIMPLE_BASE, '//g','http://g')
+        self.checkJoin(SIMPLE_BASE, '?y','http://a/b/c/d?y')
+        self.checkJoin(SIMPLE_BASE, 'g?y','http://a/b/c/g?y')
+        self.checkJoin(SIMPLE_BASE, 'g?y/./x','http://a/b/c/g?y/./x')
+        self.checkJoin(SIMPLE_BASE, '.','http://a/b/c/')
+        self.checkJoin(SIMPLE_BASE, './','http://a/b/c/')
+        self.checkJoin(SIMPLE_BASE, '..','http://a/b/')
+        self.checkJoin(SIMPLE_BASE, '../','http://a/b/')
+        self.checkJoin(SIMPLE_BASE, '../g','http://a/b/g')
+        self.checkJoin(SIMPLE_BASE, '../..','http://a/')
+        self.checkJoin(SIMPLE_BASE, '../../g','http://a/g')
+        self.checkJoin(SIMPLE_BASE, '../../../g','http://a/../g')
+        self.checkJoin(SIMPLE_BASE, './../g','http://a/b/g')
+        self.checkJoin(SIMPLE_BASE, './g/.','http://a/b/c/g/')
+        self.checkJoin(SIMPLE_BASE, '/./g','http://a/./g')
+        self.checkJoin(SIMPLE_BASE, 'g/./h','http://a/b/c/g/h')
+        self.checkJoin(SIMPLE_BASE, 'g/../h','http://a/b/c/h')
+        self.checkJoin(SIMPLE_BASE, 'http:g','http://a/b/c/g')
+        self.checkJoin(SIMPLE_BASE, 'http:','http://a/b/c/d')
+        self.checkJoin(SIMPLE_BASE, 'http:?y','http://a/b/c/d?y')
+        self.checkJoin(SIMPLE_BASE, 'http:g?y','http://a/b/c/g?y')
+        self.checkJoin(SIMPLE_BASE, 'http:g?y/./x','http://a/b/c/g?y/./x')
+
     def test_RFC2732(self):
         for url, hostname, port in [
             ('http://Test.python.org:5432/foo/', 'test.python.org', 5432),

Modified: python/branches/release27-maint/Lib/urlparse.py
==============================================================================
--- python/branches/release27-maint/Lib/urlparse.py	(original)
+++ python/branches/release27-maint/Lib/urlparse.py	Wed Jul 14 12:39:35 2010
@@ -395,72 +395,3 @@
             r.append((name, value))
 
     return r
-
-
-test_input = """
-      http://a/b/c/d
-
-      g:h        = <URL:g:h>
-      http:g     = <URL:http://a/b/c/g>
-      http:      = <URL:http://a/b/c/d>
-      g          = <URL:http://a/b/c/g>
-      ./g        = <URL:http://a/b/c/g>
-      g/         = <URL:http://a/b/c/g/>
-      /g         = <URL:http://a/g>
-      //g        = <URL:http://g>
-      ?y         = <URL:http://a/b/c/d?y>
-      g?y        = <URL:http://a/b/c/g?y>
-      g?y/./x    = <URL:http://a/b/c/g?y/./x>
-      .          = <URL:http://a/b/c/>
-      ./         = <URL:http://a/b/c/>
-      ..         = <URL:http://a/b/>
-      ../        = <URL:http://a/b/>
-      ../g       = <URL:http://a/b/g>
-      ../..      = <URL:http://a/>
-      ../../g    = <URL:http://a/g>
-      ../../../g = <URL:http://a/../g>
-      ./../g     = <URL:http://a/b/g>
-      ./g/.      = <URL:http://a/b/c/g/>
-      /./g       = <URL:http://a/./g>
-      g/./h      = <URL:http://a/b/c/g/h>
-      g/../h     = <URL:http://a/b/c/h>
-      http:g     = <URL:http://a/b/c/g>
-      http:      = <URL:http://a/b/c/d>
-      http:?y         = <URL:http://a/b/c/d?y>
-      http:g?y        = <URL:http://a/b/c/g?y>
-      http:g?y/./x    = <URL:http://a/b/c/g?y/./x>
-"""
-
-def test():
-    import sys
-    base = ''
-    if sys.argv[1:]:
-        fn = sys.argv[1]
-        if fn == '-':
-            fp = sys.stdin
-        else:
-            fp = open(fn)
-    else:
-        try:
-            from cStringIO import StringIO
-        except ImportError:
-            from StringIO import StringIO
-        fp = StringIO(test_input)
-    for line in fp:
-        words = line.split()
-        if not words:
-            continue
-        url = words[0]
-        parts = urlparse(url)
-        print '%-10s : %s' % (url, parts)
-        abs = urljoin(base, url)
-        if not base:
-            base = abs
-        wrapped = '<URL:%s>' % abs
-        print '%-10s = %s' % (url, wrapped)
-        if len(words) == 3 and words[1] == '=':
-            if wrapped != words[2]:
-                print 'EXPECTED', words[2], '!!!!!!!!!!'
-
-if __name__ == '__main__':
-    test()

From python-checkins at python.org  Wed Jul 14 12:58:12 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Wed, 14 Jul 2010 12:58:12 +0200 (CEST)
Subject: [Python-checkins] r82884 - in python/branches/release26-maint:
	Lib/test/test_urlparse.py Lib/urlparse.py
Message-ID: <20100714105812.46F31EEA35@mail.python.org>

Author: senthil.kumaran
Date: Wed Jul 14 12:58:12 2010
New Revision: 82884

Log:
Merged revisions 82883 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

................
  r82883 | senthil.kumaran | 2010-07-14 16:09:35 +0530 (Wed, 14 Jul 2010) | 9 lines
  
  Merged revisions 82881 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k
  
  ........
    r82881 | senthil.kumaran | 2010-07-14 15:51:22 +0530 (Wed, 14 Jul 2010) | 3 lines
    
    Fix Issue5842 - Moving the tests out of urllib.parse module
  ........
................


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Lib/test/test_urlparse.py
   python/branches/release26-maint/Lib/urlparse.py

Modified: python/branches/release26-maint/Lib/test/test_urlparse.py
==============================================================================
--- python/branches/release26-maint/Lib/test/test_urlparse.py	(original)
+++ python/branches/release26-maint/Lib/test/test_urlparse.py	Wed Jul 14 12:58:12 2010
@@ -7,6 +7,7 @@
 RFC1808_BASE = "http://a/b/c/d;p?q#f"
 RFC2396_BASE = "http://a/b/c/d;p?q"
 RFC3986_BASE = 'http://a/b/c/d;p?q'
+SIMPLE_BASE  = 'http://a/b/c/d'
 
 # A list of test cases.  Each test case is a a two-tuple that contains
 # a string with the query and a dictionary with the expected result.
@@ -291,6 +292,37 @@
         #self.checkJoin(RFC3986_BASE, 'http:g','http:g') # strict parser
         self.checkJoin(RFC3986_BASE, 'http:g','http://a/b/c/g') #relaxed parser
 
+    def test_urljoins(self):
+        self.checkJoin(SIMPLE_BASE, 'g:h','g:h')
+        self.checkJoin(SIMPLE_BASE, 'http:g','http://a/b/c/g')
+        self.checkJoin(SIMPLE_BASE, 'http:','http://a/b/c/d')
+        self.checkJoin(SIMPLE_BASE, 'g','http://a/b/c/g')
+        self.checkJoin(SIMPLE_BASE, './g','http://a/b/c/g')
+        self.checkJoin(SIMPLE_BASE, 'g/','http://a/b/c/g/')
+        self.checkJoin(SIMPLE_BASE, '/g','http://a/g')
+        self.checkJoin(SIMPLE_BASE, '//g','http://g')
+        self.checkJoin(SIMPLE_BASE, '?y','http://a/b/c/d?y')
+        self.checkJoin(SIMPLE_BASE, 'g?y','http://a/b/c/g?y')
+        self.checkJoin(SIMPLE_BASE, 'g?y/./x','http://a/b/c/g?y/./x')
+        self.checkJoin(SIMPLE_BASE, '.','http://a/b/c/')
+        self.checkJoin(SIMPLE_BASE, './','http://a/b/c/')
+        self.checkJoin(SIMPLE_BASE, '..','http://a/b/')
+        self.checkJoin(SIMPLE_BASE, '../','http://a/b/')
+        self.checkJoin(SIMPLE_BASE, '../g','http://a/b/g')
+        self.checkJoin(SIMPLE_BASE, '../..','http://a/')
+        self.checkJoin(SIMPLE_BASE, '../../g','http://a/g')
+        self.checkJoin(SIMPLE_BASE, '../../../g','http://a/../g')
+        self.checkJoin(SIMPLE_BASE, './../g','http://a/b/g')
+        self.checkJoin(SIMPLE_BASE, './g/.','http://a/b/c/g/')
+        self.checkJoin(SIMPLE_BASE, '/./g','http://a/./g')
+        self.checkJoin(SIMPLE_BASE, 'g/./h','http://a/b/c/g/h')
+        self.checkJoin(SIMPLE_BASE, 'g/../h','http://a/b/c/h')
+        self.checkJoin(SIMPLE_BASE, 'http:g','http://a/b/c/g')
+        self.checkJoin(SIMPLE_BASE, 'http:','http://a/b/c/d')
+        self.checkJoin(SIMPLE_BASE, 'http:?y','http://a/b/c/d?y')
+        self.checkJoin(SIMPLE_BASE, 'http:g?y','http://a/b/c/g?y')
+        self.checkJoin(SIMPLE_BASE, 'http:g?y/./x','http://a/b/c/g?y/./x')
+
     def test_urldefrag(self):
         for url, defrag, frag in [
             ('http://python.org#frag', 'http://python.org', 'frag'),

Modified: python/branches/release26-maint/Lib/urlparse.py
==============================================================================
--- python/branches/release26-maint/Lib/urlparse.py	(original)
+++ python/branches/release26-maint/Lib/urlparse.py	Wed Jul 14 12:58:12 2010
@@ -380,72 +380,3 @@
             r.append((name, value))
 
     return r
-
-
-test_input = """
-      http://a/b/c/d
-
-      g:h        = <URL:g:h>
-      http:g     = <URL:http://a/b/c/g>
-      http:      = <URL:http://a/b/c/d>
-      g          = <URL:http://a/b/c/g>
-      ./g        = <URL:http://a/b/c/g>
-      g/         = <URL:http://a/b/c/g/>
-      /g         = <URL:http://a/g>
-      //g        = <URL:http://g>
-      ?y         = <URL:http://a/b/c/d?y>
-      g?y        = <URL:http://a/b/c/g?y>
-      g?y/./x    = <URL:http://a/b/c/g?y/./x>
-      .          = <URL:http://a/b/c/>
-      ./         = <URL:http://a/b/c/>
-      ..         = <URL:http://a/b/>
-      ../        = <URL:http://a/b/>
-      ../g       = <URL:http://a/b/g>
-      ../..      = <URL:http://a/>
-      ../../g    = <URL:http://a/g>
-      ../../../g = <URL:http://a/../g>
-      ./../g     = <URL:http://a/b/g>
-      ./g/.      = <URL:http://a/b/c/g/>
-      /./g       = <URL:http://a/./g>
-      g/./h      = <URL:http://a/b/c/g/h>
-      g/../h     = <URL:http://a/b/c/h>
-      http:g     = <URL:http://a/b/c/g>
-      http:      = <URL:http://a/b/c/d>
-      http:?y         = <URL:http://a/b/c/d?y>
-      http:g?y        = <URL:http://a/b/c/g?y>
-      http:g?y/./x    = <URL:http://a/b/c/g?y/./x>
-"""
-
-def test():
-    import sys
-    base = ''
-    if sys.argv[1:]:
-        fn = sys.argv[1]
-        if fn == '-':
-            fp = sys.stdin
-        else:
-            fp = open(fn)
-    else:
-        try:
-            from cStringIO import StringIO
-        except ImportError:
-            from StringIO import StringIO
-        fp = StringIO(test_input)
-    for line in fp:
-        words = line.split()
-        if not words:
-            continue
-        url = words[0]
-        parts = urlparse(url)
-        print '%-10s : %s' % (url, parts)
-        abs = urljoin(base, url)
-        if not base:
-            base = abs
-        wrapped = '<URL:%s>' % abs
-        print '%-10s = %s' % (url, wrapped)
-        if len(words) == 3 and words[1] == '=':
-            if wrapped != words[2]:
-                print 'EXPECTED', words[2], '!!!!!!!!!!'
-
-if __name__ == '__main__':
-    test()

From python-checkins at python.org  Wed Jul 14 13:52:39 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 14 Jul 2010 13:52:39 +0200 (CEST)
Subject: [Python-checkins] r82885 - in python/branches/py3k:
	Lib/test/test_threaded_import.py Misc/NEWS
Message-ID: <20100714115239.06B15EEB01@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 14 13:52:38 2010
New Revision: 82885

Log:
Issue #9251: test_threaded_import didn't fail when run through regrtest
if the import lock was disabled.



Modified:
   python/branches/py3k/Lib/test/test_threaded_import.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/test/test_threaded_import.py
==============================================================================
--- python/branches/py3k/Lib/test/test_threaded_import.py	(original)
+++ python/branches/py3k/Lib/test/test_threaded_import.py	Wed Jul 14 13:52:38 2010
@@ -5,72 +5,67 @@
 # complains several times about module random having no attribute
 # randrange, and then Python hangs.
 
+import imp
+import sys
 import unittest
-from test.support import verbose, TestFailed, import_module
+from test.support import verbose, TestFailed, import_module, run_unittest
 thread = import_module('_thread')
 
-critical_section = thread.allocate_lock()
-done = thread.allocate_lock()
+def task(N, done, done_tasks, errors):
+    try:
+        import random
+        # This will fail if random is not completely initialized
+        x = random.randrange(1, 3)
+    except Exception as e:
+        errors.append(e.with_traceback(None))
+    finally:
+        done_tasks.append(thread.get_ident())
+        finished = len(done_tasks) == N
+        if finished:
+            done.release()
 
-def task():
-    global N, critical_section, done
-    import random
-    x = random.randrange(1, 3)
-    critical_section.acquire()
-    N -= 1
-    # Must release critical_section before releasing done, else the main
-    # thread can exit and set critical_section to None as part of global
-    # teardown; then critical_section.release() raises AttributeError.
-    finished = N == 0
-    critical_section.release()
-    if finished:
-        done.release()
 
-def test_import_hangers():
-    import sys
-    if verbose:
-        print("testing import hangers ...", end=' ')
+class ThreadedImportTests(unittest.TestCase):
 
-    import test.threaded_import_hangers
-    try:
-        if test.threaded_import_hangers.errors:
-            raise TestFailed(test.threaded_import_hangers.errors)
-        elif verbose:
-            print("OK.")
-    finally:
+    def test_parallel_module_init(self):
+        if imp.lock_held():
+            # This triggers on, e.g., from test import autotest.
+            raise unittest.SkipTest("can't run when import lock is held")
+
+        done = thread.allocate_lock()
+        done.acquire()
+        for N in (20, 50) * 3:
+            if verbose:
+                print("Trying", N, "threads ...", end=' ')
+            # Make sure that random gets reimported freshly
+            try:
+                del sys.modules['random']
+            except KeyError:
+                pass
+            errors = []
+            done_tasks = []
+            for i in range(N):
+                thread.start_new_thread(task, (N, done, done_tasks, errors,))
+            done.acquire()
+            self.assertFalse(errors)
+            if verbose:
+                print("OK.")
+        done.release()
+
+    def test_import_hangers(self):
         # In case this test is run again, make sure the helper module
         # gets loaded from scratch again.
-        del sys.modules['test.threaded_import_hangers']
+        try:
+            del sys.modules['test.threaded_import_hangers']
+        except KeyError:
+            pass
+        import test.threaded_import_hangers
+        self.assertFalse(test.threaded_import_hangers.errors)
 
-# Tricky:  When regrtest imports this module, the thread running regrtest
-# grabs the import lock and won't let go of it until this module returns.
-# All other threads attempting an import hang for the duration.  Since
-# this test spawns threads that do little *but* import, we can't do that
-# successfully until after this module finishes importing and regrtest
-# regains control.  To make this work, a special case was added to
-# regrtest to invoke a module's "test_main" function (if any) after
-# importing it.
-
-def test_main():        # magic name!  see above
-    global N, done
-
-    import imp
-    if imp.lock_held():
-        # This triggers on, e.g., from test import autotest.
-        raise unittest.SkipTest("can't run when import lock is held")
-
-    done.acquire()
-    for N in (20, 50) * 3:
-        if verbose:
-            print("Trying", N, "threads ...", end=' ')
-        for i in range(N):
-            thread.start_new_thread(task, ())
-        done.acquire()
-        if verbose:
-            print("OK.")
-    done.release()
 
-    test_import_hangers()
+def test_main():
+    run_unittest(ThreadedImportTests)
+
 
 if __name__ == "__main__":
     test_main()

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Wed Jul 14 13:52:38 2010
@@ -1650,6 +1650,9 @@
 Tests
 -----
 
+- Issue #9251: test_threaded_import didn't fail when run through regrtest
+  if the import lock was disabled.
+
 - Issue #8605: Skip test_gdb if Python is compiled with optimizations.
 
 - Issue #7449: Skip test_socketserver if threading support is disabled

From python-checkins at python.org  Wed Jul 14 13:56:06 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 14 Jul 2010 13:56:06 +0200 (CEST)
Subject: [Python-checkins] r82886 - in python/branches/release31-maint:
	Lib/test/test_threaded_import.py Misc/NEWS
Message-ID: <20100714115606.91D6DEEA93@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 14 13:56:06 2010
New Revision: 82886

Log:
Merged revisions 82885 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82885 | antoine.pitrou | 2010-07-14 13:52:38 +0200 (mer., 14 juil. 2010) | 4 lines
  
  Issue #9251: test_threaded_import didn't fail when run through regrtest
  if the import lock was disabled.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/test/test_threaded_import.py
   python/branches/release31-maint/Misc/NEWS

Modified: python/branches/release31-maint/Lib/test/test_threaded_import.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_threaded_import.py	(original)
+++ python/branches/release31-maint/Lib/test/test_threaded_import.py	Wed Jul 14 13:56:06 2010
@@ -5,71 +5,67 @@
 # complains several times about module random having no attribute
 # randrange, and then Python hangs.
 
-import _thread as thread
-from test.support import verbose, TestFailed
+import imp
+import sys
+import unittest
+from test.support import verbose, TestFailed, import_module, run_unittest
+thread = import_module('_thread')
 
-critical_section = thread.allocate_lock()
-done = thread.allocate_lock()
+def task(N, done, done_tasks, errors):
+    try:
+        import random
+        # This will fail if random is not completely initialized
+        x = random.randrange(1, 3)
+    except Exception as e:
+        errors.append(e.with_traceback(None))
+    finally:
+        done_tasks.append(thread.get_ident())
+        finished = len(done_tasks) == N
+        if finished:
+            done.release()
 
-def task():
-    global N, critical_section, done
-    import random
-    x = random.randrange(1, 3)
-    critical_section.acquire()
-    N -= 1
-    # Must release critical_section before releasing done, else the main
-    # thread can exit and set critical_section to None as part of global
-    # teardown; then critical_section.release() raises AttributeError.
-    finished = N == 0
-    critical_section.release()
-    if finished:
-        done.release()
 
-def test_import_hangers():
-    import sys
-    if verbose:
-        print("testing import hangers ...", end=' ')
+class ThreadedImportTests(unittest.TestCase):
 
-    import test.threaded_import_hangers
-    try:
-        if test.threaded_import_hangers.errors:
-            raise TestFailed(test.threaded_import_hangers.errors)
-        elif verbose:
-            print("OK.")
-    finally:
+    def test_parallel_module_init(self):
+        if imp.lock_held():
+            # This triggers on, e.g., from test import autotest.
+            raise unittest.SkipTest("can't run when import lock is held")
+
+        done = thread.allocate_lock()
+        done.acquire()
+        for N in (20, 50) * 3:
+            if verbose:
+                print("Trying", N, "threads ...", end=' ')
+            # Make sure that random gets reimported freshly
+            try:
+                del sys.modules['random']
+            except KeyError:
+                pass
+            errors = []
+            done_tasks = []
+            for i in range(N):
+                thread.start_new_thread(task, (N, done, done_tasks, errors,))
+            done.acquire()
+            self.assertFalse(errors)
+            if verbose:
+                print("OK.")
+        done.release()
+
+    def test_import_hangers(self):
         # In case this test is run again, make sure the helper module
         # gets loaded from scratch again.
-        del sys.modules['test.threaded_import_hangers']
+        try:
+            del sys.modules['test.threaded_import_hangers']
+        except KeyError:
+            pass
+        import test.threaded_import_hangers
+        self.assertFalse(test.threaded_import_hangers.errors)
 
-# Tricky:  When regrtest imports this module, the thread running regrtest
-# grabs the import lock and won't let go of it until this module returns.
-# All other threads attempting an import hang for the duration.  Since
-# this test spawns threads that do little *but* import, we can't do that
-# successfully until after this module finishes importing and regrtest
-# regains control.  To make this work, a special case was added to
-# regrtest to invoke a module's "test_main" function (if any) after
-# importing it.
-
-def test_main():        # magic name!  see above
-    global N, done
-
-    import imp
-    if imp.lock_held():
-        # This triggers on, e.g., from test import autotest.
-        raise unittest.SkipTest("can't run when import lock is held")
-
-    done.acquire()
-    for N in (20, 50) * 3:
-        if verbose:
-            print("Trying", N, "threads ...", end=' ')
-        for i in range(N):
-            thread.start_new_thread(task, ())
-        done.acquire()
-        if verbose:
-            print("OK.")
-    done.release()
 
-    test_import_hangers()
+def test_main():
+    run_unittest(ThreadedImportTests)
+
 
 if __name__ == "__main__":
     test_main()

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Wed Jul 14 13:56:06 2010
@@ -333,6 +333,9 @@
 Tests
 -----
 
+- Issue #9251: test_threaded_import didn't fail when run through regrtest
+  if the import lock was disabled.
+
 - Issue #7449: Skip test_socketserver if threading support is disabled
 
 - Issue #8672: Add a zlib test ensuring that an incomplete stream can be

From python-checkins at python.org  Wed Jul 14 15:46:57 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Wed, 14 Jul 2010 15:46:57 +0200 (CEST)
Subject: [Python-checkins] r82887 - python/branches/py3k/Lib/_strptime.py
Message-ID: <20100714134657.9EBE3EEAAA@mail.python.org>

Author: alexander.belopolsky
Date: Wed Jul 14 15:46:57 2010
New Revision: 82887

Log:
PEP 8 conformance: class_ -> cls

Modified:
   python/branches/py3k/Lib/_strptime.py

Modified: python/branches/py3k/Lib/_strptime.py
==============================================================================
--- python/branches/py3k/Lib/_strptime.py	(original)
+++ python/branches/py3k/Lib/_strptime.py	Wed Jul 14 15:46:57 2010
@@ -482,8 +482,8 @@
     tt = _strptime(data_string, format)[0]
     return time.struct_time(tt[:9])
 
-def _strptime_datetime(class_, data_string, format="%a %b %d %H:%M:%S %Y"):
-    """Return a class_ instance based on the input string and the
+def _strptime_datetime(cls, data_string, format="%a %b %d %H:%M:%S %Y"):
+    """Return a class cls instance based on the input string and the
     format string."""
     tt, fraction = _strptime(data_string, format)
     gmtoff, tzname = tt[-2:]
@@ -496,4 +496,4 @@
             tz = datetime_timezone(tzdelta)
         args += (tz,)
 
-    return class_(*args)
+    return cls(*args)

From python-checkins at python.org  Wed Jul 14 15:48:41 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Wed, 14 Jul 2010 15:48:41 +0200 (CEST)
Subject: [Python-checkins] r82888 - python/branches/release31-maint
Message-ID: <20100714134841.EF931EEACF@mail.python.org>

Author: alexander.belopolsky
Date: Wed Jul 14 15:48:41 2010
New Revision: 82888

Log:
Blocked revisions 82887 via svnmerge

........
  r82887 | alexander.belopolsky | 2010-07-14 09:46:57 -0400 (Wed, 14 Jul 2010) | 1 line
  
  PEP 8 conformance: class_ -> cls
........


Modified:
   python/branches/release31-maint/   (props changed)

From python-checkins at python.org  Wed Jul 14 15:53:01 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Wed, 14 Jul 2010 15:53:01 +0200 (CEST)
Subject: [Python-checkins] r82889 - python/branches/release27-maint
Message-ID: <20100714135301.D38F9F35B@mail.python.org>

Author: alexander.belopolsky
Date: Wed Jul 14 15:53:01 2010
New Revision: 82889

Log:
Blocked revisions 82887 via svnmerge

........
  r82887 | alexander.belopolsky | 2010-07-14 09:46:57 -0400 (Wed, 14 Jul 2010) | 1 line
  
  PEP 8 conformance: class_ -> cls
........


Modified:
   python/branches/release27-maint/   (props changed)

From python-checkins at python.org  Wed Jul 14 21:15:23 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Wed, 14 Jul 2010 21:15:23 +0200 (CEST)
Subject: [Python-checkins] r82890 -
	python/branches/py3k/Lib/urllib/request.py
Message-ID: <20100714191523.88AB0EEA63@mail.python.org>

Author: senthil.kumaran
Date: Wed Jul 14 21:15:23 2010
New Revision: 82890

Log:
Fix: Issue6853 - Get HTTPS system proxy in Windows.



Modified:
   python/branches/py3k/Lib/urllib/request.py

Modified: python/branches/py3k/Lib/urllib/request.py
==============================================================================
--- python/branches/py3k/Lib/urllib/request.py	(original)
+++ python/branches/py3k/Lib/urllib/request.py	Wed Jul 14 21:15:23 2010
@@ -2279,6 +2279,7 @@
                         proxies['http'] = proxyServer
                     else:
                         proxies['http'] = 'http://%s' % proxyServer
+                        proxies['https'] = 'http://%s' % proxyServer
                         proxies['ftp'] = 'ftp://%s' % proxyServer
             internetSettings.Close()
         except (WindowsError, ValueError, TypeError):

From python-checkins at python.org  Wed Jul 14 21:20:43 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Wed, 14 Jul 2010 21:20:43 +0200 (CEST)
Subject: [Python-checkins] r82891 - in python/branches/release31-maint:
	Lib/urllib/request.py
Message-ID: <20100714192043.2ED8EEEA64@mail.python.org>

Author: senthil.kumaran
Date: Wed Jul 14 21:20:43 2010
New Revision: 82891

Log:
Merged revisions 82890 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82890 | senthil.kumaran | 2010-07-15 00:45:23 +0530 (Thu, 15 Jul 2010) | 3 lines
  
  Fix: Issue6853 - Get HTTPS system proxy in Windows.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/urllib/request.py

Modified: python/branches/release31-maint/Lib/urllib/request.py
==============================================================================
--- python/branches/release31-maint/Lib/urllib/request.py	(original)
+++ python/branches/release31-maint/Lib/urllib/request.py	Wed Jul 14 21:20:43 2010
@@ -2283,6 +2283,7 @@
                         proxies['http'] = proxyServer
                     else:
                         proxies['http'] = 'http://%s' % proxyServer
+                        proxies['https'] = 'http://%s' % proxyServer
                         proxies['ftp'] = 'ftp://%s' % proxyServer
             internetSettings.Close()
         except (WindowsError, ValueError, TypeError):

From python-checkins at python.org  Wed Jul 14 21:25:26 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Wed, 14 Jul 2010 21:25:26 +0200 (CEST)
Subject: [Python-checkins] r82892 - in python/branches/release27-maint:
	Lib/urllib.py
Message-ID: <20100714192526.694A7EEA54@mail.python.org>

Author: senthil.kumaran
Date: Wed Jul 14 21:25:26 2010
New Revision: 82892

Log:
Merged revisions 82890 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82890 | senthil.kumaran | 2010-07-15 00:45:23 +0530 (Thu, 15 Jul 2010) | 3 lines
  
  Fix: Issue6853 - Get HTTPS system proxy in Windows.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/urllib.py

Modified: python/branches/release27-maint/Lib/urllib.py
==============================================================================
--- python/branches/release27-maint/Lib/urllib.py	(original)
+++ python/branches/release27-maint/Lib/urllib.py	Wed Jul 14 21:25:26 2010
@@ -1453,6 +1453,7 @@
                         proxies['http'] = proxyServer
                     else:
                         proxies['http'] = 'http://%s' % proxyServer
+                        proxies['https'] = 'http://%s' % proxyServer
                         proxies['ftp'] = 'ftp://%s' % proxyServer
             internetSettings.Close()
         except (WindowsError, ValueError, TypeError):

From python-checkins at python.org  Wed Jul 14 21:28:35 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Wed, 14 Jul 2010 21:28:35 +0200 (CEST)
Subject: [Python-checkins] r82893 - in python/branches/release26-maint:
	Lib/urllib.py
Message-ID: <20100714192835.2FD27EE9C0@mail.python.org>

Author: senthil.kumaran
Date: Wed Jul 14 21:28:35 2010
New Revision: 82893

Log:
Merged revisions 82892 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

................
  r82892 | senthil.kumaran | 2010-07-15 00:55:26 +0530 (Thu, 15 Jul 2010) | 9 lines
  
  Merged revisions 82890 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k
  
  ........
    r82890 | senthil.kumaran | 2010-07-15 00:45:23 +0530 (Thu, 15 Jul 2010) | 3 lines
    
    Fix: Issue6853 - Get HTTPS system proxy in Windows.
  ........
................


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Lib/urllib.py

Modified: python/branches/release26-maint/Lib/urllib.py
==============================================================================
--- python/branches/release26-maint/Lib/urllib.py	(original)
+++ python/branches/release26-maint/Lib/urllib.py	Wed Jul 14 21:28:35 2010
@@ -1448,6 +1448,7 @@
                         proxies['http'] = proxyServer
                     else:
                         proxies['http'] = 'http://%s' % proxyServer
+                        proxies['https'] = 'http://%s' % proxyServer
                         proxies['ftp'] = 'ftp://%s' % proxyServer
             internetSettings.Close()
         except (WindowsError, ValueError, TypeError):

From fdrake at acm.org  Wed Jul 14 21:32:00 2010
From: fdrake at acm.org (Fred Drake)
Date: Wed, 14 Jul 2010 15:32:00 -0400
Subject: [Python-checkins] r82892 - in python/branches/release27-maint:
	Lib/urllib.py
In-Reply-To: <20100714192526.694A7EEA54@mail.python.org>
References: <20100714192526.694A7EEA54@mail.python.org>
Message-ID: <AANLkTil9KIGiByF8ZGd0YYzVqFsHy2QHu7v7Af5uG10b@mail.gmail.com>

On Wed, Jul 14, 2010 at 3:25 PM, senthil.kumaran
<python-checkins at python.org> wrote:
> + ? ? ? ? ? ? ? ? ? ? ? ?proxies['https'] = 'http://%s' % proxyServer

Is that really right?  Shouldn't the proxy be 'https://...' ?


? -Fred

--
Fred L. Drake, Jr.? ? <fdrake at gmail.com>
"A storm broke loose in my mind."? --Albert Einstein

From python-checkins at python.org  Wed Jul 14 22:01:47 2010
From: python-checkins at python.org (fred.drake)
Date: Wed, 14 Jul 2010 22:01:47 +0200 (CEST)
Subject: [Python-checkins] r82894 - peps/trunk/pep-3147.txt
Message-ID: <20100714200147.BD83EEEA94@mail.python.org>

Author: fred.drake
Date: Wed Jul 14 22:01:47 2010
New Revision: 82894

Log:
fix markup error


Modified:
   peps/trunk/pep-3147.txt

Modified: peps/trunk/pep-3147.txt
==============================================================================
--- peps/trunk/pep-3147.txt	(original)
+++ peps/trunk/pep-3147.txt	Wed Jul 14 22:01:47 2010
@@ -543,7 +543,7 @@
 ----
 
 A proposal was floated to call the `__pycache__` directory `.pyc` or
-some other dot-file name.  This would have the effect on *nix systems
+some other dot-file name.  This would have the effect on \*nix systems
 of hiding the directory.  There are many reasons why this was
 rejected by the BDFL [20]_ including the fact that dot-files are only
 special on some platforms, and we actually do *not* want to hide these

From python-checkins at python.org  Wed Jul 14 22:10:52 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Wed, 14 Jul 2010 22:10:52 +0200 (CEST)
Subject: [Python-checkins] r82895 -
	python/branches/py3k/Lib/urllib/request.py
Message-ID: <20100714201052.46FF5EE985@mail.python.org>

Author: senthil.kumaran
Date: Wed Jul 14 22:10:52 2010
New Revision: 82895

Log:
Fix a mistake, https proxy shoud be https://



Modified:
   python/branches/py3k/Lib/urllib/request.py

Modified: python/branches/py3k/Lib/urllib/request.py
==============================================================================
--- python/branches/py3k/Lib/urllib/request.py	(original)
+++ python/branches/py3k/Lib/urllib/request.py	Wed Jul 14 22:10:52 2010
@@ -2279,7 +2279,7 @@
                         proxies['http'] = proxyServer
                     else:
                         proxies['http'] = 'http://%s' % proxyServer
-                        proxies['https'] = 'http://%s' % proxyServer
+                        proxies['https'] = 'https://%s' % proxyServer
                         proxies['ftp'] = 'ftp://%s' % proxyServer
             internetSettings.Close()
         except (WindowsError, ValueError, TypeError):

From orsenthil at gmail.com  Wed Jul 14 22:12:14 2010
From: orsenthil at gmail.com (Senthil Kumaran)
Date: Thu, 15 Jul 2010 01:42:14 +0530
Subject: [Python-checkins] r82892 - in python/branches/release27-maint:
 Lib/urllib.py
In-Reply-To: <AANLkTil9KIGiByF8ZGd0YYzVqFsHy2QHu7v7Af5uG10b@mail.gmail.com>
References: <20100714192526.694A7EEA54@mail.python.org>
	<AANLkTil9KIGiByF8ZGd0YYzVqFsHy2QHu7v7Af5uG10b@mail.gmail.com>
Message-ID: <20100714201214.GA1989@remy>

On Wed, Jul 14, 2010 at 03:32:00PM -0400, Fred Drake wrote:
> > + ? ? ? ? ? ? ? ? ? ? ? ?proxies['https'] = 'http://%s' % proxyServer
> 
> Is that really right?  Shouldn't the proxy be 'https://...' ?

Yes, it should be. Thank you for spotting that. I corrected it the
next check-in.


-- 
Senthil

Think of prototypes as a funny markup language--the interpretation is
left up to the rendering engine.
		-- Larry Wall in <199710221710.KAA24242 at wall.org>

From python-checkins at python.org  Wed Jul 14 22:13:29 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Wed, 14 Jul 2010 22:13:29 +0200 (CEST)
Subject: [Python-checkins] r82896 - in python/branches/release31-maint:
	Lib/urllib/request.py
Message-ID: <20100714201329.09291EE985@mail.python.org>

Author: senthil.kumaran
Date: Wed Jul 14 22:13:28 2010
New Revision: 82896

Log:
Merged revisions 82895 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82895 | senthil.kumaran | 2010-07-15 01:40:52 +0530 (Thu, 15 Jul 2010) | 3 lines
  
  Fix a mistake, https proxy shoud be https://
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/urllib/request.py

Modified: python/branches/release31-maint/Lib/urllib/request.py
==============================================================================
--- python/branches/release31-maint/Lib/urllib/request.py	(original)
+++ python/branches/release31-maint/Lib/urllib/request.py	Wed Jul 14 22:13:28 2010
@@ -2283,7 +2283,7 @@
                         proxies['http'] = proxyServer
                     else:
                         proxies['http'] = 'http://%s' % proxyServer
-                        proxies['https'] = 'http://%s' % proxyServer
+                        proxies['https'] = 'https://%s' % proxyServer
                         proxies['ftp'] = 'ftp://%s' % proxyServer
             internetSettings.Close()
         except (WindowsError, ValueError, TypeError):

From python-checkins at python.org  Wed Jul 14 22:22:18 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Wed, 14 Jul 2010 22:22:18 +0200 (CEST)
Subject: [Python-checkins] r82897 - in python/branches/release27-maint:
	Lib/urllib.py
Message-ID: <20100714202218.0F055EE985@mail.python.org>

Author: senthil.kumaran
Date: Wed Jul 14 22:22:17 2010
New Revision: 82897

Log:
Merged revisions 82895 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82895 | senthil.kumaran | 2010-07-15 01:40:52 +0530 (Thu, 15 Jul 2010) | 3 lines
  
  Fix a mistake, https proxy shoud be https://
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/urllib.py

Modified: python/branches/release27-maint/Lib/urllib.py
==============================================================================
--- python/branches/release27-maint/Lib/urllib.py	(original)
+++ python/branches/release27-maint/Lib/urllib.py	Wed Jul 14 22:22:17 2010
@@ -1453,7 +1453,7 @@
                         proxies['http'] = proxyServer
                     else:
                         proxies['http'] = 'http://%s' % proxyServer
-                        proxies['https'] = 'http://%s' % proxyServer
+                        proxies['https'] = 'https://%s' % proxyServer
                         proxies['ftp'] = 'ftp://%s' % proxyServer
             internetSettings.Close()
         except (WindowsError, ValueError, TypeError):

From python-checkins at python.org  Wed Jul 14 22:25:15 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Wed, 14 Jul 2010 22:25:15 +0200 (CEST)
Subject: [Python-checkins] r82898 - in python/branches/release26-maint:
	Lib/urllib.py
Message-ID: <20100714202515.57DE2EEA54@mail.python.org>

Author: senthil.kumaran
Date: Wed Jul 14 22:25:15 2010
New Revision: 82898

Log:
Merged revisions 82897 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

................
  r82897 | senthil.kumaran | 2010-07-15 01:52:17 +0530 (Thu, 15 Jul 2010) | 9 lines
  
  Merged revisions 82895 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k
  
  ........
    r82895 | senthil.kumaran | 2010-07-15 01:40:52 +0530 (Thu, 15 Jul 2010) | 3 lines
    
    Fix a mistake, https proxy shoud be https://
  ........
................


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Lib/urllib.py

Modified: python/branches/release26-maint/Lib/urllib.py
==============================================================================
--- python/branches/release26-maint/Lib/urllib.py	(original)
+++ python/branches/release26-maint/Lib/urllib.py	Wed Jul 14 22:25:15 2010
@@ -1448,7 +1448,7 @@
                         proxies['http'] = proxyServer
                     else:
                         proxies['http'] = 'http://%s' % proxyServer
-                        proxies['https'] = 'http://%s' % proxyServer
+                        proxies['https'] = 'https://%s' % proxyServer
                         proxies['ftp'] = 'ftp://%s' % proxyServer
             internetSettings.Close()
         except (WindowsError, ValueError, TypeError):

From python-checkins at python.org  Wed Jul 14 22:30:03 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Wed, 14 Jul 2010 22:30:03 +0200 (CEST)
Subject: [Python-checkins] r82899 -
	python/branches/py3k/Doc/reference/expressions.rst
Message-ID: <20100714203003.11A47EE99E@mail.python.org>

Author: senthil.kumaran
Date: Wed Jul 14 22:30:02 2010
New Revision: 82899

Log:
Fix issue9132 - Documentation for comparing dictionaries is out of date


Modified:
   python/branches/py3k/Doc/reference/expressions.rst

Modified: python/branches/py3k/Doc/reference/expressions.rst
==============================================================================
--- python/branches/py3k/Doc/reference/expressions.rst	(original)
+++ python/branches/py3k/Doc/reference/expressions.rst	Wed Jul 14 22:30:02 2010
@@ -1031,9 +1031,9 @@
   ``x <= y``.  If the corresponding element does not exist, the shorter
   sequence is ordered first (for example, ``[1,2] < [1,2,3]``).
 
-* Mappings (dictionaries) compare equal if and only if their sorted ``(key,
-  value)`` lists compare equal. [#]_ Outcomes other than equality are resolved
-  consistently, but are not otherwise defined. [#]_
+* Mappings (dictionaries) compare equal if and only if they have the same
+  ``(key, value)`` pairs. Order comparisons ``('<', '<=', '>=', '>')``
+  raise :exc:`TypeError`.
 
 * Sets and frozensets define comparison operators to mean subset and superset
   tests.  Those relations do not define total orderings (the two sets ``{1,2}``
@@ -1330,15 +1330,6 @@
    strings in a human recognizable way, compare using
    :func:`unicodedata.normalize`.
 
-.. [#] The implementation computes this efficiently, without constructing lists
-   or sorting.
-
-.. [#] Earlier versions of Python used lexicographic comparison of the sorted (key,
-   value) lists, but this was very expensive for the common case of comparing
-   for equality.  An even earlier version of Python compared dictionaries by
-   identity only, but this caused surprises because people expected to be able
-   to test a dictionary for emptiness by comparing it to ``{}``.
-
 .. [#] Due to automatic garbage-collection, free lists, and the dynamic nature of
    descriptors, you may notice seemingly unusual behaviour in certain uses of
    the :keyword:`is` operator, like those involving comparisons between instance

From python-checkins at python.org  Wed Jul 14 22:32:56 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Wed, 14 Jul 2010 22:32:56 +0200 (CEST)
Subject: [Python-checkins] r82900 - in python/branches/release31-maint:
	Doc/reference/expressions.rst
Message-ID: <20100714203256.6F73CEFC0@mail.python.org>

Author: senthil.kumaran
Date: Wed Jul 14 22:32:56 2010
New Revision: 82900

Log:
Merged revisions 82899 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82899 | senthil.kumaran | 2010-07-15 02:00:02 +0530 (Thu, 15 Jul 2010) | 2 lines
  
  Fix issue9132 - Documentation for comparing dictionaries is out of date
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Doc/reference/expressions.rst

Modified: python/branches/release31-maint/Doc/reference/expressions.rst
==============================================================================
--- python/branches/release31-maint/Doc/reference/expressions.rst	(original)
+++ python/branches/release31-maint/Doc/reference/expressions.rst	Wed Jul 14 22:32:56 2010
@@ -1026,9 +1026,9 @@
   ``x <= y``.  If the corresponding element does not exist, the shorter
   sequence is ordered first (for example, ``[1,2] < [1,2,3]``).
 
-* Mappings (dictionaries) compare equal if and only if their sorted ``(key,
-  value)`` lists compare equal. [#]_ Outcomes other than equality are resolved
-  consistently, but are not otherwise defined. [#]_
+* Mappings (dictionaries) compare equal if and only if they have the same
+  ``(key, value)`` pairs. Order comparisons ``('<', '<=', '>=', '>')``
+  raise :exc:`TypeError`.
 
 * Sets and frozensets define comparison operators to mean subset and superset
   tests.  Those relations do not define total orderings (the two sets ``{1,2}``
@@ -1310,15 +1310,6 @@
    strings in a human recognizable way, compare using
    :func:`unicodedata.normalize`.
 
-.. [#] The implementation computes this efficiently, without constructing lists
-   or sorting.
-
-.. [#] Earlier versions of Python used lexicographic comparison of the sorted (key,
-   value) lists, but this was very expensive for the common case of comparing
-   for equality.  An even earlier version of Python compared dictionaries by
-   identity only, but this caused surprises because people expected to be able
-   to test a dictionary for emptiness by comparing it to ``{}``.
-
 .. [#] Due to automatic garbage-collection, free lists, and the dynamic nature of
    descriptors, you may notice seemingly unusual behaviour in certain uses of
    the :keyword:`is` operator, like those involving comparisons between instance

From python-checkins at python.org  Wed Jul 14 22:38:12 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 14 Jul 2010 22:38:12 +0200 (CEST)
Subject: [Python-checkins] r82901 -
	python/branches/import_unicode/Python/pythonrun.c
Message-ID: <20100714203812.77919FE44@mail.python.org>

Author: victor.stinner
Date: Wed Jul 14 22:38:12 2010
New Revision: 82901

Log:
PyRun_SimpleFileExFlags() uses PyUnicode_DecodeFSDefault()

Modified:
   python/branches/import_unicode/Python/pythonrun.c

Modified: python/branches/import_unicode/Python/pythonrun.c
==============================================================================
--- python/branches/import_unicode/Python/pythonrun.c	(original)
+++ python/branches/import_unicode/Python/pythonrun.c	Wed Jul 14 22:38:12 2010
@@ -1332,7 +1332,7 @@
     d = PyModule_GetDict(m);
     if (PyDict_GetItemString(d, "__file__") == NULL) {
         PyObject *f;
-        f = PyUnicode_FromString(filename);
+        f = PyUnicode_DecodeFSDefault(filename);
         if (f == NULL)
             return -1;
         if (PyDict_SetItemString(d, "__file__", f) < 0) {

From python-checkins at python.org  Wed Jul 14 22:38:15 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 14 Jul 2010 22:38:15 +0200 (CEST)
Subject: [Python-checkins] r82902 - in python/branches/import_unicode:
	Modules/zipimport.c Python/import.c
Message-ID: <20100714203815.4C434EE9EB@mail.python.org>

Author: victor.stinner
Date: Wed Jul 14 22:38:15 2010
New Revision: 82902

Log:
get_data() takes a PyObject* for archive argument

 * Rename fopen_unicode() to _Py_fopen(), it's now a private function
 * Rename stat_unicode() to _Py_stat(), it's now a private function

Modified:
   python/branches/import_unicode/Modules/zipimport.c
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Modules/zipimport.c
==============================================================================
--- python/branches/import_unicode/Modules/zipimport.c	(original)
+++ python/branches/import_unicode/Modules/zipimport.c	Wed Jul 14 22:38:15 2010
@@ -44,8 +44,9 @@
 static PyObject *zip_directory_cache = NULL;
 
 /* forward decls */
+extern FILE* _Py_fopen(PyObject *unicode, const char *mode);
 static PyObject *read_directory(char *archive);
-static PyObject *get_data(char *archive, PyObject *toc_entry);
+static PyObject *get_data(PyObject *archive, PyObject *toc_entry);
 static PyObject *get_module_code(ZipImporter *self, char *fullname,
                                  int *p_ispackage, char **p_modpath);
 
@@ -441,7 +442,7 @@
         PyErr_SetFromErrnoWithFilename(PyExc_IOError, path);
         return NULL;
     }
-    return get_data(archive_str, toc_entry);
+    return get_data(self->archive, toc_entry);
 }
 
 static PyObject *
@@ -491,7 +492,7 @@
 
     toc_entry = PyDict_GetItemString(self->files, path);
     if (toc_entry != NULL) {
-        PyObject *bytes = get_data(_PyUnicode_AsString(self->archive), toc_entry);
+        PyObject *bytes = get_data(self->archive, toc_entry);
         PyObject *res = PyUnicode_FromString(PyBytes_AsString(bytes));
         Py_XDECREF(bytes);
         return res;
@@ -825,7 +826,7 @@
 /* Given a path to a Zip file and a toc_entry, return the (uncompressed)
    data as a new reference. */
 static PyObject *
-get_data(char *archive, PyObject *toc_entry)
+get_data(PyObject *archive, PyObject *toc_entry)
 {
     PyObject *raw_data, *data = NULL, *decompress;
     char *buf;
@@ -843,10 +844,10 @@
         return NULL;
     }
 
-    fp = fopen(archive, "rb");
+    fp = _Py_fopen(archive, "rb");
     if (!fp) {
         PyErr_Format(PyExc_IOError,
-           "zipimport: can not open file %s", archive);
+           "zipimport: can not open file %U", archive);
         return NULL;
     }
 
@@ -856,7 +857,7 @@
     if (l != 0x04034B50) {
         /* Bad: Local File Header */
         PyErr_Format(ZipImportError,
-                     "bad local file header in %s",
+                     "bad local file header in %U",
                      archive);
         fclose(fp);
         return NULL;
@@ -1086,12 +1087,8 @@
 {
     PyObject *data, *code;
     char *modpath;
-    char *archive = _PyUnicode_AsString(self->archive);
-
-    if (archive == NULL)
-        return NULL;
 
-    data = get_data(archive, toc_entry);
+    data = get_data(self->archive, toc_entry);
     if (data == NULL)
         return NULL;
 

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Wed Jul 14 22:38:15 2010
@@ -137,9 +137,9 @@
 };
 
 /* Forward */
-static FILE* fopen_unicode(PyObject *unicode, const char *mode);
+FILE* _Py_fopen(PyObject *unicode, const char *mode);
 #ifdef HAVE_STAT
-static int stat_unicode(PyObject *unicode, struct stat *statbuf);
+int _Py_stat(PyObject *unicode, struct stat *statbuf);
 static int find_init_module(PyObject *);
 #endif
 
@@ -1084,7 +1084,7 @@
     long magic;
     long pyc_mtime;
 
-    fp = fopen_unicode(cpathobj, "rb");
+    fp = _Py_fopen(cpathobj, "rb");
     if (fp == NULL)
         return NULL;
     magic = PyMarshal_ReadLongFromFile(fp);
@@ -1471,7 +1471,7 @@
             return NULL;
     }
 
-    if (stat_unicode(pyobj, &statbuf) == 0 &&
+    if (_Py_stat(pyobj, &statbuf) == 0 &&
         S_ISREG(statbuf.st_mode)) {
         return pyobj;
     }
@@ -1828,7 +1828,7 @@
         /* Check for package import (buf holds a directory name,
            and there's an __init__ module in that directory */
 #ifdef HAVE_STAT
-        if (stat_unicode(unicode, &statbuf) == 0 &&         /* it exists */
+        if (_Py_stat(unicode, &statbuf) == 0 &&         /* it exists */
             S_ISDIR(statbuf.st_mode) &&         /* it's a directory */
             case_ok(unicode, 0, namelen, name)) { /* case matches */
             if (find_init_module(unicode)) { /* and has __init__.py */
@@ -1894,7 +1894,7 @@
             filemode = fdp->mode;
             if (filemode[0] == 'U')
                 filemode = "r" PY_STDIOTEXTMODE;
-            fp = fopen_unicode(unicode, filemode);
+            fp = _Py_fopen(unicode, filemode);
             if (fp != NULL) {
                 if (case_ok(unicode, strlen(fdp->suffix), namelen, name)) {
                     break;
@@ -2024,7 +2024,7 @@
     if (fullpath_bytes == NULL)
         return 0;
 
-    done = findfirst(PyBytes_AS_STRING(fullpath_bytes), 
+    done = findfirst(PyBytes_AS_STRING(fullpath_bytes),
                      &ffblk, FA_ARCH|FA_RDONLY|FA_HIDDEN|FA_DIREC);
     Py_DECREF(fullpath_bytes);
     if (done) {
@@ -2119,9 +2119,10 @@
 #endif
 }
 
-static FILE*
-fopen_unicode(PyObject *unicode, const char *mode)
+FILE*
+_Py_fopen(PyObject *unicode, const char *mode)
 {
+    /* FIXME: use _wfopen() on Windows */
     FILE *f;
     PyObject *bytes = PyUnicode_EncodeFSDefault(unicode);
     if (bytes == NULL) {
@@ -2136,8 +2137,8 @@
 
 
 #ifdef HAVE_STAT
-static int
-stat_unicode(PyObject *unicode, struct stat *statbuf)
+int
+_Py_stat(PyObject *unicode, struct stat *statbuf)
 {
     int ret;
     PyObject *bytes = PyUnicode_EncodeFSDefault(unicode);
@@ -2159,7 +2160,7 @@
     PyObject *unicode;
 
     unicode = PyUnicode_FromFormat("%U%c__init__.py", bufobj, SEP);
-    if (stat_unicode(unicode, &statbuf) == 0) {
+    if (_Py_stat(unicode, &statbuf) == 0) {
         if (case_ok(unicode,
                     3,   /* ignore ".py" suffix */
                     8, "__init__")) {
@@ -2170,7 +2171,7 @@
     Py_DECREF(unicode);
 
     unicode = PyUnicode_FromFormat("%U%c__init__.py%c", bufobj, SEP, Py_OptimizeFlag ? "o" : "c");
-    if (stat_unicode(unicode, &statbuf) == 0) {
+    if (_Py_stat(unicode, &statbuf) == 0) {
         if (case_ok(unicode,
                     4,   /* ignore ".pyc" / ".pyo" suffix */
                     8, "__init__")) {
@@ -3467,7 +3468,7 @@
     if (mode[0] == 'U')
         mode = "r" PY_STDIOTEXTMODE;
     if (fob == NULL) {
-        fp = fopen_unicode(pathobj, mode);
+        fp = _Py_fopen(pathobj, mode);
     }
     else {
         int fd = PyObject_AsFileDescriptor(fob);
@@ -3815,7 +3816,7 @@
         struct stat statbuf;
         int rv;
 
-        rv = stat_unicode(pathobj, &statbuf);
+        rv = _Py_stat(pathobj, &statbuf);
         if (rv == 0) {
             /* it exists */
             if (S_ISDIR(statbuf.st_mode)) {

From python-checkins at python.org  Wed Jul 14 22:38:17 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 14 Jul 2010 22:38:17 +0200 (CEST)
Subject: [Python-checkins] r82903 -
	python/branches/import_unicode/Modules/zipimport.c
Message-ID: <20100714203817.81338EE9AD@mail.python.org>

Author: victor.stinner
Date: Wed Jul 14 22:38:17 2010
New Revision: 82903

Log:
get_module_code() module path is now an object

Modified:
   python/branches/import_unicode/Modules/zipimport.c

Modified: python/branches/import_unicode/Modules/zipimport.c
==============================================================================
--- python/branches/import_unicode/Modules/zipimport.c	(original)
+++ python/branches/import_unicode/Modules/zipimport.c	Wed Jul 14 22:38:17 2010
@@ -48,7 +48,7 @@
 static PyObject *read_directory(char *archive);
 static PyObject *get_data(PyObject *archive, PyObject *toc_entry);
 static PyObject *get_module_code(ZipImporter *self, char *fullname,
-                                 int *p_ispackage, char **p_modpath);
+                                 int *p_ispackage, PyObject **p_modpath);
 
 
 #define ZipImporter_Check(op) PyObject_TypeCheck(op, &ZipImporter_Type)
@@ -301,7 +301,8 @@
 {
     ZipImporter *self = (ZipImporter *)obj;
     PyObject *code, *mod, *dict;
-    char *fullname, *modpath;
+    char *fullname;
+    PyObject *modpath;
     int ispackage;
 
     if (!PyArg_ParseTuple(args, "s:zipimporter.load_module",
@@ -331,8 +332,8 @@
         int err;
 
         fullpath = PyUnicode_FromFormat("%U%c%U%s",
-                                self->archive, SEP,
-                                self->prefix, subname);
+                                        self->archive, SEP,
+                                        self->prefix, subname);
         if (fullpath == NULL)
             goto error;
 
@@ -345,11 +346,11 @@
         if (err != 0)
             goto error;
     }
-    mod = PyImport_ExecCodeModuleEx(fullname, code, modpath);
+    mod = PyImport_ExecCodeModuleWithUnicodePathnames(fullname, code, modpath, NULL);
     Py_DECREF(code);
     if (Py_VerboseFlag)
-        PySys_WriteStderr("import %s # loaded from Zip %s\n",
-                          fullname, modpath);
+        PySys_FormatStderr("import %s # loaded from Zip %U\n",
+                           fullname, modpath);
     return mod;
 error:
     Py_DECREF(code);
@@ -363,21 +364,23 @@
 {
     ZipImporter *self = (ZipImporter *)obj;
     PyObject *code;
-    char *fullname, *modpath;
+    char *fullname;
+    PyObject *modpath;
     int ispackage;
 
     if (!PyArg_ParseTuple(args, "s:zipimporter.get_filename",
                          &fullname))
-    return NULL;
+        return NULL;
 
     /* Deciding the filename requires working out where the code
        would come from if the module was actually loaded */
     code = get_module_code(self, fullname, &ispackage, &modpath);
     if (code == NULL)
-    return NULL;
+        return NULL;
     Py_DECREF(code); /* Only need the path info */
 
-    return PyUnicode_FromString(modpath);
+    Py_INCREF(modpath);
+    return modpath;
 }
 
 /* Return a bool signifying whether the module is a package or not. */
@@ -1108,7 +1111,7 @@
    'fullname'. */
 static PyObject *
 get_module_code(ZipImporter *self, char *fullname,
-                int *p_ispackage, char **p_modpath)
+                int *p_ispackage, PyObject **p_modpath)
 {
     PyObject *toc_entry;
     char *subname, path[MAXPATHLEN + 1];
@@ -1149,8 +1152,7 @@
                 continue;
             }
             if (code != NULL && p_modpath != NULL)
-                *p_modpath = _PyUnicode_AsString(
-                    PyTuple_GetItem(toc_entry, 0));
+                *p_modpath = PyTuple_GetItem(toc_entry, 0);
             return code;
         }
     }

From python-checkins at python.org  Wed Jul 14 22:38:19 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 14 Jul 2010 22:38:19 +0200 (CEST)
Subject: [Python-checkins] r82904 -
	python/branches/import_unicode/Modules/zipimport.c
Message-ID: <20100714203819.9A7B9EE9F8@mail.python.org>

Author: victor.stinner
Date: Wed Jul 14 22:38:19 2010
New Revision: 82904

Log:
read_directory() now takes an object

Modified:
   python/branches/import_unicode/Modules/zipimport.c

Modified: python/branches/import_unicode/Modules/zipimport.c
==============================================================================
--- python/branches/import_unicode/Modules/zipimport.c	(original)
+++ python/branches/import_unicode/Modules/zipimport.c	Wed Jul 14 22:38:19 2010
@@ -45,7 +45,7 @@
 
 /* forward decls */
 extern FILE* _Py_fopen(PyObject *unicode, const char *mode);
-static PyObject *read_directory(char *archive);
+static PyObject *read_directory(PyObject *archive);
 static PyObject *get_data(PyObject *archive, PyObject *toc_entry);
 static PyObject *get_module_code(ZipImporter *self, char *fullname,
                                  int *p_ispackage, PyObject **p_modpath);
@@ -121,7 +121,12 @@
         PyObject *files;
         files = PyDict_GetItemString(zip_directory_cache, path);
         if (files == NULL) {
-            files = read_directory(buf);
+            PyObject *bufobj;
+            bufobj = PyUnicode_DecodeFSDefault(buf);
+            if (bufobj == NULL)
+                return -1;
+            files = read_directory(bufobj);
+            Py_DECREF(bufobj);
             if (files == NULL)
                 return -1;
             if (PyDict_SetItemString(zip_directory_cache, path,
@@ -679,8 +684,9 @@
    data_size and file_offset are 0.
 */
 static PyObject *
-read_directory(char *archive)
+read_directory(PyObject *archive_obj)
 {
+    /* FIXME: work on Py_UNICODE* instead of char* */
     PyObject *files = NULL;
     FILE *fp;
     long compress, crc, data_size, file_size, file_offset, date, time;
@@ -691,18 +697,25 @@
     char name[MAXPATHLEN + 5];
     char *p, endof_central_dir[22];
     long arc_offset; /* offset from beginning of file to start of zip-archive */
+    PyObject *archive_bytes;
+
+    archive_bytes = PyUnicode_EncodeFSDefault(archive_obj);
+    if (archive_bytes == NULL)
+        return NULL;
 
-    if (strlen(archive) > MAXPATHLEN) {
+    if (PyBytes_GET_SIZE(archive_bytes) > MAXPATHLEN) {
         PyErr_SetString(PyExc_OverflowError,
                         "Zip path name is too long");
+        Py_DECREF(archive_bytes);
         return NULL;
     }
-    strcpy(path, archive);
+    strcpy(path, PyBytes_AsString(archive_bytes));
+    Py_DECREF(archive_bytes);
 
-    fp = fopen(archive, "rb");
+    fp = _Py_fopen(archive_obj, "rb");
     if (fp == NULL) {
         PyErr_Format(ZipImportError, "can't open Zip file: "
-                     "'%.200s'", archive);
+                     "'%.200U'", archive_obj);
         return NULL;
     }
     fseek(fp, -22, SEEK_END);
@@ -710,14 +723,14 @@
     if (fread(endof_central_dir, 1, 22, fp) != 22) {
         fclose(fp);
         PyErr_Format(ZipImportError, "can't read Zip file: "
-                     "'%.200s'", archive);
+                     "'%.200U'", archive_obj);
         return NULL;
     }
     if (get_long((unsigned char *)endof_central_dir) != 0x06054B50) {
         /* Bad: End of Central Dir signature */
         fclose(fp);
         PyErr_Format(ZipImportError, "not a Zip file: "
-                     "'%.200s'", archive);
+                     "'%.200U'", archive_obj);
         return NULL;
     }
 
@@ -783,8 +796,8 @@
     }
     fclose(fp);
     if (Py_VerboseFlag)
-        PySys_WriteStderr("# zipimport: found %ld names in %s\n",
-            count, archive);
+        PySys_FormatStderr("# zipimport: found %ld names in %U\n",
+            count, archive_obj);
     return files;
 error:
     fclose(fp);

From python-checkins at python.org  Wed Jul 14 22:38:21 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 14 Jul 2010 22:38:21 +0200 (CEST)
Subject: [Python-checkins] r82905 -
	python/branches/import_unicode/Modules/zipimport.c
Message-ID: <20100714203821.6FBE1EE9CE@mail.python.org>

Author: victor.stinner
Date: Wed Jul 14 22:38:21 2010
New Revision: 82905

Log:
use an object to access zipimport cache

Modified:
   python/branches/import_unicode/Modules/zipimport.c

Modified: python/branches/import_unicode/Modules/zipimport.c
==============================================================================
--- python/branches/import_unicode/Modules/zipimport.c	(original)
+++ python/branches/import_unicode/Modules/zipimport.c	Wed Jul 14 22:38:21 2010
@@ -64,11 +64,14 @@
     PyObject *pathbytes;
     char *path, *p, *prefix, buf[MAXPATHLEN+2];
     size_t len;
+    PyObject *files;
+    PyObject *pathobj;
 
     if (!_PyArg_NoKeywords("zipimporter()", kwds))
         return -1;
 
-    if (!PyArg_ParseTuple(args, "O&:zipimporter", PyUnicode_FSConverter, &pathbytes))
+    if (!PyArg_ParseTuple(args, "O&:zipimporter",
+                          PyUnicode_FSConverter, &pathbytes))
         return -1;
 
     len = PyBytes_GET_SIZE(pathbytes);
@@ -117,31 +120,38 @@
         *p = '\0';
         prefix = p;
     }
-    if (path != NULL) {
-        PyObject *files;
-        files = PyDict_GetItemString(zip_directory_cache, path);
-        if (files == NULL) {
-            PyObject *bufobj;
-            bufobj = PyUnicode_DecodeFSDefault(buf);
-            if (bufobj == NULL)
-                return -1;
-            files = read_directory(bufobj);
-            Py_DECREF(bufobj);
-            if (files == NULL)
-                return -1;
-            if (PyDict_SetItemString(zip_directory_cache, path,
-                                     files) != 0)
-                return -1;
-        }
-        else
-            Py_INCREF(files);
-        self->files = files;
-    }
-    else {
+    if (path == NULL)
         PyErr_SetString(ZipImportError, "not a Zip file");
         return -1;
     }
 
+    pathobj = PyUnicode_DecodeFSDefault(path);
+    if (pathobj == NULL)
+        return -1;
+    files = PyDict_GetItem(zip_directory_cache, pathobj);
+    if (files == NULL) {
+        PyObject *bufobj;
+        bufobj = PyUnicode_DecodeFSDefault(buf);
+        if (bufobj == NULL) {
+            Py_DECREF(pathobj);
+            return -1;
+        }
+        files = read_directory(bufobj);
+        Py_DECREF(bufobj);
+        if (files == NULL) {
+            Py_DECREF(pathobj);
+            return -1;
+        }
+        if (PyDict_SetItem(zip_directory_cache, pathobj, files) != 0) {
+            Py_DECREF(pathobj);
+            return -1;
+        }
+    }
+    else
+        Py_INCREF(files);
+    Py_DECREF(pathobj);
+    self->files = files;
+
     if (prefix == NULL)
         prefix = "";
     else {

From python-checkins at python.org  Wed Jul 14 22:38:22 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 14 Jul 2010 22:38:22 +0200 (CEST)
Subject: [Python-checkins] r82906 -
	python/branches/import_unicode/Modules/zipimport.c
Message-ID: <20100714203822.D54F5EE9E5@mail.python.org>

Author: victor.stinner
Date: Wed Jul 14 22:38:22 2010
New Revision: 82906

Log:
Use PyUnicode_DecodeFSDefault() for archive and prefix

Modified:
   python/branches/import_unicode/Modules/zipimport.c

Modified: python/branches/import_unicode/Modules/zipimport.c
==============================================================================
--- python/branches/import_unicode/Modules/zipimport.c	(original)
+++ python/branches/import_unicode/Modules/zipimport.c	Wed Jul 14 22:38:22 2010
@@ -70,8 +70,7 @@
     if (!_PyArg_NoKeywords("zipimporter()", kwds))
         return -1;
 
-    if (!PyArg_ParseTuple(args, "O&:zipimporter",
-                          PyUnicode_FSConverter, &pathbytes))
+    if (!PyArg_ParseTuple(args, "O&:zipimporter", PyUnicode_FSConverter, &pathbytes))
         return -1;
 
     len = PyBytes_GET_SIZE(pathbytes);
@@ -120,7 +119,7 @@
         *p = '\0';
         prefix = p;
     }
-    if (path == NULL)
+    if (path == NULL) {
         PyErr_SetString(ZipImportError, "not a Zip file");
         return -1;
     }
@@ -142,7 +141,8 @@
             Py_DECREF(pathobj);
             return -1;
         }
-        if (PyDict_SetItem(zip_directory_cache, pathobj, files) != 0) {
+        if (PyDict_SetItem(zip_directory_cache, pathobj,
+                                 files) != 0) {
             Py_DECREF(pathobj);
             return -1;
         }
@@ -164,11 +164,11 @@
         }
     }
 
-    self->archive = PyUnicode_FromString(buf);
+    self->archive = PyUnicode_DecodeFSDefault(buf);
     if (self->archive == NULL)
         return -1;
 
-    self->prefix = PyUnicode_FromString(prefix);
+    self->prefix = PyUnicode_DecodeFSDefault(prefix);
     if (self->prefix == NULL)
         return -1;
 
@@ -197,19 +197,12 @@
 static PyObject *
 zipimporter_repr(ZipImporter *self)
 {
-    char *archive = "???";
-    char *prefix = "";
-
-    if (self->archive != NULL && PyUnicode_Check(self->archive))
-        archive = _PyUnicode_AsString(self->archive);
-    if (self->prefix != NULL && PyUnicode_Check(self->prefix))
-        prefix = _PyUnicode_AsString(self->prefix);
-    if (prefix != NULL && *prefix)
-        return PyUnicode_FromFormat("<zipimporter object \"%.300s%c%.150s\">",
-                                    archive, SEP, prefix);
+    if (PyUnicode_GET_SIZE(self->prefix) != 0)
+        return PyUnicode_FromFormat("<zipimporter object \"%.300U%c%.150U\">",
+                                    self->archive, SEP, self->prefix);
     else
-        return PyUnicode_FromFormat("<zipimporter object \"%.300s\">",
-                                    archive);
+        return PyUnicode_FromFormat("<zipimporter object \"%.300U\">",
+                                    self->archive);
 }
 
 /* return fullname.split(".")[-1] */

From python-checkins at python.org  Wed Jul 14 22:38:24 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 14 Jul 2010 22:38:24 +0200 (CEST)
Subject: [Python-checkins] r82907 -
	python/branches/import_unicode/Modules/zipimport.c
Message-ID: <20100714203824.8B936EEA30@mail.python.org>

Author: victor.stinner
Date: Wed Jul 14 22:38:24 2010
New Revision: 82907

Log:
zipimport_get_source() uses PyUnicode_DecodeFSDefault()

Modified:
   python/branches/import_unicode/Modules/zipimport.c

Modified: python/branches/import_unicode/Modules/zipimport.c
==============================================================================
--- python/branches/import_unicode/Modules/zipimport.c	(original)
+++ python/branches/import_unicode/Modules/zipimport.c	Wed Jul 14 22:38:24 2010
@@ -504,7 +504,9 @@
     toc_entry = PyDict_GetItemString(self->files, path);
     if (toc_entry != NULL) {
         PyObject *bytes = get_data(self->archive, toc_entry);
-        PyObject *res = PyUnicode_FromString(PyBytes_AsString(bytes));
+        if (bytes == NULL)
+            return NULL;
+        PyObject *res = PyUnicode_DecodeFSDefault(PyBytes_AsString(bytes));
         Py_XDECREF(bytes);
         return res;
     }

From python-checkins at python.org  Wed Jul 14 22:38:26 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 14 Jul 2010 22:38:26 +0200 (CEST)
Subject: [Python-checkins] r82908 -
	python/branches/import_unicode/Modules/zipimport.c
Message-ID: <20100714203826.A5E68CA53@mail.python.org>

Author: victor.stinner
Date: Wed Jul 14 22:38:26 2010
New Revision: 82908

Log:
get_code_from_data() uses PyUnicode_EncodeFSDefault()

Modified:
   python/branches/import_unicode/Modules/zipimport.c

Modified: python/branches/import_unicode/Modules/zipimport.c
==============================================================================
--- python/branches/import_unicode/Modules/zipimport.c	(original)
+++ python/branches/import_unicode/Modules/zipimport.c	Wed Jul 14 22:38:26 2010
@@ -1107,20 +1107,23 @@
                    time_t mtime, PyObject *toc_entry)
 {
     PyObject *data, *code;
+    PyObject *modpath_bytes;
     char *modpath;
 
     data = get_data(self->archive, toc_entry);
     if (data == NULL)
         return NULL;
 
-    modpath = _PyUnicode_AsString(PyTuple_GetItem(toc_entry, 0));
+    modpath_bytes = PyUnicode_EncodeFSDefault(PyTuple_GetItem(toc_entry, 0));
+    if (modpath_bytes == NULL)
+        return NULL;
+    modpath = PyBytes_AS_STRING(modpath_bytes);
 
-    if (isbytecode) {
+    if (isbytecode)
         code = unmarshal_code(modpath, data, mtime);
-    }
-    else {
+    else
         code = compile_source(modpath, data);
-    }
+    Py_DECREF(modpath_bytes);
     Py_DECREF(data);
     return code;
 }

From nnorwitz at gmail.com  Wed Jul 14 22:56:14 2010
From: nnorwitz at gmail.com (Neal Norwitz)
Date: Wed, 14 Jul 2010 16:56:14 -0400
Subject: [Python-checkins] Python Regression Test Failures opt (1)
Message-ID: <20100714205614.GA14077@kbk-i386-bb.psfb.org>

346 tests OK.
1 test failed:
    test_subprocess
37 tests skipped:
    test_aepack test_al test_applesingle test_bsddb185 test_bsddb3
    test_cd test_cl test_curses test_epoll test_gdb test_gl
    test_imgfile test_ioctl test_kqueue test_linuxaudiodev test_macos
    test_macostools test_multiprocessing test_ossaudiodev test_pep277
    test_py3kwarn test_scriptpackages test_smtpnet test_socketserver
    test_startfile test_sunaudiodev test_tcl test_timeout test_tk
    test_ttk_guionly test_ttk_textonly test_unicode_file
    test_urllib2net test_urllibnet test_winreg test_winsound
    test_zipfile64
7 skips unexpected on linux2:
    test_epoll test_gdb test_ioctl test_multiprocessing test_tk
    test_ttk_guionly test_ttk_textonly

== CPython 2.7 (trunk:82893M, Jul 14 2010, 16:02:54) [GCC 3.3.4 20040623 (Gentoo Linux 3.3.4-r1, ssp-3.3.2-2, pie-8.7.6)]
==   Linux-2.6.9-gentoo-r1-i686-AMD_Athlon-tm-_XP_3000+-with-gentoo-1.4.16 little-endian
==   /tmp/test_python_30391
test_grammar
test_opcodes
test_dict
test_builtin
test_exceptions
test_types
test_unittest
test_doctest
test_doctest2
test_MimeWriter
test_SimpleHTTPServer
test_StringIO
test___all__
test___future__
test__locale
test_abc
test_abstract_numbers
test_aepack
test_aepack skipped -- No module named aetypes
test_aifc
test_al
test_al skipped -- No module named al
test_anydbm
test_applesingle
test_applesingle skipped -- No module named MacOS
test_argparse
test_array
test_ascii_formatd
test_ast
test_asynchat
test_asyncore
test_atexit
test_audioop
test_augassign
test_base64
test_bastion
test_bigaddrspace
test_bigmem
test_binascii
test_binhex
test_binop
test_bisect
test_bool
test_bsddb
test_bsddb185
test_bsddb185 skipped -- No module named bsddb185
test_bsddb3
test_bsddb3 skipped -- Use of the `bsddb' resource not enabled
test_buffer
test_bufio
test_bytes
test_bz2
test_calendar
test_call
test_capi
test_cd
test_cd skipped -- No module named cd
test_cfgparser
test_cgi
test_charmapcodec
test_cl
test_cl skipped -- No module named cl
test_class
test_cmath
test_cmd
test_cmd_line
test_cmd_line_script
test_code
test_codeccallbacks
test_codecencodings_cn
test_codecencodings_hk
test_codecencodings_jp
test_codecencodings_kr
test_codecencodings_tw
test_codecmaps_cn
test_codecmaps_hk
test_codecmaps_jp
test_codecmaps_kr
test_codecmaps_tw
test_codecs
test_codeop
test_coding
test_coercion
test_collections
test_colorsys
test_commands
test_compare
test_compile
test_compileall
test_compiler
test_complex
test_complex_args
test_contains
test_contextlib
test_cookie
test_cookielib
test_copy
test_copy_reg
test_cpickle
test_cprofile
test_crypt
test_csv
test_ctypes
test_curses
test_curses skipped -- Use of the `curses' resource not enabled
test_datetime
test_dbm
test_decimal
test_decorators
test_defaultdict
test_deque
test_descr
test_descrtut
test_dictcomps
test_dictviews
test_difflib
test_dircache
test_dis
test_distutils
[19646 refs]
[19646 refs]
[19646 refs]
[19646 refs]
[19643 refs]
test_dl
test_docxmlrpc
test_dumbdbm
test_dummy_thread
test_dummy_threading
test_email
test_email_codecs
test_email_renamed
test_enumerate
test_eof
test_epoll
test_epoll skipped -- kernel doesn't support epoll()
test_errno
test_exception_variations
test_extcall
test_fcntl
test_file
test_file2k
test_filecmp
test_fileinput
test_fileio
test_float
test_fnmatch
test_fork1
test_format
test_fpformat
test_fractions
test_frozen
test_ftplib
test_funcattrs
test_functools
test_future
test_future3
test_future4
test_future5
test_future_builtins
test_gc
test_gdb
test_gdb skipped -- gdb versions before 7.0 didn't support python embedding Saw:
GNU gdb 6.2.1
Copyright 2004 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB.  Type "show warranty" for details.
This GDB was configured as "i386-pc-linux-gnu".

test_gdbm
test_generators
test_genericpath
test_genexps
test_getargs
test_getargs2
test_getopt
test_gettext
test_gl
test_gl skipped -- No module named gl
test_glob
test_global
test_grp
test_gzip
test_hash
test_hashlib
test_heapq
test_hmac
test_hotshot
test_htmllib
test_htmlparser
test_httplib
test_httpservers
[15648 refs]
[15648 refs]
[15648 refs]
[25315 refs]
test_imageop
test_imaplib
test_imgfile
test_imgfile skipped -- No module named imgfile
test_imp
test_import
test_importhooks
test_importlib
test_index
test_inspect
test_int
test_int_literal
test_io
test_ioctl
test_ioctl skipped -- Unable to open /dev/tty
test_isinstance
test_iter
test_iterlen
test_itertools
test_json
test_kqueue
test_kqueue skipped -- test works only on BSD
test_largefile
test_lib2to3
test_linecache
test_linuxaudiodev
test_linuxaudiodev skipped -- Use of the `audio' resource not enabled
test_list
test_locale
test_logging
test_long
test_long_future
test_longexp
test_macos
test_macos skipped -- No module named MacOS
test_macostools
test_macostools skipped -- No module named MacOS
test_macpath
test_mailbox
test_marshal
test_math
test_md5
test_memoryio
test_memoryview
test_mhlib
test_mimetools
test_mimetypes
test_minidom
test_mmap
test_module
test_modulefinder
test_multibytecodec
test_multibytecodec_support
test_multifile
test_multiprocessing
test_multiprocessing skipped -- This platform lacks a functioning sem_open implementation, therefore, the required synchronization primitives needed will not function, see issue 3770.
test_mutants
test_mutex
test_netrc
test_new
test_nis
test_normalization
test_ntpath
test_old_mailbox
test_openpty
test_operator
test_optparse
test_os
[15648 refs]
[15648 refs]
test_ossaudiodev
test_ossaudiodev skipped -- Use of the `audio' resource not enabled
test_parser
Expecting 's_push: parser stack overflow' in next line
s_push: parser stack overflow
test_pdb
test_peepholer
test_pep247
test_pep263
test_pep277
test_pep277 skipped -- only NT+ and systems with Unicode-friendly filesystem encoding
test_pep292
test_pep352
test_pickle
test_pickletools
test_pipes
test_pkg
test_pkgimport
test_pkgutil
test_platform
[17044 refs]
[17044 refs]
test_plistlib
test_poll
test_popen
[15653 refs]
[15653 refs]
[15653 refs]
test_popen2
test_poplib
test_posix
test_posixpath
test_pow
test_pprint
test_print
test_profile
test_profilehooks
test_property
test_pstats
test_pty
test_pwd
test_py3kwarn
test_py3kwarn skipped -- test.test_py3kwarn must be run with the -3 flag
test_pyclbr
test_pydoc
[20862 refs]
[20862 refs]
[20862 refs]
[20862 refs]
[20862 refs]
[20861 refs]
[20861 refs]
test_pyexpat
test_queue
test_quopri
[18479 refs]
[18479 refs]
test_random
test_re
test_readline
test_repr
test_resource
test_rfc822
test_richcmp
test_robotparser
test_runpy
test_sax
test_scope
test_scriptpackages
test_scriptpackages skipped -- No module named aetools
test_select
test_set
test_setcomps
test_sets
test_sgmllib
test_sha
test_shelve
test_shlex
test_shutil
test_signal
test_site
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
test_slice
test_smtplib
test_smtpnet
test_smtpnet skipped -- Use of the `network' resource not enabled
test_socket
test_socketserver
test_socketserver skipped -- Use of the `network' resource not enabled
test_softspace
test_sort
test_sqlite
test_ssl
test_startfile
test_startfile skipped -- module os has no attribute startfile
test_str
test_strftime
test_string
test_stringprep
test_strop
test_strptime
test_strtod
test_struct
test_structmembers
test_structseq
test_subprocess
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15863 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
.
[15648 refs]
[15648 refs]
    this bit of output is from a test of stdout in a different process ...
[15648 refs]
[15648 refs]
[15863 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15863 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
[15648 refs]
.
[15648 refs]
[15648 refs]
    this bit of output is from a test of stdout in a different process ...
[15648 refs]
[15648 refs]
[15863 refs]
test test_subprocess failed -- multiple errors occurred; run in verbose mode for details
Re-running test 'test_subprocess' in verbose mode
test_call_kwargs (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_call_seq (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_check_call_nonzero (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_check_call_zero (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_check_output (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_check_output_nonzero (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_check_output_stderr (test.test_subprocess.ProcessTestCase) ... ERROR
test_check_output_stdout_arg (test.test_subprocess.ProcessTestCase) ... ERROR
test_communicate (test.test_subprocess.ProcessTestCase) ... ERROR
test_communicate_pipe_buf (test.test_subprocess.ProcessTestCase) ... ERROR
test_communicate_pipe_fd_leak (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_communicate_returns (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_communicate_stderr (test.test_subprocess.ProcessTestCase) ... ERROR
test_communicate_stdin (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_communicate_stdout (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_cwd (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_env (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_executable_with_cwd (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_executable_without_cwd (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_invalid_bufsize (test.test_subprocess.ProcessTestCase) ... ERROR
test_leaking_fds_on_error (test.test_subprocess.ProcessTestCase) ... ERROR
test_list2cmdline (test.test_subprocess.ProcessTestCase) ... ERROR
test_no_leaking (test.test_subprocess.ProcessTestCase) ... ERROR
test_poll (test.test_subprocess.ProcessTestCase) ... [15863 refs]
ERROR
test_stderr_filedes (test.test_subprocess.ProcessTestCase) ... ERROR
test_stderr_fileobj (test.test_subprocess.ProcessTestCase) ... ERROR
test_stderr_none (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_stderr_pipe (test.test_subprocess.ProcessTestCase) ... ERROR
test_stdin_filedes (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_stdin_fileobj (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_stdin_none (test.test_subprocess.ProcessTestCase) ... ERROR
test_stdin_pipe (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_stdout_filedes (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_stdout_filedes_of_stdout (test.test_subprocess.ProcessTestCase) ... .
[15648 refs]
ERROR
test_stdout_fileobj (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_stdout_none (test.test_subprocess.ProcessTestCase) ...     this bit of output is from a test of stdout in a different process ...
ERROR
test_stdout_pipe (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_stdout_stderr_file (test.test_subprocess.ProcessTestCase) ... ERROR
test_stdout_stderr_pipe (test.test_subprocess.ProcessTestCase) ... ERROR
test_universal_newlines (test.test_subprocess.ProcessTestCase) ... [15648 refs]
ERROR
test_universal_newlines_communicate (test.test_subprocess.ProcessTestCase) ... ERROR
test_wait (test.test_subprocess.ProcessTestCase) ... [15863 refs]
ERROR
test_writes_before_communicate (test.test_subprocess.ProcessTestCase) ... ERROR
test_args_string (test.test_subprocess.POSIXProcessTestCase) ... [15648 refs]
ERROR
test_call_string (test.test_subprocess.POSIXProcessTestCase) ... [15648 refs]
ERROR
test_exceptions (test.test_subprocess.POSIXProcessTestCase) ... ERROR
test_invalid_args (test.test_subprocess.POSIXProcessTestCase) ... ERROR
test_kill (test.test_subprocess.POSIXProcessTestCase) ... ERROR
test_preexec (test.test_subprocess.POSIXProcessTestCase) ... [15648 refs]
ERROR
test_run_abort (test.test_subprocess.POSIXProcessTestCase) ... ERROR
test_send_signal (test.test_subprocess.POSIXProcessTestCase) ... ERROR
test_shell_sequence (test.test_subprocess.POSIXProcessTestCase) ... ERROR
test_shell_string (test.test_subprocess.POSIXProcessTestCase) ... ERROR
test_terminate (test.test_subprocess.POSIXProcessTestCase) ... ERROR
test_call_string (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_close_fds (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_creationflags (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_invalid_args (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_kill (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_send_signal (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_shell_sequence (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_shell_string (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_startupinfo (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_terminate (test.test_subprocess.Win32ProcessTestCase) ... skipped 'Windows specific tests'
test_call_kwargs (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_call_seq (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_check_call_nonzero (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_check_call_zero (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_check_output (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_check_output_nonzero (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_check_output_stderr (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_check_output_stdout_arg (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_communicate (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_communicate_pipe_buf (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_communicate_pipe_fd_leak (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_communicate_returns (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_communicate_stderr (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_communicate_stdin (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_communicate_stdout (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_cwd (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_env (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_executable_with_cwd (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_executable_without_cwd (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_invalid_bufsize (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_leaking_fds_on_error (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_list2cmdline (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_no_leaking (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_poll (test.test_subprocess.ProcessTestCaseNoPoll) ... [15863 refs]
ERROR
test_stderr_filedes (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_stderr_fileobj (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_stderr_none (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_stderr_pipe (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_stdin_filedes (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_stdin_fileobj (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_stdin_none (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_stdin_pipe (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_stdout_filedes (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_stdout_filedes_of_stdout (test.test_subprocess.ProcessTestCaseNoPoll) ... .
[15648 refs]
ERROR
test_stdout_fileobj (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_stdout_none (test.test_subprocess.ProcessTestCaseNoPoll) ...     this bit of output is from a test of stdout in a different process ...
ERROR
test_stdout_pipe (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_stdout_stderr_file (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_stdout_stderr_pipe (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_universal_newlines (test.test_subprocess.ProcessTestCaseNoPoll) ... [15648 refs]
ERROR
test_universal_newlines_communicate (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_wait (test.test_subprocess.ProcessTestCaseNoPoll) ... [15863 refs]
ERROR
test_writes_before_communicate (test.test_subprocess.ProcessTestCaseNoPoll) ... ERROR
test_eintr_retry_call (test.test_subprocess.HelperFunctionTests) ... ok

======================================================================
ERROR: test_call_kwargs (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_call_seq (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_call_nonzero (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_call_zero (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_output (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_output_nonzero (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_output_stderr (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_output_stdout_arg (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_pipe_buf (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_pipe_fd_leak (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_returns (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_stderr (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_stdin (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_stdout (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_cwd (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_env (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_executable_with_cwd (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_executable_without_cwd (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_invalid_bufsize (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_leaking_fds_on_error (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_list2cmdline (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_no_leaking (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_poll (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stderr_filedes (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stderr_fileobj (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stderr_none (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stderr_pipe (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdin_filedes (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdin_fileobj (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdin_none (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdin_pipe (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_filedes (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_filedes_of_stdout (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_fileobj (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_none (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_pipe (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_stderr_file (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_stderr_pipe (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_universal_newlines (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_universal_newlines_communicate (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_wait (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_writes_before_communicate (test.test_subprocess.ProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_args_string (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_call_string (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_exceptions (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_invalid_args (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_kill (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_preexec (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_run_abort (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_send_signal (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_shell_sequence (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_shell_string (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_terminate (test.test_subprocess.POSIXProcessTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_call_kwargs (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_call_seq (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_call_nonzero (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_call_zero (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_output (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_output_nonzero (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_output_stderr (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_check_output_stdout_arg (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_pipe_buf (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_pipe_fd_leak (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_returns (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_stderr (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_stdin (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_communicate_stdout (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_cwd (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_env (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_executable_with_cwd (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_executable_without_cwd (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_invalid_bufsize (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_leaking_fds_on_error (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_list2cmdline (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_no_leaking (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_poll (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stderr_filedes (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stderr_fileobj (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stderr_none (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stderr_pipe (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdin_filedes (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdin_fileobj (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdin_none (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdin_pipe (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_filedes (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_filedes_of_stdout (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_fileobj (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_none (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_pipe (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_stderr_file (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_stdout_stderr_pipe (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_universal_newlines (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_universal_newlines_communicate (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_wait (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

======================================================================
ERROR: test_writes_before_communicate (test.test_subprocess.ProcessTestCaseNoPoll)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 800, in tearDown
    ProcessTestCase.tearDown(self)
  File "/tmp/python-test/local/lib/python2.7/test/test_subprocess.py", line 46, in tearDown
    self.assertFalse(subprocess._active, "subprocess._active not empty")
AssertionError: subprocess._active not empty

----------------------------------------------------------------------
Ran 108 tests in 176.956s

FAILED (errors=97, skipped=10)
test test_subprocess failed -- multiple errors occurred
test_sunaudiodev
test_sunaudiodev skipped -- No module named sunaudiodev
test_sundry
test_symtable
test_syntax
test_sys
[15648 refs]
[15648 refs]
[15648 refs]
[15877 refs]
[15671 refs]
test_sysconfig
[15648 refs]
[15648 refs]
test_tarfile
test_tcl
test_tcl skipped -- No module named _tkinter
test_telnetlib
test_tempfile
[15648 refs]
test_textwrap
test_thread
test_threaded_import
test_threadedtempfile
test_threading
[18948 refs]
[20223 refs]
[20035 refs]
[20035 refs]
[20035 refs]
[20035 refs]
test_threading_local
test_threadsignals
test_time
test_timeout
test_timeout skipped -- Use of the `network' resource not enabled
test_tk
test_tk skipped -- No module named _tkinter
test_tokenize
test_trace
test_traceback
test_transformer
test_ttk_guionly
test_ttk_guionly skipped -- No module named _tkinter
test_ttk_textonly
test_ttk_textonly skipped -- No module named _tkinter
test_tuple
test_typechecks
test_ucn
test_unary
test_undocumented_details
test_unicode
test_unicode_file
test_unicode_file skipped -- No Unicode filesystem semantics on this platform.
test_unicodedata
test_univnewlines
test_univnewlines2k
test_unpack
test_urllib
test_urllib2
test_urllib2_localnet
test_urllib2net
test_urllib2net skipped -- Use of the `network' resource not enabled
test_urllibnet
test_urllibnet skipped -- Use of the `network' resource not enabled
test_urlparse
test_userdict
test_userlist
test_userstring
test_uu
test_uuid
test_wait3
test_wait4
test_warnings
[15679 refs]
[15679 refs]
[15672 refs]
[15679 refs]
[15679 refs]
[15672 refs]
test_wave
test_weakref
test_weakset
test_whichdb
test_winreg
test_winreg skipped -- No module named _winreg
test_winsound
test_winsound skipped -- No module named winsound
test_with
test_wsgiref
test_xdrlib
test_xml_etree
test_xml_etree_c
test_xmllib
test_xmlrpc
test_xpickle
test_xpickle -- skipping backwards compat tests.
Use 'regrtest.py -u xpickle' to run them.
test_xrange
test_zipfile
test_zipfile64
test_zipfile64 skipped -- test requires loads of disk-space bytes and a long time to run
test_zipimport
test_zipimport_support
test_zlib
346 tests OK.
1 test failed:
    test_subprocess
37 tests skipped:
    test_aepack test_al test_applesingle test_bsddb185 test_bsddb3
    test_cd test_cl test_curses test_epoll test_gdb test_gl
    test_imgfile test_ioctl test_kqueue test_linuxaudiodev test_macos
    test_macostools test_multiprocessing test_ossaudiodev test_pep277
    test_py3kwarn test_scriptpackages test_smtpnet test_socketserver
    test_startfile test_sunaudiodev test_tcl test_timeout test_tk
    test_ttk_guionly test_ttk_textonly test_unicode_file
    test_urllib2net test_urllibnet test_winreg test_winsound
    test_zipfile64
7 skips unexpected on linux2:
    test_epoll test_gdb test_ioctl test_multiprocessing test_tk
    test_ttk_guionly test_ttk_textonly
[965814 refs]

From alexander.belopolsky at gmail.com  Wed Jul 14 23:35:05 2010
From: alexander.belopolsky at gmail.com (Alexander Belopolsky)
Date: Wed, 14 Jul 2010 17:35:05 -0400
Subject: [Python-checkins] r82899 -
	python/branches/py3k/Doc/reference/expressions.rst
In-Reply-To: <20100714203003.11A47EE99E@mail.python.org>
References: <20100714203003.11A47EE99E@mail.python.org>
Message-ID: <AANLkTil9u2cdIU9W3pBzUpH17-zngaMAvWXwQGVLIgsN@mail.gmail.com>

On Wed, Jul 14, 2010 at 4:30 PM, senthil.kumaran
<python-checkins at python.org> wrote:
..
> +* Mappings (dictionaries) compare equal if and only if they have the same

Hmm, to me *the same* implies identity (a is b) rather than equality
(a == b).  No, I cannot think of a better wording that is still
concise.

From solipsis at pitrou.net  Thu Jul 15 01:24:48 2010
From: solipsis at pitrou.net (solipsis at pitrou.net)
Date: Thu, 15 Jul 2010 01:24:48 +0200 (CEST)
Subject: [Python-checkins] Daily py3k reference leaks (r82899): sum=0
Message-ID: <20100714232448.7ABC11771C@ns6635.ovh.net>

py3k results for svn r82899 (hg cset 7ca5f460acab)
--------------------------------------------------



Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/antoine/py3k/refleaks/reflog4C_wVP', '-x']

From python-checkins at python.org  Thu Jul 15 01:38:05 2010
From: python-checkins at python.org (tarek.ziade)
Date: Thu, 15 Jul 2010 01:38:05 +0200
Subject: [Python-checkins] distutils2: Update tests to use the new API
Message-ID: <E1OZBWf-0001GB-8K@dinsdale.python.org>

tarek.ziade pushed 618d28bfd736 to distutils2:

http://hg.python.org/distutils2/rev/618d28bfd736
changeset:   351:618d28bfd736
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Mon Jul 05 11:09:27 2010 +0200
summary:     Update tests to use the new API
files:       src/distutils2/tests/test_pypi_simple.py

diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -7,7 +7,7 @@
 import tempfile
 import urllib2
 
-from distutils2.tests import support
+from distutils2.tests import support, run_unittest
 from distutils2.tests.support import unittest
 from distutils2.tests.pypi_server import use_pypi_server, PyPIServer, \
                                          PYPI_DEFAULT_STATIC_PATH
@@ -84,7 +84,7 @@
             url = 'http://example.com'
             page = ('<a href="http://www.famfamfam.com]('
                     'http://www.famfamfam.com/">')
-            index.process_index(url, page)
+            index._process_url(url, page)
 
     @use_pypi_server("test_found_links")
     def test_found_links(self, server):
@@ -214,9 +214,6 @@
         self.assertIn("%s/simple/foobar/" % server.full_address,
             index._processed_urls)  # it's the simple index page
 
-        # FIXME
-        return
-
         self.assertIn("%s/external/homepage.html" % server.full_address,
             index._processed_urls)  # the external homepage is rel="homepage"
         self.assertNotIn("%s/external/nonrel.html" % server.full_address,

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Thu Jul 15 01:38:05 2010
From: python-checkins at python.org (tarek.ziade)
Date: Thu, 15 Jul 2010 01:38:05 +0200
Subject: [Python-checkins] distutils2: Update some tests to use the new API
Message-ID: <E1OZBWf-0001GE-8q@dinsdale.python.org>

tarek.ziade pushed 22ca522e9915 to distutils2:

http://hg.python.org/distutils2/rev/22ca522e9915
changeset:   352:22ca522e9915
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Mon Jul 05 11:15:34 2010 +0200
summary:     Update some tests to use the new API
files:       src/distutils2/tests/test_pypi_simple.py

diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -213,7 +213,6 @@
         # and rel="homepage"
         self.assertIn("%s/simple/foobar/" % server.full_address,
             index._processed_urls)  # it's the simple index page
-
         self.assertIn("%s/external/homepage.html" % server.full_address,
             index._processed_urls)  # the external homepage is rel="homepage"
         self.assertNotIn("%s/external/nonrel.html" % server.full_address,

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Thu Jul 15 01:38:05 2010
From: python-checkins at python.org (tarek.ziade)
Date: Thu, 15 Jul 2010 01:38:05 +0200
Subject: [Python-checkins] distutils2: merged merwok fixes
Message-ID: <E1OZBWf-0001GN-AY@dinsdale.python.org>

tarek.ziade pushed 473fe779eb61 to distutils2:

http://hg.python.org/distutils2/rev/473fe779eb61
changeset:   354:473fe779eb61
parent:      352:22ca522e9915
parent:      353:7b78486d4c00
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Wed Jul 07 01:30:39 2010 +0200
summary:     merged merwok fixes
files:       

diff --git a/src/distutils2/_backport/tests/test_pkgutil.py b/src/distutils2/_backport/tests/test_pkgutil.py
--- a/src/distutils2/_backport/tests/test_pkgutil.py
+++ b/src/distutils2/_backport/tests/test_pkgutil.py
@@ -1,7 +1,5 @@
 # -*- coding: utf-8 -*-
 """Tests for PEP 376 pkgutil functionality"""
-import unittest2
-import unittest2.compatibility
 import sys
 import os
 import csv
@@ -17,6 +15,15 @@
 from test.test_support import run_unittest, TESTFN
 
 from distutils2._backport import pkgutil
+from distutils2.tests.support import unittest
+
+try:
+    from os.path import relpath
+except ImportError:
+    try:
+        from unittest.compatibility import relpath
+    except ImportError:
+        from unittest2.compatibility import relpath
 
 # TODO Add a test for getting a distribution that is provided by another
 #   distribution.
@@ -24,7 +31,7 @@
 # TODO Add a test for absolute pathed RECORD items (e.g. /etc/myapp/config.ini)
 
 # Adapted from Python 2.7's trunk
-class TestPkgUtilData(unittest2.TestCase):
+class TestPkgUtilData(unittest.TestCase):
 
     def setUp(self):
         self.dirname = tempfile.mkdtemp()
@@ -92,7 +99,7 @@
         del sys.modules[pkg]
 
 # Adapted from Python 2.7's trunk
-class TestPkgUtilPEP302(unittest2.TestCase):
+class TestPkgUtilPEP302(unittest.TestCase):
 
     class MyTestLoader(object):
         def load_module(self, fullname):
@@ -135,7 +142,7 @@
         del sys.modules['foo']
 
 
-class TestPkgUtilDistribution(unittest2.TestCase):
+class TestPkgUtilDistribution(unittest.TestCase):
     """Tests the pkgutil.Distribution class"""
 
     def setUp(self):
@@ -152,7 +159,7 @@
             return md5_hash.hexdigest()
 
         def record_pieces(file):
-            path = os.path.relpath(file, sys.prefix)
+            path = relpath(file, sys.prefix)
             digest = get_hexdigest(file)
             size = os.path.getsize(file)
             return [path, digest, size]
@@ -172,7 +179,7 @@
             for file in ['INSTALLER', 'METADATA', 'REQUESTED']:
                 record_writer.writerow(record_pieces(
                     os.path.join(distinfo_dir, file)))
-            record_writer.writerow([os.path.relpath(record_file, sys.prefix)])
+            record_writer.writerow([relpath(record_file, sys.prefix)])
             del record_writer # causes the RECORD file to close
             record_reader = csv.reader(open(record_file, 'rb'))
             record_data = []
@@ -227,10 +234,10 @@
             distinfo_name + '.dist-info')
         true_path = [self.fake_dists_path, distinfo_name, \
                      'grammar', 'utils.py']
-        true_path = os.path.relpath(os.path.join(*true_path), sys.prefix)
+        true_path = relpath(os.path.join(*true_path), sys.prefix)
         false_path = [self.fake_dists_path, 'towel_stuff-0.1', 'towel_stuff',
             '__init__.py']
-        false_path = os.path.relpath(os.path.join(*false_path), sys.prefix)
+        false_path = relpath(os.path.join(*false_path), sys.prefix)
 
         # Test if the distribution uses the file in question
         from distutils2._backport.pkgutil import Distribution
@@ -289,7 +296,7 @@
         self.assertEqual(sorted(found), sorted(distinfo_record_paths))
 
 
-class TestPkgUtilPEP376(unittest2.TestCase):
+class TestPkgUtilPEP376(unittest.TestCase):
     """Tests for the new functionality added in PEP 376."""
 
     def setUp(self):
@@ -528,12 +535,12 @@
 
 
 def test_suite():
-    suite = unittest2.TestSuite()
-    testcase_loader = unittest2.loader.defaultTestLoader.loadTestsFromTestCase
-    suite.addTest(testcase_loader(TestPkgUtilData))
-    suite.addTest(testcase_loader(TestPkgUtilDistribution))
-    suite.addTest(testcase_loader(TestPkgUtilPEP302))
-    suite.addTest(testcase_loader(TestPkgUtilPEP376))
+    suite = unittest.TestSuite()
+    load = unittest.defaultTestLoader.loadTestsFromTestCase
+    suite.addTest(load(TestPkgUtilData))
+    suite.addTest(load(TestPkgUtilDistribution))
+    suite.addTest(load(TestPkgUtilPEP302))
+    suite.addTest(load(TestPkgUtilPEP376))
     return suite
 
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Thu Jul 15 01:38:05 2010
From: python-checkins at python.org (tarek.ziade)
Date: Thu, 15 Jul 2010 01:38:05 +0200
Subject: [Python-checkins] distutils2: Make tests pass and restore unittest{,
	2} compat
Message-ID: <E1OZBWf-0001GI-9k@dinsdale.python.org>

tarek.ziade pushed 7b78486d4c00 to distutils2:

http://hg.python.org/distutils2/rev/7b78486d4c00
changeset:   353:7b78486d4c00
parent:      350:8b73f08d1d8d
user:        ?ric Araujo <merwok at netwok.org>
date:        Tue Jul 06 22:16:12 2010 +0200
summary:     Make tests pass and restore unittest{,2} compat
files:       src/distutils2/_backport/tests/test_pkgutil.py

diff --git a/src/distutils2/_backport/tests/test_pkgutil.py b/src/distutils2/_backport/tests/test_pkgutil.py
--- a/src/distutils2/_backport/tests/test_pkgutil.py
+++ b/src/distutils2/_backport/tests/test_pkgutil.py
@@ -1,7 +1,5 @@
 # -*- coding: utf-8 -*-
 """Tests for PEP 376 pkgutil functionality"""
-import unittest2
-import unittest2.compatibility
 import sys
 import os
 import csv
@@ -17,6 +15,15 @@
 from test.test_support import run_unittest, TESTFN
 
 from distutils2._backport import pkgutil
+from distutils2.tests.support import unittest
+
+try:
+    from os.path import relpath
+except ImportError:
+    try:
+        from unittest.compatibility import relpath
+    except ImportError:
+        from unittest2.compatibility import relpath
 
 # TODO Add a test for getting a distribution that is provided by another
 #   distribution.
@@ -24,7 +31,7 @@
 # TODO Add a test for absolute pathed RECORD items (e.g. /etc/myapp/config.ini)
 
 # Adapted from Python 2.7's trunk
-class TestPkgUtilData(unittest2.TestCase):
+class TestPkgUtilData(unittest.TestCase):
 
     def setUp(self):
         self.dirname = tempfile.mkdtemp()
@@ -92,7 +99,7 @@
         del sys.modules[pkg]
 
 # Adapted from Python 2.7's trunk
-class TestPkgUtilPEP302(unittest2.TestCase):
+class TestPkgUtilPEP302(unittest.TestCase):
 
     class MyTestLoader(object):
         def load_module(self, fullname):
@@ -135,7 +142,7 @@
         del sys.modules['foo']
 
 
-class TestPkgUtilDistribution(unittest2.TestCase):
+class TestPkgUtilDistribution(unittest.TestCase):
     """Tests the pkgutil.Distribution class"""
 
     def setUp(self):
@@ -152,7 +159,7 @@
             return md5_hash.hexdigest()
 
         def record_pieces(file):
-            path = os.path.relpath(file, sys.prefix)
+            path = relpath(file, sys.prefix)
             digest = get_hexdigest(file)
             size = os.path.getsize(file)
             return [path, digest, size]
@@ -172,7 +179,7 @@
             for file in ['INSTALLER', 'METADATA', 'REQUESTED']:
                 record_writer.writerow(record_pieces(
                     os.path.join(distinfo_dir, file)))
-            record_writer.writerow([os.path.relpath(record_file, sys.prefix)])
+            record_writer.writerow([relpath(record_file, sys.prefix)])
             del record_writer # causes the RECORD file to close
             record_reader = csv.reader(open(record_file, 'rb'))
             record_data = []
@@ -227,10 +234,10 @@
             distinfo_name + '.dist-info')
         true_path = [self.fake_dists_path, distinfo_name, \
                      'grammar', 'utils.py']
-        true_path = os.path.relpath(os.path.join(*true_path), sys.prefix)
+        true_path = relpath(os.path.join(*true_path), sys.prefix)
         false_path = [self.fake_dists_path, 'towel_stuff-0.1', 'towel_stuff',
             '__init__.py']
-        false_path = os.path.relpath(os.path.join(*false_path), sys.prefix)
+        false_path = relpath(os.path.join(*false_path), sys.prefix)
 
         # Test if the distribution uses the file in question
         from distutils2._backport.pkgutil import Distribution
@@ -289,7 +296,7 @@
         self.assertEqual(sorted(found), sorted(distinfo_record_paths))
 
 
-class TestPkgUtilPEP376(unittest2.TestCase):
+class TestPkgUtilPEP376(unittest.TestCase):
     """Tests for the new functionality added in PEP 376."""
 
     def setUp(self):
@@ -528,12 +535,12 @@
 
 
 def test_suite():
-    suite = unittest2.TestSuite()
-    testcase_loader = unittest2.loader.defaultTestLoader.loadTestsFromTestCase
-    suite.addTest(testcase_loader(TestPkgUtilData))
-    suite.addTest(testcase_loader(TestPkgUtilDistribution))
-    suite.addTest(testcase_loader(TestPkgUtilPEP302))
-    suite.addTest(testcase_loader(TestPkgUtilPEP376))
+    suite = unittest.TestSuite()
+    load = unittest.defaultTestLoader.loadTestsFromTestCase
+    suite.addTest(load(TestPkgUtilData))
+    suite.addTest(load(TestPkgUtilDistribution))
+    suite.addTest(load(TestPkgUtilPEP302))
+    suite.addTest(load(TestPkgUtilPEP376))
     return suite
 
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Thu Jul 15 01:38:05 2010
From: python-checkins at python.org (tarek.ziade)
Date: Thu, 15 Jul 2010 01:38:05 +0200
Subject: [Python-checkins] distutils2: make the code more pythonic
Message-ID: <E1OZBWf-0001GR-BQ@dinsdale.python.org>

tarek.ziade pushed 2e619c2efa22 to distutils2:

http://hg.python.org/distutils2/rev/2e619c2efa22
changeset:   355:2e619c2efa22
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Wed Jul 07 02:36:45 2010 +0200
summary:     make the code more pythonic
files:       src/distutils2/pypi/dist.py, src/distutils2/pypi/simple.py, src/distutils2/tests/test_pypi_dist.py

diff --git a/src/distutils2/pypi/dist.py b/src/distutils2/pypi/dist.py
--- a/src/distutils2/pypi/dist.py
+++ b/src/distutils2/pypi/dist.py
@@ -31,37 +31,6 @@
     information about distributions.
     """
 
-    @classmethod
-    def from_url(cls, url, probable_dist_name=None, is_external=True):
-        """Build a Distribution from a url archive (egg or zip or tgz).
-
-        :param url: complete url of the distribution
-        :param probable_dist_name: A probable name of the distribution.
-        :param is_external: Tell if the url commes from an index or from
-                            an external URL.
-        """
-        # if the url contains a md5 hash, get it.
-        md5_hash = None
-        match = MD5_HASH.match(url)
-        if match is not None:
-            md5_hash = match.group(1)
-            # remove the hash
-            url = url.replace("#md5=%s" % md5_hash, "")
-
-        # parse the archive name to find dist name and version
-        archive_name = urlparse.urlparse(url)[2].split('/')[-1]
-        extension_matched = False
-        # remove the extension from the name
-        for ext in EXTENSIONS:
-            if archive_name.endswith(ext):
-                archive_name = archive_name[:-len(ext)]
-                extension_matched = True
-
-        name, version = split_archive_name(archive_name)
-        if extension_matched is True:
-            return PyPIDistribution(name, version, url=url, url_hashname="md5",
-                                    url_hashval=md5_hash, url_is_external=is_external)
-
     def __init__(self, name, version, type=None, url=None, url_hashname=None,
                  url_hashval=None, url_is_external=True):
         """Create a new instance of PyPIDistribution.
@@ -260,24 +229,53 @@
         else:
             super(PyPIDistributions, self).append(o)
 
-    def sort_distributions(self, prefer_source=None, prefer_final=None,
+    def sort_distributions(self, prefer_source=True, prefer_final=False,
                            reverse=True, *args, **kwargs):
         """order the results with the given properties"""
 
         sort_by = []
-        if prefer_final is not None:
-            if prefer_final is True:
-                sort_by.append("is_final")
+        if prefer_final:
+            sort_by.append("is_final")
         sort_by.append("version")
 
-        if prefer_source is not None:
-            if prefer_source is True:
-                sort_by.append("is_source")
+        if prefer_source:
+            sort_by.append("is_source")
 
         super(PyPIDistributions, self).sort(
             key=lambda i: [getattr(i, arg) for arg in sort_by],
             reverse=reverse, *args, **kwargs)
 
+def create_from_url(url, probable_dist_name=None,
+                                 is_external=True):
+    """Build a Distribution from a url archive (egg or zip or tgz).
+
+    :param url: complete url of the distribution
+    :param probable_dist_name: A probable name of the distribution.
+    :param is_external: Tell if the url commes from an index or from
+                        an external URL.
+    """
+    # if the url contains a md5 hash, get it.
+    md5_hash = None
+    match = MD5_HASH.match(url)
+    if match is not None:
+        md5_hash = match.group(1)
+        # remove the hash
+        url = url.replace("#md5=%s" % md5_hash, "")
+
+    # parse the archive name to find dist name and version
+    archive_name = urlparse.urlparse(url)[2].split('/')[-1]
+    extension_matched = False
+    # remove the extension from the name
+    for ext in EXTENSIONS:
+        if archive_name.endswith(ext):
+            archive_name = archive_name[:-len(ext)]
+            extension_matched = True
+
+    name, version = split_archive_name(archive_name)
+    if extension_matched is True:
+        return PyPIDistribution(name, version, url=url, url_hashname="md5",
+                                url_hashval=md5_hash, url_is_external=is_external)
+
 def split_archive_name(archive_name, probable_name=None):
     """Split an archive name into two parts: name and version.
 
diff --git a/src/distutils2/pypi/simple.py b/src/distutils2/pypi/simple.py
--- a/src/distutils2/pypi/simple.py
+++ b/src/distutils2/pypi/simple.py
@@ -5,18 +5,18 @@
 avalaible at http://pypi.python.org/simple/
 """
 from fnmatch import translate
+import httplib
+import re
+import socket
+import sys
+import urllib2
 import urlparse
-import sys
-import re
-import urllib2
-import httplib
-import socket
 
 from distutils2.version import VersionPredicate
-from distutils2.pypi.dist import PyPIDistribution, PyPIDistributions, \
-    EXTENSIONS
-from distutils2.pypi.errors import PyPIError, DistributionNotFound, \
-    DownloadError, UnableToDownload
+from distutils2.pypi.dist import (PyPIDistribution, PyPIDistributions,
+                                  create_from_url, EXTENSIONS)
+from distutils2.pypi.errors import (PyPIError, DistributionNotFound,
+                                    DownloadError, UnableToDownload)
 from distutils2 import __version__ as __distutils2_version__
 
 # -- Constants -----------------------------------------------
@@ -60,33 +60,31 @@
 
 class SimpleIndex(object):
     """Provides useful tools to request the Python Package Index simple API
+
+    :param index_url: the url of the simple index to search on.
+    :param follow_externals: tell if following external links is needed or
+                             not. Default is False.
+    :param hosts: a list of hosts allowed to be processed while using
+                  follow_externals=True. Default behavior is to follow all
+                  hosts.
+    :param follow_externals: tell if following external links is needed or
+                             not. Default is False.
+    :param prefer_source: if there is binary and source distributions, the
+                          source prevails.
+    :param prefer_final: if the version is not mentioned, and the last
+                         version is not a "final" one (alpha, beta, etc.),
+                         pick up the last final version.
+    :param mirrors_url: the url to look on for DNS records giving mirror
+                        adresses.
+    :param mirrors: a list of mirrors to check out if problems
+                         occurs while working with the one given in "url"
+    :param timeout: time in seconds to consider a url has timeouted.
     """
 
     def __init__(self, index_url=PYPI_DEFAULT_INDEX_URL, hosts=DEFAULT_HOSTS,
                  follow_externals=False, prefer_source=True,
                  prefer_final=False, mirrors_url=PYPI_DEFAULT_MIRROR_URL,
                  mirrors=None, timeout=SOCKET_TIMEOUT):
-        """Class constructor.
-
-        :param index_url: the url of the simple index to search on.
-        :param follow_externals: tell if following external links is needed or
-                                 not. Default is False.
-        :param hosts: a list of hosts allowed to be processed while using
-                      follow_externals=True. Default behavior is to follow all
-                      hosts.
-        :param follow_externals: tell if following external links is needed or
-                                 not. Default is False.
-        :param prefer_source: if there is binary and source distributions, the
-                              source prevails.
-        :param prefer_final: if the version is not mentioned, and the last
-                             version is not a "final" one (alpha, beta, etc.),
-                             pick up the last final version.
-        :param mirrors_url: the url to look on for DNS records giving mirror
-                            adresses.
-        :param mirrors: a list of mirrors to check out if problems
-                             occurs while working with the one given in "url"
-        :param timeout: time in seconds to consider a url has timeouted.
-        """
         self.follow_externals = follow_externals
 
         if not index_url.endswith("/"):
@@ -213,7 +211,7 @@
         # local files are always considered browable resources
         if self.index_url in url or urlparse.urlparse(url)[0] == "file":
             return True
-        elif self.follow_externals is True:
+        elif self.follow_externals:
             if self._allowed_hosts(urlparse.urlparse(url)[1]):  # 1 is netloc
                 return True
             else:
@@ -266,7 +264,7 @@
                     if self._is_distribution(link) or is_download:
                         self._processed_urls.append(link)
                         # it's a distribution, so create a dist object
-                        dist = PyPIDistribution.from_url(link, project_name,
+                        dist = create_from_url(link, project_name,
                                     is_external=not self.index_url in url)
                         self._register_dist(dist)
                     else:
diff --git a/src/distutils2/tests/test_pypi_dist.py b/src/distutils2/tests/test_pypi_dist.py
--- a/src/distutils2/tests/test_pypi_dist.py
+++ b/src/distutils2/tests/test_pypi_dist.py
@@ -11,7 +11,7 @@
 from distutils2.pypi.errors import HashDoesNotMatch, UnsupportedHashName
 from distutils2.pypi.dist import (PyPIDistribution as Dist,
                                   PyPIDistributions as Dists,
-                                  split_archive_name)
+                                  split_archive_name, create_from_url)
 
 
 class TestPyPIDistribution(support.TempdirManager,
@@ -25,7 +25,7 @@
         self.assertEqual("FooBar", dist.name)
         self.assertEqual("1.1", "%s" % dist.version)
 
-    def test_from_url(self):
+    def test_create_from_url(self):
         """Test that the Distribution object can be built from a single URL"""
         url_list = {
             'FooBar-1.1.0.tar.gz': {
@@ -55,7 +55,7 @@
         }
 
         for url, attributes in url_list.items():
-            dist = Dist.from_url("http://test.tld/" + url)
+            dist = create_from_url("http://test.tld/" + url)
             for attribute, value in attributes.items():
                 if isinstance(value, dict):
                     mylist = getattr(dist, attribute)

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Thu Jul 15 01:38:05 2010
From: python-checkins at python.org (tarek.ziade)
Date: Thu, 15 Jul 2010 01:38:05 +0200
Subject: [Python-checkins] distutils2: Switch from docstrings to comments in
	tests, to uniformize output.
Message-ID: <E1OZBWf-0001GV-CC@dinsdale.python.org>

tarek.ziade pushed d7db403b144a to distutils2:

http://hg.python.org/distutils2/rev/d7db403b144a
changeset:   356:d7db403b144a
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Wed Jul 07 04:35:06 2010 +0200
summary:     Switch from docstrings to comments in tests, to uniformize output.
files:       src/distutils2/_backport/tests/test_pkgutil.py, src/distutils2/_backport/tests/test_sysconfig.py, src/distutils2/tests/test_pypi_dist.py, src/distutils2/tests/test_pypi_server.py, src/distutils2/tests/test_pypi_simple.py

diff --git a/src/distutils2/_backport/tests/test_pkgutil.py b/src/distutils2/_backport/tests/test_pkgutil.py
--- a/src/distutils2/_backport/tests/test_pkgutil.py
+++ b/src/distutils2/_backport/tests/test_pkgutil.py
@@ -26,7 +26,7 @@
         from unittest2.compatibility import relpath
 
 # TODO Add a test for getting a distribution that is provided by another
-#   distribution.
+# distribution.
 
 # TODO Add a test for absolute pathed RECORD items (e.g. /etc/myapp/config.ini)
 
@@ -143,7 +143,7 @@
 
 
 class TestPkgUtilDistribution(unittest.TestCase):
-    """Tests the pkgutil.Distribution class"""
+    # Tests the pkgutil.Distribution class
 
     def setUp(self):
         self.fake_dists_path = os.path.abspath(
@@ -196,8 +196,8 @@
             open(record_file, 'w').close()
 
     def test_instantiation(self):
-        """Test the Distribution class's instantiation provides us with usable
-        attributes."""
+        # Test the Distribution class's instantiation provides us with usable
+        # attributes.
         # Import the Distribution class
         from distutils2._backport.pkgutil import distinfo_dirname, Distribution
 
@@ -215,7 +215,7 @@
         self.assertTrue(isinstance(dist.requested, type(bool())))
 
     def test_installed_files(self):
-        """Test the iteration of installed files."""
+        # Test the iteration of installed files.
         # Test the distribution's installed files
         from distutils2._backport.pkgutil import Distribution
         for distinfo_dir in self.distinfo_dirs:
@@ -227,7 +227,7 @@
                 self.assertEqual(size, record_data[path][1])
 
     def test_uses(self):
-        """Test to determine if a distribution uses a specified file."""
+        # Test to determine if a distribution uses a specified file.
         # Criteria to test against
         distinfo_name = 'grammar-1.0a4'
         distinfo_dir = os.path.join(self.fake_dists_path,
@@ -246,7 +246,7 @@
         self.assertFalse(dist.uses(false_path))
 
     def test_get_distinfo_file(self):
-        """Test the retrieval of dist-info file objects."""
+        # Test the retrieval of dist-info file objects.
         from distutils2._backport.pkgutil import Distribution
         distinfo_name = 'choxie-2.0.0.9'
         other_distinfo_name = 'grammar-1.0a4'
@@ -279,7 +279,7 @@
                           'ENTRYPOINTS')
 
     def test_get_distinfo_files(self):
-        """Test for the iteration of RECORD path entries."""
+        # Test for the iteration of RECORD path entries.
         from distutils2._backport.pkgutil import Distribution
         distinfo_name = 'towel_stuff-0.1'
         distinfo_dir = os.path.join(self.fake_dists_path,
@@ -297,7 +297,7 @@
 
 
 class TestPkgUtilPEP376(unittest.TestCase):
-    """Tests for the new functionality added in PEP 376."""
+    # Tests for the new functionality added in PEP 376.
 
     def setUp(self):
         # Setup the path environment with our fake distributions
@@ -310,15 +310,15 @@
         sys.path[:] = self.sys_path
 
     def test_distinfo_dirname(self):
-        """Given a name and a version, we expect the distinfo_dirname function
-        to return a standard distribution information directory name."""
+        # Given a name and a version, we expect the distinfo_dirname function
+        # to return a standard distribution information directory name.
 
         items = [# (name, version, standard_dirname)
             # Test for a very simple single word name and decimal
             # version number
             ('docutils', '0.5', 'docutils-0.5.dist-info'),
             # Test for another except this time with a '-' in the name, which
-            #   needs to be transformed during the name lookup
+            # needs to be transformed during the name lookup
             ('python-ldap', '2.5', 'python_ldap-2.5.dist-info'),
             # Test for both '-' in the name and a funky version number
             ('python-ldap', '2.5 a---5', 'python_ldap-2.5 a---5.dist-info'),
@@ -333,7 +333,7 @@
             self.assertEqual(dirname, standard_dirname)
 
     def test_get_distributions(self):
-        """Lookup all distributions found in the ``sys.path``."""
+        # Lookup all distributions found in the ``sys.path``.
         # This test could potentially pick up other installed distributions
         fake_dists = [('grammar', '1.0a4'), ('choxie', '2.0.0.9'),
             ('towel-stuff', '0.1')]
@@ -374,7 +374,7 @@
         self.assertListEqual(sorted(fake_dists), sorted(found_dists))
 
     def test_get_distribution(self):
-        """Test for looking up a distribution by name."""
+        # Test for looking up a distribution by name.
         # Test the lookup of the towel-stuff distribution
         name = 'towel-stuff' # Note: This is different from the directory name
 
@@ -420,7 +420,7 @@
         self.assertEqual(dist.name, 'strawberry')
 
     def test_get_file_users(self):
-        """Test the iteration of distributions that use a file."""
+        # Test the iteration of distributions that use a file.
         from distutils2._backport.pkgutil import get_file_users, Distribution
         name = 'towel_stuff-0.1'
         path = os.path.join(self.fake_dists_path, name,
@@ -430,7 +430,7 @@
             self.assertEqual(dist.name, name)
 
     def test_provides(self):
-        """ Test for looking up distributions by what they provide """
+        # Test for looking up distributions by what they provide
         from distutils2._backport.pkgutil import provides_distribution
         from distutils2.errors import DistutilsError
 
@@ -502,7 +502,7 @@
         checkLists(l, [])
 
     def test_obsoletes(self):
-        """ Test looking for distributions based on what they obsolete """
+        # Test looking for distributions based on what they obsolete
         from distutils2._backport.pkgutil import obsoletes_distribution
         from distutils2.errors import DistutilsError
 
diff --git a/src/distutils2/_backport/tests/test_sysconfig.py b/src/distutils2/_backport/tests/test_sysconfig.py
--- a/src/distutils2/_backport/tests/test_sysconfig.py
+++ b/src/distutils2/_backport/tests/test_sysconfig.py
@@ -88,15 +88,13 @@
             shutil.rmtree(path)
 
     def test_nested_var_substitution(self):
-        """Assert that the {curly brace token} expansion pattern will replace
-        only the inner {something} on nested expressions like {py{something}} on
-        the first pass.
+        # Assert that the {curly brace token} expansion pattern will replace
+        # only the inner {something} on nested expressions like {py{something}} on
+        # the first pass.
 
-        We have no plans to make use of this, but it keeps the option open for
-        the future, at the cost only of disallowing { itself as a piece of a
-        substitution key (which would be weird).
-
-        """
+        # We have no plans to make use of this, but it keeps the option open for
+        # the future, at the cost only of disallowing { itself as a piece of a
+        # substitution key (which would be weird).
         self.assertEqual(_subst_vars('{py{version}}', {'version': '31'}), '{py31}')
 
     def test_get_paths(self):
diff --git a/src/distutils2/tests/test_pypi_dist.py b/src/distutils2/tests/test_pypi_dist.py
--- a/src/distutils2/tests/test_pypi_dist.py
+++ b/src/distutils2/tests/test_pypi_dist.py
@@ -16,17 +16,17 @@
 
 class TestPyPIDistribution(support.TempdirManager,
                            unittest.TestCase):
-    """tests the pypi.dist.PyPIDistribution class"""
+    """Tests the pypi.dist.PyPIDistribution class"""
 
     def test_instanciation(self):
-        """Test the Distribution class provides us the good attributes when
-        given on construction"""
+        # Test the Distribution class provides us the good attributes when
+        # given on construction
         dist = Dist("FooBar", "1.1")
         self.assertEqual("FooBar", dist.name)
         self.assertEqual("1.1", "%s" % dist.version)
 
     def test_create_from_url(self):
-        """Test that the Distribution object can be built from a single URL"""
+        # Test that the Distribution object can be built from a single URL
         url_list = {
             'FooBar-1.1.0.tar.gz': {
                 'name': 'foobar',  # lowercase the name
@@ -68,7 +68,7 @@
                         self.assertEqual(getattr(dist, attribute), value)
 
     def test_get_url(self):
-        """Test that the url property works well"""
+        # Test that the url property works well
 
         d = Dist("foobar", "1.1", url="test_url")
         self.assertDictEqual(d.url, {
@@ -90,7 +90,7 @@
         self.assertEqual(2, len(d._urls))
 
     def test_comparaison(self):
-        """Test that we can compare PyPIDistributions"""
+        # Test that we can compare PyPIDistributions
         foo1 = Dist("foo", "1.0")
         foo2 = Dist("foo", "2.0")
         bar = Dist("bar", "2.0")
@@ -103,7 +103,7 @@
         self.assertRaises(TypeError, foo1.__eq__, bar)
 
     def test_split_archive_name(self):
-        """Test we can split the archive names"""
+        # Test we can split the archive names
         names = {
             'foo-bar-baz-1.0-rc2': ('foo-bar-baz', '1.0c2'),
             'foo-bar-baz-1.0': ('foo-bar-baz', '1.0'),
@@ -114,7 +114,7 @@
 
     @use_pypi_server("downloads_with_md5")
     def test_download(self, server):
-        """Download is possible, and the md5 is checked if given"""
+        # Download is possible, and the md5 is checked if given
 
         url = "%s/simple/foobar/foobar-0.1.tar.gz" % server.full_address
         # check md5 if given
@@ -146,8 +146,7 @@
         shutil.rmtree(os.path.dirname(path2))
 
     def test_hashname(self):
-        """Invalid hashnames raises an exception on assignation"""
-        # should be ok
+        # Invalid hashnames raises an exception on assignation
         Dist("FooBar", "0.1", url_hashname="md5", url_hashval="value")
 
         self.assertRaises(UnsupportedHashName, Dist, "FooBar", "0.1",
@@ -155,11 +154,10 @@
 
 
 class TestPyPIDistributions(unittest.TestCase):
-    """test the pypi.distr.PyPIDistributions class"""
 
     def test_filter(self):
-        """Test we filter the distributions the right way, using version
-        predicate match method"""
+        # Test we filter the distributions the right way, using version
+        # predicate match method
         dists = Dists((
             Dist("FooBar", "1.1"),
             Dist("FooBar", "1.1.1"),
@@ -173,7 +171,6 @@
         self.assertIn(dists[1], filtered)
 
     def test_append(self):
-        """Test the append method of PyPIDistributions"""
         # When adding a new item to the list, the behavior is to test if
         # a distribution with the same name and version number already exists,
         # and if so, to add url informations to the existing PyPIDistribution
@@ -198,7 +195,7 @@
         self.assertEqual(3, len(dists))
 
     def test_prefer_final(self):
-        """Ordering support prefer_final"""
+        # Can order the distributions using prefer_final
 
         fb10 = Dist("FooBar", "1.0")  # final distribution
         fb11a = Dist("FooBar", "1.1a1")  # alpha
@@ -213,7 +210,7 @@
         self.assertEqual(fb12b, dists[0])
 
     def test_prefer_source(self):
-        """Ordering support prefer_source"""
+        # Ordering support prefer_source
         fb_source = Dist("FooBar", "1.0", type="source")
         fb_binary = Dist("FooBar", "1.0", type="binary")
         fb2_binary = Dist("FooBar", "2.0", type="binary")
@@ -230,8 +227,7 @@
         self.assertEqual(fb2_binary, dists[0])
 
     def test_get_same_name_and_version(self):
-        """PyPIDistributions can return a list of "duplicates"
-        """
+        # PyPIDistributions can return a list of "duplicates"
         fb_source = Dist("FooBar", "1.0", type="source")
         fb_binary = Dist("FooBar", "1.0", type="binary")
         fb2_binary = Dist("FooBar", "2.0", type="binary")
diff --git a/src/distutils2/tests/test_pypi_server.py b/src/distutils2/tests/test_pypi_server.py
--- a/src/distutils2/tests/test_pypi_server.py
+++ b/src/distutils2/tests/test_pypi_server.py
@@ -10,7 +10,7 @@
 class PyPIServerTest(unittest.TestCase):
 
     def test_records_requests(self):
-        """We expect that PyPIServer can log our requests"""
+        # We expect that PyPIServer can log our requests
         server = PyPIServer()
         server.start()
         self.assertEqual(len(server.requests), 0)
@@ -29,8 +29,7 @@
         server.stop()
 
     def test_serve_static_content(self):
-        """PYPI Mocked server can serve static content from disk.
-        """
+        # PYPI Mocked server can serve static content from disk.
 
         def uses_local_files_for(server, url_path):
             """Test that files are served statically (eg. the output from the
diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -20,7 +20,7 @@
                          unittest.TestCase):
 
     def _get_simple_index(self, server, base_url="/simple/", hosts=None,
-        *args, **kwargs):
+                          *args, **kwargs):
         """Build and return a SimpleSimpleIndex instance, with the test server
         urls
         """
@@ -43,9 +43,7 @@
         # issue 16
         # easy_install inquant.contentmirror.plone breaks because of a typo
         # in its home URL
-        index = simple.SimpleIndex(
-            hosts=('www.example.com',))
-
+        index = simple.SimpleIndex(hosts=('www.example.com',))
         url = 'url:%20https://svn.plone.org/svn/collective/inquant.contentmirror.plone/trunk'
         try:
             v = index._open_url(url)
@@ -88,8 +86,7 @@
 
     @use_pypi_server("test_found_links")
     def test_found_links(self, server):
-        """Browse the index, asking for a specified distribution version
-        """
+        # Browse the index, asking for a specified distribution version
         # The PyPI index contains links for version 1.0, 1.1, 2.0 and 2.0.1
         index = self._get_simple_index(server)
         last_distribution = index.get("foobar")
@@ -143,8 +140,7 @@
 
     @use_pypi_server("with_externals")
     def test_restrict_hosts(self, server):
-        """Include external pages
-        """
+        # Include external pages
         # Try to request the package index, wich contains links to "externals"
         # resources. They have to  be scanned too.
         index = self._get_simple_index(server, follow_externals=True)
@@ -154,8 +150,7 @@
 
     @use_pypi_server("with_real_externals")
     def test_restrict_hosts(self, server):
-        """Only use a list of allowed hosts is possible
-        """
+        # Only use a list of allowed hosts is possible
         # Test that telling the simple pyPI client to not retrieve external
         # works
         index = self._get_simple_index(server, follow_externals=False)
@@ -163,28 +158,22 @@
         self.assertNotIn(server.full_address + "/external/external.html",
             index._processed_urls)
 
-    @use_pypi_server("with_egg_files")
-    def test_scan_egg_files(self, server):
-        """Assert that egg files are indexed as well"""
-        pass
-
     @use_pypi_server(static_filesystem_paths=["with_externals"],
         static_uri_paths=["simple", "external"])
     def test_links_priority(self, server):
-        """
-        Download links from the pypi simple index should be used before
-        external download links.
-        http://bitbucket.org/tarek/distribute/issue/163/md5-validation-error
+        # Download links from the pypi simple index should be used before
+        # external download links.
+        # http://bitbucket.org/tarek/distribute/issue/163/md5-validation-error
+        #
+        # Usecase :
+        # - someone uploads a package on pypi, a md5 is generated
+        # - someone manually coindexes this link (with the md5 in the url) onto
+        #   an external page accessible from the package page.
+        # - someone reuploads the package (with a different md5)
+        # - while easy_installing, an MD5 error occurs because the external link
+        #   is used
+        # -> The index should use the link from pypi, not the external one.
 
-        Usecase :
-        - someone uploads a package on pypi, a md5 is generated
-        - someone manually coindexes this link (with the md5 in the url) onto
-          an external page accessible from the package page.
-        - someone reuploads the package (with a different md5)
-        - while easy_installing, an MD5 error occurs because the external link
-          is used
-        -> The index should use the link from pypi, not the external one.
-        """
         # start an index server
         index_url = server.full_address + '/simple/'
 
@@ -202,10 +191,10 @@
     @use_pypi_server(static_filesystem_paths=["with_norel_links"],
         static_uri_paths=["simple", "external"])
     def test_not_scan_all_links(self, server):
-        """Do not follow all index page links.
-        The links not tagged with rel="download" and rel="homepage" have
-        to not be processed by the package index, while processing "pages".
-        """
+        # Do not follow all index page links.
+        # The links not tagged with rel="download" and rel="homepage" have
+        # to not be processed by the package index, while processing "pages".
+
         # process the pages
         index = self._get_simple_index(server, follow_externals=True)
         index.find("foobar")
@@ -225,7 +214,7 @@
             index._processed_urls)  # linked from external homepage (rel)
 
     def test_uses_mirrors(self):
-        """When the main repository seems down, try using the given mirrors"""
+        # When the main repository seems down, try using the given mirrors"""
         server = PyPIServer("foo_bar_baz")
         mirror = PyPIServer("foo_bar_baz")
         mirror.start()  # we dont start the server here
@@ -242,7 +231,7 @@
             mirror.stop()
 
     def test_simple_link_matcher(self):
-        """Test that the simple link matcher yields the right links"""
+        # Test that the simple link matcher yields the right links"""
         index = simple.SimpleIndex(follow_externals=False)
 
         # Here, we define:
@@ -276,7 +265,7 @@
         self.assertRaises(StopIteration, generator.next)
 
     def test_browse_local_files(self):
-        """Test that we can browse local files"""
+        # Test that we can browse local files"""
         index_path = os.sep.join(["file://" + PYPI_DEFAULT_STATIC_PATH,
                                   "test_found_links", "simple"])
         index = simple.SimpleIndex(index_path)

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Thu Jul 15 01:38:05 2010
From: python-checkins at python.org (tarek.ziade)
Date: Thu, 15 Jul 2010 01:38:05 +0200
Subject: [Python-checkins] distutils2: Remove some temporary folders after
	use.
Message-ID: <E1OZBWf-0001Gf-F3@dinsdale.python.org>

tarek.ziade pushed 3882146da925 to distutils2:

http://hg.python.org/distutils2/rev/3882146da925
changeset:   359:3882146da925
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Wed Jul 07 21:09:19 2010 +0200
summary:     Remove some temporary folders after use.
files:       src/distutils2/tests/test_pypi_dist.py

diff --git a/src/distutils2/tests/test_pypi_dist.py b/src/distutils2/tests/test_pypi_dist.py
--- a/src/distutils2/tests/test_pypi_dist.py
+++ b/src/distutils2/tests/test_pypi_dist.py
@@ -5,8 +5,8 @@
 import tempfile
 
 from distutils2.tests.pypi_server import use_pypi_server
-from distutils2.tests import support
-from distutils2.tests.support import unittest
+from distutils2.tests import run_unittest
+from distutils2.tests.support import unittest, TempdirManager
 from distutils2.version import VersionPredicate
 from distutils2.pypi.errors import HashDoesNotMatch, UnsupportedHashName
 from distutils2.pypi.dist import (PyPIDistribution as Dist,
@@ -14,7 +14,7 @@
                                   split_archive_name)
 
 
-class TestPyPIDistribution(support.TempdirManager,
+class TestPyPIDistribution(TempdirManager,
                            unittest.TestCase):
     """Tests the pypi.dist.PyPIDistribution class"""
 
@@ -116,35 +116,35 @@
     def test_download(self, server):
         # Download is possible, and the md5 is checked if given
 
+        add_to_tmpdirs = lambda x: self.tempdirs.append(os.path.dirname(x))
+
         url = "%s/simple/foobar/foobar-0.1.tar.gz" % server.full_address
         # check md5 if given
         dist = Dist("FooBar", "0.1", url=url,
             url_hashname="md5", url_hashval="d41d8cd98f00b204e9800998ecf8427e")
-        dist.download()
+        add_to_tmpdirs(dist.download())
 
         # a wrong md5 fails
         dist2 = Dist("FooBar", "0.1", url=url,
             url_hashname="md5", url_hashval="wrongmd5")
+
         self.assertRaises(HashDoesNotMatch, dist2.download)
+        add_to_tmpdirs(dist2.downloaded_location)
 
         # we can omit the md5 hash
         dist3 = Dist("FooBar", "0.1", url=url)
-        dist3.download()
+        add_to_tmpdirs(dist3.download())
 
         # and specify a temporary location
         # for an already downloaded dist
-        path1 = tempfile.mkdtemp()
+        path1 = self.mkdtemp()
         dist3.download(path=path1)
         # and for a new one
-        path2_base = tempfile.mkdtemp()
+        path2_base = self.mkdtemp()
         dist4 = Dist("FooBar", "0.1", url=url)
         path2 = dist4.download(path=path2_base)
         self.assertTrue(path2_base in path2)
 
-        # remove the temp folders
-        shutil.rmtree(path1)
-        shutil.rmtree(os.path.dirname(path2))
-
     def test_hashname(self):
         # Invalid hashnames raises an exception on assignation
         Dist("FooBar", "0.1", url_hashname="md5", url_hashval="value")

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Thu Jul 15 01:38:05 2010
From: python-checkins at python.org (tarek.ziade)
Date: Thu, 15 Jul 2010 01:38:05 +0200
Subject: [Python-checkins] distutils2: Finally get back to from_url as a
	classmethod, as it's seems to be
Message-ID: <E1OZBWf-0001Gc-EY@dinsdale.python.org>

tarek.ziade pushed cbe93cce3ae7 to distutils2:

http://hg.python.org/distutils2/rev/cbe93cce3ae7
changeset:   358:cbe93cce3ae7
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Wed Jul 07 14:53:16 2010 +0200
summary:     Finally get back to from_url as a classmethod, as it's seems to be
files:       src/distutils2/pypi/dist.py, src/distutils2/pypi/simple.py, src/distutils2/tests/test_pypi_dist.py

diff --git a/src/distutils2/pypi/dist.py b/src/distutils2/pypi/dist.py
--- a/src/distutils2/pypi/dist.py
+++ b/src/distutils2/pypi/dist.py
@@ -31,6 +31,38 @@
     information about distributions.
     """
 
+    @classmethod
+    def from_url(cls, url, probable_dist_name=None, is_external=True):
+        """Build a Distribution from a url archive (egg or zip or tgz).
+
+        :param url: complete url of the distribution
+        :param probable_dist_name: A probable name of the distribution.
+        :param is_external: Tell if the url commes from an index or from
+                            an external URL.
+        """
+        # if the url contains a md5 hash, get it.
+        md5_hash = None
+        match = MD5_HASH.match(url)
+        if match is not None:
+            md5_hash = match.group(1)
+            # remove the hash
+            url = url.replace("#md5=%s" % md5_hash, "")
+
+        # parse the archive name to find dist name and version
+        archive_name = urlparse.urlparse(url)[2].split('/')[-1]
+        extension_matched = False
+        # remove the extension from the name
+        for ext in EXTENSIONS:
+            if archive_name.endswith(ext):
+                archive_name = archive_name[:-len(ext)]
+                extension_matched = True
+
+        name, version = split_archive_name(archive_name)
+        if extension_matched is True:
+            return PyPIDistribution(name, version, url=url, url_hashname="md5",
+                                    url_hashval=md5_hash,
+                                    url_is_external=is_external)
+
     def __init__(self, name, version, type=None, url=None, url_hashname=None,
                  url_hashval=None, url_is_external=True):
         """Create a new instance of PyPIDistribution.
@@ -245,36 +277,6 @@
             key=lambda i: [getattr(i, arg) for arg in sort_by],
             reverse=reverse, *args, **kwargs)
 
-def create_from_url(url, probable_dist_name=None,
-                                 is_external=True):
-    """Build a Distribution from a url archive (egg or zip or tgz).
-
-    :param url: complete url of the distribution
-    :param probable_dist_name: A probable name of the distribution.
-    :param is_external: Tell if the url commes from an index or from
-                        an external URL.
-    """
-    # if the url contains a md5 hash, get it.
-    md5_hash = None
-    match = MD5_HASH.match(url)
-    if match is not None:
-        md5_hash = match.group(1)
-        # remove the hash
-        url = url.replace("#md5=%s" % md5_hash, "")
-
-    # parse the archive name to find dist name and version
-    archive_name = urlparse.urlparse(url)[2].split('/')[-1]
-    extension_matched = False
-    # remove the extension from the name
-    for ext in EXTENSIONS:
-        if archive_name.endswith(ext):
-            archive_name = archive_name[:-len(ext)]
-            extension_matched = True
-
-    name, version = split_archive_name(archive_name)
-    if extension_matched is True:
-        return PyPIDistribution(name, version, url=url, url_hashname="md5",
-                                url_hashval=md5_hash, url_is_external=is_external)
 
 def split_archive_name(archive_name, probable_name=None):
     """Split an archive name into two parts: name and version.
diff --git a/src/distutils2/pypi/simple.py b/src/distutils2/pypi/simple.py
--- a/src/distutils2/pypi/simple.py
+++ b/src/distutils2/pypi/simple.py
@@ -14,7 +14,7 @@
 
 from distutils2.version import VersionPredicate
 from distutils2.pypi.dist import (PyPIDistribution, PyPIDistributions,
-                                  create_from_url, EXTENSIONS)
+                                  EXTENSIONS)
 from distutils2.pypi.errors import (PyPIError, DistributionNotFound,
                                     DownloadError, UnableToDownload)
 from distutils2 import __version__ as __distutils2_version__
@@ -264,7 +264,7 @@
                     if self._is_distribution(link) or is_download:
                         self._processed_urls.append(link)
                         # it's a distribution, so create a dist object
-                        dist = create_from_url(link, project_name,
+                        dist = PyPIDistribution.from_url(link, project_name,
                                     is_external=not self.index_url in url)
                         self._register_dist(dist)
                     else:
diff --git a/src/distutils2/tests/test_pypi_dist.py b/src/distutils2/tests/test_pypi_dist.py
--- a/src/distutils2/tests/test_pypi_dist.py
+++ b/src/distutils2/tests/test_pypi_dist.py
@@ -11,7 +11,7 @@
 from distutils2.pypi.errors import HashDoesNotMatch, UnsupportedHashName
 from distutils2.pypi.dist import (PyPIDistribution as Dist,
                                   PyPIDistributions as Dists,
-                                  split_archive_name, create_from_url)
+                                  split_archive_name)
 
 
 class TestPyPIDistribution(support.TempdirManager,
@@ -55,7 +55,7 @@
         }
 
         for url, attributes in url_list.items():
-            dist = create_from_url("http://test.tld/" + url)
+            dist = Dist.from_url("http://test.tld/" + url)
             for attribute, value in attributes.items():
                 if isinstance(value, dict):
                     mylist = getattr(dist, attribute)

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Thu Jul 15 01:38:05 2010
From: python-checkins at python.org (tarek.ziade)
Date: Thu, 15 Jul 2010 01:38:05 +0200
Subject: [Python-checkins] distutils2: Apply Merwok cosmetic changes
Message-ID: <E1OZBWf-0001GY-Dx@dinsdale.python.org>

tarek.ziade pushed 583174d98c05 to distutils2:

http://hg.python.org/distutils2/rev/583174d98c05
changeset:   357:583174d98c05
user:        Alexis Metaireau <ametaireau at gmail.com>
date:        Wed Jul 07 14:42:32 2010 +0200
summary:     Apply Merwok cosmetic changes
files:       src/distutils2/tests/test_pypi_simple.py

diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
--- a/src/distutils2/tests/test_pypi_simple.py
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -7,13 +7,11 @@
 import tempfile
 import urllib2
 
+from distutils2.pypi import simple
 from distutils2.tests import support, run_unittest
 from distutils2.tests.support import unittest
-from distutils2.tests.pypi_server import use_pypi_server, PyPIServer, \
-                                         PYPI_DEFAULT_STATIC_PATH
-from distutils2.pypi import simple
-
-from distutils2.errors import DistutilsError
+from distutils2.tests.pypi_server import (use_pypi_server, PyPIServer,
+                                          PYPI_DEFAULT_STATIC_PATH)
 
 
 class PyPISimpleTestCase(support.TempdirManager,
@@ -276,4 +274,4 @@
     return unittest.makeSuite(PyPISimpleTestCase)
 
 if __name__ == '__main__':
-    run_unittest(test_suite())
+    unittest.main(defaultTest="test_suite")

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Thu Jul 15 01:38:05 2010
From: python-checkins at python.org (tarek.ziade)
Date: Thu, 15 Jul 2010 01:38:05 +0200
Subject: [Python-checkins] distutils2: Get back things lost in a recent merge
Message-ID: <E1OZBWf-0001Gl-ID@dinsdale.python.org>

tarek.ziade pushed 2a9ad83361f7 to distutils2:

http://hg.python.org/distutils2/rev/2a9ad83361f7
changeset:   360:2a9ad83361f7
user:        ?ric Araujo <merwok at netwok.org>
date:        Wed Jul 07 20:19:08 2010 +0200
summary:     Get back things lost in a recent merge
files:       docs/source/metadata.rst, src/distutils2/_backport/tests/test_pkgutil.py, src/distutils2/tests/test_upload.py, src/setup.py

diff --git a/docs/source/metadata.rst b/docs/source/metadata.rst
--- a/docs/source/metadata.rst
+++ b/docs/source/metadata.rst
@@ -17,7 +17,7 @@
 Reading metadata
 ================
 
-The :class:`DistributionMetadata` class can be instanciated with the path of
+The :class:`DistributionMetadata` class can be instantiated with the path of
 the metadata file, and provides a dict-like interface to the values::
 
     >>> from distutils2.metadata import DistributionMetadata
@@ -32,14 +32,14 @@
     ["pywin32; sys.platform == 'win32'", "Sphinx"]
 
 The fields that supports environment markers can be automatically ignored if
-the object is instanciated using the ``platform_dependant`` option.
+the object is instantiated using the ``platform_dependent`` option.
 :class:`DistributionMetadata` will interpret in the case the markers and will
 automatically remove the fields that are not compliant with the running
 environment. Here's an example under Mac OS X. The win32 dependency
 we saw earlier is ignored::
 
     >>> from distutils2.metadata import DistributionMetadata
-    >>> metadata = DistributionMetadata('PKG-INFO', platform_dependant=True)
+    >>> metadata = DistributionMetadata('PKG-INFO', platform_dependent=True)
     >>> metadata['Requires-Dist']
     ['bar']
 
@@ -53,7 +53,7 @@
 
     >>> from distutils2.metadata import DistributionMetadata
     >>> context = {'sys.platform': 'win32'}
-    >>> metadata = DistributionMetadata('PKG-INFO', platform_dependant=True,
+    >>> metadata = DistributionMetadata('PKG-INFO', platform_dependent=True,
     ...                                 execution_context=context)
     ...
     >>> metadata['Requires-Dist'] = ["pywin32; sys.platform == 'win32'",
@@ -71,7 +71,7 @@
     >>> metadata.write('/to/my/PKG-INFO')
 
 The class will pick the best version for the metadata, depending on the values
-provided. If all the values provided exists in all versions, the class will
+provided. If all the values provided exist in all versions, the class will
 use :attr:`metadata.PKG_INFO_PREFERRED_VERSION`. It is set by default to 1.0.
 
 
@@ -79,7 +79,7 @@
 ==================================
 
 Some fields in :pep:`345` have to follow a version scheme in their versions
-predicate. When the scheme is violated, a warning is emited::
+predicate. When the scheme is violated, a warning is emitted::
 
     >>> from distutils2.metadata import DistributionMetadata
     >>> metadata = DistributionMetadata()
@@ -90,6 +90,3 @@
 
 
 .. TODO talk about check()
-
-
-
diff --git a/src/distutils2/_backport/tests/test_pkgutil.py b/src/distutils2/_backport/tests/test_pkgutil.py
--- a/src/distutils2/_backport/tests/test_pkgutil.py
+++ b/src/distutils2/_backport/tests/test_pkgutil.py
@@ -13,9 +13,9 @@
     from md5 import md5
 
 from test.test_support import run_unittest, TESTFN
+from distutils2.tests.support import unittest
 
 from distutils2._backport import pkgutil
-from distutils2.tests.support import unittest
 
 try:
     from os.path import relpath
@@ -34,10 +34,12 @@
 class TestPkgUtilData(unittest.TestCase):
 
     def setUp(self):
+        super(TestPkgUtilData, self).setUp()
         self.dirname = tempfile.mkdtemp()
         sys.path.insert(0, self.dirname)
 
     def tearDown(self):
+        super(TestPkgUtilData, self).tearDown()
         del sys.path[0]
         shutil.rmtree(self.dirname)
 
@@ -52,15 +54,22 @@
         os.mkdir(package_dir)
         # Empty init.py
         f = open(os.path.join(package_dir, '__init__.py'), "wb")
-        f.close()
+        try:
+            pass
+        finally:
+            f.close()
         # Resource files, res.txt, sub/res.txt
         f = open(os.path.join(package_dir, 'res.txt'), "wb")
-        f.write(RESOURCE_DATA)
-        f.close()
+        try:
+            f.write(RESOURCE_DATA)
+        finally:
+            f.close()
         os.mkdir(os.path.join(package_dir, 'sub'))
         f = open(os.path.join(package_dir, 'sub', 'res.txt'), "wb")
-        f.write(RESOURCE_DATA)
-        f.close()
+        try:
+            f.write(RESOURCE_DATA)
+        finally:
+            f.close()
 
         # Check we can read the resources
         res1 = pkgutil.get_data(pkg, 'res.txt')
@@ -80,13 +89,14 @@
         # Make a package with some resources
         zip_file = os.path.join(self.dirname, zip)
         z = zipfile.ZipFile(zip_file, 'w')
-
-        # Empty init.py
-        z.writestr(pkg + '/__init__.py', "")
-        # Resource files, res.txt, sub/res.txt
-        z.writestr(pkg + '/res.txt', RESOURCE_DATA)
-        z.writestr(pkg + '/sub/res.txt', RESOURCE_DATA)
-        z.close()
+        try:
+            # Empty init.py
+            z.writestr(pkg + '/__init__.py', "")
+            # Resource files, res.txt, sub/res.txt
+            z.writestr(pkg + '/res.txt', RESOURCE_DATA)
+            z.writestr(pkg + '/sub/res.txt', RESOURCE_DATA)
+        finally:
+            z.close()
 
         # Check we can read the resources
         sys.path.insert(0, zip_file)
@@ -121,10 +131,12 @@
             return TestPkgUtilPEP302.MyTestLoader()
 
     def setUp(self):
+        super(TestPkgUtilPEP302, self).setUp()
         sys.meta_path.insert(0, self.MyTestImporter())
 
     def tearDown(self):
         del sys.meta_path[0]
+        super(TestPkgUtilPEP302, self).tearDown()
 
     def test_getdata_pep302(self):
         # Use a dummy importer/loader
@@ -146,6 +158,7 @@
     # Tests the pkgutil.Distribution class
 
     def setUp(self):
+        super(TestPkgUtilDistribution, self).setUp()
         self.fake_dists_path = os.path.abspath(
             os.path.join(os.path.dirname(__file__), 'fake_dists'))
 
@@ -194,6 +207,7 @@
         for distinfo_dir in self.distinfo_dirs:
             record_file = os.path.join(distinfo_dir, 'RECORD')
             open(record_file, 'w').close()
+        super(TestPkgUtilDistribution, self).tearDown()
 
     def test_instantiation(self):
         # Test the Distribution class's instantiation provides us with usable
@@ -300,6 +314,7 @@
     # Tests for the new functionality added in PEP 376.
 
     def setUp(self):
+        super(TestPkgUtilPEP376, self).setUp()
         # Setup the path environment with our fake distributions
         current_path = os.path.abspath(os.path.dirname(__file__))
         self.sys_path = sys.path[:]
@@ -308,6 +323,7 @@
 
     def tearDown(self):
         sys.path[:] = self.sys_path
+        super(TestPkgUtilPEP376, self).tearDown()
 
     def test_distinfo_dirname(self):
         # Given a name and a version, we expect the distinfo_dirname function
diff --git a/src/distutils2/tests/test_upload.py b/src/distutils2/tests/test_upload.py
--- a/src/distutils2/tests/test_upload.py
+++ b/src/distutils2/tests/test_upload.py
@@ -1,12 +1,13 @@
 """Tests for distutils.command.upload."""
 # -*- encoding: utf8 -*-
-import os, sys
+import os
+import sys
 
 from distutils2.command.upload import upload
 from distutils2.core import Distribution
 
+from distutils2.tests import support
 from distutils2.tests.pypi_server import PyPIServer, PyPIServerTestCase
-from distutils2.tests import support
 from distutils2.tests.support import unittest
 from distutils2.tests.test_config import PYPIRC, PyPIRCCommandTestCase
 
@@ -29,10 +30,10 @@
         dist = Distribution()
         cmd = upload(dist)
         cmd.finalize_options()
-        for attr, waited in (('username', 'me'), ('password', 'secret'),
-                             ('realm', 'pypi'),
-                             ('repository', 'http://pypi.python.org/pypi')):
-            self.assertEqual(getattr(cmd, attr), waited)
+        for attr, expected in (('username', 'me'), ('password', 'secret'),
+                               ('realm', 'pypi'),
+                               ('repository', 'http://pypi.python.org/pypi')):
+            self.assertEqual(getattr(cmd, attr), expected)
 
     def test_saved_password(self):
         # file with no password
diff --git a/src/setup.py b/src/setup.py
--- a/src/setup.py
+++ b/src/setup.py
@@ -226,7 +226,7 @@
       packages=find_packages(),
       cmdclass={'sdist': sdist_hg, 'install': install_hg},
       package_data={'distutils2._backport': ['sysconfig.cfg']},
-      project_url=[('Mailing-list',
+      project_url=[('Mailing list',
                     'http://mail.python.org/mailman/listinfo/distutils-sig/'),
                    ('Documentation',
                     'http://packages.python.org/Distutils2'),

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Thu Jul 15 01:38:05 2010
From: python-checkins at python.org (tarek.ziade)
Date: Thu, 15 Jul 2010 01:38:05 +0200
Subject: [Python-checkins] distutils2: Begin documenting API interfaces
Message-ID: <E1OZBWf-0001Go-Ih@dinsdale.python.org>

tarek.ziade pushed abdf466ea00e to distutils2:

http://hg.python.org/distutils2/rev/abdf466ea00e
changeset:   361:abdf466ea00e
parent:      231:8fb85c777536
user:        Jeremy Kloth <jeremy.kloth at gmail.com>
date:        Tue Jun 15 10:41:18 2010 -0600
summary:     Begin documenting API interfaces
files:       src/distutils2/command/build_py.py

diff --git a/src/distutils2/command/build_py.py b/src/distutils2/command/build_py.py
--- a/src/distutils2/command/build_py.py
+++ b/src/distutils2/command/build_py.py
@@ -95,8 +95,13 @@
 
         self.byte_compile(self.get_outputs(include_bytecode=0))
 
+    # -- Top-level worker functions ------------------------------------
+
     def get_data_files(self):
-        """Generate list of '(package,src_dir,build_dir,filenames)' tuples"""
+        """Generate list of '(package,src_dir,build_dir,filenames)' tuples.
+
+        Helper function for `finalize_options()`.
+        """
         data = []
         if not self.packages:
             return data
@@ -120,7 +125,10 @@
         return data
 
     def find_data_files(self, package, src_dir):
-        """Return filenames for package's data files in 'src_dir'"""
+        """Return filenames for package's data files in 'src_dir'.
+
+        Helper function for `get_data_files()`.
+        """
         globs = (self.package_data.get('', [])
                  + self.package_data.get(package, []))
         files = []
@@ -132,7 +140,10 @@
         return files
 
     def build_package_data(self):
-        """Copy data files into build directory"""
+        """Copy data files into build directory.
+
+        Helper function for `run()`.
+        """
         for package, src_dir, build_dir, filenames in self.data_files:
             for filename in filenames:
                 target = os.path.join(build_dir, filename)
@@ -140,6 +151,8 @@
                 self.copy_file(os.path.join(src_dir, filename), target,
                                preserve_mode=False)
 
+    # XXX - this should be moved to the Distribution class as it is not
+    # only needed for build_py. It also has no dependencies on this class.
     def get_package_dir(self, package):
         """Return the directory, relative to the top of the source
            distribution, where package 'package' should be found
@@ -181,6 +194,8 @@
                     return ''
 
     def check_package(self, package, package_dir):
+        """Helper function for `find_package_modules()` and `find_modules()'.
+        """
         # Empty dir name means current directory, which we can probably
         # assume exists.  Also, os.path.exists and isdir don't know about
         # my "empty string means current dir" convention, so we have to

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Thu Jul 15 01:38:05 2010
From: python-checkins at python.org (tarek.ziade)
Date: Thu, 15 Jul 2010 01:38:05 +0200
Subject: [Python-checkins] distutils2: Merge from trunk
Message-ID: <E1OZBWf-0001Gt-NP@dinsdale.python.org>

tarek.ziade pushed a453819940a4 to distutils2:

http://hg.python.org/distutils2/rev/a453819940a4
changeset:   362:a453819940a4
parent:      361:abdf466ea00e
parent:      198:0e9228d9bbda
user:        Jeremy Kloth <jeremy.kloth at gmail.com>
date:        Tue Jun 15 10:43:16 2010 -0600
summary:     Merge from trunk
files:       src/distutils2/cmd.py, src/distutils2/command/cmd.py

diff --git a/.hgignore b/.hgignore
--- a/.hgignore
+++ b/.hgignore
@@ -1,5 +1,5 @@
 .*\.pyc$
 .*\.pyo$
+^src/build
 ^docs/build
 .*\.swp$
-
diff --git a/src/CONTRIBUTORS.txt b/src/CONTRIBUTORS.txt
--- a/src/CONTRIBUTORS.txt
+++ b/src/CONTRIBUTORS.txt
@@ -9,6 +9,7 @@
 
 Thanks to:
 
+- ??ric Araujo
 - Pior Bastida
 - Titus Brown
 - Nicolas Cadou
diff --git a/src/distutils2/_backport/tests/__init__.py b/src/distutils2/_backport/tests/__init__.py
--- a/src/distutils2/_backport/tests/__init__.py
+++ b/src/distutils2/_backport/tests/__init__.py
@@ -1,12 +1,13 @@
 import os
 import sys
-import unittest2
+
+from distutils2.tests.support import unittest
 
 
 here = os.path.dirname(__file__)
 
 def test_suite():
-    suite = unittest2.TestSuite()
+    suite = unittest.TestSuite()
     for fn in os.listdir(here):
         if fn.startswith("test") and fn.endswith(".py"):
             modname = "distutils2._backport.tests." + fn[:-3]
diff --git a/src/distutils2/_backport/tests/test_pkgutil.py b/src/distutils2/_backport/tests/test_pkgutil.py
--- a/src/distutils2/_backport/tests/test_pkgutil.py
+++ b/src/distutils2/_backport/tests/test_pkgutil.py
@@ -1,6 +1,5 @@
 # -*- coding: utf-8 -*-
 """Tests for PEP 376 pkgutil functionality"""
-import unittest2
 import sys
 import os
 import csv
@@ -12,13 +11,14 @@
 from test.test_support import run_unittest, TESTFN
 
 import distutils2._backport.pkgutil
+from distutils2.tests.support import unittest
 
 # TODO Add a test for getting a distribution that is provided by another
 #   distribution.
 
 # TODO Add a test for absolute pathed RECORD items (e.g. /etc/myapp/config.ini)
 
-class TestPkgUtilDistribution(unittest2.TestCase):
+class TestPkgUtilDistribution(unittest.TestCase):
     """Tests the pkgutil.Distribution class"""
 
     def setUp(self):
@@ -171,7 +171,7 @@
         self.assertEqual(sorted(found), sorted(distinfo_record_paths))
 
 
-class TestPkgUtilFunctions(unittest2.TestCase):
+class TestPkgUtilFunctions(unittest.TestCase):
     """Tests for the new functionality added in PEP 376."""
 
     def setUp(self):
@@ -371,8 +371,8 @@
 
 
 def test_suite():
-    suite = unittest2.TestSuite()
-    testcase_loader = unittest2.loader.defaultTestLoader.loadTestsFromTestCase
+    suite = unittest.TestSuite()
+    testcase_loader = unittest.loader.defaultTestLoader.loadTestsFromTestCase
     suite.addTest(testcase_loader(TestPkgUtilFunctions))
     suite.addTest(testcase_loader(TestPkgUtilDistribution))
     return suite
@@ -384,8 +384,8 @@
     test_main()
 
 def test_suite():
-    suite = unittest2.TestSuite()
-    testcase_loader = unittest2.loader.defaultTestLoader.loadTestsFromTestCase
+    suite = unittest.TestSuite()
+    testcase_loader = unittest.loader.defaultTestLoader.loadTestsFromTestCase
     suite.addTest(testcase_loader(TestPkgUtilFunctions))
     suite.addTest(testcase_loader(TestPkgUtilDistribution))
     return suite
diff --git a/src/distutils2/_backport/tests/test_sysconfig.py b/src/distutils2/_backport/tests/test_sysconfig.py
--- a/src/distutils2/_backport/tests/test_sysconfig.py
+++ b/src/distutils2/_backport/tests/test_sysconfig.py
@@ -4,7 +4,6 @@
 executing have not been removed.
 
 """
-import unittest2
 import sys
 import os
 import shutil
@@ -19,8 +18,9 @@
                        get_path, get_path_names,
                        _get_default_scheme, _subst_vars, _expand_vars,
                        get_scheme_names, _CONFIG_FILE)
+from distutils2.tests.support import unittest
 
-class TestSysConfig(unittest2.TestCase):
+class TestSysConfig(unittest.TestCase):
 
     def setUp(self):
         """Make a copy of sys.path"""
@@ -97,7 +97,7 @@
         substitution key (which would be weird).
 
         """
-        self.failUnlessEqual(_subst_vars('{py{version}}', {'version': '31'}), '{py31}')
+        self.assertEqual(_subst_vars('{py{version}}', {'version': '31'}), '{py31}')
 
     def test_get_paths(self):
         scheme = get_paths()
@@ -262,7 +262,7 @@
         _expand_globals(config)
 
 def test_suite():
-    return unittest2.makeSuite(TestSysConfig)
+    return unittest.makeSuite(TestSysConfig)
 
 def test_main():
     run_unittest(test_suite())
diff --git a/src/distutils2/cmd.py b/src/distutils2/command/cmd.py
rename from src/distutils2/cmd.py
rename to src/distutils2/command/cmd.py
diff --git a/src/distutils2/command/install_egg_info.py b/src/distutils2/command/install_egg_info.py
--- a/src/distutils2/command/install_egg_info.py
+++ b/src/distutils2/command/install_egg_info.py
@@ -4,7 +4,7 @@
 a package's PKG-INFO metadata."""
 
 
-from distutils2.cmd import Command
+from distutils2.command.cmd import Command
 from distutils2 import log
 from distutils2._backport.shutil import rmtree
 import os, sys, re
diff --git a/src/distutils2/config.py b/src/distutils2/config.py
--- a/src/distutils2/config.py
+++ b/src/distutils2/config.py
@@ -6,7 +6,7 @@
 import os
 from ConfigParser import ConfigParser
 
-from distutils2.cmd import Command
+from distutils2.command.cmd import Command
 
 DEFAULT_PYPIRC = """\
 [distutils]
diff --git a/src/distutils2/core.py b/src/distutils2/core.py
--- a/src/distutils2/core.py
+++ b/src/distutils2/core.py
@@ -17,7 +17,7 @@
 
 # Mainly import these so setup scripts can "from distutils2.core import" them.
 from distutils2.dist import Distribution
-from distutils2.cmd import Command
+from distutils2.command.cmd import Command
 from distutils2.config import PyPIRCCommand
 from distutils2.extension import Extension
 
diff --git a/src/distutils2/dist.py b/src/distutils2/dist.py
--- a/src/distutils2/dist.py
+++ b/src/distutils2/dist.py
@@ -493,7 +493,7 @@
         None if the user asked for help on this command.
         """
         # late import because of mutual dependence between these modules
-        from distutils2.cmd import Command
+        from distutils2.command.cmd import Command
 
         # Pull the current command from the head of the command line
         command = args[0]
@@ -607,7 +607,7 @@
         """
         # late import because of mutual dependence between these modules
         from distutils2.core import gen_usage
-        from distutils2.cmd import Command
+        from distutils2.command.cmd import Command
 
         if global_options:
             if display_options:
@@ -904,7 +904,7 @@
 
         Returns the reinitialized command object.
         """
-        from distutils2.cmd import Command
+        from distutils2.command.cmd import Command
         if not isinstance(command, Command):
             command_name = command
             command = self.get_command_obj(command_name)
diff --git a/src/distutils2/tests/__init__.py b/src/distutils2/tests/__init__.py
--- a/src/distutils2/tests/__init__.py
+++ b/src/distutils2/tests/__init__.py
@@ -1,20 +1,25 @@
-"""Test suite for distutils.
+"""Test suite for distutils2.
 
 This test suite consists of a collection of test modules in the
-distutils.tests package.  Each test module has a name starting with
+distutils2.tests package.  Each test module has a name starting with
 'test' and contains a function test_suite().  The function is expected
-to return an initialized unittest2.TestSuite instance.
+to return an initialized unittest.TestSuite instance.
 
-Tests for the command classes in the distutils.command package are
-included in distutils.tests as well, instead of using a separate
-distutils.command.tests package, since command identification is done
+Tests for the command classes in the distutils2.command package are
+included in distutils2.tests as well, instead of using a separate
+distutils2.command.tests package, since command identification is done
 by import rather than matching pre-defined names.
 
+Utility code is included in distutils2.tests.support.  Always import
+unittest from that module, it will be the right version (standard
+library unittest for 2.7 and higher, third-party unittest2 release for
+older versions).
 """
+
 import os
 import sys
 import warnings
-import unittest2
+from distutils2.tests.support import unittest
 
 from test.test_support import TESTFN    # use TESTFN from stdlib/test_support.
 
@@ -23,7 +28,7 @@
 verbose = 1
 
 def test_suite():
-    suite = unittest2.TestSuite()
+    suite = unittest.TestSuite()
     for fn in os.listdir(here):
         if fn.startswith("test") and fn.endswith(".py"):
             modname = "distutils2.tests." + fn[:-3]
@@ -42,17 +47,17 @@
 
 class BasicTestRunner:
     def run(self, test):
-        result = unittest2.TestResult()
+        result = unittest.TestResult()
         test(result)
         return result
 
 
 def _run_suite(suite, verbose_=1):
-    """Run tests from a unittest2.TestSuite-derived class."""
+    """Run tests from a unittest.TestSuite-derived class."""
     global verbose
     verbose = verbose_
     if verbose_:
-        runner = unittest2.TextTestRunner(sys.stdout, verbosity=2)
+        runner = unittest.TextTestRunner(sys.stdout, verbosity=2)
     else:
         runner = BasicTestRunner()
 
@@ -68,22 +73,22 @@
 
 
 def run_unittest(classes, verbose_=1):
-    """Run tests from unittest2.TestCase-derived classes.
+    """Run tests from unittest.TestCase-derived classes.
 
-    Extracted from stdlib test.test_support and modified to support unittest2.
+    Extracted from stdlib test.test_support and modified to support unittest.
     """
-    valid_types = (unittest2.TestSuite, unittest2.TestCase)
-    suite = unittest2.TestSuite()
+    valid_types = (unittest.TestSuite, unittest.TestCase)
+    suite = unittest.TestSuite()
     for cls in classes:
         if isinstance(cls, str):
             if cls in sys.modules:
-                suite.addTest(unittest2.findTestCases(sys.modules[cls]))
+                suite.addTest(unittest.findTestCases(sys.modules[cls]))
             else:
                 raise ValueError("str arguments must be keys in sys.modules")
         elif isinstance(cls, valid_types):
             suite.addTest(cls)
         else:
-            suite.addTest(unittest2.makeSuite(cls))
+            suite.addTest(unittest.makeSuite(cls))
     _run_suite(suite, verbose_)
 
 
@@ -128,4 +133,4 @@
 
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/support.py b/src/distutils2/tests/support.py
--- a/src/distutils2/tests/support.py
+++ b/src/distutils2/tests/support.py
@@ -1,5 +1,12 @@
-"""Support code for distutils test cases."""
+"""Support code for distutils2 test cases.
+
+Always import unittest from this module, it will be the right version
+(standard library unittest for 2.7 and higher, third-party unittest2
+release for older versions).
+"""
+
 import os
+import sys
 import shutil
 import tempfile
 from copy import deepcopy
@@ -9,6 +16,13 @@
 from distutils2.log import DEBUG, INFO, WARN, ERROR, FATAL
 from distutils2.core import Distribution
 
+if sys.version_info >= (2, 7):
+    # improved unittest package from 2.7's standard library
+    import unittest
+else:
+    # external release of same package for older versions
+    import unittest2 as unittest
+
 class LoggingSilencer(object):
 
     def setUp(self):
diff --git a/src/distutils2/tests/test_bdist.py b/src/distutils2/tests/test_bdist.py
--- a/src/distutils2/tests/test_bdist.py
+++ b/src/distutils2/tests/test_bdist.py
@@ -1,5 +1,4 @@
 """Tests for distutils.command.bdist."""
-import unittest2
 import sys
 import os
 import tempfile
@@ -10,12 +9,13 @@
 from distutils2.core import Distribution
 from distutils2.command.bdist import bdist
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 from distutils2.spawn import find_executable
 from distutils2 import spawn
 from distutils2.errors import DistutilsExecError
 
 class BuildTestCase(support.TempdirManager,
-                    unittest2.TestCase):
+                    unittest.TestCase):
 
     def test_formats(self):
 
@@ -38,7 +38,7 @@
         self.assertEquals(founded, formats)
 
 def test_suite():
-    return unittest2.makeSuite(BuildTestCase)
+    return unittest.makeSuite(BuildTestCase)
 
 if __name__ == '__main__':
     run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_bdist_dumb.py b/src/distutils2/tests/test_bdist_dumb.py
--- a/src/distutils2/tests/test_bdist_dumb.py
+++ b/src/distutils2/tests/test_bdist_dumb.py
@@ -1,6 +1,5 @@
 """Tests for distutils.command.bdist_dumb."""
 
-import unittest2
 import sys
 import os
 
@@ -12,6 +11,7 @@
     zlib = None
 
 from distutils2.tests import run_unittest
+from distutils2.tests.support import unittest
 
 from distutils2.core import Distribution
 from distutils2.command.bdist_dumb import bdist_dumb
@@ -29,7 +29,7 @@
 class BuildDumbTestCase(support.TempdirManager,
                         support.LoggingSilencer,
                         support.EnvironGuard,
-                        unittest2.TestCase):
+                        unittest.TestCase):
 
     def setUp(self):
         super(BuildDumbTestCase, self).setUp()
@@ -42,7 +42,7 @@
         sys.argv[:] = self.old_sys_argv[1]
         super(BuildDumbTestCase, self).tearDown()
 
-    @unittest2.skipUnless(zlib, "requires zlib")
+    @unittest.skipUnless(zlib, "requires zlib")
     def test_simple_built(self):
 
         # let's create a simple package
@@ -99,7 +99,7 @@
         self.assertEquals(cmd.format, default)
 
 def test_suite():
-    return unittest2.makeSuite(BuildDumbTestCase)
+    return unittest.makeSuite(BuildDumbTestCase)
 
 if __name__ == '__main__':
     run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_bdist_msi.py b/src/distutils2/tests/test_bdist_msi.py
--- a/src/distutils2/tests/test_bdist_msi.py
+++ b/src/distutils2/tests/test_bdist_msi.py
@@ -1,16 +1,16 @@
 """Tests for distutils.command.bdist_msi."""
-import unittest2
 import sys
 
 from distutils2.tests import run_unittest
 
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 class BDistMSITestCase(support.TempdirManager,
                        support.LoggingSilencer,
-                       unittest2.TestCase):
+                       unittest.TestCase):
 
-    @unittest2.skipUnless(sys.platform=="win32", "These tests are only for win32")
+    @unittest.skipUnless(sys.platform=="win32", "These tests are only for win32")
     def test_minial(self):
         # minimal test XXX need more tests
         from distutils2.command.bdist_msi import bdist_msi
@@ -19,7 +19,7 @@
         cmd.ensure_finalized()
 
 def test_suite():
-    return unittest2.makeSuite(BDistMSITestCase)
+    return unittest.makeSuite(BDistMSITestCase)
 
 if __name__ == '__main__':
     run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_bdist_wininst.py b/src/distutils2/tests/test_bdist_wininst.py
--- a/src/distutils2/tests/test_bdist_wininst.py
+++ b/src/distutils2/tests/test_bdist_wininst.py
@@ -1,14 +1,14 @@
 """Tests for distutils.command.bdist_wininst."""
-import unittest2
 
 from distutils2.tests import run_unittest
 
 from distutils2.command.bdist_wininst import bdist_wininst
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 class BuildWinInstTestCase(support.TempdirManager,
                            support.LoggingSilencer,
-                           unittest2.TestCase):
+                           unittest.TestCase):
 
     def test_get_exe_bytes(self):
 
@@ -26,7 +26,7 @@
         self.assertTrue(len(exe_file) > 10)
 
 def test_suite():
-    return unittest2.makeSuite(BuildWinInstTestCase)
+    return unittest.makeSuite(BuildWinInstTestCase)
 
 if __name__ == '__main__':
     run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_build.py b/src/distutils2/tests/test_build.py
--- a/src/distutils2/tests/test_build.py
+++ b/src/distutils2/tests/test_build.py
@@ -1,10 +1,10 @@
 """Tests for distutils.command.build."""
-import unittest2
 import os
 import sys
 
 from distutils2.command.build import build
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 try:
     from sysconfig import get_platform
 except ImportError:
@@ -12,7 +12,7 @@
 
 class BuildTestCase(support.TempdirManager,
                     support.LoggingSilencer,
-                    unittest2.TestCase):
+                    unittest.TestCase):
 
     def test_finalize_options(self):
         pkg_dir, dist = self.create_dist()
@@ -51,7 +51,7 @@
         self.assertEquals(cmd.executable, os.path.normpath(sys.executable))
 
 def test_suite():
-    return unittest2.makeSuite(BuildTestCase)
+    return unittest.makeSuite(BuildTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_build_clib.py b/src/distutils2/tests/test_build_clib.py
--- a/src/distutils2/tests/test_build_clib.py
+++ b/src/distutils2/tests/test_build_clib.py
@@ -1,5 +1,4 @@
 """Tests for distutils.command.build_clib."""
-import unittest2
 import os
 import sys
 
@@ -7,10 +6,11 @@
 from distutils2.errors import DistutilsSetupError
 from distutils2.tests import support
 from distutils2.spawn import find_executable
+from distutils2.tests.support import unittest
 
 class BuildCLibTestCase(support.TempdirManager,
                         support.LoggingSilencer,
-                        unittest2.TestCase):
+                        unittest.TestCase):
 
     def test_check_library_dist(self):
         pkg_dir, dist = self.create_dist()
@@ -137,7 +137,7 @@
         self.assertTrue('libfoo.a' in os.listdir(build_temp))
 
 def test_suite():
-    return unittest2.makeSuite(BuildCLibTestCase)
+    return unittest.makeSuite(BuildCLibTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_build_ext.py b/src/distutils2/tests/test_build_ext.py
--- a/src/distutils2/tests/test_build_ext.py
+++ b/src/distutils2/tests/test_build_ext.py
@@ -6,6 +6,7 @@
 import warnings
 
 import distutils2.tests
+from distutils2.tests.support import unittest
 from distutils2.core import Extension, Distribution
 from distutils2.command.build_ext import build_ext
 from distutils2.tests import support
@@ -17,7 +18,6 @@
 except ImportError:
     from distutils2._backport import sysconfig
 
-import unittest2
 
 # http://bugs.python.org/issue4373
 # Don't load the xx module more than once.
@@ -29,7 +29,7 @@
 
 class BuildExtTestCase(support.TempdirManager,
                        support.LoggingSilencer,
-                       unittest2.TestCase):
+                       unittest.TestCase):
     def setUp(self):
         # Create a simple test environment
         # Note that we're making changes to sys.path
@@ -46,7 +46,7 @@
             build_ext.USER_BASE = site.USER_BASE
 
     # XXX only works with 2.6 > -- dunno why yet
-    @unittest2.skipUnless(sys.version_info >= (2, 6,), 'works for >= 2.6')
+    @unittest.skipUnless(sys.version_info >= (2, 6,), 'works for >= 2.6')
     def test_build_ext(self):
         global ALREADY_TESTED
         xx_c = os.path.join(self.tmp_dir, 'xxmodule.c')
@@ -412,8 +412,8 @@
         if distutils2.tests.verbose:
             print ('test_build_ext: Cannot find source code (test'
                    ' must run in python build dir)')
-        return unittest2.TestSuite()
-    else: return unittest2.makeSuite(BuildExtTestCase)
+        return unittest.TestSuite()
+    else: return unittest.makeSuite(BuildExtTestCase)
 
 if __name__ == '__main__':
     distsutils2.tests.run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_build_py.py b/src/distutils2/tests/test_build_py.py
--- a/src/distutils2/tests/test_build_py.py
+++ b/src/distutils2/tests/test_build_py.py
@@ -3,18 +3,18 @@
 import os
 import sys
 import StringIO
-import unittest2
 
 from distutils2.command.build_py import build_py
 from distutils2.core import Distribution
 from distutils2.errors import DistutilsFileError
 
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 
 class BuildPyTestCase(support.TempdirManager,
                       support.LoggingSilencer,
-                      unittest2.TestCase):
+                      unittest.TestCase):
 
     def test_package_data(self):
         sources = self.mkdtemp()
@@ -90,7 +90,7 @@
             os.chdir(cwd)
             sys.stdout = old_stdout
 
-    @unittest2.skipUnless(hasattr(sys, 'dont_write_bytecode'),
+    @unittest.skipUnless(hasattr(sys, 'dont_write_bytecode'),
                          'dont_write_bytecode support')
     def test_dont_write_bytecode(self):
         # makes sure byte_compile is not used
@@ -109,7 +109,7 @@
         self.assertTrue('byte-compiling is disabled' in self.logs[0][1])
 
 def test_suite():
-    return unittest2.makeSuite(BuildPyTestCase)
+    return unittest.makeSuite(BuildPyTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_build_scripts.py b/src/distutils2/tests/test_build_scripts.py
--- a/src/distutils2/tests/test_build_scripts.py
+++ b/src/distutils2/tests/test_build_scripts.py
@@ -1,7 +1,6 @@
 """Tests for distutils.command.build_scripts."""
 
 import os
-import unittest2
 
 from distutils2.command.build_scripts import build_scripts
 from distutils2.core import Distribution
@@ -11,11 +10,12 @@
     from distutils2._backport import sysconfig
 
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 
 class BuildScriptsTestCase(support.TempdirManager,
                            support.LoggingSilencer,
-                           unittest2.TestCase):
+                           unittest.TestCase):
 
     def test_default_settings(self):
         cmd = self.get_build_scripts_cmd("/foo/bar", [])
@@ -106,7 +106,7 @@
             self.assertTrue(name in built)
 
 def test_suite():
-    return unittest2.makeSuite(BuildScriptsTestCase)
+    return unittest.makeSuite(BuildScriptsTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_ccompiler.py b/src/distutils2/tests/test_ccompiler.py
--- a/src/distutils2/tests/test_ccompiler.py
+++ b/src/distutils2/tests/test_ccompiler.py
@@ -1,11 +1,11 @@
 """Tests for distutils.ccompiler."""
 import os
-import unittest2
 from distutils2.tests import captured_stdout
 
 from distutils2.compiler.ccompiler import (gen_lib_options, CCompiler,
                                  get_default_compiler, customize_compiler)
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 class FakeCompiler(object):
     def library_dir_option(self, dir):
@@ -20,7 +20,7 @@
     def library_option(self, lib):
         return "-l" + lib
 
-class CCompilerTestCase(support.EnvironGuard, unittest2.TestCase):
+class CCompilerTestCase(support.EnvironGuard, unittest.TestCase):
 
     def test_gen_lib_options(self):
         compiler = FakeCompiler()
@@ -54,7 +54,7 @@
         self.assertEquals(comp.exes['archiver'], 'my_ar -arflags')
 
 def test_suite():
-    return unittest2.makeSuite(CCompilerTestCase)
+    return unittest.makeSuite(CCompilerTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_check.py b/src/distutils2/tests/test_check.py
--- a/src/distutils2/tests/test_check.py
+++ b/src/distutils2/tests/test_check.py
@@ -1,14 +1,14 @@
 """Tests for distutils.command.check."""
-import unittest2
 
 from distutils2.command.check import check
 from distutils2.metadata import _HAS_DOCUTILS
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 from distutils2.errors import DistutilsSetupError
 
 class CheckTestCase(support.LoggingSilencer,
                     support.TempdirManager,
-                    unittest2.TestCase):
+                    unittest.TestCase):
 
     def _run(self, metadata=None, **options):
         if metadata is None:
@@ -27,7 +27,7 @@
         # by default, check is checking the metadata
         # should have some warnings
         cmd = self._run()
-        self.assert_(len(cmd._warnings) > 0)
+        self.assertTrue(len(cmd._warnings) > 0)
 
         # now let's add the required fields
         # and run it again, to make sure we don't get
@@ -79,7 +79,7 @@
                                  'restructuredtext': 1})
 
 def test_suite():
-    return unittest2.makeSuite(CheckTestCase)
+    return unittest.makeSuite(CheckTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_clean.py b/src/distutils2/tests/test_clean.py
--- a/src/distutils2/tests/test_clean.py
+++ b/src/distutils2/tests/test_clean.py
@@ -1,15 +1,15 @@
 """Tests for distutils.command.clean."""
 import sys
 import os
-import unittest2
 import getpass
 
 from distutils2.command.clean import clean
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 class cleanTestCase(support.TempdirManager,
                     support.LoggingSilencer,
-                    unittest2.TestCase):
+                    unittest.TestCase):
 
     def test_simple_run(self):
         pkg_dir, dist = self.create_dist()
@@ -44,7 +44,7 @@
         cmd.run()
 
 def test_suite():
-    return unittest2.makeSuite(cleanTestCase)
+    return unittest.makeSuite(cleanTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_cmd.py b/src/distutils2/tests/test_cmd.py
--- a/src/distutils2/tests/test_cmd.py
+++ b/src/distutils2/tests/test_cmd.py
@@ -1,17 +1,17 @@
 """Tests for distutils.cmd."""
-import unittest2
 import os
 from distutils2.tests import captured_stdout, run_unittest
 
-from distutils2.cmd import Command
+from distutils2.command.cmd import Command
 from distutils2.dist import Distribution
 from distutils2.errors import DistutilsOptionError
+from distutils2.tests.support import unittest
 
 class MyCmd(Command):
     def initialize_options(self):
         pass
 
-class CommandTestCase(unittest2.TestCase):
+class CommandTestCase(unittest.TestCase):
 
     def setUp(self):
         dist = Distribution()
@@ -104,7 +104,7 @@
         self.assertRaises(DistutilsOptionError, cmd.ensure_dirname, 'option2')
 
 def test_suite():
-    return unittest2.makeSuite(CommandTestCase)
+    return unittest.makeSuite(CommandTestCase)
 
 if __name__ == '__main__':
     run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_config.py b/src/distutils2/tests/test_config.py
--- a/src/distutils2/tests/test_config.py
+++ b/src/distutils2/tests/test_config.py
@@ -1,7 +1,6 @@
 """Tests for distutils.pypirc.pypirc."""
 import sys
 import os
-import unittest2
 import tempfile
 import shutil
 
@@ -11,6 +10,7 @@
 from distutils2.log import WARN
 
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 PYPIRC = """\
 [distutils]
@@ -50,7 +50,7 @@
 class PyPIRCCommandTestCase(support.TempdirManager,
                             support.LoggingSilencer,
                             support.EnvironGuard,
-                            unittest2.TestCase):
+                            unittest.TestCase):
 
     def setUp(self):
         """Patches the environment."""
@@ -112,7 +112,7 @@
         self.assertEquals(content, WANTED)
 
 def test_suite():
-    return unittest2.makeSuite(PyPIRCCommandTestCase)
+    return unittest.makeSuite(PyPIRCCommandTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_config_cmd.py b/src/distutils2/tests/test_config_cmd.py
--- a/src/distutils2/tests/test_config_cmd.py
+++ b/src/distutils2/tests/test_config_cmd.py
@@ -1,15 +1,15 @@
 """Tests for distutils.command.config."""
-import unittest2
 import os
 import sys
 
 from distutils2.command.config import dump_file, config
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 from distutils2 import log
 
 class ConfigTestCase(support.LoggingSilencer,
                      support.TempdirManager,
-                     unittest2.TestCase):
+                     unittest.TestCase):
 
     def _info(self, msg, *args):
         for line in msg.splitlines():
@@ -83,7 +83,7 @@
             self.assertTrue(not os.path.exists(f))
 
 def test_suite():
-    return unittest2.makeSuite(ConfigTestCase)
+    return unittest.makeSuite(ConfigTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_converter.py b/src/distutils2/tests/test_converter.py
--- a/src/distutils2/tests/test_converter.py
+++ b/src/distutils2/tests/test_converter.py
@@ -1,8 +1,8 @@
 """Tests for distutils.converter."""
 import os
 import sys
-import unittest2
 from distutils2.converter import DistutilsRefactoringTool
+from distutils2.tests.support import unittest
 
 _CURDIR = os.path.dirname(__file__)
 
@@ -15,9 +15,9 @@
         f.close()
 
 
-class ConverterTestCase(unittest2.TestCase):
+class ConverterTestCase(unittest.TestCase):
 
-    @unittest2.skipUnless(not sys.version < '2.6', 'Needs Python >=2.6')
+    @unittest.skipUnless(not sys.version < '2.6', 'Needs Python >=2.6')
     def test_conversions(self):
         # for all XX_before in the conversions/ dir
         # we run the refactoring tool
@@ -33,7 +33,7 @@
             self.assertEquals(str(res), wanted)
 
 def test_suite():
-    return unittest2.makeSuite(ConverterTestCase)
+    return unittest.makeSuite(ConverterTestCase)
 
 if __name__ == '__main__':
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_core.py b/src/distutils2/tests/test_core.py
--- a/src/distutils2/tests/test_core.py
+++ b/src/distutils2/tests/test_core.py
@@ -6,8 +6,8 @@
 import shutil
 import sys
 from distutils2.tests import captured_stdout
-import unittest2
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 # setup script that uses __file__
 setup_using___file__ = """\
@@ -28,7 +28,7 @@
 """
 
 
-class CoreTestCase(support.EnvironGuard, unittest2.TestCase):
+class CoreTestCase(support.EnvironGuard, unittest.TestCase):
 
     def setUp(self):
         super(CoreTestCase, self).setUp()
@@ -92,7 +92,7 @@
         self.assertEqual(cwd, output)
 
 def test_suite():
-    return unittest2.makeSuite(CoreTestCase)
+    return unittest.makeSuite(CoreTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_cygwinccompiler.py b/src/distutils2/tests/test_cygwinccompiler.py
--- a/src/distutils2/tests/test_cygwinccompiler.py
+++ b/src/distutils2/tests/test_cygwinccompiler.py
@@ -1,5 +1,4 @@
 """Tests for distutils.cygwinccompiler."""
-import unittest2
 import sys
 import os
 import warnings
@@ -18,9 +17,10 @@
                                        get_msvcr, RE_VERSION)
 from distutils2.util import get_compiler_versions
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 class CygwinCCompilerTestCase(support.TempdirManager,
-                              unittest2.TestCase):
+                              unittest.TestCase):
 
     def setUp(self):
         super(CygwinCCompilerTestCase, self).setUp()
@@ -94,7 +94,7 @@
 
 
 def test_suite():
-    return unittest2.makeSuite(CygwinCCompilerTestCase)
+    return unittest.makeSuite(CygwinCCompilerTestCase)
 
 if __name__ == '__main__':
     run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_dist.py b/src/distutils2/tests/test_dist.py
--- a/src/distutils2/tests/test_dist.py
+++ b/src/distutils2/tests/test_dist.py
@@ -4,15 +4,15 @@
 import os
 import StringIO
 import sys
-import unittest2
 import warnings
 import textwrap
 
 from distutils2.dist import Distribution, fix_help_options, DistributionMetadata
-from distutils2.cmd import Command
+from distutils2.command.cmd import Command
 import distutils2.dist
 from distutils2.tests import TESTFN, captured_stdout
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 class test_dist(Command):
     """Sample distutils2 extension command."""
@@ -40,7 +40,7 @@
 class DistributionTestCase(support.TempdirManager,
                            support.LoggingSilencer,
                            support.EnvironGuard,
-                           unittest2.TestCase):
+                           unittest.TestCase):
 
     def setUp(self):
         super(DistributionTestCase, self).setUp()
@@ -242,7 +242,7 @@
 
 
 class MetadataTestCase(support.TempdirManager, support.EnvironGuard,
-                       unittest2.TestCase):
+                       unittest.TestCase):
 
     def setUp(self):
         super(MetadataTestCase, self).setUp()
@@ -422,10 +422,10 @@
         self.assertEquals(metadata['requires-dist'], ['foo'])
 
 def test_suite():
-    suite = unittest2.TestSuite()
-    suite.addTest(unittest2.makeSuite(DistributionTestCase))
-    suite.addTest(unittest2.makeSuite(MetadataTestCase))
+    suite = unittest.TestSuite()
+    suite.addTest(unittest.makeSuite(DistributionTestCase))
+    suite.addTest(unittest.makeSuite(MetadataTestCase))
     return suite
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_emxccompiler.py b/src/distutils2/tests/test_emxccompiler.py
--- a/src/distutils2/tests/test_emxccompiler.py
+++ b/src/distutils2/tests/test_emxccompiler.py
@@ -1,5 +1,4 @@
 """Tests for distutils.emxccompiler."""
-import unittest2
 import sys
 import os
 import warnings
@@ -10,14 +9,15 @@
 from distutils2.compiler.emxccompiler import get_versions
 from distutils2.util import get_compiler_versions
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 class EmxCCompilerTestCase(support.TempdirManager,
-                           unittest2.TestCase):
+                           unittest.TestCase):
 
     pass
 
 def test_suite():
-    return unittest2.makeSuite(EmxCCompilerTestCase)
+    return unittest.makeSuite(EmxCCompilerTestCase)
 
 if __name__ == '__main__':
     run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_extension.py b/src/distutils2/tests/test_extension.py
--- a/src/distutils2/tests/test_extension.py
+++ b/src/distutils2/tests/test_extension.py
@@ -1,16 +1,16 @@
 """Tests for distutils.extension."""
-import unittest2
 import os
 import warnings
 
 from distutils2.extension import Extension
+from distutils2.tests.support import unittest
 
-class ExtensionTestCase(unittest2.TestCase):
+class ExtensionTestCase(unittest.TestCase):
 
     pass
 
 def test_suite():
-    return unittest2.makeSuite(ExtensionTestCase)
+    return unittest.makeSuite(ExtensionTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_install.py b/src/distutils2/tests/test_install.py
--- a/src/distutils2/tests/test_install.py
+++ b/src/distutils2/tests/test_install.py
@@ -3,7 +3,6 @@
 import os
 import os.path
 import sys
-import unittest2
 import site
 
 from distutils2._backport import sysconfig
@@ -22,11 +21,12 @@
 from distutils2.errors import DistutilsOptionError
 
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 class InstallTestCase(support.TempdirManager,
                       support.EnvironGuard,
                       support.LoggingSilencer,
-                      unittest2.TestCase):
+                      unittest.TestCase):
 
     def test_home_installation_scheme(self):
         # This ensure two things:
@@ -214,7 +214,7 @@
         self.assertTrue(len(self.logs) > old_logs_len)
 
 def test_suite():
-    return unittest2.makeSuite(InstallTestCase)
+    return unittest.makeSuite(InstallTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_install_data.py b/src/distutils2/tests/test_install_data.py
--- a/src/distutils2/tests/test_install_data.py
+++ b/src/distutils2/tests/test_install_data.py
@@ -1,16 +1,16 @@
 """Tests for distutils.command.install_data."""
 import sys
 import os
-import unittest2
 import getpass
 
 from distutils2.command.install_data import install_data
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 class InstallDataTestCase(support.TempdirManager,
                           support.LoggingSilencer,
                           support.EnvironGuard,
-                          unittest2.TestCase):
+                          unittest.TestCase):
 
     def test_simple_run(self):
         pkg_dir, dist = self.create_dist()
@@ -70,7 +70,7 @@
         self.assertTrue(os.path.exists(os.path.join(inst, rone)))
 
 def test_suite():
-    return unittest2.makeSuite(InstallDataTestCase)
+    return unittest.makeSuite(InstallDataTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_install_headers.py b/src/distutils2/tests/test_install_headers.py
--- a/src/distutils2/tests/test_install_headers.py
+++ b/src/distutils2/tests/test_install_headers.py
@@ -1,16 +1,16 @@
 """Tests for distutils.command.install_headers."""
 import sys
 import os
-import unittest2
 import getpass
 
 from distutils2.command.install_headers import install_headers
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 class InstallHeadersTestCase(support.TempdirManager,
                              support.LoggingSilencer,
                              support.EnvironGuard,
-                             unittest2.TestCase):
+                             unittest.TestCase):
 
     def test_simple_run(self):
         # we have two headers
@@ -34,7 +34,7 @@
         self.assertEquals(len(cmd.get_outputs()), 2)
 
 def test_suite():
-    return unittest2.makeSuite(InstallHeadersTestCase)
+    return unittest.makeSuite(InstallHeadersTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_install_lib.py b/src/distutils2/tests/test_install_lib.py
--- a/src/distutils2/tests/test_install_lib.py
+++ b/src/distutils2/tests/test_install_lib.py
@@ -1,12 +1,12 @@
 """Tests for distutils.command.install_data."""
 import sys
 import os
-import unittest2
 
 from distutils2.command.install_lib import install_lib
 from distutils2.extension import Extension
 from distutils2.tests import support
 from distutils2.errors import DistutilsOptionError
+from distutils2.tests.support import unittest
 
 try:
     no_bytecode = sys.dont_write_bytecode
@@ -18,7 +18,7 @@
 class InstallLibTestCase(support.TempdirManager,
                          support.LoggingSilencer,
                          support.EnvironGuard,
-                         unittest2.TestCase):
+                         unittest.TestCase):
 
     def test_finalize_options(self):
         pkg_dir, dist = self.create_dist()
@@ -38,7 +38,7 @@
         cmd.finalize_options()
         self.assertEquals(cmd.optimize, 2)
 
-    @unittest2.skipIf(no_bytecode, 'byte-compile not supported')
+    @unittest.skipIf(no_bytecode, 'byte-compile not supported')
     def test_byte_compile(self):
         pkg_dir, dist = self.create_dist()
         cmd = install_lib(dist)
@@ -84,7 +84,7 @@
         # get_input should return 2 elements
         self.assertEquals(len(cmd.get_inputs()), 2)
 
-    @unittest2.skipUnless(bytecode_support, 'sys.dont_write_bytecode not supported')
+    @unittest.skipUnless(bytecode_support, 'sys.dont_write_bytecode not supported')
     def test_dont_write_bytecode(self):
         # makes sure byte_compile is not used
         pkg_dir, dist = self.create_dist()
@@ -102,7 +102,7 @@
         self.assertTrue('byte-compiling is disabled' in self.logs[0][1])
 
 def test_suite():
-    return unittest2.makeSuite(InstallLibTestCase)
+    return unittest.makeSuite(InstallLibTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_install_scripts.py b/src/distutils2/tests/test_install_scripts.py
--- a/src/distutils2/tests/test_install_scripts.py
+++ b/src/distutils2/tests/test_install_scripts.py
@@ -1,17 +1,17 @@
 """Tests for distutils.command.install_scripts."""
 
 import os
-import unittest2
 
 from distutils2.command.install_scripts import install_scripts
 from distutils2.core import Distribution
 
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 
 class InstallScriptsTestCase(support.TempdirManager,
                              support.LoggingSilencer,
-                             unittest2.TestCase):
+                             unittest.TestCase):
 
     def test_default_settings(self):
         dist = Distribution()
@@ -73,7 +73,7 @@
 
 
 def test_suite():
-    return unittest2.makeSuite(InstallScriptsTestCase)
+    return unittest.makeSuite(InstallScriptsTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_manifest.py b/src/distutils2/tests/test_manifest.py
--- a/src/distutils2/tests/test_manifest.py
+++ b/src/distutils2/tests/test_manifest.py
@@ -1,10 +1,10 @@
 """Tests for distutils.manifest."""
-import unittest2
 import os
 import sys
 import logging
 
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 from distutils2.manifest import Manifest
 
 _MANIFEST = """\
@@ -18,7 +18,7 @@
 """
 
 class ManifestTestCase(support.TempdirManager,
-                       unittest2.TestCase):
+                       unittest.TestCase):
 
     def test_manifest_reader(self):
 
@@ -51,7 +51,7 @@
 
 
 def test_suite():
-    return unittest2.makeSuite(ManifestTestCase)
+    return unittest.makeSuite(ManifestTestCase)
 
 if __name__ == '__main__':
     run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_metadata.py b/src/distutils2/tests/test_metadata.py
--- a/src/distutils2/tests/test_metadata.py
+++ b/src/distutils2/tests/test_metadata.py
@@ -1,13 +1,13 @@
 """Tests for distutils.command.bdist."""
-import unittest2
 import os
 import sys
 from StringIO import StringIO
 
 from distutils2.metadata import (DistributionMetadata, _interpret,
                                  PKG_INFO_PREFERRED_VERSION)
+from distutils2.tests.support import unittest
 
-class DistributionMetadataTestCase(unittest2.TestCase):
+class DistributionMetadataTestCase(unittest.TestCase):
 
 
     def test_interpret(self):
@@ -64,7 +64,7 @@
         res = res.read()
         f = open(PKG_INFO)
         wanted = f.read()
-        self.assert_('Keywords: keyring,password,crypt' in res)
+        self.assertTrue('Keywords: keyring,password,crypt' in res)
         f.close()
 
     def test_metadata_markers(self):
@@ -228,7 +228,7 @@
 
 
 def test_suite():
-    return unittest2.makeSuite(DistributionMetadataTestCase)
+    return unittest.makeSuite(DistributionMetadataTestCase)
 
 if __name__ == '__main__':
     run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_msvc9compiler.py b/src/distutils2/tests/test_msvc9compiler.py
--- a/src/distutils2/tests/test_msvc9compiler.py
+++ b/src/distutils2/tests/test_msvc9compiler.py
@@ -1,10 +1,10 @@
 """Tests for distutils.msvc9compiler."""
 import sys
-import unittest2
 import os
 
 from distutils2.errors import DistutilsPlatformError
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 _MANIFEST = """\
 <?xml version="1.0" encoding="UTF-8" standalone="yes"?>
@@ -62,9 +62,9 @@
 
 
 class msvc9compilerTestCase(support.TempdirManager,
-                            unittest2.TestCase):
+                            unittest.TestCase):
 
-    @unittest2.skipUnless(sys.platform=="win32", "These tests are only for win32")
+    @unittest.skipUnless(sys.platform=="win32", "These tests are only for win32")
     def test_no_compiler(self):
         # makes sure query_vcvarsall throws
         # a DistutilsPlatformError if the compiler
@@ -86,7 +86,7 @@
         finally:
             msvc9compiler.find_vcvarsall = old_find_vcvarsall
 
-    @unittest2.skipUnless(sys.platform=="win32", "These tests are only for win32")
+    @unittest.skipUnless(sys.platform=="win32", "These tests are only for win32")
     def test_reg_class(self):
         from distutils2.msvccompiler import get_build_version
         if get_build_version() < 8.0:
@@ -110,7 +110,7 @@
         keys = Reg.read_keys(HKCU, r'Control Panel')
         self.assertTrue('Desktop' in keys)
 
-    @unittest2.skipUnless(sys.platform=="win32", "These tests are only for win32")
+    @unittest.skipUnless(sys.platform=="win32", "These tests are only for win32")
     def test_remove_visual_c_ref(self):
         from distutils2.msvc9compiler import MSVCCompiler
         tempdir = self.mkdtemp()
@@ -133,7 +133,7 @@
 
 
 def test_suite():
-    return unittest2.makeSuite(msvc9compilerTestCase)
+    return unittest.makeSuite(msvc9compilerTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_pypi_versions.py b/src/distutils2/tests/test_pypi_versions.py
--- a/src/distutils2/tests/test_pypi_versions.py
+++ b/src/distutils2/tests/test_pypi_versions.py
@@ -17,9 +17,9 @@
 
 import xmlrpclib
 import os.path
-import unittest2
 
 from distutils2.version import suggest_normalized_version
+from distutils2.tests.support import unittest
 
 def test_pypi():
     #
@@ -121,11 +121,11 @@
     print "Have Suggestion : ", have_sugg, pct % (have_sugg/total_versions*100,)
     print "No Suggestion   : ", no_sugg, pct % (no_sugg/total_versions*100,)
 
-class TestPyPI(unittest2.TestCase):
+class TestPyPI(unittest.TestCase):
     pass
 
 def test_suite():
-    return unittest2.makeSuite(TestPyPI)
+    return unittest.makeSuite(TestPyPI)
 
 if __name__ == '__main__':
     run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_register.py b/src/distutils2/tests/test_register.py
--- a/src/distutils2/tests/test_register.py
+++ b/src/distutils2/tests/test_register.py
@@ -2,7 +2,6 @@
 # -*- encoding: utf8 -*-
 import sys
 import os
-import unittest2
 import getpass
 import urllib2
 import warnings
@@ -19,6 +18,7 @@
 from distutils2.errors import DistutilsSetupError
 
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 from distutils2.tests.test_config import PYPIRC, PyPIRCCommandTestCase
 
 PYPIRC_NOPASSWORD = """\
@@ -192,7 +192,7 @@
         self.assertEquals(headers['Content-length'], '290')
         self.assertTrue('tarek' in req.data)
 
-    @unittest2.skipUnless(DOCUTILS_SUPPORT, 'needs docutils')
+    @unittest.skipUnless(DOCUTILS_SUPPORT, 'needs docutils')
     def test_strict(self):
         # testing the script option
         # when on, the register command stops if
@@ -250,7 +250,7 @@
         self.assertEquals(data['requires_dist'], ['lxml'])
 
 def test_suite():
-    return unittest2.makeSuite(RegisterTestCase)
+    return unittest.makeSuite(RegisterTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_sdist.py b/src/distutils2/tests/test_sdist.py
--- a/src/distutils2/tests/test_sdist.py
+++ b/src/distutils2/tests/test_sdist.py
@@ -1,6 +1,5 @@
 """Tests for distutils.command.sdist."""
 import os
-import unittest2
 import shutil
 import zipfile
 import tarfile
@@ -29,6 +28,7 @@
 from distutils2.command.sdist import sdist
 from distutils2.command.sdist import show_formats
 from distutils2.core import Distribution
+from distutils2.tests.support import unittest
 from distutils2.tests.test_config import PyPIRCCommandTestCase
 from distutils2.errors import DistutilsExecError, DistutilsOptionError
 from distutils2.spawn import find_executable
@@ -97,7 +97,7 @@
         cmd.warn = _warn
         return dist, cmd
 
-    @unittest2.skipUnless(zlib, "requires zlib")
+    @unittest.skipUnless(zlib, "requires zlib")
     def test_prune_file_list(self):
         # this test creates a package with some vcs dirs in it
         # and launch sdist to make sure they get pruned
@@ -139,7 +139,7 @@
         # making sure everything has been pruned correctly
         self.assertEquals(len(content), 4)
 
-    @unittest2.skipUnless(zlib, "requires zlib")
+    @unittest.skipUnless(zlib, "requires zlib")
     def test_make_distribution(self):
 
         # check if tar and gzip are installed
@@ -176,7 +176,7 @@
         self.assertEquals(result,
                 ['fake-1.0.tar', 'fake-1.0.tar.gz'])
 
-    @unittest2.skipUnless(zlib, "requires zlib")
+    @unittest.skipUnless(zlib, "requires zlib")
     def test_add_defaults(self):
 
         # http://bugs.python.org/issue2279
@@ -238,7 +238,7 @@
         manifest = open(join(self.tmp_dir, 'MANIFEST')).read()
         self.assertEquals(manifest, MANIFEST % {'sep': os.sep})
 
-    @unittest2.skipUnless(zlib, "requires zlib")
+    @unittest.skipUnless(zlib, "requires zlib")
     def test_metadata_check_option(self):
         # testing the `medata-check` option
         dist, cmd = self.get_cmd(metadata={})
@@ -293,8 +293,8 @@
         cmd.formats = 'supazipa'
         self.assertRaises(DistutilsOptionError, cmd.finalize_options)
 
-    @unittest2.skipUnless(zlib, "requires zlib")
-    @unittest2.skipUnless(UID_GID_SUPPORT, "Requires grp and pwd support")
+    @unittest.skipUnless(zlib, "requires zlib")
+    @unittest.skipUnless(UID_GID_SUPPORT, "Requires grp and pwd support")
     def test_make_distribution_owner_group(self):
 
         # check if tar and gzip are installed
@@ -357,7 +357,7 @@
         self.assertIn('MANIFEST.in', cmd.filelist.files)
 
 def test_suite():
-    return unittest2.makeSuite(SDistTestCase)
+    return unittest.makeSuite(SDistTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_spawn.py b/src/distutils2/tests/test_spawn.py
--- a/src/distutils2/tests/test_spawn.py
+++ b/src/distutils2/tests/test_spawn.py
@@ -1,5 +1,4 @@
 """Tests for distutils.spawn."""
-import unittest2
 import os
 import time
 from distutils2.tests import captured_stdout
@@ -8,10 +7,11 @@
 from distutils2.spawn import spawn, find_executable
 from distutils2.errors import DistutilsExecError
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 class SpawnTestCase(support.TempdirManager,
                     support.LoggingSilencer,
-                    unittest2.TestCase):
+                    unittest.TestCase):
 
     def test_nt_quote_args(self):
 
@@ -23,7 +23,7 @@
             self.assertEquals(res, wanted)
 
 
-    @unittest2.skipUnless(os.name in ('nt', 'posix'),
+    @unittest.skipUnless(os.name in ('nt', 'posix'),
                          'Runs only under posix or nt')
     def test_spawn(self):
         tmpdir = self.mkdtemp()
@@ -54,7 +54,7 @@
         spawn([exe])  # should work without any error
 
 def test_suite():
-    return unittest2.makeSuite(SpawnTestCase)
+    return unittest.makeSuite(SpawnTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_unixccompiler.py b/src/distutils2/tests/test_unixccompiler.py
--- a/src/distutils2/tests/test_unixccompiler.py
+++ b/src/distutils2/tests/test_unixccompiler.py
@@ -1,6 +1,5 @@
 """Tests for distutils.unixccompiler."""
 import sys
-import unittest2
 
 try:
     import sysconfig
@@ -8,8 +7,9 @@
     from distutils2._backport import sysconfig
 
 from distutils2.compiler.unixccompiler import UnixCCompiler
+from distutils2.tests.support import unittest
 
-class UnixCCompilerTestCase(unittest2.TestCase):
+class UnixCCompilerTestCase(unittest.TestCase):
 
     def setUp(self):
         self._backup_platform = sys.platform
@@ -127,7 +127,7 @@
 
 
 def test_suite():
-    return unittest2.makeSuite(UnixCCompilerTestCase)
+    return unittest.makeSuite(UnixCCompilerTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_upload.py b/src/distutils2/tests/test_upload.py
--- a/src/distutils2/tests/test_upload.py
+++ b/src/distutils2/tests/test_upload.py
@@ -2,13 +2,13 @@
 # -*- encoding: utf8 -*-
 import sys
 import os
-import unittest2
 
 from distutils2.command import upload as upload_mod
 from distutils2.command.upload import upload
 from distutils2.core import Distribution
 
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 from distutils2.tests.test_config import PYPIRC, PyPIRCCommandTestCase
 
 PYPIRC_LONG_PASSWORD = """\
@@ -116,7 +116,7 @@
         # what did we send ?
         self.assertIn('d??d??', self.last_open.req.data)
         headers = dict(self.last_open.req.headers)
-        self.assert_(headers['Content-length'] > 2000)
+        self.assertTrue(int(headers['Content-length']) < 2000)
         self.assertTrue(headers['Content-type'].startswith('multipart/form-data'))
         self.assertEquals(self.last_open.req.get_method(), 'POST')
         self.assertEquals(self.last_open.req.get_full_url(),
@@ -126,7 +126,7 @@
         self.assertFalse('\n' in auth)
 
 def test_suite():
-    return unittest2.makeSuite(uploadTestCase)
+    return unittest.makeSuite(uploadTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_util.py b/src/distutils2/tests/test_util.py
--- a/src/distutils2/tests/test_util.py
+++ b/src/distutils2/tests/test_util.py
@@ -1,7 +1,6 @@
 """Tests for distutils.util."""
 import os
 import sys
-import unittest2
 from copy import copy
 from StringIO import StringIO
 import subprocess
@@ -17,6 +16,7 @@
                             byte_compile, find_packages)
 from distutils2 import util
 from distutils2.tests import support
+from distutils2.tests.support import unittest
 
 class FakePopen(object):
     test_class = None
@@ -34,7 +34,7 @@
 
 class UtilTestCase(support.EnvironGuard,
                    support.TempdirManager,
-                   unittest2.TestCase):
+                   unittest.TestCase):
 
     def setUp(self):
         super(UtilTestCase, self).setUp()
@@ -243,7 +243,7 @@
         res = get_compiler_versions()
         self.assertEquals(res[2], None)
 
-    @unittest2.skipUnless(hasattr(sys, 'dont_write_bytecode'),
+    @unittest.skipUnless(hasattr(sys, 'dont_write_bytecode'),
                           'no dont_write_bytecode support')
     def test_dont_write_bytecode(self):
         # makes sure byte_compile raise a DistutilsError
@@ -298,7 +298,7 @@
 
 
 def test_suite():
-    return unittest2.makeSuite(UtilTestCase)
+    return unittest.makeSuite(UtilTestCase)
 
 if __name__ == "__main__":
-    unittest2.main(defaultTest="test_suite")
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_version.py b/src/distutils2/tests/test_version.py
--- a/src/distutils2/tests/test_version.py
+++ b/src/distutils2/tests/test_version.py
@@ -1,5 +1,4 @@
 """Tests for distutils.version."""
-import unittest
 import doctest
 import os
 
@@ -7,6 +6,7 @@
 from distutils2.version import IrrationalVersionError
 from distutils2.version import suggest_normalized_version as suggest
 from distutils2.version import VersionPredicate
+from distutils2.tests.support import unittest
 
 class VersionTestCase(unittest.TestCase):
 
diff --git a/src/distutils2/version.py b/src/distutils2/version.py
--- a/src/distutils2/version.py
+++ b/src/distutils2/version.py
@@ -3,6 +3,10 @@
 
 from distutils2.errors import IrrationalVersionError, HugeMajorVersionNumError
 
+__all__ = ['NormalizedVersion', 'suggest_normalized_version',
+           'VersionPredicate', 'is_valid_version', 'is_valid_versions',
+           'is_valid_predicate']
+
 # A marker used in the second and third parts of the `parts` tuple, for
 # versions that don't have those segments, to sort properly. An example
 # of versions in sort order ('highest' last):
@@ -18,9 +22,9 @@
 #                                                              |
 #   'dev' < 'f' ----------------------------------------------/
 # Other letters would do, but 'f' for 'final' is kind of nice.
-FINAL_MARKER = ('f',)
+_FINAL_MARKER = ('f',)
 
-VERSION_RE = re.compile(r'''
+_VERSION_RE = re.compile(r'''
     ^
     (?P<version>\d+\.\d+)          # minimum 'N.N'
     (?P<extraversion>(?:\.\d+)*)   # any number of extra '.N' segments
@@ -70,13 +74,13 @@
         self._parse(s, error_on_huge_major_num)
 
     @classmethod
-    def from_parts(cls, version, prerelease=FINAL_MARKER,
-                   devpost=FINAL_MARKER):
+    def from_parts(cls, version, prerelease=_FINAL_MARKER,
+                   devpost=_FINAL_MARKER):
         return cls(cls.parts_to_str((version, prerelease, devpost)))
 
     def _parse(self, s, error_on_huge_major_num=True):
         """Parses a string version into parts."""
-        match = VERSION_RE.search(s)
+        match = _VERSION_RE.search(s)
         if not match:
             raise IrrationalVersionError(s)
 
@@ -98,7 +102,7 @@
                                          pad_zeros_length=1)
             parts.append(tuple(block))
         else:
-            parts.append(FINAL_MARKER)
+            parts.append(_FINAL_MARKER)
 
         # postdev
         if groups.get('postdev'):
@@ -106,14 +110,14 @@
             dev = groups.get('dev')
             postdev = []
             if post is not None:
-                postdev.extend([FINAL_MARKER[0], 'post', int(post)])
+                postdev.extend([_FINAL_MARKER[0], 'post', int(post)])
                 if dev is None:
-                    postdev.append(FINAL_MARKER[0])
+                    postdev.append(_FINAL_MARKER[0])
             if dev is not None:
                 postdev.extend(['dev', int(dev)])
             parts.append(tuple(postdev))
         else:
-            parts.append(FINAL_MARKER)
+            parts.append(_FINAL_MARKER)
         self.parts = tuple(parts)
         if error_on_huge_major_num and self.parts[0][0] > 1980:
             raise HugeMajorVersionNumError("huge major version number, %r, "
@@ -154,10 +158,10 @@
         # XXX This doesn't check for invalid tuples
         main, prerel, postdev = parts
         s = '.'.join(str(v) for v in main)
-        if prerel is not FINAL_MARKER:
+        if prerel is not _FINAL_MARKER:
             s += prerel[0]
             s += '.'.join(str(v) for v in prerel[1:])
-        if postdev and postdev is not FINAL_MARKER:
+        if postdev and postdev is not _FINAL_MARKER:
             if postdev[0] == 'f':
                 postdev = postdev[1:]
             i = 0
@@ -197,6 +201,9 @@
     def __ge__(self, other):
         return self.__eq__(other) or self.__gt__(other)
 
+    # See http://docs.python.org/reference/datamodel#object.__hash__
+    __hash__ = object.__hash__
+
 def suggest_normalized_version(s):
     """Suggest a normalized version close to the given version string.
 
@@ -361,7 +368,7 @@
                 return False
         return True
 
-class Versions(VersionPredicate):
+class _Versions(VersionPredicate):
     def __init__(self, predicate):
         predicate = predicate.strip()
         match = _PLAIN_VERSIONS.match(predicate)
@@ -372,7 +379,7 @@
         self.predicates = [_split_predicate(pred.strip())
                            for pred in predicates.split(',')]
 
-class Version(VersionPredicate):
+class _Version(VersionPredicate):
     def __init__(self, predicate):
         predicate = predicate.strip()
         match = _PLAIN_VERSIONS.match(predicate)
@@ -391,7 +398,7 @@
 
 def is_valid_versions(predicate):
     try:
-        Versions(predicate)
+        _Versions(predicate)
     except (ValueError, IrrationalVersionError):
         return False
     else:
@@ -399,7 +406,7 @@
 
 def is_valid_version(predicate):
     try:
-        Version(predicate)
+        _Version(predicate)
     except (ValueError, IrrationalVersionError):
         return False
     else:
diff --git a/src/runtests.py b/src/runtests.py
--- a/src/runtests.py
+++ b/src/runtests.py
@@ -26,9 +26,9 @@
 
 if __name__ == "__main__":
     try:
-        import unittest2
+        from distutils2.tests.support import unittest
     except ImportError:
-        print('!!! You need to install unittest2')
+        print('Error: You have to install unittest2')
         sys.exit(1)
 
     sys.exit(test_main())
diff --git a/src/setup.py b/src/setup.py
--- a/src/setup.py
+++ b/src/setup.py
@@ -17,19 +17,17 @@
     f.close()
 
 def get_tip_revision(path=os.getcwd()):
+    from subprocess import Popen, PIPE
     try:
-        from mercurial.hg import repository
-        from mercurial.ui import ui
-        from mercurial import node
-        from mercurial.error import RepoError
-    except ImportError:
+        cmd = Popen(['hg', 'tip', '--template', '{rev}', '-R', path],
+                    stdout=PIPE, stderr=PIPE)
+    except OSError:
         return 0
-    try:
-        repo = repository(ui(), path)
-        tip = repo.changelog.tip()
-        return repo.changelog.rev(tip)
-    except RepoError:
+    rev = cmd.stdout.read()
+    if rev == '':
+        # there has been an error in the command
         return 0
+    return int(rev)
 
 DEV_SUFFIX = '.dev%d' % get_tip_revision('..')
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Thu Jul 15 01:38:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Thu, 15 Jul 2010 01:38:06 +0200
Subject: [Python-checkins] distutils2: Restored logging of directory creation
Message-ID: <E1OZBWg-0001H1-8f@dinsdale.python.org>

tarek.ziade pushed a77b4594293d to distutils2:

http://hg.python.org/distutils2/rev/a77b4594293d
changeset:   364:a77b4594293d
user:        Jeremy Kloth <jeremy.kloth at gmail.com>
date:        Thu Jul 08 10:26:49 2010 -0600
summary:     Restored logging of directory creation
files:       src/distutils2/command/cmd.py, src/distutils2/compiler/ccompiler.py

diff --git a/src/distutils2/command/cmd.py b/src/distutils2/command/cmd.py
--- a/src/distutils2/command/cmd.py
+++ b/src/distutils2/command/cmd.py
@@ -373,8 +373,10 @@
         if os.path.isdir(name) or name == '':
             return
         if dry_run:
+            head = ''
             for part in name.split(os.sep):
-                self.log(part)
+                log.info("created directory %s%s", head, part)
+                head += part + os.sep
             return
         os.makedirs(name, mode)
 
diff --git a/src/distutils2/compiler/ccompiler.py b/src/distutils2/compiler/ccompiler.py
--- a/src/distutils2/compiler/ccompiler.py
+++ b/src/distutils2/compiler/ccompiler.py
@@ -940,8 +940,10 @@
         if os.path.isdir(name) or name == '':
             return
         if self.dry_run:
+            head = ''
             for part in name.split(os.sep):
-                self.log(part)
+                log.info("created directory %s%s", head, part)
+                head += part + os.sep
             return
         os.makedirs(name, mode)
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Thu Jul 15 01:38:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Thu, 15 Jul 2010 01:38:06 +0200
Subject: [Python-checkins] distutils2: Nuked support for ext_modules as list
	of tuples which was
Message-ID: <E1OZBWg-0001H4-9T@dinsdale.python.org>

tarek.ziade pushed b43b6ae02d76 to distutils2:

http://hg.python.org/distutils2/rev/b43b6ae02d76
changeset:   365:b43b6ae02d76
user:        Jeremy Kloth <jeremy.kloth at gmail.com>
date:        Thu Jul 08 11:50:14 2010 -0600
summary:     Nuked support for ext_modules as list of tuples which was
files:       src/distutils2/command/build_ext.py, src/distutils2/tests/test_build_ext.py

diff --git a/src/distutils2/command/build_ext.py b/src/distutils2/command/build_ext.py
--- a/src/distutils2/command/build_ext.py
+++ b/src/distutils2/command/build_ext.py
@@ -188,7 +188,24 @@
         if self.package is None:
             self.package = self.distribution.ext_package
 
+        # Ensure that the list of extensions is valid, i.e. it is a list of
+        # Extension objects.
         self.extensions = self.distribution.ext_modules
+        if self.extensions:
+            if not isinstance(self.extensions, (list, tuple)):
+                type_name = (self.extensions is None and 'None'
+                            or type(self.extensions).__name__)
+                raise DistutilsSetupError(
+                    "'ext_modules' must be a sequence of Extension instances,"
+                    " not %s" % (type_name,))
+            for i, ext in enumerate(self.extensions):
+                if isinstance(ext, Extension):
+                    continue                # OK! (assume type-checking done
+                                            # by Extension constructor)
+                type_name = (ext is None and 'None' or type(ext).__name__)
+                raise DistutilsSetupError(
+                    "'ext_modules' item %d must be an Extension instance,"
+                    " not %s" % (i, type_name))
 
         # Make sure Python's include directories (for Python.h, pyconfig.h,
         # etc.) are in the include search path.
@@ -396,86 +413,7 @@
         # Now actually compile and link everything.
         self.build_extensions()
 
-    def check_extensions_list(self, extensions):
-        """Ensure that the list of extensions (presumably provided as a
-        command option 'extensions') is valid, i.e. it is a list of
-        Extension objects.  We also support the old-style list of 2-tuples,
-        where the tuples are (ext_name, build_info), which are converted to
-        Extension instances here.
-
-        Raise DistutilsSetupError if the structure is invalid anywhere;
-        just returns otherwise.
-        """
-        if not isinstance(extensions, list):
-            raise DistutilsSetupError, \
-                  "'ext_modules' option must be a list of Extension instances"
-
-        for i, ext in enumerate(extensions):
-            if isinstance(ext, Extension):
-                continue                # OK! (assume type-checking done
-                                        # by Extension constructor)
-
-            if not isinstance(ext, tuple) or len(ext) != 2:
-                raise DistutilsSetupError, \
-                      ("each element of 'ext_modules' option must be an "
-                       "Extension instance or 2-tuple")
-
-            ext_name, build_info = ext
-
-            log.warn(("old-style (ext_name, build_info) tuple found in "
-                      "ext_modules for extension '%s'"
-                      "-- please convert to Extension instance" % ext_name))
-
-            if not (isinstance(ext_name, str) and
-                    extension_name_re.match(ext_name)):
-                raise DistutilsSetupError, \
-                      ("first element of each tuple in 'ext_modules' "
-                       "must be the extension name (a string)")
-
-            if not isinstance(build_info, dict):
-                raise DistutilsSetupError, \
-                      ("second element of each tuple in 'ext_modules' "
-                       "must be a dictionary (build info)")
-
-            # OK, the (ext_name, build_info) dict is type-safe: convert it
-            # to an Extension instance.
-            ext = Extension(ext_name, build_info['sources'])
-
-            # Easy stuff: one-to-one mapping from dict elements to
-            # instance attributes.
-            for key in ('include_dirs', 'library_dirs', 'libraries',
-                        'extra_objects', 'extra_compile_args',
-                        'extra_link_args'):
-                val = build_info.get(key)
-                if val is not None:
-                    setattr(ext, key, val)
-
-            # Medium-easy stuff: same syntax/semantics, different names.
-            ext.runtime_library_dirs = build_info.get('rpath')
-            if 'def_file' in build_info:
-                log.warn("'def_file' element of build info dict "
-                         "no longer supported")
-
-            # Non-trivial stuff: 'macros' split into 'define_macros'
-            # and 'undef_macros'.
-            macros = build_info.get('macros')
-            if macros:
-                ext.define_macros = []
-                ext.undef_macros = []
-                for macro in macros:
-                    if not (isinstance(macro, tuple) and len(macro) in (1, 2)):
-                        raise DistutilsSetupError, \
-                              ("'macros' element of build info dict "
-                               "must be 1- or 2-tuple")
-                    if len(macro) == 1:
-                        ext.undef_macros.append(macro[0])
-                    elif len(macro) == 2:
-                        ext.define_macros.append(macro)
-
-            extensions[i] = ext
-
     def get_source_files(self):
-        self.check_extensions_list(self.extensions)
         filenames = []
 
         # Wouldn't it be neat if we knew the names of header files too...
@@ -485,11 +423,6 @@
         return filenames
 
     def get_outputs(self):
-        # Sanity check the 'extensions' list -- can't assume this is being
-        # done in the same run as a 'build_extensions()' call (in fact, we
-        # can probably assume that it *isn't*!).
-        self.check_extensions_list(self.extensions)
-
         # And build the list of output (built) filenames.  Note that this
         # ignores the 'inplace' flag, and assumes everything goes in the
         # "build" tree.
@@ -499,9 +432,6 @@
         return outputs
 
     def build_extensions(self):
-        # First, sanity-check the 'extensions' list
-        self.check_extensions_list(self.extensions)
-
         for ext in self.extensions:
             try:
                 self.build_extension(ext)
diff --git a/src/distutils2/tests/test_build_ext.py b/src/distutils2/tests/test_build_ext.py
--- a/src/distutils2/tests/test_build_ext.py
+++ b/src/distutils2/tests/test_build_ext.py
@@ -237,53 +237,6 @@
         cmd.finalize_options()
         self.assertEqual(cmd.swig_opts, ['1', '2'])
 
-    def test_check_extensions_list(self):
-        dist = Distribution()
-        cmd = build_ext(dist)
-        cmd.finalize_options()
-
-        #'extensions' option must be a list of Extension instances
-        self.assertRaises(DistutilsSetupError, cmd.check_extensions_list, 'foo')
-
-        # each element of 'ext_modules' option must be an
-        # Extension instance or 2-tuple
-        exts = [('bar', 'foo', 'bar'), 'foo']
-        self.assertRaises(DistutilsSetupError, cmd.check_extensions_list, exts)
-
-        # first element of each tuple in 'ext_modules'
-        # must be the extension name (a string) and match
-        # a python dotted-separated name
-        exts = [('foo-bar', '')]
-        self.assertRaises(DistutilsSetupError, cmd.check_extensions_list, exts)
-
-        # second element of each tuple in 'ext_modules'
-        # must be a ary (build info)
-        exts = [('foo.bar', '')]
-        self.assertRaises(DistutilsSetupError, cmd.check_extensions_list, exts)
-
-        # ok this one should pass
-        exts = [('foo.bar', {'sources': [''], 'libraries': 'foo',
-                             'some': 'bar'})]
-        cmd.check_extensions_list(exts)
-        ext = exts[0]
-        self.assertTrue(isinstance(ext, Extension))
-
-        # check_extensions_list adds in ext the values passed
-        # when they are in ('include_dirs', 'library_dirs', 'libraries'
-        # 'extra_objects', 'extra_compile_args', 'extra_link_args')
-        self.assertEqual(ext.libraries, 'foo')
-        self.assertTrue(not hasattr(ext, 'some'))
-
-        # 'macros' element of build info dict must be 1- or 2-tuple
-        exts = [('foo.bar', {'sources': [''], 'libraries': 'foo',
-                'some': 'bar', 'macros': [('1', '2', '3'), 'foo']})]
-        self.assertRaises(DistutilsSetupError, cmd.check_extensions_list, exts)
-
-        exts[0][1]['macros'] = [('1', '2'), ('3',)]
-        cmd.check_extensions_list(exts)
-        self.assertEqual(exts[0].undef_macros, ['3'])
-        self.assertEqual(exts[0].define_macros, [('1', '2')])
-
     def test_get_source_files(self):
         modules = [Extension('foo', ['xxx'], optional=False)]
         dist = Distribution({'name': 'xx', 'ext_modules': modules})

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Thu Jul 15 01:38:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Thu, 15 Jul 2010 01:38:06 +0200
Subject: [Python-checkins] distutils2: Removed bogus uses of sub-commands
Message-ID: <E1OZBWg-0001H8-CG@dinsdale.python.org>

tarek.ziade pushed 5b9c3c536275 to distutils2:

http://hg.python.org/distutils2/rev/5b9c3c536275
changeset:   366:5b9c3c536275
user:        Jeremy Kloth <jeremy.kloth at gmail.com>
date:        Thu Jul 08 11:55:56 2010 -0600
summary:     Removed bogus uses of sub-commands
files:       src/distutils2/command/register.py, src/distutils2/command/sdist.py

diff --git a/src/distutils2/command/register.py b/src/distutils2/command/register.py
--- a/src/distutils2/command/register.py
+++ b/src/distutils2/command/register.py
@@ -28,8 +28,6 @@
     boolean_options = PyPIRCCommand.boolean_options + [
         'verify', 'list-classifiers', 'strict']
 
-    sub_commands = [('check', lambda self: True)]
-
     def initialize_options(self):
         PyPIRCCommand.initialize_options(self)
         self.list_classifiers = 0
@@ -46,9 +44,8 @@
         self.finalize_options()
         self._set_config()
 
-        # Run sub commands
-        for cmd_name in self.get_sub_commands():
-            self.run_command(cmd_name)
+        # Check the package metadata
+        self.run_command('check')
 
         if self.dry_run:
             self.verify_metadata()
diff --git a/src/distutils2/command/sdist.py b/src/distutils2/command/sdist.py
--- a/src/distutils2/command/sdist.py
+++ b/src/distutils2/command/sdist.py
@@ -45,12 +45,6 @@
 
     description = "create a source distribution (tarball, zip file, etc.)"
 
-    def checking_metadata(self):
-        """Callable used for the check sub-command.
-
-        Placed here so user_options can view it"""
-        return self.metadata_check
-
     user_options = [
         ('template=', 't',
          "name of manifest template file [default: MANIFEST.in]"),
@@ -100,8 +94,6 @@
     default_format = {'posix': 'gztar',
                       'nt': 'zip' }
 
-    sub_commands = [('check', checking_metadata)]
-
     def initialize_options(self):
         # 'template' and 'manifest' are, respectively, the names of
         # the manifest template and manifest file.
@@ -162,9 +154,9 @@
         # manifest
         self.filelist.clear()
 
-        # Run sub commands
-        for cmd_name in self.get_sub_commands():
-            self.run_command(cmd_name)
+        # Check the package metadata
+        if self.metadata_check:
+            self.run_command('check')
 
         # Do whatever it takes to get the list of files to process
         # (process the manifest template, read an existing manifest,

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Thu Jul 15 01:38:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Thu, 15 Jul 2010 01:38:06 +0200
Subject: [Python-checkins] distutils2: Merged with master repository
Message-ID: <E1OZBWg-0001Gx-5U@dinsdale.python.org>

tarek.ziade pushed a415e9205b83 to distutils2:

http://hg.python.org/distutils2/rev/a415e9205b83
changeset:   363:a415e9205b83
parent:      362:a453819940a4
parent:      350:8b73f08d1d8d
user:        Jeremy Kloth <jeremy.kloth at gmail.com>
date:        Thu Jul 08 10:05:39 2010 -0600
summary:     Merged with master repository
files:       src/distutils2/command/build_py.py, src/distutils2/command/cmd.py

diff --git a/.hgignore b/.hgignore
--- a/.hgignore
+++ b/.hgignore
@@ -1,5 +1,8 @@
-.*\.pyc$
-.*\.pyo$
-^src/build
-^docs/build
-.*\.swp$
+syntax: glob
+*.py[co]
+__pycache__/
+configure.cache
+build/
+MANIFEST
+dist/
+*.swp
diff --git a/docs/design/pep-0376.txt b/docs/design/pep-0376.txt
--- a/docs/design/pep-0376.txt
+++ b/docs/design/pep-0376.txt
@@ -402,7 +402,7 @@
   ``.egg-info`` directory that contains a PKG-INFO that matches `name` 
   for the `name` metadata.
 
-  Notice that there should be at most one result. The first result founded
+  Notice that there should be at most one result. The first result found
   is returned. If the directory is not found, returns None.
 
 - ``get_file_users(path)`` -> iterator of ``Distribution`` instances.
@@ -633,7 +633,7 @@
 
 Distributions installed using existing, pre-standardization formats do not have
 the necessary metadata available for the new API, and thus will be
-ignored. Third-party tools may of course to continue to support previous
+ignored. Third-party tools may of course continue to support previous
 formats in addition to the new format, in order to ease the transition.
 
 
diff --git a/docs/design/wiki.rst b/docs/design/wiki.rst
--- a/docs/design/wiki.rst
+++ b/docs/design/wiki.rst
@@ -282,7 +282,7 @@
   mailman/etc/*               = {config}                # 8
   mailman/foo/**/bar/*.cfg    = {config}/baz            # 9
   mailman/foo/**/*.cfg        = {config}/hmm            # 9, 10
-  some-new-semantic.txt       = {funky-crazy-category}  # 11
+  some-new-semantic.sns       = {funky-crazy-category}  # 11
 
 The glob definitions are relative paths that match files from the top
 of the source tree (the location of ``setup.cfg``). Forward slashes
@@ -399,7 +399,7 @@
   the folder hierarchy from the module until we find a setup.cfg? A setup.cfg is
   necessary if you use distutils2, is it not?
 
-  -> information founded in setup.cfg will be put in the *FILES* file upon
+  -> information found in setup.cfg will be put in the *FILES* file upon
   installation in the egg-info directory. 
   IOW in the unbuit-egg case, we would need to create that dir, then use 
   pkgutil APIs.
diff --git a/docs/source/_static/depgraph_big.png b/docs/source/_static/depgraph_big.png
new file mode 100644
index 0000000000000000000000000000000000000000..8b8a27ccc4f695a149fc67f721f40e2798f8a8eb
GIT binary patch
literal 181858
zc$|EEcRbep_x~lLvPX7=hE+yF*(#M}W$#f|X7)~2l9fb}Eqi5V&y=#0dD*fuBQk&I
z<?g;e- at ks>t?oy4d0+4M>vhg^oO51+6y<LcpP)a1!C;7Q- at 2}h!H}9`FgRSt at ZoRd
z-)i!}|L~3O+`NuCK>sCHr$=Hir!lv$OR2cT{lvO at +<P!0zR!h?6|}18?pAhUPdi4G
z&8nzE8?Gv&C`cpumZu!EeeA at n=sf#HBKD<cT(O~SpGc0pQcb*3`pVOc5SysoEubA)
zTr07Kow;(mr<--reB7;8ZzgJLYAQ~|Ja%fhI)5i6D<>xhkNp4p>U|d2O?;a5|M#+x
zrR_fV|9wXmZa{I+)BoRVzV6N({_u&`*4D(ng`~a(L%~=jEv*xQjX0Qsp+#O=qVCd~
zni|FI-tNW3#M4__Tc?GEX|-~x;fpgWt3Cnu at Me{+6>4zv#<q1XY_Rk2M7O_QcN62}
z<cur5OiKipkzc5)s!HrKi~H{C;({)Kfj^8bECi!V?(8&nbmTK75dZJlvDQDGL>#W3
zl9iQJgqum}db+mu2=lnt!vFLA7o|~QM^79(_PDlILRwlnvDYHa%j&-ssmf8~Lf at sV
zd?fNS-dp*=)>c_eAO8cvSYpiB;$lc$oh0T!=$%ro8rzvOcq}X|D;pcCKd+FHEA1KH
zs-$<SJf*3rS?;pR`pDisA~Mo<ch}>-sp({E6j%A9=@W2eSfw<{-)$6BRDNS)27dnj
zEa%SQD=8_-N9SslNLM;7#x;ywKYS5?HYz+VBt-M!!v}V5?(4U2AAc4c{Pxw07Zg+{
zPfDft)SC%3Sd>Vo_i%LU<*E_t>+9p;;YnuqvT}11B_$;tJAPcs%q*jaf{7_~cGgV5
zW+V>B=D&?i9+Fd1B5!{y68QW%5!@l$`SS$P7j>nrtuL<q`lY0-JfTuB-m$W>B6;i9
zv4)0*&K%84#g+pUe0+RQgMy at VbQlMV&3#EuUA4vV{9SY<F%C}7hPj^fh}W-!e*UyX
z&)uDNr`+QY|A`YPQVo8$ojx)Y5fSk;B;*Eszmc&qIRiuR_grl at ad9R?LqnlyjqBIH
z*{Vbt&4&)X{CvORbL4rupZ$DsSM2d;Iucs^32DM3B3RmNv(>V at s$A{O!@V3F1X$VF
z$m!{y!CIy_kB&0?`1q{+{*6DT(?*Q>YI2F2I at qByH5JdPQa7vV!jn<AYNi`EZhUHM
zyP>BSPesAQ$@zkbfs)eq_u|l<FQ=S`ee!fGSvfhgeDuT)m#N~UCO7r{`}a??vEkD2
z=v|kW4|T9+eKK(N>{<D+<3vO`g6oVF6h}(5h++B71#gvTh5U_jcJ^C?fx^)gO_Mr9
zrh>6=zehz!H<}3?!N}gZV_2eP$V+R$|2kEZ`^E+Z`DtewP982U`#~GDVIK_S<cQt~
z*@OyO4JJA+d^f{ZT=pHl{@?3vIIK-q%1yhePvoiT>661B>~AS3+?0~S!PK?15Oj5Q
znRO)})hae)IjPZUQKG!TrbIk6I;ua<^-5h1fWzTUmCCw)Fy}BlMCY6XJ13{Ko*vUh
z{hH^nX&huEnE3ek?$cQLw^x(yL|>fk&8E5PFpD824a4{9{Wt61oC|J#V%!qSa`x=e
zxw$!E_uo19^uzP>^QmI2;lt71($R5j;0B#NeHtSrCAG4?jyERyQR%r7UfxMkNNQgC
z5M_yQjXb^TXDC!2JYa!H9~c-Y(8|R;e){y7!_qL_>gwt(Il1d$Ok$nY?#}WPH?Mzt
zidAE=%I<yE7)Ux<=YzX7875&n(R4gc!kdXC5Er6;wlfJAZuc>4Xo_y7li6sE*kGv@
zVXj)XL(f_2;G-xzZO4y5;Gg$z?MRRa`}$Q5BdMZ7RpGc0V2!OvvKg(J?aOCwXl#tK
z5<ZOU%bX at fbf)cb5%}?+uMN1mx^_E36f5_JW at NCZ-yzo{AtenA3k!SF@#6;v56?|Q
z!z2NvSG&r7-|QbfvfY?ho#ShZ6}q|bJvVL4DUjq;czF2Ztt}UTEv_xg|KaE at 1R73@
zgZ{n*q^z*2>GE%%V6AV$on3WadLDX8B-wtZ9X;K|<Rm#IC2o{#nudF>8rMa=s;8Gs
z+sD4Xz9j58Pt_hT7JM&V9<YIxo&A`Y+qz^&yjWMQw>J@^u;Lr4sK1{!>l7Uw9e7H_
zX$)3T*kg-VK3?>AY+JymPq)sTJI5+6u6N>EOG`_N>#rZrN`-`k)aMCg6%<-tUos7U
z^(r7+J-=aSh{1W-j=*hmDWr0#f^ZY-%uMJ(L4KLF%;DD1eSSmu*#1gqp5IH at VA!li
zi;`prW#y-u+;(<$vO=2*J#!AiUc18bal$?jN8|m4#=@>^X8^Y0`Oy8YFOM?&;uA~D
z$vy9GJ9`LO_y%P|5%KY%k>?dy#l%j*3N>WjOSfH}xS^t=QU}RzHB{<&g09TG2OnW6
zP%s%8S>pOVvzO0h*kGMzsjO+DX)fvj4 at 70fq6j;F{P?X~x9U*h`}yIyxVm!eQT=aA
ztj+!X{kJyy^b750<bMv95Yo}nAyi!+sqV^F=UM+bK#3OE(lV#}ZZ2hDqYo at R%jL at _
zIXF0O-M!lkDT;#;bN}5`y*U!>zFbY?JnAV_Rb743`oh16*$8Q{wOBS<VBA9Tv%e4*
zQt22G(e- at 2>eT(jk!tsp8^P4FLM8GSbu0W{zNBIl_n<`=zkORuQ<IK at lG1L}b8~#e
zeYsg|a~QWoE7##(&{J#eGM!sH6V1W>Rj!u+Lj0zt?!SzVPJvve<u$-8)$CZKwFQ`v
z6)LKy4Uu8I>Tv5s7PF`e*}2H$*b2wS0>egLqo!l)OIQYg0$HJdf at c2=JRyp-PZ4Zz
z2lWwbH*Eo(W7<q0lo5o(Vk(=JLzE)G>L9S at hxY@yq^6~fug~>#PXjX;7HT8{c$Yj>
zU!U*2ll6##nVA$`X|eMEq1(6A2d1XRi-RQ^W!9&hmPZh<t}pZ{&=(9=yKid0=PAv}
zA*ys<{s<|I_9af#l>)Q`o6Pe%h at sWK+WnLJ`}?v&k^e^eJI)Iinm>jxtp4hdF#huD
z0_cmI3F4l%Xz6NpmadJ~c-nk;!U+UVK+5#8VK$JIIUp>IG=P{PpmuN3hL+C=Pk=t8
zyX_2XMeZ;K%i^z at 4&&+8sIAo&X1ncmbGT&#NK0CA4`F;#rstiVifOU^X8ns!KSyhs
zAp{#<UC>53Q&7PBoJov6^4wi%W##S*e>&I?V_G|+*P?m6F)$)7?m27}+TjNe9-ts7
zv&P0r*TF&<(`p+6xIy4bT3BQiGm5!U$)u12(ajE(@!*p%K21$MgZ4^XTwGS@^?%Wu
z(Z$8Z)i~dqz1k);I<`6Lebda0T_%LypU?QC!KZMRm0!Q^RRj at V1l$E|Y8)(i2=_n)
ztNonr5<Wz7bE1 at QqH^X-Yu3Nf+C`k1kx^G_HMG)9qi^u*N1I}%3N8QrV;@6L1w*V~
z%kJgGHsPx(D?iy-=$o8L at cszzg)8uSY}tygHc`@wy6|2y0$r2%IiQ%3kg(8g!y-rL
z0|j8BbTIV=Uc&~w=ch!dtKBvx0&iV-U}9p_7(lF1?RMGgV9&X!scCFHkolu=m{>ql
z)Ah92n}^ZD4+}#;!dPchyBC11be at FA3>>hO@$eAEC!&1}z;q|8^{b>)&zP;SEo at fm
zyLS at +c>UGxLJ&R#-rnBBt}}0M$;vjs!`jZgbzU=-{Vx*FM80_Oq(kC>@!|IzpT_aJ
za=S_12dz|R&Ys0SI{9M%d2%xA?%Efgt at -Ty0`CK_VyhwQo^*wlZ<+Ui`!XGV{ksts
zXC)+<^}YU_t=V0>qf_B<d~?{I^xfUqkF8z&X=!O1wO(R1``Zt9e-GK14ZJ`5P9Y%_
zQYA&R$V65s?CI>Q-YMz9me5m;fy`b{qqx)?`ub=lKZZ0;eU3ypiC=V(da%D)t5ITc
zQY}v>sK~U#2teKC$7 at qwldrS^IEVK at mE1E|U^OIY*<XO8U1}+1XU7LP8}ag`pARkp
zdjHeVP_>HJ4wTc=(=1o6(4hod!dBY-_)6y&5HPX7v)uBOoQ>CERuPYYfD0 at 4KZkI`
za<t}@P4x!#i^#~;=-S<*J at 4)|rrwV1x6vX5(yXV^_wZS+Jz#>Uj=Ac{spm3&wAi9o
zcJAH3TKz*5$i>YOchWD>JU0~;$(X!%?c@@~PeJ70t1u54{PSyObvjyqBG#tHr{16r
zmUnd`n1}V^MR{}$?4vhwx7;BZU3M1B3Qaqh9Ok- at c8LBWI(PRC9zw((>)Rlz8{R+o
z0<eaU!f3!Wp&4R)Y<W4~q30i<cU8~VYbKT0eO&Ue&j{dsWo->b1n;A%ThD3vj$F_x
zR;&p6_87r%2!jxc%UkmD0a)vn$2hpWPCqrCGfObRqsR(H{4f1Rtgsa)0W4aWy?1?B
zt_6&xMe9N8zvynWc)?9gShGFPRRxcY<ncMVn8!jRF224FLI39Nr_Z13pByDDcr-1G
z?%8{P#+%o1US;dg9}aBsxn`q6gFkE2F2K4Gz~dbE-i-lGncAIZO^(gz3M}3P){WNp
zqTBhs*a&dV=l<K0h+g2<oC#KtpPzpP5^18SL;RYpEid&&ou}qK at 9Kw!wEu1iRvGRN
z0_E{?%?`;>sg<u*?#I#n-$KvnFP{Ky@&iv54+FH-U*jnvXg&P&V0T7>_qAO4!Tt`P
zdG{IgF7SK*cO_=uT?Jth at 4fG82`gg!@p(O0e$^9*E-q}2HfWp{wD8`$lg#LuLG1I|
zjELgl<8xvK{)Sk^UCmrIv{o9$X5`7q$pr at WN95iJ9M{v+lNHLnT{K{+QD}5j&wcSS
z6*aZd{_dZcOQs|q9v&~P>B*HM`^>Hl0&%-+&!rc-{xTT`Cf=HnI0zE=+-Aq-45 at i}
ziH~^vy6f?K!1QTwa3eStJ^-w=*c*pzH1$@*nFQij9CGV;<Y&L6Yk@|sFFA;YyB%M*
znQ-*idhR%Z?J}5X3W^bP7X%bd|6k?<X;Oh$RoC}C8BD_?1O93y-+M1Os%YW1tLqgY
zh5*=gT%fD1-CrGiK)HWE6<%epJ$HVO1IOHi`}@%r8+gV32S^-!*g|FSHPk$M&xJ-k
zD8Y_x0dCm!={j>@xmlCXwOyOTFMlM=sPX>(@d}5zCpkIXNLYwC{WOr&HUyu!HReld
zyEJ?SsTV|=kgAG`iiv+S;Y}(C7K8QKuGw#yl%+PKp?!Mp#30KNNCAhoJQmcA2H*TK
zj5!PvfmK9=A-`(v1bQPV)CnH~raDk9QU6f}B8 at As`P;u2gTo~DX_-CO$(8TjqwCDO
z32f!E|7XT-f7g|WNt8B4CZqw95<C&wwZASNALO8(>-0&0TyWYrNbv3c`32ui0BfdE
zV1Uyecf}X9G-zA0{OaHIj3j~GciC7i-qWj|T?AY4oI!}hdv}%KM at K>!DB1d!mYZ`X
z5e$fx(Ps(JukLM*8uu3%$_gDOUdC-m>yN=aHBUioA(3SN<7<6}Qp(tP*X)nZ<Z~Ix
z$wv at ZzBqfQz-Ck&R1E<T*TmS3|5a4@)$;X1Afb}9%dEEmHoHo#w9{f8hW>2U9)#2$
z>@7ci_N)O`va7=JVOs3JR8y8QF=2#w_hu+#JxI4zc(?A{IjkTj7uPBT!K6Qpo+OBt
z0;feSKR-YA!>vy5O2v>c3O|k)b8ooMKR(-);xZkjPTOhS6hO)xT3jrE^ak6_(54}q
z+7OwSo97R2(BYy^`Qt+47NgEYsbhqMjS$cU0GsgP0XS9_=Axpa2+)3FD`^lfgZalN
zruT#QPc--wB0l;L-((=9KSK2IKK#!8bD$^?$I2hAcS{&^piD}=t~1tOVt?z}+_`@P
z;Q~8E%ihlBO)#6v%F1w`2AokRKu}(nl|6ynCk`(D>GS7dhrn`O31L0H(nu;X7Vo|!
zb+Erz2n2iP%o%p<nXIEBGQQJaq9<YNTHu~m!2>pq+oB~wdP=6*q-Y%ILT0A at t=;dR
znwA6ak0QbYP>-^v$4-D>xv*FBR?M9m68j?{tU=Mh{~`_~`AI@*?mDp4tDny&jDd3U
z1EbV^w<5Rs<42>uJY8gmU%Yr>JKM?ff3`0if^FDy6T7mx*;3=V6La>SqLC5n at 8wZa
z9v&W0ucVMdD8 at 2wE2(no?7IMgxUAF at tTqtxjD-k|JeEyUet*Tm7vydTD5*&#bj~Mu
zG$gPc-qU07sB$s9IOgW&2m#~W<~c|>*=~TwM4%NTWWxkjU=>#DG$KWl|CPaTFyeov
zC`EsLy|mTgz3&T`=E5?*uoGr<nFu;xU^gklC~S*4<k{Q*s_&-Pa_5jHI(Fhj1HevF
zm)T2e3LJKj(49%wK)$26&*-&4&;IQUyVh-YSXh^>@jyf|yT1|+ayAX}Wz&pY%HvT|
zqGSO{7UoY#!--`)%#dcQ!Q$f})dHK1q+sJs+T#Wiajfk2?+vAihw3sz#v^7#kN^*`
z^82?8kW_{9a;Ry%Ye1A}*7)?aKe$ljkU>$*j#P^X+Kl8l6#Wl!3*0xYML}X|lv-+5
z;}`4hG*?$ubym6Bqv(YE`oOVMbJEq#Z36tjO at L%xw{^3ZQBfS&GymrB#5=jyM*!;+
zy!SO%A3GV(_L&n|VIC(Yo<@3)mzKz7z3XO_H3^VQlEzu9l&+_Fx|K#NWA$58kqY!v
zF%!6V#wW(hijh<R at a->oNP}EgTI at Bn#Hmj~5RiipXec)CDR7um22Q{1!Z1H*=zOUo
za(-bryP$v!rER at 0!NcLu8IQ|F8r(?Etzex;4V#~xq!qG`>%RMcDtRw^EpDI(0l8Q8
zt6gNnE$7X4TXl6cWGx91E$^7Z;*Q!<!uN%Z@`b!gUW;$+RU6;6WrYI&t>t8ZGarw*
z$y5(m(3yGfd|GS|`?e$fHHInqc+nuQy=}vd`Cc$M<=t&8tQPB3*{NdQ`yO!HhI8F%
zvX_14>yb8u0FZlo^(1xhjkzCWhjo*3*k)rOOO0|{7V*9HdpC{@)|Iqh?MoBfysUQe
z<VjiyFHsUE(Wek+lRXL&H%v^<wSM`cq^1 at C;N`Kq+5!eAzx%G$`tI&7@*HFEs|jRp
zA;jSSk`K}^^X#2h7>u6x-UhzNyIGpDp+!_n2q2<665AF4u_vg(@Fsa|%mAf}Muh`M
zd%Re-gZ;licKp2tH&P6Go*TU~<ek>s^h3h!Aoa$2)%BYJ*$hjx4mUd`{npvV^>qh_
zh3~{5tgZnb)Bh|Fu`-F!j)GuD><(Bo3E-nvapj*1u1tO(8y_Ee%FVEkpx!ea?6b@=
z?31^J+d(Jr0n7?K{7wWRDJ3WO_3qH-qUFcubihWq-QC?}jEp(|-6CixL1h&cKeb$~
z at yg|zMqE-+#pRalyyflIC at vtBCtA4yU1n985ZHGCC&0$n?yd!Q-(~q5A?JQT=rjYO
zJOOvmfTZ3JOnLRT<9vfj5$rjM%hHD#1mh?|zh!VJ-`+v*4`;bXFOYb_l$RDMPtg6>
zQo6=~;w4Q?-t`C~Y59`l9vK4zp^c4=pTEB$cuEk3?ooT0Dx4WAoB&T%Ypq;QgMxg(
zPJ;yW%dg$9X#E}XV05Foql46bro9=6RM_~$A=j%>b89>{X?t at RQ$lI-O|CbsmMIrn
zTU%Fv3{M#v8iFX#TI(N-bB~ISo?I*&l>vvGsFt0imfe-BEmXC$R9P@!+3m!oo`;92
zf=m at +?9>BY2xy(yYcamC5bU%#I1VA?r<d<MxFpj)FkDUTRCx+1vyON%Ksr?4N3mE>
zz}y6K#(MV3uqCs{3Jxn9+bWnmPAt`bNmtRf3wsSVy$R&?Nvptz%v*o~V0Ot)ogyw8
z(0Skvy05=q1J>c^@Q0$oz3-(5+QmtwlfEyP+~dIq0ct(ZR?q*wQ~2)=e_uoS4e3Hm
z-B~)c=yI^P*;Qn!07_UoSB;H><2Zw$rER5^)g$eaxnHip9!biX;W06={U_!MA)9W=
z$@$5qHC!VVJCmDlO=N1-BKn;GwgxF>ea{WmYUbdl8JQ1PWCzu<Pr{0C!2|HQu1y7q
z7O(qU22{w=uN|4IftzNO at M26#N|GO{7-~U*>b5 at 1x$~C)!HF6W`J7>YK#=fR50`Te
zaaAvGr}iy$<!D~oUGGt#<ukhO%XnCLxdj5()vV_ovDcqz8X`LW>vKQKRk6>*F1O*$
zE!E7o=QatJh;6>t%0&f)-#-T?ORIhHi5p6-hR`N}GM)e(F6_L-I9O(*zjJ=?pzPt=
z)MxOxQA0Z-&P&h1^q*#D|CG6xXK~fkep5>>RxtM?N$W at cUq9d)sTj}u_frvC{fCj)
zK|}!mh2h6;eJiOQ8n+#T%yt>~rzv>wg&IH-1CAFTgQyyn6Tl+`&Lnt`m`i+lb)gZg
zQm|gN+xW_OqxGh4^uEoX$%mK3Z_U-rz}gp^bsgcUUZ;rn+MaV0qyAUAUWFX)OuIva
znir=;oN_iN2P_4_OC}E{N9RqRdf3BrZ%8}=R3E6C#PV{D6Yp^Qr02uLx7i`uP?2+G
z#Wr2=sbH_;S7XoH3JNztPoPVFu%AJnqIB>3SX6Nslh_I0dhnq50_#a<Q61i4mMlk&
z3xO)?tAOCLsog6rZUgI!o^#gWDrDu<^z?X2n7H(V2WR*Hba+=_fx_Ka7xHTxj0#Wl
z at F;AP?Sn^%D}89WlSf)#Utc~MDw680_s?_9!*v6J35T6WJt0l&8AY6kU_<<REgVH=
z7u;s_?nGZ~21QCnLGh$FTfNbMx6Xjqpf^jk)1m}bI#cUilUZ--=;%aEKGi%k9Xz}?
z{iV>Nm$TTS_XMZ~U-|TFV2!{dbmi+ypza1pu$QH^5x+Q{^PA^;vvDw}*0A>L*SM{)
zB0wTBgP=gx>4UL)Uve_C$B41#-jV%hKCHkw90f;Xxib08>po~PRP+To=u=iE1a#qO
z9t-Sn1T*3~bBZ=Z#=N at MqB;3SFa`q~s$9~Nap#~+$Lb>i2X^!4__$B)?ljM>+qW|}
zjsLxTE*o4P2M33kq7y%FBOO&j=tJ=JJywsNAS49Zup6tx0b%{F^>@i!bf#5;LXk<^
z3YY?13~-HVzoj62f$u%QU_8K`DoYN%tq66^;Q?hDpp|PPK}O!?DByB- at g$d;!C<j@
zfmSZDs|MyksOS^Kc-%fmq2q!Y5rZJn+pCVK5RSpXMl|g#*M@<60VmvGU3O<^=g*&d
zOF{L4W!RI>T5oT2PkDI~BV%KLJ4G!m&RiSroje=L^YVTJmLDP48GgnbGMDUYK`Z+A
zMMZ^HLxO3s%DJ(+w^5OWR?t$TB<#b;2X&Cbb`wp<b2N+aNyXN~d&Gg8{BK8|L%v7Y
zZJm90Zx7U>fne<T%*@Phmq%;Ae=iOpOVNNtMy0b&`eM at uRUi>N at S+$Yr}iI9B?vG0
z`T0?msi~y}5htYi^-8D3tYUTZYXL-3kc_JU;Ozegh%L^5>FbH1N=i*_nk%h0c&wKn
z*f{?9^Zh4{;}{GT4b7dsOu)P8>1lQzp1{xdrS0ta#CLvPI&=QK at mQVD#=(8X5}tvc
z9!l6x)b|1WS3k=;lC92@`tF at kSMrS*5$B7K53-Z7AdDa$@9a1%-pPl+*j+85WBx|F
z9dFtm7X+%{%y|#?wt|0W0nQ8c^Zi at PG2a<I*waTVm9CIr9X<(aix^x|%d?YQyyo4h
z7G>FYYUOlyJY=5J`rM5bynVw6u*6(0PM931%wY(>t4yVjFKh#LSUyJBE?FRxg`8Zq
zcZ_8b)h2mrwpbmm>Rw}_t~>BF1M$AKvvX(R?m5aMM~<Y{|CZQY!96DzGbWgLx}1<5
z!p+6i^@G^E{u;i;${MlNCQ4L(Dz(NA8nhhBU*%Td55Lw{SwwG;-&#T<sa68?z_Hn$
zUpLyTudh_(<Pbr~%kQvrarw%m_|(=)$ep9qDzWg-$>9dHd-w6R0JcQv^l1#pkofp`
z`At<dHQ7Fpqe^;uUX3%BOM1amQ}^ZKMd_|wxgx&+A at K}AiaH`9qL(QpNo;G(cV&IO
ziBw`Y(c#{I`e)w1-oW-vIC=SZ(>bmCbrvPyc(3FZ9S#`CFMvzNhI)1_Y-D6)*lMY(
zso}@A{U~m?6l^pq6h5WSIAl_Edk4)G;QL%<B;mm5Lgee_=j8Z7qzqWf+_>Ri%!EO$
zV(p(j)@35q)KO7Us!V-~p-f`Wtg#iBKNVWc6FSdC>wDG5 at EMa)QwM+}gOyCO8Y=zN
z*48XE>fw8n*|T`ivoGyHdw*;2IjIGaX9seU at 9xH?;<Ws$RQoOg7BPytkkuaSyZbUK
z<f=8oO&ai?=H)FIT%7Obxz{J25EsV?Fr`U<GZP5@)vJKc&QAKUd(jXAi|r!I*GFqS
z3-5n^;U<AtVA^r-u615P0q at IKuOQ#m|EySWqe7U9n%V at abAEk&iJXRpKte(S_I_(?
zOJ%2gP&DCV`SLc>li}gv_!1WpltoY^r=&!@d6VUE$(pkdCpd5TXz{|Ax6UI$u&b8+
z1xX*nOGT3Jy?=)jVDxw at Bwqc4`vnkI;jdo at fu)}WW<(t?fTaS!B+)-pks4*zx^Lre
zdbYN;-GDoevc~^!E`_acrJ&2qfxc;mUfxOlhGu%tm7_YkjV<6Dvf_x!Nrfu28 at QK)
z9f)~`ECs=t?PYojoB`}kmx67HT3J~MU!w6bi;_L7dr+jpdD6qfBXx57&!5|yf8e(=
zPfySB at o^kdQc{!&#!6LHRbvYagv`v$E^cnJL+&mvrv(I525k%sPIGc{j*XAA5Fc^e
zJC&7@@wm;z7oq{^hmnNfxlGD>C2G<O#)2Wi2ag9PL}G7u;Y)^6ifzj&=dq?BN~F;M
zNW)`egUic>A!7^mLN#Mor9zUHube+mVA>Jy9~ejg_XuC2Vq$P#EPG6UwzlfNd?B<b
zp at -}M1?dOs_?yLLl&@dv>LfvDH-b8bbU^0U!9f6nMhV<<fUy8m0Zj??^+)mdn2-Kl
z&OKK?KE6+1zTB&z>&*&#u5D(96a;FtfiTNb&8Ekll-B(y`n_y;Sf}aZN5M}ZINxO9
zE8KRT>plwWdF#%dhXt=~br-RfG=Q-Y*2 at D$1eneW$A_4 at Kfh*xZ$UxyNaVeUj67H`
zjGeEsY5E*_o(~+sjnKP?C4)`~gq6AUga!j%$&@a3Y~0o3L`2s$HR&EbdPGJ^$(Ac&
zJ64x4xD4=tI`cp_sCk53Y>%;!_3$x_^68zfo6OB`E<Y?V>taPcAgjUROh*ya`2O}y
z1$Fp>f`TqvyBBq2!*=r8j!B~80wz2n!bi|*aI8<?yE<vFmjI*?y!~seki!2oKyUUt
z3TY?xBu%}`bQxMW6;)M?GAT+Gx=Kno)q2a at 6vGazyc{m>?jpqeDw!&1Tp||G<Spk{
zlrIoZJ`VyLaiV08f{+I31Zf49iG`O}(X9qp1x$&pFr#v_Is7Xq{W;S~5U=1fHMpta
z&Ha|840)U>YT4t9i^LF*u(P=3)K0%QHv=IminVg7w5U6D at nomNB5Kt@XZ4EXm#cBy
zl%wbHjtdJ5Ti*r=-CMykXuz8YnET+t1NI9S{Ppsqt)ru&{-Ha){F~<H9AJ at 951f_=
zAM_pMOYhcIixMsH4YI*%+5R#qQZ6pSM{se;|Cu<N+aD-0W#i(?mX`<%NL9{6#dOrZ
z02}nmi-f0Y<;i+cmb~tEt&Lx|X9pb}-RwZoc~Bp?!Of^9^8P*loac9=!aDc@^O%uw
z46Fwz6XnZWv8pAyehtk{xSUZZKn5`}F+p^{Y#3%GyZ8zMJH1vx$xj;IFq*ivL+R~~
zqW{E+M{YK3e>M7GP;l at yd3kx?lY2k=3)#Sd1#B$zjTRr at oFAOI<h-PVraVTyb}l@!
zw_okP!(&+g<fzNcTW1SVc@*I1!RteeqvlefAQ}1Pqr^<zOPxu$Xih0tyNr{%JN}CO
zIn*Ls!d6a(N$l4FIO1T~ZieDvAdJzZ6_S?m?n_qyIgDFEPr1xzrtcrd{KEY32YWO(
z5<txGeXQkhW=^>QP!CxjZv7fY5O&HHR~)-b9;&0hFRB|8FneHf)q*{Dej9GD&v{$a
z_K9r15w>GR6AuTQ2m8#JF_3};wz)-*!1i|wVvz1Z?URMReAJ*`S^L6c-1L<EX>hPF
z_`0+^(MZCVEdm4KzmtFaY`igWa-&bbB~i*3na;pQA5`lfsb<jk-n&e~C~SMLBO5hA
z!S++!zJ2>?Xs8SX3-V(aTl(4T;^tr)G{WZN<0JfN>KKHhfMRauU17<;ROKD|&`eX$
zeBkmr7wPfJfM>eCU)RQp9x7_-VFxaoag&~_#D9(T56 at TUs1PqFoDT|5051+Mn+yAj
zzzRqB=+S0z?GlS=6;Dm3!J~?4&vK5Dw3?S_Wgg)lV?5#{%m`HdI69i<tH~uWEI*w@
z03Sgyr*VF&D>>~3Hwxfv!2cQ?bz;w4>L|@>5xImmxin2JTl7WUEPGWNa=llzHLU$P
zVB04r*{bo^3f>>nvJjxpe3G9*c8ZK_Y+WmO at 3a1$)uQY>n2MUGed_5wNyWt~T15D`
zvUl#F3AbWBrn*;w?{#qx9!)q>@A(=XCn1rswB#0`4=HZP3Q9>ybuTVd>zNeMU42Ue
zUvAY*@H^_>qMGBU@>he~L!<iv@^7!A0S8Q8NL%LJQCCQv1=+b8)3@@0z67Kbxn(2L
z`Fho4OiZCBZEtRSdzYMT7{2;e-ffhukC2d1NlWV)xWJe;6RMLZ>n%#8^ZOdj1Y~>r
zZ_w)Oi>qm9xL}QgpRFaIC<m<)7}!Frnt0;Gi5DZWRn^sLo2U9V+5hZ>xp-4At4{1`
z>O9++P_7aRy0g0fw~zGg<*<E+m|~_%f1xo!0Ldw><(615 at +i0HE}4LiGFqSQLS+op
zq%0ftVtCk>m-|b6p?jvtxP`=HYdr9loE)mTfp`m55dVXRivD|>qhYD3sTTi;67!k%
zw?N8u5Qp;{!q>0geVe816ETi*2~~3A#N11zM4mTGFy9Pi3<A3*jHO{<2%c;SV-pad
z%CFhx1iezi8RiY2RcO*ik7+;nMz-U)sa!GA7xP;;VSmPZpWyT7&)~ew!zce4$ig4>
z+&l%V=?_==XGA9b*?c?C{@${;@$S~DYN_{B#7$H#RdtoSJHxTIlkdFrphSz$WyRnp
zwz6;03m6M^!_8MR>kaDsj#Czx_q?-6FqV|Wh;PrbiEhuP$S)*#{?_EyF3lNi37Wab
z47wG`nysHj9Tt`f*UBB{D6TrrPYLi2swygmQw@@kkOau2L|6y63QsuQx^)cb9Mz_}
z%*YrSxv?}<R4NZ1^Z#obC7Eu$A_$-MF?Ovhe^+p6OvnF&ctd+$RDv$ojZvaJmiw*=
zj~{NSGBG+4(|_^(oF`?H_=Aie1AJ6JI;vl>c+q8no|DQ)c-$>2*+y{H{=$U|H$b*P
zSf`Cfym&z|=jB_E$?kpD-QCTZWVU+gq=Yj>BPtZVzj(5=xkIw*A}cK|1qJJ#TvNy0
zDnZs5c0G~4#Nyo<Zm)m|)0u5IA=Nw_3-TOloCkldlIMJ%bC6${@@{;@T9DamktSZ2
zn0VSufZj~tjYYO*@X(sY_qETsc{f#hsI8*(qw^zq);5;Fb%1kv#jW|Is!G%^AYf!=
zL07wPVJNR1`JM- at RG2yjn<^iW82Zkmj*AC*U$#!C?Zij(=!br=pGo#U*ar#uws at qa
zg#dIDSO7Gn2e6xZmXlMqB44w}<mvP04%>0%15GpO^68Ny%o}di>rCb#*$OpMj}wEO
zU-x=gsDX+EKw8M&GxT*lbVed-dip5nNnriPFJJD^Gt5j+`@`aJ5G~Ek9S52Nx({xC
z0)GSFl>S*J?`GBc;7#S#*VM3LcOS7_Y<4eQa4#!=8Q5 at ROoxfddC&6v->M)vo6K_n
zP`sM`?FBc9S8XO}ma4#hM!syb@^{I%clz&2J;h~G$aO0mq`-%xE<1b+2J_cEvs}E0
z`e)JSG93B>218wQ6v9N;j>*c(W<AF}%gSo{%&&OEO$>=B)ZFh7+tl$rrS4%ix8e3R
zUJRu!ih`xeWpQrJ_jtg0__kJp6R5?Ercd&8K0Kkmpk=T&^*KhuyJnp<i-A~K=$&U@
z^skwY6!_*X&~T}mhW|`N5P&p9Gfk)l;JiG-=_?bZ*7K(sq*jRYh%3PXq#F6jlm0Et
z-i(Aa7hmF+jNH1l<2IvLS0eufsLH4}OBI#xK==-p*@VHf8Gg^vl<)OH1)H%Uo7(22
zYk_Du(#6Fk5hxOU6iR6L^}(vI{jJC<+|0!0sCP?mwmRzhfX*{OV=e1lH at hmGmt)=9
zomR%AJpN2aqxQ;0{Tej6<S_s3c;tD7k39+!xh@)-`F$x6MyRO|Y|l<CNEvktWPQ(L
z#gwjN)p|U{a+l~s*f}^N@`AHOWl{=1`dD~*dG+GgG&}OSsOWVK>5ggVs)?vR|Eo{l
z<qG1My4GpvZD=xyI_A!ElZt)0*1o1#B-&&NQCBvEBdl)pDzNA}`)?E5+^5|6p#x-w
zrmj(@eOaA(ET*pP74MuzTc=novU^YX){hBZ*Nxr5`WNKi^qan3CnzjjKB!o{^_Hww
zh&A;*=|a#b+iK-GuOE56wKg=Yk{3vof)fiL7HFvyIq9BHz>|`%`jLF3L{#S#BjZo1
zitlCu-5cTIWa*AVgUo8m%F0#+6%~#Xe5_;3t&l;7^Y&_PBzeIe)#Vp2?(OY$rQM-H
zB?j2Ms-ziJ8X@{_afey5?ca-pAu_&qqAy;lR`_1LT(R!m_rN3kg=dFT<tc2Xa|`fD
zAt**FYU<>~L<~4Cc6Rn~YP-sU0y6bHoqE`<w2ntoxoT{pqKsW#UCNI|`1vW?-dv7s
zz_NCpVqzL$ABk*oCUUJU*UJz5eE-q>HKEsX-JYPyPXvA0>2tjLmg+=)sBD_Fq9PeM
zuCbY!^VOs7Zf*^Ied-lCVDGm{42lNEAR^Y*9)adIZ=fj}AOel<TYuWK_Jwyx7I7E8
z9f!D3Q`&CD?xov0&z8?jyszM#$m?wk`desUS;((FoK*lNOioF8Z at Hytz!G^mG!B5q
zsaMw5(F`W9X*=U{#w!$(*zE_82RHtp9_+*KIa0M=d#DXMR|q*W+mlWT%Z-NDa<oda
zJ_>#_`D*M*e$)WfN$Y(H`iATvn~p-};*6yycr~E;HE=+QcVh)*1$^dx8pkD-l_`vj
zjnOpR-wvS)4``HU&z>RZg4~>J at F)D#+A8hjBnXZQS?;YZ7fWnKiBy(1S+A<0A<O=D
zUu|-6F;!+}rjVyJ4xhn at qwSk|16%sXYWK#5E_3Uh;&T~xw9Q^Pt=8A<IQGz(cI=<=
zIThw9LnCt8R|IdTS?-gQMG>S<tG-%b8~drnvE31_e*INo(DXQyYt(($3{578?R(Rz
zp3^=LTqQM`QeUZSz6#9Hz1EX=GBMLjgZmZ3d~`U$iq5s~eQaLL5az4ysX!KzgCWD-
zm3ZSeFSiChWKVEYljpn&%(_vaCH}!Tdk{aPnRPpRYQaY4s1pztC?SwRN#*7Ji>%vu
z!NI|A^$485SvZzR%@6wJPDct4XPgpoLh~tLTkYSB74f|m_q_TFIAbQkn`yavGrqe`
z>05W&ox=N{DdNOEE7~dXk+h<sx|Oup&>*c-?RFBvG@=1JQS;=<6V6B<a7qF;BaBK)
zN at gQfSCUgw(uTlKn|;rrGcq!Iqf=taH}`z2iag=YxyajZzFqSOcrABsPJET_?71DX
z1e>*G{iUbNXN)LROSD2D{eYpb-0eqxPeDNe9Zc!zxu+#6Dw^7n+}bJ&p3wXVMRqUe
zDSe$*-!Z}1yW4r}2kjaRYn)?i^W)y{o^5Sp1ycSqhp^`|=0gAyqVUnQY+9`JWz;OS
z6lD0Gr<;xHV)SCo(t{W at ssRFv8uSrFwqh2YLnK&KBk&Vje`XT?wh{ieN*#2K>fTgN
zuT6hJb5 at Y3N-8R6h{NsF`S8jGSH5?2km^=C-GFFn{c>25Utb(5^GP_Vw1K(_flO}4
z>Pn7(bK96tvkkUrDM>C{sI%zvDalg}ZDQqo|6y%oBbWbaX=##8hzy%p0B+i0vRO;4
zooHXHD?E1bOSagT#?P!P-1&MMLv_Vf%Ih1Uu544|My>`Q`0(PURTFiy#qaf~M~@XR
z;Mt_5rFBm>Tk*L*7Qw;6QI65d?v*XOp=wO7V^6FU%$eBhpF>wuYnPnR;%G-x`byn}
zI=HeRp{l#+;aSr at 1IN3nc{YMzh+xN2I29BW$aW55D<Hb&eo4xwbfF56RzpJrs^oQc
zDyGkzqM%SdYk6U$B}&k1*X6#63G0<Be*(Dt`=Y^9p>c>$U%p&(bGu4GLGjZ1g%QcA
ztA44eXR1~|GN8tM9H&noDn5YFNw>g)YnQ)cZG92=c4FNtzsoGO(+NxkI;X<Gz=fqY
zv|bgPRGYdQk at fk0*1e;)O4HgZnEbg*svbg=%df$XDJOi8NwL$S0kxq^3J&^7_x}Dq
zvP)rMVFKjgTY20Uo1G^?=B00dkJuK!2!AxR;~@194ti#BO>b4?^l8a$60Vx%XUiX1
zSXjC at RSR>stWVcm3brkGStf=n)-Zpmz4K3;VM+M;`O%St2VY(xpDHU9IE+LtY6r(%
zb<FCL&$)H|TMRpW_eBnl`Z^z6)JTOK{PsnnG-JGKz55OtC<Fa`?#$V<9RJLCRJ1{)
zmq{(a@|zdLdT_9_OM(H7sw>IBjM)mGme8e$!h_{NXLy>Ln^6%JO`BbH`uQ{_hSt23
z&-TLZ-fT*k(a!HhG}Xxmw?fi+=7uUn_Uc5l3vfPC@@SY4UdN6dlLW8cm2{1usih?a
ze0%reJ2kshLGYAF5aZM=HDH&rI6ef_WM{9i;J?D&x at _}jO#hJ6A1V0pT#wPji<y=(
zFKpXr__Eg8E<_QN_i#f(J-MPt^f5Q-<=~*iEJC?p-VD(xZt2r&EAK_gE<8E2mRMA2
zOwk)PU>Ts7A5CrNwOAqVHH1B5d6~R^?(t*H9HZ7SwU4%)c6Ki-FRx++L>TMQ>}IQq
zKi7{HRNPq4Ehp4iaJbrUS<1OX#qzLkzMa+9qk7sz#VwDxi)PV{;~XX`pAyGXeXYRJ
z#ijh-%Ag}T%g at fvP8rJL{Yzj0ZTobc(NLVk!R{sS&QaDB7~ze%QSXDj{$lgUC%H;$
zGBWtEp}tK`*U at M(swPTHAE|I$_)*Nx=PG;S#!;jQcXvIee*9pSkYK)Y<x2N-6qkDH
zoj*IzwQ^<T<esYpX{*dS<$h~BZkWdx+r~>v#LmTa?e^^;2kc>APcpsdN-kx>&YEKh
zu?GPGfr$Ty8xj>7=cUJ%w)>hVAF|f``oiVK04n;~7~tpC1hVLfiWuSBcn!^_8k(A~
zi*x&zH_nAbP|gsnbvirN+vchf0&wCH5Zp;xe)D8}B2L=N`k#VCYN~?79u-Jl#!_qS
zGbvwuh_x33ApOvsKB>emnS$u5-{E`;1r at P5uB4tG6Dod;jTz)<myr|}7Rm|+T0h2G
zxBMI^S_N#JL}dIaf(_Q~&{Eo=6JpO#<L7=L)cS(kXLPzWj at FL@pGSm|1PH`@#gB}N
z%5Qeotk at b$e?hzSI%M?3!~`mD>efcodw>#f0s7(f-uFN~<O3k&&T98f25-Zs)}37`
zH_><$qp+>y8-YYnN~Rt0&95(+qA^U=A4Y#{4yG|I8%k2n>`cBv=rG%f13C8c$ndGF
z4w5=L!&VY=k3lS*upKp<9Xdt+E?nK30yk_LH#}SWe8Z{31rmxp#`TTrS&in<H>$e4
z;06Ea>8Q6hXLv8<yPQ9g;CNh)@-j860g{oAOF%}72fY at i`M$7W-B^OjQaS%h-<c*Z
zFSFuuuSC&+j`1fgz2-q($A^4{{&8hx&QY at 8ONccdmT27=xs<DxO;u3yX!_YbiIUw4
zQIAFXIzl3%>n<+B2;l%wfpO;eLH)t1$}d<)6*%g?eD!K#wkrjdnL-(_Ji#R(Nd)ak
zD!PKH*`CjqS#<*)hq@`q^A#E$eIsO(=MYI{H{Ng*bvI3l=wSQIhDt90V=L<Eg at QO8
z-&&mn;fcY(c4(cj+??&gt#;dxzJEWpp=0bo=fk6;!17WcnEswG<7m1->KGj4@%Z8B
z at syMl@>8cmzzUrf6olW4CI5FAF(ZThXo6}F8>iEkM(i4|GYa+UoE*7%-)Jy<qwbD6
zxWjA9tP|p{%vHPMh5IIIWdMs2=blrr*xFg2Ip^Iy=U^+$*zL^aIy<DxtD}PNjMcY#
zYES1}Z)+#d_GFy(7!wuMnYV05j~=}(WosA8*6uAWEuA{{&q*_-yYZgCHFGoyk1X`%
zw_u$|a~<rD)h%Hw`P|mcMEAEQHQwJ3It+4tDpCk?H8pCPFs2a5;oTXD1Efn|T2u5Y
z1Basl-Ip(^(D8~CnGofF=KH=K8?Y2KAA3XFf3DUOyqCGHwv-eO2+cUAeC13a!ym<X
zczAU^J(Pf}-I7P&$_GAqa%5_1s?Mkog8`5MvFr<0HAyvFO<w9}f4 at et<-q$9g6*_-
z?<QWG#@CNG2CnuShmD5_jYv8;I8-x59xE5*8+>w<aHUOXG!cBHMyVy8+wY&6lg6NB
z(9yM?bOofimBV)+F}p|KY305;GK7w0$;n;s%hN?;vT8cnY)Y?(9PNlbzi`O~Eu1}j
zR{5yJ6;jeL&icg)zbcy_b(avBZ~tn#V)NiZn#uz;6=Qp1(a{{MsifHo6)(yuDHC!%
z>XV~QiTi_=Y|Q=M)mBqdW`)FEiw$N1$)GmPBM35deZQ~mth-eY8C<Ykw7Kn<1%Ckm
zYu(s4Q7YU~g+1?j at 5w*?V=&~YqXf=<y4l(Q;brv~m2!J at PXdO2U!vx-^M8R3E^jCN
z`t?gmQSmX7nNmN&%5YNpf;?0`uKor0yP=^Ws$OWFKSz at GkO<%R-sv5{B4n81d&%hN
zf|$hI5^shvJq-(M0W%viQtgfaYXXlrL{e7v1nLz3UK}$1@%43ot+xb9D&W0(qe97C
zwG{WwrP=;MwqmUdz<^*F&>^ke&Dw*b7<7IYKuHFy4eG$}PV;!1uS;mJdZ7juh?h+1
zO|b6aZ{LPM%1q77+!@;0-j+RZeij^T&+vnUm{@g%4qMT*$=G*X8GAQ})B4MD&%#D8
ze(VieP4W8Ym1AYQ;(t=lfYW at lp2-qqFYL;z6Swd|4LAQTZ&hmQ^$LYJk!QzC^KSap
zmuOw6s;*AzGP59ZdniC}{)C2{I^Uexi2uSrUHvzVBm_*ZQ{*|imDFIl44G~lKJ3eD
zZVIC0vmO at q*xTp>kEVjc`mRjMF>~C$eY;(VJj1>14K!`|qy6m=yGe7}JUWA+AhtmT
zA696My?W-{InICPJS|R7Pfsg<(jqdT^zFYst3~9(1v16XuCQmax~eJ;zxc+D8&`-Q
z!DrLK-WePoYJZlT%!*E`Ub;j{M at RSR>(`f0f|!_?z&L!Ms-rfvEG#zbN>+-~1_{25
zrH%b}7QhJXFB;N?W#seRcKG_{G9fUCN`<)tVQXt^1Gp7*zJ*mlfGU{!LPH~vSvr9X
z&B>FueXI&o)6!Vj*l_>q9K_#Fi_d`3IjX=*dqu|6eW38o$Rf1sbKu~#Dp$_#EjeLd
zsZGPkm{XUHi1V9mlfl2WKjr+sJE~us#+;Sf<cOZo^d>b+I{&HR<l;)>w9TuZtDrD6
zQEHY`33j4+me(f8x}p<nH5HW}l(B>Yw6g*fsO;7tc4Dbo8ZSo|_ZzFh;zf-Md6x`w
zMOQy4-uqNhv6H>yd5oC&#;se&q_15gr>ECSt8jE=S@@`Rw8v7A4KxO5W91bsIebo1
zqhK!!*YL0Z3|@3ooIE-GO?}Z#-<UK~Q1J56_iU`Jsgh3U<ZqFw0xWC8+^V^`Ijx{2
zEx=^~SY^~9PcoJM63x?M-ys&raxK7Mx;1KLb2ECWXlU`4f<m;{N3g at fb`yBq)PzTl
z96?8jVf&?1x&X~i!xqB7f~xlo3?!JCoJ<>Y0`v at ziRquh--iI^DDa%?KC4sXK{qfk
zQ10>PJ=Kp_2i|hIY9e=8e_s5G)flz6aa7ALX~oOR!?CeTSZK;#D<~83^76XuHKJ!T
z5&6->>ukGE^A#I&^GuaR?Ksjez9SPlnsE&OiavA<5&n&ex|T@^M<=f6JJtFt^*>JP
z96CR*;kDr6;$AsU`_HNR?3|n?kO>8r1N>F{+Y6|ydfwQ=fFuwX#Y=X6%`?p+lW|~+
z=>vkNPv<kb{EB at ZGUwfS#a-XvPlyg6pfcGkwz90djg<!CI_C02YPB2<pFn1>!xI^%
z?q?6jo^ycOE806cqVe-SU+Xu*qGHZ7?I&erWmE;!EDy;mLZ3gs4je1+39|GEU|7M>
z;wOk;e?pqn8+ZkmE?r8iC*q7c(bd&ujoXio;Ogk;L}qPU2*#pA48{ZRA0X at LODg_!
zpvd&saG>?!IqvGsk;V>*gXh2}C^sNoU-pcB*ftQLx10aQfrb%b>%s3gu1>ZnX=oh7
zAa}l0K1qU(dIbdqq4_j$FpA5I$qScgiTa3L)yvuqcvS`xGW0b1=CKnmRNtXFoQq8u
z8$R2i%S2X()w%_$oSKmui)(%{%?7r7pS1B}n#XMUIz`yokHyEwE1!P<Zp^`|l!bMA
zZe#nZe@?Se(LmPEJ9rig6#^xRL%FTiWhogMx3nx6im54`KS5+)fj2oiFX$}W5JhTy
z4`>_@=hUA0 at nd4XHyaI!AZfiY at xHkD+gu@<uH5T{b?V+gO`**of9p_4QPk<)*<5Z}
zu006-_U)Sm?f^Pp42ziiI~GtG?H4*ABC=fN3vBQP_Zowuk`j5m$J%Wqa)143YkZhr
zt(M(!GPdp9(XxB@?x7zM5GQPRL at oOyeF#3fSi1b%#4a<mwm-horM`P7o!@sYr7Kat
z)~oziyU5|e>m at 7$nrU0HT!;p~Q&Uq5F<g58p5NR(S1q`|zn{KwIOIyPdnF}~O;_MW
zYUhQig&Vb%m4<OOH8s|aLq-#C=a(q^Z2S)zX^HU78?>CIZ20KjRJE4H(B6%k{IWTI
zQPN9{_1MKH*ogx=ixPVD(*ZU%{AE%iwlPn9ebdA-|NCJUkgR=uWvw-bg}Y<OX|7BJ
z^CZLFqEk<x*s>iWtqG%f^gq7jxQEVUaO+jkUNZTrc*qRODJRScX?XOSC>2C~&@B4=
zw at hFM!>HHxa|g!D?TyKz9eY0B1DkI(W9aa&-&ay3dzpBe)0#H<wsE(P=3wz{`;xnl
z22;hvta{?@2jc{;SXsO&HSZCydhzbu<&aE13!S>7G*nWG;dk%H`fF<7(*_s at y#BK{
zzU<om^|1`)1C6B)$LhYXRbPoZGCu8yFaN4u&NPK&*j~s#Jkeuda2gHkfQV}DX=L>y
zlRm~0S)jpP6?>Ubw<x^Ce)f$g)>A6BwpM~vGx$1djcwZ%hqfbq(n%|CXu@*UxUPzb
zxE|YCoBp!vWomj}--RA@?bOu$uL%;~(#J$tSPazyet&FfSw-jl(3zOKcZpHncx+_r
zH}7qZD%=SFba!rfxkX}sl@!fS3)}x7Ld7)Gcvt+PQmbDs54=NUe7CpVqH_xi$w4@;
ztgcEudc at Cv>C&H4hdJf>ixR%i!@~51zV2zVEPrkwfW5p}zsPW$si~>yG(UeguJ;l2
zRCYqb!m|`l1Zjy11}tgm=_^cV-&{y`BE2^H$sNm19T58T>Ep68AutesE at +$W-FVX5
zzD!3>ZpU!ocbw&3n{`<w)usDKiQL`Y^BGD at 2P`ixPcAO*eQ5knB^xLHT!xKQYhlC9
zYgEh++t=4JqlcZ*#W{WE%+w{{?|giGlr_{$OiMlro)6bIH-E|ZahqX#dT>KzCMiBS
zExUj(KT at bc1YZ%i at Pe)%Nm4u(biKj1&vI0bbLvVYF)8UJ2vcFtZ2=%kB~8uGxTTQl
zKpd5GlD|v(Ed{H#CW5yHI=liPzbDX%ffr}*Sa+RgiA2Y2b{5N4Cz%gu(Gfq?Do##L
z=2?%2Y$PTo&dADY_+USysH*D!@uLh}sj0cy5FirGaW{@XmQTO-;KA97%1Tsl0X>GK
zJh%pp50B0^8LeK}PVEV<H+Za^S!EqFWGVP-hR|P{Y09ZK{#8uOt(h~IV(epfik at KG
zhwADu5`j)LjB8F3rCT{eAJ!{-ElMizstNG%U1HqL%#Ilw8`Cl|1$3EBN!h9mha9cF
zDI>!xD*9EVC|2&={zeDCCc$!Nr{c(*`}?{wcj34_U4`Fzhq3DSmCDi45glYo>a|$i
z{narUbUuD%YrAZ~!o!2nrfL=6Yb`QCk~Lxsg6*b>$-DHx8^>5KeJ>_5$5lVS at BFjh
zCt-g*eRgiHUa0oZ-QBGgi2z0R_4UplajUC#=#UCJA=ouE6bxub58i+HdCf67_*^o6
z%8-x{sr<f_^z=Z2+<}S9r5PcPbP&qUu2zzg7*gHfW+SEoClW)Ez&x9CKTQfX0BreK
z{IK7 at 2jt~l95Lqx>xY`k^77~m*w!AHoGZ+-BV{(sXfhXZ4h9_upUBqV;}o<S6yU<0
zp9*GRU})Um+gWWFSw=rL0~A?Zbj|wNLu2;gQCfPSv#ww)v6YpTXmMMGwX=gmopl)-
ze9(svAJ9*nk(VcVWN)9zKhOO at fL@}=c(ddP>5?ut$BHC8l1HhJ4-QkfAbrTiQ|ZH-
zh5nzoRm6mah3CX)LREwY9}ISMkOFH^Hgx9rzI3PE{)Lx~9StUvjwvkUV~T&^qexxj
zL}DOfvlQJkz&I2VZ?9;WIoD!cra+(LG~(v&e%`~=((>!Hr-z3J8MwHZ7}|a3#j56y
zALX|<2G%P--{+UgQ~_}$nWPA)E19X1oSseqTn+$jQ8IW|XLe?WAW89QdATrCO=xg%
zY=Hu6#MfsRD=z7}yNi(CPq^xM0%97{8y%*9c9JVRAt4Mh0#bT1;OYf$rrzFObOKjN
zS2yH*yvq at Zoq-NA1_pQSnm!#TU?OxzZMWkOfMU5-Z7rm<g at wgoc|^>eT2)nbc4;Xf
zI+{lL-o1#32p@>X7`b!BrKQO=HB7MLb8~YVl};BxUl+P<%wOsCR7>w!#Rw7FJ3Bfi
z6%|pwefzf4qGb1jl9CeT3D$d0-K`M at S(h!rk^?3vl#h;%N{kB=Y}GdKw~sG&oN&5%
z^JY)s@|phK<FWdD^dWsj9<gmdj*qWIM36rZ30cjrtnD*B4ocU+V2s!imz0zg9b5-x
z>LMvBEKCcZQx&6o|Nfb_4~J<IKJ at WE|G31#FNxL4(LEBR+Q7Y`hK~a2L$xbTKWWjx
zFFeo0=>Fp9%CCe2MgY=yCThqqvmU|JHVtlSRBPU9=JE2Wua^?tU1>y%32QMvm(=FA
zx3}l=t4(M%P+}KHOiavWceQ0TGGQ0b&dx5Li5}8L`cn>{px_UR0+YAbHCh(p<C$hR
z+{^^%iHL~Gs{{oEUh?P}pSp&31V{hd#|r)6>&ZMi1~XbxKS5omWM>CE6HuJ+j|?b?
zjEtOnvU^}!sG+E?-a^BKPEi8);$whtm^Vk<dFB at u?0|&PVW7#6A^q<k5Te|fY5Zxg
z>+LP^dwDbr>?isG4CwrLxg8sm+bj#3Pr2lk{0)qFULMKi%a=3q@?;zxFKcOQ8<q{R
zPz0UBE1Mg3ux6mApYX9NHQ1LC3~uHhuO6<9zDu29I|6ro;r<yvvPWMA7M+MlNE+_*
zulkoRZ4Mae1UFA~V?VnxFUTx1elpn+C)CzqIC--9T1VufG0wuLPq(5yL9kOk>4gnj
zSzCK_T#ZutQ?B6Et5*$~?xu9fX^q*sxb#n#fUv1 at Vc|0Qv{dvYU}?u%{aA#DS9_8h
zb-{Owl8a$)MRY1VijEx6NE?k?P&YY?Gdp-OGBAi_&xim=7i%p#`u0I4`TF|qUIro{
z3vLeoe4i}nI{c1iyHq&+ue17SaR0C$xbOB}D_7WiUlehTl9Eqrt1Rr&--C9aIy!C|
z8nS?J=bJtiq8T_d6YowUlR`ea`$K5;XHiG{V0-Jmxo?^GT;|fEQGb%^<jL_Ubv-FH
zweK&(u4VToXJxfM0D+90q>K#yeSW&1P9hZ*6}I4XFp^GAf{;K;Iy%o$+6oK1w6=)p
z#<ZC*^~YQ8<d8WlD^rXf!NnEZy8SdAvM;*xcgXPjou#U(syw=>8ntqFLF%H{vP-9`
z^81KA?%&5VDf;sUTQ}ep`Txjz4{)yAH*8!YdnPNJBxEIIW+X{bd1UV-J9|a4vnv%6
z(niQ0$)?9nLiQ-h9)*bib?N#2-{W}S<8ai0`hGv(`@XL8I?wyQ#s|si>jlT8diS%k
z-cC<XKchNaQ&U4qPCm}PFgj|EnRk4AL}2e*Kk?__RNme)scv-WIJ5*O&z_~I+8kOy
zqY*fAq$Vz4m6w5$(brf1nBl2Y!~siDW}~Au!T2lZz)d5*!-xHR^&vh8X>G<!SxF=<
zB3667-(*itPS%0XhRJZxo;_P!Sy@?Kv`H*4m)%;53#=d6+4frB+?>d7;;6f%tq=+d
zTbm{Zo!44jETye?&NZjYHobcqsk0cfu_+iMInV^AtbE#H4ZiWHKK;W9M(_IS$6NqE
z)1N&fZfk3^b#nR^G&?`<#8J-})$rGSuGReE!+Rn6<~KPx4SjvgPck#}{Wn}Oh43$}
zP}S0+3^dW!roqZxtit~K^(*tigR08Pdo<6U&9|uAXKig=Juom3oEAOk<XMqQ@(Y_i
zVCNw&E`piQIjYyL9VvF`Crr?ooVFO&tE;O+gIahE*cn`0EG5H6t-*JPtj5uB@~mQ1
zm#w>jh9|G`LRFRGgIKfb9<C<qp)MlAnh>3m6?KyeGH&Z{6{ZotqXyHmtr82;)47J1
znny;`M7^mfDPik~7UbK?xz82Hg}&~uy}Wh<Rz4NL%C^W*AW>5>?s<>h;iI14hfjz|
zNVwL at vsNsyH8YYD>C!s0N+fBC47H~Qf6ksL^DA+y7 at A+^7#ka_y!p&c+Mcn{X3D8O
zD7}n9<5-ebB}LZ7S885f(lizYZ~A0YW4#jAmY2svl?_G>c at j<vCh~}gP)j65?_?(=
zHCX1SzkW>)i-tuguV23wla(b8I(c(z%O8(bh4ivdIyuc39-+M^$+`r*mk!$(?Ay0*
z at q1?6IDoSVER%vlAxnc(36`Wbhz4$))z#I-^J=BvU2%CnD>h&lZZd3GH$mf2Fe3zH
zP9=jDv{!Ik9F4Bv0dOc<T3UQV1}3HnOxV}cV{B_UQI<qoqQ;{PZuG4qZGnFHyr$-p
zT2kwa(g){lO8D<ytMv;F4K?w4lgi#Cw~!+8KGP<vtLs8^Y;3K$m^t_puy2T3<HSTU
z>bzlLVeq;BRULL1wl_DwNm4~ZLSkYXsE{oH|B`QEm>p1V at 5ns&LSCL25i#*nyqAlM
z7-*}Fnf at EQekE~?UD2NzM7KO<qhceQrob%{DW^dV`@_!hLrzUK!6evaWm%~X8Jd8I
z$gkB-a0b03ac;-PF&9io#?(D*KLEmPFmUG&(!dKkIw$wfrd<LtGVHgYu-o3meVYDb
zclXk2xAugP*#<cTNqGhN-G_)fw(D7K<twYJtEKKVd)}C8J<e_^0w|19ofseIqK|w&
z$@8|~LH+F6sI$x|3LGpfdk9cjC at 7?c=U=*b(VI4falz~$MGprMwgU$a;3cVPZg^T6
z=i9e$;~zX|QthW;6fPZGVY>Sv1OOiW3G{V1n3KD^B>CSKL;#Z(AMPBIT#~8lcaSnJ
zb=lJt$2qkV23Em!_XF<uphvF>EWzWIsB at J2_O_=fbyPU^e)^RD?j1V;EhlG3 at WAZV
zqr$?ib-erqHO<YWzkmPk?Cm9aXWFrs&>?NJ7rtL#&~0pKnJM4nv=(|rSYvc%W+p at X
z9FSqo84gJa3Bur-YPoZ9(b49xog>mf0x8brGQHdu&KA-Ej&W{?hMe5h!23r)8m)>A
z#pl$^uXLA1-pQu6f4v`?pK2oH-A2wv9;#xxF2==Gt(q2FS}-uTU)FfWUG~9>;M!u5
zpMDc~Z5rnZFFCJy at 1qRlCgrIeU#X)Rqlh6gX+CpRp)~^H2duB8<bjeBa2PT&GI{?w
zj>_1$xNr@=C$$>|A#FCQj*drZuB6HP(DU;0V&k+gMfMkTbwj||pFVx+?fWwAnu7zS
z^ZP$i0U$9mV12-*y=U{Au-R7I#zyzd^!$9I+4eG1<)2>zIsy!Rh1d26tk-d3y4Yim
zFH|h-?5801V-vgX7uqXMfJ;^JEngvG-PFrE1sno`)_mp4`Fr>8Th4tcaw~876~oNJ
zqU`5a(c<~_-vfmuC7A^UcbOO|sHq=IR$`etA3y)c>&#_GK<3Ej&B;&A%XcXDf2$fi
zE+f--#dPSMM%<>nzk`FY$-?7(Vd3H85dFC)9mSZv#U6!j^jAJ|YaaOX6a4ez==)=9
zyN>i<;y`>nEe9nH#4B4{fxy{qk6#Z1N_Od<R?hyz=SP9cR%fV&V}))LhaaWc=9gZs
z1PytG!fR)8^d at ox{|Id_KX{*>l|_N=Na6ywtQZ&=ynSskc(s+r7auD!j7gRA8h^Jo
ztO`p)Pfzax3Ml>QQ=+WPxsSV<%$(J3+)w}>8IEbWiHV88E7su^qe(dZX<P0eh4_X}
zrs7lqS_%O9Wse=WnbvpLxxv(?cgb$v-s>I}#&V+cKR?|{2O=yfDVhH26&VIDJ3AE!
zy8vY7B?G8FAsXU2w1NWrkt0X8s at zUrb#-*OU<C0O^N8;r*0wY^C+O<x5)crm{?U_#
z9f_WueK7pdNgiHJo+x1T`vrJ+HFb3!QBfLrw29QvD-!)jHy=-az9g6keD53kla5Cx
zf`Qs<JiEwIWmaT&z$2(^5Aa&~yf&wb^77G%iQwpH%E_sziCDS$hvHFMA7;&^QcM_Q
zPmt{AA6WJM^K|Sz0MP at dK!=PZ9Rc_Lj#_dECmidAA+*a4;MbH?R2H_j5#cWYD#bzF
zau#qE`Q`ZnutJ3N&;Qok+<et at GBP@PGD at fpHkgjChRH5qIQ^&bkLoj|+;^vEW`Z at J
zja%r%sMl32zt(+CHsrHs at ky94tze?2iPIGc7UJ9&Nf1Uk3V{@Y+>p0zyKN|&wo at v4
zIs#rb!*0qcVtz_`wlci^`gYuqQ at DEOSZeN2aG!~ZNvEykR);*qZ;2#I_;M{{TUuH|
zdI-*EEE2B8pXsz*@ThLA8kW^%vYC$UgewvDS*bC<slitZPKb%qcvZ48)v>m!qa<B1
z6w7-$7TcH9&kt45F*8Rv4it}_HTz8u9B+D!1Pc>(`|<cM?uq44AkWv2KFxW9!EEPu
z at 72{2h3=l7hP?yrL&d79s#tLpTtg5T7>E^&MkP+CfMA%&m7VddU>8jM_fY9v<@8Tq
z0zAKYdhv+p at 7c<u`gu3W1>%ir$6So1t8==i`G=Gg73Y@~_w3n|W}uPiIQvEEczF_w
ze)E8}M2)$a7O78)7C$8oO+`qhL4rEty3NYn*`*?HKR at 2T$ISAd6#6F?7hM!-rneq?
zY*T~W0-tL;N{`+)nEv^*UggHO6f6bl92tqbbBEqxpqR?2*db}9A(v#|K5TD;-5UY<
zPXoK+xUqU+SF-1s2fYYCP0C&HC7Sx0oV}A1_GpMu at cWhi{P{lMqS$~CTLJ6qM-&zo
zo*21e^rCxg^21$hmx)>is2vt>gGZwF`J8RAmy?rIo8~Tpz$8V1VNU0?dnGT85z&u;
ztR6cPMly9ZHRF<&qLW;s>^{%CiQj_PVLP~(n3%2NjEp_3G5e}}&JSO^cFq6nxpSus
z4Otn5^-qSzINj;7LsX3AREkApsgK-`2R1qaY9I&xOB0Najy^JHhsPV(OmlU*`{k=w
z%2rmqjZIB-OiWFC2a=MKz{D}v(Bn#~<;27Uwj*X`W$o<jtd=`jK+Q?PBvOw}(kvUu
zvki;LKYaM`m}5B`2(ONTK_t-2jJI#;L7f2?4L-?rG9W+^ZxAXg<tM*>?HU)8aMaKF
zY`?G6pgG9VFo*HRjT<2GL>6kwTl1^;&PX&#1)UVjXkq(f$`#^I=P<@&G$<n}83GIN
zph@^7pzV4<aX#@K)C?nae@{B@?#t at 8Do#)E$hR<Y?(5Bb@?`zSkNaw*+|nK9VvA0d
zy+1z60S8LsWM`*}R3nU3yH#2$6EMS8#Tco^@Hl0k7XN-<KR=&p!`%b!#m(&WLw>Vt
zE=#EM^b1t(w at -1-kJ-!aluD(v?Mu`QZErvC?&%49yzZ-Bj$WNz5g~!UzdsjaR6=~b
zOB+|QzVPD}d%kZ^fJ3AEosQ$c2A>%{c9;eY3<|xYp`k&)&^F;kxR255+S&v*(RgE0
zfsF at 4LPF9D3mL$D8o*uY4j!DD=02a)*r<-hpqMS<^5fGH(9L5&80+`a$lq*Bk?#0W
zq?|qQE9uzP^{u?~FM`3tKy8^9^vB=37lQpn(Onbn-~Tqv-Qf+MDv~h=4_9 at BfxK_d
z)7!^Kng8yEn<?I{qobo9a_$vO26J<BU)d*<Ky}qOHUepGzw#zEhu6HRZkG9ByN69&
zMg})?jJhfF$3NmJBa*OxE2-^q{VU<&;kg at laAd49;LzTEgZiEVH5JuYx2Llk?Qzp1
zTo`3+<f0mGA);Uj`u3kqi-LZXTYmFpBnPFPLX1V$R%u@#hfgB`Z5s?!1Uzf;t7sq#
zHbvdq+LBjLQz1>-Z*xv#Ajvcrn<!yP;rQZM9WyiY&68Z!|0W7u08hkP_@(g%Qflg7
zt2)jHQ`*A7{dNEOXAHao``;#~V8wYjFcOF1N|qF9XZ784M@@`{u65aZGsjhX4Rh{?
zza6UU6&Dx3>p5;w;*jv&o!nI0=B?N-0NBp%?!BXS%zk6Fp}3hXbH+&y<?L+KBtL%q
z(9zQi!z(LrFr*#E#%wg!x!^6+)6?B{HJKl*Re&Ka2pXH1#K2==3E}4EO6>IP$Qeyd
z@;i6#yd>OWvX*4+8Xk_lf1gqE`0=H*F8GHanbv>*#-2tePMpBIB5=g`#KcokanB=q
z>>4&!rY?XJVyXnZvCb>q#kIB5PEI0NHE`<|A at Bw+9-i9nx<iv;No_U{MvL9zQ-}ro
z#wChJoJuMxPrJLXK4qWbpZUao at Zj0^qNGX0 at M(a_F6hp&J-69oKQ&t9SDk#L&0NfQ
z#V_yFEeM>X6cnl;CCsdu85tYje;Yo-FDx9bWTR|f?xID?UDCV0E^|-w5KWk;r|0rw
zkxIr%FRzlb*VCJuH6VfyM_%T?yOb_4pp=qZciu}XC5Fk at l}2A~<m$>xAdVurG-PB!
zAe)<ZvLg;rq~x8yHM`_M{d>rcI()B!f`YvJ99!f3_wP4VGDf*)L36gINYg|{Mrvqj
zO)6PIaD*@V$PmH>qUxTX;@DDPtS<aIkPLS(FRW+-&WXo?pWm27WoKufSSxtlJ&jGF
z{8zh9W6NP-Mha}7H$HBOK8sC^EiEnO6<RT1rRU^O9X|i48eAJIo*qBuh-k0`wz>Rf
z+^x*k8HCM$M>D}BG*5tOnw^f0jI2pMZol~Hl6J<+m(6v&sW%!M8#@7sTONp3gp at c-
z+PxpQ{a-FXb?T!>T1|=dR?+IbC=0|TQ-XV^-9?%{IZzKPd}8O|XjdI at 2pHOsNmj6F
z8g!fU`f2=)<BJqCrTiyHN!G(RS|2O*PuMU0{P}g|ZF{I_zm-(rK^B$|a{h81{SaRk
z|41sG#UsaQC!jWXmxur~TD^yoni}m34+ySb7YFR#ldnHJZ81y_3y!xN|21BMXU3q5
zloX4(LHx)*;18jtd6>xUcoiZ(#3~S3)!3PRyq*k1KDIiC*EKLO_%B8M;>C+x4_aRQ
zQ#!5=7?=s`iH&Q)EkHNL{r>IyP>Ua%-YY3_3SE9v|N5V-m#WN&AkPosaoxsm!xE4+
zhy2 at KU#JMr8FIdA7-|mC0!aC(w>J}_B)*!e>RwRV1mspDP7pLnXgKP at 0<cUwM~?!i
z4DPJyJJwUDPW_GXKHaz at gM-n}pYvn)TqFX(<G&0H_>k~r=wSa`G_f;h&TLib3Wn+m
zrmkdECM3|o#{=Gqnj$kZ*emQinB at 5vdtH9s@^W>RfZYRy87>;Q(bC-7s%B{TC3<tm
zws-~vPUQZ>F3Xc=@Z10x_3-d8`~&=?$Ruvvx at B<ej5eu@$JiUc%TC2FTNOszex2}A
zBYG1?JQ&Q{j_0sJXPamZD^;6wu1X|%H-Fi<Kg7w*?BX-)FDEbWKh)rtkdPoUW;Ye;
z+dlRKB);0IQ|G*$)YLGk_UfR#9Ct{oP4AEP8MbB{x4pv5tel)Qk!lhd$8wEDfH=Jj
zSLzzHDl*LIo=)$v>jJb(Jx`LTdHmW7`*)@b65&5)X5^-?mjj9RX;tAV^qz_8vMUl%
zGIOwZc0PUiG84 at L5fKp=H+N8YUTO&}U4$ABO(4C5N&lSZr%#{oVg%cu>0G+B*UZce
z4_)j(XV+kN;Rg>+PmP~fKYbe8CSwf!`t>W2|FAZjx~+|se20DsVWVP7KtO=quEK?4
z$C81r{{F})CeiW1GB-RH{^y?q09!mfJV<%4R)C4g3SipZ)RQVWI2dz16xT=ha at cvv
zupvbH`pV9~TQfzm2qA+ZxiSXqzy)Puaj~g%_0b0{33u+?`t(T`)XmZ+Q0`UdM_NMg
zN_boF at Dcm_f(_3pt6#&zrhw$zhN7aP9G^WOJ73*=6gNOjSd)E3QpG6-uWkA`SXh!)
z#VZ!JAA0P&8Y<U6c(Vd<OOd4YTi}rD5a8*4a*wy04jf73=kQ?DS`c3Fk(?Jb7Ilk?
zHvR344Q}(lvq0=R^yE|&7cHs0!r;uunV+_{wkQ7+#;;!s3NVmk&(Vh42Lwb!8p}Oq
z;!{%2x-o~)j$0HcK2=}|RSEj?#qd*4&r(ii=3W9?X6Ek~V~!m;LU!>L!}aUe<29er
zl80VWywKU(8x9Z$Ms8{_P*!UT0{doG)Rg}BM~*BkEUPW|k;-yd<?Hi@<ao%kGmwY2
z<mhF2_^=x>2;BX^MIZU5e~+J32)mq|+~JUXBSS+mMMY{*3c`kkq<=G{)NE|{L-U>Y
z-u(~)I&r)+Lls*Tfr+6ef!Kxph%wjk&Y`n54JtHQJ-3<6{OLq17YXl+nA{HBUP*%|
zmA92QSoHSxe)wKdQu5x8 at xRs*q at gO<B8H!TAL!1A+qYHR-6cn!Z>pSxSJ>X(CNnZK
z8wuITy{Gx??876%!jyzT1Yk#=WQ7Hrp!e2?*Q&zTMglp(k-V2b9A27p4)psXx|}kD
zl+k<>xWvn>tU3sC7cX7fZBq@}3S{^0?W7J_(nXA7spY9m56^v0=y+!Zk*W2mGOeCV
zxCad#d1ypz?61xjeAaeUR8;3k^|Tp_5x-sLw%G9ILgC%JcXTW)CoHU at zV#iYxO5u4
z_e)PQtx9vNq?lNBE~jakYr`S7AA_mVVqz{Ibp!?km}QIHW!d4&*VorOK{HXaA-I<{
zCQPZxKtqF+kDe?<W$dYiz23>lP}%03XBS_s7x}K4Cw;Ic#F9~NMa6*H4(6HIqlubV
z8~k~bSm6O=ycfB!B_ErtJE`wGY~7AW=+Dok0Bh4CmUOD^?4-rvAYg2|moH<6rj7su
z!JBrAr!QXc9z9C+>eVZ}r&i-pR#73Ww_LiH{T|P5fZ%<yJI+A9n_3QNi+6NoWh_=!
zR*OG>VrwPuX8(=luV5HYe=4Wl{a{_2;#?KUF6WtET>SZK>9w!_CgL~>&^7l(&CWMy
zJ&$m^S at G;y=!lfGbY^ZY4FN41TQr~y*`E1(_ZT3K_^j`zb-h6y9vohjD(6~o1EQIT
z+mgc#fr*6$n%%p6!oT<W`TDIC0~DbuKu-X$fR_1vzjP<5*;>NF)|M1ZLcVft>*3JV
z&7t=;V$5Qal9LxhstsIR^3L8I at QM9$VzBL3MPN*5_4LDV^a<%YYl$Tluc`Lby8eQz
zT|GU-qd$y0YeOj#9zSlMx~1P`n6K|YFy+EdO(Mxk1?vH1Nmf=CV-WH_DXC?*(U1@#
z&`wM7KIhL<w_Skkmppb@^~+sxW=dNpCv_vEl(Pio at 7}S)W^MUE4B4ucGV1$Q85$N=
z+uPfF<*hV?Y4|b0=takN2PrFecLu!3XJeZ at XcW>rFhCKbzPF&ibwudzSspgay>s_2
z;9c$Bukih))fcCbCX}^X=Hq?uvy10SJr?}uqW|%EEDZeX-^WyA_Fn3lbeKL9O2OpK
zTv1iU0V0Forq9Z^C9z-Q4N*Ae1$N;zpZ`G_HfrNaZ at T|+F`xzy+ia|DY%Kr${)rcT
z<mBWwLwtI`0#JPy78Q+cOr<q>z27tmc<ibG25485yko5W%Wb5$S309b(B*h at E8##;
zP!Nt=(AQ at sBqYQFdcahoef!GY4Wn=0cJ-lITM;6-tGC3#%uL|&bD<Yx;mT1)OI8k!
z#^z?yR-0oqR&perN${WQ>+4HHx`v`oY4FjnuC8*)$sN>I+~70y`I=~NN;*G1U9%9l
z<A=$0VkvFjSUcW&tb^<L at t?<bW+SJlXlOD){2n-P;J-Tslwb)7HFB6H%`GAEv0L?!
zwIK^EeTJjd`v%5`W#JV+ec!d%|JmB|F69#ui4E5r4Cc>{^3u_X4_7?I?Q-Ic`9O7b
z_0vrt5#FP|E-tj56*p=(H#Rr5N&f*nFgfNhR~1M2IfPbXlZA5M1D%&E$Ip2xbdQYe
zlylF5&ZP;w-S2E`8{t&}4?orfYW2&NG}GKEZZUJqP8B|SZWMQ%6`b_+<2Bkc(zcSU
zg at uJ$-)H;3eEB!H58~05p+ZGKN5E%m36{3GJED`IuCOWoui5@!oF3BDL_$nVjHk0W
zK%%u(OPl7p9?$?h%{MeO?ELhpzAjxb at pgE$|BV|PJ8iSZM+F6MJA*w&L_`p{6!8kC
z94j=;sU-Lg7#bEH4$(~vur10p-|Y<Jzg9qkuu_8K<dr?kupb6qR95AAhB+7-wpU*$
zjX{{m>vzE3I9;|OpcP%+-2Qh`4+x>lRkIhlbH__J?@f5{tv*eV8;iwmH|4mw$HLyk
z&`AvG=X at qxnaggLl<S3Hu9MXx#!M-a3+(ml`}gDH<K8b*Om?iNm}kA1xl0IVKDH4K
zX0N~W>r0vQqVX^A_|$7S%mHghKkoUM*NfVtz+<SL+qt;Q=Q!S-luWzJ4(k8{lASwu
z4tzDtFy~wE@}EC{HpB~7*npZa)YjIHE-ZA!9Ra&tH*N!K6fJL`FCdg at PtnUdrL9d9
z_~!3%874G5ejE!9#&i7m0SKA8(TO0Y3CORF5P<i}85B5u2DPe9E8`5l$QVTofoJQd
zurPsfDw}x^85kHXD=Rrf+z07if6L=(_wV1kdmK_L=hJPy%dDMBK-eRf)fL1 at pJVn)
zSLK!NvK9ALhoE%7mtB86-{@ZZY%2*NPu0YP-Ok?r!sW~1#l=#6bDnDI>d2j%o0~^1
zW}~Ai$#n;9YzkO_xgp8R^I9;4SA}4lZT6l6<9Ha=L7O?AE8%rPa|0z4WmAt;0|FGc
zx2BYMPn_7eAD9Nw`b@>e0`rgGzYFA6Do9C%jg47=B*3f+H~y5iMhYe|z%l<Zq5W)Z
z@)-|<-5O?Jz7$YgV&ULGW)Tt|&I^d9tgI|AGe5i`CoB8=d2IkT_^+c;ijFxbjpyYR
z^G`~FzV9Xw{p`7nYWw5Ii~b9KBXz*Soz^$$I*zf%$miVv&G_l_=h~JQ3T$y+U_*sB
z;QIO+6ntc5WD^q;B10<rH&Z+)ER>z}2%3G`sR&e5Rei3~PSH{rIW#gpBnz!IFd&T>
zWQ73ID<4~Cy6*dy=QuYv_x at PXM{y_KFcm)^;e;KynKc#H*4Bo1h9YKFvmc-Jo`_Iz
z9+o%lObS)Kd-pE)gO*pBpB%-cr1k-97a3B$GwlfeEKE99*|EV~Gca&aC4-hQ=q3rb
zez at G!Vv)YEy1RQJiPG5M(=J=%+w3wsmH8*H*xG&%8K1_argP^)fv7AV<|>G+8>=pJ
zn_9#c;uCR7jnB?L!0aEnh06VS6a~8_K>1 at 2eA4d#BSX>p;+Lj~i?Snb=jMu<)z8TH
z&an;KtQjPT6z54xNYKOP#cS~4p|lo%uol0ngF|+cqFkA8+bX8;rfsht2Fi&wX%JvK
zZ6$3T9Yf)Tu+9CGCp=h)nVx<#;JkvZzdsv0JNw;zr3MyDfX;Zg0F)w%@u-TSA*;l(
zV+r@}sWtr5oNXbMvQZXw{>4P-FN64s7cat at 5?w`ZvokOdfygT at Ej<Gf{|`?>MEi*k
z_rzIaMD{o_4tp+)&J8vU{IPGJDM{Vf*eDAr?*C>i^V8Rif|BylxDwUw4GnQt4>>F|
z at Z|Y<CsXDR3g!y(@~tV&1TJMAti3WLG!zsmw5vAuZ@#*o7 at wHnPS6;gnwrY*I8Vww
z8I=+Li(zSbxi*$vZgDi2#@AN>v*D8J`Wj>sHHkx2YMAbWQ&hzrRowe#ANLgr7Bhxg
z+2()0y8b1tssaq{NsC2t;3L0&d2-ZaCcwtZR7RIA0fB278z^a!fl8khdHEk at s+nEp
z3&ZdFN}M^V*=30e`lAO2jj^LSo@?ppF)j?t8(k|h#C|0xBRT at N-L^(al+vm9l=Ok7
z!2Xa=pFTC;+5VGepx@$DVzo9uwEi<UFsA#3HV%9n+1Xyh4v&Tel6$ErDMxV#ljkh+
zzb!3YeSHx(ZxY~1 at XY6&%s1*GgZudJlF6o0gWv{ABEnAAPk8KeYUjJf(|F}vJvST~
z?=f$)`MtKcwP#=;3PjWD&!0=1fL==TGczf)rJknI&02b`4L3%tJCd at d)DfA6tE&l;
zlao8)-M)RZ!NYKHlkY!%+^QRMA+s!?{e>ei%xehAEk{)JvJM#LFp`H7!b8BH at Ep5+
zroxzVW at l#HeAwlR-9^aj3905!1%F at ox4=Q_ATqAb&g#H#@D?r`%U1&0t8ewS9~8RD
z#TccqMb_+KQ_%0SwSFD at me)U4 at JrXr!igw3MkSUwCrDWM>5i%m_E7-@!dAxXjnY+P
z(_`@iPhY<#$MnYSrMN)s<%Yfl<ThPL0ACL19nNS0AzRt)sjR#QqJg}E`>m{|;hdZt
z<aSn8yf{e&__4dEXRT97-N*<hj&;)3Zh}|Ia~V at XmVM|@=r`Zy?5{2|>y_o2&&J*6
z;3MPb&A6D=BhqlmfLoAJz+y_xfjv at J@ara7lE|aI!_wvXmh$_8gP6vzGcq%Kw~4LA
z5LHdDQDEHKlwa)x{swH at +nWzoUp`3X%cT&NFKE4GsrrX@=I7_h*z^7oJOFGJXuL2Z
zMc_FtEz0*g{0SQC4H8o)J#}^Oktp at 7S{C$oe*Adr#}7N;;y~RQnM91MK{5FI`^P6I
z-v*Rf7W4G<l=67Lbd!QfWY>lZuF|~D%0k(P-NxTk%&T0xHvPy@*FaA%>@<fDQB6$^
zwqQU7tW9(6v)#38H6U!c7|1cB3<%s{=L}FUY>G9J8K-zli+?v!2XhI5&XoAnPkMIQ
zN+N?8S at l(@w-iU>@^hxw3K%FYDbZsNer~r$2s&o^jj{*d#tsViwJ|;_^ZL$6N+fSv
zSI&0+bE**+ZRf3x0b6(Xdn71#-*?nP at by#?7_ccQt2_-fbf at BWWF$80rXtR}>L<<#
zSaRufZtn^~v75%d0~-_+6cediI8g-yIzU=vWTg9}C1Mc!!$;rz&8Gx`vXy`FRfu|K
z(D at g^D_>oMP7>tnQ at +wA2Za))+X(o;N3SfHsKQ07)0BM;u at XxeRaN(j`d8oPf;frs
z0A6_OYFpXfce;W<R~J99!G`RXd7&Y{-~K83-CcHaauSCp)Byax937$5H86-gAf%TL
zY7CgmV-D)FwtXZd*dZP>3$eA?^2*B6pM^k908Kne!P_RQrdYsFgBRF3f!tC1SY~FX
zNR6W>B}UYxrR#WRedGvPQc{x0P)Ya4kF*R7lhb4SC9bHet9zeDFrII2yKVII*RSPX
z%;x96iyutn!Ag+;M2L9JXIqb=;*)NA{MeQsD3bLdAt4F0Y2Ng;a26!cMwC`!_{a|u
zCFcK_=?0H}FXOX|k$~C)4CDdtv$80_Bf{zMriMO}mXlNI8VT}a&~@c5{_ND at d%@*i
zzIrv5F}pI=QD-a3BPK?R4Zp!S-xd at kBqoLexg{bY@$>C-``FjVj2GHC0cE7>2Evf8
zAl`P)m(PeWFfnzMohpD=6n(!X5wM!cIfg@#W^&)3&low<+0_*^K5iKj5;D3tRwrHb
z5_3avdf4tfGE-Ai9JsK!+O4fB@?fNq?(RU_m`nPbH`FB5Y$pv2Sg_4Ltf%$Wt5y?@
zF#<=Ao-#G%@F+QKoy^R_k|bF_H=Jse)l=&rwbGmahP!>J><~Q}u-NsZ1K-yVf at F~w
ze%3v`{%!ClUUud7W|owAsdx`}bkOAI=lky`B?1T7x(=SavuPf;myl3?^%@y>Nb7Mx
zR6Nhk%BpLKV#0D;JTanf6&Dw;<Tdm4m8G%jE*ojPF2mOO_ePsk%zzf2qnyDx^7ifV
zCR>N7-Sow;C6 at UQW7&6+^7CiCVY-yid?O<x;QV-P4|0*4mp6DPJOAVX_PhJF^ZOl8
zMk#JBs^Wos=gd#63A(X16uvoMfoHIfIe0b*O?{^iZT%{FA8MHMYgMP2LC)g`15TM4
zs_?>A=sVjBJF&U`UxiR306_?H{CsTyM9bL+BBw4~p!<);^)kIU64ij`5FDd6q4zPP
zw8(H{sS$O0LP8KWz~klJ3+}*>%<J|AzyXB1uHXT)?pmp0GjQuE5-?hx1hCsBkwM{4
zZON1o4VHp0&)8IU99)!QrUbS9XjK73 at nNE}+{+I?`zY<wZ5r}x`&F6;)w$=6sBk+X
z-!N_gaHKIZ6N&)F+n6TG7`1Wg_4>NEsfRxQT_jAffaSA;PlYbOSrqfcbG|gCxSASe
z%=|?54GV-fLU-?C5(NCv$G6!n`Ga-si?o2-CKa9*fUjb#F*w<P#zxen+`ZlOB|8gC
z4fdLWU*kBGMz5+BcEHv17X%Y=mQUroch_h8@>Av?I*2hZzEJu at _I`6{A-%^A>45*v
z=E&*u=i41w#gjBacVM&PpuHq`jR4YJnTr<hyuF5f4&(f%hzL@=^LE`E78b at o*Pe6n
z)%wwye{D+!HfDHud4V}Wur;|gn0<+Nrj>bDk9w^D%9)hkDTtS&(9VyqFwAM}?2K=*
z2NKOqAL*T*9GY{AC!uAcuTLKs)9=7Lv?|So;*yf}s(lXjwzfALER|hc#E>U~`yM)b
zpG!f39mr$(v|D|dt&2;%=y0#C<W{L>hR(?s-OBJMxQ<=Xa&Jgj*mCHmL=q)W0;xn&
z>oeO}UBPEA^OTb=GM&T2?NJ=aKX3-Fk<r0yJt0uG;viD3t*uMlGqum2&CF9slEUFR
z*#8P!o*U+<J2{Ea*mR^wcXZeBmS~(gLx!UQ02lmx1?p9CrVaM#3fvqRLDtQ=y%@er
zA(l5ILW_k21zkXY;CtdQqkKzZ>=*>Gky~2&!JDP?Zs47cj*hkt4r*?0$9~Mrbai&#
zynmlDIxY_QLK;cQv17+J`0suIYN7I+s-T}TRAo<*VGJVUuGDk)@=|?v5tI5~bT3V9
zZf^35h)^#rEmbNj`}@nA at z%AIyb_Q|%3-4>S!kZ~m+SvLGNNSRR&}1(*rX*OXYtq7
zxfd!(6PkZs&`6<G`6fkvO`8r)o0Qw_ZjW8)?(N(MAXLhs&CMDd#Z`K=DU1(qi2J_d
zz`1Yz1y?Wp^G_&n1WHQEYT!f2?_s_En~MzlRy{H@(yhv0%BqNU7YGR0Iuq38sZ*y)
z*M|L&zG^(Xs7=!~0Zh(iZOAJVVkoAp=;fNuO>?)Mu`9w$9#SeQyo#q|Ve#o+01wBh
zkXyfd->ZX|y?y)k;%q_3lfuG#jLfYNlaKBWC0!VayM3FSpc*e_E?&F^>JUVTk9SB&
zyy)%F2L`1P9^UT5`lqJ>l(%IWB`x>xX6Q6?%7vGKjXKY)133VaA;}tZwlkUdZNK6>
zQ<|wG-{RwG0Uz*7ji<h$L0NzS6H_honL)kLM-r)I#J4+Q(F|#DaPZ6QY_)UesPIbJ
z$S4YQfD9WoDG?!77t6^V{O>kRQ<FT at p1*5kWZ~`N at C~{um5iHRwv!z+<mAbnUz3kk
zU5x|fcGWoOpu`D~24P|QK+IxEgolU6Zm-g?FdUIhOu at L=-X|QLdgqRcTZy$Hut#hL
zIecTy<Wp}i9_G?0u5)n&tXq)Vbf~~9M#RQ8BEXItjr_j<B3xBoE3*rS*VSUbzK0K)
zak$Pptgqtqf>4Zl at Y}Z%I9 at b6n^N-`J>V)%Y@~5~@%*{(=%x6MI?Md@{CqmBcgoFe
z0>}k3CU|y{j35F+$WSpDalZGGrLbY)_u`HPI*CgVkjg?z5_bt`d3pKz##8mov5vKs
zl?i31GGpp8<JkIoHLT+md0;hM>7!d3)oK9i`F&yGw3!(vjS~$er55XV&*6nZtq?LA
zn$zy?l0<}iEG;ehhv09~)`R=^t19_{S$|{rzhT<ila-zQ)YYqp;e9)YhGN>=&j$?j
z^bi*rwq at PfM@)R~yYC!FJv9~8&&9*ZCV&u#(yw0&?zZq~f28c^Cs$cni5<JYwmewT
z`E}>e+3(}klE)+^L$K_jjz*~-P}Cd@)TWIUQ<O0EAw(tPMYxP3^60`s1~8KfuLX5~
z|H>EPk}r-hkmKDx7*RMi`)`#f2M}gU{|e=zpu-8Q$2#PL|D(<&F7@@@K<KV7jw;U%
zl+dIF{5}FEdgtz491S;yMDX?jA%RKdW?`F>fs7tIpZFAym``>^TY9f_&wpoSVPk9E
zt2xhJ-Vcz%@19m{m;(xazsiMniXnH|)Rwj7wKXrY%Ta1P0ZQ;ZAi5Va(;2u!;*(l!
z3Nl{3B732o)`%_-G!v8@!AVzFiTlC^wRL at 9%es>DJKZxsbzbRiJ*@o`;N>N)e){yr
z3iy)G)UnURM->%0;Dhhp>JweKaDih>ZE|-jz<)t9V95}6C<NvT#z{&;gO^3#(!X(R
zw6!&tm{@0b9sgu`tIaW-o{T+AaiR|FJBRn^9xRn<ZfRLrU2JyL1vsj#lm~gnqo6R<
zUB}&x^HR==z6(*>+33Jl3z+SV^(ehJ&H?7qIC3iZGJtWxuTEdTE{^yNBrl?2{K#*j
zNj(;3=FuQW#^EaWX?oFdMdxAp2cw?q>iaHUy7cPO5pL;f%Y3beBe1+cB*$kw_wC!a
zzH*9;tTOU{xd1#}A&Z`tmOS6uQb at qA^YX-K0$5d^r-2Wh**kdS#*OcDbEnMA(pnmS
zm7S))6&&niE+-}R`B3#-K$^nyA~#IC5&VCQ*hYexQvT*pSJ&7_AYx1paiTLH-=b#a
z?2}kyk<%(Fgv6RJPQo4t3JGBiFsw?(^XL1ltgO)QaT4<1q^>bcw4~MJ?hdXouLg0P
zXWuJ=$x6V08yXt=iX9JPBOZVQZJO&}PPmK`VBbkRp`B_^Wr$AZ|MKMv*6HKK)oyL2
zKeef;tOpO~-`MoTafBHe8SB9AVh*SI-w7ybRq?D~$0j(#@BH%p)|_C6R0J5t;RPFq
zyhdVR&3`bcH?uwOvCHV5_O=~3(Tdd=x`G*9w&wZz>-G?-=@=M at L`6mQbMzjz*;HvX
z{yj6ptbAR;P%VB{2zfNd1&hbogXvUAlXCEyj~P3T$oTyd4hs*bqoWJj+VaP%SCBto
zFewEY7J4Vcj(T3s>hY(OH==cxO4(Qe{R4z1K0clxQmr;lv at d_(f!B|WKqn(DNKR(`
z at 2U)9cK!OgOo$5if58r!Ccrw7+r{PO<<&g0GK+KP&UsJ7NH3PW|3d{V6l<lh8VZw^
z0s;aab8sXPzW?2mi+Wczv#WE*%=#9D_N~QmrFyK^#f9GdHP_tUzTMi}|9>u%g at mwb
z6tGD`Lc-g&;^oVT*jF|J;dMdn?d at bHC4aAIW2>uAqU*@X4Lm!^aA4uk<Tom6>g6H@
zc6N4=b4;>wa$UW>O=$p?*gOiXhYKh-Jp4nYKt*i&A0L0R$une-AiPeTyAsnsf1lYq
zjK`%5`$66#Z6hLt7zm_0&Y)x-zK3rNcTRzEeo2p)f(Yoy at bNJZiaT1B=ZAx9ZpOw^
z7nYQ)%=gX~gQCMisWRYbn1UXW8XsRR`VpR?^XUS^v6MC$ulYx3ukyyIV|IbC!P}7V
z{FUY9_&ayTr^oKhFmrG;p60v}k}Bu*q_UC?gcpv}!XyBP{<l}!MQ&Bm?e_Tg9w+mk
zvp1{b+*v>Uqfo95b4&8_XrDcMh9gy-Kkz7|o)5>-mx1eboH!@4Q<Y{6=P7`6aIT5?
z;Qs03$6GjWI5SQWi+C{~91m5oH?*dvrY$!Iw_CJc)+q=Hm}Mex+oRq_*->g}2IBF{
zmjajy&BaA9JUpD&?;w_>1wukCFi<JbjFXcS)(X at s@&}Ozym)u?s!vgHpDHZgbHGqM
z=EK3;TkF>gi;HmxI|Y-7T0npz)^$%#P4ON-egIzfF~_{)SD>}s-4h=-41ogY=vA<p
z$uI>*#>Au+t`zNNxr7>vyjW~zdvWm<W+<Aj-#pM2efu_Wv^OpM4_Xuyj8zmL<mZcH
z6KFJ5oeLL&CMT`^Ck~EO!^hSnJ}{LI#0kXnL@<rwBB(UU|5b8gBO9DjxYe<<)p4$E
zCRvMLS66qj^A|V*CX~y~m+n1u=n&p7y?y)krA{^F)qH(C1cVQR$4OHiX-%ONOk$Ff
zM6e=HpFUlzmZET>AR^SJNr`81w6-QnO--$|*h}GDaKig at Bq~4GvPa=1L#l&?kAvbx
zZCqSvecuf~6Y_F)7A-V9O5-{6)Au`kxyxdjapf>D&qIZ6DuaYU1n>pSY+?U@*Y<}G
zh0sb2{6CEMF{oMK1Z|>E+}+(5b&6L`9dLGb*5IS(DEFMd*;bsM&P5ZLT$WQ(!i13!
z$I>rPo~f$V7ED|wF2u>kKz;w1v=IQS$9yF`J6>wytaB={>VAF>hv#B{nZ-pHL{q<?
zJ(sbfZ~Xnv=KH1YwCzt2ta>r=``*2KyW at +aM60}8hI_3g#ufs1<WLxc at WxyUoO~?h
z(07s|-$wk|#YoJ^+|4cpds+Ye8wVHvhuoSwUi1<a`ofG4PMKeRV?4GtFfd>t$-2kb
z#HCjD(!XC_;j`gkHE7>+!i;9p#y0u-#<vWBag&Ey)+$c5KO7qyi~nF9ufZo at -PXvR
zsOhsd*Uw?c3(M!m9HWlOPvC)HzkV(CUg9AnB*aO^*{1K3nbs*NC{|W=oXfzrKykZ!
zcw|&mq%b|RjypgedX at m-w8Fj0=KV-&-0|Id!^md$V}Mye0f18_9lHwKK&7OlBy3WC
zkEzSJ%Krq`Q%#T2FNYpU)ZE%$d*76{vvn070S9vK4!ZQ5Gfa+4$#jr92t05RXKsEt
zoVL2^(kwaUBD46}$a#!JX?@SFTelX!*ymcb+&`K&^UhtC9os4axy7Id%ocb&X2m@}
zmy#Fo$CsS5jUFQ5+}zyDoE#)7BbbnOQZTXHeE^_V{o+OXKc+mBj6f0|ik<HGHg|xW
z7~Y?NOHlAONio=W9MEzua8?RN;dD##kimU61xX|&MPREQa!NL1PO0Zff@?;XeFFmn
zK?MKR6#%Q4UHM85U?EUJLgJ%p8fs%LHH`d80|HMKP6<3chdJ?ztTACExkIXbf{Fir
zt!=LKS&6 at QPjXC(n|lw4O^$88dIJ!-f6KPp+S+j3H6R<dt4b8QjE#LGBO at Pk2$m_s
zXGAdv=89dcSAYJ at 0GOmvSfwp?98jlFcdBg>eIL!aHA><=SJb<h<9wq9Of}5{hbQ1Q
zI?e;f$`_|;zkxYVly~np+MKb88 at Ttl+{gxX)pBBP=lF)fo7`L)zqR?@eCh9<+}I8d
z6YEC;mGJcKhv(U5$<MvL($@$o-o0Zd4<!V~P~CGrv3*G9QH7hvvyVq8RKUlv!i9Cz
zW=LxD_wV(UKU-T|^pW1uPK=AUqoP!S`rutoqUPAtl=PvA!+4bogaJ#;aXgfe5ax1N
zlnh|R{x7-g>%WNw<NnuNT;jccTh`c5i5077?x~<;V_>L{y7;#C9&lDzsmkw~Cr{oC
zSE5UrZ)|G1aOqO8Ra0!nlP83jNxHZwR&Oc3yt0yS*+34w>H^(4iJ4D5J!8G^w~u?y
z50V9)y!qm>FL|88FJ_N6e$3kksKC}}yVvgFt$<`4Bb&o(i@=A*u|U;1U?nBx$7_Y8
z- at kovfGbX71i%?}k;(Jla2>uPcldBP`HwaU_<x`TMaJg-rXq)ghg*HPb4b206+Ry<
zKc<lYbGFvYH_QQwnk?2a@;7M}b4duW2^pAZDN#v3BOTq?n6`5BAdoHNayR=TL%`9O
zJ$CSCK$^!7o8D;xE{JCTzse0SB_`Z6zFWV?#Xt`0{eDrL9i>7X9^y)rT+r~lt%Cqd
zy0Hb^32$lOjf;z=I2y4MlB~K%^8v*EyJ2}#Z7JapE@=YkU5^6409W%MyJ9cNN<>V|
zEiKK2iN5)k#CmUx&-|l#HU0j5M&y<Y at 76dl{peH_<5)&G2M$CeDl?v+9&3}?ozv$x
ze*!Z-{&$&kaMf(*ry15vV}JtzH?XXLtB+;X*wPV0JFYn9iE%SVIk&9wc83Eco*$~f
zg9ebkSg?T1VKj(D=i)^PAwq$WR!I;yc-3?EECtZs>l at 2gs`Tw4DpmK09g_>ZPHkSW
ziRDS#4ASimQmU_X%R|ZxZW{ePugUCMbV`LFj)meaUn^I<z;G7^k349Bw|_N511ov(
zV{dOb&IiA%tyohNj0H{6g at uwF|C$!qv_+?AyUMVEdJ{>_H6W-cDA+$VG_>`Qu+(s{
z;l#~3w&wdSUm5uL(xU3dP6BXIv&&A>{|i4u-Tpb-Ah@<x<;R_KHiv*T?j at l%2kSys
zsi(*I*ykII{9^T9-Ngqk10~K)5QIom&!3*koqRsmUqAw}MuvxnC(o*h8hG^9{mO;W
zQ7)WKGg`d|%@i;cVl9UKN}ppM{f`^8`1k*qnK7vy`1$iEkRVcOYCM()%Di#^&K;Gp
z8!Arw!|V1y6a)J$EGo(<DmsvplY^J&z^N>X4EKP^bsj2BO-{ZId(D01NF=!s?Krno
zN=$f>pRccXdd1XEzSoGw1!KpZ>i4U!>si%kp9QD$a9I-NTP)-=MJq(+Z+$%HZ|Se~
zr0a4FtJSBM`BD6yIu~Zsxt=_UxFwKING>HyG5B7|Kk32S>iKf93YWk)Q509MUY$=m
z;r6XHsch=$(@_8ojdSNt+1Uw#N(uvL&vYjbOGuy-78W+Qvm<wNb9<PaOfD+!)c^KS
zYip~yjZNd)Jd0H}^#q?|BXib>%$bV&E_bA?qQ&)!K1TLkzkXds#S`=j{r>&C-LAxr
zAJ35_8yFmX<jnUv5_U&TO>GoZ29hj1VgwzGV`PErs$49tsE|`yeEJ-p<elG|C1|mT
zh=`j>Ni2=A>``R<`O{UysV&1_6ysb5STqYzNDt+G-u8`&ouH+|vEe{L04OrAwA?>l
zoDDXGi5&-}9ICS}=i+=}%#(ie=FRb9t>>~>*4v at 9gLfr(+!P%hEoR+Ji&M<ut-mGM
zvs`)$tHCQP%bK5`9}yGN0EGGa!mtA0;lq0gAe2}*JJZO?Rgj<QV`OCf`9oQVN9Avo
z#Nzsrm3+V!GeB$Y`@0}Gl9L6ab<(r4qV69xx_ at R=3Vi=gM{8T#MusJ<VAGwucX2*A
z<}!k?1k#K3B{;TR at y72@-;Y0#z*CeHCk!;IMb0#eYy`TxN&p`3Aut3rbnI9#_^ueC
z=k&<^`}f5pB<Ou&^i41+=h)SD(nyn6uU^5j=H`l)d(7+;5)#U<v(9d$rK9FljOGkj
z-_zCAwYaiU$7JRoWctoGlAZOEvT7J*5J1uPpEa|5RSP>o2=dru4g1ys5ovuWt~%W!
zz;XqXOfjA6_U+r=_vU<`LR=LW7l+-dF0$`+nfr2lu-t=!n3&kMFOPe0aB%e7w|!n-
zUf$l`E^=}ChL}oTUtcc~IdJI1Agq&shzK^enVp>-{r;Ww^y$+wCr;GV)a;?AriS&{
z?{|RNY6s^^ccdwuHZ){?Ebns*Hg_zvQTwrcP4^4!P_>08Pk6 at sVpxygN=izKuXc2D
ziiPNFrZFpWX7!Es8ZB7to6-NrVm)|%WuA@;p+0lz(xtvHUuqj0_en`f<pJnyJ32TB
zFB^)TVqj$Ck&w8k at h={f0Nn=%;Flkqnwz5{AtB*#JQ;I_j~=&fpxCkQ!`;Ic4i3>4
zbzv5 at A!In+TWNABc6)t-1_xyJm3s(n{rS^C9e!lAEkzpJftg!dhk$7Kw0>3qlr^4F
zV6S6f@?o%L-ynP_2k%>2($~ggi0Yudm`Q;?gL8(r)|VcNSrE#4&I)*XN547{sdlTv
zYhnCzj!|udB<!oknKOF`B(HrWI)DCrUP(zf$kN3H_^n&25Vl3Nn4DWC8~octNL^h$
zOzd4ohD9jR{GEi-vv(?}sHjdE^ENVD{d at 1uom+o4*LWd9f^ph at DLjH5ivhGVuXa2Z
zJWd{?#-nU$$wN<1zqmN2D(}CZr^Vx`sd-!MV{BDbKxjg)QHj%0KFbn^{yS#H^>)Ck
z8L%Py{=?_Fz*v9({*8^D(-d!X=9($V$jC&n^H>#)_vKsRr1&>x?_;qy@<`y0!j~fZ
z!=CzDrw%YNWnSxkj=u&3<Mou$ilL*!sZhekrL at OvTC0<f+uLQ&JxELp1rOj75Fl-e
z<786rpYMJ_vyX-&y6VQ}6Ij&aZylb?VjJp}Z|AtLZNjTZL`EuGSa2sxUc1PDZD44K
z{oA*1mf@}X?-Ei{geqx&wWrDf2%OT=qWqgtlm6@(PW7n%@PRBgHrB<(<z{9k at 0Vi7
zmX3?!&uhUgLe%E(5Y2!3q>EU93CHKpGpf&uWITID4AzE?m;pCx<FK8_0?_i?rFzlg
zjZeq^FB{GOeJdjV;X{Ez8G$pku|NYYEuVZ+b1*U*kT}!tzjE@=fddD|e*c_Nxrpi%
zb6OdX8j at zd0l8G$-hO^oCy9RWogWi&f8eG#at=_XdSrw%cpufRprD|>u^uvP%C|UH
zheHSaSK6d4!|R5zHf(TcNZe=H8ZENutrkx$h?hLO9zkF)HDA84;{5B$4|g{n8O6!_
zu9C`m&Q|}L8=&Llq+w#(9nCS(6c-M@*f%Er{30JcnbS~tD4;W<KRhmu)mGyPs;W|a
zl$zT0#zd~Pv^3qM!qc|*4Z9NQg(z27*Qfk0O0aTZqTZ|D>EH(;(+8!UjQN8qR&56n
z?2a9q=ptpd(hB`LH8s(kz*|c(rlqAN=Fle=&7r`m;JbPg{*`S$EO6qP+VdJ3wA(v?
z%o&`Y>g$OdzZBB7D}3Gm7HsQA*XN)fZM6`H5##V}gVN4ca7YMIUtgbjXS(wDff7-F
ze}CN{&q$(s3P(rH(a?bWFiYCfZf}n4xVlPZ>48;a|95ykoCX66_~RzviY0=;r&s$i
z`H7b|lhw>-`a-d&aVf#MbLZR_hS>pXYes8=yWU=DVT}uV753;6o7$;U+~60zZ%kCx
z)QILs-pAo|t>om|p&=8E!FhPGwNMGY7FbWPLLC0VA0+p;MOSBM=gq{#M3<4;h6duf
zfs*i|qT}z??u#G4sH4Nh=lo!BMN?B#aD`{$W8~nzefw;O%8x?`3T0w`uywDp`QfG)
zvsJd|bu_si=SNRX&dQug5pUEtxto?2S6V8AHncWBwBKppV1<{qHy4D55hnE-5d~JM
z!<M}Y4Gj&Y*C+P1KROW%a5f5J%zNb<)o|d>AFxvSADTmW*Lm|M0iF+H<IT5kYr%HV
zy7VCuu$!j3{+czwp@`tK_5UUcAx74KtKxtN6tJ*TuX?VGRCv*3>E&SN8b;IU?iW_S
zX8Ub>UjLKRSW|xhnC|<X?Soh~Hq)Eim33JZ&9E+<X6k``p6 at hZ7LqSaGNq#ng013G
zQsO+9EExo(10qys)eXNw`(Ene;^HU#!Q%au-b*TWgkIA-ObiSJ1VG?I!~YRJdbEYF
zD=}IF-d<>#71rmE+s^h4>|R;nIi~`V;rH4?T%6)>Qh?0dq6^P}J+<(eKG;c3O{F5}
z?CMg6$M1Z8?x4PbLCwGalA{%Fk32nd*D&%y{trx61;(>Jl@=KP=n>`Bt5+#G71?;z
zqenw@^+X$*nylsr%Sb6HaTfT}M4S at Z#BAQfWc at 1t69IpIUY7U!Majy_iq?md2-Y_?
zGCXEJW62=6g`H!+s9BYoxj7fceuQ5kAqpoaCuUi90l-B{CZ-35QK7HGq at 9P+KWl9V
z=;-J)G&BeZ`uh7bz}a!A%mWdV(b?HnQDti_BG_{DOQ(^lnD5_hG1`|n4XFTZ0Q8yL
z+D3rFegeL5<xUAKwXo;+XmMhYsbr!O5RcRW#9Ve}mL{4sw6ui5CK0p#Bz_HCKQ$jM
z{@nl3(>7+Sp9c?L>!m;c(FT6<#<a>m5;m5?+4AU_=79A at EF4eI$%#P(?&_ig1csok
z3Nd(WbW|rxI at VzBEWFL%wlR23ARoBR1~K<EJY&dHhr@>t0};|>M2p$}O#vDqSd9lE
zQ2ffj46EO}f=6nN_ksi*9UBV*6=B&F%f6L&nO(yEvoIEq2?~+}vJ_QqMMqOQ43^T(
ze9oyWxY{A+zwUyq#8TzF-u%>m=&h-KXt2nhiE!`UjNII~e5)qQ->cJGIucpkmR45R
zx3_$-<^XFj?jF|07CDzMU&aYg92^`jUS8obF)@7^Z&_8;a561`$>R4eUcN(z5;y)p
zoSppkjgUa{_;Cv`xoCgG0ESNx*;h1(V5!vbz~e^`rmB$w;nPZyqJr0dlhE}%Dk^FT
z0Pq!AA2ujIckaM*$KlGdcz2WSlF(4%_w}K#vLU8h0QMkLd6JnqHa>pqLZ-UKWOD*A
z2qroCMl}eAQDy-f#GtbgVBlRz%*@R4>n!b_2kvYm+50s=B;+ZO9qR*|<ojzjQay&1
zn~zWJ-sT);X8HXR#&ap`?f_&7aDZODZ~jY>(Ba9?msnU>N?pI+UUd8U;~#4qo7(1P
zQdo=cgJs8X2=4T>0dKEMw0J*W?m#du^_aOIU&+qSel)t;-o|DhtX^qZSxs9T)%^VY
zqwEj*>~|hN7U6Svu<CX{fbNQ^DF?wxGc!&-J-xi*;`_<<ZEIB!I|BY6Q~MzyA>lI7
zNQI4h`1$v(EsUJj(u!Ubdjgthc5V(>e;k;0?eH)MkW(<ai3<%saLRB<NC=mV4D$=^
zG at QXmNk!FFWPcf((PGPTKnIz~LTW*uhi)KQF!)79QP{LT47xtixMXzSUI&yTPLKpT
z7{w&&YGTv`J6sFQH|C(Ux`sxCt>}q(4L<DH`1+qLS{4>EQBhGewYo5>iSX<nNw{wS
z&nQg+Blf~tXr(E#Pc+6%FfE7`t&KNOU>jYWm3ia0<j0R6J6~Sl_4f121ah{xv=jtr
zcKC9>#ABsEQ~v96;^O&Q7uNtnL$5)g65?}u5dHE?xraS73rl1vKP=Jj2vYC}tnhRB
z^V<U>8TRf4emb&YINg(V`3imR*=;maBnF at X2(}k8>X$RkA8d{E_35)@{i~k?7FYvB
z(q(<of|;4w9LcQ26>@;#dg>VQ3)tNW`2|>6zHJwe|F8b5gCvcUdOA8c8)FVm1a40S
zin)E$%CEEV=mhrPm187bXxBrotgQTGn3IPGfONN1&vlX%thlT6x)mj}7(p-zjXG$(
zvK-?$<V^EJ6_|MekBd_u;JIk&va_>8LL6af3!O(eWMyS{=^uDHe{}H;_ at MZ*+kZB>
zcnWR1R5ZSboEdcLdJ@{leXQ|bbj0H7bT`Nu_*~wLqkDi3$HV{AvUhT#2AKfF7EuZe
z#>Lh37I47``)X#Z*D-esL6!te at XZ<4p7{doPd60UG%(C3e`h?TNIUa@{M1fUBql(O
z{FliysW%k^pMUo3(BYBK$1s6}h6YP8>Dbt!^YX+PfFJ|zejlp9z66UkB%F&W$7b?i
zImd8lY=&w$Hfie0JZ;q&&9WtS^(Yq)(j}0(m<bJ(pn*EF=O$n_4pIPg6#CYlD&#4H
z$KY7758<?~ww4g!uCef1_u{bM0;tBUE1(VVI)sprkbodlJ*F09^yt^Ghg_|rKZt-A
zVY>-HdpcHD%pgUJ2?z+lw#fwr1-7Ei<{ovh)d>$D>ht?lOzwanY;JC*eeU at 2>6y?A
zrDu1(xgE92R+PIwPK4R1&CME=?6M3y8y$hL%Oqvr5|Wc|14kJ<tf*+J at o&J!@*#qg
zmoDwMv9U?yw_FKs$`-*o)3LhneD~>#A>rY5y}eA`&(GBX;9EdI1o;#x2+<HN4gd{1
z^b4C+J(>x4o(IlY+tfs|w6p}E!6?V3cN)TQ=Gg}#I5*96wvXFW<`z&zHn3MD(ZIkn
zGBblY6nsfD9o#-r`vH<AB^1MgE&l##4~AY1y5jNU$KO9bqX2jJnKW(Gt%h9=37LZr
zWnAG&1 at FYg^Drri?99L+QR7i2xnp`ISFT>Ah#Y|5p#A`CnIC>9G`ME>=_TUP*yJQ1
zpOc)dZ0*O7cg!N%n6uV)ncENP^B><F+C?`lN>Itl?(T(){GTvX1C^tXBq}d2$Dt&~
z#>S%n^FZA*fqr8Oa|3lsq`;9QWQZ0XKNvvx<HG|_BjvraHrv0yE&2F<IW4_l5$S>0
z|K$SUm_ at g#^GJkBoJZ7wCx}@!Qbw~Lzs>nA`xJJ$zeP;p0&#eKeLaGmI!hVnuzq}Y
z_BzNf(A~G$sc)9h8027wo??eR1X_>fS=_$020;vLPRRfA3*H|);Gl>BKm-aBjofd%
zKGOEn%LEV at dNnIk9r?~9ikOy?kdTm<msfpr&t9CgjkXpR7G at dV{3D^YRSP2M7#PBp
zJ7s6pzX5v<0$Ypc)8B4_)jQ2%8QpURFN)>7L~uqH=tRt4{D=SQix)32{|tkHzklVT
zoT4I!^Kj)J0%lg$|IQ3lcsxI#yjIoL&Mnr9%QrlwhLarweor4Bd>_cE8cw}CeDZv%
zEclC+Ad-B5XEgi#D=mybeRXEzA0``ttcB<j2L2MsPM$S^^IyQ4%y9&RtjFuVF<n$}
z)zv4l{`~OyM<>sprv`B%<~pen%`AQ^P0pNPxJu*u at H;<J3W~Z!q04yu3tSLE0k|M&
z?Xb8wnsU!MazLg0x+~FRbAObtMGjzvfvc+u>|5uvvrOJTJ~gNvU0AG&CO1~5O5LWY
zk+LidD_~Rvy#b3vK!B|mWMoja;lK_&<oWS~+{(%-f%j^3{+BvLG`;uN?|2^F)wEW}
zU*+Y|dd&|iTU$SkkG=srKSmV>6wKxNb!CW;A)J1Gezx}ZQEVsn8~uNLy>(PpTh}*y
zVK-7LE!`jpD6MpNNUMO-D5W%tN=kQkhafFs&`6hrpmcZVH`h7$``q96ykmU)amG06
z#kKd^bN*tkxz|)u!a;S8#Kha^EaR at Ot~=b^;y`fk>&$BiwS>@7)9BUAQVZhs^>t(t
z&;U_VK6`w=V#gT}KmQE%a^ABBj?`a#@#t!ZB=ze0`rG(;GFIKnGgq%(Efm46a at aTr
zxz~W+?GFuo2lfDUWX<;H#r#TQ#{r*;#=N7q3en(;m*b-7E85%JpL#ajLf1yXr2c1j
zZQ>3mXWeX93bHa}M{!xjAXPZ1UPSoGDz0zL_c38m&m%K4Go#$NWp}D~liBxM2I~}*
z=!I-F3LO3hRn?G8z^npGQ>^X7v-1iF0FY<Ry1BKv`PjnZDjy#on9;BO{Wrjzd at 3w_
z19neZSA|`j*mAVW8x58MPZi1Uz)^F!F0<TWwCIL|g9e44y?o;<px1Dj)lAptC}qe!
z^gG;+wjzEdeRBL$P<#9x7U4PpL0UJPa$9Sw2<T-%aI{lv`26_~_$Y#MFC3~}nm at UI
z;0$i_^OK;{>fc|!iH4uTKT2`fn1-mjfm=3QV%A7y&5=$CDgreFz5DP1%>+SC*3yzS
zn#=4Q5G%#Y;oraG^p86p34}yNX at b}QEj>`6-w1qy4x4~G1rZkp%=+t7(yu3X8y<c9
zzlwH7(vp%7(7Yxva%5!wnVIzHVnHlg43~<5ly9Xr+K}|jEe1xL=*?0=-P&LRAdfF2
zNB`&Ro7sT^a%UHpBv3u}k-0Y at H>l|8FMxIALyhb}a(~ggX7Ev-K at n<L>KAr)?}2UQ
zWAA^%ju^m$b$ADs5mog-T+1!KV5Dbc{0`xRrVhegCmS^SqgE3TzV$P{j`Vl$;-T#Z
z2s4yG2#*q6wh6&={lu0j)VfJT<U{Uuh?5~7Ck$(ahQ)#FizUZAwfhbr$q4G}keZGT
zwRd0wzPH^{VFQ^QiRQ4!@xBKnOM$yaiUUpaA}0^n+_c4 at eiV>0j|~h?LrOidHHTFN
z_y$T;ndEvR>&(N$gT at aOx*j=}fBAymFo8QH;A3u36 at T~;D_t%I)!xU^fH82Q?K9Zv
z_=AIk#v|qIrRIZ_{I6Rd$sMfzzzzxyPPS%q%##IWjZWbNcTd7+kIuh9 at 5*N?r64=R
zX)&a_cs<G|NRqnTeu)UZHiXW at 0OWrSEKS61?u9M5`#^GJdv(<vc=6X?j)v9k@@m&3
z-tXVPqj#QAe=+DK^nyZ01|9Mf at d8dDj6+uF?y4)$(9uz$YDt0qP1lU{bYUQUfZ`W6
zHX>lF-rxMZ%Ef1p%ldyskeC2CWw5ux!uFz~XS#cOfV5A8-G1%p`2EL^)5w5XTaQoP
zQpdvi3wjgH7f0fSjO-c<3kzgXB3PYU_g~yLx3H+c7SfljQJ{ym-BDvY2*1u84L0ZX
zW;#?L8la$zflxq at 4$K<~<_XA849E)fGv5mchcH(uVTfg<tlW<G&I8W>Kwa at scb{k5
z{>*)L9S1$;60-qHdU`MY&2Oe^j*h&~I^sWq!b?IkkJlQ>LCRr+0n6=|q{an$Qv++0
z^5SkoU3gd+jh~mx$+1gaW8(x+6ROaFZN^|wZ(DIOzWDdjk`e+?lf}jF09cyLzD3Ua
z1Ozlf!lNyW!~KN<5am8v)aLB!+Sb<AcZ at r4BMl5ff3`0t;ul3BFEu32(ZP;{wDcue
zu&LKP`_O2iPbDSj-OM|tz4V1Yg3vI at w{PEmLR%aHj`-jLQDGbn{sJv(J$5D}yS)4s
z at WMA>n)h`}V4%K#{fdEyfSk#{hlX_~K0J-s>Zkz37NpoXYJIbrzJiN~_XM5TX=O!W
z5`)K1m@^j7 at 1P8*0>HzjSM}9trP?!J_e(O^93<F4>(pHfV{gF!{=d1Y2)Qee(TFi|
zaNeLwAS^2eqphutCOoTT$p16q(RTJLRb8PpsyCv8^boSVeY-l at j29Ul{S+a~a2XqT
z{L?^!g+B~S7C}{^*K+NbhR#}8SfFYY;M at lf4-a4%3{u<yWCEXn{&x=!P+%B=rg)Q*
zLWL&(fzI(El`lbiLzcgJQI!FVR3xVfE^4=gv at R4ephP{NXx2)ec9}3`&_5;$f{lcf
zwEo)1zK6h?C`d7n!6X9YB{bo%r#p at s;ai%SnUQ6Mv(w^vi at rUNP4)+zke9D7p6mF;
z#2VZS#yvfqmYId+iK(d%@MAKnwwIa{qWUJc#n1x at 35m&)$FfJL;sYYyxG$Rk9is~o
z>`jVmzz8XM{`C24bxXr#=$xElvjO?Vz(=GTeH!`{ckWyVb$EjhU%dmhDuTxl1#3N2
znqp&P^YU=_C2FvM#P9$U)DbVh4x$DP(ECWlh07<J{v9sw<y<!n#Y?2VAcPTU1y_Rl
zDJNl17bfF>?SA)p`=6gtchR8>kS5XG7DziKf;YZ(=P^dB)SSUS(vKY{tSWM9q{0U6
z`JoPIcnm*AMo1u`>H{f6{>;pr*VWZ66fyE+k9zUHs-cmb at 87@o^7E at hEpxcIZ_#N}
ze0;=hG50(HoOtc#si~iBv{9?aB_~tH3c3)Xh7fcT at pKz8D(sv-cLk|ZP)RLR*}W<t
zw!Rk`wW8jpU&F(wl8GK~t)@=B?cy&Sxft&HER~G9Ywx`cVNK9qS^lm}<F-22-qP|N
zvKe(@qw^c}YhBS(2WVulH<mr4si}$D;}M|1vqgn|-y1CBsaWZpCMG7RuJr!>KgXT|
zuKB|AGR??VDkq1v0}TCVSF_d6Tzh at Ez5N>8)lCPMTLvWzaEpb7h2>5=tp6D^2!ayz
zR#66_ldPtvr-kF57#L9X5~{1KH#Ik-Y69q*>$h$-x#0L!djxO40O5}Qm+h~iEPC4?
ziC0+mnKKsZR4A+<MbVJ4r=H*EqpGU-(bkC2(I#HHd>o_w{S{`2C{%6$>y3g0jqZN6
zGJ1>46kpu$*0&`R^}Ntfe4X+a4<$!xK&WC+1s*6!4CLb9rKOKMqfQEwl7t3dhedw^
z4uG)&ZV&;|g6H)P?bQAH^aO4A at yRwsHi6W`Atdxhx)ZnzC<4(QES2X)JX-L>DvM|z
zQTl(^jLuapNdmA6OG;iqht2 at QQ3C0+XcfB`6cnufAnlM<udlC1)dtjv3^<6Yk#W3s
z-{DE2D0f{j7Z4XkuP&pP%^|8#N8(px7v7vcZ(US#iTBkqN#&~*e=n~Kz}Ns4co%U{
zr7KUX<U4?9-8BhT_ySb#K+hd at A;KD<nQaaJq?bXa7yd9P$pw^k+FnF6wUA%#5IP~b
ze1q8?Z2={*bD`1-TpT*a0aS<wxT0Sa?CNxB85t9>?&WSLf*>&7jlF#N5}q&?1|5wp
z%#mJQEwH$_h=c}63{LZ3ij!QA{TLXjs7|x9vqQk-uiBsub6VQBUg)G?0G|Ks>w;)`
zk3PJ#^u*9mb<Wunn;hNTNcqdta?zYw^;}PLEkQE?&D at L^bn$_|lIMOR>jNKq(h|ym
zR%(A|h0Xqc?zNVtrU#(Y*8m5fHU(0k;&&3dx1^*but0cx_-QTe?F|6t at 84OFl9Hls
zP?Sq(J6ujqu296#^v_gl1ZqY=j}G8 at 96-CXw^!8j8fhmmE80gJ9*)r`X^8Y`YHj_2
z=AN$pd59eU+}zv?d;5mYBoX#!KW`OV46CE?V7$%4L-^&@N&}Gobz<VnV6>2(Jw86Z
z1B=VY4^qVaf7N}|-O^H1{~$&Izb&62_z`tsfUrhhBrtRd9KL;|Y#Z7h_|Mz)zzToC
z1hX47JYM`&h^nZ3{Irw+=cs97u*k^E+xr_J&Fa>GehR1ruq at ek9)HE4cWQw-A<snZ
zA(O5Uab5uF^4`<a^VeT`Y7kPudkr-%2c4->cu2wQ?oI at B6`QC4*OcTYCZgl{e*_2}
zd!g;;-<5X2b_KtluybwI-_(B$W){|a!qes!9;haeOGd$^E5E;Z9_*}4!0H3PV%*V@
zRomO!1mb^63k(0=EW49_%IgA}fc9!dFUoKcV8zhT5Q8D&G{*c-(M(CQskEr*E$9=S
zR&5Yi(?6p;(4_L+-QGGzc6BRsRu`%-0CWEtAE&gx at 0S$|a{EV54<4#f0l?OQDp5Lr
zjpS=2yFp-ld^=m~<76*?e^ir2r;Pxg75>P=DbTAHe((T;lp2^9R852 at wz{+9SyEE+
z>n}T3aC|)K;k+z`=BsyhcMF4HM^h-#E(^#5YyBCHhR}k{%nJzC!AtmhoX4hwNML1U
z{eD*wZS?SoHbh=RD){*Lz-n={nm8&d>K2FL1z64Zz*DP_Y5ds3zWpy{k|P8L_r>4u
zXHlOrL<njPsHqXGs;=&WFW;Q&IS*ej!c#YCzq721&WA^w1B&d4k<oj!<M1<*9b)tb
zgh$chqWK{N1e(Bv`X}IT$8!p5#*329J$?EV)l}e$*Jp{ZU{DDFlKDEFeC&0;&{yl5
zb1KNMBD$8BzeXz;{SNqVz13ts3AEZ%@XYsL{K25R089L}vr}>LQzTJ^-NKELk&#sZ
zu_xWBQjnInm?S8J{y-=wOZ<_Smp_N|5Y&2jcsPoNk`iRFii?Zu^sY9{<3k|t?!Nv{
zso}rW6NEO=(D1yWQVVTb!ePJ>06=BwD<Bp;^72 at wI}!rKd~Yxq#rySXtgWrRc=__%
z`cx~LU0nF%iR{$m)Kn69Gvn^mi|z3Ot<CDW4bF#ajpg7r^7X12?dfyLJ{J~#3uRD2
zXEqdyWW9v%Mg;)$z(84lYi+$8LZ{FGcK^xKr<C at uW!qBI(*FKOiz1<(no1oR8JP&r
zV0B|d1d<i;B8ZMYC$vH8yvMP<{M#3v&ICx^)y>U1g#RzcML8`;9wF$UZ0+c1L=&38
zcOj1iR{vYB=I1*u`?d&N7gvFGNa#$!6(6ND*kf`nJ3BiwdzwFtx$KaFO0K&mk`2!g
zb=6-b;(Xw7?&=@lKM~Js&0GP&M|4XP(ExGiz`>Ra35C2=G&E=7k=7+X^g#KJLJ7D4
zAuecIRAqvw{@cn{D++mHn%7QEdAQLbgyL1J*cb~W(k)`*ANBQT!Mn2O>r^N|+pknk
zh$v_+E-88P?AeDPDhbpzv$#5m!UtK2+0LZXv9YlzBG9Dda>p%3q}ga>!s%HqmSu0e
z$(`|frCVCm7sjC*HZr1v&h7v~ft)HDmkxr%0L*=CV%{l31`qH`bh&wX?8cpmj5A6R
z{&I=O at hQ?d=5eSo44u?BUhhjOn#TX^*|XiL2z_+oFq&@$vJbLByw>&TItj at YJBQ~w
zy8P#IqF2hsTrXd}dezv}^l7~hfpsH=h1T7>chOeg?c1n7)dT$g@@VyeB@=}Nb4g)A
zXjW;frks*`uIRhL&(D8AT;S2khdNZumoB}o)hJ-y=};?M;o at SlpU6=@i3K|jPLUk|
z7;V|0V}|B>GJ+x_y&xk;svPi811v0mSCNs7sF)b4I3gbbw};Ml19(EkZhhB18gwKC
zD&xbapB4}?W#n&`p$!&Lz3)m&xHdO8AH%=DdGqGaIKG=`CV0`$=xlNuYiraC8^`zR
z>9 at CMZwWfz2Y&#hdxnmV&X0)el6I*%63fmmF6f9VG%*x at UT-fFqO%%>j(1O^I&4Bh
zLN{(P)*pz=58mDvYxbsN7&9iU)9$kmU&waU^Z)p<9>TS&!sa<FA!D%%YTP;5IXNaD
zDzm~C5fPbPSoqQ3e<Q(p{bF!%@MOHcy*)RojG{r|4Gj%w+wk9M=YKBssN1`qK&buw
z^T!>y=F#$0M25VDj9r>fV<UK?Rqzo=`JfX~dwM)Re!Pl~0i>p*Yvp1yB6&_TX(FM~
zcdT3O$b~Qte131v84ERLGn&w~D<aRH(bgO+^8v9uP*A|8rKJUeB|dfPlwrIlrJbXi
z5eqf-8Td%(zUJJvECdOo^Jsy-P at 3+<fW5BXKx=f-d<pnZG~|<;n;RWI2}o>iZjQF1
zK#8kH>UAwRvauBqndxT}0fC`mj at aaaqAy<D0hP46H<OS8 at O10$vn!`gpH@;;MXWAY
zb#i=w{s=wXx;oLihK8rj!8B;ULPOu>Ecnkynwn35^mc(fgTT?-H%Zeh4Gs79;-l^n
z(0X;(6gaaT{Z$$&q=E#Wn}q(3&-Moc2Mn=xuu?k#!Gey%AR!@H{edm?_2<tkXrOlU
zek~Xbbh-?zDWh|LnU($%&@7|X-)EGRl;F=5U)Qa^st(ARtunPH%NAB!xbWxq at VA!v
z6Xn#}zq%_{pnXt_8`>2 at 9RRLJ`zUTv-919zmEo0;Cx=vy88MgXJ2d(kcq_RsnQ?x7
z?M#!Kz#~G3VNAAdZDb#5wz`|-#(4c@<c#5ZSZs1tw|C}M)g-*TcIq0r4NUzW<zXHG
z00o;l+6 at WJ-zY$BrBVtC0rtxy4EFQ{xqv*swx`@c_R?%PGsZ|59oRxTr4|$@1dt1<
z_h;Wp54RjCe<%~qoCE+J^x?xfLL#DWkZj;;<w9d)neAV1f<2oD$q?4~dT*jCzjsD0
zuK-duPRLDf at p?{ZNQemEtL0RH<&g038(fwnk<WfcJ_J*lmYkeSq-|+uBo=N{_k~q^
za9%npoK{FE0nC{M0FR-uv82As76C+*?Bb_v>gnU_W`lh!8k9kbAXrV{DfhLfu#(2i
zJNqj?BT;M10e3PvGm{ZIm5C-b^Mab+n(gwy1b+T}*J89P%G$=J{#u^2-P)~C?y>LR
zg`;`x7_n|!SXk`Cb0+XfsWb@*326qc3i{~*ct2J1vxcfM2hvfdUt;_4jp~U=NMyk_
z<CBuI3|RMB46nHsXz8vb<XDa!?C-1Z%InT1IIWGJh1?0VU;OnDTtq4eOhvF>Sr0wV
zeXk at YCRTy}G1;8$+Dte(^n}Er=jDxF-`J?XR;!nt9oui~=36u$k(o))!OotwwZE-9
zqq}|tx8M2p>}6vR=Y|#*a?#PzXWVV=?3xD$qs0A5q)>0)(r9%IK!!XI50A9YWi&L_
z3((BCHJqhSzI96aWecOmy7JUU5l5+u+Cf)c-8)JuDx-J}Yio|d;o(*t*3`JMm5~Y&
zA|fJuTwKbXO4SlqyCBBgm~1j~wXg;8k3?NyX2(qQd;-fzbu|xmASVpX&1GdGSk>sS
z*`;M0eU#)GkYv-Vx=lxyAw9Fvc5keSlU!_mv!z7~z62`~fQd08EiX at wmzOs*D2T#2
zJU*WFP?yk7&7%L6U3D%FLx$cL>#0kG#Kamz0fmL^u!to;U*T8h>jM%+MMt-g8t4W^
zMM=RUrqS2e at 9yoT27=ii_hkbJ(guCr8!1Ye1p2ZE>{)Y9PY7rsNezt%$DL(45asFR
z<@XCkoS&=3;c;<s&F!vDl#u!UQr#ob=ou(=-xPfAFhgbKG_Py;{CN-jzUIb8PfkwG
zrje1 at 8>~8#Acds08@}w$rUayxKBUI=?i7a%zwP4U($d<>bhqu~u;b)9CYr;@N4PaS
z!%y{IUS)}zSuo=O4ws9*bo?=!*gmmovj6L&eJ&RLnpp7mk`fZ0w6c*804t<pYIC!P
z$YAYl|9pL;6(B~?yEP#dwD7H^Y4}{`<sILJ=$8`>UF9z)ROfGnYRJ{)FFN>>@~O`k
zDi4n3D65gusE^^15Bp`YZRPIuUH>XIG(SHPAVf<=Mb$<2Ec+)28nh6YPV_o at z3<IU
zfD5f6!;4r}<So*Yl7S|A>&vydxy&4Q?j&{y9>{jB5~|4T?d{=1v;%b|!E)eXOuLfL
z$|VTK`I8A`TW{6<`R4KdS^=n4MI9ZIshJrSpriV_x=t|qSr8OPH7XkO8|w?`43 at 60
zF33p<OjBc{MBQvE6Dw;vBu1wn;oT;PLt9(hOwcm;*RLxkdDL+CsYhjIP4$9+(R%n0
z%VMNFNUOwDY=3{B^VPDBw459PA9Ivfg;1TAy<O|oOP44?ayag;CMnP4FmsiV*a=E7
z-1X0%wwv!|IQ0NrxJpZrs+_7S{@%esNPa#mm(^&j>+aw4(lRo5 at Qbnz*dMvpR8&-O
z+RiGTdhp at Hhn3Ac{><z*@$oxBl1izlgt(p at y?P&!=J!Z7sIoyfzNp^-L>MT^Pg?qx
z7AgbftvOj+nb=1*-C%NYF<DancGYgnkQ}zlBRoUpFR5JirXtX9FZ%qM3UKDsD6QbH
z!NIOiPkzh}Th*rNezEtC4%E{38XQZ`-A|i*|8RME*>-zTv!$ga`S8%W^3 at 92(ec5`
z<lnzZ?~DoPf+Xp=xx=rK+<ySNg6bm)Pm%1~U_e?(`>v)SDia_Oh#@TWF)3<Xr3u!R
z7^@lzAW}wV=3IL%R*i8b+atBDDp^r5vZ6wVtEYe0xM=^~+E`spv|H#`GB;;N4-;5E
zPQWRgpP&B%#=*&{l`{dra(#1?8e?c?CgX`so<2~ZucEGgBaYu8cz<g_<1b%#`^x8M
zZIl`s8WjU69Cm5H89x3CANl$D#ZT)m`S|!GgIH^6YfA%F90XniD{=YvmsWnHIedy!
zn9NT0;US^Zh<23f60?ouGzTOh`Y15H>2k!Eu3Sk)pKfnYpJ)uYykq?1$B&%20FWF;
zMoTNTC$s_r)zW1S&1;J5*4;pZ7ZXtEU=%@tgeq(FJ#*0*XeA1=o-YCUj)$2aDq(v6
z{=Jf=CF|7mw6d`=z1>{*drV+Nghtc~y`!$%>Ys<FzOu2Ulwm6aq(y>`yoiA?=^83A
zn}Y;GU^4>N*cf<{({a=I?|YqV*RE}XhgX!8^lH)dXcZnRd1Y>%`=X<iyFgnl>9>4p
z<;;D`sn{4s6}5!VoGs=G3^LBn{4NJeWrok5i9_m7PQ19k at _%{(CMFPJ;@jziFo?r+
zLbA0U9<_uq3D>#<lVK%7{)9$71F$&b4m^|8Sv18!)hCx at Y0(#*nr-*4Xy`vOBZ+na
z9kD%BXmH+cw(|`}5iI&_oj0DUj!y2Tk(5f!A!uC{W#!<&Pjlnr%1kdMeZ^$$j$SKI
zI)36<FU6YQKYD6cCpA=Zw*x2u$NiF>vyoeYw2 at tCq>F;H`Q7xI-p|t-A{aW8oRoW3
zqZg<qq$kG5g#iH4z<O%=`^|n?siDGfx*qOcO87`F6p!fy4aU|}g+J(a`usI&0|NsH
z^xy?2zpR*>Y<ivfx?fsGoJ|JuNnLh+zp~w$@6%fN8l<79cpb}X%=|U@^Qc-40G5`1
zDlV=t_p)Y*>~YkJU`h6+yu1?@NR3u0ue3!}jwK^M)Fip=o-{~9tD>SJN^o2IyAuHn
zJUp6<pC&2)BP3ppl)vmPG1Exv(9kq}$}{hZWJF(E>+j!j;(oW1LXEPrLqqZRm%prJ
zg4jvT$dG{8RrAGR>)pG>EHtK3pr=%3`TGI}Ng81hkudI+c`%zFfH;gno2(Se8r3CB
z_i`<-hZd4aNlB#w{XK7w&FY)-d8j+VT0q;MQnT3+wlmvB)eUR{D&KKq`eU!jyupc%
ztE=Ge@|Paqc4bhPj`3tu(B$GG^T_Q!AU-B0CZao!F$i3W?H1H8VPhBA5_9bRiz4-3
zzq(_*l%I#H>*&Z{ym--evI%D&9Gvmt?%L%1ytJdEqp)UaVPR5|h-Wv*ig#NYo}y at C
z0<2TAzT1&n`do#Mw?xpE#?*$jLyKiZs9$1sK>=IOCKeXf8TV*T6JZ7h1`f82%{>8-
zYDg- at +_O^WT)Z?d{Y8bSa6m53KL8iP0np;{<HXJ#g5T3Vr75jtZd8<%k|H7=*U0$k
z-BwLy!@v&Y2LTse!01+3r+t5(l$3PF9ZkJ_Kara+A7P(2wX&k{yV5SlHlC*O2hh-a
zLw8&dL)4gJi(&4xUbPanu|Cft_4wrB(#dG>e;EXVPV18wv2Ft#TuNZD)^!6YJ(b9@
z6Raj?Y at E5uVSD=YX}AGg-5O_zgGZ#d0Z^%_-_`7nU$%bn;`;JSpttB!t1<bGcmX^<
z(I{2*S^>9Xo&1H^XumY>taYAzej`gZ&HY%TG)X7v!AqAfBPFr<*OMHfa8zU at qBTRZ
zJk==a(nhD1Y6&qhvF~FMEZX82<NeJ!PN!}2Qx8s#R at _>1MSgs^dJ!Ywur8Wp)T$z$
z9itvMl+7g-w%`?A#U at 7WcA{aFy<nC3k(cuR{YdyM(q at y8T*IXnIcvvTEiG>1;^(le
z^uKFqt4=-2#cm8q9`NW3vLo+rJgCq13n?~x9#6Bul<-4oaEwU>M1uXlB7yE7k+8}w
zrA-;s=|jNO)YkUV at 8d@)ZS5GXV&jKY;(nmj9zPjsah2}C#uz<&*8MZ;ZWaiT)U-4j
zXV<lHvFJNbPL-Gs1{ar<pg-^Z^2&<WVcjsQoJ`P}-j*zUnFOTDhfjXQ_be~Cu`x2@
z2II>FY-?~VL70Gm5={PZl{gp_-c?dw)%i?!KKsQ)%ionj;ELzw=YzqAqqyFj at 5{L~
zoR__oySM at N{5pmMGJbzLt*VOObg+m9E3qd-L8Ed^dh|t^^^|m0LIkrugcevf6IfRe
zm`Zbpp;?nl(o;JOI9<&lbQj%!UKH}-bXL!O^blhVkJs8IfZgsoIk`fPdR{UxQAlYi
zH+;V`g2=Kl*AS;vl|p!!A26LALMO}vJJR!~y+wrqd6Dew?&<<o^^_h<&h`Ng+S1a}
zBEmY^d67}nesjI$=g-tU?Xuv20BjRu<A0pW@!_8B#<W}-i-;!UaJX6=U#bi3!f`|z
zOHe|B0)or!rQwh_Z?Fh&-BJWnnH$h|OMCnFt*Jp!nn3!j^(2qIp&Def)6O!Dlao_-
ze}8(GN=Co5F7<*vWpEx at V8%x#Ikl<mN^7dtqN%~L73FiSRs+GM<;6={1a^YlYHK7k
z3&lmDG at 1qVhME~}$9tZfTwGJq&ond!2M0H265N74JWivF1^>!GPcLbt at 0HTt_;_wF
zJvH@$RS@^Rdl^7TO(agj!l$kha-d;jNW7)X$4N&;MOo|rLsSqCa~mB(8vDoLTsOo@
zb?puo`*JiUA&0LLbEneEMRyh%wR+7iT_6Gf4E}@4v^(`8%cD<;-7)9Z^D}a{Qb54N
zCqiOkhzSV^gJ3;ClYI0i;Yse~4k}}AQn^>4UmG9IZNaE9_~$79_ at KIKzgtLSbiCg8
zA_iP|ez!@v+wzGdHLj1hcM=c`QoP2CzX}2N7^&q?C4$F=1t$k96M!b2MzHBmca=<1
zo4H*MT3mW+^!qMFa9%z`MJ<H_v359bakfo8sEzj>8M0BrWMpLDw;oIJ{RBq<-b+ef
zKf&3>g at Kkf{d-l8wwoFb5s@@xC^$Q+sPq}@v_}=$=EDr13q*<31lcrh_!fm$-T(UY
zr)>Hh$R~NstS7l^pczkE#|i}nMQT=7mLjgYT**M_ at BG~S*jC4pR^{|C!U)cJTj$WG
zmKI7(@24js?`sNXU#1)LRXQKM>djP|oR~;?F;b+b(5z)2`a469P>t2BrBWP)CWgRY
zF1hUBG^V1ml9A-Oon2+ISsvVc7b*wM at ynDRv_`N=fhPcuM9;$VvAerlQbxuX?hExE
zhmqa$*<^~apX;Up7fK-~C%3;;Hs&Q+<ERk5uh`$BQml9P-aWd+dCzqgW at bj`^d`Hb
zt%U&qv)8K+mhUfkz-<l{n}poClnQDEGC at 9lKa5?ouD-qtlDfI0Lm_jex~ggx!j5&r
zFD*MO*Of`Cn{|xLdEhw*<AH3{!1#J`b&JZBjgC<vy$7>Pdy#UgfLaJ#rSVit7y>~|
zrikaYVx!g at peOUfrtokEfBm|SskC2;UM+Jy+Q(Vl9$Y&**uhiWHWCDuron)%(|F>#
zE*L=9dNeuT0xGPPI3Sswv547ZVA-z3KePEjS#*8JO)b6h_`^^x6#$ah@$qqLPL9m<
z^mOr)A0LQh`tuiWIfA{%N=%oFkv at vgHL9V_C!W%D)m>my>0e8o-Fj?iXmNV==xLua
z??Y`um%f&kElS5S8waD%IAS`vxsL4U2<8F8DK#|- at 8ADaXXd5268>a*NV670j43Zk
zJ&4Ea0=1WOSalfi;lnvnUfWDr^YHNSGw#XB$?s}ev`hU!TmXrt{4_of*71xxGYbn-
zZbI>E_cH8TVVse^zP>r;&OndhdbO at BJC4!^Rg-n)PwRay9T>DlbDePqO`kntg-0uU
z8xy7%O3q{bPCS52clwDCDJdyX`SqpWUo=d=CloKGOHw6)=sDxAtfIp7^BxV&dkliI
z939ua`#L&0l87%#&7}Z_BOJdeQUZ%&0&8n)Wm_N1_WiD~@kBn$+TNZP1EESq#AnZp
zelv^Judv>wby~Zy-t(y5tA%~`%$W;4+17D-uA?ByZD-m^lQzC=itx4E*KbNLFOQT@
z5R`)?qyysa?&(PhqLS$DN|7*fye`Zfrh8r8E{T5WuGo&-Ua6>_9@*aRuBz5DxF5Y|
zPSU<&;oFhPtwB at 2f*J=uKhLt6X@^9Ug0D?I*jZuXBo}Kcikr+jZ)h?>pQ97TcCTe*
zY(?vw(O`a&;*5bxU%L{Qnt$tw<d+*cmT`d368`@FTiI>8fWwUe<QaO^j;I#K!^_*9
z;C8IJRH9~~GGp**(>X6U_X39J{(aTGXXEQH(=swH-;0($3H4VY^Snk%>EhxNefPP)
z&RC8=b0K}@Qugq3k8 at XpwG_Z#sO=sAHBS0#4{a9p2An;67JRdYkuGfqcTfg%lhe_I
z&Fwb_*g}Vj(-ss9+Y?9mL;pd*_<!sa+?gux)twHtg3L^>@$vE5)xSTAZReENcXsHo
zv9V1%6W at GNfNai^2q4>Rrc3Aq=QamE*YxfsWeX-Z64kyj<w17U8r at t!C))L1G^zef
zmCJo|$mL>}1N%z5g&PDnZb<7s*PQR{W_NUQirA}CN)b0Y9{c=U0dDFVc&@_kjmx*k
zxbm|1wuh~zq@%~J)m$pF^RH4`IQ6SmD1)Lj+HlkWyNZ9~hB!z&Dhx*eu0F>l^GA<v
z{l6+BT4k1UsPr(?ksx!VH at 9&BgqWv@`}ALkq*fjkSeb9}E}94yM_SYEXzM0c;%GG(
zRTm at nW|=N~+qnrUcJ?5|NWTgVE;X{WZiVt2NesHohbc`JrRSQZ-eA$Xe%pn5>C>k>
zr$$RZPtush=^bk^v9kJOI>F*G#l4zaQtB2$C%x<+?To$OPd&8luF=<Zp|c>%kdLDT
zvChH4fvXoyVO32h=R$85R!?1^mZJ{P<M;=hNN7ZagoA?vx=nOmC01gI*?{_PJmtU^
z4I9PGLjSy&xHwW)&A)y{Ub%YJq$QLg*)P9kJ|DnyACS>_zBkLz)KvQG at R?Z6&(FFF
z4I1hB__P<VpPgDL12+yq9^W}i%4Hh58pRRcoVj?{s5ESJs^$x0!bJA)hZW&)!I^pb
zq4;b$jeH#nH#awwJ%C$C at z1VJG;-C9fL_MKXw^9L;HZ(%eA&EwQN}+hO(p^t6U}87
zeuvyhAmDuH(q%>g3!_w*d}IFZY3-Pfh=D%d<n*C$KHL95<vI3wFGz--=Ls(2m6esq
zst=V~5KB?xrm?0R3Zh!i<fLJeY?xrbjG#v9&xjMH%}S-gjuUAWa(_Cpg7Z04X2k9*
zH*;=i1{Y6RSk;>4M{u-Hg(=xv#&Q<5%M8vq7aqj#{(2 at WkTs1*Vuy6=msFg+R<>G2
zQ&UC)0s<va!K4C?p@)ZuEwcJ*Q`+keAn0acWeG;-dkgf1E?v5Wx(k(_JfTPfF>}wY
zg)Ukyn$sK83Ag#Kc587kL`PRQ3;4unbVNNkSG}f)&AV}EC^}!ST6%A9&vvSXz_cg*
zDpq1a&AyKpYTy#ZWP&iUo$aJ>+?ppKz{l@|hx_isECeabhNJqHgQ8;Ki}lu)7Tn`_
z2!m(Oo}F<=Z+l9YvEu}Vgy3Oiu0OF$E{n?_F0;xIblDFCguku<Fkn1h=WRP)cQFfl
zqDqD;DZ5PKCO-akOm|;j9>?<hxG at M{WEkdO1qTOTczw{X+0d(IZqCf2SM^BpCpulC
zE$XhPu&{9GD3;3b=GN9l_YtczEIh~BkWSyeeM2z`){WD8Qu0(w%@%l#&Fpz)0DfM#
z<Erl);R7Lb3W$|y&CJYBJ<!wB+dNpQ^%WHUkjrQm(&VV5b8OYObTU7LOLLE&h|5&e
zX?rmV!WHS=-(Oz!Aiwiq(Eq7PVN*AmiY-qO315*H2r(c}9L)O0M#{oKLF<-_QJ at hY
zgBG@`c$c-BZbg>m==$QyD;AHmvIUUah}V at 1&S!)AOwW!>3DxzVs<;ti6o=#*1ejXQ
zgRFc6L~>CaV&!kpqRK4o;(PFtd!)@=cHi9Gq-S7|%HKMg&naL;ud~wDsr9idN1>8a
zem<+q$<dCrTUlj!+EJ*Y?XX%|2hbs^s99TE8-qqaQ?*1DxV<AE$J<QuN^MFqe7|r~
zF6Sx{XQEsTx00hHFDi~18`C4_vTQ-Apcz<JLqo}N_ihSg{+esS^Z(%S&{S*0MNElt
zS913mA<E?hCk<?J!PLaW+ggQB&%&?zSP8I9n^Y<lLOZ#%BzO1O&qUB&DlcAq5~hmk
zr%@v%o;uJrw7AJM5%0}!*OsGEfNK8)<5s^uQOL#fSA5r=V#*N#N<_8RV)H=`;8*h~
z2edD9p4|Dt=Vxun5TF<Dbus0%j%`J&_g%Sl$DLL_mVRSXCqFeIMy63imsoUY%rsjk
zYu%xu$bljIun1sY#O-j+^XJ$=PY)I5yU0-YG~H5_^8O*(!i(m2C>kjoP<4_UTx#ha
z)+F4i${Ctz at s}a8;}!H8-F{fRGwPI at l=N_>Jub)I(F9+h9p6GCLQrtJBSAr4ULKX4
zK&|R1G^^N~CM)O?{i{|%%oBl$y>km6UldpYZ56$F^F~-#{kO5MTAqrXo#UQap3uo*
z2sn=H+{@Bd0dah<a53v!TLi%*_De%*Pb$0d^H#2r at k=^7I`$@uUH0+w!%q(?8x2Vp
z3 at vGSs$3V&qTSTqo-aINZ*Pz423ccDla;RE08{hw@)VCW^s5~<KA;oR^7Hf0xbwbR
zmi=PClwF8}uXzL)E&KQH-_0M`LV;ht at PR=JOiH4-O2!|F4!>g4tGdorlRdA&##X?w
zGZwlgE3THH##PO40Y2*}O$q5&AiJV%9*vux$x96Esp;w6;8n83d~c+tq=<kEN(LQ1
z at N3<EWmFD>%HX1+_+(`An5O=I^~i><t{{veL>o at fDkTu`_Qf9h{lmBXE)Um3rHuF$
zD9rp?oNpB7_v5r!o^N$BZxs#)JYMjss}qeDbm28IGYhG%7FZhmobhwEcqO^W#n{*w
zRaCPAUocxmX>8l26eXO5rPq)s_~>A(OZ0QQGsfBso)F%%)QNg$Y%=*T8W$yeWR at GO
zcv<ACKDfDF8kduuQn&0hZV<!1l}e^y+l%#eIEIzxW*A&wn#0C)*K}J911qci{AN+C
z*QwK|(_CHmH%*YRjEas9E-kf>I+svW;~HfmGx|u4gG0&Qo_lU?t_zS;4sG<!Ei7EV
zd5+w5mjWvh#IEGWnaSo5qhr0`-Rap`Nm0 at G`1msq=3}JDg<NGH49a`Y_kXVJ`2G9$
zMGT8>rRuJ>)WnajuEL?18x0K&XA+0X)FJ>{#m)M2ar#RpCAySs7*l at Yoq)og`~Lni
zf<rd_npAgp3<dj32!G{Tu%1jh81{7e<*`}?)IeEcH5Tt)7T59lc}F^Wmmu%67(tcU
zsgd`WZ?+s(r~0$H{LBIITJu>1{FAJloJpM*4n7Xfn*2t%7%&bgpS|CQD}=&BLqomg
zFAbvkUa^WvNVIa1t0gwoo2%Wd{})e}hDxX~#jjRIFKwVVp0E<H0l*OYGB2@<MNqXC
zs)Q96bM8-uC?L at d+M6C6diqM=8O_p=<gEur_$(TIVKMX5EFH(kCyT!wbmWFlgjO}u
zy4`=Oe3*^CtD`zCJn&PjDUxENQFhvI2e+9YpCa_<4;sUGMv-!kG1;UG3`AU2Zf?7A
z8*=h??Q^l)grgmeDW0C6cLrx%wl&R`O8WK83J%l1<<69ddgJzHsnB^Zy{XEy8C5MY
z3r+4RWVT>8sZmi@#tmas^9g6x3~n9lGG7_1rMq|U9-~V76#`bBNb!$EVhZv6IGnV0
z?H-x|Oj??4- at d(3R#r|ut|>9?YvIZtS(mi*Clf#w%5l1%GaB2xPx^&$96EIMYBl at M
z=hoHLP3~8MSMGf|6YihNKrXhFDB!d``Fhllhps5~tY+&ld}WBg|20zn*LNBp(fl3L
zt#*4QRh;I=ngc==2LpNGn1Ar{B-XmhX~(}kUO)zowqZAD2#kv(4P({SiuDKSCUNcB
zHBN(uD?`;z_kh>fItXSt#kQ>(%8qZCEY$myKL_C7`19>Fl1^LmeKLoS1u&?Jzq2xS
zbA^skgibzI63j*#t8S%Yy$?bCfky75AV7S|<Jy3LfI8RIk(TwtQWLe0 at 88!exR^V1
zP1}YIm++W5JM=xIu1?RjO=}sLT3nB at D$0-pnTfDH-y$Oy-1)^_b_~y*JAt@(cnlzK
z<>UFKfOu&B4D{J}xz^NZYiV_lxt*}sEVg79FjVSErT%gntvH(Y at t481Yaz!b<G)^@
z%I~yoUJ*cLB^Xjh#?nHbo}P|m=IH3?o{x6)b||xn1fk1(aHh4EiYhS!Bcl(aTK1-l
z=9LHhZ}^-D1Bo?THy15f=;-L?JCi(A6CQ3F2pj+X{vH(*(3W&llUEHuL944~v&tx7
zkA$-j!-2>xBbFWNt`EiGaaB{OR8~=08SL%d%r at 6nSNHDO&q#_jinaCP1&ur{F$kbw
z9`WrTTMGkO`n9g84FNoV3gnxzt!;7V6ZWh4q@;2nXwdQ^GzQ4%A$L1`2y*2u{~5(@
zF1230YG4M11IDMa;+DhseK#UXm)N`+^(Ts2t%lV_hXiM;#skX)X;M+NBqSv4mYH>@
zUfkQ?m;PN75pFD2Pbuc32v?b>QxWV=j%|fUc=y>AEFoGBmy_c-Sy@>yC^H$>4 at cK7
z7j2%M+KLFHY*3LTPN8h1G&LnmT9g_p+IBOdvvBiZ48?&T&eyAtLBY{5F){Ii@~R<!
z7ph&8t(<nrVz_iE<H$&E!^iR-qvK%#^-y5Kqy_f0*buKf<0Rd-TWVA)i<t{wtxYw)
z05=DdSjWc at du{oHR;^d<d49N+E8ZKZf*>GnYmE8RNWBF3bj;~4u{i)<A%SS9Th9yP
z#BK1#_3ICKd3lxZB_{~E$zw9j2aAAaZrzf^%pUG-ppD8?4>A-IXepD6j9PIqj%(xZ
z>s}WLO+kdMWwg{57P4arog18;oAb7kHu|WcudkoI!qA2eeZpNe=nJtLt5LGHE_%IQ
z$f-b!do?l8Xnfieo80(lf6LI!ET^~OVU}{*1^2qJNl{Y4y}h4I8u`~T*@{VT>y`)!
z3IAzim^!}FaH(Zuro&{g$Y`97DuP8jyx}9U#7#E6-c0X7O|>lL_n6L!#sDM7gv_78
zy$pdzM9D2imAO~*4)xeX^=h1@;KLa{1Wc$3Ut%#F4sHN{v`{~A(&XoKKYjbcPDc(a
zMf6CNU7t-5Trds>yju&`4U>6;byt5*2Z`X+RBEkCTLvI0<-PT(=Eg>e)pe=p4QeZ%
zKs<*YHN8370)y6dr<r?KrH9Fye0gW&)V_!_aFDhR-3Zy{Lj}c_2)2T)9A0WVx*p)b
z=I-ubxU+if7~HqKGv9I>UG_H_!k9GtF^XlDzpY~*hYQ$u`Mlt%r;!n0{sJh2wx_`_
zHoIKGF?kxaGcvUmW8~&zqgT*MRF-;_24JU9NO1K^xM;WyrHoc!NtMgaGAY)sk^3XP
z-B%hbW4gE&3kfn(QaC~sd<*XcrOMs%AEOQV!Ozc|fBf)>jf=|y2sC{B__ll_613ku
z&i4S;p<?gTO2t>*;c7*_z638jL~)GTU@@}{8vV~4zNXxJcM&fWFlKybVQx+dcq$82
z%H;C0{KJP2@$v9BWSkvznW(G7$GG#dySln?3h)`O$|-1oNW0-y>cu%Qox7H$T&X;@
zU~TWya&N50B?Y3?=RC(pav{C!9RenGKO}Kx*@oYGcwF3FOY~eU(i{8u at uSJk@^6mD
z+}!=><b^O%pX!;oVXOREZg=YWU?KRYS-?VP+<{TT3b2~ir|&u}bR{1qfu;hlRh#-!
zb0R!lfJSCWNj#yFIy<%>e^A`<<X8ahKxV6E3BSKYkPN}ny*|}iZ-h#UM*a#z90F#V
zu~AV{My=sAuU_rw1jGp}Z6b8nS{bdj83~X)UY7 at uGablhd3if4cPgHGD7e2UxJ)O8
z-QXNn;tdup#TPH`(9qGTSX*<PJ%3(hW4i4krY(|vB4VexCx2zQY(>Gur+H>+xQx?k
zl%G+%l$r3>txk8W)3XpS!k+b<s?N^*U^tZ$1YJ;Z9n1`}nZ at 8S`8C2+Ll>vCRaNo9
z_`G)~XWFfHIZ&^X!!h8n9!}3JeeJsZC1;(Bl~1hL!f4K>-IsrhLCt!i;VRlDM4Jkb
zSj}y1ACr at l@$H?vqz99dlF-KIDOcCd!O?1``nk=a60<XhmAuJS6F2d at Zc>|k`Tw9=
zuH&bwBYn;0?y*E^qdLg_CQzrScLOX8hunj9dF6ctS=rRiPDRt<QWmU4fN$Tj=HThD
zE=$$TU45xy3Zdp<vY)=ib{Mr51=&gf$^Nnpz5%A%f%TCV;{QKr!;C*<FflQ)o$IDS
z_GxQrSi?D_Mk0VrXo+p*eI}T-Gl!1`QA_7I!Mv!#V&JA=mx>oGqI3wY93>tep0u1?
zT4*Rf2RC;*gGzcQNVyAv&Yxw5#EWOAUG&_1?qiFJMny%9XSAUch5Pe$nU1e@=UP%a
z@=ELF90HqtMEV at 0+~M4uD)1$zap!GbyZK(W)7GCG*p~FwlYN}FGg@(Q;&^O6U=&-!
zSvC&+d8Fjzj?%ir)Ibzr9opVVuS77#q at H`dAW$b;m4QgWkrVaDJ$3En1bYOzww}&x
zXfm5=9mQ%!b6zmv|MUWgFbNI#!sQRux(V^zzaI!Vp_Zxm=2=^`7?6vyTMc7K342U<
zIIT*`{e$hdk*O>`cgyx8i>&y43=Itf0|KzoKS$DZt~<>dn_1;URlCZ=u+t7{Om at +!
zEc=3}&ITzFl#-HCxBWn`m&Yo5h)X(X?ooefWbqi109USDX&M;NX!)Er2)H at hnItmK
z9aK`nwKP^+`(WYqeC+wpZp$9$u`>;ugU`6{Z!g7|4XcSpg%o>xA>R&iZgOQMp<qH1
z&zG`bufif?%1*N*iGM2PdWfA7i<Q`sqNYYt__>qi`J*o~$reX82ir?jw^s<-yvp6=
z37mm-zyJ8r+|dz$3N-*s<7$-d<o8-hNl4J9YO&b>Yl*&{_E3>@zvD~Sm8d3_ezez6
z?|ah(<k{rnqAUR3bs-#5%;2wIowHpj^#xO0FBIa<-WS^mnD`e+W9W^7OL<e7qy$~|
zMZX3R+$|#49)bup*RcNwy5t*`Lo+k!Pr1rI4*U=nMt~Kpq1m)VrqjRhC3jCsRC9~V
zNWME<H0X%g$MUpf^0mE*1h-=x>=t4H$KvnmUcq5iTgFPqojpA%Pkwx8Cb%19zQmx}
z<DY!cOQSSZoW96 at MO4(%((=r0{Seg+#-7jN^&r)5V;tm1f4T~aTJT|t0dz at rsDf*&
zdIkpQNJ&YDK0W!twnqOnxRg7c%b}G^=_ET<N-z054ruT72n*0sU at R5P2J*vr#A!99
zr7tawRH*Rhrvj4OF8*TwMxNfrsR5Bm%47ZR+*Kk`fKB8`<zl$grdlIps=w#(TQiP1
zZq7c$pvuGe06q_`NWbHkTss<*q2u68fSL at 0DNkmX_U_;`!89Mhm8P~fxx-KXNg|#t
zhk*`Kh`+xE6h2ShjNGEQps6ve_vtYfR^mlG+C$qb4vlQ|X2GTZB204QZIh><o46x#
zkMt>vw(n&=C44#4PP+DkU~zHLutq{I)F~Q*KmaRII*id<KZK@@Ddl(OEI*w$_avRe
zu!lp?wsX!?*Y0O6k*RW_wH11MZ3_-lG)C<oLrUZ*%y`1I32AJ{@MbI;dP5xjW;-zt
z+V6GgSG|6SY5Mi+(P8v#Gc}Lh{<Qji-*Z=uY9F0Bdv<EK=Ap+q91N;u_T}rw at V;7R
z9_<I$ldWCGy0^P~9Ye(b`i1|RVeN%DUOOL*A`qjuS6$rD- at kux4I-wTq(8 at b8J`><
zjf<xkyN^`bEu`){E2YqKa43&&gN_M-pYoRsrk%D+Lt)#?zhh>4lGl6gqeC;W65pIY
ze{4Mh|HZ{b-?NHW9a9NL%C7tN&v`)AGl%=sLJjZm2??dh$jA&03`9WNsTLVZS-*Ul
zl$dxsFETPNEc1}-?%fm+b<#&QBQ1T4!>Sds5g8(q{R?PA(`3FkOEq1V5LJXhM0a+j
zNc1^=Dxv0ylTn)BTxdH(iOLeKz4Yxh0tJ=tj5%ETu~;aR>Dr%F7gAejHR+M7OwLRe
zvv6{Tnwy(vDkceYT8`W=d1S<L;1FCw(v;)RWL4S1^m}q^u;_5qN17PuA{{smFKF_A
z;|uec34|B2##v8$X$K1Q|JhbS5SOF=u|H3nt`JYVtCt5`dVl28r#mm{_5vwHTq{hX
zT~4%0Os9w<a<}eCp;t>#>muA)tIHLWr%JZJDKnOes#q6zLerQ{etn|As!-SroQiD<
zzG28>CFH#O05jX2CZoK>_PzFt<H>TcvGz%<alM*?LI?lO)G<#V0UjQD`yCbaf<Aw~
ziyH81b1MY^_7MTWj}o#X6f$A}Iq^>Tn$3<K)CEPf3=G~NO-vvbIA8pEkgfSSefM>m
zOa$%u^XIMY><EO4*%oG`RX*TOS#emn4i{&OM{k6=xY@~SbO&!%%*W=d?p4P26yl57
z{%+&_D|nAe-0${-b?G4r3W_t5u6OMJi#B2dOy2lT9okAqZVvTfcxoNweD;giv8_q(
zS?bi#89)ChrM}*uBDQxL1K0XLcq4DWx1<sw8rBnC)Obf!Lc;gF5IL`{<nqL%TKaPq
z-+_Nyt{q0r+Fb>~I5dL<0kh-EI}_X<Q+3qhd5sizW}z=f<D^)@wTCsx{vY7!!k9^U
ztb?wTai5f0D=J+e7jy<=0g{5Fv1#3Ys)b;Fw3-j~RkTHOnSk344GyNXeI)CVmY%K*
zVzi>X+)KkGN558(MXQ*8%%W=%a3@~CNgOm9rGS9Inv1q(0Y%X$vC1=TouPx}Z0C=y
zbC32YWD-6X8)hkF6(MFUG&qm46!do1tCdmPg?^TbfLDLMVx79Jr_|luy&mCW+MD_C
zYe3jdtj-`R3D+UkmDl_x`)}8!6 at p8do;oKMvoks}Kedlm6sP~vMq_qzGsyYqYPOE}
zne*pE^72^V^1E06{-Eh!vkBIY#+U2n=H$e$rIm at +5dWey=- at BOw;ocp^sP$LsYZ=Q
z$b_nk+Cpqw4R6XSU*qBAR_<gJaYp8veNjXFe}M>3p_}#g*8t-FJgt&sR6LKlXGJd{
z5ckx;0FhR$>Q~XLt0JD)q+%DtwEyZP37<Xn04QQ}Z{xYNv~)7W`bqIy3H#wv3o4AE
zm6Za<7}j^(B6*jDgyaGSAZqefKw26NY9Xs}JCO&ur<$k5bqduuMMcqW-YM5BysKMV
zX-!R%(lRnBu#Ar2P#m0`R6ajzOWlv}HmBP!Y|(;Xu0IH8)zz9OU=$;WzI>B)a(@pH
zr?<#RW~j>HF4!8WY=fBrunhI5503X2*nB>IOaoX&ZVK!6>C>kjj}IN<8%;nJlHPye
zo}7}x0JkaVdMH^_Q!~za|K+%-_sW+Se@|iEe*XOVS+~0lhIlt`N(c%H_Ey at Nfp1P5
zulMbN)t~FlQX#shsjd#IMTb#M7QMK>xtTiCk<hcVGR6S8M<(PdJ32Z#e(^13MOe;Y
zSr#ZM1uCUz10c`1Q>RX40=u1Y&sNWqMsS+;BvZ<oHv*#W_Mb+dn3b{G3|Is7matl$
zmRh>_%k3qHKALbgefh5e8JCyGY7=mAap@^2-k{&~dTsoSyL#@UAe7MLDs_N{OFulZ
zIh}W3Ag9mA$G5*T=4QjsCLBD?->h|QQmrxUp6*nb<@&z4rt8XK`qeInJe7=VAm(Wr
z5he;)Wq1D<6X*7~7C;Q#d(SoKc5);EiigI)z<^OL8^5BWVya^3iEE}4S?Fd3Aj7!)
z3cEqWBvD3APEK7pGc)sKw&#@L^v@{Noeu;MzHroCm)vCLYB_A=pxbXlFQZvx+Q)e6
zLA?(_3RsaISU?6Yt}tO?;n at zM6KcxAc5h~rhyCl^JUmgD&O(F6RKc6w at j}PH)plM*
zh!XG5h>q5}DIf<0no(-&I^A-D3LD6CRJT4pJ|0&q6Wfyb)0oH)s29y`5#5qs$zq{N
zBUAkdXGgXuoC|Gv^<*f7L`2;Dc#B<_|EscU0|A5bJD^$_CXM`5Pi%5!aq)BYAmGru
z`yP~wOH1{Q;ptbxrbi~#w`a=aa!mdF{EQp at NtIU$uFWwPm_0%}XJByfadCwU3k!+j
z{y|a|=A!+H0HImXdz*iMoR?Nm$S5zrk9Gq=n&4snDPJ|>=w{@o<%nm<$0etxQlrfJ
z`t`v};+N}_%{(<7gM(2mp$t?n!(yuFb86CO9chA|t~fAdpRg5_TPL^d8NecrcXZ5+
z7VuBDJ^u%CQ(pZ2zUO%VGczD=w?D`JzY4F3TN(Id?r?7-G%`|Bzt%Mk64;ln7>uA!
zwx?f{HUi+2!wm%hOQ-k*8VlFJVtv>BFl>_hx5-}AT2D!@2^NJr1sls~wMvg#=K(RD
zaYx;H5CH(Fs2|hzXy2}=q=e>A<E6I2=Arxq9vj-eY&Dtb-AHqk8Hi0pL_`=h@=1=i
z2J{h-B>i|w#0tsSo*<+E*K<?<rfGk!6smNi<Bkq?E5fW$qZVcomEMc=MGUL&fz>Q8
zws4bD-6M)nVO|d|>EPiL>&@tWh<0HGogeXgH-G!~hE=bMe&Z$NIDP!PGwwP%I<AAy
zC!2!CmnthO at rj8Kdh8t<C!~6AQY+|(P&S;G{vVf_>G&ymzjkuTN^y&W^6uS`!=s}t
z>!}u0pakcX*xTF7KSswOM at vJaWNb{23O$&>l9H0wG0Y}<giGRL0ViRi#Q()&0U;CO
z|6*|j$3|uHbOVEFRb1|rs-;nAr;9>t8KH^-ztk6Zj1aAf?<F-UH at IWm6(QSMMehSc
z0X_NmpJQTT>UMZ}ct8`LfDp)VAs29 at 9;J1}&6#DRq!b?CIsM>wrQN}lO<P-=+k9^P
zohQQiGY^MW7~1TE0IGa<bSrI5;tzcPc|l-(bv3ZO8Z-{tv~xV#FX?Smr3T$W(vg%(
zM?S4C6}0)>$Y at K3F3`)<^E}pVH#fI5=X(v-8%GDPcpcWMK>!9o^k+hp^=2rL^SU0E
ztHx0bjV}xo^cEUOWVPy04xUTDo at EKp=c#Df=<ny3T46JzTwy&Wsh|+xxV<R#yggRQ
zx6gN-DZ_?A%Ypx33?!AIytZc9j9P at GloZY@bE*B$pFbaDkpylH(PalPjN24h$pA&5
z&b+87LbZIIC?r`M2XBUR&Uj0fTHh|<&99o1j^;JCOkinUi5n}XSaJy4-f5QCT~E)}
zO~DWIH?r_;-jB=8%{AVf?ZVx8F64Z;8+3#D(StPUu;lOWFZX0BrOY;m&|NwqvFq^Q
zaaji4)YR1Eh3O1sP_Z640mLx0v^;K|@gn;gIQ%MD?AbF~x!C(b;7pA{d~nq2M?b=M
zXzQ@}!dx(;MmCkfymx%WR|3v(V4}rZ-casxQD~W{=#3jUCYP7vd*;ID`D8%+nVu{U
znuwgldKr(a^wa6jd^nil{LvUdJ~vWfGdVXW1q7Tn=5`Wy<?7XoyDx3Zq(W#>zyJF7
zb~?z>hX*yO)$OgyZ0&jGfz$ELUtX=CmK)UXl+_YlwapG1obT1Lmykv at hibG<h_+ at t
zDwCOXCStY43uw<Dsj=&NOFCWcNw-a#A22d;Ws^FPaq<hMk?969N#nhSgWmbE=2x$F
z+?oYNURy=CuvHBfl4FVv$Ea at rU$@}r(g;rTXh(A9ac$Lji#Ov;K$WXLW^-*q!h0oX
z>k-dVLR{{aaq>wkIojEceNMR(No^5AvovY1H~Fs^BnlcIlH?&xQP0!DC68yGyI&DE
z5!+&$d?I7!>guXm^?EJgT+OuTkhMs+>)w(En$tQrH;04Hf+56d)?1cJUNVsjw`QTH
zp3DdvD>@-0CRRZOy~f})-GuOP at u`-uRvT*pmDt%?W7H!=z@`_AiV+QdgwihenGGlj
z2?-H)Mf2Kazg89o$TAr!HnA4^$y2t_3oDAEz}D6lz2n944+N=Sw^*xM{*trLC66;e
zi=(+88>W<?+*q>!=4Z`NT3cJ2*KspbmFcwaBTY^3jxG?Bu7fqtJN_T8{sW%N|9b<+
zg{&kiUfC<7%*;@Bc8DT-l)Va(><AG_GRqca7K)Hv(Xc{DqB2S-d(`io-k<OP@&EmL
zyg!dmgO~fhpXXfXy3X@?pNnH at 6_(O!QsT>jL;-ypLu<dmtlDY+pd4X|L$1Pk{Hb(Z
z9I~Rie@(5Q?_YEaB}v-6O{cA`UG3b<bL7Yo9>@TBDDUN4(|)3mfirXOA1XC!j!0F+
zU&<HG7Mo4$>$$egXK+Ov(rVr-DW(34fCqO*LD|bD*Szg_E2qv_r>+J*EWOZnw1c;-
zp-V7#^InXcmt;}7i4SQs8F9#1GbJTuE*<ZQn^94__a8Ee0q%{>wQ9#V8QJ$%*?<!P
z4=4l at a1)4=9`p0%Wnc5B-rVV#naNE}$5Ahhjg1LPOG^)}#d;2!J}?n_;qmR;x1%9S
zM2wuA1`tvAfk`FD#!})dk=C%V-G87c at oH*nJ}|4|nf?AfIVp+meVIinN_Sb$v72)s
zW3Bi=<q;Vs at o4bT1iaJ^3R1)c9xAsg3^7nq2wE4faqf+POT~sI6r>~vHN(D*`j8#@
zz3lS0|JQ5fF*Y{-chuLSR(NRti)WYSb@=)Dm*+=iqe3`4FZyh)df=J-6d~hl*irQR
zK8S}R&;=#B=+O~L$=T*`l7TV{S*TrE>m$Y$*2nNTVN_(~c6F%dSJ#uHCRM0z1-Gh{
zLjNG?DdIR5c=O&p4(zw%@dK~%hNO+BbHY!2>VWw^IQ at _b?$#Kf2hYBPKse{h==4cc
z--_qp!6;;WfU@@il>R-B3Y+Y)j~_$lvb1z`cqWDfa&#;>lUIFyZHDA|_{|M)Lru+4
z at Gm#FR4P1Xpx&2Ca{2v**e|-KeFd2f$1U!BQUE=`#}`@eJF80cOR=KPXu3ftPiIK*
z?CKj4kyj&3$twyT!<E>dd!I*5XKruL<1tblD=Z>{J8Jd*3nxv>Eschk-XGidZM6=t
zo6dkbQ_|^V!q)SEr3n0Yc$7$nQT|uhJnypF*Zl_23t#YUXlOVWm|a~ZI5+mS*6PmC
z at 42jZUT1$nDKW3RSh&8Gk(P%0p6p5?mgPP?Zss=&-G6>~EO7MbQSEKdqn|%n?S!uV
zsLl;Nsa`$es;nGllHI-Ilc8HZ2k`&Q1t`+MgVTqPNJ|6fXrG6mK at fl0-5uxosmaQg
z;Yp#0!}${C?u+lv0p?GV&zu+GTv=W&4%yipDAZ-ffPfX}fAQ!Wa2|ZS58p6M5j2We
zUS7s%!eRmJi}ApLjqbH1;D6KRxi4kP4qdo-F%Jr;cH{FNAb-CPy^+?->`QrIlX||y
z;QU9Whd<Wds6YR at JlijH=+L2|cPZx}Sn*s4UMR?s2S)j1cuJ}%WarR at zokQK@L5LB
zLw3$7?os-W<em4NJ~<e|q_O(<mGRpL7ysNWYBcIeu8uwTP?KTDY<3S3yJCQ0rzBUf
z-mO~g%uBa>?~BGBx-U(qQJoYK5m9V>|D;KQ79Dn*fyVa(nhQ!u7|l(+G%jM;u{j$A
zl9*T-B+I+s3c41aS|bK at J6*eW?V1ey1?d^9wR(K(NA<xnp$C at uCd*#V<9H<UWpi^F
zHep*^BTmE<R1C6 at rx22b=Qo45R+9h?EFb1P+TPe)Iea-F;9sYIcsQ*{Fa4I=f!4G<
zJd6A7O{V7aomJLdIiEtdZor>Xr at okJ|EfRvChKT25x#K%KiDlsc&!t6OACLd+<jm(
z#dvo5PT{+=2Udjqem<kwySEUtDj|UectGOnj}K}<`Y$(mO}58Q`UuT_R)r*^^U9X-
zFi&`q<fnK~Dp=*$vh*v~#DXcNFY6ceMR{Wwl{_uS|A<}QWQ`f9h{tzfR~9~HL(0E=
z`}XI3@{5iKXSbL|Ez=<QU4Skg6&2^w&oDjn%xDBScG~6MYuoC4SFRWu%^UNqbQ@%x
zl+w0H3VzSCk^WtyAgiJL<<^Dk1TO)%qX+d598~|rK!-Xzg@%qUt#5W_rup5weI#Vm
zwDOE)UnVDsIXO9TtUCt>$B{;S3k;`S>FH4elHF#^4b;}tQ=jcClmNhYAFfoH at lLxI
z7at$_{J9bu3<E*P2UNz%aA=1DctXX_!Jr({<?E98u}Gh_;mR8PRJzK?kK58D9B`K5
z%=hnymw%^SgtZ{Yapa)z;S0F6sFRYC(&YUP;iO#8kM%A6{W#7gqCGDbo(d7s++e|n
zN;?wZNn!7 at mG-bx8A}2x76V!<9y=xhL?pM(Xis{2I!@8UsZ|pbwq_O<4x7ufID-W8
zD3T`e70#eWQ<sLzzi}qZ%-kFmJN;pzpFe*tuC8{1ZjYjgqKw~~4(ZtO84-|_WPs?>
zFf$|I^vLPA>^SNe7}nMB at Lf>yqH*Tg-?^cReUM7IZEsT(Ih7_yYTPJjXy9LDy#dN`
z(x!SWJr0?1cXx09%OrG4ea8EIQMt!R0}dp?2}lo3E80IL?azwPG7YjibMMCx?gZkj
z*no{+g*~|vLKprR-fM4Hzb$A)4E!J_F>z?~cejiF<wtiqPCqbd=}I}g1e&g_Y{=W(
znB(lo3Q&^DM~_tdi}b79hRTVr-#nk1m4!nMOD*f at K@)mkxPUX&(5-{!hD#EVHyreX
zHW!()UWU&@f^n$Ut5=jj5LKh^pIK+Qub46DkMq}YyywDrW3fpIy`_~EMnuiR6%nnu
z;eB1_BYze99H3)ixh`Z}Gyz#fkZ~tKu`qgcPCK{l)a9(~+nRTOCl|5c2tkNHg9)v)
zYLlw?$%Z#+o0D{9dQ*)dJ1pVh;Y4Ad+oHR-|D1j7$j!@30uuO)ue0lS;F2$2zixk>
zE(uHIQVi&h-WE7<8Xy6E6(AV7`S?iDzzL+~<%rrlU0&(;GC|9=efq at NocOAdUExwY
zRQ?M~lN!q)9R!>o(($yy)5kHxM5vj=UqGQufTy2x@~%7%$ha^!AjGB=k|p3QNrWR}
zaiBZR)e>$=NlB>K%-Cp at XROJ6_%IHD6J2v~_76PQ&tJYon%B6ZofuR~dU<vJLeTGM
z`{3>GuetW~BONR2VPCCMqV0|8?0}`o-RO>$t7b<$^{39}!r!db-uLe{=0|EIoO`%R
zts7ZV#cWPkTC#z3jh`QVlCPgBgA-L?32>TLT1H0Rg{e-$(TjYw?jIs7o_KOgOP`*p
zNP5UDX5BnA#LTOipng~0&#*}EfSkPiBvgZ9t7kGeAVf2f<1?rU9O!|=j#lSKAATkE
z&|jU-4h5b^@4qm<U!Jju at 4vA02R7LdxI)$2+e at t3S<tbeb{VZd;^V1tD(iW&ZpzsH
zj9=$O;>4^QD6U+&Qg%0IFGM6iL?pm6-X)WWxd~BHRTkmE0UY9l=3WrI7nWy?%*@Rv
zL;r4Z4h at -wyNo@RP*hZ0oO^%RvhLz;AWH3!Y(|ZNemJupBKqsH;K8O&p~RvUtD!Or
zW}r~5pwoLwj0JEonzY*>+R4I7iuwlVX+MoO<nR;ab@%YVdG^PytWFG+7~}mzycC|<
z$3;s#E%4Q7db>UCfnmpy>$uYzXWA#q)o`m;{L#4Z_pv(3llK&bhAw=2dpB+mvnUrI
zpSn?Qi8MS6oa(QeDvFOt1x+4s9Qnq^=a)&m5UJ`<JU`-qQ{SIs=>jj0M5WB0Jt!c6
zp(rXUIyr>!aq}iEtlhV*bv?q_{_DB9+<<zga~SNeUcHI~;BaE}+pMGBetusCW)8*5
zANOEhH%fIKsTLZnuo3eW)Z1&-6m;OyrAq<Z>k_W-9~|_Ra~0y at A;w9kD=SeYB|<o8
zaB1pAzFztf)#%Ow at 9%>m5CA+PB6}f7G*U&aB57iU&Vpd?1^Fy&`P6X6Fs=D%=uf9H
zr56`(rKiV&#y4n~yaWLFbG$JQa!%F8#-?F%b>aD6QR at rOaE=&<(}@#gIC$x>X*n9b
z;2fsBygVEo<mTpvbI(M~tCi!JL<~#Ks)!RSxQg8}&#(aI<2b9!^sz)#hpd+T&K`Ya
z^QphD&#>OtrS$A$GSHg)W#ss!-*R@@xw*wbk#YF;Iz3I&DWT_cG3iOSF788d%p&HI
zz!I?UM38ih at 3Eu2-k)(%YqZ<|3Kk>}&arK(D6nq|{R{DM>EC$j>37)@E`5AY>+4%z
zzt)@SNxK8w at WsaOX*yckFf<Vved6WZU$hr(cYui at cB@7gKLQA3%NY0H`SylY*s7ip
z@;(fN36vrt0Re%TxnCJBM~#L;3`Kcws~LC2AU6FRt3&^SAM^87F8%zVg4Jd!O_(6m
zdTwtQXa^khSSfHl&xl2t8JpjV+7<+Pd3iyTVp_nBC<xOW5DTi&?S$@cuaX6Z>wE+*
zEsUcv4oEO!be_mQ9&FeUa3QzN;gEXL`M`7M_+-4s<I60b at NjeQ!U at Oy{c2EMHUe{Z
zLHRVSb~eIk)xCKdP>arn2YCXVIKIt_^%$+qwR(05ueBnp?A%-cZ*+_yFg93(Z}-X7
z#rZ{eBH&iZ_Oi6IU)W$|V7LmE%^pgNmX=oYO>Q-Dzd4hE(zC`hJ!TwDK^qaV3<CD=
z?j4)?^jryY`01ab7WhAWZ=@k;;}-c- at y@iYn3>erQU9M~bOvMP7EeS=%gQ=JK|F-t
zyLV630TgcYT?K#d*}j7~s(GlwMiXS#>({S)fVEb;3&+{={@+=E5A*;H>ToG`&d$X>
z at 6Mc_8!k$@6CF(sRDW at K*0?EzSzQ05>4 at v%M9VtxK?nkVU%`0FLy#lL|01ql-BbE5
z+vN!qK=ify{P}aSX~n&s&*CC7cCDoATU%3*Jli=&`%|*A;^DSiKYnD#VcF7dO#FIj
zM7+GbxoQsPpfdO4(90)YoC%yt37|xA3f0c$idrJaF?^J{YUlJEKw%O?5=3NYb3&>{
zMeSk6p`%J7!Yv^ae-0JB!ollC`6uw;M%u9}GQg-h;_AeUIs$NHtc&cu0xRWzzun13
zdkDn<S{wlm<kP6`;(0j*g|-hL4#1z7<rRxsaK=B##?Ldo`GQvUQV0U@$xC}Z?s$cS
zgrv!Mus}ed at ml%sDz-+TXCRE9Ui?Z0fd>$xr~f8Xb{`J*jAIrHXuV2|gX*A&!0I4`
z*I$~y4%g_zahe2@@&#Zia*x+(5+un21}7lAqG;lT%pkZxn at 3_8Ly1w1ZX-CbT#dm=
z$ndPknu1wny~f*NnQx}0QGgzn@*E=!4i4fZ`kESP_)fn5B}o9(h4Euc1wRm2z7JIh
zf?)pf<3|t3EJPJRkSX~xGJ31$5K6I-t9#gK99-w&rZM*5&o2&7NR%-_TDM#!pMs|)
zX!*1f*S6C0qdX3I<WvgzvU!N%_1ErH^x%Ocu<OZDfouIm%gMP9ejyHN#4$x>X0n3L
z2#0*6WM}C=MM+7C7Famf9sqx4b{6Mm|Ih58`WyG<c%e~v6v!@dnf9_sH3LYYcTaAc
zef<6$h!6Z=w5bQpJuyH3W=fyXe at YSO0avsBhXwH4FenZ`#_AYw{37V1-o`8b at dOEC
z8XB4)pG+ZrSJy+^J3BaW9VuyTZ7oi4XJcp2t*8)#s`q9G$O}Q%lTFm}X&5>#0l5(e
zHY+b&)GBJzM6IBpfM<S9gU?^QvbijT5ebDX7kGtgbiTw9NlC29z)y7=4)wzlqxD(&
z!v~dLzMKmP47C5+mD2P6eMf<IviRj+=NJ#4K_Dg8&yu6d&CSK3;1$Zv1P7hlz@>n_
zO#&ChzeK-2_#+Q+I{1Jsml_W at x9{2IPk^%Hr0hc7RBs?=Sn+L|g7BE6?LR+M1+)tT
zIz}9#cIGDNxCnSKk=L$Ku~Rv?iX1z}jtm$sQNm+HD!0u}PaJ(%;In9%m>}8pbdjcM
zYik3;iR{T07fHE!GYqgS2XwdG#V;`s4+nzQ7kA-1bATF#`6u?G!7J2M)##3DOLuoP
zmm&@g4aGTjMMXt8>7R^-Ee7<(h35(mRJ0r%G_ta?INMi7M#k*XIm(eIUShsx at m0jj
zKwtFI#CHi8+-tA2YeT#DkQp8a^J_S7G40Zn_Q>u?{B8AVP+HsD+m#-pGB|zj%$YOD
zoU`uUts{vLdjHIS0S9^qT)xbG{5VHKLIMt(1;|9^9#&l~1&n+O3Y+Q;7U_wvUva)0
zgqdse=qM{bh$mj3B59&Zj<T|{N;&rsjgOD3M&~5~@S*uODH+*y^0s~K@{A`=oB)1s
zk(QY`y2fn?N7e at h2BH}#&gMK-r~`3tj02sqFFM>5tbns}edXdMicZrrFwnu1zV{UE
zH`L){G&RZJKRB(He&qbkjEw8(cMF$&Q1^~H)FL}4C!91s0Fs$}(-n88b5Go0!P*8Q
zjTb(!qy{~Bp?m~@+W)86q9s&tdNW+rwbay9TtTBMdkuW9j~(C81Mw<^2ZgS?I0#hG
zv^)z|^e7T1pqJ=u2cLy53TQUN&g6hoT9uWFP_{x|wY0VpUnSeynxk?R7al69-uy?v
zCQnaKoSO@!dkM1WcK}ak_3LmFa*XwY!a_Toem?r7qK8|elZwIs$2HG)>Z(X|5{#b$
z6^4h}+S^M9kcy6?=C$szK(#*q+1WxnDuW(J9R^TH at JIAKf`924u5wt~TyX>5xJ2qv
zL<`R(3Lp?qap1H+J3G4|wJ2>7s8}F{Rij&^!jh92(C`dkBv-9h3$XeH+&LWwM?$T~
zsQuhPNwIm&Ato{Fco43X>>SBBpl=V8a5Kaa$eio&^*Edy2dV?^MbRXFHBLEnS{%w9
z2rF_SA)$X*ZQjw|P6||_6Qn|w!wVXm77xNn6PVW0_J$A6n1o^pLNpqLFiv*&^h}^X
zsDrawXBN}L4xK5D0rh4FtNKs<uyqcM6dUC)uFhM4&@`PfBrrgzTzhJBuPOpfO4Kzp
za$(_Bqg$(QCnoMgq63v<k<>%l(ZwYer*A at n2LMjZ4VJ}S{{4-cn_C5`+hLm~MMMO&
zsf9<16ZwOJf(#!!>PvZzv$4v!E4#aw_ShXMs?CsciD+(CMnfbJ$IjyoM{zvl%a>YE
zz~lGdJRZD>@q-puSy@><PrPh6oLUJE5a7H26<vv_%YU|yj*X2OLbdgsAK4GJBNDC@
zE$zO3je;zpVCU&Kba8RPF|Ckzy`^U*Aqvqf1;}7$o<`jIH)-RTfdLaVjD}JtXkIOd
znhy(S6xDuud8wikXh{gNU=KT4KM|DWPDmb<s<6z?plKpgQ<>o_D4Cg=ORb)rQ9Qfc
z5yz|={ella3Wpd&NsS2wyi_oGcxEq-D*pJ)KZ+*ll?!Ny_0 at USJHlsqfc{7nNuyQ)
zDS*R!k?ZQ*5sv-&@!1so59iwB413 at h=n#y|v(|$Z2j8QAS4YQgxDcSk2hk=AXOI5f
z{=LuM-ahh=c_|+5Mc6Q_v<-(mEgI}f6+`D%;j at n-IR`td`#@;l1oRDDnWJQ<>PJV?
zs~~xBeg)w4o}I0Q=UBs#lfJ#H0L9!m3<%pKTol5pjOSPk at RQGwE`ZDgqUNqHs-Hi9
zqA4iO75x;lg|~d7?&8-)oOKJ=hkk_edYuEOfOUC)?b?r)%4jHAW?qByZ`(UM0M#5y
ziJ*K at qiavPq!TTqCS2a<^`$BNA+YY9x3!{Uv}U-f!xIu{0Hkr~HOO=b2pxU>j?Zn;
zI1s)&RV+Vf;}?#ywY0Phpl30zhegM!xqx|UzP?iMnNixw{J9Xi0JBnh&d!2pEe_y}
zv$Y8dmxO)&HZE)GU4k0o+?%&s at z^o!afCtP%!jHuT<`a=C)fE#d%HT?QQ=H&pfgyL
zUuPXfLq8^YsXK9ti{8Lk9X>S$qZ3NV*1Rm1KA at pjQ&LiTK1&?<@!>H!V3!*F^l1e^
zt6Ko*Q}CA*R8*}ffLK}MuJ1qG+R~!3ZCdX;tC)z}tYgpo_d#GW{zZoq^++HJE&sE$
zuBp+0PrLlyV=ohd5QFgp2;w|ceisn9s22Daxvs7*7E_SB04iqT@)ssa>;1l8gJ+Yc
zmBa&oSKSP3ZE5kd8sw!uNY6W$w^l_mq}{HuNV}rZGR|g-ZfE?_f8PLEE4p52YHGq~
z-s6;eXL3?il^Bf&Ak3%6>U;#xloEgdNt}C0i`friN;Udr+6eA^LS?TESh6Q-hCG~@
zoFpb8AqnS&ypLR84*+Sr8|B>N$C76%tZx8S+XIC{j}odp1rcO3gB*P)!9jK6|H`$L
zzPdVTLt`Vllfk0zwyO7S0c^vBXgd4y_Rp2?L+H>^>-C9-2c3?vYM7Lq(g3`=rmdsX
zy1nrmjTW$VU~O#;VEv%SPD1)jQ_y8xOu#xOC~CX#QqX!9XQu-9mS-!&ap5?N86?Qs
z&yQCd8;=zkWG4VGTAE<o?+?Wny)MXU*pxvCLKV0Qkd^}_pxEL`IV`@A?`+?0uTM=H
zu=w%u at mN<o2dv=%B0t0#^$-ULu<<u;^cLx}E)}?Il$jp at Isj=O^H)_(4WeU at nwpxF
z>i`8>M%C6{D=Wjs8{@BZJ?ZU*(7$a-yqeyB{{-I_K6`4$n_z5##v}ld_l{j5la!PM
zYF65e-NUKQ8!_?mjBt|&5zWi3o}ue3EI!l5udjPXs$EEku!8`R$nUKV*d7rf?)aL3
zPa0%EJ6cDL6B}#8bD^N12m`W<j>Z6*4inAc1jPVxC`M4!Bx)Go+JObyLx^It@=U2&
zI9v%8ENy96mf~f4kouA3<)S#99b$-zo!&vO<^LnK^cZj$HvkTHhU<K$Z9P3{0MF1@
z`kyp^QLSudMgVn-<NI%m*>ErjoNR|sK&cI7g=OZ~*KV8`e%~M)#TGTIOuSdZ;~;H)
z{iviQI`k~JZVaSkr|Bocg#k*84z9zaqkGsB2Jz$-eliFQ9M=nv1x1b0hGhpG9jpdA
z)1kOn7;n4^;v6*@KFGIk6GGMYuRxi_Y3slhhaa7fgLLdn8+rBeC0;E&PVdzpXGzJ)
zc|i^AWtC~?4Bg&EPEKA8xYK;5He~h30aT(;$k7!CimH~bt{Q+;sJX{8m;R!Li$yHD
zCr at 5Ol?Fu~!u6Nh at dL@m5J{*Z;D&IH|G#uUAnz*(^%1>t0Wr;S@;d09R-h9-Wfrm!
z59X6VDR784whchv0WYzfxoNfwC_ at BDuEBDvxH}>i_M^2P#nz4WkR^BW0prlN at 6Vq<
zG1srF0s|C>Y+qjZl+7vJk`@ojLgGI|U7X&CCdTd_9w0|0&Jfu<I7ozS2mBa)vJY22
zI-2~`maX2^MVyG=?^&-(4thmzr7>`Y%4hm5uEpmsUq}Hakl2gaG)2PoAa_eiPmi5?
zk%;|Q9Quw<64+P*^h`S*EDv8nm70|m4^ZDSKF;AcHvr0WAG%iE%X#x_9;g==k4d at Z
zSqS!P;<B=u`6q68|N5d0_1$%T^a;-JzOeSw3VAv<Y(d`w#|j6uq+;jj#`=TE2=y0j
z?ps<~KykLF95zMZft;hq)dpsMJZPQE_o?t6A0MCpj3!uj?b`K^)46`$*;-p%T2jHr
zQIXyOG=S^w-UDR44Ps~U&(4*|s3_El&hW&;^?S&|0EN-7e7Mq%1V=IhSXwp)?w53Y
z9R_t9JIw&dxdBU)Xukb2iMJ|v>yl!?A`$4h-An at AoV>ieIPhFbit!(+0SUZQQX(|=
z at gp|GA|oRyK(Fuq_TjNp@!7{kJ$B+E7j`!Nh_F*gnh at y{2jr~UrJoG68}=p)dIotq
zJ3AK_7nK3RCMPG+a2D<mGy3QwxPY#d!~Zi{F{!X-hl^AOJi+#PmV$r6mMbCo6t3#R
zLgK`PEiGM!tXIM=5;87%dGnbbkg8|$)G2Wf1YI3~2=?Zvko>DjmH$XU8u6kIz`rSQ
z{gO=J_y`~tweBrI84SxUkNz{&)=c>F>U^nkK|z5nFdLl3k1gpsA1A#`nH#25=XLz1
zI`=L;V8dzc|Hj3}y4*xx6{}~L))zmM;K at 2)IX$(fb#<Sc!$|_R*Q<JREB}i9=b?c>
z>g(%^NT`<~70c;fZ~}t;0t64v(8rnC*y98mnuEqveT92iWZYlw^D^3n?&sI`vZA-p
z=E`^Mki?hVe*IbxTQKPMXlu*$bFA)aYt){$|NMv7rxKM6 at 1JDPZM!`Jv^vDxLqS-W
zIyN>|@KgZ>PLOwVJ8bQjIcol39ymFUJ0GmDK`Ux1?7lpN>_PT{&0_dW{CPU2|CP36
z&gB*=+`YLl*-oh#w2rN2CR|y#uIRo(U3AQY6=IO*<G=YonvA}tK<OdUi2qd(dtn1g
zN=odr0n5i*f;Gp!*W0&m8;&SC96`GpG at DE0RI>lv8ij0fd1VCwpy-^0gaq!7K<SqV
z{S$-(?f;RB4njA%_5Zu`QY1Rw6dZ at fFpvi*7`ImEad*-8```_Qzh>d<+Z&QH?!#dK
zo?P7AyHH8NlX&mCC~m9-0>G@^R|wS`06F>{g1DlW7cR29^m8m0o)XW|qnQ3=lzaDT
z0DRHX(Y0J<H_@}SWW#(yTQ-16$kP|`2P6%jcv?XnIRbdt3gyQeo*M=pls*IxxT at YN
z2hD1aQ6{K!O!9hVx08~>kud{lf9m(0X`lW1c$Y><+N`5KWZ~iAXgc>Ik<+M9XFqyJ
z0Zd=LdzT#weiV%{cVB=1(nJdhWaUKX4c4cZfAil{yiClaTI2fO*1>`I-qVkjc5TE~
z2eQ6;z&GM1Eip0i#LSE!iuOVvvUwppo9O)hkGe)n6zV{csH&2{{h at z#m!_W>o3-%G
zAeasU;lFy`(XsVEPZjjxL^{QYd!v<URC1wz6t7Vqh=3}K-r*ooZh+Y4uMse~XFL7&
zE*fZ{3-g21htbUofDe6e<P|$jtS??Xg7{+JK?Q=}7Pv29Ss7Fg^izf#Aj7u*$Oy}o
z)gPZF;^Ipr?xF9}$PED(W?KBM at p;@n_W4GQivggOckb0K?8Tf&V0)DuY!(jueJ|W#
zmEZR>1Qr(i?e(SgPk;ZsILswccx8Ep4!6JY8Uc9z{ry9SBZX!S0TSq%ygX|RRgi^`
zmzU&x*X@>w-Fw6 at OsYc_qz9da4Rf;^*Zt@)fQ}ka?$MT6UW%WWr}eIaKP?b!$SiI-
zIRacDI!Ljxu_5py(fy|$P0HvvIrmvuSfH9=7Pah2dlelo!C!{^dtY89<Jv=T5fPzd
zJv>4*Rsdc>$!6CQ8yAO)0E<3wJw)rOsw#0QDRp2 at 9fv(ehljmG?^AjE`E{V*<oco=
zFi8z0c)B4WA;`G?Y)o^$)zr&B7QD#_;eywT*bKME;<)<b!^}v{VJO47_wPsFzRj4H
zp56jAN$nanoxEaEZfjc`TEbsh`)T*;*1<Fh2eMD8=ae=e!%o~_iE?<6 at b9ms-I(sv
z6`lFt4jVQ_A&${10qMT`Fna4;@0WDAjO}TDgM0Dldx*LRs{Fq#Zp{C+g-_AR$qb-X
zD`E9Q4A~uAumArVaXKgnv5$IC_YULpOB2nzpg87S{MtnnhC=S;OA55Afd?H)l_F8}
z75ECCm~j8_Xt$VkLl|TcdJm$FWOK9f%wpP|00 at RN<(8D8p`o{hjLD4epX`A8xS?II
zH>++?lC2baoJGnx43Hn&U-+(TJDPn$bY+i_qMt3WDf1_uA|dm?r{AGEtl;SUbsh*V
z+VjOS2t?9EeMLVo0Prxl$8zgNDmOPb^k)YG69K3Rg#Rd7Gy<%mJv$0yG^HCDV4NSR
zX#>U=jy at d!8cHQgJUl!-4VQhe1^*&}eX3U2K`$TOqX54#YyeG`=H4H^Eo#NWF7Jy|
z#L-BD{-6#8dr?=w>W}n*zEwQyurPiM=yEPdo&UP{H35^g5Uf3)UGJjR89*T#@xd>Y
zhj*YRU4!dKr$9y9Oqa;wV!?P$C5~6GUST(bh>M?}47a=!IgYi at H@?eO$~VeC0kxm@
zh-J|(_2hQk4gwN6 at X&Oh&$rhtfFXuu=F;K$4jJE2e?2-!85A6B=i#x}%gbw^%)%5=
zWwhFb0iy%f$sWK{Ttb3Qp0D#969^kvCyXjPJ3C;S@>zH9e*W=8m_xIhko*IZ2YqnI
zpNA&G`k+@>(7t$yy{WfvM*_{m5`hZ at pkVaW+YTH2UzPjTdmvnhi155e#nQSVFiZ|b
zZ3{|t*~HfgYs9icI at 4(3)}x_@*aC;gT|fA#=o3Eh;IuACMJ6Z^QO}+!;D!>=KKsu2
z19-s3U+^(8b_?Yc6nFs)fImb&e0aDb at D<}BsM7iwQc13#MxS`)UzqCDNr-Vd0yhS7
zWdg!T9EgNG8-LWzo3sG|0TZCRu}=uqlFZ4;iQ`=1H#EOxWQ+v<l%*8P0d*EhcR<R#
zjK at e0Bo;(U5|Ti)^>!O7UtC%FO3o?XDb)ph1`kPO$z3337PD3d1V$59QLFkKKr9S}
z^78T?y9ZH_{2<cFao-0(2#<-t0%Nxr811%8-z7-ce|r}%S#HFL+qY978w2VJ7a9%V
zUv!C{fuXlRo8f=>7FJcYB(?=)Yti;((U}6Pvlj+S&zjY`r{BuA%P3cmrC$dzNu+FJ
zgU54mC-mW&GHeW?ac5;^C9RxJ(a(shSFw2)7DkkondyLDHd4;CPo~mLFrhRxH9^u@
z2PuV1`0eN5!2M%9GX>mP_1UF)w8#FR4>;(+G#wpnZ9Iij<^JIa5c~u1Os1pRj}+}j
zA4`bbma#E5xR-5jfld>~#C_-ECGM~r^qq+SIfFCVhJvZ7sj)eW at MCBaFX54T=gu{?
zXc{{Z%joZgdVXo at g~Q at R3(&9_NN*7pb*xi&#B36}Q^ndkJIR3QsYAM~pO7++`SRt9
zqJSV7DJhz{;Q=Y^r~=gNH}usT#qJ<Tn%91ud-m*U2w2Js<FU^;1hfO|5~{O8U1k99
zBmX;`QROxyN?>Jm`0_Fd3DqH!VlH22;cOrbc$@+aGO>lUXU|m-D7f;VsPlk4;czxR
zP`Wgy9W0-E?}DPx3AFa}{JhA8JQ)K+yZDl;K%_07hp_)gP1FMd<lyTOTc85wL+n_Z
z8s?;>5tejD^IoA>Jw$Npf5#3+2M3u>KQN(_x4faIj+K at i(i{^QxELek^uS4VbyBD~
zy-mSaAdJ at TQ<g{5u%pE#8i?-VJ*OvwhPl`GA6^8iWx4h&^zZS~(o!nHLq#R1YKx1D
zZLEjhzmF;}7lr%Wg@#RPpPGXAK*=hCDufp3P at 8an6_Fmhqww5*k#wh9jyGJs3IH<+
zC^Y;$RO!e3JP`e=AZ-xQK(#OZ7)Gmg;5nouWYl4CaeJT+aQOzrmlO<D*f^jOAy8PT
zYE_r!&42Pb>b2pKD0KP!_{<;wE}re^shOJu2l_>X6^FA+d>#i&jPl9wl}yXE%j=8W
z5GuyXbiij%`}#^Djllx}fTqU+mz&p at rtG}D#8&!e4s=(*(`S?O#)EH9{TDdoeP_Zz
z&HUP{UgQlwPZ^%q#E&1_-L;XCq`2AA)YSA3=M)tKI?fiE#4(F?LUBpDbBF)JGbqGp
zAuB8Ugg7DICGSj`c}rIpRi=ywy*wj-v_bZ9JOzd43yz2WiAP_bUteBrGoIf2YvZ>Q
zZmR)ov`6h>UYzOWMO&wl5pzWsSmHYZwA1}1#`p2W;PTIpTdDhNVkv+>w#Cr#_B?hn
z*p)ACa*Fi-&H|i^ud)f=UZ+!0QCWvP$0kR6XJ>0)A3c;=jr%9>+B!MCAlcZX<><%{
zvi;D#r@^4dBBP^uj~!!2-BD5N;pP^H63V$ZPc^!f4(=)XeD at tZg_z8*|8EoHEnFs^
zGD72az+Tcthltwzd|o2t%$2pZOya7Y1#1u$il#9*BX|)qgZO%k%OgPWZy9$$z$vR^
z-vS$5;z#a$1Qvt)Hh=$Y?c+Qiq^ze$4YZbyf#E8KedOj3Uq3&iQZoq*7m&VCuvB8{
z4~A?Uwr-Gv*u7 at oalPgBYidVF$GU{V_0S(5{mo3AMwAv8q-YCdXB)2$>?)OV(2S`f
zO at 5kjqw_r>IZaw3bNe-arv38!FVeMy{LbOLPU7JdQ5{WrEw8?AYBg5JZ)?$4-gZ;)
z^Akt^3MzLK71C^YpMG>&(f at M6DWQza%-+7fi28bYv`o#(ArcZ2(gL1uC>Yt;cuZAQ
zwG!~}(OX@`Cy+R3 at CQs2?JXB4<HU}z)2mQlhqc`X#fo;DzkdHl^ab>{?YO at GTpL{k
zp&2M@<%OveQnIoR at T~z;H)OST1XSYf8XFte*Pj2qQPH$X1wh^!PC~AAY?n&Bc){`8
zToe&AaZ)b(lEj{G(`N^-eh3q=WQxN+ at i`|aN!kM{v{M=&ub^&4!;+!DER?T~dcW_s
zuC86a&h2tMOmXM}8?>`BBzX968-(=|4zRx)7t4*iFAd9mx<s4%+O$ue>^xl25K&tz
zgI1j_EiHOc+VQ-*ckc$O{`aE;+>f1;ldK*cox=zCVHbcz-cC)GWT#qUtAF+kOJva2
z{F9emT at lZpD~(@SI7YgAclaAeNp=p27*{v9*=zGJUuxaHecQG*k`kNUpj7E-X^8;Y
zyz4BB)Y$#7`a<;<Hmf88kx`Jzg2x{A9bYGEo@${O85z;LrcFy4J~n0<85PB?sK^GI
zeZpG~t<ynlKxrpsWo0c=<NNb4KcA$eqy$wZpdY%%b#>`fesD*3%oi`BL*$MgP5QlN
zQ+;L46B#b6jC-^Hr(gkD_b9FyrPa25N~o)Fe>6amLsd<UJ4vhgv$wd at TQ<wOi&3+*
zl#}U4ft|?8GLxh{S`1JO+MK<Qwx0)L;t6i+LB~%S8)t6 at GsQ(!R*J(T*y~_Ix+wR~
zfBG#u<SXv>6ql9lPVJ6LN!c$fESz)yKKZd@#|Q+1vZ*Ob_MLLhD_5>?Gg9>q4w5Bl
zk-~rB%3<HWeG5q(96Uo2rCRr?qNwP^Up+rRKe#Ea!c*1N)j4HlH;yN?eQs`VC#|im
z^}clJT52jYke?HCM`Bb2l3$@Fz{aA|&T at a4S~sPA^{rRuV&dZ3KDR|<SJqpOtzr9D
zH#YW*^wL{>14t$EAf0W2{z3iO`L~M)kQq2<e$m>RqiN$CE8DTa+fqIZ?SlBycf;z4
z8)$YsW at dN({QCt$m07&O at eNX-Y62Dh5rh-}|8acl7QLB;#lAGTFBj|9=N?wz?mrn1
zFNMAE_nDbypySy5#Xj5k`1tVC3%tMqXf!l5irRQP!vL2jqnNGN+hS`ze*B2V1WQgv
z1`8hi1?m%3Q?nwjp%X8A<HijLTTLGyNjWwec+P0g2sGUmI67|lfnr4$=|`ixBuXu<
ztddev-Y+zTb=V_x2P{tPQcI$9&<7;Nf6L57N7|w`ZnBp7(I-_9HdxiqoH-M)(71+;
zZ at nzJq}iSrZbNJ91c-5U4UNxG;)_d4!mnSa0-iow<RL37+gJKGesXfM()+V2DJkg$
za7Bcc>}*axdP;~i+|jlP-AS1hr;H#t1zx!VdN&dH6Bdw;jt-=FhK96EqLwj7eZDDc
zXhhA5$2#Ho`M~@0TWbrv($f1?b7?_xMBcwI&^THP>#wJ$mqz11fVSP%SLUVhxEBES
z7iksR(7tC2P3X6p2P*tj$!+UI`zo`SFJEHCZlj}bfBd1&N8ry&Md;qAlA>`Pszye1
z!XhI7A{^6p|9<I-wWpS%*croZGar*FQ}id?_{wbA5I_fDADNUyx4p9?;qy&zeSN*T
zrNvImjRO!C(zvIuj~F8Q-aQTq(uj~3RUh1)U0qd;jTy|%%?*u>Ss$fQ*mJ>?kUsur
z-2ovbVEV3q8cj48nU$N{;geZ%;h^v6KzX5T!5+G)P4IsBUcG2dTHX^nI#fWEueP?T
z;!fh<?P;a0y@|^!D+rObo*%DQJpYrCs|Amky0EYib(D^-E~2Bbd9~J0AQ^v;|8H}C
z0_vNipiTrKTPs*lg_cx>TYp6DWq0=;fEJQ_-nzQFc8-pnhUZ9C;-z7IkPIbo9J>xA
zG4k3q+?82eU3HX~vZs!Xk8dwM`*;x`(QfVMM-q<y4<BONqW1y8Vt00SZU%bi at BXnq
zMUP70+E};P^LU7HRbeKQ082#%disbvcbHJ=!auY_!Pa(QkobXDxKC-j(nf_lwYx+1
zQ_xiq9FmL&BnANuX6EK5zgIl(%=Bt{<tx|D)z#hI2JjX=^&(O2?AZfeO&k2<AzK~<
z7M3IIEE17aR8%uX^Po<pWMouzb<v+5$trL;U}@<zKI+eaJbinYz3+H-VpbNLl$4Y*
zPg2h(rO-c$cZAQj+D5|tlSAO)F-9{p0-ntK_RUdFUf#YpPXk{QxGiM-GI)*vD0(w9
z^YD|=OP-#h5b+=gnx}sQUEV7uCKhnzikh);^6X8`(t-lM$jC@#3ky~d^r3S!q({=J
zf%WqYb&&SHKofZu7OHeT{?u&#tJh<Ciy9i(n3$MI2G0fr2J##|nx6YIriveg<LA$x
zcbeXQy@*F^tE;7sPz1-scGzR1kbED9x~eKM9$WRDJ%vY5I5;@|#m*oa&ktI7eSZAz
zc$+|WP^+&Tsl>Tkw{Cf_FWR9wJ$)>xSmQE1#M~Ym8=FVEf+`bBwmrB%5+~M#T at o73
z<J1NQ21LpxCXA3Lvkq$upAvy7OYT3ULVFecyuBJ%m at rHdk8e_BhyFPV;DoxAjg2~8
z?;79}mx98T=8v at ypDh1DX3kCidG2bbcO2+?c~QVpCKDgn*z?frmX{O}gxHbM(cZDK
z?bmNgPn9srdX6=(eF{y)qYT5t1nk7KdNu5_viiuKaEE}#NO+hvMx6)B7F1&RnzTPn
zT70b#Okr>v%3QRz^CLCu6Ib`yo;%lWJJ240wL#dT*1dUQ_WO5U9-fGks<iFi4*}Xh
zpN8#@I(_<dL0T5U1lR&vV}gcDr->pE1Nx%F at haSYQ4v3U0JYci9hrZX#krj}`PDf*
zJ`VgHTl2%iX4N${DxhRRLHhQ~e9$O;4qAGBa}%;)%_7^_)Kt^XjvJ~GA^s*;OjYrt
zNA$3~pvctVF+(V!xumS)=<u*OJCB53q$+n`U*FBFtau0|d^x4Gl$yx)zfMX0<<XBH
z*}J;BUXv+Cn4CFd4 at C9l+qa##@KA_W6~1k6{5E^y$qroSRiK;-(K>(tMqI3VpBFn-
zKm$`sb~ZL4$?|w(l)OOBAcUJ7DtzAg^&;v>_mS$@76`#F?@etwfEa$c=R}lKTpSZ0
zAAg at Fj{BaCfx$lPcp%Sv|NeczY+))M)?DA%NGei at 2snWUPUqe~!~)pos~LZJetsTR
zFgBUn+tr6hMka5_jyJ<s<levk0*Z9YQvf<7qLh at B0c#)qmli_*vKrq%Ss1NFoap0#
z{R(ey at wQ%i(xnD;kHuq7{r3%CKzTy$m1{^*VnW4{NDj%pz8-LdFOBOOWy!vS6cJp4
z1=?3A-VYA8LO~PFZTsxKxU=>txx*-e>kF_~)Z>pG-?V;s>=e2-47`Nx=tWZ8o4;^D
z0s<ei2z$9~G)`$e{GG`U9voWS2$GVP&V_QL$g%J6nbPKWN`I+y4PzRwtWA!3J=f6G
zB*$tJ`sa5xp8xB&W0+D0DvQTqM at HmrmvZy+cwk*$o=hgq#}hOl+culR-Z)|_8F-9?
z+!6ayELv#XAR9vKVC?TNi~iE^`S><K&~eU%hTk-IL at ZXL6}A8- at u)u9>)a5|PgCbn
zadpKb{b>6O&jCWMd0^nS=P(U5H6KYb(U0LO63*adqv!T+6HA^EaZDnu5YC;-i2Vlv
zpJr!gF&aVO!xiQ{cyJwXiU+$_Ou}Knru*&adDVl1kDWSo3ZVAOmvaE3VclO_U%iSQ
zcNv9>^7if9iHmi5(XWv<K`zpwJ2gHH927J34CtY_xVSP<8BnLR%*>h at W3d<JF;c7S
zJB(SY!(R+NuW0y1mB6kb)6hcKZiAM94Lhr&zG}M`1FUN5>sdB8H>ppYh(Cyk0F3V|
z*|RD$VLUJ7HdtyW{-_GF-VFupp+kojH@{and><?WI_D%W96?B__4<^6T#i}Pl1tye
z{V8yzD79!B?7BO9d1-2CQH%#n?R}jt87-AhH<h8s|IhFqc{?8W7ZOUc;I*wXGc+_r
z{TCDzM2Nr571MI_&Yf!qb?@Yqm(zgwp^fWUUK at 6o!OG at 8emS~@g3>;SdnK0?3a at fT
zb+jeaBYOvj*)QDDZ7LofA|DE)xhTxxH^jumHz6e_medcDkX#LUao*9<eqp>3**kpP
zpVcSN at ib$>(oMoC4-XGKp?Ezm?km$pGhrsIg4oo9ACD|^Jb2Bj?}r-oGq@@)MMcY>
z8y}c~rAI~WK62!UR+VT=8uqK;!!E9o{(F%O>Z|v;b~B&9e}eVy%J2XB^{cqF^m%zU
zdqR6hhaFr5?pgvjU;LKoWe<viWZZx#(SH!PL7?~~Wn}2Qz2Tq^fPg1VA at Cs40M<ES
zwejQ;+%WK`&(qVgKd%qNhwV{PQmTvi6P|VK3RCwT5!Rw&K>x!ayy)23 at q89wGjU2v
z3TkMei6B)c78d?Sw;5AJoH}#ngsg|a!Gqk82LLFBr%pxu_X@$xi-57%9U!7e``tY}
zF0B4IYi4FP0r}VgnZgCaXKLx6 at M#K4r<OR_Sy)g|;3Qi9&09`TRP<VU`UhG~4cd1V
zRaJBlU>DsUG&MB=)SZJ|6yBIt`Wp+$m-p-$Yx7h%6E`=voc~X at wxk>S($doSz(uVa
zV(<XVP`TAP_Dn9amTP-i7k|G~;yrTYmefbODIF&#0X)0|kDXgol!l+5A1~87bjOdt
z$_lDdMZBn>fYjRB+S$crR`%tMw$I9^Pw#*A>Q!ZJt)rl71Q%I12%t)@PffeX?;I8s
z#L{Z{jgcaP^1sSTq?NpSQvrlJhA$ujNFe0pgJl+a(HgYv6h_w8G at +rPCZ?up!c14Y
zzuH4mtiKEWLjoAUjV+z=#w&Ep%y`}xVfC<Sc_iqNwq6DYD_B(Qg#*5<xP^VaSSKze
zg)ew5y;WFH*42&66^vc_f#;DnzNY?Nl#|K4k(0v(&-gMSk0<6&MrJ0yz!NA-Bbpjl
zUoZb_T${Qc*fOx>$rqfV?V!<tnY4>PC@=5)v$M4{m8g_gQ?p<9j&KXyFBbUCw6||h
znwgvDL2_A&GUH*An3%AL2;!?r3~ITcj{O(LHO`&m1H{84%;V$M!yi828xTPL7JvU{
z?bp!J*VhNk2{Spg^O4_!DQ-82O-4pWWdj2*0xuthy_Bn~YwP=mrgTh9*MMv&3o_!7
zY5+!6C^ustKepIcrO_l4NT53Z`t at t(Zf)|dTQ}3wDXYw=;gbWOcpT5(=jF9<pJwKO
zF30hpn6KTbSf}xDLcQO2E?SD21s^#!ENt*sw&FSwws;oq_|^}}v3kEfmZEDVBqEm;
zIgsgt#NX!=NxC$IJu4$4qnlcMwvRuo&)(0j3|w(7YUAm=iCf)3wr76+B#UkfZ)!UJ
z=<(y++}svlXNR{Pam?6vbar#Y*FN#Zlzwiy_B5cYsO<0 at u#k|@x%FKVqWc*cC+=s5
zv_j#E2U3Ik_t<D at joQ=l^<ws{6JFWe+#Jw;j!9=qd*|Sl_og5<M~#0wK*_sy?HX_e
z{>&=cc0ndOy6}mKiI6seApFa{d-q}z69wAzDeVb2Hw`1WD2Ta0YMkVvRfeKyUqx`2
zVTG%2>q<GCP7}u!L&v40gj5OQ;nv}_-ku)iz(9E<6`UdKyI`I0RQ7n_cahf_(s7Je
z=-X8JlRqOJfub~%*%^+nA3$LubiSmRtlHcR#Fr1s%H~VrZZuEvSRGozcY5Jl-JYuQ
zQ5;oKQNb4*h-`tVK at 8c-@z|GAQBy|(*L68bxQ~6G&1ILMzdLJMy1FFU?_PA?V8xw}
zzZ=<qTi{_GlCALrjwN^b7hl~A!W3{bdU|?V1_mU3d<PGz2s5?y_0a=tET0_x_U)Tq
zw7fIcJrJf5Zyfi at 9c>WfNQ{Qdo#-pnZDV9tq0N5<$_b+mlyUvdgIruhz&!De|7qPM
zf&g4n(!F~LINxk&9>@_SHo!%{f`{>&QCJ<lXvLZO`g(j_3)gmKWo2hj_^dEf)8^b^
z6k)j(lo6Mqp0EFS{En81X?4KSWB2aew^LGX#O>PGKC=cg%ywn&{Y{<1iKPH%cXvLu
z(pZ(1H?p55`VEvcfOA0A0g*56F(Ose#$pUtgJ*%T(xUKc+F7*_JbYM^ok=DB=-s<_
z7k~Z2b{h{buQo4 at geNEzd?PG+A7|(M=#$)vip0|o&wK{P7l!SLKh{vefO`Qx(!=t%
zF-oXVm%{=AV&~-NPx&WKEd`(?hEIF(MV?0`-f#c@{rDoZq2TsMOS#N9oxu8V57sU|
zfvZ#O6sHb|T2L2uc+jM>Qe5!Rp}gGOT at M~S(2G{5C2hkK6Rms1Kon4$>2*3j&Mhg4
zm3AAP;QYHz{`&Q6!u8vP3hciFN%{MehXyjMu<+nTAD`^Yzl2q&@7%meYx(SwN3*J$
zXzCtdnIK$1Gzr@@u|X;585(lpIu#M{V|liJ`EBR at lXqLM?>|gPh#@>YzO!t+y*laz
zyxdWqF~SZH#5{J|2bF+m at 7}%DPoCsIer)`cz!VpSObPo0YDV;WGcz;IP;F>gS>tAA
zoFTMwL2N<ESX^7X?>|~wRdrxM$&^2N?{{T<r39N4PVeraV?%%<N8pt!+42+9a48gf
z_p;ZnLt#xXI3<MJmk?#xY!nmQXKrq8_SBo3NyI!79%0f-uzfpVh8pn3kJWc?-;M-|
zTWKbYUe7m1p(c6z__Re*GNjRL)TMyXdi(ZgiP}MkGi;GGY|cGwV>_S{FMRp(<@L!p
zG2HZr>Qg6e0?`iihg`w`=Q(lJhyWQ7))1`#TqgimMQxg1`)aile1P7$4_EF2TtEql
zZIU827UQ|^->U+a?7w{ZGBI4`5U|w2oK6$V)mdFyDh&F1DnIdfE1(fJQV6Ej)(P1m
zo0OUfY%2QtdmZ0oV4HfPgV`n$YQk6*$X0ys4N$!K>%}FY19l+hsWnd0hFUp03x at 2h
z3^~c)c`=2jHt{44REBGEN1))wnF- at TE{L4$*#O9!bB>O!Z{K>x at bXeb3`Eldn=8=X
z62sd8WFsZs1^3o7K2Ae~72MxlR~P5*A?d}VT6k;xC$okZiq2?g>;{~qqN2k6xq*QJ
zP*hq}O)(cV`IB>@@&Tuzja7&LW at cx@wfJr+{CqXUx-v4WYiykS$8WMd7QfBJ#N^tl
z*djIHkGTI&OW#YE`koE1bl6wr+`k{)*LMntsst_flBwalq1Dw(C`b>bsB+Q1+~G8y
zPHX}I_{#M6C>ss@^kzr}D6F7-A(hvzgzPJXWW~oDSmf}*T?RF;-}k}27fjl{XXodm
z3<V<%1=WGefKrF50=4Z+!uKCP)J;uO$A<Xz(s~wxbzu$IXvl$lBR7LcGC6%Z`(f4d
z1$(k6*LouplXg5)eeM;#u;o)wqb#8t at 3>=}R2rwBR97n>O35oL+uJ;)abfQGu7HLq
z=9HH^UsG*14~FbgT&Yp{c$g>*NW`(%QSzp=>NZcjK6Pdt^#QF;1ZYy!#(g4eZq7CG
z+BF^ti3`o2`H3n+|Av&>JP%3Z-EtUMsewzkuwV_ot9oFDC8~Zk$8e!+pr+>f^{z^5
zi$pRjD=RG>9aZ4<_V)I8KmwIhb2HI at JmP9ag%}$<d-nHtA3$0l^LNPzTwmlyQiq3q
zfQ~{4A4!kr(7|;8-4qcK!Oi)DmywE_DXw>9<dA=&mX-0cbdR*G_ at eYxP=JjLDc7!%
zLA>SU<@GJxOXiCCgEeS=L>f1`6cyP3B7pv#08xVHs(rrcJtEmJ;fima0Wn<W6%wKp
z5fK6CMqEryWnK?D4m@~bW=3v~Oia|GmO53;CS=Y}WHBJ%)5`ZD36w|i?zW*a+3S3_
z4EXMuYHL%X9?I8DjOeiMtzbWR at Sw?=Gc+Uh5O}!PAH1+yx{)N1J!KLSxd|b%pF!Xx
zpF(+p8U$%^MTI8BIG%j}@xwi6xmVNM-=BOj^c_18EVyuNg95=){}N8=xU#&QQ&K_+
zWC=9Z{dbMQ3l&@69^ngyxKBSi%G%r8%O|T815AdFjg1;8WK0aj at aU){JH1N$o~_mS
zHid--+-B3uK7M!Bo5EflcoFWjV=Do^6n+-I<Y at Nnk_a0c8_~-OS;NeS+rPTScpp_2
z0YfY+D|7bn&~S3PZ|Mn8s_SX%^C|0q-`)a<e7sa==lFU*?#Z?HLqlz2V_6|I-gTCB
zHDTcZV+ipA2KU&iRk}@f#7_Sng$%DU8xwn{dq7ZWwc4C;NU3vbX(p=z2&t8T9_^G0
z&}AMT9`B!HN0FQW=3!HEcJ{zx07oLZtgP(VGyjFUi&i{IF~|S<`uRc5r*I_vId}5e
z!bWtL-NTZS-f7MUWC>g+<ma+ZSUDKIyYk>kp;(d at X|7>R%3)LD0837tGOsm_ua}Db
z^kK#H(!|5Fx+&}<^9M7wRw<2YCT}-QpHOZ}*t7nejR2w#QV=vc#F#1>{|$+$dhKU$
z+ch;c#U&*wAj5H+a9}_^bZpHhNo!ys at 8Z+q;$meZBRU`$hQ`KM at 7!U+J(>J`wb0VS
z!rhQ)Se}N5380d`bm+uWz|+O=fbf89?in7AfehVrm@^GmR%X6HEe#L2K0q6y3Gfk{
zA$ZTQChY9&1Ofp+2MXI&^VdPYzv*MIRG4YsaL&gej8{Bcwe>jGNr0BZ at a$QkwFVD&
zclF at l#<6MQ at G$Fb4w?ZU)6=I at LdxMvA=V6j`Zzc^q-A9MDQs;>7BIk&c;{OVL}FRB
zT$zh2Dva0A!GRaZzTxT98q9 at KGBSC<0zyQMqYlAcEw`?F=P<DjynmmYmq!ej2Nkh@
zp<`rZWVz->QW9N!d^~PUjf|MXb at wRc-@i`|AB6`7akJy7&$pXAER*Rm>OA-fO-&qc
z-n at ZN$mfI<rKKRfXyfMQmaNNl?5uMa-`uxGFCHB$tHw`i at av}!d<jGdj at T%$yFg7B
ztB&1Ae8-2L9>o#G&D2zKC#UI<SMyE4$WUqm;pV<k2&@1C03Lo$MVXtcD|}ZN01Dy*
ztjY3~#f6S-#p3qkDHKxT;yD!$OA-b2Ge{*RCDSr96JxCe{!&+&VPCPmvlGv;;=7E1
z?;y?`-)2#QLSNt7YJ2sH(p_dNJ)}sTCpRt_H(uD}yu+#{H$WZYW)2b$OG~!KwkPU5
z;ZL5(qG`+8nm?>D1Q}>!V3q2+y4?%M<>lpj1_xzqjk~<_=*M^4ILG|oSpeW at 7-vOA
z-xiKjQ&a!9huG-!CzL;?SDhzm0*D^A4Usd2D~Aag7jg0Nk%Br4{Mo`tDshW(|9);j
zLfqU2^ak|+zj8o+nVOmg1O`UDo4)#Px~-+f&TxE9X?sx(UxNb;`R*_Cdd1e-0(RU&
zpL)J>eB3g%1YDkklQUao<FWSfB*HvgSf$&LD4x3j;c}{kaiM8T6j<+Q$csz9zW6iV
zzkd%nc>gyy1o_2m<Bg<nGHUAGfOGXZ$z9Q~rmw!PE-x?Jy1P at O@AmMqR&74|FVoY!
z0s`b38XC(e6X9D)XdVVK>FDV6^!A3=)g3J at E1SjBk3Z(;UrpSV0M5#F`0$>Njt)T)
z5e>lXuCA_`te?vIoT1ksx9_^uR95D24h|0DMrlUI2-l8&XD#p~&^>!rS66Yv3 at TX~
z(x<)Yde=am0|DTcluXXrqL`$FHNyRUd`ES1(jHGG;%O0b&LDBy<e8 at 6t;NL9Z3i=Z
zH!m;jTX%JL8=XEK1y#0we7ZMZ)7jIrh}DKX?)r5q#2BdEkb;B*2U>c1ZvGh_8DRk(
zRFHn<whL5)OP6GJv<wV1PoAVX)~uzXvJ0MQV`Jm;!sV^`Csj`GI5vL+S*i027x1xx
z<hhvbf9X=!`+o6Uug3UWx9IOK^!4?zi0?a5#!EndZJYMugG_OF{$dZa=mc;IfN?^l
zUFI7{2dEgskIw5p&+Z)@yaocK>uxB=vB1c#E}enF!M^G2mLLy<EIB%atG>R7*jP$f
zgjpY2zdu#g)qHQp+H+=SW)RO at -P}--4xQR}djYZ%DKg^NvuDxLCm-Y*TUoKA4RH6Y
zV){bw`}a|x#p7?>cx5i9vaS8!Mvv;*vj^^y6uC09valFhSj2U9o~W;|FW!nQ{%!_Q
zQvC3tnvG5N+h8Nk?{jml?d_zFj*jOXD(F04%eVj?$|)!yg#^cC_w$nu{;YNP+?I9Z
zuK>k5hT&uZsku&lyaax|oXju4y_97OkJVWif2wF)8^3}tmq2EayU*lYPb`Vhe-My7
zBJltK0}2vxI1z%AyE|LuVt$2oGJo%lB2e_=Qc{s6B|@d8rM50El)Sw8Dw!cC;%$x{
zI|gD(71oPt&mJQqBisgt00NLy^)!_{Bymz&eN7|_GC1G=S;)m{w+Bnb6#_P11}7);
zE}5`8JN<+L0;otqN?czF1@~lmPh{YYM`pr!c<A+O>Y}0|aR~`jTkakn_{KQ{L$cGK
zC-nhKJh)X}VzL?ihU~w-&r at Kcl7E9HKllN>R94bPMMVVw13Ps1u+iDG+5$^k)_OvR
z4t4EcH{>*aXL0`g`Fgn+d0E4E?*#nAbND|kA7rHZ%5<zm=+=ObnIK~y33c&wxuIZ?
z>r}uepG*MGU=>Gudtzf_V>1hjw!uNB+S=Nk41Ih}2{?9LK=IS9lY>BJJHEV1{yaBF
z#mvkscl2oZyXoZ4VLSfhS5Q?Bnl``}L_wgLn40E4dUVQDRdmI&C?k7IM$I^uc`i^B
zR^4;jjcA9z#z=oK_*mWW$wqRLo1^I#T(muVmItm#y;nXd5 at ERaOE9y1@w6E*ZP0X<
zR#u<Ce~*KT(DwE%O=oB4ml4M-hPV#jyho27(L<cI0)^ZBs4Ocb^|JAJ*ozx)gYTa(
zzVRCB)AEa8fu&OlrbY`J5bm~*KP>;8AE<6?@*HoVB*K;*dHi1&jk4f9&nzq~xKnh1
zsOn at NF1C8Mb&}`jsO3p%g;Narqa!1m9}_`T^v3xQ_+^xrivmvD+S(E^FfgR0r2$C}
z6OKqE9v>eEqTB{}joW6saPr&e6R*FXeeoUkDk>^=hAd|zRk?rt`gQZ}U45H;!$i$Q
zj@}nW3kn7 at T5G|73okVl6=~8ek8ggF?n+)L9Jm4rcNo5- at ok?!LG#mz3v*LVYpG;p
zWa6^2(RFoomUiWhjXO^XCIh3RxG#!zjGv!uX=w at AUKE7g$Ljg%X>wE3q5Dib9WP$s
zNFpMRPc{TrR(u0>;`c_|hUx5mAT at M!bodrSYilbJ5lS^6`m^CaD=RC=kvu#^?#BF#
z%WzffL7&7w$NhH3{2wUQ^w4a7Do+dE>-$E(W2d00Kf}V at eC9jx+;=GF%hFesxd at dp
z*8-QV9#34X>*?>m3h;pCt)jx9Lh at 8a+>+euFv8t)M-G69^WI!`$x`sAfuFdzx_Sa+
zDaJl8FK>|=UyNNZ++IV2VnIOxm$dZ$l9CcU(;=^*fIW- at l8!hstf{GKadB~7H|puY
z=4;$`w|aKz*xGkp{$$*WgpYmas|ZvOQukJJvYm&=&a)Q@)j{?)eA!ymwVRuPKgJtb
zKL!2X8}?$T3BFoTOic6q`GcV0zLfi4R=j5!Qs9*kIkG#_+_Z?Ln4DKhb-r9fI>Pw-
z%V3`?1BT9XQ-ExBTiDYHJk~un#s-`Y4-+>w9upQ3dF35pHwfIMxLA;nkB{r%!QBTA
z8~}w2%H`tT_fR1&zB at X1Z!S)(sX2jNVP;F<5VJ~IqKvF;{~Hp#n8SzFpz1Y>f{v27
zFsc6Z*|WC(eun=02AaysL at sRXpa1K`;p?n;&;~vnzU8VO|5d0O!otEcH7pRPGjnr&
zN5$a+H`%CPE|~W{c_{-53(&B<yxce3GdTFzU0G51NiHrz at Mv4?t8DG;b^(WYT33h1
z$kXKg=AN3dZc0M<PAn|!BPS=fYz$0cq+%c?;W!>dAs`@tZ+M3O-TIVH6U}vkuf*i~
z%a>Z)>r;ui?Mh_ZylY;VX%9{T!rJN_dg;m)g(F9f;7OjbG0Xp*&;yONoPE@#kFVb7
zhO2phUtivLLU9HeKv|%ZT~6r9Trbh^da^kSN#)=%Algv<v4K!u&w}5$$hv#?ZZmW9
z*70$U{~eo(jwU}4Aw>oM%B`TlN=ZqnrLV7X<_r^5L-kLo7H7|fzj4e51pYLV^x(nV
z!v90ndw})aw*TWwOFK~!8d}m+N_$GQG-zmO(a_#IG_ at pBT1p!YiP8`acapYL+B at x~
zlKju}zMt=L{QmdxJcr|;KA-pdy3X@;zFyaL4(h!tIyw)Y=RweW`qOTwh=%gD+DyVv
z(z?GdX`~~4HFbYl=%sy0kn)1aOr7*MYt=%p-glS at qi1hAnD_d|hPj>HrC~Q<$oDsd
zEtC^3<zV6(_KKPMa^&IT$1M&Y8!SqlC;p+svp+dG7Qpa-+S)Ea90S<;k-(+#fcbkS
zxJX9Oj%@yu&b+iBl78!|C7*Cf8el3N8v)(h;r)*uwoHPr5_)=imXwv<efA8*vh>Kr
zg!0 at sG6|q+larGGRthdI!a(6>ejc6cpK2a-EMPmkf6^C1*Y5 at A!M8q7fGB(nIq+w0
zvY^-ml-j-$=h2UWHT-pdO%k-LmVx25-wpb`_<gXWL-_hhiJo))k3qrw at OV~QDhy%X
z9Af0@)2BghJm{qoUPM*%^l7l{w04aHY*+q^7Y~1GJoV+x-#K5{8#9;l)OSEnPj4oW
z;_2UVhhevq!GZNB at _$|_^o*LL$nu<V1bGT-dXQOpqXJZDivu~HUIJ(IT|46D=7!0B
zfc(e~PEv=(%yBuytA>EtSlzt&tfb_=Ko-1iCjCEpqud^y!Vt!R(!gHKno&_whDp1y
zAB+t1_j<Ru=m;FGy(=pJlyA3Q4G?(PkS_7!d-v{@ynda{(tXJG=FOYxPeg51Ejd*(
zhK5Whrl#<^Swcbr_b4z>7MN%IzyJkxxRUrC9j6_zob~#K1}nXVb^|lBI6Ujo(|cN7
zo$4rC42lq331ZGZX at RUauuCsrqBzIijWjeg0KqMPs?7qoiLqQ#o$I%5)y;{CiTi1O
z=}PDrA|hDd27lA8(RcXpVLYLBt#obEol{tN4QyI^jw?&IDp+4nuf26TAmcs!$jrh5
zaY}0#5SYP|mzkNj1#7ny7jJ?p_y7x+o}RwG{h_Zfd8jdo>BQjR;6KgHSnPX0BSTBv
znvsE_Q+yd7)phBu9vi0zmFSbaK%Fkb=Y at r||E2eW;>DA-Kuz745m*D)Dofmv#zx|=
zU%zJ0I7Z{Yo9zdxw*LFr*`%jMMX`>&W3H9%W7$hZk`S<f*vD68e(UZ&me@;HzAST)
zp(a0N_jYcbcD)>k*O`khdKR&*H at bA03P{g3g!wMSXPus^+4y$tpVrp%dU}VtGHEHJ
z{>!{dpEaNk-+$#w&rLfkf)m`^<mBY{A3h8Lz7NK5Qc#cv)?(iYL1~?zQ~}JX!1KT(
zCCBn4-;|ekM3v>_<oqi-P<-^?OHMw5TQV_01YE}3&kxvzqdNC2=fMNmxZ1v~qT-zB
z*|Q4ZoLEJik-_305HLDBdr?p}XuabC4*Lxd$gbVn=rtSv^J|#pO^_g<-m6z<CMG94
z?2mT;b at B0$FnRc4)Olv at c+YF)muyEtIUlfda99wQcu@#qp}f3&J$4l6-1>6LyLXcB
zTwX#*+mq=ImYUme1P23 at Jtrxtc<oxshxb$pi0pV5js;SytFA!4*hEDSWoBl!xqm$w
zvGVloTgJY=zKEzOrCYc7#>U1j>+6%kYG6gL?0Y{S9|djg6|bPHK<s6HJp43%yJZrq
z)xTH!teKf+*xK0C1?{b|C||e$#O;A8Mb=_5=~2~&Nmp|#tB{EaD~Ks^PyEN6S+^Tj
zu!HE3aMZiK9CB at rg9i-w@Q7q&W=3^-9ihR_jvQPUR0@@Ww9JFE5XjnwYb74kX6RP3
zfj5ILLW~A&1~I!1*krThDr>wdN5s8*A2)vX;`lJD8#j<~quru>{q}7$IqbHy*30uw
zPJ(!z^nnBQu9}*XmKKe%v9XZ-H$KcX4h#&;ZVoX|OG~S8o2JHmB-lbLD=T at f1at#O
zM at Kv=Oi5wp(n^=M*`Mg^>zh97p>_5({LBbA1QHAQ__`Zqqf=91z%8zC&i9c~QXcD7
z`T{U9^7AJJV4m^&zyQ6TaM}igGV^+pknnK8AZPU=vx5!SZJnILYisv{Tb~95?5=$J
zV^^3Ss?3f9HY_YH;~zZ0K1g`=%KONoGy9^myL<M}H~CyPA|hpU+FGoK$g#V~yU5i3
zI&QG!<HwI*zIrtb9v9(#VId|sI5=?Q0g*_)#$yP>moHxq1~ZUkxoT>vDSnTKhX+rS
zjEs!d-2_-!_lb##A?pIP1_chvRE9l0;qFB~hG$@6Uwr^%pS^o$B*3CV<(wTR4ov6s
z<cE9 at OhgCvA6FT4(!w1WHXb}+l4e7La%Gq%(1>?aX?DTjOQDgG&1OfxOMX{or!!$5
z_;u`IMh4Aod7N2QQBl!8%_1i!_uwZpDeya#>+d)UJzs;G19Mz2ISwr3U3y1d-BYvi
zR}eNqork$oIiH(T0?#|j%slIMWjM at z&i>J(N3Pp~@TI;&#g85x22wgYKTip3C2TUR
z935~ll4yH-TlM0_%;A%`99OTB6zGv72QV;b?{(>>Ju$Zi;x6Xi&vWsr%F|W(sf->0
zS`MqE(u;pGS52v at sh!>2_Gh$l*T~h|=!x2@<>-LpvWbY$8yXq{Kq{bM|A=zGy!>oH
z=gc)8d9I*!{cG1w>j&!?8>1$d_TRdn9+3^6jABSq(qeU at 8F>eeo3CKv&el2gs;n#?
z;0f3{cIN;S10Hq<Xei*$jJVSrS8G$#6irlF(}St08(<i$f`W0ibYB4VAb5jk!Pbq8
zjuL2TX<1oYcSyNvR~QSNJ4Oww>g(tCd*Q{){QL-rYnTk4BJOnJ>-+wEP<i-`-Q9Y=
zV?5<0FM%DrEz#h2yk=k!U3DiVJ^jSPhY#C{4Uu`k_t at F-f){|rgPSWe9$Mb^_xACL
zi5#o%cv at L`q`A4-{q}_$H at H^U)_{gn!2W$77-%`mHqh7ip}*hv$I at o|_wTAS<}(8g
zrUnKPo131?-qLt+R`tFLcxp?_^I^@@hJP^OSt~ksPp%-ed~VKeUmSB@{?rI}N at xKa
z at y^bU75jGt06Y&kf1X%IG}Fn=ANcpb?Bxi^fSmv^BJ!(Oue!V1Ri}9+yKXI;W41rQ
zyA_*9>a|h};;*R^`$typ6Q76+2h_-;jvZU^#=2lv*E2YzxV9Gv_|}#;2RHXgVPQHz
z4_Y#!%b?~H66jM?Q%6Qd<d=kG5*b&m!C+fjW{=met*ynUrNyMC9s~6bPlh-&Fu)8V
zFG<+4DXG>J6Cb~RRl)lbOG`Rt=7HjZpV-5amv`=$U*9b;!u|VQ#u_3O9G2nLz&oG*
z4RUpJi_gf2yK{%ox1Fy1*^oR9Sof>+Ey at tZOyg%#kM^PP(AK6)Tkh at XxPABT{+e`U
zPC}qENA{R$Ee(OBYo<#_g4kt0eVPJ$w!w=lDk+VQjs2VC&pR<a-K8-gEG!KCD+KJ*
zWJ}a-N*y+IgSpmid)1oHwAx_tNy580IqzlGsn(QVKS}`zPD)9o)RL%aXyinN-2U`)
z>0o3?ED+-a|1mT9Rrof2ea1oWCy9wFOaE{&L_xe!m|owRQpGaqr%#1$EDex!4i0J*
zGwagW at pyn_1;viOy0}OUOvl^XJGgB+?)mQY8nBp%*x07NvYRfWABjc9#O}0Bcf~vh
zA;Tiv`rO3C1Pg|-nd<LNhwR*3MfhIK&_yW-X!7cfv^sDU+W*o8Fh2u$%<{_0hv*|B
zcsHb;Asf&@9c|<2h at Hgf7#YI>`(eA4U_ITYzF;ENox68`zbuZirlO+4nw8h(<t}Se
zt=i&bX{=38ApRN{8jk+`^Zsu*joj{iyk}|H+55dz>a%vk_xg5Mi2)z>hw}Ti_@~}4
zKRTLH`TTh{US4t+7Z=pq@$qz9{;~lXqu{SfO7$OhpHngy9=P-FuR3TBBcr1LlrLVs
z90lzZ6cmK%925J(3?O{TsHWIm{{nCW8V{U$mcV+y45&6$jaCs at Fc9GJ_O3y$`$RXK
z9Dosk&XS~2p*qKL_DqBGeN_%T^R=?#u&}VWfB*izcvT__3JUOH%oobfCj+`eMn#3`
zUrE&4T41nvUewvuHM^z-es6T`nz at ZlL~!uFon1?F^AGM74Q at 8JwwWpXZmzEPA3X{M
zhW+~O+q<nd>c4zZFX~JByaFo}v(?HTcFo at 2{{1fXhP|QT5db5-D_3TBsXZ1~*Vm7g
z{&XSI*3%1tkb%wyihXu=7SyhLTna!D;NEq6dkT0Bc*NKUHV!2Xo<p760(Q=Bnwr#u
z12=4J$Ye1;odz%unalcR%6PiIY!hv51;#%*f+sRBlL2`HtnEyY!qR6TZ*FdDE|26-
zTmm#$Z&U}cswu<#N`OlA?8!-D78Vv*lXFs1aZ%zgOEvizPI7XFYO*$c`t+>4{4fwu
zXD_dCEVPifvA0)axETmhkQBvfVqzEsY-}R~Ol{u at acRJTdI+-EP^`2x(`&UGi0!~+
zg8HjJmdNGSHWmk~u<}Bg{Wg$G4BJyvDvX_!-$XY`i(b7t2ulI~{FG4y0&d-;>*(Cv
zB}Sz$Ny2yevS|qI$TTc0*s&Q2*ccm#!!ya03=9naYkv=vh0 at E*3zqG{&osG<ypOUV
zo&qDpF7rs`VZS;%buevVV8C*&y8v7K<Z7mo`}z5;I!Q-EFaZWLaj&rK)vL>}Js at uY
zsX at MdrsDdq1b(C9*el^q4==B-s+P^;N$^7)kCR at Nmc~~-c7sRA$>HqFxp6-yI~)7(
zz#f#mdSxMb?2b9O;d)H~>|TJW|N0DF at Fyz1TVg5cgw*@O!on(o{>ow}1mJY&E&yA{
zs7VvJwQ^J9YW}^eRW+^u%lU6PK~#1&#|}U)RirW at gi|6C5_TaW+S=M$%i7ClAubMC
zO965>Ha2R>Bz_uK02+&#CO<0O{;p?mGDKk}D>!9krU~$-kri)j`UM6zKR-WlG_=`K
zQ>pmPn?%R^c at Osa`1q`k91jQx at b>lX{Iv-GVS22XjfEaC9_;3YIsKr^@bp^n{q7mO
z<2|LDd;ho09dAt#x9jqW- at h9`?5E{ppa4^FSM2TTqVFv-!P1NiOP`fSozH=5VunfW
z at 5L&GzQM6(e+vji00Cf>-QC at +v}C9w!7edp72?`N&Gq~D{n^Way at 4OgPpj(a=(G<G
zQo%#+y!+hRN(C`3n`h<7KUQZjdT-Os)n>hU!vH(Fyt?`k5zw-U5^skO6-nde+lAO;
zA2OA~krTm}V9LJQV*zZj-Zd(d0HhI%=BK}=De36o2 at z(kSMTlo94~D4tZ?`8!n=2F
z&D6S{=lYr1*&&r~(^&t3?*lYHiF!rQ_WzeKh@}wNv$OEZ^OEZy?g+~)XXfRR?(OZ_
z*x1M`DiZn((l(H-%;sf`{a$(x;jJS-z=<;=ljG4Pb93|C5ZUvn#TTPmZ(wKit?lWI
z$yEy(8JUR4$Pd^U_r!_V&z~=WV8&kbn2Yf17t5NIzfO5xM at I+76F~cpzP<y3f`TC-
zA*|=lF==XQs)tJyYM(98lQ9G2b#--RWr&*eYynuila_W|CO at XThw>#bA>Z)r+RBpM
zU4M5^&pV&SJ3NBAePh7ajyS4P-UBZ4qrzoBJR{l!@KSS2OQj4yLPA24V`QD*zKPv9
zYM7&Q>V&knd`1wqc<SsV1YFzM9iN;$?|VX}<m*?>?7X~_5)wx)Uc9KPso7K?PD8M-
zsi_Hg|3l`3KY?@1Ns~mXaws}F7X2V81n~xNpXnB4YHI54kW^LmJK{*)#b`lbY7sFp
z=iS{!0p>4Xxk3y!h^=A34Z$o%*0xK^%1&-X?Z*UeP#R!5_d#g at 0X)ROaHrAYLt|rT
zt<%QF250>dlI94TP60kXWr!K<Cr at IO?C&=MAlM02s;LB07f&RsRQUWkNEJo1eVp9f
z2#~Ntcq)7SIxBY386Q{jJj%q0l~$9V4`Zf=tn3MJIXsjuE*8Xe{$!DxAHRJ|Jj5qM
z8mW9|XU8``KOf~4zJ<TP^w{_~+aYocJ($1t;SO<<$W7|Lf-USB#VRRz6g?@1S$r9^
z5RfS(35~+Kd77b^eV-Y7N<#yixj)UQDkvzJSzHW at iJ`(-2Y5R$Je<LILiSl!7CZ6-
zamOE+r7lB at aJYVVepz}QLY~Xty#Nj_E^HQurR4qn{ZDgph=E{WN7=HQZCo)icU5^G
zvCGMsEgGExzye}2;#vuClIwL1Oz9&C)=#ZOWh?TarLwap3aGaKy at HZ0_$?<TCG|>m
z4=ZRh3k&Z7v9WA^z^NXt6n&ZSw!C9zn%%9#WJLQ;ERolZI{zo=_5sXc8(gcW-5C%}
zZX at yor@*WKOP9#^VR8uOL{@tX1Km0j>uR&mS4O(@{_oz>qVk6uWo!w}%Ro`Eplo+A
zE54~o1^dF3mX=od{laDd9bH}f&z?P7QB~D)SQ2P?yhkBjKFF2i<Yd=IWdOpNg#}#K
zT=itdVhtZ(-`H72PYY)d8`v%YGd;oX>6n-zz}v at j9-FdH^WR}ogir_OoJ;=>Xf!h8
zq9OtKu?(Y;FXJn97NuAM>JRESXwG9te>D<x8kRhM`^IJesXlmuK3oMvJf2r?%w53_
zw_x_yOAVYoJaX1f;yKIh+XT4HJv}|Ht;_JB_nWP-pHH>Vnvm0wG;jikA4TR2hYycp
z72RxCUS?L-9e6y;!FL`I$jl{QPwRjZ!Tecl;iAg%5$m9_<{>dLQ9WF2uBW^EdRw~8
z*2*UqydOn$+nDPiqozLL_A3B_85spdeW~Rq=kX>gcsa|3{_^jyZM*UZ&#u`5)wx<^
z92pTo1{Q?<3uo_>-PsZZ-Fy1<>FQrYep|io_ZqR%Yz&hF1Z%h0zuywe)|nOgbAmej
zZYTxgSg%<S`T%wU1sV(Vs=nj&Ki&VA3otN{;z;J^?tW}+=7LzE3lKp{Z1a+z9}EK6
z)MxVW3#0OT-DA>s3k!MQeG?TBpoTq!SkvCq6AA=ICaHCB at HlXW>^Bx6 at t1OV`1y0(
zu6KuN0plHUKIiQ0j6EoCd`Tj}k~adnl`r{zt=QNGv;dGz>;~-YEX2TXgjo`~N>&f`
zc+UeR3D|3HZN=-|L;;g~u+;TkT?w<apJd`yLx5u|U$}r-tazsp)d-U7Vap1ztxmpw
zfIcx)=59|+SYb+r^a&b>wC;btug`W}td|z%kd%~k#xC5-$|L>d_E=z>o67~yb{o6`
zn*sLpc>I{MCb=umiv1+${+DMAgr4^i{M%vdNDmJSoO|}<ZWNPkXHiiRjxnM=emq9X
zYw>ArE(v at CL<>YE83)T3oEA+7lhwmMepJlP&c;?g*xVp6P?k$GRrNi=*C%HYO(z;9
zFw3~%(<iLKprfPvr=<n$8^_H^xJ}_*6Bid3gcku-j)T6wz2)}3w9L#D`;sKwxUih}
z{(ag9Gk=R~>xn=V;)RX%=R<fhC35pCc0^A8BKk#AN@{jC4?qs1FK8tIRqVb`L`)o?
z*0uvo8M83|@SHkz8@|`ndz2U<;`75(5H&cWCMgpx<)ChdxN0&p(B8fen0WQ_M+zK2
zf1 at p3gW>RA2*5x?V<TpTL`FucYG{PYvj&5~{CA2(B`z)w^!Xz&SM{ainDrh1`0<6M
z`IOYuXowkLkxCjGSYn0OnINwK7B33_$(Wd$!qf@>l}1MOaK&h6TU%SaG88a*7gAj<
z&2;oA;B<Xm-9FfshOF~ZNz^XtbmXBJ#9 at h?L3`@y>(_U5(6O<xVb3{u7Wk3&R#<fi
z<868q4S<OA^U-e8Upqqo0KilP{C;E1Mjl>&8|)e_*2&2UZ@&8bpS)0ru>erUszZqG
znAr=$9%zv~C at 2V>I2fCh;(^CLLhs#F3BOG~OyUGJf}Kfw3a=QH+x6u4mBzC+Wx<+&
zD0uer<pNFIq-!M(2}qamD{-IgB#2|%Zs>@cp#S?WEQ|yb+4PHyKfJv57GR77SUA=T
z{BJnLkv*v|A&P7=LUW%VorY%$!352AF&-W=Fh4MqiF;8|uxMr at 4UzOEuU|KOdwH$Y
zqVb-vSK3ftUnIte-<u2IsEzi;HDL||Z{i1M|K3SWJ%$}(^0YF-Vq>Xs^bsOlXZA$^
zFg`mwJ1lp{M17lEw;Hf9%m-4I=;W!FHGaP=K+)n<C&K;v at ggpQ`_|yQ!|*vOE-u2S
zu5(3j<I7-p0ili=Zr83wL4d|;KFq{65KEyFcpL at PXkQ%Z at fNEySPT{PdtkW4du7bU
z&0$R+7!B5)<2f|%>C;#ej?59em-Zc+cn?$o)jL`rw&=08#vItVIGX<s&JX-|^C5Da
zcjXuep!@%&_P@(5EGwf2&{|&}V>I=?-EDgC4}Qx(@h9<HWY%v at etwiB7qpkuci4~T
z5CH$~_8N|~SY2JkcE$Mp-V5cuKto&Hy~hd9`mcXYlZuFrRtD(G);aa##}8Us+Ta4+
zypI5{slzH2z$XDmuz@%>Oa#gVth~CW2D`DovrU<|P~W`*bg9C9_Tb0*dKY+~_07$X
z;Q0y)3Wz5_5NFGn6c-j2aHtFz1)fO&tbh}+!?r`b_8tvoR2A6MVtT{1KuU9h6`<s?
zfQce#ZXo|7KYyxesb@(0obefrc7^DSID`c}&z@=P=n!HbQc}{B92`Uh1O(Xx;RK)=
zF%Sp-+0X}_{b6u0%`qDuh^x4Y?rBZUaV!%?6r!WM4I7F5p|N}r4^4nX=PDh$iWyE`
zi^nuGWMe>KVWQpM?sk&6(+PkJDt><cTqTYjJT-wAasIR8f%Xv-6N9Z+ at gDwf(B at Yu
zzI#6dez!3sni46{ejq$5>M45kyn at Cd?aR+f8AY4&wJi_&t;RIj+^Jz!i&jyN;lD at h
zr!RWojm^AGLs+mt&En7!%bus%XTobYJa2eSyP25kpD9b7PEVdrT}vH$TYNtKM|^Vf
z#oxb{@&M^|jg60;Iz$ft=?HE*^uI2+yt#TTN9Q7VV5a4#*p7~87iPyAqd4MKdHDFE
zch+Z>AoA05$tRArrii<5&KpXC)uc&!+{0FX4=&}*b;Vz)I`*yD%xor_E}u43QQ`02
z?_yu>rk2NoUGOUxGcR9$rV_)$!J%ApGQGs|Mk6JQZ0duAgxU5?C1W+Z5w>E}nyjQG
zrv3_-lYiFQ<POo&%GdaA;2_HbQfx=*_N_xu?_deqRRH#u1F<Kcl!TP(s1<|5hxK)L
zS1+$TFpjjpZ<NBRIUcvnLa?*Cc{5MeM1qGW0vqSudTL1^?YF5OU^j|oJrIknKxqI_
z&Yc(fsC>9(9H{DR&jlm=esC#mcpRL>eCXIQzHT!@Lg|sdvYV0`x!M|biK-m#KwIqE
zpWYTqw;>=KkDN?MPuIl9!g28{<DZ&=mrHsrzK|_9OnT8Im at XS|DV&BS7s9xt?>f)q
z^fVCdG at H6$g6^`LTEKS4jM$$6^y}#99pX1A3HJB*pBZp^pL2UZ>63sUCv9yyhrxG`
zfPSli#1ZwF(|dBpn!K;S-<3b+7Y8S2#|y(Uo331q?xza-aY`%|lTe*S%hsJcg#B+E
znE8z>?&DB_d(qL^@CqsG1wOdPzlMV{Fp|NOOd7=o2M3d+OZzB-iRYqm0CTHMJjr_f
z%mh3uI8;t{_GOUPCbj;_dt2i)IK9@;&`=u~9XPxuE?Ze#c(QVVUT1Rxw=Fyimfyd6
z`!-M5vZ?)hsbw>88G9cUPC>!Aujw++9d(B)-NJG9EkN;=xX|)jeGgu|cu|qaTgOgJ
zOgu+C0J8ZR1iAah0(n4G=K9M|gFe#+4CjACWj)LD=#r%K$Uc6PcMmnwBy-X)v=^Jn
z6`Iya_O7bOhJ+CHzqX~vC4;54nQ6~l at 6k8wkXov)FBdI4dHQq|HH)k|uy^<65e471
zDXOzJ?ZIHp%Eu+$pN@=NH~*Ru3GmJ=8L;!(H+6bA=r5<fzW#<nsL4Bbx~VlXy-SXc
z0t<sxkKf$tQ at jv&f|Hb#l$?qx2ly)darpBA>Pk>NjZICRuzxdvq7(D;>VJO!$^&iI
z1=_r=TzaT}j(q{2^OZFPm`ogIO5oDtN6^sZ;X{P0h`<_1y>=Rw?SJdU0xNH5;!J()
z-2IOqKZZ~fF<jKp$l3VWs|$K{ur?qC=#9-t-EGn9b^BI-{hBsPmnX(W3fK{CY-nf)
z3_c|#l_I;ltaMb^f(W*)<(8D^+UlwetmrzJS_fE0GbmW!UyB<1NhqIgE)1kBxTP?Y
zhu-;`Dxoy+#^ITa-{xgsU&(T(;dIcn1!nI9Z`4F-nwzuX3jnIlf|OBK?yhvZIoFj(
z6B-_#5EVu7iJX{HLEy3Oty_G+ at S|c+$XsHNk5E)nvi;{i+o2OD9OH#!6f>W~PPgM?
zZLN&spz8;!9P2}VOWLZcWT7D;mqCk!n8Xn?v2k;U3z$}`;`m~S=+C`+Zhd=k#AE(@
z{5!8Df%&HV+mkJib-{;Zg7#AFPZ3Iwf#ton>*1AjpP`kIl<cg#z5mpiGY=qWS0-}F
zbCO~Y`cr^zO>rz*0A*uFPlkXGB<yT1sOc2wnP~23-KVFgr~Jr6P0VR175(qHl;<HX
zR<g^W8js8P>eY~OadL*j+W(7@@M|xP%Eyed_NhVq)P{F{X;>B(%Oa;)TU#q?*Uhst
zk)RbqZ%0Jr-BcU!J72%p^wGDKgOMQ+J`Xc6JkTlNU}wK}?V78z^HYe}=F%M>OV(GE
z0hC=_Twa^klK{s#8Lz4g4w?sGrE6f2)R(4ts|YW`GnK-zMK_u>p*)BgQP&C8xz3!Z
zW+wq`)pGQ#-JzV^+)hM_k0A#b^u;h8;gxmE3>ajdT~A at h&c at FIfKKC<G0yGnZ5%d?
zqqA{lcHr*Hg}LGP?@J0g|7Pm(&H{4efs1Ry8fc_TCj)CVcOB0M33-T#Nfqdt`%F9A
z=x=he<s_haI-Z_luxQb+9fVw;K=93kvjk!n$PT}ynhObB8i?<P#(ZB$3JMBd`)=47
zmD?T8$;<1y&!#kkvmdv2b^zwiPPj^dCJ)Dgc3^l96B3ke- at bk5=+R5K7$BvR{0Fq_
zgdSev{Iflsflnamza_-U!GY78J?6S2eb;9cSOV7Lo$X!`Kh@*41qX*cI{#~N at L5TT
z(9X`!;!s(#uz#AHGiPROj9ytC7<?a;0j_4VIp5d)!ce3rCs`B_ at 3noeK!2qhcbc@%
z1NbsxQc{FU;Mu8w*%gUAb?i6~J?L*>f0dVDnuJ?egN1Ma at W$>cFDG;hiF4<i`L%;D
zU|THExEK94=QDwYl2cGTV^a#d43DEx6x4kXXke&s43;MpI1auOz4cvaXTGd0ANI+(
zWt=lwuu~x5s4hg)nejL|1Pz=NosdhU9swp*pmkUVf=p#)Wye#6yEe-sAG)i3+=?cX
zser>gMGsL|r|cUT=z_Q#>UHZQ-ev(Z&AuHDe0G<V=E2*y;sFrT*JtxGp0q9s#j(ic
zp0)qh9!AB?BO;RUFF=q|X9F~){AoDtmRC}8KlYe}qOh<qLcXh~XKwt-3!Q)kt%XW|
zfJD3w{Nsxu6(h)7$H9ESjqVyhuQDl4r4fEmv;QtuZGco`V`F at HIB5)%@Ew)dW44=l
zTayzL?eJOMQdVHOI<R2{YL5jD9XyCJ<FIEPy%0HUi2L5I52D<03D=M!lPVQ3*c?DR
zU1en=kJ--i-&unrAmFG0TU_7Wb^|UuW)!<GoaYcZ?qX|(Tt<;;%_SE9WrCQaXYaAI
zvxj`uNPm!#p#|XqqZ*DUAtoW|e0WMX-)m`z>n&Sj@%z0?;ExlFi<<D?#K?%er>7_W
zvAMbFKK=Fhy|}nMP)fXfeA%80{n`cwOdB&DO5d$CpDKdIfJx?S8P&LqlR)GQC!u7L
zN4WJG4tQ)INl3Z8vte&k<;f2)a?iBdTOl(uQ}o6cCNK|ruIC5HmLG!ObyWMTMMS9b
zK6+rlmkX3<Bg=u2k+D!Wj~e4xp?<MuTZSAyf$!FGrcgSMymI2%bRJ+tK%a6z-{=Bg
z_gMUqOhR))9Snu_!i5V(lLwgOZ`|O*paw{%&aIis8Wa?CROALF6^rbT2{CtmUf!%^
zk()?RHWmj{68W#)Soxz4rvD7sP8WcgO;hv{yahFn*A(8{UDG-$>wk_Ds0rv=+uf~|
zkGb}f*v{hg)fZ2f-v3Rv`}X1 at v$$h6c&;<Qntf9Y^96JT``#inf?ii5A|l=qkS5*b
z_WJKUelp;Ud3y<fkPkp5ajv at i*0MQz4FXY>=R!rIbt1wbusU1d-MH5~t1W`~lU%)?
zym3E%7$ac-%ZZ^rE~yNd3wxapj1A)zaK~u??Kp7VOr&Vu%d88-?=xVxJbr%Ds42~S
z`I5i)ATiU_@@T!N_2)zH0)C5~S5X0-m=KbXZm8&?vmGb5r;6j^z+>aGva-I|If at MI
z`xoZm)ogo<O!N#4R7Xcg|HXm+^5-Q5A)ya|Ga|72{c`MoA0&r-DnmBFW@~x0yU0Y!
zJVB1){FNzz|0vFRjTEsQbgW>4{r~jzJhJy8G+J$w``eB$Jl+)ZFeOC|oHw_)cIQ&W
zl}r1>&VeHL-Iyajefo68yLT2pDqOYz>60&+)<xrxU~Jq%OS_L<5q@{2MeseVoLq*z
zEmsRLV{9Q#ONU)s;qBRgHP{;xQa(sZiU7kuDj*<$GrJ*9<H)Zf>)BR#V;nl9%5kte
zUwc0e$^f3&`-;e544`IfG at JzwLn4*$;502*3*(kE5oI^qvS6oaqZ;w#1N(ff&*~Og
zA7DzxM(I;ul9x9(!wmRT04|)ty^E}cFPWO2z@~Lru-D)J@$+YLbXlC=0wN<07K3nK
zFiwN*du2V-T|f`<6es(FR>OAMPoD<c`=rLl#}kwqL_>&k;XfaI5;hlGn*r^UXODjx
z^mh-Z4!~o(Pk?!<4A%y*#jE1T1Jw1Gavs<llHeE)?0V|x=m<|vcHqDVFwFajiB09A
z1oTs|4_i|yPM51Z9Xx`@?;SpUKR*5g{EG~xNU6k~pu!Pc1v<pQmhr3)zKMwFH4bKA
zWygUQ_2D$1epI at xz#g4sV?!~zFJwiTU_rQdrD3`)!xHwd<n3E5*8#2)SLL<j%C8pu
z??eUgKP;aD at ZmaJ`}oO|djL9^g~cp(Ys83p2|F`kL!jCKzf75^w|_3<1Rg!!Mp71k
zTroHPW-C>7b>&a7$JriT62AHMSi;P&$hf$01HLAR-W7 at LtgKIoFXbHE+uPe(tXjgq
zsPj8>=8TZ<x=lo*1<qauDKlpDpz(f2MqDTb<Hy!kDgfceT#b~9L at U|@2M#a+<4jFW
zy>4wC3c-wxi;D!fg0ULCkoFl75p2Z*0ocIM at IznUk&`OXcpV1ptZrXqWF#3C6){~V
z#5A0`U72Vl43zeNqh3nhylE02_P1R9({?K%KK^!N6k~l`8;x=Z{3Ht)SE#YTw9%7&
zZK`T&^{^YXOiX0#IA#^+oi!*igeYd#eS3KgM4rsXvETIEnt(!e@~_IuN`Zn at uYHxG
zjZ`M1p%LHqP*AsMXLkryyp3Z^!3$x(?0c!DJm=-XCUtj+zNAJm3R-~E-fy<TM&V!x
zdFlBlFH3;a>uUMVeJ?!;4Df3?2SH-<=g&Bi1RJ=4WWe^2PD9nBMv32W at J7Jz%}}s;
zYzVygBNA{O9U>Am9G+0MJ~?|Ab_DpVhU01<NUrV_3p~CM6Jt1k{=9L^!I7k_EOs1Z
z11x(#T_#u==9Klmq8x)S#F3xICMICfc!UBHJE72xh%W)V3%4ZlW1D7W_S*ovwY#fo
zQ42)**WL<%6V{hg5>Q9P#a;OM?Ij+uS~QShB9{F{LXy>y5BqVxiA5>X%WH2{{QaeS
zAGj-KULFDLfISOU<-mTX3<AbHx7<1E&Vl)?tgJ9WH2E(NG(@m=ix8;JxY3hf1{|h?
zV<%KO6yc9j*U3v(*48fkYVz{(q at Po&Iik7AneItMbLVDf<M_ss^77EQI2zzir_ at zm
zv4SO*STvF&^x(Ktki|N>x|NAn1c0{SP{+A%FHfse76qq>JAE7;X2H3u3EWy0iS<VS
zZgGyti<d8P6l8*d)faKV1b`HRWc4rzXshpz>s`Ja43PB<G at h{<9W8D9ZG=DwK+{HH
z!TX9OfExfKBc`FDX=rF56Q;TVk(q>^OFbC&5%`B`t$%Yl$4d|x4ZxOq+`^BFJ4R;&
z?Irg*$PsjbZYO?z^`#=PxNbKluf-pyzL<+Dz$0L*Jc!9Stm$T3`XqQ3 at R2xdI{e`S
z0nT-!mT#8_jKwp2-YbQ59UXKyV+vGQWg_QwcDzi*XSjQJKkWI+!rNi;@ZZT2;OqLu
zjM!H(P0Fhxk)uwCJd_Yfj?GYYrEhGZ^HvH$)azgKeRn}^gI}-D_t7&77+=?U92~kb
z-zS12)Pe2bc%k{eviO9A`f|RZRbVSP-II^u+4(4jb3i(K9}q>WYid$s^FwiQag01*
zkkyOF5?dyjK!N};W6!Ahp_-IlU*hlQ3-l}id4Lnj!oM`d9&h=5z~k at kTAZ(oC;UD3
zNh#tyu>26VjGco+dEqof1b~`vhZq>xB_)k&+)QALA(Sn{6BgwhAe%WqOaB&|ZPL?Q
zYBM*i&-G9MyBdMrqQimSV4Z9SBWVvFWT!lq%%)#ttN;R2nVk-(ZPV{_MwyqdgFB%0
z!{14IEk*;+mCwBF%&$>Lh$98@!bVO`4pt{5JbWDV68Ig at d2MYv3k!?xgh=d`{r$T^
zW-2YM&`AVuF5$+N25?Kfn23#y#VlNGkDZdzQqD&Zi at D(sIMqfL7pb|qx%rG<M|dUV
zv&i^{%=J8f2D|OOF?YDv)w`|bu^=6QmzkLvSd0t*rx_fm;IlTBfA!_juNiV}<;N1H
zA#&ghB)|?Q at m7;k3*uwPkMnm96BDT?1KcZe96XS6gPhC;Ugr8-cL6}#*37fGkVbat
zQJ`nobs!~$8A$xc@>FuP*1{{#N5;qZL(q<s-BCRkYEox|GiCq)LjWj>iv<Jswr at r~
zp|||}&!j63zSJo=1bUoNOmq^jLddA8!yY|4jIA!qkC`q{wKA)xo}&{J6T=n2AO~gx
z;FiQ=Yis*2K^Pq+U}k2<Udz=B`{~=`K;IYAuz}&6b{wj1M&#7yef0gE$7#FXA}Y|w
z at X(*K#<kNJZHM1^P=Gy$sd9V-Via<miin9*M&`KN)E7+a^A8BX2{J{TE3vSZc*$#O
zYP$93*KlZP5yMx21R>i_4%HkTcVNSht6o3%e*@z4d$muQQ1ey!7>E$y{FrCP!oqS)
z(mlfdqBU;X`fL|lLUa2oYc(7q1#rlPE&V#O&Qttjb;f_KRlGW#?+JKpxqWXIFt_`c
za&QKufNAwVu<ZmYmEJ4tK)opo^vJ=OUR--y^n-rrOC<2vO!&WX3$4&Wa6EZX03)-r
zk+9~@{1<|)J(q^4w6wIkK{LF~vtyC<XTp%6%>Dr=X{gCV;&V!{x4>G+C at DiAP?VV0
zhYVixkq-=%#qKFvzke(7KC)<ks(?8b|8fr|WhSjd*XebSXaQ((Ve4swYu7A6?0#(S
zBea9SjK}owLa!YMpNOXfe4cd0DSyW$-D!8PVA2*gz)^`gsvz*#jsKGk*um>teFuRH
zPERtG7px8k{T)s78u|bl|C3bE=YRgektBGIP?SSPMmD1gP6ObCu@?Vg`QH5eEp%3Z
za4ci0NIZG at -isG}l*c6xUA=nM;C-M>uj^+}?O49s{qh>)nPf_YJOJhjfV at l&E-oCq
z0|6L5oOX(0Iklldx%&5F752H#QG1M^M-CAjKuUo<{`&e%2SZi6IxN_qoedm#QF!$w
z-nUYu#E}cB0D2_~niXvpQ!$H?{Z#uB48U67({p%Zu7_TX|CRje+8Pdv0vehHGzW7V
zdLNuq_<6pcn3xzxAYn`Nv-VtkItQ;m!a>Jp&KzWq|99cIG{+oH?1GM$S7ws-E-9gi
zmp_0Pdw{s-BqUaDoY4VB!G7|jf-=sW_&1Jm)ajbm_=euzPkJX}xtTj-`G$TSj&=a1
z^Y!ajgHnso_I-!ikHEWOd;l647#O(qv-o{Qq8?$6P9*FHCewcV_Kiu#SM(q~JwyoB
zI$jzUSsY1t-N7LSr at 7UKQQfGSMWHe}GExUo|73|x2Rrx$z#K)Mu_0}9b2E|zV2yxZ
zm5I-$fOF3Ls1%7H1MjFdyag1ozorMmD~4cUOA>N&a?oQg`6R{x{{E~9&C8gS^x-Zk
z;3X~Xby{<2?Yg98A8D7Bu|}YAQ$2*n1bf>4Y;wH|lJF)pvRidy3`a#m?QZ`LX9=|V
z_9uoHJ`wx9f+ii+yUKvG0D;OWDk^%PJ>viEh2iq*D%O4X>VRRdU;pPm8;&6BJwrs~
zeb-pv6Br>W#As~D2NLq|@_5|j#INoD%LR}Fcr*vPW2~m__VXJ**t=1UM5a)$N-%be
zWx<gQKuuD_9F{>Cxbk0oiXCIooN<aXqo66c=Tkz^ltd7-0SkNBsR>Vg1)Bkx?gncJ
z(5HQF)*$#S4uODm13zRJ61seD_k(hCTigFa((?9n21P(P!&kie1R(j|*T*I)`NUpA
z<-+sx{aP8a=PzG|_diy0qxGnH{Da^Qq-C&eql=5NCojZ0^M{LoOD+SJer#+c4%%Ho
z1?`Q;)U{h<6cDiZOskW0`A?lXMMh43=lS!~n at huO1q=jT<^UPa9v=Adot>TMdJY~x
zP6bfx{_`6ZX!~)!sNmT(&}Jc_p*Y}GSXkJ&g<$y&9qLnjw<IBjBxTO)F46%zhd at Yz
zg`EGrxzPIN?$Yk|8Y}xK2eZh{rsyM>rT6Am at _BVa2A)Wrf(8hCtnsSL;OVaXr0}86
zo}MZ=_7KDUBpI0vhnyS>Cei_fLog3bFo^+{K>~3e=lQ_en12j805&jYlnAa<Pc|M&
zX#62>r2#l|^a&a4IJ`F1iuuTSzC>GE at F)Bxe4mi#yunxh8?&7`IA4|a*%5giorCE5
zpi|xWLp?ByQ&3RQ)FIZ~Q95v!DLp+MlRJBL_U!{XhBG^{!`r`th`TT*Ak_0oaNU6H
zSnP2rV3uYE#`f&DhHH<*w?Ee}p3)14_k>7pe)DEH=nT+ww{>L*2`5*2&4MyPSA&SA
zx9~oJB?LIumXD7T_^5FUUHcvELp(Tm{4NRAtEBC7*q$Ahq5qA;>07eA*G2W1N3Z|_
z$PxbV?e%qc_nCH-q^WUj2bRM at E~ux7Q32pkFbYf<#cCJ0$8XMd<)NVf%XzZU;32 at u
zPHP)*zw&%ST3U>GeV76t!#JqUX`|Q<AD|q-Q7Iwb;RHdIx5WUt_Pqi)v%0H`9-tW8
z!-L)7<hzDXpB``3RBpYQlhFGM45PZbx;;;eWf#J$am(}jfQ^8)BY;Iy4SN98Bh%C2
zpqIynYW!464R`#mY{pRys9Os7T-0A2tjgQW2&Z9j&BEc}Ku-Wf!i3~)vzCSb@*E=f
zxN8YUlse19%li>RKL%g$5j;T^;Scsn91fmr(rOfbbaMw at 1E~`5cf98sY?yuhT9lSf
zGxePLWl8_7^f0 at TdXbV3jAs5w@qsoRZA_O)95|dK<<Jj52c(aI&){hpwdKy>=E4Bx
zeD&y?9ceVRb#-+WvT0|HP-aIWQ(0M=I<t at Wmu{`aymTW&ZuZ9BpWisy2HRuXc5xL=
zCh-Ha!BiEjs0X<NdJX$*+Z(3Vh3}f%3im`G5ov$t0VL?*D0~z4U2HGBavDCrX2{ms
z1K5m(h8Ac8QAg_f at D*j<#a_X$u!KT(-8?`5NFn+G(As=2VdGLy3YPo3n{YTKH}U=1
z58*2f5J*G9!f at 7nIQ8){<-1&8Mt>U-l2RW-Jpk)}W4+5~arixrl$6x74Z$Rlfj16!
zYieo?1ve!;=b5r~3O)qH9 at egwrJ<tYcaswpeOS7ACZpw_ at lSE}pFfjxYo$*Zk;5wO
z13KEIxheLzLZizKk2P at OJnY)JzJ~jPvJPJtN<%`3{I at 3`<5)WExZvPG1>g<utuozu
zI>|21wjKhmf^vo5ucE$*5BG#_vUE2*Rk+Lh<!{(lti+jH--Rlws_;hk#X#r7VIZVo
z5fNmm5^iB0zo*czK+8Ih)DdXTzX!=hM at I)N3y&w5qN3I_uf8;5WjC(zE&oyj;Whsc
zKL<zlmp`ZVi-IFAKflUfb4x_xs@>v>bT7yV?4yjOV{s>B?x&}NN{`|a5ut~$s0gap
zP|jSh%lmdz6ou^8$o^xp{z+E`CT)eS9e-31+`M^n=vGz}%I_p0b3!m4RL_SF9dhPx
zo)97q(o$2aANg>1bZY8qmW<AjhougP$w+$cI}teI&wQ`;0G{=VI`khpYya)gw<qK_
zRv>=?DUn0;>O`xA=1eyya3g-;NKz4zE`ulQ9}n at Ck0m`Zl`ZjHFamOhV?XUrsXWWi
zCp#|XiNpCMBu3AiJo)U)zMIwwA+3#XPy3plK7D$~>j5|!BT#Tm{Ee{_my%+}cZA)K
z at jduTETOOO$@i~`Z7FJ|oj8yWPo(zt_Eaxk6fRr<R*K+1 at YYGZ&_rys^Kr$i*6H0c
zkDpin$9GlU@;`3r$WdcNXa;z#Hi)*1oqT)hla~72NQ;QAlDRn>gP<uTFn0`)^CY?8
z7gR&qsWih~m#O@{NpkzDeMD~Ex)m+CGJLGXL1Pxv4imIw3Fs7fAC0{o4(w73Fa5Jv
zcY8ljyokt1;U^j+4$*Xn4&mhSBLSE7>6_NoxP(2^Ehuz5$*YggqyPzNun_*2a84#2
z7Bs7Ez-}5dGh`r2F`>u%&(7K2SLZ@$w(B2Q6ymHM;7H$JSyS}te6CYIVrH0 at X-mkZ
z>#OHtvm#7U8_n(7SKOwzhwM2OVDkihlJ0h0k9U at 0raTD!LjB at c``)4>J)YLfv%iI6
zDrE*9oz_SC9)I$}(H_V5fq!V}=*;y}rqn^e;<Wf8>ykV0y8!ao_44XM{^}b&-q1&n
zu0*;_cX#V;cyN8*SNr&+ZmzllZfaVRk%0k9zzwc2_3P9w*H&Kh?(~-1>jQTWqw{-b
zGj>qTF-}5C>Zf}lFa3hB!zmt~V0Er~U at VSeJ|)s$21(W2cDGzM6z{EeiZpnAi2<Ps
zyEO`YPNh^|I}mFSy_cVAyC3Sc_>3KWz&pc at 1;h&n^@~kWmmrhgUYm}wyGCChPD4tE
zBN#c07#SEoe)vG3?8Io@|NT3wpkQ%>6RPBYXPmFUdKtxXW%HU^#PoJurQ0+}%DZ)S
z^3}c at _v|GYS{|DPMv_n+Wy;wt%wMF7RK_~rpGPm*3wr=yBS1q0bLI~{{O}SdL at p9w
zS9ofm%bz<ha?9qXWGi at u&8 at hHUTGAb^WO>l{QM|cWPey_mWZE+- at u9T@Zn?&>bJ at J
z-a4^Zok?yymagqM>ofiJ_&F&l%P&d7T-p^bCP!xO70gajl$Mr)X^{w?5jxDJ at r20B
z%S+m4HH7{*APXxy;v44M;b0ulbpWxu5NYsWwg0uP?wHe;#<yy6SWkwTFrX!mD2D*5
zyX9+2VKUJ(G4e;3IDLODA^@7IzdM^K_{#c|@#=%xfBw@{z at b&5lmh?$@dAHpXljC}
zR66pnm;Kr;^I9?mqV>$v8wA*EyB_k|+S=5Nbdw{wrhzH0yA4eZ9J=gmYy`5hvN#ac
zkiXk;;0-^Lz9MU;=u<m~QgFN;H8nMM(C*byul*sA24q3NcXR##`<=A3G=&7qisgaB
zr?28MlGoDE*?TpZr$7Il4&<xY=Sceg>xGk{1#<#11+t%VHBx{_@#VI~YnWDh<7k+=
zZ{HYzCmEDnCs+~>F6U4`8#h-Eap1a>6HdM;%6XWnaF+x_Au9PT at dF1A5YUYR2eQ`C
zx_XrXA_ETK-nV;HYvPqzMuAgbspSP)I-}P%w{`P0v5aMVdmD4C0n$Sg`LCUMn&`82
zwbofUo$2slVu0Wu6)yEQV&$Pu!{Hadtn>lg6WQC_zc8-s<k`e97pZ&);$s)jCg1I?
zEpJ%Ke@`$^MbjTY9s at xI#Ma#_Qiato`)NHu3(olg=<n|s2rW)5$;T03MMXt;>~Qla
zniwf5sS4xK6r~@zr(R}dv16eru-vVkpT*{umiyTiA>c+coL>8o*IFa<?Ul6}PS?Q<
zK+xbIr*YO3N at V2_&xQWd+&p(bb$y%>2jnb8;{+d#+piB}6O)sfj)U)*&RD62&K{YQ
zno_)RgUjyQi%6ffsUxkqxx1cVy&KhpmT%U!FJEA~1SW?fq}1}$(bgNFe~ilQLcv$Y
zz)@AouF<84JCV>m>bvz5yquLCbu&m>BD&G}zB10ht{OitTDq)f&o=%aH&p~iJl9(s
z2g_X=c&|*}@ZiCNcx(@nley7!)<08^cV(iPp7Pi^TICQ>|3}V;bMfbX$-bD-USh6P
zCT+e`wN!gIA%R{{j9u6|!(2(K%l0aGk8%ir#^aeho+U|rY{m`-iSt`QuYETEnkoVO
zzv(V14RGu0gfEQB$vr(iF}Jc;$KZlxq=}wh2w=+73vnl^d92Dmbn{y2cC~PeIsPDn
zt^QFPkYa!8h3jl5C+u#jw9f<j-!3jLtiDw(&jptHjs2^tr at gmVb9`Ubo<af8{an9|
zC+1^P5)+`n_*82O at Cy>RYkU at wm}=hFcLYMB<=|C2JKi5ZexMBg=5T~rARzMeK<@n@
zBt8&8vU77^IjR-tW%>PD%-7AMe(>Od!P}cWxyFGhzPk-m;F6;t^zu!r-drpXJ&gwd
z=H}+cEkaDX8>ilo9XNoq)+-VT>Rz`jbk2fb6OoeMjlfBdQH+9T65j~V_B>~l^q8|c
zbge5-E4->o%89{df9^HYq{79{q*b>UrZq=_ic6+>axDM7c=6%~2yNKW%E`$IO%6L2
zwH)LLJvFUuf;Dn!X~)6a3swR_VbLBw916j%?Vrz|KTe4I>~0AIZ<(K;UpGCM^D-SM
zLOq~4h at I;H^G92IdkXfk(4aX&mKH|KSZ*#@Bpj9eYteczQFeLdw at _%M>;T}xbxX^;
z4GqfQ95j-I2ow|)UJR6q0y^FW#^cI=*990GNcpYX9^h;>F5 at b+bjYYSYBbX%Z%32>
zX^t}N#sjDBVzUg9-<g at 2OtSvxp4mPDXVWl(5EKeF6+wnAej9Q#pL}o98L+GyXw*{1
z at nu&x^7}18JxLBkL`T1HPN2|jx%;F<ltqs}-)Cj)e9r8VF{!DKFAU2tJry+Kvp&A1
zs_g7+CcPU%1&!URs;a>Te3+IQZQ<0Nuf4psHr{ifp-}P at NNMs=#ee`%?yT%Mu?+7~
zBJ!1YP4)D$V>Qxc{Bj>VnC#}i8aBEl-Np!JaQT at E6>!`z`%?@!idz^6`?J2&f at Xjf
zw<Exj8Y>xeIRMp$hK8`FxL1d1@?$7P$*Wg5ysjdVu<mt07}bJ$Ucsf}^(0fD-byRb
z=kU*pc*CMYbP0PlK#&-UjXbcN0Kl4Q-I7?%d-J=zWn_&-Be}PakAlEs=4)C<$SEjF
zab^|?3C3rgf<vaJrZa<8;viN at M*ixPKiMm|{3i&|1>3_{ynCm}9-qIIA>~C4ZW#9V
zt$4Q1DZD%w4QCnqC;lYoy0q`QYRz(%_xNUixxF1TUTvA_^Be-lC7~Pbcy_^h-O9 at 9
z`quKOv#aYJP#R!yhGSm5IJxlHF-ibxc<Qa~ZJZQlU8`-~W;pbRo?A0G0yt8E3uAM2
zwY0aN-`xo3-K~{NwMTzbo{(k&LOWqZ@|=j680CIy-_r1Vxd<}3y{%`OL3yjCu#vk<
zL4TDt=6cKlp at M^hUA2}*o&`N-?B?R;e#~@##BT`8DdOU2esa%GPEBP2Ua1W9m@$9;
z9d`Jvf#_h at x24yo+Sb<A=)fx=>ev0(i_NAp`@mFNhS?3a8ZQ3+?#P|~4$iit2^b=K
zdl?}fz34vzb^z<}$!8>QL7e3EX(^A$cu{Tb^!zW=<)ts9ijO`Ba`!22QY_HJMCI|O
z7|X$LPtMwp%+G&$-l%)n{PUzZ%Q&|~&~uISo;R)8OCl?wRbESoKJRx|fr2uP=4Avb
zF7%h59qM}H|AHs3dD0b!n&SnGo15;E^xwIj)9VWZjGLRA*fbkV?(@P(4;M?KKFLgf
zV*5tlHj_TAA0?35>b`=(hb;?Zv$ftUgc>Pg#B?}bdQx+~JzGXtm`=ubJ-DVu21xBn
zGj|5ajD$x7Ob*^V>!+;F9Qm12J6PeeAHXKnK|udRf2~)*ZuXhKR$1!Z8xrOGZfIrh
zvpQ&H|22amqo!85b?b%clCu8$UQW-^2W_&yDO;bMJ>9W%`m(UoQ1u5O^EepJ+uQp(
zFvx1 at oCuAOq<cZakLEa%8>1NQtBYT;9gJ)*GLZs at B>@?2l+!Az-b7J$vyFt#e5CI7
zi=T!S&TJ86@$vE1N3 at 7uFZGhXb{ao(R`Y<(<6b?>d++%8zU8L(iluj~%O=l|PLNy1
zaDxu`gx3GlllSgRrkGiSS7(B{#e&WZlA0t7Sa#tk(WT)G4=JELDtdZ{o{PD^94`Ok
zs2*1!)T|}<2QPiDc23><*<1Y6_`+<|=bL~C`;*m4fL%;XSYfRUd%QeX0{GtJYqhgo
zk4;k7rrYXaJ1t6?S`zv9KfUJjQ50}jFoI0n at yGR^ST4X0{B6K at ktanG4GxE8F1Sp6
zO`~A{c!4EAZz{dWbKT*W?f853)_~IPO{V3)jOPFa)2_%AZYIg547R08&~N<g6;F}Q
zKcKd%(9^KRsg`K-D4ng=lA7lb`PSA}{_x{co~5dM23si!iANQVGe2IfE($0A-4Y_r
z^Z5A<M+h3cbzA%)VQ}r*ILL$>gNa33zklD2xMZX2(12AyX;NNy at +ZPBJQg$~q02O_
z at wM(0xoVuX5TG<BY$7^msiym$fMqpQY<HvA47fA)|LoP_3j>S}&Y$Q%_~w+L6V9ns
z4gsEJw=3naM#${oXfp5C7cOquL at 8?9*}(HA`ZkE51TKwb_zX9G;(cX*YSdqD6DBR3
zy!>JPXyv`3o7T3r^e0^#8zSjDUKomeo!+E)m<=>_rni_S at 8}6>@8IQS=kEgpm;rH8
zC0brZMJ2Q}Gt*M1vCG0DFr3YH=2`O3(&FA?v--xyMjK{B*YnApWgwS<N2 at GF^Tg4e
zRM9iq`l&;3URhbs`~A9HysF-f at Nn?fI}unpw^0t<OM&Ac{15Z_=YGOAfD7GdeZtwv
zqeas(&|FveRW3&-5&%5$`O=`*@OfX~cVF8IW&=~i9ORV^DxA%emrDG%U1^c46>Pk<
zObDl80lZCG_{sxxn3erUl~<ey6|Y>c$Nejlx=LY`9W!N8p7WRgCbsRFNG|+WWvMn8
z-%}glw;@wIorX1cIGE|V*qd8 at gaEEqH*c2NU10od&tWsbdMZFe|NUu`^l2cI0lzmV
zcl>~uUJ|7}Yuk~l+1(o}=95SPd~|evJ{n@)s~aadIf=n_tG^%6OmTXl)@NvF)EM;a
zfXS+|ZN|+`eX3Tu{%U=J?^Rxcz}5|queW~rqFip*<NcoJ%{ibvYy5#0&#QU<Xz8Xz
zE^Vk;WA6qCHaAKWcEQL&E{HmMXE|&eXrG{<q3PWU*uQ^&i?LbRjhH4;-j?tuqB8 at D
z?UQ}mNA))r1|Ey&0nssWWYl^K#DJBZh=@qf`;LmONkKtDzV<VfBT|hXuD6A`)_%=)
zh3nVmE2jJ^D$dZG%20n8`A#-Z%vw`Hfk4u0 at ryxnUexeUBU86l@L7=k5rb7;Zv3IY
ztZv*G10I4C8?jska34D^Si9Vz)m;(HJu2-j6p{a<*$LDj7+A(mgGJU`?h7H|;aEHL
z$Wioihlu{f{=+$qMBqZibfe3o^}czTFVhpaweCcK<Y`M;QI>xj_-D0Ne^yL at e*EFP
ziE{J$u-h{W91HO$FZ^3G<T6fj at 7_I8*)p*Wf0vj?jUrEm at -qJf8(T!LG;nLB%d^Mh
zl*<==?Idq3L~|`K23NZL{_G!TYtl6JB^g^)AU4~JEe_RCuy1mcKOf2w+o%#4j+qrc
z%%vUkOU`>ac;$8FM<&a!y!V2XuDk%bZ3mqCQks_9U(K#z{j0Gym&Rv0o?%D3nwlD4
z5pt6jdNFLM314eio*4<^;TiD%Xss~*KSLVAj$P}=o?PaC^=oBC+sx9gh*EoULDppV
zab%@n;`b9R&CLqHO(5heD05Cx&E0w9tG7 at J?>u|Ua(i_$|LV&peG>{A1`lXFXW9us
ze&}V8n;81y8K3j~cVVFHfS$}fMIdc{Et+^F5509V+mPX!N_^xM*Sz~FT;d$aHyq*h
zc|p+p9vH=ujrD6LTkBm~3d$%tgJ0!YF%EkxO^R!4YqPRPpO7(1ixnP!&}s9oO<6kk
zr3%-y;2N8mO$&F1zmCO9?Y=5(Jq%&+Y+-9>V4fat{KJRYu?hDc6BPd5nF#ItyYAp7
zQj-|00bH0g>w$fQdNTOX2Ve%LPMtDVBdGHaD$ca8{XX;m0Mtt5qE71)wt4}Iwnx(w
z7JKt+>ZrQXrAy at 1)zzQ3Zmmc#HWs`XYR at yZv<M`nx1GsKrm-}$Y>LM81MW95=aa?#
z_I7<}>7L&Wc=vRU_V;XF#`5Z<8X!3kNAtgXe`bDEmhE^wlDIclTU;!tnR-t9#vNZE
zVK_>u(l{0%qS|{!QKnA!-CduV_DoiGKC at cmfR~`5ZN9xY0y5JmYb9$gTVSv<r*nLI
zIvi(T$;q)`u at tZj1!c^e34R;3KY|QA`{PiJAKT*=?aP;g-~ZiXb{Kd=OlLmVU0@`4
z^Zc*zu`zk&<>h6bQYM`6(eqrNl+L^@U1qY@(sJ#8CDK6KPBXa$mr38cbxU6PpVro%
z*(E*MtIx0UW3G>nkB|4t7%@)ZpryNW=T7(g$+95*L+(a~hL6bjw$i<dj4KJ*EjEaH
z0{`sT{@6}d`}~~owZr{~wO5meuUt)kE-CR`uiy?m7WR-zPEJOJZftBUujg8#nZV#M
zryQ_glW?;#>t<UzlhE~j_Oll*T&N!!O5bRv?|<4V{_X8NA5FlTvbC(4VgI|W!p!Xp
z1KyqAC#L`HdMhj}{Gt!<ps`94Xz90GLNTWv5Bo=DBnb$*+lnMB{?Va>R{)-?=<Y7s
zt3x@}6vNEQ9+OAN+&OgB#J=wpCpJV&O+ALk->mHCBqi;t_lEqIY}J)ABpWaP73-<E
zRG{~<r{|GAcTI|GrHQl&`<?Rgv+2 at 4SzT+I$!Bt+i?+7B%Zu)Pz9ME=c7qCtxZ!}t
zRe{I#pFbC>e-dEH&zC7yG?Od%BA3iVU*2K1`>X$<L?OKnj))KZ;|ubX9X1!|7{bPQ
z1aiOIswtX0%ffWd`RPs#*+Wh#DQ3JN>nXf~K>)sD#V%8ZkxcmTUyDK#s-wXXyg$5&
zy{r1_Kpx|<t}4gDatByM$|GlpAa;E_*P)wormA|is`@o0>K9ga5FSTFC9YhP(Mp%r
z{+lQ`9vSTTXQ%&-OEOblH)k=>*L8rL=s<36Ze<mfM2QayDZaH8wV(e#vfcu!s;%oA
z$4)>%1wo`+Is_>tBt$||Qd(N_AR#IuAR-_lB_Lf&NVjxLhjcd_y21ZkKF|Ap-#7ku
z+%fLG-gD00d#yQtvG!VXPTY9ZcX|2T6ls1vD(rDdUA>Pm+4lJfEl$#|Y}#K-h3=B&
z^|@?@@rsO!iL%$eYZZLtB}1<_eSEf#4RsbUI^|?!hRkxP-7xipamf?C{mpq#h+nc=
zwYAbR<L2vE2*kw1x-U_xPBKr2oY?<DgT9Rl1WiwSy?uLn%f^9Kp?@qsKxAU-Z3J5O
zu#5WwdMqR{QB3s~sw{3P6;mij{MXoE#g8tpUUo1|r-Tn}fAQ=}zpU!|9iZ;+Ad0~6
z- at mue<C|AKc3TCFdx~IubG}P6+0+Lcb_$h68ZLJ9w=SUm>-_g<XAN?D%O*TbORYmV
zX`O;%W-LWM3AEB(3qGU^gQRQrk}RjQ9T(NE5d5nZ>~Xxu;FIz;24oC{#9KdN_LHw%
zi1J50{^}f&8P?Dhe-4qWF&r>AvODGiMV60B{89&<M=x+S-%FwwO>SE~>JRds=<4R=
z<e(NiO_s9p7P?Q1TU$3ZPv);}Z`V!K`uUp&v!2Dj^#K%BhVkcban6!`8jbC|dX}36
zDNkMezvk~W9hAi`8}iNf{vdmYN8Xfo`So8za5v8#9E{v^65gY>3qI$MkB=8Lu<O&!
zz$xkN*zlXd|Et6R_0U4AG_8p{jzst;l{`agJw at L4V&dYc;4bQ<K}t+&u62U^u7mXC
z(PVu9s*n;L9Sy8U_vSGfStMb_aF+flA8B_ at MJFddm_|;=1r>A_laG&Y0ViJ}Muaks
zE15X0lg<5WX<Bs&APCL~p7Rv~Ho(IN8z8rvdD2*1e>bfZJnJ|q at irxeaVy$mocn+F
z0`RJ~D^<L7 at 1@^asGg1Su+A at GM9yI2Oy)Z_uDh<=mcEP39*zzUDtePS9--D>WoZeW
zU0mWokfKc&mt_YLjClGCdrhE?hQQZP9_mqSnn|l%4O5-LgoMoclidLu18Ty!)5q8M
z1VKnYr^5e*SraALazn>Ju%I$uqn-3_T9s#)8=JnEi)3Kl*lX(St^Fs*>Vmb=3sAXe
zq42y-sgF0!>`G!H?Dkp<c75m9#X|XlnY!6SDra at F9l?=!N^h9Cyvr?6?E_qmE8kw|
zNxzsW$j|@m|G?1#P(G>Eb6=*=!P+j>DE=L={c8LIWx3nnsxLaKs$~3ZobO7$FM)8I
zWj;p0{$0-|ZmffA`L5`-{e2nZ<3@?_6#wIiioT>4GqBG-=e6=JnRd$N*X147Ty>f)
zM6!$;2>~Q*Sv at gR)6 at EYej2ZCZoc*OJeBk)Qf)%w^1}x{sk_9=0{r~1d>jVuQ3?p$
z0FF|xAnFmt$H)KC)qXB%!V47OYag^_D}GLim50Dr;_M&8h)zdS$^V6#WD&HP32g}M
zkfI0)QyH6>K=0F2Jdz&?>uKv5uYE|ec(;AZ4tH?U7s89(wv$pl>smY~`NNYf9|zT^
zT;&!g-sD>iPWZ8xKV>~aJGf*+n7=1xiiPg$gktOJ!Y&3E78Yg|H;nZEcVXf|wyKbg
zsgcpIy1KJq_uy|3(+e`%2yH&oz&N~CVU~=4;NN69k!<mz>MUGVs>EqoBTp#;yrWO8
z0a~+M%Wb8z?xH{2^!xW&rAEnt{ecb!wyquz+^7QsndR)xmmt0UISuRP%JzISmO0()
zZhpKe at c2Oi(=2GKH~2}3iIu~{<fqQzbu$TzyVc)B0!biTFqWw;bzo<<Erxwf3pACg
zLqO_3O?BqX8A=0;NJ2JWUS}xqM9PK(dPR3z1BbdQIp$^BaVwzp{Y|G68lb0_mY2bi
z%YJ3F`B0ef)1y|O`c{`3P4GtOULboe9cTcFnY>hzn6=f_Uriw_{^mJxhiW+{tz6zG
z=OT#1tpCX=tARWVZ`;W_TZ2e5IkOefZ?4~s<|5en_Rp(uls%9ibe%pLY2W6IL at PQu
z{r*Y24O5N5Xh)qqFWQSuNqtk;CAVg4!u{6&5mNsPDfqX9a0tw_15DXIGs$@D&oh9J
zwU3pgNtTa#Q&4~jr{v_}$(0aS8obCTC|J@`wmeuMPD^MueBNz$mC8rDEbOQ~R-BR=
zm58bjpg7m at HwH)VhbHe80zx7p-UX&IvY<43-4M>DrKNkD`P?ehK`G~=-!JC<{Mor6
zkXKZJ!RlTi05=gB9_~9dq`hUG0>+Pd=g6+IJ5?E()(qn;F=uDz at Q*5n$~h*Ft`Jy{
zmjCD)O?CU8m6bK*%DlTYT6{Lrtxb?C{6kN$rp15O`?J54i_CQerZ94(w_urB+Eg#?
zyvOqYpa#t&L3hsN|8jFIK-1oudFI()j>9+&opa{B`9?KHf942nwi1Qxxfz=+PNct^
zp|F~iVMB)&`m+t6Jgfky+})iFc=9j?SFY%AM(eP~=tZnWWaHI^&?fXh9<lxXj7}S%
z^V~DVRLfR9d7UNHYEC*`yzUEE#g7J6CYe<|B-VcOXK+s2l{gopwl2xZ4F9%-F)}i?
z%a-KjIeHZrGGhq|3GYduig+yZ6hh8fLpv?J1Em=EqyMfXUDs*4*~8XyZCN6%S+JuP
z>&Tk9_Y18kgP+Q~Jc3RY#CROv5D^i{%FC-&*K_v*3vSEmr~Xf{FlTFArtX|q8H8j)
z?TNwCvaQ{Aiwe~F3hHg^!iid9?9WuGSjE+!R>?o;bp{IMJ}7f3QHZ$hUeVXrHy>oq
zO`Y{yOurr7;n3Rt@?M6%tB~#P>iFD19*f1i#ijp8i8u_ko%!$Dw8cETMivg-)c*?(
zXdn8e>$jiMS+SvYMAW-5zMpxS7 at a)a*Ier=<^3l at 0425-`}zt~6#CZR;-XLSn03BY
zB})X!s0T_X#O|Q+a;5eJpH=i^?+-0X7gtUHSf@?jW``|3@!4_>P1noL2e%Z^$>(2x
z_hw^q(xiHB5&6~L9Yj3T#EQ>bxr3I5Rx1f;9-udSZ84&K|AB at v^A(AC?E2xJ{g&A?
z)ijk#AjO`Yu>TPwdNX5yNoup=52U0*pN##wqrzaUSL|>B4D>mr#z(yXj!DOS04e7P
zLY_<+wG<}sM67TV1em1<HSRULH4+aP7#e!c%{|L7);Rt9_iy9tk~#iKZm7iP!%#L<
zSb&lm=1I(5A++?@gUHDFywjFMq@?c^ME2ei4a%Ve8QIcNRl&@WvfL=1X>PykyS%Mz
zyyNJ5Zp$i?#PiIh>siI`?%lgn_mSN1+c!#i-nqN#nQ<=Sqb~lwd*u5`v6&_!AKd8}
z7(9JYG5Czctao9usG4+RV`G8}^TW}Jw}F9ITn=|RqeR>zX4+yDz~IrDnwp~XqW+}3
zva_3z=}p>^!mfNLzFi=qZ-+-`x^pKKNAdps*L6H#gv4RmKY#ryht7`DNPP?}D-#Ap
z1;2FV&!0aXVD#sXj}FiW%FD}DN}Q~s@(2k|P*PKSiQT{G_2#oI9aD(xmLqjsRh8I}
zXyr)slDWIk)=>80L$vR^qu6nA4j6fAaq%HS^u-(f=p>a&nTsRZHBeF_n3$Mon>K#5
zGs-|6SLV92u(k6zHMes#EZAd10NXHa$IQq`39Q*`38;e(DqGBxc>*V1#H!C=>R!#?
z&E_ANdXnM1w@!yn_vUS;tENB5)Gae$b=PU1AzRdPT97q<QT6#LH<sk$;IR%&$v(E?
zug1d|{MFR|0~Io%rV}gyw}j&WdVtpn+}z#g8-p2)Gq{yRKo>=2aC^lbTcz2Cq^GkG
z1j*EoxG@*_DQD^m-+g(=j}E88AepU>RVFAh@^@67W~NBh%wuMmPv18L5nLy`@0Q5*
zcPdV9%2mhW`}U<^S~|L*f`U7JT>Q!3zP*B^klI at 4iT5R9cntS2SLt~!_gM$^zk__Y
zGOrjgAJ?j+tr!2?mQ_=`ZuRG<SFzKw?8wLn&&xySVuyJ`5E|`cm7Yv&Y#*aVJ#=TV
z64XKRfBQ{F%iU>1n^VoYk|2|5dY1_Zre|k~ItDS7p=%O4?TP-)Z)3QXD0)+sz0o;I
z35nU?QC-_g&A+@~S{qhxJI8r at pR?Hy&GTaCBqf^MuIsEt&6TI-=NmQmVX%44DNc1Q
z6N~t#KBMO44e42XVmP_7BG0Z at K2_sX@$!i0xPX3>Rf?AI!f=tjcU*k at jgA%`cHL5r
zV$mpm+c$4KJsX>vno^wn4By*onW-2)VAaYu>cLYcxw})_IbbLufrp2O4#*;tL-4O=
zgsN-Wc34?i*>8{Z#hW<FEuLS-Jkhz!mKM1*-Lm2bTGM~{YA`LR?2qmTGmvN<-~>{o
z0&Z>2w30<|8lIix+YHCXKRlkSzWEit^q4thYG$S`0+*egy{Wl*((RUwq{CeMYagGF
z6rvd`>=!RwP#RS~+VQ$XJO at lecl_YY=meqRg8qNK3EKU!zP|o8n}cS2VnWQW=&QvI
zBU4kUADVfO>(e%7+tf$9=-9J5W=t$tPiQ41CI<QWU1G=&Pl71>y0t}eO1#f4`w2s~
zj!jtCAr>rt_J9=IH?mu|Zk2uvV0qKX$wWVcnTvc^9<j%<ok*&{yC@))sBV?+ba{MZ
zZQR$_my){w&!5!HRHu`SHpCKXPcir`EG$<D?Ck9yYKyvTJ}dIL<GlLx3IW<S-wl2X
zNMgZNr^Kliq`4n18Q&`(4x@%ZyRDl7b_q3Gg at ycA2p0M>@o}DY#GUau|Mzf1LBK}*
zV77Zk1ZSx~`;>$YW?^G}{Um|@-ukrWpF7uO6A5`OdQo%AqAyh!(LuGy$VmAT7cT!M
zVjXSm=Q}GpKmKM=Q?R%Pv>pF=q`9ADWr;^2f_kXs=JNDfexMM&wLFk#$m;ISh$Xv=
z1hedt#6=v_-(Ne-`cgV~SXkO%f-h57eIyrbN{AiE`w7vkIjE{sEN|5LQDpBGE_yG^
zUcUP#Z!>^g at GSiB=g(JrElI at YZN6LldGh2#f(?3;#gg;GFY{-R2TyHq-Aqrq`W3%g
ze%p!D7JTwich6*NXqymQi5NCb-x-}?pl`mUGbop&SfNq_Y(jP24>d*4NaPShOH?;*
zr1seGh6EJ58R}##K2FXW{(>XNkdcNCnf;Dxme!Pe+BBU{UvlBXg at 3OT_2=4S6`4b>
zEnGf-K1GpOeR}w-fW2aYwN7uRI^$Jz>IH0iZ3M1+$4nVFxgOUOdUHQEWBIP6NB1SJ
zMKi|!uZXz!{5c0YQj~Z3H!9k`zqP28uEyG6#Pxw(@VbnQjCM1Zz}v9*gJEg+J<n27
z3%yvEm^|w5*8s<-S9VIIV=!u^1cUvE3X~N4<u2Id9wk50eE03H&Wp8)+OvuAe$juA
z4-}5HMEn`d)AeF(D<|ge$~UUGr!;2pZ3%tT6N-Ph_xK3~tvZRJkS%tkEU?jtQ~}FD
zZ&~|HuKp8F&g8b-)_wIPKeQT86(XB$Gb+++nKF{D&1Twq_M6sG#v+ErKh}_04xN{|
zLIBfgqEcc>c;kkz^>FcxS~atZhTp&60r`MJe^_H4;r4K}d<n=(oH^u;4;ZdbgcY3K
z?BTEcf{RPb%5L~_^6InMY8jJe=#-pgU|_gH;IuqooJH<F;J&|`NoB8h`-iaFO;y3h
zkrCa at ioQHywS8I@D}Kz%*Iw+oF^^l_Z|QcwmFMHGk=s8-{%w7ot9W?5ydD^cGDMat
zNV?npk^^gwBe|a;_um_?c&JPGeTnbeI<r$#sH at 0efsJYFyI6FZQ6fRaw`+8?qQwT>
z=sD^)0GdAM0kfoh59&TVY<lQ-L3sa}QM^Yfzgt*jk}rVVuaS}Rmige5OW&H^$dghU
z)b+~UE}&(zy!{Ba7q9*K?hL*NlDT>j9ehH2aOIb{TmNezGK at dJYCP8cmBp)9xBbYy
zw}F%Qj$gBz>J7AIHP@`G at C((d2Wq}Y=zDs8e*UYO^K2C6o=QGeC#I-HuGO!1x;WO1
zLm+gpAN3DLl1Bd09R{BAd5{qwZVGhvYw~nv=m?_nd`wIiBqb%$-oh`ppPZzA^W(>l
zeN5Wfsg>apPzQ?iQSWc2yKF5`QoHR8V^CWKd3ijLOFiVR5yvZ`dM{59#EN_4QBvR5
zDOPb;NW at m-i=HOXuYQL+y|A{kQx7TtEy^Mt_3!lvTKK?O1;O{mXKZoNZ}jB>u0piX
zeXw$Xun0)LYZc%DbC;{NrM9|{S>ZgXc<H`nGdw)}qL0MS&kyxV(JghZ%)6?F4x1Tq
zJqedTX-GssaDp105kUPWz40kf!yo|xfkO8KK44vMKC6MxyOHnRzHbVP92AIz5a?~l
z8yOk-J~%^GQ(rHB`}$- at ywvxB4zKlc{lR>z`x0}51?;-zZoVMLZ;HOWnP=Y9$->c*
zFBI_p{Yiq#)>d3}lonX}=g*&6?gvh2vz>&BL0lMp$w0TSTz|**AOFf~Z+t+Nwb>Yt
zpC<8HS*TjE at m-I?dy)Tt#3Ehj8>kDeGpT%YCgn*E{bH~>+=XA8$iW9yGC6~V={qt1
zgI8{udd!AX4MFGe at Mbj68h#+<y%=@Zk=b5na6M=lkHKo;YSJKK(oOY8Fw77urnIc|
z99wo;!Na?|&Rh5Dz#_AB3m7+D^)EK>%V;%w(U)tcjA{d`l5v~<t^vAeF8QczH6j?i
zR}y!y|JipS?qPTih?FY?=xsYp8)|y9x3`DRa<oMYpP1YJ?`lbgjI{JC3F^v!Cprxr
z&|7xf$p`=4`5phjpBclIeoy8<ug%S?YC&|d$bn{{Vq`pz-j+O8R%Uq=ihJ3g7>4~q
zs>;PwmG`0SdZdrq;FI?yRyQ}NTAiuCmA(9r8Cv+%7r|rp2DP?cTU$fH3u!cEConNF
z(M=S1*J19PyjG-(ZY0L=+5f^BCj-Mdv^TEMmd1iF_w~Pb>V!|7hN4qN=9(OAE8pRg
zDdw50PEAi|n at ob!36!Sw@$)nM53Ds~3=QSHm{fv$`gL(^<j@|;!oot5#IJpPqUQTi
zRPKKSF)Jzxq<UFTU+>>3{=y)qtf+_s3O!!_q{r&m2_JB=7rrOh;Mqnd%_{y8_Ba+o
zdp60)$Q%#V+_<IPm*=`xy6Qb}!w(EDENB0Y5;O$#m6n%(prk|q4~EK4|N3?B3IPbg
zs<d0Gu8`OIcL({OkG&fBC@}SJMP(uK- at 9~l0EjxsC$IR25c5784E;%hSa8lb4($JM
zj`{DJKHl%oOv+bCNE(Wx)+X!Yfh~;r9A_>uF)=OFYDpwy$E<WIh^mef`_;nGI{pJV
z+4y-8BK!AwJq=Tb#v)Uftbp#G4Hxs$%YSXiyXvPuT<rK9gSlfKlS;4Gy63Sp{6hP4
z&+)m{w#(I7sjuQ*QUCO9p#RDA^E~r=vS*qZsm3pIEPK=*W~$v`ZSE{~9K1}|!9}DZ
zA*psE#GT+i!4nCkEBBqk!m-!@*B37}*<Q5>aaX^n(B1f{&fT)d_4P$`j(eM#brM8v
zdxA{xd7Kw6E_Q8>vn7-PA{d)l!Uj}g%FNun6O!NsQ$@kGYuADl8C9GQ=JZ|cOGD?@
z?8>K}+;L@*$|&eHS=%df3H$KjGT}uphxtyra&i6R<h;C)9f9mTi#|deQUN<!urErk
zuE at o7Gz8I<sA7FxY;+tj$A2fwrED^$)c})|nA~U{OO{|6q9P(>4YD;dE-2|Z?))0l
zqexf#F`-*vQ2YL%d?piXFG|K|S%__8(=J?nEsB}%Omx-NQdImUj5$8sbB$(UV|zJ<
z<#pX&stB<#F>z}1Enu%WTo)Jh&Nz<Nc~RblDjf8^-0YAFa4Q2r&Km?uc>wFP-uaL{
zC?<v?^OO)~WhnOBw{L8HqH!ziamu~>!(EX<Kix5?43(y)ro;ZGSyAXU5)x at ZnD0Z)
z^J(Xfa0M~$$BU>Gg|K;;!<Ng)z<?|O7XJ6y<KyG*-rntYyAnHvC4al({Sua{_336N
zMn>tC>BH4(d?QpDf at roRE`2$AMO%=f=fT1}CXU4_%sXT_s(8O}t2jL-ow}#HI}ntt
z*!Av&<m6zD+o)h$bJ_QiZwq}k#ulPmy<2f5?)1hL6+$h*_0qSmD+9ca#%@NOi%J>n
zSv15P*A4Ie9WF{1TEWCJ?3^{WQV$6U3HtKodTnj(R?~qnn`(u%Xvz at aTxUlI70%}V
zVy5Wo7L1RPiOF_>-Q7FOW0NkJg>I(K-rg?<jRjj<TRaYPiUWnV@!HMv85wRXWqU_2
z`!|G(=$)n`wo}56_qAchA4Lo8#bO7e+7e2($_4u at M9_$Xyti>bUGCuEp!n<=<DdTi
zz@#L4;hiCS*}Mxe%yY6O6DxhgSW}A4Xq~;kI%QHKbF)3!kFJrD%Hp7US3B+_9UVa;
zds9TrEG({5u8rcDOkeziGQnVmlAkiO at go(2eRGhM1ULhwFLF9t7ziE9J~|`^4EA7m
z<LkzX_jO#mEiPZaj7l51j%EzamaP8~TG^Ubst~2PDiDLJ7Y9d0MQI0{QAp}q<&IcC
zB-~mlb+_ at qFUrg&X<=nWr=_JeQ0$mK)EvQ~|7M8TRCFolXwiOnuf%bSv{fo*DnwnE
zih?3W*F#TaxX{+~tr!)sjO>cB;}+JgqEC25V;hXHjJUWbPCG-4o0!CKMf7Gq5rcf-
zyLV?*)zzDSe5e2hpdlwGUznQ<Y(Btj*?ezV&X)@Lvv>aD#W>XZav8HIBOpA#vz2?i
z?Xdh%Z!SQ1Y4#j0sl;qs47T%NllE5JUu?kga7n>ng@=TMM8S$j(=mBzOu6DYimeg+
zb$&-F9cNe9u4eX%@9kU^+nChv-)Z53gy&Cx{_>@F*}@4kJ8_lMDb{Dr(N{tYG)f}5
zpi at KH_r<=<;Mmx4;R`xVb&ZX_2Zz;FRma<I8X6=!hn{>V)m2KSGH-1jPG&Hz$bTry
zdkGjMsqVg3OU(0p>|UM>hw2$eN5_Qtc=$!~#~R at QX<c{k-u?XfvqM{(!`46i!8BEj
zx-xfyDP^LieH;37+PL=pwW;OhA3Tv8J2TGg9h0+ejp^1-Tjf-lTch~zrNSYS0z*SX
zA6s2-J?u+*6xmQ&c|T(^7T<od4ws9Qb5E-*hS%cy3CbX8hqgruwF*~@{~LS~z+`oH
zcb8%d!|%QrV<<CeB?WmN-_#^4S+{Wql_;~Dmi?KVOB*CD;o{<gia_asRTTDrU0i%u
zv&@AT)vppPFH6583?udTa8|UjrR4(;FE1GP_k5Q9_pGd}@Q;dq`>u>vdyB3LaC7Gz
z99zyF^!D|2<+O=BeE04hB()S36_v{4$Lj at s^))r^;1$h)q7qV5Wng$vr72YLI!f3z
z|6zrxdNkHyrsc-Tdq)R56<DJadI^u2PzlNJKYqa7`eL6?Euc!5&z?!6j-#2j;D7dy
zc1F#%m;M0SRJo(}S7uuaJ)a{YT+n#b-`f1j)r*ZD`S5wZq^rAI0u=9Nx3U6LLMbRX
zLcdyvv0o>=KJPH%ACMg%ZnPou9uX12<F@;JAWQ$0ql3d{ihgI*;tLB47PRmJT1Q4p
zD-vd<;m at D&eCuHs1 at XM1qVQK0*tq^4R|ECt7v0!;O#D*8=u4Vx4LJ}*cI*$CLp~)Z
zGqi?rdiGVm%Q*J;Eq=M|rb6bMVmj)Z(Y+`24-WaaiUk1<rD+K-Uc5-n&d%;-RkuDJ
zs`m*7QC6-nZGEdGm at 6-)cw at lZ-Lz!T^zqC=bM?}n9B&>To at cYaqujhZ)5XNa;r0$}
z{#K;-Nxi`5cbs<ZA9TN<YNP%`tKj2pR at N6-`zxu(jtBbFw?tol&CSg{kAoT?<GZ=b
z$Xqyc?p$zm^ex<*{DIM~D+0y#vv}-%B?AMRKeMvFIWG3Xpv6lCd{Sgg5O&>WOeoEm
zTbLcu+M3dD+nI{k&5}|#EFQIAUS8h(V;W;ykqJMwcW|H#aymdHc%}DXC0A#+y`$rP
zWo0FgLCyKRyYnmhw1gLyN6I2Ct*m5d2|FNUTt{uqW;!$?W%mYD0EiRP(;w!Tw7N`L
z?6pTN9t~$!#QY8p*;yX!o3^94bBj9eNsdVnT!@#4hq2X-TB&bQ8na(m#;YLHorl$Q
za^m};R}l?<(QNG3>x9_Y`zXk#rX(^ej_-zqhW=gNnQwA0XIKtj(bg+oKVbP3H%p|Y
zso4P0jq0Wsg>G at Io5ycCX|3yT#g`Z=mI<6~#gvr|M(|po&j4jk>Gpz^Z5SHT&S1Ff
zxN|(xQxPdNqv9g2r9~DV9-cZ(LPMYg)}8gqPl=UjhphwWioUHnjqP&xgF?y4$U8Rl
zaVu>^n at N&>*E_xn*iUKJ>7l6-5)zu)K3-_IE;-nlb35-P#>K%wNlZ-aGVVoutEj^S
z(>!lhk;$-CkZhs1FXnMuKp<T0yT*N*2Oq{3B7%d7&~E at +0y)Dd#kn3B4sQV*K4&OR
zug0vt+!rHv>vk1wnCvEZ1WNE|e|6&29D0??vZ3S=3ZWZ`(+)b&C*`1TX{Duht9K^=
z%gY~?+{tWyTlDtr4-j#T)YM{B(~f;CnWa at ntXTB|7G~zc*}e>&gI_yJDkTZ&ZKA&;
zCRS>83ymEo4qQ8~AxsR<GhCRdCDx-=P*Cvd at jA%k%+1SdtgCw$?Q#5q`S$HnY at 6eB
z31y}TM!ngi-SuDpY%>cBWjD9ds^%}r3k&9?Tt?^1Haj1>d5cMu362cT+uD`;Oex=S
z&H39+W|8yR%(Ao#9;C6qKg`_Rd>7Aa&ku+PFmKXPehFX~)pYCZ=}A0s>TElAn~*p-
z33?nJ=hZP+-N8uv$HPZ>jg!JI_N)bQ-90@>(9Q47#?12KNji@)4EOdvn5^@USBihY
z<e>OQ36DiR!_6pJy90t)c%y~8yTZexskL=l at TS+?k`Ln|PC(ix(<8Mn*pG7b at n<ew
z5A^r<7bCiZ%B&U_7kfMuF)U`Fq4C5C%**2)ja{YZ<qdIgbi7uqR14NvMo%wBEkips
ztLBMn!exI)Rr(NFGaw6qzLeF|K93^#g6>g%^C!FWNrhSU*`LetOMsMRe~%Anubf9^
z6lVm#>b{<0UHG590K{Zu4{_$|{K?$B?eHasl}X9SiewlZJBEhj4C_9i3M|=9m#Tvr
zjd)Cd-y{`u`Yv{T9fRt*nmv2gju~}N&f>a}bcdIZ57pdk`1R|X!UqA=9aBN{sD{oo
zQzQEgn_hVfMx({>sa4v67Es6h!op)iLz>apl&HG82Oz!1uP<foNo*yiO}=t#WuRwZ
z2uAJ0H(KvpWC>PM0N+fRt?@mpC?7t}#l;oZ-md8NxsUh|p|X*YL~Cm+AlxGOYH4wC
z at nwY)Mfw*nUeFTyGBPko#ZIH0Mhdqhl&`tvHs3N`zO)$7*ro?zp9fy^XF&mWI^{Ed
zx~IoZp`%n3YMq1@>o6p8_^PU^L9wy1RR`>752|mi!9ayag;=w(%LCQ=+$mIKkPcuR
zTYkU>l2fL at W|Ur8S$QDa at Z4xDhcIn$V?$qG=oeP4It-&70R at GEL3Y^4i&=6o3M8Kz
zjdHDr?*hkw1gOFAx12(Y{z*wDp?rkupHbl?oWP0-QTQDU8g|Tzw%zmRiU*!9zPCOl
zJd$u&>}9ocbWDsD_kL_(5Eo});H;*pDVvm(bo!ndeV_G6sf==xT<puE{jH+n-QR*(
z)HuP*+k1ojy=LIQu#*P}LM4SaH#e`A7i!Q4cYLCInBwvG`0BN5iS0joUPD+y2)9o)
z24BEo)hqvs4Rp5MUDaRc%cKzgd$@5?;jXNl9IcLyPJ|%FI%Y(8SHJ(BQyW07qO|mR
zG`)6qT<n}^f)WxFC3SUm-KkG_=@(NoFnFU1hyw!yn*chi<R7)K5)%{QG&MAg34V5E
zoQUoO4*^P#0SDNhvhU~BA6Z#}`&)}rV22(@3>Vxu1`^aPvZLF=j>f!y|Ng4(alX}{
zH070lDGMG($$9T`m$HC>fRT-j0zEzb=a7(E48L7Zf4?-y7dl)Dk*_rdoWci7IsaT+
zwSmhC3*}+GCr at senVEI0443TTNiL)Fmq36k0v+M43~3CiH_uJEimZ84yrhg(ltynj
ztC~$zVqlC=t^wy|WpQ|YIqHz22!eclXtaRpof{b&OUAgZUl@&TXnk|$QbSu?fWH*e
zD+y}Q$P}8z4yjemx9&PHT8)<Lj8AjzB+ohhnapt4KjPdg()0K(_~rH!F<`^8n<4A^
zj;q*P1uH4RFe8D{(b1X(ech_c%0!UiAL$jMzpri?afN^A&D2vk`TCWv`<AMg^_Lch
z80L>=jSn9_yo4(Ao<4cPsz2*B_W6D;s3+v4l#}Jx$S-!Cvgpg8qoJWm%gC4!yjmpS
zyh;gy?y}y<kXl|IjeQ;_k_P{bcAmHZ!Fu(^jYbGo8q3l0C>TU`-vLl>==I-^A3s#o
z)ur_H^^KlBy@!J&$Hf5eqKfTpC(Tl4PFxa>SD at _o@T912{kfEynVGz|F&ivR8;2IF
zqS_N??g}R3pt727-r^DxOiWCu4D8SBY^8(RPdhZhjdP$wXkOpD2QFw*@MZ{o+y0N+
zqfz_&E)ha6xNeY=J{A)@$<B%TXP`3qU@(r4upS$Po at vD;w_sMc8c#Px31UDjokjdm
z%DN&gBSQtbib}Y`n646fav7fbBdq-Rv6#HPJTiE9?Po4$W;@U)08 at 0jV#|`hB){O0
zQ|GyJr>g{8{*XGz2Q(rhJ9_{&a4HF0KTbO&Z6k(pN3lkDdSyH0F54$j$Azn?M&@-^
zO$iV_Zr;_a2|s at LU%#z^1Z;9)A^*q(zuPVc?hUTDmX?-tAlFK3D>-zif*Vn(YLqtM
zjZ_wEeNiL0ac_#wu}V)F1%&`z*Ja+cl$0^S+eyUPd_A5>q at lNc<<*LUh=bswf0zH}
z0G!5~Prhp6irMW_5Dh|J6Nu=0Or66MNd6yy$}ok?19=a}rwej(QSsoi{e|?4>{`R6
zhwIHqd&8`{c~jp3;zkvFv(6k}%+jwu%f-i+1a62Wcc`~l4SW--K<xFUStyKp(6}sb
z+iR<SYV!V3b_S{FOQEv0I<hj7pmhVsRowtBO!p)3OYD&}oWcx2(v1B4VJ$5!&g7#b
zB^dVk#l_&99Pa7a*^TSxHlC;PCDE}1h9g0C_Uu_%$0lRFe3vcr&b~fu at ZuZQKVTp`
zO$nEX2nic%YThm_Ep at w)lgEEuUtfPV)ewZ at C!Z?!;**M(!WX>60F%ANn#)`d7goF@
zBgtSqnRu?x90Sjp?QhN_#~>pk<K`XDp|jwHT*HJPyR@=$@Zr%>$c%$=(hjmv5+_ca
zP<isC(c(4$eNvu9-zON+)^Kib>(TO73n90I?LiylZLhOwOM}~oU;8xmq#|UzB>2-O
zvHbjeBWvr7t9|n*;=vdr?6n>)24rGBfrHZH<f&6kbaY<W=U3B`lHOfs*Oe2C4MeA#
z%d2DGzrSc_Zy*2q)Va<8(c^~eAtPS6MwkGAA5>rVEZ$9N)CXs4LA6|*nwmOQMOF~A
zk^m|u(97$L`_WdPWwKsU+cRtHB#^cdEv>CcKa}sykbRO(bn~XefQQ%wpR_?<Cw$db
zf#CdUBTGv;fCbgT+BGm%NL6ghylXi^D<?5A6e!=!&6!8X+ab`H7#V#es0qq;C;U`Y
zRckTQmpAS-Ha7C;SDglLnE<z|VKsOHlvI+so_omt4psqOt at 0d2%GSz?`wa_a&Zf73
zu*9l2&Z*Sx)dVvrd at 3&&@uv`tcE=BsYHDjsww<gCdj0xDZB5PU_4+?KrZO-w at RZJe
z_j|m&yn>`@7l_D0B_$={-wZAtO*0A!eT5q$7k#Ncj{1-j;xzsKt>*lv<%sF7Oe8js
z{lnp9 at Gxj2N_q8DlEn7#=e70pGIBcr_sD*R&}XkTg|L9{egqD5vscT8sh0n=J-h0S
zu1}%BOI?vR<ubrtQ<Ib8iHR_NHHQkb?}&hC+zpoQk)yMd?$(B(kRV+(zC<!oQr;r_
zb2LskS7<PxQgY3@=t1qozIuh at Wuw+0R8B7F6bSqg!lM4z-Mvh#NyufB at yyw?Xb=E~
zf9B^W-5GUL)}Q9a4~>d?1P0Z}+&t~7JHO<D8apj3tN+mOaFN#+2<Y|qe)H8J^Nu#V
z6hN~+MJmar;^+%q{ZX4Q5d$i<3)A}DA$g4R&<sKW1u$Ir+c(N$r)APnD%-fYxMyI!
z(cD)YZIg|TbI3BoSiZY_L(Ig)#9^tQv%Gp7;h#lj*#p2oH+T0xcqwuP)6>&vw}P((
zwacXcozHf^(onN+mw7t0phiSQv=28l|2s<1&AXPvPDxKsK`Y;q%Fxgd4Fj)5?`G<m
z2Q*S26HtRxEC-1|V_O at 8tE(#)4^P6V$KNRIbCNTts=*Bsl0Z*S&&PSH##}f4JtKfL
z5+hbag+ZjGq`P>2`l$Z^Qm7~I9UmWm4hpKoNZSPgIc?0gG2r6jqHX{{wo!R&H{L1n
z^TjkUzWLe4pj-I-wUDr|%)^I%XrdP8=EekRA2j9`6m$XIOM_-Z74wPNwBe7D<OONd
z8P09Gizhk#^{(T4fKX)pv5Hq`Kv>g1dh`gDR7Sd1^mx}30?}rR>F(VK97P=+a`tKN
z%QO_m#>S`bG4UN3kr<;sumGT&D<xQFRgX`59(SyVD8byLxtp3Yg7J26aA*(Jdnxna
z!2_OW?Ubm;MP7dXI{Edkdej8nG{*e_pd*sNdhT1-z$Sy9IgU$B^+g5zA$~2bya_EV
zEaD-)J^|gLMH&F6=j7z%n~e-HiWM)9p4IpeST1=7MV>l`7g$szfEX*^dbnvuNkv0K
z;xr<s+pMfiR8*iHNTPvat;KmnMMareSSAH&y%x9Cl%j}6KzO2ih~%>}NZ#ivi;Iu<
ztg5bV{PQOqMgYl=WiAW_YD&M9(?*B`iji&bbp?KKqY>)I5ajQV3+Ug4+yX{=M-_BL
zoAA!f3#U(_O&P`}ChO#c at wA;?U1kHhX2`)%{N45LT+xLYz7M!Zd);q7F;q7WK(bui
z;IlRexgWY^5ojH%t1B2R!7ko>5bFE<qhMr0{*UW&Uf=%Eit2}bSp|h;2*5x97X-ja
znitzno*mU^8rof(oX_jiofN$M;A2E&Bw93r3qKpL@^ZHeOwVLwVfp9Tz+}M>D1yWq
zr5aYZx92`u)cZ?)>eMMjks!`~lH1(SO-M at WjFa?(k?RQKH2iFCpEW`qBt3^p(63F_
z6XB5FvA%f(++n$fTg_2wu;KfmG^NumPJ%iXQ3YB|>gWifrlHxx^Hn+tG7qq?VPqs4
z+~E91o2Yu4(;#?Sv-WTIKY#jEg<)i-VPf*#+}c9-?B*T!92f#2$wge;X~C;`Yw#5$
z{&30oR<HkSE(&m58q&t`!40hzfg4CED~rM^k~~O#>+Kz!pU>AC#wnhP_dx%+gy0i|
z$Dcoclx%G|N5_ZymIn*wkuHN^LKV>60r at O*#2$<Ttx0MXSkrNHbHk(|HNvudS8k>P
zdC<L?NHmbc#g4Lqqoe9NrOrX1qoje4P~l(<e<=bZ#Eqa=_xATYJL0AO)bX_#LZ~=Q
zH4v<CZ_|SeYz}2NU4G6B%z%6G!UZaj<X{dAJo-fF+1Ng!WY*G(M5n#~@mr0bHG>%j
z)iFl`_$)3dTFl&BZ)R5pWoT1Zw at Cm=n`=KS|KFv~OT=tOzV$k&AC{e+9ZZ!3%)@Q1
zSEKb``EB0}4Gl>%UhM?4q(W`RgDxi|gv>{<f{4tSGiTs(tF)q`F*B`^QlQ1b=_C^h
z(#s at 1-iI7-7#c#>14U<^Ygk2*-SowyNs7PVY~@g|DxlzWZHD;#{4bDyvQAD;r(eO<
zkd#JtuUy<P%(}>N@%r*mQLwhQ_IVt%1F!doR)|waK6MM|kS}aHO&F37CvOVaAw4M#
z)B~2FO`b!I2G9 at nCP0WxMn(p%X+ki8Ej1$}33bN#;O{SU`+65V5|W?*(<wzT?*loe
z_%EEDui&h&u71<=I8vP+m<_O79^f4)eZd_b9v%c at x;65SLaIEGKFK>Oh+8?V()1oG
z0_!s2%YeoNvWSVE{v9 at uBpU7fMiT>nia3_5;~sTTdzIHE8FlqgJPMH(48JTpHT4^a
z_GfDowXz|WnbKr;tZ8AcL5D2o#N4+Dl%_>5XM!Q;z;msUHVa})#SQB}PV=69r|&Wf
zajo$B^=mipN)Fk17!4yUtAD*4R#%^wSI=*N(wm<S)tl)hNfYa|H#Vk2!W3B*6bk_!
zhvLR{w at G?JLPL?wqy|^Z#li6nDDtNf<252uQrZ3eeIA$1XUG+dj&HtzG;~_(zh7Kj
z%;U7AkzaSj20Dqyu<kOTIZ~{y+e`Y9qMXCMBO~&$v9VKgbJ7qECcnQnX1JR(- at g4p
zkm?Ag+th-%X9%1S6DVPv_U=?=CI}rc+C7uB%ktnLa}De95NiRqLY$)w6WL_!8`4EZ
zMgN8(D2M>DQ3S8W-1>C0n|I=4NHNqo1OOYQ09rV6T1X at XqqTW}tq9J{WIyo at bvBTf
z#~amm^aK5fDyn_``W1yV%nOr<NHjK!<czPse<DM)^Bd5q0)m3+Q>j!|R#qrJgF#RN
zn1H3rmZc5nUI%N&2swZpPDDh6k(rqcupSMlku~xG!-q05v8w56K_F$2vxeB3ucz=>
z3TC*%orap3mj69=g?qma{}X)%^?``m-w*{qX5?qHABbWQ=nfITukQs!vddq7RDsxj
z^j`sirpRhAKh;Pp9t;@T#0Dan1_!;T0G*cH)FcZz)7=0MJlCVHM~5T_C{jtk>q<zy
z>TfZJ%RR`>&fc5~RzOl7_m)uD$jbvOW+tX_L7sQ|VcsaF6;iw03^X-KA-4%|*4Nfz
z!Mf#O?@_eF%pUAu8NkY)yvJ?Q@@;d|p}g9K5yUrY at C-8sV!>~~%Bh<Si890skc2LX
zd=8n!&^LvJh1Yj>Ox=TT6}s(l?d|QMrD}5i=dZBOpUrnBUL_=~!jN39-u$T)|DXe;
zNCW&=d at c%Pv`7k>0atx`Azvu4(dg=}Tk>jZYRr)wqwVeYM%^}02h^Z at l$1c!`)wY_
z)p!q{x&X at Ix*seTOwG<BXnGKE>(S`T16E!>zIAfaE^&}?sHY7GJ6*Rm$>J9q^rKW;
zGAOLUS)zgSINpe{Y}q~mgKP%(hhEWCd*dhLi|pQGx)W>k`618(YX4$nV)6ijA=mN`
z7uq!`c>BR`M4^klKTy8a)zu*l>=dBPJwHF6Yt%?c=qrJe=&sWe$tV?*SG|1aN_R)c
z{qgZ})PI8Tq8E=*!xa~BF;A)(?E?T6+KKhtDp?!ya&qEXwel4m9eGD%Q*$9)(4&fP
zZ*QM|MgJ&50-Yp_<WMInupW;4p;ds$3oTcE6|kS%obPh;-qJwXGFlN+T2^)e2enKB
zI;?kUQWA%W+QfSNjVuvC6a=nI<#@llC_Xealo%sASne)Jrh*kWH#eUH&`E~bA;5{?
z&_9V at Re<8!#rx5Rq}fQB>pV!TgyW-wU`0k+8ylN}Vn>#1*RHLR-|DK at 0o94tdUbVm
zX|G;1`}OJsdhw*>?930}-5`&C{9-l=gPLD}BtDZ4KwU86;`cutRKO7g)77K}g5uFE
zuqH?1<?`jrbIS#zYLRQfARN&u$TLs`@Y76`l&@ayj-Ts|HgQ43gcgY)RE9{ZAs4?p
zC7(e{*<j5i<D~-Vai*rHB~fn at F!qo;B=6?0gG_etDsE$-r=#m2QV_ZiQX6?|L;^Y$
zFQc$oVY*I7`;MpyS$ISQ(wHwcW>SX^N4A#-k)}?Pk7`<&)5*}QkOyx^&&taBZ+&!}
z(SK>l($2vl;pNeO2Ru;MEKy&H7=3UDI0dxE at ii(c#KdzO^?>1S6RQ4~32~6pC-41p
z>Qx-F>;P7EAIZ^DjM@*)M5oD?CT8?Upw_`Te__>1T8}9Kh?}er=mKr|=xBe7266x)
z4(k0v*Tps53)zDDN1=6ph;W#yuAj!ujKEo2CEah~yJ#U+?$IOE0Bo~EiXz*>>y37?
z!#{BiDRg*rggT93169P#%$WYgzP0slE7_kGeQaM&{dx{#^{g*LXDEq!W at ZL?Ud0zL
z1V&>yFRJP2JOYErKtn^*+1r~0AeoJutrDifoQUxF at bK_FPEm33V-1b4rr9%F^d|@i
zT%G`DRaI53HnUfJ%gVX}QGOl|uMUD4RTPg^>>rfmlKoy<dNU{}$j8U$o|~HxXgSmb
z7M&D<D0A~pR`WT}$;sJYWY2)M1p%8FfS%R(X><Y9#4cXI{5wb?2`Q<`K0S}xAR4J}
ziHTld8zw+oze!A_8y(+DgDFFm%g^KEdjtE=w1lHG3aHF_dw2Jgpit5|)PNUs6WZT_
z8g1e{1v1yjF~Nn|eDN`3XL at GlU&D3{4H8W)Ep(u<B}SB at va*tcgJTT7Pdr at iP6kha
z>Xs7_5uF0oL5=2tlHbI~(~OQYNw~TSPE1Uo&Xf%e(kg0dFeE3z?43V<{#Sxb at KY0$
zI!xFHQWkc0Dby7lK1oLAx4v!{79Os at z4Qkaq`G$x2Q=8IK64`K%m73oB`JAkG?qgM
zH7_YYSP?`ANK-O0>OS1yUTxuSLp at rq_)55W*AFGe$DdMCQbN`Hz$|kBER}8#SfFxq
zkeR<Q&Se=nSkzn|ZY?)2ukt_bJy8|n;9#OrDlXZSyu53^zP_kKsEW!}P+?Ea%zS~5
z^gH4t&3pJl8;yMY{cC|f-gutF7ZVph<wwk32jc#Z21PZ8`Ri5&E{tD5fRwE4MR>CP
z7B>U9b9BdOxKq;8t3g1ZN=T^dI0XA7nh*#<G+6Ej+v7F9L>xRkHTG at 7-aYB+=}U!X
zE-rV`#zz=~NszCVIc*|?rkQPA;g>}ZS5Kib=_EI9Pzed$%*e<9aiKCWIEV}@>JkIC
zYZqv942x~UXvl(WxMfqeg|;=Wg3O(O*;9|mUIuxN#^B$Y?N!idjp#rVn7pe1g6`WQ
zzd44ff*Z3;+is!Fro6nosN?I1^CWJ6fB!0ZgeDcWgrwv+K&+Iu_RW*0PN74>AIJs2
zVB1|(!Eu2jLfeSZF?)E()t#MsKy^-#U>Z-JtdUd4zkToLhZ?ScL`02aCEniPHhBsI
zjf$~>%-qEb3i6zpF+p{V(aA;B7};f>u5@?8kAsVALXam9by@*=G5+iIDRdeuIp*bi
zkP_?UUz+w7we#;$q80>iJUuyGw=FM&biiJ at aV8DIX1v-PHRA!WJBfo<%~AdZ(?2qB
z?-0o-D!Ku1(+>;+laFddCO!-`sC;v#y{~UpP-un#ZGVLEc<bdg4%3eA6fO0hK7AU-
zqciB?r|n at tM{mFgB1jTOM#fjEsZ1tq(f&_zOe#l5DL|Tpnyy{O2VO(nX0BcHf=C29
zu4-sFpJUQ0_WU`=n>TMX&MBRcm6b*9QOaC*;!;x11DGHfK#eT4w6w~r4OhgegD9Dp
zE*xwxR|EYi>gfeE1~Z_}^Qg%y0Lw0(QU)k7f3y!5J|u2nKz$h(m(ylMNNB5Dd39?`
zHA2R0<`^OdReeUEI)``jo{kPVd`d}6>k<P41M0GI;=~E`*}<A=&|f!2+(^);AT{6t
zV8T)9`O|04RMynI#a;=&m6DcLg_@|3{W?)oQ*)1&5EW2%UT<U|xOsCZd5}lqZA*(>
zO<f%|8yg;Q#4}6FxBmXPfYIYvEWUz*0)*8)$}6knZ1Hb=&ezt~qOMVdgoIa!iBXM+
zc#prwa4%jJ6{1OU;U|fSiPh?hDS3FVvX`!&)YR16#q%0RTlrA!N_2V~GSv%YY4po|
zv)h`v=MC#Vj0={Eure^5t2o|gf#g+ER`z64P8utkjTV=dmX4Xf1k64;)fmhVLI(kG
zx_V|+OUq at Lm?}uz at yW?*`n7#VYB7C%Ds)>Ax?y1?xw*N&T3d0^E?Y?J at tK(<wdL2S
zqy^j=7z;*rc5hH6YTq>~=~GDFDOp&GWMhj)-KcRkv7;1_x2T7=o0}VlNefBdsCL?q
zAGi=XfZtLu%AoEce%C;EppqB2^|cy73r+w`^cUDrpcWadUv*2JIZaw4&H!zfi at RB3
zR)#9U(yndw=~biM%$Sw8-rna%^_jZ=<e08Cv1C$ma}%Ln!F)C&-?!}x0%!yU$!sc)
z_~gIxzYYw%VmVU!g!PWy3OYPm`uF$%z5RfQ%!15LJg~I1vdVVbv-_ at B5sg&@?LmiH
zyTH?l{5`Zs?@U8FJ3HU#z1%ya=eBkRBACPU_s!DvCYIHW4dwK;XOb3XW|g42{&mI%
zYl(^rkGL$dqQb%0_9++)ypE0ze!Hpj=x!Dk%uxwn2qDx)A|L>t?9^gBs%8Q537OTw
zZ_cY at l>pwT3^&2mt2Hy;`+*S=q-an7-{YN%j{bfh<K{2=tr5JIo*VVFA+jjnmIn)H
zApESYt(_<R at 2tYmrXKC)P<QwBq0)$mn*sX)yg+gm4N<wey7s<&%q{Wu&6|@5#{f4_
zYjuDk7!=fW0`fZBm3ZTh#9LG%6sFu8k6cVj>Kshle~XcA&-3mz%?#vO at H?+k0>gRQ
z&;G6iP6U|cK at FQNEG){al{M`YI?D$EY=8?Yb>5yidE&nvtY5vtadvgh5gwAS{hz%6
zzPGQ at 1~&!<2Yc at BI-_*r<RpN3e+{{WpaylU9~>kD$aV7`FilKOMjMX%D?CKeQ42CM
zGE`6(m2#}8 at Q8V``r|`ddb;iAoDzDx`T&ahC_&8F%F0K&hhLro?z!`&=KTCgSn>B*
zu&TN`=*yS0V2b3zxxC1pKkn%|+x-ai66)Rtxc~zYo0 at t76=2PMMiA|`F7wRHjEa*p
zNN|luRFpz2_6i#toA$}%n3x#N==1&E-ETqAP5^eJbOSqrG62Slnu7!FsBiD>oe`9q
z5xJ`{?dR)T=}W{A7#Vp9F3!g33dH}tY#$ww0As2h$w5bVrnR*ds1wDykgzc7l7f$q
z9~Kd at 4p{OQl<h|tbShR>T-3lGvvPA`Vc}bP`Xz9ND>Y6S42F+?<P=06DzFdk?T_P!
z6RV^7+RvUbI$=lrCngL~7x~{`1;V&Z@#y8ke(kT?(TB(eef^61w^313pQNLs%gWBK
z1Z9e9YlTHdZje)B*Vot4u*}Up`zqp2y1cxMx?Z5ZoaY2}?g`7#5*is9jR7M{%gTy@
zV1Q}hd_HzhF6wT`!y)y8gd at n^v9U@|q-8#R`t;P?{5=pHDT~^7R77)hobz%u<JCw7
z6@`EZ1 at IRCsicb9Z2gWFM%~@#=I1M`t51!LjI6G&R~6b$cJFV7!Jpq0a=rrQwmwxk
zneytz@!45#M7xk>-*vvx0==PFK(r2b2EH{+=mpnxR?W8%R4L07IS3U1TczdX5ZIxE
z0C at _%A^_6>Cm&K$7<Ug23}=5wJ$G=J61*LP$Q>LTv4DU8N_BK#1Mqg0{5rdVfkBql
zARjvA4(wlBalFMY7s=<3&01*&CZurRB+zvpKZ778iP+(zaB^Z|WoxTH_BqJ|x<(_e
zn|B?~qy2-AA3sK2L#CTUbN0SkQ&Q8?(dmPl=YKwi12Q)Ygw}VRlKeJ$@+KoLG}x%Q
zJj5Co7Z>G~^J7a(m-Y1YcK7$I>g$u+GAL9rs;a7JG9EuBM19lYS^piS=K#fG?P=h&
z at NBeqXy`+GyCSNkQ|P=#jf%1w)Ly);TS{{393IH)kf6{|Z=Dh+eaJW at P+<(Sot8D7
z_LFbs<>ym_$N>OBkP0D*XJ}}MhO09{2BeYcIlQzK5R*WSA6<t^Ujz?22+-1=MzX|W
zGMvrA(sJt4tI3H8B>7<spGP|9IAu*38ylmtU+5R%p}u8i27tBuSI=T=%Q(`?;k8(B
zE*6#PrKP90BcA*E^`?&mbzw=#XmAau-ITOQmEo<pq@;HbBT!#DkWSARds#t<+g>s{
zdGh2_ef>&IN{gODwDXN?*J^e69UUEuBxE3I;1|S|m4mU0J!)1~te=8|i)wm++2<A(
z>@G9_wA5$mSMTB}r>dx`+FsgFh!*l47|`Uud)Lmfx4!;dwKx8(V2emBSRr)oja<az
zSm-R?O`l)C?tzvNUeBEa5T~W3eJ(S2=FFKZ7rmZ4I+ih*X${MtF|)Bj+8VI~FfUB$
zJwWQoI{(S**Iiz`$bQWo-QC+u&C7c&H8oXQUcL?<seNPw at 6Me&04xSTp^yp&WD%RM
zB&hG}>W&C6*KN$Up<YD0tK<I~!jY4svh3&1o$DVQ%xNnMcxGWCe*gZ7ix)2j!pt5X
zx=l?@je+n3yOC8;FuU&wXom{U$owgz6c#4Gb?X)*8=HyXA5M?p;NWVyOQy!gwk3Aw
zPM?0dKGkUXT85p8g{2xY4 at nvThVJg}Up+n7sHmvG_vj~wi%KXf6M=xNXH@~3ku)@<
zLB+;F85I^5B9Vb|r9D=B4=;57Y^uupswy!fV`C~Bnp01nJdsvVkb-Z&!>7Q5um;XR
zZFxX9nir5HCMKQ%S2UCw4$c9*9IE!3^eD29ujMJpyAyyR#sPvcjKPhHii)p3f4-WK
zkT3?>jUwFVe13jD!jHEXh&c4CPUl%N;QW2m|Iyd?zJmh~0RQvl0bUIaji(?=hHhTa
zYDpsT^72BbTOf(xFGWqwi`?AY*49Q%pP9 at f!`wvyX8Uu^hyYDJVH^yAoKbs?fn2 at B
z$4sDT-q at 7yon=snu7L}poeYVOQdFHJYNAve7rL3i at 11QE-qDWHP<aWFiu#__Q+P<o
z%j1C{eB-#-i&T<q*|$ev%zy)&7gCc!6zYfRuCCQn81AeLGcq$zfYDJxg}l#Sx>Ox4
z?5e1$>ILLtfZ$!&mG3qy>mHs$YJ(9MDqIDM0O?&BnG2w)yn&s!ee<7Udiwe%A+v|h
z=FiX1Lxgw&@d^nEp at PnZ#l=-Hq3_<kbMtO?5Qprzga0Kw+;?-+LA%`T2B_54)zz_?
zDxiV=?*5ZqjuLMnUubUMz6_8oWnsYr)A9~vSWC39)-qrI3NAn=%5J<{LO$QW-$r@}
z{JN#*wdku?uTlb@*xK5HbRC;P4NnRMZb78yX$fX50QsW|Ai#R#Lq+z|GBTBU7JaB|
zh;{2<<ic%i+At{}NDCbouYV*Ld<Q8xxw|&`?Afz1zSU at cTWU9V_l^n=kMe32bTUTU
zT0u_ka)xf%MGz}+Yh<WvXJ#gQGwTxl$fw-DY5DorFI>2QSfR*%HZUZ_>&utxB&4Kv
zcSe6p;DO}Gm)5(_%)(-OeE4E&dYTF{1xP|1h7|<v1Ao%mWo?QpSAJdpmGbSk@>jq`
zWJx+Z9|MKyAMUQ98pvlc{cN<tG$FFKYZH>Fa3TaKJovA at joG3E6$p65n^7hsxQ|n>
zLh_ at t!I%orAFqpw1P%`mIbI$*qxOF-VVvzfJ=22PA-$ju0pU^p0A2tPA)yYs4&(zW
z|AhM1KyC}`I at oW_Jj&1~rl1G_4G8e*?CktJIdOk?_nwqga;S&k%bxD;F|Y?<Hzdu?
zQEw>9;Krw>ru9*$xk1tJ^Ydrj_tr(11Nbm9HJyYBK{O3iO9lK3aARX*gF3k=C&}pt
z-?!@oVFX~BkeEmXRs{kTT;x573<+uiOH0eKsVUUj3ALTAt2 at ib$M<ht#v$n(@_NW5
z6&8|6N=i1gwMo5vDFOzYDoFZUe7t8zyi|2Z2LV!GgTYbn=qbQnkqFq)qSk5<x8>FP
zpEb<Pn8325YO8PGo<^1WdwY9P6)gyc@%i}+U_29brPD6^oQDgb^P*rW4GayH9zQ<m
z?(UB2DMR+59D4A91c-xK1{^Awi)4kEIusQUjt at iG?$ZZ at TRaU?pG45<HtNG_YU-^N
z{~&wVCFI_tN4RX-g|99Vvr)pM1Ew`JHr}_l=XP*(<b<Jw3ACGhym0saNJj^9?*^mg
z?)(mOS0W-K(Uc(?hJZJACwQkHCiQf~w at F56!`EZ at IVT9eM~@!OjaG=FG)9dAnwwG8
zYx4pPPoKc3DF4a&0Q<<)-oC!-{{Aqn20 at g`g at sw!x%DdQ>fTFBmZ%9RnWU<^y11Sm
zg@}j<@GI(4HTdk(F8~j8QWQxU6B85X<GooV{6mIY$`TSz>+9>Io`MJ^f!oHxm0@!j
zM4QE(ocPf87Hg$F0Ft%ME(ODZTr+cSwjr2&xEX|1KJVYZfq1-rTjNzvx_WhOtrvFn
zRw`H|02dVAV88e9y!FHC!3&^XJg8JXfG7RvIL8OgAlXFzmxuhvyOa3HwD<J%AbJCL
z(E1~J!Ef<|ApYu{Tvp+W4SJXY<YPeY|N8y=Ble2w1BkUdZT;axT54)%ak6rAr)H{_
zmY2uJ#&B3zSorYjaz1{<(=M_*Gusv;1>6V`6a*R1qMm_+bNlw~v4w?;ps06ucTtlj
z06Zb7)|3dqgTuYqnB9#wQQN~^TOMBCN|*wmBBU?1wY9rVh{{#~eozyvj})S4eeUy@
zFIAnLgov)Po}K`B+r<moqX?IK2U%NFTT2C^AJjY}3ybHwcW2QJg6M|}qkdF1J_P0^
zCingzF#yg3-KeG}DR|T8<5g#*6CZyd;XDbd0ridMH0z}07Z6CDPAdAa{tV;=1U0G?
z4(Sfle*W}HbUgp~_?SfWr4Y!y{)iImg}FHdJ75)G*VIUWvWm;iy?S)8;|+*-cyz>R
zKO>jFR%Q4&f_AxLLFh%MJ!T8xF!<d-Y1%7(wa5mb`-Myfdx;MEe);m{dv)h2Am@;f
zkV1!fbb_RiYzZ!6d$4VV^9v;2_{s_%=m2LImnu|S>+Iz*0lTCERTULVDykEJUP!m<
z>r?rhuYQMHV#*)_5=KK)bGz~5kS7dwY)XocmeE(ZV;DRhK0f47S5{Vr*x10pKlKeQ
zONfs at 1!Oife+Ct|&B%ZlI}2uid}0Evv!F3>`u~`E?|82J?|=MNNg6~$MF?e<k)((Q
zAxaq`TZ(L%*$N>cMK)R4D|>|O2$7Y&GqTFc{GG$~{`|iFxNg_grPu5Ed_2y%pZonh
z&N&ZD%ik4Kd*b8c`=vPQ8X82wgkQaS1z3k!R- at -dxw(9>K*KqV4PMlI<kZx4XAISg
zo>6pmcFxYtUDwb!h4~^(a!hvfUbxT?Vg<s|bwEs{0tE&9B(68`&PPCK6HCjNnVCGW
zM&LeQEYH}ufW`Uw`C-bSroR5Bni?(N*f{{%Cr{jQj91*$Yid5JDN4h{LqLk&zCAQN
zI%+!gy{PC4##|a28Ulflkr6c9(_3APV%{#Xv1h*3*K6wO)iMP?WAGI>Szod63kX2y
z1R)0mZj?4VJDWhrP&@$f4teW3pN at _Wd_cxx%ubv^&cwvz?&DKk;c*yTP?`c_4lEu7
zfRT%Vx_Wvqii<Y{<0$zB1j>P-O?B$3o&j#9q=ffM#>B-9!K>HQ)LgT-7o4A;M^<$^
z_`+Ub`Y%F4s0$ar-U8a!9yfl)H$Od{9AqOf;Pmu#6n&yTwDo6<l$Deoj(v}`&zY_j
z249tslzci;dldKpo;m<D<9LY_<y-<`Wo1Q0{qtWUa{&Q?><uIF(}(bC%FT`S<sqM|
z_mq|KvTiM~OXNu%#2yFAxqP12GqZW0Z(x=<eJJ#*Bkk9(Uy+9b<(80;_z)TSoLj4^
zw7oq*z)QSmIv4EK+S(fL>ozq#4=Uj4!o+>>7+mPRU3fKnZf><Z1*2l&66d4yCu(bJ
zfnS!Md90(YZR#0U|Lt4<&`|jqZ2^A%PmQyf at 0?v<>cgxV$S}N{b?x8Zxr)PgORK70
z0`euo?%*I9%<cj4FNGk|c;)W_JXiYkbu<xl(e6D5X)C6C3Mr_m_xp<DJ(hSUH4BUQ
zh|nsa(<y#_DopSA__)cjRStiBhcJBwzJV>9-oo1YGDNNCz{SeDx<Wu_zO=p#QcbbI
zoRy7 at twU7k0Hau#o41IF2#QFSvu9lZnyY;14mvtIV(f-^rkAfa3(5d<mV-k>zHi<*
zLA-yWrw8hC4}cg68JRZo>!m1ALL8cfe=%wX-6n>x)X~vlGNc|Y=2L~2V1mI*{hWFX
zfX%|nI#5*uE{33W<k+!ja8aOj-5syNOme4#4V&xcft+7NECik^=6&+|&6~T44;&|4
zy?Pa*F(&q56ZG<ftSu~fF)09ORZ(5d$j-iBgh)X_VcO*>SaosvqXO(JS^74I7s!T~
z9p^V^IeGFVkV(W4W#vp;<%2y%)`!70x(6h}+@@7Z9B4`%+s6wE3(fS)kS_rMghy+y
z6bM*NpW^1GKq at FMj#Cs)AKHtT%i`V3g at u@p^!D~XTRr6n5nEbLu3{uRHZG2lkMCGY
zN=o<3io?uO!z+mfqz*IEou{W(QS1SvZELUUp5y!iK{q1F0pFanvw4$KQZyew-VG56
zU=EYCRaI5?Ie)qfEcUmjsfz)X;ZQGrjf4oIZ~&)zl|Q?}_qH}-0^WX$GOxeCbbS*3
zL9DkLkIyY25JLU^Yi;eIWo1)ya~bS%zH2?~%FC3LbD*hGE%&SPXXoZdfBJMtAys97
zSRQufo}wZ|>3!rBD@!v60Vj>(ZosPoNniM>e~_-Wt}b6>sNp6rFld64!}`jX?JFJd
z=IP(Re`{!I?E-0(ost`tn3#yu-Bk+AabidJ&|sw}7|zbwKGll?#|$mnwBYx89~Rpz
zut|ZI#_Ok<n20f^UwOvGbT|Cl8N(#~L3^-aQ!_KX40~N!MFns4^%W=DzkmPhla~(x
z><vp%U*zLMhg|u_`lr6W9zy5PTVYRl4HcEwp2s+&l9LY{@*n~mLt+>j7KZ2#76Rch
zqh~?RMUHJ`qd3f`!P4OAQ*I??<sAh0RW5FBy!2O7Tbmdj5(BHS8<}fR(217^PV59G
z|K7dt4k;-q4jMP=E%HcGNE-wv!6>p(SX6WpT>Z`0M+AbQp&^GyZ&lT3c#Msinc2nJ
zNFtyPACi(hL7Hf_B|orpE01UK5P1O{+(_p1_NWR7r{914a0Xxmdj&~z5Qs4tEP}2f
z(>njVyPlq&`H%eIuc+ji?0`20)ox;HI{3C- at XD2b*!$GySKf-B21m=+c`kycwtv60
zy86NXk`Rx;kP!Hs7q*!B13JOQ%j at dl;URJE<NI9jV7dGE2RgI$S<amsIrJ$_CI1D8
z6{A071Pm+s`kp|x%{-q&{Aw-U>`4G*VwMUmjTCk>S)SEI3)QN6{Uz$AC!YQKOY(yT
zmp)@4l<}*8KZ1qQQ&UR|Z#};xAi&;g_wfGxqq!&M#(%`&P+{0oAmK=@K-%L7!knB@
zffh9dRaLLouMZd*od&uMydG8*v!|n?(;MyG-6tVfk&_VFxp5v7aY;;M^Yi!b;kjxG
zGU~|Dqmd~oq~KJbBbE$K4-XEOQLZdF$+6u88b)^VlGD|V<z28vKmsfLSY^4a|2%qT
zYx|dfhG at -dXb{uz{1*GTxw&B-ToV#laq=T at mU0LXXNK4nRaI$#NnuBlh?rP3NJQ95
z44DH%L(a&-8yci^FLF(fja`GKmywZq1ELRvhstN0qr3Kyi2%J0c#@iy20~R(1{@6*
z6`?{(N(yX&>zfu9baeOhWN1~B6$@(y3XQBROm^?U(_4h+dn|aTFp$w&Lp42P<Fxa)
zIKj3bKYm<OS0}8k&2;|!`8Pm^sQHXAg+xnBi|5n$a3dqFI&B=R&VZ~&(@E*+Z$kF7
zv#|}fwPHTCtZXN6wWV)vK_MYp5Q70s+?%s<|FwZ23E?vyM4RB`{54v}|EnD0a(?!E
zetv%RcfQTm{P&$eunoz-&PD+w0Y-KU#)Ahi;KUcQZtWdmAvy=CF1XK_6pApt!{pfR
z-MhP)j-EJi0=&fu_?U{SYW`^+kh=a;o;;f~{a>6FKilwDJmki9A`qq!u*179<U~g9
zAmF8bV7Wa!jar6=q~Mqb=!8z%*w|dzWLg7c8EkuI*)Z2F*l4!&YpLCiTTHB4DD%Cu
zv?1UyN-i54{&r5hv`>WLIWTtwD;=FZKvp at b>%pM5vKzfg5Yqv;?#alME$nWHxatFL
z2e$Rxed|WSgPlR&jkL6)bIe}UUW<L3MF)CzcL0a#pb>2t-&^2lK<`bNXZO0^I8IHy
zwUiKcUXoH*oy$3S;R!i0DV59fU3(t(mp(s!;mI2r&Qd?|)A1=OHw_KXVNb at oGT3}=
zt%ORUC24!As_^5-XW&`#8ZHt7gM&9xu0Mp=rlzJoe&WR3vNCt#hy45?>WcA`-|@P0
z at bdi1%fL%<awd3ez8k<5j+*}Byr;aTh6MaY!i=gAya4oQVp0+w6N2A>5{!mjIePRc
z?8h}--Elz=ac5%Ef?x9~4V6{$^700QYI|GT+D!K at gzY!xKg{3N)wN^I@$=O04;D<z
zC#b36v&X{SK3%x&PQuH}yL82<JMTHXLh6PkQ}epT8~`p&Kzi?k0DsLLJ<jA5pE5I1
ziX+;yvhIbQNzcqI-=6N`4)~BQRc2shR8{GDEMH`}!I4w7zy&POG>wOgtAAoHU9((6
zTl*zoj_HEc_Yc>*OKRV9wO9)lTTD{o`l*Jbsh4;|>@p%0K}55#u;^|iJ#n50!jsD2
z3IsnyA#9Yju=O6l%uo;~gl-A$6Q7(c0*j4g?*9GV339dTufbDZzkZz;V+bT0uS$YQ
zj7KrgpYJ$u;J}BpG{;8a)=8|uo|v4Rv2JN?jRX-31RK?7JMqNr+SG)E{J72UK_pin
zR*+yFNJqM+%4);6Z?6TKR31$Gd~*g;%>cv at g4C$M5mB$KtSpmToZufSgQgJa`Q1Wq
z-?_7Sallwl?-gz$a6@>AqPuTU`Xnaq<1^~M1{^3=wXkYow*Xz(*SMX}3E;@x9G|T=
zlO~qDiiZ(AItJ9~Mn*<d)G3x{zBYz=%7Hz;q!)f^V$t#a`>gT7`<F?b$2L0(XYyE5
z7aaTPY>p90OOuP-y!Pd##2&Anc<IA;qN)OqJ4#1$Gi7v7)}Q;RcsP1Ak&nCnePDOT
zQGb at 5F9ca$9DT#&E<SVC*1>UD!0dtI(VH~2zt(N0BtwIGtckU)bi&QT&4?h7>k9&x
zbxTe at i<y{Q<G%L-(eQ6#LsVZN*bIGE{cPrwnQ8W}_-bJUh`Qf>C3J;WGos?+B(OLA
zQ&V2Y)Sm6os`fUV%_KQQ!E0e~*Y5lj4lk={j+Mk3bG%%Q&S9DU<DHN;jXgSSpc=lu
zW2QK9!UONN6||*<-3F5rzIM$?B at fhf)9EfN;9|p3!$79{VZBTRJm!}!Ubx`*SCEF^
z9B=dm_}e2OP^dJ;NIGH_BcC*4=?l*^2=DP{K_8SJXawKj;8*kW=Knf-D=K2zt`#?z
zc#TjMb4w?+q9`G!qI%WmxJiSx6X(v|47q4HOoj~vf3Er5E%0LF<h)C}#>2rOHJuDX
zdAF>rEM}d+yzz#5o5kNhH5#TyrDA1JYh#JkvuBqW7#V>P?GhFi1_larj&`i~vp|9Q
z_ at 9;;!Ts<y3?LniFp-M4PuSUA0k)6C3 at CIy<FlFG&(tb~T|@Y2-wUmbyNj$dZ;?H@
z(GrS*Z^w=ud6rXDSO!Pt7XP5`+c&>7u~;zxPp7M^%i`;*c;E7WrwKoQf&>@{q7;yv
z(rPZ90`x~ll&bdm6_9c{Ik{`<>U2g%MpQKGwhI%jAIuo*{pC6oGiwxUoqLMym8bXb
zCm42SU7u`A8QFMi*p+j$&~i!xurSy0{$_s05&D3D01S?oFJI2?EQ8oKyr`K#lWaqG
zMO9U`qpM5baL>EWB5TfDqz at z{UR7B&y6!(VY}Q>>^|9 at mck`0Z=g;dr>nZ`gKCG;)
zAWgrI9kT^ZZ2hK^L$#nL^M_>L_4*9Kf`r<yU%!BOMeEU8nIBio*7I-Uf9L0S#4BoO
z;;Lk(@h&G<*WIYwb#&Yor at HE<1Z7se=Z%x(Q>e`QtO<XX<QFW%w^cBJ#V%c4T`cKp
zZf+iCFz9dGIH_7--qhLfLylQIdQ at _vMA80__ww>HSaw;XO~a<B-P`9g at lB2y_mkvJ
z>9~}YyNJ1;VYkC{cRu5iR*~&eYWLz9LHxu;PDx40yn=#S-s`ZkCRSEDh9NtHMSK^X
zfxUgb<^1j2HzgXfZ5|LQJ;n$;rfCfHm*x(Lmk^Zdc2fh0{Bhq0aWk8P__dc8FNa(b
z1iX9qK4jd{)lKLoQInH%`SZtg_}4E`86w5k&$6=%<Nxrz5j+*g59Zm?*;yNQGg1AX
zcwO+110nMUw{C^Oa#?H3CWy<e9E!J|rlz8*z1cN7I?Bbv;|xM!O%1;N9j6@#L*7FI
z&8v51;OnH_?<*>j_)EA#xB$|oedC_^|7rp7+7qCBEF2u7IBD<iU(i*jt$O376cmo7
zDCfR(c6L at V0FK3CB{klWp%s1N=4;yI#6*edc-hY^>iot$=g*tW^a^_Y8Uv~n<_qBv
zrxgARg$Q-C?)|xuzxTz&;@av%z~59Vm|Wr)*53sL?4c(DaRU<e1t1SfQy?_1uC6RB
zEN>cnV0*nDvCiuJ5m_k;*f*v}7X1APKM61cdIp9Y;g1DCy1<IgE-qF9gTWK;`Khkl
zrqeEy>{qX<Z*=>=l1thQ4G&LU=)5j*UQto8rv2i_j~~-NeQI7lvHX{(P<u|UUOG0B
z`Lfeh<3lmjPVfl8B`^nPZEX!){`#$3djY3Vn*sQ=*$B=4vK1|O at 5{eWe)DHjQ^uDA
zE7wUqC1T$i%glh$IR4Uk{P?TuSD9ZxZB$ZbDEb at CN^F{P=GC~)pMQ~%kYIP`A&46Q
z5PX(Q!xP7kKbh)(@V9{2X)Rw}R$AIO^z^+1IpCZkAirOEdXCr<jTc|HX}r%_ld7g6
z9~K*X2E;1wzBc7tBd at j;<+^qqaAjZWzTh&eJ?sDSH^<{52C*=h|A-zj{QakYN9&as
zJ^Mf>z_$Y(YwGB2X*mK;@4u&ki;vF at Tnx5pq)|gvwMS{`Me9|o*v4BP^Ospi9^(GP
zdq+hv{43^SX5M2p-JP&-!T`t+fKZCpbeMpZd{KO8I7hY;Dh5nOHG~P&eETLMDoXt6
z)2HFb5NN7C+b-P$Oq%=$M9-A at C>Z3{HXLc at b%F1OpnhTcM7CF_l39>BjY-L48CsD7
zA*9v73xHekntC-yGttn{fE26S)g^v&yj5UC+`9VX#oFIP5_j%Yh6~!JZ<o+r2O|Zl
zUjc^6Z8QJG`{Esu7V&h!xu&ke at RMOWB7wJK<As#AGoRgpcyul<dq0H;eaUt)`O1$)
z2v@!l5eLD|G2O9UT_=W^ZTBmA<z7<jpHpeg!Yd^$LNb}_w?P|5+l%P^;Nj)<@0yP(
z`dQv7)}m*T@;B3FRs8AG4ACw%*h~1QrR7CEOS3l(`U!VpYy=`&)XExIYVp{X&T%cW
z>GCqjSJUZCz#kKH^OhQg9KF_kz!*UXKTl0fy+SSH;2_kZm)$*j&z@@<6APCwpZNLn
zCt;)L>SSkjQn$4N8-s5AF3C4cum}3>saW|4(1Tguhgw<?9@!j;J32aQc2z0RC3gUm
zIj5B66dq1HzpyYe!<<0(d|p)<#FGAOe^%Cod at n^Yu{{uduPsgY42+L^z}l9Tm+yin
zxntVjTvH=XL_~xK(^H+<viv6*65Woe+ffh`6T^zTHAj;BQa*U_z}?3uVE^}+y<!la
zM?66E%|yr86<wGuy0Bc-(r{|wR=h*K3QyXbkP!AI*}D3A|2>wCb#?DN?ewN|&3-jd
z?Y86K;4rbUc)U4;mtE`xREHQ6&-)~;g#*t%ZI`BXL#YoXM8srg^8!=U06STV;|5EP
zjEt;0f6bPT%CIhR>0MLK?Fs64 at GZHXx9a}<`EyTRo+;N^chXJEzyK4m`Q)lp$r(1b
zfqu+97cL~X&jiIs-nx4?fVuE3 at gRg#(Uj}o at TL#PO+_UoJ-a=(W}c3JUXSm)+Dya<
z%Ok9$6kJsr3CnpcPAcpIG1qu=EZ)0DW!D=-+XrG}woTa>EfSkEfcY&droDUj&KbxG
zo&@vo$Zmuu_4d9C3sapG5LHl6;OKlmu at j=gh=Zi4sOXcXDA(?*N1Q^<5;v%LbxZ?H
z1K&>A2geA}0!?LsxF9V}ff*xsr}n+#H-GX at 5*;8V1(q$Fq*~qEOAGGpx_tr3%b1wc
z$*P5v at 0jk`B=<2(%ieqNzytm)Km7q!h4_l2wzhVPe$V at J%29VhE}9y;wm3?!*uCl9
zl-cCGRiD!vykICptq4(QZ6l!ba|{(<g1RmIm*{PH at i%)8CaQIf0w>YY(E&~(7~6MB
zFp)_;MpXe+K2@}5%UXU+OiWvVI$$W!$YHV*o&2}tIi=pV{4P~ozj0qL at BOQ2cxOz#
zX=!QY-%A8Ppkkbt-26oEOH}}|FkchkLRQPgcDye3!v_MqylLG21m~A6ifjo%?g<V9
z``6NzZ|Rw2XJz^AFNx(8Gcz;udJ!{0x^XAtx<9RtL~QdVh1eS<5JmFyg!y$OB_xKW
zgh6y=zY at th`dDGck7zBeMr6h44L@&CaB%JJ0f?P{4APtP*#%>E`j3tx**x2Y*-#*2
zw{PG6o~C{ktIAHDA`}-F*B1nS{Pam?T7$vjqObNhf8xwH{1<NRZas76%%5K-3sN7P
z7QLf*4<Bqbz7(TR4Iq?d at Kc6kXMmB0fq?<uUE~#1Y-nIWRpBnzVkchNxnz6f$dQ`L
zmS)=<6C~$UL*z;_QvVhEwiNv%&a(PWe4$UQ`JxP?zxg{(Lq^rjEU=Qc*Zr+ at XXEvX
zLqkKk^xMv!RZPD&e$V}$v^0HVOGr>q0=J2%np(IhcebCOTvBWH^zI}+y%vM{H$t4T
zBduw1=gyu*2V$0webJI**j3in_MUWFP6y(1wrF1ft&CI#EIX{1=`LlEMi>YmJYZEf
z)-Nn>%emdfE4+YsUCXYdq*PW~N&utyUc)ONG4)~P0Lh(v^6^`*JUqS>+pi)*W7Z5-
z_xMGF<9NS;FQb_0DOQFkrq@#**^eJR>i6U<F0JySj!aAQ`I)Js1r#hjJKH{L)mo`V
zj~Bpf<b*P?ysaawYt`N~O_ at 4(|1|-P213eyaxPQ${{1t#ZsL_`GV}%jz1<&~No!b1
z`B!5)#bT~MjeYJDlOw2Way{YBT^o}(U#i4nW#}(oeh at y#5C5EYIh|d1>tFlCk{GYR
z^;^GQv`n*;rnlvr{bJ&I#;prVAz%;DZ3&4RN=j51*A~+Ng}yoy at 24LwBPC!Ieg6FU
zl|UB{507r9GuiiV?OM9!yvQ4HJ#t6?_ekO6+z*KWf%Ggmn4+qNz6J4+kdRz6FgUBC
zqT)y9=1wcUa3Uo(Dk|8u6*!0K^!15S^>v}-(b5MXD!>2u at vf^X=7uplJ3H{dDAhvC
zSM~jONF&S})jzK%t!&oS*1m~|;9QFHWul8OKJdNQO|<J%@NN#?fLr!jF&>mLJBDK(
zH*0wC>TJHJy;9<l^^W+7IN+S8a<=`ZT!bQ#Q<Y$D&I}!fe`hbn2M at kfRf&cti>=T$
zUp%poR~39N at T#K_R^(wxI6!}6jJlSlW at L8`pjS#$8)IuK(6O%;=OC1t(lI5J$kgwc
z%g$bHNj6N_*m$$T^FUFtw4)=ht-l1ohH8qxxxBNpKx!8yP(3Z6hj?8po;VwH=lR|Y
z6<LBHoW`w{-Qpvyx9c5Kw5PZCOD``{7jg)}rd_Tv+RABuZtuknHJW2IWYk4<KmDLQ
zr&Kd_QbyUWg|bSisml3gr-+D1%P{c#!xR)0So<s at LUicRAuMPUvR^p`z6;xhRdSIG
z6coJd;o;$Uoc*3dHPM^}{;Jn2iP$nchqszh1AVgR>({SO+EZyny{KHdwW>5qopu-;
zp=XHF<^+i}*`8JnI#pUp2`{@JT3_mO1CHDrDfkJHzM`Vy6!23#Fur~Jz}vTP32T>?
z%^Jf6o01eb^PhE9R=VE&0FQclZL!N3kAxEw6F(#+C7E9mnGby7>^v|q at Eiy2(d?tZ
zAL(<L_-#p$#cv~}zNpci_f*2I<#gDkr#tbe9!zYu{NPoM;Tm5FH<1<=!=Ima8yC$H
zgYOJXOeC6LYH`XC4*?|`Tm%o7|3u)*m6LOGb2x-A*Qkg5$dRo>E+Qh67%m0|y=E+=
z9~p7w)o)|L2atAF0516E4LL5!o`bX^Mn)V37LzY=NxlVgk9FmyReke;<(rQi9DImf
zLzqtlZ^1n8S(Uu>rpA8ImCVPCBz-`IVJi>tP+D5rv8L$%RoULXd)MoM7Ca%9^Eu at l
zjjGot{=9x7(z5)!r|_8U$J<}LsQC~zW6Od3!_Qj?GXvPjf(4(iX$7f3d=J~#$;zv!
zyteAK-t)h%mb1z^x3w(Q8En!)@#N*_(}M-{*)N4P58T^T&0q+hq-TlQ5g<5NcL7W7
z?&0xVBK8KTc=F at NSsHVCGPJ~Eh0icB*gSlBkj~q3s#6nqwS?OZ*u7I&6TP{y-tpO%
z&ztTlnZLwt(EIo}*h#=|F%e_#j#ecKgg(cp=hfr7_w at AiJBF%#ss}4Q(G+5ZZ6AUH
zqM)Ys#1uJRZV2cAV%epr=nCPQ8yQXizySR*6WeFcJ^;u2-rlZo$rPS?>Y_o1laJ5}
ziPcQ+ef}d1HhADaJ3n7B{Po>jd&$QAaTE8gpGV{4<9H2bRCIKPUMm}xgN?rl_6J4x
zaP(Uc<}h-M`x-XpL at cs^_!O;97v?#v^MgG0Cv&5^M5zmYB%O)dA0N+RoioLw)9|6F
z=;Y$!BGt_<P{X0DUa(~;;4(K-<c9*(9i-u57@~+7fs6uBR>In1tyW-gFrGAH!Gfsw
zNe}rHB~}`6K<Vg@@<}){{Nc|}k1;pGWzfNe)i>az5^jqvKp`-rl97>NI=7gu-~K at +
z0gvj9dJ3>0d3t)<#LUdOuuy>C@;8Z~pkS;pGx(+vd=?f4g46ktxw_M0^W*FK5I#mB
zAzIAH;?yblAtGF|nP<TX+rS9d=6~EB`<WTnOwTK+kfv5y$64|k90eNyb`c-A$A3DC
zI0bZ}YJoZ4GnHY~!yj87DeANZHzK@%fddHb+i(NS6+{map0E7-hr`R^Wzf(C8t{d=
z!6opjIIx;68;S``Ok{%((ZJIBOr3*=4j*p%c!wHo2v2f9TzA&MNiKo#D|fufcY^sS
z<`|p+YL4~Rm7d46D%>gRA8LGi^=e;5W#!k(E->7()+7brw{Le6P<SRKu at _iQ?E{{P
zC#S8gM}gAApUYgb0o&(zfA+x(P-%nBv6A>WK4Z21>|?D-$FRL1QrL}+hHMLeC|t-D
z^G(le{0S(*0!>_AP>%j&Ne|f1ovREnF7{A&TtA8<*R&f#6U at a$NZUZ&%r%M8p}pf(
z6Z|Ljh!kV)AV`q2`R_Z%p>plfqtjS_3$8ajJY4#J&3)IdT_db4PSuf_frn|-2lMXd
zGwLpFiWc*S-E<!dVq;?~0|tS-0<Sj!LhcOj2!P=Xc=W+>bK`q!>wW&)V1?S)h!J7G
zc!G?Ep9la3posi1)9okkcRi$}q{QPxtiAAOS1=i=Jz7y+jdxOk0*AOO3 at l^hL9VKb
z%J&pyKCH!TZk8FJ^>B~(2;wuQ#m+2v>{wy?D6tRnNfgJAyGw=f54h1e+yia2^>gIt
z(E%^M-qKXn!gTOEbSyyTO)-+^*h%vB?U`qjDY at xjNXG$XXegzf-v$J@!RHR8<o1^7
zMA&}>%ob1I!t61Mz4Tb-dze}36axcd6kb~JKSsKHHwso at q+CFk`BmVn367My)m2sS
zZ#RGsy1R!L`1u{#D%M||D$mfW1|GMUj*brNg%GB;sEDt-d;HJBXLAfVN at HzhGHt9Q
zgApsn+SP2mc4-ba2IZ!ym!wMP+2ir at XPcuC@d-|gQ(br#4><7WTNlt1*RZg#vZlin
z(ptRsE9O{uc>1&mJi{0TUY{))z;V;g4sXvz>VrSQAQDcase^dLUey;bUiguDWZ|XU
zOn1W%7#ka}EtPD#!cQ|j1 at CbjZ<1s16R$or`Ti)8XA7Mu1b6Tm_kvxyT=`*n<Hk;q
zw-CE_J32Z7Suca7fp8+nF9*i{k&;8jeRkGFL`;kw!&=kfIGF_O;e^QM;^Klx^YP<H
zJl4l}FC8uVrL9eVoXf-AUEDK-&-mfLKO;Cd8_#9`{r#l{76<fCfj^5lE-`T*gweM;
zeR#4yI5>!B at Wa1;?ZRVRyo(n0wfyJL6Jh)oH$X9BzkryS*x2MGyWJ76 at q1o$S7|{?
zFOSDXAYx76ga at oK#Iar2^z=W-s_zcs(c!Y0e*(N$t36fK#KZ)k$;Qd^7$?@9<FO8?
z&+j>g%CW)>GTj(5KigU*Kgc8o*p>qi6#35sk6_vqD8PRl2{r|14}qnFHPLT#q at u_(
z3{g7TbAy%G1OWjlW{V)*#rzH)=;E5ae2GPI2#S9juli_~zoh5lBCo2dl5o2PziQzZ
z9`4<Eb at OD%MZ*`qz7+p;6q$VR=a;y9w;JI5&6_t)Son7Obj`s8CuMbYI&W|9>{q$5
z5f6^v6#?n9x>{aWw_m?48KX*~^V_$0*=8h!yp95kq~-zgEo#J*tKZIng-_1Jq<A at y
zhPL)cb8)#nup?!&1Le|cYPi99d3h`}hkT>3ZWEXrBNNlhU35PfHGMC_XEgC(?dCh?
zyVwzj`F`MRa%@Kn%*S6~FTu9xqY<#&wxcMZi?0CdToL|To&Xb}=jW$FUL-C~^6=q9
zjYp4McI+b70AxnR!*4acjo)&20jtVa7AEmFW2{0$d{~_+Impb+94oA!^(ixxcVBMO
zrXA=UaaH2ZEZu17ys6<~7l5x1!fs at +-A=x~zGY$%Htx#8{$ngzn=*Ek-C7;K-e<e8
z-W{Zi25 at v>lK9i|+Wy~20Md9|e{6%cv9SRp^1;Gb*qtj_OabdNKivbw!2JjXCFM(a
z_*17(6Tw#^Y(LsIni!|DvKt~dZ+^`*jJRw at 4E~><n=5wj9{GP=msVC*V-pi>G#tRI
zchtQ<SN at We`b}tP@|M|H;5Fz7K|pBA0Lt+s1>X9~=GtF8I|ah4si&8?wMz3xKpYHp
z8JAQ7KKp_R?LEwd{bLY9-I^nLSsy=soUPp}B`%H?p1=fLKYcoHtt=oQkpD#4<`O at D
z?bWW14wh%9VZGJykO~4CH-HghwHsb&{PAvt_t at Bz`H5D^|2R_oiB;ep7_6|vFHpcL
zwRzyX78@(;m%OJ#CM)y0*d^54+Y75Ue><U74{h*H$i++MV{d>)GFEKS$SC3u$;k(F
z8bYzbjQqq25BQEFl$2O_3Sn(_Zti(R1RYlQ96o$Fwj6v*bPQE*Ft7gB@{$HCrC at P$
zy5pt;U{U|tzrVos#56QcVKv!6xf?|8`&(Evc0wjW4xs`YhOjIV7P`F5bw9<CBQfSr
z>Tk%)Q_9H542zkXeyZBc1jTB2=guAd`r>m(pXM6((E at Yhx_D83_6dBF1D;WVv0~4z
zR<$>5HiL~*VBm2)3md8R59HN<|9?)l7*Mjdn_WH`FVDtie&9bgAdahD_8!J50KdC(
zeWtitlI{N1^95cuW8zRNA_0p)mgK(8l1Sfxv;x9{t&_kd;ZHD~2(UNSQ^;p$XGcy>
zPEc1@&ob<iA2)$OOVN?8xeMtD(8w(r<ohx-^yGhy!=*p~@FpMxLeM;DLpZ2BFeHSX
zh6$dR^)|_K=RF4zvSEXra&!4GmH|Jw20UndHaG>sxmVOwuxmF~sm8~@$u;gN$W}>L
z{rs#D%pNTpg5`snJ7D&}1L12J+1U082?-$-FZl)qc@*2PG61(z&NaIKU&z<18f}d5
zS(t3s0_zXF^fU^r=&;O#2Og}lA2s at l?UCIPlTyV1JG!fCX?amaRh9d3?Gd}V%46`U
zUPnB$JW&34K(Gdo9mMWo>p!Ql_zW20B>{oD-CYpn$J$b;0j&MV0^oN8!3;GA%3N#e
z>JS4M^hRCCLBD+2SYLfQ_MJWy{FI-cfA;U+@)}=8Zs69>o;_pPS|DDHrluxr!iNyP
z=h(4hcFO~9SV4V&l+>ThOOO0UKg297%~w*Ql983wUk?orS4(=M%+`^rRy6g0olWHc
zk6)tUi*11Q_5&k at hre{`5)05qU1 at pwm;L=3glme5$M at uH%eKn14?YpGE2LJyr(+2(
z*avemL{A*U?yx$=2ZZig at v$>{iM at 7z@8Di5q^dl}@DF}gx$S47i61@?q@|^?yA~N_
zgwR%BrLuB)b at d)tAYhQ#NDTC7YmJWN*sofC3+DAiD~xf^%DMnyYY at PMgp~9q_)2Gv
zp(h2S=oVOCctU>m+&O=;ceoEhpj4Ap@(*?88oQ;WoI^2*<DzL;sDi+3SXo)c at 7`@+
z&bOh at O-<#1Pr*KSc=ggRU+|u9r|4)#e)BN`E)&3!ogApZy4*eoPwT&<ksm%_qE382
zJrQ;r0j<Lt<|kb_DZp~Jy)7Fe#J3Q>g94VnOC>o!$CzK5_rp at LRt5D+2=U~H4($Yj
z15w_F4I3?he})NI4OSjg$0%R14cILr at p5b`Mb9W8D2TvH!4AC%4=3BTYZqWWHG<OL
zzs}eUZRp`1OpjcuDN{~P&Lusa>!hH~5iuXyN}$G|J`RAU#c~Ycmj|vlh6_eUM at O5-
z;};OceEf)gSIfW`Bje)S9%O0{^jP)XoNP-0BMK*{C&H5c(9qW{EpnLpzI^!vuG at C8
z_H^%Oc4K2eed(;J{{B<PEF|kL{pF`<+C-KP^t=3LrVcXxtmUz2pLClAz0fPxhXidG
z<^JbxA9D=5USijMQd%)KZi8SAKkqMrJy?TFeU8-tn%H at SixkDi&W`y)*dJ_!MRz;7
z?QX&8;K5k+3hsb`Uq$6)iykjf;{Kr_JiA{83ewu at Ur7P-YwNF!=_V#7mcg@)1bu>q
zh?Gg_A0PMl;=D&|w*SlQ(vlkhN^|5P*u=x{- at S8o++6Kb%r&Cfw{IVpVdn+A*)Mxc
zEG^~vFW3-ENJ@?{7$umjg6jD58i9y~oqZQ!Y<xV at es%G`(sfX_ at pHz&ppFPp0Uy(E
ziI>JxtoP at XktyoE(N0*M%I!l=_+MruN}CPH$d at l)mRI{Wy-U_cIN at y@#c$r+UFAdP
z1ER&n&8 at t=To__~c;HcX`F^_1<vlpDJx3&wZ<|o_GUDxB-rnzZy09Vlm8Yksp&==D
zxWZzzw;xkeQ{&?1mPV~%^CHK%&pWr*W`AX6<(8hkQvp(&f{N-Dwq%1=#^XGIQtoH7
z8V9H at kN`(}_;f!aG}f;5*)P#x$Ny<&=AqaCwXzqx_kZ6?iZHr9*4D1A?DFvNz^ZU;
z0X%#dB<+jZS_y(vP|$;vvZ;xyJdRQ#cf343_p3RsTUdT#($heb0L4OzCodoGgnaxT
z3Azoe0dKg%I#+C$0o>6^h6lj%KpPwWsOV at -W8*YH=7slQuD~j=Ic|AUeG^k>MB6ef
z{)B9TT%O6mPV7?uZ{EhEN2TAtlf!~wfG;Vzy0*5ap|780o(JsXARtF|%&m8&7YHCy
zQZaZ5SRO#PLIULHM`UJZ27LQ#=HtS`!tJo-dplNUAVA&W%d4a8<cAMC(+j`U0C+!s
z$>c+Gr1UO80N0k57NyFr<`N*m{<+zinLm^&dSWGyb^QFv`V7Q`Pij=uPEawZrf?u3
z<`W$kXZn2RhN_3uH&>2<28#c`Rly`0&aBYXq-Xe}J=E~OS^!4I%TAAZ{39ZKoSX=R
zYa${<fQv`H-N)wk(sP-Q-Rmy4r~cI#;RbB9nO^#E=@v3GG`|T9WI1LK0e(6+H+L_o
z)#T~Z=V at u2J9q9xZ-U4>Ha5nB1hJK3%ax?D>x7L>f}^$=mX%OiS0~9)3k2mQ&f)hx
zgxW5Ue5cz%x+)MvNK7H<;v6Gv_%gS!@F(-3 at +d1SQ*!NUZp6Mjbf2szO*`=v5xzPp
zDQPVetmE6aZ%2goF+_DXL|i?{$;k<Nqx9#`2f+SFN<j`|0{UB+z_AGX#i0M6?&OxN
zlTu$aC<RW|KRgTo9P<HU2Ug5hR_+D{;*^n*0dOt&9zyUym{sp}Ik_V%i&IHpFCyFa
z;!eu~>S%ioMAg at K%uj(yVtv8=BzYV at f;~QKYu|H?)xA&O6!|YdJ241vig~FU!op7n
zEB0bZglwD?n(cpuj8Tv4-f!!x`3~%N8L11v9<Q;Mgu_B>v%)x}=$oO;XA*i7P!TF6
z4yq2Q*4xw6t5cM7Kgc8y2xMerre7>UJ~U<P;|+Cqk2Yw6eRR7icoi^(^PW at P8*CRf
z@(zfSm4Vz^n8?K>{jN9!2nyea!0+5~H{fHAUp9 at wbSjQt!YuOc-2)KlwA4bV!iI+*
z0prks=fnv at p!$Q!w;`s9F(Em*!gc>KoNM6jP6E1uo`>hKsHo`JkGO-dF8!b==;-J=
zIy<)%#X24SJ>QF<^ee!UvGdEt)fM}d+uGWWb7?qTQNF0tgz)!qtTS65b5%J8KX+o2
zX%NpIoJx|Mf-xyL at YSn*r<s^)Z+795fCj~ey6(@m#|TJ-ftA$V>}u5+$jQi%UM6vm
zPp}=24ImIo*8X^w`I6%u*w~wki at dy4vRS`#%kCLuz>$t3x98*IGx+)G=D11xj?0$6
zkK;_nqK%bF{M{6AItwP3QV2iT<d&(^XiCtriP{d;06SZ*BL>UB!9jL|1T#)uUDVtC
zfv$yx0$b)D;2!_rAqv(V*p`DseSozOEYA%AQF9IoIsxlOe)8n&U*Wc7_AB!PPpZ+$
z4bWNj+fyq65AgsPW#Zq_`k`yQ^S6gbNA->vDnxh`f_5QjXlX at 5FrJ<!!|pc$0iHf7
zk%f=)NmzsBxHd!wy#6CN2(LbRBvx8LK*IqpySMngNKAZdl7d#{tE29op2A7;DX`sC
zfuC)c9(LzFT?T=b at uZ0Xf>h!=FKRw)lE$H7EA!*g;4u~dc}#d7pd{Wn3Ly-oEQDEU
z6_wYN=afFEmpI^30AB$CT^*zgzf6HP!FzXnd>m&n01%)MIeq5LmyV9(NCe5qs+ at N0
z!X(*oDk?S_HZ<HQ9J^%wR~wTey1Kg0fPBvlxY2!jN6dO7MV|4 at nVTFP#ik>#*35vS
zxR7Hp2CV*-D_8or+vdzC+eiVSB;4+bc-4&mh#Q!i@&YS2ANzhC at FX||-*>RW1LuFV
zrK;Bb&p_lH9E3tcLowfXJNybjPyBPG5(hz?ss?fL;r9>MAxh8VxI4_ at 7g_&JHy at XQ
zAjR~bc$#lXkYlT;s;YJE0-_==F7AY_ld7r%Lqm*>dG$YYJlw;*r at C^rYW-N_6BA`n
z4A{KF5jR8CKG=|rJ?bD$G at m?icUWK90(}z`uA<MMH8Hy!za`M#f<OZz{t5tU0Q4ug
z6`|L0)Bg2w?iZEkTTTHMVG8}f^2Uc+T2KG|c?=PdKuAtb#zd3(OfSwYK!pMec1eSc
zLU#d0HNaDz`S{!BNOE4~5I*Co%~i+EgLsh&UU;l8wY@^!%n^1kz^ZX<#k%$QbEeyY
zP9PI;0^j8)dg(yA(6nE?*cmEl`~L8Q at 3LREgN{r^PoWi#2Z3J>6|z?fH|zq*jnlC_
zbL{QxG<0<nFdBRH0?V3f5STf|$$1yEqGNk&8yc#BD-M_a2BtqYIT>x-w`DV>&vy+>
zPVSeLmEAh_VbC)NFvhc!lb212jEsyhmx9?zMn*;;Dv4*9nfE{}0Hc4mKX+(w5QobU
z(layRzhZ-OdwO~_wY5K*42#5^26VtQmE`-g()>rFw8!!IN{r6l1;?cAKgJ>V|D=w;
zeEqr?p_u7TFod3#(vjg|Y=U)ga9IBLD*~9Vc+wfk_gm-Y7@$T+F<mUOeZ=QZMFcF^
z^BCt~OM)yq84~}%z<BPi985ug_n^7ZVtyw5+1$e|Co;{?&nG!}Fu?PZmX=mzQj%xy
zXIsWgmyTn**?|Kt at PD^BEd}ohcvU>yv;5t$<NrZd9oG+{^ykjq;pl$9@!51_tI01q
z+S!{%K<|i=75I#5q2<W!Ln8I139 at l`sc%Jv$fe7dS!u|mGo$2_aF`elhEsQ3=S40T
zE4;lDI|<2ZpQ&)#IgUB8x3|Z9LO+N%tntT5m$!q?1KPS at KxPH<<O^Wj>}ZHF8x0w+
zX0ieOR$k`1x;h=eA=aB_Jg%dJ=#P`D3aw`Fs#_c at jlm!6Ha#b&YIdM}uF-MhFg-ne
zO=F`1{~D8|m>5nv^x}W{(&PZTi4QwObsT%Mht7 at wYyv_QA{;sxj at tu;g1zdP2m%tv
z^{|43^2t_~S`)rvd1G}3ClcViiji9XvG0+)MjOI#F=ESYAITKown00DwR#3&$7Ey8
zI<`EC$FLgcx0jIpibh-VgG)~bT^@4to(2rUH4s|v--VYh<0vAWxC3v48S~04-5(gw
z;9>f++tr-o;+U}>6MPhB?redv?UE6mlIvwFjhE}T1=rdxUEf$~FDa|878~c%%ePrz
z1C2p!Y-~I>F(K5px`sox!0>J6$98~c9H*v^mJHy)sdX|kGFd{^5Ud<myZR=mH|((~
zlw{w&htDrI at yIJFy~3Fo5O8q%(EcsW<sN_iXNJ}u at bG&{*9z%Jg(7lZ1}E)wg7?Pc
zGbJTudU`sZ0$I-Vs{Z;_T~V=>{{v!xT~XU%T5Iw5Pw_i<T313>M-s8+0^Hys;9X5^
zt<6}mH3#om(s31==Ks#1ZwGQ+Z`wHp!3FP at PEpEkuHn?t(HR0c?ds)qCuOaqdB*pp
zO at ok!%K8s}oy+eqDNMmj)ojouEiHY?e&qrv1mD0w5>-`Itp3Jev at M<igO8S6*WDM>
z7x^}qbipf<{^TxcVVpN|R<+<rt}zy40u(!a`*vGIRFq*m=Z^E_lu}+oU?Dzm+zCh>
zPO7?i at uGyA#8$*UN=b<;hQl<QfBcvm^yD0LJ}7i8Ldf2aEJcdekB^s&>)MkiEIW7Y
zikK*l5J^qt_}2$=R||v|sKdmijftp?UkvPQY<Li1T{VS6<hCc}k>4n7X*n`IJ-t=E
zZZWwv5~i at Qyth|v{CkM&$3MNFBmDf104>yjCkVWt+brAFIP`t7(6Obj(s{$-{rmSA
zY;BgNQ!?x@&&92^)sD9~)pfObgO-Vt^C0lcjv{Lw*xp$Y^$_=IPGD(x&V;#IFv!xb
zE=)9UA5Oy)fzYjjI6p8l5}T2l at k@yAzHq~Ryh{_zZNt7D)Z|QL+S~_~`|xd`0q}vn
zMh at 2AKYHv~c~1|`^XJbyK0THUTdX&`WlcJ#Vk6^1j$Ma1stZ8%!%a`h?1v8Wzc5Si
ziblY)<<-ncfO=Kq)lIv;U#ttXx;4Lk|7d7!{}noNdn?qoJQ2?{Yy at 0W&NaGu<HpYO
z=g-gfZEhT66!WeC<}wR0q!hjvQ_F}JX^!jr#k@}r7<BpEB54{sOn+zKxO=$E+qZ|d
z4)%ya>;U+>(>W~f_z`zk4EQqADbRQ)z_Kvh;0RD0&;`UF69rY?H2<*%*9VG<8ofo<
z*xiDsUcljS)B%>cg2_<b;gF~e*=7X~Yu?buCnwu6Wgfe7tJP-+z}x-x>k2$t0653&
z^-z_U{sqUwu_vgg${>#5*gm24d08y3*t>Tx_IU%B!2wfP<$3NLDbB=tNk+48yK$Eh
z%oC>)Lc}luQ8GL{%wm7(j=Ze(OdC#q2F%jhsz;IW1e4y_SSlHUm+ at JrvR_@Sgm{D9
z+}IVLo=#rvLnqc(Y>#z5+2uodPlqr{fP`shwoAd&F8DIOnm#^h*`uI-fv800<sFZc
z3Ii<JN{sPd!hwrGq^RnhV(!Wc*WW+G#PqN!>L$U-=jz7s%a<?LeEs^Rn$Gb6K{3yS
z{_Wehv at 86dZ5GNB?<dW|AAi96xMO2+E+C%mba!`)HQWyh3c{<YJ8}%EfkRGMy}5Is
zF;*vJ$q!=>j;nK68E5v|X?VuP#ieF5u~jd>egbx$;Pn0bJy7Jlfg97L3!%HiZ>MhP
zhGmA{+KR!!e%Mo1V7_-dQ&|Ow|HjYnbYspGC}BX=hMXSv3%SyN-NWDGc`yWHaVaUp
ziYRSK;Gw{?1_lQH82kis%XRUhD6AR9cDEKz87nKhfhlr48FqR3@?mS at eS9MvFy^>n
zhqE?8M at hKJ*RB6Cw3_a|-j!phsi*g9aBxtkU?y%(iJO}n at ENaE#5w9X1$CiU$6iZM
z4+~*BIuxx(r}FYH;n9M%wKaiYYikQ44(W%W&AgMF+nxf;soLj6bh5Djda{a&?m32C
zXn_Fm3MmWW;o&&x1uxA6Qaq4QaZlmyiWQFg16#tdk(HIzwjv!ZN`Ln30a$u29v)kU
z)w8Uu&q1j}ICJK&e<>`o0LUp+v@($doWBy_UK44Y_?FBN$;K8$OH0e;#V%w06n&}u
zMz at i~35GLgc4M{_9>!$6IhL8S>+KoI*{{qI8qG1c at VKz50BhJ_EpYtldk)nV#Wct9
z<}>qSO*n*sg_SiSdqXonu_;<?XtX{AtAHXSBirm}RctQQFQ_E?fJ(&#7G??Bg32Nm
zF0)MCO_A4cId{b%|Im!@XlPQ}%;mbEtq2!?i<^t<jZRl at vAqB$u6%u;V|@g+t%61x
z`I)1ft#_s|hlYyk6)@V^az|}BwxjZCYNWs?6crUqOG^n;o!R9;@7%%2tOaOj!jxQp
zH-B1NvqL?Zo|Pp60t*j-2&F)Oab`_mNJumgcg$HSO%jXSc!7}=%~esu?_#YzmR$m%
zwT>UhI;8%74XpO7sUZmyu&R2od%s?Nu-f0_6E?q(9zBYu0yc}k_d=Kt3=WRcX$(g>
z&Bo52JuMt7GZ5_~bOYiHC<ziEv`9WI_-Uhbt^+QQlY`+O1 at b8yC#4BiiKTTo`(t3>
zIlwSBnBZjEXU{I-KtozuZ}eGVH?qfD-Yw0Dm)_vn*I=dRSXVA5&LU-IX2wgFytG9%
z4R<MlSw94Ulb)OF_Wbz{0|Nsb7y_(ZM60l1xLePXD`6u(HPyPrZTx2 at GbM)#2?129
zSjvJ=u)1Bhini*X*VW!M&Y+xr|Nfmtsrx=40OB4k&T(H~>Z2Q9n!WI|qvI6NUQp*%
zRbJFM+qb)01*e#Oj}*q4jY>HN9(b;V1K?(7o$~W9+bm4DBAr7gw3%l)&%@&{<R_ZK
zT4ZH4)9VFcfQXn_PnC|E+7p%lX*~w4EM3{?e+Kr)b-!bZB4yIq*;!gn&IRyNFvA|4
zG9W%azI1;4M?{DIAz#ZIBzAwkGGo#Ux$(&>&))o}xLY*V=c?((dJ30z<t9jmXQk$|
z_BXDzKycX$hJ=mh5bJ@$Q{}(W=?VA?*1h at _@DWHd`SIf<u(mkO76`r1Jk>!n(9VQX
z{j|^W`e9kgLOc(wCR+9qrUkx_mYd$6Y)cU~H9Zgf^Y3^2IhBH|;gUX_5GJsY4H&>|
zukB2$l6rt$?)+v{6hqPaVpn=b#`EH0hm?g*Idugy0pLw3t4YOcniFr`()6!d8wQ)D
z7We*2KDCv}*Jfj5tAGGh*_o|h$(~lkz;8ZwihyJbDJ#x-p6jvd^8lF^DWBv|wv>Pa
zSBacfg--Kss at wmJmyXsGH2&2P7Iim5DEqx}cb at d5{%cyng0>u>$%7CLm%Og0$a^Y)
zkSNXoHF9V8*RQ^%e- at V)KYcn6xcaQm5Qx|*cnZBKRM1MfefefOf}0Nkhd_se{YIJ}
z7ZJIm^z>;4u%E!?>Z=>{c-{!oj)~w7$`S{lc=GDEx&QsuxZhtQc3V-;`^Cn^dAjaD
zwvU~i9e)?}J|obeSaZc2?w$Y6`%9bwP(~WDn*fiYsNWTg*H$Apw?>Pm%_%OnzG^r)
zG_;#A(BFTG)3$s4iWJ9R=|8WFi;H(UDQ4^4PZ?4yv`itKT3MW0EgdqfsjaKKdHeQj
zEL5p<qjO++dEf*?XO1EJCI@(8<%jFe6(H>^t@`Yl8zQc9*4EbgkriIoWt&ko1WkLE
z;2tgJqo-P8HKT^0qWSP4LrE%F>~UT_KRk050m2sMU}a%{rM7S2slT6}``FkMiX%r(
z6Wouo%RlLv28Y|}6v%I(_j!rcg&bVq=sBe<a-c5u3~6E_O92rPoK>$+Fun$Q0&L=c
zbwFd!K`;^#aK?AaBpvhJmGo}8n>SsLpMO-d%j|;0{D_ at mYU1e3R9u9^p=eRBL&U_y
ze;jV#xl>u;akvupcV at NFE>s2JFNKKU8x-_l(t_vwd7 at bzI{)fy>%ZDNt{tGgqByxy
zFr2vYuhDUzq@<*8SXhdMVB?voXB at 5Su&tmjd}in8k5pDxD&;2B&HQrs^`w5`pv}75
z)Z}&UC${TiU(4IGMX!a71||ixEQLPasvDv3UH<#>Pi~*X;hB=n;@PF80At$Zl$6RE
zU&hL-oB!@o96H2cYG&q7w&3x`dnU6{=rBaQsKi8x?n0}(zq at kp9yoDcq*tno^mSIU
zabGc3q(#NVh!idL+U|4$5%j&F2&9IRI_7`}0S}^zic~`REhNl;cSsY+_fwwDGd=);
zC7W;xI1x^)xv8cW4%k5s&Vb1RFp)fy0s9E`04?<#gCN2wz%N#i?l_;~rlw|ON3ne&
z2!=yW+HS16bW&1McrY53oP7Jyqet!mYBskB<Qp3s3u~Sft$k59Juz81?=mtx><&of
zv3oxy2^}3>zHi0XueVT$W8v>$tv~yAdRk;38w(4$y4{?siMjcam6er2;N6t)V6qoJ
z?y<1s0#3;+75XNZM#wG?ODSsV>JGE{OW-%B6|YD%9F&-NXYAL)Hxq{~l7v&YZ{MEy
z98vxCv5-Xg9Kf+ at TpV_%@q!796k1LZ5$3v#O9W at 1QAZ~ypQWIpV#9+aOi|*&pQeGq
zSptc at z5T%SGzXJZC<~UkWjJm+_+o`h&8sa3=<ey+?c}y$zkmP!N(fCXK!Dli=jVs>
z&YnGsQ!7No#ogPUH&Mw{rv<N0bYu+e^5m11ks%@gN-b6Q>R9k-e<O#0+BRk)A#ztu
zMJ05sGdt-?U-4)6SFdFL%kD(ENk~ai5-R at v{gs{S=yyd|)l3%b_N|!!Ze&eE1Lerb
z2m>3NMA2;7{u?BZS4 at S@D#Xep;FkxUfBcGzk1t>;AuWv)pUz=zPE1-+ at mXJc+TB!%
ziGZkcxw6vI^aK)N;c_?H=!<i6b7jCo$HvCo)6=<9x?0XC+?SI(4dD-Sj9IC_^5?H!
z2?$r|l_jA^#-^#S&yJ)yuczF!sCk8qhTpx<aZ?q at d;7nCoozkFt6~o}Sge-|M%S}-
zrSexM!YeM<ZRg<_&LN0HB&-PL)vrJhKS at Kgd-m6uerjqe#>+tvV26G+MjZ3^_m^;6
zt%-S51u_|*qk^x`j?oBETfcBm&snuRlj~~l{rwk8M at C0SmDGm-9;xbLz1 at wBjF!A2
z9XB_2I)NfLPkHvExf#|zRx)5!sDz_>dVHLf;2z<ylF%juBoJ^pT+p_yQ|#HZytZes
zgL<FInlcw5QavptFhtn^i4DedOy%nnJR(ceJ+APW-<z5!KnWZ+E?Lh($|!#4&Q>8Q
zgeWKAKb7#YB8}mK0M>wbmQ$TLDI(nZEU&<&vGH*a_`k0(3;=o1GoS_o6!oGygU~y_
zu)rIx0~Tnk5MdV`gOL8p({uNS2Qlk;tJa|^qtT-$J{J`M{(+dX4!Yo(6A}_~FRAt5
zmEMn0pFYV{R9Aa~AbW6Dtw<3orcAwQozQEqU%x&WX4Olv37Cc}e_>x&NFpfsR?!Kk
z<@6J|Nrh&=yfHcyIgU$5{8XKYDyAvX+o16R at ckCVGb3PDbfKf--4mTl4r_Vv?)fvX
z6-DMB;l%qgpgllbUcG*O;C~hOiSIxZ?6?M+=GNqj*w=S$ynL<EVh49zGw3N;D1OS?
zv_&u=7V4jr-P~B;isj%8v at 4wp3t9X0In|5Lg5AMOKARg{HBx9-X?5KAOQ-qh(dk2n
z4(X)~{b!1s6aMT9`@mD5{_e<dGg?hQnU=ND|Hav7ZEep&VcO*66XpNM0&Tv&<JYm+
z;AGi)L>u>S-)>{k3r>oJ1u<Ft{QzsQDk_NVSLUz7Q;mqZk<s{(Ej<Sfg^8h<xH$c@
zbQcYdV=^(zsE3LGf^-)l-l6N%IoYjB;H1B%hG{rs>+s>j*4DRf?fv%cn}&hG>*i*e
zltw-t9^qTJJc<tL<}aI(vk{ztmVz5(p*<QseE0(B1txRhsk90KtdQEnvz$3|hH4e`
zx(fvHs>M$42=Tqhi|y>Lz*FI|e!MgbCucPd?fWx)3)4pY{8SK1U44BiR_4cd0qBBE
z$H_e at O4%}7CBF=MA~%GE=Kx^`VXaI|P0PCTx4<v!Nz-7^?~>GiiC{q|K=;?~%C+?i
z?J8eX?JJ5vfWUM;@GU1$^)^$#o|*>v1nYEX>Kw$e*&Q7n{jj50i3TbhPZY6 at 7Q*|4
zy-)UKK<sjyGY{NWGJs<MXgi+#NJ~?gTUcP#E3KdnszzE`T8OMJnVFeuR=U4~Yi$*%
z1qB7;TlGLZUsq6AdS1=^wYIhlR9*%S@`muH)>~kq4=)fWl{R-NT&|%cKAr>uM>&Kb
zEXTvSv6Jm-Y1v^)DKEprY2oXn at 84hOSY<zV&K30O%F>LFVOLHovpGY7&!0c0SahLS
z;vm>-zod#4J$S{i+s!QoEyuif at 7}TQe4d7I!Tp?^X-bom!OZr4-F^uPtP+xvfUP8E
zW at af}Y}?j-(XjIrJG<mYNp3D5)16?Ky@#1<{$<5gwzX0I{P`2RpF=JhUTAx)tzFj8
zFzx$StEs68=Xm<Qd*_U^(TA$Mmg*DQcoEQtRi?aY1TM7AsjSe2OD;6p?kpWBcf(=Y
zcJMa>1+1opa~(VS?kcIPs2q@${UK(v;k@##y?aCY{Z$Ft;#6x#!(ipqja%WFXPcR}
ztpB!NHkCrE$`@eLBh at 2y0k9~$1q1}<C)*F=P)T^^qN1X$($E?z(3>~`j7O)TAE(IL
zEpz?(`xh&+e)oQ+E}e6lwfJ8xz>mTQ$#<(oS%dy?H7|B9Zl-HFc0anUYMkfc$7L}g
z2TbPR>e7q?rY?wCd3X-vy$k;~e-{4Di=Vh?(0^>TINqJ*#ls4r@=`*o;A78+KEeJ<
zALT!4eh at mR{eZ(z{$ozy<5#hP at Ab8F0wojDsA|t25MVGOF=Bb~`x{rul8;&gwa31F
zC*M!5{`;mg|1Icpd+CmWMXQL_S<U~hEVzIFzRA<4=-*pY0A~8=U?nIiDNU>K{3gm0
z6Y`kR!rEN$O3VX*9|7gNn at O7+xxnV==!iqz9XD6$KpPQTSy?G6Dcy+s at uj<at)M-x
z$3`pVWBlrR*B&}Mr^{xe{lKlO!7XOSE}d(NIp^Xe5sPE at DbFerV7W1$Vq#?_CM88i
z{dq!@>$u583+riSW=Ya+5MOv6bAWgkaPioeQ=OxRt~YjwiHYS$^C{W>>rC+p3!_1^
zla}5gDOpE7t)QgzYN{)DsBosZ9B6K2`{WM7e{<ikn#YaSCceS-(PmIdq|y8A1pN(c
zH}jzt{GK4o47jJZ^z7!%-8kP)N5>5)c+cWfJdws*aDavub)1yu!O^am8o<%f05V!t
zuYCm$ZP%bh`}z5mf;<L86nXY+*}{h}OR6)lQ_k#STklt;2Ibboc^kpS=^oFS;#D%d
zT1i)zm4M0RgF?%v at ad$ZMSEridv2xZ^_M<ho=huA*Kg+l!W+}zfQK(Y;kZHYgPcvj
zp0-}cS<>&p9DdE(`tnDquvfqUCqDk8l1=I?wilqa{muO(TZzSALRX+McGKrqb6C5e
ztgP(QM!}{2jB46)pv~0tp)W7$@-067tj4*1RZ2Fk(Aa>cnBHP@#XLsI{#``o!omJ6
zFNx}sR#IYHoay7l8-VA=qJ6M6X2-R_z(7o|VM#Tf_4&Si%P?>Ax%=tlCF?&igj2AJ
zv4(_-5=X;$V`Z(}cB8IreTA?oSezpSm$bAQ(g-@>`ULlumKNgAY1AVtagHTcu`<9&
z>O=U{b7w&85D_9}5~$2yzdRWce+IA48H8tlCqPUgo%Er7(p5eQ`)6VUxinv(7 at 6l$
z*3kF>#PJY_ at 59KH1@Hiy!h)cXrgqem^4y{6>1l{;B3D+%qVevFdds4hZAlQFh_E_R
zT4xpHfN|0&lMA`1sOV0o_=JRfqs@=0tGd}rn#RVQ1osr>+}^pTMse?dcJN1&- at yA?
z2;^b>7KeHYt&R~YenmL;en_6*M3L9gp@=gSFreU~lYq&No3#M1=6+Z~GB+Ax{OV4~
zMGviN?^av29Hyxe{am3f^-)t_pRzPZBVg&y9APJ;sH`kGgTe3sosfz_NBVUfAUD6D
zAMxY6vaniG0z__60-l5L*<U$zsz1jkL%W^|34+7wRBo38DMLZV!esk?TU%SyRRCYC
zR{wG8uF=6khWF=`<!7`=Q{G+tcoFaw1VN8gUk}&2llqs9?<cLqK`_qHEGOo+TQ(X0
z+}_rP_ljh|GgFV|g(<bdmlF}V9oCj7)vVT*=Rm~5-kvktKW9~6)6ybGAcr+A)8Oj<
zbJ;85{)q#`u$L8q+*%U`zw+g-hD&<EUwVq!il-1DJ%JYzU!BojFrqhF|CgPeeUu=g
zrluA)wTJ9v{8B!j&F3nEk{|qb_Gq4dO7>=<nZJ+xmkQBo3_q%zYjhMmAaS|Taf8vo
zz<`JV9?U#lI3t7?P|XebTz$2+wnkZJ4twElG1=w;lye^#KBzjpx{8fJ!p_dFTQa)x
zIKwg|E-vnp&3t@(YAU6e_eo!>7&Y6eY-%CLg^AXkP5 at NL02z+v=jQ|7d4N^`+Qm&l
zLvs#-sAKsrzTwZDDtY$^<f;Yc6d`=Zl(Dg~QCV4X2Ef+MCt5%sf+ZG=*p{?viJa42
z{{NV|4tT8F at BNe_2^mp at P%2wSLX at 41jF4=i%(9bEW{Oft*%{fZjF6C!GQuNeM97Sg
zRl@%q^#1;Re17A7#q+(tXI$60&$-UjxBm3hxcgw!;k+RsA=u?xgXIQ4YUx93;18=9
zZ>p<nB!9l*od5BKGsmcm%UuVbe5}$%BKGsEt1m`+D=G-MY^w%%byr%z0m_z-o at XJd
z!Rzr^@j``KO=DwzWU-=DM~l$Fo0^(7`=UCwysN8g>_E?@p4|DD0Q6|_5c3KO=m~ED
z6^--eyzCk1R_A+>@2be|4-P>C0Ib#4b)plmCwgaoDNs5Cyg*T++Y7xuCU;mQbr=8q
zQK7g3-mlR6`g#p9UArb^7tdRuY6nzQR0hL%FA12G%{>91CG1=I`>Q$nwL2##C&snu
zc0H3 at A0h>MXDnLwIusAdSiakte|^afRp%j&D`;wle=fgP3^0GQ?`{c~i{;Hr at O!rs
zrW=HtyLFATb$PIm09NTR{#)L~0x%<%KHt21r;Jy)9oK#KuXn-PA`j%!h%@|W`g4$-
zPTL;U%{AuBF)ZN#Q<S%}%fbTAPT9D^nBt7HJ?{V<)aaR9KR`;!v|sdw*j#4|L>NRm
zn at -@Wro_CwGmt+*kl_=63cNZ=Le9Wvu>bo~@9IiN- at wr#-9mtBK&lN>^B)ekW=|WI
z76Jq$*U->tDRwl`)X^cP4S&u0%J()~rh<fw3SPkHe2ba at 3P$-07g9c#x#x~Rka>|^
z=BUc@`YC#WKX^gxe_miF5vvXnKi|A&JIZQ*8y-G*Cp2^`;oHy<m*QPkzDK<0I98fh
zf6ElVN+eEXEPqdYeO_Mvt?N?zLN{}{m0VGW12cps*SeoyUZlQs`u<)mUIubo^y?s>
zNuQ*O-XG<9y1F6PCQF6`mmo-7cV(y(QxFPFE1_f-Q^iuo<ZgL+K5Sp*;NnsQoQSDm
z)S6E~KmgxQFi?~bi1fNL2STxYC6SY)bT{P6KrOkju<$nsMc6EUWwbSg>*@r=<e}zx
z6G!_W^#|i$Wk*Htl$Bcg@#1Lq_a610pH{_FQ<txRD~^Bdn%Fz+X7eO8TlkYfu|sS@
z^UbXk4!=thC6Uqh4-6={xD-5=zFzurIPO~V;UBfNwQoLrIE^-n!VgH=I at Ibe9$pK+
z=+cC at _Ya&N!$^7poBMIiL9y3r<J0_~*{>&#Uh|IRH^SxE<#oo-EpHL_y at vO$I4pf_
zJ`3kyJw~1e`5Wgt&i=YJQfqEuQDfxZb08)rW-Qin9XJyi_<)VREGz3+(w7 at g4oz1(
ztEYFHjMby>nJDA2PflMmblZ1YH^tDf0kdLJBYz#PA^m$WknJx1SLrhI`OJ*+UG(&Z
z3kIRtYvELfyM|6f>Pk=kResSO%_-x%cJ0C$i|ra3&0T*gUYu_)bDaC#^XjT|d9%--
z(NxDjLVt>D9UL#RtXSP$eKBbgA7-3c at xb{}?B=!4mwf^q?1lAR{XEkfyM at gCT_X4a
zG9Yp1;N-*#DJshd&wY2RsjUqGL`XqRO>Jgl!wumUOD&5(UNn#K5-;VLFxVFFVr0Ax
zr>>3>4=7Qv66;m at 7^$nzPv(|kq11Es;<vFeZb-fUIE1~hzzDelD``i)yybeYfyL{;
zy!`YGcLWZR>YqDD4!8ioN76Wh`i9jpi6{fjd(Yw*oL0ZQG at J%uaXSI8`F89UJkbzy
zOxv!jNX_t-hvwv3U0FWV4VW$Ff|e3zYcl{|&s3jqXlZFpj)*cF<hG8_p$rtm at oV$+
zR#sM66*ph``#UisW30&5K|w)*<H!*$oPT=!<VkyUO1^7FXLNpkK9=dYBi8CpxDVmz
zb)V4VO%tDvFl$?9xemUg*>vLbmoM#Img<eorh>sSF?*-0sbzwJrJ|lkh;g~+sd$Tn
z2M<nHl28Xf5w;G&!~~?$Dn6mp__f=(wxJ;}hP0_~)y%&&FYQswpBjh%-T%cpepk8Y
z+<oa!+XH0r%C8>?$x*3!_;=Wwk(64>xN8 at AfGj)xc7>#a^*dWaM_nYaVF=$AtM{TA
z&X<>Pu8%2;KDl#e$EFh(MWV)8Z7;?a2o)Y?UvbDWdgYO?BU6~4PqWbDZW2pjUN~#e
z^vO7rw*M!2JH{sf9dBxD6?g0vKLbad{q=hE^HbAqHfJ#tgQtPXBsGC$7E;-d8q&%J
zXJl}%<#~Wuw+HI(9vv$wbU6Fd;oY!1xH7iTz{!#yJ$e)q*3-7pDEpr4*@47w%mxd&
z8n!al-|bpow;!{eT{VtbD}Nh20sK9HJEc5l5B2x=C$wC|8C^0mGXH>pr;fd^TeHEV
zo$BoAA-NDvIIl`*>FSc6KYt!*N+SF at cUOVfIXBV82v5g`u(l*=Cryu7x$)hC0+CVc
z<c(H*!QDHMm{fQ#W7jd#!VQ|Fw1$jrXup5|jx7jS8^wkyYN;s#?#bN2EQX at kPIa~&
z1{sfjYoHv%(Oxe<c**AV_Yd&7LG({d<bZ$;U%+%Y*B2_F_<u>svCW$|CupT7=y22Q
z-n}hCWh>Ghi(<SN*DrJ9<2NeZ!pzKEGElJn!i5VXorOGjuYxmCl-|atPLZ>Ej<637
z52Ks_n6q8&LwOb at H#adZ^%xxGL8$=jL2o8szZ#xUfQQ>jOWT at eAb;(xLHnFTvF-)I
z-6&!Sp^3nwn{~0)KbsE1OV4D9j<z<c2b&Hz#o|IeFfZX}KVEt?8RxPO^G2_CDs^s)
zRVRa7EV{p at riLA8B;jMKM!bYt*U6J7vr7-BA7BvEJ$v>WPM`pT*#Sf{IVDB3D?hk%
zvy6<4)bGIuk7aybcWVX*Gw})tPy;!Oh<Yx{MEU;xdt6s?@7~Vz!)5S-fxd=I@~`o(
z7j&|w9XUpevuzr<AGG`WLeE;4D&(EZPvwO#jdHDtF!-#li2JNawYIfEcI at D&OX*5&
z8$}`FhY#uD9mr+2qu6UeKmdUNk?j!u_B0)C_`;d!>T18x>#O&pqbVSt;j#}%{tqS9
zGqF1r>x)PVQO;^M2HvXF->GUOtvDyf#ujg-eu}sP at fuh4qVNdd&^+&DPmgQYKCs+g
zpQb#wHUlVRL8=}r5#Z2MTF;^+mUY(XC%99gjZZG57`vddv;+7Ono5#8#_^)*$LD;K
zioY`yq|~h44TmN9<6d;wzB~4TYpt}LD3+C%m#5w~-;$)XQIXlQb?Yg*;_5eVE`!N`
zV9{%(rC?qyskE=Ux%IKIwfYN|+i84|#}1;P at vGU^(@sk492t7&&h4S4rA>yn{8Ai@
z3e{bKrnLJ1fJF1Aql1x)i)+Id?w?guB*igbARN`*-PpACSj?WOyQk-S*r%KV%Lnr^
zwo%kFe+9jl=Q^xpxJ2V$y^>15bRJF}jQ`lNW8iCf#l?|a_B;_Pc<fuYR6_8d{QNvh
z*qREEq6RkevZS0H>Bwbc^_n^fIXvOylF4hYJ#s-*H7TbfQGrG*oL2M(E^(O|tj7)y
zY!?Q{ql2LCRJYzCX(QI&-i~XvAbL#y9*+OOqNJO3w!QtcLAy-EqepKV8!4TK8mhoR
zJ9ms49F05eDmrh7t7~6g89+Y%7eK4sM~6*P*5fw+J8(BHNy+mzpQRxj6&UcjNU&hH
zOkSQK_T2B^kGp6tySY6&U~!t)PDD%$=ajHr7tZ4z1y@?N^X1Ex*v$)B^#e;l at HSGv
zO?&q2!3J5T=WVU6XTa}Rk9#UJMvcP56KzC+84QN0uAvGWBRGLf-Rw7vCSEtNsMw;!
zwkF20vb^+xCEMuYD3?8?@owkl<5<!2oE>TTQJ4Jatbswu`noaY#x5MB5)cSVPe1JC
z<&~U{u6ib22d`qf^V2DD;SG+|0 at e8oNO@{%>YgO6X0(;yZfv>6R?x3szhdXzW)hO`
zVH;qF<5P~&?Zev7_DM^#LW-OeC#vJ_OKfJoMR(8?5|R5}8&2H2h!wQn3N~z0LU8TA
zKhZOPy(1_sHZv>BUoq*Vq2XR!R?WtC7jh`!M6Pi;cJhDxcm_{bT>P=x8TpBa;o&qW
zyCh^$`wos!BH`C?GpDq)anyYl_<Q&7(*l>4%hu)O<yDQkKLM!N-EiGgPNlY)eR#x&
z(E}Gpz&k_pN8J?xQ)OaEJCTTU-?)uKMddzI at vmRMHWB#w`Qd<EnTn4eKaLk=&Q3Ob
zI5Gt8mgN5PE8hF|0ltA}5#tjUro*9ez at tL$lZFj($D?IVuh)x$IQ%Qi7rx>K?!R*#
zCU{RJNn1z)%o0~{1X42N<V4B98D<<pP-44I50C?=lmygOvza!+twaJ~p0{;%r`5Wq
zr>Ak&N>x?0Z(x8^)FCJ^kV8=L9tCx9=ga<nb6m~?fg5|4C8=l%o%M2!QCkAtdkI`|
z_ntkcWUPtm^}2`XUrDS7R$;7iuE|?04&24|N0i8ems(k0e`@lgJn_WVty at 3Z6yp9H
zc%fkT&!0a}=omM%wWUbWA}=)H6TM8fWy|F%JF)_sc5YnjbysR;ZQkF)f)fuXGcj>9
z1p{v-;O>Mohj3)^P=rc}#X2&n?_ZoA0JaBDiQh07CO3h)@|S at Qu7!dCgAzD7)Z`#1
zC&y7?lzt2kb_h}H{3llf*qm6v!pIo(_N_e1l|$;rJ}yA3#8(p&6O;fBN;v5>Ng$0^
zf#2h(edP@>VmbK<Wk{tWOq94z<niNRfB!82%m at Tr49mn60$FIZUx12=>d3*h;5Vj^
zgGCoUJo+}CTcLR2!T~J6sm4m|!iy&TUQQc7ZXO=&=n7RN3034&>){7?d-7xt{q`DH
z?P7a-`|n|DjVE$DEFfMjLVQ6>LH(1zuhm`1+$K74{2RN=ewHt=V!f4)j_$uxvYF^C
z^qBu*jbi+Fb_>aZt5Kmt`H at j8f@%Jl8$SrHmV)Dy=5M#bQ<yf#9fwo6FG<_Nx8Z5Q
zW03Q4cN{!t`c;U at iKV4X@OW_gC{;ytX at 6m@0VmZ$iSv8Cy=;dMA3k%4-l>i~Q7q&6
zb6SJ!q9*?zSy{ZhcaXY_f67*ub5mB{784u$eLYIauWCRh5wq{~2OfwUsHLT)@wEy!
zOZnjh5zF1+N!&mN_H*yy!!6mm6xq7i<BeBApFi)8i;Dwy+eSsDc=@s*s!X^|aJ$=;
zl3?i`1qT+V%y2rzqAn}~Ba<t at Tl8V<sYQ{SX6dCJ5cuv%GOYVu<}@6!1$<20N0XYG
z+I=OB1^Y^Y*M_P^i`9c=J%vDnAPxdCu5pfxj0DSK@>l!UMFc5CSzY~6RMEh|K*~@S
zpxV<9ifkQYC0y at I9--NUs1y!92w+ELPmj^0xIB?`WX=<i&MC?-;HNl!q at L=Rlw=WW
z^#kJA%aW3MmM3d(f=M~E9Xs9erd89(C=^19#TYRimH4|`g44+1S7fNS73pRwcza6|
zaD!!4RaI+&J>Wz at El=pQ<?!X>Zs=RL$nmy&3mMr5mRo6RM|SU61j>s%z-6>`ZA&zS
z7%q3-^@5;oD)=JW%*m6RsP>9inn}CV;oK)SM_M<>1>C-U8_sFhF-WS4VSCBpcVQyj
zyHnPF`$i}l2}2jVb?X)!tI~SN#_L1~ELCl7FVqLz&Yanqm6ZjFi-_xT$#6K@%#2+i
z?u2}Y(C!`Y-oMWfW|us6Y!49Gyuw1<@g^%Pn;dATp`igkq at HEnR42~>dqXX+k&%ly
zm)P7)t(~fleJ$r?_SvN3${H{s^s=u__4PNZOiw`68yB(9T$7=w0;`(*yd?pQ_UEf#
zR15;fcQlg}(QN~hl9IXsh9<*zAfv;@jSWwP at w`@653E^EiyBS_E_}Tvuu;f|r2LNM
zR_B^hT4|~Ir>t|2k6$Z-bBaKuj`x8(_8i&L>$7?sNb+EqJn at j-`pr`!mpg>9OPy|T
zP7r}nP3dt|l<d<p`$R-^Y`)GfEDUo9Bw7{OcZo@@EDn4epZn{{Io4-mV}o9roV<TO
z7ncgfi_<D9re<b0Dl01`=$>Eq@>;y^x0X0ba0if!iwGW5R1_2L5QC48Q-5efPa+;Z
zRHiGoO?9xgv^<g8R at K?bv^`)mWc}8`3143VaHRJvWpcmaPy{c$r$_BAAT987*YduO
zme$RyzlOfyB#rMTlBsq*ld9fclS%z`LSg>~FF09Q!vG1R7#Ls!C#6I&J8(qQd;<>`
zKa}X#Ha>@w5QrC1z7jyFpqSVrNe0cNe-X(5SBpaQO^F5oaP{gjIKZ9s^!r8agMi=x
zKX<A-V55g1141JzDniaTxTl^f=jkaS!W8OqT3uZc{#HX{<I^*Ta1N at k&(uFKP&QrX
zO+$m~<ma`|x<3g>;<u%r6?N|;c%HV~Jk`MqgB!RYrK;-Wq=!6F77z-=Z7xxJ+d=~X
zO9=%9^v|9>oBH)DQ`l8fTzvQZ{Jc}0gF9+Yz at eb34WiMIT|9l0ll<<yiDGr8f+%a~
zJlqtZo{IOeQ$JobJJE=*hayg}b+f^D;kX01Hk28+IhAl~GDpP3?1)eaSYBEf_<HSU
zgd|mcazX-Opf*$vB06?`BTsTt;wZJ>uqTdl7+z9SPq$%2!c0?cj3PPk$STDEX;2J<
z<w;tjL6O}aoPddu at iCq}Bu`Yuyc88>kjgFzC>)|%UQtn?R{BGx=YBVDLQ4L8Z*AC8
z{^%}9q7b{y+n%Wrd=JKXYzCiG(AR(5P#>&J1yQo at E?bad(hUepT3WRC at 87@td*%W-
zExswOfg*!!MeycQ*KscQA7ZC+JMiWVPb4Mf0D9o`Kpk+QG_}qO9~o?_LFEGuj?L(x
zF)=ZN|1hQ1Sv at _}C&ST9+uZU%_peuj#x+KY;31KV^m;F4cWWw>;wm!83Z at PYVK_qo
zW{oGs`Oouolh&6Xw-#W}_F7iQj?{H^CY)#lxE>Z32ASQ48IKBeDFNH#X>CzCeHusV
zAwpafVcLJ-0B%$DEJuZyZ98^6WRN2xBSTFMun?!Z_1`9jyVJ%ya+SavwY0Q=*W(p1
z-etwaoR)vr*chOj311Oo^Z@$^1U)^yw04`(H&-C2Uv_nkICf<mGV>H<rf+~0aZU78
z|7`)RO+kc=kYv(~cJ6;mj>uYOWswtj>NFtq>gHyRf9^P;2)`K56!@k^PpRwQ*xO7K
z8}biW1`-z+f5#GJ5L{(iiFdUq1j(s*r*f=4vxFh0bF0|e>@lQFx4E&ims(LR48qmD
z_;_l7PTj5 at Mu36+{QU5;#oaPd*LwcP7nCA{?->k}l at HZ1G`s^|1HtLVix;@d1qg2Y
zldx>vOk8S;D#jM;KVNbeGxKe<uU|urxYi2Z-%jvEcqt`GG at 3k45Fg>hM>jh&g_0gr
zk0B6VzI>@Xcc-)Tby-<i!>#XdNO>1NFkl}o&NJY~WQg0dqisrnQ|;9uW1y0mhV<_(
zI8z2_mWGv;|6l5JI}89H7(K4i!*AOoY*qR0o${nUQ4y!fmF}`)13U*OCjdjfeF6g1
z54hCcRQZwRTh`O6#!DMbD$A<{{(}ZXF*4kz)R+JD0{9r)e9kF1Ds>^Rt*xD!o(|`-
z5Aydvk)j1b*I at FUJl_x;wV<eIwsX}@h-8*lR#-_q!o`_*zP`CRl1t-Kp#d1$x54 at c
zJ7p`sd?{%nrb8(6 at bGxI0T?9GAkum*O%$WN3q(Jlhj<^1%NhH5Xx8?9H7Sf>PyK at j
zBz{%CIYwol(yykqk5OqZ?36X_Dzf*0zl;s{sd1j$;r#s&?X(1U?uBoD{CGDKJO~UP
z3q9`erf`%EUdh^@LR+wX3P%37Wj|k26dHu$ROa%Zxd)OAg*3Pyu(hpid`EnJQ&TMt
zlY9$en_2kdW?3CbiG=}FL!BG%07{1F<kY{g-x<)|ZJ2H*DI`P#I7#b8MtpocKzLeO
zlofj%9gR`!9JbIu7Bt&@=gu7nvtfX~2Hpfv;M9kX4m~Q7jE!w)WMrKE@#O(U>Mcv4
zB;gSeTLDrMi8^qYCjs9=I3 at Ib`?d*SQ;ODMK2<k#HqrH<yiCO+;ScD-?#uJ7R#sN1
zH4A2f*~eXk!{$32AZfPldF!!xR0jNK)I}_N_wLZpXm%3{O$Ts+IUku5#AIBniPD9A
z<KsN;B8g&`R#%oizDMZYi;4;f4&L_aSCiyrC#U=W&}mR&yUjEJwE<uv9y9d^(aM0d
z%Yiw*XOZ&QNi8+yTi|W;d*(aV;Cw3t{*3hX?Q at nb)eZ;<z>`IoeW~r4Sxcfq-t*`1
zb2T4WtdHl>6rKm?NfWj|pVZRVS!j!UFSOF-FJ4R&ah2S+&sVP}$&LN!QO0u5xr`T$
zpFitMRLprE-$6=3=mRWJ3pRLOic?cbYpr(hP@*j#YZK?r;T_K6 at 82g=wAxFKNl3ik
z0BUJ1+bJlfCyIL!&H}oV)z{w*=P2mVEspC60P}YSKYxA%%b!9*nl{}{4GsRrxk2US
z*JE#WLHwJVpAQ7a+nvw6Uq*)QL!<!4P!$yw`~j%(hXh24)8+w~8S*7;()jl68#*`M
ze*JT9s);%_pSM5{kT+r<ltB3`?HU?W$kuUl6T_ZYY|%Qz&3!XJUkDHu;cq!rYh$C@
zkiScr|4uBLST)6vJ3Big$Dlqwdk;`z6y*l!O{*3J=-Sluh|B)q{{1Jwva+&vJbLs9
z%L7~vON)z at Yk*Te^iz@&HDukDXG at qC;k{^ozZG at H+xs=~PIkKuW)+|uK%ao7tEhdx
zapT4r?uc#Mw)J)FTYqz>aTz)s0Jh!fpF8($G&R1KLf}>9;e2_Q^W5Cr_wL{40F)xa
z1SI1mV7gDA7`P);Y&vx_+ac^+I?H(BI6!v<T$mKq)zxv{jhL?e?Z*#Zw8~KkmTLl=
z-pihc=(nS!5S-?}qFLf9(t~)f8FCyFv9-0u7l6iQhul>$`^4F^yNVop`PCZ5L`4DU
zlsQxY8^jvty?c{*%n&`Xc^GB7mKS=5bzE!~KCcTo$g>y#^TS;*_wMZzbp+6ol##K$
zsHg}zYE)6XjVShDSz7LQw at ZwU62HPB=Fq*Z+}Pyj02db*&a4?<pKE7OS0qXU1llX(
z{i?~u4qW}_&CLYgva%)Dw*2E#QkLrLzl%5n{2<S&kqNunQq^PMTwmxxT~e$ftknoP
z6UDNC`vSk!YI8F;=g`vD#-=cM;zJQC>6Z}lt*or@{kc92ar^W03ly(c{@CJXPhQ>~
z0<RZ2a82y(d77SQ5EuL4NzrDzdwQk;#+m*2oWF*0nbyN+dlCd8)LO-{VK6VGtc$&?
zE4b{fBq>}~Jw_Zgo(Vp=W?8XOMr@&qehCQ)k3tmH65xPveH6Y66B84Ze*oG5DWC7=
zMjDb|O>7&z_vlgJ;-W|Tl`=Q(AU{76$PDl$D<?-oOZ%o)(#BOc+aTcGyLZeCLiYCd
zS0=lUeg5*L at bZzwqEz@~a0IRnxB)LK%YR)dV@$+?jFdDHklzNn;jlRi!NKk4*U;}D
zA_t0lz5TGRsS$Sv!rB_81s5leRS%k-bxFSjF=roOOjJS4%k!0edM~O-4nE4xogb4!
zyF^7raoXA3+&s^sjtcjXLCAY7;mQpxy2G-Lee7aLNJ!qr77nbUQ&CYpKI+JdiUR*H
z8)=fzGccHswx;0X*gK)2*m@(cpz!qC!X$)=J+J0IoyQ#JSkeorTj=7)1O`ak-|ot+
zoPeZ(^0e-~*BOPa8cZhwCFnPB4~^)F3l|OmZrjPmCZN{S^zI#2DN0>NIn+9H?7Jtw
zyh!a1Pbc}%)D&1;>|TBs2P1(5dO$omFTP82{Mq^i!@8K4Y-T%$sOYZY=6EuK%S7j{
zbsxvCL-gqBgPiNGrYPY&v;)PbJ?Cub>FG at ur@o=D%26MGd+M<~aS*sCu&Cr`_MwU-
zVCZ;5s;|FKQ1DE~1aMpk#G;<lSYU+@;up$%w7<XR_RH0kB_XpnTaZfvaI}Bsu1H*o
zh>WyA5$TbnyV#!FVdBHmmZ}jH5fRbdvOa7)B8I|}znW$0B2Bz;f2cXW6(I{!0f&xF
z3W|z2!gVw at c^Q&JKcshuDjgTenP0!AYndHQ0SMqwG1zwP?O$3!Z`bSO9En{zy1I})
zZ~OaixisbmkpV0Si2>i(b}FhsgY3Y}%uGB}+z<l!BNrvJ!vGVwjDJ!K)RE7w#(nmX
zIR^(VQR^BG6b}94s0;2Dh>vHZ- at X;TbhvlHe*zL14BPPgoDmlZyRITi)MH#(Uc!AX
zV5BH60ih7x-Yx6dGew*3oa_l;dOA8fl|YIifR+l{Dl2dE^76_l3w{gvJ`Wf<;sVqi
znV+v+oF2eyCM@#cnM&*E=?z3XzOKTZIrs0=V&e&FJMP`PcQ8zG-80v`_k59ERcp#=
zEI0$k#Rf>6{Nv^29Sl?AOL+E-6bq%`Sd_8qdOxM{yGL4*&WrC6)zQ?XLCS<!%c#t)
z=J|yWf=+|VB^EUG{~)Nm8VdTL2$;EfqR5q{x3}m3I at Z)EAQBzkHIP<<`oUerj=9c-
zryzApEp$8U=bH!IsSh?#2XlvfioJZIcr}e<$!n|xQUV0PEu^HfPtQ~W#eraX<lsS$
z_2MT4Z=9?|5DL)(8&_y{?W!T?mM5BnvBJ+wDfI^CaR0#rO(P at 9^XJcRAtS>&IqsPQ
zL&c@@amTOq<IM#C|Bn?uJ|(@IHKS8I#!)w%S3i$%_3Bmh0XSN6YisNOTwWN<QD9<$
z+2 at Pjw6xIRbY1eP2S8lmxEr&8=)tp5Lk!QgC5iz&R8`%9qb%AIrQu^^HoPXU_o6_u
z|4y8a6SSz^u?8?hm39BIY8xvB1qCc`LnaVmq683)I!fEv`ygqG+I5IIE{wIGo_t57
z!mxxR;<KLR1pu6O<J4qoYAPIQR2)z%+?E8vxS&7;;(*HS3xKM!va(RIDMjmC>fy4}
zk^D|d>!+FMbBKp$O=-mIw-ZuK$T;AZb;xLe#<)vyjG4Mfupr38Hc|jchxqve!^3I(
zcVf{DFxGmEf2y0=lB&U0!bw+n3gA1A{!Uksu_1i{+YuRRj&Dw%UAGsOcL5ZPd2SO1
zNOgI5C~OLunx&U`<2=U>yG(YUoAe;6@*BSNO-R at W;mGjl9V_rncpDFeSe75`+{}E`
z;cd(@XWWzxK$iq}DC}in0Z>5X&f5U>%A_4RMt+J(;Za4ROq9R{%>;M%4G!v9ce-7^
zEGQ)<H5jI9e at 8b{;mnzPv+lUd7}9a{QO8F at yINu;eO6u(gN?4mLm-`*al&$9bv4QM
zfX$mWZCa0?FPA3EaPH3CW#fj=-$(sS{)-pY09bLS;@9$Py)z|v!Gv?^Uh`k9^NNZb
zR`TSD4!Bk0=ht$aW}RO8S-}aXV{B~f at prmz8gOKFZ!cSYeLVm<YfDS?%2-Kv-0~<Z
zD~lTDkVUz)6Qk2|a%%B%A9o8ta7P0ayf*)Xboso-1_#%(v*XpxH4cSnESuVf^`6Io
zV{l6|1qDUPfF|8(H8llb1=z*xS4F at bDRme@I*hNW4Y_O~;O=#+#;5>@dDsbsFin49
z{~yMGS*nosz<~qn*Hd)B^I~K9_$L1h7uc(RPdG2Md&gxL7x`1CZWX8{wGc0LbJ;Y8
z+`c`oTANcYz{|UZfPvMfHYfJ}#PBc~_~w%b>pxUeyVPTv10e!x9s<8w{OPlL8~1jA
z8Tp5YYnC at Z5xGPQDI0H;v78RCf-fd2xbr^byY$0LeN}$IFDrdFZKI~HLsiG~-qeQ$
z1e741#@0?))P+@<3Zkl<O=nK*%O0TF44&5V+1e0+ at RuswX7~od2NhOvhIb84gs!J0
zgK)@sc0HxLgoK2U1O1qo5V9d2!kU;z-wm?0%SCw`4xg3Z2T_rs+?bkxTa+P7g#ku6
z|J9V`@y^YX?vrnW*t~<~f~aukseZvln$4Rx52D%r<DzN1a6CdcGvb;3K`t&1 at C<+{
zaF`F7ifyBdf0mBtF!QUYgbhRLs|8YsNfh21BLI#FN!k|IjEoFO{+g-mlE;s;tW|da
z*exzCaqio9qV#Ztiukc(^)WFUk91!>wHo}=ALXV|D=#vV5oMRH<F1m<&(F7d%Wgv*
zt*u+P;x=sXcGRK(bd&F(C{OSOZiH+RD=(-|F!=p at RZY!SfbV$nAD_y78yWFXsylS}
zaJbDMynTSwpsr5kzY}tbQtYz{$Ux$fl1E$0IBm*tvjz)GD4hP&w6tw_i7#r)U}$9Y
z{a^aZHuUzE#))`T at p~7}IRbf_)#vXOtTAI{Tvb(dFihEf1RL-n^RWBl$hWcAv;wa|
zf%2q23%vD86qbTRLv>T^2tegpY}=Z=DF5HRW^m_YaCU&MMOP(-h3P!YS;NA at fDvT0
zy#&ba?!3u{tk7fH9ChUgYpb#sdVMMyZl!T%=Hw^>sz5#w7soGZH9R<YB6qDL;q~ja
zc~dS+vMpN at Ap86mado)>MWA-nr(6 at PYpKSF(;yN8H~~SQXyY2Y913{S?z1571~V#k
z2|p}KQ~w>3EDF8PPjm&ztp24-Pfy1R#KnuaN;&Xbb&y<&mgqH>ayUk&P(_>0u)BBb
zz<V|kCMPFRNf*BH&6_t6N5&0(Z>ANtjL%WV>7)7&xSMJH{R5C3v3qv9(>C|%`C^C0
znEUId(wQV|-GovzpI=_#gwKb&th^R4<ED9B|6W*F*iADk#s~ZjyG{IeV%`Kt`W{9!
z19V~mxC}`Xj<TR|L+H`3CrnKfu->!fzi@*0p~|ODrMJC|jf<lK{{h2=l+7tFu5Z&l
zKQnW4jh3h1faDH*{EU{?&5)4oDAWic5+Z%(3ljx}t=v33DHUuo-g}*g8g4{H?85eA
z`{Lr_pj)@dfi>b9Gf7EFU>u?zCwH|p>YP1`dWHbzlr=Pzbafe8TU${&<<d9fc<Jkq
zDos+^yFPrlJ-lzDZ%qY|5b^~`lY at hUeXo}$QM3tf1jfe3B$lF{Jb4OC1rLdonw3|(
zzso{!_ZdS&l!iuWm4X6#iCFc{6~_w~E_9Up5FmL5Jb7|1)*csLzIZ`G`26{EC0GrZ
zouIJrgCe>@QTR at Amyw<L!mlsQvkM8)$Zi8fWIokbg>sXV<qKbRuY{SDIBg{-C&wEY
zFE70IM`E>`nYp^IPRXX5ijp!wCFLfBgYNKA23$G$d*%WzgHKN0;#qFo=^(h9kf472
zD}WgQEa+Sk?voI+lTrN*FoU4&r>y8n_tDk^2syAQ;_51zn3yQ&Jfx}~wqej*xI7Q1
zU{d3~Wg+)8G&WB6`BC?WE3sAu({L_IUSt+=9n-<p at EY;bcuksPT+V%5S4YRcQ|z73
zr!spxJ5vh_ENne>9&XCZ&ksLv?Ys`yCYTf{o6lb%6thIt;JFTyx*qplKCB=F{KKCU
zIuOGF>^*I{m|Qynpa-xwyl3`9hm_#}wr{_A_b&BXnCO|r=9S^5SUF>3)*Qo<o4DN=
zZ1ttnccAIplVu{ore40(H*o2qMIkE-3lda^?(BS1Ur!ExcoK{)MFtLID}eaRuC4)@
znLMc)uBN+o?E+eRVtv(oGWi!E`m(^=8X6k<Mn}m}qj-I1C5lB(Por9P_w;(yNn?S0
ziT6~m7yBtJt*lT}d~VX(+<dcNl`l3DIaVyY-4=Z_I_HT}HH`d*l_{qqQ8Nm+myES%
zx~}|fUV_Ml($f$i($do}g~!b07Zg0?pRop$S5i`P86&RU;x079BisPS3_RsyiL><=
zC+)iX@$q*7s8IMU&oP=?SilJ;w2h)lIDiv22yrqizv5R5o)}me8^;mzuFsBW?&soy
zBicPQG;}6ir?mis`QqZ@)XgkD<<vIxD1f4J4i5aN`i(u45K;~uI&{L!>?!f#^)etP
z8|zvU^>U3-S02#V<qjivlobP>tv2hxq_ORWp2_uHSp4y=5cqtqH-H4A2Nu)87;j`}
z^I0`KzLA!8Xr=7Lx&btz;5fuYVet7I$FI%fZ7V7p<(XCQWM!o)G-#9vX at KB?yKH{^
zu#1tt&IAr2er=xD$=Nv^jsU1;Vsi3e*j)=Vvm3z8 at tzKM>~#0^q;wg%_iUAssi5k8
z7aJQJn)~|CcKjH|xplZw<B6~}>aVO72LBb9h&M2}95|q0)2$Av&%l5wBQp~Pnge7v
z6&g at RsBA4X2=(<Ps1IG-!j8M=?t%Nn#O%e-1;hhC7qHqZ^FPi3FH~MSudh$M<h^1o
zT#%ieEpoE@{rl~ZqFuR6NN~C3*xrhw`akg~Qx1SonOl&89d{cMT)CW0%G at ZyG&eR<
zs+>NZLOOHv=1qyl!Lw)2wvL}acdn|ZhXo~rqNAC+ySq`?*~8Nl<)`2f5puoodX;_w
z0<B8uXAG35Ps at Rg!<(f9)adN&jQ86e9gF~sH{^0(m6zv6H?3cv$rEF+%7H{n_XBDJ
z;H_w5^id>Y?k)}3I at a(+_+4Y80$2-vC+Z;r?8BZ(+^z-RVC4RnS at c-PSNx={t+c at 2
zv4WhGl%%tiUsUvn-**UrhS5aLn>U+3efrdKVR`srxItlC$tGn+idUX<+`wDlsmDcb
z>*?qicAQyhmVqP5ZnBMzj>cs#@Z=|o9eVou`)^M6lqCU<f_ENC^wKxcO=O1jdJ^6o
z6`S;(&!6>Cp(|A*J}^D~@MaPcA=63%00;uMFTf8N4C~tOzY|5*Ao|iUG6pR!dRzoj
zfA(zE*RQO=Zp_Tgwoy>XK?a6X9t`W|NBdxB--KSbu_wKuK^3Cb<HwJ&!=n7!!lS4(
zTiYeT|A7x4&`G$CVGHi)s{YEc<7I9cZuaG`v$A%e>VANc!1^XKG#->Rgp{AMdLBKC
ze*4ykDhD9vBf)i<=q&sq!wP_MV|mIBQj$A&?x2J<1Z{&(rOb9z>B0&lmJi?a<f1Cm
zU+E1>!IuNv7Dz at FYN6rQcCb@y1qM!W)2uqs?A<+fzL_N#C0#8oj`rt3fKv*s3vH>(
zT*q-%6TmcDT24+54)Q{R*#SHRs~UJG>*wcJFyHd^t1*-OY%@j7ryM_CUfwKc9tau^
z4mmp2&CSh0>ZyK!K|L2HDSdo=0Mg23>!QNn#OoX@!QE3`jumU&0q_!0NdjOhs;3Ur
z%mq*G{0HwO#ADTifG1%8GZz<@#sCaLf<=R>sF1no=a*97rIJEhR8-U$m7bg|ZzJyz
ziikQcE(~ap<%J8U)V{!<w5+TxQwy8FbY=>amXthIy32|U+s$#uQw|dwm#=JDiF32N
zaG^ap_Z(!ezkj^|8?(#Em~fj8egA%n;=V0&6*zH9puL8M#^RqpvgtZAn+e%xu3Qnq
zRwM+F77(@3F?P!0zAeZYTEaz;zi@@i=;#;o2X*VHEZAWPuSh5aUVTicZ?m$p%FD~E
z>@e{OJs9WFAGm}YZzLs|QKWx*pe|*Ub=4tCSV#yJyZ~IcKnPAh_PbZ at eN$6YcIl++
z!5yT2M8m~qt~vO~K)RK?&>>O8>?-bzDl01^zU%>Zj}G6}#SDn at 6~UA&GBPp<8 at NvX
z24D|p|AwC*2~KveEH7DnOgK4$dl{m<liEh_#l#q<UO`+<>bHrXpC8gj=aP at f)Po#7
zBco6fGOE72GAor&<$`eC5a6ZRv380ZH*RQI6lkTth49<n&zL6VKDielT5{ktwvnc$
zCIR+?(`l_CkFp5A3ZMl0TkX4wPOa850r+WdrUnoGT<o~kZI2_<@B=+23OfX?8g_pQ
z-N?S}%4%vgKPy(l(4>Q?SZ*5U1{q}6l2A+e;=X{s${Q>hMZ~>Wjp!+JKh3DuuwId!
zEFI_H81Ua7u-V(&8}E)%QogI1<d&9})^{*6?I0Z-8sY#VcInb3T<Gb56VR@%*!?aD
z(L-wXquiN8^lF}7C0IqC8*4{RFU*-JDF0`6)MN2m&@rR9v|Imb8o at c*+S-1Qz1BMc
zPAw=XC;<Ks1cWPMUp5Z_F26pE4MHp=793Ak`%_wsdVa{i)YTQ8oQ?nwz84X(72q{4
zzFS-*UXn#6cL1>f=`9=iicy0Sjpi;}Zrj^9$aO$_A3u45Z+zo^X!~~Aix<;=#QzNI
z1MJai`D1q0#nRFei#SL)SK`8qyL^$?v$3%qqTlY};i0U at abTlL61uX~yNYt;ls2s^
z2f>k1+Yki-;Y^W12 at 9v(j%+6qaY7j)U{0HAH+VBn!IUbSoId)a at WO!u2OwY6q-&=p
zKU?VRLj8#8eg%S$3E?ArM%4IU*Pr(C@&XJ~ch_Cg<@@_x>Fj0mE8Qi|cv|o;W#!~j
zPJMxQW_V&z0Dj`fiy~`{&&AC#$4CfMv$GG^FADbV?H*)Yu)7Vwa0)U#UN=+&>po*_
zOckLLaMF*|3~U?~nAh0qRXIfKPJ`PpVN?fE4*(5L%gB)JeSICrS?}x?!upZB%N at XK
zQuCiJfIB;uufG1XxQK at biKG_G1@Q8x`k6M;XGX-MNB!s+xS-ClR~lkaRYSuLNG4cb
zztVes0i{;5V!IJ*!ms5Ot*|Yesl4^!UoU_GPzqG#c6C4wDLKXmHh40r1=V?jLqltT
zG9vbY1dXDnI1`RL3;X)^xf4l+ckjST2lAauZCsC|qb(pC0JB@`Oq>qxWckP4H2nZt
zszenTUGw?&W-GxrC at 5&n9BXP)TIxQI&$Vto4G<KkK$?#)7~bZ62|xK=L&IC(f+;)_
zRu?Xu+_nj)A0kvPnPblmn0|GQVyLCy?#nZSr!KBFK`Cp?05jMMOc|^e$0~M<U*3me
z?qFr!cO2Uyc=YathqH at Xbck8SOS)5`OghQ`mSLcKF5t5G)yVQ|cRx(8tP)mAGtZ|S
za=EN55$}30-QWCydA{1)+6J at WMk^~Zx3{WRR)J4m2spiay<hi!j3mv0Wd9J$kTau~
z+5Rl&D=j!)uaR}{Fe;C`Tetl9=XH~gxvNHB+!w05hOd`5zO0H^_t&Txc-8BIB at l@`
zO!y<^_#->YuHX_i$TA^TeNiUo4GhE<iBz7YWYnxEd90UXc+$%1S$_v+nv9H_j~+3A
zZ}trhk>Ya5($Y78S|Gs4Dy4dI^YY5u*ktsxKgFy+H{P*{09hc!svq7bBGN{;&E?OR
zfiH<W58dgHh>5ua_~xXX90^dDkW*ddo^z1%<ZeIqGBh?03l83f^{pUkwm|rmfB*hP
z**<I~R8&-yRq`?e6p5;8iPKUC50XJ7toxjAfl}QGkieT=%OMLQey^{uzo(L-;RSh~
z3OopGH~~ODr7QEgLCc4Z0hPs8)i4&xK%962!t#+w1e;>?x_%uu!9c(uW=Iz6w70jT
zq%U?NPfcCMW_9caUteQ}TY^QRCH{Q>5Q!U<fK(2=y+xOnlcRXz1Oc9jX7}#2+4^yK
z7DYgieSLjXhQpA=DDhI`5#RZHk&$EoN$@7dqV|s2a*Xn=f4E&<;5&T4!e{pJGCr&j
zm<Yb`K|e(mO^l6gcXs||Z7cxs95N<>5U#uhzO1dSaW^o847;hn at W!xv9a}k2&>hmp
z$F??XZRxKeZ+^efS49H20|$O^VT!VHCB%r9*VnJ%x(di5kOTcwQc^x at o>lkuer=vK
zgDekX7cK_^yu6iAaH&nP#n;~2n(W}ggP*&)lFk=9*!gboZ!5euLG5P|QBgBHyB+rS
z_GoxuDM-E4)u~Ww_s*RiQXW$T8yg$MJOBhd=f<|8S-pIjpq;8N+q?a^pAL5oE+&nN
z%D6bCrlzJ~X2wn=5>0`+I66AkkPs#&CIZ5``|H0uSz23n{8?7f&``p*-~<5cmz|xp
zMJ`m#yyrpTFmcyk at b(G~)Vfgd;@r7CE-o%;$KdNoF{-Oc;GOJ}aPUJNYwwlc*8Twj
z-vFv$VFP=pD=RCl%IYQHyt<0)G45Xe`StjYz2c~Ho7GV;4#vE;Y6+}>JzP1?Ge&to
z?p4dm%3_j`Ua#1M?*nQ at PuX?l$`xF3GyW+X(V;?{Tdr}r!Q1M|wad<)QdjJ_aQOO)
z9L at qSsyLOx{PBu!`e9B%0X at oy0~Vs8qa&fDq(og)AhnKDWF4pW)8hIkY=awWj1CH*
zU|bnL2I&NM<YLUkT}TibwWJDg9XVyx-Qcm>jXOyWo68G7dr>{eR;>r at DbSD~cbR<~
z8q&2LD+&z_HFx at nJKQd{CL6Sciin8pdr*%}C>TX+Yik8!vH|=H2?>RCZ0rlbeL5th
zJA5IP1Hl$QrEls~vNuw|L=g)@{r$L^6tEn2aTInKS0|kG6TIAaB1I;ACZ1K=YuoYT
z<%;Kj#v|B;oDXIKiH4n%Q%>mA7({7;FTBUYT;^R}UDfsV<jBE9MR)BN-Z<l(K(MaN
zj%db6xNgU?hl<K}ZXO=UQ8&JpU#kWti5rMZyBh|Fhk3^Ck&}~$kP(=enBIK&Kym%g
z><!4rKsTP|<nX(6Ta5#71gwO at IJi}?pg;t-(xY2^{OB7S%YvO&KYnCAzUcr=%H;=x
z26qH5W7&W=*aSM-df=Vq5o|Zd_y%!heJw=J!zjB;mk#4nwwW{aIIi{O`30<{0 at J`g
zG4qDULKh)Z`{kx&?58D9OiaXuu(;d|021C;ugPZ~>%sU15JN#-y-Aq4U{k``OkG^u
zted6(G;47q$lWaUUY5WB_wuDUF6k<E><xr~*nct{RhDLEXzfg%967LmziCa#F0_ at 8
z5CpIQJWw0z2e1BP%NA`z!#lvgaj!0l#$z)j>LcwIwIQC?plNPus(lyEc?Gg9JgT^)
zBqtwVc9R7vi9<9*pbO>}V91Yk__Zz&Kq*;%bfQQ8%h*YW`1sDMo=$&KY9xH)q at Erl
zm>`gtIxwayi&JL6O^;up%-7!m0RXQ;*}SGsmd7ie)6%;6?wvADWCe$W+*IX~i<iDm
z-}oAy&F;PK6)-r|e*#PauHpv-I4v_1_iX`Wd~=Uo?&C8x=9DZWK>N6TZ;Y6Z9tq+3
z;d82~1;tN1dO3%T!RY`BR)#Vk-4~(4PJ0UrMOb?S4h>cpVp7piO|e0KqB6$DY_Mxr
zLV^jmZK$gH8}bK1OhfT42vxW$Gu<cdU^{2z2b6O_>jc}wZGBG8&KP$wl&mjUlY`G4
z5Z1Xr;^D(XO!>&Bu at KRkavEYMfiN;YUft3{gZqkHT?3+{nIRu1q@*0U?39sv|2}Pv
z95vY8y at wC|CEX`aY-Cr;cetydu<-im&=}Y<kJw+JilXn1#{f3LF$h3aINpn0w1BJm
zR(@{A at +Phcf+&DneQ`^n``4>V)6OMV21s$RqhjUHxCwK?YWmBUyZ`+8^FCxFV<jH}
z-bhcqJ at oEfe1A`Y6p at jU)%eNcpqRCF>O-<t(LgN20Y_G_v*VRIejHbAA{>p1ijq~*
zJ`Cvy#e*Rt<D@!<G_*v0Uhhd;U{jRZ&s1;7b7&GR<vD0j7d6eIL&OR|9`;3pXJVb<
ztM`)q7BaHpv>;mAbf$l17H*2I-f?lpwqv*#{EUvyY{M^1CYLVVg8%=fqocUV0-O!P
zVV>)_?%TI-2?X4ZLr?EtS$V>N!KR;8%ENbgX~6=>HEKO8 at 4pE|9i@;UH!yO$eER<5
z%ZqK~<W*l at SZ^ZW;^aKDYU8meutX#k at 87?{IP&Dlt%e3woMl=Yw1v!wCiUqP1DF)5
z6o-aV{`m1DAv=4yb>Uy#>S07=BrcO&UH<8V9)Zh;hK7c4`5-`$>9NdQCl#^d>;Q#O
zNy}wqXBg}85WpLW;Cb7&ZL6I6R*j3}pK at lzK-dL-M9eU`04xdq!kTy;ydm*p$EvET
zHUX3hN=o8X1^cB`R|$`f&b-(O7NocyMW5#8Qfp{vARCU0i<_F7@|dK2dh+B+3>*7s
zZ{e&2Iw&bAsldZu5Snm<Q{J@&2S8DML2TY0W4)XOmbj0}z4v+$-}xfcisI4D^3Bf|
z>hA77bM|aNs{LUi#MaVMiHyulcJR8$51R;>(xjJX+iG8JT~AjW5toOXSa2(cv^49f
zQ>U;%`ToNPbR#SXeE4u$Q}WU~_Wig`9=ul&aFr)@+3%yHqoX+GNV6^!|E=ujYYU?k
zR at _CPaGSvl%nYv^82AkQHIo$E$GgFu{Qdn+;Z5$mC0}TBWNd6~5s+U?y7obM)2X@^
z^T9Nnc=aTm at JLzVKVGa8sLF+k(Qsg^bNY~^l~c6Pz5$X1q^BSDtw@#1I$ul`EoLtZ
za0UxjeSM~=(T(dIpF7^zGBwh|DX*xA!-^Q$!oz8>lyK?+7qG1#0XgS%bo`Gw51F<-
zQ#+%l7o@`- at a!2Ew%Rtt%TyS5V0Bej_Y>WNd|_c>yig|~&nz!5Pux>Bbiu}^v;Sb}
z&h=!ujCajQscbv=acBxqHS$NiZbqttGN8dkfB5*=-v^Nm{>n?Q-$xo68&SXlFWevu
zsWhX&Ua&mJ+B#s{<G1BGs^HM+sp%Ttzn4{1B*lA;{(f`930z4?y%Un2KbRPVEmJG2
zS|(izu5jhclikO#4R#%kzI2;lB2?Dal0wu9c?J-3*jie at t+kc5BiDqC5L{!9SW at Qi
z6v;EyCt5#!>cCX;uM8-n>Ng-g9F?4^Wp)(#a_em90ZvX59A>^jLN;dC0P#X}$BrGQ
zfKVVIzWMxlH_!<|5fT3zH#Xrvx3t_sM at NTBxxSM7)-6qoPgL|lR8UV1oPdA;HJHF<
zC#Rdm#l?wM=Jxi>ZwhatG#r37Tm=&)YAa}3>C4z;sH+=@`@SKo1GgpfUi=mmS!#Sb
zB0Bo^Rtg4ub9m0W*#=vgnWIx#edd3hLklT(=((xGeZAk=)zwv8LINAtrsn5$MXWkh
zKXGxiSN(%)T6 at 5fFT1&UKXpE+7Uds()Uh929E}%64!}^RPu<{aXpBCJ;#Igtee2e(
z7zY?lyf!&H{`8C$h)}_95A1AH0u$Izd#hg<_=Aj$46v6D3G9B{vu97CSRODq4tQ*!
zKRa*NV27EXpP#uzk)I!lot at oBYy>a&*=Ud*8tN`_UYscI&9iQ1MIks~x)4&ZR|7IX
z+0y|{VFXctve&jd0m_CrhbR)kMqpqdWs$n8t0)z#RPfJ=RaP*A#i3}&DfoF}pXcHg
zIl?hop<*Jy4WI<ODw^u+HHnp~YHGV0KLaX4_3-R$zEZb|jM+OxDS8G5Zl<CK0KtYD
z_DFdx?(FRB1WchP;&9E=lbb6CFV-NPXozgfOi#ziSFyUXR3Y7vke0R$!o^8QO<1SH
zE at 1eZDrL<>id+xA;Yt;K{oMwIHUYqg%=;>DfWe)})Xnw}Q`(5tgWrdTaTmiznQJ3(
zrlumt!#yy-gi&PxfZmvEBbGW29M}q0b+6tMPYMe=Jb^vL4zMaAw;%fW$!|08dFBKS
zY4*_4(cv0p!xE>hzlNLZAO_ir4gg8OTh2Gb at jg`H$~xA5ZlPdT0GE@#t{U$Q7yOS)
za&PPy=CC|+E;hbHwKMID$RPkV0EDDGXSWghMn=|Fj`)#L2bPw44)}9L?0-;yd8+Ru
zuBE}%YPb$1w_=59`On<7v14_2+3J7<A}M_LeMaJLOIurY%rWPCkl+xqTUqr3_6liU
zIY!w9*$v-NL4nuM!`&MpJR|%0bJ}m;zUdd*?1BeF at y~@`A2v?KJGdtC<8;~+05E%T
zRVZH30Kb5jge%g3B+k9*c&NM|7(W at tYRINFwe|Pz?aVSRFFolb1C+9)qy&dxFbb)C
zMw at ADJZm>AEGXz35wQyx8J1&gZEY2Vi^9Ux{VdrtAkBn_hihhDSpt{^uL4~O*{Fqu
zMb^0&&zgRTu5q3zAlCzRY(DZpEaJq6{{H at j?<^rFCoC^7FY?c-&1s;<$8lg3_f6vh
zN?KYDm;D<KVufQ7NIt<Pn&OV%e)y0cQt=7yh(z#-!l<J4y<fWHUQbj>Sy)(fbalDE
z(%r5(qpe-7XL3E at +zYFdr%#VR*@NNEvYyuK_xDJT<+%>juzqS$7xw;T55p<$2y^Uc
zo%tZ}ZFm?{EG~h6={!vJ_3ALoUP-qf>DTC2J0KcjJ>|ebuEfay7gTVH6nu{=Uq9>O
zTU|Xp6mT`QwWVNYX09QXCFmE~?ZHmPO(di^4H0<v?%jLSrxaL#Q=mEGT1?YQ!rwpR
zxtA|rKI7$$(yuf12k|l)U{>6pd%jFqyDu+4{$paoR at CB!@`h#3Iss=x;V$e)q-K>u
zyn at ZkvlCr<BNvIAu}`-yj0JayX=!PFd||DDav}Ka<JUu<KR<#I?&;H~_?UnUvqRl+
zz4`0wOJlCtkaq?L2lpL4$_O^`60ob2lM{Vz6dCzu1RcD5d}j9c*=ebtUd_(Ujjzh>
z)4B;sAKPl#**77uxc)~37-`y~<c(6uJMaZLmQC2I;fxeLc<QP=76<_i=j90&IQG8I
zcS!nzLK)ODe|_;<7TXpDD*kHOwyr}^8k)m(K=#(k8J)N%_lU4C`-6H at Q`3Zt-EQvg
zK|njP&HPmu03g6%Q*g{2f`WNX77#-)<?a=~On at iF5s%-)J_k;zs>)x!EC{EI!(O;p
z4fuX at R=_gKgA029T}+i8yceuP^6$<``NEpfO!aNDv$x+BP1 at MdV57#)FCb8Nb1TJE
zn}!T_O+l1ECbb2%@*xUM*B^|>=C{YkuZdYT$Aya8e(D3@<fp@(t at llPKdsr%ujRYN
z?DRSJb8XZT_5g9mX3Djp5+0sv_*~d2NR}A>9y>{aDq{l)uyKo)M1_%#++OXw0^Hnh
z-be7{d;NBL_~?=J+*4lw7RqXBTc=jlKCgPUtJbdca{W+`m8c>TQJ<>N7&0!duPi7m
zWS%EGkt`1}yQ(Qxa&_+8lgi%cJAAynD5jvipB5LIASl<6S>EF6I;W`I<%Fk*(F9^4
zbK&~Q1aHwYF!=NK?L<}<FTv-h#WjQD3TkSU0NuU4y&*(Jm>;5=au;5Vc6 at z{NYwe!
zRg7(tp-uvP=T4s71SBRrGLo$CCsSl#AgON!+v9_g9{|0^U!^QDZ2Ge!eEYxoBO1{?
z_4V~QZ-&b|0w@^GI}0v>hto+gh_b_P!#(GyV*<fdX5lw5A`}(F6+H)%IX8~?D%D<b
z-#x-s36P}O+1VdUy7L2e at efnbNgW)t#9=k1u)S4%Resn$kG+l4R~;Pqu}2tKjx2Y?
z%6P2=gcod?7ZCV)i|a>J?RaTaeqlkuL$|x~p^fV}%V+Wz5G=%<z~Xq=cWuHe+OHl1
zlnpV*=z4ut%RgS(puXEGla%fXUTWQs<<*Rgu_tvQHRjl8HUHxaemG2)ReofgfW#oM
z1I2MnCi0y(Hf9B1#<@<s7G%3VOM_e0!CiH9bP`^@+FMgogU!+B&h44(t+)%vfkUeR
zV+g(wHgOuPPm(<`)$G4xU8~9gq`8Zo8nzud3Tq@<UJars12(<db-$$1N~ecR8<HAg
zbMfK{BO{ixXV0c(T5{0d!v2zxk&)VsUU_+97T_0Lowu=F3- at unf!98wFF)rr_>RVF
zwngcs>o|jMmj0%rM~`A>7(AVM{%+b^$;KU;5U?Q;#$Bwnq7y5m2T}^a-afWazi%~l
z`>Fl3 at 4h<`e8JNp=(VJrW<-iH)bQjYShPj0gJxn<60U8+jy?$8yCkSKBF82~4RB;}
zORuD_|H0ukEFR$Q`FnTbQZh2O0}n$rX(y*6v$M1QVe;~n|I`}flV@;cV9jmD0lbw4
zU)OQ+EK=UT**DdJ%Xa%7oo}p64-EP1US0i#+8#JMw>b3;fEM-<;x%x1IPSE8H$zyw
z)@IkJSovs1cq5H#1!6NEF-`N(;lovUae1`YOZ%ggsAxD7oVDPQBY>5U;)oewABu-R
zeiL2U0CokGlpK?ilG?d<?+!LLwv$Ih>kL#B6t-gB5;eyS3=9ql2#`%J$PrfUZ_dl$
ztpWTfEnQWP2)F*X<co`oQq#R5ZWEnV at 855yqN2)s{`_V at z*bD{7cK+{S=9EeEH9yK
z4;Wv2-;<~h?y_<Q1s7>>`m3No#ATxM?1=bAeT*;aKe`rwH3mq%K5w8Qf9}~_Ofo3H
zdC}JPs7KL|A+^IHc$cV4URWqPKiR|MvX{eV- at i_Ij)?ghMibNDGvBeD?53YzK-e#x
zj5DT|59V*<>O$0a#OXT<Mt-mUU%0((tUVLY7#Pyb%nXiXjf{^|+<Ua{RcV)QhlIQV
ze%UuLfI}=mR`f at dHADuNAL#RL$eN at Q3z-vHe`PJD(d^ubyFU~&Z43N8wG9pT)`l_%
ziP?9WjuF$bWQR%=zlPL+q}9;(r5^KHkuo$gVipn#AH$;Ko9b#aM at MQN9v*!IgE!#+
z at P@K~EtX$f=*O9BFn+9jrC8eV_mWelmzFX?OvI{CkwXu+%f<rIL<YW%yNDH52 at Zza
z$lV{*69^E(1cimQiRW?QD-h>9Y%;8CE8G|RNK!6nw|53+V8Ok+rw5hXcd$t4z>6Ug
z{)bv^fh^!2BZww`m6a!+4x~vj95XSLm)~;p=FL3kVRcjn{_kZrYQe+fZb!&m0|1Eu
z$(DW$CrF*s4zIVwVjjS`og4Ijae5&1xtpBvX#n;BRq@|KQbHwXj6o=t<}|2W9#zUL
zq3!%{uQ?7f0GJQ@`_~KW=?B$gPQbf({^G@^zdeKJ&Yn#XUjSI36LVT#<KTn(J^;1A
z?QzM8%K6k1uY)25w;2UcU<)^)KX~xqzaVOm1t{0z at 8Wdcix+b6#0X0~reXoEmb#42
zygh~-c6|kWGO;v|yF*N;F;U?*)~S+HQ{~_foWSh;|DD<>JV{<*VGU3+OyJvCQcszN
z|N4$d9zKDhl@%B6OZxBSS$zjwfAjWjim#sIe1M}3M#KxHvY*X#nD`*HgxG{TQ<z-s
zFBtNOitdt at miCw(p#i(0p{Jj1wS8^(Y$!P|Zzos`Zm4vd?Ed|B7 at ndUFU0>YUcoK6
zk^DwCRk?7kL(uHamQA#DqI}qaPE6On0+xxlNteb&B7bhVu#Jie?;kuoUVha4?{qRU
zxC?CBGrrdKmN=6Q4GqCR^o@^?r+tv_Z7&yr at 9TWA1Gcc^bPGI|=P4UmvJDEK at u9fa
zn}Upt3{^2d?u9HMvph at ZIsfApc*qG}h9u|VCTt|Iv&-ri4_BUjtIcKm_1fYTJG$)1
z=x9}CCE at fV-c3^rwT<vLN5;n9z(4sl)OaoN#5rf&*QNScaFeH}=f$>Xl-SA*h?6+Z
zT4divYGPsnPAv!dT~}8ZQ~4A?TZs<Ouh()jG7t%2?@OtxEfzF#a&mA36MMMwn4uhZ
zL_Z2C!J&Q3ILq_7tIMM8nc7?_+l42hwhREDajJNGl;1|IIQ9nsp_zq+pNY@%*^xtY
zx*hW2xK9n|Ero<`K)?v1V!<T$`0-<dy7ttZndb2hqUu<NP<wIFFGu(gLe@=t36)_i
zl5vZFui7dL802PX at Ezvor{W0T7AxT at d-dvXx(*zKymgBla6#DKV>5;_WEX@)MC2_k
z_pi^kt{jy9`SB?UHh*9E at Q9s{j}nr2a9G%oyg{O8C$NCcPg1rq{_6$Uy!l~g!ZL5;
zgyc5`tbLGCv+5W;kl4s}%VootK9c$#zczou#Dp!kd_L+v?eTQ!!ht`Z&iYo^@}Te^
zE}SzmGQzWa{rYwAojW>v+lcdmHkU53A3C(HtJu-;;asqCiTHnYIdx4<8bGx)tgP7A
zur2CB|Jz%iffQ)V<o>pBHPX0|byWb+w(iNDi`V4N!P|j!Yg1kUky-W0k?*%`Av(?8
zYK-OL;=;zmrTK~RhBHP+p^$F&@$zOh{Rj&SL**Yx_Ty3&(hJ7t&S9%<jQEwrCJSss
zf_#3Gm!ZSJ0^ZB#QrB_1I>64%z<52zqTVym(A>a&^iSF6L*jWhj-fnecGl%(X(<OE
zpV|Rtt<_>KjtBL~uyOiLQ&SVA0D+PjkAE)B$#?&0jEYSsmdCg?%5)ot1<rS+pN8y$
zEv|*OdV}A;9}e>gcQsBno|7uv at 960Gwyw@%v^C|isI6}HfcNs89*!^J5ZC<&58^#5
zWEaOWbMt&wy_26klm4~vICv}<Fc>K*DYleDf{Q5Cb^fh>Ca|?tYc~@U2QM$GJ3edu
z#J&+!0tX<U9Xxn2%dkXi)V5rC;$@EQ-xoO?;mT9M5j(!Tq7<1pen;spYfa at 1lJNL=
zTs)eRt0e!v`qV$%;sj_2uMprEpC%o8wtac3ni^@I<&izOcoIUk=WrSPe=IQqM&#l0
z(})qLv9Ynyc(qy(YJhvkwX|rl?*I5PYCv|Ex>5jt(lA1bxQ&7WDFl)cFg_egM42am
zE7<I*aGRl=SS|pB6k~ju+XRl~0xC>MPDU-mj~yL#@V0=k%pA<QJ~Ma<@8+<z_0XX+
z^rCuM`kHs+(`L48*>Yk(ExUw-!N<M-5>!1Paip*jb{|(#qek~IF)`tIl8z4jTXP`>
zUi}1du6_GZjvr}Svesc7*T$5jr#ArbR&TFfeCd0c=yY4ADI4R;4;^k}<_->2eLsWt
zfAVPAc(=*OJqdZ1_4f_5&qG$v_n0;(VDWF-aG`fMMm4;~#Jx%q+cjT=$Hs=j*;QkM
z%qKhYUh?MCdU}kH9z6mh*B+DD_}s+HP4O~!S&w^)j780*vvd^aJO{%5_U+pQwP-Pf
z$sr*jsM5E%=z)xj5H-oF2+;VQ<&jSY!BLyk|3gW_*XHf^%6J<dc-99Lqz{0~)WSkg
zY%B}*l?)7!@bF|B)E&3IKfk`DoRn~gm)Ac$yhXdWqrF|>)Tw&94sn-}Guh9kMn*1T
z%iGtltdRmHIl?&nuyOTeT<+QaJ{l~Fj*bp@(ZbJc`ZezS6G$Zr1g_<$Ee}xW5Tz$v
zUC-L7WS_&PjghY6pLAD&lKFvIw$Gh!xrmpYxE%x%0F4OQ#?+<rH~658jEq%mP4gnF
zlKFB#CISNkaUK>h2aa-bOG`QpZo^q5+%2S_pm0wmC8W8apa8qI`o7z!h#D6aX-C4j
z%PA?5b#``oK<+=C{flERV>tUj`OJGBywkurS4jPY#nr=|^YimBT}JP#sAq~DIr0X_
z?MB;DDOdv1&f$d*UIR^ilS@$#A&qKwzxf>3jzXOJoM+~z)$z|EBNs+mI3c02b8~NB
zSy at pr`FYi*ZZ11>J-|>RgOdFDN_)6e?JBiJw3z+W2Z0LHH~!?I&sk{>ll at mj_>MY;
zd;YJj>yD>-|NloulroA2rz8=jVI`4Bk&$fKqhw2=?2!{HgrW!`vyfRSyOebuSw;3W
zN+^_BzONUD-|vs>(I2;aa?bg at ->>m}t><goHe3-vR=fQD)8RYYgz08yXA8Qfr>Eb8
zd~e>+)?ocHQZf0Qj!yRUyt85d&;Ix1)>iE|#(4?9<7ulGrHtXRA;#5yrZ)LkI at KW%
zLPjmE7FRFNF++WQ4%8$;K~hM_+01?nlTuV-^kTehTJI|GkjdmUXAe&?)+09o<5!lK
zQ$5wrWc`0vGdOq?!p*j5uLQ+!`Vlx(e>!x>pSu1yaC;6mrM7md2S8qKVwFopB`hXp
zUVY?^Lh(0On=|nHI0uW_YH;CLJOwRy={YZkw3?3nRZ_o<NEhtvf*CgPrv3ZGAu+`g
zFwQbu(V4rk4WbOc+~$Sj{s95|oQah9$j$YIeWz?&mSNc4y98BL)#PlQ?)IOP&02ad
z?1LvJu7F87p`-KSAM==r9j8HDF}%*RsJBMfb+p-kBW0jqs`O59u;cWYVS+h<;Nj_s
z+kpU>AuSmWnM1_KOe6*aIBgr*zG(Y4t6_daa>}{A5!GDqC#rT=RILpg^B&e}jRuE6
zckk#$LHmqwfL~=8Rcu``O{XDdMlhCO{w9BRawrLqzOY3dPLb5q)~bQ4y>!yxWV6<i
z<a*BkrkZh=Ef8w6+KW?Hd3nNUk&V4(X_1|3wBrTt55{T3yUL50ItMe0e5!!yqJyK|
zGBXy6smPKgn~daSWT+&Zzdf09rhvWS0gx>j9#;GfS^4w<*KYzfJX1SPzcL7UeQi7u
zz?$D6S7Oxr)x71=uC6W|m;npqAF`%r-0V?kC=;ft3}3tSF3e66d at 2K&j_BQt;8Sg#
z>s}&HdJq#M{fkpxUY>Bm&8^5R`m|y8rP&wmFLH9I2n(}A at v4v9d&;N?KAD+zp7v^c
zHnX$i at T2G46pO?<ilF6T=htrekP`=@))NPkkr$5u(epYE4h~BzD?A%(!*3cI84qjJ
z+8R=JSsz=x7a56^t%S1S#w266*e&TgQVxP=;6XvnXb1~aO{%(Yt{*1!G^t9Hboc*_
zI>ufFDh{5&mx`K>9yI+kfH^>+Q+!&D&83 at +njvIhE8R)Cy(jIMXaNK#=B6^xZT-GM
zLnl3b%XoL`V)}?^i_{%jT3XPa;nvI;xPsw5Ijg8BBs&{t7ICgWAql%PajUg`XA$i~
z<(qwdX7JX86G^t$Gq6mLEYaLyT>ShDO{uDD2{d$c$A3MtYD%FZU>S39VMrP`kAv4j
zmwW#{$$6wT;o5kYsuaQ@;heoZJT$$%y;z|<e|{T;@5#ExTB+Mx04 at q*^F9bRDMXgy
zaBOj)5=;b6mimWGs55TZKt;f+A0~uAbFf1C<)x!4r+#;|dcElMXvg$hn&zm?(dK9F
zl#O3;ad8- at Aoh9zkxg=&YnWTh?dI^COglHaySpQ|)#Jwmu-r>a>)V*yL#6))swJIv
zoMTX<&VBy;1n4c!y94$RDH!*^Ng0c2ZJ;C#qwiua{6pC8;G)6OmXP?=EFqWgx&-gG
z(>R?zKRu2_E<r&-*C*?bP>FO(-NDL6Q`3#tuU{`+LQn?BRr#Y7fs)EJF5O}5`HKzV
z2gv^V$Rvv^rwoxz1oLgJj*AQCE{5Y$<MX29v~+abLPEK%KbNtn#)8K;3Gb>q+ps(x
z?;0xlesO&E*Q at E88bwA<@gG5{#;v#%JK<yiL=#U_Ef%`n*)lX8$GfD&j~+Gs+4s(w
z-GlN-!|wzE^W0_^tN{P3X}{)@rwwbj$K-~=hWN#|^GT<Jap{_kVN!``GT3x;h=No(
zHLij%k4oMyEtPip{)x)d^N-$m=?95}>e+5xdg{S;9fka2{Wy;a|9Za7ovi}Q`H%9o
zJ8^4s!=t at t<r*B|yW{3WWh3>-!HL+%jEL_e_n-H}{nI1I0w|`zA6I<iXgtW&DR5VS
zKs{|x(!&&Z=;`UNTPL|XJ3B{oR&L$8H4wzBS9)<a^kW-0_dp-dinTa{%%JJ%{8b^u
z^$(SKbwm99X|MtbK_zx|g at kOx$Pjy9etl~Rj>PTX$h6g7V6&D;BqG2D3o_G2_nU2>
zHkX(fyNrwsR*L|<<rft64i5){I_%=-|C&D3I at Bd9DvJFY*eCbTKmU+Fe)KCW6nhjK
z%d&mD-cX937EeSqw*P}U>hHYPy_)P(%M&)QV#K at o(Nq5N+ISq<(dYxtQaF_UVTUVc
zzWr`xx<XLDP}0%X(?$^|Ar%zpA*`Uw(Or1!Y?M-n6A7N4o=D7zc_6o`y@`p5ah7DS
zWuTZ?z++=$vs{+f(9u*;R%S%rv}C2IT$vaJ2HeL5XJhc{%BkVKgy^no6}92D1Bm^3
zl^>MRQO+}wmj~b~got>l+wJPTpAw~24NvW7$9@&*zcWdRDbJFUu=yDq3{EtCsk}bg
zE{bS**REZ%^1Bl0%2v^SE_Q_i#N>kTZKO5RCXB;-`$+Yjix)4FTUtnwk?i1x?iWR7
zgw+mU?jC;bhKE<E<Wnr5;shX|snL!tT3T9xw{G>tm&c at 4J>pTvPH_N2)JPxz4=2`M
z^oxz1{U(TSe^Ti2its&3{vjuuZG%CsgdQ at 2eUg>8a;&JRz(%~jxnCoov71)Yuu=h7
zz at D4>iUn*C3O*;!HApK=Fp0gL_ONGF%+9m3v*X|~7-Z}a0o%Whn0`Lv*|WEBLj(c<
zi#4fM2Fc0EOO#XY)ApR=mwjR?DOz|>K~<GWK7`Y^q~y5U)UYqO)AhulXrC`%4A4Ch
zd;o&bKT0Xg%@snDWh{OG&II=(;KY+?ft2>2O1-7Qi10bxV;=y=jBPJS<SJ%KsYHp&
zU|VT$@CII$=;k!0?k7cdokeV6VPPIUd}_iKtIpL596k1FVTpEhG#IR3!fw%HdU|@u
z>OwG?8g7K=kUV1l<pjjCj)QgE-oAT>CSO*b0)VW4gW<*vQFl0wYy`-1>hk47V8IVM
z?N?Xh4x_iQuz;AUtf%}2Hu^j9klwzP1A)acgA>=Ui-Xz8ejs#KuErE_7t($7zYiVi
zW)oorGyrjTSzR6DPOq7S)jZlTZJ-=C=K8S)l6dO2e at IhethDzIEN#L?i=Vrhwr#a&
zkS=-f;6Ze0``IfyJBb`(wmqAO9N;C92MwjA;rP7ru(%@51feH`@Q~cxj2U)GNlE{Z
zmF_Gv at b{*(69a)kK{6g5lAJ{1&ynm>T-iKq`NU5|L<Gm(p6S0h3u@#ObHu at H;@@gq
zR{tzB6PqTN@|U`Df>m&XJp%)zva%C^H`WnT&p*w|0=p=T%`r$MYnG!stU!_VDZlUE
zzhhS~QqF)*CXpTWp6t`f;yQGQ8R;wG*FdP{HYjM;8;(-Gr`)}$DD$x|uduuB&6_tk
zrB>e9s5Y$I?2ldD$Y%<{GLEv6)^Cy3*ViA`T}hE^J33-)>LY5?3XpMYx7WNN+D}o@
z<5%M}m5`A~P|qJ<{P2MuYr&s#%#TXNC=ej<8SXBXZf<U7=HOU=^5jVr9wcLdI6eoy
zJ^Ju6AF5ShNu+yj`&kHGH#z{!5hwKLZ%t1}_Gf0an=CB8t(p3u3xV627qufOZW#N^
zk&GG3i>|ImQDc~n&kz84vr24FfBEK(B!(<FS1QL{jufH*v2tW607$`Pw7sA*_hK#P
zRj_e6f>*`w#tmxZ;D9KQff!1E{``3?xndJB at t6yjh{$I6Kggx^`LPNQGc`3e2)zt%
z1dg75{`^^1-dbBgK;V$q!gZ8H_vx{VH^f%>dM2v*;!Or>?jwoRVBJINbWH)LCl&{V
zgjjLkW<bDNq at Jd;OSA}FDc&^y3ra?P=FA!095Y((-MdY#tb)+T966!`@@01A$`vG3
z#V{8vF2q8zva)Lo3=K&ud=+wwnEk<yGfcq(*u!efg9i_AsOzQ2Y%cj{8Y!&$!g5A+
zCy3Q$fWAmDi5Bko!jF^r)-8&|fpUUz2?@LU`ufO)cKS$uhkFJk)blQnv<l*07%&|;
zjuRgrPY$CfW126?-`|8%xu;{?>a(!m0Ro_;p1v7ZYht3Xr^f^vctNc;OR%%>Q>_G<
zX=l_mG%D}$Yv5!#Qq=(rex8$44)N&?+&e}LrO^bf1bfmu`l9<L*fDHlbNgkDeJ6jW
zy91)P*Y2V`e at 0DJ6`f7%v+_*+7tb&E%3?M;BqRh?P&>;Y2*MzGVpw!e-lYb!-`_u@
zW7T&kCc*Xhf>pw{&ANxe2~VHybCTnp!$KeKxdE?_{Rg;D3AAcG at oDusCS27bg*g!1
zklLbS%qGFQQuRLrMh^+{0Z*eh2Dpp6^(3_O!lI+sIXF1rMiqcM at 1jL*xY%!#z_ciG
zXAFAFP!40;-Q{){Bs2q;{OR*&)0%tyD1g}Vn9#4b8$iO2Lfg-N1vYKm!or)1-KVRX
z(>1VV0Dj=KC^uqs!r56A8}PyZVR@%9Ngh*(AnrI52B-;7;$?om2)1p%b{VaJ_rs+#
z*q}H+_8QKHSz-rEOVeO|%o5Konam81`bKtkHaRT%PR9!eT#^c(Q-+mTK=m>ZVy#+{
zW_k$ak*E@%7GkM#cv=$&hl5Vfc%SCyZ${_~_kqOf*wEqP;v)N=n|sgyRnCtg*#Nj)
zEEFL>SwjO8T*Fogr at d3Zevv}tH>ACMxfMXSJopWy4oFTWA)_6HeK)bx@^eKhsc;sR
zmzNHrNl+1*nwlD%Gsjd?clXY*v9YCcJa78Tj~~@hR46UA_FrC7DJdz at tX*sV?}xC<
zj*cu6+$_b9L8Q^Pfb?%YY(`VE_}3k`brF>=;;h_Iymu8iGI|gic%-=c(G`2z{5kIu
z3oEN1oHV*c<ebF`kDO?n#zU6PgoK2<ipdxc@}7M}5AfDWgxi{CrPi2e@$}g<M)*uz
z`hc5ZK3ovO{&6 at C94rHkBZVj&>;a?>N?H*o<&M^O!-fs;j3#dbS at 6gJtir!pK~CBb
zWWIbUE_dn_9t;?_Ln0z?eW<7pS(3SCW at aw06QUyUurlBbE~p&-o7A(heg+5R!Rcch
z?BhcKxn4(%2;5FC at tntQlnWOxngZ~`4glO%;I#ifMzX&~+AxhN4^|o at bR;VP&nhcV
zX>Q=AsvfukQRHGd#O*i>2so!Fui^NOSFc`?8#T3Yhl1 at VABduHatPaNF2#G&fdNiP
zVo}>|wjwG8^f>!?{RbBo4VUTBv9pDG?Z&rdzSOvG5L`9`G2+`f_)1_-5XT=rd>C_s
zPEJmngFj;r9GD35VG15oIC1poQKT1kauNacLBbboAjO#xfJbl$ph!%C at 5~aC!oxdD
zy>~J&Fa(B%zORnpL(&AW5S0K_u)BYHIsh#2Qh{DkPLb!nB3sNYhwhT#-pT-6eSzct
z<VKpSL;ny_Nx*2yoH at gQOGh9YyV7)sVozGQ!xWJ189Qunqc+$njLwmQRz*c+___en
zUX=d%%^~K{#TX=qTXnG2GXTTS#y|zasvNPWLV)`7=TC$#0P>KZJ3SpW at VdKuB|M#x
zk&$J0NikVX|4DRkJ`OB+&WQjNL`KkS*RG+=;V|RGgzQoeq#{M01QZ1BV at 46)`PS)9
zP!UTT8$?+4_VyH#l1Gmo>6%+sU?`7kYD7iZz{%snf&2S(kY|jKk8kR0)7q}CE*xZr
z*T{}R3Ix_5Odreu0>)##?&fB;F!OUM|Ahu1)d}D`ZZHAdi)*m*ZQ3?MkU`JD;0vm<
zs{ab4>I*^$JRrUc3=SrdNVM4E0bzf6YpeF~_24WpBiL!r0!A`PJ`@Y$2p|0!Hj%Ro
zk0tc8g!!x%S_O>&{Dg2G^#kqU;o(8<mb<kB&-3zbP-cDl`qrQg{Pt}f at x^(5LBV(L
zZf^-&W^d%-<!x<O7#SViuyf~5<Z-f%=A*!%Es;@C0k~j!*DfkN7$v1j_$J^Q03;KZ
zN8~W?U07I%tJwMc((Z!Yz$vMH`;^p_6jfAC85!+Z+N_+Fm38^*Rcz?$>FxCuvume+
z6dkBp4-$dQf;jLDhwAI=i|c+C78atlMvV_dVIMXeG`gIzuHxIE(PkC?44COmNl78I
zP?RDqsJgE1Ed)q85WXPgR!a*Td`sNYjCANulF9u0_sfCCR8&^tU?k`VBT-^A^7!!G
z-W0ZG#n%V^l?NdY4-d!vM&RDyT>8pytVLq!<m5feGgge3E?p8A7e~C0F!+;z3U^o!
zd^8sA0j(k_0?1K+RWJt}zhzq%5zijLbXT$a)!l+CZdePvAH<kOHtVkH>21dx7Xf4P
z%uGRBTU+dOJa+6D04W at 0JOm~|+;fsAOaU7wu!nG&n+X$*Lc+q>?)A!GV~#H4zJ2?!
z9qBMjXMKGz)Aa7dA!LUJvtSCI5o2xuzu3cpWAx<5GAdVCHBC(sNBscrzzvWM1n{zx
z(_28}SV=6yHlx55kIAU1ql3-FSSg1`L5rAW_}UEK2km^R*B=ucJgeKHZDfR`5O1rh
z`r!0Ik9?P>W|83};bg#)jeXC<%~rV**Co24-N2KJj7*ui<l-WBL1-I<Dkc^dbT&3N
za0rzEnzXgG4~dGN0vxTUtTqM#Gm at i#WaZ}N;T+5N at 3ts&M~-Y+?ySPq3CQMSXo%gs
z5OrR?v6XW4>HY|g8o3_bzfY>GQ`XehUPAyEguD+A9zH~RuL2{iGlT2WSC$%fMKU3H
zgAWZ2^7i%z0OQ^Th2S`1kDZ)2z^2SgO>i0}Q{m7|L4npAfUPE`rsV;9Sk-R0&HUJf
zL+ at b3`@k|T*N<4<6c?l6zwYXagt+9!Miz+-92$d1M^?+ZDJz`G(#bOLO-bSP at bJJI
zrTK at uyAHu8N(fLML-gv`uU{#C0J#xrY80_$X4z8y`0>N1xVTucweF(OHh>GhfCIUf
zCmOH^2mEKzwegMK-ri_7K|n~6k&$TDNUOQnVOlA#g{|OzZ=L=(+S{nujSV7cEzZ>J
zY^-D<WiApX1qCr6V+nXY0BqRbd&@iH>C^Ss<yFJ0FYFZ%r~qV#!>LHB1%Qo~mKKw;
z_TXh85c+p$2uFc$P2Rb4$Hd$`$<p8j+}i#7Y_ at iG37`l)M3$9Ihcv|4_#1<R+;9dM
z;=|SBRm8;HhA<WbT*o(ASqJdT0K#$l{)q*y&<~;ncyF8{L<2lGHwWq%+t+4oZM~+e
zcQXCiGdV}coW5VSr5_du1X?ES1?VZKMxNcjf97AhPai;jgEnLDUO*dI>jev>KQfe*
znz~j)Lj&0#zI-wGIReTH|G`O=aAN*37MPYCUORx3nqb#hz>1(xmzLg%qs2##;20jB
zJjnGs?^2|$2~&7-Pm*cpaU=o62a1gizP;t(Ut{;_$%v}$B*z7oZFKv0mokc(4?nl@
zYggIQp8t0?IOtd+*nMxS;e4*XL*k6S-wZ`2BN?ta#u#PWjZa71ePhn!D!iVZZ|JD}
z`}gm^IZ52=*A3~nbZ}5zyLPSH&py)c>1prko0kaq^J9_uR2V>vsfht4B9Zv@$B%P=
zyo?_zpRTvi04ro&py#$QlOx%nk<@rDsqwB|$2FJKjt>3V5{GXy02s7Vn;O4<O$QN4
z77w}DQ5bjJGviW2+}GL3q4tiBh>@aeE_?*<9kT69ANIwUi{8n_#bQler~MJ<F_+it
z{7k;Ow^(a&aBy7VluuSaLz0z&;}M9eP0!;QxbiA_lz3lp&o4>NjT#MYZEY#LT*Nu8
zg&tZ8Jp at mtsJ5B^7I*OJ-6YdAuhP;f&#(H!Uz`mg2Dh at YQ35qrjFoC=f{RxzmR|aJ
zpYCd~jUL at m<MVS1vw3rWN{TwO+=}B$JZ8DOr|VrE$(@Pn=@(=>UKyMX=}$YicW-<{
ztt038$mA=p9%V3Of{eA8f-O at 1v-faJy~vKxsIfP?@!K~B3 at -9U1oerxT8ezM71Lm*
zxz+#4KbJ3}`r{wLTJ-`wtzx67k)rl5Uqps at p8fsNd-7G*Yt70vKE^Gxv#;7m3qwLf
zTkKyrU%8Ta#G&hf#A~~b3CRY=-n16$4EUq{G}li5otavAeqkYRN1;Y(ann~#cyz|f
z=Eu$DCXXj4uWG~_pFgjb)JXpQyD(1DHL<hPaCB at 8ewja}zN_Ij)zZ_?tOd)R8WFK6
zHZCrs>1$fk*I_`xMI|MXQ<-C7j1nr=0 at d#wIR_?P^EMBHYSs)~2``yG<@IMyE2&Yz
zk#l5X;v?fG6}npoHtN<}<V{Gbc64+MHdz<s=jH9L%^m9<CCq3rf270D>WQnf$kS|U
zF*$TDyRGTnJ5p7D4j=&KKnW|a;KUCim-p1?20iI{Y3b=HRS at f*WBHJ#bUn#W;QlQJ
z-zddBM?;eIGE?AY+NTVkw(M0{v-AK9Pf9yvi)SQb&x-B{Wjhs?(v<GUMfNzZ!%V~5
z%*XG1wr=Pu$@IaTK5v9tx~5){{3`=74wKq}-2JtRq9uYiD<@|cKK>s6*C1Up at +QNY
zviNU<gQo$Id<<aPIraCCdu(iMdfUiDiN%3|WV#!^@b*^XJ3`+_*IR^_FzFZUSAt7;
zXLeXCp!?G&(1mk(pEdJ7-y3qhR;0FW=?nIhx;Q$TM)r6t%&3)&#}p~D7H+I0H-Ff}
z&v>$a=<1p7n{gIG+l2F*3r97}UaP048D+;L3&#yrJ9lax+(TH<r?;Xe{MSzz(}yq9
zwEy`Mq34>c0>^*9<VDzJ_usF~P at O+BI$uKb->)tl_`60>I$_dln#utGe&*C!xhz=|
GzyAU5D?ecX

diff --git a/docs/source/conf.py b/docs/source/conf.py
--- a/docs/source/conf.py
+++ b/docs/source/conf.py
@@ -17,12 +17,13 @@
 # add these directories to sys.path here. If the directory is relative to the
 # documentation root, use os.path.abspath to make it absolute, like shown here.
 #sys.path.append(os.path.abspath('.'))
+sys.path.append(os.path.abspath(os.path.join(__file__, '..', '..', '..', 'src')))
 
 # -- General configuration -----------------------------------------------------
 
 # Add any Sphinx extension module names here, as strings. They can be extensions
 # coming with Sphinx (named 'sphinx.ext.*') or your custom ones.
-extensions = []
+extensions = ['sphinx.ext.autodoc']
 
 # Add any paths that contain templates here, relative to this directory.
 templates_path = ['_templates']
diff --git a/docs/source/depgraph.rst b/docs/source/depgraph.rst
new file mode 100644
--- /dev/null
+++ b/docs/source/depgraph.rst
@@ -0,0 +1,125 @@
+===============================
+Dependency Graph Builder Module
+===============================
+
+Introduction
+------------
+
+This module provides the means to analyse the dependencies between various
+distributions and furthermore to create a graph representing the relationships
+from a list of :class:`distutils2._backport.pkgutil.Distribution` and
+:class:`distutils2._backport.pkgutil.EggInfoDistribution` instances. The graph
+is represented by the :class:`distutils2.depgraph.DependencyGraph` class that
+keeps internally an adjacency list. Several functions are provided that
+generate a graph in different manners. First, all of them are documented and
+then several use case examples are provided along with
+`graphviz <http://www.graphviz.org/>`_ illustrations
+of the generated graphs.
+
+API
+---
+
+.. automodule:: distutils2.depgraph
+   :members:
+
+Example Usage
+-------------
+
+Depict all Dependenciess in the System
+++++++++++++++++++++++++++++++++++++++
+
+First, we shall generate a graph of all the distributions on the system
+and then create an image out of it using the tools provided by
+`graphviz <http://www.graphviz.org/>`_. For obtaining the list of
+installed distributions, we will use the functions provided by the module
+:mod:`distutils2._backport.pkgutil`::
+
+  from distutils2._backport.pkgutil import get_distributions
+  from distutils2.depgraph import generate_graph
+
+  dists = list(pkgutil.get_distributions())
+  graph = generate_graph(dists)
+
+Now, it would be of interest to print out the missing requirements. This
+can be done as follows::
+
+  for dist, reqs in graph.missing.iteritems():
+      if len(reqs) > 0:
+          reqs_s = " ,".join(reqs)
+          print("Missing dependencies for %s: %s" % (dist.name, reqs_s))
+
+Example output on my configuration is:
+
+.. code-block:: none
+
+  Missing dependencies for TurboCheetah: Cheetah
+  Missing dependencies for TurboGears: ConfigObj, DecoratorTools, RuleDispatch
+  Missing dependencies for jockey: PyKDE4.kdecore, PyKDE4.kdeui, PyQt4.QtCore, PyQt4.QtGui
+  Missing dependencies for TurboKid: kid
+  Missing dependencies for TurboJson: DecoratorTools, RuleDispatch
+
+Now, we proceed with generating a graphical representation of the graph. First
+we write it to a file, and then we generate a PNG image using the ``dot``
+command line tool::
+
+  from distutils2.depgraph import graph_to_dot
+  f = open('output.dot', 'w')
+  # we only show the interesting distributions, skipping the disconnected ones
+  graph_to_dot(graph, f, skip_disconnected=True)
+
+Now, we can create the actual picture using:
+
+.. code-block:: bash
+
+  $ dot -Tpng output.dot > output.png
+
+An example output image is:
+
+.. figure:: images/depgraph_output.png
+   :alt: An example dot output
+
+If you want to include ``.egg`` and ``.egg-info`` distributions as well, then
+the code requires only one change, namely the line::
+
+  dists = list(pkgutil.get_distributions())
+
+has to be replaced with::
+
+  dists = list(pkgutil.get_distributions(use_egg_info=True))
+
+Then, on most platforms, a richer graph is obtained because at the moment most
+distributions are provided in the ``.egg``/``.egg-info`` rather than the
+``.dist-info`` format. An example of a more involved graph for illustrative
+reasons can be seen `here <_static/depgraph_big.png>`_.
+
+
+List all Dependent Distributions
+++++++++++++++++++++++++++++++++
+
+We will list all distributions that are dependent on some given distibution.
+This time, ``.egg``/``.egg-info`` distributions will be considered as well::
+
+  from distutils2._backport.pkgutil import get_distributions
+  from distutils2.depgraph import dependent_dists
+  import sys
+
+  dists = list(get_distributions(use_egg_info=True))
+  dist = None
+  for d in dists:
+      if d.name == 'bacon':
+          dist = d
+          break
+  if dist is None:
+     print('No such distribution in the system')
+     sys.exit(1)
+  deps = dependent_dists(dists, dist)
+  deps_s = ", ".join([x.name for x in deps])
+  print("The following distributions depend on %s: %s" % (dist.name, deps_s))
+
+And this is example output with the dependency relationships as in the
+`previous section <_static/depgraph_big.png>`_:
+
+.. code-block:: none
+
+  The following distributions depend on bacon: towel-stuff, choxie, grammar
+
diff --git a/docs/source/images/depgraph_output.png b/docs/source/images/depgraph_output.png
new file mode 100644
index 0000000000000000000000000000000000000000..960bb1b5639e7ff615ddbd6cbf8a542be31d5fe7
GIT binary patch
literal 24719
zc%00<Wn5KT_xHU4ML<9rX#u4bK|lmS1f--vq`O<BLr`e}2|>D3x*H at FQBqn;>Fx%f
zvCg^v&-J;VcyqtH&%wiiz4ux(#u&fvZ_c?4Qk0j(!=c1Mp-_0zQsT-e6j~DeVaLLR
z-&~f}{RZDK4WCMiqy8fQrZ(n8p-?wa(&CR*T@%+P-LzHrFCJ};;GW)HV<`K|kcx%F
za>LfzBD>aDIw9D7-Soc4&y0!mXp=+d^#wI&_EX(*9+!0PYHg+VZ<k(Nd7?yyb~*Zo
z0fm2 at +uw_=p5vf<);KRkrh2~^(SAPmT-bH%jcapD>3xv$fRUa7-6(|1^pbKZhJFaw
zu$TRS0sP35=y=^Igd#L=6zy)OK(c6>$df0R+1S~Yo<G0hCyIu;!Jw+9hO=+XO~Jvz
zfkKgyk%e~|msD2=ynTy7O-+3xLP|U+Md%^_^4(!Si%QM8#l=4ZSx;FwIK-8d2yt+5
z?AOP6Nk~a&hYPiGb8~$(3vEAMtgWxlE-hg_fBt-LW0H!NmbO(vPfzdCrAt`S at 3_Qn
zKJ*AJEkZsZrJ(R7PbDWjI@)h-&Dz+^ETgCh|8tV>%*F=3f`WpptLx{sHVk=r`Bs)E
zhb1K?K1oSQeMOCpjd$NDv5<(q`IMjkRyj}n$&)7;k{TKmGBPrXs;a)}=?o-fWa*ii
zm&7j-6BE0-xutpky}-i76_=92VqjpH+1%{SxQXc#g>RHZL^%6B?jFD662ZXWU_?xe
z|Jj+RRF?A5 at i9FM3l<t08uGEf7w7%ec6y_uqk&|_GFhUpUNwc|_NF{~f0g{sOsn73
z+`>Y?#zrB3T#7e=fmq^_l16i1BVZKMJG$H3qytLj(@RPSP7gOF9UQpvX&(kA+_!6m
zenmt?wBieLaNt%|RT&11TUu69#>K_)zdGZGw!AnydXXsTg1 at k^a9!AwU*bJ&Xn%$z
z<HLuElA at EBm}xr2KO`i)OYy#-cyzu;97gvDUr<o6y`uwL at K2N1X)ujpJv%4omo%|}
z at WjNBxjAF)T8GOp5wYARIN7psZwMJBMA+toDzCB23xrT;JbOmUXEW~0Yc)L69z?t}
zluw at Qxg*{F<qIPd6DAWAlVO0lDmkYv3DG7 at S6}{fPT3S;8t?N}z5ZgoN2P5>>me|h
z5)RPjv+2L*bgzzPufQli{Cl=5B_%b(cv?TVu)x5 at ML6wsNVBuELqJUYmR!%1P~p)z
z&-d@&+wtF5Vn92bO5R?%_WRHKM-nnJIGERHLms?b&|mqPm#%xXxmKZhJ2EP&)P3L5
zdb)|m*4EbbbiI1J^(yD*fdS>Utx=4bHug?|ud{6dT-Kv^LuiF4q1{r_(k(&6%v%Fb
zIbXC|yjErFh~+Zcf;;@)neUFhYk)rW>eS}w=t#dKm~{Q1meJG`vc<~md9m&jPtS+H
zfB)9EvwL4r!9&e&k18!KH4GTO<>v1GJHNO|R9Balh=`~}zXjdb*H?eK$;&XHa{~h=
zW;{qOzXMW)kbuCa+^pBIH(6-3-0T`e2ZOWuwaet><QzJ6B#S*sV$HA4?>eukKivI(
zuf}OLW;CsoZ^lX>r1GuRNRdHj2*uI);q>-<?=uX$hHcF6KYn!JuclRb|9u4!dj$^<
zPrdn-kVGiuEoZ3$QrFD3Q49fxIjlEt-zq^4WJtdMo#Zk#>uYq$s57Kxc`%2Nl+=&Q
zs8hc`{fS{fyXI?{uxt8X!f37wI&;u^tZ^kibR&Y{Clk!H?U<UP4GIc6+VuYW?!E8{
zR&m1)4i9DMs6naemQFTBy at FPi^<z;{bX|{iUeA-|98W58!p5WD at 3ywKJ`XB9!t)b-
z?X)_)ecCU*%}dj{A*QKGX=r4$y-~kuINR!H7|?dXeET-ud!d8tx?cOw&Q|sQ5(!e9
zqL8s$I at w>v!p3fiQV<S$mai5>`EJ8p_q+F>9)h+>=Z#-MImzxg$lX0X8%AeKJfOrC
zgl5Yn`2YTGxHjFaw}*|9WDqb-VOTSglkzHfzWd{!-yQFCR*HU2q9K=`sh)n-1|eX7
zu%>Zzdist=KIhl03#Pa)gpE;qAfbAbXDwW}9=R?a10y4s=aGGZda3A|BFcWiq6adN
zkc#Tv(~oyrCMLAj>_`7 at eX!`%Ts}MAMM~M)Sb69`ie_P__r)%qZ$N<9osVKnQS^L#
zWP&amJnAmcd3zU^fZEyzCx at HwE3HOWhE(!Y)ys^rMa9JEqq=THbrDcf2Ag!pq@|^y
zp$O>ct_6v|LBqfl$sfY$9kQ&veperLlL$N6>p;VxR-A#0U&Z=QXJ}w(C{li)e9tT{
z2Dt4mq<#AINk~oMo&PnuFk!C~{lRQGiRJf|L|mPkCeQ{R{m-b3y)5NC^Y5`-mp;#k
zNgle0ioR)amzI_m)yYzp?(grHtD+z$mt>0$hMM}dzdr)i2NUDBSgchYhEjxxUS9Z-
z;Lf#VUa1)`;A9zfe(CaMrDxBcT`6u66h7M$16|;j(JOF5R6_=un4wW_N|99z^FfR%
zD=keM%6~pss8yZY)RY`XD<l)etU$K1>^wI$HRU+-8BLij+VT9v$sIxq8x>0ZKsmSG
zb=weK1Ro#2!lau_nS~@%k;Qy>;fJxEodU{yc_2$%yuzRj6J11DSU6KT4 at ytO`}fz-
zQsEYS!TVF0<xn-~vu!4-luS%+<*BlL`}#E%Dj)$W{*m`9q|Fdo(x}|*>`s>R(wG-8
z`O%oT*G1?&w_Yoq;9z6Zqbgqf=`>wkc0Q5}qqeAM$jz0K(iCWEX}RP>L`=LtK2~8t
zjF#Hhr)V&CwEYJgRj6GX*>tkPjxM5-BX7fvne_Fm6uO9og at wD5(~INN)BC4&<K1!h
z%&<^8HTEsYO}E9-J2kro2Z at nf*Y!T%$KQO)6qN>gh~tOuP}|VuUT4J$Nq?7 at msdUe
zl)SpE><vhtlM`=2VIkE}VNsEPXeeGmL4l%{)>RbCojVeWiUjbZ?*aF{dmzT6){lmV
zhe4 at Go1C?(6E~>m&{0r&Q<d|^%VajXVM1o-=lvf2JtGziz+d0GQ15nSKr<?wTUiMP
zJuPj!yENrHNJ~va(_e0;5E~blR#ipT)6?^VNrXnnu-}}g556ob;Vaa1beZY4V+GVO
zG}NzOzd{D+tSBJ&G`qUHuU@`<Sz<X~ulW^f<CkJZ<KFrNIjY3t(6+Hmr{0-&g&eie
zoAN#}k(xuZ0&BDB^iEH*(7VHF at 8*tIZ{EB?p`d6IOWgOT?Nn4$MBTf0PhKD?XuEfb
zYnayy#9(f2PMj_JQ`_mu-k<_DYHVz*Kc3erx1b<oWp#DA_t9S-{)o at v9^H<_*oldW
zEUc_Szz=zKb at 4ETT`etX{o-#huHc7 at Go(Svg-c|l)zt~??CssmQ<Y(5W&LqbZu)}~
zT_ia<xx#i*pw4}tjhBZfTsbfE9XaQ~{b^6JHh(;;FO@?T5bOP;rADCx<W at AGYL&_W
zTyY5rgGD;^0u7weGEXujLM`_{Scq-hR$+?<b{aBH0C7>((^HtApP$><kVlzAPg{4L
zqVOIH3FW1yr;j0(a){~F;?BZBrYc+4 at NiXsg&mVHa`_g$zbDV#_xdG62R_PVr9<^l
zvbVphp{W^JS;^bc**OqKBj_t&BP}5A;@l2xB^$3OD@&i5(bt5+Gc+<nOxXXjS4a|N
z4)rfIIe*BKm4V?6s&BXV(X=6MS63Ga9i76)#)h$}sf3!E8VL!B_{+r}UjFjJMkLCA
zO-{lP*0r4)7Jexoxo-6q;3nN}x=Fp>`8Jb6GKVvN0k(;W$<jzMUDMI;tFlP~&!BGV
zAca9f&vW}S*=<cktXg|}72x+mJd}8}-e)g<4h_-s@<um*FN~VaY*VLNcDC&3=r{u5
z);BX_c9vSZQ`~q!3G;TGni}<+bugqdA}9zuGBPrC$dc>TnX9U*Y9I|!TT<_p#=jS*
z>XnwH at EZTDqcX^Q{WWK2=cD}*Jta-e;H4#V5_0l)G|JW4{pMT0-*Fm_m6?n}HiKYw
z;0N&{GI`kb<;#`kzZZhCa&ib;WM*dv!v|EAmES;Bpl4;pu6Nm7D9+uOgI*n=w6(Rv
z<Ku&Y1Vbo!aW<R(3jegM95ryF@;F#~3<L`NgS at VkArVAFWQE8QZFphf?N=AaW}fH!
zBkHAwm;Ldm{EfbZ1<?$P6#~WrMRY=0;R06pN2M|Ghy*!?aS&8te!1|yr7Mb={$n?y
zRZ!3|MvC>w1q%X#WwKfk4ZJu%{WH_ at R!T;uwa#f(-^hs3*}Nbc8UHXS=>Z=;ker?E
z_Dadg`878`nsWVvmu$Y}qsm6wbi6<X#6I)(n($}#;)XZ7i at m%133mb?<+Q0^$;;1g
z`}8b-YdI(7Hz+|YuNB#p=dRp8lB)R;l)w+v&G2|<9=L|gv^P1_Wy<XeWDdee#L#1b
zSZEz2_WN5;V{UMAUW4#=-R>aY{;8tS(%LGzwk41S{WsWO8G?4B+b{l{r?P86b8&Hr
zU^KcGjId_<kmU=IiUQq6N<rr}9~i)7w*`uSiB;k#`^Ux#2>;RXVoF48Y#;<IJ#aPx
ztlQApT^%}F1{gi!#}*c>eAc5rnVC0 at jg3Edbzv(AAKy%N{f*t;-d?dnB~ULAD)AQ7
z34yqy?Jsl at adFt!u3eLj<GzfIjg58nDzTvWFd7UdQ(6A=V#C$p!Xp at x@R%4(5N at dX
z2!_6YxP#T&+6v;{j=%dFEEM=0MqXZ0NG*QPqr2a_yJtbRnUh`7x3{+gLqg`L9EFic
z8V*u at _6%Ea9)b57d$jEAY()MMlR+=$ufH|Zf`x_U6CE8rI?E at WKg96p5uI62;$wGr
zfq{X6 at Q4UBFn4BVW)FA1GG6DkBmo0~M at L6j+GZlrj#!d4n0%?HPusTteDR}k0shZa
zW^HI}M0oxZCMJ@>OOy5Yq5RB(7xVBq#eZMe)I<wTi-nD?71Sc)<HrC92}VIdDk!wM
z#l-=6d90ZD)Pc}zdJ9ZU%pLP=sy~ps<S<haU{&DFf50oXGBPvg^iLMeJ3K%x4leP^
zl>murPoF;hI`06wg6-tg{9H{<wE8I&ZeNLv$N!i+H at 63fxl>kV*fTRQcnwf<<L+Go
z#GYYciGV#LyncN&Pty4N5t4!7xkW{t&_pG5^}zpl!G;EHPK%!-BYzIoCqSuCeM>G{
zT3QBVNf0-nf_srM85xZ3`zx0 at IXVC1I)ywANY<Uoi~JzkaEXF0CGgq&E_ODh3A20K
zcHQW^Iozr9`aMdhz+9an{_v&{t_-ME|3|)58t_J(;k7sep<sw<^v92k^z>*#?t4g<
z|F*GodrHQk?qksAPr-3I|51;Cl=Lzc6&0A>8*FSi6IC`=Dh)8h>J6 at 3{`fTGpK9R0
z!PkV3QjaBqxQ5r!Qsol(2vIAa>ixn38yhD)FOC=e_%?0p^X7lBvasB^diCmf&5I1(
zF|!n}?2+V+Qp1j5wJ%mn$3HZ{zJN+8D=8^OIk_TDmMC-C)Sb-%`;>I<KG>a54%8PL
zAOAZqC)u*%;r;vMn+?18D8#|AaB?Cs0pf8975m6r7;35}shz(54Fo<xsj5C*s3-_s
zh9ran!=sZo#QL7=pb!1MW5_^ukx-%9#HV?=TCY>SE at STahi+xMV{a~B#X><(<ch7w
zD=aSgOnLu3KMiZVMa(37qgGpfXlMwt+YQ=?fs%zc9qG7u%}uS|WQ)dm8pH9}+?>Uj
zn*zQ7<4di_xOsSZu=aO$cj+Ufz&~CqRAc9O^l~AjZvlwUcV4rQ?BL)aT<Yf6-d;PB
zk<L&m at Dx`)JUrCZ)j#+4UIVTbd-iO|t68m5GZY*$$?WWGNoncpxHt;HDKG`^p&}U2
zR8$DZ*M~W}j5Y2%j!4F9O(g=~B2)(S?rs9Wg7TrbXSx)<977}QSu=J(CvI)c0l-uG
za#6mGbUbn6y3sWno);H%E;N&D(K5kdVYX%R+S=5B$7ynq2IwL=@<~K7w>8Rq!E#97
zB+{&N%quzjnWu8er=g+2*F>{0CR;vf+}Qi^<Ht~`vJNRVjSkkvZZ}QEeA8Mfue7QB
z89+anqri+pn7*-$R>++I#m36o&tCuPTmUUKS;(WxQiyloJ^tlVA3eIrJ+mGj^&Uo=
z&g&?|v9xZA|6D>trFdUF2%~<0e&xy)>#?%S0K at PFjNuQCM_E~U3o86o8>rSO)T`6=
z`vzkmcZA at f!ovv=Kd4 at mPfSI{&T$Ifv&3!J)YZeo2tpbuEizf65)zmaA=h6Q6mXoK
zPXBEx9;pL&f%?px;`t^rlE|Eg@<VjA0g%}Bd!`{J`0-kFHZL}J<1LFTy$~5j^;O&H
zHF8g4_4W5F6 at LfuKq+cyj4h1xX})^(3gtu0EPuVIZ^7})b&sV~(h)~+=GdrPOiU}$
z%gI!flx+Vr3JcZo<qI*PlAIi#hPrwm7)gt%1~+313+c+rN^x<7y-?;q;(1$l85tR|
zQCV48%lpq&RmD+Xr>C3yt<2Z&{P^-E0M!S)xeQ2?W7m8h<KW<cTl8^Z+`8zOqK3-;
z1d$+Ng!Ua~RH%fzfT^kJrPTgh70E1RLKOG?`*u!xHld|OMPaD$(9qCX=kZ_njMmZy
z<1>w*2`F6M8)_^hmwXT(Ve-64n;tdy`?sjLIPeY=%5iJvv5dSt6RHBTmVv?H#S41W
zhlB+7hk8s=aBVIwdSYVYNSUmg(K6Bu5m+zc<KlST at m#(Xot|z4$jiv!)YjLpM!%;K
zjFEZzG{JhX&z+cY8LS(+2xX{5vFzHQ6=kSDDqNL~m)gTBKsJPHn9S>B at A2kTBLZ&V
z!40OHyh^I7I=@CRayiUNeKqD at o2b6_;K2iy+qeC4a+p0mJ$)Ak|H^Y&@KC~Jf>&)^
zZg)}8HGelcs*bE<Zq2r-*EzBx;4nHiW&l@?g&=&}X#XoNw=V+n*+-CAa3f1_-uX}*
zF9|4&@6#Y5cnLHVSSifz2T)%x)oG}>^zsk(uje{%P8B|{vKqOA5@~*QPI<few5|t<
zMq*-&moHyJa^GNL!aP0PBrjTAT3G?ZjE3q-63lVI^VNK at o{?1t0v{L@)G}e$97^}<
z2o16i9YxJ=pZaVS12yOH0F_!(Q`5f#tjtI8++%BR?-J@)gPTe*$DKR)sFGLb?huM)
z^~Fh*_jP1-b#?#KJg{h}4cV?n7ZNU~6;;&dZqDWqI`1<=aFMxrc`er-o%z(hT;%(d
zkugzgGhTsBen*?&&H%>eSS}-ERTdIU!pX_eCXDV8?dnLeq=5nbPfK1Hf)~pJw-EMf
z3%J%?JVHT36B<f&KRq`Wi!v0Gl$3O|$xG0ulNbwf21P(bgpP9k6MA0}rtRd!o%*3G
z0St3PagC-GPz=ha`y=OS^p9jAs)dCG1idfLH@!2`)6wBNM}YjL7K2Q&2?>8Vy^o&;
zbU4p{BSnd*tH;jb*a(**xi#ZG;%NFKerT?eAWK;q)I?5x8)7-@<m5zRePM2n=hMrJ
zGq*GjcJ_RwvH9+g`@h}yhn{N*RUS<>Yq>Yd2`zk&WBh4l>qS6#>((s_jub8<i2&U}
zjcQvOv{ZOM6^GphAAgG1fkytR-FjW~#Lj#-Hp*tIVNCx4plocoBO5#WGAJh#R80f2
zpOb>*&<-$K+2FCtDk^TWv$Nyi<7a~sE#0D~0enOkX#~$(_PwnwqiZxWDvDKDI2n9Y
z*Z1!skZCL|EP)F>NfYL#T3XJ(yf2QK8 at F1o&i(l#1`%RxV<QK8um@!?3WQ{(vO+j8
zaJWE23|I(pXmH_dzpLcRVq#(eOi7<Uu|;+n^BA=HGOb!q)<r|LdkmqAT!c{~1X&)T
zESN}^Hf#08Mt+Cm*Q#GT=Z$x-vyKa-6Xb9?iPk7-<X7#>Qn0+gK*zh at 9dxRxs9*qD
zeuo*;^V*MY5%81?;OeYPMHnBZU#r at dRFI@)xKJw%%$WDZ8Ph7V6H$8SE3&%Lbj;Lr
zabi^}Mp&R-OJvu$Pv(BOF=<Sr!q^D}P(b5#w*AF0pzmv`(HG5vHbf1KB-*6m`mz20
z%YFtP^Z(XM-YNJ-li#)M8vRsq<XgJBq}30=R5EaL6M;KH`Ak&XwSD`B!~VR;Hz6TG
zJ*_N&04`P+=Q8a at 4l`=|OlL}{<e)QW<oc?q=I(kbM6B!I^TSXAyj5F381?%<K@;X6
z=CqjE*{{BS{aPY8(24xvfI$IO{DYTFfSR~5Af=G}FIo*~-T<ymB1ZdKrKU~E_xAQn
zhz2@{7u3%TT7@{N5E2sdzxeGtWsC;1(a`YF)y*x4gu)|sTPmQ`mJJ||iHRvRMmkFw
zmx9}u7=QIHbaFKtDQD)D;}YE<+bNnxG>BcAR4vwRBp at cvxIgZ8vYhh-T6G0^bEoO|
zSbFV2E-tReApP!pOY{z^YX+TQ5Ny|gx`-qM#C?Gx0;vf-Mv-MPs`ARpj+z&LP$&?R
zSRkjbgrxyVd825KWPnJm8J|AkKvX_^_AGQoRZWdAUf0sV04+|I>CDE^5FLmcXs)%h
z6BFT|?)kg- at 82I^)tAjuPRq{5thSr>|93$Pb-qrHVq;^QSzNq=+S=Z3nQHVfu(2tu
z{_*W!i!jlUltjb9!Exhbcj#{VsMw#hvqXHtS#UIfRFdH7z$@7AF38O)fklUKOoL`Y
z^Fa02`dncS;B&JxGblkp!M&BCt571B`ZEHy>dVXTK`i1y7K0i%nLjMB>a)r3=;#30
zB&RhoF>!Am`S~-osHjKZaM_%Xj}O_90{^hKwl)JchMAd}|G|S6V1c8F{(eHR5555b
zSG2XY^)0`BX_!>{C?_D3MMTc2EAH&fcSonr_w3A5yV0HW`1lyvO!M+;R2+eL94$Ru
z8za7R=Z?LzvwvY>VK{C<VIjDC_%Z;NMA_ThoBzxS4VBCA{xCj1E^)pTb{A+uQAZ~<
zI2h;W&!1R~x!=CYsH&>Y09lHnbKJh&0i{_{ON+h22rhsa95=U?i3zHysi~#u=0`AU
zSKyaYS)OVY`6?_VH4gJQ4liH&*&M(t$U5SS7hFX;^~2^9+`POt6~xURk{i%nD=Vu;
z!obi_Jh*{@kr9T7_vz)0MBRH_TrJaoFI+fu8?KX)k+n+wwXHLe$x at cG-QVAD4g3Ti
z68H at QgoT}*IC6FpduPK&n=QJvqvNuzt*!mbmv8Iq1+{f_X8-)b(9_e~TOPd92)^vI
zi%(G2ATtfPab%;=*!af1d&HnspP|-;r=$R4Uj>alIy%A<G)FcLJv^>@*VokWxV_Ko
z$%NqY^YeqUD)RX8>u`~)I5_Z;_uvdsV4q3R(a{MA2z)@P8F_dR*0y(YB0VN8 at fZTT
z2!{3Rw{M~_V=yF$i-pQCv+yGUwXx<oIXT&1?_yU|SJyf at sbeR(21Z#+YXU#C5^e}p
zmS$X1M~B9Csv!t&)p#jiOHEM`m8Fc03hy!&Mc>)qM;2VrF)+C7XG9t?M2DRm>oQ*+
z at 0es}Wz8OK|AEZEB%dVU7avav&UkNgnhwgzKNEJa{w_831`JJ(7VaP^&gHW5@`$9Q
z(5vKkZh#xt(9l4pkcScvaSODxVOv}FzJ7iNM_aS4fuT+fptgVDvq-m0O_{jw-FpG|
z%pI~+*VJsCp4J<Ey+Yg6)U<VQuweDdg`PpX-dXPY<&{@Z=@^)quRt20P>{E{*YBb*
zFmXi`6$wN|MG2^>f8D~rTQArj)syw~qy5}hLTX{p>v?&3ENpCVWSPYN-FPJ+BlCyW
z?5&O{Y>HX0^;`-BdjMREK|w)*bQiMX5>z^n&Gybt2B0OdGM<j3^0vQfUhnNWQa^e{
z{pCHKq`SMow{PF{EiKuoaq+1631D2X at bM*5guPrRY?wjAMU{KMeQO1-u?OL2A;AIm
z((g-s92giVA})RjSwU&?Iu&Jz2&Cbwf7c%+Z)0=&U~~GtPQ9~1Z?aI0^ExFpHT7LU
z%Jz;9dI14SdV2av!@U6|+v%pzM;AvJ&@X<6IV at z;wNULHSd9?CqbMdhQCC;~zxS$?
zA!G1q1boFF&X(r=&sl%~{8_+Fs80bwLD&#;ca6Uih>MHg0Of&koKbp*6^P~L=GOXJ
z#NEBl-oL%42Om%- at ZCExD9kX#NAbB|Ln0lKp#YzUxDqLOd4J792}egBCfQgYK-wt+
z0d**!VYEWQ7+3I-y}cU8Wg>`zy9TY7WaD|TfgSxJ6Gjc3DS at r{?PpNnpzOhXY1cVo
zJJeDmHv|IhOB3UA-O`^$7h3fM+UrSq^d3h4>|nwU*(VPU4xX8t^No$Y{?F3VJyYDX
z^Yae;%8BKYOqJ))1CZAJ=?ooh_I?D^+8IXkPCAOoU~RP2^OL0b$2&T$VKjmzK*?tp
zr<*0^<!vu}g#sXyjYdlh3<H>h- at HLb_ICxH*F<1~krh|ZBl}sb2i)9|lG8fc^73K!
zHzFb<-%#DRMV5eI)LZAeqChEIt~5*ZZAdS|Pe&{H#U)OwYCngEe{X-GlN1$2havhb
zIhfS9zdC{}2s*6}Bm4eahlgL_+xEayPNBWS8Dvd=xLEH!va6n>kRtJ(R>-%1+UUK|
z at Y37nwOLVdWN*I8ZaQpzqM8x(y{M=tmj49<vj3Qy%OZUKODUHIjK(;?FUSTV3I`7_
ztGMYT7@~DN!_D-jS#NSy^_2U&#Kc4?rl|hJ2QOI!1+~^n#-l~kQlUxxphMDZ(UA!W
zRE^u6R3GBwWh^T-*S2D%?&vkiLgMBawD}XFpmb!t7d{EZ!NF<AUr*Z=k(SKp?}Mt8
zXE)shZp-5AXdBE~%Rz8x5EMUugtt~kis=A6;-R at oLLO>sCA1Je_&?(A1;8XB&T~-V
zkzDh|2|F`o_99e09`6fJD0z%|sxojZmi4K|gbK~V(KI2F4Z-D`%*=zSkKanMMGq8c
zl+)HR8agkg8bZABK-hre<MB9HyW=wVF(yVrC0mvrt^tO44+wb&E3q<IvoKrF`#c)d
zNs=hY7hJJLn?GKra$Y-r;Zv)Z1FURp>Dzz45ThvX+ftjDnmUVX7ZCSbR#t$0-`U=#
z2b$KbFel{Y;R&QsE0<4xC=FrFgyMmGI)TZS&gc)MAzMxFNqiuNEIz2HsO;?Qq{F=L
zeq9u<WL>_=#1xs4!MMLZ at e!h5faT_}22}dNfvl%p- at XMiFfuN~{~88lF2aEFxNK-6
z{c&7U6mVP$AEhr1kb3n>_<Y^Yo4BetBm}oqY{fw{+7S$qMP~@b%?PQb=;yj=@$vCt
zVegy_p%jzQ(#mtkh45LA-f-rB144}4ySA=w;BeZzS2DezE%BjSWO at cN?GS{GfIm)#
z0Y6PDVjVwy`UL752C-reKDRS4$O(-tI(W)$ktR~=W`*amgOX8RetyP-mkY>_)rg*V
z5($o9?smz><NTok2+6NqT|uH^VuJHuo9|ijEr0m<QN`Sx1xidmXl4j0 at v11Y!Vfk*
z?Ra<b-GH}}5+NS#!`QX;bqkou^!C0Xv{dJlJ at EimcJ>TN(5R>=Vu&F6sq7dkmB)|K
z`~m`^8XJYqPgaX1Hk<P;c4!7HD|ud>x#HmBW`g1wb+*^VLZU>~);{n8MwVEwE-4B3
z^zvfm;rRfe6*_*{Sl1d>&q_n+;Zc2jw%dEvW@^$|r8`tEPjvt+N7t`kiLZ|5$(0P3
z*y23{$Zl%_IXOFmS{Ze=)#XOBKaZ4U$^c@%p4AK at N%_$A?h^?KWfPMur4ws>=A<y2
z!IqZSs6L=aDxG{Q%Ubm*k4+))Lyl)hFG|IhD{!YU7k_uWL$pI#nJC%ycTiX at DZ0lJ
za<&Bum8hnsX27)QTw(Q8E0~#xm>5hmvqk*%d|t=mk<ZO4IU|In)h-Ej2Njw&88-QO
z^M>)fR%B4!{Gm--h0VBJAA+)F<3xLfPYVcNrZhqR4z>poca4w7Tie)J9ITBwo*vk|
zl~k-V!SQ>e>%M%;!O`(1F)?x4S at ra*e})`bC at wx82^pc2r3_4FiKX#R1qGRL_sk*z
zJRdxG&>wm0X<*<I<kH^vd%fY6mAuF=JvJM2TzrRUZGI_ymuq}w#LCIZ?BU^oq=#Td
z%J=WG-1qP2K-M^$KPvR|kl+0LIWHNExn_eam!GdM_<mA|TqV%#+d6d>{Q-EpJ3Cm&
zb}H|Qc0V-Bk53^*_%_FgIghcKnIsDExL8S>iqAR~GyzzI5l9RvQP2r3J7e=D?@!3S
zH4?EnSFbYGYRiL-om*Lvfzpr$I#uEB#WS)vY4W3TNSV$MLYmRr+Z!@GVl>TB#(h%}
z#5688R?*V()AenF{-j<_0In4qSz=~+YrbVrwY)LBv#9_`0T3Pv>(z}W$M>0;H{n);
zS{Fc~)otVEPyZPO_v87GtbBY`VyCfOMwoK&NwHJ&cQMXBGFeClCKM2plaH`$1055h
z1e{hrlvsB~-hz~tVu<L2lHfO=BcH^8nwy`$#W4tx?zHlg=@eKp9ZYuPk=B2fJ~xsb
z!H*YeNea2|rH+qlLYz77O{G;@jYNiqhUOk-x~~>iW1#^0l{7Ueq421{x8su&z68gI
zLPE<kEpJuz^rmw?!E0BT_0qy at ->lS>8!6Vyfp{lE1qTM+9aRH}F)=e!0c<%x-}Gii
z85<cH*&m0~i;Xvio(9$Cjo8V8e%>G^j>1o<Dp1b*+S%z3x;9O}K3%-v><UJf2{kt}
zBcfGhP46fA7^08QZb}wR*`-t^hBz1;v{cTfBT;oe%bzTiJmv&`{{90XO at 8yphAmnu
zgda|~TCuJi>g&WrLY2*gM09jC6mS+s#<$1{1B&topLKW7IFwUVA9ISQC?Hv8zd5oI
zkDRkG>I|`PN+^^QxIMHf8?;~xX296eQby2agGBIyyyWJj`{{Z$>%)iIA?laTPxcv6
z72ba at B)SjRCq~vVYdsEa->%;R0fJP at dtR&yA01!H{W>r(ps&TOkW3OYS at R-eY~G=^
zVRE7-B{sqRyb~!I=zRv{%H?HcSg0gHmm;z8X1yQlXa8N1wHdDn@)2|~9XbBL*JS&X
zq=~`7AVL^_-;P#8QxiwBZ+m<DlFx8~2IIp=03;|yP0e^+?<Y^5T=D at 3(jq-=^z`(^
zLWR-@#^`#VJDmVBu`)B~$gO5%?F7&-6A)g$bgA#r#rCz*wj~IQLO|4kuMvzLflu at e
z3?8HQ{Ihv^d4Jf3`)_;*kBw#dK5h5IUb}j2Dnj>Z_e6Q<X=v at eU##gB6(3NJqswDP
zU&07ez0XChp)26Ruu;#RKVOYzCBZ=#QBzZEEd#GYfI^mK8PX@~U5HTvu3I-?<cRIf
zs>lc<-o3kuX3k at x{r_65rLs{0b1&C9_3I!eGE#J7W5azOZj}W?tXB9w<{Mp1CNTwt
zv|ZzV%&38*fV*DQNV%E9hv?|Z`kJSA_p-F;y!HqKv5*b=JGu>lb#(#+gsN=M%55Nx
zD=geRs$Mj6x%hL^)W7(H3H=D+e*XM93OO4SP%%yMkKyw;lmHWtcx12BgO>4eN=C`B
zb{Hr6sIC&bY2mrKxzvgZlH}y%t?g~XWQD&8T-p8Rsqi&qnKooe_2Z_Mxp__TsG*Tj
z1Q at -xfNMse>8=ZLX8LAk0dRFgfP-rXtuGd>vgycx)LIX|Z8+GGjaOQcLmPSO`pm8C
z2CeGG?h2e9ZrTkh5fDbi#ZjR3)!$dvny~ENxnh&iXh_!PGVLV^ibUfs*9Un(#;N<u
zV14 at 9Bui^wQh`&>NMvN>-nT02v2+*r&_MH(z2!?jP3H$%D`4S-E{^_)j at 7*Q1Ey4p
zAMay$c;8YjSWg>)%g3n`3bwX)&~C)ZGI9K~DA_gdfS!8TvJ2imz%Vm|s&EJI@&0n7
z7L+Y&8XAlB at k$2YmtPX>nphKEgaJ_VRM`aJ+wtjXwOh{W!h!_Ke74n3RZUHzva+%Q
zW>TX3kLF+uSCqU*2Z#R`T8|K2*I7UO^*?%lPspJ#Iep^PUmYuFMoR?_*4R8%`~NyG
z<9F@;lu(3E;L?O$^DR_s_4cca<3-W2U{ZE06rxqo*VB!9iC&#@Q#>>mrkDb)>aZf+
zMj23oOjWjJklno>_ow?x4BC88Y{p)tr>C0(1*)p6OIcRNf9vk{bKMYrGWzS}27`fU
z+r8e0Vn_{9Wn+EtAg-aYkuEs%p54?#s7>Jzf-;!Xh29stDI_E$H^(`02L_Z^hVnm7
zx&6T=W+Ff#s}`!wwRm<g2$;P}j}CvmOyXOvc>cZ988`s^Z$?2u0seZ<r%%CPmCVP=
zOlZaxw7$plcF%Z7I4+9ZVPpHI{{R&n8ai~U*WJ}+p}Kf>yek at K%k8&euvA13fNvRf
zK3U&bVRC9w)CfFQVfjBx$s<30t$vec78Fc?{5;0&lDm6k3;DUTwUs)e>+xfB8w-M5
z{Lh#7%yF_3RaI4cADhpfTrU$46ukdw;+KDpTK+QK-=iyTuCCz_fcwjX0c3ZyrS|vt
z|I-6i4Grmwi;KPA?>H&5np0k#1bzShUEs$^#&b2bfo{%bWFw$;xxapz`2TN3n#aw4
zad!4GD%WLmN-hR&#6zLt;J^(9 at E`byi;DmnxP19?KL8ok4{NLLgWbgx?}4kF4aiwL
zq;3xn4<nEZvEr5yVfJ+O?NU!5$Bi34&=TaL751~DmgeS51JXGS4T(@CEygP>3<Cx%
zc#mz*p*|Wzt+9{KQFqM26F6xLD6px}ye)J*7X_1P{FB1b`|s&ALQME{!qyeS6jt8~
zwW`g*%k$XJh$s~Yxj6rzWhhWDee?atk0;U(6&`M1ItR*uNXaWJi{!KY#o`&h0RzF~
z{_y+VXqhJ}Ir6u(w6wnU^zd0uQ-=DR_9n|gk*{!=m&rKzF*+%q$p0AnMM6LzGE$^N
zESM-?*zscd&E>1$b@)ykE4X-^ruaEHWPd9BD*syyserf`e0=;pNP{4vw`%$S7&BQ}
zS##(m=A)}^>XMp`?z|u#0{|UM00xEugUgjG#!KXSo{yFDK0xYS&cLJcKHvZ62-m)?
z+Sj(>o>?e^B-GThU@#pI)-+1n(&pxj9h{w+9zA*lRqEsDY>Dj+X667Ug=AT{2n5CG
z+ at AfgdVcX|o9U(`_k%TQ=tcSz?uVeKg=&L{1}`iukkHV`!BiRs+yrBhZI<E{<0F_7
z-t4-aju at M8@SvBkl7;*EUA~No=^Wq1q4Ojm+v`J2%mA^%L!_+>-{aaA at 6^YA_&`F*
zZH$F<s?4NYU_C-dMg|8MAreR*ag@#V$(8RWs_h1M=DWLwhN2Nm(Gg7gv#;-lv${@?
z*AHHx8%8&G_ekwKImqe0whfl=0s#dD9J5x$V57t!?7#%2`}<=eg9jxkLX|CxRV7=|
z-Mwzqp!7~|W~MK~@@3W4%v4lV?z=zkBj-^K1BOf{4=)`8b|FXG#wI4BK;+JMQ~qXB
z-M6hUvz9fcprGKf8RvnTkoEX2W`DAfM>~Gr4<1Uq^P at lSNeBraqgdG3f`^NAy06F-
zXh4AAQ}c at qX3H@?efkuXH0}MPi&!=`Hn6JQ9y1EhCL7$M$T-wR0QPRcJL2z|$p?5d
zBso!sN{BU|?Y!H{GJ1RES~l1|=W!Jan+T9o<YJJ1k=^q=R>K at Ja&lfQx>QzlA3621
zuL&Rd00c?{Zw3-Fn(wa+xgS0 at J}jOT{-&j(B9^aKnC|0)0{2HkO8NwT8aTDo9)6%x
zWB(c%jQ^ZejAD{Y7Yo1-qEQO~;L3cKuhtJm77JBdUq6Uc1dqdw`Q at X@zOiyMi~jT{
zBxGa~;KC91e0xp!uB!+A!OTr at m&wfr8UjM(Sp}!nVS|Y(8*_8>&zaIucfC#=A3}LA
zDJ!$=-5A9{*2=1ECxh{*_-+UZ(M(NENl8h4{vLO4c6F8A*;NABCOq5e=0x at bfgf&s
z?B0ffHX6v3M$RNboU%Fx79_Do%TNoslhiw}w;rrdL?k3!wX?H(xZO at 1`_PRW(40Pc
z#cTDd74v$s<A$1!&h53I;M{?jUiW868vSk$LJsn~!*D6-lrtp?IN<<Fw)Z3nzWCja
zjqK}w`@o72KWNO22&r~_C7!ITtkF`VE67R?4AtX2RffBF2{I%?MXjwl02*dzW^Opg
z7KBn#QijLG_(Qm7W at gUxKKd)}<t2oPN9ija#WWkloGfB)o;zybOhpq9%~DhI7xFmJ
zU+hU5t+t~h!MWVp+Dbr99sr(5{K=D+&d$#3X^!j=%TjW39kXo#5s8T*;F6C{P7IMA
z at V{Vy_UW6NGC9X4g(>H?ml}NukBoc+?W2c*wXCEBtJIFH%>8HG at b29bIlZ+n0=nSa
z!{{D`?d>^c=HxKmzKsV6iJXdpzR<eO5w^FtZ?UM|lz1xOqLQbIEL$FIOuhpqLKcC;
zrEXpazK1?@IW8&s`uVkVcXyAbIaz>AfBp8YHHet`pS9cv4ws<iC8ed0RaHrlcZ1Ep
zgoznUWBE}<Mdj`Ptcg0WUFYWJwg*@Hb>0E4M$f>24)HIcuYcp at aPu1Y!cwtUON4|w
zySpGOe&uGpI6(_g&+hW_c6@$~K}bOn6u~ITaOV!bt*xzL!1o)<1GOXT6V)%E4Vl^5
z47|Lg;KAN#7W$W$-$TZimNu-U<Sw!n4du?$7(2D9ij167w>?iK#~vo%*w}bxZLD08
zEgD&91b`$YCVmSOf=m1GI+#$yfHTE@%gP$pZH9rtLBFI)$O$Dqy=%~=qvK<Pna{7W
zaBxHw6$#MM(fe`~QslRzf9r&57D^Zx&^tIdfFuVYhuT0H`Mob50vpjsNX<aphX at DY
zQoMosvc3OD{*%hhBo7Gi#0M`i|Fg^uBJ&zPC|0P}*4m2x?;$pVFt*c!^{WaBy#X7Z
z^=kP~q@}TuEr+C}P-Mfy#Drc at kP3?T4TjP2aYN+r-_@(cL=`D1deu#QjAWIXp+Nh{
zrU*Fquiw9mLm)vDQ`Pd(P#_{(`}=;y#hk#P$fnBC(NQWWv;B(~%T<h-fJ+}fd_WFi
z*4Eb6cph`ZZC}iPBLx9YMV>;KnVo%|ntB8G--UMM2qHQLh5>lb9d`Y~A=aocdM76*
zWB~4H*W6h1K^}|h{>XWiDzmWgc0)tM%>4Xi$Yxh}_tv2yQsmTGem;9*Qqt_&+Ar23
z1AtnD<sl<$+;*9e&63tuQMff0vOkln%7*euN}@r=zuDV+)(i=xz2&xhHUbZ#W&EJW
zJnl=*xo&Q5xy8lgD|IzBso)5=wziOa{F<Dc{rwx=%*>2mK%lj&OZsRmPErm68#z0s
zp+SK_=Cs|fn%7n%MQJrP*TG{;LN3Y3$oQ^nGemrDYZK$-Iz2gQ?d`n=)|jE0`yjPW
zcECbLOblaZXUCUjA2|h?lEQhWqpUozltPK3rly{OJ0geY#>ODyF&%2D;S0E~q_Wa4
zC<vQ_gM;$iNJ~?bkFspK$%~3IRN|<<tc*z97fce2#0*rw+)tlGot$_rEiF?OS$r?4
zJbymI!UV|T>gtNsHPGGt^k^(cfv%>m&bg1$*l+;C_tm)okacimBoXo$-oKl6ot0K2
z$b)&H-suGevYWLrF)^a(Wo2b=U{(qW3k?hm(2x^n8X7GF14REGM+=w9f;1Vl`m<r9
znbD85C3NNI&=8bI^qPB&&+=4%Bh4!;^i$3gH!-=jva+K1{5kW at SdR%5%F4>Brl#ie
zj~{qo7ww&#`~m{5Kt=YOYV^oJretU5ji28Wj{?nMX{Y1NK}uwYSb_u?`JY*zZo2)@
zG3fC}7GOV1D=Q@(olwyJug2WScBZ5;3S@&SG~U|UTHn$V2Xt0lORKG?2OkDDynDW+
zs>&Zah3s92)F6sN-Jz+e8C+_qP7Y=eFO1sp(dx<y3?{NgdUE2f`s^8UN)8}8H#Zk$
zzeo}z&R|jTW2#;bqfb!I2E12dQyE+{E9=(U+L~|X0;mCX>ZBm$OGihNSBu82KDLpL
zwCw?OuL1$Mu_50K16oJD>g{LIAu*6M$sx2tnrrsz_dCfkG{d8!25p3sdTDj1z5fb<
zb=m&P_&$&ZXLW&pxnky9H2c=!A-N!K=SG5NVW&SH)m@`b++ at dA@h at 8XS6wsP>gqsL
zkOV0={Xry1(#gNf0~k$BN=o{0z(8jlYA14b5INT?W)Ko3|1$X9J2VtHT(Sp5og2Nt
z9b2o#&A&OlkKc`^-Np2Yk}rp_NAr0U-dR at wueN^shNF@#O9)p+o^MivX+ at p)-o88h
z%=7R0q3cHN;*qd%!Qf2hU`TK<4n)g#z_|Ly56Q`4FBf~Z^LU5z;8POtsrL@*ANS7>
zrx{sTup%NNU_ at f!X7>P<_?JuW<P{a+YgJj#S_~8n&SnncM$2?WVZn7ieDDQU`yR_B
zwx)R9$~Q{>DhZC?DE-}GQ$6o<-fY>pR=uw80yOdhw`We%%8Hm688McZmtE(hpSR=t
zaS0_lUibKloFbv8r at zi+=nGxU4nQ7jizqAOKHF-$7Axpd1iv&z&T!B&K%V;n+-A!s
zu{pDZP_#?uPyiH%HigBP_Q)2h{b~Mt&T0BR7OhtMjx+ziuglBJV>Txwm-d*i4i|py
z?rzI3ZbClH8w8z$SnHAwaH3#WEBK5YHU&0^StyXsHoVdox4+bn0itbUVq)*$KqN at r
z8Sl6@%0(^cjO%@IoCA#-=Ad2X5&LNFsQ2Qtx>=>;vNA4>013!FDNIIu2i&|);pHwX
z-*B8xlP3kx^gj+C?p+xO_jY7Vg1Q~SXi#8*9FT;f_%<imT?QFOLcKu?BzVEhhJXCX
zjC9ozDnWrpIcAF20Y#>=u)lE<on6*#P at T^m at 5npw-*UZ;xuX+IE99<!ur>yBr=Gth
zm2(+{36z8cf^!dMY6foK0zhv72?E1YiSyo3li>RGKv1XBHuFpfN^ETGrT&a7naZp;
ziLg;K;B{gJoY)c*6M?$+Xbm+>^wA)|nJjVrjm^xAwtshA7jQJsa2g{6Y3*!q+a(|)
z!>rZDzk2m5sPAm_bKS=V25Bve$gt3PZna=^G+B>vLoNU9N3M(Vk>3`k(lVPYLd|^C
zdxc;LQXY(q-sa}ASWi?H4)u~-)vwsM$2&nnqKnMW&&v&?iF9g51I9J<ElzqpI$bMw
z at 7^qA^nddPy{x1p71AM?OiX3mNQk$(hlZMZ2`JXR_rHgf`@K4#(2H`%g*2V*JZ<E!
zd)L#T<8?4*CTB|wAz=QD%Fz0GXP;M9-zl<;fm#A*V^<bd`gaU2hJ{+6tp8Y{Sy=LG
zYAP9}2o~FdhqC{7{RqP3V4+pDwAeLq<*;#aGp{{53+KLfZ_`5vkCG=ahEq=-peH>(
zo)QQn<NWk6s=4_QT53%Vf8(O+^XF0E&l^ux@()Q5wd at +7064)fnK_70z4)qn>U4B!
z9sIx>W20_sRs=%s=Ya`Fc2ZDZfBi}XrPs2S3|aqny{?z+1O=+R4CTVM?APeuwYrN9
zdHzC;JN^)QQL;=KT~W+b9CZBlH*V`T#Ny!Kgiyyd9q1~U^(4}x08dOxJX?LSZ+`4X
z9tM(I2TzS`&TuvzrT6b_d~nnpFVf6NP4yw;)RhI0wf^!o5b;eY#3s9&Af5=dcX7Ho
z?Q%*}sY7uULf9`&EZ{f96K3~@aKO}0UXsb_p`oFpiRlYWW at hHt`1qCR`$J7dYg6uZ
zRs=t8+jaf-9QADCHiVPnvuC${D&-Fy?F}jr($eCCEXT&hwSYuzZ*NP<%iC6+|M$Q&
zW_NFrU;uJ}4J=GxSXk$$XZf`8f+B$D|2!s#8}p^JGrg{k60C)_O8r&xJH81CR2K);
z%@iEZSx91qJorE?koW_yO$4kQx`OO)gBkJxC4>vi_!Nw*{Lfi{xjA3tN!-5|=kDuy
zLzZWYkN!S(b>*LPcwo#;0eDT-QK+0(V%$aK>+dhh6!n-Xik^puq&w!e82poQ;5d;D
zV$te%6?rrRpc*+F0>#Z{NWk2p-crWR?SWdMrr*+%xvPiApCp&5mfDw#<3A;yf at wmb
z?wNk?V;{UN=y|l|RXwdV5#1k{v|slIF}7!CXSgwGq$7%s-U`UR%Fe>LU0(5m;oYB2
zYWZ2f)O^PcK!Tq>ePUop2UeawtV~}X%ppbz*w6U1?9d20_m%ewKLsl!>S1bX%Jbs4
zxQwi<_4d!M^Mdy=<CdJr)8A;Rl5}1_^^akCu~0t}ACwQ-Bnx}RL)AEkiE=lQvoS+`
zZEJh`?c2A#m8nzUk_wv%zO-bStaKPNo_gVEnOjdJCGTr6-?<Z}$`-9V4Rr@|mHF1K
z${Jb-vkf&Zt*1djLCD=9i-=Dmx{UYM%6pNg;NrDE9+vAsX?}}Rl$H+Ctq0~~M*RdI
zLiWd2MiSB&iasfFjvQU2qoV_}TWefn`J2#x_hnWVso1POz0!&))@_VOg;!VKhY$*7
zV`rDnQdW|Xzzn_4^#+9;;tmS*_1)@XPJ!ECL!!ewGLi&ckTsxq&q;BYqBxZK_FkN5
zTI&Al$e_`CAqgfXCeiPFe0(3m!$rWN->TG{oU>ZBmPXbZk*8e2qa#aTB|QM0*eJZ~
zcT<Pf!)S%#{jbp#x!^=E{r$4Jy86_@!UB1U at JkpCBQOEF$nL at q;*~I}`(aSm#gI)J
zwftoOL}a-OU4)R3P_%Kr`{UKkww0m$WvJK4XC>d$lFHj4N$w*^UR`ZHCW_<t2JEcO
zRD(uMPYvi)BY$IK<1+X)WFZ%>wg>Y at jG}ztVEk at 1Eo)t-P;C&fkCmSO^_)X(Urt^g
ztI)xyQN^oscV!k5xvIZ6CZg|@P9~W<sj`uxxfHGbdLK4k6atFM&cnkK4!B>aQ$MWl
zW);=5F<C!xb`~Zh07!AkXJBApZ-tzcln}LZc*x8^<gAJ8N+28G$g=vkt}d=K!_BnQ
zfRWI6&5VkQ56I(&^b(VQ9*5y;Wi&?0gC(l1t0O|WA5OX?6g-D2xz?v;)SE1XY}+GG
zhpHF7^9C$;@5K;-GD{xw|JY?R7Xos}*%uWR#S$sM*Z#k59&P;{uRpDc-4Vi{Nr|&?
zkGKBRojx8|?}jxtD&r89GfJY?l|5SUrG(sNbFy)jl+%{A;MT{$Csd+=JBs4sawbA+
zHpFE0IP+<d$=PvTn-{Y^+8L7h%d&q?Up4luRtheToHdyx#%&&-_Z%i-+!Yk87kH7K
zm6ccVY|Z7z at x%D>Vgcv1+%h?Qe0&ZDhBx4|a=Q*+X|2cztc(|HQr${2MUH;#?(Y}6
z=)8=QH!+!~`rB9+8J3ntkMa2uIvO$7rI6mmd%`Kbt`C#T8WmZLL%3q%<G+D+tWr<b
z*4N)ZzO}G>jO!!AtD^#>(9UQ3>nk-4&4l(q*VnK5;HGg=3A|Qnjf_Noz;u{C+s9{2
z)6egGEN}8U<+iii#GGLH?d>H47DV}0+}cF7!u)h?DWgjVOLeKz>!v1Q48_f$)twy&
z%-rNXwLSW%F3Mm2**ukcjmeI1n=z>FTDO*SjojsQPD#mGmU#xNY$jq$OSz(?ZaNL~
z-d}E5c8`<IuFBk9dGKN3><G+*^Tw|SRqp$i3xEE!g at lGupcdxmhi06)4*oI at Jb&u3
zs-mmgV`yj?TVG%QEuuOif-nJaMbK?0vs0lM=9Mk#cx|kl>fAtoZNAojfi}W+wB$A5
zXDJBc{n6}NX-O{LO8Z$+=iLRl`R35yEDe>B@<tPVAF>bF<77MKlOJ{gYHFIf|CpRi
z>S%AzwQ;ur-;SFr_}e(7bZ*WVlgAE6>2cD3y7TAD`#6ys`cS!?hFQZXsRULmSKehF
zDL1@&JaW9dI27ZEiBJ6%RRTPP`N#76WW5W;){e2Y5y>NakemG?1t}^0rlzKEoTOV0
zzVoxQEr^a#)3BM#42;#j%;DDPNw{C!MTgm`8?%Nk*V{-=LQ>+qK3+OAGb1s<`rFRZ
zU5 at K~_vG3E)cF-pOlN0jT+~FlnZnz1w(OA=c2r<!sENq=pjD`im<mLj?lTRIF1STp
zK!9YDC_2<eGR~XkU6B?}8_%`VGIuedFzMPo2{dH!Ug}G8Uwn~p!oVv3>-6ZT1dKgX
zgp|6B49 at C!W&U_%4^-ysTV*x$0s`^#i;MSQ at Bkp|!5F>^4mJb?58>+6*49>+mBnRe
zVbOP}O^p8Vp=7i9Z&MGUwHL`V3A5hhLYGDcKHB$R{BbE-85kMI26GgsycC>{Go;P-
zSBAbd>(tmAl>aTNeI}u+tNXpPGkZ$uAfJS552OItg)2|>8)!vbWhF1hg`QMj)>Cp!
zpUYRSl>h#1sP^nx2N4m`hsCS9=eSL$8-kcV6jW5 at Aa*6lt12pO?S?O41n%9t=M1Q+
zrliyYu4fdY{sTGp$aAT_Q9homyz@`=3Ms-SP{*vlhSRgDgxp>pdM73(D!S6hzt at u_
zDA)A9?vTAmalj2jtJ>B?LPBB<CY`MqKGfKsk&^Aa2GsQP^bnZb@~J=wSCxLt8&jGV
z-l|u_1sb8ROaWnVQCjuRHjS$`Ha25GFZ}iVdvcNzq30evRSzMkdpua#+731*C-gl7
zuF=_6DCaftLsC7@`TP87_S?5_S3~<pMyh)bK*{)k4J01X1&O!9uv1$9Sy}m{TslT1
zodex?05$mw;6+vp;bWuf7L96Kouq at pM!gq7kv&u_?CeGi5rH;iWl!e!VOB~3Jh<7}
z*`sIcMPD{a2|BOogPk=NO$!I4E}NOruS_0Pl9HA-h+>j61TQAbLc)!Kfiaw;K*wh{
z_0Ys5L7l at CfJJ|@&dEqXKtT2;QR!=pD`g;>BOb}~&tU4l_w|J%yH?|2?AlDovlUE4
zL6i at xx6Y1Vni?3iWGkdxNK7cJsda)kj|&RICJqw+-rfClzKR{<=(b){qK3Nq-899o
z1%-tg_V#!4^YZdsP60-=%1!ZBM~dkNM at Gz`OCTJdWS>gkBm%I_wl>n&{|v(SrM0bX
z`h at 89iTTN^j?r!mOfxexxP7eDO~am^p0%ynw*4xbALqeV9~`F?_4RL*S643zgz%Yc
z{Hld!D;n8#w_NTJDtq(h4W;!x?+ef2G_fl^*;KqV>(8Bd&6<Bx-b%XvAt~txs^sTS
z)gm#Adn7Ix7u#RxYVn^KogMvoFRP$X+(JAHYH#$A_I*cZXO)Ao%@yRsm3kKFv_>JX
z8wl^;zrVDy!obQJw5m8BV%L052J_;l&aogs5<0WA6iCP at F@ya50ZC5R=!B8K=m>ep
zTwGG}C4tVw<J0I*od at i4kEIu{0Wf#|=b50#Z!urc;P*3BR#skr;Bb>x*z=!-5Qtgr
z0CA~z>6yD%9taAGDJWcnTg?lQq(1 at 4ECJ*Q>@dW=e%EK!9?Rmzp84BH=Ii5?{4N{R
zkpG=dEaFeK8(awy$ORO>MV(%0p1Dir>FH^|zw&HuZcao`FJ;wUipu%#g%>|R|L3VG
zT0j+>;qq`Si+3!`4SrXy(ITsVM?2k|rEs&NHhLRX4GjtiVHAq|j<yeI#TP=tbYY&x
z12RxGLL#C!z*3ByobQp#g8#be?d^@Mpn>Fnun!#0i0IM!R9NVjk-><Ii;J8jjwhqH
zGoEU2L;L~Y at 9fOXXD5E&T$)hFsYsC<BH%%}xw#1_D7u{ZGx5e6-0oFbkG%m5m*JmF
zdMurD8Cj=>{<Zz?c!!+1Mp}rh(v#iM4uo)!jpqqMhK84?aai=a(a~I#0{Hg;_UA>P
z+uMCt!vp<gz)b*hB2Vj9T8%i+;L{fdN6KGiyK at J57}DCtCL%UgYSBfC225EC@(?4y
zIU_fB<N at JMPug-NnX##<KiI}ZVK2>$O)v2ps77f29+o!(Q2b#3RWL|Gz_P at wm-f=7
zOZuk=>qlq1y|wrO<~~Rxz&L at q41D+Qt;mfTft#ZIX>CrZ at vptT*Dhba9Q*IXEi=o@
zT~1QIVj3D6$iZN0YHH-X2GV~CMjBq^*k*cq`VwV`%P3SgF56%HHFb5rfg~O4Z|?MD
zWM)W&a2=hV{(%|b_d5B%TDa~&s at pGqjqH_5GO}JGD-mzVxOmNLMzRuRMcJz)T&~TP
z8!da3n^hDcn~acKq(n#=$qdQ(o#*$ye^%e_z0dhR=QEz?oTKn#VWE~St||}p-=iBN
zrK^0v*MbL`;BPtl%!DF%mV;G)e5j<6k&!_&vIZ#>^nBXpW-uzq1~x=<mqhunyM(ml
zd$W)RGBRwa2D+|J#oWRIo_#a|a5k~9kaYf;%SBkQKN}(SoavzZJ}|K#>Vjm&j50BR
zDWSaqyg*q?tNrRqHEL2)EC9osf5!KclX#WP&npO(=N3>5KtutbqEn#YkdTYRW%iH_
zySK_;Z7~6tip$7Yh1knvpeixca)^cwz&lSC{!9CrS(m@$<m&GaF9Rf%1Lp9cMEQb#
zjyA_1lALj?NqXJe+t#?Za}9F~Y(J_vKbk^_k7wH4+!U0SW+kxkM`mOk?j&;N)jkSM
z5<ypJpjza>{QMZLy=~xr>VdX!L}{rstdqXhj!i8)rJ~}*=I`Ilt*tcC(b4F(E=Mz}
zhK<CL^p8G4>wDkBR*3#vK;X9a_FRi%V^{?bvZC7B+8Qz$k{3WIZc?&$X)P=)P(3~%
z^VA6}_9LBvG9N1FQ9X0U?ELvSV5_E)k-PJ$udhH(e3_h#JS^+*s9uJHA*Z;Qum_HJ
zW~@O)_wD|{=cYnbouTVL;!s29m!=4AMNe*|Ed%tv9s<L^w7N>(sJHd<k_K!1tv{Px
zIZ&h^M10l_<LvB=?xj#d<>@4#$wwt6$IDuUF}uI}!zXul{yZuz1zBJu)(VQI!sX39
ztK`Jg=W#{D3cZjQa+kXg`p*ndf!AukMMYD?rjNeh-x&ju at ImQ!?i?QmvNAgDd*b9t
zbB8ywoN<DjagI(-)UfhcZQ<&U;N?jj7JlWbk^D?F0cK5H2$z%Q&MNQwRnM+2kac%=
z*W1?@wFE$dTQ9B|>*(x*-Cq&B?8GMQm1!r_369Wven=V}6rjE(O--tm)z#%M%{1uL
zzP^6D^v_!gB?|nLvi}SQX!OQXGktd<{NkC<bsGZ4Mqub5JT%$;*SyndFxXs<lXqD#
zOB54zvq7N(#&mxLbe9jNAGKXH?krqG7+z*yH&E%UOA1uqUZS}pbv_b6R7grH@%DUJ
zcTW$xZi4q9bnF;Cx(;!0IO^fyfx(~|CTg2sXi$qtb2Ak%^@&rbEM>s6a>Qw$P*kML
z-t&FF(XcUQF{b<tGP_$~2bBETbLS*=_2Q?yZXZL-Gd5;J_mlwB8e+F2I;V24442JN
zirLYn)b;Dv(Iw5`;M1;!t#9|QEdh5bGZH$dGzB?e*X0K-j-jjIjue5En>$<8UpozA
zhw^4mu^!_}SRQfB&VBik$IQ&E+jX@*c$s==Xs9M=@#y9aHJ82zak%@6k9X8A=;-MD
z9WZwgAVP>M!DnYH;N8$r9)z- at x>{#(tO1SpL!%18fj_VdUz==a{`&p9h?3Ib#_bhm
zUq3&RAr8k1tT6$U*&IcW=KIZjdQsjzo~<5!v^9p+!k8?fMu9!v1o0Ug8{<~LsS1+9
zHtZ9F`@HyU6%|Mo#7z77`IFo!Md)G_lt$)>i;>;zQw93BZMM0#fKT;IOmxl7In&b8
z(B%#6GKflvl#7dt^ic6jbV1Zjl6>9W9ceA~PiUy}QPp4!Gc$8dT^$--=I!GXQCoZZ
zRz^M#KR-VUuObVuvIK<ef at _sNLx_C3|Ho83p0 at YF=>v3hbSD%P+DdI-oGl6m+(N?z
zwFqnt?-`=xV%FCDe~dQ@`1tq)ug&b=T3=ue3JOBs0m60Q{d=!C+~g?gqImEiv7q3u
zg2F;gvD at er4MMjGuMYK^)NhPkw<GAs4&@Y3I#h3Na|42RfEJvyFVD}9g7{5+sC0H4
zF8j>e5OndP5IDJ*g-<W?b8;{Mm1sgC(Eb2W+zWPx)<eL^b&IdGU at yW}&Vm1rMj~ix
zYNFQQTDBC9QVsoL)|K`3j<@-`Buk6(15DBIYGjSG)mSYoC}QI#Qd}YsZiJMON-n*b
z5%2;gTG`b-X9qqd4i1{5eq`I{p3Nl5b0^#e1beg91%Nti<hw=i%n-E9?d==Zwzf((
z9nAGcs^GhHAt$0X{0&V^I#;H9h-qomy1Kfw`}U!py3o+jIfjNe=A-t?z%}+S{hMEm
z-%Nlnd`r<_{7o~HOzYJPs+&~Uo<!`rRO9!Z4gB2FpT$P<lb)`w=9#zo=zd?X(7d3i
z2rI?qJooN#y7wH~b at 0V#c!_sVkglmIho`3}V%9P(k%Bk5L}OlaGqdH6 at 9uc4kP!~&
z1I8Cp)u}!HkoAkJ{Ayb>B2R3pnSi}L;;pX32eAX6qXLPEZ{K1dKN=_&M0`D?fwI@@
z^2XhGcp$%`f&;9F4!oQ!)etlcfn?h$&Jc|~(UJaQYH_qyzJyFhV-`RLDP3vP=DBnC
zZro|VZ>Z%E%`ZZu&q4(S01ylEMIGvcBZ^J3OQawNgIB&Xq6+z*4B33Y?|tZe6r{`W
zga>P_T{0>)nH?-dl_7cg`JdWO8REU8(~`MBlsBhKSa|>VAcrRqrpm*2B>CgW`YtYF
zVkTKEKxwcKKYaM$&Ce;s6PK2Jr=+B$1<!Z at r4V8WaFBL{f?x3o2 at _LOQKO@FAb?)>
zEd($KrTbT<oju+Tn~`+*^!nWUg!$C5BZzd?*JG at uI@%Ks&ptc$^%Oj23NfFYn(~=w
z*%z9465Vl#;qKw at D_s*-QQ__A=m3_C?SREz27$8Fe)xjmsZ)od=-I6d1>#q+Rr3l@
z^75#Ylan`Bv-c7~o`Cezb~b(UgiAtFvNcWAu;S`x<!VvBK~)gZu_-B?cjaAEQ&W+;
zbYtO}bt8svPYLMZa5`32cWlV-+zKc+a&vQI<Ki>{RQ>&D?#AIw2j6l+d}oGAc~SqD
zS^htdt7Q$8-y9?AIbOaT4O*i&Q!Wbh<pdwD+&vC{TVB|ob#6H)ZSt at 44i3ppO&Z+1
zy!rNZfdER_hkxqrL5u^1p&nOLas1u8NB&9TK|Ry&L)76_vGOW90$Bq!M&{-U+1cBJ
z^nBzgLNY)n?J6oN+&nxj0|Ok8&eqRda1ZOQt_G_v1*^WE_NAq#Zyg=w1A&F6j0XN3
zeSD|?YrxjlR;GeGC#WOJ at tys7%iM37!hiQ%u!gvk*Y at 3_a?Hxg>fW=G0{iu(rl!XD
z)9_(%`^kK#^E{7{3`QQA<@0#<b31d!;aP!^n-u|)oy{Lt8IQvfemz(B{-Qy?;Q)5z
zGIhi)K$L2OaJ<_|h- at 7BT=`^vikioJH9tQ;RFrb>9uMf`JaFMP;d^S}@JL=AiAT^3
zs7cuDsXf!;Qc`V2xcg1qT6%i1Pwlm}|KFtv(hrqM1#hiM06e4;i4n_FoirHKO;%N<
zkeQh&sHn&_Jv}WhE898NaQ*G8u<MZGPCHw_0w(LdC#Vrz#c4;jv|vGwOaAd>Ma`9<
zdQe%d)ocn!Qb)j$$hGn(PZFsdkC|k7l6IFGiCACXr`_}L$4eONa)*ZW^1_0G3ZJQt
zO87-`KmsIU3?SK4d$4s!FwcSQ=>9B#r=@8EAUWsyIqS at IZj6ho>$zin at Eb|bHx|cT
zG1ed<$AUlYiHnGctOH6`*+8g{AOCBB8vdD3SSZ@&B*(|k|9&hxFEyv6gbBE$dzf5O
z0^%sAygX%Y&I7TBMUjzv)fbJ*KG+11P_SwOk!aT7#md4WNYpbmGovEv6+AUw{y-0Y
z|E#6u<2H%gU2|uXs{`fmAX?B+I8nnpRp>Pxzvn=q=2r=%P3dfHC9s3}1zeMC;%;q;
zX7*YAh69-qB}e!8?8g!xZ~KyMpl-xC0KF3mY*+;<QGV$hGC2VvG69;f^y6YQ^Wn3*
z>Y$x51Esb~7}@MS8qCDR#O6{fdv19-I|dx7-NOy at QBrLsv&12G0~OHH4+FqtTZ9PU
z%&n~%?i$g77%S5f!utU-pnIUc2d60|VRZZUp&}g+VPJn4n3$+BiM&c|m;rG0n5Gg-
zGE$Zhp1;JzoIX~$2aNTMj|*T<TpLpbrG7wER5aC~5Q%+2E(`|U)M6Zej at F_E(94%a
zk4s8cqQS5Cl}rkg#Ps#`Bf!p{>5S#L#l%>JM%7X%6lc&F9*fL-Dw^t7M{EB1_f$vP
za{bpYONnTgAD6q<KvmtBx*O=@<6e<OCQG8jjsvdFp9XkTRaG+^8^g(KI<ZAy!$kk$
zWSsw4Eos|$v*F_(^KHir5~&gp!o|s{MO%3P>dJ}}C^F0Ew|k~Nf3HmUwzNbtMq^#P
zyxv4cMV-gta1H8P=8w}$&p(R+rI0c^KQB+zYXg)xkHtoI2gXE4OMK$y=B at xOSGlz0
zO~S8wJV^42dinWP*nM=unwZE^y`iILtMUGd%_Z2p5X=@}CZ+=i4ghlATDgaAKa)j-
zwB-TzgcIfdR#8E0@^|o(;F){(oaN-?kRM%HSxIj{gs-1}l&X=i^siGy6Vlw=#D^#9
zDX3-!%@_VP_dea5A!J_?pgaT;qz7sprKO|u<?mDZ4ul3ENsGn=0XF)8=jPF=xB-i{
zqt~>tf~0f2bP2zYnX9R{mqj3aTW(mWK^`=A%tR}pb!dpIE@&~vRQxO0Wd)`)mIj8>
z4)xrSF{rK#<USUQ{k!qxg- at blV#0!g7_hO(UT0+;9TwW3=QHylnd`nvkc^YFb7jb?
zOHFO<1V|^KGSXMy)2OEt`OKBCU8pG+OfHXxzHSdLOZ5n9U2}Jr`0?Wh2R}c~WZP2{
z%xKiJXYy#mb7m&D=g8COqIV4Lw4x7#H2Sm7fFMS02v&?Lv_e8c><I6REYF at rbGA@*
zAczXom<`4j+?Do}*<gq+_mLw=!oogB<x1$($nRUPa!Q6+61l9b1I_h2dL|T&yCD*Z
zHGpoYR3g9)Uj?RQOb)!Ewv`nZcp>j^y+?q!(LUc?crjkXhRe6KvJ!?*J`=ju^xp_&
zUZtyF5 at cWHBzQB1(D*;>Q!x^|{a;fy*osad at BIf4;(_oY`irObV^Benu!u<Im0_9J
zmrnNHJH?D at NyftW!jM?P<rQd|+S(?xwP|~f)gO_NkU- at _;H+<ig`HaYtuSY0V-t at W
zM!*vw!rl1!S<bbO4`MTTW`QBZFx@{k9$4S$&-wytp9a5n(bbh8VNujXp;(anZB;{7
zXdydM*82LMgvfy(hDB(IsAFw7MET8SJ5Yrrl5TCCom22SE2^s6K=n=h{F$7qMK~@n
zIK*Xn?i`w`2fqy&w&|V>fTa{zA9~kIYioEQ+~JFyc>&_HsV-ezU7G(IhSaRAWZ*9U
z#W8*mJm5Sb6=_0$%@q>W-+{A0@%5e^6h)PU^704HmDxoYiD`|$3Paya6wz&IZ`b*)
z=Y@?8H60YX_5^@^WMl-{k-$JjKzvYv)F8BAXScwvZn(bj->_LncXviZL&Iu^!hUWo
zP>Vl5*AXHksYh%5=utJ{%^hXPLO^_Q{q)G>oj%<>GIE5CjcsPIFyW(<F#es6KA|cW
z(quB@!p&ynEAHP<0F5^KGJqNUEyh7D5S>EIfaL7$?RE6<D6#ukeT6<OEDQhyg8?-1
z4+x;DSAEQ0&sVA%zRSlMM+;agE-j5JD8bJ!Z7w?>Ja|y!IEfnL at 9&SqkXqQ5cm|L#
zB<P)=TpA3W;<Fspp}^dzE;Tzl1&A^i1aNzMdp919nnGUI(h|)m!I(NYu+`eJVg7G=
z`{`3UwO}XY78e(%RTvo=A!m(Ddd1VHktEWgm;STV5gJX?E(cI at 5;(AW<XmM_EQ1hj
zV>=fHgAqiv0HrZ$Y(n@{E3JPYcuTb;Kdi}tWe8E)K3tKWdQ$O$+9*7W(bqB7F2g!T
F{tpk90)YSk

diff --git a/docs/source/index.rst b/docs/source/index.rst
--- a/docs/source/index.rst
+++ b/docs/source/index.rst
@@ -12,6 +12,12 @@
    :maxdepth: 2
 
    metadata
+   pkgutil
+   depgraph
+   new_commands
+   test_framework
+   pypi
+   version
 
 Indices and tables
 ==================
diff --git a/docs/source/metadata.rst b/docs/source/metadata.rst
--- a/docs/source/metadata.rst
+++ b/docs/source/metadata.rst
@@ -5,13 +5,13 @@
 Distutils2 provides a :class:`DistributionMetadata` class that can read and
 write Metadata files. This class is compatible with all versions of Metadata:
 
-- 1.0 : PEP 241
-- 1.1 : PEP 314
-- 1.2 : PEP 345
+* 1.0 : :pep:`241`
+* 1.1 : :pep:`314`
+* 1.2 : :pep:`345`
 
-The PEP 345 implementation supports the micro-language for the environment
+The :pep:`345` implementation supports the micro-language for the environment
 markers, and displays warnings when versions that are supposed to be
-PEP 386 are violating the scheme.
+:pep:`386` are violating the scheme.
 
 
 Reading metadata
@@ -71,14 +71,14 @@
     >>> metadata.write('/to/my/PKG-INFO')
 
 The class will pick the best version for the metadata, depending on the values
-provided. If all the values provided exists in all versions, teh class will
-used :attr:`metadata.PKG_INFO_PREFERRED_VERSION`. It is set by default to 1.0.
+provided. If all the values provided exists in all versions, the class will
+use :attr:`metadata.PKG_INFO_PREFERRED_VERSION`. It is set by default to 1.0.
 
 
 Conflict checking and best version
 ==================================
 
-Some fields in PEP 345 have to follow a version scheme in their versions
+Some fields in :pep:`345` have to follow a version scheme in their versions
 predicate. When the scheme is violated, a warning is emited::
 
     >>> from distutils2.metadata import DistributionMetadata
@@ -89,7 +89,7 @@
 
 
 
-XXX talk about check()
+.. TODO talk about check()
 
 
 
diff --git a/docs/source/new_commands.rst b/docs/source/new_commands.rst
new file mode 100644
--- /dev/null
+++ b/docs/source/new_commands.rst
@@ -0,0 +1,65 @@
+========
+Commands
+========
+
+Distutils2 provides a set of commands that are not present in distutils itself.
+You might recognize some of them from other projects, like Distribute or
+Setuptools.
+
+``upload_docs`` - Upload package documentation to PyPI
+======================================================
+
+PyPI now supports uploading project documentation to the dedicated URL
+http://packages.python.org/<project>/.
+
+The ``upload_docs`` command will create the necessary zip file out of a
+documentation directory and will post to the repository.
+
+Note that to upload the documentation of a project, the corresponding version
+must already be registered with PyPI, using the distutils ``register``
+command -- just like the ``upload`` command.
+
+Assuming there is an ``Example`` project with documentation in the
+subdirectory ``docs``, e.g.::
+
+  Example/
+  |-- example.py
+  |-- setup.cfg
+  |-- setup.py
+  |-- docs
+  |   |-- build
+  |   |   `-- html
+  |   |   |   |-- index.html
+  |   |   |   `-- tips_tricks.html
+  |   |-- conf.py
+  |   |-- index.txt
+  |   `-- tips_tricks.txt
+
+You can simply pass the documentation directory path to the ``upload_docs``
+command::
+
+    python setup.py upload_docs --upload-dir=docs/build/html
+
+As with any other ``setuptools`` based command, you can define useful
+defaults in the ``setup.cfg`` of your Python project, e.g.:
+
+.. code-block:: ini
+
+    [upload_docs]
+    upload-dir = docs/build/html
+
+The ``upload_docs`` command has the following options:
+
+``--upload-dir``
+    The directory to be uploaded to the repository. The default value is
+    ``docs`` in project root.
+
+``--show-response``
+    Display the full response text from server; this is useful for debugging
+    PyPI problems.
+
+``--repository=URL, -r URL``
+    The URL of the repository to upload to.  Defaults to
+    http://pypi.python.org/pypi (i.e., the main PyPI installation).
+
+
diff --git a/docs/source/pkgutil.rst b/docs/source/pkgutil.rst
new file mode 100644
--- /dev/null
+++ b/docs/source/pkgutil.rst
@@ -0,0 +1,143 @@
+=======
+pkgutil
+=======
+
+Introduction
+============
+
+This module provides the necessary functions to provide support for
+the "Importer Protocol" as described in :pep:`302` and for working with
+the database of installed Python distributions which is specified in
+:pep:`376`. In addition to the functions required in :pep:`376`, back support
+for older ``.egg`` and ``.egg-info`` distributions is provided as well. These
+distributions are represented by the class
+:class:`distutils2._backport.pkgutil.EggInfoDistribution` and
+most functions provide an extra argument ``use_egg_info`` which indicates if
+they should consider these old styled distributions. In this document,
+first a complete documentation of the functions and classes
+is provided and then several use cases are presented.
+
+API Reference
+=============
+
+.. automodule:: distutils2._backport.pkgutil
+   :members:
+
+Example Usage
+=============
+
+Print All Information About a Distribution
+++++++++++++++++++++++++++++++++++++++++++
+
+Given a path to a ``.dist-info`` distribution, we shall print out all
+information that can be obtained using functions provided in this module::
+
+  from distutils2._backport import pkgutil
+  import sys
+
+  path = raw_input() # read the path from the keyboard
+  # first create the Distribution instance
+  try:
+      dist = pkgutil.Distribution(path)
+  except IOError:
+      print('No such distribution')
+      sys.exit(1)
+
+  print('Information about %s' % dist.name)
+  print('Files')
+  print('=====')
+  for (path, md5, size) in dist.get_installed_files():
+      print('* Path: %s' % path)
+      print('  Hash %s, Size: %s bytes' % (md5, size)) 
+  print('Metadata')
+  print('========')
+  for key, value in dist.metadata.items():
+      print('%20s: %s' % (key, value))
+  print('Extra')
+  print('=====')
+  if dist.requested:
+      print('* It was installed by user request')
+  else:
+      print('* It was installed as a dependency')
+
+If we save the script above as ``print_info.py`` and we are intested in the
+distribution located at
+``/home/josip/dev/distutils2/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9``
+then by typing in the console:
+
+.. code-block:: bash
+
+  $ echo /home/josip/dev/distutils2/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info | python print_info.py
+
+we get the following output:
+
+.. code-block:: none
+
+  Information about choxie
+  Files
+  =====
+  * Path: ../home/josip/dev/distutils2/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9/truffles.py
+    Hash 5e052db6a478d06bad9ae033e6bc08af, Size: 111 bytes
+  * Path: ../home/josip/dev/distutils2/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9/choxie/chocolate.py
+    Hash ac56bf496d8d1d26f866235b95f31030, Size: 214 bytes
+  * Path: ../home/josip/dev/distutils2/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9/choxie/__init__.py
+    Hash 416aab08dfa846f473129e89a7625bbc, Size: 25 bytes
+  * Path: ../home/josip/dev/distutils2/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info/INSTALLER
+    Hash d41d8cd98f00b204e9800998ecf8427e, Size: 0 bytes
+  * Path: ../home/josip/dev/distutils2/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info/METADATA
+    Hash 696a209967fef3c8b8f5a7bb10386385, Size: 225 bytes
+  * Path: ../home/josip/dev/distutils2/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info/REQUESTED
+    Hash d41d8cd98f00b204e9800998ecf8427e, Size: 0 bytes
+  * Path: ../home/josip/dev/distutils2/src/distutils2/_backport/tests/fake_dists/choxie-2.0.0.9.dist-info/RECORD
+    Hash None, Size: None bytes
+  Metadata
+  ========
+      Metadata-Version: 1.2
+                  Name: choxie
+               Version: 2.0.0.9
+              Platform: []
+    Supported-Platform: UNKNOWN
+               Summary: Chocolate with a kick!
+           Description: UNKNOWN
+              Keywords: []
+             Home-page: UNKNOWN
+                Author: UNKNOWN
+          Author-email: UNKNOWN
+            Maintainer: UNKNOWN
+      Maintainer-email: UNKNOWN
+               License: UNKNOWN
+            Classifier: []
+          Download-URL: UNKNOWN
+        Obsoletes-Dist: ['truffles (<=0.8,>=0.5)', 'truffles (<=0.9,>=0.6)']
+           Project-URL: []
+         Provides-Dist: ['truffles (1.0)']
+         Requires-Dist: ['towel-stuff (0.1)']
+       Requires-Python: UNKNOWN
+     Requires-External: []
+  Extra
+  =====
+  * It was installed as a dependency
+
+Find Out Obsoleted Distributions
+++++++++++++++++++++++++++++++++
+
+Now, we take tackle a different problem, we are interested in finding out
+which distributions have been obsoleted. This can be easily done as follows::
+
+  from distutils2._backport import pkgutil
+
+  # iterate over all distributions in the system
+  for dist in pkgutil.get_distributions():
+      name = dist.name
+      version = dist.metadata['Version']
+      # find out which distributions obsolete this name/version combination
+      for obsoleted_by in pkgutil.obsoletes_distribution(name, version):
+          print('%s(%s) is obsoleted by %s' % (name, version, obsoleted_by.name))
+
+This is how the output might look like:
+
+.. code-block:: none
+
+  strawberry(0.6) is obsoleted by choxie
+  grammar(1.0a4) is obsoleted by towel-stuff
+
diff --git a/docs/source/pypi.rst b/docs/source/pypi.rst
new file mode 100644
--- /dev/null
+++ b/docs/source/pypi.rst
@@ -0,0 +1,195 @@
+=========================================
+Tools to query PyPI: the PyPI package
+=========================================
+
+Distutils2 comes with a module (eg. `distutils2.pypi`) which contains
+facilities to access the Python Package Index (named "pypi", and avalaible on
+the url `http://pypi.python.org`.
+
+There is two ways to retrieve data from pypi: using the *simple* API, and using
+*XML-RPC*. The first one is in fact a set of HTML pages avalaible at
+`http://pypi.python.org/simple/`, and the second one contains a set of XML-RPC
+methods. In order to reduce the overload caused by running distant methods on 
+the pypi server (by using the XML-RPC methods), the best way to retrieve 
+informations is by using the simple API, when it contains the information you 
+need.
+
+Distutils2 provides two python modules to ease the work with those two APIs:
+`distutils2.pypi.simple` and `distutils2.pypi.xmlrpc`. Both of them depends on 
+another python module: `distutils2.pypi.dist`.
+
+
+Requesting information via the "simple" API `distutils2.pypi.simple`
+====================================================================
+
+`distutils2.pypi.simple` can process the Python Package Index and return and 
+download urls of distributions, for specific versions or latests, but it also 
+can process external html pages, with the goal to find *pypi unhosted* versions 
+of python distributions.
+
+You should use `distutils2.pypi.simple` for: 
+
+    * Search distributions by name and versions.
+    * Process pypi external pages.
+    * Download distributions by name and versions.
+
+And should not be used to:
+
+    * Things that will end up in too long index processing (like "finding all
+      distributions with a specific version, no matters the name")
+
+API
+----
+
+Here is a complete overview of the APIs of the SimpleIndex class.
+
+.. autoclass:: distutils2.pypi.simple.SimpleIndex
+    :members:
+
+Usage Exemples
+---------------
+
+To help you understand how using the `SimpleIndex` class, here are some basic
+usages.
+
+Request PyPI to get a specific distribution
+++++++++++++++++++++++++++++++++++++++++++++
+
+Supposing you want to scan the PyPI index to get a list of distributions for 
+the "foobar" project. You can use the "find" method for that::
+
+    >>> from distutils2.pypi import SimpleIndex
+    >>> client = SimpleIndex()
+    >>> client.find("foobar")
+    [<PyPIDistribution "Foobar 1.1">, <PyPIDistribution "Foobar 1.2">]
+    
+Note that you also can request the client about specific versions, using version
+specifiers (described in `PEP 345 
+<http://www.python.org/dev/peps/pep-0345/#version-specifiers>`_)::
+
+    >>> client.find("foobar < 1.2")
+    [<PyPIDistribution "foobar 1.1">, ]
+
+`find` returns a list of distributions, but you also can get the last
+distribution (the more up to date) that fullfil your requirements, like this::
+    
+    >>> client.get("foobar < 1.2")
+    <PyPIDistribution "foobar 1.1">
+
+Download distributions
++++++++++++++++++++++++
+
+As it can get the urls of distributions provided by PyPI, the `SimpleIndex` 
+client also can download the distributions and put it for you in a temporary
+destination::
+
+    >>> client.download("foobar")
+    /tmp/temp_dir/foobar-1.2.tar.gz
+
+You also can specify the directory you want to download to::
+    
+    >>> client.download("foobar", "/path/to/my/dir")
+    /path/to/my/dir/foobar-1.2.tar.gz
+
+While downloading, the md5 of the archive will be checked, if not matches, it
+will try another time, then if fails again, raise `MD5HashDoesNotMatchError`.
+
+Internally, that's not the SimpleIndex which download the distributions, but the
+`PyPIDistribution` class. Please refer to this documentation for more details.
+
+Following PyPI external links
+++++++++++++++++++++++++++++++
+
+The default behavior for distutils2 is to *not* follow the links provided
+by HTML pages in the "simple index", to find distributions related
+downloads.
+
+It's possible to tell the PyPIClient to follow external links by setting the 
+`follow_externals` attribute, on instanciation or after::
+
+    >>> client = SimpleIndex(follow_externals=True)
+
+or ::
+
+    >>> client = SimpleIndex()
+    >>> client.follow_externals = True
+
+Working with external indexes, and mirrors
++++++++++++++++++++++++++++++++++++++++++++
+
+The default `SimpleIndex` behavior is to rely on the Python Package index stored
+on PyPI (http://pypi.python.org/simple).
+
+As you can need to work with a local index, or private indexes, you can specify
+it using the index_url parameter::
+
+    >>> client = SimpleIndex(index_url="file://filesystem/path/")
+
+or ::
+
+    >>> client = SimpleIndex(index_url="http://some.specific.url/")
+
+You also can specify mirrors to fallback on in case the first index_url you
+provided doesnt respond, or not correctly. The default behavior for
+`SimpleIndex` is to use the list provided by Python.org DNS records, as
+described in the :pep:`381` about mirroring infrastructure.
+
+If you don't want to rely on these, you could specify the list of mirrors you
+want to try by specifying the `mirrors` attribute. It's a simple iterable::
+
+    >>> mirrors = ["http://first.mirror","http://second.mirror"]
+    >>> client = SimpleIndex(mirrors=mirrors)
+
+
+Requesting informations via XML-RPC (`distutils2.pypi.XmlRpcIndex`)
+==========================================================================
+
+The other method to request the Python package index, is using the XML-RPC
+methods. Distutils2 provides a simple wrapper around `xmlrpclib
+<http://docs.python.org/library/xmlrpclib.html>`_, that can return you
+`PyPIDistribution` objects.
+
+::
+    >>> from distutils2.pypi import XmlRpcIndex()
+    >>> client = XmlRpcIndex()
+
+
+PyPI Distributions
+==================
+
+Both `SimpleIndex` and `XmlRpcIndex` classes works with the classes provided
+in the `pypi.dist` package.
+
+`PyPIDistribution`
+------------------
+
+`PyPIDistribution` is a simple class that defines the following attributes:
+
+:name:
+    The name of the package. `foobar` in our exemples here
+:version:
+    The version of the package
+:location:
+    If the files from the archive has been downloaded, here is the path where
+    you can find them.
+:url:
+    The url of the distribution
+
+.. autoclass:: distutils2.pypi.dist.PyPIDistribution
+    :members:
+
+`PyPIDistributions`
+-------------------
+
+The `dist` module also provides another class, to work with lists of 
+`PyPIDistribution` classes. It allow to filter results and is used as a 
+container of 
+
+.. autoclass:: distutils2.pypi.dist.PyPIDistributions
+    :members:
+
+At a higher level
+=================
+
+XXX : A description about a wraper around PyPI simple and XmlRpc Indexes
+(PyPIIndex ?) 
diff --git a/docs/source/test_framework.rst b/docs/source/test_framework.rst
new file mode 100644
--- /dev/null
+++ b/docs/source/test_framework.rst
@@ -0,0 +1,88 @@
+==============
+Test Framework
+==============
+
+When you are testing code that works with distutils, you might find these tools
+useful.
+
+``PyPIServer``
+==============
+
+PyPIServer is a class that implements an HTTP server running in a separate
+thread. All it does is record the requests for further inspection. The recorded
+data is available under ``requests`` attribute. The default
+HTTP response can be overriden with the ``default_response_status``,
+``default_response_headers`` and ``default_response_data`` attributes.
+
+By default, when accessing the server with urls beginning with `/simple/`, 
+the server also record your requests, but will look for files under 
+the `/tests/pypiserver/simple/` path.
+
+You can tell the sever to serve static files for other paths. This could be 
+accomplished by using the `static_uri_paths` parameter, as below::
+
+    server = PyPIServer(static_uri_paths=["first_path", "second_path"])
+
+You need to create the content that will be served under the 
+`/tests/pypiserver/default` path. If you want to serve content from another 
+place, you also can specify another filesystem path (wich need to be under
+`tests/pypiserver/`. This will replace the default behavior of the server, and
+it will not serve content from the `default` dir ::
+
+    server = PyPIServer(static_filesystem_paths=["path/to/your/dir"])
+
+If you just need to add some paths to the existing ones, you can do as shown, 
+keeping in mind that the server will alwas try to load paths in reverse order 
+(e.g here, try "another/super/path" then the default one) ::
+
+    server = PyPIServer(test_static_path="another/super/path")
+    server = PyPIServer("another/super/path")
+    # or 
+    server.static_filesystem_paths.append("another/super/path")
+
+As a result of what, in your tests, while you need to use the PyPIServer, in
+order to isolates the test cases, the best practice is to place the common files
+in the `default` folder, and to create a directory for each specific test case::
+
+    server = PyPIServer(static_filesystem_paths = ["default", "test_pypi_server"],
+        static_uri_paths=["simple", "external"])
+
+``PyPIServerTestCase``
+======================
+
+``PyPIServerTestCase`` is a test case class with setUp and tearDown methods that
+take care of a single PyPIServer instance attached as a ``pypi`` attribute on
+the test class. Use it as one of the base classes in your test case::
+
+  class UploadTestCase(PyPIServerTestCase):
+      def test_something(self):
+          cmd = self.prepare_command()
+          cmd.ensure_finalized()
+          cmd.repository = self.pypi.full_address
+          cmd.run()
+
+          environ, request_data = self.pypi.requests[-1]
+          self.assertEqual(request_data, EXPECTED_REQUEST_DATA)
+
+The ``use_pypi_server`` decorator
+=================================
+
+You also can use a decorator for your tests, if you do not need the same server
+instance along all you test case. So, you can specify, for each test method,
+some initialisation parameters for the server.
+
+For this, you need to add a `server` parameter to your method, like this::
+
+    class SampleTestCase(TestCase):
+        @use_pypi_server()
+        def test_somthing(self, server):
+            # your tests goes here
+
+The decorator will instanciate the server for you, and run and stop it just
+before and after your method call. You also can pass the server initializer,
+just like this::
+
+    class SampleTestCase(TestCase):
+        @use_pypi_server("test_case_name")
+        def test_something(self, server):
+            # something
diff --git a/docs/source/version.rst b/docs/source/version.rst
new file mode 100644
--- /dev/null
+++ b/docs/source/version.rst
@@ -0,0 +1,64 @@
+======================
+Working with versions
+======================
+
+Distutils2 ships with a python package capable to work with version numbers.
+It's an implementation of version specifiers `as defined in PEP 345
+<http://www.python.org/dev/peps/pep-0345/#version-specifiers>`_ about
+Metadata.
+
+`distutils2.version.NormalizedVersion`
+======================================
+
+A Normalized version corresponds to a specific version of a distribution, as
+described in the PEP 345. So, you can work with the `NormalizedVersion` like
+this::
+
+    >>> NormalizedVersion("1.2b1")
+    NormalizedVersion('1.2b1')
+
+If you try to use irrational version specifiers, an `IrrationalVersionError`
+will be raised::
+
+    >>> NormalizedVersion("irrational_version_number")
+    ...
+    IrrationalVersionError: irrational_version_number
+
+You can compare NormalizedVersion objects, like this::
+
+    >>> NormalizedVersion("1.2b1") < NormalizedVersion("1.2")
+    True
+
+NormalizedVersion is used internally by `VersionPredicate` to do his stuff.
+
+`distutils2.version.suggest_normalized_version`
+-----------------------------------------------
+
+You also can let the normalized version be suggested to you, using the
+`suggest_normalized_version` function::
+
+    >>> suggest_normalized_version('2.1-rc1') 
+    2.1c1
+
+If `suggest_normalized_version` can't actually suggest you a version, it will
+return `None`::
+
+    >>> print suggest_normalized_version('not a version')
+    None
+
+`distutils2.version.VersionPredicate`
+=====================================
+
+`VersionPredicate` knows how to parse stuff like "ProjectName (>=version)", the
+class also provides a `match` method to test if a version number is the version
+predicate::
+
+    >>> version = VersionPredicate("ProjectName (<1.2,>1.0")
+    >>> version.match("1.2.1")
+    False
+    >>> version.match("1.1.1")
+    True
+
+`is_valid_predicate`
+--------------------
+
diff --git a/src/CONTRIBUTORS.txt b/src/CONTRIBUTORS.txt
--- a/src/CONTRIBUTORS.txt
+++ b/src/CONTRIBUTORS.txt
@@ -5,7 +5,7 @@
 Distutils2 is a project that was started and that is maintained by
 Tarek Ziad??, and many people are contributing to the project.
 
-If you did, please add your name below in alphabetical order !
+If you did, please add your name below in alphabetical order!
 
 Thanks to:
 
@@ -15,12 +15,13 @@
 - Nicolas Cadou
 - Josip Djolonga
 - Yannick Gringas
+- Jeremy Kloth
+- Martin von L??wis
 - Carl Meyer
+- Alexis Metaireau
 - Michael Mulich
 - George Peris
 - Sean Reifschneider
 - Erik Rose
 - Brian Rosner
 - Alexandre Vassalotti
-- Martin von L??wis
-
diff --git a/src/DEVNOTES.txt b/src/DEVNOTES.txt
--- a/src/DEVNOTES.txt
+++ b/src/DEVNOTES.txt
@@ -3,8 +3,8 @@
 
 - Distutils2 runs from 2.4 to 3.2 (3.x not implemented yet), so
   make sure you don't use a syntax that doesn't work under
-  a specific Python version.
+  one of these Python versions.
 
 - Always run tests.sh before you push a change. This implies
-  that you have all Python versions installed.
+  that you have all Python versions installed from 2.4 to 2.6.
 
diff --git a/src/Modules/_hashopenssl.c b/src/Modules/_hashopenssl.c
new file mode 100644
--- /dev/null
+++ b/src/Modules/_hashopenssl.c
@@ -0,0 +1,524 @@
+/* Module that wraps all OpenSSL hash algorithms */
+
+/*
+ * Copyright (C) 2005   Gregory P. Smith (greg at krypto.org)
+ * Licensed to PSF under a Contributor Agreement.
+ *
+ * Derived from a skeleton of shamodule.c containing work performed by:
+ *
+ * Andrew Kuchling (amk at amk.ca)
+ * Greg Stein (gstein at lyra.org)
+ *
+ */
+
+#define PY_SSIZE_T_CLEAN
+
+#include "Python.h"
+#include "structmember.h"
+
+#if (PY_VERSION_HEX < 0x02050000)
+#define Py_ssize_t      int
+#endif
+
+/* EVP is the preferred interface to hashing in OpenSSL */
+#include <openssl/evp.h>
+
+#define MUNCH_SIZE INT_MAX
+
+
+#ifndef HASH_OBJ_CONSTRUCTOR
+#define HASH_OBJ_CONSTRUCTOR 0
+#endif
+
+typedef struct {
+    PyObject_HEAD
+    PyObject            *name;  /* name of this hash algorithm */
+    EVP_MD_CTX          ctx;    /* OpenSSL message digest context */
+} EVPobject;
+
+
+static PyTypeObject EVPtype;
+
+
+#define DEFINE_CONSTS_FOR_NEW(Name)  \
+    static PyObject *CONST_ ## Name ## _name_obj; \
+    static EVP_MD_CTX CONST_new_ ## Name ## _ctx; \
+    static EVP_MD_CTX *CONST_new_ ## Name ## _ctx_p = NULL;
+
+DEFINE_CONSTS_FOR_NEW(md5)
+DEFINE_CONSTS_FOR_NEW(sha1)
+DEFINE_CONSTS_FOR_NEW(sha224)
+DEFINE_CONSTS_FOR_NEW(sha256)
+DEFINE_CONSTS_FOR_NEW(sha384)
+DEFINE_CONSTS_FOR_NEW(sha512)
+
+
+static EVPobject *
+newEVPobject(PyObject *name)
+{
+    EVPobject *retval = (EVPobject *)PyObject_New(EVPobject, &EVPtype);
+
+    /* save the name for .name to return */
+    if (retval != NULL) {
+        Py_INCREF(name);
+        retval->name = name;
+    }
+
+    return retval;
+}
+
+/* Internal methods for a hash object */
+
+static void
+EVP_dealloc(PyObject *ptr)
+{
+    EVP_MD_CTX_cleanup(&((EVPobject *)ptr)->ctx);
+    Py_XDECREF(((EVPobject *)ptr)->name);
+    PyObject_Del(ptr);
+}
+
+
+/* External methods for a hash object */
+
+PyDoc_STRVAR(EVP_copy__doc__, "Return a copy of the hash object.");
+
+static PyObject *
+EVP_copy(EVPobject *self, PyObject *unused)
+{
+    EVPobject *newobj;
+
+    if ( (newobj = newEVPobject(self->name))==NULL)
+        return NULL;
+
+    EVP_MD_CTX_copy(&newobj->ctx, &self->ctx);
+    return (PyObject *)newobj;
+}
+
+PyDoc_STRVAR(EVP_digest__doc__,
+"Return the digest value as a string of binary data.");
+
+static PyObject *
+EVP_digest(EVPobject *self, PyObject *unused)
+{
+    unsigned char digest[EVP_MAX_MD_SIZE];
+    EVP_MD_CTX temp_ctx;
+    PyObject *retval;
+    unsigned int digest_size;
+
+    EVP_MD_CTX_copy(&temp_ctx, &self->ctx);
+    digest_size = EVP_MD_CTX_size(&temp_ctx);
+    EVP_DigestFinal(&temp_ctx, digest, NULL);
+
+    retval = PyString_FromStringAndSize((const char *)digest, digest_size);
+    EVP_MD_CTX_cleanup(&temp_ctx);
+    return retval;
+}
+
+PyDoc_STRVAR(EVP_hexdigest__doc__,
+"Return the digest value as a string of hexadecimal digits.");
+
+static PyObject *
+EVP_hexdigest(EVPobject *self, PyObject *unused)
+{
+    unsigned char digest[EVP_MAX_MD_SIZE];
+    EVP_MD_CTX temp_ctx;
+    PyObject *retval;
+    char *hex_digest;
+    unsigned int i, j, digest_size;
+
+    /* Get the raw (binary) digest value */
+    EVP_MD_CTX_copy(&temp_ctx, &self->ctx);
+    digest_size = EVP_MD_CTX_size(&temp_ctx);
+    EVP_DigestFinal(&temp_ctx, digest, NULL);
+
+    EVP_MD_CTX_cleanup(&temp_ctx);
+
+    /* Create a new string */
+    /* NOTE: not thread safe! modifying an already created string object */
+    /* (not a problem because we hold the GIL by default) */
+    retval = PyString_FromStringAndSize(NULL, digest_size * 2);
+    if (!retval)
+	    return NULL;
+    hex_digest = PyString_AsString(retval);
+    if (!hex_digest) {
+	    Py_DECREF(retval);
+	    return NULL;
+    }
+
+    /* Make hex version of the digest */
+    for(i=j=0; i<digest_size; i++) {
+        char c;
+        c = (digest[i] >> 4) & 0xf;
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+        c = (digest[i] & 0xf);
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+    }
+    return retval;
+}
+
+PyDoc_STRVAR(EVP_update__doc__,
+"Update this hash object's state with the provided string.");
+
+static PyObject *
+EVP_update(EVPobject *self, PyObject *args)
+{
+    unsigned char *cp;
+    Py_ssize_t len;
+
+    if (!PyArg_ParseTuple(args, "s#:update", &cp, &len))
+        return NULL;
+
+    if (len > 0 && len <= MUNCH_SIZE) {
+    EVP_DigestUpdate(&self->ctx, cp, Py_SAFE_DOWNCAST(len, Py_ssize_t,
+                                                      unsigned int));
+    } else {
+        Py_ssize_t offset = 0;
+        while (len) {
+            unsigned int process = len > MUNCH_SIZE ? MUNCH_SIZE : len;
+            EVP_DigestUpdate(&self->ctx, cp + offset, process);
+            len -= process;
+            offset += process;
+        }
+    }
+    Py_INCREF(Py_None);
+    return Py_None;
+}
+
+static PyMethodDef EVP_methods[] = {
+    {"update",	  (PyCFunction)EVP_update,    METH_VARARGS, EVP_update__doc__},
+    {"digest",	  (PyCFunction)EVP_digest,    METH_NOARGS,  EVP_digest__doc__},
+    {"hexdigest", (PyCFunction)EVP_hexdigest, METH_NOARGS,  EVP_hexdigest__doc__},
+    {"copy",	  (PyCFunction)EVP_copy,      METH_NOARGS,  EVP_copy__doc__},
+    {NULL,	  NULL}		/* sentinel */
+};
+
+static PyObject *
+EVP_get_block_size(EVPobject *self, void *closure)
+{
+    return PyInt_FromLong(EVP_MD_CTX_block_size(&((EVPobject *)self)->ctx));
+}
+
+static PyObject *
+EVP_get_digest_size(EVPobject *self, void *closure)
+{
+    return PyInt_FromLong(EVP_MD_CTX_size(&((EVPobject *)self)->ctx));
+}
+
+static PyMemberDef EVP_members[] = {
+    {"name", T_OBJECT, offsetof(EVPobject, name), READONLY, PyDoc_STR("algorithm name.")},
+    {NULL}  /* Sentinel */
+};
+
+static PyGetSetDef EVP_getseters[] = {
+    {"digest_size",
+     (getter)EVP_get_digest_size, NULL,
+     NULL,
+     NULL},
+    {"block_size",
+     (getter)EVP_get_block_size, NULL,
+     NULL,
+     NULL},
+    /* the old md5 and sha modules support 'digest_size' as in PEP 247.
+     * the old sha module also supported 'digestsize'.  ugh. */
+    {"digestsize",
+     (getter)EVP_get_digest_size, NULL,
+     NULL,
+     NULL},
+    {NULL}  /* Sentinel */
+};
+
+
+static PyObject *
+EVP_repr(PyObject *self)
+{
+    char buf[100];
+    PyOS_snprintf(buf, sizeof(buf), "<%s HASH object @ %p>",
+            PyString_AsString(((EVPobject *)self)->name), self);
+    return PyString_FromString(buf);
+}
+
+#if HASH_OBJ_CONSTRUCTOR
+static int
+EVP_tp_init(EVPobject *self, PyObject *args, PyObject *kwds)
+{
+    static char *kwlist[] = {"name", "string", NULL};
+    PyObject *name_obj = NULL;
+    char *nameStr;
+    unsigned char *cp = NULL;
+    Py_ssize_t len = 0;
+    const EVP_MD *digest;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwds, "O|s#:HASH", kwlist,
+                                     &name_obj, &cp, &len)) {
+        return -1;
+    }
+
+    if (!PyArg_Parse(name_obj, "s", &nameStr)) {
+        PyErr_SetString(PyExc_TypeError, "name must be a string");
+        return -1;
+    }
+
+    digest = EVP_get_digestbyname(nameStr);
+    if (!digest) {
+        PyErr_SetString(PyExc_ValueError, "unknown hash function");
+        return -1;
+    }
+    EVP_DigestInit(&self->ctx, digest);
+
+    self->name = name_obj;
+    Py_INCREF(self->name);
+
+    if (cp && len) {
+        if (len > 0 && len <= MUNCH_SIZE) {
+        EVP_DigestUpdate(&self->ctx, cp, Py_SAFE_DOWNCAST(len, Py_ssize_t,
+                                                          unsigned int));
+        } else {
+            Py_ssize_t offset = 0;
+            while (len) {
+                unsigned int process = len > MUNCH_SIZE ? MUNCH_SIZE : len;
+                EVP_DigestUpdate(&self->ctx, cp + offset, process);
+                len -= process;
+                offset += process;
+            }
+        }
+    }
+    
+    return 0;
+}
+#endif
+
+
+PyDoc_STRVAR(hashtype_doc,
+"A hash represents the object used to calculate a checksum of a\n\
+string of information.\n\
+\n\
+Methods:\n\
+\n\
+update() -- updates the current digest with an additional string\n\
+digest() -- return the current digest value\n\
+hexdigest() -- return the current digest as a string of hexadecimal digits\n\
+copy() -- return a copy of the current hash object\n\
+\n\
+Attributes:\n\
+\n\
+name -- the hash algorithm being used by this object\n\
+digest_size -- number of bytes in this hashes output\n");
+
+static PyTypeObject EVPtype = {
+    PyObject_HEAD_INIT(NULL)
+    0,			/*ob_size*/
+    "_hashlib.HASH",    /*tp_name*/
+    sizeof(EVPobject),	/*tp_basicsize*/
+    0,			/*tp_itemsize*/
+    /* methods */
+    EVP_dealloc,	/*tp_dealloc*/
+    0,			/*tp_print*/
+    0,                  /*tp_getattr*/
+    0,                  /*tp_setattr*/
+    0,                  /*tp_compare*/
+    EVP_repr,           /*tp_repr*/
+    0,                  /*tp_as_number*/
+    0,                  /*tp_as_sequence*/
+    0,                  /*tp_as_mapping*/
+    0,                  /*tp_hash*/
+    0,                  /*tp_call*/
+    0,                  /*tp_str*/
+    0,                  /*tp_getattro*/
+    0,                  /*tp_setattro*/
+    0,                  /*tp_as_buffer*/
+    Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE, /*tp_flags*/
+    hashtype_doc,       /*tp_doc*/
+    0,                  /*tp_traverse*/
+    0,			/*tp_clear*/
+    0,			/*tp_richcompare*/
+    0,			/*tp_weaklistoffset*/
+    0,			/*tp_iter*/
+    0,			/*tp_iternext*/
+    EVP_methods,	/* tp_methods */
+    EVP_members,	/* tp_members */
+    EVP_getseters,      /* tp_getset */
+#if 1
+    0,                  /* tp_base */
+    0,                  /* tp_dict */
+    0,                  /* tp_descr_get */
+    0,                  /* tp_descr_set */
+    0,                  /* tp_dictoffset */
+#endif
+#if HASH_OBJ_CONSTRUCTOR
+    (initproc)EVP_tp_init, /* tp_init */
+#endif
+};
+
+static PyObject *
+EVPnew(PyObject *name_obj,
+       const EVP_MD *digest, const EVP_MD_CTX *initial_ctx,
+       const unsigned char *cp, Py_ssize_t len)
+{
+    EVPobject *self;
+
+    if (!digest && !initial_ctx) {
+        PyErr_SetString(PyExc_ValueError, "unsupported hash type");
+        return NULL;
+    }
+
+    if ((self = newEVPobject(name_obj)) == NULL)
+        return NULL;
+
+    if (initial_ctx) {
+        EVP_MD_CTX_copy(&self->ctx, initial_ctx);
+    } else {
+        EVP_DigestInit(&self->ctx, digest);
+    }
+
+    if (cp && len) {
+        if (len > 0 && len <= MUNCH_SIZE) {
+            EVP_DigestUpdate(&self->ctx, cp, Py_SAFE_DOWNCAST(len, Py_ssize_t,
+                                                              unsigned int));
+        } else {
+            Py_ssize_t offset = 0;
+            while (len) {
+                unsigned int process = len > MUNCH_SIZE ? MUNCH_SIZE : len;
+                EVP_DigestUpdate(&self->ctx, cp + offset, process);
+                len -= process;
+                offset += process;
+            }
+        }
+    }
+
+    return (PyObject *)self;
+}
+
+
+/* The module-level function: new() */
+
+PyDoc_STRVAR(EVP_new__doc__,
+"Return a new hash object using the named algorithm.\n\
+An optional string argument may be provided and will be\n\
+automatically hashed.\n\
+\n\
+The MD5 and SHA1 algorithms are always supported.\n");
+
+static PyObject *
+EVP_new(PyObject *self, PyObject *args, PyObject *kwdict)
+{
+    static char *kwlist[] = {"name", "string", NULL};
+    PyObject *name_obj = NULL;
+    char *name;
+    const EVP_MD *digest;
+    unsigned char *cp = NULL;
+    Py_ssize_t len = 0;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwdict, "O|s#:new", kwlist,
+                                     &name_obj, &cp, &len)) {
+        return NULL;
+    }
+
+    if (!PyArg_Parse(name_obj, "s", &name)) {
+        PyErr_SetString(PyExc_TypeError, "name must be a string");
+        return NULL;
+    }
+
+    digest = EVP_get_digestbyname(name);
+
+    return EVPnew(name_obj, digest, NULL, cp, len);
+}
+
+/*
+ *  This macro generates constructor function definitions for specific
+ *  hash algorithms.  These constructors are much faster than calling
+ *  the generic one passing it a python string and are noticably
+ *  faster than calling a python new() wrapper.  Thats important for
+ *  code that wants to make hashes of a bunch of small strings.
+ */
+#define GEN_CONSTRUCTOR(NAME)  \
+    static PyObject * \
+    EVP_new_ ## NAME (PyObject *self, PyObject *args) \
+    { \
+        unsigned char *cp = NULL; \
+        Py_ssize_t len = 0; \
+     \
+        if (!PyArg_ParseTuple(args, "|s#:" #NAME , &cp, &len)) { \
+            return NULL; \
+        } \
+     \
+        return EVPnew( \
+                CONST_ ## NAME ## _name_obj, \
+                NULL, \
+                CONST_new_ ## NAME ## _ctx_p, \
+                cp, len); \
+    }
+
+/* a PyMethodDef structure for the constructor */
+#define CONSTRUCTOR_METH_DEF(NAME)  \
+    {"openssl_" #NAME, (PyCFunction)EVP_new_ ## NAME, METH_VARARGS, \
+        PyDoc_STR("Returns a " #NAME \
+                  " hash object; optionally initialized with a string") \
+    }
+
+/* used in the init function to setup a constructor */
+#define INIT_CONSTRUCTOR_CONSTANTS(NAME)  do { \
+    CONST_ ## NAME ## _name_obj = PyString_FromString(#NAME); \
+    if (EVP_get_digestbyname(#NAME)) { \
+        CONST_new_ ## NAME ## _ctx_p = &CONST_new_ ## NAME ## _ctx; \
+        EVP_DigestInit(CONST_new_ ## NAME ## _ctx_p, EVP_get_digestbyname(#NAME)); \
+    } \
+} while (0);
+
+GEN_CONSTRUCTOR(md5)
+GEN_CONSTRUCTOR(sha1)
+GEN_CONSTRUCTOR(sha224)
+GEN_CONSTRUCTOR(sha256)
+GEN_CONSTRUCTOR(sha384)
+GEN_CONSTRUCTOR(sha512)
+
+/* List of functions exported by this module */
+
+static struct PyMethodDef EVP_functions[] = {
+    {"new", (PyCFunction)EVP_new, METH_VARARGS|METH_KEYWORDS, EVP_new__doc__},
+    CONSTRUCTOR_METH_DEF(md5),
+    CONSTRUCTOR_METH_DEF(sha1),
+    CONSTRUCTOR_METH_DEF(sha224),
+    CONSTRUCTOR_METH_DEF(sha256),
+    CONSTRUCTOR_METH_DEF(sha384),
+    CONSTRUCTOR_METH_DEF(sha512),
+    {NULL,	NULL}		 /* Sentinel */
+};
+
+
+/* Initialize this module. */
+
+PyMODINIT_FUNC
+init_hashlib(void)
+{
+    PyObject *m;
+
+    OpenSSL_add_all_digests();
+
+    /* TODO build EVP_functions openssl_* entries dynamically based
+     * on what hashes are supported rather than listing many
+     * but having some be unsupported.  Only init appropriate
+     * constants. */
+
+    EVPtype.ob_type = &PyType_Type;
+    if (PyType_Ready(&EVPtype) < 0)
+        return;
+
+    m = Py_InitModule("_hashlib", EVP_functions);
+    if (m == NULL)
+        return;
+
+#if HASH_OBJ_CONSTRUCTOR
+    Py_INCREF(&EVPtype);
+    PyModule_AddObject(m, "HASH", (PyObject *)&EVPtype);
+#endif
+
+    /* these constants are used by the convenience constructors */
+    INIT_CONSTRUCTOR_CONSTANTS(md5);
+    INIT_CONSTRUCTOR_CONSTANTS(sha1);
+    INIT_CONSTRUCTOR_CONSTANTS(sha224);
+    INIT_CONSTRUCTOR_CONSTANTS(sha256);
+    INIT_CONSTRUCTOR_CONSTANTS(sha384);
+    INIT_CONSTRUCTOR_CONSTANTS(sha512);
+}
diff --git a/src/Modules/md5.c b/src/Modules/md5.c
new file mode 100644
--- /dev/null
+++ b/src/Modules/md5.c
@@ -0,0 +1,381 @@
+/*
+  Copyright (C) 1999, 2000, 2002 Aladdin Enterprises.  All rights reserved.
+
+  This software is provided 'as-is', without any express or implied
+  warranty.  In no event will the authors be held liable for any damages
+  arising from the use of this software.
+
+  Permission is granted to anyone to use this software for any purpose,
+  including commercial applications, and to alter it and redistribute it
+  freely, subject to the following restrictions:
+
+  1. The origin of this software must not be misrepresented; you must not
+     claim that you wrote the original software. If you use this software
+     in a product, an acknowledgment in the product documentation would be
+     appreciated but is not required.
+  2. Altered source versions must be plainly marked as such, and must not be
+     misrepresented as being the original software.
+  3. This notice may not be removed or altered from any source distribution.
+
+  L. Peter Deutsch
+  ghost at aladdin.com
+
+ */
+/* $Id: md5.c,v 1.6 2002/04/13 19:20:28 lpd Exp $ */
+/*
+  Independent implementation of MD5 (RFC 1321).
+
+  This code implements the MD5 Algorithm defined in RFC 1321, whose
+  text is available at
+	http://www.ietf.org/rfc/rfc1321.txt
+  The code is derived from the text of the RFC, including the test suite
+  (section A.5) but excluding the rest of Appendix A.  It does not include
+  any code or documentation that is identified in the RFC as being
+  copyrighted.
+
+  The original and principal author of md5.c is L. Peter Deutsch
+  <ghost at aladdin.com>.  Other authors are noted in the change history
+  that follows (in reverse chronological order):
+
+  2002-04-13 lpd Clarified derivation from RFC 1321; now handles byte order
+	either statically or dynamically; added missing #include <string.h>
+	in library.
+  2002-03-11 lpd Corrected argument list for main(), and added int return
+	type, in test program and T value program.
+  2002-02-21 lpd Added missing #include <stdio.h> in test program.
+  2000-07-03 lpd Patched to eliminate warnings about "constant is
+	unsigned in ANSI C, signed in traditional"; made test program
+	self-checking.
+  1999-11-04 lpd Edited comments slightly for automatic TOC extraction.
+  1999-10-18 lpd Fixed typo in header comment (ansi2knr rather than md5).
+  1999-05-03 lpd Original version.
+ */
+
+#include "md5.h"
+#include <string.h>
+
+#undef BYTE_ORDER	/* 1 = big-endian, -1 = little-endian, 0 = unknown */
+#ifdef ARCH_IS_BIG_ENDIAN
+#  define BYTE_ORDER (ARCH_IS_BIG_ENDIAN ? 1 : -1)
+#else
+#  define BYTE_ORDER 0
+#endif
+
+#define T_MASK ((md5_word_t)~0)
+#define T1 /* 0xd76aa478 */ (T_MASK ^ 0x28955b87)
+#define T2 /* 0xe8c7b756 */ (T_MASK ^ 0x173848a9)
+#define T3    0x242070db
+#define T4 /* 0xc1bdceee */ (T_MASK ^ 0x3e423111)
+#define T5 /* 0xf57c0faf */ (T_MASK ^ 0x0a83f050)
+#define T6    0x4787c62a
+#define T7 /* 0xa8304613 */ (T_MASK ^ 0x57cfb9ec)
+#define T8 /* 0xfd469501 */ (T_MASK ^ 0x02b96afe)
+#define T9    0x698098d8
+#define T10 /* 0x8b44f7af */ (T_MASK ^ 0x74bb0850)
+#define T11 /* 0xffff5bb1 */ (T_MASK ^ 0x0000a44e)
+#define T12 /* 0x895cd7be */ (T_MASK ^ 0x76a32841)
+#define T13    0x6b901122
+#define T14 /* 0xfd987193 */ (T_MASK ^ 0x02678e6c)
+#define T15 /* 0xa679438e */ (T_MASK ^ 0x5986bc71)
+#define T16    0x49b40821
+#define T17 /* 0xf61e2562 */ (T_MASK ^ 0x09e1da9d)
+#define T18 /* 0xc040b340 */ (T_MASK ^ 0x3fbf4cbf)
+#define T19    0x265e5a51
+#define T20 /* 0xe9b6c7aa */ (T_MASK ^ 0x16493855)
+#define T21 /* 0xd62f105d */ (T_MASK ^ 0x29d0efa2)
+#define T22    0x02441453
+#define T23 /* 0xd8a1e681 */ (T_MASK ^ 0x275e197e)
+#define T24 /* 0xe7d3fbc8 */ (T_MASK ^ 0x182c0437)
+#define T25    0x21e1cde6
+#define T26 /* 0xc33707d6 */ (T_MASK ^ 0x3cc8f829)
+#define T27 /* 0xf4d50d87 */ (T_MASK ^ 0x0b2af278)
+#define T28    0x455a14ed
+#define T29 /* 0xa9e3e905 */ (T_MASK ^ 0x561c16fa)
+#define T30 /* 0xfcefa3f8 */ (T_MASK ^ 0x03105c07)
+#define T31    0x676f02d9
+#define T32 /* 0x8d2a4c8a */ (T_MASK ^ 0x72d5b375)
+#define T33 /* 0xfffa3942 */ (T_MASK ^ 0x0005c6bd)
+#define T34 /* 0x8771f681 */ (T_MASK ^ 0x788e097e)
+#define T35    0x6d9d6122
+#define T36 /* 0xfde5380c */ (T_MASK ^ 0x021ac7f3)
+#define T37 /* 0xa4beea44 */ (T_MASK ^ 0x5b4115bb)
+#define T38    0x4bdecfa9
+#define T39 /* 0xf6bb4b60 */ (T_MASK ^ 0x0944b49f)
+#define T40 /* 0xbebfbc70 */ (T_MASK ^ 0x4140438f)
+#define T41    0x289b7ec6
+#define T42 /* 0xeaa127fa */ (T_MASK ^ 0x155ed805)
+#define T43 /* 0xd4ef3085 */ (T_MASK ^ 0x2b10cf7a)
+#define T44    0x04881d05
+#define T45 /* 0xd9d4d039 */ (T_MASK ^ 0x262b2fc6)
+#define T46 /* 0xe6db99e5 */ (T_MASK ^ 0x1924661a)
+#define T47    0x1fa27cf8
+#define T48 /* 0xc4ac5665 */ (T_MASK ^ 0x3b53a99a)
+#define T49 /* 0xf4292244 */ (T_MASK ^ 0x0bd6ddbb)
+#define T50    0x432aff97
+#define T51 /* 0xab9423a7 */ (T_MASK ^ 0x546bdc58)
+#define T52 /* 0xfc93a039 */ (T_MASK ^ 0x036c5fc6)
+#define T53    0x655b59c3
+#define T54 /* 0x8f0ccc92 */ (T_MASK ^ 0x70f3336d)
+#define T55 /* 0xffeff47d */ (T_MASK ^ 0x00100b82)
+#define T56 /* 0x85845dd1 */ (T_MASK ^ 0x7a7ba22e)
+#define T57    0x6fa87e4f
+#define T58 /* 0xfe2ce6e0 */ (T_MASK ^ 0x01d3191f)
+#define T59 /* 0xa3014314 */ (T_MASK ^ 0x5cfebceb)
+#define T60    0x4e0811a1
+#define T61 /* 0xf7537e82 */ (T_MASK ^ 0x08ac817d)
+#define T62 /* 0xbd3af235 */ (T_MASK ^ 0x42c50dca)
+#define T63    0x2ad7d2bb
+#define T64 /* 0xeb86d391 */ (T_MASK ^ 0x14792c6e)
+
+
+static void
+md5_process(md5_state_t *pms, const md5_byte_t *data /*[64]*/)
+{
+    md5_word_t
+	a = pms->abcd[0], b = pms->abcd[1],
+	c = pms->abcd[2], d = pms->abcd[3];
+    md5_word_t t;
+#if BYTE_ORDER > 0
+    /* Define storage only for big-endian CPUs. */
+    md5_word_t X[16];
+#else
+    /* Define storage for little-endian or both types of CPUs. */
+    md5_word_t xbuf[16];
+    const md5_word_t *X;
+#endif
+
+    {
+#if BYTE_ORDER == 0
+	/*
+	 * Determine dynamically whether this is a big-endian or
+	 * little-endian machine, since we can use a more efficient
+	 * algorithm on the latter.
+	 */
+	static const int w = 1;
+
+	if (*((const md5_byte_t *)&w)) /* dynamic little-endian */
+#endif
+#if BYTE_ORDER <= 0		/* little-endian */
+	{
+	    /*
+	     * On little-endian machines, we can process properly aligned
+	     * data without copying it.
+	     */
+	    if (!((data - (const md5_byte_t *)0) & 3)) {
+		/* data are properly aligned */
+		X = (const md5_word_t *)data;
+	    } else {
+		/* not aligned */
+		memcpy(xbuf, data, 64);
+		X = xbuf;
+	    }
+	}
+#endif
+#if BYTE_ORDER == 0
+	else			/* dynamic big-endian */
+#endif
+#if BYTE_ORDER >= 0		/* big-endian */
+	{
+	    /*
+	     * On big-endian machines, we must arrange the bytes in the
+	     * right order.
+	     */
+	    const md5_byte_t *xp = data;
+	    int i;
+
+#  if BYTE_ORDER == 0
+	    X = xbuf;		/* (dynamic only) */
+#  else
+#    define xbuf X		/* (static only) */
+#  endif
+	    for (i = 0; i < 16; ++i, xp += 4)
+		xbuf[i] = xp[0] + (xp[1] << 8) + (xp[2] << 16) + (xp[3] << 24);
+	}
+#endif
+    }
+
+#define ROTATE_LEFT(x, n) (((x) << (n)) | ((x) >> (32 - (n))))
+
+    /* Round 1. */
+    /* Let [abcd k s i] denote the operation
+       a = b + ((a + F(b,c,d) + X[k] + T[i]) <<< s). */
+#define F(x, y, z) (((x) & (y)) | (~(x) & (z)))
+#define SET(a, b, c, d, k, s, Ti)\
+  t = a + F(b,c,d) + X[k] + Ti;\
+  a = ROTATE_LEFT(t, s) + b
+    /* Do the following 16 operations. */
+    SET(a, b, c, d,  0,  7,  T1);
+    SET(d, a, b, c,  1, 12,  T2);
+    SET(c, d, a, b,  2, 17,  T3);
+    SET(b, c, d, a,  3, 22,  T4);
+    SET(a, b, c, d,  4,  7,  T5);
+    SET(d, a, b, c,  5, 12,  T6);
+    SET(c, d, a, b,  6, 17,  T7);
+    SET(b, c, d, a,  7, 22,  T8);
+    SET(a, b, c, d,  8,  7,  T9);
+    SET(d, a, b, c,  9, 12, T10);
+    SET(c, d, a, b, 10, 17, T11);
+    SET(b, c, d, a, 11, 22, T12);
+    SET(a, b, c, d, 12,  7, T13);
+    SET(d, a, b, c, 13, 12, T14);
+    SET(c, d, a, b, 14, 17, T15);
+    SET(b, c, d, a, 15, 22, T16);
+#undef SET
+
+     /* Round 2. */
+     /* Let [abcd k s i] denote the operation
+          a = b + ((a + G(b,c,d) + X[k] + T[i]) <<< s). */
+#define G(x, y, z) (((x) & (z)) | ((y) & ~(z)))
+#define SET(a, b, c, d, k, s, Ti)\
+  t = a + G(b,c,d) + X[k] + Ti;\
+  a = ROTATE_LEFT(t, s) + b
+     /* Do the following 16 operations. */
+    SET(a, b, c, d,  1,  5, T17);
+    SET(d, a, b, c,  6,  9, T18);
+    SET(c, d, a, b, 11, 14, T19);
+    SET(b, c, d, a,  0, 20, T20);
+    SET(a, b, c, d,  5,  5, T21);
+    SET(d, a, b, c, 10,  9, T22);
+    SET(c, d, a, b, 15, 14, T23);
+    SET(b, c, d, a,  4, 20, T24);
+    SET(a, b, c, d,  9,  5, T25);
+    SET(d, a, b, c, 14,  9, T26);
+    SET(c, d, a, b,  3, 14, T27);
+    SET(b, c, d, a,  8, 20, T28);
+    SET(a, b, c, d, 13,  5, T29);
+    SET(d, a, b, c,  2,  9, T30);
+    SET(c, d, a, b,  7, 14, T31);
+    SET(b, c, d, a, 12, 20, T32);
+#undef SET
+
+     /* Round 3. */
+     /* Let [abcd k s t] denote the operation
+          a = b + ((a + H(b,c,d) + X[k] + T[i]) <<< s). */
+#define H(x, y, z) ((x) ^ (y) ^ (z))
+#define SET(a, b, c, d, k, s, Ti)\
+  t = a + H(b,c,d) + X[k] + Ti;\
+  a = ROTATE_LEFT(t, s) + b
+     /* Do the following 16 operations. */
+    SET(a, b, c, d,  5,  4, T33);
+    SET(d, a, b, c,  8, 11, T34);
+    SET(c, d, a, b, 11, 16, T35);
+    SET(b, c, d, a, 14, 23, T36);
+    SET(a, b, c, d,  1,  4, T37);
+    SET(d, a, b, c,  4, 11, T38);
+    SET(c, d, a, b,  7, 16, T39);
+    SET(b, c, d, a, 10, 23, T40);
+    SET(a, b, c, d, 13,  4, T41);
+    SET(d, a, b, c,  0, 11, T42);
+    SET(c, d, a, b,  3, 16, T43);
+    SET(b, c, d, a,  6, 23, T44);
+    SET(a, b, c, d,  9,  4, T45);
+    SET(d, a, b, c, 12, 11, T46);
+    SET(c, d, a, b, 15, 16, T47);
+    SET(b, c, d, a,  2, 23, T48);
+#undef SET
+
+     /* Round 4. */
+     /* Let [abcd k s t] denote the operation
+          a = b + ((a + I(b,c,d) + X[k] + T[i]) <<< s). */
+#define I(x, y, z) ((y) ^ ((x) | ~(z)))
+#define SET(a, b, c, d, k, s, Ti)\
+  t = a + I(b,c,d) + X[k] + Ti;\
+  a = ROTATE_LEFT(t, s) + b
+     /* Do the following 16 operations. */
+    SET(a, b, c, d,  0,  6, T49);
+    SET(d, a, b, c,  7, 10, T50);
+    SET(c, d, a, b, 14, 15, T51);
+    SET(b, c, d, a,  5, 21, T52);
+    SET(a, b, c, d, 12,  6, T53);
+    SET(d, a, b, c,  3, 10, T54);
+    SET(c, d, a, b, 10, 15, T55);
+    SET(b, c, d, a,  1, 21, T56);
+    SET(a, b, c, d,  8,  6, T57);
+    SET(d, a, b, c, 15, 10, T58);
+    SET(c, d, a, b,  6, 15, T59);
+    SET(b, c, d, a, 13, 21, T60);
+    SET(a, b, c, d,  4,  6, T61);
+    SET(d, a, b, c, 11, 10, T62);
+    SET(c, d, a, b,  2, 15, T63);
+    SET(b, c, d, a,  9, 21, T64);
+#undef SET
+
+     /* Then perform the following additions. (That is increment each
+        of the four registers by the value it had before this block
+        was started.) */
+    pms->abcd[0] += a;
+    pms->abcd[1] += b;
+    pms->abcd[2] += c;
+    pms->abcd[3] += d;
+}
+
+void
+md5_init(md5_state_t *pms)
+{
+    pms->count[0] = pms->count[1] = 0;
+    pms->abcd[0] = 0x67452301;
+    pms->abcd[1] = /*0xefcdab89*/ T_MASK ^ 0x10325476;
+    pms->abcd[2] = /*0x98badcfe*/ T_MASK ^ 0x67452301;
+    pms->abcd[3] = 0x10325476;
+}
+
+void
+md5_append(md5_state_t *pms, const md5_byte_t *data, int nbytes)
+{
+    const md5_byte_t *p = data;
+    int left = nbytes;
+    int offset = (pms->count[0] >> 3) & 63;
+    md5_word_t nbits = (md5_word_t)(nbytes << 3);
+
+    if (nbytes <= 0)
+	return;
+
+    /* Update the message length. */
+    pms->count[1] += nbytes >> 29;
+    pms->count[0] += nbits;
+    if (pms->count[0] < nbits)
+	pms->count[1]++;
+
+    /* Process an initial partial block. */
+    if (offset) {
+	int copy = (offset + nbytes > 64 ? 64 - offset : nbytes);
+
+	memcpy(pms->buf + offset, p, copy);
+	if (offset + copy < 64)
+	    return;
+	p += copy;
+	left -= copy;
+	md5_process(pms, pms->buf);
+    }
+
+    /* Process full blocks. */
+    for (; left >= 64; p += 64, left -= 64)
+	md5_process(pms, p);
+
+    /* Process a final partial block. */
+    if (left)
+	memcpy(pms->buf, p, left);
+}
+
+void
+md5_finish(md5_state_t *pms, md5_byte_t digest[16])
+{
+    static const md5_byte_t pad[64] = {
+	0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
+    };
+    md5_byte_t data[8];
+    int i;
+
+    /* Save the length before padding. */
+    for (i = 0; i < 8; ++i)
+	data[i] = (md5_byte_t)(pms->count[i >> 2] >> ((i & 3) << 3));
+    /* Pad to 56 bytes mod 64. */
+    md5_append(pms, pad, ((55 - (pms->count[0] >> 3)) & 63) + 1);
+    /* Append the length. */
+    md5_append(pms, data, 8);
+    for (i = 0; i < 16; ++i)
+	digest[i] = (md5_byte_t)(pms->abcd[i >> 2] >> ((i & 3) << 3));
+}
diff --git a/src/Modules/md5.h b/src/Modules/md5.h
new file mode 100644
--- /dev/null
+++ b/src/Modules/md5.h
@@ -0,0 +1,91 @@
+/*
+  Copyright (C) 1999, 2002 Aladdin Enterprises.  All rights reserved.
+
+  This software is provided 'as-is', without any express or implied
+  warranty.  In no event will the authors be held liable for any damages
+  arising from the use of this software.
+
+  Permission is granted to anyone to use this software for any purpose,
+  including commercial applications, and to alter it and redistribute it
+  freely, subject to the following restrictions:
+
+  1. The origin of this software must not be misrepresented; you must not
+     claim that you wrote the original software. If you use this software
+     in a product, an acknowledgment in the product documentation would be
+     appreciated but is not required.
+  2. Altered source versions must be plainly marked as such, and must not be
+     misrepresented as being the original software.
+  3. This notice may not be removed or altered from any source distribution.
+
+  L. Peter Deutsch
+  ghost at aladdin.com
+
+ */
+/* $Id: md5.h 43594 2006-04-03 16:27:50Z matthias.klose $ */
+/*
+  Independent implementation of MD5 (RFC 1321).
+
+  This code implements the MD5 Algorithm defined in RFC 1321, whose
+  text is available at
+	http://www.ietf.org/rfc/rfc1321.txt
+  The code is derived from the text of the RFC, including the test suite
+  (section A.5) but excluding the rest of Appendix A.  It does not include
+  any code or documentation that is identified in the RFC as being
+  copyrighted.
+
+  The original and principal author of md5.h is L. Peter Deutsch
+  <ghost at aladdin.com>.  Other authors are noted in the change history
+  that follows (in reverse chronological order):
+
+  2002-04-13 lpd Removed support for non-ANSI compilers; removed
+	references to Ghostscript; clarified derivation from RFC 1321;
+	now handles byte order either statically or dynamically.
+  1999-11-04 lpd Edited comments slightly for automatic TOC extraction.
+  1999-10-18 lpd Fixed typo in header comment (ansi2knr rather than md5);
+	added conditionalization for C++ compilation from Martin
+	Purschke <purschke at bnl.gov>.
+  1999-05-03 lpd Original version.
+ */
+
+#ifndef md5_INCLUDED
+#  define md5_INCLUDED
+
+/*
+ * This package supports both compile-time and run-time determination of CPU
+ * byte order.  If ARCH_IS_BIG_ENDIAN is defined as 0, the code will be
+ * compiled to run only on little-endian CPUs; if ARCH_IS_BIG_ENDIAN is
+ * defined as non-zero, the code will be compiled to run only on big-endian
+ * CPUs; if ARCH_IS_BIG_ENDIAN is not defined, the code will be compiled to
+ * run on either big- or little-endian CPUs, but will run slightly less
+ * efficiently on either one than if ARCH_IS_BIG_ENDIAN is defined.
+ */
+
+typedef unsigned char md5_byte_t; /* 8-bit byte */
+typedef unsigned int md5_word_t; /* 32-bit word */
+
+/* Define the state of the MD5 Algorithm. */
+typedef struct md5_state_s {
+    md5_word_t count[2];	/* message length in bits, lsw first */
+    md5_word_t abcd[4];		/* digest buffer */
+    md5_byte_t buf[64];		/* accumulate block */
+} md5_state_t;
+
+#ifdef __cplusplus
+extern "C" 
+{
+#endif
+
+/* Initialize the algorithm. */
+void md5_init(md5_state_t *pms);
+
+/* Append a string to the message. */
+void md5_append(md5_state_t *pms, const md5_byte_t *data, int nbytes);
+
+/* Finish the message and return the digest. */
+void md5_finish(md5_state_t *pms, md5_byte_t digest[16]);
+
+#ifdef __cplusplus
+}  /* end extern "C" */
+#endif
+
+#endif /* md5_INCLUDED */
diff --git a/src/Modules/md5module.c b/src/Modules/md5module.c
new file mode 100644
--- /dev/null
+++ b/src/Modules/md5module.c
@@ -0,0 +1,312 @@
+
+/* MD5 module */
+
+/* This module provides an interface to the RSA Data Security,
+   Inc. MD5 Message-Digest Algorithm, described in RFC 1321.
+   It requires the files md5c.c and md5.h (which are slightly changed
+   from the versions in the RFC to avoid the "global.h" file.) */
+
+
+/* MD5 objects */
+
+#include "Python.h"
+#include "structmember.h"
+#include "md5.h"
+
+typedef struct {
+	PyObject_HEAD
+        md5_state_t	md5;		/* the context holder */
+} md5object;
+
+static PyTypeObject MD5type;
+
+#define is_md5object(v)		((v)->ob_type == &MD5type)
+
+static md5object *
+newmd5object(void)
+{
+	md5object *md5p;
+
+	md5p = PyObject_New(md5object, &MD5type);
+	if (md5p == NULL)
+		return NULL;
+
+	md5_init(&md5p->md5);	/* actual initialisation */
+	return md5p;
+}
+
+
+/* MD5 methods */
+
+static void
+md5_dealloc(md5object *md5p)
+{
+	PyObject_Del(md5p);
+}
+
+
+/* MD5 methods-as-attributes */
+
+static PyObject *
+md5_update(md5object *self, PyObject *args)
+{
+	unsigned char *cp;
+	int len;
+
+	if (!PyArg_ParseTuple(args, "s#:update", &cp, &len))
+		return NULL;
+
+	md5_append(&self->md5, cp, len);
+
+	Py_INCREF(Py_None);
+	return Py_None;
+}
+
+PyDoc_STRVAR(update_doc,
+"update (arg)\n\
+\n\
+Update the md5 object with the string arg. Repeated calls are\n\
+equivalent to a single call with the concatenation of all the\n\
+arguments.");
+
+
+static PyObject *
+md5_digest(md5object *self)
+{
+ 	md5_state_t mdContext;
+	unsigned char aDigest[16];
+
+	/* make a temporary copy, and perform the final */
+	mdContext = self->md5;
+	md5_finish(&mdContext, aDigest);
+
+	return PyString_FromStringAndSize((char *)aDigest, 16);
+}
+
+PyDoc_STRVAR(digest_doc,
+"digest() -> string\n\
+\n\
+Return the digest of the strings passed to the update() method so\n\
+far. This is a 16-byte string which may contain non-ASCII characters,\n\
+including null bytes.");
+
+
+static PyObject *
+md5_hexdigest(md5object *self)
+{
+ 	md5_state_t mdContext;
+	unsigned char digest[16];
+	unsigned char hexdigest[32];
+	int i, j;
+
+	/* make a temporary copy, and perform the final */
+	mdContext = self->md5;
+	md5_finish(&mdContext, digest);
+
+	/* Make hex version of the digest */
+	for(i=j=0; i<16; i++) {
+		char c;
+		c = (digest[i] >> 4) & 0xf;
+		c = (c>9) ? c+'a'-10 : c + '0';
+		hexdigest[j++] = c;
+		c = (digest[i] & 0xf);
+		c = (c>9) ? c+'a'-10 : c + '0';
+		hexdigest[j++] = c;
+	}
+	return PyString_FromStringAndSize((char*)hexdigest, 32);
+}
+
+
+PyDoc_STRVAR(hexdigest_doc,
+"hexdigest() -> string\n\
+\n\
+Like digest(), but returns the digest as a string of hexadecimal digits.");
+
+
+static PyObject *
+md5_copy(md5object *self)
+{
+	md5object *md5p;
+
+	if ((md5p = newmd5object()) == NULL)
+		return NULL;
+
+	md5p->md5 = self->md5;
+
+	return (PyObject *)md5p;
+}
+
+PyDoc_STRVAR(copy_doc,
+"copy() -> md5 object\n\
+\n\
+Return a copy (``clone'') of the md5 object.");
+
+
+static PyMethodDef md5_methods[] = {
+	{"update",    (PyCFunction)md5_update,    METH_VARARGS, update_doc},
+	{"digest",    (PyCFunction)md5_digest,    METH_NOARGS,  digest_doc},
+	{"hexdigest", (PyCFunction)md5_hexdigest, METH_NOARGS,  hexdigest_doc},
+	{"copy",      (PyCFunction)md5_copy,      METH_NOARGS,  copy_doc},
+	{NULL, NULL}			     /* sentinel */
+};
+
+static PyObject *
+md5_get_block_size(PyObject *self, void *closure)
+{
+    return PyInt_FromLong(64);
+}
+
+static PyObject *
+md5_get_digest_size(PyObject *self, void *closure)
+{
+    return PyInt_FromLong(16);
+}
+
+static PyObject *
+md5_get_name(PyObject *self, void *closure)
+{
+    return PyString_FromStringAndSize("MD5", 3);
+}
+
+static PyGetSetDef md5_getseters[] = {
+    {"digest_size",
+     (getter)md5_get_digest_size, NULL,
+     NULL,
+     NULL},
+    {"block_size",
+     (getter)md5_get_block_size, NULL,
+     NULL,
+     NULL},
+    {"name",
+     (getter)md5_get_name, NULL,
+     NULL,
+     NULL},
+    /* the old md5 and sha modules support 'digest_size' as in PEP 247.
+     * the old sha module also supported 'digestsize'.  ugh. */
+    {"digestsize",
+     (getter)md5_get_digest_size, NULL,
+     NULL,
+     NULL},
+    {NULL}  /* Sentinel */
+};
+
+
+PyDoc_STRVAR(module_doc,
+"This module implements the interface to RSA's MD5 message digest\n\
+algorithm (see also Internet RFC 1321). Its use is quite\n\
+straightforward: use the new() to create an md5 object. You can now\n\
+feed this object with arbitrary strings using the update() method, and\n\
+at any point you can ask it for the digest (a strong kind of 128-bit\n\
+checksum, a.k.a. ``fingerprint'') of the concatenation of the strings\n\
+fed to it so far using the digest() method.\n\
+\n\
+Functions:\n\
+\n\
+new([arg]) -- return a new md5 object, initialized with arg if provided\n\
+md5([arg]) -- DEPRECATED, same as new, but for compatibility\n\
+\n\
+Special Objects:\n\
+\n\
+MD5Type -- type object for md5 objects");
+
+PyDoc_STRVAR(md5type_doc,
+"An md5 represents the object used to calculate the MD5 checksum of a\n\
+string of information.\n\
+\n\
+Methods:\n\
+\n\
+update() -- updates the current digest with an additional string\n\
+digest() -- return the current digest value\n\
+hexdigest() -- return the current digest as a string of hexadecimal digits\n\
+copy() -- return a copy of the current md5 object");
+
+static PyTypeObject MD5type = {
+	PyObject_HEAD_INIT(NULL)
+	0,			  /*ob_size*/
+	"_md5.md5",		  /*tp_name*/
+	sizeof(md5object),	  /*tp_size*/
+	0,			  /*tp_itemsize*/
+	/* methods */
+	(destructor)md5_dealloc,  /*tp_dealloc*/
+	0,			  /*tp_print*/
+	0,                        /*tp_getattr*/
+	0,			  /*tp_setattr*/
+	0,			  /*tp_compare*/
+	0,			  /*tp_repr*/
+        0,			  /*tp_as_number*/
+	0,                        /*tp_as_sequence*/
+	0,			  /*tp_as_mapping*/
+	0, 			  /*tp_hash*/
+	0,			  /*tp_call*/
+	0,			  /*tp_str*/
+	0,			  /*tp_getattro*/
+	0,			  /*tp_setattro*/
+	0,	                  /*tp_as_buffer*/
+	Py_TPFLAGS_DEFAULT,	  /*tp_flags*/
+	md5type_doc,		  /*tp_doc*/
+        0,                        /*tp_traverse*/
+        0,			  /*tp_clear*/
+        0,			  /*tp_richcompare*/
+        0,			  /*tp_weaklistoffset*/
+        0,			  /*tp_iter*/
+        0,			  /*tp_iternext*/
+        md5_methods,	          /*tp_methods*/
+        0,      	          /*tp_members*/
+        md5_getseters,            /*tp_getset*/
+};
+
+
+/* MD5 functions */
+
+static PyObject *
+MD5_new(PyObject *self, PyObject *args)
+{
+	md5object *md5p;
+	unsigned char *cp = NULL;
+	int len = 0;
+
+	if (!PyArg_ParseTuple(args, "|s#:new", &cp, &len))
+		return NULL;
+
+	if ((md5p = newmd5object()) == NULL)
+		return NULL;
+
+	if (cp)
+		md5_append(&md5p->md5, cp, len);
+
+	return (PyObject *)md5p;
+}
+
+PyDoc_STRVAR(new_doc,
+"new([arg]) -> md5 object\n\
+\n\
+Return a new md5 object. If arg is present, the method call update(arg)\n\
+is made.");
+
+
+/* List of functions exported by this module */
+
+static PyMethodDef md5_functions[] = {
+	{"new",		(PyCFunction)MD5_new, METH_VARARGS, new_doc},
+	{NULL,		NULL}	/* Sentinel */
+};
+
+
+/* Initialize this module. */
+
+PyMODINIT_FUNC
+init_md5(void)
+{
+	PyObject *m, *d;
+
+        MD5type.ob_type = &PyType_Type;
+        if (PyType_Ready(&MD5type) < 0)
+            return;
+	m = Py_InitModule3("_md5", md5_functions, module_doc);
+	if (m == NULL)
+	    return;
+	d = PyModule_GetDict(m);
+	PyDict_SetItemString(d, "MD5Type", (PyObject *)&MD5type);
+	PyModule_AddIntConstant(m, "digest_size", 16);
+	/* No need to check the error here, the caller will do that */
+}
diff --git a/src/Modules/sha256module.c b/src/Modules/sha256module.c
new file mode 100644
--- /dev/null
+++ b/src/Modules/sha256module.c
@@ -0,0 +1,701 @@
+/* SHA256 module */
+
+/* This module provides an interface to NIST's SHA-256 and SHA-224 Algorithms */
+
+/* See below for information about the original code this module was
+   based upon. Additional work performed by:
+
+   Andrew Kuchling (amk at amk.ca)
+   Greg Stein (gstein at lyra.org)
+   Trevor Perrin (trevp at trevp.net)
+
+   Copyright (C) 2005   Gregory P. Smith (greg at krypto.org)
+   Licensed to PSF under a Contributor Agreement.
+
+*/
+
+/* SHA objects */
+
+#include "Python.h"
+#include "structmember.h"
+
+
+/* Endianness testing and definitions */
+#define TestEndianness(variable) {int i=1; variable=PCT_BIG_ENDIAN;\
+	if (*((char*)&i)==1) variable=PCT_LITTLE_ENDIAN;}
+
+#define PCT_LITTLE_ENDIAN 1
+#define PCT_BIG_ENDIAN 0
+
+/* Some useful types */
+
+typedef unsigned char SHA_BYTE;
+
+#if SIZEOF_INT == 4
+typedef unsigned int SHA_INT32;	/* 32-bit integer */
+#else
+/* not defined. compilation will die. */
+#endif
+
+/* The SHA block size and message digest sizes, in bytes */
+
+#define SHA_BLOCKSIZE    64
+#define SHA_DIGESTSIZE  32
+
+/* The structure for storing SHA info */
+
+typedef struct {
+    PyObject_HEAD
+    SHA_INT32 digest[8];		/* Message digest */
+    SHA_INT32 count_lo, count_hi;	/* 64-bit bit count */
+    SHA_BYTE data[SHA_BLOCKSIZE];	/* SHA data buffer */
+    int Endianness;
+    int local;				/* unprocessed amount in data */
+    int digestsize;
+} SHAobject;
+
+/* When run on a little-endian CPU we need to perform byte reversal on an
+   array of longwords. */
+
+static void longReverse(SHA_INT32 *buffer, int byteCount, int Endianness)
+{
+    SHA_INT32 value;
+
+    if ( Endianness == PCT_BIG_ENDIAN )
+	return;
+
+    byteCount /= sizeof(*buffer);
+    while (byteCount--) {
+        value = *buffer;
+        value = ( ( value & 0xFF00FF00L ) >> 8  ) | \
+                ( ( value & 0x00FF00FFL ) << 8 );
+        *buffer++ = ( value << 16 ) | ( value >> 16 );
+    }
+}
+
+static void SHAcopy(SHAobject *src, SHAobject *dest)
+{
+    dest->Endianness = src->Endianness;
+    dest->local = src->local;
+    dest->digestsize = src->digestsize;
+    dest->count_lo = src->count_lo;
+    dest->count_hi = src->count_hi;
+    memcpy(dest->digest, src->digest, sizeof(src->digest));
+    memcpy(dest->data, src->data, sizeof(src->data));
+}
+
+
+/* ------------------------------------------------------------------------
+ *
+ * This code for the SHA-256 algorithm was noted as public domain. The
+ * original headers are pasted below.
+ *
+ * Several changes have been made to make it more compatible with the
+ * Python environment and desired interface.
+ *
+ */
+
+/* LibTomCrypt, modular cryptographic library -- Tom St Denis
+ *
+ * LibTomCrypt is a library that provides various cryptographic
+ * algorithms in a highly modular and flexible manner.
+ *
+ * The library is free for all purposes without any express
+ * gurantee it works.
+ *
+ * Tom St Denis, tomstdenis at iahu.ca, http://libtomcrypt.org
+ */
+
+
+/* SHA256 by Tom St Denis */
+
+/* Various logical functions */
+#define ROR(x, y)\
+( ((((unsigned long)(x)&0xFFFFFFFFUL)>>(unsigned long)((y)&31)) | \
+((unsigned long)(x)<<(unsigned long)(32-((y)&31)))) & 0xFFFFFFFFUL)
+#define Ch(x,y,z)       (z ^ (x & (y ^ z)))
+#define Maj(x,y,z)      (((x | y) & z) | (x & y)) 
+#define S(x, n)         ROR((x),(n))
+#define R(x, n)         (((x)&0xFFFFFFFFUL)>>(n))
+#define Sigma0(x)       (S(x, 2) ^ S(x, 13) ^ S(x, 22))
+#define Sigma1(x)       (S(x, 6) ^ S(x, 11) ^ S(x, 25))
+#define Gamma0(x)       (S(x, 7) ^ S(x, 18) ^ R(x, 3))
+#define Gamma1(x)       (S(x, 17) ^ S(x, 19) ^ R(x, 10))
+
+
+static void
+sha_transform(SHAobject *sha_info)
+{
+    int i;
+	SHA_INT32 S[8], W[64], t0, t1;
+
+    memcpy(W, sha_info->data, sizeof(sha_info->data));
+    longReverse(W, (int)sizeof(sha_info->data), sha_info->Endianness);
+
+    for (i = 16; i < 64; ++i) {
+		W[i] = Gamma1(W[i - 2]) + W[i - 7] + Gamma0(W[i - 15]) + W[i - 16];
+    }
+    for (i = 0; i < 8; ++i) {
+        S[i] = sha_info->digest[i];
+    }
+
+    /* Compress */
+#define RND(a,b,c,d,e,f,g,h,i,ki)                    \
+     t0 = h + Sigma1(e) + Ch(e, f, g) + ki + W[i];   \
+     t1 = Sigma0(a) + Maj(a, b, c);                  \
+     d += t0;                                        \
+     h  = t0 + t1;
+
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],0,0x428a2f98);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],1,0x71374491);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],2,0xb5c0fbcf);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],3,0xe9b5dba5);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],4,0x3956c25b);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],5,0x59f111f1);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],6,0x923f82a4);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],7,0xab1c5ed5);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],8,0xd807aa98);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],9,0x12835b01);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],10,0x243185be);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],11,0x550c7dc3);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],12,0x72be5d74);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],13,0x80deb1fe);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],14,0x9bdc06a7);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],15,0xc19bf174);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],16,0xe49b69c1);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],17,0xefbe4786);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],18,0x0fc19dc6);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],19,0x240ca1cc);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],20,0x2de92c6f);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],21,0x4a7484aa);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],22,0x5cb0a9dc);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],23,0x76f988da);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],24,0x983e5152);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],25,0xa831c66d);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],26,0xb00327c8);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],27,0xbf597fc7);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],28,0xc6e00bf3);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],29,0xd5a79147);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],30,0x06ca6351);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],31,0x14292967);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],32,0x27b70a85);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],33,0x2e1b2138);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],34,0x4d2c6dfc);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],35,0x53380d13);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],36,0x650a7354);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],37,0x766a0abb);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],38,0x81c2c92e);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],39,0x92722c85);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],40,0xa2bfe8a1);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],41,0xa81a664b);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],42,0xc24b8b70);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],43,0xc76c51a3);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],44,0xd192e819);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],45,0xd6990624);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],46,0xf40e3585);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],47,0x106aa070);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],48,0x19a4c116);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],49,0x1e376c08);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],50,0x2748774c);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],51,0x34b0bcb5);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],52,0x391c0cb3);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],53,0x4ed8aa4a);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],54,0x5b9cca4f);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],55,0x682e6ff3);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],56,0x748f82ee);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],57,0x78a5636f);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],58,0x84c87814);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],59,0x8cc70208);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],60,0x90befffa);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],61,0xa4506ceb);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],62,0xbef9a3f7);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],63,0xc67178f2);
+
+#undef RND     
+    
+    /* feedback */
+    for (i = 0; i < 8; i++) {
+        sha_info->digest[i] = sha_info->digest[i] + S[i];
+    }
+
+}
+
+
+
+/* initialize the SHA digest */
+
+static void
+sha_init(SHAobject *sha_info)
+{
+    TestEndianness(sha_info->Endianness)
+    sha_info->digest[0] = 0x6A09E667L;
+    sha_info->digest[1] = 0xBB67AE85L;
+    sha_info->digest[2] = 0x3C6EF372L;
+    sha_info->digest[3] = 0xA54FF53AL;
+    sha_info->digest[4] = 0x510E527FL;
+    sha_info->digest[5] = 0x9B05688CL;
+    sha_info->digest[6] = 0x1F83D9ABL;
+    sha_info->digest[7] = 0x5BE0CD19L;
+    sha_info->count_lo = 0L;
+    sha_info->count_hi = 0L;
+    sha_info->local = 0;
+    sha_info->digestsize = 32;
+}
+
+static void
+sha224_init(SHAobject *sha_info)
+{
+    TestEndianness(sha_info->Endianness)
+    sha_info->digest[0] = 0xc1059ed8L;
+    sha_info->digest[1] = 0x367cd507L;
+    sha_info->digest[2] = 0x3070dd17L;
+    sha_info->digest[3] = 0xf70e5939L;
+    sha_info->digest[4] = 0xffc00b31L;
+    sha_info->digest[5] = 0x68581511L;
+    sha_info->digest[6] = 0x64f98fa7L;
+    sha_info->digest[7] = 0xbefa4fa4L;
+    sha_info->count_lo = 0L;
+    sha_info->count_hi = 0L;
+    sha_info->local = 0;
+    sha_info->digestsize = 28;
+}
+
+
+/* update the SHA digest */
+
+static void
+sha_update(SHAobject *sha_info, SHA_BYTE *buffer, int count)
+{
+    int i;
+    SHA_INT32 clo;
+
+    clo = sha_info->count_lo + ((SHA_INT32) count << 3);
+    if (clo < sha_info->count_lo) {
+        ++sha_info->count_hi;
+    }
+    sha_info->count_lo = clo;
+    sha_info->count_hi += (SHA_INT32) count >> 29;
+    if (sha_info->local) {
+        i = SHA_BLOCKSIZE - sha_info->local;
+        if (i > count) {
+            i = count;
+        }
+        memcpy(((SHA_BYTE *) sha_info->data) + sha_info->local, buffer, i);
+        count -= i;
+        buffer += i;
+        sha_info->local += i;
+        if (sha_info->local == SHA_BLOCKSIZE) {
+            sha_transform(sha_info);
+        }
+        else {
+            return;
+        }
+    }
+    while (count >= SHA_BLOCKSIZE) {
+        memcpy(sha_info->data, buffer, SHA_BLOCKSIZE);
+        buffer += SHA_BLOCKSIZE;
+        count -= SHA_BLOCKSIZE;
+        sha_transform(sha_info);
+    }
+    memcpy(sha_info->data, buffer, count);
+    sha_info->local = count;
+}
+
+/* finish computing the SHA digest */
+
+static void
+sha_final(unsigned char digest[SHA_DIGESTSIZE], SHAobject *sha_info)
+{
+    int count;
+    SHA_INT32 lo_bit_count, hi_bit_count;
+
+    lo_bit_count = sha_info->count_lo;
+    hi_bit_count = sha_info->count_hi;
+    count = (int) ((lo_bit_count >> 3) & 0x3f);
+    ((SHA_BYTE *) sha_info->data)[count++] = 0x80;
+    if (count > SHA_BLOCKSIZE - 8) {
+	memset(((SHA_BYTE *) sha_info->data) + count, 0,
+	       SHA_BLOCKSIZE - count);
+	sha_transform(sha_info);
+	memset((SHA_BYTE *) sha_info->data, 0, SHA_BLOCKSIZE - 8);
+    }
+    else {
+	memset(((SHA_BYTE *) sha_info->data) + count, 0,
+	       SHA_BLOCKSIZE - 8 - count);
+    }
+
+    /* GJS: note that we add the hi/lo in big-endian. sha_transform will
+       swap these values into host-order. */
+    sha_info->data[56] = (hi_bit_count >> 24) & 0xff;
+    sha_info->data[57] = (hi_bit_count >> 16) & 0xff;
+    sha_info->data[58] = (hi_bit_count >>  8) & 0xff;
+    sha_info->data[59] = (hi_bit_count >>  0) & 0xff;
+    sha_info->data[60] = (lo_bit_count >> 24) & 0xff;
+    sha_info->data[61] = (lo_bit_count >> 16) & 0xff;
+    sha_info->data[62] = (lo_bit_count >>  8) & 0xff;
+    sha_info->data[63] = (lo_bit_count >>  0) & 0xff;
+    sha_transform(sha_info);
+    digest[ 0] = (unsigned char) ((sha_info->digest[0] >> 24) & 0xff);
+    digest[ 1] = (unsigned char) ((sha_info->digest[0] >> 16) & 0xff);
+    digest[ 2] = (unsigned char) ((sha_info->digest[0] >>  8) & 0xff);
+    digest[ 3] = (unsigned char) ((sha_info->digest[0]      ) & 0xff);
+    digest[ 4] = (unsigned char) ((sha_info->digest[1] >> 24) & 0xff);
+    digest[ 5] = (unsigned char) ((sha_info->digest[1] >> 16) & 0xff);
+    digest[ 6] = (unsigned char) ((sha_info->digest[1] >>  8) & 0xff);
+    digest[ 7] = (unsigned char) ((sha_info->digest[1]      ) & 0xff);
+    digest[ 8] = (unsigned char) ((sha_info->digest[2] >> 24) & 0xff);
+    digest[ 9] = (unsigned char) ((sha_info->digest[2] >> 16) & 0xff);
+    digest[10] = (unsigned char) ((sha_info->digest[2] >>  8) & 0xff);
+    digest[11] = (unsigned char) ((sha_info->digest[2]      ) & 0xff);
+    digest[12] = (unsigned char) ((sha_info->digest[3] >> 24) & 0xff);
+    digest[13] = (unsigned char) ((sha_info->digest[3] >> 16) & 0xff);
+    digest[14] = (unsigned char) ((sha_info->digest[3] >>  8) & 0xff);
+    digest[15] = (unsigned char) ((sha_info->digest[3]      ) & 0xff);
+    digest[16] = (unsigned char) ((sha_info->digest[4] >> 24) & 0xff);
+    digest[17] = (unsigned char) ((sha_info->digest[4] >> 16) & 0xff);
+    digest[18] = (unsigned char) ((sha_info->digest[4] >>  8) & 0xff);
+    digest[19] = (unsigned char) ((sha_info->digest[4]      ) & 0xff);
+    digest[20] = (unsigned char) ((sha_info->digest[5] >> 24) & 0xff);
+    digest[21] = (unsigned char) ((sha_info->digest[5] >> 16) & 0xff);
+    digest[22] = (unsigned char) ((sha_info->digest[5] >>  8) & 0xff);
+    digest[23] = (unsigned char) ((sha_info->digest[5]      ) & 0xff);
+    digest[24] = (unsigned char) ((sha_info->digest[6] >> 24) & 0xff);
+    digest[25] = (unsigned char) ((sha_info->digest[6] >> 16) & 0xff);
+    digest[26] = (unsigned char) ((sha_info->digest[6] >>  8) & 0xff);
+    digest[27] = (unsigned char) ((sha_info->digest[6]      ) & 0xff);
+    digest[28] = (unsigned char) ((sha_info->digest[7] >> 24) & 0xff);
+    digest[29] = (unsigned char) ((sha_info->digest[7] >> 16) & 0xff);
+    digest[30] = (unsigned char) ((sha_info->digest[7] >>  8) & 0xff);
+    digest[31] = (unsigned char) ((sha_info->digest[7]      ) & 0xff);
+}
+
+/*
+ * End of copied SHA code.
+ *
+ * ------------------------------------------------------------------------
+ */
+
+static PyTypeObject SHA224type;
+static PyTypeObject SHA256type;
+
+
+static SHAobject *
+newSHA224object(void)
+{
+    return (SHAobject *)PyObject_New(SHAobject, &SHA224type);
+}
+
+static SHAobject *
+newSHA256object(void)
+{
+    return (SHAobject *)PyObject_New(SHAobject, &SHA256type);
+}
+
+/* Internal methods for a hash object */
+
+static void
+SHA_dealloc(PyObject *ptr)
+{
+    PyObject_Del(ptr);
+}
+
+
+/* External methods for a hash object */
+
+PyDoc_STRVAR(SHA256_copy__doc__, "Return a copy of the hash object.");
+
+static PyObject *
+SHA256_copy(SHAobject *self, PyObject *unused)
+{
+    SHAobject *newobj;
+
+    if (((PyObject*)self)->ob_type == &SHA256type) {
+        if ( (newobj = newSHA256object())==NULL)
+            return NULL;
+    } else {
+        if ( (newobj = newSHA224object())==NULL)
+            return NULL;
+    }
+
+    SHAcopy(self, newobj);
+    return (PyObject *)newobj;
+}
+
+PyDoc_STRVAR(SHA256_digest__doc__,
+"Return the digest value as a string of binary data.");
+
+static PyObject *
+SHA256_digest(SHAobject *self, PyObject *unused)
+{
+    unsigned char digest[SHA_DIGESTSIZE];
+    SHAobject temp;
+
+    SHAcopy(self, &temp);
+    sha_final(digest, &temp);
+    return PyString_FromStringAndSize((const char *)digest, self->digestsize);
+}
+
+PyDoc_STRVAR(SHA256_hexdigest__doc__,
+"Return the digest value as a string of hexadecimal digits.");
+
+static PyObject *
+SHA256_hexdigest(SHAobject *self, PyObject *unused)
+{
+    unsigned char digest[SHA_DIGESTSIZE];
+    SHAobject temp;
+    PyObject *retval;
+    char *hex_digest;
+    int i, j;
+
+    /* Get the raw (binary) digest value */
+    SHAcopy(self, &temp);
+    sha_final(digest, &temp);
+
+    /* Create a new string */
+    retval = PyString_FromStringAndSize(NULL, self->digestsize * 2);
+    if (!retval)
+	    return NULL;
+    hex_digest = PyString_AsString(retval);
+    if (!hex_digest) {
+	    Py_DECREF(retval);
+	    return NULL;
+    }
+
+    /* Make hex version of the digest */
+    for(i=j=0; i<self->digestsize; i++) {
+        char c;
+        c = (digest[i] >> 4) & 0xf;
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+        c = (digest[i] & 0xf);
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+    }
+    return retval;
+}
+
+PyDoc_STRVAR(SHA256_update__doc__,
+"Update this hash object's state with the provided string.");
+
+static PyObject *
+SHA256_update(SHAobject *self, PyObject *args)
+{
+    unsigned char *cp;
+    int len;
+
+    if (!PyArg_ParseTuple(args, "s#:update", &cp, &len))
+        return NULL;
+
+    sha_update(self, cp, len);
+
+    Py_INCREF(Py_None);
+    return Py_None;
+}
+
+static PyMethodDef SHA_methods[] = {
+    {"copy",	  (PyCFunction)SHA256_copy,      METH_NOARGS,  SHA256_copy__doc__},
+    {"digest",	  (PyCFunction)SHA256_digest,    METH_NOARGS,  SHA256_digest__doc__},
+    {"hexdigest", (PyCFunction)SHA256_hexdigest, METH_NOARGS,  SHA256_hexdigest__doc__},
+    {"update",	  (PyCFunction)SHA256_update,    METH_VARARGS, SHA256_update__doc__},
+    {NULL,	  NULL}		/* sentinel */
+};
+
+static PyObject *
+SHA256_get_block_size(PyObject *self, void *closure)
+{
+    return PyInt_FromLong(SHA_BLOCKSIZE);
+}
+
+static PyObject *
+SHA256_get_name(PyObject *self, void *closure)
+{
+    if (((SHAobject *)self)->digestsize == 32)
+        return PyString_FromStringAndSize("SHA256", 6);
+    else
+        return PyString_FromStringAndSize("SHA224", 6);
+}
+
+static PyGetSetDef SHA_getseters[] = {
+    {"block_size",
+     (getter)SHA256_get_block_size, NULL,
+     NULL,
+     NULL},
+    {"name",
+     (getter)SHA256_get_name, NULL,
+     NULL,
+     NULL},
+    {NULL}  /* Sentinel */
+};
+
+static PyMemberDef SHA_members[] = {
+    {"digest_size", T_INT, offsetof(SHAobject, digestsize), READONLY, NULL},
+    /* the old md5 and sha modules support 'digest_size' as in PEP 247.
+     * the old sha module also supported 'digestsize'.  ugh. */
+    {"digestsize", T_INT, offsetof(SHAobject, digestsize), READONLY, NULL},
+    {NULL}  /* Sentinel */
+};
+
+static PyTypeObject SHA224type = {
+    PyObject_HEAD_INIT(NULL)
+    0,			/*ob_size*/
+    "_sha256.sha224",	/*tp_name*/
+    sizeof(SHAobject),	/*tp_size*/
+    0,			/*tp_itemsize*/
+    /* methods */
+    SHA_dealloc,	/*tp_dealloc*/
+    0,			/*tp_print*/
+    0,          	/*tp_getattr*/
+    0,                  /*tp_setattr*/
+    0,                  /*tp_compare*/
+    0,                  /*tp_repr*/
+    0,                  /*tp_as_number*/
+    0,                  /*tp_as_sequence*/
+    0,                  /*tp_as_mapping*/
+    0,                  /*tp_hash*/
+    0,                  /*tp_call*/
+    0,                  /*tp_str*/
+    0,                  /*tp_getattro*/
+    0,                  /*tp_setattro*/
+    0,                  /*tp_as_buffer*/
+    Py_TPFLAGS_DEFAULT, /*tp_flags*/
+    0,                  /*tp_doc*/
+    0,                  /*tp_traverse*/
+    0,			/*tp_clear*/
+    0,			/*tp_richcompare*/
+    0,			/*tp_weaklistoffset*/
+    0,			/*tp_iter*/
+    0,			/*tp_iternext*/
+    SHA_methods,	/* tp_methods */
+    SHA_members,	/* tp_members */
+    SHA_getseters,      /* tp_getset */
+};
+
+static PyTypeObject SHA256type = {
+    PyObject_HEAD_INIT(NULL)
+    0,			/*ob_size*/
+    "_sha256.sha256",	/*tp_name*/
+    sizeof(SHAobject),	/*tp_size*/
+    0,			/*tp_itemsize*/
+    /* methods */
+    SHA_dealloc,	/*tp_dealloc*/
+    0,			/*tp_print*/
+    0,          	/*tp_getattr*/
+    0,                  /*tp_setattr*/
+    0,                  /*tp_compare*/
+    0,                  /*tp_repr*/
+    0,                  /*tp_as_number*/
+    0,                  /*tp_as_sequence*/
+    0,                  /*tp_as_mapping*/
+    0,                  /*tp_hash*/
+    0,                  /*tp_call*/
+    0,                  /*tp_str*/
+    0,                  /*tp_getattro*/
+    0,                  /*tp_setattro*/
+    0,                  /*tp_as_buffer*/
+    Py_TPFLAGS_DEFAULT, /*tp_flags*/
+    0,                  /*tp_doc*/
+    0,                  /*tp_traverse*/
+    0,			/*tp_clear*/
+    0,			/*tp_richcompare*/
+    0,			/*tp_weaklistoffset*/
+    0,			/*tp_iter*/
+    0,			/*tp_iternext*/
+    SHA_methods,	/* tp_methods */
+    SHA_members,	/* tp_members */
+    SHA_getseters,      /* tp_getset */
+};
+
+
+/* The single module-level function: new() */
+
+PyDoc_STRVAR(SHA256_new__doc__,
+"Return a new SHA-256 hash object; optionally initialized with a string.");
+
+static PyObject *
+SHA256_new(PyObject *self, PyObject *args, PyObject *kwdict)
+{
+    static char *kwlist[] = {"string", NULL};
+    SHAobject *new;
+    unsigned char *cp = NULL;
+    int len;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwdict, "|s#:new", kwlist,
+                                     &cp, &len)) {
+        return NULL;
+    }
+
+    if ((new = newSHA256object()) == NULL)
+        return NULL;
+
+    sha_init(new);
+
+    if (PyErr_Occurred()) {
+        Py_DECREF(new);
+        return NULL;
+    }
+    if (cp)
+        sha_update(new, cp, len);
+
+    return (PyObject *)new;
+}
+
+PyDoc_STRVAR(SHA224_new__doc__,
+"Return a new SHA-224 hash object; optionally initialized with a string.");
+
+static PyObject *
+SHA224_new(PyObject *self, PyObject *args, PyObject *kwdict)
+{
+    static char *kwlist[] = {"string", NULL};
+    SHAobject *new;
+    unsigned char *cp = NULL;
+    int len;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwdict, "|s#:new", kwlist,
+                                     &cp, &len)) {
+        return NULL;
+    }
+
+    if ((new = newSHA224object()) == NULL)
+        return NULL;
+
+    sha224_init(new);
+
+    if (PyErr_Occurred()) {
+        Py_DECREF(new);
+        return NULL;
+    }
+    if (cp)
+        sha_update(new, cp, len);
+
+    return (PyObject *)new;
+}
+
+
+/* List of functions exported by this module */
+
+static struct PyMethodDef SHA_functions[] = {
+    {"sha256", (PyCFunction)SHA256_new, METH_VARARGS|METH_KEYWORDS, SHA256_new__doc__},
+    {"sha224", (PyCFunction)SHA224_new, METH_VARARGS|METH_KEYWORDS, SHA224_new__doc__},
+    {NULL,	NULL}		 /* Sentinel */
+};
+
+
+/* Initialize this module. */
+
+#define insint(n,v) { PyModule_AddIntConstant(m,n,v); }
+
+PyMODINIT_FUNC
+init_sha256(void)
+{
+    PyObject *m;
+
+    SHA224type.ob_type = &PyType_Type;
+    if (PyType_Ready(&SHA224type) < 0)
+        return;
+    SHA256type.ob_type = &PyType_Type;
+    if (PyType_Ready(&SHA256type) < 0)
+        return;
+    m = Py_InitModule("_sha256", SHA_functions);
+    if (m == NULL)
+	return;
+}
diff --git a/src/Modules/sha512module.c b/src/Modules/sha512module.c
new file mode 100644
--- /dev/null
+++ b/src/Modules/sha512module.c
@@ -0,0 +1,769 @@
+/* SHA512 module */
+
+/* This module provides an interface to NIST's SHA-512 and SHA-384 Algorithms */
+
+/* See below for information about the original code this module was
+   based upon. Additional work performed by:
+
+   Andrew Kuchling (amk at amk.ca)
+   Greg Stein (gstein at lyra.org)
+   Trevor Perrin (trevp at trevp.net)
+
+   Copyright (C) 2005   Gregory P. Smith (greg at krypto.org)
+   Licensed to PSF under a Contributor Agreement.
+
+*/
+
+/* SHA objects */
+
+#include "Python.h"
+#include "structmember.h"
+
+#ifdef PY_LONG_LONG /* If no PY_LONG_LONG, don't compile anything! */
+
+/* Endianness testing and definitions */
+#define TestEndianness(variable) {int i=1; variable=PCT_BIG_ENDIAN;\
+	if (*((char*)&i)==1) variable=PCT_LITTLE_ENDIAN;}
+
+#define PCT_LITTLE_ENDIAN 1
+#define PCT_BIG_ENDIAN 0
+
+/* Some useful types */
+
+typedef unsigned char SHA_BYTE;
+
+#if SIZEOF_INT == 4
+typedef unsigned int SHA_INT32;	/* 32-bit integer */
+typedef unsigned PY_LONG_LONG SHA_INT64;	/* 64-bit integer */
+#else
+/* not defined. compilation will die. */
+#endif
+
+/* The SHA block size and message digest sizes, in bytes */
+
+#define SHA_BLOCKSIZE   128
+#define SHA_DIGESTSIZE  64
+
+/* The structure for storing SHA info */
+
+typedef struct {
+    PyObject_HEAD
+    SHA_INT64 digest[8];		/* Message digest */
+    SHA_INT32 count_lo, count_hi;	/* 64-bit bit count */
+    SHA_BYTE data[SHA_BLOCKSIZE];	/* SHA data buffer */
+    int Endianness;
+    int local;				/* unprocessed amount in data */
+    int digestsize;
+} SHAobject;
+
+/* When run on a little-endian CPU we need to perform byte reversal on an
+   array of longwords. */
+
+static void longReverse(SHA_INT64 *buffer, int byteCount, int Endianness)
+{
+    SHA_INT64 value;
+
+    if ( Endianness == PCT_BIG_ENDIAN )
+	return;
+
+    byteCount /= sizeof(*buffer);
+    while (byteCount--) {
+        value = *buffer;
+
+		((unsigned char*)buffer)[0] = (unsigned char)(value >> 56) & 0xff;
+		((unsigned char*)buffer)[1] = (unsigned char)(value >> 48) & 0xff;
+		((unsigned char*)buffer)[2] = (unsigned char)(value >> 40) & 0xff;
+		((unsigned char*)buffer)[3] = (unsigned char)(value >> 32) & 0xff;
+		((unsigned char*)buffer)[4] = (unsigned char)(value >> 24) & 0xff;
+		((unsigned char*)buffer)[5] = (unsigned char)(value >> 16) & 0xff;
+		((unsigned char*)buffer)[6] = (unsigned char)(value >>  8) & 0xff;
+		((unsigned char*)buffer)[7] = (unsigned char)(value      ) & 0xff;
+        
+		buffer++;
+    }
+}
+
+static void SHAcopy(SHAobject *src, SHAobject *dest)
+{
+    dest->Endianness = src->Endianness;
+    dest->local = src->local;
+    dest->digestsize = src->digestsize;
+    dest->count_lo = src->count_lo;
+    dest->count_hi = src->count_hi;
+    memcpy(dest->digest, src->digest, sizeof(src->digest));
+    memcpy(dest->data, src->data, sizeof(src->data));
+}
+
+
+/* ------------------------------------------------------------------------
+ *
+ * This code for the SHA-512 algorithm was noted as public domain. The
+ * original headers are pasted below.
+ *
+ * Several changes have been made to make it more compatible with the
+ * Python environment and desired interface.
+ *
+ */
+
+/* LibTomCrypt, modular cryptographic library -- Tom St Denis
+ *
+ * LibTomCrypt is a library that provides various cryptographic
+ * algorithms in a highly modular and flexible manner.
+ *
+ * The library is free for all purposes without any express
+ * gurantee it works.
+ *
+ * Tom St Denis, tomstdenis at iahu.ca, http://libtomcrypt.org
+ */
+
+
+/* SHA512 by Tom St Denis */
+
+/* Various logical functions */
+#define ROR64(x, y) \
+    ( ((((x) & 0xFFFFFFFFFFFFFFFFULL)>>((unsigned PY_LONG_LONG)(y) & 63)) | \
+      ((x)<<((unsigned PY_LONG_LONG)(64-((y) & 63))))) & 0xFFFFFFFFFFFFFFFFULL)
+#define Ch(x,y,z)       (z ^ (x & (y ^ z)))
+#define Maj(x,y,z)      (((x | y) & z) | (x & y)) 
+#define S(x, n)         ROR64((x),(n))
+#define R(x, n)         (((x) & 0xFFFFFFFFFFFFFFFFULL) >> ((unsigned PY_LONG_LONG)n))
+#define Sigma0(x)       (S(x, 28) ^ S(x, 34) ^ S(x, 39))
+#define Sigma1(x)       (S(x, 14) ^ S(x, 18) ^ S(x, 41))
+#define Gamma0(x)       (S(x, 1) ^ S(x, 8) ^ R(x, 7))
+#define Gamma1(x)       (S(x, 19) ^ S(x, 61) ^ R(x, 6))
+
+
+static void
+sha512_transform(SHAobject *sha_info)
+{
+    int i;
+    SHA_INT64 S[8], W[80], t0, t1;
+
+    memcpy(W, sha_info->data, sizeof(sha_info->data));
+    longReverse(W, (int)sizeof(sha_info->data), sha_info->Endianness);
+
+    for (i = 16; i < 80; ++i) {
+		W[i] = Gamma1(W[i - 2]) + W[i - 7] + Gamma0(W[i - 15]) + W[i - 16];
+    }
+    for (i = 0; i < 8; ++i) {
+        S[i] = sha_info->digest[i];
+    }
+
+    /* Compress */
+#define RND(a,b,c,d,e,f,g,h,i,ki)                    \
+     t0 = h + Sigma1(e) + Ch(e, f, g) + ki + W[i];   \
+     t1 = Sigma0(a) + Maj(a, b, c);                  \
+     d += t0;                                        \
+     h  = t0 + t1;
+
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],0,0x428a2f98d728ae22ULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],1,0x7137449123ef65cdULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],2,0xb5c0fbcfec4d3b2fULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],3,0xe9b5dba58189dbbcULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],4,0x3956c25bf348b538ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],5,0x59f111f1b605d019ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],6,0x923f82a4af194f9bULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],7,0xab1c5ed5da6d8118ULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],8,0xd807aa98a3030242ULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],9,0x12835b0145706fbeULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],10,0x243185be4ee4b28cULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],11,0x550c7dc3d5ffb4e2ULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],12,0x72be5d74f27b896fULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],13,0x80deb1fe3b1696b1ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],14,0x9bdc06a725c71235ULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],15,0xc19bf174cf692694ULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],16,0xe49b69c19ef14ad2ULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],17,0xefbe4786384f25e3ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],18,0x0fc19dc68b8cd5b5ULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],19,0x240ca1cc77ac9c65ULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],20,0x2de92c6f592b0275ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],21,0x4a7484aa6ea6e483ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],22,0x5cb0a9dcbd41fbd4ULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],23,0x76f988da831153b5ULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],24,0x983e5152ee66dfabULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],25,0xa831c66d2db43210ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],26,0xb00327c898fb213fULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],27,0xbf597fc7beef0ee4ULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],28,0xc6e00bf33da88fc2ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],29,0xd5a79147930aa725ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],30,0x06ca6351e003826fULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],31,0x142929670a0e6e70ULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],32,0x27b70a8546d22ffcULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],33,0x2e1b21385c26c926ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],34,0x4d2c6dfc5ac42aedULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],35,0x53380d139d95b3dfULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],36,0x650a73548baf63deULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],37,0x766a0abb3c77b2a8ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],38,0x81c2c92e47edaee6ULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],39,0x92722c851482353bULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],40,0xa2bfe8a14cf10364ULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],41,0xa81a664bbc423001ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],42,0xc24b8b70d0f89791ULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],43,0xc76c51a30654be30ULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],44,0xd192e819d6ef5218ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],45,0xd69906245565a910ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],46,0xf40e35855771202aULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],47,0x106aa07032bbd1b8ULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],48,0x19a4c116b8d2d0c8ULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],49,0x1e376c085141ab53ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],50,0x2748774cdf8eeb99ULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],51,0x34b0bcb5e19b48a8ULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],52,0x391c0cb3c5c95a63ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],53,0x4ed8aa4ae3418acbULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],54,0x5b9cca4f7763e373ULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],55,0x682e6ff3d6b2b8a3ULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],56,0x748f82ee5defb2fcULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],57,0x78a5636f43172f60ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],58,0x84c87814a1f0ab72ULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],59,0x8cc702081a6439ecULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],60,0x90befffa23631e28ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],61,0xa4506cebde82bde9ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],62,0xbef9a3f7b2c67915ULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],63,0xc67178f2e372532bULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],64,0xca273eceea26619cULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],65,0xd186b8c721c0c207ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],66,0xeada7dd6cde0eb1eULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],67,0xf57d4f7fee6ed178ULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],68,0x06f067aa72176fbaULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],69,0x0a637dc5a2c898a6ULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],70,0x113f9804bef90daeULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],71,0x1b710b35131c471bULL);
+    RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],72,0x28db77f523047d84ULL);
+    RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],73,0x32caab7b40c72493ULL);
+    RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],74,0x3c9ebe0a15c9bebcULL);
+    RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],75,0x431d67c49c100d4cULL);
+    RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],76,0x4cc5d4becb3e42b6ULL);
+    RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],77,0x597f299cfc657e2aULL);
+    RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],78,0x5fcb6fab3ad6faecULL);
+    RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],79,0x6c44198c4a475817ULL);
+
+#undef RND     
+    
+    /* feedback */
+    for (i = 0; i < 8; i++) {
+        sha_info->digest[i] = sha_info->digest[i] + S[i];
+    }
+
+}
+
+
+
+/* initialize the SHA digest */
+
+static void
+sha512_init(SHAobject *sha_info)
+{
+    TestEndianness(sha_info->Endianness)
+    sha_info->digest[0] = 0x6a09e667f3bcc908ULL;
+    sha_info->digest[1] = 0xbb67ae8584caa73bULL;
+    sha_info->digest[2] = 0x3c6ef372fe94f82bULL;
+    sha_info->digest[3] = 0xa54ff53a5f1d36f1ULL;
+    sha_info->digest[4] = 0x510e527fade682d1ULL;
+    sha_info->digest[5] = 0x9b05688c2b3e6c1fULL;
+    sha_info->digest[6] = 0x1f83d9abfb41bd6bULL;
+    sha_info->digest[7] = 0x5be0cd19137e2179ULL;
+    sha_info->count_lo = 0L;
+    sha_info->count_hi = 0L;
+    sha_info->local = 0;
+    sha_info->digestsize = 64;
+}
+
+static void
+sha384_init(SHAobject *sha_info)
+{
+    TestEndianness(sha_info->Endianness)
+    sha_info->digest[0] = 0xcbbb9d5dc1059ed8ULL;
+    sha_info->digest[1] = 0x629a292a367cd507ULL;
+    sha_info->digest[2] = 0x9159015a3070dd17ULL;
+    sha_info->digest[3] = 0x152fecd8f70e5939ULL;
+    sha_info->digest[4] = 0x67332667ffc00b31ULL;
+    sha_info->digest[5] = 0x8eb44a8768581511ULL;
+    sha_info->digest[6] = 0xdb0c2e0d64f98fa7ULL;
+    sha_info->digest[7] = 0x47b5481dbefa4fa4ULL;
+    sha_info->count_lo = 0L;
+    sha_info->count_hi = 0L;
+    sha_info->local = 0;
+    sha_info->digestsize = 48;
+}
+
+
+/* update the SHA digest */
+
+static void
+sha512_update(SHAobject *sha_info, SHA_BYTE *buffer, int count)
+{
+    int i;
+    SHA_INT32 clo;
+
+    clo = sha_info->count_lo + ((SHA_INT32) count << 3);
+    if (clo < sha_info->count_lo) {
+        ++sha_info->count_hi;
+    }
+    sha_info->count_lo = clo;
+    sha_info->count_hi += (SHA_INT32) count >> 29;
+    if (sha_info->local) {
+        i = SHA_BLOCKSIZE - sha_info->local;
+        if (i > count) {
+            i = count;
+        }
+        memcpy(((SHA_BYTE *) sha_info->data) + sha_info->local, buffer, i);
+        count -= i;
+        buffer += i;
+        sha_info->local += i;
+        if (sha_info->local == SHA_BLOCKSIZE) {
+            sha512_transform(sha_info);
+        }
+        else {
+            return;
+        }
+    }
+    while (count >= SHA_BLOCKSIZE) {
+        memcpy(sha_info->data, buffer, SHA_BLOCKSIZE);
+        buffer += SHA_BLOCKSIZE;
+        count -= SHA_BLOCKSIZE;
+        sha512_transform(sha_info);
+    }
+    memcpy(sha_info->data, buffer, count);
+    sha_info->local = count;
+}
+
+/* finish computing the SHA digest */
+
+static void
+sha512_final(unsigned char digest[SHA_DIGESTSIZE], SHAobject *sha_info)
+{
+    int count;
+    SHA_INT32 lo_bit_count, hi_bit_count;
+
+    lo_bit_count = sha_info->count_lo;
+    hi_bit_count = sha_info->count_hi;
+    count = (int) ((lo_bit_count >> 3) & 0x7f);
+    ((SHA_BYTE *) sha_info->data)[count++] = 0x80;
+    if (count > SHA_BLOCKSIZE - 16) {
+	memset(((SHA_BYTE *) sha_info->data) + count, 0,
+	       SHA_BLOCKSIZE - count);
+	sha512_transform(sha_info);
+	memset((SHA_BYTE *) sha_info->data, 0, SHA_BLOCKSIZE - 16);
+    }
+    else {
+	memset(((SHA_BYTE *) sha_info->data) + count, 0,
+	       SHA_BLOCKSIZE - 16 - count);
+    }
+
+    /* GJS: note that we add the hi/lo in big-endian. sha512_transform will
+       swap these values into host-order. */
+    sha_info->data[112] = 0;
+    sha_info->data[113] = 0;
+    sha_info->data[114] = 0;
+    sha_info->data[115] = 0;
+    sha_info->data[116] = 0;
+    sha_info->data[117] = 0;
+    sha_info->data[118] = 0;
+    sha_info->data[119] = 0;
+    sha_info->data[120] = (hi_bit_count >> 24) & 0xff;
+    sha_info->data[121] = (hi_bit_count >> 16) & 0xff;
+    sha_info->data[122] = (hi_bit_count >>  8) & 0xff;
+    sha_info->data[123] = (hi_bit_count >>  0) & 0xff;
+    sha_info->data[124] = (lo_bit_count >> 24) & 0xff;
+    sha_info->data[125] = (lo_bit_count >> 16) & 0xff;
+    sha_info->data[126] = (lo_bit_count >>  8) & 0xff;
+    sha_info->data[127] = (lo_bit_count >>  0) & 0xff;
+    sha512_transform(sha_info);
+    digest[ 0] = (unsigned char) ((sha_info->digest[0] >> 56) & 0xff);
+    digest[ 1] = (unsigned char) ((sha_info->digest[0] >> 48) & 0xff);
+    digest[ 2] = (unsigned char) ((sha_info->digest[0] >> 40) & 0xff);
+    digest[ 3] = (unsigned char) ((sha_info->digest[0] >> 32) & 0xff);
+    digest[ 4] = (unsigned char) ((sha_info->digest[0] >> 24) & 0xff);
+    digest[ 5] = (unsigned char) ((sha_info->digest[0] >> 16) & 0xff);
+    digest[ 6] = (unsigned char) ((sha_info->digest[0] >>  8) & 0xff);
+    digest[ 7] = (unsigned char) ((sha_info->digest[0]      ) & 0xff);
+    digest[ 8] = (unsigned char) ((sha_info->digest[1] >> 56) & 0xff);
+    digest[ 9] = (unsigned char) ((sha_info->digest[1] >> 48) & 0xff);
+    digest[10] = (unsigned char) ((sha_info->digest[1] >> 40) & 0xff);
+    digest[11] = (unsigned char) ((sha_info->digest[1] >> 32) & 0xff);
+    digest[12] = (unsigned char) ((sha_info->digest[1] >> 24) & 0xff);
+    digest[13] = (unsigned char) ((sha_info->digest[1] >> 16) & 0xff);
+    digest[14] = (unsigned char) ((sha_info->digest[1] >>  8) & 0xff);
+    digest[15] = (unsigned char) ((sha_info->digest[1]      ) & 0xff);
+    digest[16] = (unsigned char) ((sha_info->digest[2] >> 56) & 0xff);
+    digest[17] = (unsigned char) ((sha_info->digest[2] >> 48) & 0xff);
+    digest[18] = (unsigned char) ((sha_info->digest[2] >> 40) & 0xff);
+    digest[19] = (unsigned char) ((sha_info->digest[2] >> 32) & 0xff);
+    digest[20] = (unsigned char) ((sha_info->digest[2] >> 24) & 0xff);
+    digest[21] = (unsigned char) ((sha_info->digest[2] >> 16) & 0xff);
+    digest[22] = (unsigned char) ((sha_info->digest[2] >>  8) & 0xff);
+    digest[23] = (unsigned char) ((sha_info->digest[2]      ) & 0xff);
+    digest[24] = (unsigned char) ((sha_info->digest[3] >> 56) & 0xff);
+    digest[25] = (unsigned char) ((sha_info->digest[3] >> 48) & 0xff);
+    digest[26] = (unsigned char) ((sha_info->digest[3] >> 40) & 0xff);
+    digest[27] = (unsigned char) ((sha_info->digest[3] >> 32) & 0xff);
+    digest[28] = (unsigned char) ((sha_info->digest[3] >> 24) & 0xff);
+    digest[29] = (unsigned char) ((sha_info->digest[3] >> 16) & 0xff);
+    digest[30] = (unsigned char) ((sha_info->digest[3] >>  8) & 0xff);
+    digest[31] = (unsigned char) ((sha_info->digest[3]      ) & 0xff);
+    digest[32] = (unsigned char) ((sha_info->digest[4] >> 56) & 0xff);
+    digest[33] = (unsigned char) ((sha_info->digest[4] >> 48) & 0xff);
+    digest[34] = (unsigned char) ((sha_info->digest[4] >> 40) & 0xff);
+    digest[35] = (unsigned char) ((sha_info->digest[4] >> 32) & 0xff);
+    digest[36] = (unsigned char) ((sha_info->digest[4] >> 24) & 0xff);
+    digest[37] = (unsigned char) ((sha_info->digest[4] >> 16) & 0xff);
+    digest[38] = (unsigned char) ((sha_info->digest[4] >>  8) & 0xff);
+    digest[39] = (unsigned char) ((sha_info->digest[4]      ) & 0xff);
+    digest[40] = (unsigned char) ((sha_info->digest[5] >> 56) & 0xff);
+    digest[41] = (unsigned char) ((sha_info->digest[5] >> 48) & 0xff);
+    digest[42] = (unsigned char) ((sha_info->digest[5] >> 40) & 0xff);
+    digest[43] = (unsigned char) ((sha_info->digest[5] >> 32) & 0xff);
+    digest[44] = (unsigned char) ((sha_info->digest[5] >> 24) & 0xff);
+    digest[45] = (unsigned char) ((sha_info->digest[5] >> 16) & 0xff);
+    digest[46] = (unsigned char) ((sha_info->digest[5] >>  8) & 0xff);
+    digest[47] = (unsigned char) ((sha_info->digest[5]      ) & 0xff);
+    digest[48] = (unsigned char) ((sha_info->digest[6] >> 56) & 0xff);
+    digest[49] = (unsigned char) ((sha_info->digest[6] >> 48) & 0xff);
+    digest[50] = (unsigned char) ((sha_info->digest[6] >> 40) & 0xff);
+    digest[51] = (unsigned char) ((sha_info->digest[6] >> 32) & 0xff);
+    digest[52] = (unsigned char) ((sha_info->digest[6] >> 24) & 0xff);
+    digest[53] = (unsigned char) ((sha_info->digest[6] >> 16) & 0xff);
+    digest[54] = (unsigned char) ((sha_info->digest[6] >>  8) & 0xff);
+    digest[55] = (unsigned char) ((sha_info->digest[6]      ) & 0xff);
+    digest[56] = (unsigned char) ((sha_info->digest[7] >> 56) & 0xff);
+    digest[57] = (unsigned char) ((sha_info->digest[7] >> 48) & 0xff);
+    digest[58] = (unsigned char) ((sha_info->digest[7] >> 40) & 0xff);
+    digest[59] = (unsigned char) ((sha_info->digest[7] >> 32) & 0xff);
+    digest[60] = (unsigned char) ((sha_info->digest[7] >> 24) & 0xff);
+    digest[61] = (unsigned char) ((sha_info->digest[7] >> 16) & 0xff);
+    digest[62] = (unsigned char) ((sha_info->digest[7] >>  8) & 0xff);
+    digest[63] = (unsigned char) ((sha_info->digest[7]      ) & 0xff);
+}
+
+/*
+ * End of copied SHA code.
+ *
+ * ------------------------------------------------------------------------
+ */
+
+static PyTypeObject SHA384type;
+static PyTypeObject SHA512type;
+
+
+static SHAobject *
+newSHA384object(void)
+{
+    return (SHAobject *)PyObject_New(SHAobject, &SHA384type);
+}
+
+static SHAobject *
+newSHA512object(void)
+{
+    return (SHAobject *)PyObject_New(SHAobject, &SHA512type);
+}
+
+/* Internal methods for a hash object */
+
+static void
+SHA512_dealloc(PyObject *ptr)
+{
+    PyObject_Del(ptr);
+}
+
+
+/* External methods for a hash object */
+
+PyDoc_STRVAR(SHA512_copy__doc__, "Return a copy of the hash object.");
+
+static PyObject *
+SHA512_copy(SHAobject *self, PyObject *unused)
+{
+    SHAobject *newobj;
+
+    if (((PyObject*)self)->ob_type == &SHA512type) {
+        if ( (newobj = newSHA512object())==NULL)
+            return NULL;
+    } else {
+        if ( (newobj = newSHA384object())==NULL)
+            return NULL;
+    }
+
+    SHAcopy(self, newobj);
+    return (PyObject *)newobj;
+}
+
+PyDoc_STRVAR(SHA512_digest__doc__,
+"Return the digest value as a string of binary data.");
+
+static PyObject *
+SHA512_digest(SHAobject *self, PyObject *unused)
+{
+    unsigned char digest[SHA_DIGESTSIZE];
+    SHAobject temp;
+
+    SHAcopy(self, &temp);
+    sha512_final(digest, &temp);
+    return PyString_FromStringAndSize((const char *)digest, self->digestsize);
+}
+
+PyDoc_STRVAR(SHA512_hexdigest__doc__,
+"Return the digest value as a string of hexadecimal digits.");
+
+static PyObject *
+SHA512_hexdigest(SHAobject *self, PyObject *unused)
+{
+    unsigned char digest[SHA_DIGESTSIZE];
+    SHAobject temp;
+    PyObject *retval;
+    char *hex_digest;
+    int i, j;
+
+    /* Get the raw (binary) digest value */
+    SHAcopy(self, &temp);
+    sha512_final(digest, &temp);
+
+    /* Create a new string */
+    retval = PyString_FromStringAndSize(NULL, self->digestsize * 2);
+    if (!retval)
+	    return NULL;
+    hex_digest = PyString_AsString(retval);
+    if (!hex_digest) {
+	    Py_DECREF(retval);
+	    return NULL;
+    }
+
+    /* Make hex version of the digest */
+    for (i=j=0; i<self->digestsize; i++) {
+        char c;
+        c = (digest[i] >> 4) & 0xf;
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+        c = (digest[i] & 0xf);
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+    }
+    return retval;
+}
+
+PyDoc_STRVAR(SHA512_update__doc__,
+"Update this hash object's state with the provided string.");
+
+static PyObject *
+SHA512_update(SHAobject *self, PyObject *args)
+{
+    unsigned char *cp;
+    int len;
+
+    if (!PyArg_ParseTuple(args, "s#:update", &cp, &len))
+        return NULL;
+
+    sha512_update(self, cp, len);
+
+    Py_INCREF(Py_None);
+    return Py_None;
+}
+
+static PyMethodDef SHA_methods[] = {
+    {"copy",	  (PyCFunction)SHA512_copy,      METH_NOARGS, SHA512_copy__doc__},
+    {"digest",	  (PyCFunction)SHA512_digest,    METH_NOARGS, SHA512_digest__doc__},
+    {"hexdigest", (PyCFunction)SHA512_hexdigest, METH_NOARGS, SHA512_hexdigest__doc__},
+    {"update",	  (PyCFunction)SHA512_update,    METH_VARARGS, SHA512_update__doc__},
+    {NULL,	  NULL}		/* sentinel */
+};
+
+static PyObject *
+SHA512_get_block_size(PyObject *self, void *closure)
+{
+    return PyInt_FromLong(SHA_BLOCKSIZE);
+}
+
+static PyObject *
+SHA512_get_name(PyObject *self, void *closure)
+{
+    if (((SHAobject *)self)->digestsize == 64)
+        return PyString_FromStringAndSize("SHA512", 6);
+    else
+        return PyString_FromStringAndSize("SHA384", 6);
+}
+
+static PyGetSetDef SHA_getseters[] = {
+    {"block_size",
+     (getter)SHA512_get_block_size, NULL,
+     NULL,
+     NULL},
+    {"name",
+     (getter)SHA512_get_name, NULL,
+     NULL,
+     NULL},
+    {NULL}  /* Sentinel */
+};
+
+static PyMemberDef SHA_members[] = {
+    {"digest_size", T_INT, offsetof(SHAobject, digestsize), READONLY, NULL},
+    /* the old md5 and sha modules support 'digest_size' as in PEP 247.
+     * the old sha module also supported 'digestsize'.  ugh. */
+    {"digestsize", T_INT, offsetof(SHAobject, digestsize), READONLY, NULL},
+    {NULL}  /* Sentinel */
+};
+
+static PyTypeObject SHA384type = {
+    PyObject_HEAD_INIT(NULL)
+    0,			/*ob_size*/
+    "_sha512.sha384",	/*tp_name*/
+    sizeof(SHAobject),	/*tp_size*/
+    0,			/*tp_itemsize*/
+    /* methods */
+    SHA512_dealloc,	/*tp_dealloc*/
+    0,			/*tp_print*/
+    0,          	/*tp_getattr*/
+    0,                  /*tp_setattr*/
+    0,                  /*tp_compare*/
+    0,                  /*tp_repr*/
+    0,                  /*tp_as_number*/
+    0,                  /*tp_as_sequence*/
+    0,                  /*tp_as_mapping*/
+    0,                  /*tp_hash*/
+    0,                  /*tp_call*/
+    0,                  /*tp_str*/
+    0,                  /*tp_getattro*/
+    0,                  /*tp_setattro*/
+    0,                  /*tp_as_buffer*/
+    Py_TPFLAGS_DEFAULT, /*tp_flags*/
+    0,                  /*tp_doc*/
+    0,                  /*tp_traverse*/
+    0,			/*tp_clear*/
+    0,			/*tp_richcompare*/
+    0,			/*tp_weaklistoffset*/
+    0,			/*tp_iter*/
+    0,			/*tp_iternext*/
+    SHA_methods,	/* tp_methods */
+    SHA_members,	/* tp_members */
+    SHA_getseters,      /* tp_getset */
+};
+
+static PyTypeObject SHA512type = {
+    PyObject_HEAD_INIT(NULL)
+    0,			/*ob_size*/
+    "_sha512.sha512",	/*tp_name*/
+    sizeof(SHAobject),	/*tp_size*/
+    0,			/*tp_itemsize*/
+    /* methods */
+    SHA512_dealloc,	/*tp_dealloc*/
+    0,			/*tp_print*/
+    0,          	/*tp_getattr*/
+    0,                  /*tp_setattr*/
+    0,                  /*tp_compare*/
+    0,                  /*tp_repr*/
+    0,                  /*tp_as_number*/
+    0,                  /*tp_as_sequence*/
+    0,                  /*tp_as_mapping*/
+    0,                  /*tp_hash*/
+    0,                  /*tp_call*/
+    0,                  /*tp_str*/
+    0,                  /*tp_getattro*/
+    0,                  /*tp_setattro*/
+    0,                  /*tp_as_buffer*/
+    Py_TPFLAGS_DEFAULT, /*tp_flags*/
+    0,                  /*tp_doc*/
+    0,                  /*tp_traverse*/
+    0,			/*tp_clear*/
+    0,			/*tp_richcompare*/
+    0,			/*tp_weaklistoffset*/
+    0,			/*tp_iter*/
+    0,			/*tp_iternext*/
+    SHA_methods,	/* tp_methods */
+    SHA_members,	/* tp_members */
+    SHA_getseters,      /* tp_getset */
+};
+
+
+/* The single module-level function: new() */
+
+PyDoc_STRVAR(SHA512_new__doc__,
+"Return a new SHA-512 hash object; optionally initialized with a string.");
+
+static PyObject *
+SHA512_new(PyObject *self, PyObject *args, PyObject *kwdict)
+{
+    static char *kwlist[] = {"string", NULL};
+    SHAobject *new;
+    unsigned char *cp = NULL;
+    int len;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwdict, "|s#:new", kwlist,
+                                     &cp, &len)) {
+        return NULL;
+    }
+
+    if ((new = newSHA512object()) == NULL)
+        return NULL;
+
+    sha512_init(new);
+
+    if (PyErr_Occurred()) {
+        Py_DECREF(new);
+        return NULL;
+    }
+    if (cp)
+        sha512_update(new, cp, len);
+
+    return (PyObject *)new;
+}
+
+PyDoc_STRVAR(SHA384_new__doc__,
+"Return a new SHA-384 hash object; optionally initialized with a string.");
+
+static PyObject *
+SHA384_new(PyObject *self, PyObject *args, PyObject *kwdict)
+{
+    static char *kwlist[] = {"string", NULL};
+    SHAobject *new;
+    unsigned char *cp = NULL;
+    int len;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwdict, "|s#:new", kwlist,
+                                     &cp, &len)) {
+        return NULL;
+    }
+
+    if ((new = newSHA384object()) == NULL)
+        return NULL;
+
+    sha384_init(new);
+
+    if (PyErr_Occurred()) {
+        Py_DECREF(new);
+        return NULL;
+    }
+    if (cp)
+        sha512_update(new, cp, len);
+
+    return (PyObject *)new;
+}
+
+
+/* List of functions exported by this module */
+
+static struct PyMethodDef SHA_functions[] = {
+    {"sha512", (PyCFunction)SHA512_new, METH_VARARGS|METH_KEYWORDS, SHA512_new__doc__},
+    {"sha384", (PyCFunction)SHA384_new, METH_VARARGS|METH_KEYWORDS, SHA384_new__doc__},
+    {NULL,	NULL}		 /* Sentinel */
+};
+
+
+/* Initialize this module. */
+
+#define insint(n,v) { PyModule_AddIntConstant(m,n,v); }
+
+PyMODINIT_FUNC
+init_sha512(void)
+{
+    PyObject *m;
+
+    SHA384type.ob_type = &PyType_Type;
+    if (PyType_Ready(&SHA384type) < 0)
+        return;
+    SHA512type.ob_type = &PyType_Type;
+    if (PyType_Ready(&SHA512type) < 0)
+        return;
+    m = Py_InitModule("_sha512", SHA_functions);
+    if (m == NULL)
+	return;
+}
+
+#endif
diff --git a/src/Modules/shamodule.c b/src/Modules/shamodule.c
new file mode 100644
--- /dev/null
+++ b/src/Modules/shamodule.c
@@ -0,0 +1,593 @@
+/* SHA module */
+
+/* This module provides an interface to NIST's Secure Hash Algorithm */
+
+/* See below for information about the original code this module was
+   based upon. Additional work performed by:
+
+   Andrew Kuchling (amk at amk.ca)
+   Greg Stein (gstein at lyra.org)
+
+   Copyright (C) 2005   Gregory P. Smith (greg at krypto.org)
+   Licensed to PSF under a Contributor Agreement.
+
+*/
+
+/* SHA objects */
+
+#include "Python.h"
+#include "structmember.h"
+
+
+/* Endianness testing and definitions */
+#define TestEndianness(variable) {int i=1; variable=PCT_BIG_ENDIAN;\
+	if (*((char*)&i)==1) variable=PCT_LITTLE_ENDIAN;}
+
+#define PCT_LITTLE_ENDIAN 1
+#define PCT_BIG_ENDIAN 0
+
+/* Some useful types */
+
+typedef unsigned char SHA_BYTE;
+
+#if SIZEOF_INT == 4
+typedef unsigned int SHA_INT32;	/* 32-bit integer */
+#else
+/* not defined. compilation will die. */
+#endif
+
+/* The SHA block size and message digest sizes, in bytes */
+
+#define SHA_BLOCKSIZE    64
+#define SHA_DIGESTSIZE  20
+
+/* The structure for storing SHS info */
+
+typedef struct {
+    PyObject_HEAD
+    SHA_INT32 digest[5];		/* Message digest */
+    SHA_INT32 count_lo, count_hi;	/* 64-bit bit count */
+    SHA_BYTE data[SHA_BLOCKSIZE];	/* SHA data buffer */
+    int Endianness;
+    int local;				/* unprocessed amount in data */
+} SHAobject;
+
+/* When run on a little-endian CPU we need to perform byte reversal on an
+   array of longwords. */
+
+static void longReverse(SHA_INT32 *buffer, int byteCount, int Endianness)
+{
+    SHA_INT32 value;
+
+    if ( Endianness == PCT_BIG_ENDIAN )
+	return;
+
+    byteCount /= sizeof(*buffer);
+    while (byteCount--) {
+        value = *buffer;
+        value = ( ( value & 0xFF00FF00L ) >> 8  ) | \
+                ( ( value & 0x00FF00FFL ) << 8 );
+        *buffer++ = ( value << 16 ) | ( value >> 16 );
+    }
+}
+
+static void SHAcopy(SHAobject *src, SHAobject *dest)
+{
+    dest->Endianness = src->Endianness;
+    dest->local = src->local;
+    dest->count_lo = src->count_lo;
+    dest->count_hi = src->count_hi;
+    memcpy(dest->digest, src->digest, sizeof(src->digest));
+    memcpy(dest->data, src->data, sizeof(src->data));
+}
+
+
+/* ------------------------------------------------------------------------
+ *
+ * This code for the SHA algorithm was noted as public domain. The original
+ * headers are pasted below.
+ *
+ * Several changes have been made to make it more compatible with the
+ * Python environment and desired interface.
+ *
+ */
+
+/* NIST Secure Hash Algorithm */
+/* heavily modified by Uwe Hollerbach <uh at alumni.caltech edu> */
+/* from Peter C. Gutmann's implementation as found in */
+/* Applied Cryptography by Bruce Schneier */
+/* Further modifications to include the "UNRAVEL" stuff, below */
+
+/* This code is in the public domain */
+
+/* UNRAVEL should be fastest & biggest */
+/* UNROLL_LOOPS should be just as big, but slightly slower */
+/* both undefined should be smallest and slowest */
+
+#define UNRAVEL
+/* #define UNROLL_LOOPS */
+
+/* The SHA f()-functions.  The f1 and f3 functions can be optimized to
+   save one boolean operation each - thanks to Rich Schroeppel,
+   rcs at cs.arizona.edu for discovering this */
+
+/*#define f1(x,y,z)	((x & y) | (~x & z))		// Rounds  0-19 */
+#define f1(x,y,z)	(z ^ (x & (y ^ z)))		/* Rounds  0-19 */
+#define f2(x,y,z)	(x ^ y ^ z)			/* Rounds 20-39 */
+/*#define f3(x,y,z)	((x & y) | (x & z) | (y & z))	// Rounds 40-59 */
+#define f3(x,y,z)	((x & y) | (z & (x | y)))	/* Rounds 40-59 */
+#define f4(x,y,z)	(x ^ y ^ z)			/* Rounds 60-79 */
+
+/* SHA constants */
+
+#define CONST1		0x5a827999L			/* Rounds  0-19 */
+#define CONST2		0x6ed9eba1L			/* Rounds 20-39 */
+#define CONST3		0x8f1bbcdcL			/* Rounds 40-59 */
+#define CONST4		0xca62c1d6L			/* Rounds 60-79 */
+
+/* 32-bit rotate */
+
+#define R32(x,n)	((x << n) | (x >> (32 - n)))
+
+/* the generic case, for when the overall rotation is not unraveled */
+
+#define FG(n)	\
+    T = R32(A,5) + f##n(B,C,D) + E + *WP++ + CONST##n;	\
+    E = D; D = C; C = R32(B,30); B = A; A = T
+
+/* specific cases, for when the overall rotation is unraveled */
+
+#define FA(n)	\
+    T = R32(A,5) + f##n(B,C,D) + E + *WP++ + CONST##n; B = R32(B,30)
+
+#define FB(n)	\
+    E = R32(T,5) + f##n(A,B,C) + D + *WP++ + CONST##n; A = R32(A,30)
+
+#define FC(n)	\
+    D = R32(E,5) + f##n(T,A,B) + C + *WP++ + CONST##n; T = R32(T,30)
+
+#define FD(n)	\
+    C = R32(D,5) + f##n(E,T,A) + B + *WP++ + CONST##n; E = R32(E,30)
+
+#define FE(n)	\
+    B = R32(C,5) + f##n(D,E,T) + A + *WP++ + CONST##n; D = R32(D,30)
+
+#define FT(n)	\
+    A = R32(B,5) + f##n(C,D,E) + T + *WP++ + CONST##n; C = R32(C,30)
+
+/* do SHA transformation */
+
+static void
+sha_transform(SHAobject *sha_info)
+{
+    int i;
+    SHA_INT32 T, A, B, C, D, E, W[80], *WP;
+
+    memcpy(W, sha_info->data, sizeof(sha_info->data));
+    longReverse(W, (int)sizeof(sha_info->data), sha_info->Endianness);
+
+    for (i = 16; i < 80; ++i) {
+	W[i] = W[i-3] ^ W[i-8] ^ W[i-14] ^ W[i-16];
+
+	/* extra rotation fix */
+	W[i] = R32(W[i], 1);
+    }
+    A = sha_info->digest[0];
+    B = sha_info->digest[1];
+    C = sha_info->digest[2];
+    D = sha_info->digest[3];
+    E = sha_info->digest[4];
+    WP = W;
+#ifdef UNRAVEL
+    FA(1); FB(1); FC(1); FD(1); FE(1); FT(1); FA(1); FB(1); FC(1); FD(1);
+    FE(1); FT(1); FA(1); FB(1); FC(1); FD(1); FE(1); FT(1); FA(1); FB(1);
+    FC(2); FD(2); FE(2); FT(2); FA(2); FB(2); FC(2); FD(2); FE(2); FT(2);
+    FA(2); FB(2); FC(2); FD(2); FE(2); FT(2); FA(2); FB(2); FC(2); FD(2);
+    FE(3); FT(3); FA(3); FB(3); FC(3); FD(3); FE(3); FT(3); FA(3); FB(3);
+    FC(3); FD(3); FE(3); FT(3); FA(3); FB(3); FC(3); FD(3); FE(3); FT(3);
+    FA(4); FB(4); FC(4); FD(4); FE(4); FT(4); FA(4); FB(4); FC(4); FD(4);
+    FE(4); FT(4); FA(4); FB(4); FC(4); FD(4); FE(4); FT(4); FA(4); FB(4);
+    sha_info->digest[0] += E;
+    sha_info->digest[1] += T;
+    sha_info->digest[2] += A;
+    sha_info->digest[3] += B;
+    sha_info->digest[4] += C;
+#else /* !UNRAVEL */
+#ifdef UNROLL_LOOPS
+    FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1);
+    FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1); FG(1);
+    FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2);
+    FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2); FG(2);
+    FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3);
+    FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3); FG(3);
+    FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4);
+    FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4); FG(4);
+#else /* !UNROLL_LOOPS */
+    for (i =  0; i < 20; ++i) { FG(1); }
+    for (i = 20; i < 40; ++i) { FG(2); }
+    for (i = 40; i < 60; ++i) { FG(3); }
+    for (i = 60; i < 80; ++i) { FG(4); }
+#endif /* !UNROLL_LOOPS */
+    sha_info->digest[0] += A;
+    sha_info->digest[1] += B;
+    sha_info->digest[2] += C;
+    sha_info->digest[3] += D;
+    sha_info->digest[4] += E;
+#endif /* !UNRAVEL */
+}
+
+/* initialize the SHA digest */
+
+static void
+sha_init(SHAobject *sha_info)
+{
+    TestEndianness(sha_info->Endianness)
+
+    sha_info->digest[0] = 0x67452301L;
+    sha_info->digest[1] = 0xefcdab89L;
+    sha_info->digest[2] = 0x98badcfeL;
+    sha_info->digest[3] = 0x10325476L;
+    sha_info->digest[4] = 0xc3d2e1f0L;
+    sha_info->count_lo = 0L;
+    sha_info->count_hi = 0L;
+    sha_info->local = 0;
+}
+
+/* update the SHA digest */
+
+static void
+sha_update(SHAobject *sha_info, SHA_BYTE *buffer, int count)
+{
+    int i;
+    SHA_INT32 clo;
+
+    clo = sha_info->count_lo + ((SHA_INT32) count << 3);
+    if (clo < sha_info->count_lo) {
+        ++sha_info->count_hi;
+    }
+    sha_info->count_lo = clo;
+    sha_info->count_hi += (SHA_INT32) count >> 29;
+    if (sha_info->local) {
+        i = SHA_BLOCKSIZE - sha_info->local;
+        if (i > count) {
+            i = count;
+        }
+        memcpy(((SHA_BYTE *) sha_info->data) + sha_info->local, buffer, i);
+        count -= i;
+        buffer += i;
+        sha_info->local += i;
+        if (sha_info->local == SHA_BLOCKSIZE) {
+            sha_transform(sha_info);
+        }
+        else {
+            return;
+        }
+    }
+    while (count >= SHA_BLOCKSIZE) {
+        memcpy(sha_info->data, buffer, SHA_BLOCKSIZE);
+        buffer += SHA_BLOCKSIZE;
+        count -= SHA_BLOCKSIZE;
+        sha_transform(sha_info);
+    }
+    memcpy(sha_info->data, buffer, count);
+    sha_info->local = count;
+}
+
+/* finish computing the SHA digest */
+
+static void
+sha_final(unsigned char digest[20], SHAobject *sha_info)
+{
+    int count;
+    SHA_INT32 lo_bit_count, hi_bit_count;
+
+    lo_bit_count = sha_info->count_lo;
+    hi_bit_count = sha_info->count_hi;
+    count = (int) ((lo_bit_count >> 3) & 0x3f);
+    ((SHA_BYTE *) sha_info->data)[count++] = 0x80;
+    if (count > SHA_BLOCKSIZE - 8) {
+	memset(((SHA_BYTE *) sha_info->data) + count, 0,
+	       SHA_BLOCKSIZE - count);
+	sha_transform(sha_info);
+	memset((SHA_BYTE *) sha_info->data, 0, SHA_BLOCKSIZE - 8);
+    }
+    else {
+	memset(((SHA_BYTE *) sha_info->data) + count, 0,
+	       SHA_BLOCKSIZE - 8 - count);
+    }
+
+    /* GJS: note that we add the hi/lo in big-endian. sha_transform will
+       swap these values into host-order. */
+    sha_info->data[56] = (hi_bit_count >> 24) & 0xff;
+    sha_info->data[57] = (hi_bit_count >> 16) & 0xff;
+    sha_info->data[58] = (hi_bit_count >>  8) & 0xff;
+    sha_info->data[59] = (hi_bit_count >>  0) & 0xff;
+    sha_info->data[60] = (lo_bit_count >> 24) & 0xff;
+    sha_info->data[61] = (lo_bit_count >> 16) & 0xff;
+    sha_info->data[62] = (lo_bit_count >>  8) & 0xff;
+    sha_info->data[63] = (lo_bit_count >>  0) & 0xff;
+    sha_transform(sha_info);
+    digest[ 0] = (unsigned char) ((sha_info->digest[0] >> 24) & 0xff);
+    digest[ 1] = (unsigned char) ((sha_info->digest[0] >> 16) & 0xff);
+    digest[ 2] = (unsigned char) ((sha_info->digest[0] >>  8) & 0xff);
+    digest[ 3] = (unsigned char) ((sha_info->digest[0]      ) & 0xff);
+    digest[ 4] = (unsigned char) ((sha_info->digest[1] >> 24) & 0xff);
+    digest[ 5] = (unsigned char) ((sha_info->digest[1] >> 16) & 0xff);
+    digest[ 6] = (unsigned char) ((sha_info->digest[1] >>  8) & 0xff);
+    digest[ 7] = (unsigned char) ((sha_info->digest[1]      ) & 0xff);
+    digest[ 8] = (unsigned char) ((sha_info->digest[2] >> 24) & 0xff);
+    digest[ 9] = (unsigned char) ((sha_info->digest[2] >> 16) & 0xff);
+    digest[10] = (unsigned char) ((sha_info->digest[2] >>  8) & 0xff);
+    digest[11] = (unsigned char) ((sha_info->digest[2]      ) & 0xff);
+    digest[12] = (unsigned char) ((sha_info->digest[3] >> 24) & 0xff);
+    digest[13] = (unsigned char) ((sha_info->digest[3] >> 16) & 0xff);
+    digest[14] = (unsigned char) ((sha_info->digest[3] >>  8) & 0xff);
+    digest[15] = (unsigned char) ((sha_info->digest[3]      ) & 0xff);
+    digest[16] = (unsigned char) ((sha_info->digest[4] >> 24) & 0xff);
+    digest[17] = (unsigned char) ((sha_info->digest[4] >> 16) & 0xff);
+    digest[18] = (unsigned char) ((sha_info->digest[4] >>  8) & 0xff);
+    digest[19] = (unsigned char) ((sha_info->digest[4]      ) & 0xff);
+}
+
+/*
+ * End of copied SHA code.
+ *
+ * ------------------------------------------------------------------------
+ */
+
+static PyTypeObject SHAtype;
+
+
+static SHAobject *
+newSHAobject(void)
+{
+    return (SHAobject *)PyObject_New(SHAobject, &SHAtype);
+}
+
+/* Internal methods for a hashing object */
+
+static void
+SHA_dealloc(PyObject *ptr)
+{
+    PyObject_Del(ptr);
+}
+
+
+/* External methods for a hashing object */
+
+PyDoc_STRVAR(SHA_copy__doc__, "Return a copy of the hashing object.");
+
+static PyObject *
+SHA_copy(SHAobject *self, PyObject *unused)
+{
+    SHAobject *newobj;
+
+    if ( (newobj = newSHAobject())==NULL)
+        return NULL;
+
+    SHAcopy(self, newobj);
+    return (PyObject *)newobj;
+}
+
+PyDoc_STRVAR(SHA_digest__doc__,
+"Return the digest value as a string of binary data.");
+
+static PyObject *
+SHA_digest(SHAobject *self, PyObject *unused)
+{
+    unsigned char digest[SHA_DIGESTSIZE];
+    SHAobject temp;
+
+    SHAcopy(self, &temp);
+    sha_final(digest, &temp);
+    return PyString_FromStringAndSize((const char *)digest, sizeof(digest));
+}
+
+PyDoc_STRVAR(SHA_hexdigest__doc__,
+"Return the digest value as a string of hexadecimal digits.");
+
+static PyObject *
+SHA_hexdigest(SHAobject *self, PyObject *unused)
+{
+    unsigned char digest[SHA_DIGESTSIZE];
+    SHAobject temp;
+    PyObject *retval;
+    char *hex_digest;
+    int i, j;
+
+    /* Get the raw (binary) digest value */
+    SHAcopy(self, &temp);
+    sha_final(digest, &temp);
+
+    /* Create a new string */
+    retval = PyString_FromStringAndSize(NULL, sizeof(digest) * 2);
+    if (!retval)
+	    return NULL;
+    hex_digest = PyString_AsString(retval);
+    if (!hex_digest) {
+	    Py_DECREF(retval);
+	    return NULL;
+    }
+
+    /* Make hex version of the digest */
+    for(i=j=0; i<sizeof(digest); i++) {
+        char c;
+        c = (digest[i] >> 4) & 0xf;
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+        c = (digest[i] & 0xf);
+	c = (c>9) ? c+'a'-10 : c + '0';
+        hex_digest[j++] = c;
+    }
+    return retval;
+}
+
+PyDoc_STRVAR(SHA_update__doc__,
+"Update this hashing object's state with the provided string.");
+
+static PyObject *
+SHA_update(SHAobject *self, PyObject *args)
+{
+    unsigned char *cp;
+    int len;
+
+    if (!PyArg_ParseTuple(args, "s#:update", &cp, &len))
+        return NULL;
+
+    sha_update(self, cp, len);
+
+    Py_INCREF(Py_None);
+    return Py_None;
+}
+
+static PyMethodDef SHA_methods[] = {
+    {"copy",	  (PyCFunction)SHA_copy,      METH_NOARGS,  SHA_copy__doc__},
+    {"digest",	  (PyCFunction)SHA_digest,    METH_NOARGS,  SHA_digest__doc__},
+    {"hexdigest", (PyCFunction)SHA_hexdigest, METH_NOARGS,  SHA_hexdigest__doc__},
+    {"update",	  (PyCFunction)SHA_update,    METH_VARARGS, SHA_update__doc__},
+    {NULL,	  NULL}		/* sentinel */
+};
+
+static PyObject *
+SHA_get_block_size(PyObject *self, void *closure)
+{
+    return PyInt_FromLong(SHA_BLOCKSIZE);
+}
+
+static PyObject *
+SHA_get_digest_size(PyObject *self, void *closure)
+{
+    return PyInt_FromLong(SHA_DIGESTSIZE);
+}
+
+static PyObject *
+SHA_get_name(PyObject *self, void *closure)
+{
+    return PyString_FromStringAndSize("SHA1", 4);
+}
+
+static PyGetSetDef SHA_getseters[] = {
+    {"digest_size",
+     (getter)SHA_get_digest_size, NULL,
+     NULL,
+     NULL},
+    {"block_size",
+     (getter)SHA_get_block_size, NULL,
+     NULL,
+     NULL},
+    {"name",
+     (getter)SHA_get_name, NULL,
+     NULL,
+     NULL},
+    /* the old md5 and sha modules support 'digest_size' as in PEP 247.
+     * the old sha module also supported 'digestsize'.  ugh. */
+    {"digestsize",
+     (getter)SHA_get_digest_size, NULL,
+     NULL,
+     NULL},
+    {NULL}  /* Sentinel */
+};
+
+static PyTypeObject SHAtype = {
+    PyObject_HEAD_INIT(NULL)
+    0,			/*ob_size*/
+    "_sha.sha",		/*tp_name*/
+    sizeof(SHAobject),	/*tp_size*/
+    0,			/*tp_itemsize*/
+    /* methods */
+    SHA_dealloc,	/*tp_dealloc*/
+    0,			/*tp_print*/
+    0,                  /*tp_getattr*/
+    0,                  /*tp_setattr*/
+    0,                  /*tp_compare*/
+    0,                  /*tp_repr*/
+    0,                  /*tp_as_number*/
+    0,                  /*tp_as_sequence*/
+    0,                  /*tp_as_mapping*/
+    0,                  /*tp_hash*/
+    0,                  /*tp_call*/
+    0,                  /*tp_str*/
+    0,                  /*tp_getattro*/
+    0,                  /*tp_setattro*/
+    0,                  /*tp_as_buffer*/
+    Py_TPFLAGS_DEFAULT, /*tp_flags*/
+    0,                  /*tp_doc*/
+    0,                  /*tp_traverse*/
+    0,			/*tp_clear*/
+    0,			/*tp_richcompare*/
+    0,			/*tp_weaklistoffset*/
+    0,			/*tp_iter*/
+    0,			/*tp_iternext*/
+    SHA_methods,	/* tp_methods */
+    0,                  /* tp_members */
+    SHA_getseters,      /* tp_getset */
+};
+
+
+/* The single module-level function: new() */
+
+PyDoc_STRVAR(SHA_new__doc__,
+"Return a new SHA hashing object.  An optional string argument\n\
+may be provided; if present, this string will be automatically\n\
+hashed.");
+
+static PyObject *
+SHA_new(PyObject *self, PyObject *args, PyObject *kwdict)
+{
+    static char *kwlist[] = {"string", NULL};
+    SHAobject *new;
+    unsigned char *cp = NULL;
+    int len;
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwdict, "|s#:new", kwlist,
+                                     &cp, &len)) {
+        return NULL;
+    }
+
+    if ((new = newSHAobject()) == NULL)
+        return NULL;
+
+    sha_init(new);
+
+    if (PyErr_Occurred()) {
+        Py_DECREF(new);
+        return NULL;
+    }
+    if (cp)
+        sha_update(new, cp, len);
+
+    return (PyObject *)new;
+}
+
+
+/* List of functions exported by this module */
+
+static struct PyMethodDef SHA_functions[] = {
+    {"new", (PyCFunction)SHA_new, METH_VARARGS|METH_KEYWORDS, SHA_new__doc__},
+    {NULL,	NULL}		 /* Sentinel */
+};
+
+
+/* Initialize this module. */
+
+#define insint(n,v) { PyModule_AddIntConstant(m,n,v); }
+
+PyMODINIT_FUNC
+init_sha(void)
+{
+    PyObject *m;
+
+    SHAtype.ob_type = &PyType_Type;
+    if (PyType_Ready(&SHAtype) < 0)
+        return;
+    m = Py_InitModule("_sha", SHA_functions);
+    if (m == NULL)
+	return;
+
+    /* Add some symbolic constants to the module */
+    insint("blocksize", 1);  /* For future use, in case some hash
+                                functions require an integral number of
+                                blocks */ 
+    insint("digestsize", 20);
+    insint("digest_size", 20);
+}
diff --git a/src/README.txt b/src/README.txt
--- a/src/README.txt
+++ b/src/README.txt
@@ -2,7 +2,7 @@
 Distutils2
 ==========
 
-Welcome to Distutils2 !
+Welcome to Distutils2!
 
 Distutils2 is the new version of Distutils. It's not backward compatible with
 Distutils but provides more features, and implement most new packaging
@@ -10,6 +10,6 @@
 
 See the documentation at http://packages.python.org/Distutils2 for more info.
 
-**Beware that Distutils2 is its in early stage and should not be used in
+**Beware that Distutils2 is in its early stage and should not be used in
 production. Its API is subject to changes**
 
diff --git a/src/distutils2/README b/src/distutils2/README
--- a/src/distutils2/README
+++ b/src/distutils2/README
@@ -1,4 +1,4 @@
-This directory contains the Distutils package.
+This directory contains the Distutils2 package.
 
 There's a full documentation available at:
 
@@ -8,6 +8,6 @@
 
     http://www.python.org/sigs/distutils-sig/
 
-WARNING: Distutils2 must remain compatible with 2.4
+WARNING: Distutils2 must remain compatible with Python 2.4
 
 $Id: README 70017 2009-02-27 12:53:34Z tarek.ziade $
diff --git a/src/distutils2/__init__.py b/src/distutils2/__init__.py
--- a/src/distutils2/__init__.py
+++ b/src/distutils2/__init__.py
@@ -1,16 +1,26 @@
 """distutils
 
-The main package for the Python Module Distribution Utilities.  Normally
-used from a setup script as
+The main package for the Python Distribution Utilities 2.  Setup
+scripts should import the setup function from distutils2.core:
 
-   from distutils.core import setup
+    from distutils2.core import setup
 
-   setup (...)
+    setup(name=..., version=..., ...)
+
+Third-party tools can use parts of Distutils2 as building blocks
+without causing the other modules to be imported:
+
+    import distutils2.version
+    import distutils2.pypi.simple
+    import distutils2.tests.pypi_server
 """
-__all__ = ['__version__', 'setup']
+__all__ = ['__version__']
 
 __revision__ = "$Id: __init__.py 78020 2010-02-06 16:37:32Z benjamin.peterson $"
 __version__ = "1.0a2"
 
-from distutils2.core import setup
 
+# when set to True, converts doctests by default too 
+run_2to3_on_doctests = True
+# Standard package names for fixer packages
+lib2to3_fixer_packages = ['lib2to3.fixes']
diff --git a/src/distutils2/_backport/hashlib.py b/src/distutils2/_backport/hashlib.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/hashlib.py
@@ -0,0 +1,143 @@
+# $Id$
+#
+#  Copyright (C) 2005   Gregory P. Smith (greg at krypto.org)
+#  Licensed to PSF under a Contributor Agreement.
+#
+
+__doc__ = """hashlib module - A common interface to many hash functions.
+
+new(name, string='') - returns a new hash object implementing the
+                       given hash function; initializing the hash
+                       using the given string data.
+
+Named constructor functions are also available, these are much faster
+than using new():
+
+md5(), sha1(), sha224(), sha256(), sha384(), and sha512()
+
+More algorithms may be available on your platform but the above are
+guaranteed to exist.
+
+NOTE: If you want the adler32 or crc32 hash functions they are available in
+the zlib module.
+
+Choose your hash function wisely.  Some have known collision weaknesses.
+sha384 and sha512 will be slow on 32 bit platforms.
+
+Hash objects have these methods:
+ - update(arg): Update the hash object with the string arg. Repeated calls
+                are equivalent to a single call with the concatenation of all
+                the arguments.
+ - digest():    Return the digest of the strings passed to the update() method
+                so far. This may contain non-ASCII characters, including
+                NUL bytes.
+ - hexdigest(): Like digest() except the digest is returned as a string of
+                double length, containing only hexadecimal digits.
+ - copy():      Return a copy (clone) of the hash object. This can be used to
+                efficiently compute the digests of strings that share a common
+                initial substring.
+
+For example, to obtain the digest of the string 'Nobody inspects the
+spammish repetition':
+
+    >>> import hashlib
+    >>> m = hashlib.md5()
+    >>> m.update("Nobody inspects")
+    >>> m.update(" the spammish repetition")
+    >>> m.digest()
+    '\\xbbd\\x9c\\x83\\xdd\\x1e\\xa5\\xc9\\xd9\\xde\\xc9\\xa1\\x8d\\xf0\\xff\\xe9'
+
+More condensed:
+
+    >>> hashlib.sha224("Nobody inspects the spammish repetition").hexdigest()
+    'a4337bc45a8fc544c03f52dc550cd6e1e87021bc896588bd79e901e2'
+
+"""
+
+# This tuple and __get_builtin_constructor() must be modified if a new
+# always available algorithm is added.
+__always_supported = ('md5', 'sha1', 'sha224', 'sha256', 'sha384', 'sha512')
+
+algorithms = __always_supported
+
+__all__ = __always_supported + ('new', 'algorithms')
+
+
+def __get_builtin_constructor(name):
+    if name in ('SHA1', 'sha1'):
+        import _sha
+        return _sha.new
+    elif name in ('MD5', 'md5'):
+        import _md5
+        return _md5.new
+    elif name in ('SHA256', 'sha256', 'SHA224', 'sha224'):
+        import _sha256
+        bs = name[3:]
+        if bs == '256':
+            return _sha256.sha256
+        elif bs == '224':
+            return _sha256.sha224
+    elif name in ('SHA512', 'sha512', 'SHA384', 'sha384'):
+        import _sha512
+        bs = name[3:]
+        if bs == '512':
+            return _sha512.sha512
+        elif bs == '384':
+            return _sha512.sha384
+
+    raise ValueError('unsupported hash type %s' % name)
+
+
+def __get_openssl_constructor(name):
+    try:
+        f = getattr(_hashlib, 'openssl_' + name)
+        # Allow the C module to raise ValueError.  The function will be
+        # defined but the hash not actually available thanks to OpenSSL.
+        f()
+        # Use the C function directly (very fast)
+        return f
+    except (AttributeError, ValueError):
+        return __get_builtin_constructor(name)
+
+
+def __py_new(name, string=''):
+    """new(name, string='') - Return a new hashing object using the named algorithm;
+    optionally initialized with a string.
+    """
+    return __get_builtin_constructor(name)(string)
+
+
+def __hash_new(name, string=''):
+    """new(name, string='') - Return a new hashing object using the named algorithm;
+    optionally initialized with a string.
+    """
+    try:
+        return _hashlib.new(name, string)
+    except ValueError:
+        # If the _hashlib module (OpenSSL) doesn't support the named
+        # hash, try using our builtin implementations.
+        # This allows for SHA224/256 and SHA384/512 support even though
+        # the OpenSSL library prior to 0.9.8 doesn't provide them.
+        return __get_builtin_constructor(name)(string)
+
+
+try:
+    import _hashlib
+    new = __hash_new
+    __get_hash = __get_openssl_constructor
+except ImportError:
+    new = __py_new
+    __get_hash = __get_builtin_constructor
+
+for __func_name in __always_supported:
+    # try them all, some may not work due to the OpenSSL
+    # version not supporting that algorithm.
+    try:
+        globals()[__func_name] = __get_hash(__func_name)
+    except ValueError:
+        import logging
+        logging.exception('code for hash %s was not found.', __func_name)
+
+# Cleanup locals()
+del __always_supported, __func_name, __get_hash
+del __py_new, __hash_new, __get_openssl_constructor
diff --git a/src/distutils2/_backport/pkgutil.py b/src/distutils2/_backport/pkgutil.py
--- a/src/distutils2/_backport/pkgutil.py
+++ b/src/distutils2/_backport/pkgutil.py
@@ -12,10 +12,17 @@
 from distutils2.errors import DistutilsError
 from distutils2.metadata import DistributionMetadata
 from distutils2.version import suggest_normalized_version, VersionPredicate
+import zipimport
+try:
+    import cStringIO as StringIO
+except ImportError:
+    import StringIO
+import re
+import warnings
 
 __all__ = [
     'get_importer', 'iter_importers', 'get_loader', 'find_loader',
-    'walk_packages', 'iter_modules',
+    'walk_packages', 'iter_modules', 'get_data',
     'ImpImporter', 'ImpLoader', 'read_code', 'extend_path',
     'Distribution', 'EggInfoDistribution', 'distinfo_dirname',
     'get_distributions', 'get_distribution', 'get_file_users',
@@ -24,7 +31,7 @@
 
 
 def read_code(stream):
-    # This helper is needed in order for the PEP 302 emulation to
+    # This helper is needed in order for the :pep:`302` emulation to
     # correctly handle compiled files
     import marshal
 
@@ -79,32 +86,34 @@
 
 
 def walk_packages(path=None, prefix='', onerror=None):
-    """Yields (module_loader, name, ispkg) for all modules recursively
-    on path, or, if path is None, all accessible modules.
+    """Yields ``(module_loader, name, ispkg)`` for all modules recursively
+    on *path*, or, if *path* is ``None``, all accessible modules.
 
-    'path' should be either None or a list of paths to look for
-    modules in.
+    :parameter path: should be either ``None`` or a list of paths to look for
+                     modules in.
+    :parameter prefix: is a string to output on the front of every module name
+                       on output.
 
-    'prefix' is a string to output on the front of every module name
-    on output.
-
-    Note that this function must import all *packages* (NOT all
-    modules!) on the given path, in order to access the __path__
+    Note that this function must import all packages (NOT all
+    modules!) on the given path, in order to access the ``__path__``
     attribute to find submodules.
 
-    'onerror' is a function which gets called with one argument (the
+    *onerror* is a function which gets called with one argument (the
     name of the package which was being imported) if any exception
     occurs while trying to import a package.  If no onerror function is
-    supplied, ImportErrors are caught and ignored, while all other
+    supplied, ``ImportErrors`` are caught and ignored, while all other
     exceptions are propagated, terminating the search.
 
     Examples:
 
-    # list all modules python can access
-    walk_packages()
+    * list all modules python can access::
 
-    # list all submodules of ctypes
-    walk_packages(ctypes.__path__, ctypes.__name__+'.')
+        walk_packages()
+
+    * list all submodules of ctypes::
+
+        walk_packages(ctypes.__path__, ctypes.__name__+'.')
+
     """
 
     def seen(p, m={}):
@@ -137,14 +146,14 @@
 
 
 def iter_modules(path=None, prefix=''):
-    """Yields (module_loader, name, ispkg) for all submodules on path,
-    or, if path is None, all top-level modules on sys.path.
+    """Yields ``(module_loader, name, ispkg)`` for all submodules on path,
+    or, if *path* is ``None``, all top-level modules on ``sys.path``.
 
-    'path' should be either None or a list of paths to look for
-    modules in.
+    :parameter path: should be either None or a list of paths to look for
+                     modules in.
+    :parameter prefix: is a string to output on the front of every module name
+                       on output.
 
-    'prefix' is a string to output on the front of every module name
-    on output.
     """
 
     if path is None:
@@ -162,6 +171,7 @@
 
 #@simplegeneric
 def iter_importer_modules(importer, prefix=''):
+    ""
     if not hasattr(importer, 'iter_modules'):
         return []
     return importer.iter_modules(prefix)
@@ -169,15 +179,16 @@
 iter_importer_modules = simplegeneric(iter_importer_modules)
 
 
-class ImpImporter:
-    """PEP 302 Importer that wraps Python's "classic" import algorithm
+class ImpImporter(object):
+    """:pep:`302` Importer that wraps Python's "classic" import algorithm
 
-    ImpImporter(dirname) produces a PEP 302 importer that searches that
-    directory.  ImpImporter(None) produces a PEP 302 importer that searches
-    the current sys.path, plus any modules that are frozen or built-in.
+    ``ImpImporter(dirname)`` produces a :pep:`302` importer that searches that
+    directory. ``ImpImporter(None)`` produces a :pep:`302` importer that
+    searches the current ``sys.path``, plus any modules that are frozen
+    or built-in.
 
-    Note that ImpImporter does not currently support being used by placement
-    on sys.meta_path.
+    Note that :class:`ImpImporter` does not currently support being used by placement
+    on ``sys.meta_path``.
     """
 
     def __init__(self, path=None):
@@ -231,9 +242,9 @@
                 yield prefix + modname, ispkg
 
 
-class ImpLoader:
-    """PEP 302 Loader that wraps Python's "classic" import algorithm
-    """
+class ImpLoader(object):
+    """:pep:`302` Loader that wraps Python's "classic" import algorithm """
+
     code = source = None
 
     def __init__(self, fullname, file, filename, etc):
@@ -365,17 +376,17 @@
 
 
 def get_importer(path_item):
-    """Retrieve a PEP 302 importer for the given path item
+    """Retrieve a  :pep:`302` importer for the given path item
 
-    The returned importer is cached in sys.path_importer_cache
+    The returned importer is cached in ``sys.path_importer_cache``
     if it was newly created by a path hook.
 
     If there is no importer, a wrapper around the basic import
     machinery is returned. This wrapper is never inserted into
-    the importer cache (None is inserted instead).
+    the importer cache (``None`` is inserted instead).
 
     The cache (or part of it) can be cleared manually if a
-    rescan of sys.path_hooks is necessary.
+    rescan of ``sys.path_hooks`` is necessary.
     """
     try:
         importer = sys.path_importer_cache[path_item]
@@ -399,7 +410,7 @@
 
 
 def iter_importers(fullname=""):
-    """Yield PEP 302 importers for the given module name
+    """Yield :pep:`302` importers for the given module name
 
     If fullname contains a '.', the importers will be for the package
     containing fullname, otherwise they will be importers for sys.meta_path,
@@ -407,20 +418,20 @@
     the named module is in a package, that package is imported as a side
     effect of invoking this function.
 
-    Non PEP 302 mechanisms (e.g. the Windows registry) used by the
+    Non :pep:`302` mechanisms (e.g. the Windows registry) used by the
     standard import machinery to find files in alternative locations
-    are partially supported, but are searched AFTER sys.path. Normally,
-    these locations are searched BEFORE sys.path, preventing sys.path
+    are partially supported, but are searched AFTER ``sys.path``. Normally,
+    these locations are searched BEFORE sys.path, preventing ``sys.path``
     entries from shadowing them.
 
     For this to cause a visible difference in behaviour, there must
     be a module or package name that is accessible via both sys.path
-    and one of the non PEP 302 file system mechanisms. In this case,
+    and one of the non :pep:`302` file system mechanisms. In this case,
     the emulation will find the former version, while the builtin
     import mechanism will find the latter.
 
     Items of the following types can be affected by this discrepancy:
-        imp.C_EXTENSION, imp.PY_SOURCE, imp.PY_COMPILED, imp.PKG_DIRECTORY
+        ``imp.C_EXTENSION, imp.PY_SOURCE, imp.PY_COMPILED, imp.PKG_DIRECTORY``
     """
     if fullname.startswith('.'):
         raise ImportError("Relative module names not supported")
@@ -441,15 +452,15 @@
 
 
 def get_loader(module_or_name):
-    """Get a PEP 302 "loader" object for module_or_name
+    """Get a :pep:`302` "loader" object for module_or_name
 
     If the module or package is accessible via the normal import
     mechanism, a wrapper around the relevant part of that machinery
     is returned.  Returns None if the module cannot be found or imported.
     If the named module is not already imported, its containing package
-    (if any) is imported, in order to establish the package __path__.
+    (if any) is imported, in order to establish the package ``__path__``.
 
-    This function uses iter_importers(), and is thus subject to the same
+    This function uses :func:`iter_importers`, and is thus subject to the same
     limitations regarding platform-specific special import locations such
     as the Windows registry.
     """
@@ -467,12 +478,13 @@
 
 
 def find_loader(fullname):
-    """Find a PEP 302 "loader" object for fullname
+    """Find a :pep:`302` "loader" object for fullname
 
-    If fullname contains dots, path must be the containing package's __path__.
-    Returns None if the module cannot be found or imported. This function uses
-    iter_importers(), and is thus subject to the same limitations regarding
-    platform-specific special import locations such as the Windows registry.
+    If fullname contains dots, path must be the containing package's
+    ``__path__``. Returns ``None`` if the module cannot be found or imported.
+    This function uses :func:`iter_importers`, and is thus subject to the same
+    limitations regarding platform-specific special import locations such as
+    the Windows registry.
     """
     for importer in iter_importers(fullname):
         loader = importer.find_module(fullname)
@@ -485,21 +497,22 @@
 def extend_path(path, name):
     """Extend a package's path.
 
-    Intended use is to place the following code in a package's __init__.py:
+    Intended use is to place the following code in a package's
+    ``__init__.py``::
 
         from pkgutil import extend_path
         __path__ = extend_path(__path__, __name__)
 
-    This will add to the package's __path__ all subdirectories of
-    directories on sys.path named after the package.  This is useful
+    This will add to the package's ``__path__`` all subdirectories of
+    directories on ``sys.path`` named after the package.  This is useful
     if one wants to distribute different parts of a single logical
     package as multiple directories.
 
-    It also looks for *.pkg files beginning where * matches the name
-    argument.  This feature is similar to *.pth files (see site.py),
-    except that it doesn't special-case lines starting with 'import'.
-    A *.pkg file is trusted at face value: apart from checking for
-    duplicates, all entries found in a *.pkg file are added to the
+    It also looks for ``*.pkg`` files beginning where ``*`` matches the name
+    argument.  This feature is similar to ``*.pth`` files (see ``site.py``),
+    except that it doesn't special-case lines starting with ``import``.
+    A ``*.pkg`` file is trusted at face value: apart from checking for
+    duplicates, all entries found in a ``*.pkg`` file are added to the
     path, regardless of whether they are exist the filesystem.  (This
     is a feature.)
 
@@ -512,7 +525,7 @@
     are not (unicode or 8-bit) strings referring to existing
     directories are ignored.  Unicode items of sys.path that cause
     errors when used as filenames may cause this function to raise an
-    exception (in line with os.path.isdir() behavior).
+    exception (in line with ``os.path.isdir()`` behavior).
     """
 
     if not isinstance(path, list):
@@ -560,23 +573,23 @@
 def get_data(package, resource):
     """Get a resource from a package.
 
-    This is a wrapper round the PEP 302 loader get_data API. The package
+    This is a wrapper round the :pep:`302` loader get_data API. The package
     argument should be the name of a package, in standard module format
-    (foo.bar). The resource argument should be in the form of a relative
-    filename, using '/' as the path separator. The parent directory name '..'
-    is not allowed, and nor is a rooted name (starting with a '/').
+    (``foo.bar``). The resource argument should be in the form of a relative
+    filename, using ``'/'`` as the path separator. The parent directory name
+    ``'..'`` is not allowed, and nor is a rooted name (starting with a ``'/'``).
 
     The function returns a binary string, which is the contents of the
     specified resource.
 
     For packages located in the filesystem, which have already been imported,
-    this is the rough equivalent of
+    this is the rough equivalent of::
 
         d = os.path.dirname(sys.modules[package].__file__)
         data = open(os.path.join(d, resource), 'rb').read()
 
-    If the package cannot be located or loaded, or it uses a PEP 302 loader
-    which does not support get_data(), then None is returned.
+    If the package cannot be located or loaded, or it uses a :pep:`302` loader
+    which does not support :func:`get_data`, then ``None`` is returned.
     """
 
     loader = get_loader(package)
@@ -603,7 +616,7 @@
 
 class Distribution(object):
     """Created with the *path* of the ``.dist-info`` directory provided to the
-    constructor. It reads the metadata contained in METADATA when it is
+    constructor. It reads the metadata contained in ``METADATA`` when it is
     instantiated."""
 
     # Attribute documenting for Sphinx style documentation, see for more info:
@@ -612,9 +625,9 @@
     """The name of the distribution."""
     metadata = None
     """A :class:`distutils2.metadata.DistributionMetadata` instance loaded with
-    the distribution's METADATA file."""
+    the distribution's ``METADATA`` file."""
     requested = False
-    """A boolean that indicates whether the REQUESTED metadata file is present
+    """A boolean that indicates whether the ``REQUESTED`` metadata file is present
     (in other words, whether the package was installed by user request)."""
 
     def __init__(self, path):
@@ -635,15 +648,17 @@
 
     def get_installed_files(self, local=False):
         """
-        Iterates over the RECORD entries and returns a tuple (path, md5, size)
-        for each line. If *local* is ``True``, the returned path is transformed
-        into a local absolute path. Otherwise the raw value from RECORD is
-        returned.
+        Iterates over the ``RECORD`` entries and returns a tuple
+        ``(path, md5, size)`` for each line. If *local* is ``True``,
+        the returned path is transformed into a local absolute path.
+        Otherwise the raw value from RECORD is returned.
 
         A local absolute path is an absolute path in which occurrences of
         ``'/'`` have been replaced by the system separator given by ``os.sep``.
+
         :parameter local: flag to say if the path should be returned a local
                           absolute path
+
         :type local: boolean
         :returns: iterator of (path, md5, size)
         """
@@ -651,7 +666,7 @@
 
     def uses(self, path):
         """
-        Returns ``True`` if path is listed in RECORD. *path* can be a local
+        Returns ``True`` if path is listed in ``RECORD``. *path* can be a local
         absolute path or a relative ``'/'``-separated path.
 
         :rtype: boolean
@@ -673,8 +688,8 @@
                          directory path, a :class:`DistutilsError` is raised
         :type path: string
         :parameter binary: If *binary* is ``True``, opens the file in read-only
-                           binary mode (rb), otherwise opens it in read-only
-                           mode (r).
+                           binary mode (``rb``), otherwise opens it in read-only
+                           mode (``r``).
         :rtype: file object
         """
         open_flags = 'r'
@@ -701,37 +716,129 @@
 
     def get_distinfo_files(self, local=False):
         """
-        Iterates over the RECORD entries and returns paths for each line if the
-        path is pointing to a file located in the ``.dist-info`` directory or
-        one of its subdirectories.
+        Iterates over the ``RECORD`` entries and returns paths for each line if
+        the path is pointing to a file located in the ``.dist-info`` directory
+        or one of its subdirectories.
 
         :parameter local: If *local* is ``True``, each returned path is
                           transformed into a local absolute path. Otherwise the
-                          raw value from RECORD is returned.
+                          raw value from ``RECORD`` is returned.
         :type local: boolean
         :returns: iterator of paths
         """
         for path, md5, size in self._get_records(local):
             yield path
 
+    def __eq__(self, other):
+        return isinstance(other, Distribution) and self.path == other.path
+
+    # See http://docs.python.org/reference/datamodel#object.__hash__
+    __hash__ = object.__hash__
+
 
 class EggInfoDistribution(object):
     """Created with the *path* of the ``.egg-info`` directory or file provided
     to the constructor. It reads the metadata contained in the file itself, or
     if the given path happens to be a directory, the metadata is read from the
-    file PKG-INFO under that directory."""
+    file ``PKG-INFO`` under that directory."""
 
     name = ''
     """The name of the distribution."""
     metadata = None
     """A :class:`distutils2.metadata.DistributionMetadata` instance loaded with
-    the distribution's METADATA file."""
+    the distribution's ``METADATA`` file."""
+    _REQUIREMENT = re.compile( \
+        r'(?P<name>[-A-Za-z0-9_.]+)\s*' \
+        r'(?P<first>(?:<|<=|!=|==|>=|>)[-A-Za-z0-9_.]+)?\s*' \
+        r'(?P<rest>(?:\s*,\s*(?:<|<=|!=|==|>=|>)[-A-Za-z0-9_.]+)*)\s*' \
+        r'(?P<extras>\[.*\])?')
 
     def __init__(self, path):
-        if os.path.isdir(path):
-            path = os.path.join(path, 'PKG-INFO')
-        self.metadata = DistributionMetadata(path=path)
-        self.name = self.metadata['name']
+        self.path = path
+
+        # reused from Distribute's pkg_resources
+        def yield_lines(strs):
+            """Yield non-empty/non-comment lines of a ``basestring`` or sequence"""
+            if isinstance(strs, basestring):
+                for s in strs.splitlines():
+                    s = s.strip()
+                    if s and not s.startswith('#'): # skip blank lines/comments
+                        yield s
+            else:
+                for ss in strs:
+                    for s in yield_lines(ss):
+                        yield s
+
+        requires = None
+        if path.endswith('.egg'):
+            if os.path.isdir(path):
+                meta_path = os.path.join(path, 'EGG-INFO', 'PKG-INFO')
+                self.metadata = DistributionMetadata(path=meta_path)
+                try:
+                    req_path = os.path.join(path, 'EGG-INFO', 'requires.txt')
+                    requires = open(req_path, 'r').read()
+                except IOError:
+                    requires = None
+            else:
+                zipf = zipimport.zipimporter(path)
+                fileobj = StringIO.StringIO(zipf.get_data('EGG-INFO/PKG-INFO'))
+                self.metadata = DistributionMetadata(fileobj=fileobj)
+                try:
+                    requires = zipf.get_data('EGG-INFO/requires.txt')
+                except IOError:
+                    requires = None
+            self.name = self.metadata['name']
+        elif path.endswith('.egg-info'):
+            if os.path.isdir(path):
+                path = os.path.join(path, 'PKG-INFO')
+                try:
+                    req_f = open(os.path.join(path, 'requires.txt'), 'r')
+                    requires = req_f.read()
+                except IOError:
+                    requires = None
+            self.metadata = DistributionMetadata(path=path)
+            self.name = self.metadata['name']
+        else:
+            raise ValueError('The path must end with .egg-info or .egg')
+
+        provides = "%s (%s)" % (self.metadata['name'],
+                                self.metadata['version'])
+        if self.metadata['Metadata-Version'] == '1.2':
+            self.metadata['Provides-Dist'] += (provides,)
+        else:
+            self.metadata['Provides'] += (provides,)
+        reqs = []
+        if requires is not None:
+            for line in yield_lines(requires):
+                if line[0] == '[':
+                    warnings.warn('distutils2 does not support extensions in requires.txt')
+                    break
+                else:
+                    match = self._REQUIREMENT.match(line.strip())
+                    if not match:
+                        raise ValueError('Distribution %s has ill formed '
+                                         'requires.txt file (%s)' %
+                                         (self.name, line))
+                    else:
+                        if match.group('extras'):
+                            s = 'Distribution %s uses extra requirements which'\
+                                ' are not supported in distutils' % (self.name)
+                            warnings.warn(s)
+                        name = match.group('name')
+                        version = None
+                        if match.group('first'):
+                            version = match.group('first')
+                            if match.group('rest'):
+                                version += match.group('rest')
+                            version = version.replace(' ', '') # trim spaces
+                        if version is None:
+                            reqs.append(name)
+                        else:
+                            reqs.append('%s (%s)' % (name, version))
+            if self.metadata['Metadata-Version'] == '1.2':
+                self.metadata['Requires-Dist'] += reqs
+            else:
+                self.metadata['Requires'] += reqs
 
     def get_installed_files(self, local=False):
         return []
@@ -739,6 +846,13 @@
     def uses(self, path):
         return False
 
+    def __eq__(self, other):
+        return isinstance(other, EggInfoDistribution) and \
+               self.path == other.path
+
+    # See http://docs.python.org/reference/datamodel#object.__hash__
+    __hash__ = object.__hash__
+
 
 def _normalize_dist_name(name):
     """Returns a normalized name from the given *name*.
@@ -792,7 +906,8 @@
             if dir.endswith('.dist-info'):
                 dist = Distribution(os.path.join(realpath, dir))
                 yield dist
-            elif use_egg_info and dir.endswith('.egg-info'):
+            elif use_egg_info and (dir.endswith('.egg-info') or
+                                   dir.endswith('.egg')):
                 dist = EggInfoDistribution(os.path.join(realpath, dir))
                 yield dist
 
@@ -801,18 +916,18 @@
     """
     Scans all elements in ``sys.path`` and looks for all directories ending with
     ``.dist-info``. Returns a :class:`Distribution` corresponding to the
-    ``.dist-info`` directory that contains the METADATA that matches *name* for
-    the *name* metadata field.
+    ``.dist-info`` directory that contains the ``METADATA`` that matches *name*
+    for the *name* metadata field.
     If no distribution exists with the given *name* and the parameter
     *use_egg_info* is set to ``True``, then all files and directories ending
     with ``.egg-info`` are scanned. A :class:`EggInfoDistribution` instance is
     returned if one is found that has metadata that matches *name* for the
     *name* metadata field.
 
-    This function only returns the first result founded, as no more than one
+    This function only returns the first result found, as no more than one
     value is expected. If the directory is not found, ``None`` is returned.
 
-    :rtype: :class:`Distribution` or :class:`EggInfoDistribution: or None"""
+    :rtype: :class:`Distribution` or :class:`EggInfoDistribution` or None"""
     found = None
     for dist in get_distributions():
         if dist.name == name:
@@ -867,7 +982,7 @@
     then all files and directories ending with ``.egg-info`` are considered
     as well and returns an :class:`EggInfoDistribution` instance.
 
-    This function only returns the first result founded, since no more than
+    This function only returns the first result found, since no more than
     one values are expected. If the directory is not found, returns ``None``.
 
     :parameter version: a version specifier that indicates the version
@@ -887,7 +1002,7 @@
         provided = dist.metadata['Provides-Dist'] + dist.metadata['Provides']
 
         for p in provided:
-            p_components = p.split(' ', 1)
+            p_components = p.rsplit(' ', 1)
             if len(p_components) == 1 or predicate is None:
                 if name == p_components[0]:
                     yield dist
@@ -896,7 +1011,8 @@
                 p_name, p_ver = p_components
                 if len(p_ver) < 2 or p_ver[0] != '(' or p_ver[-1] != ')':
                     raise DistutilsError(('Distribution %s has invalid ' +
-                                          'provides field') % (dist.name,))
+                                          'provides field: %s') \
+                                           % (dist.name, p))
                 p_ver = p_ver[1:-1] # trim off the parenthesis
                 if p_name == name and predicate.match(p_ver):
                     yield dist
diff --git a/src/distutils2/_backport/sysconfig.cfg b/src/distutils2/_backport/sysconfig.cfg
--- a/src/distutils2/_backport/sysconfig.cfg
+++ b/src/distutils2/_backport/sysconfig.cfg
@@ -1,6 +1,6 @@
 [globals]
 # These are the useful categories that are sometimes referenced at runtime,
-# using pkgutils.open():
+# using pkgutil.open():
 config             = {confdir}/{distribution.name}         # Configuration files
 appdata            = {datadir}/{distribution.name}         # Non-writable data that is independent of architecture (images, many xml/text files)
 appdata.arch       = {libdir}/{distribution.name}          # Non-writable data that is architecture-dependent (some binary data formats)
diff --git a/src/distutils2/_backport/sysconfig.py b/src/distutils2/_backport/sysconfig.py
--- a/src/distutils2/_backport/sysconfig.py
+++ b/src/distutils2/_backport/sysconfig.py
@@ -5,7 +5,7 @@
 import os
 import re
 from os.path import pardir, abspath
-from ConfigParser import ConfigParser
+from ConfigParser import RawConfigParser
 
 _PREFIX = os.path.normpath(sys.prefix)
 _EXEC_PREFIX = os.path.normpath(sys.exec_prefix)
@@ -14,7 +14,7 @@
 # XXX _CONFIG_DIR will be set by the Makefile later
 _CONFIG_DIR = os.path.normpath(os.path.dirname(__file__))
 _CONFIG_FILE = os.path.join(_CONFIG_DIR, 'sysconfig.cfg')
-_SCHEMES = ConfigParser()
+_SCHEMES = RawConfigParser()
 _SCHEMES.read(_CONFIG_FILE)
 _VAR_REPL = re.compile(r'\{([^{]*?)\}')
 
@@ -580,10 +580,11 @@
                 # behaviour.
                 pass
             else:
-                m = re.search(
-                        r'<key>ProductUserVisibleVersion</key>\s*' +
-                        r'<string>(.*?)</string>', f.read())
-                f.close()
+                try:
+                    m = re.search(r'<key>ProductUserVisibleVersion</key>\s*'
+                                  r'<string>(.*?)</string>', f.read())
+                finally:
+                    f.close()
                 if m is not None:
                     macrelease = '.'.join(m.group(1).split('.')[:2])
                 # else: fall back to the default behaviour
diff --git a/src/distutils2/_backport/tarfile.py b/src/distutils2/_backport/tarfile.py
--- a/src/distutils2/_backport/tarfile.py
+++ b/src/distutils2/_backport/tarfile.py
@@ -370,7 +370,7 @@
 #---------------------------
 # internal stream interface
 #---------------------------
-class _LowLevelFile:
+class _LowLevelFile(object):
     """Low-level file object. Supports reading and writing.
        It is used instead of a regular file object for streaming
        access.
@@ -394,7 +394,7 @@
     def write(self, s):
         os.write(self.fd, s)
 
-class _Stream:
+class _Stream(object):
     """Class that serves as an adapter between TarFile and
        a stream-like object.  The stream-like object only
        needs to have a read() or write() method and is accessed
@@ -2003,8 +2003,10 @@
         # Append the tar header and data to the archive.
         if tarinfo.isreg():
             f = bltn_open(name, "rb")
-            self.addfile(tarinfo, f)
-            f.close()
+            try:
+                self.addfile(tarinfo, f)
+            finally:
+                f.close()
 
         elif tarinfo.isdir():
             self.addfile(tarinfo)
@@ -2214,9 +2216,11 @@
         """
         source = self.extractfile(tarinfo)
         target = bltn_open(targetpath, "wb")
-        copyfileobj(source, target)
-        source.close()
-        target.close()
+        try:
+            copyfileobj(source, target)
+        finally:
+            source.close()
+            target.close()
 
     def makeunknown(self, tarinfo, targetpath):
         """Make a file from a TarInfo object with an unknown type
@@ -2423,7 +2427,7 @@
             print >> sys.stderr, msg
 # class TarFile
 
-class TarIter:
+class TarIter(object):
     """Iterator Class.
 
        for tarinfo in TarFile(...):
@@ -2460,7 +2464,7 @@
         return tarinfo
 
 # Helper classes for sparse file support
-class _section:
+class _section(object):
     """Base class for _data and _hole.
     """
     def __init__(self, offset, size):
@@ -2507,7 +2511,7 @@
 #---------------------------------------------
 TAR_PLAIN = 0           # zipfile.ZIP_STORED
 TAR_GZIPPED = 8         # zipfile.ZIP_DEFLATED
-class TarFileCompat:
+class TarFileCompat(object):
     """TarFile class compatible with standard module zipfile's
        ZipFile class.
     """
@@ -2564,8 +2568,10 @@
        are able to handle, else return False.
     """
     try:
-        t = open(name)
-        t.close()
+        try:
+            t = open(name)
+        finally:
+            t.close()
         return True
     except TarError:
         return False
diff --git a/src/distutils2/_backport/tests/fake_dists/bacon-0.1.egg-info/PKG-INFO b/src/distutils2/_backport/tests/fake_dists/bacon-0.1.egg-info/PKG-INFO
--- a/src/distutils2/_backport/tests/fake_dists/bacon-0.1.egg-info/PKG-INFO
+++ b/src/distutils2/_backport/tests/fake_dists/bacon-0.1.egg-info/PKG-INFO
@@ -2,4 +2,5 @@
 Name: bacon
 Version: 0.1
 Provides-Dist: truffles (2.0)
+Provides-Dist: bacon (0.1)
 Obsoletes-Dist: truffles (>=0.9,<=1.5)
diff --git a/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/PKG-INFO b/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/PKG-INFO
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/PKG-INFO
@@ -0,0 +1,18 @@
+Metadata-Version: 1.0
+Name: banana
+Version: 0.4
+Summary: A yellow fruit
+Home-page: http://en.wikipedia.org/wiki/Banana
+Author: Josip Djolonga
+Author-email: foo at nbar.com
+License: BSD
+Description: A fruit
+Keywords: foo bar
+Platform: UNKNOWN
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: Intended Audience :: Science/Research
+Classifier: License :: OSI Approved :: BSD License
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python
+Classifier: Topic :: Scientific/Engineering :: GIS
diff --git a/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/SOURCES.txt b/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/SOURCES.txt
new file mode 100644
diff --git a/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/dependency_links.txt b/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/dependency_links.txt
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/dependency_links.txt
@@ -0,0 +1,1 @@
+
diff --git a/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/entry_points.txt b/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/entry_points.txt
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/entry_points.txt
@@ -0,0 +1,3 @@
+
+      # -*- Entry points: -*-
+      
\ No newline at end of file
diff --git a/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/not-zip-safe b/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/not-zip-safe
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/not-zip-safe
@@ -0,0 +1,1 @@
+
diff --git a/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/requires.txt b/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/requires.txt
new file mode 100644
--- /dev/null
+++ b/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/requires.txt
@@ -0,0 +1,6 @@
+# this should be ignored
+
+strawberry >=0.5
+
+[section ignored]
+foo ==0.5
diff --git a/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/top_level.txt b/src/distutils2/_backport/tests/fake_dists/banana-0.4.egg/EGG-INFO/top_level.txt
new file mode 100644
diff --git a/src/distutils2/_backport/tests/fake_dists/strawberry-0.6.egg b/src/distutils2/_backport/tests/fake_dists/strawberry-0.6.egg
new file mode 100644
index 0000000000000000000000000000000000000000..6d160e8b161031ae52638514843592187925b757
GIT binary patch
literal 1402
zc$`yK)KALH(=X28%1l#;R!B%nEKbc!%uQ8LF-TCbRZuEUEh#N1$<NOz)-}*GOExsr
zEvPioGuBH at Pghci<toWY%~MFNNKIBKs4U6I&jU&+=q4*DW#$&-7nLX!R~G9i<QH3m
z7<vVXB^i2|dBv$kB^m}GVTed#QZb0uP**Wf*VMFDNGr<ERX|v)mz<xQo0ylPmzr2y
z84uK6l9-dD05v};Kfk27q$sffVnb?0W{Cz|OhZ#sNkK`)L5V9hr#O{MK_N9cBOmB5
zh0J1wy!;XcpxYIa^NWg7lS>r9UQWv|0ty0Ufu2)H%gjmDgJ}xL0otCbPy`8 at OrXVy
z$=M1e`3iV~M*-+)g_5F5g~as4%sjABpm0h{1UV)xlPkcRnT3l11j?rGw_!j6Vhl12
zuI}!-o_=or`X%`V at j0nwsX2Nj6(yk|oD9qiubF*7xU_<sfsy3}GXn#dK$usBW}XPL
zdBOgnLC&thaML(|CUIalO$4ZygTbxvn9cJCa*K8Yd7pq-5ZSZ<ZwLcyl*$Bd9}q?}
z%GU3+-(d%yJ- at a8FngYAJv`50)he!AtAbSdVkX?4u~u<++)e34e{3~_X5O|jF|J;B
zC3xwx-m|X72inV6*}Y^gtiD*1wpsGu-O%40%MDdGSw1j37$$7q`bcR1wJ4|Sg?Z{f
zrybfGuv%=l^lLM>i9RnLxBqqeU~_GC)B*8Q%^m+PITr6Z&8t31F+o4>xK^{d-p7?^
zx~J#&ZkWuS9<SEFoOjj3--Yb2mtJ3TW&iZ at 1KQ^&N4;mX5B=!V_JJD_rYtj1!&DME
z6jJkm at f=@}pP5%u3=ddwV4#ZQ4p;_;F99Leli98tF#_2jEDOYNBU4ffQu9($^O7s$
zb29U?!NvlU3?q{qGp<A<0cUS%tYT!qnS at v&NeIJT2(vL05VF~)kj+L(POOmRgw<qR
zi3wsdBiLjKCI)OtixrZz at HiMVb`TC;(kREwfG at 6CA#sJpco7ENF@*?^LS_Q-7U0dw
Q22#ojgxi7o*D`~60Pqom)c^nh

diff --git a/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1.dist-info/METADATA b/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1.dist-info/METADATA
--- a/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1.dist-info/METADATA
+++ b/src/distutils2/_backport/tests/fake_dists/towel_stuff-0.1.dist-info/METADATA
@@ -2,4 +2,6 @@
 Name: towel-stuff
 Version: 0.1
 Provides-Dist: truffles (1.1.2)
+Provides-Dist: towel-stuff (0.1)
 Obsoletes-Dist: truffles (!=0.8,<1.0)
+Requires-Dist: bacon (<=0.2)
diff --git a/src/distutils2/_backport/tests/test_pkgutil.py b/src/distutils2/_backport/tests/test_pkgutil.py
--- a/src/distutils2/_backport/tests/test_pkgutil.py
+++ b/src/distutils2/_backport/tests/test_pkgutil.py
@@ -1,8 +1,14 @@
 # -*- coding: utf-8 -*-
 """Tests for PEP 376 pkgutil functionality"""
+import unittest2
+import unittest2.compatibility
 import sys
 import os
 import csv
+import imp
+import tempfile
+import shutil
+import zipfile
 try:
     from hashlib import md5
 except ImportError:
@@ -10,31 +16,141 @@
 
 from test.test_support import run_unittest, TESTFN
 
-import distutils2._backport.pkgutil
-from distutils2.tests.support import unittest
+from distutils2._backport import pkgutil
 
 # TODO Add a test for getting a distribution that is provided by another
 #   distribution.
 
 # TODO Add a test for absolute pathed RECORD items (e.g. /etc/myapp/config.ini)
 
-class TestPkgUtilDistribution(unittest.TestCase):
+# Adapted from Python 2.7's trunk
+class TestPkgUtilData(unittest2.TestCase):
+
+    def setUp(self):
+        self.dirname = tempfile.mkdtemp()
+        sys.path.insert(0, self.dirname)
+
+    def tearDown(self):
+        del sys.path[0]
+        shutil.rmtree(self.dirname)
+
+    def test_getdata_filesys(self):
+        pkg = 'test_getdata_filesys'
+
+        # Include a LF and a CRLF, to test that binary data is read back
+        RESOURCE_DATA = 'Hello, world!\nSecond line\r\nThird line'
+
+        # Make a package with some resources
+        package_dir = os.path.join(self.dirname, pkg)
+        os.mkdir(package_dir)
+        # Empty init.py
+        f = open(os.path.join(package_dir, '__init__.py'), "wb")
+        f.close()
+        # Resource files, res.txt, sub/res.txt
+        f = open(os.path.join(package_dir, 'res.txt'), "wb")
+        f.write(RESOURCE_DATA)
+        f.close()
+        os.mkdir(os.path.join(package_dir, 'sub'))
+        f = open(os.path.join(package_dir, 'sub', 'res.txt'), "wb")
+        f.write(RESOURCE_DATA)
+        f.close()
+
+        # Check we can read the resources
+        res1 = pkgutil.get_data(pkg, 'res.txt')
+        self.assertEqual(res1, RESOURCE_DATA)
+        res2 = pkgutil.get_data(pkg, 'sub/res.txt')
+        self.assertEqual(res2, RESOURCE_DATA)
+
+        del sys.modules[pkg]
+
+    def test_getdata_zipfile(self):
+        zip = 'test_getdata_zipfile.zip'
+        pkg = 'test_getdata_zipfile'
+
+        # Include a LF and a CRLF, to test that binary data is read back
+        RESOURCE_DATA = 'Hello, world!\nSecond line\r\nThird line'
+
+        # Make a package with some resources
+        zip_file = os.path.join(self.dirname, zip)
+        z = zipfile.ZipFile(zip_file, 'w')
+
+        # Empty init.py
+        z.writestr(pkg + '/__init__.py', "")
+        # Resource files, res.txt, sub/res.txt
+        z.writestr(pkg + '/res.txt', RESOURCE_DATA)
+        z.writestr(pkg + '/sub/res.txt', RESOURCE_DATA)
+        z.close()
+
+        # Check we can read the resources
+        sys.path.insert(0, zip_file)
+        res1 = pkgutil.get_data(pkg, 'res.txt')
+        self.assertEqual(res1, RESOURCE_DATA)
+        res2 = pkgutil.get_data(pkg, 'sub/res.txt')
+        self.assertEqual(res2, RESOURCE_DATA)
+        del sys.path[0]
+
+        del sys.modules[pkg]
+
+# Adapted from Python 2.7's trunk
+class TestPkgUtilPEP302(unittest2.TestCase):
+
+    class MyTestLoader(object):
+        def load_module(self, fullname):
+            # Create an empty module
+            mod = sys.modules.setdefault(fullname, imp.new_module(fullname))
+            mod.__file__ = "<%s>" % self.__class__.__name__
+            mod.__loader__ = self
+            # Make it a package
+            mod.__path__ = []
+            # Count how many times the module is reloaded
+            mod.__dict__['loads'] = mod.__dict__.get('loads',0) + 1
+            return mod
+
+        def get_data(self, path):
+            return "Hello, world!"
+
+    class MyTestImporter(object):
+        def find_module(self, fullname, path=None):
+            return TestPkgUtilPEP302.MyTestLoader()
+
+    def setUp(self):
+        sys.meta_path.insert(0, self.MyTestImporter())
+
+    def tearDown(self):
+        del sys.meta_path[0]
+
+    def test_getdata_pep302(self):
+        # Use a dummy importer/loader
+        self.assertEqual(pkgutil.get_data('foo', 'dummy'), "Hello, world!")
+        del sys.modules['foo']
+
+    def test_alreadyloaded(self):
+        # Ensure that get_data works without reloading - the "loads" module
+        # variable in the example loader should count how many times a reload
+        # occurs.
+        import foo
+        self.assertEqual(foo.loads, 1)
+        self.assertEqual(pkgutil.get_data('foo', 'dummy'), "Hello, world!")
+        self.assertEqual(foo.loads, 1)
+        del sys.modules['foo']
+
+
+class TestPkgUtilDistribution(unittest2.TestCase):
     """Tests the pkgutil.Distribution class"""
 
     def setUp(self):
-        super(TestPkgUtilDistribution, self).setUp()
-
         self.fake_dists_path = os.path.abspath(
             os.path.join(os.path.dirname(__file__), 'fake_dists'))
-        self.distinfo_dirs = [ os.path.join(self.fake_dists_path, dir)
+
+        self.distinfo_dirs = [os.path.join(self.fake_dists_path, dir)
             for dir in os.listdir(self.fake_dists_path)
-            if dir.endswith('.dist-info')
-            ]
+            if dir.endswith('.dist-info')]
 
         def get_hexdigest(file):
             md5_hash = md5()
             md5_hash.update(open(file).read())
             return md5_hash.hexdigest()
+
         def record_pieces(file):
             path = os.path.relpath(file, sys.prefix)
             digest = get_hexdigest(file)
@@ -51,7 +167,8 @@
 
             for path, dirs, files in os.walk(dist_location):
                 for f in files:
-                    record_writer.writerow(record_pieces(os.path.join(path, f)))
+                    record_writer.writerow(record_pieces(
+                                           os.path.join(path, f)))
             for file in ['INSTALLER', 'METADATA', 'REQUESTED']:
                 record_writer.writerow(record_pieces(
                     os.path.join(distinfo_dir, file)))
@@ -60,7 +177,8 @@
             record_reader = csv.reader(open(record_file, 'rb'))
             record_data = []
             for row in record_reader:
-                path, md5_, size = row[:] + [ None for i in xrange(len(row), 3) ]
+                path, md5_, size = row[:] + \
+                                   [None for i in xrange(len(row), 3)]
                 record_data.append([path, (md5_, size,)])
             self.records[distinfo_dir] = dict(record_data)
 
@@ -69,7 +187,6 @@
         for distinfo_dir in self.distinfo_dirs:
             record_file = os.path.join(distinfo_dir, 'RECORD')
             open(record_file, 'w').close()
-        super(TestPkgUtilDistribution, self).tearDown()
 
     def test_instantiation(self):
         """Test the Distribution class's instantiation provides us with usable
@@ -108,7 +225,8 @@
         distinfo_name = 'grammar-1.0a4'
         distinfo_dir = os.path.join(self.fake_dists_path,
             distinfo_name + '.dist-info')
-        true_path = [self.fake_dists_path, distinfo_name, 'grammar', 'utils.py']
+        true_path = [self.fake_dists_path, distinfo_name, \
+                     'grammar', 'utils.py']
         true_path = os.path.relpath(os.path.join(*true_path), sys.prefix)
         false_path = [self.fake_dists_path, 'towel_stuff-0.1', 'towel_stuff',
             '__init__.py']
@@ -150,7 +268,8 @@
         self.assertRaises(DistutilsError, dist.get_distinfo_file,
             other_distinfo_file)
         # Test for a file that does not exist and should not exist
-        self.assertRaises(DistutilsError, dist.get_distinfo_file, 'ENTRYPOINTS')
+        self.assertRaises(DistutilsError, dist.get_distinfo_file, \
+                          'ENTRYPOINTS')
 
     def test_get_distinfo_files(self):
         """Test for the iteration of RECORD path entries."""
@@ -161,21 +280,19 @@
         dist = Distribution(distinfo_dir)
         # Test for the iteration of the raw path
         distinfo_record_paths = self.records[distinfo_dir].keys()
-        found = [ path for path in dist.get_distinfo_files() ]
+        found = [path for path in dist.get_distinfo_files()]
         self.assertEqual(sorted(found), sorted(distinfo_record_paths))
         # Test for the iteration of local absolute paths
-        distinfo_record_paths = [ os.path.join(sys.prefix, path)
-            for path in self.records[distinfo_dir].keys()
-            ]
-        found = [ path for path in dist.get_distinfo_files(local=True) ]
+        distinfo_record_paths = [os.path.join(sys.prefix, path)
+            for path in self.records[distinfo_dir].keys()]
+        found = [path for path in dist.get_distinfo_files(local=True)]
         self.assertEqual(sorted(found), sorted(distinfo_record_paths))
 
 
-class TestPkgUtilFunctions(unittest.TestCase):
+class TestPkgUtilPEP376(unittest2.TestCase):
     """Tests for the new functionality added in PEP 376."""
 
     def setUp(self):
-        super(TestPkgUtilFunctions, self).setUp()
         # Setup the path environment with our fake distributions
         current_path = os.path.abspath(os.path.dirname(__file__))
         self.sys_path = sys.path[:]
@@ -183,15 +300,15 @@
         sys.path[0:0] = [self.fake_dists_path]
 
     def tearDown(self):
-        super(TestPkgUtilFunctions, self).tearDown()
         sys.path[:] = self.sys_path
 
     def test_distinfo_dirname(self):
         """Given a name and a version, we expect the distinfo_dirname function
         to return a standard distribution information directory name."""
 
-        items = [ # (name, version, standard_dirname)
-            # Test for a very simple single word name and decimal version number
+        items = [# (name, version, standard_dirname)
+            # Test for a very simple single word name and decimal
+            # version number
             ('docutils', '0.5', 'docutils-0.5.dist-info'),
             # Test for another except this time with a '-' in the name, which
             #   needs to be transformed during the name lookup
@@ -221,7 +338,7 @@
                                                  EggInfoDistribution
 
         # Verify the fake dists have been found.
-        dists = [ dist for dist in get_distributions() ]
+        dists = [dist for dist in get_distributions()]
         for dist in dists:
             if not isinstance(dist, Distribution):
                 self.fail("item received was not a Distribution instance: "
@@ -234,10 +351,11 @@
         self.assertListEqual(sorted(found_dists), sorted(fake_dists))
 
         # Now, test if the egg-info distributions are found correctly as well
-        fake_dists += [('bacon', '0.1'), ('cheese', '2.0.2')]
+        fake_dists += [('bacon', '0.1'), ('cheese', '2.0.2'),
+                       ('banana', '0.4'), ('strawberry', '0.6')]
         found_dists = []
 
-        dists = [ dist for dist in get_distributions(use_egg_info=True) ]
+        dists = [dist for dist in get_distributions(use_egg_info=True)]
         for dist in dists:
             if not (isinstance(dist, Distribution) or \
                     isinstance(dist, EggInfoDistribution)):
@@ -248,7 +366,6 @@
 
         self.assertListEqual(sorted(fake_dists), sorted(found_dists))
 
-
     def test_get_distribution(self):
         """Test for looking up a distribution by name."""
         # Test the lookup of the towel-stuff distribution
@@ -274,6 +391,8 @@
         # instructed to
         self.assertEqual(None, get_distribution('bacon'))
         self.assertEqual(None, get_distribution('cheese'))
+        self.assertEqual(None, get_distribution('strawberry'))
+        self.assertEqual(None, get_distribution('banana'))
 
         # Now check that it works well in both situations, when egg-info
         # is a file and directory respectively.
@@ -285,6 +404,14 @@
         self.assertTrue(isinstance(dist, EggInfoDistribution))
         self.assertEqual(dist.name, 'bacon')
 
+        dist = get_distribution('banana', use_egg_info=True)
+        self.assertTrue(isinstance(dist, EggInfoDistribution))
+        self.assertEqual(dist.name, 'banana')
+
+        dist = get_distribution('strawberry', use_egg_info=True)
+        self.assertTrue(isinstance(dist, EggInfoDistribution))
+        self.assertEqual(dist.name, 'strawberry')
+
     def test_get_file_users(self):
         """Test the iteration of distributions that use a file."""
         from distutils2._backport.pkgutil import get_file_users, Distribution
@@ -300,7 +427,7 @@
         from distutils2._backport.pkgutil import provides_distribution
         from distutils2.errors import DistutilsError
 
-        checkLists = lambda x,y: self.assertListEqual(sorted(x), sorted(y))
+        checkLists = lambda x, y: self.assertListEqual(sorted(x), sorted(y))
 
         l = [dist.name for dist in provides_distribution('truffles')]
         checkLists(l, ['choxie', 'towel-stuff'])
@@ -318,10 +445,12 @@
         l = [dist.name for dist in provides_distribution('truffles', '1.1')]
         checkLists(l, ['towel-stuff'])
 
-        l = [dist.name for dist in provides_distribution('truffles', '!=1.1,<=2.0')]
+        l = [dist.name for dist in provides_distribution('truffles', \
+                                                         '!=1.1,<=2.0')]
         checkLists(l, ['choxie'])
 
-        l = [dist.name for dist in provides_distribution('truffles', '!=1.1,<=2.0',
+        l = [dist.name for dist in provides_distribution('truffles', \
+                                                         '!=1.1,<=2.0',
                                                           use_egg_info=True)]
         checkLists(l, ['choxie', 'bacon', 'cheese'])
 
@@ -338,12 +467,39 @@
         l = [dist.name for dist in provides_distribution('truffles', '>=1.0')]
         checkLists(l, ['choxie', 'towel-stuff'])
 
+        l = [dist.name for dist in provides_distribution('strawberry', '0.6',
+                                                         use_egg_info=True)]
+        checkLists(l, ['strawberry'])
+
+        l = [dist.name for dist in provides_distribution('strawberry', '>=0.5',
+                                                         use_egg_info=True)]
+        checkLists(l, ['strawberry'])
+
+
+        l = [dist.name for dist in provides_distribution('strawberry', '>0.6',
+                                                         use_egg_info=True)]
+        checkLists(l, [])
+
+
+        l = [dist.name for dist in provides_distribution('banana', '0.4',
+                                                         use_egg_info=True)]
+        checkLists(l, ['banana'])
+
+        l = [dist.name for dist in provides_distribution('banana', '>=0.3',
+                                                         use_egg_info=True)]
+        checkLists(l, ['banana'])
+
+
+        l = [dist.name for dist in provides_distribution('banana', '!=0.4',
+                                                         use_egg_info=True)]
+        checkLists(l, [])
+
     def test_obsoletes(self):
         """ Test looking for distributions based on what they obsolete """
         from distutils2._backport.pkgutil import obsoletes_distribution
         from distutils2.errors import DistutilsError
 
-        checkLists = lambda x,y: self.assertListEqual(sorted(x), sorted(y))
+        checkLists = lambda x, y: self.assertListEqual(sorted(x), sorted(y))
 
         l = [dist.name for dist in obsoletes_distribution('truffles', '1.0')]
         checkLists(l, [])
@@ -363,7 +519,8 @@
         l = [dist.name for dist in obsoletes_distribution('truffles', '0.9.6')]
         checkLists(l, ['choxie', 'towel-stuff'])
 
-        l = [dist.name for dist in obsoletes_distribution('truffles', '0.5.2.3')]
+        l = [dist.name for dist in obsoletes_distribution('truffles', \
+                                                          '0.5.2.3')]
         checkLists(l, ['choxie', 'towel-stuff'])
 
         l = [dist.name for dist in obsoletes_distribution('truffles', '0.2')]
@@ -371,27 +528,18 @@
 
 
 def test_suite():
-    suite = unittest.TestSuite()
-    testcase_loader = unittest.loader.defaultTestLoader.loadTestsFromTestCase
-    suite.addTest(testcase_loader(TestPkgUtilFunctions))
+    suite = unittest2.TestSuite()
+    testcase_loader = unittest2.loader.defaultTestLoader.loadTestsFromTestCase
+    suite.addTest(testcase_loader(TestPkgUtilData))
     suite.addTest(testcase_loader(TestPkgUtilDistribution))
+    suite.addTest(testcase_loader(TestPkgUtilPEP302))
+    suite.addTest(testcase_loader(TestPkgUtilPEP376))
     return suite
 
+
 def test_main():
     run_unittest(test_suite())
 
-if __name__ == "__main__":
-    test_main()
-
-def test_suite():
-    suite = unittest.TestSuite()
-    testcase_loader = unittest.loader.defaultTestLoader.loadTestsFromTestCase
-    suite.addTest(testcase_loader(TestPkgUtilFunctions))
-    suite.addTest(testcase_loader(TestPkgUtilDistribution))
-    return suite
-
-def test_main():
-    run_unittest(test_suite())
 
 if __name__ == "__main__":
     test_main()
diff --git a/src/distutils2/_backport/tests/test_sysconfig.py b/src/distutils2/_backport/tests/test_sysconfig.py
--- a/src/distutils2/_backport/tests/test_sysconfig.py
+++ b/src/distutils2/_backport/tests/test_sysconfig.py
@@ -8,7 +8,7 @@
 import os
 import shutil
 from copy import copy, deepcopy
-from ConfigParser import ConfigParser
+from ConfigParser import RawConfigParser
 
 from test.test_support import run_unittest, TESTFN
 
@@ -107,7 +107,7 @@
         wanted.sort()
         scheme = scheme.items()
         scheme.sort()
-        self.assertEquals(scheme, wanted)
+        self.assertEqual(scheme, wanted)
 
     def test_get_config_vars(self):
         cvars = get_config_vars()
@@ -120,21 +120,21 @@
         sys.version = ('2.4.4 (#71, Oct 18 2006, 08:34:43) '
                        '[MSC v.1310 32 bit (Intel)]')
         sys.platform = 'win32'
-        self.assertEquals(get_platform(), 'win32')
+        self.assertEqual(get_platform(), 'win32')
 
         # windows XP, amd64
         os.name = 'nt'
         sys.version = ('2.4.4 (#71, Oct 18 2006, 08:34:43) '
                        '[MSC v.1310 32 bit (Amd64)]')
         sys.platform = 'win32'
-        self.assertEquals(get_platform(), 'win-amd64')
+        self.assertEqual(get_platform(), 'win-amd64')
 
         # windows XP, itanium
         os.name = 'nt'
         sys.version = ('2.4.4 (#71, Oct 18 2006, 08:34:43) '
                        '[MSC v.1310 32 bit (Itanium)]')
         sys.platform = 'win32'
-        self.assertEquals(get_platform(), 'win-ia64')
+        self.assertEqual(get_platform(), 'win-ia64')
 
         # macbook
         os.name = 'posix'
@@ -153,9 +153,9 @@
         maxint = sys.maxint
         try:
             sys.maxint = 2147483647
-            self.assertEquals(get_platform(), 'macosx-10.3-ppc')
+            self.assertEqual(get_platform(), 'macosx-10.3-ppc')
             sys.maxint = 9223372036854775807
-            self.assertEquals(get_platform(), 'macosx-10.3-ppc64')
+            self.assertEqual(get_platform(), 'macosx-10.3-ppc64')
         finally:
             sys.maxint = maxint
 
@@ -173,9 +173,9 @@
         maxint = sys.maxint
         try:
             sys.maxint = 2147483647
-            self.assertEquals(get_platform(), 'macosx-10.3-i386')
+            self.assertEqual(get_platform(), 'macosx-10.3-i386')
             sys.maxint = 9223372036854775807
-            self.assertEquals(get_platform(), 'macosx-10.3-x86_64')
+            self.assertEqual(get_platform(), 'macosx-10.3-x86_64')
         finally:
             sys.maxint = maxint
 
@@ -186,33 +186,33 @@
                                        '-fno-strict-aliasing -fno-common '
                                        '-dynamic -DNDEBUG -g -O3')
 
-        self.assertEquals(get_platform(), 'macosx-10.4-fat')
+        self.assertEqual(get_platform(), 'macosx-10.4-fat')
 
         get_config_vars()['CFLAGS'] = ('-arch x86_64 -arch i386 -isysroot '
                                        '/Developer/SDKs/MacOSX10.4u.sdk  '
                                        '-fno-strict-aliasing -fno-common '
                                        '-dynamic -DNDEBUG -g -O3')
 
-        self.assertEquals(get_platform(), 'macosx-10.4-intel')
+        self.assertEqual(get_platform(), 'macosx-10.4-intel')
 
         get_config_vars()['CFLAGS'] = ('-arch x86_64 -arch ppc -arch i386 -isysroot '
                                        '/Developer/SDKs/MacOSX10.4u.sdk  '
                                        '-fno-strict-aliasing -fno-common '
                                        '-dynamic -DNDEBUG -g -O3')
-        self.assertEquals(get_platform(), 'macosx-10.4-fat3')
+        self.assertEqual(get_platform(), 'macosx-10.4-fat3')
 
         get_config_vars()['CFLAGS'] = ('-arch ppc64 -arch x86_64 -arch ppc -arch i386 -isysroot '
                                        '/Developer/SDKs/MacOSX10.4u.sdk  '
                                        '-fno-strict-aliasing -fno-common '
                                        '-dynamic -DNDEBUG -g -O3')
-        self.assertEquals(get_platform(), 'macosx-10.4-universal')
+        self.assertEqual(get_platform(), 'macosx-10.4-universal')
 
         get_config_vars()['CFLAGS'] = ('-arch x86_64 -arch ppc64 -isysroot '
                                        '/Developer/SDKs/MacOSX10.4u.sdk  '
                                        '-fno-strict-aliasing -fno-common '
                                        '-dynamic -DNDEBUG -g -O3')
 
-        self.assertEquals(get_platform(), 'macosx-10.4-fat64')
+        self.assertEqual(get_platform(), 'macosx-10.4-fat64')
 
         for arch in ('ppc', 'i386', 'x86_64', 'ppc64'):
             get_config_vars()['CFLAGS'] = ('-arch %s -isysroot '
@@ -220,7 +220,7 @@
                                            '-fno-strict-aliasing -fno-common '
                                            '-dynamic -DNDEBUG -g -O3'%(arch,))
 
-            self.assertEquals(get_platform(), 'macosx-10.4-%s'%(arch,))
+            self.assertEqual(get_platform(), 'macosx-10.4-%s'%(arch,))
 
         # linux debian sarge
         os.name = 'posix'
@@ -230,7 +230,7 @@
         self._set_uname(('Linux', 'aglae', '2.6.21.1dedibox-r7',
                     '#1 Mon Apr 30 17:25:38 CEST 2007', 'i686'))
 
-        self.assertEquals(get_platform(), 'linux-i686')
+        self.assertEqual(get_platform(), 'linux-i686')
 
         # XXX more platforms to tests here
 
@@ -241,11 +241,11 @@
     def test_get_scheme_names(self):
         wanted = ('nt', 'nt_user', 'os2', 'os2_home', 'posix_home',
                   'posix_prefix', 'posix_user')
-        self.assertEquals(get_scheme_names(), wanted)
+        self.assertEqual(get_scheme_names(), wanted)
 
     def test_expand_globals(self):
 
-        config = ConfigParser()
+        config = RawConfigParser()
         config.add_section('globals')
         config.set('globals', 'foo', 'ok')
         config.add_section('posix')
@@ -254,8 +254,8 @@
 
         _expand_globals(config)
 
-        self.assertEquals(config.get('posix', 'foo'), 'ok')
-        self.assertEquals(config.get('posix', 'more'), '/etc/ok')
+        self.assertEqual(config.get('posix', 'foo'), 'ok')
+        self.assertEqual(config.get('posix', 'more'), '/etc/ok')
 
         # we might not have globals after all
         # extending again (==no more globals section)
diff --git a/src/distutils2/command/bdist.py b/src/distutils2/command/bdist.py
--- a/src/distutils2/command/bdist.py
+++ b/src/distutils2/command/bdist.py
@@ -62,7 +62,7 @@
                       'os2': 'zip'}
 
     # Establish the preferred order (for the --help-formats option).
-    format_commands = ['rpm', 'gztar', 'bztar', 'ztar', 'tar',
+    format_commands = ['gztar', 'bztar', 'ztar', 'tar',
                        'wininst', 'zip', 'msi']
 
     # And the real information.
@@ -96,7 +96,7 @@
 
         # 'bdist_base' -- parent of per-built-distribution-format
         # temporary directories (eg. we'll probably have
-        # "build/bdist.<plat>/dumb", "build/bdist.<plat>/rpm", etc.)
+        # "build/bdist.<plat>/dumb", etc.)
         if self.bdist_base is None:
             build_base = self.get_finalized_command('build').build_base
             self.bdist_base = os.path.join(build_base,
@@ -126,7 +126,7 @@
         # Reinitialize and run each command.
         for i in range(len(self.formats)):
             cmd_name = commands[i]
-            sub_cmd = self.reinitialize_command(cmd_name)
+            sub_cmd = self.get_reinitialized_command(cmd_name)
 
             # passing the owner and group names for tar archiving
             if cmd_name == 'bdist_dumb':
diff --git a/src/distutils2/command/bdist_dumb.py b/src/distutils2/command/bdist_dumb.py
--- a/src/distutils2/command/bdist_dumb.py
+++ b/src/distutils2/command/bdist_dumb.py
@@ -85,7 +85,7 @@
         if not self.skip_build:
             self.run_command('build')
 
-        install = self.reinitialize_command('install', reinit_subcommands=1)
+        install = self.get_reinitialized_command('install', reinit_subcommands=1)
         install.root = self.bdist_dir
         install.skip_build = self.skip_build
         install.warn_dir = 0
diff --git a/src/distutils2/command/bdist_msi.py b/src/distutils2/command/bdist_msi.py
--- a/src/distutils2/command/bdist_msi.py
+++ b/src/distutils2/command/bdist_msi.py
@@ -177,12 +177,12 @@
         if not self.skip_build:
             self.run_command('build')
 
-        install = self.reinitialize_command('install', reinit_subcommands=1)
+        install = self.get_reinitialized_command('install', reinit_subcommands=1)
         install.prefix = self.bdist_dir
         install.skip_build = self.skip_build
         install.warn_dir = 0
 
-        install_lib = self.reinitialize_command('install_lib')
+        install_lib = self.get_reinitialized_command('install_lib')
         # we do not want to include pyc or pyo files
         install_lib.compile = 0
         install_lib.optimize = 0
diff --git a/src/distutils2/command/bdist_wininst.py b/src/distutils2/command/bdist_wininst.py
--- a/src/distutils2/command/bdist_wininst.py
+++ b/src/distutils2/command/bdist_wininst.py
@@ -126,13 +126,13 @@
         if not self.skip_build:
             self.run_command('build')
 
-        install = self.reinitialize_command('install', reinit_subcommands=1)
+        install = self.get_reinitialized_command('install', reinit_subcommands=1)
         install.root = self.bdist_dir
         install.skip_build = self.skip_build
         install.warn_dir = 0
         install.plat_name = self.plat_name
 
-        install_lib = self.reinitialize_command('install_lib')
+        install_lib = self.get_reinitialized_command('install_lib')
         # we do not want to include pyc or pyo files
         install_lib.compile = 0
         install_lib.optimize = 0
diff --git a/src/distutils2/command/build_py.py b/src/distutils2/command/build_py.py
--- a/src/distutils2/command/build_py.py
+++ b/src/distutils2/command/build_py.py
@@ -6,14 +6,65 @@
 
 import os
 import sys
+import logging
 from glob import glob
 
 from distutils2.core import Command
 from distutils2.errors import DistutilsOptionError, DistutilsFileError
 from distutils2.util import convert_path
-from distutils2 import log
+from distutils2.converter.refactor import DistutilsRefactoringTool
 
-class build_py(Command):
+# marking public APIs
+__all__ = ['Mixin2to3', 'build_py']
+
+try:
+    from distutils2.util import Mixin2to3 as _Mixin2to3
+    from lib2to3.refactor import get_fixers_from_package
+    _CONVERT = True
+    _KLASS = _Mixin2to3
+except ImportError:
+    _CONVERT = False
+    _KLASS = object
+
+class Mixin2to3(_KLASS):
+    """ The base class which can be used for refactoring. When run under
+    Python 3.0, the run_2to3 method provided by Mixin2to3 is overridden.
+    When run on Python 2.x, it merely creates a class which overrides run_2to3,
+    yet does nothing in particular with it.
+    """
+    if _CONVERT:
+        def _run_2to3(self, files, doctests=[]):
+            """ Takes a list of files and doctests, and performs conversion
+            on those.
+              - First, the files which contain the code(`files`) are converted.
+              - Second, the doctests in `files` are converted.
+              - Thirdly, the doctests in `doctests` are converted.
+            """
+
+            # Convert the ".py" files.
+            logging.info("Converting Python code")
+            _KLASS.run_2to3(self, files)
+
+            # Convert the doctests in the ".py" files.
+            logging.info("Converting doctests with '.py' files")
+            _KLASS.run_2to3(self, files, doctests_only=True)
+
+            # If the following conditions are met, then convert:-
+            # 1. User has specified the 'convert_2to3_doctests' option. So, we
+            #    can expect that the list 'doctests' is not empty.
+            # 2. The default is allow distutils2 to allow conversion of text files
+            #    containing doctests. It is set as
+            #    distutils2.run_2to3_on_doctests
+
+            if doctests != [] and distutils2.run_2to3_on_doctests:
+                logging.info("Converting text files which contain doctests")
+                _KLASS.run_2to3(self, doctests, doctests_only=True)
+    else:
+        # If run on Python 2.x, there is nothing to do.
+        def _run_2to3(self, files, doctests=[]):
+            pass
+
+class build_py(Command, Mixin2to3):
 
     description = "\"build\" pure Python modules (copy to build directory)"
 
@@ -39,6 +90,8 @@
         self.compile = 0
         self.optimize = 0
         self.force = None
+        self._updated_files = []
+        self._doctests_2to3 = []
 
     def finalize_options(self):
         self.set_undefined_options('build',
@@ -93,6 +146,9 @@
             self.build_packages()
             self.build_package_data()
 
+        if self.distribution.use_2to3 and self_updated_files:
+            self.run_2to3(self._updated_files, self._doctests_2to3)
+
         self.byte_compile(self.get_outputs(include_bytecode=0))
 
     # -- Top-level worker functions ------------------------------------
@@ -148,8 +204,10 @@
             for filename in filenames:
                 target = os.path.join(build_dir, filename)
                 self.mkpath(os.path.dirname(target))
-                self.copy_file(os.path.join(src_dir, filename), target,
-                               preserve_mode=False)
+                outf, copied = self.copy_file(os.path.join(src_dir, filename),
+                               target, preserve_mode=False)
+                if copied and srcfile in self.distribution.convert_2to3.doctests:
+                    self._doctests_2to3.append(outf)
 
     # XXX - this should be moved to the Distribution class as it is not
     # only needed for build_py. It also has no dependencies on this class.
@@ -215,8 +273,8 @@
             if os.path.isfile(init_py):
                 return init_py
             else:
-                log.warn(("package init file '%s' not found " +
-                          "(or not a regular file)"), init_py)
+                logging.warning(("package init file '%s' not found " +
+                                 "(or not a regular file)"), init_py)
 
         # Either not in a package at all (__init__.py not expected), or
         # __init__.py doesn't exist -- so don't return the filename.
@@ -224,7 +282,8 @@
 
     def check_module(self, module, module_file):
         if not os.path.isfile(module_file):
-            log.warn("file %s (for module %s) not found", module_file, module)
+            logging.warning("file %s (for module %s) not found",
+                            module_file, module)
             return False
         else:
             return True
diff --git a/src/distutils2/command/cmd.py b/src/distutils2/command/cmd.py
--- a/src/distutils2/command/cmd.py
+++ b/src/distutils2/command/cmd.py
@@ -19,7 +19,7 @@
 except ImportError:
     from distutils2._backport.shutil import make_archive
 
-class Command:
+class Command(object):
     """Abstract base class for defining command classes, the "worker bees"
     of the Distutils.  A useful analogy for command classes is to think of
     them as subroutines with local variables called "options".  The options
@@ -57,8 +57,7 @@
     def __init__(self, dist):
         """Create and initialize a new Command object.  Most importantly,
         invokes the 'initialize_options()' method, which is the real
-        initializer and depends on the actual command being
-        instantiated.
+        initializer and depends on the actual command being instantiated.
         """
         # late import because of mutual dependence between these classes
         from distutils2.dist import Distribution
@@ -333,10 +332,8 @@
         cmd_obj.ensure_finalized()
         return cmd_obj
 
-    # XXX rename to 'get_reinitialized_command()'? (should do the
-    # same in dist.py, if so)
-    def reinitialize_command(self, command, reinit_subcommands=0):
-        return self.distribution.reinitialize_command(
+    def get_reinitialized_command(self, command, reinit_subcommands=0):
+        return self.distribution.get_reinitialized_command(
             command, reinit_subcommands)
 
     def run_command(self, command):
diff --git a/src/distutils2/command/sdist.py b/src/distutils2/command/sdist.py
--- a/src/distutils2/command/sdist.py
+++ b/src/distutils2/command/sdist.py
@@ -20,7 +20,7 @@
 from distutils2.core import Command
 from distutils2 import util
 from distutils2.errors import (DistutilsPlatformError, DistutilsOptionError,
-                              DistutilsTemplateError)
+                               DistutilsTemplateError)
 from distutils2.manifest import Manifest
 from distutils2 import log
 from distutils2.util import convert_path, newer
diff --git a/src/distutils2/command/upload.py b/src/distutils2/command/upload.py
--- a/src/distutils2/command/upload.py
+++ b/src/distutils2/command/upload.py
@@ -160,7 +160,7 @@
         # send the data
         try:
             result = urlopen(request)
-            status = result.getcode()
+            status = result.code
             reason = result.msg
         except socket.error, e:
             self.announce(str(e), log.ERROR)
diff --git a/src/distutils2/command/upload_docs.py b/src/distutils2/command/upload_docs.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/command/upload_docs.py
@@ -0,0 +1,135 @@
+import base64, httplib, os.path, socket, tempfile, urlparse, zipfile
+from cStringIO import StringIO
+from distutils2 import log
+from distutils2.command.upload import upload
+from distutils2.core import PyPIRCCommand
+from distutils2.errors import DistutilsFileError
+
+def zip_dir(directory):
+    """Compresses recursively contents of directory into a StringIO object"""
+    destination = StringIO()
+    zip_file = zipfile.ZipFile(destination, "w")
+    for root, dirs, files in os.walk(directory):
+        for name in files:
+            full = os.path.join(root, name)
+            relative = root[len(directory):].lstrip(os.path.sep)
+            dest = os.path.join(relative, name)
+            zip_file.write(full, dest)
+    zip_file.close()
+    return destination
+
+# grabbed from
+#    http://code.activestate.com/recipes/146306-http-client-to-post-using-multipartform-data/
+def encode_multipart(fields, files, boundary=None):
+    """
+    fields is a sequence of (name, value) elements for regular form fields.
+    files is a sequence of (name, filename, value) elements for data to be uploaded as files
+    Return (content_type, body) ready for httplib.HTTP instance
+    """
+    if boundary is None:
+        boundary = '--------------GHSKFJDLGDS7543FJKLFHRE75642756743254'
+    l = []
+    for (key, value) in fields:
+        l.extend([
+            '--' + boundary,
+            'Content-Disposition: form-data; name="%s"' % key,
+            '',
+            value])
+    for (key, filename, value) in files:
+        l.extend([
+            '--' + boundary,
+            'Content-Disposition: form-data; name="%s"; filename="%s"' % (key, filename),
+            '',
+            value])
+    l.append('--' + boundary + '--')
+    l.append('')
+    body =  '\r\n'.join(l)
+    content_type = 'multipart/form-data; boundary=%s' % boundary
+    return content_type, body
+
+class upload_docs(PyPIRCCommand):
+
+    user_options = [
+        ('repository=', 'r', "url of repository [default: %s]" % upload.DEFAULT_REPOSITORY),
+        ('show-response', None, 'display full response text from server'),
+        ('upload-dir=', None, 'directory to upload'),
+        ]
+
+    def initialize_options(self):
+        PyPIRCCommand.initialize_options(self)
+        self.upload_dir = "build/docs"
+
+    def finalize_options(self):
+        PyPIRCCommand.finalize_options(self)
+        if self.upload_dir == None:
+            build = self.get_finalized_command('build')
+            self.upload_dir = os.path.join(build.build_base, "docs")
+        self.announce('Using upload directory %s' % self.upload_dir)
+        self.verify_upload_dir(self.upload_dir)
+        config = self._read_pypirc()
+        if config != {}:
+            self.username = config['username']
+            self.password = config['password']
+            self.repository = config['repository']
+            self.realm = config['realm']
+
+    def verify_upload_dir(self, upload_dir):
+        self.ensure_dirname('upload_dir')
+        index_location = os.path.join(upload_dir, "index.html")
+        if not os.path.exists(index_location):
+            mesg = "No 'index.html found in docs directory (%s)"
+            raise DistutilsFileError(mesg % upload_dir)
+
+    def run(self):
+        tmp_dir = tempfile.mkdtemp()
+        name = self.distribution.metadata['Name']
+        zip_file = zip_dir(self.upload_dir)
+
+        fields = {':action': 'doc_upload', 'name': name}.items()
+        files = [('content', name, zip_file.getvalue())]
+        content_type, body = encode_multipart(fields, files)
+
+        credentials = self.username + ':' + self.password
+        auth = "Basic " + base64.encodestring(credentials).strip()
+
+        self.announce("Submitting documentation to %s" % (self.repository),
+                      log.INFO)
+
+        schema, netloc, url, params, query, fragments = \
+            urlparse.urlparse(self.repository)
+        if schema == "http":
+            conn = httplib.HTTPConnection(netloc)
+        elif schema == "https":
+            conn = httplib.HTTPSConnection(netloc)
+        else:
+            raise AssertionError("unsupported schema "+schema)
+
+        try:
+            conn.connect()
+            conn.putrequest("POST", url)
+            conn.putheader('Content-type', content_type)
+            conn.putheader('Content-length', str(len(body)))
+            conn.putheader('Authorization', auth)
+            conn.endheaders()
+            conn.send(body)
+        except socket.error, e:
+            self.announce(str(e), log.ERROR)
+            return
+
+        r = conn.getresponse()
+
+        if r.status == 200:
+            self.announce('Server response (%s): %s' % (r.status, r.reason),
+                          log.INFO)
+        elif r.status == 301:
+            location = r.getheader('Location')
+            if location is None:
+                location = 'http://packages.python.org/%s/' % meta.get_name()
+            self.announce('Upload successful. Visit %s' % location,
+                          log.INFO)
+        else:
+            self.announce('Upload failed (%s): %s' % (r.status, r.reason),
+                          log.ERROR)
+
+        if self.show_response:
+            print "\n".join(['-'*75, r.read(), '-'*75])
diff --git a/src/distutils2/compiler/bcppcompiler.py b/src/distutils2/compiler/bcppcompiler.py
--- a/src/distutils2/compiler/bcppcompiler.py
+++ b/src/distutils2/compiler/bcppcompiler.py
@@ -16,7 +16,7 @@
 import os
 
 from distutils2.errors import (DistutilsExecError, CompileError, LibError,
-                              LinkError, UnknownFileError)
+                               LinkError, UnknownFileError)
 from distutils2.compiler.ccompiler import CCompiler, gen_preprocess_options
 from distutils2.file_util import write_file
 from distutils2.dep_util import newer
diff --git a/src/distutils2/compiler/ccompiler.py b/src/distutils2/compiler/ccompiler.py
--- a/src/distutils2/compiler/ccompiler.py
+++ b/src/distutils2/compiler/ccompiler.py
@@ -10,7 +10,7 @@
 import re
 
 from distutils2.errors import (CompileError, LinkError, UnknownFileError,
-                              DistutilsPlatformError, DistutilsModuleError)
+                               DistutilsPlatformError, DistutilsModuleError)
 from distutils2.spawn import spawn
 from distutils2.util import split_quoted, execute, newer_group
 from distutils2 import log
@@ -76,7 +76,7 @@
 
         compiler.shared_lib_extension = so_ext
 
-class CCompiler:
+class CCompiler(object):
     """Abstract base class to define the interface that must be implemented
     by real compiler classes.  Also has some utility methods used by
     several compiler classes.
@@ -800,14 +800,16 @@
             library_dirs = []
         fd, fname = tempfile.mkstemp(".c", funcname, text=True)
         f = os.fdopen(fd, "w")
-        for incl in includes:
-            f.write("""#include "%s"\n""" % incl)
-        f.write("""\
+        try:
+            for incl in includes:
+                f.write("""#include "%s"\n""" % incl)
+            f.write("""\
 main (int argc, char **argv) {
     %s();
 }
 """ % funcname)
-        f.close()
+        finally:
+            f.close()
         try:
             objects = self.compile([fname], include_dirs=include_dirs)
         except CompileError:
@@ -1048,7 +1050,7 @@
         module_name = "distutils2.compiler." + module_name
         __import__ (module_name)
         module = sys.modules[module_name]
-        klass = vars(module)[class_name]
+        cls = vars(module)[class_name]
     except ImportError:
         raise DistutilsModuleError, \
               "can't compile C/C++ code: unable to load module '%s'" % \
@@ -1061,7 +1063,7 @@
     # XXX The None is necessary to preserve backwards compatibility
     # with classes that expect verbose to be the first positional
     # argument.
-    return klass(None, dry_run, force)
+    return cls(None, dry_run, force)
 
 
 def gen_preprocess_options(macros, include_dirs):
diff --git a/src/distutils2/compiler/emxccompiler.py b/src/distutils2/compiler/emxccompiler.py
--- a/src/distutils2/compiler/emxccompiler.py
+++ b/src/distutils2/compiler/emxccompiler.py
@@ -25,9 +25,8 @@
 from warnings import warn
 
 from distutils2.compiler.unixccompiler import UnixCCompiler
-from distutils2.util import write_file
 from distutils2.errors import DistutilsExecError, CompileError, UnknownFileError
-from distutils2.util import get_compiler_versions
+from distutils2.util import get_compiler_versions, write_file
 
 class EMXCCompiler (UnixCCompiler):
 
@@ -273,8 +272,10 @@
         # It would probably better to read single lines to search.
         # But we do this only once, and it is fast enough
         f = open(fn)
-        s = f.read()
-        f.close()
+        try:
+            s = f.read()
+        finally:
+            f.close()
 
     except IOError, exc:
         # if we can't read this file, we cannot say it is wrong
diff --git a/src/distutils2/compiler/msvc9compiler.py b/src/distutils2/compiler/msvc9compiler.py
--- a/src/distutils2/compiler/msvc9compiler.py
+++ b/src/distutils2/compiler/msvc9compiler.py
@@ -20,7 +20,7 @@
 import re
 
 from distutils2.errors import (DistutilsExecError, DistutilsPlatformError,
-                              CompileError, LibError, LinkError)
+                               CompileError, LibError, LinkError)
 from distutils2.compiler.ccompiler import CCompiler, gen_lib_options
 from distutils2 import log
 from distutils2.util import get_platform
@@ -50,7 +50,7 @@
     'win-ia64' : 'ia64',
 }
 
-class Reg:
+class Reg(object):
     """Helper class to read values from the registry
     """
 
@@ -112,7 +112,7 @@
         return s
     convert_mbcs = staticmethod(convert_mbcs)
 
-class MacroExpander:
+class MacroExpander(object):
 
     def __init__(self, version):
         self.macros = {}
diff --git a/src/distutils2/compiler/msvccompiler.py b/src/distutils2/compiler/msvccompiler.py
--- a/src/distutils2/compiler/msvccompiler.py
+++ b/src/distutils2/compiler/msvccompiler.py
@@ -15,7 +15,7 @@
 import string
 
 from distutils2.errors import (DistutilsExecError, DistutilsPlatformError,
-                              CompileError, LibError, LinkError)
+                               CompileError, LibError, LinkError)
 from distutils2.compiler.ccompiler import CCompiler, gen_lib_options
 from distutils2 import log
 
@@ -104,7 +104,7 @@
             pass
     return s
 
-class MacroExpander:
+class MacroExpander(object):
 
     def __init__(self, version):
         self.macros = {}
diff --git a/src/distutils2/compiler/unixccompiler.py b/src/distutils2/compiler/unixccompiler.py
--- a/src/distutils2/compiler/unixccompiler.py
+++ b/src/distutils2/compiler/unixccompiler.py
@@ -19,10 +19,10 @@
 from types import StringType, NoneType
 
 from distutils2.util import newer
-from distutils2.compiler.ccompiler import \
-     CCompiler, gen_preprocess_options, gen_lib_options
-from distutils2.errors import \
-     DistutilsExecError, CompileError, LibError, LinkError
+from distutils2.compiler.ccompiler import (CCompiler, gen_preprocess_options,
+                                           gen_lib_options)
+from distutils2.errors import (DistutilsExecError, CompileError,
+                               LibError, LinkError)
 from distutils2 import log
 
 try:
diff --git a/src/distutils2/config.py b/src/distutils2/config.py
--- a/src/distutils2/config.py
+++ b/src/distutils2/config.py
@@ -4,7 +4,7 @@
 that uses .pypirc in the distutils.command package.
 """
 import os
-from ConfigParser import ConfigParser
+from ConfigParser import RawConfigParser
 
 from distutils2.command.cmd import Command
 
@@ -59,7 +59,7 @@
         if os.path.exists(rc):
             self.announce('Using PyPI login from %s' % rc)
             repository = self.repository or self.DEFAULT_REPOSITORY
-            config = ConfigParser()
+            config = RawConfigParser()
             config.read(rc)
             sections = config.sections()
             if 'distutils' in sections:
diff --git a/src/distutils2/core.py b/src/distutils2/core.py
--- a/src/distutils2/core.py
+++ b/src/distutils2/core.py
@@ -2,8 +2,9 @@
 
 The only module that needs to be imported to use the Distutils; provides
 the 'setup' function (which is to be called from the setup script).  Also
-indirectly provides the Distribution and Command classes, although they are
-really defined in distutils2.dist and distutils2.cmd.
+exports useful classes so that setup scripts can import them from here
+although they are really defined in other modules: Distribution, Command,
+PyPIRCommand, Extension, find_packages.
 """
 
 __revision__ = "$Id: core.py 77704 2010-01-23 09:23:15Z tarek.ziade $"
@@ -12,7 +13,7 @@
 import os
 
 from distutils2.errors import (DistutilsSetupError, DistutilsArgError,
-                              DistutilsError, CCompilerError)
+                               DistutilsError, CCompilerError)
 from distutils2.util import grok_environment_error
 
 # Mainly import these so setup scripts can "from distutils2.core import" them.
@@ -20,6 +21,7 @@
 from distutils2.command.cmd import Command
 from distutils2.config import PyPIRCCommand
 from distutils2.extension import Extension
+from distutils2.util import find_packages
 
 # This is a barebones help message generated displayed when the user
 # runs the setup script with no arguments at all.  More useful help
@@ -47,7 +49,8 @@
                   'maintainer', 'maintainer_email', 'url', 'license',
                   'description', 'long_description', 'keywords',
                   'platforms', 'classifiers', 'download_url',
-                  'requires', 'provides', 'obsoletes',
+                  'requires', 'provides', 'obsoletes', 'use_2to3',
+                  'convert_2to3_doctests',
                   )
 
 # Legal keyword arguments for the Extension constructor
@@ -94,11 +97,7 @@
 
     # Determine the distribution class -- either caller-supplied or
     # our Distribution (see below).
-    klass = attrs.get('distclass')
-    if klass:
-        del attrs['distclass']
-    else:
-        klass = Distribution
+    distclass = attrs.pop('distclass', Distribution)
 
     if 'script_name' not in attrs:
         attrs['script_name'] = os.path.basename(sys.argv[0])
@@ -108,7 +107,7 @@
     # Create the Distribution instance, using the remaining arguments
     # (ie. everything except distclass) to initialize it
     try:
-        _setup_distribution = dist = klass(attrs)
+        _setup_distribution = dist = distclass(attrs)
     except DistutilsSetupError, msg:
         if 'name' in attrs:
             raise SystemExit, "error in %s setup command: %s" % \
diff --git a/src/distutils2/depgraph.py b/src/distutils2/depgraph.py
--- a/src/distutils2/depgraph.py
+++ b/src/distutils2/depgraph.py
@@ -1,47 +1,49 @@
-"""
-A dependency graph generator. The graph is represented as an instance of
-:class:`DependencyGraph`, and DOT output is possible as well.
+"""Analyse the relationships between the distributions in the system and generate
+a dependency graph.
 """
 
-from distutils2._backport import pkgutil
 from distutils2.errors import DistutilsError
 from distutils2.version import VersionPredicate
 
-__all__ = ['DependencyGraph', 'generate_graph']
+__all__ = ['DependencyGraph', 'generate_graph', 'dependent_dists',
+           'graph_to_dot']
 
 
 class DependencyGraph(object):
     """
     Represents a dependency graph between distributions.
 
-    The depedency relationships are stored in an *adjacency_list* that maps
+    The dependency relationships are stored in an ``adjacency_list`` that maps
     distributions to a list of ``(other, label)`` tuples where  ``other``
     is a distribution and the edge is labelled with ``label`` (i.e. the version
-    specifier, if such was provided). If any missing depencies are found,
-    they are stored in ``missing``. It maps distributions to a list of
-    requirements that were not provided by any other distributions.
+    specifier, if such was provided). Also, for more efficient traversal, for
+    every distribution ``x``, a list of predecessors is kept in
+    ``reverse_list[x]``. An edge from distribution ``a`` to
+    distribution ``b`` means that ``a`` depends on ``b``. If any missing
+    depencies are found, they are stored in ``missing``, which is a dictionary
+    that maps distributions to a list of requirements that were not provided by
+    any other distributions.
     """
 
     def __init__(self):
         self.adjacency_list = {}
+        self.reverse_list = {}
         self.missing = {}
 
     def add_distribution(self, distribution):
-        """
-        Add distribution *x* to the graph.
+        """Add the *distribution* to the graph.
 
         :type distribution: :class:`pkgutil.Distribution` or
                             :class:`pkgutil.EggInfoDistribution`
         """
         self.adjacency_list[distribution] = list()
+        self.reverse_list[distribution] = list()
         self.missing[distribution] = list()
 
     def add_edge(self, x, y, label=None):
-        """
-        Add an edge from distribution *x* to distribution *y* with the given
+        """Add an edge from distribution *x* to distribution *y* with the given
         *label*.
 
-
         :type x: :class:`pkgutil.Distribution` or
                  :class:`pkgutil.EggInfoDistribution`
         :type y: :class:`pkgutil.Distribution` or
@@ -49,6 +51,9 @@
         :type label: ``str`` or ``None``
         """
         self.adjacency_list[x].append((y, label))
+        # multiple edges are allowed, so be careful
+        if not x in self.reverse_list[y]:
+            self.reverse_list[y].append(x)
 
     def add_missing(self, distribution, requirement):
         """
@@ -60,45 +65,42 @@
         """
         self.missing[distribution].append(requirement)
 
-    def to_dot(self, f, skip_disconnected=True):
-        """
-        Writes a DOT output for the graph to the provided *file*.
-        If *skip_disconnected* is set to ``True``, then all distributions
-        that are not dependent on any other distributions are skipped.
 
-        :type f: ``file``
-        ;type skip_disconnected: ``bool``
-        """
-        if not isinstance(f, file):
-            raise TypeError('the argument has to be of type file')
+def graph_to_dot(graph, f, skip_disconnected=True):
+    """Writes a DOT output for the graph to the provided file *f*.
 
-        disconnected = []
+    If *skip_disconnected* is set to ``True``, then all distributions
+    that are not dependent on any other distribution are skipped.
 
-        f.write("digraph dependencies {\n")
-        for dist, adjs in self.adjacency_list.iteritems():
-            if len(adjs) == 0 and not skip_disconnected:
-                disconnected.append(dist)
-            for (other, label) in adjs:
-                if not label is None:
-                    f.write('"%s" -> "%s" [label="%s"]\n' %
-                                                (dist.name, other.name, label))
-                else:
-                    f.write('"%s" -> "%s"\n' % (dist.name, other.name))
-        if not skip_disconnected and len(disconnected) > 0:
-            f.write('subgraph disconnected {\n')
-            f.write('label = "Disconnected"\n')
-            f.write('bgcolor = red\n')
+    :type f: has to support ``file``-like operations
+    :type skip_disconnected: ``bool``
+    """
+    disconnected = []
 
-            for dist in disconnected:
-                f.write('"%s"' % dist.name)
-                f.write('\n')
-            f.write('}\n')
+    f.write("digraph dependencies {\n")
+    for dist, adjs in graph.adjacency_list.iteritems():
+        if len(adjs) == 0 and not skip_disconnected:
+            disconnected.append(dist)
+        for (other, label) in adjs:
+            if not label is None:
+                f.write('"%s" -> "%s" [label="%s"]\n' %
+                                            (dist.name, other.name, label))
+            else:
+                f.write('"%s" -> "%s"\n' % (dist.name, other.name))
+    if not skip_disconnected and len(disconnected) > 0:
+        f.write('subgraph disconnected {\n')
+        f.write('label = "Disconnected"\n')
+        f.write('bgcolor = red\n')
+
+        for dist in disconnected:
+            f.write('"%s"' % dist.name)
+            f.write('\n')
         f.write('}\n')
+    f.write('}\n')
 
 
 def generate_graph(dists):
-    """
-    Generates a dependency graph from the given distributions.
+    """Generates a dependency graph from the given distributions.
 
     :parameter dists: a list of distributions
     :type dists: list of :class:`pkgutil.Distribution` and
@@ -115,7 +117,7 @@
         provides = dist.metadata['Provides-Dist'] + dist.metadata['Provides']
 
         for p in provides:
-            comps = p.split(" ", 1)
+            comps = p.strip().rsplit(" ", 1)
             name = comps[0]
             version = None
             if len(comps) == 2:
@@ -133,22 +135,26 @@
         requires = dist.metadata['Requires-Dist'] + dist.metadata['Requires']
         for req in requires:
             predicate = VersionPredicate(req)
-            comps = req.split(" ", 1)
+            comps = req.strip().rsplit(" ", 1)
             name = comps[0]
 
             if not name in provided:
                 graph.add_missing(dist, req)
             else:
+                matched = False
                 for (version, provider) in provided[name]:
                     if predicate.match(version):
                         graph.add_edge(dist, provider, req)
+                        matched = True
+                        break
+                if not matched:
+                    graph.add_missing(dist, req)
 
     return graph
 
 
 def dependent_dists(dists, dist):
-    """
-    Recursively generate a list of distributions from *dists* that are
+    """Recursively generate a list of distributions from *dists* that are
     dependent on *dist*.
 
     :param dists: a list of distributions
@@ -158,24 +164,26 @@
         raise ValueError('The given distribution is not a member of the list')
     graph = generate_graph(dists)
 
-    dep = [dist]
-    fringe = [dist] # list of nodes we should expand
+    dep = [dist] # dependent distributions
+    fringe = graph.reverse_list[dist] # list of nodes we should inspect
+
     while not len(fringe) == 0:
-        next = graph.adjacency_list[fringe.pop()]
-        for (dist, label) in next:
-            if not dist in dep: # avoid infinite loops
-                dep.append(dist)
-                fringe.append(dist)
+        node = fringe.pop()
+        dep.append(node)
+        for prev in graph.reverse_list[node]:
+            if not prev in dep:
+                fringe.append(prev)
 
-    dep.pop()
+    dep.pop(0) # remove dist from dep, was there to prevent infinite loops
     return dep
 
 if __name__ == '__main__':
-    dists = list(pkgutil.get_distributions(use_egg_info=True))
+    from distutils2._backport.pkgutil import get_distributions
+    dists = list(get_distributions(use_egg_info=True))
     graph = generate_graph(dists)
     for dist, reqs in graph.missing.iteritems():
         if len(reqs) > 0:
             print("Missing dependencies for %s: %s" % (dist.name,
                                                        ", ".join(reqs)))
     f = open('output.dot', 'w')
-    graph.to_dot(f, True)
+    graph_to_dot(graph, f, True)
diff --git a/src/distutils2/dist.py b/src/distutils2/dist.py
--- a/src/distutils2/dist.py
+++ b/src/distutils2/dist.py
@@ -13,8 +13,10 @@
 except ImportError:
     warnings = None
 
+from ConfigParser import RawConfigParser
+
 from distutils2.errors import (DistutilsOptionError, DistutilsArgError,
-                              DistutilsModuleError, DistutilsClassError)
+                               DistutilsModuleError, DistutilsClassError)
 from distutils2.fancy_getopt import FancyGetopt, translate_longopt
 from distutils2.util import check_environ, strtobool
 from distutils2 import log
@@ -110,7 +112,11 @@
         ('requires', None,
          "print the list of packages/modules required"),
         ('obsoletes', None,
-         "print the list of packages/modules made obsolete")
+         "print the list of packages/modules made obsolete"),
+        ('use-2to3', None,
+         "use 2to3 to make source python 3.x compatible"),
+        ('convert-2to3-doctests', None,
+         "use 2to3 to convert doctests in seperate text files"), 
         ]
     display_option_names = map(lambda x: translate_longopt(x[0]),
                                display_options)
@@ -144,8 +150,8 @@
         # information here (and enough command-line options) that it's
         # worth it.  Also delegate 'get_XXX()' methods to the 'metadata'
         # object in a sneaky and underhanded (but efficient!) way.
+        self.metadata = DistributionMetadata()
 
-        self.metadata = DistributionMetadata()
         #for basename in self.metadata._METHOD_BASENAMES:
         #    method_name = "get_" + basename
         #    setattr(self, method_name, getattr(self.metadata, method_name))
@@ -204,6 +210,8 @@
         self.scripts = None
         self.data_files = None
         self.password = ''
+        self.use_2to3 = False
+        self.convert_2to3_doctests = []
 
         # And now initialize bookkeeping stuff that can't be supplied by
         # the caller at all.  'command_obj' maps command names to
@@ -248,7 +256,7 @@
                 elif hasattr(self, key):
                     setattr(self, key, val)
                 else:
-                    msg = "Unknown distribution option: %s" % repr(key)
+                    msg = "Unknown distribution option: %r" % key
                     if warnings is not None:
                         warnings.warn(msg)
                     else:
@@ -362,14 +370,12 @@
         return files
 
     def parse_config_files(self, filenames=None):
-        from ConfigParser import ConfigParser
-
         if filenames is None:
             filenames = self.find_config_files()
 
         log.debug("Distribution.parse_config_files():")
 
-        parser = ConfigParser()
+        parser = RawConfigParser()
         for filename in filenames:
             log.debug("  reading %s" % filename)
             parser.read(filename)
@@ -383,7 +389,7 @@
                         opt = opt.replace('-', '_')
                         opt_dict[opt] = (filename, val)
 
-            # Make the ConfigParser forget everything (so we retain
+            # Make the RawConfigParser forget everything (so we retain
             # the original filenames that options come from)
             parser.__init__()
 
@@ -581,15 +587,11 @@
         instance, analogous to the .finalize_options() method of Command
         objects.
         """
-
-        # XXX conversion -- removed
-        #for attr in ('keywords', 'platforms'):
-        #    value = self.metadata.get_field(attr)
-        #    if value is None:
-        #        continue
-        #    if isinstance(value, str):
-        #        value = [elm.strip() for elm in value.split(',')]
-        #        setattr(self.metadata, attr, value)
+        if getattr(self, 'convert_2to3_doctests', None):
+            self.convert_2to3_doctests = [os.path.join(p) 
+                                for p in self.convert_2to3_doctests]
+        else:
+            self.convert_2to3_doctests = []
 
     def _show_help(self, parser, global_options=1, display_options=1,
                    commands=[]):
@@ -627,16 +629,16 @@
 
         for command in self.commands:
             if isinstance(command, type) and issubclass(command, Command):
-                klass = command
+                cls = command
             else:
-                klass = self.get_command_class(command)
-            if (hasattr(klass, 'help_options') and
-                isinstance(klass.help_options, list)):
-                parser.set_option_table(klass.user_options +
-                                        fix_help_options(klass.help_options))
+                cls = self.get_command_class(command)
+            if (hasattr(cls, 'help_options') and
+                isinstance(cls.help_options, list)):
+                parser.set_option_table(cls.user_options +
+                                        fix_help_options(cls.help_options))
             else:
-                parser.set_option_table(klass.user_options)
-            parser.print_help("Options for '%s' command:" % klass.__name__)
+                parser.set_option_table(cls.user_options)
+            parser.print_help("Options for '%s' command:" % cls.__name__)
             print('')
 
         print(gen_usage(self.script_name))
@@ -688,11 +690,11 @@
         print(header + ":")
 
         for cmd in commands:
-            klass = self.cmdclass.get(cmd)
-            if not klass:
-                klass = self.get_command_class(cmd)
+            cls = self.cmdclass.get(cmd)
+            if not cls:
+                cls = self.get_command_class(cmd)
             try:
-                description = klass.description
+                description = cls.description
             except AttributeError:
                 description = "(no description available)"
 
@@ -754,11 +756,11 @@
 
         rv = []
         for cmd in (std_commands + extra_commands):
-            klass = self.cmdclass.get(cmd)
-            if not klass:
-                klass = self.get_command_class(cmd)
+            cls = self.cmdclass.get(cmd)
+            if not cls:
+                cls = self.get_command_class(cmd)
             try:
-                description = klass.description
+                description = cls.description
             except AttributeError:
                 description = "(no description available)"
             rv.append((cmd, description))
@@ -790,13 +792,13 @@
         Raises DistutilsModuleError if the expected module could not be
         found, or if that module does not define the expected class.
         """
-        klass = self.cmdclass.get(command)
-        if klass:
-            return klass
+        cls = self.cmdclass.get(command)
+        if cls:
+            return cls
 
         for pkgname in self.get_command_packages():
             module_name = "%s.%s" % (pkgname, command)
-            klass_name = command
+            class_name = command
 
             try:
                 __import__ (module_name)
@@ -805,14 +807,14 @@
                 continue
 
             try:
-                klass = getattr(module, klass_name)
+                cls = getattr(module, class_name)
             except AttributeError:
                 raise DistutilsModuleError, \
                       "invalid command '%s' (no class '%s' in module '%s')" \
-                      % (command, klass_name, module_name)
+                      % (command, class_name, module_name)
 
-            self.cmdclass[command] = klass
-            return klass
+            self.cmdclass[command] = cls
+            return cls
 
         raise DistutilsModuleError("invalid command '%s'" % command)
 
@@ -828,8 +830,8 @@
             log.debug("Distribution.get_command_obj(): " \
                       "creating '%s' command object" % command)
 
-            klass = self.get_command_class(command)
-            cmd_obj = self.command_obj[command] = klass(self)
+            cls = self.get_command_class(command)
+            cmd_obj = self.command_obj[command] = cls(self)
             self.have_run[command] = 0
 
             # Set any options that were supplied in config files
@@ -885,7 +887,7 @@
             except ValueError, msg:
                 raise DistutilsOptionError, msg
 
-    def reinitialize_command(self, command, reinit_subcommands=0):
+    def get_reinitialized_command(self, command, reinit_subcommands=0):
         """Reinitializes a command to the state it was in when first
         returned by 'get_command_obj()': ie., initialized but not yet
         finalized.  This provides the opportunity to sneak option
@@ -920,7 +922,7 @@
 
         if reinit_subcommands:
             for sub in command.get_sub_commands():
-                self.reinitialize_command(sub, reinit_subcommands)
+                self.get_reinitialized_command(sub, reinit_subcommands)
 
         return command
 
diff --git a/src/distutils2/extension.py b/src/distutils2/extension.py
--- a/src/distutils2/extension.py
+++ b/src/distutils2/extension.py
@@ -23,7 +23,7 @@
 # import that large-ish module (indirectly, through distutils.core) in
 # order to do anything.
 
-class Extension:
+class Extension(object):
     """Just a collection of attributes that describes an extension
     module and everything needed to build it (hopefully in a portable
     way, but there are hooks that let you be as unportable as you need).
diff --git a/src/distutils2/fancy_getopt.py b/src/distutils2/fancy_getopt.py
--- a/src/distutils2/fancy_getopt.py
+++ b/src/distutils2/fancy_getopt.py
@@ -30,7 +30,7 @@
 # (for use as attributes of some object).
 longopt_xlate = string.maketrans('-', '_')
 
-class FancyGetopt:
+class FancyGetopt(object):
     """Wrapper around the standard 'getopt()' module that provides some
     handy extra functionality:
       * short and long options are tied together
@@ -473,7 +473,7 @@
     return string.translate(opt, longopt_xlate)
 
 
-class OptionDummy:
+class OptionDummy(object):
     """Dummy class just used as a place to hold command-line option
     values as instance attributes."""
 
diff --git a/src/distutils2/log.py b/src/distutils2/log.py
--- a/src/distutils2/log.py
+++ b/src/distutils2/log.py
@@ -11,14 +11,14 @@
 
 import sys
 
-class Log:
+class Log(object):
 
     def __init__(self, threshold=WARN):
         self.threshold = threshold
 
     def _log(self, level, msg, args):
         if level not in (DEBUG, INFO, WARN, ERROR, FATAL):
-            raise ValueError('%s wrong log level' % str(level))
+            raise ValueError('%s wrong log level' % level)
 
         if level >= self.threshold:
             if args:
diff --git a/src/distutils2/metadata.py b/src/distutils2/metadata.py
--- a/src/distutils2/metadata.py
+++ b/src/distutils2/metadata.py
@@ -105,7 +105,6 @@
     keys = fields.keys()
     possible_versions = ['1.0', '1.1', '1.2']
 
-
     # first let's try to see if a field is not part of one of the version
     for key in keys:
         if key not in _241_FIELDS and '1.0' in possible_versions:
@@ -128,9 +127,9 @@
         raise MetadataConflictError('You used incompatible 1.1 and 1.2 fields')
 
     # we have the choice, either 1.0, or 1.2
-    #   - 1.0 has a broken Summary field but work with all tools
+    #   - 1.0 has a broken Summary field but works with all tools
     #   - 1.1 is to avoid
-    #   - 1.2 fixes Summary but is not spreaded yet
+    #   - 1.2 fixes Summary but is not widespread yet
     if not is_1_1 and not is_1_2:
         # we couldn't find any specific marker
         if PKG_INFO_PREFERRED_VERSION in possible_versions:
@@ -185,14 +184,16 @@
 class DistributionMetadata(object):
     """Distribution meta-data class (1.0 or 1.2).
     """
-    def __init__(self, path=None, platform_dependant=False,
-                 execution_context=None):
+    def __init__(self, path=None, platform_dependent=False,
+                 execution_context=None, fileobj=None):
         self._fields = {}
         self.version = None
         self.docutils_support = _HAS_DOCUTILS
-        self.platform_dependant = platform_dependant
+        self.platform_dependent = platform_dependent
         if path is not None:
             self.read(path)
+        elif fileobj is not None:
+            self.read_file(fileobj)
         self.execution_context = execution_context
 
     def _set_best_version(self):
@@ -261,7 +262,7 @@
         return reporter.messages
 
     def _platform(self, value):
-        if not self.platform_dependant or ';' not in value:
+        if not self.platform_dependent or ';' not in value:
             return True, value
         value, marker = value.split(';')
         return _interpret(marker, self.execution_context), value
@@ -516,7 +517,7 @@
             raise NameError(value)
 
     def _nonsense_op(self):
-        msg = 'This operation is not supported : "%s"' % str(self)
+        msg = 'This operation is not supported : "%s"' % self
         raise SyntaxError(msg)
 
     def __call__(self):
@@ -633,4 +634,3 @@
     operations = _CHAIN(execution_context)
     tokenize(StringIO(marker).readline, operations.eat)
     return operations.result()
-
diff --git a/src/distutils2/mkpkg.py b/src/distutils2/mkpkg.py
--- a/src/distutils2/mkpkg.py
+++ b/src/distutils2/mkpkg.py
@@ -675,7 +675,7 @@
 troveDict = buildTroveDict(troveList)
 
 
-class SetupClass:
+class SetupClass(object):
     def __init__(self):
         self.config = None
         self.classifierDict = {}
@@ -717,14 +717,16 @@
 
     def inspectFile(self, path):
         fp = open(path, 'r')
-        for line in [ fp.readline() for x in range(10) ]:
-            m = re.match(r'^#!.*python((?P<major>\d)(\.\d+)?)?$', line)
-            if m:
-                if m.group('major') == '3':
-                    self.classifierDict['Programming Language :: Python :: 3'] = 1
-                else:
-                    self.classifierDict['Programming Language :: Python :: 2'] = 1
-        fp.close()
+        try:
+            for line in [ fp.readline() for x in range(10) ]:
+                m = re.match(r'^#!.*python((?P<major>\d)(\.\d+)?)?$', line)
+                if m:
+                    if m.group('major') == '3':
+                        self.classifierDict['Programming Language :: Python :: 3'] = 1
+                    else:
+                        self.classifierDict['Programming Language :: Python :: 2'] = 1
+        finally:
+            fp.close()
 
 
     def inspectDirectory(self):
@@ -885,38 +887,33 @@
         if os.path.exists('setup.py'): shutil.move('setup.py', 'setup.py.old')
 
         fp = open('setup.py', 'w')
-        fp.write('#!/usr/bin/env python\n\n')
-        fp.write('from distutils2.core import setup\n\n')
+        try:
+            fp.write('#!/usr/bin/env python\n\n')
+            fp.write('from distutils2.core import setup\n\n')
+            fp.write('setup(name=%s,\n' % repr(self.setupData['name']))
+            fp.write('      version=%s,\n' % repr(self.setupData['version']))
+            fp.write('      description=%s,\n'
+                    % repr(self.setupData['description']))
+            fp.write('      author=%s,\n' % repr(self.setupData['author']))
+            fp.write('      author_email=%s,\n'
+                    % repr(self.setupData['author_email']))
+            if self.setupData['url']:
+                fp.write('      url=%s,\n' % repr(self.setupData['url']))
+            if self.setupData['classifier']:
+                fp.write('      classifier=[\n')
+                for classifier in sorted(self.setupData['classifier'].keys()):
+                    fp.write('            %s,\n' % repr(classifier))
+                fp.write('         ],\n')
+            if self.setupData['packages']:
+                fp.write('      packages=%s,\n'
+                        % repr(self._dotted_packages(self.setupData['packages'])))
+                fp.write('      package_dir=%s,\n'
+                        % repr(self.setupData['packages']))
+            fp.write('      #scripts=[\'path/to/script\']\n')
 
-        fp.write('from sys import version\n')
-        fp.write('if version < \'2.2.3\':\n')
-        fp.write('    from distutils2.dist import DistributionMetadata\n')
-        fp.write('    DistributionMetadata.classifier = None\n')
-        fp.write('    DistributionMetadata.download_url = None\n')
-
-        fp.write('setup(name = %s,\n' % repr(self.setupData['name']))
-        fp.write('        version = %s,\n' % repr(self.setupData['version']))
-        fp.write('        description = %s,\n'
-                % repr(self.setupData['description']))
-        fp.write('        author = %s,\n' % repr(self.setupData['author']))
-        fp.write('        author_email = %s,\n'
-                % repr(self.setupData['author_email']))
-        if self.setupData['url']:
-            fp.write('        url = %s,\n' % repr(self.setupData['url']))
-        if self.setupData['classifier']:
-            fp.write('        classifier = [\n')
-            for classifier in sorted(self.setupData['classifier'].keys()):
-                fp.write('              %s,\n' % repr(classifier))
-            fp.write('           ],\n')
-        if self.setupData['packages']:
-            fp.write('        packages = %s,\n'
-                    % repr(self._dotted_packages(self.setupData['packages'])))
-            fp.write('        package_dir = %s,\n'
-                    % repr(self.setupData['packages']))
-        fp.write('        #scripts = [\'path/to/script\']\n')
-
-        fp.write('        )\n')
-        fp.close()
+            fp.write('      )\n')
+        finally:
+            fp.close()
         os.chmod('setup.py', 0755)
 
         print 'Wrote "setup.py".'
diff --git a/src/distutils2/pypi/__init__.py b/src/distutils2/pypi/__init__.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/pypi/__init__.py
@@ -0,0 +1,8 @@
+"""distutils2.pypi
+
+Package containing ways to interact with the PyPI APIs.
+""" 
+
+__all__ = ['simple',
+           'dist',
+]
diff --git a/src/distutils2/pypi/dist.py b/src/distutils2/pypi/dist.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/pypi/dist.py
@@ -0,0 +1,315 @@
+"""distutils2.pypi.dist
+
+Provides the PyPIDistribution class thats represents a distribution retrieved
+on PyPI.
+"""
+import re
+import urlparse
+import urllib
+import tempfile
+from operator import attrgetter
+
+try:
+    import hashlib
+except ImportError:
+    from distutils2._backport import hashlib
+
+from distutils2.version import suggest_normalized_version, NormalizedVersion
+from distutils2.pypi.errors import HashDoesNotMatch, UnsupportedHashName
+
+EXTENSIONS = ".tar.gz .tar.bz2 .tar .zip .tgz .egg".split()
+MD5_HASH = re.compile(r'^.*#md5=([a-f0-9]+)$')
+
+
+class PyPIDistribution(object):
+    """Represents a distribution retrieved from PyPI.
+
+    This is a simple container for various attributes as name, version,
+    downloaded_location, url etc.
+
+    The PyPIDistribution class is used by the pypi.*Index class to return
+    information about distributions.
+    """
+
+    @classmethod
+    def from_url(cls, url, probable_dist_name=None, is_external=True):
+        """Build a Distribution from a url archive (egg or zip or tgz).
+
+        :param url: complete url of the distribution
+        :param probable_dist_name: A probable name of the distribution.
+        :param is_external: Tell if the url commes from an index or from
+                            an external URL.
+        """
+        # if the url contains a md5 hash, get it.
+        md5_hash = None
+        match = MD5_HASH.match(url)
+        if match is not None:
+            md5_hash = match.group(1)
+            # remove the hash
+            url = url.replace("#md5=%s" % md5_hash, "")
+
+        # parse the archive name to find dist name and version
+        archive_name = urlparse.urlparse(url)[2].split('/')[-1]
+        extension_matched = False
+        # remove the extension from the name
+        for ext in EXTENSIONS:
+            if archive_name.endswith(ext):
+                archive_name = archive_name[:-len(ext)]
+                extension_matched = True
+
+        name, version = split_archive_name(archive_name)
+        if extension_matched is True:
+            return PyPIDistribution(name, version, url=url, url_hashname="md5",
+                                    url_hashval=md5_hash, url_is_external=is_external)
+
+    def __init__(self, name, version, type=None, url=None, url_hashname=None,
+                 url_hashval=None, url_is_external=True):
+        """Create a new instance of PyPIDistribution.
+
+        :param name: the name of the distribution
+        :param version: the version of the distribution
+        :param type: the type of the dist (eg. source, bin-*, etc.)
+        :param url: URL where we found this distribution
+        :param url_hashname: the name of the hash we want to use. Refer to the
+                         hashlib.new documentation for more information.
+        :param url_hashval: the hash value.
+        :param url_is_external: we need to know if the provided url comes from an
+                            index browsing, or from an external resource.
+
+        """
+        self.name = name
+        self.version = NormalizedVersion(version)
+        self.type = type
+        # set the downloaded path to None by default. The goal here
+        # is to not download distributions multiple times
+        self.downloaded_location = None
+        # We store urls in dict, because we need to have a bit more informations
+        # than the simple URL. It will be used later to find the good url to
+        # use.
+        # We have two _url* attributes: _url and _urls. _urls contains a list of
+        # dict for the different urls, and _url contains the choosen url, in
+        # order to dont make the selection process multiple times.
+        self._urls = []
+        self._url = None
+        self.add_url(url, url_hashname, url_hashval, url_is_external)
+
+    def add_url(self, url, hashname=None, hashval=None, is_external=True):
+        """Add a new url to the list of urls"""
+        if hashname is not None:
+            try:
+                hashlib.new(hashname)
+            except ValueError:
+                raise UnsupportedHashName(hashname)
+
+        self._urls.append({
+            'url': url,
+            'hashname': hashname,
+            'hashval': hashval,
+            'is_external': is_external,
+        })
+        # reset the url selection process
+        self._url = None
+
+    @property
+    def url(self):
+        """Pick up the right url for the list of urls in self.urls"""
+        # We return internal urls over externals.
+        # If there is more than one internal or external, return the first
+        # one.
+        if self._url is None:
+            if len(self._urls) > 1:
+                internals_urls = [u for u in self._urls \
+                                  if u['is_external'] == False]
+                if len(internals_urls) >= 1:
+                    self._url = internals_urls[0]
+            if self._url is None:
+                self._url = self._urls[0]
+        return self._url
+
+    @property
+    def is_source(self):
+        """return if the distribution is a source one or not"""
+        return self.type == 'source'
+
+    @property
+    def is_final(self):
+        """proxy to version.is_final"""
+        return self.version.is_final
+
+    def download(self, path=None):
+        """Download the distribution to a path, and return it.
+
+        If the path is given in path, use this, otherwise, generates a new one
+        """
+        if path is None:
+            path = tempfile.mkdtemp()
+
+        # if we do not have downloaded it yet, do it.
+        if self.downloaded_location is None:
+            url = self.url['url']
+            archive_name = urlparse.urlparse(url)[2].split('/')[-1]
+            filename, headers = urllib.urlretrieve(url,
+                                                   path + "/" + archive_name)
+            self.downloaded_location = filename
+            self._check_md5(filename)
+        return self.downloaded_location
+
+    def _check_md5(self, filename):
+        """Check that the md5 checksum of the given file matches the one in
+        url param"""
+        hashname = self.url['hashname']
+        expected_hashval = self.url['hashval']
+        if not None in (expected_hashval, hashname):
+            f = open(filename)
+            hashval = hashlib.new(hashname)
+            hashval.update(f.read())
+            if hashval.hexdigest() != expected_hashval:
+                raise HashDoesNotMatch("got %s instead of %s"
+                    % (hashval.hexdigest(), expected_hashval))
+
+    def __repr__(self):
+        return "%s %s %s %s" \
+            % (self.__class__.__name__, self.name, self.version,
+               self.type or "")
+
+    def _check_is_comparable(self, other):
+        if not isinstance(other, PyPIDistribution):
+            raise TypeError("cannot compare %s and %s"
+                % (type(self).__name__, type(other).__name__))
+        elif self.name != other.name:
+            raise TypeError("cannot compare %s and %s"
+                % (self.name, other.name))
+
+    def __eq__(self, other):
+        self._check_is_comparable(other)
+        return self.version == other.version
+
+    def __lt__(self, other):
+        self._check_is_comparable(other)
+        return self.version < other.version
+
+    def __ne__(self, other):
+        return not self.__eq__(other)
+
+    def __gt__(self, other):
+        return not (self.__lt__(other) or self.__eq__(other))
+
+    def __le__(self, other):
+        return self.__eq__(other) or self.__lt__(other)
+
+    def __ge__(self, other):
+        return self.__eq__(other) or self.__gt__(other)
+
+    # See http://docs.python.org/reference/datamodel#object.__hash__
+    __hash__ = object.__hash__
+
+
+class PyPIDistributions(list):
+    """A container of PyPIDistribution objects.
+
+    Contains methods and facilities to sort and filter distributions.
+    """
+    def __init__(self, list=[]):
+        # To disable the ability to pass lists on instanciation
+        super(PyPIDistributions, self).__init__()
+        for item in list:
+            self.append(item)
+
+    def filter(self, predicate):
+        """Filter the distributions and return a subset of distributions that
+        match the given predicate
+        """
+        return PyPIDistributions(
+            [dist for dist in self if dist.name == predicate.name and
+            predicate.match(dist.version)])
+
+    def get_last(self, predicate, prefer_source=None, prefer_final=None):
+        """Return the most up to date version, that satisfy the given
+        predicate
+        """
+        distributions = self.filter(predicate)
+        distributions.sort_distributions(prefer_source, prefer_final, reverse=True)
+        return distributions[0]
+
+    def get_same_name_and_version(self):
+        """Return lists of PyPIDistribution objects that refer to the same
+        name and version number. This do not consider the type (source, binary,
+        etc.)"""
+        processed = []
+        duplicates = []
+        for dist in self:
+            if (dist.name, dist.version) not in processed:
+                processed.append((dist.name, dist.version))
+                found_duplicates = [d for d in self if d.name == dist.name and
+                                    d.version == dist.version]
+                if len(found_duplicates) > 1:
+                    duplicates.append(found_duplicates)
+        return duplicates
+
+    def append(self, o):
+        """Append a new distribution to the list.
+
+        If a distribution with the same name and version exists, just grab the
+        URL informations and add a new new url for the existing one.
+        """
+        similar_dists = [d for d in self if d.name == o.name and
+                         d.version == o.version and d.type == o.type]
+        if len(similar_dists) > 0:
+            dist = similar_dists[0]
+            dist.add_url(**o.url)
+        else:
+            super(PyPIDistributions, self).append(o)
+
+    def sort_distributions(self, prefer_source=None, prefer_final=None,
+                           reverse=True, *args, **kwargs):
+        """order the results with the given properties"""
+
+        sort_by = []
+        if prefer_final is not None:
+            if prefer_final is True:
+                sort_by.append("is_final")
+        sort_by.append("version")
+
+        if prefer_source is not None:
+            if prefer_source is True:
+                sort_by.append("is_source")
+
+        super(PyPIDistributions, self).sort(
+            key=lambda i: [getattr(i, arg) for arg in sort_by],
+            reverse=reverse, *args, **kwargs)
+
+def split_archive_name(archive_name, probable_name=None):
+    """Split an archive name into two parts: name and version.
+
+    Return the tuple (name, version)
+    """
+    # Try to determine wich part is the name and wich is the version using the
+    # "-" separator. Take the larger part to be the version number then reduce
+    # if this not works.
+    def eager_split(str, maxsplit=2):
+        # split using the "-" separator
+        splits = str.rsplit("-", maxsplit)
+        name = splits[0]
+        version = "-".join(splits[1:])
+        if version.startswith("-"):
+            version = version[1:]
+        if suggest_normalized_version(version) is None and maxsplit >= 0:
+            # we dont get a good version number: recurse !
+            return eager_split(str, maxsplit - 1)
+        else:
+            return (name, version)
+    if probable_name is not None:
+        probable_name = probable_name.lower()
+    name = None
+    if probable_name is not None and probable_name in archive_name:
+        # we get the name from probable_name, if given.
+        name = probable_name
+        version = archive_name.lstrip(name)
+    else:
+        name, version = eager_split(archive_name)
+
+    version = suggest_normalized_version(version)
+    if version != "" and name != "":
+        return (name.lower(), version)
+    else:
+        raise CantParseArchiveName(archive_name)
diff --git a/src/distutils2/pypi/errors.py b/src/distutils2/pypi/errors.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/pypi/errors.py
@@ -0,0 +1,33 @@
+"""distutils2.pypi.errors
+
+All errors and exceptions raised by PyPiIndex classes.
+"""
+from distutils2.errors import DistutilsError
+
+
+class PyPIError(DistutilsError):
+    """The base class for errors of the pypi python package."""
+
+
+class DistributionNotFound(PyPIError):
+    """No distribution match the given requirements."""
+
+
+class CantParseArchiveName(PyPIError):
+    """An archive name can't be parsed to find distribution name and version"""
+
+
+class DownloadError(PyPIError):
+    """An error has occurs while downloading"""
+
+
+class HashDoesNotMatch(DownloadError):
+    """Compared hashes does not match"""
+
+
+class UnsupportedHashName(PyPIError):
+    """A unsupported hashname has been used"""
+
+
+class UnableToDownload(PyPIError):
+    """All mirrors have been tried, without success"""
diff --git a/src/distutils2/pypi/simple.py b/src/distutils2/pypi/simple.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/pypi/simple.py
@@ -0,0 +1,395 @@
+"""pypi.simple
+
+Contains the class "SimpleIndex", a simple spider to find and retrieve
+distributions on the Python Package Index, using it's "simple" API,
+avalaible at http://pypi.python.org/simple/
+"""
+from fnmatch import translate
+import urlparse
+import sys
+import re
+import urllib2
+import httplib
+import socket
+
+from distutils2.version import VersionPredicate
+from distutils2.pypi.dist import PyPIDistribution, PyPIDistributions, \
+    EXTENSIONS
+from distutils2.pypi.errors import PyPIError, DistributionNotFound, \
+    DownloadError, UnableToDownload
+from distutils2 import __version__ as __distutils2_version__
+
+# -- Constants -----------------------------------------------
+PYPI_DEFAULT_INDEX_URL = "http://pypi.python.org/simple/"
+PYPI_DEFAULT_MIRROR_URL = "mirrors.pypi.python.org"
+DEFAULT_HOSTS = ("*",)
+SOCKET_TIMEOUT = 15
+USER_AGENT = "Python-urllib/%s distutils2/%s" % (
+    sys.version[:3], __distutils2_version__)
+
+# -- Regexps -------------------------------------------------
+EGG_FRAGMENT = re.compile(r'^egg=([-A-Za-z0-9_.]+)$')
+HREF = re.compile("""href\\s*=\\s*['"]?([^'"> ]+)""", re.I)
+PYPI_MD5 = re.compile(
+    '<a href="([^"#]+)">([^<]+)</a>\n\s+\\(<a (?:title="MD5 hash"\n\s+)'
+    'href="[^?]+\?:action=show_md5&amp;digest=([0-9a-f]{32})">md5</a>\\)')
+URL_SCHEME = re.compile('([-+.a-z0-9]{2,}):', re.I).match
+
+# This pattern matches a character entity reference (a decimal numeric
+# references, a hexadecimal numeric reference, or a named reference).
+ENTITY_SUB = re.compile(r'&(#(\d+|x[\da-fA-F]+)|[\w.:-]+);?').sub
+REL = re.compile("""<([^>]*\srel\s*=\s*['"]?([^'">]+)[^>]*)>""", re.I)
+
+
+def socket_timeout(timeout=SOCKET_TIMEOUT):
+    """Decorator to add a socket timeout when requesting pages on PyPI.
+    """
+    def _socket_timeout(func):
+        def _socket_timeout(self, *args, **kwargs):
+            old_timeout = socket.getdefaulttimeout()
+            if hasattr(self, "_timeout"):
+                timeout = self._timeout
+            socket.setdefaulttimeout(timeout)
+            try:
+                return func(self, *args, **kwargs)
+            finally:
+                socket.setdefaulttimeout(old_timeout)
+        return _socket_timeout
+    return _socket_timeout
+
+
+class SimpleIndex(object):
+    """Provides useful tools to request the Python Package Index simple API
+    """
+
+    def __init__(self, index_url=PYPI_DEFAULT_INDEX_URL, hosts=DEFAULT_HOSTS,
+                 follow_externals=False, prefer_source=True,
+                 prefer_final=False, mirrors_url=PYPI_DEFAULT_MIRROR_URL,
+                 mirrors=None, timeout=SOCKET_TIMEOUT):
+        """Class constructor.
+
+        :param index_url: the url of the simple index to search on.
+        :param follow_externals: tell if following external links is needed or
+                                 not. Default is False.
+        :param hosts: a list of hosts allowed to be processed while using
+                      follow_externals=True. Default behavior is to follow all
+                      hosts.
+        :param follow_externals: tell if following external links is needed or
+                                 not. Default is False.
+        :param prefer_source: if there is binary and source distributions, the
+                              source prevails.
+        :param prefer_final: if the version is not mentioned, and the last
+                             version is not a "final" one (alpha, beta, etc.),
+                             pick up the last final version.
+        :param mirrors_url: the url to look on for DNS records giving mirror
+                            adresses.
+        :param mirrors: a list of mirrors to check out if problems
+                             occurs while working with the one given in "url"
+        :param timeout: time in seconds to consider a url has timeouted.
+        """
+        self.follow_externals = follow_externals
+
+        if not index_url.endswith("/"):
+            index_url += "/"
+        self._index_urls = [index_url]
+        # if no mirrors are defined, use the method described in PEP 381.
+        if mirrors is None:
+            try:
+                mirrors = socket.gethostbyname_ex(mirrors_url)[-1]
+            except socket.gaierror:
+                mirrors = []
+        self._index_urls.extend(mirrors)
+        self._current_index_url = 0
+        self._timeout = timeout
+        self._prefer_source = prefer_source
+        self._prefer_final = prefer_final
+
+        # create a regexp to match all given hosts
+        self._allowed_hosts = re.compile('|'.join(map(translate, hosts))).match
+
+        # we keep an index of pages we have processed, in order to avoid
+        # scanning them multple time (eg. if there is multiple pages pointing
+        # on one)
+        self._processed_urls = []
+        self._distributions = {}
+
+    def find(self, requirements, prefer_source=None, prefer_final=None):
+        """Browse the PyPI to find distributions that fullfil the given
+        requirements.
+
+        :param requirements: A project name and it's distribution, using
+                             version specifiers, as described in PEP345.
+        :type requirements:  You can pass either a version.VersionPredicate
+                             or a string.
+        :param prefer_source: if there is binary and source distributions, the
+                              source prevails.
+        :param prefer_final: if the version is not mentioned, and the last
+                             version is not a "final" one (alpha, beta, etc.),
+                             pick up the last final version.
+        """
+        requirements = self._get_version_predicate(requirements)
+        if prefer_source is None:
+            prefer_source = self._prefer_source
+        if prefer_final is None:
+            prefer_final = self._prefer_final
+
+        # process the index for this project
+        self._process_pypi_page(requirements.name)
+
+        # filter with requirements and return the results
+        if requirements.name in self._distributions:
+            dists = self._distributions[requirements.name].filter(requirements)
+            dists.sort_distributions(prefer_source=prefer_source,
+                                     prefer_final=prefer_final)
+        else:
+            dists = []
+
+        return dists
+
+    def get(self, requirements, *args, **kwargs):
+        """Browse the PyPI index to find distributions that fullfil the
+        given requirements, and return the most recent one.
+
+        You can specify prefer_final and prefer_source arguments here.
+        If not, the default one will be used.
+        """
+        predicate = self._get_version_predicate(requirements)
+        dists = self.find(predicate, *args, **kwargs)
+
+        if len(dists) == 0:
+            raise DistributionNotFound(requirements)
+
+        return dists.get_last(predicate)
+
+    def download(self, requirements, temp_path=None, *args, **kwargs):
+        """Download the distribution, using the requirements.
+
+        If more than one distribution match the requirements, use the last
+        version.
+        Download the distribution, and put it in the temp_path. If no temp_path
+        is given, creates and return one.
+
+        Returns the complete absolute path to the downloaded archive.
+
+        :param requirements: The same as the find attribute of `find`.
+
+        You can specify prefer_final and prefer_source arguments here.
+        If not, the default one will be used.
+        """
+        return self.get(requirements, *args, **kwargs)\
+                   .download(path=temp_path)
+
+    def _get_version_predicate(self, requirements):
+        """Return a VersionPredicate object, from a string or an already
+        existing object.
+        """
+        if isinstance(requirements, str):
+            requirements = VersionPredicate(requirements)
+        return requirements
+
+    @property
+    def index_url(self):
+        return self._index_urls[self._current_index_url]
+
+    def _switch_to_next_mirror(self):
+        """Switch to the next mirror (eg. point self.index_url to the next
+        url.
+        """
+        # Internally, iter over the _index_url iterable, if we have read all
+        # of the available indexes, raise an exception.
+        if self._current_index_url < len(self._index_urls):
+            self._current_index_url = self._current_index_url + 1
+        else:
+            raise UnableToDownload("All mirrors fails")
+
+    def _is_browsable(self, url):
+        """Tell if the given URL can be browsed or not.
+
+        It uses the follow_externals and the hosts list to tell if the given
+        url is browsable or not.
+        """
+        # if _index_url is contained in the given URL, we are browsing the
+        # index, and it's always "browsable".
+        # local files are always considered browable resources
+        if self.index_url in url or urlparse.urlparse(url)[0] == "file":
+            return True
+        elif self.follow_externals is True:
+            if self._allowed_hosts(urlparse.urlparse(url)[1]):  # 1 is netloc
+                return True
+            else:
+                return False
+        return False
+
+    def _is_distribution(self, link):
+        """Tell if the given URL matches to a distribution name or not.
+        """
+        #XXX find a better way to check that links are distributions
+        # Using a regexp ?
+        for ext in EXTENSIONS:
+            if ext in link:
+                return True
+        return False
+
+    def _register_dist(self, dist):
+        """Register a distribution as a part of fetched distributions for
+        SimpleIndex.
+
+        Return the PyPIDistributions object for the specified project name
+        """
+        # Internally, check if a entry exists with the project name, if not,
+        # create a new one, and if exists, add the dist to the pool.
+        if not dist.name in self._distributions:
+            self._distributions[dist.name] = PyPIDistributions()
+        self._distributions[dist.name].append(dist)
+        return self._distributions[dist.name]
+
+    def _process_url(self, url, project_name=None, follow_links=True):
+        """Process an url and search for distributions packages.
+
+        For each URL found, if it's a download, creates a PyPIdistribution
+        object. If it's a homepage and we can follow links, process it too.
+
+        :param url: the url to process
+        :param project_name: the project name we are searching for.
+        :param follow_links: Do not want to follow links more than from one
+                             level. This parameter tells if we want to follow
+                             the links we find (eg. run recursively this
+                             method on it)
+        """
+        f = self._open_url(url)
+        base_url = f.url
+        if url not in self._processed_urls:
+            self._processed_urls.append(url)
+            link_matcher = self._get_link_matcher(url)
+            for link, is_download in link_matcher(f.read(), base_url):
+                if link not in self._processed_urls:
+                    if self._is_distribution(link) or is_download:
+                        self._processed_urls.append(link)
+                        # it's a distribution, so create a dist object
+                        dist = PyPIDistribution.from_url(link, project_name,
+                                    is_external=not self.index_url in url)
+                        self._register_dist(dist)
+                    else:
+                        if self._is_browsable(link) and follow_links:
+                            self._process_url(link, project_name,
+                                follow_links=False)
+
+    def _get_link_matcher(self, url):
+        """Returns the right link matcher function of the given url
+        """
+        if self.index_url in url:
+            return self._simple_link_matcher
+        else:
+            return self._default_link_matcher
+
+    def _simple_link_matcher(self, content, base_url):
+        """Yield all links with a rel="download" or rel="homepage".
+
+        This matches the simple index requirements for matching links.
+        If follow_externals is set to False, dont yeld the external
+        urls.
+        """
+        for match in REL.finditer(content):
+            tag, rel = match.groups()
+            rels = map(str.strip, rel.lower().split(','))
+            if 'homepage' in rels or 'download' in rels:
+                for match in HREF.finditer(tag):
+                    url = urlparse.urljoin(base_url,
+                                           self._htmldecode(match.group(1)))
+                    if 'download' in rels or self._is_browsable(url):
+                        # yield a list of (url, is_download)
+                        yield (urlparse.urljoin(base_url, url),
+                               'download' in rels)
+
+    def _default_link_matcher(self, content, base_url):
+        """Yield all links found on the page.
+        """
+        for match in HREF.finditer(content):
+            url = urlparse.urljoin(base_url, self._htmldecode(match.group(1)))
+            if self._is_browsable(url):
+                yield (url, False)
+
+    def _process_pypi_page(self, name):
+        """Find and process a PyPI page for the given project name.
+
+        :param name: the name of the project to find the page
+        """
+        try:
+            # Browse and index the content of the given PyPI page.
+            url = self.index_url + name + "/"
+            self._process_url(url, name)
+        except DownloadError:
+            # if an error occurs, try with the next index_url
+            # (provided by the mirrors)
+            self._switch_to_next_mirror()
+            self._distributions.clear()
+            self._process_pypi_page(name)
+
+    @socket_timeout()
+    def _open_url(self, url):
+        """Open a urllib2 request, handling HTTP authentication, and local
+        files support.
+
+        """
+        try:
+            scheme, netloc, path, params, query, frag = urlparse.urlparse(url)
+
+            if scheme in ('http', 'https'):
+                auth, host = urllib2.splituser(netloc)
+            else:
+                auth = None
+
+            # add index.html automatically for filesystem paths
+            if scheme == 'file':
+                if url.endswith('/'):
+                    url += "index.html"
+
+            if auth:
+                auth = "Basic " + \
+                    urllib2.unquote(auth).encode('base64').strip()
+                new_url = urlparse.urlunparse((
+                    scheme, host, path, params, query, frag))
+                request = urllib2.Request(new_url)
+                request.add_header("Authorization", auth)
+            else:
+                request = urllib2.Request(url)
+            request.add_header('User-Agent', USER_AGENT)
+            fp = urllib2.urlopen(request)
+
+            if auth:
+                # Put authentication info back into request URL if same host,
+                # so that links found on the page will work
+                s2, h2, path2, param2, query2, frag2 = \
+                    urlparse.urlparse(fp.url)
+                if s2 == scheme and h2 == host:
+                    fp.url = urlparse.urlunparse(
+                        (s2, netloc, path2, param2, query2, frag2))
+
+            return fp
+        except (ValueError, httplib.InvalidURL), v:
+            msg = ' '.join([str(arg) for arg in v.args])
+            raise PyPIError('%s %s' % (url, msg))
+        except urllib2.HTTPError, v:
+            return v
+        except urllib2.URLError, v:
+            raise DownloadError("Download error for %s: %s" % (url, v.reason))
+        except httplib.BadStatusLine, v:
+            raise DownloadError('%s returned a bad status line. '
+                'The server might be down, %s' % (url, v.line))
+        except httplib.HTTPException, v:
+            raise DownloadError("Download error for %s: %s" % (url, v))
+
+    def _decode_entity(self, match):
+        what = match.group(1)
+        if what.startswith('#x'):
+            what = int(what[2:], 16)
+        elif what.startswith('#'):
+            what = int(what[1:])
+        else:
+            from htmlentitydefs import name2codepoint
+            what = name2codepoint.get(what, match.group(0))
+        return unichr(what)
+
+    def _htmldecode(self, text):
+        """Decode HTML entities in the given text."""
+        return ENTITY_SUB(self._decode_entity, text)
diff --git a/src/distutils2/spawn.py b/src/distutils2/spawn.py
--- a/src/distutils2/spawn.py
+++ b/src/distutils2/spawn.py
@@ -14,7 +14,7 @@
 from distutils2.errors import DistutilsPlatformError, DistutilsExecError
 from distutils2 import log
 
-def spawn(cmd, search_path=1, verbose=0, dry_run=0):
+def spawn(cmd, search_path=1, verbose=0, dry_run=0, env=None):
     """Run another program, specified as a command list 'cmd', in a new process.
 
     'cmd' is just the argument list for the new process, ie.
@@ -27,15 +27,18 @@
     must be the exact path to the executable.  If 'dry_run' is true,
     the command will not actually be run.
 
+    If 'env' is given, it's a environment dictionary used for the execution
+    environment.
+
     Raise DistutilsExecError if running the program fails in any way; just
     return on success.
     """
     if os.name == 'posix':
-        _spawn_posix(cmd, search_path, dry_run=dry_run)
+        _spawn_posix(cmd, search_path, dry_run=dry_run, env=env)
     elif os.name == 'nt':
-        _spawn_nt(cmd, search_path, dry_run=dry_run)
+        _spawn_nt(cmd, search_path, dry_run=dry_run, env=env)
     elif os.name == 'os2':
-        _spawn_os2(cmd, search_path, dry_run=dry_run)
+        _spawn_os2(cmd, search_path, dry_run=dry_run, env=env)
     else:
         raise DistutilsPlatformError, \
               "don't know how to spawn programs on platform '%s'" % os.name
@@ -56,7 +59,7 @@
             args[i] = '"%s"' % arg
     return args
 
-def _spawn_nt(cmd, search_path=1, verbose=0, dry_run=0):
+def _spawn_nt(cmd, search_path=1, verbose=0, dry_run=0, env=None):
     executable = cmd[0]
     cmd = _nt_quote_args(cmd)
     if search_path:
@@ -66,7 +69,11 @@
     if not dry_run:
         # spawn for NT requires a full path to the .exe
         try:
-            rc = os.spawnv(os.P_WAIT, executable, cmd)
+            if env is None:
+                rc = os.spawnv(os.P_WAIT, executable, cmd)
+            else:
+                rc = os.spawnve(os.P_WAIT, executable, cmd, env)
+
         except OSError, exc:
             # this seems to happen when the command isn't found
             raise DistutilsExecError, \
@@ -76,7 +83,7 @@
             raise DistutilsExecError, \
                   "command '%s' failed with exit status %d" % (cmd[0], rc)
 
-def _spawn_os2(cmd, search_path=1, verbose=0, dry_run=0):
+def _spawn_os2(cmd, search_path=1, verbose=0, dry_run=0, env=None):
     executable = cmd[0]
     if search_path:
         # either we find one or it stays the same
@@ -85,7 +92,11 @@
     if not dry_run:
         # spawnv for OS/2 EMX requires a full path to the .exe
         try:
-            rc = os.spawnv(os.P_WAIT, executable, cmd)
+            if env is None:
+                rc = os.spawnv(os.P_WAIT, executable, cmd)
+            else:
+                rc = os.spawnve(os.P_WAIT, executable, cmd, env)
+
         except OSError, exc:
             # this seems to happen when the command isn't found
             raise DistutilsExecError, \
@@ -97,16 +108,24 @@
                   "command '%s' failed with exit status %d" % (cmd[0], rc)
 
 
-def _spawn_posix(cmd, search_path=1, verbose=0, dry_run=0):
+def _spawn_posix(cmd, search_path=1, verbose=0, dry_run=0, env=None):
     log.info(' '.join(cmd))
     if dry_run:
         return
-    exec_fn = search_path and os.execvp or os.execv
+
+    if env is None:
+        exec_fn = search_path and os.execvp or os.execv
+    else:
+        exec_fn = search_path and os.execvpe or os.execve
+
     pid = os.fork()
 
     if pid == 0:  # in the child
         try:
-            exec_fn(cmd[0], cmd)
+            if env is None:
+                exec_fn(cmd[0], cmd)
+            else:
+                exec_fn(cmd[0], cmd, env)
         except OSError, e:
             sys.stderr.write("unable to execute %s: %s\n" %
                              (cmd[0], e.strerror))
diff --git a/src/distutils2/tests/__init__.py b/src/distutils2/tests/__init__.py
--- a/src/distutils2/tests/__init__.py
+++ b/src/distutils2/tests/__init__.py
@@ -45,7 +45,7 @@
     """Test failed."""
 
 
-class BasicTestRunner:
+class BasicTestRunner(object):
     def run(self, test):
         result = unittest.TestResult()
         test(result)
diff --git a/src/distutils2/tests/pypi_server.py b/src/distutils2/tests/pypi_server.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypi_server.py
@@ -0,0 +1,195 @@
+"""Mocked PyPI Server implementation, to use in tests.
+
+This module also provides a simple test case to extend if you need to use
+the PyPIServer all along your test case. Be sure to read the documentation 
+before any use.
+"""
+
+import Queue
+import threading
+import time
+import urllib2
+from BaseHTTPServer import HTTPServer
+from SimpleHTTPServer import SimpleHTTPRequestHandler
+import os.path
+import select
+
+from distutils2.tests.support import unittest
+
+PYPI_DEFAULT_STATIC_PATH = os.path.dirname(os.path.abspath(__file__)) + "/pypiserver"
+
+def use_pypi_server(*server_args, **server_kwargs):
+    """Decorator to make use of the PyPIServer for test methods, 
+    just when needed, and not for the entire duration of the testcase.
+    """
+    def wrapper(func):
+        def wrapped(*args, **kwargs):
+            server = PyPIServer(*server_args, **server_kwargs)
+            server.start()
+            try:
+                func(server=server, *args, **kwargs)
+            finally:
+                server.stop()
+        return wrapped
+    return wrapper
+
+class PyPIServerTestCase(unittest.TestCase):
+
+    def setUp(self):
+        super(PyPIServerTestCase, self).setUp()
+        self.pypi = PyPIServer()
+        self.pypi.start()
+
+    def tearDown(self):
+        super(PyPIServerTestCase, self).tearDown()
+        self.pypi.stop()
+
+class PyPIServer(threading.Thread):
+    """PyPI Mocked server.
+    Provides a mocked version of the PyPI API's, to ease tests.
+
+    Support serving static content and serving previously given text.
+    """
+
+    def __init__(self, test_static_path=None,
+                 static_filesystem_paths=["default"], static_uri_paths=["simple"]):
+        """Initialize the server.
+
+        static_uri_paths and static_base_path are parameters used to provides
+        respectively the http_paths to serve statically, and where to find the
+        matching files on the filesystem.
+        """
+        threading.Thread.__init__(self)
+        self._run = True
+        self.httpd = HTTPServer(('', 0), PyPIRequestHandler)
+        self.httpd.RequestHandlerClass.log_request = lambda *_: None
+        self.httpd.RequestHandlerClass.pypi_server = self
+        self.address = (self.httpd.server_name, self.httpd.server_port)
+        self.request_queue = Queue.Queue()
+        self._requests = []
+        self.default_response_status = 200
+        self.default_response_headers = [('Content-type', 'text/plain')]
+        self.default_response_data = "hello"
+        
+        # initialize static paths / filesystems
+        self.static_uri_paths = static_uri_paths
+        if test_static_path is not None:
+            static_filesystem_paths.append(test_static_path)
+        self.static_filesystem_paths = [PYPI_DEFAULT_STATIC_PATH + "/" + path
+            for path in static_filesystem_paths]
+
+    def run(self):
+        # loop because we can't stop it otherwise, for python < 2.6
+        while self._run:
+            r, w, e = select.select([self.httpd], [], [], 0.5)
+            if r:
+                self.httpd.handle_request()
+
+    def stop(self):
+        """self shutdown is not supported for python < 2.6"""
+        self._run = False
+
+    def get_next_response(self):
+        return (self.default_response_status,
+                self.default_response_headers,
+                self.default_response_data)
+
+    @property
+    def requests(self):
+        """Use this property to get all requests that have been made
+        to the server
+        """
+        while True:
+            try:
+                self._requests.append(self.request_queue.get_nowait())
+            except Queue.Empty:
+                break
+        return self._requests
+
+    @property
+    def full_address(self):
+        return "http://%s:%s" % self.address
+
+
+class PyPIRequestHandler(SimpleHTTPRequestHandler):
+    # we need to access the pypi server while serving the content
+    pypi_server = None
+
+    def do_POST(self):
+        return self.serve_request()
+    def do_GET(self):
+        return self.serve_request()
+    def do_DELETE(self):
+        return self.serve_request()
+    def do_PUT(self):
+        return self.serve_request()
+
+    def serve_request(self):
+        """Serve the content.
+
+        Also record the requests to be accessed later. If trying to access an
+        url matching a static uri, serve static content, otherwise serve
+        what is provided by the `get_next_response` method.
+        """
+        # record the request. Read the input only on PUT or POST requests
+        if self.command in ("PUT", "POST"):
+            if 'content-length' in self.headers.dict:
+                request_data = self.rfile.read(
+                    int(self.headers['content-length']))
+            else:
+                request_data = self.rfile.read()
+        elif self.command in ("GET", "DELETE"):
+            request_data = ''
+
+        self.pypi_server.request_queue.put((self, request_data))
+
+        # serve the content from local disc if we request an URL beginning
+        # by a pattern defined in `static_paths`
+        url_parts = self.path.split("/")
+        if (len(url_parts) > 1 and 
+                url_parts[1] in self.pypi_server.static_uri_paths):
+            data = None
+            # always take the last first.
+            fs_paths = []
+            fs_paths.extend(self.pypi_server.static_filesystem_paths)
+            fs_paths.reverse()
+            relative_path = self.path
+            for fs_path in fs_paths:
+                try:
+                    if self.path.endswith("/"):
+                        relative_path += "index.html"
+                    file = open(fs_path + relative_path)
+                    data = file.read()
+                    if relative_path.endswith('.tar.gz'):
+                        headers=[('Content-type', 'application/x-gtar')]
+                    else:
+                        headers=[('Content-type', 'text/html')]
+                    self.make_response(data, headers=headers)
+                except IOError:
+                    pass
+
+            if data is None:
+                self.make_response("Not found", 404)
+
+        # otherwise serve the content from get_next_response
+        else:
+            # send back a response
+            status, headers, data = self.pypi_server.get_next_response()
+            self.make_response(data, status, headers)
+
+    def make_response(self, data, status=200,
+                      headers=[('Content-type', 'text/html')]):
+        """Send the response to the HTTP client"""
+        if not isinstance(status, int):
+            try:
+                status = int(status)
+            except ValueError:
+                # we probably got something like YYY Codename. 
+                # Just get the first 3 digits
+                status = int(status[:3])
+
+        self.send_response(status)
+        for header, value in headers:
+            self.send_header(header, value)
+        self.end_headers()
+        self.wfile.write(data)
diff --git a/src/distutils2/tests/pypiserver/downloads_with_md5/simple/badmd5/badmd5-0.1.tar.gz b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/badmd5/badmd5-0.1.tar.gz
new file mode 100644
diff --git a/src/distutils2/tests/pypiserver/downloads_with_md5/simple/badmd5/index.html b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/badmd5/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/badmd5/index.html
@@ -0,0 +1,3 @@
+<html><body>
+<a href="badmd5-0.1.tar.gz#md5=3e3d86693d6564c807272b11b3069dfe" rel="download">badmd5-0.1.tar.gz</a><br/>
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/downloads_with_md5/simple/foobar/foobar-0.1.tar.gz b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/foobar/foobar-0.1.tar.gz
new file mode 100644
diff --git a/src/distutils2/tests/pypiserver/downloads_with_md5/simple/foobar/index.html b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/foobar/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/foobar/index.html
@@ -0,0 +1,3 @@
+<html><body>
+<a href="foobar-0.1.tar.gz#md5=d41d8cd98f00b204e9800998ecf8427e" rel="download">foobar-0.1.tar.gz</a><br/>
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/downloads_with_md5/simple/index.html b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/downloads_with_md5/simple/index.html
@@ -0,0 +1,2 @@
+<a href="foobar/">foobar/</a> 
+<a href="badmd5/">badmd5/</a> 
diff --git a/src/distutils2/tests/pypiserver/foo_bar_baz/simple/bar/index.html b/src/distutils2/tests/pypiserver/foo_bar_baz/simple/bar/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/foo_bar_baz/simple/bar/index.html
@@ -0,0 +1,6 @@
+<html><head><title>Links for bar</title></head><body><h1>Links for bar</h1>
+<a rel="download" href="../../packages/source/F/bar/bar-1.0.tar.gz">bar-1.0.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/bar/bar-1.0.1.tar.gz">bar-1.0.1.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/bar/bar-2.0.tar.gz">bar-2.0.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/bar/bar-2.0.1.tar.gz">bar-2.0.1.tar.gz</a><br/> 
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/foo_bar_baz/simple/baz/index.html b/src/distutils2/tests/pypiserver/foo_bar_baz/simple/baz/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/foo_bar_baz/simple/baz/index.html
@@ -0,0 +1,6 @@
+<html><head><title>Links for baz</title></head><body><h1>Links for baz</h1>
+<a rel="download" href="../../packages/source/F/baz/baz-1.0.tar.gz">baz-1.0.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/baz/baz-1.0.1.tar.gz">baz-1.0.1.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/baz/baz-2.0.tar.gz">baz-2.0.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/baz/baz-2.0.1.tar.gz">baz-2.0.1.tar.gz</a><br/> 
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/foo_bar_baz/simple/foo/index.html b/src/distutils2/tests/pypiserver/foo_bar_baz/simple/foo/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/foo_bar_baz/simple/foo/index.html
@@ -0,0 +1,6 @@
+<html><head><title>Links for foo</title></head><body><h1>Links for foo</h1>
+<a rel="download" href="../../packages/source/F/foo/foo-1.0.tar.gz">foo-1.0.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/foo/foo-1.0.1.tar.gz">foo-1.0.1.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/foo/foo-2.0.tar.gz">foo-2.0.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/foo/foo-2.0.1.tar.gz">foo-2.0.1.tar.gz</a><br/> 
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/foo_bar_baz/simple/index.html b/src/distutils2/tests/pypiserver/foo_bar_baz/simple/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/foo_bar_baz/simple/index.html
@@ -0,0 +1,3 @@
+<a href="foo/">foo/</a> 
+<a href="bar/">bar/</a> 
+<a href="baz/">baz/</a> 
diff --git a/src/distutils2/tests/pypiserver/test_found_links/simple/foobar/index.html b/src/distutils2/tests/pypiserver/test_found_links/simple/foobar/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/test_found_links/simple/foobar/index.html
@@ -0,0 +1,6 @@
+<html><head><title>Links for Foobar</title></head><body><h1>Links for Foobar</h1>
+<a rel="download" href="../../packages/source/F/Foobar/Foobar-1.0.tar.gz#md5=98fa833fdabcdd78d00245aead66c174">Foobar-1.0.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/Foobar/Foobar-1.0.1.tar.gz#md5=2351efb20f6b7b5d9ce80fa4cb1bd9ca">Foobar-1.0.1.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/Foobar/Foobar-2.0.tar.gz#md5=98fa833fdabcdd78d00245aead66c274">Foobar-2.0.tar.gz</a><br/> 
+<a rel="download" href="../../packages/source/F/Foobar/Foobar-2.0.1.tar.gz#md5=2352efb20f6b7b5d9ce80fa4cb2bd9ca">Foobar-2.0.1.tar.gz</a><br/> 
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/test_found_links/simple/index.html b/src/distutils2/tests/pypiserver/test_found_links/simple/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/test_found_links/simple/index.html
@@ -0,0 +1,1 @@
+<a href="foobar/">foobar/</a> 
diff --git a/src/distutils2/tests/pypiserver/test_pypi_server/external/index.html b/src/distutils2/tests/pypiserver/test_pypi_server/external/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/test_pypi_server/external/index.html
@@ -0,0 +1,1 @@
+index.html from external server
diff --git a/src/distutils2/tests/pypiserver/test_pypi_server/simple/index.html b/src/distutils2/tests/pypiserver/test_pypi_server/simple/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/test_pypi_server/simple/index.html
@@ -0,0 +1,1 @@
+Yeah
diff --git a/src/distutils2/tests/pypiserver/with_externals/external/external.html b/src/distutils2/tests/pypiserver/with_externals/external/external.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/with_externals/external/external.html
@@ -0,0 +1,3 @@
+<html><body>
+<a href="/foobar-0.1.tar.gz#md5=1__bad_md5___">bad old link</a>
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/with_externals/simple/foobar/index.html b/src/distutils2/tests/pypiserver/with_externals/simple/foobar/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/with_externals/simple/foobar/index.html
@@ -0,0 +1,4 @@
+<html><body>
+<a rel ="download" href="/foobar-0.1.tar.gz#md5=12345678901234567">foobar-0.1.tar.gz</a><br/>
+<a href="../../external/external.html" rel="homepage">external homepage</a><br/>
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/with_externals/simple/index.html b/src/distutils2/tests/pypiserver/with_externals/simple/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/with_externals/simple/index.html
@@ -0,0 +1,1 @@
+<a href="foobar/">foobar/</a> 
diff --git a/src/distutils2/tests/pypiserver/with_norel_links/external/homepage.html b/src/distutils2/tests/pypiserver/with_norel_links/external/homepage.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/with_norel_links/external/homepage.html
@@ -0,0 +1,7 @@
+<html>
+<body>
+<p>a rel=homepage HTML page</p>
+<a href="/foobar-2.0.tar.gz">foobar 2.0</a>
+</body>
+</html>
+
diff --git a/src/distutils2/tests/pypiserver/with_norel_links/external/nonrel.html b/src/distutils2/tests/pypiserver/with_norel_links/external/nonrel.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/with_norel_links/external/nonrel.html
@@ -0,0 +1,1 @@
+A page linked without rel="download" or rel="homepage" link.
diff --git a/src/distutils2/tests/pypiserver/with_norel_links/simple/foobar/index.html b/src/distutils2/tests/pypiserver/with_norel_links/simple/foobar/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/with_norel_links/simple/foobar/index.html
@@ -0,0 +1,6 @@
+<html><body>
+<a rel="download" href="/foobar-0.1.tar.gz" rel="download">foobar-0.1.tar.gz</a><br/>
+<a href="../../external/homepage.html" rel="homepage">external homepage</a><br/>
+<a href="../../external/nonrel.html">unrelated link</a><br/>
+<a href="/unrelated-0.2.tar.gz">unrelated download</a></br/>
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/with_norel_links/simple/index.html b/src/distutils2/tests/pypiserver/with_norel_links/simple/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/with_norel_links/simple/index.html
@@ -0,0 +1,1 @@
+<a href="foobar/">foobar/</a> 
diff --git a/src/distutils2/tests/pypiserver/with_real_externals/simple/foobar/index.html b/src/distutils2/tests/pypiserver/with_real_externals/simple/foobar/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/with_real_externals/simple/foobar/index.html
@@ -0,0 +1,4 @@
+<html><body>
+<a rel="download" href="/foobar-0.1.tar.gz#md5=0_correct_md5">foobar-0.1.tar.gz</a><br/>
+<a href="http://a-really-external-website/external/external.html" rel="homepage">external homepage</a><br/>
+</body></html>
diff --git a/src/distutils2/tests/pypiserver/with_real_externals/simple/index.html b/src/distutils2/tests/pypiserver/with_real_externals/simple/index.html
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/pypiserver/with_real_externals/simple/index.html
@@ -0,0 +1,1 @@
+<a href="foobar/">foobar/</a> 
diff --git a/src/distutils2/tests/support.py b/src/distutils2/tests/support.py
--- a/src/distutils2/tests/support.py
+++ b/src/distutils2/tests/support.py
@@ -14,7 +14,6 @@
 
 from distutils2 import log
 from distutils2.log import DEBUG, INFO, WARN, ERROR, FATAL
-from distutils2.core import Distribution
 
 if sys.version_info >= (2, 7):
     # improved unittest package from 2.7's standard library
@@ -42,7 +41,7 @@
 
     def _log(self, level, msg, args):
         if level not in (DEBUG, INFO, WARN, ERROR, FATAL):
-            raise ValueError('%s wrong log level' % str(level))
+            raise ValueError('%s wrong log level' % level)
         self.logs.append((level, msg, args))
 
     def get_logs(self, *levels):
@@ -65,12 +64,22 @@
     def setUp(self):
         super(TempdirManager, self).setUp()
         self.tempdirs = []
+        self.tempfiles = []
 
     def tearDown(self):
         super(TempdirManager, self).tearDown()
         while self.tempdirs:
             d = self.tempdirs.pop()
             shutil.rmtree(d, os.name in ('nt', 'cygwin'))
+        for file_ in self.tempfiles:
+            if os.path.exists(file_):
+                os.remove(file_)
+
+    def mktempfile(self):
+        """Create a temporary file that will be cleaned up."""
+        tempfile_ = tempfile.NamedTemporaryFile()
+        self.tempfiles.append(tempfile_.name)
+        return tempfile_
 
     def mkdtemp(self):
         """Create a temporary directory that will be cleaned up.
@@ -105,6 +114,7 @@
         It returns the package directory and the distribution
         instance.
         """
+        from distutils2.dist import Distribution
         tmp_dir = self.mkdtemp()
         pkg_dir = os.path.join(tmp_dir, pkg_name)
         os.mkdir(pkg_dir)
diff --git a/src/distutils2/tests/test_Mixin2to3.py b/src/distutils2/tests/test_Mixin2to3.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/test_Mixin2to3.py
@@ -0,0 +1,54 @@
+"""Tests for distutils.command.build_py."""
+import sys
+import tempfile
+
+import distutils2
+from distutils2.tests import support
+from distutils2.tests.support import unittest
+from distutils2.command.build_py import Mixin2to3
+
+
+class Mixin2to3TestCase(support.TempdirManager, unittest.TestCase):
+
+    @unittest.skipUnless(sys.version > '2.6', 'Need >= 2.6')
+    def test_convert_code_only(self):
+        # used to check if code gets converted properly.
+        code_content = "print 'test'\n"
+        code_handle = self.mktempfile()
+        code_name = code_handle.name
+
+        code_handle.write(code_content)
+        code_handle.flush()
+
+        mixin2to3 = Mixin2to3()
+        mixin2to3._run_2to3([code_name])
+        converted_code_content = "print('test')\n"
+        new_code_content = "".join(open(code_name).readlines())
+
+        self.assertEquals(new_code_content, converted_code_content)
+
+    @unittest.skipUnless(sys.version > '2.6', 'Need >= 2.6')
+    def test_doctests_only(self):
+        # used to check if doctests gets converted properly.
+        doctest_content = '"""\n>>> print test\ntest\n"""\nprint test\n\n'
+        doctest_handle = self.mktempfile()
+        doctest_name = doctest_handle.name
+
+        doctest_handle.write(doctest_content)
+        doctest_handle.flush()
+
+        mixin2to3 = Mixin2to3()
+        mixin2to3._run_2to3([doctest_name])
+
+        converted_doctest_content = ['"""', '>>> print(test)', 'test', '"""',
+                                     'print(test)', '', '', '']
+        converted_doctest_content = '\n'.join(converted_doctest_content)
+        new_doctest_content = "".join(open(doctest_name).readlines())
+
+        self.assertEquals(new_doctest_content, converted_doctest_content)
+
+def test_suite():
+    return unittest.makeSuite(Mixin2to3TestCase)
+
+if __name__ == "__main__":
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_bdist.py b/src/distutils2/tests/test_bdist.py
--- a/src/distutils2/tests/test_bdist.py
+++ b/src/distutils2/tests/test_bdist.py
@@ -1,8 +1,6 @@
 """Tests for distutils.command.bdist."""
 import sys
 import os
-import tempfile
-import shutil
 
 from distutils2.tests import run_unittest
 
@@ -25,7 +23,7 @@
         cmd = bdist(dist)
         cmd.formats = ['msi']
         cmd.ensure_finalized()
-        self.assertEquals(cmd.formats, ['msi'])
+        self.assertEqual(cmd.formats, ['msi'])
 
         # what format bdist offers ?
         # XXX an explicit list in bdist is
@@ -33,9 +31,9 @@
         # we should add a registry
         formats = ['zip', 'gztar', 'bztar', 'ztar', 'tar', 'wininst', 'msi']
         formats.sort()
-        founded = cmd.format_command.keys()
-        founded.sort()
-        self.assertEquals(founded, formats)
+        found = cmd.format_command.keys()
+        found.sort()
+        self.assertEqual(found, formats)
 
 def test_suite():
     return unittest.makeSuite(BuildTestCase)
diff --git a/src/distutils2/tests/test_bdist_dumb.py b/src/distutils2/tests/test_bdist_dumb.py
--- a/src/distutils2/tests/test_bdist_dumb.py
+++ b/src/distutils2/tests/test_bdist_dumb.py
@@ -78,7 +78,7 @@
             base = base.replace(':', '-')
 
         wanted = ['%s.zip' % base]
-        self.assertEquals(dist_created, wanted)
+        self.assertEqual(dist_created, wanted)
 
         # now let's check what we have in the zip file
         # XXX to be done
@@ -87,16 +87,16 @@
         pkg_dir, dist = self.create_dist()
         os.chdir(pkg_dir)
         cmd = bdist_dumb(dist)
-        self.assertEquals(cmd.bdist_dir, None)
+        self.assertEqual(cmd.bdist_dir, None)
         cmd.finalize_options()
 
         # bdist_dir is initialized to bdist_base/dumb if not set
         base = cmd.get_finalized_command('bdist').bdist_base
-        self.assertEquals(cmd.bdist_dir, os.path.join(base, 'dumb'))
+        self.assertEqual(cmd.bdist_dir, os.path.join(base, 'dumb'))
 
         # the format is set to a default value depending on the os.name
         default = cmd.default_format[os.name]
-        self.assertEquals(cmd.format, default)
+        self.assertEqual(cmd.format, default)
 
 def test_suite():
     return unittest.makeSuite(BuildDumbTestCase)
diff --git a/src/distutils2/tests/test_build.py b/src/distutils2/tests/test_build.py
--- a/src/distutils2/tests/test_build.py
+++ b/src/distutils2/tests/test_build.py
@@ -20,11 +20,11 @@
         cmd.finalize_options()
 
         # if not specified, plat_name gets the current platform
-        self.assertEquals(cmd.plat_name, get_platform())
+        self.assertEqual(cmd.plat_name, get_platform())
 
         # build_purelib is build + lib
         wanted = os.path.join(cmd.build_base, 'lib')
-        self.assertEquals(cmd.build_purelib, wanted)
+        self.assertEqual(cmd.build_purelib, wanted)
 
         # build_platlib is 'build/lib.platform-x.x[-pydebug]'
         # examples:
@@ -34,21 +34,21 @@
             self.assertTrue(cmd.build_platlib.endswith('-pydebug'))
             plat_spec += '-pydebug'
         wanted = os.path.join(cmd.build_base, 'lib' + plat_spec)
-        self.assertEquals(cmd.build_platlib, wanted)
+        self.assertEqual(cmd.build_platlib, wanted)
 
         # by default, build_lib = build_purelib
-        self.assertEquals(cmd.build_lib, cmd.build_purelib)
+        self.assertEqual(cmd.build_lib, cmd.build_purelib)
 
         # build_temp is build/temp.<plat>
         wanted = os.path.join(cmd.build_base, 'temp' + plat_spec)
-        self.assertEquals(cmd.build_temp, wanted)
+        self.assertEqual(cmd.build_temp, wanted)
 
         # build_scripts is build/scripts-x.x
         wanted = os.path.join(cmd.build_base, 'scripts-' +  sys.version[0:3])
-        self.assertEquals(cmd.build_scripts, wanted)
+        self.assertEqual(cmd.build_scripts, wanted)
 
         # executable is os.path.normpath(sys.executable)
-        self.assertEquals(cmd.executable, os.path.normpath(sys.executable))
+        self.assertEqual(cmd.executable, os.path.normpath(sys.executable))
 
 def test_suite():
     return unittest.makeSuite(BuildTestCase)
diff --git a/src/distutils2/tests/test_build_clib.py b/src/distutils2/tests/test_build_clib.py
--- a/src/distutils2/tests/test_build_clib.py
+++ b/src/distutils2/tests/test_build_clib.py
@@ -55,14 +55,14 @@
         self.assertRaises(DistutilsSetupError, cmd.get_source_files)
 
         cmd.libraries = [('name', {'sources': ['a', 'b']})]
-        self.assertEquals(cmd.get_source_files(), ['a', 'b'])
+        self.assertEqual(cmd.get_source_files(), ['a', 'b'])
 
         cmd.libraries = [('name', {'sources': ('a', 'b')})]
-        self.assertEquals(cmd.get_source_files(), ['a', 'b'])
+        self.assertEqual(cmd.get_source_files(), ['a', 'b'])
 
         cmd.libraries = [('name', {'sources': ('a', 'b')}),
                          ('name2', {'sources': ['c', 'd']})]
-        self.assertEquals(cmd.get_source_files(), ['a', 'b', 'c', 'd'])
+        self.assertEqual(cmd.get_source_files(), ['a', 'b', 'c', 'd'])
 
     def test_build_libraries(self):
 
@@ -91,11 +91,11 @@
 
         cmd.include_dirs = 'one-dir'
         cmd.finalize_options()
-        self.assertEquals(cmd.include_dirs, ['one-dir'])
+        self.assertEqual(cmd.include_dirs, ['one-dir'])
 
         cmd.include_dirs = None
         cmd.finalize_options()
-        self.assertEquals(cmd.include_dirs, [])
+        self.assertEqual(cmd.include_dirs, [])
 
         cmd.distribution.libraries = 'WONTWORK'
         self.assertRaises(DistutilsSetupError, cmd.finalize_options)
diff --git a/src/distutils2/tests/test_build_ext.py b/src/distutils2/tests/test_build_ext.py
--- a/src/distutils2/tests/test_build_ext.py
+++ b/src/distutils2/tests/test_build_ext.py
@@ -1,6 +1,5 @@
 import sys
 import os
-import tempfile
 import shutil
 from StringIO import StringIO
 import warnings
@@ -81,11 +80,11 @@
         for attr in ('error', 'foo', 'new', 'roj'):
             self.assertTrue(hasattr(xx, attr))
 
-        self.assertEquals(xx.foo(2, 5), 7)
-        self.assertEquals(xx.foo(13,15), 28)
-        self.assertEquals(xx.new().demo(), None)
+        self.assertEqual(xx.foo(2, 5), 7)
+        self.assertEqual(xx.foo(13,15), 28)
+        self.assertEqual(xx.new().demo(), None)
         doc = 'This is a template module just for instruction.'
-        self.assertEquals(xx.__doc__, doc)
+        self.assertEqual(xx.__doc__, doc)
         self.assertTrue(isinstance(xx.Null(), xx.Null))
         self.assertTrue(isinstance(xx.Str(), xx.Str))
 
@@ -195,7 +194,7 @@
         cmd = build_ext(dist)
         cmd.libraries = 'my_lib'
         cmd.finalize_options()
-        self.assertEquals(cmd.libraries, ['my_lib'])
+        self.assertEqual(cmd.libraries, ['my_lib'])
 
         # make sure cmd.library_dirs is turned into a list
         # if it's a string
@@ -209,7 +208,7 @@
         cmd = build_ext(dist)
         cmd.rpath = os.pathsep.join(['one', 'two'])
         cmd.finalize_options()
-        self.assertEquals(cmd.rpath, ['one', 'two'])
+        self.assertEqual(cmd.rpath, ['one', 'two'])
 
         # XXX more tests to perform for win32
 
@@ -218,25 +217,25 @@
         cmd = build_ext(dist)
         cmd.define = 'one,two'
         cmd.finalize_options()
-        self.assertEquals(cmd.define, [('one', '1'), ('two', '1')])
+        self.assertEqual(cmd.define, [('one', '1'), ('two', '1')])
 
         # make sure undef is turned into a list of
         # strings if they are ','-separated strings
         cmd = build_ext(dist)
         cmd.undef = 'one,two'
         cmd.finalize_options()
-        self.assertEquals(cmd.undef, ['one', 'two'])
+        self.assertEqual(cmd.undef, ['one', 'two'])
 
         # make sure swig_opts is turned into a list
         cmd = build_ext(dist)
         cmd.swig_opts = None
         cmd.finalize_options()
-        self.assertEquals(cmd.swig_opts, [])
+        self.assertEqual(cmd.swig_opts, [])
 
         cmd = build_ext(dist)
         cmd.swig_opts = '1 2'
         cmd.finalize_options()
-        self.assertEquals(cmd.swig_opts, ['1', '2'])
+        self.assertEqual(cmd.swig_opts, ['1', '2'])
 
     def test_check_extensions_list(self):
         dist = Distribution()
@@ -272,7 +271,7 @@
         # check_extensions_list adds in ext the values passed
         # when they are in ('include_dirs', 'library_dirs', 'libraries'
         # 'extra_objects', 'extra_compile_args', 'extra_link_args')
-        self.assertEquals(ext.libraries, 'foo')
+        self.assertEqual(ext.libraries, 'foo')
         self.assertTrue(not hasattr(ext, 'some'))
 
         # 'macros' element of build info dict must be 1- or 2-tuple
@@ -282,15 +281,15 @@
 
         exts[0][1]['macros'] = [('1', '2'), ('3',)]
         cmd.check_extensions_list(exts)
-        self.assertEquals(exts[0].undef_macros, ['3'])
-        self.assertEquals(exts[0].define_macros, [('1', '2')])
+        self.assertEqual(exts[0].undef_macros, ['3'])
+        self.assertEqual(exts[0].define_macros, [('1', '2')])
 
     def test_get_source_files(self):
         modules = [Extension('foo', ['xxx'], optional=False)]
         dist = Distribution({'name': 'xx', 'ext_modules': modules})
         cmd = build_ext(dist)
         cmd.ensure_finalized()
-        self.assertEquals(cmd.get_source_files(), ['xxx'])
+        self.assertEqual(cmd.get_source_files(), ['xxx'])
 
     def test_compiler_option(self):
         # cmd.compiler is an option and
@@ -301,7 +300,7 @@
         cmd.compiler = 'unix'
         cmd.ensure_finalized()
         cmd.run()
-        self.assertEquals(cmd.compiler, 'unix')
+        self.assertEqual(cmd.compiler, 'unix')
 
     def test_get_outputs(self):
         tmp_dir = self.mkdtemp()
@@ -312,7 +311,7 @@
                              'ext_modules': [ext]})
         cmd = build_ext(dist)
         cmd.ensure_finalized()
-        self.assertEquals(len(cmd.get_outputs()), 1)
+        self.assertEqual(len(cmd.get_outputs()), 1)
 
         if os.name == "nt":
             cmd.debug = sys.executable.endswith("_d.exe")
@@ -332,19 +331,19 @@
         finally:
             os.chdir(old_wd)
         self.assertTrue(os.path.exists(so_file))
-        self.assertEquals(os.path.splitext(so_file)[-1],
+        self.assertEqual(os.path.splitext(so_file)[-1],
                           sysconfig.get_config_var('SO'))
         so_dir = os.path.dirname(so_file)
-        self.assertEquals(so_dir, other_tmp_dir)
+        self.assertEqual(so_dir, other_tmp_dir)
 
         cmd.inplace = 0
         cmd.run()
         so_file = cmd.get_outputs()[0]
         self.assertTrue(os.path.exists(so_file))
-        self.assertEquals(os.path.splitext(so_file)[-1],
+        self.assertEqual(os.path.splitext(so_file)[-1],
                           sysconfig.get_config_var('SO'))
         so_dir = os.path.dirname(so_file)
-        self.assertEquals(so_dir, cmd.build_lib)
+        self.assertEqual(so_dir, cmd.build_lib)
 
         # inplace = 0, cmd.package = 'bar'
         build_py = cmd.get_finalized_command('build_py')
@@ -352,7 +351,7 @@
         path = cmd.get_ext_fullpath('foo')
         # checking that the last directory is the build_dir
         path = os.path.split(path)[0]
-        self.assertEquals(path, cmd.build_lib)
+        self.assertEqual(path, cmd.build_lib)
 
         # inplace = 1, cmd.package = 'bar'
         cmd.inplace = 1
@@ -366,7 +365,7 @@
         # checking that the last directory is bar
         path = os.path.split(path)[0]
         lastdir = os.path.split(path)[-1]
-        self.assertEquals(lastdir, 'bar')
+        self.assertEqual(lastdir, 'bar')
 
     def test_ext_fullpath(self):
         ext = sysconfig.get_config_vars()['SO']
@@ -382,14 +381,14 @@
         curdir = os.getcwd()
         wanted = os.path.join(curdir, 'src', 'lxml', 'etree' + ext)
         path = cmd.get_ext_fullpath('lxml.etree')
-        self.assertEquals(wanted, path)
+        self.assertEqual(wanted, path)
 
         # building lxml.etree not inplace
         cmd.inplace = 0
         cmd.build_lib = os.path.join(curdir, 'tmpdir')
         wanted = os.path.join(curdir, 'tmpdir', 'lxml', 'etree' + ext)
         path = cmd.get_ext_fullpath('lxml.etree')
-        self.assertEquals(wanted, path)
+        self.assertEqual(wanted, path)
 
         # building twisted.runner.portmap not inplace
         build_py = cmd.get_finalized_command('build_py')
@@ -398,13 +397,13 @@
         path = cmd.get_ext_fullpath('twisted.runner.portmap')
         wanted = os.path.join(curdir, 'tmpdir', 'twisted', 'runner',
                               'portmap' + ext)
-        self.assertEquals(wanted, path)
+        self.assertEqual(wanted, path)
 
         # building twisted.runner.portmap inplace
         cmd.inplace = 1
         path = cmd.get_ext_fullpath('twisted.runner.portmap')
         wanted = os.path.join(curdir, 'twisted', 'runner', 'portmap' + ext)
-        self.assertEquals(wanted, path)
+        self.assertEqual(wanted, path)
 
 def test_suite():
     src = _get_source_filename()
diff --git a/src/distutils2/tests/test_build_py.py b/src/distutils2/tests/test_build_py.py
--- a/src/distutils2/tests/test_build_py.py
+++ b/src/distutils2/tests/test_build_py.py
@@ -19,11 +19,15 @@
     def test_package_data(self):
         sources = self.mkdtemp()
         f = open(os.path.join(sources, "__init__.py"), "w")
-        f.write("# Pretend this is a package.")
-        f.close()
+        try:
+            f.write("# Pretend this is a package.")
+        finally:
+            f.close()
         f = open(os.path.join(sources, "README.txt"), "w")
-        f.write("Info about this package")
-        f.close()
+        try:
+            f.write("Info about this package")
+        finally:
+            f.close()
 
         destination = self.mkdtemp()
 
diff --git a/src/distutils2/tests/test_build_scripts.py b/src/distutils2/tests/test_build_scripts.py
--- a/src/distutils2/tests/test_build_scripts.py
+++ b/src/distutils2/tests/test_build_scripts.py
@@ -74,8 +74,10 @@
 
     def write_script(self, dir, name, text):
         f = open(os.path.join(dir, name), "w")
-        f.write(text)
-        f.close()
+        try:
+            f.write(text)
+        finally:
+            f.close()
 
     def test_version_int(self):
         source = self.mkdtemp()
diff --git a/src/distutils2/tests/test_ccompiler.py b/src/distutils2/tests/test_ccompiler.py
--- a/src/distutils2/tests/test_ccompiler.py
+++ b/src/distutils2/tests/test_ccompiler.py
@@ -31,7 +31,7 @@
         opts = gen_lib_options(compiler, libdirs, runlibdirs, libs)
         wanted = ['-Llib1', '-Llib2', '-cool', '-Rrunlib1', 'found',
                   '-lname2']
-        self.assertEquals(opts, wanted)
+        self.assertEqual(opts, wanted)
 
     def test_customize_compiler(self):
 
@@ -51,7 +51,7 @@
 
         comp = compiler()
         customize_compiler(comp)
-        self.assertEquals(comp.exes['archiver'], 'my_ar -arflags')
+        self.assertEqual(comp.exes['archiver'], 'my_ar -arflags')
 
 def test_suite():
     return unittest.makeSuite(CCompilerTestCase)
diff --git a/src/distutils2/tests/test_check.py b/src/distutils2/tests/test_check.py
--- a/src/distutils2/tests/test_check.py
+++ b/src/distutils2/tests/test_check.py
@@ -37,7 +37,7 @@
                     'name': 'xxx', 'version': 'xxx'
                     }
         cmd = self._run(metadata)
-        self.assertEquals(len(cmd._warnings), 0)
+        self.assertEqual(len(cmd._warnings), 0)
 
         # now with the strict mode, we should
         # get an error if there are missing metadata
@@ -45,7 +45,7 @@
 
         # and of course, no error when all metadata are present
         cmd = self._run(metadata, strict=1)
-        self.assertEquals(len(cmd._warnings), 0)
+        self.assertEqual(len(cmd._warnings), 0)
 
     def test_check_restructuredtext(self):
         if not _HAS_DOCUTILS: # won't test without docutils
@@ -55,7 +55,7 @@
         pkg_info, dist = self.create_dist(description=broken_rest)
         cmd = check(dist)
         cmd.check_restructuredtext()
-        self.assertEquals(len(cmd._warnings), 1)
+        self.assertEqual(len(cmd._warnings), 1)
 
         # let's see if we have an error with strict=1
         metadata = {'home_page': 'xxx', 'author': 'xxx',
@@ -69,7 +69,7 @@
         # and non-broken rest
         metadata['description'] = 'title\n=====\n\ntest'
         cmd = self._run(metadata, strict=1, restructuredtext=1)
-        self.assertEquals(len(cmd._warnings), 0)
+        self.assertEqual(len(cmd._warnings), 0)
 
     def test_check_all(self):
 
diff --git a/src/distutils2/tests/test_cmd.py b/src/distutils2/tests/test_cmd.py
--- a/src/distutils2/tests/test_cmd.py
+++ b/src/distutils2/tests/test_cmd.py
@@ -43,7 +43,7 @@
 
         # making sure execute gets called properly
         def _execute(func, args, exec_msg, level):
-            self.assertEquals(exec_msg, 'generating out from in')
+            self.assertEqual(exec_msg, 'generating out from in')
         cmd.force = True
         cmd.execute = _execute
         cmd.make_file(infiles='in', outfile='out', func='func', args=())
@@ -62,7 +62,7 @@
 
         wanted = ["command options for 'MyCmd':", '  option1 = 1',
                   '  option2 = 1']
-        self.assertEquals(msgs, wanted)
+        self.assertEqual(msgs, wanted)
 
     def test_ensure_string(self):
         cmd = self.cmd
@@ -80,7 +80,7 @@
         cmd = self.cmd
         cmd.option1 = 'ok,dok'
         cmd.ensure_string_list('option1')
-        self.assertEquals(cmd.option1, ['ok', 'dok'])
+        self.assertEqual(cmd.option1, ['ok', 'dok'])
 
         cmd.option2 = ['xxx', 'www']
         cmd.ensure_string_list('option2')
diff --git a/src/distutils2/tests/test_config.py b/src/distutils2/tests/test_config.py
--- a/src/distutils2/tests/test_config.py
+++ b/src/distutils2/tests/test_config.py
@@ -1,7 +1,6 @@
 """Tests for distutils.pypirc.pypirc."""
 import sys
 import os
-import tempfile
 import shutil
 
 from distutils2.core import PyPIRCCommand
@@ -87,20 +86,20 @@
 
         config = config.items()
         config.sort()
-        waited = [('password', 'secret'), ('realm', 'pypi'),
-                  ('repository', 'http://pypi.python.org/pypi'),
-                  ('server', 'server1'), ('username', 'me')]
-        self.assertEquals(config, waited)
+        expected = [('password', 'secret'), ('realm', 'pypi'),
+                    ('repository', 'http://pypi.python.org/pypi'),
+                    ('server', 'server1'), ('username', 'me')]
+        self.assertEqual(config, expected)
 
         # old format
         self.write_file(self.rc, PYPIRC_OLD)
         config = cmd._read_pypirc()
         config = config.items()
         config.sort()
-        waited = [('password', 'secret'), ('realm', 'pypi'),
-                  ('repository', 'http://pypi.python.org/pypi'),
-                  ('server', 'server-login'), ('username', 'tarek')]
-        self.assertEquals(config, waited)
+        expected = [('password', 'secret'), ('realm', 'pypi'),
+                    ('repository', 'http://pypi.python.org/pypi'),
+                    ('server', 'server-login'), ('username', 'tarek')]
+        self.assertEqual(config, expected)
 
     def test_server_empty_registration(self):
         cmd = self._cmd(self.dist)
@@ -109,7 +108,7 @@
         cmd._store_pypirc('tarek', 'xxx')
         self.assertTrue(os.path.exists(rc))
         content = open(rc).read()
-        self.assertEquals(content, WANTED)
+        self.assertEqual(content, WANTED)
 
 def test_suite():
     return unittest.makeSuite(PyPIRCCommandTestCase)
diff --git a/src/distutils2/tests/test_config_cmd.py b/src/distutils2/tests/test_config_cmd.py
--- a/src/distutils2/tests/test_config_cmd.py
+++ b/src/distutils2/tests/test_config_cmd.py
@@ -34,7 +34,7 @@
             f.close()
 
         dump_file(this_file, 'I am the header')
-        self.assertEquals(len(self._logs), numlines+1)
+        self.assertEqual(len(self._logs), numlines+1)
 
     def test_search_cpp(self):
         if sys.platform == 'win32':
@@ -44,10 +44,10 @@
 
         # simple pattern searches
         match = cmd.search_cpp(pattern='xxx', body='// xxx')
-        self.assertEquals(match, 0)
+        self.assertEqual(match, 0)
 
         match = cmd.search_cpp(pattern='_configtest', body='// xxx')
-        self.assertEquals(match, 1)
+        self.assertEqual(match, 1)
 
     def test_finalize_options(self):
         # finalize_options does a bit of transformation
@@ -59,9 +59,9 @@
         cmd.library_dirs = 'three%sfour' % os.pathsep
         cmd.ensure_finalized()
 
-        self.assertEquals(cmd.include_dirs, ['one', 'two'])
-        self.assertEquals(cmd.libraries, ['one'])
-        self.assertEquals(cmd.library_dirs, ['three', 'four'])
+        self.assertEqual(cmd.include_dirs, ['one', 'two'])
+        self.assertEqual(cmd.libraries, ['one'])
+        self.assertEqual(cmd.library_dirs, ['three', 'four'])
 
     def test_clean(self):
         # _clean removes files
diff --git a/src/distutils2/tests/test_converter.py b/src/distutils2/tests/test_converter.py
--- a/src/distutils2/tests/test_converter.py
+++ b/src/distutils2/tests/test_converter.py
@@ -30,7 +30,7 @@
             wanted = file_.replace('before', 'after')
             wanted = _read_file(os.path.join(convdir, wanted))
             res = ref.refactor_string(original, 'setup.py')
-            self.assertEquals(str(res), wanted)
+            self.assertEqual(str(res), wanted)
 
 def test_suite():
     return unittest.makeSuite(ConverterTestCase)
diff --git a/src/distutils2/tests/test_cygwinccompiler.py b/src/distutils2/tests/test_cygwinccompiler.py
--- a/src/distutils2/tests/test_cygwinccompiler.py
+++ b/src/distutils2/tests/test_cygwinccompiler.py
@@ -44,48 +44,48 @@
         sys.version = ('2.6.1 (r261:67515, Dec  6 2008, 16:42:21) \n[GCC '
                        '4.0.1 (Apple Computer, Inc. build 5370)]')
 
-        self.assertEquals(check_config_h()[0], CONFIG_H_OK)
+        self.assertEqual(check_config_h()[0], CONFIG_H_OK)
 
         # then it tries to see if it can find "__GNUC__" in pyconfig.h
         sys.version = 'something without the *CC word'
 
         # if the file doesn't exist it returns  CONFIG_H_UNCERTAIN
-        self.assertEquals(check_config_h()[0], CONFIG_H_UNCERTAIN)
+        self.assertEqual(check_config_h()[0], CONFIG_H_UNCERTAIN)
 
         # if it exists but does not contain __GNUC__, it returns CONFIG_H_NOTOK
         self.write_file(self.python_h, 'xxx')
-        self.assertEquals(check_config_h()[0], CONFIG_H_NOTOK)
+        self.assertEqual(check_config_h()[0], CONFIG_H_NOTOK)
 
         # and CONFIG_H_OK if __GNUC__ is found
         self.write_file(self.python_h, 'xxx __GNUC__ xxx')
-        self.assertEquals(check_config_h()[0], CONFIG_H_OK)
+        self.assertEqual(check_config_h()[0], CONFIG_H_OK)
 
     def test_get_msvcr(self):
 
         # none
         sys.version  = ('2.6.1 (r261:67515, Dec  6 2008, 16:42:21) '
                         '\n[GCC 4.0.1 (Apple Computer, Inc. build 5370)]')
-        self.assertEquals(get_msvcr(), None)
+        self.assertEqual(get_msvcr(), None)
 
         # MSVC 7.0
         sys.version = ('2.5.1 (r251:54863, Apr 18 2007, 08:51:08) '
                        '[MSC v.1300 32 bits (Intel)]')
-        self.assertEquals(get_msvcr(), ['msvcr70'])
+        self.assertEqual(get_msvcr(), ['msvcr70'])
 
         # MSVC 7.1
         sys.version = ('2.5.1 (r251:54863, Apr 18 2007, 08:51:08) '
                        '[MSC v.1310 32 bits (Intel)]')
-        self.assertEquals(get_msvcr(), ['msvcr71'])
+        self.assertEqual(get_msvcr(), ['msvcr71'])
 
         # VS2005 / MSVC 8.0
         sys.version = ('2.5.1 (r251:54863, Apr 18 2007, 08:51:08) '
                        '[MSC v.1400 32 bits (Intel)]')
-        self.assertEquals(get_msvcr(), ['msvcr80'])
+        self.assertEqual(get_msvcr(), ['msvcr80'])
 
         # VS2008 / MSVC 9.0
         sys.version = ('2.5.1 (r251:54863, Apr 18 2007, 08:51:08) '
                        '[MSC v.1500 32 bits (Intel)]')
-        self.assertEquals(get_msvcr(), ['msvcr90'])
+        self.assertEqual(get_msvcr(), ['msvcr90'])
 
         # unknown
         sys.version = ('2.5.1 (r251:54863, Apr 18 2007, 08:51:08) '
diff --git a/src/distutils2/tests/test_depgraph.py b/src/distutils2/tests/test_depgraph.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/test_depgraph.py
@@ -0,0 +1,187 @@
+"""Tests for distutils.depgraph """
+
+from distutils2.tests import support
+from distutils2.tests.support import unittest
+from distutils2 import depgraph
+from distutils2._backport import pkgutil
+
+import os
+import sys
+import re
+try:
+    import cStringIO as StringIO
+except ImportErorr:
+    import StringIO
+
+class DepGraphTestCase(support.LoggingSilencer,
+                       unittest.TestCase):
+
+    DISTROS_DIST = ('choxie', 'grammar', 'towel-stuff')
+    DISTROS_EGG  = ('bacon', 'banana', 'strawberry', 'cheese')
+    EDGE = re.compile(
+           r'"(?P<from>.*)" -> "(?P<to>.*)" \[label="(?P<label>.*)"\]'
+           )
+
+    def checkLists(self, l1, l2):
+        """ Compare two lists without taking the order into consideration """
+        self.assertListEqual(sorted(l1), sorted(l2))
+
+    def setUp(self):
+        super(DepGraphTestCase, self).setUp()
+        path = os.path.join(os.path.dirname(__file__), '..', '_backport',
+                            'tests', 'fake_dists')
+        path = os.path.abspath(path)
+        self.sys_path = sys.path[:]
+        sys.path[0:0] = [path]
+
+    def test_generate_graph(self):
+        dists = []
+        for name in self.DISTROS_DIST:
+            dist = pkgutil.get_distribution(name)
+            self.assertNotEqual(dist, None)
+            dists.append(dist)
+
+        choxie, grammar, towel = dists
+
+        graph = depgraph.generate_graph(dists)
+
+        deps = [(x.name, y) for (x,y) in graph.adjacency_list[choxie]]
+        self.checkLists([('towel-stuff', 'towel-stuff (0.1)')], deps)
+        self.assertTrue(choxie in graph.reverse_list[towel])
+        self.checkLists(graph.missing[choxie], [])
+
+        deps = [(x.name, y) for (x,y) in graph.adjacency_list[grammar]]
+        self.checkLists([], deps)
+        self.checkLists(graph.missing[grammar], ['truffles (>=1.2)'])
+
+        deps = [(x.name, y) for (x,y) in graph.adjacency_list[towel]]
+        self.checkLists([], deps)
+        self.checkLists(graph.missing[towel], ['bacon (<=0.2)'])
+
+    def test_generate_graph_egg(self):
+        dists = []
+        for name in self.DISTROS_DIST + self.DISTROS_EGG:
+            dist = pkgutil.get_distribution(name, use_egg_info=True)
+            self.assertNotEqual(dist, None)
+            dists.append(dist)
+
+        choxie, grammar, towel, bacon, banana, strawberry, cheese = dists
+
+        graph = depgraph.generate_graph(dists)
+
+        deps = [(x.name, y) for (x,y) in graph.adjacency_list[choxie]]
+        self.checkLists([('towel-stuff', 'towel-stuff (0.1)')], deps)
+        self.assertTrue(choxie in graph.reverse_list[towel])
+        self.checkLists(graph.missing[choxie], [])
+
+        deps = [(x.name, y) for (x,y) in graph.adjacency_list[grammar]]
+        self.checkLists([('bacon', 'truffles (>=1.2)')], deps)
+        self.checkLists(graph.missing[grammar], [])
+        self.assertTrue(grammar in graph.reverse_list[bacon])
+
+        deps = [(x.name, y) for (x,y) in graph.adjacency_list[towel]]
+        self.checkLists([('bacon', 'bacon (<=0.2)')], deps)
+        self.checkLists(graph.missing[towel], [])
+        self.assertTrue(towel in graph.reverse_list[bacon])
+
+        deps = [(x.name, y) for (x,y) in graph.adjacency_list[bacon]]
+        self.checkLists([], deps)
+        self.checkLists(graph.missing[bacon], [])
+
+        deps = [(x.name, y) for (x,y) in graph.adjacency_list[banana]]
+        self.checkLists([('strawberry', 'strawberry (>=0.5)')], deps)
+        self.checkLists(graph.missing[banana], [])
+        self.assertTrue(banana in graph.reverse_list[strawberry])
+
+        deps = [(x.name, y) for (x,y) in graph.adjacency_list[strawberry]]
+        self.checkLists([], deps)
+        self.checkLists(graph.missing[strawberry], [])
+
+        deps = [(x.name, y) for (x,y) in graph.adjacency_list[cheese]]
+        self.checkLists([], deps)
+        self.checkLists(graph.missing[cheese], [])
+
+    def test_dependent_dists(self):
+        dists = []
+        for name in self.DISTROS_DIST:
+            dist = pkgutil.get_distribution(name)
+            self.assertNotEqual(dist, None)
+            dists.append(dist)
+
+        choxie, grammar, towel = dists
+
+        deps = [d.name for d in depgraph.dependent_dists(dists, choxie)]
+        self.checkLists([], deps)
+
+        deps = [d.name for d in depgraph.dependent_dists(dists, grammar)]
+        self.checkLists([], deps)
+
+        deps = [d.name for d in depgraph.dependent_dists(dists, towel)]
+        self.checkLists(['choxie'], deps)
+
+
+    def test_dependent_dists_egg(self):
+        dists = []
+        for name in self.DISTROS_DIST + self.DISTROS_EGG:
+            dist = pkgutil.get_distribution(name, use_egg_info=True)
+            self.assertNotEqual(dist, None)
+            dists.append(dist)
+
+        choxie, grammar, towel, bacon, banana, strawberry, cheese = dists
+
+        deps = [d.name for d in depgraph.dependent_dists(dists, choxie)]
+        self.checkLists([], deps)
+
+        deps = [d.name for d in depgraph.dependent_dists(dists, grammar)]
+        self.checkLists([], deps)
+
+        deps = [d.name for d in depgraph.dependent_dists(dists, towel)]
+        self.checkLists(['choxie'], deps)
+
+        deps = [d.name for d in depgraph.dependent_dists(dists, bacon)]
+        self.checkLists(['choxie', 'towel-stuff', 'grammar'], deps)
+
+        deps = [d.name for d in depgraph.dependent_dists(dists, strawberry)]
+        self.checkLists(['banana'], deps)
+
+        deps = [d.name for d in depgraph.dependent_dists(dists, cheese)]
+        self.checkLists([], deps)
+
+    def test_graph_to_dot(self):
+        expected = (
+            ('towel-stuff', 'bacon', 'bacon (<=0.2)'),
+            ('grammar', 'bacon', 'truffles (>=1.2)'),
+            ('choxie', 'towel-stuff', 'towel-stuff (0.1)'),
+            ('banana', 'strawberry', 'strawberry (>=0.5)')
+        )
+
+        dists = []
+        for name in self.DISTROS_DIST + self.DISTROS_EGG:
+            dist = pkgutil.get_distribution(name, use_egg_info=True)
+            self.assertNotEqual(dist, None)
+            dists.append(dist)
+
+        graph = depgraph.generate_graph(dists)
+        buf = StringIO.StringIO()
+        depgraph.graph_to_dot(graph, buf)
+        buf.seek(0)
+        matches = []
+        lines = buf.readlines()
+        for line in lines[1:-1]: # skip the first and the last lines
+            if line[-1] == '\n':
+                line = line[:-1]
+            match = self.EDGE.match(line.strip())
+            self.assertTrue(match is not None)
+            matches.append(match.groups())
+
+        self.checkLists(matches, expected)
+
+    def tearDown(self):
+        super(DepGraphTestCase, self).tearDown()
+        sys.path = self.sys_path
+
+def test_suite():
+    return unittest.makeSuite(DepGraphTestCase)
+
+if __name__ == "__main__":
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_dist.py b/src/distutils2/tests/test_dist.py
--- a/src/distutils2/tests/test_dist.py
+++ b/src/distutils2/tests/test_dist.py
@@ -71,11 +71,11 @@
         sys.argv.append("build")
 
         __, stdout = captured_stdout(self.create_distribution, files)
-        self.assertEquals(stdout, '')
+        self.assertEqual(stdout, '')
         distutils2.dist.DEBUG = True
         try:
             __, stdout = captured_stdout(self.create_distribution, files)
-            self.assertEquals(stdout, '')
+            self.assertEqual(stdout, '')
         finally:
             distutils2.dist.DEBUG = False
 
@@ -129,13 +129,13 @@
         # Check DistributionMetadata handling of Unicode fields
         tmp_dir = self.mkdtemp()
         my_file = os.path.join(tmp_dir, 'f')
-        klass = Distribution
+        cls = Distribution
 
-        dist = klass(attrs={'author': u'Mister Caf??',
-                            'name': 'my.package',
-                            'maintainer': u'Caf?? Junior',
-                            'summary': u'Caf?? torr??fi??',
-                            'description': u'H??h??h??'})
+        dist = cls(attrs={'author': u'Mister Caf??',
+                          'name': 'my.package',
+                          'maintainer': u'Caf?? Junior',
+                          'summary': u'Caf?? torr??fi??',
+                          'description': u'H??h??h??'})
 
 
         # let's make sure the file can be written
@@ -144,11 +144,11 @@
         dist.metadata.write_file(open(my_file, 'w'))
 
         # regular ascii is of course always usable
-        dist = klass(attrs={'author': 'Mister Cafe',
-                            'name': 'my.package',
-                            'maintainer': 'Cafe Junior',
-                            'summary': 'Cafe torrefie',
-                            'description': 'Hehehe'})
+        dist = cls(attrs={'author': 'Mister Cafe',
+                          'name': 'my.package',
+                          'maintainer': 'Cafe Junior',
+                          'summary': 'Cafe torrefie',
+                          'description': 'Hehehe'})
 
         my_file2 = os.path.join(tmp_dir, 'f2')
         dist.metadata.write_file(open(my_file, 'w'))
@@ -156,7 +156,7 @@
     def test_empty_options(self):
         # an empty options dictionary should not stay in the
         # list of attributes
-        klass = Distribution
+        cls = Distribution
 
         # catching warnings
         warns = []
@@ -166,15 +166,15 @@
         old_warn = warnings.warn
         warnings.warn = _warn
         try:
-            dist = klass(attrs={'author': 'xxx',
-                                'name': 'xxx',
-                                'version': 'xxx',
-                                'url': 'xxxx',
-                                'options': {}})
+            dist = cls(attrs={'author': 'xxx',
+                              'name': 'xxx',
+                              'version': 'xxx',
+                              'url': 'xxxx',
+                              'options': {}})
         finally:
             warnings.warn = old_warn
 
-        self.assertEquals(len(warns), 0)
+        self.assertEqual(len(warns), 0)
 
     def test_finalize_options(self):
 
@@ -185,20 +185,20 @@
         dist.finalize_options()
 
         # finalize_option splits platforms and keywords
-        self.assertEquals(dist.metadata['platform'], ['one', 'two'])
-        self.assertEquals(dist.metadata['keywords'], ['one', 'two'])
+        self.assertEqual(dist.metadata['platform'], ['one', 'two'])
+        self.assertEqual(dist.metadata['keywords'], ['one', 'two'])
 
     def test_get_command_packages(self):
         dist = Distribution()
-        self.assertEquals(dist.command_packages, None)
+        self.assertEqual(dist.command_packages, None)
         cmds = dist.get_command_packages()
-        self.assertEquals(cmds, ['distutils2.command'])
-        self.assertEquals(dist.command_packages,
+        self.assertEqual(cmds, ['distutils2.command'])
+        self.assertEqual(dist.command_packages,
                           ['distutils2.command'])
 
         dist.command_packages = 'one,two'
         cmds = dist.get_command_packages()
-        self.assertEquals(cmds, ['distutils2.command', 'one', 'two'])
+        self.assertEqual(cmds, ['distutils2.command', 'one', 'two'])
 
 
     def test_announce(self):
@@ -238,7 +238,7 @@
             os.path.expanduser = old_expander
 
         # make sure --no-user-cfg disables the user cfg file
-        self.assertEquals(len(all_files)-1, len(files))
+        self.assertEqual(len(all_files)-1, len(files))
 
 
 class MetadataTestCase(support.TempdirManager, support.EnvironGuard,
@@ -340,8 +340,10 @@
         temp_dir = self.mkdtemp()
         user_filename = os.path.join(temp_dir, user_filename)
         f = open(user_filename, 'w')
-        f.write('.')
-        f.close()
+        try:
+            f.write('.')
+        finally:
+            f.close()
 
         try:
             dist = Distribution()
@@ -365,8 +367,8 @@
     def test_fix_help_options(self):
         help_tuples = [('a', 'b', 'c', 'd'), (1, 2, 3, 4)]
         fancy_options = fix_help_options(help_tuples)
-        self.assertEquals(fancy_options[0], ('a', 'b', 'c'))
-        self.assertEquals(fancy_options[1], (1, 2, 3))
+        self.assertEqual(fancy_options[0], ('a', 'b', 'c'))
+        self.assertEqual(fancy_options[1], (1, 2, 3))
 
     def test_show_help(self):
         # smoke test, just makes sure some help is displayed
@@ -412,14 +414,14 @@
         PKG_INFO.seek(0)
 
         metadata.read_file(PKG_INFO)
-        self.assertEquals(metadata['name'], "package")
-        self.assertEquals(metadata['version'], "1.0")
-        self.assertEquals(metadata['summary'], "xxx")
-        self.assertEquals(metadata['download_url'], 'http://example.com')
-        self.assertEquals(metadata['keywords'], ['one', 'two'])
-        self.assertEquals(metadata['platform'], [])
-        self.assertEquals(metadata['obsoletes'], [])
-        self.assertEquals(metadata['requires-dist'], ['foo'])
+        self.assertEqual(metadata['name'], "package")
+        self.assertEqual(metadata['version'], "1.0")
+        self.assertEqual(metadata['summary'], "xxx")
+        self.assertEqual(metadata['download_url'], 'http://example.com')
+        self.assertEqual(metadata['keywords'], ['one', 'two'])
+        self.assertEqual(metadata['platform'], [])
+        self.assertEqual(metadata['obsoletes'], [])
+        self.assertEqual(metadata['requires-dist'], ['foo'])
 
 def test_suite():
     suite = unittest.TestSuite()
diff --git a/src/distutils2/tests/test_install.py b/src/distutils2/tests/test_install.py
--- a/src/distutils2/tests/test_install.py
+++ b/src/distutils2/tests/test_install.py
@@ -139,23 +139,23 @@
 
         # two elements
         cmd.handle_extra_path()
-        self.assertEquals(cmd.extra_path, ['path', 'dirs'])
-        self.assertEquals(cmd.extra_dirs, 'dirs')
-        self.assertEquals(cmd.path_file, 'path')
+        self.assertEqual(cmd.extra_path, ['path', 'dirs'])
+        self.assertEqual(cmd.extra_dirs, 'dirs')
+        self.assertEqual(cmd.path_file, 'path')
 
         # one element
         cmd.extra_path = ['path']
         cmd.handle_extra_path()
-        self.assertEquals(cmd.extra_path, ['path'])
-        self.assertEquals(cmd.extra_dirs, 'path')
-        self.assertEquals(cmd.path_file, 'path')
+        self.assertEqual(cmd.extra_path, ['path'])
+        self.assertEqual(cmd.extra_dirs, 'path')
+        self.assertEqual(cmd.path_file, 'path')
 
         # none
         dist.extra_path = cmd.extra_path = None
         cmd.handle_extra_path()
-        self.assertEquals(cmd.extra_path, None)
-        self.assertEquals(cmd.extra_dirs, '')
-        self.assertEquals(cmd.path_file, None)
+        self.assertEqual(cmd.extra_path, None)
+        self.assertEqual(cmd.extra_dirs, '')
+        self.assertEqual(cmd.path_file, None)
 
         # three elements (no way !)
         cmd.extra_path = 'path,dirs,again'
@@ -199,7 +199,7 @@
         # line (the egg info file)
         f = open(cmd.record)
         try:
-            self.assertEquals(len(f.readlines()), 1)
+            self.assertEqual(len(f.readlines()), 1)
         finally:
             f.close()
 
diff --git a/src/distutils2/tests/test_install_data.py b/src/distutils2/tests/test_install_data.py
--- a/src/distutils2/tests/test_install_data.py
+++ b/src/distutils2/tests/test_install_data.py
@@ -27,14 +27,14 @@
         self.write_file(two, 'xxx')
 
         cmd.data_files = [one, (inst2, [two])]
-        self.assertEquals(cmd.get_inputs(), [one, (inst2, [two])])
+        self.assertEqual(cmd.get_inputs(), [one, (inst2, [two])])
 
         # let's run the command
         cmd.ensure_finalized()
         cmd.run()
 
         # let's check the result
-        self.assertEquals(len(cmd.get_outputs()), 2)
+        self.assertEqual(len(cmd.get_outputs()), 2)
         rtwo = os.path.split(two)[-1]
         self.assertTrue(os.path.exists(os.path.join(inst2, rtwo)))
         rone = os.path.split(one)[-1]
@@ -47,7 +47,7 @@
         cmd.run()
 
         # let's check the result
-        self.assertEquals(len(cmd.get_outputs()), 2)
+        self.assertEqual(len(cmd.get_outputs()), 2)
         self.assertTrue(os.path.exists(os.path.join(inst2, rtwo)))
         self.assertTrue(os.path.exists(os.path.join(inst, rone)))
         cmd.outfiles = []
@@ -65,7 +65,7 @@
         cmd.run()
 
         # let's check the result
-        self.assertEquals(len(cmd.get_outputs()), 4)
+        self.assertEqual(len(cmd.get_outputs()), 4)
         self.assertTrue(os.path.exists(os.path.join(inst2, rtwo)))
         self.assertTrue(os.path.exists(os.path.join(inst, rone)))
 
diff --git a/src/distutils2/tests/test_install_headers.py b/src/distutils2/tests/test_install_headers.py
--- a/src/distutils2/tests/test_install_headers.py
+++ b/src/distutils2/tests/test_install_headers.py
@@ -23,7 +23,7 @@
 
         pkg_dir, dist = self.create_dist(headers=headers)
         cmd = install_headers(dist)
-        self.assertEquals(cmd.get_inputs(), headers)
+        self.assertEqual(cmd.get_inputs(), headers)
 
         # let's run the command
         cmd.install_dir = os.path.join(pkg_dir, 'inst')
@@ -31,7 +31,7 @@
         cmd.run()
 
         # let's check the results
-        self.assertEquals(len(cmd.get_outputs()), 2)
+        self.assertEqual(len(cmd.get_outputs()), 2)
 
 def test_suite():
     return unittest.makeSuite(InstallHeadersTestCase)
diff --git a/src/distutils2/tests/test_install_lib.py b/src/distutils2/tests/test_install_lib.py
--- a/src/distutils2/tests/test_install_lib.py
+++ b/src/distutils2/tests/test_install_lib.py
@@ -25,8 +25,8 @@
         cmd = install_lib(dist)
 
         cmd.finalize_options()
-        self.assertEquals(cmd.compile, 1)
-        self.assertEquals(cmd.optimize, 0)
+        self.assertEqual(cmd.compile, 1)
+        self.assertEqual(cmd.optimize, 0)
 
         # optimize must be 0, 1, or 2
         cmd.optimize = 'foo'
@@ -36,7 +36,7 @@
 
         cmd.optimize = '2'
         cmd.finalize_options()
-        self.assertEquals(cmd.optimize, 2)
+        self.assertEqual(cmd.optimize, 2)
 
     @unittest.skipIf(no_bytecode, 'byte-compile not supported')
     def test_byte_compile(self):
@@ -82,7 +82,7 @@
         cmd.distribution.script_name = 'setup.py'
 
         # get_input should return 2 elements
-        self.assertEquals(len(cmd.get_inputs()), 2)
+        self.assertEqual(len(cmd.get_inputs()), 2)
 
     @unittest.skipUnless(bytecode_support, 'sys.dont_write_bytecode not supported')
     def test_dont_write_bytecode(self):
diff --git a/src/distutils2/tests/test_install_scripts.py b/src/distutils2/tests/test_install_scripts.py
--- a/src/distutils2/tests/test_install_scripts.py
+++ b/src/distutils2/tests/test_install_scripts.py
@@ -42,8 +42,10 @@
         def write_script(name, text):
             expected.append(name)
             f = open(os.path.join(source, name), "w")
-            f.write(text)
-            f.close()
+            try:
+                f.write(text)
+            finally:
+                f.close()
 
         write_script("script1.py", ("#! /usr/bin/env python2.3\n"
                                     "# bogus script w/ Python sh-bang\n"
diff --git a/src/distutils2/tests/test_manifest.py b/src/distutils2/tests/test_manifest.py
--- a/src/distutils2/tests/test_manifest.py
+++ b/src/distutils2/tests/test_manifest.py
@@ -44,7 +44,7 @@
 
         # the manifest should have been read
         # and 3 warnings issued (we ddidn't provided the files)
-        self.assertEquals(len(warns), 3)
+        self.assertEqual(len(warns), 3)
         for warn in warns:
             self.assertIn('warning: no files found matching', warn)
 
diff --git a/src/distutils2/tests/test_metadata.py b/src/distutils2/tests/test_metadata.py
--- a/src/distutils2/tests/test_metadata.py
+++ b/src/distutils2/tests/test_metadata.py
@@ -5,9 +5,9 @@
 
 from distutils2.metadata import (DistributionMetadata, _interpret,
                                  PKG_INFO_PREFERRED_VERSION)
-from distutils2.tests.support import unittest
+from distutils2.tests.support import unittest, LoggingSilencer
 
-class DistributionMetadataTestCase(unittest.TestCase):
+class DistributionMetadataTestCase(LoggingSilencer, unittest.TestCase):
 
 
     def test_interpret(self):
@@ -15,15 +15,15 @@
         version = sys.version.split()[0]
         os_name = os.name
 
-        assert _interpret("sys.platform == '%s'" % platform)
-        assert _interpret("sys.platform == '%s' or python_version == '2.4'" \
-                % platform)
-        assert _interpret("sys.platform == '%s' and "
-                          "python_full_version == '%s'"\
-                % (platform, version))
-        assert _interpret("'%s' == sys.platform" % platform)
+        self.assertTrue(_interpret("sys.platform == '%s'" % platform))
+        self.assertTrue(_interpret(
+            "sys.platform == '%s' or python_version == '2.4'" % platform))
+        self.assertTrue(_interpret(
+            "sys.platform == '%s' and python_full_version == '%s'" %
+            (platform, version)))
+        self.assertTrue(_interpret("'%s' == sys.platform" % platform))
 
-        assert _interpret('os.name == "%s"' % os_name)
+        self.assertTrue(_interpret('os.name == "%s"' % os_name))
 
         # stuff that need to raise a syntax error
         ops = ('os.name == os.name', 'os.name == 2', "'2' == '2'",
@@ -35,24 +35,25 @@
         OP = 'os.name == "%s"' % os_name
         AND = ' and '
         OR = ' or '
-        assert _interpret(OP+AND+OP)
-        assert _interpret(OP+AND+OP+AND+OP)
-        assert _interpret(OP+OR+OP)
-        assert _interpret(OP+OR+OP+OR+OP)
+        self.assertTrue(_interpret(OP + AND + OP))
+        self.assertTrue(_interpret(OP + AND + OP + AND + OP))
+        self.assertTrue(_interpret(OP + OR + OP))
+        self.assertTrue(_interpret(OP + OR + OP + OR + OP))
 
         # other operators
-        assert _interpret("os.name != 'buuuu'")
-        assert _interpret("python_version > '1.0'")
-        assert _interpret("python_version < '5.0'")
-        assert _interpret("python_version <= '5.0'")
-        assert _interpret("python_version >= '1.0'")
-        assert _interpret("'%s' in os.name" % os_name)
-        assert _interpret("'buuuu' not in os.name")
-        assert _interpret("'buuuu' not in os.name and '%s' in os.name" \
-                            % os_name)
+        self.assertTrue(_interpret("os.name != 'buuuu'"))
+        self.assertTrue(_interpret("python_version > '1.0'"))
+        self.assertTrue(_interpret("python_version < '5.0'"))
+        self.assertTrue(_interpret("python_version <= '5.0'"))
+        self.assertTrue(_interpret("python_version >= '1.0'"))
+        self.assertTrue(_interpret("'%s' in os.name" % os_name))
+        self.assertTrue(_interpret("'buuuu' not in os.name"))
+        self.assertTrue(_interpret(
+            "'buuuu' not in os.name and '%s' in os.name" % os_name))
 
         # execution context
-        assert _interpret('python_version == "0.1"', {'python_version': '0.1'})
+        self.assertTrue(_interpret('python_version == "0.1"',
+                                   {'python_version': '0.1'}))
 
     def test_metadata_read_write(self):
 
@@ -63,25 +64,28 @@
         res.seek(0)
         res = res.read()
         f = open(PKG_INFO)
-        wanted = f.read()
+        try:
+            # XXX this is not used
+            wanted = f.read()
+        finally:
+            f.close()
         self.assertTrue('Keywords: keyring,password,crypt' in res)
-        f.close()
 
     def test_metadata_markers(self):
         # see if we can be platform-aware
         PKG_INFO = os.path.join(os.path.dirname(__file__), 'PKG-INFO')
         content = open(PKG_INFO).read()
         content = content % sys.platform
-        metadata = DistributionMetadata(platform_dependant=True)
+        metadata = DistributionMetadata(platform_dependent=True)
         metadata.read_file(StringIO(content))
-        self.assertEquals(metadata['Requires-Dist'], ['bar'])
+        self.assertEqual(metadata['Requires-Dist'], ['bar'])
 
         # test with context
         context = {'sys.platform': 'okook'}
-        metadata = DistributionMetadata(platform_dependant=True,
+        metadata = DistributionMetadata(platform_dependent=True,
                                         execution_context=context)
         metadata.read_file(StringIO(content))
-        self.assertEquals(metadata['Requires-Dist'], ['foo'])
+        self.assertEqual(metadata['Requires-Dist'], ['foo'])
 
     def test_description(self):
         PKG_INFO = os.path.join(os.path.dirname(__file__), 'PKG-INFO')
@@ -93,14 +97,14 @@
         # see if we can read the description now
         DESC = os.path.join(os.path.dirname(__file__), 'LONG_DESC.txt')
         wanted = open(DESC).read()
-        self.assertEquals(wanted, metadata['Description'])
+        self.assertEqual(wanted, metadata['Description'])
 
         # save the file somewhere and make sure we can read it back
         out = StringIO()
         metadata.write_file(out)
         out.seek(0)
         metadata.read_file(out)
-        self.assertEquals(wanted, metadata['Description'])
+        self.assertEqual(wanted, metadata['Description'])
 
     def test_mapper_apis(self):
         PKG_INFO = os.path.join(os.path.dirname(__file__), 'PKG-INFO')
@@ -115,25 +119,25 @@
     def test_versions(self):
         metadata = DistributionMetadata()
         metadata['Obsoletes'] = 'ok'
-        self.assertEquals(metadata['Metadata-Version'], '1.1')
+        self.assertEqual(metadata['Metadata-Version'], '1.1')
 
         del metadata['Obsoletes']
         metadata['Obsoletes-Dist'] = 'ok'
-        self.assertEquals(metadata['Metadata-Version'], '1.2')
+        self.assertEqual(metadata['Metadata-Version'], '1.2')
 
         del metadata['Obsoletes-Dist']
         metadata['Version'] = '1'
-        self.assertEquals(metadata['Metadata-Version'], '1.0')
+        self.assertEqual(metadata['Metadata-Version'], '1.0')
 
         PKG_INFO = os.path.join(os.path.dirname(__file__),
                                 'SETUPTOOLS-PKG-INFO')
         metadata.read_file(StringIO(open(PKG_INFO).read()))
-        self.assertEquals(metadata['Metadata-Version'], '1.0')
+        self.assertEqual(metadata['Metadata-Version'], '1.0')
 
         PKG_INFO = os.path.join(os.path.dirname(__file__),
                                 'SETUPTOOLS-PKG-INFO2')
         metadata.read_file(StringIO(open(PKG_INFO).read()))
-        self.assertEquals(metadata['Metadata-Version'], '1.1')
+        self.assertEqual(metadata['Metadata-Version'], '1.1')
 
     def test_warnings(self):
         metadata = DistributionMetadata()
@@ -161,7 +165,7 @@
 
         # we should have a certain amount of warnings
         num_wanted = len(values)
-        self.assertEquals(num_wanted, res)
+        self.assertEqual(num_wanted, res)
 
     def test_multiple_predicates(self):
         metadata = DistributionMetadata()
@@ -184,29 +188,29 @@
             res = m.warns
             del m.warns
 
-        self.assertEquals(res, 0)
+        self.assertEqual(res, 0)
 
     def test_project_url(self):
         metadata = DistributionMetadata()
         metadata['Project-URL'] = [('one', 'http://ok')]
-        self.assertEquals(metadata['Project-URL'],
+        self.assertEqual(metadata['Project-URL'],
                           [('one', 'http://ok')])
-        self.assertEquals(metadata.version, '1.2')
+        self.assertEqual(metadata.version, '1.2')
 
     def test_check(self):
         metadata = DistributionMetadata()
         metadata['Version'] = 'rr'
         metadata['Requires-dist'] = ['Foo (a)']
         missing, warnings = metadata.check()
-        self.assertEquals(missing, ['Name', 'Home-page'])
-        self.assertEquals(len(warnings), 2)
+        self.assertEqual(missing, ['Name', 'Home-page'])
+        self.assertEqual(len(warnings), 2)
 
     def test_best_choice(self):
         metadata = DistributionMetadata()
         metadata['Version'] = '1.0'
-        self.assertEquals(metadata.version, PKG_INFO_PREFERRED_VERSION)
+        self.assertEqual(metadata.version, PKG_INFO_PREFERRED_VERSION)
         metadata['Classifier'] = ['ok']
-        self.assertEquals(metadata.version, '1.2')
+        self.assertEqual(metadata.version, '1.2')
 
     def test_project_urls(self):
         # project-url is a bit specific, make sure we write it
@@ -214,7 +218,7 @@
         metadata = DistributionMetadata()
         metadata['Version'] = '1.0'
         metadata['Project-Url'] = [('one', 'http://ok')]
-        self.assertEquals(metadata['Project-Url'], [('one', 'http://ok')])
+        self.assertEqual(metadata['Project-Url'], [('one', 'http://ok')])
         file_ = StringIO()
         metadata.write_file(file_)
         file_.seek(0)
@@ -224,7 +228,7 @@
         file_.seek(0)
         metadata = DistributionMetadata()
         metadata.read_file(file_)
-        self.assertEquals(metadata['Project-Url'], [('one', 'http://ok')])
+        self.assertEqual(metadata['Project-Url'], [('one', 'http://ok')])
 
 
 def test_suite():
diff --git a/src/distutils2/tests/test_msvc9compiler.py b/src/distutils2/tests/test_msvc9compiler.py
--- a/src/distutils2/tests/test_msvc9compiler.py
+++ b/src/distutils2/tests/test_msvc9compiler.py
@@ -105,7 +105,7 @@
         import _winreg
         HKCU = _winreg.HKEY_CURRENT_USER
         keys = Reg.read_keys(HKCU, 'xxxx')
-        self.assertEquals(keys, None)
+        self.assertEqual(keys, None)
 
         keys = Reg.read_keys(HKCU, r'Control Panel')
         self.assertTrue('Desktop' in keys)
@@ -116,20 +116,24 @@
         tempdir = self.mkdtemp()
         manifest = os.path.join(tempdir, 'manifest')
         f = open(manifest, 'w')
-        f.write(_MANIFEST)
-        f.close()
+        try:
+            f.write(_MANIFEST)
+        finally:
+            f.close()
 
         compiler = MSVCCompiler()
         compiler._remove_visual_c_ref(manifest)
 
         # see what we got
         f = open(manifest)
-        # removing trailing spaces
-        content = '\n'.join([line.rstrip() for line in f.readlines()])
-        f.close()
+        try:
+            # removing trailing spaces
+            content = '\n'.join([line.rstrip() for line in f.readlines()])
+        finally:
+            f.close()
 
         # makes sure the manifest was properly cleaned
-        self.assertEquals(content, _CLEANED_MANIFEST)
+        self.assertEqual(content, _CLEANED_MANIFEST)
 
 
 def test_suite():
diff --git a/src/distutils2/tests/test_pypi_dist.py b/src/distutils2/tests/test_pypi_dist.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/test_pypi_dist.py
@@ -0,0 +1,251 @@
+"""Tests for the distutils2.pypi.dist module."""
+
+import os
+import shutil
+import tempfile
+
+from distutils2.tests.pypi_server import use_pypi_server
+from distutils2.tests import support
+from distutils2.tests.support import unittest
+from distutils2.version import VersionPredicate
+from distutils2.pypi.errors import HashDoesNotMatch, UnsupportedHashName
+from distutils2.pypi.dist import (PyPIDistribution as Dist,
+                                  PyPIDistributions as Dists,
+                                  split_archive_name)
+
+
+class TestPyPIDistribution(support.TempdirManager,
+                           unittest.TestCase):
+    """tests the pypi.dist.PyPIDistribution class"""
+
+    def test_instanciation(self):
+        """Test the Distribution class provides us the good attributes when
+        given on construction"""
+        dist = Dist("FooBar", "1.1")
+        self.assertEqual("FooBar", dist.name)
+        self.assertEqual("1.1", "%s" % dist.version)
+
+    def test_from_url(self):
+        """Test that the Distribution object can be built from a single URL"""
+        url_list = {
+            'FooBar-1.1.0.tar.gz': {
+                'name': 'foobar',  # lowercase the name
+                'version': '1.1',
+            },
+            'Foo-Bar-1.1.0.zip': {
+                'name': 'foo-bar',  # keep the dash
+                'version': '1.1',
+            },
+            'foobar-1.1b2.tar.gz#md5=123123123123123': {
+                'name': 'foobar',
+                'version': '1.1b2',
+                'url': {
+                    'url': 'http://test.tld/foobar-1.1b2.tar.gz',  # no hash
+                    'hashval': '123123123123123',
+                    'hashname': 'md5',
+                }
+            },
+            'foobar-1.1-rc2.tar.gz': {  # use suggested name
+                'name': 'foobar',
+                'version': '1.1c2',
+                'url': {
+                    'url': 'http://test.tld/foobar-1.1-rc2.tar.gz',
+                }
+            }
+        }
+
+        for url, attributes in url_list.items():
+            dist = Dist.from_url("http://test.tld/" + url)
+            for attribute, value in attributes.items():
+                if isinstance(value, dict):
+                    mylist = getattr(dist, attribute)
+                    for val in value.keys():
+                        self.assertEqual(value[val], mylist[val])
+                else:
+                    if attribute == "version":
+                        self.assertEqual("%s" % getattr(dist, "version"), value)
+                    else:
+                        self.assertEqual(getattr(dist, attribute), value)
+
+    def test_get_url(self):
+        """Test that the url property works well"""
+
+        d = Dist("foobar", "1.1", url="test_url")
+        self.assertDictEqual(d.url, {
+            "url": "test_url",
+            "is_external": True,
+            "hashname": None,
+            "hashval": None,
+        })
+
+        # add a new url
+        d.add_url(url="internal_url", is_external=False)
+        self.assertEqual(d._url, None)
+        self.assertDictEqual(d.url, {
+            "url": "internal_url",
+            "is_external": False,
+            "hashname": None,
+            "hashval": None,
+        })
+        self.assertEqual(2, len(d._urls))
+
+    def test_comparaison(self):
+        """Test that we can compare PyPIDistributions"""
+        foo1 = Dist("foo", "1.0")
+        foo2 = Dist("foo", "2.0")
+        bar = Dist("bar", "2.0")
+        # assert we use the version to compare
+        self.assertTrue(foo1 < foo2)
+        self.assertFalse(foo1 > foo2)
+        self.assertFalse(foo1 == foo2)
+
+        # assert we can't compare dists with different names
+        self.assertRaises(TypeError, foo1.__eq__, bar)
+
+    def test_split_archive_name(self):
+        """Test we can split the archive names"""
+        names = {
+            'foo-bar-baz-1.0-rc2': ('foo-bar-baz', '1.0c2'),
+            'foo-bar-baz-1.0': ('foo-bar-baz', '1.0'),
+            'foobarbaz-1.0': ('foobarbaz', '1.0'),
+        }
+        for name, results in names.items():
+            self.assertEqual(results, split_archive_name(name))
+
+    @use_pypi_server("downloads_with_md5")
+    def test_download(self, server):
+        """Download is possible, and the md5 is checked if given"""
+
+        url = "%s/simple/foobar/foobar-0.1.tar.gz" % server.full_address
+        # check md5 if given
+        dist = Dist("FooBar", "0.1", url=url,
+            url_hashname="md5", url_hashval="d41d8cd98f00b204e9800998ecf8427e")
+        dist.download()
+
+        # a wrong md5 fails
+        dist2 = Dist("FooBar", "0.1", url=url,
+            url_hashname="md5", url_hashval="wrongmd5")
+        self.assertRaises(HashDoesNotMatch, dist2.download)
+
+        # we can omit the md5 hash
+        dist3 = Dist("FooBar", "0.1", url=url)
+        dist3.download()
+
+        # and specify a temporary location
+        # for an already downloaded dist
+        path1 = tempfile.mkdtemp()
+        dist3.download(path=path1)
+        # and for a new one
+        path2_base = tempfile.mkdtemp()
+        dist4 = Dist("FooBar", "0.1", url=url)
+        path2 = dist4.download(path=path2_base)
+        self.assertTrue(path2_base in path2)
+
+        # remove the temp folders
+        shutil.rmtree(path1)
+        shutil.rmtree(os.path.dirname(path2))
+
+    def test_hashname(self):
+        """Invalid hashnames raises an exception on assignation"""
+        # should be ok
+        Dist("FooBar", "0.1", url_hashname="md5", url_hashval="value")
+
+        self.assertRaises(UnsupportedHashName, Dist, "FooBar", "0.1",
+                          url_hashname="invalid_hashname", url_hashval="value")
+
+
+class TestPyPIDistributions(unittest.TestCase):
+    """test the pypi.distr.PyPIDistributions class"""
+
+    def test_filter(self):
+        """Test we filter the distributions the right way, using version
+        predicate match method"""
+        dists = Dists((
+            Dist("FooBar", "1.1"),
+            Dist("FooBar", "1.1.1"),
+            Dist("FooBar", "1.2"),
+            Dist("FooBar", "1.2.1"),
+        ))
+        filtered = dists.filter(VersionPredicate("FooBar (<1.2)"))
+        self.assertNotIn(dists[2], filtered)
+        self.assertNotIn(dists[3], filtered)
+        self.assertIn(dists[0], filtered)
+        self.assertIn(dists[1], filtered)
+
+    def test_append(self):
+        """Test the append method of PyPIDistributions"""
+        # When adding a new item to the list, the behavior is to test if
+        # a distribution with the same name and version number already exists,
+        # and if so, to add url informations to the existing PyPIDistribution
+        # object.
+        # If no object matches, just add "normally" the object to the list.
+
+        dists = Dists([
+            Dist("FooBar", "1.1", url="external_url", type="source"),
+        ])
+        self.assertEqual(1, len(dists))
+        dists.append(Dist("FooBar", "1.1", url="internal_url",
+                          url_is_external=False, type="source"))
+        self.assertEqual(1, len(dists))
+        self.assertEqual(2, len(dists[0]._urls))
+
+        dists.append(Dist("Foobar", "1.1.1", type="source"))
+        self.assertEqual(2, len(dists))
+
+        # when adding a distribution whith a different type, a new distribution
+        # has to be added.
+        dists.append(Dist("Foobar", "1.1.1", type="binary"))
+        self.assertEqual(3, len(dists))
+
+    def test_prefer_final(self):
+        """Ordering support prefer_final"""
+
+        fb10 = Dist("FooBar", "1.0")  # final distribution
+        fb11a = Dist("FooBar", "1.1a1")  # alpha
+        fb12a = Dist("FooBar", "1.2a1")  # alpha
+        fb12b = Dist("FooBar", "1.2b1")  # beta
+        dists = Dists([fb10, fb11a, fb12a, fb12b])
+
+        dists.sort_distributions(prefer_final=True)
+        self.assertEqual(fb10, dists[0])
+
+        dists.sort_distributions(prefer_final=False)
+        self.assertEqual(fb12b, dists[0])
+
+    def test_prefer_source(self):
+        """Ordering support prefer_source"""
+        fb_source = Dist("FooBar", "1.0", type="source")
+        fb_binary = Dist("FooBar", "1.0", type="binary")
+        fb2_binary = Dist("FooBar", "2.0", type="binary")
+        dists = Dists([fb_binary, fb_source])
+
+        dists.sort_distributions(prefer_source=True)
+        self.assertEqual(fb_source, dists[0])
+
+        dists.sort_distributions(prefer_source=False)
+        self.assertEqual(fb_binary, dists[0])
+
+        dists.append(fb2_binary)
+        dists.sort_distributions(prefer_source=True)
+        self.assertEqual(fb2_binary, dists[0])
+
+    def test_get_same_name_and_version(self):
+        """PyPIDistributions can return a list of "duplicates"
+        """
+        fb_source = Dist("FooBar", "1.0", type="source")
+        fb_binary = Dist("FooBar", "1.0", type="binary")
+        fb2_binary = Dist("FooBar", "2.0", type="binary")
+        dists = Dists([fb_binary, fb_source, fb2_binary])
+        duplicates = dists.get_same_name_and_version()
+        self.assertTrue(1, len(duplicates))
+        self.assertIn(fb_source, duplicates[0])
+
+
+def test_suite():
+    suite = unittest.TestSuite()
+    suite.addTest(unittest.makeSuite(TestPyPIDistribution))
+    suite.addTest(unittest.makeSuite(TestPyPIDistributions))
+    return suite
+
+if __name__ == '__main__':
+    run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_pypi_server.py b/src/distutils2/tests/test_pypi_server.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/test_pypi_server.py
@@ -0,0 +1,70 @@
+"""Tests for distutils.command.bdist."""
+import urllib
+import urllib2
+import os.path
+
+from distutils2.tests.pypi_server import PyPIServer, PYPI_DEFAULT_STATIC_PATH
+from distutils2.tests.support import unittest
+
+
+class PyPIServerTest(unittest.TestCase):
+
+    def test_records_requests(self):
+        """We expect that PyPIServer can log our requests"""
+        server = PyPIServer()
+        server.start()
+        self.assertEqual(len(server.requests), 0)
+
+        data = "Rock Around The Bunker"
+        headers = {"X-test-header": "Mister Iceberg"}
+
+        request = urllib2.Request(server.full_address, data, headers)
+        urllib2.urlopen(request)
+        self.assertEqual(len(server.requests), 1)
+        handler, request_data = server.requests[-1]
+        self.assertIn("Rock Around The Bunker", request_data)
+        self.assertIn("x-test-header", handler.headers.dict)
+        self.assertEqual(handler.headers.dict["x-test-header"],
+                         "Mister Iceberg")
+        server.stop()
+
+    def test_serve_static_content(self):
+        """PYPI Mocked server can serve static content from disk.
+        """
+
+        def uses_local_files_for(server, url_path):
+            """Test that files are served statically (eg. the output from the
+            server is the same than the one made by a simple file read.
+            """
+            url = server.full_address + url_path
+            request = urllib2.Request(url)
+            response = urllib2.urlopen(request)
+            file = open(PYPI_DEFAULT_STATIC_PATH + "/test_pypi_server" +
+               url_path)
+            return response.read() == file.read()
+
+        server = PyPIServer(static_uri_paths=["simple", "external"],
+            static_filesystem_paths=["test_pypi_server"])
+        server.start()
+
+        # the file does not exists on the disc, so it might not be served
+        url = server.full_address + "/simple/unexisting_page"
+        request = urllib2.Request(url)
+        try:
+            urllib2.urlopen(request)
+        except urllib2.HTTPError,e:
+            self.assertEqual(e.code, 404)
+
+        # now try serving a content that do exists
+        self.assertTrue(uses_local_files_for(server, "/simple/index.html"))
+
+        # and another one in another root path
+        self.assertTrue(uses_local_files_for(server, "/external/index.html"))
+        server.stop()
+
+
+def test_suite():
+    return unittest.makeSuite(PyPIServerTest)
+
+if __name__ == '__main__':
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_pypi_simple.py b/src/distutils2/tests/test_pypi_simple.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/test_pypi_simple.py
@@ -0,0 +1,294 @@
+"""Tests for the pypi.simple module.
+
+"""
+import sys
+import os
+import shutil
+import tempfile
+import urllib2
+
+from distutils2.tests import support
+from distutils2.tests.support import unittest
+from distutils2.tests.pypi_server import use_pypi_server, PyPIServer, \
+                                         PYPI_DEFAULT_STATIC_PATH
+from distutils2.pypi import simple
+
+from distutils2.errors import DistutilsError
+
+
+class PyPISimpleTestCase(support.TempdirManager,
+                         unittest.TestCase):
+
+    def _get_simple_index(self, server, base_url="/simple/", hosts=None,
+        *args, **kwargs):
+        """Build and return a SimpleSimpleIndex instance, with the test server
+        urls
+        """
+        if hosts is None:
+            hosts = (server.full_address.strip("http://"),)
+        kwargs['hosts'] = hosts
+        return simple.SimpleIndex(server.full_address + base_url, *args,
+            **kwargs)
+
+    def test_bad_urls(self):
+        index = simple.SimpleIndex()
+        url = 'http://127.0.0.1:0/nonesuch/test_simple'
+        try:
+            v = index._open_url(url)
+        except Exception, v:
+            self.assertTrue(url in str(v))
+        else:
+            self.assertTrue(isinstance(v, urllib2.HTTPError))
+
+        # issue 16
+        # easy_install inquant.contentmirror.plone breaks because of a typo
+        # in its home URL
+        index = simple.SimpleIndex(
+            hosts=('www.example.com',))
+
+        url = 'url:%20https://svn.plone.org/svn/collective/inquant.contentmirror.plone/trunk'
+        try:
+            v = index._open_url(url)
+        except Exception, v:
+            self.assertTrue(url in str(v))
+        else:
+            self.assertTrue(isinstance(v, urllib2.HTTPError))
+
+        def _urlopen(*args):
+            import httplib
+            raise httplib.BadStatusLine('line')
+
+        old_urlopen = urllib2.urlopen
+        urllib2.urlopen = _urlopen
+        url = 'http://example.com'
+        try:
+            try:
+                v = index._open_url(url)
+            except Exception, v:
+                self.assertTrue('line' in str(v))
+            else:
+                raise AssertionError('Should have raise here!')
+        finally:
+            urllib2.urlopen = old_urlopen
+
+        # issue 20
+        url = 'http://http://svn.pythonpaste.org/Paste/wphp/trunk'
+        try:
+            index._open_url(url)
+        except Exception, v:
+            self.assertTrue('nonnumeric port' in str(v))
+
+        # issue #160
+        if sys.version_info[0] == 2 and sys.version_info[1] == 7:
+            # this should not fail
+            url = 'http://example.com'
+            page = ('<a href="http://www.famfamfam.com]('
+                    'http://www.famfamfam.com/">')
+            index.process_index(url, page)
+
+    @use_pypi_server("test_found_links")
+    def test_found_links(self, server):
+        """Browse the index, asking for a specified distribution version
+        """
+        # The PyPI index contains links for version 1.0, 1.1, 2.0 and 2.0.1
+        index = self._get_simple_index(server)
+        last_distribution = index.get("foobar")
+
+        # we have scanned the index page
+        self.assertIn(server.full_address + "/simple/foobar/",
+            index._processed_urls)
+
+        # we have found 4 distributions in this page
+        self.assertEqual(len(index._distributions["foobar"]), 4)
+
+        # and returned the most recent one
+        self.assertEqual("%s" % last_distribution.version, '2.0.1')
+
+    def test_is_browsable(self):
+        index = simple.SimpleIndex(follow_externals=False)
+        self.assertTrue(index._is_browsable(index.index_url + "test"))
+
+        # Now, when following externals, we can have a list of hosts to trust.
+        # and don't follow other external links than the one described here.
+        index = simple.SimpleIndex(hosts=["pypi.python.org", "test.org"],
+                                   follow_externals=True)
+        good_urls = (
+            "http://pypi.python.org/foo/bar",
+            "http://pypi.python.org/simple/foobar",
+            "http://test.org",
+            "http://test.org/",
+            "http://test.org/simple/",
+        )
+        bad_urls = (
+            "http://python.org",
+            "http://test.tld",
+        )
+
+        for url in good_urls:
+            self.assertTrue(index._is_browsable(url))
+
+        for url in bad_urls:
+            self.assertFalse(index._is_browsable(url))
+
+        # allow all hosts
+        index = simple.SimpleIndex(follow_externals=True, hosts=("*",))
+        self.assertTrue(index._is_browsable("http://an-external.link/path"))
+        self.assertTrue(index._is_browsable("pypi.test.tld/a/path"))
+
+        # specify a list of hosts we want to allow
+        index = simple.SimpleIndex(follow_externals=True,
+                                   hosts=("*.test.tld",))
+        self.assertFalse(index._is_browsable("http://an-external.link/path"))
+        self.assertTrue(index._is_browsable("http://pypi.test.tld/a/path"))
+
+    @use_pypi_server("with_externals")
+    def test_restrict_hosts(self, server):
+        """Include external pages
+        """
+        # Try to request the package index, wich contains links to "externals"
+        # resources. They have to  be scanned too.
+        index = self._get_simple_index(server, follow_externals=True)
+        index.get("foobar")
+        self.assertIn(server.full_address + "/external/external.html",
+            index._processed_urls)
+
+    @use_pypi_server("with_real_externals")
+    def test_restrict_hosts(self, server):
+        """Only use a list of allowed hosts is possible
+        """
+        # Test that telling the simple pyPI client to not retrieve external
+        # works
+        index = self._get_simple_index(server, follow_externals=False)
+        index.get("foobar")
+        self.assertNotIn(server.full_address + "/external/external.html",
+            index._processed_urls)
+
+    @use_pypi_server("with_egg_files")
+    def test_scan_egg_files(self, server):
+        """Assert that egg files are indexed as well"""
+        pass
+
+    @use_pypi_server(static_filesystem_paths=["with_externals"],
+        static_uri_paths=["simple", "external"])
+    def test_links_priority(self, server):
+        """
+        Download links from the pypi simple index should be used before
+        external download links.
+        http://bitbucket.org/tarek/distribute/issue/163/md5-validation-error
+
+        Usecase :
+        - someone uploads a package on pypi, a md5 is generated
+        - someone manually coindexes this link (with the md5 in the url) onto
+          an external page accessible from the package page.
+        - someone reuploads the package (with a different md5)
+        - while easy_installing, an MD5 error occurs because the external link
+          is used
+        -> The index should use the link from pypi, not the external one.
+        """
+        # start an index server
+        index_url = server.full_address + '/simple/'
+
+        # scan a test index
+        index = simple.SimpleIndex(index_url, follow_externals=True)
+        dists = index.find("foobar")
+        server.stop()
+
+        # we have only one link, because links are compared without md5
+        self.assertEqual(len(dists), 1)
+        # the link should be from the index
+        self.assertEqual('12345678901234567', dists[0].url['hashval'])
+        self.assertEqual('md5', dists[0].url['hashname'])
+
+    @use_pypi_server(static_filesystem_paths=["with_norel_links"],
+        static_uri_paths=["simple", "external"])
+    def test_not_scan_all_links(self, server):
+        """Do not follow all index page links.
+        The links not tagged with rel="download" and rel="homepage" have
+        to not be processed by the package index, while processing "pages".
+        """
+        # process the pages
+        index = self._get_simple_index(server, follow_externals=True)
+        index.find("foobar")
+        # now it should have processed only pages with links rel="download"
+        # and rel="homepage"
+        self.assertIn("%s/simple/foobar/" % server.full_address,
+            index._processed_urls)  # it's the simple index page
+
+        # FIXME
+        return
+
+        self.assertIn("%s/external/homepage.html" % server.full_address,
+            index._processed_urls)  # the external homepage is rel="homepage"
+        self.assertNotIn("%s/external/nonrel.html" % server.full_address,
+            index._processed_urls)  # this link contains no rel=*
+        self.assertNotIn("%s/unrelated-0.2.tar.gz" % server.full_address,
+            index._processed_urls)  # linked from simple index (no rel)
+        self.assertIn("%s/foobar-0.1.tar.gz" % server.full_address,
+            index._processed_urls)  # linked from simple index (rel)
+        self.assertIn("%s/foobar-2.0.tar.gz" % server.full_address,
+            index._processed_urls)  # linked from external homepage (rel)
+
+    def test_uses_mirrors(self):
+        """When the main repository seems down, try using the given mirrors"""
+        server = PyPIServer("foo_bar_baz")
+        mirror = PyPIServer("foo_bar_baz")
+        mirror.start()  # we dont start the server here
+
+        try:
+            # create the index using both servers
+            index = simple.SimpleIndex(server.full_address + "/simple/",
+                hosts=('*',), timeout=1,  # set the timeout to 1s for the tests
+                mirrors=[mirror.full_address + "/simple/",])
+
+            # this should not raise a timeout
+            self.assertEqual(4, len(index.find("foo")))
+        finally:
+            mirror.stop()
+
+    def test_simple_link_matcher(self):
+        """Test that the simple link matcher yields the right links"""
+        index = simple.SimpleIndex(follow_externals=False)
+
+        # Here, we define:
+        #   1. one link that must be followed, cause it's a download one
+        #   2. one link that must *not* be followed, cause the is_browsable
+        #      returns false for it.
+        #   3. one link that must be followed cause it's a homepage that is
+        #      browsable
+        self.assertTrue(index._is_browsable("%stest" % index.index_url))
+        self.assertFalse(index._is_browsable("http://dl-link2"))
+        content = """
+        <a href="http://dl-link1" rel="download">download_link1</a>
+        <a href="http://dl-link2" rel="homepage">homepage_link1</a>
+        <a href="%stest" rel="homepage">homepage_link2</a>
+        """ % index.index_url
+
+        # Test that the simple link matcher yield the good links.
+        generator = index._simple_link_matcher(content, index.index_url)
+        self.assertEqual(('http://dl-link1', True), generator.next())
+        self.assertEqual(('%stest' % index.index_url, False),
+                         generator.next())
+        self.assertRaises(StopIteration, generator.next)
+
+        # Follow the external links is possible
+        index.follow_externals = True
+        generator = index._simple_link_matcher(content, index.index_url)
+        self.assertEqual(('http://dl-link1', True), generator.next())
+        self.assertEqual(('http://dl-link2', False), generator.next())
+        self.assertEqual(('%stest' % index.index_url, False),
+                         generator.next())
+        self.assertRaises(StopIteration, generator.next)
+
+    def test_browse_local_files(self):
+        """Test that we can browse local files"""
+        index_path = os.sep.join(["file://" + PYPI_DEFAULT_STATIC_PATH,
+                                  "test_found_links", "simple"])
+        index = simple.SimpleIndex(index_path)
+        dists = index.find("foobar")
+        self.assertEqual(4, len(dists))
+
+def test_suite():
+    return unittest.makeSuite(PyPISimpleTestCase)
+
+if __name__ == '__main__':
+    run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_register.py b/src/distutils2/tests/test_register.py
--- a/src/distutils2/tests/test_register.py
+++ b/src/distutils2/tests/test_register.py
@@ -124,7 +124,7 @@
 
         # with the content similar to WANTED_PYPIRC
         content = open(self.rc).read()
-        self.assertEquals(content, WANTED_PYPIRC)
+        self.assertEqual(content, WANTED_PYPIRC)
 
         # now let's make sure the .pypirc file generated
         # really works : we shouldn't be asked anything
@@ -141,7 +141,7 @@
         self.assertTrue(self.conn.reqs, 2)
         req1 = dict(self.conn.reqs[0].headers)
         req2 = dict(self.conn.reqs[1].headers)
-        self.assertEquals(req2['Content-length'], req1['Content-length'])
+        self.assertEqual(req2['Content-length'], req1['Content-length'])
         self.assertTrue('xxx' in self.conn.reqs[1].data)
 
     def test_password_not_in_file(self):
@@ -154,7 +154,7 @@
 
         # dist.password should be set
         # therefore used afterwards by other commands
-        self.assertEquals(cmd.distribution.password, 'password')
+        self.assertEqual(cmd.distribution.password, 'password')
 
     def test_registering(self):
         # this test runs choice 2
@@ -171,7 +171,7 @@
         self.assertTrue(self.conn.reqs, 1)
         req = self.conn.reqs[0]
         headers = dict(req.headers)
-        self.assertEquals(headers['Content-length'], '608')
+        self.assertEqual(headers['Content-length'], '608')
         self.assertTrue('tarek' in req.data)
 
     def test_password_reset(self):
@@ -189,7 +189,7 @@
         self.assertTrue(self.conn.reqs, 1)
         req = self.conn.reqs[0]
         headers = dict(req.headers)
-        self.assertEquals(headers['Content-length'], '290')
+        self.assertEqual(headers['Content-length'], '290')
         self.assertTrue('tarek' in req.data)
 
     @unittest.skipUnless(DOCUTILS_SUPPORT, 'needs docutils')
@@ -246,8 +246,8 @@
         cmd.ensure_finalized()
         cmd.distribution.metadata['Requires-Dist'] = ['lxml']
         data = cmd.build_post_data('submit')
-        self.assertEquals(data['metadata_version'], '1.2')
-        self.assertEquals(data['requires_dist'], ['lxml'])
+        self.assertEqual(data['metadata_version'], '1.2')
+        self.assertEqual(data['requires_dist'], ['lxml'])
 
 def test_suite():
     return unittest.makeSuite(RegisterTestCase)
diff --git a/src/distutils2/tests/test_sdist.py b/src/distutils2/tests/test_sdist.py
--- a/src/distutils2/tests/test_sdist.py
+++ b/src/distutils2/tests/test_sdist.py
@@ -20,7 +20,6 @@
 
 from os.path import join
 import sys
-import tempfile
 import warnings
 
 from distutils2.tests import captured_stdout
@@ -128,7 +127,7 @@
         # now let's check what we have
         dist_folder = join(self.tmp_dir, 'dist')
         files = os.listdir(dist_folder)
-        self.assertEquals(files, ['fake-1.0.zip'])
+        self.assertEqual(files, ['fake-1.0.zip'])
 
         zip_file = zipfile.ZipFile(join(dist_folder, 'fake-1.0.zip'))
         try:
@@ -137,7 +136,7 @@
             zip_file.close()
 
         # making sure everything has been pruned correctly
-        self.assertEquals(len(content), 4)
+        self.assertEqual(len(content), 4)
 
     @unittest.skipUnless(zlib, "requires zlib")
     def test_make_distribution(self):
@@ -159,7 +158,7 @@
         dist_folder = join(self.tmp_dir, 'dist')
         result = os.listdir(dist_folder)
         result.sort()
-        self.assertEquals(result,
+        self.assertEqual(result,
                           ['fake-1.0.tar', 'fake-1.0.tar.gz'] )
 
         os.remove(join(dist_folder, 'fake-1.0.tar'))
@@ -173,7 +172,7 @@
 
         result = os.listdir(dist_folder)
         result.sort()
-        self.assertEquals(result,
+        self.assertEqual(result,
                 ['fake-1.0.tar', 'fake-1.0.tar.gz'])
 
     @unittest.skipUnless(zlib, "requires zlib")
@@ -223,7 +222,7 @@
         # now let's check what we have
         dist_folder = join(self.tmp_dir, 'dist')
         files = os.listdir(dist_folder)
-        self.assertEquals(files, ['fake-1.0.zip'])
+        self.assertEqual(files, ['fake-1.0.zip'])
 
         zip_file = zipfile.ZipFile(join(dist_folder, 'fake-1.0.zip'))
         try:
@@ -232,11 +231,11 @@
             zip_file.close()
 
         # making sure everything was added
-        self.assertEquals(len(content), 11)
+        self.assertEqual(len(content), 11)
 
         # checking the MANIFEST
         manifest = open(join(self.tmp_dir, 'MANIFEST')).read()
-        self.assertEquals(manifest, MANIFEST % {'sep': os.sep})
+        self.assertEqual(manifest, MANIFEST % {'sep': os.sep})
 
     @unittest.skipUnless(zlib, "requires zlib")
     def test_metadata_check_option(self):
@@ -248,7 +247,7 @@
         cmd.ensure_finalized()
         cmd.run()
         warnings = self.get_logs(WARN)
-        self.assertEquals(len(warnings), 1)
+        self.assertEqual(len(warnings), 1)
 
         # trying with a complete set of metadata
         self.clear_logs()
@@ -260,7 +259,7 @@
         # removing manifest generated warnings
         warnings = [warn for warn in warnings if
                     not warn.endswith('-- skipping')]
-        self.assertEquals(len(warnings), 0)
+        self.assertEqual(len(warnings), 0)
 
 
     def test_show_formats(self):
@@ -270,7 +269,7 @@
         num_formats = len(get_archive_formats())
         output = [line for line in stdout.split('\n')
                   if line.strip().startswith('--formats=')]
-        self.assertEquals(len(output), num_formats)
+        self.assertEqual(len(output), num_formats)
 
     def test_finalize_options(self):
 
@@ -278,9 +277,9 @@
         cmd.finalize_options()
 
         # default options set by finalize
-        self.assertEquals(cmd.manifest, 'MANIFEST')
-        self.assertEquals(cmd.template, 'MANIFEST.in')
-        self.assertEquals(cmd.dist_dir, 'dist')
+        self.assertEqual(cmd.manifest, 'MANIFEST')
+        self.assertEqual(cmd.template, 'MANIFEST.in')
+        self.assertEqual(cmd.dist_dir, 'dist')
 
         # formats has to be a string splitable on (' ', ',') or
         # a stringlist
@@ -317,8 +316,8 @@
         archive = tarfile.open(archive_name)
         try:
             for member in archive.getmembers():
-                self.assertEquals(member.uid, 0)
-                self.assertEquals(member.gid, 0)
+                self.assertEqual(member.uid, 0)
+                self.assertEqual(member.gid, 0)
         finally:
             archive.close()
 
@@ -339,7 +338,7 @@
         # rights (see #7408)
         try:
             for member in archive.getmembers():
-                self.assertEquals(member.uid, os.getuid())
+                self.assertEqual(member.uid, os.getuid())
         finally:
             archive.close()
 
diff --git a/src/distutils2/tests/test_spawn.py b/src/distutils2/tests/test_spawn.py
--- a/src/distutils2/tests/test_spawn.py
+++ b/src/distutils2/tests/test_spawn.py
@@ -20,7 +20,7 @@
                                (['nochange', 'nospace'],
                                 ['nochange', 'nospace'])):
             res = _nt_quote_args(args)
-            self.assertEquals(res, wanted)
+            self.assertEqual(res, wanted)
 
 
     @unittest.skipUnless(os.name in ('nt', 'posix'),
diff --git a/src/distutils2/tests/test_upload.py b/src/distutils2/tests/test_upload.py
--- a/src/distutils2/tests/test_upload.py
+++ b/src/distutils2/tests/test_upload.py
@@ -1,34 +1,15 @@
 """Tests for distutils.command.upload."""
 # -*- encoding: utf8 -*-
-import sys
-import os
+import os, sys
 
-from distutils2.command import upload as upload_mod
 from distutils2.command.upload import upload
 from distutils2.core import Distribution
 
+from distutils2.tests.pypi_server import PyPIServer, PyPIServerTestCase
 from distutils2.tests import support
 from distutils2.tests.support import unittest
 from distutils2.tests.test_config import PYPIRC, PyPIRCCommandTestCase
 
-PYPIRC_LONG_PASSWORD = """\
-[distutils]
-
-index-servers =
-    server1
-    server2
-
-[server1]
-username:me
-password:aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
-
-[server2]
-username:meagain
-password: secret
-realm:acme
-repository:http://another.pypi/
-"""
-
 
 PYPIRC_NOPASSWORD = """\
 [distutils]
@@ -40,38 +21,9 @@
 username:me
 """
 
-class FakeOpen(object):
-
-    def __init__(self, url):
-        self.url = url
-        if not isinstance(url, str):
-            self.req = url
-        else:
-            self.req = None
-        self.msg = 'OK'
-
-    def getcode(self):
-        return 200
-
-
-class uploadTestCase(PyPIRCCommandTestCase):
-
-    def setUp(self):
-        super(uploadTestCase, self).setUp()
-        self.old_open = upload_mod.urlopen
-        upload_mod.urlopen = self._urlopen
-        self.last_open = None
-
-    def tearDown(self):
-        upload_mod.urlopen = self.old_open
-        super(uploadTestCase, self).tearDown()
-
-    def _urlopen(self, url):
-        self.last_open = FakeOpen(url)
-        return self.last_open
+class UploadTestCase(PyPIServerTestCase, PyPIRCCommandTestCase):
 
     def test_finalize_options(self):
-
         # new format
         self.write_file(self.rc, PYPIRC)
         dist = Distribution()
@@ -80,7 +32,7 @@
         for attr, waited in (('username', 'me'), ('password', 'secret'),
                              ('realm', 'pypi'),
                              ('repository', 'http://pypi.python.org/pypi')):
-            self.assertEquals(getattr(cmd, attr), waited)
+            self.assertEqual(getattr(cmd, attr), waited)
 
     def test_saved_password(self):
         # file with no password
@@ -89,44 +41,42 @@
         # make sure it passes
         dist = Distribution()
         cmd = upload(dist)
-        cmd.finalize_options()
-        self.assertEquals(cmd.password, None)
+        cmd.ensure_finalized()
+        self.assertEqual(cmd.password, None)
 
         # make sure we get it as well, if another command
         # initialized it at the dist level
         dist.password = 'xxx'
         cmd = upload(dist)
         cmd.finalize_options()
-        self.assertEquals(cmd.password, 'xxx')
+        self.assertEqual(cmd.password, 'xxx')
 
     def test_upload(self):
-        tmp = self.mkdtemp()
-        path = os.path.join(tmp, 'xxx')
+        path = os.path.join(self.tmp_dir, 'xxx')
         self.write_file(path)
         command, pyversion, filename = 'xxx', '2.6', path
         dist_files = [(command, pyversion, filename)]
-        self.write_file(self.rc, PYPIRC_LONG_PASSWORD)
 
         # lets run it
         pkg_dir, dist = self.create_dist(dist_files=dist_files, author=u'd??d??')
         cmd = upload(dist)
         cmd.ensure_finalized()
+        cmd.repository = self.pypi.full_address
         cmd.run()
 
         # what did we send ?
-        self.assertIn('d??d??', self.last_open.req.data)
-        headers = dict(self.last_open.req.headers)
-        self.assertTrue(int(headers['Content-length']) < 2000)
-        self.assertTrue(headers['Content-type'].startswith('multipart/form-data'))
-        self.assertEquals(self.last_open.req.get_method(), 'POST')
-        self.assertEquals(self.last_open.req.get_full_url(),
-                          'http://pypi.python.org/pypi')
-        self.assertTrue('xxx' in self.last_open.req.data)
-        auth = self.last_open.req.headers['Authorization']
-        self.assertFalse('\n' in auth)
+        handler, request_data = self.pypi.requests[-1]
+        headers = handler.headers.dict
+        self.assertIn('d??d??', request_data)
+        self.assertIn('xxx', request_data)
+        self.assertEqual(int(headers['content-length']), len(request_data))
+        self.assertTrue(int(headers['content-length']) < 2000)
+        self.assertTrue(headers['content-type'].startswith('multipart/form-data'))
+        self.assertEqual(handler.command, 'POST')
+        self.assertNotIn('\n', headers['authorization'])
 
 def test_suite():
-    return unittest.makeSuite(uploadTestCase)
+    return unittest.makeSuite(UploadTestCase)
 
 if __name__ == "__main__":
     unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_upload_docs.py b/src/distutils2/tests/test_upload_docs.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/test_upload_docs.py
@@ -0,0 +1,198 @@
+"""Tests for distutils.command.upload_docs."""
+# -*- encoding: utf8 -*-
+import httplib, os, os.path, shutil, sys, tempfile, zipfile
+from cStringIO import StringIO
+
+from distutils2.command import upload_docs as upload_docs_mod
+from distutils2.command.upload_docs import (upload_docs, zip_dir,
+                                    encode_multipart)
+from distutils2.core import Distribution
+
+from distutils2.errors import DistutilsFileError, DistutilsOptionError
+
+from distutils2.tests import support
+from distutils2.tests.pypi_server import PyPIServer, PyPIServerTestCase
+from distutils2.tests.test_config import PyPIRCCommandTestCase
+from distutils2.tests.support import unittest
+
+
+EXPECTED_MULTIPART_OUTPUT = "\r\n".join([
+'---x',
+'Content-Disposition: form-data; name="a"',
+'',
+'b',
+'---x',
+'Content-Disposition: form-data; name="c"',
+'',
+'d',
+'---x',
+'Content-Disposition: form-data; name="e"; filename="f"',
+'',
+'g',
+'---x',
+'Content-Disposition: form-data; name="h"; filename="i"',
+'',
+'j',
+'---x--',
+'',
+])
+
+PYPIRC = """\
+[distutils]
+index-servers = server1
+
+[server1]
+repository = %s
+username = real_slim_shady
+password = long_island
+"""
+
+class UploadDocsTestCase(PyPIServerTestCase, PyPIRCCommandTestCase):
+
+    def setUp(self):
+        super(UploadDocsTestCase, self).setUp()
+        self.dist = Distribution()
+        self.dist.metadata['Name'] = "distr-name"
+        self.cmd = upload_docs(self.dist)
+
+    def test_default_uploaddir(self):
+        sandbox = tempfile.mkdtemp()
+        previous = os.getcwd()
+        os.chdir(sandbox)
+        try:
+            os.mkdir("build")
+            self.prepare_sample_dir("build")
+            self.cmd.ensure_finalized()
+            self.assertEqual(self.cmd.upload_dir, os.path.join("build", "docs"))
+        finally:
+            os.chdir(previous)
+
+    def prepare_sample_dir(self, sample_dir=None):
+        if sample_dir is None:
+            sample_dir = tempfile.mkdtemp()
+        os.mkdir(os.path.join(sample_dir, "docs"))
+        self.write_file(os.path.join(sample_dir, "docs", "index.html"), "Ce mortel ennui")
+        self.write_file(os.path.join(sample_dir, "index.html"), "Oh la la")
+        return sample_dir
+
+    def test_zip_dir(self):
+        source_dir = self.prepare_sample_dir()
+        compressed = zip_dir(source_dir)
+
+        zip_f = zipfile.ZipFile(compressed)
+        self.assertEqual(zip_f.namelist(), ['index.html', 'docs/index.html'])
+
+    def test_encode_multipart(self):
+        fields = [("a", "b"), ("c", "d")]
+        files = [("e", "f", "g"), ("h", "i", "j")]
+        content_type, body = encode_multipart(fields, files, "-x")
+        self.assertEqual(content_type, "multipart/form-data; boundary=-x")
+        self.assertEqual(body, EXPECTED_MULTIPART_OUTPUT)
+
+    def prepare_command(self):
+        self.cmd.upload_dir = self.prepare_sample_dir()
+        self.cmd.ensure_finalized()
+        self.cmd.repository = self.pypi.full_address
+        self.cmd.username = "username"
+        self.cmd.password = "password"
+
+    def test_upload(self):
+        self.prepare_command()
+        self.cmd.run()
+
+        self.assertEqual(len(self.pypi.requests), 1)
+        handler, request_data = self.pypi.requests[-1]
+        self.assertIn("content", request_data)
+        self.assertIn("Basic", handler.headers.dict['authorization'])
+        self.assertTrue(handler.headers.dict['content-type']
+            .startswith('multipart/form-data;'))
+
+        action, name, content =\
+            request_data.split("----------------GHSKFJDLGDS7543FJKLFHRE75642756743254")[1:4]
+
+        # check that we picked the right chunks
+        self.assertIn('name=":action"', action)
+        self.assertIn('name="name"', name)
+        self.assertIn('name="content"', content)
+
+        # check their contents
+        self.assertIn("doc_upload", action)
+        self.assertIn("distr-name", name)
+        self.assertIn("docs/index.html", content)
+        self.assertIn("Ce mortel ennui", content)
+
+    def test_https_connection(self):
+        https_called = False
+        orig_https = upload_docs_mod.httplib.HTTPSConnection
+        def https_conn_wrapper(*args):
+            https_called = True
+            return upload_docs_mod.httplib.HTTPConnection(*args) # the testing server is http
+        upload_docs_mod.httplib.HTTPSConnection = https_conn_wrapper
+        try:
+            self.prepare_command()
+            self.cmd.run()
+            self.assertFalse(https_called)
+
+            self.cmd.repository = self.cmd.repository.replace("http", "https")
+            self.cmd.run()
+            self.assertFalse(https_called)
+        finally:
+            upload_docs_mod.httplib.HTTPSConnection = orig_https
+
+    def test_handling_response(self):
+        calls = []
+        def aggr(*args):
+            calls.append(args)
+        self.pypi.default_response_status = '403 Forbidden'
+        self.prepare_command()
+        self.cmd.announce = aggr
+        self.cmd.run()
+        message, _ = calls[-1]
+        self.assertIn('Upload failed (403): Forbidden', message)
+
+        calls = []
+        self.pypi.default_response_status = '301 Moved Permanently'
+        self.pypi.default_response_headers.append(("Location", "brand_new_location"))
+        self.cmd.run()
+        message, _ = calls[-1]
+        self.assertIn('brand_new_location', message)
+
+    def test_reads_pypirc_data(self):
+        self.write_file(self.rc, PYPIRC % self.pypi.full_address)
+        self.cmd.repository = self.pypi.full_address
+        self.cmd.upload_dir = self.prepare_sample_dir()
+        self.cmd.ensure_finalized()
+        self.assertEqual(self.cmd.username, "real_slim_shady")
+        self.assertEqual(self.cmd.password, "long_island")
+
+    def test_checks_index_html_presence(self):
+        self.cmd.upload_dir = self.prepare_sample_dir()
+        os.remove(os.path.join(self.cmd.upload_dir, "index.html"))
+        self.assertRaises(DistutilsFileError, self.cmd.ensure_finalized)
+
+    def test_checks_upload_dir(self):
+        self.cmd.upload_dir = self.prepare_sample_dir()
+        shutil.rmtree(os.path.join(self.cmd.upload_dir))
+        self.assertRaises(DistutilsOptionError, self.cmd.ensure_finalized)
+
+    def test_show_response(self):
+        orig_stdout = sys.stdout
+        write_args = []
+        class MockStdIn(object):
+            def write(self, arg):
+                write_args.append(arg)
+        sys.stdout = MockStdIn()
+        try:
+            self.prepare_command()
+            self.cmd.show_response = True
+            self.cmd.run()
+        finally:
+            sys.stdout = orig_stdout
+        self.assertTrue(write_args[0], "should report the response")
+        self.assertIn(self.pypi.default_response_data + "\n", write_args[0])
+
+def test_suite():
+    return unittest.makeSuite(UploadDocsTestCase)
+
+if __name__ == "__main__":
+    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_util.py b/src/distutils2/tests/test_util.py
--- a/src/distutils2/tests/test_util.py
+++ b/src/distutils2/tests/test_util.py
@@ -100,7 +100,7 @@
             return '/'.join(path)
         os.path.join = _join
 
-        self.assertEquals(convert_path('/home/to/my/stuff'),
+        self.assertEqual(convert_path('/home/to/my/stuff'),
                           '/home/to/my/stuff')
 
         # win
@@ -112,9 +112,9 @@
         self.assertRaises(ValueError, convert_path, '/home/to/my/stuff')
         self.assertRaises(ValueError, convert_path, 'home/to/my/stuff/')
 
-        self.assertEquals(convert_path('home/to/my/stuff'),
+        self.assertEqual(convert_path('home/to/my/stuff'),
                           'home\\to\\my\\stuff')
-        self.assertEquals(convert_path('.'),
+        self.assertEqual(convert_path('.'),
                           os.curdir)
 
     def test_change_root(self):
@@ -127,9 +127,9 @@
             return '/'.join(path)
         os.path.join = _join
 
-        self.assertEquals(change_root('/root', '/old/its/here'),
+        self.assertEqual(change_root('/root', '/old/its/here'),
                           '/root/old/its/here')
-        self.assertEquals(change_root('/root', 'its/here'),
+        self.assertEqual(change_root('/root', 'its/here'),
                           '/root/its/here')
 
         # windows
@@ -146,9 +146,9 @@
             return '\\'.join(path)
         os.path.join = _join
 
-        self.assertEquals(change_root('c:\\root', 'c:\\old\\its\\here'),
+        self.assertEqual(change_root('c:\\root', 'c:\\old\\its\\here'),
                           'c:\\root\\old\\its\\here')
-        self.assertEquals(change_root('c:\\root', 'its\\here'),
+        self.assertEqual(change_root('c:\\root', 'its\\here'),
                           'c:\\root\\its\\here')
 
         # BugsBunny os (it's a great os)
@@ -159,7 +159,7 @@
         # XXX platforms to be covered: os2, mac
 
     def test_split_quoted(self):
-        self.assertEquals(split_quoted('""one"" "two" \'three\' \\four'),
+        self.assertEqual(split_quoted('""one"" "two" \'three\' \\four'),
                           ['one', 'two', 'three', 'four'])
 
     def test_strtobool(self):
@@ -177,7 +177,7 @@
         res = rfc822_escape(header)
         wanted = ('I am a%(8s)spoor%(8s)slonesome%(8s)s'
                   'header%(8s)s') % {'8s': '\n'+8*' '}
-        self.assertEquals(res, wanted)
+        self.assertEqual(res, wanted)
 
     def test_find_exe_version(self):
         # the ld version scheme under MAC OS is:
@@ -195,7 +195,7 @@
                                 ('@(#)PROGRAM:ld  PROJECT:ld64-95.2.12',
                                  '95.2.12')):
             result = _MAC_OS_X_LD_VERSION.search(output)
-            self.assertEquals(result.group(1), version)
+            self.assertEqual(result.group(1), version)
 
     def _find_executable(self, name):
         if name in self._exes:
@@ -205,43 +205,43 @@
     def test_get_compiler_versions(self):
         # get_versions calls distutils.spawn.find_executable on
         # 'gcc', 'ld' and 'dllwrap'
-        self.assertEquals(get_compiler_versions(), (None, None, None))
+        self.assertEqual(get_compiler_versions(), (None, None, None))
 
         # Let's fake we have 'gcc' and it returns '3.4.5'
         self._exes['gcc'] = 'gcc (GCC) 3.4.5 (mingw special)\nFSF'
         res = get_compiler_versions()
-        self.assertEquals(str(res[0]), '3.4.5')
+        self.assertEqual(str(res[0]), '3.4.5')
 
         # and let's see what happens when the version
         # doesn't match the regular expression
         # (\d+\.\d+(\.\d+)*)
         self._exes['gcc'] = 'very strange output'
         res = get_compiler_versions()
-        self.assertEquals(res[0], None)
+        self.assertEqual(res[0], None)
 
         # same thing for ld
         if sys.platform != 'darwin':
             self._exes['ld'] = 'GNU ld version 2.17.50 20060824'
             res = get_compiler_versions()
-            self.assertEquals(str(res[1]), '2.17.50')
+            self.assertEqual(str(res[1]), '2.17.50')
             self._exes['ld'] = '@(#)PROGRAM:ld  PROJECT:ld64-77'
             res = get_compiler_versions()
-            self.assertEquals(res[1], None)
+            self.assertEqual(res[1], None)
         else:
             self._exes['ld'] = 'GNU ld version 2.17.50 20060824'
             res = get_compiler_versions()
-            self.assertEquals(res[1], None)
+            self.assertEqual(res[1], None)
             self._exes['ld'] = '@(#)PROGRAM:ld  PROJECT:ld64-77'
             res = get_compiler_versions()
-            self.assertEquals(str(res[1]), '77')
+            self.assertEqual(str(res[1]), '77')
 
         # and dllwrap
         self._exes['dllwrap'] = 'GNU dllwrap 2.17.50 20060824\nFSF'
         res = get_compiler_versions()
-        self.assertEquals(str(res[2]), '2.17.50')
+        self.assertEqual(str(res[2]), '2.17.50')
         self._exes['dllwrap'] = 'Cheese Wrap'
         res = get_compiler_versions()
-        self.assertEquals(res[2], None)
+        self.assertEqual(res[2], None)
 
     @unittest.skipUnless(hasattr(sys, 'dont_write_bytecode'),
                           'no dont_write_bytecode support')
@@ -294,7 +294,38 @@
         self.write_file(os.path.join(pkg5, '__init__.py'))
 
         res = find_packages([root], ['pkg1.pkg2'])
-        self.assertEquals(set(res), set(['pkg1', 'pkg5', 'pkg1.pkg3', 'pkg1.pkg3.pkg6']))
+        self.assertEqual(set(res), set(['pkg1', 'pkg5', 'pkg1.pkg3', 'pkg1.pkg3.pkg6']))
+
+    @unittest.skipUnless(sys.version > '2.6', 'Need Python 2.6 or more')
+    def test_run_2to3_on_code(self):
+        content = "print 'test'"
+        converted_content = "print('test')"
+        file_handle = self.mktempfile()
+        file_name = file_handle.name
+        file_handle.write(content)
+        file_handle.flush()
+        file_handle.seek(0)
+        from distutils2.util import run_2to3
+        run_2to3([file_name])
+        new_content = "".join(file_handle.read())
+        file_handle.close()
+        self.assertEquals(new_content, converted_content)
+
+    @unittest.skipUnless(sys.version > '2.6', 'Need Python 2.6 or more')
+    def test_run_2to3_on_doctests(self):
+        # to check if text files containing doctests only get converted.
+        content = ">>> print 'test'\ntest\n"
+        converted_content = ">>> print('test')\ntest\n\n"
+        file_handle = self.mktempfile()
+        file_name = file_handle.name
+        file_handle.write(content)
+        file_handle.flush()
+        file_handle.seek(0)
+        from distutils2.util import run_2to3
+        run_2to3([file_name], doctests_only=True)
+        new_content = "".join(file_handle.readlines())
+        file_handle.close()
+        self.assertEquals(new_content, converted_content)
 
 
 def test_suite():
diff --git a/src/distutils2/tests/test_version.py b/src/distutils2/tests/test_version.py
--- a/src/distutils2/tests/test_version.py
+++ b/src/distutils2/tests/test_version.py
@@ -25,15 +25,15 @@
     def test_basic_versions(self):
 
         for v, s in self.versions:
-            self.assertEquals(str(v), s)
+            self.assertEqual(str(v), s)
 
     def test_from_parts(self):
 
         for v, s in self.versions:
             parts = v.parts
             v2 = V.from_parts(*v.parts)
-            self.assertEquals(v, v2)
-            self.assertEquals(str(v), str(v2))
+            self.assertEqual(v, v2)
+            self.assertEqual(str(v), str(v2))
 
     def test_irrational_versions(self):
 
@@ -96,34 +96,34 @@
 
     def test_suggest_normalized_version(self):
 
-        self.assertEquals(suggest('1.0'), '1.0')
-        self.assertEquals(suggest('1.0-alpha1'), '1.0a1')
-        self.assertEquals(suggest('1.0c2'), '1.0c2')
-        self.assertEquals(suggest('walla walla washington'), None)
-        self.assertEquals(suggest('2.4c1'), '2.4c1')
+        self.assertEqual(suggest('1.0'), '1.0')
+        self.assertEqual(suggest('1.0-alpha1'), '1.0a1')
+        self.assertEqual(suggest('1.0c2'), '1.0c2')
+        self.assertEqual(suggest('walla walla washington'), None)
+        self.assertEqual(suggest('2.4c1'), '2.4c1')
 
         # from setuptools
-        self.assertEquals(suggest('0.4a1.r10'), '0.4a1.post10')
-        self.assertEquals(suggest('0.7a1dev-r66608'), '0.7a1.dev66608')
-        self.assertEquals(suggest('0.6a9.dev-r41475'), '0.6a9.dev41475')
-        self.assertEquals(suggest('2.4preview1'), '2.4c1')
-        self.assertEquals(suggest('2.4pre1') , '2.4c1')
-        self.assertEquals(suggest('2.1-rc2'), '2.1c2')
+        self.assertEqual(suggest('0.4a1.r10'), '0.4a1.post10')
+        self.assertEqual(suggest('0.7a1dev-r66608'), '0.7a1.dev66608')
+        self.assertEqual(suggest('0.6a9.dev-r41475'), '0.6a9.dev41475')
+        self.assertEqual(suggest('2.4preview1'), '2.4c1')
+        self.assertEqual(suggest('2.4pre1') , '2.4c1')
+        self.assertEqual(suggest('2.1-rc2'), '2.1c2')
 
         # from pypi
-        self.assertEquals(suggest('0.1dev'), '0.1.dev0')
-        self.assertEquals(suggest('0.1.dev'), '0.1.dev0')
+        self.assertEqual(suggest('0.1dev'), '0.1.dev0')
+        self.assertEqual(suggest('0.1.dev'), '0.1.dev0')
 
         # we want to be able to parse Twisted
         # development versions are like post releases in Twisted
-        self.assertEquals(suggest('9.0.0+r2363'), '9.0.0.post2363')
+        self.assertEqual(suggest('9.0.0+r2363'), '9.0.0.post2363')
 
         # pre-releases are using markers like "pre1"
-        self.assertEquals(suggest('9.0.0pre1'), '9.0.0c1')
+        self.assertEqual(suggest('9.0.0pre1'), '9.0.0c1')
 
         # we want to be able to parse Tcl-TK
         # they us "p1" "p2" for post releases
-        self.assertEquals(suggest('1.4p1'), '1.4.post1')
+        self.assertEqual(suggest('1.4p1'), '1.4.post1')
 
     def test_predicate(self):
         # VersionPredicate knows how to parse stuff like:
@@ -154,6 +154,17 @@
         # XXX need to silent the micro version in this case
         #assert not VersionPredicate('Ho (<3.0,!=2.6)').match('2.6.3')
 
+    def test_is_final(self):
+        # VersionPredicate knows is a distribution is a final one or not.
+        final_versions = ('1.0', '1.0.post456')
+        other_versions = ('1.0.dev1', '1.0a2', '1.0c3')
+
+        for version in final_versions:
+            self.assertTrue(V(version).is_final)
+        for version in other_versions:
+            self.assertFalse(V(version).is_final)
+
+
 def test_suite():
     #README = os.path.join(os.path.dirname(__file__), 'README.txt')
     #suite = [doctest.DocFileSuite(README), unittest.makeSuite(VersionTestCase)]
@@ -162,4 +173,3 @@
 
 if __name__ == "__main__":
     unittest.main(defaultTest="test_suite")
-
diff --git a/src/distutils2/util.py b/src/distutils2/util.py
--- a/src/distutils2/util.py
+++ b/src/distutils2/util.py
@@ -1,12 +1,15 @@
 """distutils.util
 
-Miscellaneous utility functions -- anything that doesn't fit into
-one of the other *util.py modules.
+Miscellaneous utility functions.
 """
 
 __revision__ = "$Id: util.py 77761 2010-01-26 22:46:15Z tarek.ziade $"
 
-import sys, os, string, re
+import sys
+import os
+import string
+import re
+from copy import copy
 from fnmatch import fnmatchcase
 
 from distutils2.errors import (DistutilsPlatformError, DistutilsFileError,
@@ -17,6 +20,7 @@
 
 _PLATFORM = None
 
+
 def newer(source, target):
     """Tells if the target is newer than the source.
 
@@ -37,6 +41,7 @@
 
     return os.stat(source).st_mtime > os.stat(target).st_mtime
 
+
 def get_platform():
     """Return a string that identifies the current platform.
 
@@ -48,6 +53,7 @@
         _PLATFORM = _sysconfig.get_platform()
     return _PLATFORM
 
+
 def set_platform(identifier):
     """Sets the platform string identifier returned by get_platform().
 
@@ -57,6 +63,7 @@
     global _PLATFORM
     _PLATFORM = identifier
 
+
 def convert_path(pathname):
     """Return 'pathname' as a name that will work on the native filesystem.
 
@@ -125,6 +132,7 @@
 
 _environ_checked = 0
 
+
 def check_environ():
     """Ensure that 'os.environ' has all the environment variables needed.
 
@@ -147,6 +155,7 @@
 
     _environ_checked = 1
 
+
 def subst_vars(s, local_vars):
     """Perform shell/Perl-style variable substitution on 'string'.
 
@@ -158,7 +167,8 @@
     variables not found in either 'local_vars' or 'os.environ'.
     """
     check_environ()
-    def _subst (match, local_vars=local_vars):
+
+    def _subst(match, local_vars=local_vars):
         var_name = match.group(1)
         if var_name in local_vars:
             return str(local_vars[var_name])
@@ -170,6 +180,7 @@
     except KeyError, var:
         raise ValueError("invalid variable '$%s'" % var)
 
+
 def grok_environment_error(exc, prefix="error: "):
     """Generate a useful error message from an EnvironmentError.
 
@@ -196,12 +207,14 @@
 # Needed by 'split_quoted()'
 _wordchars_re = _squote_re = _dquote_re = None
 
+
 def _init_regex():
     global _wordchars_re, _squote_re, _dquote_re
     _wordchars_re = re.compile(r'[^\\\'\"%s ]*' % string.whitespace)
     _squote_re = re.compile(r"'(?:[^'\\]|\\.)*'")
     _dquote_re = re.compile(r'"(?:[^"\\]|\\.)*"')
 
+
 def split_quoted(s):
     """Split a string up according to Unix shell-like rules for quotes and
     backslashes.
@@ -217,7 +230,8 @@
     # This is a nice algorithm for splitting up a single string, since it
     # doesn't require character-by-character examination.  It was a little
     # bit of a brain-bender to get it working right, though...
-    if _wordchars_re is None: _init_regex()
+    if _wordchars_re is None:
+        _init_regex()
 
     s = s.strip()
     words = []
@@ -237,8 +251,8 @@
 
         elif s[end] == '\\':            # preserve whatever is being escaped;
                                         # will become part of the current word
-            s = s[:end] + s[end+1:]
-            pos = end+1
+            s = s[:end] + s[end + 1:]
+            pos = end + 1
 
         else:
             if s[end] == "'":           # slurp singly-quoted string
@@ -253,7 +267,7 @@
                 raise ValueError("bad string (mismatched %s quotes?)" % s[end])
 
             (beg, end) = m.span()
-            s = s[:beg] + s[beg+1:end-1] + s[end:]
+            s = s[:beg] + s[beg + 1:end - 1] + s[end:]
             pos = m.end() - 2
 
         if pos >= len(s):
@@ -296,7 +310,7 @@
     elif val in ('n', 'no', 'f', 'false', 'off', '0'):
         return 0
     else:
-        raise ValueError, "invalid truth value %r" % (val,)
+        raise ValueError("invalid truth value %r" % (val,))
 
 
 def byte_compile(py_files, optimize=0, force=0, prefix=None, base_dir=None,
@@ -350,12 +364,8 @@
     # "Indirect" byte-compilation: write a temporary script and then
     # run it with the appropriate flags.
     if not direct:
-        try:
-            from tempfile import mkstemp
-            (script_fd, script_name) = mkstemp(".py")
-        except ImportError:
-            from tempfile import mktemp
-            (script_fd, script_name) = None, mktemp(".py")
+        from tempfile import mkstemp
+        script_fd, script_name = mkstemp(".py")
         log.info("writing byte-compilation script '%s'", script_name)
         if not dry_run:
             if script_fd is not None:
@@ -363,43 +373,50 @@
             else:
                 script = open(script_name, "w")
 
-            script.write("""\
-from distutils.util import byte_compile
+            try:
+                script.write("""\
+from distutils2.util import byte_compile
 files = [
 """)
 
-            # XXX would be nice to write absolute filenames, just for
-            # safety's sake (script should be more robust in the face of
-            # chdir'ing before running it).  But this requires abspath'ing
-            # 'prefix' as well, and that breaks the hack in build_lib's
-            # 'byte_compile()' method that carefully tacks on a trailing
-            # slash (os.sep really) to make sure the prefix here is "just
-            # right".  This whole prefix business is rather delicate -- the
-            # problem is that it's really a directory, but I'm treating it
-            # as a dumb string, so trailing slashes and so forth matter.
+                # XXX would be nice to write absolute filenames, just for
+                # safety's sake (script should be more robust in the face of
+                # chdir'ing before running it).  But this requires abspath'ing
+                # 'prefix' as well, and that breaks the hack in build_lib's
+                # 'byte_compile()' method that carefully tacks on a trailing
+                # slash (os.sep really) to make sure the prefix here is "just
+                # right".  This whole prefix business is rather delicate -- the
+                # problem is that it's really a directory, but I'm treating it
+                # as a dumb string, so trailing slashes and so forth matter.
 
-            #py_files = map(os.path.abspath, py_files)
-            #if prefix:
-            #    prefix = os.path.abspath(prefix)
+                #py_files = map(os.path.abspath, py_files)
+                #if prefix:
+                #    prefix = os.path.abspath(prefix)
 
-            script.write(",\n".join(map(repr, py_files)) + "]\n")
-            script.write("""
+                script.write(",\n".join(map(repr, py_files)) + "]\n")
+                script.write("""
 byte_compile(files, optimize=%r, force=%r,
              prefix=%r, base_dir=%r,
              verbose=%r, dry_run=0,
              direct=1)
 """ % (optimize, force, prefix, base_dir, verbose))
 
-            script.close()
+            finally:
+                script.close()
 
         cmd = [sys.executable, script_name]
         if optimize == 1:
             cmd.insert(1, "-O")
         elif optimize == 2:
             cmd.insert(1, "-OO")
-        spawn(cmd, dry_run=dry_run)
-        execute(os.remove, (script_name,), "removing %s" % script_name,
-                dry_run=dry_run)
+
+        env = copy(os.environ)
+        env['PYTHONPATH'] = ':'.join(sys.path)
+        try:
+            spawn(cmd, dry_run=dry_run, env=env)
+        finally:
+            execute(os.remove, (script_name,), "removing %s" % script_name,
+                    dry_run=dry_run)
 
     # "Direct" byte-compilation: use the py_compile module to compile
     # right here, right now.  Note that the script generated in indirect
@@ -447,7 +464,9 @@
     return sep.join(lines)
 
 _RE_VERSION = re.compile('(\d+\.\d+(\.\d+)*)')
-_MAC_OS_X_LD_VERSION = re.compile('^@\(#\)PROGRAM:ld  PROJECT:ld64-((\d+)(\.\d+)*)')
+_MAC_OS_X_LD_VERSION = re.compile('^@\(#\)PROGRAM:ld  '
+                                  'PROJECT:ld64-((\d+)(\.\d+)*)')
+
 
 def _find_ld_version():
     """Finds the ld version. The version scheme differs under Mac OSX."""
@@ -456,6 +475,7 @@
     else:
         return _find_exe_version('ld -v')
 
+
 def _find_exe_version(cmd, pattern=_RE_VERSION):
     """Find the version of an executable by running `cmd` in the shell.
 
@@ -485,6 +505,7 @@
         return None
     return result.group(1)
 
+
 def get_compiler_versions():
     """Returns a tuple providing the versions of gcc, ld and dllwrap
 
@@ -496,6 +517,7 @@
     dllwrap = _find_exe_version('dllwrap --version')
     return gcc, ld, dllwrap
 
+
 def newer_group(sources, target, missing='error'):
     """Return true if 'target' is out-of-date with respect to any file
     listed in 'sources'.
@@ -535,14 +557,18 @@
 
     return False
 
+
 def write_file(filename, contents):
     """Create a file with the specified name and write 'contents' (a
     sequence of strings without line terminators) to it.
     """
-    f = open(filename, "w")
-    for line in contents:
-        f.write(line + "\n")
-    f.close()
+    try:
+        f = open(filename, "w")
+        for line in contents:
+            f.write(line + "\n")
+    finally:
+        f.close()
+
 
 def _is_package(path):
     """Returns True if path is a package (a dir with an __init__ file."""
@@ -550,6 +576,7 @@
         return False
     return os.path.isfile(os.path.join(path, '__init__.py'))
 
+
 def _under(path, root):
     path = path.split(os.sep)
     root = root.split(os.sep)
@@ -560,12 +587,14 @@
             return False
     return True
 
+
 def _package_name(root_path, path):
     """Returns a dotted package name, given a subpath."""
     if not _under(path, root_path):
         raise ValueError('"%s" is not a subpath of "%s"' % (path, root_path))
     return path[len(root_path) + 1:].replace(os.sep, '.')
 
+
 def find_packages(paths=('.',), exclude=()):
     """Return a list all Python packages found recursively within
     directories 'paths'
@@ -580,6 +609,7 @@
     """
     packages = []
     discarded = []
+
     def _discarded(path):
         for discard in discarded:
             if _under(path, discard):
@@ -611,3 +641,50 @@
                 packages.append(package_name)
     return packages
 
+
+# utility functions for 2to3 support
+
+def run_2to3(files, doctests_only=False, fixer_names=None, options=None,
+                                                            explicit=None):
+    """ Wrapper function around the refactor() class which
+    performs the conversions on a list of python files.
+    Invoke 2to3 on a list of Python files. The files should all come
+    from the build area, as the modification is done in-place."""
+
+    if not files:
+        return
+
+    # Make this class local, to delay import of 2to3
+    from lib2to3.refactor import get_fixers_from_package
+    from distutils2.converter.refactor import DistutilsRefactoringTool
+
+    if fixer_names is None:
+        fixer_names = get_fixers_from_package('lib2to3.fixes')
+
+    r = DistutilsRefactoringTool(fixer_names, options=options)
+    if doctests_only:
+        r.refactor(files, doctests_only=True, write=True)
+    else:
+        r.refactor(files, write=True)
+
+
+class Mixin2to3:
+    """ Wrapper class for commands that run 2to3.
+    To configure 2to3, setup scripts may either change
+    the class variables, or inherit from this class
+    to override how 2to3 is invoked.
+    """
+    # provide list of fixers to run.
+    # defaults to all from lib2to3.fixers
+    fixer_names = None
+
+    # options dictionary
+    options = None
+
+    # list of fixers to invoke even though they are marked as explicit
+    explicit = None
+
+    def run_2to3(self, files, doctests_only=False):
+        """ Issues a call to util.run_2to3. """
+        return run_2to3(files, doctests_only, self.fixer_names,
+                        self.options, self.explicit)
diff --git a/src/distutils2/version.py b/src/distutils2/version.py
--- a/src/distutils2/version.py
+++ b/src/distutils2/version.py
@@ -36,6 +36,7 @@
     (?P<postdev>(\.post(?P<post>\d+))?(\.dev(?P<dev>\d+))?)?
     $''', re.VERBOSE)
 
+
 class NormalizedVersion(object):
     """A rational version.
 
@@ -61,7 +62,8 @@
         @param error_on_huge_major_num {bool} Whether to consider an
             apparent use of a year or full date as the major version number
             an error. Default True. One of the observed patterns on PyPI before
-            the introduction of `NormalizedVersion` was version numbers like this:
+            the introduction of `NormalizedVersion` was version numbers like
+            this:
                 2009.01.03
                 20040603
                 2005.01
@@ -71,6 +73,7 @@
             the possibility of using a version number like "1.0" (i.e.
             where the major number is less than that huge major number).
         """
+        self.is_final = True  # by default, consider a version as final.
         self._parse(s, error_on_huge_major_num)
 
     @classmethod
@@ -101,6 +104,7 @@
             block += self._parse_numdots(groups.get('prerelversion'), s,
                                          pad_zeros_length=1)
             parts.append(tuple(block))
+            self.is_final = False
         else:
             parts.append(_FINAL_MARKER)
 
@@ -115,6 +119,7 @@
                     postdev.append(_FINAL_MARKER[0])
             if dev is not None:
                 postdev.extend(['dev', int(dev)])
+                self.is_final = False
             parts.append(tuple(postdev))
         else:
             parts.append(_FINAL_MARKER)
@@ -204,6 +209,7 @@
     # See http://docs.python.org/reference/datamodel#object.__hash__
     __hash__ = object.__hash__
 
+
 def suggest_normalized_version(s):
     """Suggest a normalized version close to the given version string.
 
@@ -215,7 +221,7 @@
     on observation of versions currently in use on PyPI. Given a dump of
     those version during PyCon 2009, 4287 of them:
     - 2312 (53.93%) match NormalizedVersion without change
-    - with the automatic suggestion
+      with the automatic suggestion
     - 3474 (81.04%) match when using this suggestion method
 
     @param s {str} An irrational version string.
@@ -305,7 +311,6 @@
     # PyPI stats: ~21 (0.62%) better
     rs = re.sub(r"\.?(pre|preview|-c)(\d+)$", r"c\g<2>", rs)
 
-
     # Tcl/Tk uses "px" for their post release markers
     rs = re.sub(r"p(\d+)$", r".post\1", rs)
 
@@ -322,6 +327,7 @@
 _PLAIN_VERSIONS = re.compile(r"^\s*(.*)\s*$")
 _SPLIT_CMP = re.compile(r"^\s*(<=|>=|<|>|!=|==)\s*([^\s,]+)\s*$")
 
+
 def _split_predicate(predicate):
     match = _SPLIT_CMP.match(predicate)
     if match is None:
@@ -368,6 +374,7 @@
                 return False
         return True
 
+
 class _Versions(VersionPredicate):
     def __init__(self, predicate):
         predicate = predicate.strip()
@@ -379,6 +386,7 @@
         self.predicates = [_split_predicate(pred.strip())
                            for pred in predicates.split(',')]
 
+
 class _Version(VersionPredicate):
     def __init__(self, predicate):
         predicate = predicate.strip()
@@ -388,6 +396,7 @@
         self.name = None
         self.predicates = _split_predicate(match.groups()[0])
 
+
 def is_valid_predicate(predicate):
     try:
         VersionPredicate(predicate)
@@ -396,6 +405,7 @@
     else:
         return True
 
+
 def is_valid_versions(predicate):
     try:
         _Versions(predicate)
@@ -404,6 +414,7 @@
     else:
         return True
 
+
 def is_valid_version(predicate):
     try:
         _Version(predicate)
@@ -411,4 +422,3 @@
         return False
     else:
         return True
-
diff --git a/src/runtests.py b/src/runtests.py
--- a/src/runtests.py
+++ b/src/runtests.py
@@ -1,6 +1,6 @@
 """Tests for distutils2.
 
-The tests for distutils2 are defined in the distutils2.tests package;
+The tests for distutils2 are defined in the distutils2.tests package.
 """
 import sys
 
@@ -8,8 +8,7 @@
     import distutils2.tests
     from distutils2.tests import run_unittest, reap_children, TestFailed
     from distutils2._backport.tests import test_suite as btest_suite
-    # just supporting -q right now
-    # to enable detailed/quiet output
+    # XXX just supporting -q right now to enable detailed/quiet output
     if len(sys.argv) > 1:
         verbose = sys.argv[-1] != '-q'
     else:
@@ -17,7 +16,7 @@
     try:
         try:
             run_unittest([distutils2.tests.test_suite(), btest_suite()],
-                    verbose_=verbose)
+                         verbose_=verbose)
             return 0
         except TestFailed:
             return 1
@@ -28,7 +27,7 @@
     try:
         from distutils2.tests.support import unittest
     except ImportError:
-        print('Error: You have to install unittest2')
+        sys.stderr.write('Error: You have to install unittest2')
         sys.exit(1)
 
     sys.exit(test_main())
diff --git a/src/setup.py b/src/setup.py
--- a/src/setup.py
+++ b/src/setup.py
@@ -3,8 +3,12 @@
 __revision__ = "$Id$"
 import sys
 import os
+import re
 
-from distutils2.core import setup
+from distutils2 import __version__ as VERSION
+from distutils2 import log
+from distutils2.core import setup, Extension
+from distutils2.compiler.ccompiler import new_compiler
 from distutils2.command.sdist import sdist
 from distutils2.command.install import install
 from distutils2 import __version__ as VERSION
@@ -31,6 +35,7 @@
 
 DEV_SUFFIX = '.dev%d' % get_tip_revision('..')
 
+
 class install_hg(install):
 
     user_options = install.user_options + [
@@ -62,10 +67,141 @@
             self.distribution.metadata.version += DEV_SUFFIX
         sdist.run(self)
 
+
+# additional paths to check, set from the command line
+SSL_INCDIR = ''   # --openssl-incdir=
+SSL_LIBDIR = ''   # --openssl-libdir=
+SSL_DIR = ''      # --openssl-prefix=
+
+def add_dir_to_list(dirlist, dir):
+    """Add the directory 'dir' to the list 'dirlist' (at the front) if
+    'dir' actually exists and is a directory.  If 'dir' is already in
+    'dirlist' it is moved to the front.
+    """
+    if dir is not None and os.path.isdir(dir) and dir not in dirlist:
+        if dir in dirlist:
+            dirlist.remove(dir)
+        dirlist.insert(0, dir)
+
+
+def prepare_hashlib_extensions():
+    """Decide which C extensions to build and create the appropriate
+    Extension objects to build them.  Return a list of Extensions.
+    """
+    # this CCompiler object is only used to locate include files
+    compiler = new_compiler()
+
+    # Ensure that these paths are always checked
+    if os.name == 'posix':
+        add_dir_to_list(compiler.library_dirs, '/usr/local/lib')
+        add_dir_to_list(compiler.include_dirs, '/usr/local/include')
+
+        add_dir_to_list(compiler.library_dirs, '/usr/local/ssl/lib')
+        add_dir_to_list(compiler.include_dirs, '/usr/local/ssl/include')
+
+        add_dir_to_list(compiler.library_dirs, '/usr/contrib/ssl/lib')
+        add_dir_to_list(compiler.include_dirs, '/usr/contrib/ssl/include')
+
+        add_dir_to_list(compiler.library_dirs, '/usr/lib')
+        add_dir_to_list(compiler.include_dirs, '/usr/include')
+
+    # look in command line supplied paths
+    if SSL_LIBDIR:
+        add_dir_to_list(compiler.library_dirs, SSL_LIBDIR)
+    if SSL_INCDIR:
+        add_dir_to_list(compiler.include_dirs, SSL_INCDIR)
+    if SSL_DIR:
+        if os.name == 'nt':
+            add_dir_to_list(compiler.library_dirs, os.path.join(SSL_DIR, 'out32dll'))
+            # prefer the static library
+            add_dir_to_list(compiler.library_dirs, os.path.join(SSL_DIR, 'out32'))
+        else:
+            add_dir_to_list(compiler.library_dirs, os.path.join(SSL_DIR, 'lib'))
+        add_dir_to_list(compiler.include_dirs, os.path.join(SSL_DIR, 'include'))
+
+    oslibs = {'posix': ['ssl', 'crypto'],
+              'nt': ['libeay32',  'gdi32', 'advapi32', 'user32']}
+
+    if os.name not in oslibs:
+        sys.stderr.write(
+            'unknown operating system, impossible to compile _hashlib')
+        sys.exit(1)
+
+    exts = []
+
+    ssl_inc_dirs = []
+    ssl_incs = []
+    for inc_dir in compiler.include_dirs:
+        f = os.path.join(inc_dir, 'openssl', 'ssl.h')
+        if os.path.exists(f):
+            ssl_incs.append(f)
+            ssl_inc_dirs.append(inc_dir)
+
+    ssl_lib = compiler.find_library_file(compiler.library_dirs, oslibs[os.name][0])
+
+    # find out which version of OpenSSL we have
+    openssl_ver = 0
+    openssl_ver_re = re.compile(
+        '^\s*#\s*define\s+OPENSSL_VERSION_NUMBER\s+(0x[0-9a-fA-F]+)' )
+    ssl_inc_dir = ''
+    for ssl_inc_dir in ssl_inc_dirs:
+        name = os.path.join(ssl_inc_dir, 'openssl', 'opensslv.h')
+        if os.path.isfile(name):
+            try:
+                incfile = open(name, 'r')
+                for line in incfile:
+                    m = openssl_ver_re.match(line)
+                    if m:
+                        openssl_ver = int(m.group(1), 16)
+                        break
+            except IOError:
+                pass
+
+        # first version found is what we'll use
+        if openssl_ver:
+            break
+
+    if (ssl_inc_dir and ssl_lib is not None and openssl_ver >= 0x00907000):
+
+        log.info('Using OpenSSL version 0x%08x from', openssl_ver)
+        log.info(' Headers:\t%s', ssl_inc_dir)
+        log.info(' Library:\t%s', ssl_lib)
+
+        # The _hashlib module wraps optimized implementations
+        # of hash functions from the OpenSSL library.
+        exts.append(Extension('_hashlib', ['_hashopenssl.c'],
+                              include_dirs = [ssl_inc_dir],
+                              library_dirs = [os.path.dirname(ssl_lib)],
+                              libraries = oslibs[os.name]))
+    else:
+        exts.append(Extension('_sha', ['shamodule.c']) )
+        exts.append(Extension('_md5',
+                              sources=['md5module.c', 'md5.c'],
+                              depends=['md5.h']) )
+
+    if (not ssl_lib or openssl_ver < 0x00908000):
+        # OpenSSL doesn't do these until 0.9.8 so we'll bring our own
+        exts.append(Extension('_sha256', ['sha256module.c']))
+        exts.append(Extension('_sha512', ['sha512module.c']))
+
+    def prepend_modules(filename):
+        return os.path.join('Modules', filename)
+
+    # all the C code is in the Modules subdirectory, prepend the path
+    for ext in exts:
+        ext.sources = [prepend_modules(fn) for fn in ext.sources]
+        if hasattr(ext, 'depends') and ext.depends is not None:
+            ext.depends = [prepend_modules(fn) for fn in ext.depends]
+
+    return exts
+
 setup_kwargs = {}
 if sys.version < '2.6':
     setup_kwargs['scripts'] = ['distutils2/mkpkg.py']
 
+if sys.version < '2.5':
+    setup_kwargs['ext_modules'] = prepare_hashlib_extensions()
+
 _CLASSIFIERS = """\
 Development Status :: 3 - Alpha
 Intended Audience :: Developers
@@ -77,26 +213,23 @@
 Topic :: System :: Systems Administration
 Topic :: Utilities"""
 
-setup (name="Distutils2",
-       version=VERSION,
-       summary="Python Distribution Utilities",
-       keywords=['packaging', 'distutils'],
-       author="Tarek Ziade",
-       author_email="tarek at ziade.org",
-       home_page="http://bitbucket.org/tarek/distutils2/wiki/Home",
-       license="PSF",
-       description=README,
-       classifier=_CLASSIFIERS.split('\n'),
-       packages=find_packages(),
-       cmdclass={'sdist': sdist_hg, 'install': install_hg},
-       package_data={'distutils2._backport': ['sysconfig.cfg']},
-       project_url=[('Mailing-list',
+setup(name="Distutils2",
+      version=VERSION,
+      summary="Python Distribution Utilities",
+      keywords=['packaging', 'distutils'],
+      author="Tarek Ziade",
+      author_email="tarek at ziade.org",
+      home_page="http://bitbucket.org/tarek/distutils2/wiki/Home",
+      license="PSF",
+      description=README,
+      classifier=_CLASSIFIERS.split('\n'),
+      packages=find_packages(),
+      cmdclass={'sdist': sdist_hg, 'install': install_hg},
+      package_data={'distutils2._backport': ['sysconfig.cfg']},
+      project_url=[('Mailing-list',
                     'http://mail.python.org/mailman/listinfo/distutils-sig/'),
-                    ('Documentation',
-                     'http://packages.python.org/Distutils2'),
-                    ('Repository', 'http://hg.python.org/distutils2'),
-                    ('Bug tracker', 'http://bugs.python.org')],
-       **setup_kwargs
-       )
-
-
+                   ('Documentation',
+                    'http://packages.python.org/Distutils2'),
+                   ('Repository', 'http://hg.python.org/distutils2'),
+                   ('Bug tracker', 'http://bugs.python.org')],
+      **setup_kwargs)
diff --git a/src/tests.sh b/src/tests.sh
--- a/src/tests.sh
+++ b/src/tests.sh
@@ -1,28 +1,40 @@
 #!/bin/sh
-echo -n "Running tests for Python 2.4..."
-python2.4 runtests.py -q > /dev/null 2> /dev/null
+echo -n "Running tests for Python 2.4... "
+rm -rf _hashlib.so
+python2.4 setup.py build_ext -i -q 2> /dev/null > /dev/null
+python2.4 -Wd runtests.py -q 2> /dev/null
 if [ $? -ne 0 ];then
     echo "Failed"
-    exit $1
+    exit 1
 else
     echo "Success"
 fi
 
-echo -n "Running tests for Python 2.5..."
-python2.5 runtests.py -q > /dev/null 2> /dev/null
+echo -n "Running tests for Python 2.5... "
+rm -rf _hashlib.so
+python2.5 setup.py build_ext -i -q 2> /dev/null > /dev/null
+python2.5 -Wd runtests.py -q 2> /dev/null
 if [ $? -ne 0 ];then
     echo "Failed"
-    exit $1
+    exit 1
 else
     echo "Success"
 fi
 
-echo -n "Running tests for Python 2.6..."
-python2.6 runtests.py -q > /dev/null 2> /dev/null
+echo -n "Running tests for Python 2.6... "
+python2.6 -Wd -bb -3 runtests.py -q 2> /dev/null
 if [ $? -ne 0 ];then
     echo "Failed"
-    exit $1
+    exit 1
 else
     echo "Success"
 fi
 
+echo -n "Running tests for Python 2.7... "
+python2.7 -Wd -bb -3 runtests.py -q 2> /dev/null
+if [ $? -ne 0 ];then
+    echo "Failed"
+    exit 1
+else
+    echo "Success"
+fi

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Thu Jul 15 01:38:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Thu, 15 Jul 2010 01:38:06 +0200
Subject: [Python-checkins] distutils2: Pull fixes from jkloth
Message-ID: <E1OZBWg-0001HD-HZ@dinsdale.python.org>

tarek.ziade pushed 70b98f177413 to distutils2:

http://hg.python.org/distutils2/rev/70b98f177413
changeset:   367:70b98f177413
tag:         tip
parent:      360:2a9ad83361f7
parent:      366:5b9c3c536275
user:        ?ric Araujo <merwok at netwok.org>
date:        Wed Jul 14 15:59:45 2010 +0200
summary:     Pull fixes from jkloth
files:       

diff --git a/src/distutils2/command/build_ext.py b/src/distutils2/command/build_ext.py
--- a/src/distutils2/command/build_ext.py
+++ b/src/distutils2/command/build_ext.py
@@ -188,7 +188,24 @@
         if self.package is None:
             self.package = self.distribution.ext_package
 
+        # Ensure that the list of extensions is valid, i.e. it is a list of
+        # Extension objects.
         self.extensions = self.distribution.ext_modules
+        if self.extensions:
+            if not isinstance(self.extensions, (list, tuple)):
+                type_name = (self.extensions is None and 'None'
+                            or type(self.extensions).__name__)
+                raise DistutilsSetupError(
+                    "'ext_modules' must be a sequence of Extension instances,"
+                    " not %s" % (type_name,))
+            for i, ext in enumerate(self.extensions):
+                if isinstance(ext, Extension):
+                    continue                # OK! (assume type-checking done
+                                            # by Extension constructor)
+                type_name = (ext is None and 'None' or type(ext).__name__)
+                raise DistutilsSetupError(
+                    "'ext_modules' item %d must be an Extension instance,"
+                    " not %s" % (i, type_name))
 
         # Make sure Python's include directories (for Python.h, pyconfig.h,
         # etc.) are in the include search path.
@@ -396,86 +413,7 @@
         # Now actually compile and link everything.
         self.build_extensions()
 
-    def check_extensions_list(self, extensions):
-        """Ensure that the list of extensions (presumably provided as a
-        command option 'extensions') is valid, i.e. it is a list of
-        Extension objects.  We also support the old-style list of 2-tuples,
-        where the tuples are (ext_name, build_info), which are converted to
-        Extension instances here.
-
-        Raise DistutilsSetupError if the structure is invalid anywhere;
-        just returns otherwise.
-        """
-        if not isinstance(extensions, list):
-            raise DistutilsSetupError, \
-                  "'ext_modules' option must be a list of Extension instances"
-
-        for i, ext in enumerate(extensions):
-            if isinstance(ext, Extension):
-                continue                # OK! (assume type-checking done
-                                        # by Extension constructor)
-
-            if not isinstance(ext, tuple) or len(ext) != 2:
-                raise DistutilsSetupError, \
-                      ("each element of 'ext_modules' option must be an "
-                       "Extension instance or 2-tuple")
-
-            ext_name, build_info = ext
-
-            log.warn(("old-style (ext_name, build_info) tuple found in "
-                      "ext_modules for extension '%s'"
-                      "-- please convert to Extension instance" % ext_name))
-
-            if not (isinstance(ext_name, str) and
-                    extension_name_re.match(ext_name)):
-                raise DistutilsSetupError, \
-                      ("first element of each tuple in 'ext_modules' "
-                       "must be the extension name (a string)")
-
-            if not isinstance(build_info, dict):
-                raise DistutilsSetupError, \
-                      ("second element of each tuple in 'ext_modules' "
-                       "must be a dictionary (build info)")
-
-            # OK, the (ext_name, build_info) dict is type-safe: convert it
-            # to an Extension instance.
-            ext = Extension(ext_name, build_info['sources'])
-
-            # Easy stuff: one-to-one mapping from dict elements to
-            # instance attributes.
-            for key in ('include_dirs', 'library_dirs', 'libraries',
-                        'extra_objects', 'extra_compile_args',
-                        'extra_link_args'):
-                val = build_info.get(key)
-                if val is not None:
-                    setattr(ext, key, val)
-
-            # Medium-easy stuff: same syntax/semantics, different names.
-            ext.runtime_library_dirs = build_info.get('rpath')
-            if 'def_file' in build_info:
-                log.warn("'def_file' element of build info dict "
-                         "no longer supported")
-
-            # Non-trivial stuff: 'macros' split into 'define_macros'
-            # and 'undef_macros'.
-            macros = build_info.get('macros')
-            if macros:
-                ext.define_macros = []
-                ext.undef_macros = []
-                for macro in macros:
-                    if not (isinstance(macro, tuple) and len(macro) in (1, 2)):
-                        raise DistutilsSetupError, \
-                              ("'macros' element of build info dict "
-                               "must be 1- or 2-tuple")
-                    if len(macro) == 1:
-                        ext.undef_macros.append(macro[0])
-                    elif len(macro) == 2:
-                        ext.define_macros.append(macro)
-
-            extensions[i] = ext
-
     def get_source_files(self):
-        self.check_extensions_list(self.extensions)
         filenames = []
 
         # Wouldn't it be neat if we knew the names of header files too...
@@ -485,11 +423,6 @@
         return filenames
 
     def get_outputs(self):
-        # Sanity check the 'extensions' list -- can't assume this is being
-        # done in the same run as a 'build_extensions()' call (in fact, we
-        # can probably assume that it *isn't*!).
-        self.check_extensions_list(self.extensions)
-
         # And build the list of output (built) filenames.  Note that this
         # ignores the 'inplace' flag, and assumes everything goes in the
         # "build" tree.
@@ -499,9 +432,6 @@
         return outputs
 
     def build_extensions(self):
-        # First, sanity-check the 'extensions' list
-        self.check_extensions_list(self.extensions)
-
         for ext in self.extensions:
             try:
                 self.build_extension(ext)
diff --git a/src/distutils2/command/build_py.py b/src/distutils2/command/build_py.py
--- a/src/distutils2/command/build_py.py
+++ b/src/distutils2/command/build_py.py
@@ -151,8 +151,13 @@
 
         self.byte_compile(self.get_outputs(include_bytecode=0))
 
+    # -- Top-level worker functions ------------------------------------
+
     def get_data_files(self):
-        """Generate list of '(package,src_dir,build_dir,filenames)' tuples"""
+        """Generate list of '(package,src_dir,build_dir,filenames)' tuples.
+
+        Helper function for `finalize_options()`.
+        """
         data = []
         if not self.packages:
             return data
@@ -176,7 +181,10 @@
         return data
 
     def find_data_files(self, package, src_dir):
-        """Return filenames for package's data files in 'src_dir'"""
+        """Return filenames for package's data files in 'src_dir'.
+
+        Helper function for `get_data_files()`.
+        """
         globs = (self.package_data.get('', [])
                  + self.package_data.get(package, []))
         files = []
@@ -188,7 +196,10 @@
         return files
 
     def build_package_data(self):
-        """Copy data files into build directory"""
+        """Copy data files into build directory.
+
+        Helper function for `run()`.
+        """
         for package, src_dir, build_dir, filenames in self.data_files:
             for filename in filenames:
                 target = os.path.join(build_dir, filename)
@@ -198,6 +209,8 @@
                 if copied and srcfile in self.distribution.convert_2to3.doctests:
                     self._doctests_2to3.append(outf)
 
+    # XXX - this should be moved to the Distribution class as it is not
+    # only needed for build_py. It also has no dependencies on this class.
     def get_package_dir(self, package):
         """Return the directory, relative to the top of the source
            distribution, where package 'package' should be found
@@ -239,6 +252,8 @@
                     return ''
 
     def check_package(self, package, package_dir):
+        """Helper function for `find_package_modules()` and `find_modules()'.
+        """
         # Empty dir name means current directory, which we can probably
         # assume exists.  Also, os.path.exists and isdir don't know about
         # my "empty string means current dir" convention, so we have to
diff --git a/src/distutils2/command/cmd.py b/src/distutils2/command/cmd.py
--- a/src/distutils2/command/cmd.py
+++ b/src/distutils2/command/cmd.py
@@ -373,8 +373,10 @@
         if os.path.isdir(name) or name == '':
             return
         if dry_run:
+            head = ''
             for part in name.split(os.sep):
-                self.log(part)
+                log.info("created directory %s%s", head, part)
+                head += part + os.sep
             return
         os.makedirs(name, mode)
 
diff --git a/src/distutils2/command/register.py b/src/distutils2/command/register.py
--- a/src/distutils2/command/register.py
+++ b/src/distutils2/command/register.py
@@ -28,8 +28,6 @@
     boolean_options = PyPIRCCommand.boolean_options + [
         'verify', 'list-classifiers', 'strict']
 
-    sub_commands = [('check', lambda self: True)]
-
     def initialize_options(self):
         PyPIRCCommand.initialize_options(self)
         self.list_classifiers = 0
@@ -46,9 +44,8 @@
         self.finalize_options()
         self._set_config()
 
-        # Run sub commands
-        for cmd_name in self.get_sub_commands():
-            self.run_command(cmd_name)
+        # Check the package metadata
+        self.run_command('check')
 
         if self.dry_run:
             self.verify_metadata()
diff --git a/src/distutils2/command/sdist.py b/src/distutils2/command/sdist.py
--- a/src/distutils2/command/sdist.py
+++ b/src/distutils2/command/sdist.py
@@ -45,12 +45,6 @@
 
     description = "create a source distribution (tarball, zip file, etc.)"
 
-    def checking_metadata(self):
-        """Callable used for the check sub-command.
-
-        Placed here so user_options can view it"""
-        return self.metadata_check
-
     user_options = [
         ('template=', 't',
          "name of manifest template file [default: MANIFEST.in]"),
@@ -100,8 +94,6 @@
     default_format = {'posix': 'gztar',
                       'nt': 'zip' }
 
-    sub_commands = [('check', checking_metadata)]
-
     def initialize_options(self):
         # 'template' and 'manifest' are, respectively, the names of
         # the manifest template and manifest file.
@@ -162,9 +154,9 @@
         # manifest
         self.filelist.clear()
 
-        # Run sub commands
-        for cmd_name in self.get_sub_commands():
-            self.run_command(cmd_name)
+        # Check the package metadata
+        if self.metadata_check:
+            self.run_command('check')
 
         # Do whatever it takes to get the list of files to process
         # (process the manifest template, read an existing manifest,
diff --git a/src/distutils2/compiler/ccompiler.py b/src/distutils2/compiler/ccompiler.py
--- a/src/distutils2/compiler/ccompiler.py
+++ b/src/distutils2/compiler/ccompiler.py
@@ -940,8 +940,10 @@
         if os.path.isdir(name) or name == '':
             return
         if self.dry_run:
+            head = ''
             for part in name.split(os.sep):
-                self.log(part)
+                log.info("created directory %s%s", head, part)
+                head += part + os.sep
             return
         os.makedirs(name, mode)
 
diff --git a/src/distutils2/tests/test_build_ext.py b/src/distutils2/tests/test_build_ext.py
--- a/src/distutils2/tests/test_build_ext.py
+++ b/src/distutils2/tests/test_build_ext.py
@@ -237,53 +237,6 @@
         cmd.finalize_options()
         self.assertEqual(cmd.swig_opts, ['1', '2'])
 
-    def test_check_extensions_list(self):
-        dist = Distribution()
-        cmd = build_ext(dist)
-        cmd.finalize_options()
-
-        #'extensions' option must be a list of Extension instances
-        self.assertRaises(DistutilsSetupError, cmd.check_extensions_list, 'foo')
-
-        # each element of 'ext_modules' option must be an
-        # Extension instance or 2-tuple
-        exts = [('bar', 'foo', 'bar'), 'foo']
-        self.assertRaises(DistutilsSetupError, cmd.check_extensions_list, exts)
-
-        # first element of each tuple in 'ext_modules'
-        # must be the extension name (a string) and match
-        # a python dotted-separated name
-        exts = [('foo-bar', '')]
-        self.assertRaises(DistutilsSetupError, cmd.check_extensions_list, exts)
-
-        # second element of each tuple in 'ext_modules'
-        # must be a ary (build info)
-        exts = [('foo.bar', '')]
-        self.assertRaises(DistutilsSetupError, cmd.check_extensions_list, exts)
-
-        # ok this one should pass
-        exts = [('foo.bar', {'sources': [''], 'libraries': 'foo',
-                             'some': 'bar'})]
-        cmd.check_extensions_list(exts)
-        ext = exts[0]
-        self.assertTrue(isinstance(ext, Extension))
-
-        # check_extensions_list adds in ext the values passed
-        # when they are in ('include_dirs', 'library_dirs', 'libraries'
-        # 'extra_objects', 'extra_compile_args', 'extra_link_args')
-        self.assertEqual(ext.libraries, 'foo')
-        self.assertTrue(not hasattr(ext, 'some'))
-
-        # 'macros' element of build info dict must be 1- or 2-tuple
-        exts = [('foo.bar', {'sources': [''], 'libraries': 'foo',
-                'some': 'bar', 'macros': [('1', '2', '3'), 'foo']})]
-        self.assertRaises(DistutilsSetupError, cmd.check_extensions_list, exts)
-
-        exts[0][1]['macros'] = [('1', '2'), ('3',)]
-        cmd.check_extensions_list(exts)
-        self.assertEqual(exts[0].undef_macros, ['3'])
-        self.assertEqual(exts[0].define_macros, [('1', '2')])
-
     def test_get_source_files(self):
         modules = [Extension('foo', ['xxx'], optional=False)]
         dist = Distribution({'name': 'xx', 'ext_modules': modules})

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Thu Jul 15 01:56:40 2010
From: python-checkins at python.org (barry.warsaw)
Date: Thu, 15 Jul 2010 01:56:40 +0200 (CEST)
Subject: [Python-checkins] r82909 - peps/trunk/pep-3149.txt
Message-ID: <20100714235640.6E55DEE981@mail.python.org>

Author: barry.warsaw
Date: Thu Jul 15 01:56:40 2010
New Revision: 82909

Log:
Added PEP 3149


Added:
   peps/trunk/pep-3149.txt

Added: peps/trunk/pep-3149.txt
==============================================================================
--- (empty file)
+++ peps/trunk/pep-3149.txt	Thu Jul 15 01:56:40 2010
@@ -0,0 +1,267 @@
+PEP: 3149
+Title: ABI version tagged .so files
+Version: $Revision: 81577 $
+Last-Modified: $Date: 2010-05-27 19:54:25 -0400 (Thu, 27 May 2010) $
+Author: Barry Warsaw <barry at python.org>
+Status: Draft
+Type: Standards Track
+Content-Type: text/x-rst
+Created: 2010-07-09
+Python-Version: 3.2
+Post-History: 2010-07-14
+Resolution: TBD
+
+
+Abstract
+========
+
+PEP 3147 [1]_ described an extension to Python's import machinery that
+improved the sharing of Python source code, by allowing more than one
+byte compilation file (.pyc) to be co-located with each source file.
+
+This PEP defines an adjunct feature which allows the co-location of
+extension module files (.so) in a similar manner.  This optional,
+build-time feature will enable downstream distributions of Python to
+more easily provide more than one Python major version at a time.
+
+
+Background
+==========
+
+PEP 3147 defined the file system layout for a pure-Python package,
+where multiple versions of Python are available on the system.  For
+example, where the `alpha` package containing source modules `one.py`
+and `two.py` exist on a system with Python 3.2 and 3.3, the post-byte
+compilation file system layout would be::
+
+    alpha/
+        __pycache__/
+            __init__.cpython-32.pyc
+            __init__.cpython-33.pyc
+            one.cpython-32.pyc
+            one.cpython-33.pyc
+            two.cpython-32.pyc
+            two.cpython-33.pyc
+        __init__.py
+        one.py
+        two.py
+
+For packages with extension modules, a similar differentiation is
+needed for the module's .so files.  Extension modules compiled for
+different Python major versions are incompatible with each other due
+to changes in the ABI.  While PEP 384 [2]_ defines a stable ABI, it
+will minimize, but not eliminate extension module incompatibilities
+between Python major versions.  Thus a mechanism for discriminating
+extension module file names is proposed.
+
+
+Rationale
+=========
+
+Linux distributions such as Ubuntu [3]_ and Debian [4]_ provide more
+than one Python version at the same time to their users.  For example,
+Ubuntu 9.10 Karmic Koala users can install Python 2.5, 2.6, and 3.1,
+with Python 2.6 being the default.
+
+In order to share as much as possible between the available Python
+versions, these distributions install third party (i.e. non-standard
+library) packages into `/usr/share/pyshared` and symlink to them from
+`/usr/lib/pythonX.Y/dist-packages`.  The symlinks exist because in a
+pre-PEP 3147 world (i.e < Python 3.2), the `.pyc` files resulting from
+byte compilation by the various installed Pythons will name collide
+with each other.  For Python versions >= 3.2, all pure-Python packages
+can be shared, because the `.pyc` files will no longer cause file
+system naming conflicts.  Eliminating these symlinks makes for a
+simpler, more robust Python distribution.
+
+A similar situation arises with shared library extensions.  Because
+extension modules are typically named `foo.so` for a `foo` extension
+module, these would also name collide if `foo` was provided for more
+than one Python version.  There are several approaches that could be
+taken to avoid this, which will be explored below, but this PEP
+proposes a fairly simple compile-time option to allow extension
+modules to live in the same file system directory and avoid any name
+collisions.
+
+
+Proposal
+========
+
+A new configure option is added for building Python, called
+`--with-so-abi-tag`.  This takes as an argument a unique, but
+arbitrary string, e.g.::
+
+    ./configure --with-so-abi-tag=cpython-32
+
+This string is passed into the `Makefile` and affects two aspects of
+the Python build.  First, it is compiled into `Python/dynload_shlib.c`
+where it defines some additional `.so` file names to search for when
+importing extension modules.  Second, it modifies the `Makefile`'s
+`$SO` variable, which in turn controls the `distutils` module's default
+filename when compiling extension modules.
+
+When `--with-so-abi-tag` is not given to `configure` nothing changes
+in the way the Python executable is built, or acts.  Thus, this
+configure switch is completely optional and has no effect if not used.
+
+What this allows is for distributions that want to distinguish among
+extension modules built for different versions of Python, but shared
+in the same file system path, to arrange for `.so` names that are
+unique and non-colliding.
+
+For example, let's say Python 3.2 was built with::
+
+    ./configure --with-so-abi-tag=cpython-32
+
+and Python 3.3 was built with::
+
+    ./configure --with-so-abi-tag=cpython-33
+
+For an arbitrary package `foo`, you would see these files when the
+distribution package was installed::
+
+    /usr/share/pyshared/foo.cpython-32.so
+    /usr/share/pyshared/foo.cpython-33.so
+
+
+Proven approach
+===============
+
+The approach described here is already proven, in a sense, on Debian
+and Ubuntu system where different extensions are used for debug builds
+of Python and extension modules.  Debug builds on Windows also already
+use a different file extension for dynamic libraries.
+
+
+PEP 384
+=======
+
+PEP 384 defines a stable ABI for extension modules.  Universal
+adoption of PEP 384 would eliminate the need for this PEP because all
+extension modules would be compatible with any Python version.  In
+practice of course, it will be impossible to achieve universal
+adoption.  Older extensions may not be ported to PEP 384, or an
+extension may require Python APIs outside of PEP 384 definition.
+Therefore there will always be a (hopefully diminishing, but never
+zero) need for ABI version tagged shared libraries.
+
+Further, it is anticipated that the stable ABI will evolve over time,
+meaning that existing PEP 384 compatible extension modules may be
+incompatible with future versions of Python.  While a complete
+specification is reserved for PEP 384, here is a discussion of the
+relevant issues.
+
+PEP 384 describes a change to ``PyModule_Create()`` where ``3`` is
+passed as the API version if the extension was complied with
+``Py_LIMITED_API``.  This should be formalized into an official macro
+called ``PYTHON_ABI_VERSION`` to mirror ``PYTHON_API_VERSION``.  If
+and when the ABI changes in an incompatible way, this version number
+would be bumped.  To facilitate sharing, Python would be extended to
+search for extension modules with the ``PYTHON_ABI_VERSION`` number in
+its name.  The prefix ``abi`` is reserved for Python's use.
+
+Thus for example, an initial implementation of PEP 384, compiled with
+`--with-so-abi-tag=cpython-xy` would search for the following file
+names when extension module `foo` is imported (in this order)::
+
+    foo.abi3.so
+    foo.cpython-xy.so
+    foo.so
+
+The distutils [7]_ ``build_ext`` command would also have to be
+extended to compile to shared library files with the ``abi3`` tag,
+when the module author indicates that their extension supports that
+version of the ABI.  This could be done in a backward compatible way
+by adding a keyword argument to the ``Extension`` class, such as::
+
+    Extension('foo', ['foo.c'], abi=3)
+
+
+Alternatives
+============
+
+In the initial python-dev thread [8]_ where this idea was first
+introduced, several alternatives were suggested.  For completeness
+they are listed here, along with the reasons for not adopting them.
+
+
+Independent directories or symlinks
+-----------------------------------
+
+Debian and Ubuntu could simply add a version-specific directory to
+``sys.path`` that would contain just the extension modules for that
+version of Python.  Or the symlink trick eliminated in PEP 3147 could
+be retained for just shared libraries.  This approach is rejected
+because it propagates the essential complexity that PEP 3147 tries to
+avoid, and adds yet another directory to search for all modules, even
+when the number of extension modules is much fewer than the total
+number of Python packages.  It also makes for more robust management
+when all of a package's module files live in the same directory,
+because it allows systems such as `dpkg` to detect file conflicts
+between distribution packages.
+
+
+Don't share packages with extension modules
+-------------------------------------------
+
+It has been suggested that Python packages with extension modules not
+be shared among all supported Python versions on a distribution.  Even
+with adoption of PEP 3149, extension modules will have to be compiled
+for every supported Python version, so perhaps sharing of such
+packages isn't useful anyway.  Not sharing packages with extensions
+though is infeasible for several reasons.
+
+If a pure-Python package is shared in one version, should it suddenly
+be not-shared if the next release adds an extension module for speed?
+Also, even though all extension shared libraries will be compiled and
+distributed once for every supported Python, there's a big difference
+between duplicating the `.so` files and duplicating all `.py` files.
+The extra space increases the download time for such packages, and
+more immediately, increases the space pressures on already constrained
+distribution CD-ROMs.
+
+
+Reference implementation
+========================
+
+Work on this code is tracked in a Bazaar branch on Launchpad [5]_
+until it's ready for merge into Python 3.2.  The work-in-progress diff
+can also be viewed [6]_ and is updated automatically as new changes
+are uploaded.
+
+
+References
+==========
+
+.. [1] PEP 3147
+
+.. [2] PEP 384
+
+.. [3] Ubuntu: <http://www.ubuntu.com>
+
+.. [4] Debian: <http://www.debian.org>
+
+.. [5] https://code.edge.launchpad.net/~barry/python/sovers
+
+.. [6] https://code.edge.launchpad.net/~barry/python/sovers/+merge/29411
+
+.. [7] http://docs.python.org/py3k/distutils/index.html
+
+.. [8] http://mail.python.org/pipermail/python-dev/2010-June/100998.html
+
+
+Copyright
+=========
+
+This document has been placed in the public domain.
+
+
+
+..
+   Local Variables:
+   mode: indented-text
+   indent-tabs-mode: nil
+   sentence-end-double-space: t
+   fill-column: 70
+   coding: utf-8
+   End:

From python-checkins at python.org  Thu Jul 15 08:24:04 2010
From: python-checkins at python.org (brett.cannon)
Date: Thu, 15 Jul 2010 08:24:04 +0200 (CEST)
Subject: [Python-checkins] r82910 -
	python/branches/py3k/Lib/importlib/test/benchmark.py
Message-ID: <20100715062404.A169BEE989@mail.python.org>

Author: brett.cannon
Date: Thu Jul 15 08:24:04 2010
New Revision: 82910

Log:
Touch up comments and code along with outputting what the unit of measure is.


Modified:
   python/branches/py3k/Lib/importlib/test/benchmark.py

Modified: python/branches/py3k/Lib/importlib/test/benchmark.py
==============================================================================
--- python/branches/py3k/Lib/importlib/test/benchmark.py	(original)
+++ python/branches/py3k/Lib/importlib/test/benchmark.py	Thu Jul 15 08:24:04 2010
@@ -1,9 +1,10 @@
 """Benchmark some basic import use-cases."""
 # XXX
-#    - Bench from source (turn off bytecode generation)
-#    - Bench from bytecode (remove existence of source)
-#    - Bench bytecode generation
-#    - Bench extensions
+#    - from source
+#        + sys.dont_write_bytecode = True
+#        + sys.dont_write_bytecode = False
+#    - from bytecode
+#    - extensions
 from . import util
 from .source import util as source_util
 import imp
@@ -30,7 +31,7 @@
             # One execution too far
             if total_time > seconds:
                 count -= 1
-        yield count
+        yield count // seconds
 
 def from_cache(repeat):
     """sys.modules"""
@@ -49,13 +50,15 @@
     name = 'errno'
     if name in sys.modules:
         del sys.modules[name]
+    # Relying on built-in importer being implicit.
     for result in bench(name, lambda: sys.modules.pop(name), repeat=repeat):
         yield result
 
 
-def main(import_, repeat=3):
+def main(import_, *, repeat=3):
     __builtins__.__import__ = import_
     benchmarks = from_cache, builtin_mod
+    print("Measuring imports/second\n")
     for benchmark in benchmarks:
         print(benchmark.__doc__, "[", end=' ')
         sys.stdout.flush()
@@ -75,7 +78,7 @@
                         default=False, help="use the built-in __import__")
     options, args = parser.parse_args()
     if args:
-        raise RuntimeError("unrecognized args: {0}".format(args))
+        raise RuntimeError("unrecognized args: {}".format(args))
     import_ = __import__
     if not options.builtin:
         import_ = importlib.__import__

From python-checkins at python.org  Thu Jul 15 18:35:58 2010
From: python-checkins at python.org (jean-paul.calderone)
Date: Thu, 15 Jul 2010 18:35:58 +0200 (CEST)
Subject: [Python-checkins] r82911 - python/branches/signalfd-issue8407
Message-ID: <20100715163558.57E1FEE981@mail.python.org>

Author: jean-paul.calderone
Date: Thu Jul 15 18:35:58 2010
New Revision: 82911

Log:
this can never be merged; see py3k-signalfd-issue8407

Removed:
   python/branches/signalfd-issue8407/

From python-checkins at python.org  Thu Jul 15 23:40:14 2010
From: python-checkins at python.org (vinay.sajip)
Date: Thu, 15 Jul 2010 23:40:14 +0200 (CEST)
Subject: [Python-checkins] r82912 -
	python/branches/release31-maint/Lib/logging/handlers.py
Message-ID: <20100715214014.11B40F42B@mail.python.org>

Author: vinay.sajip
Date: Thu Jul 15 23:40:13 2010
New Revision: 82912

Log:
Backported SysLogHandler fix for issue #7077.

Modified:
   python/branches/release31-maint/Lib/logging/handlers.py

Modified: python/branches/release31-maint/Lib/logging/handlers.py
==============================================================================
--- python/branches/release31-maint/Lib/logging/handlers.py	(original)
+++ python/branches/release31-maint/Lib/logging/handlers.py	Thu Jul 15 23:40:13 2010
@@ -202,7 +202,11 @@
 
         self.extMatch = re.compile(self.extMatch, re.ASCII)
         self.interval = self.interval * interval # multiply by units requested
-        self.rolloverAt = self.computeRollover(int(time.time()))
+        if os.path.exists(filename):
+            t = os.stat(filename)[ST_MTIME]
+        else:
+            t = int(time.time())
+        self.rolloverAt = self.computeRollover(t)
 
     def computeRollover(self, currentTime):
         """
@@ -774,6 +778,9 @@
             self.encodePriority(self.facility,
                                 self.mapPriority(record.levelname)),
                                 msg)
+        msg = msg.encode('utf-8')
+        if codecs:
+            msg = codecs.BOM_UTF8 + msg
         try:
             if self.unixsocket:
                 try:

From python-checkins at python.org  Fri Jul 16 01:23:38 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 16 Jul 2010 01:23:38 +0200 (CEST)
Subject: [Python-checkins] r82913 -
	python/branches/import_unicode/Modules/zipimport.c
Message-ID: <20100715232338.99FB3DBF3@mail.python.org>

Author: victor.stinner
Date: Fri Jul 16 01:23:38 2010
New Revision: 82913

Log:
zipimport: import surrogates support

Modified:
   python/branches/import_unicode/Modules/zipimport.c

Modified: python/branches/import_unicode/Modules/zipimport.c
==============================================================================
--- python/branches/import_unicode/Modules/zipimport.c	(original)
+++ python/branches/import_unicode/Modules/zipimport.c	Fri Jul 16 01:23:38 2010
@@ -418,20 +418,26 @@
 zipimporter_get_data(PyObject *obj, PyObject *args)
 {
     ZipImporter *self = (ZipImporter *)obj;
-    char *path;
+    PyObject *pathbytes;
+    /* FIXME: use Py_UNICODE* instead of char* */
+    const char *path;
 #ifdef ALTSEP
     char *p, buf[MAXPATHLEN + 1];
 #endif
     PyObject *toc_entry;
     Py_ssize_t len;
+    PyObject *archive_bytes;
     char *archive_str;
 
-    if (!PyArg_ParseTuple(args, "s:zipimporter.get_data", &path))
+    if (!PyArg_ParseTuple(args, "O&:zipimporter.get_data",
+                          PyUnicode_FSConverter, &pathbytes))
         return NULL;
 
+    path = PyBytes_AsString(pathbytes);
 #ifdef ALTSEP
     if (strlen(path) >= MAXPATHLEN) {
         PyErr_SetString(ZipImportError, "path too long");
+        Py_DECREF(pathbytes);
         return NULL;
     }
     strcpy(buf, path);
@@ -441,18 +447,23 @@
     }
     path = buf;
 #endif
-    archive_str = _PyUnicode_AsStringAndSize(self->archive, &len);
+    archive_bytes = PyUnicode_EncodeFSDefault(self->archive);
+    archive_str = PyBytes_AS_STRING(archive_bytes);
+    len = PyBytes_GET_SIZE(archive_bytes);
     if ((size_t)len < strlen(path) &&
         strncmp(path, archive_str, len) == 0 &&
         path[len] == SEP) {
         path = path + len + 1;
     }
+    Py_DECREF(archive_bytes);
 
     toc_entry = PyDict_GetItemString(self->files, path);
     if (toc_entry == NULL) {
         PyErr_SetFromErrnoWithFilename(PyExc_IOError, path);
+        Py_DECREF(pathbytes);
         return NULL;
     }
+    Py_DECREF(pathbytes);
     return get_data(self->archive, toc_entry);
 }
 
@@ -703,6 +714,7 @@
     char *p, endof_central_dir[22];
     long arc_offset; /* offset from beginning of file to start of zip-archive */
     PyObject *archive_bytes;
+    PyObject *pathobj;
 
     archive_bytes = PyUnicode_EncodeFSDefault(archive_obj);
     if (archive_bytes == NULL)
@@ -789,7 +801,10 @@
 
         strncpy(path + length + 1, name, MAXPATHLEN - length - 1);
 
-        t = Py_BuildValue("siiiiiii", path, compress, data_size,
+        pathobj = PyUnicode_DecodeFSDefault(path);
+        if (pathobj == NULL)
+            goto error;
+        t = Py_BuildValue("Niiiiiii", pathobj, compress, data_size,
                           file_size, file_offset, time, date, crc);
         if (t == NULL)
             goto error;
@@ -855,15 +870,17 @@
     int err;
     Py_ssize_t bytes_read = 0;
     long l;
-    char *datapath;
+    PyObject *datapath;
     long compress, data_size, file_size, file_offset, bytes_size;
     long time, date, crc;
 
-    if (!PyArg_ParseTuple(toc_entry, "slllllll", &datapath, &compress,
+    if (!PyArg_ParseTuple(toc_entry, "O&lllllll",
+                          PyUnicode_FSConverter, &datapath, &compress,
                           &data_size, &file_size, &file_offset, &time,
                           &date, &crc)) {
         return NULL;
     }
+    Py_DECREF(datapath);
 
     fp = _Py_fopen(archive, "rb");
     if (!fp) {

From solipsis at pitrou.net  Fri Jul 16 01:25:09 2010
From: solipsis at pitrou.net (solipsis at pitrou.net)
Date: Fri, 16 Jul 2010 01:25:09 +0200 (CEST)
Subject: [Python-checkins] Daily py3k reference leaks (r82910): sum=0
Message-ID: <20100715232509.9F5D31775D@ns6635.ovh.net>

py3k results for svn r82910 (hg cset 618e27a6c8fd)
--------------------------------------------------



Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/antoine/py3k/refleaks/reflogvPW9Jg', '-x']

From python-checkins at python.org  Fri Jul 16 09:53:23 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Fri, 16 Jul 2010 09:53:23 +0200 (CEST)
Subject: [Python-checkins] r82914 - peps/trunk/pep-0381.txt
Message-ID: <20100716075323.D5D5DEEA00@mail.python.org>

Author: martin.v.loewis
Date: Fri Jul 16 09:53:23 2010
New Revision: 82914

Log:
Document signing process.


Modified:
   peps/trunk/pep-0381.txt

Modified: peps/trunk/pep-0381.txt
==============================================================================
--- peps/trunk/pep-0381.txt	(original)
+++ peps/trunk/pep-0381.txt	Fri Jul 16 09:53:23 2010
@@ -83,6 +83,59 @@
 - /stats/months/2008-11.bz2
 - /stats/months/2008-10.bz2
 
+Mirror Authenticity
+===================
+
+With a distributed mirroring system, clients may want to verify that
+the mirrored copies are authentic. There are multiple threats to
+consider:
+
+1. the central index may get compromised
+2. the central index is assumed to be trusted, but the mirrors might
+   be tampered.
+3. a man in the middle between the central index and the end user,
+   or between a mirror and the end user might tamper with datagrams.
+
+This specification only deals with the second threat. Some provisions
+are made to detect man-in-the-middle attacks. To detect the first
+attack, package authors need to sign their packages using PGP keys, so
+that users verify that the package comes from the author they trust.
+
+The central index provides a DSA key at the URL /serverkey, in the PEM
+format as generated by "openssl dsa -pubout" (i.e. RFC 3280
+SubjectPublicKeyInfo, with the algorithm 1.3.14.3.2.12). This URL must
+*not* be mirrored, and clients must fetch the official serverkey from
+PyPI directly, or use the copy that came with the PyPI client software.
+Mirrors should still download the key, to detect a key rollover.
+
+For each package, a mirrored signature is provided at
+/serversig/<package>. This is the DSA signature of the parallel URL
+/simple/<package>, in DER form, using SHA-1 with DSA (i.e. as a RFC
+3279 Dsa-Sig-Value, created by algorithm 1.2.840.10040.4.3)
+
+Clients using a mirror need to perform the following steps to verify
+a package:
+
+1. download the /simple page, and compute its SHA-1 hash
+2. compute the DSA signature of that hash
+3. download the corresponding /serversig, and compare it (byte-for-byte)
+   with the value computed in step 2.
+4. compute and verify (against the /simple page) the MD-5 hashes
+   of all files they download from the mirror.
+
+An implementation of the verification algorithm is available from
+https://svn.python.org/packages/trunk/pypi/tools/verify.py
+
+Verification is not needed when downloading from central index, and
+should be avoided to reduce the computation overhead.
+
+About once a year, the key will be replaced with a new one. Mirrors
+will have to re-fetch all /serversig pages. Clients using mirrors need
+to find a trusted copy of the new server key. One way to obtain one
+is to download it from https://pypi.python.org/serverkey. To detect
+man-in-the-middle attacks, clients need to verify the SSL server
+certificate, which will be signed by the CACert authority.
+
 
 Special pages a mirror needs to provide
 =======================================

From python-checkins at python.org  Fri Jul 16 16:39:45 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Fri, 16 Jul 2010 16:39:45 +0200 (CEST)
Subject: [Python-checkins] r82915 - python/branches/py3k/Lib/pickletools.py
Message-ID: <20100716143945.C929FEE9AB@mail.python.org>

Author: alexander.belopolsky
Date: Fri Jul 16 16:39:45 2010
New Revision: 82915

Log:
Corrected TUPLE<N> opcodes' docs.

Modified:
   python/branches/py3k/Lib/pickletools.py

Modified: python/branches/py3k/Lib/pickletools.py
==============================================================================
--- python/branches/py3k/Lib/pickletools.py	(original)
+++ python/branches/py3k/Lib/pickletools.py	Fri Jul 16 16:39:45 2010
@@ -1250,10 +1250,11 @@
       stack_before=[anyobject],
       stack_after=[pytuple],
       proto=2,
-      doc="""One-tuple.
+      doc="""Build a one-tuple out of the topmost item on the stack.
 
       This code pops one value off the stack and pushes a tuple of
-      length 1 whose one item is that value back onto it.  IOW:
+      length 1 whose one item is that value back onto it.  In other
+      words:
 
           stack[-1] = tuple(stack[-1:])
       """),
@@ -1264,10 +1265,11 @@
       stack_before=[anyobject, anyobject],
       stack_after=[pytuple],
       proto=2,
-      doc="""One-tuple.
+      doc="""Build a two-tuple out of the top two items on the stack.
 
-      This code pops two values off the stack and pushes a tuple
-      of length 2 whose items are those values back onto it.  IOW:
+      This code pops two values off the stack and pushes a tuple of
+      length 2 whose items are those values back onto it.  In other
+      words:
 
           stack[-2:] = [tuple(stack[-2:])]
       """),
@@ -1278,10 +1280,11 @@
       stack_before=[anyobject, anyobject, anyobject],
       stack_after=[pytuple],
       proto=2,
-      doc="""One-tuple.
+      doc="""Build a three-tuple out of the top three items on the stack.
 
-      This code pops three values off the stack and pushes a tuple
-      of length 3 whose items are those values back onto it.  IOW:
+      This code pops three values off the stack and pushes a tuple of
+      length 3 whose items are those values back onto it.  In other
+      words:
 
           stack[-3:] = [tuple(stack[-3:])]
       """),

From python-checkins at python.org  Fri Jul 16 16:52:25 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Fri, 16 Jul 2010 16:52:25 +0200 (CEST)
Subject: [Python-checkins] r82916 - in python/branches/release31-maint:
	Lib/pickletools.py
Message-ID: <20100716145225.DD8B3C9E3@mail.python.org>

Author: alexander.belopolsky
Date: Fri Jul 16 16:52:25 2010
New Revision: 82916

Log:
Merged revisions 82915 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82915 | alexander.belopolsky | 2010-07-16 10:39:45 -0400 (Fri, 16 Jul 2010) | 1 line
  
  Corrected TUPLE<N> opcodes' docs.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/pickletools.py

Modified: python/branches/release31-maint/Lib/pickletools.py
==============================================================================
--- python/branches/release31-maint/Lib/pickletools.py	(original)
+++ python/branches/release31-maint/Lib/pickletools.py	Fri Jul 16 16:52:25 2010
@@ -1250,10 +1250,11 @@
       stack_before=[anyobject],
       stack_after=[pytuple],
       proto=2,
-      doc="""One-tuple.
+      doc="""Build a one-tuple out of the topmost item on the stack.
 
       This code pops one value off the stack and pushes a tuple of
-      length 1 whose one item is that value back onto it.  IOW:
+      length 1 whose one item is that value back onto it.  In other
+      words:
 
           stack[-1] = tuple(stack[-1:])
       """),
@@ -1264,10 +1265,11 @@
       stack_before=[anyobject, anyobject],
       stack_after=[pytuple],
       proto=2,
-      doc="""One-tuple.
+      doc="""Build a two-tuple out of the top two items on the stack.
 
-      This code pops two values off the stack and pushes a tuple
-      of length 2 whose items are those values back onto it.  IOW:
+      This code pops two values off the stack and pushes a tuple of
+      length 2 whose items are those values back onto it.  In other
+      words:
 
           stack[-2:] = [tuple(stack[-2:])]
       """),
@@ -1278,10 +1280,11 @@
       stack_before=[anyobject, anyobject, anyobject],
       stack_after=[pytuple],
       proto=2,
-      doc="""One-tuple.
+      doc="""Build a three-tuple out of the top three items on the stack.
 
-      This code pops three values off the stack and pushes a tuple
-      of length 3 whose items are those values back onto it.  IOW:
+      This code pops three values off the stack and pushes a tuple of
+      length 3 whose items are those values back onto it.  In other
+      words:
 
           stack[-3:] = [tuple(stack[-3:])]
       """),

From python-checkins at python.org  Fri Jul 16 16:53:52 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Fri, 16 Jul 2010 16:53:52 +0200 (CEST)
Subject: [Python-checkins] r82917 - in python/branches/release27-maint:
	Lib/pickletools.py
Message-ID: <20100716145352.363FCD6B9@mail.python.org>

Author: alexander.belopolsky
Date: Fri Jul 16 16:53:52 2010
New Revision: 82917

Log:
Merged revisions 82915 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82915 | alexander.belopolsky | 2010-07-16 10:39:45 -0400 (Fri, 16 Jul 2010) | 1 line
  
  Corrected TUPLE<N> opcodes' docs.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/pickletools.py

Modified: python/branches/release27-maint/Lib/pickletools.py
==============================================================================
--- python/branches/release27-maint/Lib/pickletools.py	(original)
+++ python/branches/release27-maint/Lib/pickletools.py	Fri Jul 16 16:53:52 2010
@@ -1214,10 +1214,11 @@
       stack_before=[anyobject],
       stack_after=[pytuple],
       proto=2,
-      doc="""One-tuple.
+      doc="""Build a one-tuple out of the topmost item on the stack.
 
       This code pops one value off the stack and pushes a tuple of
-      length 1 whose one item is that value back onto it.  IOW:
+      length 1 whose one item is that value back onto it.  In other
+      words:
 
           stack[-1] = tuple(stack[-1:])
       """),
@@ -1228,10 +1229,11 @@
       stack_before=[anyobject, anyobject],
       stack_after=[pytuple],
       proto=2,
-      doc="""One-tuple.
+      doc="""Build a two-tuple out of the top two items on the stack.
 
-      This code pops two values off the stack and pushes a tuple
-      of length 2 whose items are those values back onto it.  IOW:
+      This code pops two values off the stack and pushes a tuple of
+      length 2 whose items are those values back onto it.  In other
+      words:
 
           stack[-2:] = [tuple(stack[-2:])]
       """),
@@ -1242,10 +1244,11 @@
       stack_before=[anyobject, anyobject, anyobject],
       stack_after=[pytuple],
       proto=2,
-      doc="""One-tuple.
+      doc="""Build a three-tuple out of the top three items on the stack.
 
-      This code pops three values off the stack and pushes a tuple
-      of length 3 whose items are those values back onto it.  IOW:
+      This code pops three values off the stack and pushes a tuple of
+      length 3 whose items are those values back onto it.  In other
+      words:
 
           stack[-3:] = [tuple(stack[-3:])]
       """),

From python-checkins at python.org  Fri Jul 16 21:04:29 2010
From: python-checkins at python.org (brett.cannon)
Date: Fri, 16 Jul 2010 21:04:29 +0200 (CEST)
Subject: [Python-checkins] r82918 -
	python/branches/py3k/Lib/importlib/test/benchmark.py
Message-ID: <20100716190429.966B4EC35@mail.python.org>

Author: brett.cannon
Date: Fri Jul 16 21:04:29 2010
New Revision: 82918

Log:
Add benchmarks for importing just source w/o writing bytecode, importing source
while writing bytecode, and importing bytecode with source existing (don't care
about sourceless imports).


Modified:
   python/branches/py3k/Lib/importlib/test/benchmark.py

Modified: python/branches/py3k/Lib/importlib/test/benchmark.py
==============================================================================
--- python/branches/py3k/Lib/importlib/test/benchmark.py	(original)
+++ python/branches/py3k/Lib/importlib/test/benchmark.py	Fri Jul 16 21:04:29 2010
@@ -1,14 +1,15 @@
-"""Benchmark some basic import use-cases."""
-# XXX
-#    - from source
-#        + sys.dont_write_bytecode = True
-#        + sys.dont_write_bytecode = False
-#    - from bytecode
-#    - extensions
+"""Benchmark some basic import use-cases.
+
+The assumption is made that this benchmark is run in a fresh interpreter and
+thus has no external changes made to import-related attributes in sys.
+
+"""
 from . import util
 from .source import util as source_util
 import imp
 import importlib
+import os
+import py_compile
 import sys
 import timeit
 
@@ -33,7 +34,7 @@
                 count -= 1
         yield count // seconds
 
-def from_cache(repeat):
+def from_cache(seconds, repeat):
     """sys.modules"""
     name = '<benchmark import>'
     module = imp.new_module(name)
@@ -41,32 +42,74 @@
     module.__package__ = ''
     with util.uncache(name):
         sys.modules[name] = module
-        for result in bench(name, repeat=repeat):
+        for result in bench(name, repeat=repeat, seconds=seconds):
             yield result
 
 
-def builtin_mod(repeat):
+def builtin_mod(seconds, repeat):
     """Built-in module"""
     name = 'errno'
     if name in sys.modules:
         del sys.modules[name]
     # Relying on built-in importer being implicit.
-    for result in bench(name, lambda: sys.modules.pop(name), repeat=repeat):
+    for result in bench(name, lambda: sys.modules.pop(name), repeat=repeat,
+                        seconds=seconds):
         yield result
 
 
-def main(import_, *, repeat=3):
+def source_wo_bytecode(seconds, repeat):
+    """Source w/o bytecode"""
+    sys.dont_write_bytecode = True
+    try:
+        name = '__importlib_test_benchmark__'
+        # Clears out sys.modules and puts an entry at the front of sys.path.
+        with source_util.create_modules(name) as mapping:
+            assert not os.path.exists(imp.cache_from_source(mapping[name]))
+            for result in bench(name, lambda: sys.modules.pop(name), repeat=repeat,
+                                seconds=seconds):
+                yield result
+    finally:
+        sys.dont_write_bytecode = False
+
+
+def source_writing_bytecode(seconds, repeat):
+    """Source writing bytecode"""
+    assert not sys.dont_write_bytecode
+    name = '__importlib_test_benchmark__'
+    with source_util.create_modules(name) as mapping:
+        def cleanup():
+            sys.modules.pop(name)
+            os.unlink(imp.cache_from_source(mapping[name]))
+        for result in bench(name, cleanup, repeat=repeat, seconds=seconds):
+            assert not os.path.exists(imp.cache_from_source(mapping[name]))
+            yield result
+
+
+def source_using_bytecode(seconds, repeat):
+    """Bytecode w/ source"""
+    name = '__importlib_test_benchmark__'
+    with source_util.create_modules(name) as mapping:
+        py_compile.compile(mapping[name])
+        assert os.path.exists(imp.cache_from_source(mapping[name]))
+        for result in bench(name, lambda: sys.modules.pop(name), repeat=repeat,
+                            seconds=seconds):
+            yield result
+
+
+def main(import_):
     __builtins__.__import__ = import_
-    benchmarks = from_cache, builtin_mod
+    benchmarks = (from_cache, builtin_mod, source_using_bytecode,
+                  source_wo_bytecode, source_writing_bytecode,)
     print("Measuring imports/second\n")
     for benchmark in benchmarks:
         print(benchmark.__doc__, "[", end=' ')
         sys.stdout.flush()
         results = []
-        for result in benchmark(repeat):
+        for result in benchmark(seconds=1, repeat=3):
             results.append(result)
             print(result, end=' ')
             sys.stdout.flush()
+        assert not sys.dont_write_bytecode
         print("]", "best is", max(results))
 
 

From python-checkins at python.org  Fri Jul 16 21:10:38 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Fri, 16 Jul 2010 21:10:38 +0200 (CEST)
Subject: [Python-checkins] r82919 -
	python/branches/py3k/Lib/test/test_threaded_import.py
Message-ID: <20100716191038.D55E0F1ED@mail.python.org>

Author: antoine.pitrou
Date: Fri Jul 16 21:10:38 2010
New Revision: 82919

Log:
Fix possible failure in pickling tests due to different instantiations
of the random module being around.



Modified:
   python/branches/py3k/Lib/test/test_threaded_import.py

Modified: python/branches/py3k/Lib/test/test_threaded_import.py
==============================================================================
--- python/branches/py3k/Lib/test/test_threaded_import.py	(original)
+++ python/branches/py3k/Lib/test/test_threaded_import.py	Fri Jul 16 21:10:38 2010
@@ -27,6 +27,16 @@
 
 class ThreadedImportTests(unittest.TestCase):
 
+    def setUp(self):
+        self.old_random = sys.modules.pop('random', None)
+
+    def tearDown(self):
+        # If the `random` module was already initialized, we restore the
+        # old module at the end so that pickling tests don't fail.
+        # See http://bugs.python.org/issue3657#msg110461
+        if self.old_random is not None:
+            sys.modules['random'] = self.old_random
+
     def test_parallel_module_init(self):
         if imp.lock_held():
             # This triggers on, e.g., from test import autotest.

From python-checkins at python.org  Fri Jul 16 21:11:23 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Fri, 16 Jul 2010 21:11:23 +0200 (CEST)
Subject: [Python-checkins] r82920 - in python/branches/release31-maint:
	Lib/test/test_threaded_import.py
Message-ID: <20100716191123.6C105F3BA@mail.python.org>

Author: antoine.pitrou
Date: Fri Jul 16 21:11:23 2010
New Revision: 82920

Log:
Merged revisions 82919 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82919 | antoine.pitrou | 2010-07-16 21:10:38 +0200 (ven., 16 juil. 2010) | 4 lines
  
  Fix possible failure in pickling tests due to different instantiations
  of the random module being around.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/test/test_threaded_import.py

Modified: python/branches/release31-maint/Lib/test/test_threaded_import.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_threaded_import.py	(original)
+++ python/branches/release31-maint/Lib/test/test_threaded_import.py	Fri Jul 16 21:11:23 2010
@@ -27,6 +27,16 @@
 
 class ThreadedImportTests(unittest.TestCase):
 
+    def setUp(self):
+        self.old_random = sys.modules.pop('random', None)
+
+    def tearDown(self):
+        # If the `random` module was already initialized, we restore the
+        # old module at the end so that pickling tests don't fail.
+        # See http://bugs.python.org/issue3657#msg110461
+        if self.old_random is not None:
+            sys.modules['random'] = self.old_random
+
     def test_parallel_module_init(self):
         if imp.lock_held():
             # This triggers on, e.g., from test import autotest.

From python-checkins at python.org  Fri Jul 16 21:26:23 2010
From: python-checkins at python.org (brett.cannon)
Date: Fri, 16 Jul 2010 21:26:23 +0200 (CEST)
Subject: [Python-checkins] r82921 -
	python/branches/py3k/Lib/importlib/test/benchmark.py
Message-ID: <20100716192623.CF18FEE9A1@mail.python.org>

Author: brett.cannon
Date: Fri Jul 16 21:26:23 2010
New Revision: 82921

Log:
Add comma grouping to max result so it's easier to read.

Modified:
   python/branches/py3k/Lib/importlib/test/benchmark.py

Modified: python/branches/py3k/Lib/importlib/test/benchmark.py
==============================================================================
--- python/branches/py3k/Lib/importlib/test/benchmark.py	(original)
+++ python/branches/py3k/Lib/importlib/test/benchmark.py	Fri Jul 16 21:26:23 2010
@@ -110,7 +110,7 @@
             print(result, end=' ')
             sys.stdout.flush()
         assert not sys.dont_write_bytecode
-        print("]", "best is", max(results))
+        print("]", "best is", format(max(results), ',d'))
 
 
 if __name__ == '__main__':

From solipsis at pitrou.net  Sat Jul 17 01:24:11 2010
From: solipsis at pitrou.net (solipsis at pitrou.net)
Date: Sat, 17 Jul 2010 01:24:11 +0200 (CEST)
Subject: [Python-checkins] Daily py3k reference leaks (r82921): sum=0
Message-ID: <20100716232411.702921770A@ns6635.ovh.net>

py3k results for svn r82921 (hg cset 32f7cc1aa27b)
--------------------------------------------------



Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/antoine/py3k/refleaks/reflogepYJ7N', '-x']

From python-checkins at python.org  Sat Jul 17 03:19:57 2010
From: python-checkins at python.org (r.david.murray)
Date: Sat, 17 Jul 2010 03:19:57 +0200 (CEST)
Subject: [Python-checkins] r82922 - in python/branches/py3k:
	Lib/email/feedparser.py Lib/email/test/test_email.py Misc/NEWS
Message-ID: <20100717011957.CE216EE983@mail.python.org>

Author: r.david.murray
Date: Sat Jul 17 03:19:57 2010
New Revision: 82922

Log:
#1555570: correctly handle a \r\n that is split by the read buffer.

Patch and test by Tony Nelson.


Modified:
   python/branches/py3k/Lib/email/feedparser.py
   python/branches/py3k/Lib/email/test/test_email.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/email/feedparser.py
==============================================================================
--- python/branches/py3k/Lib/email/feedparser.py	(original)
+++ python/branches/py3k/Lib/email/feedparser.py	Sat Jul 17 03:19:57 2010
@@ -104,6 +104,10 @@
         # data after the final RE.  In the case of a NL/CR terminated string,
         # this is the empty string.
         self._partial = parts.pop()
+        #GAN 29Mar09  bugs 1555570, 1721862  Confusion at 8K boundary ending with \r:
+        # is there a \n to follow later?
+        if not self._partial and parts and parts[-1].endswith('\r'):
+            self._partial = parts.pop(-2)+parts.pop()
         # parts is a list of strings, alternating between the line contents
         # and the eol character(s).  Gather up a list of lines after
         # re-attaching the newlines.

Modified: python/branches/py3k/Lib/email/test/test_email.py
==============================================================================
--- python/branches/py3k/Lib/email/test/test_email.py	(original)
+++ python/branches/py3k/Lib/email/test/test_email.py	Sat Jul 17 03:19:57 2010
@@ -2454,6 +2454,39 @@
 -Me
 """)
 
+    def test_pushCR_LF(self):
+        '''FeedParser BufferedSubFile.push() assumed it received complete
+           line endings.  A CR ending one push() followed by a LF starting
+           the next push() added an empty line.
+        '''
+        imt = [
+            ("a\r \n",  2),
+            ("b",       0),
+            ("c\n",     1),
+            ("",        0),
+            ("d\r\n",   1),
+            ("e\r",     0),
+            ("\nf",     1),
+            ("\r\n",    1),
+          ]
+        from email.feedparser import BufferedSubFile, NeedMoreData
+        bsf = BufferedSubFile()
+        om = []
+        nt = 0
+        for il, n in imt:
+            bsf.push(il)
+            nt += n
+            n1 = 0
+            while True:
+                ol = bsf.readline()
+                if ol == NeedMoreData:
+                    break
+                om.append(ol)
+                n1 += 1
+            self.assertTrue(n == n1)
+        self.assertTrue(len(om) == nt)
+        self.assertTrue(''.join([il for il, n in imt]) == ''.join(om))
+
 
 
 class TestParsers(TestEmailBase):

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sat Jul 17 03:19:57 2010
@@ -473,6 +473,9 @@
 Library
 -------
 
+- Issue #1555570: email no longer inserts extra blank lines when a \r\n
+  combo crosses an 8192 byte boundary.
+
 - Issue #9243: Fix sndhdr module and add unit tests, contributed by James Lee.
 
 - ``ast.literal_eval()`` now allows byte literals.

From python-checkins at python.org  Sat Jul 17 03:28:05 2010
From: python-checkins at python.org (r.david.murray)
Date: Sat, 17 Jul 2010 03:28:05 +0200 (CEST)
Subject: [Python-checkins] r82923 - in python/branches/release31-maint:
	Lib/email/feedparser.py Lib/email/test/test_email.py Misc/NEWS
Message-ID: <20100717012805.1D4C7EE987@mail.python.org>

Author: r.david.murray
Date: Sat Jul 17 03:28:04 2010
New Revision: 82923

Log:
Merged revisions 82922 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82922 | r.david.murray | 2010-07-16 21:19:57 -0400 (Fri, 16 Jul 2010) | 4 lines
  
  #1555570: correctly handle a \r\n that is split by the read buffer.
  
  Patch and test by Tony Nelson.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/email/feedparser.py
   python/branches/release31-maint/Lib/email/test/test_email.py
   python/branches/release31-maint/Misc/NEWS

Modified: python/branches/release31-maint/Lib/email/feedparser.py
==============================================================================
--- python/branches/release31-maint/Lib/email/feedparser.py	(original)
+++ python/branches/release31-maint/Lib/email/feedparser.py	Sat Jul 17 03:28:04 2010
@@ -104,6 +104,10 @@
         # data after the final RE.  In the case of a NL/CR terminated string,
         # this is the empty string.
         self._partial = parts.pop()
+        #GAN 29Mar09  bugs 1555570, 1721862  Confusion at 8K boundary ending with \r:
+        # is there a \n to follow later?
+        if not self._partial and parts and parts[-1].endswith('\r'):
+            self._partial = parts.pop(-2)+parts.pop()
         # parts is a list of strings, alternating between the line contents
         # and the eol character(s).  Gather up a list of lines after
         # re-attaching the newlines.

Modified: python/branches/release31-maint/Lib/email/test/test_email.py
==============================================================================
--- python/branches/release31-maint/Lib/email/test/test_email.py	(original)
+++ python/branches/release31-maint/Lib/email/test/test_email.py	Sat Jul 17 03:28:04 2010
@@ -2450,6 +2450,39 @@
 -Me
 """)
 
+    def test_pushCR_LF(self):
+        '''FeedParser BufferedSubFile.push() assumed it received complete
+           line endings.  A CR ending one push() followed by a LF starting
+           the next push() added an empty line.
+        '''
+        imt = [
+            ("a\r \n",  2),
+            ("b",       0),
+            ("c\n",     1),
+            ("",        0),
+            ("d\r\n",   1),
+            ("e\r",     0),
+            ("\nf",     1),
+            ("\r\n",    1),
+          ]
+        from email.feedparser import BufferedSubFile, NeedMoreData
+        bsf = BufferedSubFile()
+        om = []
+        nt = 0
+        for il, n in imt:
+            bsf.push(il)
+            nt += n
+            n1 = 0
+            while True:
+                ol = bsf.readline()
+                if ol == NeedMoreData:
+                    break
+                om.append(ol)
+                n1 += 1
+            self.assertTrue(n == n1)
+        self.assertTrue(len(om) == nt)
+        self.assertTrue(''.join([il for il, n in imt]) == ''.join(om))
+
 
 
 class TestParsers(TestEmailBase):

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Sat Jul 17 03:28:04 2010
@@ -78,6 +78,9 @@
 Library
 -------
 
+- Issue #1555570: email no longer inserts extra blank lines when a \r\n
+  combo crosses an 8192 byte boundary.
+
 - Issue #9243: Fix sndhdr module and add unit tests, contributed by James Lee.
 
 - ``ast.literal_eval()`` now allows byte literals.

From python-checkins at python.org  Sat Jul 17 03:35:16 2010
From: python-checkins at python.org (r.david.murray)
Date: Sat, 17 Jul 2010 03:35:16 +0200 (CEST)
Subject: [Python-checkins] r82924 - in python/branches/release27-maint:
	Lib/email/feedparser.py Lib/email/test/test_email.py Misc/NEWS
Message-ID: <20100717013516.9FD84EE98D@mail.python.org>

Author: r.david.murray
Date: Sat Jul 17 03:35:16 2010
New Revision: 82924

Log:
Merged revisions 82922 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82922 | r.david.murray | 2010-07-16 21:19:57 -0400 (Fri, 16 Jul 2010) | 4 lines
  
  #1555570: correctly handle a \r\n that is split by the read buffer.
  
  Patch and test by Tony Nelson.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/email/feedparser.py
   python/branches/release27-maint/Lib/email/test/test_email.py
   python/branches/release27-maint/Misc/NEWS

Modified: python/branches/release27-maint/Lib/email/feedparser.py
==============================================================================
--- python/branches/release27-maint/Lib/email/feedparser.py	(original)
+++ python/branches/release27-maint/Lib/email/feedparser.py	Sat Jul 17 03:35:16 2010
@@ -104,6 +104,10 @@
         # data after the final RE.  In the case of a NL/CR terminated string,
         # this is the empty string.
         self._partial = parts.pop()
+        #GAN 29Mar09  bugs 1555570, 1721862  Confusion at 8K boundary ending with \r:
+        # is there a \n to follow later?
+        if not self._partial and parts and parts[-1].endswith('\r'):
+            self._partial = parts.pop(-2)+parts.pop()
         # parts is a list of strings, alternating between the line contents
         # and the eol character(s).  Gather up a list of lines after
         # re-attaching the newlines.

Modified: python/branches/release27-maint/Lib/email/test/test_email.py
==============================================================================
--- python/branches/release27-maint/Lib/email/test/test_email.py	(original)
+++ python/branches/release27-maint/Lib/email/test/test_email.py	Sat Jul 17 03:35:16 2010
@@ -2465,6 +2465,39 @@
 -Me
 """)
 
+    def test_pushCR_LF(self):
+        '''FeedParser BufferedSubFile.push() assumed it received complete
+           line endings.  A CR ending one push() followed by a LF starting
+           the next push() added an empty line.
+        '''
+        imt = [
+            ("a\r \n",  2),
+            ("b",       0),
+            ("c\n",     1),
+            ("",        0),
+            ("d\r\n",   1),
+            ("e\r",     0),
+            ("\nf",     1),
+            ("\r\n",    1),
+          ]
+        from email.feedparser import BufferedSubFile, NeedMoreData
+        bsf = BufferedSubFile()
+        om = []
+        nt = 0
+        for il, n in imt:
+            bsf.push(il)
+            nt += n
+            n1 = 0
+            while True:
+                ol = bsf.readline()
+                if ol == NeedMoreData:
+                    break
+                om.append(ol)
+                n1 += 1
+            self.assertTrue(n == n1)
+        self.assertTrue(len(om) == nt)
+        self.assertTrue(''.join([il for il, n in imt]) == ''.join(om))
+
 
 
 class TestParsers(TestEmailBase):

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Sat Jul 17 03:35:16 2010
@@ -833,6 +833,8 @@
 
 Library
 -------
+- Issue #1555570: email no longer inserts extra blank lines when a \r\n
+  combo crosses an 8192 byte boundary.
 
 - Issue #6906: Tk should not set Unicode environment variables on Windows.
 

From python-checkins at python.org  Sat Jul 17 03:40:30 2010
From: python-checkins at python.org (r.david.murray)
Date: Sat, 17 Jul 2010 03:40:30 +0200 (CEST)
Subject: [Python-checkins] r82925 - in python/branches/release26-maint:
	Lib/email/feedparser.py Lib/email/test/test_email.py Misc/NEWS
Message-ID: <20100717014030.903B8EE98F@mail.python.org>

Author: r.david.murray
Date: Sat Jul 17 03:40:30 2010
New Revision: 82925

Log:
Merged revisions 82924 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

................
  r82924 | r.david.murray | 2010-07-16 21:35:16 -0400 (Fri, 16 Jul 2010) | 11 lines
  
  Merged revisions 82922 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k
  
  ........
    r82922 | r.david.murray | 2010-07-16 21:19:57 -0400 (Fri, 16 Jul 2010) | 4 lines
    
    #1555570: correctly handle a \r\n that is split by the read buffer.
    
    Patch and test by Tony Nelson.
  ........
................


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Lib/email/feedparser.py
   python/branches/release26-maint/Lib/email/test/test_email.py
   python/branches/release26-maint/Misc/NEWS

Modified: python/branches/release26-maint/Lib/email/feedparser.py
==============================================================================
--- python/branches/release26-maint/Lib/email/feedparser.py	(original)
+++ python/branches/release26-maint/Lib/email/feedparser.py	Sat Jul 17 03:40:30 2010
@@ -104,6 +104,10 @@
         # data after the final RE.  In the case of a NL/CR terminated string,
         # this is the empty string.
         self._partial = parts.pop()
+        #GAN 29Mar09  bugs 1555570, 1721862  Confusion at 8K boundary ending with \r:
+        # is there a \n to follow later?
+        if not self._partial and parts and parts[-1].endswith('\r'):
+            self._partial = parts.pop(-2)+parts.pop()
         # parts is a list of strings, alternating between the line contents
         # and the eol character(s).  Gather up a list of lines after
         # re-attaching the newlines.

Modified: python/branches/release26-maint/Lib/email/test/test_email.py
==============================================================================
--- python/branches/release26-maint/Lib/email/test/test_email.py	(original)
+++ python/branches/release26-maint/Lib/email/test/test_email.py	Sat Jul 17 03:40:30 2010
@@ -2445,6 +2445,39 @@
 -Me
 """)
 
+    def test_pushCR_LF(self):
+        '''FeedParser BufferedSubFile.push() assumed it received complete
+           line endings.  A CR ending one push() followed by a LF starting
+           the next push() added an empty line.
+        '''
+        imt = [
+            ("a\r \n",  2),
+            ("b",       0),
+            ("c\n",     1),
+            ("",        0),
+            ("d\r\n",   1),
+            ("e\r",     0),
+            ("\nf",     1),
+            ("\r\n",    1),
+          ]
+        from email.feedparser import BufferedSubFile, NeedMoreData
+        bsf = BufferedSubFile()
+        om = []
+        nt = 0
+        for il, n in imt:
+            bsf.push(il)
+            nt += n
+            n1 = 0
+            while True:
+                ol = bsf.readline()
+                if ol == NeedMoreData:
+                    break
+                om.append(ol)
+                n1 += 1
+            self.assertTrue(n == n1)
+        self.assertTrue(len(om) == nt)
+        self.assertTrue(''.join([il for il, n in imt]) == ''.join(om))
+
 
 
 class TestParsers(TestEmailBase):

Modified: python/branches/release26-maint/Misc/NEWS
==============================================================================
--- python/branches/release26-maint/Misc/NEWS	(original)
+++ python/branches/release26-maint/Misc/NEWS	Sat Jul 17 03:40:30 2010
@@ -81,6 +81,9 @@
 Library
 -------
 
+- Issue #1555570: email no longer inserts extra blank lines when a \r\n
+  combo crosses an 8192 byte boundary.
+
 - Issue #9164: Ensure sysconfig handles dupblice archs while building on OSX
 
 - Issue #7646: The fnmatch pattern cache no longer grows without bound.

From solipsis at pitrou.net  Sat Jul 17 10:58:29 2010
From: solipsis at pitrou.net (Antoine Pitrou)
Date: Sat, 17 Jul 2010 10:58:29 +0200
Subject: [Python-checkins] r82922 - in python/branches/py3k:
 Lib/email/feedparser.py Lib/email/test/test_email.py Misc/NEWS
References: <20100717011957.CE216EE983@mail.python.org>
Message-ID: <20100717105829.30f70970@pitrou.net>

On Sat, 17 Jul 2010 03:19:57 +0200 (CEST)
r.david.murray <python-checkins at python.org> wrote:
>          self._partial = parts.pop()
> +        #GAN 29Mar09  bugs 1555570, 1721862  Confusion at 8K boundary ending with \r:
> +        # is there a \n to follow later?

What is "GAN" supposed to mean?



From python-checkins at python.org  Sat Jul 17 12:36:12 2010
From: python-checkins at python.org (stefan.krah)
Date: Sat, 17 Jul 2010 12:36:12 +0200 (CEST)
Subject: [Python-checkins] r82926 - in python/branches/py3k-cdecimal:
	Demo/classes/Complex.py Demo/classes/Dates.py
	Demo/classes/Dbm.py Demo/classes/README Demo/classes/Range.py
	Demo/classes/Rev.py Demo/classes/Vec.py Demo/embed/Makefile
	Demo/embed/demo.c Demo/md5test/md5driver.py
	Demo/parser/test_parser.py Demo/parser/test_unparse.py
	Demo/parser/unparse.py Doc/Makefile Doc/README.txt
	Doc/c-api/arg.rst Doc/c-api/datetime.rst Doc/c-api/file.rst
	Doc/c-api/init.rst Doc/c-api/intro.rst Doc/c-api/number.rst
	Doc/c-api/unicode.rst Doc/data/refcounts.dat
	Doc/distutils/apiref.rst Doc/distutils/builtdist.rst
	Doc/distutils/extending.rst Doc/distutils/setupscript.rst
	Doc/distutils/sourcedist.rst Doc/documenting/building.rst
	Doc/documenting/index.rst Doc/documenting/markup.rst
	Doc/documenting/style.rst Doc/faq/extending.rst
	Doc/faq/gui.rst Doc/glossary.rst Doc/howto/functional.rst
	Doc/howto/unicode.rst Doc/library/2to3.rst
	Doc/library/ast.rst Doc/library/calendar.rst
	Doc/library/cmath.rst Doc/library/codecs.rst
	Doc/library/contextlib.rst Doc/library/datetime.rst
	Doc/library/dis.rst Doc/library/doctest.rst
	Doc/library/fnmatch.rst Doc/library/functions.rst
	Doc/library/http.cookiejar.rst Doc/library/http.server.rst
	Doc/library/importlib.rst Doc/library/logging.rst
	Doc/library/math.rst Doc/library/multiprocessing.rst
	Doc/library/os.path.rst Doc/library/os.rst
	Doc/library/pdb.rst Doc/library/re.rst Doc/library/select.rst
	Doc/library/socket.rst Doc/library/sqlite3.rst
	Doc/library/ssl.rst Doc/library/stdtypes.rst
	Doc/library/string.rst Doc/library/struct.rst
	Doc/library/subprocess.rst Doc/library/sys.rst
	Doc/library/sysconfig.rst Doc/library/unittest.rst
	Doc/library/urllib.parse.rst Doc/library/urllib.request.rst
	Doc/library/warnings.rst Doc/reference/compound_stmts.rst
	Doc/reference/datamodel.rst Doc/reference/expressions.rst
	Doc/reference/introduction.rst Doc/reference/lexical_analysis.rst
	Doc/tools/sphinxext/indexsidebar.html
	Doc/tools/sphinxext/pyspecific.py Doc/tutorial/classes.rst
	Doc/tutorial/controlflow.rst Doc/tutorial/introduction.rst
	Doc/using/cmdline.rst Doc/using/mac.rst Doc/whatsnew/2.7.rst
	Doc/whatsnew/3.0.rst Doc/whatsnew/3.2.rst Grammar/Grammar
	Include/longobject.h Include/pyctype.h Include/structseq.h
	Include/traceback.h Include/unicodeobject.h Lib/_abcoll.py
	Lib/_strptime.py Lib/ast.py Lib/contextlib.py
	Lib/ctypes/__init__.py Lib/ctypes/test/test_callbacks.py
	Lib/ctypes/test/test_win32.py Lib/decimal.py Lib/dis.py
	Lib/distutils/command/install.py Lib/distutils/msvc9compiler.py
	Lib/distutils/tests/test_sdist.py
	Lib/distutils/unixccompiler.py Lib/email/feedparser.py
	Lib/email/test/test_email.py Lib/encodings/aliases.py
	Lib/encodings/cp858.py Lib/fnmatch.py Lib/heapq.py
	Lib/http/cookiejar.py Lib/idlelib/textView.py
	Lib/importlib/__init__.py Lib/importlib/_bootstrap.py
	Lib/importlib/abc.py Lib/importlib/test/benchmark.py
	Lib/importlib/test/extension/test_case_sensitivity.py
	Lib/importlib/test/extension/test_finder.py
	Lib/importlib/test/extension/test_loader.py
	Lib/importlib/test/extension/test_path_hook.py
	Lib/importlib/test/import_/test_path.py
	Lib/importlib/test/regrtest.py
	Lib/importlib/test/source/test_abc_loader.py
	Lib/importlib/test/source/test_case_sensitivity.py
	Lib/importlib/test/source/test_file_loader.py
	Lib/importlib/test/source/test_finder.py
	Lib/importlib/test/source/test_path_hook.py
	Lib/importlib/test/source/test_source_encoding.py
	Lib/importlib/test/test_abc.py Lib/importlib/test/test_api.py
	Lib/importlib/test/test_util.py Lib/inspect.py Lib/lib2to3
	Lib/lib2to3/fixes/fix_itertools_imports.py
	Lib/lib2to3/tests/test_fixers.py Lib/logging/config.py
	Lib/macpath.py Lib/ntpath.py Lib/os2emxpath.py Lib/pdb.doc
	Lib/pdb.py Lib/pickletools.py Lib/posixpath.py Lib/pydoc.py
	Lib/site.py Lib/smtpd.py Lib/sndhdr.py Lib/socket.py
	Lib/subprocess.py Lib/sysconfig.py Lib/tarfile.py
	Lib/test/decimaltestdata/extra.decTest
	Lib/test/math_testcases.txt Lib/test/regrtest.py
	Lib/test/sndhdrdata Lib/test/support.py
	Lib/test/symlink_support.py Lib/test/test_ast.py
	Lib/test/test_audioop.py Lib/test/test_calendar.py
	Lib/test/test_cmath.py Lib/test/test_codeccallbacks.py
	Lib/test/test_codecs.py Lib/test/test_collections.py
	Lib/test/test_complex.py Lib/test/test_contextlib.py
	Lib/test/test_datetime.py Lib/test/test_decimal.py
	Lib/test/test_descr.py Lib/test/test_dict.py
	Lib/test/test_dis.py Lib/test/test_enumerate.py
	Lib/test/test_exceptions.py Lib/test/test_extcall.py
	Lib/test/test_float.py Lib/test/test_fnmatch.py
	Lib/test/test_gdb.py Lib/test/test_genericpath.py
	Lib/test/test_getargs2.py Lib/test/test_glob.py
	Lib/test/test_http_cookiejar.py Lib/test/test_httpservers.py
	Lib/test/test_imp.py Lib/test/test_import.py
	Lib/test/test_inspect.py Lib/test/test_iter.py
	Lib/test/test_keywordonlyarg.py Lib/test/test_logging.py
	Lib/test/test_math.py Lib/test/test_optparse.py
	Lib/test/test_os.py Lib/test/test_parser.py
	Lib/test/test_platform.py Lib/test/test_posixpath.py
	Lib/test/test_property.py Lib/test/test_pyexpat.py
	Lib/test/test_runpy.py Lib/test/test_shutil.py
	Lib/test/test_sndhdr.py Lib/test/test_strtod.py
	Lib/test/test_struct.py Lib/test/test_structseq.py
	Lib/test/test_subprocess.py Lib/test/test_syntax.py
	Lib/test/test_sysconfig.py Lib/test/test_tarfile.py
	Lib/test/test_threaded_import.py Lib/test/test_tokenize.py
	Lib/test/test_tuple.py Lib/test/test_unicode.py
	Lib/test/test_unpack.py Lib/test/test_urllib.py
	Lib/test/test_urllib2.py Lib/test/test_urlparse.py
	Lib/test/test_xmlrpc_net.py Lib/unittest/case.py
	Lib/unittest/test/test_case.py Lib/urllib/parse.py
	Lib/urllib/request.py Lib/uuid.py Lib/warnings.py
	Lib/xml/dom/expatbuilder.py Makefile.pre.in Misc/ACKS
	Misc/NEWS Misc/Vim/vimrc Misc/maintainers.rst
	Modules/_ctypes/callproc.c Modules/_ctypes/libffi/fficonfig.py.in
	Modules/_ctypes/libffi_msvc/LICENSE
	Modules/_ctypes/libffi_msvc/README
	Modules/_ctypes/libffi_msvc/README.ctypes
	Modules/_ctypes/libffi_msvc/ffi.c Modules/_ctypes/libffi_msvc/ffi.h
	Modules/_ctypes/libffi_msvc/ffi_common.h
	Modules/_ctypes/libffi_msvc/fficonfig.h
	Modules/_ctypes/libffi_msvc/ffitarget.h
	Modules/_ctypes/libffi_msvc/prep_cif.c
	Modules/_ctypes/libffi_msvc/types.c
	Modules/_ctypes/libffi_msvc/win32.c
	Modules/_ctypes/libffi_msvc/win64.asm Modules/_io/_iomodule.c
	Modules/_math.c Modules/_scproxy.c Modules/_ssl.c
	Modules/_struct.c Modules/_testcapimodule.c Modules/_time.c
	Modules/_time.h Modules/audioop.c Modules/cmathmodule.c
	Modules/datetimemodule.c Modules/makesetup
	Modules/mathmodule.c Modules/parsermodule.c
	Modules/posixmodule.c Modules/selectmodule.c
	Modules/timemodule.c Modules/xxmodule.c Objects/abstract.c
	Objects/bytearrayobject.c Objects/bytes_methods.c
	Objects/descrobject.c Objects/enumobject.c
	Objects/exceptions.c Objects/floatobject.c
	Objects/frameobject.c Objects/memoryobject.c
	Objects/stringlib/formatter.h Objects/structseq.c
	Objects/typeobject.c Objects/unicodeobject.c
	PC/VS8.0/pythoncore.vcproj PCbuild/_ctypes.vcproj
	PCbuild/pythoncore.vcproj Parser/Python.asdl Parser/asdl_c.py
	Python/Python-ast.c Python/_warnings.c Python/ast.c
	Python/ceval.c Python/compile.c Python/getargs.c
	Python/graminit.c Python/import.c Python/pythonrun.c
	Python/sysmodule.c Python/traceback.c README Tools/README
	Tools/gdb/libpython.py Tools/i18n/msgfmt.py Tools/scripts
	Tools/scripts/patchcheck.py Tools/webchecker configure
	configure.in setup.py
Message-ID: <20100717103612.C11E2EE99B@mail.python.org>

Author: stefan.krah
Date: Sat Jul 17 12:36:09 2010
New Revision: 82926

Log:
Merged revisions 81981,81988-81990,81994,81999-82001,82003-82005,82008,82011,82014-82015,82022,82028,82034-82035,82037-82038,82041,82043,82045,82049,82051,82053,82055,82057,82059,82061,82068,82070-82071,82073,82076,82084,82089-82090,82113-82115,82119,82125,82128,82131,82138,82146,82148,82152,82156,82159,82162-82163,82166,82168,82171,82180-82181,82184-82186,82190,82193-82194,82200,82202,82204,82206,82208,82211,82214,82220,82223,82226,82228,82230-82232,82234,82238,82244,82247,82252,82255,82259,82262-82263,82269-82270,82273,82278,82281,82283,82286,82288-82302,82304-82316,82318,82320-82321,82326,82328,82331,82333,82336,82342-82343,82346-82347,82349,82351,82354,82357-82360,82365,82370-82371,82373,82375-82377,82381,82383-82384,82386-82387,82391-82394,82402,82405,82407,82410,82413,82416-82417,82422,82430,82434,82437-82438,82440,82442-82443,82448,82452,82456,82462,82464-82465,82467-82468,82471-82472,82478,82480-82481,82484-82485,82487-82488,82495,82507,82510,82514-82524,82526-82527,82533,82535,82537-82538,82540-82542,82547,82550-82553,82555,82557-82561,82567,82570,82573,82576-82578,82585,82587,82589,82591,82593-82595,82597,82599-82601,82604,82607,82609-82610,82615,82617,82624-82626,82628-82630,82632-82633,82636-82637,82642,82644,82646-82647,82649-82650,82654,82659,82661-82663,82723-82724,82730-82731,82735,82737,82739,82741,82743-82746,82748-82750,82753,82757-82764,82766-82767,82771,82777,82780,82784-82785,82789-82790,82796,82798-82799,82801,82804-82806,82814,82817-82822,82825,82828,82830-82832,82834-82835,82837-82839,82842-82843,82848-82850,82856-82857,82871-82874,82879,82881,82885,82887,82890,82895,82899,82910,82915,82918-82919,82921-82922 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

................
  r81981 | alexander.belopolsky | 2010-06-14 16:15:50 +0200 (Mon, 14 Jun 2010) | 3 lines
  
  Issue #5094: The ``datetime`` module now has a simple concrete class
  implementing ``datetime.tzinfo`` interface.
................
  r81988 | alexander.belopolsky | 2010-06-14 19:32:03 +0200 (Mon, 14 Jun 2010) | 1 line
  
  Issue 6280: Tests and simpler implementation for calendar.timegm
................
  r81989 | alexander.belopolsky | 2010-06-14 20:33:19 +0200 (Mon, 14 Jun 2010) | 1 line
  
  Undo r81988 code change leaving added test.
................
  r81990 | brett.cannon | 2010-06-15 00:22:54 +0200 (Tue, 15 Jun 2010) | 5 lines
  
  Switch the __import__ state check from using __builtins__ to builtins to be
  nicer to other VMs.
  
  Thanks to Philip Jenvey for the pointer.
................
  r81994 | mark.dickinson | 2010-06-15 10:42:37 +0200 (Tue, 15 Jun 2010) | 10 lines
  
  Merged revisions 81992 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r81992 | mark.dickinson | 2010-06-15 09:33:03 +0100 (Tue, 15 Jun 2010) | 3 lines
    
    Issue #8469:  Further clarifications and improvements to struct module
    documentation.  Thanks Mads Kiilerich.
  ........
................
  r81999 | ronald.oussoren | 2010-06-15 18:05:20 +0200 (Tue, 15 Jun 2010) | 5 lines
  
  Fix for issue #8577: without this patch test_distutils
  will fail when builddir != srcdir (that is, when you
  run configure in a directory that is not the top of
  the source tree).
................
  r82000 | antoine.pitrou | 2010-06-15 19:00:21 +0200 (Tue, 15 Jun 2010) | 3 lines
  
  Fixes to the PyFile_FromFd() doc, by Renato Cunha.
................
  r82001 | antoine.pitrou | 2010-06-15 19:30:16 +0200 (Tue, 15 Jun 2010) | 3 lines
  
  Further refinements to the C file API.
................
  r82003 | alexander.belopolsky | 2010-06-15 20:40:23 +0200 (Tue, 15 Jun 2010) | 1 line
  
  Minor changes to the choice of assert methods
................
  r82004 | alexander.belopolsky | 2010-06-15 21:24:52 +0200 (Tue, 15 Jun 2010) | 1 line
  
  Issue 5094: minor documentation fixes
................
  r82005 | ronald.oussoren | 2010-06-15 23:19:50 +0200 (Tue, 15 Jun 2010) | 3 lines
  
  Fix for buildbot failure in  r81999.
................
  r82008 | r.david.murray | 2010-06-16 01:46:40 +0200 (Wed, 16 Jun 2010) | 9 lines
  
  Merged revisions 81634 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r81634 | r.david.murray | 2010-05-31 21:42:41 -0400 (Mon, 31 May 2010) | 2 lines
    
    #7583: clarify discussion of hard tab expansion in doctests.
  ........
................
  r82011 | r.david.murray | 2010-06-16 04:19:40 +0200 (Wed, 16 Jun 2010) | 17 lines
  
  Merged revisions 81675 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r81675 | r.david.murray | 2010-06-03 11:43:20 -0400 (Thu, 03 Jun 2010) | 10 lines
    
    #5610: use \Z not $ so we don't eat extra chars when body part ends with \r\n.
    
    If a body part ended with \r\n, feedparser, using '$' to terminate its
    search for the newline, would match on the \r\n, and think that it needed
    to strip two characters in order to account for the line end before the
    boundary.  That made it chop one too many characters off the end of
    the body part.  Using \Z makes the match correct.
    
    Patch and test by Tony Nelson.
  ........
................
  r82014 | r.david.murray | 2010-06-16 14:53:07 +0200 (Wed, 16 Jun 2010) | 20 lines
  
  Blocked revisions 81571,81678 via svnmerge
  
  I'm going to merge 81678 by hand.
  
  ........
    r81571 | victor.stinner | 2010-05-27 18:29:48 -0400 (Thu, 27 May 2010) | 3 lines
    
    Issue #8835: test_support.transient_internet() catchs gaierror(EAI_NONAME) and
    gaierror(EAI_NODATA)
  ........
    r81678 | r.david.murray | 2010-06-03 16:19:25 -0400 (Thu, 03 Jun 2010) | 7 lines
    
    #8889: rewrite transient_internet so we don't use EAI_NODATA on FreeBSD.
    
    FreeBSD doesn't have socket.EAI_NODATA.  I rewrote the routine because
    there's no easy way to conditionally include a context manager in a
    with statement.  As a side benefit, instead of a stack of context
    managers there's now only one.
  ........
................
  r82015 | r.david.murray | 2010-06-16 14:56:31 +0200 (Wed, 16 Jun 2010) | 2 lines
  
  Remove versionadded accidentally introduced by r82008.
................
  r82022 | senthil.kumaran | 2010-06-16 18:41:11 +0200 (Wed, 16 Jun 2010) | 9 lines
  
  Merged revisions 82018 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82018 | senthil.kumaran | 2010-06-16 20:25:31 +0530 (Wed, 16 Jun 2010) | 3 lines
    
    Fix Issue8937 - SimpleHTTPServer should contain usage example
  ........
................
  r82028 | senthil.kumaran | 2010-06-16 19:44:57 +0200 (Wed, 16 Jun 2010) | 9 lines
  
  Merged revisions 82026 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82026 | senthil.kumaran | 2010-06-16 23:07:32 +0530 (Wed, 16 Jun 2010) | 3 lines
    
    Addressing RDM's review comments on the doc change.
  ........
................
  r82034 | alexander.belopolsky | 2010-06-17 00:38:15 +0200 (Thu, 17 Jun 2010) | 6 lines
  
  Issue #9012: "Separate compilation of time and datetime modules."
  Segregated code shared between time and datetime modules into
  Modules/_time.c.  Added a new header file, Modules/_time.h, which
  will be used instead of Include/timefuncs.h for declarations shared
  between time and datetime modules.
................
  r82035 | victor.stinner | 2010-06-17 01:05:06 +0200 (Thu, 17 Jun 2010) | 4 lines
  
  Issue #9012: Add _time.c and _time.h to the Visual Studio project file
  
  Add these new files, added by r82034, to the pythoncore project.
................
  r82037 | victor.stinner | 2010-06-17 01:33:54 +0200 (Thu, 17 Jun 2010) | 4 lines
  
  Issue #850997: mbcs encoding (Windows only) handles errors argument: strict
  mode raises unicode errors. The encoder only supports "strict" and "replace"
  error handlers, the decoder only supports "strict" and "ignore" error handlers.
................
  r82038 | victor.stinner | 2010-06-17 01:48:49 +0200 (Thu, 17 Jun 2010) | 2 lines
  
  Issue #850997: Oops, I forgot the author of the patch: Mark Hammond
................
  r82041 | r.david.murray | 2010-06-17 04:04:29 +0200 (Thu, 17 Jun 2010) | 16 lines
  
  Merged revisions 82039 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82039 | r.david.murray | 2010-06-16 21:36:52 -0400 (Wed, 16 Jun 2010) | 10 lines
    
    #8720: fix inspect regression by teaching getsourcefile about linecache.
    
    The fix for issue 4050 caused a regression:  before that fix, source
    lines in the linecache would eventually be found by inspect.  After the
    fix inspect reports an error earlier, and the source isn't found.
    The fix for the fix is to have getsourcefile look in the linecache for
    the file and return the psuedo-filename if the source is there, just as
    it already returns it if there is a PEP 302 loader.
  ........
................
  r82043 | mark.dickinson | 2010-06-17 14:33:22 +0200 (Thu, 17 Jun 2010) | 6 lines
  
  Issue #9011: Remove buggy and unnecessary ST->AST compilation code
  dealing with unary minus applied to a constant.  The removed code was
  mutating the ST, causing a second compilation to fail.  (The peephole
  optimizer already takes care of optimizing this case, so there's no
  lost optimization opportunity here.)
................
  r82045 | r.david.murray | 2010-06-17 15:23:18 +0200 (Thu, 17 Jun 2010) | 5 lines
  
  Don't use os.normcase when the result we are expecting is None.
  
  This worked fine on linux but fails on Windows.  That may or may
  not be a but in normcase.
................
  r82049 | senthil.kumaran | 2010-06-17 18:48:06 +0200 (Thu, 17 Jun 2010) | 9 lines
  
  Merged revisions 82047 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82047 | senthil.kumaran | 2010-06-17 22:08:34 +0530 (Thu, 17 Jun 2010) | 3 lines
    
    Fix Issue4452 - Incorrect docstring of os.setpgrp
  ........
................
  r82051 | mark.dickinson | 2010-06-17 20:24:52 +0200 (Thu, 17 Jun 2010) | 2 lines
  
  Add note about changes to the `round` function between 2.x and 3.x.
................
  r82053 | alexander.belopolsky | 2010-06-17 20:30:34 +0200 (Thu, 17 Jun 2010) | 2 lines
  
  Issue #6641: The datetime.strptime method now supports the %z directive.
................
  r82055 | barry.warsaw | 2010-06-17 20:38:20 +0200 (Thu, 17 Jun 2010) | 2 lines
  
  Typo repair.
................
  r82057 | victor.stinner | 2010-06-17 23:43:33 +0200 (Thu, 17 Jun 2010) | 2 lines
  
  Issue #8203: Fix IDLE Credits dialog: view_file() uses its encoding argument.
................
  r82059 | victor.stinner | 2010-06-18 01:08:50 +0200 (Fri, 18 Jun 2010) | 5 lines
  
  Issue #6543: Write the traceback in the terminal encoding instead of utf-8.
  Fix the encoding of the modules filename.
  
  Reindent also traceback.h, just because I hate tabs :-)
................
  r82061 | victor.stinner | 2010-06-18 01:17:37 +0200 (Fri, 18 Jun 2010) | 2 lines
  
  Issue #6543: Mention the author of the patch, Amaury Forgeot d'Arc
................
  r82068 | senthil.kumaran | 2010-06-18 17:08:18 +0200 (Fri, 18 Jun 2010) | 3 lines
  
  Fix Issue1368368 - prompt_user_passwd() in FancyURLopener masks 401 Unauthorized error page
................
  r82070 | alexander.belopolsky | 2010-06-18 18:22:00 +0200 (Fri, 18 Jun 2010) | 1 line
  
  Fixed a typo in a comment.
................
  r82071 | alexander.belopolsky | 2010-06-18 18:57:49 +0200 (Fri, 18 Jun 2010) | 1 line
  
  Added a new line at the end of the file.
................
  r82073 | alexander.belopolsky | 2010-06-18 20:44:37 +0200 (Fri, 18 Jun 2010) | 4 lines
  
  Issue #6641: Original commit for this issue, r82053, introduced a
  regression making datetime subclass' strptime return datetime rather
  than subclass instances.  Fixed this bug and a few typos.
................
  r82076 | jean-paul.calderone | 2010-06-18 22:03:54 +0200 (Fri, 18 Jun 2010) | 19 lines
  
  Merged revisions 82075 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82075 | jean-paul.calderone | 2010-06-18 16:00:17 -0400 (Fri, 18 Jun 2010) | 12 lines
    
    Revert r60115
    
    This revision introduced quoting for strings containing | based
    on a misunderstanding of the commonly used quoting rules used
    on Windows.
    
    | is interpreted by cmd.exe, not by the MS C runtime argv initializer.
    It only needs to be quoted if it is part of an argument passed through
    cmd.exe.
    
    See issue1300, issue7839, and issue8972.
  ........
................
  r82084 | victor.stinner | 2010-06-19 01:59:45 +0200 (Sat, 19 Jun 2010) | 6 lines
  
  Issue #8939: Improve arg.rst
  
   * Add :ctype: to Py_BEGIN_ALLOW_THREADS and int
   * "s" and "s#" formats of Py_BuildValue(): specify that the Python object type
     is str in the description
................
  r82089 | jean-paul.calderone | 2010-06-19 21:54:48 +0200 (Sat, 19 Jun 2010) | 1 line
  
  merge forward from the python 2.x branch
................
  r82090 | jean-paul.calderone | 2010-06-19 21:58:37 +0200 (Sat, 19 Jun 2010) | 1 line
  
  Revert r82089.  Commit was intended for a branch.
................
  r82113 | benjamin.peterson | 2010-06-20 17:25:14 +0200 (Sun, 20 Jun 2010) | 8 lines
  
  Blocked revisions 82112 via svnmerge
  
  ........
    r82112 | benjamin.peterson | 2010-06-20 10:12:04 -0500 (Sun, 20 Jun 2010) | 1 line
    
    update emacs section
  ........
................
  r82114 | benjamin.peterson | 2010-06-20 17:27:07 +0200 (Sun, 20 Jun 2010) | 1 line
  
  update release schedule url
................
  r82115 | benjamin.peterson | 2010-06-20 17:27:42 +0200 (Sun, 20 Jun 2010) | 1 line
  
  add copyright years
................
  r82119 | mark.dickinson | 2010-06-20 22:01:04 +0200 (Sun, 20 Jun 2010) | 9 lines
  
  Merged revisions 82117 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82117 | mark.dickinson | 2010-06-20 19:50:19 +0100 (Sun, 20 Jun 2010) | 1 line
    
    Merge test_strtod and test_float string-to-float conversion tests.
  ........
................
  r82125 | brett.cannon | 2010-06-21 04:49:35 +0200 (Mon, 21 Jun 2010) | 2 lines
  
  Move over to assertIs.
................
  r82128 | alexander.belopolsky | 2010-06-21 17:21:14 +0200 (Mon, 21 Jun 2010) | 2 lines
  
  Issue #9005: Prevent utctimetuple() from producing year 0 or year 10,000.
................
  r82131 | benjamin.peterson | 2010-06-21 17:37:16 +0200 (Mon, 21 Jun 2010) | 9 lines
  
  Merged revisions 82130 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82130 | benjamin.peterson | 2010-06-21 10:27:46 -0500 (Mon, 21 Jun 2010) | 1 line
    
    fix finding visual studio 2008 on 64 bit #8854
  ........
................
  r82138 | thomas.heller | 2010-06-21 18:00:31 +0200 (Mon, 21 Jun 2010) | 13 lines
  
  Merged revisions 82126-82127 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82126 | thomas.heller | 2010-06-21 16:00:24 +0200 (Mo, 21 Jun 2010) | 1 line
    
    Fix #8959 by reverting revision 80761.
  ........
    r82127 | thomas.heller | 2010-06-21 17:01:18 +0200 (Mo, 21 Jun 2010) | 2 lines
    
    Add tests for problems reported in issue 8959.
  ........
................
  r82146 | senthil.kumaran | 2010-06-22 04:42:52 +0200 (Tue, 22 Jun 2010) | 3 lines
  
  Minor docs issue.
................
  r82148 | senthil.kumaran | 2010-06-22 04:57:23 +0200 (Tue, 22 Jun 2010) | 3 lines
  
  Minor comment formatting.
................
  r82152 | alexander.belopolsky | 2010-06-22 16:07:33 +0200 (Tue, 22 Jun 2010) | 1 line
  
  Added more test cases
................
  r82156 | benjamin.peterson | 2010-06-22 20:11:34 +0200 (Tue, 22 Jun 2010) | 10 lines
  
  Blocked revisions 82155 via svnmerge
  
  ........
    r82155 | benjamin.peterson | 2010-06-22 13:09:02 -0500 (Tue, 22 Jun 2010) | 4 lines
    
    keep UserDict an old-style class
    
    Be generous in abc.py to allow this.
  ........
................
  r82159 | benjamin.peterson | 2010-06-22 21:21:52 +0200 (Tue, 22 Jun 2010) | 9 lines
  
  Merged revisions 82157 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82157 | benjamin.peterson | 2010-06-22 14:16:37 -0500 (Tue, 22 Jun 2010) | 1 line
    
    remove INT_MAX assertions; they can fail with large Py_ssize_t #9058
  ........
................
  r82162 | benjamin.peterson | 2010-06-22 21:52:02 +0200 (Tue, 22 Jun 2010) | 8 lines
  
  Blocked revisions 82161 via svnmerge
  
  ........
    r82161 | benjamin.peterson | 2010-06-22 14:49:47 -0500 (Tue, 22 Jun 2010) | 1 line
    
    bump revision
  ........
................
  r82163 | benjamin.peterson | 2010-06-22 22:02:39 +0200 (Tue, 22 Jun 2010) | 9 lines
  
  Merged revisions 82160 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82160 | benjamin.peterson | 2010-06-22 14:45:51 -0500 (Tue, 22 Jun 2010) | 1 line
    
    spacing nit; this isn't C
  ........
................
  r82166 | benjamin.peterson | 2010-06-22 22:29:32 +0200 (Tue, 22 Jun 2010) | 9 lines
  
  Merged revisions 82165 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82165 | benjamin.peterson | 2010-06-22 15:26:20 -0500 (Tue, 22 Jun 2010) | 1 line
    
    must force gc here
  ........
................
  r82168 | benjamin.peterson | 2010-06-22 22:34:34 +0200 (Tue, 22 Jun 2010) | 9 lines
  
  Merged revisions 82167 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82167 | benjamin.peterson | 2010-06-22 15:32:02 -0500 (Tue, 22 Jun 2010) | 1 line
    
    mark ref counting as impl detail
  ........
................
  r82171 | antoine.pitrou | 2010-06-22 23:49:39 +0200 (Tue, 22 Jun 2010) | 10 lines
  
  Merged revisions 82169 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82169 | antoine.pitrou | 2010-06-22 23:42:05 +0200 (mar., 22 juin 2010) | 4 lines
    
    Fix misindents in compile.c (for Benjamin).
    Of course, whoever used the wrong indentation rules needs to be spanked.
  ........
................
  r82180 | stefan.krah | 2010-06-23 20:35:33 +0200 (Wed, 23 Jun 2010) | 8 lines
  
  Blocked revisions 82177 via svnmerge
  
  ........
    r82177 | stefan.krah | 2010-06-23 20:12:09 +0200 (Wed, 23 Jun 2010) | 3 lines
    
    Issue #8930: Remaining indentation fixes after the Grand Unified Indenting.
  ........
................
  r82181 | stefan.krah | 2010-06-23 20:42:39 +0200 (Wed, 23 Jun 2010) | 3 lines
  
  Issue #8930: Remaining indentation fixes after the Grand Unified Indenting.
................
  r82184 | alexander.belopolsky | 2010-06-23 23:40:15 +0200 (Wed, 23 Jun 2010) | 1 line
  
  Issue #9051: Instances of timezone class can now be pickled.
................
  r82185 | alexander.belopolsky | 2010-06-24 00:29:48 +0200 (Thu, 24 Jun 2010) | 1 line
  
  Test future pickle protocols.  Thanks Antoine Pitrou for suggestion.
................
  r82186 | alexander.belopolsky | 2010-06-24 00:58:49 +0200 (Thu, 24 Jun 2010) | 1 line
  
  Added more test cases
................
  r82190 | benjamin.peterson | 2010-06-24 02:17:03 +0200 (Thu, 24 Jun 2010) | 9 lines
  
  Merged revisions 82189 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82189 | benjamin.peterson | 2010-06-23 19:12:40 -0500 (Wed, 23 Jun 2010) | 1 line
    
    prevent assignment to set literals
  ........
................
  r82193 | stefan.krah | 2010-06-24 12:22:10 +0200 (Thu, 24 Jun 2010) | 10 lines
  
  Blocked revisions 82191 via svnmerge
  
  ........
    r82191 | stefan.krah | 2010-06-24 11:33:05 +0200 (Thu, 24 Jun 2010) | 5 lines
    
    Issue #9020: The Py_IS* macros from pyctype.h should generally only be
    used with signed/unsigned char arguments. For integer arguments, EOF
    has to be handled separately.
  ........
................
  r82194 | stefan.krah | 2010-06-24 12:25:08 +0200 (Thu, 24 Jun 2010) | 3 lines
  
  Add specification for the Py_IS* macros in pyctype.h.
................
  r82200 | victor.stinner | 2010-06-25 00:08:25 +0200 (Fri, 25 Jun 2010) | 3 lines
  
  Issue #8949: "z" format of PyArg_Parse*() functions doesn't accept bytes
  objects, as described in the documentation.
................
  r82202 | victor.stinner | 2010-06-25 00:31:12 +0200 (Fri, 25 Jun 2010) | 2 lines
  
  PyArg_Parse*() functions: factorize code for s/z and u/Z formats
................
  r82204 | antoine.pitrou | 2010-06-25 00:34:04 +0200 (Fri, 25 Jun 2010) | 5 lines
  
  Issue #8682: The ssl module now temporary increments the reference count of
  a socket object got through `PyWeakref_GetObject`, so as to avoid possible
  deallocation while the object is still being used.
................
  r82206 | victor.stinner | 2010-06-25 00:57:10 +0200 (Fri, 25 Jun 2010) | 2 lines
  
  getbuffer(): release the buffer on error (if the buffer is not contiguous)
................
  r82208 | victor.stinner | 2010-06-25 02:02:38 +0200 (Fri, 25 Jun 2010) | 3 lines
  
  Issue #8850: Remove "w" and "w#" formats from PyArg_Parse*() functions, use
  "w*" format instead. Add tests for "w*" format.
................
  r82211 | antoine.pitrou | 2010-06-25 02:07:34 +0200 (Fri, 25 Jun 2010) | 10 lines
  
  Merged revisions 82210 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82210 | antoine.pitrou | 2010-06-25 02:03:21 +0200 (ven., 25 juin 2010) | 4 lines
    
    Issue #9075: In the ssl module, remove the setting of a `debug` flag
    on an OpenSSL structure.
  ........
................
  r82214 | ezio.melotti | 2010-06-25 12:56:11 +0200 (Fri, 25 Jun 2010) | 1 line
  
  #9018: os.path.normcase() now raises a TypeError if the argument is not str or bytes.
................
  r82220 | benjamin.peterson | 2010-06-25 21:30:21 +0200 (Fri, 25 Jun 2010) | 1 line
  
  only take into account positional arguments count in related error messages
................
  r82223 | mark.dickinson | 2010-06-25 22:22:24 +0200 (Fri, 25 Jun 2010) | 9 lines
  
  Merged revisions 82221 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82221 | mark.dickinson | 2010-06-25 21:19:48 +0100 (Fri, 25 Jun 2010) | 1 line
    
    Fix indentation of Python code example in C comment.
  ........
................
  r82226 | benjamin.peterson | 2010-06-25 22:37:22 +0200 (Fri, 25 Jun 2010) | 9 lines
  
  Merged revisions 82225 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82225 | benjamin.peterson | 2010-06-25 15:34:01 -0500 (Fri, 25 Jun 2010) | 1 line
    
    mark implementation detail as such
  ........
................
  r82228 | benjamin.peterson | 2010-06-25 23:24:10 +0200 (Fri, 25 Jun 2010) | 8 lines
  
  Blocked revisions 82227 via svnmerge
  
  ........
    r82227 | benjamin.peterson | 2010-06-25 16:19:04 -0500 (Fri, 25 Jun 2010) | 1 line
    
    this must be a typo
  ........
................
  r82230 | benjamin.peterson | 2010-06-26 01:02:11 +0200 (Sat, 26 Jun 2010) | 9 lines
  
  Merged revisions 82229 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82229 | benjamin.peterson | 2010-06-25 17:58:47 -0500 (Fri, 25 Jun 2010) | 1 line
    
    mark implementation detail as such
  ........
................
  r82231 | benjamin.peterson | 2010-06-26 01:24:35 +0200 (Sat, 26 Jun 2010) | 1 line
  
  fix typos
................
  r82232 | benjamin.peterson | 2010-06-26 01:37:15 +0200 (Sat, 26 Jun 2010) | 1 line
  
  remove old, outdated tool
................
  r82234 | r.david.murray | 2010-06-26 02:17:12 +0200 (Sat, 26 Jun 2010) | 9 lines
  
  Merged revisions 82233 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82233 | r.david.murray | 2010-06-25 20:06:44 -0400 (Fri, 25 Jun 2010) | 2 lines
    
    #4640: add a test to optparse that proves issue is invalid.
  ........
................
  r82238 | alexander.belopolsky | 2010-06-26 04:15:07 +0200 (Sat, 26 Jun 2010) | 11 lines
  
  Merged revisions 82236 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82236 | alexander.belopolsky | 2010-06-25 22:05:19 -0400 (Fri, 25 Jun 2010) | 4 lines
    
    Issue #9024: Fixed mark-up for PyDateTime_IMPORT macro and added that
    it is typically used in module initialization function. Thanks Tim
    Golden for the patch.
  ........
................
  r82244 | r.david.murray | 2010-06-26 05:34:33 +0200 (Sat, 26 Jun 2010) | 9 lines
  
  Merged revisions 82242 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82242 | r.david.murray | 2010-06-25 23:27:32 -0400 (Fri, 25 Jun 2010) | 2 lines
    
    Fix indentation in recently added test.
  ........
................
  r82247 | r.david.murray | 2010-06-26 20:44:14 +0200 (Sat, 26 Jun 2010) | 14 lines
  
  Blocked revisions 82246 via svnmerge
  
  ........
    r82246 | r.david.murray | 2010-06-26 14:39:50 -0400 (Sat, 26 Jun 2010) | 8 lines
    
    #9085: email versions have gotten out of sync, 2.7 is actually 4.0.3.
    
    In 2.5 the email version was bumped to 4.0.2 just before release but
    after the maintenance branch was set up.  The bump was not backported
    to trunk, and 2.6 went out the door with a version number of 4.0.1.
    I bumped this to 4.0.2 because of a behavior change, but that makes
    it look the same as 2.5.  So bump it again to 4.0.3.
  ........
................
  r82252 | ezio.melotti | 2010-06-26 20:50:39 +0200 (Sat, 26 Jun 2010) | 9 lines
  
  Merged revisions 82248 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82248 | ezio.melotti | 2010-06-26 21:44:42 +0300 (Sat, 26 Jun 2010) | 1 line
    
    Fix extra space.
  ........
................
  r82255 | alexander.belopolsky | 2010-06-26 20:57:02 +0200 (Sat, 26 Jun 2010) | 9 lines
  
  Merged revisions 82251 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82251 | alexander.belopolsky | 2010-06-26 14:49:10 -0400 (Sat, 26 Jun 2010) | 1 line
    
    Issue 9024: Added :cdata: tag to PyDateTimeAPI
  ........
................
  r82259 | brett.cannon | 2010-06-27 00:29:06 +0200 (Sun, 27 Jun 2010) | 6 lines
  
  Having CFLAGS come before OPT prevents silencing certain classes of warnings
  from the fact that OPT contains -Wall be default. This is annoying when
  compilers like clang have thorough debugging information about things that
  Python does extensively (e.g. -Wunused-value for unused return values caused by
  a macro use).
................
  r82262 | georg.brandl | 2010-06-27 12:17:12 +0200 (Sun, 27 Jun 2010) | 1 line
  
  #9078: fix some Unicode C API descriptions, in comments and docs.
................
  r82263 | georg.brandl | 2010-06-27 12:37:48 +0200 (Sun, 27 Jun 2010) | 1 line
  
  #9064: accept number of frames for "up" and "down" commands in pdb.
................
  r82269 | georg.brandl | 2010-06-27 12:59:19 +0200 (Sun, 27 Jun 2010) | 1 line
  
  Wording fix.
................
  r82270 | georg.brandl | 2010-06-27 13:24:03 +0200 (Sun, 27 Jun 2010) | 1 line
  
  Untabify.
................
  r82273 | ronald.oussoren | 2010-06-27 14:37:46 +0200 (Sun, 27 Jun 2010) | 15 lines
  
  Merged revisions 82272 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82272 | ronald.oussoren | 2010-06-27 14:36:16 +0200 (Sun, 27 Jun 2010) | 8 lines
    
    Two small fixes for the support for SDKs on MacOSX:
    
    1) The code that checks if an path should be located in the SDK
       explicitly excludes /usr/local. This fixes issue9046
    
    2) The SDK variant for filtering "db_dirs_to_check" in setup.py
       was not doing anything because of a missing assignment.
  ........
................
  r82278 | ronald.oussoren | 2010-06-27 14:45:47 +0200 (Sun, 27 Jun 2010) | 12 lines
  
  Merged revisions 82276 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82276 | ronald.oussoren | 2010-06-27 14:43:47 +0200 (Sun, 27 Jun 2010) | 5 lines
    
    The uuid library on OSX 10.5 seems to contain the same bug as the one
    on OSX 10.6, therefore don't use it there either.
    
    This fixes issue8621.
  ........
................
  r82281 | ronald.oussoren | 2010-06-27 14:51:31 +0200 (Sun, 27 Jun 2010) | 19 lines
  
  Merged revisions 82150 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82150 | ronald.oussoren | 2010-06-22 11:32:22 +0200 (Tue, 22 Jun 2010) | 12 lines
    
    The code in _scproxy (a mac specific helper module to 
    detect proxy settings) had the wrong logic for detecting
    if the checkbox 'Exclude simple hostnames' is checked. This
    checkin fixes that.
    
    As a result the test failure 'Issue8455' goes away on systems
    where the checkbox is not checked.
    
    I'm carefully avoiding saying that is fixes that issue,
    test_urllib2_localnet assumes that system proxy settings are
    empty (not just on OSX, see Issue8455 for details).
  ........
................
  r82283 | georg.brandl | 2010-06-27 15:28:06 +0200 (Sun, 27 Jun 2010) | 8 lines
  
  Blocked revisions 82271 via svnmerge
  
  ........
    r82271 | georg.brandl | 2010-06-27 13:24:15 +0200 (So, 27 Jun 2010) | 1 line
    
    Untabify.
  ........
................
  r82286 | ronald.oussoren | 2010-06-27 16:26:30 +0200 (Sun, 27 Jun 2010) | 11 lines
  
  Merged revisions 82284 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82284 | ronald.oussoren | 2010-06-27 15:59:39 +0200 (Sun, 27 Jun 2010) | 4 lines
    
    Fix for Issue8883: without this patch test_urllib will fail
    when there is a bare IP address in the "Bypass proxy settings
    for these Hosts & Domains" list on MacOSX.
  ........
................
  r82288 | mark.dickinson | 2010-06-27 19:36:54 +0200 (Sun, 27 Jun 2010) | 1 line
  
  Issue #9089: PyNumber_Int is no more.  Remove it from the docs.
................
  r82289 | mark.dickinson | 2010-06-27 20:18:01 +0200 (Sun, 27 Jun 2010) | 1 line
  
  Issue #9089: Remove intobject.h from MSVC project files.
................
  r82290 | mark.dickinson | 2010-06-27 20:19:09 +0200 (Sun, 27 Jun 2010) | 1 line
  
  Issue #9089: Remove references to intobject.c and intobject.h from comments.
................
  r82291 | brett.cannon | 2010-06-27 22:40:18 +0200 (Sun, 27 Jun 2010) | 1 line
  
  Revert r82259; still no good way to override instead of replace OPT, but changing order alters pydebug optimization flags.
................
  r82292 | benjamin.peterson | 2010-06-27 22:50:37 +0200 (Sun, 27 Jun 2010) | 451 lines
  
  Blocked revisions 79620,79624-79625,79633-79637,79639-79640,79650-79651,79654,79673,79693,79695,79697-79698,79704,79706-79707,79710,79712-79714,79718-79719,79722,79730,79744,79750,79755,79760,79769,79808,79811,79890,80062,80065,80082,80086,80171,80189,80242-80243,80374,80411-80412,80445,80579-80581,80700,80737,80745,80752,80794,80813-80814,80823,80963,80982,81012-81013,81030,81035,81088,81140-81141,81149-81151,81275,81484,81502,81517,81561,81705,81847,81919,81964,82056,82136-82137,82142,82149 via svnmerge
  
  ........
    r79620 | brian.curtin | 2010-04-02 16:18:14 -0500 (Fri, 02 Apr 2010) | 9 lines
    
    Implement #7347. Add CreateKeyEx, DeleteKeyEx, and update _winreg tests.
    
    *ReflectionKey functions used to not be documented or tested, but they are
    now sufficiently documented and tested on platforms where they apply. 
    Additionally, fixed a bug in QueryReflectionKey which was returning an
    incorrect value.
    
    All tests pass from XP through Windows 7, on 32 and 64-bit platforms.
  ........
    r79624 | ezio.melotti | 2010-04-02 16:43:10 -0500 (Fri, 02 Apr 2010) | 1 line
    
    Fix test_compiler.py that was using unittest.__file__ to find Lib/ (unittest is now a package).
  ........
    r79625 | brian.curtin | 2010-04-02 16:51:37 -0500 (Fri, 02 Apr 2010) | 2 lines
    
    Add a line about #7347 to Misc\News
  ........
    r79633 | brian.curtin | 2010-04-02 18:26:06 -0500 (Fri, 02 Apr 2010) | 14 lines
    
    Implement #1220212. Add os.kill support for Windows.
    
    os.kill takes one of two newly added signals, CTRL_C_EVENT and 
    CTRL_BREAK_EVENT, or any integer value. The events are a special case
    which work with subprocess console applications which implement a 
    special console control handler. Any other value but those two will 
    cause os.kill to use TerminateProcess, outright killing the process.
    
    This change adds win_console_handler.py, which is a script to implement 
    SetConsoleCtrlHandler and applicable handler function, using ctypes.
    
    subprocess also gets another attribute which is a necessary flag to 
    creationflags in Popen in order to send the CTRL events.
  ........
    r79634 | brian.curtin | 2010-04-02 18:31:28 -0500 (Fri, 02 Apr 2010) | 2 lines
    
    Add note about #1220212 (os.kill on Windows)
  ........
    r79635 | benjamin.peterson | 2010-04-02 18:31:40 -0500 (Fri, 02 Apr 2010) | 1 line
    
    set svn:eol-style on new file
  ........
    r79636 | benjamin.peterson | 2010-04-02 18:59:41 -0500 (Fri, 02 Apr 2010) | 4 lines
    
    always check _PyString_Resize for error
    
    also normalize how this error is checked
  ........
    r79637 | raymond.hettinger | 2010-04-02 19:39:26 -0500 (Fri, 02 Apr 2010) | 1 line
    
    Clear cyclical references in list based OrderedDict.
  ........
    r79639 | benjamin.peterson | 2010-04-02 19:57:33 -0500 (Fri, 02 Apr 2010) | 1 line
    
    more _PyString_Resize error checking
  ........
    r79640 | brian.curtin | 2010-04-02 19:59:32 -0500 (Fri, 02 Apr 2010) | 3 lines
    
    Fix assertRaises usage on reflection functions which should raise
    NotImplementedError on Windows XP and below.
  ........
    r79650 | raymond.hettinger | 2010-04-02 22:14:28 -0500 (Fri, 02 Apr 2010) | 1 line
    
    Improve clear() method.  Keeps key/value refcnts >= 1 until final dict.clear() so that decrefs to zero won't trigger arbitrary code .  Also runs a bit faster.
  ........
    r79651 | raymond.hettinger | 2010-04-03 02:57:09 -0500 (Sat, 03 Apr 2010) | 1 line
    
    Factor-out constant expressions
  ........
    r79654 | victor.stinner | 2010-04-03 03:40:16 -0500 (Sat, 03 Apr 2010) | 5 lines
    
    Issue #8227: Fix C API documentation, argument parsing
    
     * 'z', 'z#', 'z*' does also accept Unicode
     * unify types name: replace "string or Unicode objet" by "string or Unicode"
  ........
    r79673 | brian.curtin | 2010-04-03 08:58:39 -0500 (Sat, 03 Apr 2010) | 2 lines
    
    Add missing return statement in an error condition.
  ........
    r79693 | benjamin.peterson | 2010-04-03 10:38:38 -0500 (Sat, 03 Apr 2010) | 1 line
    
    wrap
  ........
    r79695 | benjamin.peterson | 2010-04-03 10:40:29 -0500 (Sat, 03 Apr 2010) | 1 line
    
    remove unneeded argument
  ........
    r79697 | benjamin.peterson | 2010-04-03 10:44:56 -0500 (Sat, 03 Apr 2010) | 1 line
    
    silence PyCObject warnings in bsddb
  ........
    r79698 | benjamin.peterson | 2010-04-03 10:45:59 -0500 (Sat, 03 Apr 2010) | 1 line
    
    spelling
  ........
    r79704 | benjamin.peterson | 2010-04-03 10:58:15 -0500 (Sat, 03 Apr 2010) | 1 line
    
    remove deprecation warnings silence attempting
  ........
    r79706 | benjamin.peterson | 2010-04-03 11:06:42 -0500 (Sat, 03 Apr 2010) | 1 line
    
    stop CObject deprecation warnings in test___all__
  ........
    r79707 | mark.dickinson | 2010-04-03 11:41:20 -0500 (Sat, 03 Apr 2010) | 5 lines
    
    Ensure 'module removed' warning messages contain the word 'module' or 'package'.
    
    This should fix the test_py3kwarn failure on OS X.  test_support.import_module
    also requires this.
  ........
    r79710 | mark.dickinson | 2010-04-03 11:54:02 -0500 (Sat, 03 Apr 2010) | 1 line
    
    Replace backquotes with repr(), to silence a SyntaxWarning.
  ........
    r79712 | raymond.hettinger | 2010-04-03 12:10:05 -0500 (Sat, 03 Apr 2010) | 1 line
    
    Silence a compiler warning.
  ........
    r79713 | raymond.hettinger | 2010-04-03 13:10:37 -0500 (Sat, 03 Apr 2010) | 1 line
    
    Add count() method to collections.deque().
  ........
    r79714 | mark.dickinson | 2010-04-03 13:17:54 -0500 (Sat, 03 Apr 2010) | 1 line
    
    Silence DeprecationWarnings from uses of has_key and <> in plat-mac.
  ........
    r79718 | antoine.pitrou | 2010-04-03 16:42:18 -0500 (Sat, 03 Apr 2010) | 3 lines
    
    Remove useless (?) import from r79706
  ........
    r79719 | benjamin.peterson | 2010-04-03 16:50:40 -0500 (Sat, 03 Apr 2010) | 1 line
    
    import bsddb more robustly
  ........
    r79722 | raymond.hettinger | 2010-04-03 17:34:15 -0500 (Sat, 03 Apr 2010) | 1 line
    
    Expand test coverage for deque.count().
  ........
    r79730 | raymond.hettinger | 2010-04-03 20:24:59 -0500 (Sat, 03 Apr 2010) | 1 line
    
    Issue 5479:  Add functools.total_ordering class decorator.
  ........
    r79744 | raymond.hettinger | 2010-04-04 02:33:46 -0500 (Sun, 04 Apr 2010) | 1 line
    
    Documentation nit
  ........
    r79750 | raymond.hettinger | 2010-04-04 13:34:45 -0500 (Sun, 04 Apr 2010) | 1 line
    
    Add functools.CmpToKey()
  ........
    r79755 | raymond.hettinger | 2010-04-04 16:45:01 -0500 (Sun, 04 Apr 2010) | 6 lines
    
    Add tests for cmp_to_key.
    Adopt PEP 8 compliant function name.
    Factor-out existing uses cmp_to_key.
    Update documentation to use internal pointers instead of external resource.
  ........
    r79760 | raymond.hettinger | 2010-04-04 17:24:03 -0500 (Sun, 04 Apr 2010) | 1 line
    
    Add tests for functools.total_ordering.
  ........
    r79769 | benjamin.peterson | 2010-04-04 18:23:22 -0500 (Sun, 04 Apr 2010) | 1 line
    
    fix dis on new style classes #8310
  ........
    r79808 | raymond.hettinger | 2010-04-05 13:53:43 -0500 (Mon, 05 Apr 2010) | 1 line
    
    Classes that override __eq__ also need to define __hash__.
  ........
    r79811 | brian.curtin | 2010-04-05 14:04:23 -0500 (Mon, 05 Apr 2010) | 7 lines
    
    Fix a failing test on an apparently slow Windows buildbot.
    
    On slower Windows machines, waiting 0.1 seconds can sometimes not be 
    enough for a subprocess to start and be ready to accept signals, causing 
    the test to fail. One buildbot is also choking on input()/EOFError so 
    that was changed to not depend on input.
  ........
    r79890 | mark.dickinson | 2010-04-07 05:18:27 -0500 (Wed, 07 Apr 2010) | 4 lines
    
    Use some more interesting test values for (unsigned) long long ctypes tests,
    in the hope of getting more information about the test_ctypes failures
    on Sparc (see issue #8314).
  ........
    r80062 | r.david.murray | 2010-04-13 15:57:40 -0500 (Tue, 13 Apr 2010) | 2 lines
    
    Issue #5277: Fix quote counting when parsing RFC 2231 encoded parameters.
  ........
    r80065 | brian.curtin | 2010-04-13 21:24:24 -0500 (Tue, 13 Apr 2010) | 6 lines
    
    Add a small sleep to let a subprocess start before terminating it.
    
    David Bolen's buildbot isn't know for it's speed, and it seems that we
    may have been trying to kill the subprocess before it was fully 
    initialized. I ran with this change on the bot itself and it seemed to work.
  ........
    r80082 | georg.brandl | 2010-04-14 16:36:49 -0500 (Wed, 14 Apr 2010) | 1 line
    
    #8370: fix module name in backported doc addition.
  ........
    r80086 | brian.curtin | 2010-04-14 19:40:40 -0500 (Wed, 14 Apr 2010) | 9 lines
    
    Fix os.kill tests to be more robust and work with slower machines.
    
    Rather than depending on some sleep value, start up an interpreter
    as a subprocess and communicate with it. Because subprocess pipes 
    can't be read from until EOF and I want to read from them before that,
    use ctypes to peek by using PeekNamedPipe. Once the subprocess has 
    written the message, then it is ready to roll and accept signals.
    After that, kill it.
  ........
    r80171 | antoine.pitrou | 2010-04-18 06:16:24 -0500 (Sun, 18 Apr 2010) | 3 lines
    
    Fix bootstrap after r80166
  ........
    r80189 | victor.stinner | 2010-04-18 13:22:25 -0500 (Sun, 18 Apr 2010) | 1 line
    
    Revert r80166 (and r80171), restore Lib/platform.py. subprocess cannot be used in platform.py
  ........
    r80242 | ronald.oussoren | 2010-04-20 03:53:12 -0500 (Tue, 20 Apr 2010) | 8 lines
    
    The PythonLauncher change is needed due
    to changes in how the BASECFLAGS and CFLAGS
    variables get filled by configure.
    
    The Mac/Makefile.in change ensures that 
    pythonw gets build with the rigth deployment
    targets.
  ........
    r80243 | ronald.oussoren | 2010-04-20 03:54:48 -0500 (Tue, 20 Apr 2010) | 3 lines
    
    This patch fixes the handling of a weak-linked
    variable and should fix issue #8095.
  ........
    r80374 | antoine.pitrou | 2010-04-22 12:44:41 -0500 (Thu, 22 Apr 2010) | 3 lines
    
    Temporarily display OpenSSL version in verbose run -- trying to diagnose failure on Neal's buildbot.
  ........
    r80411 | florent.xicluna | 2010-04-23 12:59:10 -0500 (Fri, 23 Apr 2010) | 2 lines
    
    Remove ImportWarnings filters.  They become obsolete after r79310, issue #8205.
  ........
    r80412 | florent.xicluna | 2010-04-23 13:10:12 -0500 (Fri, 23 Apr 2010) | 2 lines
    
    Fix the "regrtest -s" switch.
  ........
    r80445 | brian.curtin | 2010-04-24 12:10:22 -0500 (Sat, 24 Apr 2010) | 2 lines
    
    Fix #5774. Some _winreg functions are documented as taking kwargs but don't.
  ........
    r80579 | nick.coghlan | 2010-04-28 09:34:30 -0500 (Wed, 28 Apr 2010) | 1 line
    
    Also mention patch submitter's name in NEWS, not just in the commit message
  ........
    r80580 | nick.coghlan | 2010-04-28 09:51:08 -0500 (Wed, 28 Apr 2010) | 1 line
    
    Issue 8202: when using the -m command line switch, sys.argv[0] is now '-m' instead of '-c' while searching for the module to be executed
  ........
    r80581 | nick.coghlan | 2010-04-28 09:53:59 -0500 (Wed, 28 Apr 2010) | 1 line
    
    Add version changed note for -m tinkering with sys.argv[0] during the search process
  ........
    r80700 | ronald.oussoren | 2010-05-02 04:55:57 -0500 (Sun, 02 May 2010) | 3 lines
    
    Small update to r80698 to ensure that webbrowser.open
    uses the default browser.
  ........
    r80737 | brett.cannon | 2010-05-03 18:57:15 -0500 (Mon, 03 May 2010) | 4 lines
    
    Fix two potential uninitialization errors and an unneeded assignment.
    
    Found using Clang's static analyzer.
  ........
    r80745 | brett.cannon | 2010-05-03 20:04:53 -0500 (Mon, 03 May 2010) | 5 lines
    
    Mention the code clean-up thanks to Clang's static analyzer in Modules.
    
    Was not applied to modules that will not compile under OS X, dbmmodule.c,
    getaddrinfo.c, and getnameinfo.c.
  ........
    r80752 | victor.stinner | 2010-05-04 06:35:36 -0500 (Tue, 04 May 2010) | 4 lines
    
    _pyio: Fix TextIOWrapper constructor: os has no device_encoding() function
    
    _io module doesn't call this function which was introduced in Python3.
  ........
    r80794 | barry.warsaw | 2010-05-05 11:17:22 -0500 (Wed, 05 May 2010) | 1 line
    
    NEWS
  ........
    r80813 | brett.cannon | 2010-05-05 15:20:19 -0500 (Wed, 05 May 2010) | 4 lines
    
    Remove three unneeded variable assignments.
    
    Found using Clang's static analyzer.
  ........
    r80814 | brett.cannon | 2010-05-05 15:24:30 -0500 (Wed, 05 May 2010) | 1 line
    
    Partially revert the over-reaching r80813.
  ........
    r80823 | brett.cannon | 2010-05-05 15:54:53 -0500 (Wed, 05 May 2010) | 1 line
    
    Mention how Clang's static anaylzer was run over Objects/ and Python/.
  ........
    r80963 | ronald.oussoren | 2010-05-08 03:44:37 -0500 (Sat, 08 May 2010) | 4 lines
    
    Fix for issue #7724: make it possible to build using
    the OSX 10.4u SDK on MacOSX 10.6 by honoring the specified
    SDK when looking for files.
  ........
    r80982 | antoine.pitrou | 2010-05-08 10:23:57 -0500 (Sat, 08 May 2010) | 3 lines
    
    Revert r80963 - it broke compilation everywhere
  ........
    r81012 | gregory.p.smith | 2010-05-08 18:38:49 -0500 (Sat, 08 May 2010) | 2 lines
    
    Fixes [issue7245] Better Ctrl-C support in pdb.
  ........
    r81013 | gregory.p.smith | 2010-05-08 20:20:20 -0500 (Sat, 08 May 2010) | 5 lines
    
    Revert r81012.  buildbot problems and its questionable of me to even
    add this to trunk while we're on the way to 2.7rc1.
    
    When fixed this can go into py3k first.  Sorry.
  ........
    r81030 | antoine.pitrou | 2010-05-09 10:15:11 -0500 (Sun, 09 May 2010) | 3 lines
    
    Fixup indentation of PC/_msi.c
  ........
    r81035 | antoine.pitrou | 2010-05-09 13:32:48 -0500 (Sun, 09 May 2010) | 3 lines
    
    Remove spurious newlines, and add version number.
  ........
    r81088 | brian.curtin | 2010-05-11 14:13:13 -0500 (Tue, 11 May 2010) | 13 lines
    
    #8575 - Update and reorganize some _winreg contents.
    
    I've removed the hopeful note about a future higher-level module since
    it's been in there for quite a long time and nothing of the sort has 
    come up. There are a few places where markup was added to cross-reference
    other sections, and many of the external links have been removed and now
    point to newly created sections containing previously undocumented
    information.
    
    The Value Types section was created and it's contents were taken from
    a function-specific area, since it applies to more than just that
    function. It fits in better with the other newly documented constants.
  ........
    r81140 | florent.xicluna | 2010-05-13 12:05:29 -0500 (Thu, 13 May 2010) | 3 lines
    
    Add sensible information about the OS X platform to diagnose issue #8423:
    test_pep277 fails on "x86 Tiger" buildbot but not on "PPC Tiger".
  ........
    r81141 | florent.xicluna | 2010-05-13 13:16:06 -0500 (Thu, 13 May 2010) | 2 lines
    
    Revert the additional OS X information (r81140).  Keep the endianness information.
  ........
    r81149 | florent.xicluna | 2010-05-13 16:40:01 -0500 (Thu, 13 May 2010) | 2 lines
    
    Better test skipping, with message in the log.
  ........
    r81150 | florent.xicluna | 2010-05-13 16:41:05 -0500 (Thu, 13 May 2010) | 2 lines
    
    Improve test feedback to troubleshoot issue #8423 on OS X.
  ........
    r81151 | florent.xicluna | 2010-05-13 18:46:48 -0500 (Thu, 13 May 2010) | 2 lines
    
    Revert changeset r81150 which helped diagnose issue #8423 on some OS X buildbot.
  ........
    r81275 | antoine.pitrou | 2010-05-17 14:56:59 -0500 (Mon, 17 May 2010) | 4 lines
    
    Issue #7079: Fix a possible crash when closing a file object while using
    it from another thread.  Patch by Daniel Stutzbach.
  ........
    r81484 | georg.brandl | 2010-05-22 16:26:21 -0500 (Sat, 22 May 2010) | 1 line
    
    NEWS rewrap and punctuation consistency.
  ........
    r81502 | georg.brandl | 2010-05-24 16:37:54 -0500 (Mon, 24 May 2010) | 1 line
    
    #8616: update module name
  ........
    r81517 | brian.curtin | 2010-05-25 10:06:15 -0500 (Tue, 25 May 2010) | 5 lines
    
    Fix #2810 - handle the case where some registry calls return
    ERROR_MORE_DATA, requiring another call to get the remaining data.
    
    Patch by Daniel Stutzbach
  ........
    r81561 | alexander.belopolsky | 2010-05-26 15:57:04 -0500 (Wed, 26 May 2010) | 1 line
    
    Added Andrej Krpic. (Thanks for issue #7879 patch.)
  ........
    r81705 | r.david.murray | 2010-06-04 14:51:06 -0500 (Fri, 04 Jun 2010) | 9 lines
    
    #4487: have Charset check with codecs for possible aliases.
    
    Previously, unexpected results occurred when email was passed, for example,
    'utf8' as a charset name, since email would accept it but would *not* use
    the 'utf-8' codec for it, even though Python itself recognises that as
    an alias for utf-8.  Now Charset checks with codecs for aliases as well
    as its own internal table.  Issue 8898 has been opened to change this
    further in py3k so that all aliasing is routed through the codecs module.
  ........
    r81847 | brian.curtin | 2010-06-08 16:15:06 -0500 (Tue, 08 Jun 2010) | 3 lines
    
    Move a variable declration outside of a loop to match what was
    done in r81843 for py3k.
  ........
    r81919 | vinay.sajip | 2010-06-11 17:56:50 -0500 (Fri, 11 Jun 2010) | 1 line
    
    Issue #8924: logging: Improved error handling for Unicode in exception text.
  ........
    r81964 | nick.coghlan | 2010-06-13 01:50:39 -0500 (Sun, 13 Jun 2010) | 1 line
    
    Revert r80580 due to some unintended side effects. See issue #8202 for details.
  ........
    r82056 | florent.xicluna | 2010-06-17 15:30:56 -0500 (Thu, 17 Jun 2010) | 2 lines
    
    Add few words about test.test_genericpath.CommonTest
  ........
    r82136 | benjamin.peterson | 2010-06-21 10:56:45 -0500 (Mon, 21 Jun 2010) | 1 line
    
    update pydoc-topics
  ........
    r82137 | benjamin.peterson | 2010-06-21 10:57:57 -0500 (Mon, 21 Jun 2010) | 1 line
    
    bump verson to 2.7rc2
  ........
    r82142 | benjamin.peterson | 2010-06-21 12:28:30 -0500 (Mon, 21 Jun 2010) | 1 line
    
    towards final release
  ........
    r82149 | ronald.oussoren | 2010-06-22 04:18:28 -0500 (Tue, 22 Jun 2010) | 11 lines
    
    Fix for issue8446:
    
    * Don't import 'ic' in webbrowser, that module is no longer used
    * Remove 'MacOS' from the list of modules that should emit a Py3kWarning on import.
      This is needed because one of the earlier tests triggers and import of this
      extension, and that causes a failure in test_py3kwarn (running test_py3kwarn 
      separately worked fine)
    
    With these changes 'make tests' no longer says that test_py3kwarn fails.
  ........
................
  r82293 | benjamin.peterson | 2010-06-27 22:54:28 +0200 (Sun, 27 Jun 2010) | 13 lines
  
  Merged revisions 80712,81651 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r80712 | vinay.sajip | 2010-05-03 10:11:53 -0500 (Mon, 03 May 2010) | 1 line
    
    Issue #8576: logging updated to remove usage of find_unused_port().
  ........
    r81651 | vinay.sajip | 2010-06-02 05:05:31 -0500 (Wed, 02 Jun 2010) | 1 line
    
    Logging: improved error reporting for BaseConfigurator.resolve().
  ........
................
  r82294 | benjamin.peterson | 2010-06-27 23:05:30 +0200 (Sun, 27 Jun 2010) | 174 lines
  
  Blocked revisions 80731-80736,80739,80741-80744,80746-80747,80807-80812,80815-80822,80843,80874 via svnmerge
  
  ........
    r80731 | brett.cannon | 2010-05-03 18:41:23 -0500 (Mon, 03 May 2010) | 4 lines
    
    Remove unneeded variable initialization.
    
    Found using Clang's static analyzer.
  ........
    r80732 | brett.cannon | 2010-05-03 18:41:51 -0500 (Mon, 03 May 2010) | 4 lines
    
    Remove an unused variable.
    
    Found using Clang's static analyzer.
  ........
    r80733 | brett.cannon | 2010-05-03 18:42:40 -0500 (Mon, 03 May 2010) | 3 lines
    
    Clean up whitespace and remove unneeded variable initialization as found by
    Clang.
  ........
    r80734 | brett.cannon | 2010-05-03 18:43:49 -0500 (Mon, 03 May 2010) | 4 lines
    
    Remove unneeded variable mutation and initializations.
    
    Found using Clang's static analyzer.
  ........
    r80735 | brett.cannon | 2010-05-03 18:44:54 -0500 (Mon, 03 May 2010) | 4 lines
    
    Remove unneeded variable assignment.
    
    Found using Clang's static analyzer.
  ........
    r80736 | brett.cannon | 2010-05-03 18:51:28 -0500 (Mon, 03 May 2010) | 4 lines
    
    Remove unused variables and a variable initialization.
    
    Found using Clang's static analyzer.
  ........
    r80739 | brett.cannon | 2010-05-03 19:36:00 -0500 (Mon, 03 May 2010) | 4 lines
    
    Prevent a possible NULL de-reference and an unneeded variable assignment.
    
    Found using Clang's static analyzer.
  ........
    r80741 | brett.cannon | 2010-05-03 19:48:11 -0500 (Mon, 03 May 2010) | 4 lines
    
    Fix a Py_DECREF to a Py_XDECREF.
    
    Found using Clang's static analyzer.
  ........
    r80742 | brett.cannon | 2010-05-03 19:52:41 -0500 (Mon, 03 May 2010) | 4 lines
    
    Strip out extraneous whitespace, cast a some `const char *` to `void *` when
    passed to free() and make a `char *` to a `const char *` as found by Clang's
    static analyzer.
  ........
    r80743 | brett.cannon | 2010-05-03 19:57:44 -0500 (Mon, 03 May 2010) | 4 lines
    
    Remove an unneeded variable assignment.
    
    Found using Clang's static analyzer.
  ........
    r80744 | brett.cannon | 2010-05-03 20:01:00 -0500 (Mon, 03 May 2010) | 2 lines
    
    Fix some whitespace.
  ........
    r80746 | brett.cannon | 2010-05-03 20:16:51 -0500 (Mon, 03 May 2010) | 4 lines
    
    Remove an unneeded variable and assignment.
    
    Found using Clang's static analyzer.
  ........
    r80747 | brett.cannon | 2010-05-03 20:23:36 -0500 (Mon, 03 May 2010) | 4 lines
    
    Pull a NULL pointer check up to cover more cases in the function.
    
    Found using Clang's static analyzer.
  ........
    r80807 | brett.cannon | 2010-05-05 15:14:27 -0500 (Wed, 05 May 2010) | 4 lines
    
    Remove an unnecessary variable.
    
    Found using Clang's static analyzer.
  ........
    r80808 | brett.cannon | 2010-05-05 15:15:14 -0500 (Wed, 05 May 2010) | 2 lines
    
    Remove extraneous whitespace.
  ........
    r80809 | brett.cannon | 2010-05-05 15:16:09 -0500 (Wed, 05 May 2010) | 4 lines
    
    Remove an unneeded variable increment.
    
    Found using Clang's static analyzer.
  ........
    r80810 | brett.cannon | 2010-05-05 15:16:50 -0500 (Wed, 05 May 2010) | 4 lines
    
    Remove an unneeded variable.
    
    Found using Clang's static analyzer.
  ........
    r80811 | brett.cannon | 2010-05-05 15:18:23 -0500 (Wed, 05 May 2010) | 4 lines
    
    Remove an unneeded variable assignment.
    
    Found using Clang's static analyzer.
  ........
    r80812 | brett.cannon | 2010-05-05 15:19:26 -0500 (Wed, 05 May 2010) | 4 lines
    
    Remove an unneeded assignment.
    
    Found using Clang's static analyzer.
  ........
    r80815 | brett.cannon | 2010-05-05 15:25:47 -0500 (Wed, 05 May 2010) | 1 line
    
    Fix whitespace.
  ........
    r80816 | brett.cannon | 2010-05-05 15:26:58 -0500 (Wed, 05 May 2010) | 4 lines
    
    Remove an unneeded variable and fix a little whitespace.
    
    Found using Clang's static analyzer.
  ........
    r80817 | brett.cannon | 2010-05-05 15:27:49 -0500 (Wed, 05 May 2010) | 4 lines
    
    Remove an unneeded variable increment.
    
    Found using Clang's static analyzer.
  ........
    r80818 | brett.cannon | 2010-05-05 15:30:30 -0500 (Wed, 05 May 2010) | 8 lines
    
    Remove two unneeded branches to an 'if' statement by applying De Morgan's Law
    and creating a single 'if' statement along with a NULL default value for a
    variable.
    
    Also clean up a bunch of whitespace.
    
    Found using Clang's static analyzer.
  ........
    r80819 | brett.cannon | 2010-05-05 15:35:24 -0500 (Wed, 05 May 2010) | 4 lines
    
    Remove an unnecessary variable assignment.
    
    Found using Clang's static analyzer.
  ........
    r80820 | brett.cannon | 2010-05-05 15:38:52 -0500 (Wed, 05 May 2010) | 4 lines
    
    Change to a Py_XDECREF and fix some whitespace.
    
    Found using Clang's static analyzer.
  ........
    r80821 | brett.cannon | 2010-05-05 15:50:03 -0500 (Wed, 05 May 2010) | 4 lines
    
    Explicitly add stdio.h and string.h to make strtod.c work standalone.
    
    Found using Clang's static analyzer.
  ........
    r80822 | brett.cannon | 2010-05-05 15:53:20 -0500 (Wed, 05 May 2010) | 4 lines
    
    Remove an unneeded variable assignment.
    
    Found using Clang's static analyzer.
  ........
    r80843 | brett.cannon | 2010-05-05 18:11:08 -0500 (Wed, 05 May 2010) | 1 line
    
    Revert a change where an expression is not needed now, but could be in the future.
  ........
    r80874 | brett.cannon | 2010-05-06 12:56:36 -0500 (Thu, 06 May 2010) | 5 lines
    
    Add back in a line that was unneeded which advanced a pointer, but commented
    out as it is currently unneeded.
    
    This effectively adds back in the line removed in r80809 as a comment.
  ........
................
  r82295 | benjamin.peterson | 2010-06-27 23:15:09 +0200 (Sun, 27 Jun 2010) | 16 lines
  
  Blocked revisions 79925-79926,80397 via svnmerge
  
  ........
    r79925 | nick.coghlan | 2010-04-10 09:24:36 -0500 (Sat, 10 Apr 2010) | 1 line
    
    Try to turn some buildbots green by allowing test_multiprocessing to pass even if it hits the sys.exc_clear code in the threading module, and improve the test coverage by making the ctypes dependencies a bit more granular (two of the cited ctypes objects don't exist on my system)
  ........
    r79926 | nick.coghlan | 2010-04-10 10:01:54 -0500 (Sat, 10 Apr 2010) | 1 line
    
    Fix typo in comment
  ........
    r80397 | sean.reifschneider | 2010-04-23 03:38:24 -0500 (Fri, 23 Apr 2010) | 2 lines
    
    Small comment documentation change to clarify "ident" selection.
  ........
................
  r82296 | benjamin.peterson | 2010-06-27 23:22:34 +0200 (Sun, 27 Jun 2010) | 12 lines
  
  Blocked revisions 80649,80710 via svnmerge
  
  ........
    r80649 | tarek.ziade | 2010-04-30 07:15:12 -0500 (Fri, 30 Apr 2010) | 1 line
    
    Fixed #8577. distutils.sysconfig.get_python_inc() now differenciates buildir and srcdir
  ........
    r80710 | brett.cannon | 2010-05-02 16:51:23 -0500 (Sun, 02 May 2010) | 1 line
    
    Update the Vim syntax highlight file.
  ........
................
  r82297 | benjamin.peterson | 2010-06-27 23:27:49 +0200 (Sun, 27 Jun 2010) | 16 lines
  
  Blocked revisions 80787,80793 via svnmerge
  
  ........
    r80787 | ronald.oussoren | 2010-05-05 10:32:39 -0500 (Wed, 05 May 2010) | 5 lines
    
    Force exit using os._exit instead of sys.exit,
    this makes sure that the child does not continue
    testing.
  ........
    r80793 | barry.warsaw | 2010-05-05 11:15:09 -0500 (Wed, 05 May 2010) | 5 lines
    
    Bug 7755: audiotest.au is arguably copyrighted material, but definitely makes
    Debian unhappy.  The actual contents of the audio clip are unimportant, so
    replace it with something that we know is okay.  Guido likes woodpeckers.
  ........
................
  r82298 | benjamin.peterson | 2010-06-27 23:45:24 +0200 (Sun, 27 Jun 2010) | 37 lines
  
  Merged revisions 81465-81466,81468,81679,81735,81760,81868,82183 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r81465 | georg.brandl | 2010-05-22 06:29:19 -0500 (Sat, 22 May 2010) | 2 lines
    
    Issue #3924: Ignore cookies with invalid "version" field in cookielib.
  ........
    r81466 | georg.brandl | 2010-05-22 06:31:16 -0500 (Sat, 22 May 2010) | 1 line
    
    Underscore the name of an internal utility function.
  ........
    r81468 | georg.brandl | 2010-05-22 06:43:25 -0500 (Sat, 22 May 2010) | 1 line
    
    #8635: document enumerate() start parameter in docstring.
  ........
    r81679 | benjamin.peterson | 2010-06-03 16:21:03 -0500 (Thu, 03 Jun 2010) | 1 line
    
    use a set for membership testing
  ........
    r81735 | michael.foord | 2010-06-05 06:46:59 -0500 (Sat, 05 Jun 2010) | 1 line
    
    Extract error message truncating into a method (unittest.TestCase._truncateMessage).
  ........
    r81760 | michael.foord | 2010-06-05 14:38:42 -0500 (Sat, 05 Jun 2010) | 1 line
    
    Issue 8302. SkipTest exception is setUpClass or setUpModule is now reported as a skip rather than an error.
  ........
    r81868 | benjamin.peterson | 2010-06-09 14:45:04 -0500 (Wed, 09 Jun 2010) | 1 line
    
    fix code formatting
  ........
    r82183 | benjamin.peterson | 2010-06-23 15:29:26 -0500 (Wed, 23 Jun 2010) | 1 line
    
    cpython only gc tests
  ........
................
  r82299 | benjamin.peterson | 2010-06-27 23:48:35 +0200 (Sun, 27 Jun 2010) | 13 lines
  
  Merged revisions 81881-81882 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r81881 | andrew.kuchling | 2010-06-10 19:16:08 -0500 (Thu, 10 Jun 2010) | 1 line
    
    #5753: update demo.c to use PySys_SetArgvEx(), and add a comment
  ........
    r81882 | andrew.kuchling | 2010-06-10 19:23:01 -0500 (Thu, 10 Jun 2010) | 1 line
    
    #5753: Suggest PySys_SetArgvEx() instead of PySys_SetArgv()
  ........
................
  r82300 | brett.cannon | 2010-06-27 23:49:22 +0200 (Sun, 27 Jun 2010) | 4 lines
  
  Add a link to PEP 3147 from the importlib docs.
  
  Closes issue 8667. Thanks Ashley Sands for the patch.
................
  r82301 | benjamin.peterson | 2010-06-28 00:32:30 +0200 (Mon, 28 Jun 2010) | 303 lines
  
  Merged revisions 80605-80609,80642-80646,80651-80652,80674,80684-80686,80748,80852,80854,80870,80872-80873,80907,80915-80916,80951-80952,80976-80977,80985,81038-81040,81042,81053,81070,81104-81105,81114,81125,81245,81285,81402,81463,81516,81562-81563,81567,81593,81635,81680-81681,81684,81801,81888,81931-81933,81939-81942,81963,81984,81991,82120,82188,82264-82267 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r80605 | andrew.kuchling | 2010-04-28 19:22:16 -0500 (Wed, 28 Apr 2010) | 1 line
    
    Add various items
  ........
    r80606 | andrew.kuchling | 2010-04-28 20:44:30 -0500 (Wed, 28 Apr 2010) | 6 lines
    
    Fix doubled 'the'.
    Markup fixes to use :exc:, :option: in a few places.
      (Glitch: unittest.main's -c ends up a link to the Python
      interpreter's -c option.  Should we skip using :option: for that
      switch, or disable the auto-linking somehow?)
  ........
    r80607 | andrew.kuchling | 2010-04-28 20:45:41 -0500 (Wed, 28 Apr 2010) | 1 line
    
    Add various unittest items
  ........
    r80608 | benjamin.peterson | 2010-04-28 22:18:05 -0500 (Wed, 28 Apr 2010) | 1 line
    
    update pypy description
  ........
    r80609 | benjamin.peterson | 2010-04-28 22:30:59 -0500 (Wed, 28 Apr 2010) | 1 line
    
    update pypy url
  ........
    r80642 | andrew.kuchling | 2010-04-29 19:49:09 -0500 (Thu, 29 Apr 2010) | 1 line
    
    Always add space after RFC; reword paragraph
  ........
    r80643 | andrew.kuchling | 2010-04-29 19:52:31 -0500 (Thu, 29 Apr 2010) | 6 lines
    
    Reword paragraph to make its meaning clearer.
    
    Antoine Pitrou: is my version of the paragraph still correct?
    
    R. David Murray: is this more understandable than the previous version?
  ........
    r80644 | andrew.kuchling | 2010-04-29 20:02:15 -0500 (Thu, 29 Apr 2010) | 1 line
    
    Fix typos
  ........
    r80645 | andrew.kuchling | 2010-04-29 20:32:47 -0500 (Thu, 29 Apr 2010) | 1 line
    
    Markup fix; clarify by adding 'in that order'
  ........
    r80646 | andrew.kuchling | 2010-04-29 20:33:40 -0500 (Thu, 29 Apr 2010) | 1 line
    
    Add various items; rearrange unittest section a bit
  ........
    r80651 | andrew.kuchling | 2010-04-30 08:46:55 -0500 (Fri, 30 Apr 2010) | 1 line
    
    Minor grammar re-wording
  ........
    r80652 | andrew.kuchling | 2010-04-30 08:47:34 -0500 (Fri, 30 Apr 2010) | 1 line
    
    Add item
  ........
    r80674 | andrew.kuchling | 2010-04-30 20:19:16 -0500 (Fri, 30 Apr 2010) | 1 line
    
    Add various items
  ........
    r80684 | andrew.kuchling | 2010-05-01 07:05:52 -0500 (Sat, 01 May 2010) | 1 line
    
    Minor grammar fix
  ........
    r80685 | andrew.kuchling | 2010-05-01 07:06:51 -0500 (Sat, 01 May 2010) | 1 line
    
    Describe memoryview
  ........
    r80686 | antoine.pitrou | 2010-05-01 07:16:39 -0500 (Sat, 01 May 2010) | 4 lines
    
    Fix attribution. Travis didn't do much and he did a bad work.
    (yes, this is a sensitive subject, sorry)
  ........
    r80748 | andrew.kuchling | 2010-05-03 20:24:22 -0500 (Mon, 03 May 2010) | 1 line
    
    Add some more items; the urlparse change is added twice
  ........
    r80852 | andrew.kuchling | 2010-05-05 20:09:47 -0500 (Wed, 05 May 2010) | 1 line
    
    Reword paragraph; fix filename, which should be pyconfig.h
  ........
    r80854 | andrew.kuchling | 2010-05-05 20:10:56 -0500 (Wed, 05 May 2010) | 1 line
    
    Add various items
  ........
    r80870 | andrew.kuchling | 2010-05-06 09:14:09 -0500 (Thu, 06 May 2010) | 1 line
    
    Describe ElementTree 1.3; rearrange new-module sections; describe dict views as sets; small edits and items
  ........
    r80872 | andrew.kuchling | 2010-05-06 12:21:59 -0500 (Thu, 06 May 2010) | 1 line
    
    Add 2 items; record ideas for two initial sections; clarify wording
  ........
    r80873 | andrew.kuchling | 2010-05-06 12:27:57 -0500 (Thu, 06 May 2010) | 1 line
    
    Change section title; point to unittest2
  ........
    r80907 | andrew.kuchling | 2010-05-06 20:45:14 -0500 (Thu, 06 May 2010) | 1 line
    
    Add a new section on the development plan; add an item
  ........
    r80915 | antoine.pitrou | 2010-05-07 05:15:51 -0500 (Fri, 07 May 2010) | 3 lines
    
    Fix some markup and a class name. Also, wrap a long line.
  ........
    r80916 | andrew.kuchling | 2010-05-07 06:30:47 -0500 (Fri, 07 May 2010) | 1 line
    
    Re-word text
  ........
    r80951 | andrew.kuchling | 2010-05-07 20:15:26 -0500 (Fri, 07 May 2010) | 1 line
    
    Add two items
  ........
    r80952 | andrew.kuchling | 2010-05-07 20:35:55 -0500 (Fri, 07 May 2010) | 1 line
    
    Get accents correct
  ........
    r80976 | andrew.kuchling | 2010-05-08 08:28:03 -0500 (Sat, 08 May 2010) | 1 line
    
    Add logging.dictConfig example; give up on writing a Ttk example
  ........
    r80977 | andrew.kuchling | 2010-05-08 08:29:46 -0500 (Sat, 08 May 2010) | 1 line
    
    Markup fixes
  ........
    r80985 | andrew.kuchling | 2010-05-08 10:39:46 -0500 (Sat, 08 May 2010) | 7 lines
    
    Write summary of the 2.7 release; rewrite the future section some more;
    mention PYTHONWARNINGS env. var; tweak some examples for readability.
    
    And with this commit, the "What's New" is done... except for a
    complete read-through to polish the text, and fixing any reported errors,
    but those tasks can easily wait until after beta2.
  ........
    r81038 | benjamin.peterson | 2010-05-09 16:09:40 -0500 (Sun, 09 May 2010) | 1 line
    
    finish clause
  ........
    r81039 | andrew.kuchling | 2010-05-10 09:18:27 -0500 (Mon, 10 May 2010) | 1 line
    
    Markup fix; re-word a sentence
  ........
    r81040 | andrew.kuchling | 2010-05-10 09:20:12 -0500 (Mon, 10 May 2010) | 1 line
    
    Use title case
  ........
    r81042 | andrew.kuchling | 2010-05-10 10:03:35 -0500 (Mon, 10 May 2010) | 1 line
    
    Link to unittest2 article
  ........
    r81053 | florent.xicluna | 2010-05-10 14:59:22 -0500 (Mon, 10 May 2010) | 2 lines
    
    Add a link on maketrans().
  ........
    r81070 | andrew.kuchling | 2010-05-10 18:13:41 -0500 (Mon, 10 May 2010) | 1 line
    
    Fix typo
  ........
    r81104 | andrew.kuchling | 2010-05-11 19:38:44 -0500 (Tue, 11 May 2010) | 1 line
    
    Revision pass: lots of edits, typo fixes, rearrangements
  ........
    r81105 | andrew.kuchling | 2010-05-11 19:40:47 -0500 (Tue, 11 May 2010) | 1 line
    
    Let's call this done
  ........
    r81114 | andrew.kuchling | 2010-05-12 08:56:07 -0500 (Wed, 12 May 2010) | 1 line
    
    Grammar fix
  ........
    r81125 | andrew.kuchling | 2010-05-12 13:56:48 -0500 (Wed, 12 May 2010) | 1 line
    
    #8696: add documentation for logging.config.dictConfig (PEP 391)
  ........
    r81245 | andrew.kuchling | 2010-05-16 18:31:16 -0500 (Sun, 16 May 2010) | 1 line
    
    Add cross-reference to later section
  ........
    r81285 | vinay.sajip | 2010-05-18 03:16:27 -0500 (Tue, 18 May 2010) | 1 line
    
    Fixed minor typo in ReST markup.
  ........
    r81402 | vinay.sajip | 2010-05-21 12:41:34 -0500 (Fri, 21 May 2010) | 1 line
    
    Updated logging documentation with more dictConfig information.
  ........
    r81463 | georg.brandl | 2010-05-22 03:17:23 -0500 (Sat, 22 May 2010) | 1 line
    
    #8785: less confusing description of regex.find*.
  ........
    r81516 | andrew.kuchling | 2010-05-25 08:34:08 -0500 (Tue, 25 May 2010) | 1 line
    
    Add three items
  ........
    r81562 | andrew.kuchling | 2010-05-27 08:22:53 -0500 (Thu, 27 May 2010) | 1 line
    
    Rewrite wxWidgets section
  ........
    r81563 | andrew.kuchling | 2010-05-27 08:30:09 -0500 (Thu, 27 May 2010) | 1 line
    
    Remove top-level 'General Questions' section, pushing up the questions it contains
  ........
    r81567 | andrew.kuchling | 2010-05-27 16:29:59 -0500 (Thu, 27 May 2010) | 1 line
    
    Add item
  ........
    r81593 | georg.brandl | 2010-05-29 03:46:18 -0500 (Sat, 29 May 2010) | 1 line
    
    #8616: add new turtle demo "nim".
  ........
    r81635 | georg.brandl | 2010-06-01 02:25:23 -0500 (Tue, 01 Jun 2010) | 1 line
    
    Put docs for RegexObject.search() before RegexObject.match() to mirror re.search() and re.match() order.
  ........
    r81680 | vinay.sajip | 2010-06-03 17:34:42 -0500 (Thu, 03 Jun 2010) | 1 line
    
    Issue #8890: Documentation changed to avoid reference to temporary files.
  ........
    r81681 | sean.reifschneider | 2010-06-03 20:51:26 -0500 (Thu, 03 Jun 2010) | 2 lines
    
    Issue8810: Clearing up docstring for tzinfo.utcoffset.
  ........
    r81684 | vinay.sajip | 2010-06-04 08:41:02 -0500 (Fri, 04 Jun 2010) | 1 line
    
    Issue #8890: Documentation changed to avoid reference to temporary files - other cases covered.
  ........
    r81801 | andrew.kuchling | 2010-06-07 08:38:40 -0500 (Mon, 07 Jun 2010) | 1 line
    
    #8875: Remove duplicated paragraph
  ........
    r81888 | andrew.kuchling | 2010-06-10 20:54:58 -0500 (Thu, 10 Jun 2010) | 1 line
    
    Add a few more items
  ........
    r81931 | georg.brandl | 2010-06-12 01:26:54 -0500 (Sat, 12 Jun 2010) | 1 line
    
    Fix punctuation.
  ........
    r81932 | georg.brandl | 2010-06-12 01:28:58 -0500 (Sat, 12 Jun 2010) | 1 line
    
    Document that an existing directory raises in mkdir().
  ........
    r81933 | georg.brandl | 2010-06-12 01:45:33 -0500 (Sat, 12 Jun 2010) | 1 line
    
    Update version in README.
  ........
    r81939 | georg.brandl | 2010-06-12 04:45:01 -0500 (Sat, 12 Jun 2010) | 1 line
    
    Use newer toctree syntax.
  ........
    r81940 | georg.brandl | 2010-06-12 04:45:28 -0500 (Sat, 12 Jun 2010) | 1 line
    
    Add document on how to build.
  ........
    r81941 | georg.brandl | 2010-06-12 04:45:58 -0500 (Sat, 12 Jun 2010) | 1 line
    
    Fix gratuitous indentation.
  ........
    r81942 | georg.brandl | 2010-06-12 04:46:03 -0500 (Sat, 12 Jun 2010) | 1 line
    
    Update README.
  ........
    r81963 | andrew.kuchling | 2010-06-12 15:00:55 -0500 (Sat, 12 Jun 2010) | 1 line
    
    Grammar fix
  ........
    r81984 | georg.brandl | 2010-06-14 10:58:39 -0500 (Mon, 14 Jun 2010) | 1 line
    
    #8993: fix reference.
  ........
    r81991 | andrew.kuchling | 2010-06-14 19:38:58 -0500 (Mon, 14 Jun 2010) | 1 line
    
    Add another bunch of items
  ........
    r82120 | andrew.kuchling | 2010-06-20 16:45:45 -0500 (Sun, 20 Jun 2010) | 1 line
    
    Note that Python 3.x isn't covered; add forward ref. for UTF-8; note error in 2.5 and up
  ........
    r82188 | benjamin.peterson | 2010-06-23 19:02:46 -0500 (Wed, 23 Jun 2010) | 1 line
    
    remove reverted changed
  ........
    r82264 | georg.brandl | 2010-06-27 05:47:47 -0500 (Sun, 27 Jun 2010) | 1 line
    
    Confusing punctuation.
  ........
    r82265 | georg.brandl | 2010-06-27 05:49:23 -0500 (Sun, 27 Jun 2010) | 1 line
    
    Use designated syntax for optional grammar element.
  ........
    r82266 | georg.brandl | 2010-06-27 05:51:44 -0500 (Sun, 27 Jun 2010) | 1 line
    
    Fix URL.
  ........
    r82267 | georg.brandl | 2010-06-27 05:55:38 -0500 (Sun, 27 Jun 2010) | 1 line
    
    Two typos.
  ........
................
  r82302 | benjamin.peterson | 2010-06-28 00:37:28 +0200 (Mon, 28 Jun 2010) | 15 lines
  
  Merged revisions 81380 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r81380 | brett.cannon | 2010-05-20 13:37:55 -0500 (Thu, 20 May 2010) | 8 lines
    
    Turned out that if you used explicit relative import syntax
    (e.g. from .os import sep) and it failed, import would still try the implicit
    relative import semantics of an absolute import (from os import sep). That's
    not right, so when level is negative, only do explicit relative import
    semantics.
    
    Fixes issue #7902. Thanks to Meador Inge for the patch.
  ........
................
  r82304 | benjamin.peterson | 2010-06-28 00:41:29 +0200 (Mon, 28 Jun 2010) | 13 lines
  
  Merged revisions 81499,81506 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r81499 | georg.brandl | 2010-05-24 16:29:07 -0500 (Mon, 24 May 2010) | 1 line
    
    #8016: add the CP858 codec (approved by Benjamin).  (Also add CP720 to the tests, it was missing there.)
  ........
    r81506 | benjamin.peterson | 2010-05-24 17:04:53 -0500 (Mon, 24 May 2010) | 1 line
    
    set svn:eol-style
  ........
................
  r82305 | benjamin.peterson | 2010-06-28 00:44:51 +0200 (Mon, 28 Jun 2010) | 9 lines
  
  Merged revisions 82303 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82303 | benjamin.peterson | 2010-06-27 17:40:26 -0500 (Sun, 27 Jun 2010) | 1 line
    
    mark tracking tests as implementation details
  ........
................
  r82306 | benjamin.peterson | 2010-06-28 01:16:51 +0200 (Mon, 28 Jun 2010) | 9 lines
  
  Blocked revisions 81700 via svnmerge
  
  ........
    r81700 | martin.v.loewis | 2010-06-04 14:15:32 -0500 (Fri, 04 Jun 2010) | 2 lines
    
    Display installer warning that Windows 2000 won't 
    be supported in future releases.
  ........
................
  r82307 | benjamin.peterson | 2010-06-28 01:21:13 +0200 (Mon, 28 Jun 2010) | 9 lines
  
  Merged revisions 81880 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r81880 | andrew.kuchling | 2010-06-10 19:14:34 -0500 (Thu, 10 Jun 2010) | 1 line
    
    Edit comments for current Python; bump version number of Python
  ........
................
  r82308 | benjamin.peterson | 2010-06-28 01:26:04 +0200 (Mon, 28 Jun 2010) | 9 lines
  
  Recorded merge of revisions 80866 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r80866 | andrew.kuchling | 2010-05-06 08:03:39 -0500 (Thu, 06 May 2010) | 1 line
    
    Use anonymous hyperlinks
  ........
................
  r82309 | benjamin.peterson | 2010-06-28 01:31:03 +0200 (Mon, 28 Jun 2010) | 8 lines
  
  Blocked revisions 80396 via svnmerge
  
  ........
    r80396 | sean.reifschneider | 2010-04-23 03:31:55 -0500 (Fri, 23 Apr 2010) | 2 lines
    
    issue8451: Making syslog module use sys.argv[0] for "ident".
  ........
................
  r82310 | benjamin.peterson | 2010-06-28 01:34:03 +0200 (Mon, 28 Jun 2010) | 9 lines
  
  Merged revisions 81004 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r81004 | mark.dickinson | 2010-05-08 14:13:21 -0500 (Sat, 08 May 2010) | 1 line
    
    Fix configure bug that was misreporting availability of MacOS X 10.5 SDK as available on Linux but not on OS X.
  ........
................
  r82311 | benjamin.peterson | 2010-06-28 01:49:45 +0200 (Mon, 28 Jun 2010) | 24 lines
  
  Merged revisions 79894-79895,80098,80120 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r79894 | jeroen.ruigrok | 2010-04-07 09:33:37 -0500 (Wed, 07 Apr 2010) | 5 lines
    
    FreeBSD is not a sysv platform, so use freebsd.S instead of sysv.S (as is
    also the case in FreeBSD's port of libffi).
    
    Reviewed by: dickinsm
  ........
    r79895 | jeroen.ruigrok | 2010-04-07 11:34:08 -0500 (Wed, 07 Apr 2010) | 2 lines
    
    Document the libffi FreeBSD fix.
  ........
    r80098 | benjamin.peterson | 2010-04-15 16:42:16 -0500 (Thu, 15 Apr 2010) | 1 line
    
    add space
  ........
    r80120 | antoine.pitrou | 2010-04-16 16:34:02 -0500 (Fri, 16 Apr 2010) | 3 lines
    
    Ignore jinja2
  ........
................
  r82312 | benjamin.peterson | 2010-06-28 01:53:17 +0200 (Mon, 28 Jun 2010) | 8 lines
  
  Blocked revisions 79596 via svnmerge
  
  ........
    r79596 | eric.smith | 2010-04-02 07:30:56 -0500 (Fri, 02 Apr 2010) | 1 line
    
    Issue 7994: Make object.__format__ with a non-empty format string a PendingDecprecationWarning. Still need to remove uses of this from various tests.
  ........
................
  r82313 | brett.cannon | 2010-06-28 01:57:46 +0200 (Mon, 28 Jun 2010) | 29 lines
  
  Implement importlib.abc.SourceLoader and deprecate PyLoader and PyPycLoader.
  
  SourceLoader is a simplification of both PyLoader and PyPycLoader. If one only
  wants to use source, then they need to only implement get_data and
  get_filename. To also use bytecode -- sourceless loading is not supported --
  then two abstract methods -- path_mtime and set_data -- need to be implemented.
  Compared to PyLoader and PyPycLoader, there are less abstract methods
  introduced and bytecode files become an optimization controlled by the ABC and
  hidden from the user (this need came about as PEP 3147 showed that not treating
  bytecode as an optimization can cause problems for compatibility).
  
  PyLoader is deprecated in favor of SourceLoader. To be compatible from Python
  3.1 onwards, a subclass need only use simple methods for source_path and
  is_package. Otherwise conditional subclassing based on whether Python 3.1 or
  Python 3.2 is being is the only change. The documentation and docstring for
  PyLoader explain what is exactly needed.
  
  PyPycLoader is deprecated also in favor of SourceLoader. Because PEP 3147
  shifted bytecode path details so much, there is no foolproof way to provide
  backwards-compatibility with SourceLoader. Because of this the class is simply
  deprecated and users should move to SourceLoader (and optionally PyLoader for
  Python 3.1). This does lead to a loss of support for sourceless loading
  unfortunately.
  
  At some point before Python 3.2 is released, SourceLoader will be moved over to
  importlib._bootstrap so that the core code of importlib relies on the new code
  instead of the old PyPycLoader code. This commit is being done now so that
  there is no issue in having the API in Python 3.1a1.
................
  r82314 | benjamin.peterson | 2010-06-28 02:01:59 +0200 (Mon, 28 Jun 2010) | 32 lines
  
  Merged revisions 77402,77505,77510 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r77402 | brett.cannon | 2010-01-09 20:56:19 -0600 (Sat, 09 Jan 2010) | 12 lines
    
    DeprecationWarning is now silent by default.
    
    This was originally suggested by Guido, discussed on the stdlib-sig mailing
    list, and given the OK by Guido directly to me. What this change essentially
    means is that Python has taken a policy of silencing warnings that are only
    of interest to developers by default. This should prevent users from seeing
    warnings which are triggered by an application being run against a new
    interpreter before the app developer has a chance to update their code.
    
    Closes issue #7319. Thanks to Antoine Pitrou, Ezio Melotti, and Brian Curtin
    for helping with the issue.
  ........
    r77505 | brett.cannon | 2010-01-14 14:00:28 -0600 (Thu, 14 Jan 2010) | 7 lines
    
    The silencing of DeprecationWarning was not taking -3 into consideration. Since
    Py3K warnings are DeprecationWarning by default this was causing -3 to
    essentially be a no-op. Now DeprecationWarning is only silenced if -3 is not
    used.
    
    Closes issue #7700. Thanks Ezio Melotti and Florent Xicluna for patch help.
  ........
    r77510 | brett.cannon | 2010-01-14 19:31:45 -0600 (Thu, 14 Jan 2010) | 1 line
    
    Remove C++/C99-style comments.
  ........
................
  r82315 | benjamin.peterson | 2010-06-28 02:16:12 +0200 (Mon, 28 Jun 2010) | 1 line
  
  fix versionchanged
................
  r82316 | benjamin.peterson | 2010-06-28 02:24:13 +0200 (Mon, 28 Jun 2010) | 1 line
  
  fix test with more obviously incorrect bytecode
................
  r82318 | benjamin.peterson | 2010-06-28 05:07:10 +0200 (Mon, 28 Jun 2010) | 9 lines
  
  Merged revisions 82317 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82317 | benjamin.peterson | 2010-06-27 21:58:25 -0500 (Sun, 27 Jun 2010) | 1 line
    
    remove unused last argument to property_copy
  ........
................
  r82320 | brett.cannon | 2010-06-28 06:57:24 +0200 (Mon, 28 Jun 2010) | 11 lines
  
  Move importlib.abc.SourceLoader to _bootstrap.
  
  Required updating code relying on other modules to switch to _bootstrap's
  unique module requirements. This led to the realization that
  get_code was being too liberal in its exception catching when calling set_data
  by blindly grabbing IOError. Shifted the responsibility of safely ignoring
  writes to a read-only path to set_data.
  
  Importlib is still not relying on SourceLoader yet; requires creating a
  SourcelessLoader and updating the source finder.
................
  r82321 | brett.cannon | 2010-06-28 07:46:25 +0200 (Mon, 28 Jun 2010) | 2 lines
  
  Add an inheritance test for importlib.abc.SourceLoader.
................
  r82326 | senthil.kumaran | 2010-06-28 16:08:00 +0200 (Mon, 28 Jun 2010) | 9 lines
  
  Merged revisions 82324 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82324 | senthil.kumaran | 2010-06-28 19:26:46 +0530 (Mon, 28 Jun 2010) | 3 lines
    
    Fix Issue8653 - Docstring for urlunsplit function.
  ........
................
  r82328 | alexander.belopolsky | 2010-06-28 16:36:55 +0200 (Mon, 28 Jun 2010) | 1 line
  
  Added missing white space
................
  r82331 | benjamin.peterson | 2010-06-28 17:39:55 +0200 (Mon, 28 Jun 2010) | 9 lines
  
  Merged revisions 82330 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82330 | benjamin.peterson | 2010-06-28 10:36:40 -0500 (Mon, 28 Jun 2010) | 1 line
    
    testcapi tests are definitely cpython only
  ........
................
  r82333 | benjamin.peterson | 2010-06-28 17:43:25 +0200 (Mon, 28 Jun 2010) | 9 lines
  
  Merged revisions 82332 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82332 | benjamin.peterson | 2010-06-28 10:41:06 -0500 (Mon, 28 Jun 2010) | 1 line
    
    fix skipping condition
  ........
................
  r82336 | senthil.kumaran | 2010-06-28 19:31:40 +0200 (Mon, 28 Jun 2010) | 9 lines
  
  Merged revisions 82334 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82334 | senthil.kumaran | 2010-06-28 22:37:40 +0530 (Mon, 28 Jun 2010) | 3 lines
    
    Documentation correction. urllib2.urlopen.info returns a mimetools.Message instance not HTTPMessage object.
  ........
................
  r82342 | benjamin.peterson | 2010-06-28 21:43:42 +0200 (Mon, 28 Jun 2010) | 1 line
  
  update error message
................
  r82343 | mark.dickinson | 2010-06-28 21:44:20 +0200 (Mon, 28 Jun 2010) | 23 lines
  
  Merged revisions 82338,82340-82341 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82338 | mark.dickinson | 2010-06-28 20:31:41 +0100 (Mon, 28 Jun 2010) | 9 lines
    
    Fix some shallow bugs in Demo/parser/unparse.py, and add tests:
    
      - insert commas between entries in del statement
      - left and right shifts were represented as >> and << (respectively); reverse
      - unindent properly after for: else:  or while: else:
      - add parens around the result of an unary operation
      - add parens around negative numbers, to avoid turning (-1)**2 into -1**2.
  ........
    r82340 | mark.dickinson | 2010-06-28 20:34:15 +0100 (Mon, 28 Jun 2010) | 1 line
    
    Fix typo in test_unparse.py.
  ........
    r82341 | mark.dickinson | 2010-06-28 20:38:19 +0100 (Mon, 28 Jun 2010) | 1 line
    
    Set svn:eol-style on test_unparse.py.
  ........
................
  r82346 | mark.dickinson | 2010-06-28 22:09:18 +0200 (Mon, 28 Jun 2010) | 9 lines
  
  Merged revisions 82345 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82345 | mark.dickinson | 2010-06-28 20:54:19 +0100 (Mon, 28 Jun 2010) | 1 line
    
    unparse.py:  fix mispaced parentheses in chained comparisons
  ........
................
  r82347 | mark.dickinson | 2010-06-28 23:14:17 +0200 (Mon, 28 Jun 2010) | 18 lines
  
  Update Demo/parser/unparse.py to current Python 3.x syntax.  Additions:
   - relative imports
   - keyword-only arguments
   - function annotations
   - class decorators
   - raise ... from ...
   - except ... as ...
   - nonlocal
   - bytes literals
   - set literals
   - set comprehensions
   - dict comprehensions
  Removals:
   - print statement.
  
  Some of this should be backported to 2.x.
................
  r82349 | mark.dickinson | 2010-06-28 23:39:51 +0200 (Mon, 28 Jun 2010) | 1 line
  
  unparse.py: Typo fix.
................
  r82351 | brian.curtin | 2010-06-29 02:14:28 +0200 (Tue, 29 Jun 2010) | 3 lines
  
  Update md5driver.py for 3.x.
  Changed an import, replaced md5.new() with md5(), and added an encode where needed.
................
  r82354 | benjamin.peterson | 2010-06-29 03:27:35 +0200 (Tue, 29 Jun 2010) | 1 line
  
  rephrase and remove uneeded statement
................
  r82357 | mark.dickinson | 2010-06-29 09:38:37 +0200 (Tue, 29 Jun 2010) | 9 lines
  
  Merged revisions 82356 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82356 | mark.dickinson | 2010-06-29 08:37:25 +0100 (Tue, 29 Jun 2010) | 1 line
    
    Spelling.
  ........
................
  r82358 | mark.dickinson | 2010-06-29 09:48:23 +0200 (Tue, 29 Jun 2010) | 1 line
  
  unparse.py: respect coding cookie in input files
................
  r82359 | mark.dickinson | 2010-06-29 10:52:36 +0200 (Tue, 29 Jun 2010) | 1 line
  
  Add parentheses around numeric literals, to avoid turning 3 .bit_length() into 3.bit_length().
................
  r82360 | mark.dickinson | 2010-06-29 12:01:48 +0200 (Tue, 29 Jun 2010) | 8 lines
  
  More unparse.py fixes:
  
   - parenthesize lambdas, to avoid turning (lambda : int)() into lambda: int()
   - unparse an infinite float literals in the AST as an overflowing finite value
  
  unparse.py now successfully round-trips on all valid Lib/*.py and Lib/test/*.py files.
................
  r82365 | mark.dickinson | 2010-06-29 16:08:23 +0200 (Tue, 29 Jun 2010) | 1 line
  
  test_unparse.py: Do roundtrip testing for all Python files in Lib and Lib/test.
................
  r82370 | vinay.sajip | 2010-06-29 17:31:15 +0200 (Tue, 29 Jun 2010) | 1 line
  
  Added information about pickle security and SocketHandler, and some cross-reference targets.
................
  r82371 | benjamin.peterson | 2010-06-29 17:32:34 +0200 (Tue, 29 Jun 2010) | 8 lines
  
  Blocked revisions 82368 via svnmerge
  
  ........
    r82368 | benjamin.peterson | 2010-06-29 10:18:02 -0500 (Tue, 29 Jun 2010) | 1 line
    
    group cStringIO docs under its factory function
  ........
................
  r82373 | benjamin.peterson | 2010-06-29 17:57:57 +0200 (Tue, 29 Jun 2010) | 1 line
  
  note that byte strings can be raw, too #9114
................
  r82375 | brett.cannon | 2010-06-29 20:26:11 +0200 (Tue, 29 Jun 2010) | 1 line
  
  Make a sentence a little less awkward.
................
  r82376 | benjamin.peterson | 2010-06-29 20:36:39 +0200 (Tue, 29 Jun 2010) | 1 line
  
  update for nonlocal keyword
................
  r82377 | mark.dickinson | 2010-06-29 20:38:59 +0200 (Tue, 29 Jun 2010) | 1 line
  
  Fix typo in unparsing of a class definition.
................
  r82381 | mark.dickinson | 2010-06-29 22:10:42 +0200 (Tue, 29 Jun 2010) | 9 lines
  
  Merged revisions 82379 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82379 | mark.dickinson | 2010-06-29 21:09:12 +0100 (Tue, 29 Jun 2010) | 1 line
    
    Issue #1789:  clarify that the 'size' column in struct docs refers to standard size.
  ........
................
  r82383 | mark.dickinson | 2010-06-30 10:32:11 +0200 (Wed, 30 Jun 2010) | 1 line
  
  Collapse else: if: ... into elif: 
................
  r82384 | mark.dickinson | 2010-06-30 10:46:53 +0200 (Wed, 30 Jun 2010) | 1 line
  
  Output try-except-finally statements where appropriate.
................
  r82386 | mark.dickinson | 2010-06-30 11:05:47 +0200 (Wed, 30 Jun 2010) | 1 line
  
  Better solution for attribute access on integer literals.
................
  r82387 | mark.dickinson | 2010-06-30 11:14:43 +0200 (Wed, 30 Jun 2010) | 1 line
  
  Remove backtick syntax for repr.
................
  r82391 | mark.dickinson | 2010-06-30 13:13:36 +0200 (Wed, 30 Jun 2010) | 1 line
  
  Issue #9011:  Tests for Python 3.2's treatment of negated imaginary literals.
................
  r82392 | mark.dickinson | 2010-06-30 13:14:30 +0200 (Wed, 30 Jun 2010) | 1 line
  
  Revert accidental extra changes included in r82391.
................
  r82393 | mark.dickinson | 2010-06-30 13:45:53 +0200 (Wed, 30 Jun 2010) | 1 line
  
  Unparse infinite imaginary literals correctly.  Add some more numeric tests.
................
  r82394 | michael.foord | 2010-06-30 14:17:50 +0200 (Wed, 30 Jun 2010) | 1 line
  
  Issue 9110. Adding ContextDecorator to contextlib. This enables the creation of APIs that act as decorators as well as context managers. contextlib.contextmanager changed to use ContextDecorator.
................
  r82402 | mark.dickinson | 2010-06-30 18:36:12 +0200 (Wed, 30 Jun 2010) | 12 lines
  
  Blocked revisions 82399-82400 via svnmerge
  
  ........
    r82399 | mark.dickinson | 2010-06-30 15:19:56 +0100 (Wed, 30 Jun 2010) | 1 line
    
    Update Demo/parser directory; backport unparse fixes from py3k.
  ........
    r82400 | mark.dickinson | 2010-06-30 17:27:57 +0100 (Wed, 30 Jun 2010) | 2 lines
    
    Issue #9125:  Update parser module for "except ... as ..." syntax.
  ........
................
  r82405 | benjamin.peterson | 2010-06-30 19:39:45 +0200 (Wed, 30 Jun 2010) | 9 lines
  
  Merged revisions 82403 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82403 | benjamin.peterson | 2010-06-30 12:11:08 -0500 (Wed, 30 Jun 2010) | 1 line
    
    mark test depending on ref counting
  ........
................
  r82407 | giampaolo.rodola | 2010-06-30 19:47:39 +0200 (Wed, 30 Jun 2010) | 9 lines
  
  Merged revisions 82404 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82404 | giampaolo.rodola | 2010-06-30 19:38:28 +0200 (mer, 30 giu 2010) | 1 line
    
    fix issue #6589: cleanup asyncore.socket_map if smtpd.SMTPServer constructor raises an exception
  ........
................
  r82410 | benjamin.peterson | 2010-06-30 20:44:04 +0200 (Wed, 30 Jun 2010) | 9 lines
  
  Merged revisions 82409 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82409 | benjamin.peterson | 2010-06-30 13:41:08 -0500 (Wed, 30 Jun 2010) | 1 line
    
    an AttributeError is perfectly acceptable here
  ........
................
  r82413 | ezio.melotti | 2010-07-01 09:32:02 +0200 (Thu, 01 Jul 2010) | 13 lines
  
  Update PyUnicode_DecodeUTF8 from RFC 2279 to RFC 3629.
  
  1) #8271: when a byte sequence is invalid, only the start byte and all the
     valid continuation bytes are now replaced by U+FFFD, instead of replacing
     the number of bytes specified by the start byte.
     See http://www.unicode.org/versions/Unicode5.2.0/ch03.pdf (pages 94-95);
  2) 5- and 6-bytes-long UTF-8 sequences are now considered invalid (no changes
     in behavior);
  3) Change the error messages "unexpected code byte" to "invalid start byte"
     and "invalid data" to "invalid continuation byte";
  4) Add an extensive set of tests in test_unicode;
  5) Fix test_codeccallbacks because it was failing after this change.
................
  r82416 | benjamin.peterson | 2010-07-01 17:09:14 +0200 (Thu, 01 Jul 2010) | 8 lines
  
  Blocked revisions 82415 via svnmerge
  
  ........
    r82415 | benjamin.peterson | 2010-07-01 10:07:15 -0500 (Thu, 01 Jul 2010) | 1 line
    
    remove docs about delegating to special methods; it does no such thing
  ........
................
  r82417 | benjamin.peterson | 2010-07-01 17:16:55 +0200 (Thu, 01 Jul 2010) | 1 line
  
  correctly lookup __trunc__ and __floor__
................
  r82422 | benjamin.peterson | 2010-07-01 19:49:01 +0200 (Thu, 01 Jul 2010) | 9 lines
  
  Merged revisions 82420 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82420 | benjamin.peterson | 2010-07-01 12:45:52 -0500 (Thu, 01 Jul 2010) | 1 line
    
    fix fixer name
  ........
................
  r82430 | benjamin.peterson | 2010-07-02 01:39:26 +0200 (Fri, 02 Jul 2010) | 8 lines
  
  Blocked revisions 82429 via svnmerge
  
  ........
    r82429 | benjamin.peterson | 2010-07-01 18:35:37 -0500 (Thu, 01 Jul 2010) | 1 line
    
    trunc does use the special method, though
  ........
................
  r82434 | georg.brandl | 2010-07-02 09:41:51 +0200 (Fri, 02 Jul 2010) | 9 lines
  
  Merged revisions 82433 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82433 | georg.brandl | 2010-07-02 09:33:50 +0200 (Fr, 02 Jul 2010) | 1 line
    
    Grammar and markup fixes.
  ........
................
  r82437 | benjamin.peterson | 2010-07-02 15:35:17 +0200 (Fri, 02 Jul 2010) | 1 line
  
  account for different ref counting semantics of _PyObject_LookupSpecial
................
  r82438 | benjamin.peterson | 2010-07-02 15:46:42 +0200 (Fri, 02 Jul 2010) | 1 line
  
  fix lookup of __ceil__
................
  r82440 | nick.coghlan | 2010-07-02 17:57:50 +0200 (Fri, 02 Jul 2010) | 9 lines
  
  Merged revisions 82439 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82439 | nick.coghlan | 2010-07-03 01:50:14 +1000 (Sat, 03 Jul 2010) | 1 line
    
    Fix a couple of minor nits in What's New before the release goes out
  ........
................
  r82442 | mark.dickinson | 2010-07-02 18:05:15 +0200 (Fri, 02 Jul 2010) | 1 line
  
  Style/consistency nit: make math_floor and math_ceil code look the same.
................
  r82443 | nick.coghlan | 2010-07-02 18:14:05 +0200 (Fri, 02 Jul 2010) | 9 lines
  
  Merged revisions 82441 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82441 | nick.coghlan | 2010-07-03 02:01:53 +1000 (Sat, 03 Jul 2010) | 1 line
    
    Issue 8202 actually caused an unintended behavioural change and was reverted in 2.7 - remove reference from What's New
  ........
................
  r82448 | benjamin.peterson | 2010-07-02 21:45:07 +0200 (Fri, 02 Jul 2010) | 9 lines
  
  Merged revisions 82447 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82447 | benjamin.peterson | 2010-07-02 14:41:39 -0500 (Fri, 02 Jul 2010) | 1 line
    
    add space
  ........
................
  r82452 | mark.dickinson | 2010-07-02 22:26:07 +0200 (Fri, 02 Jul 2010) | 9 lines
  
  Merged revisions 82446 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82446 | mark.dickinson | 2010-07-02 19:06:52 +0100 (Fri, 02 Jul 2010) | 1 line
    
    Clarify sys.float_info documentation.
  ........
................
  r82456 | brett.cannon | 2010-07-03 00:03:00 +0200 (Sat, 03 Jul 2010) | 7 lines
  
  Some people have mistaken Python bytecode as being stable and unchanging. In
  reality it's simply an implementation detail for CPython. This point is now
  clearly documented in both the docs for dis and the glossary.
  
  Closes issue #7829. Thanks to Terry Reedy for some initial suggestions on
  wording.
................
  r82462 | ezio.melotti | 2010-07-03 01:18:51 +0200 (Sat, 03 Jul 2010) | 17 lines
  
  Merged revisions 82455,82457,82459 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82455 | eric.smith | 2010-07-03 00:44:16 +0300 (Sat, 03 Jul 2010) | 1 line
    
    Moved period outside paren, where it belongs.
  ........
    r82457 | ezio.melotti | 2010-07-03 01:17:29 +0300 (Sat, 03 Jul 2010) | 1 line
    
    #9139: Add examples for str.format().
  ........
    r82459 | ezio.melotti | 2010-07-03 01:50:39 +0300 (Sat, 03 Jul 2010) | 1 line
    
    #9139: the thousands separator is new in 2.7.  Also add a missing variable in the example.
  ........
................
  r82464 | benjamin.peterson | 2010-07-03 01:25:44 +0200 (Sat, 03 Jul 2010) | 9 lines
  
  Merged revisions 82461 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82461 | benjamin.peterson | 2010-07-02 18:05:27 -0500 (Fri, 02 Jul 2010) | 1 line
    
    don't require the presence of __getformat__ or __setformat__; use requires_IEEE_754 globally
  ........
................
  r82465 | brett.cannon | 2010-07-03 03:32:48 +0200 (Sat, 03 Jul 2010) | 3 lines
  
  Make test_import a little bit more robust for cleaning up after itself in the
  face of a failure.
................
  r82467 | alexander.belopolsky | 2010-07-03 05:35:27 +0200 (Sat, 03 Jul 2010) | 9 lines
  
  Merged revisions 82466 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82466 | alexander.belopolsky | 2010-07-02 23:27:12 -0400 (Fri, 02 Jul 2010) | 1 line
    
    Revert r81681 (issue 8810).
  ........
................
  r82468 | ezio.melotti | 2010-07-03 06:52:19 +0200 (Sat, 03 Jul 2010) | 1 line
  
  Update comment about surrogates.
................
  r82471 | nick.coghlan | 2010-07-03 09:36:51 +0200 (Sat, 03 Jul 2010) | 1 line
  
  Issue 6507: accept source strings directly in dis.dis(). Original patch by Daniel Urban
................
  r82472 | nick.coghlan | 2010-07-03 09:38:28 +0200 (Sat, 03 Jul 2010) | 1 line
  
  Issue 6507: missing patch submitter from NEWS entry in previous checkin
................
  r82478 | mark.dickinson | 2010-07-03 11:17:16 +0200 (Sat, 03 Jul 2010) | 9 lines
  
  Merged revisions 82476 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82476 | mark.dickinson | 2010-07-03 10:15:09 +0100 (Sat, 03 Jul 2010) | 1 line
    
    Fix typo in sys.float_info docs.
  ........
................
  r82480 | georg.brandl | 2010-07-03 12:21:50 +0200 (Sat, 03 Jul 2010) | 1 line
  
  Wrap and use the correct directive.
................
  r82481 | georg.brandl | 2010-07-03 12:22:10 +0200 (Sat, 03 Jul 2010) | 1 line
  
  Use the right role.
................
  r82484 | georg.brandl | 2010-07-03 12:26:17 +0200 (Sat, 03 Jul 2010) | 9 lines
  
  Recorded merge of revisions 82474 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82474 | georg.brandl | 2010-07-03 10:40:13 +0200 (Sa, 03 Jul 2010) | 1 line
    
    Fix role name.
  ........
................
  r82485 | georg.brandl | 2010-07-03 12:26:54 +0200 (Sat, 03 Jul 2010) | 9 lines
  
  Merged revisions 82483 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82483 | georg.brandl | 2010-07-03 12:25:54 +0200 (Sa, 03 Jul 2010) | 1 line
    
    Add link to bytecode docs.
  ........
................
  r82487 | georg.brandl | 2010-07-03 12:33:26 +0200 (Sat, 03 Jul 2010) | 1 line
  
  Fix markup.
................
  r82488 | georg.brandl | 2010-07-03 12:41:33 +0200 (Sat, 03 Jul 2010) | 1 line
  
  Remove the need for a "()" empty argument list after opcodes.
................
  r82495 | victor.stinner | 2010-07-03 15:44:22 +0200 (Sat, 03 Jul 2010) | 10 lines
  
  Merged revisions 82492 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/trunk
  
  ........
    r82492 | victor.stinner | 2010-07-03 15:36:19 +0200 (sam., 03 juil. 2010) | 3 lines
    
    Issue #7673: Fix security vulnerability (CVE-2010-2089) in the audioop module,
    ensure that the input string length is a multiple of the frame size
  ........
................
  r82507 | benjamin.peterson | 2010-07-03 17:09:20 +0200 (Sat, 03 Jul 2010) | 3 lines
  
  Removed merge tracking for "svnmerge" for 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k/python/trunk
................
  r82510 | senthil.kumaran | 2010-07-03 19:48:22 +0200 (Sat, 03 Jul 2010) | 4 lines
  
  Fix Issue5468 - urlencode to handle bytes and other alternate encodings.
  (Extensive tests provided). Patch by Dan Mahn.
................
  r82514 | alexander.belopolsky | 2010-07-03 22:35:53 +0200 (Sat, 03 Jul 2010) | 3 lines
  
  Issue #9094: Make python -m pickletools disassemble pickles given in
  the command line.
................
  r82515 | mark.dickinson | 2010-07-03 23:00:51 +0200 (Sat, 03 Jul 2010) | 1 line
  
  Remove nonexistent tools from Tools README.
................
  r82516 | brett.cannon | 2010-07-03 23:37:28 +0200 (Sat, 03 Jul 2010) | 2 lines
  
  Ignore __pycache__.
................
  r82517 | alexander.belopolsky | 2010-07-03 23:42:47 +0200 (Sat, 03 Jul 2010) | 3 lines
  
  Issue #9151: Demo/classes/Dates.py does not work in 3.x
  Made minimal changes to make included test pass.
................
  r82518 | brett.cannon | 2010-07-03 23:48:25 +0200 (Sat, 03 Jul 2010) | 34 lines
  
  Make importlib.abc.SourceLoader the primary mechanism for importlib.
  
  This required moving the class from importlib/abc.py into
  importlib/_bootstrap.py and jiggering some code to work better with the class.
  This included changing how the file finder worked to better meet import
  semantics. This also led to fixing importlib to handle the empty string from
  sys.path as import currently does (and making me wish we didn't support that
  instead just required people to insert '.' instead to represent cwd).
  
  It also required making the new set_data abstractmethod create
  any needed subdirectories implicitly thanks to __pycache__ (it was either this
  or grow the SourceLoader ABC to gain an 'exists' method and either a mkdir
  method or have set_data with no data arg mean to create a directory).
  
  Lastly, as an optimization the file loaders cache the file path where the
  finder found something to use for loading (this is thanks to having a
  sourceless loader separate from the source loader to simplify the code and
  cut out stat calls).
  Unfortunately test_runpy assumed a loader would always work for a module, even
  if you changed from underneath it what it was expected to work with. By simply
  dropping the previous loader in test_runpy so the proper loader can be returned
  by the finder fixed the failure.
  
  At this point importlib deviates from import on two points:
  
  1. The exception raised when trying to import a file is different (import does
  an explicit file check to print a special message, importlib just says the path
  cannot be imported as if it was just some module name).
  
  2. the co_filename on a code object is not being set to where bytecode was
  actually loaded from instead of where the marshalled code object originally
  came from (a solution for this has already been agreed upon on python-dev but has
  not been implemented yet; issue8611).
................
  r82519 | brett.cannon | 2010-07-03 23:50:52 +0200 (Sat, 03 Jul 2010) | 1 line
  
  Trailing whitespace is bad for .rst files.
................
  r82520 | brett.cannon | 2010-07-04 00:03:16 +0200 (Sun, 04 Jul 2010) | 1 line
  
  Fix a spelling mistake in a comment.
................
  r82521 | alexander.belopolsky | 2010-07-04 00:05:41 +0200 (Sun, 04 Jul 2010) | 5 lines
  
  Issue #9151: Demo/classes/Dates.py does not work in 3.x Converted
  descriptive comment into a docstring.  Cast attributes to int in
  __init__.  Use __new__ instead of deleting attributes to
  "uninitialize".
................
  r82522 | brett.cannon | 2010-07-04 00:18:47 +0200 (Sun, 04 Jul 2010) | 34 lines
  
  Make importlib.abc.SourceLoader the primary mechanism for importlib.
  
  This required moving the class from importlib/abc.py into
  importlib/_bootstrap.py and jiggering some code to work better with the class.
  This included changing how the file finder worked to better meet import
  semantics. This also led to fixing importlib to handle the empty string from
  sys.path as import currently does (and making me wish we didn't support that
  instead just required people to insert '.' instead to represent cwd).
  
  It also required making the new set_data abstractmethod create
  any needed subdirectories implicitly thanks to __pycache__ (it was either this
  or grow the SourceLoader ABC to gain an 'exists' method and either a mkdir
  method or have set_data with no data arg mean to create a directory).
  
  Lastly, as an optimization the file loaders cache the file path where the
  finder found something to use for loading (this is thanks to having a
  sourceless loader separate from the source loader to simplify the code and
  cut out stat calls).
  Unfortunately test_runpy assumed a loader would always work for a module, even
  if you changed from underneath it what it was expected to work with. By simply
  dropping the previous loader in test_runpy so the proper loader can be returned
  by the finder fixed the failure.
  
  At this point importlib deviates from import on two points:
  
  1. The exception raised when trying to import a file is different (import does
  an explicit file check to print a special message, importlib just says the path
  cannot be imported as if it was just some module name).
  
  2. the co_filename on a code object is not being set to where bytecode was
  actually loaded from instead of where the marshalled code object originally
  came from (a solution for this has already been agreed upon on python-dev but has
  not been implemented yet; issue8611).
................
  r82523 | brett.cannon | 2010-07-04 00:32:41 +0200 (Sun, 04 Jul 2010) | 34 lines
  
  Make importlib.abc.SourceLoader the primary mechanism for importlib.
  
  This required moving the class from importlib/abc.py into
  importlib/_bootstrap.py and jiggering some code to work better with the class.
  This included changing how the file finder worked to better meet import
  semantics. This also led to fixing importlib to handle the empty string from
  sys.path as import currently does (and making me wish we didn't support that
  instead just required people to insert '.' instead to represent cwd).
  
  It also required making the new set_data abstractmethod create
  any needed subdirectories implicitly thanks to __pycache__ (it was either this
  or grow the SourceLoader ABC to gain an 'exists' method and either a mkdir
  method or have set_data with no data arg mean to create a directory).
  
  Lastly, as an optimization the file loaders cache the file path where the
  finder found something to use for loading (this is thanks to having a
  sourceless loader separate from the source loader to simplify the code and
  cut out stat calls).
  Unfortunately test_runpy assumed a loader would always work for a module, even
  if you changed from underneath it what it was expected to work with. By simply
  dropping the previous loader in test_runpy so the proper loader can be returned
  by the finder fixed the failure.
  
  At this point importlib deviates from import on two points:
  
  1. The exception raised when trying to import a file is different (import does
  an explicit file check to print a special message, importlib just says the path
  cannot be imported as if it was just some module name).
  
  2. the co_filename on a code object is not being set to where bytecode was
  actually loaded from instead of where the marshalled code object originally
  came from (a solution for this has already been agreed upon on python-dev but has
  not been implemented yet; issue8611).
................
  r82524 | alexander.belopolsky | 2010-07-04 00:36:06 +0200 (Sun, 04 Jul 2010) | 1 line
  
  Made minimal modifications to pass included tests
................
  r82526 | georg.brandl | 2010-07-04 09:25:18 +0200 (Sun, 04 Jul 2010) | 1 line
  
  Remove old and unsafe Dbm demo class.
................
  r82527 | mark.dickinson | 2010-07-04 12:15:11 +0200 (Sun, 04 Jul 2010) | 1 line
  
  Fix refleak in Modules/audioop.c.
................
  r82533 | alexander.belopolsky | 2010-07-04 18:28:08 +0200 (Sun, 04 Jul 2010) | 1 line
  
  Issue #9152: Removed dead code in datetime module
................
  r82535 | benjamin.peterson | 2010-07-04 18:33:07 +0200 (Sun, 04 Jul 2010) | 3 lines
  
  Removed merge tracking for "svnmerge" for 
  svn+ssh://pythondev at svn.python.org/sandbox/trunk/2to3/lib2to3
................
  r82537 | mark.dickinson | 2010-07-04 18:37:31 +0200 (Sun, 04 Jul 2010) | 2 lines
  
  Issue #9128: Fix validation of class decorators in parser module.
................
  r82538 | benjamin.peterson | 2010-07-04 18:37:39 +0200 (Sun, 04 Jul 2010) | 3 lines
  
  Initialized merge tracking via "svnmerge" with revisions "0-80937" from 
  svn+ssh://pythondev at svn.python.org/sandbox/trunk/2to3
................
  r82540 | benjamin.peterson | 2010-07-04 18:40:27 +0200 (Sun, 04 Jul 2010) | 3 lines
  
  Removed merge tracking for "svnmerge" for 
  svn+ssh://pythondev at svn.python.org/sandbox/trunk/2to3
................
  r82541 | benjamin.peterson | 2010-07-04 18:41:33 +0200 (Sun, 04 Jul 2010) | 3 lines
  
  Initialized merge tracking via "svnmerge" with revisions "0-80937" from 
  svn+ssh://pythondev at svn.python.org/sandbox/trunk/2to3/lib2to3
................
  r82542 | benjamin.peterson | 2010-07-04 18:44:15 +0200 (Sun, 04 Jul 2010) | 17 lines
  
  Merged revisions 81478,82530-82531 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/sandbox/trunk/2to3/lib2to3
  
  ........
    r81478 | benjamin.peterson | 2010-05-22 13:47:39 -0500 (Sat, 22 May 2010) | 1 line
    
    ensure doctests have some future_features
  ........
    r82530 | benjamin.peterson | 2010-07-04 11:11:41 -0500 (Sun, 04 Jul 2010) | 1 line
    
    simplify ignore star imports from itertools #8892
  ........
    r82531 | benjamin.peterson | 2010-07-04 11:13:20 -0500 (Sun, 04 Jul 2010) | 1 line
    
    wrap with parenthesis not \
  ........
................
  r82547 | alexander.belopolsky | 2010-07-04 19:00:20 +0200 (Sun, 04 Jul 2010) | 3 lines
  
  Issue #9118: help(None) will now return NoneType doc instead of
  starting interactive help.
................
  r82550 | alexander.belopolsky | 2010-07-04 19:38:32 +0200 (Sun, 04 Jul 2010) | 1 line
  
  Fixed doctests
................
  r82551 | alexander.belopolsky | 2010-07-04 19:47:30 +0200 (Sun, 04 Jul 2010) | 1 line
  
  Fixed the test
................
  r82552 | mark.dickinson | 2010-07-04 20:11:51 +0200 (Sun, 04 Jul 2010) | 2 lines
  
  Issue #9130: Fix validation of relative imports in parser module.
................
  r82553 | mark.dickinson | 2010-07-04 20:15:26 +0200 (Sun, 04 Jul 2010) | 1 line
  
  Fix symbol numbers in test_parser test.
................
  r82555 | mark.dickinson | 2010-07-04 20:38:57 +0200 (Sun, 04 Jul 2010) | 2 lines
  
  Issue #9130: Validate ellipsis tokens in relative imports.
................
  r82557 | mark.dickinson | 2010-07-04 20:49:18 +0200 (Sun, 04 Jul 2010) | 1 line
  
  Make Demo/parser/test_parser.py run.
................
  r82558 | mark.dickinson | 2010-07-04 21:23:49 +0200 (Sun, 04 Jul 2010) | 1 line
  
  Remove coding cookie from heapq.py.
................
  r82559 | mark.dickinson | 2010-07-04 21:27:12 +0200 (Sun, 04 Jul 2010) | 1 line
  
  Re-encode shlex.py in UTF-8, and remove coding cookie.
................
  r82560 | mark.dickinson | 2010-07-04 22:07:09 +0200 (Sun, 04 Jul 2010) | 1 line
  
  Revert -r82559;  it's not clear that this is the right thing to do, and the change obscures the original intentions.
................
  r82561 | andrew.kuchling | 2010-07-04 22:47:00 +0200 (Sun, 04 Jul 2010) | 1 line
  
  Add brief explanation; remove no-longer-working link
................
  r82567 | brett.cannon | 2010-07-05 00:05:34 +0200 (Mon, 05 Jul 2010) | 5 lines
  
  Do some basic refactoring. Along the way also list what files had their
  whitespace cleaned up and do the proper plurality for "N file(s)" based on N.
  
  Refactoring closes issue8912. Thanks to ??ric Araujo for the patch.
................
  r82570 | senthil.kumaran | 2010-07-05 13:41:42 +0200 (Mon, 05 Jul 2010) | 3 lines
  
  Fix: Issue9091 Minor documentation clarification.
................
  r82573 | senthil.kumaran | 2010-07-05 14:00:56 +0200 (Mon, 05 Jul 2010) | 3 lines
  
  Fix the docstrings of the capitalize method.
................
  r82576 | benjamin.peterson | 2010-07-05 16:59:40 +0200 (Mon, 05 Jul 2010) | 1 line
  
  pydoc still has a silly encoding
................
  r82577 | benjamin.peterson | 2010-07-05 17:01:22 +0200 (Mon, 05 Jul 2010) | 1 line
  
  cleanup basicsize logic #3268
................
  r82578 | alexander.belopolsky | 2010-07-05 17:05:33 +0200 (Mon, 05 Jul 2010) | 1 line
  
  Added more tests for utctimetuple()
................
  r82585 | benjamin.peterson | 2010-07-05 19:11:05 +0200 (Mon, 05 Jul 2010) | 1 line
  
  looking up on the type is correct, so this isn't an XXX
................
  r82587 | georg.brandl | 2010-07-05 19:48:38 +0200 (Mon, 05 Jul 2010) | 1 line
  
  #9166: NotImplemented is not an exception.
................
  r82589 | alexander.belopolsky | 2010-07-05 19:57:31 +0200 (Mon, 05 Jul 2010) | 1 line
  
  Added two more test cases for datetime
................
  r82591 | benjamin.peterson | 2010-07-05 22:04:54 +0200 (Mon, 05 Jul 2010) | 1 line
  
  untabify
................
  r82593 | georg.brandl | 2010-07-05 22:13:06 +0200 (Mon, 05 Jul 2010) | 1 line
  
  Remove Dbm.
................
  r82594 | georg.brandl | 2010-07-05 22:13:41 +0200 (Mon, 05 Jul 2010) | 1 line
  
  Update Vec class constructor, remove indirection via function, use operator module.
................
  r82595 | mark.dickinson | 2010-07-05 22:14:26 +0200 (Mon, 05 Jul 2010) | 1 line
  
  Post-detabification cleanup.
................
  r82597 | victor.stinner | 2010-07-05 23:36:21 +0200 (Mon, 05 Jul 2010) | 2 lines
  
  #9158: Fix y* format of PyArg_Parse*() functions documentation
................
  r82599 | alexander.belopolsky | 2010-07-05 23:44:05 +0200 (Mon, 05 Jul 2010) | 1 line
  
  "Modernized" the demo a little.
................
  r82600 | brett.cannon | 2010-07-06 00:11:16 +0200 (Tue, 06 Jul 2010) | 1 line
  
  Fix test_xmlrpc_net to no longer fail since there are no more buildbots for trunk.
................
  r82601 | senthil.kumaran | 2010-07-06 04:08:36 +0200 (Tue, 06 Jul 2010) | 3 lines
  
  Change 'lowered' to 'lowercased'
................
  r82604 | matthias.klose | 2010-07-06 12:53:30 +0200 (Tue, 06 Jul 2010) | 2 lines
  
  - sysmodule.c (get_hash_info): Define as static function.
................
  r82607 | mark.dickinson | 2010-07-06 17:00:40 +0200 (Tue, 06 Jul 2010) | 1 line
  
  Indentation and PEP 7 fixes.
................
  r82609 | vinay.sajip | 2010-07-06 17:08:55 +0200 (Tue, 06 Jul 2010) | 1 line
  
  Minor improvements to logging documentation.
................
  r82610 | mark.dickinson | 2010-07-06 17:11:44 +0200 (Tue, 06 Jul 2010) | 1 line
  
  Style nit.
................
  r82615 | georg.brandl | 2010-07-07 00:58:50 +0200 (Wed, 07 Jul 2010) | 1 line
  
  Fix typo.
................
  r82617 | alexander.belopolsky | 2010-07-07 01:19:45 +0200 (Wed, 07 Jul 2010) | 1 line
  
  Issue #9000: datetime.timezone objects now have eval-friendly repr.
................
  r82624 | mark.dickinson | 2010-07-07 18:10:44 +0200 (Wed, 07 Jul 2010) | 1 line
  
  Kill gratuitous space.
................
  r82625 | mark.dickinson | 2010-07-07 18:17:31 +0200 (Wed, 07 Jul 2010) | 1 line
  
  Minor refactoring in lgamma code, for clarity.
................
  r82626 | mark.dickinson | 2010-07-07 18:21:29 +0200 (Wed, 07 Jul 2010) | 1 line
  
  Issue #9186:  log1p(-1.0) should raise ValueError, not OverflowError.
................
  r82628 | benjamin.peterson | 2010-07-07 20:44:05 +0200 (Wed, 07 Jul 2010) | 1 line
  
  this needn't be in the loop
................
  r82629 | georg.brandl | 2010-07-07 20:51:43 +0200 (Wed, 07 Jul 2010) | 1 line
  
  Make comment out of an awkward note.
................
  r82630 | benjamin.peterson | 2010-07-07 20:54:59 +0200 (Wed, 07 Jul 2010) | 1 line
  
  don't ignore exceptions from PyObject_IsTrue
................
  r82632 | georg.brandl | 2010-07-07 21:04:36 +0200 (Wed, 07 Jul 2010) | 1 line
  
  Turn more notes into comments.
................
  r82633 | benjamin.peterson | 2010-07-07 21:04:48 +0200 (Wed, 07 Jul 2010) | 1 line
  
  add NEWS
................
  r82636 | benjamin.peterson | 2010-07-07 22:54:01 +0200 (Wed, 07 Jul 2010) | 4 lines
  
  make struct sequences subclass tuple; kill lots of code
  
  This fixes #8413.
................
  r82637 | benjamin.peterson | 2010-07-08 00:45:06 +0200 (Thu, 08 Jul 2010) | 1 line
  
  ValueError in this case is also acceptable
................
  r82642 | alexander.belopolsky | 2010-07-08 01:56:38 +0200 (Thu, 08 Jul 2010) | 1 line
  
  Issue #5288: Eliminated round-trips between timdelta and int offsets
................
  r82644 | ezio.melotti | 2010-07-08 17:03:02 +0200 (Thu, 08 Jul 2010) | 1 line
  
  Fix typo in a comment in mathmodule.c.
................
  r82646 | mark.dickinson | 2010-07-08 19:23:40 +0200 (Thu, 08 Jul 2010) | 1 line
  
  In test_decimal, convert heuristic for skipping tests into an explicit skiplist.
................
  r82647 | antoine.pitrou | 2010-07-08 20:51:30 +0200 (Thu, 08 Jul 2010) | 3 lines
  
  Issue #8605: Skip test_gdb if Python is compiled with optimizations.
................
  r82649 | mark.dickinson | 2010-07-08 21:03:34 +0200 (Thu, 08 Jul 2010) | 1 line
  
  Fix a performance issue in Decimal.pow.  Thanks Stefan Krah for finding this.
................
  r82650 | mark.dickinson | 2010-07-08 21:09:16 +0200 (Thu, 08 Jul 2010) | 1 line
  
  Fix misplaced exactness check that was causing unnecessary work in Decimal.__pow__.
................
  r82654 | mark.dickinson | 2010-07-08 23:15:36 +0200 (Thu, 08 Jul 2010) | 3 lines
  
  Issue #9136: Profiling Decimal gave 'dictionary changed size during iteration'.
  Remove the use of locals() that caused this error.
................
  r82659 | brian.curtin | 2010-07-08 23:39:08 +0200 (Thu, 08 Jul 2010) | 17 lines
  
  Implement #1578269. Patch by Jason R. Coombs.
  
  Added Windows support for os.symlink when run on Windows 6.0 or greater,
  aka Vista. Previous Windows versions will raise NotImplementedError
  when trying to symlink.
  
  Includes numerous test updates and additions to test_os, including
  a symlink_support module because of the fact that privilege escalation
  is required in order to run the tests to ensure that the user is able
  to create symlinks. By default, accounts do not have the required
  privilege, so the escalation code will have to be exposed later (or
  documented on how to do so). I'll be following up with that work next.
  
  Note that the tests use ctypes, which was agreed on during the PyCon
  language summit.
................
  r82661 | brian.curtin | 2010-07-09 00:13:25 +0200 (Fri, 09 Jul 2010) | 2 lines
  
  Undo inadvertant checkin.
................
  r82662 | benjamin.peterson | 2010-07-09 00:16:05 +0200 (Fri, 09 Jul 2010) | 1 line
  
  assert tuple inheritance
................
  r82663 | benjamin.peterson | 2010-07-09 00:33:03 +0200 (Fri, 09 Jul 2010) | 1 line
  
  fix repr of complicated structseqs #9206
................
  r82723 | benjamin.peterson | 2010-07-09 03:58:26 +0200 (Fri, 09 Jul 2010) | 1 line
  
  use assert method
................
  r82724 | georg.brandl | 2010-07-09 09:33:15 +0200 (Fri, 09 Jul 2010) | 1 line
  
  2.7 is now stable.
................
  r82730 | r.david.murray | 2010-07-09 14:23:21 +0200 (Fri, 09 Jul 2010) | 4 lines
  
  7846: limit fnmatch pattern cache to _MAXCACHE=100 entries.
  
  Patch by Andrew Clegg.
................
  r82731 | r.david.murray | 2010-07-09 15:14:03 +0200 (Fri, 09 Jul 2010) | 2 lines
  
  Fix sort order mistake in Misc/ACKS.
................
  r82735 | benjamin.peterson | 2010-07-09 15:20:40 +0200 (Fri, 09 Jul 2010) | 1 line
  
  OverflowError is fine
................
  r82737 | brian.curtin | 2010-07-09 15:22:07 +0200 (Fri, 09 Jul 2010) | 2 lines
  
  Reorder Jason "two O's" Coombs
................
  r82739 | benjamin.peterson | 2010-07-09 15:28:42 +0200 (Fri, 09 Jul 2010) | 1 line
  
  allow more exceptions
................
  r82741 | benjamin.peterson | 2010-07-09 15:31:11 +0200 (Fri, 09 Jul 2010) | 1 line
  
  wrap
................
  r82743 | brian.curtin | 2010-07-09 15:54:27 +0200 (Fri, 09 Jul 2010) | 8 lines
  
  Fix the breakage of Lib/tarfile.py on non-Windows platforms due to 
  using WindowsError in a try/except. Only add WindowsError to the list of
  exceptions to catch when we are actually running on Windows.
  
  Additionally, add a call that was left out in test_posixpath.
  
  Thanks Amaury, Antoine, and Jason.
................
  r82744 | brian.curtin | 2010-07-09 17:15:09 +0200 (Fri, 09 Jul 2010) | 3 lines
  
  Adding versionchanged to the various Windows os.symlink additions, along
  with a few minor touchups.
................
  r82745 | brian.curtin | 2010-07-09 17:58:59 +0200 (Fri, 09 Jul 2010) | 2 lines
  
  Re-flow several long lines from #1578269.
................
  r82746 | jeffrey.yasskin | 2010-07-09 18:30:58 +0200 (Fri, 09 Jul 2010) | 12 lines
  
  Issue #9189: Allow users to set $CFLAGS, $CPPFLAGS, and $LDFLAGS when running
  configure to append to Python's default values for those variables, and
  similarly allow users to set $XXFLAGS on the make command line to append to the
  values set by configure.
  
  In the makefile, this renames the variables that used to be $XXFLAGS to
  $PY_XXFLAGS, and renames the old $PY_CFLAGS to $PY_CORE_CFLAGS.  To compensate,
  sysconfig now aliases $XXFLAGS=$PY_XXFLAGS so that scripts using it keep
  working.  I see that as the right interface, not a backward-compatibility hack,
  since these are logically the $XXFLAGS variables; we just use a different name
  in the makefile to deal with make's semantics.
................
  r82748 | benjamin.peterson | 2010-07-09 20:28:25 +0200 (Fri, 09 Jul 2010) | 1 line
  
  remove SocketIO.__del__; close() can handle it
................
  r82749 | mark.dickinson | 2010-07-09 21:25:48 +0200 (Fri, 09 Jul 2010) | 1 line
  
  Silence gcc warning. (In function 'bytearray_init': warning: 'value' may be used uninitialized in this function).
................
  r82750 | benjamin.peterson | 2010-07-09 21:37:00 +0200 (Fri, 09 Jul 2010) | 1 line
  
  be more robust across platforms
................
  r82753 | jeffrey.yasskin | 2010-07-09 21:55:05 +0200 (Fri, 09 Jul 2010) | 2 lines
  
  Oops. Fix distutils tests that r82746 broke.
................
  r82757 | georg.brandl | 2010-07-10 10:58:37 +0200 (Sat, 10 Jul 2010) | 1 line
  
  Fix markup.
................
  r82758 | georg.brandl | 2010-07-10 12:23:40 +0200 (Sat, 10 Jul 2010) | 1 line
  
  Emphasize role of count for Pascal string.
................
  r82759 | georg.brandl | 2010-07-10 12:32:36 +0200 (Sat, 10 Jul 2010) | 1 line
  
  #3071: tell how many values were expected when unpacking too many.
................
  r82760 | georg.brandl | 2010-07-10 12:39:57 +0200 (Sat, 10 Jul 2010) | 1 line
  
  #3214: improve description of duck-typing in glossary.
................
  r82761 | georg.brandl | 2010-07-10 13:40:13 +0200 (Sat, 10 Jul 2010) | 1 line
  
  #1434090: properly append child in expatbuilder doctype handler.
................
  r82762 | georg.brandl | 2010-07-10 13:51:06 +0200 (Sat, 10 Jul 2010) | 1 line
  
  #8338: fix outdated class name.
................
  r82763 | georg.brandl | 2010-07-10 14:01:34 +0200 (Sat, 10 Jul 2010) | 1 line
  
  #8456: fix signature of sqlite3.connect().
................
  r82764 | georg.brandl | 2010-07-10 14:20:38 +0200 (Sat, 10 Jul 2010) | 1 line
  
  #8564: update docs on integrating doctest/unittest with unittest(2) test discovery.
................
  r82766 | r.david.murray | 2010-07-10 15:52:13 +0200 (Sat, 10 Jul 2010) | 5 lines
  
  Fix 'refleak' introduced by fnmatch cache purge tests.
  
  This introduces a 'purge' function for the fnmatch module analogous
  to the 'purge' function in the re module.
................
  r82767 | michael.foord | 2010-07-10 15:52:22 +0200 (Sat, 10 Jul 2010) | 2 lines
  
  Fix error message for comparing single line strings in unittest.TestCase.assertEqual.
  Issue 9174.
................
  r82771 | r.david.murray | 2010-07-10 16:23:36 +0200 (Sat, 10 Jul 2010) | 2 lines
  
  Add missing docs for re module 'purge' function.
................
  r82777 | benjamin.peterson | 2010-07-10 17:14:45 +0200 (Sat, 10 Jul 2010) | 1 line
  
  ValueError is eventually what we want to move to, I suppose
................
  r82780 | senthil.kumaran | 2010-07-11 05:12:43 +0200 (Sun, 11 Jul 2010) | 3 lines
  
  Stricter verification for file based url scheme and reliance on ftp protocol.
................
  r82784 | benjamin.peterson | 2010-07-11 05:36:35 +0200 (Sun, 11 Jul 2010) | 1 line
  
  remove unneeded error check
................
  r82785 | senthil.kumaran | 2010-07-11 07:01:52 +0200 (Sun, 11 Jul 2010) | 3 lines
  
  Issue2007: Removing reference to MSIE Cookie handling by mechanize. Suggested by John J Lee.
................
  r82789 | georg.brandl | 2010-07-11 10:33:16 +0200 (Sun, 11 Jul 2010) | 1 line
  
  Silence makeindex.
................
  r82790 | georg.brandl | 2010-07-11 10:36:20 +0200 (Sun, 11 Jul 2010) | 1 line
  
  #3214 followup: add link to ABC entry.
................
  r82796 | ronald.oussoren | 2010-07-11 11:05:07 +0200 (Sun, 11 Jul 2010) | 2 lines
  
  Fix for issue 9164
................
  r82798 | georg.brandl | 2010-07-11 11:23:11 +0200 (Sun, 11 Jul 2010) | 1 line
  
  #6774: explain shutdown() behavior varying with platform.
................
  r82799 | georg.brandl | 2010-07-11 11:26:57 +0200 (Sun, 11 Jul 2010) | 1 line
  
  Fix typo.
................
  r82801 | georg.brandl | 2010-07-11 11:33:39 +0200 (Sun, 11 Jul 2010) | 1 line
  
  #9184: fix default value for "buffering" param of open().
................
  r82804 | georg.brandl | 2010-07-11 11:41:21 +0200 (Sun, 11 Jul 2010) | 1 line
  
  Allow set literals in literal_eval().
................
  r82805 | georg.brandl | 2010-07-11 11:42:10 +0200 (Sun, 11 Jul 2010) | 1 line
  
  #7935: cross-reference to ast.literal_eval() from eval() docs.
................
  r82806 | georg.brandl | 2010-07-11 12:22:44 +0200 (Sun, 11 Jul 2010) | 1 line
  
  #9223: link to Command class reference, and move Command interface docs nearer to class docs.
................
  r82814 | antoine.pitrou | 2010-07-11 14:12:00 +0200 (Sun, 11 Jul 2010) | 4 lines
  
  Issue #7616: Fix copying of overlapping memoryview slices with the Intel
  compiler.
................
  r82817 | benjamin.peterson | 2010-07-11 14:59:24 +0200 (Sun, 11 Jul 2010) | 1 line
  
  test for set literals
................
  r82818 | mark.dickinson | 2010-07-11 19:38:24 +0200 (Sun, 11 Jul 2010) | 2 lines
  
  Issue #9165: Add math.isfinite and cmath.isfinite.
................
  r82819 | martin.v.loewis | 2010-07-11 19:39:46 +0200 (Sun, 11 Jul 2010) | 2 lines
  
  Issue #8974: fix print calls in msgfmt.py.
................
  r82820 | mark.dickinson | 2010-07-11 20:13:41 +0200 (Sun, 11 Jul 2010) | 1 line
  
  Improve docstrings for isnan, isinf and isfinite.
................
  r82821 | mark.dickinson | 2010-07-11 20:53:06 +0200 (Sun, 11 Jul 2010) | 3 lines
  
  Issue #9137: Fix issue in MutableMapping.update, which incorrectly
  treated keyword arguments called 'self' or 'other' specially.
................
  r82822 | mark.dickinson | 2010-07-11 21:12:10 +0200 (Sun, 11 Jul 2010) | 1 line
  
  assertEquals -> assertEqual
................
  r82825 | mark.dickinson | 2010-07-11 21:47:37 +0200 (Sun, 11 Jul 2010) | 1 line
  
  Include versionadded in (c)math.isfinite docs (thanks Ezio Melotti).  Other minor doc cleanups.
................
  r82828 | benjamin.peterson | 2010-07-12 01:06:06 +0200 (Mon, 12 Jul 2010) | 1 line
  
  allow byte literals
................
  r82830 | brian.curtin | 2010-07-12 01:32:11 +0200 (Mon, 12 Jul 2010) | 2 lines
  
  Put my name down for winreg.
................
  r82831 | martin.v.loewis | 2010-07-12 08:03:18 +0200 (Mon, 12 Jul 2010) | 2 lines
  
  Remove myself.
................
  r82832 | georg.brandl | 2010-07-12 11:00:29 +0200 (Mon, 12 Jul 2010) | 1 line
  
  Take care of duplicate target name warnings.
................
  r82834 | georg.brandl | 2010-07-12 11:06:13 +0200 (Mon, 12 Jul 2010) | 1 line
  
  Use raw string literals for regexes containing backlash.
................
  r82835 | mark.dickinson | 2010-07-12 11:37:40 +0200 (Mon, 12 Jul 2010) | 1 line
  
  Remove mention of execfile from the tutorial.
................
  r82837 | mark.dickinson | 2010-07-12 16:14:18 +0200 (Mon, 12 Jul 2010) | 1 line
  
  Issue #2009: refactor varargslist and typedargslist productions to make them more friendly for third-party parsers.
................
  r82838 | mark.dickinson | 2010-07-12 16:18:21 +0200 (Mon, 12 Jul 2010) | 1 line
  
  Regenerate Python/graminit.c.
................
  r82839 | ezio.melotti | 2010-07-12 21:49:41 +0200 (Mon, 12 Jul 2010) | 1 line
  
  #6026: skip test_get_file_list when zlib is not available.
................
  r82842 | antoine.pitrou | 2010-07-12 22:01:52 +0200 (Mon, 12 Jul 2010) | 3 lines
  
  Fix definition of len() and indexing for memoryview objects (part of #7696).
................
  r82843 | mark.dickinson | 2010-07-12 22:03:24 +0200 (Mon, 12 Jul 2010) | 1 line
  
  Yield more information on failure in test_struct boolean test.
................
  r82848 | georg.brandl | 2010-07-13 08:38:10 +0200 (Tue, 13 Jul 2010) | 1 line
  
  Add bytes in literal_eval doc.
................
  r82849 | jeroen.ruigrok | 2010-07-13 16:47:01 +0200 (Tue, 13 Jul 2010) | 2 lines
  
  Fix documentation typo: wprite() -> write().
................
  r82850 | alexander.belopolsky | 2010-07-13 16:50:16 +0200 (Tue, 13 Jul 2010) | 1 line
  
  Set sys.modules[name] to None instead of 0 to block module import.
................
  r82856 | victor.stinner | 2010-07-14 01:04:56 +0200 (Wed, 14 Jul 2010) | 2 lines
  
  Issue #9243: Fix sndhdr module and add unit tests, contributed by James Lee.
................
  r82857 | victor.stinner | 2010-07-14 01:08:01 +0200 (Wed, 14 Jul 2010) | 2 lines
  
  Woops, test_sndhdr.py contains the same code twice: fix it
................
  r82871 | georg.brandl | 2010-07-14 10:00:22 +0200 (Wed, 14 Jul 2010) | 1 line
  
  #9258: fix copy-paste errors.
................
  r82872 | georg.brandl | 2010-07-14 10:53:18 +0200 (Wed, 14 Jul 2010) | 1 line
  
  Remove XXX from text.
................
  r82873 | georg.brandl | 2010-07-14 10:53:36 +0200 (Wed, 14 Jul 2010) | 1 line
  
  Remove unused code that would raise a NameError.
................
  r82874 | georg.brandl | 2010-07-14 10:54:40 +0200 (Wed, 14 Jul 2010) | 1 line
  
  #9235: fix missing import of sys.
................
  r82879 | stefan.krah | 2010-07-14 12:16:11 +0200 (Wed, 14 Jul 2010) | 3 lines
  
  High byte is the exit status.
................
  r82881 | senthil.kumaran | 2010-07-14 12:21:22 +0200 (Wed, 14 Jul 2010) | 3 lines
  
  Fix Issue5842 - Moving the tests out of urllib.parse module
................
  r82885 | antoine.pitrou | 2010-07-14 13:52:38 +0200 (Wed, 14 Jul 2010) | 4 lines
  
  Issue #9251: test_threaded_import didn't fail when run through regrtest
  if the import lock was disabled.
................
  r82887 | alexander.belopolsky | 2010-07-14 15:46:57 +0200 (Wed, 14 Jul 2010) | 1 line
  
  PEP 8 conformance: class_ -> cls
................
  r82890 | senthil.kumaran | 2010-07-14 21:15:23 +0200 (Wed, 14 Jul 2010) | 3 lines
  
  Fix: Issue6853 - Get HTTPS system proxy in Windows.
................
  r82895 | senthil.kumaran | 2010-07-14 22:10:52 +0200 (Wed, 14 Jul 2010) | 3 lines
  
  Fix a mistake, https proxy shoud be https://
................
  r82899 | senthil.kumaran | 2010-07-14 22:30:02 +0200 (Wed, 14 Jul 2010) | 2 lines
  
  Fix issue9132 - Documentation for comparing dictionaries is out of date
................
  r82910 | brett.cannon | 2010-07-15 08:24:04 +0200 (Thu, 15 Jul 2010) | 2 lines
  
  Touch up comments and code along with outputting what the unit of measure is.
................
  r82915 | alexander.belopolsky | 2010-07-16 16:39:45 +0200 (Fri, 16 Jul 2010) | 1 line
  
  Corrected TUPLE<N> opcodes' docs.
................
  r82918 | brett.cannon | 2010-07-16 21:04:29 +0200 (Fri, 16 Jul 2010) | 4 lines
  
  Add benchmarks for importing just source w/o writing bytecode, importing source
  while writing bytecode, and importing bytecode with source existing (don't care
  about sourceless imports).
................
  r82919 | antoine.pitrou | 2010-07-16 21:10:38 +0200 (Fri, 16 Jul 2010) | 4 lines
  
  Fix possible failure in pickling tests due to different instantiations
  of the random module being around.
................
  r82921 | brett.cannon | 2010-07-16 21:26:23 +0200 (Fri, 16 Jul 2010) | 1 line
  
  Add comma grouping to max result so it's easier to read.
................
  r82922 | r.david.murray | 2010-07-17 03:19:57 +0200 (Sat, 17 Jul 2010) | 4 lines
  
  #1555570: correctly handle a \r\n that is split by the read buffer.
  
  Patch and test by Tony Nelson.
................


Added:
   python/branches/py3k-cdecimal/Demo/parser/test_unparse.py
      - copied unchanged from r82922, /python/branches/py3k/Demo/parser/test_unparse.py
   python/branches/py3k-cdecimal/Doc/documenting/building.rst
      - copied unchanged from r82922, /python/branches/py3k/Doc/documenting/building.rst
   python/branches/py3k-cdecimal/Lib/encodings/cp858.py
      - copied unchanged from r82922, /python/branches/py3k/Lib/encodings/cp858.py
   python/branches/py3k-cdecimal/Lib/test/sndhdrdata/
      - copied from r82922, /python/branches/py3k/Lib/test/sndhdrdata/
   python/branches/py3k-cdecimal/Lib/test/symlink_support.py
      - copied unchanged from r82922, /python/branches/py3k/Lib/test/symlink_support.py
   python/branches/py3k-cdecimal/Lib/test/test_sndhdr.py
      - copied unchanged from r82922, /python/branches/py3k/Lib/test/test_sndhdr.py
   python/branches/py3k-cdecimal/Modules/_ctypes/libffi_msvc/LICENSE
      - copied unchanged from r82922, /python/branches/py3k/Modules/_ctypes/libffi_msvc/LICENSE
   python/branches/py3k-cdecimal/Modules/_ctypes/libffi_msvc/README
      - copied unchanged from r82922, /python/branches/py3k/Modules/_ctypes/libffi_msvc/README
   python/branches/py3k-cdecimal/Modules/_ctypes/libffi_msvc/README.ctypes
      - copied unchanged from r82922, /python/branches/py3k/Modules/_ctypes/libffi_msvc/README.ctypes
   python/branches/py3k-cdecimal/Modules/_ctypes/libffi_msvc/ffi.c
      - copied unchanged from r82922, /python/branches/py3k/Modules/_ctypes/libffi_msvc/ffi.c
   python/branches/py3k-cdecimal/Modules/_ctypes/libffi_msvc/ffi_common.h
      - copied unchanged from r82922, /python/branches/py3k/Modules/_ctypes/libffi_msvc/ffi_common.h
   python/branches/py3k-cdecimal/Modules/_ctypes/libffi_msvc/ffitarget.h
      - copied unchanged from r82922, /python/branches/py3k/Modules/_ctypes/libffi_msvc/ffitarget.h
   python/branches/py3k-cdecimal/Modules/_ctypes/libffi_msvc/prep_cif.c
      - copied unchanged from r82922, /python/branches/py3k/Modules/_ctypes/libffi_msvc/prep_cif.c
   python/branches/py3k-cdecimal/Modules/_ctypes/libffi_msvc/types.c
      - copied unchanged from r82922, /python/branches/py3k/Modules/_ctypes/libffi_msvc/types.c
   python/branches/py3k-cdecimal/Modules/_ctypes/libffi_msvc/win32.c
      - copied unchanged from r82922, /python/branches/py3k/Modules/_ctypes/libffi_msvc/win32.c
   python/branches/py3k-cdecimal/Modules/_ctypes/libffi_msvc/win64.asm
      - copied unchanged from r82922, /python/branches/py3k/Modules/_ctypes/libffi_msvc/win64.asm
   python/branches/py3k-cdecimal/Modules/_time.c
      - copied unchanged from r82922, /python/branches/py3k/Modules/_time.c
   python/branches/py3k-cdecimal/Modules/_time.h
      - copied unchanged from r82922, /python/branches/py3k/Modules/_time.h
Removed:
   python/branches/py3k-cdecimal/Demo/classes/Dbm.py
   python/branches/py3k-cdecimal/Tools/webchecker/
Modified:
   python/branches/py3k-cdecimal/   (props changed)
   python/branches/py3k-cdecimal/Demo/classes/Complex.py
   python/branches/py3k-cdecimal/Demo/classes/Dates.py
   python/branches/py3k-cdecimal/Demo/classes/README
   python/branches/py3k-cdecimal/Demo/classes/Range.py
   python/branches/py3k-cdecimal/Demo/classes/Rev.py
   python/branches/py3k-cdecimal/Demo/classes/Vec.py
   python/branches/py3k-cdecimal/Demo/embed/Makefile
   python/branches/py3k-cdecimal/Demo/embed/demo.c
   python/branches/py3k-cdecimal/Demo/md5test/md5driver.py
   python/branches/py3k-cdecimal/Demo/parser/test_parser.py
   python/branches/py3k-cdecimal/Demo/parser/unparse.py
   python/branches/py3k-cdecimal/Doc/Makefile
   python/branches/py3k-cdecimal/Doc/README.txt
   python/branches/py3k-cdecimal/Doc/c-api/arg.rst
   python/branches/py3k-cdecimal/Doc/c-api/datetime.rst
   python/branches/py3k-cdecimal/Doc/c-api/file.rst
   python/branches/py3k-cdecimal/Doc/c-api/init.rst
   python/branches/py3k-cdecimal/Doc/c-api/intro.rst
   python/branches/py3k-cdecimal/Doc/c-api/number.rst
   python/branches/py3k-cdecimal/Doc/c-api/unicode.rst
   python/branches/py3k-cdecimal/Doc/data/refcounts.dat
   python/branches/py3k-cdecimal/Doc/distutils/apiref.rst
   python/branches/py3k-cdecimal/Doc/distutils/builtdist.rst
   python/branches/py3k-cdecimal/Doc/distutils/extending.rst
   python/branches/py3k-cdecimal/Doc/distutils/setupscript.rst
   python/branches/py3k-cdecimal/Doc/distutils/sourcedist.rst
   python/branches/py3k-cdecimal/Doc/documenting/index.rst
   python/branches/py3k-cdecimal/Doc/documenting/markup.rst
   python/branches/py3k-cdecimal/Doc/documenting/style.rst
   python/branches/py3k-cdecimal/Doc/faq/extending.rst
   python/branches/py3k-cdecimal/Doc/faq/gui.rst
   python/branches/py3k-cdecimal/Doc/glossary.rst
   python/branches/py3k-cdecimal/Doc/howto/functional.rst
   python/branches/py3k-cdecimal/Doc/howto/unicode.rst
   python/branches/py3k-cdecimal/Doc/library/2to3.rst
   python/branches/py3k-cdecimal/Doc/library/ast.rst
   python/branches/py3k-cdecimal/Doc/library/calendar.rst
   python/branches/py3k-cdecimal/Doc/library/cmath.rst
   python/branches/py3k-cdecimal/Doc/library/codecs.rst
   python/branches/py3k-cdecimal/Doc/library/contextlib.rst
   python/branches/py3k-cdecimal/Doc/library/datetime.rst
   python/branches/py3k-cdecimal/Doc/library/dis.rst
   python/branches/py3k-cdecimal/Doc/library/doctest.rst
   python/branches/py3k-cdecimal/Doc/library/fnmatch.rst
   python/branches/py3k-cdecimal/Doc/library/functions.rst
   python/branches/py3k-cdecimal/Doc/library/http.cookiejar.rst
   python/branches/py3k-cdecimal/Doc/library/http.server.rst
   python/branches/py3k-cdecimal/Doc/library/importlib.rst
   python/branches/py3k-cdecimal/Doc/library/logging.rst
   python/branches/py3k-cdecimal/Doc/library/math.rst
   python/branches/py3k-cdecimal/Doc/library/multiprocessing.rst
   python/branches/py3k-cdecimal/Doc/library/os.path.rst
   python/branches/py3k-cdecimal/Doc/library/os.rst
   python/branches/py3k-cdecimal/Doc/library/pdb.rst
   python/branches/py3k-cdecimal/Doc/library/re.rst
   python/branches/py3k-cdecimal/Doc/library/select.rst
   python/branches/py3k-cdecimal/Doc/library/socket.rst
   python/branches/py3k-cdecimal/Doc/library/sqlite3.rst
   python/branches/py3k-cdecimal/Doc/library/ssl.rst
   python/branches/py3k-cdecimal/Doc/library/stdtypes.rst
   python/branches/py3k-cdecimal/Doc/library/string.rst
   python/branches/py3k-cdecimal/Doc/library/struct.rst
   python/branches/py3k-cdecimal/Doc/library/subprocess.rst
   python/branches/py3k-cdecimal/Doc/library/sys.rst
   python/branches/py3k-cdecimal/Doc/library/sysconfig.rst
   python/branches/py3k-cdecimal/Doc/library/unittest.rst
   python/branches/py3k-cdecimal/Doc/library/urllib.parse.rst
   python/branches/py3k-cdecimal/Doc/library/urllib.request.rst
   python/branches/py3k-cdecimal/Doc/library/warnings.rst
   python/branches/py3k-cdecimal/Doc/reference/compound_stmts.rst
   python/branches/py3k-cdecimal/Doc/reference/datamodel.rst
   python/branches/py3k-cdecimal/Doc/reference/expressions.rst
   python/branches/py3k-cdecimal/Doc/reference/introduction.rst
   python/branches/py3k-cdecimal/Doc/reference/lexical_analysis.rst
   python/branches/py3k-cdecimal/Doc/tools/sphinxext/indexsidebar.html
   python/branches/py3k-cdecimal/Doc/tools/sphinxext/pyspecific.py
   python/branches/py3k-cdecimal/Doc/tutorial/classes.rst
   python/branches/py3k-cdecimal/Doc/tutorial/controlflow.rst
   python/branches/py3k-cdecimal/Doc/tutorial/introduction.rst
   python/branches/py3k-cdecimal/Doc/using/cmdline.rst
   python/branches/py3k-cdecimal/Doc/using/mac.rst
   python/branches/py3k-cdecimal/Doc/whatsnew/2.7.rst
   python/branches/py3k-cdecimal/Doc/whatsnew/3.0.rst
   python/branches/py3k-cdecimal/Doc/whatsnew/3.2.rst
   python/branches/py3k-cdecimal/Grammar/Grammar
   python/branches/py3k-cdecimal/Include/longobject.h
   python/branches/py3k-cdecimal/Include/pyctype.h
   python/branches/py3k-cdecimal/Include/structseq.h
   python/branches/py3k-cdecimal/Include/traceback.h
   python/branches/py3k-cdecimal/Include/unicodeobject.h
   python/branches/py3k-cdecimal/Lib/_abcoll.py
   python/branches/py3k-cdecimal/Lib/_strptime.py
   python/branches/py3k-cdecimal/Lib/ast.py
   python/branches/py3k-cdecimal/Lib/contextlib.py
   python/branches/py3k-cdecimal/Lib/ctypes/__init__.py
   python/branches/py3k-cdecimal/Lib/ctypes/test/test_callbacks.py
   python/branches/py3k-cdecimal/Lib/ctypes/test/test_win32.py
   python/branches/py3k-cdecimal/Lib/decimal.py
   python/branches/py3k-cdecimal/Lib/dis.py
   python/branches/py3k-cdecimal/Lib/distutils/command/install.py
   python/branches/py3k-cdecimal/Lib/distutils/msvc9compiler.py
   python/branches/py3k-cdecimal/Lib/distutils/tests/test_sdist.py
   python/branches/py3k-cdecimal/Lib/distutils/unixccompiler.py
   python/branches/py3k-cdecimal/Lib/email/feedparser.py
   python/branches/py3k-cdecimal/Lib/email/test/test_email.py
   python/branches/py3k-cdecimal/Lib/encodings/aliases.py
   python/branches/py3k-cdecimal/Lib/fnmatch.py
   python/branches/py3k-cdecimal/Lib/heapq.py
   python/branches/py3k-cdecimal/Lib/http/cookiejar.py
   python/branches/py3k-cdecimal/Lib/idlelib/textView.py
   python/branches/py3k-cdecimal/Lib/importlib/__init__.py
   python/branches/py3k-cdecimal/Lib/importlib/_bootstrap.py
   python/branches/py3k-cdecimal/Lib/importlib/abc.py
   python/branches/py3k-cdecimal/Lib/importlib/test/benchmark.py
   python/branches/py3k-cdecimal/Lib/importlib/test/extension/test_case_sensitivity.py
   python/branches/py3k-cdecimal/Lib/importlib/test/extension/test_finder.py
   python/branches/py3k-cdecimal/Lib/importlib/test/extension/test_loader.py
   python/branches/py3k-cdecimal/Lib/importlib/test/extension/test_path_hook.py
   python/branches/py3k-cdecimal/Lib/importlib/test/import_/test_path.py
   python/branches/py3k-cdecimal/Lib/importlib/test/regrtest.py
   python/branches/py3k-cdecimal/Lib/importlib/test/source/test_abc_loader.py
   python/branches/py3k-cdecimal/Lib/importlib/test/source/test_case_sensitivity.py
   python/branches/py3k-cdecimal/Lib/importlib/test/source/test_file_loader.py
   python/branches/py3k-cdecimal/Lib/importlib/test/source/test_finder.py
   python/branches/py3k-cdecimal/Lib/importlib/test/source/test_path_hook.py
   python/branches/py3k-cdecimal/Lib/importlib/test/source/test_source_encoding.py
   python/branches/py3k-cdecimal/Lib/importlib/test/test_abc.py
   python/branches/py3k-cdecimal/Lib/importlib/test/test_api.py
   python/branches/py3k-cdecimal/Lib/importlib/test/test_util.py
   python/branches/py3k-cdecimal/Lib/inspect.py
   python/branches/py3k-cdecimal/Lib/lib2to3/   (props changed)
   python/branches/py3k-cdecimal/Lib/lib2to3/fixes/fix_itertools_imports.py
   python/branches/py3k-cdecimal/Lib/lib2to3/tests/test_fixers.py
   python/branches/py3k-cdecimal/Lib/logging/config.py
   python/branches/py3k-cdecimal/Lib/macpath.py
   python/branches/py3k-cdecimal/Lib/ntpath.py
   python/branches/py3k-cdecimal/Lib/os2emxpath.py
   python/branches/py3k-cdecimal/Lib/pdb.doc
   python/branches/py3k-cdecimal/Lib/pdb.py
   python/branches/py3k-cdecimal/Lib/pickletools.py
   python/branches/py3k-cdecimal/Lib/posixpath.py
   python/branches/py3k-cdecimal/Lib/pydoc.py
   python/branches/py3k-cdecimal/Lib/site.py
   python/branches/py3k-cdecimal/Lib/smtpd.py
   python/branches/py3k-cdecimal/Lib/sndhdr.py
   python/branches/py3k-cdecimal/Lib/socket.py
   python/branches/py3k-cdecimal/Lib/subprocess.py
   python/branches/py3k-cdecimal/Lib/sysconfig.py
   python/branches/py3k-cdecimal/Lib/tarfile.py
   python/branches/py3k-cdecimal/Lib/test/decimaltestdata/extra.decTest
   python/branches/py3k-cdecimal/Lib/test/math_testcases.txt
   python/branches/py3k-cdecimal/Lib/test/regrtest.py
   python/branches/py3k-cdecimal/Lib/test/support.py
   python/branches/py3k-cdecimal/Lib/test/test_ast.py
   python/branches/py3k-cdecimal/Lib/test/test_audioop.py
   python/branches/py3k-cdecimal/Lib/test/test_calendar.py
   python/branches/py3k-cdecimal/Lib/test/test_cmath.py
   python/branches/py3k-cdecimal/Lib/test/test_codeccallbacks.py
   python/branches/py3k-cdecimal/Lib/test/test_codecs.py
   python/branches/py3k-cdecimal/Lib/test/test_collections.py
   python/branches/py3k-cdecimal/Lib/test/test_complex.py
   python/branches/py3k-cdecimal/Lib/test/test_contextlib.py
   python/branches/py3k-cdecimal/Lib/test/test_datetime.py
   python/branches/py3k-cdecimal/Lib/test/test_decimal.py
   python/branches/py3k-cdecimal/Lib/test/test_descr.py
   python/branches/py3k-cdecimal/Lib/test/test_dict.py
   python/branches/py3k-cdecimal/Lib/test/test_dis.py
   python/branches/py3k-cdecimal/Lib/test/test_enumerate.py
   python/branches/py3k-cdecimal/Lib/test/test_exceptions.py
   python/branches/py3k-cdecimal/Lib/test/test_extcall.py
   python/branches/py3k-cdecimal/Lib/test/test_float.py
   python/branches/py3k-cdecimal/Lib/test/test_fnmatch.py
   python/branches/py3k-cdecimal/Lib/test/test_gdb.py
   python/branches/py3k-cdecimal/Lib/test/test_genericpath.py
   python/branches/py3k-cdecimal/Lib/test/test_getargs2.py
   python/branches/py3k-cdecimal/Lib/test/test_glob.py
   python/branches/py3k-cdecimal/Lib/test/test_http_cookiejar.py
   python/branches/py3k-cdecimal/Lib/test/test_httpservers.py
   python/branches/py3k-cdecimal/Lib/test/test_imp.py
   python/branches/py3k-cdecimal/Lib/test/test_import.py
   python/branches/py3k-cdecimal/Lib/test/test_inspect.py
   python/branches/py3k-cdecimal/Lib/test/test_iter.py
   python/branches/py3k-cdecimal/Lib/test/test_keywordonlyarg.py
   python/branches/py3k-cdecimal/Lib/test/test_logging.py
   python/branches/py3k-cdecimal/Lib/test/test_math.py
   python/branches/py3k-cdecimal/Lib/test/test_optparse.py
   python/branches/py3k-cdecimal/Lib/test/test_os.py
   python/branches/py3k-cdecimal/Lib/test/test_parser.py
   python/branches/py3k-cdecimal/Lib/test/test_platform.py
   python/branches/py3k-cdecimal/Lib/test/test_posixpath.py
   python/branches/py3k-cdecimal/Lib/test/test_property.py
   python/branches/py3k-cdecimal/Lib/test/test_pyexpat.py
   python/branches/py3k-cdecimal/Lib/test/test_runpy.py
   python/branches/py3k-cdecimal/Lib/test/test_shutil.py
   python/branches/py3k-cdecimal/Lib/test/test_strtod.py
   python/branches/py3k-cdecimal/Lib/test/test_struct.py
   python/branches/py3k-cdecimal/Lib/test/test_structseq.py
   python/branches/py3k-cdecimal/Lib/test/test_subprocess.py
   python/branches/py3k-cdecimal/Lib/test/test_syntax.py
   python/branches/py3k-cdecimal/Lib/test/test_sysconfig.py
   python/branches/py3k-cdecimal/Lib/test/test_tarfile.py
   python/branches/py3k-cdecimal/Lib/test/test_threaded_import.py
   python/branches/py3k-cdecimal/Lib/test/test_tokenize.py
   python/branches/py3k-cdecimal/Lib/test/test_tuple.py
   python/branches/py3k-cdecimal/Lib/test/test_unicode.py
   python/branches/py3k-cdecimal/Lib/test/test_unpack.py
   python/branches/py3k-cdecimal/Lib/test/test_urllib.py
   python/branches/py3k-cdecimal/Lib/test/test_urllib2.py
   python/branches/py3k-cdecimal/Lib/test/test_urlparse.py
   python/branches/py3k-cdecimal/Lib/test/test_xmlrpc_net.py
   python/branches/py3k-cdecimal/Lib/unittest/case.py
   python/branches/py3k-cdecimal/Lib/unittest/test/test_case.py
   python/branches/py3k-cdecimal/Lib/urllib/parse.py
   python/branches/py3k-cdecimal/Lib/urllib/request.py
   python/branches/py3k-cdecimal/Lib/uuid.py
   python/branches/py3k-cdecimal/Lib/warnings.py
   python/branches/py3k-cdecimal/Lib/xml/dom/expatbuilder.py
   python/branches/py3k-cdecimal/Makefile.pre.in
   python/branches/py3k-cdecimal/Misc/ACKS
   python/branches/py3k-cdecimal/Misc/NEWS
   python/branches/py3k-cdecimal/Misc/Vim/vimrc
   python/branches/py3k-cdecimal/Misc/maintainers.rst
   python/branches/py3k-cdecimal/Modules/_ctypes/callproc.c
   python/branches/py3k-cdecimal/Modules/_ctypes/libffi/fficonfig.py.in
   python/branches/py3k-cdecimal/Modules/_ctypes/libffi_msvc/ffi.h
   python/branches/py3k-cdecimal/Modules/_ctypes/libffi_msvc/fficonfig.h
   python/branches/py3k-cdecimal/Modules/_io/_iomodule.c
   python/branches/py3k-cdecimal/Modules/_math.c
   python/branches/py3k-cdecimal/Modules/_scproxy.c
   python/branches/py3k-cdecimal/Modules/_ssl.c
   python/branches/py3k-cdecimal/Modules/_struct.c
   python/branches/py3k-cdecimal/Modules/_testcapimodule.c
   python/branches/py3k-cdecimal/Modules/audioop.c
   python/branches/py3k-cdecimal/Modules/cmathmodule.c
   python/branches/py3k-cdecimal/Modules/datetimemodule.c
   python/branches/py3k-cdecimal/Modules/makesetup
   python/branches/py3k-cdecimal/Modules/mathmodule.c
   python/branches/py3k-cdecimal/Modules/parsermodule.c
   python/branches/py3k-cdecimal/Modules/posixmodule.c
   python/branches/py3k-cdecimal/Modules/selectmodule.c
   python/branches/py3k-cdecimal/Modules/timemodule.c
   python/branches/py3k-cdecimal/Modules/xxmodule.c
   python/branches/py3k-cdecimal/Objects/abstract.c
   python/branches/py3k-cdecimal/Objects/bytearrayobject.c
   python/branches/py3k-cdecimal/Objects/bytes_methods.c
   python/branches/py3k-cdecimal/Objects/descrobject.c
   python/branches/py3k-cdecimal/Objects/enumobject.c
   python/branches/py3k-cdecimal/Objects/exceptions.c
   python/branches/py3k-cdecimal/Objects/floatobject.c
   python/branches/py3k-cdecimal/Objects/frameobject.c
   python/branches/py3k-cdecimal/Objects/memoryobject.c
   python/branches/py3k-cdecimal/Objects/stringlib/formatter.h
   python/branches/py3k-cdecimal/Objects/structseq.c
   python/branches/py3k-cdecimal/Objects/typeobject.c
   python/branches/py3k-cdecimal/Objects/unicodeobject.c
   python/branches/py3k-cdecimal/PC/VS8.0/pythoncore.vcproj
   python/branches/py3k-cdecimal/PCbuild/_ctypes.vcproj
   python/branches/py3k-cdecimal/PCbuild/pythoncore.vcproj
   python/branches/py3k-cdecimal/Parser/Python.asdl
   python/branches/py3k-cdecimal/Parser/asdl_c.py
   python/branches/py3k-cdecimal/Python/Python-ast.c
   python/branches/py3k-cdecimal/Python/_warnings.c
   python/branches/py3k-cdecimal/Python/ast.c
   python/branches/py3k-cdecimal/Python/ceval.c
   python/branches/py3k-cdecimal/Python/compile.c
   python/branches/py3k-cdecimal/Python/getargs.c
   python/branches/py3k-cdecimal/Python/graminit.c
   python/branches/py3k-cdecimal/Python/import.c
   python/branches/py3k-cdecimal/Python/pythonrun.c
   python/branches/py3k-cdecimal/Python/sysmodule.c
   python/branches/py3k-cdecimal/Python/traceback.c
   python/branches/py3k-cdecimal/README
   python/branches/py3k-cdecimal/Tools/README
   python/branches/py3k-cdecimal/Tools/gdb/libpython.py
   python/branches/py3k-cdecimal/Tools/i18n/msgfmt.py
   python/branches/py3k-cdecimal/Tools/scripts/   (props changed)
   python/branches/py3k-cdecimal/Tools/scripts/patchcheck.py
   python/branches/py3k-cdecimal/configure
   python/branches/py3k-cdecimal/configure.in
   python/branches/py3k-cdecimal/setup.py

Modified: python/branches/py3k-cdecimal/Demo/classes/Complex.py
==============================================================================
--- python/branches/py3k-cdecimal/Demo/classes/Complex.py	(original)
+++ python/branches/py3k-cdecimal/Demo/classes/Complex.py	Sat Jul 17 12:36:09 2010
@@ -151,13 +151,9 @@
             raise ValueError("can't convert Complex with nonzero im to float")
         return float(self.re)
 
-    def __cmp__(self, other):
+    def __eq__(self, other):
         other = ToComplex(other)
-        return cmp((self.re, self.im), (other.re, other.im))
-
-    def __rcmp__(self, other):
-        other = ToComplex(other)
-        return cmp(other, self)
+        return (self.re, self.im) == (other.re, other.im)
 
     def __bool__(self):
         return not (self.re == self.im == 0)
@@ -190,14 +186,14 @@
 
     __rmul__ = __mul__
 
-    def __div__(self, other):
+    def __truediv__(self, other):
         other = ToComplex(other)
         d = float(other.re*other.re + other.im*other.im)
         if not d: raise ZeroDivisionError('Complex division')
         return Complex((self.re*other.re + self.im*other.im) / d,
                        (self.im*other.re - self.re*other.im) / d)
 
-    def __rdiv__(self, other):
+    def __rtruediv__(self, other):
         other = ToComplex(other)
         return other / self
 
@@ -226,8 +222,9 @@
     print('       ', a, 'and', b, end=' ')
     try:
         result = eval(expr)
-    except:
-        result = sys.exc_info()[0]
+    except Exception as e:
+        print('!!\t!!\t!! error: {}'.format(e))
+        return
     print('->', result)
     if isinstance(result, str) or isinstance(value, str):
         ok = (result == value)
@@ -295,13 +292,6 @@
                     (Complex(1), Complex(0,10), 1),
                     (2, Complex(4,0), 16),
             ],
-            'cmp(a,b)': [
-                    (1, 10, -1),
-                    (1, Complex(0,10), 1),
-                    (Complex(0,10), 1, -1),
-                    (Complex(0,10), Complex(1), -1),
-                    (Complex(1), Complex(0,10), 1),
-            ],
     }
     for expr in sorted(testsuite):
         print(expr + ':')

Modified: python/branches/py3k-cdecimal/Demo/classes/Dates.py
==============================================================================
--- python/branches/py3k-cdecimal/Demo/classes/Dates.py	(original)
+++ python/branches/py3k-cdecimal/Demo/classes/Dates.py	Sat Jul 17 12:36:09 2010
@@ -1,45 +1,47 @@
-# Class Date supplies date objects that support date arithmetic.
-#
-# Date(month,day,year) returns a Date object.  An instance prints as,
-# e.g., 'Mon 16 Aug 1993'.
-#
-# Addition, subtraction, comparison operators, min, max, and sorting
-# all work as expected for date objects:  int+date or date+int returns
-# the date `int' days from `date'; date+date raises an exception;
-# date-int returns the date `int' days before `date'; date2-date1 returns
-# an integer, the number of days from date1 to date2; int-date raises an
-# exception; date1 < date2 is true iff date1 occurs before date2 (&
-# similarly for other comparisons); min(date1,date2) is the earlier of
-# the two dates and max(date1,date2) the later; and date objects can be
-# used as dictionary keys.
-#
-# Date objects support one visible method, date.weekday().  This returns
-# the day of the week the date falls on, as a string.
-#
-# Date objects also have 4 read-only data attributes:
-#   .month  in 1..12
-#   .day    in 1..31
-#   .year   int or long int
-#   .ord    the ordinal of the date relative to an arbitrary staring point
-#
-# The Dates module also supplies function today(), which returns the
-# current date as a date object.
-#
-# Those entranced by calendar trivia will be disappointed, as no attempt
-# has been made to accommodate the Julian (etc) system.  On the other
-# hand, at least this package knows that 2000 is a leap year but 2100
-# isn't, and works fine for years with a hundred decimal digits <wink>.
-
-# Tim Peters   tim at ksr.com
-# not speaking for Kendall Square Research Corp
+"""
+Class Date supplies date objects that support date arithmetic.
 
-# Adapted to Python 1.1 (where some hacks to overcome coercion are unnecessary)
-# by Guido van Rossum
+Date(month,day,year) returns a Date object.  An instance prints as,
+e.g., 'Mon 16 Aug 1993'.
 
-# Note that as of Python 2.3, a datetime module is included in the stardard
-# library.
+Addition, subtraction, comparison operators, min, max, and sorting
+all work as expected for date objects:  int+date or date+int returns
+the date `int' days from `date'; date+date raises an exception;
+date-int returns the date `int' days before `date'; date2-date1 returns
+an integer, the number of days from date1 to date2; int-date raises an
+exception; date1 < date2 is true iff date1 occurs before date2 (&
+similarly for other comparisons); min(date1,date2) is the earlier of
+the two dates and max(date1,date2) the later; and date objects can be
+used as dictionary keys.
+
+Date objects support one visible method, date.weekday().  This returns
+the day of the week the date falls on, as a string.
+
+Date objects also have 4 read-only data attributes:
+  .month  in 1..12
+  .day    in 1..31
+  .year   int or long int
+  .ord    the ordinal of the date relative to an arbitrary staring point
+
+The Dates module also supplies function today(), which returns the
+current date as a date object.
+
+Those entranced by calendar trivia will be disappointed, as no attempt
+has been made to accommodate the Julian (etc) system.  On the other
+hand, at least this package knows that 2000 is a leap year but 2100
+isn't, and works fine for years with a hundred decimal digits <wink>.
+
+Tim Peters   tim at ksr.com
+not speaking for Kendall Square Research Corp
+
+Adapted to Python 1.1 (where some hacks to overcome coercion are unnecessary)
+by Guido van Rossum
+
+Note that as of Python 2.3, a datetime module is included in the stardard
+library.
+"""
 
-# vi:set tabsize=8:
+import functools
 
 _MONTH_NAMES = [ 'January', 'February', 'March', 'April', 'May',
                  'June', 'July', 'August', 'September', 'October',
@@ -57,8 +59,6 @@
     dbm = dbm + dim
 del dbm, dim
 
-_INT_TYPES = type(1), type(1)
-
 def _is_leap(year):           # 1 if leap year, else 0
     if year % 4 != 0: return 0
     if year % 400 == 0: return 1
@@ -85,11 +85,12 @@
 _DI400Y = _days_before_year(400)      # number of days in 400 years
 
 def _num2date(n):             # return date with ordinal n
-    if type(n) not in _INT_TYPES:
+    if not isinstance(n, int):
         raise TypeError('argument must be integer: %r' % type(n))
 
-    ans = Date(1,1,1)   # arguments irrelevant; just getting a Date obj
-    del ans.ord, ans.month, ans.day, ans.year # un-initialize it
+    # Get uninitialized Date object.  This is necesary because once
+    # attributes are set, they cannot be changed.
+    ans = Date.__new__(Date)
     ans.ord = n
 
     n400 = (n-1)//_DI400Y                # # of 400-year blocks preceding
@@ -99,11 +100,7 @@
     if dby >= n:
         more = more - 1
         dby = dby - _days_in_year(more)
-    year, n = year + more, int(n - dby)
-
-    try: year = int(year)               # chop to int, if it fits
-    except (ValueError, OverflowError): pass
-
+    year, n = year + more, n - dby
     month = min(n//29 + 1, 12)
     dbm = _days_before_month(month, year)
     if dbm >= n:
@@ -114,9 +111,9 @@
     return ans
 
 def _num2day(n):      # return weekday name of day with ordinal n
-    return _DAY_NAMES[ int(n % 7) ]
-
+    return _DAY_NAMES[n % 7]
 
+ at functools.total_ordering
 class Date:
     def __init__(self, month, day, year):
         if not 1 <= month <= 12:
@@ -124,7 +121,7 @@
         dim = _days_in_month(month, year)
         if not 1 <= day <= dim:
             raise ValueError('day must be in 1..%r: %r' % (dim, day))
-        self.month, self.day, self.year = month, day, year
+        self.month, self.day, self.year = map(int, (month, day, year))
         self.ord = _date2num(self)
 
     # don't allow setting existing attributes
@@ -133,8 +130,11 @@
             raise AttributeError('read-only attribute ' + name)
         self.__dict__[name] = value
 
-    def __cmp__(self, other):
-        return cmp(self.ord, other.ord)
+    def __eq__(self, other):
+        return self.ord == other.ord
+
+    def __lt__(self, other):
+        return self.ord < other.ord
 
     # define a hash function so dates can be used as dictionary keys
     def __hash__(self):
@@ -150,14 +150,14 @@
 
     # Python 1.1 coerces neither int+date nor date+int
     def __add__(self, n):
-        if type(n) not in _INT_TYPES:
+        if not isinstance(n, int):
             raise TypeError('can\'t add %r to date' % type(n))
         return _num2date(self.ord + n)
     __radd__ = __add__ # handle int+date
 
     # Python 1.1 coerces neither date-int nor date-date
     def __sub__(self, other):
-        if type(other) in _INT_TYPES:           # date-int
+        if isinstance(other, int):           # date-int
             return _num2date(self.ord - other)
         else:
             return self.ord - other.ord         # date-date

Deleted: python/branches/py3k-cdecimal/Demo/classes/Dbm.py
==============================================================================
--- python/branches/py3k-cdecimal/Demo/classes/Dbm.py	Sat Jul 17 12:36:09 2010
+++ (empty file)
@@ -1,66 +0,0 @@
-# A wrapper around the (optional) built-in class dbm, supporting keys
-# and values of almost any type instead of just string.
-# (Actually, this works only for keys and values that can be read back
-# correctly after being converted to a string.)
-
-
-class Dbm:
-
-    def __init__(self, filename, mode, perm):
-        import dbm.ndbm
-        self.db = dbm.ndbm.open(filename, mode, perm)
-
-    def __repr__(self):
-        s = ''
-        for key in self.keys():
-            t = repr(key) + ': ' + repr(self[key])
-            if s: t = ', ' + t
-            s = s + t
-        return '{' + s + '}'
-
-    def __len__(self):
-        return len(self.db)
-
-    def __getitem__(self, key):
-        return eval(self.db[repr(key)])
-
-    def __setitem__(self, key, value):
-        self.db[repr(key)] = repr(value)
-
-    def __delitem__(self, key):
-        del self.db[repr(key)]
-
-    def keys(self):
-        res = []
-        for key in self.db.keys():
-            res.append(eval(key))
-        return res
-
-    def has_key(self, key):
-        return repr(key) in self.db
-
-
-def test():
-    d = Dbm('@dbm', 'rw', 0o600)
-    print(d)
-    while 1:
-        try:
-            key = eval(input('key: '))
-            if key in d:
-                value = d[key]
-                print('currently:', value)
-            value = eval(input('value: '))
-            if value is None:
-                del d[key]
-            else:
-                d[key] = value
-        except KeyboardInterrupt:
-            print('')
-            print(d)
-        except EOFError:
-            print('[eof]')
-            break
-    print(d)
-
-
-test()

Modified: python/branches/py3k-cdecimal/Demo/classes/README
==============================================================================
--- python/branches/py3k-cdecimal/Demo/classes/README	(original)
+++ python/branches/py3k-cdecimal/Demo/classes/README	Sat Jul 17 12:36:09 2010
@@ -2,7 +2,6 @@
 
 Complex.py	Complex numbers
 Dates.py	Date manipulation package by Tim Peters
-Dbm.py		Wrapper around built-in dbm, supporting	arbitrary values
 Range.py	Example of a generator: re-implement built-in range()
 Rev.py		Yield the reverse of a sequence
 Vec.py		A simple vector class

Modified: python/branches/py3k-cdecimal/Demo/classes/Range.py
==============================================================================
--- python/branches/py3k-cdecimal/Demo/classes/Range.py	(original)
+++ python/branches/py3k-cdecimal/Demo/classes/Range.py	Sat Jul 17 12:36:09 2010
@@ -66,7 +66,7 @@
 def test():
     import time, builtins
     #Just a quick sanity check
-    correct_result = builtins.range(5, 100, 3)
+    correct_result = list(builtins.range(5, 100, 3))
     oldrange_result = list(oldrange(5, 100, 3))
     genrange_result = list(genrange(5, 100, 3))
     if genrange_result != correct_result or oldrange_result != correct_result:

Modified: python/branches/py3k-cdecimal/Demo/classes/Rev.py
==============================================================================
--- python/branches/py3k-cdecimal/Demo/classes/Rev.py	(original)
+++ python/branches/py3k-cdecimal/Demo/classes/Rev.py	Sat Jul 17 12:36:09 2010
@@ -4,8 +4,7 @@
 
 It works on mutable or inmutable sequences.
 
->>> chars = list(Rev('Hello World!'))
->>> print ''.join(chars)
+>>> Rev('Hello World!')
 !dlroW olleH
 
 The .forw is so you can use anonymous sequences in __init__, and still
@@ -16,16 +15,16 @@
 confusion. Maybe it should be change to copy input sequence to break
 the connection completely ? )
 
->>> nnn = range(3)
+>>> nnn = list(range(3))
 >>> rnn = Rev(nnn)
->>> for n in rnn: print n
+>>> for n in rnn: n
 ...
 2
 1
 0
 >>> for n in range(4, 6): nnn.append(n)   # update nnn
 ...
->>> for n in rnn: print n     # prints reversed updated values
+>>> for n in rnn: n     # prints reversed updated values
 ...
 5
 4
@@ -35,7 +34,7 @@
 >>> nnn = nnn[1:-1]
 >>> nnn
 [1, 2, 4]
->>> for n in rnn: print n     # prints reversed values of old nnn
+>>> for n in rnn: n     # prints reversed values of old nnn
 ...
 5
 4
@@ -45,16 +44,15 @@
 
 #
 >>> WH = Rev('Hello World!')
->>> print WH.forw, WH.back
+>>> print(WH.forw, WH.back)
 Hello World! !dlroW olleH
->>> nnn = Rev(range(1, 10))
->>> print nnn.forw
+>>> nnn = Rev(list(range(1, 10)))
+>>> print(nnn.forw)
 [1, 2, 3, 4, 5, 6, 7, 8, 9]
->>> print nnn.back
+>>> print(nnn.back)
 [9, 8, 7, 6, 5, 4, 3, 2, 1]
 
->>> rrr = Rev(nnn)
->>> rrr
+>>> Rev(nnn)
 <1, 2, 3, 4, 5, 6, 7, 8, 9>
 
 '''

Modified: python/branches/py3k-cdecimal/Demo/classes/Vec.py
==============================================================================
--- python/branches/py3k-cdecimal/Demo/classes/Vec.py	(original)
+++ python/branches/py3k-cdecimal/Demo/classes/Vec.py	Sat Jul 17 12:36:09 2010
@@ -1,23 +1,41 @@
-# A simple vector class
+class Vec:
+    """ A simple vector class
 
+    Instances of the Vec class  can be constructed from numbers
 
-def vec(*v):
-    return Vec(*v)
+    >>> a = Vec(1, 2, 3)
+    >>> b = Vec(3, 2, 1)
 
+    added
+    >>> a + b
+    Vec(4, 4, 4)
 
-class Vec:
+    subtracted
+    >>> a - b
+    Vec(-2, 0, 2)
+
+    and multiplied by a scalar on the left
+    >>> 3.0 * a
+    Vec(3.0, 6.0, 9.0)
 
+    or on the right
+    >>> a * 3.0
+    Vec(3.0, 6.0, 9.0)
+    """
     def __init__(self, *v):
         self.v = list(v)
 
-    def fromlist(self, v):
+    @classmethod
+    def fromlist(cls, v):
         if not isinstance(v, list):
             raise TypeError
-        self.v = v[:]
-        return self
+        inst = cls()
+        inst.v = v
+        return inst
 
     def __repr__(self):
-        return 'vec(' + repr(self.v)[1:-1] + ')'
+        args = ', '.join(repr(x) for x in self.v)
+        return 'Vec({})'.format(args)
 
     def __len__(self):
         return len(self.v)
@@ -27,28 +45,24 @@
 
     def __add__(self, other):
         # Element-wise addition
-        v = list(map(lambda x, y: x+y, self, other))
-        return Vec().fromlist(v)
+        v = [x + y for x, y in zip(self.v, other.v)]
+        return Vec.fromlist(v)
 
     def __sub__(self, other):
         # Element-wise subtraction
-        v = list(map(lambda x, y: x-y, self, other))
-        return Vec().fromlist(v)
+        v = [x - y for x, y in zip(self.v, other.v)]
+        return Vec.fromlist(v)
 
     def __mul__(self, scalar):
         # Multiply by scalar
-        v = [x*scalar for x in self.v]
-        return Vec().fromlist(v)
+        v = [x * scalar for x in self.v]
+        return Vec.fromlist(v)
 
+    __rmul__ = __mul__
 
 
 def test():
-    a = vec(1, 2, 3)
-    b = vec(3, 2, 1)
-    print(a)
-    print(b)
-    print(a+b)
-    print(a-b)
-    print(a*3.0)
+    import doctest
+    doctest.testmod()
 
 test()

Modified: python/branches/py3k-cdecimal/Demo/embed/Makefile
==============================================================================
--- python/branches/py3k-cdecimal/Demo/embed/Makefile	(original)
+++ python/branches/py3k-cdecimal/Demo/embed/Makefile	Sat Jul 17 12:36:09 2010
@@ -1,5 +1,5 @@
 # Makefile for embedded Python use demo.
-# (This version tailored for my Red Hat Linux 6.1 setup;
+# (This version originally written on Red Hat Linux 6.1;
 # edit lines marked with XXX.)
 
 # XXX The compiler you are using
@@ -10,7 +10,7 @@
 srcdir=		../..
 
 # Python version
-VERSION=	3.0
+VERSION=	3.2
 
 # Compiler flags
 OPT=		-g
@@ -21,7 +21,7 @@
 # The Python library
 LIBPYTHON=	$(blddir)/libpython$(VERSION).a
 
-# XXX edit LIBS (in particular) to match $(blddir)/Modules/Makefile
+# XXX edit LIBS (in particular) to match $(blddir)/Makefile
 LIBS=		-lnsl -ldl -lreadline -ltermcap -lieee -lpthread -lutil
 LDFLAGS=	-Xlinker -export-dynamic
 SYSLIBS=	-lm

Modified: python/branches/py3k-cdecimal/Demo/embed/demo.c
==============================================================================
--- python/branches/py3k-cdecimal/Demo/embed/demo.c	(original)
+++ python/branches/py3k-cdecimal/Demo/embed/demo.c	Sat Jul 17 12:36:09 2010
@@ -20,10 +20,19 @@
     Py_Initialize();
 
     /* Define sys.argv.  It is up to the application if you
-       want this; you can also let it undefined (since the Python
+       want this; you can also leave it undefined (since the Python
        code is generally not a main program it has no business
-       touching sys.argv...) */
-    PySys_SetArgv(2, args);
+       touching sys.argv...) 
+
+       If the third argument is true, sys.path is modified to include
+       either the directory containing the script named by argv[0], or
+       the current working directory.  This can be risky; if you run
+       an application embedding Python in a directory controlled by
+       someone else, attackers could put a Trojan-horse module in the
+       directory (say, a file named os.py) that your application would
+       then import and run.
+    */
+    PySys_SetArgvEx(argc, argv, 0);
 
     /* Do some application specific code */
     printf("Hello, brave new world\n\n");

Modified: python/branches/py3k-cdecimal/Demo/md5test/md5driver.py
==============================================================================
--- python/branches/py3k-cdecimal/Demo/md5test/md5driver.py	(original)
+++ python/branches/py3k-cdecimal/Demo/md5test/md5driver.py	Sat Jul 17 12:36:09 2010
@@ -1,11 +1,10 @@
+from hashlib import md5
 import string
-import md5
 from sys import argv
 
 def MDPrint(str):
     outstr = ''
-    for i in str:
-        o = ord(i)
+    for o in str:
         outstr = (outstr
                   + string.hexdigits[(o >> 4) & 0xF]
                   + string.hexdigits[o & 0xF])
@@ -42,7 +41,7 @@
     print('MD5 time trial. Processing', TEST_BYTES, 'characters...')
     t1 = time()
 
-    mdContext = md5.new()
+    mdContext = md5()
 
     for i in range(TEST_BLOCKS):
         mdContext.update(data)
@@ -57,13 +56,13 @@
 
 
 def MDString(str):
-    MDPrint(md5.new(str).digest())
+    MDPrint(md5(str.encode("utf-8")).digest())
     print('"' + str + '"')
 
 
 def MDFile(filename):
     f = open(filename, 'rb')
-    mdContext = md5.new()
+    mdContext = md5()
 
     while 1:
         data = f.read(1024)
@@ -78,7 +77,7 @@
 import sys
 
 def MDFilter():
-    mdContext = md5.new()
+    mdContext = md5()
 
     while 1:
         data = sys.stdin.read(16)

Modified: python/branches/py3k-cdecimal/Demo/parser/test_parser.py
==============================================================================
--- python/branches/py3k-cdecimal/Demo/parser/test_parser.py	(original)
+++ python/branches/py3k-cdecimal/Demo/parser/test_parser.py	Sat Jul 17 12:36:09 2010
@@ -11,19 +11,19 @@
     global _numFailed
     print('----', fileName, end=' ')
     try:
-        ast = parser.suite(t)
-        tup = parser.ast2tuple(ast)
-        # this discards the first AST; a huge memory savings when running
+        st = parser.suite(t)
+        tup = parser.st2tuple(st)
+        # this discards the first ST; a huge memory savings when running
         # against a large source file like Tkinter.py.
-        ast = None
-        new = parser.tuple2ast(tup)
+        st = None
+        new = parser.tuple2st(tup)
     except parser.ParserError as err:
         print()
         print('parser module raised exception on input file', fileName + ':')
         traceback.print_exc()
         _numFailed = _numFailed + 1
     else:
-        if tup != parser.ast2tuple(new):
+        if tup != parser.st2tuple(new):
             print()
             print('parser module failed on input file', fileName)
             _numFailed = _numFailed + 1

Modified: python/branches/py3k-cdecimal/Demo/parser/unparse.py
==============================================================================
--- python/branches/py3k-cdecimal/Demo/parser/unparse.py	(original)
+++ python/branches/py3k-cdecimal/Demo/parser/unparse.py	Sat Jul 17 12:36:09 2010
@@ -1,9 +1,15 @@
 "Usage: unparse.py <path to source file>"
 import sys
-import _ast
+import math
+import ast
+import tokenize
 import io
 import os
 
+# Large float and imaginary literals get turned into infinities in the AST.
+# We unparse those infinities to INFSTR.
+INFSTR = "1e" + repr(sys.float_info.max_10_exp + 1)
+
 def interleave(inter, f, seq):
     """Call f on each item in seq, calling inter() in between.
     """
@@ -20,7 +26,7 @@
 class Unparser:
     """Methods in this class recursively traverse an AST and
     output source code for the abstract syntax; original formatting
-    is disregarged. """
+    is disregarded. """
 
     def __init__(self, tree, file = sys.stdout):
         """Unparser(tree, file=sys.stdout) -> None.
@@ -80,10 +86,11 @@
 
     def _ImportFrom(self, t):
         self.fill("from ")
-        self.write(t.module)
+        self.write("." * t.level)
+        if t.module:
+            self.write(t.module)
         self.write(" import ")
         interleave(lambda: self.write(", "), self.dispatch, t.names)
-        # XXX(jpe) what is level for?
 
     def _Assign(self, t):
         self.fill()
@@ -115,7 +122,7 @@
 
     def _Delete(self, t):
         self.fill("del ")
-        self.dispatch(t.targets)
+        interleave(lambda: self.write(", "), self.dispatch, t.targets)
 
     def _Assert(self, t):
         self.fill("assert ")
@@ -124,24 +131,14 @@
             self.write(", ")
             self.dispatch(t.msg)
 
-    def _Print(self, t):
-        self.fill("print ")
-        do_comma = False
-        if t.dest:
-            self.write(">>")
-            self.dispatch(t.dest)
-            do_comma = True
-        for e in t.values:
-            if do_comma:self.write(", ")
-            else:do_comma=True
-            self.dispatch(e)
-        if not t.nl:
-            self.write(",")
-
     def _Global(self, t):
         self.fill("global ")
         interleave(lambda: self.write(", "), self.write, t.names)
 
+    def _Nonlocal(self, t):
+        self.fill("nonlocal ")
+        interleave(lambda: self.write(", "), self.write, t.names)
+
     def _Yield(self, t):
         self.write("(")
         self.write("yield")
@@ -151,15 +148,15 @@
         self.write(")")
 
     def _Raise(self, t):
-        self.fill('raise ')
-        if t.type:
-            self.dispatch(t.type)
-        if t.inst:
-            self.write(", ")
-            self.dispatch(t.inst)
-        if t.tback:
-            self.write(", ")
-            self.dispatch(t.tback)
+        self.fill("raise")
+        if not t.exc:
+            assert not t.cause
+            return
+        self.write(" ")
+        self.dispatch(t.exc)
+        if t.cause:
+            self.write(" from ")
+            self.dispatch(t.cause)
 
     def _TryExcept(self, t):
         self.fill("try")
@@ -176,10 +173,14 @@
             self.leave()
 
     def _TryFinally(self, t):
-        self.fill("try")
-        self.enter()
-        self.dispatch(t.body)
-        self.leave()
+        if len(t.body) == 1 and isinstance(t.body[0], ast.TryExcept):
+            # try-except-finally
+            self.dispatch(t.body)
+        else:
+            self.fill("try")
+            self.enter()
+            self.dispatch(t.body)
+            self.leave()
 
         self.fill("finally")
         self.enter()
@@ -192,21 +193,40 @@
             self.write(" ")
             self.dispatch(t.type)
         if t.name:
-            self.write(", ")
-            self.dispatch(t.name)
+            self.write(" as ")
+            self.write(t.name)
         self.enter()
         self.dispatch(t.body)
         self.leave()
 
     def _ClassDef(self, t):
         self.write("\n")
+        for deco in t.decorator_list:
+            self.fill("@")
+            self.dispatch(deco)
         self.fill("class "+t.name)
-        if t.bases:
-            self.write("(")
-            for a in t.bases:
-                self.dispatch(a)
-                self.write(", ")
-            self.write(")")
+        self.write("(")
+        comma = False
+        for e in t.bases:
+            if comma: self.write(", ")
+            else: comma = True
+            self.dispatch(e)
+        for e in t.keywords:
+            if comma: self.write(", ")
+            else: comma = True
+            self.dispatch(e)
+        if t.starargs:
+            if comma: self.write(", ")
+            else: comma = True
+            self.write("*")
+            self.dispatch(t.starargs)
+        if t.kwargs:
+            if comma: self.write(", ")
+            else: comma = True
+            self.write("**")
+            self.dispatch(t.kwargs)
+        self.write(")")
+
         self.enter()
         self.dispatch(t.body)
         self.leave()
@@ -219,6 +239,9 @@
         self.fill("def "+t.name + "(")
         self.dispatch(t.args)
         self.write(")")
+        if t.returns:
+            self.write(" -> ")
+            self.dispatch(t.returns)
         self.enter()
         self.dispatch(t.body)
         self.leave()
@@ -235,15 +258,24 @@
             self.fill("else")
             self.enter()
             self.dispatch(t.orelse)
-            self.leave
+            self.leave()
 
     def _If(self, t):
         self.fill("if ")
         self.dispatch(t.test)
         self.enter()
-        # XXX elif?
         self.dispatch(t.body)
         self.leave()
+        # collapse nested ifs into equivalent elifs.
+        while (t.orelse and len(t.orelse) == 1 and
+               isinstance(t.orelse[0], ast.If)):
+            t = t.orelse[0]
+            self.fill("elif ")
+            self.dispatch(t.test)
+            self.enter()
+            self.dispatch(t.body)
+            self.leave()
+        # final else
         if t.orelse:
             self.fill("else")
             self.enter()
@@ -260,7 +292,7 @@
             self.fill("else")
             self.enter()
             self.dispatch(t.orelse)
-            self.leave
+            self.leave()
 
     def _With(self, t):
         self.fill("with ")
@@ -273,19 +305,18 @@
         self.leave()
 
     # expr
+    def _Bytes(self, t):
+        self.write(repr(t.s))
+
     def _Str(self, tree):
         self.write(repr(tree.s))
 
     def _Name(self, t):
         self.write(t.id)
 
-    def _Repr(self, t):
-        self.write("`")
-        self.dispatch(t.value)
-        self.write("`")
-
     def _Num(self, t):
-        self.write(repr(t.n))
+        # Substitute overflowing decimal literal for AST infinities.
+        self.write(repr(t.n).replace("inf", INFSTR))
 
     def _List(self, t):
         self.write("[")
@@ -306,6 +337,22 @@
             self.dispatch(gen)
         self.write(")")
 
+    def _SetComp(self, t):
+        self.write("{")
+        self.dispatch(t.elt)
+        for gen in t.generators:
+            self.dispatch(gen)
+        self.write("}")
+
+    def _DictComp(self, t):
+        self.write("{")
+        self.dispatch(t.key)
+        self.write(": ")
+        self.dispatch(t.value)
+        for gen in t.generators:
+            self.dispatch(gen)
+        self.write("}")
+
     def _comprehension(self, t):
         self.write(" for ")
         self.dispatch(t.target)
@@ -324,14 +371,20 @@
         self.dispatch(t.orelse)
         self.write(")")
 
+    def _Set(self, t):
+        assert(t.elts) # should be at least one element
+        self.write("{")
+        interleave(lambda: self.write(", "), self.dispatch, t.elts)
+        self.write("}")
+
     def _Dict(self, t):
         self.write("{")
-        def writem(xxx_todo_changeme):
-            (k, v) = xxx_todo_changeme
+        def write_pair(pair):
+            (k, v) = pair
             self.dispatch(k)
             self.write(": ")
             self.dispatch(v)
-        interleave(lambda: self.write(", "), writem, zip(t.keys, t.values))
+        interleave(lambda: self.write(", "), write_pair, zip(t.keys, t.values))
         self.write("}")
 
     def _Tuple(self, t):
@@ -346,13 +399,14 @@
 
     unop = {"Invert":"~", "Not": "not", "UAdd":"+", "USub":"-"}
     def _UnaryOp(self, t):
-        self.write(self.unop[t.op.__class__.__name__])
         self.write("(")
+        self.write(self.unop[t.op.__class__.__name__])
+        self.write(" ")
         self.dispatch(t.operand)
         self.write(")")
 
     binop = { "Add":"+", "Sub":"-", "Mult":"*", "Div":"/", "Mod":"%",
-                    "LShift":">>", "RShift":"<<", "BitOr":"|", "BitXor":"^", "BitAnd":"&",
+                    "LShift":"<<", "RShift":">>", "BitOr":"|", "BitXor":"^", "BitAnd":"&",
                     "FloorDiv":"//", "Pow": "**"}
     def _BinOp(self, t):
         self.write("(")
@@ -369,9 +423,9 @@
         for o, e in zip(t.ops, t.comparators):
             self.write(" " + self.cmpops[o.__class__.__name__] + " ")
             self.dispatch(e)
-            self.write(")")
+        self.write(")")
 
-    boolops = {_ast.And: 'and', _ast.Or: 'or'}
+    boolops = {ast.And: 'and', ast.Or: 'or'}
     def _BoolOp(self, t):
         self.write("(")
         s = " %s " % self.boolops[t.op.__class__]
@@ -380,6 +434,11 @@
 
     def _Attribute(self,t):
         self.dispatch(t.value)
+        # Special case: 3.__abs__() is a syntax error, so if t.value
+        # is an integer literal then we need to either parenthesize
+        # it or add an extra space to get 3 .__abs__().
+        if isinstance(t.value, ast.Num) and isinstance(t.value.n, int):
+            self.write(" ")
         self.write(".")
         self.write(t.attr)
 
@@ -433,28 +492,55 @@
     def _ExtSlice(self, t):
         interleave(lambda: self.write(', '), self.dispatch, t.dims)
 
+    # argument
+    def _arg(self, t):
+        self.write(t.arg)
+        if t.annotation:
+            self.write(": ")
+            self.dispatch(t.annotation)
+
     # others
     def _arguments(self, t):
         first = True
-        nonDef = len(t.args)-len(t.defaults)
-        for a in t.args[0:nonDef]:
+        # normal arguments
+        defaults = [None] * (len(t.args) - len(t.defaults)) + t.defaults
+        for a, d in zip(t.args, defaults):
             if first:first = False
             else: self.write(", ")
             self.dispatch(a)
-        for a,d in zip(t.args[nonDef:], t.defaults):
-            if first:first = False
-            else: self.write(", ")
-            self.dispatch(a),
-            self.write("=")
-            self.dispatch(d)
-        if t.vararg:
+            if d:
+                self.write("=")
+                self.dispatch(d)
+
+        # varargs, or bare '*' if no varargs but keyword-only arguments present
+        if t.vararg or t.kwonlyargs:
             if first:first = False
             else: self.write(", ")
-            self.write("*"+t.vararg)
+            self.write("*")
+            if t.vararg:
+                self.write(t.vararg)
+                if t.varargannotation:
+                    self.write(": ")
+                    self.dispatch(t.varargannotation)
+
+        # keyword-only arguments
+        if t.kwonlyargs:
+            for a, d in zip(t.kwonlyargs, t.kw_defaults):
+                if first:first = False
+                else: self.write(", ")
+                self.dispatch(a),
+                if d:
+                    self.write("=")
+                    self.dispatch(d)
+
+        # kwargs
         if t.kwarg:
             if first:first = False
             else: self.write(", ")
             self.write("**"+t.kwarg)
+            if t.kwargannotation:
+                self.write(": ")
+                self.dispatch(t.kwargannotation)
 
     def _keyword(self, t):
         self.write(t.arg)
@@ -462,10 +548,12 @@
         self.dispatch(t.value)
 
     def _Lambda(self, t):
+        self.write("(")
         self.write("lambda ")
         self.dispatch(t.args)
         self.write(": ")
         self.dispatch(t.body)
+        self.write(")")
 
     def _alias(self, t):
         self.write(t.name)
@@ -473,8 +561,11 @@
             self.write(" as "+t.asname)
 
 def roundtrip(filename, output=sys.stdout):
-    source = open(filename).read()
-    tree = compile(source, filename, "exec", _ast.PyCF_ONLY_AST)
+    with open(filename, "rb") as pyfile:
+        encoding = tokenize.detect_encoding(pyfile.readline)[0]
+    with open(filename, "r", encoding=encoding) as pyfile:
+        source = pyfile.read()
+    tree = compile(source, filename, "exec", ast.PyCF_ONLY_AST)
     Unparser(tree, output)
 
 

Modified: python/branches/py3k-cdecimal/Doc/Makefile
==============================================================================
--- python/branches/py3k-cdecimal/Doc/Makefile	(original)
+++ python/branches/py3k-cdecimal/Doc/Makefile	Sat Jul 17 12:36:09 2010
@@ -144,6 +144,7 @@
 	# archive the A4 latex
 	-rm -r build/latex
 	make latex PAPER=a4
+	-sed -i 's/makeindex/makeindex -q/' build/latex/Makefile
 	(cd build/latex; make clean && make all-pdf && make FMT=pdf zip bz2)
 	cp build/latex/docs-pdf.zip dist/python-$(DISTVERSION)-docs-pdf-a4.zip
 	cp build/latex/docs-pdf.tar.bz2 dist/python-$(DISTVERSION)-docs-pdf-a4.tar.bz2
@@ -151,6 +152,7 @@
 	# archive the letter latex
 	rm -r build/latex
 	make latex PAPER=letter
+	-sed -i 's/makeindex/makeindex -q/' build/latex/Makefile
 	(cd build/latex; make clean && make all-pdf && make FMT=pdf zip bz2)
 	cp build/latex/docs-pdf.zip dist/python-$(DISTVERSION)-docs-pdf-letter.zip
 	cp build/latex/docs-pdf.tar.bz2 dist/python-$(DISTVERSION)-docs-pdf-letter.tar.bz2

Modified: python/branches/py3k-cdecimal/Doc/README.txt
==============================================================================
--- python/branches/py3k-cdecimal/Doc/README.txt	(original)
+++ python/branches/py3k-cdecimal/Doc/README.txt	Sat Jul 17 12:36:09 2010
@@ -14,12 +14,11 @@
 Building the docs
 =================
 
-You need to install Python 2.4 or higher (but Python 3.0 is not supported yet);
-the toolset used to build the docs are written in Python.  The toolset used
-to build the documentation is called *Sphinx*, it is not included in this
-tree, but maintained separately in the Python Subversion repository.  Also
-needed are Jinja, a templating engine (included in Sphinx as a Subversion
-external), and optionally Pygments, a code highlighter.
+You need to have Python 2.4 or higher installed; the toolset used to build the
+docs is written in Python.  It is called *Sphinx*, it is not included in this
+tree, but maintained separately.  Also needed are the docutils, supplying the
+base markup that Sphinx uses, Jinja, a templating engine, and optionally
+Pygments, a code highlighter.
 
 
 Using make
@@ -47,29 +46,29 @@
    convert them into a single Compiled HTML (.chm) file -- these are popular
    under Microsoft Windows, but very handy on every platform.
 
-   To create the CHM file, you need to run the Microsoft HTML Help Workshop
-   over the generated project (.hhp) file.
+   To create the CHM file, you need to run the Microsoft HTML Help Workshop over
+   the generated project (.hhp) file.
 
- * "latex", which builds LaTeX source files that can be run with "pdflatex"
-   to produce PDF documents.
+ * "latex", which builds LaTeX source files as input to "pdflatex" to produce
+   PDF documents.
 
  * "text", which builds a plain text file for each source file.
 
  * "linkcheck", which checks all external references to see whether they are
-   broken, redirected or malformed, and outputs this information to stdout
-   as well as a plain-text (.txt) file.
+   broken, redirected or malformed, and outputs this information to stdout as
+   well as a plain-text (.txt) file.
 
  * "changes", which builds an overview over all versionadded/versionchanged/
    deprecated items in the current version. This is meant as a help for the
    writer of the "What's New" document.
 
- * "coverage", which builds a coverage overview for standard library modules
-   and C API.
+ * "coverage", which builds a coverage overview for standard library modules and
+   C API.
 
- * "pydoc-topics", which builds a Python module containing a dictionary
-   with plain text documentation for the labels defined in
-   `tools/sphinxext/pyspecific.py` -- pydoc needs these to show topic
-   and keyword help.
+ * "pydoc-topics", which builds a Python module containing a dictionary with
+   plain text documentation for the labels defined in
+   `tools/sphinxext/pyspecific.py` -- pydoc needs these to show topic and
+   keyword help.
 
 A "make update" updates the Subversion checkouts in `tools/`.
 
@@ -95,7 +94,7 @@
 
 or by installing it from PyPI.
 
-You can optionally also install Pygments, either as a checkout via :: 
+You can optionally also install Pygments, either as a checkout via ::
 
    svn co http://svn.python.org/projects/external/Pygments-1.3.1/pygments tools/pygments
 

Modified: python/branches/py3k-cdecimal/Doc/c-api/arg.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/c-api/arg.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/c-api/arg.rst	Sat Jul 17 12:36:09 2010
@@ -36,7 +36,7 @@
 
 However, when a :ctype:`Py_buffer` structure gets filled, the underlying
 buffer is locked so that the caller can subsequently use the buffer even
-inside a ``Py_BEGIN_ALLOW_THREADS`` block without the risk of mutable data
+inside a :ctype:`Py_BEGIN_ALLOW_THREADS` block without the risk of mutable data
 being resized or destroyed.  As a result, **you have to call**
 :cfunc:`PyBuffer_Release` after you have finished processing the data (or
 in any early abort case).
@@ -48,9 +48,9 @@
    the length argument (int or :ctype:`Py_ssize_t`) is controlled by
    defining the macro :cmacro:`PY_SSIZE_T_CLEAN` before including
    :file:`Python.h`.  If the macro was defined, length is a
-   :ctype:`Py_ssize_t` rather than an int.  This behavior will change
+   :ctype:`Py_ssize_t` rather than an :ctype:`int`. This behavior will change
    in a future Python version to only support :ctype:`Py_ssize_t` and
-   drop int support.  It is best to always define :cmacro:`PY_SSIZE_T_CLEAN`.
+   drop :ctype:`int` support. It is best to always define :cmacro:`PY_SSIZE_T_CLEAN`.
 
 
 ``s`` (:class:`str`) [const char \*]
@@ -100,7 +100,7 @@
    contain embedded NUL bytes; if it does, a :exc:`TypeError`
    exception is raised.
 
-``y*`` (:class:`bytes`, :class:`bytearray` or buffer compatible object) [Py_buffer \*]
+``y*`` (:class:`bytes`, :class:`bytearray` or buffer compatible object) [Py_buffer]
    This variant on ``s*`` doesn't accept Unicode objects, only objects
    supporting the buffer protocol.  **This is the recommended way to accept
    binary data.**
@@ -150,21 +150,11 @@
    any conversion.  Raises :exc:`TypeError` if the object is not a Unicode
    object.  The C variable may also be declared as :ctype:`PyObject\*`.
 
-``w`` (:class:`bytearray` or read-write character buffer) [char \*]
-   Similar to ``y``, but accepts any object which implements the read-write buffer
-   interface.  The caller must determine the length of the buffer by other means,
-   or use ``w#`` instead.  Only single-segment buffer objects are accepted;
-   :exc:`TypeError` is raised for all others.
-
 ``w*`` (:class:`bytearray` or read-write byte-oriented buffer) [Py_buffer]
-   This is to ``w`` what ``y*`` is to ``y``.
-
-``w#`` (:class:`bytearray` or read-write character buffer) [char \*, int]
-   Like ``y#``, but accepts any object which implements the read-write buffer
-   interface.  The :ctype:`char \*` variable is set to point to the first byte
-   of the buffer, and the :ctype:`int` is set to the length of the buffer.
-   Only single-segment buffer objects are accepted; :exc:`TypeError` is raised
-   for all others.
+   This format accepts any object which implements the read-write buffer
+   interface. It fills a :ctype:`Py_buffer` structure provided by the caller.
+   The buffer may contain embedded null bytes. The caller have to call
+   :cfunc:`PyBuffer_Release` when it is done with the buffer.
 
 ``es`` (:class:`str`) [const char \*encoding, char \*\*buffer]
    This variant on ``s`` is used for encoding Unicode into a character buffer.
@@ -492,11 +482,11 @@
    strings a tad more readable.
 
    ``s`` (:class:`str` or ``None``) [char \*]
-      Convert a null-terminated C string to a Python object using ``'utf-8'``
+      Convert a null-terminated C string to a Python :class:`str` object using ``'utf-8'``
       encoding. If the C string pointer is *NULL*, ``None`` is used.
 
    ``s#`` (:class:`str` or ``None``) [char \*, int]
-      Convert a C string and its length to a Python object using ``'utf-8'``
+      Convert a C string and its length to a Python :class:`str` object using ``'utf-8'``
       encoding. If the C string pointer is *NULL*, the length is ignored and
       ``None`` is returned.
 

Modified: python/branches/py3k-cdecimal/Doc/c-api/datetime.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/c-api/datetime.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/c-api/datetime.rst	Sat Jul 17 12:36:09 2010
@@ -8,9 +8,10 @@
 Various date and time objects are supplied by the :mod:`datetime` module.
 Before using any of these functions, the header file :file:`datetime.h` must be
 included in your source (note that this is not included by :file:`Python.h`),
-and the macro :cfunc:`PyDateTime_IMPORT` must be invoked.  The macro puts a
-pointer to a C structure into a static variable,  ``PyDateTimeAPI``, that is
-used by the following macros.
+and the macro :cmacro:`PyDateTime_IMPORT` must be invoked, usually as part of
+the module initialisation function.  The macro puts a pointer to a C structure
+into a static variable, :cdata:`PyDateTimeAPI`, that is used by the following
+macros.
 
 Type-check macros:
 

Modified: python/branches/py3k-cdecimal/Doc/c-api/file.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/c-api/file.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/c-api/file.rst	Sat Jul 17 12:36:09 2010
@@ -7,24 +7,29 @@
 
 .. index:: object: file
 
-Python's built-in file objects are implemented entirely on the :ctype:`FILE\*`
-support from the C standard library.  This is an implementation detail and may
-change in future releases of Python.  The ``PyFile_`` APIs are a wrapper over
-the :mod:`io` module.
-
-
-.. cfunction:: PyFile_FromFd(int fd, char *name, char *mode, int buffering, char *encoding, char *newline, int closefd)
-
-   Create a new :ctype:`PyFileObject` from the file descriptor of an already
-   opened file *fd*. The arguments *name*, *encoding* and *newline* can be
-   *NULL* to use the defaults; *buffering* can be *-1* to use the default.
-   Return *NULL* on failure.
+These APIs are a minimal emulation of the Python 2 C API for built-in file
+objects, which used to rely on the buffered I/O (:ctype:`FILE\*`) support
+from the C standard library.  In Python 3, files and streams use the new
+:mod:`io` module, which defines several layers over the low-level unbuffered
+I/O of the operating system.  The functions described below are
+convenience C wrappers over these new APIs, and meant mostly for internal
+error reporting in the interpreter; third-party code is advised to access
+the :mod:`io` APIs instead.
+
+
+.. cfunction:: PyFile_FromFd(int fd, char *name, char *mode, int buffering, char *encoding, char *errors, char *newline, int closefd)
+
+   Create a Python file object from the file descriptor of an already
+   opened file *fd*.  The arguments *name*, *encoding*, *errors* and *newline*
+   can be *NULL* to use the defaults; *buffering* can be *-1* to use the
+   default.  Return *NULL* on failure.  For a more comprehensive description of
+   the arguments, please refer to the :func:`io.open` function documentation.
 
    .. warning::
 
-     Take care when you are mixing streams and descriptors! For more
-     information, see `the GNU C Library docs
-     <http://www.gnu.org/software/libc/manual/html_node/Stream_002fDescriptor-Precautions.html#Stream_002fDescriptor-Precautions>`_.
+     Since Python streams have their own buffering layer, mixing them with
+     OS-level file descriptors can produce various issues (such as unexpected
+     ordering of data).
 
 
 .. cfunction:: int PyObject_AsFileDescriptor(PyObject *p)

Modified: python/branches/py3k-cdecimal/Doc/c-api/init.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/c-api/init.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/c-api/init.rst	Sat Jul 17 12:36:09 2010
@@ -385,7 +385,7 @@
 
 .. cfunction:: void PySys_SetArgv(int argc, wchar_t **argv)
 
-   This function works like :cfunc:`PySys_SetArgv` with *updatepath* set to 1.
+   This function works like :cfunc:`PySys_SetArgvEx` with *updatepath* set to 1.
 
 
 .. cfunction:: void Py_SetPythonHome(wchar_t *home)

Modified: python/branches/py3k-cdecimal/Doc/c-api/intro.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/c-api/intro.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/c-api/intro.rst	Sat Jul 17 12:36:09 2010
@@ -519,12 +519,12 @@
 :mod:`builtins`, :mod:`__main__`, :mod:`sys`, and :mod:`exceptions`.  It also
 initializes the module search path (``sys.path``).
 
-.. index:: single: PySys_SetArgv()
+.. index:: single: PySys_SetArgvEx()
 
 :cfunc:`Py_Initialize` does not set the "script argument list"  (``sys.argv``).
-If this variable is needed by Python code that  will be executed later, it must
-be set explicitly with a call to  ``PySys_SetArgv(argc, argv)`` subsequent to
-the call to :cfunc:`Py_Initialize`.
+If this variable is needed by Python code that will be executed later, it must
+be set explicitly with a call to  ``PySys_SetArgvEx(argc, argv, updatepath)``
+after the call to :cfunc:`Py_Initialize`.
 
 On most systems (in particular, on Unix and Windows, although the details are
 slightly different), :cfunc:`Py_Initialize` calculates the module search path

Modified: python/branches/py3k-cdecimal/Doc/c-api/number.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/c-api/number.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/c-api/number.rst	Sat Jul 17 12:36:09 2010
@@ -215,18 +215,6 @@
    the Python statement ``o1 |= o2``.
 
 
-.. cfunction:: PyObject* PyNumber_Int(PyObject *o)
-
-   Returns the *o* converted to an integer object on success, or *NULL* on
-   failure.  This is the equivalent of the Python expression ``int(o)``.
-
-   .. note::
-
-     This function is defined in the transitional :file:`intobject.h`
-     header file.  It will be removed completely in Python 3.1.  Use
-     the :cfunc:`PyNumber_Long` function instead.
-
-
 .. cfunction:: PyObject* PyNumber_Long(PyObject *o)
 
    .. index:: builtin: int

Modified: python/branches/py3k-cdecimal/Doc/c-api/unicode.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/c-api/unicode.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/c-api/unicode.rst	Sat Jul 17 12:36:09 2010
@@ -346,10 +346,10 @@
    Coerce an encoded object *obj* to an Unicode object and return a reference with
    incremented refcount.
 
-   String and other char buffer compatible objects are decoded according to the
-   given encoding and using the error handling defined by errors.  Both can be
-   *NULL* to have the interface use the default values (see the next section for
-   details).
+   :class:`bytes`, :class:`bytearray` and other char buffer compatible objects
+   are decoded according to the given encoding and using the error handling
+   defined by errors. Both can be *NULL* to have the interface use the default
+   values (see the next section for details).
 
    All other objects, including Unicode objects, cause a :exc:`TypeError` to be
    set.

Modified: python/branches/py3k-cdecimal/Doc/data/refcounts.dat
==============================================================================
--- python/branches/py3k-cdecimal/Doc/data/refcounts.dat	(original)
+++ python/branches/py3k-cdecimal/Doc/data/refcounts.dat	Sat Jul 17 12:36:09 2010
@@ -846,9 +846,6 @@
 PyNumber_InPlaceXor:PyObject*:v:0:
 PyNumber_InPlaceXor:PyObject*:w:0:
 
-PyNumber_Int:PyObject*::+1:
-PyNumber_Int:PyObject*:o:0:
-
 PyNumber_Invert:PyObject*::+1:
 PyNumber_Invert:PyObject*:o:0:
 

Modified: python/branches/py3k-cdecimal/Doc/distutils/apiref.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/distutils/apiref.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/distutils/apiref.rst	Sat Jul 17 12:36:09 2010
@@ -21,7 +21,9 @@
 .. function:: setup(arguments)
 
    The basic do-everything function that does most everything you could ever ask
-   for from a Distutils method. See XXXXX
+   for from a Distutils method.
+
+   .. See XXXXX
 
    The setup function takes a large number of arguments. These are laid out in the
    following table.
@@ -147,11 +149,11 @@
 In addition, the :mod:`distutils.core` module exposed a number of  classes that
 live elsewhere.
 
-* :class:`Extension` from :mod:`distutils.extension`
+* :class:`~distutils.extension.Extension` from :mod:`distutils.extension`
 
-* :class:`Command` from :mod:`distutils.cmd`
+* :class:`~distutils.cmd.Command` from :mod:`distutils.cmd`
 
-* :class:`Distribution` from :mod:`distutils.dist`
+* :class:`~distutils.dist.Distribution` from :mod:`distutils.dist`
 
 A short description of each of these follows, but see the relevant module for
 the full reference.
@@ -995,7 +997,7 @@
    errors are ignored (apart from being reported to ``sys.stdout`` if *verbose* is
    true).
 
-**\*\*** Some of this could be replaced with the shutil module? **\*\***
+.. XXX Some of this could be replaced with the shutil module?
 
 
 :mod:`distutils.file_util` --- Single file operations
@@ -1311,8 +1313,7 @@
   the "negative alias" of :option:`--verbose`, then :option:`--quiet` on the
   command line sets *verbose* to false.
 
-**\*\*** Should be replaced with :mod:`optik` (which is also now known as
-:mod:`optparse` in Python 2.3 and later). **\*\***
+.. XXX Should be replaced with :mod:`optparse`.
 
 
 .. function:: fancy_getopt(options, negative_opt, object, args)
@@ -1680,8 +1681,8 @@
 ===================================================================
 
 .. module:: distutils.cmd
-   :synopsis: This module provides the abstract base class Command. This class is subclassed
-              by the modules in the distutils.command  subpackage.
+   :synopsis: This module provides the abstract base class Command. This class
+              is subclassed by the modules in the distutils.command subpackage.
 
 
 This module supplies the abstract base class :class:`Command`.
@@ -1691,20 +1692,84 @@
 
    Abstract base class for defining command classes, the "worker bees" of the
    Distutils.  A useful analogy for command classes is to think of them as
-   subroutines with local variables called *options*.  The options are declared in
-   :meth:`initialize_options` and defined (given their final values) in
-   :meth:`finalize_options`, both of which must be defined by every command class.
-   The distinction between the two is necessary because option values might come
-   from the outside world (command line, config file, ...), and any options
-   dependent on other options must be computed after these outside influences have
-   been processed --- hence :meth:`finalize_options`.  The body of the subroutine,
-   where it does all its work based on the values of its options, is the
-   :meth:`run` method, which must also be implemented by every command class.
+   subroutines with local variables called *options*.  The options are declared
+   in :meth:`initialize_options` and defined (given their final values) in
+   :meth:`finalize_options`, both of which must be defined by every command
+   class.  The distinction between the two is necessary because option values
+   might come from the outside world (command line, config file, ...), and any
+   options dependent on other options must be computed after these outside
+   influences have been processed --- hence :meth:`finalize_options`.  The body
+   of the subroutine, where it does all its work based on the values of its
+   options, is the :meth:`run` method, which must also be implemented by every
+   command class.
 
-   The class constructor takes a single argument *dist*, a  :class:`Distribution`
+   The class constructor takes a single argument *dist*, a :class:`Distribution`
    instance.
 
 
+Creating a new Distutils command
+================================
+
+This section outlines the steps to create a new Distutils command.
+
+A new command lives in a module in the :mod:`distutils.command` package. There
+is a sample template in that directory called :file:`command_template`.  Copy
+this file to a new module with the same name as the new command you're
+implementing.  This module should implement a class with the same name as the
+module (and the command).  So, for instance, to create the command
+``peel_banana`` (so that users can run ``setup.py peel_banana``), you'd copy
+:file:`command_template` to :file:`distutils/command/peel_banana.py`, then edit
+it so that it's implementing the class :class:`peel_banana`, a subclass of
+:class:`distutils.cmd.Command`.
+
+Subclasses of :class:`Command` must define the following methods.
+
+.. method:: Command.initialize_options()
+
+   Set default values for all the options that this command supports.  Note that
+   these defaults may be overridden by other commands, by the setup script, by
+   config files, or by the command-line.  Thus, this is not the place to code
+   dependencies between options; generally, :meth:`initialize_options`
+   implementations are just a bunch of ``self.foo = None`` assignments.
+
+
+.. method:: Command.finalize_options()
+
+   Set final values for all the options that this command supports. This is
+   always called as late as possible, ie.  after any option assignments from the
+   command-line or from other commands have been done.  Thus, this is the place
+   to to code option dependencies: if *foo* depends on *bar*, then it is safe to
+   set *foo* from *bar* as long as *foo* still has the same value it was
+   assigned in :meth:`initialize_options`.
+
+
+.. method:: Command.run()
+
+   A command's raison d'etre: carry out the action it exists to perform,
+   controlled by the options initialized in :meth:`initialize_options`,
+   customized by other commands, the setup script, the command-line, and config
+   files, and finalized in :meth:`finalize_options`.  All terminal output and
+   filesystem interaction should be done by :meth:`run`.
+
+
+.. attribute:: Command.sub_commands
+
+   *sub_commands* formalizes the notion of a "family" of commands,
+   e.g. ``install`` as the parent with sub-commands ``install_lib``,
+   ``install_headers``, etc.  The parent of a family of commands defines
+   *sub_commands* as a class attribute; it's a list of 2-tuples ``(command_name,
+   predicate)``, with *command_name* a string and *predicate* a function, a
+   string or ``None``.  *predicate* is a method of the parent command that
+   determines whether the corresponding command is applicable in the current
+   situation.  (E.g. we ``install_headers`` is only applicable if we have any C
+   header files to install.)  If *predicate* is ``None``, that command is always
+   applicable.
+
+   *sub_commands* is usually defined at the *end* of a class, because
+   predicates can be methods of the class, so they must already have been
+   defined.  The canonical example is the :command:`install` command.
+
+
 :mod:`distutils.command` --- Individual Distutils commands
 ==========================================================
 
@@ -1943,63 +2008,3 @@
 This is described in more detail in :pep:`301`.
 
 .. % todo
-
-
-Creating a new Distutils command
-================================
-
-This section outlines the steps to create a new Distutils command.
-
-A new command lives in a module in the :mod:`distutils.command` package. There
-is a sample template in that directory called  :file:`command_template`. Copy
-this file to a new module with the same name as the new command you're
-implementing. This module should implement a class with the same name as the
-module (and the command). So, for instance, to create the command
-``peel_banana`` (so that users can run ``setup.py peel_banana``), you'd copy
-:file:`command_template`  to :file:`distutils/command/peel_banana.py`, then edit
-it so that it's implementing the class :class:`peel_banana`, a subclass of
-:class:`distutils.cmd.Command`.
-
-Subclasses of :class:`Command` must define the following methods.
-
-
-.. method:: Command.initialize_options()
-
-   Set default values for all the options that this command supports.  Note that
-   these defaults may be overridden by other commands, by the setup script, by
-   config files, or by the command-line.  Thus, this is not the place to code
-   dependencies between options; generally, :meth:`initialize_options`
-   implementations are just a bunch of ``self.foo = None`` assignments.
-
-
-.. method:: Command.finalize_options()
-
-   Set final values for all the options that this command supports. This is
-   always called as late as possible, ie.  after any option assignments from the
-   command-line or from other commands have been done.  Thus, this is the place
-   to to code option dependencies: if *foo* depends on *bar*, then it is safe to
-   set *foo* from *bar* as long as *foo* still has the same value it was
-   assigned in :meth:`initialize_options`.
-
-
-.. method:: Command.run()
-
-   A command's raison d'etre: carry out the action it exists to perform, controlled
-   by the options initialized in :meth:`initialize_options`, customized by other
-   commands, the setup script, the command-line, and config files, and finalized in
-   :meth:`finalize_options`.  All terminal output and filesystem interaction should
-   be done by :meth:`run`.
-
-*sub_commands* formalizes the notion of a "family" of commands, eg. ``install``
-as the parent with sub-commands ``install_lib``, ``install_headers``, etc.  The
-parent of a family of commands defines *sub_commands* as a class attribute; it's
-a list of 2-tuples ``(command_name, predicate)``, with *command_name* a string
-and *predicate* a function, a string or None. *predicate* is a method of
-the parent command that determines whether the corresponding command is
-applicable in the current situation.  (Eg. we ``install_headers`` is only
-applicable if we have any C header files to install.)  If *predicate* is None,
-that command is always applicable.
-
-*sub_commands* is usually defined at the \*end\* of a class, because predicates
-can be methods of the class, so they must already have been defined.  The
-canonical example is the :command:`install` command.

Modified: python/branches/py3k-cdecimal/Doc/distutils/builtdist.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/distutils/builtdist.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/distutils/builtdist.rst	Sat Jul 17 12:36:09 2010
@@ -146,8 +146,8 @@
 Creating dumb built distributions
 =================================
 
-**\*\*** Need to document absolute vs. prefix-relative packages here, but first
-I have to implement it! **\*\***
+.. XXX Need to document absolute vs. prefix-relative packages here, but first
+       I have to implement it!
 
 
 .. _creating-rpms:
@@ -322,7 +322,7 @@
 option.
 
 By default the installer will display the cool "Python Powered" logo when it is
-run, but you can also supply your own 152x161 bitmap which must be a Windows
+run, but you can also supply your own 152x261 bitmap which must be a Windows
 :file:`.bmp` file with the :option:`--bitmap` option.
 
 The installer will also display a large title on the desktop background window
@@ -375,7 +375,7 @@
 The Postinstallation script
 ---------------------------
 
-Starting with Python 2.3, a postinstallation script can be specified which the
+Starting with Python 2.3, a postinstallation script can be specified with the
 :option:`--install-script` option.  The basename of the script must be
 specified, and the script filename must also be listed in the scripts argument
 to the setup function.

Modified: python/branches/py3k-cdecimal/Doc/distutils/extending.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/distutils/extending.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/distutils/extending.rst	Sat Jul 17 12:36:09 2010
@@ -15,8 +15,8 @@
 should be copied into packages in addition to :file:`.py` files as a
 convenience.
 
-Most distutils command implementations are subclasses of the :class:`Command`
-class from :mod:`distutils.cmd`.  New commands may directly inherit from
+Most distutils command implementations are subclasses of the
+:class:`distutils.cmd.Command` class.  New commands may directly inherit from
 :class:`Command`, while replacements often derive from :class:`Command`
 indirectly, directly subclassing the command they are replacing.  Commands are
 required to derive from :class:`Command`.

Modified: python/branches/py3k-cdecimal/Doc/distutils/setupscript.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/distutils/setupscript.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/distutils/setupscript.rst	Sat Jul 17 12:36:09 2010
@@ -207,7 +207,7 @@
 SWIG on the interface file and compile the resulting C/C++ file into your
 extension.
 
-**\*\*** SWIG support is rough around the edges and largely untested! **\*\***
+.. XXX SWIG support is rough around the edges and largely untested!
 
 This warning notwithstanding, options to SWIG can be currently passed like
 this::
@@ -326,7 +326,7 @@
 (Again, this sort of non-portable construct should be avoided if you intend to
 distribute your code.)
 
-**\*\*** Should mention clib libraries here or somewhere else! **\*\***
+.. XXX Should mention clib libraries here or somewhere else!
 
 
 Other options

Modified: python/branches/py3k-cdecimal/Doc/distutils/sourcedist.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/distutils/sourcedist.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/distutils/sourcedist.rst	Sat Jul 17 12:36:09 2010
@@ -76,10 +76,10 @@
   :option:`packages` options
 
 * all C source files mentioned in the :option:`ext_modules` or
-  :option:`libraries` options (
+  :option:`libraries` options
 
-  **\*\*** getting C library sources currently broken---no
-  :meth:`get_source_files` method in :file:`build_clib.py`! **\*\***)
+  .. XXX Getting C library sources is currently broken -- no
+     :meth:`get_source_files` method in :file:`build_clib.py`!
 
 * scripts identified by the :option:`scripts` option
   See :ref:`distutils-installing-scripts`.

Modified: python/branches/py3k-cdecimal/Doc/documenting/index.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/documenting/index.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/documenting/index.rst	Sat Jul 17 12:36:09 2010
@@ -10,9 +10,9 @@
 `reStructuredText`_, developed by the `docutils`_ project, amended by custom
 directives and using a toolset named `Sphinx`_ to postprocess the HTML output.
 
-This document describes the style guide for our documentation, the custom
-reStructuredText markup introduced to support Python documentation and how it
-should be used, as well as the Sphinx build system.
+This document describes the style guide for our documentation as well as the
+custom reStructuredText markup introduced by Sphinx to support Python
+documentation and how it should be used.
 
 .. _reStructuredText: http://docutils.sf.net/rst.html
 .. _docutils: http://docutils.sf.net/
@@ -35,3 +35,4 @@
    rest.rst
    markup.rst
    fromlatex.rst
+   building.rst

Modified: python/branches/py3k-cdecimal/Doc/documenting/markup.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/documenting/markup.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/documenting/markup.rst	Sat Jul 17 12:36:09 2010
@@ -698,10 +698,10 @@
       .. toctree::
          :maxdepth: 2
 
-         intro.rst
-         strings.rst
-         datatypes.rst
-         numeric.rst
+         intro
+         strings
+         datatypes
+         numeric
          (many more files listed here)
 
    This accomplishes two things:
@@ -709,8 +709,8 @@
    * Tables of contents from all those files are inserted, with a maximum depth
      of two, that means one nested heading.  ``toctree`` directives in those
      files are also taken into account.
-   * Sphinx knows that the relative order of the files ``intro.rst``,
-     ``strings.rst`` and so forth, and it knows that they are children of the
+   * Sphinx knows that the relative order of the files ``intro``,
+     ``strings`` and so forth, and it knows that they are children of the
      shown file, the library index.  From this information it generates "next
      chapter", "previous chapter" and "parent chapter" links.
 

Modified: python/branches/py3k-cdecimal/Doc/documenting/style.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/documenting/style.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/documenting/style.rst	Sat Jul 17 12:36:09 2010
@@ -7,7 +7,7 @@
 wherever possible. This particular style guide was selected mostly because it
 seems reasonable and is easy to get online.
 
-Topics which are not covered in the Apple's style guide will be discussed in
+Topics which are not covered in Apple's style guide will be discussed in
 this document.
 
 All reST files use an indentation of 3 spaces.  The maximum line length is 80

Modified: python/branches/py3k-cdecimal/Doc/faq/extending.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/faq/extending.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/faq/extending.rst	Sat Jul 17 12:36:09 2010
@@ -29,6 +29,8 @@
 C++ objects with constructors are probably not a good idea.
 
 
+.. _c-wrapper-software:
+
 Writing C is hard; are there any alternatives?
 ----------------------------------------------
 
@@ -200,11 +202,7 @@
 whole lot of difference between C and C++ -- so the strategy of building a new
 Python type around a C structure (pointer) type will also work for C++ objects.
 
-For C++ libraries, you can look at `SIP
-<http://www.riverbankcomputing.co.uk/software/sip/>`_, `CXX
-<http://cxx.sourceforge.net/>`_, `Boost
-<http://www.boost.org/libs/python/doc/index.html>`_, `Weave
-<http://www.scipy.org/Weave>`_ or `SWIG <http://www.swig.org>`_
+For C++ libraries, see :ref:`c-wrapper-software`.
 
 
 I added a module using the Setup file and the make fails; why?

Modified: python/branches/py3k-cdecimal/Doc/faq/gui.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/faq/gui.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/faq/gui.rst	Sat Jul 17 12:36:09 2010
@@ -13,14 +13,14 @@
 =====================
 
 What platform-independent GUI toolkits exist for Python?
---------------------------------------------------------
+========================================================
 
 Depending on what platform(s) you are aiming at, there are several.
 
 .. XXX check links
 
 Tkinter
-'''''''
+-------
 
 Standard builds of Python include an object-oriented interface to the Tcl/Tk
 widget set, called Tkinter.  This is probably the easiest to install and use.
@@ -29,22 +29,27 @@
 Unix platforms.
 
 wxWidgets
-'''''''''
+---------
 
-wxWidgets is a GUI class library written in C++ that's a portable
-interface to various platform-specific libraries, and that has a
-Python interface called `wxPython <http://www.wxpython.org>`__.
-
-wxWidgets preserves the look and feel of the
-underlying graphics toolkit, and has a large set of widgets and
-collection of GDI classes.  See `the wxWidgets page
-<http://www.wxwidgets.org>`_ for more details.
+wxWidgets (http://www.wxwidgets.org) is a free, portable GUI class
+library written in C++ that provides a native look and feel on a
+number of platforms, with Windows, MacOS X, GTK, X11, all listed as
+current stable targets.  Language bindings are available for a number
+of languages including Python, Perl, Ruby, etc.
+
+wxPython (http://www.wxpython.org) is the Python binding for
+wxwidgets.  While it often lags slightly behind the official wxWidgets
+releases, it also offers a number of features via pure Python
+extensions that are not available in other language bindings.  There
+is an active wxPython user and developer community.
+
+Both wxWidgets and wxPython are free, open source, software with
+permissive licences that allow their use in commercial products as
+well as in freeware or shareware.
 
-wxWidgets supports Windows and MacOS; on Unix variants,
-it supports both GTk+ and Motif toolkits.
 
 Qt
-'''
+---
 
 There are bindings available for the Qt toolkit (`PyQt
 <http://www.riverbankcomputing.co.uk/software/pyqt/>`_) and for KDE (`PyKDE <http://www.riverbankcomputing.co.uk/software/pykde/intro>`__).  If
@@ -55,13 +60,13 @@
 <http://www.trolltech.com>`_.
 
 Gtk+
-''''
+----
 
 PyGtk bindings for the `Gtk+ toolkit <http://www.gtk.org>`_ have been
 implemented by James Henstridge; see <http://www.pygtk.org>.
 
 FLTK
-''''
+----
 
 Python bindings for `the FLTK toolkit <http://www.fltk.org>`_, a simple yet
 powerful and mature cross-platform windowing system, are available from `the
@@ -69,7 +74,7 @@
 
 
 FOX
-'''
+----
 
 A wrapper for `the FOX toolkit <http://www.fox-toolkit.org/>`_ called `FXpy
 <http://fxpy.sourceforge.net/>`_ is available.  FOX supports both Unix variants
@@ -77,13 +82,13 @@
 
 
 OpenGL
-''''''
+------
 
 For OpenGL bindings, see `PyOpenGL <http://pyopengl.sourceforge.net>`_.
 
 
 What platform-specific GUI toolkits exist for Python?
------------------------------------------------------
+========================================================
 
 `The Mac port <http://python.org/download/mac>`_ by Jack Jansen has a rich and
 ever-growing set of modules that support the native Mac toolbox calls.  The port

Modified: python/branches/py3k-cdecimal/Doc/glossary.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/glossary.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/glossary.rst	Sat Jul 17 12:36:09 2010
@@ -57,11 +57,17 @@
 
    bytecode
       Python source code is compiled into bytecode, the internal representation
-      of a Python program in the interpreter.  The bytecode is also cached in
-      ``.pyc`` and ``.pyo`` files so that executing the same file is faster the
-      second time (recompilation from source to bytecode can be avoided).  This
-      "intermediate language" is said to run on a :term:`virtual machine`
-      that executes the machine code corresponding to each bytecode.
+      of a Python program in the CPython interpreter.  The bytecode is also
+      cached in ``.pyc`` and ``.pyo`` files so that executing the same file is
+      faster the second time (recompilation from source to bytecode can be
+      avoided).  This "intermediate language" is said to run on a
+      :term:`virtual machine` that executes the machine code corresponding to
+      each bytecode. Do note that bytecodes are not expected to work between
+      different Python virtual machines, nor to be stable between Python
+      releases.
+
+      A list of bytecode instructions can be found in the documentation for
+      :ref:`the dis module <bytecodes>`.
 
    class
       A template for creating user-defined objects. Class definitions
@@ -148,15 +154,15 @@
       object.
 
    duck-typing
-      A pythonic programming style which determines an object's type by inspection
-      of its method or attribute signature rather than by explicit relationship
-      to some type object ("If it looks like a duck and quacks like a duck, it
+      A programming style which does not look at an object's type to determine
+      if it has the right interface; instead, the method or attribute is simply
+      called or used ("If it looks like a duck and quacks like a duck, it
       must be a duck.")  By emphasizing interfaces rather than specific types,
       well-designed code improves its flexibility by allowing polymorphic
       substitution.  Duck-typing avoids tests using :func:`type` or
-      :func:`isinstance`. (Note, however, that duck-typing can be complemented
-      with abstract base classes.) Instead, it typically employs :func:`hasattr`
-      tests or :term:`EAFP` programming.
+      :func:`isinstance`.  (Note, however, that duck-typing can be complemented
+      with :term:`abstract base class`\ es.)  Instead, it typically employs
+      :func:`hasattr` tests or :term:`EAFP` programming.
 
    EAFP
       Easier to ask for forgiveness than permission.  This common Python coding
@@ -424,10 +430,11 @@
    nested scope
       The ability to refer to a variable in an enclosing definition.  For
       instance, a function defined inside another function can refer to
-      variables in the outer function.  Note that nested scopes work only for
-      reference and not for assignment which will always write to the innermost
-      scope.  In contrast, local variables both read and write in the innermost
-      scope.  Likewise, global variables read and write to the global namespace.
+      variables in the outer function.  Note that nested scopes by default work
+      only for reference and not for assignment.  Local variables both read and
+      write in the innermost scope.  Likewise, global variables read and write
+      to the global namespace.  The :keyword:`nonlocal` allows writing to outer
+      scopes.
 
    new-style class
       Old name for the flavor of classes now used for all class objects.  In

Modified: python/branches/py3k-cdecimal/Doc/howto/functional.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/howto/functional.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/howto/functional.rst	Sat Jul 17 12:36:09 2010
@@ -5,9 +5,6 @@
 :Author: A. M. Kuchling
 :Release: 0.31
 
-(This is a first draft.  Please send comments/error reports/suggestions to
-amk at amk.ca.)
-
 In this document, we'll take a tour of Python's features suitable for
 implementing programs in a functional style.  After an introduction to the
 concepts of functional programming, we'll look at language features such as

Modified: python/branches/py3k-cdecimal/Doc/howto/unicode.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/howto/unicode.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/howto/unicode.rst	Sat Jul 17 12:36:09 2010
@@ -4,10 +4,12 @@
   Unicode HOWTO
 *****************
 
-:Release: 1.1
+:Release: 1.11
 
-This HOWTO discusses Python's support for Unicode, and explains various problems
-that people commonly encounter when trying to work with Unicode.
+This HOWTO discusses Python 2.x's support for Unicode, and explains
+various problems that people commonly encounter when trying to work
+with Unicode.  (This HOWTO has not yet been updated to cover the 3.x
+versions of Python.)
 
 
 Introduction to Unicode
@@ -146,8 +148,9 @@
 4. Many Internet standards are defined in terms of textual data, and can't
    handle content with embedded zero bytes.
 
-Generally people don't use this encoding, instead choosing other encodings that
-are more efficient and convenient.
+Generally people don't use this encoding, instead choosing other
+encodings that are more efficient and convenient.  UTF-8 is probably
+the most commonly supported encoding; it will be discussed below.
 
 Encodings don't have to handle every possible Unicode character, and most
 encodings don't.  The rules for converting a Unicode string into the ASCII
@@ -223,8 +226,8 @@
 <http://en.wikipedia.org/wiki/UTF-8>, for example.
 
 
-Python's Unicode Support
-========================
+Python 2.x's Unicode Support
+============================
 
 Now that you've learned the rudiments of Unicode, we can look at Python's
 Unicode features.
@@ -266,8 +269,8 @@
     >>> b'\x80abc'.decode("utf-8", "ignore")
     'abc'
 
-Encodings are specified as strings containing the encoding's name.  Python comes
-with roughly 100 different encodings; see the Python Library Reference at
+Encodings are specified as strings containing the encoding's name.  Python 3.2
+comes with roughly 100 different encodings; see the Python Library Reference at
 :ref:`standard-encodings` for a list.  Some encodings have multiple names; for
 example, 'latin-1', 'iso_8859_1' and '8859' are all synonyms for the same
 encoding.
@@ -626,7 +629,10 @@
 
 Version 1.1: Feb-Nov 2008.  Updates the document with respect to Python 3 changes.
 
+Version 1.11: posted June 20 2010.  Notes that Python 3.x is not covered,
+and that the HOWTO only covers 2.x.
 
+.. comment Describe Python 3.x support (new section? new document?)
 .. comment Additional topic: building Python w/ UCS2 or UCS4 support
 .. comment Describe use of codecs.StreamRecoder and StreamReaderWriter
 

Modified: python/branches/py3k-cdecimal/Doc/library/2to3.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/2to3.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/2to3.rst	Sat Jul 17 12:36:09 2010
@@ -276,7 +276,7 @@
 
    Converts the :keyword:`print` statement to the :func:`print` function.
 
-.. 2to3fixer:: raises
+.. 2to3fixer:: raise
 
    Converts ``raise E, V`` to ``raise E(V)``, and ``raise E, V, T`` to ``raise
    E(V).with_traceback(T)``.  If ``E`` is a tuple, the translation will be

Modified: python/branches/py3k-cdecimal/Doc/library/ast.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/ast.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/ast.rst	Sat Jul 17 12:36:09 2010
@@ -119,12 +119,15 @@
 
    Safely evaluate an expression node or a string containing a Python
    expression.  The string or node provided may only consist of the following
-   Python literal structures: strings, numbers, tuples, lists, dicts, booleans,
-   and ``None``.
+   Python literal structures: strings, bytes, numbers, tuples, lists, dicts,
+   sets, booleans, and ``None``.
 
    This can be used for safely evaluating strings containing Python expressions
    from untrusted sources without the need to parse the values oneself.
 
+   .. versionchanged:: 3.2
+      Now allows bytes and set literals.
+
 
 .. function:: get_docstring(node, clean=True)
 

Modified: python/branches/py3k-cdecimal/Doc/library/calendar.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/calendar.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/calendar.rst	Sat Jul 17 12:36:09 2010
@@ -15,7 +15,7 @@
 are given as integers. For related
 functionality, see also the :mod:`datetime` and :mod:`time` modules.
 
-Most of these functions and classses rely on the :mod:`datetime` module which
+Most of these functions and classes rely on the :mod:`datetime` module which
 uses an idealized calendar, the current Gregorian calendar indefinitely extended
 in both directions.  This matches the definition of the "proleptic Gregorian"
 calendar in Dershowitz and Reingold's book "Calendrical Calculations", where

Modified: python/branches/py3k-cdecimal/Doc/library/cmath.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/cmath.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/cmath.rst	Sat Jul 17 12:36:09 2010
@@ -187,15 +187,24 @@
 Classification functions
 ------------------------
 
+.. function:: isfinite(x)
+
+   Return ``True`` if both the real and imaginary parts of *x* are finite, and
+   ``False`` otherwise.
+
+   .. versionadded:: 3.2
+
+
 .. function:: isinf(x)
 
-   Return *True* if the real or the imaginary part of x is positive
-   or negative infinity.
+   Return ``True`` if either the real or the imaginary part of *x* is an
+   infinity, and ``False`` otherwise.
 
 
 .. function:: isnan(x)
 
-   Return *True* if the real or imaginary part of x is not a number (NaN).
+   Return ``True`` if either the real or the imaginary part of *x* is a NaN,
+   and ``False`` otherwise.
 
 
 Constants

Modified: python/branches/py3k-cdecimal/Doc/library/codecs.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/codecs.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/codecs.rst	Sat Jul 17 12:36:09 2010
@@ -953,6 +953,8 @@
 +-----------------+--------------------------------+--------------------------------+
 | cp857           | 857, IBM857                    | Turkish                        |
 +-----------------+--------------------------------+--------------------------------+
+| cp858           | 858, IBM858                    | Western Europe                 |
++-----------------+--------------------------------+--------------------------------+
 | cp860           | 860, IBM860                    | Portuguese                     |
 +-----------------+--------------------------------+--------------------------------+
 | cp861           | 861, CP-IS, IBM861             | Icelandic                      |
@@ -1223,6 +1225,23 @@
    Convert a label to Unicode, as specified in :rfc:`3490`.
 
 
+:mod:`encodings.mbcs` --- Windows ANSI codepage
+-----------------------------------------------
+
+.. module:: encodings.mbcs
+   :synopsis: Windows ANSI codepage
+
+Encode operand according to the ANSI codepage (CP_ACP). This codec only
+supports ``'strict'`` and ``'replace'`` error handlers to encode, and
+``'strict'`` and ``'ignore'`` error handlers to decode.
+
+Availability: Windows only.
+
+.. versionchanged:: 3.2
+   Before 3.2, the *errors* argument was ignored; ``'replace'`` was always used
+   to encode, and ``'ignore'`` to decode.
+
+
 :mod:`encodings.utf_8_sig` --- UTF-8 codec with BOM signature
 -------------------------------------------------------------
 

Modified: python/branches/py3k-cdecimal/Doc/library/contextlib.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/contextlib.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/contextlib.rst	Sat Jul 17 12:36:09 2010
@@ -51,6 +51,11 @@
    the exception has been handled, and execution will resume with the statement
    immediately following the :keyword:`with` statement.
 
+   contextmanager uses :class:`ContextDecorator` so the context managers it
+   creates can be used as decorators as well as in :keyword:`with` statements.
+
+   .. versionchanged:: 3.2
+      Use of :class:`ContextDecorator`.
 
 .. function:: closing(thing)
 
@@ -79,6 +84,58 @@
    ``page.close()`` will be called when the :keyword:`with` block is exited.
 
 
+.. class:: ContextDecorator()
+
+   A base class that enables a context manager to also be used as a decorator.
+
+   Context managers inheriting from ``ContextDecorator`` have to implement
+   ``__enter__`` and ``__exit__`` as normal. ``__exit__`` retains its optional
+   exception handling even when used as a decorator.
+
+   Example::
+
+      from contextlib import ContextDecorator
+
+      class mycontext(ContextDecorator):
+         def __enter__(self):
+            print('Starting')
+            return self
+
+         def __exit__(self, *exc):
+            print('Finishing')
+            return False
+
+      >>> @mycontext()
+      ... def function():
+      ...    print('The bit in the middle')
+      ...
+      >>> function()
+      Starting
+      The bit in the middle
+      Finishing
+
+      >>> with mycontext():
+      ...    print('The bit in the middle')
+      ...
+      Starting
+      The bit in the middle
+      Finishing
+
+   Existing context managers that already have a base class can be extended by
+   using ``ContextDecorator`` as a mixin class::
+
+      from contextlib import ContextDecorator
+
+      class mycontext(ContextBaseClass, ContextDecorator):
+         def __enter__(self):
+            return self
+
+         def __exit__(self, *exc):
+            return False
+
+   .. versionadded:: 3.2
+
+
 .. seealso::
 
    :pep:`0343` - The "with" statement

Modified: python/branches/py3k-cdecimal/Doc/library/datetime.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/datetime.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/datetime.rst	Sat Jul 17 12:36:09 2010
@@ -28,11 +28,14 @@
 have an optional time zone information member, :attr:`tzinfo`, that can contain
 an instance of a subclass of the abstract :class:`tzinfo` class.  These
 :class:`tzinfo` objects capture information about the offset from UTC time, the
-time zone name, and whether Daylight Saving Time is in effect.  Note that no
-concrete :class:`tzinfo` classes are supplied by the :mod:`datetime` module.
-Supporting timezones at whatever level of detail is required is up to the
-application.  The rules for time adjustment across the world are more political
-than rational, and there is no standard suitable for every application.
+time zone name, and whether Daylight Saving Time is in effect.  Note that only
+one concrete :class:`tzinfo` class, the :class:`timezone` class, is supplied by the
+:mod:`datetime` module.  The :class:`timezone` class can reprsent simple
+timezones with fixed offset from UTC such as UTC itself or North American EST and
+EDT timezones.  Supporting timezones at whatever level of detail is
+required is up to the application.  The rules for time adjustment across the
+world are more political than rational, change frequently, and there is no
+standard suitable for every application aside from UTC.
 
 The :mod:`datetime` module exports the following constants:
 
@@ -99,6 +102,14 @@
    time adjustment (for example, to account for time zone and/or daylight saving
    time).
 
+.. class:: timezone
+
+   A class that implements the :class:`tzinfo` abstract base class as a
+   fixed offset from the UTC.
+
+   .. versionadded:: 3.2
+
+
 Objects of these types are immutable.
 
 Objects of the :class:`date` type are always naive.
@@ -116,6 +127,7 @@
    object
        timedelta
        tzinfo
+           timezone
        time
        date
            datetime
@@ -660,8 +672,8 @@
 
    Return the current UTC date and time, with :attr:`tzinfo` ``None``. This is like
    :meth:`now`, but returns the current UTC date and time, as a naive
-   :class:`datetime` object. See also :meth:`now`.
-
+   :class:`datetime` object.  An aware current UTC datetime can be obtained by
+   calling ``datetime.now(timezone.utc)``.  See also :meth:`now`.
 
 .. classmethod:: datetime.fromtimestamp(timestamp, tz=None)
 
@@ -962,10 +974,10 @@
    ``d.dst()`` returns.  DST is never in effect for a UTC time.
 
    If *d* is aware, *d* is normalized to UTC time, by subtracting
-   ``d.utcoffset()``, and a :class:`time.struct_time` for the normalized time is
-   returned.  :attr:`tm_isdst` is forced to 0. Note that the result's
-   :attr:`tm_year` member may be :const:`MINYEAR`\ -1 or :const:`MAXYEAR`\ +1, if
-   *d*.year was ``MINYEAR`` or ``MAXYEAR`` and UTC adjustment spills over a year
+   ``d.utcoffset()``, and a :class:`time.struct_time` for the
+   normalized time is returned.  :attr:`tm_isdst` is forced to 0. Note
+   that an :exc:`OverflowError` may be raised if *d*.year was
+   ``MINYEAR`` or ``MAXYEAR`` and UTC adjustment spills over a year
    boundary.
 
 
@@ -1318,8 +1330,10 @@
 :class:`tzinfo` is an abstract base class, meaning that this class should not be
 instantiated directly.  You need to derive a concrete subclass, and (at least)
 supply implementations of the standard :class:`tzinfo` methods needed by the
-:class:`datetime` methods you use.  The :mod:`datetime` module does not supply
-any concrete subclasses of :class:`tzinfo`.
+:class:`datetime` methods you use.  The :mod:`datetime` module supplies
+a simple concrete subclass of :class:`tzinfo` :class:`timezone` which can reprsent
+timezones with fixed offset from UTC such as UTC itself or North American EST and
+EDT.
 
 An instance of (a concrete subclass of) :class:`tzinfo` can be passed to the
 constructors for :class:`datetime` and :class:`time` objects. The latter objects
@@ -1337,7 +1351,7 @@
 :mod:`datetime` objects.  If in doubt, simply implement all of them.
 
 
-.. method:: tzinfo.utcoffset(self, dt)
+.. method:: tzinfo.utcoffset(dt)
 
    Return offset of local time from UTC, in minutes east of UTC.  If local time is
    west of UTC, this should be negative.  Note that this is intended to be the
@@ -1359,7 +1373,7 @@
    :exc:`NotImplementedError`.
 
 
-.. method:: tzinfo.dst(self, dt)
+.. method:: tzinfo.dst(dt)
 
    Return the daylight saving time (DST) adjustment, in minutes east of UTC, or
    ``None`` if DST information isn't known.  Return ``timedelta(0)`` if DST is not
@@ -1407,7 +1421,7 @@
    The default implementation of :meth:`dst` raises :exc:`NotImplementedError`.
 
 
-.. method:: tzinfo.tzname(self, dt)
+.. method:: tzinfo.tzname(dt)
 
    Return the time zone name corresponding to the :class:`datetime` object *dt*, as
    a string. Nothing about string names is defined by the :mod:`datetime` module,
@@ -1443,7 +1457,7 @@
 There is one more :class:`tzinfo` method that a subclass may wish to override:
 
 
-.. method:: tzinfo.fromutc(self, dt)
+.. method:: tzinfo.fromutc(dt)
 
    This is called from the default :class:`datetime.astimezone()` implementation.
    When called from that, ``dt.tzinfo`` is *self*, and *dt*'s date and time members
@@ -1520,9 +1534,65 @@
 standard local time.
 
 Applications that can't bear such ambiguities should avoid using hybrid
-:class:`tzinfo` subclasses; there are no ambiguities when using UTC, or any
-other fixed-offset :class:`tzinfo` subclass (such as a class representing only
-EST (fixed offset -5 hours), or only EDT (fixed offset -4 hours)).
+:class:`tzinfo` subclasses; there are no ambiguities when using :class:`timezone`,
+or any other fixed-offset :class:`tzinfo` subclass (such as a class representing
+only EST (fixed offset -5 hours), or only EDT (fixed offset -4 hours)).
+
+
+.. _datetime-timezone:
+
+:class:`timezone` Objects
+--------------------------
+
+A :class:`timezone` object represents a timezone that is defined by a
+fixed offset from UTC.  Note that objects of this class cannot be used
+to represent timezone information in the locations where different
+offsets are used in different days of the year or where historical
+changes have been made to civil time.
+
+
+.. class:: timezone(offset[, name])
+
+  The *offset* argument must be specified as a :class:`timedelta`
+  object representing the difference between the local time and UTC.  It must
+  be strictly between ``-timedelta(hours=24)`` and
+  ``timedelta(hours=24)`` and represent a whole number of minutes,
+  otherwise :exc:`ValueError` is raised.
+
+  The *name* argument is optional.  If specified it must be a string that
+  is used as the value returned by the ``tzname(dt)`` method.  Otherwise,
+  ``tzname(dt)`` returns a string 'UTCsHH:MM', where s is the sign of
+  *offset*, HH and MM are two digits of ``offset.hours`` and
+  ``offset.minutes`` respectively.
+
+.. method:: timezone.utcoffset(dt)
+
+  Return the fixed value specified when the :class:`timezone` instance is
+  constructed.  The *dt* argument is ignored.  The return value is a
+  :class:`timedelta` instance equal to the difference between the
+  local time and UTC.
+
+.. method:: timezone.tzname(dt)
+
+  Return the fixed value specified when the :class:`timezone` instance is
+  constructed or a string 'UTCsHH:MM', where s is the sign of
+  *offset*, HH and MM are two digits of ``offset.hours`` and
+  ``offset.minutes`` respectively.
+
+.. method:: timezone.dst(dt)
+
+  Always returns ``None``.
+
+.. method:: timezone.fromutc(dt)
+
+  Return ``dt + offset``.  The *dt* argument must be an aware
+  :class:`datetime` instance, with ``tzinfo`` set to ``self``.
+
+Class attributes:
+
+.. attribute:: timezone.utc
+
+   The UTC timezone, ``timezone(timedelta(0))``.
 
 
 .. _strftime-strptime-behavior:
@@ -1690,3 +1760,10 @@
 (5)
    For example, if :meth:`utcoffset` returns ``timedelta(hours=-3, minutes=-30)``,
    ``%z`` is replaced with the string ``'-0330'``.
+
+.. versionadded:: 3.2
+
+   When the ``%z`` directive is provided to the :meth:`strptime`
+   method, an aware :class:`datetime` object will be produced.  The
+   ``tzinfo`` of the result will be set to a :class:`timezone`
+   instance.

Modified: python/branches/py3k-cdecimal/Doc/library/dis.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/dis.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/dis.rst	Sat Jul 17 12:36:09 2010
@@ -5,12 +5,19 @@
    :synopsis: Disassembler for Python bytecode.
 
 
-The :mod:`dis` module supports the analysis of Python :term:`bytecode` by
-disassembling it.  Since there is no Python assembler, this module defines the
-Python assembly language.  The Python bytecode which this module takes as an
+The :mod:`dis` module supports the analysis of CPython :term:`bytecode` by
+disassembling it. The CPython bytecode which this module takes as an
 input is defined in the file :file:`Include/opcode.h` and used by the compiler
 and the interpreter.
 
+.. impl-detail::
+
+   Bytecode is an implementation detail of the CPython interpreter!  No
+   guarantees are made that bytecode will not be added, removed, or changed
+   between versions of Python.  Use of this module should not be considered to
+   work across Python VMs or Python releases.
+
+
 Example: Given the function :func:`myfunc`::
 
    def myfunc(alist):
@@ -31,23 +38,27 @@
 
 .. function:: dis(x=None)
 
-   Disassemble the *x* object. *x* can denote either a module, a
-   class, a method, a function, or a code object.   For a module, it disassembles
-   all functions.  For a class, it disassembles all methods.  For a single code
-   sequence, it prints one line per bytecode instruction.  If no object is
-   provided, it disassembles the last traceback.
+   Disassemble the *x* object.  *x* can denote either a module, a
+   class, a method, a function, a code object, a string of source code or a
+   byte sequence of raw bytecode.  For a module, it disassembles all
+   functions.  For a class, it disassembles all methods.  For a code object
+   or sequence of raw bytecode, it prints one line per bytecode instruction.
+   Strings are first compiled to code objects with the :func:`compile`
+   built-in function before being disassembled.  If no object is provided,
+   this function disassembles the last traceback.
 
 
 .. function:: distb(tb=None)
 
-   Disassembles the top-of-stack function of a traceback, using the last traceback
-   if none was passed.  The instruction causing the exception is indicated.
+   Disassemble the top-of-stack function of a traceback, using the last
+   traceback if none was passed.  The instruction causing the exception is
+   indicated.
 
 
 .. function:: disassemble(code, lasti=-1)
               disco(code, lasti=-1)
 
-   Disassembles a code object, indicating the last instruction if *lasti* was
+   Disassemble a code object, indicating the last instruction if *lasti* was
    provided.  The output is divided in the following columns:
 
    #. the line number, for the first instruction of each line
@@ -133,225 +144,233 @@
 The Python compiler currently generates the following bytecode instructions.
 
 
-.. opcode:: STOP_CODE ()
+**General instructions**
+
+.. opcode:: STOP_CODE
 
    Indicates end-of-code to the compiler, not used by the interpreter.
 
 
-.. opcode:: NOP ()
+.. opcode:: NOP
 
    Do nothing code.  Used as a placeholder by the bytecode optimizer.
 
 
-.. opcode:: POP_TOP ()
+.. opcode:: POP_TOP
 
    Removes the top-of-stack (TOS) item.
 
 
-.. opcode:: ROT_TWO ()
+.. opcode:: ROT_TWO
 
    Swaps the two top-most stack items.
 
 
-.. opcode:: ROT_THREE ()
+.. opcode:: ROT_THREE
 
    Lifts second and third stack item one position up, moves top down to position
    three.
 
 
-.. opcode:: ROT_FOUR ()
+.. opcode:: ROT_FOUR
 
    Lifts second, third and forth stack item one position up, moves top down to
    position four.
 
 
-.. opcode:: DUP_TOP ()
+.. opcode:: DUP_TOP
 
    Duplicates the reference on top of the stack.
 
-Unary Operations take the top of the stack, apply the operation, and push the
-result back on the stack.
 
+**Unary operations**
+
+Unary operations take the top of the stack, apply the operation, and push the
+result back on the stack.
 
-.. opcode:: UNARY_POSITIVE ()
+.. opcode:: UNARY_POSITIVE
 
    Implements ``TOS = +TOS``.
 
 
-.. opcode:: UNARY_NEGATIVE ()
+.. opcode:: UNARY_NEGATIVE
 
    Implements ``TOS = -TOS``.
 
 
-.. opcode:: UNARY_NOT ()
+.. opcode:: UNARY_NOT
 
    Implements ``TOS = not TOS``.
 
 
-.. opcode:: UNARY_INVERT ()
+.. opcode:: UNARY_INVERT
 
    Implements ``TOS = ~TOS``.
 
 
-.. opcode:: GET_ITER ()
+.. opcode:: GET_ITER
 
    Implements ``TOS = iter(TOS)``.
 
+
+**Binary operations**
+
 Binary operations remove the top of the stack (TOS) and the second top-most
 stack item (TOS1) from the stack.  They perform the operation, and put the
 result back on the stack.
 
-
-.. opcode:: BINARY_POWER ()
+.. opcode:: BINARY_POWER
 
    Implements ``TOS = TOS1 ** TOS``.
 
 
-.. opcode:: BINARY_MULTIPLY ()
+.. opcode:: BINARY_MULTIPLY
 
    Implements ``TOS = TOS1 * TOS``.
 
 
-.. opcode:: BINARY_FLOOR_DIVIDE ()
+.. opcode:: BINARY_FLOOR_DIVIDE
 
    Implements ``TOS = TOS1 // TOS``.
 
 
-.. opcode:: BINARY_TRUE_DIVIDE ()
+.. opcode:: BINARY_TRUE_DIVIDE
 
    Implements ``TOS = TOS1 / TOS``.
 
 
-.. opcode:: BINARY_MODULO ()
+.. opcode:: BINARY_MODULO
 
    Implements ``TOS = TOS1 % TOS``.
 
 
-.. opcode:: BINARY_ADD ()
+.. opcode:: BINARY_ADD
 
    Implements ``TOS = TOS1 + TOS``.
 
 
-.. opcode:: BINARY_SUBTRACT ()
+.. opcode:: BINARY_SUBTRACT
 
    Implements ``TOS = TOS1 - TOS``.
 
 
-.. opcode:: BINARY_SUBSCR ()
+.. opcode:: BINARY_SUBSCR
 
    Implements ``TOS = TOS1[TOS]``.
 
 
-.. opcode:: BINARY_LSHIFT ()
+.. opcode:: BINARY_LSHIFT
 
    Implements ``TOS = TOS1 << TOS``.
 
 
-.. opcode:: BINARY_RSHIFT ()
+.. opcode:: BINARY_RSHIFT
 
    Implements ``TOS = TOS1 >> TOS``.
 
 
-.. opcode:: BINARY_AND ()
+.. opcode:: BINARY_AND
 
    Implements ``TOS = TOS1 & TOS``.
 
 
-.. opcode:: BINARY_XOR ()
+.. opcode:: BINARY_XOR
 
    Implements ``TOS = TOS1 ^ TOS``.
 
 
-.. opcode:: BINARY_OR ()
+.. opcode:: BINARY_OR
 
    Implements ``TOS = TOS1 | TOS``.
 
+
+**In-place operations**
+
 In-place operations are like binary operations, in that they remove TOS and
 TOS1, and push the result back on the stack, but the operation is done in-place
 when TOS1 supports it, and the resulting TOS may be (but does not have to be)
 the original TOS1.
 
-
-.. opcode:: INPLACE_POWER ()
+.. opcode:: INPLACE_POWER
 
    Implements in-place ``TOS = TOS1 ** TOS``.
 
 
-.. opcode:: INPLACE_MULTIPLY ()
+.. opcode:: INPLACE_MULTIPLY
 
    Implements in-place ``TOS = TOS1 * TOS``.
 
 
-.. opcode:: INPLACE_FLOOR_DIVIDE ()
+.. opcode:: INPLACE_FLOOR_DIVIDE
 
    Implements in-place ``TOS = TOS1 // TOS``.
 
 
-.. opcode:: INPLACE_TRUE_DIVIDE ()
+.. opcode:: INPLACE_TRUE_DIVIDE
 
    Implements in-place ``TOS = TOS1 / TOS``.
 
 
-.. opcode:: INPLACE_MODULO ()
+.. opcode:: INPLACE_MODULO
 
    Implements in-place ``TOS = TOS1 % TOS``.
 
 
-.. opcode:: INPLACE_ADD ()
+.. opcode:: INPLACE_ADD
 
    Implements in-place ``TOS = TOS1 + TOS``.
 
 
-.. opcode:: INPLACE_SUBTRACT ()
+.. opcode:: INPLACE_SUBTRACT
 
    Implements in-place ``TOS = TOS1 - TOS``.
 
 
-.. opcode:: INPLACE_LSHIFT ()
+.. opcode:: INPLACE_LSHIFT
 
    Implements in-place ``TOS = TOS1 << TOS``.
 
 
-.. opcode:: INPLACE_RSHIFT ()
+.. opcode:: INPLACE_RSHIFT
 
    Implements in-place ``TOS = TOS1 >> TOS``.
 
 
-.. opcode:: INPLACE_AND ()
+.. opcode:: INPLACE_AND
 
    Implements in-place ``TOS = TOS1 & TOS``.
 
 
-.. opcode:: INPLACE_XOR ()
+.. opcode:: INPLACE_XOR
 
    Implements in-place ``TOS = TOS1 ^ TOS``.
 
 
-.. opcode:: INPLACE_OR ()
+.. opcode:: INPLACE_OR
 
    Implements in-place ``TOS = TOS1 | TOS``.
 
 
-.. opcode:: STORE_SUBSCR ()
+.. opcode:: STORE_SUBSCR
 
    Implements ``TOS1[TOS] = TOS2``.
 
 
-.. opcode:: DELETE_SUBSCR ()
+.. opcode:: DELETE_SUBSCR
 
    Implements ``del TOS1[TOS]``.
 
-Miscellaneous opcodes.
 
+**Miscellaneous opcodes**
 
-.. opcode:: PRINT_EXPR ()
+.. opcode:: PRINT_EXPR
 
    Implements the expression statement for the interactive mode.  TOS is removed
    from the stack and printed.  In non-interactive mode, an expression statement is
    terminated with ``POP_STACK``.
 
 
-.. opcode:: BREAK_LOOP ()
+.. opcode:: BREAK_LOOP
 
    Terminates a loop due to a :keyword:`break` statement.
 
@@ -377,36 +396,35 @@
    Calls ``dict.setitem(TOS1[-i], TOS, TOS1)``.  Used to implement dict
    comprehensions.
 
-
 For all of the SET_ADD, LIST_APPEND and MAP_ADD instructions, while the
 added value or key/value pair is popped off, the container object remains on
 the stack so that it is available for further iterations of the loop.
 
 
-.. opcode:: RETURN_VALUE ()
+.. opcode:: RETURN_VALUE
 
    Returns with TOS to the caller of the function.
 
 
-.. opcode:: YIELD_VALUE ()
+.. opcode:: YIELD_VALUE
 
    Pops ``TOS`` and yields it from a :term:`generator`.
 
 
-.. opcode:: IMPORT_STAR ()
+.. opcode:: IMPORT_STAR
 
    Loads all symbols not starting with ``'_'`` directly from the module TOS to the
    local namespace. The module is popped after loading all names. This opcode
    implements ``from module import *``.
 
 
-.. opcode:: POP_BLOCK ()
+.. opcode:: POP_BLOCK
 
    Removes one block from the block stack.  Per frame, there is a  stack of blocks,
    denoting nested loops, try statements, and such.
 
 
-.. opcode:: POP_EXCEPT ()
+.. opcode:: POP_EXCEPT
 
    Removes one block from the block stack. The popped block must be an exception
    handler block, as implicitly created when entering an except handler.
@@ -414,14 +432,14 @@
    last three popped values are used to restore the exception state.
 
 
-.. opcode:: END_FINALLY ()
+.. opcode:: END_FINALLY
 
    Terminates a :keyword:`finally` clause.  The interpreter recalls whether the
    exception has to be re-raised, or whether the function returns, and continues
    with the outer-next block.
 
 
-.. opcode:: LOAD_BUILD_CLASS ()
+.. opcode:: LOAD_BUILD_CLASS
 
    Pushes :func:`builtins.__build_class__` onto the stack.  It is later called
    by ``CALL_FUNCTION`` to construct a class.
@@ -439,7 +457,7 @@
    :opcode:`UNPACK_SEQUENCE`).
 
 
-.. opcode:: WITH_CLEANUP ()
+.. opcode:: WITH_CLEANUP
 
    Cleans up the stack when a :keyword:`with` statement block exits.  TOS is
    the context manager's :meth:`__exit__` bound method. Below TOS are 1--3
@@ -649,7 +667,7 @@
    Pushes a try block from a try-except clause onto the block stack. *delta* points
    to the finally block.
 
-.. opcode:: STORE_MAP ()
+.. opcode:: STORE_MAP
 
    Store a key and value pair in a dictionary.  Pops the key and value while leaving
    the dictionary on the stack.
@@ -765,7 +783,7 @@
    variable-arguments tuple, followed by explicit keyword and positional arguments.
 
 
-.. opcode:: HAVE_ARGUMENT ()
+.. opcode:: HAVE_ARGUMENT
 
    This is not really an opcode.  It identifies the dividing line between opcodes
    which don't take arguments ``< HAVE_ARGUMENT`` and those which do ``>=

Modified: python/branches/py3k-cdecimal/Doc/library/doctest.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/doctest.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/doctest.rst	Sat Jul 17 12:36:09 2010
@@ -282,11 +282,8 @@
 How are Docstring Examples Recognized?
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
-In most cases a copy-and-paste of an interactive console session works fine, but
-doctest isn't trying to do an exact emulation of any specific Python shell.  All
-hard tab characters are expanded to spaces, using 8-column tab stops.  If you
-don't believe tabs should mean that, too bad:  don't use hard tabs, or write
-your own :class:`DocTestParser` class.
+In most cases a copy-and-paste of an interactive console session works fine,
+but doctest isn't trying to do an exact emulation of any specific Python shell.
 
 ::
 
@@ -317,6 +314,17 @@
   blank line, put ``<BLANKLINE>`` in your doctest example each place a blank line
   is expected.
 
+* All hard tab characters are expanded to spaces, using 8-column tab stops.
+  Tabs in output generated by the tested code are not modified.  Because any
+  hard tabs in the sample output *are* expanded, this means that if the code
+  output includes hard tabs, the only way the doctest can pass is if the
+  :const:`NORMALIZE_WHITESPACE` option or directive is in effect.
+  Alternatively, the test can be rewritten to capture the output and compare it
+  to an expected value as part of the test.  This handling of tabs in the
+  source was arrived at through trial and error, and has proven to be the least
+  error prone way of handling them.  It is possible to use a different
+  algorithm for handling tabs by writing a custom :class:`DocTestParser` class.
+
 * Output to stdout is captured, but not output to stderr (exception tracebacks
   are captured via a different means).
 
@@ -905,18 +913,16 @@
 As your collection of doctest'ed modules grows, you'll want a way to run all
 their doctests systematically.  :mod:`doctest` provides two functions that can
 be used to create :mod:`unittest` test suites from modules and text files
-containing doctests.  These test suites can then be run using :mod:`unittest`
-test runners::
+containing doctests.  To integrate with :mod:`unittest` test discovery, include
+a :func:`load_tests` function in your test module::
 
    import unittest
    import doctest
-   import my_module_with_doctests, and_another
+   import my_module_with_doctests
 
-   suite = unittest.TestSuite()
-   for mod in my_module_with_doctests, and_another:
-       suite.addTest(doctest.DocTestSuite(mod))
-   runner = unittest.TextTestRunner()
-   runner.run(suite)
+   def load_tests(loader, tests, ignore):
+       tests.addTests(doctest.DocTestSuite(my_module_with_doctests))
+       return test
 
 There are two main functions for creating :class:`unittest.TestSuite` instances
 from text files and modules with doctests:
@@ -1778,4 +1784,3 @@
 .. [#] Examples containing both expected output and an exception are not supported.
    Trying to guess where one ends and the other begins is too error-prone, and that
    also makes for a confusing test.
-

Modified: python/branches/py3k-cdecimal/Doc/library/fnmatch.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/fnmatch.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/fnmatch.rst	Sat Jul 17 12:36:09 2010
@@ -82,6 +82,13 @@
       <_sre.SRE_Match object at 0x...>
 
 
+.. function:: purge()
+
+   Clear the internal pattern cache.
+
+   .. versionadded:: 3.2
+
+
 .. seealso::
 
    Module :mod:`glob`

Modified: python/branches/py3k-cdecimal/Doc/library/functions.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/functions.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/functions.rst	Sat Jul 17 12:36:09 2010
@@ -335,6 +335,9 @@
    returns the current global and local dictionary, respectively, which may be
    useful to pass around for use by :func:`eval` or :func:`exec`.
 
+   See :func:`ast.literal_eval` for a function that can safely evaluate strings
+   with expressions containing only literals.
+
 
 .. function:: exec(object[, globals[, locals]])
 
@@ -676,7 +679,7 @@
    :meth:`__index__` method that returns an integer.
 
 
-.. function:: open(file, mode='r', buffering=None, encoding=None, errors=None, newline=None, closefd=True)
+.. function:: open(file, mode='r', buffering=-1, encoding=None, errors=None, newline=None, closefd=True)
 
    Open *file* and return a corresponding stream.  If the file cannot be opened,
    an :exc:`IOError` is raised.

Modified: python/branches/py3k-cdecimal/Doc/library/http.cookiejar.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/http.cookiejar.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/http.cookiejar.rst	Sat Jul 17 12:36:09 2010
@@ -108,10 +108,6 @@
       :mod:`http.cookiejar` and :mod:`http.cookies` modules do not depend on each
       other.
 
-   http://wwwsearch.sourceforge.net/mechanize/
-      Extensions to this module, including a class for reading Microsoft Internet
-      Explorer cookies on Windows.
-
    http://wp.netscape.com/newsref/std/cookie_spec.html
       The specification of the original Netscape cookie protocol.  Though this is
       still the dominant protocol, the 'Netscape cookie protocol' implemented by all
@@ -292,11 +288,8 @@
 FileCookieJar subclasses and co-operation with web browsers
 -----------------------------------------------------------
 
-The following :class:`CookieJar` subclasses are provided for reading and writing
-.  Further :class:`CookieJar` subclasses, including one that reads Microsoft
-Internet Explorer cookies, are available at
-http://wwwsearch.sourceforge.net/mechanize/.
-
+The following :class:`CookieJar` subclasses are provided for reading and
+writing .
 
 .. class:: MozillaCookieJar(filename, delayload=None, policy=None)
 

Modified: python/branches/py3k-cdecimal/Doc/library/http.server.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/http.server.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/http.server.rst	Sat Jul 17 12:36:09 2010
@@ -281,7 +281,31 @@
       contents of the file are output. If the file's MIME type starts with
       ``text/`` the file is opened in text mode; otherwise binary mode is used.
 
-      For example usage, see the implementation of the :func:`test` function.
+      For example usage, see the implementation of the :func:`test` function
+      invocation in the :mod:`http.server` module.
+
+
+The :class:`SimpleHTTPRequestHandler` class can be used in the following
+manner in order to create a very basic webserver serving files relative to
+the current directory. ::
+
+   import http.server
+   import socketserver
+
+   PORT = 8000
+
+   Handler = http.server.SimpleHTTPRequestHandler
+
+   httpd = socketserver.TCPServer(("", PORT), Handler)
+
+   print("serving at port", PORT)
+   httpd.serve_forever()
+
+:mod:`http.server` can also be invoked directly using the :option:`-m`
+switch of the interpreter a with ``port number`` argument.  Similar to
+the previous example, this serves files relative to the current directory. ::
+
+        python -m http.server 8000
 
 
 .. class:: CGIHTTPRequestHandler(request, client_address, server)

Modified: python/branches/py3k-cdecimal/Doc/library/importlib.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/importlib.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/importlib.rst	Sat Jul 17 12:36:09 2010
@@ -18,12 +18,12 @@
 :func:`__import__` function) in Python source code. This provides an
 implementation of :keyword:`import` which is portable to any Python
 interpreter. This also provides a reference implementation which is easier to
-comprehend than one in a programming language other than Python.
+comprehend than one implemented in a programming language other than Python.
 
-Two, the components to implement :keyword:`import` can be exposed in this
+Two, the components to implement :keyword:`import` are exposed in this
 package, making it easier for users to create their own custom objects (known
 generically as an :term:`importer`) to participate in the import process.
-Details on providing custom importers can be found in :pep:`302`.
+Details on custom importers can be found in :pep:`302`.
 
 .. seealso::
 
@@ -36,8 +36,7 @@
         in :data:`sys.modules`).
 
     The :func:`.__import__` function
-        The built-in function for which the :keyword:`import` statement is
-        syntactic sugar.
+        The :keyword:`import` statement is syntactic sugar for this function.
 
     :pep:`235`
         Import on Case-Insensitive Platforms
@@ -46,7 +45,7 @@
         Defining Python Source Code Encodings
 
     :pep:`302`
-        New Import Hooks.
+        New Import Hooks
 
     :pep:`328`
         Imports: Multi-Line and Absolute/Relative
@@ -57,14 +56,16 @@
     :pep:`3120`
         Using UTF-8 as the Default Source Encoding
 
+    :pep:`3147`
+        PYC Repository Directories
+
 
 Functions
 ---------
 
 .. function:: __import__(name, globals={}, locals={}, fromlist=list(), level=0)
 
-    An implementation of the built-in :func:`__import__` function. See the
-    built-in function's documentation for usage instructions.
+    An implementation of the built-in :func:`__import__` function.
 
 .. function:: import_module(name, package=None)
 
@@ -210,22 +211,112 @@
 
     .. method:: get_filename(fullname)
 
-        An abstract method that is to return the value for :attr:`__file__` for
+        An abstract method that is to return the value of :attr:`__file__` for
         the specified module. If no path is available, :exc:`ImportError` is
         raised.
 
+        If source code is available, then the method should return the path to
+        the source file, regardless of whether a bytecode was used to load the
+        module.
+
+
+.. class:: SourceLoader
+
+    An abstract base class for implementing source (and optionally bytecode)
+    file loading. The class inherits from both :class:`ResourceLoader` and
+    :class:`ExecutionLoader`, requiring the implementation of:
+
+    * :meth:`ResourceLoader.get_data`
+    * :meth:`ExecutionLoader.get_filename`
+          Implement to only return the path to the source file; sourceless
+          loading is not supported.
+
+    The abstract methods defined by this class are to add optional bytecode
+    file support. Not implementing these optional methods causes the loader to
+    only work with source code. Implementing the methods allows the loader to
+    work with source *and* bytecode files; it does not allow for *sourceless*
+    loading where only bytecode is provided.  Bytecode files are an
+    optimization to speed up loading by removing the parsing step of Python's
+    compiler, and so no bytecode-specific API is exposed.
+
+    .. method:: path_mtime(self, path)
+
+        Optional abstract method which returns the modification time for the
+        specified path.
+
+    .. method:: set_data(self, path, data)
+
+        Optional abstract method which writes the specified bytes to a file
+        path. Any intermediate directories which do not exist are to be created
+        automatically.
+
+        When writing to the path fails because the path is read-only
+        (:attr:`errno.EACCES`), do not propagate the exception.
+
+    .. method:: get_code(self, fullname)
+
+        Concrete implementation of :meth:`InspectLoader.get_code`.
+
+    .. method:: load_module(self, fullname)
+
+        Concrete implementation of :meth:`Loader.load_module`.
+
+    .. method:: get_source(self, fullname)
+
+        Concrete implementation of :meth:`InspectLoader.get_source`.
+
+    .. method:: is_package(self, fullname)
+
+        Concrete implementation of :meth:`InspectLoader.is_package`. A module
+        is determined to be a package if its file path is a file named
+        ``__init__`` when the file extension is removed.
+
 
 .. class:: PyLoader
 
     An abstract base class inheriting from
-    :class:`importlib.abc.ExecutionLoader` and
-    :class:`importlib.abc.ResourceLoader` designed to ease the loading of
+    :class:`ExecutionLoader` and
+    :class:`ResourceLoader` designed to ease the loading of
     Python source modules (bytecode is not handled; see
-    :class:`importlib.abc.PyPycLoader` for a source/bytecode ABC). A subclass
+    :class:`SourceLoader` for a source/bytecode ABC). A subclass
     implementing this ABC will only need to worry about exposing how the source
     code is stored; all other details for loading Python source code will be
     handled by the concrete implementations of key methods.
 
+    .. deprecated:: 3.2
+        This class has been deprecated in favor of :class:`SourceLoader` and is
+        slated for removal in Python 3.4. See below for how to create a
+        subclass that is compatbile with Python 3.1 onwards.
+
+    If compatibility with Python 3.1 is required, then use the following idiom
+    to implement a subclass that will work with Python 3.1 onwards (make sure
+    to implement :meth:`ExecutionLoader.get_filename`)::
+
+        try:
+            from importlib.abc import SourceLoader
+        except ImportError:
+            from importlib.abc import PyLoader as SourceLoader
+
+
+        class CustomLoader(SourceLoader):
+            def get_filename(self, fullname):
+                """Return the path to the source file."""
+                # Implement ...
+
+            def source_path(self, fullname):
+                """Implement source_path in terms of get_filename."""
+                try:
+                    return self.get_filename(fullname)
+                except ImportError:
+                    return None
+
+            def is_package(self, fullname):
+                """Implement is_package by looking for an __init__ file
+                name as returned by get_filename."""
+                filename = os.path.basename(self.get_filename(fullname))
+                return os.path.splitext(filename)[0] == '__init__'
+
+
     .. method:: source_path(fullname)
 
         An abstract method that returns the path to the source code for a
@@ -267,10 +358,18 @@
 
 .. class:: PyPycLoader
 
-    An abstract base class inheriting from :class:`importlib.abc.PyLoader`.
+    An abstract base class inheriting from :class:`PyLoader`.
     This ABC is meant to help in creating loaders that support both Python
     source and bytecode.
 
+    .. deprecated:: 3.2
+        This class has been deprecated in favor of :class:`SourceLoader` and to
+        properly support :pep:`3147`. If compatibility is required with
+        Python 3.1, implement both :class:`SourceLoader` and :class:`PyLoader`;
+        instructions on how to do so are included in the documentation for
+        :class:`PyLoader`. Do note that this solution will not support
+        sourceless/bytecode-only loading; only source *and* bytecode loading.
+
     .. method:: source_mtime(fullname)
 
         An abstract method which returns the modification time for the source
@@ -289,8 +388,8 @@
     .. method:: get_filename(fullname)
 
         A concrete implementation of
-        :meth:`importlib.abc.ExecutionLoader.get_filename` that relies on
-        :meth:`importlib.abc.PyLoader.source_path` and :meth:`bytecode_path`.
+        :meth:`ExecutionLoader.get_filename` that relies on
+        :meth:`PyLoader.source_path` and :meth:`bytecode_path`.
         If :meth:`source_path` returns a path, then that value is returned.
         Else if :meth:`bytecode_path` returns a path, that path will be
         returned. If a path is not available from both methods,
@@ -417,100 +516,3 @@
     attribute to be used at the global level of the module during
     initialization.
 
-
-Example
--------
-
-Below is an example meta path importer that uses a dict for back-end storage
-for source code. While not an optimal solution -- manipulations of
-:attr:`__path__` on packages does not influence import -- it does illustrate
-what little is required to implement an importer.
-
-.. testcode::
-
-    """An importer where source is stored in a dict."""
-    from importlib import abc
-
-
-    class DictImporter(abc.Finder, abc.PyLoader):
-
-        """A meta path importer that stores source code in a dict.
-
-        The keys are the module names -- packages must end in ``.__init__``.
-        The values must be something that can be passed to 'bytes'.
-
-        """
-
-        def __init__(self, memory):
-            """Store the dict."""
-            self.memory = memory
-
-        def contains(self, name):
-            """See if a module or package is in the dict."""
-            if name in self.memory:
-                return name
-            package_name = '{}.__init__'.format(name)
-            if  package_name in self.memory:
-                return package_name
-            return False
-
-        __contains__ = contains  # Convenience.
-
-        def find_module(self, fullname, path=None):
-            """Find the module in the dict."""
-            if fullname in self:
-                return self
-            return None
-
-        def source_path(self, fullname):
-            """Return the module name if the module is in the dict."""
-            if not fullname in self:
-                raise ImportError
-            return fullname
-
-        def get_data(self, path):
-            """Return the bytes for the source.
-
-            The value found in the dict is passed through 'bytes' before being
-            returned.
-
-            """
-            name = self.contains(path)
-            if not name:
-                raise IOError
-            return bytes(self.memory[name])
-
-        def is_package(self, fullname):
-            """Tell if module is a package based on whether the dict contains the
-            name with ``.__init__`` appended to it."""
-            if fullname not in self:
-                raise ImportError
-            if fullname in self.memory:
-                return False
-            # If name is in this importer but not as it is then it must end in
-            # ``__init__``.
-            else:
-                return True
-
-.. testcode::
-    :hide:
-
-    import importlib
-    import sys
-
-
-    # Build the dict; keys of name, value of __package__.
-    names = {'_top_level': '', '_pkg.__init__': '_pkg', '_pkg.mod': '_pkg'}
-    source = {name: "name = {!r}".format(name).encode() for name in names}
-
-    # Register the meta path importer.
-    importer = DictImporter(source)
-    sys.meta_path.append(importer)
-
-    # Sanity check.
-    for name in names:
-        module = importlib.import_module(name)
-        assert module.__name__ == name
-        assert getattr(module, 'name') == name
-        assert module.__loader__ is importer
-        assert module.__package__ == names[name]

Modified: python/branches/py3k-cdecimal/Doc/library/logging.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/logging.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/logging.rst	Sat Jul 17 12:36:09 2010
@@ -53,10 +53,12 @@
 
 Most applications are probably going to want to log to a file, so let's start
 with that case. Using the :func:`basicConfig` function, we can set up the
-default handler so that debug messages are written to a file::
+default handler so that debug messages are written to a file (in the example,
+we assume that you have the appropriate permissions to create a file called
+*example.log* in the current directory)::
 
    import logging
-   LOG_FILENAME = '/tmp/logging_example.out'
+   LOG_FILENAME = 'example.log'
    logging.basicConfig(filename=LOG_FILENAME,level=logging.DEBUG)
 
    logging.debug('This message should go to the log file')
@@ -75,7 +77,7 @@
    import logging
    import logging.handlers
 
-   LOG_FILENAME = '/tmp/logging_rotatingfile_example.out'
+   LOG_FILENAME = 'logging_rotatingfile_example.out'
 
    # Set up a specific logger with our desired output level
    my_logger = logging.getLogger('MyLogger')
@@ -100,14 +102,14 @@
 The result should be 6 separate files, each with part of the log history for the
 application::
 
-   /tmp/logging_rotatingfile_example.out
-   /tmp/logging_rotatingfile_example.out.1
-   /tmp/logging_rotatingfile_example.out.2
-   /tmp/logging_rotatingfile_example.out.3
-   /tmp/logging_rotatingfile_example.out.4
-   /tmp/logging_rotatingfile_example.out.5
+   logging_rotatingfile_example.out
+   logging_rotatingfile_example.out.1
+   logging_rotatingfile_example.out.2
+   logging_rotatingfile_example.out.3
+   logging_rotatingfile_example.out.4
+   logging_rotatingfile_example.out.5
 
-The most current file is always :file:`/tmp/logging_rotatingfile_example.out`,
+The most current file is always :file:`logging_rotatingfile_example.out`,
 and each time it reaches the size limit it is renamed with the suffix
 ``.1``. Each of the existing backup files is renamed to increment the suffix
 (``.1`` becomes ``.2``, etc.)  and the ``.6`` file is erased.
@@ -244,7 +246,7 @@
   methods listed above, but this is how to log at custom log levels.
 
 :func:`getLogger` returns a reference to a logger instance with the specified
-if it is provided, or ``root`` if not.  The names are period-separated
+name if it is provided, or ``root`` if not.  The names are period-separated
 hierarchical structures.  Multiple calls to :func:`getLogger` with the same name
 will return a reference to the same logger object.  Loggers that are further
 down in the hierarchical list are children of loggers higher up in the list.
@@ -321,24 +323,34 @@
 Configuring Logging
 ^^^^^^^^^^^^^^^^^^^
 
-Programmers can configure logging either by creating loggers, handlers, and
-formatters explicitly in a main module with the configuration methods listed
-above (using Python code), or by creating a logging config file.  The following
-code is an example of configuring a very simple logger, a console handler, and a
-simple formatter in a Python module::
+Programmers can configure logging in three ways:
+
+1. Creating loggers, handlers, and formatters explicitly using Python
+   code that calls the configuration methods listed above.
+2. Creating a logging config file and reading it using the :func:`fileConfig`
+   function.
+3. Creating a dictionary of configuration information and passing it
+   to the :func:`dictConfig` function.
+
+The following example configures a very simple logger, a console
+handler, and a simple formatter using Python code::
 
     import logging
 
     # create logger
     logger = logging.getLogger("simple_example")
     logger.setLevel(logging.DEBUG)
+
     # create console handler and set level to debug
     ch = logging.StreamHandler()
     ch.setLevel(logging.DEBUG)
+
     # create formatter
     formatter = logging.Formatter("%(asctime)s - %(name)s - %(levelname)s - %(message)s")
+
     # add formatter to ch
     ch.setFormatter(formatter)
+
     # add ch to logger
     logger.addHandler(ch)
 
@@ -428,6 +440,45 @@
 class defined in package `mypackage` and module `mymodule`, where `mypackage`
 is available on the Python import path).
 
+.. versionchanged:: 3.2
+
+In Python 3.2, a new means of configuring logging has been introduced, using
+dictionaries to hold configuration information. This provides a superset of the
+functionality of the config-file-based approach outlined above, and is the
+recommended configuration method for new applications and deployments. Because
+a Python dictionary is used to hold configuration information, and since you
+can populate that dictionary using different means, you have more options for
+configuration. For example, you can use a configuration file in JSON format,
+or, if you have access to YAML processing functionality, a file in YAML
+format, to populate the configuration dictionary. Or, of course, you can
+construct the dictionary in Python code, receive it in pickled form over a
+socket, or use whatever approach makes sense for your application.
+
+Here's an example of the same configuration as above, in YAML format for
+the new dictionary-based approach::
+
+    version: 1
+    formatters:
+      simple:
+        format: format=%(asctime)s - %(name)s - %(levelname)s - %(message)s
+    handlers:
+      console:
+        class: logging.StreamHandler
+        level: DEBUG
+        formatter: simple
+        stream: ext://sys.stdout
+    loggers:
+      simpleExample:
+        level: DEBUG
+        handlers: [console]
+        propagate: no
+    root:
+        level: DEBUG
+        handlers: [console]
+
+For more information about logging using a dictionary, see
+:ref:`logging-config-api`.
+
 .. _library-config:
 
 Configuring Logging for a Library
@@ -1065,14 +1116,14 @@
 
    logging.basicConfig(level=logging.DEBUG,
                        format='%(asctime)s %(levelname)s %(message)s',
-                       filename='/tmp/myapp.log',
+                       filename='myapp.log',
                        filemode='w')
    logging.debug('A debug message')
    logging.info('Some information')
    logging.warning('A shot across the bows')
 
 The :meth:`basicConfig` method is used to change the configuration defaults,
-which results in output (written to ``/tmp/myapp.log``) which should look
+which results in output (written to ``myapp.log``) which should look
 something like the following::
 
    2004-07-02 13:00:08,743 DEBUG A debug message
@@ -1364,6 +1415,8 @@
    2008-01-18 14:49:54,033 d.e.f WARNING  IP: 127.0.0.1       User: jim      A message at WARNING level with 2 parameters
 
 
+.. _multiple-processes:
+
 Logging to a single file from multiple processes
 ------------------------------------------------
 
@@ -1519,6 +1572,11 @@
       69 myapp.area2     WARNING  Jail zesty vixen who grabbed pay from quack.
       69 myapp.area2     ERROR    The five boxing wizards jump quickly.
 
+Note that there are some security issues with pickle in some scenarios. If
+these affect you, you can use an alternative serialization scheme by overriding
+the :meth:`makePickle` method and implementing your alternative there, as
+well as adapting the above script to use your alternative serialization.
+
 Using arbitrary objects as messages
 -----------------------------------
 
@@ -1674,6 +1732,8 @@
    :exc:`NotImplementedError`.
 
 
+.. _stream-handler:
+
 StreamHandler
 ^^^^^^^^^^^^^
 
@@ -1707,6 +1767,8 @@
       no output, so an explicit :meth:`flush` call may be needed at times.
 
 
+.. _file-handler:
+
 FileHandler
 ^^^^^^^^^^^
 
@@ -1715,7 +1777,7 @@
 :class:`StreamHandler`.
 
 
-.. class:: FileHandler(filename, mode='a', encoding=None, delay=0)
+.. class:: FileHandler(filename, mode='a', encoding=None, delay=False)
 
    Returns a new instance of the :class:`FileHandler` class. The specified file is
    opened and used as the stream for logging. If *mode* is not specified,
@@ -1733,6 +1795,7 @@
 
       Outputs the record to the file.
 
+.. _null-handler:
 
 NullHandler
 ^^^^^^^^^^^
@@ -1756,6 +1819,8 @@
 See :ref:`library-config` for more information on how to use
 :class:`NullHandler`.
 
+.. _watched-file-handler:
+
 WatchedFileHandler
 ^^^^^^^^^^^^^^^^^^
 
@@ -1794,6 +1859,7 @@
       changed.  If it has, the existing stream is flushed and closed and the
       file opened again, before outputting the record to the file.
 
+.. _rotating-file-handler:
 
 RotatingFileHandler
 ^^^^^^^^^^^^^^^^^^^
@@ -1834,6 +1900,7 @@
       Outputs the record to the file, catering for rollover as described
       previously.
 
+.. _timed-rotating-file-handler:
 
 TimedRotatingFileHandler
 ^^^^^^^^^^^^^^^^^^^^^^^^
@@ -1843,7 +1910,7 @@
 timed intervals.
 
 
-.. class:: TimedRotatingFileHandler(filename, when='h', interval=1, backupCount=0, encoding=None, delay=0, utc=False)
+.. class:: TimedRotatingFileHandler(filename, when='h', interval=1, backupCount=0, encoding=None, delay=False, utc=False)
 
    Returns a new instance of the :class:`TimedRotatingFileHandler` class. The
    specified file is opened and used as the stream for logging. On rotating it also
@@ -1886,6 +1953,9 @@
    one is deleted. The deletion logic uses the interval to determine which
    files to delete, so changing the interval may leave old files lying around.
 
+   If *delay* is true, then file opening is deferred until the first call to
+   :meth:`emit`.
+
 
    .. method:: doRollover()
 
@@ -1897,6 +1967,8 @@
       Outputs the record to the file, catering for rollover as described above.
 
 
+.. _socket-handler:
+
 SocketHandler
 ^^^^^^^^^^^^^
 
@@ -1943,6 +2015,11 @@
       Pickles the record's attribute dictionary in binary format with a length
       prefix, and returns it ready for transmission across the socket.
 
+      Note that pickles aren't completely secure. If you are concerned about
+      security, you may want to override this method to implement a more secure
+      mechanism. For example, you can sign pickles using HMAC and then verify
+      them on the receiving end, or alternatively you can disable unpickling of
+      global objects on the receiving end.
 
    .. method:: send(packet)
 
@@ -1950,6 +2027,8 @@
       partial sends which can happen when the network is busy.
 
 
+.. _datagram-handler:
+
 DatagramHandler
 ^^^^^^^^^^^^^^^
 
@@ -1983,6 +2062,8 @@
       Send a pickled string to a socket.
 
 
+.. _syslog-handler:
+
 SysLogHandler
 ^^^^^^^^^^^^^
 
@@ -2172,6 +2253,7 @@
       lookup to get the message ID. This version returns 1, which is the base
       message ID in :file:`win32service.pyd`.
 
+.. _smtp-handler:
 
 SMTPHandler
 ^^^^^^^^^^^
@@ -2200,6 +2282,7 @@
       If you want to specify a subject line which is record-dependent, override
       this method.
 
+.. _memory-handler:
 
 MemoryHandler
 ^^^^^^^^^^^^^
@@ -2270,6 +2353,8 @@
       Checks for buffer full or a record at the *flushLevel* or higher.
 
 
+.. _http-handler:
+
 HTTPHandler
 ^^^^^^^^^^^
 
@@ -2412,6 +2497,7 @@
       just uses :func:`traceback.print_exception`. The resulting string is
       returned.
 
+.. _filter:
 
 Filter Objects
 --------------
@@ -2437,6 +2523,7 @@
       yes. If deemed appropriate, the record may be modified in-place by this
       method.
 
+.. _log-record:
 
 LogRecord Objects
 -----------------
@@ -2468,6 +2555,7 @@
       Returns the message for this :class:`LogRecord` instance after merging any
       user-supplied arguments with the message.
 
+.. _logger-adapter:
 
 LoggerAdapter Objects
 ---------------------
@@ -2553,19 +2641,57 @@
 in :mod:`logging` itself) and defining handlers which are declared either in
 :mod:`logging` or :mod:`logging.handlers`.
 
+.. function:: dictConfig(config)
+
+    Takes the logging configuration from a dictionary.  The contents of
+    this dictionary are described in :ref:`logging-config-dictschema`
+    below.
+
+    If an error is encountered during configuration, this function will
+    raise a :exc:`ValueError`, :exc:`TypeError`, :exc:`AttributeError`
+    or :exc:`ImportError` with a suitably descriptive message.  The
+    following is a (possibly incomplete) list of conditions which will
+    raise an error:
+
+    * A ``level`` which is not a string or which is a string not
+      corresponding to an actual logging level.
+    * A ``propagate`` value which is not a boolean.
+    * An id which does not have a corresponding destination.
+    * A non-existent handler id found during an incremental call.
+    * An invalid logger name.
+    * Inability to resolve to an internal or external object.
+
+    Parsing is performed by the :class:`DictConfigurator` class, whose
+    constructor is passed the dictionary used for configuration, and
+    has a :meth:`configure` method.  The :mod:`logging.config` module
+    has a callable attribute :attr:`dictConfigClass`
+    which is initially set to :class:`DictConfigurator`.
+    You can replace the value of :attr:`dictConfigClass` with a
+    suitable implementation of your own.
+
+    :func:`dictConfig` calls :attr:`dictConfigClass` passing
+    the specified dictionary, and then calls the :meth:`configure` method on
+    the returned object to put the configuration into effect::
+
+          def dictConfig(config):
+              dictConfigClass(config).configure()
+
+    For example, a subclass of :class:`DictConfigurator` could call
+    ``DictConfigurator.__init__()`` in its own :meth:`__init__()`, then
+    set up custom prefixes which would be usable in the subsequent
+    :meth:`configure` call. :attr:`dictConfigClass` would be bound to
+    this new subclass, and then :func:`dictConfig` could be called exactly as
+    in the default, uncustomized state.
 
-.. function:: fileConfig(fname, defaults=None, disable_existing_loggers=True)
+.. function:: fileConfig(fname[, defaults])
 
    Reads the logging configuration from a :mod:`configparser`\-format file named
    *fname*. This function can be called several times from an application,
-   allowing an end user the ability to select from various pre-canned
+   allowing an end user to select from various pre-canned
    configurations (if the developer provides a mechanism to present the choices
    and load the chosen configuration). Defaults to be passed to the ConfigParser
    can be specified in the *defaults* argument.
 
-   If *disable_existing_loggers* is true, any existing loggers that are not
-   children of named loggers will be disabled.
-
 
 .. function:: listen(port=DEFAULT_LOGGING_CONFIG_PORT)
 
@@ -2589,6 +2715,402 @@
    :func:`listen`.
 
 
+.. _logging-config-dictschema:
+
+Configuration dictionary schema
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Describing a logging configuration requires listing the various
+objects to create and the connections between them; for example, you
+may create a handler named "console" and then say that the logger
+named "startup" will send its messages to the "console" handler.
+These objects aren't limited to those provided by the :mod:`logging`
+module because you might write your own formatter or handler class.
+The parameters to these classes may also need to include external
+objects such as ``sys.stderr``.  The syntax for describing these
+objects and connections is defined in :ref:`logging-config-dict-connections`
+below.
+
+Dictionary Schema Details
+"""""""""""""""""""""""""
+
+The dictionary passed to :func:`dictConfig` must contain the following
+keys:
+
+* `version` - to be set to an integer value representing the schema
+  version.  The only valid value at present is 1, but having this key
+  allows the schema to evolve while still preserving backwards
+  compatibility.
+
+All other keys are optional, but if present they will be interpreted
+as described below.  In all cases below where a 'configuring dict' is
+mentioned, it will be checked for the special ``'()'`` key to see if a
+custom instantiation is required.  If so, the mechanism described in
+:ref:`logging-config-dict-userdef` below is used to create an instance;
+otherwise, the context is used to determine what to instantiate.
+
+* `formatters` - the corresponding value will be a dict in which each
+  key is a formatter id and each value is a dict describing how to
+  configure the corresponding Formatter instance.
+
+  The configuring dict is searched for keys ``format`` and ``datefmt``
+  (with defaults of ``None``) and these are used to construct a
+  :class:`logging.Formatter` instance.
+
+* `filters` - the corresponding value will be a dict in which each key
+  is a filter id and each value is a dict describing how to configure
+  the corresponding Filter instance.
+
+  The configuring dict is searched for the key ``name`` (defaulting to the
+  empty string) and this is used to construct a :class:`logging.Filter`
+  instance.
+
+* `handlers` - the corresponding value will be a dict in which each
+  key is a handler id and each value is a dict describing how to
+  configure the corresponding Handler instance.
+
+  The configuring dict is searched for the following keys:
+
+  * ``class`` (mandatory).  This is the fully qualified name of the
+    handler class.
+
+  * ``level`` (optional).  The level of the handler.
+
+  * ``formatter`` (optional).  The id of the formatter for this
+    handler.
+
+  * ``filters`` (optional).  A list of ids of the filters for this
+    handler.
+
+  All *other* keys are passed through as keyword arguments to the
+  handler's constructor.  For example, given the snippet::
+
+      handlers:
+        console:
+          class : logging.StreamHandler
+          formatter: brief
+          level   : INFO
+          filters: [allow_foo]
+          stream  : ext://sys.stdout
+        file:
+          class : logging.handlers.RotatingFileHandler
+          formatter: precise
+          filename: logconfig.log
+          maxBytes: 1024
+          backupCount: 3
+
+  the handler with id ``console`` is instantiated as a
+  :class:`logging.StreamHandler`, using ``sys.stdout`` as the underlying
+  stream.  The handler with id ``file`` is instantiated as a
+  :class:`logging.handlers.RotatingFileHandler` with the keyword arguments
+  ``filename='logconfig.log', maxBytes=1024, backupCount=3``.
+
+* `loggers` - the corresponding value will be a dict in which each key
+  is a logger name and each value is a dict describing how to
+  configure the corresponding Logger instance.
+
+  The configuring dict is searched for the following keys:
+
+  * ``level`` (optional).  The level of the logger.
+
+  * ``propagate`` (optional).  The propagation setting of the logger.
+
+  * ``filters`` (optional).  A list of ids of the filters for this
+    logger.
+
+  * ``handlers`` (optional).  A list of ids of the handlers for this
+    logger.
+
+  The specified loggers will be configured according to the level,
+  propagation, filters and handlers specified.
+
+* `root` - this will be the configuration for the root logger.
+  Processing of the configuration will be as for any logger, except
+  that the ``propagate`` setting will not be applicable.
+
+* `incremental` - whether the configuration is to be interpreted as
+  incremental to the existing configuration.  This value defaults to
+  ``False``, which means that the specified configuration replaces the
+  existing configuration with the same semantics as used by the
+  existing :func:`fileConfig` API.
+
+  If the specified value is ``True``, the configuration is processed
+  as described in the section on :ref:`logging-config-dict-incremental`.
+
+* `disable_existing_loggers` - whether any existing loggers are to be
+  disabled. This setting mirrors the parameter of the same name in
+  :func:`fileConfig`. If absent, this parameter defaults to ``True``.
+  This value is ignored if `incremental` is ``True``.
+
+.. _logging-config-dict-incremental:
+
+Incremental Configuration
+"""""""""""""""""""""""""
+
+It is difficult to provide complete flexibility for incremental
+configuration.  For example, because objects such as filters
+and formatters are anonymous, once a configuration is set up, it is
+not possible to refer to such anonymous objects when augmenting a
+configuration.
+
+Furthermore, there is not a compelling case for arbitrarily altering
+the object graph of loggers, handlers, filters, formatters at
+run-time, once a configuration is set up; the verbosity of loggers and
+handlers can be controlled just by setting levels (and, in the case of
+loggers, propagation flags).  Changing the object graph arbitrarily in
+a safe way is problematic in a multi-threaded environment; while not
+impossible, the benefits are not worth the complexity it adds to the
+implementation.
+
+Thus, when the ``incremental`` key of a configuration dict is present
+and is ``True``, the system will completely ignore any ``formatters`` and
+``filters`` entries, and process only the ``level``
+settings in the ``handlers`` entries, and the ``level`` and
+``propagate`` settings in the ``loggers`` and ``root`` entries.
+
+Using a value in the configuration dict lets configurations to be sent
+over the wire as pickled dicts to a socket listener. Thus, the logging
+verbosity of a long-running application can be altered over time with
+no need to stop and restart the application.
+
+.. _logging-config-dict-connections:
+
+Object connections
+""""""""""""""""""
+
+The schema describes a set of logging objects - loggers,
+handlers, formatters, filters - which are connected to each other in
+an object graph.  Thus, the schema needs to represent connections
+between the objects.  For example, say that, once configured, a
+particular logger has attached to it a particular handler.  For the
+purposes of this discussion, we can say that the logger represents the
+source, and the handler the destination, of a connection between the
+two.  Of course in the configured objects this is represented by the
+logger holding a reference to the handler.  In the configuration dict,
+this is done by giving each destination object an id which identifies
+it unambiguously, and then using the id in the source object's
+configuration to indicate that a connection exists between the source
+and the destination object with that id.
+
+So, for example, consider the following YAML snippet::
+
+    formatters:
+      brief:
+        # configuration for formatter with id 'brief' goes here
+      precise:
+        # configuration for formatter with id 'precise' goes here
+    handlers:
+      h1: #This is an id
+       # configuration of handler with id 'h1' goes here
+       formatter: brief
+      h2: #This is another id
+       # configuration of handler with id 'h2' goes here
+       formatter: precise
+    loggers:
+      foo.bar.baz:
+        # other configuration for logger 'foo.bar.baz'
+        handlers: [h1, h2]
+
+(Note: YAML used here because it's a little more readable than the
+equivalent Python source form for the dictionary.)
+
+The ids for loggers are the logger names which would be used
+programmatically to obtain a reference to those loggers, e.g.
+``foo.bar.baz``.  The ids for Formatters and Filters can be any string
+value (such as ``brief``, ``precise`` above) and they are transient,
+in that they are only meaningful for processing the configuration
+dictionary and used to determine connections between objects, and are
+not persisted anywhere when the configuration call is complete.
+
+The above snippet indicates that logger named ``foo.bar.baz`` should
+have two handlers attached to it, which are described by the handler
+ids ``h1`` and ``h2``. The formatter for ``h1`` is that described by id
+``brief``, and the formatter for ``h2`` is that described by id
+``precise``.
+
+
+.. _logging-config-dict-userdef:
+
+User-defined objects
+""""""""""""""""""""
+
+The schema supports user-defined objects for handlers, filters and
+formatters.  (Loggers do not need to have different types for
+different instances, so there is no support in this configuration
+schema for user-defined logger classes.)
+
+Objects to be configured are described by dictionaries
+which detail their configuration.  In some places, the logging system
+will be able to infer from the context how an object is to be
+instantiated, but when a user-defined object is to be instantiated,
+the system will not know how to do this.  In order to provide complete
+flexibility for user-defined object instantiation, the user needs
+to provide a 'factory' - a callable which is called with a
+configuration dictionary and which returns the instantiated object.
+This is signalled by an absolute import path to the factory being
+made available under the special key ``'()'``.  Here's a concrete
+example::
+
+    formatters:
+      brief:
+        format: '%(message)s'
+      default:
+        format: '%(asctime)s %(levelname)-8s %(name)-15s %(message)s'
+        datefmt: '%Y-%m-%d %H:%M:%S'
+      custom:
+          (): my.package.customFormatterFactory
+          bar: baz
+          spam: 99.9
+          answer: 42
+
+The above YAML snippet defines three formatters.  The first, with id
+``brief``, is a standard :class:`logging.Formatter` instance with the
+specified format string.  The second, with id ``default``, has a
+longer format and also defines the time format explicitly, and will
+result in a :class:`logging.Formatter` initialized with those two format
+strings.  Shown in Python source form, the ``brief`` and ``default``
+formatters have configuration sub-dictionaries::
+
+    {
+      'format' : '%(message)s'
+    }
+
+and::
+
+    {
+      'format' : '%(asctime)s %(levelname)-8s %(name)-15s %(message)s',
+      'datefmt' : '%Y-%m-%d %H:%M:%S'
+    }
+
+respectively, and as these dictionaries do not contain the special key
+``'()'``, the instantiation is inferred from the context: as a result,
+standard :class:`logging.Formatter` instances are created.  The
+configuration sub-dictionary for the third formatter, with id
+``custom``, is::
+
+  {
+    '()' : 'my.package.customFormatterFactory',
+    'bar' : 'baz',
+    'spam' : 99.9,
+    'answer' : 42
+  }
+
+and this contains the special key ``'()'``, which means that
+user-defined instantiation is wanted.  In this case, the specified
+factory callable will be used. If it is an actual callable it will be
+used directly - otherwise, if you specify a string (as in the example)
+the actual callable will be located using normal import mechanisms.
+The callable will be called with the **remaining** items in the
+configuration sub-dictionary as keyword arguments.  In the above
+example, the formatter with id ``custom`` will be assumed to be
+returned by the call::
+
+    my.package.customFormatterFactory(bar='baz', spam=99.9, answer=42)
+
+The key ``'()'`` has been used as the special key because it is not a
+valid keyword parameter name, and so will not clash with the names of
+the keyword arguments used in the call.  The ``'()'`` also serves as a
+mnemonic that the corresponding value is a callable.
+
+
+.. _logging-config-dict-externalobj:
+
+Access to external objects
+""""""""""""""""""""""""""
+
+There are times where a configuration needs to refer to objects
+external to the configuration, for example ``sys.stderr``.  If the
+configuration dict is constructed using Python code, this is
+straightforward, but a problem arises when the configuration is
+provided via a text file (e.g. JSON, YAML).  In a text file, there is
+no standard way to distinguish ``sys.stderr`` from the literal string
+``'sys.stderr'``.  To facilitate this distinction, the configuration
+system looks for certain special prefixes in string values and
+treat them specially.  For example, if the literal string
+``'ext://sys.stderr'`` is provided as a value in the configuration,
+then the ``ext://`` will be stripped off and the remainder of the
+value processed using normal import mechanisms.
+
+The handling of such prefixes is done in a way analogous to protocol
+handling: there is a generic mechanism to look for prefixes which
+match the regular expression ``^(?P<prefix>[a-z]+)://(?P<suffix>.*)$``
+whereby, if the ``prefix`` is recognised, the ``suffix`` is processed
+in a prefix-dependent manner and the result of the processing replaces
+the string value.  If the prefix is not recognised, then the string
+value will be left as-is.
+
+
+.. _logging-config-dict-internalobj:
+
+Access to internal objects
+""""""""""""""""""""""""""
+
+As well as external objects, there is sometimes also a need to refer
+to objects in the configuration.  This will be done implicitly by the
+configuration system for things that it knows about.  For example, the
+string value ``'DEBUG'`` for a ``level`` in a logger or handler will
+automatically be converted to the value ``logging.DEBUG``, and the
+``handlers``, ``filters`` and ``formatter`` entries will take an
+object id and resolve to the appropriate destination object.
+
+However, a more generic mechanism is needed for user-defined
+objects which are not known to the :mod:`logging` module.  For
+example, consider :class:`logging.handlers.MemoryHandler`, which takes
+a ``target`` argument which is another handler to delegate to. Since
+the system already knows about this class, then in the configuration,
+the given ``target`` just needs to be the object id of the relevant
+target handler, and the system will resolve to the handler from the
+id.  If, however, a user defines a ``my.package.MyHandler`` which has
+an ``alternate`` handler, the configuration system would not know that
+the ``alternate`` referred to a handler.  To cater for this, a generic
+resolution system allows the user to specify::
+
+    handlers:
+      file:
+        # configuration of file handler goes here
+
+      custom:
+        (): my.package.MyHandler
+        alternate: cfg://handlers.file
+
+The literal string ``'cfg://handlers.file'`` will be resolved in an
+analogous way to strings with the ``ext://`` prefix, but looking
+in the configuration itself rather than the import namespace.  The
+mechanism allows access by dot or by index, in a similar way to
+that provided by ``str.format``.  Thus, given the following snippet::
+
+    handlers:
+      email:
+        class: logging.handlers.SMTPHandler
+        mailhost: localhost
+        fromaddr: my_app at domain.tld
+        toaddrs:
+          - support_team at domain.tld
+          - dev_team at domain.tld
+        subject: Houston, we have a problem.
+
+in the configuration, the string ``'cfg://handlers'`` would resolve to
+the dict with key ``handlers``, the string ``'cfg://handlers.email``
+would resolve to the dict with key ``email`` in the ``handlers`` dict,
+and so on.  The string ``'cfg://handlers.email.toaddrs[1]`` would
+resolve to ``'dev_team.domain.tld'`` and the string
+``'cfg://handlers.email.toaddrs[0]'`` would resolve to the value
+``'support_team at domain.tld'``. The ``subject`` value could be accessed
+using either ``'cfg://handlers.email.subject'`` or, equivalently,
+``'cfg://handlers.email[subject]'``.  The latter form only needs to be
+used if the key contains spaces or non-alphanumeric characters.  If an
+index value consists only of decimal digits, access will be attempted
+using the corresponding integer value, falling back to the string
+value if needed.
+
+Given a string ``cfg://handlers.myhandler.mykey.123``, this will
+resolve to ``config_dict['handlers']['myhandler']['mykey']['123']``.
+If the string is specified as ``cfg://handlers.myhandler.mykey[123]``,
+the system will attempt to retrieve the value from
+``config_dict['handlers']['myhandler']['mykey'][123]``, and fall back
+to ``config_dict['handlers']['myhandler']['mykey']['123']`` if that
+fails.
+
 .. _logging-config-fileformat:
 
 Configuration file format

Modified: python/branches/py3k-cdecimal/Doc/library/math.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/math.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/math.rst	Sat Jul 17 12:36:09 2010
@@ -97,15 +97,23 @@
    <http://code.activestate.com/recipes/393090/>`_\.
 
 
+.. function:: isfinite(x)
+
+   Return ``True`` if *x* is neither an infinity nor a NaN, and
+   ``False`` otherwise.  (Note that ``0.0`` *is* considered finite.)
+
+   .. versionadded:: 3.2
+
+
 .. function:: isinf(x)
 
-   Check if the float *x* is positive or negative infinity.
+   Return ``True`` if *x* is a positive or negative infinity, and
+   ``False`` otherwise.
 
 
 .. function:: isnan(x)
 
-   Check if the float *x* is a NaN (not a number).  For more information
-   on NaNs, see the IEEE 754 standards.
+   Return ``True`` if *x* is a NaN (not a number), and ``False`` otherwise.
 
 
 .. function:: ldexp(x, i)

Modified: python/branches/py3k-cdecimal/Doc/library/multiprocessing.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/multiprocessing.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/multiprocessing.rst	Sat Jul 17 12:36:09 2010
@@ -2228,8 +2228,8 @@
 
 
 An example of how a pool of worker processes can each run a
-:class:`SimpleHTTPServer.HttpServer` instance while sharing a single listening
-socket.
+:class:`~http.server.SimpleHTTPRequestHandler` instance while sharing a single
+listening socket.
 
 .. literalinclude:: ../includes/mp_webserver.py
 

Modified: python/branches/py3k-cdecimal/Doc/library/os.path.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/os.path.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/os.path.rst	Sat Jul 17 12:36:09 2010
@@ -201,6 +201,7 @@
    Normalize the case of a pathname.  On Unix and Mac OS X, this returns the
    path unchanged; on case-insensitive filesystems, it converts the path to
    lowercase.  On Windows, it also converts forward slashes to backward slashes.
+   Raise a TypeError if the type of *path* is not ``str`` or ``bytes``.
 
 
 .. function:: normpath(path)
@@ -230,11 +231,18 @@
 
 .. function:: samefile(path1, path2)
 
-   Return ``True`` if both pathname arguments refer to the same file or directory
-   (as indicated by device number and i-node number). Raise an exception if a
-   :func:`os.stat` call on either pathname fails.
+   Return ``True`` if both pathname arguments refer to the same file or directory.
+   On Unix, this is determined by the device number and i-node number and raises an
+   exception if a :func:`os.stat` call on either pathname fails.
+
+   On Windows, two files are the same if they resolve to the same final path
+   name using the Windows API call GetFinalPathNameByHandle. This function
+   raises an exception if handles cannot be obtained to either file.
 
-   Availability: Unix.
+   Availability: Windows, Unix.
+
+   .. versionchanged:: 3.2
+      Added Windows support.
 
 
 .. function:: sameopenfile(fp1, fp2)

Modified: python/branches/py3k-cdecimal/Doc/library/os.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/os.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/os.rst	Sat Jul 17 12:36:09 2010
@@ -723,7 +723,7 @@
 
       This function is intended for low-level I/O.  For normal usage, use the
       built-in function :func:`open`, which returns a "file object" with
-      :meth:`~file.read` and :meth:`~file.wprite` methods (and many more).  To
+      :meth:`~file.read` and :meth:`~file.write` methods (and many more).  To
       wrap a file descriptor in a "file object", use :func:`fdopen`.
 
 
@@ -1064,8 +1064,10 @@
 .. function:: lstat(path)
 
    Like :func:`stat`, but do not follow symbolic links.  This is an alias for
-   :func:`stat` on platforms that do not support symbolic links, such as
-   Windows.
+   :func:`stat` on platforms that do not support symbolic links.
+
+   .. versionchanged:: 3.2
+      Added support for Windows 6.0 (Vista) symbolic links.
 
 
 .. function:: mkfifo(path[, mode])
@@ -1115,7 +1117,8 @@
 
    Create a directory named *path* with numeric mode *mode*. The default *mode*
    is ``0o777`` (octal).  On some systems, *mode* is ignored.  Where it is used,
-   the current umask value is first masked out.
+   the current umask value is first masked out.  If the directory already
+   exists, :exc:`OSError` is raised.
 
    It is also possible to create temporary directories; see the
    :mod:`tempfile` module's :func:`tempfile.mkdtemp` function.
@@ -1180,7 +1183,10 @@
    and the call may raise an UnicodeDecodeError. If the *path* is a bytes
    object, the result will be a bytes object.
 
-   Availability: Unix.
+   Availability: Unix, Windows
+
+   .. versionchanged:: 3.2
+      Added support for Windows 6.0 (Vista) symbolic links.
 
 
 .. function:: remove(path)
@@ -1342,7 +1348,27 @@
 
    Create a symbolic link pointing to *source* named *link_name*.
 
-   Availability: Unix.
+   On Windows, symlink version takes an additional, optional parameter,
+   *target_is_directory*, which defaults to False.
+
+   symlink(source, link_name, target_is_directory=False)
+
+   On Windows, a symlink represents a file or a directory, and does not
+   morph to the target dynamically.  For this reason, when creating a
+   symlink on Windows, if the target is not already present, the symlink
+   will default to being a file symlink.  If *target_is_directory* is set to
+   True, the symlink will be created as a directory symlink.  This
+   parameter is ignored if the target exists (and the symlink is created
+   with the same type as the target).
+
+   Symbolic link support was introduced in Windows 6.0 (Vista). *symlink*
+   will raise a NotImplementedError on Windows versions earlier than 6.0. The
+   SeCreateSymbolicLinkPrivilege is required in order to create symlinks.
+
+   Availability:  Unix, Windows
+
+   .. versionchanged:: 3.2
+      Added support for Windows 6.0 (Vista) symbolic links.
 
 
 .. function:: unlink(path)

Modified: python/branches/py3k-cdecimal/Doc/library/pdb.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/pdb.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/pdb.rst	Sat Jul 17 12:36:09 2010
@@ -210,11 +210,13 @@
    Print a stack trace, with the most recent frame at the bottom.  An arrow
    indicates the current frame, which determines the context of most commands.
 
-d(own)
-   Move the current frame one level down in the stack trace (to a newer frame).
-
-u(p)
-   Move the current frame one level up in the stack trace (to an older frame).
+d(own) [*count*]
+   Move the current frame *count* (default one) levels down in the stack trace
+   (to a newer frame).
+
+u(p) [*count*]
+   Move the current frame *count* (default one) levels up in the stack trace
+   (to an older frame).
 
 b(reak) [[*filename*:]\ *lineno* | *function*\ [, *condition*]]
    With a *lineno* argument, set a break there in the current file.  With a

Modified: python/branches/py3k-cdecimal/Doc/library/re.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/re.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/re.rst	Sat Jul 17 12:36:09 2010
@@ -692,6 +692,11 @@
    metacharacters in it.
 
 
+.. function:: purge()
+
+   Clear the regular expression cache.
+
+
 .. exception:: error
 
    Exception raised when a string passed to one of the functions here is not a
@@ -709,18 +714,12 @@
 
    The :class:`RegexObject` class supports the following methods and attributes:
 
+   .. method:: RegexObject.search(string[, pos[, endpos]])
 
-   .. method:: RegexObject.match(string[, pos[, endpos]])
-
-      If zero or more characters at the beginning of *string* match this regular
-      expression, return a corresponding :class:`MatchObject` instance.  Return
-      ``None`` if the string does not match the pattern; note that this is different
-      from a zero-length match.
-
-      .. note::
-
-         If you want to locate a match anywhere in *string*, use
-         :meth:`~RegexObject.search` instead.
+      Scan through *string* looking for a location where this regular expression
+      produces a match, and return a corresponding :class:`MatchObject` instance.
+      Return ``None`` if no position in the string matches the pattern; note that this
+      is different from finding a zero-length match at some point in the string.
 
       The optional second parameter *pos* gives an index in the string where the
       search is to start; it defaults to ``0``.  This is not completely equivalent to
@@ -732,24 +731,34 @@
       will be as if the string is *endpos* characters long, so only the characters
       from *pos* to ``endpos - 1`` will be searched for a match.  If *endpos* is less
       than *pos*, no match will be found, otherwise, if *rx* is a compiled regular
-      expression object, ``rx.match(string, 0, 50)`` is equivalent to
-      ``rx.match(string[:50], 0)``.
+      expression object, ``rx.search(string, 0, 50)`` is equivalent to
+      ``rx.search(string[:50], 0)``.
 
-         >>> pattern = re.compile("o")
-         >>> pattern.match("dog")      # No match as "o" is not at the start of "dog."
-         >>> pattern.match("dog", 1)   # Match as "o" is the 2nd character of "dog".
-         <_sre.SRE_Match object at ...>
+      >>> pattern = re.compile("d")
+      >>> pattern.search("dog")     # Match at index 0
+      <_sre.SRE_Match object at ...>
+      >>> pattern.search("dog", 1)  # No match; search doesn't include the "d"
 
 
-   .. method:: RegexObject.search(string[, pos[, endpos]])
+   .. method:: RegexObject.match(string[, pos[, endpos]])
 
-      Scan through *string* looking for a location where this regular expression
-      produces a match, and return a corresponding :class:`MatchObject` instance.
-      Return ``None`` if no position in the string matches the pattern; note that this
-      is different from finding a zero-length match at some point in the string.
+      If zero or more characters at the *beginning* of *string* match this regular
+      expression, return a corresponding :class:`MatchObject` instance.  Return
+      ``None`` if the string does not match the pattern; note that this is different
+      from a zero-length match.
 
       The optional *pos* and *endpos* parameters have the same meaning as for the
-      :meth:`~RegexObject.match` method.
+      :meth:`~RegexObject.search` method.
+
+      .. note::
+
+         If you want to locate a match anywhere in *string*, use
+         :meth:`~RegexObject.search` instead.
+
+      >>> pattern = re.compile("o")
+      >>> pattern.match("dog")      # No match as "o" is not at the start of "dog".
+      >>> pattern.match("dog", 1)   # Match as "o" is the 2nd character of "dog".
+      <_sre.SRE_Match object at ...>
 
 
    .. method:: RegexObject.split(string[, maxsplit=0])
@@ -759,12 +768,16 @@
 
    .. method:: RegexObject.findall(string[, pos[, endpos]])
 
-      Identical to the :func:`findall` function, using the compiled pattern.
+      Similar to the :func:`findall` function, using the compiled pattern, but
+      also accepts optional *pos* and *endpos* parameters that limit the search
+      region like for :meth:`match`.
 
 
    .. method:: RegexObject.finditer(string[, pos[, endpos]])
 
-      Identical to the :func:`finditer` function, using the compiled pattern.
+      Similar to the :func:`finditer` function, using the compiled pattern, but
+      also accepts optional *pos* and *endpos* parameters that limit the search
+      region like for :meth:`match`.
 
 
    .. method:: RegexObject.sub(repl, string[, count=0])
@@ -1217,9 +1230,9 @@
    ...   random.shuffle(inner_word)
    ...   return m.group(1) + "".join(inner_word) + m.group(3)
    >>> text = "Professor Abdolmalek, please report your absences promptly."
-   >>> re.sub("(\w)(\w+)(\w)", repl, text)
+   >>> re.sub(r"(\w)(\w+)(\w)", repl, text)
    'Poefsrosr Aealmlobdk, pslaee reorpt your abnseces plmrptoy.'
-   >>> re.sub("(\w)(\w+)(\w)", repl, text)
+   >>> re.sub(r"(\w)(\w+)(\w)", repl, text)
    'Pofsroser Aodlambelk, plasee reoprt yuor asnebces potlmrpy.'
 
 

Modified: python/branches/py3k-cdecimal/Doc/library/select.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/select.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/select.rst	Sat Jul 17 12:36:09 2010
@@ -41,14 +41,14 @@
 
 .. function:: kqueue()
 
-   (Only supported on BSD.)  Returns a kernel queue object object; see section
+   (Only supported on BSD.)  Returns a kernel queue object; see section
    :ref:`kqueue-objects` below for the methods supported by kqueue objects.
 
 
 .. function:: kevent(ident, filter=KQ_FILTER_READ, flags=KQ_EV_ADD, fflags=0, data=0, udata=0)
 
-   (Only supported on BSD.)  Returns a kernel event object object; see section
-   :ref:`kevent-objects` below for the methods supported by kqueue objects.
+   (Only supported on BSD.)  Returns a kernel event object; see section
+   :ref:`kevent-objects` below for the methods supported by kevent objects.
 
 
 .. function:: select(rlist, wlist, xlist[, timeout])

Modified: python/branches/py3k-cdecimal/Doc/library/socket.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/socket.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/socket.rst	Sat Jul 17 12:36:09 2010
@@ -71,18 +71,21 @@
 tuple, and the fields depend on the address type. The general tuple form is
 ``(addr_type, v1, v2, v3 [, scope])``, where:
 
-   - *addr_type* is one of TIPC_ADDR_NAMESEQ, TIPC_ADDR_NAME, or
-     TIPC_ADDR_ID.
-   - *scope* is one of TIPC_ZONE_SCOPE, TIPC_CLUSTER_SCOPE, and
-     TIPC_NODE_SCOPE.
-   - If *addr_type* is TIPC_ADDR_NAME, then *v1* is the server type, *v2* is
-     the port identifier, and *v3* should be 0.
+- *addr_type* is one of TIPC_ADDR_NAMESEQ, TIPC_ADDR_NAME, or
+  TIPC_ADDR_ID.
+- *scope* is one of TIPC_ZONE_SCOPE, TIPC_CLUSTER_SCOPE, and
+  TIPC_NODE_SCOPE.
+- If *addr_type* is TIPC_ADDR_NAME, then *v1* is the server type, *v2* is
+  the port identifier, and *v3* should be 0.
 
-     If *addr_type* is TIPC_ADDR_NAMESEQ, then *v1* is the server type, *v2*
-     is the lower port number, and *v3* is the upper port number.
+  If *addr_type* is TIPC_ADDR_NAMESEQ, then *v1* is the server type, *v2*
+  is the lower port number, and *v3* is the upper port number.
 
-     If *addr_type* is TIPC_ADDR_ID, then *v1* is the node, *v2* is the
-     reference, and *v3* should be set to 0.
+  If *addr_type* is TIPC_ADDR_ID, then *v1* is the node, *v2* is the
+  reference, and *v3* should be set to 0.
+
+  If *addr_type* is TIPC_ADDR_ID, then *v1* is the node, *v2* is the
+  reference, and *v3* should be set to 0.
 
 
 All errors raise exceptions.  The normal exceptions for invalid argument types
@@ -690,7 +693,7 @@
 
    Set a timeout on blocking socket operations.  The *value* argument can be a
    nonnegative float expressing seconds, or ``None``. If a float is given,
-   subsequent socket operations will raise an :exc:`timeout` exception if the
+   subsequent socket operations will raise a :exc:`timeout` exception if the
    timeout period *value* has elapsed before the operation has completed.  Setting
    a timeout of ``None`` disables timeouts on socket operations.
    ``s.settimeout(0.0)`` is equivalent to ``s.setblocking(0)``;
@@ -745,7 +748,9 @@
    Shut down one or both halves of the connection.  If *how* is :const:`SHUT_RD`,
    further receives are disallowed.  If *how* is :const:`SHUT_WR`, further sends
    are disallowed.  If *how* is :const:`SHUT_RDWR`, further sends and receives are
-   disallowed.
+   disallowed.  Depending on the platform, shutting down one half of the connection
+   can also close the opposite half (e.g. on Mac OS X, ``shutdown(SHUT_WR)`` does
+   not allow further reads on the other end of the connection).
 
 Note that there are no methods :meth:`read` or :meth:`write`; use
 :meth:`~socket.recv` and :meth:`~socket.send` without *flags* argument instead.

Modified: python/branches/py3k-cdecimal/Doc/library/sqlite3.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/sqlite3.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/sqlite3.rst	Sat Jul 17 12:36:09 2010
@@ -136,7 +136,7 @@
    first blank for the column name: the column name would simply be "x".
 
 
-.. function:: connect(database[, timeout, isolation_level, detect_types, factory])
+.. function:: connect(database[, timeout, detect_types, isolation_level, check_same_thread, factory, cached_statements])
 
    Opens a connection to the SQLite database file *database*. You can use
    ``":memory:"`` to open a database connection to a database that resides in RAM

Modified: python/branches/py3k-cdecimal/Doc/library/ssl.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/ssl.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/ssl.rst	Sat Jul 17 12:36:09 2010
@@ -110,13 +110,16 @@
 
    .. note::
 
-      This information varies depending on the version of OpenSSL.
-      For instance, in some older versions of OpenSSL (such as 0.9.7l on
-      OS X 10.4), an SSLv2 client could not connect to an SSLv23 server.
-      Conversely, starting from 1.0.0, an SSLv23 client will actually
-      try the SSLv3 protocol unless you explicitly enable SSLv2 ciphers.
+      Which connections succeed will vary depending on the version of
+      OpenSSL.  For instance, in some older versions of OpenSSL (such
+      as 0.9.7l on OS X 10.4), an SSLv2 client could not connect to an
+      SSLv23 server.  Another example: beginning with OpenSSL 1.0.0,
+      an SSLv23 client will not actually attempt SSLv2 connections
+      unless you explicitly enable SSLv2 ciphers; for example, you
+      might specify ``"ALL"`` or ``"SSLv2"`` as the *ciphers* parameter
+      to enable them.
 
-   The parameter ``ciphers`` sets the available ciphers for this SSL object.
+   The *ciphers* parameter sets the available ciphers for this SSL object.
    It should be a string in the `OpenSSL cipher list format
    <http://www.openssl.org/docs/apps/ciphers.html#CIPHER_LIST_FORMAT>`_.
 

Modified: python/branches/py3k-cdecimal/Doc/library/stdtypes.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/stdtypes.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/stdtypes.rst	Sat Jul 17 12:36:09 2010
@@ -452,7 +452,7 @@
     Equivalent to::
 
         def bit_length(self):
-            s = bin(x)          # binary representation:  bin(-37) --> '-0b100101'
+            s = bin(self)       # binary representation:  bin(-37) --> '-0b100101'
             s = s.lstrip('-0b') # remove leading zeros and minus sign
             return len(s)       # len('100101') --> 6
 
@@ -966,7 +966,8 @@
 
 .. method:: str.capitalize()
 
-   Return a copy of the string with only its first character capitalized.
+   Return a copy of the string with its first character capitalized and the
+   rest lowercased.
 
 
 .. method:: str.center(width[, fillchar])
@@ -1329,6 +1330,10 @@
    You can use :meth:`str.maketrans` to create a translation map from
    character-to-character mappings in different formats.
 
+   You can use the :func:`~string.maketrans` helper function in the :mod:`string`
+   module to create a translation table. For string objects, set the *table*
+   argument to ``None`` for translations that only delete characters:
+
    .. note::
 
       An even more flexible approach is to create a custom character mapping
@@ -2259,10 +2264,19 @@
    buffer protocol.  Builtin objects that support the buffer protocol include
    :class:`bytes` and :class:`bytearray`.
 
-   ``len(view)`` returns the total number of bytes in the memoryview, *view*.
+   A :class:`memoryview` has the notion of an *element*, which is the
+   atomic memory unit handled by the originating object *obj*.  For many
+   simple types such as :class:`bytes` and :class:`bytearray`, an element
+   is a single byte, but other types such as :class:`array.array` may have
+   bigger elements.
+
+   ``len(view)`` returns the total number of elements in the memoryview,
+   *view*.  The :class:`~memoryview.itemsize` attribute will give you the
+   number of bytes in a single element.
 
    A :class:`memoryview` supports slicing to expose its data.  Taking a single
-   index will return a single byte.  Full slicing will result in a subview::
+   index will return a single element as a :class:`bytes` object.  Full
+   slicing will result in a subview::
 
       >>> v = memoryview(b'abcefg')
       >>> v[1]
@@ -2273,11 +2287,8 @@
       <memory at 0x77ab28>
       >>> bytes(v[1:4])
       b'bce'
-      >>> v[3:-1]
-      <memory at 0x744f18>
-      >>> bytes(v[4:-1])
 
-   If the object the memory view is over supports changing its data, the
+   If the object the memoryview is over supports changing its data, the
    memoryview supports slice assignment::
 
       >>> data = bytearray(b'abcefg')
@@ -2295,14 +2306,20 @@
       File "<stdin>", line 1, in <module>
       ValueError: cannot modify size of memoryview object
 
-   Notice how the size of the memoryview object can not be changed.
-
+   Notice how the size of the memoryview object cannot be changed.
 
    :class:`memoryview` has two methods:
 
    .. method:: tobytes()
 
-      Return the data in the buffer as a bytestring.
+      Return the data in the buffer as a bytestring.  This is equivalent to
+      calling the :class:`bytes` constructor on the memoryview. ::
+
+         >>> m = memoryview(b"abc")
+         >>> m.tobytes()
+         b'abc'
+         >>> bytes(m)
+         b'abc'
 
    .. method:: tolist()
 
@@ -2320,7 +2337,15 @@
 
    .. attribute:: itemsize
 
-      The size in bytes of each element of the memoryview.
+      The size in bytes of each element of the memoryview::
+
+         >>> m = memoryview(array.array('H', [1,2,3]))
+         >>> m.itemsize
+         2
+         >>> m[0]
+         b'\x01\x00'
+         >>> len(m[0]) == m.itemsize
+         True
 
    .. attribute:: shape
 

Modified: python/branches/py3k-cdecimal/Doc/library/string.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/string.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/string.rst	Sat Jul 17 12:36:09 2010
@@ -174,7 +174,7 @@
    .. method:: convert_field(value, conversion)
 
       Converts the value (returned by :meth:`get_field`) given a conversion type
-      (as in the tuple returned by the :meth:`parse` method.)  The default
+      (as in the tuple returned by the :meth:`parse` method).  The default
       version understands 'r' (repr) and 's' (str) conversion types.
 
 
@@ -185,7 +185,7 @@
 
 The :meth:`str.format` method and the :class:`Formatter` class share the same
 syntax for format strings (although in the case of :class:`Formatter`,
-subclasses can define their own format string syntax.)
+subclasses can define their own format string syntax).
 
 Format strings contain "replacement fields" surrounded by curly braces ``{}``.
 Anything that is not contained in braces is considered literal text, which is
@@ -197,7 +197,7 @@
    .. productionlist:: sf
       replacement_field: "{" [`field_name`] ["!" `conversion`] [":" `format_spec`] "}"
       field_name: arg_name ("." `attribute_name` | "[" `element_index` "]")*
-      arg_name: (`identifier` | `integer`)?
+      arg_name: [`identifier` | `integer`]
       attribute_name: `identifier`
       element_index: `integer` | `index_string`
       index_string: <any source character except "]"> +
@@ -211,6 +211,8 @@
 preceded by an exclamation point ``'!'``, and a *format_spec*, which is preceded
 by a colon ``':'``.  These specify a non-default format for the replacement value.
 
+See also the :ref:`formatspec` section.
+
 The *field_name* itself begins with an *arg_name* that is either either a number or a
 keyword.  If it's a number, it refers to a positional argument, and if it's a keyword,
 it refers to a named keyword argument.  If the numerical arg_names in a format string
@@ -221,6 +223,10 @@
 attribute using :func:`getattr`, while an expression of the form ``'[index]'``
 does an index lookup using :func:`__getitem__`.
 
+.. versionchanged:: 3.1
+   The positional argument specifiers can be omitted, so ``'{} {}'`` is
+   equivalent to ``'{0} {1}'``.
+
 Some simple format string examples::
 
    "First, thou shalt count to {0}" # References first positional argument
@@ -261,26 +267,7 @@
 format_spec are substituted before the *format_spec* string is interpreted.
 This allows the formatting of a value to be dynamically specified.
 
-For example, suppose you wanted to have a replacement field whose field width is
-determined by another variable::
-
-   "A man with two {0:{1}}".format("noses", 10)
-
-This would first evaluate the inner replacement field, making the format string
-effectively::
-
-   "A man with two {0:10}"
-
-Then the outer replacement field would be evaluated, producing::
-
-   "noses     "
-
-Which is substituted into the string, yielding::
-
-   "A man with two noses     "
-
-(The extra space is because we specified a field width of 10, and because left
-alignment is the default for strings.)
+See the :ref:`formatexamples` section for some examples.
 
 
 .. _formatspec:
@@ -290,7 +277,7 @@
 
 "Format specifications" are used within replacement fields contained within a
 format string to define how individual values are presented (see
-:ref:`formatstrings`.)  They can also be passed directly to the built-in
+:ref:`formatstrings`).  They can also be passed directly to the built-in
 :func:`format` function.  Each formattable type may define how the format
 specification is to be interpreted.
 
@@ -324,7 +311,7 @@
    | Option  | Meaning                                                  |
    +=========+==========================================================+
    | ``'<'`` | Forces the field to be left-aligned within the available |
-   |         | space (This is the default.)                             |
+   |         | space (this is the default).                             |
    +---------+----------------------------------------------------------+
    | ``'>'`` | Forces the field to be right-aligned within the          |
    |         | available space.                                         |
@@ -366,6 +353,9 @@
 For a locale aware separator, use the ``'n'`` integer presentation type
 instead.
 
+.. versionchanged:: 3.1
+   Added the ``','`` option (see also :pep:`378`).
+
 *width* is a decimal integer defining the minimum field width.  If not
 specified, then the field width will be determined by the content.
 
@@ -484,6 +474,147 @@
    +---------+----------------------------------------------------------+
 
 
+.. _formatexamples:
+
+Format examples
+^^^^^^^^^^^^^^^
+
+This section contains examples of the new format syntax and comparison with
+the old ``%``-formatting.
+
+In most of the cases the syntax is similar to the old ``%``-formatting, with the
+addition of the ``{}`` and with ``:`` used instead of ``%``.
+For example, ``'%03.2f'`` can be translated to ``'{:03.2f}'``.
+
+The new format syntax also supports new and different options, shown in the
+follow examples.
+
+Accessing arguments by position::
+
+   >>> '{0}, {1}, {2}'.format('a', 'b', 'c')
+   'a, b, c'
+   >>> '{}, {}, {}'.format('a', 'b', 'c')  # 3.1+ only
+   'a, b, c'
+   >>> '{2}, {1}, {0}'.format('a', 'b', 'c')
+   'c, b, a'
+   >>> '{2}, {1}, {0}'.format(*'abc')      # unpacking argument sequence
+   'c, b, a'
+   >>> '{0}{1}{0}'.format('abra', 'cad')   # arguments' indices can be repeated
+   'abracadabra'
+
+Accessing arguments by name::
+
+   >>> 'Coordinates: {latitude}, {longitude}'.format(latitude='37.24N', longitude='-115.81W')
+   'Coordinates: 37.24N, -115.81W'
+   >>> coord = {'latitude': '37.24N', 'longitude': '-115.81W'}
+   >>> 'Coordinates: {latitude}, {longitude}'.format(**coord)
+   'Coordinates: 37.24N, -115.81W'
+
+Accessing arguments' attributes::
+
+   >>> c = 3-5j
+   >>> ('The complex number {0} is formed from the real part {0.real} '
+   ...  'and the imaginary part {0.imag}.').format(c)
+   'The complex number (3-5j) is formed from the real part 3.0 and the imaginary part -5.0.'
+   >>> class Point:
+   ...     def __init__(self, x, y):
+   ...         self.x, self.y = x, y
+   ...     def __str__(self):
+   ...         return 'Point({self.x}, {self.y})'.format(self=self)
+   ...
+   >>> str(Point(4, 2))
+   'Point(4, 2)'
+
+Accessing arguments' items::
+
+   >>> coord = (3, 5)
+   >>> 'X: {0[0]};  Y: {0[1]}'.format(coord)
+   'X: 3;  Y: 5'
+
+Replacing ``%s`` and ``%r``::
+
+   >>> "repr() shows quotes: {!r}; str() doesn't: {!s}".format('test1', 'test2')
+   "repr() shows quotes: 'test1'; str() doesn't: test2"
+
+Aligning the text and specifying a width::
+
+   >>> '{:<30}'.format('left aligned')
+   'left aligned                  '
+   >>> '{:>30}'.format('right aligned')
+   '                 right aligned'
+   >>> '{:^30}'.format('centered')
+   '           centered           '
+   >>> '{:*^30}'.format('centered')  # use '*' as a fill char
+   '***********centered***********'
+
+Replacing ``%+f``, ``%-f``, and ``% f`` and specifying a sign::
+
+   >>> '{:+f}; {:+f}'.format(3.14, -3.14)  # show it always
+   '+3.140000; -3.140000'
+   >>> '{: f}; {: f}'.format(3.14, -3.14)  # show a space for positive numbers
+   ' 3.140000; -3.140000'
+   >>> '{:-f}; {:-f}'.format(3.14, -3.14)  # show only the minus -- same as '{:f}; {:f}'
+   '3.140000; -3.140000'
+
+Replacing ``%x`` and ``%o`` and converting the value to different bases::
+
+   >>> # format also supports binary numbers
+   >>> "int: {0:d};  hex: {0:x};  oct: {0:o};  bin: {0:b}".format(42)
+   'int: 42;  hex: 2a;  oct: 52;  bin: 101010'
+   >>> # with 0x, 0o, or 0b as prefix:
+   >>> "int: {0:d};  hex: {0:#x};  oct: {0:#o};  bin: {0:#b}".format(42)
+   'int: 42;  hex: 0x2a;  oct: 0o52;  bin: 0b101010'
+
+Using the comma as a thousands separator::
+
+   >>> '{:,}'.format(1234567890)
+   '1,234,567,890'
+
+Expressing a percentage::
+
+   >>> points = 19
+   >>> total = 22
+   >>> 'Correct answers: {:.2%}.'.format(points/total)
+   'Correct answers: 86.36%'
+
+Using type-specific formatting::
+
+   >>> import datetime
+   >>> d = datetime.datetime(2010, 7, 4, 12, 15, 58)
+   >>> '{:%Y-%m-%d %H:%M:%S}'.format(d)
+   '2010-07-04 12:15:58'
+
+Nesting arguments and more complex examples::
+
+   >>> for align, text in zip('<^>', ['left', 'center', 'right']):
+   ...     '{0:{align}{fill}16}'.format(text, fill=align, align=align)
+   ...
+   'left<<<<<<<<<<<<'
+   '^^^^^center^^^^^'
+   '>>>>>>>>>>>right'
+   >>>
+   >>> octets = [192, 168, 0, 1]
+   >>> '{:02X}{:02X}{:02X}{:02X}'.format(*octets)
+   'C0A80001'
+   >>> int(_, 16)
+   3232235521
+   >>>
+   >>> width = 5
+   >>> for num in range(5,12):
+   ...     for base in 'dXob':
+   ...         print('{0:{width}{base}}'.format(num, base=base, width=width), end=' ')
+   ...     print()
+   ...
+       5     5     5   101
+       6     6     6   110
+       7     7     7   111
+       8     8    10  1000
+       9     9    11  1001
+      10     A    12  1010
+      11     B    13  1011
+
+
+
 .. _template-strings:
 
 Template strings

Modified: python/branches/py3k-cdecimal/Doc/library/struct.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/struct.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/struct.rst	Sat Jul 17 12:36:09 2010
@@ -20,9 +20,9 @@
    order to maintain proper alignment for the C types involved; similarly,
    alignment is taken into account when unpacking.  This behavior is chosen so
    that the bytes of a packed struct correspond exactly to the layout in memory
-   of the corresponding C struct.  To omit pad bytes, use `standard` size and
-   alignment instead of `native` size and alignment: see :ref:`struct-alignment`
-   for details.
+   of the corresponding C struct.  To handle platform-independent data formats
+   or omit implicit pad bytes, use `standard` size and alignment instead of
+   `native` size and alignment: see :ref:`struct-alignment` for details.
 
 Functions and Exceptions
 ------------------------
@@ -95,19 +95,19 @@
 the byte order, size and alignment of the packed data, according to the
 following table:
 
-+-----------+------------------------+--------------------+
-| Character | Byte order             | Size and alignment |
-+===========+========================+====================+
-| ``@``     | native                 | native             |
-+-----------+------------------------+--------------------+
-| ``=``     | native                 | standard           |
-+-----------+------------------------+--------------------+
-| ``<``     | little-endian          | standard           |
-+-----------+------------------------+--------------------+
-| ``>``     | big-endian             | standard           |
-+-----------+------------------------+--------------------+
-| ``!``     | network (= big-endian) | standard           |
-+-----------+------------------------+--------------------+
++-----------+------------------------+----------+-----------+
+| Character | Byte order             | Size     | Alignment |
++===========+========================+==========+===========+
+| ``@``     | native                 | native   | native    |
++-----------+------------------------+----------+-----------+
+| ``=``     | native                 | standard | none      |
++-----------+------------------------+----------+-----------+
+| ``<``     | little-endian          | standard | none      |
++-----------+------------------------+----------+-----------+
+| ``>``     | big-endian             | standard | none      |
++-----------+------------------------+----------+-----------+
+| ``!``     | network (= big-endian) | standard | none      |
++-----------+------------------------+----------+-----------+
 
 If the first character is not one of these, ``'@'`` is assumed.
 
@@ -120,11 +120,8 @@
 Native size and alignment are determined using the C compiler's
 ``sizeof`` expression.  This is always combined with native byte order.
 
-Standard size and alignment are as follows: no alignment is required for any
-type (so you have to use pad bytes); :ctype:`short` is 2 bytes; :ctype:`int` and
-:ctype:`long` are 4 bytes; :ctype:`long long` (:ctype:`__int64` on Windows) is 8
-bytes; :ctype:`float` and :ctype:`double` are 32-bit and 64-bit IEEE floating
-point numbers, respectively. :ctype:`_Bool` is 1 byte.
+Standard size depends only on the format character;  see the table in
+the :ref:`format-characters` section.
 
 Note the difference between ``'@'`` and ``'='``: both use native byte order, but
 the size and alignment of the latter is standardized.
@@ -135,12 +132,6 @@
 There is no way to indicate non-native byte order (force byte-swapping); use the
 appropriate choice of ``'<'`` or ``'>'``.
 
-The ``'P'`` format character is only available for the native byte ordering
-(selected as the default or with the ``'@'`` byte order character). The byte
-order character ``'='`` chooses to use little- or big-endian ordering based on
-the host system. The struct module does not interpret this as native ordering,
-so the ``'P'`` format is not available.
-
 Notes:
 
 (1) Padding is only automatically added between successive structure members.
@@ -160,7 +151,11 @@
 ^^^^^^^^^^^^^^^^^
 
 Format characters have the following meaning; the conversion between C and
-Python values should be obvious given their types:
+Python values should be obvious given their types.  The 'Standard size' column
+refers to the size of the packed value in bytes when using standard size; that
+is, when the format string starts with one of ``'<'``, ``'>'``, ``'!'`` or
+``'='``.  When using native size, the size of the packed value is
+platform-dependent.
 
 +--------+-------------------------+--------------------+----------------+------------+
 | Format | C Type                  | Python type        | Standard size  | Notes      |
@@ -192,15 +187,15 @@
 | ``Q``  | :ctype:`unsigned long   | integer            | 8              | \(3), \(4) |
 |        | long`                   |                    |                |            |
 +--------+-------------------------+--------------------+----------------+------------+
-| ``f``  | :ctype:`float`          | float              | 4              |            |
+| ``f``  | :ctype:`float`          | float              | 4              | \(5)       |
 +--------+-------------------------+--------------------+----------------+------------+
-| ``d``  | :ctype:`double`         | float              | 8              |            |
+| ``d``  | :ctype:`double`         | float              | 8              | \(5)       |
 +--------+-------------------------+--------------------+----------------+------------+
 | ``s``  | :ctype:`char[]`         | bytes              |                | \(1)       |
 +--------+-------------------------+--------------------+----------------+------------+
 | ``p``  | :ctype:`char[]`         | bytes              |                | \(1)       |
 +--------+-------------------------+--------------------+----------------+------------+
-| ``P``  | :ctype:`void \*`        | integer            |                |            |
+| ``P``  | :ctype:`void \*`        | integer            |                | \(6)       |
 +--------+-------------------------+--------------------+----------------+------------+
 
 Notes:
@@ -228,6 +223,18 @@
    .. versionchanged:: 3.2
       Use of the :meth:`__index__` method for non-integers is new in 3.2.
 
+(5)
+   For the ``'f'`` and ``'d'`` conversion codes, the packed representation uses
+   the IEEE 754 binary32 (for ``'f'``) or binary64 (for ``'d'``) format,
+   regardless of the floating-point format used by the platform.
+
+(6)
+   The ``'P'`` format character is only available for the native byte ordering
+   (selected as the default or with the ``'@'`` byte order character). The byte
+   order character ``'='`` chooses to use little- or big-endian ordering based
+   on the host system. The struct module does not interpret this as native
+   ordering, so the ``'P'`` format is not available.
+
 
 A format character may be preceded by an integral repeat count.  For example,
 the format string ``'4h'`` means exactly the same as ``'hhhh'``.
@@ -252,18 +259,16 @@
    In 3.0, some of the integer formats wrapped out-of-range values and
    raised :exc:`DeprecationWarning` instead of :exc:`struct.error`.
 
-
 The ``'p'`` format character encodes a "Pascal string", meaning a short
-variable-length string stored in a fixed number of bytes. The count is the total
-number of bytes stored.  The first byte stored is the length of the string, or
-255, whichever is smaller.  The bytes of the string follow.  If the string
-passed in to :func:`pack` is too long (longer than the count minus 1), only the
-leading count-1 bytes of the string are stored.  If the string is shorter than
-count-1, it is padded with null bytes so that exactly count bytes in all are
-used.  Note that for :func:`unpack`, the ``'p'`` format character consumes count
-bytes, but that the string returned can never contain more than 255 bytes.
-
-
+variable-length string stored in a *fixed number of bytes*, given by the count.
+The first byte stored is the length of the string, or 255, whichever is
+smaller.  The bytes of the string follow.  If the string passed in to
+:func:`pack` is too long (longer than the count minus 1), only the leading
+``count-1`` bytes of the string are stored.  If the string is shorter than
+``count-1``, it is padded with null bytes so that exactly count bytes in all
+are used.  Note that for :func:`unpack`, the ``'p'`` format character consumes
+``count`` bytes, but that the string returned can never contain more than 255
+bytes.
 
 For the ``'?'`` format character, the return value is either :const:`True` or
 :const:`False`. When packing, the truth value of the argument object is used.

Modified: python/branches/py3k-cdecimal/Doc/library/subprocess.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/subprocess.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/subprocess.rst	Sat Jul 17 12:36:09 2010
@@ -575,7 +575,7 @@
    pipe = os.popen(cmd, 'w')
    ...
    rc = pipe.close()
-   if rc is not None and rc % 256:
+   if rc is not None and rc >> 8:
        print("There were some errors")
    ==>
    process = Popen(cmd, 'w', stdin=PIPE)

Modified: python/branches/py3k-cdecimal/Doc/library/sys.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/sys.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/sys.rst	Sat Jul 17 12:36:09 2010
@@ -223,44 +223,65 @@
 .. data:: float_info
 
    A structseq holding information about the float type. It contains low level
-   information about the precision and internal representation. Please study
-   your system's :file:`float.h` for more information.
-
-   +---------------------+--------------------------------------------------+
-   | attribute           |  explanation                                     |
-   +=====================+==================================================+
-   | :const:`epsilon`    | Difference between 1 and the next representable  |
-   |                     | floating point number                            |
-   +---------------------+--------------------------------------------------+
-   | :const:`dig`        | digits (see :file:`float.h`)                     |
-   +---------------------+--------------------------------------------------+
-   | :const:`mant_dig`   | mantissa digits (see :file:`float.h`)            |
-   +---------------------+--------------------------------------------------+
-   | :const:`max`        | maximum representable finite float               |
-   +---------------------+--------------------------------------------------+
-   | :const:`max_exp`    | maximum int e such that radix**(e-1) is in the   |
-   |                     | range of finite representable floats             |
-   +---------------------+--------------------------------------------------+
-   | :const:`max_10_exp` | maximum int e such that 10**e is in the          |
-   |                     | range of finite representable floats             |
-   +---------------------+--------------------------------------------------+
-   | :const:`min`        | Minimum positive normalizer float                |
-   +---------------------+--------------------------------------------------+
-   | :const:`min_exp`    | minimum int e such that radix**(e-1) is a        |
-   |                     | normalized float                                 |
-   +---------------------+--------------------------------------------------+
-   | :const:`min_10_exp` | minimum int e such that 10**e is a normalized    |
-   |                     | float                                            |
-   +---------------------+--------------------------------------------------+
-   | :const:`radix`      | radix of exponent                                |
-   +---------------------+--------------------------------------------------+
-   | :const:`rounds`     | addition rounds (see :file:`float.h`)            |
-   +---------------------+--------------------------------------------------+
-
-   .. note::
-
-      The information in the table is simplified.
+   information about the precision and internal representation.  The values
+   correspond to the various floating-point constants defined in the standard
+   header file :file:`float.h` for the 'C' programming language; see section
+   5.2.4.2.2 of the 1999 ISO/IEC C standard [C99]_, 'Characteristics of
+   floating types', for details.
+
+   +---------------------+----------------+--------------------------------------------------+
+   | attribute           | float.h macro  | explanation                                      |
+   +=====================+================+==================================================+
+   | :const:`epsilon`    | DBL_EPSILON    | difference between 1 and the least value greater |
+   |                     |                | than 1 that is representable as a float          |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`dig`        | DBL_DIG        | maximum number of decimal digits that can be     |
+   |                     |                | faithfully represented in a float;  see below    |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`mant_dig`   | DBL_MANT_DIG   | float precision: the number of base-``radix``    |
+   |                     |                | digits in the significand of a float             |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`max`        | DBL_MAX        | maximum representable finite float               |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`max_exp`    | DBL_MAX_EXP    | maximum integer e such that ``radix**(e-1)`` is  |
+   |                     |                | a representable finite float                     |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`max_10_exp` | DBL_MAX_10_EXP | maximum integer e such that ``10**e`` is in the  |
+   |                     |                | range of representable finite floats             |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`min`        | DBL_MIN        | minimum positive normalized float                |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`min_exp`    | DBL_MIN_EXP    | minimum integer e such that ``radix**(e-1)`` is  |
+   |                     |                | a normalized float                               |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`min_10_exp` | DBL_MIN_10_EXP | minimum integer e such that ``10**e`` is a       |
+   |                     |                | normalized float                                 |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`radix`      | FLT_RADIX      | radix of exponent representation                 |
+   +---------------------+----------------+--------------------------------------------------+
+   | :const:`rounds`     | FLT_ROUNDS     | constant representing rounding mode              |
+   |                     |                | used for arithmetic operations                   |
+   +---------------------+----------------+--------------------------------------------------+
+
+   The attribute :attr:`sys.float_info.dig` needs further explanation.  If
+   ``s`` is any string representing a decimal number with at most
+   :attr:`sys.float_info.dig` significant digits, then converting ``s`` to a
+   float and back again will recover a string representing the same decimal
+   value::
 
+      >>> import sys
+      >>> sys.float_info.dig
+      15
+      >>> s = '3.14159265358979'    # decimal string with 15 significant digits
+      >>> format(float(s), '.15g')  # convert to float and back -> same value
+      '3.14159265358979'
+
+   But for strings with more than :attr:`sys.float_info.dig` significant digits,
+   this isn't always true::
+
+      >>> s = '9876543211234567'    # 16 significant digits is too many!
+      >>> format(float(s), '.16g')  # conversion changes value
+      '9876543211234568'
 
 .. data:: float_repr_style
 
@@ -957,3 +978,8 @@
    first three characters of :const:`version`.  It is provided in the :mod:`sys`
    module for informational purposes; modifying this value has no effect on the
    registry keys used by Python. Availability: Windows.
+
+.. rubric:: Citations
+
+.. [C99] ISO/IEC 9899:1999.  "Programming languages -- C."  A public draft of this standard is available at http://www.open-std.org/jtc1/sc22/wg14/www/docs/n1256.pdf .
+

Modified: python/branches/py3k-cdecimal/Doc/library/sysconfig.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/sysconfig.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/sysconfig.rst	Sat Jul 17 12:36:09 2010
@@ -16,9 +16,9 @@
 Configuration variables
 -----------------------
 
-A Python distribution contains a :file:`Makefile` file and a :file:`python.h`
-that are necessary to build the Python binary itself, but also any C extension
-created in a third party project and compiled using :mod:`distutils`.
+A Python distribution contains a :file:`Makefile` and a :file:`pyconfig.h`
+header file that are necessary to build both the Python binary itself and
+third-party C extensions compiled using :mod:`distutils`.
 
 :mod:`sysconfig` puts all variables found in these files in a dictionary that
 can be accessed using :func:`get_config_vars` or :func:`get_config_var`.

Modified: python/branches/py3k-cdecimal/Doc/library/unittest.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/unittest.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/unittest.rst	Sat Jul 17 12:36:09 2010
@@ -220,26 +220,26 @@
 
 unittest supports three command options.
 
-* -f / --failfast
+* :option:`-b` / :option:`--buffer`
 
-  Stop the test run on the first error or failure.
+  The standard output and standard error streams are buffered during the test
+  run. Output during a passing test is discarded. Output is echoed normally
+  on test fail or error and is added to the failure messages.
 
-* -c / --catch
+* :option:`-c` / :option:`--catch`
 
-  Control-c during the test run waits for the current test to end and then
-  reports all the results so far. A second control-c raises the normal
-  ``KeyboardInterrupt`` exception.
+  Control-C during the test run waits for the current test to end and then
+  reports all the results so far. A second control-C raises the normal
+  :exc:`KeyboardInterrupt` exception.
 
   See `Signal Handling`_ for the functions that provide this functionality.
 
-* -b / --buffer
+* :option:`-f` / :option:`--failfast`
 
-  The standard out and standard error streams are buffered during the test
-  run. Output during a passing test is discarded. Output is echoed normally
-  on test fail or error and is added to the failure messages.
+  Stop the test run on the first error or failure.
 
 .. versionadded:: 3.2
-   The command line options ``-c``, ``-b`` and ``-f`` where added.
+   The command line options ``-c``, ``-b`` and ``-f`` were added.
 
 The command line can also be used for test discovery, for running all of the
 tests in a project or just a subset.
@@ -270,8 +270,9 @@
    -t directory     Top level directory of project (default to
                     start directory)
 
-The -s, -p, & -t options can be passsed in as positional arguments. The
-following two command lines are equivalent::
+The :option:`-s`, :option:`-p`, and :option:`-t` options can be passed in
+as positional arguments in that order. The following two command lines
+are equivalent::
 
    python -m unittest discover -s project_directory -p '*_test.py'
    python -m unittest discover project_directory '*_test.py'
@@ -829,6 +830,11 @@
       compare equal, the test will fail with the explanation given by *msg*, or
       :const:`None`.
 
+      If *delta* is supplied instead of *places* then the difference
+      between *first* and *second* must be less than *delta*.
+
+      Supplying both *delta* and *places* raises a ``TypeError``.
+
       .. versionchanged:: 3.2
          Objects that compare equal are automatically almost equal.
          Added the ``delta`` keyword argument.
@@ -849,7 +855,7 @@
       compare equal, the test will fail with the explanation given by *msg*, or
       :const:`None`.
 
-      If *delta* is supplied instead of *places* then the the difference
+      If *delta* is supplied instead of *places* then the difference
       between *first* and *second* must be more than *delta*.
 
       Supplying both *delta* and *places* raises a ``TypeError``.
@@ -1902,12 +1908,12 @@
 Signal Handling
 ---------------
 
-The -c / --catch command line option to unittest, along with the ``catchbreak``
+The :option:`-c`/:option:`--catch` command line option to unittest, along with the ``catchbreak``
 parameter to :func:`unittest.main()`, provide more friendly handling of
-control-c during a test run. With catch break behavior enabled control-c will
+control-C during a test run. With catch break behavior enabled control-C will
 allow the currently running test to complete, and the test run will then end
 and report all the results so far. A second control-c will raise a
-``KeyboardInterrupt`` in the usual way.
+:exc:`KeyboardInterrupt` in the usual way.
 
 The control-c handling signal handler attempts to remain compatible with code or
 tests that install their own :const:`signal.SIGINT` handler. If the ``unittest``

Modified: python/branches/py3k-cdecimal/Doc/library/urllib.parse.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/urllib.parse.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/urllib.parse.rst	Sat Jul 17 12:36:09 2010
@@ -310,23 +310,29 @@
    ``b'a&\xef'``.
 
 
-.. function:: urlencode(query, doseq=False)
+.. function:: urlencode(query, doseq=False, safe='', encoding=None, errors=None)
 
-   Convert a mapping object or a sequence of two-element tuples to a
-   "url-encoded" string, suitable to pass to :func:`urlopen` above as the
-   optional *data* argument.  This is useful to pass a dictionary of form
-   fields to a ``POST`` request.  The resulting string is a series of
-   ``key=value`` pairs separated by ``'&'`` characters, where both *key* and
-   *value* are quoted using :func:`quote_plus` above. When a sequence of
-   two-element tuples is used as the *query* argument, the first element of
-   each tuple is a key and the second is a value. The value element in itself
-   can be a sequence and in that case, if the optional parameter *doseq* is
-   evaluates to *True*, individual ``key=value`` pairs separated by ``'&'`` are
-   generated for each element of the value sequence for the key.  The order of
-   parameters in the encoded string will match the order of parameter tuples in
-   the sequence. This module provides the functions :func:`parse_qs` and
-   :func:`parse_qsl` which are used to parse query strings into Python data
-   structures.
+   Convert a mapping object or a sequence of two-element tuples, which may
+   either be a :class:`str` or a :class:`bytes`,  to a "url-encoded" string,
+   suitable to pass to :func:`urlopen` above as the optional *data* argument.
+   This is useful to pass a dictionary of form fields to a ``POST`` request.
+   The resulting string is a series of ``key=value`` pairs separated by ``'&'``
+   characters, where both *key* and *value* are quoted using :func:`quote_plus`
+   above. When a sequence of two-element tuples is used as the *query*
+   argument, the first element of each tuple is a key and the second is a
+   value. The value element in itself can be a sequence and in that case, if
+   the optional parameter *doseq* is evaluates to *True*, individual
+   ``key=value`` pairs separated by ``'&'`` are generated for each element of
+   the value sequence for the key.  The order of parameters in the encoded
+   string will match the order of parameter tuples in the sequence. This module
+   provides the functions :func:`parse_qs` and :func:`parse_qsl` which are used
+   to parse query strings into Python data structures.
+
+   When *query* parameter is a :class:`str`, the *safe*, *encoding* and *error*
+   parameters are sent the :func:`quote_plus` for encoding.
+
+   .. versionchanged:: 3.2
+      query paramater supports bytes and string.
 
 
 .. seealso::

Modified: python/branches/py3k-cdecimal/Doc/library/urllib.request.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/urllib.request.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/urllib.request.rst	Sat Jul 17 12:36:09 2010
@@ -40,8 +40,8 @@
      commonly used to determine if a redirect was followed
 
    * :meth:`info` --- return the meta-information of the page, such as headers,
-     in the form of an :class:`http.client.HTTPMessage` instance (see `Quick
-     Reference to HTTP Headers <http://www.cs.tut.fi/~jkorpela/http.html>`_)
+     in the form of an :func:`email.message_from_string` instance (see
+     `Quick Reference to HTTP Headers <http://www.cs.tut.fi/~jkorpela/http.html>`_)
 
    Raises :exc:`URLError` on errors.
 

Modified: python/branches/py3k-cdecimal/Doc/library/warnings.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/library/warnings.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/library/warnings.rst	Sat Jul 17 12:36:09 2010
@@ -57,7 +57,7 @@
 | :exc:`UserWarning`               | The default category for :func:`warn`.        |
 +----------------------------------+-----------------------------------------------+
 | :exc:`DeprecationWarning`        | Base category for warnings about deprecated   |
-|                                  | features.                                     |
+|                                  | features (ignored by default).                |
 +----------------------------------+-----------------------------------------------+
 | :exc:`SyntaxWarning`             | Base category for warnings about dubious      |
 |                                  | syntactic features.                           |
@@ -91,6 +91,9 @@
 standard warning categories.  A warning category must always be a subclass of
 the :exc:`Warning` class.
 
+.. versionchanged:: 3.2
+   :exc:`DeprecationWarning` is ignored by default.
+
 
 .. _warning-filter:
 
@@ -150,14 +153,6 @@
 :mod:`warnings` module parses these when it is first imported (invalid options
 are ignored, after printing a message to ``sys.stderr``).
 
-The warnings that are ignored by default may be enabled by passing :option:`-Wd`
-to the interpreter. This enables default handling for all warnings, including
-those that are normally ignored by default. This is particular useful for
-enabling ImportWarning when debugging problems importing a developed package.
-ImportWarning can also be enabled explicitly in Python code using::
-
-   warnings.simplefilter('default', ImportWarning)
-
 
 .. _warning-suppress:
 
@@ -233,6 +228,37 @@
 entries from the warnings list before each new operation).
 
 
+Updating Code For New Versions of Python
+----------------------------------------
+
+Warnings that are only of interest to the developer are ignored by default. As
+such you should make sure to test your code with typically ignored warnings
+made visible. You can do this from the command-line by passing :option:`-Wd`
+to the interpreter (this is shorthand for :option:`-W default`).  This enables
+default handling for all warnings, including those that are ignored by default.
+To change what action is taken for encountered warnings you simply change what
+argument is passed to :option:`-W`, e.g. :option:`-W error`. See the
+:option:`-W` flag for more details on what is possible.
+
+To programmatically do the same as :option:`-Wd`, use::
+
+  warnings.simplefilter('default')
+
+Make sure to execute this code as soon as possible. This prevents the
+registering of what warnings have been raised from unexpectedly influencing how
+future warnings are treated.
+
+Having certain warnings ignored by default is done to prevent a user from
+seeing warnings that are only of interest to the developer. As you do not
+necessarily have control over what interpreter a user uses to run their code,
+it is possible that a new version of Python will be released between your
+release cycles.  The new interpreter release could trigger new warnings in your
+code that were not there in an older interpreter, e.g.
+:exc:`DeprecationWarning` for a module that you are using. While you as a
+developer want to be notified that your code is using a deprecated module, to a
+user this information is essentially noise and provides no benefit to them.
+
+
 .. _warning-functions:
 
 Available Functions

Modified: python/branches/py3k-cdecimal/Doc/reference/compound_stmts.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/reference/compound_stmts.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/reference/compound_stmts.rst	Sat Jul 17 12:36:09 2010
@@ -272,14 +272,12 @@
        try:
            foo
        finally:
-           N = None
            del N
 
-That means that you have to assign the exception to a different name if you want
-to be able to refer to it after the except clause.  The reason for this is that
-with the traceback attached to them, exceptions will form a reference cycle with
-the stack frame, keeping all locals in that frame alive until the next garbage
-collection occurs.
+This means the exception must be assigned to a different name to be able to
+refer to it after the except clause.  Exceptions are cleared because with the
+traceback attached to them, they form a reference cycle with the stack frame,
+keeping all locals in that frame alive until the next garbage collection occurs.
 
 .. index::
    module: sys

Modified: python/branches/py3k-cdecimal/Doc/reference/datamodel.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/reference/datamodel.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/reference/datamodel.rst	Sat Jul 17 12:36:09 2010
@@ -1595,7 +1595,7 @@
 
 In particular, the metaclass :class:`abc.ABCMeta` implements these methods in
 order to allow the addition of Abstract Base Classes (ABCs) as "virtual base
-classes" to any class or type (including built-in types), and including to other
+classes" to any class or type (including built-in types), including other
 ABCs.
 
 .. method:: class.__instancecheck__(self, instance)
@@ -1614,7 +1614,7 @@
 
 Note that these methods are looked up on the type (metaclass) of a class.  They
 cannot be defined as class methods in the actual class.  This is consistent with
-the lookup of special methods that are called on instances, only that in this
+the lookup of special methods that are called on instances, only in this
 case the instance is itself a class.
 
 .. seealso::

Modified: python/branches/py3k-cdecimal/Doc/reference/expressions.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/reference/expressions.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/reference/expressions.rst	Sat Jul 17 12:36:09 2010
@@ -1031,9 +1031,9 @@
   ``x <= y``.  If the corresponding element does not exist, the shorter
   sequence is ordered first (for example, ``[1,2] < [1,2,3]``).
 
-* Mappings (dictionaries) compare equal if and only if their sorted ``(key,
-  value)`` lists compare equal. [#]_ Outcomes other than equality are resolved
-  consistently, but are not otherwise defined. [#]_
+* Mappings (dictionaries) compare equal if and only if they have the same
+  ``(key, value)`` pairs. Order comparisons ``('<', '<=', '>=', '>')``
+  raise :exc:`TypeError`.
 
 * Sets and frozensets define comparison operators to mean subset and superset
   tests.  Those relations do not define total orderings (the two sets ``{1,2}``
@@ -1330,15 +1330,6 @@
    strings in a human recognizable way, compare using
    :func:`unicodedata.normalize`.
 
-.. [#] The implementation computes this efficiently, without constructing lists
-   or sorting.
-
-.. [#] Earlier versions of Python used lexicographic comparison of the sorted (key,
-   value) lists, but this was very expensive for the common case of comparing
-   for equality.  An even earlier version of Python compared dictionaries by
-   identity only, but this caused surprises because people expected to be able
-   to test a dictionary for emptiness by comparing it to ``{}``.
-
 .. [#] Due to automatic garbage-collection, free lists, and the dynamic nature of
    descriptors, you may notice seemingly unusual behaviour in certain uses of
    the :keyword:`is` operator, like those involving comparisons between instance

Modified: python/branches/py3k-cdecimal/Doc/reference/introduction.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/reference/introduction.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/reference/introduction.rst	Sat Jul 17 12:36:09 2010
@@ -69,12 +69,12 @@
    more information, see `the IronPython website <http://www.ironpython.com/>`_.
 
 PyPy
-   An implementation of Python written in Python; even the bytecode interpreter is
-   written in Python.  This is executed using CPython as the underlying
-   interpreter.  One of the goals of the project is to encourage experimentation
-   with the language itself by making it easier to modify the interpreter (since it
-   is written in Python).  Additional information is available on `the PyPy
-   project's home page <http://codespeak.net/pypy/>`_.
+   An implementation of Python written completely in Python. It supports several
+   advanced features not found in other implementations like stackless support
+   and a Just in Time compiler. One of the goals of the project is to encourage
+   experimentation with the language itself by making it easier to modify the
+   interpreter (since it is written in Python).  Additional information is
+   available on `the PyPy project's home page <http://pypy.org/>`_.
 
 Each of these implementations varies in some way from the language as documented
 in this manual, or introduces specific information beyond what's covered in the

Modified: python/branches/py3k-cdecimal/Doc/reference/lexical_analysis.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/reference/lexical_analysis.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/reference/lexical_analysis.rst	Sat Jul 17 12:36:09 2010
@@ -407,7 +407,7 @@
 
 .. productionlist::
    bytesliteral: `bytesprefix`(`shortbytes` | `longbytes`)
-   bytesprefix: "b" | "B"
+   bytesprefix: "b" | "B" | "br" | "Br" | "bR" | "BR"
    shortbytes: "'" `shortbytesitem`* "'" | '"' `shortbytesitem`* '"'
    longbytes: "'''" `longbytesitem`* "'''" | '"""' `longbytesitem`* '"""'
    shortbytesitem: `shortbyteschar` | `bytesescapeseq`
@@ -431,16 +431,16 @@
 characters that otherwise have a special meaning, such as newline, backslash
 itself, or the quote character.
 
-String literals may optionally be prefixed with a letter ``'r'`` or ``'R'``;
-such strings are called :dfn:`raw strings` and treat backslashes as literal
-characters.  As a result, ``'\U'`` and ``'\u'`` escapes in raw strings are not
-treated specially.
-
 Bytes literals are always prefixed with ``'b'`` or ``'B'``; they produce an
 instance of the :class:`bytes` type instead of the :class:`str` type.  They
 may only contain ASCII characters; bytes with a numeric value of 128 or greater
 must be expressed with escapes.
 
+Both string and bytes literals may optionally be prefixed with a letter ``'r'``
+or ``'R'``; such strings are called :dfn:`raw strings` and treat backslashes as
+literal characters.  As a result, in string literals, ``'\U'`` and ``'\u'``
+escapes in raw strings are not treated specially.
+
 In triple-quoted strings, unescaped newlines and quotes are allowed (and are
 retained), except that three unescaped quotes in a row terminate the string.  (A
 "quote" is the character used to open the string, i.e. either ``'`` or ``"``.)
@@ -545,9 +545,9 @@
 String literal concatenation
 ----------------------------
 
-Multiple adjacent string literals (delimited by whitespace), possibly using
-different quoting conventions, are allowed, and their meaning is the same as
-their concatenation.  Thus, ``"hello" 'world'`` is equivalent to
+Multiple adjacent string or bytes literals (delimited by whitespace), possibly
+using different quoting conventions, are allowed, and their meaning is the same
+as their concatenation.  Thus, ``"hello" 'world'`` is equivalent to
 ``"helloworld"``.  This feature can be used to reduce the number of backslashes
 needed, to split long strings conveniently across long lines, or even to add
 comments to parts of strings, for example::

Modified: python/branches/py3k-cdecimal/Doc/tools/sphinxext/indexsidebar.html
==============================================================================
--- python/branches/py3k-cdecimal/Doc/tools/sphinxext/indexsidebar.html	(original)
+++ python/branches/py3k-cdecimal/Doc/tools/sphinxext/indexsidebar.html	Sat Jul 17 12:36:09 2010
@@ -2,8 +2,7 @@
             <p><a href="{{ pathto('download') }}">Download these documents</a></p>
 	    <h3>Docs for other versions</h3>
 	    <ul>
-	      <li><a href="http://docs.python.org/2.6/">Python 2.6 (stable)</a></li>
-	      <li><a href="http://docs.python.org/2.7/">Python 2.7 (in development)</a></li>
+	      <li><a href="http://docs.python.org/2.7/">Python 2.7 (stable)</a></li>
 	      <li><a href="http://docs.python.org/3.1/">Python 3.1 (stable)</a></li>
               <li><a href="http://www.python.org/doc/versions/">Old versions</a></li>
             </ul>

Modified: python/branches/py3k-cdecimal/Doc/tools/sphinxext/pyspecific.py
==============================================================================
--- python/branches/py3k-cdecimal/Doc/tools/sphinxext/pyspecific.py	(original)
+++ python/branches/py3k-cdecimal/Doc/tools/sphinxext/pyspecific.py	Sat Jul 17 12:36:09 2010
@@ -5,7 +5,7 @@
 
     Sphinx extension with Python doc-specific markup.
 
-    :copyright: 2008, 2009 by Georg Brandl.
+    :copyright: 2008, 2009, 2010 by Georg Brandl.
     :license: Python license.
 """
 
@@ -149,7 +149,7 @@
 import re
 from sphinx import addnodes
 
-opcode_sig_re = re.compile(r'(\w+(?:\+\d)?)\s*\((.*)\)')
+opcode_sig_re = re.compile(r'(\w+(?:\+\d)?)(?:\s*\((.*)\))?')
 
 def parse_opcode_signature(env, sig, signode):
     """Transform an opcode signature into RST nodes."""
@@ -158,9 +158,10 @@
         raise ValueError
     opname, arglist = m.groups()
     signode += addnodes.desc_name(opname, opname)
-    paramlist = addnodes.desc_parameterlist()
-    signode += paramlist
-    paramlist += addnodes.desc_parameter(arglist, arglist)
+    if arglist is not None:
+        paramlist = addnodes.desc_parameterlist()
+        signode += paramlist
+        paramlist += addnodes.desc_parameter(arglist, arglist)
     return opname.strip()
 
 

Modified: python/branches/py3k-cdecimal/Doc/tutorial/classes.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/tutorial/classes.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/tutorial/classes.rst	Sat Jul 17 12:36:09 2010
@@ -580,7 +580,7 @@
 =================
 
 "Private" instance variables that cannot be accessed except from inside an
-object, don't exist in Python.  However, there is a convention that is followed
+object don't exist in Python.  However, there is a convention that is followed
 by most Python code: a name prefixed with an underscore (e.g. ``_spam``) should
 be treated as a non-public part of the API (whether it is a function, a method
 or a data member).  It should be considered an implementation detail and subject
@@ -599,12 +599,12 @@
 possible to access or modify a variable that is considered private.  This can
 even be useful in special circumstances, such as in the debugger.
 
-Notice that code passed to ``exec()``, ``eval()`` or ``execfile()`` does not
-consider the classname of the invoking  class to be the current class; this is
-similar to the effect of the  ``global`` statement, the effect of which is
-likewise restricted to  code that is byte-compiled together.  The same
-restriction applies to ``getattr()``, ``setattr()`` and ``delattr()``, as well
-as when referencing ``__dict__`` directly.
+Notice that code passed to ``exec()`` or ``eval()`` does not consider the
+classname of the invoking class to be the current class; this is similar to the
+effect of the ``global`` statement, the effect of which is likewise restricted
+to code that is byte-compiled together.  The same restriction applies to
+``getattr()``, ``setattr()`` and ``delattr()``, as well as when referencing
+``__dict__`` directly.
 
 
 .. _tut-odds:

Modified: python/branches/py3k-cdecimal/Doc/tutorial/controlflow.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/tutorial/controlflow.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/tutorial/controlflow.rst	Sat Jul 17 12:36:09 2010
@@ -87,7 +87,6 @@
 If you do need to iterate over a sequence of numbers, the built-in function
 :func:`range` comes in handy.  It generates arithmetic progressions::
 
-
     >>> for i in range(5):
     ...     print(i)
     ...
@@ -97,9 +96,7 @@
     3
     4
 
-
-
-The given end point is never part of the generated list; ``range(10)`` generates
+The given end point is never part of the generated sequence; ``range(10)`` generates
 10 values, the legal indices for items of a sequence of length 10.  It
 is possible to let the range start at another number, or to specify a different
 increment (even negative; sometimes this is called the 'step')::

Modified: python/branches/py3k-cdecimal/Doc/tutorial/introduction.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/tutorial/introduction.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/tutorial/introduction.rst	Sat Jul 17 12:36:09 2010
@@ -208,7 +208,7 @@
 
    print(hello)
 
-Note that newlines still need to be embedded in the string using ``\n``; the
+Note that newlines still need to be embedded in the string using ``\n`` -- the
 newline following the trailing backslash is discarded.  This example would print
 the following:
 

Modified: python/branches/py3k-cdecimal/Doc/using/cmdline.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/using/cmdline.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/using/cmdline.rst	Sat Jul 17 12:36:09 2010
@@ -61,7 +61,7 @@
 
 .. cmdoption:: -c <command>
 
-   Execute the Python code in *command*.  *command* can be one ore more
+   Execute the Python code in *command*.  *command* can be one or more
    statements separated by newlines, with significant leading whitespace as in
    normal module code.
 

Modified: python/branches/py3k-cdecimal/Doc/using/mac.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/using/mac.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/using/mac.rst	Sat Jul 17 12:36:09 2010
@@ -114,8 +114,8 @@
 =======
 
 MacPython ships with the standard IDLE development environment. A good
-introduction to using IDLE can be found at http://hkn.eecs.berkeley.edu/
-dyoo/python/idle_intro/index.html.
+introduction to using IDLE can be found at
+http://hkn.eecs.berkeley.edu/~dyoo/python/idle_intro/index.html.
 
 
 .. _mac-package-manager:

Modified: python/branches/py3k-cdecimal/Doc/whatsnew/2.7.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/whatsnew/2.7.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/whatsnew/2.7.rst	Sat Jul 17 12:36:09 2010
@@ -6,12 +6,11 @@
 :Release: |release|
 :Date: |today|
 
-.. Fix accents on Kristjan Valur Jonsson, Fuerstenau
-
-.. Big jobs: ElementTree 1.3, pep 391, sysconfig
-..  unittest test discovery
 ..  hyperlink all the methods & functions.
 
+.. T_STRING_INPLACE not described in main docs
+.. "Format String Syntax" in string.rst could use many more examples.
+
 .. $Id$
    Rules for maintenance:
 
@@ -54,20 +53,77 @@
    when researching a change.
 
 This article explains the new features in Python 2.7.  The final
-release of 2.7 is currently scheduled for June 2010; the detailed
+release of 2.7 is currently scheduled for July 2010; the detailed
 schedule is described in :pep:`373`.
 
-Python 2.7 is planned to be the last major release in the 2.x series.
-Though more major releases have not been absolutely ruled out, the
-Python maintainers are planning to focus more on Python 3.x.  Despite
-that, it's likely that the 2.7 release will have a longer period of
-maintenance compared to earlier 2.x versions.
-
-.. Compare with previous release in 2 - 3 sentences here.
-   add hyperlink when the documentation becomes available online.
+Numeric handling has been improved in many ways, for both
+floating-point numbers and for the :class:`Decimal` class.  There are
+some useful additions to the standard library, such as a greatly
+enhanced :mod:`unittest` module, the :mod:`argparse` module for
+parsing command-line options, convenient ordered-dictionary and
+:class:`Counter` classes in the :mod:`collections` module, and many
+other improvements.
+
+Python 2.7 is planned to be the last of the 2.x releases, so we worked
+on making it a good release for the long term.  To help with porting
+to Python 3, several new features from the Python 3.x series have been
+included in 2.7.
+
+This article doesn't attempt to provide a complete specification of
+the new features, but instead provides a convenient overview.  For
+full details, you should refer to the documentation for Python 2.7 at
+http://docs.python.org. If you want to understand the rationale for
+the design and implementation, refer to the PEP for a particular new
+feature or the issue on http://bugs.python.org in which a change was
+discussed.  Whenever possible, "What's New in Python" links to the
+bug/patch item for each change.
 
 .. _whatsnew27-python31:
 
+The Future for Python 2.x
+=========================
+
+Python 2.7 is intended to be the last major release in the 2.x series.
+The Python maintainers are planning to focus their future efforts on
+the Python 3.x series.
+
+This means that 2.7 will remain in place for a long time, running
+production systems that have not been ported to Python 3.x.
+Two consequences of the long-term significance of 2.7 are:
+
+* It's very likely the 2.7 release will have a longer period of
+  maintenance compared to earlier 2.x versions.  Python 2.7 will
+  continue to be maintained while the transition to 3.x continues, and
+  the developers are planning to support Python 2.7 with bug-fix
+  releases beyond the typical two years.
+
+* A policy decision was made to silence warnings only of interest to
+  developers.  :exc:`DeprecationWarning` and its
+  descendants are now ignored unless otherwise requested, preventing
+  users from seeing warnings triggered by an application.  This change
+  was also made in the branch that will become Python 3.2. (Discussed
+  on stdlib-sig and carried out in :issue:`7319`.)
+
+  In previous releases, :exc:`DeprecationWarning` messages were
+  enabled by default, providing Python developers with a clear
+  indication of where their code may break in a future major version
+  of Python.
+
+  However, there are increasingly many users of Python-based
+  applications who are not directly involved in the development of
+  those applications.  :exc:`DeprecationWarning` messages are
+  irrelevant to such users, making them worry about an application
+  that's actually working correctly and burdening application developers
+  with responding to these concerns.
+
+  You can re-enable display of :exc:`DeprecationWarning` messages by
+  running Python with the :option:`-Wdefault` (short form:
+  :option:`-Wd`) switch, or by setting the :envvar:`PYTHONWARNINGS`
+  environment variable to ``"default"`` (or ``"d"``) before running
+  Python.  Python code can also re-enable them
+  by calling ``warnings.simplefilter('default')``.
+
+
 Python 3.1 Features
 =======================
 
@@ -78,25 +134,31 @@
 
 A partial list of 3.1 features that were backported to 2.7:
 
-* A version of the :mod:`io` library, rewritten in C for performance.
+* The syntax for set literals (``{1,2,3}`` is a mutable set).
+* Dictionary and set comprehensions (``{ i: i*2 for i in range(3)}``).
+* Multiple context managers in a single :keyword:`with` statement.
+* A new version of the :mod:`io` library, rewritten in C for performance.
 * The ordered-dictionary type described in :ref:`pep-0372`.
-* The new format specifier described in :ref:`pep-0378`.
+* The new ``","`` format specifier described in :ref:`pep-0378`.
 * The :class:`memoryview` object.
-* A small subset of the :mod:`importlib` module `described below <#importlib-section>`__.
+* A small subset of the :mod:`importlib` module,
+  `described below <#importlib-section>`__.
 * Float-to-string and string-to-float conversions now round their
-  results more correctly.  And :func:`repr` of a floating-point
+  results more correctly, and :func:`repr` of a floating-point
   number *x* returns a result that's guaranteed to round back to the
-  same number when converted back to a string.
+  same number when converted back to a float.
+* The :ctype:`PyCapsule` type, used to provide a C API for extension modules.
 * The :cfunc:`PyLong_AsLongAndOverflow` C API function.
 
-One porting change: the :option:`-3` switch now automatically
-enables the :option:`-Qwarn` switch that causes warnings
-about using classic division with integers and long integers.
-
 Other new Python3-mode warnings include:
 
 * :func:`operator.isCallable` and :func:`operator.sequenceIncludes`,
-  which are not supported in 3.x.
+  which are not supported in 3.x, now trigger warnings.
+* The :option:`-3` switch now automatically
+  enables the :option:`-Qwarn` switch that causes warnings
+  about using classic division with integers and long integers.
+
+
 
 .. ========================================================================
 .. Large, PEP-level features and changes should be described here.
@@ -104,22 +166,22 @@
 
 .. _pep-0372:
 
-PEP 372: Adding an ordered dictionary to collections
+PEP 372: Adding an Ordered Dictionary to collections
 ====================================================
 
 Regular Python dictionaries iterate over key/value pairs in arbitrary order.
 Over the years, a number of authors have written alternative implementations
 that remember the order that the keys were originally inserted.  Based on
-the experiences from those implementations, a new
-:class:`~collections.OrderedDict` class has been introduced in the
-:mod:`collections` module.
+the experiences from those implementations, 2.7 introduces a new
+:class:`~collections.OrderedDict` class in the :mod:`collections` module.
 
-The :class:`~collections.OrderedDict` API is substantially the same as regular
-dictionaries but will iterate over keys and values in a guaranteed order
+The :class:`~collections.OrderedDict` API provides the same interface as regular
+dictionaries but iterates over keys and values in a guaranteed order
 depending on when a key was first inserted::
 
     >>> from collections import OrderedDict
-    >>> d = OrderedDict([('first', 1), ('second', 2),
+    >>> d = OrderedDict([('first', 1),
+    ...                  ('second', 2),
     ...                  ('third', 3)])
     >>> d.items()
     [('first', 1), ('second', 2), ('third', 3)]
@@ -156,9 +218,11 @@
 Comparing two ordered dictionaries checks both the keys and values,
 and requires that the insertion order was the same::
 
-    >>> od1 = OrderedDict([('first', 1), ('second', 2),
+    >>> od1 = OrderedDict([('first', 1),
+    ...                    ('second', 2),
     ...                    ('third', 3)])
-    >>> od2 = OrderedDict([('third', 3), ('first', 1),
+    >>> od2 = OrderedDict([('third', 3),
+    ...                    ('first', 1),
     ...                    ('second', 2)])
     >>> od1 == od2
     False
@@ -176,17 +240,12 @@
 deletion doesn't have to traverse the entire linked list and therefore
 remains O(1).
 
-.. XXX check O(1)-ness with Raymond
-..     Also check if the 'somenamedtuple' in the collection module should
-..     be replaced/removed in order to use
-..     :meth:`~collections.namedtuple._asdict()` (see below)
-
 The standard library now supports use of ordered dictionaries in several
 modules.
 
-* The :mod:`ConfigParser` module uses them by default, letting
-  configuration files be read, modified, and then written back in their original
-  order.
+* The :mod:`ConfigParser` module uses them by default, meaning that
+  configuration files can now read, modified, and then written back
+  in their original order.
 
 * The :meth:`~collections.somenamedtuple._asdict()` method for
   :func:`collections.namedtuple` now returns an ordered dictionary with the
@@ -210,7 +269,7 @@
 =================================================
 
 To make program output more readable, it can be useful to add
-separators to large numbers and render them as
+separators to large numbers, rendering them as
 18,446,744,073,709,551,616 instead of 18446744073709551616.
 
 The fully general solution for doing this is the :mod:`locale` module,
@@ -237,8 +296,6 @@
 comma-formatting mechanism isn't as general as the :mod:`locale`
 module, but it's easier to use.
 
-.. XXX "Format String Syntax" in string.rst could use many more examples.
-
 .. seealso::
 
    :pep:`378` - Format Specifier for Thousands Separator
@@ -248,13 +305,13 @@
 ======================================================
 
 The :mod:`argparse` module for parsing command-line arguments was
-added, intended as a more powerful replacement for the
+added as a more powerful replacement for the
 :mod:`optparse` module.
 
 This means Python now supports three different modules for parsing
 command-line arguments: :mod:`getopt`, :mod:`optparse`, and
 :mod:`argparse`.  The :mod:`getopt` module closely resembles the C
-:cfunc:`getopt` function, so it remains useful if you're writing a
+library's :cfunc:`getopt` function, so it remains useful if you're writing a
 Python prototype that will eventually be rewritten in C.
 :mod:`optparse` becomes redundant, but there are no plans to remove it
 because there are many scripts still using it, and there's no
@@ -306,23 +363,28 @@
       -o FILE     direct output to FILE instead of stdout
       -C NUM      display NUM lines of added context
 
-Similarly to :mod:`optparse`, the command-line switches and arguments
+As with :mod:`optparse`, the command-line switches and arguments
 are returned as an object with attributes named by the *dest* parameters::
 
     -> ./python.exe argparse-example.py -v
-    {'output': None, 'is_verbose': True, 'context': 0, 'inputs': []}
+    {'output': None,
+     'is_verbose': True,
+     'context': 0,
+     'inputs': []}
 
     -> ./python.exe argparse-example.py -v -o /tmp/output -C 4 file1 file2
-    {'output': '/tmp/output', 'is_verbose': True, 'context': 4,
+    {'output': '/tmp/output',
+     'is_verbose': True,
+     'context': 4,
      'inputs': ['file1', 'file2']}
 
 :mod:`argparse` has much fancier validation than :mod:`optparse`; you
 can specify an exact number of arguments as an integer, 0 or more
 arguments by passing ``'*'``, 1 or more by passing ``'+'``, or an
 optional argument with ``'?'``.  A top-level parser can contain
-sub-parsers, so you can define subcommands that have different sets of
+sub-parsers to define subcommands that have different sets of
 switches, as in ``svn commit``, ``svn checkout``, etc.  You can
-specify an argument type as :class:`~argparse.FileType`, which will
+specify an argument's type as :class:`~argparse.FileType`, which will
 automatically open files for you and understands that ``'-'`` means
 standard input or output.
 
@@ -331,6 +393,8 @@
    `argparse module documentation <http://docs.python.org/dev/library/argparse.html>`__
 
    `Upgrading optparse code to use argparse <http://docs.python.org/dev/library/argparse.html#upgrading-optparse-code>`__
+     Part of the Python documentation, describing how to convert
+     code that uses :mod:`optparse`.
 
    :pep:`389` - argparse - New Command Line Parsing Module
      PEP written and implemented by Steven Bethard.
@@ -338,34 +402,88 @@
 PEP 391: Dictionary-Based Configuration For Logging
 ====================================================
 
-.. not documented in library reference yet.
+.. XXX not documented in library reference yet; add link here once it's added.
 
-The :mod:`logging` module is very flexible; an application can define
+The :mod:`logging` module is very flexible; applications can define
 a tree of logging subsystems, and each logger in this tree can filter
 out certain messages, format them differently, and direct messages to
 a varying number of handlers.
 
 All this flexibility can require a lot of configuration.  You can
 write Python statements to create objects and set their properties,
-but a complex set-up would require verbose but boring code.
+but a complex set-up requires verbose but boring code.
 :mod:`logging` also supports a :func:`~logging.config.fileConfig`
 function that parses a file, but the file format doesn't support
 configuring filters, and it's messier to generate programmatically.
 
 Python 2.7 adds a :func:`~logging.config.dictConfig` function that
-uses a dictionary, and there are many ways to produce a dictionary
-from different sources.  You can construct one with code, of course.
-Python's standard library now includes a JSON parser, so you could
-parse a file containing JSON, or you could use a YAML parsing library
-if one is installed.
+uses a dictionary to configure logging.  There are many ways to
+produce a dictionary from different sources: construct one with code;
+parse a file containing JSON; or use a YAML parsing library if one is
+installed.
+
+The following example configures two loggers, the root logger and a
+logger named "network".   Messages sent to the root logger will be
+sent to the system log using the syslog protocol, and messages
+to the "network" logger will be written to a :file:`network.log` file
+that will be rotated once the log reaches 1Mb.
+
+::
+
+    import logging
+    import logging.config
 
-XXX describe an example.
+    configdict = {
+     'version': 1,    # Configuration schema in use; must be 1 for now
+     'formatters': {
+         'standard': {
+             'format': ('%(asctime)s %(name)-15s '
+                        '%(levelname)-8s %(message)s')}},
+
+     'handlers': {'netlog': {'backupCount': 10,
+                         'class': 'logging.handlers.RotatingFileHandler',
+                         'filename': '/logs/network.log',
+                         'formatter': 'standard',
+                         'level': 'INFO',
+                         'maxBytes': 1024*1024},
+                  'syslog': {'class': 'logging.handlers.SysLogHandler',
+                             'formatter': 'standard',
+                             'level': 'ERROR'}},
+
+     # Specify all the subordinate loggers
+     'loggers': {
+                 'network': {
+                             'handlers': ['netlog']
+                 }
+     },
+     # Specify properties of the root logger
+     'root': {
+              'handlers': ['syslog']
+     },
+    }
+
+    # Set up configuration
+    logging.config.dictConfig(configdict)
+
+    # As an example, log two error messages
+    logger = logging.getLogger('/')
+    logger.error('Database not found')
 
-Two smaller enhancements to the logging module are:
+    netlogger = logging.getLogger('network')
+    netlogger.error('Connection failed')
+
+Three smaller enhancements to the :mod:`logging` module, all
+implemented by Vinay Sajip, are:
 
 .. rev79293
 
-* :class:`Logger` instances gained a :meth:`getChild` that retrieves a
+* The :class:`~logging.handlers.SysLogHandler` class now supports
+  syslogging over TCP.  The constructor has a *socktype* parameter
+  giving the type of socket to use, either :const:`socket.SOCK_DGRAM`
+  for UDP or :const:`socket.SOCK_STREAM` for TCP.  The default
+  protocol remains UDP.
+
+* :class:`Logger` instances gained a :meth:`getChild` method that retrieves a
   descendant logger using a relative path.  For example,
   once you retrieve a logger by doing ``log = getLogger('app')``,
   calling ``log.getChild('network.listen')`` is equivalent to
@@ -387,12 +505,10 @@
 are different in Python 3.x.  They return an object called a :dfn:`view`
 instead of a fully materialized list.
 
-.. Views can be iterated over, but they also behave like sets.  XXX not working.
-
 It's not possible to change the return values of :meth:`keys`,
 :meth:`values`, and :meth:`items` in Python 2.7 because too much code
 would break.  Instead the 3.x versions were added under the new names
-of :meth:`viewkeys`, :meth:`viewvalues`, and :meth:`viewitems`.
+:meth:`viewkeys`, :meth:`viewvalues`, and :meth:`viewitems`.
 
 ::
 
@@ -402,6 +518,17 @@
     >>> d.viewkeys()
     dict_keys([0, 130, 10, 140, 20, 150, 30, ..., 250])
 
+Views can be iterated over, but the key and item views also behave
+like sets.  The ``&`` operator performs intersection, and ``|``
+performs a union::
+
+    >>> d1 = dict((i*10, chr(65+i)) for i in range(26))
+    >>> d2 = dict((i**.5, i) for i in range(1000))
+    >>> d1.viewkeys() & d2.viewkeys()
+    set([0.0, 10.0, 20.0, 30.0])
+    >>> d1.viewkeys() | range(0, 30)
+    set([0, 1, 130, 3, 4, 5, 6, ..., 120, 250])
+
 The view keeps track of the dictionary and its contents change as the
 dictionary is modified::
 
@@ -433,6 +560,58 @@
      Backported to 2.7 by Alexandre Vassalotti; :issue:`1967`.
 
 
+PEP 3137: The memoryview Object
+====================================================
+
+The :class:`memoryview` object provides a view of another object's
+memory content that matches the :class:`bytes` type's interface.
+
+    >>> import string
+    >>> m = memoryview(string.letters)
+    >>> m
+    <memory at 0x37f850>
+    >>> len(m)           # Returns length of underlying object
+    52
+    >>> m[0], m[25], m[26]   # Indexing returns one byte
+    ('a', 'z', 'A')
+    >>> m2 = m[0:26]         # Slicing returns another memoryview
+    >>> m2
+    <memory at 0x37f080>
+
+The content of the view can be converted to a string of bytes or
+a list of integers:
+
+    >>> m2.tobytes()
+    'abcdefghijklmnopqrstuvwxyz'
+    >>> m2.tolist()
+    [97, 98, 99, 100, 101, 102, 103, ... 121, 122]
+    >>>
+
+:class:`memoryview` objects allow modifying the underlying object if
+it's a mutable object.
+
+    >>> m2[0] = 75
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    TypeError: cannot modify read-only memory
+    >>> b = bytearray(string.letters)  # Creating a mutable object
+    >>> b
+    bytearray(b'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ')
+    >>> mb = memoryview(b)
+    >>> mb[0] = '*'         # Assign to view, changing the bytearray.
+    >>> b[0:5]              # The bytearray has been changed.
+    bytearray(b'*bcde')
+    >>>
+
+.. seealso::
+
+   :pep:`3137` - Immutable Bytes and Mutable Buffer
+     PEP written by Guido van Rossum.
+     Implemented by Travis Oliphant, Antoine Pitrou and others.
+     Backported to 2.7 by Antoine Pitrou; :issue:`2396`.
+
+
+
 Other Language Changes
 ======================
 
@@ -458,9 +637,9 @@
   3.x, generalizing list/generator comprehensions to use
   the literal syntax for sets and dictionaries.
 
-    >>> {x:x*x for x in range(6)}
+    >>> {x: x*x for x in range(6)}
     {0: 0, 1: 1, 2: 4, 3: 9, 4: 16, 5: 25}
-    >>> {'a'*x for x in range(6)}
+    >>> {('a'*x) for x in range(6)}
     set(['', 'a', 'aa', 'aaa', 'aaaa', 'aaaaa'])
 
   Backported by Alexandre Vassalotti; :issue:`2333`.
@@ -490,8 +669,8 @@
   in many different places: :func:`str` on
   floats and complex numbers; the :class:`float` and :class:`complex`
   constructors;
-  numeric formatting; serialization and
-  deserialization of floats and complex numbers using the
+  numeric formatting; serializing and
+  deserializing floats and complex numbers using the
   :mod:`marshal`, :mod:`pickle`
   and :mod:`json` modules;
   parsing of float and imaginary literals in Python code;
@@ -506,7 +685,7 @@
   .. maybe add an example?
 
   The rounding library responsible for this improvement works on
-  Windows, and on Unix platforms using the gcc, icc, or suncc
+  Windows and on Unix platforms using the gcc, icc, or suncc
   compilers.  There may be a small number of platforms where correct
   operation of this code cannot be guaranteed, so the code is not
   used on such systems.  You can find out which code is being used
@@ -516,6 +695,37 @@
   Implemented by Eric Smith and Mark Dickinson, using David Gay's
   :file:`dtoa.c` library; :issue:`7117`.
 
+* Conversions from long integers and regular integers to floating
+  point now round differently, returning the floating-point number
+  closest to the number.  This doesn't matter for small integers that
+  can be converted exactly, but for large numbers that will
+  unavoidably lose precision, Python 2.7 now approximates more
+  closely.  For example, Python 2.6 computed the following::
+
+    >>> n = 295147905179352891391
+    >>> float(n)
+    2.9514790517935283e+20
+    >>> n - long(float(n))
+    65535L
+
+  Python 2.7's floating-point result is larger, but much closer to the
+  true value::
+
+    >>> n = 295147905179352891391
+    >>> float(n)
+    2.9514790517935289e+20
+    >>> n - long(float(n))
+    -1L
+
+  (Implemented by Mark Dickinson; :issue:`3166`.)
+
+  Integer division is also more accurate in its rounding behaviours.  (Also
+  implemented by Mark Dickinson; :issue:`1811`.)
+
+* Implicit coercion for complex numbers has been removed; the interpreter
+  will no longer ever attempt to call a :meth:`__coerce__` method on complex
+  objects.  (Removed by Meador Inge and Mark Dickinson; :issue:`5211`.)
+
 * The :meth:`str.format` method now supports automatic numbering of the replacement
   fields.  This makes using :meth:`str.format` more closely resemble using
   ``%s`` formatting::
@@ -543,6 +753,16 @@
   so it will now produce 'INF' and 'NAN'.
   (Contributed by Eric Smith; :issue:`3382`.)
 
+  A low-level change: the :meth:`object.__format__` method now triggers
+  a :exc:`PendingDeprecationWarning` if it's passed a format string,
+  because the :meth:`__format__` method for :class:`object` converts
+  the object to a string representation and formats that.  Previously
+  the method silently applied the format string to the string
+  representation, but that could hide mistakes in Python code.  If
+  you're supplying formatting information such as an alignment or
+  precision, presumably you're expecting the formatting to be applied
+  in some object-specific way.  (Fixed by Eric Smith; :issue:`7994`.)
+
 * The :func:`int` and :func:`long` types gained a ``bit_length``
   method that returns the number of bits necessary to represent
   its argument in binary::
@@ -560,32 +780,11 @@
 
   (Contributed by Fredrik Johansson and Victor Stinner; :issue:`3439`.)
 
-* Conversions from long integers and regular integers to floating
-  point now round differently, returning the floating-point number
-  closest to the number.  This doesn't matter for small integers that
-  can be converted exactly, but for large numbers that will
-  unavoidably lose precision, Python 2.7 now approximates more
-  closely.  For example, Python 2.6 computed the following::
-
-    >>> n = 295147905179352891391
-    >>> float(n)
-    2.9514790517935283e+20
-    >>> n - long(float(n))
-    65535L
-
-  Python 2.7's floating-point result is larger, but much closer to the
-  true value::
-
-    >>> n = 295147905179352891391
-    >>> float(n)
-    2.9514790517935289e+20
-    >>> n - long(float(n))
-    -1L
-
-  (Implemented by Mark Dickinson; :issue:`3166`.)
-
-  Integer division is also more accurate in its rounding behaviours.  (Also
-  implemented by Mark Dickinson; :issue:`1811`.)
+* The :keyword:`import` statement will no longer try a relative import
+  if an absolute import (e.g. ``from .os import sep``) fails.  This
+  fixes a bug, but could possibly break certain :keyword:`import`
+  statements that were only working by accident.  (Fixed by Meador Inge;
+  :issue:`7902`.)
 
 * It's now possible for a subclass of the built-in :class:`unicode` type
   to override the :meth:`__unicode__` method.  (Implemented by
@@ -603,9 +802,15 @@
   (Contributed by Amaury Forgeot d'Arc, after a suggestion by
   George Sakkis; :issue:`5982`.)
 
-* A new encoding named "cp720", used primarily for Arabic text, is now
-  supported.  (Contributed by Alexander Belchenko and Amaury Forgeot
-  d'Arc; :issue:`1616979`.)
+* When a restricted set of attributes were set using ``__slots__``,
+  deleting an unset attribute would not raise :exc:`AttributeError`
+  as you would expect.  Fixed by Benjamin Peterson; :issue:`7604`.)
+
+* Two new encodings are now supported: "cp720", used primarily for
+  Arabic text; and "cp858", a variant of CP 850 that adds the euro
+  symbol.  (CP720 contributed by Alexander Belchenko and Amaury
+  Forgeot d'Arc in :issue:`1616979`; CP858 contributed by Tim Hatch in
+  :issue:`8016`.)
 
 * The :class:`file` object will now set the :attr:`filename` attribute
   on the :exc:`IOError` exception when trying to open a directory
@@ -615,7 +820,7 @@
   (fixed by Stefan Krah; :issue:`5677`).
 
 * The Python tokenizer now translates line endings itself, so the
-  :func:`compile` built-in function can now accept code using any
+  :func:`compile` built-in function now accepts code using any
   line-ending convention.  Additionally, it no longer requires that the
   code end in a newline.
 
@@ -648,13 +853,12 @@
 For example, the following setting will print warnings every time
 they occur, but turn warnings from the :mod:`Cookie` module into an
 error.  (The exact syntax for setting an environment variable varies
-across operating systems and shells, so it may be different for you.)
+across operating systems and shells.)
 
 ::
 
   export PYTHONWARNINGS=all,error:::Cookie:0
 
-
 .. ======================================================================
 
 
@@ -678,7 +882,7 @@
   any of them.  This would previously take quadratic
   time for garbage collection, but now the number of full garbage collections
   is reduced as the number of objects on the heap grows.
-  The new logic is to only perform a full garbage collection pass when
+  The new logic only performs a full garbage collection pass when
   the middle generation has been collected 10 times and when the
   number of survivor objects from the middle generation exceeds 10% of
   the number of objects in the oldest generation.  (Suggested by Martin
@@ -788,11 +992,11 @@
   The new version features better Python 3.x compatibility, various bug fixes,
   and adds several new BerkeleyDB flags and methods.
   (Updated by Jes??s Cea Avi??n; :issue:`8156`.  The pybsddb
-  changelog can be browsed at http://hg.jcea.es/pybsddb/file/tip/ChangeLog.)
+  changelog can be read at http://hg.jcea.es/pybsddb/file/tip/ChangeLog.)
 
 * The :mod:`bz2` module's :class:`~bz2.BZ2File` now supports the context
-  management protocol, so you can write ``with bz2.BZ2File(...) as f: ...``.
-  (Contributed by Hagen Fuerstenau; :issue:`3860`.)
+  management protocol, so you can write ``with bz2.BZ2File(...) as f:``.
+  (Contributed by Hagen F??rstenau; :issue:`3860`.)
 
 * New class: the :class:`~collections.Counter` class in the :mod:`collections`
   module is useful for tallying data.  :class:`~collections.Counter` instances
@@ -816,7 +1020,7 @@
      >>> c['z']
      0
 
-  There are three additional :class:`~collections.Counter` methods:
+  There are three additional :class:`~collections.Counter` methods.
   :meth:`~collections.Counter.most_common` returns the N most common
   elements and their counts.  :meth:`~collections.Counter.elements`
   returns an iterator over the contained elements, repeating each
@@ -843,12 +1047,20 @@
 
   .. revision 79660
 
-  The new :class:`~collections.OrderedDict` class is described in the earlier
+  New class: :class:`~collections.OrderedDict` is described in the earlier
   section :ref:`pep-0372`.
 
+  New method: The :class:`~collections.deque` data type now has a
+  :meth:`~collections.deque.count` method that returns the number of
+  contained elements equal to the supplied argument *x*, and a
+  :meth:`~collections.deque.reverse` method that reverses the elements
+  of the deque in-place.  :class:`deque` also exposes its maximum
+  length as the read-only :attr:`~collections.deque.maxlen` attribute.
+  (Both features added by Raymond Hettinger.)
+
   The :class:`~collections.namedtuple` class now has an optional *rename* parameter.
   If *rename* is true, field names that are invalid because they've
-  been repeated or that aren't legal Python identifiers will be
+  been repeated or aren't legal Python identifiers will be
   renamed to legal names that are derived from the field's
   position within the list of fields:
 
@@ -859,13 +1071,43 @@
 
   (Added by Raymond Hettinger; :issue:`1818`.)
 
-  The :class:`~collections.deque` data type now has a
-  :meth:`~collections.deque.count` method that returns the number of
-  contained elements equal to the supplied argument *x*, and a
-  :meth:`~collections.deque.reverse` method that reverses the elements
-  of the deque in-place.  :class:`deque` also exposes its maximum
-  length as the read-only :attr:`~collections.deque.maxlen` attribute.
-  (Both features added by Raymond Hettinger.)
+  Finally, the :class:`~collections.Mapping` abstract base class now
+  returns :const:`NotImplemented` if a mapping is compared to
+  another type that isn't a :class:`Mapping`.
+  (Fixed by Daniel Stutzbach; :issue:`8729`.)
+
+* Constructors for the parsing classes in the :mod:`ConfigParser` module now
+  take a *allow_no_value* parameter, defaulting to false; if true,
+  options without values will be allowed.  For example::
+
+    >>> import ConfigParser, StringIO
+    >>> sample_config = """
+    ... [mysqld]
+    ... user = mysql
+    ... pid-file = /var/run/mysqld/mysqld.pid
+    ... skip-bdb
+    ... """
+    >>> config = ConfigParser.RawConfigParser(allow_no_value=True)
+    >>> config.readfp(StringIO.StringIO(sample_config))
+    >>> config.get('mysqld', 'user')
+    'mysql'
+    >>> print config.get('mysqld', 'skip-bdb')
+    None
+    >>> print config.get('mysqld', 'unknown')
+    Traceback (most recent call last):
+      ...
+    NoOptionError: No option 'unknown' in section: 'mysqld'
+
+  (Contributed by Mats Kindahl; :issue:`7005`.)
+
+* Deprecated function: :func:`contextlib.nested`, which allows
+  handling more than one context manager with a single :keyword:`with`
+  statement, has been deprecated, because the :keyword:`with` statement
+  now supports multiple context managers.
+
+* The :mod:`cookielib` module now ignores cookies that have an invalid
+  version field, one that doesn't contain an integer value.  (Fixed by
+  John J. Lee; :issue:`3924`.)
 
 * The :mod:`copy` module's :func:`~copy.deepcopy` function will now
   correctly copy bound instance methods.  (Implemented by
@@ -885,7 +1127,7 @@
 * New method: the :class:`~decimal.Decimal` class gained a
   :meth:`~decimal.Decimal.from_float` class method that performs an exact
   conversion of a floating-point number to a :class:`~decimal.Decimal`.
-  Note that this is an **exact** conversion that strives for the
+  This exact conversion strives for the
   closest decimal approximation to the floating-point representation's value;
   the resulting decimal value will therefore still include the inaccuracy,
   if any.
@@ -893,27 +1135,57 @@
   ``Decimal('0.1000000000000000055511151231257827021181583404541015625')``.
   (Implemented by Raymond Hettinger; :issue:`4796`.)
 
-  Most of the methods of the :class:`~decimal.Context` class now accept integers
-  as well as :class:`~decimal.Decimal` instances; the only exceptions are the
-  :meth:`~decimal.Context.canonical` and :meth:`~decimal.Context.is_canonical`
-  methods.  (Patch by Juan Jos?? Conti; :issue:`7633`.)
+  Comparing instances of :class:`Decimal` with floating-point
+  numbers now produces sensible results based on the numeric values
+  of the operands.  Previously such comparisons would fall back to
+  Python's default rules for comparing objects, which produced arbitrary
+  results based on their type.  Note that you still cannot combine
+  :class:`Decimal` and floating-point in other operations such as addition,
+  since you should be explicitly choosing how to convert between float and
+  :class:`Decimal`.
+  (Fixed by Mark Dickinson; :issue:`2531`.)
 
   The constructor for :class:`~decimal.Decimal` now accepts
   floating-point numbers (added by Raymond Hettinger; :issue:`8257`)
   and non-European Unicode characters such as Arabic-Indic digits
   (contributed by Mark Dickinson; :issue:`6595`).
 
+  Most of the methods of the :class:`~decimal.Context` class now accept integers
+  as well as :class:`~decimal.Decimal` instances; the only exceptions are the
+  :meth:`~decimal.Context.canonical` and :meth:`~decimal.Context.is_canonical`
+  methods.  (Patch by Juan Jos?? Conti; :issue:`7633`.)
+
   When using :class:`~decimal.Decimal` instances with a string's
   :meth:`~str.format` method, the default alignment was previously
-  left-alignment.  This has been changed to right-alignment, which seems
+  left-alignment.  This has been changed to right-alignment, which is
   more sensible for numeric types.  (Changed by Mark Dickinson; :issue:`6857`.)
 
+  Comparisons involving a signaling NaN value (or ``sNAN``) now signal
+  :const:`InvalidOperation` instead of silently returning a true or
+  false value depending on the comparison operator.  Quiet NaN values
+  (or ``NaN``) are now hashable.  (Fixed by Mark Dickinson;
+  :issue:`7279`.)
+
 * The :mod:`difflib` module now produces output that is more
-  compatible with modern :command:`diff`/:command:`patch` tools thanks
-  to two changes: 1) the header giving the filename now uses a tab
-  character instead of spaces as a separator, and 2) the date format
-  used is now ISO-8601 style, ``2005-01-26 23:30:50``.  (Fixed by
-  Anatoly Techtonik; :issue:`7585`.)
+  compatible with modern :command:`diff`/:command:`patch` tools
+  through one small change, using a tab character instead of spaces as
+  a separator in the header giving the filename.  (Fixed by Anatoly
+  Techtonik; :issue:`7585`.)
+
+* The Distutils ``sdist`` command now always regenerates the
+  :file:`MANIFEST` file, since even if the :file:`MANIFEST.in` or
+  :file:`setup.py` files haven't been modified, the user might have
+  created some new files that should be included.
+  (Fixed by Tarek Ziad??; :issue:`8688`.)
+
+* The :mod:`doctest` module's :const:`IGNORE_EXCEPTION_DETAIL` flag
+  will now ignore the name of the module containing the exception
+  being tested.  (Patch by Lennart Regebro; :issue:`7490`.)
+
+* The :mod:`email` module's :class:`~email.message.Message` class will
+  now accept a Unicode-valued payload, automatically converting the
+  payload to the encoding specified by :attr:`output_charset`.
+  (Added by R. David Murray; :issue:`1368247`.)
 
 * The :class:`~fractions.Fraction` class now accepts a single float or
   :class:`~decimal.Decimal` instance, or two rational numbers, as
@@ -921,17 +1193,18 @@
   rationals added in :issue:`5812`, and float/decimal in
   :issue:`8294`.)
 
-  An oversight was fixed, making the :class:`Fraction` match the other
-  numeric types; ordering comparisons (``<``, ``<=``, ``>``, ``>=``) between
+  Ordering comparisons (``<``, ``<=``, ``>``, ``>=``) between
   fractions and complex numbers now raise a :exc:`TypeError`.
+  This fixes an oversight, making the :class:`Fraction` match the other
+  numeric types.
 
   .. revision 79455
 
-* New class: a new :class:`~ftplib.FTP_TLS` class in
+* New class: :class:`~ftplib.FTP_TLS` in
   the :mod:`ftplib` module provides secure FTP
   connections using TLS encapsulation of authentication as well as
   subsequent control and data transfers.
-  (Contributed by Giampaolo Rodola', :issue:`2054`.)
+  (Contributed by Giampaolo Rodola; :issue:`2054`.)
 
   The :meth:`~ftplib.FTP.storbinary` method for binary uploads can now restart
   uploads thanks to an added *rest* parameter (patch by Pablo Mouzo;
@@ -957,8 +1230,8 @@
   otherwise. (Contributed by Antoine Pitrou; :issue:`4688`.)
 
 * The :mod:`gzip` module's :class:`~gzip.GzipFile` now supports the context
-  management protocol, so you can write ``with gzip.GzipFile(...) as f: ...``
-  (contributed by Hagen Fuerstenau; :issue:`3860`), and it now implements
+  management protocol, so you can write ``with gzip.GzipFile(...) as f:``
+  (contributed by Hagen F??rstenau; :issue:`3860`), and it now implements
   the :class:`io.BufferedIOBase` ABC, so you can wrap it with
   :class:`io.BufferedReader` for faster processing
   (contributed by Nir Aides; :issue:`7471`).
@@ -973,18 +1246,25 @@
 * New attribute: the :mod:`hashlib` module now has an :attr:`~hashlib.hashlib.algorithms`
   attribute containing a tuple naming the supported algorithms.
   In Python 2.7, ``hashlib.algorithms`` contains
-  ``('md5', 'sha1', 'sha224', 'sha256', 'sha384', 'sha512')``
+  ``('md5', 'sha1', 'sha224', 'sha256', 'sha384', 'sha512')``.
   (Contributed by Carl Chenet; :issue:`7418`.)
 
 * The default :class:`~httplib.HTTPResponse` class used by the :mod:`httplib` module now
   supports buffering, resulting in much faster reading of HTTP responses.
-  (Contributed by Kristjan Valur Jonsson; :issue:`4879`.)
+  (Contributed by Kristj??n Valur J??nsson; :issue:`4879`.)
 
   The :class:`~httplib.HTTPConnection` and :class:`~httplib.HTTPSConnection` classes
   now support a *source_address* parameter, a ``(host, port)`` 2-tuple
   giving the source address that will be used for the connection.
   (Contributed by Eldon Ziegler; :issue:`3972`.)
 
+* The :mod:`ihooks` module now supports relative imports.  Note that
+  :mod:`ihooks` is an older module for customizing imports,
+  superseded by the :mod:`imputil` module added in Python 2.0.
+  (Relative import support added by Neil Schemenauer.)
+
+  .. revision 75423
+
 * The :mod:`imaplib` module now supports IPv6 addresses.
   (Contributed by Derek Morr; :issue:`1655`.)
 
@@ -997,9 +1277,9 @@
     >>> def f(a, b=1, *pos, **named):
     ...     pass
     >>> getcallargs(f, 1, 2, 3)
-    {'a': 1, 'named': {}, 'b': 2, 'pos': (3,)}
+    {'a': 1, 'b': 2, 'pos': (3,), 'named': {}}
     >>> getcallargs(f, a=2, x=4)
-    {'a': 2, 'named': {'x': 4}, 'b': 1, 'pos': ()}
+    {'a': 2, 'b': 1, 'pos': (), 'named': {'x': 4}}
     >>> getcallargs(f)
     Traceback (most recent call last):
     ...
@@ -1050,8 +1330,8 @@
   floats or :class:`~decimal.Decimal` instances.  (Implemented by Raymond
   Hettinger; :issue:`5032`.)
 
-  :func:`itertools.combinations` and :func:`itertools.product` were
-  previously raising :exc:`ValueError` for values of *r* larger than
+  :func:`itertools.combinations` and :func:`itertools.product`
+  previously raised :exc:`ValueError` for values of *r* larger than
   the input iterable.  This was deemed a specification error, so they
   now return an empty iterator.  (Fixed by Raymond Hettinger; :issue:`4816`.)
 
@@ -1065,6 +1345,12 @@
   with any object literal that decodes to a list of pairs.
   (Contributed by Raymond Hettinger; :issue:`5381`.)
 
+* The :mod:`mailbox` module's :class:`Maildir` class now records the
+  timestamp on the directories it reads, and only re-reads them if the
+  modification time has subsequently changed.  This improves
+  performance by avoiding unneeded directory scans.  (Fixed by
+  A.M. Kuchling and Antoine Pitrou; :issue:`1607951`, :issue:`6896`.)
+
 * New functions: the :mod:`math` module gained
   :func:`~math.erf` and :func:`~math.erfc` for the error function and the complementary error function,
   :func:`~math.expm1` which computes ``e**x - 1`` with more precision than
@@ -1095,7 +1381,8 @@
   real, effective, and saved GIDs and UIDs;
   :func:`~os.setresgid` and :func:`~os.setresuid`, which set
   real, effective, and saved GIDs and UIDs to new values;
-  :func:`~os.initgroups`.  (GID/UID functions
+  :func:`~os.initgroups`, which initialize the group access list
+  for the current process.  (GID/UID functions
   contributed by Travis H.; :issue:`6508`.  Support for initgroups added
   by Jean-Paul Calderone; :issue:`7333`.)
 
@@ -1117,6 +1404,20 @@
   now accept an optional *flags* argument, for consistency with the
   other functions in the module.  (Added by Gregory P. Smith.)
 
+* New function: :func:`~runpy.run_path` in the :mod:`runpy` module
+  will execute the code at a provided *path* argument.  *path* can be
+  the path of a Python source file (:file:`example.py`), a compiled
+  bytecode file (:file:`example.pyc`), a directory
+  (:file:`./package/`), or a zip archive (:file:`example.zip`).  If a
+  directory or zip path is provided, it will be added to the front of
+  ``sys.path`` and the module :mod:`__main__` will be imported.  It's
+  expected that the directory or zip contains a :file:`__main__.py`;
+  if it doesn't, some other :file:`__main__.py` might be imported from
+  a location later in ``sys.path``.  This makes more of the machinery
+  of :mod:`runpy` available to scripts that want to mimic the way
+  Python's command line processes an explicit path name.
+  (Added by Nick Coghlan; :issue:`6816`.)
+
 * New function: in the :mod:`shutil` module, :func:`~shutil.make_archive`
   takes a filename, archive type (zip or tar-format), and a directory
   path, and creates an archive containing the directory's contents.
@@ -1128,12 +1429,17 @@
   named pipes like a regular file by opening them for reading, and
   this would block indefinitely.  (Fixed by Antoine Pitrou; :issue:`3002`.)
 
+* The :mod:`signal` module no longer re-installs the signal handler
+  unless this is truly necessary, which fixes a bug that could make it
+  impossible to catch the EINTR signal robustly.  (Fixed by
+  Charles-Francois Natali; :issue:`8354`.)
+
 * New functions: in the :mod:`site` module, three new functions
   return various site- and user-specific paths.
   :func:`~site.getsitepackages` returns a list containing all
-  global site-packages directories, and
+  global site-packages directories,
   :func:`~site.getusersitepackages` returns the path of the user's
-  site-packages directory.
+  site-packages directory, and
   :func:`~site.getuserbase` returns the value of the :envvar:`USER_BASE`
   environment variable, giving the path to a directory that can be used
   to store data.
@@ -1144,20 +1450,7 @@
   catch and swallow the :exc:`KeyboardInterrupt` exception.  (Fixed by
   Victor Stinner; :issue:`3137`.)
 
-* The :mod:`socket` module's :class:`~ssl.SSL` objects now support the
-  buffer API, which fixed a test suite failure (fix by Antoine Pitrou;
-  :issue:`7133`).  :class:`SSL` objects also now automatically set
-  OpenSSL's :cmacro:`SSL_MODE_AUTO_RETRY`, which will prevent an error
-  code being returned from :meth:`recv` operations that trigger an SSL
-  renegotiation (fix by Antoine Pitrou; :issue:`8222`).
-
-  The version of OpenSSL being used is now available as the module
-  attributes :attr:`OPENSSL_VERSION` (a string),
-  :attr:`OPENSSL_VERSION_INFO` (a 5-tuple), and
-  :attr:`OPENSSL_VERSION_NUMBER` (an integer).  (Added by Antoine
-  Pitrou; :issue:`8321`.)
-
-  The :func:`~socket.create_connection` function
+* The :func:`~socket.create_connection` function
   gained a *source_address* parameter, a ``(host, port)`` 2-tuple
   giving the source address that will be used for the connection.
   (Contributed by Eldon Ziegler; :issue:`3972`.)
@@ -1168,11 +1461,16 @@
   Antoine Pitrou; :issue:`8104`.)
 
 * The :mod:`SocketServer` module's :class:`~SocketServer.TCPServer` class now
-  has a :attr:`~SocketServer.TCPServer.disable_nagle_algorithm` class attribute.
-  The default value is False; if overridden to be True,
+  supports socket timeouts and disabling the Nagle algorithm.
+  The :attr:`~SocketServer.TCPServer.disable_nagle_algorithm` class attribute
+  defaults to False; if overridden to be True,
   new request connections will have the TCP_NODELAY option set to
   prevent buffering many small sends into a single TCP packet.
-  (Contributed by Kristjan Valur Jonsson; :issue:`6192`.)
+  The :attr:`~SocketServer.TCPServer.timeout` class attribute can hold
+  a timeout in seconds that will be applied to the request socket; if
+  no request is received within that time, :meth:`handle_timeout`
+  will be called and :meth:`handle_request` will return.
+  (Contributed by Kristj??n Valur J??nsson; :issue:`6192` and :issue:`6267`.)
 
 * Updated module: the :mod:`sqlite3` module has been updated to
   version 2.6.0 of the `pysqlite package <http://code.google.com/p/pysqlite/>`__. Version 2.6.0 includes a number of bugfixes, and adds
@@ -1181,6 +1479,32 @@
   and then call :meth:`~sqlite3.Connection.load_extension` to load a particular shared library.
   (Updated by Gerhard H??ring.)
 
+* The :mod:`ssl` module's :class:`ssl.SSLSocket` objects now support the
+  buffer API, which fixed a test suite failure (fix by Antoine Pitrou;
+  :issue:`7133`) and automatically set
+  OpenSSL's :cmacro:`SSL_MODE_AUTO_RETRY`, which will prevent an error
+  code being returned from :meth:`recv` operations that trigger an SSL
+  renegotiation (fix by Antoine Pitrou; :issue:`8222`).
+
+  The :func:`ssl.wrap_socket` constructor function now takes a
+  *ciphers* argument that's a string listing the encryption algorithms
+  to be allowed; the format of the string is described
+  `in the OpenSSL documentation
+  <http://www.openssl.org/docs/apps/ciphers.html#CIPHER_LIST_FORMAT>`__.
+  (Added by Antoine Pitrou; :issue:`8322`.)
+
+  Another change makes the extension load all of OpenSSL's ciphers and
+  digest algorithms so that they're all available.  Some SSL
+  certificates couldn't be verified, reporting an "unknown algorithm"
+  error.  (Reported by Beda Kosata, and fixed by Antoine Pitrou;
+  :issue:`8484`.)
+
+  The version of OpenSSL being used is now available as the module
+  attributes :data:`ssl.OPENSSL_VERSION` (a string),
+  :data:`ssl.OPENSSL_VERSION_INFO` (a 5-tuple), and
+  :data:`ssl.OPENSSL_VERSION_NUMBER` (an integer).  (Added by Antoine
+  Pitrou; :issue:`8321`.)
+
 * The :mod:`struct` module will no longer silently ignore overflow
   errors when a value is too large for a particular integer format
   code (one of ``bBhHiIlLqQ``); it now always raises a
@@ -1216,6 +1540,10 @@
   false for ones that are implicitly global.
   (Contributed by Jeremy Hylton.)
 
+* The :mod:`syslog` module will now use the value of ``sys.argv[0]`` as the
+  identifier instead of the previous default value of ``'python'``.
+  (Changed by Sean Reifschneider; :issue:`8451`.)
+
 * The ``sys.version_info`` value is now a named tuple, with attributes
   named :attr:`major`, :attr:`minor`, :attr:`micro`,
   :attr:`releaselevel`, and :attr:`serial`.  (Contributed by Ross
@@ -1237,7 +1565,7 @@
 
   :mod:`tarfile` now supports filtering the :class:`~tarfile.TarInfo`
   objects being added to a tar file.  When you call :meth:`~tarfile.TarFile.add`,
-  instance, you may supply an optional *filter* argument
+  you may supply an optional *filter* argument
   that's a callable.  The *filter* callable will be passed the
   :class:`~tarfile.TarInfo` for every file being added, and can modify and return it.
   If the callable returns ``None``, the file will be excluded from the
@@ -1262,8 +1590,39 @@
   and has been updated to version 5.2.0 (updated by
   Florent Xicluna; :issue:`8024`).
 
-* The :class:`~UserDict.UserDict` class is now a new-style class.  (Changed by
-  Benjamin Peterson.)
+* The :mod:`urlparse` module's :func:`~urlparse.urlsplit` now handles
+  unknown URL schemes in a fashion compliant with :rfc:`3986`: if the
+  URL is of the form ``"<something>://..."``, the text before the
+  ``://`` is treated as the scheme, even if it's a made-up scheme that
+  the module doesn't know about.  This change may break code that
+  worked around the old behaviour.  For example, Python 2.6.4 or 2.5
+  will return the following:
+
+    >>> import urlparse
+    >>> urlparse.urlsplit('invented://host/filename?query')
+    ('invented', '', '//host/filename?query', '', '')
+
+  Python 2.7 (and Python 2.6.5) will return:
+
+    >>> import urlparse
+    >>> urlparse.urlsplit('invented://host/filename?query')
+    ('invented', 'host', '/filename?query', '', '')
+
+  (Python 2.7 actually produces slightly different output, since it
+  returns a named tuple instead of a standard tuple.)
+
+  The :mod:`urlparse` module also supports IPv6 literal addresses as defined by
+  :rfc:`2732` (contributed by Senthil Kumaran; :issue:`2987`). ::
+
+    >>> urlparse.urlparse('http://[1080::8:800:200C:417A]/foo')
+    ParseResult(scheme='http', netloc='[1080::8:800:200C:417A]',
+                path='/foo', params='', query='', fragment='')
+
+* New class: the :class:`~weakref.WeakSet` class in the :mod:`weakref`
+  module is a set that only holds weak references to its elements; elements
+  will be removed once there are no references pointing to them.
+  (Originally implemented in Python 3.x by Raymond Hettinger, and backported
+  to 2.7 by Michael Foord.)
 
 * The ElementTree library, :mod:`xml.etree`, no longer escapes
   ampersands and angle brackets when outputting an XML processing
@@ -1271,13 +1630,22 @@
   or comment (which looks like ``<!-- comment -->``).
   (Patch by Neil Muller; :issue:`2746`.)
 
+* The XML-RPC client and server, provided by the :mod:`xmlrpclib` and
+  :mod:`SimpleXMLRPCServer` modules, have improved performance by
+  supporting HTTP/1.1 keep-alive and by optionally using gzip encoding
+  to compress the XML being exchanged.  The gzip compression is
+  controlled by the :attr:`encode_threshold` attribute of
+  :class:`SimpleXMLRPCRequestHandler`, which contains a size in bytes;
+  responses larger than this will be compressed.
+  (Contributed by Kristj??n Valur J??nsson; :issue:`6267`.)
+
 * The :mod:`zipfile` module's :class:`~zipfile.ZipFile` now supports the context
-  management protocol, so you can write ``with zipfile.ZipFile(...) as f: ...``.
+  management protocol, so you can write ``with zipfile.ZipFile(...) as f:``.
   (Contributed by Brian Curtin; :issue:`5511`.)
 
-  :mod:`zipfile` now supports archiving empty directories and
+  :mod:`zipfile` now also supports archiving empty directories and
   extracts them correctly.  (Fixed by Kuba Wieczorek; :issue:`4710`.)
-  Reading files out of an archive is now faster, and interleaving
+  Reading files out of an archive is faster, and interleaving
   :meth:`~zipfile.ZipFile.read` and :meth:`~zipfile.ZipFile.readline` now works correctly.
   (Contributed by Nir Aides; :issue:`7610`.)
 
@@ -1291,36 +1659,157 @@
   :issue:`6003`.)
 
 
+.. ======================================================================
+.. whole new modules get described in subsections here
+
+
+.. _importlib-section:
+
+New module: importlib
+------------------------------
+
+Python 3.1 includes the :mod:`importlib` package, a re-implementation
+of the logic underlying Python's :keyword:`import` statement.
+:mod:`importlib` is useful for implementors of Python interpreters and
+to users who wish to write new importers that can participate in the
+import process.  Python 2.7 doesn't contain the complete
+:mod:`importlib` package, but instead has a tiny subset that contains
+a single function, :func:`~importlib.import_module`.
+
+``import_module(name, package=None)`` imports a module.  *name* is
+a string containing the module or package's name.  It's possible to do
+relative imports by providing a string that begins with a ``.``
+character, such as ``..utils.errors``.  For relative imports, the
+*package* argument must be provided and is the name of the package that
+will be used as the anchor for
+the relative import.  :func:`~importlib.import_module` both inserts the imported
+module into ``sys.modules`` and returns the module object.
+
+Here are some examples::
+
+    >>> from importlib import import_module
+    >>> anydbm = import_module('anydbm')  # Standard absolute import
+    >>> anydbm
+    <module 'anydbm' from '/p/python/Lib/anydbm.py'>
+    >>> # Relative import
+    >>> file_util = import_module('..file_util', 'distutils.command')
+    >>> file_util
+    <module 'distutils.file_util' from '/python/Lib/distutils/file_util.pyc'>
+
+:mod:`importlib` was implemented by Brett Cannon and introduced in
+Python 3.1.
+
+
 New module: sysconfig
 ---------------------------------
 
-XXX A new :mod:`sysconfig` module has been extracted from
-:mod:`distutils` and put in the standard library.
+The :mod:`sysconfig` module has been pulled out of the Distutils
+package, becoming a new top-level module in its own right.
+:mod:`sysconfig` provides functions for getting information about
+Python's build process: compiler switches, installation paths, the
+platform name, and whether Python is running from its source
+directory.
+
+Some of the functions in the module are:
+
+* :func:`~sysconfig.get_config_var` returns variables from Python's
+  Makefile and the :file:`pyconfig.h` file.
+* :func:`~sysconfig.get_config_vars` returns a dictionary containing
+  all of the configuration variables.
+* :func:`~sysconfig.getpath` returns the configured path for
+  a particular type of module: the standard library,
+  site-specific modules, platform-specific modules, etc.
+* :func:`~sysconfig.is_python_build` returns true if you're running a
+  binary from a Python source tree, and false otherwise.
+
+Consult the :mod:`sysconfig` documentation for more details and for
+a complete list of functions.
+
+The Distutils package and :mod:`sysconfig` are now maintained by Tarek
+Ziad??, who has also started a Distutils2 package (source repository at
+http://hg.python.org/distutils2/) for developing a next-generation
+version of Distutils.
 
-The :mod:`sysconfig` module provides access to Python's configuration
-information like the list of installation paths and the configuration
-variables relevant for the current platform. (contributed by Tarek)
 
-Updated module: ElementTree 1.3
----------------------------------
+ttk: Themed Widgets for Tk
+--------------------------
+
+Tcl/Tk 8.5 includes a set of themed widgets that re-implement basic Tk
+widgets but have a more customizable appearance and can therefore more
+closely resemble the native platform's widgets.  This widget
+set was originally called Tile, but was renamed to Ttk (for "themed Tk")
+on being added to Tcl/Tck release 8.5.
 
-XXX write this.
+To learn more, read the :mod:`ttk` module documentation.  You may also
+wish to read the Tcl/Tk manual page describing the
+Ttk theme engine, available at
+http://www.tcl.tk/man/tcl8.5/TkCmd/ttk_intro.htm. Some
+screenshots of the Python/Ttk code in use are at
+http://code.google.com/p/python-ttk/wiki/Screenshots.
 
-.. ======================================================================
-.. whole new modules get described in subsections here
+The :mod:`ttk` module was written by Guilherme Polo and added in
+:issue:`2983`.  An alternate version called ``Tile.py``, written by
+Martin Franklin and maintained by Kevin Walzer, was proposed for
+inclusion in :issue:`2618`, but the authors argued that Guilherme
+Polo's work was more comprehensive.
 
 
-Unit Testing Enhancements
+.. _unittest-section:
+
+Updated module: unittest
 ---------------------------------
 
-The :mod:`unittest` module was enhanced in several ways.
-The progress messages now shows 'x' for expected failures
+The :mod:`unittest` module was greatly enhanced; many
+new features were added.  Most of these features were implemented
+by Michael Foord, unless otherwise noted.  The enhanced version of
+the module is downloadable separately for use with Python versions 2.4 to 2.6,
+packaged as the :mod:`unittest2` package, from
+http://pypi.python.org/pypi/unittest2.
+
+When used from the command line, the module can automatically discover
+tests.  It's not as fancy as `py.test <http://pytest.org>`__ or
+`nose <http://code.google.com/p/python-nose/>`__, but provides a simple way
+to run tests kept within a set of package directories.  For example,
+the following command will search the :file:`test/` subdirectory for
+any importable test files named ``test*.py``::
+
+   python -m unittest discover -s test
+
+Consult the :mod:`unittest` module documentation for more details.
+(Developed in :issue:`6001`.)
+
+The :func:`main` function supports some other new options:
+
+* :option:`-b` or :option:`--buffer` will buffer the standard output
+  and standard error streams during each test.  If the test passes,
+  any resulting output will be discarded; on failure, the buffered
+  output will be displayed.
+
+* :option:`-c` or :option:`--catch` will cause the control-C interrupt
+  to be handled more gracefully.  Instead of interrupting the test
+  process immediately, the currently running test will be completed
+  and then the partial results up to the interruption will be reported.
+  If you're impatient, a second press of control-C will cause an immediate
+  interruption.
+
+  This control-C handler tries to avoid causing problems when the code
+  being tested or the tests being run have defined a signal handler of
+  their own, by noticing that a signal handler was already set and
+  calling it.  If this doesn't work for you, there's a
+  :func:`removeHandler` decorator that can be used to mark tests that
+  should have the control-C handling disabled.
+
+* :option:`-f` or :option:`--failfast` makes
+  test execution stop immediately when a test fails instead of
+  continuing to execute further tests.  (Suggested by Cliff Dyer and
+  implemented by Michael Foord; :issue:`8074`.)
+
+The progress messages now show 'x' for expected failures
 and 'u' for unexpected successes when run in verbose mode.
 (Contributed by Benjamin Peterson.)
-Test cases can raise the :exc:`~unittest.SkipTest` exception to skip a test.
-(:issue:`1034053`.)
 
-.. XXX describe test discovery (Contributed by Michael Foord; :issue:`6001`.)
+Test cases can raise the :exc:`~unittest.SkipTest` exception to skip a
+test (:issue:`1034053`).
 
 The error messages for :meth:`~unittest.TestCase.assertEqual`,
 :meth:`~unittest.TestCase.assertTrue`, and :meth:`~unittest.TestCase.assertFalse`
@@ -1330,7 +1819,7 @@
 provide will be printed for failures.  (Added by Michael Foord; :issue:`5663`.)
 
 The :meth:`~unittest.TestCase.assertRaises` method now
-return a context handler when called without providing a callable
+returns a context handler when called without providing a callable
 object to run.  For example, you can write this::
 
   with self.assertRaises(KeyError):
@@ -1350,7 +1839,7 @@
 
 The methods :meth:`~unittest.TestCase.addCleanup` and
 :meth:`~unittest.TestCase.doCleanups` were added.
-:meth:`~unittest.TestCase.addCleanup` allows you to add cleanup functions that
+:meth:`~unittest.TestCase.addCleanup` lets you add cleanup functions that
 will be called unconditionally (after :meth:`~unittest.TestCase.setUp` if
 :meth:`~unittest.TestCase.setUp` fails, otherwise after :meth:`~unittest.TestCase.tearDown`). This allows
 for much simpler resource allocation and deallocation during tests
@@ -1382,10 +1871,10 @@
   differences in the two strings.  This comparison is now used by
   default when Unicode strings are compared with :meth:`~unittest.TestCase.assertEqual`.
 
-* :meth:`~unittest.TestCase.assertRegexpMatches` checks whether its first argument is a
-  string matching a regular expression provided as its second argument.
-
-  .. XXX add assertNotRegexpMatches see issue 8038
+* :meth:`~unittest.TestCase.assertRegexpMatches` and
+  :meth:`~unittest.TestCase.assertNotRegexpMatches` checks whether the
+  first argument is a string matching or not matching the regular
+  expression provided as the second argument (:issue:`8038`).
 
 * :meth:`~unittest.TestCase.assertRaisesRegexp` checks whether a particular exception
   is raised, and then also checks that the string representation of
@@ -1414,9 +1903,10 @@
   all of the key/value pairs in *first* are found in *second*.
 
 * :meth:`~unittest.TestCase.assertAlmostEqual` and :meth:`~unittest.TestCase.assertNotAlmostEqual` test
-  whether *first* and *second* are approximately equal by computing
-  their difference, rounding the result to an optionally-specified number
-  of *places* (the default is 7), and comparing to zero.
+  whether *first* and *second* are approximately equal.  This method
+  can either round their difference to an optionally-specified number
+  of *places* (the default is 7) and compare it to zero, or require
+  the difference to be smaller than a supplied *delta* value.
 
 * :meth:`~unittest.TestLoader.loadTestsFromName` properly honors the
   :attr:`~unittest.TestLoader.suiteClass` attribute of
@@ -1428,20 +1918,13 @@
   objects being compared are of the specified type.  This function
   should compare the two objects and raise an exception if they don't
   match; it's a good idea for the function to provide additional
-  information about why the two objects are matching, much as the new
+  information about why the two objects aren't matching, much as the new
   sequence comparison methods do.
 
 :func:`unittest.main` now takes an optional ``exit`` argument.  If
-False, :func:`~unittest.main` doesn't call :func:`sys.exit`, allowing it to be
-used from the interactive interpreter. (Contributed by J. Pablo
-Fern??ndez; :issue:`3379`.)
-
-A new command-line switch, :option:`-f` or :option:`--failfast`, makes
-test execution stop immediately when a test fails instead of
-continuing to execute further tests.  (Suggested by Cliff Dyer and
-implemented by Michael Foord; :issue:`8074`.)
-
-.. XXX document the other new switches
+False, :func:`~unittest.main` doesn't call :func:`sys.exit`, allowing
+:func:`main` to be used from the interactive interpreter.
+(Contributed by J. Pablo Fern??ndez; :issue:`3379`.)
 
 :class:`~unittest.TestResult` has new :meth:`~unittest.TestResult.startTestRun` and
 :meth:`~unittest.TestResult.stopTestRun` methods that are called immediately before
@@ -1450,70 +1933,98 @@
 With all these changes, the :file:`unittest.py` was becoming awkwardly
 large, so the module was turned into a package and the code split into
 several files (by Benjamin Peterson).  This doesn't affect how the
-module is imported.
-
-
-.. _importlib-section:
-
-importlib: Importing Modules
-------------------------------
+module is imported or used.
 
-Python 3.1 includes the :mod:`importlib` package, a re-implementation
-of the logic underlying Python's :keyword:`import` statement.
-:mod:`importlib` is useful for implementors of Python interpreters and
-to users who wish to write new importers that can participate in the
-import process.  Python 2.7 doesn't contain the complete
-:mod:`importlib` package, but instead has a tiny subset that contains
-a single function, :func:`~importlib.import_module`.
-
-``import_module(name, package=None)`` imports a module.  *name* is
-a string containing the module or package's name.  It's possible to do
-relative imports by providing a string that begins with a ``.``
-character, such as ``..utils.errors``.  For relative imports, the
-*package* argument must be provided and is the name of the package that
-will be used as the anchor for
-the relative import.  :func:`~importlib.import_module` both inserts the imported
-module into ``sys.modules`` and returns the module object.
-
-Here are some examples::
-
-    >>> from importlib import import_module
-    >>> anydbm = import_module('anydbm')  # Standard absolute import
-    >>> anydbm
-    <module 'anydbm' from '/p/python/Lib/anydbm.py'>
-    >>> # Relative import
-    >>> sysconfig = import_module('..sysconfig', 'distutils.command')
-    >>> sysconfig
-    <module 'distutils.sysconfig' from '/p/python/Lib/distutils/sysconfig.pyc'>
-
-:mod:`importlib` was implemented by Brett Cannon and introduced in
-Python 3.1.
-
-
-ttk: Themed Widgets for Tk
---------------------------
-
-Tcl/Tk 8.5 includes a set of themed widgets that re-implement basic Tk
-widgets but have a more customizable appearance and can therefore more
-closely resemble the native platform's widgets.  This widget
-set was originally called Tile, but was renamed to Ttk (for "themed Tk")
-on being added to Tcl/Tck release 8.5.
+.. seealso::
 
-XXX write a brief discussion and an example here.
+  http://www.voidspace.org.uk/python/articles/unittest2.shtml
+    Describes the new features, how to use them, and the
+    rationale for various design decisions.  (By Michael Foord.)
 
-The :mod:`ttk` module was written by Guilherme Polo and added in
-:issue:`2983`.  An alternate version called ``Tile.py``, written by
-Martin Franklin and maintained by Kevin Walzer, was proposed for
-inclusion in :issue:`2618`, but the authors argued that Guilherme
-Polo's work was more comprehensive.
+.. _elementtree-section:
 
+Updated module: ElementTree 1.3
+---------------------------------
 
-Deprecations and Removals
-=========================
+The version of the ElementTree library included with Python was updated to
+version 1.3.  Some of the new features are:
 
-* :func:`contextlib.nested`, which allows handling more than one context manager
-  with one :keyword:`with` statement, has been deprecated; :keyword:`with`
-  supports multiple context managers syntactically now.
+* The various parsing functions now take a *parser* keyword argument
+  giving an :class:`XMLParser` instance that will
+  be used.  This makes it possible to override the file's internal encoding::
+
+    p = ET.XMLParser(encoding='utf-8')
+    t = ET.XML("""<root/>""", parser=p)
+
+  Errors in parsing XML now raise a :exc:`ParseError` exception, whose
+  instances have a :attr:`position` attribute
+  containing a (*line*, *column*) tuple giving the location of the problem.
+
+* ElementTree's code for converting trees to a string has been
+  significantly reworked, making it roughly twice as fast in many
+  cases.  The :class:`ElementTree` :meth:`write` and :class:`Element`
+  :meth:`write` methods now have a *method* parameter that can be
+  "xml" (the default), "html", or "text".  HTML mode will output empty
+  elements as ``<empty></empty>`` instead of ``<empty/>``, and text
+  mode will skip over elements and only output the text chunks.  If
+  you set the :attr:`tag` attribute of an element to ``None`` but
+  leave its children in place, the element will be omitted when the
+  tree is written out, so you don't need to do more extensive rearrangement
+  to remove a single element.
+
+  Namespace handling has also been improved.  All ``xmlns:<whatever>``
+  declarations are now output on the root element, not scattered throughout
+  the resulting XML.  You can set the default namespace for a tree
+  by setting the :attr:`default_namespace` attribute and can
+  register new prefixes with :meth:`register_namespace`.  In XML mode,
+  you can use the true/false *xml_declaration* parameter to suppress the
+  XML declaration.
+
+* New :class:`Element` method: :meth:`extend` appends the items from a
+  sequence to the element's children.  Elements themselves behave like
+  sequences, so it's easy to move children from one element to
+  another::
+
+    from xml.etree import ElementTree as ET
+
+    t = ET.XML("""<list>
+      <item>1</item> <item>2</item>  <item>3</item>
+    </list>""")
+    new = ET.XML('<root/>')
+    new.extend(t)
+
+    # Outputs <root><item>1</item>...</root>
+    print ET.tostring(new)
+
+* New :class:`Element` method: :meth:`iter` yields the children of the
+  element as a generator.  It's also possible to write ``for child in
+  elem:`` to loop over an element's children.  The existing method
+  :meth:`getiterator` is now deprecated, as is :meth:`getchildren`
+  which constructs and returns a list of children.
+
+* New :class:`Element` method: :meth:`itertext` yields all chunks of
+  text that are descendants of the element.  For example::
+
+    t = ET.XML("""<list>
+      <item>1</item> <item>2</item>  <item>3</item>
+    </list>""")
+
+    # Outputs ['\n  ', '1', ' ', '2', '  ', '3', '\n']
+    print list(t.itertext())
+
+* Deprecated: using an element as a Boolean (i.e., ``if elem:``) would
+  return true if the element had any children, or false if there were
+  no children.  This behaviour is confusing -- ``None`` is false, but
+  so is a childless element? -- so it will now trigger a
+  :exc:`FutureWarning`.  In your code, you should be explicit: write
+  ``len(elem) != 0`` if you're interested in the number of children,
+  or ``elem is not None``.
+
+Fredrik Lundh develops ElementTree and produced the 1.3 version;
+you can read his article describing 1.3 at
+http://effbot.org/zone/elementtree-13-intro.htm.
+Florent Xicluna updated the version included with
+Python, after discussions on python-dev and in :issue:`6472`.)
 
 .. ======================================================================
 
@@ -1528,9 +2039,9 @@
   <http://sourceware.org/gdb/current/onlinedocs/gdb/Python.html>`__.
   When you begin debugging an executable program P, GDB will look for
   a file named ``P-gdb.py`` and automatically read it.  Dave Malcolm
-  contributed a :file:`python-gdb.py` that adds a number of useful
-  commands when debugging Python itself.  For example, there are
-  ``py-up`` and ``py-down`` that go up or down one Python stack frame,
+  contributed a :file:`python-gdb.py` that adds a number of
+  commands useful when debugging Python itself.  For example,
+  ``py-up`` and ``py-down`` go up or down one Python stack frame,
   which usually corresponds to several C stack frames.  ``py-print``
   prints the value of a Python variable, and ``py-bt`` prints the
   Python stack trace.  (Added as a result of :issue:`8032`.)
@@ -1543,11 +2054,11 @@
 * :cfunc:`Py_AddPendingCall` is now thread-safe, letting any
   worker thread submit notifications to the main Python thread.  This
   is particularly useful for asynchronous IO operations.
-  (Contributed by Kristjan Valur Jonsson; :issue:`4293`.)
+  (Contributed by Kristj??n Valur J??nsson; :issue:`4293`.)
 
 * New function: :cfunc:`PyCode_NewEmpty` creates an empty code object;
   only the filename, function name, and first line number are required.
-  This is useful to extension modules that are attempting to
+  This is useful for extension modules that are attempting to
   construct a more useful traceback stack.  Previously such
   extensions needed to call :cfunc:`PyCode_New`, which had many
   more arguments.  (Added by Jeffrey Yasskin.)
@@ -1555,7 +2066,7 @@
 * New function: :cfunc:`PyErr_NewExceptionWithDoc` creates a new
   exception class, just as the existing :cfunc:`PyErr_NewException` does,
   but takes an extra ``char *`` argument containing the docstring for the
-  new exception class.  (Added by the 'lekma' user on the Python bug tracker;
+  new exception class.  (Added by 'lekma' on the Python bug tracker;
   :issue:`7033`.)
 
 * New function: :cfunc:`PyFrame_GetLineNumber` takes a frame object
@@ -1576,6 +2087,28 @@
   :cfunc:`PyOS_ascii_strtod` and :cfunc:`PyOS_ascii_atof` functions
   are now deprecated.
 
+* New function: :cfunc:`PySys_SetArgvEx` sets the value of
+  ``sys.argv`` and can optionally update ``sys.path`` to include the
+  directory containing the script named by ``sys.argv[0]`` depending
+  on the value of an *updatepath* parameter.
+
+  This function was added to close a security hole for applications
+  that embed Python.  The old function, :cfunc:`PySys_SetArgv`, would
+  always update ``sys.path``, and sometimes it would add the current
+  directory.  This meant that, if you ran an application embedding
+  Python in a directory controlled by someone else, attackers could
+  put a Trojan-horse module in the directory (say, a file named
+  :file:`os.py`) that your application would then import and run.
+
+  If you maintain a C/C++ application that embeds Python, check
+  whether you're calling :cfunc:`PySys_SetArgv` and carefully consider
+  whether the application should be using :cfunc:`PySys_SetArgvEx`
+  with *updatepath* set to false.
+
+  Security issue reported as `CVE-2008-5983
+  <http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2008-5983>`_;
+  discussed in :issue:`5753`, and fixed by Antoine Pitrou.
+
 * New macros: the Python header files now define the following macros:
   :cmacro:`Py_ISALNUM`,
   :cmacro:`Py_ISALPHA`,
@@ -1594,9 +2127,14 @@
 
   .. XXX these macros don't seem to be described in the c-api docs.
 
+* Removed function: :cmacro:`PyEval_CallObject` is now only available
+  as a macro.  A function version was being kept around to preserve
+  ABI linking compatibility, but that was in 1997; it can certainly be
+  deleted by now.  (Removed by Antoine Pitrou; :issue:`8276`.)
+
 * New format codes: the :cfunc:`PyFormat_FromString`,
-  :cfunc:`PyFormat_FromStringV`, and :cfunc:`PyErr_Format` now
-  accepts ``%lld`` and ``%llu`` format codes for displaying values of
+  :cfunc:`PyFormat_FromStringV`, and :cfunc:`PyErr_Format` functions now
+  accept ``%lld`` and ``%llu`` format codes for displaying
   C's :ctype:`long long` types.
   (Contributed by Mark Dickinson; :issue:`7228`.)
 
@@ -1610,7 +2148,7 @@
   ever release the lock, since the other threads weren't replicated,
   and the child process would no longer be able to perform imports.
 
-  Python 2.7 now acquires the import lock before performing an
+  Python 2.7 acquires the import lock before performing an
   :func:`os.fork`, and will also clean up any locks created using the
   :mod:`threading` module.  C extension modules that have internal
   locks, or that call :cfunc:`fork()` themselves, will not benefit
@@ -1623,6 +2161,12 @@
   being raised when an interpreter shuts down.
   (Patch by Adam Olsen; :issue:`1722344`.)
 
+* When using the :ctype:`PyMemberDef` structure to define attributes
+  of a type, Python will no longer let you try to delete or set a
+  :const:`T_STRING_INPLACE` attribute.
+
+  .. rev 79644
+
 * Global symbols defined by the :mod:`ctypes` module are now prefixed
   with ``Py``, or with ``_ctypes``.  (Implemented by Thomas
   Heller; :issue:`3102`.)
@@ -1631,15 +2175,15 @@
   building the :mod:`pyexpat` module to use the system Expat library.
   (Contributed by Arfrever Frehtes Taifersar Arahesis; :issue:`7609`.)
 
-* New configure option: compiling Python with the
+* New configure option: the
   :option:`--with-valgrind` option will now disable the pymalloc
   allocator, which is difficult for the Valgrind memory-error detector
   to analyze correctly.
   Valgrind will therefore be better at detecting memory leaks and
   overruns. (Contributed by James Henstridge; :issue:`2422`.)
 
-* New configure option: you can now supply no arguments to
-  :option:`--with-dbmliborder=` in order to build none of the various
+* New configure option: you can now supply an empty string to
+  :option:`--with-dbmliborder=` in order to disable all of the various
   DBM modules.  (Added by Arfrever Frehtes Taifersar Arahesis;
   :issue:`6491`.)
 
@@ -1660,6 +2204,52 @@
   Arfrever Frehtes Taifersar Arahesis; :issue:`6094`.)
 
 
+.. _whatsnew27-capsules:
+
+Capsules
+-------------------
+
+Python 3.1 adds a new C datatype, :ctype:`PyCapsule`, for providing a
+C API to an extension module.  A capsule is essentially the holder of
+a C ``void *`` pointer, and is made available as a module attribute; for
+example, the :mod:`socket` module's API is exposed as ``socket.CAPI``,
+and :mod:`unicodedata` exposes ``ucnhash_CAPI``.  Other extensions
+can import the module, access its dictionary to get the capsule
+object, and then get the ``void *`` pointer, which will usually point
+to an array of pointers to the module's various API functions.
+
+There is an existing data type already used for this,
+:ctype:`PyCObject`, but it doesn't provide type safety.  Evil code
+written in pure Python could cause a segmentation fault by taking a
+:ctype:`PyCObject` from module A and somehow substituting it for the
+:ctype:`PyCObject` in module B.   Capsules know their own name,
+and getting the pointer requires providing the name::
+
+   void *vtable;
+
+   if (!PyCapsule_IsValid(capsule, "mymodule.CAPI") {
+           PyErr_SetString(PyExc_ValueError, "argument type invalid");
+           return NULL;
+   }
+
+   vtable = PyCapsule_GetPointer(capsule, "mymodule.CAPI");
+
+You are assured that ``vtable`` points to whatever you're expecting.
+If a different capsule was passed in, :cfunc:`PyCapsule_IsValid` would
+detect the mismatched name and return false.  Refer to
+:ref:`using-capsules` for more information on using these objects.
+
+Python 2.7 now uses capsules internally to provide various
+extension-module APIs, but the :cfunc:`PyCObject_AsVoidPtr` was
+modified to handle capsules, preserving compile-time compatibility
+with the :ctype:`CObject` interface.  Use of
+:cfunc:`PyCObject_AsVoidPtr` will signal a
+:exc:`PendingDeprecationWarning`, which is silent by default.
+
+Implemented in Python 3.1 and backported to 2.7 by Larry Hastings;
+discussed in :issue:`5630`.
+
+
 .. ======================================================================
 
 Port-Specific Changes: Windows
@@ -1672,16 +2262,24 @@
   and :data:`LIBRARIES_ASSEMBLY_NAME_PREFIX`.
   (Contributed by David Cournapeau; :issue:`4365`.)
 
+* The :mod:`_winreg` module for accessing the registry now implements
+  the :func:`CreateKeyEx` and :func:`DeleteKeyEx` functions, extended
+  versions of previously-supported functions that take several extra
+  arguments.  The :func:`DisableReflectionKey`,
+  :func:`EnableReflectionKey`, and :func:`QueryReflectionKey` were also
+  tested and documented.
+  (Implemented by Brian Curtin: :issue:`7347`.)
+
 * The new :cfunc:`_beginthreadex` API is used to start threads, and
   the native thread-local storage functions are now used.
-  (Contributed by Kristjan Valur Jonsson; :issue:`3582`.)
+  (Contributed by Kristj??n Valur J??nsson; :issue:`3582`.)
 
 * The :func:`os.kill` function now works on Windows.  The signal value
   can be the constants :const:`CTRL_C_EVENT`,
-  :const:`CTRL_BREAK_EVENT`, or any integer.  The Control-C and
-  Control-Break keystroke events can be sent to subprocesses; any
-  other value will use the :cfunc:`TerminateProcess` API.
-  (Contributed by Miki Tebeka; :issue:`1220212`.)
+  :const:`CTRL_BREAK_EVENT`, or any integer.  The first two constants
+  will send Control-C and Control-Break keystroke events to
+  subprocesses; any other value will use the :cfunc:`TerminateProcess`
+  API.  (Contributed by Miki Tebeka; :issue:`1220212`.)
 
 * The :func:`os.listdir` function now correctly fails
   for an empty path.  (Fixed by Hirokazu Yamamoto; :issue:`5913`.)
@@ -1700,18 +2298,29 @@
   installation and a user-installed copy of the same version.
   (Changed by Ronald Oussoren; :issue:`4865`.)
 
+Port-Specific Changes: FreeBSD
+-----------------------------------
+
+* FreeBSD 7.1's :const:`SO_SETFIB` constant, used with
+  :func:`~socket.getsockopt`/:func:`~socket.setsockopt` to select an
+  alternate routing table, is now available in the :mod:`socket`
+  module.  (Added by Kyle VanderBeek; :issue:`8235`.)
 
 Other Changes and Fixes
 =======================
 
 * Two benchmark scripts, :file:`iobench` and :file:`ccbench`, were
   added to the :file:`Tools` directory.  :file:`iobench` measures the
-  speed of built-in file I/O objects (as returned by :func:`open`)
+  speed of the built-in file I/O objects returned by :func:`open`
   while performing various operations, and :file:`ccbench` is a
   concurrency benchmark that tries to measure computing throughput,
   thread switching latency, and IO processing bandwidth when
   performing several tasks using a varying number of threads.
 
+* The :file:`Tools/i18n/msgfmt.py` script now understands plural
+  forms in :file:`.po` files.  (Fixed by Martin von L??wis;
+  :issue:`5464`.)
+
 * When importing a module from a :file:`.pyc` or :file:`.pyo` file
   with an existing :file:`.py` counterpart, the :attr:`co_filename`
   attributes of the resulting code objects are overwritten when the
@@ -1747,14 +2356,13 @@
 This section lists previously described changes and other bugfixes
 that may require changes to your code:
 
-* When using :class:`Decimal` instances with a string's
-  :meth:`format` method, the default alignment was previously
-  left-alignment.  This has been changed to right-alignment, which might
-  change the output of your programs.
-  (Changed by Mark Dickinson; :issue:`6857`.)
+* The :func:`range` function processes its arguments more
+  consistently; it will now call :meth:`__int__` on non-float,
+  non-integer arguments that are supplied to it.  (Fixed by Alexander
+  Belopolsky; :issue:`1533`.)
 
-  Another :meth:`format`-related change: the default precision used
-  for floating-point and complex numbers was changed from 6 decimal
+* The string :meth:`format` method changed the default precision used
+  for floating-point and complex numbers from 6 decimal
   places to 12, which matches the precision used by :func:`str`.
   (Changed by Eric Smith; :issue:`5920`.)
 
@@ -1764,18 +2372,79 @@
   affects new-style classes (derived from :class:`object`) and C extension
   types.  (:issue:`6101`.)
 
-* The :meth:`readline` method of :class:`StringIO` objects now does
-  nothing when a negative length is requested, as other file-like
-  objects do.  (:issue:`7348`).
+* Due to a bug in Python 2.6, the *exc_value* parameter to
+  :meth:`__exit__` methods was often the string representation of the
+  exception, not an instance.  This was fixed in 2.7, so *exc_value*
+  will be an instance as expected.  (Fixed by Florent Xicluna;
+  :issue:`7853`.)
+
+* When a restricted set of attributes were set using ``__slots__``,
+  deleting an unset attribute would not raise :exc:`AttributeError`
+  as you would expect.  Fixed by Benjamin Peterson; :issue:`7604`.)
 
 In the standard library:
 
+* Operations with :class:`datetime` instances that resulted in a year
+  falling outside the supported range didn't always raise
+  :exc:`OverflowError`.  Such errors are now checked more carefully
+  and will now raise the exception. (Reported by Mark Leander, patch
+  by Anand B. Pillai and Alexander Belopolsky; :issue:`7150`.)
+
+* When using :class:`Decimal` instances with a string's
+  :meth:`format` method, the default alignment was previously
+  left-alignment.  This has been changed to right-alignment, which might
+  change the output of your programs.
+  (Changed by Mark Dickinson; :issue:`6857`.)
+
+  Comparisons involving a signaling NaN value (or ``sNAN``) now signal
+  :const:`InvalidOperation` instead of silently returning a true or
+  false value depending on the comparison operator.  Quiet NaN values
+  (or ``NaN``) are now hashable.  (Fixed by Mark Dickinson;
+  :issue:`7279`.)
+
 * The ElementTree library, :mod:`xml.etree`, no longer escapes
   ampersands and angle brackets when outputting an XML processing
   instruction (which looks like `<?xml-stylesheet href="#style1"?>`)
   or comment (which looks like `<!-- comment -->`).
   (Patch by Neil Muller; :issue:`2746`.)
 
+* The :meth:`readline` method of :class:`StringIO` objects now does
+  nothing when a negative length is requested, as other file-like
+  objects do.  (:issue:`7348`).
+
+* The :mod:`syslog` module will now use the value of ``sys.argv[0]`` as the
+  identifier instead of the previous default value of ``'python'``.
+  (Changed by Sean Reifschneider; :issue:`8451`.)
+
+* The :mod:`tarfile` module's default error handling has changed, to
+  no longer suppress fatal errors.  The default error level was previously 0,
+  which meant that errors would only result in a message being written to the
+  debug log, but because the debug log is not activated by default,
+  these errors go unnoticed.  The default error level is now 1,
+  which raises an exception if there's an error.
+  (Changed by Lars Gust??bel; :issue:`7357`.)
+
+* The :mod:`urlparse` module's :func:`~urlparse.urlsplit` now handles
+  unknown URL schemes in a fashion compliant with :rfc:`3986`: if the
+  URL is of the form ``"<something>://..."``, the text before the
+  ``://`` is treated as the scheme, even if it's a made-up scheme that
+  the module doesn't know about.  This change may break code that
+  worked around the old behaviour.  For example, Python 2.6.4 or 2.5
+  will return the following:
+
+    >>> import urlparse
+    >>> urlparse.urlsplit('invented://host/filename?query')
+    ('invented', '', '//host/filename?query', '', '')
+
+  Python 2.7 (and Python 2.6.5) will return:
+
+    >>> import urlparse
+    >>> urlparse.urlsplit('invented://host/filename?query')
+    ('invented', 'host', '/filename?query', '', '')
+
+  (Python 2.7 actually produces slightly different output, since it
+  returns a named tuple instead of a standard tuple.)
+
 For C extensions:
 
 * C extensions that use integer format codes with the ``PyArg_Parse*``
@@ -1786,6 +2455,14 @@
   :cfunc:`PyOS_ascii_strtod` and :cfunc:`PyOS_ascii_atof` functions,
   which are now deprecated.
 
+For applications that embed Python:
+
+* The :cfunc:`PySys_SetArgvEx` function was added, letting
+  applications close a security hole when the existing
+  :cfunc:`PySys_SetArgv` function was used.  Check whether you're
+  calling :cfunc:`PySys_SetArgv` and carefully consider whether the
+  application should be using :cfunc:`PySys_SetArgvEx` with
+  *updatepath* set to false.
 
 .. ======================================================================
 
@@ -1797,5 +2474,6 @@
 
 The author would like to thank the following people for offering
 suggestions, corrections and assistance with various drafts of this
-article: Ryan Lovett, R. David Murray, Hugh Secker-Walker.
+article: Nick Coghlan, Philip Jenvey, Ryan Lovett, R. David Murray,
+Hugh Secker-Walker.
 

Modified: python/branches/py3k-cdecimal/Doc/whatsnew/3.0.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/whatsnew/3.0.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/whatsnew/3.0.rst	Sat Jul 17 12:36:09 2010
@@ -798,6 +798,15 @@
 * A new built-in function :func:`next` was added to call the
   :meth:`__next__` method on an object.
 
+* The :func:`round` function rounding strategy and return type have
+  changed.  Exact halfway cases are now rounded to the nearest even
+  result instead of away from zero.  (For example, ``round(2.5)`` now
+  returns ``2`` rather than ``3``.)  :func:`round(x[, n])` now
+  delegates to ``x.__round__([n])`` instead of always returning a
+  float.  It generally returns an integer when called with a single
+  argument and a value of the same type as ``x`` when called with two
+  arguments.
+
 * Moved :func:`intern` to :func:`sys.intern`.
 
 * Removed: :func:`apply`.  Instead of ``apply(f, args)`` use

Modified: python/branches/py3k-cdecimal/Doc/whatsnew/3.2.rst
==============================================================================
--- python/branches/py3k-cdecimal/Doc/whatsnew/3.2.rst	(original)
+++ python/branches/py3k-cdecimal/Doc/whatsnew/3.2.rst	Sat Jul 17 12:36:09 2010
@@ -173,7 +173,8 @@
 
 * bytearray objects cannot be used anymore as filenames: convert them to bytes
 
-* "t#" format of PyArg_Parse*() functions has been removed: use "s#" or "s*"
-  instead
+* PyArg_Parse*() functions:
+
+  * "t#" format has been removed: use "s#" or "s*" instead
+  * "w" and "w#" formats has been removed: use "w*" instead
 
-* Stub

Modified: python/branches/py3k-cdecimal/Grammar/Grammar
==============================================================================
--- python/branches/py3k-cdecimal/Grammar/Grammar	(original)
+++ python/branches/py3k-cdecimal/Grammar/Grammar	Sat Jul 17 12:36:09 2010
@@ -11,9 +11,9 @@
 # "How to Change Python's Grammar"
 
 # Start symbols for the grammar:
-#	single_input is a single interactive statement;
-#	file_input is a module or sequence of commands read from an input file;
-#	eval_input is the input for the eval() and input() functions.
+#       single_input is a single interactive statement;
+#       file_input is a module or sequence of commands read from an input file;
+#       eval_input is the input for the eval() and input() functions.
 # NB: compound_stmt in single_input is followed by extra NEWLINE!
 single_input: NEWLINE | simple_stmt | compound_stmt NEWLINE
 file_input: (NEWLINE | stmt)* ENDMARKER
@@ -24,13 +24,13 @@
 decorated: decorators (classdef | funcdef)
 funcdef: 'def' NAME parameters ['->' test] ':' suite
 parameters: '(' [typedargslist] ')'
-typedargslist: ((tfpdef ['=' test] ',')*
-                ('*' [tfpdef] (',' tfpdef ['=' test])* [',' '**' tfpdef] | '**' tfpdef)
-                | tfpdef ['=' test] (',' tfpdef ['=' test])* [','])
+typedargslist: (tfpdef ['=' test] (',' tfpdef ['=' test])* [','
+       ['*' [tfpdef] (',' tfpdef ['=' test])* [',' '**' tfpdef] | '**' tfpdef]]
+     |  '*' [tfpdef] (',' tfpdef ['=' test])* [',' '**' tfpdef] | '**' tfpdef)
 tfpdef: NAME [':' test]
-varargslist: ((vfpdef ['=' test] ',')*
-              ('*' [vfpdef] (',' vfpdef ['=' test])*  [',' '**' vfpdef] | '**' vfpdef)
-              | vfpdef ['=' test] (',' vfpdef ['=' test])* [','])
+varargslist: (vfpdef ['=' test] (',' vfpdef ['=' test])* [','
+       ['*' [vfpdef] (',' vfpdef ['=' test])* [',' '**' vfpdef] | '**' vfpdef]]
+     |  '*' [vfpdef] (',' vfpdef ['=' test])* [',' '**' vfpdef] | '**' vfpdef)
 vfpdef: NAME
 
 stmt: simple_stmt | compound_stmt
@@ -71,9 +71,9 @@
 for_stmt: 'for' exprlist 'in' testlist ':' suite ['else' ':' suite]
 try_stmt: ('try' ':' suite
            ((except_clause ':' suite)+
-	    ['else' ':' suite]
-	    ['finally' ':' suite] |
-	   'finally' ':' suite))
+            ['else' ':' suite]
+            ['finally' ':' suite] |
+           'finally' ':' suite))
 with_stmt: 'with' with_item (',' with_item)*  ':' suite
 with_item: test ['as' expr]
 # NB compile.c makes sure that the default except clause is last

Modified: python/branches/py3k-cdecimal/Include/longobject.h
==============================================================================
--- python/branches/py3k-cdecimal/Include/longobject.h	(original)
+++ python/branches/py3k-cdecimal/Include/longobject.h	Sat Jul 17 12:36:09 2010
@@ -58,7 +58,7 @@
 #error "sizeof(pid_t) is neither sizeof(int), sizeof(long) or sizeof(long long)"
 #endif /* SIZEOF_PID_T */
 
-/* For use by intobject.c only */
+/* Used by Python/mystrtoul.c. */
 PyAPI_DATA(unsigned char) _PyLong_DigitValue[256];
 
 /* _PyLong_Frexp returns a double x and an exponent e such that the

Modified: python/branches/py3k-cdecimal/Include/pyctype.h
==============================================================================
--- python/branches/py3k-cdecimal/Include/pyctype.h	(original)
+++ python/branches/py3k-cdecimal/Include/pyctype.h	Sat Jul 17 12:36:09 2010
@@ -11,6 +11,9 @@
 
 extern const unsigned int _Py_ctype_table[256];
 
+/* Unlike their C counterparts, the following macros are not meant to
+ * handle an int with any of the values [EOF, 0-UCHAR_MAX]. The argument
+ * must be a signed/unsigned char. */
 #define Py_ISLOWER(c)  (_Py_ctype_table[Py_CHARMASK(c)] & PY_CTF_LOWER)
 #define Py_ISUPPER(c)  (_Py_ctype_table[Py_CHARMASK(c)] & PY_CTF_UPPER)
 #define Py_ISALPHA(c)  (_Py_ctype_table[Py_CHARMASK(c)] & PY_CTF_ALPHA)

Modified: python/branches/py3k-cdecimal/Include/structseq.h
==============================================================================
--- python/branches/py3k-cdecimal/Include/structseq.h	(original)
+++ python/branches/py3k-cdecimal/Include/structseq.h	Sat Jul 17 12:36:09 2010
@@ -26,17 +26,12 @@
 
 PyAPI_FUNC(PyObject *) PyStructSequence_New(PyTypeObject* type);
 
-typedef struct {
-    PyObject_VAR_HEAD
-    PyObject *ob_item[1];
-} PyStructSequence;
+typedef PyTupleObject PyStructSequence;
 
 /* Macro, *only* to be used to fill in brand new objects */
-#define PyStructSequence_SET_ITEM(op, i, v) \
-    (((PyStructSequence *)(op))->ob_item[i] = v)
+#define PyStructSequence_SET_ITEM(op, i, v) PyTuple_SET_ITEM(op, i, v)
 
-#define PyStructSequence_GET_ITEM(op, i) \
-    (((PyStructSequence *)(op))->ob_item[i])
+#define PyStructSequence_GET_ITEM(op, i) PyTuple_GET_ITEM(op, i)
 
 
 #ifdef __cplusplus

Modified: python/branches/py3k-cdecimal/Include/traceback.h
==============================================================================
--- python/branches/py3k-cdecimal/Include/traceback.h	(original)
+++ python/branches/py3k-cdecimal/Include/traceback.h	Sat Jul 17 12:36:09 2010
@@ -10,16 +10,16 @@
 /* Traceback interface */
 
 typedef struct _traceback {
-	PyObject_HEAD
-	struct _traceback *tb_next;
-	struct _frame *tb_frame;
-	int tb_lasti;
-	int tb_lineno;
+    PyObject_HEAD
+    struct _traceback *tb_next;
+    struct _frame *tb_frame;
+    int tb_lasti;
+    int tb_lineno;
 } PyTracebackObject;
 
 PyAPI_FUNC(int) PyTraceBack_Here(struct _frame *);
 PyAPI_FUNC(int) PyTraceBack_Print(PyObject *, PyObject *);
-PyAPI_FUNC(int) _Py_DisplaySourceLine(PyObject *, const char *, int, int);
+PyAPI_FUNC(int) _Py_DisplaySourceLine(PyObject *, PyObject *, int, int);
 
 /* Reveal traceback type so we can typecheck traceback objects */
 PyAPI_DATA(PyTypeObject) PyTraceBack_Type;

Modified: python/branches/py3k-cdecimal/Include/unicodeobject.h
==============================================================================
--- python/branches/py3k-cdecimal/Include/unicodeobject.h	(original)
+++ python/branches/py3k-cdecimal/Include/unicodeobject.h	Sat Jul 17 12:36:09 2010
@@ -498,14 +498,14 @@
     Py_ssize_t size             /* size of buffer */
     );
 
-/* Similar to PyUnicode_FromUnicode(), but u points to Latin-1 encoded bytes */
+/* Similar to PyUnicode_FromUnicode(), but u points to UTF-8 encoded bytes */
 PyAPI_FUNC(PyObject*) PyUnicode_FromStringAndSize(
     const char *u,        /* char buffer */
     Py_ssize_t size       /* size of buffer */
     );
 
 /* Similar to PyUnicode_FromUnicode(), but u points to null-terminated
-   Latin-1 encoded bytes */
+   UTF-8 encoded bytes */
 PyAPI_FUNC(PyObject*) PyUnicode_FromString(
     const char *u        /* string */
     );
@@ -550,7 +550,7 @@
 
    Coercion is done in the following way:
 
-   1. String and other char buffer compatible objects are decoded
+   1. bytes, bytearray and other char buffer compatible objects are decoded
       under the assumptions that they contain data using the current
       default encoding. Decoding is done in "strict" mode.
 
@@ -574,7 +574,7 @@
    Unicode objects are passed back as-is (subclasses are converted to
    true Unicode objects), all other objects are delegated to
    PyUnicode_FromEncodedObject(obj, NULL, "strict") which results in
-   using the default encoding as basis for decoding the object.
+   using UTF-8 encoding as basis for decoding the object.
 
    The API returns NULL in case of an error. The caller is responsible
    for decref'ing the returned objects.
@@ -606,7 +606,7 @@
 
 #ifdef HAVE_WCHAR_H
 
-/* Create a Unicode Object from the whcar_t buffer w of the given
+/* Create a Unicode Object from the wchar_t buffer w of the given
    size.
 
    The buffer is copied into the new object. */
@@ -665,7 +665,7 @@
    parameters encoding and errors have the same semantics as the ones
    of the builtin unicode() API.
 
-   Setting encoding to NULL causes the default encoding to be used.
+   Setting encoding to NULL causes the default encoding (UTF-8) to be used.
 
    Error handling is set by errors which may also be set to NULL
    meaning to use the default handling defined for the codec. Default

Modified: python/branches/py3k-cdecimal/Lib/_abcoll.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/_abcoll.py	(original)
+++ python/branches/py3k-cdecimal/Lib/_abcoll.py	Sat Jul 17 12:36:09 2010
@@ -480,7 +480,15 @@
         except KeyError:
             pass
 
-    def update(self, other=(), **kwds):
+    def update(*args, **kwds):
+        if len(args) > 2:
+            raise TypeError("update() takes at most 2 positional "
+                            "arguments ({} given)".format(len(args)))
+        elif not args:
+            raise TypeError("update() takes at least 1 argument (0 given)")
+        self = args[0]
+        other = args[1] if len(args) >= 2 else ()
+
         if isinstance(other, Mapping):
             for key in other:
                 self[key] = other[key]

Modified: python/branches/py3k-cdecimal/Lib/_strptime.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/_strptime.py	(original)
+++ python/branches/py3k-cdecimal/Lib/_strptime.py	Sat Jul 17 12:36:09 2010
@@ -16,7 +16,9 @@
 from re import compile as re_compile
 from re import IGNORECASE, ASCII
 from re import escape as re_escape
-from datetime import date as datetime_date
+from datetime import (date as datetime_date,
+                      timedelta as datetime_timedelta,
+                      timezone as datetime_timezone)
 try:
     from _thread import allocate_lock as _thread_allocate_lock
 except:
@@ -204,6 +206,7 @@
             #XXX: Does 'Y' need to worry about having less or more than
             #     4 digits?
             'Y': r"(?P<Y>\d\d\d\d)",
+            'z': r"(?P<z>[+-]\d\d[0-5]\d)",
             'A': self.__seqToRE(self.locale_time.f_weekday, 'A'),
             'a': self.__seqToRE(self.locale_time.a_weekday, 'a'),
             'B': self.__seqToRE(self.locale_time.f_month[1:], 'B'),
@@ -293,7 +296,9 @@
 
 
 def _strptime(data_string, format="%a %b %d %H:%M:%S %Y"):
-    """Return a time struct based on the input string and the format string."""
+    """Return a 2-tuple consisting of a time struct and an int containing
+    the number of microseconds based on the input string and the
+    format string."""
 
     for index, arg in enumerate([data_string, format]):
         if not isinstance(arg, str):
@@ -333,10 +338,12 @@
     if len(data_string) != found.end():
         raise ValueError("unconverted data remains: %s" %
                           data_string[found.end():])
+
     year = 1900
     month = day = 1
     hour = minute = second = fraction = 0
     tz = -1
+    tzoffset = None
     # Default to -1 to signify that values not known; not critical to have,
     # though
     week_of_year = -1
@@ -417,6 +424,11 @@
             else:
                 # W starts week on Monday.
                 week_of_year_start = 0
+        elif group_key == 'z':
+            z = found_dict['z']
+            tzoffset = int(z[1:3]) * 60 + int(z[3:5])
+            if z.startswith("-"):
+                tzoffset = -tzoffset
         elif group_key == 'Z':
             # Since -1 is default value only need to worry about setting tz if
             # it can be something other than -1.
@@ -453,9 +465,35 @@
         day = datetime_result.day
     if weekday == -1:
         weekday = datetime_date(year, month, day).weekday()
-    return (time.struct_time((year, month, day,
-                              hour, minute, second,
-                              weekday, julian, tz)), fraction)
+    # Add timezone info
+    tzname = found_dict.get("Z")
+    if tzoffset is not None:
+        gmtoff = tzoffset * 60
+    else:
+        gmtoff = None
+
+    return (year, month, day,
+            hour, minute, second,
+            weekday, julian, tz, gmtoff, tzname), fraction
 
 def _strptime_time(data_string, format="%a %b %d %H:%M:%S %Y"):
-    return _strptime(data_string, format)[0]
+    """Return a time struct based on the input string and the
+    format string."""
+    tt = _strptime(data_string, format)[0]
+    return time.struct_time(tt[:9])
+
+def _strptime_datetime(cls, data_string, format="%a %b %d %H:%M:%S %Y"):
+    """Return a class cls instance based on the input string and the
+    format string."""
+    tt, fraction = _strptime(data_string, format)
+    gmtoff, tzname = tt[-2:]
+    args = tt[:6] + (fraction,)
+    if gmtoff is not None:
+        tzdelta = datetime_timedelta(seconds=gmtoff)
+        if tzname:
+            tz = datetime_timezone(tzdelta, tzname)
+        else:
+            tz = datetime_timezone(tzdelta)
+        args += (tz,)
+
+    return cls(*args)

Modified: python/branches/py3k-cdecimal/Lib/ast.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/ast.py	(original)
+++ python/branches/py3k-cdecimal/Lib/ast.py	Sat Jul 17 12:36:09 2010
@@ -50,7 +50,7 @@
     if isinstance(node_or_string, Expression):
         node_or_string = node_or_string.body
     def _convert(node):
-        if isinstance(node, Str):
+        if isinstance(node, (Str, Bytes)):
             return node.s
         elif isinstance(node, Num):
             return node.n
@@ -58,6 +58,8 @@
             return tuple(map(_convert, node.elts))
         elif isinstance(node, List):
             return list(map(_convert, node.elts))
+        elif isinstance(node, Set):
+            return set(map(_convert, node.elts))
         elif isinstance(node, Dict):
             return dict((_convert(k), _convert(v)) for k, v
                         in zip(node.keys, node.values))

Modified: python/branches/py3k-cdecimal/Lib/contextlib.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/contextlib.py	(original)
+++ python/branches/py3k-cdecimal/Lib/contextlib.py	Sat Jul 17 12:36:09 2010
@@ -4,9 +4,20 @@
 from functools import wraps
 from warnings import warn
 
-__all__ = ["contextmanager", "closing"]
+__all__ = ["contextmanager", "closing", "ContextDecorator"]
 
-class GeneratorContextManager(object):
+
+class ContextDecorator(object):
+    "A base class or mixin that enables context managers to work as decorators."
+    def __call__(self, func):
+        @wraps(func)
+        def inner(*args, **kwds):
+            with self:
+                return func(*args, **kwds)
+        return inner
+
+
+class GeneratorContextManager(ContextDecorator):
     """Helper for @contextmanager decorator."""
 
     def __init__(self, gen):

Modified: python/branches/py3k-cdecimal/Lib/ctypes/__init__.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/ctypes/__init__.py	(original)
+++ python/branches/py3k-cdecimal/Lib/ctypes/__init__.py	Sat Jul 17 12:36:09 2010
@@ -265,7 +265,7 @@
     pass
 else:
     if _os.name in ("nt", "ce"):
-        set_conversion_mode("mbcs", "ignore")
+        set_conversion_mode("mbcs", "strict")
     else:
         set_conversion_mode("ascii", "strict")
 

Modified: python/branches/py3k-cdecimal/Lib/ctypes/test/test_callbacks.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/ctypes/test/test_callbacks.py	(original)
+++ python/branches/py3k-cdecimal/Lib/ctypes/test/test_callbacks.py	Sat Jul 17 12:36:09 2010
@@ -166,6 +166,41 @@
 
         self.assertTrue(diff < 0.01, "%s not less than 0.01" % diff)
 
+    def test_issue_8959_a(self):
+        from ctypes.util import find_library
+        libc_path = find_library("c")
+        if not libc_path:
+            return # cannot test
+        libc = CDLL(libc_path)
+
+        @CFUNCTYPE(c_int, POINTER(c_int), POINTER(c_int))
+        def cmp_func(a, b):
+            return a[0] - b[0]
+
+        array = (c_int * 5)(5, 1, 99, 7, 33)
+
+        libc.qsort(array, len(array), sizeof(c_int), cmp_func)
+        self.assertEqual(array[:], [1, 5, 7, 33, 99])
+
+    try:
+        WINFUNCTYPE
+    except NameError:
+        pass
+    else:
+        def test_issue_8959_b(self):
+            from ctypes.wintypes import BOOL, HWND, LPARAM
+            global windowCount
+            windowCount = 0
+
+            @WINFUNCTYPE(BOOL, HWND, LPARAM)
+            def EnumWindowsCallbackFunc(hwnd, lParam):
+                global windowCount
+                windowCount += 1
+                return True #Allow windows to keep enumerating
+
+            windll.user32.EnumWindows(EnumWindowsCallbackFunc, 0)
+            self.assertFalse(windowCount == 0)
+
 ################################################################
 
 if __name__ == '__main__':

Modified: python/branches/py3k-cdecimal/Lib/ctypes/test/test_win32.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/ctypes/test/test_win32.py	(original)
+++ python/branches/py3k-cdecimal/Lib/ctypes/test/test_win32.py	Sat Jul 17 12:36:09 2010
@@ -6,6 +6,32 @@
 
 import _ctypes_test
 
+if sys.platform == "win32" and sizeof(c_void_p) == sizeof(c_int):
+    # Only windows 32-bit has different calling conventions.
+
+    class WindowsTestCase(unittest.TestCase):
+        def test_callconv_1(self):
+            # Testing stdcall function
+
+            IsWindow = windll.user32.IsWindow
+            # ValueError: Procedure probably called with not enough arguments (4 bytes missing)
+            self.assertRaises(ValueError, IsWindow)
+
+            # This one should succeeed...
+            self.assertEqual(0, IsWindow(0))
+
+            # ValueError: Procedure probably called with too many arguments (8 bytes in excess)
+            self.assertRaises(ValueError, IsWindow, 0, 0, 0)
+
+        def test_callconv_2(self):
+            # Calling stdcall function as cdecl
+
+            IsWindow = cdll.user32.IsWindow
+
+            # ValueError: Procedure called with not enough arguments (4 bytes missing)
+            # or wrong calling convention
+            self.assertRaises(ValueError, IsWindow, None)
+
 if sys.platform == "win32":
     class FunctionCallTestCase(unittest.TestCase):
 

Modified: python/branches/py3k-cdecimal/Lib/decimal.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/decimal.py	(original)
+++ python/branches/py3k-cdecimal/Lib/decimal.py	Sat Jul 17 12:36:09 2010
@@ -2047,12 +2047,14 @@
         # case where xc == 1: result is 10**(xe*y), with xe*y
         # required to be an integer
         if xc == 1:
-            if ye >= 0:
-                exponent = xe*yc*10**ye
-            else:
-                exponent, remainder = divmod(xe*yc, 10**-ye)
-                if remainder:
-                    return None
+            xe *= yc
+            # result is now 10**(xe * 10**ye);  xe * 10**ye must be integral
+            while xe % 10 == 0:
+                xe //= 10
+                ye += 1
+            if ye < 0:
+                return None
+            exponent = xe * 10**ye
             if y.sign == 1:
                 exponent = -exponent
             # if other is a nonnegative integer, use ideal exponent
@@ -2325,9 +2327,10 @@
         # try for an exact result with precision +1
         if ans is None:
             ans = self._power_exact(other, context.prec + 1)
-            if ans is not None and result_sign == 1:
-                ans = _dec_from_triple(1, ans._int, ans._exp)
-            exact = True
+            if ans is not None:
+                if result_sign == 1:
+                    ans = _dec_from_triple(1, ans._int, ans._exp)
+                exact = True
 
         # usual case: inexact result, x**y computed directly as exp(y*log(x))
         if ans is None:
@@ -3810,20 +3813,38 @@
                  Emin=None, Emax=None,
                  capitals=None, clamp=None,
                  _ignored_flags=None):
-        if flags is None:
-            flags = []
+        # Set defaults; for everything except flags and _ignored_flags,
+        # inherit from DefaultContext.
+        try:
+            dc = DefaultContext
+        except NameError:
+            pass
+
+        self.prec = prec if prec is not None else dc.prec
+        self.rounding = rounding if rounding is not None else dc.rounding
+        self.Emin = Emin if Emin is not None else dc.Emin
+        self.Emax = Emax if Emax is not None else dc.Emax
+        self.capitals = capitals if capitals is not None else dc.capitals
+        self.clamp = clamp if clamp is not None else dc.clamp
+
         if _ignored_flags is None:
-            _ignored_flags = []
-        if not isinstance(flags, dict):
-            flags = dict([(s, int(s in flags)) for s in _signals])
-        if traps is not None and not isinstance(traps, dict):
-            traps = dict([(s, int(s in traps)) for s in _signals])
-        for name, val in locals().items():
-            if val is None:
-                setattr(self, name, _copy.copy(getattr(DefaultContext, name)))
-            else:
-                setattr(self, name, val)
-        del self.self
+            self._ignored_flags = []
+        else:
+            self._ignored_flags = _ignored_flags
+
+        if traps is None:
+            self.traps = dc.traps.copy()
+        elif not isinstance(traps, dict):
+            self.traps = dict((s, int(s in traps)) for s in _signals)
+        else:
+            self.traps = traps
+
+        if flags is None:
+            self.flags = dict.fromkeys(_signals, 0)
+        elif not isinstance(flags, dict):
+            self.flags = dict((s, int(s in flags)) for s in _signals)
+        else:
+            self.flags = flags
 
     def __repr__(self):
         """Show the current context."""

Modified: python/branches/py3k-cdecimal/Lib/dis.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/dis.py	(original)
+++ python/branches/py3k-cdecimal/Lib/dis.py	Sat Jul 17 12:36:09 2010
@@ -12,6 +12,22 @@
 
 _have_code = (types.MethodType, types.FunctionType, types.CodeType, type)
 
+def _try_compile(source, name):
+    """Attempts to compile the given source, first as an expression and
+       then as a statement if the first approach fails.
+
+       Utility function to accept strings in functions that otherwise
+       expect code objects
+    """
+    # ncoghlan: currently only used by dis(), but plan to add an
+    # equivalent for show_code() as well (but one that returns a
+    # string rather than printing directly to the console)
+    try:
+        c = compile(source, name, 'eval')
+    except SyntaxError:
+        c = compile(source, name, 'exec')
+    return c
+
 def dis(x=None):
     """Disassemble classes, methods, functions, or code.
 
@@ -38,7 +54,9 @@
     elif hasattr(x, 'co_code'):
         disassemble(x)
     elif isinstance(x, (bytes, bytearray)):
-        disassemble_string(x)
+        _disassemble_bytes(x)
+    elif isinstance(x, str):
+        _disassemble_str(x)
     else:
         raise TypeError("don't know how to disassemble %s objects" %
                         type(x).__name__)
@@ -157,7 +175,7 @@
                 print('(' + free[oparg] + ')', end=' ')
         print()
 
-def disassemble_string(code, lasti=-1, varnames=None, names=None,
+def _disassemble_bytes(code, lasti=-1, varnames=None, names=None,
                        constants=None):
     labels = findlabels(code)
     n = len(code)
@@ -196,6 +214,10 @@
                 print('(' + cmp_op[oparg] + ')', end=' ')
         print()
 
+def _disassemble_str(source):
+    """Compile the source string, then disassemble the code object."""
+    disassemble(_try_compile(source, '<dis>'))
+
 disco = disassemble                     # XXX For backwards compatibility
 
 def findlabels(code):

Modified: python/branches/py3k-cdecimal/Lib/distutils/command/install.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/distutils/command/install.py	(original)
+++ python/branches/py3k-cdecimal/Lib/distutils/command/install.py	Sat Jul 17 12:36:09 2010
@@ -302,8 +302,8 @@
         # about needing recursive variable expansion (shudder).
 
         py_version = sys.version.split()[0]
-        prefix, exec_prefix, srcdir = get_config_vars('prefix', 'exec_prefix',
-                                                      'srcdir')
+        prefix, exec_prefix, srcdir, projectbase = get_config_vars('prefix', 'exec_prefix',
+                                                      'srcdir', 'projectbase')
 
         self.config_vars = {'dist_name': self.distribution.get_name(),
                             'dist_version': self.distribution.get_version(),
@@ -316,6 +316,7 @@
                             'sys_exec_prefix': exec_prefix,
                             'exec_prefix': exec_prefix,
                             'srcdir': srcdir,
+                            'projectbase': projectbase,
                            }
 
         self.config_vars['userbase'] = self.install_userbase

Modified: python/branches/py3k-cdecimal/Lib/distutils/msvc9compiler.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/distutils/msvc9compiler.py	(original)
+++ python/branches/py3k-cdecimal/Lib/distutils/msvc9compiler.py	Sat Jul 17 12:36:09 2010
@@ -37,9 +37,18 @@
          winreg.HKEY_LOCAL_MACHINE,
          winreg.HKEY_CLASSES_ROOT)
 
-VS_BASE = r"Software\Microsoft\VisualStudio\%0.1f"
-WINSDK_BASE = r"Software\Microsoft\Microsoft SDKs\Windows"
-NET_BASE = r"Software\Microsoft\.NETFramework"
+NATIVE_WIN64 = (sys.platform == 'win32' and sys.maxsize > 2**32)
+if NATIVE_WIN64:
+    # Visual C++ is a 32-bit application, so we need to look in
+    # the corresponding registry branch, if we're running a
+    # 64-bit Python on Win64
+    VS_BASE = r"Software\Wow6432Node\Microsoft\VisualStudio\%0.1f"
+    WINSDK_BASE = r"Software\Wow6432Node\Microsoft\Microsoft SDKs\Windows"
+    NET_BASE = r"Software\Wow6432Node\Microsoft\.NETFramework"
+else:
+    VS_BASE = r"Software\Microsoft\VisualStudio\%0.1f"
+    WINSDK_BASE = r"Software\Microsoft\Microsoft SDKs\Windows"
+    NET_BASE = r"Software\Microsoft\.NETFramework"
 
 # A map keyed by get_platform() return values to values accepted by
 # 'vcvarsall.bat'.  Note a cross-compile may combine these (eg, 'x86_amd64' is

Modified: python/branches/py3k-cdecimal/Lib/distutils/tests/test_sdist.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/distutils/tests/test_sdist.py	(original)
+++ python/branches/py3k-cdecimal/Lib/distutils/tests/test_sdist.py	Sat Jul 17 12:36:09 2010
@@ -346,6 +346,7 @@
         finally:
             archive.close()
 
+    @unittest.skipUnless(zlib, "requires zlib")
     def test_get_file_list(self):
         # make sure MANIFEST is recalculated
         dist, cmd = self.get_cmd()

Modified: python/branches/py3k-cdecimal/Lib/distutils/unixccompiler.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/distutils/unixccompiler.py	(original)
+++ python/branches/py3k-cdecimal/Lib/distutils/unixccompiler.py	Sat Jul 17 12:36:09 2010
@@ -340,7 +340,9 @@
             static = os.path.join(dir, static_f)
 
             if sys.platform == 'darwin' and (
-                    dir.startswith('/System/') or dir.startswith('/usr/')):
+                dir.startswith('/System/') or (
+                dir.startswith('/usr/') and not dir.startswith('/usr/local/'))):
+
                 shared = os.path.join(sysroot, dir[1:], shared_f)
                 dylib = os.path.join(sysroot, dir[1:], dylib_f)
                 static = os.path.join(sysroot, dir[1:], static_f)

Modified: python/branches/py3k-cdecimal/Lib/email/feedparser.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/email/feedparser.py	(original)
+++ python/branches/py3k-cdecimal/Lib/email/feedparser.py	Sat Jul 17 12:36:09 2010
@@ -28,7 +28,7 @@
 
 NLCRE = re.compile('\r\n|\r|\n')
 NLCRE_bol = re.compile('(\r\n|\r|\n)')
-NLCRE_eol = re.compile('(\r\n|\r|\n)$')
+NLCRE_eol = re.compile('(\r\n|\r|\n)\Z')
 NLCRE_crack = re.compile('(\r\n|\r|\n)')
 # RFC 2822 $3.6.8 Optional fields.  ftext is %d33-57 / %d59-126, Any character
 # except controls, SP, and ":".
@@ -104,6 +104,10 @@
         # data after the final RE.  In the case of a NL/CR terminated string,
         # this is the empty string.
         self._partial = parts.pop()
+        #GAN 29Mar09  bugs 1555570, 1721862  Confusion at 8K boundary ending with \r:
+        # is there a \n to follow later?
+        if not self._partial and parts and parts[-1].endswith('\r'):
+            self._partial = parts.pop(-2)+parts.pop()
         # parts is a list of strings, alternating between the line contents
         # and the eol character(s).  Gather up a list of lines after
         # re-attaching the newlines.

Modified: python/branches/py3k-cdecimal/Lib/email/test/test_email.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/email/test/test_email.py	(original)
+++ python/branches/py3k-cdecimal/Lib/email/test/test_email.py	Sat Jul 17 12:36:09 2010
@@ -2454,6 +2454,39 @@
 -Me
 """)
 
+    def test_pushCR_LF(self):
+        '''FeedParser BufferedSubFile.push() assumed it received complete
+           line endings.  A CR ending one push() followed by a LF starting
+           the next push() added an empty line.
+        '''
+        imt = [
+            ("a\r \n",  2),
+            ("b",       0),
+            ("c\n",     1),
+            ("",        0),
+            ("d\r\n",   1),
+            ("e\r",     0),
+            ("\nf",     1),
+            ("\r\n",    1),
+          ]
+        from email.feedparser import BufferedSubFile, NeedMoreData
+        bsf = BufferedSubFile()
+        om = []
+        nt = 0
+        for il, n in imt:
+            bsf.push(il)
+            nt += n
+            n1 = 0
+            while True:
+                ol = bsf.readline()
+                if ol == NeedMoreData:
+                    break
+                om.append(ol)
+                n1 += 1
+            self.assertTrue(n == n1)
+        self.assertTrue(len(om) == nt)
+        self.assertTrue(''.join([il for il, n in imt]) == ''.join(om))
+
 
 
 class TestParsers(TestEmailBase):
@@ -2588,6 +2621,24 @@
         eq(headers, ['A', 'B', 'CC'])
         eq(msg.get_payload(), 'body')
 
+    def test_CRLFLF_at_end_of_part(self):
+        # issue 5610: feedparser should not eat two chars from body part ending
+        # with "\r\n\n".
+        m = (
+            "From: foo at bar.com\n"
+            "To: baz\n"
+            "Mime-Version: 1.0\n"
+            "Content-Type: multipart/mixed; boundary=BOUNDARY\n"
+            "\n"
+            "--BOUNDARY\n"
+            "Content-Type: text/plain\n"
+            "\n"
+            "body ending with CRLF newline\r\n"
+            "\n"
+            "--BOUNDARY--\n"
+          )
+        msg = email.message_from_string(m)
+        self.assertTrue(msg.get_payload(0).get_payload().endswith('\r\n'))
 
 
 class TestBase64(unittest.TestCase):

Modified: python/branches/py3k-cdecimal/Lib/encodings/aliases.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/encodings/aliases.py	(original)
+++ python/branches/py3k-cdecimal/Lib/encodings/aliases.py	Sat Jul 17 12:36:09 2010
@@ -146,6 +146,11 @@
     'csibm857'           : 'cp857',
     'ibm857'             : 'cp857',
 
+    # cp858 codec
+    '858'                : 'cp858',
+    'csibm858'           : 'cp858',
+    'ibm858'             : 'cp858',
+
     # cp860 codec
     '860'                : 'cp860',
     'csibm860'           : 'cp860',

Modified: python/branches/py3k-cdecimal/Lib/fnmatch.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/fnmatch.py	(original)
+++ python/branches/py3k-cdecimal/Lib/fnmatch.py	Sat Jul 17 12:36:09 2010
@@ -12,10 +12,16 @@
 
 import re
 
-__all__ = ["filter", "fnmatch","fnmatchcase","translate"]
+__all__ = ["filter", "fnmatch", "fnmatchcase", "purge", "translate"]
 
 _cache = {}  # Maps text patterns to compiled regexen.
 _cacheb = {}  # Ditto for bytes patterns.
+_MAXCACHE = 100 # Maximum size of caches
+
+def purge():
+    """Clear the pattern cache"""
+    _cache.clear()
+    _cacheb.clear()
 
 def fnmatch(name, pat):
     """Test whether FILENAME matches PATTERN.
@@ -48,6 +54,8 @@
             res = bytes(res_str, 'ISO-8859-1')
         else:
             res = translate(pat)
+        if len(cache) >= _MAXCACHE:
+            cache.clear()
         cache[pat] = regex = re.compile(res)
     return regex.match
 

Modified: python/branches/py3k-cdecimal/Lib/heapq.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/heapq.py	(original)
+++ python/branches/py3k-cdecimal/Lib/heapq.py	Sat Jul 17 12:36:09 2010
@@ -1,5 +1,3 @@
-# -*- coding: latin-1 -*-
-
 """Heap queue algorithm (a.k.a. priority queue).
 
 Heaps are arrays for which a[k] <= a[2*k+1] and a[k] <= a[2*k+2] for
@@ -34,7 +32,7 @@
 
 __about__ = """Heap queues
 
-[explanation by Fran?ois Pinard]
+[explanation by Fran??ois Pinard]
 
 Heaps are arrays for which a[k] <= a[2*k+1] and a[k] <= a[2*k+2] for
 all k, counting elements from 0.  For the sake of comparison,

Modified: python/branches/py3k-cdecimal/Lib/http/cookiejar.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/http/cookiejar.py	(original)
+++ python/branches/py3k-cdecimal/Lib/http/cookiejar.py	Sat Jul 17 12:36:09 2010
@@ -436,6 +436,13 @@
         if attr: headers.append("; ".join(attr))
     return ", ".join(headers)
 
+def strip_quotes(text):
+    if text.startswith('"'):
+        text = text[1:]
+    if text.endswith('"'):
+        text = text[:-1]
+    return text
+
 def parse_ns_headers(ns_headers):
     """Ad-hoc parser for Netscape protocol cookie-attributes.
 
@@ -453,7 +460,7 @@
     """
     known_attrs = ("expires", "domain", "path", "secure",
                    # RFC 2109 attrs (may turn up in Netscape cookies, too)
-                   "port", "max-age")
+                   "version", "port", "max-age")
 
     result = []
     for ns_header in ns_headers:
@@ -473,12 +480,11 @@
                     k = lc
                 if k == "version":
                     # This is an RFC 2109 cookie.
+                    v = strip_quotes(v)
                     version_set = True
                 if k == "expires":
                     # convert expires date to seconds since epoch
-                    if v.startswith('"'): v = v[1:]
-                    if v.endswith('"'): v = v[:-1]
-                    v = http2time(v)  # None if invalid
+                    v = http2time(strip_quotes(v))  # None if invalid
             pairs.append((k, v))
 
         if pairs:
@@ -1446,7 +1452,11 @@
 
         # set the easy defaults
         version = standard.get("version", None)
-        if version is not None: version = int(version)
+        if version is not None:
+            try:
+                version = int(version)
+            except ValueError:
+                return None  # invalid version, ignore cookie
         secure = standard.get("secure", False)
         # (discard is also set if expires is Absent)
         discard = standard.get("discard", False)

Modified: python/branches/py3k-cdecimal/Lib/idlelib/textView.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/idlelib/textView.py	(original)
+++ python/branches/py3k-cdecimal/Lib/idlelib/textView.py	Sat Jul 17 12:36:09 2010
@@ -62,11 +62,7 @@
 
 def view_file(parent, title, filename, encoding=None):
     try:
-        if encoding:
-            import codecs
-            textFile = codecs.open(filename, 'r')
-        else:
-            textFile = open(filename, 'r')
+        textFile = open(filename, 'r', encoding=encoding)
     except IOError:
         import tkinter.messagebox as tkMessageBox
         tkMessageBox.showerror(title='File Load Error',

Modified: python/branches/py3k-cdecimal/Lib/importlib/__init__.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/importlib/__init__.py	(original)
+++ python/branches/py3k-cdecimal/Lib/importlib/__init__.py	Sat Jul 17 12:36:09 2010
@@ -36,7 +36,7 @@
     """
     if 'PYTHONCASEOK' in os.environ:
         return True
-    elif check in os.listdir(directory):
+    elif check in os.listdir(directory if directory else os.getcwd()):
         return True
     return False
 

Modified: python/branches/py3k-cdecimal/Lib/importlib/_bootstrap.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/importlib/_bootstrap.py	(original)
+++ python/branches/py3k-cdecimal/Lib/importlib/_bootstrap.py	Sat Jul 17 12:36:09 2010
@@ -22,7 +22,7 @@
 def _path_join(*args):
     """Replacement for os.path.join."""
     return path_sep.join(x[:-len(path_sep)] if x.endswith(path_sep) else x
-                            for x in args)
+                            for x in args if x)
 
 
 def _path_exists(path):
@@ -53,6 +53,8 @@
 # XXX Could also expose Modules/getpath.c:isdir()
 def _path_isdir(path):
     """Replacement for os.path.isdir."""
+    if not path:
+        path = _os.getcwd()
     return _path_is_mode_type(path, 0o040000)
 
 
@@ -99,6 +101,8 @@
     new.__dict__.update(old.__dict__)
 
 
+code_type = type(_wrap.__code__)
+
 # Finder/loader utility code ##################################################
 
 def set_package(fxn):
@@ -138,7 +142,7 @@
     the second argument.
 
     """
-    def decorated(self, fullname):
+    def decorated(self, fullname, *args, **kwargs):
         module = sys.modules.get(fullname)
         is_reload = bool(module)
         if not is_reload:
@@ -148,7 +152,7 @@
             module = imp.new_module(fullname)
             sys.modules[fullname] = module
         try:
-            return fxn(self, module)
+            return fxn(self, module, *args, **kwargs)
         except:
             if not is_reload:
                 del sys.modules[fullname]
@@ -301,258 +305,227 @@
         return imp.is_frozen_package(fullname)
 
 
-class PyLoader:
+class _LoaderBasics:
 
-    """Loader base class for Python source code.
+    """Base class of common code needed by both SourceLoader and
+    _SourcelessFileLoader."""
 
-    Subclasses need to implement the methods:
+    def is_package(self, fullname):
+        """Concrete implementation of InspectLoader.is_package by checking if
+        the path returned by get_filename has a filename of '__init__.py'."""
+        filename = self.get_filename(fullname).rpartition(path_sep)[2]
+        return filename.rsplit('.', 1)[0] == '__init__'
+
+    def _bytes_from_bytecode(self, fullname, data, source_mtime):
+        """Return the marshalled bytes from bytecode, verifying the magic
+        number and timestamp alon the way.
 
-    - source_path
-    - get_data
-    - is_package
+        If source_mtime is None then skip the timestamp check.
 
-    """
+        """
+        magic = data[:4]
+        raw_timestamp = data[4:8]
+        if len(magic) != 4 or magic != imp.get_magic():
+            raise ImportError("bad magic number in {}".format(fullname))
+        elif len(raw_timestamp) != 4:
+            raise EOFError("bad timestamp in {}".format(fullname))
+        elif source_mtime is not None:
+            if marshal._r_long(raw_timestamp) != source_mtime:
+                raise ImportError("bytecode is stale for {}".format(fullname))
+        # Can't return the code object as errors from marshal loading need to
+        # propagate even when source is available.
+        return data[8:]
 
     @module_for_loader
-    def load_module(self, module):
-        """Initialize the module."""
+    def _load_module(self, module, *, sourceless=False):
+        """Helper for load_module able to handle either source or sourceless
+        loading."""
         name = module.__name__
-        code_object = self.get_code(module.__name__)
+        code_object = self.get_code(name)
         module.__file__ = self.get_filename(name)
+        if not sourceless:
+            module.__cached__ = imp.cache_from_source(module.__file__)
+        else:
+            module.__cached__ = module.__file__
+        module.__package__ = name
         if self.is_package(name):
-            module.__path__  = [module.__file__.rsplit(path_sep, 1)[0]]
-        module.__package__ = module.__name__
-        if not hasattr(module, '__path__'):
+            module.__path__ = [module.__file__.rsplit(path_sep, 1)[0]]
+        else:
             module.__package__ = module.__package__.rpartition('.')[0]
         module.__loader__ = self
         exec(code_object, module.__dict__)
         return module
 
-    def get_filename(self, fullname):
-        """Return the path to the source file, else raise ImportError."""
-        path = self.source_path(fullname)
-        if path is not None:
-            return path
-        else:
-            raise ImportError("no source path available for "
-                                "{0!r}".format(fullname))
 
-    def get_code(self, fullname):
-        """Get a code object from source."""
-        source_path = self.source_path(fullname)
-        if source_path is None:
-            message = "a source path must exist to load {0}".format(fullname)
-            raise ImportError(message)
-        source = self.get_data(source_path)
-        return compile(source, source_path, 'exec', dont_inherit=True)
+class SourceLoader(_LoaderBasics):
 
-    # Never use in implementing import! Imports code within the method.
-    def get_source(self, fullname):
-        """Return the source code for a module.
+    def path_mtime(self, path:str) -> int:
+        """Optional method that returns the modification time for the specified
+        path.
 
-        self.source_path() and self.get_data() are used to implement this
-        method.
+        Implementing this method allows the loader to read bytecode files.
 
         """
-        path = self.source_path(fullname)
-        if path is None:
-            return None
-        try:
-            source_bytes = self.get_data(path)
-        except IOError:
-            return ImportError("source not available through get_data()")
-        import io
-        import tokenize
-        encoding = tokenize.detect_encoding(io.BytesIO(source_bytes).readline)
-        return source_bytes.decode(encoding[0])
+        raise NotImplementedError
 
+    def set_data(self, path:str, data:bytes) -> None:
+        """Optional method which writes data to a file path.
 
-class PyPycLoader(PyLoader):
+        Implementing this method allows for the writing of bytecode files.
 
-    """Loader base class for Python source and bytecode.
+        """
+        raise NotImplementedError
 
-    Requires implementing the methods needed for PyLoader as well as
-    source_mtime, bytecode_path, and write_bytecode.
 
-    """
-
-    def get_filename(self, fullname):
-        """Return the source or bytecode file path."""
-        path = self.source_path(fullname)
-        if path is not None:
-            return path
-        path = self.bytecode_path(fullname)
-        if path is not None:
-            return path
-        raise ImportError("no source or bytecode path available for "
-                            "{0!r}".format(fullname))
+    def get_source(self, fullname):
+        """Concrete implementation of InspectLoader.get_source."""
+        import tokenize
+        path = self.get_filename(fullname)
+        try:
+            source_bytes = self.get_data(path)
+        except IOError:
+            raise ImportError("source not available through get_data()")
+        encoding = tokenize.detect_encoding(_io.BytesIO(source_bytes).readline)
+        newline_decoder = _io.IncrementalNewlineDecoder(None, True)
+        return newline_decoder.decode(source_bytes.decode(encoding[0]))
 
     def get_code(self, fullname):
-        """Get a code object from source or bytecode."""
-        # XXX Care enough to make sure this call does not happen if the magic
-        #     number is bad?
-        source_timestamp = self.source_mtime(fullname)
-        # Try to use bytecode if it is available.
-        bytecode_path = self.bytecode_path(fullname)
-        if bytecode_path:
-            data = self.get_data(bytecode_path)
+        """Concrete implementation of InspectLoader.get_code.
+
+        Reading of bytecode requires path_mtime to be implemented. To write
+        bytecode, set_data must also be implemented.
+
+        """
+        source_path = self.get_filename(fullname)
+        bytecode_path = imp.cache_from_source(source_path)
+        source_mtime = None
+        if bytecode_path is not None:
             try:
-                magic = data[:4]
-                if len(magic) < 4:
-                    raise ImportError("bad magic number in {}".format(fullname))
-                raw_timestamp = data[4:8]
-                if len(raw_timestamp) < 4:
-                    raise EOFError("bad timestamp in {}".format(fullname))
-                pyc_timestamp = marshal._r_long(raw_timestamp)
-                bytecode = data[8:]
-                # Verify that the magic number is valid.
-                if imp.get_magic() != magic:
-                    raise ImportError("bad magic number in {}".format(fullname))
-                # Verify that the bytecode is not stale (only matters when
-                # there is source to fall back on.
-                if source_timestamp:
-                    if pyc_timestamp < source_timestamp:
-                        raise ImportError("bytecode is stale")
-            except (ImportError, EOFError):
-                # If source is available give it a shot.
-                if source_timestamp is not None:
+                source_mtime = self.path_mtime(source_path)
+            except NotImplementedError:
+                pass
+            else:
+                try:
+                    data = self.get_data(bytecode_path)
+                except IOError:
                     pass
                 else:
-                    raise
-            else:
-                # Bytecode seems fine, so try to use it.
-                # XXX If the bytecode is ill-formed, would it be beneficial to
-                #     try for using source if available and issue a warning?
-                return marshal.loads(bytecode)
-        elif source_timestamp is None:
-            raise ImportError("no source or bytecode available to create code "
-                                "object for {0!r}".format(fullname))
-        # Use the source.
-        code_object = super().get_code(fullname)
-        # Generate bytecode and write it out.
-        if not sys.dont_write_bytecode:
+                    try:
+                        bytes_data = self._bytes_from_bytecode(fullname, data,
+                                                               source_mtime)
+                    except (ImportError, EOFError):
+                        pass
+                    else:
+                        found = marshal.loads(bytes_data)
+                        if isinstance(found, code_type):
+                            return found
+                        else:
+                            msg = "Non-code object in {}"
+                            raise ImportError(msg.format(bytecode_path))
+        source_bytes = self.get_data(source_path)
+        code_object = compile(source_bytes, source_path, 'exec',
+                                dont_inherit=True)
+        if (not sys.dont_write_bytecode and bytecode_path is not None and
+                source_mtime is not None):
+            # If e.g. Jython ever implements imp.cache_from_source to have
+            # their own cached file format, this block of code will most likely
+            # throw an exception.
             data = bytearray(imp.get_magic())
-            data.extend(marshal._w_long(source_timestamp))
+            data.extend(marshal._w_long(source_mtime))
             data.extend(marshal.dumps(code_object))
-            self.write_bytecode(fullname, data)
+            try:
+                self.set_data(bytecode_path, data)
+            except NotImplementedError:
+                pass
         return code_object
 
+    def load_module(self, fullname):
+        """Concrete implementation of Loader.load_module.
 
-class _PyFileLoader(PyLoader):
+        Requires ExecutionLoader.get_filename and ResourceLoader.get_data to be
+        implemented to load source code. Use of bytecode is dictated by whether
+        get_code uses/writes bytecode.
 
-    """Load a Python source file."""
+        """
+        return self._load_module(fullname)
 
-    def __init__(self, name, path, is_pkg):
-        self._name = name
-        self._is_pkg = is_pkg
-        # Figure out the base path based on whether it was source or bytecode
-        # that was found.
-        try:
-            self._base_path = _path_without_ext(path, imp.PY_SOURCE)
-        except ValueError:
-            self._base_path = _path_without_ext(path, imp.PY_COMPILED)
-
-    def _find_path(self, ext_type):
-        """Find a path from the base path and the specified extension type that
-        exists, returning None if one is not found."""
-        for suffix in _suffix_list(ext_type):
-            path = self._base_path + suffix
-            if _path_exists(path):
-                return path
-        else:
-            return None
+
+class _FileLoader:
+
+    """Base file loader class which implements the loader protocol methods that
+    require file system usage."""
+
+    def __init__(self, fullname, path):
+        """Cache the module name and the path to the file found by the
+        finder."""
+        self._name = fullname
+        self._path = path
 
     @_check_name
-    def source_path(self, fullname):
-        """Return the path to an existing source file for the module, or None
-        if one cannot be found."""
-        # Not a property so that it is easy to override.
-        return self._find_path(imp.PY_SOURCE)
+    def get_filename(self, fullname):
+        """Return the path to the source file as found by the finder."""
+        return self._path
 
     def get_data(self, path):
         """Return the data from path as raw bytes."""
-        return _io.FileIO(path, 'r').read()  # Assuming bytes.
+        with _closing(_io.FileIO(path, 'r')) as file:
+            return file.read()
 
-    @_check_name
-    def is_package(self, fullname):
-        """Return a boolean based on whether the module is a package.
 
-        Raises ImportError (like get_source) if the loader cannot handle the
-        package.
+class _SourceFileLoader(_FileLoader, SourceLoader):
 
-        """
-        return self._is_pkg
+    """Concrete implementation of SourceLoader using the file system."""
 
+    def path_mtime(self, path):
+        """Return the modification time for the path."""
+        return int(_os.stat(path).st_mtime)
 
-class _PyPycFileLoader(PyPycLoader, _PyFileLoader):
+    def set_data(self, path, data):
+        """Write bytes data to a file."""
+        try:
+            with _closing(_io.FileIO(path, 'wb')) as file:
+                file.write(data)
+        except IOError as exc:
+            if exc.errno == errno.ENOENT:
+                directory, _, filename = path.rpartition(path_sep)
+                sub_directories = []
+                while not _path_isdir(directory):
+                    directory, _, sub_dir = directory.rpartition(path_sep)
+                    sub_directories.append(sub_dir)
+                    for part in reversed(sub_directories):
+                        directory = _path_join(directory, part)
+                        try:
+                            _os.mkdir(directory)
+                        except IOError as exc:
+                            if exc.errno != errno.EACCES:
+                                raise
+                            else:
+                                return
+                return self.set_data(path, data)
+            elif exc.errno != errno.EACCES:
+                raise
 
-    """Load a module from a source or bytecode file."""
 
-    def _find_path(self, ext_type):
-        """Return PEP 3147 path if ext_type is PY_COMPILED, otherwise
-        super()._find_path() is called."""
-        if ext_type == imp.PY_COMPILED:
-            # We don't really care what the extension on self._base_path is,
-            # as long as it has exactly one dot.
-            source_path = self._base_path + '.py'
-            pycache_path = imp.cache_from_source(source_path)
-            legacy_path = self._base_path + '.pyc'
-            # The rule is: if the source file exists, then Python always uses
-            # the __pycache__/foo.<tag>.pyc file.  If the source file does not
-            # exist, then Python uses the legacy path.
-            pyc_path = (pycache_path
-                        if _path_exists(source_path)
-                        else legacy_path)
-            return (pyc_path if _path_exists(pyc_path) else None)
-        return super()._find_path(ext_type)
+class _SourcelessFileLoader(_FileLoader, _LoaderBasics):
 
-    @_check_name
-    def source_mtime(self, name):
-        """Return the modification time of the source for the specified
-        module."""
-        source_path = self.source_path(name)
-        if not source_path:
-            return None
-        return int(_os.stat(source_path).st_mtime)
+    """Loader which handles sourceless file imports."""
 
-    @_check_name
-    def bytecode_path(self, fullname):
-        """Return the path to a bytecode file, or None if one does not
-        exist."""
-        # Not a property for easy overriding.
-        return self._find_path(imp.PY_COMPILED)
-
-    @_check_name
-    def write_bytecode(self, name, data):
-        """Write out 'data' for the specified module, returning a boolean
-        signifying if the write-out actually occurred.
+    def load_module(self, fullname):
+        return self._load_module(fullname, sourceless=True)
 
-        Raises ImportError (just like get_source) if the specified module
-        cannot be handled by the loader.
+    def get_code(self, fullname):
+        path = self.get_filename(fullname)
+        data = self.get_data(path)
+        bytes_data = self._bytes_from_bytecode(fullname, data, None)
+        found = marshal.loads(bytes_data)
+        if isinstance(found, code_type):
+            return found
+        else:
+            raise ImportError("Non-code object in {}".format(path))
 
-        """
-        bytecode_path = self.bytecode_path(name)
-        if not bytecode_path:
-            source_path = self.source_path(name)
-            bytecode_path = imp.cache_from_source(source_path)
-            # Ensure that the __pycache__ directory exists.  We can't use
-            # os.path.dirname() here.
-            dirname, sep, basename = bytecode_path.rpartition(path_sep)
-            try:
-                _os.mkdir(dirname)
-            except OSError as error:
-                if error.errno != errno.EEXIST:
-                    raise
-        try:
-            # Assuming bytes.
-            with _closing(_io.FileIO(bytecode_path, 'w')) as bytecode_file:
-                bytecode_file.write(data)
-                return True
-        except IOError as exc:
-            if exc.errno == errno.EACCES:
-                return False
-            else:
-                raise
+    def get_source(self, fullname):
+        """Return None as there is no source code."""
+        return None
 
 
 class _ExtensionFileLoader:
@@ -563,7 +536,7 @@
 
     """
 
-    def __init__(self, name, path, is_pkg):
+    def __init__(self, name, path):
         """Initialize the loader.
 
         If is_pkg is True then an exception is raised as extension modules
@@ -572,8 +545,6 @@
         """
         self._name = name
         self._path = path
-        if is_pkg:
-            raise ValueError("extension modules cannot be packages")
 
     @_check_name
     @set_package
@@ -671,147 +642,88 @@
             return None
 
 
-class _ChainedFinder:
-
-    """Finder that sequentially calls other finders."""
-
-    def __init__(self, *finders):
-        self._finders = finders
-
-    def find_module(self, fullname, path=None):
-        for finder in self._finders:
-            result = finder.find_module(fullname, path)
-            if result:
-                return result
-        else:
-            return None
-
-
 class _FileFinder:
 
-    """Base class for file finders.
-
-    Subclasses are expected to define the following attributes:
-
-        * _suffixes
-            Sequence of file suffixes whose order will be followed.
+    """File-based finder.
 
-        * _possible_package
-            True if importer should check for packages.
-
-        * _loader
-            A callable that takes the module name, a file path, and whether
-            the path points to a package and returns a loader for the module
-            found at that path.
+    Constructor takes a list of objects detailing what file extensions their
+    loader supports along with whether it can be used for a package.
 
     """
 
-    def __init__(self, path_entry):
-        """Initialize an importer for the passed-in sys.path entry (which is
-        assumed to have already been verified as an existing directory).
-
-        Can be used as an entry on sys.path_hook.
+    def __init__(self, path, *details):
+        """Initialize with finder details."""
+        packages = []
+        modules = []
+        for detail in details:
+            modules.extend((suffix, detail.loader) for suffix in detail.suffixes)
+            if detail.supports_packages:
+                packages.extend((suffix, detail.loader)
+                                for suffix in detail.suffixes)
+        self.packages = packages
+        self.modules = modules
+        self.path = path
 
-        """
-        absolute_path = _path_absolute(path_entry)
-        if not _path_isdir(absolute_path):
-            raise ImportError("only directories are supported")
-        self._path_entry = absolute_path
-
-    def find_module(self, fullname, path=None):
+    def find_module(self, fullname):
+        """Try to find a loader for the specified module."""
         tail_module = fullname.rpartition('.')[2]
-        package_directory = None
-        if self._possible_package:
-            for ext in self._suffixes:
-                package_directory = _path_join(self._path_entry, tail_module)
-                init_filename = '__init__' + ext
-                package_init = _path_join(package_directory, init_filename)
-                if (_path_isfile(package_init) and
-                        _case_ok(self._path_entry, tail_module) and
-                        _case_ok(package_directory, init_filename)):
-                    return self._loader(fullname, package_init, True)
-        for ext in self._suffixes:
-            file_name = tail_module + ext
-            file_path = _path_join(self._path_entry, file_name)
-            if (_path_isfile(file_path) and
-                    _case_ok(self._path_entry, file_name)):
-                return self._loader(fullname, file_path, False)
-        else:
-            # Raise a warning if it matches a directory w/o an __init__ file.
-            if (package_directory is not None and
-                    _path_isdir(package_directory) and
-                    _case_ok(self._path_entry, tail_module)):
-                _warnings.warn("Not importing directory %s: missing __init__"
-                                % package_directory, ImportWarning)
-            return None
-
-
-class _PyFileFinder(_FileFinder):
-
-    """Importer for source/bytecode files."""
-
-    _possible_package = True
-    _loader = _PyFileLoader
-
-    def __init__(self, path_entry):
-        # Lack of imp during class creation means _suffixes is set here.
-        # Make sure that Python source files are listed first!  Needed for an
-        # optimization by the loader.
-        self._suffixes = _suffix_list(imp.PY_SOURCE)
-        super().__init__(path_entry)
-
-
-class _PyPycFileFinder(_PyFileFinder):
+        base_path = _path_join(self.path, tail_module)
+        if _path_isdir(base_path) and _case_ok(self.path, tail_module):
+            for suffix, loader in self.packages:
+                init_filename = '__init__' + suffix
+                full_path = _path_join(base_path, init_filename)
+                if (_path_isfile(full_path) and
+                        _case_ok(base_path, init_filename)):
+                    return loader(fullname, full_path)
+            else:
+                msg = "Not importing directory {}: missing __init__"
+                _warnings.warn(msg.format(base_path), ImportWarning)
+        for suffix, loader in self.modules:
+            mod_filename = tail_module + suffix
+            full_path = _path_join(self.path, mod_filename)
+            if _path_isfile(full_path) and _case_ok(self.path, mod_filename):
+                return loader(fullname, full_path)
+        return None
 
-    """Finder for source and bytecode files."""
+class _SourceFinderDetails:
 
-    _loader = _PyPycFileLoader
+    loader = _SourceFileLoader
+    supports_packages = True
 
-    def __init__(self, path_entry):
-        super().__init__(path_entry)
-        self._suffixes += _suffix_list(imp.PY_COMPILED)
+    def __init__(self):
+        self.suffixes = _suffix_list(imp.PY_SOURCE)
 
+class _SourcelessFinderDetails:
 
+    loader = _SourcelessFileLoader
+    supports_packages = True
 
+    def __init__(self):
+        self.suffixes = _suffix_list(imp.PY_COMPILED)
 
-class _ExtensionFileFinder(_FileFinder):
 
-    """Importer for extension files."""
+class _ExtensionFinderDetails:
 
-    _possible_package = False
-    _loader = _ExtensionFileLoader
+    loader = _ExtensionFileLoader
+    supports_packages = False
 
-    def __init__(self, path_entry):
-        # Assigning to _suffixes here instead of at the class level because
-        # imp is not imported at the time of class creation.
-        self._suffixes = _suffix_list(imp.C_EXTENSION)
-        super().__init__(path_entry)
+    def __init__(self):
+        self.suffixes = _suffix_list(imp.C_EXTENSION)
 
 
 # Import itself ###############################################################
 
-def _chained_path_hook(*path_hooks):
-    """Create a closure which sequentially checks path hooks to see which ones
-    (if any) can work with a path."""
-    def path_hook(entry):
-        """Check to see if 'entry' matches any of the enclosed path hooks."""
-        finders = []
-        for hook in path_hooks:
-            try:
-                finder = hook(entry)
-            except ImportError:
-                continue
-            else:
-                finders.append(finder)
-        if not finders:
-            raise ImportError("no finder found")
-        else:
-            return _ChainedFinder(*finders)
-
-    return path_hook
+def _file_path_hook(path):
+    """If the path is a directory, return a file-based finder."""
+    if _path_isdir(path):
+        return _FileFinder(path, _ExtensionFinderDetails(),
+                           _SourceFinderDetails(),
+                           _SourcelessFinderDetails())
+    else:
+        raise ImportError("only directories are supported")
 
 
-_DEFAULT_PATH_HOOK = _chained_path_hook(_ExtensionFileFinder, _PyPycFileFinder)
+_DEFAULT_PATH_HOOK = _file_path_hook
 
 class _DefaultPathFinder(PathFinder):
 

Modified: python/branches/py3k-cdecimal/Lib/importlib/abc.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/importlib/abc.py	(original)
+++ python/branches/py3k-cdecimal/Lib/importlib/abc.py	Sat Jul 17 12:36:09 2010
@@ -1,8 +1,16 @@
 """Abstract base classes related to import."""
 from . import _bootstrap
 from . import machinery
+from . import util
 import abc
+import imp
+import io
+import marshal
+import os.path
+import sys
+import tokenize
 import types
+import warnings
 
 
 class Loader(metaclass=abc.ABCMeta):
@@ -58,19 +66,19 @@
     def is_package(self, fullname:str) -> bool:
         """Abstract method which when implemented should return whether the
         module is a package."""
-        return NotImplementedError
+        raise NotImplementedError
 
     @abc.abstractmethod
     def get_code(self, fullname:str) -> types.CodeType:
         """Abstract method which when implemented should return the code object
         for the module"""
-        return NotImplementedError
+        raise NotImplementedError
 
     @abc.abstractmethod
     def get_source(self, fullname:str) -> str:
         """Abstract method which should return the source code for the
         module."""
-        return NotImplementedError
+        raise NotImplementedError
 
 InspectLoader.register(machinery.BuiltinImporter)
 InspectLoader.register(machinery.FrozenImporter)
@@ -92,33 +100,170 @@
         raise NotImplementedError
 
 
-class PyLoader(_bootstrap.PyLoader, ResourceLoader, ExecutionLoader):
+class SourceLoader(_bootstrap.SourceLoader, ResourceLoader, ExecutionLoader):
 
-    """Abstract base class to assist in loading source code by requiring only
-    back-end storage methods to be implemented.
+    """Abstract base class for loading source code (and optionally any
+    corresponding bytecode).
+
+    To support loading from source code, the abstractmethods inherited from
+    ResourceLoader and ExecutionLoader need to be implemented. To also support
+    loading from bytecode, the optional methods specified directly by this ABC
+    is required.
+
+    Inherited abstractmethods not implemented in this ABC:
+
+        * ResourceLoader.get_data
+        * ExecutionLoader.get_filename
+
+    """
 
-    The methods get_code, get_source, and load_module are implemented for the
-    user.
+
+class PyLoader(SourceLoader):
+
+    """Implement the deprecated PyLoader ABC in terms of SourceLoader.
+
+    This class has been deprecated! It is slated for removal in Python 3.4.
+    If compatibility with Python 3.1 is not needed then implement the
+    SourceLoader ABC instead of this class. If Python 3.1 compatibility is
+    needed, then use the following idiom to have a single class that is
+    compatible with Python 3.1 onwards::
+
+        try:
+            from importlib.abc import SourceLoader
+        except ImportError:
+            from importlib.abc import PyLoader as SourceLoader
+
+
+        class CustomLoader(SourceLoader):
+            def get_filename(self, fullname):
+                # Implement ...
+
+            def source_path(self, fullname):
+                '''Implement source_path in terms of get_filename.'''
+                try:
+                    return self.get_filename(fullname)
+                except ImportError:
+                    return None
+
+            def is_package(self, fullname):
+                filename = os.path.basename(self.get_filename(fullname))
+                return os.path.splitext(filename)[0] == '__init__'
 
     """
 
     @abc.abstractmethod
+    def is_package(self, fullname):
+        raise NotImplementedError
+
+    @abc.abstractmethod
     def source_path(self, fullname:str) -> object:
         """Abstract method which when implemented should return the path to the
-        sourced code for the module."""
+        source code for the module."""
         raise NotImplementedError
 
+    def get_filename(self, fullname):
+        """Implement get_filename in terms of source_path.
+
+        As get_filename should only return a source file path there is no
+        chance of the path not existing but loading still being possible, so
+        ImportError should propagate instead of being turned into returning
+        None.
+
+        """
+        warnings.warn("importlib.abc.PyLoader is deprecated and is "
+                            "slated for removal in Python 3.4; "
+                            "use SourceLoader instead. "
+                            "See the importlib documentation on how to be "
+                            "compatible with Python 3.1 onwards.",
+                        PendingDeprecationWarning)
+        path = self.source_path(fullname)
+        if path is None:
+            raise ImportError
+        else:
+            return path
+
 
-class PyPycLoader(_bootstrap.PyPycLoader, PyLoader):
+class PyPycLoader(PyLoader):
 
     """Abstract base class to assist in loading source and bytecode by
     requiring only back-end storage methods to be implemented.
 
+    This class has been deprecated! Removal is slated for Python 3.4. Implement
+    the SourceLoader ABC instead. If Python 3.1 compatibility is needed, see
+    PyLoader.
+
     The methods get_code, get_source, and load_module are implemented for the
     user.
 
     """
 
+    def get_filename(self, fullname):
+        """Return the source or bytecode file path."""
+        path = self.source_path(fullname)
+        if path is not None:
+            return path
+        path = self.bytecode_path(fullname)
+        if path is not None:
+            return path
+        raise ImportError("no source or bytecode path available for "
+                            "{0!r}".format(fullname))
+
+    def get_code(self, fullname):
+        """Get a code object from source or bytecode."""
+        warnings.warn("importlib.abc.PyPycLoader is deprecated and slated for "
+                            "removal in Python 3.4; use SourceLoader instead. "
+                            "If Python 3.1 compatibility is required, see the "
+                            "latest documentation for PyLoader.",
+                        PendingDeprecationWarning)
+        source_timestamp = self.source_mtime(fullname)
+        # Try to use bytecode if it is available.
+        bytecode_path = self.bytecode_path(fullname)
+        if bytecode_path:
+            data = self.get_data(bytecode_path)
+            try:
+                magic = data[:4]
+                if len(magic) < 4:
+                    raise ImportError("bad magic number in {}".format(fullname))
+                raw_timestamp = data[4:8]
+                if len(raw_timestamp) < 4:
+                    raise EOFError("bad timestamp in {}".format(fullname))
+                pyc_timestamp = marshal._r_long(raw_timestamp)
+                bytecode = data[8:]
+                # Verify that the magic number is valid.
+                if imp.get_magic() != magic:
+                    raise ImportError("bad magic number in {}".format(fullname))
+                # Verify that the bytecode is not stale (only matters when
+                # there is source to fall back on.
+                if source_timestamp:
+                    if pyc_timestamp < source_timestamp:
+                        raise ImportError("bytecode is stale")
+            except (ImportError, EOFError):
+                # If source is available give it a shot.
+                if source_timestamp is not None:
+                    pass
+                else:
+                    raise
+            else:
+                # Bytecode seems fine, so try to use it.
+                return marshal.loads(bytecode)
+        elif source_timestamp is None:
+            raise ImportError("no source or bytecode available to create code "
+                                "object for {0!r}".format(fullname))
+        # Use the source.
+        source_path = self.source_path(fullname)
+        if source_path is None:
+            message = "a source path must exist to load {0}".format(fullname)
+            raise ImportError(message)
+        source = self.get_data(source_path)
+        code_object = compile(source, source_path, 'exec', dont_inherit=True)
+        # Generate bytecode and write it out.
+        if not sys.dont_write_bytecode:
+            data = bytearray(imp.get_magic())
+            data.extend(marshal._w_long(source_timestamp))
+            data.extend(marshal.dumps(code_object))
+            self.write_bytecode(fullname, data)
+        return code_object
+
     @abc.abstractmethod
     def source_mtime(self, fullname:str) -> int:
         """Abstract method which when implemented should return the

Modified: python/branches/py3k-cdecimal/Lib/importlib/test/benchmark.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/importlib/test/benchmark.py	(original)
+++ python/branches/py3k-cdecimal/Lib/importlib/test/benchmark.py	Sat Jul 17 12:36:09 2010
@@ -1,13 +1,15 @@
-"""Benchmark some basic import use-cases."""
-# XXX
-#    - Bench from source (turn off bytecode generation)
-#    - Bench from bytecode (remove existence of source)
-#    - Bench bytecode generation
-#    - Bench extensions
+"""Benchmark some basic import use-cases.
+
+The assumption is made that this benchmark is run in a fresh interpreter and
+thus has no external changes made to import-related attributes in sys.
+
+"""
 from . import util
 from .source import util as source_util
 import imp
 import importlib
+import os
+import py_compile
 import sys
 import timeit
 
@@ -30,9 +32,9 @@
             # One execution too far
             if total_time > seconds:
                 count -= 1
-        yield count
+        yield count // seconds
 
-def from_cache(repeat):
+def from_cache(seconds, repeat):
     """sys.modules"""
     name = '<benchmark import>'
     module = imp.new_module(name)
@@ -40,31 +42,75 @@
     module.__package__ = ''
     with util.uncache(name):
         sys.modules[name] = module
-        for result in bench(name, repeat=repeat):
+        for result in bench(name, repeat=repeat, seconds=seconds):
             yield result
 
 
-def builtin_mod(repeat):
+def builtin_mod(seconds, repeat):
     """Built-in module"""
     name = 'errno'
     if name in sys.modules:
         del sys.modules[name]
-    for result in bench(name, lambda: sys.modules.pop(name), repeat=repeat):
+    # Relying on built-in importer being implicit.
+    for result in bench(name, lambda: sys.modules.pop(name), repeat=repeat,
+                        seconds=seconds):
         yield result
 
 
-def main(import_, repeat=3):
+def source_wo_bytecode(seconds, repeat):
+    """Source w/o bytecode"""
+    sys.dont_write_bytecode = True
+    try:
+        name = '__importlib_test_benchmark__'
+        # Clears out sys.modules and puts an entry at the front of sys.path.
+        with source_util.create_modules(name) as mapping:
+            assert not os.path.exists(imp.cache_from_source(mapping[name]))
+            for result in bench(name, lambda: sys.modules.pop(name), repeat=repeat,
+                                seconds=seconds):
+                yield result
+    finally:
+        sys.dont_write_bytecode = False
+
+
+def source_writing_bytecode(seconds, repeat):
+    """Source writing bytecode"""
+    assert not sys.dont_write_bytecode
+    name = '__importlib_test_benchmark__'
+    with source_util.create_modules(name) as mapping:
+        def cleanup():
+            sys.modules.pop(name)
+            os.unlink(imp.cache_from_source(mapping[name]))
+        for result in bench(name, cleanup, repeat=repeat, seconds=seconds):
+            assert not os.path.exists(imp.cache_from_source(mapping[name]))
+            yield result
+
+
+def source_using_bytecode(seconds, repeat):
+    """Bytecode w/ source"""
+    name = '__importlib_test_benchmark__'
+    with source_util.create_modules(name) as mapping:
+        py_compile.compile(mapping[name])
+        assert os.path.exists(imp.cache_from_source(mapping[name]))
+        for result in bench(name, lambda: sys.modules.pop(name), repeat=repeat,
+                            seconds=seconds):
+            yield result
+
+
+def main(import_):
     __builtins__.__import__ = import_
-    benchmarks = from_cache, builtin_mod
+    benchmarks = (from_cache, builtin_mod, source_using_bytecode,
+                  source_wo_bytecode, source_writing_bytecode,)
+    print("Measuring imports/second\n")
     for benchmark in benchmarks:
         print(benchmark.__doc__, "[", end=' ')
         sys.stdout.flush()
         results = []
-        for result in benchmark(repeat):
+        for result in benchmark(seconds=1, repeat=3):
             results.append(result)
             print(result, end=' ')
             sys.stdout.flush()
-        print("]", "best is", max(results))
+        assert not sys.dont_write_bytecode
+        print("]", "best is", format(max(results), ',d'))
 
 
 if __name__ == '__main__':
@@ -75,7 +121,7 @@
                         default=False, help="use the built-in __import__")
     options, args = parser.parse_args()
     if args:
-        raise RuntimeError("unrecognized args: {0}".format(args))
+        raise RuntimeError("unrecognized args: {}".format(args))
     import_ = __import__
     if not options.builtin:
         import_ = importlib.__import__

Modified: python/branches/py3k-cdecimal/Lib/importlib/test/extension/test_case_sensitivity.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/importlib/test/extension/test_case_sensitivity.py	(original)
+++ python/branches/py3k-cdecimal/Lib/importlib/test/extension/test_case_sensitivity.py	Sat Jul 17 12:36:09 2010
@@ -13,7 +13,8 @@
         good_name = ext_util.NAME
         bad_name = good_name.upper()
         assert good_name != bad_name
-        finder = _bootstrap._ExtensionFileFinder(ext_util.PATH)
+        finder = _bootstrap._FileFinder(ext_util.PATH,
+                                        _bootstrap._ExtensionFinderDetails())
         return finder.find_module(bad_name)
 
     def test_case_sensitive(self):

Modified: python/branches/py3k-cdecimal/Lib/importlib/test/extension/test_finder.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/importlib/test/extension/test_finder.py	(original)
+++ python/branches/py3k-cdecimal/Lib/importlib/test/extension/test_finder.py	Sat Jul 17 12:36:09 2010
@@ -9,7 +9,8 @@
     """Test the finder for extension modules."""
 
     def find_module(self, fullname):
-        importer = _bootstrap._ExtensionFileFinder(util.PATH)
+        importer = _bootstrap._FileFinder(util.PATH,
+                                          _bootstrap._ExtensionFinderDetails())
         return importer.find_module(fullname)
 
     def test_module(self):

Modified: python/branches/py3k-cdecimal/Lib/importlib/test/extension/test_loader.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/importlib/test/extension/test_loader.py	(original)
+++ python/branches/py3k-cdecimal/Lib/importlib/test/extension/test_loader.py	Sat Jul 17 12:36:09 2010
@@ -13,7 +13,7 @@
 
     def load_module(self, fullname):
         loader = _bootstrap._ExtensionFileLoader(ext_util.NAME,
-                                                ext_util.FILEPATH, False)
+                                                ext_util.FILEPATH)
         return loader.load_module(fullname)
 
     def test_module(self):

Modified: python/branches/py3k-cdecimal/Lib/importlib/test/extension/test_path_hook.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/importlib/test/extension/test_path_hook.py	(original)
+++ python/branches/py3k-cdecimal/Lib/importlib/test/extension/test_path_hook.py	Sat Jul 17 12:36:09 2010
@@ -14,7 +14,7 @@
     # XXX Should it only work for directories containing an extension module?
 
     def hook(self, entry):
-        return _bootstrap._ExtensionFileFinder(entry)
+        return _bootstrap._file_path_hook(entry)
 
     def test_success(self):
         # Path hook should handle a directory where a known extension module

Modified: python/branches/py3k-cdecimal/Lib/importlib/test/import_/test_path.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/importlib/test/import_/test_path.py	(original)
+++ python/branches/py3k-cdecimal/Lib/importlib/test/import_/test_path.py	Sat Jul 17 12:36:09 2010
@@ -5,6 +5,7 @@
 import imp
 import os
 import sys
+import tempfile
 from test import support
 from types import MethodType
 import unittest
@@ -80,23 +81,28 @@
 
     def test_implicit_hooks(self):
         # Test that the implicit path hooks are used.
-        existing_path = os.path.dirname(support.TESTFN)
         bad_path = '<path>'
         module = '<module>'
         assert not os.path.exists(bad_path)
-        with util.import_state():
-            nothing = _bootstrap._DefaultPathFinder.find_module(module,
-                                                           path=[existing_path])
-            self.assertTrue(nothing is None)
-            self.assertTrue(existing_path in sys.path_importer_cache)
-            self.assertTrue(not isinstance(sys.path_importer_cache[existing_path],
-                                        imp.NullImporter))
-            nothing = _bootstrap._DefaultPathFinder.find_module(module,
-                                                                path=[bad_path])
-            self.assertTrue(nothing is None)
-            self.assertTrue(bad_path in sys.path_importer_cache)
-            self.assertTrue(isinstance(sys.path_importer_cache[bad_path],
-                                    imp.NullImporter))
+        existing_path = tempfile.mkdtemp()
+        try:
+            with util.import_state():
+                nothing = _bootstrap._DefaultPathFinder.find_module(module,
+                                                        path=[existing_path])
+                self.assertTrue(nothing is None)
+                self.assertTrue(existing_path in sys.path_importer_cache)
+                result = isinstance(sys.path_importer_cache[existing_path],
+                                    imp.NullImporter)
+                self.assertFalse(result)
+                nothing = _bootstrap._DefaultPathFinder.find_module(module,
+                                                            path=[bad_path])
+                self.assertTrue(nothing is None)
+                self.assertTrue(bad_path in sys.path_importer_cache)
+                self.assertTrue(isinstance(sys.path_importer_cache[bad_path],
+                                           imp.NullImporter))
+        finally:
+            os.rmdir(existing_path)
+
 
     def test_path_importer_cache_has_None(self):
         # Test that the default hook is used when sys.path_importer_cache

Modified: python/branches/py3k-cdecimal/Lib/importlib/test/regrtest.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/importlib/test/regrtest.py	(original)
+++ python/branches/py3k-cdecimal/Lib/importlib/test/regrtest.py	Sat Jul 17 12:36:09 2010
@@ -6,9 +6,11 @@
 this script.
 
 XXX FAILING
-    test_import
-        execution bit
+  * test_import
+    - test_incorrect_code_name
         file name differing between __file__ and co_filename (r68360 on trunk)
+    - test_import_by_filename
+        exception for trying to import by file name does not match
 
 """
 import importlib

Modified: python/branches/py3k-cdecimal/Lib/importlib/test/source/test_abc_loader.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/importlib/test/source/test_abc_loader.py	(original)
+++ python/branches/py3k-cdecimal/Lib/importlib/test/source/test_abc_loader.py	Sat Jul 17 12:36:09 2010
@@ -1,14 +1,68 @@
 import importlib
 from importlib import abc
+
 from .. import abc as testing_abc
 from .. import util
 from . import util as source_util
+
 import imp
+import inspect
+import io
 import marshal
 import os
 import sys
 import types
 import unittest
+import warnings
+
+
+class SourceOnlyLoaderMock(abc.SourceLoader):
+
+    # Globals that should be defined for all modules.
+    source = (b"_ = '::'.join([__name__, __file__, __cached__, __package__, "
+              b"repr(__loader__)])")
+
+    def __init__(self, path):
+        self.path = path
+
+    def get_data(self, path):
+        assert self.path == path
+        return self.source
+
+    def get_filename(self, fullname):
+        return self.path
+
+
+class SourceLoaderMock(SourceOnlyLoaderMock):
+
+    source_mtime = 1
+
+    def __init__(self, path, magic=imp.get_magic()):
+        super().__init__(path)
+        self.bytecode_path = imp.cache_from_source(self.path)
+        data = bytearray(magic)
+        data.extend(marshal._w_long(self.source_mtime))
+        code_object = compile(self.source, self.path, 'exec',
+                                dont_inherit=True)
+        data.extend(marshal.dumps(code_object))
+        self.bytecode = bytes(data)
+        self.written = {}
+
+    def get_data(self, path):
+        if path == self.path:
+            return super().get_data(path)
+        elif path == self.bytecode_path:
+            return self.bytecode
+        else:
+            raise IOError
+
+    def path_mtime(self, path):
+        assert path == self.path
+        return self.source_mtime
+
+    def set_data(self, path, data):
+        self.written[path] = bytes(data)
+        return path == self.bytecode_path
 
 
 class PyLoaderMock(abc.PyLoader):
@@ -33,17 +87,42 @@
         return self.source
 
     def is_package(self, name):
+        filename = os.path.basename(self.get_filename(name))
+        return os.path.splitext(filename)[0] == '__init__'
+
+    def source_path(self, name):
         try:
-            return '__init__' in self.module_paths[name]
+            return self.module_paths[name]
         except KeyError:
             raise ImportError
 
-    def source_path(self, name):
+    def get_filename(self, name):
+        """Silence deprecation warning."""
+        with warnings.catch_warnings(record=True) as w:
+            warnings.simplefilter("always")
+            path = super().get_filename(name)
+            assert len(w) == 1
+            assert issubclass(w[0].category, PendingDeprecationWarning)
+            return path
+
+
+class PyLoaderCompatMock(PyLoaderMock):
+
+    """Mock that matches what is suggested to have a loader that is compatible
+    from Python 3.1 onwards."""
+
+    def get_filename(self, fullname):
         try:
-            return self.module_paths[name]
+            return self.module_paths[fullname]
         except KeyError:
             raise ImportError
 
+    def source_path(self, fullname):
+        try:
+            return self.get_filename(fullname)
+        except ImportError:
+            return None
+
 
 class PyPycLoaderMock(abc.PyPycLoader, PyLoaderMock):
 
@@ -114,6 +193,13 @@
         except TypeError:
             return '__init__' in self.bytecode_to_path[name]
 
+    def get_code(self, name):
+        with warnings.catch_warnings(record=True) as w:
+            warnings.simplefilter("always")
+            code_object = super().get_code(name)
+            assert len(w) == 1
+            assert issubclass(w[0].category, PendingDeprecationWarning)
+            return code_object
 
 class PyLoaderTests(testing_abc.LoaderTests):
 
@@ -200,6 +286,14 @@
         return mock
 
 
+class PyLoaderCompatTests(PyLoaderTests):
+
+    """Test that the suggested code to make a loader that is compatible from
+    Python 3.1 forward works."""
+
+    mocker = PyLoaderCompatMock
+
+
 class PyLoaderInterfaceTests(unittest.TestCase):
 
     """Tests for importlib.abc.PyLoader to make sure that when source_path()
@@ -234,30 +328,6 @@
             mock.get_filename(name)
 
 
-class PyLoaderGetSourceTests(unittest.TestCase):
-
-    """Tests for importlib.abc.PyLoader.get_source()."""
-
-    def test_default_encoding(self):
-        # Should have no problems with UTF-8 text.
-        name = 'mod'
-        mock = PyLoaderMock({name: os.path.join('path', 'to', 'mod')})
-        source = 'x = "??"'
-        mock.source = source.encode('utf-8')
-        returned_source = mock.get_source(name)
-        self.assertEqual(returned_source, source)
-
-    def test_decoded_source(self):
-        # Decoding should work.
-        name = 'mod'
-        mock = PyLoaderMock({name: os.path.join('path', 'to', 'mod')})
-        source = "# coding: Latin-1\nx='??'"
-        assert source.encode('latin-1') != source.encode('utf-8')
-        mock.source = source.encode('latin-1')
-        returned_source = mock.get_source(name)
-        self.assertEqual(returned_source, source)
-
-
 class PyPycLoaderTests(PyLoaderTests):
 
     """Tests for importlib.abc.PyPycLoader."""
@@ -413,7 +483,7 @@
     def test_bad_bytecode(self):
         # Malformed code object bytecode should lead to a ValueError.
         name = 'mod'
-        bc = {name: {'path': os.path.join('path', 'to', 'mod'), 'bc': b'XXX'}}
+        bc = {name: {'path': os.path.join('path', 'to', 'mod'), 'bc': b'1234'}}
         mock = PyPycLoaderMock({name: None}, bc)
         with util.uncache(name), self.assertRaises(ValueError):
             mock.load_module(name)
@@ -465,12 +535,341 @@
             mock.load_module(name)
 
 
+class SourceLoaderTestHarness(unittest.TestCase):
+
+    def setUp(self, *, is_package=True, **kwargs):
+        self.package = 'pkg'
+        if is_package:
+            self.path = os.path.join(self.package, '__init__.py')
+            self.name = self.package
+        else:
+            module_name = 'mod'
+            self.path = os.path.join(self.package, '.'.join(['mod', 'py']))
+            self.name = '.'.join([self.package, module_name])
+        self.cached = imp.cache_from_source(self.path)
+        self.loader = self.loader_mock(self.path, **kwargs)
+
+    def verify_module(self, module):
+        self.assertEqual(module.__name__, self.name)
+        self.assertEqual(module.__file__, self.path)
+        self.assertEqual(module.__cached__, self.cached)
+        self.assertEqual(module.__package__, self.package)
+        self.assertEqual(module.__loader__, self.loader)
+        values = module._.split('::')
+        self.assertEqual(values[0], self.name)
+        self.assertEqual(values[1], self.path)
+        self.assertEqual(values[2], self.cached)
+        self.assertEqual(values[3], self.package)
+        self.assertEqual(values[4], repr(self.loader))
+
+    def verify_code(self, code_object):
+        module = imp.new_module(self.name)
+        module.__file__ = self.path
+        module.__cached__ = self.cached
+        module.__package__ = self.package
+        module.__loader__ = self.loader
+        module.__path__ = []
+        exec(code_object, module.__dict__)
+        self.verify_module(module)
+
+
+class SourceOnlyLoaderTests(SourceLoaderTestHarness):
+
+    """Test importlib.abc.SourceLoader for source-only loading.
+
+    Reload testing is subsumed by the tests for
+    importlib.util.module_for_loader.
+
+    """
+
+    loader_mock = SourceOnlyLoaderMock
+
+    def test_get_source(self):
+        # Verify the source code is returned as a string.
+        # If an IOError is raised by get_data then raise ImportError.
+        expected_source = self.loader.source.decode('utf-8')
+        self.assertEqual(self.loader.get_source(self.name), expected_source)
+        def raise_IOError(path):
+            raise IOError
+        self.loader.get_data = raise_IOError
+        with self.assertRaises(ImportError):
+            self.loader.get_source(self.name)
+
+    def test_is_package(self):
+        # Properly detect when loading a package.
+        self.setUp(is_package=True)
+        self.assertTrue(self.loader.is_package(self.name))
+        self.setUp(is_package=False)
+        self.assertFalse(self.loader.is_package(self.name))
+
+    def test_get_code(self):
+        # Verify the code object is created.
+        code_object = self.loader.get_code(self.name)
+        self.verify_code(code_object)
+
+    def test_load_module(self):
+        # Loading a module should set __name__, __loader__, __package__,
+        # __path__ (for packages), __file__, and __cached__.
+        # The module should also be put into sys.modules.
+        with util.uncache(self.name):
+            module = self.loader.load_module(self.name)
+            self.verify_module(module)
+            self.assertEqual(module.__path__, [os.path.dirname(self.path)])
+            self.assertTrue(self.name in sys.modules)
+
+    def test_package_settings(self):
+        # __package__ needs to be set, while __path__ is set on if the module
+        # is a package.
+        # Testing the values for a package are covered by test_load_module.
+        self.setUp(is_package=False)
+        with util.uncache(self.name):
+            module = self.loader.load_module(self.name)
+            self.verify_module(module)
+            self.assertTrue(not hasattr(module, '__path__'))
+
+    def test_get_source_encoding(self):
+        # Source is considered encoded in UTF-8 by default unless otherwise
+        # specified by an encoding line.
+        source = "_ = '??'"
+        self.loader.source = source.encode('utf-8')
+        returned_source = self.loader.get_source(self.name)
+        self.assertEqual(returned_source, source)
+        source = "# coding: latin-1\n_ = ??"
+        self.loader.source = source.encode('latin-1')
+        returned_source = self.loader.get_source(self.name)
+        self.assertEqual(returned_source, source)
+
+
+ at unittest.skipIf(sys.dont_write_bytecode, "sys.dont_write_bytecode is true")
+class SourceLoaderBytecodeTests(SourceLoaderTestHarness):
+
+    """Test importlib.abc.SourceLoader's use of bytecode.
+
+    Source-only testing handled by SourceOnlyLoaderTests.
+
+    """
+
+    loader_mock = SourceLoaderMock
+
+    def verify_code(self, code_object, *, bytecode_written=False):
+        super().verify_code(code_object)
+        if bytecode_written:
+            self.assertIn(self.cached, self.loader.written)
+            data = bytearray(imp.get_magic())
+            data.extend(marshal._w_long(self.loader.source_mtime))
+            data.extend(marshal.dumps(code_object))
+            self.assertEqual(self.loader.written[self.cached], bytes(data))
+
+    def test_code_with_everything(self):
+        # When everything should work.
+        code_object = self.loader.get_code(self.name)
+        self.verify_code(code_object)
+
+    def test_no_bytecode(self):
+        # If no bytecode exists then move on to the source.
+        self.loader.bytecode_path = "<does not exist>"
+        # Sanity check
+        with self.assertRaises(IOError):
+            bytecode_path = imp.cache_from_source(self.path)
+            self.loader.get_data(bytecode_path)
+        code_object = self.loader.get_code(self.name)
+        self.verify_code(code_object, bytecode_written=True)
+
+    def test_code_bad_timestamp(self):
+        # Bytecode is only used when the timestamp matches the source EXACTLY.
+        for source_mtime in (0, 2):
+            assert source_mtime != self.loader.source_mtime
+            original = self.loader.source_mtime
+            self.loader.source_mtime = source_mtime
+            # If bytecode is used then EOFError would be raised by marshal.
+            self.loader.bytecode = self.loader.bytecode[8:]
+            code_object = self.loader.get_code(self.name)
+            self.verify_code(code_object, bytecode_written=True)
+            self.loader.source_mtime = original
+
+    def test_code_bad_magic(self):
+        # Skip over bytecode with a bad magic number.
+        self.setUp(magic=b'0000')
+        # If bytecode is used then EOFError would be raised by marshal.
+        self.loader.bytecode = self.loader.bytecode[8:]
+        code_object = self.loader.get_code(self.name)
+        self.verify_code(code_object, bytecode_written=True)
+
+    def test_dont_write_bytecode(self):
+        # Bytecode is not written if sys.dont_write_bytecode is true.
+        # Can assume it is false already thanks to the skipIf class decorator.
+        try:
+            sys.dont_write_bytecode = True
+            self.loader.bytecode_path = "<does not exist>"
+            code_object = self.loader.get_code(self.name)
+            self.assertNotIn(self.cached, self.loader.written)
+        finally:
+            sys.dont_write_bytecode = False
+
+    def test_no_set_data(self):
+        # If set_data is not defined, one can still read bytecode.
+        self.setUp(magic=b'0000')
+        original_set_data = self.loader.__class__.set_data
+        try:
+            del self.loader.__class__.set_data
+            code_object = self.loader.get_code(self.name)
+            self.verify_code(code_object)
+        finally:
+            self.loader.__class__.set_data = original_set_data
+
+    def test_set_data_raises_exceptions(self):
+        # Raising NotImplementedError or IOError is okay for set_data.
+        def raise_exception(exc):
+            def closure(*args, **kwargs):
+                raise exc
+            return closure
+
+        self.setUp(magic=b'0000')
+        self.loader.set_data = raise_exception(NotImplementedError)
+        code_object = self.loader.get_code(self.name)
+        self.verify_code(code_object)
+
+
+class SourceLoaderGetSourceTests(unittest.TestCase):
+
+    """Tests for importlib.abc.SourceLoader.get_source()."""
+
+    def test_default_encoding(self):
+        # Should have no problems with UTF-8 text.
+        name = 'mod'
+        mock = SourceOnlyLoaderMock('mod.file')
+        source = 'x = "??"'
+        mock.source = source.encode('utf-8')
+        returned_source = mock.get_source(name)
+        self.assertEqual(returned_source, source)
+
+    def test_decoded_source(self):
+        # Decoding should work.
+        name = 'mod'
+        mock = SourceOnlyLoaderMock("mod.file")
+        source = "# coding: Latin-1\nx='??'"
+        assert source.encode('latin-1') != source.encode('utf-8')
+        mock.source = source.encode('latin-1')
+        returned_source = mock.get_source(name)
+        self.assertEqual(returned_source, source)
+
+    def test_universal_newlines(self):
+        # PEP 302 says universal newlines should be used.
+        name = 'mod'
+        mock = SourceOnlyLoaderMock('mod.file')
+        source = "x = 42\r\ny = -13\r\n"
+        mock.source = source.encode('utf-8')
+        expect = io.IncrementalNewlineDecoder(None, True).decode(source)
+        self.assertEqual(mock.get_source(name), expect)
+
+class AbstractMethodImplTests(unittest.TestCase):
+
+    """Test the concrete abstractmethod implementations."""
+
+    class Loader(abc.Loader):
+        def load_module(self, fullname):
+            super().load_module(fullname)
+
+    class Finder(abc.Finder):
+        def find_module(self, _):
+            super().find_module(_)
+
+    class ResourceLoader(Loader, abc.ResourceLoader):
+        def get_data(self, _):
+            super().get_data(_)
+
+    class InspectLoader(Loader, abc.InspectLoader):
+        def is_package(self, _):
+            super().is_package(_)
+
+        def get_code(self, _):
+            super().get_code(_)
+
+        def get_source(self, _):
+            super().get_source(_)
+
+    class ExecutionLoader(InspectLoader, abc.ExecutionLoader):
+        def get_filename(self, _):
+            super().get_filename(_)
+
+    class SourceLoader(ResourceLoader, ExecutionLoader, abc.SourceLoader):
+        pass
+
+    class PyLoader(ResourceLoader, InspectLoader, abc.PyLoader):
+        def source_path(self, _):
+            super().source_path(_)
+
+    class PyPycLoader(PyLoader, abc.PyPycLoader):
+        def bytecode_path(self, _):
+            super().bytecode_path(_)
+
+        def source_mtime(self, _):
+            super().source_mtime(_)
+
+        def write_bytecode(self, _, _2):
+            super().write_bytecode(_, _2)
+
+    def raises_NotImplementedError(self, ins, *args):
+        for method_name in args:
+            method = getattr(ins, method_name)
+            arg_count = len(inspect.getfullargspec(method)[0]) - 1
+            args = [''] * arg_count
+            try:
+                method(*args)
+            except NotImplementedError:
+                pass
+            else:
+                msg = "{}.{} did not raise NotImplementedError"
+                self.fail(msg.format(ins.__class__.__name__, method_name))
+
+    def test_Loader(self):
+        self.raises_NotImplementedError(self.Loader(), 'load_module')
+
+    # XXX misplaced; should be somewhere else
+    def test_Finder(self):
+        self.raises_NotImplementedError(self.Finder(), 'find_module')
+
+    def test_ResourceLoader(self):
+        self.raises_NotImplementedError(self.ResourceLoader(), 'load_module',
+                                        'get_data')
+
+    def test_InspectLoader(self):
+        self.raises_NotImplementedError(self.InspectLoader(), 'load_module',
+                                        'is_package', 'get_code', 'get_source')
+
+    def test_ExecutionLoader(self):
+        self.raises_NotImplementedError(self.ExecutionLoader(), 'load_module',
+                                        'is_package', 'get_code', 'get_source',
+                                        'get_filename')
+
+    def test_SourceLoader(self):
+        ins = self.SourceLoader()
+        # Required abstractmethods.
+        self.raises_NotImplementedError(ins, 'get_filename', 'get_data')
+        # Optional abstractmethods.
+        self.raises_NotImplementedError(ins,'path_mtime', 'set_data')
+
+    def test_PyLoader(self):
+        self.raises_NotImplementedError(self.PyLoader(), 'source_path',
+                                        'get_data', 'is_package')
+
+    def test_PyPycLoader(self):
+        self.raises_NotImplementedError(self.PyPycLoader(), 'source_path',
+                                        'source_mtime', 'bytecode_path',
+                                        'write_bytecode')
+
+
 def test_main():
     from test.support import run_unittest
-    run_unittest(PyLoaderTests, PyLoaderInterfaceTests, PyLoaderGetSourceTests,
+    run_unittest(PyLoaderTests, PyLoaderCompatTests,
+                    PyLoaderInterfaceTests,
                     PyPycLoaderTests, PyPycLoaderInterfaceTests,
                     SkipWritingBytecodeTests, RegeneratedBytecodeTests,
-                    BadBytecodeFailureTests, MissingPathsTests)
+                    BadBytecodeFailureTests, MissingPathsTests,
+                    SourceOnlyLoaderTests,
+                    SourceLoaderBytecodeTests,
+                    SourceLoaderGetSourceTests,
+                    AbstractMethodImplTests)
 
 
 if __name__ == '__main__':

Modified: python/branches/py3k-cdecimal/Lib/importlib/test/source/test_case_sensitivity.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/importlib/test/source/test_case_sensitivity.py	(original)
+++ python/branches/py3k-cdecimal/Lib/importlib/test/source/test_case_sensitivity.py	Sat Jul 17 12:36:09 2010
@@ -19,7 +19,9 @@
     assert name != name.lower()
 
     def find(self, path):
-        finder = _bootstrap._PyPycFileFinder(path)
+        finder = _bootstrap._FileFinder(path,
+                                        _bootstrap._SourceFinderDetails(),
+                                        _bootstrap._SourcelessFinderDetails())
         return finder.find_module(self.name)
 
     def sensitivity_test(self):
@@ -27,7 +29,7 @@
         sensitive_pkg = 'sensitive.{0}'.format(self.name)
         insensitive_pkg = 'insensitive.{0}'.format(self.name.lower())
         context = source_util.create_modules(insensitive_pkg, sensitive_pkg)
-        with  context as mapping:
+        with context as mapping:
             sensitive_path = os.path.join(mapping['.root'], 'sensitive')
             insensitive_path = os.path.join(mapping['.root'], 'insensitive')
             return self.find(sensitive_path), self.find(insensitive_path)
@@ -37,7 +39,7 @@
             env.unset('PYTHONCASEOK')
             sensitive, insensitive = self.sensitivity_test()
             self.assertTrue(hasattr(sensitive, 'load_module'))
-            self.assertIn(self.name, sensitive._base_path)
+            self.assertIn(self.name, sensitive.get_filename(self.name))
             self.assertIsNone(insensitive)
 
     def test_insensitive(self):
@@ -45,9 +47,9 @@
             env.set('PYTHONCASEOK', '1')
             sensitive, insensitive = self.sensitivity_test()
             self.assertTrue(hasattr(sensitive, 'load_module'))
-            self.assertIn(self.name, sensitive._base_path)
+            self.assertIn(self.name, sensitive.get_filename(self.name))
             self.assertTrue(hasattr(insensitive, 'load_module'))
-            self.assertIn(self.name, insensitive._base_path)
+            self.assertIn(self.name, insensitive.get_filename(self.name))
 
 
 def test_main():

Modified: python/branches/py3k-cdecimal/Lib/importlib/test/source/test_file_loader.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/importlib/test/source/test_file_loader.py	(original)
+++ python/branches/py3k-cdecimal/Lib/importlib/test/source/test_file_loader.py	Sat Jul 17 12:36:09 2010
@@ -1,9 +1,11 @@
 import importlib
 from importlib import _bootstrap
 from .. import abc
+from .. import util
 from . import util as source_util
 
 import imp
+import marshal
 import os
 import py_compile
 import stat
@@ -23,8 +25,7 @@
     # [basic]
     def test_module(self):
         with source_util.create_modules('_temp') as mapping:
-            loader = _bootstrap._PyPycFileLoader('_temp', mapping['_temp'],
-                                                    False)
+            loader = _bootstrap._SourceFileLoader('_temp', mapping['_temp'])
             module = loader.load_module('_temp')
             self.assertTrue('_temp' in sys.modules)
             check = {'__name__': '_temp', '__file__': mapping['_temp'],
@@ -34,9 +35,8 @@
 
     def test_package(self):
         with source_util.create_modules('_pkg.__init__') as mapping:
-            loader = _bootstrap._PyPycFileLoader('_pkg',
-                                                 mapping['_pkg.__init__'],
-                                                 True)
+            loader = _bootstrap._SourceFileLoader('_pkg',
+                                                 mapping['_pkg.__init__'])
             module = loader.load_module('_pkg')
             self.assertTrue('_pkg' in sys.modules)
             check = {'__name__': '_pkg', '__file__': mapping['_pkg.__init__'],
@@ -48,8 +48,8 @@
 
     def test_lacking_parent(self):
         with source_util.create_modules('_pkg.__init__', '_pkg.mod')as mapping:
-            loader = _bootstrap._PyPycFileLoader('_pkg.mod',
-                                                    mapping['_pkg.mod'], False)
+            loader = _bootstrap._SourceFileLoader('_pkg.mod',
+                                                    mapping['_pkg.mod'])
             module = loader.load_module('_pkg.mod')
             self.assertTrue('_pkg.mod' in sys.modules)
             check = {'__name__': '_pkg.mod', '__file__': mapping['_pkg.mod'],
@@ -63,8 +63,7 @@
 
     def test_module_reuse(self):
         with source_util.create_modules('_temp') as mapping:
-            loader = _bootstrap._PyPycFileLoader('_temp', mapping['_temp'],
-                                                    False)
+            loader = _bootstrap._SourceFileLoader('_temp', mapping['_temp'])
             module = loader.load_module('_temp')
             module_id = id(module)
             module_dict_id = id(module.__dict__)
@@ -74,7 +73,7 @@
             # everything that has happened above can be too fast;
             # force an mtime on the source that is guaranteed to be different
             # than the original mtime.
-            loader.source_mtime = self.fake_mtime(loader.source_mtime)
+            loader.path_mtime = self.fake_mtime(loader.path_mtime)
             module = loader.load_module('_temp')
             self.assertTrue('testing_var' in module.__dict__,
                          "'testing_var' not in "
@@ -94,8 +93,7 @@
                 setattr(orig_module, attr, value)
             with open(mapping[name], 'w') as file:
                 file.write('+++ bad syntax +++')
-            loader = _bootstrap._PyPycFileLoader('_temp', mapping['_temp'],
-                                                    False)
+            loader = _bootstrap._SourceFileLoader('_temp', mapping['_temp'])
             with self.assertRaises(SyntaxError):
                 loader.load_module(name)
             for attr in attributes:
@@ -106,17 +104,32 @@
         with source_util.create_modules('_temp') as mapping:
             with open(mapping['_temp'], 'w') as file:
                 file.write('=')
-            loader = _bootstrap._PyPycFileLoader('_temp', mapping['_temp'],
-                                                    False)
+            loader = _bootstrap._SourceFileLoader('_temp', mapping['_temp'])
             with self.assertRaises(SyntaxError):
                 loader.load_module('_temp')
             self.assertTrue('_temp' not in sys.modules)
 
+    def test_file_from_empty_string_dir(self):
+        # Loading a module found from an empty string entry on sys.path should
+        # not only work, but keep all attributes relative.
+        with open('_temp.py', 'w') as file:
+            file.write("# test file for importlib")
+        try:
+            with util.uncache('_temp'):
+                loader = _bootstrap._SourceFileLoader('_temp', '_temp.py')
+                mod = loader.load_module('_temp')
+                self.assertEqual('_temp.py', mod.__file__)
+                self.assertEqual(imp.cache_from_source('_temp.py'),
+                                 mod.__cached__)
+
+        finally:
+            os.unlink('_temp.py')
+
 
 class BadBytecodeTest(unittest.TestCase):
 
     def import_(self, file, module_name):
-        loader = _bootstrap._PyPycFileLoader(module_name, file, False)
+        loader = self.loader(module_name, file)
         module = loader.load_module(module_name)
         self.assertTrue(module_name in sys.modules)
 
@@ -129,101 +142,156 @@
         except KeyError:
             pass
         py_compile.compile(mapping[name])
-        bytecode_path = imp.cache_from_source(mapping[name])
-        with open(bytecode_path, 'rb') as file:
-            bc = file.read()
-        new_bc = manipulator(bc)
-        with open(bytecode_path, 'wb') as file:
-            if new_bc:
-                file.write(new_bc)
-        if del_source:
+        if not del_source:
+            bytecode_path = imp.cache_from_source(mapping[name])
+        else:
             os.unlink(mapping[name])
-            make_legacy_pyc(mapping[name])
+            bytecode_path = make_legacy_pyc(mapping[name])
+        if manipulator:
+            with open(bytecode_path, 'rb') as file:
+                bc = file.read()
+                new_bc = manipulator(bc)
+            with open(bytecode_path, 'wb') as file:
+                if new_bc is not None:
+                    file.write(new_bc)
         return bytecode_path
 
-    @source_util.writes_bytecode_files
-    def test_empty_file(self):
-        # When a .pyc is empty, regenerate it if possible, else raise
-        # ImportError.
+    def _test_empty_file(self, test, *, del_source=False):
         with source_util.create_modules('_temp') as mapping:
             bc_path = self.manipulate_bytecode('_temp', mapping,
-                                                lambda bc: None)
-            self.import_(mapping['_temp'], '_temp')
-            with open(bc_path, 'rb') as file:
-                self.assertGreater(len(file.read()), 8)
-            self.manipulate_bytecode('_temp', mapping, lambda bc: None,
-                                        del_source=True)
-            with self.assertRaises(ImportError):
-                self.import_(mapping['_temp'], '_temp')
+                                                lambda bc: b'',
+                                                del_source=del_source)
+            test('_temp', mapping, bc_path)
 
     @source_util.writes_bytecode_files
-    def test_partial_magic(self):
+    def _test_partial_magic(self, test, *, del_source=False):
         # When their are less than 4 bytes to a .pyc, regenerate it if
         # possible, else raise ImportError.
         with source_util.create_modules('_temp') as mapping:
             bc_path = self.manipulate_bytecode('_temp', mapping,
-                                                lambda bc: bc[:3])
-            self.import_(mapping['_temp'], '_temp')
-            with open(bc_path, 'rb') as file:
-                self.assertGreater(len(file.read()), 8)
-            self.manipulate_bytecode('_temp', mapping, lambda bc: bc[:3],
-                                        del_source=True)
+                                                lambda bc: bc[:3],
+                                                del_source=del_source)
+            test('_temp', mapping, bc_path)
+
+    def _test_magic_only(self, test, *, del_source=False):
+        with source_util.create_modules('_temp') as mapping:
+            bc_path = self.manipulate_bytecode('_temp', mapping,
+                                                lambda bc: bc[:4],
+                                                del_source=del_source)
+            test('_temp', mapping, bc_path)
+
+    def _test_partial_timestamp(self, test, *, del_source=False):
+        with source_util.create_modules('_temp') as mapping:
+            bc_path = self.manipulate_bytecode('_temp', mapping,
+                                                lambda bc: bc[:7],
+                                                del_source=del_source)
+            test('_temp', mapping, bc_path)
+
+    def _test_no_marshal(self, *, del_source=False):
+        with source_util.create_modules('_temp') as mapping:
+            bc_path = self.manipulate_bytecode('_temp', mapping,
+                                                lambda bc: bc[:8],
+                                                del_source=del_source)
+            file_path = mapping['_temp'] if not del_source else bc_path
+            with self.assertRaises(EOFError):
+                self.import_(file_path, '_temp')
+
+    def _test_non_code_marshal(self, *, del_source=False):
+        with source_util.create_modules('_temp') as mapping:
+            bytecode_path = self.manipulate_bytecode('_temp', mapping,
+                                    lambda bc: bc[:8] + marshal.dumps(b'abcd'),
+                                    del_source=del_source)
+            file_path = mapping['_temp'] if not del_source else bytecode_path
             with self.assertRaises(ImportError):
-                self.import_(mapping['_temp'], '_temp')
+                self.import_(file_path, '_temp')
+
+    def _test_bad_marshal(self, *, del_source=False):
+        with source_util.create_modules('_temp') as mapping:
+            bytecode_path = self.manipulate_bytecode('_temp', mapping,
+                                                lambda bc: bc[:8] + b'<test>',
+                                                del_source=del_source)
+            file_path = mapping['_temp'] if not del_source else bytecode_path
+            with self.assertRaises(ValueError):
+                self.import_(file_path, '_temp')
+
+    def _test_bad_magic(self, test, *, del_source=False):
+        with source_util.create_modules('_temp') as mapping:
+            bc_path = self.manipulate_bytecode('_temp', mapping,
+                                    lambda bc: b'\x00\x00\x00\x00' + bc[4:])
+            test('_temp', mapping, bc_path)
+
+
+class SourceLoaderBadBytecodeTest(BadBytecodeTest):
+
+    loader = _bootstrap._SourceFileLoader
+
+    @source_util.writes_bytecode_files
+    def test_empty_file(self):
+        # When a .pyc is empty, regenerate it if possible, else raise
+        # ImportError.
+        def test(name, mapping, bytecode_path):
+            self.import_(mapping[name], name)
+            with open(bytecode_path, 'rb') as file:
+                self.assertGreater(len(file.read()), 8)
+
+        self._test_empty_file(test)
+
+    def test_partial_magic(self):
+        def test(name, mapping, bytecode_path):
+            self.import_(mapping[name], name)
+            with open(bytecode_path, 'rb') as file:
+                self.assertGreater(len(file.read()), 8)
+
+        self._test_partial_magic(test)
 
     @source_util.writes_bytecode_files
     def test_magic_only(self):
         # When there is only the magic number, regenerate the .pyc if possible,
         # else raise EOFError.
-        with source_util.create_modules('_temp') as mapping:
-            bc_path = self.manipulate_bytecode('_temp', mapping,
-                                                lambda bc: bc[:4])
-            self.import_(mapping['_temp'], '_temp')
-            with open(bc_path, 'rb') as file:
+        def test(name, mapping, bytecode_path):
+            self.import_(mapping[name], name)
+            with open(bytecode_path, 'rb') as file:
                 self.assertGreater(len(file.read()), 8)
-            self.manipulate_bytecode('_temp', mapping, lambda bc: bc[:4],
-                                        del_source=True)
-            with self.assertRaises(EOFError):
-                self.import_(mapping['_temp'], '_temp')
+
+    @source_util.writes_bytecode_files
+    def test_bad_magic(self):
+        # When the magic number is different, the bytecode should be
+        # regenerated.
+        def test(name, mapping, bytecode_path):
+            self.import_(mapping[name], name)
+            with open(bytecode_path, 'rb') as bytecode_file:
+                self.assertEqual(bytecode_file.read(4), imp.get_magic())
+
+        self._test_bad_magic(test)
 
     @source_util.writes_bytecode_files
     def test_partial_timestamp(self):
         # When the timestamp is partial, regenerate the .pyc, else
         # raise EOFError.
-        with source_util.create_modules('_temp') as mapping:
-            bc_path = self.manipulate_bytecode('_temp', mapping,
-                                                lambda bc: bc[:7])
-            self.import_(mapping['_temp'], '_temp')
+        def test(name, mapping, bc_path):
+            self.import_(mapping[name], name)
             with open(bc_path, 'rb') as file:
                 self.assertGreater(len(file.read()), 8)
-            self.manipulate_bytecode('_temp', mapping, lambda bc: bc[:7],
-                                        del_source=True)
-            with self.assertRaises(EOFError):
-                self.import_(mapping['_temp'], '_temp')
 
     @source_util.writes_bytecode_files
     def test_no_marshal(self):
         # When there is only the magic number and timestamp, raise EOFError.
-        with source_util.create_modules('_temp') as mapping:
-            bc_path = self.manipulate_bytecode('_temp', mapping,
-                                                lambda bc: bc[:8])
-            with self.assertRaises(EOFError):
-                self.import_(mapping['_temp'], '_temp')
+        self._test_no_marshal()
 
     @source_util.writes_bytecode_files
-    def test_bad_magic(self):
-        # When the magic number is different, the bytecode should be
-        # regenerated.
-        with source_util.create_modules('_temp') as mapping:
-            bc_path = self.manipulate_bytecode('_temp', mapping,
-                                    lambda bc: b'\x00\x00\x00\x00' + bc[4:])
-            self.import_(mapping['_temp'], '_temp')
-            with open(bc_path, 'rb') as bytecode_file:
-                self.assertEqual(bytecode_file.read(4), imp.get_magic())
+    def test_non_code_marshal(self):
+        self._test_non_code_marshal()
+        # XXX ImportError when sourceless
+
+    # [bad marshal]
+    @source_util.writes_bytecode_files
+    def test_bad_marshal(self):
+        # Bad marshal data should raise a ValueError.
+        self._test_bad_marshal()
 
     # [bad timestamp]
     @source_util.writes_bytecode_files
-    def test_bad_bytecode(self):
+    def test_old_timestamp(self):
         # When the timestamp is older than the source, bytecode should be
         # regenerated.
         zeros = b'\x00\x00\x00\x00'
@@ -240,23 +308,6 @@
                 bytecode_file.seek(4)
                 self.assertEqual(bytecode_file.read(4), source_timestamp)
 
-    # [bad marshal]
-    @source_util.writes_bytecode_files
-    def test_bad_marshal(self):
-        # Bad marshal data should raise a ValueError.
-        with source_util.create_modules('_temp') as mapping:
-            bytecode_path = imp.cache_from_source(mapping['_temp'])
-            source_mtime = os.path.getmtime(mapping['_temp'])
-            source_timestamp = importlib._w_long(source_mtime)
-            source_util.ensure_bytecode_path(bytecode_path)
-            with open(bytecode_path, 'wb') as bytecode_file:
-                bytecode_file.write(imp.get_magic())
-                bytecode_file.write(source_timestamp)
-                bytecode_file.write(b'AAAA')
-            with self.assertRaises(ValueError):
-                self.import_(mapping['_temp'], '_temp')
-            self.assertTrue('_temp' not in sys.modules)
-
     # [bytecode read-only]
     @source_util.writes_bytecode_files
     def test_read_only_bytecode(self):
@@ -279,9 +330,57 @@
                 os.chmod(bytecode_path, stat.S_IWUSR)
 
 
+class SourcelessLoaderBadBytecodeTest(BadBytecodeTest):
+
+    loader = _bootstrap._SourcelessFileLoader
+
+    def test_empty_file(self):
+        def test(name, mapping, bytecode_path):
+            with self.assertRaises(ImportError):
+                self.import_(bytecode_path, name)
+
+        self._test_empty_file(test, del_source=True)
+
+    def test_partial_magic(self):
+        def test(name, mapping, bytecode_path):
+            with self.assertRaises(ImportError):
+                self.import_(bytecode_path, name)
+        self._test_partial_magic(test, del_source=True)
+
+    def test_magic_only(self):
+        def test(name, mapping, bytecode_path):
+            with self.assertRaises(EOFError):
+                self.import_(bytecode_path, name)
+
+        self._test_magic_only(test, del_source=True)
+
+    def test_bad_magic(self):
+        def test(name, mapping, bytecode_path):
+            with self.assertRaises(ImportError):
+                self.import_(bytecode_path, name)
+
+        self._test_bad_magic(test, del_source=True)
+
+    def test_partial_timestamp(self):
+        def test(name, mapping, bytecode_path):
+            with self.assertRaises(EOFError):
+                self.import_(bytecode_path, name)
+
+        self._test_partial_timestamp(test, del_source=True)
+
+    def test_no_marshal(self):
+        self._test_no_marshal(del_source=True)
+
+    def test_non_code_marshal(self):
+        self._test_non_code_marshal(del_source=True)
+
+
 def test_main():
     from test.support import run_unittest
-    run_unittest(SimpleTest, BadBytecodeTest)
+    run_unittest(SimpleTest,
+                 SourceLoaderBadBytecodeTest,
+                 SourcelessLoaderBadBytecodeTest
+                 )
 
 
 if __name__ == '__main__':

Modified: python/branches/py3k-cdecimal/Lib/importlib/test/source/test_finder.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/importlib/test/source/test_finder.py	(original)
+++ python/branches/py3k-cdecimal/Lib/importlib/test/source/test_finder.py	Sat Jul 17 12:36:09 2010
@@ -34,7 +34,9 @@
     """
 
     def import_(self, root, module):
-        finder = _bootstrap._PyPycFileFinder(root)
+        finder = _bootstrap._FileFinder(root,
+                                        _bootstrap._SourceFinderDetails(),
+                                        _bootstrap._SourcelessFinderDetails())
         return finder.find_module(module)
 
     def run_test(self, test, create=None, *, compile_=None, unlink=None):
@@ -113,10 +115,9 @@
 
     # [package over modules]
     def test_package_over_module(self):
-        # XXX This is not a blackbox test!
         name = '_temp'
         loader = self.run_test(name, {'{0}.__init__'.format(name), name})
-        self.assertTrue('__init__' in loader._base_path)
+        self.assertTrue('__init__' in loader.get_filename(name))
 
 
     def test_failure(self):
@@ -131,6 +132,17 @@
             with self.assertRaises(ImportWarning):
                 self.run_test('pkg', {'pkg.__init__'}, unlink={'pkg.__init__'})
 
+    def test_empty_string_for_dir(self):
+        # The empty string from sys.path means to search in the cwd.
+        finder = _bootstrap._FileFinder('', _bootstrap._SourceFinderDetails())
+        with open('mod.py', 'w') as file:
+            file.write("# test file for importlib")
+        try:
+            loader = finder.find_module('mod')
+            self.assertTrue(hasattr(loader, 'load_module'))
+        finally:
+            os.unlink('mod.py')
+
 
 def test_main():
     from test.support import run_unittest

Modified: python/branches/py3k-cdecimal/Lib/importlib/test/source/test_path_hook.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/importlib/test/source/test_path_hook.py	(original)
+++ python/branches/py3k-cdecimal/Lib/importlib/test/source/test_path_hook.py	Sat Jul 17 12:36:09 2010
@@ -8,11 +8,14 @@
     """Test the path hook for source."""
 
     def test_success(self):
-        # XXX Only work on existing directories?
         with source_util.create_modules('dummy') as mapping:
-            self.assertTrue(hasattr(_bootstrap._FileFinder(mapping['.root']),
+            self.assertTrue(hasattr(_bootstrap._file_path_hook(mapping['.root']),
                                  'find_module'))
 
+    def test_empty_string(self):
+        # The empty string represents the cwd.
+        self.assertTrue(hasattr(_bootstrap._file_path_hook(''), 'find_module'))
+
 
 def test_main():
     from test.support import run_unittest

Modified: python/branches/py3k-cdecimal/Lib/importlib/test/source/test_source_encoding.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/importlib/test/source/test_source_encoding.py	(original)
+++ python/branches/py3k-cdecimal/Lib/importlib/test/source/test_source_encoding.py	Sat Jul 17 12:36:09 2010
@@ -35,8 +35,8 @@
         with source_util.create_modules(self.module_name) as mapping:
             with open(mapping[self.module_name], 'wb') as file:
                 file.write(source)
-            loader = _bootstrap._PyPycFileLoader(self.module_name,
-                                       mapping[self.module_name], False)
+            loader = _bootstrap._SourceFileLoader(self.module_name,
+                                       mapping[self.module_name])
             return loader.load_module(self.module_name)
 
     def create_source(self, encoding):
@@ -97,8 +97,8 @@
         with source_util.create_modules(module_name) as mapping:
             with open(mapping[module_name], 'wb') as file:
                 file.write(source)
-            loader = _bootstrap._PyPycFileLoader(module_name,
-                                                 mapping[module_name], False)
+            loader = _bootstrap._SourceFileLoader(module_name,
+                                                 mapping[module_name])
             return loader.load_module(module_name)
 
     # [cr]

Modified: python/branches/py3k-cdecimal/Lib/importlib/test/test_abc.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/importlib/test/test_abc.py	(original)
+++ python/branches/py3k-cdecimal/Lib/importlib/test/test_abc.py	Sat Jul 17 12:36:09 2010
@@ -59,6 +59,11 @@
     subclasses = [abc.PyLoader]
 
 
+class SourceLoader(InheritanceTests, unittest.TestCase):
+
+    superclasses = [abc.ResourceLoader, abc.ExecutionLoader]
+
+
 class PyLoader(InheritanceTests, unittest.TestCase):
 
     superclasses = [abc.Loader, abc.ResourceLoader, abc.ExecutionLoader]

Modified: python/branches/py3k-cdecimal/Lib/importlib/test/test_api.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/importlib/test/test_api.py	(original)
+++ python/branches/py3k-cdecimal/Lib/importlib/test/test_api.py	Sat Jul 17 12:36:09 2010
@@ -27,7 +27,7 @@
                 self.assertEqual(module.__name__, name)
 
     def test_shallow_relative_package_import(self):
-        # Test importing a module from a package through a relatve import.
+        # Test importing a module from a package through a relative import.
         pkg_name = 'pkg'
         pkg_long_name = '{0}.__init__'.format(pkg_name)
         module_name = 'mod'

Modified: python/branches/py3k-cdecimal/Lib/importlib/test/test_util.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/importlib/test/test_util.py	(original)
+++ python/branches/py3k-cdecimal/Lib/importlib/test/test_util.py	Sat Jul 17 12:36:09 2010
@@ -40,7 +40,7 @@
         with test_util.uncache(name):
             sys.modules[name] = module
             returned_module = self.return_module(name)
-            self.assertTrue(sys.modules[name] is returned_module)
+            self.assertIs(returned_module, sys.modules[name])
 
     def test_new_module_failure(self):
         # Test that a module is removed from sys.modules if added but an
@@ -57,7 +57,7 @@
         with test_util.uncache(name):
             sys.modules[name] = module
             self.raise_exception(name)
-            self.assertTrue(sys.modules[name] is module)
+            self.assertIs(module, sys.modules[name])
 
 
 class SetPackageTests(unittest.TestCase):

Modified: python/branches/py3k-cdecimal/Lib/inspect.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/inspect.py	(original)
+++ python/branches/py3k-cdecimal/Lib/inspect.py	Sat Jul 17 12:36:09 2010
@@ -437,7 +437,9 @@
     if info: return info[0]
 
 def getsourcefile(object):
-    """Return the Python source file an object was defined in, if it exists."""
+    """Return the filename that can be used to locate an object's source.
+    Return None if no way can be identified to get the source.
+    """
     filename = getfile(object)
     if filename[-4:].lower() in ('.pyc', '.pyo'):
         filename = filename[:-4] + '.py'
@@ -450,6 +452,9 @@
     # only return a non-existent filename if the module has a PEP 302 loader
     if hasattr(getmodule(object, filename), '__loader__'):
         return filename
+    # or it is in the linecache
+    if filename in linecache.cache:
+        return filename
 
 def getabsfile(object, _filename=None):
     """Return an absolute path to the source or compiled file for an object.
@@ -954,7 +959,7 @@
         else:
             arg2value[varargs] = ()
     elif 0 < num_args < num_pos:
-        raise TypeError('%s() takes %s %d %s (%d given)' % (
+        raise TypeError('%s() takes %s %d positional %s (%d given)' % (
             f_name, 'at most' if defaults else 'exactly', num_args,
             'arguments' if num_args > 1 else 'argument', num_total))
     elif num_args == 0 and num_total:

Modified: python/branches/py3k-cdecimal/Lib/lib2to3/fixes/fix_itertools_imports.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/lib2to3/fixes/fix_itertools_imports.py	(original)
+++ python/branches/py3k-cdecimal/Lib/lib2to3/fixes/fix_itertools_imports.py	Sat Jul 17 12:36:09 2010
@@ -20,6 +20,9 @@
             if child.type == token.NAME:
                 member = child.value
                 name_node = child
+            elif child.type == token.STAR:
+                # Just leave the import as is.
+                return
             else:
                 assert child.type == syms.import_as_name
                 name_node = child.children[0]
@@ -44,8 +47,8 @@
             children[-1].remove()
 
         # If there are no imports left, just get rid of the entire statement
-        if not (imports.children or getattr(imports, 'value', None)) or \
-                imports.parent is None:
+        if (not (imports.children or getattr(imports, 'value', None)) or
+            imports.parent is None):
             p = node.prefix
             node = BlankLine()
             node.prefix = p

Modified: python/branches/py3k-cdecimal/Lib/lib2to3/tests/test_fixers.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/lib2to3/tests/test_fixers.py	(original)
+++ python/branches/py3k-cdecimal/Lib/lib2to3/tests/test_fixers.py	Sat Jul 17 12:36:09 2010
@@ -3670,6 +3670,10 @@
         a = "from itertools import bar, filterfalse, foo"
         self.check(b, a)
 
+    def test_import_star(self):
+        s = "from itertools import *"
+        self.unchanged(s)
+
 
     def test_unchanged(self):
         s = "from itertools import foo"

Modified: python/branches/py3k-cdecimal/Lib/logging/config.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/logging/config.py	(original)
+++ python/branches/py3k-cdecimal/Lib/logging/config.py	Sat Jul 17 12:36:09 2010
@@ -386,15 +386,21 @@
         """
         name = s.split('.')
         used = name.pop(0)
-        found = self.importer(used)
-        for frag in name:
-            used += '.' + frag
-            try:
-                found = getattr(found, frag)
-            except AttributeError:
-                self.importer(used)
-                found = getattr(found, frag)
-        return found
+        try:
+            found = self.importer(used)
+            for frag in name:
+                used += '.' + frag
+                try:
+                    found = getattr(found, frag)
+                except AttributeError:
+                    self.importer(used)
+                    found = getattr(found, frag)
+            return found
+        except ImportError:
+            e, tb = sys.exc_info()[1:]
+            v = ValueError('Cannot resolve %r: %s' % (s, e))
+            v.__cause__, v.__traceback__ = e, tb
+            raise v
 
     def ext_convert(self, value):
         """Default converter for the ext:// protocol."""
@@ -873,6 +879,8 @@
         def run(self):
             server = self.rcvr(port=self.port, handler=self.hdlr,
                                ready=self.ready)
+            if self.port == 0:
+                self.port = server.server_address[1]
             self.ready.set()
             global _listener
             logging._acquireLock()

Modified: python/branches/py3k-cdecimal/Lib/macpath.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/macpath.py	(original)
+++ python/branches/py3k-cdecimal/Lib/macpath.py	Sat Jul 17 12:36:09 2010
@@ -32,6 +32,9 @@
 # Normalize the case of a pathname.  Dummy in Posix, but <s>.lower() here.
 
 def normcase(path):
+    if not isinstance(path, (bytes, str)):
+        raise TypeError("normcase() argument must be str or bytes, "
+                        "not '{}'".format(path.__class__.__name__))
     return path.lower()
 
 

Modified: python/branches/py3k-cdecimal/Lib/ntpath.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/ntpath.py	(original)
+++ python/branches/py3k-cdecimal/Lib/ntpath.py	Sat Jul 17 12:36:09 2010
@@ -16,7 +16,8 @@
            "getatime","getctime", "islink","exists","lexists","isdir","isfile",
            "ismount", "expanduser","expandvars","normpath","abspath",
            "splitunc","curdir","pardir","sep","pathsep","defpath","altsep",
-           "extsep","devnull","realpath","supports_unicode_filenames","relpath"]
+           "extsep","devnull","realpath","supports_unicode_filenames","relpath",
+           "samefile",]
 
 # strings representing various path-related bits and pieces
 # These are primarily for export; internally, they are hardcoded.
@@ -78,6 +79,9 @@
     """Normalize case of pathname.
 
     Makes all characters lowercase and all slashes into backslashes."""
+    if not isinstance(s, (bytes, str)):
+        raise TypeError("normcase() argument must be str or bytes, "
+                        "not '{}'".format(s.__class__.__name__))
     return s.replace(_get_altsep(s), _get_sep(s)).lower()
 
 
@@ -306,16 +310,28 @@
     return split(p)[0]
 
 # Is a path a symbolic link?
-# This will always return false on systems where posix.lstat doesn't exist.
+# This will always return false on systems where os.lstat doesn't exist.
 
 def islink(path):
-    """Test for symbolic link.
-    On WindowsNT/95 and OS/2 always returns false
+    """Test whether a path is a symbolic link.
+    This will always return false for Windows prior to 6.0
+    and for OS/2.
     """
-    return False
-
-# alias exists to lexists
-lexists = exists
+    try:
+        st = os.lstat(path)
+    except (os.error, AttributeError):
+        return False
+    return stat.S_ISLNK(st.st_mode)
+
+# Being true for dangling symbolic links is also useful.
+
+def lexists(path):
+    """Test whether a path exists.  Returns True for broken symbolic links"""
+    try:
+        st = os.lstat(path)
+    except (os.error, WindowsError):
+        return False
+    return True
 
 # Is a path a mount point?  Either a root (with or without drive letter)
 # or an UNC path with at most a / or \ after the mount point.
@@ -609,3 +625,17 @@
     if not rel_list:
         return _get_dot(path)
     return join(*rel_list)
+
+
+# determine if two files are in fact the same file
+def samefile(f1, f2):
+    "Test whether two pathnames reference the same actual file"
+    try:
+        from nt import _getfinalpathname
+        return _getfinalpathname(f1) == _getfinalpathname(f2)
+    except (NotImplementedError, ImportError):
+        # On Windows XP and earlier, two files are the same if their
+        #  absolute pathnames are the same.
+        # Also, on other operating systems, fake this method with a
+        #  Windows-XP approximation.
+        return abspath(f1) == abspath(f2)

Modified: python/branches/py3k-cdecimal/Lib/os2emxpath.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/os2emxpath.py	(original)
+++ python/branches/py3k-cdecimal/Lib/os2emxpath.py	Sat Jul 17 12:36:09 2010
@@ -36,6 +36,9 @@
     """Normalize case of pathname.
 
     Makes all characters lowercase and all altseps into seps."""
+    if not isinstance(s, (bytes, str)):
+        raise TypeError("normcase() argument must be str or bytes, "
+                        "not '{}'".format(s.__class__.__name__))
     return s.replace('\\', '/').lower()
 
 

Modified: python/branches/py3k-cdecimal/Lib/pdb.doc
==============================================================================
--- python/branches/py3k-cdecimal/Lib/pdb.doc	(original)
+++ python/branches/py3k-cdecimal/Lib/pdb.doc	Sat Jul 17 12:36:09 2010
@@ -68,13 +68,13 @@
         An arrow indicates the "current frame", which determines the
         context of most commands.
 
-d(own)
-        Move the current frame one level down in the stack trace
-        (to a newer frame).
-
-u(p)
-        Move the current frame one level up in the stack trace
-        (to an older frame).
+d(own) [ count ]
+        Move the current frame count (default one) levels down in the
+        stack trace (to a newer frame).
+
+u(p) [ count ]
+        Move the current frame count (default one) levels up in the
+        stack trace (to an older frame).
 
 b(reak) [ ([filename:]lineno | function) [, condition] ]
         With a filename:line number argument, set a break there.  If
@@ -138,8 +138,8 @@
 run [args...]
         Restart the debugged python program. If a string is supplied it is
         splitted with "shlex", and the result is used as the new sys.argv.
-	History, breakpoints, actions and debugger options are preserved.
-	"restart" is an alias for "run".
+        History, breakpoints, actions and debugger options are preserved.
+        "restart" is an alias for "run".
 
 c(ont(inue))
         Continue execution, only stop when a breakpoint is encountered.

Modified: python/branches/py3k-cdecimal/Lib/pdb.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/pdb.py	(original)
+++ python/branches/py3k-cdecimal/Lib/pdb.py	Sat Jul 17 12:36:09 2010
@@ -618,26 +618,44 @@
     do_w = do_where
     do_bt = do_where
 
+    def _select_frame(self, number):
+        assert 0 <= number < len(self.stack)
+        self.curindex = number
+        self.curframe = self.stack[self.curindex][0]
+        self.curframe_locals = self.curframe.f_locals
+        self.print_stack_entry(self.stack[self.curindex])
+        self.lineno = None
+
     def do_up(self, arg):
         if self.curindex == 0:
             print('*** Oldest frame', file=self.stdout)
+            return
+        try:
+            count = int(arg or 1)
+        except ValueError:
+            print('*** Invalid frame count (%s)' % arg, file=self.stdout)
+            return
+        if count < 0:
+            newframe = 0
         else:
-            self.curindex = self.curindex - 1
-            self.curframe = self.stack[self.curindex][0]
-            self.curframe_locals = self.curframe.f_locals
-            self.print_stack_entry(self.stack[self.curindex])
-            self.lineno = None
+            newframe = max(0, self.curindex - count)
+        self._select_frame(newframe)
     do_u = do_up
 
     def do_down(self, arg):
         if self.curindex + 1 == len(self.stack):
             print('*** Newest frame', file=self.stdout)
+            return
+        try:
+            count = int(arg or 1)
+        except ValueError:
+            print('*** Invalid frame count (%s)' % arg, file=self.stdout)
+            return
+        if count < 0:
+            newframe = len(self.stack) - 1
         else:
-            self.curindex = self.curindex + 1
-            self.curframe = self.stack[self.curindex][0]
-            self.curframe_locals = self.curframe.f_locals
-            self.print_stack_entry(self.stack[self.curindex])
-            self.lineno = None
+            newframe = min(len(self.stack) - 1, self.curindex + count)
+        self._select_frame(newframe)
     do_d = do_down
 
     def do_until(self, arg):

Modified: python/branches/py3k-cdecimal/Lib/pickletools.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/pickletools.py	(original)
+++ python/branches/py3k-cdecimal/Lib/pickletools.py	Sat Jul 17 12:36:09 2010
@@ -1250,10 +1250,11 @@
       stack_before=[anyobject],
       stack_after=[pytuple],
       proto=2,
-      doc="""One-tuple.
+      doc="""Build a one-tuple out of the topmost item on the stack.
 
       This code pops one value off the stack and pushes a tuple of
-      length 1 whose one item is that value back onto it.  IOW:
+      length 1 whose one item is that value back onto it.  In other
+      words:
 
           stack[-1] = tuple(stack[-1:])
       """),
@@ -1264,10 +1265,11 @@
       stack_before=[anyobject, anyobject],
       stack_after=[pytuple],
       proto=2,
-      doc="""One-tuple.
+      doc="""Build a two-tuple out of the top two items on the stack.
 
-      This code pops two values off the stack and pushes a tuple
-      of length 2 whose items are those values back onto it.  IOW:
+      This code pops two values off the stack and pushes a tuple of
+      length 2 whose items are those values back onto it.  In other
+      words:
 
           stack[-2:] = [tuple(stack[-2:])]
       """),
@@ -1278,10 +1280,11 @@
       stack_before=[anyobject, anyobject, anyobject],
       stack_after=[pytuple],
       proto=2,
-      doc="""One-tuple.
+      doc="""Build a three-tuple out of the top three items on the stack.
 
-      This code pops three values off the stack and pushes a tuple
-      of length 3 whose items are those values back onto it.  IOW:
+      This code pops three values off the stack and pushes a tuple of
+      length 3 whose items are those values back onto it.  In other
+      words:
 
           stack[-3:] = [tuple(stack[-3:])]
       """),
@@ -2330,4 +2333,43 @@
     return doctest.testmod()
 
 if __name__ == "__main__":
-    _test()
+    import sys, argparse
+    parser = argparse.ArgumentParser(
+        description='disassemble one or more pickle files')
+    parser.add_argument(
+        'pickle_file', type=argparse.FileType('br'),
+        nargs='*', help='the pickle file')
+    parser.add_argument(
+        '-o', '--output', default=sys.stdout, type=argparse.FileType('w'),
+        help='the file where the output should be written')
+    parser.add_argument(
+        '-m', '--memo', action='store_true',
+        help='preserve memo between disassemblies')
+    parser.add_argument(
+        '-l', '--indentlevel', default=4, type=int,
+        help='the number of blanks by which to indent a new MARK level')
+    parser.add_argument(
+        '-p', '--preamble', default="==> {name} <==",
+        help='if more than one pickle file is specified, print this before'
+        ' each disassembly')
+    parser.add_argument(
+        '-t', '--test', action='store_true',
+        help='run self-test suite')
+    parser.add_argument(
+        '-v', action='store_true',
+        help='run verbosely; only affects self-test run')
+    args = parser.parse_args()
+    if args.test:
+        _test()
+    else:
+        if not args.pickle_file:
+            parser.print_help()
+        elif len(args.pickle_file) == 1:
+            dis(args.pickle_file[0], args.output,
+                indentlevel=args.indentlevel)
+        else:
+            memo = {} if args.memo else None
+            for f in args.pickle_file:
+                preamble = args.preamble.format(name=f.name)
+                args.output.write(preamble + '\n')
+                dis(f, args.output, memo, args.indentlevel)

Modified: python/branches/py3k-cdecimal/Lib/posixpath.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/posixpath.py	(original)
+++ python/branches/py3k-cdecimal/Lib/posixpath.py	Sat Jul 17 12:36:09 2010
@@ -49,6 +49,9 @@
 def normcase(s):
     """Normalize case of pathname.  Has no effect under Posix"""
     # TODO: on Mac OS X, this should really return s.lower().
+    if not isinstance(s, (bytes, str)):
+        raise TypeError("normcase() argument must be str or bytes, "
+                        "not '{}'".format(s.__class__.__name__))
     return s
 
 

Modified: python/branches/py3k-cdecimal/Lib/pydoc.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/pydoc.py	(original)
+++ python/branches/py3k-cdecimal/Lib/pydoc.py	Sat Jul 17 12:36:09 2010
@@ -1705,8 +1705,9 @@
             return ''
         return '<pydoc.Helper instance>'
 
-    def __call__(self, request=None):
-        if request is not None:
+    _GoInteractive = object()
+    def __call__(self, request=_GoInteractive):
+        if request is not self._GoInteractive:
             self.help(request)
         else:
             self.intro()

Modified: python/branches/py3k-cdecimal/Lib/site.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/site.py	(original)
+++ python/branches/py3k-cdecimal/Lib/site.py	Sat Jul 17 12:36:09 2010
@@ -273,12 +273,12 @@
     environment, and will return a list of full paths.
     """
     sitepackages = []
-    seen = []
+    seen = set()
 
     for prefix in PREFIXES:
         if not prefix or prefix in seen:
             continue
-        seen.append(prefix)
+        seen.add(prefix)
 
         if sys.platform in ('os2emx', 'riscos'):
             sitepackages.append(os.path.join(prefix, "Lib", "site-packages"))

Modified: python/branches/py3k-cdecimal/Lib/smtpd.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/smtpd.py	(original)
+++ python/branches/py3k-cdecimal/Lib/smtpd.py	Sat Jul 17 12:36:09 2010
@@ -274,14 +274,19 @@
         self._localaddr = localaddr
         self._remoteaddr = remoteaddr
         asyncore.dispatcher.__init__(self)
-        self.create_socket(socket.AF_INET, socket.SOCK_STREAM)
-        # try to re-use a server port if possible
-        self.set_reuse_addr()
-        self.bind(localaddr)
-        self.listen(5)
-        print('%s started at %s\n\tLocal addr: %s\n\tRemote addr:%s' % (
-            self.__class__.__name__, time.ctime(time.time()),
-            localaddr, remoteaddr), file=DEBUGSTREAM)
+        try:
+            self.create_socket(socket.AF_INET, socket.SOCK_STREAM)
+            # try to re-use a server port if possible
+            self.set_reuse_addr()
+            self.bind(localaddr)
+            self.listen(5)
+        except:
+            self.close()
+            raise
+        else:
+            print('%s started at %s\n\tLocal addr: %s\n\tRemote addr:%s' % (
+                self.__class__.__name__, time.ctime(time.time()),
+                localaddr, remoteaddr), file=DEBUGSTREAM)
 
     def handle_accept(self):
         conn, addr = self.accept()

Modified: python/branches/py3k-cdecimal/Lib/sndhdr.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/sndhdr.py	(original)
+++ python/branches/py3k-cdecimal/Lib/sndhdr.py	Sat Jul 17 12:36:09 2010
@@ -57,12 +57,12 @@
 
 def test_aifc(h, f):
     import aifc
-    if h.startswith(b'FORM'):
+    if not h.startswith(b'FORM'):
         return None
     if h[8:12] == b'AIFC':
         fmt = 'aifc'
     elif h[8:12] == b'AIFF':
-        fmt = b'aiff'
+        fmt = 'aiff'
     else:
         return None
     f.seek(0)
@@ -123,7 +123,7 @@
 
 
 def test_voc(h, f):
-    if h.startswith(b'Creative Voice File\032'):
+    if not h.startswith(b'Creative Voice File\032'):
         return None
     sbseek = get_short_le(h[20:22])
     rate = 0
@@ -150,7 +150,7 @@
 
 
 def test_8svx(h, f):
-    if h.startswith(b'FORM') or h[8:12] != b'8SVX':
+    if not h.startswith(b'FORM') or h[8:12] != b'8SVX':
         return None
     # Should decode it to get #channels -- assume always 1
     return '8svx', 0, 1, 0, 8

Modified: python/branches/py3k-cdecimal/Lib/socket.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/socket.py	(original)
+++ python/branches/py3k-cdecimal/Lib/socket.py	Sat Jul 17 12:36:09 2010
@@ -252,10 +252,6 @@
         self._sock._decref_socketios()
         self._sock = None
 
-    def __del__(self):
-        if not self.closed:
-            self._sock._decref_socketios()
-
 
 def getfqdn(name=''):
     """Get fully qualified domain name from name.

Modified: python/branches/py3k-cdecimal/Lib/subprocess.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/subprocess.py	(original)
+++ python/branches/py3k-cdecimal/Lib/subprocess.py	Sat Jul 17 12:36:09 2010
@@ -500,8 +500,8 @@
 
     2) A string surrounded by double quotation marks is
        interpreted as a single argument, regardless of white space
-       or pipe characters contained within.  A quoted string can be
-       embedded in an argument.
+       contained within.  A quoted string can be embedded in an
+       argument.
 
     3) A double quotation mark preceded by a backslash is
        interpreted as a literal double quotation mark.
@@ -529,7 +529,7 @@
         if result:
             result.append(' ')
 
-        needquote = (" " in arg) or ("\t" in arg) or ("|" in arg) or not arg
+        needquote = (" " in arg) or ("\t" in arg) or not arg
         if needquote:
             result.append('"')
 

Modified: python/branches/py3k-cdecimal/Lib/sysconfig.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/sysconfig.py	(original)
+++ python/branches/py3k-cdecimal/Lib/sysconfig.py	Sat Jul 17 12:36:09 2010
@@ -123,8 +123,8 @@
 
 if _PYTHON_BUILD:
     for scheme in ('posix_prefix', 'posix_home'):
-        _INSTALL_SCHEMES[scheme]['include'] = '{projectbase}/Include'
-        _INSTALL_SCHEMES[scheme]['platinclude'] = '{srcdir}'
+        _INSTALL_SCHEMES[scheme]['include'] = '{srcdir}/Include'
+        _INSTALL_SCHEMES[scheme]['platinclude'] = '{projectbase}/.'
 
 def _subst_vars(s, local_vars):
     try:
@@ -259,6 +259,13 @@
                 # bogus variable reference; just drop it since we can't deal
                 variables.remove(name)
 
+    # Add in CFLAGS, LDFLAGS, and CPPFLAGS, which are named with a
+    # prefix in the Makefile.
+    for var in ('CFLAGS', 'LDFLAGS', 'CPPFLAGS'):
+        makefile_value = done.get('PY_' + var)
+        if makefile_value is not None:
+            done[var] = makefile_value
+
     # save the results in the global dictionary
     vars.update(done)
     return vars
@@ -432,6 +439,8 @@
 
         if 'srcdir' not in _CONFIG_VARS:
             _CONFIG_VARS['srcdir'] = _PROJECT_BASE
+        else:
+            _CONFIG_VARS['srcdir'] = realpath(_CONFIG_VARS['srcdir'])
 
 
         # Convert srcdir into an absolute path if it appears necessary.
@@ -651,8 +660,7 @@
                 cflags = get_config_vars().get('CFLAGS')
 
                 archs = re.findall('-arch\s+(\S+)', cflags)
-                archs.sort()
-                archs = tuple(archs)
+                archs = tuple(sorted(set(archs)))
 
                 if len(archs) == 1:
                     machine = archs[0]

Modified: python/branches/py3k-cdecimal/Lib/tarfile.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/tarfile.py	(original)
+++ python/branches/py3k-cdecimal/Lib/tarfile.py	Sat Jul 17 12:36:09 2010
@@ -55,6 +55,15 @@
 except ImportError:
     grp = pwd = None
 
+# os.symlink on Windows prior to 6.0 raises NotImplementedError
+symlink_exception = (AttributeError, NotImplementedError)
+try:
+    # WindowsError (1314) will be raised if the caller does not hold the
+    # SeCreateSymbolicLinkPrivilege privilege
+    symlink_exception += (WindowsError,)
+except NameError:
+    pass
+
 # from tarfile import *
 __all__ = ["TarFile", "TarInfo", "is_tarfile", "TarError"]
 
@@ -2273,7 +2282,7 @@
           (platform limitation), we try to make a copy of the referenced file
           instead of a link.
         """
-        if hasattr(os, "symlink") and hasattr(os, "link"):
+        try:
             # For systems that support symbolic and hard links.
             if tarinfo.issym():
                 os.symlink(tarinfo.linkname, targetpath)
@@ -2282,10 +2291,17 @@
                 if os.path.exists(tarinfo._link_target):
                     os.link(tarinfo._link_target, targetpath)
                 else:
-                    self._extract_member(self._find_link_target(tarinfo), targetpath)
+                    self._extract_mem
+        except symlink_exception:
+            if tarinfo.issym():
+                linkpath = os.path.join(os.path.dirname(tarinfo.name),
+                                        tarinfo.linkname)
+            else:
+                linkpath = tarinfo.linkname
         else:
             try:
-                self._extract_member(self._find_link_target(tarinfo), targetpath)
+                self._extract_member(self._find_link_target(tarinfo),
+                                     targetpath)
             except KeyError:
                 raise ExtractError("unable to resolve link inside archive")
 

Modified: python/branches/py3k-cdecimal/Lib/test/decimaltestdata/extra.decTest
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/decimaltestdata/extra.decTest	(original)
+++ python/branches/py3k-cdecimal/Lib/test/decimaltestdata/extra.decTest	Sat Jul 17 12:36:09 2010
@@ -213,7 +213,20 @@
 extr1659 shift 1234567 4 -> 0
 extr1660 shift 1234567 5 -> NaN Invalid_operation
 
+-- Cases where the power function was impossibly slow to determine that the
+-- result is inexact.  Thanks Stefan Krah for identifying this problem.
+precision: 16
+maxExponent: 999999999
+minExponent: -999999999
+extr1700 power 10 1e-999999999 -> 1.000000000000000 Inexact Rounded
+extr1701 power 100.0 -557.71e-742888888 -> 1.000000000000000 Inexact Rounded
+extr1702 power 10 1e-100 -> 1.000000000000000 Inexact Rounded
 
+-- A couple of interesting exact cases for power.  Note that the specification
+-- requires these to be reported as Inexact.
+extr1710 power 1e375 56e-3 -> 1.000000000000000E+21 Inexact Rounded
+extr1711 power 10000 0.75 -> 1000.000000000000 Inexact Rounded
+extr1712 power 1e-24 0.875 -> 1.000000000000000E-21 Inexact Rounded
 
 -- Tests for the is_* boolean operations
 precision: 9

Modified: python/branches/py3k-cdecimal/Lib/test/math_testcases.txt
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/math_testcases.txt	(original)
+++ python/branches/py3k-cdecimal/Lib/test/math_testcases.txt	Sat Jul 17 12:36:09 2010
@@ -371,6 +371,78 @@
 
 
 -----------------------------------------------------------
+-- log1p: log(1 + x), without precision loss for small x --
+-----------------------------------------------------------
+
+-- special values
+log1p0000 log1p 0.0 -> 0.0
+log1p0001 log1p -0.0 -> -0.0
+log1p0002 log1p inf -> inf
+log1p0003 log1p -inf -> nan             invalid
+log1p0004 log1p nan -> nan
+
+-- singularity at -1.0
+log1p0010 log1p -1.0 -> -inf            divide-by-zero
+log1p0011 log1p -0.9999999999999999 -> -36.736800569677101
+
+-- finite values < 1.0 are invalid
+log1p0020 log1p -1.0000000000000002 -> nan invalid
+log1p0021 log1p -1.1 -> nan invalid
+log1p0022 log1p -2.0 -> nan invalid
+log1p0023 log1p -1e300 -> nan invalid
+
+-- tiny x: log1p(x) ~ x
+log1p0110 log1p 5e-324 -> 5e-324
+log1p0111 log1p 1e-320 -> 1e-320
+log1p0112 log1p 1e-300 -> 1e-300
+log1p0113 log1p 1e-150 -> 1e-150
+log1p0114 log1p 1e-20 -> 1e-20
+
+log1p0120 log1p -5e-324 -> -5e-324
+log1p0121 log1p -1e-320 -> -1e-320
+log1p0122 log1p -1e-300 -> -1e-300
+log1p0123 log1p -1e-150 -> -1e-150
+log1p0124 log1p -1e-20 -> -1e-20
+
+-- some (mostly) random small and moderate-sized values
+log1p0200 log1p -0.89156889782277482 -> -2.2216403106762863
+log1p0201 log1p -0.23858496047770464 -> -0.27257668276980057
+log1p0202 log1p -0.011641726191307515 -> -0.011710021654495657
+log1p0203 log1p -0.0090126398571693817 -> -0.0090534993825007650
+log1p0204 log1p -0.00023442805985712781 -> -0.00023445554240995693
+log1p0205 log1p -1.5672870980936349e-5 -> -1.5672993801662046e-5
+log1p0206 log1p -7.9650013274825295e-6 -> -7.9650330482740401e-6
+log1p0207 log1p -2.5202948343227410e-7 -> -2.5202951519170971e-7
+log1p0208 log1p -8.2446372820745855e-11 -> -8.2446372824144559e-11
+log1p0209 log1p -8.1663670046490789e-12 -> -8.1663670046824230e-12
+log1p0210 log1p 7.0351735084656292e-18 -> 7.0351735084656292e-18
+log1p0211 log1p 5.2732161907375226e-12 -> 5.2732161907236188e-12
+log1p0212 log1p 1.0000000000000000e-10 -> 9.9999999995000007e-11
+log1p0213 log1p 2.1401273266000197e-9 -> 2.1401273243099470e-9
+log1p0214 log1p 1.2668914653979560e-8 -> 1.2668914573728861e-8
+log1p0215 log1p 1.6250007816299069e-6 -> 1.6249994613175672e-6
+log1p0216 log1p 8.3740495645839399e-6 -> 8.3740145024266269e-6
+log1p0217 log1p 3.0000000000000001e-5 -> 2.9999550008999799e-5
+log1p0218 log1p 0.0070000000000000001 -> 0.0069756137364252423
+log1p0219 log1p 0.013026235315053002 -> 0.012942123564008787
+log1p0220 log1p 0.013497160797236184 -> 0.013406885521915038
+log1p0221 log1p 0.027625599078135284 -> 0.027250897463483054
+log1p0222 log1p 0.14179687245544870 -> 0.13260322540908789
+
+-- large values
+log1p0300 log1p 1.7976931348623157e+308 -> 709.78271289338397
+log1p0301 log1p 1.0000000000000001e+300 -> 690.77552789821368
+log1p0302 log1p 1.0000000000000001e+70 -> 161.18095650958321
+log1p0303 log1p 10000000000.000000 -> 23.025850930040455
+
+-- other values transferred from testLog1p in test_math
+log1p0400 log1p -0.63212055882855767 -> -1.0000000000000000
+log1p0401 log1p 1.7182818284590451 -> 1.0000000000000000
+log1p0402 log1p 1.0000000000000000 -> 0.69314718055994529
+log1p0403 log1p 1.2379400392853803e+27 -> 62.383246250395075
+
+
+-----------------------------------------------------------
 -- expm1: exp(x) - 1, without precision loss for small x --
 -----------------------------------------------------------
 

Modified: python/branches/py3k-cdecimal/Lib/test/regrtest.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/regrtest.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/regrtest.py	Sat Jul 17 12:36:09 2010
@@ -152,6 +152,7 @@
 option '-uall,-gui'.
 """
 
+import builtins
 import getopt
 import json
 import os
@@ -859,15 +860,9 @@
         sys.path_hooks[:] = saved_hooks[2]
 
     def get___import__(self):
-        if isinstance(__builtins__, dict):
-            return __builtins__['__import__']
-        else:
-            return __builtins__.__import__
+        return builtins.__import__
     def restore___import__(self, import_):
-        if isinstance(__builtins__, dict):
-            __builtins__['__import__'] = import_
-        else:
-            __builtins__.__import__ = import_
+        builtins.__import__ = import_
 
     def get_warnings_filters(self):
         return id(warnings.filters), warnings.filters, warnings.filters[:]

Modified: python/branches/py3k-cdecimal/Lib/test/support.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/support.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/support.py	Sat Jul 17 12:36:09 2010
@@ -17,6 +17,7 @@
 import importlib
 import collections
 import re
+import subprocess
 import imp
 import time
 try:
@@ -38,8 +39,7 @@
     "set_memlimit", "bigmemtest", "bigaddrspacetest", "BasicTestRunner",
     "run_unittest", "run_doctest", "threading_setup", "threading_cleanup",
     "reap_children", "cpython_only", "check_impl_detail", "get_attribute",
-    "swap_item", "swap_attr",
-    ]
+    "swap_item", "swap_attr", "can_symlink", "skip_unless_symlink"]
 
 
 class Error(Exception):
@@ -109,7 +109,7 @@
         orig_modules[name] = sys.modules[name]
     except KeyError:
         saved = False
-    sys.modules[name] = 0
+    sys.modules[name] = None
     return saved
 
 
@@ -1169,6 +1169,27 @@
             except:
                 break
 
+try:
+    from .symlink_support import enable_symlink_privilege
+except:
+    enable_symlink_privilege = lambda: True
+
+def can_symlink():
+    """It's no longer sufficient to test for the presence of symlink in the
+    os module - on Windows XP and earlier, os.symlink exists but a
+    NotImplementedError is thrown.
+    """
+    has_symlink = hasattr(os, 'symlink')
+    is_old_windows = sys.platform == "win32" and sys.getwindowsversion().major < 6
+    has_privilege = False if is_old_windows else enable_symlink_privilege()
+    return has_symlink and (not is_old_windows) and has_privilege
+
+def skip_unless_symlink(test):
+    """Skip decorator for tests that require functional symlink"""
+    selector = can_symlink()
+    msg = "Requires functional symlink implementation"
+    return [unittest.skip(msg)(test), test][selector]
+
 @contextlib.contextmanager
 def swap_attr(obj, attr, new_val):
     """Temporary swap out an attribute with a new object.

Modified: python/branches/py3k-cdecimal/Lib/test/test_ast.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_ast.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_ast.py	Sat Jul 17 12:36:09 2010
@@ -285,6 +285,8 @@
         self.assertEqual(ast.literal_eval('[1, 2, 3]'), [1, 2, 3])
         self.assertEqual(ast.literal_eval('{"foo": 42}'), {"foo": 42})
         self.assertEqual(ast.literal_eval('(True, False, None)'), (True, False, None))
+        self.assertEqual(ast.literal_eval('{1, 2, 3}'), {1, 2, 3})
+        self.assertEqual(ast.literal_eval('b"hi"'), b"hi")
         self.assertRaises(ValueError, ast.literal_eval, 'foo()')
 
     def test_literal_eval_issue4907(self):

Modified: python/branches/py3k-cdecimal/Lib/test/test_audioop.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_audioop.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_audioop.py	Sat Jul 17 12:36:09 2010
@@ -20,6 +20,12 @@
 
 data = [gendata1(), gendata2(), gendata4()]
 
+INVALID_DATA = [
+    ('abc', 0),
+    ('abc', 2),
+    ('abc', 4),
+]
+
 
 class TestAudioop(unittest.TestCase):
 
@@ -168,6 +174,33 @@
         self.assertRaises(audioop.error,
             audioop.findmax, ''.join(chr(x) for x in range(256)), -2392392)
 
+    def test_issue7673(self):
+        state = None
+        for data, size in INVALID_DATA:
+            size2 = size
+            self.assertRaises(audioop.error, audioop.getsample, data, size, 0)
+            self.assertRaises(audioop.error, audioop.max, data, size)
+            self.assertRaises(audioop.error, audioop.minmax, data, size)
+            self.assertRaises(audioop.error, audioop.avg, data, size)
+            self.assertRaises(audioop.error, audioop.rms, data, size)
+            self.assertRaises(audioop.error, audioop.avgpp, data, size)
+            self.assertRaises(audioop.error, audioop.maxpp, data, size)
+            self.assertRaises(audioop.error, audioop.cross, data, size)
+            self.assertRaises(audioop.error, audioop.mul, data, size, 1.0)
+            self.assertRaises(audioop.error, audioop.tomono, data, size, 0.5, 0.5)
+            self.assertRaises(audioop.error, audioop.tostereo, data, size, 0.5, 0.5)
+            self.assertRaises(audioop.error, audioop.add, data, data, size)
+            self.assertRaises(audioop.error, audioop.bias, data, size, 0)
+            self.assertRaises(audioop.error, audioop.reverse, data, size)
+            self.assertRaises(audioop.error, audioop.lin2lin, data, size, size2)
+            self.assertRaises(audioop.error, audioop.ratecv, data, size, 1, 1, 1, state)
+            self.assertRaises(audioop.error, audioop.lin2ulaw, data, size)
+            self.assertRaises(audioop.error, audioop.ulaw2lin, data, size)
+            self.assertRaises(audioop.error, audioop.lin2alaw, data, size)
+            self.assertRaises(audioop.error, audioop.alaw2lin, data, size)
+            self.assertRaises(audioop.error, audioop.lin2adpcm, data, size, state)
+            self.assertRaises(audioop.error, audioop.adpcm2lin, data, size, state)
+
 def test_main():
     run_unittest(TestAudioop)
 

Modified: python/branches/py3k-cdecimal/Lib/test/test_calendar.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_calendar.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_calendar.py	Sat Jul 17 12:36:09 2010
@@ -2,7 +2,7 @@
 import unittest
 
 from test import support
-
+import time
 
 result_2004_text = """
                                   2004
@@ -381,13 +381,21 @@
         # A 31-day december starting on friday (2+7+7+7+7+1 days)
         self.check_weeks(1995, 12, (2, 7, 7, 7, 7, 1))
 
+class TimegmTestCase(unittest.TestCase):
+    TIMESTAMPS = [0, 10, 100, 1000, 10000, 100000, 1000000,
+                  1234567890, 1262304000, 1275785153,]
+    def test_timegm(self):
+        for secs in self.TIMESTAMPS:
+            tuple = time.gmtime(secs)
+            self.assertEqual(secs, calendar.timegm(tuple))
 
 def test_main():
     support.run_unittest(
         OutputTestCase,
         CalendarTestCase,
         MondayTestCase,
-        SundayTestCase
+        SundayTestCase,
+        TimegmTestCase,
     )
 
 

Modified: python/branches/py3k-cdecimal/Lib/test/test_cmath.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_cmath.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_cmath.py	Sat Jul 17 12:36:09 2010
@@ -442,6 +442,15 @@
         self.assertCEqual(rect(1, pi/2), (0, 1.))
         self.assertCEqual(rect(1, -pi/2), (0, -1.))
 
+    def test_isfinite(self):
+        real_vals = [float('-inf'), -2.3, -0.0,
+                     0.0, 2.3, float('inf'), float('nan')]
+        for x in real_vals:
+            for y in real_vals:
+                z = complex(x, y)
+                self.assertEqual(cmath.isfinite(z),
+                                  math.isfinite(x) and math.isfinite(y))
+
     def test_isnan(self):
         self.assertFalse(cmath.isnan(1))
         self.assertFalse(cmath.isnan(1j))

Modified: python/branches/py3k-cdecimal/Lib/test/test_codeccallbacks.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_codeccallbacks.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_codeccallbacks.py	Sat Jul 17 12:36:09 2010
@@ -153,28 +153,30 @@
             sout += bytes("\\U%08x" % sys.maxunicode, "ascii")
         self.assertEqual(sin.encode("iso-8859-15", "backslashreplace"), sout)
 
-    def test_decoderelaxedutf8(self):
-        # This is the test for a decoding callback handler,
-        # that relaxes the UTF-8 minimal encoding restriction.
-        # A null byte that is encoded as "\xc0\x80" will be
-        # decoded as a null byte. All other illegal sequences
-        # will be handled strictly.
+    def test_decoding_callbacks(self):
+        # This is a test for a decoding callback handler
+        # that allows the decoding of the invalid sequence
+        # "\xc0\x80" and returns "\x00" instead of raising an error.
+        # All other illegal sequences will be handled strictly.
         def relaxedutf8(exc):
             if not isinstance(exc, UnicodeDecodeError):
                 raise TypeError("don't know how to handle %r" % exc)
-            if exc.object[exc.start:exc.end].startswith(b"\xc0\x80"):
+            if exc.object[exc.start:exc.start+2] == b"\xc0\x80":
                 return ("\x00", exc.start+2) # retry after two bytes
             else:
                 raise exc
 
-        codecs.register_error(
-            "test.relaxedutf8", relaxedutf8)
+        codecs.register_error("test.relaxedutf8", relaxedutf8)
 
+        # all the "\xc0\x80" will be decoded to "\x00"
         sin = b"a\x00b\xc0\x80c\xc3\xbc\xc0\x80\xc0\x80"
         sout = "a\x00b\x00c\xfc\x00\x00"
         self.assertEqual(sin.decode("utf-8", "test.relaxedutf8"), sout)
+
+        # "\xc0\x81" is not valid and a UnicodeDecodeError will be raised
         sin = b"\xc0\x80\xc0\x81"
-        self.assertRaises(UnicodeError, sin.decode, "utf-8", "test.relaxedutf8")
+        self.assertRaises(UnicodeDecodeError, sin.decode,
+                          "utf-8", "test.relaxedutf8")
 
     def test_charmapencode(self):
         # For charmap encodings the replacement string will be

Modified: python/branches/py3k-cdecimal/Lib/test/test_codecs.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_codecs.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_codecs.py	Sat Jul 17 12:36:09 2010
@@ -1263,6 +1263,7 @@
     "cp424",
     "cp437",
     "cp500",
+    "cp720",
     "cp737",
     "cp775",
     "cp850",
@@ -1270,6 +1271,7 @@
     "cp855",
     "cp856",
     "cp857",
+    "cp858",
     "cp860",
     "cp861",
     "cp862",
@@ -1358,11 +1360,6 @@
     "idna",
 ]
 
-# The following encodings only support "strict" mode
-only_strict_mode = [
-    "idna",
-]
-
 class BasicUnicodeTest(unittest.TestCase, MixInCheckStateHandling):
     def test_basics(self):
         s = "abc123" # all codecs should be able to encode these
@@ -1437,7 +1434,7 @@
                     result = "".join(codecs.iterdecode(codecs.iterencode("", encoding), encoding))
                     self.assertEqual(result, "")
 
-                if encoding not in only_strict_mode:
+                if encoding not in ("idna", "mbcs"):
                     # check incremental decoder/encoder with errors argument
                     try:
                         encoder = codecs.getincrementalencoder(encoding)("ignore")

Modified: python/branches/py3k-cdecimal/Lib/test/test_collections.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_collections.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_collections.py	Sat Jul 17 12:36:09 2010
@@ -758,6 +758,19 @@
         od.update([('a', 1), ('b', 2), ('c', 9), ('d', 4)], c=3, e=5)
         self.assertEqual(list(od.items()), pairs)                                   # mixed input
 
+        # Issue 9137: Named argument called 'other' or 'self'
+        # shouldn't be treated specially.
+        od = OrderedDict()
+        od.update(self=23)
+        self.assertEqual(list(od.items()), [('self', 23)])
+        od = OrderedDict()
+        od.update(other={})
+        self.assertEqual(list(od.items()), [('other', {})])
+        od = OrderedDict()
+        od.update(red=5, blue=6, other=7, self=8)
+        self.assertEqual(sorted(list(od.items())),
+                         [('blue', 6), ('other', 7), ('red', 5), ('self', 8)])
+
         # Make sure that direct calls to update do not clear previous contents
         # add that updates items are not moved to the end
         d = OrderedDict([('a', 1), ('b', 2), ('c', 3), ('d', 44), ('e', 55)])

Modified: python/branches/py3k-cdecimal/Lib/test/test_complex.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_complex.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_complex.py	Sat Jul 17 12:36:09 2010
@@ -437,6 +437,23 @@
 
     @unittest.skipUnless(float.__getformat__("double").startswith("IEEE"),
                          "test requires IEEE 754 doubles")
+    def test_negated_imaginary_literal(self):
+        z0 = -0j
+        z1 = -7j
+        z2 = -1e1000j
+        # Note: In versions of Python < 3.2, a negated imaginary literal
+        # accidentally ended up with real part 0.0 instead of -0.0, thanks to a
+        # modification during CST -> AST translation (see issue #9011).  That's
+        # fixed in Python 3.2.
+        self.assertFloatsAreIdentical(z0.real, -0.0)
+        self.assertFloatsAreIdentical(z0.imag, -0.0)
+        self.assertFloatsAreIdentical(z1.real, -0.0)
+        self.assertFloatsAreIdentical(z1.imag, -7.0)
+        self.assertFloatsAreIdentical(z2.real, -0.0)
+        self.assertFloatsAreIdentical(z2.imag, -INF)
+
+    @unittest.skipUnless(float.__getformat__("double").startswith("IEEE"),
+                         "test requires IEEE 754 doubles")
     def test_overflow(self):
         self.assertEqual(complex("1e500"), complex(INF, 0.0))
         self.assertEqual(complex("-1e500j"), complex(0.0, -INF))

Modified: python/branches/py3k-cdecimal/Lib/test/test_contextlib.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_contextlib.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_contextlib.py	Sat Jul 17 12:36:09 2010
@@ -202,6 +202,169 @@
                 return True
         self.boilerPlate(lock, locked)
 
+
+class mycontext(ContextDecorator):
+    started = False
+    exc = None
+    catch = False
+
+    def __enter__(self):
+        self.started = True
+        return self
+
+    def __exit__(self, *exc):
+        self.exc = exc
+        return self.catch
+
+
+class TestContextDecorator(unittest.TestCase):
+
+    def test_contextdecorator(self):
+        context = mycontext()
+        with context as result:
+            self.assertIs(result, context)
+            self.assertTrue(context.started)
+
+        self.assertEqual(context.exc, (None, None, None))
+
+
+    def test_contextdecorator_with_exception(self):
+        context = mycontext()
+
+        with self.assertRaisesRegexp(NameError, 'foo'):
+            with context:
+                raise NameError('foo')
+        self.assertIsNotNone(context.exc)
+        self.assertIs(context.exc[0], NameError)
+
+        context = mycontext()
+        context.catch = True
+        with context:
+            raise NameError('foo')
+        self.assertIsNotNone(context.exc)
+        self.assertIs(context.exc[0], NameError)
+
+
+    def test_decorator(self):
+        context = mycontext()
+
+        @context
+        def test():
+            self.assertIsNone(context.exc)
+            self.assertTrue(context.started)
+        test()
+        self.assertEqual(context.exc, (None, None, None))
+
+
+    def test_decorator_with_exception(self):
+        context = mycontext()
+
+        @context
+        def test():
+            self.assertIsNone(context.exc)
+            self.assertTrue(context.started)
+            raise NameError('foo')
+
+        with self.assertRaisesRegexp(NameError, 'foo'):
+            test()
+        self.assertIsNotNone(context.exc)
+        self.assertIs(context.exc[0], NameError)
+
+
+    def test_decorating_method(self):
+        context = mycontext()
+
+        class Test(object):
+
+            @context
+            def method(self, a, b, c=None):
+                self.a = a
+                self.b = b
+                self.c = c
+
+        # these tests are for argument passing when used as a decorator
+        test = Test()
+        test.method(1, 2)
+        self.assertEqual(test.a, 1)
+        self.assertEqual(test.b, 2)
+        self.assertEqual(test.c, None)
+
+        test = Test()
+        test.method('a', 'b', 'c')
+        self.assertEqual(test.a, 'a')
+        self.assertEqual(test.b, 'b')
+        self.assertEqual(test.c, 'c')
+
+        test = Test()
+        test.method(a=1, b=2)
+        self.assertEqual(test.a, 1)
+        self.assertEqual(test.b, 2)
+
+
+    def test_typo_enter(self):
+        class mycontext(ContextDecorator):
+            def __unter__(self):
+                pass
+            def __exit__(self, *exc):
+                pass
+
+        with self.assertRaises(AttributeError):
+            with mycontext():
+                pass
+
+
+    def test_typo_exit(self):
+        class mycontext(ContextDecorator):
+            def __enter__(self):
+                pass
+            def __uxit__(self, *exc):
+                pass
+
+        with self.assertRaises(AttributeError):
+            with mycontext():
+                pass
+
+
+    def test_contextdecorator_as_mixin(self):
+        class somecontext(object):
+            started = False
+            exc = None
+
+            def __enter__(self):
+                self.started = True
+                return self
+
+            def __exit__(self, *exc):
+                self.exc = exc
+
+        class mycontext(somecontext, ContextDecorator):
+            pass
+
+        context = mycontext()
+        @context
+        def test():
+            self.assertIsNone(context.exc)
+            self.assertTrue(context.started)
+        test()
+        self.assertEqual(context.exc, (None, None, None))
+
+
+    def test_contextmanager_as_decorator(self):
+        state = []
+        @contextmanager
+        def woohoo(y):
+            state.append(y)
+            yield
+            state.append(999)
+
+        @woohoo(1)
+        def test(x):
+            self.assertEqual(state, [1])
+            state.append(x)
+        test('something')
+        self.assertEqual(state, [1, 'something', 999])
+
+
 # This is needed to make the test actually run under regrtest.py!
 def test_main():
     support.run_unittest(__name__)

Modified: python/branches/py3k-cdecimal/Lib/test/test_datetime.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_datetime.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_datetime.py	Sat Jul 17 12:36:09 2010
@@ -15,10 +15,13 @@
 from datetime import timedelta
 from datetime import tzinfo
 from datetime import time
+from datetime import timezone
 from datetime import date, datetime
+import time as _time
 
-pickle_choices = [(pickle, pickle, proto) for proto in range(3)]
-assert len(pickle_choices) == 3
+pickle_choices = [(pickle, pickle, proto)
+                  for proto in range(pickle.HIGHEST_PROTOCOL + 1)]
+assert len(pickle_choices) == pickle.HIGHEST_PROTOCOL + 1
 
 # An arbitrary collection of objects of non-datetime types, for testing
 # mixed-type comparisons.
@@ -49,6 +52,7 @@
 # tzinfo tests
 
 class FixedOffset(tzinfo):
+
     def __init__(self, offset, name, dstoffset=42):
         if isinstance(offset, int):
             offset = timedelta(minutes=offset)
@@ -67,6 +71,7 @@
         return self.__dstoffset
 
 class PicklableFixedOffset(FixedOffset):
+
     def __init__(self, offset=None, name=None, dstoffset=None):
         FixedOffset.__init__(self, offset, name, dstoffset)
 
@@ -118,18 +123,126 @@
     def test_pickling_subclass(self):
         # Make sure we can pickle/unpickle an instance of a subclass.
         offset = timedelta(minutes=-300)
-        orig = PicklableFixedOffset(offset, 'cookie')
-        self.assertIsInstance(orig, tzinfo)
-        self.assertTrue(type(orig) is PicklableFixedOffset)
-        self.assertEqual(orig.utcoffset(None), offset)
-        self.assertEqual(orig.tzname(None), 'cookie')
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertIsInstance(derived, tzinfo)
-            self.assertTrue(type(derived) is PicklableFixedOffset)
-            self.assertEqual(derived.utcoffset(None), offset)
-            self.assertEqual(derived.tzname(None), 'cookie')
+        for otype, args in [
+            (PicklableFixedOffset, (offset, 'cookie')),
+            (timezone, (offset,)),
+            (timezone, (offset, "EST"))]:
+            orig = otype(*args)
+            oname = orig.tzname(None)
+            self.assertIsInstance(orig, tzinfo)
+            self.assertIs(type(orig), otype)
+            self.assertEqual(orig.utcoffset(None), offset)
+            self.assertEqual(orig.tzname(None), oname)
+            for pickler, unpickler, proto in pickle_choices:
+                green = pickler.dumps(orig, proto)
+                derived = unpickler.loads(green)
+                self.assertIsInstance(derived, tzinfo)
+                self.assertIs(type(derived), otype)
+                self.assertEqual(derived.utcoffset(None), offset)
+                self.assertEqual(derived.tzname(None), oname)
+
+class TestTimeZone(unittest.TestCase):
+
+    def setUp(self):
+        self.ACDT = timezone(timedelta(hours=9.5), 'ACDT')
+        self.EST = timezone(-timedelta(hours=5), 'EST')
+        self.DT = datetime(2010, 1, 1)
+
+    def test_str(self):
+        for tz in [self.ACDT, self.EST, timezone.utc,
+                   timezone.min, timezone.max]:
+            self.assertEqual(str(tz), tz.tzname(None))
+
+    def test_repr(self):
+        import datetime
+        for tz in [self.ACDT, self.EST, timezone.utc,
+                   timezone.min, timezone.max]:
+            # test round-trip
+            tzrep = repr(tz)
+            self.assertEqual(tz, eval(tzrep))
+
+
+    def test_class_members(self):
+        limit = timedelta(hours=23, minutes=59)
+        self.assertEqual(timezone.utc.utcoffset(None), ZERO)
+        self.assertEqual(timezone.min.utcoffset(None), -limit)
+        self.assertEqual(timezone.max.utcoffset(None), limit)
+
+
+    def test_constructor(self):
+        self.assertEqual(timezone.utc, timezone(timedelta(0)))
+        # invalid offsets
+        for invalid in [timedelta(microseconds=1), timedelta(1, 1),
+                        timedelta(seconds=1), timedelta(1), -timedelta(1)]:
+            self.assertRaises(ValueError, timezone, invalid)
+            self.assertRaises(ValueError, timezone, -invalid)
+
+        with self.assertRaises(TypeError): timezone(None)
+        with self.assertRaises(TypeError): timezone(42)
+        with self.assertRaises(TypeError): timezone(ZERO, None)
+        with self.assertRaises(TypeError): timezone(ZERO, 42)
+        with self.assertRaises(TypeError): timezone(ZERO, 'ABC', 'extra')
+
+    def test_inheritance(self):
+        self.assertIsInstance(timezone.utc, tzinfo)
+        self.assertIsInstance(self.EST, tzinfo)
+
+    def test_utcoffset(self):
+        dummy = self.DT
+        for h in [0, 1.5, 12]:
+            offset = h * HOUR
+            self.assertEqual(offset, timezone(offset).utcoffset(dummy))
+            self.assertEqual(-offset, timezone(-offset).utcoffset(dummy))
+
+        with self.assertRaises(TypeError): self.EST.utcoffset('')
+        with self.assertRaises(TypeError): self.EST.utcoffset(5)
+
+
+    def test_dst(self):
+        self.assertEqual(None, timezone.utc.dst(self.DT))
+
+        with self.assertRaises(TypeError): self.EST.dst('')
+        with self.assertRaises(TypeError): self.EST.dst(5)
+
+    def test_tzname(self):
+        self.assertEqual('UTC+00:00', timezone(ZERO).tzname(None))
+        self.assertEqual('UTC-05:00', timezone(-5 * HOUR).tzname(None))
+        self.assertEqual('UTC+09:30', timezone(9.5 * HOUR).tzname(None))
+        self.assertEqual('UTC-00:01', timezone(timedelta(minutes=-1)).tzname(None))
+        self.assertEqual('XYZ', timezone(-5 * HOUR, 'XYZ').tzname(None))
+
+        with self.assertRaises(TypeError): self.EST.tzname('')
+        with self.assertRaises(TypeError): self.EST.tzname(5)
+
+    def test_fromutc(self):
+        with self.assertRaises(ValueError):
+            timezone.utc.fromutc(self.DT)
+        with self.assertRaises(TypeError):
+            timezone.utc.fromutc('not datetime')
+        for tz in [self.EST, self.ACDT, Eastern]:
+            utctime = self.DT.replace(tzinfo=tz)
+            local = tz.fromutc(utctime)
+            self.assertEqual(local - utctime, tz.utcoffset(local))
+            self.assertEqual(local,
+                             self.DT.replace(tzinfo=timezone.utc))
+
+    def test_comparison(self):
+        self.assertNotEqual(timezone(ZERO), timezone(HOUR))
+        self.assertEqual(timezone(HOUR), timezone(HOUR))
+        self.assertEqual(timezone(-5 * HOUR), timezone(-5 * HOUR, 'EST'))
+        with self.assertRaises(TypeError): timezone(ZERO) < timezone(ZERO)
+        self.assertIn(timezone(ZERO), {timezone(ZERO)})
+
+    def test_aware_datetime(self):
+        # test that timezone instances can be used by datetime
+        t = datetime(1, 1, 1)
+        for tz in [timezone.min, timezone.max, timezone.utc]:
+            self.assertEqual(tz.tzname(t),
+                             t.replace(tzinfo=tz).tzname())
+            self.assertEqual(tz.utcoffset(t),
+                             t.replace(tzinfo=tz).utcoffset())
+            self.assertEqual(tz.dst(t),
+                             t.replace(tzinfo=tz).dst())
 
 #############################################################################
 # Base clase for testing a particular aspect of timedelta, time, date and
@@ -206,6 +319,7 @@
         c = td(0, 0, 1000) # One millisecond
         eq(a+b+c, td(7, 60, 1000))
         eq(a-b, td(6, 24*3600 - 60))
+        eq(b.__rsub__(a), td(6, 24*3600 - 60))
         eq(-a, td(-7))
         eq(+a, td(7))
         eq(-b, td(-1, 24*3600 - 60))
@@ -283,6 +397,7 @@
         self.assertRaises(ZeroDivisionError, lambda: a // zero)
         self.assertRaises(ZeroDivisionError, lambda: a / zero)
         self.assertRaises(ZeroDivisionError, lambda: a / 0.0)
+        self.assertRaises(TypeError, lambda: a / '')
 
     @requires_IEEE_754
     def test_disallowed_special(self):
@@ -408,10 +523,20 @@
                    microseconds=999999)),
            "999999999 days, 23:59:59.999999")
 
+    def test_repr(self):
+        name = 'datetime.' + self.theclass.__name__
+        self.assertEqual(repr(self.theclass(1)),
+                         "%s(1)" % name)
+        self.assertEqual(repr(self.theclass(10, 2)),
+                         "%s(10, 2)" % name)
+        self.assertEqual(repr(self.theclass(-10, 2, 400000)),
+                         "%s(-10, 2, 400000)" % name)
+
     def test_roundtrip(self):
         for td in (timedelta(days=999999999, hours=23, minutes=59,
                              seconds=59, microseconds=999999),
                    timedelta(days=-999999999),
+                   timedelta(days=-999999999, seconds=1),
                    timedelta(days=1, seconds=2, microseconds=3)):
 
             # Verify td -> string -> td identity.
@@ -765,6 +890,7 @@
     def test_computations(self):
         a = self.theclass(2002, 1, 31)
         b = self.theclass(1956, 1, 31)
+        c = self.theclass(2001,2,1)
 
         diff = a-b
         self.assertEqual(diff.days, 46*365 + len(range(1956, 2002, 4)))
@@ -790,6 +916,7 @@
         self.assertEqual(a - (a + day), -day)
         self.assertEqual(a - (a - week), week)
         self.assertEqual(a - (a - day), day)
+        self.assertEqual(c - (c - day), day)
 
         # Add/sub ints or floats should be illegal
         for i in 1, 1.0:
@@ -1005,8 +1132,13 @@
             self.assertEqual(b.__format__(fmt), 'B')
 
     def test_resolution_info(self):
-        self.assertIsInstance(self.theclass.min, self.theclass)
-        self.assertIsInstance(self.theclass.max, self.theclass)
+        # XXX: Should min and max respect subclassing?
+        if issubclass(self.theclass, datetime):
+            expected_class = datetime
+        else:
+            expected_class = date
+        self.assertIsInstance(self.theclass.min, expected_class)
+        self.assertIsInstance(self.theclass.max, expected_class)
         self.assertIsInstance(self.theclass.resolution, timedelta)
         self.assertTrue(self.theclass.max > self.theclass.min)
 
@@ -1226,6 +1358,10 @@
         for month_byte in b'9', b'\0', b'\r', b'\xff':
             self.assertRaises(TypeError, self.theclass,
                                          base[:2] + month_byte + base[3:])
+        # Good bytes, but bad tzinfo:
+        self.assertRaises(TypeError, self.theclass,
+                          bytes([1] * len(base)), 'EST')
+
         for ord_byte in range(1, 13):
             # This shouldn't blow up because of the month byte alone.  If
             # the implementation changes to do more-careful checking, it may
@@ -1587,8 +1723,8 @@
     def test_microsecond_rounding(self):
         # Test whether fromtimestamp "rounds up" floats that are less
         # than one microsecond smaller than an integer.
-        self.assertEquals(self.theclass.fromtimestamp(0.9999999),
-                          self.theclass.fromtimestamp(1))
+        self.assertEqual(self.theclass.fromtimestamp(0.9999999),
+                         self.theclass.fromtimestamp(1))
 
     def test_insane_fromtimestamp(self):
         # It's possible that some platform maps time_t to double,
@@ -1616,7 +1752,7 @@
     @unittest.skipIf(sys.platform == "win32", "Windows doesn't accept negative timestamps")
     def test_negative_float_utcfromtimestamp(self):
         d = self.theclass.utcfromtimestamp(-1.05)
-        self.assertEquals(d, self.theclass(1969, 12, 31, 23, 59, 58, 950000))
+        self.assertEqual(d, self.theclass(1969, 12, 31, 23, 59, 58, 950000))
 
     def test_utcnow(self):
         import time
@@ -1637,10 +1773,42 @@
 
         string = '2004-12-01 13:02:47.197'
         format = '%Y-%m-%d %H:%M:%S.%f'
-        result, frac = _strptime._strptime(string, format)
-        expected = self.theclass(*(result[0:6]+(frac,)))
+        expected = _strptime._strptime_datetime(self.theclass, string, format)
         got = self.theclass.strptime(string, format)
         self.assertEqual(expected, got)
+        self.assertIs(type(expected), self.theclass)
+        self.assertIs(type(got), self.theclass)
+
+        strptime = self.theclass.strptime
+        self.assertEqual(strptime("+0002", "%z").utcoffset(), 2 * MINUTE)
+        self.assertEqual(strptime("-0002", "%z").utcoffset(), -2 * MINUTE)
+        # Only local timezone and UTC are supported
+        for tzseconds, tzname in ((0, 'UTC'), (0, 'GMT'),
+                                 (-_time.timezone, _time.tzname[0])):
+            if tzseconds < 0:
+                sign = '-'
+                seconds = -tzseconds
+            else:
+                sign ='+'
+                seconds = tzseconds
+            hours, minutes = divmod(seconds//60, 60)
+            dtstr = "{}{:02d}{:02d} {}".format(sign, hours, minutes, tzname)
+            dt = strptime(dtstr, "%z %Z")
+            self.assertEqual(dt.utcoffset(), timedelta(seconds=tzseconds))
+            self.assertEqual(dt.tzname(), tzname)
+        # Can produce inconsistent datetime
+        dtstr, fmt = "+1234 UTC", "%z %Z"
+        dt = strptime(dtstr, fmt)
+        self.assertEqual(dt.utcoffset(), 12 * HOUR + 34 * MINUTE)
+        self.assertEqual(dt.tzname(), 'UTC')
+        # yet will roundtrip
+        self.assertEqual(dt.strftime(fmt), dtstr)
+
+        # Produce naive datetime if no %z is provided
+        self.assertEqual(strptime("UTC", "%Z").tzinfo, None)
+
+        with self.assertRaises(ValueError): strptime("-2400", "%z")
+        with self.assertRaises(ValueError): strptime("-000", "%z")
 
     def test_more_timetuple(self):
         # This tests fields beyond those tested by the TestDate.test_timetuple.
@@ -1695,6 +1863,8 @@
         self.assertRaises(TypeError, combine, t, d) # args reversed
         self.assertRaises(TypeError, combine, d, t, 1) # too many args
         self.assertRaises(TypeError, combine, "date", "time") # wrong types
+        self.assertRaises(TypeError, combine, d, "time") # wrong type
+        self.assertRaises(TypeError, combine, "date", t) # wrong type
 
     def test_replace(self):
         cls = self.theclass
@@ -1737,6 +1907,8 @@
             def dst(self, dt): return timedelta(0)
         bog = Bogus()
         self.assertRaises(ValueError, dt.astimezone, bog)   # naive
+        self.assertRaises(ValueError,
+                          dt.replace(tzinfo=bog).astimezone, f)
 
         class AlsoBogus(tzinfo):
             def utcoffset(self, dt): return timedelta(0)
@@ -1771,6 +1943,12 @@
         self.assertEqual(dt2.newmeth(-7), dt1.year + dt1.month +
                                           dt1.second - 7)
 
+class TestSubclassDateTime(TestDateTime):
+    theclass = SubclassDatetime
+    # Override tests not designed for subclass
+    def test_roundtrip(self):
+        pass
+
 class SubclassTime(time):
     sub_var = 1
 
@@ -2707,6 +2885,11 @@
         maxdiff = max - min
         self.assertEqual(maxdiff, self.theclass.max - self.theclass.min +
                                   timedelta(minutes=2*1439))
+        # Different tzinfo, but the same offset
+        tza = timezone(HOUR, 'A')
+        tzb = timezone(HOUR, 'B')
+        delta = min.replace(tzinfo=tza) - max.replace(tzinfo=tzb)
+        self.assertEqual(delta, self.theclass.min - self.theclass.max)
 
     def test_tzinfo_now(self):
         meth = self.theclass.now
@@ -2729,20 +2912,21 @@
         # We don't know which time zone we're in, and don't have a tzinfo
         # class to represent it, so seeing whether a tz argument actually
         # does a conversion is tricky.
-        weirdtz = FixedOffset(timedelta(hours=15, minutes=58), "weirdtz", 0)
         utc = FixedOffset(0, "utc", 0)
-        for dummy in range(3):
-            now = datetime.now(weirdtz)
-            self.assertTrue(now.tzinfo is weirdtz)
-            utcnow = datetime.utcnow().replace(tzinfo=utc)
-            now2 = utcnow.astimezone(weirdtz)
-            if abs(now - now2) < timedelta(seconds=30):
-                break
-            # Else the code is broken, or more than 30 seconds passed between
-            # calls; assuming the latter, just try again.
-        else:
-            # Three strikes and we're out.
-            self.fail("utcnow(), now(tz), or astimezone() may be broken")
+        for weirdtz in [FixedOffset(timedelta(hours=15, minutes=58), "weirdtz", 0),
+                        timezone(timedelta(hours=15, minutes=58), "weirdtz"),]:
+            for dummy in range(3):
+                now = datetime.now(weirdtz)
+                self.assertTrue(now.tzinfo is weirdtz)
+                utcnow = datetime.utcnow().replace(tzinfo=utc)
+                now2 = utcnow.astimezone(weirdtz)
+                if abs(now - now2) < timedelta(seconds=30):
+                    break
+                # Else the code is broken, or more than 30 seconds passed between
+                # calls; assuming the latter, just try again.
+            else:
+                # Three strikes and we're out.
+                self.fail("utcnow(), now(tz), or astimezone() may be broken")
 
     def test_tzinfo_fromtimestamp(self):
         import time
@@ -2839,7 +3023,7 @@
 
     def test_utctimetuple(self):
         class DST(tzinfo):
-            def __init__(self, dstvalue):
+            def __init__(self, dstvalue=0):
                 if isinstance(dstvalue, int):
                     dstvalue = timedelta(minutes=dstvalue)
                 self.dstvalue = dstvalue
@@ -2858,8 +3042,6 @@
             def utcoffset(self, dt):
                 return self.uofs
 
-        # Ensure tm_isdst is 0 regardless of what dst() says:  DST is never
-        # in effect for a UTC time.
         for dstvalue in -33, 33, 0, None:
             d = cls(1, 2, 3, 10, 20, 30, 40, tzinfo=UOFS(-53, dstvalue))
             t = d.utctimetuple()
@@ -2872,34 +3054,52 @@
             self.assertEqual(d.weekday(), t.tm_wday)
             self.assertEqual(d.toordinal() - date(1, 1, 1).toordinal() + 1,
                              t.tm_yday)
+            # Ensure tm_isdst is 0 regardless of what dst() says: DST
+            # is never in effect for a UTC time.
             self.assertEqual(0, t.tm_isdst)
 
-        # At the edges, UTC adjustment can normalize into years out-of-range
-        # for a datetime object.  Ensure that a correct timetuple is
-        # created anyway.
+        # For naive datetime, utctimetuple == timetuple except for isdst
+        d = cls(1, 2, 3, 10, 20, 30, 40)
+        t = d.utctimetuple()
+        self.assertEqual(t[:-1], d.timetuple()[:-1])
+        self.assertEqual(0, t.tm_isdst)
+        # Same if utcoffset is None
+        class NOFS(DST):
+            def utcoffset(self, dt):
+                return None
+        d = cls(1, 2, 3, 10, 20, 30, 40, tzinfo=NOFS())
+        t = d.utctimetuple()
+        self.assertEqual(t[:-1], d.timetuple()[:-1])
+        self.assertEqual(0, t.tm_isdst)
+        # Check that bad tzinfo is detected
+        class BOFS(DST):
+            def utcoffset(self, dt):
+                return "EST"
+        d = cls(1, 2, 3, 10, 20, 30, 40, tzinfo=BOFS())
+        self.assertRaises(TypeError, d.utctimetuple)
+
+        # Check that utctimetuple() is the same as
+        # astimezone(utc).timetuple()
+        d = cls(2010, 11, 13, 14, 15, 16, 171819)
+        for tz in [timezone.min, timezone.utc, timezone.max]:
+            dtz = d.replace(tzinfo=tz)
+            self.assertEqual(dtz.utctimetuple()[:-1],
+                             dtz.astimezone(timezone.utc).timetuple()[:-1])
+        # At the edges, UTC adjustment can produce years out-of-range
+        # for a datetime object.  Ensure that an OverflowError is
+        # raised.
         tiny = cls(MINYEAR, 1, 1, 0, 0, 37, tzinfo=UOFS(1439))
         # That goes back 1 minute less than a full day.
-        t = tiny.utctimetuple()
-        self.assertEqual(t.tm_year, MINYEAR-1)
-        self.assertEqual(t.tm_mon, 12)
-        self.assertEqual(t.tm_mday, 31)
-        self.assertEqual(t.tm_hour, 0)
-        self.assertEqual(t.tm_min, 1)
-        self.assertEqual(t.tm_sec, 37)
-        self.assertEqual(t.tm_yday, 366)    # "year 0" is a leap year
-        self.assertEqual(t.tm_isdst, 0)
+        self.assertRaises(OverflowError, tiny.utctimetuple)
 
         huge = cls(MAXYEAR, 12, 31, 23, 59, 37, 999999, tzinfo=UOFS(-1439))
         # That goes forward 1 minute less than a full day.
-        t = huge.utctimetuple()
-        self.assertEqual(t.tm_year, MAXYEAR+1)
-        self.assertEqual(t.tm_mon, 1)
-        self.assertEqual(t.tm_mday, 1)
-        self.assertEqual(t.tm_hour, 23)
-        self.assertEqual(t.tm_min, 58)
-        self.assertEqual(t.tm_sec, 37)
-        self.assertEqual(t.tm_yday, 1)
-        self.assertEqual(t.tm_isdst, 0)
+        self.assertRaises(OverflowError, huge.utctimetuple)
+        # More overflow cases
+        tiny = cls.min.replace(tzinfo=timezone(MINUTE))
+        self.assertRaises(OverflowError, tiny.utctimetuple)
+        huge = cls.max.replace(tzinfo=timezone(-MINUTE))
+        self.assertRaises(OverflowError, huge.utctimetuple)
 
     def test_tzinfo_isoformat(self):
         zero = FixedOffset(0, "+00:00")
@@ -3101,6 +3301,7 @@
     return dt
 
 ZERO = timedelta(0)
+MINUTE = timedelta(minutes=1)
 HOUR = timedelta(hours=1)
 DAY = timedelta(days=1)
 # In the US, DST starts at 2am (standard time) on the first Sunday in April.
@@ -3344,6 +3545,19 @@
             def dst(self, dt): return None
         self.assertRaises(ValueError, now.astimezone, notok())
 
+        # Sometimes blow up. In the following, tzinfo.dst()
+        # implementation may return None or not None depending on
+        # whether DST is assumed to be in effect.  In this situation,
+        # a ValueError should be raised by astimezone().
+        class tricky_notok(ok):
+            def dst(self, dt):
+                if dt.year == 2000:
+                    return None
+                else:
+                    return 10*HOUR
+        dt = self.theclass(2001, 1, 1).replace(tzinfo=utc_real)
+        self.assertRaises(ValueError, dt.astimezone, tricky_notok())
+
     def test_fromutc(self):
         self.assertRaises(TypeError, Eastern.fromutc)   # not enough args
         now = datetime.utcnow().replace(tzinfo=utc_real)

Modified: python/branches/py3k-cdecimal/Lib/test/test_decimal.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_decimal.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_decimal.py	Sat Jul 17 12:36:09 2010
@@ -78,10 +78,41 @@
 
 # list of individual .decTest test ids that correspond to tests that
 # we're skipping for one reason or another.
-skipped_test_ids = [
-    'scbx164',  # skipping apparently implementation-specific scaleb
-    'scbx165',  # tests, pending clarification of scaleb rules.
-]
+skipped_test_ids = set([
+    # Skip implementation-specific scaleb tests.
+    'scbx164',
+    'scbx165',
+
+    # For some operations (currently exp, ln, log10, power), the decNumber
+    # reference implementation imposes additional restrictions on the context
+    # and operands.  These restrictions are not part of the specification;
+    # however, the effect of these restrictions does show up in some of the
+    # testcases.  We skip testcases that violate these restrictions, since
+    # Decimal behaves differently from decNumber for these testcases so these
+    # testcases would otherwise fail.
+    'expx901',
+    'expx902',
+    'expx903',
+    'expx905',
+    'lnx901',
+    'lnx902',
+    'lnx903',
+    'lnx905',
+    'logx901',
+    'logx902',
+    'logx903',
+    'logx905',
+    'powx1183',
+    'powx1184',
+    'powx4001',
+    'powx4002',
+    'powx4003',
+    'powx4005',
+    'powx4008',
+    'powx4010',
+    'powx4012',
+    'powx4014',
+    ])
 
 # Make sure it actually raises errors when not expected and caught in flags
 # Slower, since it runs some things several times.
@@ -172,27 +203,6 @@
     'same_quantum',
     )
 
-# For some operations (currently exp, ln, log10, power), the decNumber
-# reference implementation imposes additional restrictions on the
-# context and operands.  These restrictions are not part of the
-# specification; however, the effect of these restrictions does show
-# up in some of the testcases.  We skip testcases that violate these
-# restrictions, since Decimal behaves differently from decNumber for
-# these testcases so these testcases would otherwise fail.
-
-decNumberRestricted = ('power', 'ln', 'log10', 'exp')
-DEC_MAX_MATH = 999999
-def outside_decNumber_bounds(v, context):
-    if (context.prec > DEC_MAX_MATH or
-        context.Emax > DEC_MAX_MATH or
-        -context.Emin > DEC_MAX_MATH):
-        return True
-    if not v._is_special and v and (
-        v.adjusted() > DEC_MAX_MATH or
-        v.adjusted() < 1-2*DEC_MAX_MATH):
-        return True
-    return False
-
 class DecimalTest(unittest.TestCase):
     """Class which tests the Decimal class against the test cases.
 
@@ -330,22 +340,6 @@
 
         ans = FixQuotes(ans)
 
-        # skip tests that are related to bounds imposed in the decNumber
-        # reference implementation
-        if fname in decNumberRestricted:
-            if fname == 'power':
-                if not (vals[1]._isinteger() and
-                        -1999999997 <= vals[1] <= 999999999):
-                    if outside_decNumber_bounds(vals[0], self.context) or \
-                            outside_decNumber_bounds(vals[1], self.context):
-                        #print "Skipping test %s" % s
-                        return
-            else:
-                if outside_decNumber_bounds(vals[0], self.context):
-                    #print "Skipping test %s" % s
-                    return
-
-
         if EXTENDEDERRORTEST and fname not in ('to_sci_string', 'to_eng_string'):
             for error in theirexceptions:
                 self.context.traps[error] = 1

Modified: python/branches/py3k-cdecimal/Lib/test/test_descr.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_descr.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_descr.py	Sat Jul 17 12:36:09 2010
@@ -1,6 +1,7 @@
 import builtins
 import sys
 import types
+import math
 import unittest
 
 from copy import deepcopy
@@ -1578,6 +1579,9 @@
             ("__exit__", run_context, swallow, set(), {"__enter__" : iden}),
             ("__complex__", complex, complex_num, set(), {}),
             ("__format__", format, format_impl, set(), {}),
+            ("__floor__", math.floor, zero, set(), {}),
+            ("__trunc__", math.trunc, zero, set(), {}),
+            ("__ceil__", math.ceil, zero, set(), {}),
             ]
 
         class Checker(object):

Modified: python/branches/py3k-cdecimal/Lib/test/test_dict.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_dict.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_dict.py	Sat Jul 17 12:36:09 2010
@@ -664,6 +664,7 @@
         gc.collect()
         self.assertTrue(gc.is_tracked(t), t)
 
+    @support.cpython_only
     def test_track_literals(self):
         # Test GC-optimization of dict literals
         x, y, z, w = 1.5, "a", (1, None), []
@@ -681,6 +682,7 @@
         self._tracked({1: {}})
         self._tracked({1: set()})
 
+    @support.cpython_only
     def test_track_dynamic(self):
         # Test GC-optimization of dynamically-created dicts
         class MyObject(object):
@@ -744,6 +746,7 @@
         d.update([(x, y), (z, w)])
         self._tracked(d)
 
+    @support.cpython_only
     def test_track_subtypes(self):
         # Dict subtypes are always tracked
         class MyDict(dict):

Modified: python/branches/py3k-cdecimal/Lib/test/test_dis.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_dis.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_dis.py	Sat Jul 17 12:36:09 2010
@@ -96,6 +96,46 @@
 
 """
 
+expr_str = "x + 1"
+
+dis_expr_str = """\
+  1           0 LOAD_NAME                0 (x)
+              3 LOAD_CONST               0 (1)
+              6 BINARY_ADD
+              7 RETURN_VALUE
+"""
+
+simple_stmt_str = "x = x + 1"
+
+dis_simple_stmt_str = """\
+  1           0 LOAD_NAME                0 (x)
+              3 LOAD_CONST               0 (1)
+              6 BINARY_ADD
+              7 STORE_NAME               0 (x)
+             10 LOAD_CONST               1 (None)
+             13 RETURN_VALUE
+"""
+
+compound_stmt_str = """\
+x = 0
+while 1:
+    x += 1"""
+# Trailing newline has been deliberately omitted
+
+dis_compound_stmt_str = """\
+  1           0 LOAD_CONST               0 (0)
+              3 STORE_NAME               0 (x)
+
+  2           6 SETUP_LOOP              13 (to 22)
+
+  3     >>    9 LOAD_NAME                0 (x)
+             12 LOAD_CONST               1 (1)
+             15 INPLACE_ADD
+             16 STORE_NAME               0 (x)
+             19 JUMP_ABSOLUTE            9
+        >>   22 LOAD_CONST               2 (None)
+             25 RETURN_VALUE
+"""
 
 class DisTests(unittest.TestCase):
     def do_disassembly_test(self, func, expected):
@@ -166,6 +206,11 @@
         from test import dis_module
         self.do_disassembly_test(dis_module, dis_module_expected_results)
 
+    def test_disassemble_str(self):
+        self.do_disassembly_test(expr_str, dis_expr_str)
+        self.do_disassembly_test(simple_stmt_str, dis_simple_stmt_str)
+        self.do_disassembly_test(compound_stmt_str, dis_compound_stmt_str)
+
 def test_main():
     run_unittest(DisTests)
 

Modified: python/branches/py3k-cdecimal/Lib/test/test_enumerate.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_enumerate.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_enumerate.py	Sat Jul 17 12:36:09 2010
@@ -103,6 +103,7 @@
         self.assertRaises(TypeError, self.enum, 'abc', 'a') # wrong type
         self.assertRaises(TypeError, self.enum, 'abc', 2, 3) # too many arguments
 
+    @support.cpython_only
     def test_tuple_reuse(self):
         # Tests an implementation detail where tuple is reused
         # whenever nothing else holds a reference to it
@@ -145,6 +146,7 @@
         x = range(1)
         self.assertEqual(type(reversed(x)), type(iter(x)))
 
+    @support.cpython_only
     def test_len(self):
         # This is an implementation detail, not an interface requirement
         from test.test_iterlen import len

Modified: python/branches/py3k-cdecimal/Lib/test/test_exceptions.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_exceptions.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_exceptions.py	Sat Jul 17 12:36:09 2010
@@ -7,7 +7,7 @@
 import weakref
 
 from test.support import (TESTFN, unlink, run_unittest, captured_output,
-                          gc_collect)
+                          gc_collect, cpython_only)
 
 # XXX This is not really enough, each *operation* should be tested!
 
@@ -137,6 +137,7 @@
         ckmsg(s, "'continue' not properly in loop")
         ckmsg("continue\n", "'continue' not properly in loop")
 
+    @cpython_only
     def testSettingException(self):
         # test that setting an exception at the C level works even if the
         # exception object can't be constructed.
@@ -669,6 +670,7 @@
         tb2 = raiseMemError()
         self.assertEqual(tb1, tb2)
 
+    @cpython_only
     def test_exception_with_doc(self):
         import _testcapi
         doc2 = "This is a test docstring."

Modified: python/branches/py3k-cdecimal/Lib/test/test_extcall.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_extcall.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_extcall.py	Sat Jul 17 12:36:09 2010
@@ -279,13 +279,13 @@
     >>> f(6, a=4, *(1, 2, 3))
     Traceback (most recent call last):
       ...
-    TypeError: f() takes exactly 1 argument (5 given)
+    TypeError: f() takes exactly 1 positional argument (5 given)
     >>> def f(a, *, kw):
     ...    pass
     >>> f(6, 4, kw=4)
     Traceback (most recent call last):
       ...
-    TypeError: f() takes exactly 2 arguments (3 given)
+    TypeError: f() takes exactly 1 positional argument (3 given)
 """
 
 import sys

Modified: python/branches/py3k-cdecimal/Lib/test/test_float.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_float.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_float.py	Sat Jul 17 12:36:09 2010
@@ -7,13 +7,17 @@
 from math import isinf, isnan, copysign, ldexp
 import operator
 import random, fractions
-import re
 
 INF = float("inf")
 NAN = float("nan")
 
+have_getformat = hasattr(float, "__getformat__")
+requires_getformat = unittest.skipUnless(have_getformat,
+                                         "requires __getformat__")
+requires_setformat = unittest.skipUnless(hasattr(float, "__setformat__"),
+                                         "requires __setformat__")
 # decorator for skipping tests on non-IEEE 754 platforms
-requires_IEEE_754 = unittest.skipUnless(
+requires_IEEE_754 = unittest.skipUnless(have_getformat and
     float.__getformat__("double").startswith("IEEE"),
     "test requires IEEE 754 doubles")
 
@@ -21,74 +25,6 @@
 test_dir = os.path.dirname(__file__) or os.curdir
 format_testfile = os.path.join(test_dir, 'formatfloat_testcases.txt')
 
-finite_decimal_parser = re.compile(r"""    # A numeric string consists of:
-    (?P<sign>[-+])?          # an optional sign, followed by
-    (?=\d|\.\d)              # a number with at least one digit
-    (?P<int>\d*)             # having a (possibly empty) integer part
-    (?:\.(?P<frac>\d*))?     # followed by an optional fractional part
-    (?:E(?P<exp>[-+]?\d+))?  # and an optional exponent
-    \Z
-""", re.VERBOSE | re.IGNORECASE | re.UNICODE).match
-
-# Pure Python version of correctly rounded string->float conversion.
-# Avoids any use of floating-point by returning the result as a hex string.
-def strtod(s, mant_dig=53, min_exp = -1021, max_exp = 1024):
-    """Convert a finite decimal string to a hex string representing an
-    IEEE 754 binary64 float.  Return 'inf' or '-inf' on overflow.
-    This function makes no use of floating-point arithmetic at any
-    stage."""
-
-    # parse string into a pair of integers 'a' and 'b' such that
-    # abs(decimal value) = a/b, and a boolean 'negative'.
-    m = finite_decimal_parser(s)
-    if m is None:
-        raise ValueError('invalid numeric string')
-    fraction = m.group('frac') or ''
-    intpart = int(m.group('int') + fraction)
-    exp = int(m.group('exp') or '0') - len(fraction)
-    negative = m.group('sign') == '-'
-    a, b = intpart*10**max(exp, 0), 10**max(0, -exp)
-
-    # quick return for zeros
-    if not a:
-        return '-0x0.0p+0' if negative else '0x0.0p+0'
-
-    # compute exponent e for result; may be one too small in the case
-    # that the rounded value of a/b lies in a different binade from a/b
-    d = a.bit_length() - b.bit_length()
-    d += (a >> d if d >= 0 else a << -d) >= b
-    e = max(d, min_exp) - mant_dig
-
-    # approximate a/b by number of the form q * 2**e; adjust e if necessary
-    a, b = a << max(-e, 0), b << max(e, 0)
-    q, r = divmod(a, b)
-    if 2*r > b or 2*r == b and q & 1:
-        q += 1
-        if q.bit_length() == mant_dig+1:
-            q //= 2
-            e += 1
-
-    # double check that (q, e) has the right form
-    assert q.bit_length() <= mant_dig and e >= min_exp - mant_dig
-    assert q.bit_length() == mant_dig or e == min_exp - mant_dig
-
-    # check for overflow and underflow
-    if e + q.bit_length() > max_exp:
-        return '-inf' if negative else 'inf'
-    if not q:
-        return '-0x0.0p+0' if negative else '0x0.0p+0'
-
-    # for hex representation, shift so # bits after point is a multiple of 4
-    hexdigs = 1 + (mant_dig-2)//4
-    shift = 3 - (mant_dig-2)%4
-    q, e = q << shift, e - shift
-    return '{}0x{:x}.{:0{}x}p{:+d}'.format(
-        '-' if negative else '',
-        q // 16**hexdigs,
-        q % 16**hexdigs,
-        hexdigs,
-        e + 4*hexdigs)
-
 class GeneralFloatCases(unittest.TestCase):
 
     def test_float(self):
@@ -446,6 +382,7 @@
             #self.assertTrue(0.0 > pow_op(-2.0, -1047) > -1e-315)
 
 
+ at requires_setformat
 class FormatFunctionsTestCase(unittest.TestCase):
 
     def setUp(self):
@@ -496,6 +433,7 @@
 # on non-IEEE platforms, attempting to unpack a bit pattern
 # representing an infinity or a NaN should raise an exception.
 
+ at requires_setformat
 class UnknownFormatTestCase(unittest.TestCase):
     def setUp(self):
         self.save_formats = {'double':float.__getformat__('double'),
@@ -528,36 +466,25 @@
 # let's also try to guarantee that -0.0 and 0.0 don't get confused.
 
 class IEEEFormatTestCase(unittest.TestCase):
-    if float.__getformat__("double").startswith("IEEE"):
-        def test_double_specials_do_unpack(self):
-            for fmt, data in [('>d', BE_DOUBLE_INF),
-                              ('>d', BE_DOUBLE_NAN),
-                              ('<d', LE_DOUBLE_INF),
-                              ('<d', LE_DOUBLE_NAN)]:
-                struct.unpack(fmt, data)
-
-    if float.__getformat__("float").startswith("IEEE"):
-        def test_float_specials_do_unpack(self):
-            for fmt, data in [('>f', BE_FLOAT_INF),
-                              ('>f', BE_FLOAT_NAN),
-                              ('<f', LE_FLOAT_INF),
-                              ('<f', LE_FLOAT_NAN)]:
-                struct.unpack(fmt, data)
-
-    if float.__getformat__("double").startswith("IEEE"):
-        def test_negative_zero(self):
-            def pos_pos():
-                return 0.0, math.atan2(0.0, -1)
-            def pos_neg():
-                return 0.0, math.atan2(-0.0, -1)
-            def neg_pos():
-                return -0.0, math.atan2(0.0, -1)
-            def neg_neg():
-                return -0.0, math.atan2(-0.0, -1)
-            self.assertEquals(pos_pos(), neg_pos())
-            self.assertEquals(pos_neg(), neg_neg())
+
+    @requires_IEEE_754
+    def test_double_specials_do_unpack(self):
+        for fmt, data in [('>d', BE_DOUBLE_INF),
+                          ('>d', BE_DOUBLE_NAN),
+                          ('<d', LE_DOUBLE_INF),
+                          ('<d', LE_DOUBLE_NAN)]:
+            struct.unpack(fmt, data)
+
+    @requires_IEEE_754
+    def test_float_specials_do_unpack(self):
+        for fmt, data in [('>f', BE_FLOAT_INF),
+                          ('>f', BE_FLOAT_NAN),
+                          ('<f', LE_FLOAT_INF),
+                          ('<f', LE_FLOAT_NAN)]:
+            struct.unpack(fmt, data)
 
 class FormatTestCase(unittest.TestCase):
+
     def test_format(self):
         # these should be rewritten to use both format(x, spec) and
         # x.__format__(spec)
@@ -611,8 +538,7 @@
         self.assertEqual(format(INF, 'f'), 'inf')
         self.assertEqual(format(INF, 'F'), 'INF')
 
-    @unittest.skipUnless(float.__getformat__("double").startswith("IEEE"),
-                         "test requires IEEE 754 doubles")
+    @requires_IEEE_754
     def test_format_testfile(self):
         for line in open(format_testfile):
             if line.startswith('--'):
@@ -692,9 +618,10 @@
             self.assertEqual(s, repr(float(s)))
             self.assertEqual(negs, repr(float(negs)))
 
+
+ at requires_IEEE_754
 class RoundTestCase(unittest.TestCase):
-    @unittest.skipUnless(float.__getformat__("double").startswith("IEEE"),
-                         "test requires IEEE 754 doubles")
+
     def test_inf_nan(self):
         self.assertRaises(OverflowError, round, INF)
         self.assertRaises(OverflowError, round, -INF)
@@ -704,8 +631,6 @@
         self.assertRaises(TypeError, round, NAN, "ceci n'est pas un integer")
         self.assertRaises(TypeError, round, -0.0, 1j)
 
-    @unittest.skipUnless(float.__getformat__("double").startswith("IEEE"),
-                         "test requires IEEE 754 doubles")
     def test_large_n(self):
         for n in [324, 325, 400, 2**31-1, 2**31, 2**32, 2**100]:
             self.assertEqual(round(123.456, n), 123.456)
@@ -718,8 +643,6 @@
         self.assertEqual(round(1e150, 309), 1e150)
         self.assertEqual(round(1.4e-315, 315), 1e-315)
 
-    @unittest.skipUnless(float.__getformat__("double").startswith("IEEE"),
-                         "test requires IEEE 754 doubles")
     def test_small_n(self):
         for n in [-308, -309, -400, 1-2**31, -2**31, -2**31-1, -2**100]:
             self.assertEqual(round(123.456, n), 0.0)
@@ -727,8 +650,6 @@
             self.assertEqual(round(1e300, n), 0.0)
             self.assertEqual(round(1e-320, n), 0.0)
 
-    @unittest.skipUnless(float.__getformat__("double").startswith("IEEE"),
-                         "test requires IEEE 754 doubles")
     def test_overflow(self):
         self.assertRaises(OverflowError, round, 1.6e308, -308)
         self.assertRaises(OverflowError, round, -1.7e308, -308)
@@ -776,9 +697,6 @@
             self.assertEqual(float(format(x, '.2f')), round(x, 2))
             self.assertEqual(float(format(x, '.3f')), round(x, 3))
 
-
-    @unittest.skipUnless(float.__getformat__("double").startswith("IEEE"),
-                         "test requires IEEE 754 doubles")
     def test_format_specials(self):
         # Test formatting of nans and infs.
 
@@ -1334,38 +1252,6 @@
             else:
                 self.identical(x, fromHex(toHex(x)))
 
-class StrtodTestCase(unittest.TestCase):
-    def check_string(self, s):
-        expected = strtod(s)
-        try:
-            fs = float(s)
-        except OverflowError:
-            got = '-inf' if s[0] == '-' else 'inf'
-        else:
-            got = fs.hex()
-        self.assertEqual(expected, got,
-                         "Incorrectly rounded str->float conversion for "
-                         "{}: expected {}, got {}".format(s, expected, got))
-
-    @unittest.skipUnless(getattr(sys, 'float_repr_style', '') == 'short',
-                         "applies only when using short float repr style")
-    def test_bug7632(self):
-        # check a few particular values that gave incorrectly rounded
-        # results with previous versions of dtoa.c
-        test_strings = [
-            '94393431193180696942841837085033647913224148539854e-358',
-            '12579816049008305546974391768996369464963024663104e-357',
-            '17489628565202117263145367596028389348922981857013e-357',
-            '18487398785991994634182916638542680759613590482273e-357',
-            '32002864200581033134358724675198044527469366773928e-358',
-            '73608278998966969345824653500136787876436005957953e-358',
-            '64774478836417299491718435234611299336288082136054e-358',
-            '13704940134126574534878641876947980878824688451169e-357',
-            '46697445774047060960624497964425416610480524760471e-358',
-            ]
-        for s in test_strings:
-            self.check_string(s)
-
 
 def test_main():
     support.run_unittest(
@@ -1378,7 +1264,6 @@
         RoundTestCase,
         InfNanTest,
         HexFloatTestCase,
-        StrtodTestCase,
         )
 
 if __name__ == '__main__':

Modified: python/branches/py3k-cdecimal/Lib/test/test_fnmatch.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_fnmatch.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_fnmatch.py	Sat Jul 17 12:36:09 2010
@@ -3,10 +3,14 @@
 from test import support
 import unittest
 
-from fnmatch import fnmatch, fnmatchcase
+from fnmatch import fnmatch, fnmatchcase, _MAXCACHE, _cache, _cacheb, purge
 
 
 class FnmatchTestCase(unittest.TestCase):
+
+    def tearDown(self):
+        purge()
+
     def check_match(self, filename, pattern, should_match=1, fn=fnmatch):
         if should_match:
             self.assertTrue(fn(filename, pattern),
@@ -60,6 +64,21 @@
         self.check_match(b'test\xff', b'te*\xff')
         self.check_match(b'foo\nbar', b'foo*')
 
+    def test_cache_clearing(self):
+        # check that caches do not grow too large
+        # http://bugs.python.org/issue7846
+
+        # string pattern cache
+        for i in range(_MAXCACHE + 1):
+            fnmatch('foo', '?' * i)
+
+        self.assertLessEqual(len(_cache), _MAXCACHE)
+
+        # bytes pattern cache
+        for i in range(_MAXCACHE + 1):
+            fnmatch(b'foo', b'?' * i)
+        self.assertLessEqual(len(_cacheb), _MAXCACHE)
+
 
 def test_main():
     support.run_unittest(FnmatchTestCase)

Modified: python/branches/py3k-cdecimal/Lib/test/test_gdb.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_gdb.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_gdb.py	Sat Jul 17 12:36:09 2010
@@ -9,6 +9,7 @@
 import sys
 import unittest
 import locale
+import sysconfig
 
 from test.support import run_unittest, findfile
 
@@ -664,6 +665,15 @@
                                     r".*\na = 1\nb = 2\nc = 3\n.*")
 
 def test_main():
+    cflags = sysconfig.get_config_vars()['PY_CFLAGS']
+    final_opt = ""
+    for opt in cflags.split():
+        if opt.startswith('-O'):
+            final_opt = opt
+    if final_opt and final_opt != '-O0':
+        raise unittest.SkipTest("Python was built with compiler optimizations, "
+                                "tests can't reliably succeed")
+
     run_unittest(PrettyPrintTests,
                  PyListTests,
                  StackNavigationTests,

Modified: python/branches/py3k-cdecimal/Lib/test/test_genericpath.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_genericpath.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_genericpath.py	Sat Jul 17 12:36:09 2010
@@ -194,14 +194,18 @@
     ]
 
     def test_normcase(self):
-        # Check that normcase() is idempotent
-        p = "FoO/./BaR"
-        p = self.pathmodule.normcase(p)
-        self.assertEqual(p, self.pathmodule.normcase(p))
-
-        p = b"FoO/./BaR"
-        p = self.pathmodule.normcase(p)
-        self.assertEqual(p, self.pathmodule.normcase(p))
+        normcase = self.pathmodule.normcase
+        # check that normcase() is idempotent
+        for p in ["FoO/./BaR", b"FoO/./BaR"]:
+            p = normcase(p)
+            self.assertEqual(p, normcase(p))
+
+        self.assertEqual(normcase(''), '')
+        self.assertEqual(normcase(b''), b'')
+
+        # check that normcase raises a TypeError for invalid types
+        for path in (None, True, 0, 2.5, [], bytearray(b''), {'o','o'}):
+            self.assertRaises(TypeError, normcase, path)
 
     def test_splitdrive(self):
         # splitdrive for non-NT paths

Modified: python/branches/py3k-cdecimal/Lib/test/test_getargs2.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_getargs2.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_getargs2.py	Sat Jul 17 12:36:09 2010
@@ -325,7 +325,7 @@
         from _testcapi import getargs_z
         self.assertEqual(getargs_z('abc\xe9'), b'abc\xc3\xa9')
         self.assertRaises(TypeError, getargs_z, 'nul:\0')
-        self.assertEqual(getargs_z(b'bytes'), b'bytes')
+        self.assertRaises(TypeError, getargs_z, b'bytes')
         self.assertRaises(TypeError, getargs_z, bytearray(b'bytearray'))
         self.assertRaises(TypeError, getargs_z, memoryview(b'memoryview'))
         self.assertIsNone(getargs_z(None))
@@ -375,6 +375,16 @@
         self.assertRaises(TypeError, getargs_y_hash, memoryview(b'memoryview'))
         self.assertRaises(TypeError, getargs_y_hash, None)
 
+    def test_w_star(self):
+        # getargs_w_star() modifies first and last byte
+        from _testcapi import getargs_w_star
+        self.assertRaises(TypeError, getargs_w_star, 'abc\xe9')
+        self.assertRaises(TypeError, getargs_w_star, b'bytes')
+        self.assertRaises(TypeError, getargs_w_star, b'nul:\0')
+        self.assertEqual(getargs_w_star(bytearray(b'bytearray')), b'[ytearra]')
+        self.assertEqual(getargs_w_star(memoryview(b'memoryview')), b'[emoryvie]')
+        self.assertRaises(TypeError, getargs_w_star, None)
+
 
 class Unicode_TestCase(unittest.TestCase):
     def test_u(self):

Modified: python/branches/py3k-cdecimal/Lib/test/test_glob.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_glob.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_glob.py	Sat Jul 17 12:36:09 2010
@@ -1,5 +1,5 @@
 import unittest
-from test.support import run_unittest, TESTFN
+from test.support import run_unittest, TESTFN, skip_unless_symlink, can_symlink
 import glob
 import os
 import shutil
@@ -25,7 +25,7 @@
         self.mktemp('ZZZ')
         self.mktemp('a', 'bcd', 'EF')
         self.mktemp('a', 'bcd', 'efg', 'ha')
-        if hasattr(os, 'symlink'):
+        if can_symlink():
             os.symlink(self.norm('broken'), self.norm('sym1'))
             os.symlink(self.norm('broken'), self.norm('sym2'))
 
@@ -98,12 +98,12 @@
         # either of these results are reasonable
         self.assertIn(res[0], [self.tempdir, self.tempdir + os.sep])
 
+    @skip_unless_symlink
     def test_glob_broken_symlinks(self):
-        if hasattr(os, 'symlink'):
-            eq = self.assertSequencesEqual_noorder
-            eq(self.glob('sym*'), [self.norm('sym1'), self.norm('sym2')])
-            eq(self.glob('sym1'), [self.norm('sym1')])
-            eq(self.glob('sym2'), [self.norm('sym2')])
+        eq = self.assertSequencesEqual_noorder
+        eq(self.glob('sym*'), [self.norm('sym1'), self.norm('sym2')])
+        eq(self.glob('sym1'), [self.norm('sym1')])
+        eq(self.glob('sym2'), [self.norm('sym2')])
 
 
 def test_main():

Modified: python/branches/py3k-cdecimal/Lib/test/test_http_cookiejar.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_http_cookiejar.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_http_cookiejar.py	Sat Jul 17 12:36:09 2010
@@ -97,6 +97,7 @@
 
 
 class HeaderTests(TestCase):
+
     def test_parse_ns_headers(self):
         # quotes should be stripped
         expected = [[('foo', 'bar'), ('expires', 2209069412), ('version', '0')]]
@@ -106,6 +107,16 @@
             ]:
             self.assertEquals(parse_ns_headers([hdr]), expected)
 
+    def test_parse_ns_headers_version(self):
+
+        # quotes should be stripped
+        expected = [[('foo', 'bar'), ('version', '1')]]
+        for hdr in [
+            'foo=bar; version="1"',
+            'foo=bar; Version="1"',
+            ]:
+            self.assertEquals(parse_ns_headers([hdr]), expected)
+
     def test_parse_ns_headers_special_names(self):
         # names such as 'expires' are not special in first name=value pair
         # of Set-Cookie: header
@@ -1020,6 +1031,8 @@
             ["Set-Cookie2: a=foo; path=/; Version=1; domain"],
             # bad max-age
             ["Set-Cookie: b=foo; max-age=oops"],
+            # bad version
+            ["Set-Cookie: b=foo; version=spam"],
             ]:
             c = cookiejar_from_cookie_headers(headers)
             # these bad cookies shouldn't be set

Modified: python/branches/py3k-cdecimal/Lib/test/test_httpservers.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_httpservers.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_httpservers.py	Sat Jul 17 12:36:09 2010
@@ -299,7 +299,7 @@
 
         # The shebang line should be pure ASCII: use symlink if possible.
         # See issue #7668.
-        if hasattr(os, 'symlink'):
+        if support.can_symlink():
             self.pythonexe = os.path.join(self.parent_dir, 'python')
             os.symlink(sys.executable, self.pythonexe)
         else:

Modified: python/branches/py3k-cdecimal/Lib/test/test_imp.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_imp.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_imp.py	Sat Jul 17 12:36:09 2010
@@ -44,7 +44,7 @@
 class ImportTests(unittest.TestCase):
 
     def test_find_module_encoding(self):
-        fd = imp.find_module("heapq")[0]
+        fd = imp.find_module("pydoc")[0]
         self.assertEqual(fd.encoding, "iso-8859-1")
 
     def test_issue1267(self):

Modified: python/branches/py3k-cdecimal/Lib/test/test_import.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_import.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_import.py	Sat Jul 17 12:36:09 2010
@@ -147,22 +147,24 @@
         filename = module + '.py'
 
         # Create a file with a list of 65000 elements.
-        with open(filename, 'w+') as f:
+        with open(filename, 'w') as f:
             f.write('d = [\n')
             for i in range(65000):
                 f.write('"",\n')
             f.write(']')
 
-        # Compile & remove .py file; we only need .pyc (or .pyo).
-        # Bytecode must be relocated from the PEP 3147 bytecode-only location.
-        py_compile.compile(filename)
-        unlink(filename)
-        make_legacy_pyc(filename)
+        try:
+            # Compile & remove .py file; we only need .pyc (or .pyo).
+            # Bytecode must be relocated from the PEP 3147 bytecode-only location.
+            py_compile.compile(filename)
+        finally:
+            unlink(filename)
 
         # Need to be able to load from current dir.
         sys.path.append('')
 
         try:
+            make_legacy_pyc(filename)
             # This used to crash.
             exec('import ' + module)
         finally:
@@ -468,6 +470,17 @@
         ns = dict(__package__=object())
         self.assertRaises(ValueError, check_relative)
 
+    def test_absolute_import_without_future(self):
+        # If absolute import syntax is used, then do not try to perform
+        # a relative import in the face of failure.
+        # Issue #7902.
+        try:
+            from .os import sep
+        except ImportError:
+            pass
+        else:
+            self.fail("explicit relative import triggered an "
+                      "implicit relative import")
 
 class OverridingImportBuiltinTests(unittest.TestCase):
     def test_override_builtin(self):

Modified: python/branches/py3k-cdecimal/Lib/test/test_inspect.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_inspect.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_inspect.py	Sat Jul 17 12:36:09 2010
@@ -3,6 +3,7 @@
 import types
 import unittest
 import inspect
+import linecache
 import datetime
 import collections
 from os.path import normcase
@@ -275,6 +276,11 @@
     def test_getsourcefile(self):
         self.assertEqual(normcase(inspect.getsourcefile(mod.spam)), modfile)
         self.assertEqual(normcase(inspect.getsourcefile(git.abuse)), modfile)
+        fn = "_non_existing_filename_used_for_sourcefile_test.py"
+        co = compile("None", fn, "exec")
+        self.assertEqual(inspect.getsourcefile(co), None)
+        linecache.cache[co.co_filename] = (1, None, "None", co.co_filename)
+        self.assertEqual(normcase(inspect.getsourcefile(co)), fn)
 
     def test_getfile(self):
         self.assertEqual(inspect.getfile(mod.StupidGit), mod.__file__)
@@ -373,6 +379,15 @@
         self.assertRaises(IOError, inspect.getsource, unicodedata)
         self.assertRaises(IOError, inspect.findsource, unicodedata)
 
+    def test_findsource_code_in_linecache(self):
+        lines = ["x=1"]
+        co = compile(lines[0], "_dynamically_created_file", "exec")
+        self.assertRaises(IOError, inspect.findsource, co)
+        self.assertRaises(IOError, inspect.getsource, co)
+        linecache.cache[co.co_filename] = (1, None, lines, co.co_filename)
+        self.assertEquals(inspect.findsource(co), (lines,0))
+        self.assertEquals(inspect.getsource(co), lines[0])
+
 # Helper for testing classify_class_attrs.
 def attrs_wo_objs(cls):
     return [t[:3] for t in inspect.classify_class_attrs(cls)]

Modified: python/branches/py3k-cdecimal/Lib/test/test_iter.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_iter.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_iter.py	Sat Jul 17 12:36:09 2010
@@ -1,7 +1,7 @@
 # Test iterators.
 
 import unittest
-from test.support import run_unittest, TESTFN, unlink
+from test.support import run_unittest, TESTFN, unlink, cpython_only
 
 # Test result of triple loop (too big to inline)
 TRIPLETS = [(0, 0, 0), (0, 0, 1), (0, 0, 2),
@@ -761,8 +761,9 @@
         (a, b), (c,) = IteratingSequenceClass(2), {42: 24}
         self.assertEqual((a, b, c), (0, 1, 42))
 
-        # Test reference count behavior
 
+    @cpython_only
+    def test_ref_counting_behavior(self):
         class C(object):
             count = 0
             def __new__(cls):

Modified: python/branches/py3k-cdecimal/Lib/test/test_keywordonlyarg.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_keywordonlyarg.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_keywordonlyarg.py	Sat Jul 17 12:36:09 2010
@@ -73,6 +73,14 @@
         fundef3 += "lastarg):\n  pass\n"
         compile(fundef3, "<test>", "single")
 
+    def testTooManyPositionalErrorMessage(self):
+        def f(a, b=None, *, c=None):
+            pass
+        with self.assertRaises(TypeError) as exc:
+            f(1, 2, 3)
+        expected = "f() takes at most 2 positional arguments (3 given)"
+        self.assertEqual(str(exc.exception), expected)
+
     def testSyntaxErrorForFunctionCall(self):
         self.assertRaisesSyntaxError("f(p, k=1, p2)")
         self.assertRaisesSyntaxError("f(p, k1=50, *(1,2), k1=100)")

Modified: python/branches/py3k-cdecimal/Lib/test/test_logging.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_logging.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_logging.py	Sat Jul 17 12:36:09 2010
@@ -1670,6 +1670,8 @@
         t = logging.config.listen(port)
         t.start()
         t.ready.wait()
+        # Now get the port allocated
+        port = t.port
         t.ready.clear()
         try:
             sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)

Modified: python/branches/py3k-cdecimal/Lib/test/test_math.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_math.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_math.py	Sat Jul 17 12:36:09 2010
@@ -647,15 +647,7 @@
 
     def testLog1p(self):
         self.assertRaises(TypeError, math.log1p)
-        self.ftest('log1p(1/e -1)', math.log1p(1/math.e-1), -1)
-        self.ftest('log1p(0)', math.log1p(0), 0)
-        self.ftest('log1p(e-1)', math.log1p(math.e-1), 1)
-        self.ftest('log1p(1)', math.log1p(1), math.log(2))
-        self.assertEquals(math.log1p(INF), INF)
-        self.assertRaises(ValueError, math.log1p, NINF)
-        self.assertTrue(math.isnan(math.log1p(NAN)))
         n= 2**90
-        self.assertAlmostEquals(math.log1p(n), 62.383246250395075)
         self.assertAlmostEquals(math.log1p(n), math.log1p(float(n)))
 
     def testLog10(self):
@@ -923,12 +915,14 @@
         self.assertRaises(TypeError, math.trunc, 1, 2)
         self.assertRaises(TypeError, math.trunc, TestNoTrunc())
 
-        # XXX Doesn't work because the method is looked up on
-        #     the type only.
-        #t = TestNoTrunc()
-        #t.__trunc__ = lambda *args: args
-        #self.assertEquals((), math.trunc(t))
-        #self.assertRaises(TypeError, math.trunc, t, 0)
+    def testIsfinite(self):
+        self.assertTrue(math.isfinite(0.0))
+        self.assertTrue(math.isfinite(-0.0))
+        self.assertTrue(math.isfinite(1.0))
+        self.assertTrue(math.isfinite(-1.0))
+        self.assertFalse(math.isfinite(float("nan")))
+        self.assertFalse(math.isfinite(float("inf")))
+        self.assertFalse(math.isfinite(float("-inf")))
 
     def testIsnan(self):
         self.assertTrue(math.isnan(float("nan")))

Modified: python/branches/py3k-cdecimal/Lib/test/test_optparse.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_optparse.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_optparse.py	Sat Jul 17 12:36:09 2010
@@ -754,6 +754,11 @@
                            {'a': "-b3", 'boo': None, 'foo': None},
                            [])
 
+    def test_combined_single_invalid_option(self):
+        self.parser.add_option("-t", action="store_true")
+        self.assertParseFail(["-test"],
+                             "no such option: -e")
+
 class TestBool(BaseTest):
     def setUp(self):
         options = [make_option("-v",

Modified: python/branches/py3k-cdecimal/Lib/test/test_os.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_os.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_os.py	Sat Jul 17 12:36:09 2010
@@ -58,6 +58,7 @@
         os.closerange(first, first + 2)
         self.assertRaises(OSError, os.write, first, b"a")
 
+    @support.cpython_only
     def test_rename(self):
         path = support.TESTFN
         old = sys.getrefcount(path)
@@ -519,7 +520,7 @@
             f = open(path, "w")
             f.write("I'm " + path + " and proud of it.  Blame test_os.\n")
             f.close()
-        if hasattr(os, "symlink"):
+        if support.can_symlink():
             os.symlink(os.path.abspath(t2_path), link_path)
             sub2_tree = (sub2_path, ["link"], ["tmp3"])
         else:
@@ -563,7 +564,7 @@
         self.assertEqual(all[flipped + 1], (sub1_path, ["SUB11"], ["tmp2"]))
         self.assertEqual(all[2 - 2 * flipped], sub2_tree)
 
-        if hasattr(os, "symlink"):
+        if support.can_symlink():
             # Walk, following symlinks.
             for root, dirs, files in os.walk(walk_path, followlinks=True):
                 if root == link_path:
@@ -1032,6 +1033,84 @@
         self._kill_with_event(signal.CTRL_BREAK_EVENT, "CTRL_BREAK_EVENT")
 
 
+def skipUnlessWindows6(test):
+    if (hasattr(sys, 'getwindowsversion')
+        and sys.getwindowsversion().major >= 6):
+        return test
+    return unittest.skip("Requires Windows Vista or later")(test)
+
+ at unittest.skipUnless(sys.platform == "win32", "Win32 specific tests")
+ at support.skip_unless_symlink
+class Win32SymlinkTests(unittest.TestCase):
+    filelink = 'filelinktest'
+    filelink_target = os.path.abspath(__file__)
+    dirlink = 'dirlinktest'
+    dirlink_target = os.path.dirname(filelink_target)
+    missing_link = 'missing link'
+
+    def setUp(self):
+        assert os.path.exists(self.dirlink_target)
+        assert os.path.exists(self.filelink_target)
+        assert not os.path.exists(self.dirlink)
+        assert not os.path.exists(self.filelink)
+        assert not os.path.exists(self.missing_link)
+
+    def tearDown(self):
+        if os.path.exists(self.filelink):
+            os.remove(self.filelink)
+        if os.path.exists(self.dirlink):
+            os.rmdir(self.dirlink)
+        if os.path.lexists(self.missing_link):
+            os.remove(self.missing_link)
+
+    def test_directory_link(self):
+        os.symlink(self.dirlink_target, self.dirlink)
+        self.assertTrue(os.path.exists(self.dirlink))
+        self.assertTrue(os.path.isdir(self.dirlink))
+        self.assertTrue(os.path.islink(self.dirlink))
+        self.check_stat(self.dirlink, self.dirlink_target)
+
+    def test_file_link(self):
+        os.symlink(self.filelink_target, self.filelink)
+        self.assertTrue(os.path.exists(self.filelink))
+        self.assertTrue(os.path.isfile(self.filelink))
+        self.assertTrue(os.path.islink(self.filelink))
+        self.check_stat(self.filelink, self.filelink_target)
+
+    def _create_missing_dir_link(self):
+        'Create a "directory" link to a non-existent target'
+        linkname = self.missing_link
+        if os.path.lexists(linkname):
+            os.remove(linkname)
+        target = r'c:\\target does not exist.29r3c740'
+        assert not os.path.exists(target)
+        target_is_dir = True
+        os.symlink(target, linkname, target_is_dir)
+
+    def test_remove_directory_link_to_missing_target(self):
+        self._create_missing_dir_link()
+        # For compatibility with Unix, os.remove will check the
+        #  directory status and call RemoveDirectory if the symlink
+        #  was created with target_is_dir==True.
+        os.remove(self.missing_link)
+
+    @unittest.skip("currently fails; consider for improvement")
+    def test_isdir_on_directory_link_to_missing_target(self):
+        self._create_missing_dir_link()
+        # consider having isdir return true for directory links
+        self.assertTrue(os.path.isdir(self.missing_link))
+
+    @unittest.skip("currently fails; consider for improvement")
+    def test_rmdir_on_directory_link_to_missing_target(self):
+        self._create_missing_dir_link()
+        # consider allowing rmdir to remove directory links
+        os.rmdir(self.missing_link)
+
+    def check_stat(self, link, target):
+        self.assertEqual(os.stat(link), os.stat(target))
+        self.assertNotEqual(os.lstat(link), os.stat(link))
+
+
 class MiscTests(unittest.TestCase):
 
     @unittest.skipIf(os.name == "nt", "POSIX specific test")
@@ -1055,6 +1134,7 @@
         PosixUidGidTests,
         Pep383Tests,
         Win32KillTests,
+        Win32SymlinkTests,
         MiscTests,
     )
 

Modified: python/branches/py3k-cdecimal/Lib/test/test_parser.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_parser.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_parser.py	Sat Jul 17 12:36:09 2010
@@ -149,6 +149,13 @@
     def test_class_defs(self):
         self.check_suite("class foo():pass")
         self.check_suite("class foo(object):pass")
+        self.check_suite("@class_decorator\n"
+                         "class foo():pass")
+        self.check_suite("@class_decorator(arg)\n"
+                         "class foo():pass")
+        self.check_suite("@decorator1\n"
+                         "@decorator2\n"
+                         "class foo():pass")
 
     def test_import_from_statement(self):
         self.check_suite("from sys.path import *")
@@ -182,6 +189,18 @@
         self.check_suite("import sys as system, math")
         self.check_suite("import sys, math as my_math")
 
+    def test_relative_imports(self):
+        self.check_suite("from . import name")
+        self.check_suite("from .. import name")
+        # check all the way up to '....', since '...' is tokenized
+        # differently from '.' (it's an ellipsis token).
+        self.check_suite("from ... import name")
+        self.check_suite("from .... import name")
+        self.check_suite("from .pkg import name")
+        self.check_suite("from ..pkg import name")
+        self.check_suite("from ...pkg import name")
+        self.check_suite("from ....pkg import name")
+
     def test_pep263(self):
         self.check_suite("# -*- coding: iso-8859-1 -*-\n"
                          "pass\n")
@@ -457,6 +476,20 @@
                 (0, ''))
         self.check_bad_tree(tree, "malformed global ast")
 
+    def test_missing_import_source(self):
+        # from import fred
+        tree = \
+            (257,
+             (268,
+              (269,
+               (270,
+                (282,
+                 (284, (1, 'from'), (1, 'import'),
+                  (287, (285, (1, 'fred')))))),
+               (4, ''))),
+             (4, ''), (0, ''))
+        self.check_bad_tree(tree, "from import fred")
+
 
 class CompileTestCase(unittest.TestCase):
 
@@ -484,8 +517,18 @@
         st = parser.suite('a = "\\u1"')
         self.assertRaises(SyntaxError, parser.compilest, st)
 
+    def test_issue_9011(self):
+        # Issue 9011: compilation of an unary minus expression changed
+        # the meaning of the ST, so that a second compilation produced
+        # incorrect results.
+        st = parser.expr('-3')
+        code1 = parser.compilest(st)
+        self.assertEqual(eval(code1), -3)
+        code2 = parser.compilest(st)
+        self.assertEqual(eval(code2), -3)
+
 class ParserStackLimitTestCase(unittest.TestCase):
-    """try to push the parser to/over it's limits.
+    """try to push the parser to/over its limits.
     see http://bugs.python.org/issue1881 for a discussion
     """
     def _nested_expression(self, level):

Modified: python/branches/py3k-cdecimal/Lib/test/test_platform.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_platform.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_platform.py	Sat Jul 17 12:36:09 2010
@@ -10,20 +10,26 @@
     def test_architecture(self):
         res = platform.architecture()
 
-    if hasattr(os, "symlink"):
-        def test_architecture_via_symlink(self): # issue3762
-            def get(python):
-                cmd = [python, '-c',
-                    'import platform; print(platform.architecture())']
-                p = subprocess.Popen(cmd, stdout=subprocess.PIPE)
-                return p.communicate()
-            real = os.path.realpath(sys.executable)
-            link = os.path.abspath(support.TESTFN)
-            os.symlink(real, link)
-            try:
-                self.assertEqual(get(real), get(link))
-            finally:
-                os.remove(link)
+    @support.skip_unless_symlink
+    def test_architecture_via_symlink(self): # issue3762
+        # On Windows, the EXE needs to know where pythonXY.dll is at so we have
+        # to add the directory to the path.
+        if sys.platform == "win32":
+            os.environ["Path"] = "{};{}".format(
+                os.path.dirname(sys.executable), os.environ["Path"])
+
+        def get(python):
+            cmd = [python, '-c',
+                'import platform; print(platform.architecture())']
+            p = subprocess.Popen(cmd, stdout=subprocess.PIPE)
+            return p.communicate()
+        real = os.path.realpath(sys.executable)
+        link = os.path.abspath(support.TESTFN)
+        os.symlink(real, link)
+        try:
+            self.assertEqual(get(real), get(link))
+        finally:
+            os.remove(link)
 
     def test_platform(self):
         for aliased in (False, True):

Modified: python/branches/py3k-cdecimal/Lib/test/test_posixpath.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_posixpath.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_posixpath.py	Sat Jul 17 12:36:09 2010
@@ -1,7 +1,9 @@
 import unittest
 from test import support, test_genericpath
 
-import posixpath, os
+import posixpath
+import os
+import sys
 from posixpath import realpath, abspath, dirname, basename
 
 # An absolute path to a temporary filename for testing. We can't rely on TESTFN
@@ -9,6 +11,16 @@
 
 ABSTFN = abspath(support.TESTFN)
 
+def skip_if_ABSTFN_contains_backslash(test):
+    """
+    On Windows, posixpath.abspath still returns paths with backslashes
+    instead of posix forward slashes. If this is the case, several tests
+    fail, so skip them.
+    """
+    found_backslash = '\\' in ABSTFN
+    msg = "ABSTFN is not a posix path - tests fail"
+    return [test, unittest.skip(msg)(test)][found_backslash]
+
 def safe_rmdir(dirname):
     try:
         os.rmdir(dirname)
@@ -143,7 +155,7 @@
             f.write(b"foo")
             f.close()
             self.assertIs(posixpath.islink(support.TESTFN + "1"), False)
-            if hasattr(os, "symlink"):
+            if support.can_symlink():
                 os.symlink(support.TESTFN + "1", support.TESTFN + "2")
                 self.assertIs(posixpath.islink(support.TESTFN + "2"), True)
                 os.remove(support.TESTFN + "1")
@@ -154,85 +166,60 @@
             if not f.close():
                 f.close()
 
+    @staticmethod
+    def _create_file(filename):
+        with open(filename, 'wb') as f:
+            f.write(b'foo')
+
     def test_samefile(self):
-        f = open(support.TESTFN + "1", "wb")
-        try:
-            f.write(b"foo")
-            f.close()
-            self.assertIs(
-                posixpath.samefile(
-                    support.TESTFN + "1",
-                    support.TESTFN + "1"
-                ),
-                True
-            )
-            # If we don't have links, assume that os.stat doesn't return resonable
-            # inode information and thus, that samefile() doesn't work
-            if hasattr(os, "symlink"):
-                os.symlink(
-                    support.TESTFN + "1",
-                    support.TESTFN + "2"
-                )
-                self.assertIs(
-                    posixpath.samefile(
-                        support.TESTFN + "1",
-                        support.TESTFN + "2"
-                    ),
-                    True
-                )
-                os.remove(support.TESTFN + "2")
-                f = open(support.TESTFN + "2", "wb")
-                f.write(b"bar")
-                f.close()
-                self.assertIs(
-                    posixpath.samefile(
-                        support.TESTFN + "1",
-                        support.TESTFN + "2"
-                    ),
-                    False
-                )
-        finally:
-            if not f.close():
-                f.close()
+        test_fn = support.TESTFN + "1"
+        self._create_file(test_fn)
+        self.assertTrue(posixpath.samefile(test_fn, test_fn))
+        self.assertRaises(TypeError, posixpath.samefile)
+
+    @unittest.skipIf(
+        sys.platform.startswith('win'),
+        "posixpath.samefile does not work on links in Windows")
+    @support.skip_unless_symlink
+    def test_samefile_on_links(self):
+        test_fn1 = support.TESTFN + "1"
+        test_fn2 = support.TESTFN + "2"
+        self._create_file(test_fn1)
+
+        os.symlink(test_fn1, test_fn2)
+        self.assertTrue(posixpath.samefile(test_fn1, test_fn2))
+        os.remove(test_fn2)
+
+        self._create_file(test_fn2)
+        self.assertFalse(posixpath.samefile(test_fn1, test_fn2))
+
 
     def test_samestat(self):
-        f = open(support.TESTFN + "1", "wb")
-        try:
-            f.write(b"foo")
-            f.close()
-            self.assertIs(
-                posixpath.samestat(
-                    os.stat(support.TESTFN + "1"),
-                    os.stat(support.TESTFN + "1")
-                ),
-                True
-            )
-            # If we don't have links, assume that os.stat() doesn't return resonable
-            # inode information and thus, that samefile() doesn't work
-            if hasattr(os, "symlink"):
-                if hasattr(os, "symlink"):
-                    os.symlink(support.TESTFN + "1", support.TESTFN + "2")
-                    self.assertIs(
-                        posixpath.samestat(
-                            os.stat(support.TESTFN + "1"),
-                            os.stat(support.TESTFN + "2")
-                        ),
-                        True
-                    )
-                    os.remove(support.TESTFN + "2")
-                f = open(support.TESTFN + "2", "wb")
-                f.write(b"bar")
-                f.close()
-                self.assertIs(
-                    posixpath.samestat(
-                        os.stat(support.TESTFN + "1"),
-                        os.stat(support.TESTFN + "2")
-                    ),
-                    False
-                )
-        finally:
-            if not f.close():
-                f.close()
+        test_fn = support.TESTFN + "1"
+        self._create_file(test_fn)
+        test_fns = [test_fn]*2
+        stats = map(os.stat, test_fns)
+        self.assertTrue(posixpath.samestat(*stats))
+
+    @unittest.skipIf(
+        sys.platform.startswith('win'),
+        "posixpath.samestat does not work on links in Windows")
+    @support.skip_unless_symlink
+    def test_samestat_on_links(self):
+        test_fn1 = support.TESTFN + "1"
+        test_fn2 = support.TESTFN + "2"
+        self._create_file(test_fn1)
+        test_fns = (test_fn1, test_fn2)
+        os.symlink(*test_fns)
+        stats = map(os.stat, test_fns)
+        self.assertTrue(posixpath.samestat(*stats))
+        os.remove(test_fn2)
+
+        self._create_file(test_fn2)
+        stats = map(os.stat, test_fns)
+        self.assertFalse(posixpath.samestat(*stats))
+
+        self.assertRaises(TypeError, posixpath.samestat)
 
     def test_ismount(self):
         self.assertIs(posixpath.ismount("/"), True)
@@ -286,103 +273,112 @@
         self.assertEqual(posixpath.normpath(b"///..//./foo/.//bar"),
                          b"/foo/bar")
 
-    if hasattr(os, "symlink"):
-        def test_realpath_basic(self):
-            # Basic operation.
-            try:
-                os.symlink(ABSTFN+"1", ABSTFN)
-                self.assertEqual(realpath(ABSTFN), ABSTFN+"1")
-            finally:
-                support.unlink(ABSTFN)
-
-        def test_realpath_symlink_loops(self):
-            # Bug #930024, return the path unchanged if we get into an infinite
-            # symlink loop.
-            try:
-                old_path = abspath('.')
-                os.symlink(ABSTFN, ABSTFN)
-                self.assertEqual(realpath(ABSTFN), ABSTFN)
-
-                os.symlink(ABSTFN+"1", ABSTFN+"2")
-                os.symlink(ABSTFN+"2", ABSTFN+"1")
-                self.assertEqual(realpath(ABSTFN+"1"), ABSTFN+"1")
-                self.assertEqual(realpath(ABSTFN+"2"), ABSTFN+"2")
-
-                # Test using relative path as well.
-                os.chdir(dirname(ABSTFN))
-                self.assertEqual(realpath(basename(ABSTFN)), ABSTFN)
-            finally:
-                os.chdir(old_path)
-                support.unlink(ABSTFN)
-                support.unlink(ABSTFN+"1")
-                support.unlink(ABSTFN+"2")
-
-        def test_realpath_resolve_parents(self):
-            # We also need to resolve any symlinks in the parents of a relative
-            # path passed to realpath. E.g.: current working directory is
-            # /usr/doc with 'doc' being a symlink to /usr/share/doc. We call
-            # realpath("a"). This should return /usr/share/doc/a/.
-            try:
-                old_path = abspath('.')
-                os.mkdir(ABSTFN)
-                os.mkdir(ABSTFN + "/y")
-                os.symlink(ABSTFN + "/y", ABSTFN + "/k")
-
-                os.chdir(ABSTFN + "/k")
-                self.assertEqual(realpath("a"), ABSTFN + "/y/a")
-            finally:
-                os.chdir(old_path)
-                support.unlink(ABSTFN + "/k")
-                safe_rmdir(ABSTFN + "/y")
-                safe_rmdir(ABSTFN)
-
-        def test_realpath_resolve_before_normalizing(self):
-            # Bug #990669: Symbolic links should be resolved before we
-            # normalize the path. E.g.: if we have directories 'a', 'k' and 'y'
-            # in the following hierarchy:
-            # a/k/y
-            #
-            # and a symbolic link 'link-y' pointing to 'y' in directory 'a',
-            # then realpath("link-y/..") should return 'k', not 'a'.
-            try:
-                old_path = abspath('.')
-                os.mkdir(ABSTFN)
-                os.mkdir(ABSTFN + "/k")
-                os.mkdir(ABSTFN + "/k/y")
-                os.symlink(ABSTFN + "/k/y", ABSTFN + "/link-y")
-
-                # Absolute path.
-                self.assertEqual(realpath(ABSTFN + "/link-y/.."), ABSTFN + "/k")
-                # Relative path.
-                os.chdir(dirname(ABSTFN))
-                self.assertEqual(realpath(basename(ABSTFN) + "/link-y/.."),
-                                 ABSTFN + "/k")
-            finally:
-                os.chdir(old_path)
-                support.unlink(ABSTFN + "/link-y")
-                safe_rmdir(ABSTFN + "/k/y")
-                safe_rmdir(ABSTFN + "/k")
-                safe_rmdir(ABSTFN)
-
-        def test_realpath_resolve_first(self):
-            # Bug #1213894: The first component of the path, if not absolute,
-            # must be resolved too.
-
-            try:
-                old_path = abspath('.')
-                os.mkdir(ABSTFN)
-                os.mkdir(ABSTFN + "/k")
-                os.symlink(ABSTFN, ABSTFN + "link")
-                os.chdir(dirname(ABSTFN))
-
-                base = basename(ABSTFN)
-                self.assertEqual(realpath(base + "link"), ABSTFN)
-                self.assertEqual(realpath(base + "link/k"), ABSTFN + "/k")
-            finally:
-                os.chdir(old_path)
-                support.unlink(ABSTFN + "link")
-                safe_rmdir(ABSTFN + "/k")
-                safe_rmdir(ABSTFN)
+    @support.skip_unless_symlink
+    @skip_if_ABSTFN_contains_backslash
+    def test_realpath_basic(self):
+        # Basic operation.
+        try:
+            os.symlink(ABSTFN+"1", ABSTFN)
+            self.assertEqual(realpath(ABSTFN), ABSTFN+"1")
+        finally:
+            support.unlink(ABSTFN)
+
+    @support.skip_unless_symlink
+    @skip_if_ABSTFN_contains_backslash
+    def test_realpath_symlink_loops(self):
+        # Bug #930024, return the path unchanged if we get into an infinite
+        # symlink loop.
+        try:
+            old_path = abspath('.')
+            os.symlink(ABSTFN, ABSTFN)
+            self.assertEqual(realpath(ABSTFN), ABSTFN)
+
+            os.symlink(ABSTFN+"1", ABSTFN+"2")
+            os.symlink(ABSTFN+"2", ABSTFN+"1")
+            self.assertEqual(realpath(ABSTFN+"1"), ABSTFN+"1")
+            self.assertEqual(realpath(ABSTFN+"2"), ABSTFN+"2")
+
+            # Test using relative path as well.
+            os.chdir(dirname(ABSTFN))
+            self.assertEqual(realpath(basename(ABSTFN)), ABSTFN)
+        finally:
+            os.chdir(old_path)
+            support.unlink(ABSTFN)
+            support.unlink(ABSTFN+"1")
+            support.unlink(ABSTFN+"2")
+
+    @support.skip_unless_symlink
+    @skip_if_ABSTFN_contains_backslash
+    def test_realpath_resolve_parents(self):
+        # We also need to resolve any symlinks in the parents of a relative
+        # path passed to realpath. E.g.: current working directory is
+        # /usr/doc with 'doc' being a symlink to /usr/share/doc. We call
+        # realpath("a"). This should return /usr/share/doc/a/.
+        try:
+            old_path = abspath('.')
+            os.mkdir(ABSTFN)
+            os.mkdir(ABSTFN + "/y")
+            os.symlink(ABSTFN + "/y", ABSTFN + "/k")
+
+            os.chdir(ABSTFN + "/k")
+            self.assertEqual(realpath("a"), ABSTFN + "/y/a")
+        finally:
+            os.chdir(old_path)
+            support.unlink(ABSTFN + "/k")
+            safe_rmdir(ABSTFN + "/y")
+            safe_rmdir(ABSTFN)
+
+    @support.skip_unless_symlink
+    @skip_if_ABSTFN_contains_backslash
+    def test_realpath_resolve_before_normalizing(self):
+        # Bug #990669: Symbolic links should be resolved before we
+        # normalize the path. E.g.: if we have directories 'a', 'k' and 'y'
+        # in the following hierarchy:
+        # a/k/y
+        #
+        # and a symbolic link 'link-y' pointing to 'y' in directory 'a',
+        # then realpath("link-y/..") should return 'k', not 'a'.
+        try:
+            old_path = abspath('.')
+            os.mkdir(ABSTFN)
+            os.mkdir(ABSTFN + "/k")
+            os.mkdir(ABSTFN + "/k/y")
+            os.symlink(ABSTFN + "/k/y", ABSTFN + "/link-y")
+
+            # Absolute path.
+            self.assertEqual(realpath(ABSTFN + "/link-y/.."), ABSTFN + "/k")
+            # Relative path.
+            os.chdir(dirname(ABSTFN))
+            self.assertEqual(realpath(basename(ABSTFN) + "/link-y/.."),
+                             ABSTFN + "/k")
+        finally:
+            os.chdir(old_path)
+            support.unlink(ABSTFN + "/link-y")
+            safe_rmdir(ABSTFN + "/k/y")
+            safe_rmdir(ABSTFN + "/k")
+            safe_rmdir(ABSTFN)
+
+    @support.skip_unless_symlink
+    @skip_if_ABSTFN_contains_backslash
+    def test_realpath_resolve_first(self):
+        # Bug #1213894: The first component of the path, if not absolute,
+        # must be resolved too.
+
+        try:
+            old_path = abspath('.')
+            os.mkdir(ABSTFN)
+            os.mkdir(ABSTFN + "/k")
+            os.symlink(ABSTFN, ABSTFN + "link")
+            os.chdir(dirname(ABSTFN))
+
+            base = basename(ABSTFN)
+            self.assertEqual(realpath(base + "link"), ABSTFN)
+            self.assertEqual(realpath(base + "link/k"), ABSTFN + "/k")
+        finally:
+            os.chdir(old_path)
+            support.unlink(ABSTFN + "link")
+            safe_rmdir(ABSTFN + "/k")
+            safe_rmdir(ABSTFN)
 
     def test_relpath(self):
         (real_getcwd, os.getcwd) = (os.getcwd, lambda: r"/home/user/bar")

Modified: python/branches/py3k-cdecimal/Lib/test/test_property.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_property.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_property.py	Sat Jul 17 12:36:09 2010
@@ -118,7 +118,7 @@
         self.assertEqual(base.__class__.spam.__doc__, "spam spam spam")
         self.assertEqual(sub.__class__.spam.__doc__, "spam spam spam")
 
-    @unittest.skipIf(sys.flags.optimize >= 1,
+    @unittest.skipIf(sys.flags.optimize >= 2,
                      "Docstrings are omitted with -O2 and above")
     def test_property_getter_doc_override(self):
         newgettersub = PropertySubNewGetter()

Modified: python/branches/py3k-cdecimal/Lib/test/test_pyexpat.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_pyexpat.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_pyexpat.py	Sat Jul 17 12:36:09 2010
@@ -184,7 +184,7 @@
             self.fail()
         except TypeError as e:
             self.assertEquals(str(e),
-                'ParserCreate() argument 2 must be string or None, not int')
+                'ParserCreate() argument 2 must be str or None, not int')
 
         try:
             expat.ParserCreate(namespace_separator='too long')

Modified: python/branches/py3k-cdecimal/Lib/test/test_runpy.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_runpy.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_runpy.py	Sat Jul 17 12:36:09 2010
@@ -6,7 +6,7 @@
 import re
 import tempfile
 import py_compile
-from test.support import forget, make_legacy_pyc, run_unittest, verbose
+from test.support import forget, make_legacy_pyc, run_unittest, unload, verbose
 from test.script_helper import (
     make_pkg, make_script, make_zip_pkg, make_zip_script, temp_dir)
 
@@ -174,6 +174,7 @@
             __import__(mod_name)
             os.remove(mod_fname)
             make_legacy_pyc(mod_fname)
+            unload(mod_name)  # In case loader caches paths
             if verbose: print("Running from compiled:", mod_name)
             d2 = run_module(mod_name) # Read from bytecode
             self.assertIn("x", d2)
@@ -197,6 +198,7 @@
             __import__(mod_name)
             os.remove(mod_fname)
             make_legacy_pyc(mod_fname)
+            unload(mod_name)  # In case loader caches paths
             if verbose: print("Running from compiled:", pkg_name)
             d2 = run_module(pkg_name) # Read from bytecode
             self.assertIn("x", d2)
@@ -252,6 +254,7 @@
             __import__(mod_name)
             os.remove(mod_fname)
             make_legacy_pyc(mod_fname)
+            unload(mod_name)  # In case the loader caches paths
             if verbose: print("Running from compiled:", mod_name)
             d2 = run_module(mod_name, run_name=run_name) # Read from bytecode
             self.assertIn("__package__", d2)
@@ -405,7 +408,11 @@
 
 
 def test_main():
-    run_unittest(RunModuleCodeTest, RunModuleTest, RunPathTest)
+    run_unittest(
+                 RunModuleCodeTest,
+                 RunModuleTest,
+                 RunPathTest
+                 )
 
 if __name__ == "__main__":
     test_main()

Modified: python/branches/py3k-cdecimal/Lib/test/test_shutil.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_shutil.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_shutil.py	Sat Jul 17 12:36:09 2010
@@ -271,7 +271,7 @@
             shutil.rmtree(src_dir)
             shutil.rmtree(os.path.dirname(dst_dir))
 
-    @unittest.skipUnless(hasattr(os, 'symlink'), 'requires os.symlink')
+    @support.skip_unless_symlink
     def test_dont_copy_file_onto_link_to_itself(self):
         # bug 851123.
         os.mkdir(TESTFN)
@@ -282,10 +282,11 @@
             f.write('cheddar')
             f.close()
 
-            os.link(src, dst)
-            self.assertRaises(shutil.Error, shutil.copyfile, src, dst)
-            self.assertEqual(open(src,'r').read(), 'cheddar')
-            os.remove(dst)
+            if hasattr(os, "link"):
+                os.link(src, dst)
+                self.assertRaises(shutil.Error, shutil.copyfile, src, dst)
+                self.assertEqual(open(src,'r').read(), 'cheddar')
+                os.remove(dst)
 
             # Using `src` here would mean we end up with a symlink pointing
             # to TESTFN/TESTFN/cheese, while it should point at
@@ -300,30 +301,30 @@
             except OSError:
                 pass
 
-        @unittest.skipUnless(hasattr(os, 'symlink'), 'requires os.symlink')
-        def test_rmtree_on_symlink(self):
-            # bug 1669.
-            os.mkdir(TESTFN)
+    @support.skip_unless_symlink
+    def test_rmtree_on_symlink(self):
+        # bug 1669.
+        os.mkdir(TESTFN)
+        try:
+            src = os.path.join(TESTFN, 'cheese')
+            dst = os.path.join(TESTFN, 'shop')
+            os.mkdir(src)
+            os.symlink(src, dst)
+            self.assertRaises(OSError, shutil.rmtree, dst)
+        finally:
+            shutil.rmtree(TESTFN, ignore_errors=True)
+
+    if hasattr(os, "mkfifo"):
+        # Issue #3002: copyfile and copytree block indefinitely on named pipes
+        def test_copyfile_named_pipe(self):
+            os.mkfifo(TESTFN)
             try:
-                src = os.path.join(TESTFN, 'cheese')
-                dst = os.path.join(TESTFN, 'shop')
-                os.mkdir(src)
-                os.symlink(src, dst)
-                self.assertRaises(OSError, shutil.rmtree, dst)
+                self.assertRaises(shutil.SpecialFileError,
+                                  shutil.copyfile, TESTFN, TESTFN2)
+                self.assertRaises(shutil.SpecialFileError,
+                                  shutil.copyfile, __file__, TESTFN)
             finally:
-                shutil.rmtree(TESTFN, ignore_errors=True)
-
-    @unittest.skipUnless(hasattr(os, 'mkfifo'), 'requires os.mkfifo')
-    # Issue #3002: copyfile and copytree block indefinitely on named pipes
-    def test_copyfile_named_pipe(self):
-        os.mkfifo(TESTFN)
-        try:
-            self.assertRaises(shutil.SpecialFileError,
-                                shutil.copyfile, TESTFN, TESTFN2)
-            self.assertRaises(shutil.SpecialFileError,
-                                shutil.copyfile, __file__, TESTFN)
-        finally:
-            os.remove(TESTFN)
+                os.remove(TESTFN)
 
     @unittest.skipUnless(hasattr(os, 'mkfifo'), 'requires os.mkfifo')
     def test_copytree_named_pipe(self):
@@ -361,7 +362,7 @@
         shutil.copytree(src_dir, dst_dir, copy_function=_copy)
         self.assertEquals(len(copied), 2)
 
-    @unittest.skipUnless(hasattr(os, 'symlink'), 'requires os.symlink')
+    @support.skip_unless_symlink
     def test_copytree_dangling_symlinks(self):
 
         # a dangling symlink raises an error at the end

Modified: python/branches/py3k-cdecimal/Lib/test/test_strtod.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_strtod.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_strtod.py	Sat Jul 17 12:36:09 2010
@@ -23,6 +23,8 @@
     \Z
 """, re.VERBOSE | re.IGNORECASE).match
 
+# Pure Python version of correctly rounded string->float conversion.
+# Avoids any use of floating-point by returning the result as a hex string.
 def strtod(s, mant_dig=53, min_exp = -1021, max_exp = 1024):
     """Convert a finite decimal string to a hex string representing an
     IEEE 754 binary64 float.  Return 'inf' or '-inf' on overflow.
@@ -259,6 +261,10 @@
             '18487398785991994634182916638542680759613590482273e-357',
             '32002864200581033134358724675198044527469366773928e-358',
             '94393431193180696942841837085033647913224148539854e-358',
+            '73608278998966969345824653500136787876436005957953e-358',
+            '64774478836417299491718435234611299336288082136054e-358',
+            '13704940134126574534878641876947980878824688451169e-357',
+            '46697445774047060960624497964425416610480524760471e-358',
             # failing case for bug introduced by METD in r77451 (attempted
             # fix for issue 7632, bug 2), and fixed in r77482.
             '28639097178261763178489759107321392745108491825303e-311',

Modified: python/branches/py3k-cdecimal/Lib/test/test_struct.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_struct.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_struct.py	Sat Jul 17 12:36:09 2010
@@ -47,8 +47,8 @@
 
         self.assertRaises(struct.error, struct.pack, 'iii', 3)
         self.assertRaises(struct.error, struct.pack, 'i', 3, 3, 3)
-        self.assertRaises(struct.error, struct.pack, 'i', 'foo')
-        self.assertRaises(struct.error, struct.pack, 'P', 'foo')
+        self.assertRaises((TypeError, struct.error), struct.pack, 'i', 'foo')
+        self.assertRaises((TypeError, struct.error), struct.pack, 'P', 'foo')
         self.assertRaises(struct.error, struct.unpack, 'd', b'flap')
         s = struct.pack('ii', 1, 2)
         self.assertRaises(struct.error, struct.unpack, 'iii', s)
@@ -234,7 +234,8 @@
                                                                  b'\x01' + got)
                 else:
                     # x is out of range -- verify pack realizes that.
-                    self.assertRaises(struct.error, pack, format, x)
+                    self.assertRaises((OverflowError, ValueError, struct.error),
+                                      pack, format, x)
 
             def run(self):
                 from random import randrange
@@ -438,12 +439,15 @@
 
         # Go beyond boundaries.
         small_buf = array.array('b', b' '*10)
-        self.assertRaises(struct.error, s.pack_into, small_buf, 0, test_string)
-        self.assertRaises(struct.error, s.pack_into, small_buf, 2, test_string)
+        self.assertRaises((ValueError, struct.error), s.pack_into, small_buf, 0,
+                          test_string)
+        self.assertRaises((ValueError, struct.error), s.pack_into, small_buf, 2,
+                          test_string)
 
         # Test bogus offset (issue 3694)
         sb = small_buf
-        self.assertRaises(TypeError, struct.pack_into, b'', sb, None)
+        self.assertRaises((TypeError, struct.error), struct.pack_into, b'', sb,
+                          None)
 
     def test_pack_into_fn(self):
         test_string = b'Reykjavik rocks, eow!'
@@ -463,8 +467,10 @@
 
         # Go beyond boundaries.
         small_buf = array.array('b', b' '*10)
-        self.assertRaises(struct.error, pack_into, small_buf, 0, test_string)
-        self.assertRaises(struct.error, pack_into, small_buf, 2, test_string)
+        self.assertRaises((ValueError, struct.error), pack_into, small_buf, 0,
+                          test_string)
+        self.assertRaises((ValueError, struct.error), pack_into, small_buf, 2,
+                          test_string)
 
     def test_unpack_with_buffer(self):
         # SF bug 1563759: struct.unpack doens't support buffer protocol objects
@@ -475,6 +481,9 @@
             self.assertEqual(value, 0x12345678)
 
     def test_bool(self):
+        class ExplodingBool(object):
+            def __bool__(self):
+                raise IOError
         for prefix in tuple("<>!=")+('',):
             false = (), [], [], '', 0
             true = [1], 'test', 5, -1, 0xffffffff+1, 0xffffffff/2
@@ -503,8 +512,16 @@
                 self.assertFalse(prefix, msg='encoded bool is not one byte: %r'
                                              %packed)
 
-            for c in [b'\x01', b'\x7f', b'\xff', b'\x0f', b'\xf0']:
-                self.assertTrue(struct.unpack('>?', c)[0])
+            try:
+                struct.pack(prefix + '?', ExplodingBool())
+            except IOError:
+                pass
+            else:
+                self.fail("Expected IOError: struct.pack(%r, "
+                          "ExplodingBool())" % (prefix + '?'))
+
+        for c in [b'\x01', b'\x7f', b'\xff', b'\x0f', b'\xf0']:
+            self.assertTrue(struct.unpack('>?', c)[0])
 
     def test_count_overflow(self):
         hugecount = '{}b'.format(sys.maxsize+1)

Modified: python/branches/py3k-cdecimal/Lib/test/test_structseq.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_structseq.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_structseq.py	Sat Jul 17 12:36:09 2010
@@ -1,12 +1,14 @@
+import os
+import time
 import unittest
 from test import support
 
-import time
 
 class StructSeqTest(unittest.TestCase):
 
     def test_tuple(self):
         t = time.gmtime()
+        self.assertIsInstance(t, tuple)
         astuple = tuple(t)
         self.assertEqual(len(t), len(astuple))
         self.assertEqual(t, astuple)
@@ -33,6 +35,13 @@
         self.assertEqual(repr(t),
             "time.struct_time(tm_year=1970, tm_mon=1, tm_mday=1, tm_hour=0, "
             "tm_min=0, tm_sec=0, tm_wday=3, tm_yday=1, tm_isdst=0)")
+        # os.stat() gives a complicated struct sequence.
+        st = os.stat(__file__)
+        rep = repr(st)
+        self.assertTrue(rep.startswith(os.name + ".stat_result"))
+        self.assertIn("st_mode=", rep)
+        self.assertIn("st_ino=", rep)
+        self.assertIn("st_dev=", rep)
 
     def test_concat(self):
         t1 = time.gmtime()

Modified: python/branches/py3k-cdecimal/Lib/test/test_subprocess.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_subprocess.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_subprocess.py	Sat Jul 17 12:36:09 2010
@@ -478,8 +478,6 @@
                          '"a\\\\b\\ c" d e')
         self.assertEqual(subprocess.list2cmdline(['ab', '']),
                          'ab ""')
-        self.assertEqual(subprocess.list2cmdline(['echo', 'foo|bar']),
-                         'echo "foo|bar"')
 
 
     def test_poll(self):

Modified: python/branches/py3k-cdecimal/Lib/test/test_syntax.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_syntax.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_syntax.py	Sat Jul 17 12:36:09 2010
@@ -478,10 +478,12 @@
 
 >>> del ()
 Traceback (most recent call last):
-   ...
-  File "<doctest test.test_syntax[50]>", line 1
 SyntaxError: can't delete ()
 
+>>> {1, 2, 3} = 42
+Traceback (most recent call last):
+SyntaxError: can't assign to literal
+
 """
 
 import re

Modified: python/branches/py3k-cdecimal/Lib/test/test_sysconfig.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_sysconfig.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_sysconfig.py	Sat Jul 17 12:36:09 2010
@@ -12,7 +12,7 @@
 from copy import copy, deepcopy
 
 from test.support import (run_unittest, TESTFN, unlink, get_attribute,
-                          captured_stdout)
+                          captured_stdout, skip_unless_symlink)
 
 import sysconfig
 from sysconfig import (get_paths, get_platform, get_config_vars,
@@ -239,17 +239,23 @@
                   'posix_home', 'posix_prefix', 'posix_user')
         self.assertEquals(get_scheme_names(), wanted)
 
+    @skip_unless_symlink
     def test_symlink(self):
+        # On Windows, the EXE needs to know where pythonXY.dll is at so we have
+        # to add the directory to the path.
+        if sys.platform == "win32":
+            os.environ["Path"] = "{};{}".format(
+                os.path.dirname(sys.executable), os.environ["Path"])
+
         # Issue 7880
-        symlink = get_attribute(os, "symlink")
         def get(python):
             cmd = [python, '-c',
                    'import sysconfig; print(sysconfig.get_platform())']
-            p = subprocess.Popen(cmd, stdout=subprocess.PIPE)
+            p = subprocess.Popen(cmd, stdout=subprocess.PIPE, env=os.environ)
             return p.communicate()
         real = os.path.realpath(sys.executable)
         link = os.path.abspath(TESTFN)
-        symlink(real, link)
+        os.symlink(real, link)
         try:
             self.assertEqual(get(real), get(link))
         finally:

Modified: python/branches/py3k-cdecimal/Lib/test/test_tarfile.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_tarfile.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_tarfile.py	Sat Jul 17 12:36:09 2010
@@ -291,6 +291,9 @@
         self.assertTrue(self.tar.getmembers()[-1].name == "misc/eof",
                 "could not find all members")
 
+    @unittest.skipUnless(hasattr(os, "link"),
+                         "Missing hardlink implementation")
+    @support.skip_unless_symlink
     def test_extract_hardlink(self):
         # Test hardlink extraction (e.g. bug #857297).
         tar = tarfile.open(tarname, errorlevel=1, encoding="iso8859-1")
@@ -695,16 +698,16 @@
                 os.remove(target)
                 os.remove(link)
 
+    @support.skip_unless_symlink
     def test_symlink_size(self):
-        if hasattr(os, "symlink"):
-            path = os.path.join(TEMPDIR, "symlink")
-            os.symlink("link_target", path)
-            try:
-                tar = tarfile.open(tmpname, self.mode)
-                tarinfo = tar.gettarinfo(path)
-                self.assertEqual(tarinfo.size, 0)
-            finally:
-                os.remove(path)
+        path = os.path.join(TEMPDIR, "symlink")
+        os.symlink("link_target", path)
+        try:
+            tar = tarfile.open(tmpname, self.mode)
+            tarinfo = tar.gettarinfo(path)
+            self.assertEqual(tarinfo.size, 0)
+        finally:
+            os.remove(path)
 
     def test_add_self(self):
         # Test for #1257255.
@@ -1408,15 +1411,26 @@
         data = open(os.path.join(TEMPDIR, name), "rb").read()
         self.assertEqual(md5sum(data), md5_regtype)
 
+    # When 8879 gets fixed, this will need to change. Currently on Windows
+    # we have os.path.islink but no os.link, so these tests fail without the
+    # following skip until link is completed.
+    @unittest.skipIf(hasattr(os.path, "islink"),
+                     "Skip emulation - has os.path.islink but not os.link")
     def test_hardlink_extraction1(self):
         self._test_link_extraction("ustar/lnktype")
 
+    @unittest.skipIf(hasattr(os.path, "islink"),
+                     "Skip emulation - has os.path.islink but not os.link")
     def test_hardlink_extraction2(self):
         self._test_link_extraction("./ustar/linktest2/lnktype")
 
+    @unittest.skipIf(hasattr(os, "symlink"),
+                     "Skip emulation if symlink exists")
     def test_symlink_extraction1(self):
         self._test_link_extraction("ustar/symtype")
 
+    @unittest.skipIf(hasattr(os, "symlink"),
+                     "Skip emulation if symlink exists")
     def test_symlink_extraction2(self):
         self._test_link_extraction("./ustar/linktest2/symtype")
 

Modified: python/branches/py3k-cdecimal/Lib/test/test_threaded_import.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_threaded_import.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_threaded_import.py	Sat Jul 17 12:36:09 2010
@@ -5,72 +5,77 @@
 # complains several times about module random having no attribute
 # randrange, and then Python hangs.
 
+import imp
+import sys
 import unittest
-from test.support import verbose, TestFailed, import_module
+from test.support import verbose, TestFailed, import_module, run_unittest
 thread = import_module('_thread')
 
-critical_section = thread.allocate_lock()
-done = thread.allocate_lock()
+def task(N, done, done_tasks, errors):
+    try:
+        import random
+        # This will fail if random is not completely initialized
+        x = random.randrange(1, 3)
+    except Exception as e:
+        errors.append(e.with_traceback(None))
+    finally:
+        done_tasks.append(thread.get_ident())
+        finished = len(done_tasks) == N
+        if finished:
+            done.release()
+
+
+class ThreadedImportTests(unittest.TestCase):
+
+    def setUp(self):
+        self.old_random = sys.modules.pop('random', None)
+
+    def tearDown(self):
+        # If the `random` module was already initialized, we restore the
+        # old module at the end so that pickling tests don't fail.
+        # See http://bugs.python.org/issue3657#msg110461
+        if self.old_random is not None:
+            sys.modules['random'] = self.old_random
+
+    def test_parallel_module_init(self):
+        if imp.lock_held():
+            # This triggers on, e.g., from test import autotest.
+            raise unittest.SkipTest("can't run when import lock is held")
 
-def task():
-    global N, critical_section, done
-    import random
-    x = random.randrange(1, 3)
-    critical_section.acquire()
-    N -= 1
-    # Must release critical_section before releasing done, else the main
-    # thread can exit and set critical_section to None as part of global
-    # teardown; then critical_section.release() raises AttributeError.
-    finished = N == 0
-    critical_section.release()
-    if finished:
+        done = thread.allocate_lock()
+        done.acquire()
+        for N in (20, 50) * 3:
+            if verbose:
+                print("Trying", N, "threads ...", end=' ')
+            # Make sure that random gets reimported freshly
+            try:
+                del sys.modules['random']
+            except KeyError:
+                pass
+            errors = []
+            done_tasks = []
+            for i in range(N):
+                thread.start_new_thread(task, (N, done, done_tasks, errors,))
+            done.acquire()
+            self.assertFalse(errors)
+            if verbose:
+                print("OK.")
         done.release()
 
-def test_import_hangers():
-    import sys
-    if verbose:
-        print("testing import hangers ...", end=' ')
-
-    import test.threaded_import_hangers
-    try:
-        if test.threaded_import_hangers.errors:
-            raise TestFailed(test.threaded_import_hangers.errors)
-        elif verbose:
-            print("OK.")
-    finally:
+    def test_import_hangers(self):
         # In case this test is run again, make sure the helper module
         # gets loaded from scratch again.
-        del sys.modules['test.threaded_import_hangers']
+        try:
+            del sys.modules['test.threaded_import_hangers']
+        except KeyError:
+            pass
+        import test.threaded_import_hangers
+        self.assertFalse(test.threaded_import_hangers.errors)
 
-# Tricky:  When regrtest imports this module, the thread running regrtest
-# grabs the import lock and won't let go of it until this module returns.
-# All other threads attempting an import hang for the duration.  Since
-# this test spawns threads that do little *but* import, we can't do that
-# successfully until after this module finishes importing and regrtest
-# regains control.  To make this work, a special case was added to
-# regrtest to invoke a module's "test_main" function (if any) after
-# importing it.
-
-def test_main():        # magic name!  see above
-    global N, done
-
-    import imp
-    if imp.lock_held():
-        # This triggers on, e.g., from test import autotest.
-        raise unittest.SkipTest("can't run when import lock is held")
-
-    done.acquire()
-    for N in (20, 50) * 3:
-        if verbose:
-            print("Trying", N, "threads ...", end=' ')
-        for i in range(N):
-            thread.start_new_thread(task, ())
-        done.acquire()
-        if verbose:
-            print("OK.")
-    done.release()
 
-    test_import_hangers()
+def test_main():
+    run_unittest(ThreadedImportTests)
+
 
 if __name__ == "__main__":
     test_main()

Modified: python/branches/py3k-cdecimal/Lib/test/test_tokenize.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_tokenize.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_tokenize.py	Sat Jul 17 12:36:09 2010
@@ -42,7 +42,7 @@
         ...
     IndentationError: unindent does not match any outer indentation level
 
-There are some standard formattig practises that are easy to get right.
+There are some standard formatting practices that are easy to get right.
 
     >>> roundtrip("if x == 1:\\n"
     ...           "    print(x)\\n")

Modified: python/branches/py3k-cdecimal/Lib/test/test_tuple.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_tuple.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_tuple.py	Sat Jul 17 12:36:09 2010
@@ -96,6 +96,7 @@
         gc.collect()
         self.assertTrue(gc.is_tracked(t), t)
 
+    @support.cpython_only
     def test_track_literals(self):
         # Test GC-optimization of tuple literals
         x, y, z = 1.5, "a", []
@@ -136,16 +137,19 @@
         self._tracked(tp(tuple([obj]) for obj in [x, y, z]))
         self._tracked(tuple(tp([obj]) for obj in [x, y, z]))
 
+    @support.cpython_only
     def test_track_dynamic(self):
         # Test GC-optimization of dynamically constructed tuples.
         self.check_track_dynamic(tuple, False)
 
+    @support.cpython_only
     def test_track_subtypes(self):
         # Tuple subtypes must always be tracked
         class MyTuple(tuple):
             pass
         self.check_track_dynamic(MyTuple, True)
 
+    @support.cpython_only
     def test_bug7466(self):
         # Trying to untrack an unfinished tuple could crash Python
         self._not_tracked(tuple(gc.collect() for i in range(101)))

Modified: python/branches/py3k-cdecimal/Lib/test/test_unicode.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_unicode.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_unicode.py	Sat Jul 17 12:36:09 2010
@@ -942,6 +942,159 @@
         # * strict decoding testing for all of the
         #   UTF8_ERROR cases in PyUnicode_DecodeUTF8
 
+    def test_utf8_decode_valid_sequences(self):
+        sequences = [
+            # single byte
+            (b'\x00', '\x00'), (b'a', 'a'), (b'\x7f', '\x7f'),
+            # 2 bytes
+            (b'\xc2\x80', '\x80'), (b'\xdf\xbf', '\u07ff'),
+            # 3 bytes
+            (b'\xe0\xa0\x80', '\u0800'), (b'\xed\x9f\xbf', '\ud7ff'),
+            (b'\xee\x80\x80', '\uE000'), (b'\xef\xbf\xbf', '\uffff'),
+            # 4 bytes
+            (b'\xF0\x90\x80\x80', '\U00010000'),
+            (b'\xf4\x8f\xbf\xbf', '\U0010FFFF')
+        ]
+        for seq, res in sequences:
+            self.assertEqual(seq.decode('utf-8'), res)
+
+
+    def test_utf8_decode_invalid_sequences(self):
+        # continuation bytes in a sequence of 2, 3, or 4 bytes
+        continuation_bytes = [bytes([x]) for x in range(0x80, 0xC0)]
+        # start bytes of a 2-byte sequence equivalent to codepoints < 0x7F
+        invalid_2B_seq_start_bytes = [bytes([x]) for x in range(0xC0, 0xC2)]
+        # start bytes of a 4-byte sequence equivalent to codepoints > 0x10FFFF
+        invalid_4B_seq_start_bytes = [bytes([x]) for x in range(0xF5, 0xF8)]
+        invalid_start_bytes = (
+            continuation_bytes + invalid_2B_seq_start_bytes +
+            invalid_4B_seq_start_bytes + [bytes([x]) for x in range(0xF7, 0x100)]
+        )
+
+        for byte in invalid_start_bytes:
+            self.assertRaises(UnicodeDecodeError, byte.decode, 'utf-8')
+
+        for sb in invalid_2B_seq_start_bytes:
+            for cb in continuation_bytes:
+                self.assertRaises(UnicodeDecodeError, (sb+cb).decode, 'utf-8')
+
+        for sb in invalid_4B_seq_start_bytes:
+            for cb1 in continuation_bytes[:3]:
+                for cb3 in continuation_bytes[:3]:
+                    self.assertRaises(UnicodeDecodeError,
+                                      (sb+cb1+b'\x80'+cb3).decode, 'utf-8')
+
+        for cb in [bytes([x]) for x in range(0x80, 0xA0)]:
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xE0'+cb+b'\x80').decode, 'utf-8')
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xE0'+cb+b'\xBF').decode, 'utf-8')
+        # surrogates
+        for cb in [bytes([x]) for x in range(0xA0, 0xC0)]:
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xED'+cb+b'\x80').decode, 'utf-8')
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xED'+cb+b'\xBF').decode, 'utf-8')
+        for cb in [bytes([x]) for x in range(0x80, 0x90)]:
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xF0'+cb+b'\x80\x80').decode, 'utf-8')
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xF0'+cb+b'\xBF\xBF').decode, 'utf-8')
+        for cb in [bytes([x]) for x in range(0x90, 0xC0)]:
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xF4'+cb+b'\x80\x80').decode, 'utf-8')
+            self.assertRaises(UnicodeDecodeError,
+                              (b'\xF4'+cb+b'\xBF\xBF').decode, 'utf-8')
+
+    def test_issue8271(self):
+        # Issue #8271: during the decoding of an invalid UTF-8 byte sequence,
+        # only the start byte and the continuation byte(s) are now considered
+        # invalid, instead of the number of bytes specified by the start byte.
+        # See http://www.unicode.org/versions/Unicode5.2.0/ch03.pdf (page 95,
+        # table 3-8, Row 2) for more information about the algorithm used.
+        FFFD = '\ufffd'
+        sequences = [
+            # invalid start bytes
+            (b'\x80', FFFD), # continuation byte
+            (b'\x80\x80', FFFD*2), # 2 continuation bytes
+            (b'\xc0', FFFD),
+            (b'\xc0\xc0', FFFD*2),
+            (b'\xc1', FFFD),
+            (b'\xc1\xc0', FFFD*2),
+            (b'\xc0\xc1', FFFD*2),
+            # with start byte of a 2-byte sequence
+            (b'\xc2', FFFD), # only the start byte
+            (b'\xc2\xc2', FFFD*2), # 2 start bytes
+            (b'\xc2\xc2\xc2', FFFD*3), # 2 start bytes
+            (b'\xc2\x41', FFFD+'A'), # invalid continuation byte
+            # with start byte of a 3-byte sequence
+            (b'\xe1', FFFD), # only the start byte
+            (b'\xe1\xe1', FFFD*2), # 2 start bytes
+            (b'\xe1\xe1\xe1', FFFD*3), # 3 start bytes
+            (b'\xe1\xe1\xe1\xe1', FFFD*4), # 4 start bytes
+            (b'\xe1\x80', FFFD), # only 1 continuation byte
+            (b'\xe1\x41', FFFD+'A'), # invalid continuation byte
+            (b'\xe1\x41\x80', FFFD+'A'+FFFD), # invalid cb followed by valid cb
+            (b'\xe1\x41\x41', FFFD+'AA'), # 2 invalid continuation bytes
+            (b'\xe1\x80\x41', FFFD+'A'), # only 1 valid continuation byte
+            (b'\xe1\x80\xe1\x41', FFFD*2+'A'), # 1 valid and the other invalid
+            (b'\xe1\x41\xe1\x80', FFFD+'A'+FFFD), # 1 invalid and the other valid
+            # with start byte of a 4-byte sequence
+            (b'\xf1', FFFD), # only the start byte
+            (b'\xf1\xf1', FFFD*2), # 2 start bytes
+            (b'\xf1\xf1\xf1', FFFD*3), # 3 start bytes
+            (b'\xf1\xf1\xf1\xf1', FFFD*4), # 4 start bytes
+            (b'\xf1\xf1\xf1\xf1\xf1', FFFD*5), # 5 start bytes
+            (b'\xf1\x80', FFFD), # only 1 continuation bytes
+            (b'\xf1\x80\x80', FFFD), # only 2 continuation bytes
+            (b'\xf1\x80\x41', FFFD+'A'), # 1 valid cb and 1 invalid
+            (b'\xf1\x80\x41\x41', FFFD+'AA'), # 1 valid cb and 1 invalid
+            (b'\xf1\x80\x80\x41', FFFD+'A'), # 2 valid cb and 1 invalid
+            (b'\xf1\x41\x80', FFFD+'A'+FFFD), # 1 invalid cv and 1 valid
+            (b'\xf1\x41\x80\x80', FFFD+'A'+FFFD*2), # 1 invalid cb and 2 invalid
+            (b'\xf1\x41\x80\x41', FFFD+'A'+FFFD+'A'), # 2 invalid cb and 1 invalid
+            (b'\xf1\x41\x41\x80', FFFD+'AA'+FFFD), # 1 valid cb and 1 invalid
+            (b'\xf1\x41\xf1\x80', FFFD+'A'+FFFD),
+            (b'\xf1\x41\x80\xf1', FFFD+'A'+FFFD*2),
+            (b'\xf1\xf1\x80\x41', FFFD*2+'A'),
+            (b'\xf1\x41\xf1\xf1', FFFD+'A'+FFFD*2),
+            # with invalid start byte of a 4-byte sequence (rfc2279)
+            (b'\xf5', FFFD), # only the start byte
+            (b'\xf5\xf5', FFFD*2), # 2 start bytes
+            (b'\xf5\x80', FFFD*2), # only 1 continuation byte
+            (b'\xf5\x80\x80', FFFD*3), # only 2 continuation byte
+            (b'\xf5\x80\x80\x80', FFFD*4), # 3 continuation bytes
+            (b'\xf5\x80\x41', FFFD*2+'A'), #  1 valid cb and 1 invalid
+            (b'\xf5\x80\x41\xf5', FFFD*2+'A'+FFFD),
+            (b'\xf5\x41\x80\x80\x41', FFFD+'A'+FFFD*2+'A'),
+            # with invalid start byte of a 5-byte sequence (rfc2279)
+            (b'\xf8', FFFD), # only the start byte
+            (b'\xf8\xf8', FFFD*2), # 2 start bytes
+            (b'\xf8\x80', FFFD*2), # only one continuation byte
+            (b'\xf8\x80\x41', FFFD*2 + 'A'), # 1 valid cb and 1 invalid
+            (b'\xf8\x80\x80\x80\x80', FFFD*5), # invalid 5 bytes seq with 5 bytes
+            # with invalid start byte of a 6-byte sequence (rfc2279)
+            (b'\xfc', FFFD), # only the start byte
+            (b'\xfc\xfc', FFFD*2), # 2 start bytes
+            (b'\xfc\x80\x80', FFFD*3), # only 2 continuation bytes
+            (b'\xfc\x80\x80\x80\x80\x80', FFFD*6), # 6 continuation bytes
+            # invalid start byte
+            (b'\xfe', FFFD),
+            (b'\xfe\x80\x80', FFFD*3),
+            # other sequences
+            (b'\xf1\x80\x41\x42\x43', '\ufffd\x41\x42\x43'),
+            (b'\xf1\x80\xff\x42\x43', '\ufffd\ufffd\x42\x43'),
+            (b'\xf1\x80\xc2\x81\x43', '\ufffd\x81\x43'),
+            (b'\x61\xF1\x80\x80\xE1\x80\xC2\x62\x80\x63\x80\xBF\x64',
+             '\x61\uFFFD\uFFFD\uFFFD\x62\uFFFD\x63\uFFFD\uFFFD\x64'),
+        ]
+        for n, (seq, res) in enumerate(sequences):
+            self.assertRaises(UnicodeDecodeError, seq.decode, 'utf-8', 'strict')
+            self.assertEqual(seq.decode('utf-8', 'replace'), res)
+            self.assertEqual((seq+b'b').decode('utf-8', 'replace'), res+'b')
+            self.assertEqual(seq.decode('utf-8', 'ignore'),
+                             res.replace('\uFFFD', ''))
+
     def test_codecs_idna(self):
         # Test whether trailing dot is preserved
         self.assertEqual("www.python.org.".encode("idna"), b"www.python.org.")
@@ -1039,8 +1192,8 @@
         s = bytes(range(128))
         for encoding in (
             'cp037', 'cp1026',
-            'cp437', 'cp500', 'cp737', 'cp775', 'cp850',
-            'cp852', 'cp855', 'cp860', 'cp861', 'cp862',
+            'cp437', 'cp500', 'cp720', 'cp737', 'cp775', 'cp850',
+            'cp852', 'cp855', 'cp858', 'cp860', 'cp861', 'cp862',
             'cp863', 'cp865', 'cp866',
             'iso8859_10', 'iso8859_13', 'iso8859_14', 'iso8859_15',
             'iso8859_2', 'iso8859_3', 'iso8859_4', 'iso8859_5', 'iso8859_6',
@@ -1067,8 +1220,8 @@
         s = bytes(range(128, 256))
         for encoding in (
             'cp037', 'cp1026',
-            'cp437', 'cp500', 'cp737', 'cp775', 'cp850',
-            'cp852', 'cp855', 'cp860', 'cp861', 'cp862',
+            'cp437', 'cp500', 'cp720', 'cp737', 'cp775', 'cp850',
+            'cp852', 'cp855', 'cp858', 'cp860', 'cp861', 'cp862',
             'cp863', 'cp865', 'cp866',
             'iso8859_10', 'iso8859_13', 'iso8859_14', 'iso8859_15',
             'iso8859_2', 'iso8859_4', 'iso8859_5',

Modified: python/branches/py3k-cdecimal/Lib/test/test_unpack.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_unpack.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_unpack.py	Sat Jul 17 12:36:09 2010
@@ -62,14 +62,14 @@
     >>> a, b = t
     Traceback (most recent call last):
       ...
-    ValueError: too many values to unpack
+    ValueError: too many values to unpack (expected 2)
 
 Unpacking tuple of wrong size
 
     >>> a, b = l
     Traceback (most recent call last):
       ...
-    ValueError: too many values to unpack
+    ValueError: too many values to unpack (expected 2)
 
 Unpacking sequence too short
 
@@ -83,7 +83,7 @@
     >>> a, b = Seq()
     Traceback (most recent call last):
       ...
-    ValueError: too many values to unpack
+    ValueError: too many values to unpack (expected 2)
 
 Unpacking a sequence where the test for too long raises a different kind of
 error

Modified: python/branches/py3k-cdecimal/Lib/test/test_urllib.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_urllib.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_urllib.py	Sat Jul 17 12:36:09 2010
@@ -795,6 +795,116 @@
         self.assertEqual("a=a&a=b",
                          urllib.parse.urlencode({"a": {"a": 1, "b": 1}}, True))
 
+    def test_urlencode_encoding(self):
+        # ASCII encoding. Expect %3F with errors="replace'
+        given = (('\u00a0', '\u00c1'),)
+        expect = '%3F=%3F'
+        result = urllib.parse.urlencode(given, encoding="ASCII", errors="replace")
+        self.assertEqual(expect, result)
+
+        # Default is UTF-8 encoding.
+        given = (('\u00a0', '\u00c1'),)
+        expect = '%C2%A0=%C3%81'
+        result = urllib.parse.urlencode(given)
+        self.assertEqual(expect, result)
+
+        # Latin-1 encoding.
+        given = (('\u00a0', '\u00c1'),)
+        expect = '%A0=%C1'
+        result = urllib.parse.urlencode(given, encoding="latin-1")
+        self.assertEqual(expect, result)
+
+    def test_urlencode_encoding_doseq(self):
+        # ASCII Encoding. Expect %3F with errors="replace'
+        given = (('\u00a0', '\u00c1'),)
+        expect = '%3F=%3F'
+        result = urllib.parse.urlencode(given, doseq=True,
+                                        encoding="ASCII", errors="replace")
+        self.assertEqual(expect, result)
+
+        # ASCII Encoding. On a sequence of values.
+        given = (("\u00a0", (1, "\u00c1")),)
+        expect = '%3F=1&%3F=%3F'
+        result = urllib.parse.urlencode(given, True,
+                                        encoding="ASCII", errors="replace")
+        self.assertEqual(expect, result)
+
+        # Utf-8
+        given = (("\u00a0", "\u00c1"),)
+        expect = '%C2%A0=%C3%81'
+        result = urllib.parse.urlencode(given, True)
+        self.assertEqual(expect, result)
+
+        given = (("\u00a0", (42, "\u00c1")),)
+        expect = '%C2%A0=42&%C2%A0=%C3%81'
+        result = urllib.parse.urlencode(given, True)
+        self.assertEqual(expect, result)
+
+        # latin-1
+        given = (("\u00a0", "\u00c1"),)
+        expect = '%A0=%C1'
+        result = urllib.parse.urlencode(given, True, encoding="latin-1")
+        self.assertEqual(expect, result)
+
+        given = (("\u00a0", (42, "\u00c1")),)
+        expect = '%A0=42&%A0=%C1'
+        result = urllib.parse.urlencode(given, True, encoding="latin-1")
+        self.assertEqual(expect, result)
+
+    def test_urlencode_bytes(self):
+        given = ((b'\xa0\x24', b'\xc1\x24'),)
+        expect = '%A0%24=%C1%24'
+        result = urllib.parse.urlencode(given)
+        self.assertEqual(expect, result)
+        result = urllib.parse.urlencode(given, True)
+        self.assertEqual(expect, result)
+
+        # Sequence of values
+        given = ((b'\xa0\x24', (42, b'\xc1\x24')),)
+        expect = '%A0%24=42&%A0%24=%C1%24'
+        result = urllib.parse.urlencode(given, True)
+        self.assertEqual(expect, result)
+
+    def test_urlencode_encoding_safe_parameter(self):
+
+        # Send '$' (\x24) as safe character
+        # Default utf-8 encoding
+
+        given = ((b'\xa0\x24', b'\xc1\x24'),)
+        result = urllib.parse.urlencode(given, safe=":$")
+        expect = '%A0$=%C1$'
+        self.assertEqual(expect, result)
+
+        given = ((b'\xa0\x24', b'\xc1\x24'),)
+        result = urllib.parse.urlencode(given, doseq=True, safe=":$")
+        expect = '%A0$=%C1$'
+        self.assertEqual(expect, result)
+
+        # Safe parameter in sequence
+        given = ((b'\xa0\x24', (b'\xc1\x24', 0xd, 42)),)
+        expect = '%A0$=%C1$&%A0$=13&%A0$=42'
+        result = urllib.parse.urlencode(given, True, safe=":$")
+        self.assertEqual(expect, result)
+
+        # Test all above in latin-1 encoding
+
+        given = ((b'\xa0\x24', b'\xc1\x24'),)
+        result = urllib.parse.urlencode(given, safe=":$",
+                                        encoding="latin-1")
+        expect = '%A0$=%C1$'
+        self.assertEqual(expect, result)
+
+        given = ((b'\xa0\x24', b'\xc1\x24'),)
+        expect = '%A0$=%C1$'
+        result = urllib.parse.urlencode(given, doseq=True, safe=":$",
+                                        encoding="latin-1")
+
+        given = ((b'\xa0\x24', (b'\xc1\x24', 0xd, 42)),)
+        expect = '%A0$=%C1$&%A0$=13&%A0$=42'
+        result = urllib.parse.urlencode(given, True, safe=":$",
+                                        encoding="latin-1")
+        self.assertEqual(expect, result)
+
 class Pathname_Tests(unittest.TestCase):
     """Test pathname2url() and url2pathname()"""
 

Modified: python/branches/py3k-cdecimal/Lib/test/test_urllib2.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_urllib2.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_urllib2.py	Sat Jul 17 12:36:09 2010
@@ -731,6 +731,8 @@
             ("file://ftp.example.com///foo.txt", False),
 # XXXX bug: fails with OSError, should be URLError
             ("file://ftp.example.com/foo.txt", False),
+            ("file://somehost//foo/something.txt", True),
+            ("file://localhost//foo/something.txt", False),
             ]:
             req = Request(url)
             try:
@@ -741,6 +743,7 @@
             else:
                 self.assertTrue(o.req is req)
                 self.assertEqual(req.type, "ftp")
+            self.assertEqual(req.type is "ftp", ftp)
 
     def test_http(self):
 

Modified: python/branches/py3k-cdecimal/Lib/test/test_urlparse.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_urlparse.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_urlparse.py	Sat Jul 17 12:36:09 2010
@@ -7,6 +7,7 @@
 RFC1808_BASE = "http://a/b/c/d;p?q#f"
 RFC2396_BASE = "http://a/b/c/d;p?q"
 RFC3986_BASE = 'http://a/b/c/d;p?q'
+SIMPLE_BASE  = 'http://a/b/c/d'
 
 # A list of test cases.  Each test case is a a two-tuple that contains
 # a string with the query and a dictionary with the expected result.
@@ -295,6 +296,37 @@
         #self.checkJoin(RFC3986_BASE, 'http:g','http:g') # strict parser
         self.checkJoin(RFC3986_BASE, 'http:g','http://a/b/c/g') #relaxed parser
 
+    def test_urljoins(self):
+        self.checkJoin(SIMPLE_BASE, 'g:h','g:h')
+        self.checkJoin(SIMPLE_BASE, 'http:g','http://a/b/c/g')
+        self.checkJoin(SIMPLE_BASE, 'http:','http://a/b/c/d')
+        self.checkJoin(SIMPLE_BASE, 'g','http://a/b/c/g')
+        self.checkJoin(SIMPLE_BASE, './g','http://a/b/c/g')
+        self.checkJoin(SIMPLE_BASE, 'g/','http://a/b/c/g/')
+        self.checkJoin(SIMPLE_BASE, '/g','http://a/g')
+        self.checkJoin(SIMPLE_BASE, '//g','http://g')
+        self.checkJoin(SIMPLE_BASE, '?y','http://a/b/c/d?y')
+        self.checkJoin(SIMPLE_BASE, 'g?y','http://a/b/c/g?y')
+        self.checkJoin(SIMPLE_BASE, 'g?y/./x','http://a/b/c/g?y/./x')
+        self.checkJoin(SIMPLE_BASE, '.','http://a/b/c/')
+        self.checkJoin(SIMPLE_BASE, './','http://a/b/c/')
+        self.checkJoin(SIMPLE_BASE, '..','http://a/b/')
+        self.checkJoin(SIMPLE_BASE, '../','http://a/b/')
+        self.checkJoin(SIMPLE_BASE, '../g','http://a/b/g')
+        self.checkJoin(SIMPLE_BASE, '../..','http://a/')
+        self.checkJoin(SIMPLE_BASE, '../../g','http://a/g')
+        self.checkJoin(SIMPLE_BASE, '../../../g','http://a/../g')
+        self.checkJoin(SIMPLE_BASE, './../g','http://a/b/g')
+        self.checkJoin(SIMPLE_BASE, './g/.','http://a/b/c/g/')
+        self.checkJoin(SIMPLE_BASE, '/./g','http://a/./g')
+        self.checkJoin(SIMPLE_BASE, 'g/./h','http://a/b/c/g/h')
+        self.checkJoin(SIMPLE_BASE, 'g/../h','http://a/b/c/h')
+        self.checkJoin(SIMPLE_BASE, 'http:g','http://a/b/c/g')
+        self.checkJoin(SIMPLE_BASE, 'http:','http://a/b/c/d')
+        self.checkJoin(SIMPLE_BASE, 'http:?y','http://a/b/c/d?y')
+        self.checkJoin(SIMPLE_BASE, 'http:g?y','http://a/b/c/g?y')
+        self.checkJoin(SIMPLE_BASE, 'http:g?y/./x','http://a/b/c/g?y/./x')
+
     def test_RFC2732(self):
         for url, hostname, port in [
             ('http://Test.python.org:5432/foo/', 'test.python.org', 5432),

Modified: python/branches/py3k-cdecimal/Lib/test/test_xmlrpc_net.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_xmlrpc_net.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_xmlrpc_net.py	Sat Jul 17 12:36:09 2010
@@ -48,7 +48,7 @@
         # Perform a minimal sanity check on the result, just to be sure
         # the request means what we think it means.
         self.assertIsInstance(builders, collections.Sequence)
-        self.assertTrue([x for x in builders if "trunk" in x], builders)
+        self.assertTrue([x for x in builders if "3.x" in x], builders)
 
 
 def test_main():

Modified: python/branches/py3k-cdecimal/Lib/unittest/case.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/unittest/case.py	(original)
+++ python/branches/py3k-cdecimal/Lib/unittest/case.py	Sat Jul 17 12:36:09 2010
@@ -940,9 +940,14 @@
                 'Second argument is not a string'))
 
         if first != second:
-            standardMsg = '%s != %s' % (safe_repr(first, True), safe_repr(second, True))
-            diff = '\n' + ''.join(difflib.ndiff(first.splitlines(True),
-                                                       second.splitlines(True)))
+            firstlines = first.splitlines(True)
+            secondlines = second.splitlines(True)
+            if len(firstlines) == 1 and first.strip('\r\n') == first:
+                firstlines = [first + '\n']
+                secondlines = [second + '\n']
+            standardMsg = '%s != %s' % (safe_repr(first, True),
+                                        safe_repr(second, True))
+            diff = '\n' + ''.join(difflib.ndiff(firstlines, secondlines))
             standardMsg = self._truncateMessage(standardMsg, diff)
             self.fail(self._formatMessage(msg, standardMsg))
 

Modified: python/branches/py3k-cdecimal/Lib/unittest/test/test_case.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/unittest/test/test_case.py	(original)
+++ python/branches/py3k-cdecimal/Lib/unittest/test/test_case.py	Sat Jul 17 12:36:09 2010
@@ -841,6 +841,21 @@
             # so can't use assertEqual either. Just use assertTrue.
             self.assertTrue(sample_text_error == error)
 
+    def testAsertEqualSingleLine(self):
+        sample_text = "laden swallows fly slowly"
+        revised_sample_text = "unladen swallows fly quickly"
+        sample_text_error = """\
+- laden swallows fly slowly
+?                    ^^^^
++ unladen swallows fly quickly
+? ++                   ^^^^^
+"""
+        try:
+            self.assertEqual(sample_text, revised_sample_text)
+        except self.failureException as e:
+            error = str(e).split('\n', 1)[1]
+            self.assertTrue(sample_text_error == error)
+
     def testAssertIsNone(self):
         self.assertIsNone(None)
         self.assertRaises(self.failureException, self.assertIsNone, False)

Modified: python/branches/py3k-cdecimal/Lib/urllib/parse.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/urllib/parse.py	(original)
+++ python/branches/py3k-cdecimal/Lib/urllib/parse.py	Sat Jul 17 12:36:09 2010
@@ -8,21 +8,22 @@
 RFC 2732 : "Format for Literal IPv6 Addresses in URL's by R.Hinden, B.Carpenter
 and L.Masinter, December 1999.
 
-RFC2396:  "Uniform Resource Identifiers (URI)": Generic Syntax by T.
+RFC 2396:  "Uniform Resource Identifiers (URI)": Generic Syntax by T.
 Berners-Lee, R. Fielding, and L. Masinter, August 1998.
 
-RFC2368: "The mailto URL scheme", by P.Hoffman , L Masinter, J. Zwinski, July 1998.
+RFC 2368: "The mailto URL scheme", by P.Hoffman , L Masinter, J. Zwinski, July 1998.
 
 RFC 1808: "Relative Uniform Resource Locators", by R. Fielding, UC Irvine, June
 1995.
 
-RFC1738: "Uniform Resource Locators (URL)" by T. Berners-Lee, L. Masinter, M.
+RFC 1738: "Uniform Resource Locators (URL)" by T. Berners-Lee, L. Masinter, M.
 McCahill, December 1994
 
-RFC 3986 is considered the current standard and any changes to urlparse module
-should conform to this.  urlparse module is not entirely compliant with this.
-The defacto scenarios of parsing are considered sometimes and for backward
-compatiblity purposes, older RFC uses of parsing are retained. The testcases in
+RFC 3986 is considered the current standard and any future changes to
+urlparse module should conform with it.  The urlparse module is
+currently not entirely compliant with this RFC due to defacto
+scenarios for parsing, and for backward compatibility purposes, some
+parsing quirks from older RFCs are retained. The testcases in
 test_urlparse.py provides a good indicator of parsing behavior.
 """
 
@@ -220,6 +221,11 @@
     return urlunsplit((scheme, netloc, url, query, fragment))
 
 def urlunsplit(components):
+    """Combine the elements of a tuple as returned by urlsplit() into a
+    complete URL as a string. The data argument can be any five-item iterable.
+    This may result in a slightly different, but equivalent URL, if the URL that
+    was parsed originally had unnecessary delimiters (for example, a ? with an
+    empty query; the RFC states that these are equivalent)."""
     scheme, netloc, url, query, fragment = components
     if netloc or (scheme and scheme in uses_netloc and url[:2] != '//'):
         if url and url[:1] != '/': url = '/' + url
@@ -553,7 +559,7 @@
         _safe_quoters[safe] = quoter = Quoter(safe).__getitem__
     return ''.join([quoter(char) for char in bs])
 
-def urlencode(query, doseq=False):
+def urlencode(query, doseq=False, safe='', encoding=None, errors=None):
     """Encode a sequence of two-element tuples or dictionary into a URL query string.
 
     If any values in the query arg are sequences and doseq is true, each
@@ -562,6 +568,10 @@
     If the query arg is a sequence of two-element tuples, the order of the
     parameters in the output will match the order of parameters in the
     input.
+
+    The query arg may be either a string or a bytes type. When query arg is a
+    string, the safe, encoding and error parameters are sent the quote_plus for
+    encoding.
     """
 
     if hasattr(query, "items"):
@@ -586,14 +596,28 @@
     l = []
     if not doseq:
         for k, v in query:
-            k = quote_plus(str(k))
-            v = quote_plus(str(v))
+            if isinstance(k, bytes):
+                k = quote_plus(k, safe)
+            else:
+                k = quote_plus(str(k), safe, encoding, errors)
+
+            if isinstance(v, bytes):
+                v = quote_plus(v, safe)
+            else:
+                v = quote_plus(str(v), safe, encoding, errors)
             l.append(k + '=' + v)
     else:
         for k, v in query:
-            k = quote_plus(str(k))
-            if isinstance(v, str):
-                v = quote_plus(v)
+            if isinstance(k, bytes):
+                k = quote_plus(k, safe)
+            else:
+                k = quote_plus(str(k), safe, encoding, errors)
+
+            if isinstance(v, bytes):
+                v = quote_plus(v, safe)
+                l.append(k + '=' + v)
+            elif isinstance(v, str):
+                v = quote_plus(v, safe, encoding, errors)
                 l.append(k + '=' + v)
             else:
                 try:
@@ -601,12 +625,16 @@
                     x = len(v)
                 except TypeError:
                     # not a sequence
-                    v = quote_plus(str(v))
+                    v = quote_plus(str(v), safe, encoding, errors)
                     l.append(k + '=' + v)
                 else:
                     # loop over the sequence
                     for elt in v:
-                        l.append(k + '=' + quote_plus(str(elt)))
+                        if isinstance(elt, bytes):
+                            elt = quote_plus(elt, safe)
+                        else:
+                            elt = quote_plus(str(elt), safe, encoding, errors)
+                        l.append(k + '=' + elt)
     return '&'.join(l)
 
 # Utilities to parse URLs (most of these return None for missing parts):
@@ -771,67 +799,3 @@
     match = _valueprog.match(attr)
     if match: return match.group(1, 2)
     return attr, None
-
-test_input = """
-      http://a/b/c/d
-
-      g:h        = <URL:g:h>
-      http:g     = <URL:http://a/b/c/g>
-      http:      = <URL:http://a/b/c/d>
-      g          = <URL:http://a/b/c/g>
-      ./g        = <URL:http://a/b/c/g>
-      g/         = <URL:http://a/b/c/g/>
-      /g         = <URL:http://a/g>
-      //g        = <URL:http://g>
-      ?y         = <URL:http://a/b/c/d?y>
-      g?y        = <URL:http://a/b/c/g?y>
-      g?y/./x    = <URL:http://a/b/c/g?y/./x>
-      .          = <URL:http://a/b/c/>
-      ./         = <URL:http://a/b/c/>
-      ..         = <URL:http://a/b/>
-      ../        = <URL:http://a/b/>
-      ../g       = <URL:http://a/b/g>
-      ../..      = <URL:http://a/>
-      ../../g    = <URL:http://a/g>
-      ../../../g = <URL:http://a/../g>
-      ./../g     = <URL:http://a/b/g>
-      ./g/.      = <URL:http://a/b/c/g/>
-      /./g       = <URL:http://a/./g>
-      g/./h      = <URL:http://a/b/c/g/h>
-      g/../h     = <URL:http://a/b/c/h>
-      http:g     = <URL:http://a/b/c/g>
-      http:      = <URL:http://a/b/c/d>
-      http:?y         = <URL:http://a/b/c/d?y>
-      http:g?y        = <URL:http://a/b/c/g?y>
-      http:g?y/./x    = <URL:http://a/b/c/g?y/./x>
-"""
-
-def test():
-    base = ''
-    if sys.argv[1:]:
-        fn = sys.argv[1]
-        if fn == '-':
-            fp = sys.stdin
-        else:
-            fp = open(fn)
-    else:
-        from io import StringIO
-        fp = StringIO(test_input)
-    for line in fp:
-        words = line.split()
-        if not words:
-            continue
-        url = words[0]
-        parts = urlparse(url)
-        print('%-10s : %s' % (url, parts))
-        abs = urljoin(base, url)
-        if not base:
-            base = abs
-        wrapped = '<URL:%s>' % abs
-        print('%-10s = %s' % (url, wrapped))
-        if len(words) == 3 and words[1] == '=':
-            if wrapped != words[2]:
-                print('EXPECTED', words[2], '!!!!!!!!!!')
-
-if __name__ == '__main__':
-    test()

Modified: python/branches/py3k-cdecimal/Lib/urllib/request.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/urllib/request.py	(original)
+++ python/branches/py3k-cdecimal/Lib/urllib/request.py	Sat Jul 17 12:36:09 2010
@@ -1188,7 +1188,8 @@
     # Use local file or FTP depending on form of URL
     def file_open(self, req):
         url = req.selector
-        if url[:2] == '//' and url[2:3] != '/':
+        if url[:2] == '//' and url[2:3] != '/' and (req.host and
+                req.host != 'localhost'):
             req.type = 'ftp'
             return self.parent.open(req)
         else:
@@ -1866,7 +1867,8 @@
         else:
             return self.http_error_default(url, fp, errcode, errmsg, headers)
 
-    def http_error_401(self, url, fp, errcode, errmsg, headers, data=None):
+    def http_error_401(self, url, fp, errcode, errmsg, headers, data=None,
+            retry=False):
         """Error 401 -- authentication required.
         This function supports Basic authentication only."""
         if not 'www-authenticate' in headers:
@@ -1882,13 +1884,17 @@
         if scheme.lower() != 'basic':
             URLopener.http_error_default(self, url, fp,
                                          errcode, errmsg, headers)
+        if not retry:
+            URLopener.http_error_default(self, url, fp, errcode, errmsg,
+                    headers)
         name = 'retry_' + self.type + '_basic_auth'
         if data is None:
             return getattr(self,name)(url, realm)
         else:
             return getattr(self,name)(url, realm, data)
 
-    def http_error_407(self, url, fp, errcode, errmsg, headers, data=None):
+    def http_error_407(self, url, fp, errcode, errmsg, headers, data=None,
+            retry=False):
         """Error 407 -- proxy authentication required.
         This function supports Basic authentication only."""
         if not 'proxy-authenticate' in headers:
@@ -1904,6 +1910,9 @@
         if scheme.lower() != 'basic':
             URLopener.http_error_default(self, url, fp,
                                          errcode, errmsg, headers)
+        if not retry:
+            URLopener.http_error_default(self, url, fp, errcode, errmsg,
+                    headers)
         name = 'retry_proxy_' + self.type + '_basic_auth'
         if data is None:
             return getattr(self,name)(url, realm)
@@ -2196,8 +2205,13 @@
                         continue
 
                 base = ip2num(m.group(1))
-                mask = int(m.group(2)[1:])
-                mask = 32 - mask
+                mask = m.group(2)
+                if mask is None:
+                    mask = 8 * (m.group(1).count('.') + 1)
+
+                else:
+                    mask = int(mask[1:])
+                    mask = 32 - mask
 
                 if (hostIP >> mask) == (base >> mask):
                     return True
@@ -2265,6 +2279,7 @@
                         proxies['http'] = proxyServer
                     else:
                         proxies['http'] = 'http://%s' % proxyServer
+                        proxies['https'] = 'https://%s' % proxyServer
                         proxies['ftp'] = 'ftp://%s' % proxyServer
             internetSettings.Close()
         except (WindowsError, ValueError, TypeError):

Modified: python/branches/py3k-cdecimal/Lib/uuid.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/uuid.py	(original)
+++ python/branches/py3k-cdecimal/Lib/uuid.py	Sat Jul 17 12:36:09 2010
@@ -427,17 +427,17 @@
         if hasattr(lib, 'uuid_generate_time'):
             _uuid_generate_time = lib.uuid_generate_time
 
-    # The uuid_generate_* functions are broken on MacOS X 10.6, as noted
+    # The uuid_generate_* functions are broken on MacOS X 10.5, as noted
     # in issue #8621 the function generates the same sequence of values
     # in the parent process and all children created using fork (unless
     # those children use exec as well).
     #
-    # Assume that the uuid_generate functions are broken from 10.6 onward,
+    # Assume that the uuid_generate functions are broken from 10.5 onward,
     # the test can be adjusted when a later version is fixed.
     import sys
     if sys.platform == 'darwin':
         import os
-        if int(os.uname()[2].split('.')[0]) >= 10:
+        if int(os.uname()[2].split('.')[0]) >= 9:
             _uuid_generate_random = _uuid_generate_time = None
 
     # On Windows prior to 2000, UuidCreate gives a UUID containing the

Modified: python/branches/py3k-cdecimal/Lib/warnings.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/warnings.py	(original)
+++ python/branches/py3k-cdecimal/Lib/warnings.py	Sat Jul 17 12:36:09 2010
@@ -371,8 +371,10 @@
 # Module initialization
 _processoptions(sys.warnoptions)
 if not _warnings_defaults:
-    simplefilter("ignore", category=PendingDeprecationWarning, append=1)
-    simplefilter("ignore", category=ImportWarning, append=1)
+    silence = [ImportWarning, PendingDeprecationWarning]
+    silence.append(DeprecationWarning)
+    for cls in silence:
+        simplefilter("ignore", category=cls)
     bytes_warning = sys.flags.bytes_warning
     if bytes_warning > 1:
         bytes_action = "error"

Modified: python/branches/py3k-cdecimal/Lib/xml/dom/expatbuilder.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/xml/dom/expatbuilder.py	(original)
+++ python/branches/py3k-cdecimal/Lib/xml/dom/expatbuilder.py	Sat Jul 17 12:36:09 2010
@@ -242,7 +242,7 @@
         doctype = self.document.implementation.createDocumentType(
             doctypeName, publicId, systemId)
         doctype.ownerDocument = self.document
-        self.document.childNodes.append(doctype)
+        _append_child(self.document, doctype)
         self.document.doctype = doctype
         if self._filter and self._filter.acceptNode(doctype) == FILTER_REJECT:
             self.document.doctype = None

Modified: python/branches/py3k-cdecimal/Makefile.pre.in
==============================================================================
--- python/branches/py3k-cdecimal/Makefile.pre.in	(original)
+++ python/branches/py3k-cdecimal/Makefile.pre.in	Sat Jul 17 12:36:09 2010
@@ -59,12 +59,18 @@
 # Compiler options
 OPT=		@OPT@
 BASECFLAGS=	@BASECFLAGS@
-CFLAGS=		$(BASECFLAGS) @CFLAGS@ $(OPT) $(EXTRA_CFLAGS)
+CONFIGURE_CFLAGS=	@CFLAGS@
+CONFIGURE_CPPFLAGS=	@CPPFLAGS@
+CONFIGURE_LDFLAGS=	@LDFLAGS@
+# Avoid assigning CFLAGS, LDFLAGS, etc. so users can use them on the
+# command line to append to these values without stomping the pre-set
+# values.
+PY_CFLAGS=	$(BASECFLAGS) $(OPT) $(CONFIGURE_CFLAGS) $(CFLAGS) $(EXTRA_CFLAGS)
 # Both CPPFLAGS and LDFLAGS need to contain the shell's value for setup.py to
 # be able to build extension modules using the directories specified in the
 # environment variables
-CPPFLAGS=	-I. -IInclude -I$(srcdir)/Include @CPPFLAGS@
-LDFLAGS=	@LDFLAGS@
+PY_CPPFLAGS=	-I. -IInclude -I$(srcdir)/Include $(CONFIGURE_CPPFLAGS) $(CPPFLAGS)
+PY_LDFLAGS=	$(CONFIGURE_LDFLAGS) $(LDFLAGS)
 LDLAST=		@LDLAST@
 SGI_ABI=	@SGI_ABI@
 CCSHARED=	@CCSHARED@
@@ -73,7 +79,7 @@
 # Extra C flags added for building the interpreter object files.
 CFLAGSFORSHARED=@CFLAGSFORSHARED@
 # C flags used for building the interpreter object files
-PY_CFLAGS=	$(CFLAGS) $(CPPFLAGS) $(CFLAGSFORSHARED) -DPy_BUILD_CORE
+PY_CORE_CFLAGS=	$(PY_CFLAGS) $(PY_CPPFLAGS) $(CFLAGSFORSHARED) -DPy_BUILD_CORE
 
 
 # Machine-dependent subdirectories
@@ -411,7 +417,7 @@
 
 # Build the interpreter
 $(BUILDPYTHON):	Modules/python.o $(LIBRARY) $(LDLIBRARY)
-		$(LINKCC) $(LDFLAGS) $(LINKFORSHARED) -o $@ \
+		$(LINKCC) $(PY_LDFLAGS) $(LINKFORSHARED) -o $@ \
 			Modules/python.o \
 			$(BLDLIBRARY) $(LIBS) $(MODLIBS) $(SYSLIBS) $(LDLAST)
 
@@ -422,8 +428,8 @@
 # Build the shared modules
 sharedmods: $(BUILDPYTHON)
 	@case $$MAKEFLAGS in \
-	*s*) $(RUNSHARED) CC='$(CC)' LDSHARED='$(BLDSHARED)' LDFLAGS='$(LDFLAGS)' OPT='$(OPT)' ./$(BUILDPYTHON) -E $(srcdir)/setup.py -q build;; \
-	*) $(RUNSHARED) CC='$(CC)' LDSHARED='$(BLDSHARED)' LDFLAGS='$(LDFLAGS)' OPT='$(OPT)' ./$(BUILDPYTHON) -E $(srcdir)/setup.py build;; \
+	*s*) $(RUNSHARED) CC='$(CC)' LDSHARED='$(BLDSHARED)' LDFLAGS='$(PY_LDFLAGS)' OPT='$(OPT)' ./$(BUILDPYTHON) -E $(srcdir)/setup.py -q build;; \
+	*) $(RUNSHARED) CC='$(CC)' LDSHARED='$(BLDSHARED)' LDFLAGS='$(PY_LDFLAGS)' OPT='$(OPT)' ./$(BUILDPYTHON) -E $(srcdir)/setup.py build;; \
 	esac
 
 # Build static library
@@ -440,18 +446,18 @@
 
 libpython$(VERSION).so: $(LIBRARY_OBJS)
 	if test $(INSTSONAME) != $(LDLIBRARY); then \
-		$(LDSHARED) $(LDFLAGS) -Wl,-h$(INSTSONAME) -o $(INSTSONAME) $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
+		$(LDSHARED) $(PY_LDFLAGS) -Wl,-h$(INSTSONAME) -o $(INSTSONAME) $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
 		$(LN) -f $(INSTSONAME) $@; \
 	else \
-		$(LDSHARED) $(LDFLAGS) -o $@ $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
+		$(LDSHARED) $(PY_LDFLAGS) -o $@ $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
 	fi
 
 libpython$(VERSION).dylib: $(LIBRARY_OBJS)
-	 $(CC) -dynamiclib -Wl,-single_module $(LDFLAGS) -undefined dynamic_lookup -Wl,-install_name,$(prefix)/lib/libpython$(VERSION).dylib -Wl,-compatibility_version,$(VERSION) -Wl,-current_version,$(VERSION) -o $@ $(LIBRARY_OBJS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
+	 $(CC) -dynamiclib -Wl,-single_module $(PY_LDFLAGS) -undefined dynamic_lookup -Wl,-install_name,$(prefix)/lib/libpython$(VERSION).dylib -Wl,-compatibility_version,$(VERSION) -Wl,-current_version,$(VERSION) -o $@ $(LIBRARY_OBJS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
 		 
 
 libpython$(VERSION).sl: $(LIBRARY_OBJS)
-	$(LDSHARED) $(LDFLAGS) -o $@ $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST)
+	$(LDSHARED) $(PY_LDFLAGS) -o $@ $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST)
 
 # Copy up the gdb python hooks into a position where they can be automatically
 # loaded by gdb during Lib/test/test_gdb.py
@@ -497,7 +503,7 @@
 # for a shared core library; otherwise, this rule is a noop.
 $(DLLLIBRARY) libpython$(VERSION).dll.a: $(LIBRARY_OBJS)
 	if test -n "$(DLLLIBRARY)"; then \
-		$(LDSHARED) $(LDFLAGS) -Wl,--out-implib=$@ -o $(DLLLIBRARY) $^ \
+		$(LDSHARED) $(PY_LDFLAGS) -Wl,--out-implib=$@ -o $(DLLLIBRARY) $^ \
 			$(LIBS) $(MODLIBS) $(SYSLIBS) $(LDLAST); \
 	else true; \
 	fi
@@ -541,10 +547,10 @@
 		$(SIGNAL_OBJS) \
 		$(MODOBJS) \
 		$(srcdir)/Modules/getbuildinfo.c
-	$(CC) -c $(PY_CFLAGS) -DSVNVERSION="\"`LC_ALL=C $(SVNVERSION)`\"" -o $@ $(srcdir)/Modules/getbuildinfo.c
+	$(CC) -c $(PY_CORE_CFLAGS) -DSVNVERSION="\"`LC_ALL=C $(SVNVERSION)`\"" -o $@ $(srcdir)/Modules/getbuildinfo.c
 
 Modules/getpath.o: $(srcdir)/Modules/getpath.c Makefile
-	$(CC) -c $(PY_CFLAGS) -DPYTHONPATH='"$(PYTHONPATH)"' \
+	$(CC) -c $(PY_CORE_CFLAGS) -DPYTHONPATH='"$(PYTHONPATH)"' \
 		-DPREFIX='"$(prefix)"' \
 		-DEXEC_PREFIX='"$(exec_prefix)"' \
 		-DVERSION='"$(VERSION)"' \
@@ -552,7 +558,7 @@
 		-o $@ $(srcdir)/Modules/getpath.c
 
 Modules/python.o: $(srcdir)/Modules/python.c
-	$(MAINCC) -c $(PY_CFLAGS) -o $@ $(srcdir)/Modules/python.c
+	$(MAINCC) -c $(PY_CORE_CFLAGS) -o $@ $(srcdir)/Modules/python.c
 
 $(IO_OBJS): $(IO_H)
 
@@ -561,7 +567,7 @@
 		-$(PGEN) $(GRAMMAR_INPUT) $(GRAMMAR_H) $(GRAMMAR_C)
 
 $(PGEN):	$(PGENOBJS)
-		$(CC) $(OPT) $(LDFLAGS) $(PGENOBJS) $(LIBS) -o $(PGEN)
+		$(CC) $(OPT) $(PY_LDFLAGS) $(PGENOBJS) $(LIBS) -o $(PGEN)
 
 Parser/grammar.o:	$(srcdir)/Parser/grammar.c \
 				$(srcdir)/Include/token.h \
@@ -581,10 +587,10 @@
 Python/compile.o Python/symtable.o Python/ast.o: $(GRAMMAR_H) $(AST_H)
 
 Python/getplatform.o: $(srcdir)/Python/getplatform.c
-		$(CC) -c $(PY_CFLAGS) -DPLATFORM='"$(MACHDEP)"' -o $@ $(srcdir)/Python/getplatform.c
+		$(CC) -c $(PY_CORE_CFLAGS) -DPLATFORM='"$(MACHDEP)"' -o $@ $(srcdir)/Python/getplatform.c
 
 Python/importdl.o: $(srcdir)/Python/importdl.c
-		$(CC) -c $(PY_CFLAGS) -I$(DLINCLDIR) -o $@ $(srcdir)/Python/importdl.c
+		$(CC) -c $(PY_CORE_CFLAGS) -I$(DLINCLDIR) -o $@ $(srcdir)/Python/importdl.c
 
 Objects/unicodectype.o:	$(srcdir)/Objects/unicodectype.c \
 				$(srcdir)/Objects/unicodetype_db.h
@@ -1131,7 +1137,7 @@
 
 # Some make's put the object file in the current directory
 .c.o:
-	$(CC) -c $(PY_CFLAGS) -o $@ $<
+	$(CC) -c $(PY_CORE_CFLAGS) -o $@ $<
 
 # Run reindent on the library
 reindent:

Modified: python/branches/py3k-cdecimal/Misc/ACKS
==============================================================================
--- python/branches/py3k-cdecimal/Misc/ACKS	(original)
+++ python/branches/py3k-cdecimal/Misc/ACKS	Sat Jul 17 12:36:09 2010
@@ -145,6 +145,7 @@
 Vadim Chugunov
 David Cinege
 Mike Clarkson
+Andrew Clegg
 Brad Clements
 Steve Clift
 Nick Coghlan
@@ -158,6 +159,7 @@
 Juan Jos?? Conti
 Matt Conway
 David M. Cooke
+Jason R. Coombs
 Greg Copeland
 Aldo Cortesi
 David Costanzo
@@ -406,6 +408,7 @@
 Kurt B. Kaiser
 Tamito Kajiyama
 Peter van Kampen
+Rafe Kaplan
 Jacob Kaplan-Moss
 Lou Kates
 Hiroaki Kawai
@@ -451,6 +454,7 @@
 Simon Law
 Chris Lawrence
 Brian Leair
+James Lee
 John J. Lee
 Inyeol Lee
 Thomas Lee
@@ -796,6 +800,7 @@
 Eren T??rkay
 Lionel Ulmer
 Roger Upole
+Daniel Urban
 Michael Urman
 Hector Urtubia
 Andi Vajda

Modified: python/branches/py3k-cdecimal/Misc/NEWS
==============================================================================
--- python/branches/py3k-cdecimal/Misc/NEWS	(original)
+++ python/branches/py3k-cdecimal/Misc/NEWS	Sat Jul 17 12:36:09 2010
@@ -12,6 +12,45 @@
 Core and Builtins
 -----------------
 
+- Issue #7616: Fix copying of overlapping memoryview slices with the Intel
+  compiler.
+
+- Issue #8413: structsequence now subclasses tuple.
+
+- Issue #8271: during the decoding of an invalid UTF-8 byte sequence, only the
+  start byte and the continuation byte(s) are now considered invalid, instead
+  of the number of bytes specified by the start byte.
+  E.g.: '\xf1\x80AB'.decode('utf-8', 'replace') now returns u'\ufffdAB' and
+  replaces with U+FFFD only the start byte ('\xf1') and the continuation byte
+  ('\x80') even if '\xf1' is the start byte of a 4-bytes sequence.
+  Previous versions returned a single u'\ufffd'.
+
+- Issue #9011: A negated imaginary literal (e.g., "-7j") now has real
+  part -0.0 rather than 0.0.  So "-7j" is now exactly equivalent to
+  "-(7j)".
+
+- Be more specific in error messages about positional arguments.
+
+- Issue #8949: "z" format of PyArg_Parse*() functions doesn't accept bytes
+  objects, as described in the documentation.
+
+- Issue #6543: Write the traceback in the terminal encoding instead of utf-8.
+  Fix the encoding of the modules filename. Patch written by Amaury Forgeot
+  d'Arc.
+
+- Issue #9011: Remove buggy and unnecessary (in 3.x) ST->AST
+  compilation code dealing with unary minus applied to a constant.
+  The removed code was mutating the ST, causing a second compilation
+  to fail.
+
+- Issue #850997: mbcs encoding (Windows only) handles errors argument: strict
+  mode raises unicode errors. The encoder only supports "strict" and "replace"
+  error handlers, the decoder only supports "strict" and "ignore" error
+  handlers. Patch written by Mark Hammond.
+
+- Issue #8850: Remove "w" and "w#" formats from PyArg_Parse*() functions, use
+  "w*" format instead. Add tests for "w*" format.
+
 - Issue #8592: PyArg_Parse*() functions raise a TypeError for "y", "u" and "Z"
   formats if the string contains a null byte/character. Write unit tests for
   string formats.
@@ -23,6 +62,8 @@
 - Issue #8969: On Windows, use mbcs codec in strict mode to encode and decode
   filenames and enable os.fsencode().
 
+- Issue #9058: Remove assertions about INT_MAX in UnicodeDecodeError.
+
 - Issue #8941: decoding big endian UTF-32 data in UCS-2 builds could crash
   the interpreter with characters outside the Basic Multilingual Plane
   (higher than 0x10000).
@@ -432,6 +473,84 @@
 Library
 -------
 
+- Issue #1555570: email no longer inserts extra blank lines when a \r\n
+  combo crosses an 8192 byte boundary.
+
+- Issue #9243: Fix sndhdr module and add unit tests, contributed by James Lee.
+
+- ``ast.literal_eval()`` now allows byte literals.
+
+- Issue #9137: Fix issue in MutableMapping.update, which incorrectly
+  treated keyword arguments called 'self' or 'other' specially.
+
+- ``ast.literal_eval()`` now allows set literals.
+
+- Issue #9164: Ensure that sysconfig handles duplicate -arch flags in CFLAGS.
+
+- Issue #7646: The fnmatch pattern cache no longer grows without bound.
+
+- Issue #9136: Fix 'dictionary changed size during iteration'
+  RuntimeError produced when profiling the decimal module.  This was
+  due to a dangerous iteration over 'locals()' in Context.__init__.
+
+- Fix extreme speed issue in Decimal.pow when the base is an exact
+  power of 10 and the exponent is tiny (for example,
+  Decimal(10) ** Decimal('1e-999999999')).
+
+- Issue #9186: Fix math.log1p(-1.0) to raise ValueError, not OverflowError.
+
+- Issue #9130: Fix validation of relative imports in parser module.
+
+- Issue #9128: Fix validation of class decorators in parser module.
+
+- Issue #9094: python -m pickletools will now disassemble pickle files
+  listed in the command line arguments.  See output of python -m
+  pickletools -h for more details.
+
+- Issue #5468: urlencode to handle bytes type and other encodings in its query
+  parameter. Patch by Dan Mahn.
+
+- Issue #7673: Fix security vulnerability (CVE-2010-2089) in the audioop
+  module, ensure that the input string length is a multiple of the frame size
+
+- Issue #6507: Accept source strings in dis.dis(). Original patch by
+  Daniel Urban.
+
+- Issue #7829: Clearly document that the dis module is exposing an
+  implementation detail that is not stable between Python VMs or releases.
+
+- Issue #6589: cleanup asyncore.socket_map in case smtpd.SMTPServer constructor
+  raises an exception.
+
+- Issue #9110: Addition of ContextDecorator to contextlib, for creating APIs
+  that act as both context managers and decorators. contextmanager changes
+  to use ContextDecorator.
+
+- Implement importlib.abc.SourceLoader and deprecate PyLoader and PyPycLoader
+  for removal in Python 3.4.
+
+- Issue #9064: pdb's "up" and "down" commands now accept an optional argument.
+
+- Issue #9018: os.path.normcase() now raises a TypeError if the argument is
+  not ``str`` or ``bytes``.
+
+- Issue #9075: In the ssl module, remove the setting of a ``debug`` flag
+  on an OpenSSL structure.
+
+- Issue #8682: The ssl module now temporary increments the reference count of
+  a socket object got through ``PyWeakref_GetObject``, so as to avoid possible
+  deallocation while the object is still being used.
+
+- Issue #1368368: FancyURLOpener class changed to throw an Exception on wrong
+  password instead of presenting an interactive prompt. Older behavior can be
+  obtained by passing retry=True to http_error_xxx methods of FancyURLOpener.
+
+- Issue #8720: fix regression caused by fix for #4050 by making getsourcefile
+  smart enough to find source files in the linecache.
+
+- Issue #5610: feedparser no longer eats extra characters at the end of
+  a body part if the body part ends with a \r\n.
+
 - Issue #8986: math.erfc was incorrectly raising OverflowError for
   values between -27.3 and -30.0 on some platforms.
 
@@ -1306,6 +1425,41 @@
 Extension Modules
 -----------------
 
+- Issue #9165: Add new functions math.isfinite and cmath.isfinite, to
+  accompany existing isinf and isnan functions.
+
+- Issue #1578269: Implement os.symlink for Windows 6.0+. Patch by
+  Jason R. Coombs
+
+- In struct.pack, correctly propogate exceptions from computing the truth of an
+  object in the '?' format.
+
+- Issue #9000: datetime.timezone objects now have eval-friendly repr.
+
+- In the math module, correctly lookup __trunc__, __ceil__, and __floor__ as
+  special methods.
+
+- Issue #9005: Prevent utctimetuple() from producing year 0 or year
+  10,000.  Prior to this change, timezone adjustment in utctimetuple()
+  could produce tm_year value of 0 or 10,000.  Now an OverflowError is
+  raised in these edge cases.
+
+- Issue #6641: The ``datetime.strptime`` method now supports the
+  ``%z`` directive.  When the ``%z`` directive is present in the
+  format string, an aware ``datetime`` object is returned with
+  ``tzinfo`` bound to a ``datetime.timezone`` instance constructed
+  from the parsed offset.  If both ``%z`` and ``%Z`` are present, the
+  data in ``%Z`` field is used for timezone name, but ``%Z`` data
+  without ``%z`` is discarded.
+
+- Issue #5094: The ``datetime`` module now has a simple concrete class
+  implementing ``datetime.tzinfo`` interface.  Instances of the new
+  class, ``datetime.timezone``, return fixed name and UTC offset from
+  their ``tzname(dt)`` and ``utcoffset(dt)`` methods.  The ``dst(dt)``
+  method always returns ``None``.  A class attribute, ``utc`` contains
+  an instance representing the UTC timezone.  Original patch by Rafe
+  Kaplan.
+
 - Issue #8973: Add __all__ to struct module; this ensures that
   help(struct) includes documentation for the struct.Struct class.
 
@@ -1396,6 +1550,14 @@
 Build
 -----
 
+- Issue #9189: Make a user-specified CFLAGS, CPPFLAGS, or LDFLAGS
+  setting override the configure and makefile defaults, without
+  deleting options the user didn't intend to override.  Developers
+  should no longer need to specify OPT or EXTRA_CFLAGS, although those
+  variables are still present for backward-compatibility.
+
+- Issue #8854: Fix finding Visual Studio 2008 on Windows x64.
+
 - Issue #1759169, #8864: Drop _XOPEN_SOURCE on Solaris, define it for
   multiprocessing only.
 
@@ -1491,6 +1653,11 @@
 Tests
 -----
 
+- Issue #9251: test_threaded_import didn't fail when run through regrtest
+  if the import lock was disabled.
+
+- Issue #8605: Skip test_gdb if Python is compiled with optimizations.
+
 - Issue #7449: Skip test_socketserver if threading support is disabled
 
 - Issue #8672: Add a zlib test ensuring that an incomplete stream can be
@@ -1619,7 +1786,7 @@
 Tools/Demos
 -----------
 
-- Issue #5464: Implement plural forms in msgfmt.py.
+- Issue #5464, #8974: Implement plural forms in msgfmt.py.
 
 - iobench (a file I/O benchmark) and ccbench (a concurrency benchmark) were
   added to the `Tools/` directory.  They were previously living in the
@@ -2015,6 +2182,8 @@
 Library
 -------
 
+- Issue #8203: Fix IDLE Credits dialog: view_file() uses its encoding argument.
+
 - Issue #5311: bdist_msi can now build packages that do not depend on a
   specific Python version.
 

Modified: python/branches/py3k-cdecimal/Misc/Vim/vimrc
==============================================================================
--- python/branches/py3k-cdecimal/Misc/Vim/vimrc	(original)
+++ python/branches/py3k-cdecimal/Misc/Vim/vimrc	Sat Jul 17 12:36:09 2010
@@ -43,7 +43,7 @@
 " Display tabs at the beginning of a line in Python mode as bad.
 au BufRead,BufNewFile *.py,*.pyw match BadWhitespace /^\t\+/
 " Make trailing whitespace be flagged as bad.
-au BufRead,BufNewFile *.py,*.pyw,*.c,*.h match BadWhitespace /\s\+$/
+au BufRead,BufNewFile *.py,*.pyw,*.c,*.h,*.rst match BadWhitespace /\s\+$/
 
 " Wrap text after a certain number of characters
 " Python: 79 

Modified: python/branches/py3k-cdecimal/Misc/maintainers.rst
==============================================================================
--- python/branches/py3k-cdecimal/Misc/maintainers.rst	(original)
+++ python/branches/py3k-cdecimal/Misc/maintainers.rst	Sat Jul 17 12:36:09 2010
@@ -89,7 +89,7 @@
 doctest             tim_one (inactive)
 dummy_threading     brett.cannon
 email               barry, r.david.murray
-encodings           lemburg, loewis
+encodings           lemburg
 errno
 exceptions
 fcntl
@@ -104,7 +104,7 @@
 gc                  pitrou
 getopt
 getpass
-gettext             loewis
+gettext
 glob
 grp
 gzip
@@ -125,7 +125,7 @@
 keyword
 lib2to3             benjamin.peterson
 linecache
-locale              loewis, lemburg
+locale              lemburg
 logging             vsajip
 macpath
 mailbox             andrew.kuchling
@@ -135,7 +135,7 @@
 mimetypes
 mmap
 modulefinder        theller, jvr
-msilib              loewis
+msilib
 msvcrt
 multiprocessing     jnoller
 netrc
@@ -144,7 +144,7 @@
 numbers
 operator
 optparse            aronacher
-os                  loewis
+os
 ossaudiodev
 parser
 pdb
@@ -217,7 +217,7 @@
 tty
 turtle              gregorlingl
 types
-unicodedata         loewis, lemburg, ezio.melotti
+unicodedata         lemburg, ezio.melotti
 unittest            michael.foord
 urllib              orsenthil
 uu
@@ -226,13 +226,13 @@
 wave
 weakref             fdrake
 webbrowser          georg.brandl
-winreg
+winreg              brian.curtin
 winsound            effbot (inactive)
 wsgiref             pje
 xdrlib
-xml                 loewis
+xml
 xml.etree           effbot (inactive)
-xmlrpc              loewis
+xmlrpc
 zipfile
 zipimport
 zlib
@@ -277,13 +277,13 @@
 i18n                lemburg
 import machinery    brett.cannon, ncoghlan
 io                  pitrou, benjamin.peterson
-locale              lemburg, loewis
+locale              lemburg
 mathematics         mark.dickinson, eric.smith, lemburg
 memory management   tim_one, lemburg
 networking          giampaolo.rodola
 packaging           tarek, lemburg
 py3 transition      benjamin.peterson
-release management  tarek, lemburg, benjamin.peterson, barry, loewis,
+release management  tarek, lemburg, benjamin.peterson, barry,
                     gvanrossum, anthonybaxter
 str.format          eric.smith
 time and dates      lemburg

Modified: python/branches/py3k-cdecimal/Modules/_ctypes/callproc.c
==============================================================================
--- python/branches/py3k-cdecimal/Modules/_ctypes/callproc.c	(original)
+++ python/branches/py3k-cdecimal/Modules/_ctypes/callproc.c	Sat Jul 17 12:36:09 2010
@@ -761,6 +761,7 @@
     ffi_cif cif;
     int cc;
 #ifdef MS_WIN32
+    int delta;
 #ifndef DONT_USE_SEH
     DWORD dwExceptionCode = 0;
     EXCEPTION_RECORD record;
@@ -811,8 +812,9 @@
 #ifndef DONT_USE_SEH
     __try {
 #endif
+        delta =
 #endif
-        ffi_call(&cif, (void *)pProc, resmem, avalues);
+                ffi_call(&cif, (void *)pProc, resmem, avalues);
 #ifdef MS_WIN32
 #ifndef DONT_USE_SEH
     }
@@ -844,6 +846,35 @@
         return -1;
     }
 #endif
+#ifdef MS_WIN64
+    if (delta != 0) {
+        PyErr_Format(PyExc_RuntimeError,
+                 "ffi_call failed with code %d",
+                 delta);
+        return -1;
+    }
+#else
+    if (delta < 0) {
+        if (flags & FUNCFLAG_CDECL)
+            PyErr_Format(PyExc_ValueError,
+                     "Procedure called with not enough "
+                     "arguments (%d bytes missing) "
+                     "or wrong calling convention",
+                     -delta);
+        else
+            PyErr_Format(PyExc_ValueError,
+                     "Procedure probably called with not enough "
+                     "arguments (%d bytes missing)",
+                     -delta);
+        return -1;
+    } else if (delta > 0) {
+        PyErr_Format(PyExc_ValueError,
+                 "Procedure probably called with too many "
+                 "arguments (%d bytes in excess)",
+                 delta);
+        return -1;
+    }
+#endif
 #endif
     if ((flags & FUNCFLAG_PYTHONAPI) && PyErr_Occurred())
         return -1;
@@ -1116,7 +1147,11 @@
     }
     for (i = 0; i < argcount; ++i) {
         atypes[i] = args[i].ffi_type;
-        if (atypes[i]->type == FFI_TYPE_STRUCT)
+        if (atypes[i]->type == FFI_TYPE_STRUCT 
+#ifdef _WIN64
+            && atypes[i]->size <= sizeof(void *)
+#endif
+            )
             avalues[i] = (void *)args[i].value.p;
         else
             avalues[i] = (void *)&args[i].value;

Modified: python/branches/py3k-cdecimal/Modules/_ctypes/libffi/fficonfig.py.in
==============================================================================
--- python/branches/py3k-cdecimal/Modules/_ctypes/libffi/fficonfig.py.in	(original)
+++ python/branches/py3k-cdecimal/Modules/_ctypes/libffi/fficonfig.py.in	Sat Jul 17 12:36:09 2010
@@ -6,7 +6,7 @@
     'MIPS_IRIX': ['src/mips/ffi.c', 'src/mips/o32.S', 'src/mips/n32.S'],
     'MIPS_LINUX': ['src/mips/ffi.c', 'src/mips/o32.S'],
     'X86': ['src/x86/ffi.c', 'src/x86/sysv.S'],
-    'X86_FREEBSD': ['src/x86/ffi.c', 'src/x86/sysv.S'],
+    'X86_FREEBSD': ['src/x86/ffi.c', 'src/x86/freebsd.S'],
     'X86_WIN32': ['src/x86/ffi.c', 'src/x86/win32.S'],
     'SPARC': ['src/sparc/ffi.c', 'src/sparc/v8.S', 'src/sparc/v9.S'],
     'ALPHA': ['src/alpha/ffi.c', 'src/alpha/osf.S'],

Modified: python/branches/py3k-cdecimal/Modules/_ctypes/libffi_msvc/ffi.h
==============================================================================
--- python/branches/py3k-cdecimal/Modules/_ctypes/libffi_msvc/ffi.h	(original)
+++ python/branches/py3k-cdecimal/Modules/_ctypes/libffi_msvc/ffi.h	Sat Jul 17 12:36:09 2010
@@ -1,5 +1,5 @@
 /* -----------------------------------------------------------------*-C-*-
-   libffi @VERSION@ - Copyright (c) 1996-2003, 2007, 2008  Red Hat, Inc.
+   libffi 2.00-beta - Copyright (c) 1996-2003  Red Hat, Inc.
 
    Permission is hereby granted, free of charge, to any person obtaining
    a copy of this software and associated documentation files (the
@@ -12,14 +12,13 @@
    The above copyright notice and this permission notice shall be included
    in all copies or substantial portions of the Software.
 
-   THE SOFTWARE IS PROVIDED ``AS IS'', WITHOUT WARRANTY OF ANY KIND,
-   EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
-   MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
-   NONINFRINGEMENT.  IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
-   HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
-   WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
-   OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
-   DEALINGS IN THE SOFTWARE.
+   THE SOFTWARE IS PROVIDED ``AS IS'', WITHOUT WARRANTY OF ANY KIND, EXPRESS
+   OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+   MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
+   IN NO EVENT SHALL CYGNUS SOLUTIONS BE LIABLE FOR ANY CLAIM, DAMAGES OR
+   OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+   ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+   OTHER DEALINGS IN THE SOFTWARE.
 
    ----------------------------------------------------------------------- */
 
@@ -57,7 +56,7 @@
 #endif
 
 /* Specify which architecture libffi is configured for. */
-/* #define @TARGET@ */
+//XXX #define X86
 
 /* ---- System configuration information --------------------------------- */
 
@@ -65,10 +64,6 @@
 
 #ifndef LIBFFI_ASM
 
-#ifdef _MSC_VER
-#define __attribute__(X)
-#endif
-
 #include <stddef.h>
 #include <limits.h>
 
@@ -84,21 +79,12 @@
 #  ifdef __GNUC__
 #   define FFI_LONG_LONG_MAX __LONG_LONG_MAX__
 #  endif
+#  ifdef _MSC_VER
+#   define FFI_LONG_LONG_MAX _I64_MAX
+#  endif
 # endif
 #endif
 
-/* The closure code assumes that this works on pointers, i.e. a size_t	*/
-/* can hold a pointer.							*/
-
-typedef struct _ffi_type
-{
-  size_t size;
-  unsigned short alignment;
-  unsigned short type;
-  struct _ffi_type **elements;
-} ffi_type;
-
-#ifndef LIBFFI_HIDE_BASIC_TYPES
 #if SCHAR_MAX == 127
 # define ffi_type_uchar                ffi_type_uint8
 # define ffi_type_schar                ffi_type_sint8
@@ -129,23 +115,26 @@
  #error "int size not supported"
 #endif
 
+#define ffi_type_ulong         ffi_type_uint64
+#define ffi_type_slong         ffi_type_sint64
 #if LONG_MAX == 2147483647
 # if FFI_LONG_LONG_MAX != 9223372036854775807
- #error "no 64-bit data type supported"
+  #error "no 64-bit data type supported"
 # endif
 #elif LONG_MAX != 9223372036854775807
  #error "long size not supported"
 #endif
 
-#if LONG_MAX == 2147483647
-# define ffi_type_ulong        ffi_type_uint32
-# define ffi_type_slong        ffi_type_sint32
-#elif LONG_MAX == 9223372036854775807
-# define ffi_type_ulong        ffi_type_uint64
-# define ffi_type_slong        ffi_type_sint64
-#else
- #error "long size not supported"
-#endif
+/* The closure code assumes that this works on pointers, i.e. a size_t	*/
+/* can hold a pointer.							*/
+
+typedef struct _ffi_type
+{
+  size_t size;
+  unsigned short alignment;
+  unsigned short type;
+  /*@null@*/ struct _ffi_type **elements;
+} ffi_type;
 
 /* These are defined in types.c */
 extern ffi_type ffi_type_void;
@@ -159,19 +148,14 @@
 extern ffi_type ffi_type_sint64;
 extern ffi_type ffi_type_float;
 extern ffi_type ffi_type_double;
+extern ffi_type ffi_type_longdouble;
 extern ffi_type ffi_type_pointer;
 
-#if HAVE_LONG_DOUBLE
-extern ffi_type ffi_type_longdouble;
-#else
-#define ffi_type_longdouble ffi_type_double
-#endif
-#endif /* LIBFFI_HIDE_BASIC_TYPES */
 
 typedef enum {
   FFI_OK = 0,
   FFI_BAD_TYPEDEF,
-  FFI_BAD_ABI
+  FFI_BAD_ABI 
 } ffi_status;
 
 typedef unsigned FFI_TYPE;
@@ -179,8 +163,8 @@
 typedef struct {
   ffi_abi abi;
   unsigned nargs;
-  ffi_type **arg_types;
-  ffi_type *rtype;
+  /*@dependent@*/ ffi_type **arg_types;
+  /*@dependent@*/ ffi_type *rtype;
   unsigned bytes;
   unsigned flags;
 #ifdef FFI_EXTRA_CIF_FIELDS
@@ -190,16 +174,10 @@
 
 /* ---- Definitions for the raw API -------------------------------------- */
 
-#ifndef FFI_SIZEOF_ARG
-# if LONG_MAX == 2147483647
-#  define FFI_SIZEOF_ARG        4
-# elif LONG_MAX == 9223372036854775807
-#  define FFI_SIZEOF_ARG        8
-# endif
-#endif
-
-#ifndef FFI_SIZEOF_JAVA_RAW
-#  define FFI_SIZEOF_JAVA_RAW FFI_SIZEOF_ARG
+#ifdef _WIN64
+#define FFI_SIZEOF_ARG 8
+#else
+#define FFI_SIZEOF_ARG 4
 #endif
 
 typedef union {
@@ -210,25 +188,10 @@
   void*     ptr;
 } ffi_raw;
 
-#if FFI_SIZEOF_JAVA_RAW == 4 && FFI_SIZEOF_ARG == 8
-/* This is a special case for mips64/n32 ABI (and perhaps others) where
-   sizeof(void *) is 4 and FFI_SIZEOF_ARG is 8.  */
-typedef union {
-  signed int	sint;
-  unsigned int	uint;
-  float		flt;
-  char		data[FFI_SIZEOF_JAVA_RAW];
-  void*		ptr;
-} ffi_java_raw;
-#else
-typedef ffi_raw ffi_java_raw;
-#endif
-
-
-void ffi_raw_call (ffi_cif *cif,
-		   void (*fn)(void),
-		   void *rvalue,
-		   ffi_raw *avalue);
+void ffi_raw_call (/*@dependent@*/ ffi_cif *cif, 
+		   void (*fn)(), 
+		   /*@out@*/ void *rvalue, 
+		   /*@dependent@*/ ffi_raw *avalue);
 
 void ffi_ptrarray_to_raw (ffi_cif *cif, void **args, ffi_raw *raw);
 void ffi_raw_to_ptrarray (ffi_cif *cif, ffi_raw *raw, void **args);
@@ -238,35 +201,25 @@
 /* packing, even on 64-bit machines.  I.e. on 64-bit machines		*/
 /* longs and doubles are followed by an empty 64-bit word.		*/
 
-void ffi_java_raw_call (ffi_cif *cif,
-			void (*fn)(void),
-			void *rvalue,
-			ffi_java_raw *avalue);
+void ffi_java_raw_call (/*@dependent@*/ ffi_cif *cif, 
+		        void (*fn)(), 
+		        /*@out@*/ void *rvalue, 
+		        /*@dependent@*/ ffi_raw *avalue);
 
-void ffi_java_ptrarray_to_raw (ffi_cif *cif, void **args, ffi_java_raw *raw);
-void ffi_java_raw_to_ptrarray (ffi_cif *cif, ffi_java_raw *raw, void **args);
+void ffi_java_ptrarray_to_raw (ffi_cif *cif, void **args, ffi_raw *raw);
+void ffi_java_raw_to_ptrarray (ffi_cif *cif, ffi_raw *raw, void **args);
 size_t ffi_java_raw_size (ffi_cif *cif);
 
 /* ---- Definitions for closures ----------------------------------------- */
 
 #if FFI_CLOSURES
 
-#ifdef _MSC_VER
-__declspec(align(8))
-#endif
 typedef struct {
   char tramp[FFI_TRAMPOLINE_SIZE];
   ffi_cif   *cif;
   void     (*fun)(ffi_cif*,void*,void**,void*);
   void      *user_data;
-#ifdef __GNUC__
-} ffi_closure __attribute__((aligned (8)));
-#else
 } ffi_closure;
-#endif
-
-void *ffi_closure_alloc (size_t size, void **code);
-void ffi_closure_free (void *);
 
 ffi_status
 ffi_prep_closure (ffi_closure*,
@@ -274,13 +227,6 @@
 		  void (*fun)(ffi_cif*,void*,void**,void*),
 		  void *user_data);
 
-ffi_status
-ffi_prep_closure_loc (ffi_closure*,
-		      ffi_cif *,
-		      void (*fun)(ffi_cif*,void*,void**,void*),
-		      void *user_data,
-		      void*codeloc);
-
 typedef struct {
   char tramp[FFI_TRAMPOLINE_SIZE];
 
@@ -302,27 +248,6 @@
 
 } ffi_raw_closure;
 
-typedef struct {
-  char tramp[FFI_TRAMPOLINE_SIZE];
-
-  ffi_cif   *cif;
-
-#if !FFI_NATIVE_RAW_API
-
-  /* if this is enabled, then a raw closure has the same layout 
-     as a regular closure.  We use this to install an intermediate 
-     handler to do the transaltion, void** -> ffi_raw*. */
-
-  void     (*translate_args)(ffi_cif*,void*,void**,void*);
-  void      *this_closure;
-
-#endif
-
-  void     (*fun)(ffi_cif*,void*,ffi_java_raw*,void*);
-  void      *user_data;
-
-} ffi_java_raw_closure;
-
 ffi_status
 ffi_prep_raw_closure (ffi_raw_closure*,
 		      ffi_cif *cif,
@@ -330,42 +255,29 @@
 		      void *user_data);
 
 ffi_status
-ffi_prep_raw_closure_loc (ffi_raw_closure*,
-			  ffi_cif *cif,
-			  void (*fun)(ffi_cif*,void*,ffi_raw*,void*),
-			  void *user_data,
-			  void *codeloc);
-
-ffi_status
-ffi_prep_java_raw_closure (ffi_java_raw_closure*,
+ffi_prep_java_raw_closure (ffi_raw_closure*,
 		           ffi_cif *cif,
-		           void (*fun)(ffi_cif*,void*,ffi_java_raw*,void*),
+		           void (*fun)(ffi_cif*,void*,ffi_raw*,void*),
 		           void *user_data);
 
-ffi_status
-ffi_prep_java_raw_closure_loc (ffi_java_raw_closure*,
-			       ffi_cif *cif,
-			       void (*fun)(ffi_cif*,void*,ffi_java_raw*,void*),
-			       void *user_data,
-			       void *codeloc);
-
 #endif /* FFI_CLOSURES */
 
 /* ---- Public interface definition -------------------------------------- */
 
-ffi_status ffi_prep_cif(ffi_cif *cif,
+ffi_status ffi_prep_cif(/*@out@*/ /*@partial@*/ ffi_cif *cif, 
 			ffi_abi abi,
-			unsigned int nargs,
-			ffi_type *rtype,
-			ffi_type **atypes);
-
-void ffi_call(ffi_cif *cif,
-	      void (*fn)(void),
-	      void *rvalue,
-	      void **avalue);
+			unsigned int nargs, 
+			/*@dependent@*/ /*@out@*/ /*@partial@*/ ffi_type *rtype, 
+			/*@dependent@*/ ffi_type **atypes);
+
+int
+ffi_call(/*@dependent@*/ ffi_cif *cif, 
+	 void (*fn)(), 
+	 /*@out@*/ void *rvalue, 
+	 /*@dependent@*/ void **avalue);
 
 /* Useful for eliminating compiler warnings */
-#define FFI_FN(f) ((void (*)(void))f)
+#define FFI_FN(f) ((void (*)())f)
 
 /* ---- Definitions shared with assembly code ---------------------------- */
 
@@ -376,7 +288,7 @@
 #define FFI_TYPE_INT        1
 #define FFI_TYPE_FLOAT      2    
 #define FFI_TYPE_DOUBLE     3
-#if HAVE_LONG_DOUBLE
+#if 1
 #define FFI_TYPE_LONGDOUBLE 4
 #else
 #define FFI_TYPE_LONGDOUBLE FFI_TYPE_DOUBLE
@@ -400,3 +312,4 @@
 #endif
 
 #endif
+

Modified: python/branches/py3k-cdecimal/Modules/_ctypes/libffi_msvc/fficonfig.h
==============================================================================
--- python/branches/py3k-cdecimal/Modules/_ctypes/libffi_msvc/fficonfig.h	(original)
+++ python/branches/py3k-cdecimal/Modules/_ctypes/libffi_msvc/fficonfig.h	Sat Jul 17 12:36:09 2010
@@ -1,186 +1,96 @@
 /* fficonfig.h.  Originally created by configure, now hand_maintained for MSVC. */
 
-/* fficonfig.h.in.  Generated from configure.ac by autoheader.  */
+/* fficonfig.h.  Generated automatically by configure.  */
+/* fficonfig.h.in.  Generated automatically from configure.in by autoheader.  */
 
-/* Define if building universal (internal helper macro) */
-/* #undef AC_APPLE_UNIVERSAL_BUILD */
+/* Define this for MSVC, but not for mingw32! */
+#ifdef _MSC_VER
+#define __attribute__(x) /* */
+#endif
+#define alloca _alloca
 
-/* Define to one of `_getb67', `GETB67', `getb67' for Cray-2 and Cray-YMP
-   systems. This function is required for `alloca.c' support on those systems.
-   */
-/* #undef CRAY_STACKSEG_END */
+/*----------------------------------------------------------------*/
 
-/* Define to 1 if using `alloca.c'. */
+/* Define if using alloca.c.  */
 /* #undef C_ALLOCA */
 
-/* Define to the flags needed for the .section .eh_frame directive. */
-/* #undef EH_FRAME_FLAGS */
-
-/* Define this if you want extra debugging. */
-/* #undef FFI_DEBUG */
-
-/* Cannot use malloc on this target, so, we revert to alternative means */
-/* #undef FFI_MMAP_EXEC_WRIT */
-
-/* Define this is you do not want support for the raw API. */
-#define FFI_NO_RAW_API 1
-
-/* Define this is you do not want support for aggregate types. */
-/* #undef FFI_NO_STRUCTS */
+/* Define to one of _getb67, GETB67, getb67 for Cray-2 and Cray-YMP systems.
+   This function is required for alloca.c support on those systems.  */
+/* #undef CRAY_STACKSEG_END */
 
-/* Define to 1 if you have `alloca', as a function or macro. */
+/* Define if you have alloca, as a function or macro.  */
 #define HAVE_ALLOCA 1
 
-/* Define to 1 if you have <alloca.h> and it should be used (not on Ultrix).
-   */
-/* #undef HAVE_ALLOCA_H */
-
-/* Define if your assembler supports .cfi_* directives. */
-/* #undef HAVE_AS_CFI_PSEUDO_OP */
+/* Define if you have <alloca.h> and it should be used (not on Ultrix).  */
+/* #define HAVE_ALLOCA_H 1 */
 
-/* Define if your assembler supports .register. */
-/* #undef HAVE_AS_REGISTER_PSEUDO_OP */
-
-/* Define if your assembler and linker support unaligned PC relative relocs.
-   */
-/* #undef HAVE_AS_SPARC_UA_PCREL */
-
-/* Define if your assembler supports PC relative relocs. */
-/* #undef HAVE_AS_X86_PCREL */
-
-/* Define to 1 if you have the <dlfcn.h> header file. */
-/* #undef HAVE_DLFCN_H */
-
-/* Define if __attribute__((visibility("hidden"))) is supported. */
-/* #undef HAVE_HIDDEN_VISIBILITY_ATTRIBUTE */
-
-/* Define to 1 if you have the <inttypes.h> header file. */
-/* #undef HAVE_INTTYPES_H */
+/* If using the C implementation of alloca, define if you know the
+   direction of stack growth for your system; otherwise it will be
+   automatically deduced at run-time.
+ STACK_DIRECTION > 0 => grows toward higher addresses
+ STACK_DIRECTION < 0 => grows toward lower addresses
+ STACK_DIRECTION = 0 => direction of growth unknown
+ */
+/* #undef STACK_DIRECTION */
 
-/* Define if you have the long double type and it is bigger than a double */
-/* #undef HAVE_LONG_DOUBLE */
+/* Define if you have the ANSI C header files.  */
+#define STDC_HEADERS 1
 
-/* Define to 1 if you have the `memcpy' function. */
+/* Define if you have the memcpy function.  */
 #define HAVE_MEMCPY 1
 
-/* Define to 1 if you have the <memory.h> header file. */
-/* #undef HAVE_MEMORY_H */
-
-/* Define to 1 if you have the `mmap' function. */
-/* #undef HAVE_MMAP */
-
-/* Define if mmap with MAP_ANON(YMOUS) works. */
-/* #undef HAVE_MMAP_ANON */
-
-/* Define if mmap of /dev/zero works. */
-/* #undef HAVE_MMAP_DEV_ZERO */
-
 /* Define if read-only mmap of a plain file works. */
-/* #undef HAVE_MMAP_FILE */
-
-/* Define if .eh_frame sections should be read-only. */
-/* #undef HAVE_RO_EH_FRAME */
-
-/* Define to 1 if you have the <stdint.h> header file. */
-#undef HAVE_STDINT_H
-
-/* Define to 1 if you have the <stdlib.h> header file. */
-#undef HAVE_STDLIB_H
-
-/* Define to 1 if you have the <strings.h> header file. */
-#undef HAVE_STRINGS_H
-
-/* Define to 1 if you have the <string.h> header file. */
-#undef HAVE_STRING_H
+//#define HAVE_MMAP_FILE 1
 
-/* Define to 1 if you have the <sys/mman.h> header file. */
-/* #undef HAVE_SYS_MMAN_H */
-
-/* Define to 1 if you have the <sys/stat.h> header file. */
-/* #undef HAVE_SYS_STAT_H */
+/* Define if mmap of /dev/zero works. */
+//#define HAVE_MMAP_DEV_ZERO 1
 
-/* Define to 1 if you have the <sys/types.h> header file. */
-/* #undef HAVE_SYS_TYPES_H */
+/* Define if mmap with MAP_ANON(YMOUS) works. */
+//#define HAVE_MMAP_ANON 1
 
-/* Define to 1 if you have the <unistd.h> header file. */
-/* #undef HAVE_UNISTD_H */
+/* The number of bytes in type double */
+#define SIZEOF_DOUBLE 8
 
-/* Define to the sub-directory in which libtool stores uninstalled libraries.
-   */
-/* #undef LT_OBJDIR */
+/* The number of bytes in type long double */
+#define SIZEOF_LONG_DOUBLE 12
 
-/* Define to 1 if your C compiler doesn't accept -c and -o together. */
-/* #undef NO_MINUS_C_MINUS_O */
+/* Define if you have the long double type and it is bigger than a double */
+#define HAVE_LONG_DOUBLE 1
 
-/* Name of package */
-/* #undef PACKAGE */
+/* whether byteorder is bigendian */
+/* #undef WORDS_BIGENDIAN */
 
-/* Define to the address where bug reports for this package should be sent. */
-/* #undef PACKAGE_BUGREPORT */
+/* Define if the host machine stores words of multi-word integers in
+   big-endian order. */
+/* #undef HOST_WORDS_BIG_ENDIAN */
 
-/* Define to the full name of this package. */
-/* #undef PACKAGE_NAME */
+/* 1234 = LIL_ENDIAN, 4321 = BIGENDIAN */
+#define BYTEORDER 1234
 
-/* Define to the full name and version of this package. */
-/* #undef PACKAGE_STRING */
+/* Define if your assembler and linker support unaligned PC relative relocs. */
+/* #undef HAVE_AS_SPARC_UA_PCREL */
 
-/* Define to the one symbol short name of this package. */
-/* #undef PACKAGE_TARNAME */
+/* Define if your assembler supports .register. */
+/* #undef HAVE_AS_REGISTER_PSEUDO_OP */
 
-/* Define to the home page for this package. */
-/* #undef PACKAGE_URL */
+/* Define if .eh_frame sections should be read-only. */
+/* #undef HAVE_RO_EH_FRAME */
 
-/* Define to the version of this package. */
-/* #undef PACKAGE_VERSION */
+/* Define to the flags needed for the .section .eh_frame directive. */
+/* #define EH_FRAME_FLAGS "aw" */
 
-/* The size of `double', as computed by sizeof. */
-#define SIZEOF_DOUBLE 8
+/* Define to the flags needed for the .section .eh_frame directive. */
+/* #define EH_FRAME_FLAGS "aw" */
 
-/* The size of `long double', as computed by sizeof. */
-#undef SIZEOF_LONG_DOUBLE
+/* Define this if you want extra debugging. */
+/* #undef FFI_DEBUG */
 
-/* If using the C implementation of alloca, define if you know the
-   direction of stack growth for your system; otherwise it will be
-   automatically deduced at runtime.
-	STACK_DIRECTION > 0 => grows toward higher addresses
-	STACK_DIRECTION < 0 => grows toward lower addresses
-	STACK_DIRECTION = 0 => direction of growth unknown */
-/* #undef STACK_DIRECTION */
+/* Define this is you do not want support for aggregate types. */
+/* #undef FFI_NO_STRUCTS */
 
-/* Define to 1 if you have the ANSI C header files. */
-#define STDC_HEADERS 1
+/* Define this is you do not want support for the raw API. */
+/* #undef FFI_NO_RAW_API */
 
-/* Define this if you are using Purify and want to suppress spurious messages.
-   */
+/* Define this if you are using Purify and want to suppress spurious messages. */
 /* #undef USING_PURIFY */
 
-/* Version number of package */
-/* #undef VERSION */
-
-/* Define WORDS_BIGENDIAN to 1 if your processor stores words with the most
-   significant byte first (like Motorola and SPARC, unlike Intel). */
-#if defined AC_APPLE_UNIVERSAL_BUILD
-# if defined __BIG_ENDIAN__
-#  define WORDS_BIGENDIAN 1
-# endif
-#else
-# ifndef WORDS_BIGENDIAN
-#  undef WORDS_BIGENDIAN
-# endif
-#endif
-
-
-#ifdef HAVE_HIDDEN_VISIBILITY_ATTRIBUTE
-#ifdef LIBFFI_ASM
-#define FFI_HIDDEN(name) .hidden name
-#else
-#define FFI_HIDDEN __attribute__ ((visibility ("hidden")))
-#endif
-#else
-#ifdef LIBFFI_ASM
-#define FFI_HIDDEN(name)
-#else
-#define FFI_HIDDEN
-#endif
-#endif
-

Modified: python/branches/py3k-cdecimal/Modules/_io/_iomodule.c
==============================================================================
--- python/branches/py3k-cdecimal/Modules/_io/_iomodule.c	(original)
+++ python/branches/py3k-cdecimal/Modules/_io/_iomodule.c	Sat Jul 17 12:36:09 2010
@@ -176,7 +176,7 @@
  * The main open() function
  */
 PyDoc_STRVAR(open_doc,
-"open(file, mode='r', buffering=None, encoding=None,\n"
+"open(file, mode='r', buffering=-1, encoding=None,\n"
 "     errors=None, newline=None, closefd=True) -> file object\n"
 "\n"
 "Open file and return a stream.  Raise IOError upon failure.\n"

Modified: python/branches/py3k-cdecimal/Modules/_math.c
==============================================================================
--- python/branches/py3k-cdecimal/Modules/_math.c	(original)
+++ python/branches/py3k-cdecimal/Modules/_math.c	Sat Jul 17 12:36:09 2010
@@ -55,14 +55,15 @@
     else if (x >= two_pow_p28) {        /* x > 2**28 */
         if (Py_IS_INFINITY(x)) {
             return x+x;
-        } else {
-            return log(x)+ln2;                  /* acosh(huge)=log(2x) */
+        }
+        else {
+            return log(x)+ln2;          /* acosh(huge)=log(2x) */
         }
     }
     else if (x == 1.) {
-        return 0.0;                             /* acosh(1) = 0 */
+        return 0.0;                     /* acosh(1) = 0 */
     }
-    else if (x > 2.) {                          /* 2 < x < 2**28 */
+    else if (x > 2.) {                  /* 2 < x < 2**28 */
         double t = x*x;
         return log(2.0*x - 1.0 / (x + sqrt(t - 1.0)));
     }
@@ -94,7 +95,7 @@
         return x+x;
     }
     if (absx < two_pow_m28) {           /* |x| < 2**-28 */
-        return x;               /* return x inexact except 0 */
+        return x;                       /* return x inexact except 0 */
     }
     if (absx > two_pow_p28) {           /* |x| > 2**28 */
         w = log(absx)+ln2;
@@ -114,9 +115,9 @@
  * Method :
  *    1.Reduced x to positive by atanh(-x) = -atanh(x)
  *    2.For x>=0.5
- *                1           2x                          x
- *      atanh(x) = --- * log(1 + -------) = 0.5 * log1p(2 * --------)
- *                2          1 - x                    1 - x
+ *                  1              2x                          x
+ *      atanh(x) = --- * log(1 + -------) = 0.5 * log1p(2 * -------)
+ *                  2             1 - x                      1 - x
  *
  *      For x<0.5
  *      atanh(x) = 0.5*log1p(2x+2x*x/(1-x))
@@ -173,15 +174,15 @@
     */
 
     if (fabs(x) < 0.7) {
-    double u;
-    u = exp(x);
-    if (u == 1.0)
-        return x;
-    else
-        return (u - 1.0) * x / log(u);
+        double u;
+        u = exp(x);
+        if (u == 1.0)
+            return x;
+        else
+            return (u - 1.0) * x / log(u);
     }
     else
-    return exp(x) - 1.0;
+        return exp(x) - 1.0;
 }
 
 /* log1p(x) = log(1+x).  The log1p function is designed to avoid the
@@ -194,7 +195,7 @@
     /* For x small, we use the following approach.  Let y be the nearest float
        to 1+x, then
 
-      1+x = y * (1 - (y-1-x)/y)
+         1+x = y * (1 - (y-1-x)/y)
 
        so log(1+x) = log(y) + log(1-(y-1-x)/y).  Since (y-1-x)/y is tiny, the
        second term is well approximated by (y-1-x)/y.  If abs(x) >=
@@ -213,17 +214,19 @@
 
     double y;
     if (fabs(x) < DBL_EPSILON/2.) {
-    return x;
-    } else if (-0.5 <= x && x <= 1.) {
-    /* WARNING: it's possible than an overeager compiler
-       will incorrectly optimize the following two lines
-       to the equivalent of "return log(1.+x)". If this
-       happens, then results from log1p will be inaccurate
-       for small x. */
-    y = 1.+x;
-    return log(y)-((y-1.)-x)/y;
-    } else {
-    /* NaNs and infinities should end up here */
-    return log(1.+x);
+        return x;
+    }
+    else if (-0.5 <= x && x <= 1.) {
+        /* WARNING: it's possible than an overeager compiler
+           will incorrectly optimize the following two lines
+           to the equivalent of "return log(1.+x)". If this
+           happens, then results from log1p will be inaccurate
+           for small x. */
+        y = 1.+x;
+        return log(y)-((y-1.)-x)/y;
+    }
+    else {
+        /* NaNs and infinities should end up here */
+        return log(1.+x);
     }
 }

Modified: python/branches/py3k-cdecimal/Modules/_scproxy.c
==============================================================================
--- python/branches/py3k-cdecimal/Modules/_scproxy.c	(original)
+++ python/branches/py3k-cdecimal/Modules/_scproxy.c	Sat Jul 17 12:36:09 2010
@@ -75,7 +75,7 @@
         aNum = CFDictionaryGetValue(proxyDict,
             kSCPropNetProxiesExcludeSimpleHostnames);
         if (aNum == NULL) {
-            v = PyBool_FromLong(1);
+            v = PyBool_FromLong(0);
         } else {
             v = PyBool_FromLong(cfnum_to_int32(aNum));
         }

Modified: python/branches/py3k-cdecimal/Modules/_ssl.c
==============================================================================
--- python/branches/py3k-cdecimal/Modules/_ssl.c	(original)
+++ python/branches/py3k-cdecimal/Modules/_ssl.c	Sat Jul 17 12:36:09 2010
@@ -212,8 +212,11 @@
                     errstr = "EOF occurred in violation of protocol";
                 } else if (ret == -1) {
                     /* underlying BIO reported an I/O error */
+                    Py_INCREF(s);
                     ERR_clear_error();
-                    return s->errorhandler();
+                    v = s->errorhandler();
+                    Py_DECREF(s);
+                    return v;
                 } else { /* possible? */
                     p = PY_SSL_ERROR_SYSCALL;
                     errstr = "Some I/O error occurred";
@@ -334,6 +337,7 @@
                      PY_SSL_ERROR_NO_SOCKET, __FILE__, __LINE__);
         return NULL;
     }
+    Py_INCREF(sock);
 
     /* just in case the blocking state of the socket has been changed */
     nonblocking = (sock->sock_timeout >= 0.0);
@@ -348,9 +352,8 @@
         ret = SSL_do_handshake(self->ssl);
         err = SSL_get_error(self->ssl, ret);
         PySSL_END_ALLOW_THREADS
-        if(PyErr_CheckSignals()) {
-            return NULL;
-        }
+        if (PyErr_CheckSignals())
+            goto error;
         if (err == SSL_ERROR_WANT_READ) {
             sockstate = check_socket_and_wait_for_timeout(sock, 0);
         } else if (err == SSL_ERROR_WANT_WRITE) {
@@ -361,22 +364,22 @@
         if (sockstate == SOCKET_HAS_TIMED_OUT) {
             PyErr_SetString(PySSLErrorObject,
                             ERRSTR("The handshake operation timed out"));
-            return NULL;
+            goto error;
         } else if (sockstate == SOCKET_HAS_BEEN_CLOSED) {
             PyErr_SetString(PySSLErrorObject,
                             ERRSTR("Underlying socket has been closed."));
-            return NULL;
+            goto error;
         } else if (sockstate == SOCKET_TOO_LARGE_FOR_SELECT) {
             PyErr_SetString(PySSLErrorObject,
                             ERRSTR("Underlying socket too large for select()."));
-            return NULL;
+            goto error;
         } else if (sockstate == SOCKET_IS_NONBLOCKING) {
             break;
         }
     } while (err == SSL_ERROR_WANT_READ || err == SSL_ERROR_WANT_WRITE);
+    Py_DECREF(sock);
     if (ret < 1)
         return PySSL_SetError(self, ret, __FILE__, __LINE__);
-    self->ssl->debug = 1;
 
     if (self->peer_cert)
         X509_free (self->peer_cert);
@@ -386,6 +389,10 @@
 
     Py_INCREF(Py_None);
     return Py_None;
+
+error:
+    Py_DECREF(sock);
+    return NULL;
 }
 
 static PyObject *
@@ -1051,9 +1058,12 @@
                      PY_SSL_ERROR_NO_SOCKET, __FILE__, __LINE__);
         return NULL;
     }
+    Py_INCREF(sock);
 
-    if (!PyArg_ParseTuple(args, "y*:write", &buf))
+    if (!PyArg_ParseTuple(args, "y*:write", &buf)) {
+        Py_DECREF(sock);
         return NULL;
+    }
 
     /* just in case the blocking state of the socket has been changed */
     nonblocking = (sock->sock_timeout >= 0.0);
@@ -1103,6 +1113,7 @@
         }
     } while (err == SSL_ERROR_WANT_READ || err == SSL_ERROR_WANT_WRITE);
 
+    Py_DECREF(sock);
     PyBuffer_Release(&buf);
     if (len > 0)
         return PyLong_FromLong(len);
@@ -1110,6 +1121,7 @@
         return PySSL_SetError(self, len, __FILE__, __LINE__);
 
 error:
+    Py_DECREF(sock);
     PyBuffer_Release(&buf);
     return NULL;
 }
@@ -1159,21 +1171,23 @@
                      PY_SSL_ERROR_NO_SOCKET, __FILE__, __LINE__);
         return NULL;
     }
+    Py_INCREF(sock);
 
     if (!PyArg_ParseTuple(args, "|Oi:read", &dest, &count))
-        return NULL;
+        goto error;
+
     if ((dest == NULL) || (dest == Py_None)) {
         if (!(dest = PyByteArray_FromStringAndSize((char *) 0, len)))
-            return NULL;
+            goto error;
         mem = PyByteArray_AS_STRING(dest);
     } else if (PyLong_Check(dest)) {
         len = PyLong_AS_LONG(dest);
         if (!(dest = PyByteArray_FromStringAndSize((char *) 0, len)))
-            return NULL;
+            goto error;
         mem = PyByteArray_AS_STRING(dest);
     } else {
         if (PyObject_GetBuffer(dest, &buf, PyBUF_CONTIG) < 0)
-            return NULL;
+            goto error;
         mem = buf.buf;
         len = buf.len;
         if ((count > 0) && (count <= len))
@@ -1240,6 +1254,7 @@
         goto error;
     }
   done:
+    Py_DECREF(sock);
     if (!buf_passed) {
         PyObject *res = PyBytes_FromStringAndSize(mem, count);
         Py_DECREF(dest);
@@ -1249,8 +1264,9 @@
         return PyLong_FromLong(count);
     }
   error:
+    Py_DECREF(sock);
     if (!buf_passed) {
-        Py_DECREF(dest);
+        Py_XDECREF(dest);
     } else {
         PyBuffer_Release(&buf);
     }
@@ -1275,6 +1291,7 @@
                      PY_SSL_ERROR_NO_SOCKET, __FILE__, __LINE__);
         return NULL;
     }
+    Py_INCREF(sock);
 
     /* Just in case the blocking state of the socket has been changed */
     nonblocking = (sock->sock_timeout >= 0.0);
@@ -1324,24 +1341,29 @@
             else
                 PyErr_SetString(PySSLErrorObject,
                                 "The write operation timed out");
-            return NULL;
+            goto error;
         }
         else if (sockstate == SOCKET_TOO_LARGE_FOR_SELECT) {
             PyErr_SetString(PySSLErrorObject,
                             "Underlying socket too large for select().");
-            return NULL;
+            goto error;
         }
         else if (sockstate != SOCKET_OPERATION_OK)
             /* Retain the SSL error code */
             break;
     }
 
-    if (err < 0)
+    if (err < 0) {
+        Py_DECREF(sock);
         return PySSL_SetError(self, err, __FILE__, __LINE__);
-    else {
-        Py_INCREF(sock);
-        return (PyObject *) sock;
     }
+    else
+        /* It's already INCREF'ed */
+        return (PyObject *) sock;
+
+error:
+    Py_DECREF(sock);
+    return NULL;
 }
 
 PyDoc_STRVAR(PySSL_SSLshutdown_doc,

Modified: python/branches/py3k-cdecimal/Modules/_struct.c
==============================================================================
--- python/branches/py3k-cdecimal/Modules/_struct.c	(original)
+++ python/branches/py3k-cdecimal/Modules/_struct.c	Sat Jul 17 12:36:09 2010
@@ -591,9 +591,13 @@
 static int
 np_bool(char *p, PyObject *v, const formatdef *f)
 {
-    BOOL_TYPE y;
+    int y;
+    BOOL_TYPE x;
     y = PyObject_IsTrue(v);
-    memcpy(p, (char *)&y, sizeof y);
+    if (y < 0)
+        return -1;
+    x = y;
+    memcpy(p, (char *)&x, sizeof x);
     return 0;
 }
 
@@ -865,6 +869,8 @@
 {
     char y;
     y = PyObject_IsTrue(v);
+    if (y < 0)
+        return -1;
     memcpy(p, (char *)&y, sizeof y);
     return 0;
 }

Modified: python/branches/py3k-cdecimal/Modules/_testcapimodule.c
==============================================================================
--- python/branches/py3k-cdecimal/Modules/_testcapimodule.c	(original)
+++ python/branches/py3k-cdecimal/Modules/_testcapimodule.c	Sat Jul 17 12:36:09 2010
@@ -1386,6 +1386,28 @@
 }
 
 static PyObject *
+getargs_w_star(PyObject *self, PyObject *args)
+{
+    Py_buffer buffer;
+    PyObject *result;
+    char *str;
+
+    if (!PyArg_ParseTuple(args, "w*:getargs_w_star", &buffer))
+        return NULL;
+
+    if (2 <= buffer.len) {
+        str = buffer.buf;
+        str[0] = '[';
+        str[buffer.len-1] = ']';
+    }
+
+    result = PyBytes_FromStringAndSize(buffer.buf, buffer.len);
+    PyBuffer_Release(&buffer);
+    return result;
+}
+
+
+static PyObject *
 test_empty_argparse(PyObject *self)
 {
     /* Test that formats can begin with '|'. See issue #4720. */
@@ -2227,6 +2249,7 @@
     {"getargs_u_hash",          getargs_u_hash,                  METH_VARARGS},
     {"getargs_Z",               getargs_Z,                       METH_VARARGS},
     {"getargs_Z_hash",          getargs_Z_hash,                  METH_VARARGS},
+    {"getargs_w_star",          getargs_w_star,                  METH_VARARGS},
     {"codec_incrementalencoder",
      (PyCFunction)codec_incrementalencoder,      METH_VARARGS},
     {"codec_incrementaldecoder",

Modified: python/branches/py3k-cdecimal/Modules/audioop.c
==============================================================================
--- python/branches/py3k-cdecimal/Modules/audioop.c	(original)
+++ python/branches/py3k-cdecimal/Modules/audioop.c	Sat Jul 17 12:36:09 2010
@@ -297,6 +297,29 @@
 
 static PyObject *AudioopError;
 
+static int
+audioop_check_size(int size)
+{
+    if (size != 1 && size != 2 && size != 4) {
+        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
+        return 0;
+    }
+    else
+        return 1;
+}
+
+static int
+audioop_check_parameters(int len, int size)
+{
+    if (!audioop_check_size(size))
+        return 0;
+    if (len % size != 0) {
+        PyErr_SetString(AudioopError, "not a whole number of frames");
+        return 0;
+    }
+    return 1;
+}
+
 static PyObject *
 audioop_getsample(PyObject *self, PyObject *args)
 {
@@ -306,10 +329,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#in:getsample", &cp, &len, &size, &i) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     if ( i < 0 || i >= len/size ) {
         PyErr_SetString(AudioopError, "Index out of range");
         return 0;
@@ -330,10 +351,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:max", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     for ( i=0; i<len; i+= size) {
         if ( size == 1 )      val = (int)*CHARP(cp, i);
         else if ( size == 2 ) val = (int)*SHORTP(cp, i);
@@ -354,10 +373,8 @@
 
     if (!PyArg_ParseTuple(args, "s#i:minmax", &cp, &len, &size))
         return NULL;
-    if (size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
+    if (!audioop_check_parameters(len, size))
         return NULL;
-    }
     for (i = 0; i < len; i += size) {
         if (size == 1) val = (int) *CHARP(cp, i);
         else if (size == 2) val = (int) *SHORTP(cp, i);
@@ -378,10 +395,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:avg", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     for ( i=0; i<len; i+= size) {
         if ( size == 1 )      val = (int)*CHARP(cp, i);
         else if ( size == 2 ) val = (int)*SHORTP(cp, i);
@@ -405,10 +420,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:rms", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     for ( i=0; i<len; i+= size) {
         if ( size == 1 )      val = (int)*CHARP(cp, i);
         else if ( size == 2 ) val = (int)*SHORTP(cp, i);
@@ -616,10 +629,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:avgpp", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     /* Compute first delta value ahead. Also automatically makes us
     ** skip the first extreme value
     */
@@ -673,10 +684,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:maxpp", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     /* Compute first delta value ahead. Also automatically makes us
     ** skip the first extreme value
     */
@@ -725,10 +734,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#i:cross", &cp, &len, &size) )
         return 0;
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
     ncross = -1;
     prevval = 17; /* Anything <> 0,1 */
     for ( i=0; i<len; i+= size) {
@@ -753,6 +760,8 @@
 
     if ( !PyArg_ParseTuple(args, "s#id:mul", &cp, &len, &size, &factor ) )
         return 0;
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     if ( size == 1 ) maxval = (double) 0x7f;
     else if ( size == 2 ) maxval = (double) 0x7fff;
@@ -798,6 +807,15 @@
         return 0;
     cp = pcp.buf;
     len = pcp.len;
+    if (!audioop_check_parameters(len, size)) {
+        PyBuffer_Release(&pcp);
+        return NULL;
+    }
+    if (((len / size) & 1) != 0) {
+        PyErr_SetString(AudioopError, "not a whole number of frames");
+        PyBuffer_Release(&pcp);
+        return NULL;
+    }
 
     if ( size == 1 ) maxval = (double) 0x7f;
     else if ( size == 2 ) maxval = (double) 0x7fff;
@@ -809,8 +827,10 @@
     }
 
     rv = PyBytes_FromStringAndSize(NULL, len/2);
-    if ( rv == 0 )
+    if ( rv == 0 ) {
+        PyBuffer_Release(&pcp);
         return 0;
+    }
     ncp = (signed char *)PyBytes_AsString(rv);
 
 
@@ -845,6 +865,8 @@
     if ( !PyArg_ParseTuple(args, "s#idd:tostereo",
                            &cp, &len, &size, &fac1, &fac2 ) )
         return 0;
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     if ( size == 1 ) maxval = (double) 0x7f;
     else if ( size == 2 ) maxval = (double) 0x7fff;
@@ -903,7 +925,8 @@
     if ( !PyArg_ParseTuple(args, "s#s#i:add",
                       &cp1, &len1, &cp2, &len2, &size ) )
         return 0;
-
+    if (!audioop_check_parameters(len1, size))
+        return NULL;
     if ( len1 != len2 ) {
         PyErr_SetString(AudioopError, "Lengths should be the same");
         return 0;
@@ -958,10 +981,8 @@
                       &cp, &len, &size , &bias) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     rv = PyBytes_FromStringAndSize(NULL, len);
     if ( rv == 0 )
@@ -994,10 +1015,8 @@
                       &cp, &len, &size) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4 ) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     rv = PyBytes_FromStringAndSize(NULL, len);
     if ( rv == 0 )
@@ -1031,11 +1050,10 @@
                       &cp, &len, &size, &size2) )
         return 0;
 
-    if ( (size != 1 && size != 2 && size != 4) ||
-         (size2 != 1 && size2 != 2 && size2 != 4)) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
+    if (!audioop_check_size(size2))
+        return NULL;
 
     if (len/size > PY_SSIZE_T_MAX/size2) {
         PyErr_SetString(PyExc_MemoryError,
@@ -1086,10 +1104,8 @@
                           &nchannels, &inrate, &outrate, &state,
                           &weightA, &weightB))
         return NULL;
-    if (size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
+    if (!audioop_check_size(size))
         return NULL;
-    }
     if (nchannels < 1) {
         PyErr_SetString(AudioopError, "# of channels should be >= 1");
         return NULL;
@@ -1265,10 +1281,8 @@
                            &cp, &len, &size) )
         return 0 ;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     rv = PyBytes_FromStringAndSize(NULL, len/size);
     if ( rv == 0 )
@@ -1299,10 +1313,8 @@
                            &cp, &len, &size) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     if (len > PY_SSIZE_T_MAX/size) {
         PyErr_SetString(PyExc_MemoryError,
@@ -1338,10 +1350,8 @@
                            &cp, &len, &size) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     rv = PyBytes_FromStringAndSize(NULL, len/size);
     if ( rv == 0 )
@@ -1372,10 +1382,8 @@
                            &cp, &len, &size) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     if (len > PY_SSIZE_T_MAX/size) {
         PyErr_SetString(PyExc_MemoryError,
@@ -1413,11 +1421,8 @@
                            &cp, &len, &size, &state) )
         return 0;
 
-
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     str = PyBytes_FromStringAndSize(NULL, len/(size*2));
     if ( str == 0 )
@@ -1522,10 +1527,8 @@
                            &cp, &len, &size, &state) )
         return 0;
 
-    if ( size != 1 && size != 2 && size != 4) {
-        PyErr_SetString(AudioopError, "Size should be 1, 2 or 4");
-        return 0;
-    }
+    if (!audioop_check_parameters(len, size))
+        return NULL;
 
     /* Decode state, should have (value, step) */
     if ( state == Py_None ) {

Modified: python/branches/py3k-cdecimal/Modules/cmathmodule.c
==============================================================================
--- python/branches/py3k-cdecimal/Modules/cmathmodule.c	(original)
+++ python/branches/py3k-cdecimal/Modules/cmathmodule.c	Sat Jul 17 12:36:09 2010
@@ -1024,6 +1024,19 @@
 Convert from polar coordinates to rectangular coordinates.");
 
 static PyObject *
+cmath_isfinite(PyObject *self, PyObject *args)
+{
+    Py_complex z;
+    if (!PyArg_ParseTuple(args, "D:isfinite", &z))
+        return NULL;
+    return PyBool_FromLong(Py_IS_FINITE(z.real) && Py_IS_FINITE(z.imag));
+}
+
+PyDoc_STRVAR(cmath_isfinite_doc,
+"isfinite(z) -> bool\n\
+Return True if both the real and imaginary parts of z are finite, else False.");
+
+static PyObject *
 cmath_isnan(PyObject *self, PyObject *args)
 {
     Py_complex z;
@@ -1065,6 +1078,7 @@
     {"cos",    cmath_cos,   METH_VARARGS, c_cos_doc},
     {"cosh",   cmath_cosh,  METH_VARARGS, c_cosh_doc},
     {"exp",    cmath_exp,   METH_VARARGS, c_exp_doc},
+    {"isfinite", cmath_isfinite, METH_VARARGS, cmath_isfinite_doc},
     {"isinf",  cmath_isinf, METH_VARARGS, cmath_isinf_doc},
     {"isnan",  cmath_isnan, METH_VARARGS, cmath_isnan_doc},
     {"log",    cmath_log,   METH_VARARGS, cmath_log_doc},

Modified: python/branches/py3k-cdecimal/Modules/datetimemodule.c
==============================================================================
--- python/branches/py3k-cdecimal/Modules/datetimemodule.c	(original)
+++ python/branches/py3k-cdecimal/Modules/datetimemodule.c	Sat Jul 17 12:36:09 2010
@@ -88,8 +88,11 @@
 /* p is a pointer to a time or a datetime object; HASTZINFO(p) returns
  * p->hastzinfo.
  */
-#define HASTZINFO(p)            (((_PyDateTime_BaseTZInfo *)(p))->hastzinfo)
-
+#define HASTZINFO(p) (((_PyDateTime_BaseTZInfo *)(p))->hastzinfo)
+#define GET_TIME_TZINFO(p) (HASTZINFO(p) ? \
+                            ((PyDateTime_Time *)(p))->tzinfo : Py_None)
+#define GET_DT_TZINFO(p) (HASTZINFO(p) ? \
+                          ((PyDateTime_DateTime *)(p))->tzinfo : Py_None)
 /* M is a char or int claiming to be a valid month.  The macro is equivalent
  * to the two-sided Python test
  *      1 <= M <= 12
@@ -102,6 +105,7 @@
 static PyTypeObject PyDateTime_DeltaType;
 static PyTypeObject PyDateTime_TimeType;
 static PyTypeObject PyDateTime_TZInfoType;
+static PyTypeObject PyDateTime_TimeZoneType;
 
 /* ---------------------------------------------------------------------------
  * Math utilities.
@@ -771,6 +775,54 @@
 #define new_delta(d, s, us, normalize)  \
     new_delta_ex(d, s, us, normalize, &PyDateTime_DeltaType)
 
+
+typedef struct
+{
+    PyObject_HEAD
+    PyObject *offset;
+    PyObject *name;
+} PyDateTime_TimeZone;
+
+PyObject *PyDateTime_TimeZone_UTC;
+
+/* Create new timezone instance checking offset range.  This
+   function does not check the name argument.  Caller must assure
+   that offset is a timedelta instance and name is either NULL
+   or a unicode object. */
+static PyObject *
+new_timezone(PyObject *offset, PyObject *name)
+{
+    PyDateTime_TimeZone *self;
+    PyTypeObject *type = &PyDateTime_TimeZoneType;
+
+    assert(offset != NULL);
+    assert(PyDelta_Check(offset));
+    assert(name == NULL || PyUnicode_Check(name));
+
+    if (GET_TD_MICROSECONDS(offset) != 0 || GET_TD_SECONDS(offset) % 60 != 0) {
+        PyErr_Format(PyExc_ValueError, "offset must be a timedelta"
+                     " representing a whole number of minutes");
+        return NULL;
+    }
+    if ((GET_TD_DAYS(offset) == -1 && GET_TD_SECONDS(offset) == 0) ||
+        GET_TD_DAYS(offset) < -1 || GET_TD_DAYS(offset) >= 1) {
+        PyErr_Format(PyExc_ValueError, "offset must be a timedelta"
+                     " strictly between -timedelta(hours=24) and"
+                     " timedelta(hours=24).");
+        return NULL;
+    }
+
+    self = (PyDateTime_TimeZone *)(type->tp_alloc(type, 0));
+    if (self == NULL) {
+        return NULL;
+    }
+    Py_INCREF(offset);
+    self->offset = offset;
+    Py_XINCREF(name);
+    self->name = name;
+    return (PyObject *)self;
+}
+
 /* ---------------------------------------------------------------------------
  * tzinfo helpers.
  */
@@ -790,25 +842,6 @@
     return -1;
 }
 
-/* Return tzinfo.methname(tzinfoarg), without any checking of results.
- * If tzinfo is None, returns None.
- */
-static PyObject *
-call_tzinfo_method(PyObject *tzinfo, char *methname, PyObject *tzinfoarg)
-{
-    PyObject *result;
-
-    assert(tzinfo && methname && tzinfoarg);
-    assert(check_tzinfo_subclass(tzinfo) >= 0);
-    if (tzinfo == Py_None) {
-        result = Py_None;
-        Py_INCREF(result);
-    }
-    else
-        result = PyObject_CallMethod(tzinfo, methname, "O", tzinfoarg);
-    return result;
-}
-
 /* If self has a tzinfo member, return a BORROWED reference to it.  Else
  * return NULL, which is NOT AN ERROR.  There are no error returns here,
  * and the caller must not decref the result.
@@ -826,69 +859,53 @@
     return tzinfo;
 }
 
-/* Call getattr(tzinfo, name)(tzinfoarg), and extract an int from the
- * result.  tzinfo must be an instance of the tzinfo class.  If the method
- * returns None, this returns 0 and sets *none to 1.  If the method doesn't
- * return None or timedelta, TypeError is raised and this returns -1.  If it
- * returnsa timedelta and the value is out of range or isn't a whole number
- * of minutes, ValueError is raised and this returns -1.
- * Else *none is set to 0 and the integer method result is returned.
+/* Call getattr(tzinfo, name)(tzinfoarg), and check the result.  tzinfo must
+ * be an instance of the tzinfo class.  If the method returns None, this
+ * returns None.  If the method doesn't return None or timedelta, TypeError is
+ * raised and this returns NULL.  If it returns a timedelta and the value is
+ * out of range or isn't a whole number of minutes, ValueError is raised and
+ * this returns NULL.  Else result is returned.
  */
-static int
-call_utc_tzinfo_method(PyObject *tzinfo, char *name, PyObject *tzinfoarg,
-                       int *none)
+static PyObject *
+call_tzinfo_method(PyObject *tzinfo, char *name, PyObject *tzinfoarg)
 {
-    PyObject *u;
-    int result = -1;
+    PyObject *offset;
 
     assert(tzinfo != NULL);
-    assert(PyTZInfo_Check(tzinfo));
+    assert(PyTZInfo_Check(tzinfo) || tzinfo == Py_None);
     assert(tzinfoarg != NULL);
 
-    *none = 0;
-    u = call_tzinfo_method(tzinfo, name, tzinfoarg);
-    if (u == NULL)
-        return -1;
-
-    else if (u == Py_None) {
-        result = 0;
-        *none = 1;
-    }
-    else if (PyDelta_Check(u)) {
-        const int days = GET_TD_DAYS(u);
-        if (days < -1 || days > 0)
-            result = 24*60;             /* trigger ValueError below */
-        else {
-            /* next line can't overflow because we know days
-             * is -1 or 0 now
-             */
-            int ss = days * 24 * 3600 + GET_TD_SECONDS(u);
-            result = divmod(ss, 60, &ss);
-            if (ss || GET_TD_MICROSECONDS(u)) {
-                PyErr_Format(PyExc_ValueError,
-                             "tzinfo.%s() must return a "
-                             "whole number of minutes",
-                             name);
-                result = -1;
-            }
+    if (tzinfo == Py_None)
+        Py_RETURN_NONE;
+    offset = PyObject_CallMethod(tzinfo, name, "O", tzinfoarg);
+    if (offset == Py_None || offset == NULL)
+        return offset;
+    if (PyDelta_Check(offset)) {
+        if (GET_TD_MICROSECONDS(offset) != 0 || GET_TD_SECONDS(offset) % 60 != 0) {
+            Py_DECREF(offset);
+            PyErr_Format(PyExc_ValueError, "offset must be a timedelta"
+                         " representing a whole number of minutes");
+            return NULL;
+        }
+        if ((GET_TD_DAYS(offset) == -1 && GET_TD_SECONDS(offset) == 0) ||
+            GET_TD_DAYS(offset) < -1 || GET_TD_DAYS(offset) >= 1) {
+            Py_DECREF(offset);
+            PyErr_Format(PyExc_ValueError, "offset must be a timedelta"
+                         " strictly between -timedelta(hours=24) and"
+                         " timedelta(hours=24).");
+            return NULL;
         }
     }
     else {
+        Py_DECREF(offset);
         PyErr_Format(PyExc_TypeError,
                      "tzinfo.%s() must return None or "
-                     "timedelta, not '%s'",
-                     name, Py_TYPE(u)->tp_name);
+                     "timedelta, not '%.200s'",
+                     name, Py_TYPE(offset)->tp_name);
+        return NULL;
     }
 
-    Py_DECREF(u);
-    if (result < -1439 || result > 1439) {
-        PyErr_Format(PyExc_ValueError,
-                     "tzinfo.%s() returned %d; must be in "
-                     "-1439 .. 1439",
-                     name, result);
-        result = -1;
-    }
-    return result;
+    return offset;
 }
 
 /* Call tzinfo.utcoffset(tzinfoarg), and extract an integer from the
@@ -899,37 +916,10 @@
  * # of minutes), ValueError is raised and this returns -1.  Else *none is
  * set to 0 and the offset is returned (as int # of minutes east of UTC).
  */
-static int
-call_utcoffset(PyObject *tzinfo, PyObject *tzinfoarg, int *none)
-{
-    return call_utc_tzinfo_method(tzinfo, "utcoffset", tzinfoarg, none);
-}
-
-/* Call tzinfo.name(tzinfoarg), and return the offset as a timedelta or None.
- */
 static PyObject *
-offset_as_timedelta(PyObject *tzinfo, char *name, PyObject *tzinfoarg) {
-    PyObject *result;
-
-    assert(tzinfo && name && tzinfoarg);
-    if (tzinfo == Py_None) {
-        result = Py_None;
-        Py_INCREF(result);
-    }
-    else {
-        int none;
-        int offset = call_utc_tzinfo_method(tzinfo, name, tzinfoarg,
-                                            &none);
-        if (offset < 0 && PyErr_Occurred())
-            return NULL;
-        if (none) {
-            result = Py_None;
-            Py_INCREF(result);
-        }
-        else
-            result = new_delta(0, offset * 60, 0, 1);
-    }
-    return result;
+call_utcoffset(PyObject *tzinfo, PyObject *tzinfoarg)
+{
+    return call_tzinfo_method(tzinfo, "utcoffset", tzinfoarg);
 }
 
 /* Call tzinfo.dst(tzinfoarg), and extract an integer from the
@@ -940,10 +930,10 @@
  * ValueError is raised and this returns -1.  Else *none is set to 0 and
  * the offset is returned (as an int # of minutes east of UTC).
  */
-static int
-call_dst(PyObject *tzinfo, PyObject *tzinfoarg, int *none)
+static PyObject *
+call_dst(PyObject *tzinfo, PyObject *tzinfoarg)
 {
-    return call_utc_tzinfo_method(tzinfo, "dst", tzinfoarg, none);
+    return call_tzinfo_method(tzinfo, "dst", tzinfoarg);
 }
 
 /* Call tzinfo.tzname(tzinfoarg), and return the result.  tzinfo must be
@@ -961,107 +951,23 @@
     assert(check_tzinfo_subclass(tzinfo) >= 0);
     assert(tzinfoarg != NULL);
 
-    if (tzinfo == Py_None) {
-        result = Py_None;
-        Py_INCREF(result);
-    }
-    else
-        result = PyObject_CallMethod(tzinfo, "tzname", "O", tzinfoarg);
+    if (tzinfo == Py_None)
+        Py_RETURN_NONE;
 
-    if (result != NULL && result != Py_None) {
-        if (!PyUnicode_Check(result)) {
-            PyErr_Format(PyExc_TypeError, "tzinfo.tzname() must "
-                         "return None or a string, not '%s'",
-                         Py_TYPE(result)->tp_name);
-            Py_DECREF(result);
-            result = NULL;
-        }
-        else if (!PyUnicode_Check(result)) {
-            PyObject *temp = PyUnicode_FromObject(result);
-            Py_DECREF(result);
-            result = temp;
-        }
-    }
-    return result;
-}
+    result = PyObject_CallMethod(tzinfo, "tzname", "O", tzinfoarg);
 
-typedef enum {
-              /* an exception has been set; the caller should pass it on */
-          OFFSET_ERROR,
-
-          /* type isn't date, datetime, or time subclass */
-          OFFSET_UNKNOWN,
-
-          /* date,
-           * datetime with !hastzinfo
-           * datetime with None tzinfo,
-           * datetime where utcoffset() returns None
-           * time with !hastzinfo
-           * time with None tzinfo,
-           * time where utcoffset() returns None
-           */
-          OFFSET_NAIVE,
-
-          /* time or datetime where utcoffset() doesn't return None */
-          OFFSET_AWARE
-} naivety;
-
-/* Classify an object as to whether it's naive or offset-aware.  See
- * the "naivety" typedef for details.  If the type is aware, *offset is set
- * to minutes east of UTC (as returned by the tzinfo.utcoffset() method).
- * If the type is offset-naive (or unknown, or error), *offset is set to 0.
- * tzinfoarg is the argument to pass to the tzinfo.utcoffset() method.
- */
-static naivety
-classify_utcoffset(PyObject *op, PyObject *tzinfoarg, int *offset)
-{
-    int none;
-    PyObject *tzinfo;
+    if (result == NULL || result == Py_None)
+        return result;
 
-    assert(tzinfoarg != NULL);
-    *offset = 0;
-    tzinfo = get_tzinfo_member(op);     /* NULL means no tzinfo, not error */
-    if (tzinfo == Py_None)
-        return OFFSET_NAIVE;
-    if (tzinfo == NULL) {
-        /* note that a datetime passes the PyDate_Check test */
-        return (PyTime_Check(op) || PyDate_Check(op)) ?
-               OFFSET_NAIVE : OFFSET_UNKNOWN;
-    }
-    *offset = call_utcoffset(tzinfo, tzinfoarg, &none);
-    if (*offset == -1 && PyErr_Occurred())
-        return OFFSET_ERROR;
-    return none ? OFFSET_NAIVE : OFFSET_AWARE;
-}
-
-/* Classify two objects as to whether they're naive or offset-aware.
- * This isn't quite the same as calling classify_utcoffset() twice:  for
- * binary operations (comparison and subtraction), we generally want to
- * ignore the tzinfo members if they're identical.  This is by design,
- * so that results match "naive" expectations when mixing objects from a
- * single timezone.  So in that case, this sets both offsets to 0 and
- * both naiveties to OFFSET_NAIVE.
- * The function returns 0 if everything's OK, and -1 on error.
- */
-static int
-classify_two_utcoffsets(PyObject *o1, int *offset1, naivety *n1,
-                        PyObject *tzinfoarg1,
-                        PyObject *o2, int *offset2, naivety *n2,
-                        PyObject *tzinfoarg2)
-{
-    if (get_tzinfo_member(o1) == get_tzinfo_member(o2)) {
-        *offset1 = *offset2 = 0;
-        *n1 = *n2 = OFFSET_NAIVE;
-    }
-    else {
-        *n1 = classify_utcoffset(o1, tzinfoarg1, offset1);
-        if (*n1 == OFFSET_ERROR)
-            return -1;
-        *n2 = classify_utcoffset(o2, tzinfoarg2, offset2);
-        if (*n2 == OFFSET_ERROR)
-            return -1;
+    if (!PyUnicode_Check(result)) {
+        PyErr_Format(PyExc_TypeError, "tzinfo.tzname() must "
+                     "return None or a string, not '%s'",
+                     Py_TYPE(result)->tp_name);
+        Py_DECREF(result);
+        result = NULL;
     }
-    return 0;
+
+    return result;
 }
 
 /* repr is like "someclass(arg1, arg2)".  If tzinfo isn't None,
@@ -1113,6 +1019,8 @@
                                 GET_YEAR(date));
 }
 
+static PyObject *delta_negative(PyDateTime_Delta *self);
+
 /* Add an hours & minutes UTC offset string to buf.  buf has no more than
  * buflen bytes remaining.  The UTC offset is gotten by calling
  * tzinfo.uctoffset(tzinfoarg).  If that returns None, \0 is stored into
@@ -1127,28 +1035,41 @@
 format_utcoffset(char *buf, size_t buflen, const char *sep,
                 PyObject *tzinfo, PyObject *tzinfoarg)
 {
-    int offset;
-    int hours;
-    int minutes;
+    PyObject *offset;
+    int hours, minutes, seconds;
     char sign;
-    int none;
 
     assert(buflen >= 1);
 
-    offset = call_utcoffset(tzinfo, tzinfoarg, &none);
-    if (offset == -1 && PyErr_Occurred())
+    offset = call_utcoffset(tzinfo, tzinfoarg);
+    if (offset == NULL)
         return -1;
-    if (none) {
+    if (offset == Py_None) {
+        Py_DECREF(offset);
         *buf = '\0';
         return 0;
     }
-    sign = '+';
-    if (offset < 0) {
+    /* Offset is normalized, so it is negative if days < 0 */
+    if (GET_TD_DAYS(offset) < 0) {
+        PyObject *temp = offset;
         sign = '-';
-        offset = - offset;
+        offset = delta_negative((PyDateTime_Delta *)offset);
+        Py_DECREF(temp);
+        if (offset == NULL)
+            return -1;
+    }
+    else {
+        sign = '+';
     }
-    hours = divmod(offset, 60, &minutes);
+    /* Offset is not negative here. */
+    seconds = GET_TD_SECONDS(offset);
+    Py_DECREF(offset);
+    minutes = divmod(seconds, 60, &seconds);
+    hours = divmod(minutes, 60, &minutes);
+    assert(seconds == 0);
+    /* XXX ignore sub-minute data, curently not allowed. */
     PyOS_snprintf(buf, buflen, "%c%02d%s%02d", sign, hours, sep, minutes);
+
     return 0;
 }
 
@@ -1390,7 +1311,7 @@
         format = PyUnicode_FromString(PyBytes_AS_STRING(newfmt));
         if (format != NULL) {
             result = PyObject_CallMethod(time, "strftime", "OO",
-                                         format, timetuple);
+                                         format, timetuple, NULL);
             Py_DECREF(format);
         }
         Py_DECREF(time);
@@ -1893,17 +1814,24 @@
     return result;
 }
 
+static int
+delta_cmp(PyObject *self, PyObject *other)
+{
+    int diff = GET_TD_DAYS(self) - GET_TD_DAYS(other);
+    if (diff == 0) {
+        diff = GET_TD_SECONDS(self) - GET_TD_SECONDS(other);
+        if (diff == 0)
+            diff = GET_TD_MICROSECONDS(self) -
+                GET_TD_MICROSECONDS(other);
+    }
+    return diff;
+}
+
 static PyObject *
 delta_richcompare(PyObject *self, PyObject *other, int op)
 {
     if (PyDelta_Check(other)) {
-        int diff = GET_TD_DAYS(self) - GET_TD_DAYS(other);
-        if (diff == 0) {
-            diff = GET_TD_SECONDS(self) - GET_TD_SECONDS(other);
-            if (diff == 0)
-                diff = GET_TD_MICROSECONDS(self) -
-                       GET_TD_MICROSECONDS(other);
-        }
+        int diff = delta_cmp(self, other);
         return diff_to_bool(diff, op);
     }
     else {
@@ -3075,76 +3003,73 @@
     return tzinfo_nogo("dst");
 }
 
+
+static PyObject *add_datetime_timedelta(PyDateTime_DateTime *date,
+                                        PyDateTime_Delta *delta,
+                                        int factor);
+static PyObject *datetime_utcoffset(PyObject *self, PyObject *);
+static PyObject *datetime_dst(PyObject *self, PyObject *);
+
 static PyObject *
-tzinfo_fromutc(PyDateTime_TZInfo *self, PyDateTime_DateTime *dt)
+tzinfo_fromutc(PyDateTime_TZInfo *self, PyObject *dt)
 {
-    int y, m, d, hh, mm, ss, us;
-
-    PyObject *result;
-    int off, dst;
-    int none;
-    int delta;
+    PyObject *result = NULL;
+    PyObject *off = NULL, *dst = NULL;
+    PyDateTime_Delta *delta = NULL;
 
-    if (! PyDateTime_Check(dt)) {
+    if (!PyDateTime_Check(dt)) {
         PyErr_SetString(PyExc_TypeError,
                         "fromutc: argument must be a datetime");
         return NULL;
     }
-    if (! HASTZINFO(dt) || dt->tzinfo != (PyObject *)self) {
+    if (GET_DT_TZINFO(dt) != (PyObject *)self) {
         PyErr_SetString(PyExc_ValueError, "fromutc: dt.tzinfo "
                         "is not self");
         return NULL;
     }
 
-    off = call_utcoffset(dt->tzinfo, (PyObject *)dt, &none);
-    if (off == -1 && PyErr_Occurred())
+    off = datetime_utcoffset(dt, NULL);
+    if (off == NULL)
         return NULL;
-    if (none) {
+    if (off == Py_None) {
         PyErr_SetString(PyExc_ValueError, "fromutc: non-None "
                         "utcoffset() result required");
-        return NULL;
+        goto Fail;
     }
 
-    dst = call_dst(dt->tzinfo, (PyObject *)dt, &none);
-    if (dst == -1 && PyErr_Occurred())
-        return NULL;
-    if (none) {
+    dst = datetime_dst(dt, NULL);
+    if (dst == NULL)
+        goto Fail;
+    if (dst == Py_None) {
         PyErr_SetString(PyExc_ValueError, "fromutc: non-None "
                         "dst() result required");
-        return NULL;
+        goto Fail;
     }
 
-    y = GET_YEAR(dt);
-    m = GET_MONTH(dt);
-    d = GET_DAY(dt);
-    hh = DATE_GET_HOUR(dt);
-    mm = DATE_GET_MINUTE(dt);
-    ss = DATE_GET_SECOND(dt);
-    us = DATE_GET_MICROSECOND(dt);
-
-    delta = off - dst;
-    mm += delta;
-    if ((mm < 0 || mm >= 60) &&
-        normalize_datetime(&y, &m, &d, &hh, &mm, &ss, &us) < 0)
-        return NULL;
-    result = new_datetime(y, m, d, hh, mm, ss, us, dt->tzinfo);
+    delta = (PyDateTime_Delta *)delta_subtract(off, dst);
+    if (delta == NULL)
+        goto Fail;
+    result = add_datetime_timedelta((PyDateTime_DateTime *)dt, delta, 1);
     if (result == NULL)
-        return result;
-
-    dst = call_dst(dt->tzinfo, result, &none);
-    if (dst == -1 && PyErr_Occurred())
         goto Fail;
-    if (none)
-        goto Inconsistent;
-    if (dst == 0)
-        return result;
 
-    mm += dst;
-    if ((mm < 0 || mm >= 60) &&
-        normalize_datetime(&y, &m, &d, &hh, &mm, &ss, &us) < 0)
+    Py_DECREF(dst);
+    dst = call_dst(GET_DT_TZINFO(dt), result);
+    if (dst == NULL)
         goto Fail;
-    Py_DECREF(result);
-    result = new_datetime(y, m, d, hh, mm, ss, us, dt->tzinfo);
+    if (dst == Py_None)
+        goto Inconsistent;
+    if (delta_bool(delta) != 0) {
+        PyObject *temp = result;
+        result = add_datetime_timedelta((PyDateTime_DateTime *)result,
+                                        (PyDateTime_Delta *)dst, 1);
+        Py_DECREF(temp);
+        if (result == NULL)
+            goto Fail;
+    }
+    Py_DECREF(delta);
+    Py_DECREF(dst);
+    Py_DECREF(off);
     return result;
 
 Inconsistent:
@@ -3153,7 +3078,10 @@
 
     /* fall thru to failure */
 Fail:
-    Py_DECREF(result);
+    Py_XDECREF(off);
+    Py_XDECREF(dst);
+    Py_XDECREF(delta);
+    Py_XDECREF(result);
     return NULL;
 }
 
@@ -3261,7 +3189,7 @@
     PyObject_GenericGetAttr,                    /* tp_getattro */
     0,                                          /* tp_setattro */
     0,                                          /* tp_as_buffer */
-    Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE, /* tp_flags */
+    Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE,   /* tp_flags */
     tzinfo_doc,                                 /* tp_doc */
     0,                                          /* tp_traverse */
     0,                                          /* tp_clear */
@@ -3283,6 +3211,231 @@
     0,                                          /* tp_free */
 };
 
+static char *timezone_kws[] = {"offset", "name", NULL};
+
+static PyObject *
+timezone_new(PyTypeObject *type, PyObject *args, PyObject *kw)
+{
+    PyObject *offset;
+    PyObject *name = NULL;
+    if (PyArg_ParseTupleAndKeywords(args, kw, "O!|O!:timezone", timezone_kws,
+                                    &PyDateTime_DeltaType, &offset,
+                                    &PyUnicode_Type, &name))
+        return new_timezone(offset, name);
+
+    return NULL;
+}
+
+static void
+timezone_dealloc(PyDateTime_TimeZone *self)
+{
+    Py_CLEAR(self->offset);
+    Py_CLEAR(self->name);
+    Py_TYPE(self)->tp_free((PyObject *)self);
+}
+
+static PyObject *
+timezone_richcompare(PyDateTime_TimeZone *self,
+                     PyDateTime_TimeZone *other, int op)
+{
+    if (op != Py_EQ && op != Py_NE) {
+        Py_INCREF(Py_NotImplemented);
+        return Py_NotImplemented;
+    }
+    return delta_richcompare(self->offset, other->offset, op);
+}
+
+static long
+timezone_hash(PyDateTime_TimeZone *self)
+{
+    return delta_hash((PyDateTime_Delta *)self->offset);
+}
+
+/* Check argument type passed to tzname, utcoffset, or dst methods.
+   Returns 0 for good argument.  Returns -1 and sets exception info
+   otherwise.
+ */
+static int
+_timezone_check_argument(PyObject *dt, const char *meth)
+{
+    if (dt == Py_None || PyDateTime_Check(dt))
+        return 0;
+    PyErr_Format(PyExc_TypeError, "%s(dt) argument must be a datetime instance"
+                 " or None, not %.200s", meth, Py_TYPE(dt)->tp_name);
+    return -1;
+}
+
+static PyObject *
+timezone_repr(PyDateTime_TimeZone *self)
+{
+    /* Note that although timezone is not subclassable, it is convenient
+       to use Py_TYPE(self)->tp_name here. */
+    const char *type_name = Py_TYPE(self)->tp_name;
+
+    if (((PyObject *)self) == PyDateTime_TimeZone_UTC)
+        return PyUnicode_FromFormat("%s.utc", type_name);
+
+    if (self->name == NULL)
+        return PyUnicode_FromFormat("%s(%R)", type_name, self->offset);
+
+    return PyUnicode_FromFormat("%s(%R, %R)", type_name, self->offset,
+                                self->name);
+}
+
+
+static PyObject *
+timezone_str(PyDateTime_TimeZone *self)
+{
+    char buf[10];
+    int hours, minutes, seconds;
+    PyObject *offset;
+    char sign;
+
+    if (self->name != NULL) {
+        Py_INCREF(self->name);
+        return self->name;
+    }
+    /* Offset is normalized, so it is negative if days < 0 */
+    if (GET_TD_DAYS(self->offset) < 0) {
+        sign = '-';
+        offset = delta_negative((PyDateTime_Delta *)self->offset);
+        if (offset == NULL)
+            return NULL;
+    }
+    else {
+        sign = '+';
+        offset = self->offset;
+        Py_INCREF(offset);
+    }
+    /* Offset is not negative here. */
+    seconds = GET_TD_SECONDS(offset);
+    Py_DECREF(offset);
+    minutes = divmod(seconds, 60, &seconds);
+    hours = divmod(minutes, 60, &minutes);
+    assert(seconds == 0);
+    /* XXX ignore sub-minute data, curently not allowed. */
+    PyOS_snprintf(buf, sizeof(buf), "UTC%c%02d:%02d", sign, hours, minutes);
+
+    return PyUnicode_FromString(buf);
+}
+
+static PyObject *
+timezone_tzname(PyDateTime_TimeZone *self, PyObject *dt)
+{
+    if (_timezone_check_argument(dt, "tzname") == -1)
+        return NULL;
+
+    return timezone_str(self);
+}
+
+static PyObject *
+timezone_utcoffset(PyDateTime_TimeZone *self, PyObject *dt)
+{
+    if (_timezone_check_argument(dt, "utcoffset") == -1)
+        return NULL;
+
+    Py_INCREF(self->offset);
+    return self->offset;
+}
+
+static PyObject *
+timezone_dst(PyObject *self, PyObject *dt)
+{
+    if (_timezone_check_argument(dt, "dst") == -1)
+        return NULL;
+
+    Py_RETURN_NONE;
+}
+
+static PyObject *
+timezone_fromutc(PyDateTime_TimeZone *self, PyDateTime_DateTime *dt)
+{
+    if (!PyDateTime_Check(dt)) {
+        PyErr_SetString(PyExc_TypeError,
+                        "fromutc: argument must be a datetime");
+        return NULL;
+    }
+    if (!HASTZINFO(dt) || dt->tzinfo != (PyObject *)self) {
+        PyErr_SetString(PyExc_ValueError, "fromutc: dt.tzinfo "
+                        "is not self");
+        return NULL;
+    }
+
+    return add_datetime_timedelta(dt, (PyDateTime_Delta *)self->offset, 1);
+}
+
+static PyObject *
+timezone_getinitargs(PyDateTime_TimeZone *self)
+{
+    if (self->name == NULL)
+        return Py_BuildValue("(O)", self->offset);
+    return Py_BuildValue("(OO)", self->offset, self->name);
+}
+
+static PyMethodDef timezone_methods[] = {
+    {"tzname", (PyCFunction)timezone_tzname, METH_O,
+     PyDoc_STR("If name is specified when timezone is created, returns the name."
+               "  Otherwise returns offset as 'UTC(+|-)HH:MM'.")},
+
+    {"utcoffset", (PyCFunction)timezone_utcoffset, METH_O,
+     PyDoc_STR("Return fixed offset.")},
+
+    {"dst", (PyCFunction)timezone_dst, METH_O,
+     PyDoc_STR("Return None.")},
+
+    {"fromutc", (PyCFunction)timezone_fromutc, METH_O,
+     PyDoc_STR("datetime in UTC -> datetime in local time.")},
+
+    {"__getinitargs__", (PyCFunction)timezone_getinitargs, METH_NOARGS,
+     PyDoc_STR("pickle support")},
+
+    {NULL, NULL}
+};
+
+static char timezone_doc[] =
+PyDoc_STR("Fixed offset from UTC implementation of tzinfo.");
+
+static PyTypeObject PyDateTime_TimeZoneType = {
+    PyVarObject_HEAD_INIT(NULL, 0)
+    "datetime.timezone",              /* tp_name */
+    sizeof(PyDateTime_TimeZone),      /* tp_basicsize */
+    0,                                /* tp_itemsize */
+    (destructor)timezone_dealloc,     /* tp_dealloc */
+    0,                                /* tp_print */
+    0,                                /* tp_getattr */
+    0,                                /* tp_setattr */
+    0,                                /* tp_reserved */
+    (reprfunc)timezone_repr,          /* tp_repr */
+    0,                                /* tp_as_number */
+    0,                                /* tp_as_sequence */
+    0,                                /* tp_as_mapping */
+    (hashfunc)timezone_hash,          /* tp_hash */
+    0,                                /* tp_call */
+    (reprfunc)timezone_str,           /* tp_str */
+    0,                                /* tp_getattro */
+    0,                                /* tp_setattro */
+    0,                                /* tp_as_buffer */
+    Py_TPFLAGS_DEFAULT,               /* tp_flags */
+    timezone_doc,                     /* tp_doc */
+    0,                                /* tp_traverse */
+    0,                                /* tp_clear */
+    (richcmpfunc)timezone_richcompare,/* tp_richcompare */
+    0,                                /* tp_weaklistoffset */
+    0,                                /* tp_iter */
+    0,                                /* tp_iternext */
+    timezone_methods,                 /* tp_methods */
+    0,                                /* tp_members */
+    0,                                /* tp_getset */
+    &PyDateTime_TZInfoType,           /* tp_base */
+    0,                                /* tp_dict */
+    0,                                /* tp_descr_get */
+    0,                                /* tp_descr_set */
+    0,                                /* tp_dictoffset */
+    0,                                /* tp_init */
+    0,                                /* tp_alloc */
+    timezone_new,                     /* tp_new */
+};
+
 /*
  * PyDateTime_Time implementation.
  */
@@ -3416,21 +3569,18 @@
 
 /* These are all METH_NOARGS, so don't need to check the arglist. */
 static PyObject *
-time_utcoffset(PyDateTime_Time *self, PyObject *unused) {
-    return offset_as_timedelta(HASTZINFO(self) ? self->tzinfo : Py_None,
-                               "utcoffset", Py_None);
+time_utcoffset(PyObject *self, PyObject *unused) {
+    return call_utcoffset(GET_TIME_TZINFO(self), Py_None);
 }
 
 static PyObject *
-time_dst(PyDateTime_Time *self, PyObject *unused) {
-    return offset_as_timedelta(HASTZINFO(self) ? self->tzinfo : Py_None,
-                               "dst", Py_None);
+time_dst(PyObject *self, PyObject *unused) {
+    return call_dst(GET_TIME_TZINFO(self), Py_None);
 }
 
 static PyObject *
 time_tzname(PyDateTime_Time *self, PyObject *unused) {
-    return call_tzname(HASTZINFO(self) ? self->tzinfo : Py_None,
-                       Py_None);
+    return call_tzname(GET_TIME_TZINFO(self), Py_None);
 }
 
 /*
@@ -3485,7 +3635,7 @@
                                       TIME_GET_MINUTE(self),
                                       TIME_GET_SECOND(self));
 
-    if (result == NULL || ! HASTZINFO(self) || self->tzinfo == Py_None)
+    if (result == NULL || !HASTZINFO(self) || self->tzinfo == Py_None)
         return result;
 
     /* We need to append the UTC offset. */
@@ -3536,98 +3686,108 @@
 static PyObject *
 time_richcompare(PyObject *self, PyObject *other, int op)
 {
+    PyObject *result = NULL;
+    PyObject *offset1, *offset2;
     int diff;
-    naivety n1, n2;
-    int offset1, offset2;
 
     if (! PyTime_Check(other)) {
         Py_INCREF(Py_NotImplemented);
         return Py_NotImplemented;
     }
-    if (classify_two_utcoffsets(self, &offset1, &n1, Py_None,
-                                other, &offset2, &n2, Py_None) < 0)
+
+    if (GET_TIME_TZINFO(self) == GET_TIME_TZINFO(other)) {
+        diff = memcmp(((PyDateTime_Time *)self)->data,
+                      ((PyDateTime_Time *)other)->data,
+                      _PyDateTime_TIME_DATASIZE);
+        return diff_to_bool(diff, op);
+    }
+    offset1 = time_utcoffset(self, NULL);
+    if (offset1 == NULL)
         return NULL;
-    assert(n1 != OFFSET_UNKNOWN && n2 != OFFSET_UNKNOWN);
+    offset2 = time_utcoffset(other, NULL);
+    if (offset2 == NULL)
+        goto done;
     /* If they're both naive, or both aware and have the same offsets,
      * we get off cheap.  Note that if they're both naive, offset1 ==
-     * offset2 == 0 at this point.
+     * offset2 == Py_None at this point.
      */
-    if (n1 == n2 && offset1 == offset2) {
+    if ((offset1 == offset2) ||
+        (PyDelta_Check(offset1) && PyDelta_Check(offset2) &&
+         delta_cmp(offset1, offset2) == 0)) {
         diff = memcmp(((PyDateTime_Time *)self)->data,
                       ((PyDateTime_Time *)other)->data,
                       _PyDateTime_TIME_DATASIZE);
-        return diff_to_bool(diff, op);
+        result = diff_to_bool(diff, op);
     }
-
-    if (n1 == OFFSET_AWARE && n2 == OFFSET_AWARE) {
-        assert(offset1 != offset2);             /* else last "if" handled it */
-        /* Convert everything except microseconds to seconds.  These
-         * can't overflow (no more than the # of seconds in 2 days).
-         */
-        offset1 = TIME_GET_HOUR(self) * 3600 +
-                  (TIME_GET_MINUTE(self) - offset1) * 60 +
-                  TIME_GET_SECOND(self);
-        offset2 = TIME_GET_HOUR(other) * 3600 +
-                  (TIME_GET_MINUTE(other) - offset2) * 60 +
-                  TIME_GET_SECOND(other);
-        diff = offset1 - offset2;
+    /* The hard case: both aware with different UTC offsets */
+    else if (offset1 != Py_None && offset2 != Py_None) {
+        int offsecs1, offsecs2;
+        assert(offset1 != offset2); /* else last "if" handled it */
+        offsecs1 = TIME_GET_HOUR(self) * 3600 +
+                   TIME_GET_MINUTE(self) * 60 +
+                   TIME_GET_SECOND(self) -
+                   GET_TD_DAYS(offset1) * 86400 -
+                   GET_TD_SECONDS(offset1);
+        offsecs2 = TIME_GET_HOUR(other) * 3600 +
+                   TIME_GET_MINUTE(other) * 60 +
+                   TIME_GET_SECOND(other) -
+                   GET_TD_DAYS(offset2) * 86400 -
+                   GET_TD_SECONDS(offset2);
+        diff = offsecs1 - offsecs2;
         if (diff == 0)
             diff = TIME_GET_MICROSECOND(self) -
                    TIME_GET_MICROSECOND(other);
-        return diff_to_bool(diff, op);
+        result = diff_to_bool(diff, op);
     }
-
-    assert(n1 != n2);
-    PyErr_SetString(PyExc_TypeError,
-                    "can't compare offset-naive and "
-                    "offset-aware times");
-    return NULL;
+    else {
+        PyErr_SetString(PyExc_TypeError,
+                        "can't compare offset-naive and "
+                        "offset-aware times");
+    }
+ done:
+    Py_DECREF(offset1);
+    Py_XDECREF(offset2);
+    return result;
 }
 
 static long
 time_hash(PyDateTime_Time *self)
 {
     if (self->hashcode == -1) {
-        naivety n;
-        int offset;
-        PyObject *temp;
-
-        n = classify_utcoffset((PyObject *)self, Py_None, &offset);
-        assert(n != OFFSET_UNKNOWN);
-        if (n == OFFSET_ERROR)
+        PyObject *offset;
+
+        offset = time_utcoffset((PyObject *)self, NULL);
+
+        if (offset == NULL)
             return -1;
 
         /* Reduce this to a hash of another object. */
-        if (offset == 0) {
+        if (offset == Py_None)
             self->hashcode = generic_hash(
                 (unsigned char *)self->data, _PyDateTime_TIME_DATASIZE);
-            return self->hashcode;
-        }
         else {
-            int hour;
-            int minute;
-
-            assert(n == OFFSET_AWARE);
+            PyObject *temp1, *temp2;
+            int seconds, microseconds;
             assert(HASTZINFO(self));
-            hour = divmod(TIME_GET_HOUR(self) * 60 +
-                            TIME_GET_MINUTE(self) - offset,
-                          60,
-                          &minute);
-            if (0 <= hour && hour < 24)
-                temp = new_time(hour, minute,
-                                TIME_GET_SECOND(self),
-                                TIME_GET_MICROSECOND(self),
-                                Py_None);
-            else
-                temp = Py_BuildValue("iiii",
-                           hour, minute,
-                           TIME_GET_SECOND(self),
-                           TIME_GET_MICROSECOND(self));
-        }
-        if (temp != NULL) {
-            self->hashcode = PyObject_Hash(temp);
-            Py_DECREF(temp);
+            seconds = TIME_GET_HOUR(self) * 3600 +
+                      TIME_GET_MINUTE(self) * 60 +
+                      TIME_GET_SECOND(self);
+            microseconds = TIME_GET_MICROSECOND(self);
+            temp1 = new_delta(0, seconds, microseconds, 1);
+            if (temp1 == NULL) {
+                Py_DECREF(offset);
+                return -1;
+            }
+            temp2 = delta_subtract(temp1, offset);
+            Py_DECREF(temp1);
+            if (temp2 == NULL) {
+                Py_DECREF(offset);
+                return -1;
+            }
+            self->hashcode = PyObject_Hash(temp2);
+            Py_DECREF(temp2);
         }
+        Py_DECREF(offset);
     }
     return self->hashcode;
 }
@@ -3656,10 +3816,10 @@
 }
 
 static int
-time_bool(PyDateTime_Time *self)
+time_bool(PyObject *self)
 {
-    int offset;
-    int none;
+    PyObject *offset, *tzinfo;
+    int offsecs = 0;
 
     if (TIME_GET_SECOND(self) || TIME_GET_MICROSECOND(self)) {
         /* Since utcoffset is in whole minutes, nothing can
@@ -3667,13 +3827,15 @@
          */
         return 1;
     }
-    offset = 0;
-    if (HASTZINFO(self) && self->tzinfo != Py_None) {
-        offset = call_utcoffset(self->tzinfo, Py_None, &none);
-        if (offset == -1 && PyErr_Occurred())
+    tzinfo = GET_TIME_TZINFO(self);
+    if (tzinfo != Py_None) {
+        offset = call_utcoffset(tzinfo, Py_None);
+        if (offset == NULL)
             return -1;
+        offsecs = GET_TD_DAYS(offset)*86400 + GET_TD_SECONDS(offset);
+        Py_DECREF(offset);
     }
-    return (TIME_GET_MINUTE(self) - offset + TIME_GET_HOUR(self)*60) != 0;
+    return (TIME_GET_MINUTE(self)*60 - offsecs + TIME_GET_HOUR(self)*3600) != 0;
 }
 
 /* Pickle support, a simple use of __reduce__. */
@@ -4115,82 +4277,23 @@
     return result;
 }
 
-/* Return new datetime from time.strptime(). */
+/* Return new datetime from _strptime.strptime_datetime(). */
 static PyObject *
 datetime_strptime(PyObject *cls, PyObject *args)
 {
     static PyObject *module = NULL;
-    PyObject *result = NULL, *obj, *st = NULL, *frac = NULL;
     const Py_UNICODE *string, *format;
 
     if (!PyArg_ParseTuple(args, "uu:strptime", &string, &format))
         return NULL;
 
-    if (module == NULL &&
-        (module = PyImport_ImportModuleNoBlock("_strptime")) == NULL)
-        return NULL;
-
-    /* _strptime._strptime returns a two-element tuple.  The first
-       element is a time.struct_time object.  The second is the
-       microseconds (which are not defined for time.struct_time). */
-    obj = PyObject_CallMethod(module, "_strptime", "uu", string, format);
-    if (obj != NULL) {
-        int i, good_timetuple = 1;
-        long int ia[7];
-        if (PySequence_Check(obj) && PySequence_Size(obj) == 2) {
-            st = PySequence_GetItem(obj, 0);
-            frac = PySequence_GetItem(obj, 1);
-            if (st == NULL || frac == NULL)
-                good_timetuple = 0;
-            /* copy y/m/d/h/m/s values out of the
-               time.struct_time */
-            if (good_timetuple &&
-                PySequence_Check(st) &&
-                PySequence_Size(st) >= 6) {
-                for (i=0; i < 6; i++) {
-                    PyObject *p = PySequence_GetItem(st, i);
-                    if (p == NULL) {
-                        good_timetuple = 0;
-                        break;
-                    }
-                    if (PyLong_Check(p))
-                        ia[i] = PyLong_AsLong(p);
-                    else
-                        good_timetuple = 0;
-                    Py_DECREF(p);
-                }
-/*                              if (PyLong_CheckExact(p)) {
-                                        ia[i] = PyLong_AsLongAndOverflow(p, &overflow);
-                                        if (overflow)
-                                                good_timetuple = 0;
-                                }
-                                else
-                                        good_timetuple = 0;
-                                Py_DECREF(p);
-*/                      }
-                        else
-                                good_timetuple = 0;
-                        /* follow that up with a little dose of microseconds */
-            if (PyLong_Check(frac))
-                ia[6] = PyLong_AsLong(frac);
-            else
-                good_timetuple = 0;
-        }
-        else
-            good_timetuple = 0;
-        if (good_timetuple)
-            result = PyObject_CallFunction(cls, "iiiiiii",
-                                           ia[0], ia[1], ia[2],
-                                           ia[3], ia[4], ia[5],
-                                           ia[6]);
-        else
-            PyErr_SetString(PyExc_ValueError,
-                "unexpected value from _strptime._strptime");
+    if (module == NULL) {
+        module = PyImport_ImportModuleNoBlock("_strptime");
+        if (module == NULL)
+            return NULL;
     }
-    Py_XDECREF(obj);
-    Py_XDECREF(st);
-    Py_XDECREF(frac);
-    return result;
+    return PyObject_CallMethod(module, "_strptime_datetime", "Ouu",
+                               cls, string, format);
 }
 
 /* Return new datetime from date/datetime and time arguments. */
@@ -4241,21 +4344,18 @@
 
 /* These are all METH_NOARGS, so don't need to check the arglist. */
 static PyObject *
-datetime_utcoffset(PyDateTime_DateTime *self, PyObject *unused) {
-    return offset_as_timedelta(HASTZINFO(self) ? self->tzinfo : Py_None,
-                               "utcoffset", (PyObject *)self);
+datetime_utcoffset(PyObject *self, PyObject *unused) {
+    return call_utcoffset(GET_DT_TZINFO(self), self);
 }
 
 static PyObject *
-datetime_dst(PyDateTime_DateTime *self, PyObject *unused) {
-    return offset_as_timedelta(HASTZINFO(self) ? self->tzinfo : Py_None,
-                               "dst", (PyObject *)self);
+datetime_dst(PyObject *self, PyObject *unused) {
+    return call_dst(GET_DT_TZINFO(self), self);
 }
 
 static PyObject *
-datetime_tzname(PyDateTime_DateTime *self, PyObject *unused) {
-    return call_tzname(HASTZINFO(self) ? self->tzinfo : Py_None,
-                       (PyObject *)self);
+datetime_tzname(PyObject *self, PyObject *unused) {
+    return call_tzname(GET_DT_TZINFO(self), self);
 }
 
 /*
@@ -4322,21 +4422,43 @@
         /* datetime - ??? */
         if (PyDateTime_Check(right)) {
             /* datetime - datetime */
-            naivety n1, n2;
-            int offset1, offset2;
+            PyObject *offset1, *offset2, *offdiff = NULL;
             int delta_d, delta_s, delta_us;
 
-            if (classify_two_utcoffsets(left, &offset1, &n1, left,
-                                        right, &offset2, &n2,
-                                        right) < 0)
-                return NULL;
-            assert(n1 != OFFSET_UNKNOWN && n2 != OFFSET_UNKNOWN);
-            if (n1 != n2) {
-                PyErr_SetString(PyExc_TypeError,
-                    "can't subtract offset-naive and "
-                    "offset-aware datetimes");
-                return NULL;
+            if (GET_DT_TZINFO(left) == GET_DT_TZINFO(right)) {
+                offset2 = offset1 = Py_None;
+                Py_INCREF(offset1);
+                Py_INCREF(offset2);
+            }
+            else {
+                offset1 = datetime_utcoffset(left, NULL);
+                if (offset1 == NULL)
+                    return NULL;
+                offset2 = datetime_utcoffset(right, NULL);
+                if (offset2 == NULL) {
+                    Py_DECREF(offset1);
+                    return NULL;
+                }
+                if ((offset1 != Py_None) != (offset2 != Py_None)) {
+                    PyErr_SetString(PyExc_TypeError,
+                                    "can't subtract offset-naive and "
+                                    "offset-aware datetimes");
+                    Py_DECREF(offset1);
+                    Py_DECREF(offset2);
+                    return NULL;
+                }
             }
+            if ((offset1 != offset2) &&
+                delta_cmp(offset1, offset2) != 0) {
+                offdiff = delta_subtract(offset1, offset2);
+                if (offdiff == NULL) {
+                    Py_DECREF(offset1);
+                    Py_DECREF(offset2);
+                    return NULL;
+                }
+            }
+            Py_DECREF(offset1);
+            Py_DECREF(offset2);
             delta_d = ymd_to_ord(GET_YEAR(left),
                                  GET_MONTH(left),
                                  GET_DAY(left)) -
@@ -4355,11 +4477,13 @@
                        DATE_GET_SECOND(right));
             delta_us = DATE_GET_MICROSECOND(left) -
                        DATE_GET_MICROSECOND(right);
-            /* (left - offset1) - (right - offset2) =
-             * (left - right) + (offset2 - offset1)
-             */
-            delta_s += (offset2 - offset1) * 60;
             result = new_delta(delta_d, delta_s, delta_us, 1);
+            if (offdiff != NULL) {
+                PyObject *temp = result;
+                result = delta_subtract(result, offdiff);
+                Py_DECREF(temp);
+                Py_DECREF(offdiff);
+            }
         }
         else if (PyDelta_Check(right)) {
             /* datetime - delta */
@@ -4469,9 +4593,9 @@
 static PyObject *
 datetime_richcompare(PyObject *self, PyObject *other, int op)
 {
+    PyObject *result = NULL;
+    PyObject *offset1, *offset2;
     int diff;
-    naivety n1, n2;
-    int offset1, offset2;
 
     if (! PyDateTime_Check(other)) {
         if (PyDate_Check(other)) {
@@ -4492,85 +4616,99 @@
         return Py_NotImplemented;
     }
 
-    if (classify_two_utcoffsets(self, &offset1, &n1, self,
-                                other, &offset2, &n2, other) < 0)
+    if (GET_DT_TZINFO(self) == GET_DT_TZINFO(other)) {
+        diff = memcmp(((PyDateTime_DateTime *)self)->data,
+                      ((PyDateTime_DateTime *)other)->data,
+                      _PyDateTime_DATETIME_DATASIZE);
+        return diff_to_bool(diff, op);
+    }
+    offset1 = datetime_utcoffset(self, NULL);
+    if (offset1 == NULL)
         return NULL;
-    assert(n1 != OFFSET_UNKNOWN && n2 != OFFSET_UNKNOWN);
+    offset2 = datetime_utcoffset(other, NULL);
+    if (offset2 == NULL)
+        goto done;
     /* If they're both naive, or both aware and have the same offsets,
      * we get off cheap.  Note that if they're both naive, offset1 ==
-     * offset2 == 0 at this point.
+     * offset2 == Py_None at this point.
      */
-    if (n1 == n2 && offset1 == offset2) {
+    if ((offset1 == offset2) ||
+        (PyDelta_Check(offset1) && PyDelta_Check(offset2) &&
+         delta_cmp(offset1, offset2) == 0)) {
         diff = memcmp(((PyDateTime_DateTime *)self)->data,
                       ((PyDateTime_DateTime *)other)->data,
                       _PyDateTime_DATETIME_DATASIZE);
-        return diff_to_bool(diff, op);
+        result = diff_to_bool(diff, op);
     }
-
-    if (n1 == OFFSET_AWARE && n2 == OFFSET_AWARE) {
+    else if (offset1 != Py_None && offset2 != Py_None) {
         PyDateTime_Delta *delta;
 
-        assert(offset1 != offset2);             /* else last "if" handled it */
+        assert(offset1 != offset2); /* else last "if" handled it */
         delta = (PyDateTime_Delta *)datetime_subtract((PyObject *)self,
                                                        other);
         if (delta == NULL)
-            return NULL;
+            goto done;
         diff = GET_TD_DAYS(delta);
         if (diff == 0)
             diff = GET_TD_SECONDS(delta) |
                    GET_TD_MICROSECONDS(delta);
         Py_DECREF(delta);
-        return diff_to_bool(diff, op);
+        result = diff_to_bool(diff, op);
     }
-
-    assert(n1 != n2);
-    PyErr_SetString(PyExc_TypeError,
-                    "can't compare offset-naive and "
-                    "offset-aware datetimes");
-    return NULL;
+    else {
+        PyErr_SetString(PyExc_TypeError,
+                        "can't compare offset-naive and "
+                        "offset-aware datetimes");
+    }
+ done:
+    Py_DECREF(offset1);
+    Py_XDECREF(offset2);
+    return result;
 }
 
 static long
 datetime_hash(PyDateTime_DateTime *self)
 {
     if (self->hashcode == -1) {
-        naivety n;
-        int offset;
-        PyObject *temp;
-
-        n = classify_utcoffset((PyObject *)self, (PyObject *)self,
-                               &offset);
-        assert(n != OFFSET_UNKNOWN);
-        if (n == OFFSET_ERROR)
+        PyObject *offset;
+
+        offset = datetime_utcoffset((PyObject *)self, NULL);
+
+        if (offset == NULL)
             return -1;
 
         /* Reduce this to a hash of another object. */
-        if (n == OFFSET_NAIVE) {
+        if (offset == Py_None)
             self->hashcode = generic_hash(
                 (unsigned char *)self->data, _PyDateTime_DATETIME_DATASIZE);
-            return self->hashcode;
-        }
         else {
-            int days;
-            int seconds;
+            PyObject *temp1, *temp2;
+            int days, seconds;
 
-            assert(n == OFFSET_AWARE);
             assert(HASTZINFO(self));
             days = ymd_to_ord(GET_YEAR(self),
                               GET_MONTH(self),
                               GET_DAY(self));
             seconds = DATE_GET_HOUR(self) * 3600 +
-                      (DATE_GET_MINUTE(self) - offset) * 60 +
+                      DATE_GET_MINUTE(self) * 60 +
                       DATE_GET_SECOND(self);
-            temp = new_delta(days,
-                             seconds,
-                             DATE_GET_MICROSECOND(self),
-                             1);
-        }
-        if (temp != NULL) {
-            self->hashcode = PyObject_Hash(temp);
-            Py_DECREF(temp);
+            temp1 = new_delta(days, seconds,
+                              DATE_GET_MICROSECOND(self),
+                              1);
+            if (temp1 == NULL) {
+                Py_DECREF(offset);
+                return -1;
+            }
+            temp2 = delta_subtract(temp1, offset);
+            Py_DECREF(temp1);
+            if (temp2 == NULL) {
+                Py_DECREF(offset);
+                return -1;
+            }
+            self->hashcode = PyObject_Hash(temp2);
+            Py_DECREF(temp2);
         }
+        Py_DECREF(offset);
     }
     return self->hashcode;
 }
@@ -4605,10 +4743,9 @@
 static PyObject *
 datetime_astimezone(PyDateTime_DateTime *self, PyObject *args, PyObject *kw)
 {
-    int y, m, d, hh, mm, ss, us;
     PyObject *result;
-    int offset, none;
-
+    PyObject *offset;
+    PyObject *temp;
     PyObject *tzinfo;
     static char *keywords[] = {"tz", NULL};
 
@@ -4626,39 +4763,35 @@
     }
 
     /* Convert self to UTC. */
-    offset = call_utcoffset(self->tzinfo, (PyObject *)self, &none);
-    if (offset == -1 && PyErr_Occurred())
+    offset = datetime_utcoffset((PyObject *)self, NULL);
+    if (offset == NULL)
         return NULL;
-    if (none)
-        goto NeedAware;
+    if (offset == Py_None) {
+        Py_DECREF(offset);
+      NeedAware:
+        PyErr_SetString(PyExc_ValueError, "astimezone() cannot be applied to "
+                        "a naive datetime");
+        return NULL;
+    }
 
-    y = GET_YEAR(self);
-    m = GET_MONTH(self);
-    d = GET_DAY(self);
-    hh = DATE_GET_HOUR(self);
-    mm = DATE_GET_MINUTE(self);
-    ss = DATE_GET_SECOND(self);
-    us = DATE_GET_MICROSECOND(self);
-
-    mm -= offset;
-    if ((mm < 0 || mm >= 60) &&
-        normalize_datetime(&y, &m, &d, &hh, &mm, &ss, &us) < 0)
+    /* result = self - offset */
+    result = add_datetime_timedelta(self,
+                (PyDateTime_Delta *)offset, -1);
+    Py_DECREF(offset);
+    if (result == NULL)
         return NULL;
 
     /* Attach new tzinfo and let fromutc() do the rest. */
-    result = new_datetime(y, m, d, hh, mm, ss, us, tzinfo);
-    if (result != NULL) {
-        PyObject *temp = result;
+    temp = ((PyDateTime_DateTime *)result)->tzinfo;
+    ((PyDateTime_DateTime *)result)->tzinfo = tzinfo;
+    Py_INCREF(tzinfo);
+    Py_DECREF(temp);
 
-        result = PyObject_CallMethod(tzinfo, "fromutc", "O", temp);
-        Py_DECREF(temp);
-    }
-    return result;
+    temp = result;
+    result = PyObject_CallMethod(tzinfo, "fromutc", "O", temp);
+    Py_DECREF(temp);
 
-NeedAware:
-    PyErr_SetString(PyExc_ValueError, "astimezone() cannot be applied to "
-                                      "a naive datetime");
-    return NULL;
+    return result;
 }
 
 static PyObject *
@@ -4667,17 +4800,15 @@
     int dstflag = -1;
 
     if (HASTZINFO(self) && self->tzinfo != Py_None) {
-        int none;
+        PyObject * dst;
 
-        dstflag = call_dst(self->tzinfo, (PyObject *)self, &none);
-        if (dstflag == -1 && PyErr_Occurred())
+        dst = call_dst(self->tzinfo, (PyObject *)self);
+        if (dst == NULL)
             return NULL;
 
-        if (none)
-            dstflag = -1;
-        else if (dstflag != 0)
-            dstflag = 1;
-
+        if (dst != Py_None)
+            dstflag = delta_bool((PyDateTime_Delta *)dst);
+        Py_DECREF(dst);
     }
     return build_struct_time(GET_YEAR(self),
                              GET_MONTH(self),
@@ -4713,47 +4844,47 @@
                     DATE_GET_MINUTE(self),
                     DATE_GET_SECOND(self),
                     DATE_GET_MICROSECOND(self),
-                    HASTZINFO(self) ? self->tzinfo : Py_None);
+                    GET_DT_TZINFO(self));
 }
 
 static PyObject *
 datetime_utctimetuple(PyDateTime_DateTime *self)
 {
-    int y = GET_YEAR(self);
-    int m = GET_MONTH(self);
-    int d = GET_DAY(self);
-    int hh = DATE_GET_HOUR(self);
-    int mm = DATE_GET_MINUTE(self);
-    int ss = DATE_GET_SECOND(self);
-    int us = 0;         /* microseconds are ignored in a timetuple */
-    int offset = 0;
-
-    if (HASTZINFO(self) && self->tzinfo != Py_None) {
-        int none;
+    int y, m, d, hh, mm, ss;
+    PyObject *tzinfo;
+    PyDateTime_DateTime *utcself;
 
-        offset = call_utcoffset(self->tzinfo, (PyObject *)self, &none);
-        if (offset == -1 && PyErr_Occurred())
-            return NULL;
+    tzinfo = GET_DT_TZINFO(self);
+    if (tzinfo == Py_None) {
+        utcself = self;
+        Py_INCREF(utcself);
     }
-    /* Even if offset is 0, don't call timetuple() -- tm_isdst should be
-     * 0 in a UTC timetuple regardless of what dst() says.
-     */
-    if (offset) {
-        /* Subtract offset minutes & normalize. */
-        int stat;
-
-        mm -= offset;
-        stat = normalize_datetime(&y, &m, &d, &hh, &mm, &ss, &us);
-        if (stat < 0) {
-            /* At the edges, it's possible we overflowed
-             * beyond MINYEAR or MAXYEAR.
-             */
-            if (PyErr_ExceptionMatches(PyExc_OverflowError))
-                PyErr_Clear();
-            else
+    else {
+        PyObject *offset;
+        offset = call_utcoffset(tzinfo, (PyObject *)self);
+        if (offset == NULL)
+            return NULL;
+        if (offset == Py_None) {
+            Py_DECREF(offset);
+            utcself = self;
+            Py_INCREF(utcself);
+        }
+        else {
+            utcself = (PyDateTime_DateTime *)add_datetime_timedelta(self,
+                                                (PyDateTime_Delta *)offset, -1);
+            Py_DECREF(offset);
+            if (utcself == NULL)
                 return NULL;
         }
     }
+    y = GET_YEAR(utcself);
+    m = GET_MONTH(utcself);
+    d = GET_DAY(utcself);
+    hh = DATE_GET_HOUR(utcself);
+    mm = DATE_GET_MINUTE(utcself);
+    ss = DATE_GET_SECOND(utcself);
+
+    Py_DECREF(utcself);
     return build_struct_time(y, m, d, hh, mm, ss, 0);
 }
 
@@ -4971,6 +5102,7 @@
     PyObject *m;        /* a module object */
     PyObject *d;        /* its dict */
     PyObject *x;
+    PyObject *delta;
 
     m = PyModule_Create(&datetimemodule);
     if (m == NULL)
@@ -4986,6 +5118,8 @@
         return NULL;
     if (PyType_Ready(&PyDateTime_TZInfoType) < 0)
         return NULL;
+    if (PyType_Ready(&PyDateTime_TimeZoneType) < 0)
+        return NULL;
 
     /* timedelta values */
     d = PyDateTime_DeltaType.tp_dict;
@@ -5059,6 +5193,36 @@
         return NULL;
     Py_DECREF(x);
 
+    /* timezone values */
+    d = PyDateTime_TimeZoneType.tp_dict;
+
+    delta = new_delta(0, 0, 0, 0);
+    if (delta == NULL)
+        return NULL;
+    x = new_timezone(delta, NULL);
+    Py_DECREF(delta);
+    if (x == NULL || PyDict_SetItemString(d, "utc", x) < 0)
+        return NULL;
+    PyDateTime_TimeZone_UTC = x;
+
+    delta = new_delta(-1, 60, 0, 1); /* -23:59 */
+    if (delta == NULL)
+        return NULL;
+    x = new_timezone(delta, NULL);
+    Py_DECREF(delta);
+    if (x == NULL || PyDict_SetItemString(d, "min", x) < 0)
+        return NULL;
+    Py_DECREF(x);
+
+    delta = new_delta(0, (23 * 60 + 59) * 60, 0, 0); /* +23:59 */
+    if (delta == NULL)
+        return NULL;
+    x = new_timezone(delta, NULL);
+    Py_DECREF(delta);
+    if (x == NULL || PyDict_SetItemString(d, "max", x) < 0)
+        return NULL;
+    Py_DECREF(x);
+
     /* module initialization */
     PyModule_AddIntConstant(m, "MINYEAR", MINYEAR);
     PyModule_AddIntConstant(m, "MAXYEAR", MAXYEAR);
@@ -5079,6 +5243,9 @@
     Py_INCREF(&PyDateTime_TZInfoType);
     PyModule_AddObject(m, "tzinfo", (PyObject *) &PyDateTime_TZInfoType);
 
+    Py_INCREF(&PyDateTime_TimeZoneType);
+    PyModule_AddObject(m, "timezone", (PyObject *) &PyDateTime_TimeZoneType);
+
     x = PyCapsule_New(&CAPI, PyDateTime_CAPSULE_NAME, NULL);
     if (x == NULL)
         return NULL;

Modified: python/branches/py3k-cdecimal/Modules/makesetup
==============================================================================
--- python/branches/py3k-cdecimal/Modules/makesetup	(original)
+++ python/branches/py3k-cdecimal/Modules/makesetup	Sat Jul 17 12:36:09 2010
@@ -219,7 +219,7 @@
 			case $doconfig in
 			no)	cc="$cc \$(CCSHARED) \$(CFLAGS) \$(CPPFLAGS)";;
 			*)
-				cc="$cc \$(PY_CFLAGS)";;
+				cc="$cc \$(PY_CORE_CFLAGS)";;
 			esac
 			rule="$obj: $src; $cc $cpps -c $src -o $obj"
 			echo "$rule" >>$rulesf

Modified: python/branches/py3k-cdecimal/Modules/mathmodule.c
==============================================================================
--- python/branches/py3k-cdecimal/Modules/mathmodule.c	(original)
+++ python/branches/py3k-cdecimal/Modules/mathmodule.c	Sat Jul 17 12:36:09 2010
@@ -69,6 +69,7 @@
 
 static const double pi = 3.141592653589793238462643383279502884197;
 static const double sqrtpi = 1.772453850905516027298167483341145182798;
+static const double logpi = 1.144729885849400174143427351353058711647;
 
 static double
 sinpi(double x)
@@ -356,20 +357,15 @@
     if (absx < 1e-20)
         return -log(absx);
 
-    /* Lanczos' formula */
-    if (x > 0.0) {
-        /* we could save a fraction of a ulp in accuracy by having a
-           second set of numerator coefficients for lanczos_sum that
-           absorbed the exp(-lanczos_g) term, and throwing out the
-           lanczos_g subtraction below; it's probably not worth it. */
-        r = log(lanczos_sum(x)) - lanczos_g +
-            (x-0.5)*(log(x+lanczos_g-0.5)-1);
-    }
-    else {
-        r = log(pi) - log(fabs(sinpi(absx))) - log(absx) -
-            (log(lanczos_sum(absx)) - lanczos_g +
-             (absx-0.5)*(log(absx+lanczos_g-0.5)-1));
-    }
+    /* Lanczos' formula.  We could save a fraction of a ulp in accuracy by
+       having a second set of numerator coefficients for lanczos_sum that
+       absorbed the exp(-lanczos_g) term, and throwing out the lanczos_g
+       subtraction below; it's probably not worth it. */
+    r = log(lanczos_sum(absx)) - lanczos_g;
+    r += (absx - 0.5) * (log(absx + lanczos_g - 0.5) - 1);
+    if (x < 0.0)
+        /* Use reflection formula to get value for negative x. */
+        r = logpi - log(fabs(sinpi(absx))) - log(absx) - r;
     if (Py_IS_INFINITY(r))
         errno = ERANGE;
     return r;
@@ -841,19 +837,17 @@
 
 static PyObject * math_ceil(PyObject *self, PyObject *number) {
     static PyObject *ceil_str = NULL;
-    PyObject *method;
+    PyObject *method, *result;
 
-    if (ceil_str == NULL) {
-        ceil_str = PyUnicode_InternFromString("__ceil__");
-        if (ceil_str == NULL)
+    method = _PyObject_LookupSpecial(number, "__ceil__", &ceil_str);
+    if (method == NULL) {
+        if (PyErr_Occurred())
             return NULL;
-    }
-
-    method = _PyType_Lookup(Py_TYPE(number), ceil_str);
-    if (method == NULL)
         return math_1_to_int(number, ceil, 0);
-    else
-        return PyObject_CallFunction(method, "O", number);
+    }
+    result = PyObject_CallFunctionObjArgs(method, NULL);
+    Py_DECREF(method);
+    return result;
 }
 
 PyDoc_STRVAR(math_ceil_doc,
@@ -881,19 +875,17 @@
 
 static PyObject * math_floor(PyObject *self, PyObject *number) {
     static PyObject *floor_str = NULL;
-    PyObject *method;
+    PyObject *method, *result;
 
-    if (floor_str == NULL) {
-        floor_str = PyUnicode_InternFromString("__floor__");
-        if (floor_str == NULL)
+    method = _PyObject_LookupSpecial(number, "__floor__", &floor_str);
+    if (method == NULL) {
+        if (PyErr_Occurred())
             return NULL;
-    }
-
-    method = _PyType_Lookup(Py_TYPE(number), floor_str);
-    if (method == NULL)
         return math_1_to_int(number, floor, 0);
-    else
-        return PyObject_CallFunction(method, "O", number);
+    }
+    result = PyObject_CallFunctionObjArgs(method, NULL);
+    Py_DECREF(method);
+    return result;
 }
 
 PyDoc_STRVAR(math_floor_doc,
@@ -904,7 +896,7 @@
       "gamma(x)\n\nGamma function at x.")
 FUNC1A(lgamma, m_lgamma,
       "lgamma(x)\n\nNatural logarithm of absolute value of Gamma function at x.")
-FUNC1(log1p, m_log1p, 1,
+FUNC1(log1p, m_log1p, 0,
       "log1p(x)\n\nReturn the natural logarithm of 1+x (base e).\n"
       "The result is computed in a way which is accurate for x near zero.")
 FUNC1(sin, sin, 0,
@@ -989,17 +981,17 @@
    def msum(iterable):
        partials = []  # sorted, non-overlapping partial sums
        for x in iterable:
-       i = 0
-       for y in partials:
-           if abs(x) < abs(y):
-           x, y = y, x
-           hi = x + y
-           lo = y - (hi - x)
-           if lo:
-           partials[i] = lo
-           i += 1
-           x = hi
-       partials[i:] = [x]
+           i = 0
+           for y in partials:
+               if abs(x) < abs(y):
+                   x, y = y, x
+               hi = x + y
+               lo = y - (hi - x)
+               if lo:
+                   partials[i] = lo
+                   i += 1
+               x = hi
+           partials[i:] = [x]
        return sum_exact(partials)
 
    Rounded x+y stored in hi with the roundoff stored in lo.  Together hi+lo
@@ -1178,7 +1170,7 @@
  * http://www.luschny.de/math/factorial/binarysplitfact.html
  *
  * Faster algorithms exist, but they're more complicated and depend on
- * a fast prime factoriazation algorithm.
+ * a fast prime factorization algorithm.
  *
  * Notes on the algorithm
  * ----------------------
@@ -1420,27 +1412,24 @@
 math_trunc(PyObject *self, PyObject *number)
 {
     static PyObject *trunc_str = NULL;
-    PyObject *trunc;
+    PyObject *trunc, *result;
 
     if (Py_TYPE(number)->tp_dict == NULL) {
         if (PyType_Ready(Py_TYPE(number)) < 0)
             return NULL;
     }
 
-    if (trunc_str == NULL) {
-        trunc_str = PyUnicode_InternFromString("__trunc__");
-        if (trunc_str == NULL)
-            return NULL;
-    }
-
-    trunc = _PyType_Lookup(Py_TYPE(number), trunc_str);
+    trunc = _PyObject_LookupSpecial(number, "__trunc__", &trunc_str);
     if (trunc == NULL) {
-        PyErr_Format(PyExc_TypeError,
-                     "type %.100s doesn't define __trunc__ method",
-                     Py_TYPE(number)->tp_name);
+        if (!PyErr_Occurred())
+            PyErr_Format(PyExc_TypeError,
+                         "type %.100s doesn't define __trunc__ method",
+                         Py_TYPE(number)->tp_name);
         return NULL;
     }
-    return PyObject_CallFunctionObjArgs(trunc, number, NULL);
+    result = PyObject_CallFunctionObjArgs(trunc, NULL);
+    Py_DECREF(trunc);
+    return result;
 }
 
 PyDoc_STRVAR(math_trunc_doc,
@@ -1829,6 +1818,19 @@
 Convert angle x from degrees to radians.");
 
 static PyObject *
+math_isfinite(PyObject *self, PyObject *arg)
+{
+    double x = PyFloat_AsDouble(arg);
+    if (x == -1.0 && PyErr_Occurred())
+        return NULL;
+    return PyBool_FromLong((long)Py_IS_FINITE(x));
+}
+
+PyDoc_STRVAR(math_isfinite_doc,
+"isfinite(x) -> bool\n\n\
+Return True if x is neither an infinity nor a NaN, and False otherwise.");
+
+static PyObject *
 math_isnan(PyObject *self, PyObject *arg)
 {
     double x = PyFloat_AsDouble(arg);
@@ -1839,7 +1841,7 @@
 
 PyDoc_STRVAR(math_isnan_doc,
 "isnan(x) -> bool\n\n\
-Check if float x is not a number (NaN).");
+Return True if x is a NaN (not a number), and False otherwise.");
 
 static PyObject *
 math_isinf(PyObject *self, PyObject *arg)
@@ -1852,7 +1854,7 @@
 
 PyDoc_STRVAR(math_isinf_doc,
 "isinf(x) -> bool\n\n\
-Check if float x is infinite (positive or negative).");
+Return True if x is a positive or negative infinity, and False otherwise.");
 
 static PyMethodDef math_methods[] = {
     {"acos",            math_acos,      METH_O,         math_acos_doc},
@@ -1879,6 +1881,7 @@
     {"fsum",            math_fsum,      METH_O,         math_fsum_doc},
     {"gamma",           math_gamma,     METH_O,         math_gamma_doc},
     {"hypot",           math_hypot,     METH_VARARGS,   math_hypot_doc},
+    {"isfinite",        math_isfinite,  METH_O,         math_isfinite_doc},
     {"isinf",           math_isinf,     METH_O,         math_isinf_doc},
     {"isnan",           math_isnan,     METH_O,         math_isnan_doc},
     {"ldexp",           math_ldexp,     METH_VARARGS,   math_ldexp_doc},

Modified: python/branches/py3k-cdecimal/Modules/parsermodule.c
==============================================================================
--- python/branches/py3k-cdecimal/Modules/parsermodule.c	(original)
+++ python/branches/py3k-cdecimal/Modules/parsermodule.c	Sat Jul 17 12:36:09 2010
@@ -1754,21 +1754,21 @@
                 && validate_dotted_as_names(CHILD(tree, 1)));
 }
 
-/* Helper function to count the number of leading dots in
+/* Helper function to count the number of leading dots (or ellipsis tokens) in
  * 'from ...module import name'
  */
 static int
 count_from_dots(node *tree)
 {
-        int i;
-        for (i = 1; i < NCH(tree); i++)
-                if (TYPE(CHILD(tree, i)) != DOT)
-                        break;
-        return i-1;
+    int i;
+    for (i = 1; i < NCH(tree); i++)
+        if (TYPE(CHILD(tree, i)) != DOT && TYPE(CHILD(tree, i)) != ELLIPSIS)
+            break;
+    return i - 1;
 }
 
-/* 'from' ('.'* dotted_name | '.') 'import' ('*' | '(' import_as_names ')' |
- *     import_as_names
+/* import_from: ('from' ('.'* dotted_name | '.'+)
+ *               'import' ('*' | '(' import_as_names ')' | import_as_names))
  */
 static int
 validate_import_from(node *tree)
@@ -1778,7 +1778,8 @@
         int havename = (TYPE(CHILD(tree, ndots + 1)) == dotted_name);
         int offset = ndots + havename;
         int res = validate_ntype(tree, import_from)
-                && (nch >= 4 + ndots)
+                && (offset >= 1)
+                && (nch >= 3 + offset)
                 && validate_name(CHILD(tree, 0), "from")
                 && (!havename || validate_dotted_name(CHILD(tree, ndots + 1)))
                 && validate_name(CHILD(tree, offset + 1), "import");
@@ -2509,14 +2510,15 @@
 static int
 validate_decorated(node *tree)
 {
-  int nch = NCH(tree);
-  int ok = (validate_ntype(tree, decorated)
-            && (nch == 2)
-            && validate_decorators(RCHILD(tree, -2))
-            && (validate_funcdef(RCHILD(tree, -1))
-                || validate_class(RCHILD(tree, -1)))
-            );
-  return ok;
+    int nch = NCH(tree);
+    int ok = (validate_ntype(tree, decorated)
+              && (nch == 2)
+              && validate_decorators(RCHILD(tree, -2)));
+    if (TYPE(RCHILD(tree, -1)) == funcdef)
+        ok = ok && validate_funcdef(RCHILD(tree, -1));
+    else
+        ok = ok && validate_class(RCHILD(tree, -1));
+    return ok;
 }
 
 static int

Modified: python/branches/py3k-cdecimal/Modules/posixmodule.c
==============================================================================
--- python/branches/py3k-cdecimal/Modules/posixmodule.c	(original)
+++ python/branches/py3k-cdecimal/Modules/posixmodule.c	Sat Jul 17 12:36:09 2010
@@ -571,7 +571,7 @@
 
 #ifdef MS_WINDOWS
 static PyObject *
-win32_error(char* function, char* filename)
+win32_error(char* function, const char* filename)
 {
     /* XXX We should pass the function name along in the future.
        (winreg.c also wants to pass the function name.)
@@ -978,12 +978,28 @@
     return TRUE;
 }
 
-static int
-win32_stat(const char* path, struct win32_stat *result)
+/* About the following functions: win32_lstat, win32_lstat_w, win32_stat,
+   win32_stat_w
+
+   In Posix, stat automatically traverses symlinks and returns the stat
+   structure for the target.  In Windows, the equivalent GetFileAttributes by
+   default does not traverse symlinks and instead returns attributes for
+   the symlink.
+
+   Therefore, win32_lstat will get the attributes traditionally, and
+   win32_stat will first explicitly resolve the symlink target and then will
+   call win32_lstat on that result.
+
+   The _w represent Unicode equivalents of the aformentioned ANSI functions. */
+
+static int 
+win32_lstat(const char* path, struct win32_stat *result)
 {
     WIN32_FILE_ATTRIBUTE_DATA info;
     int code;
     char *dot;
+    WIN32_FIND_DATAA find_data;
+    HANDLE find_data_handle;
     if (!GetFileAttributesExA(path, GetFileExInfoStandard, &info)) {
         if (GetLastError() != ERROR_SHARING_VIOLATION) {
             /* Protocol violation: we explicitly clear errno, instead of
@@ -1000,27 +1016,44 @@
             }
         }
     }
+
     code = attribute_data_to_stat(&info, result);
     if (code != 0)
         return code;
+
+    /* Get WIN32_FIND_DATA structure for the path to determine if
+       it is a symlink */
+    if(info.dwFileAttributes & FILE_ATTRIBUTE_REPARSE_POINT) {
+        find_data_handle = FindFirstFileA(path, &find_data);
+        if(find_data_handle != INVALID_HANDLE_VALUE) {
+            if(find_data.dwReserved0 == IO_REPARSE_TAG_SYMLINK) {
+                /* first clear the S_IFMT bits */
+                result->st_mode ^= (result->st_mode & 0170000);
+                /* now set the bits that make this a symlink */
+                result->st_mode |= 0120000;
+            }
+            FindClose(find_data_handle);
+        }
+    }
+
     /* Set S_IFEXEC if it is an .exe, .bat, ... */
     dot = strrchr(path, '.');
     if (dot) {
-        if (stricmp(dot, ".bat") == 0 ||
-        stricmp(dot, ".cmd") == 0 ||
-        stricmp(dot, ".exe") == 0 ||
-        stricmp(dot, ".com") == 0)
+        if (stricmp(dot, ".bat") == 0 || stricmp(dot, ".cmd") == 0 ||
+            stricmp(dot, ".exe") == 0 || stricmp(dot, ".com") == 0)
             result->st_mode |= 0111;
     }
     return code;
 }
 
 static int
-win32_wstat(const wchar_t* path, struct win32_stat *result)
+win32_lstat_w(const wchar_t* path, struct win32_stat *result)
 {
     int code;
     const wchar_t *dot;
     WIN32_FILE_ATTRIBUTE_DATA info;
+    WIN32_FIND_DATAW find_data;
+    HANDLE find_data_handle;
     if (!GetFileAttributesExW(path, GetFileExInfoStandard, &info)) {
         if (GetLastError() != ERROR_SHARING_VIOLATION) {
             /* Protocol violation: we explicitly clear errno, instead of
@@ -1028,8 +1061,8 @@
             errno = 0;
             return -1;
         } else {
-            /* Could not get attributes on open file. Fall back to
-               reading the directory. */
+            /* Could not get attributes on open file. Fall back to reading
+            the directory. */
             if (!attributes_from_dir_w(path, &info)) {
                 /* Very strange. This should not fail now */
                 errno = 0;
@@ -1040,18 +1073,199 @@
     code = attribute_data_to_stat(&info, result);
     if (code < 0)
         return code;
+
+    /* Get WIN32_FIND_DATA structure for the path to determine if
+       it is a symlink */
+    if(info.dwFileAttributes & FILE_ATTRIBUTE_REPARSE_POINT) {
+        find_data_handle = FindFirstFileW(path, &find_data);
+        if(find_data_handle != INVALID_HANDLE_VALUE) {
+            if(find_data.dwReserved0 == IO_REPARSE_TAG_SYMLINK) {
+                /* first clear the S_IFMT bits */
+                result->st_mode ^= (result->st_mode & 0170000);
+                /* now set the bits that make this a symlink */
+                result->st_mode |= 0120000;
+            }
+            FindClose(find_data_handle);
+        }
+    }
+
     /* Set IFEXEC if it is an .exe, .bat, ... */
     dot = wcsrchr(path, '.');
     if (dot) {
-        if (_wcsicmp(dot, L".bat") == 0 ||
-            _wcsicmp(dot, L".cmd") == 0 ||
-            _wcsicmp(dot, L".exe") == 0 ||
-            _wcsicmp(dot, L".com") == 0)
+        if (_wcsicmp(dot, L".bat") == 0 || _wcsicmp(dot, L".cmd") == 0 ||
+            _wcsicmp(dot, L".exe") == 0 || _wcsicmp(dot, L".com") == 0)
             result->st_mode |= 0111;
     }
     return code;
 }
 
+/* Grab GetFinalPathNameByHandle dynamically from kernel32 */
+static int has_GetFinalPathNameByHandle = 0;
+static DWORD (CALLBACK *Py_GetFinalPathNameByHandleA)(HANDLE, LPSTR, DWORD,
+                                                      DWORD);
+static DWORD (CALLBACK *Py_GetFinalPathNameByHandleW)(HANDLE, LPWSTR, DWORD,
+                                                      DWORD);
+static int
+check_GetFinalPathNameByHandle()
+{
+    HINSTANCE hKernel32;
+    /* only recheck */
+    if (!has_GetFinalPathNameByHandle)
+    {
+        hKernel32 = GetModuleHandle("KERNEL32");
+        *(FARPROC*)&Py_GetFinalPathNameByHandleA = GetProcAddress(hKernel32,
+                                                "GetFinalPathNameByHandleA");
+        *(FARPROC*)&Py_GetFinalPathNameByHandleW = GetProcAddress(hKernel32,
+                                                "GetFinalPathNameByHandleW");
+        has_GetFinalPathNameByHandle = Py_GetFinalPathNameByHandleA &&
+                                       Py_GetFinalPathNameByHandleW;
+    }
+    return has_GetFinalPathNameByHandle;
+}
+
+static int
+win32_stat(const char* path, struct win32_stat *result)
+{
+    /* Traverse the symlink to the target using
+    GetFinalPathNameByHandle()
+    */
+    int code;
+    HANDLE hFile;
+    int buf_size;
+    char *target_path;
+    int result_length;
+    WIN32_FILE_ATTRIBUTE_DATA info;
+    
+    if(!check_GetFinalPathNameByHandle()) {
+        /* if the OS doesn't have GetFinalPathNameByHandle, it doesn't
+           have symlinks, so just fall back to the traditional behavior
+           found in lstat. */
+        return win32_lstat(path, result);
+    }
+
+    hFile = CreateFileA(
+        path,
+        0, /* desired access */
+        0, /* share mode */
+        NULL, /* security attributes */
+        OPEN_EXISTING,
+        /* FILE_FLAG_BACKUP_SEMANTICS is required to open a directory */
+        FILE_FLAG_BACKUP_SEMANTICS,
+        NULL);
+    
+    if(hFile == INVALID_HANDLE_VALUE) {
+        /* Either the target doesn't exist, or we don't have access to
+           get a handle to it. If the former, we need to return an error.
+           If the latter, we can use attributes_from_dir. */
+        if (GetLastError() != ERROR_SHARING_VIOLATION) {
+            /* Protocol violation: we explicitly clear errno, instead of
+               setting it to a POSIX error. Callers should use GetLastError. */
+            errno = 0;
+            return -1;
+        } else {
+            /* Could not get attributes on open file. Fall back to
+               reading the directory. */
+            if (!attributes_from_dir(path, &info)) {
+                /* Very strange. This should not fail now */
+                errno = 0;
+                return -1;
+            }
+        }
+        code = attribute_data_to_stat(&info, result);
+    }
+    
+    buf_size = Py_GetFinalPathNameByHandleA(hFile, 0, 0, VOLUME_NAME_DOS);
+    if(!buf_size) return -1;
+    target_path = (char *)malloc((buf_size+1)*sizeof(char));
+    result_length = Py_GetFinalPathNameByHandleA(hFile, target_path,
+                                                 buf_size, VOLUME_NAME_DOS);
+    
+    if(!result_length)
+        return -1;
+
+    if(!CloseHandle(hFile))
+        return -1;
+
+    target_path[result_length] = 0;
+    code = win32_lstat(target_path, result);
+    free(target_path);
+    
+    return code;
+}
+
+static int 
+win32_stat_w(const wchar_t* path, struct win32_stat *result)
+{
+    /* Traverse the symlink to the target using GetFinalPathNameByHandle() */
+    int code;
+    HANDLE hFile;
+    int buf_size;
+    wchar_t *target_path;
+    int result_length;
+    WIN32_FILE_ATTRIBUTE_DATA info;
+    
+    if(!check_GetFinalPathNameByHandle()) {
+        /* If the OS doesn't have GetFinalPathNameByHandle, it doesn't have
+           symlinks, so just fall back to the traditional behavior found
+           in lstat. */
+        return win32_lstat_w(path, result);
+    }
+
+    hFile = CreateFileW(
+        path,
+        0, /* desired access */
+        0, /* share mode */
+        NULL, /* security attributes */
+        OPEN_EXISTING,
+        /* FILE_FLAG_BACKUP_SEMANTICS is required to open a directory */
+        FILE_ATTRIBUTE_NORMAL|FILE_FLAG_BACKUP_SEMANTICS,
+        NULL);
+    
+    if(hFile == INVALID_HANDLE_VALUE) {
+        /* Either the target doesn't exist, or we don't have access to
+           get a handle to it. If the former, we need to return an error.
+           If the latter, we can use attributes_from_dir. */
+        if (GetLastError() != ERROR_SHARING_VIOLATION) {
+            /* Protocol violation: we explicitly clear errno, instead of
+               setting it to a POSIX error. Callers should use GetLastError. */
+            errno = 0;
+            return -1;
+        } else {
+            /* Could not get attributes on open file. Fall back to
+               reading the directory. */
+            if (!attributes_from_dir_w(path, &info)) {
+                /* Very strange. This should not fail now */
+                errno = 0;
+                return -1;
+            }
+        }
+        code = attribute_data_to_stat(&info, result);
+    }
+    else {
+        /* We have a good handle to the target, use it to determine the target
+           path name (then we'll call lstat on it). */
+        buf_size = Py_GetFinalPathNameByHandleW(hFile, 0, 0, VOLUME_NAME_DOS);
+        if(!buf_size)
+            return -1;
+
+        target_path = (wchar_t *)malloc((buf_size+1)*sizeof(wchar_t));
+        result_length = Py_GetFinalPathNameByHandleW(hFile, target_path,
+                                                buf_size, VOLUME_NAME_DOS);
+        
+        if(!result_length)
+            return -1;
+
+        if(!CloseHandle(hFile))
+            return -1;
+
+        target_path[result_length] = 0;
+        code = win32_lstat_w(target_path, result);
+        free(target_path);
+    }
+    
+    return code;
+}
+
 static int
 win32_fstat(int file_number, struct win32_stat *result)
 {
@@ -1097,9 +1311,12 @@
     /* similar to stat() */
     result->st_mode = attributes_to_mode(info.dwFileAttributes);
     result->st_size = (((__int64)info.nFileSizeHigh)<<32) + info.nFileSizeLow;
-    FILE_TIME_to_time_t_nsec(&info.ftCreationTime, &result->st_ctime, &result->st_ctime_nsec);
-    FILE_TIME_to_time_t_nsec(&info.ftLastWriteTime, &result->st_mtime, &result->st_mtime_nsec);
-    FILE_TIME_to_time_t_nsec(&info.ftLastAccessTime, &result->st_atime, &result->st_atime_nsec);
+    FILE_TIME_to_time_t_nsec(&info.ftCreationTime, &result->st_ctime,
+                             &result->st_ctime_nsec);
+    FILE_TIME_to_time_t_nsec(&info.ftLastWriteTime, &result->st_mtime,
+                             &result->st_mtime_nsec);
+    FILE_TIME_to_time_t_nsec(&info.ftLastAccessTime, &result->st_atime,
+                             &result->st_atime_nsec);
     /* specific to fstat() */
     result->st_nlink = info.nNumberOfLinks;
     result->st_ino = (((__int64)info.nFileIndexHigh)<<32) + info.nFileIndexLow;
@@ -2443,6 +2660,70 @@
     }
     return PyBytes_FromString(outbuf);
 } /* end of posix__getfullpathname */
+
+/* A helper function for samepath on windows */
+static PyObject *
+posix__getfinalpathname(PyObject *self, PyObject *args)
+{
+    HANDLE hFile;
+    int buf_size;
+    wchar_t *target_path;
+    int result_length;
+    PyObject *result;
+    wchar_t *path;
+    
+    if (!PyArg_ParseTuple(args, "u|:_getfullpathname", &path)) {
+        return NULL;
+    }
+
+    if(!check_GetFinalPathNameByHandle()) {
+        /* If the OS doesn't have GetFinalPathNameByHandle, return a
+           NotImplementedError. */
+        return PyErr_Format(PyExc_NotImplementedError,
+            "GetFinalPathNameByHandle not available on this platform");
+    }
+
+    hFile = CreateFileW(
+        path,
+        0, /* desired access */
+        0, /* share mode */
+        NULL, /* security attributes */
+        OPEN_EXISTING,
+        /* FILE_FLAG_BACKUP_SEMANTICS is required to open a directory */
+        FILE_FLAG_BACKUP_SEMANTICS,
+        NULL);
+    
+    if(hFile == INVALID_HANDLE_VALUE) {
+        return win32_error_unicode("GetFinalPathNamyByHandle", path);
+        return PyErr_Format(PyExc_RuntimeError,
+                            "Could not get a handle to file.");
+    }
+
+    /* We have a good handle to the target, use it to determine the
+       target path name. */
+    buf_size = Py_GetFinalPathNameByHandleW(hFile, 0, 0, VOLUME_NAME_NT);
+
+    if(!buf_size)
+        return win32_error_unicode("GetFinalPathNameByHandle", path);
+
+    target_path = (wchar_t *)malloc((buf_size+1)*sizeof(wchar_t));
+    if(!target_path)
+        return PyErr_NoMemory();
+
+    result_length = Py_GetFinalPathNameByHandleW(hFile, target_path,
+                                                 buf_size, VOLUME_NAME_DOS);
+    if(!result_length)
+        return win32_error_unicode("GetFinalPathNamyByHandle", path);
+
+    if(!CloseHandle(hFile))
+        return win32_error_unicode("GetFinalPathNameByHandle", path);
+
+    target_path[result_length] = 0;
+    result = PyUnicode_FromUnicode(target_path, result_length);
+    free(target_path);
+    return result;
+
+} /* end of posix__getfinalpathname */
 #endif /* MS_WINDOWS */
 
 PyDoc_STRVAR(posix_mkdir__doc__,
@@ -2623,7 +2904,7 @@
 posix_stat(PyObject *self, PyObject *args)
 {
 #ifdef MS_WINDOWS
-    return posix_do_stat(self, args, "O&:stat", STAT, "U:stat", win32_wstat);
+    return posix_do_stat(self, args, "O&:stat", STAT, "U:stat", win32_stat_w);
 #else
     return posix_do_stat(self, args, "O&:stat", STAT, NULL, NULL);
 #endif
@@ -2681,6 +2962,41 @@
     return PyLong_FromLong((long)i);
 }
 
+#ifdef MS_WINDOWS
+
+/* override the default DeleteFileW behavior so that directory
+symlinks can be removed with this function, the same as with
+Unix symlinks */
+BOOL WINAPI Py_DeleteFileW(LPCWSTR lpFileName)
+{
+    WIN32_FILE_ATTRIBUTE_DATA info;
+    WIN32_FIND_DATAW find_data;
+    HANDLE find_data_handle;
+    int is_directory = 0;
+    int is_link = 0;
+
+    if (GetFileAttributesExW(lpFileName, GetFileExInfoStandard, &info)) {
+        is_directory = info.dwFileAttributes & FILE_ATTRIBUTE_DIRECTORY;
+        
+        /* Get WIN32_FIND_DATA structure for the path to determine if
+           it is a symlink */
+        if(is_directory &&
+           info.dwFileAttributes & FILE_ATTRIBUTE_REPARSE_POINT) {
+            find_data_handle = FindFirstFileW(lpFileName, &find_data);
+
+            if(find_data_handle != INVALID_HANDLE_VALUE) {
+                is_link = find_data.dwReserved0 == IO_REPARSE_TAG_SYMLINK;
+                FindClose(find_data_handle);
+            }
+        }
+    }
+
+    if (is_directory && is_link)
+        return RemoveDirectoryW(lpFileName);
+
+    return DeleteFileW(lpFileName);
+}
+#endif /* MS_WINDOWS */
 
 PyDoc_STRVAR(posix_unlink__doc__,
 "unlink(path)\n\n\
@@ -2694,7 +3010,8 @@
 posix_unlink(PyObject *self, PyObject *args)
 {
 #ifdef MS_WINDOWS
-    return win32_1str(args, "remove", "y:remove", DeleteFileA, "U:remove", DeleteFileW);
+    return win32_1str(args, "remove", "y:remove", DeleteFileA,
+                      "U:remove", Py_DeleteFileW);
 #else
     return posix_1str(args, "O&:remove", unlink);
 #endif
@@ -3925,7 +4242,7 @@
 #ifdef HAVE_SETPGRP
 PyDoc_STRVAR(posix_setpgrp__doc__,
 "setpgrp()\n\n\
-Make this process a session leader.");
+Make this process the process group leader.");
 
 static PyObject *
 posix_setpgrp(PyObject *self, PyObject *noargs)
@@ -4544,7 +4861,8 @@
     return posix_do_stat(self, args, "O&:lstat", lstat, NULL, NULL);
 #else /* !HAVE_LSTAT */
 #ifdef MS_WINDOWS
-    return posix_do_stat(self, args, "O&:lstat", STAT, "U:lstat", win32_wstat);
+    return posix_do_stat(self, args, "O&:lstat", STAT, "U:lstat",
+                         win32_lstat_w);
 #else
     return posix_do_stat(self, args, "O&:lstat", STAT, NULL, NULL);
 #endif
@@ -4609,6 +4927,198 @@
 }
 #endif /* HAVE_SYMLINK */
 
+#if !defined(HAVE_READLINK) && defined(MS_WINDOWS)
+
+PyDoc_STRVAR(win_readlink__doc__,
+"readlink(path) -> path\n\n\
+Return a string representing the path to which the symbolic link points.");
+
+/* The following structure was copied from
+   http://msdn.microsoft.com/en-us/library/ms791514.aspx as the required
+   include doesn't seem to be present in the Windows SDK (at least as included
+   with Visual Studio Express). */
+typedef struct _REPARSE_DATA_BUFFER {
+    ULONG ReparseTag;
+    USHORT ReparseDataLength;
+    USHORT Reserved;
+    union {
+        struct {
+            USHORT SubstituteNameOffset;
+            USHORT SubstituteNameLength;
+            USHORT PrintNameOffset;
+            USHORT PrintNameLength;
+            ULONG Flags;
+            WCHAR PathBuffer[1];
+        } SymbolicLinkReparseBuffer;
+
+        struct {
+            USHORT SubstituteNameOffset;
+            USHORT  SubstituteNameLength;
+            USHORT  PrintNameOffset;
+            USHORT  PrintNameLength;
+            WCHAR  PathBuffer[1];
+        } MountPointReparseBuffer;
+
+        struct {
+            UCHAR  DataBuffer[1];
+        } GenericReparseBuffer;
+    };
+} REPARSE_DATA_BUFFER, *PREPARSE_DATA_BUFFER;
+
+#define REPARSE_DATA_BUFFER_HEADER_SIZE  FIELD_OFFSET(REPARSE_DATA_BUFFER,\
+                                                      GenericReparseBuffer)
+
+#define MAXIMUM_REPARSE_DATA_BUFFER_SIZE  ( 16 * 1024 )
+
+/* Windows readlink implementation */
+static PyObject *
+win_readlink(PyObject *self, PyObject *args)
+{
+    wchar_t *path;
+    DWORD n_bytes_returned;
+    DWORD io_result;
+    PyObject *result;
+    HANDLE reparse_point_handle;
+
+    char target_buffer[MAXIMUM_REPARSE_DATA_BUFFER_SIZE];
+    REPARSE_DATA_BUFFER *rdb = (REPARSE_DATA_BUFFER *)target_buffer;
+    wchar_t *print_name;
+
+    if (!PyArg_ParseTuple(args,
+                  "u:readlink",
+                  &path))
+        return NULL;
+
+    /* First get a handle to the reparse point */
+    Py_BEGIN_ALLOW_THREADS
+    reparse_point_handle = CreateFileW(
+        path,
+        0,
+        0,
+        0,
+        OPEN_EXISTING,
+        FILE_FLAG_OPEN_REPARSE_POINT|FILE_FLAG_BACKUP_SEMANTICS,
+        0);
+    Py_END_ALLOW_THREADS
+    
+    if (reparse_point_handle==INVALID_HANDLE_VALUE)
+    {
+        return win32_error_unicode("readlink", path);
+    }
+    
+    Py_BEGIN_ALLOW_THREADS
+    /* New call DeviceIoControl to read the reparse point */
+    io_result = DeviceIoControl(
+        reparse_point_handle,
+        FSCTL_GET_REPARSE_POINT,
+        0, 0, /* in buffer */
+        target_buffer, sizeof(target_buffer),
+        &n_bytes_returned,
+        0 /* we're not using OVERLAPPED_IO */
+        );
+    CloseHandle(reparse_point_handle);
+    Py_END_ALLOW_THREADS
+
+    if (io_result==0)
+    {
+        return win32_error_unicode("readlink", path);
+    }
+
+    if (rdb->ReparseTag != IO_REPARSE_TAG_SYMLINK)
+    {
+        PyErr_SetString(PyExc_ValueError,
+                "not a symbolic link");
+        return NULL;
+    }
+    print_name = rdb->SymbolicLinkReparseBuffer.PathBuffer +
+                 rdb->SymbolicLinkReparseBuffer.PrintNameOffset;
+
+    result = PyUnicode_FromWideChar(print_name,
+                    rdb->SymbolicLinkReparseBuffer.PrintNameLength/2);
+    return result;
+}
+
+#endif /* !defined(HAVE_READLINK) && defined(MS_WINDOWS) */
+
+#if !defined(HAVE_SYMLINK) && defined(MS_WINDOWS)
+
+/* Grab CreateSymbolicLinkW dynamically from kernel32 */
+static int has_CreateSymbolicLinkW = 0;
+static DWORD (CALLBACK *Py_CreateSymbolicLinkW)(LPWSTR, LPWSTR, DWORD);
+static int
+check_CreateSymbolicLinkW()
+{
+    HINSTANCE hKernel32;
+    /* only recheck */
+    if (has_CreateSymbolicLinkW)
+        return has_CreateSymbolicLinkW;
+    hKernel32 = GetModuleHandle("KERNEL32");
+    *(FARPROC*)&Py_CreateSymbolicLinkW = GetProcAddress(hKernel32,
+                                                        "CreateSymbolicLinkW");
+    if (Py_CreateSymbolicLinkW)
+        has_CreateSymbolicLinkW = 1;
+    return has_CreateSymbolicLinkW;
+}
+
+PyDoc_STRVAR(win_symlink__doc__,
+"symlink(src, dst, target_is_directory=False)\n\n\
+Create a symbolic link pointing to src named dst.\n\
+target_is_directory is required if the target is to be interpreted as\n\
+a directory.\n\
+This function requires Windows 6.0 or greater, and raises a\n\
+NotImplementedError otherwise.");
+
+static PyObject *
+win_symlink(PyObject *self, PyObject *args, PyObject *kwargs)
+{
+    static char *kwlist[] = {"src", "dest", "target_is_directory", NULL};
+    PyObject *src, *dest;
+    int target_is_directory = 0;
+    DWORD res;
+    WIN32_FILE_ATTRIBUTE_DATA src_info;
+    
+    if (!check_CreateSymbolicLinkW())
+    {
+        /* raise NotImplementedError */
+        return PyErr_Format(PyExc_NotImplementedError,
+            "CreateSymbolicLinkW not found");
+    }
+    if (!PyArg_ParseTupleAndKeywords(args, kwargs, "OO|i:symlink",
+        kwlist, &src, &dest, &target_is_directory))
+        return NULL;
+    if (!convert_to_unicode(&src)) { return NULL; }
+    if (!convert_to_unicode(&dest)) {
+        Py_DECREF(src);
+        return NULL;
+    }
+    
+    /* if src is a directory, ensure target_is_directory==1 */
+    if(
+        GetFileAttributesExW(
+            PyUnicode_AsUnicode(src), GetFileExInfoStandard, &src_info
+        ))
+    {
+        target_is_directory = target_is_directory ||
+            (src_info.dwFileAttributes & FILE_ATTRIBUTE_DIRECTORY);
+    }
+
+    Py_BEGIN_ALLOW_THREADS
+    res = Py_CreateSymbolicLinkW(
+        PyUnicode_AsUnicode(dest),
+        PyUnicode_AsUnicode(src),
+        target_is_directory);
+    Py_END_ALLOW_THREADS
+    Py_DECREF(src);
+    Py_DECREF(dest);
+    if (!res)
+    {
+        return win32_error_unicode("symlink", PyUnicode_AsUnicode(src));
+    }
+    
+    Py_INCREF(Py_None);
+    return Py_None;
+}
+#endif /* !defined(HAVE_SYMLINK) && defined(MS_WINDOWS) */
 
 #ifdef HAVE_TIMES
 #if defined(PYCC_VACPP) && defined(PYOS_OS2)
@@ -7076,13 +7586,20 @@
 #ifdef HAVE_READLINK
     {"readlink",        posix_readlink, METH_VARARGS, posix_readlink__doc__},
 #endif /* HAVE_READLINK */
-    {"rename",          posix_rename, METH_VARARGS, posix_rename__doc__},
-    {"rmdir",           posix_rmdir, METH_VARARGS, posix_rmdir__doc__},
-    {"stat",            posix_stat, METH_VARARGS, posix_stat__doc__},
+#if !defined(HAVE_READLINK) && defined(MS_WINDOWS)
+    {"readlink",	win_readlink, METH_VARARGS, win_readlink__doc__},
+#endif /* !defined(HAVE_READLINK) && defined(MS_WINDOWS) */
+    {"rename",	posix_rename, METH_VARARGS, posix_rename__doc__},
+    {"rmdir",	posix_rmdir, METH_VARARGS, posix_rmdir__doc__},
+    {"stat",	posix_stat, METH_VARARGS, posix_stat__doc__},
     {"stat_float_times", stat_float_times, METH_VARARGS, stat_float_times__doc__},
 #ifdef HAVE_SYMLINK
     {"symlink",         posix_symlink, METH_VARARGS, posix_symlink__doc__},
 #endif /* HAVE_SYMLINK */
+#if !defined(HAVE_SYMLINK) && defined(MS_WINDOWS)
+    {"symlink", (PyCFunction)win_symlink, METH_VARARGS | METH_KEYWORDS,
+                win_symlink__doc__},
+#endif /* !defined(HAVE_SYMLINK) && defined(MS_WINDOWS) */
 #ifdef HAVE_SYSTEM
     {"system",          posix_system, METH_VARARGS, posix_system__doc__},
 #endif
@@ -7307,6 +7824,7 @@
     {"abort",           posix_abort, METH_NOARGS, posix_abort__doc__},
 #ifdef MS_WINDOWS
     {"_getfullpathname",        posix__getfullpathname, METH_VARARGS, NULL},
+    {"_getfinalpathname",       posix__getfinalpathname, METH_VARARGS, NULL},
 #endif
 #ifdef HAVE_GETLOADAVG
     {"getloadavg",      posix_getloadavg, METH_NOARGS, posix_getloadavg__doc__},

Modified: python/branches/py3k-cdecimal/Modules/selectmodule.c
==============================================================================
--- python/branches/py3k-cdecimal/Modules/selectmodule.c	(original)
+++ python/branches/py3k-cdecimal/Modules/selectmodule.c	Sat Jul 17 12:36:09 2010
@@ -91,7 +91,7 @@
     fd2obj[0].obj = (PyObject*)0;            /* set list to zero size */
     FD_ZERO(set);
 
-    fast_seq=PySequence_Fast(seq, "arguments 1-3 must be sequences");
+    fast_seq = PySequence_Fast(seq, "arguments 1-3 must be sequences");
     if (!fast_seq)
         return -1;
 

Modified: python/branches/py3k-cdecimal/Modules/timemodule.c
==============================================================================
--- python/branches/py3k-cdecimal/Modules/timemodule.c	(original)
+++ python/branches/py3k-cdecimal/Modules/timemodule.c	Sat Jul 17 12:36:09 2010
@@ -90,32 +90,6 @@
 /* For Y2K check */
 static PyObject *moddict;
 
-/* Exposed in timefuncs.h. */
-time_t
-_PyTime_DoubleToTimet(double x)
-{
-    time_t result;
-    double diff;
-
-    result = (time_t)x;
-    /* How much info did we lose?  time_t may be an integral or
-     * floating type, and we don't know which.  If it's integral,
-     * we don't know whether C truncates, rounds, returns the floor,
-     * etc.  If we lost a second or more, the C rounding is
-     * unreasonable, or the input just doesn't fit in a time_t;
-     * call it an error regardless.  Note that the original cast to
-     * time_t can cause a C error too, but nothing we can do to
-     * worm around that.
-     */
-    diff = x - (double)result;
-    if (diff <= -1.0 || diff >= 1.0) {
-        PyErr_SetString(PyExc_ValueError,
-                        "timestamp out of range for platform time_t");
-        result = (time_t)-1;
-    }
-    return result;
-}
-
 static PyObject *
 time_time(PyObject *self, PyObject *unused)
 {

Modified: python/branches/py3k-cdecimal/Modules/xxmodule.c
==============================================================================
--- python/branches/py3k-cdecimal/Modules/xxmodule.c	(original)
+++ python/branches/py3k-cdecimal/Modules/xxmodule.c	Sat Jul 17 12:36:09 2010
@@ -10,7 +10,7 @@
    your own types of attributes instead.  Maybe you want to name your
    local variables other than 'self'.  If your object type is needed in
    other files, you'll have to create a file "foobarobject.h"; see
-   intobject.h for an example. */
+   floatobject.h for an example. */
 
 /* Xxo objects */
 

Modified: python/branches/py3k-cdecimal/Objects/abstract.c
==============================================================================
--- python/branches/py3k-cdecimal/Objects/abstract.c	(original)
+++ python/branches/py3k-cdecimal/Objects/abstract.c	Sat Jul 17 12:36:09 2010
@@ -2522,8 +2522,7 @@
     }
     else {
         if (!check_class(cls,
-            "isinstance() arg 2 must be a class, type,"
-            " or tuple of classes and types"))
+            "isinstance() arg 2 must be a type or tuple of types"))
             return -1;
         icls = PyObject_GetAttr(inst, __class__);
         if (icls == NULL) {

Modified: python/branches/py3k-cdecimal/Objects/bytearrayobject.c
==============================================================================
--- python/branches/py3k-cdecimal/Objects/bytearrayobject.c	(original)
+++ python/branches/py3k-cdecimal/Objects/bytearrayobject.c	Sat Jul 17 12:36:09 2010
@@ -33,6 +33,7 @@
         PyObject *index = PyNumber_Index(arg);
         if (index == NULL) {
             PyErr_Format(PyExc_TypeError, "an integer is required");
+            *value = -1;
             return 0;
         }
         face_value = PyLong_AsLong(index);
@@ -42,6 +43,7 @@
     if (face_value < 0 || face_value >= 256) {
         /* this includes the OverflowError in case the long is too large */
         PyErr_SetString(PyExc_ValueError, "byte must be in range(0, 256)");
+        *value = -1;
         return 0;
     }
 

Modified: python/branches/py3k-cdecimal/Objects/bytes_methods.c
==============================================================================
--- python/branches/py3k-cdecimal/Objects/bytes_methods.c	(original)
+++ python/branches/py3k-cdecimal/Objects/bytes_methods.c	Sat Jul 17 12:36:09 2010
@@ -332,7 +332,8 @@
 PyDoc_STRVAR_shared(_Py_capitalize__doc__,
 "B.capitalize() -> copy of B\n\
 \n\
-Return a copy of B with only its first character capitalized (ASCII).");
+Return a copy of B with only its first character capitalized (ASCII)\n\
+and the rest lower-cased.");
 
 void
 _Py_bytes_capitalize(char *result, char *s, Py_ssize_t len)

Modified: python/branches/py3k-cdecimal/Objects/descrobject.c
==============================================================================
--- python/branches/py3k-cdecimal/Objects/descrobject.c	(original)
+++ python/branches/py3k-cdecimal/Objects/descrobject.c	Sat Jul 17 12:36:09 2010
@@ -1115,7 +1115,7 @@
 } propertyobject;
 
 static PyObject * property_copy(PyObject *, PyObject *, PyObject *,
-                                  PyObject *, PyObject *);
+                                  PyObject *);
 
 static PyMemberDef property_members[] = {
     {"fget", T_OBJECT, offsetof(propertyobject, prop_get), READONLY},
@@ -1132,7 +1132,7 @@
 static PyObject *
 property_getter(PyObject *self, PyObject *getter)
 {
-    return property_copy(self, getter, NULL, NULL, NULL);
+    return property_copy(self, getter, NULL, NULL);
 }
 
 
@@ -1142,7 +1142,7 @@
 static PyObject *
 property_setter(PyObject *self, PyObject *setter)
 {
-    return property_copy(self, NULL, setter, NULL, NULL);
+    return property_copy(self, NULL, setter, NULL);
 }
 
 
@@ -1152,7 +1152,7 @@
 static PyObject *
 property_deleter(PyObject *self, PyObject *deleter)
 {
-    return property_copy(self, NULL, NULL, deleter, NULL);
+    return property_copy(self, NULL, NULL, deleter);
 }
 
 
@@ -1221,11 +1221,10 @@
 }
 
 static PyObject *
-property_copy(PyObject *old, PyObject *get, PyObject *set, PyObject *del,
-                PyObject *doc)
+property_copy(PyObject *old, PyObject *get, PyObject *set, PyObject *del)
 {
     propertyobject *pold = (propertyobject *)old;
-    PyObject *new, *type;
+    PyObject *new, *type, *doc;
 
     type = PyObject_Type(old);
     if (type == NULL)
@@ -1243,15 +1242,12 @@
         Py_XDECREF(del);
         del = pold->prop_del ? pold->prop_del : Py_None;
     }
-    if (doc == NULL || doc == Py_None) {
-        Py_XDECREF(doc);
-        if (pold->getter_doc && get != Py_None) {
-            /* make _init use __doc__ from getter */
-            doc = Py_None;
-        }
-        else {
-            doc = pold->prop_doc ? pold->prop_doc : Py_None;
-        }
+    if (pold->getter_doc && get != Py_None) {
+        /* make _init use __doc__ from getter */
+        doc = Py_None;
+    }
+    else {
+        doc = pold->prop_doc ? pold->prop_doc : Py_None;
     }
 
     new =  PyObject_CallFunction(type, "OOOO", get, set, del, doc);

Modified: python/branches/py3k-cdecimal/Objects/enumobject.c
==============================================================================
--- python/branches/py3k-cdecimal/Objects/enumobject.c	(original)
+++ python/branches/py3k-cdecimal/Objects/enumobject.c	Sat Jul 17 12:36:09 2010
@@ -159,12 +159,13 @@
 }
 
 PyDoc_STRVAR(enum_doc,
-"enumerate(iterable) -> iterator for index, value of iterable\n"
+"enumerate(iterable[, start]) -> iterator for index, value of iterable\n"
 "\n"
 "Return an enumerate object.  iterable must be another object that supports\n"
 "iteration.  The enumerate object yields pairs containing a count (from\n"
-"zero) and a value yielded by the iterable argument.  enumerate is useful\n"
-"for obtaining an indexed list: (0, seq[0]), (1, seq[1]), (2, seq[2]), ...");
+"start, which defaults to zero) and a value yielded by the iterable argument.\n"
+"enumerate is useful for obtaining an indexed list:\n"
+"    (0, seq[0]), (1, seq[1]), (2, seq[2]), ...");
 
 PyTypeObject PyEnum_Type = {
     PyVarObject_HEAD_INIT(&PyType_Type, 0)

Modified: python/branches/py3k-cdecimal/Objects/exceptions.c
==============================================================================
--- python/branches/py3k-cdecimal/Objects/exceptions.c	(original)
+++ python/branches/py3k-cdecimal/Objects/exceptions.c	Sat Jul 17 12:36:09 2010
@@ -1622,9 +1622,6 @@
     const char *encoding, const char *object, Py_ssize_t length,
     Py_ssize_t start, Py_ssize_t end, const char *reason)
 {
-    assert(length < INT_MAX);
-    assert(start < INT_MAX);
-    assert(end < INT_MAX);
     return PyObject_CallFunction(PyExc_UnicodeDecodeError, "sy#nns",
                                  encoding, object, length, start, end, reason);
 }

Modified: python/branches/py3k-cdecimal/Objects/floatobject.c
==============================================================================
--- python/branches/py3k-cdecimal/Objects/floatobject.c	(original)
+++ python/branches/py3k-cdecimal/Objects/floatobject.c	Sat Jul 17 12:36:09 2010
@@ -21,7 +21,22 @@
 extern int finite(double);
 #endif
 
-/* Special free list -- see comments for same code in intobject.c. */
+/* Special free list
+
+   Since some Python programs can spend much of their time allocating
+   and deallocating floats, these operations should be very fast.
+   Therefore we use a dedicated allocation scheme with a much lower
+   overhead (in space and time) than straight malloc(): a simple
+   dedicated free list, filled when necessary with memory from malloc().
+
+   block_list is a singly-linked list of all PyFloatBlocks ever allocated,
+   linked via their next members.  PyFloatBlocks are never returned to the
+   system before shutdown (PyFloat_Fini).
+
+   free_list is a singly-linked list of available PyFloatObjects, linked
+   via abuse of their ob_type members.
+*/
+
 #define BLOCK_SIZE      1000    /* 1K less typical malloc overhead */
 #define BHEAD_SIZE      8       /* Enough for a 64-bit pointer */
 #define N_FLOATOBJECTS  ((BLOCK_SIZE - BHEAD_SIZE) / sizeof(PyFloatObject))
@@ -1157,7 +1172,7 @@
         return float_str((PyFloatObject *)v);
 
     if (x == 0.0) {
-        if(copysign(1.0, x) == -1.0)
+        if (copysign(1.0, x) == -1.0)
             return PyUnicode_FromString("-0x0.0p+0");
         else
             return PyUnicode_FromString("0x0.0p+0");
@@ -1407,7 +1422,7 @@
                     round_up = 1;
                     break;
                 }
-        if (round_up == 1) {
+        if (round_up) {
             x += 2*half_eps;
             if (top_exp == DBL_MAX_EXP &&
                 x == ldexp((double)(2*half_eps), DBL_MANT_DIG))

Modified: python/branches/py3k-cdecimal/Objects/frameobject.c
==============================================================================
--- python/branches/py3k-cdecimal/Objects/frameobject.c	(original)
+++ python/branches/py3k-cdecimal/Objects/frameobject.c	Sat Jul 17 12:36:09 2010
@@ -391,7 +391,7 @@
        the local variables in f_localsplus are NULL.
 
    2. We also maintain a separate free list of stack frames (just like
-   integers are allocated in a special way -- see intobject.c).  When
+   floats are allocated in a special way -- see floatobject.c).  When
    a stack frame is on the free list, only the following members have
    a meaning:
     ob_type             == &Frametype

Modified: python/branches/py3k-cdecimal/Objects/memoryobject.c
==============================================================================
--- python/branches/py3k-cdecimal/Objects/memoryobject.c	(original)
+++ python/branches/py3k-cdecimal/Objects/memoryobject.c	Sat Jul 17 12:36:09 2010
@@ -624,7 +624,7 @@
 static int
 memory_ass_sub(PyMemoryViewObject *self, PyObject *key, PyObject *value)
 {
-    Py_ssize_t start, len, bytelen, i;
+    Py_ssize_t start, len, bytelen;
     Py_buffer srcview;
     Py_buffer *view = &(self->view);
     char *srcbuf, *destbuf;
@@ -694,16 +694,8 @@
     if (destbuf + bytelen < srcbuf || srcbuf + bytelen < destbuf)
         /* No overlapping */
         memcpy(destbuf, srcbuf, bytelen);
-    else if (destbuf < srcbuf) {
-        /* Copy in ascending order */
-        for (i = 0; i < bytelen; i++)
-            destbuf[i] = srcbuf[i];
-    }
-    else {
-        /* Copy in descencing order */
-        for (i = bytelen - 1; i >= 0; i--)
-            destbuf[i] = srcbuf[i];
-    }
+    else
+        memmove(destbuf, srcbuf, bytelen);
 
     PyBuffer_Release(&srcview);
     return 0;

Modified: python/branches/py3k-cdecimal/Objects/stringlib/formatter.h
==============================================================================
--- python/branches/py3k-cdecimal/Objects/stringlib/formatter.h	(original)
+++ python/branches/py3k-cdecimal/Objects/stringlib/formatter.h	Sat Jul 17 12:36:09 2010
@@ -776,14 +776,6 @@
             goto done;
         }
 
-        /* Error to specify a comma. */
-        if (format->thousands_separators) {
-            PyErr_SetString(PyExc_ValueError,
-                            "Thousands separators not allowed with integer"
-                            " format specifier 'c'");
-            goto done;
-        }
-
         /* taken from unicodeobject.c formatchar() */
         /* Integer input truncated to a character */
 /* XXX: won't work for int */

Modified: python/branches/py3k-cdecimal/Objects/structseq.c
==============================================================================
--- python/branches/py3k-cdecimal/Objects/structseq.c	(original)
+++ python/branches/py3k-cdecimal/Objects/structseq.c	Sat Jul 17 12:36:09 2010
@@ -30,113 +30,30 @@
 PyStructSequence_New(PyTypeObject *type)
 {
     PyStructSequence *obj;
+    Py_ssize_t size = REAL_SIZE_TP(type), i;
 
-    obj = PyObject_New(PyStructSequence, type);
+    obj = PyObject_GC_NewVar(PyStructSequence, type, size);
     if (obj == NULL)
         return NULL;
+    /* Hack the size of the variable object, so invisible fields don't appear
+     to Python code. */
     Py_SIZE(obj) = VISIBLE_SIZE_TP(type);
+    for (i = 0; i < size; i++)
+        obj->ob_item[i] = NULL;
 
-    return (PyObject*) obj;
+    return (PyObject*)obj;
 }
 
 static void
 structseq_dealloc(PyStructSequence *obj)
 {
     Py_ssize_t i, size;
-
+    
     size = REAL_SIZE(obj);
     for (i = 0; i < size; ++i) {
         Py_XDECREF(obj->ob_item[i]);
     }
-    PyObject_Del(obj);
-}
-
-static Py_ssize_t
-structseq_length(PyStructSequence *obj)
-{
-    return VISIBLE_SIZE(obj);
-}
-
-static PyObject*
-structseq_item(PyStructSequence *obj, Py_ssize_t i)
-{
-    if (i < 0 || i >= VISIBLE_SIZE(obj)) {
-        PyErr_SetString(PyExc_IndexError, "tuple index out of range");
-        return NULL;
-    }
-    Py_INCREF(obj->ob_item[i]);
-    return obj->ob_item[i];
-}
-
-static PyObject*
-structseq_slice(PyStructSequence *obj, Py_ssize_t low, Py_ssize_t high)
-{
-    PyTupleObject *np;
-    Py_ssize_t i;
-
-    if (low < 0)
-        low = 0;
-    if (high > VISIBLE_SIZE(obj))
-        high = VISIBLE_SIZE(obj);
-    if (high < low)
-        high = low;
-    np = (PyTupleObject *)PyTuple_New(high-low);
-    if (np == NULL)
-        return NULL;
-    for(i = low; i < high; ++i) {
-        PyObject *v = obj->ob_item[i];
-        Py_INCREF(v);
-        PyTuple_SET_ITEM(np, i-low, v);
-    }
-    return (PyObject *) np;
-}
-
-static PyObject *
-structseq_subscript(PyStructSequence *self, PyObject *item)
-{
-    if (PyIndex_Check(item)) {
-        Py_ssize_t i = PyNumber_AsSsize_t(item, PyExc_IndexError);
-        if (i == -1 && PyErr_Occurred())
-            return NULL;
-
-        if (i < 0)
-            i += VISIBLE_SIZE(self);
-
-        if (i < 0 || i >= VISIBLE_SIZE(self)) {
-            PyErr_SetString(PyExc_IndexError,
-                "tuple index out of range");
-            return NULL;
-        }
-        Py_INCREF(self->ob_item[i]);
-        return self->ob_item[i];
-    }
-    else if (PySlice_Check(item)) {
-        Py_ssize_t start, stop, step, slicelen, cur, i;
-        PyObject *result;
-
-        if (PySlice_GetIndicesEx((PySliceObject *)item,
-                                 VISIBLE_SIZE(self), &start, &stop,
-                                 &step, &slicelen) < 0) {
-            return NULL;
-        }
-        if (slicelen <= 0)
-            return PyTuple_New(0);
-        result = PyTuple_New(slicelen);
-        if (result == NULL)
-            return NULL;
-        for (cur = start, i = 0; i < slicelen;
-             cur += step, i++) {
-            PyObject *v = self->ob_item[cur];
-            Py_INCREF(v);
-            PyTuple_SET_ITEM(result, i, v);
-        }
-        return result;
-    }
-    else {
-        PyErr_SetString(PyExc_TypeError,
-                        "structseq index must be integer");
-        return NULL;
-    }
+    PyObject_GC_Del(obj);
 }
 
 static PyObject *
@@ -223,11 +140,6 @@
     return (PyObject*) res;
 }
 
-static PyObject *
-make_tuple(PyStructSequence *obj)
-{
-    return structseq_slice(obj, 0, VISIBLE_SIZE(obj));
-}
 
 static PyObject *
 structseq_repr(PyStructSequence *obj)
@@ -236,7 +148,6 @@
 #define REPR_BUFFER_SIZE 512
 #define TYPE_MAXSIZE 100
 
-    PyObject *tup;
     PyTypeObject *typ = Py_TYPE(obj);
     int i, removelast = 0;
     Py_ssize_t len;
@@ -246,10 +157,6 @@
     /* pointer to end of writeable buffer; safes space for "...)\0" */
     endofbuf= &buf[REPR_BUFFER_SIZE-5];
 
-    if ((tup = make_tuple(obj)) == NULL) {
-        return NULL;
-    }
-
     /* "typename(", limited to  TYPE_MAXSIZE */
     len = strlen(typ->tp_name) > TYPE_MAXSIZE ? TYPE_MAXSIZE :
                             strlen(typ->tp_name);
@@ -262,19 +169,17 @@
         char *cname, *crepr;
 
         cname = typ->tp_members[i].name;
-
-        val = PyTuple_GetItem(tup, i);
-        if (cname == NULL || val == NULL) {
+        if (cname == NULL) {
+            PyErr_Format(PyExc_SystemError, "In structseq_repr(), member %d name is NULL"
+                         " for type %.500s", i, typ->tp_name);
             return NULL;
         }
+        val = PyStructSequence_GET_ITEM(obj, i);
         repr = PyObject_Repr(val);
-        if (repr == NULL) {
-            Py_DECREF(tup);
+        if (repr == NULL)
             return NULL;
-        }
         crepr = _PyUnicode_AsString(repr);
         if (crepr == NULL) {
-            Py_DECREF(tup);
             Py_DECREF(repr);
             return NULL;
         }
@@ -300,7 +205,6 @@
             break;
         }
     }
-    Py_DECREF(tup);
     if (removelast) {
         /* overwrite last ", " */
         pbuf-=2;
@@ -312,62 +216,6 @@
 }
 
 static PyObject *
-structseq_concat(PyStructSequence *obj, PyObject *b)
-{
-    PyObject *tup, *result;
-    tup = make_tuple(obj);
-    result = PySequence_Concat(tup, b);
-    Py_DECREF(tup);
-    return result;
-}
-
-static PyObject *
-structseq_repeat(PyStructSequence *obj, Py_ssize_t n)
-{
-    PyObject *tup, *result;
-    tup = make_tuple(obj);
-    result = PySequence_Repeat(tup, n);
-    Py_DECREF(tup);
-    return result;
-}
-
-static int
-structseq_contains(PyStructSequence *obj, PyObject *o)
-{
-    PyObject *tup;
-    int result;
-    tup = make_tuple(obj);
-    if (!tup)
-        return -1;
-    result = PySequence_Contains(tup, o);
-    Py_DECREF(tup);
-    return result;
-}
-
-static long
-structseq_hash(PyObject *obj)
-{
-    PyObject *tup;
-    long result;
-    tup = make_tuple((PyStructSequence*) obj);
-    if (!tup)
-        return -1;
-    result = PyObject_Hash(tup);
-    Py_DECREF(tup);
-    return result;
-}
-
-static PyObject *
-structseq_richcompare(PyObject *obj, PyObject *o2, int op)
-{
-    PyObject *tup, *result;
-    tup = make_tuple((PyStructSequence*) obj);
-    result = PyObject_RichCompare(tup, o2, op);
-    Py_DECREF(tup);
-    return result;
-}
-
-static PyObject *
 structseq_reduce(PyStructSequence* self)
 {
     PyObject* tup;
@@ -409,33 +257,16 @@
     return result;
 }
 
-static PySequenceMethods structseq_as_sequence = {
-    (lenfunc)structseq_length,
-    (binaryfunc)structseq_concat,           /* sq_concat */
-    (ssizeargfunc)structseq_repeat,         /* sq_repeat */
-    (ssizeargfunc)structseq_item,               /* sq_item */
-    0,                                          /* sq_slice */
-    0,                                          /* sq_ass_item */
-    0,                                          /* sq_ass_slice */
-    (objobjproc)structseq_contains,             /* sq_contains */
-};
-
-static PyMappingMethods structseq_as_mapping = {
-    (lenfunc)structseq_length,
-    (binaryfunc)structseq_subscript,
-};
-
 static PyMethodDef structseq_methods[] = {
-    {"__reduce__", (PyCFunction)structseq_reduce,
-     METH_NOARGS, NULL},
+    {"__reduce__", (PyCFunction)structseq_reduce, METH_NOARGS, NULL},
     {NULL, NULL}
 };
 
 static PyTypeObject _struct_sequence_template = {
     PyVarObject_HEAD_INIT(&PyType_Type, 0)
     NULL,                                       /* tp_name */
-    0,                                          /* tp_basicsize */
-    0,                                          /* tp_itemsize */
+    sizeof(PyStructSequence) - sizeof(PyObject *), /* tp_basicsize */
+    sizeof(PyObject *),                         /* tp_itemsize */
     (destructor)structseq_dealloc,              /* tp_dealloc */
     0,                                          /* tp_print */
     0,                                          /* tp_getattr */
@@ -443,19 +274,19 @@
     0,                                          /* tp_reserved */
     (reprfunc)structseq_repr,                   /* tp_repr */
     0,                                          /* tp_as_number */
-    &structseq_as_sequence,                     /* tp_as_sequence */
-    &structseq_as_mapping,                      /* tp_as_mapping */
-    structseq_hash,                             /* tp_hash */
+    0,                                          /* tp_as_sequence */
+    0,                                          /* tp_as_mapping */
+    0,                                          /* tp_hash */
     0,                                          /* tp_call */
     0,                                          /* tp_str */
     0,                                          /* tp_getattro */
     0,                                          /* tp_setattro */
     0,                                          /* tp_as_buffer */
-    Py_TPFLAGS_DEFAULT,                     /* tp_flags */
+    Py_TPFLAGS_DEFAULT,                         /* tp_flags */
     NULL,                                       /* tp_doc */
     0,                                          /* tp_traverse */
     0,                                          /* tp_clear */
-    structseq_richcompare,                      /* tp_richcompare */
+    0,                                          /* tp_richcompare */
     0,                                          /* tp_weaklistoffset */
     0,                                          /* tp_iter */
     0,                                          /* tp_iternext */
@@ -494,11 +325,9 @@
     n_members = i;
 
     memcpy(type, &_struct_sequence_template, sizeof(PyTypeObject));
+    type->tp_base = &PyTuple_Type;
     type->tp_name = desc->name;
     type->tp_doc = desc->doc;
-    type->tp_basicsize = sizeof(PyStructSequence)+
-        sizeof(PyObject*)*(n_members-1);
-    type->tp_itemsize = 0;
 
     members = PyMem_NEW(PyMemberDef, n_members-n_unnamed_members+1);
     if (members == NULL)

Modified: python/branches/py3k-cdecimal/Objects/typeobject.c
==============================================================================
--- python/branches/py3k-cdecimal/Objects/typeobject.c	(original)
+++ python/branches/py3k-cdecimal/Objects/typeobject.c	Sat Jul 17 12:36:09 2010
@@ -3476,11 +3476,8 @@
 static void
 inherit_special(PyTypeObject *type, PyTypeObject *base)
 {
-    Py_ssize_t oldsize, newsize;
 
     /* Copying basicsize is connected to the GC flags */
-    oldsize = base->tp_basicsize;
-    newsize = type->tp_basicsize ? type->tp_basicsize : oldsize;
     if (!(type->tp_flags & Py_TPFLAGS_HAVE_GC) &&
         (base->tp_flags & Py_TPFLAGS_HAVE_GC) &&
         (!type->tp_traverse && !type->tp_clear)) {
@@ -3507,7 +3504,8 @@
                 type->tp_new = base->tp_new;
         }
     }
-    type->tp_basicsize = newsize;
+    if (type->tp_basicsize == 0)
+        type->tp_basicsize = base->tp_basicsize;
 
     /* Copy other non-function slots */
 

Modified: python/branches/py3k-cdecimal/Objects/unicodeobject.c
==============================================================================
--- python/branches/py3k-cdecimal/Objects/unicodeobject.c	(original)
+++ python/branches/py3k-cdecimal/Objects/unicodeobject.c	Sat Jul 17 12:36:09 2010
@@ -1273,7 +1273,7 @@
            case of a TypeError. */
         if (PyErr_ExceptionMatches(PyExc_TypeError))
             PyErr_Format(PyExc_TypeError,
-                         "coercing to str: need string or buffer, "
+                         "coercing to str: need bytes, bytearray or char buffer, "
                          "%.80s found",
                          Py_TYPE(obj)->tp_name);
         goto onError;
@@ -1767,6 +1767,33 @@
     return 0;
 }
 
+/* create or adjust a UnicodeDecodeError */
+static void
+make_decode_exception(PyObject **exceptionObject,
+                      const char *encoding,
+                      const char *input, Py_ssize_t length,
+                      Py_ssize_t startpos, Py_ssize_t endpos,
+                      const char *reason)
+{
+    if (*exceptionObject == NULL) {
+        *exceptionObject = PyUnicodeDecodeError_Create(
+            encoding, input, length, startpos, endpos, reason);
+    }
+    else {
+        if (PyUnicodeDecodeError_SetStart(*exceptionObject, startpos))
+            goto onError;
+        if (PyUnicodeDecodeError_SetEnd(*exceptionObject, endpos))
+            goto onError;
+        if (PyUnicodeDecodeError_SetReason(*exceptionObject, reason))
+            goto onError;
+    }
+    return;
+
+onError:
+    Py_DECREF(*exceptionObject);
+    *exceptionObject = NULL;
+}
+
 /* error handling callback helper:
    build arguments, call the callback and check the arguments,
    if no exception occurred, copy the replacement to the output
@@ -1800,20 +1827,13 @@
             goto onError;
     }
 
-    if (*exceptionObject == NULL) {
-        *exceptionObject = PyUnicodeDecodeError_Create(
-            encoding, *input, *inend-*input, *startinpos, *endinpos, reason);
-        if (*exceptionObject == NULL)
-            goto onError;
-    }
-    else {
-        if (PyUnicodeDecodeError_SetStart(*exceptionObject, *startinpos))
-            goto onError;
-        if (PyUnicodeDecodeError_SetEnd(*exceptionObject, *endinpos))
-            goto onError;
-        if (PyUnicodeDecodeError_SetReason(*exceptionObject, reason))
-            goto onError;
-    }
+    make_decode_exception(exceptionObject,
+        encoding,
+        *input, *inend - *input,
+        *startinpos, *endinpos,
+        reason);
+    if (*exceptionObject == NULL)
+        goto onError;
 
     restuple = PyObject_CallFunctionObjArgs(*errorHandler, *exceptionObject, NULL);
     if (restuple == NULL)
@@ -2265,24 +2285,24 @@
 
 static
 char utf8_code_length[256] = {
-    /* Map UTF-8 encoded prefix byte to sequence length.  zero means
-       illegal prefix.  see RFC 2279 for details */
-    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
-    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+    /* Map UTF-8 encoded prefix byte to sequence length.  Zero means
+       illegal prefix.  See RFC 3629 for details */
+    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 00-0F */
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
     1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
-    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 70-7F */
+    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 80-8F */
     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
-    2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
-    3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
-    4, 4, 4, 4, 4, 4, 4, 4, 5, 5, 5, 5, 6, 6, 0, 0
+    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* B0-BF */
+    0, 0, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, /* C0-C1 + C2-CF */
+    2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, /* D0-DF */
+    3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, /* E0-EF */
+    4, 4, 4, 4, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0  /* F0-F4 + F5-FF */
 };
 
 PyObject *PyUnicode_DecodeUTF8(const char *s,
@@ -2312,6 +2332,7 @@
 {
     const char *starts = s;
     int n;
+    int k;
     Py_ssize_t startinpos;
     Py_ssize_t endinpos;
     Py_ssize_t outpos;
@@ -2395,7 +2416,9 @@
             else {
                 errmsg = "unexpected end of data";
                 startinpos = s-starts;
-                endinpos = size;
+                endinpos = startinpos+1;
+                for (k=1; (k < size-startinpos) && ((s[k]&0xC0) == 0x80); k++)
+                    endinpos++;
                 goto utf8Error;
             }
         }
@@ -2403,7 +2426,7 @@
         switch (n) {
 
         case 0:
-            errmsg = "unexpected code byte";
+            errmsg = "invalid start byte";
             startinpos = s-starts;
             endinpos = startinpos+1;
             goto utf8Error;
@@ -2416,63 +2439,67 @@
 
         case 2:
             if ((s[1] & 0xc0) != 0x80) {
-                errmsg = "invalid data";
+                errmsg = "invalid continuation byte";
                 startinpos = s-starts;
-                endinpos = startinpos+2;
+                endinpos = startinpos + 1;
                 goto utf8Error;
             }
             ch = ((s[0] & 0x1f) << 6) + (s[1] & 0x3f);
-            if (ch < 0x80) {
-                startinpos = s-starts;
-                endinpos = startinpos+2;
-                errmsg = "illegal encoding";
-                goto utf8Error;
-            }
-            else
-                *p++ = (Py_UNICODE)ch;
+            assert ((ch > 0x007F) && (ch <= 0x07FF));
+            *p++ = (Py_UNICODE)ch;
             break;
 
         case 3:
+            /* Decoding UTF-8 sequences in range \xed\xa0\x80-\xed\xbf\xbf
+               will result in surrogates in range d800-dfff. Surrogates are
+               not valid UTF-8 so they are rejected.
+               See http://www.unicode.org/versions/Unicode5.2.0/ch03.pdf
+               (table 3-7) and http://www.rfc-editor.org/rfc/rfc3629.txt */
             if ((s[1] & 0xc0) != 0x80 ||
-                (s[2] & 0xc0) != 0x80) {
-                errmsg = "invalid data";
+                (s[2] & 0xc0) != 0x80 ||
+                ((unsigned char)s[0] == 0xE0 &&
+                 (unsigned char)s[1] < 0xA0) ||
+                ((unsigned char)s[0] == 0xED &&
+                 (unsigned char)s[1] > 0x9F)) {
+                errmsg = "invalid continuation byte";
                 startinpos = s-starts;
-                endinpos = startinpos+3;
+                endinpos = startinpos + 1;
+
+                /* if s[1] first two bits are 1 and 0, then the invalid
+                   continuation byte is s[2], so increment endinpos by 1,
+                   if not, s[1] is invalid and endinpos doesn't need to
+                   be incremented. */
+                if ((s[1] & 0xC0) == 0x80)
+                    endinpos++;
                 goto utf8Error;
             }
             ch = ((s[0] & 0x0f) << 12) + ((s[1] & 0x3f) << 6) + (s[2] & 0x3f);
-            if (ch < 0x0800 || (ch >= 0xd800 && ch <= 0xDFFF)) {
-                errmsg = "illegal encoding";
-                startinpos = s-starts;
-                endinpos = startinpos+3;
-                goto utf8Error;
-            }
-            else
-                *p++ = (Py_UNICODE)ch;
+            assert ((ch > 0x07FF) && (ch <= 0xFFFF));
+            *p++ = (Py_UNICODE)ch;
             break;
 
         case 4:
             if ((s[1] & 0xc0) != 0x80 ||
                 (s[2] & 0xc0) != 0x80 ||
-                (s[3] & 0xc0) != 0x80) {
-                errmsg = "invalid data";
+                (s[3] & 0xc0) != 0x80 ||
+                ((unsigned char)s[0] == 0xF0 &&
+                 (unsigned char)s[1] < 0x90) ||
+                ((unsigned char)s[0] == 0xF4 &&
+                 (unsigned char)s[1] > 0x8F)) {
+                errmsg = "invalid continuation byte";
                 startinpos = s-starts;
-                endinpos = startinpos+4;
+                endinpos = startinpos + 1;
+                if ((s[1] & 0xC0) == 0x80) {
+                    endinpos++;
+                    if ((s[2] & 0xC0) == 0x80)
+                        endinpos++;
+                }
                 goto utf8Error;
             }
             ch = ((s[0] & 0x7) << 18) + ((s[1] & 0x3f) << 12) +
-                ((s[2] & 0x3f) << 6) + (s[3] & 0x3f);
-            /* validate and convert to UTF-16 */
-            if ((ch < 0x10000)        /* minimum value allowed for 4
-                                         byte encoding */
-                || (ch > 0x10ffff))   /* maximum value allowed for
-                                         UTF-16 */
-            {
-                errmsg = "illegal encoding";
-                startinpos = s-starts;
-                endinpos = startinpos+4;
-                goto utf8Error;
-            }
+                 ((s[2] & 0x3f) << 6) + (s[3] & 0x3f);
+            assert ((ch > 0xFFFF) && (ch <= 0x10ffff));
+
 #ifdef Py_UNICODE_WIDE
             *p++ = (Py_UNICODE)ch;
 #else
@@ -2488,13 +2515,6 @@
             *p++ = (Py_UNICODE)(0xDC00 + (ch & 0x03FF));
 #endif
             break;
-
-        default:
-            /* Other sizes are only needed for UCS-4 */
-            errmsg = "unsupported Unicode code range";
-            startinpos = s-starts;
-            endinpos = startinpos+n;
-            goto utf8Error;
         }
         s += n;
         continue;
@@ -4552,32 +4572,46 @@
 static int decode_mbcs(PyUnicodeObject **v,
                        const char *s, /* MBCS string */
                        int size, /* sizeof MBCS string */
-                       int final)
+                       int final,
+                       const char *errors)
 {
     Py_UNICODE *p;
-    Py_ssize_t n = 0;
-    int usize = 0;
+    Py_ssize_t n;
+    DWORD usize;
+    DWORD flags;
 
     assert(size >= 0);
 
+    /* check and handle 'errors' arg */
+    if (errors==NULL || strcmp(errors, "strict")==0)
+        flags = MB_ERR_INVALID_CHARS;
+    else if (strcmp(errors, "ignore")==0)
+        flags = 0;
+    else {
+        PyErr_Format(PyExc_ValueError,
+                     "mbcs encoding does not support errors='%s'",
+                     errors);
+        return -1;
+    }
+
     /* Skip trailing lead-byte unless 'final' is set */
     if (!final && size >= 1 && is_dbcs_lead_byte(s, size - 1))
         --size;
 
     /* First get the size of the result */
     if (size > 0) {
-        usize = MultiByteToWideChar(CP_ACP, 0, s, size, NULL, 0);
-        if (usize == 0) {
-            PyErr_SetFromWindowsErrWithFilename(0, NULL);
-            return -1;
-        }
-    }
+        usize = MultiByteToWideChar(CP_ACP, flags, s, size, NULL, 0);
+        if (usize==0)
+            goto mbcs_decode_error;
+    } else
+        usize = 0;
 
     if (*v == NULL) {
         /* Create unicode object */
         *v = _PyUnicode_New(usize);
         if (*v == NULL)
             return -1;
+        n = 0;
     }
     else {
         /* Extend unicode object */
@@ -4587,15 +4621,35 @@
     }
 
     /* Do the conversion */
-    if (size > 0) {
+    if (usize > 0) {
         p = PyUnicode_AS_UNICODE(*v) + n;
-        if (0 == MultiByteToWideChar(CP_ACP, 0, s, size, p, usize)) {
-            PyErr_SetFromWindowsErrWithFilename(0, NULL);
-            return -1;
+        if (0 == MultiByteToWideChar(CP_ACP, flags, s, size, p, usize)) {
+            goto mbcs_decode_error;
         }
     }
-
     return size;
+
+mbcs_decode_error:
+    /* If the last error was ERROR_NO_UNICODE_TRANSLATION, then
+       we raise a UnicodeDecodeError - else it is a 'generic'
+       windows error
+     */
+    if (GetLastError()==ERROR_NO_UNICODE_TRANSLATION) {
+        /* Ideally, we should get reason from FormatMessage - this
+           is the Windows 2000 English version of the message
+        */
+        PyObject *exc = NULL;
+        const char *reason = "No mapping for the Unicode character exists "
+                             "in the target multi-byte code page.";
+        make_decode_exception(&exc, "mbcs", s, size, 0, 0, reason);
+        if (exc != NULL) {
+            PyCodec_StrictErrors(exc);
+            Py_DECREF(exc);
+        }
+    } else {
+        PyErr_SetFromWindowsErrWithFilename(0, NULL);
+    }
+    return -1;
 }
 
 PyObject *PyUnicode_DecodeMBCSStateful(const char *s,
@@ -4612,10 +4666,10 @@
 #ifdef NEED_RETRY
   retry:
     if (size > INT_MAX)
-        done = decode_mbcs(&v, s, INT_MAX, 0);
+        done = decode_mbcs(&v, s, INT_MAX, 0, errors);
     else
 #endif
-        done = decode_mbcs(&v, s, (int)size, !consumed);
+        done = decode_mbcs(&v, s, (int)size, !consumed, errors);
 
     if (done < 0) {
         Py_XDECREF(v);
@@ -4649,20 +4703,45 @@
  */
 static int encode_mbcs(PyObject **repr,
                        const Py_UNICODE *p, /* unicode */
-                       int size) /* size of unicode */
+                       int size, /* size of unicode */
+                       const char* errors)
 {
-    int mbcssize = 0;
-    Py_ssize_t n = 0;
+    BOOL usedDefaultChar = FALSE;
+    BOOL *pusedDefaultChar;
+    int mbcssize;
+    Py_ssize_t n;
+    PyObject *exc = NULL;
+    DWORD flags;
 
     assert(size >= 0);
 
+    /* check and handle 'errors' arg */
+    if (errors==NULL || strcmp(errors, "strict")==0) {
+        flags = WC_NO_BEST_FIT_CHARS;
+        pusedDefaultChar = &usedDefaultChar;
+    } else if (strcmp(errors, "replace")==0) {
+        flags = 0;
+        pusedDefaultChar = NULL;
+    } else {
+         PyErr_Format(PyExc_ValueError,
+                      "mbcs encoding does not support errors='%s'",
+                      errors);
+         return -1;
+    }
+
     /* First get the size of the result */
     if (size > 0) {
-        mbcssize = WideCharToMultiByte(CP_ACP, 0, p, size, NULL, 0, NULL, NULL);
+        mbcssize = WideCharToMultiByte(CP_ACP, flags, p, size, NULL, 0,
+                                       NULL, pusedDefaultChar);
         if (mbcssize == 0) {
             PyErr_SetFromWindowsErrWithFilename(0, NULL);
             return -1;
         }
+        /* If we used a default char, then we failed! */
+        if (pusedDefaultChar && *pusedDefaultChar)
+            goto mbcs_encode_error;
+    } else {
+        mbcssize = 0;
     }
 
     if (*repr == NULL) {
@@ -4670,6 +4749,7 @@
         *repr = PyBytes_FromStringAndSize(NULL, mbcssize);
         if (*repr == NULL)
             return -1;
+        n = 0;
     }
     else {
         /* Extend string object */
@@ -4681,13 +4761,20 @@
     /* Do the conversion */
     if (size > 0) {
         char *s = PyBytes_AS_STRING(*repr) + n;
-        if (0 == WideCharToMultiByte(CP_ACP, 0, p, size, s, mbcssize, NULL, NULL)) {
+        if (0 == WideCharToMultiByte(CP_ACP, flags, p, size, s, mbcssize,
+                                     NULL, pusedDefaultChar)) {
             PyErr_SetFromWindowsErrWithFilename(0, NULL);
             return -1;
         }
+        if (pusedDefaultChar && *pusedDefaultChar)
+            goto mbcs_encode_error;
     }
-
     return 0;
+
+mbcs_encode_error:
+    raise_encode_exception(&exc, "mbcs", p, size, 0, 0, "invalid character");
+    Py_XDECREF(exc);
+    return -1;
 }
 
 PyObject *PyUnicode_EncodeMBCS(const Py_UNICODE *p,
@@ -4700,10 +4787,10 @@
 #ifdef NEED_RETRY
   retry:
     if (size > INT_MAX)
-        ret = encode_mbcs(&repr, p, INT_MAX);
+        ret = encode_mbcs(&repr, p, INT_MAX, errors);
     else
 #endif
-        ret = encode_mbcs(&repr, p, (int)size);
+        ret = encode_mbcs(&repr, p, (int)size, errors);
 
     if (ret < 0) {
         Py_XDECREF(repr);
@@ -6618,7 +6705,7 @@
              "S.capitalize() -> str\n\
 \n\
 Return a capitalized version of S, i.e. make the first character\n\
-have upper case.");
+have upper case and the rest lower case.");
 
 static PyObject*
 unicode_capitalize(PyUnicodeObject *self)
@@ -7890,7 +7977,7 @@
 }
 
 PyDoc_STRVAR(replace__doc__,
-             "S.replace (old, new[, count]) -> str\n\
+             "S.replace(old, new[, count]) -> str\n\
 \n\
 Return a copy of S with all occurrences of substring\n\
 old replaced by new.  If the optional argument count is\n\

Modified: python/branches/py3k-cdecimal/PC/VS8.0/pythoncore.vcproj
==============================================================================
--- python/branches/py3k-cdecimal/PC/VS8.0/pythoncore.vcproj	(original)
+++ python/branches/py3k-cdecimal/PC/VS8.0/pythoncore.vcproj	Sat Jul 17 12:36:09 2010
@@ -759,10 +759,6 @@
 				>
 			</File>
 			<File
-				RelativePath="..\..\Include\intobject.h"
-				>
-			</File>
-			<File
 				RelativePath="..\..\Include\intrcheck.h"
 				>
 			</File>

Modified: python/branches/py3k-cdecimal/PCbuild/_ctypes.vcproj
==============================================================================
--- python/branches/py3k-cdecimal/PCbuild/_ctypes.vcproj	(original)
+++ python/branches/py3k-cdecimal/PCbuild/_ctypes.vcproj	Sat Jul 17 12:36:09 2010
@@ -1,7 +1,7 @@
 <?xml version="1.0" encoding="Windows-1252"?>
 <VisualStudioProject
 	ProjectType="Visual C++"
-	Version="9,00"
+	Version="9.00"
 	Name="_ctypes"
 	ProjectGUID="{0E9791DB-593A-465F-98BC-681011311618}"
 	RootNamespace="_ctypes"
@@ -42,8 +42,7 @@
 			/>
 			<Tool
 				Name="VCCLCompilerTool"
-				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc;..\Modules\_ctypes\libffi\include;..\Modules\_ctypes\libffi\src\x86"
-				PreprocessorDefinitions="X86_WIN32"
+				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc"
 			/>
 			<Tool
 				Name="VCManagedResourceCompilerTool"
@@ -104,8 +103,7 @@
 			/>
 			<Tool
 				Name="VCCLCompilerTool"
-				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc;..\Modules\_ctypes\libffi\include;..\Modules\_ctypes\libffi\src\x86"
-				PreprocessorDefinitions="X86_WIN64"
+				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc"
 			/>
 			<Tool
 				Name="VCManagedResourceCompilerTool"
@@ -166,8 +164,7 @@
 			/>
 			<Tool
 				Name="VCCLCompilerTool"
-				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc;..\Modules\_ctypes\libffi\include;..\Modules\_ctypes\libffi\src\x86"
-				PreprocessorDefinitions="X86_WIN32"
+				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc"
 			/>
 			<Tool
 				Name="VCManagedResourceCompilerTool"
@@ -231,8 +228,7 @@
 			/>
 			<Tool
 				Name="VCCLCompilerTool"
-				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc;..\Modules\_ctypes\libffi\include;..\Modules\_ctypes\libffi\src\x86"
-				PreprocessorDefinitions="X86_WIN64"
+				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc"
 			/>
 			<Tool
 				Name="VCManagedResourceCompilerTool"
@@ -295,8 +291,7 @@
 			/>
 			<Tool
 				Name="VCCLCompilerTool"
-				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc;..\Modules\_ctypes\libffi\include;..\Modules\_ctypes\libffi\src\x86"
-				PreprocessorDefinitions="X86_WIN32"
+				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc"
 			/>
 			<Tool
 				Name="VCManagedResourceCompilerTool"
@@ -360,8 +355,7 @@
 			/>
 			<Tool
 				Name="VCCLCompilerTool"
-				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc;..\Modules\_ctypes\libffi\include;..\Modules\_ctypes\libffi\src\x86"
-				PreprocessorDefinitions="X86_WIN64"
+				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc"
 			/>
 			<Tool
 				Name="VCManagedResourceCompilerTool"
@@ -425,8 +419,7 @@
 			/>
 			<Tool
 				Name="VCCLCompilerTool"
-				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc;..\Modules\_ctypes\libffi\include;..\Modules\_ctypes\libffi\src\x86"
-				PreprocessorDefinitions="X86_WIN32"
+				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc"
 			/>
 			<Tool
 				Name="VCManagedResourceCompilerTool"
@@ -490,8 +483,7 @@
 			/>
 			<Tool
 				Name="VCCLCompilerTool"
-				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc;..\Modules\_ctypes\libffi\include;..\Modules\_ctypes\libffi\src\x86"
-				PreprocessorDefinitions="X86_WIN64"
+				AdditionalIncludeDirectories="..\Modules\_ctypes\libffi_msvc"
 			/>
 			<Tool
 				Name="VCManagedResourceCompilerTool"
@@ -551,7 +543,7 @@
 				>
 			</File>
 			<File
-				RelativePath="..\Modules\_ctypes\libffi\include\ffi_common.h"
+				RelativePath="..\Modules\_ctypes\libffi_msvc\ffi_common.h"
 				>
 			</File>
 			<File
@@ -559,7 +551,7 @@
 				>
 			</File>
 			<File
-				RelativePath="..\Modules\_ctypes\libffi\src\x86\ffitarget.h"
+				RelativePath="..\Modules\_ctypes\libffi_msvc\ffitarget.h"
 				>
 			</File>
 		</Filter>
@@ -583,23 +575,15 @@
 				>
 			</File>
 			<File
-				RelativePath="..\Modules\_ctypes\libffi\src\x86\ffi.c"
+				RelativePath="..\Modules\_ctypes\libffi_msvc\ffi.c"
 				>
-				<FileConfiguration
-					Name="Debug|x64"
-					>
-					<Tool
-						Name="VCCLCompilerTool"
-						PreprocessorDefinitions="X86_WIN64"
-					/>
-				</FileConfiguration>
 			</File>
 			<File
 				RelativePath="..\Modules\_ctypes\malloc_closure.c"
 				>
 			</File>
 			<File
-				RelativePath="..\Modules\_ctypes\libffi\src\prep_cif.c"
+				RelativePath="..\Modules\_ctypes\libffi_msvc\prep_cif.c"
 				>
 			</File>
 			<File
@@ -607,32 +591,14 @@
 				>
 			</File>
 			<File
-				RelativePath="..\Modules\_ctypes\libffi\src\x86\win32.S"
+				RelativePath="..\Modules\_ctypes\libffi_msvc\win32.c"
 				>
 				<FileConfiguration
-					Name="Debug|Win32"
-					>
-					<Tool
-						Name="VCCustomBuildTool"
-						CommandLine="cl /nologo /EP /I ..\Modules\_ctypes\libffi_msvc /I ..\Modules\_ctypes\libffi\src\x86 $(InputPath) &gt; $(IntDir)$(InputName).asm&#x0D;&#x0A;ml /nologo /Zi /c /Fo $(IntDir)$(InputName).obj $(IntDir)$(InputName).asm&#x0D;&#x0A;"
-						Outputs="$(IntDir)$(InputName).obj"
-					/>
-				</FileConfiguration>
-				<FileConfiguration
 					Name="Debug|x64"
 					ExcludedFromBuild="true"
 					>
 					<Tool
-						Name="VCCustomBuildTool"
-					/>
-				</FileConfiguration>
-				<FileConfiguration
-					Name="Release|Win32"
-					>
-					<Tool
-						Name="VCCustomBuildTool"
-						CommandLine="cl /nologo /EP /I ..\Modules\_ctypes\libffi_msvc /I ..\Modules\_ctypes\libffi\src\x86 $(InputPath) &gt; $(IntDir)$(InputName).asm&#x0D;&#x0A;ml /nologo /c /Fo $(IntDir)$(InputName).obj $(IntDir)$(InputName).asm&#x0D;&#x0A;"
-						Outputs="$(IntDir)$(InputName).obj"
+						Name="VCCLCompilerTool"
 					/>
 				</FileConfiguration>
 				<FileConfiguration
@@ -640,16 +606,7 @@
 					ExcludedFromBuild="true"
 					>
 					<Tool
-						Name="VCCustomBuildTool"
-					/>
-				</FileConfiguration>
-				<FileConfiguration
-					Name="PGInstrument|Win32"
-					>
-					<Tool
-						Name="VCCustomBuildTool"
-						CommandLine="cl /nologo /EP /I ..\Modules\_ctypes\libffi_msvc /I ..\Modules\_ctypes\libffi\src\x86 $(InputPath) &gt; $(IntDir)$(InputName).asm&#x0D;&#x0A;ml /nologo /Zi /c /Fo $(IntDir)$(InputName).obj $(IntDir)$(InputName).asm&#x0D;&#x0A;"
-						Outputs="$(IntDir)$(InputName).obj"
+						Name="VCCLCompilerTool"
 					/>
 				</FileConfiguration>
 				<FileConfiguration
@@ -657,16 +614,7 @@
 					ExcludedFromBuild="true"
 					>
 					<Tool
-						Name="VCCustomBuildTool"
-					/>
-				</FileConfiguration>
-				<FileConfiguration
-					Name="PGUpdate|Win32"
-					>
-					<Tool
-						Name="VCCustomBuildTool"
-						CommandLine="cl /nologo /EP /I ..\Modules\_ctypes\libffi_msvc /I ..\Modules\_ctypes\libffi\src\x86 $(InputPath) &gt; $(IntDir)$(InputName).asm&#x0D;&#x0A;ml /nologo /Zi /c /Fo $(IntDir)$(InputName).obj $(IntDir)$(InputName).asm&#x0D;&#x0A;"
-						Outputs="$(IntDir)$(InputName).obj"
+						Name="VCCLCompilerTool"
 					/>
 				</FileConfiguration>
 				<FileConfiguration
@@ -674,12 +622,12 @@
 					ExcludedFromBuild="true"
 					>
 					<Tool
-						Name="VCCustomBuildTool"
+						Name="VCCLCompilerTool"
 					/>
 				</FileConfiguration>
 			</File>
 			<File
-				RelativePath="..\Modules\_ctypes\libffi\src\x86\win64.S"
+				RelativePath="..\Modules\_ctypes\libffi_msvc\win64.asm"
 				>
 				<FileConfiguration
 					Name="Debug|Win32"
@@ -694,8 +642,8 @@
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="cl /nologo /EP /I ..\Modules\_ctypes\libffi_msvc /I ..\Modules\_ctypes\libffi\src\x86 $(InputPath) &gt; $(IntDir)$(InputName).asm&#x0D;&#x0A;ml64 /Zi /nologo /c /Fo $(IntDir)$(InputName).obj $(IntDir)$(InputName).asm&#x0D;&#x0A;"
-						Outputs="$(IntDir)$(InputName).obj"
+						CommandLine="ml64 /nologo /c /Zi /Fo &quot;$(IntDir)\win64.obj&quot; &quot;$(InputPath)&quot;&#x0D;&#x0A;"
+						Outputs="$(IntDir)\win64.obj"
 					/>
 				</FileConfiguration>
 				<FileConfiguration
@@ -711,8 +659,8 @@
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="cl /nologo /EP /I ..\Modules\_ctypes\libffi_msvc /I ..\Modules\_ctypes\libffi\src\x86 $(InputPath) &gt; $(IntDir)$(InputName).asm&#x0D;&#x0A;ml64 /nologo /c /Fo $(IntDir)$(InputName).obj $(IntDir)$(InputName).asm&#x0D;&#x0A;"
-						Outputs="$(IntDir)$(InputName).obj"
+						CommandLine="ml64 /nologo /c /Fo &quot;$(IntDir)\win64.obj&quot; &quot;$(InputPath)&quot;&#x0D;&#x0A;"
+						Outputs="$(IntDir)\win64.obj"
 					/>
 				</FileConfiguration>
 				<FileConfiguration
@@ -728,8 +676,8 @@
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="cl /nologo /EP /I ..\Modules\_ctypes\libffi_msvc /I ..\Modules\_ctypes\libffi\src\x86 $(InputPath) &gt; $(IntDir)$(InputName).asm&#x0D;&#x0A;ml64 /nologo /c /Fo $(IntDir)$(InputName).obj $(IntDir)$(InputName).asm&#x0D;&#x0A;"
-						Outputs="$(IntDir)$(InputName).obj"
+						CommandLine="ml64 /nologo /c /Fo &quot;$(IntDir)\win64.obj&quot; &quot;$(InputPath)&quot;&#x0D;&#x0A;"
+						Outputs="$(IntDir)\win64.obj"
 					/>
 				</FileConfiguration>
 				<FileConfiguration
@@ -745,8 +693,8 @@
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="cl /nologo /EP /I ..\Modules\_ctypes\libffi_msvc /I ..\Modules\_ctypes\libffi\src\x86 $(InputPath) &gt; $(IntDir)$(InputName).asm&#x0D;&#x0A;ml64 /nologo /c /Fo $(IntDir)$(InputName).obj $(IntDir)$(InputName).asm&#x0D;&#x0A;"
-						Outputs="$(IntDir)$(InputName).obj"
+						CommandLine="ml64 /nologo /c /Fo &quot;$(IntDir)\win64.obj&quot; &quot;$(InputPath)&quot;&#x0D;&#x0A;"
+						Outputs="$(IntDir)\win64.obj"
 					/>
 				</FileConfiguration>
 			</File>

Modified: python/branches/py3k-cdecimal/PCbuild/pythoncore.vcproj
==============================================================================
--- python/branches/py3k-cdecimal/PCbuild/pythoncore.vcproj	(original)
+++ python/branches/py3k-cdecimal/PCbuild/pythoncore.vcproj	Sat Jul 17 12:36:09 2010
@@ -755,10 +755,6 @@
 				>
 			</File>
 			<File
-				RelativePath="..\Include\intobject.h"
-				>
-			</File>
-			<File
 				RelativePath="..\Include\intrcheck.h"
 				>
 			</File>
@@ -1040,6 +1036,14 @@
 				>
 			</File>
 			<File
+				RelativePath="..\Modules\_time.c"
+				>
+			</File>
+			<File
+				RelativePath="..\Modules\_time.h"
+				>
+			</File>
+			<File
 				RelativePath="..\Modules\_weakref.c"
 				>
 			</File>

Modified: python/branches/py3k-cdecimal/Parser/Python.asdl
==============================================================================
--- python/branches/py3k-cdecimal/Parser/Python.asdl	(original)
+++ python/branches/py3k-cdecimal/Parser/Python.asdl	Sat Jul 17 12:36:09 2010
@@ -17,7 +17,7 @@
 			 expr? starargs,
 			 expr? kwargs,
 			 stmt* body,
-			 expr *decorator_list)
+			 expr* decorator_list)
 	      | Return(expr? value)
 
 	      | Delete(expr* targets)

Modified: python/branches/py3k-cdecimal/Parser/asdl_c.py
==============================================================================
--- python/branches/py3k-cdecimal/Parser/asdl_c.py	(original)
+++ python/branches/py3k-cdecimal/Parser/asdl_c.py	Sat Jul 17 12:36:09 2010
@@ -732,8 +732,9 @@
 {
     int i, result;
     PyObject *s, *l = PyTuple_New(num_fields);
-    if (!l) return 0;
-    for(i = 0; i < num_fields; i++) {
+    if (!l)
+        return 0;
+    for (i = 0; i < num_fields; i++) {
         s = PyUnicode_FromString(attrs[i]);
         if (!s) {
             Py_DECREF(l);

Modified: python/branches/py3k-cdecimal/Python/Python-ast.c
==============================================================================
--- python/branches/py3k-cdecimal/Python/Python-ast.c	(original)
+++ python/branches/py3k-cdecimal/Python/Python-ast.c	Sat Jul 17 12:36:09 2010
@@ -2,7 +2,7 @@
 
 
 /*
-   __version__ 73626.
+   __version__ 82163.
 
    This module must be committed separately after each AST grammar change;
    The __version__ number is set to the revision number of the commit
@@ -537,8 +537,9 @@
 {
     int i, result;
     PyObject *s, *l = PyTuple_New(num_fields);
-    if (!l) return 0;
-    for(i = 0; i < num_fields; i++) {
+    if (!l)
+        return 0;
+    for (i = 0; i < num_fields; i++) {
         s = PyUnicode_FromString(attrs[i]);
         if (!s) {
             Py_DECREF(l);
@@ -6773,7 +6774,7 @@
             NULL;
         if (PyModule_AddIntConstant(m, "PyCF_ONLY_AST", PyCF_ONLY_AST) < 0)
                 return NULL;
-        if (PyModule_AddStringConstant(m, "__version__", "73626") < 0)
+        if (PyModule_AddStringConstant(m, "__version__", "82163") < 0)
                 return NULL;
         if (PyDict_SetItemString(d, "mod", (PyObject*)mod_type) < 0) return
             NULL;

Modified: python/branches/py3k-cdecimal/Python/_warnings.c
==============================================================================
--- python/branches/py3k-cdecimal/Python/_warnings.c	(original)
+++ python/branches/py3k-cdecimal/Python/_warnings.c	Sat Jul 17 12:36:09 2010
@@ -251,7 +251,7 @@
 
     name = PyObject_GetAttrString(category, "__name__");
     if (name == NULL)  /* XXX Can an object lack a '__name__' attribute? */
-            return;
+        return;
 
     f_stderr = PySys_GetObject("stderr");
     if (f_stderr == NULL) {
@@ -282,8 +282,7 @@
         PyFile_WriteString("\n", f_stderr);
     }
     else
-        if (_Py_DisplaySourceLine(f_stderr, _PyUnicode_AsString(filename),
-                              lineno, 2) < 0)
+        if (_Py_DisplaySourceLine(f_stderr, filename, lineno, 2) < 0)
                 return;
     PyErr_Clear();
 }
@@ -847,28 +846,35 @@
 static PyObject *
 init_filters(void)
 {
-    PyObject *filters = PyList_New(3);
+    /* Don't silence DeprecationWarning if -3 was used. */
+    PyObject *filters = PyList_New(4);
+    unsigned int pos = 0;  /* Post-incremented in each use. */
+    unsigned int x;
     const char *bytes_action;
+
     if (filters == NULL)
         return NULL;
 
-    PyList_SET_ITEM(filters, 0,
+    PyList_SET_ITEM(filters, pos++,
+                    create_filter(PyExc_DeprecationWarning, "ignore"));
+    PyList_SET_ITEM(filters, pos++,
                     create_filter(PyExc_PendingDeprecationWarning, "ignore"));
-    PyList_SET_ITEM(filters, 1, create_filter(PyExc_ImportWarning, "ignore"));
+    PyList_SET_ITEM(filters, pos++,
+                    create_filter(PyExc_ImportWarning, "ignore"));
     if (Py_BytesWarningFlag > 1)
         bytes_action = "error";
     else if (Py_BytesWarningFlag)
         bytes_action = "default";
     else
         bytes_action = "ignore";
-    PyList_SET_ITEM(filters, 2, create_filter(PyExc_BytesWarning,
+    PyList_SET_ITEM(filters, pos++, create_filter(PyExc_BytesWarning,
                     bytes_action));
 
-    if (PyList_GET_ITEM(filters, 0) == NULL ||
-        PyList_GET_ITEM(filters, 1) == NULL ||
-        PyList_GET_ITEM(filters, 2) == NULL) {
-        Py_DECREF(filters);
-        return NULL;
+    for (x = 0; x < pos; x += 1) {
+        if (PyList_GET_ITEM(filters, x) == NULL) {
+            Py_DECREF(filters);
+            return NULL;
+        }
     }
 
     return filters;

Modified: python/branches/py3k-cdecimal/Python/ast.c
==============================================================================
--- python/branches/py3k-cdecimal/Python/ast.c	(original)
+++ python/branches/py3k-cdecimal/Python/ast.c	Sat Jul 17 12:36:09 2010
@@ -1678,34 +1678,8 @@
 static expr_ty
 ast_for_factor(struct compiling *c, const node *n)
 {
-    node *pfactor, *ppower, *patom, *pnum;
     expr_ty expression;
 
-    /* If the unary - operator is applied to a constant, don't generate
-       a UNARY_NEGATIVE opcode.  Just store the approriate value as a
-       constant.  The peephole optimizer already does something like
-       this but it doesn't handle the case where the constant is
-       (sys.maxint - 1).  In that case, we want a PyIntObject, not a
-       PyLongObject.
-    */
-    if (TYPE(CHILD(n, 0)) == MINUS
-        && NCH(n) == 2
-        && TYPE((pfactor = CHILD(n, 1))) == factor
-        && NCH(pfactor) == 1
-        && TYPE((ppower = CHILD(pfactor, 0))) == power
-        && NCH(ppower) == 1
-        && TYPE((patom = CHILD(ppower, 0))) == atom
-        && TYPE((pnum = CHILD(patom, 0))) == NUMBER) {
-        char *s = PyObject_MALLOC(strlen(STR(pnum)) + 2);
-        if (s == NULL)
-            return NULL;
-        s[0] = '-';
-        strcpy(s + 1, STR(pnum));
-        PyObject_FREE(STR(pnum));
-        STR(pnum) = s;
-        return ast_for_atom(c, patom);
-    }
-
     expression = ast_for_expr(c, CHILD(n, 1));
     if (!expression)
         return NULL;

Modified: python/branches/py3k-cdecimal/Python/ceval.c
==============================================================================
--- python/branches/py3k-cdecimal/Python/ceval.c	(original)
+++ python/branches/py3k-cdecimal/Python/ceval.c	Sat Jul 17 12:36:09 2010
@@ -87,7 +87,7 @@
     inst = inst1 - inst0 - intr;
     loop = loop1 - loop0 - intr;
     fprintf(stderr, "opcode=%03d t=%d inst=%06lld loop=%06lld\n",
-        opcode, ticked, inst, loop);
+            opcode, ticked, inst, loop);
 }
 
 #endif
@@ -126,10 +126,10 @@
 static int call_trace(Py_tracefunc, PyObject *, PyFrameObject *,
                       int, PyObject *);
 static int call_trace_protected(Py_tracefunc, PyObject *,
-                                 PyFrameObject *, int, PyObject *);
+                                PyFrameObject *, int, PyObject *);
 static void call_exc_trace(Py_tracefunc, PyObject *, PyFrameObject *);
 static int maybe_call_line_trace(Py_tracefunc, PyObject *,
-                                  PyFrameObject *, int *, int *, int *);
+                                 PyFrameObject *, int *, int *, int *);
 
 static PyObject * cmp_outcome(int, PyObject *, PyObject *);
 static PyObject * import_from(PyObject *, PyObject *);
@@ -718,14 +718,14 @@
 
 /* Status code for main loop (reason for stack unwind) */
 enum why_code {
-                WHY_NOT =       0x0001, /* No error */
-        WHY_EXCEPTION = 0x0002,         /* Exception occurred */
-        WHY_RERAISE =           0x0004, /* Exception re-raised by 'finally' */
-        WHY_RETURN =            0x0008, /* 'return' statement */
-        WHY_BREAK =             0x0010, /* 'break' statement */
-        WHY_CONTINUE =          0x0020, /* 'continue' statement */
-        WHY_YIELD =             0x0040, /* 'yield' operator */
-        WHY_SILENCED = 0x0080 /* Exception silenced by 'with' */
+        WHY_NOT =       0x0001, /* No error */
+        WHY_EXCEPTION = 0x0002, /* Exception occurred */
+        WHY_RERAISE =   0x0004, /* Exception re-raised by 'finally' */
+        WHY_RETURN =    0x0008, /* 'return' statement */
+        WHY_BREAK =     0x0010, /* 'break' statement */
+        WHY_CONTINUE =  0x0020, /* 'continue' statement */
+        WHY_YIELD =     0x0040, /* 'yield' operator */
+        WHY_SILENCED =  0x0080  /* Exception silenced by 'with' */
 };
 
 static enum why_code do_raise(PyObject *, PyObject *);
@@ -1005,38 +1005,38 @@
 
 /* The stack can grow at most MAXINT deep, as co_nlocals and
    co_stacksize are ints. */
-#define STACK_LEVEL()   ((int)(stack_pointer - f->f_valuestack))
-#define EMPTY()         (STACK_LEVEL() == 0)
-#define TOP()           (stack_pointer[-1])
-#define SECOND()        (stack_pointer[-2])
-#define THIRD()         (stack_pointer[-3])
-#define FOURTH()        (stack_pointer[-4])
-#define PEEK(n)         (stack_pointer[-(n)])
-#define SET_TOP(v)      (stack_pointer[-1] = (v))
-#define SET_SECOND(v)   (stack_pointer[-2] = (v))
-#define SET_THIRD(v)    (stack_pointer[-3] = (v))
-#define SET_FOURTH(v)   (stack_pointer[-4] = (v))
-#define SET_VALUE(n, v) (stack_pointer[-(n)] = (v))
-#define BASIC_STACKADJ(n)       (stack_pointer += n)
-#define BASIC_PUSH(v)   (*stack_pointer++ = (v))
-#define BASIC_POP()     (*--stack_pointer)
+#define STACK_LEVEL()     ((int)(stack_pointer - f->f_valuestack))
+#define EMPTY()           (STACK_LEVEL() == 0)
+#define TOP()             (stack_pointer[-1])
+#define SECOND()          (stack_pointer[-2])
+#define THIRD()           (stack_pointer[-3])
+#define FOURTH()          (stack_pointer[-4])
+#define PEEK(n)           (stack_pointer[-(n)])
+#define SET_TOP(v)        (stack_pointer[-1] = (v))
+#define SET_SECOND(v)     (stack_pointer[-2] = (v))
+#define SET_THIRD(v)      (stack_pointer[-3] = (v))
+#define SET_FOURTH(v)     (stack_pointer[-4] = (v))
+#define SET_VALUE(n, v)   (stack_pointer[-(n)] = (v))
+#define BASIC_STACKADJ(n) (stack_pointer += n)
+#define BASIC_PUSH(v)     (*stack_pointer++ = (v))
+#define BASIC_POP()       (*--stack_pointer)
 
 #ifdef LLTRACE
 #define PUSH(v)         { (void)(BASIC_PUSH(v), \
-                   lltrace && prtrace(TOP(), "push")); \
-                   assert(STACK_LEVEL() <= co->co_stacksize); }
+                          lltrace && prtrace(TOP(), "push")); \
+                          assert(STACK_LEVEL() <= co->co_stacksize); }
 #define POP()           ((void)(lltrace && prtrace(TOP(), "pop")), \
-             BASIC_POP())
+                         BASIC_POP())
 #define STACKADJ(n)     { (void)(BASIC_STACKADJ(n), \
-                   lltrace && prtrace(TOP(), "stackadj")); \
-                   assert(STACK_LEVEL() <= co->co_stacksize); }
+                          lltrace && prtrace(TOP(), "stackadj")); \
+                          assert(STACK_LEVEL() <= co->co_stacksize); }
 #define EXT_POP(STACK_POINTER) ((void)(lltrace && \
-                prtrace((STACK_POINTER)[-1], "ext_pop")), \
-                *--(STACK_POINTER))
+                                prtrace((STACK_POINTER)[-1], "ext_pop")), \
+                                *--(STACK_POINTER))
 #else
-#define PUSH(v)         BASIC_PUSH(v)
-#define POP()           BASIC_POP()
-#define STACKADJ(n)     BASIC_STACKADJ(n)
+#define PUSH(v)                BASIC_PUSH(v)
+#define POP()                  BASIC_POP()
+#define STACKADJ(n)            BASIC_STACKADJ(n)
 #define EXT_POP(STACK_POINTER) (*--(STACK_POINTER))
 #endif
 
@@ -1051,8 +1051,8 @@
    accessed by other code (e.g. a __del__ method or gc.collect()) and the
    variable would be pointing to already-freed memory. */
 #define SETLOCAL(i, value)      do { PyObject *tmp = GETLOCAL(i); \
-                     GETLOCAL(i) = value; \
-                     Py_XDECREF(tmp); } while (0)
+                                     GETLOCAL(i) = value; \
+                                     Py_XDECREF(tmp); } while (0)
 
 
 #define UNWIND_BLOCK(b) \
@@ -1318,7 +1318,7 @@
             it doesn't have to be remembered across a full loop */
         if (HAS_ARG(opcode))
             oparg = NEXTARG();
-      dispatch_opcode:
+    dispatch_opcode:
 #ifdef DYNAMIC_EXECUTION_PROFILE
 #ifdef DXPAIRS
         dxpairs[lastopcode][opcode]++;
@@ -2174,8 +2174,8 @@
                 break;
             if (oparg < PyTuple_GET_SIZE(co->co_cellvars)) {
                 v = PyTuple_GET_ITEM(co->co_cellvars,
-                                       oparg);
-                   format_exc_check_arg(
+                                     oparg);
+                format_exc_check_arg(
                        PyExc_UnboundLocalError,
                        UNBOUNDLOCAL_ERROR_MSG,
                        v);
@@ -2695,7 +2695,7 @@
             func = *pfunc;
 
             if (PyMethod_Check(func)
-            && PyMethod_GET_SELF(func) != NULL) {
+                && PyMethod_GET_SELF(func) != NULL) {
                 PyObject *self = PyMethod_GET_SELF(func);
                 Py_INCREF(self);
                 func = PyMethod_GET_FUNCTION(func);
@@ -3100,11 +3100,11 @@
             if (!(co->co_flags & CO_VARARGS)) {
                 PyErr_Format(PyExc_TypeError,
                     "%U() takes %s %d "
-                    "argument%s (%d given)",
+                    "positional argument%s (%d given)",
                     co->co_name,
                     defcount ? "at most" : "exactly",
-                    total_args,
-                    total_args == 1 ? "" : "s",
+                    co->co_argcount,
+                    co->co_argcount == 1 ? "" : "s",
                     argcount + kwcount);
                 goto fail;
             }
@@ -3464,7 +3464,8 @@
             return 1;
         }
         Py_DECREF(w);
-        PyErr_SetString(PyExc_ValueError, "too many values to unpack");
+        PyErr_Format(PyExc_ValueError, "too many values to unpack "
+                     "(expected %d)", argcnt);
         goto Error;
     }
 

Modified: python/branches/py3k-cdecimal/Python/compile.c
==============================================================================
--- python/branches/py3k-cdecimal/Python/compile.c	(original)
+++ python/branches/py3k-cdecimal/Python/compile.c	Sat Jul 17 12:36:09 2010
@@ -968,7 +968,7 @@
 {
     int arg = compiler_add_o(c, dict, o);
     if (arg < 0)
-    return 0;
+        return 0;
     return compiler_addop_i(c, opcode, arg);
 }
 
@@ -979,11 +979,11 @@
     int arg;
     PyObject *mangled = _Py_Mangle(c->u->u_private, o);
     if (!mangled)
-    return 0;
+        return 0;
     arg = compiler_add_o(c, dict, mangled);
     Py_DECREF(mangled);
     if (arg < 0)
-    return 0;
+        return 0;
     return compiler_addop_i(c, opcode, arg);
 }
 
@@ -1134,7 +1134,7 @@
 compiler_isdocstring(stmt_ty s)
 {
     if (s->kind != Expr_kind)
-    return 0;
+        return 0;
     return s->v.Expr.value->kind == Str_kind;
 }
 
@@ -1240,11 +1240,11 @@
     PyObject *k, *v;
     k = PyTuple_Pack(2, name, name->ob_type);
     if (k == NULL)
-    return -1;
+        return -1;
     v = PyDict_GetItem(dict, k);
     Py_DECREF(k);
     if (v == NULL)
-    return -1;
+        return -1;
     return PyLong_AS_LONG(v);
 }
 
@@ -3073,7 +3073,7 @@
     block = compiler_new_block(c);
     finally = compiler_new_block(c);
     if (!block || !finally)
-    return 0;
+        return 0;
 
     /* Evaluate EXPR */
     VISIT(c, expr, s->v.With.context_expr);
@@ -3082,15 +3082,15 @@
     /* SETUP_WITH pushes a finally block. */
     compiler_use_next_block(c, block);
     if (!compiler_push_fblock(c, FINALLY_TRY, block)) {
-    return 0;
+        return 0;
     }
 
     if (s->v.With.optional_vars) {
-    VISIT(c, expr, s->v.With.optional_vars);
+        VISIT(c, expr, s->v.With.optional_vars);
     }
     else {
     /* Discard result from context.__enter__() */
-    ADDOP(c, POP_TOP);
+        ADDOP(c, POP_TOP);
     }
 
     /* BLOCK code */
@@ -3103,7 +3103,7 @@
     ADDOP_O(c, LOAD_CONST, Py_None, consts);
     compiler_use_next_block(c, finally);
     if (!compiler_push_fblock(c, FINALLY_END, finally))
-    return 0;
+        return 0;
 
     /* Finally block starts; context.__exit__ is on the stack under
        the exception or return information. Just issue our magic
@@ -3942,7 +3942,7 @@
     freevars = dict_keys_inorder(c->u->u_freevars, PyTuple_Size(cellvars));
     if (!freevars)
         goto error;
-    filename = PyUnicode_DecodeFSDefault(c->c_filename);
+    filename = PyUnicode_FromString(c->c_filename);
     if (!filename)
         goto error;
 

Modified: python/branches/py3k-cdecimal/Python/getargs.c
==============================================================================
--- python/branches/py3k-cdecimal/Python/getargs.c	(original)
+++ python/branches/py3k-cdecimal/Python/getargs.c	Sat Jul 17 12:36:09 2010
@@ -853,70 +853,6 @@
     /* XXX WAAAAH!  's', 'y', 'z', 'u', 'Z', 'e', 'w' codes all
        need to be cleaned up! */
 
-    case 's': {/* text string */
-        if (*format == '*') {
-            Py_buffer *p = (Py_buffer *)va_arg(*p_va, Py_buffer *);
-
-            if (PyUnicode_Check(arg)) {
-                uarg = UNICODE_DEFAULT_ENCODING(arg);
-                if (uarg == NULL)
-                    return converterr(CONV_UNICODE,
-                                      arg, msgbuf, bufsize);
-                PyBuffer_FillInfo(p, arg,
-                                  PyBytes_AS_STRING(uarg), PyBytes_GET_SIZE(uarg),
-                                  1, 0);
-            }
-            else { /* any buffer-like object */
-                char *buf;
-                if (getbuffer(arg, p, &buf) < 0)
-                    return converterr(buf, arg, msgbuf, bufsize);
-            }
-            if (addcleanup(p, freelist, cleanup_buffer)) {
-                return converterr(
-                    "(cleanup problem)",
-                    arg, msgbuf, bufsize);
-            }
-            format++;
-        } else if (*format == '#') {
-            void **p = (void **)va_arg(*p_va, char **);
-            FETCH_SIZE;
-
-            if (PyUnicode_Check(arg)) {
-                uarg = UNICODE_DEFAULT_ENCODING(arg);
-                if (uarg == NULL)
-                    return converterr(CONV_UNICODE,
-                                      arg, msgbuf, bufsize);
-                *p = PyBytes_AS_STRING(uarg);
-                STORE_SIZE(PyBytes_GET_SIZE(uarg));
-            }
-            else { /* any buffer-like object */
-                /* XXX Really? */
-                char *buf;
-                Py_ssize_t count = convertbuffer(arg, p, &buf);
-                if (count < 0)
-                    return converterr(buf, arg, msgbuf, bufsize);
-                STORE_SIZE(count);
-            }
-            format++;
-        } else {
-            char **p = va_arg(*p_va, char **);
-
-            if (PyUnicode_Check(arg)) {
-                uarg = UNICODE_DEFAULT_ENCODING(arg);
-                if (uarg == NULL)
-                    return converterr(CONV_UNICODE,
-                                      arg, msgbuf, bufsize);
-                *p = PyBytes_AS_STRING(uarg);
-            }
-            else
-                return converterr("string", arg, msgbuf, bufsize);
-            if ((Py_ssize_t) strlen(*p) != PyBytes_GET_SIZE(uarg))
-                return converterr("string without null bytes",
-                                  arg, msgbuf, bufsize);
-        }
-        break;
-    }
-
     case 'y': {/* any buffer-like object, but not PyUnicode */
         void **p = (void **)va_arg(*p_va, char **);
         char *buf;
@@ -948,11 +884,14 @@
         break;
     }
 
-    case 'z': {/* like 's' or 's#', but None is okay, stored as NULL */
+    case 's': /* text string */
+    case 'z': /* text string or None */
+    {
         if (*format == '*') {
+            /* "s*" or "z*" */
             Py_buffer *p = (Py_buffer *)va_arg(*p_va, Py_buffer *);
 
-            if (arg == Py_None)
+            if (c == 'z' && arg == Py_None)
                 PyBuffer_FillInfo(p, NULL, NULL, 0, 1, 0);
             else if (PyUnicode_Check(arg)) {
                 uarg = UNICODE_DEFAULT_ENCODING(arg);
@@ -975,11 +914,12 @@
             }
             format++;
         } else if (*format == '#') { /* any buffer-like object */
+            /* "s#" or "z#" */
             void **p = (void **)va_arg(*p_va, char **);
             FETCH_SIZE;
 
-            if (arg == Py_None) {
-                *p = 0;
+            if (c == 'z' && arg == Py_None) {
+                *p = NULL;
                 STORE_SIZE(0);
             }
             else if (PyUnicode_Check(arg)) {
@@ -1000,16 +940,12 @@
             }
             format++;
         } else {
+            /* "s" or "z" */
             char **p = va_arg(*p_va, char **);
             uarg = NULL;
 
-            if (arg == Py_None)
-                *p = 0;
-            else if (PyBytes_Check(arg)) {
-                /* Enable null byte check below */
-                uarg = arg;
-                *p = PyBytes_AS_STRING(arg);
-            }
+            if (c == 'z' && arg == Py_None)
+                *p = NULL;
             else if (PyUnicode_Check(arg)) {
                 uarg = UNICODE_DEFAULT_ENCODING(arg);
                 if (uarg == NULL)
@@ -1018,24 +954,28 @@
                 *p = PyBytes_AS_STRING(uarg);
             }
             else
-                return converterr("string or None",
+                return converterr(c == 'z' ? "str or None" : "str",
                                   arg, msgbuf, bufsize);
             if (*p != NULL && uarg != NULL &&
                 (Py_ssize_t) strlen(*p) != PyBytes_GET_SIZE(uarg))
                 return converterr(
-                    "string without null bytes or None",
+                    c == 'z' ? "str without null bytes or None"
+                             : "str without null bytes",
                     arg, msgbuf, bufsize);
         }
         break;
     }
 
-    case 'Z': {/* unicode, may be NULL (None) */
+    case 'u': /* raw unicode buffer (Py_UNICODE *) */
+    case 'Z': /* raw unicode buffer or None */
+    {
         if (*format == '#') { /* any buffer-like object */
+            /* "s#" or "Z#" */
             Py_UNICODE **p = va_arg(*p_va, Py_UNICODE **);
             FETCH_SIZE;
 
-            if (arg == Py_None) {
-                *p = 0;
+            if (c == 'Z' && arg == Py_None) {
+                *p = NULL;
                 STORE_SIZE(0);
             }
             else if (PyUnicode_Check(arg)) {
@@ -1046,10 +986,11 @@
                 return converterr("str or None", arg, msgbuf, bufsize);
             format++;
         } else {
+            /* "s" or "Z" */
             Py_UNICODE **p = va_arg(*p_va, Py_UNICODE **);
 
-            if (arg == Py_None)
-                *p = 0;
+            if (c == 'Z' && arg == Py_None)
+                *p = NULL;
             else if (PyUnicode_Check(arg)) {
                 *p = PyUnicode_AS_UNICODE(arg);
                 if (Py_UNICODE_strlen(*p) != PyUnicode_GET_SIZE(arg))
@@ -1057,7 +998,8 @@
                         "str without null character or None",
                         arg, msgbuf, bufsize);
             } else
-                return converterr("str or None", arg, msgbuf, bufsize);
+                return converterr(c == 'Z' ? "str or None" : "str",
+                                  arg, msgbuf, bufsize);
         }
         break;
     }
@@ -1227,24 +1169,6 @@
         break;
     }
 
-    case 'u': {/* raw unicode buffer (Py_UNICODE *) */
-        Py_UNICODE **p = va_arg(*p_va, Py_UNICODE **);
-        if (!PyUnicode_Check(arg))
-            return converterr("str", arg, msgbuf, bufsize);
-        *p = PyUnicode_AS_UNICODE(arg);
-        if (*format == '#') { /* store pointer and size */
-            FETCH_SIZE;
-            STORE_SIZE(PyUnicode_GET_SIZE(arg));
-            format++;
-        } else {
-            if (Py_UNICODE_strlen(*p) != PyUnicode_GET_SIZE(arg))
-                return converterr(
-                    "str without null character",
-                    arg, msgbuf, bufsize);
-        }
-        break;
-    }
-
     case 'S': { /* PyBytes object */
         PyObject **p = va_arg(*p_va, PyObject **);
         if (PyBytes_Check(arg))
@@ -1307,58 +1231,28 @@
     }
 
 
-    case 'w': { /* memory buffer, read-write access */
+    case 'w': { /* "w*": memory buffer, read-write access */
         void **p = va_arg(*p_va, void **);
-        PyBufferProcs *pb = arg->ob_type->tp_as_buffer;
-        Py_ssize_t count;
-        int temp=-1;
-        Py_buffer view;
 
-        if (pb && pb->bf_releasebuffer && *format != '*')
-            /* Buffer must be released, yet caller does not use
-               the Py_buffer protocol. */
-            return converterr("pinned buffer", arg, msgbuf, bufsize);
-
-
-        if (pb && pb->bf_getbuffer && *format == '*') {
-            /* Caller is interested in Py_buffer, and the object
-               supports it directly. */
-            format++;
-            if (PyObject_GetBuffer(arg, (Py_buffer*)p, PyBUF_WRITABLE) < 0) {
-                PyErr_Clear();
-                return converterr("read-write buffer", arg, msgbuf, bufsize);
-            }
-            if (addcleanup(p, freelist, cleanup_buffer)) {
-                return converterr(
-                    "(cleanup problem)",
-                    arg, msgbuf, bufsize);
-            }
-            if (!PyBuffer_IsContiguous((Py_buffer*)p, 'C'))
-                return converterr("contiguous buffer", arg, msgbuf, bufsize);
-            break;
-        }
+        if (*format != '*')
+            return converterr(
+                "invalid use of 'w' format character",
+                arg, msgbuf, bufsize);
+        format++;
 
-        /* Here we have processed w*, only w and w# remain. */
-        if (pb == NULL ||
-            pb->bf_getbuffer == NULL ||
-            ((temp = PyObject_GetBuffer(arg, &view,
-                                        PyBUF_SIMPLE)) != 0) ||
-            view.readonly == 1) {
-            if (temp==0) {
-                PyBuffer_Release(&view);
-            }
-            return converterr("single-segment read-write buffer",
-                              arg, msgbuf, bufsize);
+        /* Caller is interested in Py_buffer, and the object
+           supports it directly. */
+        if (PyObject_GetBuffer(arg, (Py_buffer*)p, PyBUF_WRITABLE) < 0) {
+            PyErr_Clear();
+            return converterr("read-write buffer", arg, msgbuf, bufsize);
         }
-
-        if ((count = view.len) < 0)
-            return converterr("(unspecified)", arg, msgbuf, bufsize);
-        *p = view.buf;
-        if (*format == '#') {
-            FETCH_SIZE;
-            STORE_SIZE(count);
-            format++;
+        if (addcleanup(p, freelist, cleanup_buffer)) {
+            return converterr(
+                "(cleanup problem)",
+                arg, msgbuf, bufsize);
         }
+        if (!PyBuffer_IsContiguous((Py_buffer*)p, 'C'))
+            return converterr("contiguous buffer", arg, msgbuf, bufsize);
         break;
     }
 
@@ -1416,6 +1310,7 @@
         return -1;
     }
     if (!PyBuffer_IsContiguous(view, 'C')) {
+        PyBuffer_Release(view);
         *errmsg = "contiguous buffer";
         return -1;
     }

Modified: python/branches/py3k-cdecimal/Python/graminit.c
==============================================================================
--- python/branches/py3k-cdecimal/Python/graminit.c	(original)
+++ python/branches/py3k-cdecimal/Python/graminit.c	Sat Jul 17 12:36:09 2010
@@ -180,8 +180,8 @@
     {24, 9},
 };
 static arc arcs_8_5[4] = {
-    {28, 1},
-    {31, 2},
+    {28, 10},
+    {31, 11},
     {32, 3},
     {0, 5},
 };
@@ -190,7 +190,7 @@
     {0, 6},
 };
 static arc arcs_8_7[2] = {
-    {28, 10},
+    {28, 12},
     {32, 3},
 };
 static arc arcs_8_8[1] = {
@@ -201,14 +201,40 @@
     {0, 9},
 };
 static arc arcs_8_10[3] = {
-    {30, 7},
-    {29, 11},
+    {30, 5},
+    {29, 4},
     {0, 10},
 };
-static arc arcs_8_11[1] = {
+static arc arcs_8_11[3] = {
+    {28, 13},
+    {30, 14},
+    {0, 11},
+};
+static arc arcs_8_12[3] = {
+    {30, 7},
+    {29, 15},
+    {0, 12},
+};
+static arc arcs_8_13[2] = {
+    {30, 14},
+    {0, 13},
+};
+static arc arcs_8_14[2] = {
+    {28, 16},
+    {32, 3},
+};
+static arc arcs_8_15[1] = {
     {24, 6},
 };
-static state states_8[12] = {
+static arc arcs_8_16[3] = {
+    {30, 14},
+    {29, 17},
+    {0, 16},
+};
+static arc arcs_8_17[1] = {
+    {24, 13},
+};
+static state states_8[18] = {
     {3, arcs_8_0},
     {3, arcs_8_1},
     {3, arcs_8_2},
@@ -220,7 +246,13 @@
     {1, arcs_8_8},
     {2, arcs_8_9},
     {3, arcs_8_10},
-    {1, arcs_8_11},
+    {3, arcs_8_11},
+    {3, arcs_8_12},
+    {2, arcs_8_13},
+    {2, arcs_8_14},
+    {1, arcs_8_15},
+    {3, arcs_8_16},
+    {1, arcs_8_17},
 };
 static arc arcs_9_0[1] = {
     {21, 1},
@@ -263,8 +295,8 @@
     {24, 9},
 };
 static arc arcs_10_5[4] = {
-    {34, 1},
-    {31, 2},
+    {34, 10},
+    {31, 11},
     {32, 3},
     {0, 5},
 };
@@ -273,7 +305,7 @@
     {0, 6},
 };
 static arc arcs_10_7[2] = {
-    {34, 10},
+    {34, 12},
     {32, 3},
 };
 static arc arcs_10_8[1] = {
@@ -284,14 +316,40 @@
     {0, 9},
 };
 static arc arcs_10_10[3] = {
-    {30, 7},
-    {29, 11},
+    {30, 5},
+    {29, 4},
     {0, 10},
 };
-static arc arcs_10_11[1] = {
+static arc arcs_10_11[3] = {
+    {34, 13},
+    {30, 14},
+    {0, 11},
+};
+static arc arcs_10_12[3] = {
+    {30, 7},
+    {29, 15},
+    {0, 12},
+};
+static arc arcs_10_13[2] = {
+    {30, 14},
+    {0, 13},
+};
+static arc arcs_10_14[2] = {
+    {34, 16},
+    {32, 3},
+};
+static arc arcs_10_15[1] = {
     {24, 6},
 };
-static state states_10[12] = {
+static arc arcs_10_16[3] = {
+    {30, 14},
+    {29, 17},
+    {0, 16},
+};
+static arc arcs_10_17[1] = {
+    {24, 13},
+};
+static state states_10[18] = {
     {3, arcs_10_0},
     {3, arcs_10_1},
     {3, arcs_10_2},
@@ -303,7 +361,13 @@
     {1, arcs_10_8},
     {2, arcs_10_9},
     {3, arcs_10_10},
-    {1, arcs_10_11},
+    {3, arcs_10_11},
+    {3, arcs_10_12},
+    {2, arcs_10_13},
+    {2, arcs_10_14},
+    {1, arcs_10_15},
+    {3, arcs_10_16},
+    {1, arcs_10_17},
 };
 static arc arcs_11_0[1] = {
     {21, 1},
@@ -1755,11 +1819,11 @@
      "\000\000\020\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000"},
     {263, "parameters", 0, 4, states_7,
      "\000\040\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000"},
-    {264, "typedargslist", 0, 12, states_8,
+    {264, "typedargslist", 0, 18, states_8,
      "\000\000\040\200\001\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000"},
     {265, "tfpdef", 0, 4, states_9,
      "\000\000\040\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000"},
-    {266, "varargslist", 0, 12, states_10,
+    {266, "varargslist", 0, 18, states_10,
      "\000\000\040\200\001\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000"},
     {267, "vfpdef", 0, 2, states_11,
      "\000\000\040\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000"},

Modified: python/branches/py3k-cdecimal/Python/import.c
==============================================================================
--- python/branches/py3k-cdecimal/Python/import.c	(original)
+++ python/branches/py3k-cdecimal/Python/import.c	Sat Jul 17 12:36:09 2010
@@ -558,10 +558,10 @@
    copy can be retrieved from there by calling
    _PyImport_FindExtension().
 
-   Modules which do support multiple multiple initialization set
-   their m_size field to a non-negative number (indicating the size
-   of the module-specific state). They are still recorded in the
-   extensions dictionary, to avoid loading shared libraries twice.
+   Modules which do support multiple initialization set their m_size
+   field to a non-negative number (indicating the size of the
+   module-specific state). They are still recorded in the extensions
+   dictionary, to avoid loading shared libraries twice.
 */
 
 int
@@ -2385,7 +2385,8 @@
     if (parent == NULL)
         return NULL;
 
-    head = load_next(parent, Py_None, &name, buf, &buflen);
+    head = load_next(parent, level < 0 ? Py_None : parent, &name, buf,
+                        &buflen);
     if (head == NULL)
         return NULL;
 

Modified: python/branches/py3k-cdecimal/Python/pythonrun.c
==============================================================================
--- python/branches/py3k-cdecimal/Python/pythonrun.c	(original)
+++ python/branches/py3k-cdecimal/Python/pythonrun.c	Sat Jul 17 12:36:09 2010
@@ -1190,7 +1190,7 @@
     d = PyModule_GetDict(m);
     if (PyDict_GetItemString(d, "__file__") == NULL) {
         PyObject *f;
-        f = PyUnicode_DecodeFSDefault(filename);
+        f = PyUnicode_FromString(filename);
         if (f == NULL)
             return -1;
         if (PyDict_SetItemString(d, "__file__", f) < 0) {

Modified: python/branches/py3k-cdecimal/Python/sysmodule.c
==============================================================================
--- python/branches/py3k-cdecimal/Python/sysmodule.c	(original)
+++ python/branches/py3k-cdecimal/Python/sysmodule.c	Sat Jul 17 12:36:09 2010
@@ -595,7 +595,7 @@
     5,
 };
 
-PyObject *
+static PyObject *
 get_hash_info(void)
 {
     PyObject *hash_info;

Modified: python/branches/py3k-cdecimal/Python/traceback.c
==============================================================================
--- python/branches/py3k-cdecimal/Python/traceback.c	(original)
+++ python/branches/py3k-cdecimal/Python/traceback.c	Sat Jul 17 12:36:09 2010
@@ -133,33 +133,38 @@
     return 0;
 }
 
-static int
-_Py_FindSourceFile(const char* filename, char* namebuf, size_t namelen, int open_flags)
+static PyObject *
+_Py_FindSourceFile(PyObject *filename, char* namebuf, size_t namelen, PyObject *io)
 {
-    int i;
-    int fd = -1;
+    Py_ssize_t i;
+    PyObject *binary;
     PyObject *v;
-    Py_ssize_t _npath;
-    int npath;
+    Py_ssize_t npath;
     size_t taillen;
     PyObject *syspath;
     const char* path;
     const char* tail;
+    const char* filepath;
     Py_ssize_t len;
 
+    filepath = _PyUnicode_AsString(filename);
+    if (filepath == NULL) {
+        PyErr_Clear();
+        return NULL;
+    }
+
     /* Search tail of filename in sys.path before giving up */
-    tail = strrchr(filename, SEP);
+    tail = strrchr(filepath, SEP);
     if (tail == NULL)
-        tail = filename;
+        tail = filepath;
     else
         tail++;
     taillen = strlen(tail);
 
     syspath = PySys_GetObject("path");
     if (syspath == NULL || !PyList_Check(syspath))
-        return -1;
-    _npath = PyList_Size(syspath);
-    npath = Py_SAFE_DOWNCAST(_npath, Py_ssize_t, int);
+        return NULL;
+    npath = PyList_Size(syspath);
 
     for (i = 0; i < npath; i++) {
         v = PyList_GetItem(syspath, i);
@@ -170,6 +175,10 @@
         if (!PyUnicode_Check(v))
             continue;
         path = _PyUnicode_AsStringAndSize(v, &len);
+        if (path == NULL) {
+            PyErr_Clear();
+            continue;
+        }
         if (len + 1 + (Py_ssize_t)taillen >= (Py_ssize_t)namelen - 1)
             continue; /* Too long */
         strcpy(namebuf, path);
@@ -178,31 +187,27 @@
         if (len > 0 && namebuf[len-1] != SEP)
             namebuf[len++] = SEP;
         strcpy(namebuf+len, tail);
-        Py_BEGIN_ALLOW_THREADS
-        fd = open(namebuf, open_flags);
-        Py_END_ALLOW_THREADS
-        if (0 <= fd) {
-            return fd;
-        }
+
+        binary = PyObject_CallMethod(io, "open", "ss", namebuf, "rb");
+        if (binary != NULL)
+            return binary;
+        PyErr_Clear();
     }
-    return -1;
+    return NULL;
 }
 
 int
-_Py_DisplaySourceLine(PyObject *f, const char *filename, int lineno, int indent)
+_Py_DisplaySourceLine(PyObject *f, PyObject *filename, int lineno, int indent)
 {
     int err = 0;
     int fd;
     int i;
     char *found_encoding;
     char *encoding;
+    PyObject *io;
+    PyObject *binary;
     PyObject *fob = NULL;
     PyObject *lineobj = NULL;
-#ifdef O_BINARY
-    const int open_flags = O_RDONLY | O_BINARY;   /* necessary for Windows */
-#else
-    const int open_flags = O_RDONLY;
-#endif
     char buf[MAXPATHLEN+1];
     Py_UNICODE *u, *p;
     Py_ssize_t len;
@@ -210,27 +215,32 @@
     /* open the file */
     if (filename == NULL)
         return 0;
-    Py_BEGIN_ALLOW_THREADS
-    fd = open(filename, open_flags);
-    Py_END_ALLOW_THREADS
-    if (fd < 0) {
-        fd = _Py_FindSourceFile(filename, buf, sizeof(buf), open_flags);
-        if (fd < 0)
+
+    io = PyImport_ImportModuleNoBlock("io");
+    if (io == NULL)
+        return -1;
+    binary = PyObject_CallMethod(io, "open", "Os", filename, "rb");
+
+    if (binary == NULL) {
+        binary = _Py_FindSourceFile(filename, buf, sizeof(buf), io);
+        if (binary == NULL) {
+            Py_DECREF(io);
             return 0;
-        filename = buf;
+        }
     }
 
     /* use the right encoding to decode the file as unicode */
+    fd = PyObject_AsFileDescriptor(binary);
     found_encoding = PyTokenizer_FindEncoding(fd);
-    encoding = (found_encoding != NULL) ? found_encoding :
-        (char*)PyUnicode_GetDefaultEncoding();
+    encoding = (found_encoding != NULL) ? found_encoding : "utf-8";
     lseek(fd, 0, 0); /* Reset position */
-    fob = PyFile_FromFd(fd, (char*)filename, "r", -1, (char*)encoding,
-        NULL, NULL, 1);
+    fob = PyObject_CallMethod(io, "TextIOWrapper", "Os", binary, encoding);
+    Py_DECREF(io);
+    Py_DECREF(binary);
     PyMem_FREE(found_encoding);
+
     if (fob == NULL) {
         PyErr_Clear();
-        close(fd);
         return 0;
     }
 
@@ -287,17 +297,19 @@
 }
 
 static int
-tb_displayline(PyObject *f, const char *filename, int lineno, const char *name)
+tb_displayline(PyObject *f, PyObject *filename, int lineno, PyObject *name)
 {
-    int err = 0;
-    char linebuf[2000];
+    int err;
+    PyObject *line;
 
     if (filename == NULL || name == NULL)
         return -1;
-    /* This is needed by Emacs' compile command */
-#define FMT "  File \"%.500s\", line %d, in %.500s\n"
-    PyOS_snprintf(linebuf, sizeof(linebuf), FMT, filename, lineno, name);
-    err = PyFile_WriteString(linebuf, f);
+    line = PyUnicode_FromFormat("  File \"%U\", line %d, in %U\n",
+                                filename, lineno, name);
+    if (line == NULL)
+        return -1;
+    err = PyFile_WriteObject(line, f, Py_PRINT_RAW);
+    Py_DECREF(line);
     if (err != 0)
         return err;
     return _Py_DisplaySourceLine(f, filename, lineno, 4);
@@ -316,10 +328,9 @@
     while (tb != NULL && err == 0) {
         if (depth <= limit) {
             err = tb_displayline(f,
-                _PyUnicode_AsString(
-                    tb->tb_frame->f_code->co_filename),
-                tb->tb_lineno,
-                _PyUnicode_AsString(tb->tb_frame->f_code->co_name));
+                                 tb->tb_frame->f_code->co_filename,
+                                 tb->tb_lineno,
+                                 tb->tb_frame->f_code->co_name);
         }
         depth--;
         tb = tb->tb_next;

Modified: python/branches/py3k-cdecimal/README
==============================================================================
--- python/branches/py3k-cdecimal/README	(original)
+++ python/branches/py3k-cdecimal/README	Sat Jul 17 12:36:09 2010
@@ -172,13 +172,13 @@
 Release Schedule
 ----------------
 
-See PEP 375 for release details: http://www.python.org/dev/peps/pep-0375/
+See PEP 392 for release details: http://www.python.org/dev/peps/pep-0392/
 
 
 Copyright and License Information
 ---------------------------------
 
-Copyright (c) 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008
+Copyright (c) 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010
 Python Software Foundation.
 All rights reserved.
 

Modified: python/branches/py3k-cdecimal/Tools/README
==============================================================================
--- python/branches/py3k-cdecimal/Tools/README	(original)
+++ python/branches/py3k-cdecimal/Tools/README	Sat Jul 17 12:36:09 2010
@@ -1,18 +1,7 @@
 This directory contains a number of Python programs that are useful
 while building or extending Python.
 
-audiopy		Audiopy is a program to control the Solaris audio
-		device, allowing you to choose both the input and
-		output devices, and to set the output volume, that can
-		be run either as a command-line script, or as a
-		Tkinter application.
-
-compiler	Tools used to maintain the compiler package in the
-		standard library.
-
-faqwiz		FAQ Wizard.
-		See http://www.python.org/cgi-bin/faqw.py
-		for a live example.
+faqwiz		FAQ Wizard: a CGI script for a user-editable FAQ.
 
 freeze		Create a stand-alone executable from a Python program.
 
@@ -36,8 +25,6 @@
 versioncheck	A tool to automate checking whether you have the latest
 		version of a package (by Jack Jansen).
 
-webchecker	A link checker for web sites.
-
 world		Script to take a list of Internet addresses and print
 		out where in the world those addresses originate from,
 		based on the top-level domain country code found in

Modified: python/branches/py3k-cdecimal/Tools/gdb/libpython.py
==============================================================================
--- python/branches/py3k-cdecimal/Tools/gdb/libpython.py	(original)
+++ python/branches/py3k-cdecimal/Tools/gdb/libpython.py	Sat Jul 17 12:36:09 2010
@@ -43,6 +43,7 @@
 from __future__ import with_statement
 import gdb
 import locale
+import sys
 
 # Look up the gdb.Type for some standard types:
 _type_char_ptr = gdb.lookup_type('char').pointer() # char*
@@ -55,7 +56,6 @@
 
 Py_TPFLAGS_HEAPTYPE = (1L << 9)
 
-Py_TPFLAGS_INT_SUBCLASS      = (1L << 23)
 Py_TPFLAGS_LONG_SUBCLASS     = (1L << 24)
 Py_TPFLAGS_LIST_SUBCLASS     = (1L << 25)
 Py_TPFLAGS_TUPLE_SUBCLASS    = (1L << 26)
@@ -312,8 +312,6 @@
         if tp_flags & Py_TPFLAGS_HEAPTYPE:
             return HeapTypeObjectPtr
 
-        if tp_flags & Py_TPFLAGS_INT_SUBCLASS:
-            return PyIntObjectPtr
         if tp_flags & Py_TPFLAGS_LONG_SUBCLASS:
             return PyLongObjectPtr
         if tp_flags & Py_TPFLAGS_LIST_SUBCLASS:

Modified: python/branches/py3k-cdecimal/Tools/i18n/msgfmt.py
==============================================================================
--- python/branches/py3k-cdecimal/Tools/i18n/msgfmt.py	(original)
+++ python/branches/py3k-cdecimal/Tools/i18n/msgfmt.py	Sat Jul 17 12:36:09 2010
@@ -165,7 +165,7 @@
             section = STR
             if l.startswith('msgstr['):
                 if not is_plural:
-                    print(sys.stderr, 'plural without msgid_plural on %s:%d' % (infile, lno),
+                    print('plural without msgid_plural on %s:%d' % (infile, lno),
                           file=sys.stderr)
                     sys.exit(1)
                 l = l.split(']', 1)[1]
@@ -173,7 +173,7 @@
                     msgstr += b'\0' # Separator of the various plural forms
             else:
                 if is_plural:
-                    print(sys.stderr, 'indexed msgstr required for plural on  %s:%d' % (infile, lno),
+                    print('indexed msgstr required for plural on  %s:%d' % (infile, lno),
                           file=sys.stderr)
                     sys.exit(1)
                 l = l[6:]

Modified: python/branches/py3k-cdecimal/Tools/scripts/patchcheck.py
==============================================================================
--- python/branches/py3k-cdecimal/Tools/scripts/patchcheck.py	(original)
+++ python/branches/py3k-cdecimal/Tools/scripts/patchcheck.py	Sat Jul 17 12:36:09 2010
@@ -5,6 +5,10 @@
 import reindent
 
 
+def n_files_str(count):
+    """Return 'N file(s)' with the proper plurality on 'file'."""
+    return "{} file{}".format(count, "s" if count != 1 else "")
+
 def status(message, modal=False, info=None):
     """Decorator to output status info to stdout."""
     def decorated_fxn(fxn):
@@ -26,7 +30,7 @@
     return decorated_fxn
 
 @status("Getting the list of files that have been added/changed",
-            info=lambda x: "%s files" % len(x))
+            info=lambda x: n_files_str(len(x)))
 def changed_files():
     """Run ``svn status`` and return a set of files that have been
     changed/added."""
@@ -44,20 +48,30 @@
             files.add(path)
     return files
 
- at status("Fixing whitespace", info=lambda x: "%s files" % x)
+def report_modified_files(file_paths):
+    count = len(file_paths)
+    if count == 0:
+        return n_files_str(count)
+    else:
+        lines = ["{}:".format(n_files_str(count))]
+        for path in file_paths:
+            lines.append("  {}".format(path))
+        return "\n".join(lines)
+
+ at status("Fixing whitespace", info=report_modified_files)
 def normalize_whitespace(file_paths):
     """Make sure that the whitespace for .py files have been normalized."""
     reindent.makebackup = False  # No need to create backups.
-    result = list(map(reindent.check, (x for x in file_paths if x.endswith('.py'))))
-    return sum(result)
+    fixed = []
+    for path in (x for x in file_paths if x.endswith('.py')):
+        if reindent.check(path):
+            fixed.append(path)
+    return fixed
 
 @status("Docs modified", modal=True)
 def docs_modified(file_paths):
-    """Report if any files in the Docs directory."""
-    for path in file_paths:
-        if path.startswith("Doc"):
-            return True
-    return False
+    """Report if any file in the Doc directory has been changed."""
+    return bool(file_paths)
 
 @status("Misc/ACKS updated", modal=True)
 def credit_given(file_paths):
@@ -72,14 +86,18 @@
 
 def main():
     file_paths = changed_files()
-    # PEP 7/8 verification.
-    normalize_whitespace(file_paths)
+    python_files = [fn for fn in file_paths if fn.endswith('.py')]
+    c_files = [fn for fn in file_paths if fn.endswith(('.c', '.h'))]
+    docs = [fn for fn in file_paths if fn.startswith('Doc')]
+    special_files = {'Misc/ACKS', 'Misc/NEWS'} & set(file_paths)
+    # PEP 8 whitespace rules enforcement.
+    normalize_whitespace(python_files)
     # Docs updated.
-    docs_modified(file_paths)
+    docs_modified(docs)
     # Misc/ACKS changed.
-    credit_given(file_paths)
+    credit_given(special_files)
     # Misc/NEWS changed.
-    reported_news(file_paths)
+    reported_news(special_files)
 
     # Test suite run and passed.
     print()

Modified: python/branches/py3k-cdecimal/configure
==============================================================================
--- python/branches/py3k-cdecimal/configure	(original)
+++ python/branches/py3k-cdecimal/configure	Sat Jul 17 12:36:09 2010
@@ -1,5 +1,5 @@
 #! /bin/sh
-# From configure.in Revision: 81803 .
+# From configure.in Revision: 81804 .
 # Guess values for system-dependent variables and create Makefiles.
 # Generated by GNU Autoconf 2.65 for python 3.2.
 #
@@ -3158,9 +3158,12 @@
 (it is also a good idea to do 'make clean' before compiling)" "$LINENO" 5
 fi
 
-# If the user set CFLAGS, use this instead of the automatically
-# determined setting
-preset_cflags="$CFLAGS"
+# Don't let AC_PROG_CC set the default CFLAGS. It normally sets -g -O2
+# when the compiler supports them, but we don't always want -O2, and
+# we set -g later.
+if test -z "$CFLAGS"; then
+        CFLAGS=
+fi
 ac_ext=c
 ac_cpp='$CPP $CPPFLAGS'
 ac_compile='$CC -c $CFLAGS $CPPFLAGS conftest.$ac_ext >&5'
@@ -3952,10 +3955,6 @@
 ac_link='$CC -o conftest$ac_exeext $CFLAGS $CPPFLAGS $LDFLAGS conftest.$ac_ext $LIBS >&5'
 ac_compiler_gnu=$ac_cv_c_compiler_gnu
 
-if test ! -z "$preset_cflags"
-then
-	CFLAGS=$preset_cflags
-fi
 
 
 

Modified: python/branches/py3k-cdecimal/configure.in
==============================================================================
--- python/branches/py3k-cdecimal/configure.in	(original)
+++ python/branches/py3k-cdecimal/configure.in	Sat Jul 17 12:36:09 2010
@@ -461,14 +461,13 @@
 (it is also a good idea to do 'make clean' before compiling)])
 fi
 
-# If the user set CFLAGS, use this instead of the automatically
-# determined setting
-preset_cflags="$CFLAGS"
-AC_PROG_CC
-if test ! -z "$preset_cflags"
-then
-	CFLAGS=$preset_cflags
+# Don't let AC_PROG_CC set the default CFLAGS. It normally sets -g -O2
+# when the compiler supports them, but we don't always want -O2, and
+# we set -g later.
+if test -z "$CFLAGS"; then
+        CFLAGS=
 fi
+AC_PROG_CC
 
 AC_SUBST(CXX)
 AC_SUBST(MAINCC)

Modified: python/branches/py3k-cdecimal/setup.py
==============================================================================
--- python/branches/py3k-cdecimal/setup.py	(original)
+++ python/branches/py3k-cdecimal/setup.py	Sat Jul 17 12:36:09 2010
@@ -46,7 +46,7 @@
     """
     Returns True if 'path' can be located in an OSX SDK
     """
-    return path.startswith('/usr/') or path.startswith('/System/')
+    return (path.startswith('/usr/') and not path.startswith('/usr/local')) or path.startswith('/System/')
 
 def find_file(filename, std_dirs, paths):
     """Searches for the directory where a given file is located,
@@ -450,9 +450,9 @@
                                depends=['_math.h'],
                                libraries=math_libs) )
         # time operations and variables
-        exts.append( Extension('time', ['timemodule.c'],
+        exts.append( Extension('time', ['timemodule.c', '_time.c'],
                                libraries=math_libs) )
-        exts.append( Extension('datetime', ['datetimemodule.c', 'timemodule.c'],
+        exts.append( Extension('datetime', ['datetimemodule.c', '_time.c'],
                                libraries=math_libs) )
         # fast iterator tools implemented in C
         exts.append( Extension("itertools", ["itertoolsmodule.c"]) )
@@ -862,6 +862,7 @@
                         else:
                             if os.path.isdir(dn):
                                 tmp.append(dn)
+                    db_dirs_to_check = tmp
 
                     db_dirs_to_check = tmp
 

From python-checkins at python.org  Sat Jul 17 13:46:52 2010
From: python-checkins at python.org (stefan.krah)
Date: Sat, 17 Jul 2010 13:46:52 +0200 (CEST)
Subject: [Python-checkins] r82927 - python/branches/py3k/setup.py
Message-ID: <20100717114652.D2497EE992@mail.python.org>

Author: stefan.krah
Date: Sat Jul 17 13:46:52 2010
New Revision: 82927

Log:
Issue #7384: On Gentoo, libreadline.so is a "fake library", so ldd fails.
In that case, do not attempt to parse stderr output.



Modified:
   python/branches/py3k/setup.py

Modified: python/branches/py3k/setup.py
==============================================================================
--- python/branches/py3k/setup.py	(original)
+++ python/branches/py3k/setup.py	Sat Jul 17 13:46:52 2010
@@ -534,18 +534,19 @@
             tmpfile = os.path.join(self.build_temp, 'readline_termcap_lib')
             if not os.path.exists(self.build_temp):
                 os.makedirs(self.build_temp)
-            os.system("ldd %s > %s" % (do_readline, tmpfile))
-            fp = open(tmpfile)
-            for ln in fp:
-                if 'curses' in ln:
-                    readline_termcap_library = re.sub(
-                        r'.*lib(n?cursesw?)\.so.*', r'\1', ln
-                    ).rstrip()
-                    break
-                if 'tinfo' in ln: # termcap interface split out from ncurses
-                    readline_termcap_library = 'tinfo'
-                    break
-            fp.close()
+            ret = os.system("ldd %s > %s" % (do_readline, tmpfile))
+            if ret >> 8 == 0:
+                fp = open(tmpfile)
+                for ln in fp:
+                    if 'curses' in ln:
+                        readline_termcap_library = re.sub(
+                            r'.*lib(n?cursesw?)\.so.*', r'\1', ln
+                        ).rstrip()
+                        break
+                    if 'tinfo' in ln: # termcap interface split out from ncurses
+                        readline_termcap_library = 'tinfo'
+                        break
+                fp.close()
             os.unlink(tmpfile)
         # Issue 7384: If readline is already linked against curses,
         # use the same library for the readline and curses modules.

From python-checkins at python.org  Sat Jul 17 13:48:58 2010
From: python-checkins at python.org (stefan.krah)
Date: Sat, 17 Jul 2010 13:48:58 +0200 (CEST)
Subject: [Python-checkins] r82928 - in python/branches/release31-maint:
	setup.py
Message-ID: <20100717114858.A9666EE9B2@mail.python.org>

Author: stefan.krah
Date: Sat Jul 17 13:48:58 2010
New Revision: 82928

Log:
Merged revisions 82927 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82927 | stefan.krah | 2010-07-17 13:46:52 +0200 (Sat, 17 Jul 2010) | 4 lines
  
  Issue #7384: On Gentoo, libreadline.so is a "fake library", so ldd fails.
  In that case, do not attempt to parse stderr output.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/setup.py

Modified: python/branches/release31-maint/setup.py
==============================================================================
--- python/branches/release31-maint/setup.py	(original)
+++ python/branches/release31-maint/setup.py	Sat Jul 17 13:48:58 2010
@@ -547,18 +547,19 @@
             tmpfile = os.path.join(self.build_temp, 'readline_termcap_lib')
             if not os.path.exists(self.build_temp):
                 os.makedirs(self.build_temp)
-            os.system("ldd %s > %s" % (do_readline, tmpfile))
-            fp = open(tmpfile)
-            for ln in fp:
-                if 'curses' in ln:
-                    readline_termcap_library = re.sub(
-                        r'.*lib(n?cursesw?)\.so.*', r'\1', ln
-                    ).rstrip()
-                    break
-                if 'tinfo' in ln: # termcap interface split out from ncurses
-                    readline_termcap_library = 'tinfo'
-                    break
-            fp.close()
+            ret = os.system("ldd %s > %s" % (do_readline, tmpfile))
+            if ret >> 8 == 0:
+                fp = open(tmpfile)
+                for ln in fp:
+                    if 'curses' in ln:
+                        readline_termcap_library = re.sub(
+                            r'.*lib(n?cursesw?)\.so.*', r'\1', ln
+                        ).rstrip()
+                        break
+                    if 'tinfo' in ln: # termcap interface split out from ncurses
+                        readline_termcap_library = 'tinfo'
+                        break
+                fp.close()
             os.unlink(tmpfile)
         # Issue 7384: If readline is already linked against curses,
         # use the same library for the readline and curses modules.

From python-checkins at python.org  Sat Jul 17 14:21:08 2010
From: python-checkins at python.org (stefan.krah)
Date: Sat, 17 Jul 2010 14:21:08 +0200 (CEST)
Subject: [Python-checkins] r82929 - python/branches/release27-maint/setup.py
Message-ID: <20100717122108.747D1EE99B@mail.python.org>

Author: stefan.krah
Date: Sat Jul 17 14:21:08 2010
New Revision: 82929

Log:
Issue #7384: On Gentoo, libreadline.so is a "fake library", so ldd fails.
In that case, do not attempt to parse stderr output.



Modified:
   python/branches/release27-maint/setup.py

Modified: python/branches/release27-maint/setup.py
==============================================================================
--- python/branches/release27-maint/setup.py	(original)
+++ python/branches/release27-maint/setup.py	Sat Jul 17 14:21:08 2010
@@ -594,16 +594,18 @@
         # Determine if readline is already linked against curses or tinfo.
         if do_readline and find_executable('ldd'):
             fp = os.popen("ldd %s" % do_readline)
-            for ln in fp:
-                if 'curses' in ln:
-                    readline_termcap_library = re.sub(
-                        r'.*lib(n?cursesw?)\.so.*', r'\1', ln
-                    ).rstrip()
-                    break
-                if 'tinfo' in ln: # termcap interface split out from ncurses
-                    readline_termcap_library = 'tinfo'
-                    break
-            fp.close()
+            ldd_output = fp.readlines()
+            ret = fp.close()
+            if ret is None or ret >> 8 == 0:
+                for ln in ldd_output:
+                    if 'curses' in ln:
+                        readline_termcap_library = re.sub(
+                            r'.*lib(n?cursesw?)\.so.*', r'\1', ln
+                        ).rstrip()
+                        break
+                    if 'tinfo' in ln: # termcap interface split out from ncurses
+                        readline_termcap_library = 'tinfo'
+                        break
         # Issue 7384: If readline is already linked against curses,
         # use the same library for the readline and curses modules.
         if 'curses' in readline_termcap_library:

From python-checkins at python.org  Sat Jul 17 14:31:09 2010
From: python-checkins at python.org (stefan.krah)
Date: Sat, 17 Jul 2010 14:31:09 +0200 (CEST)
Subject: [Python-checkins] r82930 - in python/branches/release26-maint:
	setup.py
Message-ID: <20100717123109.68384EEA6F@mail.python.org>

Author: stefan.krah
Date: Sat Jul 17 14:31:09 2010
New Revision: 82930

Log:
Merged revisions 82929 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

........
  r82929 | stefan.krah | 2010-07-17 14:21:08 +0200 (Sat, 17 Jul 2010) | 4 lines
  
  Issue #7384: On Gentoo, libreadline.so is a "fake library", so ldd fails.
  In that case, do not attempt to parse stderr output.
........


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/setup.py

Modified: python/branches/release26-maint/setup.py
==============================================================================
--- python/branches/release26-maint/setup.py	(original)
+++ python/branches/release26-maint/setup.py	Sat Jul 17 14:31:09 2010
@@ -607,16 +607,18 @@
         # Determine if readline is already linked against curses or tinfo.
         if do_readline and find_executable('ldd'):
             fp = os.popen("ldd %s" % do_readline)
-            for ln in fp:
-                if 'curses' in ln:
-                    readline_termcap_library = re.sub(
-                        r'.*lib(n?cursesw?)\.so.*', r'\1', ln
-                    ).rstrip()
-                    break
-                if 'tinfo' in ln: # termcap interface split out from ncurses
-                    readline_termcap_library = 'tinfo'
-                    break
-            fp.close()
+            ldd_output = fp.readlines()
+            ret = fp.close()
+            if ret is None or ret >> 8 == 0:
+                for ln in ldd_output:
+                    if 'curses' in ln:
+                        readline_termcap_library = re.sub(
+                            r'.*lib(n?cursesw?)\.so.*', r'\1', ln
+                        ).rstrip()
+                        break
+                    if 'tinfo' in ln: # termcap interface split out from ncurses
+                        readline_termcap_library = 'tinfo'
+                        break
         # Issue 7384: If readline is already linked against curses,
         # use the same library for the readline and curses modules.
         # Disabled since applications relying on ncursesw might break.

From python-checkins at python.org  Sat Jul 17 17:51:21 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Sat, 17 Jul 2010 17:51:21 +0200 (CEST)
Subject: [Python-checkins] r82931 - in python/branches/py3k:
	Doc/library/pickletools.rst Lib/pickletools.py Misc/NEWS
Message-ID: <20100717155121.EB6A8EE983@mail.python.org>

Author: alexander.belopolsky
Date: Sat Jul 17 17:51:21 2010
New Revision: 82931

Log:
Issue #9268: Add annotation option to pickletools.dis


Modified:
   python/branches/py3k/Doc/library/pickletools.rst
   python/branches/py3k/Lib/pickletools.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Doc/library/pickletools.rst
==============================================================================
--- python/branches/py3k/Doc/library/pickletools.rst	(original)
+++ python/branches/py3k/Doc/library/pickletools.rst	Sat Jul 17 17:51:21 2010
@@ -13,14 +13,18 @@
 :mod:`pickletools` module relevant.
 
 
-.. function:: dis(pickle, out=None, memo=None, indentlevel=4)
+.. function:: dis(pickle, out=None, memo=None, indentlevel=4, annotate=0)
 
-   Outputs a symbolic disassembly of the pickle to the file-like object *out*,
-   defaulting to ``sys.stdout``.  *pickle* can be a string or a file-like object.
-   *memo* can be a Python dictionary that will be used as the pickle's memo; it can
-   be used to perform disassemblies across multiple pickles created by the same
-   pickler. Successive levels, indicated by ``MARK`` opcodes in the stream, are
-   indented by *indentlevel* spaces.
+   Outputs a symbolic disassembly of the pickle to the file-like
+   object *out*, defaulting to ``sys.stdout``.  *pickle* can be a
+   string or a file-like object.  *memo* can be a Python dictionary
+   that will be used as the pickle's memo; it can be used to perform
+   disassemblies across multiple pickles created by the same
+   pickler. Successive levels, indicated by ``MARK`` opcodes in the
+   stream, are indented by *indentlevel* spaces.  If a nonzero value
+   is given to *annotate*, each opcode in the output is annotated with
+   a short description.  The value of *annotate* is used as a hint for
+   the column where annotation should start.
 
 .. function:: genops(pickle)
 

Modified: python/branches/py3k/Lib/pickletools.py
==============================================================================
--- python/branches/py3k/Lib/pickletools.py	(original)
+++ python/branches/py3k/Lib/pickletools.py	Sat Jul 17 17:51:21 2010
@@ -1907,7 +1907,7 @@
 ##############################################################################
 # A symbolic pickle disassembler.
 
-def dis(pickle, out=None, memo=None, indentlevel=4):
+def dis(pickle, out=None, memo=None, indentlevel=4, annotate=0):
     """Produce a symbolic disassembly of a pickle.
 
     'pickle' is a file-like object, or string, containing a (at least one)
@@ -1923,9 +1923,15 @@
     to proceed across multiple pickles that were all created by the same
     pickler with the same memo.  Ordinarily you don't need to worry about this.
 
-    Optional arg indentlevel is the number of blanks by which to indent
+    Optional arg 'indentlevel' is the number of blanks by which to indent
     a new MARK level.  It defaults to 4.
 
+    Optional arg 'annotate' if nonzero instructs dis() to add short
+    description of the opcode on each line of disassembled output.
+    The value given to 'annotate' must be an integer and is used as a
+    hint for the column where annotation should start.  The default
+    value is 0, meaning no annotations.
+
     In addition to printing the disassembly, some sanity checks are made:
 
     + All embedded opcode arguments "make sense".
@@ -1953,6 +1959,7 @@
     markstack = []      # bytecode positions of MARK opcodes
     indentchunk = ' ' * indentlevel
     errormsg = None
+    annocol = annotate  # columnt hint for annotations
     for opcode, arg, pos in genops(pickle):
         if pos is not None:
             print("%5d:" % pos, end=' ', file=out)
@@ -2020,6 +2027,13 @@
                 line += ' ' + repr(arg)
             if markmsg:
                 line += ' ' + markmsg
+        if annotate:
+            line += ' ' * (annocol - len(line))
+            # make a mild effort to align annotations
+            annocol = len(line)
+            if annocol > 50:
+                annocol = annotate
+            line += ' ' + opcode.doc.split('\n', 1)[0]
         print(line, file=out)
 
         if errormsg:
@@ -2293,6 +2307,22 @@
    12: h    BINGET     1
    14: .    STOP
 highest protocol among opcodes = 2
+
+Try protocol 3 with annotations:
+
+>>> dis(pickle.dumps(T, 3), annotate=1)
+    0: \x80 PROTO      3 Protocol version indicator.
+    2: ]    EMPTY_LIST   Push an empty list.
+    3: q    BINPUT     0 Store the stack top into the memo.  The stack is not popped.
+    5: h    BINGET     0 Read an object from the memo and push it on the stack.
+    7: \x85 TUPLE1       Build a one-tuple out of the topmost item on the stack.
+    8: q    BINPUT     1 Store the stack top into the memo.  The stack is not popped.
+   10: a    APPEND       Append an object to a list.
+   11: 0    POP          Discard the top stack item, shrinking the stack by one item.
+   12: h    BINGET     1 Read an object from the memo and push it on the stack.
+   14: .    STOP         Stop the unpickling machine.
+highest protocol among opcodes = 2
+
 """
 
 _memo_test = r"""
@@ -2349,6 +2379,9 @@
         '-l', '--indentlevel', default=4, type=int,
         help='the number of blanks by which to indent a new MARK level')
     parser.add_argument(
+        '-a', '--annotate',  action='store_true',
+        help='annotate each line with a short opcode description')
+    parser.add_argument(
         '-p', '--preamble', default="==> {name} <==",
         help='if more than one pickle file is specified, print this before'
         ' each disassembly')
@@ -2362,14 +2395,15 @@
     if args.test:
         _test()
     else:
+        annotate = 30 if args.annotate else 0
         if not args.pickle_file:
             parser.print_help()
         elif len(args.pickle_file) == 1:
-            dis(args.pickle_file[0], args.output,
-                indentlevel=args.indentlevel)
+            dis(args.pickle_file[0], args.output, None,
+                args.indentlevel, annotate)
         else:
             memo = {} if args.memo else None
             for f in args.pickle_file:
                 preamble = args.preamble.format(name=f.name)
                 args.output.write(preamble + '\n')
-                dis(f, args.output, memo, args.indentlevel)
+                dis(f, args.output, memo, args.indentlevel, annotate)

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sat Jul 17 17:51:21 2010
@@ -473,6 +473,10 @@
 Library
 -------
 
+- Issue #9268: ``pickletools.dis()`` now has an optional *annotate*
+  argument which controls printing of opcode descriptions in ``dis()``
+  output.
+
 - Issue #1555570: email no longer inserts extra blank lines when a \r\n
   combo crosses an 8192 byte boundary.
 

From python-checkins at python.org  Sat Jul 17 17:59:37 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Sat, 17 Jul 2010 17:59:37 +0200 (CEST)
Subject: [Python-checkins] r82932 - python/branches/release31-maint
Message-ID: <20100717155937.CF377EE98D@mail.python.org>

Author: alexander.belopolsky
Date: Sat Jul 17 17:59:37 2010
New Revision: 82932

Log:
Blocked revisions 82931 via svnmerge

........
  r82931 | alexander.belopolsky | 2010-07-17 11:51:21 -0400 (Sat, 17 Jul 2010) | 2 lines
  
  Issue #9268: Add annotation option to pickletools.dis
........


Modified:
   python/branches/release31-maint/   (props changed)

From python-checkins at python.org  Sat Jul 17 18:01:06 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Sat, 17 Jul 2010 18:01:06 +0200 (CEST)
Subject: [Python-checkins] r82933 - python/branches/release27-maint
Message-ID: <20100717160106.BB596EE98D@mail.python.org>

Author: alexander.belopolsky
Date: Sat Jul 17 18:01:06 2010
New Revision: 82933

Log:
Blocked revisions 82931 via svnmerge

........
  r82931 | alexander.belopolsky | 2010-07-17 11:51:21 -0400 (Sat, 17 Jul 2010) | 2 lines
  
  Issue #9268: Add annotation option to pickletools.dis
........


Modified:
   python/branches/release27-maint/   (props changed)

From python-checkins at python.org  Sat Jul 17 22:39:24 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sat, 17 Jul 2010 22:39:24 +0200 (CEST)
Subject: [Python-checkins] r82934 - in python/branches/py3k: Makefile.pre.in
	Misc/NEWS
Message-ID: <20100717203924.340F5EE9BF@mail.python.org>

Author: benjamin.peterson
Date: Sat Jul 17 22:39:23 2010
New Revision: 82934

Log:
sharedinstall should depend on sharedmods #9280

Modified:
   python/branches/py3k/Makefile.pre.in
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Makefile.pre.in
==============================================================================
--- python/branches/py3k/Makefile.pre.in	(original)
+++ python/branches/py3k/Makefile.pre.in	Sat Jul 17 22:39:23 2010
@@ -1045,7 +1045,7 @@
 
 # Install the dynamically loadable modules
 # This goes into $(exec_prefix)
-sharedinstall:
+sharedinstall: sharedmods
 	$(RUNSHARED) ./$(BUILDPYTHON) -E $(srcdir)/setup.py install \
 	   	--prefix=$(prefix) \
 		--install-scripts=$(BINDIR) \

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sat Jul 17 22:39:23 2010
@@ -1554,6 +1554,8 @@
 Build
 -----
 
+- Issue #9280: Make sharedinstall depend on sharedmods.
+
 - Issue #9189: Make a user-specified CFLAGS, CPPFLAGS, or LDFLAGS
   setting override the configure and makefile defaults, without
   deleting options the user didn't intend to override.  Developers

From python-checkins at python.org  Sat Jul 17 22:41:16 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sat, 17 Jul 2010 22:41:16 +0200 (CEST)
Subject: [Python-checkins] r82935 - in python/branches/release27-maint:
	Makefile.pre.in Misc/NEWS
Message-ID: <20100717204116.2C46CEE9BF@mail.python.org>

Author: benjamin.peterson
Date: Sat Jul 17 22:41:16 2010
New Revision: 82935

Log:
Merged revisions 82934 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82934 | benjamin.peterson | 2010-07-17 15:39:23 -0500 (Sat, 17 Jul 2010) | 1 line
  
  sharedinstall should depend on sharedmods #9280
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Makefile.pre.in
   python/branches/release27-maint/Misc/NEWS

Modified: python/branches/release27-maint/Makefile.pre.in
==============================================================================
--- python/branches/release27-maint/Makefile.pre.in	(original)
+++ python/branches/release27-maint/Makefile.pre.in	Sat Jul 17 22:41:16 2010
@@ -1042,7 +1042,7 @@
 
 # Install the dynamically loadable modules
 # This goes into $(exec_prefix)
-sharedinstall:
+sharedinstall: sharedmods
 	$(RUNSHARED) ./$(BUILDPYTHON) -E $(srcdir)/setup.py install \
 	   	--prefix=$(prefix) \
 		--install-scripts=$(BINDIR) \

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Sat Jul 17 22:41:16 2010
@@ -40,6 +40,11 @@
 Extension Modules
 -----------------
 
+Build
+-----
+
+- Issue #9280: Make sharedinstall depend on sharedmods.
+
 Tests
 -----
 

From python-checkins at python.org  Sat Jul 17 22:44:50 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sat, 17 Jul 2010 22:44:50 +0200 (CEST)
Subject: [Python-checkins] r82936 - in python/branches/release31-maint:
	Makefile.pre.in Misc/NEWS
Message-ID: <20100717204450.E38D9EE9BF@mail.python.org>

Author: benjamin.peterson
Date: Sat Jul 17 22:44:50 2010
New Revision: 82936

Log:
Merged revisions 82934 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82934 | benjamin.peterson | 2010-07-17 15:39:23 -0500 (Sat, 17 Jul 2010) | 1 line
  
  sharedinstall should depend on sharedmods #9280
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Makefile.pre.in
   python/branches/release31-maint/Misc/NEWS

Modified: python/branches/release31-maint/Makefile.pre.in
==============================================================================
--- python/branches/release31-maint/Makefile.pre.in	(original)
+++ python/branches/release31-maint/Makefile.pre.in	Sat Jul 17 22:44:50 2010
@@ -1018,7 +1018,7 @@
 
 # Install the dynamically loadable modules
 # This goes into $(exec_prefix)
-sharedinstall:
+sharedinstall: sharedmods
 	$(RUNSHARED) ./$(BUILDPYTHON) -E $(srcdir)/setup.py install \
 	   	--prefix=$(prefix) \
 		--install-scripts=$(BINDIR) \

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Sat Jul 17 22:44:50 2010
@@ -829,6 +829,8 @@
 Build
 -----
 
+- Issue #9280: Make sharedinstall depend on sharedmods.
+
 - Issue #8854: Fix finding Visual Studio 2008 on Windows x64.
 
 - Issue #3920, #7903: Define _BSD_SOURCE on OpenBSD 4.4 through 4.9.

From python-checkins at python.org  Sun Jul 18 00:50:45 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Sun, 18 Jul 2010 00:50:45 +0200 (CEST)
Subject: [Python-checkins] r82937 - in python/branches/py3k: Lib/pickle.py
	Lib/test/pickletester.py Misc/NEWS Modules/_pickle.c
Message-ID: <20100717225045.BA1FED36E@mail.python.org>

Author: alexander.belopolsky
Date: Sun Jul 18 00:50:45 2010
New Revision: 82937

Log:
Issue #5180: Fixed a bug that prevented loading 2.x pickles in 3.x
python when they contain instances of old-style classes.


Modified:
   python/branches/py3k/Lib/pickle.py
   python/branches/py3k/Lib/test/pickletester.py
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/Modules/_pickle.c

Modified: python/branches/py3k/Lib/pickle.py
==============================================================================
--- python/branches/py3k/Lib/pickle.py	(original)
+++ python/branches/py3k/Lib/pickle.py	Sun Jul 18 00:50:45 2010
@@ -1034,19 +1034,15 @@
     def _instantiate(self, klass, k):
         args = tuple(self.stack[k+1:])
         del self.stack[k:]
-        instantiated = False
-        if (not args and
-                isinstance(klass, type) and
-                not hasattr(klass, "__getinitargs__")):
-            value = _EmptyClass()
-            value.__class__ = klass
-            instantiated = True
-        if not instantiated:
+        if (args or not isinstance(klass, type) or
+            hasattr(klass, "__getinitargs__")):
             try:
                 value = klass(*args)
             except TypeError as err:
                 raise TypeError("in constructor for %s: %s" %
                                 (klass.__name__, str(err)), sys.exc_info()[2])
+        else:
+            value = klass.__new__(klass)
         self.append(value)
 
     def load_inst(self):
@@ -1239,14 +1235,7 @@
         raise _Stop(value)
     dispatch[STOP[0]] = load_stop
 
-# Helper class for load_inst/load_obj
-
-class _EmptyClass:
-    pass
-
-# Encode/decode longs in linear time.
-
-import binascii as _binascii
+# Encode/decode longs.
 
 def encode_long(x):
     r"""Encode a long to a two's complement little-endian binary string.

Modified: python/branches/py3k/Lib/test/pickletester.py
==============================================================================
--- python/branches/py3k/Lib/test/pickletester.py	(original)
+++ python/branches/py3k/Lib/test/pickletester.py	Sun Jul 18 00:50:45 2010
@@ -65,9 +65,21 @@
     def __eq__(self, other):
         return self.__dict__ == other.__dict__
 
+class D(C):
+    def __init__(self, arg):
+        pass
+
+class E(C):
+    def __getinitargs__(self):
+        return ()
+
 import __main__
 __main__.C = C
 C.__module__ = "__main__"
+__main__.D = D
+D.__module__ = "__main__"
+__main__.E = E
+E.__module__ = "__main__"
 
 class myint(int):
     def __init__(self, x):
@@ -425,6 +437,63 @@
     def test_load_from_data2(self):
         self.assertEqual(self._testdata, self.loads(DATA2))
 
+    def test_load_classic_instance(self):
+        # See issue5180.  Test loading 2.x pickles that
+        # contain an instance of old style class.
+        for X, args in [(C, ()), (D, ('x',)), (E, ())]:
+            xname = X.__name__.encode('ascii')
+            # Protocol 0 (text mode pickle):
+            """
+            0: (    MARK
+            1: i        INST       '__main__ X' (MARK at 0)
+            15: p    PUT        0
+            18: (    MARK
+            19: d        DICT       (MARK at 18)
+            20: p    PUT        1
+            23: b    BUILD
+            24: .    STOP
+            """
+            pickle0 = (b"(i__main__\n"
+                       b"X\n"
+                       b"p0\n"
+                       b"(dp1\nb.").replace(b'X', xname)
+            self.assertEqual(X(*args), self.loads(pickle0))
+
+            # Protocol 1 (binary mode pickle)
+            """
+            0: (    MARK
+            1: c        GLOBAL     '__main__ X'
+            15: q        BINPUT     0
+            17: o        OBJ        (MARK at 0)
+            18: q    BINPUT     1
+            20: }    EMPTY_DICT
+            21: q    BINPUT     2
+            23: b    BUILD
+            24: .    STOP
+            """
+            pickle1 = (b'(c__main__\n'
+                       b'X\n'
+                       b'q\x00oq\x01}q\x02b.').replace(b'X', xname)
+            self.assertEqual(X(*args), self.loads(pickle1))
+
+            # Protocol 2 (pickle2 = b'\x80\x02' + pickle1)
+            """
+            0: \x80 PROTO      2
+            2: (    MARK
+            3: c        GLOBAL     '__main__ X'
+            17: q        BINPUT     0
+            19: o        OBJ        (MARK at 2)
+            20: q    BINPUT     1
+            22: }    EMPTY_DICT
+            23: q    BINPUT     2
+            25: b    BUILD
+            26: .    STOP
+            """
+            pickle2 = (b'\x80\x02(c__main__\n'
+                       b'X\n'
+                       b'q\x00oq\x01}q\x02b.').replace(b'X', xname)
+            self.assertEqual(X(*args), self.loads(pickle2))
+
     # There are gratuitous differences between pickles produced by
     # pickle and cPickle, largely because cPickle starts PUT indices at
     # 1 and pickle starts them at 0.  See XXX comment in cPickle's put2() --

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sun Jul 18 00:50:45 2010
@@ -1429,6 +1429,9 @@
 Extension Modules
 -----------------
 
+- Issue #5180: Fixed a bug that prevented loading 2.x pickles in 3.x
+  python when they contain instances of old-style classes.
+
 - Issue #9165: Add new functions math.isfinite and cmath.isfinite, to
   accompany existing isinf and isnan functions.
 

Modified: python/branches/py3k/Modules/_pickle.c
==============================================================================
--- python/branches/py3k/Modules/_pickle.c	(original)
+++ python/branches/py3k/Modules/_pickle.c	Sun Jul 18 00:50:45 2010
@@ -3466,29 +3466,19 @@
 static PyObject *
 instantiate(PyObject *cls, PyObject *args)
 {
-    PyObject *r = NULL;
-
-    /* XXX: The pickle.py module does not create instances this way when the
-       args tuple is empty. See Unpickler._instantiate(). */
-    if ((r = PyObject_CallObject(cls, args)))
-        return r;
-
-    /* XXX: Is this still nescessary? */
-    {
-        PyObject *tp, *v, *tb, *tmp_value;
-
-        PyErr_Fetch(&tp, &v, &tb);
-        tmp_value = v;
-        /* NULL occurs when there was a KeyboardInterrupt */
-        if (tmp_value == NULL)
-            tmp_value = Py_None;
-        if ((r = PyTuple_Pack(3, tmp_value, cls, args))) {
-            Py_XDECREF(v);
-            v = r;
-        }
-        PyErr_Restore(tp, v, tb);
+    PyObject *result = NULL;
+    /* Caller must assure args are a tuple.  Normally, args come from
+       Pdata_poptuple which packs objects from the top of the stack
+       into a newly created tuple. */
+    assert(PyTuple_Check(args));
+    if (Py_SIZE(args) > 0 || !PyType_Check(cls) ||
+        PyObject_HasAttrString(cls, "__getinitargs__")) {
+        result = PyObject_CallObject(cls, args);
+    }
+    else {
+        result = PyObject_CallMethod(cls, "__new__", "O", cls);
     }
-    return NULL;
+    return result;
 }
 
 static int
@@ -3547,7 +3537,7 @@
         if (len < 2)
             return bad_readline();
         class_name = PyUnicode_DecodeASCII(s, len - 1, "strict");
-        if (class_name == NULL) {
+        if (class_name != NULL) {
             cls = find_class(self, module_name, class_name);
             Py_DECREF(class_name);
         }
@@ -4276,7 +4266,7 @@
         return -1;
     PDATA_POP(self->stack, callable);
     if (callable) {
-        obj = instantiate(callable, argtup);
+        obj = PyObject_CallObject(callable, argtup);
         Py_DECREF(callable);
     }
     Py_DECREF(argtup);

From python-checkins at python.org  Sun Jul 18 01:01:39 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Sun, 18 Jul 2010 01:01:39 +0200 (CEST)
Subject: [Python-checkins] r82938 - in python/branches/release31-maint:
	Lib/pickle.py Lib/test/pickletester.py Misc/NEWS Modules/_pickle.c
Message-ID: <20100717230139.E0777F1F2@mail.python.org>

Author: alexander.belopolsky
Date: Sun Jul 18 01:01:39 2010
New Revision: 82938

Log:
Merged revisions 82937 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82937 | alexander.belopolsky | 2010-07-17 18:50:45 -0400 (Sat, 17 Jul 2010) | 3 lines
  
  Issue #5180: Fixed a bug that prevented loading 2.x pickles in 3.x
  python when they contain instances of old-style classes.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/pickle.py
   python/branches/release31-maint/Lib/test/pickletester.py
   python/branches/release31-maint/Misc/NEWS
   python/branches/release31-maint/Modules/_pickle.c

Modified: python/branches/release31-maint/Lib/pickle.py
==============================================================================
--- python/branches/release31-maint/Lib/pickle.py	(original)
+++ python/branches/release31-maint/Lib/pickle.py	Sun Jul 18 01:01:39 2010
@@ -1034,19 +1034,15 @@
     def _instantiate(self, klass, k):
         args = tuple(self.stack[k+1:])
         del self.stack[k:]
-        instantiated = False
-        if (not args and
-                isinstance(klass, type) and
-                not hasattr(klass, "__getinitargs__")):
-            value = _EmptyClass()
-            value.__class__ = klass
-            instantiated = True
-        if not instantiated:
+        if (args or not isinstance(klass, type) or
+            hasattr(klass, "__getinitargs__")):
             try:
                 value = klass(*args)
             except TypeError as err:
                 raise TypeError("in constructor for %s: %s" %
                                 (klass.__name__, str(err)), sys.exc_info()[2])
+        else:
+            value = klass.__new__(klass)
         self.append(value)
 
     def load_inst(self):
@@ -1239,11 +1235,6 @@
         raise _Stop(value)
     dispatch[STOP[0]] = load_stop
 
-# Helper class for load_inst/load_obj
-
-class _EmptyClass:
-    pass
-
 # Encode/decode longs in linear time.
 
 import binascii as _binascii

Modified: python/branches/release31-maint/Lib/test/pickletester.py
==============================================================================
--- python/branches/release31-maint/Lib/test/pickletester.py	(original)
+++ python/branches/release31-maint/Lib/test/pickletester.py	Sun Jul 18 01:01:39 2010
@@ -65,9 +65,21 @@
     def __eq__(self, other):
         return self.__dict__ == other.__dict__
 
+class D(C):
+    def __init__(self, arg):
+        pass
+
+class E(C):
+    def __getinitargs__(self):
+        return ()
+
 import __main__
 __main__.C = C
 C.__module__ = "__main__"
+__main__.D = D
+D.__module__ = "__main__"
+__main__.E = E
+E.__module__ = "__main__"
 
 class myint(int):
     def __init__(self, x):
@@ -425,6 +437,63 @@
     def test_load_from_data2(self):
         self.assertEqual(self._testdata, self.loads(DATA2))
 
+    def test_load_classic_instance(self):
+        # See issue5180.  Test loading 2.x pickles that
+        # contain an instance of old style class.
+        for X, args in [(C, ()), (D, ('x',)), (E, ())]:
+            xname = X.__name__.encode('ascii')
+            # Protocol 0 (text mode pickle):
+            """
+            0: (    MARK
+            1: i        INST       '__main__ X' (MARK at 0)
+            15: p    PUT        0
+            18: (    MARK
+            19: d        DICT       (MARK at 18)
+            20: p    PUT        1
+            23: b    BUILD
+            24: .    STOP
+            """
+            pickle0 = (b"(i__main__\n"
+                       b"X\n"
+                       b"p0\n"
+                       b"(dp1\nb.").replace(b'X', xname)
+            self.assertEqual(X(*args), self.loads(pickle0))
+
+            # Protocol 1 (binary mode pickle)
+            """
+            0: (    MARK
+            1: c        GLOBAL     '__main__ X'
+            15: q        BINPUT     0
+            17: o        OBJ        (MARK at 0)
+            18: q    BINPUT     1
+            20: }    EMPTY_DICT
+            21: q    BINPUT     2
+            23: b    BUILD
+            24: .    STOP
+            """
+            pickle1 = (b'(c__main__\n'
+                       b'X\n'
+                       b'q\x00oq\x01}q\x02b.').replace(b'X', xname)
+            self.assertEqual(X(*args), self.loads(pickle1))
+
+            # Protocol 2 (pickle2 = b'\x80\x02' + pickle1)
+            """
+            0: \x80 PROTO      2
+            2: (    MARK
+            3: c        GLOBAL     '__main__ X'
+            17: q        BINPUT     0
+            19: o        OBJ        (MARK at 2)
+            20: q    BINPUT     1
+            22: }    EMPTY_DICT
+            23: q    BINPUT     2
+            25: b    BUILD
+            26: .    STOP
+            """
+            pickle2 = (b'\x80\x02(c__main__\n'
+                       b'X\n'
+                       b'q\x00oq\x01}q\x02b.').replace(b'X', xname)
+            self.assertEqual(X(*args), self.loads(pickle2))
+
     # There are gratuitous differences between pickles produced by
     # pickle and cPickle, largely because cPickle starts PUT indices at
     # 1 and pickle starts them at 0.  See XXX comment in cPickle's put2() --

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Sun Jul 18 01:01:39 2010
@@ -753,6 +753,9 @@
 Extension Modules
 -----------------
 
+- Issue #5180: Fixed a bug that prevented loading 2.x pickles in 3.x
+  python when they contain instances of old-style classes.
+
 - Issue #7670: sqlite3: Fixed crashes when operating on closed connections.
 
 - Stop providing crtassem.h symbols when compiling with Visual Studio 2010, as

Modified: python/branches/release31-maint/Modules/_pickle.c
==============================================================================
--- python/branches/release31-maint/Modules/_pickle.c	(original)
+++ python/branches/release31-maint/Modules/_pickle.c	Sun Jul 18 01:01:39 2010
@@ -3466,29 +3466,19 @@
 static PyObject *
 instantiate(PyObject *cls, PyObject *args)
 {
-    PyObject *r = NULL;
-
-    /* XXX: The pickle.py module does not create instances this way when the
-       args tuple is empty. See Unpickler._instantiate(). */
-    if ((r = PyObject_CallObject(cls, args)))
-        return r;
-
-    /* XXX: Is this still nescessary? */
-    {
-        PyObject *tp, *v, *tb, *tmp_value;
-
-        PyErr_Fetch(&tp, &v, &tb);
-        tmp_value = v;
-        /* NULL occurs when there was a KeyboardInterrupt */
-        if (tmp_value == NULL)
-            tmp_value = Py_None;
-        if ((r = PyTuple_Pack(3, tmp_value, cls, args))) {
-            Py_XDECREF(v);
-            v = r;
-        }
-        PyErr_Restore(tp, v, tb);
+    PyObject *result = NULL;
+    /* Caller must assure args are a tuple.  Normally, args come from
+       Pdata_poptuple which packs objects from the top of the stack
+       into a newly created tuple. */
+    assert(PyTuple_Check(args));
+    if (Py_SIZE(args) > 0 || !PyType_Check(cls) ||
+        PyObject_HasAttrString(cls, "__getinitargs__")) {
+        result = PyObject_CallObject(cls, args);
+    }
+    else {
+        result = PyObject_CallMethod(cls, "__new__", "O", cls);
     }
-    return NULL;
+    return result;
 }
 
 static int
@@ -3547,7 +3537,7 @@
         if (len < 2)
             return bad_readline();
         class_name = PyUnicode_DecodeASCII(s, len - 1, "strict");
-        if (class_name == NULL) {
+        if (class_name != NULL) {
             cls = find_class(self, module_name, class_name);
             Py_DECREF(class_name);
         }
@@ -4276,7 +4266,7 @@
         return -1;
     PDATA_POP(self->stack, callable);
     if (callable) {
-        obj = instantiate(callable, argtup);
+        obj = PyObject_CallObject(callable, argtup);
         Py_DECREF(callable);
     }
     Py_DECREF(argtup);

From python-checkins at python.org  Sun Jul 18 01:13:43 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Sun, 18 Jul 2010 01:13:43 +0200 (CEST)
Subject: [Python-checkins] r82939 - python/branches/release27-maint
Message-ID: <20100717231343.299A3EE9FC@mail.python.org>

Author: alexander.belopolsky
Date: Sun Jul 18 01:13:43 2010
New Revision: 82939

Log:
Blocked revisions 82937 via svnmerge

........
  r82937 | alexander.belopolsky | 2010-07-17 18:50:45 -0400 (Sat, 17 Jul 2010) | 3 lines
  
  Issue #5180: Fixed a bug that prevented loading 2.x pickles in 3.x
  python when they contain instances of old-style classes.
........


Modified:
   python/branches/release27-maint/   (props changed)

From solipsis at pitrou.net  Sun Jul 18 01:25:09 2010
From: solipsis at pitrou.net (solipsis at pitrou.net)
Date: Sun, 18 Jul 2010 01:25:09 +0200 (CEST)
Subject: [Python-checkins] Daily py3k reference leaks (r82934): sum=0
Message-ID: <20100717232510.063E81775D@ns6635.ovh.net>

py3k results for svn r82934 (hg cset 45b3a207b85e)
--------------------------------------------------



Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/antoine/py3k/refleaks/reflog6D8wR-', '-x']

From python-checkins at python.org  Sun Jul 18 04:27:10 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Sun, 18 Jul 2010 04:27:10 +0200 (CEST)
Subject: [Python-checkins] r82940 - in python/branches/release27-maint:
	Doc/library/urllib.rst Lib/test/test_urllib.py Lib/urllib.py
	Misc/NEWS
Message-ID: <20100718022710.A97A1EE992@mail.python.org>

Author: senthil.kumaran
Date: Sun Jul 18 04:27:10 2010
New Revision: 82940

Log:
Fixing Issue1712522 -  urllib.quote to support Unicode. The default
encoding='utf-8' and errors='strict'.



Modified:
   python/branches/release27-maint/Doc/library/urllib.rst
   python/branches/release27-maint/Lib/test/test_urllib.py
   python/branches/release27-maint/Lib/urllib.py
   python/branches/release27-maint/Misc/NEWS

Modified: python/branches/release27-maint/Doc/library/urllib.rst
==============================================================================
--- python/branches/release27-maint/Doc/library/urllib.rst	(original)
+++ python/branches/release27-maint/Doc/library/urllib.rst	Sun Jul 18 04:27:10 2010
@@ -202,24 +202,40 @@
 Utility functions
 -----------------
 
-.. function:: quote(string[, safe])
+.. function:: quote(string[, safe[, encoding[, errors]]])
 
    Replace special characters in *string* using the ``%xx`` escape. Letters,
    digits, and the characters ``'_.-'`` are never quoted. By default, this
-   function is intended for quoting the path section of the URL.The optional
+   function is intended for quoting the path section of the URL. The optional
    *safe* parameter specifies additional characters that should not be quoted
    --- its default value is ``'/'``.
 
+   *string* may be either a :class:`str` or a :class:`unicode`.
+
+   The optional *encoding* and *errors* parameters specify how to deal with
+   non-ASCII characters, as accepted by the :meth:`unicode.encode` method.
+   *encoding* defaults to ``'utf-8'``.
+   *errors* defaults to ``'strict'``, meaning unsupported characters raise a
+   :class:`UnicodeEncodeError`.
+   Non-Unicode strings are not encoded by default, and all bytes are allowed.
+
    Example: ``quote('/~connolly/')`` yields ``'/%7econnolly/'``.
 
+   Example: ``quote(u'/El Ni?o/')`` yields ``'/El%20Ni%C3%B1o/'``.
+
+   .. versionchanged:: 2.7.1
+      Added *encoding* and *errors* parameters.
 
-.. function:: quote_plus(string[, safe])
+
+.. function:: quote_plus(string[, safe[, encoding[, errors]]])
 
    Like :func:`quote`, but also replaces spaces by plus signs, as required for
    quoting HTML form values when building up a query string to go into a URL.
    Plus signs in the original string are escaped unless they are included in
    *safe*.  It also does not have *safe* default to ``'/'``.
 
+   Example: ``quote_plus(u'/El Ni?o/')`` yields ``'%2FEl+Ni%C3%B1o%2F'``.
+
 
 .. function:: unquote(string)
 

Modified: python/branches/release27-maint/Lib/test/test_urllib.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_urllib.py	(original)
+++ python/branches/release27-maint/Lib/test/test_urllib.py	Sun Jul 18 04:27:10 2010
@@ -355,6 +355,38 @@
         self.assertEqual(quote_by_default, result,
                          "using quote_plus(): %s != %s" %
                          (quote_by_default, result))
+        # Safe expressed as unicode rather than str
+        result = urllib.quote(quote_by_default, safe=u"<>")
+        self.assertEqual(quote_by_default, result,
+                         "using quote(): %r != %r" % (quote_by_default, result))
+        # "Safe" non-ASCII bytes should still work
+        # (Technically disallowed by the URI standard, but allowed for
+        # backwards compatibility with previous versions of Python)
+        result = urllib.quote(b"a\xfcb", safe=b"\xfc")
+        expect = b"a\xfcb"
+        self.assertEqual(expect, result,
+                         "using quote(): %r != %r" %
+                         (expect, result))
+        # Same as above, but with 'safe' as a unicode rather than str
+        # "Safe" non-ASCII unicode characters should have no effect
+        # (Since URIs are not allowed to have non-ASCII characters)
+        result = urllib.quote(b"a\xfcb", safe=u"\xfc")
+        expect = urllib.quote(b"a\xfcb", safe="")
+        self.assertEqual(expect, result,
+                         "using quote(): %r != %r" %
+                         (expect, result))
+        # Same as above, but quoting a unicode rather than a str
+        result = urllib.quote(u"a\xfcb", encoding="latin-1", safe=b"\xfc")
+        expect = b"a\xfcb"
+        self.assertEqual(expect, result,
+                         "using quote(): %r != %r" %
+                         (expect, result))
+        # Same as above, but with both the quoted value and 'safe' as unicode
+        result = urllib.quote(u"a\xfcb", encoding="latin-1", safe=u"\xfc")
+        expect = urllib.quote(u"a\xfcb", encoding="latin-1", safe="")
+        self.assertEqual(expect, result,
+                         "using quote(): %r != %r" %
+                         (expect, result))
 
     def test_default_quoting(self):
         # Make sure all characters that should be quoted are by default sans
@@ -406,6 +438,81 @@
                          'alpha%2Bbeta+gamma')
         self.assertEqual(urllib.quote_plus('alpha+beta gamma', '+'),
                          'alpha+beta+gamma')
+        # Test with unicode
+        self.assertEqual(urllib.quote_plus(u'alpha+beta gamma'),
+                         'alpha%2Bbeta+gamma')
+        # Test with safe unicode
+        self.assertEqual(urllib.quote_plus('alpha+beta gamma', u'+'),
+                         'alpha+beta+gamma')
+
+    def test_quote_bytes(self):
+        # Non-ASCII bytes should quote directly to percent-encoded values
+        given = b"\xa2\xd8ab\xff"
+        expect = "%A2%D8ab%FF"
+        result = urllib.quote(given)
+        self.assertEqual(expect, result,
+                         "using quote(): %r != %r" % (expect, result))
+        # Encoding argument should raise UnicodeDecodeError on bytes input
+        # with non-ASCII characters (just as with str.encode).
+        self.assertRaises(UnicodeDecodeError, urllib.quote, given,
+                            encoding="latin-1")
+
+    def test_quote_with_unicode(self):
+        # Characters in Latin-1 range, encoded by default in UTF-8
+        given = u"\xa2\xd8ab\xff"
+        expect = "%C2%A2%C3%98ab%C3%BF"
+        result = urllib.quote(given)
+        self.assertEqual(expect, result,
+                         "using quote(): %r != %r" % (expect, result))
+        # Characters in Latin-1 range, encoded by with None (default)
+        result = urllib.quote(given, encoding=None, errors=None)
+        self.assertEqual(expect, result,
+                         "using quote(): %r != %r" % (expect, result))
+        # Characters in Latin-1 range, encoded with Latin-1
+        given = u"\xa2\xd8ab\xff"
+        expect = "%A2%D8ab%FF"
+        result = urllib.quote(given, encoding="latin-1")
+        self.assertEqual(expect, result,
+                         "using quote(): %r != %r" % (expect, result))
+        # Characters in BMP, encoded by default in UTF-8
+        given = u"\u6f22\u5b57"              # "Kanji"
+        expect = "%E6%BC%A2%E5%AD%97"
+        result = urllib.quote(given)
+        self.assertEqual(expect, result,
+                         "using quote(): %r != %r" % (expect, result))
+        # Characters in BMP, encoded with Latin-1
+        given = u"\u6f22\u5b57"
+        self.assertRaises(UnicodeEncodeError, urllib.quote, given,
+                                    encoding="latin-1")
+        # Characters in BMP, encoded with Latin-1, with replace error handling
+        given = u"\u6f22\u5b57"
+        expect = "%3F%3F"                    # "??"
+        result = urllib.quote(given, encoding="latin-1",
+                                    errors="replace")
+        self.assertEqual(expect, result,
+                         "using quote(): %r != %r" % (expect, result))
+        # Characters in BMP, Latin-1, with xmlcharref error handling
+        given = u"\u6f22\u5b57"
+        expect = "%26%2328450%3B%26%2323383%3B"      # "&#28450;&#23383;"
+        result = urllib.quote(given, encoding="latin-1",
+                                    errors="xmlcharrefreplace")
+        self.assertEqual(expect, result,
+                         "using quote(): %r != %r" % (expect, result))
+
+    def test_quote_plus_with_unicode(self):
+        # Encoding (latin-1) test for quote_plus
+        given = u"\xa2\xd8 \xff"
+        expect = "%A2%D8+%FF"
+        result = urllib.quote_plus(given, encoding="latin-1")
+        self.assertEqual(expect, result,
+                         "using quote_plus(): %r != %r" % (expect, result))
+        # Errors test for quote_plus
+        given = u"ab\u6f22\u5b57 cd"
+        expect = "ab%3F%3F+cd"
+        result = urllib.quote_plus(given, encoding="latin-1",
+                                         errors="replace")
+        self.assertEqual(expect, result,
+                         "using quote_plus(): %r != %r" % (expect, result))
 
 class UnquotingTests(unittest.TestCase):
     """Tests for unquote() and unquote_plus()

Modified: python/branches/release27-maint/Lib/urllib.py
==============================================================================
--- python/branches/release27-maint/Lib/urllib.py	(original)
+++ python/branches/release27-maint/Lib/urllib.py	Sun Jul 18 04:27:10 2010
@@ -1193,7 +1193,7 @@
     _safe_map[c] = c if (i < 128 and c in always_safe) else '%{:02X}'.format(i)
 _safe_quoters = {}
 
-def quote(s, safe='/'):
+def quote(s, safe='/', encoding=None, errors=None):
     """quote('abc def') -> 'abc%20def'
 
     Each part of a URL, e.g. the path info, the query, etc., has a
@@ -1213,10 +1213,28 @@
     is reserved, but in typical usage the quote function is being
     called on a path where the existing slash characters are used as
     reserved characters.
+
+    string and safe may be either str or unicode objects.
+
+    The optional encoding and errors parameters specify how to deal with the
+    non-ASCII characters, as accepted by the unicode.encode method.
+    By default, encoding='utf-8' (characters are encoded with UTF-8), and
+    errors='strict' (unsupported characters raise a UnicodeEncodeError).
     """
     # fastpath
     if not s:
         return s
+
+    if encoding is not None or isinstance(s, unicode):
+        if encoding is None:
+            encoding = 'utf-8'
+        if errors is None:
+            errors = 'strict'
+        s = s.encode(encoding, errors)
+    if isinstance(safe, unicode):
+        # Normalize 'safe' by converting to str and removing non-ASCII chars
+        safe = safe.encode('ascii', 'ignore')
+
     cachekey = (safe, always_safe)
     try:
         (quoter, safe) = _safe_quoters[cachekey]
@@ -1230,12 +1248,12 @@
         return s
     return ''.join(map(quoter, s))
 
-def quote_plus(s, safe=''):
+def quote_plus(s, safe='', encoding=None, errors=None):
     """Quote the query fragment of a URL; replacing ' ' with '+'"""
     if ' ' in s:
-        s = quote(s, safe + ' ')
+        s = quote(s, safe + ' ', encoding, errors)
         return s.replace(' ', '+')
-    return quote(s, safe)
+    return quote(s, safe, encoding, errors)
 
 def urlencode(query, doseq=0):
     """Encode a sequence of two-element tuples or dictionary into a URL query string.

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Sun Jul 18 04:27:10 2010
@@ -18,6 +18,10 @@
 Library
 -------
 
+- Issue 1712522: urllib.quote supports Unicode String with encoding and errors
+  parameter. The encoding parameter defaults to utf-8 and errors to strict.
+  Patch by Matt Giuca.
+
 - Issue #7646: The fnmatch pattern cache no longer grows without bound.
 
 - Issue #9136: Fix 'dictionary changed size during iteration'

From python-checkins at python.org  Sun Jul 18 09:29:03 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun, 18 Jul 2010 09:29:03 +0200 (CEST)
Subject: [Python-checkins] r82941 - python/branches/py3k/Modules/_struct.c
Message-ID: <20100718072903.07530EE986@mail.python.org>

Author: mark.dickinson
Date: Sun Jul 18 09:29:02 2010
New Revision: 82941

Log:
Issue #9277: Struct module: standard bool packing was incorrect if
char is unsigned.  Thanks Stefan Krah for the patch.


Modified:
   python/branches/py3k/Modules/_struct.c

Modified: python/branches/py3k/Modules/_struct.c
==============================================================================
--- python/branches/py3k/Modules/_struct.c	(original)
+++ python/branches/py3k/Modules/_struct.c	Sun Jul 18 09:29:02 2010
@@ -867,11 +867,11 @@
 static int
 bp_bool(char *p, PyObject *v, const formatdef *f)
 {
-    char y;
+    int y;
     y = PyObject_IsTrue(v);
     if (y < 0)
         return -1;
-    memcpy(p, (char *)&y, sizeof y);
+    *p = (char)y;
     return 0;
 }
 

From python-checkins at python.org  Sun Jul 18 09:42:29 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun, 18 Jul 2010 09:42:29 +0200 (CEST)
Subject: [Python-checkins] r82942 - in python/branches/release31-maint:
	Modules/_struct.c
Message-ID: <20100718074229.6D0FAF5FF@mail.python.org>

Author: mark.dickinson
Date: Sun Jul 18 09:42:29 2010
New Revision: 82942

Log:
Merged revisions 82941 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82941 | mark.dickinson | 2010-07-18 08:29:02 +0100 (Sun, 18 Jul 2010) | 3 lines
  
  Issue #9277: Struct module: standard bool packing was incorrect if
  char is unsigned.  Thanks Stefan Krah for the patch.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Modules/_struct.c

Modified: python/branches/release31-maint/Modules/_struct.c
==============================================================================
--- python/branches/release31-maint/Modules/_struct.c	(original)
+++ python/branches/release31-maint/Modules/_struct.c	Sun Jul 18 09:42:29 2010
@@ -856,11 +856,11 @@
 static int
 bp_bool(char *p, PyObject *v, const formatdef *f)
 {
-    char y;
+    int y;
     y = PyObject_IsTrue(v);
     if (y < 0)
         return -1;
-    memcpy(p, (char *)&y, sizeof y);
+    *p = (char)y;
     return 0;
 }
 

From python-checkins at python.org  Sun Jul 18 09:48:20 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun, 18 Jul 2010 09:48:20 +0200 (CEST)
Subject: [Python-checkins] r82943 - python/branches/py3k/Misc/NEWS
Message-ID: <20100718074820.CE914C44E@mail.python.org>

Author: mark.dickinson
Date: Sun Jul 18 09:48:20 2010
New Revision: 82943

Log:
Misc/NEWS entry for r82941.

Modified:
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sun Jul 18 09:48:20 2010
@@ -1429,6 +1429,11 @@
 Extension Modules
 -----------------
 
+- Issue #9277: Fix bug in struct.pack for bools in standard mode
+  (e.g., struct.pack('>?')):  if conversion to bool raised an exception
+  then that exception wasn't properly propagated on machines with
+  unsigned char.
+
 - Issue #5180: Fixed a bug that prevented loading 2.x pickles in 3.x
   python when they contain instances of old-style classes.
 

From python-checkins at python.org  Sun Jul 18 09:51:41 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun, 18 Jul 2010 09:51:41 +0200 (CEST)
Subject: [Python-checkins] r82944 - in python/branches/release31-maint:
	Misc/NEWS
Message-ID: <20100718075141.C9615EEA10@mail.python.org>

Author: mark.dickinson
Date: Sun Jul 18 09:51:41 2010
New Revision: 82944

Log:
Merged revisions 82943 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82943 | mark.dickinson | 2010-07-18 08:48:20 +0100 (Sun, 18 Jul 2010) | 1 line
  
  Misc/NEWS entry for r82941.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Misc/NEWS

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Sun Jul 18 09:51:41 2010
@@ -753,6 +753,11 @@
 Extension Modules
 -----------------
 
+- Issue #9277: Fix bug in struct.pack for bools in standard mode
+  (e.g., struct.pack('>?')):  if conversion to bool raised an exception
+  then that exception wasn't properly propagated on machines with
+  unsigned char.
+
 - Issue #5180: Fixed a bug that prevented loading 2.x pickles in 3.x
   python when they contain instances of old-style classes.
 

From python-checkins at python.org  Sun Jul 18 09:55:55 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun, 18 Jul 2010 09:55:55 +0200 (CEST)
Subject: [Python-checkins] r82945 - in python/branches/release27-maint:
	Misc/NEWS Modules/_struct.c
Message-ID: <20100718075555.CF7DDD7CB@mail.python.org>

Author: mark.dickinson
Date: Sun Jul 18 09:55:55 2010
New Revision: 82945

Log:
Merged revisions 82941,82943 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82941 | mark.dickinson | 2010-07-18 08:29:02 +0100 (Sun, 18 Jul 2010) | 3 lines
  
  Issue #9277: Struct module: standard bool packing was incorrect if
  char is unsigned.  Thanks Stefan Krah for the patch.
........
  r82943 | mark.dickinson | 2010-07-18 08:48:20 +0100 (Sun, 18 Jul 2010) | 1 line
  
  Misc/NEWS entry for r82941.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Misc/NEWS
   python/branches/release27-maint/Modules/_struct.c

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Sun Jul 18 09:55:55 2010
@@ -44,6 +44,11 @@
 Extension Modules
 -----------------
 
+- Issue #9277: Fix bug in struct.pack for bools in standard mode
+  (e.g., struct.pack('>?')):  if conversion to bool raised an exception
+  then that exception wasn't properly propagated on machines where
+  char is unsigned.
+
 Build
 -----
 

Modified: python/branches/release27-maint/Modules/_struct.c
==============================================================================
--- python/branches/release27-maint/Modules/_struct.c	(original)
+++ python/branches/release27-maint/Modules/_struct.c	Sun Jul 18 09:55:55 2010
@@ -912,11 +912,11 @@
 static int
 bp_bool(char *p, PyObject *v, const formatdef *f)
 {
-    char y;
+    int y;
     y = PyObject_IsTrue(v);
     if (y < 0)
         return -1;
-    memcpy(p, (char *)&y, sizeof y);
+    *p = (char)y;
     return 0;
 }
 

From python-checkins at python.org  Sun Jul 18 10:01:37 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun, 18 Jul 2010 10:01:37 +0200 (CEST)
Subject: [Python-checkins] r82946 - in python/branches/release26-maint:
	Misc/NEWS Modules/_struct.c
Message-ID: <20100718080137.DED28D7CB@mail.python.org>

Author: mark.dickinson
Date: Sun Jul 18 10:01:37 2010
New Revision: 82946

Log:
Merged revisions 82945 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

................
  r82945 | mark.dickinson | 2010-07-18 08:55:55 +0100 (Sun, 18 Jul 2010) | 14 lines
  
  Merged revisions 82941,82943 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k
  
  ........
    r82941 | mark.dickinson | 2010-07-18 08:29:02 +0100 (Sun, 18 Jul 2010) | 3 lines
    
    Issue #9277: Struct module: standard bool packing was incorrect if
    char is unsigned.  Thanks Stefan Krah for the patch.
  ........
    r82943 | mark.dickinson | 2010-07-18 08:48:20 +0100 (Sun, 18 Jul 2010) | 1 line
    
    Misc/NEWS entry for r82941.
  ........
................


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Misc/NEWS
   python/branches/release26-maint/Modules/_struct.c

Modified: python/branches/release26-maint/Misc/NEWS
==============================================================================
--- python/branches/release26-maint/Misc/NEWS	(original)
+++ python/branches/release26-maint/Misc/NEWS	Sun Jul 18 10:01:37 2010
@@ -302,6 +302,11 @@
 Extension Modules
 -----------------
 
+- Issue #9277: Fix bug in struct.pack for bools in standard mode
+  (e.g., struct.pack('>?')):  if conversion to bool raised an exception
+  then that exception wasn't properly propagated on machines where
+  char is unsigned.
+
 - Issue #7384: If the system readline library is linked against
   ncurses, do not link the readline module against ncursesw. The
   additional restriction of linking the readline and curses modules

Modified: python/branches/release26-maint/Modules/_struct.c
==============================================================================
--- python/branches/release26-maint/Modules/_struct.c	(original)
+++ python/branches/release26-maint/Modules/_struct.c	Sun Jul 18 10:01:37 2010
@@ -1013,9 +1013,9 @@
 static int
 bp_bool(char *p, PyObject *v, const formatdef *f)
 {
-    char y;
+    int y;
     y = PyObject_IsTrue(v);
-    memcpy(p, (char *)&y, sizeof y);
+    *p = (char)y;
     return 0;
 }
 

From python-checkins at python.org  Sun Jul 18 10:03:10 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun, 18 Jul 2010 10:03:10 +0200 (CEST)
Subject: [Python-checkins] r82947 - python/branches/py3k/Misc/NEWS
Message-ID: <20100718080310.E3B26EE982@mail.python.org>

Author: mark.dickinson
Date: Sun Jul 18 10:03:10 2010
New Revision: 82947

Log:
Clarify Misc/NEWS entry.

Modified:
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sun Jul 18 10:03:10 2010
@@ -1431,8 +1431,8 @@
 
 - Issue #9277: Fix bug in struct.pack for bools in standard mode
   (e.g., struct.pack('>?')):  if conversion to bool raised an exception
-  then that exception wasn't properly propagated on machines with
-  unsigned char.
+  then that exception wasn't properly propagated on machines where
+  char is unsigned.
 
 - Issue #5180: Fixed a bug that prevented loading 2.x pickles in 3.x
   python when they contain instances of old-style classes.

From python-checkins at python.org  Sun Jul 18 10:50:32 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun, 18 Jul 2010 10:50:32 +0200 (CEST)
Subject: [Python-checkins] r82948 - in python/branches/release31-maint:
	Misc/NEWS
Message-ID: <20100718085032.E6D06EEA02@mail.python.org>

Author: mark.dickinson
Date: Sun Jul 18 10:50:32 2010
New Revision: 82948

Log:
Merged revisions 82947 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82947 | mark.dickinson | 2010-07-18 09:03:10 +0100 (Sun, 18 Jul 2010) | 1 line
  
  Clarify Misc/NEWS entry.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Misc/NEWS

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Sun Jul 18 10:50:32 2010
@@ -755,8 +755,8 @@
 
 - Issue #9277: Fix bug in struct.pack for bools in standard mode
   (e.g., struct.pack('>?')):  if conversion to bool raised an exception
-  then that exception wasn't properly propagated on machines with
-  unsigned char.
+  then that exception wasn't properly propagated on machines where
+  char is unsigned.
 
 - Issue #5180: Fixed a bug that prevented loading 2.x pickles in 3.x
   python when they contain instances of old-style classes.

From python-checkins at python.org  Sun Jul 18 12:11:03 2010
From: python-checkins at python.org (georg.brandl)
Date: Sun, 18 Jul 2010 12:11:03 +0200 (CEST)
Subject: [Python-checkins] r82949 - in python/branches/py3k:
	Doc/documenting/markup.rst Doc/library/pdb.rst
	Doc/tools/sphinxext/pyspecific.py Lib/pdb.doc Lib/pdb.py
Message-ID: <20100718101103.4A4D2EE986@mail.python.org>

Author: georg.brandl
Date: Sun Jul 18 12:11:03 2010
New Revision: 82949

Log:
#9279: remove the pdb.doc file, put its contents in pdb.__doc__.  Also sync this and the pdb docs, introduce a new directive for pdb commands and a role to link to them.

Removed:
   python/branches/py3k/Lib/pdb.doc
Modified:
   python/branches/py3k/Doc/documenting/markup.rst
   python/branches/py3k/Doc/library/pdb.rst
   python/branches/py3k/Doc/tools/sphinxext/pyspecific.py
   python/branches/py3k/Lib/pdb.py

Modified: python/branches/py3k/Doc/documenting/markup.rst
==============================================================================
--- python/branches/py3k/Doc/documenting/markup.rst	(original)
+++ python/branches/py3k/Doc/documenting/markup.rst	Sun Jul 18 12:11:03 2010
@@ -503,7 +503,7 @@
 .. describe:: option
 
    A command-line option to an executable program.  The leading hyphen(s) must
-   be included.
+   be included.  If a matching ``cmdoption`` directive exists, it is linked to.
 
 .. describe:: program
 

Modified: python/branches/py3k/Doc/library/pdb.rst
==============================================================================
--- python/branches/py3k/Doc/library/pdb.rst	(original)
+++ python/branches/py3k/Doc/library/pdb.rst	Sun Jul 18 12:11:03 2010
@@ -20,7 +20,7 @@
    module: bdb
    module: cmd
 
-The debugger is extensible --- it is actually defined as the class :class:`Pdb`.
+The debugger is extensible -- it is actually defined as the class :class:`Pdb`.
 This is currently undocumented but easily understood by reading the source.  The
 extension interface uses the modules :mod:`bdb` and :mod:`cmd`.
 
@@ -55,8 +55,8 @@
    import pdb; pdb.set_trace()
 
 at the location you want to break into the debugger.  You can then step through
-the code following this statement, and continue running without the debugger using
-the ``c`` command.
+the code following this statement, and continue running without the debugger
+using the :pdbcmd:`continue` command.
 
 The typical usage to inspect a crashed program is::
 
@@ -82,26 +82,28 @@
 .. function:: run(statement, globals=None, locals=None)
 
    Execute the *statement* (given as a string) under debugger control.  The
-   debugger prompt appears before any code is executed; you can set breakpoints and
-   type ``continue``, or you can step through the statement using ``step`` or
-   ``next`` (all these commands are explained below).  The optional *globals* and
-   *locals* arguments specify the environment in which the code is executed; by
-   default the dictionary of the module :mod:`__main__` is used.  (See the
-   explanation of the built-in :func:`exec` or :func:`eval` functions.)
+   debugger prompt appears before any code is executed; you can set breakpoints
+   and type :pdbcmd:`continue`, or you can step through the statement using
+   :pdbcmd:`step` or :pdbcmd:`next` (all these commands are explained below).
+   The optional *globals* and *locals* arguments specify the environment in
+   which the code is executed; by default the dictionary of the module
+   :mod:`__main__` is used.  (See the explanation of the built-in :func:`exec`
+   or :func:`eval` functions.)
 
 
 .. function:: runeval(expression, globals=None, locals=None)
 
    Evaluate the *expression* (given as a string) under debugger control.  When
-   :func:`runeval` returns, it returns the value of the expression.  Otherwise this
-   function is similar to :func:`run`.
+   :func:`runeval` returns, it returns the value of the expression.  Otherwise
+   this function is similar to :func:`run`.
 
 
 .. function:: runcall(function, *args, **kwds)
 
-   Call the *function* (a function or method object, not a string) with the given
-   arguments.  When :func:`runcall` returns, it returns whatever the function call
-   returned.  The debugger prompt appears as soon as the function is entered.
+   Call the *function* (a function or method object, not a string) with the
+   given arguments.  When :func:`runcall` returns, it returns whatever the
+   function call returned.  The debugger prompt appears as soon as the function
+   is entered.
 
 
 .. function:: set_trace()
@@ -160,16 +162,17 @@
 Debugger Commands
 =================
 
-The debugger recognizes the following commands.  Most commands can be
-abbreviated to one or two letters; e.g. ``h(elp)`` means that either ``h`` or
-``help`` can be used to enter the help command (but not ``he`` or ``hel``, nor
-``H`` or ``Help`` or ``HELP``).  Arguments to commands must be separated by
-whitespace (spaces or tabs).  Optional arguments are enclosed in square brackets
-(``[]``) in the command syntax; the square brackets must not be typed.
-Alternatives in the command syntax are separated by a vertical bar (``|``).
+The commands recognized by the debugger are listed below.  Most commands can be
+abbreviated to one or two letters as indicated; e.g. ``h(elp)`` means that
+either ``h`` or ``help`` can be used to enter the help command (but not ``he``
+or ``hel``, nor ``H`` or ``Help`` or ``HELP``).  Arguments to commands must be
+separated by whitespace (spaces or tabs).  Optional arguments are enclosed in
+square brackets (``[]``) in the command syntax; the square brackets must not be
+typed.  Alternatives in the command syntax are separated by a vertical bar
+(``|``).
 
 Entering a blank line repeats the last command entered.  Exception: if the last
-command was a ``list`` command, the next 11 lines are listed.
+command was a :pdbcmd:`list` command, the next 11 lines are listed.
 
 Commands that the debugger doesn't recognize are assumed to be Python statements
 and are executed in the context of the program being debugged.  Python
@@ -179,93 +182,107 @@
 statement, the exception name is printed but the debugger's state is not
 changed.
 
+The debugger supports :ref:`aliases <debugger-aliases>`.  Aliases can have
+parameters which allows one a certain level of adaptability to the context under
+examination.
+
 Multiple commands may be entered on a single line, separated by ``;;``.  (A
 single ``;`` is not used as it is the separator for multiple commands in a line
-that is passed to the Python parser.) No intelligence is applied to separating
+that is passed to the Python parser.)  No intelligence is applied to separating
 the commands; the input is split at the first ``;;`` pair, even if it is in the
 middle of a quoted string.
 
-The debugger supports aliases.  Aliases can have parameters which allows one a
-certain level of adaptability to the context under examination.
-
 .. index::
    pair: .pdbrc; file
    triple: debugger; configuration; file
 
-If a file :file:`.pdbrc`  exists in the user's home directory or in the current
+If a file :file:`.pdbrc` exists in the user's home directory or in the current
 directory, it is read in and executed as if it had been typed at the debugger
-prompt. This is particularly useful for aliases.  If both files exist, the one
+prompt.  This is particularly useful for aliases.  If both files exist, the one
 in the home directory is read first and aliases defined there can be overridden
 by the local file.
 
-h(elp) [*command*]
+
+.. pdbcommand:: h(elp) [command]
+
    Without argument, print the list of available commands.  With a *command* as
    argument, print help about that command.  ``help pdb`` displays the full
-   documentation file; if the environment variable :envvar:`PAGER` is defined, the
-   file is piped through that command instead.  Since the *command* argument must
-   be an identifier, ``help exec`` must be entered to get help on the ``!``
-   command.
+   documentation (the docstring of the :mod:`pdb` module); if the environment
+   variable :envvar:`PAGER` is defined, the string is piped through that command
+   instead.  Since the *command* argument must be an identifier, ``help exec``
+   must be entered to get help on the ``!`` command.
+
+.. pdbcommand:: w(here)
 
-w(here)
    Print a stack trace, with the most recent frame at the bottom.  An arrow
    indicates the current frame, which determines the context of most commands.
 
-d(own) [*count*]
+.. pdbcommand:: d(own) [count]
+
    Move the current frame *count* (default one) levels down in the stack trace
    (to a newer frame).
 
-u(p) [*count*]
-   Move the current frame *count* (default one) levels up in the stack trace
-   (to an older frame).
+.. pdbcommand:: u(p) [count]
+
+   Move the current frame *count* (default one) levels up in the stack trace (to
+   an older frame).
+
+.. pdbcommand:: b(reak) [([filename:]lineno | function) [, condition]]
 
-b(reak) [[*filename*:]\ *lineno* | *function*\ [, *condition*]]
    With a *lineno* argument, set a break there in the current file.  With a
-   *function* argument, set a break at the first executable statement within that
-   function. The line number may be prefixed with a filename and a colon, to
-   specify a breakpoint in another file (probably one that hasn't been loaded yet).
-   The file is searched on ``sys.path``. Note that each breakpoint is assigned a
-   number to which all the other breakpoint commands refer.
-
-   If a second argument is present, it is an expression which must evaluate to true
-   before the breakpoint is honored.
-
-   Without argument, list all breaks, including for each breakpoint, the number of
-   times that breakpoint has been hit, the current ignore count, and the associated
-   condition if any.
-
-tbreak [[*filename*:]\ *lineno* | *function*\ [, *condition*]]
-   Temporary breakpoint, which is removed automatically when it is first hit.  The
-   arguments are the same as break.
+   *function* argument, set a break at the first executable statement within
+   that function.  The line number may be prefixed with a filename and a colon,
+   to specify a breakpoint in another file (probably one that hasn't been loaded
+   yet).  The file is searched on :data:`sys.path`.  Note that each breakpoint
+   is assigned a number to which all the other breakpoint commands refer.
+
+   If a second argument is present, it is an expression which must evaluate to
+   true before the breakpoint is honored.
+
+   Without argument, list all breaks, including for each breakpoint, the number
+   of times that breakpoint has been hit, the current ignore count, and the
+   associated condition if any.
+
+.. pdbcommand:: tbreak [([filename:]lineno | function) [, condition]]
+
+   Temporary breakpoint, which is removed automatically when it is first hit.
+   The arguments are the same as for :pdbcmd:`break`.
+
+.. pdbcommand:: cl(ear) [bpnumber [bpnumber ...]]
 
-cl(ear) [*bpnumber* [*bpnumber ...*]]
    With a space separated list of breakpoint numbers, clear those breakpoints.
    Without argument, clear all breaks (but first ask confirmation).
 
-disable [*bpnumber* [*bpnumber ...*]]
-   Disables the breakpoints given as a space separated list of breakpoint numbers.
-   Disabling a breakpoint means it cannot cause the program to stop execution, but
-   unlike clearing a breakpoint, it remains in the list of breakpoints and can be
-   (re-)enabled.
-
-enable [*bpnumber* [*bpnumber ...*]]
-   Enables the breakpoints specified.
-
-ignore *bpnumber* [*count*]
-   Sets the ignore count for the given breakpoint number.  If count is omitted, the
-   ignore count is set to 0.  A breakpoint becomes active when the ignore count is
-   zero.  When non-zero, the count is decremented each time the breakpoint is
-   reached and the breakpoint is not disabled and any associated condition
-   evaluates to true.
-
-condition *bpnumber* [*condition*]
-   Condition is an expression which must evaluate to true before the breakpoint is
-   honored.  If condition is absent, any existing condition is removed; i.e., the
-   breakpoint is made unconditional.
+.. pdbcommand:: disable [bpnumber [bpnumber ...]]
+
+   Disable the breakpoints given as a space separated list of breakpoint
+   numbers.  Disabling a breakpoint means it cannot cause the program to stop
+   execution, but unlike clearing a breakpoint, it remains in the list of
+   breakpoints and can be (re-)enabled.
+
+.. pdbcommand:: enable [bpnumber [bpnumber ...]]
+
+   Enable the breakpoints specified.
+
+.. pdbcommand:: ignore bpnumber [count]
+
+   Set the ignore count for the given breakpoint number.  If count is omitted,
+   the ignore count is set to 0.  A breakpoint becomes active when the ignore
+   count is zero.  When non-zero, the count is decremented each time the
+   breakpoint is reached and the breakpoint is not disabled and any associated
+   condition evaluates to true.
+
+.. pdbcommand:: condition bpnumber [condition]
+
+   Set a new *condition* for the breakpoint, an expression which must evaluate
+   to true before the breakpoint is honored.  If *condition* is absent, any
+   existing condition is removed; i.e., the breakpoint is made unconditional.
+
+.. pdbcommand:: commands [bpnumber]
 
-commands [*bpnumber*]
    Specify a list of commands for breakpoint number *bpnumber*.  The commands
-   themselves appear on the following lines.  Type a line containing just 'end' to
-   terminate the commands. An example::
+   themselves appear on the following lines.  Type a line containing just
+   ``end`` to terminate the commands. An example::
 
       (Pdb) commands 1
       (com) print some_variable
@@ -273,12 +290,12 @@
       (Pdb)
 
    To remove all commands from a breakpoint, type commands and follow it
-   immediately with  end; that is, give no commands.
+   immediately with ``end``; that is, give no commands.
 
    With no *bpnumber* argument, commands refers to the last breakpoint set.
 
-   You can use breakpoint commands to start your program up again. Simply use the
-   continue command, or step, or any other command that resumes execution.
+   You can use breakpoint commands to start your program up again.  Simply use
+   the continue command, or step, or any other command that resumes execution.
 
    Specifying any command resuming execution (currently continue, step, next,
    return, jump, quit and their abbreviations) terminates the command list (as if
@@ -292,63 +309,81 @@
    that are to print a specific message and then continue.  If none of the other
    commands print anything, you see no sign that the breakpoint was reached.
 
-s(tep)
+.. pdbcommand:: s(tep)
+
    Execute the current line, stop at the first possible occasion (either in a
    function that is called or on the next line in the current function).
 
-n(ext)
-   Continue execution until the next line in the current function is reached or it
-   returns.  (The difference between ``next`` and ``step`` is that ``step`` stops
-   inside a called function, while ``next`` executes called functions at (nearly)
-   full speed, only stopping at the next line in the current function.)
+.. pdbcommand:: n(ext)
+
+   Continue execution until the next line in the current function is reached or
+   it returns.  (The difference between :pdbcmd:`next` and :pdbcmd:`step` is
+   that :pdbcmd:`step` stops inside a called function, while :pdbcmd:`next`
+   executes called functions at (nearly) full speed, only stopping at the next
+   line in the current function.)
+
+.. pdbcommand:: unt(il)
 
-unt(il)
    Continue execution until the line with the line number greater than the
    current one is reached or when returning from current frame.
 
-r(eturn)
+.. pdbcommand:: r(eturn)
+
    Continue execution until the current function returns.
 
-c(ont(inue))
+.. pdbcommand:: c(ont(inue))
+
    Continue execution, only stop when a breakpoint is encountered.
 
-j(ump) *lineno*
+.. pdbcommand:: j(ump) lineno
+
    Set the next line that will be executed.  Only available in the bottom-most
-   frame.  This lets you jump back and execute code again, or jump forward to skip
-   code that you don't want to run.
+   frame.  This lets you jump back and execute code again, or jump forward to
+   skip code that you don't want to run.
 
-   It should be noted that not all jumps are allowed --- for instance it is not
+   It should be noted that not all jumps are allowed -- for instance it is not
    possible to jump into the middle of a :keyword:`for` loop or out of a
    :keyword:`finally` clause.
 
-l(ist) [*first*\ [, *last*]]
-   List source code for the current file.  Without arguments, list 11 lines around
-   the current line or continue the previous listing.  With one argument, list 11
-   lines around at that line.  With two arguments, list the given range; if the
-   second argument is less than the first, it is interpreted as a count.
+.. pdbcommand:: l(ist) [first[, last]]
+
+   List source code for the current file.  Without arguments, list 11 lines
+   around the current line or continue the previous listing.  With one argument,
+   list 11 lines around at that line.  With two arguments, list the given range;
+   if the second argument is less than the first, it is interpreted as a count.
+
+.. pdbcommand:: a(rgs)
 
-a(rgs)
    Print the argument list of the current function.
 
-p(rint) *expression*
+.. pdbcommand:: p(rint) expression
+
    Evaluate the *expression* in the current context and print its value.
 
-pp *expression*
-   Like the ``p`` command, except the value of the expression is pretty-printed
-   using the :mod:`pprint` module.
-
-alias [*name* [command]]
-   Creates an alias called *name* that executes *command*.  The command must *not*
-   be enclosed in quotes.  Replaceable parameters can be indicated by ``%1``,
-   ``%2``, and so on, while ``%*`` is replaced by all the parameters.  If no
-   command is given, the current alias for *name* is shown. If no arguments are
-   given, all aliases are listed.
-
-   Aliases may be nested and can contain anything that can be legally typed at the
-   pdb prompt.  Note that internal pdb commands *can* be overridden by aliases.
-   Such a command is then hidden until the alias is removed.  Aliasing is
-   recursively applied to the first word of the command line; all other words in
-   the line are left alone.
+.. pdbcommand:: pp expression
+
+   Like the :pdbcmd:`print` command, except the value of the expression is
+   pretty-printed using the :mod:`pprint` module.
+
+.. pdbcommand:: whatis expression
+
+   Print the type of the *expression*.
+
+.. _debugger-aliases:
+
+.. pdbcommand:: alias [name [command]]
+
+   Create an alias called *name* that executes *command*.  The command must
+   *not* be enclosed in quotes.  Replaceable parameters can be indicated by
+   ``%1``, ``%2``, and so on, while ``%*`` is replaced by all the parameters.
+   If no command is given, the current alias for *name* is shown. If no
+   arguments are given, all aliases are listed.
+
+   Aliases may be nested and can contain anything that can be legally typed at
+   the pdb prompt.  Note that internal pdb commands *can* be overridden by
+   aliases.  Such a command is then hidden until the alias is removed.  Aliasing
+   is recursively applied to the first word of the command line; all other words
+   in the line are left alone.
 
    As an example, here are two useful aliases (especially when placed in the
    :file:`.pdbrc` file)::
@@ -358,25 +393,32 @@
       #Print instance variables in self
       alias ps pi self
 
-unalias *name*
-   Deletes the specified alias.
+.. pdbcommand:: unalias name
+
+   Delete the specified alias.
+
+.. pdbcommand:: ! statement
 
-[!]\ *statement*
    Execute the (one-line) *statement* in the context of the current stack frame.
    The exclamation point can be omitted unless the first word of the statement
-   resembles a debugger command. To set a global variable, you can prefix the
-   assignment command with a ``global`` command on the same line, e.g.::
+   resembles a debugger command.  To set a global variable, you can prefix the
+   assignment command with a :keyword:`global` statement on the same line,
+   e.g.::
 
       (Pdb) global list_options; list_options = ['-l']
       (Pdb)
 
-run [*args* ...]
-   Restart the debugged Python program. If an argument is supplied, it is split
-   with "shlex" and the result is used as the new sys.argv. History, breakpoints,
-   actions and debugger options are preserved. "restart" is an alias for "run".
+.. pdbcommand:: run [args ...]
+                restart [args ...]
+
+   Restart the debugged Python program.  If an argument is supplied, it is split
+   with :mod:`shlex` and the result is used as the new :data:`sys.argv`.
+   History, breakpoints, actions and debugger options are preserved.
+   :pdbcmd:`restart` is an alias for :pdbcmd:`run`.
+
+.. pdbcommand:: q(uit)
 
-q(uit)
-   Quit from the debugger. The program being executed is aborted.
+   Quit from the debugger.  The program being executed is aborted.
 
 
 .. rubric:: Footnotes

Modified: python/branches/py3k/Doc/tools/sphinxext/pyspecific.py
==============================================================================
--- python/branches/py3k/Doc/tools/sphinxext/pyspecific.py	(original)
+++ python/branches/py3k/Doc/tools/sphinxext/pyspecific.py	Sun Jul 18 12:11:03 2010
@@ -165,6 +165,28 @@
     return opname.strip()
 
 
+pdbcmd_sig_re = re.compile(r'([a-z()!]+)\s*(.*)')
+
+# later...
+#pdbargs_tokens_re = re.compile(r'''[a-zA-Z]+  |  # identifiers
+#                                   [.,:]+     |  # punctuation
+#                                   [\[\]()]   |  # parens
+#                                   \s+           # whitespace
+#                                   ''', re.X)
+
+def parse_pdb_command(env, sig, signode):
+    """Transform a pdb command signature into RST nodes."""
+    m = pdbcmd_sig_re.match(sig)
+    if m is None:
+        raise ValueError
+    name, args = m.groups()
+    fullname = name.replace('(', '').replace(')', '')
+    signode += addnodes.desc_name(name, name)
+    if args:
+        signode += addnodes.desc_addname(' '+args, ' '+args)
+    return fullname
+
+
 def setup(app):
     app.add_role('issue', issue_role)
     app.add_directive('impl-detail', ImplementationDetail)
@@ -172,4 +194,6 @@
     app.add_builder(suspicious.CheckSuspiciousMarkupBuilder)
     app.add_description_unit('opcode', 'opcode', '%s (opcode)',
                              parse_opcode_signature)
+    app.add_description_unit('pdbcommand', 'pdbcmd', '%s (pdb command)',
+                             parse_pdb_command)
     app.add_description_unit('2to3fixer', '2to3fixer', '%s (2to3 fixer)')

Deleted: python/branches/py3k/Lib/pdb.doc
==============================================================================
--- python/branches/py3k/Lib/pdb.doc	Sun Jul 18 12:11:03 2010
+++ (empty file)
@@ -1,202 +0,0 @@
-The Python Debugger Pdb
-=======================
-
-To use the debugger in its simplest form:
-
-        >>> import pdb
-        >>> pdb.run('<a statement>')
-
-The debugger's prompt is '(Pdb) '.  This will stop in the first
-function call in <a statement>.
-
-Alternatively, if a statement terminated with an unhandled exception,
-you can use pdb's post-mortem facility to inspect the contents of the
-traceback:
-
-        >>> <a statement>
-        <exception traceback>
-        >>> import pdb
-        >>> pdb.pm()
-
-The commands recognized by the debugger are listed in the next
-section.  Most can be abbreviated as indicated; e.g., h(elp) means
-that 'help' can be typed as 'h' or 'help' (but not as 'he' or 'hel',
-nor as 'H' or 'Help' or 'HELP').  Optional arguments are enclosed in
-square brackets.
-
-A blank line repeats the previous command literally, except for
-'list', where it lists the next 11 lines.
-
-Commands that the debugger doesn't recognize are assumed to be Python
-statements and are executed in the context of the program being
-debugged.  Python statements can also be prefixed with an exclamation
-point ('!').  This is a powerful way to inspect the program being
-debugged; it is even possible to change variables.  When an exception
-occurs in such a statement, the exception name is printed but the
-debugger's state is not changed.
-
-The debugger supports aliases, which can save typing.  And aliases can
-have parameters (see the alias help entry) which allows one a certain
-level of adaptability to the context under examination.
-
-Multiple commands may be entered on a single line, separated by the
-pair ';;'.  No intelligence is applied to separating the commands; the
-input is split at the first ';;', even if it is in the middle of a
-quoted string.
-
-If a file ".pdbrc" exists in your home directory or in the current
-directory, it is read in and executed as if it had been typed at the
-debugger prompt.  This is particularly useful for aliases.  If both
-files exist, the one in the home directory is read first and aliases
-defined there can be overriden by the local file.
-
-Aside from aliases, the debugger is not directly programmable; but it
-is implemented as a class from which you can derive your own debugger
-class, which you can make as fancy as you like.
-
-
-Debugger commands
-=================
-
-h(elp)
-        Without argument, print the list of available commands.  With
-        a command name as argument, print help about that command
-        (this is currently not implemented).
-
-w(here)
-        Print a stack trace, with the most recent frame at the bottom.
-        An arrow indicates the "current frame", which determines the
-        context of most commands.
-
-d(own) [ count ]
-        Move the current frame count (default one) levels down in the
-        stack trace (to a newer frame).
-
-u(p) [ count ]
-        Move the current frame count (default one) levels up in the
-        stack trace (to an older frame).
-
-b(reak) [ ([filename:]lineno | function) [, condition] ]
-        With a filename:line number argument, set a break there.  If
-        filename is omitted, use the current file.  With a function
-        name, set a break at the first executable line of that
-        function.  Without argument, list all breaks.  Each breakpoint
-        is assigned a number to which all the other breakpoint
-        commands refer.
-
-        The condition argument, if present, is a string which must
-        evaluate to true in order for the breakpoint to be honored.
-
-tbreak [ ([filename:]lineno | function) [, condition] ]
-        Temporary breakpoint, which is removed automatically when it
-        is first hit.  The arguments are the same as break.
-
-cl(ear) [bpnumber [bpnumber ...] ]
-        With a space separated list of breakpoint numbers, clear those
-        breakpoints.  Without argument, clear all breaks (but first
-        ask confirmation).
-
-disable bpnumber [bpnumber ...]
-        Disables the breakpoints given as a space separated list of
-        breakpoint numbers.  Disabling a breakpoint means it cannot
-        cause the program to stop execution, but unlike clearing a
-        breakpoint, it remains in the list of breakpoints and can be
-        (re-)enabled.
-
-enable bpnumber [bpnumber ...]
-        Enables the breakpoints specified.
-
-ignore bpnumber count
-        Sets the ignore count for the given breakpoint number.  If
-        count is omitted, the ignore count is set to 0.  A breakpoint
-        becomes active when the ignore count is zero.  When non-zero,
-        the count is decremented each time the breakpoint is reached
-        and the breakpoint is not disabled and any associated
-        condition evaluates to true.
-
-condition bpnumber condition
-        condition is an expression which must evaluate to true before
-        the breakpoint is honored.  If condition is absent, any
-        existing condition is removed; i.e., the breakpoint is made
-        unconditional.
-
-s(tep)
-        Execute the current line, stop at the first possible occasion
-        (either in a function that is called or in the current function).
-
-n(ext)
-        Continue execution until the next line in the current function
-        is reached or it returns.
-
-unt(il)
-        Continue execution until the line with a number greater than the 
-        current one is reached or until the current frame returns.
-
-r(eturn)
-        Continue execution until the current function returns.
-
-run [args...]
-        Restart the debugged python program. If a string is supplied it is
-        splitted with "shlex", and the result is used as the new sys.argv.
-        History, breakpoints, actions and debugger options are preserved.
-        "restart" is an alias for "run".
-
-c(ont(inue))
-        Continue execution, only stop when a breakpoint is encountered.
-
-l(ist) [first [,last]]
-        List source code for the current file.
-        Without arguments, list 11 lines around the current line
-        or continue the previous listing.
-        With one argument, list 11 lines starting at that line.
-        With two arguments, list the given range;
-        if the second argument is less than the first, it is a count.
-
-a(rgs)
-        Print the argument list of the current function.
-
-p expression
-        Print the value of the expression.
-
-(!) statement
-        Execute the (one-line) statement in the context of the current
-        stack frame.  The exclamation point can be omitted unless the
-        first word of the statement resembles a debugger command.  To
-        assign to a global variable you must always prefix the command
-        with a 'global' command, e.g.:
-        (Pdb) global list_options; list_options = ['-l']
-        (Pdb)
-
-
-whatis arg
-         Prints the type of the argument.
-
-alias [name [command]]
-        Creates an alias called 'name' that executes 'command'.  The
-        command must *not* be enclosed in quotes.  Replaceable
-        parameters can be indicated by %1, %2, and so on, while %* is
-        replaced by all the parameters.  If no command is given, the
-        current alias for name is shown. If no name is given, all
-        aliases are listed.
-
-        Aliases may be nested and can contain anything that can be
-        legally typed at the pdb prompt.  Note!  You *can* override
-        internal pdb commands with aliases!  Those internal commands
-        are then hidden until the alias is removed.  Aliasing is
-        recursively applied to the first word of the command line; all
-        other words in the line are left alone.
-
-        As an example, here are two useful aliases (especially when
-        placed in the .pdbrc file):
-
-        #Print instance variables (usage "pi classInst")
-        alias pi for k in %1.__dict__.keys(): print "%1.",k,"=",%1.__dict__[k]
-        #Print instance variables in self
-        alias ps pi self
-                
-unalias name
-        Deletes the specified alias.
-
-q(uit)
-        Quit from the debugger.
-        The program being executed is aborted.

Modified: python/branches/py3k/Lib/pdb.py
==============================================================================
--- python/branches/py3k/Lib/pdb.py	(original)
+++ python/branches/py3k/Lib/pdb.py	Sun Jul 18 12:11:03 2010
@@ -1,8 +1,217 @@
 #! /usr/bin/env python3
 
-"""A Python debugger."""
-
-# (See pdb.doc for documentation.)
+"""
+The Python Debugger Pdb
+=======================
+
+To use the debugger in its simplest form:
+
+        >>> import pdb
+        >>> pdb.run('<a statement>')
+
+The debugger's prompt is '(Pdb) '.  This will stop in the first
+function call in <a statement>.
+
+Alternatively, if a statement terminated with an unhandled exception,
+you can use pdb's post-mortem facility to inspect the contents of the
+traceback:
+
+        >>> <a statement>
+        <exception traceback>
+        >>> import pdb
+        >>> pdb.pm()
+
+The commands recognized by the debugger are listed in the next
+section.  Most can be abbreviated as indicated; e.g., h(elp) means
+that 'help' can be typed as 'h' or 'help' (but not as 'he' or 'hel',
+nor as 'H' or 'Help' or 'HELP').  Optional arguments are enclosed in
+square brackets.  Alternatives in the command syntax are separated
+by a vertical bar (|).
+
+A blank line repeats the previous command literally, except for
+'list', where it lists the next 11 lines.
+
+Commands that the debugger doesn't recognize are assumed to be Python
+statements and are executed in the context of the program being
+debugged.  Python statements can also be prefixed with an exclamation
+point ('!').  This is a powerful way to inspect the program being
+debugged; it is even possible to change variables or call functions.
+When an exception occurs in such a statement, the exception name is
+printed but the debugger's state is not changed.
+
+The debugger supports aliases, which can save typing.  And aliases can
+have parameters (see the alias help entry) which allows one a certain
+level of adaptability to the context under examination.
+
+Multiple commands may be entered on a single line, separated by the
+pair ';;'.  No intelligence is applied to separating the commands; the
+input is split at the first ';;', even if it is in the middle of a
+quoted string.
+
+If a file ".pdbrc" exists in your home directory or in the current
+directory, it is read in and executed as if it had been typed at the
+debugger prompt.  This is particularly useful for aliases.  If both
+files exist, the one in the home directory is read first and aliases
+defined there can be overriden by the local file.
+
+Aside from aliases, the debugger is not directly programmable; but it
+is implemented as a class from which you can derive your own debugger
+class, which you can make as fancy as you like.
+
+
+Debugger commands
+=================
+
+h(elp)
+        Without argument, print the list of available commands.  With
+        a command name as argument, print help about that command.
+
+w(here)
+        Print a stack trace, with the most recent frame at the bottom.
+        An arrow indicates the "current frame", which determines the
+        context of most commands.
+
+d(own) [ count ]
+        Move the current frame count (default one) levels down in the
+        stack trace (to a newer frame).
+
+u(p) [ count ]
+        Move the current frame count (default one) levels up in the
+        stack trace (to an older frame).
+
+b(reak) [ ([filename:]lineno | function) [, condition] ]
+        With a filename:lineno argument, set a break there.  If
+        filename is omitted, use the current file.  With a function
+        name, set a break at the first executable line of that
+        function.  Without argument, list all breaks.  Each breakpoint
+        is assigned a number to which all the other breakpoint
+        commands refer.
+
+        The condition argument, if present, is a string which must
+        evaluate to true in order for the breakpoint to be honored.
+
+tbreak [ ([filename:]lineno | function) [, condition] ]
+        Temporary breakpoint, which is removed automatically when it
+        is first hit.  The arguments are the same as for break.
+
+cl(ear) [bpnumber [bpnumber ...] ]
+        With a space separated list of breakpoint numbers, clear those
+        breakpoints.  Without argument, clear all breaks (but first
+        ask confirmation).
+
+disable bpnumber [bpnumber ...]
+        Disable the breakpoints given as a space separated list of
+        breakpoint numbers.  Disabling a breakpoint means it cannot
+        cause the program to stop execution, but unlike clearing a
+        breakpoint, it remains in the list of breakpoints and can be
+        (re-)enabled.
+
+enable bpnumber [bpnumber ...]
+        Enable the breakpoints specified.
+
+ignore bpnumber [count]
+        Set the ignore count for the given breakpoint number.  If
+        count is omitted, the ignore count is set to 0.  A breakpoint
+        becomes active when the ignore count is zero.  When non-zero,
+        the count is decremented each time the breakpoint is reached
+        and the breakpoint is not disabled and any associated
+        condition evaluates to true.
+
+condition bpnumber [condition]
+        Set a new condition for the breakpoint, an expression which
+        must evaluate to true before the breakpoint is honored.  If
+        condition is absent, any existing condition is removed; i.e.,
+        the breakpoint is made unconditional.
+
+commands [bpnumber]
+        Specify a list of commands for the breakpoint.  Type a line
+        containing just 'end' to terminate the commands.  The commands
+        are executed when the breakpoint is hit.
+
+        With no breakpoint number argument, refers to the last
+        breakpoint set.
+
+s(tep)
+        Execute the current line, stop at the first possible occasion
+        (either in a function that is called or in the current
+        function).
+
+n(ext)
+        Continue execution until the next line in the current function
+        is reached or it returns.
+
+unt(il)
+        Continue execution until the line with a number greater than
+        the current one is reached or until the current frame returns.
+
+r(eturn)
+        Continue execution until the current function returns.
+
+run [args...]
+        Restart the debugged python program. If a string is supplied
+        it is splitted with "shlex", and the result is used as the new
+        sys.argv.  History, breakpoints, actions and debugger options
+        are preserved.  "restart" is an alias for "run".
+
+c(ont(inue))
+        Continue execution, only stop when a breakpoint is encountered.
+
+l(ist) [first [,last]]
+        List source code for the current file.
+        Without arguments, list 11 lines around the current line
+        or continue the previous listing.
+        With one argument, list 11 lines starting at that line.
+        With two arguments, list the given range;
+        if the second argument is less than the first, it is a count.
+
+a(rgs)
+        Print the argument list of the current function.
+
+p expression
+        Print the value of the expression.
+
+(!) statement
+        Execute the (one-line) statement in the context of the current
+        stack frame.  The exclamation point can be omitted unless the
+        first word of the statement resembles a debugger command.  To
+        assign to a global variable you must always prefix the command
+        with a 'global' command, e.g.:
+        (Pdb) global list_options; list_options = ['-l']
+        (Pdb)
+
+
+whatis arg
+        Print the type of the argument.
+
+alias [name [command]]
+        Creates an alias called 'name' that executes 'command'.  The
+        command must *not* be enclosed in quotes.  Replaceable
+        parameters can be indicated by %1, %2, and so on, while %* is
+        replaced by all the parameters.  If no command is given, the
+        current alias for name is shown. If no name is given, all
+        aliases are listed.
+
+        Aliases may be nested and can contain anything that can be
+        legally typed at the pdb prompt.  Note!  You *can* override
+        internal pdb commands with aliases!  Those internal commands
+        are then hidden until the alias is removed.  Aliasing is
+        recursively applied to the first word of the command line; all
+        other words in the line are left alone.
+
+        As an example, here are two useful aliases (especially when
+        placed in the .pdbrc file):
+
+        # Print instance variables (usage "pi classInst")
+        alias pi for k in %1.__dict__.keys(): print "%1.",k,"=",%1.__dict__[k]
+        # Print instance variables in self
+        alias ps pi self
+
+unalias name
+        Delete the specified alias.
+
+q(uit)
+        Quit from the debugger. The program being executed is aborted.
+"""
 
 import sys
 import linecache
@@ -905,7 +1114,7 @@
                                                      prompt_prefix), file=self.stdout)
 
 
-    # Help methods (derived from pdb.doc)
+    # Help methods (derived from docstring)
 
     def help_help(self):
         self.help_h()
@@ -1281,15 +1490,8 @@
 
 # print help
 def help():
-    for dirname in sys.path:
-        fullname = os.path.join(dirname, 'pdb.doc')
-        if os.path.exists(fullname):
-            sts = os.system('${PAGER-more} '+fullname)
-            if sts: print('*** Pager exit status:', sts)
-            break
-    else:
-        print('Sorry, can\'t find the help file "pdb.doc"', end=' ')
-        print('along the Python search path')
+    import pydoc
+    pydoc.pager(__doc__)
 
 def main():
     if not sys.argv[1:] or sys.argv[1] in ("--help", "-h"):

From ncoghlan at gmail.com  Sun Jul 18 13:45:55 2010
From: ncoghlan at gmail.com (Nick Coghlan)
Date: Sun, 18 Jul 2010 21:45:55 +1000
Subject: [Python-checkins] r82907 -
	python/branches/import_unicode/Modules/zipimport.c
In-Reply-To: <20100714203824.8B936EEA30@mail.python.org>
References: <20100714203824.8B936EEA30@mail.python.org>
Message-ID: <AANLkTilyKbkKEG5IBe0ewvBYT6lJM-hlIPz6GNk1yqvu@mail.gmail.com>

On Thu, Jul 15, 2010 at 6:38 AM, victor.stinner
<python-checkins at python.org> wrote:
> Author: victor.stinner
> Date: Wed Jul 14 22:38:24 2010
> New Revision: 82907
>
> Log:
> zipimport_get_source() uses PyUnicode_DecodeFSDefault()

Victor,

All these changes to zipimport.c on the branch are making me nervous
regarding backwards compatibility with legacy PEP 302 importers.
Zipimport needs to work as well as it did in 3.1 if all the other
changes are made to support Unicode paths in the import system but
zipimport itself remains unpatched.

Cheers,
Nick.

-- 
Nick Coghlan?? |?? ncoghlan at gmail.com?? |?? Brisbane, Australia

From solipsis at pitrou.net  Sun Jul 18 14:51:07 2010
From: solipsis at pitrou.net (Antoine Pitrou)
Date: Sun, 18 Jul 2010 14:51:07 +0200
Subject: [Python-checkins] r82940 - in python/branches/release27-maint:
 Doc/library/urllib.rst Lib/test/test_urllib.py Lib/urllib.py Misc/NEWS
References: <20100718022710.A97A1EE992@mail.python.org>
Message-ID: <20100718145107.10b69a91@pitrou.net>

On Sun, 18 Jul 2010 04:27:10 +0200 (CEST)
senthil.kumaran <python-checkins at python.org> wrote:
>  Utility functions
>  -----------------
>  
> -.. function:: quote(string[, safe])
> +.. function:: quote(string[, safe[, encoding[, errors]]])

New APIs shouldn't be added in bugfix releases. Is there a strong
reason to do this?

Looking at the issue (which in itself was quite old), you could as well
have fixed the robotparser module instead.

Thanks

Antoine.



From python-checkins at python.org  Sun Jul 18 15:12:03 2010
From: python-checkins at python.org (georg.brandl)
Date: Sun, 18 Jul 2010 15:12:03 +0200 (CEST)
Subject: [Python-checkins] r82950 - peps/trunk/pep-3148.txt
Message-ID: <20100718131203.786CEEE9B2@mail.python.org>

Author: georg.brandl
Date: Sun Jul 18 15:12:03 2010
New Revision: 82950

Log:
Update from Brian; make links work.

Modified:
   peps/trunk/pep-3148.txt

Modified: peps/trunk/pep-3148.txt
==============================================================================
--- peps/trunk/pep-3148.txt	(original)
+++ peps/trunk/pep-3148.txt	Sun Jul 18 15:12:03 2010
@@ -200,7 +200,7 @@
 
 ``result(timeout=None)``
 
-    Return the value returned by the call. If the call hasn't yet
+    Return the value returned by the call.  If the call hasn't yet
     completed then this method will wait up to *timeout* seconds.  If
     the call hasn't completed in *timeout* seconds then a
     `TimeoutError` will be raised.  If *timeout* is not specified or
@@ -372,12 +372,12 @@
     
             for future in futures.as_completed(future_to_url):
                 url = future_to_url[future]
-                if future.exception() is not None:
-                    print('%r generated an exception: %s' % (
-                              url, future.exception()))
-                else:
+                try:
                     print('%r page is %d bytes' % (
                               url, len(future.result())))
+                except Exception as e:
+                    print('%r generated an exception: %s' % (
+                              url, e))
 
     if __name__ == '__main__':
         main()
@@ -419,11 +419,22 @@
 Anh Hai Trinh proposed a simpler but more limited API concept [5]_ and
 the API has been discussed in some detail on stdlib-sig [6]_.
 
+The proposed design was discussed on the Python-Dev mailing list [7]_.
+Following those discussions, the following changes were made:
+
+* The `Executor` class was made into an abstract base class
+* The `Future.remove_done_callback` method was removed due to a lack
+  of convincing use cases
+* The `Future.add_done_callback` method was modified to allow the
+  same callable to be added many times
+* The `Future` class's mutation methods were better documented to
+  indicate that they are private to the `Executor` that created them
+
 ========================
 Reference Implementation
 ========================
 
-The reference implementation [7]_ contains a complete implementation
+The reference implementation [8]_ contains a complete implementation
 of the proposed design.  It has been tested on Linux and Mac OS X.
 
 ==========
@@ -432,31 +443,35 @@
 
 .. [1]
    `java.util.concurrent` package documentation
-   `http://java.sun.com/j2se/1.5.0/docs/api/java/util/concurrent/package-summary.html`
+   http://java.sun.com/j2se/1.5.0/docs/api/java/util/concurrent/package-summary.html
 
 .. [2]
    Python Cookbook recipe 84317, "Easy threading with Futures"
-   `http://code.activestate.com/recipes/84317/`
+   http://code.activestate.com/recipes/84317/
 
 .. [3]
    `Python-3000` thread, "mechanism for handling asynchronous concurrency"
-   `http://mail.python.org/pipermail/python-3000/2006-April/000960.html`
+   http://mail.python.org/pipermail/python-3000/2006-April/000960.html
 
 .. [4]
    `Python 3000` thread, "Futures in Python 3000 (was Re: mechanism for handling asynchronous concurrency)"
-   `http://mail.python.org/pipermail/python-3000/2006-April/000970.html`
+   http://mail.python.org/pipermail/python-3000/2006-April/000970.html
 
 .. [5]
    A discussion of `stream`, a similar concept proposed by Anh Hai Trinh
-   `http://www.mail-archive.com/stdlib-sig at python.org/msg00480.html`
+   http://www.mail-archive.com/stdlib-sig at python.org/msg00480.html
 
 .. [6]
    A discussion of the proposed API on stdlib-sig
-   `http://mail.python.org/pipermail/stdlib-sig/2009-November/000731.html`
+   http://mail.python.org/pipermail/stdlib-sig/2009-November/000731.html
 
 .. [7]
+   A discussion of the PEP on python-dev
+   http://mail.python.org/pipermail/python-dev/2010-March/098169.html
+
+.. [8]
    Reference `futures` implementation
-   `http://code.google.com/p/pythonfutures/source/browse/#svn/branches/feedback`
+   http://code.google.com/p/pythonfutures/source/browse/#svn/branches/feedback
 
 =========
 Copyright

From python-checkins at python.org  Sun Jul 18 15:43:32 2010
From: python-checkins at python.org (georg.brandl)
Date: Sun, 18 Jul 2010 15:43:32 +0200 (CEST)
Subject: [Python-checkins] r82951 -
	python/branches/py3k/Doc/library/contextlib.rst
Message-ID: <20100718134332.686E9F400@mail.python.org>

Author: georg.brandl
Date: Sun Jul 18 15:43:32 2010
New Revision: 82951

Log:
#9110: update to ContextDecorator doc.

Modified:
   python/branches/py3k/Doc/library/contextlib.rst

Modified: python/branches/py3k/Doc/library/contextlib.rst
==============================================================================
--- python/branches/py3k/Doc/library/contextlib.rst	(original)
+++ python/branches/py3k/Doc/library/contextlib.rst	Sun Jul 18 15:43:32 2010
@@ -57,6 +57,7 @@
    .. versionchanged:: 3.2
       Use of :class:`ContextDecorator`.
 
+
 .. function:: closing(thing)
 
    Return a context manager that closes *thing* upon completion of the block.  This
@@ -92,22 +93,25 @@
    ``__enter__`` and ``__exit__`` as normal. ``__exit__`` retains its optional
    exception handling even when used as a decorator.
 
-   Example::
+   ``ContextDecorator`` is used by :func:`contextmanager`, so you get this
+   functionality automatically.
+
+   Example of ``ContextDecorator``::
 
       from contextlib import ContextDecorator
 
       class mycontext(ContextDecorator):
-         def __enter__(self):
-            print('Starting')
-            return self
-
-         def __exit__(self, *exc):
-            print('Finishing')
-            return False
+          def __enter__(self):
+              print('Starting')
+              return self
+
+          def __exit__(self, *exc):
+              print('Finishing')
+              return False
 
       >>> @mycontext()
       ... def function():
-      ...    print('The bit in the middle')
+      ...     print('The bit in the middle')
       ...
       >>> function()
       Starting
@@ -115,23 +119,38 @@
       Finishing
 
       >>> with mycontext():
-      ...    print('The bit in the middle')
+      ...     print('The bit in the middle')
       ...
       Starting
       The bit in the middle
       Finishing
 
+   This change is just syntactic sugar for any construct of the following form::
+
+      def f():
+          with cm():
+              # Do stuff
+
+   ``ContextDecorator`` lets you instead write::
+
+      @cm()
+      def f():
+          # Do stuff
+
+   It makes it clear that the ``cm`` applies to the whole function, rather than
+   just a piece of it (and saving an indentation level is nice, too).
+
    Existing context managers that already have a base class can be extended by
    using ``ContextDecorator`` as a mixin class::
 
       from contextlib import ContextDecorator
 
       class mycontext(ContextBaseClass, ContextDecorator):
-         def __enter__(self):
-            return self
+          def __enter__(self):
+              return self
 
-         def __exit__(self, *exc):
-            return False
+          def __exit__(self, *exc):
+              return False
 
    .. versionadded:: 3.2
 

From python-checkins at python.org  Sun Jul 18 16:23:36 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sun, 18 Jul 2010 16:23:36 +0200 (CEST)
Subject: [Python-checkins] r82952 -
	python/branches/py3k/Doc/library/collections.rst
Message-ID: <20100718142336.DA83AEE9E5@mail.python.org>

Author: benjamin.peterson
Date: Sun Jul 18 16:23:36 2010
New Revision: 82952

Log:
use classmethod

Modified:
   python/branches/py3k/Doc/library/collections.rst

Modified: python/branches/py3k/Doc/library/collections.rst
==============================================================================
--- python/branches/py3k/Doc/library/collections.rst	(original)
+++ python/branches/py3k/Doc/library/collections.rst	Sun Jul 18 16:23:36 2010
@@ -763,7 +763,7 @@
 three additional methods and one attribute.  To prevent conflicts with
 field names, the method and attribute names start with an underscore.
 
-.. method:: somenamedtuple._make(iterable)
+.. classmethod:: somenamedtuple._make(iterable)
 
    Class method that makes a new instance from an existing sequence or iterable.
 

From python-checkins at python.org  Sun Jul 18 16:26:35 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sun, 18 Jul 2010 16:26:35 +0200 (CEST)
Subject: [Python-checkins] r82953 - in python/branches/release27-maint:
	Doc/library/collections.rst
Message-ID: <20100718142635.1484AF09D@mail.python.org>

Author: benjamin.peterson
Date: Sun Jul 18 16:26:34 2010
New Revision: 82953

Log:
Merged revisions 82952 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82952 | benjamin.peterson | 2010-07-18 09:23:36 -0500 (Sun, 18 Jul 2010) | 1 line
  
  use classmethod
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Doc/library/collections.rst

Modified: python/branches/release27-maint/Doc/library/collections.rst
==============================================================================
--- python/branches/release27-maint/Doc/library/collections.rst	(original)
+++ python/branches/release27-maint/Doc/library/collections.rst	Sun Jul 18 16:26:34 2010
@@ -787,7 +787,7 @@
 three additional methods and one attribute.  To prevent conflicts with
 field names, the method and attribute names start with an underscore.
 
-.. method:: somenamedtuple._make(iterable)
+.. classmethod:: somenamedtuple._make(iterable)
 
    Class method that makes a new instance from an existing sequence or iterable.
 

From python-checkins at python.org  Sun Jul 18 16:27:02 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sun, 18 Jul 2010 16:27:02 +0200 (CEST)
Subject: [Python-checkins] r82954 -
	python/branches/py3k/Doc/library/collections.rst
Message-ID: <20100718142702.40899F09D@mail.python.org>

Author: benjamin.peterson
Date: Sun Jul 18 16:27:02 2010
New Revision: 82954

Log:
nest method and attribute doc

Modified:
   python/branches/py3k/Doc/library/collections.rst

Modified: python/branches/py3k/Doc/library/collections.rst
==============================================================================
--- python/branches/py3k/Doc/library/collections.rst	(original)
+++ python/branches/py3k/Doc/library/collections.rst	Sun Jul 18 16:27:02 2010
@@ -551,7 +551,7 @@
    :class:`defaultdict` objects support the following method in addition to the
    standard :class:`dict` operations:
 
-   .. method:: defaultdict.__missing__(key)
+   .. method:: __missing__(key)
 
       If the :attr:`default_factory` attribute is ``None``, this raises a
       :exc:`KeyError` exception with the *key* as argument.
@@ -571,7 +571,7 @@
    :class:`defaultdict` objects support the following instance variable:
 
 
-   .. attribute:: defaultdict.default_factory
+   .. attribute:: default_factory
 
       This attribute is used by the :meth:`__missing__` method; it is
       initialized from the first argument to the constructor, if present, or to
@@ -891,11 +891,11 @@
 
    .. versionadded:: 3.1
 
-.. method:: OrderedDict.popitem(last=True)
+   .. method:: popitem(last=True)
 
-   The :meth:`popitem` method for ordered dictionaries returns and removes
-   a (key, value) pair.  The pairs are returned in LIFO order if *last* is
-   true or FIFO order if false.
+      The :meth:`popitem` method for ordered dictionaries returns and removes a
+      (key, value) pair.  The pairs are returned in LIFO order if *last* is true
+      or FIFO order if false.
 
 In addition to the usual mapping methods, ordered dictionaries also support
 reverse iteration using :func:`reversed`.
@@ -956,12 +956,13 @@
    initialized with its contents; note that a reference to *initialdata* will not
    be kept, allowing it be used for other purposes.
 
-In addition to supporting the methods and operations of mappings,
-:class:`UserDict` instances provide the following attribute:
+   In addition to supporting the methods and operations of mappings,
+   :class:`UserDict` instances provide the following attribute:
 
-.. attribute:: UserDict.data
+   .. attribute:: data
 
-   A real dictionary used to store the contents of the :class:`UserDict` class.
+      A real dictionary used to store the contents of the :class:`UserDict`
+      class.
 
 
 
@@ -985,13 +986,13 @@
    defaulting to the empty list ``[]``.  *list* can be any iterable, for
    example a real Python list or a :class:`UserList` object.
 
-In addition to supporting the methods and operations of mutable sequences,
-:class:`UserList` instances provide the following attribute:
+   In addition to supporting the methods and operations of mutable sequences,
+   :class:`UserList` instances provide the following attribute:
 
-.. attribute:: UserList.data
+   .. attribute:: data
 
-   A real :class:`list` object used to store the contents of the
-   :class:`UserList` class.
+      A real :class:`list` object used to store the contents of the
+      :class:`UserList` class.
 
 **Subclassing requirements:** Subclasses of :class:`UserList` are expect to
 offer a constructor which can be called with either no arguments or one

From python-checkins at python.org  Sun Jul 18 16:28:27 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sun, 18 Jul 2010 16:28:27 +0200 (CEST)
Subject: [Python-checkins] r82955 - in python/branches/release31-maint:
	Doc/library/collections.rst
Message-ID: <20100718142827.0CF61EE98C@mail.python.org>

Author: benjamin.peterson
Date: Sun Jul 18 16:28:26 2010
New Revision: 82955

Log:
Merged revisions 82952,82954 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82952 | benjamin.peterson | 2010-07-18 09:23:36 -0500 (Sun, 18 Jul 2010) | 1 line
  
  use classmethod
........
  r82954 | benjamin.peterson | 2010-07-18 09:27:02 -0500 (Sun, 18 Jul 2010) | 1 line
  
  nest method and attribute doc
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Doc/library/collections.rst

Modified: python/branches/release31-maint/Doc/library/collections.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/collections.rst	(original)
+++ python/branches/release31-maint/Doc/library/collections.rst	Sun Jul 18 16:28:26 2010
@@ -527,7 +527,7 @@
    :class:`defaultdict` objects support the following method in addition to the
    standard :class:`dict` operations:
 
-   .. method:: defaultdict.__missing__(key)
+   .. method:: __missing__(key)
 
       If the :attr:`default_factory` attribute is ``None``, this raises a
       :exc:`KeyError` exception with the *key* as argument.
@@ -547,7 +547,7 @@
    :class:`defaultdict` objects support the following instance variable:
 
 
-   .. attribute:: defaultdict.default_factory
+   .. attribute:: default_factory
 
       This attribute is used by the :meth:`__missing__` method; it is
       initialized from the first argument to the constructor, if present, or to
@@ -736,7 +736,7 @@
 three additional methods and one attribute.  To prevent conflicts with
 field names, the method and attribute names start with an underscore.
 
-.. method:: somenamedtuple._make(iterable)
+.. classmethod:: somenamedtuple._make(iterable)
 
    Class method that makes a new instance from an existing sequence or iterable.
 
@@ -864,11 +864,11 @@
 
    .. versionadded:: 3.1
 
-.. method:: OrderedDict.popitem(last=True)
+   .. method:: popitem(last=True)
 
-   The :meth:`popitem` method for ordered dictionaries returns and removes
-   a (key, value) pair.  The pairs are returned in LIFO order if *last* is
-   true or FIFO order if false.
+      The :meth:`popitem` method for ordered dictionaries returns and removes a
+      (key, value) pair.  The pairs are returned in LIFO order if *last* is true
+      or FIFO order if false.
 
 In addition to the usual mapping methods, ordered dictionaries also support
 reverse iteration using :func:`reversed`.
@@ -929,12 +929,13 @@
    initialized with its contents; note that a reference to *initialdata* will not
    be kept, allowing it be used for other purposes.
 
-In addition to supporting the methods and operations of mappings,
-:class:`UserDict` instances provide the following attribute:
+   In addition to supporting the methods and operations of mappings,
+   :class:`UserDict` instances provide the following attribute:
 
-.. attribute:: UserDict.data
+   .. attribute:: data
 
-   A real dictionary used to store the contents of the :class:`UserDict` class.
+      A real dictionary used to store the contents of the :class:`UserDict`
+      class.
 
 
 
@@ -958,13 +959,13 @@
    defaulting to the empty list ``[]``.  *list* can be any iterable, for
    example a real Python list or a :class:`UserList` object.
 
-In addition to supporting the methods and operations of mutable sequences,
-:class:`UserList` instances provide the following attribute:
+   In addition to supporting the methods and operations of mutable sequences,
+   :class:`UserList` instances provide the following attribute:
 
-.. attribute:: UserList.data
+   .. attribute:: data
 
-   A real :class:`list` object used to store the contents of the
-   :class:`UserList` class.
+      A real :class:`list` object used to store the contents of the
+      :class:`UserList` class.
 
 **Subclassing requirements:** Subclasses of :class:`UserList` are expect to
 offer a constructor which can be called with either no arguments or one

From python-checkins at python.org  Sun Jul 18 16:36:12 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Sun, 18 Jul 2010 16:36:12 +0200 (CEST)
Subject: [Python-checkins] r82956 - in python/branches/release31-maint:
	Lib/ast.py Lib/test/test_ast.py Misc/NEWS
Message-ID: <20100718143612.45CD5EEA42@mail.python.org>

Author: benjamin.peterson
Date: Sun Jul 18 16:36:12 2010
New Revision: 82956

Log:
remove support for byte literals; a new feature

Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/ast.py
   python/branches/release31-maint/Lib/test/test_ast.py
   python/branches/release31-maint/Misc/NEWS

Modified: python/branches/release31-maint/Lib/ast.py
==============================================================================
--- python/branches/release31-maint/Lib/ast.py	(original)
+++ python/branches/release31-maint/Lib/ast.py	Sun Jul 18 16:36:12 2010
@@ -50,7 +50,7 @@
     if isinstance(node_or_string, Expression):
         node_or_string = node_or_string.body
     def _convert(node):
-        if isinstance(node, (Str, Bytes)):
+        if isinstance(node, Str):
             return node.s
         elif isinstance(node, Num):
             return node.n

Modified: python/branches/release31-maint/Lib/test/test_ast.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_ast.py	(original)
+++ python/branches/release31-maint/Lib/test/test_ast.py	Sun Jul 18 16:36:12 2010
@@ -271,7 +271,6 @@
         self.assertEqual(ast.literal_eval('[1, 2, 3]'), [1, 2, 3])
         self.assertEqual(ast.literal_eval('{"foo": 42}'), {"foo": 42})
         self.assertEqual(ast.literal_eval('(True, False, None)'), (True, False, None))
-        self.assertEqual(ast.literal_eval('b"hi"'), b"hi")
         self.assertRaises(ValueError, ast.literal_eval, 'foo()')
 
     def test_literal_eval_issue4907(self):

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Sun Jul 18 16:36:12 2010
@@ -83,8 +83,6 @@
 
 - Issue #9243: Fix sndhdr module and add unit tests, contributed by James Lee.
 
-- ``ast.literal_eval()`` now allows byte literals.
-
 - Issue #9137: Fix issue in MutableMapping.update, which incorrectly
   treated keyword arguments called 'self' or 'other' specially.
 

From python-checkins at python.org  Sun Jul 18 18:01:36 2010
From: python-checkins at python.org (mark.dickinson)
Date: Sun, 18 Jul 2010 18:01:36 +0200 (CEST)
Subject: [Python-checkins] r82957 - in python/branches/py3k: configure
	configure.in
Message-ID: <20100718160136.C8399EE98C@mail.python.org>

Author: mark.dickinson
Date: Sun Jul 18 18:01:36 2010
New Revision: 82957

Log:
Add configure check for a bug with gcc soft floating-point and subnormals, to help debug issue 8265.

Modified:
   python/branches/py3k/configure
   python/branches/py3k/configure.in

Modified: python/branches/py3k/configure
==============================================================================
--- python/branches/py3k/configure	(original)
+++ python/branches/py3k/configure	Sun Jul 18 18:01:36 2010
@@ -1,5 +1,5 @@
 #! /bin/sh
-# From configure.in Revision: 82090 .
+# From configure.in Revision: 82746 .
 # Guess values for system-dependent variables and create Makefiles.
 # Generated by GNU Autoconf 2.65 for python 3.2.
 #
@@ -11641,6 +11641,45 @@
 
 fi
 
+# Temporary check for a floating-point bug that appeared on ARM.
+{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for buggy subnormal floating-point" >&5
+$as_echo_n "checking for buggy subnormal floating-point... " >&6; }
+if test "${ac_cv_buggy_subnormal+set}" = set; then :
+  $as_echo_n "(cached) " >&6
+else
+
+if test "$cross_compiling" = yes; then :
+  ac_cv_buggy_subnormal=no
+else
+  cat confdefs.h - <<_ACEOF >conftest.$ac_ext
+/* end confdefs.h.  */
+
+int main() {
+    double MIN = 2.2250738585072014e-308;  /* 2**-1022 */
+    double TINY = 4.9406564584124654e-324; /* 2**-1074 */
+    double TRIPLE_TINY = 1.4821969375237396e-323;  /* 3 * TINY */
+
+    double diff = MIN - TRIPLE_TINY;
+    double expected_diff = 2.2250738585071999e-308;
+
+    return !(diff == expected_diff);
+}
+
+_ACEOF
+if ac_fn_c_try_run "$LINENO"; then :
+  ac_cv_buggy_subnormal=no
+else
+  ac_cv_buggy_subnormal=yes
+fi
+rm -f core *.core core.conftest.* gmon.out bb.out conftest$ac_exeext \
+  conftest.$ac_objext conftest.beam conftest.$ac_ext
+fi
+
+fi
+
+{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_buggy_subnormal" >&5
+$as_echo "$ac_cv_buggy_subnormal" >&6; }
+
 # The short float repr introduced in Python 3.1 requires the
 # correctly-rounded string <-> double conversion functions from
 # Python/dtoa.c, which in turn require that the FPU uses 53-bit

Modified: python/branches/py3k/configure.in
==============================================================================
--- python/branches/py3k/configure.in	(original)
+++ python/branches/py3k/configure.in	Sun Jul 18 18:01:36 2010
@@ -3334,6 +3334,26 @@
    in ARM mixed-endian order (byte order 45670123)])
 fi
 
+# Temporary check for a floating-point bug that appeared on ARM.
+AC_MSG_CHECKING(for buggy subnormal floating-point)
+AC_CACHE_VAL(ac_cv_buggy_subnormal, [
+AC_RUN_IFELSE([AC_LANG_SOURCE([[
+int main() {
+    double MIN = 2.2250738585072014e-308;  /* 2**-1022 */
+    double TINY = 4.9406564584124654e-324; /* 2**-1074 */
+    double TRIPLE_TINY = 1.4821969375237396e-323;  /* 3 * TINY */
+
+    double diff = MIN - TRIPLE_TINY;
+    double expected_diff = 2.2250738585071999e-308;
+
+    return !(diff == expected_diff);    
+}
+]])],
+[ac_cv_buggy_subnormal=no],
+[ac_cv_buggy_subnormal=yes],
+[ac_cv_buggy_subnormal=no])])
+AC_MSG_RESULT($ac_cv_buggy_subnormal)
+
 # The short float repr introduced in Python 3.1 requires the
 # correctly-rounded string <-> double conversion functions from
 # Python/dtoa.c, which in turn require that the FPU uses 53-bit

From python-checkins at python.org  Sun Jul 18 18:13:27 2010
From: python-checkins at python.org (jean-paul.calderone)
Date: Sun, 18 Jul 2010 18:13:27 +0200 (CEST)
Subject: [Python-checkins] r82958 -
	python/branches/py3k/Doc/library/select.rst
Message-ID: <20100718161327.B7B41F400@mail.python.org>

Author: jean-paul.calderone
Date: Sun Jul 18 18:13:27 2010
New Revision: 82958

Log:
There is no method named "register(fd, eventmask)"; fix markup to just indicate this is code.

Modified:
   python/branches/py3k/Doc/library/select.rst

Modified: python/branches/py3k/Doc/library/select.rst
==============================================================================
--- python/branches/py3k/Doc/library/select.rst	(original)
+++ python/branches/py3k/Doc/library/select.rst	Sun Jul 18 18:13:27 2010
@@ -233,7 +233,7 @@
 .. method:: poll.modify(fd, eventmask)
 
    Modifies an already registered fd. This has the same effect as
-   :meth:`register(fd, eventmask)`.  Attempting to modify a file descriptor
+   ``register(fd, eventmask)``.  Attempting to modify a file descriptor
    that was never registered causes an :exc:`IOError` exception with errno
    :const:`ENOENT` to be raised.
 

From python-checkins at python.org  Sun Jul 18 18:30:32 2010
From: python-checkins at python.org (jean-paul.calderone)
Date: Sun, 18 Jul 2010 18:30:32 +0200 (CEST)
Subject: [Python-checkins] r82959 -
	python/branches/py3k/Doc/library/select.rst
Message-ID: <20100718163032.23429EE986@mail.python.org>

Author: jean-paul.calderone
Date: Sun Jul 18 18:30:31 2010
New Revision: 82959

Log:
Document the extra epoll flags

Modified:
   python/branches/py3k/Doc/library/select.rst

Modified: python/branches/py3k/Doc/library/select.rst
==============================================================================
--- python/branches/py3k/Doc/library/select.rst	(original)
+++ python/branches/py3k/Doc/library/select.rst	Sun Jul 18 18:30:31 2010
@@ -131,15 +131,15 @@
    | :const:`EPOLLONESHOT` | Set one-shot behavior. After one event is     |
    |                       | pulled out, the fd is internally disabled     |
    +-----------------------+-----------------------------------------------+
-   | :const:`EPOLLRDNORM`  | ???                                           |
+   | :const:`EPOLLRDNORM`  | Equivalent to :const:`EPOLLIN`                |
    +-----------------------+-----------------------------------------------+
-   | :const:`EPOLLRDBAND`  | ???                                           |
+   | :const:`EPOLLRDBAND`  | Priority data band can be read.               |
    +-----------------------+-----------------------------------------------+
-   | :const:`EPOLLWRNORM`  | ???                                           |
+   | :const:`EPOLLWRNORM`  | Equivalent to :const:`EPOLLOUT`               |
    +-----------------------+-----------------------------------------------+
-   | :const:`EPOLLWRBAND`  | ???                                           |
+   | :const:`EPOLLWRBAND`  | Priority data may be written.                 |
    +-----------------------+-----------------------------------------------+
-   | :const:`EPOLLMSG`     | ???                                           |
+   | :const:`EPOLLMSG`     | Ignored.                                      |
    +-----------------------+-----------------------------------------------+
 
 

From solipsis at pitrou.net  Mon Jul 19 01:23:53 2010
From: solipsis at pitrou.net (solipsis at pitrou.net)
Date: Mon, 19 Jul 2010 01:23:53 +0200 (CEST)
Subject: [Python-checkins] Daily py3k reference leaks (r82959): sum=0
Message-ID: <20100718232353.957D01770A@ns6635.ovh.net>

py3k results for svn r82959 (hg cset 00a2008ac58d)
--------------------------------------------------



Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/antoine/py3k/refleaks/reflogCoGAkJ', '-x']

From alexander.belopolsky at gmail.com  Mon Jul 19 06:16:39 2010
From: alexander.belopolsky at gmail.com (Alexander Belopolsky)
Date: Mon, 19 Jul 2010 00:16:39 -0400
Subject: [Python-checkins] r82949 - in python/branches/py3k:
	Doc/documenting/markup.rst Doc/library/pdb.rst
	Doc/tools/sphinxext/pyspecific.py Lib/pdb.doc Lib/pdb.py
In-Reply-To: <20100718101103.4A4D2EE986@mail.python.org>
References: <20100718101103.4A4D2EE986@mail.python.org>
Message-ID: <AANLkTilEK4l0DJHiuQ_JymnAKYZ-uT1I2RIeMwR55Lyk@mail.gmail.com>

On Sun, Jul 18, 2010 at 6:11 AM, georg.brandl
<python-checkins at python.org> wrote:
..
> + ? documentation (the docstring of the :mod:`pdb` module); if the environment
> + ? variable :envvar:`PAGER` is defined, the string is piped through that command
> + ? instead. ..
..
> + ? ?import pydoc
> + ? ?pydoc.pager(__doc__)

pydoc.pager() is a bit more sophisticated than the pdb documentation
suggests.  BTW, pydoc.pager docstring look like a code comment
mechanically converted to a docstring.

From python-checkins at python.org  Mon Jul 19 08:52:35 2010
From: python-checkins at python.org (georg.brandl)
Date: Mon, 19 Jul 2010 08:52:35 +0200 (CEST)
Subject: [Python-checkins] r82960 - python/branches/py3k/Doc/library/io.rst
Message-ID: <20100719065235.5F633E7F9@mail.python.org>

Author: georg.brandl
Date: Mon Jul 19 08:52:35 2010
New Revision: 82960

Log:
Clarify.

Modified:
   python/branches/py3k/Doc/library/io.rst

Modified: python/branches/py3k/Doc/library/io.rst
==============================================================================
--- python/branches/py3k/Doc/library/io.rst	(original)
+++ python/branches/py3k/Doc/library/io.rst	Mon Jul 19 08:52:35 2010
@@ -61,8 +61,8 @@
    Open *file* and return a corresponding stream.  If the file cannot be opened,
    an :exc:`IOError` is raised.
 
-   *file* is either a string or bytes object giving the name (and the path if
-   the file isn't in the current working directory) of the file to be opened or
+   *file* is either a string or bytes object giving the pathname (absolute or
+   relative to the current working directory) of the file to be opened or
    an integer file descriptor of the file to be wrapped.  (If a file descriptor
    is given, it is closed when the returned I/O object is closed, unless
    *closefd* is set to ``False``.)

From python-checkins at python.org  Mon Jul 19 08:57:52 2010
From: python-checkins at python.org (georg.brandl)
Date: Mon, 19 Jul 2010 08:57:52 +0200 (CEST)
Subject: [Python-checkins] r82961 -
	python/branches/py3k/Doc/documenting/markup.rst
Message-ID: <20100719065752.27221F003@mail.python.org>

Author: georg.brandl
Date: Mon Jul 19 08:57:52 2010
New Revision: 82961

Log:
Clarify :option: description.

Modified:
   python/branches/py3k/Doc/documenting/markup.rst

Modified: python/branches/py3k/Doc/documenting/markup.rst
==============================================================================
--- python/branches/py3k/Doc/documenting/markup.rst	(original)
+++ python/branches/py3k/Doc/documenting/markup.rst	Mon Jul 19 08:57:52 2010
@@ -502,8 +502,9 @@
 
 .. describe:: option
 
-   A command-line option to an executable program.  The leading hyphen(s) must
-   be included.  If a matching ``cmdoption`` directive exists, it is linked to.
+   A command-line option of Python.  The leading hyphen(s) must be included.
+   If a matching ``cmdoption`` directive exists, it is linked to.  For options
+   of other programs or scripts, use simple ````code```` markup.
 
 .. describe:: program
 

From python-checkins at python.org  Mon Jul 19 09:31:40 2010
From: python-checkins at python.org (mark.dickinson)
Date: Mon, 19 Jul 2010 09:31:40 +0200 (CEST)
Subject: [Python-checkins] r82962 - in python/branches/py3k: configure
	configure.in
Message-ID: <20100719073140.6550DF2E3@mail.python.org>

Author: mark.dickinson
Date: Mon Jul 19 09:31:40 2010
New Revision: 82962

Log:
Remove temporary debugging code.

Modified:
   python/branches/py3k/configure
   python/branches/py3k/configure.in

Modified: python/branches/py3k/configure
==============================================================================
--- python/branches/py3k/configure	(original)
+++ python/branches/py3k/configure	Mon Jul 19 09:31:40 2010
@@ -1,5 +1,5 @@
 #! /bin/sh
-# From configure.in Revision: 82746 .
+# From configure.in Revision: 82090 .
 # Guess values for system-dependent variables and create Makefiles.
 # Generated by GNU Autoconf 2.65 for python 3.2.
 #
@@ -11641,45 +11641,6 @@
 
 fi
 
-# Temporary check for a floating-point bug that appeared on ARM.
-{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for buggy subnormal floating-point" >&5
-$as_echo_n "checking for buggy subnormal floating-point... " >&6; }
-if test "${ac_cv_buggy_subnormal+set}" = set; then :
-  $as_echo_n "(cached) " >&6
-else
-
-if test "$cross_compiling" = yes; then :
-  ac_cv_buggy_subnormal=no
-else
-  cat confdefs.h - <<_ACEOF >conftest.$ac_ext
-/* end confdefs.h.  */
-
-int main() {
-    double MIN = 2.2250738585072014e-308;  /* 2**-1022 */
-    double TINY = 4.9406564584124654e-324; /* 2**-1074 */
-    double TRIPLE_TINY = 1.4821969375237396e-323;  /* 3 * TINY */
-
-    double diff = MIN - TRIPLE_TINY;
-    double expected_diff = 2.2250738585071999e-308;
-
-    return !(diff == expected_diff);
-}
-
-_ACEOF
-if ac_fn_c_try_run "$LINENO"; then :
-  ac_cv_buggy_subnormal=no
-else
-  ac_cv_buggy_subnormal=yes
-fi
-rm -f core *.core core.conftest.* gmon.out bb.out conftest$ac_exeext \
-  conftest.$ac_objext conftest.beam conftest.$ac_ext
-fi
-
-fi
-
-{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_buggy_subnormal" >&5
-$as_echo "$ac_cv_buggy_subnormal" >&6; }
-
 # The short float repr introduced in Python 3.1 requires the
 # correctly-rounded string <-> double conversion functions from
 # Python/dtoa.c, which in turn require that the FPU uses 53-bit

Modified: python/branches/py3k/configure.in
==============================================================================
--- python/branches/py3k/configure.in	(original)
+++ python/branches/py3k/configure.in	Mon Jul 19 09:31:40 2010
@@ -3334,26 +3334,6 @@
    in ARM mixed-endian order (byte order 45670123)])
 fi
 
-# Temporary check for a floating-point bug that appeared on ARM.
-AC_MSG_CHECKING(for buggy subnormal floating-point)
-AC_CACHE_VAL(ac_cv_buggy_subnormal, [
-AC_RUN_IFELSE([AC_LANG_SOURCE([[
-int main() {
-    double MIN = 2.2250738585072014e-308;  /* 2**-1022 */
-    double TINY = 4.9406564584124654e-324; /* 2**-1074 */
-    double TRIPLE_TINY = 1.4821969375237396e-323;  /* 3 * TINY */
-
-    double diff = MIN - TRIPLE_TINY;
-    double expected_diff = 2.2250738585071999e-308;
-
-    return !(diff == expected_diff);    
-}
-]])],
-[ac_cv_buggy_subnormal=no],
-[ac_cv_buggy_subnormal=yes],
-[ac_cv_buggy_subnormal=no])])
-AC_MSG_RESULT($ac_cv_buggy_subnormal)
-
 # The short float repr introduced in Python 3.1 requires the
 # correctly-rounded string <-> double conversion functions from
 # Python/dtoa.c, which in turn require that the FPU uses 53-bit

From python-checkins at python.org  Mon Jul 19 09:35:16 2010
From: python-checkins at python.org (mark.dickinson)
Date: Mon, 19 Jul 2010 09:35:16 +0200 (CEST)
Subject: [Python-checkins] r82963 - in python/branches/release31-maint:
	configure configure.in
Message-ID: <20100719073516.6467ED326@mail.python.org>

Author: mark.dickinson
Date: Mon Jul 19 09:35:16 2010
New Revision: 82963

Log:
Revert temporary debugging commit from r82756.

Modified:
   python/branches/release31-maint/configure
   python/branches/release31-maint/configure.in

Modified: python/branches/release31-maint/configure
==============================================================================
--- python/branches/release31-maint/configure	(original)
+++ python/branches/release31-maint/configure	Mon Jul 19 09:35:16 2010
@@ -1,5 +1,5 @@
 #! /bin/sh
-# From configure.in Revision: 82141 .
+# From configure.in Revision: 80702 .
 # Guess values for system-dependent variables and create Makefiles.
 # Generated by GNU Autoconf 2.65 for python 3.1.
 #
@@ -1923,11 +1923,11 @@
        cat confdefs.h - <<_ACEOF >conftest.$ac_ext
 /* end confdefs.h.  */
 $ac_includes_default
-	     enum { N = $2 / 2 - 1 };
 int
 main ()
 {
-static int test_array [1 - 2 * !(0 < ($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1))];
+static int test_array [1 - 2 * !(enum { N = $2 / 2 - 1 };
+	     0 < ($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1))];
 test_array [0] = 0
 
   ;
@@ -1938,11 +1938,11 @@
   cat confdefs.h - <<_ACEOF >conftest.$ac_ext
 /* end confdefs.h.  */
 $ac_includes_default
-	        enum { N = $2 / 2 - 1 };
 int
 main ()
 {
-static int test_array [1 - 2 * !(($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1)
+static int test_array [1 - 2 * !(enum { N = $2 / 2 - 1 };
+		($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1)
 		 < ($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 2))];
 test_array [0] = 0
 
@@ -4126,15 +4126,6 @@
   " >&2;}
 fi
 
-# If using GCC, give result of 'gcc -v' in configure output.
-if test "$GCC" = yes; then
-{ $as_echo "$as_me:${as_lineno-$LINENO}: checking gcc version" >&5
-$as_echo_n "checking gcc version... " >&6; }
-gcc_version=`$CC -v`
-{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $gcc_version" >&5
-$as_echo "$gcc_version" >&6; }
-fi
-
 
 # checks for UNIX variants that set C preprocessor variables
 

Modified: python/branches/release31-maint/configure.in
==============================================================================
--- python/branches/release31-maint/configure.in	(original)
+++ python/branches/release31-maint/configure.in	Mon Jul 19 09:35:16 2010
@@ -504,13 +504,6 @@
   ])
 fi
 
-# If using GCC, give result of 'gcc -v' in configure output.
-if test "$GCC" = yes; then
-AC_MSG_CHECKING(gcc version)
-gcc_version=`$CC -v`
-AC_MSG_RESULT($gcc_version)
-fi
-
 
 # checks for UNIX variants that set C preprocessor variables
 AC_AIX

From python-checkins at python.org  Mon Jul 19 10:02:46 2010
From: python-checkins at python.org (georg.brandl)
Date: Mon, 19 Jul 2010 10:02:46 +0200 (CEST)
Subject: [Python-checkins] r82964 - in python/branches/py3k:
	Doc/library/pdb.rst Lib/pdb.py
Message-ID: <20100719080246.BC649EDD9@mail.python.org>

Author: georg.brandl
Date: Mon Jul 19 10:02:46 2010
New Revision: 82964

Log:
pydoc.pager does not promise to use $PAGER.

Modified:
   python/branches/py3k/Doc/library/pdb.rst
   python/branches/py3k/Lib/pdb.py

Modified: python/branches/py3k/Doc/library/pdb.rst
==============================================================================
--- python/branches/py3k/Doc/library/pdb.rst	(original)
+++ python/branches/py3k/Doc/library/pdb.rst	Mon Jul 19 10:02:46 2010
@@ -207,10 +207,9 @@
 
    Without argument, print the list of available commands.  With a *command* as
    argument, print help about that command.  ``help pdb`` displays the full
-   documentation (the docstring of the :mod:`pdb` module); if the environment
-   variable :envvar:`PAGER` is defined, the string is piped through that command
-   instead.  Since the *command* argument must be an identifier, ``help exec``
-   must be entered to get help on the ``!`` command.
+   documentation (the docstring of the :mod:`pdb` module).  Since the *command*
+   argument must be an identifier, ``help exec`` must be entered to get help on
+   the ``!`` command.
 
 .. pdbcommand:: w(here)
 

Modified: python/branches/py3k/Lib/pdb.py
==============================================================================
--- python/branches/py3k/Lib/pdb.py	(original)
+++ python/branches/py3k/Lib/pdb.py	Mon Jul 19 10:02:46 2010
@@ -1123,7 +1123,7 @@
         print("""h(elp)
 Without argument, print the list of available commands.
 With a command name as argument, print help about that command
-"help pdb" pipes the full documentation file to the $PAGER
+"help pdb" shows the full pdb documentation
 "help exec" gives help on the ! command""", file=self.stdout)
 
     def help_where(self):

From python-checkins at python.org  Mon Jul 19 13:28:05 2010
From: python-checkins at python.org (georg.brandl)
Date: Mon, 19 Jul 2010 13:28:05 +0200 (CEST)
Subject: [Python-checkins] r82965 -
	python/branches/py3k/Doc/tutorial/classes.rst
Message-ID: <20100719112805.CD0FAF490@mail.python.org>

Author: georg.brandl
Date: Mon Jul 19 13:28:05 2010
New Revision: 82965

Log:
Clarification.  Yay importlib!

Modified:
   python/branches/py3k/Doc/tutorial/classes.rst

Modified: python/branches/py3k/Doc/tutorial/classes.rst
==============================================================================
--- python/branches/py3k/Doc/tutorial/classes.rst	(original)
+++ python/branches/py3k/Doc/tutorial/classes.rst	Mon Jul 19 13:28:05 2010
@@ -737,7 +737,7 @@
    StopIteration
 
 Having seen the mechanics behind the iterator protocol, it is easy to add
-iterator behavior to your classes.  Define a :meth:`__iter__` method which
+iterator behavior to your classes.  Define an :meth:`__iter__` method which
 returns an object with a :meth:`__next__` method.  If the class defines
 :meth:`__next__`, then :meth:`__iter__` can just return ``self``::
 
@@ -754,7 +754,10 @@
            self.index = self.index - 1
            return self.data[self.index]
 
-   >>> for char in Reverse('spam'):
+   >>> rev = Reverse('spam')
+   >>> iter(rev)
+   <__main__.Reverse object at 0x00A1DB50>
+   >>> for char in rev:
    ...     print(char)
    ...
    m

From python-checkins at python.org  Mon Jul 19 14:36:57 2010
From: python-checkins at python.org (stefan.krah)
Date: Mon, 19 Jul 2010 14:36:57 +0200 (CEST)
Subject: [Python-checkins] r82966 - python/branches/py3k/Include/Python.h
Message-ID: <20100719123657.6BF54EE983@mail.python.org>

Author: stefan.krah
Date: Mon Jul 19 14:36:57 2010
New Revision: 82966

Log:
Issue #9036: Throughout the code base, Py_CHARMASK is used on 8-bit wide
signed/unsigned chars or on integers directly derived from those. In all
cases, it could be replaced by a simple cast to (unsigned char). Reasons
for the change:

  a) Make the comment more explicit.

  b) If char is unsigned, the cast is optimized away.

  c) If char is unsigned, gcc emits spurious "array subscript
     has type 'char'" warnings.



Modified:
   python/branches/py3k/Include/Python.h

Modified: python/branches/py3k/Include/Python.h
==============================================================================
--- python/branches/py3k/Include/Python.h	(original)
+++ python/branches/py3k/Include/Python.h	Mon Jul 19 14:36:57 2010
@@ -134,13 +134,8 @@
 }
 #endif
 
-/* Convert a possibly signed character to a nonnegative int */
-/* XXX This assumes characters are 8 bits wide */
-#ifdef __CHAR_UNSIGNED__
-#define Py_CHARMASK(c)		(c)
-#else
+/* Argument must be a char or an int in [-128, 127] or [0, 255]. */
 #define Py_CHARMASK(c)		((unsigned char)((c) & 0xff))
-#endif
 
 #include "pyfpe.h"
 

From python-checkins at python.org  Mon Jul 19 15:00:36 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Mon, 19 Jul 2010 15:00:36 +0200 (CEST)
Subject: [Python-checkins] r82967 -
	python/branches/py3k/Lib/test/test_sysconfig.py
Message-ID: <20100719130036.329ABEE997@mail.python.org>

Author: ronald.oussoren
Date: Mon Jul 19 15:00:36 2010
New Revision: 82967

Log:
This patch adds a testcase that demonstrates a problem
with the expansion of LDSHARED when accessing that value
through sysconfig.

The problem is probably caused by the 9189.

A fix will follow shortly.


Modified:
   python/branches/py3k/Lib/test/test_sysconfig.py

Modified: python/branches/py3k/Lib/test/test_sysconfig.py
==============================================================================
--- python/branches/py3k/Lib/test/test_sysconfig.py	(original)
+++ python/branches/py3k/Lib/test/test_sysconfig.py	Mon Jul 19 15:00:36 2010
@@ -277,6 +277,12 @@
             _main()
         self.assertTrue(len(output.getvalue().split('\n')) > 0)
 
+    def test_ldshared_value(self):
+        ldflags = sysconfig.get_config_var('LDFLAGS')
+        ldshared = sysconfig.get_config_var('LDSHARED')
+
+        self.assertIn(ldflags, ldshared)
+
 
 def test_main():
     run_unittest(TestSysConfig)

From python-checkins at python.org  Mon Jul 19 15:00:48 2010
From: python-checkins at python.org (stefan.krah)
Date: Mon, 19 Jul 2010 15:00:48 +0200 (CEST)
Subject: [Python-checkins] r82968 - in python/branches/release31-maint:
	Include/Python.h
Message-ID: <20100719130048.9801AEE984@mail.python.org>

Author: stefan.krah
Date: Mon Jul 19 15:00:48 2010
New Revision: 82968

Log:
Merged revisions 82966 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82966 | stefan.krah | 2010-07-19 14:36:57 +0200 (Mon, 19 Jul 2010) | 13 lines
  
  Issue #9036: Throughout the code base, Py_CHARMASK is used on 8-bit wide
  signed/unsigned chars or on integers directly derived from those. In all
  cases, it could be replaced by a simple cast to (unsigned char). Reasons
  for the change:
  
    a) Make the comment more explicit.
  
    b) If char is unsigned, the cast is optimized away.
  
    c) If char is unsigned, gcc emits spurious "array subscript
       has type 'char'" warnings.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Include/Python.h

Modified: python/branches/release31-maint/Include/Python.h
==============================================================================
--- python/branches/release31-maint/Include/Python.h	(original)
+++ python/branches/release31-maint/Include/Python.h	Mon Jul 19 15:00:48 2010
@@ -126,13 +126,8 @@
 /* _Py_Mangle is defined in compile.c */
 PyAPI_FUNC(PyObject*) _Py_Mangle(PyObject *p, PyObject *name);
 
-/* Convert a possibly signed character to a nonnegative int */
-/* XXX This assumes characters are 8 bits wide */
-#ifdef __CHAR_UNSIGNED__
-#define Py_CHARMASK(c)		(c)
-#else
+/* Argument must be a char or an int in [-128, 127] or [0, 255]. */
 #define Py_CHARMASK(c)		((unsigned char)((c) & 0xff))
-#endif
 
 #include "pyfpe.h"
 

From python-checkins at python.org  Mon Jul 19 15:14:01 2010
From: python-checkins at python.org (stefan.krah)
Date: Mon, 19 Jul 2010 15:14:01 +0200 (CEST)
Subject: [Python-checkins] r82969 -
	python/branches/release27-maint/Include/Python.h
Message-ID: <20100719131401.9DB15F0FF@mail.python.org>

Author: stefan.krah
Date: Mon Jul 19 15:14:01 2010
New Revision: 82969

Log:
Issue #9036: Throughout the code base, Py_CHARMASK is used on 8-bit wide
signed/unsigned chars or on integers directly derived from those. In all
cases, it could be replaced by a simple cast to (unsigned char). Reasons
for the change:

  a) Make the comment more explicit.

  b) If char is unsigned, the cast is optimized away.

  c) If char is unsigned, gcc emits spurious "array subscript
     has type 'char'" warnings.




Modified:
   python/branches/release27-maint/Include/Python.h

Modified: python/branches/release27-maint/Include/Python.h
==============================================================================
--- python/branches/release27-maint/Include/Python.h	(original)
+++ python/branches/release27-maint/Include/Python.h	Mon Jul 19 15:14:01 2010
@@ -150,13 +150,8 @@
    Set ml_flags in the PyMethodDef to METH_NOARGS. */
 #define PyArg_NoArgs(v)		PyArg_Parse(v, "")
 
-/* Convert a possibly signed character to a nonnegative int */
-/* XXX This assumes characters are 8 bits wide */
-#ifdef __CHAR_UNSIGNED__
-#define Py_CHARMASK(c)		(c)
-#else
+/* Argument must be a char or an int in [-128, 127] or [0, 255]. */
 #define Py_CHARMASK(c)		((unsigned char)((c) & 0xff))
-#endif
 
 #include "pyfpe.h"
 

From python-checkins at python.org  Mon Jul 19 15:36:13 2010
From: python-checkins at python.org (stefan.krah)
Date: Mon, 19 Jul 2010 15:36:13 +0200 (CEST)
Subject: [Python-checkins] r82970 - in python/branches/release26-maint:
	Include/Python.h
Message-ID: <20100719133613.9EA69F128@mail.python.org>

Author: stefan.krah
Date: Mon Jul 19 15:36:13 2010
New Revision: 82970

Log:
Merged revisions 82969 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

........
  r82969 | stefan.krah | 2010-07-19 15:14:01 +0200 (Mon, 19 Jul 2010) | 14 lines
  
  Issue #9036: Throughout the code base, Py_CHARMASK is used on 8-bit wide
  signed/unsigned chars or on integers directly derived from those. In all
  cases, it could be replaced by a simple cast to (unsigned char). Reasons
  for the change:
  
    a) Make the comment more explicit.
  
    b) If char is unsigned, the cast is optimized away.
  
    c) If char is unsigned, gcc emits spurious "array subscript
       has type 'char'" warnings.
........


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Include/Python.h

Modified: python/branches/release26-maint/Include/Python.h
==============================================================================
--- python/branches/release26-maint/Include/Python.h	(original)
+++ python/branches/release26-maint/Include/Python.h	Mon Jul 19 15:36:13 2010
@@ -147,13 +147,8 @@
    Set ml_flags in the PyMethodDef to METH_NOARGS. */
 #define PyArg_NoArgs(v)		PyArg_Parse(v, "")
 
-/* Convert a possibly signed character to a nonnegative int */
-/* XXX This assumes characters are 8 bits wide */
-#ifdef __CHAR_UNSIGNED__
-#define Py_CHARMASK(c)		(c)
-#else
+/* Argument must be a char or an int in [-128, 127] or [0, 255]. */
 #define Py_CHARMASK(c)		((unsigned char)((c) & 0xff))
-#endif
 
 #include "pyfpe.h"
 

From python-checkins at python.org  Mon Jul 19 16:20:53 2010
From: python-checkins at python.org (stefan.krah)
Date: Mon, 19 Jul 2010 16:20:53 +0200 (CEST)
Subject: [Python-checkins] r82971 - in python/branches/py3k/Lib:
	subprocess.py test/test_subprocess.py
Message-ID: <20100719142053.5083DF300@mail.python.org>

Author: stefan.krah
Date: Mon Jul 19 16:20:53 2010
New Revision: 82971

Log:
Issue #9265: Incorrect name passed as arg[0] when shell=True
and executable specified.



Modified:
   python/branches/py3k/Lib/subprocess.py
   python/branches/py3k/Lib/test/test_subprocess.py

Modified: python/branches/py3k/Lib/subprocess.py
==============================================================================
--- python/branches/py3k/Lib/subprocess.py	(original)
+++ python/branches/py3k/Lib/subprocess.py	Mon Jul 19 16:20:53 2010
@@ -1073,6 +1073,8 @@
 
             if shell:
                 args = ["/bin/sh", "-c"] + args
+                if executable:
+                    args[0] = executable
 
             if executable is None:
                 executable = args[0]

Modified: python/branches/py3k/Lib/test/test_subprocess.py
==============================================================================
--- python/branches/py3k/Lib/test/test_subprocess.py	(original)
+++ python/branches/py3k/Lib/test/test_subprocess.py	Mon Jul 19 16:20:53 2010
@@ -750,6 +750,25 @@
         os.remove(fname)
         self.assertEqual(rc, 47)
 
+    def test_specific_shell(self):
+        # Issue #9265: Incorrect name passed as arg[0].
+        shells = []
+        for prefix in ['/bin', '/usr/bin/', '/usr/local/bin']:
+            for name in ['bash', 'ksh']:
+                sh = os.path.join(prefix, name)
+                if os.path.isfile(sh):
+                    shells.append(sh)
+        if not shells: # Will probably work for any shell but csh.
+            self.skipTest("bash or ksh required for this test")
+        sh = '/bin/sh'
+        if os.path.isfile(sh) and not os.path.islink(sh):
+            # Test will fail if /bin/sh is a symlink to csh.
+            shells.append(sh)
+        for sh in shells:
+            p = subprocess.Popen("echo $0", executable=sh, shell=True,
+                                 stdout=subprocess.PIPE)
+            self.assertEqual(p.stdout.read().strip(), bytes(sh, 'ascii'))
+
     def _kill_process(self, method, *args):
         # Do not inherit file handles from the parent.
         # It should fix failures on some platforms.

From python-checkins at python.org  Mon Jul 19 16:39:36 2010
From: python-checkins at python.org (stefan.krah)
Date: Mon, 19 Jul 2010 16:39:36 +0200 (CEST)
Subject: [Python-checkins] r82972 - in python/branches/release31-maint:
	Lib/subprocess.py Lib/test/test_subprocess.py
Message-ID: <20100719143936.B92A5EF3D@mail.python.org>

Author: stefan.krah
Date: Mon Jul 19 16:39:36 2010
New Revision: 82972

Log:
Merged revisions 82971 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82971 | stefan.krah | 2010-07-19 16:20:53 +0200 (Mon, 19 Jul 2010) | 4 lines
  
  Issue #9265: Incorrect name passed as arg[0] when shell=True
  and executable specified.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/subprocess.py
   python/branches/release31-maint/Lib/test/test_subprocess.py

Modified: python/branches/release31-maint/Lib/subprocess.py
==============================================================================
--- python/branches/release31-maint/Lib/subprocess.py	(original)
+++ python/branches/release31-maint/Lib/subprocess.py	Mon Jul 19 16:39:36 2010
@@ -1044,6 +1044,8 @@
 
             if shell:
                 args = ["/bin/sh", "-c"] + args
+                if executable:
+                    args[0] = executable
 
             if executable is None:
                 executable = args[0]

Modified: python/branches/release31-maint/Lib/test/test_subprocess.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_subprocess.py	(original)
+++ python/branches/release31-maint/Lib/test/test_subprocess.py	Mon Jul 19 16:39:36 2010
@@ -663,6 +663,25 @@
             os.remove(fname)
             self.assertEqual(rc, 47)
 
+        def test_specific_shell(self):
+            # Issue #9265: Incorrect name passed as arg[0].
+            shells = []
+            for prefix in ['/bin', '/usr/bin/', '/usr/local/bin']:
+                for name in ['bash', 'ksh']:
+                    sh = os.path.join(prefix, name)
+                    if os.path.isfile(sh):
+                        shells.append(sh)
+            if not shells: # Will probably work for any shell but csh.
+                self.skipTest("bash or ksh required for this test")
+            sh = '/bin/sh'
+            if os.path.isfile(sh) and not os.path.islink(sh):
+                # Test will fail if /bin/sh is a symlink to csh.
+                shells.append(sh)
+            for sh in shells:
+                p = subprocess.Popen("echo $0", executable=sh, shell=True,
+                                     stdout=subprocess.PIPE)
+                self.assertEqual(p.stdout.read().strip(), bytes(sh, 'ascii'))
+
         def DISABLED_test_send_signal(self):
             p = subprocess.Popen([sys.executable,
                               "-c", "input()"])

From python-checkins at python.org  Mon Jul 19 16:41:08 2010
From: python-checkins at python.org (stefan.krah)
Date: Mon, 19 Jul 2010 16:41:08 +0200 (CEST)
Subject: [Python-checkins] r82973 - in python/branches/release27-maint/Lib:
	subprocess.py test/test_subprocess.py
Message-ID: <20100719144108.7BEA3F594@mail.python.org>

Author: stefan.krah
Date: Mon Jul 19 16:41:08 2010
New Revision: 82973

Log:
Issue #9265: Incorrect name passed as arg[0] when shell=True
and executable specified.



Modified:
   python/branches/release27-maint/Lib/subprocess.py
   python/branches/release27-maint/Lib/test/test_subprocess.py

Modified: python/branches/release27-maint/Lib/subprocess.py
==============================================================================
--- python/branches/release27-maint/Lib/subprocess.py	(original)
+++ python/branches/release27-maint/Lib/subprocess.py	Mon Jul 19 16:41:08 2010
@@ -1091,6 +1091,8 @@
 
             if shell:
                 args = ["/bin/sh", "-c"] + args
+                if executable:
+                    args[0] = executable
 
             if executable is None:
                 executable = args[0]

Modified: python/branches/release27-maint/Lib/test/test_subprocess.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_subprocess.py	(original)
+++ python/branches/release27-maint/Lib/test/test_subprocess.py	Mon Jul 19 16:41:08 2010
@@ -641,6 +641,25 @@
         os.remove(fname)
         self.assertEqual(rc, 47)
 
+    def test_specific_shell(self):
+        # Issue #9265: Incorrect name passed as arg[0].
+        shells = []
+        for prefix in ['/bin', '/usr/bin/', '/usr/local/bin']:
+            for name in ['bash', 'ksh']:
+                sh = os.path.join(prefix, name)
+                if os.path.isfile(sh):
+                    shells.append(sh)
+        if not shells: # Will probably work for any shell but csh.
+            self.skipTest("bash or ksh required for this test")
+        sh = '/bin/sh'
+        if os.path.isfile(sh) and not os.path.islink(sh):
+            # Test will fail if /bin/sh is a symlink to csh.
+            shells.append(sh)
+        for sh in shells:
+            p = subprocess.Popen("echo $0", executable=sh, shell=True,
+                                 stdout=subprocess.PIPE)
+            self.assertEqual(p.stdout.read().strip(), sh)
+
     def _kill_process(self, method, *args):
         # Do not inherit file handles from the parent.
         # It should fix failures on some platforms.

From python-checkins at python.org  Mon Jul 19 16:49:38 2010
From: python-checkins at python.org (stefan.krah)
Date: Mon, 19 Jul 2010 16:49:38 +0200 (CEST)
Subject: [Python-checkins] r82974 - in python/branches/release26-maint:
	Lib/subprocess.py Lib/test/test_subprocess.py
Message-ID: <20100719144938.75833F4E7@mail.python.org>

Author: stefan.krah
Date: Mon Jul 19 16:49:38 2010
New Revision: 82974

Log:
Merged revisions 82973 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

........
  r82973 | stefan.krah | 2010-07-19 16:41:08 +0200 (Mon, 19 Jul 2010) | 4 lines
  
  Issue #9265: Incorrect name passed as arg[0] when shell=True
  and executable specified.
........


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Lib/subprocess.py
   python/branches/release26-maint/Lib/test/test_subprocess.py

Modified: python/branches/release26-maint/Lib/subprocess.py
==============================================================================
--- python/branches/release26-maint/Lib/subprocess.py	(original)
+++ python/branches/release26-maint/Lib/subprocess.py	Mon Jul 19 16:49:38 2010
@@ -1029,6 +1029,8 @@
 
             if shell:
                 args = ["/bin/sh", "-c"] + args
+                if executable:
+                    args[0] = executable
 
             if executable is None:
                 executable = args[0]

Modified: python/branches/release26-maint/Lib/test/test_subprocess.py
==============================================================================
--- python/branches/release26-maint/Lib/test/test_subprocess.py	(original)
+++ python/branches/release26-maint/Lib/test/test_subprocess.py	Mon Jul 19 16:49:38 2010
@@ -613,6 +613,25 @@
             os.remove(fname)
             self.assertEqual(rc, 47)
 
+        def test_specific_shell(self):
+            # Issue #9265: Incorrect name passed as arg[0].
+            shells = []
+            for prefix in ['/bin', '/usr/bin/', '/usr/local/bin']:
+                for name in ['bash', 'ksh']:
+                    sh = os.path.join(prefix, name)
+                    if os.path.isfile(sh):
+                        shells.append(sh)
+            if not shells: # Will probably work for any shell but csh.
+                self.skipTest("bash or ksh required for this test")
+            sh = '/bin/sh'
+            if os.path.isfile(sh) and not os.path.islink(sh):
+                # Test will fail if /bin/sh is a symlink to csh.
+                shells.append(sh)
+            for sh in shells:
+                p = subprocess.Popen("echo $0", executable=sh, shell=True,
+                                     stdout=subprocess.PIPE)
+                self.assertEqual(p.stdout.read().strip(), sh)
+
         def DISABLED_test_send_signal(self):
             p = subprocess.Popen([sys.executable,
                               "-c", "input()"])

From python-checkins at python.org  Mon Jul 19 17:43:23 2010
From: python-checkins at python.org (stefan.krah)
Date: Mon, 19 Jul 2010 17:43:23 +0200 (CEST)
Subject: [Python-checkins] r82975 - in python/branches/release26-maint:
	Lib/test/test_posix.py Modules/posixmodule.c
Message-ID: <20100719154323.69537DF0B@mail.python.org>

Author: stefan.krah
Date: Mon Jul 19 17:43:23 2010
New Revision: 82975

Log:
Merged revisions 82853-82854 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

........
  r82853 | stefan.krah | 2010-07-13 21:17:08 +0200 (Tue, 13 Jul 2010) | 4 lines
  
  Issue #9185: On Solaris and OpenBSD, posix_getcwd() could loop indefinitely
  if the path length exceeded PATH_MAX.
........
  r82854 | stefan.krah | 2010-07-13 21:40:00 +0200 (Tue, 13 Jul 2010) | 3 lines
  
  Remove PYOS_OS2 special cases from the Solaris/OpenBSD section.
........


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Lib/test/test_posix.py
   python/branches/release26-maint/Modules/posixmodule.c

Modified: python/branches/release26-maint/Lib/test/test_posix.py
==============================================================================
--- python/branches/release26-maint/Lib/test/test_posix.py	(original)
+++ python/branches/release26-maint/Lib/test/test_posix.py	Mon Jul 19 17:43:23 2010
@@ -11,6 +11,7 @@
 import os
 import pwd
 import shutil
+import sys
 import unittest
 import warnings
 warnings.filterwarnings('ignore', '.* potential security risk .*',
@@ -287,8 +288,13 @@
                     os.chdir(dirname)
                     try:
                         os.getcwd()
-                        if current_path_length < 1027:
+                        if current_path_length < 4099:
                             _create_and_do_getcwd(dirname, current_path_length + len(dirname) + 1)
+                    except OSError as e:
+                        expected_errno = errno.ENAMETOOLONG
+                        if 'sunos' in sys.platform or 'openbsd' in sys.platform:
+                            expected_errno = errno.ERANGE # Issue 9185
+                        self.assertEqual(e.errno, expected_errno)
                     finally:
                         os.chdir('..')
                         os.rmdir(dirname)

Modified: python/branches/release26-maint/Modules/posixmodule.c
==============================================================================
--- python/branches/release26-maint/Modules/posixmodule.c	(original)
+++ python/branches/release26-maint/Modules/posixmodule.c	Mon Jul 19 17:43:23 2010
@@ -1970,6 +1970,24 @@
 "getcwd() -> path\n\n\
 Return a string representing the current working directory.");
 
+#if (defined(__sun) && defined(__SVR4)) || defined(__OpenBSD__)
+/* Issue 9185: getcwd() returns NULL/ERANGE indefinitely. */
+static PyObject *
+posix_getcwd(PyObject *self, PyObject *noargs)
+{
+    char buf[PATH_MAX+2];
+    char *res;
+
+    Py_BEGIN_ALLOW_THREADS
+    res = getcwd(buf, sizeof buf);
+    Py_END_ALLOW_THREADS
+
+    if (res == NULL)
+        return posix_error();
+
+    return PyString_FromString(buf);
+}
+#else
 static PyObject *
 posix_getcwd(PyObject *self, PyObject *noargs)
 {
@@ -2006,6 +2024,7 @@
 
     return dynamic_return;
 }
+#endif /* getcwd() NULL/ERANGE workaround. */
 
 #ifdef Py_USING_UNICODE
 PyDoc_STRVAR(posix_getcwdu__doc__,

From python-checkins at python.org  Mon Jul 19 19:19:04 2010
From: python-checkins at python.org (ezio.melotti)
Date: Mon, 19 Jul 2010 19:19:04 +0200 (CEST)
Subject: [Python-checkins] r82976 - in tracker:
	instances/python-dev/html/issue.search.html
	instances/python-dev/html/page.html
	roundup-src/roundup/cgi/templating.py
Message-ID: <20100719171904.8AE40F36F@mail.python.org>

Author: ezio.melotti
Date: Mon Jul 19 19:19:02 2010
New Revision: 82976

Log:
#262: Ignore attachments content while searching. Also add a checkbox to enable/disable the option in the search page. Original patch by Daniel Diniz.

Modified:
   tracker/instances/python-dev/html/issue.search.html
   tracker/instances/python-dev/html/page.html
   tracker/roundup-src/roundup/cgi/templating.py

Modified: tracker/instances/python-dev/html/issue.search.html
==============================================================================
--- tracker/instances/python-dev/html/issue.search.html	(original)
+++ tracker/instances/python-dev/html/issue.search.html	Mon Jul 19 19:19:02 2010
@@ -42,6 +42,15 @@
   <td>&nbsp;</td>
 </tr>
 
+<tr tal:define="name string:ignore">
+  <th i18n:translate="">Ignore attachments:</th>
+  <td><input type="checkbox" checked="checked"
+             name="ignore" id="ignore" value="file:content"/></td>
+  <td>&nbsp;</td>
+  <td>&nbsp;</td>
+  <td>&nbsp;</td>
+</tr>
+
 <tr tal:define="name string:title">
   <th i18n:translate="">Title:</th>
   <td metal:use-macro="search_input"></td>

Modified: tracker/instances/python-dev/html/page.html
==============================================================================
--- tracker/instances/python-dev/html/page.html	(original)
+++ tracker/instances/python-dev/html/page.html	Mon Jul 19 19:19:02 2010
@@ -42,6 +42,7 @@
        <input type="hidden" name="@sort" value="-activity" />
        <input type="hidden" name="@filter" value="status"/>
        <input type="hidden" name="@action" value="searchid"/>
+       <input type="hidden" name="ignore" value="file:content"/>
        <input tal:attributes="value python:request.form.getvalue('@search_text') or nothing;"
               class="input-text" id="search-text" name="@search_text" size="10" />
        <input type="submit" id="submit" value="search" name="submit"

Modified: tracker/roundup-src/roundup/cgi/templating.py
==============================================================================
--- tracker/roundup-src/roundup/cgi/templating.py	(original)
+++ tracker/roundup-src/roundup/cgi/templating.py	Mon Jul 19 19:19:02 2010
@@ -2658,12 +2658,32 @@
 </script>
 """%self.base
 
-    def batch(self):
+    def get_ignored(self, to_ignore='ignore'):
+        """ Return a dict of '(class, property):None' items to ignore
+
+        'ignore' is built from a form value, its name is passed as to_ignore
+        and it follows a 'class1:prop1,class2:prop2' format
+        """
+        ignore = {}
+        if not to_ignore or to_ignore not in self.form:
+            return ignore
+        for clprop in self.form[to_ignore].value.split(','):
+            clprop = clprop.strip()
+            if clprop.count(':') != 1:
+                continue
+            klass, prop = [kp.strip() for kp in clprop.split(':')]
+            if not klass or not prop:
+                continue
+            ignore[(klass, prop)] = None
+        return ignore
+
+    def batch(self, to_ignore='ignore'):
         """ Return a batch object for results from the "current search"
         """
         filterspec = self.filterspec
         sort = self.sort
         group = self.group
+        ignore = self.get_ignored(to_ignore)
 
         # get the list of ids we're batching over
         klass = self.client.db.getclass(self.classname)
@@ -2672,7 +2692,7 @@
                 [w.upper().encode("utf-8", "replace") for w in re.findall(
                     r'(?u)\b\w{2,25}\b',
                     unicode(self.search_text, "utf-8", "replace")
-                )], klass)
+                )], klass, ignore)
         else:
             matches = None
 

From python-checkins at python.org  Mon Jul 19 19:35:50 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Mon, 19 Jul 2010 19:35:50 +0200 (CEST)
Subject: [Python-checkins] r82977 - in python/branches/release27-maint/Lib:
	test/test_urllib.py urllib.py
Message-ID: <20100719173550.5639ACA1C@mail.python.org>

Author: senthil.kumaran
Date: Mon Jul 19 19:35:50 2010
New Revision: 82977

Log:
Fix Issue9301 - urllib.quote(None) to raise TypeError



Modified:
   python/branches/release27-maint/Lib/test/test_urllib.py
   python/branches/release27-maint/Lib/urllib.py

Modified: python/branches/release27-maint/Lib/test/test_urllib.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_urllib.py	(original)
+++ python/branches/release27-maint/Lib/test/test_urllib.py	Mon Jul 19 19:35:50 2010
@@ -413,6 +413,7 @@
                          "using quote(): %s != %s" % (expected, result))
         self.assertEqual(expected, result,
                          "using quote_plus(): %s != %s" % (expected, result))
+        self.assertRaises(TypeError, urllib.quote, None)
 
     def test_quoting_space(self):
         # Make sure quote() and quote_plus() handle spaces as specified in

Modified: python/branches/release27-maint/Lib/urllib.py
==============================================================================
--- python/branches/release27-maint/Lib/urllib.py	(original)
+++ python/branches/release27-maint/Lib/urllib.py	Mon Jul 19 19:35:50 2010
@@ -1223,6 +1223,8 @@
     """
     # fastpath
     if not s:
+        if s is None:
+            raise TypeError('None object cannot be quoted')
         return s
 
     if encoding is not None or isinstance(s, unicode):

From python-checkins at python.org  Mon Jul 19 19:58:26 2010
From: python-checkins at python.org (stefan.krah)
Date: Mon, 19 Jul 2010 19:58:26 +0200 (CEST)
Subject: [Python-checkins] r82978 - in python/branches/py3k:
	Lib/test/test_unicode.py Objects/unicodeobject.c
Message-ID: <20100719175826.DBA69EE99A@mail.python.org>

Author: stefan.krah
Date: Mon Jul 19 19:58:26 2010
New Revision: 82978

Log:
Sub-issue of #9036: Fix incorrect use of Py_CHARMASK.



Modified:
   python/branches/py3k/Lib/test/test_unicode.py
   python/branches/py3k/Objects/unicodeobject.c

Modified: python/branches/py3k/Lib/test/test_unicode.py
==============================================================================
--- python/branches/py3k/Lib/test/test_unicode.py	(original)
+++ python/branches/py3k/Lib/test/test_unicode.py	Mon Jul 19 19:58:26 2010
@@ -757,6 +757,7 @@
         self.assertRaises(OverflowError, "%c".__mod__, (0x110000,))
         self.assertEqual('%c' % '\U00021483', '\U00021483')
         self.assertRaises(TypeError, "%c".__mod__, "aa")
+        self.assertRaises(ValueError, "%.1\u1032f".__mod__, (1.0/3))
 
         # formatting jobs delegated from the string implementation:
         self.assertEqual('...%(foo)s...' % {'foo':"abc"}, '...abc...')

Modified: python/branches/py3k/Objects/unicodeobject.c
==============================================================================
--- python/branches/py3k/Objects/unicodeobject.c	(original)
+++ python/branches/py3k/Objects/unicodeobject.c	Mon Jul 19 19:58:26 2010
@@ -9242,7 +9242,7 @@
                 else if (c >= '0' && c <= '9') {
                     prec = c - '0';
                     while (--fmtcnt >= 0) {
-                        c = Py_CHARMASK(*fmt++);
+                        c = *fmt++;
                         if (c < '0' || c > '9')
                             break;
                         if ((prec*10) / 10 != prec) {

From python-checkins at python.org  Mon Jul 19 20:01:13 2010
From: python-checkins at python.org (stefan.krah)
Date: Mon, 19 Jul 2010 20:01:13 +0200 (CEST)
Subject: [Python-checkins] r82979 - in python/branches/release31-maint:
	Lib/test/test_unicode.py Objects/unicodeobject.c
Message-ID: <20100719180113.58480EE9BC@mail.python.org>

Author: stefan.krah
Date: Mon Jul 19 20:01:13 2010
New Revision: 82979

Log:
Merged revisions 82978 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82978 | stefan.krah | 2010-07-19 19:58:26 +0200 (Mon, 19 Jul 2010) | 3 lines
  
  Sub-issue of #9036: Fix incorrect use of Py_CHARMASK.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/test/test_unicode.py
   python/branches/release31-maint/Objects/unicodeobject.c

Modified: python/branches/release31-maint/Lib/test/test_unicode.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_unicode.py	(original)
+++ python/branches/release31-maint/Lib/test/test_unicode.py	Mon Jul 19 20:01:13 2010
@@ -763,6 +763,7 @@
         self.assertRaises(OverflowError, "%c".__mod__, (0x110000,))
         self.assertEqual('%c' % '\U00021483', '\U00021483')
         self.assertRaises(TypeError, "%c".__mod__, "aa")
+        self.assertRaises(ValueError, "%.1\u1032f".__mod__, (1.0/3))
 
         # formatting jobs delegated from the string implementation:
         self.assertEqual('...%(foo)s...' % {'foo':"abc"}, '...abc...')

Modified: python/branches/release31-maint/Objects/unicodeobject.c
==============================================================================
--- python/branches/release31-maint/Objects/unicodeobject.c	(original)
+++ python/branches/release31-maint/Objects/unicodeobject.c	Mon Jul 19 20:01:13 2010
@@ -9304,7 +9304,7 @@
                 else if (c >= '0' && c <= '9') {
                     prec = c - '0';
                     while (--fmtcnt >= 0) {
-                        c = Py_CHARMASK(*fmt++);
+                        c = *fmt++;
                         if (c < '0' || c > '9')
                             break;
                         if ((prec*10) / 10 != prec) {

From python-checkins at python.org  Mon Jul 19 20:06:46 2010
From: python-checkins at python.org (stefan.krah)
Date: Mon, 19 Jul 2010 20:06:46 +0200 (CEST)
Subject: [Python-checkins] r82980 - in python/branches/release27-maint:
	Lib/test/test_unicode.py Objects/unicodeobject.c
Message-ID: <20100719180646.AF224DC80@mail.python.org>

Author: stefan.krah
Date: Mon Jul 19 20:06:46 2010
New Revision: 82980

Log:
Sub-issue of #9036: Fix incorrect use of Py_CHARMASK.



Modified:
   python/branches/release27-maint/Lib/test/test_unicode.py
   python/branches/release27-maint/Objects/unicodeobject.c

Modified: python/branches/release27-maint/Lib/test/test_unicode.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_unicode.py	(original)
+++ python/branches/release27-maint/Lib/test/test_unicode.py	Mon Jul 19 20:06:46 2010
@@ -394,6 +394,7 @@
 
         self.assertEqual(u'%c' % 0x1234, u'\u1234')
         self.assertRaises(OverflowError, u"%c".__mod__, (sys.maxunicode+1,))
+        self.assertRaises(ValueError, u"%.1\u1032f".__mod__, (1.0/3))
 
         for num in range(0x00,0x80):
             char = chr(num)

Modified: python/branches/release27-maint/Objects/unicodeobject.c
==============================================================================
--- python/branches/release27-maint/Objects/unicodeobject.c	(original)
+++ python/branches/release27-maint/Objects/unicodeobject.c	Mon Jul 19 20:06:46 2010
@@ -8417,7 +8417,7 @@
                 else if (c >= '0' && c <= '9') {
                     prec = c - '0';
                     while (--fmtcnt >= 0) {
-                        c = Py_CHARMASK(*fmt++);
+                        c = *fmt++;
                         if (c < '0' || c > '9')
                             break;
                         if ((prec*10) / 10 != prec) {

From python-checkins at python.org  Mon Jul 19 20:10:42 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Mon, 19 Jul 2010 20:10:42 +0200 (CEST)
Subject: [Python-checkins] r82981 -
	python/branches/release27-maint/Doc/library/stdtypes.rst
Message-ID: <20100719181042.590A0F44A@mail.python.org>

Author: antoine.pitrou
Date: Mon Jul 19 20:10:42 2010
New Revision: 82981

Log:
Issue #9304: fix example in the 2.x memoryview documentation.



Modified:
   python/branches/release27-maint/Doc/library/stdtypes.rst

Modified: python/branches/release27-maint/Doc/library/stdtypes.rst
==============================================================================
--- python/branches/release27-maint/Doc/library/stdtypes.rst	(original)
+++ python/branches/release27-maint/Doc/library/stdtypes.rst	Mon Jul 19 20:10:42 2010
@@ -2574,7 +2574,7 @@
       'g'
       >>> v[1:4]
       <memory at 0x77ab28>
-      >>> str(v[1:4])
+      >>> v[1:4].tobytes()
       'bce'
 
    If the object the memoryview is over supports changing its data, the
@@ -2612,7 +2612,7 @@
 
       Return the data in the buffer as a list of integers. ::
 
-         >>> memoryview(b'abc').tolist()
+         >>> memoryview("abc").tolist()
          [97, 98, 99]
 
    There are also several readonly attributes available:

From python-checkins at python.org  Mon Jul 19 20:15:41 2010
From: python-checkins at python.org (stefan.krah)
Date: Mon, 19 Jul 2010 20:15:41 +0200 (CEST)
Subject: [Python-checkins] r82982 -
	python/branches/release26-maint/Lib/test/test_posix.py
Message-ID: <20100719181541.752C4EE9AE@mail.python.org>

Author: stefan.krah
Date: Mon Jul 19 20:15:41 2010
New Revision: 82982

Log:
Add missing import of errno.



Modified:
   python/branches/release26-maint/Lib/test/test_posix.py

Modified: python/branches/release26-maint/Lib/test/test_posix.py
==============================================================================
--- python/branches/release26-maint/Lib/test/test_posix.py	(original)
+++ python/branches/release26-maint/Lib/test/test_posix.py	Mon Jul 19 20:15:41 2010
@@ -7,6 +7,7 @@
 except ImportError:
     raise test_support.TestSkipped, "posix is not available"
 
+import errno
 import time
 import os
 import pwd

From python-checkins at python.org  Mon Jul 19 20:17:19 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Mon, 19 Jul 2010 20:17:19 +0200 (CEST)
Subject: [Python-checkins] r82983 - in python/branches/py3k/Lib:
	test/test_urllib.py urllib/parse.py
Message-ID: <20100719181719.4588BEE9C4@mail.python.org>

Author: senthil.kumaran
Date: Mon Jul 19 20:17:19 2010
New Revision: 82983

Log:
Fix Issue9301 - urllib.parse.unquote and unquote_to_byte to raise TypeError for None.



Modified:
   python/branches/py3k/Lib/test/test_urllib.py
   python/branches/py3k/Lib/urllib/parse.py

Modified: python/branches/py3k/Lib/test/test_urllib.py
==============================================================================
--- python/branches/py3k/Lib/test/test_urllib.py	(original)
+++ python/branches/py3k/Lib/test/test_urllib.py	Mon Jul 19 20:17:19 2010
@@ -554,6 +554,7 @@
         self.assertEqual(result.count('%'), 1,
                          "using unquote(): not all characters escaped: "
                          "%s" % result)
+        self.assertRaises(TypeError, urllib.parse.unquote, None)
 
     def test_unquoting_badpercent(self):
         # Test unquoting on bad percent-escapes
@@ -589,6 +590,8 @@
         self.assertEqual(expect, result, "using unquote_to_bytes(): %r != %r"
                          % (expect, result))
 
+        self.assertRaises(TypeError, urllib.parse.unquote_to_bytes, None)
+
     def test_unquoting_mixed_case(self):
         # Test unquoting on mixed-case hex digits in the percent-escapes
         given = '%Ab%eA'

Modified: python/branches/py3k/Lib/urllib/parse.py
==============================================================================
--- python/branches/py3k/Lib/urllib/parse.py	(original)
+++ python/branches/py3k/Lib/urllib/parse.py	Mon Jul 19 20:17:19 2010
@@ -314,6 +314,8 @@
     # Note: strings are encoded as UTF-8. This is only an issue if it contains
     # unescaped non-ASCII characters, which URIs should not.
     if not string:
+        if string is None:
+            raise TypeError('None object is invalid for unquote_to_bytes()')
         return b''
     if isinstance(string, str):
         string = string.encode('utf-8')
@@ -339,6 +341,8 @@
     unquote('abc%20def') -> 'abc def'.
     """
     if not string:
+        if string is None:
+            raise TypeError('None object is invalid for unquote() function.')
         return string
     res = string.split('%')
     if len(res) == 1:

From python-checkins at python.org  Mon Jul 19 20:24:18 2010
From: python-checkins at python.org (stefan.krah)
Date: Mon, 19 Jul 2010 20:24:18 +0200 (CEST)
Subject: [Python-checkins] r82984 - in python/branches/release26-maint:
	Lib/test/test_unicode.py Objects/unicodeobject.c
Message-ID: <20100719182418.984D2EE9AE@mail.python.org>

Author: stefan.krah
Date: Mon Jul 19 20:24:18 2010
New Revision: 82984

Log:
Merged revisions 82980 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

........
  r82980 | stefan.krah | 2010-07-19 20:06:46 +0200 (Mon, 19 Jul 2010) | 3 lines
  
  Sub-issue of #9036: Fix incorrect use of Py_CHARMASK.
........


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Lib/test/test_unicode.py
   python/branches/release26-maint/Objects/unicodeobject.c

Modified: python/branches/release26-maint/Lib/test/test_unicode.py
==============================================================================
--- python/branches/release26-maint/Lib/test/test_unicode.py	(original)
+++ python/branches/release26-maint/Lib/test/test_unicode.py	Mon Jul 19 20:24:18 2010
@@ -392,6 +392,7 @@
 
         self.assertEqual(u'%c' % 0x1234, u'\u1234')
         self.assertRaises(OverflowError, u"%c".__mod__, (sys.maxunicode+1,))
+        self.assertRaises(ValueError, u"%.1\u1032f".__mod__, (1.0/3))
 
         for num in range(0x00,0x80):
             char = chr(num)

Modified: python/branches/release26-maint/Objects/unicodeobject.c
==============================================================================
--- python/branches/release26-maint/Objects/unicodeobject.c	(original)
+++ python/branches/release26-maint/Objects/unicodeobject.c	Mon Jul 19 20:24:18 2010
@@ -8607,7 +8607,7 @@
                 else if (c >= '0' && c <= '9') {
                     prec = c - '0';
                     while (--fmtcnt >= 0) {
-                        c = Py_CHARMASK(*fmt++);
+                        c = *fmt++;
                         if (c < '0' || c > '9')
                             break;
                         if ((prec*10) / 10 != prec) {

From python-checkins at python.org  Tue Jul 20 01:17:22 2010
From: python-checkins at python.org (gregory.p.smith)
Date: Tue, 20 Jul 2010 01:17:22 +0200 (CEST)
Subject: [Python-checkins] r82985 - in python/branches/py3k:
	Lib/http/cookiejar.py Lib/test/test_http_cookiejar.py Misc/NEWS
Message-ID: <20100719231722.3EB7FF10E@mail.python.org>

Author: gregory.p.smith
Date: Tue Jul 20 01:17:22 2010
New Revision: 82985

Log:
Fixes Issue #3704: http.cookiejar was not properly handling URLs with a / in
the parameters.  (This is jjlee's issue3704.patch ported to py3k)


Modified:
   python/branches/py3k/Lib/http/cookiejar.py
   python/branches/py3k/Lib/test/test_http_cookiejar.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/http/cookiejar.py
==============================================================================
--- python/branches/py3k/Lib/http/cookiejar.py	(original)
+++ python/branches/py3k/Lib/http/cookiejar.py	Tue Jul 20 01:17:22 2010
@@ -609,17 +609,14 @@
     return req_host, erhn
 
 def request_path(request):
-    """request-URI, as defined by RFC 2965."""
+    """Path component of request-URI, as defined by RFC 2965."""
     url = request.get_full_url()
-    path, parameters, query, frag = urllib.parse.urlparse(url)[2:]
-    if parameters:
-        path = "%s;%s" % (path, parameters)
-    path = escape_path(path)
-    req_path = urllib.parse.urlunparse(("", "", path, "", query, frag))
-    if not req_path.startswith("/"):
+    parts = urllib.parse.urlsplit(url)
+    path = escape_path(parts.path)
+    if not path.startswith("/"):
         # fix bad RFC 2396 absoluteURI
-        req_path = "/"+req_path
-    return req_path
+        path = "/" + path
+    return path
 
 def request_port(request):
     host = request.get_host()

Modified: python/branches/py3k/Lib/test/test_http_cookiejar.py
==============================================================================
--- python/branches/py3k/Lib/test/test_http_cookiejar.py	(original)
+++ python/branches/py3k/Lib/test/test_http_cookiejar.py	Tue Jul 20 01:17:22 2010
@@ -1,18 +1,21 @@
 """Tests for http/cookiejar.py."""
 
-import re, os, time, urllib.request
-from unittest import TestCase
+import os
+import re
+import test.support
+import time
+import unittest
+import urllib.request
+
+from http.cookiejar import (time2isoz, http2time, time2netscape,
+     parse_ns_headers, join_header_words, split_header_words, Cookie,
+     CookieJar, DefaultCookiePolicy, LWPCookieJar, MozillaCookieJar,
+     LoadError, lwp_cookie_str, DEFAULT_HTTP_PORT, escape_path,
+     reach, is_HDN, domain_match, user_domain_match, request_path,
+     request_port, request_host)
 
-from test import support
 
-from http.cookiejar import time2isoz, http2time, time2netscape, \
-     parse_ns_headers, join_header_words, split_header_words, Cookie, \
-     CookieJar, DefaultCookiePolicy, LWPCookieJar, MozillaCookieJar, \
-     LoadError, lwp_cookie_str, DEFAULT_HTTP_PORT, escape_path, \
-     reach, is_HDN, domain_match, user_domain_match, request_path, \
-     request_port, request_host
-
-class DateTimeTests(TestCase):
+class DateTimeTests(unittest.TestCase):
 
     def test_time2isoz(self):
         base = 1019227000
@@ -96,7 +99,7 @@
                          )
 
 
-class HeaderTests(TestCase):
+class HeaderTests(unittest.TestCase):
 
     def test_parse_ns_headers(self):
         # quotes should be stripped
@@ -228,10 +231,10 @@
     return cookie_hdr
 
 
-class FileCookieJarTests(TestCase):
+class FileCookieJarTests(unittest.TestCase):
     def test_lwp_valueless_cookie(self):
         # cookies with no value should be saved and loaded consistently
-        filename = support.TESTFN
+        filename = test.support.TESTFN
         c = LWPCookieJar()
         interact_netscape(c, "http://www.acme.com/", 'boo')
         self.assertEqual(c._cookies["www.acme.com"]["/"]["boo"].value, None)
@@ -246,7 +249,7 @@
 
     def test_bad_magic(self):
         # IOErrors (eg. file doesn't exist) are allowed to propagate
-        filename = support.TESTFN
+        filename = test.support.TESTFN
         for cookiejar_class in LWPCookieJar, MozillaCookieJar:
             c = cookiejar_class()
             try:
@@ -269,7 +272,7 @@
             try: os.unlink(filename)
             except OSError: pass
 
-class CookieTests(TestCase):
+class CookieTests(unittest.TestCase):
     # XXX
     # Get rid of string comparisons where not actually testing str / repr.
     # .clear() etc.
@@ -349,7 +352,7 @@
     def test_missing_value(self):
         # missing = sign in Cookie: header is regarded by Mozilla as a missing
         # name, and by http.cookiejar as a missing value
-        filename = support.TESTFN
+        filename = test.support.TESTFN
         c = MozillaCookieJar(filename)
         interact_netscape(c, "http://www.acme.com/", 'eggs')
         interact_netscape(c, "http://www.acme.com/", '"spam"; path=/foo/')
@@ -534,6 +537,16 @@
         interact_netscape(c, "http://www.acme.com/blah/rhubarb/", 'eggs="bar"')
         self.assertIn("/blah/rhubarb", c._cookies["www.acme.com"])
 
+    def test_default_path_with_query(self):
+        cj = CookieJar()
+        uri = "http://example.com/?spam/eggs"
+        value = 'eggs="bar"'
+        interact_netscape(cj, uri, value)
+        # Default path does not include query, so is "/", not "/?spam".
+        self.assertIn("/", cj._cookies["example.com"])
+        # Cookie is sent back to the same URI.
+        self.assertEquals(interact_netscape(cj, uri), value)
+
     def test_escape_path(self):
         cases = [
             # quoted safe
@@ -562,16 +575,15 @@
     def test_request_path(self):
         # with parameters
         req = urllib.request.Request(
-            "http://www.example.com/rheum/rhaponicum;"
+            "http://www.example.com/rheum/rhaponticum;"
             "foo=bar;sing=song?apples=pears&spam=eggs#ni")
-        self.assertEquals(request_path(req), "/rheum/rhaponicum;"
-                     "foo=bar;sing=song?apples=pears&spam=eggs#ni")
+        self.assertEquals(request_path(req),
+                          "/rheum/rhaponticum;foo=bar;sing=song")
         # without parameters
         req = urllib.request.Request(
-            "http://www.example.com/rheum/rhaponicum?"
+            "http://www.example.com/rheum/rhaponticum?"
             "apples=pears&spam=eggs#ni")
-        self.assertEquals(request_path(req), "/rheum/rhaponicum?"
-                     "apples=pears&spam=eggs#ni")
+        self.assertEquals(request_path(req), "/rheum/rhaponticum")
         # missing final slash
         req = urllib.request.Request("http://www.example.com")
         self.assertEquals(request_path(req), "/")
@@ -1045,7 +1057,7 @@
         self.assertTrue(cookie.expires is None)
 
 
-class LWPCookieTests(TestCase):
+class LWPCookieTests(unittest.TestCase):
     # Tests taken from libwww-perl, with a few modifications and additions.
 
     def test_netscape_example_1(self):
@@ -1435,7 +1447,7 @@
         self.assertEquals(len(c), 6)
 
         # save and restore
-        filename = support.TESTFN
+        filename = test.support.TESTFN
 
         try:
             c.save(filename, ignore_discard=True)
@@ -1475,7 +1487,7 @@
         # Save / load Mozilla/Netscape cookie file format.
         year_plus_one = time.localtime()[0] + 1
 
-        filename = support.TESTFN
+        filename = test.support.TESTFN
 
         c = MozillaCookieJar(filename,
                              policy=DefaultCookiePolicy(rfc2965=True))
@@ -1637,7 +1649,7 @@
 
 
 def test_main(verbose=None):
-    support.run_unittest(
+    test.support.run_unittest(
         DateTimeTests,
         HeaderTests,
         CookieTests,

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Tue Jul 20 01:17:22 2010
@@ -473,6 +473,9 @@
 Library
 -------
 
+- Issue #3704: http.cookiejar was not properly handling URLs with a / in the
+  parameters.
+
 - Issue #9268: ``pickletools.dis()`` now has an optional *annotate*
   argument which controls printing of opcode descriptions in ``dis()``
   output.

From solipsis at pitrou.net  Tue Jul 20 01:24:39 2010
From: solipsis at pitrou.net (solipsis at pitrou.net)
Date: Tue, 20 Jul 2010 01:24:39 +0200 (CEST)
Subject: [Python-checkins] Daily py3k reference leaks (r82983): sum=0
Message-ID: <20100719232439.6DAE21770A@ns6635.ovh.net>

py3k results for svn r82983 (hg cset 77f02ab68a8a)
--------------------------------------------------



Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/antoine/py3k/refleaks/reflogfbctIu', '-x']

From python-checkins at python.org  Tue Jul 20 10:07:44 2010
From: python-checkins at python.org (stefan.krah)
Date: Tue, 20 Jul 2010 10:07:44 +0200 (CEST)
Subject: [Python-checkins] r82986 -
	python/branches/py3k-cdecimal/Lib/test/test_curses.py
Message-ID: <20100720080744.5ACBEF022@mail.python.org>

Author: stefan.krah
Date: Tue Jul 20 10:07:44 2010
New Revision: 82986

Log:
Try to fix Issue #8433.

Modified:
   python/branches/py3k-cdecimal/Lib/test/test_curses.py

Modified: python/branches/py3k-cdecimal/Lib/test/test_curses.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_curses.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_curses.py	Tue Jul 20 10:07:44 2010
@@ -218,7 +218,7 @@
         (availmask, oldmask) = curses.mousemask(curses.BUTTON1_PRESSED)
         # availmask indicates that mouse stuff not available.
         if availmask != 0:
-            curses.mouseinterval(10)
+            curses.mouseinterval(10000)
             # just verify these don't cause errors
             m = curses.getmouse()
             curses.ungetmouse(*m)

From python-checkins at python.org  Tue Jul 20 12:08:40 2010
From: python-checkins at python.org (stefan.krah)
Date: Tue, 20 Jul 2010 12:08:40 +0200 (CEST)
Subject: [Python-checkins] r82987 -
	python/branches/py3k-cdecimal/Lib/test/test_curses.py
Message-ID: <20100720100840.6330BD2AE@mail.python.org>

Author: stefan.krah
Date: Tue Jul 20 12:08:40 2010
New Revision: 82987

Log:
Skip issue 8433

Modified:
   python/branches/py3k-cdecimal/Lib/test/test_curses.py

Modified: python/branches/py3k-cdecimal/Lib/test/test_curses.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_curses.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_curses.py	Tue Jul 20 12:08:40 2010
@@ -16,13 +16,14 @@
 # option.  If not available, nothing after this line will be executed.
 
 import unittest
-from test.support import requires, import_module
+from test.support import requires, import_module, verbose
 requires('curses')
 
 # If either of these don't exist, skip the tests.
 curses = import_module('curses')
 curses.panel = import_module('curses.panel')
 
+HAVE_ISSUE_8433 = False
 
 # XXX: if newterm was supported we could use it instead of initscr and not exit
 term = os.environ.get('TERM')
@@ -215,13 +216,17 @@
         curses.has_key(13)
 
     if hasattr(curses, 'getmouse'):
+        global HAVE_ISSUE_8433
         (availmask, oldmask) = curses.mousemask(curses.BUTTON1_PRESSED)
         # availmask indicates that mouse stuff not available.
         if availmask != 0:
-            curses.mouseinterval(10000)
+            curses.mouseinterval(10)
             # just verify these don't cause errors
-            m = curses.getmouse()
-            curses.ungetmouse(*m)
+            try:
+                m = curses.getmouse()
+                curses.ungetmouse(*m)
+            except curses.error:
+                HAVE_ISSUE_8433 = True
 
     if hasattr(curses, 'is_term_resized'):
         curses.is_term_resized(*stdscr.getmaxyx())
@@ -274,6 +279,8 @@
         test_issue6243(stdscr)
     finally:
         curses.resetty()
+        if HAVE_ISSUE_8433:
+            raise unittest.SkipTest("test getmouse() skipped: see issue 8433")
 
 def test_main():
     if not sys.stdout.isatty():

From python-checkins at python.org  Tue Jul 20 12:13:43 2010
From: python-checkins at python.org (stefan.krah)
Date: Tue, 20 Jul 2010 12:13:43 +0200 (CEST)
Subject: [Python-checkins] r82988 -
	python/branches/py3k-cdecimal/Lib/test/test_curses.py
Message-ID: <20100720101343.0CF2ED422@mail.python.org>

Author: stefan.krah
Date: Tue Jul 20 12:13:42 2010
New Revision: 82988

Log:
Better place for the skip in issue 8433

Modified:
   python/branches/py3k-cdecimal/Lib/test/test_curses.py

Modified: python/branches/py3k-cdecimal/Lib/test/test_curses.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_curses.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_curses.py	Tue Jul 20 12:13:42 2010
@@ -279,8 +279,6 @@
         test_issue6243(stdscr)
     finally:
         curses.resetty()
-        if HAVE_ISSUE_8433:
-            raise unittest.SkipTest("test getmouse() skipped: see issue 8433")
 
 def test_main():
     if not sys.stdout.isatty():
@@ -294,6 +292,8 @@
     finally:
         curses.endwin()
     unit_tests()
+    if HAVE_ISSUE_8433:
+        raise unittest.SkipTest("test getmouse() skipped: see issue 8433")
 
 if __name__ == '__main__':
     curses.wrapper(main)

From python-checkins at python.org  Tue Jul 20 15:17:13 2010
From: python-checkins at python.org (nick.coghlan)
Date: Tue, 20 Jul 2010 15:17:13 +0200 (CEST)
Subject: [Python-checkins] r82991 - in peps/trunk: pep-3150.txt pep0/pep.py
Message-ID: <20100720131713.BB5DAF300@mail.python.org>

Author: nick.coghlan
Date: Tue Jul 20 15:17:13 2010
New Revision: 82991

Log:
Make error message slightly more helpful in diagnosing missing angle brackets around an email address in the PEP header

Added:
   peps/trunk/pep-3150.txt   (contents, props changed)
Modified:
   peps/trunk/pep0/pep.py

Added: peps/trunk/pep-3150.txt
==============================================================================
--- (empty file)
+++ peps/trunk/pep-3150.txt	Tue Jul 20 15:17:13 2010
@@ -0,0 +1,274 @@
+PEP: 3150
+Title: Statement local namespaces (aka "where" clause)
+Version: $Revision$
+Last-Modified: $Date$
+Author: Nick Coghlan <ncoghlan at gmail.com>
+Status: Deferred
+Type: Standards Track
+Content-Type: text/x-rst
+Created: 2010-07-09
+Python-Version: 3.3
+Post-History: 2010-07-14
+Resolution: TBD
+
+
+Abstract
+========
+
+A recurring proposal on python-ideas is the addition of some form of
+statement local namespace.
+
+This PEP is intended to serve as a focal point for those ideas, so we
+can hopefully avoid retreading the same ground a couple of times a
+year. Even if the proposal is never accepted having a PEP to point
+people to can be valuable (e.g. having PEP 315 helps greatly in avoiding
+endless rehashing of loop-and-a-half arguments).
+
+The ideas in this PEP are just a sketch of a way this concept might work.
+They avoid some pitfalls that have been encountered in the past, but
+have not themselves been subject to the test of implementation.
+
+
+PEP Deferral
+============
+
+This PEP is currently deferred at least until the language moratorium
+(PEP 3003) is officially lifted by Guido. Even after that, it will
+require input from at least the four major Python implementations
+(CPython, PyPy, Jython, IronPython) on the feasibility of implementing
+the proposed semantics to get it moving again.
+
+Proposal
+========
+
+This PEP proposes the addition of an optional "where" clause to the
+syntax for simple statements which may contain an expression. The
+current list of simple statements that would be affected by this
+addition is as follows:
+
+* expression statement
+* assignment statement
+* augmented assignment statement
+* del statement
+* return statement
+* yield statement
+* raise statement
+* assert statement
+
+The ``where`` clause would allow subexpressions to be referenced by
+name in the header line, with the actual definitions following in
+the indented clause. As a simple example::
+
+ c = sqrt(a*a + b*b) where:
+    a = retrieve_a()
+    b = retrieve_b()
+
+Torture Test
+============
+
+An implementation of this PEP must support execution of the following
+code at module, class and function scope::
+
+  b = {}
+  a = b[f(a)] = x where:
+    x = 42
+    def f(x):
+      return x
+  assert "x" not in locals()
+  assert "f" not in locals()
+  assert a == 42
+  assert d[42] == 42 where:
+    d = b
+  assert "d" not in locals()
+
+Most naive implementations will choke on the first complex assignment,
+while less naive but still broken implementations will fail when
+the torture test is executed at class scope.
+
+And yes, that's a perfectly well-defined assignment statement. Insane,
+I agree, but legal::
+
+  >>> def f(x): return x
+  ... 
+  >>> x = 42
+  >>> b = {}
+  >>> a = b[f(a)] = x
+  >>> a
+  42
+  >>> b
+  {42: 42}
+
+
+Syntax Change
+=============
+
+Current::
+
+  expr_stmt: testlist_star_expr (augassign (yield_expr|testlist) |
+               ('=' (yield_expr|testlist_star_expr))*)
+  del_stmt: 'del' exprlist
+  return_stmt: 'return' [testlist]
+  yield_stmt: yield_expr
+  raise_stmt: 'raise' [test ['from' test]]
+  assert_stmt: 'assert' test [',' test]
+
+
+New::
+
+  expr_stmt: testlist_star_expr (augassign (yield_expr|testlist) |
+               ('=' (yield_expr|testlist_star_expr))*) [where_clause]
+  del_stmt: 'del' exprlist [where_clause]
+  return_stmt: 'return' [testlist] [where_clause]
+  yield_stmt: yield_expr [where_clause]
+  raise_stmt: 'raise' [test ['from' test]] [where_clause]
+  assert_stmt: 'assert' test [',' test] [where_clause]
+  where_clause: "where" ":" suite
+
+(Note that expr_stmt in the grammar covers assignment and augmented
+assignment in addition to simple expression statements)
+
+The new clause is added as an optional element of the existing statements
+rather than as a new kind of compound statement in order to avoid creating
+an ambiguity in the grammar. It is applied only to the specific elements
+listed so that nonsense like the following is disallowed::
+
+   pass where:
+       a = b = 1
+
+However, even this is inadequate, as it creates problems for the definition
+of simple_stmt (which allows chaining of multiple single line statements
+with ";" rather than "\n").
+
+So the above syntax change should instead be taken as a statement of intent.
+Any actual proposal would need to resolve the simple_stmt parsing problem
+before it could be seriously considered. This would likely require a
+non-trivial restructuring of the grammar, breaking up small_stmt and
+flow_stmt to separate the statements that potentially contain arbitrary
+subexpressions and then allowing a single one of those statements with
+a ``where`` clause at the simple_stmt level. Something along the lines of::
+
+  simple_stmt: small_stmt (';' small_stmt)* [';'] NEWLINE
+  where_stmt: subexpr_stmt (where_clause | (';' small_stmt)* [';']) NEWLINE
+  subexpr_stmt: expr_stmt | del_stmt | flow_subexpr_stmt | assert_stmt
+  small_stmt: (pass_stmt | flow_stmt | import_stmt |
+               global_stmt | nonlocal_stmt)
+  flow_stmt: break_stmt | continue_stmt
+  flow_subexpr_stmt: return_stmt | raise_stmt | yield_stmt
+  where_clause: "where" ":" suite
+
+
+For reference, here are the current definitions at that level::
+
+  simple_stmt: small_stmt (';' small_stmt)* [';'] NEWLINE
+  small_stmt: (expr_stmt | del_stmt | pass_stmt | flow_stmt |
+               import_stmt | global_stmt | nonlocal_stmt | assert_stmt)
+  flow_stmt: break_stmt | continue_stmt | return_stmt | raise_stmt | yield_stmt
+
+
+Common Objections
+=================
+
+* Two Ways To Do It: a lot of code may now be written with values
+  defined either before the expression where they are used or
+  afterwards in a ``where`` clause, creating two ways to do it,
+  without an obvious way of choosing between them.
+
+* Out of Order Execution: the ``where`` clause makes execution
+  jump around a little strangely, as the body of the ``where``
+  clause is executed before the simple statement in the clause
+  header. The closest any other part of Python comes to this
+  before is the out of order evaluation in conditional
+  expressions.
+  
+
+Possible Additions
+==================
+
+* The current proposal allows the addition of a ``where`` clause only
+  for simple statements. Extending the idea to allow the use of
+  compound statements would be quite possible, but doing so raises
+  serious readability concerns (as values defined in the ``where``
+  clause may be used well before they are defined, exactly the kind
+  of readability trap that decorators were designed to eliminate)
+
+* Currently only the outermost clause of comprehensions and generator
+  expressions can reference the surrounding namespace when executed
+  at class level. If this proposal is implemented successfully, the
+  associated namespace semantics could allow that restriction to be
+  lifted. There would be backwards compatibility implications in doing
+  so as existing code may be relying on the behaviour of ignoring
+  class level variables.
+
+Possible Implementation Strategy
+================================
+
+AKA How Class Scopes Screw You When Attempting To Implement This
+
+The natural idea when setting out to implement this concept is to
+use an ordinary nested function scope. This doesn't work for the
+two reasons mentioned in the Torture Test section above:
+
+* Non-local variables are not your friend because they ignore class scopes
+  and (when writing back to the outer scope) aren't really on speaking
+  terms with module scopes either.
+
+* Return-based semantics struggle with complex assignment statements
+  like the one in the torture test
+
+The most promising approach is one based on symtable analysis and
+copy-in-copy-out referencing semantics to move any required name
+bindings between the inner and outer scopes. The torture test above
+would then translate to something like the following::
+
+  b = {}
+  def _anon1(b): # 'b' reference copied in
+    x = 42
+    def f(x):
+      return x
+    a = b[f(a)] = x
+    return a # 'a' reference copied out
+  a = _anon1(b)
+  assert "x" not in locals()
+  assert "f" not in locals()
+  assert a == 42
+  def _anon2(b) # 'b' reference copied in
+    d = b
+    assert d[42] == 42
+    # Nothing to copy out (not an assignment)
+  _anon2()
+  assert "d" not in locals()
+
+However, as noted in the abstract, an actual implementation of
+this idea has never been tried.
+
+
+Reference implementation
+========================
+
+None as yet. If you want a crash course in Python namespace
+semantics and code compilation, feel free to try ;)
+
+
+References
+==========
+
+.. [1] http://mail.python.org/pipermail/python-ideas/2010-June/007476.html
+
+.. [2] http://mail.python.org/pipermail/python-ideas/2010-July/007584.html
+
+
+Copyright
+=========
+
+This document has been placed in the public domain.
+
+
+
+..
+   Local Variables:
+   mode: indented-text
+   indent-tabs-mode: nil
+   sentence-end-double-space: t
+   fill-column: 70
+   coding: utf-8
+   End:

Modified: peps/trunk/pep0/pep.py
==============================================================================
--- peps/trunk/pep0/pep.py	(original)
+++ peps/trunk/pep0/pep.py	Tue Jul 20 15:17:13 2010
@@ -99,7 +99,8 @@
             if part[0].isupper():
                 break
         else:
-            raise ValueError("last name missing a capital letter")
+            raise ValueError("last name missing a capital letter: %r"
+	                                                       % name_parts)
         base = u' '.join(name_parts[index:]).lower()
         return unicodedata.normalize('NFKD', base).encode('ASCII', 'ignore')
 

From python-checkins at python.org  Tue Jul 20 15:20:51 2010
From: python-checkins at python.org (nick.coghlan)
Date: Tue, 20 Jul 2010 15:20:51 +0200 (CEST)
Subject: [Python-checkins] r82991 - svn:log
Message-ID: <20100720132051.303DFF32B@mail.python.org>

Author: nick.coghlan
Revision: 82991
Property Name: svn:log
Action: modified

Property diff:
--- old property value
+++ new property value
@@ -1 +1,3 @@
-Make error message slightly more helpful in diagnosing missing angle brackets around an email address in the PEP header
\ No newline at end of file
+Make error message slightly more helpful in diagnosing missing angle brackets around an email address in the PEP header
+
+(Also added PEP 3150, which was meant to be a separate checkin. C'est la vie)

From python-checkins at python.org  Tue Jul 20 15:24:27 2010
From: python-checkins at python.org (nick.coghlan)
Date: Tue, 20 Jul 2010 15:24:27 +0200 (CEST)
Subject: [Python-checkins] r82992 - peps/trunk/pep-3150.txt
Message-ID: <20100720132427.84F1BF32B@mail.python.org>

Author: nick.coghlan
Date: Tue Jul 20 15:24:27 2010
New Revision: 82992

Log:
Typo fix

Modified:
   peps/trunk/pep-3150.txt

Modified: peps/trunk/pep-3150.txt
==============================================================================
--- peps/trunk/pep-3150.txt	(original)
+++ peps/trunk/pep-3150.txt	Tue Jul 20 15:24:27 2010
@@ -137,7 +137,7 @@
 
 However, even this is inadequate, as it creates problems for the definition
 of simple_stmt (which allows chaining of multiple single line statements
-with ";" rather than "\n").
+with ";" rather than "\\n").
 
 So the above syntax change should instead be taken as a statement of intent.
 Any actual proposal would need to resolve the simple_stmt parsing problem

From python-checkins at python.org  Tue Jul 20 17:07:28 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Tue, 20 Jul 2010 17:07:28 +0200 (CEST)
Subject: [Python-checkins] r82993 - in tracker/instances/python-dev:
	extensions/openid_login.py lib/openid.py lib/openid2rp.py
Message-ID: <20100720150728.DC4FDF2B5@mail.python.org>

Author: martin.v.loewis
Date: Tue Jul 20 17:07:28 2010
New Revision: 82993

Log:
Rename openid.py to openid2rp.py.


Added:
   tracker/instances/python-dev/lib/openid2rp.py
      - copied unchanged from r82992, /tracker/instances/python-dev/lib/openid.py
Removed:
   tracker/instances/python-dev/lib/openid.py
Modified:
   tracker/instances/python-dev/extensions/openid_login.py

Modified: tracker/instances/python-dev/extensions/openid_login.py
==============================================================================
--- tracker/instances/python-dev/extensions/openid_login.py	(original)
+++ tracker/instances/python-dev/extensions/openid_login.py	Tue Jul 20 17:07:28 2010
@@ -1,11 +1,11 @@
-import openid, urllib, cgi, collections, calendar, time
+import openid2rp, urllib, cgi, collections, calendar, time
 from roundup.cgi.actions import Action, LoginAction, RegisterAction
 from roundup.cgi.exceptions import *
 from roundup import date, password
 
 good_providers = ['Google', 'myOpenID', 'Launchpad']
 providers = {}
-for p in openid.providers:
+for p in openid2rp.providers:
     if p[0] not in good_providers: continue
     providers[p[0]] = p
 
@@ -30,9 +30,9 @@
         if discovered:
             stypes, url, op_local = discovered
         else:
-            stypes, url, op_local = openid.discover(provider)
+            stypes, url, op_local = openid2rp.discover(provider)
         now = date.Date('.')
-        session_data = openid.associate(stypes, url)
+        session_data = openid2rp.associate(stypes, url)
         # check whether a session has expired a day ago
         sessions = self.db.openid_session.filter(None, {'expires':'to -1d'})
         if sessions:
@@ -52,23 +52,23 @@
     def authenticate(self, session, query):
         '''Authenticate an OpenID indirect response, and return the claimed ID'''
         try:
-            signed = openid.authenticate(session, query)
+            signed = openid2rp.authenticate(session, query)
         except Exception, e:
             raise ValueError, "Authentication failed: "+str(e)
-        if openid.is_op_endpoint(session.stypes):
+        if openid2rp.is_op_endpoint(session.stypes):
             # Provider-guided login: provider ought to report claimed ID
             if 'openid.claimed_id' in query:
                 claimed = query['openid.claimed_id'][0]
             else:
                 raise ValueError, 'incomplete response'
             # OpenID 11.2: verify that provider is authorized to assert ID
-            discovered = openid.discover(claimed)
+            discovered = openid2rp.discover(claimed)
             if not discovered or discovered[1] != session.url:
                 raise ValueError, "Provider %s is not authorized to make assertions about %s" % (session.url, claimed)
         else:
             # User entered claimed ID, stored in session object
             claimed = session.provider_id
-            if not openid.is_compat_1x(session.stypes):
+            if not openid2rp.is_compat_1x(session.stypes):
                 # can only check correct claimed ID for OpenID 2.0
                 if 'openid.claimed_id' not in query or claimed != query['openid.claimed_id'][0]:
                     # assertion is not about an ID, or about a different ID; refuse to accept
@@ -97,17 +97,17 @@
         else:
             return LoginAction.handle(self)
         # Login an OpenID
-        type, claimed = openid.normalize_uri(username)
+        type, claimed = openid2rp.normalize_uri(username)
         if type == 'xri':
             raise ValueError, "XRIs are not supported"
-        discovered = openid.discover(claimed)
+        discovered = openid2rp.discover(claimed)
         if not discovered:
             raise ValueError, "OpenID provider discovery failed"
         stypes, url, op_local = discovered
         session = self.get_session(claimed, discovered) # one session per claimed id
         realm = self.base+"?@action=openid_return"
         return_to = realm + "&__came_from=%s" % urllib.quote(self.client.path)
-        url = openid.request_authentication(session.stypes, session.url,
+        url = openid2rp.request_authentication(session.stypes, session.url,
                                             session.assoc_handle, return_to, realm=realm,
                                             claimed=claimed, op_local=op_local)
         raise Redirect, url
@@ -127,7 +127,7 @@
         session = self.get_session(provider_id)
         realm = self.base+"?@action=openid_return"
         return_to = realm + "&__came_from=%s" % urllib.quote(self.client.path)
-        url = openid.request_authentication(session.stypes, session.url,
+        url = openid2rp.request_authentication(session.stypes, session.url,
                                             session.assoc_handle, return_to, realm=realm)
         raise Redirect, url
 
@@ -151,7 +151,7 @@
             return payload
         if 'openid.response_nonce' in query:
             nonce = query['openid.response_nonce'][0]
-            stamp = openid.parse_nonce(nonce)
+            stamp = openid2rp.parse_nonce(nonce)
             utc = calendar.timegm(stamp.utctimetuple())
             if utc < time.time()-3600:
                 # Old nonce
@@ -222,7 +222,7 @@
             if key.startswith('openid'):
                 openid_fields.append((key, self.form.getfirst(key)))
         pt = self.client.instance.templates.get('user', 'openid')
-        username = openid.get_username(query)
+        username = openid2rp.get_username(query)
         realname = None
         if username:
             if isinstance(username, tuple):
@@ -232,7 +232,7 @@
         result = pt.render(self.client, None, None,
                            realname=realname,
                            username=username,
-                           email=openid.get_email(query),
+                           email=openid2rp.get_email(query),
                            claimed=claimed,
                            openid_fields=openid_fields)
         self.client.additional_headers['Content-Type'] = pt.content_type

Deleted: tracker/instances/python-dev/lib/openid.py
==============================================================================
--- tracker/instances/python-dev/lib/openid.py	Tue Jul 20 17:07:28 2010
+++ (empty file)
@@ -1,701 +0,0 @@
-# -*- coding: utf-8 -*-
-# OpenID relying party library
-# Copyright Martin v. L?wis, 2009
-# Licensed under the Academic Free License, version 3
-
-# This library implements OpenID Authentication 2.0,
-# in the role of a relying party
-# It has the following assumptions and limitations:
-# - service discovery requires YADIS (HTML discovery not implemented)
-# - only provider-directed mode (identifier_select) is supported
-# - direct requests require https
-# - as a signature algorithm, HMAC-SHA1 is requested
-
-import urlparse, urllib, httplib, BeautifulSoup, time
-import cStringIO, base64, hmac, sha, datetime, re, binascii, struct
-import itertools
-
-try:
-    from xml.etree import ElementTree
-except ImportError:
-    from elementtree import ElementTree
-
-# Importing M2Crypto patches urllib; don't let them do that
-orig = urllib.URLopener.open_https.im_func
-from M2Crypto import DH
-urllib.URLopener.open_https = orig
-
-# Don't use urllib2, since it breaks in 2.5
-# for https://login.launchpad.net//+xrds
-
-# Don't use urllib, since it sometimes selects HTTP/1.1 (e.g. in PyPI)
-# and then fails to parse chunked responses.
-
-def normalize_uri(uri):
-    """Normalize an uri according to OpenID section 7.2. Return a pair
-    type,value, where type can be either 'xri' or 'uri'."""
-    
-    # 7.2 Normalization
-    if uri.startswith('xri://'):
-        uri = uri[6:]
-    if uri[0] in ("=", "@", "+", "$", "!", ")"):
-        return 'xri', uri
-    if not uri.startswith('http'):
-        uri = 'http://' + uri
-    # RFC 3986, section 6
-
-    # 6.2.2.1 case normalization
-    parts = urlparse.urlparse(uri) # already lower-cases scheme
-    if '@' in parts[1]: #netloc
-        userinfo,hostname = parts[1].rsplit('@', 1)
-    else:
-        userinfo,hostname = None,parts[1]
-    if ':' in hostname:
-        host,port = hostname.rsplit(':', 1)
-        if ']' in port:
-            # IPv6
-            host,port = hostname,None
-    else:
-        host,port = hostname,None
-    netloc = hostname = host.lower()
-    if port:
-        netloc = hostname = host+':'+port
-    if userinfo:
-        netloc = userinfo + '@' + hostname
-    parts = list(parts)
-    parts[1] = netloc
-    uri = urlparse.urlunparse(parts)
-
-    # 6.2.2.2. normalize case in % escapes
-    # XXX should restrict search to parts that can be pct-encoded
-    for match in re.findall('%[0-9a-fA-F][0-9a-fA-F]', uri):
-        m2 = match.upper()
-        if m2 != match:
-            uri = uri.replace(match, m2)
-
-    # 6.2.2.3 remove dot segments
-    parts = urlparse.urlparse(uri)
-    path = parts[2] #path
-    newpath = ''
-    while path:
-        if path.startswith('../'):
-            path = path[3:]
-        elif path.startswith('./'):
-            path = path[2:]
-        elif path.startswith('/./'):
-            newpath += '/'; path = path[3:]
-        elif path == '/.':
-            newpath += '/'; path = ''
-        elif path.startswith('/../'):
-            newpath = newpath.rsplit('/', 1)[0]
-            path = path[3:] # leave /
-        elif path == '/..':
-            newpath = newpath.rsplit('/', 1)[0]
-            path = '/'
-        elif path == '.' or path=='..':
-            path = ''
-        else:
-            pos = path.find('/', 1)
-            if pos == -1:
-                pos = len(path)
-            newpath += path[:pos]
-            path = path[pos:]
-    parts = list(parts)
-    parts[2] = newpath
-    uri = urlparse.urlunparse(parts)
-
-    # 6.2.3 scheme based normalization
-
-    parts = urlparse.urlparse(uri)    
-    netloc = parts[1]
-    if netloc.endswith(':'):
-        netloc = netloc[:-1]
-    elif parts[0] == 'http' and netloc.endswith(':80'):
-        netloc = netloc[:-3]
-    elif parts[0] == 'https' and netloc.endswith(':443'):
-        netloc = netloc[:-4]
-    # other default ports not considered here
-
-    path = parts[2]
-    if parts[0] in ('http', 'https') and parts[2]=='':
-        path = '/'
-
-    # 6.2.5 protocol-based normalization not done, as it
-    # is not appropriate to resolve the URL just for normalization
-    # it seems like a bug in the OpenID spec that it doesn't specify
-    # which normalizations exactly should be performed
-
-    parts = list(parts)
-    parts[1] = netloc
-    parts[2] = path
-    return 'uri', urlparse.urlunparse(parts)
-
-
-def parse_response(s):
-    '''Parse a key-value form (OpenID section 4.1.1) into a dictionary'''
-    res = {}
-    for line in s.splitlines():
-        k,v = line.split(':', 1)
-        res[k] = v
-    return res
-
-def discover(url):
-    '''Perform service discovery on the OP URL.
-    Return list of service types, and the auth/2.0 URL,
-    or None if discovery fails.'''
-    scheme, netloc, path, query, fragment = urlparse.urlsplit(url)
-    assert not fragment
-    if scheme == 'https':
-        conn = httplib.HTTPSConnection(netloc)
-    elif scheme == 'http':
-        conn = httplib.HTTPConnection(netloc)
-    else:
-        raise ValueError, "Unsupported scheme "+scheme
-    # conn.set_debuglevel(1)
-    if query:
-        path += '?'+query
-    try:
-        conn.connect()
-    except:
-        # DNS or TCP error
-        return None
-    # httplib in 2.5 incorrectly sends https port in Host
-    # header even if it is 443
-    conn.putrequest("GET", path, skip_host=1)
-    conn.putheader('Host', netloc)
-    conn.putheader('Accept', "text/html; q=0.3, "+
-                   "application/xhtml+xml; q=0.5, "+
-                   "application/xrds+xml")
-    conn.endheaders()
-
-    res = conn.getresponse()
-    data = res.read()
-    conn.close()
-
-    if res.status in (301, 302, 303, 307):
-        return discover(res.msg.get('location'))
-
-    content_type = res.msg.gettype()
-
-    # Yadis 6.2.5 option 2 and 3: header includes x-xrds-location
-    xrds_loc = res.msg.get('x-xrds-location')
-    if xrds_loc and content_type != 'application/xrds+xml':
-        return discover(xrds_loc)
-
-    if content_type in ('text/html', 'application/xhtml+xml'):
-        soup = BeautifulSoup.BeautifulSoup(data)
-        # Yadis 6.2.5 option 1: meta tag
-        meta = soup.find('meta', {'http-equiv':lambda v:v and v.lower()=='x-xrds-location'})
-        if meta:
-            xrds_loc = meta['content']
-            return discover(xrds_loc)
-        # OpenID 7.3.3: attempt html based discovery
-        op_endpoint = soup.find('link', {'rel':lambda v:v and 'openid2.provider' in v.lower()})
-        if op_endpoint:
-            op_endpoint = op_endpoint['href']
-            op_local = soup.find('link', {'rel':lambda v:v and 'openid2.local_id' in v.lower()})
-            if op_local:
-                op_local = op_local['href']
-            else:
-                op_local = None
-            return ['http://specs.openid.net/auth/2.0/signon'], op_endpoint, op_local
-        # 14.2.1: 1.1 compatibility
-        op_endpoint = soup.find('link', {'rel':lambda v:v and 'openid.server' in v.lower()})
-        if op_endpoint:
-            op_endpoint = op_endpoint['href']
-            op_local = soup.find('link', {'rel':lambda v:v and 'openid.delegate' in v.lower()})
-            if op_local:
-                op_local = op_local['href']
-            else:
-                op_local = None
-            return ['http://openid.net/signon/1.1'], op_endpoint, op_local
-        # Discovery failed
-        return None
-
-    elif content_type == 'application/xrds+xml':
-        # Yadis 6.2.5 option 4
-        doc = ElementTree.fromstring(data)
-        for svc in doc.findall(".//{xri://$xrd*($v*2.0)}Service"):
-            services = [x.text for x in svc.findall("{xri://$xrd*($v*2.0)}Type")]
-            if 'http://specs.openid.net/auth/2.0/server' in services:
-                # 7.3.2.1.1 OP Identifier Element
-                uri = svc.find("{xri://$xrd*($v*2.0)}URI")
-                if uri is not None:
-                    op_local = None
-                    op_endpoint = uri.text
-                    break
-            elif 'http://specs.openid.net/auth/2.0/signon' in services:
-                # 7.3.2.1.2.  Claimed Identifier Element
-                op_local = svc.find("{xri://$xrd*($v*2.0)}LocalID")
-                if op_local is not None:
-                    op_local = op_local.text
-                uri = svc.find("{xri://$xrd*($v*2.0)}URI")
-                if uri is not None:
-                    op_endpoint = uri.text
-                    break
-            elif 'http://openid.net/server/1.0' in services or \
-                 'http://openid.net/server/1.1' in services or \
-                 'http://openid.net/signon/1.0' in services or \
-                 'http://openid.net/signon/1.1' in services:
-                # 14.2.1 says we also need to check for the 1.x types;
-                # XXX should check 1.x only if no 2.0 service is found
-                op_local = svc.find("{http://openid.net/xmlns/1.0}Delegate")
-                if op_local is not None:
-                    op_local = op_local.text
-                uri = svc.find("{xri://$xrd*($v*2.0)}URI")
-                if uri is not None:
-                    op_endpoint = uri.text
-                    break
-        else:
-            return None # No OpenID 2.0 service found
-    else:
-        # unknown content type
-        return None
-    return services, op_endpoint, op_local
-
-def is_compat_1x(services):
-    for uri in ('http://specs.openid.net/auth/2.0/signon',
-                'http://specs.openid.net/auth/2.0/server'):
-        if uri in services:
-            return False
-    for uri in ('http://openid.net/signon/1.0',
-                'http://openid.net/signon/1.1',
-                'http://openid.net/server/1.0',
-                'http://openid.net/server/1.1'):
-        if uri in services:
-            return True
-    raise ValueError, "Neither 1.x nor 2.0 service found"
-
-def is_op_endpoint(services):
-    for uri in ('http://specs.openid.net/auth/2.0/server',
-                'http://openid.net/server/1.0',
-                'http://openid.net/server/1.1'):
-        if uri in services:
-            return True
-    return False
-
-# OpenSSL MPI integer representation
-def bin2mpi(bin):
-    if ord(bin[0]) >= 128:
-        # avoid interpretation as a negative number
-        bin = "\x00" + bin
-    return struct.pack(">i", len(bin))+bin
-def mpi2bin(mpi):
-    assert len(mpi)-4 == struct.unpack(">i", mpi[:4])[0]
-    return mpi[4:]
-
-# Appendix B; DH default prime
-dh_prime = """
-DCF93A0B883972EC0E19989AC5A2CE310E1D37717E8D9571BB7623731866E61E
-F75A2E27898B057F9891C2E27A639C3F29B60814581CD3B2CA3986D268370557
-7D45C2E7E52DC81C7A171876E5CEA74B1448BFDFAF18828EFD2519F14E45E382
-6634AF1949E5B535CC829A483B8A76223E5D490A257F05BDFF16F2FB22C583AB
-"""
-dh_prime = binascii.unhexlify("".join(dh_prime.split()))
-# OpenSSL MPI representation: dh_prime, 2
-dh = DH.set_params(bin2mpi(dh_prime), '\x00\x00\x00\x01\x02')
-dh.gen_key()
-dh_public_base64 = base64.b64encode(mpi2bin(dh.pub))
-
-def string_xor(s1, s2):
-    res = []
-    for c1, c2 in itertools.izip(s1, s2):
-        res.append(chr(ord(c1) ^ ord(c2)))
-    return ''.join(res)
-
-def associate(services, url):
-    '''Create an association (OpenID section 8) between RP and OP.
-    Return response as a dictionary.'''
-    data = {
-        'openid.ns':"http://specs.openid.net/auth/2.0",
-        'openid.mode':"associate",
-        'openid.assoc_type':"HMAC-SHA1",
-        'openid.session_type':"no-encryption",
-        }
-    if url.startswith('http:'):
-        # Use DH exchange
-        data['openid.session_type'] = "DH-SHA1"
-        # No need to send key and generator
-        data['openid.dh_consumer_public'] = dh_public_base64
-    if is_compat_1x(services):
-        # 14.2.1: clear session_type in 1.1 compatibility mode
-        if data['openid.session_type'] == "no-encryption":
-            data['openid.session_type'] = ''
-        del data['openid.ns']
-    res = urllib.urlopen(url, urllib.urlencode(data))
-    data = parse_response(res.read())
-    if 'error' in data:
-        raise ValueError, "associate failed: "+data['error']
-    if url.startswith('http:'):
-        enc_mac_key = base64.b64decode(data['enc_mac_key'])
-        dh_server_public = base64.b64decode(data['dh_server_public'])
-        # compute_key does not return an MPI
-        shared_secret = dh.compute_key(bin2mpi(dh_server_public))
-        if ord(shared_secret[0]) >= 128:
-            # btwoc: add leading zero if number would otherwise be negative
-            shared_secret = '\x00' + shared_secret
-        shared_secret = sha.new(shared_secret).digest()
-        if len(shared_secret) != len(enc_mac_key):
-            raise ValueError, "incorrect DH key size"
-        # Fake mac_key result
-        data['mac_key'] = base64.b64encode(string_xor(enc_mac_key, shared_secret))
-    return data
-
-def request_authentication(services, url, assoc_handle, return_to,
-                           claimed=None, op_local=None, realm=None):
-    '''Request authentication (OpenID section 9).
-    services is the list of discovered service types,
-    url the OP service URL, assoc_handle the established session
-    dictionary, and return_to the return URL.
-
-    The return_to URL will also be passed as realm, and the
-    OP may perform RP discovery on it; always request these
-    data through SREG 1.0 as well.
-
-    If AX or SREG 1.1 are supported, request email address,
-    first/last name, or nickname.
-
-    Return the URL that the browser should be redirected to.'''
-
-    if is_op_endpoint(services):
-        # claimed is an OP identifier
-        claimed = op_local = None
-
-    if claimed is None:
-        claimed = "http://specs.openid.net/auth/2.0/identifier_select"
-    if op_local is None:
-        op_local = "http://specs.openid.net/auth/2.0/identifier_select"
-    if realm is None:
-        realm = return_to
-    data = {
-        'openid.ns':"http://specs.openid.net/auth/2.0",
-        'openid.mode':"checkid_setup",
-        'openid.assoc_handle':assoc_handle,
-        'openid.return_to':return_to,
-        'openid.claimed_id':claimed,
-        'openid.identity':op_local,
-        'openid.realm':realm,
-        'openid.ns.sreg':"http://openid.net/sreg/1.0",
-        'openid.sreg.required':'nickname,email',
-        }
-    if is_compat_1x(services):
-        del data['openid.ns']
-        del data['openid.claimed_id']
-        del data['openid.realm']
-        data['openid.trust_root'] = return_to
-    if "http://openid.net/srv/ax/1.0" in services:
-        data.update({
-            'openid.ns.ax':"http://openid.net/srv/ax/1.0",
-            'openid.ax.mode':'fetch_request',
-            'openid.ax.required':'email,first,last',
-            'openid.ax.type.email':'http://axschema.org/contact/email',
-            'openid.ax.type.first':"http://axschema.org/namePerson/first",
-            'openid.ax.type.last':"http://axschema.org/namePerson/last",
-            })
-    if "http://openid.net/extensions/sreg/1.1" in services:
-        data.update({
-            'openid.ns.sreg11':"http://openid.net/extensions/sreg/1.1",
-            'openid.sreg11.required':'nickname,email'
-            })
-    if '?' in url:
-        return url+'&'+urllib.urlencode(data)
-    else:
-        return url+"?"+urllib.urlencode(data)
-
-class NotAuthenticated(Exception):
-    pass
-
-def authenticate(session, response):
-    '''Process an authentication response.
-    session must be the established session (minimally including
-    assoc_handle and mac_key), response is the query string as parsed
-    by cgi.parse_qs.
-    If authentication succeeds, return the list of signed fields.
-    If the user was not authenticated, NotAuthenticated is raised.
-    If the HTTP request is invalid (missing parameters, failure to
-    validate signature), different exceptions will be raised, typically
-    ValueError.
-
-    Callers must check openid.response_nonce for replay attacks.
-    '''
-
-    # 1.1 compat: openid.ns may not be sent
-    # if response['openid.ns'][0] != 'http://specs.openid.net/auth/2.0':
-    #    raise ValueError('missing openid.ns')
-    if session['assoc_handle'] != response['openid.assoc_handle'][0]:
-        raise ValueError('incorrect session')
-    if response['openid.mode'][0] == 'cancel':
-        raise NotAuthenticated('provider did not authenticate user (cancelled)')
-    if response['openid.mode'][0] != 'id_res':
-        raise ValueError('invalid openid.mode')
-    if  'openid.identity' not in response:
-        raise ValueError('missing openid.identity')
-
-    # Won't check nonce value - caller must verify this is not a replay
-
-    signed = response['openid.signed'][0].split(',')
-    query = []
-    for name in signed:
-        if isinstance(name, unicode):
-            name = name.encode('ascii')
-        value = response['openid.'+name][0]
-        if isinstance(value, unicode):
-            value = value.encode('utf-8')
-        query.append('%s:%s\n' % (name, value))
-    query = ''.join(query)
-
-    mac_key = base64.decodestring(session['mac_key'])
-    transmitted_sig = base64.decodestring(response['openid.sig'][0])
-    computed_sig = hmac.new(mac_key, query, sha).digest()
-
-    if transmitted_sig != computed_sig:
-        raise ValueError('Invalid signature')
-
-    # Check that all critical fields are signed. OpenID 2.0 says
-    # that in a positive assertion, op_endpoint, return_to,
-    # response_nonce and assoc_handle must be signed, and claimed_id
-    # and identity if present in the response. 1.1 compatibility
-    # says that response_nonce and op_endpoint may be missing.
-    # In addition, OpenID 1.1 providers apparently fail to sign
-    # assoc_handle often.
-    if response['openid.mode'][0] == 'id_res':
-        if 'return_to' not in signed or \
-           ('openid.identity' in response and 'identity' not in signed) or \
-           ('openid.claimed_id' in response and 'claimed_id' not in signed):
-            raise ValueError, "Critical field missing in signature"
-
-    return signed
-
-def parse_nonce(nonce):
-    '''Split a nonce into a (timestamp, ID) pair'''
-    stamp = nonce.split('Z', 1)[0]
-    stamp = time.strptime(stamp, "%Y-%m-%dT%H:%M:%S")[:6]
-    stamp = datetime.datetime(*stamp)
-    return stamp
-
-def get_namespaces(resp):
-    res = {}
-    for k, v in resp.items():
-        if k.startswith('openid.ns.'):
-            k = k.rsplit('.', 1)[1]
-            res[v[0]] = k
-    return res
-
-def get_ax(resp, ns, validated):
-    if "http://openid.net/srv/ax/1.0" not in ns:
-        return {}
-    ax = ns["http://openid.net/srv/ax/1.0"]+"."
-    oax = "openid."+ax
-    res = {}
-    for k, v in resp.items():
-        if k.startswith(oax+"type."):
-            k = k.rsplit('.',1)[1]
-            value_name = oax+"value."+k
-            if ax+"value."+k not in validated:
-                continue
-            res[v[0]] = resp[value_name][0]
-    return res
-    
-
-def get_email(resp):
-    "Return the email address embedded response, or None."
-
-    validated = resp['openid.signed'][0]
-
-    # SREG 1.0; doesn't require namespace, as the protocol doesn't
-    # specify one
-    if 'openid.sreg.email' in resp and \
-       'sreg.email' in validated:
-        return resp['openid.sreg.email'][0]
-
-    ns = get_namespaces(resp)
-
-    ax = get_ax(resp, ns, validated)
-    if "http://axschema.org/contact/email" in ax:
-        return ax["http://axschema.org/contact/email"]
-
-    # TODO: SREG 1.1
-    return None
-
-def get_username(resp):
-    "Return either nickname or (first, last) or None."
-
-    validated = resp['openid.signed'][0]
-    if 'openid.sreg.nickname' in resp and \
-       'sreg.nickname' in validated:
-        return resp['openid.sreg.nickname'][0]
-
-    ns = get_namespaces(resp)
-
-    ax = get_ax(resp, ns, validated)
-    if "http://axschema.org/namePerson/first" in ax and \
-       "http://axschema.org/namePerson/last" in ax:
-        return (ax["http://axschema.org/namePerson/first"],
-                ax["http://axschema.org/namePerson/last"])
-
-    # TODO: SREG 1.1
-    return
-
-
-################ Test Server #################################
-
-import BaseHTTPServer, cgi
-
-# supported providers
-providers = (
-    ('Google', 'http://www.google.com/favicon.ico', 'https://www.google.com/accounts/o8/id'),
-    ('Yahoo', 'http://www.yahoo.com/favicon.ico', 'http://yahoo.com/'),
-    # Verisigns service URL is not https
-    #('Verisign', 'https://pip.verisignlabs.com/favicon.ico', 'https://pip.verisignlabs.com')
-    ('myOpenID', 'https://www.myopenid.com/favicon.ico', 'https://www.myopenid.com/'),
-    ('Launchpad', 'https://login.launchpad.net/favicon.ico', 'https://login.launchpad.net/')
-    )
-             
-sessions = []
-class Handler(BaseHTTPServer.BaseHTTPRequestHandler):
-
-    def write(self, payload, type):
-        self.send_response(200)
-        self.send_header("Content-type", type)
-        self.send_header("Content-length", str(len(payload)))
-        self.end_headers()
-        self.wfile.write(payload)
-
-    def do_GET(self):
-        if self.path == '/':
-            return self.root()
-        path = self.path
-        i = path.rfind('?')
-        if i >= 0:
-            query = cgi.parse_qs(path[i+1:])
-            path = path[:i]
-        else:
-            query = {}
-        if path == '/':
-            if 'provider' in query:
-                prov = [p for p in providers if p[0]  == query['provider'][0]]
-                if len(prov) != 1:
-                    return self.not_found()
-                prov = prov[0]
-                services, url, op_local = discover(prov[2])
-                session = associate(services, url)
-                sessions.append(session)
-                self.send_response(307) # temporary redirect - do not cache
-                self.send_header("Location", request_authentication
-                                 (services, url, session['assoc_handle'],
-                                  self.base_url+"?returned=1"))
-                self.end_headers()
-                return
-            if 'claimed' in query:
-                kind, claimed = normalize_uri(query['claimed'][0])
-                if kind == 'xri':
-                    return self.error('XRI resolution not supported')
-                res = discover(claimed)
-                if res is None:
-                    return self.error('Discovery failed')
-                services, url, op_local = res
-                session = associate(services, url)
-                sessions.append(session)
-                self.send_response(307)
-                self.send_header("Location", request_authentication
-                                 (services, url, session['assoc_handle'],
-                                  self.base_url+"?returned=1",
-                                  claimed, op_local))
-                self.end_headers()
-                return                
-            if 'returned' in query:
-                if 'openid.identity' not in query:
-                    return self.rp_discovery()
-                handle = query['openid.assoc_handle'][0]
-                for session in sessions:
-                    if session['assoc_handle'] == handle:
-                        break
-                else:
-                    session = None
-                if not session:
-                    return self.error('Not authenticated (no session)')
-                try:
-                    signed = authenticate(session, query)
-                except Exception, e:
-                    self.error("Authentication failed: "+repr(e))
-                    return
-                if 'openid.claimed_id' in query:
-                    if 'claimed_id' not in signed:
-                        return self.error('Incomplete signature')
-                    claimed = query['openid.claimed_id'][0]
-                else:
-                    # OpenID 1, claimed ID not reported - should set cookie
-                    if 'identity' not in signed:
-                        return self.error('Incomplete signature')
-                    claimed = query['openid.identity'][0]
-                payload = "Hello "+claimed+"\n"
-                email = get_email(query)
-                if email:
-                    payload += 'Your email is '+email+"\n"
-                else:
-                    payload += 'No email address is known\n'
-                username = get_username(query)
-                if isinstance(username, tuple):
-                    username = " ".join(username)
-                if username:
-                    payload += 'Your nickname is '+username+'\n'
-                else:
-                    payload += 'No nickname is known\n'
-                return self.write(payload, "text/plain")
-                
-        return self.not_found()
-
-    
-
-    def debug(self, value):
-        payload = repr(value)
-        self.write(payload, "text/plain")
-
-    def error(self, text):
-        self.write(text, "text/plain")
-
-    def root(self):
-        payload = "<html><head><title>OpenID login</title></head><body>\n"
-        
-        for name, icon, provider in providers:
-            payload += "<p><a href='%s?provider=%s'><img src='%s' alt='%s'></a></p>\n" % (
-                self.base_url, name, icon, name)
-        payload += "<form>Type your OpenID:<input name='claimed'/><input type='submit'/></form>\n"
-        payload += "</body></html>"
-        self.write(payload, "text/html")
-
-    def rp_discovery(self):
-        payload = '''<xrds:XRDS  
-                xmlns:xrds="xri://$xrds"  
-                xmlns="xri://$xrd*($v*2.0)">  
-                <XRD>  
-                     <Service priority="1">  
-                              <Type>http://specs.openid.net/auth/2.0/return_to</Type>  
-                              <URI>%s</URI>  
-                     </Service>  
-                </XRD>  
-                </xrds:XRDS>
-        ''' % (self.base_url+"/?returned=1")
-        self.write(payload, 'application/xrds+xml')
-
-    def not_found(self):
-        self.send_response(404)
-        self.end_headers()
-        
-# OpenID providers often attempt relying-party discovery
-# This requires the test server to use a globally valid URL
-# If Python cannot correctly determine the base URL, you
-# can pass it as command line argument
-def test_server():
-    import socket, sys
-    if len(sys.argv) > 1:
-        base_url = sys.argv[1]
-    else:
-        base_url = "http://" + socket.getfqdn() + ":8000/"
-    Handler.base_url = base_url
-    BaseHTTPServer.HTTPServer.address_family = socket.AF_INET6
-    httpd = BaseHTTPServer.HTTPServer(('', 8000), Handler)
-    httpd.serve_forever()
-
-if __name__ == '__main__':
-    test_server()

From python-checkins at python.org  Tue Jul 20 17:09:01 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Tue, 20 Jul 2010 17:09:01 +0200 (CEST)
Subject: [Python-checkins] r82994 -
	tracker/instances/python-dev/lib/openid2rp.py
Message-ID: <20100720150901.3916FF322@mail.python.org>

Author: martin.v.loewis
Date: Tue Jul 20 17:09:01 2010
New Revision: 82994

Log:
Import 6d3ad02c7b41 from the openid2rp package.


Modified:
   tracker/instances/python-dev/lib/openid2rp.py

Modified: tracker/instances/python-dev/lib/openid2rp.py
==============================================================================
--- tracker/instances/python-dev/lib/openid2rp.py	(original)
+++ tracker/instances/python-dev/lib/openid2rp.py	Tue Jul 20 17:09:01 2010
@@ -5,32 +5,31 @@
 
 # This library implements OpenID Authentication 2.0,
 # in the role of a relying party
-# It has the following assumptions and limitations:
-# - service discovery requires YADIS (HTML discovery not implemented)
-# - only provider-directed mode (identifier_select) is supported
-# - direct requests require https
-# - as a signature algorithm, HMAC-SHA1 is requested
-
-import urlparse, urllib, httplib, BeautifulSoup, time
-import cStringIO, base64, hmac, sha, datetime, re, binascii, struct
-import itertools
+
+import urlparse, urllib, httplib, time, cgi, htmllib, formatter
+import cStringIO, base64, hmac, hashlib, datetime, re, random
+import itertools, cPickle, sys
 
 try:
     from xml.etree import ElementTree
 except ImportError:
     from elementtree import ElementTree
 
-# Importing M2Crypto patches urllib; don't let them do that
-orig = urllib.URLopener.open_https.im_func
-from M2Crypto import DH
-urllib.URLopener.open_https = orig
-
 # Don't use urllib2, since it breaks in 2.5
 # for https://login.launchpad.net//+xrds
 
 # Don't use urllib, since it sometimes selects HTTP/1.1 (e.g. in PyPI)
 # and then fails to parse chunked responses.
 
+# 3.x portability
+
+if sys.version_info < (3,):
+    def b(s):
+        return s
+else:
+    def b(s):
+        return s.encode('ascii')
+
 def normalize_uri(uri):
     """Normalize an uri according to OpenID section 7.2. Return a pair
     type,value, where type can be either 'xri' or 'uri'."""
@@ -134,11 +133,27 @@
 def parse_response(s):
     '''Parse a key-value form (OpenID section 4.1.1) into a dictionary'''
     res = {}
-    for line in s.splitlines():
+    for line in s.decode('utf-8').splitlines():
         k,v = line.split(':', 1)
         res[k] = v
     return res
 
+class OpenIDParser(htmllib.HTMLParser):
+    def __init__(self):
+        htmllib.HTMLParser.__init__(self, formatter.NullFormatter())
+        self.links = {}
+        self.xrds_location=None
+
+    def do_link(self, attrs):
+        attrs = dict(attrs)
+        self.links[attrs['rel']] = attrs['href']
+
+    def do_meta(self, attrs):
+        attrs = dict(attrs)
+        # Yadis 6.2.5 option 1: meta tag
+        if attrs['http-equiv'].lower() == 'x-xrds-location':
+            self.xrds_location = attrs['content']
+
 def discover(url):
     '''Perform service discovery on the OP URL.
     Return list of service types, and the auth/2.0 URL,
@@ -175,7 +190,10 @@
     if res.status in (301, 302, 303, 307):
         return discover(res.msg.get('location'))
 
-    content_type = res.msg.gettype()
+    if sys.version_info < (3,0):
+        content_type = res.msg.gettype()
+    else:
+        content_type = res.msg.get_content_type()
 
     # Yadis 6.2.5 option 2 and 3: header includes x-xrds-location
     xrds_loc = res.msg.get('x-xrds-location')
@@ -183,31 +201,21 @@
         return discover(xrds_loc)
 
     if content_type in ('text/html', 'application/xhtml+xml'):
-        soup = BeautifulSoup.BeautifulSoup(data)
+        parser = OpenIDParser()
+        parser.feed(data)
+        parser.close()
         # Yadis 6.2.5 option 1: meta tag
-        meta = soup.find('meta', {'http-equiv':lambda v:v and v.lower()=='x-xrds-location'})
-        if meta:
-            xrds_loc = meta['content']
-            return discover(xrds_loc)
+        if parser.xrds_location:
+            return discover(parser.xrds_location)
         # OpenID 7.3.3: attempt html based discovery
-        op_endpoint = soup.find('link', {'rel':lambda v:v and 'openid2.provider' in v.lower()})
+        op_endpoint = parser.links.get('openid2.provider')
         if op_endpoint:
-            op_endpoint = op_endpoint['href']
-            op_local = soup.find('link', {'rel':lambda v:v and 'openid2.local_id' in v.lower()})
-            if op_local:
-                op_local = op_local['href']
-            else:
-                op_local = None
+            op_local = parser.links.get('openid2.local_id')
             return ['http://specs.openid.net/auth/2.0/signon'], op_endpoint, op_local
         # 14.2.1: 1.1 compatibility
-        op_endpoint = soup.find('link', {'rel':lambda v:v and 'openid.server' in v.lower()})
+        op_endpoint = parser.links.get('openid.server')
         if op_endpoint:
-            op_endpoint = op_endpoint['href']
-            op_local = soup.find('link', {'rel':lambda v:v and 'openid.delegate' in v.lower()})
-            if op_local:
-                op_local = op_local['href']
-            else:
-                op_local = None
+            op_local = parser.links.get('openid.delegate')
             return ['http://openid.net/signon/1.1'], op_endpoint, op_local
         # Discovery failed
         return None
@@ -273,16 +281,20 @@
         if uri in services:
             return True
     return False
+is_op_identifier = is_op_endpoint
+
+# 4.1.3: Binary two's complement
+def btwoc(l):
+    res = cPickle.dumps(l, 2)
+    # Pickle result: proto 2, long1 (integer < 256 bytes)
+    # number of bytes, little-endian integer, stop
+    assert res[:3] == '\x80\x02\x8a' 
+    # btwoc ought to produce the shortest representation in two's
+    # complement. Fortunately, pickle already does that.
+    return res[3+ord(res[3]):3:-1]
 
-# OpenSSL MPI integer representation
-def bin2mpi(bin):
-    if ord(bin[0]) >= 128:
-        # avoid interpretation as a negative number
-        bin = "\x00" + bin
-    return struct.pack(">i", len(bin))+bin
-def mpi2bin(mpi):
-    assert len(mpi)-4 == struct.unpack(">i", mpi[:4])[0]
-    return mpi[4:]
+def unbtwoc(b):
+    return cPickle.loads('\x80\x02\x8a'+chr(len(b))+b[::-1]+'.')
 
 # Appendix B; DH default prime
 dh_prime = """
@@ -291,11 +303,7 @@
 7D45C2E7E52DC81C7A171876E5CEA74B1448BFDFAF18828EFD2519F14E45E382
 6634AF1949E5B535CC829A483B8A76223E5D490A257F05BDFF16F2FB22C583AB
 """
-dh_prime = binascii.unhexlify("".join(dh_prime.split()))
-# OpenSSL MPI representation: dh_prime, 2
-dh = DH.set_params(bin2mpi(dh_prime), '\x00\x00\x00\x01\x02')
-dh.gen_key()
-dh_public_base64 = base64.b64encode(mpi2bin(dh.pub))
+dh_prime = long("".join(dh_prime.split()), 16)
 
 def string_xor(s1, s2):
     res = []
@@ -315,6 +323,11 @@
     if url.startswith('http:'):
         # Use DH exchange
         data['openid.session_type'] = "DH-SHA1"
+        # Private key: random number between 1 and dh_prime-1
+        priv = random.randrange(1, dh_prime-1)
+        # Public key: 2^priv mod prime
+        pubkey = pow(2L, priv, dh_prime)
+        dh_public_base64 = base64.b64encode(btwoc(pubkey))
         # No need to send key and generator
         data['openid.dh_consumer_public'] = dh_public_base64
     if is_compat_1x(services):
@@ -328,21 +341,45 @@
         raise ValueError, "associate failed: "+data['error']
     if url.startswith('http:'):
         enc_mac_key = base64.b64decode(data['enc_mac_key'])
-        dh_server_public = base64.b64decode(data['dh_server_public'])
-        # compute_key does not return an MPI
-        shared_secret = dh.compute_key(bin2mpi(dh_server_public))
-        if ord(shared_secret[0]) >= 128:
-            # btwoc: add leading zero if number would otherwise be negative
-            shared_secret = '\x00' + shared_secret
-        shared_secret = sha.new(shared_secret).digest()
+        dh_server_public = unbtwoc(base64.b64decode(data['dh_server_public']))
+        # shared secret: sha1(2^(server_priv*priv) mod prime) xor enc_mac_key
+        shared_secret = btwoc(pow(dh_server_public, priv, dh_prime))
+        shared_secret = hashlib.sha1(shared_secret).digest()
         if len(shared_secret) != len(enc_mac_key):
             raise ValueError, "incorrect DH key size"
         # Fake mac_key result
         data['mac_key'] = base64.b64encode(string_xor(enc_mac_key, shared_secret))
     return data
 
+class _AX:
+    def __init__(self):
+        self.__dict__['_reverse'] = {}
+    def __setattr__(self, name, value):
+        self.__dict__[name] = value
+        self._reverse[value] = name
+    def lookup(self, value):
+        try:
+            return self._reverse[value]
+        except KeyError:
+            return 'a%d' % (hash(value) % 1000000000)
+
+AX = _AX()
+AX.nickname  = "http://axschema.org/namePerson/friendly"
+AX.email     = "http://axschema.org/contact/email"
+AX.fullname  = "http://axschema.org/namePerson"
+AX.dob       = "http://axschema.org/birthDate"
+AX.gender    = "http://axschema.org/person/gender"
+AX.postcode  = "http://axschema.org/contact/postalCode/home"
+AX.country   = "http://axschema.org/contact/country/home"
+AX.language  = "http://axschema.org/pref/language"
+AX.timezone  = "http://axschema.org/pref/timezone"
+AX.first     = "http://axschema.org/namePerson/first"
+AX.last      = "http://axschema.org/namePerson/last"
+
 def request_authentication(services, url, assoc_handle, return_to,
-                           claimed=None, op_local=None, realm=None):
+                           claimed=None, op_local=None, realm=None,
+                           sreg = (('nickname', 'email'), ()),
+                           ax = ((AX.email, AX.first, AX.last), ())):
     '''Request authentication (OpenID section 9).
     services is the list of discovered service types,
     url the OP service URL, assoc_handle the established session
@@ -375,28 +412,38 @@
         'openid.claimed_id':claimed,
         'openid.identity':op_local,
         'openid.realm':realm,
-        'openid.ns.sreg':"http://openid.net/sreg/1.0",
         'openid.sreg.required':'nickname,email',
         }
+    sreg_req, sreg_opt = sreg
+    sreg11 = {}
+    if sreg_req or sreg_opt:
+        data['openid.ns.sreg'] = "http://openid.net/sreg/1.0"
+        if sreg_req:
+            data['openid.sreg.required'] = sreg11['openid.sreg11.required'] = ','.join(sreg_req)
+        if sreg_opt:
+            data['openid.sreg.optional'] =  sreg11['openid.sreg11.optional'] =','.join(sreg_opt)
     if is_compat_1x(services):
         del data['openid.ns']
         del data['openid.claimed_id']
         del data['openid.realm']
         data['openid.trust_root'] = return_to
-    if "http://openid.net/srv/ax/1.0" in services:
+    ax_req, ax_opt = ax
+    if "http://openid.net/srv/ax/1.0" in services and (ax_req or ax_opt):
         data.update({
             'openid.ns.ax':"http://openid.net/srv/ax/1.0",
             'openid.ax.mode':'fetch_request',
-            'openid.ax.required':'email,first,last',
-            'openid.ax.type.email':'http://axschema.org/contact/email',
-            'openid.ax.type.first':"http://axschema.org/namePerson/first",
-            'openid.ax.type.last':"http://axschema.org/namePerson/last",
             })
-    if "http://openid.net/extensions/sreg/1.1" in services:
+        for uri in ax_req + ax_opt:
+            data['openid.ax.type.'+AX.lookup(uri)] = uri
+        if ax_req:
+            data['openid.ax.required'] = ','.join(AX.lookup(uri) for uri in ax_req)
+        if ax_opt:
+            data['openid.ax.if_available'] = ','.join(AX.lookup(uri) for uri in ax_req)
+    if "http://openid.net/extensions/sreg/1.1" in services and sreg11:
         data.update({
             'openid.ns.sreg11':"http://openid.net/extensions/sreg/1.1",
-            'openid.sreg11.required':'nickname,email'
             })
+        data.update(sreg11)
     if '?' in url:
         return url+'&'+urllib.urlencode(data)
     else:
@@ -405,20 +452,29 @@
 class NotAuthenticated(Exception):
     pass
 
+def _prepare_response(response):
+    if isinstance(response, str):
+        return cgi.parse_qs(response)
+    # backwards compatibility: allow caller to pass parse_qs result
+    # already
+    pass
+    return response
+
 def authenticate(session, response):
-    '''Process an authentication response.
-    session must be the established session (minimally including
-    assoc_handle and mac_key), response is the query string as parsed
-    by cgi.parse_qs.
-    If authentication succeeds, return the list of signed fields.
-    If the user was not authenticated, NotAuthenticated is raised.
-    If the HTTP request is invalid (missing parameters, failure to
-    validate signature), different exceptions will be raised, typically
-    ValueError.
+    '''Process an authentication response.  session must be the
+    established session (minimally including assoc_handle and
+    mac_key), response the query string as given in the original URL
+    (i.e. as the CGI variable QUERY_STRING).  If authentication
+    succeeds, return the list of signed fields.  If the user was not
+    authenticated, NotAuthenticated is raised.  If the HTTP request is
+    invalid (missing parameters, failure to validate signature),
+    different exceptions will be raised, typically ValueError.
 
     Callers must check openid.response_nonce for replay attacks.
     '''
 
+    response = _prepare_response(response)
+
     # 1.1 compat: openid.ns may not be sent
     # if response['openid.ns'][0] != 'http://specs.openid.net/auth/2.0':
     #    raise ValueError('missing openid.ns')
@@ -431,22 +487,21 @@
     if  'openid.identity' not in response:
         raise ValueError('missing openid.identity')
 
-    # Won't check nonce value - caller must verify this is not a replay
+    # Will not check nonce value - caller must verify this is not a replay
 
     signed = response['openid.signed'][0].split(',')
     query = []
     for name in signed:
-        if isinstance(name, unicode):
-            name = name.encode('ascii')
         value = response['openid.'+name][0]
-        if isinstance(value, unicode):
+        value = '%s:%s\n' % (name, value)
+        if sys.version_info >= (3,):
             value = value.encode('utf-8')
-        query.append('%s:%s\n' % (name, value))
-    query = ''.join(query)
+        query.append(value)
+    query = b('').join(query)
 
-    mac_key = base64.decodestring(session['mac_key'])
-    transmitted_sig = base64.decodestring(response['openid.sig'][0])
-    computed_sig = hmac.new(mac_key, query, sha).digest()
+    mac_key = base64.decodestring(b(session['mac_key']))
+    transmitted_sig = base64.decodestring(b(response['openid.sig'][0]))
+    computed_sig = hmac.new(mac_key, query, hashlib.sha1).digest()
 
     if transmitted_sig != computed_sig:
         raise ValueError('Invalid signature')
@@ -474,6 +529,7 @@
     return stamp
 
 def get_namespaces(resp):
+    resp = _prepare_response(resp)
     res = {}
     for k, v in resp.items():
         if k.startswith('openid.ns.'):
@@ -487,6 +543,7 @@
     ax = ns["http://openid.net/srv/ax/1.0"]+"."
     oax = "openid."+ax
     res = {}
+    resp = _prepare_response(resp)
     for k, v in resp.items():
         if k.startswith(oax+"type."):
             k = k.rsplit('.',1)[1]
@@ -495,11 +552,23 @@
                 continue
             res[v[0]] = resp[value_name][0]
     return res
-    
+
+def get_sreg(resp, validated):
+    """Return the dictionary of simple registration parameters in resp,
+    with the openid.sreg. prefix stripped."""
+    res = {}
+    resp = _prepare_response(resp)
+    for k, v in resp.items():
+        if k.startswith('openid.sreg.'):
+            k = k[len('openid.sreg.'):]
+            if 'sreg.'+k in validated:
+                res[k] = v[0]
+    return res
 
 def get_email(resp):
     "Return the email address embedded response, or None."
 
+    resp = _prepare_response(resp)
     validated = resp['openid.signed'][0]
 
     # SREG 1.0; doesn't require namespace, as the protocol doesn't
@@ -518,8 +587,11 @@
     return None
 
 def get_username(resp):
-    "Return either nickname or (first, last) or None."
+    """Return either nickname or (first, last) or None.
+    This function is deprecated; use get_ax and get_sreg instead.
+    """
 
+    resp = _prepare_response(resp)
     validated = resp['openid.signed'][0]
     if 'openid.sreg.nickname' in resp and \
        'sreg.nickname' in validated:
@@ -535,167 +607,3 @@
 
     # TODO: SREG 1.1
     return
-
-
-################ Test Server #################################
-
-import BaseHTTPServer, cgi
-
-# supported providers
-providers = (
-    ('Google', 'http://www.google.com/favicon.ico', 'https://www.google.com/accounts/o8/id'),
-    ('Yahoo', 'http://www.yahoo.com/favicon.ico', 'http://yahoo.com/'),
-    # Verisigns service URL is not https
-    #('Verisign', 'https://pip.verisignlabs.com/favicon.ico', 'https://pip.verisignlabs.com')
-    ('myOpenID', 'https://www.myopenid.com/favicon.ico', 'https://www.myopenid.com/'),
-    ('Launchpad', 'https://login.launchpad.net/favicon.ico', 'https://login.launchpad.net/')
-    )
-             
-sessions = []
-class Handler(BaseHTTPServer.BaseHTTPRequestHandler):
-
-    def write(self, payload, type):
-        self.send_response(200)
-        self.send_header("Content-type", type)
-        self.send_header("Content-length", str(len(payload)))
-        self.end_headers()
-        self.wfile.write(payload)
-
-    def do_GET(self):
-        if self.path == '/':
-            return self.root()
-        path = self.path
-        i = path.rfind('?')
-        if i >= 0:
-            query = cgi.parse_qs(path[i+1:])
-            path = path[:i]
-        else:
-            query = {}
-        if path == '/':
-            if 'provider' in query:
-                prov = [p for p in providers if p[0]  == query['provider'][0]]
-                if len(prov) != 1:
-                    return self.not_found()
-                prov = prov[0]
-                services, url, op_local = discover(prov[2])
-                session = associate(services, url)
-                sessions.append(session)
-                self.send_response(307) # temporary redirect - do not cache
-                self.send_header("Location", request_authentication
-                                 (services, url, session['assoc_handle'],
-                                  self.base_url+"?returned=1"))
-                self.end_headers()
-                return
-            if 'claimed' in query:
-                kind, claimed = normalize_uri(query['claimed'][0])
-                if kind == 'xri':
-                    return self.error('XRI resolution not supported')
-                res = discover(claimed)
-                if res is None:
-                    return self.error('Discovery failed')
-                services, url, op_local = res
-                session = associate(services, url)
-                sessions.append(session)
-                self.send_response(307)
-                self.send_header("Location", request_authentication
-                                 (services, url, session['assoc_handle'],
-                                  self.base_url+"?returned=1",
-                                  claimed, op_local))
-                self.end_headers()
-                return                
-            if 'returned' in query:
-                if 'openid.identity' not in query:
-                    return self.rp_discovery()
-                handle = query['openid.assoc_handle'][0]
-                for session in sessions:
-                    if session['assoc_handle'] == handle:
-                        break
-                else:
-                    session = None
-                if not session:
-                    return self.error('Not authenticated (no session)')
-                try:
-                    signed = authenticate(session, query)
-                except Exception, e:
-                    self.error("Authentication failed: "+repr(e))
-                    return
-                if 'openid.claimed_id' in query:
-                    if 'claimed_id' not in signed:
-                        return self.error('Incomplete signature')
-                    claimed = query['openid.claimed_id'][0]
-                else:
-                    # OpenID 1, claimed ID not reported - should set cookie
-                    if 'identity' not in signed:
-                        return self.error('Incomplete signature')
-                    claimed = query['openid.identity'][0]
-                payload = "Hello "+claimed+"\n"
-                email = get_email(query)
-                if email:
-                    payload += 'Your email is '+email+"\n"
-                else:
-                    payload += 'No email address is known\n'
-                username = get_username(query)
-                if isinstance(username, tuple):
-                    username = " ".join(username)
-                if username:
-                    payload += 'Your nickname is '+username+'\n'
-                else:
-                    payload += 'No nickname is known\n'
-                return self.write(payload, "text/plain")
-                
-        return self.not_found()
-
-    
-
-    def debug(self, value):
-        payload = repr(value)
-        self.write(payload, "text/plain")
-
-    def error(self, text):
-        self.write(text, "text/plain")
-
-    def root(self):
-        payload = "<html><head><title>OpenID login</title></head><body>\n"
-        
-        for name, icon, provider in providers:
-            payload += "<p><a href='%s?provider=%s'><img src='%s' alt='%s'></a></p>\n" % (
-                self.base_url, name, icon, name)
-        payload += "<form>Type your OpenID:<input name='claimed'/><input type='submit'/></form>\n"
-        payload += "</body></html>"
-        self.write(payload, "text/html")
-
-    def rp_discovery(self):
-        payload = '''<xrds:XRDS  
-                xmlns:xrds="xri://$xrds"  
-                xmlns="xri://$xrd*($v*2.0)">  
-                <XRD>  
-                     <Service priority="1">  
-                              <Type>http://specs.openid.net/auth/2.0/return_to</Type>  
-                              <URI>%s</URI>  
-                     </Service>  
-                </XRD>  
-                </xrds:XRDS>
-        ''' % (self.base_url+"/?returned=1")
-        self.write(payload, 'application/xrds+xml')
-
-    def not_found(self):
-        self.send_response(404)
-        self.end_headers()
-        
-# OpenID providers often attempt relying-party discovery
-# This requires the test server to use a globally valid URL
-# If Python cannot correctly determine the base URL, you
-# can pass it as command line argument
-def test_server():
-    import socket, sys
-    if len(sys.argv) > 1:
-        base_url = sys.argv[1]
-    else:
-        base_url = "http://" + socket.getfqdn() + ":8000/"
-    Handler.base_url = base_url
-    BaseHTTPServer.HTTPServer.address_family = socket.AF_INET6
-    httpd = BaseHTTPServer.HTTPServer(('', 8000), Handler)
-    httpd.serve_forever()
-
-if __name__ == '__main__':
-    test_server()

From python-checkins at python.org  Tue Jul 20 18:07:10 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Tue, 20 Jul 2010 18:07:10 +0200 (CEST)
Subject: [Python-checkins] r82995 - python/branches/py3k/Lib/sysconfig.py
Message-ID: <20100720160710.DE804EE99B@mail.python.org>

Author: ronald.oussoren
Date: Tue Jul 20 18:07:10 2010
New Revision: 82995

Log:
Without this patch the value of sysconfig.get_config_var('LDSHARED')
is wrong when PY_LDFLAGS is not empty.

The bug was caused by LDSHARED getting expanded *before* sysconfig
renamed PY_LDSHARED (and simular values) to names without a PY_
prefix.

The patch tries to maintain the intended behaviour of allowing users
to set LDFLAGS in the environment and have that affect the build.

Without this patch a universal build on OSX cannot build universal
(fat binary) extensions.



Modified:
   python/branches/py3k/Lib/sysconfig.py

Modified: python/branches/py3k/Lib/sysconfig.py
==============================================================================
--- python/branches/py3k/Lib/sysconfig.py	(original)
+++ python/branches/py3k/Lib/sysconfig.py	Tue Jul 20 18:07:10 2010
@@ -225,6 +225,12 @@
     # do variable interpolation here
     variables = list(notdone.keys())
 
+    # Variables with a 'PY_' prefix in the makefile. These need to
+    # be made available without that prefix through sysconfig.
+    # Special care is needed to ensure that variable expansion works, even
+    # if the expansion uses the name without a prefix.
+    renamed_variables = ('CFLAGS', 'LDFLAGS', 'CPPFLAGS')
+
     while len(variables) > 0:
         for name in tuple(variables):
             value = notdone[name]
@@ -240,8 +246,20 @@
                 elif n in os.environ:
                     # do it like make: fall back to environment
                     item = os.environ[n]
+
+                elif n in renamed_variables:
+                    if name.startswith('PY_') and name[3:] in renamed_variables:
+                        item = ""
+
+                    elif 'PY_' + n in notdone:
+                        found = False
+
+                    else:
+                        item = str(done['PY_' + n])
+
                 else:
                     done[n] = item = ""
+
                 if found:
                     after = value[m.end():]
                     value = value[:m.start()] + item + after
@@ -255,17 +273,19 @@
                         else:
                             done[name] = value
                         variables.remove(name)
+
+                        if name.startswith('PY_') \
+                                and name[3:] in renamed_variables:
+
+                            name = name[3:]
+                            if name not in done:
+                                done[name] = value
+
+
             else:
                 # bogus variable reference; just drop it since we can't deal
                 variables.remove(name)
 
-    # Add in CFLAGS, LDFLAGS, and CPPFLAGS, which are named with a
-    # prefix in the Makefile.
-    for var in ('CFLAGS', 'LDFLAGS', 'CPPFLAGS'):
-        makefile_value = done.get('PY_' + var)
-        if makefile_value is not None:
-            done[var] = makefile_value
-
     # save the results in the global dictionary
     vars.update(done)
     return vars

From python-checkins at python.org  Tue Jul 20 18:35:43 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Tue, 20 Jul 2010 18:35:43 +0200 (CEST)
Subject: [Python-checkins] r82996 -
	tracker/instances/python-dev/extensions/openid_login.py
Message-ID: <20100720163543.A7449EE997@mail.python.org>

Author: martin.v.loewis
Date: Tue Jul 20 18:35:43 2010
New Revision: 82996

Log:
Add missing list of providers.


Modified:
   tracker/instances/python-dev/extensions/openid_login.py

Modified: tracker/instances/python-dev/extensions/openid_login.py
==============================================================================
--- tracker/instances/python-dev/extensions/openid_login.py	(original)
+++ tracker/instances/python-dev/extensions/openid_login.py	Tue Jul 20 18:35:43 2010
@@ -3,10 +3,12 @@
 from roundup.cgi.exceptions import *
 from roundup import date, password
 
-good_providers = ['Google', 'myOpenID', 'Launchpad']
 providers = {}
-for p in openid2rp.providers:
-    if p[0] not in good_providers: continue
+for p in (
+    ('Google', 'http://www.google.com/favicon.ico', 'https://www.google.com/accounts/o8/id'),
+    ('myOpenID', 'https://www.myopenid.com/favicon.ico', 'https://www.myopenid.com/'),
+    ('Launchpad', 'https://login.launchpad.net/favicon.ico', 'https://login.launchpad.net/')
+    ):
     providers[p[0]] = p
 
 class Openid:

From stefan-usenet at bytereef.org  Tue Jul 20 21:37:22 2010
From: stefan-usenet at bytereef.org (Stefan Krah)
Date: Tue, 20 Jul 2010 21:37:22 +0200
Subject: [Python-checkins] r82995 -	python/branches/py3k/Lib/sysconfig.py
In-Reply-To: <20100720160710.DE804EE99B@mail.python.org>
References: <20100720160710.DE804EE99B@mail.python.org>
Message-ID: <20100720193722.GA23269@yoda.bytereef.org>

ronald.oussoren <python-checkins at python.org> wrote:
> Without this patch the value of sysconfig.get_config_var('LDSHARED')
> is wrong when PY_LDFLAGS is not empty.

It looks like the recently added test_ldshared_value is failing on
the Windows buildbots.


Stefan Krah



From python-checkins at python.org  Tue Jul 20 21:55:19 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Tue, 20 Jul 2010 21:55:19 +0200 (CEST)
Subject: [Python-checkins] r82997 - in python/branches/py3k: Lib/trace.py
	Misc/ACKS Misc/NEWS
Message-ID: <20100720195519.005C0EE9A3@mail.python.org>

Author: alexander.belopolsky
Date: Tue Jul 20 21:55:18 2010
New Revision: 82997

Log:
Issue #9282: Fixed --listfuncs option of trace.py.  Thanks Eli
Bendersky for the patch.


Modified:
   python/branches/py3k/Lib/trace.py
   python/branches/py3k/Misc/ACKS
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/trace.py
==============================================================================
--- python/branches/py3k/Lib/trace.py	(original)
+++ python/branches/py3k/Lib/trace.py	Tue Jul 20 21:55:18 2010
@@ -257,7 +257,8 @@
         if self.calledfuncs:
             print()
             print("functions called:")
-            for filename, modulename, funcname in sorted(calls.keys()):
+            calls = self.calledfuncs.keys()
+            for filename, modulename, funcname in sorted(calls):
                 print(("filename: %s, modulename: %s, funcname: %s"
                        % (filename, modulename, funcname)))
 

Modified: python/branches/py3k/Misc/ACKS
==============================================================================
--- python/branches/py3k/Misc/ACKS	(original)
+++ python/branches/py3k/Misc/ACKS	Tue Jul 20 21:55:18 2010
@@ -63,6 +63,7 @@
 Ben Bell
 Thomas Bellman
 Alexander Belopolsky
+Eli Bendersky
 Andrew Bennetts
 Andy Bensky
 Michel Van den Bergh

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Tue Jul 20 21:55:18 2010
@@ -473,6 +473,9 @@
 Library
 -------
 
+- Issue #9282: Fixed --listfuncs option of trace.py.  Thanks Eli
+  Bendersky for the patch.
+
 - Issue #3704: http.cookiejar was not properly handling URLs with a / in the
   parameters.
 

From python-checkins at python.org  Tue Jul 20 22:11:03 2010
From: python-checkins at python.org (stefan.krah)
Date: Tue, 20 Jul 2010 22:11:03 +0200 (CEST)
Subject: [Python-checkins] r82998 -
	python/branches/py3k-cdecimal/Lib/test/test_curses.py
Message-ID: <20100720201103.C0F0FEE995@mail.python.org>

Author: stefan.krah
Date: Tue Jul 20 22:11:03 2010
New Revision: 82998

Log:
Revert changes for testing issue #8433.

Modified:
   python/branches/py3k-cdecimal/Lib/test/test_curses.py

Modified: python/branches/py3k-cdecimal/Lib/test/test_curses.py
==============================================================================
--- python/branches/py3k-cdecimal/Lib/test/test_curses.py	(original)
+++ python/branches/py3k-cdecimal/Lib/test/test_curses.py	Tue Jul 20 22:11:03 2010
@@ -16,14 +16,13 @@
 # option.  If not available, nothing after this line will be executed.
 
 import unittest
-from test.support import requires, import_module, verbose
+from test.support import requires, import_module
 requires('curses')
 
 # If either of these don't exist, skip the tests.
 curses = import_module('curses')
 curses.panel = import_module('curses.panel')
 
-HAVE_ISSUE_8433 = False
 
 # XXX: if newterm was supported we could use it instead of initscr and not exit
 term = os.environ.get('TERM')
@@ -216,17 +215,13 @@
         curses.has_key(13)
 
     if hasattr(curses, 'getmouse'):
-        global HAVE_ISSUE_8433
         (availmask, oldmask) = curses.mousemask(curses.BUTTON1_PRESSED)
         # availmask indicates that mouse stuff not available.
         if availmask != 0:
             curses.mouseinterval(10)
             # just verify these don't cause errors
-            try:
-                m = curses.getmouse()
-                curses.ungetmouse(*m)
-            except curses.error:
-                HAVE_ISSUE_8433 = True
+            m = curses.getmouse()
+            curses.ungetmouse(*m)
 
     if hasattr(curses, 'is_term_resized'):
         curses.is_term_resized(*stdscr.getmaxyx())
@@ -292,8 +287,6 @@
     finally:
         curses.endwin()
     unit_tests()
-    if HAVE_ISSUE_8433:
-        raise unittest.SkipTest("test getmouse() skipped: see issue 8433")
 
 if __name__ == '__main__':
     curses.wrapper(main)

From python-checkins at python.org  Tue Jul 20 22:13:45 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Tue, 20 Jul 2010 22:13:45 +0200 (CEST)
Subject: [Python-checkins] r82999 - in python/branches/release31-maint:
	Lib/trace.py Misc/ACKS Misc/NEWS
Message-ID: <20100720201345.4BBDDEE9E9@mail.python.org>

Author: alexander.belopolsky
Date: Tue Jul 20 22:13:45 2010
New Revision: 82999

Log:
Merged revisions 82997 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82997 | alexander.belopolsky | 2010-07-20 15:55:18 -0400 (Tue, 20 Jul 2010) | 3 lines
  
  Issue #9282: Fixed --listfuncs option of trace.py.  Thanks Eli
  Bendersky for the patch.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/trace.py
   python/branches/release31-maint/Misc/ACKS
   python/branches/release31-maint/Misc/NEWS

Modified: python/branches/release31-maint/Lib/trace.py
==============================================================================
--- python/branches/release31-maint/Lib/trace.py	(original)
+++ python/branches/release31-maint/Lib/trace.py	Tue Jul 20 22:13:45 2010
@@ -257,7 +257,8 @@
         if self.calledfuncs:
             print()
             print("functions called:")
-            for filename, modulename, funcname in sorted(calls.keys()):
+            calls = self.calledfuncs.keys()
+            for filename, modulename, funcname in sorted(calls):
                 print(("filename: %s, modulename: %s, funcname: %s"
                        % (filename, modulename, funcname)))
 

Modified: python/branches/release31-maint/Misc/ACKS
==============================================================================
--- python/branches/release31-maint/Misc/ACKS	(original)
+++ python/branches/release31-maint/Misc/ACKS	Tue Jul 20 22:13:45 2010
@@ -60,6 +60,7 @@
 Ben Bell
 Thomas Bellman
 Alexander Belopolsky
+Eli Bendersky
 Andrew Bennetts
 Andy Bensky
 Michel Van den Bergh

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Tue Jul 20 22:13:45 2010
@@ -78,6 +78,9 @@
 Library
 -------
 
+- Issue #9282: Fixed --listfuncs option of trace.py.  Thanks Eli
+  Bendersky for the patch.
+
 - Issue #1555570: email no longer inserts extra blank lines when a \r\n
   combo crosses an 8192 byte boundary.
 

From python-checkins at python.org  Tue Jul 20 22:18:15 2010
From: python-checkins at python.org (vinay.sajip)
Date: Tue, 20 Jul 2010 22:18:15 +0200 (CEST)
Subject: [Python-checkins] r83000 - in python/branches/release26-maint/Lib:
	logging/config.py test/test_logging.py
Message-ID: <20100720201815.25AA9EE995@mail.python.org>

Author: vinay.sajip
Date: Tue Jul 20 22:18:14 2010
New Revision: 83000

Log:
Fixed #9310: backported logging fix from 2.7.

Modified:
   python/branches/release26-maint/Lib/logging/config.py
   python/branches/release26-maint/Lib/test/test_logging.py

Modified: python/branches/release26-maint/Lib/logging/config.py
==============================================================================
--- python/branches/release26-maint/Lib/logging/config.py	(original)
+++ python/branches/release26-maint/Lib/logging/config.py	Tue Jul 20 22:18:14 2010
@@ -104,6 +104,9 @@
 def _strip_spaces(alist):
     return map(lambda x: string.strip(x), alist)
 
+def _encoded(s):
+    return s if isinstance(s, str) else s.encode('utf-8')
+
 def _create_formatters(cp):
     """Create and return formatters"""
     flist = cp.get("formatters", "keys")
@@ -214,7 +217,7 @@
     #avoid disabling child loggers of explicitly
     #named loggers. With a sorted list it is easier
     #to find the child loggers.
-    existing.sort()
+    existing.sort(key=_encoded)
     #We'll keep the list of existing loggers
     #which are children of named loggers here...
     child_loggers = []

Modified: python/branches/release26-maint/Lib/test/test_logging.py
==============================================================================
--- python/branches/release26-maint/Lib/test/test_logging.py	(original)
+++ python/branches/release26-maint/Lib/test/test_logging.py	Tue Jul 20 22:18:14 2010
@@ -69,6 +69,12 @@
         finally:
             logging._releaseLock()
 
+        # Set two unused loggers: one non-ASCII and one Unicode.
+        # This is to test correct operation when sorting existing
+        # loggers in the configuration code. See issues 8201, 9310.
+        logging.getLogger("\xab\xd7\xbb")
+        logging.getLogger(u"\u013f\u00d6\u0047")
+
         self.root_logger = logging.getLogger("")
         self.original_logging_level = self.root_logger.getEffectiveLevel()
 

From python-checkins at python.org  Tue Jul 20 22:30:13 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Tue, 20 Jul 2010 22:30:13 +0200 (CEST)
Subject: [Python-checkins] r83001 - python/branches/release27-maint
Message-ID: <20100720203013.6B715F12E@mail.python.org>

Author: alexander.belopolsky
Date: Tue Jul 20 22:30:13 2010
New Revision: 83001

Log:
Blocked revisions 82997 via svnmerge

........
  r82997 | alexander.belopolsky | 2010-07-20 15:55:18 -0400 (Tue, 20 Jul 2010) | 3 lines
  
  Issue #9282: Fixed --listfuncs option of trace.py.  Thanks Eli
  Bendersky for the patch.
........


Modified:
   python/branches/release27-maint/   (props changed)

From python-checkins at python.org  Wed Jul 21 00:08:46 2010
From: python-checkins at python.org (nick.coghlan)
Date: Wed, 21 Jul 2010 00:08:46 +0200 (CEST)
Subject: [Python-checkins] r83002 - peps/trunk/pep-3150.txt
Message-ID: <20100720220846.77C3DEE99B@mail.python.org>

Author: nick.coghlan
Date: Wed Jul 21 00:08:46 2010
New Revision: 83002

Log:
Updates based on python-ideas discussion

Modified:
   peps/trunk/pep-3150.txt

Modified: peps/trunk/pep-3150.txt
==============================================================================
--- peps/trunk/pep-3150.txt	(original)
+++ peps/trunk/pep-3150.txt	Wed Jul 21 00:08:46 2010
@@ -1,5 +1,5 @@
 PEP: 3150
-Title: Statement local namespaces (aka "where" clause)
+Title: Statement local namespaces (aka "given" clause)
 Version: $Revision$
 Last-Modified: $Date$
 Author: Nick Coghlan <ncoghlan at gmail.com>
@@ -15,7 +15,7 @@
 Abstract
 ========
 
-A recurring proposal on python-ideas is the addition of some form of
+A recurring proposal ([1], [2], [3]) on python-ideas is the addition of some form of
 statement local namespace.
 
 This PEP is intended to serve as a focal point for those ideas, so we
@@ -41,7 +41,7 @@
 Proposal
 ========
 
-This PEP proposes the addition of an optional "where" clause to the
+This PEP proposes the addition of an optional ``given`` clause to the
 syntax for simple statements which may contain an expression. The
 current list of simple statements that would be affected by this
 addition is as follows:
@@ -55,49 +55,107 @@
 * raise statement
 * assert statement
 
-The ``where`` clause would allow subexpressions to be referenced by
+The ``given`` clause would allow subexpressions to be referenced by
 name in the header line, with the actual definitions following in
 the indented clause. As a simple example::
 
- c = sqrt(a*a + b*b) where:
+ c = sqrt(a*a + b*b) given:
     a = retrieve_a()
     b = retrieve_b()
 
-Torture Test
-============
-
-An implementation of this PEP must support execution of the following
-code at module, class and function scope::
-
-  b = {}
-  a = b[f(a)] = x where:
-    x = 42
-    def f(x):
-      return x
-  assert "x" not in locals()
-  assert "f" not in locals()
-  assert a == 42
-  assert d[42] == 42 where:
-    d = b
-  assert "d" not in locals()
-
-Most naive implementations will choke on the first complex assignment,
-while less naive but still broken implementations will fail when
-the torture test is executed at class scope.
-
-And yes, that's a perfectly well-defined assignment statement. Insane,
-I agree, but legal::
-
-  >>> def f(x): return x
-  ... 
-  >>> x = 42
-  >>> b = {}
-  >>> a = b[f(a)] = x
-  >>> a
-  42
-  >>> b
-  {42: 42}
+Rationale
+=========
 
+Some past language features (specifically function decorators
+and list comprehensions) were motivated, at least in part, by
+the desire to give the important parts of a statement more
+prominence when reading code. In the case of function decorators,
+information such as whether or not a method is a class or static
+method can now be found in the function definition rather than
+after the function body. List comprehensions similarly take the
+expression being assigned to each member of the list and move it
+to the beginning of the expression rather than leaving it buried
+inside a ``for`` loop.
+
+The rationale for the ``given`` clause is similar. Currently,
+breaking out a subexpression requires naming that subexpression
+*before* the actual statement of interest. The ``given`` clause
+is designed to allow a programmer to highlight for the reader
+the statement which is actually of interest (and presumably has
+significance for later code) while hiding the most likely irrelevant
+"calculation details" inside an indented suite.
+
+Using the simple example from the proposal section, the current Python
+equivalent would be::
+
+  a = retrieve_a()
+  b = retrieve_b()
+  c = sqrt(a*a + b*b)
+
+If later code is only interested in the value of c, then the
+details involved in retrieving the values of a and b may be an
+unnecessary distraction to the reader (particularly if those
+details are more complicated than the simple function calls
+shown in the example).
+
+To use a more illustrative example (courtesy of Alex Light),
+which of the following is easier to comprehend?
+
+Subexpressions up front?::
+
+  sea = water()
+  temp = get_temperature(sea)
+  depth = get_depth(sea)
+  purity = get_purity(sea)
+  saltiness = get_salinity(sea)
+  size = get_size(sea)
+  density = get_density(sea)
+  desired_property = calc_value(temp, depth, purity,
+                                salinity, size, density)
+  # Further operations using desired_property
+
+Or subexpressions indented?::
+
+  desired_property = calc_value(temp, depth, purity,
+                                salinity, size, density) given:
+    sea = water()
+    temp = get_temperature(sea)
+    depth = get_depth(sea)
+    purity = get_purity(sea)
+    saltiness = get_salinity(sea)
+    size = get_size(sea)
+    density = get_density(sea)
+  # Further operations using desired_property
+
+The ``given`` clause may also function as a more readable
+alternative to some uses of lambda expressions and similar
+constructs when passing one-off functions to operations
+like ``sorted``.
+
+One way to think of the proposed clause is as a middle
+ground between normal in-line code and separation of an
+operation out into a dedicated function.
+
+Keyword Choice
+==============
+
+This proposal initially used ``where`` based on the name of a similar
+construct in Haskell. However, it has been pointed out that there
+are existing Python libraries (such as Numpy [4]) that already use
+``where`` in the SQL query condition sense, making that keyword choice
+potentially confusing.
+
+While ``given`` may also be used as a variable name (and hence would be
+deprecated using the usual ``__future__`` dance for introducing
+new keywords), it is associated much more strongly with the desired
+"here are some extra variables this expression may use" semantics
+for the new clause.
+
+Reusing the ``with`` keyword has also been proposed. This has the
+advantage of avoiding the addition of a new keyword, but also has
+a high potential for confusion as the ``with`` clause and ``with``
+statement would look similar but do completely different things.
+That way lies C++ and Perl :)
 
 Syntax Change
 =============
@@ -117,12 +175,12 @@
 
   expr_stmt: testlist_star_expr (augassign (yield_expr|testlist) |
                ('=' (yield_expr|testlist_star_expr))*) [where_clause]
-  del_stmt: 'del' exprlist [where_clause]
-  return_stmt: 'return' [testlist] [where_clause]
-  yield_stmt: yield_expr [where_clause]
-  raise_stmt: 'raise' [test ['from' test]] [where_clause]
-  assert_stmt: 'assert' test [',' test] [where_clause]
-  where_clause: "where" ":" suite
+  del_stmt: 'del' exprlist [given_clause]
+  return_stmt: 'return' [testlist] [given_clause]
+  yield_stmt: yield_expr [given_clause]
+  raise_stmt: 'raise' [test ['from' test]] [given_clause]
+  assert_stmt: 'assert' test [',' test] [given_clause]
+  given_clause: "given" ":" suite
 
 (Note that expr_stmt in the grammar covers assignment and augmented
 assignment in addition to simple expression statements)
@@ -145,20 +203,22 @@
 non-trivial restructuring of the grammar, breaking up small_stmt and
 flow_stmt to separate the statements that potentially contain arbitrary
 subexpressions and then allowing a single one of those statements with
-a ``where`` clause at the simple_stmt level. Something along the lines of::
+a ``given`` clause at the simple_stmt level. Something along the lines of::
 
+  stmt: simple_stmt | given_stmt | compound_stmt
   simple_stmt: small_stmt (';' small_stmt)* [';'] NEWLINE
-  where_stmt: subexpr_stmt (where_clause | (';' small_stmt)* [';']) NEWLINE
-  subexpr_stmt: expr_stmt | del_stmt | flow_subexpr_stmt | assert_stmt
   small_stmt: (pass_stmt | flow_stmt | import_stmt |
                global_stmt | nonlocal_stmt)
   flow_stmt: break_stmt | continue_stmt
+  given_stmt: subexpr_stmt (given_clause |
+                (';' (small_stmt | subexpr_stmt))* [';']) NEWLINE
+  subexpr_stmt: expr_stmt | del_stmt | flow_subexpr_stmt | assert_stmt
   flow_subexpr_stmt: return_stmt | raise_stmt | yield_stmt
-  where_clause: "where" ":" suite
-
+  given_clause: "given" ":" suite
 
 For reference, here are the current definitions at that level::
 
+  stmt: simple_stmt | compound_stmt
   simple_stmt: small_stmt (';' small_stmt)* [';'] NEWLINE
   small_stmt: (expr_stmt | del_stmt | pass_stmt | flow_stmt |
                import_stmt | global_stmt | nonlocal_stmt | assert_stmt)
@@ -177,19 +237,29 @@
   jump around a little strangely, as the body of the ``where``
   clause is executed before the simple statement in the clause
   header. The closest any other part of Python comes to this
-  before is the out of order evaluation in conditional
-  expressions.
-  
+  is the out of order evaluation in list comprehensions,
+  generator expressions and conditional expressions.
+
+These objections should not be dismissed lightly - the proposal
+in this PEP needs to be subjected to the test of application to
+a large code base (such as the standard library) in a search
+for examples where the readability of real world code is genuinely
+enhanced.
+
+New PEP 8 guidelines would also need to be developed to provide
+appropriate direction on when to use the ``given`` clause over
+ordinary variable assignments.
 
 Possible Additions
 ==================
 
-* The current proposal allows the addition of a ``where`` clause only
+* The current proposal allows the addition of a ``given`` clause only
   for simple statements. Extending the idea to allow the use of
   compound statements would be quite possible, but doing so raises
-  serious readability concerns (as values defined in the ``where``
+  serious readability concerns (as values defined in the ``given``
   clause may be used well before they are defined, exactly the kind
-  of readability trap that decorators were designed to eliminate)
+  of readability trap that other features like decorators and ``with``
+  statements are designed to eliminate)
 
 * Currently only the outermost clause of comprehensions and generator
   expressions can reference the surrounding namespace when executed
@@ -197,7 +267,43 @@
   associated namespace semantics could allow that restriction to be
   lifted. There would be backwards compatibility implications in doing
   so as existing code may be relying on the behaviour of ignoring
-  class level variables.
+  class level variables, but the idea is worth considering.
+
+Torture Test
+============
+
+An implementation of this PEP must support execution of the following
+code at module, class and function scope::
+
+  b = {}
+  a = b[f(a)] = x given:
+    x = 42
+    def f(x):
+      return x
+  assert "x" not in locals()
+  assert "f" not in locals()
+  assert a == 42
+  assert d[42] == 42 given:
+    d = b
+  assert "d" not in locals()
+
+Most naive implementations will choke on the first complex assignment,
+while less naive but still broken implementations will fail when
+the torture test is executed at class scope.
+
+And yes, that's a perfectly well-defined assignment statement. Insane,
+you might rightly say, but legal::
+
+  >>> def f(x): return x
+  ... 
+  >>> x = 42
+  >>> b = {}
+  >>> a = b[f(a)] = x
+  >>> a
+  42
+  >>> b
+  {42: 42}
+
 
 Possible Implementation Strategy
 ================================
@@ -256,6 +362,9 @@
 
 .. [2] http://mail.python.org/pipermail/python-ideas/2010-July/007584.html
 
+.. [3] http://mail.python.org/pipermail/python-ideas/2009-July/005132.html
+
+.. [4] http://mail.python.org/pipermail/python-ideas/2010-July/007596.html
 
 Copyright
 =========

From python-checkins at python.org  Wed Jul 21 00:34:39 2010
From: python-checkins at python.org (nick.coghlan)
Date: Wed, 21 Jul 2010 00:34:39 +0200 (CEST)
Subject: [Python-checkins] r83003 - peps/trunk/pep-3150.txt
Message-ID: <20100720223439.7315CEE992@mail.python.org>

Author: nick.coghlan
Date: Wed Jul 21 00:34:39 2010
New Revision: 83003

Log:
Tidy up code indenting using 3-space top-level indents and 4-space indents within the sample Python code

Modified:
   peps/trunk/pep-3150.txt

Modified: peps/trunk/pep-3150.txt
==============================================================================
--- peps/trunk/pep-3150.txt	(original)
+++ peps/trunk/pep-3150.txt	Wed Jul 21 00:34:39 2010
@@ -59,9 +59,9 @@
 name in the header line, with the actual definitions following in
 the indented clause. As a simple example::
 
- c = sqrt(a*a + b*b) given:
-    a = retrieve_a()
-    b = retrieve_b()
+   c = sqrt(a*a + b*b) given:
+       a = retrieve_a()
+       b = retrieve_b()
 
 Rationale
 =========
@@ -88,9 +88,9 @@
 Using the simple example from the proposal section, the current Python
 equivalent would be::
 
-  a = retrieve_a()
-  b = retrieve_b()
-  c = sqrt(a*a + b*b)
+   a = retrieve_a()
+   b = retrieve_b()
+   c = sqrt(a*a + b*b)
 
 If later code is only interested in the value of c, then the
 details involved in retrieving the values of a and b may be an
@@ -103,29 +103,29 @@
 
 Subexpressions up front?::
 
-  sea = water()
-  temp = get_temperature(sea)
-  depth = get_depth(sea)
-  purity = get_purity(sea)
-  saltiness = get_salinity(sea)
-  size = get_size(sea)
-  density = get_density(sea)
-  desired_property = calc_value(temp, depth, purity,
-                                salinity, size, density)
-  # Further operations using desired_property
+   sea = water()
+   temp = get_temperature(sea)
+   depth = get_depth(sea)
+   purity = get_purity(sea)
+   saltiness = get_salinity(sea)
+   size = get_size(sea)
+   density = get_density(sea)
+   desired_property = calc_value(temp, depth, purity,
+                                 salinity, size, density)
+   # Further operations using desired_property
 
 Or subexpressions indented?::
 
-  desired_property = calc_value(temp, depth, purity,
+   desired_property = calc_value(temp, depth, purity,
                                 salinity, size, density) given:
-    sea = water()
-    temp = get_temperature(sea)
-    depth = get_depth(sea)
-    purity = get_purity(sea)
-    saltiness = get_salinity(sea)
-    size = get_size(sea)
-    density = get_density(sea)
-  # Further operations using desired_property
+       sea = water()
+       temp = get_temperature(sea)
+       depth = get_depth(sea)
+       purity = get_purity(sea)
+       saltiness = get_salinity(sea)
+       size = get_size(sea)
+       density = get_density(sea)
+   # Further operations using desired_property
 
 The ``given`` clause may also function as a more readable
 alternative to some uses of lambda expressions and similar
@@ -162,25 +162,25 @@
 
 Current::
 
-  expr_stmt: testlist_star_expr (augassign (yield_expr|testlist) |
-               ('=' (yield_expr|testlist_star_expr))*)
-  del_stmt: 'del' exprlist
-  return_stmt: 'return' [testlist]
-  yield_stmt: yield_expr
-  raise_stmt: 'raise' [test ['from' test]]
-  assert_stmt: 'assert' test [',' test]
+   expr_stmt: testlist_star_expr (augassign (yield_expr|testlist) |
+                ('=' (yield_expr|testlist_star_expr))*)
+   del_stmt: 'del' exprlist
+   return_stmt: 'return' [testlist]
+   yield_stmt: yield_expr
+   raise_stmt: 'raise' [test ['from' test]]
+   assert_stmt: 'assert' test [',' test]
 
 
 New::
 
-  expr_stmt: testlist_star_expr (augassign (yield_expr|testlist) |
-               ('=' (yield_expr|testlist_star_expr))*) [where_clause]
-  del_stmt: 'del' exprlist [given_clause]
-  return_stmt: 'return' [testlist] [given_clause]
-  yield_stmt: yield_expr [given_clause]
-  raise_stmt: 'raise' [test ['from' test]] [given_clause]
-  assert_stmt: 'assert' test [',' test] [given_clause]
-  given_clause: "given" ":" suite
+   expr_stmt: testlist_star_expr (augassign (yield_expr|testlist) |
+                ('=' (yield_expr|testlist_star_expr))*) [where_clause]
+   del_stmt: 'del' exprlist [given_clause]
+   return_stmt: 'return' [testlist] [given_clause]
+   yield_stmt: yield_expr [given_clause]
+   raise_stmt: 'raise' [test ['from' test]] [given_clause]
+   assert_stmt: 'assert' test [',' test] [given_clause]
+   given_clause: "given" ":" suite
 
 (Note that expr_stmt in the grammar covers assignment and augmented
 assignment in addition to simple expression statements)
@@ -205,24 +205,24 @@
 subexpressions and then allowing a single one of those statements with
 a ``given`` clause at the simple_stmt level. Something along the lines of::
 
-  stmt: simple_stmt | given_stmt | compound_stmt
-  simple_stmt: small_stmt (';' small_stmt)* [';'] NEWLINE
-  small_stmt: (pass_stmt | flow_stmt | import_stmt |
-               global_stmt | nonlocal_stmt)
-  flow_stmt: break_stmt | continue_stmt
-  given_stmt: subexpr_stmt (given_clause |
-                (';' (small_stmt | subexpr_stmt))* [';']) NEWLINE
-  subexpr_stmt: expr_stmt | del_stmt | flow_subexpr_stmt | assert_stmt
-  flow_subexpr_stmt: return_stmt | raise_stmt | yield_stmt
-  given_clause: "given" ":" suite
+   stmt: simple_stmt | given_stmt | compound_stmt
+   simple_stmt: small_stmt (';' small_stmt)* [';'] NEWLINE
+   small_stmt: (pass_stmt | flow_stmt | import_stmt |
+                global_stmt | nonlocal_stmt)
+   flow_stmt: break_stmt | continue_stmt
+   given_stmt: subexpr_stmt (given_clause |
+                 (';' (small_stmt | subexpr_stmt))* [';']) NEWLINE
+   subexpr_stmt: expr_stmt | del_stmt | flow_subexpr_stmt | assert_stmt
+   flow_subexpr_stmt: return_stmt | raise_stmt | yield_stmt
+   given_clause: "given" ":" suite
 
 For reference, here are the current definitions at that level::
 
-  stmt: simple_stmt | compound_stmt
-  simple_stmt: small_stmt (';' small_stmt)* [';'] NEWLINE
-  small_stmt: (expr_stmt | del_stmt | pass_stmt | flow_stmt |
-               import_stmt | global_stmt | nonlocal_stmt | assert_stmt)
-  flow_stmt: break_stmt | continue_stmt | return_stmt | raise_stmt | yield_stmt
+   stmt: simple_stmt | compound_stmt
+   simple_stmt: small_stmt (';' small_stmt)* [';'] NEWLINE
+   small_stmt: (expr_stmt | del_stmt | pass_stmt | flow_stmt |
+                import_stmt | global_stmt | nonlocal_stmt | assert_stmt)
+   flow_stmt: break_stmt | continue_stmt | return_stmt | raise_stmt | yield_stmt
 
 
 Common Objections
@@ -275,17 +275,17 @@
 An implementation of this PEP must support execution of the following
 code at module, class and function scope::
 
-  b = {}
-  a = b[f(a)] = x given:
-    x = 42
-    def f(x):
-      return x
-  assert "x" not in locals()
-  assert "f" not in locals()
-  assert a == 42
-  assert d[42] == 42 given:
-    d = b
-  assert "d" not in locals()
+   b = {}
+   a = b[f(a)] = x given:
+       x = 42
+       def f(x):
+           return x
+   assert "x" not in locals()
+   assert "f" not in locals()
+   assert a == 42
+   assert d[42] == 42 given:
+       d = b
+   assert "d" not in locals()
 
 Most naive implementations will choke on the first complex assignment,
 while less naive but still broken implementations will fail when
@@ -294,15 +294,15 @@
 And yes, that's a perfectly well-defined assignment statement. Insane,
 you might rightly say, but legal::
 
-  >>> def f(x): return x
-  ... 
-  >>> x = 42
-  >>> b = {}
-  >>> a = b[f(a)] = x
-  >>> a
-  42
-  >>> b
-  {42: 42}
+   >>> def f(x): return x
+   ... 
+   >>> x = 42
+   >>> b = {}
+   >>> a = b[f(a)] = x
+   >>> a
+   42
+   >>> b
+   {42: 42}
 
 
 Possible Implementation Strategy
@@ -326,23 +326,23 @@
 bindings between the inner and outer scopes. The torture test above
 would then translate to something like the following::
 
-  b = {}
-  def _anon1(b): # 'b' reference copied in
-    x = 42
-    def f(x):
-      return x
-    a = b[f(a)] = x
-    return a # 'a' reference copied out
-  a = _anon1(b)
-  assert "x" not in locals()
-  assert "f" not in locals()
-  assert a == 42
-  def _anon2(b) # 'b' reference copied in
-    d = b
-    assert d[42] == 42
-    # Nothing to copy out (not an assignment)
-  _anon2()
-  assert "d" not in locals()
+   b = {}
+   def _anon1(b): # 'b' reference copied in
+       x = 42
+       def f(x):
+           return x
+       a = b[f(a)] = x
+       return a # 'a' reference copied out
+   a = _anon1(b)
+   assert "x" not in locals()
+   assert "f" not in locals()
+   assert a == 42
+   def _anon2(b) # 'b' reference copied in
+       d = b
+       assert d[42] == 42
+       # Nothing to copy out (not an assignment)
+   _anon2()
+   assert "d" not in locals()
 
 However, as noted in the abstract, an actual implementation of
 this idea has never been tried.

From python-checkins at python.org  Wed Jul 21 00:36:51 2010
From: python-checkins at python.org (nick.coghlan)
Date: Wed, 21 Jul 2010 00:36:51 +0200 (CEST)
Subject: [Python-checkins] r83004 - peps/trunk/pep-3150.txt
Message-ID: <20100720223651.D4FBAEE992@mail.python.org>

Author: nick.coghlan
Date: Wed Jul 21 00:36:51 2010
New Revision: 83004

Log:
One more indentation nit

Modified:
   peps/trunk/pep-3150.txt

Modified: peps/trunk/pep-3150.txt
==============================================================================
--- peps/trunk/pep-3150.txt	(original)
+++ peps/trunk/pep-3150.txt	Wed Jul 21 00:36:51 2010
@@ -117,7 +117,7 @@
 Or subexpressions indented?::
 
    desired_property = calc_value(temp, depth, purity,
-                                salinity, size, density) given:
+                                 salinity, size, density) given:
        sea = water()
        temp = get_temperature(sea)
        depth = get_depth(sea)

From python-checkins at python.org  Wed Jul 21 00:37:20 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Wed, 21 Jul 2010 00:37:20 +0200 (CEST)
Subject: [Python-checkins] r83005 - in python/branches/py3k:
	Include/opcode.h Include/symtable.h
	Lib/test/test_line_tracing.py Lib/test/test_scope.py
	Lib/test/test_trace.py Objects/typeobject.c Python/ceval.c
	Python/compile.c Python/symtable.c
Message-ID: <20100720223720.36E9FEE992@mail.python.org>

Author: benjamin.peterson
Date: Wed Jul 21 00:37:19 2010
New Revision: 83005

Log:
move test_trace.py so as not to conflict with future tests for the trace module

Added:
   python/branches/py3k/Lib/test/test_line_tracing.py
      - copied unchanged from r83003, /python/branches/py3k/Lib/test/test_trace.py
Removed:
   python/branches/py3k/Lib/test/test_trace.py
Modified:
   python/branches/py3k/Include/opcode.h
   python/branches/py3k/Include/symtable.h
   python/branches/py3k/Lib/test/test_scope.py
   python/branches/py3k/Objects/typeobject.c
   python/branches/py3k/Python/ceval.c
   python/branches/py3k/Python/compile.c
   python/branches/py3k/Python/symtable.c

Modified: python/branches/py3k/Include/opcode.h
==============================================================================
--- python/branches/py3k/Include/opcode.h	(original)
+++ python/branches/py3k/Include/opcode.h	Wed Jul 21 00:37:19 2010
@@ -123,6 +123,7 @@
 #define LOAD_CLOSURE    135     /* Load free variable from closure */
 #define LOAD_DEREF      136     /* Load and dereference from closure cell */ 
 #define STORE_DEREF     137     /* Store into cell */ 
+#define LOAD_NAME_LOCAL_ONLY 138
 
 /* The next 3 opcodes must be contiguous and satisfy
    (CALL_FUNCTION_VAR - CALL_FUNCTION) & 3 == 1  */

Modified: python/branches/py3k/Include/symtable.h
==============================================================================
--- python/branches/py3k/Include/symtable.h	(original)
+++ python/branches/py3k/Include/symtable.h	Wed Jul 21 00:37:19 2010
@@ -88,6 +88,7 @@
 #define GLOBAL_IMPLICIT 3
 #define FREE 4
 #define CELL 5
+#define LOCAL_ONLY 6
 
 /* The following two names are used for the ste_unoptimized bit field */
 #define OPT_IMPORT_STAR 1

Modified: python/branches/py3k/Lib/test/test_scope.py
==============================================================================
--- python/branches/py3k/Lib/test/test_scope.py	(original)
+++ python/branches/py3k/Lib/test/test_scope.py	Wed Jul 21 00:37:19 2010
@@ -690,6 +690,14 @@
         h = g()
         self.assertEqual(h(), 3)
 
+    def testLocalClosureShadowing(self):
+        exec("""
+x = 4
+def f(x):
+    class C:
+         x = x
+raises(NameError, f, 3)""", {"raises" : self.assertRaises})
+
 
 def test_main():
     run_unittest(ScopeTests)

Deleted: python/branches/py3k/Lib/test/test_trace.py
==============================================================================
--- python/branches/py3k/Lib/test/test_trace.py	Wed Jul 21 00:37:19 2010
+++ (empty file)
@@ -1,790 +0,0 @@
-# Testing the line trace facility.
-
-from test import support
-import unittest
-import sys
-import difflib
-import gc
-
-# A very basic example.  If this fails, we're in deep trouble.
-def basic():
-    return 1
-
-basic.events = [(0, 'call'),
-                (1, 'line'),
-                (1, 'return')]
-
-# Many of the tests below are tricky because they involve pass statements.
-# If there is implicit control flow around a pass statement (in an except
-# clause or else caluse) under what conditions do you set a line number
-# following that clause?
-
-
-# The entire "while 0:" statement is optimized away.  No code
-# exists for it, so the line numbers skip directly from "del x"
-# to "x = 1".
-def arigo_example():
-    x = 1
-    del x
-    while 0:
-        pass
-    x = 1
-
-arigo_example.events = [(0, 'call'),
-                        (1, 'line'),
-                        (2, 'line'),
-                        (5, 'line'),
-                        (5, 'return')]
-
-# check that lines consisting of just one instruction get traced:
-def one_instr_line():
-    x = 1
-    del x
-    x = 1
-
-one_instr_line.events = [(0, 'call'),
-                         (1, 'line'),
-                         (2, 'line'),
-                         (3, 'line'),
-                         (3, 'return')]
-
-def no_pop_tops():      # 0
-    x = 1               # 1
-    for a in range(2):  # 2
-        if a:           # 3
-            x = 1       # 4
-        else:           # 5
-            x = 1       # 6
-
-no_pop_tops.events = [(0, 'call'),
-                      (1, 'line'),
-                      (2, 'line'),
-                      (3, 'line'),
-                      (6, 'line'),
-                      (2, 'line'),
-                      (3, 'line'),
-                      (4, 'line'),
-                      (2, 'line'),
-                      (2, 'return')]
-
-def no_pop_blocks():
-    y = 1
-    while not y:
-        bla
-    x = 1
-
-no_pop_blocks.events = [(0, 'call'),
-                        (1, 'line'),
-                        (2, 'line'),
-                        (4, 'line'),
-                        (4, 'return')]
-
-def called(): # line -3
-    x = 1
-
-def call():   # line 0
-    called()
-
-call.events = [(0, 'call'),
-               (1, 'line'),
-               (-3, 'call'),
-               (-2, 'line'),
-               (-2, 'return'),
-               (1, 'return')]
-
-def raises():
-    raise Exception
-
-def test_raise():
-    try:
-        raises()
-    except Exception as exc:
-        x = 1
-
-test_raise.events = [(0, 'call'),
-                     (1, 'line'),
-                     (2, 'line'),
-                     (-3, 'call'),
-                     (-2, 'line'),
-                     (-2, 'exception'),
-                     (-2, 'return'),
-                     (2, 'exception'),
-                     (3, 'line'),
-                     (4, 'line'),
-                     (4, 'return')]
-
-def _settrace_and_return(tracefunc):
-    sys.settrace(tracefunc)
-    sys._getframe().f_back.f_trace = tracefunc
-def settrace_and_return(tracefunc):
-    _settrace_and_return(tracefunc)
-
-settrace_and_return.events = [(1, 'return')]
-
-def _settrace_and_raise(tracefunc):
-    sys.settrace(tracefunc)
-    sys._getframe().f_back.f_trace = tracefunc
-    raise RuntimeError
-def settrace_and_raise(tracefunc):
-    try:
-        _settrace_and_raise(tracefunc)
-    except RuntimeError as exc:
-        pass
-
-settrace_and_raise.events = [(2, 'exception'),
-                             (3, 'line'),
-                             (4, 'line'),
-                             (4, 'return')]
-
-# implicit return example
-# This test is interesting because of the else: pass
-# part of the code.  The code generate for the true
-# part of the if contains a jump past the else branch.
-# The compiler then generates an implicit "return None"
-# Internally, the compiler visits the pass statement
-# and stores its line number for use on the next instruction.
-# The next instruction is the implicit return None.
-def ireturn_example():
-    a = 5
-    b = 5
-    if a == b:
-        b = a+1
-    else:
-        pass
-
-ireturn_example.events = [(0, 'call'),
-                          (1, 'line'),
-                          (2, 'line'),
-                          (3, 'line'),
-                          (4, 'line'),
-                          (6, 'line'),
-                          (6, 'return')]
-
-# Tight loop with while(1) example (SF #765624)
-def tightloop_example():
-    items = range(0, 3)
-    try:
-        i = 0
-        while 1:
-            b = items[i]; i+=1
-    except IndexError:
-        pass
-
-tightloop_example.events = [(0, 'call'),
-                            (1, 'line'),
-                            (2, 'line'),
-                            (3, 'line'),
-                            (4, 'line'),
-                            (5, 'line'),
-                            (5, 'line'),
-                            (5, 'line'),
-                            (5, 'line'),
-                            (5, 'exception'),
-                            (6, 'line'),
-                            (7, 'line'),
-                            (7, 'return')]
-
-def tighterloop_example():
-    items = range(1, 4)
-    try:
-        i = 0
-        while 1: i = items[i]
-    except IndexError:
-        pass
-
-tighterloop_example.events = [(0, 'call'),
-                            (1, 'line'),
-                            (2, 'line'),
-                            (3, 'line'),
-                            (4, 'line'),
-                            (4, 'line'),
-                            (4, 'line'),
-                            (4, 'line'),
-                            (4, 'exception'),
-                            (5, 'line'),
-                            (6, 'line'),
-                            (6, 'return')]
-
-def generator_function():
-    try:
-        yield True
-        "continued"
-    finally:
-        "finally"
-def generator_example():
-    # any() will leave the generator before its end
-    x = any(generator_function())
-
-    # the following lines were not traced
-    for x in range(10):
-        y = x
-
-generator_example.events = ([(0, 'call'),
-                             (2, 'line'),
-                             (-6, 'call'),
-                             (-5, 'line'),
-                             (-4, 'line'),
-                             (-4, 'return'),
-                             (-4, 'call'),
-                             (-4, 'exception'),
-                             (-1, 'line'),
-                             (-1, 'return')] +
-                            [(5, 'line'), (6, 'line')] * 10 +
-                            [(5, 'line'), (5, 'return')])
-
-
-class Tracer:
-    def __init__(self):
-        self.events = []
-    def trace(self, frame, event, arg):
-        self.events.append((frame.f_lineno, event))
-        return self.trace
-    def traceWithGenexp(self, frame, event, arg):
-        (o for o in [1])
-        self.events.append((frame.f_lineno, event))
-        return self.trace
-
-class TraceTestCase(unittest.TestCase):
-
-    # Disable gc collection when tracing, otherwise the
-    # deallocators may be traced as well.
-    def setUp(self):
-        self.using_gc = gc.isenabled()
-        gc.disable()
-
-    def tearDown(self):
-        if self.using_gc:
-            gc.enable()
-
-    def compare_events(self, line_offset, events, expected_events):
-        events = [(l - line_offset, e) for (l, e) in events]
-        if events != expected_events:
-            self.fail(
-                "events did not match expectation:\n" +
-                "\n".join(difflib.ndiff([str(x) for x in expected_events],
-                                        [str(x) for x in events])))
-
-    def run_and_compare(self, func, events):
-        tracer = Tracer()
-        sys.settrace(tracer.trace)
-        func()
-        sys.settrace(None)
-        self.compare_events(func.__code__.co_firstlineno,
-                            tracer.events, events)
-
-    def run_test(self, func):
-        self.run_and_compare(func, func.events)
-
-    def run_test2(self, func):
-        tracer = Tracer()
-        func(tracer.trace)
-        sys.settrace(None)
-        self.compare_events(func.__code__.co_firstlineno,
-                            tracer.events, func.events)
-
-    def set_and_retrieve_none(self):
-        sys.settrace(None)
-        assert sys.gettrace() is None
-
-    def set_and_retrieve_func(self):
-        def fn(*args):
-            pass
-
-        sys.settrace(fn)
-        try:
-            assert sys.gettrace() is fn
-        finally:
-            sys.settrace(None)
-
-    def test_01_basic(self):
-        self.run_test(basic)
-    def test_02_arigo(self):
-        self.run_test(arigo_example)
-    def test_03_one_instr(self):
-        self.run_test(one_instr_line)
-    def test_04_no_pop_blocks(self):
-        self.run_test(no_pop_blocks)
-    def test_05_no_pop_tops(self):
-        self.run_test(no_pop_tops)
-    def test_06_call(self):
-        self.run_test(call)
-    def test_07_raise(self):
-        self.run_test(test_raise)
-
-    def test_08_settrace_and_return(self):
-        self.run_test2(settrace_and_return)
-    def test_09_settrace_and_raise(self):
-        self.run_test2(settrace_and_raise)
-    def test_10_ireturn(self):
-        self.run_test(ireturn_example)
-    def test_11_tightloop(self):
-        self.run_test(tightloop_example)
-    def test_12_tighterloop(self):
-        self.run_test(tighterloop_example)
-
-    def test_13_genexp(self):
-        self.run_test(generator_example)
-        # issue1265: if the trace function contains a generator,
-        # and if the traced function contains another generator
-        # that is not completely exhausted, the trace stopped.
-        # Worse: the 'finally' clause was not invoked.
-        tracer = Tracer()
-        sys.settrace(tracer.traceWithGenexp)
-        generator_example()
-        sys.settrace(None)
-        self.compare_events(generator_example.__code__.co_firstlineno,
-                            tracer.events, generator_example.events)
-
-    def test_14_onliner_if(self):
-        def onliners():
-            if True: False
-            else: True
-            return 0
-        self.run_and_compare(
-            onliners,
-            [(0, 'call'),
-             (1, 'line'),
-             (3, 'line'),
-             (3, 'return')])
-
-    def test_15_loops(self):
-        # issue1750076: "while" expression is skipped by debugger
-        def for_example():
-            for x in range(2):
-                pass
-        self.run_and_compare(
-            for_example,
-            [(0, 'call'),
-             (1, 'line'),
-             (2, 'line'),
-             (1, 'line'),
-             (2, 'line'),
-             (1, 'line'),
-             (1, 'return')])
-
-        def while_example():
-            # While expression should be traced on every loop
-            x = 2
-            while x > 0:
-                x -= 1
-        self.run_and_compare(
-            while_example,
-            [(0, 'call'),
-             (2, 'line'),
-             (3, 'line'),
-             (4, 'line'),
-             (3, 'line'),
-             (4, 'line'),
-             (3, 'line'),
-             (3, 'return')])
-
-    def test_16_blank_lines(self):
-        namespace = {}
-        exec("def f():\n" + "\n" * 256 + "    pass", namespace)
-        self.run_and_compare(
-            namespace["f"],
-            [(0, 'call'),
-             (257, 'line'),
-             (257, 'return')])
-
-
-class RaisingTraceFuncTestCase(unittest.TestCase):
-    def trace(self, frame, event, arg):
-        """A trace function that raises an exception in response to a
-        specific trace event."""
-        if event == self.raiseOnEvent:
-            raise ValueError # just something that isn't RuntimeError
-        else:
-            return self.trace
-
-    def f(self):
-        """The function to trace; raises an exception if that's the case
-        we're testing, so that the 'exception' trace event fires."""
-        if self.raiseOnEvent == 'exception':
-            x = 0
-            y = 1/x
-        else:
-            return 1
-
-    def run_test_for_event(self, event):
-        """Tests that an exception raised in response to the given event is
-        handled OK."""
-        self.raiseOnEvent = event
-        try:
-            for i in range(sys.getrecursionlimit() + 1):
-                sys.settrace(self.trace)
-                try:
-                    self.f()
-                except ValueError:
-                    pass
-                else:
-                    self.fail("exception not thrown!")
-        except RuntimeError:
-            self.fail("recursion counter not reset")
-
-    # Test the handling of exceptions raised by each kind of trace event.
-    def test_call(self):
-        self.run_test_for_event('call')
-    def test_line(self):
-        self.run_test_for_event('line')
-    def test_return(self):
-        self.run_test_for_event('return')
-    def test_exception(self):
-        self.run_test_for_event('exception')
-
-    def test_trash_stack(self):
-        def f():
-            for i in range(5):
-                print(i)  # line tracing will raise an exception at this line
-
-        def g(frame, why, extra):
-            if (why == 'line' and
-                frame.f_lineno == f.__code__.co_firstlineno + 2):
-                raise RuntimeError("i am crashing")
-            return g
-
-        sys.settrace(g)
-        try:
-            f()
-        except RuntimeError:
-            # the test is really that this doesn't segfault:
-            import gc
-            gc.collect()
-        else:
-            self.fail("exception not propagated")
-
-
-# 'Jump' tests: assigning to frame.f_lineno within a trace function
-# moves the execution position - it's how debuggers implement a Jump
-# command (aka. "Set next statement").
-
-class JumpTracer:
-    """Defines a trace function that jumps from one place to another,
-    with the source and destination lines of the jump being defined by
-    the 'jump' property of the function under test."""
-
-    def __init__(self, function):
-        self.function = function
-        self.jumpFrom = function.jump[0]
-        self.jumpTo = function.jump[1]
-        self.done = False
-
-    def trace(self, frame, event, arg):
-        if not self.done and frame.f_code == self.function.__code__:
-            firstLine = frame.f_code.co_firstlineno
-            if event == 'line' and frame.f_lineno == firstLine + self.jumpFrom:
-                # Cope with non-integer self.jumpTo (because of
-                # no_jump_to_non_integers below).
-                try:
-                    frame.f_lineno = firstLine + self.jumpTo
-                except TypeError:
-                    frame.f_lineno = self.jumpTo
-                self.done = True
-        return self.trace
-
-# The first set of 'jump' tests are for things that are allowed:
-
-def jump_simple_forwards(output):
-    output.append(1)
-    output.append(2)
-    output.append(3)
-
-jump_simple_forwards.jump = (1, 3)
-jump_simple_forwards.output = [3]
-
-def jump_simple_backwards(output):
-    output.append(1)
-    output.append(2)
-
-jump_simple_backwards.jump = (2, 1)
-jump_simple_backwards.output = [1, 1, 2]
-
-def jump_out_of_block_forwards(output):
-    for i in 1, 2:
-        output.append(2)
-        for j in [3]:  # Also tests jumping over a block
-            output.append(4)
-    output.append(5)
-
-jump_out_of_block_forwards.jump = (3, 5)
-jump_out_of_block_forwards.output = [2, 5]
-
-def jump_out_of_block_backwards(output):
-    output.append(1)
-    for i in [1]:
-        output.append(3)
-        for j in [2]:  # Also tests jumping over a block
-            output.append(5)
-        output.append(6)
-    output.append(7)
-
-jump_out_of_block_backwards.jump = (6, 1)
-jump_out_of_block_backwards.output = [1, 3, 5, 1, 3, 5, 6, 7]
-
-def jump_to_codeless_line(output):
-    output.append(1)
-    # Jumping to this line should skip to the next one.
-    output.append(3)
-
-jump_to_codeless_line.jump = (1, 2)
-jump_to_codeless_line.output = [3]
-
-def jump_to_same_line(output):
-    output.append(1)
-    output.append(2)
-    output.append(3)
-
-jump_to_same_line.jump = (2, 2)
-jump_to_same_line.output = [1, 2, 3]
-
-# Tests jumping within a finally block, and over one.
-def jump_in_nested_finally(output):
-    try:
-        output.append(2)
-    finally:
-        output.append(4)
-        try:
-            output.append(6)
-        finally:
-            output.append(8)
-        output.append(9)
-
-jump_in_nested_finally.jump = (4, 9)
-jump_in_nested_finally.output = [2, 9]
-
-# The second set of 'jump' tests are for things that are not allowed:
-
-def no_jump_too_far_forwards(output):
-    try:
-        output.append(2)
-        output.append(3)
-    except ValueError as e:
-        output.append('after' in str(e))
-
-no_jump_too_far_forwards.jump = (3, 6)
-no_jump_too_far_forwards.output = [2, True]
-
-def no_jump_too_far_backwards(output):
-    try:
-        output.append(2)
-        output.append(3)
-    except ValueError as e:
-        output.append('before' in str(e))
-
-no_jump_too_far_backwards.jump = (3, -1)
-no_jump_too_far_backwards.output = [2, True]
-
-# Test each kind of 'except' line.
-def no_jump_to_except_1(output):
-    try:
-        output.append(2)
-    except:
-        e = sys.exc_info()[1]
-        output.append('except' in str(e))
-
-no_jump_to_except_1.jump = (2, 3)
-no_jump_to_except_1.output = [True]
-
-def no_jump_to_except_2(output):
-    try:
-        output.append(2)
-    except ValueError:
-        e = sys.exc_info()[1]
-        output.append('except' in str(e))
-
-no_jump_to_except_2.jump = (2, 3)
-no_jump_to_except_2.output = [True]
-
-def no_jump_to_except_3(output):
-    try:
-        output.append(2)
-    except ValueError as e:
-        output.append('except' in str(e))
-
-no_jump_to_except_3.jump = (2, 3)
-no_jump_to_except_3.output = [True]
-
-def no_jump_to_except_4(output):
-    try:
-        output.append(2)
-    except (ValueError, RuntimeError) as e:
-        output.append('except' in str(e))
-
-no_jump_to_except_4.jump = (2, 3)
-no_jump_to_except_4.output = [True]
-
-def no_jump_forwards_into_block(output):
-    try:
-        output.append(2)
-        for i in 1, 2:
-            output.append(4)
-    except ValueError as e:
-        output.append('into' in str(e))
-
-no_jump_forwards_into_block.jump = (2, 4)
-no_jump_forwards_into_block.output = [True]
-
-def no_jump_backwards_into_block(output):
-    try:
-        for i in 1, 2:
-            output.append(3)
-        output.append(4)
-    except ValueError as e:
-        output.append('into' in str(e))
-
-no_jump_backwards_into_block.jump = (4, 3)
-no_jump_backwards_into_block.output = [3, 3, True]
-
-def no_jump_into_finally_block(output):
-    try:
-        try:
-            output.append(3)
-            x = 1
-        finally:
-            output.append(6)
-    except ValueError as e:
-        output.append('finally' in str(e))
-
-no_jump_into_finally_block.jump = (4, 6)
-no_jump_into_finally_block.output = [3, 6, True]  # The 'finally' still runs
-
-def no_jump_out_of_finally_block(output):
-    try:
-        try:
-            output.append(3)
-        finally:
-            output.append(5)
-            output.append(6)
-    except ValueError as e:
-        output.append('finally' in str(e))
-
-no_jump_out_of_finally_block.jump = (5, 1)
-no_jump_out_of_finally_block.output = [3, True]
-
-# This verifies the line-numbers-must-be-integers rule.
-def no_jump_to_non_integers(output):
-    try:
-        output.append(2)
-    except ValueError as e:
-        output.append('integer' in str(e))
-
-no_jump_to_non_integers.jump = (2, "Spam")
-no_jump_to_non_integers.output = [True]
-
-# This verifies that you can't set f_lineno via _getframe or similar
-# trickery.
-def no_jump_without_trace_function():
-    try:
-        previous_frame = sys._getframe().f_back
-        previous_frame.f_lineno = previous_frame.f_lineno
-    except ValueError as e:
-        # This is the exception we wanted; make sure the error message
-        # talks about trace functions.
-        if 'trace' not in str(e):
-            raise
-    else:
-        # Something's wrong - the expected exception wasn't raised.
-        raise RuntimeError("Trace-function-less jump failed to fail")
-
-
-class JumpTestCase(unittest.TestCase):
-    def compare_jump_output(self, expected, received):
-        if received != expected:
-            self.fail( "Outputs don't match:\n" +
-                       "Expected: " + repr(expected) + "\n" +
-                       "Received: " + repr(received))
-
-    def run_test(self, func):
-        tracer = JumpTracer(func)
-        sys.settrace(tracer.trace)
-        output = []
-        func(output)
-        sys.settrace(None)
-        self.compare_jump_output(func.output, output)
-
-    def test_01_jump_simple_forwards(self):
-        self.run_test(jump_simple_forwards)
-    def test_02_jump_simple_backwards(self):
-        self.run_test(jump_simple_backwards)
-    def test_03_jump_out_of_block_forwards(self):
-        self.run_test(jump_out_of_block_forwards)
-    def test_04_jump_out_of_block_backwards(self):
-        self.run_test(jump_out_of_block_backwards)
-    def test_05_jump_to_codeless_line(self):
-        self.run_test(jump_to_codeless_line)
-    def test_06_jump_to_same_line(self):
-        self.run_test(jump_to_same_line)
-    def test_07_jump_in_nested_finally(self):
-        self.run_test(jump_in_nested_finally)
-    def test_08_no_jump_too_far_forwards(self):
-        self.run_test(no_jump_too_far_forwards)
-    def test_09_no_jump_too_far_backwards(self):
-        self.run_test(no_jump_too_far_backwards)
-    def test_10_no_jump_to_except_1(self):
-        self.run_test(no_jump_to_except_1)
-    def test_11_no_jump_to_except_2(self):
-        self.run_test(no_jump_to_except_2)
-    def test_12_no_jump_to_except_3(self):
-        self.run_test(no_jump_to_except_3)
-    def test_13_no_jump_to_except_4(self):
-        self.run_test(no_jump_to_except_4)
-    def test_14_no_jump_forwards_into_block(self):
-        self.run_test(no_jump_forwards_into_block)
-    def test_15_no_jump_backwards_into_block(self):
-        self.run_test(no_jump_backwards_into_block)
-    def test_16_no_jump_into_finally_block(self):
-        self.run_test(no_jump_into_finally_block)
-    def test_17_no_jump_out_of_finally_block(self):
-        self.run_test(no_jump_out_of_finally_block)
-    def test_18_no_jump_to_non_integers(self):
-        self.run_test(no_jump_to_non_integers)
-    def test_19_no_jump_without_trace_function(self):
-        no_jump_without_trace_function()
-
-    def test_20_large_function(self):
-        d = {}
-        exec("""def f(output):        # line 0
-            x = 0                     # line 1
-            y = 1                     # line 2
-            '''                       # line 3
-            %s                        # lines 4-1004
-            '''                       # line 1005
-            x += 1                    # line 1006
-            output.append(x)          # line 1007
-            return""" % ('\n' * 1000,), d)
-        f = d['f']
-
-        f.jump = (2, 1007)
-        f.output = [0]
-        self.run_test(f)
-
-    def test_jump_to_firstlineno(self):
-        # This tests that PDB can jump back to the first line in a
-        # file.  See issue #1689458.  It can only be triggered in a
-        # function call if the function is defined on a single line.
-        code = compile("""
-# Comments don't count.
-output.append(2)  # firstlineno is here.
-output.append(3)
-output.append(4)
-""", "<fake module>", "exec")
-        class fake_function:
-            __code__ = code
-            jump = (2, 0)
-        tracer = JumpTracer(fake_function)
-        sys.settrace(tracer.trace)
-        namespace = {"output": []}
-        exec(code, namespace)
-        sys.settrace(None)
-        self.compare_jump_output([2, 3, 2, 3, 4], namespace["output"])
-
-
-def test_main():
-    support.run_unittest(
-        TraceTestCase,
-        RaisingTraceFuncTestCase,
-        JumpTestCase
-    )
-
-if __name__ == "__main__":
-    test_main()

Modified: python/branches/py3k/Objects/typeobject.c
==============================================================================
--- python/branches/py3k/Objects/typeobject.c	(original)
+++ python/branches/py3k/Objects/typeobject.c	Wed Jul 21 00:37:19 2010
@@ -6094,6 +6094,7 @@
     PyErr_SetString(PyExc_TypeError,
                     "super(type, obj): "
                     "obj must be an instance or subtype of type");
+    printf("%s\n", type->tp_name);
     return NULL;
 }
 

Modified: python/branches/py3k/Python/ceval.c
==============================================================================
--- python/branches/py3k/Python/ceval.c	(original)
+++ python/branches/py3k/Python/ceval.c	Wed Jul 21 00:37:19 2010
@@ -2052,6 +2052,7 @@
             break;
 
         TARGET(LOAD_NAME)
+        TARGET(LOAD_NAME_LOCAL_ONLY)
             w = GETITEM(names, oparg);
             if ((v = f->f_locals) == NULL) {
                 PyErr_Format(PyExc_SystemError,
@@ -2073,15 +2074,14 @@
                 }
             }
             if (x == NULL) {
-                x = PyDict_GetItem(f->f_globals, w);
+                if (opcode != LOAD_NAME_LOCAL_ONLY) {
+                    x = PyDict_GetItem(f->f_globals, w);
+                    if (x == NULL)
+                        x = PyDict_GetItem(f->f_builtins, w);
+                }
                 if (x == NULL) {
-                    x = PyDict_GetItem(f->f_builtins, w);
-                    if (x == NULL) {
-                        format_exc_check_arg(
-                                    PyExc_NameError,
-                                    NAME_ERROR_MSG, w);
-                        break;
-                    }
+                    format_exc_check_arg(PyExc_NameError, NAME_ERROR_MSG, w);
+                    break;
                 }
                 Py_INCREF(x);
             }

Modified: python/branches/py3k/Python/compile.c
==============================================================================
--- python/branches/py3k/Python/compile.c	(original)
+++ python/branches/py3k/Python/compile.c	Wed Jul 21 00:37:19 2010
@@ -787,6 +787,7 @@
         case LOAD_CONST:
             return 1;
         case LOAD_NAME:
+        case LOAD_NAME_LOCAL_ONLY:
             return 1;
         case BUILD_TUPLE:
         case BUILD_LIST:
@@ -2481,6 +2482,7 @@
         optype = OP_DEREF;
         break;
     case LOCAL:
+    case LOCAL_ONLY:
         if (c->u->u_ste->ste_type == FunctionBlock)
             optype = OP_FAST;
         break;
@@ -2556,7 +2558,7 @@
         break;
     case OP_NAME:
         switch (ctx) {
-        case Load: op = LOAD_NAME; break;
+        case Load: op = (scope == LOCAL_ONLY) ? LOAD_NAME_LOCAL_ONLY : LOAD_NAME; break;
         case Store: op = STORE_NAME; break;
         case Del: op = DELETE_NAME; break;
         case AugLoad:

Modified: python/branches/py3k/Python/symtable.c
==============================================================================
--- python/branches/py3k/Python/symtable.c	(original)
+++ python/branches/py3k/Python/symtable.c	Wed Jul 21 00:37:19 2010
@@ -432,7 +432,14 @@
         return PySet_Add(free, name) >= 0;
     }
     if (flags & DEF_BOUND) {
-        SET_SCOPE(scopes, name, LOCAL);
+        if (ste->ste_type == ClassBlock &&
+            !(flags & DEF_PARAM) &&
+            bound && PySet_Contains(bound, name)) {
+            SET_SCOPE(scopes, name, LOCAL_ONLY);
+        }
+        else {
+            SET_SCOPE(scopes, name, LOCAL);
+        }
         if (PySet_Add(local, name) < 0)
             return 0;
         if (PySet_Discard(global, name) < 0)
@@ -489,7 +496,7 @@
         long scope;
         assert(PyLong_Check(v));
         scope = PyLong_AS_LONG(v);
-        if (scope != LOCAL)
+        if (scope != LOCAL && scope != LOCAL_ONLY)
             continue;
         if (!PySet_Contains(free, name))
             continue;

From python-checkins at python.org  Wed Jul 21 00:39:35 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Wed, 21 Jul 2010 00:39:35 +0200 (CEST)
Subject: [Python-checkins] r83006 - in python/branches/py3k:
	Include/opcode.h Include/symtable.h Lib/test/test_scope.py
	Objects/typeobject.c Python/ceval.c Python/compile.c
	Python/symtable.c
Message-ID: <20100720223935.2EAEBD422@mail.python.org>

Author: benjamin.peterson
Date: Wed Jul 21 00:39:34 2010
New Revision: 83006

Log:
revert unintended changes

Modified:
   python/branches/py3k/Include/opcode.h
   python/branches/py3k/Include/symtable.h
   python/branches/py3k/Lib/test/test_scope.py
   python/branches/py3k/Objects/typeobject.c
   python/branches/py3k/Python/ceval.c
   python/branches/py3k/Python/compile.c
   python/branches/py3k/Python/symtable.c

Modified: python/branches/py3k/Include/opcode.h
==============================================================================
--- python/branches/py3k/Include/opcode.h	(original)
+++ python/branches/py3k/Include/opcode.h	Wed Jul 21 00:39:34 2010
@@ -123,7 +123,6 @@
 #define LOAD_CLOSURE    135     /* Load free variable from closure */
 #define LOAD_DEREF      136     /* Load and dereference from closure cell */ 
 #define STORE_DEREF     137     /* Store into cell */ 
-#define LOAD_NAME_LOCAL_ONLY 138
 
 /* The next 3 opcodes must be contiguous and satisfy
    (CALL_FUNCTION_VAR - CALL_FUNCTION) & 3 == 1  */

Modified: python/branches/py3k/Include/symtable.h
==============================================================================
--- python/branches/py3k/Include/symtable.h	(original)
+++ python/branches/py3k/Include/symtable.h	Wed Jul 21 00:39:34 2010
@@ -88,7 +88,6 @@
 #define GLOBAL_IMPLICIT 3
 #define FREE 4
 #define CELL 5
-#define LOCAL_ONLY 6
 
 /* The following two names are used for the ste_unoptimized bit field */
 #define OPT_IMPORT_STAR 1

Modified: python/branches/py3k/Lib/test/test_scope.py
==============================================================================
--- python/branches/py3k/Lib/test/test_scope.py	(original)
+++ python/branches/py3k/Lib/test/test_scope.py	Wed Jul 21 00:39:34 2010
@@ -690,14 +690,6 @@
         h = g()
         self.assertEqual(h(), 3)
 
-    def testLocalClosureShadowing(self):
-        exec("""
-x = 4
-def f(x):
-    class C:
-         x = x
-raises(NameError, f, 3)""", {"raises" : self.assertRaises})
-
 
 def test_main():
     run_unittest(ScopeTests)

Modified: python/branches/py3k/Objects/typeobject.c
==============================================================================
--- python/branches/py3k/Objects/typeobject.c	(original)
+++ python/branches/py3k/Objects/typeobject.c	Wed Jul 21 00:39:34 2010
@@ -6094,7 +6094,6 @@
     PyErr_SetString(PyExc_TypeError,
                     "super(type, obj): "
                     "obj must be an instance or subtype of type");
-    printf("%s\n", type->tp_name);
     return NULL;
 }
 

Modified: python/branches/py3k/Python/ceval.c
==============================================================================
--- python/branches/py3k/Python/ceval.c	(original)
+++ python/branches/py3k/Python/ceval.c	Wed Jul 21 00:39:34 2010
@@ -2052,7 +2052,6 @@
             break;
 
         TARGET(LOAD_NAME)
-        TARGET(LOAD_NAME_LOCAL_ONLY)
             w = GETITEM(names, oparg);
             if ((v = f->f_locals) == NULL) {
                 PyErr_Format(PyExc_SystemError,
@@ -2074,14 +2073,15 @@
                 }
             }
             if (x == NULL) {
-                if (opcode != LOAD_NAME_LOCAL_ONLY) {
-                    x = PyDict_GetItem(f->f_globals, w);
-                    if (x == NULL)
-                        x = PyDict_GetItem(f->f_builtins, w);
-                }
+                x = PyDict_GetItem(f->f_globals, w);
                 if (x == NULL) {
-                    format_exc_check_arg(PyExc_NameError, NAME_ERROR_MSG, w);
-                    break;
+                    x = PyDict_GetItem(f->f_builtins, w);
+                    if (x == NULL) {
+                        format_exc_check_arg(
+                                    PyExc_NameError,
+                                    NAME_ERROR_MSG, w);
+                        break;
+                    }
                 }
                 Py_INCREF(x);
             }

Modified: python/branches/py3k/Python/compile.c
==============================================================================
--- python/branches/py3k/Python/compile.c	(original)
+++ python/branches/py3k/Python/compile.c	Wed Jul 21 00:39:34 2010
@@ -787,7 +787,6 @@
         case LOAD_CONST:
             return 1;
         case LOAD_NAME:
-        case LOAD_NAME_LOCAL_ONLY:
             return 1;
         case BUILD_TUPLE:
         case BUILD_LIST:
@@ -2482,7 +2481,6 @@
         optype = OP_DEREF;
         break;
     case LOCAL:
-    case LOCAL_ONLY:
         if (c->u->u_ste->ste_type == FunctionBlock)
             optype = OP_FAST;
         break;
@@ -2558,7 +2556,7 @@
         break;
     case OP_NAME:
         switch (ctx) {
-        case Load: op = (scope == LOCAL_ONLY) ? LOAD_NAME_LOCAL_ONLY : LOAD_NAME; break;
+        case Load: op = LOAD_NAME; break;
         case Store: op = STORE_NAME; break;
         case Del: op = DELETE_NAME; break;
         case AugLoad:

Modified: python/branches/py3k/Python/symtable.c
==============================================================================
--- python/branches/py3k/Python/symtable.c	(original)
+++ python/branches/py3k/Python/symtable.c	Wed Jul 21 00:39:34 2010
@@ -432,14 +432,7 @@
         return PySet_Add(free, name) >= 0;
     }
     if (flags & DEF_BOUND) {
-        if (ste->ste_type == ClassBlock &&
-            !(flags & DEF_PARAM) &&
-            bound && PySet_Contains(bound, name)) {
-            SET_SCOPE(scopes, name, LOCAL_ONLY);
-        }
-        else {
-            SET_SCOPE(scopes, name, LOCAL);
-        }
+        SET_SCOPE(scopes, name, LOCAL);
         if (PySet_Add(local, name) < 0)
             return 0;
         if (PySet_Discard(global, name) < 0)
@@ -496,7 +489,7 @@
         long scope;
         assert(PyLong_Check(v));
         scope = PyLong_AS_LONG(v);
-        if (scope != LOCAL && scope != LOCAL_ONLY)
+        if (scope != LOCAL)
             continue;
         if (!PySet_Contains(free, name))
             continue;

From ncoghlan at gmail.com  Wed Jul 21 00:39:49 2010
From: ncoghlan at gmail.com (Nick Coghlan)
Date: Wed, 21 Jul 2010 08:39:49 +1000
Subject: [Python-checkins] r83005 - in python/branches/py3k:
	Include/opcode.h Include/symtable.h
	Lib/test/test_line_tracing.py Lib/test/test_scope.py
	Lib/test/test_trace.py Objects/typeobject.c Python/ceval.c
	Python/compile.c Python/symtable.c
In-Reply-To: <20100720223720.36E9FEE992@mail.python.org>
References: <20100720223720.36E9FEE992@mail.python.org>
Message-ID: <AANLkTikNpiPYU6SKb5uv4u3lXUZIf9izVjgVB3b2JNET@mail.gmail.com>

On Wed, Jul 21, 2010 at 8:37 AM, benjamin.peterson
<python-checkins at python.org> wrote:
> Author: benjamin.peterson
> Date: Wed Jul 21 00:37:19 2010
> New Revision: 83005
>
> Log:
> move test_trace.py so as not to conflict with future tests for the trace module
>
> Added:
> ? python/branches/py3k/Lib/test/test_line_tracing.py
> ? ? ?- copied unchanged from r83003, /python/branches/py3k/Lib/test/test_trace.py
> Removed:
> ? python/branches/py3k/Lib/test/test_trace.py
> Modified:
> ? python/branches/py3k/Include/opcode.h
> ? python/branches/py3k/Include/symtable.h
> ? python/branches/py3k/Lib/test/test_scope.py
> ? python/branches/py3k/Objects/typeobject.c
> ? python/branches/py3k/Python/ceval.c
> ? python/branches/py3k/Python/compile.c
> ? python/branches/py3k/Python/symtable.c

I have a feeling this checkin picked up some experimental code you
were playing with locally...

Cheers,
Nick.

-- 
Nick Coghlan?? |?? ncoghlan at gmail.com?? |?? Brisbane, Australia

From ncoghlan at gmail.com  Wed Jul 21 00:40:39 2010
From: ncoghlan at gmail.com (Nick Coghlan)
Date: Wed, 21 Jul 2010 08:40:39 +1000
Subject: [Python-checkins] r83005 - in python/branches/py3k:
	Include/opcode.h Include/symtable.h
	Lib/test/test_line_tracing.py Lib/test/test_scope.py
	Lib/test/test_trace.py Objects/typeobject.c Python/ceval.c
	Python/compile.c Python/symtable.c
In-Reply-To: <AANLkTikNpiPYU6SKb5uv4u3lXUZIf9izVjgVB3b2JNET@mail.gmail.com>
References: <20100720223720.36E9FEE992@mail.python.org>
	<AANLkTikNpiPYU6SKb5uv4u3lXUZIf9izVjgVB3b2JNET@mail.gmail.com>
Message-ID: <AANLkTin8I9jCptNpdkrD49E6_yEtjXH-0cP5oQSJZhEY@mail.gmail.com>

> I have a feeling this checkin picked up some experimental code you
> were playing with locally...

And in the time it took me to write that, you'd already fixed it :)

-- 
Nick Coghlan?? |?? ncoghlan at gmail.com?? |?? Brisbane, Australia

From solipsis at pitrou.net  Wed Jul 21 01:24:27 2010
From: solipsis at pitrou.net (solipsis at pitrou.net)
Date: Wed, 21 Jul 2010 01:24:27 +0200 (CEST)
Subject: [Python-checkins] Daily py3k reference leaks (r82997): sum=0
Message-ID: <20100720232427.4E86C1770A@ns6635.ovh.net>

py3k results for svn r82997 (hg cset 1bdbc5d595da)
--------------------------------------------------



Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/antoine/py3k/refleaks/reflogjusxod', '-x']

From python-checkins at python.org  Wed Jul 21 03:35:46 2010
From: python-checkins at python.org (brian.curtin)
Date: Wed, 21 Jul 2010 03:35:46 +0200 (CEST)
Subject: [Python-checkins] r83007 -
	python/branches/release27-maint/Modules/zipimport.c
Message-ID: <20100721013546.5320FEEA11@mail.python.org>

Author: brian.curtin
Date: Wed Jul 21 03:35:46 2010
New Revision: 83007

Log:
Fix #9316. if/is grammar corrections.


Modified:
   python/branches/release27-maint/Modules/zipimport.c

Modified: python/branches/release27-maint/Modules/zipimport.c
==============================================================================
--- python/branches/release27-maint/Modules/zipimport.c	(original)
+++ python/branches/release27-maint/Modules/zipimport.c	Wed Jul 21 03:35:46 2010
@@ -536,19 +536,19 @@
 "is_package(fullname) -> bool.\n\
 \n\
 Return True if the module specified by fullname is a package.\n\
-Raise ZipImportError is the module couldn't be found.");
+Raise ZipImportError if the module couldn't be found.");
 
 PyDoc_STRVAR(doc_get_code,
 "get_code(fullname) -> code object.\n\
 \n\
 Return the code object for the specified module. Raise ZipImportError\n\
-is the module couldn't be found.");
+if the module couldn't be found.");
 
 PyDoc_STRVAR(doc_get_source,
 "get_source(fullname) -> source string.\n\
 \n\
 Return the source code for the specified module. Raise ZipImportError\n\
-is the module couldn't be found, return None if the archive does\n\
+if the module couldn't be found, return None if the archive does\n\
 contain the module, but has no source for it.");
 
 

From python-checkins at python.org  Wed Jul 21 03:40:15 2010
From: python-checkins at python.org (brian.curtin)
Date: Wed, 21 Jul 2010 03:40:15 +0200 (CEST)
Subject: [Python-checkins] r83008 - in python/branches/release26-maint:
	Modules/zipimport.c
Message-ID: <20100721014015.A3304EE9BF@mail.python.org>

Author: brian.curtin
Date: Wed Jul 21 03:40:15 2010
New Revision: 83008

Log:
Merged revisions 83007 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

........
  r83007 | brian.curtin | 2010-07-20 20:35:46 -0500 (Tue, 20 Jul 2010) | 2 lines
  
  Fix #9316. if/is grammar corrections.
........


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Modules/zipimport.c

Modified: python/branches/release26-maint/Modules/zipimport.c
==============================================================================
--- python/branches/release26-maint/Modules/zipimport.c	(original)
+++ python/branches/release26-maint/Modules/zipimport.c	Wed Jul 21 03:40:15 2010
@@ -536,19 +536,19 @@
 "is_package(fullname) -> bool.\n\
 \n\
 Return True if the module specified by fullname is a package.\n\
-Raise ZipImportError is the module couldn't be found.");
+Raise ZipImportError if the module couldn't be found.");
 
 PyDoc_STRVAR(doc_get_code,
 "get_code(fullname) -> code object.\n\
 \n\
 Return the code object for the specified module. Raise ZipImportError\n\
-is the module couldn't be found.");
+if the module couldn't be found.");
 
 PyDoc_STRVAR(doc_get_source,
 "get_source(fullname) -> source string.\n\
 \n\
 Return the source code for the specified module. Raise ZipImportError\n\
-is the module couldn't be found, return None if the archive does\n\
+if the module couldn't be found, return None if the archive does\n\
 contain the module, but has no source for it.");
 
 

From python-checkins at python.org  Wed Jul 21 03:44:19 2010
From: python-checkins at python.org (brian.curtin)
Date: Wed, 21 Jul 2010 03:44:19 +0200 (CEST)
Subject: [Python-checkins] r83009 - python/branches/py3k/Modules/zipimport.c
Message-ID: <20100721014419.8389BEE9AE@mail.python.org>

Author: brian.curtin
Date: Wed Jul 21 03:44:19 2010
New Revision: 83009

Log:
Fix #9316. if/is grammar corrections.


Modified:
   python/branches/py3k/Modules/zipimport.c

Modified: python/branches/py3k/Modules/zipimport.c
==============================================================================
--- python/branches/py3k/Modules/zipimport.c	(original)
+++ python/branches/py3k/Modules/zipimport.c	Wed Jul 21 03:44:19 2010
@@ -523,19 +523,19 @@
 "is_package(fullname) -> bool.\n\
 \n\
 Return True if the module specified by fullname is a package.\n\
-Raise ZipImportError is the module couldn't be found.");
+Raise ZipImportError if the module couldn't be found.");
 
 PyDoc_STRVAR(doc_get_code,
 "get_code(fullname) -> code object.\n\
 \n\
 Return the code object for the specified module. Raise ZipImportError\n\
-is the module couldn't be found.");
+if the module couldn't be found.");
 
 PyDoc_STRVAR(doc_get_source,
 "get_source(fullname) -> source string.\n\
 \n\
 Return the source code for the specified module. Raise ZipImportError\n\
-is the module couldn't be found, return None if the archive does\n\
+if the module couldn't be found, return None if the archive does\n\
 contain the module, but has no source for it.");
 
 

From python-checkins at python.org  Wed Jul 21 03:46:44 2010
From: python-checkins at python.org (brian.curtin)
Date: Wed, 21 Jul 2010 03:46:44 +0200 (CEST)
Subject: [Python-checkins] r83010 - in python/branches/release31-maint:
	Modules/zipimport.c
Message-ID: <20100721014644.F33CDEE9B7@mail.python.org>

Author: brian.curtin
Date: Wed Jul 21 03:46:44 2010
New Revision: 83010

Log:
Merged revisions 83009 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83009 | brian.curtin | 2010-07-20 20:44:19 -0500 (Tue, 20 Jul 2010) | 2 lines
  
  Fix #9316. if/is grammar corrections.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Modules/zipimport.c

Modified: python/branches/release31-maint/Modules/zipimport.c
==============================================================================
--- python/branches/release31-maint/Modules/zipimport.c	(original)
+++ python/branches/release31-maint/Modules/zipimport.c	Wed Jul 21 03:46:44 2010
@@ -523,19 +523,19 @@
 "is_package(fullname) -> bool.\n\
 \n\
 Return True if the module specified by fullname is a package.\n\
-Raise ZipImportError is the module couldn't be found.");
+Raise ZipImportError if the module couldn't be found.");
 
 PyDoc_STRVAR(doc_get_code,
 "get_code(fullname) -> code object.\n\
 \n\
 Return the code object for the specified module. Raise ZipImportError\n\
-is the module couldn't be found.");
+if the module couldn't be found.");
 
 PyDoc_STRVAR(doc_get_source,
 "get_source(fullname) -> source string.\n\
 \n\
 Return the source code for the specified module. Raise ZipImportError\n\
-is the module couldn't be found, return None if the archive does\n\
+if the module couldn't be found, return None if the archive does\n\
 contain the module, but has no source for it.");
 
 

From python-checkins at python.org  Wed Jul 21 11:48:31 2010
From: python-checkins at python.org (brett.cannon)
Date: Wed, 21 Jul 2010 11:48:31 +0200 (CEST)
Subject: [Python-checkins] r83011 -
	python/branches/py3k/Doc/library/importlib.rst
Message-ID: <20100721094831.D4001C386@mail.python.org>

Author: brett.cannon
Date: Wed Jul 21 11:48:31 2010
New Revision: 83011

Log:
Minor clarification about importlib.abc.SourceLoader.get_filename.


Modified:
   python/branches/py3k/Doc/library/importlib.rst

Modified: python/branches/py3k/Doc/library/importlib.rst
==============================================================================
--- python/branches/py3k/Doc/library/importlib.rst	(original)
+++ python/branches/py3k/Doc/library/importlib.rst	Wed Jul 21 11:48:31 2010
@@ -228,7 +228,7 @@
 
     * :meth:`ResourceLoader.get_data`
     * :meth:`ExecutionLoader.get_filename`
-          Implement to only return the path to the source file; sourceless
+          Should only return the path to the source file; sourceless
           loading is not supported.
 
     The abstract methods defined by this class are to add optional bytecode

From python-checkins at python.org  Wed Jul 21 11:50:42 2010
From: python-checkins at python.org (brett.cannon)
Date: Wed, 21 Jul 2010 11:50:42 +0200 (CEST)
Subject: [Python-checkins] r83012 - in python/branches/release31-maint:
	Doc/library/dis.rst Misc/NEWS
Message-ID: <20100721095042.91087D879@mail.python.org>

Author: brett.cannon
Date: Wed Jul 21 11:50:42 2010
New Revision: 83012

Log:
Backport r82456


Modified:
   python/branches/release31-maint/Doc/library/dis.rst
   python/branches/release31-maint/Misc/NEWS

Modified: python/branches/release31-maint/Doc/library/dis.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/dis.rst	(original)
+++ python/branches/release31-maint/Doc/library/dis.rst	Wed Jul 21 11:50:42 2010
@@ -5,12 +5,18 @@
    :synopsis: Disassembler for Python bytecode.
 
 
-The :mod:`dis` module supports the analysis of Python :term:`bytecode` by
-disassembling it.  Since there is no Python assembler, this module defines the
-Python assembly language.  The Python bytecode which this module takes as an
+The :mod:`dis` module supports the analysis of CPython :term:`bytecode` by
+disassembling it. The CPython bytecode which this module takes as an
 input is defined in the file :file:`Include/opcode.h` and used by the compiler
 and the interpreter.
 
+.. impl-detail::
+
+   Bytecode is an implementation detail of the CPython interpreter!  No
+   guarantees are made that bytecode will not be added, removed, or changed
+   between versions of Python.  Use of this module should not be considered to
+   work across Python VMs or Python releases.
+
 Example: Given the function :func:`myfunc`::
 
    def myfunc(alist):

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Wed Jul 21 11:50:42 2010
@@ -363,6 +363,12 @@
 - Issue #8193: Fix test_zlib failure with zlib 1.2.4.
 
 
+Documentation
+-------------
+
+- Issue #7829: Document in dis that bytecode is an implementation detail.
+
+
 What's New in Python 3.1.2?
 ===========================
 

From python-checkins at python.org  Wed Jul 21 11:52:10 2010
From: python-checkins at python.org (brett.cannon)
Date: Wed, 21 Jul 2010 11:52:10 +0200 (CEST)
Subject: [Python-checkins] r83013 - in python/branches/release27-maint:
	Doc/library/dis.rst Misc/NEWS
Message-ID: <20100721095210.D8795C399@mail.python.org>

Author: brett.cannon
Date: Wed Jul 21 11:52:10 2010
New Revision: 83013

Log:
Backport r82456.


Modified:
   python/branches/release27-maint/Doc/library/dis.rst
   python/branches/release27-maint/Misc/NEWS

Modified: python/branches/release27-maint/Doc/library/dis.rst
==============================================================================
--- python/branches/release27-maint/Doc/library/dis.rst	(original)
+++ python/branches/release27-maint/Doc/library/dis.rst	Wed Jul 21 11:52:10 2010
@@ -6,11 +6,17 @@
    :synopsis: Disassembler for Python bytecode.
 
 
-The :mod:`dis` module supports the analysis of Python :term:`bytecode` by disassembling
-it.  Since there is no Python assembler, this module defines the Python assembly
-language.  The Python bytecode which this module takes as an input is defined
-in the file  :file:`Include/opcode.h` and used by the compiler and the
-interpreter.
+The :mod:`dis` module supports the analysis of CPython :term:`bytecode` by
+disassembling it. The CPython bytecode which this module takes as an
+input is defined in the file :file:`Include/opcode.h` and used by the compiler
+and the interpreter.
+
+.. impl-detail::
+
+   Bytecode is an implementation detail of the CPython interpreter!  No
+   guarantees are made that bytecode will not be added, removed, or changed
+   between versions of Python.  Use of this module should not be considered to
+   work across Python VMs or Python releases.
 
 Example: Given the function :func:`myfunc`::
 

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Wed Jul 21 11:52:10 2010
@@ -59,6 +59,11 @@
 
 - Issue #8605: Skip test_gdb if Python is compiled with optimizations.
 
+Documentation
+-------------
+
+- Issue 7829: Document in dis that bytecode is an implementation detail.
+
 
 What's New in Python 2.7?
 =========================

From python-checkins at python.org  Wed Jul 21 14:26:35 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 21 Jul 2010 14:26:35 +0200 (CEST)
Subject: [Python-checkins] r83014 - sandbox/trunk/errnopep
Message-ID: <20100721122635.B0EC3E8DD@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 21 14:26:35 2010
New Revision: 83014

Log:
A hypothetical PEP for the OS exception hierarchy



Added:
   sandbox/trunk/errnopep/

From python-checkins at python.org  Wed Jul 21 14:27:36 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 21 Jul 2010 14:27:36 +0200 (CEST)
Subject: [Python-checkins] r83015 - sandbox/trunk/errnopep/pepXXXX.txt
Message-ID: <20100721122736.52052E8DD@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 21 14:27:36 2010
New Revision: 83015

Log:
Draft PEP



Added:
   sandbox/trunk/errnopep/pepXXXX.txt   (contents, props changed)

Added: sandbox/trunk/errnopep/pepXXXX.txt
==============================================================================
--- (empty file)
+++ sandbox/trunk/errnopep/pepXXXX.txt	Wed Jul 21 14:27:36 2010
@@ -0,0 +1,533 @@
+PEP: XXX
+Title: Reworking the OS and IO exception hierarchy
+Version: $Revision: $
+Last-Modified: $Date: $
+Author: Antoine Pitrou <solipsis at pitrou.net>
+Status: 
+Type: Standards Track
+Content-Type: text/x-rst
+Created: 
+Post-History:
+
+
+Abstract
+========
+
+Rationale
+=========
+
+Confusing set of OS-related exceptions
+--------------------------------------
+
+OS-related (or system call-related) exceptions are currently a diversity
+of classes, arranged in the following subhierarchies::
+
+    +-- EnvironmentError
+        +-- IOError
+            +-- io.BlockingIOError
+            +-- io.UnsupportedOperation (also inherits from ValueError)
+            +-- socket.error
+        +-- OSError
+            +-- WindowsError
+
+    +-- select.error
+
+While some of these distinctions can be explained by implementation
+considerations, they are often not very logical at a higher level.  The
+line separating OSError and IOError, for example, is often blurry.  Consider
+the following::
+
+    >>> os.remove("fff")
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    OSError: [Errno 2] No such file or directory: 'fff'
+    >>> open("fff")
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: [Errno 2] No such file or directory: 'fff'
+
+The same error condition (a non-existing file) gets cast as two different
+exceptions depending on which library function was called.  The reason
+for this is that the `os` module exclusively raises OSError (or its
+subclass WindowsError) while the `io` module mostly raises IOError.
+However, the user is interested in the nature of the error, not in which
+part of the interpreter it comes from (since the latter is obvious from
+reading the traceback message or application source code).
+
+A further proof of the ambiguity of this segmentation is that the standard
+library itself sometimes has problems deciding.  For example, in the
+``select`` module, similar failures will raise either ``select.error``,
+``OSError`` or ``IOError`` depending on whether you are using select(),
+a poll object, a kqueue object, or an epoll object.
+
+As for WindowsError, it seems to be a pointless distinction.  First, it
+only exists on Windows systems, which requires tedious compatibility code
+in cross-platform applications.  Second, since it inherits from OSError
+and is raised for similar errors as OSError is raised for on other systems.
+Third, the user wanting low-level access to exception specifics has to
+use the ``errno`` or ``winerror`` attribute anyway.
+
+
+Lack of fine-grained exceptions
+-------------------------------
+
+The current variety of OS-related exceptions doesn't allow the user to filter
+easily for the desired kinds of failures.  As an example, consider the task
+of deleting a file if it exists.  The Look Before You Leap (LBYL) idiom
+suffers from an obvious race condition::
+
+    if os.path.exists(filename):
+        os.remove(filename)
+
+If a file named as `filename` is created by another thread or process
+between the calls to `os.path.exists` and `os.remove`, it won't be deleted.
+This can produce bugs in the application, or even security issues.
+
+Therefore, the solution is to try to remove the file, and ignore the error
+if the file doesn't exist (an idiom known as Easier to Ask Forgiveness
+than to get Permission, or EAFP).  Careful code will read like the following
+(which works under both POSIX and Windows systems)::
+
+    try:
+        os.remove(filename)
+    except OSError as e:
+        if e.errno != errno.ENOENT:
+            raise
+
+or even::
+
+    try:
+        os.remove(filename)
+    except EnvironmentError as e:
+        if e.errno != errno.ENOENT:
+            raise
+
+This is a lot more to type, and also forces the user to remember the various
+cryptic mnemonics from the errno module.  It imposes an additional cognitive
+burden and gets tiresome rather quickly.  Consequently, many programmers
+will instead write the following code, which silences exceptions too
+broadly::
+
+    try:
+        os.remove(filename)
+    except OSError:
+        pass
+
+What the programmer would like to write instead is something such as::
+
+    try:
+        os.remove(filename)
+    except FileNotFound:
+        pass
+
+
+Step 1: coalesce exception types
+================================
+
+The first step of the resolution is to coalesce existing exception types.
+The extent of this step is not yet fully determined.  A number of possible
+changes are listed hereafter::
+
+* alias both socket.error and select.error to IOError
+* alias OSError to IOError
+* alias WindowsError to OSError
+
+Each of these changes doesn't preserve exact compatibility, but it does
+preserve *useful compatibility* (see later for a definition of useful
+compatibility).
+
+Not only does this first step present the user a simpler landscape, but
+it also allows for a better and more complete resolution of step 2
+(see "Prerequisite" below).
+
+
+Step 2: define additional subclasses
+====================================
+
+The second step of the resolution is to extend the hierarchy by defining
+subclasses which will be raised, rather than their parent, for specific
+errno values.  Which errno values is not decided yet, but a survey of
+existing exception matching practices (see Appendix A) will help us
+choose a reasonable subset of all values.  Trying to map all errno
+mnemonics, indeed, seems foolish, pointless, and would pollute the root
+namespace.
+
+Furthermore, in a couple of cases, different errno values could raise
+the same exception subclass.  For example, EAGAIN, EALREADY, EWOULDBLOCK
+and EINPROGRESS are all used to signal that an operation on a non-blocking
+socket would block (and therefore needs trying again later).  They could
+therefore all raise an identical subclass (perhaps even the existing
+``io.BlockingIOError``) and let the user examine the ``errno`` attribute
+if (s)he so desires (see below "exception attributes").
+
+
+Prerequisite
+------------
+
+Step 1 is a loose prerequisite for this.
+
+Prerequisite, because some errnos can currently be attached to different
+exception classes: for example, EBADF can be attached to both OSError and
+IOError, depending on the context.  If we don't want to break *useful
+compatibility*, we can't make an ``except OSError`` (or IOError) fail
+to match an exception where it would succeed today.
+
+Loose, because we could decide for a partial resolution of step 2
+if existing exception classes are not coalesced: for example, EBADF could
+raise a hypothetical BadFileDescriptor where an IOError was previously
+raised, but continue to raise OSError otherwise.
+
+The dependency on step 1 could be totally removed if the new subclasses
+used multiple inheritance to match with all of the existing superclasses
+(or, at least, OSError and IOError, which are arguable the most prevalent
+ones).  It would, however, make the hierarchy more complicated and
+therefore harder to grasp for the user.
+
+
+Exception attributes
+--------------------
+
+In order to preserve *useful compatibility*, these subclasses should still
+set adequate values for the various exception attributes defined on the
+superclass (for example ``errno``, ``filename``, and optionally
+``winerror``).
+
+
+
+Compatibility concerns
+======================
+
+Reworking the exception hierarchy will obviously change the exact semantics
+of at least some existing code.  While it is not possible to improve on the
+current situation without changing exact semantics, it is possible to define
+a narrower type of compatibility, which we will call **useful compatibility**,
+and define as follows:
+
+* *useful compatibility* doesn't make exception catching any narrower, but
+  it can be broader for *na?ve* exception-catching code.  Given the following
+  kind of snippet, all exceptions caught before this PEP will also be
+  caught after this PEP, but the reverse may be false::
+  
+      try:
+          os.remove(filename)
+      except OSError:
+          pass
+
+* *useful compatibility* doesn't alter the behaviour of *careful*
+  exception-catching code.  Given the following kind of snippet, the same
+  errors should be silenced or reraised, regardless of whether this PEP
+  has been implemented or not::
+
+      try:
+          os.remove(filename)
+      except OSError as e:
+          if e.errno != errno.ENOENT:
+              raise
+
+The rationale for this compromise is that dangerous (or "na?ve") code
+can't really be helped, but at least code which "works" won't suddenly
+raise errors and crash.  This is important since such code is likely to
+be present in scripts used as cron tasks or automated system administration
+programs.
+
+Careful code should not be penalized.
+
+
+Possible alternative
+====================
+
+Pattern matching
+----------------
+
+Another possibility would be to introduce an advanced pattern matching
+syntax when catching exceptions.  For example::
+
+    try:
+        os.remove(filename)
+    except OSError as e if e.errno == errno.ENOENT:
+        pass
+
+Several problems with this proposal:
+
+* it introduces new syntax, which is perceived by the author to be a heavier
+  change compared to reworking the exception hierarchy
+* it doesn't diminish typing effort significantly
+* it doesn't relieve the programmer from the burden of having to remember
+  errno mnemonics
+
+
+Exceptions ignored by this PEP
+==============================
+
+This PEP ignores ``EOFError``, which signals a truncated input stream in
+various protocol and file format implementations (for example ``GzipFile``).
+``EOFError`` is not OS- or IO-related, it is a logical error raised at
+a higher level.
+
+This PEP also ignores ``SSLError``, which is raised by the ``ssl`` module
+in order to propagate errors signalled by the ``OpenSSL`` library.  Ideally,
+``SSLError`` would benefit from a similar but separate treatment since it
+defines its own constants for error types (``ssl.SSL_ERROR_WANT_READ``,
+etc.).
+
+
+Appendix A: Survey of common errnos
+===================================
+
+This is a quick recension of the various errno mnemonics checked for in
+the standard library and its tests, as part of ``except`` clauses.
+
+Common errnos with OSError
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+* ``EBADF``: bad file descriptor (usually means the file descriptor was
+             closed)
+
+* ``EEXIST``: file or directory exists
+
+* ``EINTR``: interrupted function call
+
+* ``ENOTDIR``: not a directory
+
+* ``EOPNOTSUPP``: operation not supported on socket
+  (possible confusion with the existing io.UnsupportedOperation)
+
+* ``EPERM``: operation not permitted (when using e.g. os.setuid())
+
+Common errnos with IOError
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+* ``EACCES``: permission denied (for filesystem operations)
+
+* ``EBADF``: bad file descriptor (with select.epoll); read operation on a
+  write-only GzipFile, or vice-versa
+
+* ``EBUSY``: device or resource busy
+
+* ``EISDIR``: is a directory (when trying to open())
+
+* ``ENODEV``: no such device
+
+* ``ENOENT``: no such file or directory
+
+* ``ETIMEDOUT``: connection timed out
+
+Common errnos with socket.error
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+All these errors may also be associated with a plain IOError, for example
+when calling read() on a socket file descriptor.
+
+* ``EAGAIN``: resource temporarily unavailable (during a non-blocking socket
+              call except connect())
+
+* ``EALREADY``: connection already in progress (during a non-blocking connect())
+
+* ``EINPROGRESS``: operation in progress (during a non-blocking connect())
+
+* ``EINTR``: interrupted function call
+
+* ``EISCONN``: the socket is connected
+
+* ``ECONNABORTED``: connection aborted by peer (during an accept() call)
+
+* ``ECONNREFUSED``: connection refused by peer
+
+* ``ECONNRESET``: connection reset by peer
+
+* ``ENOTCONN``: socket not connected
+
+* ``ESHUTDOWN``: cannot send after transport endpoint shutdown
+
+* ``EWOULDBLOCK``: same reasons as ``EAGAIN``
+
+Common errnos with select.error
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+* ``EINTR``: interrupted function call
+
+
+
+Appendix B: Survey of raised OS and IO errors
+=============================================
+
+Interpreter core
+~~~~~~~~~~~~~~~~
+
+Handling of PYTHONSTARTUP raises IOError (but the error gets discarded)::
+
+    $ PYTHONSTARTUP=foox ./python
+    Python 3.2a0 (py3k:82920M, Jul 16 2010, 22:53:23) 
+    [GCC 4.4.3] on linux2
+    Type "help", "copyright", "credits" or "license" for more information.
+    Could not open PYTHONSTARTUP
+    IOError: [Errno 2] No such file or directory: 'foox'
+
+``PyObject_Print()`` raises IOError when ferror() signals an error on the
+`FILE *` parameter (which, in the source tree, is always either stdout or
+stderr).
+
+
+Modules
+~~~~~~~
+
+bz2
+---
+
+Raises IOError throughout (OSError is unused)::
+
+    >>> bz2.BZ2File("foox", "rb")
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: [Errno 2] No such file or directory
+    >>> bz2.BZ2File("LICENSE", "rb").read()
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: invalid data stream
+    >>> bz2.BZ2File("/tmp/zzz.bz2", "wb").read()
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: file is not ready for reading
+
+curses
+------
+
+Not examined.
+
+dbm.gnu, dbm.ndbm
+-----------------
+
+_dbm.error and _gdbm.error inherit from IOError::
+
+    >>> dbm.gnu.open("foox")
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    _gdbm.error: [Errno 2] No such file or directory
+
+fcntl
+-----
+
+Raises IOError throughout (OSError is unused).
+
+imp module
+----------
+
+Raises IOError for bad file descriptors::
+
+    >>> imp.load_source("foo", "foo", 123)
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: [Errno 9] Bad file descriptor
+
+io module
+---------
+
+Raises IOError when trying to open a directory under Unix::
+
+    >>> open("Python/", "r")
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: [Errno 21] Is a directory: 'Python/'
+
+Raises IOError for unsupported operations::
+
+    >>> open("LICENSE").write("bar")
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: not writable
+    >>> io.StringIO().fileno()
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    io.UnsupportedOperation: fileno
+    >>> open("LICENSE").seek(1, 1)
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: can't do nonzero cur-relative seeks
+
+(io.UnsupportedOperation inherits from IOError)
+
+Raises either IOError or TypeError when the inferior I/O layer misbehaves
+(i.e. violates the API it is expected to implement).
+
+Raises IOError when the underlying OS resource becomes invalid::
+
+    >>> f = open("LICENSE")
+    >>> os.close(f.fileno())
+    >>> f.read()
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: [Errno 9] Bad file descriptor
+
+...or for implementation-specific optimizations::
+
+    >>> f = open("LICENSE")
+    >>> next(f)
+    'A. HISTORY OF THE SOFTWARE\n'
+    >>> f.tell()
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: telling position disabled by next() call
+
+Raises BlockingIOError (inherited from IOError) when a call on a non-blocking
+object would block.
+
+multiprocessing
+---------------
+
+Not examined.
+
+ossaudiodev
+-----------
+
+Raises IOError throughout (OSError is unused)::
+
+    >>> ossaudiodev.open("foo", "r")
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: [Errno 2] No such file or directory: 'foo'
+
+readline
+--------
+
+Raises IOError in various file-handling functions::
+
+    >>> readline.read_history_file("foo")
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: [Errno 2] No such file or directory
+    >>> readline.read_init_file("foo")
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: [Errno 2] No such file or directory
+    >>> readline.write_history_file("/dev/nonexistent")
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: [Errno 13] Permission denied
+
+select
+------
+
+select() and poll objects raise select.error, which doesn't inherit from
+anything (but poll.modify() which raises IOError).
+epoll objects raise IOError.
+kqueue objects raise both OSError and IOError.
+
+signal
+------
+
+signal.ItimerError inherits from IOError.
+
+socket
+------
+
+socket.error inherits from IOError.
+
+time
+----
+
+Raises IOError for internal errors in time.time() and time.sleep().
+
+zipimport
+---------
+
+zipimporter.get_data() can raise IOError.

From python-checkins at python.org  Wed Jul 21 14:29:04 2010
From: python-checkins at python.org (doug.hellmann)
Date: Wed, 21 Jul 2010 14:29:04 +0200 (CEST)
Subject: [Python-checkins] r83016 - in python/branches/py3k:
	Lib/json/tests/test_fail.py Modules/_json.c
Message-ID: <20100721122904.74BF5E9E6@mail.python.org>

Author: doug.hellmann
Date: Wed Jul 21 14:29:04 2010
New Revision: 83016

Log:
Apply patch from Ray Allen for issue 9296

Modified:
   python/branches/py3k/Lib/json/tests/test_fail.py
   python/branches/py3k/Modules/_json.c

Modified: python/branches/py3k/Lib/json/tests/test_fail.py
==============================================================================
--- python/branches/py3k/Lib/json/tests/test_fail.py	(original)
+++ python/branches/py3k/Lib/json/tests/test_fail.py	Wed Jul 21 14:29:04 2010
@@ -74,3 +74,12 @@
                 pass
             else:
                 self.fail("Expected failure for fail{0}.json: {1!r}".format(idx, doc))
+
+    def test_non_string_keys_dict(self):
+        data = {'a' : 1, (1, 2) : 2}
+
+        #This is for c encoder
+        self.assertRaises(TypeError, json.dumps, data)
+
+        #This is for python encoder
+        self.assertRaises(TypeError, json.dumps, data, indent=True)

Modified: python/branches/py3k/Modules/_json.c
==============================================================================
--- python/branches/py3k/Modules/_json.c	(original)
+++ python/branches/py3k/Modules/_json.c	Wed Jul 21 14:29:04 2010
@@ -1454,7 +1454,7 @@
         }
         else {
             /* TODO: include repr of key */
-            PyErr_SetString(PyExc_ValueError, "keys must be a string");
+            PyErr_SetString(PyExc_TypeError, "keys must be a string");
             goto bail;
         }
 

From python-checkins at python.org  Wed Jul 21 14:35:38 2010
From: python-checkins at python.org (doug.hellmann)
Date: Wed, 21 Jul 2010 14:35:38 +0200 (CEST)
Subject: [Python-checkins] r83017 - in python/branches/release31-maint:
	Lib/json/tests/test_fail.py Modules/_json.c
Message-ID: <20100721123538.30DABE8D4@mail.python.org>

Author: doug.hellmann
Date: Wed Jul 21 14:35:38 2010
New Revision: 83017

Log:
Merged revisions 83016 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83016 | doug.hellmann | 2010-07-21 08:29:04 -0400 (Wed, 21 Jul 2010) | 1 line
  
  Apply patch from Ray Allen for issue 9296
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/json/tests/test_fail.py
   python/branches/release31-maint/Modules/_json.c

Modified: python/branches/release31-maint/Lib/json/tests/test_fail.py
==============================================================================
--- python/branches/release31-maint/Lib/json/tests/test_fail.py	(original)
+++ python/branches/release31-maint/Lib/json/tests/test_fail.py	Wed Jul 21 14:35:38 2010
@@ -74,3 +74,12 @@
                 pass
             else:
                 self.fail("Expected failure for fail{0}.json: {1!r}".format(idx, doc))
+
+    def test_non_string_keys_dict(self):
+        data = {'a' : 1, (1, 2) : 2}
+
+        #This is for c encoder
+        self.assertRaises(TypeError, json.dumps, data)
+
+        #This is for python encoder
+        self.assertRaises(TypeError, json.dumps, data, indent=True)

Modified: python/branches/release31-maint/Modules/_json.c
==============================================================================
--- python/branches/release31-maint/Modules/_json.c	(original)
+++ python/branches/release31-maint/Modules/_json.c	Wed Jul 21 14:35:38 2010
@@ -1454,7 +1454,7 @@
         }
         else {
             /* TODO: include repr of key */
-            PyErr_SetString(PyExc_ValueError, "keys must be a string");
+            PyErr_SetString(PyExc_TypeError, "keys must be a string");
             goto bail;
         }
 

From python-checkins at python.org  Wed Jul 21 14:36:33 2010
From: python-checkins at python.org (doug.hellmann)
Date: Wed, 21 Jul 2010 14:36:33 +0200 (CEST)
Subject: [Python-checkins] r83018 - in python/branches/release27-maint:
	Lib/json/tests/test_fail.py Modules/_json.c
Message-ID: <20100721123633.37EF8E8D4@mail.python.org>

Author: doug.hellmann
Date: Wed Jul 21 14:36:33 2010
New Revision: 83018

Log:
Apply patch from Ray Allen for issue 9296

Modified:
   python/branches/release27-maint/Lib/json/tests/test_fail.py
   python/branches/release27-maint/Modules/_json.c

Modified: python/branches/release27-maint/Lib/json/tests/test_fail.py
==============================================================================
--- python/branches/release27-maint/Lib/json/tests/test_fail.py	(original)
+++ python/branches/release27-maint/Lib/json/tests/test_fail.py	Wed Jul 21 14:36:33 2010
@@ -74,3 +74,12 @@
                 pass
             else:
                 self.fail("Expected failure for fail{0}.json: {1!r}".format(idx, doc))
+
+    def test_non_string_keys_dict(self):
+        data = {'a' : 1, (1, 2) : 2}
+
+        #This is for c encoder
+        self.assertRaises(TypeError, json.dumps, data)
+
+        #This is for python encoder
+        self.assertRaises(TypeError, json.dumps, data, indent=True)

Modified: python/branches/release27-maint/Modules/_json.c
==============================================================================
--- python/branches/release27-maint/Modules/_json.c	(original)
+++ python/branches/release27-maint/Modules/_json.c	Wed Jul 21 14:36:33 2010
@@ -2147,7 +2147,7 @@
         }
         else {
             /* TODO: include repr of key */
-            PyErr_SetString(PyExc_ValueError, "keys must be a string");
+            PyErr_SetString(PyExc_TypeError, "keys must be a string");
             goto bail;
         }
 

From python-checkins at python.org  Wed Jul 21 14:45:12 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 21 Jul 2010 14:45:12 +0200 (CEST)
Subject: [Python-checkins] r83019 - sandbox/trunk/errnopep/pepXXXX.txt
Message-ID: <20100721124512.05B68DAC0@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 21 14:45:11 2010
New Revision: 83019

Log:
A couple of changes



Modified:
   sandbox/trunk/errnopep/pepXXXX.txt

Modified: sandbox/trunk/errnopep/pepXXXX.txt
==============================================================================
--- sandbox/trunk/errnopep/pepXXXX.txt	(original)
+++ sandbox/trunk/errnopep/pepXXXX.txt	Wed Jul 21 14:45:11 2010
@@ -62,10 +62,10 @@
 
 As for WindowsError, it seems to be a pointless distinction.  First, it
 only exists on Windows systems, which requires tedious compatibility code
-in cross-platform applications.  Second, since it inherits from OSError
-and is raised for similar errors as OSError is raised for on other systems.
-Third, the user wanting low-level access to exception specifics has to
-use the ``errno`` or ``winerror`` attribute anyway.
+in cross-platform applications.  Second, it inherits from OSError and
+is raised for similar errors as OSError is raised for on other systems.
+Third, the user wanting access to low-level exception specifics has to
+examine the ``errno`` or ``winerror`` attribute anyway.
 
 
 Lack of fine-grained exceptions
@@ -251,7 +251,7 @@
 
 * it introduces new syntax, which is perceived by the author to be a heavier
   change compared to reworking the exception hierarchy
-* it doesn't diminish typing effort significantly
+* it doesn't decrease typing effort significantly
 * it doesn't relieve the programmer from the burden of having to remember
   errno mnemonics
 
@@ -287,8 +287,12 @@
 
 * ``EINTR``: interrupted function call
 
+* ``EISDIR``: is a directory
+
 * ``ENOTDIR``: not a directory
 
+* ``ENOENT``: no such file or directory
+
 * ``EOPNOTSUPP``: operation not supported on socket
   (possible confusion with the existing io.UnsupportedOperation)
 
@@ -308,7 +312,7 @@
 
 * ``ENODEV``: no such device
 
-* ``ENOENT``: no such file or directory
+* ``ENOENT``: no such file or directory (when trying to open())
 
 * ``ETIMEDOUT``: connection timed out
 
@@ -316,7 +320,7 @@
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
 All these errors may also be associated with a plain IOError, for example
-when calling read() on a socket file descriptor.
+when calling read() on a socket's file descriptor.
 
 * ``EAGAIN``: resource temporarily unavailable (during a non-blocking socket
               call except connect())

From python-checkins at python.org  Wed Jul 21 15:40:35 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 21 Jul 2010 15:40:35 +0200 (CEST)
Subject: [Python-checkins] r83020 - sandbox/trunk/errnopep/pepXXXX.txt
Message-ID: <20100721134035.CC077EFA5@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 21 15:40:35 2010
New Revision: 83020

Log:
Add some considerations



Modified:
   sandbox/trunk/errnopep/pepXXXX.txt

Modified: sandbox/trunk/errnopep/pepXXXX.txt
==============================================================================
--- sandbox/trunk/errnopep/pepXXXX.txt	(original)
+++ sandbox/trunk/errnopep/pepXXXX.txt	Wed Jul 21 15:40:35 2010
@@ -13,6 +13,24 @@
 Abstract
 ========
 
+The standard exception hierarchy is an important part of the Python
+language.  It has two defining qualities: it is both generic and
+selective.  Generic in that the same exception type can be raised
+- and handled - regardless of the context (for example, whether you are
+trying to add something to an integer, to call a string method, or to write
+an object on a socket, a TypeError will be raised for bad argument types).
+Selective in that it allows the user to easily handle (silence, examine,
+process, store or encapsulate...) specific kinds of error conditions
+while letting other errors bubble up to higher calling contexts.  For
+example, you can choose to catch ZeroDivisionErrors without affecting
+the default handling of other ArithmeticErrors (such as OverflowErrors).
+
+This PEP proposes changes to a part of the exception hierarchy in order
+to better embody the qualities mentioned above: the errors related to
+operating system calls (OSError, IOError, select.error, and all their
+subclasses).
+
+
 Rationale
 =========
 
@@ -29,7 +47,6 @@
             +-- socket.error
         +-- OSError
             +-- WindowsError
-
     +-- select.error
 
 While some of these distinctions can be explained by implementation
@@ -54,11 +71,17 @@
 part of the interpreter it comes from (since the latter is obvious from
 reading the traceback message or application source code).
 
+In fact, it is hard to think of any situation where OSError should be
+caught but not IOError, or the reverse.
+
 A further proof of the ambiguity of this segmentation is that the standard
 library itself sometimes has problems deciding.  For example, in the
 ``select`` module, similar failures will raise either ``select.error``,
 ``OSError`` or ``IOError`` depending on whether you are using select(),
-a poll object, a kqueue object, or an epoll object.
+a poll object, a kqueue object, or an epoll object.  This makes user code
+uselessly complicated since it has to be prepared to catch various
+exception types, depending on which exact implementation of a single
+primitive it chooses to use at runtime.
 
 As for WindowsError, it seems to be a pointless distinction.  First, it
 only exists on Windows systems, which requires tedious compatibility code
@@ -121,6 +144,13 @@
         pass
 
 
+Moratorium
+==========
+
+The moratorium in effect on language builtins means this PEP has little
+chance to be accepted for Python 3.2.
+
+
 Step 1: coalesce exception types
 ================================
 
@@ -129,7 +159,7 @@
 changes are listed hereafter::
 
 * alias both socket.error and select.error to IOError
-* alias OSError to IOError
+* alias IOError to OSError
 * alias WindowsError to OSError
 
 Each of these changes doesn't preserve exact compatibility, but it does
@@ -140,15 +170,21 @@
 it also allows for a better and more complete resolution of step 2
 (see "Prerequisite" below).
 
+Deprecation of names
+--------------------
+
+It is not yet decided whether the old names will be deprecated (then removed)
+or all alternative names will continue living in the root namespace.
+
 
 Step 2: define additional subclasses
 ====================================
 
 The second step of the resolution is to extend the hierarchy by defining
 subclasses which will be raised, rather than their parent, for specific
-errno values.  Which errno values is not decided yet, but a survey of
-existing exception matching practices (see Appendix A) will help us
-choose a reasonable subset of all values.  Trying to map all errno
+errno values.  Which errno values is subject to discussion, but a survey
+of existing exception matching practices (see Appendix A) helps us
+propose a reasonable subset of all values.  Trying to map all errno
 mnemonics, indeed, seems foolish, pointless, and would pollute the root
 namespace.
 
@@ -156,10 +192,9 @@
 the same exception subclass.  For example, EAGAIN, EALREADY, EWOULDBLOCK
 and EINPROGRESS are all used to signal that an operation on a non-blocking
 socket would block (and therefore needs trying again later).  They could
-therefore all raise an identical subclass (perhaps even the existing
-``io.BlockingIOError``) and let the user examine the ``errno`` attribute
-if (s)he so desires (see below "exception attributes").
-
+therefore all raise an identical subclass and let the user examine the
+``errno`` attribute if (s)he so desires (see below "exception
+attributes").
 
 Prerequisite
 ------------
@@ -183,6 +218,51 @@
 ones).  It would, however, make the hierarchy more complicated and
 therefore harder to grasp for the user.
 
+New exception classes
+---------------------
+
+The following temptative list of subclasses, along with a description and
+the list of errnos mapped to them, is submitted to discussion:
+
+* FileAlreadyExists: trying to create a file or directory which already
+  exists (EEXIST)
+
+* FileNotFound: for all circumstances where a file and directory is
+  requested but doesn't exist (ENOENT)
+
+* IsADirectory: file-level operation (open(), os.remove()...) requested
+  on a directory (EISDIR)
+
+* NotADirectory: directory-level operation requested on something else
+  (ENOTDIR)
+
+* PermissionDenied: trying to run an operation without the adequate access
+  rights - for example filesystem permissions (EACCESS, optionally EPERM)
+
+* BlockingIOError: an operation would block on an object (e.g. socket) set
+  for non-blocking operation (EAGAIN, EALREADY, EWOULDBLOCK, EINPROGRESS);
+  this is the existing ``io.BlockingIOError`` with an extended role
+
+* BadFileDescriptor: operation on an invalid file descriptor (EBADF);
+  the default error message could point out that most causes are that
+  an existing file descriptor has been closed
+
+* ConnectionAborted: connection attempt aborted by peer (ECONNABORTED)
+
+* ConnectionRefused: connection reset by peer (ECONNREFUSED)
+
+* ConnectionReset: connection reset by peer (ECONNRESET)
+
+* TimeoutError: connection timed out (ECONNTIMEOUT); this could be re-cast
+  as a generic timeout exception, useful for other types of timeout (for
+  example in Lock.acquire())
+
+This list assumes step 1 is accepted in full; the exception classes
+described above would all derive from the now unified exception type
+OSError.  It will need reworking if a partial version of step 1 is accepted
+instead (again, see appendix A for the current distribution of errnos
+and exception types).
+
 
 Exception attributes
 --------------------
@@ -192,6 +272,27 @@
 superclass (for example ``errno``, ``filename``, and optionally
 ``winerror``).
 
+Implementation
+--------------
+
+Since it is proposed that the subclasses are raised based purely on the
+value of ``errno``, little or no changes should be required in extension
+modules (either standard or third-party).  As long as they use the
+``PyErr_SetFromErrno()`` family of functions (or the
+``PyErr_SetFromWindowsErr()`` family of functions under Windows), they
+should automatically benefit from the new, finer-grained exception classes.
+
+Library modules written in Python, though, will have to be adapted where
+they currently use the following idiom::
+
+    raise OSError(errno.ENOTDIR, "execve called")
+
+Fortunately, such Python code is quite rare since raising OSError or IOError
+with an errno value normally happens when interfacing with system calls,
+which is usually done in C extensions.
+
+If there is popular demand, the subroutine choosing an exception type based
+on the errno value could be exposed for use in pure Python.
 
 
 Compatibility concerns
@@ -233,6 +334,28 @@
 Careful code should not be penalized.
 
 
+Possible objections
+===================
+
+Namespace pollution
+-------------------
+
+Making the exception hierarchy finer-grained makes the root (or builtins)
+namespace larger.  This is to be moderated, however, as:
+
+* only a handful of additional classes are proposed; 
+
+* while standard exception types live in the root namespace, they are
+  visually distinguished by the fact that they use the CamelCase convention,
+  while almost all other builtins use lowercase naming (except True, False,
+  None, Ellipsis and NotImplemented)
+
+An alternative would be to provide a separate module containing the
+finer-grained exceptions, but that would defeat the purpose of
+encouraging careful code over careless code, since the user would first
+have to import the new module instead of using names already accessible.
+
+
 Possible alternative
 ====================
 

From python-checkins at python.org  Wed Jul 21 15:50:26 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 21 Jul 2010 15:50:26 +0200 (CEST)
Subject: [Python-checkins] r83022 - sandbox/trunk/errnopep/pepXXXX.txt
Message-ID: <20100721135026.BD7BCE8D4@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 21 15:50:26 2010
New Revision: 83022

Log:
Fix rest code, and an example



Modified:
   sandbox/trunk/errnopep/pepXXXX.txt

Modified: sandbox/trunk/errnopep/pepXXXX.txt
==============================================================================
--- sandbox/trunk/errnopep/pepXXXX.txt	(original)
+++ sandbox/trunk/errnopep/pepXXXX.txt	Wed Jul 21 15:50:26 2010
@@ -1,4 +1,4 @@
-PEP: XXX
+PEP: 3789
 Title: Reworking the OS and IO exception hierarchy
 Version: $Revision: $
 Last-Modified: $Date: $
@@ -283,9 +283,9 @@
 should automatically benefit from the new, finer-grained exception classes.
 
 Library modules written in Python, though, will have to be adapted where
-they currently use the following idiom::
+they currently use the following idiom (seen in ``Lib/tempfile.py``)::
 
-    raise OSError(errno.ENOTDIR, "execve called")
+    raise IOError(_errno.EEXIST, "No usable temporary file name found")
 
 Fortunately, such Python code is quite rare since raising OSError or IOError
 with an errno value normally happens when interfacing with system calls,
@@ -401,7 +401,7 @@
 the standard library and its tests, as part of ``except`` clauses.
 
 Common errnos with OSError
-~~~~~~~~~~~~~~~~~~~~~~~~~~
+--------------------------
 
 * ``EBADF``: bad file descriptor (usually means the file descriptor was
              closed)
@@ -422,7 +422,7 @@
 * ``EPERM``: operation not permitted (when using e.g. os.setuid())
 
 Common errnos with IOError
-~~~~~~~~~~~~~~~~~~~~~~~~~~
+--------------------------
 
 * ``EACCES``: permission denied (for filesystem operations)
 
@@ -440,7 +440,7 @@
 * ``ETIMEDOUT``: connection timed out
 
 Common errnos with socket.error
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+-------------------------------
 
 All these errors may also be associated with a plain IOError, for example
 when calling read() on a socket's file descriptor.
@@ -469,7 +469,7 @@
 * ``EWOULDBLOCK``: same reasons as ``EAGAIN``
 
 Common errnos with select.error
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+-------------------------------
 
 * ``EINTR``: interrupted function call
 
@@ -479,7 +479,7 @@
 =============================================
 
 Interpreter core
-~~~~~~~~~~~~~~~~
+----------------
 
 Handling of PYTHONSTARTUP raises IOError (but the error gets discarded)::
 
@@ -496,10 +496,10 @@
 
 
 Modules
-~~~~~~~
+-------
 
 bz2
----
+~~~
 
 Raises IOError throughout (OSError is unused)::
 
@@ -517,12 +517,12 @@
     IOError: file is not ready for reading
 
 curses
-------
+~~~~~~
 
 Not examined.
 
 dbm.gnu, dbm.ndbm
------------------
+~~~~~~~~~~~~~~~~~
 
 _dbm.error and _gdbm.error inherit from IOError::
 
@@ -532,12 +532,12 @@
     _gdbm.error: [Errno 2] No such file or directory
 
 fcntl
------
+~~~~~
 
 Raises IOError throughout (OSError is unused).
 
 imp module
-----------
+~~~~~~~~~~
 
 Raises IOError for bad file descriptors::
 
@@ -547,7 +547,7 @@
     IOError: [Errno 9] Bad file descriptor
 
 io module
----------
+~~~~~~~~~
 
 Raises IOError when trying to open a directory under Unix::
 
@@ -599,12 +599,12 @@
 object would block.
 
 multiprocessing
----------------
+~~~~~~~~~~~~~~~
 
 Not examined.
 
 ossaudiodev
------------
+~~~~~~~~~~~
 
 Raises IOError throughout (OSError is unused)::
 
@@ -614,7 +614,7 @@
     IOError: [Errno 2] No such file or directory: 'foo'
 
 readline
---------
+~~~~~~~~
 
 Raises IOError in various file-handling functions::
 
@@ -632,7 +632,7 @@
     IOError: [Errno 13] Permission denied
 
 select
-------
+~~~~~~
 
 select() and poll objects raise select.error, which doesn't inherit from
 anything (but poll.modify() which raises IOError).
@@ -640,21 +640,21 @@
 kqueue objects raise both OSError and IOError.
 
 signal
-------
+~~~~~~
 
 signal.ItimerError inherits from IOError.
 
 socket
-------
+~~~~~~
 
 socket.error inherits from IOError.
 
 time
-----
+~~~~
 
 Raises IOError for internal errors in time.time() and time.sleep().
 
 zipimport
----------
+~~~~~~~~~
 
 zipimporter.get_data() can raise IOError.

From python-checkins at python.org  Wed Jul 21 16:46:44 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 21 Jul 2010 16:46:44 +0200 (CEST)
Subject: [Python-checkins] r83023 - sandbox/trunk/errnopep/pepXXXX.txt
Message-ID: <20100721144644.85BFBED8F@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 21 16:46:44 2010
New Revision: 83023

Log:
add references, copyright, and a couple of cosmetic fixes



Modified:
   sandbox/trunk/errnopep/pepXXXX.txt

Modified: sandbox/trunk/errnopep/pepXXXX.txt
==============================================================================
--- sandbox/trunk/errnopep/pepXXXX.txt	(original)
+++ sandbox/trunk/errnopep/pepXXXX.txt	Wed Jul 21 16:46:44 2010
@@ -144,19 +144,12 @@
         pass
 
 
-Moratorium
-==========
-
-The moratorium in effect on language builtins means this PEP has little
-chance to be accepted for Python 3.2.
-
-
 Step 1: coalesce exception types
 ================================
 
 The first step of the resolution is to coalesce existing exception types.
 The extent of this step is not yet fully determined.  A number of possible
-changes are listed hereafter::
+changes are listed hereafter:
 
 * alias both socket.error and select.error to IOError
 * alias IOError to OSError
@@ -356,6 +349,27 @@
 have to import the new module instead of using names already accessible.
 
 
+Earlier discussion
+==================
+
+While this is the first time such as formal proposal is made, the idea
+has received informal support in the past [1]_; both the introduction
+of finer-grained exception classes and the coalescing of OSError and
+IOError.
+
+The removal of WindowsError alone has been discussed and rejected
+as part of another PEP [2]_, but there seemed to be a consensus that the
+distinction with OSError wasn't meaningful.  This supports at least its
+aliasing with OSError.
+
+
+Moratorium
+==========
+
+The moratorium in effect on language builtins means this PEP has little
+chance to be accepted for Python 3.2.
+
+
 Possible alternative
 ====================
 
@@ -658,3 +672,19 @@
 ~~~~~~~~~
 
 zipimporter.get_data() can raise IOError.
+
+
+References
+==========
+
+.. [1] "IO module precisions and exception hierarchy"
+   http://mail.python.org/pipermail/python-dev/2009-September/092130.html
+
+.. [2] Discussion of "Removing WindowsError" in PEP 348
+   http://www.python.org/dev/peps/pep-0348/#removing-windowserror
+
+Copyright
+=========
+
+This document has been placed in the public domain.
+

From python-checkins at python.org  Wed Jul 21 17:36:19 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 21 Jul 2010 17:36:19 +0200 (CEST)
Subject: [Python-checkins] r83024 - sandbox/trunk/errnopep/pepXXXX.txt
Message-ID: <20100721153619.6EB66CA3F@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 21 17:36:19 2010
New Revision: 83024

Log:
Some cosmetic fixes



Modified:
   sandbox/trunk/errnopep/pepXXXX.txt

Modified: sandbox/trunk/errnopep/pepXXXX.txt
==============================================================================
--- sandbox/trunk/errnopep/pepXXXX.txt	(original)
+++ sandbox/trunk/errnopep/pepXXXX.txt	Wed Jul 21 17:36:19 2010
@@ -1,13 +1,15 @@
-PEP: 3789
+PEP: 3151
 Title: Reworking the OS and IO exception hierarchy
 Version: $Revision: $
 Last-Modified: $Date: $
 Author: Antoine Pitrou <solipsis at pitrou.net>
-Status: 
+Status: Draft
 Type: Standards Track
 Content-Type: text/x-rst
-Created: 
+Created: 2010-07-21
+Python-Version: 3.2 or 3.3
 Post-History:
+Resolution: TBD
 
 
 Abstract
@@ -217,36 +219,36 @@
 The following temptative list of subclasses, along with a description and
 the list of errnos mapped to them, is submitted to discussion:
 
-* FileAlreadyExists: trying to create a file or directory which already
+* ``FileAlreadyExists``: trying to create a file or directory which already
   exists (EEXIST)
 
-* FileNotFound: for all circumstances where a file and directory is
+* ``FileNotFound``: for all circumstances where a file and directory is
   requested but doesn't exist (ENOENT)
 
-* IsADirectory: file-level operation (open(), os.remove()...) requested
+* ``IsADirectory``: file-level operation (open(), os.remove()...) requested
   on a directory (EISDIR)
 
-* NotADirectory: directory-level operation requested on something else
+* ``NotADirectory``: directory-level operation requested on something else
   (ENOTDIR)
 
-* PermissionDenied: trying to run an operation without the adequate access
+* ``PermissionDenied``: trying to run an operation without the adequate access
   rights - for example filesystem permissions (EACCESS, optionally EPERM)
 
-* BlockingIOError: an operation would block on an object (e.g. socket) set
+* ``BlockingIOError``: an operation would block on an object (e.g. socket) set
   for non-blocking operation (EAGAIN, EALREADY, EWOULDBLOCK, EINPROGRESS);
   this is the existing ``io.BlockingIOError`` with an extended role
 
-* BadFileDescriptor: operation on an invalid file descriptor (EBADF);
+* ``BadFileDescriptor``: operation on an invalid file descriptor (EBADF);
   the default error message could point out that most causes are that
   an existing file descriptor has been closed
 
-* ConnectionAborted: connection attempt aborted by peer (ECONNABORTED)
+* ``ConnectionAborted``: connection attempt aborted by peer (ECONNABORTED)
 
-* ConnectionRefused: connection reset by peer (ECONNREFUSED)
+* ``ConnectionRefused``: connection reset by peer (ECONNREFUSED)
 
-* ConnectionReset: connection reset by peer (ECONNRESET)
+* ``ConnectionReset``: connection reset by peer (ECONNRESET)
 
-* TimeoutError: connection timed out (ECONNTIMEOUT); this could be re-cast
+* ``TimeoutError``: connection timed out (ECONNTIMEOUT); this could be re-cast
   as a generic timeout exception, useful for other types of timeout (for
   example in Lock.acquire())
 
@@ -418,7 +420,7 @@
 --------------------------
 
 * ``EBADF``: bad file descriptor (usually means the file descriptor was
-             closed)
+  closed)
 
 * ``EEXIST``: file or directory exists
 
@@ -460,9 +462,10 @@
 when calling read() on a socket's file descriptor.
 
 * ``EAGAIN``: resource temporarily unavailable (during a non-blocking socket
-              call except connect())
+  call except connect())
 
-* ``EALREADY``: connection already in progress (during a non-blocking connect())
+* ``EALREADY``: connection already in progress (during a non-blocking
+  connect())
 
 * ``EINPROGRESS``: operation in progress (during a non-blocking connect())
 
@@ -488,7 +491,6 @@
 * ``EINTR``: interrupted function call
 
 
-
 Appendix B: Survey of raised OS and IO errors
 =============================================
 
@@ -508,12 +510,11 @@
 `FILE *` parameter (which, in the source tree, is always either stdout or
 stderr).
 
-
-Modules
--------
+Standard library
+----------------
 
 bz2
-~~~
+'''
 
 Raises IOError throughout (OSError is unused)::
 
@@ -531,12 +532,12 @@
     IOError: file is not ready for reading
 
 curses
-~~~~~~
+''''''
 
 Not examined.
 
 dbm.gnu, dbm.ndbm
-~~~~~~~~~~~~~~~~~
+'''''''''''''''''
 
 _dbm.error and _gdbm.error inherit from IOError::
 
@@ -546,12 +547,12 @@
     _gdbm.error: [Errno 2] No such file or directory
 
 fcntl
-~~~~~
+'''''
 
 Raises IOError throughout (OSError is unused).
 
 imp module
-~~~~~~~~~~
+''''''''''
 
 Raises IOError for bad file descriptors::
 
@@ -561,7 +562,7 @@
     IOError: [Errno 9] Bad file descriptor
 
 io module
-~~~~~~~~~
+'''''''''
 
 Raises IOError when trying to open a directory under Unix::
 
@@ -613,12 +614,12 @@
 object would block.
 
 multiprocessing
-~~~~~~~~~~~~~~~
+'''''''''''''''
 
 Not examined.
 
 ossaudiodev
-~~~~~~~~~~~
+'''''''''''
 
 Raises IOError throughout (OSError is unused)::
 
@@ -628,7 +629,7 @@
     IOError: [Errno 2] No such file or directory: 'foo'
 
 readline
-~~~~~~~~
+''''''''
 
 Raises IOError in various file-handling functions::
 
@@ -646,7 +647,7 @@
     IOError: [Errno 13] Permission denied
 
 select
-~~~~~~
+''''''
 
 select() and poll objects raise select.error, which doesn't inherit from
 anything (but poll.modify() which raises IOError).
@@ -654,22 +655,22 @@
 kqueue objects raise both OSError and IOError.
 
 signal
-~~~~~~
+''''''
 
 signal.ItimerError inherits from IOError.
 
 socket
-~~~~~~
+''''''
 
 socket.error inherits from IOError.
 
 time
-~~~~
+''''
 
 Raises IOError for internal errors in time.time() and time.sleep().
 
 zipimport
-~~~~~~~~~
+'''''''''
 
 zipimporter.get_data() can raise IOError.
 
@@ -688,3 +689,13 @@
 
 This document has been placed in the public domain.
 
+
+
+..
+   Local Variables:
+   mode: indented-text
+   indent-tabs-mode: nil
+   sentence-end-double-space: t
+   fill-column: 70
+   coding: utf-8
+   End:

From python-checkins at python.org  Wed Jul 21 17:54:48 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 21 Jul 2010 17:54:48 +0200 (CEST)
Subject: [Python-checkins] r83025 -
	python/branches/py3k/Doc/library/string.rst
Message-ID: <20100721155448.5625BEF29@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 21 17:54:48 2010
New Revision: 83025

Log:
Remove outdated mention of deprecated functions in the string module - they have been removed in 3.x.



Modified:
   python/branches/py3k/Doc/library/string.rst

Modified: python/branches/py3k/Doc/library/string.rst
==============================================================================
--- python/branches/py3k/Doc/library/string.rst	(original)
+++ python/branches/py3k/Doc/library/string.rst	Wed Jul 21 17:54:48 2010
@@ -7,13 +7,13 @@
 
 .. index:: module: re
 
-The :mod:`string` module contains a number of useful constants and classes, as
-well as some deprecated legacy functions that are also available as methods on
-strings. In addition, Python's built-in string classes support the sequence type
-methods described in the :ref:`typesseq` section, and also the string-specific
-methods described in the :ref:`string-methods` section. To output formatted
-strings, see the :ref:`string-formatting` section. Also, see the :mod:`re`
-module for string functions based on regular expressions.
+The :mod:`string` module contains a number of useful constants and classes
+for string formatting.  In addition, Python's built-in string classes
+support the sequence type methods described in the :ref:`typesseq`
+section, and also the string-specific methods described in the
+:ref:`string-methods` section.  To output formatted strings, see the
+:ref:`string-formatting` section.  Also, see the :mod:`re` module for
+string functions based on regular expressions.
 
 
 String constants

From python-checkins at python.org  Wed Jul 21 17:55:40 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 21 Jul 2010 17:55:40 +0200 (CEST)
Subject: [Python-checkins] r83026 - in python/branches/release31-maint:
	Doc/library/string.rst
Message-ID: <20100721155540.374BCEEF8@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 21 17:55:40 2010
New Revision: 83026

Log:
Merged revisions 83025 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83025 | antoine.pitrou | 2010-07-21 17:54:48 +0200 (mer., 21 juil. 2010) | 3 lines
  
  Remove outdated mention of deprecated functions in the string module - they have been removed in 3.x.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Doc/library/string.rst

Modified: python/branches/release31-maint/Doc/library/string.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/string.rst	(original)
+++ python/branches/release31-maint/Doc/library/string.rst	Wed Jul 21 17:55:40 2010
@@ -7,13 +7,13 @@
 
 .. index:: module: re
 
-The :mod:`string` module contains a number of useful constants and classes, as
-well as some deprecated legacy functions that are also available as methods on
-strings. In addition, Python's built-in string classes support the sequence type
-methods described in the :ref:`typesseq` section, and also the string-specific
-methods described in the :ref:`string-methods` section. To output formatted
-strings, see the :ref:`string-formatting` section. Also, see the :mod:`re`
-module for string functions based on regular expressions.
+The :mod:`string` module contains a number of useful constants and classes
+for string formatting.  In addition, Python's built-in string classes
+support the sequence type methods described in the :ref:`typesseq`
+section, and also the string-specific methods described in the
+:ref:`string-methods` section.  To output formatted strings, see the
+:ref:`string-formatting` section.  Also, see the :mod:`re` module for
+string functions based on regular expressions.
 
 
 String constants

From python-checkins at python.org  Wed Jul 21 17:56:03 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 21 Jul 2010 17:56:03 +0200 (CEST)
Subject: [Python-checkins] r83027 - python/branches/release27-maint
Message-ID: <20100721155603.A394FEFAF@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 21 17:56:03 2010
New Revision: 83027

Log:
Blocked revisions 83025 via svnmerge

........
  r83025 | antoine.pitrou | 2010-07-21 17:54:48 +0200 (mer., 21 juil. 2010) | 3 lines
  
  Remove outdated mention of deprecated functions in the string module - they have been removed in 3.x.
........


Modified:
   python/branches/release27-maint/   (props changed)

From python-checkins at python.org  Wed Jul 21 18:12:40 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 21 Jul 2010 18:12:40 +0200 (CEST)
Subject: [Python-checkins] r83028 - sandbox/trunk/errnopep/pepXXXX.txt
Message-ID: <20100721161240.54E87EE2C@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 21 18:12:40 2010
New Revision: 83028

Log:
Explain an example, and put forward the compatibility section



Modified:
   sandbox/trunk/errnopep/pepXXXX.txt

Modified: sandbox/trunk/errnopep/pepXXXX.txt
==============================================================================
--- sandbox/trunk/errnopep/pepXXXX.txt	(original)
+++ sandbox/trunk/errnopep/pepXXXX.txt	Wed Jul 21 18:12:40 2010
@@ -138,7 +138,12 @@
     except OSError:
         pass
 
-What the programmer would like to write instead is something such as::
+``os.remove`` can raise an OSError not only when the file doesn't exist,
+but in other possible situations (for example, the filename points to a
+directory, or the current process doesn't have permission to remove
+the file), which all indicate bugs in the application logic and therefore
+shouldn't be silenced.  What the programmer would like to write instead is
+something such as::
 
     try:
         os.remove(filename)
@@ -146,6 +151,45 @@
         pass
 
 
+Compatibility concerns
+======================
+
+Reworking the exception hierarchy will obviously change the exact semantics
+of at least some existing code.  While it is not possible to improve on the
+current situation without changing exact semantics, it is possible to define
+a narrower type of compatibility, which we will call **useful compatibility**,
+and define as follows:
+
+* *useful compatibility* doesn't make exception catching any narrower, but
+  it can be broader for *na?ve* exception-catching code.  Given the following
+  kind of snippet, all exceptions caught before this PEP will also be
+  caught after this PEP, but the reverse may be false::
+  
+      try:
+          os.remove(filename)
+      except OSError:
+          pass
+
+* *useful compatibility* doesn't alter the behaviour of *careful*
+  exception-catching code.  Given the following kind of snippet, the same
+  errors should be silenced or reraised, regardless of whether this PEP
+  has been implemented or not::
+
+      try:
+          os.remove(filename)
+      except OSError as e:
+          if e.errno != errno.ENOENT:
+              raise
+
+The rationale for this compromise is that careless (or "na?ve") code
+can't really be helped, but at least code which "works" won't suddenly
+raise errors and crash.  This is important since such code is likely to
+be present in scripts used as cron tasks or automated system administration
+programs.
+
+Careful code should not be penalized.
+
+
 Step 1: coalesce exception types
 ================================
 
@@ -290,45 +334,6 @@
 on the errno value could be exposed for use in pure Python.
 
 
-Compatibility concerns
-======================
-
-Reworking the exception hierarchy will obviously change the exact semantics
-of at least some existing code.  While it is not possible to improve on the
-current situation without changing exact semantics, it is possible to define
-a narrower type of compatibility, which we will call **useful compatibility**,
-and define as follows:
-
-* *useful compatibility* doesn't make exception catching any narrower, but
-  it can be broader for *na?ve* exception-catching code.  Given the following
-  kind of snippet, all exceptions caught before this PEP will also be
-  caught after this PEP, but the reverse may be false::
-  
-      try:
-          os.remove(filename)
-      except OSError:
-          pass
-
-* *useful compatibility* doesn't alter the behaviour of *careful*
-  exception-catching code.  Given the following kind of snippet, the same
-  errors should be silenced or reraised, regardless of whether this PEP
-  has been implemented or not::
-
-      try:
-          os.remove(filename)
-      except OSError as e:
-          if e.errno != errno.ENOENT:
-              raise
-
-The rationale for this compromise is that dangerous (or "na?ve") code
-can't really be helped, but at least code which "works" won't suddenly
-raise errors and crash.  This is important since such code is likely to
-be present in scripts used as cron tasks or automated system administration
-programs.
-
-Careful code should not be penalized.
-
-
 Possible objections
 ===================
 

From python-checkins at python.org  Wed Jul 21 18:16:19 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 21 Jul 2010 18:16:19 +0200 (CEST)
Subject: [Python-checkins] r83029 - sandbox/trunk/errnopep/pepXXXX.txt
Message-ID: <20100721161619.766D3EFB5@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 21 18:16:19 2010
New Revision: 83029

Log:
Cosmetics



Modified:
   sandbox/trunk/errnopep/pepXXXX.txt

Modified: sandbox/trunk/errnopep/pepXXXX.txt
==============================================================================
--- sandbox/trunk/errnopep/pepXXXX.txt	(original)
+++ sandbox/trunk/errnopep/pepXXXX.txt	Wed Jul 21 18:16:19 2010
@@ -202,8 +202,7 @@
 * alias WindowsError to OSError
 
 Each of these changes doesn't preserve exact compatibility, but it does
-preserve *useful compatibility* (see later for a definition of useful
-compatibility).
+preserve *useful compatibility* (see "compatibility" section above).
 
 Not only does this first step present the user a simpler landscape, but
 it also allows for a better and more complete resolution of step 2
@@ -243,8 +242,8 @@
 Prerequisite, because some errnos can currently be attached to different
 exception classes: for example, EBADF can be attached to both OSError and
 IOError, depending on the context.  If we don't want to break *useful
-compatibility*, we can't make an ``except OSError`` (or IOError) fail
-to match an exception where it would succeed today.
+compatibility*, we can't make an ``except OSError`` (or IOError) fail to
+match an exception where it would succeed today.
 
 Loose, because we could decide for a partial resolution of step 2
 if existing exception classes are not coalesced: for example, EBADF could

From python-checkins at python.org  Wed Jul 21 18:41:31 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 21 Jul 2010 18:41:31 +0200 (CEST)
Subject: [Python-checkins] r83030 -
	python/branches/py3k/Lib/test/test_array.py
Message-ID: <20100721164131.A2491F29A@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 21 18:41:31 2010
New Revision: 83030

Log:
Issue #5395: check that array.fromfile() re-raises an IOError instead of replacing it
with EOFError.
(this is only an added test, but 2.x will get a fix too)



Modified:
   python/branches/py3k/Lib/test/test_array.py

Modified: python/branches/py3k/Lib/test/test_array.py
==============================================================================
--- python/branches/py3k/Lib/test/test_array.py	(original)
+++ python/branches/py3k/Lib/test/test_array.py	Wed Jul 21 18:41:31 2010
@@ -326,6 +326,17 @@
                 f.close()
             support.unlink(support.TESTFN)
 
+    def test_fromfile_ioerror(self):
+        # Issue #5395: Check if fromfile raises a proper IOError
+        # instead of EOFError.
+        a = array.array(self.typecode)
+        f = open(support.TESTFN, 'wb')
+        try:
+            self.assertRaises(IOError, a.fromfile, f, len(self.example))
+        finally:
+            f.close()
+            support.unlink(support.TESTFN)
+
     def test_filewrite(self):
         a = array.array(self.typecode, 2*self.example)
         f = open(support.TESTFN, 'wb')

From python-checkins at python.org  Wed Jul 21 18:47:28 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 21 Jul 2010 18:47:28 +0200 (CEST)
Subject: [Python-checkins] r83031 - in python/branches/release27-maint:
	Lib/test/test_array.py Misc/ACKS Misc/NEWS Modules/arraymodule.c
Message-ID: <20100721164728.C5BF7C386@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 21 18:47:28 2010
New Revision: 83031

Log:
Merged revisions 83030 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83030 | antoine.pitrou | 2010-07-21 18:41:31 +0200 (mer., 21 juil. 2010) | 5 lines
  
  Issue #5395: check that array.fromfile() re-raises an IOError instead of replacing it
  with EOFError.
  (this is only an added test, but 2.x will get a fix too)
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/test/test_array.py
   python/branches/release27-maint/Misc/ACKS
   python/branches/release27-maint/Misc/NEWS
   python/branches/release27-maint/Modules/arraymodule.c

Modified: python/branches/release27-maint/Lib/test/test_array.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_array.py	(original)
+++ python/branches/release27-maint/Lib/test/test_array.py	Wed Jul 21 18:47:28 2010
@@ -188,6 +188,17 @@
                 f.close()
             test_support.unlink(test_support.TESTFN)
 
+    def test_fromfile_ioerror(self):
+        # Issue #5395: Check if fromfile raises a proper IOError
+        # instead of EOFError.
+        a = array.array(self.typecode)
+        f = open(test_support.TESTFN, 'wb')
+        try:
+            self.assertRaises(IOError, a.fromfile, f, len(self.example))
+        finally:
+            f.close()
+            test_support.unlink(test_support.TESTFN)
+
     def test_filewrite(self):
         a = array.array(self.typecode, 2*self.example)
         f = open(test_support.TESTFN, 'wb')

Modified: python/branches/release27-maint/Misc/ACKS
==============================================================================
--- python/branches/release27-maint/Misc/ACKS	(original)
+++ python/branches/release27-maint/Misc/ACKS	Wed Jul 21 18:47:28 2010
@@ -355,6 +355,7 @@
 Brian Hooper
 Randall Hopper
 Nadav Horesh
+Jan Hosang
 Ken Howard
 Brad Howes
 Chih-Hao Huang

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Wed Jul 21 18:47:28 2010
@@ -18,6 +18,10 @@
 Library
 -------
 
+- Issue #5395: array.fromfile() would raise a spurious EOFError when an
+  I/O error occurred.  Now an IOError is raised instead.  Patch by chuck
+  (Jan Hosang).
+
 - Issue 1712522: urllib.quote supports Unicode String with encoding and errors
   parameter. The encoding parameter defaults to utf-8 and errors to strict.
   Patch by Matt Giuca.

Modified: python/branches/release27-maint/Modules/arraymodule.c
==============================================================================
--- python/branches/release27-maint/Modules/arraymodule.c	(original)
+++ python/branches/release27-maint/Modules/arraymodule.c	Wed Jul 21 18:47:28 2010
@@ -1228,8 +1228,14 @@
             PyMem_RESIZE(item, char, Py_SIZE(self)*itemsize);
             self->ob_item = item;
             self->allocated = Py_SIZE(self);
-            PyErr_SetString(PyExc_EOFError,
-                             "not enough items in file");
+            if (ferror(fp)) {
+                PyErr_SetFromErrno(PyExc_IOError);
+                clearerr(fp);
+            }
+            else {
+                PyErr_SetString(PyExc_EOFError,
+                                "not enough items in file");
+            }
             return NULL;
         }
     }

From python-checkins at python.org  Wed Jul 21 18:50:53 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 21 Jul 2010 18:50:53 +0200 (CEST)
Subject: [Python-checkins] r83032 - in python/branches/release26-maint:
	Lib/test/test_array.py Misc/ACKS Misc/NEWS Modules/arraymodule.c
Message-ID: <20100721165053.2859DEB2A@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 21 18:50:52 2010
New Revision: 83032

Log:
Merged revisions 83031 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

................
  r83031 | antoine.pitrou | 2010-07-21 18:47:28 +0200 (mer., 21 juil. 2010) | 11 lines
  
  Merged revisions 83030 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k
  
  ........
    r83030 | antoine.pitrou | 2010-07-21 18:41:31 +0200 (mer., 21 juil. 2010) | 5 lines
    
    Issue #5395: check that array.fromfile() re-raises an IOError instead of replacing it
    with EOFError.
    (this is only an added test, but 2.x will get a fix too)
  ........
................


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Lib/test/test_array.py
   python/branches/release26-maint/Misc/ACKS
   python/branches/release26-maint/Misc/NEWS
   python/branches/release26-maint/Modules/arraymodule.c

Modified: python/branches/release26-maint/Lib/test/test_array.py
==============================================================================
--- python/branches/release26-maint/Lib/test/test_array.py	(original)
+++ python/branches/release26-maint/Lib/test/test_array.py	Wed Jul 21 18:50:52 2010
@@ -188,6 +188,17 @@
                 f.close()
             test_support.unlink(test_support.TESTFN)
 
+    def test_fromfile_ioerror(self):
+        # Issue #5395: Check if fromfile raises a proper IOError
+        # instead of EOFError.
+        a = array.array(self.typecode)
+        f = open(test_support.TESTFN, 'wb')
+        try:
+            self.assertRaises(IOError, a.fromfile, f, len(self.example))
+        finally:
+            f.close()
+            test_support.unlink(test_support.TESTFN)
+
     def test_tofromlist(self):
         a = array.array(self.typecode, 2*self.example)
         b = array.array(self.typecode)

Modified: python/branches/release26-maint/Misc/ACKS
==============================================================================
--- python/branches/release26-maint/Misc/ACKS	(original)
+++ python/branches/release26-maint/Misc/ACKS	Wed Jul 21 18:50:52 2010
@@ -328,6 +328,7 @@
 Brian Hooper
 Randall Hopper
 Nadav Horesh
+Jan Hosang
 Ken Howard
 Brad Howes
 Chih-Hao Huang

Modified: python/branches/release26-maint/Misc/NEWS
==============================================================================
--- python/branches/release26-maint/Misc/NEWS	(original)
+++ python/branches/release26-maint/Misc/NEWS	Wed Jul 21 18:50:52 2010
@@ -81,6 +81,10 @@
 Library
 -------
 
+- Issue #5395: array.fromfile() would raise a spurious EOFError when an
+  I/O error occurred.  Now an IOError is raised instead.  Patch by chuck
+  (Jan Hosang).
+
 - Issue #1555570: email no longer inserts extra blank lines when a \r\n
   combo crosses an 8192 byte boundary.
 

Modified: python/branches/release26-maint/Modules/arraymodule.c
==============================================================================
--- python/branches/release26-maint/Modules/arraymodule.c	(original)
+++ python/branches/release26-maint/Modules/arraymodule.c	Wed Jul 21 18:50:52 2010
@@ -1262,8 +1262,14 @@
             PyMem_RESIZE(item, char, Py_SIZE(self)*itemsize);
             self->ob_item = item;
             self->allocated = Py_SIZE(self);
-            PyErr_SetString(PyExc_EOFError,
-                             "not enough items in file");
+            if (ferror(fp)) {
+                PyErr_SetFromErrno(PyExc_IOError);
+                clearerr(fp);
+            }
+            else {
+                PyErr_SetString(PyExc_EOFError,
+                                "not enough items in file");
+            }
             return NULL;
         }
     }

From python-checkins at python.org  Wed Jul 21 18:53:04 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 21 Jul 2010 18:53:04 +0200 (CEST)
Subject: [Python-checkins] r83033 - in python/branches/release31-maint:
	Lib/test/test_array.py
Message-ID: <20100721165304.7A181F41E@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 21 18:53:04 2010
New Revision: 83033

Log:
Merged revisions 83030 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83030 | antoine.pitrou | 2010-07-21 18:41:31 +0200 (mer., 21 juil. 2010) | 5 lines
  
  Issue #5395: check that array.fromfile() re-raises an IOError instead of replacing it
  with EOFError.
  (this is only an added test, but 2.x will get a fix too)
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/test/test_array.py

Modified: python/branches/release31-maint/Lib/test/test_array.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_array.py	(original)
+++ python/branches/release31-maint/Lib/test/test_array.py	Wed Jul 21 18:53:04 2010
@@ -182,6 +182,17 @@
                 f.close()
             support.unlink(support.TESTFN)
 
+    def test_fromfile_ioerror(self):
+        # Issue #5395: Check if fromfile raises a proper IOError
+        # instead of EOFError.
+        a = array.array(self.typecode)
+        f = open(support.TESTFN, 'wb')
+        try:
+            self.assertRaises(IOError, a.fromfile, f, len(self.example))
+        finally:
+            f.close()
+            support.unlink(support.TESTFN)
+
     def test_tofromlist(self):
         a = array.array(self.typecode, 2*self.example)
         b = array.array(self.typecode)

From python-checkins at python.org  Wed Jul 21 19:18:39 2010
From: python-checkins at python.org (georg.brandl)
Date: Wed, 21 Jul 2010 19:18:39 +0200 (CEST)
Subject: [Python-checkins] r83034 - peps/trunk/pep-3151.txt
Message-ID: <20100721171839.D07EFD765@mail.python.org>

Author: georg.brandl
Date: Wed Jul 21 19:18:39 2010
New Revision: 83034

Log:
Add PEP 3151 -- reworking the OS/IO exception hierarchy.

Added:
   peps/trunk/pep-3151.txt   (contents, props changed)

Added: peps/trunk/pep-3151.txt
==============================================================================
--- (empty file)
+++ peps/trunk/pep-3151.txt	Wed Jul 21 19:18:39 2010
@@ -0,0 +1,705 @@
+PEP: 3151
+Title: Reworking the OS and IO exception hierarchy
+Version: $Revision$
+Last-Modified: $Date$
+Author: Antoine Pitrou <solipsis at pitrou.net>
+Status: Draft
+Type: Standards Track
+Content-Type: text/x-rst
+Created: 2010-07-21
+Python-Version: 3.2 or 3.3
+Post-History:
+Resolution: TBD
+
+
+Abstract
+========
+
+The standard exception hierarchy is an important part of the Python
+language.  It has two defining qualities: it is both generic and
+selective.  Generic in that the same exception type can be raised
+- and handled - regardless of the context (for example, whether you are
+trying to add something to an integer, to call a string method, or to write
+an object on a socket, a TypeError will be raised for bad argument types).
+Selective in that it allows the user to easily handle (silence, examine,
+process, store or encapsulate...) specific kinds of error conditions
+while letting other errors bubble up to higher calling contexts.  For
+example, you can choose to catch ZeroDivisionErrors without affecting
+the default handling of other ArithmeticErrors (such as OverflowErrors).
+
+This PEP proposes changes to a part of the exception hierarchy in
+order to better embody the qualities mentioned above: the errors
+related to operating system calls (OSError, IOError, select.error, and
+all their subclasses).
+
+
+Rationale
+=========
+
+Confusing set of OS-related exceptions
+--------------------------------------
+
+OS-related (or system call-related) exceptions are currently a diversity
+of classes, arranged in the following subhierarchies::
+
+    +-- EnvironmentError
+        +-- IOError
+            +-- io.BlockingIOError
+            +-- io.UnsupportedOperation (also inherits from ValueError)
+            +-- socket.error
+        +-- OSError
+            +-- WindowsError
+    +-- select.error
+
+While some of these distinctions can be explained by implementation
+considerations, they are often not very logical at a higher level.  The
+line separating OSError and IOError, for example, is often blurry.  Consider
+the following::
+
+    >>> os.remove("fff")
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    OSError: [Errno 2] No such file or directory: 'fff'
+    >>> open("fff")
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: [Errno 2] No such file or directory: 'fff'
+
+The same error condition (a non-existing file) gets cast as two different
+exceptions depending on which library function was called.  The reason
+for this is that the `os` module exclusively raises OSError (or its
+subclass WindowsError) while the `io` module mostly raises IOError.
+However, the user is interested in the nature of the error, not in which
+part of the interpreter it comes from (since the latter is obvious from
+reading the traceback message or application source code).
+
+In fact, it is hard to think of any situation where OSError should be
+caught but not IOError, or the reverse.
+
+A further proof of the ambiguity of this segmentation is that the standard
+library itself sometimes has problems deciding.  For example, in the
+``select`` module, similar failures will raise either ``select.error``,
+``OSError`` or ``IOError`` depending on whether you are using select(),
+a poll object, a kqueue object, or an epoll object.  This makes user code
+uselessly complicated since it has to be prepared to catch various
+exception types, depending on which exact implementation of a single
+primitive it chooses to use at runtime.
+
+As for WindowsError, it seems to be a pointless distinction.  First, it
+only exists on Windows systems, which requires tedious compatibility code
+in cross-platform applications.  Second, it inherits from OSError and
+is raised for similar errors as OSError is raised for on other systems.
+Third, the user wanting access to low-level exception specifics has to
+examine the ``errno`` or ``winerror`` attribute anyway.
+
+
+Lack of fine-grained exceptions
+-------------------------------
+
+The current variety of OS-related exceptions doesn't allow the user to filter
+easily for the desired kinds of failures.  As an example, consider the task
+of deleting a file if it exists.  The Look Before You Leap (LBYL) idiom
+suffers from an obvious race condition::
+
+    if os.path.exists(filename):
+        os.remove(filename)
+
+If a file named as `filename` is created by another thread or process
+between the calls to `os.path.exists` and `os.remove`, it won't be deleted.
+This can produce bugs in the application, or even security issues.
+
+Therefore, the solution is to try to remove the file, and ignore the error
+if the file doesn't exist (an idiom known as Easier to Ask Forgiveness
+than to get Permission, or EAFP).  Careful code will read like the following
+(which works under both POSIX and Windows systems)::
+
+    try:
+        os.remove(filename)
+    except OSError as e:
+        if e.errno != errno.ENOENT:
+            raise
+
+or even::
+
+    try:
+        os.remove(filename)
+    except EnvironmentError as e:
+        if e.errno != errno.ENOENT:
+            raise
+
+This is a lot more to type, and also forces the user to remember the various
+cryptic mnemonics from the errno module.  It imposes an additional cognitive
+burden and gets tiresome rather quickly.  Consequently, many programmers
+will instead write the following code, which silences exceptions too
+broadly::
+
+    try:
+        os.remove(filename)
+    except OSError:
+        pass
+
+``os.remove`` can raise an OSError not only when the file doesn't exist,
+but in other possible situations (for example, the filename points to a
+directory, or the current process doesn't have permission to remove
+the file), which all indicate bugs in the application logic and therefore
+shouldn't be silenced.  What the programmer would like to write instead is
+something such as::
+
+    try:
+        os.remove(filename)
+    except FileNotFound:
+        pass
+
+
+Compatibility concerns
+======================
+
+Reworking the exception hierarchy will obviously change the exact semantics
+of at least some existing code.  While it is not possible to improve on the
+current situation without changing exact semantics, it is possible to define
+a narrower type of compatibility, which we will call **useful compatibility**,
+and define as follows:
+
+* *useful compatibility* doesn't make exception catching any narrower, but
+  it can be broader for *na?ve* exception-catching code.  Given the following
+  kind of snippet, all exceptions caught before this PEP will also be
+  caught after this PEP, but the reverse may be false::
+  
+      try:
+          os.remove(filename)
+      except OSError:
+          pass
+
+* *useful compatibility* doesn't alter the behaviour of *careful*
+  exception-catching code.  Given the following kind of snippet, the same
+  errors should be silenced or reraised, regardless of whether this PEP
+  has been implemented or not::
+
+      try:
+          os.remove(filename)
+      except OSError as e:
+          if e.errno != errno.ENOENT:
+              raise
+
+The rationale for this compromise is that careless (or "na?ve") code
+can't really be helped, but at least code which "works" won't suddenly
+raise errors and crash.  This is important since such code is likely to
+be present in scripts used as cron tasks or automated system administration
+programs.
+
+Careful code should not be penalized.
+
+
+Step 1: coalesce exception types
+================================
+
+The first step of the resolution is to coalesce existing exception types.
+The extent of this step is not yet fully determined.  A number of possible
+changes are listed hereafter:
+
+* alias both socket.error and select.error to IOError
+* alias IOError to OSError
+* alias WindowsError to OSError
+
+Each of these changes doesn't preserve exact compatibility, but it does
+preserve *useful compatibility* (see "compatibility" section above).
+
+Not only does this first step present the user a simpler landscape, but
+it also allows for a better and more complete resolution of step 2
+(see "Prerequisite" below).
+
+Deprecation of names
+--------------------
+
+It is not yet decided whether the old names will be deprecated (then removed)
+or all alternative names will continue living in the root namespace.
+
+
+Step 2: define additional subclasses
+====================================
+
+The second step of the resolution is to extend the hierarchy by defining
+subclasses which will be raised, rather than their parent, for specific
+errno values.  Which errno values is subject to discussion, but a survey
+of existing exception matching practices (see Appendix A) helps us
+propose a reasonable subset of all values.  Trying to map all errno
+mnemonics, indeed, seems foolish, pointless, and would pollute the root
+namespace.
+
+Furthermore, in a couple of cases, different errno values could raise
+the same exception subclass.  For example, EAGAIN, EALREADY, EWOULDBLOCK
+and EINPROGRESS are all used to signal that an operation on a non-blocking
+socket would block (and therefore needs trying again later).  They could
+therefore all raise an identical subclass and let the user examine the
+``errno`` attribute if (s)he so desires (see below "exception
+attributes").
+
+Prerequisite
+------------
+
+Step 1 is a loose prerequisite for this.
+
+Prerequisite, because some errnos can currently be attached to different
+exception classes: for example, EBADF can be attached to both OSError and
+IOError, depending on the context.  If we don't want to break *useful
+compatibility*, we can't make an ``except OSError`` (or IOError) fail to
+match an exception where it would succeed today.
+
+Loose, because we could decide for a partial resolution of step 2
+if existing exception classes are not coalesced: for example, EBADF could
+raise a hypothetical BadFileDescriptor where an IOError was previously
+raised, but continue to raise OSError otherwise.
+
+The dependency on step 1 could be totally removed if the new subclasses
+used multiple inheritance to match with all of the existing superclasses
+(or, at least, OSError and IOError, which are arguable the most prevalent
+ones).  It would, however, make the hierarchy more complicated and
+therefore harder to grasp for the user.
+
+New exception classes
+---------------------
+
+The following tentative list of subclasses, along with a description and
+the list of errnos mapped to them, is submitted to discussion:
+
+* ``FileAlreadyExists``: trying to create a file or directory which already
+  exists (EEXIST)
+
+* ``FileNotFound``: for all circumstances where a file and directory is
+  requested but doesn't exist (ENOENT)
+
+* ``IsADirectory``: file-level operation (open(), os.remove()...) requested
+  on a directory (EISDIR)
+
+* ``NotADirectory``: directory-level operation requested on something else
+  (ENOTDIR)
+
+* ``PermissionDenied``: trying to run an operation without the adequate access
+  rights - for example filesystem permissions (EACCESS, optionally EPERM)
+
+* ``BlockingIOError``: an operation would block on an object (e.g. socket) set
+  for non-blocking operation (EAGAIN, EALREADY, EWOULDBLOCK, EINPROGRESS);
+  this is the existing ``io.BlockingIOError`` with an extended role
+
+* ``BadFileDescriptor``: operation on an invalid file descriptor (EBADF);
+  the default error message could point out that most causes are that
+  an existing file descriptor has been closed
+
+* ``ConnectionAborted``: connection attempt aborted by peer (ECONNABORTED)
+
+* ``ConnectionRefused``: connection reset by peer (ECONNREFUSED)
+
+* ``ConnectionReset``: connection reset by peer (ECONNRESET)
+
+* ``TimeoutError``: connection timed out (ECONNTIMEOUT); this could be re-cast
+  as a generic timeout exception, useful for other types of timeout (for
+  example in Lock.acquire())
+
+This list assumes step 1 is accepted in full; the exception classes
+described above would all derive from the now unified exception type
+OSError.  It will need reworking if a partial version of step 1 is accepted
+instead (again, see appendix A for the current distribution of errnos
+and exception types).
+
+
+Exception attributes
+--------------------
+
+In order to preserve *useful compatibility*, these subclasses should still
+set adequate values for the various exception attributes defined on the
+superclass (for example ``errno``, ``filename``, and optionally
+``winerror``).
+
+Implementation
+--------------
+
+Since it is proposed that the subclasses are raised based purely on the
+value of ``errno``, little or no changes should be required in extension
+modules (either standard or third-party).  As long as they use the
+``PyErr_SetFromErrno()`` family of functions (or the
+``PyErr_SetFromWindowsErr()`` family of functions under Windows), they
+should automatically benefit from the new, finer-grained exception classes.
+
+Library modules written in Python, though, will have to be adapted where
+they currently use the following idiom (seen in ``Lib/tempfile.py``)::
+
+    raise IOError(_errno.EEXIST, "No usable temporary file name found")
+
+Fortunately, such Python code is quite rare since raising OSError or IOError
+with an errno value normally happens when interfacing with system calls,
+which is usually done in C extensions.
+
+If there is popular demand, the subroutine choosing an exception type based
+on the errno value could be exposed for use in pure Python.
+
+
+Possible objections
+===================
+
+Namespace pollution
+-------------------
+
+Making the exception hierarchy finer-grained makes the root (or builtins)
+namespace larger.  This is to be moderated, however, as:
+
+* only a handful of additional classes are proposed; 
+
+* while standard exception types live in the root namespace, they are
+  visually distinguished by the fact that they use the CamelCase convention,
+  while almost all other builtins use lowercase naming (except True, False,
+  None, Ellipsis and NotImplemented)
+
+An alternative would be to provide a separate module containing the
+finer-grained exceptions, but that would defeat the purpose of
+encouraging careful code over careless code, since the user would first
+have to import the new module instead of using names already accessible.
+
+
+Earlier discussion
+==================
+
+While this is the first time such as formal proposal is made, the idea
+has received informal support in the past [1]_; both the introduction
+of finer-grained exception classes and the coalescing of OSError and
+IOError.
+
+The removal of WindowsError alone has been discussed and rejected
+as part of another PEP [2]_, but there seemed to be a consensus that the
+distinction with OSError wasn't meaningful.  This supports at least its
+aliasing with OSError.
+
+
+Moratorium
+==========
+
+The moratorium in effect on language builtins means this PEP has little
+chance to be accepted for Python 3.2.
+
+
+Possible alternative
+====================
+
+Pattern matching
+----------------
+
+Another possibility would be to introduce an advanced pattern matching
+syntax when catching exceptions.  For example::
+
+    try:
+        os.remove(filename)
+    except OSError as e if e.errno == errno.ENOENT:
+        pass
+
+Several problems with this proposal:
+
+* it introduces new syntax, which is perceived by the author to be a heavier
+  change compared to reworking the exception hierarchy
+* it doesn't decrease typing effort significantly
+* it doesn't relieve the programmer from the burden of having to remember
+  errno mnemonics
+
+
+Exceptions ignored by this PEP
+==============================
+
+This PEP ignores ``EOFError``, which signals a truncated input stream in
+various protocol and file format implementations (for example ``GzipFile``).
+``EOFError`` is not OS- or IO-related, it is a logical error raised at
+a higher level.
+
+This PEP also ignores ``SSLError``, which is raised by the ``ssl`` module
+in order to propagate errors signalled by the ``OpenSSL`` library.  Ideally,
+``SSLError`` would benefit from a similar but separate treatment since it
+defines its own constants for error types (``ssl.SSL_ERROR_WANT_READ``,
+etc.).
+
+
+Appendix A: Survey of common errnos
+===================================
+
+This is a quick recension of the various errno mnemonics checked for in
+the standard library and its tests, as part of ``except`` clauses.
+
+Common errnos with OSError
+--------------------------
+
+* ``EBADF``: bad file descriptor (usually means the file descriptor was
+  closed)
+
+* ``EEXIST``: file or directory exists
+
+* ``EINTR``: interrupted function call
+
+* ``EISDIR``: is a directory
+
+* ``ENOTDIR``: not a directory
+
+* ``ENOENT``: no such file or directory
+
+* ``EOPNOTSUPP``: operation not supported on socket
+  (possible confusion with the existing io.UnsupportedOperation)
+
+* ``EPERM``: operation not permitted (when using e.g. os.setuid())
+
+Common errnos with IOError
+--------------------------
+
+* ``EACCES``: permission denied (for filesystem operations)
+
+* ``EBADF``: bad file descriptor (with select.epoll); read operation on a
+  write-only GzipFile, or vice-versa
+
+* ``EBUSY``: device or resource busy
+
+* ``EISDIR``: is a directory (when trying to open())
+
+* ``ENODEV``: no such device
+
+* ``ENOENT``: no such file or directory (when trying to open())
+
+* ``ETIMEDOUT``: connection timed out
+
+Common errnos with socket.error
+-------------------------------
+
+All these errors may also be associated with a plain IOError, for example
+when calling read() on a socket's file descriptor.
+
+* ``EAGAIN``: resource temporarily unavailable (during a non-blocking socket
+  call except connect())
+
+* ``EALREADY``: connection already in progress (during a non-blocking
+  connect())
+
+* ``EINPROGRESS``: operation in progress (during a non-blocking connect())
+
+* ``EINTR``: interrupted function call
+
+* ``EISCONN``: the socket is connected
+
+* ``ECONNABORTED``: connection aborted by peer (during an accept() call)
+
+* ``ECONNREFUSED``: connection refused by peer
+
+* ``ECONNRESET``: connection reset by peer
+
+* ``ENOTCONN``: socket not connected
+
+* ``ESHUTDOWN``: cannot send after transport endpoint shutdown
+
+* ``EWOULDBLOCK``: same reasons as ``EAGAIN``
+
+Common errnos with select.error
+-------------------------------
+
+* ``EINTR``: interrupted function call
+
+
+Appendix B: Survey of raised OS and IO errors
+=============================================
+
+Interpreter core
+----------------
+
+Handling of PYTHONSTARTUP raises IOError (but the error gets discarded)::
+
+    $ PYTHONSTARTUP=foox ./python
+    Python 3.2a0 (py3k:82920M, Jul 16 2010, 22:53:23) 
+    [GCC 4.4.3] on linux2
+    Type "help", "copyright", "credits" or "license" for more information.
+    Could not open PYTHONSTARTUP
+    IOError: [Errno 2] No such file or directory: 'foox'
+
+``PyObject_Print()`` raises IOError when ferror() signals an error on the
+`FILE *` parameter (which, in the source tree, is always either stdout or
+stderr).
+
+Standard library
+----------------
+
+bz2
+'''
+
+Raises IOError throughout (OSError is unused)::
+
+    >>> bz2.BZ2File("foox", "rb")
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: [Errno 2] No such file or directory
+    >>> bz2.BZ2File("LICENSE", "rb").read()
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: invalid data stream
+    >>> bz2.BZ2File("/tmp/zzz.bz2", "wb").read()
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: file is not ready for reading
+
+curses
+''''''
+
+Not examined.
+
+dbm.gnu, dbm.ndbm
+'''''''''''''''''
+
+_dbm.error and _gdbm.error inherit from IOError::
+
+    >>> dbm.gnu.open("foox")
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    _gdbm.error: [Errno 2] No such file or directory
+
+fcntl
+'''''
+
+Raises IOError throughout (OSError is unused).
+
+imp module
+''''''''''
+
+Raises IOError for bad file descriptors::
+
+    >>> imp.load_source("foo", "foo", 123)
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: [Errno 9] Bad file descriptor
+
+io module
+'''''''''
+
+Raises IOError when trying to open a directory under Unix::
+
+    >>> open("Python/", "r")
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: [Errno 21] Is a directory: 'Python/'
+
+Raises IOError for unsupported operations::
+
+    >>> open("LICENSE").write("bar")
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: not writable
+    >>> io.StringIO().fileno()
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    io.UnsupportedOperation: fileno
+    >>> open("LICENSE").seek(1, 1)
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: can't do nonzero cur-relative seeks
+
+(io.UnsupportedOperation inherits from IOError)
+
+Raises either IOError or TypeError when the inferior I/O layer misbehaves
+(i.e. violates the API it is expected to implement).
+
+Raises IOError when the underlying OS resource becomes invalid::
+
+    >>> f = open("LICENSE")
+    >>> os.close(f.fileno())
+    >>> f.read()
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: [Errno 9] Bad file descriptor
+
+...or for implementation-specific optimizations::
+
+    >>> f = open("LICENSE")
+    >>> next(f)
+    'A. HISTORY OF THE SOFTWARE\n'
+    >>> f.tell()
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: telling position disabled by next() call
+
+Raises BlockingIOError (inherited from IOError) when a call on a non-blocking
+object would block.
+
+multiprocessing
+'''''''''''''''
+
+Not examined.
+
+ossaudiodev
+'''''''''''
+
+Raises IOError throughout (OSError is unused)::
+
+    >>> ossaudiodev.open("foo", "r")
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: [Errno 2] No such file or directory: 'foo'
+
+readline
+''''''''
+
+Raises IOError in various file-handling functions::
+
+    >>> readline.read_history_file("foo")
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: [Errno 2] No such file or directory
+    >>> readline.read_init_file("foo")
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: [Errno 2] No such file or directory
+    >>> readline.write_history_file("/dev/nonexistent")
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    IOError: [Errno 13] Permission denied
+
+select
+''''''
+
+select() and poll objects raise select.error, which doesn't inherit from
+anything (but poll.modify() which raises IOError).
+epoll objects raise IOError.
+kqueue objects raise both OSError and IOError.
+
+signal
+''''''
+
+signal.ItimerError inherits from IOError.
+
+socket
+''''''
+
+socket.error inherits from IOError.
+
+time
+''''
+
+Raises IOError for internal errors in time.time() and time.sleep().
+
+zipimport
+'''''''''
+
+zipimporter.get_data() can raise IOError.
+
+
+References
+==========
+
+.. [1] "IO module precisions and exception hierarchy"
+   http://mail.python.org/pipermail/python-dev/2009-September/092130.html
+
+.. [2] Discussion of "Removing WindowsError" in PEP 348
+   http://www.python.org/dev/peps/pep-0348/#removing-windowserror
+
+Copyright
+=========
+
+This document has been placed in the public domain.
+
+
+
+..
+   Local Variables:
+   mode: indented-text
+   indent-tabs-mode: nil
+   sentence-end-double-space: t
+   fill-column: 70
+   coding: utf-8
+   End:

From python-checkins at python.org  Wed Jul 21 19:43:43 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Wed, 21 Jul 2010 19:43:43 +0200 (CEST)
Subject: [Python-checkins] r83035 - in python/branches/py3k: Lib/trace.py
	Misc/NEWS Misc/maintainers.rst
Message-ID: <20100721174343.20B1CF303@mail.python.org>

Author: alexander.belopolsky
Date: Wed Jul 21 19:43:42 2010
New Revision: 83035

Log:
Issue #9323: Fixed a bug in trace.py that resulted in loosing the name
of the script being traced.  Patch by Eli Bendersky.


Modified:
   python/branches/py3k/Lib/trace.py
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/Misc/maintainers.rst

Modified: python/branches/py3k/Lib/trace.py
==============================================================================
--- python/branches/py3k/Lib/trace.py	(original)
+++ python/branches/py3k/Lib/trace.py	Wed Jul 21 19:43:42 2010
@@ -797,12 +797,9 @@
                   ignoredirs=ignore_dirs, infile=counts_file,
                   outfile=counts_file, timing=timing)
         try:
-            fp = open(progname)
-            try:
-                script = fp.read()
-            finally:
-                fp.close()
-            t.run('exec(%r)' % (script,))
+            with open(progname) as fp:
+                code = compile(fp.read(), progname, 'exec')
+            t.run(code)
         except IOError as err:
             _err_exit("Cannot run file %r because: %s" % (sys.argv[0], err))
         except SystemExit:

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Wed Jul 21 19:43:42 2010
@@ -473,6 +473,9 @@
 Library
 -------
 
+- Issue #9323: Fixed a bug in trace.py that resulted in loosing the
+  name of the script being traced.  Patch by Eli Bendersky.
+
 - Issue #9282: Fixed --listfuncs option of trace.py.  Thanks Eli
   Bendersky for the patch.
 

Modified: python/branches/py3k/Misc/maintainers.rst
==============================================================================
--- python/branches/py3k/Misc/maintainers.rst	(original)
+++ python/branches/py3k/Misc/maintainers.rst	Wed Jul 21 19:43:42 2010
@@ -212,7 +212,7 @@
 tkinter             gpolo
 token               georg.brandl
 tokenize
-trace
+trace               alexander.belopolsky
 traceback           georg.brandl
 tty
 turtle              gregorlingl

From python-checkins at python.org  Wed Jul 21 19:44:30 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 21 Jul 2010 19:44:30 +0200 (CEST)
Subject: [Python-checkins] r83036 - peps/trunk/pep-3151.txt
Message-ID: <20100721174430.646CBEF29@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 21 19:44:30 2010
New Revision: 83036

Log:
Wording and markup changes.



Modified:
   peps/trunk/pep-3151.txt

Modified: peps/trunk/pep-3151.txt
==============================================================================
--- peps/trunk/pep-3151.txt	(original)
+++ peps/trunk/pep-3151.txt	Wed Jul 21 19:44:30 2010
@@ -87,10 +87,11 @@
 
 As for WindowsError, it seems to be a pointless distinction.  First, it
 only exists on Windows systems, which requires tedious compatibility code
-in cross-platform applications.  Second, it inherits from OSError and
-is raised for similar errors as OSError is raised for on other systems.
-Third, the user wanting access to low-level exception specifics has to
-examine the ``errno`` or ``winerror`` attribute anyway.
+in cross-platform applications (such code can be found in ``Lib/shutil.py``).
+Second, it inherits from OSError and is raised for similar errors as OSError
+is raised for on other systems. Third, the user wanting access to low-level
+exception specifics has to examine the ``errno`` or ``winerror`` attribute
+anyway.
 
 
 Lack of fine-grained exceptions
@@ -104,9 +105,9 @@
     if os.path.exists(filename):
         os.remove(filename)
 
-If a file named as `filename` is created by another thread or process
-between the calls to `os.path.exists` and `os.remove`, it won't be deleted.
-This can produce bugs in the application, or even security issues.
+If a file named as ``filename`` is created by another thread or process
+between the calls to ``os.path.exists`` and ``os.remove``, it won't be
+deleted.  This can produce bugs in the application, or even security issues.
 
 Therefore, the solution is to try to remove the file, and ignore the error
 if the file doesn't exist (an idiom known as Easier to Ask Forgiveness
@@ -128,10 +129,10 @@
             raise
 
 This is a lot more to type, and also forces the user to remember the various
-cryptic mnemonics from the errno module.  It imposes an additional cognitive
-burden and gets tiresome rather quickly.  Consequently, many programmers
-will instead write the following code, which silences exceptions too
-broadly::
+cryptic mnemonics from the ``errno`` module.  It imposes an additional
+cognitive burden and gets tiresome rather quickly.  Consequently, many
+programmers will instead write the following code, which silences exceptions
+too broadly::
 
     try:
         os.remove(filename)
@@ -151,7 +152,7 @@
         pass
 
 
-Compatibility concerns
+Compatibility strategy
 ======================
 
 Reworking the exception hierarchy will obviously change the exact semantics
@@ -213,6 +214,8 @@
 
 It is not yet decided whether the old names will be deprecated (then removed)
 or all alternative names will continue living in the root namespace.
+Deprecation of names from the root namespace presents some implementation
+challenges, especially where performance is important.
 
 
 Step 2: define additional subclasses
@@ -575,7 +578,8 @@
       File "<stdin>", line 1, in <module>
     IOError: [Errno 21] Is a directory: 'Python/'
 
-Raises IOError for unsupported operations::
+Raises IOError or io.UnsupportedOperation (which inherits from the former)
+for unsupported operations::
 
     >>> open("LICENSE").write("bar")
     Traceback (most recent call last):
@@ -590,8 +594,6 @@
       File "<stdin>", line 1, in <module>
     IOError: can't do nonzero cur-relative seeks
 
-(io.UnsupportedOperation inherits from IOError)
-
 Raises either IOError or TypeError when the inferior I/O layer misbehaves
 (i.e. violates the API it is expected to implement).
 
@@ -653,10 +655,10 @@
 select
 ''''''
 
-select() and poll objects raise select.error, which doesn't inherit from
-anything (but poll.modify() which raises IOError).
-epoll objects raise IOError.
-kqueue objects raise both OSError and IOError.
+* select() and poll objects raise select.error, which doesn't inherit from
+  anything (but poll.modify() raises IOError);
+* epoll objects raise IOError;
+* kqueue objects raise both OSError and IOError.
 
 signal
 ''''''

From python-checkins at python.org  Wed Jul 21 19:50:34 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Wed, 21 Jul 2010 19:50:34 +0200 (CEST)
Subject: [Python-checkins] r83037 - in python/branches/release31-maint:
	Lib/trace.py Misc/NEWS Misc/maintainers.rst
Message-ID: <20100721175034.6FFB8E7F9@mail.python.org>

Author: alexander.belopolsky
Date: Wed Jul 21 19:50:34 2010
New Revision: 83037

Log:
Merged revisions 83035 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83035 | alexander.belopolsky | 2010-07-21 13:43:42 -0400 (Wed, 21 Jul 2010) | 3 lines
  
  Issue #9323: Fixed a bug in trace.py that resulted in loosing the name
  of the script being traced.  Patch by Eli Bendersky.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/trace.py
   python/branches/release31-maint/Misc/NEWS
   python/branches/release31-maint/Misc/maintainers.rst

Modified: python/branches/release31-maint/Lib/trace.py
==============================================================================
--- python/branches/release31-maint/Lib/trace.py	(original)
+++ python/branches/release31-maint/Lib/trace.py	Wed Jul 21 19:50:34 2010
@@ -797,12 +797,9 @@
                   ignoredirs=ignore_dirs, infile=counts_file,
                   outfile=counts_file, timing=timing)
         try:
-            fp = open(progname)
-            try:
-                script = fp.read()
-            finally:
-                fp.close()
-            t.run('exec(%r)' % (script,))
+            with open(progname) as fp:
+                code = compile(fp.read(), progname, 'exec')
+            t.run(code)
         except IOError as err:
             _err_exit("Cannot run file %r because: %s" % (sys.argv[0], err))
         except SystemExit:

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Wed Jul 21 19:50:34 2010
@@ -78,6 +78,9 @@
 Library
 -------
 
+- Issue #9323: Fixed a bug in trace.py that resulted in loosing the
+  name of the script being traced.  Patch by Eli Bendersky.
+
 - Issue #9282: Fixed --listfuncs option of trace.py.  Thanks Eli
   Bendersky for the patch.
 

Modified: python/branches/release31-maint/Misc/maintainers.rst
==============================================================================
--- python/branches/release31-maint/Misc/maintainers.rst	(original)
+++ python/branches/release31-maint/Misc/maintainers.rst	Wed Jul 21 19:50:34 2010
@@ -212,7 +212,7 @@
 tkinter             gpolo
 token               georg.brandl
 tokenize
-trace
+trace               alexander.belopolsky
 traceback           georg.brandl
 tty
 turtle              gregorlingl

From python-checkins at python.org  Wed Jul 21 19:54:21 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 21 Jul 2010 19:54:21 +0200 (CEST)
Subject: [Python-checkins] r83038 - peps/trunk/pep-3151.txt
Message-ID: <20100721175421.56735F29A@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 21 19:54:21 2010
New Revision: 83038

Log:
Add mention of the os module.



Modified:
   peps/trunk/pep-3151.txt

Modified: peps/trunk/pep-3151.txt
==============================================================================
--- peps/trunk/pep-3151.txt	(original)
+++ peps/trunk/pep-3151.txt	Wed Jul 21 19:54:21 2010
@@ -624,6 +624,12 @@
 
 Not examined.
 
+os / posix module
+'''''''''''''''''
+
+The ``os`` (or ``posix``) module raises OSError throughout, except under
+Windows where WindosError can be raised instead.
+
 ossaudiodev
 '''''''''''
 

From python-checkins at python.org  Wed Jul 21 20:07:45 2010
From: python-checkins at python.org (andrew.kuchling)
Date: Wed, 21 Jul 2010 20:07:45 +0200 (CEST)
Subject: [Python-checkins] r83039 - peps/trunk/pep-0327.txt
Message-ID: <20100721180745.A0B49EFAF@mail.python.org>

Author: andrew.kuchling
Date: Wed Jul 21 20:07:45 2010
New Revision: 83039

Log:
Update links to decimal specification

Modified:
   peps/trunk/pep-0327.txt

Modified: peps/trunk/pep-0327.txt
==============================================================================
--- peps/trunk/pep-0327.txt	(original)
+++ peps/trunk/pep-0327.txt	Wed Jul 21 20:07:45 2010
@@ -156,7 +156,7 @@
     (your memory permitting) should work just as well.  Most of the
     expense of programming simplicity can be hidden from application
     programs and placed in a suitable decimal arithmetic type.  As per
-    http://www2.hursley.ibm.com/decimal/, *a single data type can be
+    http://speleotrove.com/decimal/, *a single data type can be
     used for integer, fixed-point, and floating-point decimal
     arithmetic* -- and for money arithmetic which doesn't drive the
     application programmer crazy.
@@ -1217,8 +1217,8 @@
    http://www.rexxla.org/Standards/ansi.html
 
 .. [2] General Decimal Arithmetic specification (Cowlishaw):
-   http://www2.hursley.ibm.com/decimal/decarith.html (related
-   documents and links at http://www2.hursley.ibm.com/decimal/)
+   http://speleotrove.com/decimal/decarith.html (related
+   documents and links at http://speleotrove.com/decimal/)
 
 .. [3] ANSI/IEEE standard 854-1987 (Radix-Independent Floating-Point
    Arithmetic):

From python-checkins at python.org  Wed Jul 21 20:49:25 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 21 Jul 2010 20:49:25 +0200 (CEST)
Subject: [Python-checkins] r83040 - peps/trunk/pep-3151.txt
Message-ID: <20100721184925.521E5F11D@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 21 20:49:25 2010
New Revision: 83040

Log:
Add items to survey.



Modified:
   peps/trunk/pep-3151.txt

Modified: peps/trunk/pep-3151.txt
==============================================================================
--- peps/trunk/pep-3151.txt	(original)
+++ peps/trunk/pep-3151.txt	Wed Jul 21 20:49:25 2010
@@ -49,6 +49,7 @@
             +-- socket.error
         +-- OSError
             +-- WindowsError
+        +-- mmap.error
     +-- select.error
 
 While some of these distinctions can be explained by implementation
@@ -199,6 +200,7 @@
 changes are listed hereafter:
 
 * alias both socket.error and select.error to IOError
+* alias mmap.error to OSError
 * alias IOError to OSError
 * alias WindowsError to OSError
 
@@ -517,6 +519,9 @@
 `FILE *` parameter (which, in the source tree, is always either stdout or
 stderr).
 
+Unicode encoding and decoding using the ``mbcs`` encoding can raise
+WindowsError for some error conditions.
+
 Standard library
 ----------------
 
@@ -616,16 +621,54 @@
       File "<stdin>", line 1, in <module>
     IOError: telling position disabled by next() call
 
-Raises BlockingIOError (inherited from IOError) when a call on a non-blocking
+Raises BlockingIOError (inheriting from IOError) when a call on a non-blocking
 object would block.
 
+mmap
+''''
+
+Undex Unix, raises its own ``mmap.error`` (inheriting from EnvironmentError)
+throughout::
+
+    >>> mmap.mmap(123, 10)
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    mmap.error: [Errno 9] Bad file descriptor
+    >>> mmap.mmap(os.open("/tmp", os.O_RDONLY), 10)
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    mmap.error: [Errno 13] Permission denied
+
+Under Windows, however, it mostly raises WindowsError (the source code
+also shows a few occurrences of ``mmap.error``)::
+
+    >>> fd = os.open("LICENSE", os.O_RDONLY)
+    >>> m = mmap.mmap(fd, 16384)
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    WindowsError: [Error 5] Acc?s refus?
+    >>> sys.last_value.errno
+    13
+    >>> errno.errorcode[13]
+    'EACCES'
+
+    >>> m = mmap.mmap(-1, 4096)
+    >>> m.resize(16384)
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    WindowsError: [Error 87] Param?tre incorrect
+    >>> sys.last_value.errno
+    22
+    >>> errno.errorcode[22]
+    'EINVAL'
+
 multiprocessing
 '''''''''''''''
 
 Not examined.
 
-os / posix module
-'''''''''''''''''
+os / posix
+''''''''''
 
 The ``os`` (or ``posix``) module raises OSError throughout, except under
 Windows where WindosError can be raised instead.
@@ -676,6 +719,12 @@
 
 socket.error inherits from IOError.
 
+sys
+'''
+
+``sys.getwindowsversion()`` raises WindowsError with a bogus error number
+if the ``GetVersionEx()`` call fails.
+
 time
 ''''
 

From python-checkins at python.org  Wed Jul 21 20:59:23 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 21 Jul 2010 20:59:23 +0200 (CEST)
Subject: [Python-checkins] r83041 - peps/trunk/pep-3151.txt
Message-ID: <20100721185923.EEC46D879@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 21 20:59:23 2010
New Revision: 83041

Log:
Markup



Modified:
   peps/trunk/pep-3151.txt

Modified: peps/trunk/pep-3151.txt
==============================================================================
--- peps/trunk/pep-3151.txt	(original)
+++ peps/trunk/pep-3151.txt	Wed Jul 21 20:59:23 2010
@@ -704,7 +704,7 @@
 select
 ''''''
 
-* select() and poll objects raise select.error, which doesn't inherit from
+* select() and poll objects raise ``select.error``, which doesn't inherit from
   anything (but poll.modify() raises IOError);
 * epoll objects raise IOError;
 * kqueue objects raise both OSError and IOError.
@@ -712,12 +712,12 @@
 signal
 ''''''
 
-signal.ItimerError inherits from IOError.
+``signal.ItimerError`` inherits from IOError.
 
 socket
 ''''''
 
-socket.error inherits from IOError.
+``socket.error`` inherits from IOError.
 
 sys
 '''

From python-checkins at python.org  Wed Jul 21 21:16:49 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 21 Jul 2010 21:16:49 +0200 (CEST)
Subject: [Python-checkins] r83042 - peps/trunk/pep-3151.txt
Message-ID: <20100721191649.57D90ECFB@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 21 21:16:49 2010
New Revision: 83042

Log:
Markup again



Modified:
   peps/trunk/pep-3151.txt

Modified: peps/trunk/pep-3151.txt
==============================================================================
--- peps/trunk/pep-3151.txt	(original)
+++ peps/trunk/pep-3151.txt	Wed Jul 21 21:16:49 2010
@@ -68,8 +68,8 @@
 
 The same error condition (a non-existing file) gets cast as two different
 exceptions depending on which library function was called.  The reason
-for this is that the `os` module exclusively raises OSError (or its
-subclass WindowsError) while the `io` module mostly raises IOError.
+for this is that the ``os`` module exclusively raises OSError (or its
+subclass WindowsError) while the ``io`` module mostly raises IOError.
 However, the user is interested in the nature of the error, not in which
 part of the interpreter it comes from (since the latter is obvious from
 reading the traceback message or application source code).

From python-checkins at python.org  Wed Jul 21 21:41:06 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 21 Jul 2010 21:41:06 +0200 (CEST)
Subject: [Python-checkins] r83043 - peps/trunk/pep-3151.txt
Message-ID: <20100721194106.27729D981@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 21 21:41:06 2010
New Revision: 83043

Log:
Stress appendices



Modified:
   peps/trunk/pep-3151.txt

Modified: peps/trunk/pep-3151.txt
==============================================================================
--- peps/trunk/pep-3151.txt	(original)
+++ peps/trunk/pep-3151.txt	Wed Jul 21 21:41:06 2010
@@ -94,6 +94,10 @@
 exception specifics has to examine the ``errno`` or ``winerror`` attribute
 anyway.
 
+.. note::
+    **Appendix B** surveys the use of the various exception types accross
+    the interpreter and the standard library.
+
 
 Lack of fine-grained exceptions
 -------------------------------
@@ -226,7 +230,7 @@
 The second step of the resolution is to extend the hierarchy by defining
 subclasses which will be raised, rather than their parent, for specific
 errno values.  Which errno values is subject to discussion, but a survey
-of existing exception matching practices (see Appendix A) helps us
+of existing exception matching practices (see **Appendix A**) helps us
 propose a reasonable subset of all values.  Trying to map all errno
 mnemonics, indeed, seems foolish, pointless, and would pollute the root
 namespace.

From python-checkins at python.org  Thu Jul 22 00:11:04 2010
From: python-checkins at python.org (nick.coghlan)
Date: Thu, 22 Jul 2010 00:11:04 +0200 (CEST)
Subject: [Python-checkins] r83044 - peps/trunk/pep-3150.txt
Message-ID: <20100721221104.8B1D8D765@mail.python.org>

Author: nick.coghlan
Date: Thu Jul 22 00:11:04 2010
New Revision: 83044

Log:
Fix some minor typos, add notes on topics not yet covered and point out that the most likely outcome for this PEP is eventual rejection

Modified:
   peps/trunk/pep-3150.txt

Modified: peps/trunk/pep-3150.txt
==============================================================================
--- peps/trunk/pep-3150.txt	(original)
+++ peps/trunk/pep-3150.txt	Thu Jul 22 00:11:04 2010
@@ -36,7 +36,35 @@
 (PEP 3003) is officially lifted by Guido. Even after that, it will
 require input from at least the four major Python implementations
 (CPython, PyPy, Jython, IronPython) on the feasibility of implementing
-the proposed semantics to get it moving again.
+the proposed semantics to get it moving again. Input from related
+projects with a vested interest in Python's syntax (e.g. Cython) will
+also be valuable.
+
+That said, if a decision on acceptance or rejection had to be made
+immediately, rejection would be far more likely. Unlike the previous
+major syntax addition to Python (PEP 343's ``with`` statement), this
+PEP has no "killer application" of code that is clearly and obviously
+improved through the use of the new syntax. The ``with`` statement (in
+conjunction with the generator enhancements in PEP 342) allowed
+exception handling to be factored out into context managers in a way
+that had never before been possible. Code using the new statement was
+not only easier to read, but much easier to write correctly in the
+first place.
+
+In the case of this PEP. however, the "Two Ways to Do It" objection is a
+strong one. While the ability to break out subexpresions of a statement
+without having to worry about name clashes with the rest of a
+function or script and without distracting from the operation that is
+the ultimate aim of the statement is potentially nice to have as a
+language feature, it doesn't really provide significant expressive power
+over and above what is already possible by assigning subexpressions to
+ordinary local variables before the statement of interest. In particular,
+explaining to new Python programmers when it is best to use a ``given``
+clause and when to use normal local variables is likely to be challenging
+and an unnecessary distraction.
+
+"It might be kinda, sorta, nice to have, sometimes" really isn't a strong
+argument for a new syntactic construct (particularly one this complicated).
 
 Proposal
 ========
@@ -174,7 +202,7 @@
 New::
 
    expr_stmt: testlist_star_expr (augassign (yield_expr|testlist) |
-                ('=' (yield_expr|testlist_star_expr))*) [where_clause]
+                ('=' (yield_expr|testlist_star_expr))*) [given_clause]
    del_stmt: 'del' exprlist [given_clause]
    return_stmt: 'return' [testlist] [given_clause]
    yield_stmt: yield_expr [given_clause]
@@ -190,7 +218,7 @@
 an ambiguity in the grammar. It is applied only to the specific elements
 listed so that nonsense like the following is disallowed::
 
-   pass where:
+   pass given:
        a = b = 1
 
 However, even this is inadequate, as it creates problems for the definition
@@ -230,11 +258,11 @@
 
 * Two Ways To Do It: a lot of code may now be written with values
   defined either before the expression where they are used or
-  afterwards in a ``where`` clause, creating two ways to do it,
+  afterwards in a ``given`` clause, creating two ways to do it,
   without an obvious way of choosing between them.
 
-* Out of Order Execution: the ``where`` clause makes execution
-  jump around a little strangely, as the body of the ``where``
+* Out of Order Execution: the ``given`` clause makes execution
+  jump around a little strangely, as the body of the ``given``
   clause is executed before the simple statement in the clause
   header. The closest any other part of Python comes to this
   is the out of order evaluation in list comprehensions,
@@ -348,13 +376,24 @@
 this idea has never been tried.
 
 
-Reference implementation
+Reference Implementation
 ========================
 
 None as yet. If you want a crash course in Python namespace
 semantics and code compilation, feel free to try ;)
 
 
+TO-DO
+=====
+
+* Mention two-suite in-order variants (and explain why they're even more
+  pointless than the specific idea in the PEP)
+* Mention PEP 359 and possible uses for locals() in the ``given`` clause
+* Describe the expected semantics of ``break``, ``continue``, ``return``
+  and ``yield`` in a ``given`` clause (i.e. syntax errors at the clause
+  level, but allowed inside the appropriate compound statements)
+
+
 References
 ==========
 

From solipsis at pitrou.net  Thu Jul 22 01:23:45 2010
From: solipsis at pitrou.net (solipsis at pitrou.net)
Date: Thu, 22 Jul 2010 01:23:45 +0200 (CEST)
Subject: [Python-checkins] Daily py3k reference leaks (r83035): sum=0
Message-ID: <20100721232345.5972B1771F@ns6635.ovh.net>

py3k results for svn r83035 (hg cset ac84ab89f963)
--------------------------------------------------



Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/antoine/py3k/refleaks/reflogVXrjX0', '-x']

From python-checkins at python.org  Thu Jul 22 03:47:30 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Thu, 22 Jul 2010 03:47:30 +0200 (CEST)
Subject: [Python-checkins] r83045 - in python/branches/release27-maint:
	Doc/library/urllib.rst Lib/test/test_urllib.py Lib/urllib.py
	Misc/NEWS
Message-ID: <20100722014730.9ED7AD7D7@mail.python.org>

Author: senthil.kumaran
Date: Thu Jul 22 03:47:30 2010
New Revision: 83045

Log:
Reverting the checkin made in revision 82940, as it was adding new parameters to quote function in a bugfix release.
Discussed in issue1712522



Modified:
   python/branches/release27-maint/Doc/library/urllib.rst
   python/branches/release27-maint/Lib/test/test_urllib.py
   python/branches/release27-maint/Lib/urllib.py
   python/branches/release27-maint/Misc/NEWS

Modified: python/branches/release27-maint/Doc/library/urllib.rst
==============================================================================
--- python/branches/release27-maint/Doc/library/urllib.rst	(original)
+++ python/branches/release27-maint/Doc/library/urllib.rst	Thu Jul 22 03:47:30 2010
@@ -202,40 +202,24 @@
 Utility functions
 -----------------
 
-.. function:: quote(string[, safe[, encoding[, errors]]])
+.. function:: quote(string[, safe])
 
    Replace special characters in *string* using the ``%xx`` escape. Letters,
    digits, and the characters ``'_.-'`` are never quoted. By default, this
-   function is intended for quoting the path section of the URL. The optional
+   function is intended for quoting the path section of the URL.The optional
    *safe* parameter specifies additional characters that should not be quoted
    --- its default value is ``'/'``.
 
-   *string* may be either a :class:`str` or a :class:`unicode`.
-
-   The optional *encoding* and *errors* parameters specify how to deal with
-   non-ASCII characters, as accepted by the :meth:`unicode.encode` method.
-   *encoding* defaults to ``'utf-8'``.
-   *errors* defaults to ``'strict'``, meaning unsupported characters raise a
-   :class:`UnicodeEncodeError`.
-   Non-Unicode strings are not encoded by default, and all bytes are allowed.
-
    Example: ``quote('/~connolly/')`` yields ``'/%7econnolly/'``.
 
-   Example: ``quote(u'/El Ni?o/')`` yields ``'/El%20Ni%C3%B1o/'``.
-
-   .. versionchanged:: 2.7.1
-      Added *encoding* and *errors* parameters.
 
-
-.. function:: quote_plus(string[, safe[, encoding[, errors]]])
+.. function:: quote_plus(string[, safe])
 
    Like :func:`quote`, but also replaces spaces by plus signs, as required for
    quoting HTML form values when building up a query string to go into a URL.
    Plus signs in the original string are escaped unless they are included in
    *safe*.  It also does not have *safe* default to ``'/'``.
 
-   Example: ``quote_plus(u'/El Ni?o/')`` yields ``'%2FEl+Ni%C3%B1o%2F'``.
-
 
 .. function:: unquote(string)
 

Modified: python/branches/release27-maint/Lib/test/test_urllib.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_urllib.py	(original)
+++ python/branches/release27-maint/Lib/test/test_urllib.py	Thu Jul 22 03:47:30 2010
@@ -355,38 +355,6 @@
         self.assertEqual(quote_by_default, result,
                          "using quote_plus(): %s != %s" %
                          (quote_by_default, result))
-        # Safe expressed as unicode rather than str
-        result = urllib.quote(quote_by_default, safe=u"<>")
-        self.assertEqual(quote_by_default, result,
-                         "using quote(): %r != %r" % (quote_by_default, result))
-        # "Safe" non-ASCII bytes should still work
-        # (Technically disallowed by the URI standard, but allowed for
-        # backwards compatibility with previous versions of Python)
-        result = urllib.quote(b"a\xfcb", safe=b"\xfc")
-        expect = b"a\xfcb"
-        self.assertEqual(expect, result,
-                         "using quote(): %r != %r" %
-                         (expect, result))
-        # Same as above, but with 'safe' as a unicode rather than str
-        # "Safe" non-ASCII unicode characters should have no effect
-        # (Since URIs are not allowed to have non-ASCII characters)
-        result = urllib.quote(b"a\xfcb", safe=u"\xfc")
-        expect = urllib.quote(b"a\xfcb", safe="")
-        self.assertEqual(expect, result,
-                         "using quote(): %r != %r" %
-                         (expect, result))
-        # Same as above, but quoting a unicode rather than a str
-        result = urllib.quote(u"a\xfcb", encoding="latin-1", safe=b"\xfc")
-        expect = b"a\xfcb"
-        self.assertEqual(expect, result,
-                         "using quote(): %r != %r" %
-                         (expect, result))
-        # Same as above, but with both the quoted value and 'safe' as unicode
-        result = urllib.quote(u"a\xfcb", encoding="latin-1", safe=u"\xfc")
-        expect = urllib.quote(u"a\xfcb", encoding="latin-1", safe="")
-        self.assertEqual(expect, result,
-                         "using quote(): %r != %r" %
-                         (expect, result))
 
     def test_default_quoting(self):
         # Make sure all characters that should be quoted are by default sans
@@ -439,81 +407,6 @@
                          'alpha%2Bbeta+gamma')
         self.assertEqual(urllib.quote_plus('alpha+beta gamma', '+'),
                          'alpha+beta+gamma')
-        # Test with unicode
-        self.assertEqual(urllib.quote_plus(u'alpha+beta gamma'),
-                         'alpha%2Bbeta+gamma')
-        # Test with safe unicode
-        self.assertEqual(urllib.quote_plus('alpha+beta gamma', u'+'),
-                         'alpha+beta+gamma')
-
-    def test_quote_bytes(self):
-        # Non-ASCII bytes should quote directly to percent-encoded values
-        given = b"\xa2\xd8ab\xff"
-        expect = "%A2%D8ab%FF"
-        result = urllib.quote(given)
-        self.assertEqual(expect, result,
-                         "using quote(): %r != %r" % (expect, result))
-        # Encoding argument should raise UnicodeDecodeError on bytes input
-        # with non-ASCII characters (just as with str.encode).
-        self.assertRaises(UnicodeDecodeError, urllib.quote, given,
-                            encoding="latin-1")
-
-    def test_quote_with_unicode(self):
-        # Characters in Latin-1 range, encoded by default in UTF-8
-        given = u"\xa2\xd8ab\xff"
-        expect = "%C2%A2%C3%98ab%C3%BF"
-        result = urllib.quote(given)
-        self.assertEqual(expect, result,
-                         "using quote(): %r != %r" % (expect, result))
-        # Characters in Latin-1 range, encoded by with None (default)
-        result = urllib.quote(given, encoding=None, errors=None)
-        self.assertEqual(expect, result,
-                         "using quote(): %r != %r" % (expect, result))
-        # Characters in Latin-1 range, encoded with Latin-1
-        given = u"\xa2\xd8ab\xff"
-        expect = "%A2%D8ab%FF"
-        result = urllib.quote(given, encoding="latin-1")
-        self.assertEqual(expect, result,
-                         "using quote(): %r != %r" % (expect, result))
-        # Characters in BMP, encoded by default in UTF-8
-        given = u"\u6f22\u5b57"              # "Kanji"
-        expect = "%E6%BC%A2%E5%AD%97"
-        result = urllib.quote(given)
-        self.assertEqual(expect, result,
-                         "using quote(): %r != %r" % (expect, result))
-        # Characters in BMP, encoded with Latin-1
-        given = u"\u6f22\u5b57"
-        self.assertRaises(UnicodeEncodeError, urllib.quote, given,
-                                    encoding="latin-1")
-        # Characters in BMP, encoded with Latin-1, with replace error handling
-        given = u"\u6f22\u5b57"
-        expect = "%3F%3F"                    # "??"
-        result = urllib.quote(given, encoding="latin-1",
-                                    errors="replace")
-        self.assertEqual(expect, result,
-                         "using quote(): %r != %r" % (expect, result))
-        # Characters in BMP, Latin-1, with xmlcharref error handling
-        given = u"\u6f22\u5b57"
-        expect = "%26%2328450%3B%26%2323383%3B"      # "&#28450;&#23383;"
-        result = urllib.quote(given, encoding="latin-1",
-                                    errors="xmlcharrefreplace")
-        self.assertEqual(expect, result,
-                         "using quote(): %r != %r" % (expect, result))
-
-    def test_quote_plus_with_unicode(self):
-        # Encoding (latin-1) test for quote_plus
-        given = u"\xa2\xd8 \xff"
-        expect = "%A2%D8+%FF"
-        result = urllib.quote_plus(given, encoding="latin-1")
-        self.assertEqual(expect, result,
-                         "using quote_plus(): %r != %r" % (expect, result))
-        # Errors test for quote_plus
-        given = u"ab\u6f22\u5b57 cd"
-        expect = "ab%3F%3F+cd"
-        result = urllib.quote_plus(given, encoding="latin-1",
-                                         errors="replace")
-        self.assertEqual(expect, result,
-                         "using quote_plus(): %r != %r" % (expect, result))
 
 class UnquotingTests(unittest.TestCase):
     """Tests for unquote() and unquote_plus()

Modified: python/branches/release27-maint/Lib/urllib.py
==============================================================================
--- python/branches/release27-maint/Lib/urllib.py	(original)
+++ python/branches/release27-maint/Lib/urllib.py	Thu Jul 22 03:47:30 2010
@@ -1193,7 +1193,7 @@
     _safe_map[c] = c if (i < 128 and c in always_safe) else '%{:02X}'.format(i)
 _safe_quoters = {}
 
-def quote(s, safe='/', encoding=None, errors=None):
+def quote(s, safe='/'):
     """quote('abc def') -> 'abc%20def'
 
     Each part of a URL, e.g. the path info, the query, etc., has a
@@ -1213,30 +1213,12 @@
     is reserved, but in typical usage the quote function is being
     called on a path where the existing slash characters are used as
     reserved characters.
-
-    string and safe may be either str or unicode objects.
-
-    The optional encoding and errors parameters specify how to deal with the
-    non-ASCII characters, as accepted by the unicode.encode method.
-    By default, encoding='utf-8' (characters are encoded with UTF-8), and
-    errors='strict' (unsupported characters raise a UnicodeEncodeError).
     """
     # fastpath
     if not s:
         if s is None:
             raise TypeError('None object cannot be quoted')
         return s
-
-    if encoding is not None or isinstance(s, unicode):
-        if encoding is None:
-            encoding = 'utf-8'
-        if errors is None:
-            errors = 'strict'
-        s = s.encode(encoding, errors)
-    if isinstance(safe, unicode):
-        # Normalize 'safe' by converting to str and removing non-ASCII chars
-        safe = safe.encode('ascii', 'ignore')
-
     cachekey = (safe, always_safe)
     try:
         (quoter, safe) = _safe_quoters[cachekey]
@@ -1250,12 +1232,12 @@
         return s
     return ''.join(map(quoter, s))
 
-def quote_plus(s, safe='', encoding=None, errors=None):
+def quote_plus(s, safe=''):
     """Quote the query fragment of a URL; replacing ' ' with '+'"""
     if ' ' in s:
-        s = quote(s, safe + ' ', encoding, errors)
+        s = quote(s, safe + ' ')
         return s.replace(' ', '+')
-    return quote(s, safe, encoding, errors)
+    return quote(s, safe)
 
 def urlencode(query, doseq=0):
     """Encode a sequence of two-element tuples or dictionary into a URL query string.

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Thu Jul 22 03:47:30 2010
@@ -22,10 +22,6 @@
   I/O error occurred.  Now an IOError is raised instead.  Patch by chuck
   (Jan Hosang).
 
-- Issue 1712522: urllib.quote supports Unicode String with encoding and errors
-  parameter. The encoding parameter defaults to utf-8 and errors to strict.
-  Patch by Matt Giuca.
-
 - Issue #7646: The fnmatch pattern cache no longer grows without bound.
 
 - Issue #9136: Fix 'dictionary changed size during iteration'

From python-checkins at python.org  Thu Jul 22 09:40:56 2010
From: python-checkins at python.org (brett.cannon)
Date: Thu, 22 Jul 2010 09:40:56 +0200 (CEST)
Subject: [Python-checkins] r83046 -
	python/branches/py3k/Lib/importlib/test/benchmark.py
Message-ID: <20100722074056.B6DADED8F@mail.python.org>

Author: brett.cannon
Date: Thu Jul 22 09:40:56 2010
New Revision: 83046

Log:
Add importlib benchmarks which try to be "realistic" by importing the decimal
module which is the largest module in the stdlib.


Modified:
   python/branches/py3k/Lib/importlib/test/benchmark.py

Modified: python/branches/py3k/Lib/importlib/test/benchmark.py
==============================================================================
--- python/branches/py3k/Lib/importlib/test/benchmark.py	(original)
+++ python/branches/py3k/Lib/importlib/test/benchmark.py	Thu Jul 22 09:40:56 2010
@@ -6,6 +6,7 @@
 """
 from . import util
 from .source import util as source_util
+import decimal
 import imp
 import importlib
 import os
@@ -58,7 +59,7 @@
 
 
 def source_wo_bytecode(seconds, repeat):
-    """Source w/o bytecode"""
+    """Source w/o bytecode: simple"""
     sys.dont_write_bytecode = True
     try:
         name = '__importlib_test_benchmark__'
@@ -72,8 +73,23 @@
         sys.dont_write_bytecode = False
 
 
+def decimal_wo_bytecode(seconds, repeat):
+    """Source w/o bytecode: decimal"""
+    name = 'decimal'
+    decimal_bytecode = imp.cache_from_source(decimal.__file__)
+    if os.path.exists(decimal_bytecode):
+        os.unlink(decimal_bytecode)
+    sys.dont_write_bytecode = True
+    try:
+        for result in bench(name, lambda: sys.modules.pop(name), repeat=repeat,
+                            seconds=seconds):
+            yield result
+    finally:
+        sys.dont_write_bytecode = False
+
+
 def source_writing_bytecode(seconds, repeat):
-    """Source writing bytecode"""
+    """Source writing bytecode: simple"""
     assert not sys.dont_write_bytecode
     name = '__importlib_test_benchmark__'
     with source_util.create_modules(name) as mapping:
@@ -85,8 +101,19 @@
             yield result
 
 
+def decimal_writing_bytecode(seconds, repeat):
+    """Source writing bytecode: decimal"""
+    assert not sys.dont_write_bytecode
+    name = 'decimal'
+    def cleanup():
+        sys.modules.pop(name)
+        os.unlink(imp.cache_from_source(decimal.__file__))
+    for result in bench(name, cleanup, repeat=repeat, seconds=seconds):
+        yield result
+
+
 def source_using_bytecode(seconds, repeat):
-    """Bytecode w/ source"""
+    """Bytecode w/ source: simple"""
     name = '__importlib_test_benchmark__'
     with source_util.create_modules(name) as mapping:
         py_compile.compile(mapping[name])
@@ -96,16 +123,32 @@
             yield result
 
 
+def decimal_using_bytecode(seconds, repeat):
+    """Bytecode w/ source: decimal"""
+    name = 'decimal'
+    py_compile.compile(decimal.__file__)
+    for result in bench(name, lambda: sys.modules.pop(name), repeat=repeat,
+                        seconds=seconds):
+        yield result
+
+
 def main(import_):
     __builtins__.__import__ = import_
-    benchmarks = (from_cache, builtin_mod, source_using_bytecode,
-                  source_wo_bytecode, source_writing_bytecode,)
-    print("Measuring imports/second\n")
+    benchmarks = (from_cache, builtin_mod,
+                  source_using_bytecode, source_wo_bytecode,
+                  source_writing_bytecode,
+                  decimal_using_bytecode, decimal_writing_bytecode,
+                  decimal_wo_bytecode,)
+    seconds = 1
+    seconds_plural = 's' if seconds > 1 else ''
+    repeat = 3
+    header = "Measuring imports/second over {} second{}, best out of {}\n"
+    print(header.format(seconds, seconds_plural, repeat))
     for benchmark in benchmarks:
         print(benchmark.__doc__, "[", end=' ')
         sys.stdout.flush()
         results = []
-        for result in benchmark(seconds=1, repeat=3):
+        for result in benchmark(seconds=seconds, repeat=repeat):
             results.append(result)
             print(result, end=' ')
             sys.stdout.flush()

From python-checkins at python.org  Thu Jul 22 12:52:13 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Thu, 22 Jul 2010 12:52:13 +0200 (CEST)
Subject: [Python-checkins] r83048 - peps/trunk/pep-3151.txt
Message-ID: <20100722105213.3D0DFEE98C@mail.python.org>

Author: antoine.pitrou
Date: Thu Jul 22 12:52:13 2010
New Revision: 83048

Log:
Use a proper word (Rob Cliffe)



Modified:
   peps/trunk/pep-3151.txt

Modified: peps/trunk/pep-3151.txt
==============================================================================
--- peps/trunk/pep-3151.txt	(original)
+++ peps/trunk/pep-3151.txt	Thu Jul 22 12:52:13 2010
@@ -426,7 +426,7 @@
 Appendix A: Survey of common errnos
 ===================================
 
-This is a quick recension of the various errno mnemonics checked for in
+This is a quick inventory of the various errno mnemonics checked for in
 the standard library and its tests, as part of ``except`` clauses.
 
 Common errnos with OSError

From python-checkins at python.org  Thu Jul 22 13:41:49 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Thu, 22 Jul 2010 13:41:49 +0200 (CEST)
Subject: [Python-checkins] r83049 - peps/trunk/pep-3151.txt
Message-ID: <20100722114149.64D93F1D3@mail.python.org>

Author: antoine.pitrou
Date: Thu Jul 22 13:41:49 2010
New Revision: 83049

Log:
Nick's input



Modified:
   peps/trunk/pep-3151.txt

Modified: peps/trunk/pep-3151.txt
==============================================================================
--- peps/trunk/pep-3151.txt	(original)
+++ peps/trunk/pep-3151.txt	Thu Jul 22 13:41:49 2010
@@ -95,7 +95,7 @@
 anyway.
 
 .. note::
-    **Appendix B** surveys the use of the various exception types accross
+    `Appendix B`_ surveys the use of the various exception types accross
     the interpreter and the standard library.
 
 
@@ -196,24 +196,55 @@
 Careful code should not be penalized.
 
 
+.. _Step 1:
+
 Step 1: coalesce exception types
 ================================
 
 The first step of the resolution is to coalesce existing exception types.
-The extent of this step is not yet fully determined.  A number of possible
-changes are listed hereafter:
+The following changes are proposed:
 
 * alias both socket.error and select.error to IOError
 * alias mmap.error to OSError
-* alias IOError to OSError
 * alias WindowsError to OSError
+* alias OSError to IOError
+* coalesce EnvironmentError into IOError
 
 Each of these changes doesn't preserve exact compatibility, but it does
 preserve *useful compatibility* (see "compatibility" section above).
 
-Not only does this first step present the user a simpler landscape, but
-it also allows for a better and more complete resolution of step 2
-(see "Prerequisite" below).
+Each of these changes can be accepted or refused invididually, but of course
+it is considered that the greatest impact can be achieved if this first step
+is accepted in full.  In this case, the IO exception subhierarchy would
+become::
+
+    +-- IOError   (replacing OSError, WindowsError, EnvironmentError, etc.)
+        +-- io.BlockingIOError
+        +-- io.UnsupportedOperation (also inherits from ValueError)
+
+Justification
+-------------
+
+Not only does this first step present the user a simpler landscape as
+explained in the rationale_ section, but it also allows for a better
+and more complete resolution of `Step 2`_ (see Prerequisite_).
+
+The rationale for keeping ``IOError`` as the official name for generic
+OS-related exceptions is the survey in `Appendix B`_, which shows it is the
+dominant error today in the standard library.  ``EnvironmentError`` might
+be more accurate, but it is more tedious to type and also much lesser-known.
+As for third-party Python code, Google Code Search shows IOError
+being ten times more popular than EnvironmentError in user code, and
+three times more popular than OSError [3]_.
+
+Exception attributes
+--------------------
+
+Coalescing WindowsError would mean the ``winerr`` attribute would be
+present on all platforms, just set to ``None`` if the platform
+isn't Windows.  Indeed, ``errno``, ``filename`` and ``strerror`` can all
+already be None, as is often the case when IOError is raised directly
+by Python code.
 
 Deprecation of names
 --------------------
@@ -224,13 +255,15 @@
 challenges, especially where performance is important.
 
 
+.. _Step 2:
+
 Step 2: define additional subclasses
 ====================================
 
 The second step of the resolution is to extend the hierarchy by defining
 subclasses which will be raised, rather than their parent, for specific
 errno values.  Which errno values is subject to discussion, but a survey
-of existing exception matching practices (see **Appendix A**) helps us
+of existing exception matching practices (see `Appendix A`_) helps us
 propose a reasonable subset of all values.  Trying to map all errno
 mnemonics, indeed, seems foolish, pointless, and would pollute the root
 namespace.
@@ -246,7 +279,7 @@
 Prerequisite
 ------------
 
-Step 1 is a loose prerequisite for this.
+`Step 1`_ is a loose prerequisite for this.
 
 Prerequisite, because some errnos can currently be attached to different
 exception classes: for example, EBADF can be attached to both OSError and
@@ -304,13 +337,12 @@
   as a generic timeout exception, useful for other types of timeout (for
   example in Lock.acquire())
 
-This list assumes step 1 is accepted in full; the exception classes
+This list assumes `Step 1`_ is accepted in full; the exception classes
 described above would all derive from the now unified exception type
 OSError.  It will need reworking if a partial version of step 1 is accepted
 instead (again, see appendix A for the current distribution of errnos
 and exception types).
 
-
 Exception attributes
 --------------------
 
@@ -423,6 +455,8 @@
 etc.).
 
 
+.. _Appendix A:
+
 Appendix A: Survey of common errnos
 ===================================
 
@@ -504,6 +538,8 @@
 * ``EINTR``: interrupted function call
 
 
+.. _Appendix B:
+
 Appendix B: Survey of raised OS and IO errors
 =============================================
 
@@ -740,15 +776,27 @@
 zipimporter.get_data() can raise IOError.
 
 
+Acknowledgments
+===============
+
+Significant input has been received from Nick Coghlan.
+
 References
 ==========
 
-.. [1] "IO module precisions and exception hierarchy"
+.. [1] "IO module precisions and exception hierarchy":
    http://mail.python.org/pipermail/python-dev/2009-September/092130.html
 
-.. [2] Discussion of "Removing WindowsError" in PEP 348
+.. [2] Discussion of "Removing WindowsError" in PEP 348:
    http://www.python.org/dev/peps/pep-0348/#removing-windowserror
 
+.. [3] Google Code Search of ``IOError`` in Python code: `around 40000 results
+   <http://www.google.com/codesearch?q=lang%3Apython%20IOError>`_;
+   ``OSError``: `around 15200 results
+   <http://www.google.com/codesearch?q=lang%3Apython%20OSError>`_;
+   ``EnvironmentError``: `around 3000 results
+   <http://www.google.com/codesearch?q=lang%3Apython%20EnvironmentError>`_
+
 Copyright
 =========
 

From python-checkins at python.org  Thu Jul 22 13:48:29 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Thu, 22 Jul 2010 13:48:29 +0200 (CEST)
Subject: [Python-checkins] r83050 - peps/trunk/pep-3151.txt
Message-ID: <20100722114829.7A535ECFB@mail.python.org>

Author: antoine.pitrou
Date: Thu Jul 22 13:48:29 2010
New Revision: 83050

Log:
The attribute is named winerror, not winerr



Modified:
   peps/trunk/pep-3151.txt

Modified: peps/trunk/pep-3151.txt
==============================================================================
--- peps/trunk/pep-3151.txt	(original)
+++ peps/trunk/pep-3151.txt	Thu Jul 22 13:48:29 2010
@@ -240,7 +240,7 @@
 Exception attributes
 --------------------
 
-Coalescing WindowsError would mean the ``winerr`` attribute would be
+Coalescing WindowsError would mean the ``winerror`` attribute would be
 present on all platforms, just set to ``None`` if the platform
 isn't Windows.  Indeed, ``errno``, ``filename`` and ``strerror`` can all
 already be None, as is often the case when IOError is raised directly

From python-checkins at python.org  Thu Jul 22 14:21:07 2010
From: python-checkins at python.org (nick.coghlan)
Date: Thu, 22 Jul 2010 14:21:07 +0200 (CEST)
Subject: [Python-checkins] r83051 - peps/trunk/pep-3150.txt
Message-ID: <20100722122107.7F36EF188@mail.python.org>

Author: nick.coghlan
Date: Thu Jul 22 14:21:07 2010
New Revision: 83051

Log:
Add a couple more TODO list items

Modified:
   peps/trunk/pep-3150.txt

Modified: peps/trunk/pep-3150.txt
==============================================================================
--- peps/trunk/pep-3150.txt	(original)
+++ peps/trunk/pep-3150.txt	Thu Jul 22 14:21:07 2010
@@ -389,9 +389,13 @@
 * Mention two-suite in-order variants (and explain why they're even more
   pointless than the specific idea in the PEP)
 * Mention PEP 359 and possible uses for locals() in the ``given`` clause
-* Describe the expected semantics of ``break``, ``continue``, ``return``
+* Define the expected semantics of ``break``, ``continue``, ``return``
   and ``yield`` in a ``given`` clause (i.e. syntax errors at the clause
   level, but allowed inside the appropriate compound statements)
+* Define the expected semantics of ``nonlocal`` and ``global`` in the
+  ``given`` clause
+* Define the name lookup semantics for function definitions in a
+  ``given`` clause at function, class and module scope.
 
 
 References

From python-checkins at python.org  Thu Jul 22 14:21:11 2010
From: python-checkins at python.org (benjamin.peterson)
Date: Thu, 22 Jul 2010 14:21:11 +0200 (CEST)
Subject: [Python-checkins] r83052 - peps/trunk/Makefile
Message-ID: <20100722122111.214F1F188@mail.python.org>

Author: benjamin.peterson
Date: Thu Jul 22 14:21:10 2010
New Revision: 83052

Log:
2.5 is now default on dinsdale, so don't require version suffix

Modified:
   peps/trunk/Makefile

Modified: peps/trunk/Makefile
==============================================================================
--- peps/trunk/Makefile	(original)
+++ peps/trunk/Makefile	Thu Jul 22 14:21:10 2010
@@ -5,7 +5,7 @@
 
 PEP2HTML=pep2html.py
 
-PYTHON=python2.5
+PYTHON=python
 
 .SUFFIXES: .txt .html
 

From python-checkins at python.org  Thu Jul 22 14:50:06 2010
From: python-checkins at python.org (tarek.ziade)
Date: Thu, 22 Jul 2010 14:50:06 +0200 (CEST)
Subject: [Python-checkins] r83053 - in python/branches/py3k:
	Lib/distutils/archive_util.py Lib/distutils/bcppcompiler.py
	Lib/distutils/ccompiler.py Lib/distutils/cmd.py
	Lib/distutils/command/bdist.py Lib/distutils/command/bdist_dumb.py
	Lib/distutils/command/bdist_msi.py
	Lib/distutils/command/bdist_rpm.py
	Lib/distutils/command/bdist_wininst.py
	Lib/distutils/command/build.py Lib/distutils/command/build_clib.py
	Lib/distutils/command/build_ext.py Lib/distutils/command/build_py.py
	Lib/distutils/command/build_scripts.py
	Lib/distutils/command/config.py Lib/distutils/command/install.py
	Lib/distutils/command/register.py Lib/distutils/command/sdist.py
	Lib/distutils/command/upload.py Lib/distutils/config.py
	Lib/distutils/core.py Lib/distutils/cygwinccompiler.py
	Lib/distutils/dep_util.py Lib/distutils/dir_util.py
	Lib/distutils/dist.py Lib/distutils/emxccompiler.py
	Lib/distutils/errors.py Lib/distutils/extension.py
	Lib/distutils/fancy_getopt.py Lib/distutils/file_util.py
	Lib/distutils/filelist.py Lib/distutils/msvc9compiler.py
	Lib/distutils/msvccompiler.py Lib/distutils/sysconfig.py
	Lib/distutils/tests/support.py
	Lib/distutils/tests/test_archive_util.py
	Lib/distutils/tests/test_bdist.py
	Lib/distutils/tests/test_bdist_dumb.py
	Lib/distutils/tests/test_bdist_rpm.py
	Lib/distutils/tests/test_bdist_wininst.py
	Lib/distutils/tests/test_build_clib.py
	Lib/distutils/tests/test_build_ext.py
	Lib/distutils/tests/test_build_py.py
	Lib/distutils/tests/test_build_scripts.py
	Lib/distutils/tests/test_ccompiler.py
	Lib/distutils/tests/test_cmd.py
	Lib/distutils/tests/test_cygwinccompiler.py
	Lib/distutils/tests/test_dist.py
	Lib/distutils/tests/test_emxccompiler.py
	Lib/distutils/tests/test_extension.py
	Lib/distutils/tests/test_file_util.py
	Lib/distutils/tests/test_filelist.py
	Lib/distutils/tests/test_install.py
	Lib/distutils/tests/test_install_lib.py
	Lib/distutils/tests/test_register.py
	Lib/distutils/tests/test_sdist.py
	Lib/distutils/tests/test_sysconfig.py
	Lib/distutils/tests/test_unixccompiler.py
	Lib/distutils/tests/test_upload.py
	Lib/distutils/tests/test_util.py Lib/distutils/text_file.py
	Lib/distutils/unixccompiler.py Lib/distutils/util.py setup.py
Message-ID: <20100722125006.1290DF417@mail.python.org>

Author: tarek.ziade
Date: Thu Jul 22 14:50:05 2010
New Revision: 83053

Log:
reverted distutils its 3.1 state. All new work is now happening in disutils2, and distutils is now feature-frozen.

Removed:
   python/branches/py3k/Lib/distutils/tests/test_ccompiler.py
   python/branches/py3k/Lib/distutils/tests/test_emxccompiler.py
Modified:
   python/branches/py3k/Lib/distutils/archive_util.py
   python/branches/py3k/Lib/distutils/bcppcompiler.py
   python/branches/py3k/Lib/distutils/ccompiler.py
   python/branches/py3k/Lib/distutils/cmd.py
   python/branches/py3k/Lib/distutils/command/bdist.py
   python/branches/py3k/Lib/distutils/command/bdist_dumb.py
   python/branches/py3k/Lib/distutils/command/bdist_msi.py
   python/branches/py3k/Lib/distutils/command/bdist_rpm.py
   python/branches/py3k/Lib/distutils/command/bdist_wininst.py
   python/branches/py3k/Lib/distutils/command/build.py
   python/branches/py3k/Lib/distutils/command/build_clib.py
   python/branches/py3k/Lib/distutils/command/build_ext.py
   python/branches/py3k/Lib/distutils/command/build_py.py
   python/branches/py3k/Lib/distutils/command/build_scripts.py
   python/branches/py3k/Lib/distutils/command/config.py
   python/branches/py3k/Lib/distutils/command/install.py
   python/branches/py3k/Lib/distutils/command/register.py
   python/branches/py3k/Lib/distutils/command/sdist.py
   python/branches/py3k/Lib/distutils/command/upload.py
   python/branches/py3k/Lib/distutils/config.py
   python/branches/py3k/Lib/distutils/core.py
   python/branches/py3k/Lib/distutils/cygwinccompiler.py
   python/branches/py3k/Lib/distutils/dep_util.py
   python/branches/py3k/Lib/distutils/dir_util.py
   python/branches/py3k/Lib/distutils/dist.py
   python/branches/py3k/Lib/distutils/emxccompiler.py
   python/branches/py3k/Lib/distutils/errors.py
   python/branches/py3k/Lib/distutils/extension.py
   python/branches/py3k/Lib/distutils/fancy_getopt.py
   python/branches/py3k/Lib/distutils/file_util.py
   python/branches/py3k/Lib/distutils/filelist.py
   python/branches/py3k/Lib/distutils/msvc9compiler.py
   python/branches/py3k/Lib/distutils/msvccompiler.py
   python/branches/py3k/Lib/distutils/sysconfig.py
   python/branches/py3k/Lib/distutils/tests/support.py
   python/branches/py3k/Lib/distutils/tests/test_archive_util.py
   python/branches/py3k/Lib/distutils/tests/test_bdist.py
   python/branches/py3k/Lib/distutils/tests/test_bdist_dumb.py
   python/branches/py3k/Lib/distutils/tests/test_bdist_rpm.py
   python/branches/py3k/Lib/distutils/tests/test_bdist_wininst.py
   python/branches/py3k/Lib/distutils/tests/test_build_clib.py
   python/branches/py3k/Lib/distutils/tests/test_build_ext.py
   python/branches/py3k/Lib/distutils/tests/test_build_py.py
   python/branches/py3k/Lib/distutils/tests/test_build_scripts.py
   python/branches/py3k/Lib/distutils/tests/test_cmd.py
   python/branches/py3k/Lib/distutils/tests/test_cygwinccompiler.py
   python/branches/py3k/Lib/distutils/tests/test_dist.py
   python/branches/py3k/Lib/distutils/tests/test_extension.py
   python/branches/py3k/Lib/distutils/tests/test_file_util.py
   python/branches/py3k/Lib/distutils/tests/test_filelist.py
   python/branches/py3k/Lib/distutils/tests/test_install.py
   python/branches/py3k/Lib/distutils/tests/test_install_lib.py
   python/branches/py3k/Lib/distutils/tests/test_register.py
   python/branches/py3k/Lib/distutils/tests/test_sdist.py
   python/branches/py3k/Lib/distutils/tests/test_sysconfig.py
   python/branches/py3k/Lib/distutils/tests/test_unixccompiler.py
   python/branches/py3k/Lib/distutils/tests/test_upload.py
   python/branches/py3k/Lib/distutils/tests/test_util.py
   python/branches/py3k/Lib/distutils/text_file.py
   python/branches/py3k/Lib/distutils/unixccompiler.py
   python/branches/py3k/Lib/distutils/util.py
   python/branches/py3k/setup.py

Modified: python/branches/py3k/Lib/distutils/archive_util.py
==============================================================================
--- python/branches/py3k/Lib/distutils/archive_util.py	(original)
+++ python/branches/py3k/Lib/distutils/archive_util.py	Thu Jul 22 14:50:05 2010
@@ -14,55 +14,15 @@
 from distutils.dir_util import mkpath
 from distutils import log
 
-try:
-    from pwd import getpwnam
-except ImportError:
-    getpwnam = None
-
-try:
-    from grp import getgrnam
-except ImportError:
-    getgrnam = None
-
-def _get_gid(name):
-    """Returns a gid, given a group name."""
-    if getgrnam is None or name is None:
-        return None
-    try:
-        result = getgrnam(name)
-    except KeyError:
-        result = None
-    if result is not None:
-        return result[2]
-    return None
-
-def _get_uid(name):
-    """Returns an uid, given a user name."""
-    if getpwnam is None or name is None:
-        return None
-    try:
-        result = getpwnam(name)
-    except KeyError:
-        result = None
-    if result is not None:
-        return result[2]
-    return None
-
-def make_tarball(base_name, base_dir, compress="gzip", verbose=0, dry_run=0,
-                 owner=None, group=None):
+def make_tarball(base_name, base_dir, compress="gzip", verbose=0, dry_run=0):
     """Create a (possibly compressed) tar file from all the files under
     'base_dir'.
 
     'compress' must be "gzip" (the default), "compress", "bzip2", or None.
-    (compress will be deprecated in Python 3.2)
-
-    'owner' and 'group' can be used to define an owner and a group for the
-    archive that is being built. If not provided, the current owner and group
-    will be used.
-
+    Both "tar" and the compression utility named by 'compress' must be on
+    the default program search path, so this is probably Unix-specific.
     The output tar file will be named 'base_dir' +  ".tar", possibly plus
     the appropriate compression extension (".gz", ".bz2" or ".Z").
-
     Returns the output filename.
     """
     tar_compression = {'gzip': 'gz', 'bzip2': 'bz2', None: '', 'compress': ''}
@@ -84,23 +44,10 @@
     import tarfile  # late import so Python build itself doesn't break
 
     log.info('Creating tar archive')
-
-    uid = _get_uid(owner)
-    gid = _get_gid(group)
-
-    def _set_uid_gid(tarinfo):
-        if gid is not None:
-            tarinfo.gid = gid
-            tarinfo.gname = group
-        if uid is not None:
-            tarinfo.uid = uid
-            tarinfo.uname = owner
-        return tarinfo
-
     if not dry_run:
         tar = tarfile.open(archive_name, 'w|%s' % tar_compression[compress])
         try:
-            tar.add(base_dir, filter=_set_uid_gid)
+            tar.add(base_dir)
         finally:
             tar.close()
 
@@ -190,7 +137,7 @@
     return None
 
 def make_archive(base_name, format, root_dir=None, base_dir=None, verbose=0,
-                 dry_run=0, owner=None, group=None):
+                 dry_run=0):
     """Create an archive file (eg. zip or tar).
 
     'base_name' is the name of the file to create, minus any format-specific
@@ -203,9 +150,6 @@
     ie. 'base_dir' will be the common prefix of all files and
     directories in the archive.  'root_dir' and 'base_dir' both default
     to the current directory.  Returns the name of the archive file.
-
-    'owner' and 'group' are used when creating a tar archive. By default,
-    uses the current owner and group.
     """
     save_cwd = os.getcwd()
     if root_dir is not None:
@@ -227,11 +171,6 @@
     func = format_info[0]
     for arg, val in format_info[1]:
         kwargs[arg] = val
-
-    if format != 'zip':
-        kwargs['owner'] = owner
-        kwargs['group'] = group
-
     try:
         filename = func(base_name, base_dir, **kwargs)
     finally:

Modified: python/branches/py3k/Lib/distutils/bcppcompiler.py
==============================================================================
--- python/branches/py3k/Lib/distutils/bcppcompiler.py	(original)
+++ python/branches/py3k/Lib/distutils/bcppcompiler.py	Thu Jul 22 14:50:05 2010
@@ -13,11 +13,13 @@
 
 __revision__ = "$Id$"
 
-import os
 
-from distutils.errors import (DistutilsExecError, CompileError, LibError,
-                              LinkError, UnknownFileError)
-from distutils.ccompiler import CCompiler, gen_preprocess_options
+import os
+from distutils.errors import \
+     DistutilsExecError, DistutilsPlatformError, \
+     CompileError, LibError, LinkError, UnknownFileError
+from distutils.ccompiler import \
+     CCompiler, gen_preprocess_options, gen_lib_options
 from distutils.file_util import write_file
 from distutils.dep_util import newer
 from distutils import log

Modified: python/branches/py3k/Lib/distutils/ccompiler.py
==============================================================================
--- python/branches/py3k/Lib/distutils/ccompiler.py	(original)
+++ python/branches/py3k/Lib/distutils/ccompiler.py	Thu Jul 22 14:50:05 2010
@@ -5,71 +5,15 @@
 
 __revision__ = "$Id$"
 
-import sys
-import os
-import re
-
-from distutils.errors import (CompileError, LinkError, UnknownFileError,
-                              DistutilsPlatformError, DistutilsModuleError)
+import sys, os, re
+from distutils.errors import *
 from distutils.spawn import spawn
 from distutils.file_util import move_file
 from distutils.dir_util import mkpath
-from distutils.dep_util import newer_group
+from distutils.dep_util import newer_pairwise, newer_group
 from distutils.util import split_quoted, execute
 from distutils import log
 
-_sysconfig = __import__('sysconfig')
-
-def customize_compiler(compiler):
-    """Do any platform-specific customization of a CCompiler instance.
-
-    Mainly needed on Unix, so we can plug in the information that
-    varies across Unices and is stored in Python's Makefile.
-    """
-    if compiler.compiler_type == "unix":
-        (cc, cxx, opt, cflags, ccshared, ldshared, so_ext, ar, ar_flags) = \
-            _sysconfig.get_config_vars('CC', 'CXX', 'OPT', 'CFLAGS',
-                                       'CCSHARED', 'LDSHARED', 'SO', 'AR',
-                                       'ARFLAGS')
-
-        if 'CC' in os.environ:
-            cc = os.environ['CC']
-        if 'CXX' in os.environ:
-            cxx = os.environ['CXX']
-        if 'LDSHARED' in os.environ:
-            ldshared = os.environ['LDSHARED']
-        if 'CPP' in os.environ:
-            cpp = os.environ['CPP']
-        else:
-            cpp = cc + " -E"           # not always
-        if 'LDFLAGS' in os.environ:
-            ldshared = ldshared + ' ' + os.environ['LDFLAGS']
-        if 'CFLAGS' in os.environ:
-            cflags = opt + ' ' + os.environ['CFLAGS']
-            ldshared = ldshared + ' ' + os.environ['CFLAGS']
-        if 'CPPFLAGS' in os.environ:
-            cpp = cpp + ' ' + os.environ['CPPFLAGS']
-            cflags = cflags + ' ' + os.environ['CPPFLAGS']
-            ldshared = ldshared + ' ' + os.environ['CPPFLAGS']
-        if 'AR' in os.environ:
-            ar = os.environ['AR']
-        if 'ARFLAGS' in os.environ:
-            archiver = ar + ' ' + os.environ['ARFLAGS']
-        else:
-            archiver = ar + ' ' + ar_flags
-
-        cc_cmd = cc + ' ' + cflags
-        compiler.set_executables(
-            preprocessor=cpp,
-            compiler=cc_cmd,
-            compiler_so=cc_cmd + ' ' + ccshared,
-            compiler_cxx=cxx,
-            linker_so=ldshared,
-            linker_exe=cc,
-            archiver=archiver)
-
-        compiler.shared_lib_extension = so_ext
-
 class CCompiler:
     """Abstract base class to define the interface that must be implemented
     by real compiler classes.  Also has some utility methods used by
@@ -449,6 +393,22 @@
 
         return output_dir, macros, include_dirs
 
+    def _prep_compile(self, sources, output_dir, depends=None):
+        """Decide which souce files must be recompiled.
+
+        Determine the list of object files corresponding to 'sources',
+        and figure out which ones really need to be recompiled.
+        Return a list of all object files and a dictionary telling
+        which source files can be skipped.
+        """
+        # Get the list of expected output (object) files
+        objects = self.object_filenames(sources, output_dir=output_dir)
+        assert len(objects) == len(sources)
+
+        # Return an empty dict for the "which source files can be skipped"
+        # return value to preserve API compatibility.
+        return objects, {}
+
     def _fix_object_args(self, objects, output_dir):
         """Typecheck and fix up some arguments supplied to various methods.
         Specifically: ensure that 'objects' is a list; if output_dir is
@@ -650,15 +610,26 @@
         """
         pass
 
+
     # values for target_desc parameter in link()
     SHARED_OBJECT = "shared_object"
     SHARED_LIBRARY = "shared_library"
     EXECUTABLE = "executable"
 
-    def link(self, target_desc, objects, output_filename, output_dir=None,
-             libraries=None, library_dirs=None, runtime_library_dirs=None,
-             export_symbols=None, debug=0, extra_preargs=None,
-             extra_postargs=None, build_temp=None, target_lang=None):
+    def link(self,
+             target_desc,
+             objects,
+             output_filename,
+             output_dir=None,
+             libraries=None,
+             library_dirs=None,
+             runtime_library_dirs=None,
+             export_symbols=None,
+             debug=0,
+             extra_preargs=None,
+             extra_postargs=None,
+             build_temp=None,
+             target_lang=None):
         """Link a bunch of stuff together to create an executable or
         shared library file.
 
@@ -707,11 +678,19 @@
 
     # Old 'link_*()' methods, rewritten to use the new 'link()' method.
 
-    def link_shared_lib(self, objects, output_libname, output_dir=None,
-                        libraries=None, library_dirs=None,
-                        runtime_library_dirs=None, export_symbols=None,
-                        debug=0, extra_preargs=None, extra_postargs=None,
-                        build_temp=None, target_lang=None):
+    def link_shared_lib(self,
+                        objects,
+                        output_libname,
+                        output_dir=None,
+                        libraries=None,
+                        library_dirs=None,
+                        runtime_library_dirs=None,
+                        export_symbols=None,
+                        debug=0,
+                        extra_preargs=None,
+                        extra_postargs=None,
+                        build_temp=None,
+                        target_lang=None):
         self.link(CCompiler.SHARED_LIBRARY, objects,
                   self.library_filename(output_libname, lib_type='shared'),
                   output_dir,
@@ -720,11 +699,19 @@
                   extra_preargs, extra_postargs, build_temp, target_lang)
 
 
-    def link_shared_object(self, objects, output_filename, output_dir=None,
-                           libraries=None, library_dirs=None,
-                           runtime_library_dirs=None, export_symbols=None,
-                           debug=0, extra_preargs=None, extra_postargs=None,
-                           build_temp=None, target_lang=None):
+    def link_shared_object(self,
+                           objects,
+                           output_filename,
+                           output_dir=None,
+                           libraries=None,
+                           library_dirs=None,
+                           runtime_library_dirs=None,
+                           export_symbols=None,
+                           debug=0,
+                           extra_preargs=None,
+                           extra_postargs=None,
+                           build_temp=None,
+                           target_lang=None):
         self.link(CCompiler.SHARED_OBJECT, objects,
                   output_filename, output_dir,
                   libraries, library_dirs, runtime_library_dirs,
@@ -732,10 +719,17 @@
                   extra_preargs, extra_postargs, build_temp, target_lang)
 
 
-    def link_executable(self, objects, output_progname, output_dir=None,
-                        libraries=None, library_dirs=None,
-                        runtime_library_dirs=None, debug=0, extra_preargs=None,
-                        extra_postargs=None, target_lang=None):
+    def link_executable(self,
+                        objects,
+                        output_progname,
+                        output_dir=None,
+                        libraries=None,
+                        library_dirs=None,
+                        runtime_library_dirs=None,
+                        debug=0,
+                        extra_preargs=None,
+                        extra_postargs=None,
+                        target_lang=None):
         self.link(CCompiler.EXECUTABLE, objects,
                   self.executable_filename(output_progname), output_dir,
                   libraries, library_dirs, runtime_library_dirs, None,
@@ -917,7 +911,7 @@
     def move_file(self, src, dst):
         return move_file(src, dst, dry_run=self.dry_run)
 
-    def mkpath(self, name, mode=0o777):
+    def mkpath (self, name, mode=0o777):
         mkpath(name, mode, dry_run=self.dry_run)
 
 
@@ -1085,14 +1079,12 @@
     return pp_opts
 
 
-def gen_lib_options(compiler, library_dirs, runtime_library_dirs, libraries):
+def gen_lib_options (compiler, library_dirs, runtime_library_dirs, libraries):
     """Generate linker options for searching library directories and
-    linking with specific libraries.
-
-    'libraries' and 'library_dirs' are, respectively, lists of library names
-    (not filenames!) and search directories.  Returns a list of command-line
-    options suitable for use with some compiler (depending on the two format
-    strings passed in).
+    linking with specific libraries.  'libraries' and 'library_dirs' are,
+    respectively, lists of library names (not filenames!) and search
+    directories.  Returns a list of command-line options suitable for use
+    with some compiler (depending on the two format strings passed in).
     """
     lib_opts = []
 
@@ -1102,7 +1094,7 @@
     for dir in runtime_library_dirs:
         opt = compiler.runtime_library_dir_option(dir)
         if isinstance(opt, list):
-            lib_opts.extend(opt)
+            lib_opts = lib_opts + opt
         else:
             lib_opts.append(opt)
 
@@ -1113,14 +1105,14 @@
     # pretty nasty way to arrange your C code.
 
     for lib in libraries:
-        lib_dir, lib_name = os.path.split(lib)
-        if lib_dir != '':
+        (lib_dir, lib_name) = os.path.split(lib)
+        if lib_dir:
             lib_file = compiler.find_library_file([lib_dir], lib_name)
-            if lib_file is not None:
+            if lib_file:
                 lib_opts.append(lib_file)
             else:
                 compiler.warn("no library file corresponding to "
                               "'%s' found (skipping)" % lib)
         else:
-            lib_opts.append(compiler.library_option(lib))
+            lib_opts.append(compiler.library_option (lib))
     return lib_opts

Modified: python/branches/py3k/Lib/distutils/cmd.py
==============================================================================
--- python/branches/py3k/Lib/distutils/cmd.py	(original)
+++ python/branches/py3k/Lib/distutils/cmd.py	Thu Jul 22 14:50:05 2010
@@ -367,11 +367,9 @@
         from distutils.spawn import spawn
         spawn(cmd, search_path, dry_run=self.dry_run)
 
-    def make_archive(self, base_name, format, root_dir=None, base_dir=None,
-                     owner=None, group=None):
-        return archive_util.make_archive(base_name, format, root_dir,
-                                         base_dir, dry_run=self.dry_run,
-                                         owner=owner, group=group)
+    def make_archive(self, base_name, format, root_dir=None, base_dir=None):
+        return archive_util.make_archive(base_name, format, root_dir, base_dir,
+                                         dry_run=self.dry_run)
 
     def make_file(self, infiles, outfile, func, args,
                   exec_msg=None, skip_msg=None, level=1):

Modified: python/branches/py3k/Lib/distutils/command/bdist.py
==============================================================================
--- python/branches/py3k/Lib/distutils/command/bdist.py	(original)
+++ python/branches/py3k/Lib/distutils/command/bdist.py	Thu Jul 22 14:50:05 2010
@@ -6,10 +6,9 @@
 __revision__ = "$Id$"
 
 import os
-
-from distutils.util import get_platform
 from distutils.core import Command
-from distutils.errors import DistutilsPlatformError, DistutilsOptionError
+from distutils.errors import *
+from distutils.util import get_platform
 
 
 def show_formats():
@@ -40,12 +39,6 @@
                      "[default: dist]"),
                     ('skip-build', None,
                      "skip rebuilding everything (for testing/debugging)"),
-                    ('owner=', 'u',
-                     "Owner name used when creating a tar file"
-                     " [default: current user]"),
-                    ('group=', 'g',
-                     "Group name used when creating a tar file"
-                     " [default: current group]"),
                    ]
 
     boolean_options = ['skip-build']
@@ -87,8 +80,6 @@
         self.formats = None
         self.dist_dir = None
         self.skip_build = 0
-        self.group = None
-        self.owner = None
 
     def finalize_options(self):
         # have to finalize 'plat_name' before 'bdist_base'
@@ -134,11 +125,6 @@
             if cmd_name not in self.no_format_option:
                 sub_cmd.format = self.formats[i]
 
-            # passing the owner and group names for tar archiving
-            if cmd_name == 'bdist_dumb':
-                sub_cmd.owner = self.owner
-                sub_cmd.group = self.group
-
             # If we're going to need to run this command again, tell it to
             # keep its temporary files around so subsequent runs go faster.
             if cmd_name in commands[i+1:]:

Modified: python/branches/py3k/Lib/distutils/command/bdist_dumb.py
==============================================================================
--- python/branches/py3k/Lib/distutils/command/bdist_dumb.py	(original)
+++ python/branches/py3k/Lib/distutils/command/bdist_dumb.py	Thu Jul 22 14:50:05 2010
@@ -7,18 +7,16 @@
 __revision__ = "$Id$"
 
 import os
-
-from sysconfig import get_python_version
-
-from distutils.util import get_platform
 from distutils.core import Command
+from distutils.util import get_platform
 from distutils.dir_util import remove_tree, ensure_relative
-from distutils.errors import DistutilsPlatformError
+from distutils.errors import *
+from distutils.sysconfig import get_python_version
 from distutils import log
 
 class bdist_dumb(Command):
 
-    description = 'create a "dumb" built distribution'
+    description = "create a \"dumb\" built distribution"
 
     user_options = [('bdist-dir=', 'd',
                      "temporary directory for creating the distribution"),
@@ -37,12 +35,6 @@
                     ('relative', None,
                      "build the archive using relative paths"
                      "(default: false)"),
-                    ('owner=', 'u',
-                     "Owner name used when creating a tar file"
-                     " [default: current user]"),
-                    ('group=', 'g',
-                     "Group name used when creating a tar file"
-                     " [default: current group]"),
                    ]
 
     boolean_options = ['keep-temp', 'skip-build', 'relative']
@@ -59,8 +51,6 @@
         self.dist_dir = None
         self.skip_build = 0
         self.relative = 0
-        self.owner = None
-        self.group = None
 
     def finalize_options(self):
         if self.bdist_dir is None:
@@ -118,8 +108,7 @@
 
         # Make the archive
         filename = self.make_archive(pseudoinstall_root,
-                                     self.format, root_dir=archive_root,
-                                     owner=self.owner, group=self.group)
+                                     self.format, root_dir=archive_root)
         if self.distribution.has_ext_modules():
             pyversion = get_python_version()
         else:

Modified: python/branches/py3k/Lib/distutils/command/bdist_msi.py
==============================================================================
--- python/branches/py3k/Lib/distutils/command/bdist_msi.py	(original)
+++ python/branches/py3k/Lib/distutils/command/bdist_msi.py	Thu Jul 22 14:50:05 2010
@@ -6,15 +6,15 @@
 """
 Implements the bdist_msi command.
 """
-import sys, os
-from sysconfig import get_python_version, get_platform
 
+import sys, os
 from distutils.core import Command
 from distutils.dir_util import remove_tree
+from distutils.sysconfig import get_python_version
 from distutils.version import StrictVersion
 from distutils.errors import DistutilsOptionError
+from distutils.util import get_platform
 from distutils import log
-
 import msilib
 from msilib import schema, sequence, text
 from msilib import Directory, Feature, Dialog, add_data
@@ -28,6 +28,7 @@
         default, cancel, bitmap=true)"""
         Dialog.__init__(self, *args)
         ruler = self.h - 36
+        bmwidth = 152*ruler/328
         #if kw.get("bitmap", True):
         #    self.bitmap("Bitmap", 0, 0, bmwidth, ruler, "PythonWin")
         self.line("BottomLine", 0, ruler, self.w, 0)
@@ -418,6 +419,7 @@
         # see "Dialog Style Bits"
         modal = 3      # visible | modal
         modeless = 1   # visible
+        track_disk_space = 32
 
         # UI customization properties
         add_data(db, "Property",

Modified: python/branches/py3k/Lib/distutils/command/bdist_rpm.py
==============================================================================
--- python/branches/py3k/Lib/distutils/command/bdist_rpm.py	(original)
+++ python/branches/py3k/Lib/distutils/command/bdist_rpm.py	Thu Jul 22 14:50:05 2010
@@ -5,14 +5,13 @@
 
 __revision__ = "$Id$"
 
-import sys
-import os
-
+import sys, os
 from distutils.core import Command
 from distutils.debug import DEBUG
+from distutils.util import get_platform
 from distutils.file_util import write_file
-from distutils.errors import (DistutilsOptionError, DistutilsPlatformError,
-                              DistutilsFileError, DistutilsExecError)
+from distutils.errors import *
+from distutils.sysconfig import get_python_version
 from distutils import log
 
 class bdist_rpm(Command):

Modified: python/branches/py3k/Lib/distutils/command/bdist_wininst.py
==============================================================================
--- python/branches/py3k/Lib/distutils/command/bdist_wininst.py	(original)
+++ python/branches/py3k/Lib/distutils/command/bdist_wininst.py	Thu Jul 22 14:50:05 2010
@@ -5,16 +5,13 @@
 
 __revision__ = "$Id$"
 
-import sys
-import os
-
-from sysconfig import get_python_version
-
+import sys, os
 from distutils.core import Command
-from distutils.dir_util import remove_tree
-from distutils.errors import DistutilsOptionError, DistutilsPlatformError
-from distutils import log
 from distutils.util import get_platform
+from distutils.dir_util import create_tree, remove_tree
+from distutils.errors import *
+from distutils.sysconfig import get_python_version
+from distutils import log
 
 class bdist_wininst(Command):
 

Modified: python/branches/py3k/Lib/distutils/command/build.py
==============================================================================
--- python/branches/py3k/Lib/distutils/command/build.py	(original)
+++ python/branches/py3k/Lib/distutils/command/build.py	Thu Jul 22 14:50:05 2010
@@ -5,15 +5,16 @@
 __revision__ = "$Id$"
 
 import sys, os
-
-from distutils.util import get_platform
 from distutils.core import Command
 from distutils.errors import DistutilsOptionError
+from distutils.util import get_platform
+
 
 def show_compilers():
     from distutils.ccompiler import show_compilers
     show_compilers()
 
+
 class build(Command):
 
     description = "build everything needed to install"
@@ -126,6 +127,7 @@
         for cmd_name in self.get_sub_commands():
             self.run_command(cmd_name)
 
+
     # -- Predicates for the sub-command list ---------------------------
 
     def has_pure_modules(self):
@@ -140,6 +142,7 @@
     def has_scripts(self):
         return self.distribution.has_scripts()
 
+
     sub_commands = [('build_py',      has_pure_modules),
                     ('build_clib',    has_c_libraries),
                     ('build_ext',     has_ext_modules),

Modified: python/branches/py3k/Lib/distutils/command/build_clib.py
==============================================================================
--- python/branches/py3k/Lib/distutils/command/build_clib.py	(original)
+++ python/branches/py3k/Lib/distutils/command/build_clib.py	Thu Jul 22 14:50:05 2010
@@ -18,8 +18,8 @@
 
 import os
 from distutils.core import Command
-from distutils.errors import DistutilsSetupError
-from distutils.ccompiler import customize_compiler
+from distutils.errors import *
+from distutils.sysconfig import customize_compiler
 from distutils import log
 
 def show_compilers():

Modified: python/branches/py3k/Lib/distutils/command/build_ext.py
==============================================================================
--- python/branches/py3k/Lib/distutils/command/build_ext.py	(original)
+++ python/branches/py3k/Lib/distutils/command/build_ext.py	Thu Jul 22 14:50:05 2010
@@ -7,14 +7,12 @@
 __revision__ = "$Id$"
 
 import sys, os, re
-from warnings import warn
-
-from distutils.util import get_platform
 from distutils.core import Command
 from distutils.errors import *
-from distutils.ccompiler import customize_compiler
+from distutils.sysconfig import customize_compiler, get_python_version
 from distutils.dep_util import newer_group
 from distutils.extension import Extension
+from distutils.util import get_platform
 from distutils import log
 
 # this keeps compatibility from 2.3 to 2.5
@@ -114,39 +112,6 @@
          "list available compilers", show_compilers),
         ]
 
-    # making 'compiler' a property to deprecate
-    # its usage as something else than a compiler type
-    # e.g. like a compiler instance
-    def __init__(self, dist):
-        self._compiler = None
-        Command.__init__(self, dist)
-
-    def __setattr__(self, name, value):
-        # need this to make sure setattr() (used in distutils)
-        # doesn't kill our property
-        if name == 'compiler':
-            self._set_compiler(value)
-        else:
-            self.__dict__[name] = value
-
-    def _set_compiler(self, compiler):
-        if not isinstance(compiler, str) and compiler is not None:
-            # we don't want to allow that anymore in the future
-            warn("'compiler' specifies the compiler type in build_ext. "
-                 "If you want to get the compiler object itself, "
-                 "use 'compiler_obj'", DeprecationWarning)
-        self._compiler = compiler
-
-    def _get_compiler(self):
-        if not isinstance(self._compiler, str) and self._compiler is not None:
-            # we don't want to allow that anymore in the future
-            warn("'compiler' specifies the compiler type in build_ext. "
-                 "If you want to get the compiler object itself, "
-                 "use 'compiler_obj'", DeprecationWarning)
-        return self._compiler
-
-    compiler = property(_get_compiler, _set_compiler)
-
     def initialize_options(self):
         self.extensions = None
         self.build_lib = None
@@ -171,7 +136,8 @@
         self.user = None
 
     def finalize_options(self):
-        _sysconfig = __import__('sysconfig')
+        from distutils import sysconfig
+
         self.set_undefined_options('build',
                                    ('build_lib', 'build_lib'),
                                    ('build_temp', 'build_temp'),
@@ -188,8 +154,8 @@
 
         # Make sure Python's include directories (for Python.h, pyconfig.h,
         # etc.) are in the include search path.
-        py_include = _sysconfig.get_path('include')
-        plat_py_include = _sysconfig.get_path('platinclude')
+        py_include = sysconfig.get_python_inc()
+        plat_py_include = sysconfig.get_python_inc(plat_specific=1)
         if self.include_dirs is None:
             self.include_dirs = self.distribution.include_dirs or []
         if isinstance(self.include_dirs, str):
@@ -261,13 +227,13 @@
         if os.name == 'os2':
             self.library_dirs.append(os.path.join(sys.exec_prefix, 'Config'))
 
-        # for extensions under Cygwin Python's library directory must be
+        # for extensions under Cygwin and AtheOS Python's library directory must be
         # appended to library_dirs
-        if sys.platform[:6] == 'cygwin':
+        if sys.platform[:6] == 'cygwin' or sys.platform[:6] == 'atheos':
             if sys.executable.startswith(os.path.join(sys.exec_prefix, "bin")):
                 # building third party extensions
                 self.library_dirs.append(os.path.join(sys.prefix, "lib",
-                                  "python" + _sysconfig.get_python_version(),
+                                                      "python" + get_python_version(),
                                                       "config"))
             else:
                 # building python standard extensions
@@ -275,13 +241,13 @@
 
         # for extensions under Linux or Solaris with a shared Python library,
         # Python's library directory must be appended to library_dirs
-        _sysconfig.get_config_var('Py_ENABLE_SHARED')
+        sysconfig.get_config_var('Py_ENABLE_SHARED')
         if ((sys.platform.startswith('linux') or sys.platform.startswith('gnu')
              or sys.platform.startswith('sunos'))
-            and _sysconfig.get_config_var('Py_ENABLE_SHARED')):
+            and sysconfig.get_config_var('Py_ENABLE_SHARED')):
             if sys.executable.startswith(os.path.join(sys.exec_prefix, "bin")):
                 # building third party extensions
-                self.library_dirs.append(_sysconfig.get_config_var('LIBDIR'))
+                self.library_dirs.append(sysconfig.get_config_var('LIBDIR'))
             else:
                 # building python standard extensions
                 self.library_dirs.append('.')
@@ -344,50 +310,38 @@
 
         # Setup the CCompiler object that we'll use to do all the
         # compiling and linking
-
-        # used to prevent the usage of an existing compiler for the
-        # compiler option when calling new_compiler()
-        # this will be removed in 3.3 and 2.8
-        if not isinstance(self._compiler, str):
-            self._compiler = None
-
-        self.compiler_obj = new_compiler(compiler=self._compiler,
-                                         verbose=self.verbose,
-                                         dry_run=self.dry_run,
-                                         force=self.force)
-
-        # used to keep the compiler object reachable with
-        # "self.compiler". this will be removed in 3.3 and 2.8
-        self._compiler = self.compiler_obj
-
-        customize_compiler(self.compiler_obj)
+        self.compiler = new_compiler(compiler=self.compiler,
+                                     verbose=self.verbose,
+                                     dry_run=self.dry_run,
+                                     force=self.force)
+        customize_compiler(self.compiler)
         # If we are cross-compiling, init the compiler now (if we are not
         # cross-compiling, init would not hurt, but people may rely on
         # late initialization of compiler even if they shouldn't...)
         if os.name == 'nt' and self.plat_name != get_platform():
-            self.compiler_obj.initialize(self.plat_name)
+            self.compiler.initialize(self.plat_name)
 
         # And make sure that any compile/link-related options (which might
         # come from the command-line or from the setup script) are set in
         # that CCompiler object -- that way, they automatically apply to
         # all compiling and linking done here.
         if self.include_dirs is not None:
-            self.compiler_obj.set_include_dirs(self.include_dirs)
+            self.compiler.set_include_dirs(self.include_dirs)
         if self.define is not None:
             # 'define' option is a list of (name,value) tuples
             for (name, value) in self.define:
-                self.compiler_obj.define_macro(name, value)
+                self.compiler.define_macro(name, value)
         if self.undef is not None:
             for macro in self.undef:
-                self.compiler_obj.undefine_macro(macro)
+                self.compiler.undefine_macro(macro)
         if self.libraries is not None:
-            self.compiler_obj.set_libraries(self.libraries)
+            self.compiler.set_libraries(self.libraries)
         if self.library_dirs is not None:
-            self.compiler_obj.set_library_dirs(self.library_dirs)
+            self.compiler.set_library_dirs(self.library_dirs)
         if self.rpath is not None:
-            self.compiler_obj.set_runtime_library_dirs(self.rpath)
+            self.compiler.set_runtime_library_dirs(self.rpath)
         if self.link_objects is not None:
-            self.compiler_obj.set_link_objects(self.link_objects)
+            self.compiler.set_link_objects(self.link_objects)
 
         # Now actually compile and link everything.
         self.build_extensions()
@@ -548,13 +502,13 @@
         for undef in ext.undef_macros:
             macros.append((undef,))
 
-        objects = self.compiler_obj.compile(sources,
-                                            output_dir=self.build_temp,
-                                            macros=macros,
-                                            include_dirs=ext.include_dirs,
-                                            debug=self.debug,
-                                            extra_postargs=extra_args,
-                                            depends=ext.depends)
+        objects = self.compiler.compile(sources,
+                                         output_dir=self.build_temp,
+                                         macros=macros,
+                                         include_dirs=ext.include_dirs,
+                                         debug=self.debug,
+                                         extra_postargs=extra_args,
+                                         depends=ext.depends)
 
         # XXX -- this is a Vile HACK!
         #
@@ -575,9 +529,9 @@
         extra_args = ext.extra_link_args or []
 
         # Detect target language, if not provided
-        language = ext.language or self.compiler_obj.detect_language(sources)
+        language = ext.language or self.compiler.detect_language(sources)
 
-        self.compiler_obj.link_shared_object(
+        self.compiler.link_shared_object(
             objects, ext_path,
             libraries=self.get_libraries(ext),
             library_dirs=ext.library_dirs,
@@ -710,13 +664,13 @@
         of the file from which it will be loaded (eg. "foo/bar.so", or
         "foo\bar.pyd").
         """
-        _sysconfig = __import__('sysconfig')
+        from distutils.sysconfig import get_config_var
         ext_path = ext_name.split('.')
         # OS/2 has an 8 character module (extension) limit :-(
         if os.name == "os2":
             ext_path[len(ext_path) - 1] = ext_path[len(ext_path) - 1][:8]
         # extensions in debug_mode are named 'module_d.pyd' under windows
-        so_ext = _sysconfig.get_config_var('SO')
+        so_ext = get_config_var('SO')
         if os.name == 'nt' and self.debug:
             return os.path.join(*ext_path) + '_d' + so_ext
         return os.path.join(*ext_path) + so_ext
@@ -744,7 +698,7 @@
         # Append '_d' to the python import library on debug builds.
         if sys.platform == "win32":
             from distutils.msvccompiler import MSVCCompiler
-            if not isinstance(self.compiler_obj, MSVCCompiler):
+            if not isinstance(self.compiler, MSVCCompiler):
                 template = "python%d%d"
                 if self.debug:
                     template = template + '_d'
@@ -775,12 +729,28 @@
             # don't extend ext.libraries, it may be shared with other
             # extensions, it is a reference to the original list
             return ext.libraries + [pythonlib]
+        elif sys.platform[:6] == "atheos":
+            from distutils import sysconfig
+
+            template = "python%d.%d"
+            pythonlib = (template %
+                   (sys.hexversion >> 24, (sys.hexversion >> 16) & 0xff))
+            # Get SHLIBS from Makefile
+            extra = []
+            for lib in sysconfig.get_config_var('SHLIBS').split():
+                if lib.startswith('-l'):
+                    extra.append(lib[2:])
+                else:
+                    extra.append(lib)
+            # don't extend ext.libraries, it may be shared with other
+            # extensions, it is a reference to the original list
+            return ext.libraries + [pythonlib, "m"] + extra
         elif sys.platform == 'darwin':
             # Don't use the default code below
             return ext.libraries
         else:
-            _sysconfig = __import__('sysconfig')
-            if _sysconfig.get_config_var('Py_ENABLE_SHARED'):
+            from distutils import sysconfig
+            if sysconfig.get_config_var('Py_ENABLE_SHARED'):
                 template = "python%d.%d"
                 pythonlib = (template %
                              (sys.hexversion >> 24, (sys.hexversion >> 16) & 0xff))

Modified: python/branches/py3k/Lib/distutils/command/build_py.py
==============================================================================
--- python/branches/py3k/Lib/distutils/command/build_py.py	(original)
+++ python/branches/py3k/Lib/distutils/command/build_py.py	Thu Jul 22 14:50:05 2010
@@ -4,16 +4,16 @@
 
 __revision__ = "$Id$"
 
-import os
+import sys, os
 import sys
 from glob import glob
 
 from distutils.core import Command
-from distutils.errors import DistutilsOptionError, DistutilsFileError
+from distutils.errors import *
 from distutils.util import convert_path, Mixin2to3
 from distutils import log
 
-class build_py(Command):
+class build_py (Command):
 
     description = "\"build\" pure Python modules (copy to build directory)"
 
@@ -133,6 +133,7 @@
 
     def build_package_data(self):
         """Copy data files into build directory"""
+        lastdir = None
         for package, src_dir, build_dir, filenames in self.data_files:
             for filename in filenames:
                 target = os.path.join(build_dir, filename)

Modified: python/branches/py3k/Lib/distutils/command/build_scripts.py
==============================================================================
--- python/branches/py3k/Lib/distutils/command/build_scripts.py	(original)
+++ python/branches/py3k/Lib/distutils/command/build_scripts.py	Thu Jul 22 14:50:05 2010
@@ -6,6 +6,7 @@
 
 import os, re
 from stat import ST_MODE
+from distutils import sysconfig
 from distutils.core import Command
 from distutils.dep_util import newer
 from distutils.util import convert_path, Mixin2to3
@@ -56,7 +57,6 @@
         ie. starts with "\#!" and contains "python"), then adjust the first
         line to refer to the current Python interpreter as we copy.
         """
-        _sysconfig = __import__('sysconfig')
         self.mkpath(self.build_dir)
         outfiles = []
         updated_files = []
@@ -96,16 +96,16 @@
                 updated_files.append(outfile)
                 if not self.dry_run:
                     outf = open(outfile, "w")
-                    if not _sysconfig.is_python_build():
+                    if not sysconfig.python_build:
                         outf.write("#!%s%s\n" %
                                    (self.executable,
                                     post_interp))
                     else:
                         outf.write("#!%s%s\n" %
                                    (os.path.join(
-                            _sysconfig.get_config_var("BINDIR"),
-                           "python%s%s" % (_sysconfig.get_config_var("VERSION"),
-                                           _sysconfig.get_config_var("EXE"))),
+                            sysconfig.get_config_var("BINDIR"),
+                           "python%s%s" % (sysconfig.get_config_var("VERSION"),
+                                           sysconfig.get_config_var("EXE"))),
                                     post_interp))
                     outf.writelines(f.readlines())
                     outf.close()

Modified: python/branches/py3k/Lib/distutils/command/config.py
==============================================================================
--- python/branches/py3k/Lib/distutils/command/config.py	(original)
+++ python/branches/py3k/Lib/distutils/command/config.py	Thu Jul 22 14:50:05 2010
@@ -11,12 +11,11 @@
 
 __revision__ = "$Id$"
 
-import os
-import re
+import sys, os, re
 
 from distutils.core import Command
 from distutils.errors import DistutilsExecError
-from distutils.ccompiler import customize_compiler
+from distutils.sysconfig import customize_compiler
 from distutils import log
 
 LANG_EXT = {"c": ".c", "c++": ".cxx"}

Modified: python/branches/py3k/Lib/distutils/command/install.py
==============================================================================
--- python/branches/py3k/Lib/distutils/command/install.py	(original)
+++ python/branches/py3k/Lib/distutils/command/install.py	Thu Jul 22 14:50:05 2010
@@ -7,17 +7,26 @@
 import sys
 import os
 
-from sysconfig import get_config_vars, get_paths, get_path, get_config_var
-
 from distutils import log
 from distutils.core import Command
 from distutils.debug import DEBUG
+from distutils.sysconfig import get_config_vars
 from distutils.errors import DistutilsPlatformError
 from distutils.file_util import write_file
-from distutils.util import convert_path, change_root, get_platform
+from distutils.util import convert_path, subst_vars, change_root
+from distutils.util import get_platform
 from distutils.errors import DistutilsOptionError
 
-# kept for backward compat, will be removed in 3.2
+# this keeps compatibility from 2.3 to 2.5
+if sys.version < "2.6":
+    USER_BASE = None
+    USER_SITE = None
+    HAS_USER_SITE = False
+else:
+    from site import USER_BASE
+    from site import USER_SITE
+    HAS_USER_SITE = True
+
 if sys.version < "2.2":
     WINDOWS_SCHEME = {
         'purelib': '$base',
@@ -50,21 +59,15 @@
         'scripts': '$base/bin',
         'data'   : '$base',
         },
-    'unix_user': {
-        'purelib': '$usersite',
-        'platlib': '$usersite',
-        'headers': '$userbase/include/python$py_version_short/$dist_name',
-        'scripts': '$userbase/bin',
-        'data'   : '$userbase',
-        },
     'nt': WINDOWS_SCHEME,
-    'nt_user': {
-        'purelib': '$usersite',
-        'platlib': '$usersite',
-        'headers': '$userbase/Python$py_version_nodot/Include/$dist_name',
-        'scripts': '$userbase/Scripts',
-        'data'   : '$userbase',
+    'mac': {
+        'purelib': '$base/Lib/site-packages',
+        'platlib': '$base/Lib/site-packages',
+        'headers': '$base/Include/$dist_name',
+        'scripts': '$base/Scripts',
+        'data'   : '$base',
         },
+
     'os2': {
         'purelib': '$base/Lib/site-packages',
         'platlib': '$base/Lib/site-packages',
@@ -72,26 +75,47 @@
         'scripts': '$base/Scripts',
         'data'   : '$base',
         },
-    'os2_home': {
+    }
+
+# user site schemes
+if HAS_USER_SITE:
+    INSTALL_SCHEMES['nt_user'] = {
+        'purelib': '$usersite',
+        'platlib': '$usersite',
+        'headers': '$userbase/Python$py_version_nodot/Include/$dist_name',
+        'scripts': '$userbase/Scripts',
+        'data'   : '$userbase',
+        }
+
+    INSTALL_SCHEMES['unix_user'] = {
         'purelib': '$usersite',
         'platlib': '$usersite',
         'headers': '$userbase/include/python$py_version_short/$dist_name',
         'scripts': '$userbase/bin',
         'data'   : '$userbase',
-        },
-    }
+        }
+
+    INSTALL_SCHEMES['mac_user'] = {
+        'purelib': '$usersite',
+        'platlib': '$usersite',
+        'headers': '$userbase/$py_version_short/include/$dist_name',
+        'scripts': '$userbase/bin',
+        'data'   : '$userbase',
+        }
 
+    INSTALL_SCHEMES['os2_home'] = {
+        'purelib': '$usersite',
+        'platlib': '$usersite',
+        'headers': '$userbase/include/python$py_version_short/$dist_name',
+        'scripts': '$userbase/bin',
+        'data'   : '$userbase',
+        }
+
+# The keys to an installation scheme; if any new types of files are to be
+# installed, be sure to add an entry to every installation scheme above,
+# and to SCHEME_KEYS here.
 SCHEME_KEYS = ('purelib', 'platlib', 'headers', 'scripts', 'data')
-# end of backward compat
 
-def _subst_vars(s, local_vars):
-    try:
-        return s.format(**local_vars)
-    except KeyError:
-        try:
-            return s.format(**os.environ)
-        except KeyError as var:
-            raise AttributeError('{%s}' % var)
 
 class install(Command):
 
@@ -158,10 +182,11 @@
 
     boolean_options = ['compile', 'force', 'skip-build']
 
-    user_options.append(('user', None,
-                        "install in user site-package '%s'" % \
-                            get_path('purelib', '%s_user' % os.name)))
-    boolean_options.append('user')
+    if HAS_USER_SITE:
+        user_options.append(('user', None,
+                             "install in user site-package '%s'" % USER_SITE))
+        boolean_options.append('user')
+
     negative_opt = {'no-compile' : 'compile'}
 
 
@@ -191,8 +216,8 @@
         self.install_lib = None         # set to either purelib or platlib
         self.install_scripts = None
         self.install_data = None
-        self.install_userbase = get_config_var('userbase')
-        self.install_usersite = get_path('purelib', '%s_user' % os.name)
+        self.install_userbase = USER_BASE
+        self.install_usersite = USER_SITE
 
         self.compile = None
         self.optimize = None
@@ -302,9 +327,7 @@
         # about needing recursive variable expansion (shudder).
 
         py_version = sys.version.split()[0]
-        prefix, exec_prefix, srcdir, projectbase = get_config_vars('prefix', 'exec_prefix',
-                                                      'srcdir', 'projectbase')
-
+        (prefix, exec_prefix) = get_config_vars('prefix', 'exec_prefix')
         self.config_vars = {'dist_name': self.distribution.get_name(),
                             'dist_version': self.distribution.get_version(),
                             'dist_fullname': self.distribution.get_fullname(),
@@ -315,12 +338,12 @@
                             'prefix': prefix,
                             'sys_exec_prefix': exec_prefix,
                             'exec_prefix': exec_prefix,
-                            'srcdir': srcdir,
-                            'projectbase': projectbase,
                            }
 
-        self.config_vars['userbase'] = self.install_userbase
-        self.config_vars['usersite'] = self.install_usersite
+        if HAS_USER_SITE:
+            self.config_vars['userbase'] = self.install_userbase
+            self.config_vars['usersite'] = self.install_usersite
+
         self.expand_basedirs()
 
         self.dump_dirs("post-expand_basedirs()")
@@ -424,10 +447,10 @@
                 raise DistutilsPlatformError(
                     "User base directory is not specified")
             self.install_base = self.install_platbase = self.install_userbase
-            self.select_scheme("posix_user")
+            self.select_scheme("unix_user")
         elif self.home is not None:
             self.install_base = self.install_platbase = self.home
-            self.select_scheme("posix_home")
+            self.select_scheme("unix_home")
         else:
             if self.prefix is None:
                 if self.exec_prefix is not None:
@@ -443,7 +466,7 @@
 
             self.install_base = self.prefix
             self.install_platbase = self.exec_prefix
-            self.select_scheme("posix_prefix")
+            self.select_scheme("unix_prefix")
 
     def finalize_other(self):
         """Finalizes options for non-posix platforms"""
@@ -455,7 +478,7 @@
             self.select_scheme(os.name + "_user")
         elif self.home is not None:
             self.install_base = self.install_platbase = self.home
-            self.select_scheme("posix_home")
+            self.select_scheme("unix_home")
         else:
             if self.prefix is None:
                 self.prefix = os.path.normpath(sys.prefix)
@@ -470,15 +493,11 @@
     def select_scheme(self, name):
         """Sets the install directories by applying the install schemes."""
         # it's the caller's problem if they supply a bad name!
-        scheme = get_paths(name, expand=False)
-        for key, value in scheme.items():
-            if key == 'platinclude':
-                key = 'headers'
-                value = os.path.join(value, self.distribution.get_name())
+        scheme = INSTALL_SCHEMES[name]
+        for key in SCHEME_KEYS:
             attrname = 'install_' + key
-            if hasattr(self, attrname):
-                if getattr(self, attrname) is None:
-                    setattr(self, attrname, value)
+            if getattr(self, attrname) is None:
+                setattr(self, attrname, scheme[key])
 
     def _expand_attrs(self, attrs):
         for attr in attrs:
@@ -486,7 +505,7 @@
             if val is not None:
                 if os.name == 'posix' or os.name == 'nt':
                     val = os.path.expanduser(val)
-                val = _subst_vars(val, self.config_vars)
+                val = subst_vars(val, self.config_vars)
                 setattr(self, attr, val)
 
     def expand_basedirs(self):

Modified: python/branches/py3k/Lib/distutils/command/register.py
==============================================================================
--- python/branches/py3k/Lib/distutils/command/register.py	(original)
+++ python/branches/py3k/Lib/distutils/command/register.py	Thu Jul 22 14:50:05 2010
@@ -7,15 +7,13 @@
 
 __revision__ = "$Id$"
 
-import os
-import string
-import getpass
+import os, string, getpass
 import io
-import urllib.parse
-import urllib.request
+import urllib.parse, urllib.request
 from warnings import warn
 
 from distutils.core import PyPIRCCommand
+from distutils.errors import *
 from distutils import log
 
 class register(PyPIRCCommand):

Modified: python/branches/py3k/Lib/distutils/command/sdist.py
==============================================================================
--- python/branches/py3k/Lib/distutils/command/sdist.py	(original)
+++ python/branches/py3k/Lib/distutils/command/sdist.py	Thu Jul 22 14:50:05 2010
@@ -7,14 +7,14 @@
 import os
 import string
 import sys
+from types import *
 from glob import glob
 from warnings import warn
 
 from distutils.core import Command
 from distutils import dir_util, dep_util, file_util, archive_util
 from distutils.text_file import TextFile
-from distutils.errors import (DistutilsPlatformError, DistutilsOptionError,
-                              DistutilsTemplateError)
+from distutils.errors import *
 from distutils.filelist import FileList
 from distutils import log
 from distutils.util import convert_path
@@ -76,10 +76,6 @@
         ('medata-check', None,
          "Ensure that all required elements of meta-data "
          "are supplied. Warn if any missing. [default]"),
-        ('owner=', 'u',
-         "Owner name used when creating a tar file [default: current user]"),
-        ('group=', 'g',
-         "Group name used when creating a tar file [default: current group]"),
         ]
 
     boolean_options = ['use-defaults', 'prune',
@@ -119,8 +115,6 @@
 
         self.archive_files = None
         self.metadata_check = 1
-        self.owner = None
-        self.group = None
 
     def finalize_options(self):
         if self.manifest is None:
@@ -424,8 +418,7 @@
             self.formats.append(self.formats.pop(self.formats.index('tar')))
 
         for fmt in self.formats:
-            file = self.make_archive(base_name, fmt, base_dir=base_dir,
-                                     owner=self.owner, group=self.group)
+            file = self.make_archive(base_name, fmt, base_dir=base_dir)
             archive_files.append(file)
             self.distribution.dist_files.append(('sdist', '', file))
 

Modified: python/branches/py3k/Lib/distutils/command/upload.py
==============================================================================
--- python/branches/py3k/Lib/distutils/command/upload.py	(original)
+++ python/branches/py3k/Lib/distutils/command/upload.py	Thu Jul 22 14:50:05 2010
@@ -1,19 +1,25 @@
 """distutils.command.upload
 
 Implements the Distutils 'upload' subcommand (upload package to PyPI)."""
-import os
-import io
-import socket
-import platform
-from urllib.request import urlopen, Request, HTTPError
-from base64 import standard_b64encode
-from urllib.parse import urlparse
-from hashlib import md5
 
-from distutils.errors import DistutilsOptionError
+from distutils.errors import *
 from distutils.core import PyPIRCCommand
 from distutils.spawn import spawn
 from distutils import log
+import sys
+import os, io
+import socket
+import platform
+import configparser
+import http.client as httpclient
+from base64 import standard_b64encode
+import urllib.parse
+
+# this keeps compatibility for 2.3 and 2.4
+if sys.version < "2.5":
+    from md5 import md5
+else:
+    from hashlib import md5
 
 class upload(PyPIRCCommand):
 
@@ -60,15 +66,6 @@
             self.upload_file(command, pyversion, filename)
 
     def upload_file(self, command, pyversion, filename):
-        # Makes sure the repository URL is compliant
-        schema, netloc, url, params, query, fragments = \
-            urlparse(self.repository)
-        if params or query or fragments:
-            raise AssertionError("Incompatible url %s" % self.repository)
-
-        if schema not in ('http', 'https'):
-            raise AssertionError("unsupported schema " + schema)
-
         # Sign if requested
         if self.sign:
             gpg_args = ["gpg", "--detach-sign", "-a", filename]
@@ -140,10 +137,10 @@
         for key, value in data.items():
             title = '\nContent-Disposition: form-data; name="%s"' % key
             # handle multiple entries for the same name
-            if not isinstance(value, list):
+            if type(value) != type([]):
                 value = [value]
             for value in value:
-                if isinstance(value, tuple):
+                if type(value) is tuple:
                     title += '; filename="%s"' % value[0]
                     value = value[1]
                 else:
@@ -161,30 +158,40 @@
         self.announce("Submitting %s to %s" % (filename, self.repository), log.INFO)
 
         # build the Request
-        headers = {'Content-type':
-                        'multipart/form-data; boundary=%s' % boundary,
-                   'Content-length': str(len(body)),
-                   'Authorization': auth}
-
-        request = Request(self.repository, data=body,
-                          headers=headers)
-        # send the data
+        # We can't use urllib since we need to send the Basic
+        # auth right with the first request
+        # TODO(jhylton): Can we fix urllib?
+        schema, netloc, url, params, query, fragments = \
+            urllib.parse.urlparse(self.repository)
+        assert not params and not query and not fragments
+        if schema == 'http':
+            http = httpclient.HTTPConnection(netloc)
+        elif schema == 'https':
+            http = httpclient.HTTPSConnection(netloc)
+        else:
+            raise AssertionError("unsupported schema "+schema)
+
+        data = ''
+        loglevel = log.INFO
         try:
-            result = urlopen(request)
-            status = result.getcode()
-            reason = result.msg
+            http.connect()
+            http.putrequest("POST", url)
+            http.putheader('Content-type',
+                           'multipart/form-data; boundary=%s'%boundary)
+            http.putheader('Content-length', str(len(body)))
+            http.putheader('Authorization', auth)
+            http.endheaders()
+            http.send(body)
         except socket.error as e:
             self.announce(str(e), log.ERROR)
             return
-        except HTTPError as e:
-            status = e.code
-            reason = e.msg
 
-        if status == 200:
-            self.announce('Server response (%s): %s' % (status, reason),
+        r = http.getresponse()
+        if r.status == 200:
+            self.announce('Server response (%s): %s' % (r.status, r.reason),
                           log.INFO)
         else:
-            self.announce('Upload failed (%s): %s' % (status, reason),
+            self.announce('Upload failed (%s): %s' % (r.status, r.reason),
                           log.ERROR)
         if self.show_response:
-            self.announce('-'*75, result.read(), '-'*75)
+            self.announce('-'*75, r.read(), '-'*75)

Modified: python/branches/py3k/Lib/distutils/config.py
==============================================================================
--- python/branches/py3k/Lib/distutils/config.py	(original)
+++ python/branches/py3k/Lib/distutils/config.py	Thu Jul 22 14:50:05 2010
@@ -4,6 +4,7 @@
 that uses .pypirc in the distutils.command package.
 """
 import os
+import sys
 from configparser import ConfigParser
 
 from distutils.cmd import Command
@@ -59,6 +60,8 @@
         if os.path.exists(rc):
             self.announce('Using PyPI login from %s' % rc)
             repository = self.repository or self.DEFAULT_REPOSITORY
+            realm = self.realm or self.DEFAULT_REALM
+
             config = ConfigParser()
             config.read(rc)
             sections = config.sections()

Modified: python/branches/py3k/Lib/distutils/core.py
==============================================================================
--- python/branches/py3k/Lib/distutils/core.py	(original)
+++ python/branches/py3k/Lib/distutils/core.py	Thu Jul 22 14:50:05 2010
@@ -8,12 +8,10 @@
 
 __revision__ = "$Id$"
 
-import sys
-import os
+import sys, os
 
 from distutils.debug import DEBUG
-from distutils.errors import (DistutilsSetupError, DistutilsArgError,
-                              DistutilsError, CCompilerError)
+from distutils.errors import *
 from distutils.util import grok_environment_error
 
 # Mainly import these so setup scripts can "from distutils.core import" them.
@@ -33,9 +31,9 @@
    or: %(script)s cmd --help
 """
 
-def gen_usage(script_name):
+def gen_usage (script_name):
     script = os.path.basename(script_name)
-    return USAGE % {'script': script}
+    return USAGE % vars()
 
 
 # Some mild magic to control the behaviour of 'setup()' from 'run_setup()'.
@@ -58,7 +56,7 @@
                       'extra_objects', 'extra_compile_args', 'extra_link_args',
                       'swig_opts', 'export_symbols', 'depends', 'language')
 
-def setup(**attrs):
+def setup (**attrs):
     """The gateway to the Distutils: do everything your setup script needs
     to do, in a highly flexible and user-driven way.  Briefly: create a
     Distribution instance; find and parse config files; parse the command
@@ -131,9 +129,8 @@
     if _setup_stop_after == "config":
         return dist
 
-    # Parse the command line and override config files; any
-    # command-line errors are the end user's fault, so turn them into
-    # SystemExit to suppress tracebacks.
+    # Parse the command line; any command-line errors are the end user's
+    # fault, so turn them into SystemExit to suppress tracebacks.
     try:
         ok = dist.parse_command_line()
     except DistutilsArgError as msg:
@@ -170,8 +167,10 @@
 
     return dist
 
+# setup ()
 
-def run_setup(script_name, script_args=None, stop_after="run"):
+
+def run_setup (script_name, script_args=None, stop_after="run"):
     """Run a setup script in a somewhat controlled environment, and
     return the Distribution instance that drives things.  This is useful
     if you need to find out the distribution meta-data (passed as
@@ -234,4 +233,7 @@
 
     # I wonder if the setup script's namespace -- g and l -- would be of
     # any interest to callers?
+    #print "_setup_distribution:", _setup_distribution
     return _setup_distribution
+
+# run_setup ()

Modified: python/branches/py3k/Lib/distutils/cygwinccompiler.py
==============================================================================
--- python/branches/py3k/Lib/distutils/cygwinccompiler.py	(original)
+++ python/branches/py3k/Lib/distutils/cygwinccompiler.py	Thu Jul 22 14:50:05 2010
@@ -50,13 +50,16 @@
 import os
 import sys
 import copy
+from subprocess import Popen, PIPE
 import re
-from warnings import warn
 
+from distutils.ccompiler import gen_preprocess_options, gen_lib_options
 from distutils.unixccompiler import UnixCCompiler
 from distutils.file_util import write_file
 from distutils.errors import DistutilsExecError, CompileError, UnknownFileError
-from distutils.util import get_compiler_versions
+from distutils import log
+from distutils.version import LooseVersion
+from distutils.spawn import find_executable
 
 def get_msvcr():
     """Include the appropriate MSVC runtime library if Python was built
@@ -107,7 +110,7 @@
                 % details)
 
         self.gcc_version, self.ld_version, self.dllwrap_version = \
-            get_compiler_versions()
+            get_versions()
         self.debug_print(self.compiler_type + ": gcc %s, ld %s, dllwrap %s\n" %
                          (self.gcc_version,
                           self.ld_version,
@@ -337,7 +340,7 @@
     # XXX since this function also checks sys.version, it's not strictly a
     # "pyconfig.h" check -- should probably be renamed...
 
-    _sysconfig = __import__('sysconfig')
+    from distutils import sysconfig
 
     # if sys.version contains GCC then python was compiled with GCC, and the
     # pyconfig.h file should be OK
@@ -345,7 +348,7 @@
         return CONFIG_H_OK, "sys.version mentions 'GCC'"
 
     # let's see if __GNUC__ is mentioned in python.h
-    fn = _sysconfig.get_config_h_filename()
+    fn = sysconfig.get_config_h_filename()
     try:
         with open(fn) as config_h:
             if "__GNUC__" in config_h.read():
@@ -356,27 +359,33 @@
         return (CONFIG_H_UNCERTAIN,
                 "couldn't read '%s': %s" % (fn, exc.strerror))
 
-class _Deprecated_SRE_Pattern(object):
-    def __init__(self, pattern):
-        self.pattern = pattern
-
-    def __getattr__(self, name):
-        if name in ('findall', 'finditer', 'match', 'scanner', 'search',
-                    'split', 'sub', 'subn'):
-            warn("'distutils.cygwinccompiler.RE_VERSION' is deprecated "
-                 "and will be removed in the next version", DeprecationWarning)
-        return getattr(self.pattern, name)
+RE_VERSION = re.compile(b'(\d+\.\d+(\.\d+)*)')
 
+def _find_exe_version(cmd):
+    """Find the version of an executable by running `cmd` in the shell.
 
-RE_VERSION = _Deprecated_SRE_Pattern(re.compile('(\d+\.\d+(\.\d+)*)'))
+    If the command is not found, or the output does not match
+    `RE_VERSION`, returns None.
+    """
+    executable = cmd.split()[0]
+    if find_executable(executable) is None:
+        return None
+    out = Popen(cmd, shell=True, stdout=PIPE).stdout
+    try:
+        out_string = out.read()
+    finally:
+        out.close()
+    result = RE_VERSION.search(out_string)
+    if result is None:
+        return None
+    # LooseVersion works with strings
+    # so we need to decode our bytes
+    return LooseVersion(result.group(1).decode())
 
 def get_versions():
     """ Try to find out the versions of gcc, ld and dllwrap.
 
     If not possible it returns None for it.
     """
-    warn("'distutils.cygwinccompiler.get_versions' is deprecated "
-         "use 'distutils.util.get_compiler_versions' instead",
-         DeprecationWarning)
-
-    return get_compiler_versions()
+    commands = ['gcc -dumpversion', 'ld -v', 'dllwrap --version']
+    return tuple([_find_exe_version(cmd) for cmd in commands])

Modified: python/branches/py3k/Lib/distutils/dep_util.py
==============================================================================
--- python/branches/py3k/Lib/distutils/dep_util.py	(original)
+++ python/branches/py3k/Lib/distutils/dep_util.py	Thu Jul 22 14:50:05 2010
@@ -9,27 +9,29 @@
 import os
 from distutils.errors import DistutilsFileError
 
-def newer(source, target):
-    """Tells if the target is newer than the source.
 
-    Return true if 'source' exists and is more recently modified than
-    'target', or if 'source' exists and 'target' doesn't.
-
-    Return false if both exist and 'target' is the same age or younger
-    than 'source'. Raise DistutilsFileError if 'source' does not exist.
-
-    Note that this test is not very accurate: files created in the same second
-    will have the same "age".
+def newer (source, target):
+    """Return true if 'source' exists and is more recently modified than
+    'target', or if 'source' exists and 'target' doesn't.  Return false if
+    both exist and 'target' is the same age or younger than 'source'.
+    Raise DistutilsFileError if 'source' does not exist.
     """
     if not os.path.exists(source):
         raise DistutilsFileError("file '%s' does not exist" %
                                  os.path.abspath(source))
     if not os.path.exists(target):
-        return True
+        return 1
+
+    from stat import ST_MTIME
+    mtime1 = os.stat(source)[ST_MTIME]
+    mtime2 = os.stat(target)[ST_MTIME]
+
+    return mtime1 > mtime2
+
+# newer ()
 
-    return os.stat(source).st_mtime > os.stat(target).st_mtime
 
-def newer_pairwise(sources, targets):
+def newer_pairwise (sources, targets):
     """Walk two filename lists in parallel, testing if each source is newer
     than its corresponding target.  Return a pair of lists (sources,
     targets) where source is newer than target, according to the semantics
@@ -41,18 +43,19 @@
     # build a pair of lists (sources, targets) where  source is newer
     n_sources = []
     n_targets = []
-    for source, target in zip(sources, targets):
-        if newer(source, target):
-            n_sources.append(source)
-            n_targets.append(target)
+    for i in range(len(sources)):
+        if newer(sources[i], targets[i]):
+            n_sources.append(sources[i])
+            n_targets.append(targets[i])
 
-    return n_sources, n_targets
+    return (n_sources, n_targets)
+
+# newer_pairwise ()
 
-def newer_group(sources, target, missing='error'):
-    """Return true if 'target' is out-of-date with respect to any file
-    listed in 'sources'.
 
-    In other words, if 'target' exists and is newer
+def newer_group (sources, target, missing='error'):
+    """Return true if 'target' is out-of-date with respect to any file
+    listed in 'sources'.  In other words, if 'target' exists and is newer
     than every file in 'sources', return false; otherwise return true.
     'missing' controls what we do when a source file is missing; the
     default ("error") is to blow up with an OSError from inside 'stat()';
@@ -65,14 +68,14 @@
     """
     # If the target doesn't even exist, then it's definitely out-of-date.
     if not os.path.exists(target):
-        return True
+        return 1
 
     # Otherwise we have to find out the hard way: if *any* source file
     # is more recent than 'target', then 'target' is out-of-date and
     # we can immediately return true.  If we fall through to the end
     # of the loop, then 'target' is up-to-date and we return false.
-    target_mtime = os.stat(target).st_mtime
-
+    from stat import ST_MTIME
+    target_mtime = os.stat(target)[ST_MTIME]
     for source in sources:
         if not os.path.exists(source):
             if missing == 'error':      # blow up when we stat() the file
@@ -80,9 +83,12 @@
             elif missing == 'ignore':   # missing source dropped from
                 continue                #  target's dependency list
             elif missing == 'newer':    # missing source means target is
-                return True             #  out-of-date
+                return 1                #  out-of-date
 
-        if os.stat(source).st_mtime > target_mtime:
-            return True
+        source_mtime = os.stat(source)[ST_MTIME]
+        if source_mtime > target_mtime:
+            return 1
+    else:
+        return 0
 
-    return False
+# newer_group ()

Modified: python/branches/py3k/Lib/distutils/dir_util.py
==============================================================================
--- python/branches/py3k/Lib/distutils/dir_util.py	(original)
+++ python/branches/py3k/Lib/distutils/dir_util.py	Thu Jul 22 14:50:05 2010
@@ -4,7 +4,7 @@
 
 __revision__ = "$Id$"
 
-import os
+import os, sys
 from distutils.errors import DistutilsFileError, DistutilsInternalError
 from distutils import log
 

Modified: python/branches/py3k/Lib/distutils/dist.py
==============================================================================
--- python/branches/py3k/Lib/distutils/dist.py	(original)
+++ python/branches/py3k/Lib/distutils/dist.py	Thu Jul 22 14:50:05 2010
@@ -7,15 +7,13 @@
 __revision__ = "$Id$"
 
 import sys, os, re
-from email import message_from_file
 
 try:
     import warnings
 except ImportError:
     warnings = None
 
-from distutils.errors import (DistutilsOptionError, DistutilsArgError,
-                              DistutilsModuleError, DistutilsClassError)
+from distutils.errors import *
 from distutils.fancy_getopt import FancyGetopt, translate_longopt
 from distutils.util import check_environ, strtobool, rfc822_escape
 from distutils import log
@@ -55,9 +53,7 @@
                       ('quiet', 'q', "run quietly (turns verbosity off)"),
                       ('dry-run', 'n', "don't actually do anything"),
                       ('help', 'h', "show detailed help message"),
-                      ('no-user-cfg', None,
-                       'ignore pydistutils.cfg in your home directory'),
-    ]
+                     ]
 
     # 'common_usage' is a short (2-3 line) string describing the common
     # usage of the setup script.
@@ -264,22 +260,6 @@
                     else:
                         sys.stderr.write(msg + "\n")
 
-        # no-user-cfg is handled before other command line args
-        # because other args override the config files, and this
-        # one is needed before we can load the config files.
-        # If attrs['script_args'] wasn't passed, assume false.
-        #
-        # This also make sure we just look at the global options
-        self.want_user_cfg = True
-
-        if self.script_args is not None:
-            for arg in self.script_args:
-                if not arg.startswith('-'):
-                    break
-                if arg == '--no-user-cfg':
-                    self.want_user_cfg = False
-                    break
-
         self.finalize_options()
 
     def get_option_dict(self, command):
@@ -331,10 +311,7 @@
         Distutils installation directory (ie. where the top-level
         Distutils __inst__.py file lives), a file in the user's home
         directory named .pydistutils.cfg on Unix and pydistutils.cfg
-        on Windows/Mac; and setup.cfg in the current directory.
-
-        The file in the user's home directory can be disabled with the
-        --no-user-cfg option.
+        on Windows/Mac, and setup.cfg in the current directory.
         """
         files = []
         check_environ()
@@ -354,19 +331,15 @@
             user_filename = "pydistutils.cfg"
 
         # And look for the user config file
-        if self.want_user_cfg:
-            user_file = os.path.join(os.path.expanduser('~'), user_filename)
-            if os.path.isfile(user_file):
-                files.append(user_file)
+        user_file = os.path.join(os.path.expanduser('~'), user_filename)
+        if os.path.isfile(user_file):
+            files.append(user_file)
 
         # All platforms support local setup.cfg
         local_file = "setup.cfg"
         if os.path.isfile(local_file):
             files.append(local_file)
 
-        if DEBUG:
-            self.announce("using config files: %s" % ', '.join(files))
-
         return files
 
     def parse_config_files(self, filenames=None):
@@ -1016,80 +989,25 @@
                          "provides", "requires", "obsoletes",
                          )
 
-    def __init__(self, path=None):
-        if path is not None:
-            self.read_pkg_file(open(path))
-        else:
-            self.name = None
-            self.version = None
-            self.author = None
-            self.author_email = None
-            self.maintainer = None
-            self.maintainer_email = None
-            self.url = None
-            self.license = None
-            self.description = None
-            self.long_description = None
-            self.keywords = None
-            self.platforms = None
-            self.classifiers = None
-            self.download_url = None
-            # PEP 314
-            self.provides = None
-            self.requires = None
-            self.obsoletes = None
-
-    def read_pkg_file(self, file):
-        """Reads the metadata values from a file object."""
-        msg = message_from_file(file)
-
-        def _read_field(name):
-            value = msg[name]
-            if value == 'UNKNOWN':
-                return None
-            return value
-
-        def _read_list(name):
-            values = msg.get_all(name, None)
-            if values == []:
-                return None
-            return values
-
-        metadata_version = msg['metadata-version']
-        self.name = _read_field('name')
-        self.version = _read_field('version')
-        self.description = _read_field('summary')
-        # we are filling author only.
-        self.author = _read_field('author')
+    def __init__ (self):
+        self.name = None
+        self.version = None
+        self.author = None
+        self.author_email = None
         self.maintainer = None
-        self.author_email = _read_field('author-email')
         self.maintainer_email = None
-        self.url = _read_field('home-page')
-        self.license = _read_field('license')
-
-        if 'download-url' in msg:
-            self.download_url = _read_field('download-url')
-        else:
-            self.download_url = None
-
-        self.long_description = _read_field('description')
-        self.description = _read_field('summary')
-
-        if 'keywords' in msg:
-            self.keywords = _read_field('keywords').split(',')
-
-        self.platforms = _read_list('platform')
-        self.classifiers = _read_list('classifier')
-
-        # PEP 314 - these fields only exist in 1.1
-        if metadata_version == '1.1':
-            self.requires = _read_list('requires')
-            self.provides = _read_list('provides')
-            self.obsoletes = _read_list('obsoletes')
-        else:
-            self.requires = None
-            self.provides = None
-            self.obsoletes = None
+        self.url = None
+        self.license = None
+        self.description = None
+        self.long_description = None
+        self.keywords = None
+        self.platforms = None
+        self.classifiers = None
+        self.download_url = None
+        # PEP 314
+        self.provides = None
+        self.requires = None
+        self.obsoletes = None
 
     def write_pkg_info(self, base_dir):
         """Write the PKG-INFO file into the release tree.

Modified: python/branches/py3k/Lib/distutils/emxccompiler.py
==============================================================================
--- python/branches/py3k/Lib/distutils/emxccompiler.py	(original)
+++ python/branches/py3k/Lib/distutils/emxccompiler.py	Thu Jul 22 14:50:05 2010
@@ -21,13 +21,12 @@
 
 __revision__ = "$Id$"
 
-import os, sys, copy
-from warnings import warn
-
+import os,sys,copy
+from distutils.ccompiler import gen_preprocess_options, gen_lib_options
 from distutils.unixccompiler import UnixCCompiler
 from distutils.file_util import write_file
 from distutils.errors import DistutilsExecError, CompileError, UnknownFileError
-from distutils.util import get_compiler_versions
+from distutils import log
 
 class EMXCCompiler (UnixCCompiler):
 
@@ -56,8 +55,8 @@
                 ("Reason: %s." % details) +
                 "Compiling may fail because of undefined preprocessor macros.")
 
-        gcc_version, ld_version, dllwrap_version = get_compiler_versions()
-        self.gcc_version, self.ld_version = gcc_version, ld_version
+        (self.gcc_version, self.ld_version) = \
+            get_versions()
         self.debug_print(self.compiler_type + ": gcc %s, ld %s\n" %
                          (self.gcc_version,
                           self.ld_version) )
@@ -292,11 +291,23 @@
     """ Try to find out the versions of gcc and ld.
         If not possible it returns None for it.
     """
-    warn("'distutils.emxccompiler.get_versions' is deprecated "
-         "use 'distutils.util.get_compiler_versions' instead",
-         DeprecationWarning)
-
+    from distutils.version import StrictVersion
+    from distutils.spawn import find_executable
+    import re
+
+    gcc_exe = find_executable('gcc')
+    if gcc_exe:
+        out = os.popen(gcc_exe + ' -dumpversion','r')
+        out_string = out.read()
+        out.close()
+        result = re.search('(\d+\.\d+\.\d+)', out_string, re.ASCII)
+        if result:
+            gcc_version = StrictVersion(result.group(1))
+        else:
+            gcc_version = None
+    else:
+        gcc_version = None
     # EMX ld has no way of reporting version number, and we use GCC
     # anyway - so we can link OMF DLLs
-    gcc_version, ld_version, dllwrap_version = get_compiler_versions()
-    return gcc_version, None
+    ld_version = None
+    return (gcc_version, ld_version)

Modified: python/branches/py3k/Lib/distutils/errors.py
==============================================================================
--- python/branches/py3k/Lib/distutils/errors.py	(original)
+++ python/branches/py3k/Lib/distutils/errors.py	Thu Jul 22 14:50:05 2010
@@ -10,79 +10,90 @@
 
 __revision__ = "$Id$"
 
-class DistutilsError(Exception):
+class DistutilsError (Exception):
     """The root of all Distutils evil."""
+    pass
 
-class DistutilsModuleError(DistutilsError):
+class DistutilsModuleError (DistutilsError):
     """Unable to load an expected module, or to find an expected class
     within some module (in particular, command modules and classes)."""
+    pass
 
-class DistutilsClassError(DistutilsError):
+class DistutilsClassError (DistutilsError):
     """Some command class (or possibly distribution class, if anyone
     feels a need to subclass Distribution) is found not to be holding
     up its end of the bargain, ie. implementing some part of the
     "command "interface."""
+    pass
 
-class DistutilsGetoptError(DistutilsError):
+class DistutilsGetoptError (DistutilsError):
     """The option table provided to 'fancy_getopt()' is bogus."""
+    pass
 
-class DistutilsArgError(DistutilsError):
+class DistutilsArgError (DistutilsError):
     """Raised by fancy_getopt in response to getopt.error -- ie. an
     error in the command line usage."""
+    pass
 
-class DistutilsFileError(DistutilsError):
+class DistutilsFileError (DistutilsError):
     """Any problems in the filesystem: expected file not found, etc.
     Typically this is for problems that we detect before IOError or
     OSError could be raised."""
+    pass
 
-class DistutilsOptionError(DistutilsError):
+class DistutilsOptionError (DistutilsError):
     """Syntactic/semantic errors in command options, such as use of
     mutually conflicting options, or inconsistent options,
     badly-spelled values, etc.  No distinction is made between option
     values originating in the setup script, the command line, config
     files, or what-have-you -- but if we *know* something originated in
     the setup script, we'll raise DistutilsSetupError instead."""
+    pass
 
-class DistutilsSetupError(DistutilsError):
+class DistutilsSetupError (DistutilsError):
     """For errors that can be definitely blamed on the setup script,
     such as invalid keyword arguments to 'setup()'."""
+    pass
 
-class DistutilsPlatformError(DistutilsError):
+class DistutilsPlatformError (DistutilsError):
     """We don't know how to do something on the current platform (but
     we do know how to do it on some platform) -- eg. trying to compile
     C files on a platform not supported by a CCompiler subclass."""
+    pass
 
-class DistutilsExecError(DistutilsError):
+class DistutilsExecError (DistutilsError):
     """Any problems executing an external program (such as the C
     compiler, when compiling C files)."""
+    pass
 
-class DistutilsInternalError(DistutilsError):
+class DistutilsInternalError (DistutilsError):
     """Internal inconsistencies or impossibilities (obviously, this
     should never be seen if the code is working!)."""
+    pass
 
-class DistutilsTemplateError(DistutilsError):
+class DistutilsTemplateError (DistutilsError):
     """Syntax error in a file list template."""
 
 class DistutilsByteCompileError(DistutilsError):
     """Byte compile error."""
 
 # Exception classes used by the CCompiler implementation classes
-class CCompilerError(Exception):
+class CCompilerError (Exception):
     """Some compile/link operation failed."""
 
-class PreprocessError(CCompilerError):
+class PreprocessError (CCompilerError):
     """Failure to preprocess one or more C/C++ files."""
 
-class CompileError(CCompilerError):
+class CompileError (CCompilerError):
     """Failure to compile one or more C/C++ source files."""
 
-class LibError(CCompilerError):
+class LibError (CCompilerError):
     """Failure to create a static library from one or more C/C++ object
     files."""
 
-class LinkError(CCompilerError):
+class LinkError (CCompilerError):
     """Failure to link one or more C/C++ object files into an executable
     or shared library file."""
 
-class UnknownFileError(CCompilerError):
+class UnknownFileError (CCompilerError):
     """Attempt to process an unknown file type."""

Modified: python/branches/py3k/Lib/distutils/extension.py
==============================================================================
--- python/branches/py3k/Lib/distutils/extension.py	(original)
+++ python/branches/py3k/Lib/distutils/extension.py	Thu Jul 22 14:50:05 2010
@@ -6,6 +6,7 @@
 __revision__ = "$Id$"
 
 import os
+import sys
 import warnings
 
 # This class is really only used by the "build_ext" command, so it might
@@ -134,17 +135,14 @@
 
 def read_setup_file(filename):
     """Reads a Setup file and returns Extension instances."""
-    warnings.warn('distutils.extensions.read_setup_file is deprecated. '
-                  'It will be removed in the next Python release.')
-    _sysconfig = __import__('sysconfig')
-    from distutils.sysconfig import (expand_makefile_vars,
+    from distutils.sysconfig import (parse_makefile, expand_makefile_vars,
                                      _variable_rx)
 
     from distutils.text_file import TextFile
     from distutils.util import split_quoted
 
     # First pass over the file to gather "VAR = VALUE" assignments.
-    vars = _sysconfig._parse_makefile(filename)
+    vars = parse_makefile(filename)
 
     # Second pass to gobble up the real content: lines of the form
     #   <module> ... [<sourcefile> ...] [<cpparg> ...] [<library> ...]
@@ -164,10 +162,7 @@
             file.warn("'%s' lines not handled yet" % line)
             continue
 
-        with warnings.catch_warnings():
-            warnings.simplefilter("ignore")
-            line = expand_makefile_vars(line, vars)
-
+        line = expand_makefile_vars(line, vars)
         words = split_quoted(line)
 
         # NB. this parses a slightly different syntax than the old

Modified: python/branches/py3k/Lib/distutils/fancy_getopt.py
==============================================================================
--- python/branches/py3k/Lib/distutils/fancy_getopt.py	(original)
+++ python/branches/py3k/Lib/distutils/fancy_getopt.py	Thu Jul 22 14:50:05 2010
@@ -10,11 +10,9 @@
 
 __revision__ = "$Id$"
 
-import sys
-import string
-import re
+import sys, string, re
 import getopt
-from distutils.errors import DistutilsGetoptError, DistutilsArgError
+from distutils.errors import *
 
 # Much like command_re in distutils.core, this is close to but not quite
 # the same as a Python NAME -- except, in the spirit of most GNU
@@ -446,3 +444,16 @@
         'options' will be initialized to None."""
         for opt in options:
             setattr(self, opt, None)
+
+
+if __name__ == "__main__":
+    text = """\
+Tra-la-la, supercalifragilisticexpialidocious.
+How *do* you spell that odd word, anyways?
+(Someone ask Mary -- she'll know [or she'll
+say, "How should I know?"].)"""
+
+    for w in (10, 20, 30, 40):
+        print("width: %d" % w)
+        print("\n".join(wrap_text(text, w)))
+        print()

Modified: python/branches/py3k/Lib/distutils/file_util.py
==============================================================================
--- python/branches/py3k/Lib/distutils/file_util.py	(original)
+++ python/branches/py3k/Lib/distutils/file_util.py	Thu Jul 22 14:50:05 2010
@@ -10,18 +10,17 @@
 from distutils import log
 
 # for generating verbose output in 'copy_file()'
-_copy_action = {None: 'copying',
-                'hard': 'hard linking',
-                'sym': 'symbolically linking'}
+_copy_action = { None:   'copying',
+                 'hard': 'hard linking',
+                 'sym':  'symbolically linking' }
 
 
 def _copy_file_contents(src, dst, buffer_size=16*1024):
-    """Copy the file 'src' to 'dst'.
-
-    Both must be filenames. Any error opening either file, reading from
-    'src', or writing to 'dst', raises DistutilsFileError.  Data is
-    read/written in chunks of 'buffer_size' bytes (default 16k).  No attempt
-    is made to handle anything apart from regular files.
+    """Copy the file 'src' to 'dst'; both must be filenames.  Any error
+    opening either file, reading from 'src', or writing to 'dst', raises
+    DistutilsFileError.  Data is read/written in chunks of 'buffer_size'
+    bytes (default 16k).  No attempt is made to handle anything apart from
+    regular files.
     """
     # Stolen from shutil module in the standard library, but with
     # custom error-handling added.
@@ -69,16 +68,15 @@
 
 def copy_file(src, dst, preserve_mode=1, preserve_times=1, update=0,
               link=None, verbose=1, dry_run=0):
-    """Copy a file 'src' to 'dst'.
-
-    If 'dst' is a directory, then 'src' is copied there with the same name;
-    otherwise, it must be a filename.  (If the file exists, it will be
-    ruthlessly clobbered.)  If 'preserve_mode' is true (the default),
-    the file's mode (type and permission bits, or whatever is analogous on
-    the current platform) is copied.  If 'preserve_times' is true (the
-    default), the last-modified and last-access times are copied as well.
-    If 'update' is true, 'src' will only be copied if 'dst' does not exist,
-    or if 'dst' does exist but is older than 'src'.
+    """Copy a file 'src' to 'dst'.  If 'dst' is a directory, then 'src' is
+    copied there with the same name; otherwise, it must be a filename.  (If
+    the file exists, it will be ruthlessly clobbered.)  If 'preserve_mode'
+    is true (the default), the file's mode (type and permission bits, or
+    whatever is analogous on the current platform) is copied.  If
+    'preserve_times' is true (the default), the last-modified and
+    last-access times are copied as well.  If 'update' is true, 'src' will
+    only be copied if 'dst' does not exist, or if 'dst' does exist but is
+    older than 'src'.
 
     'link' allows you to make hard links (os.link) or symbolic links
     (os.symlink) instead of copying: set it to "hard" or "sym"; if it is
@@ -132,6 +130,15 @@
     if dry_run:
         return (dst, 1)
 
+    # On Mac OS, use the native file copy routine
+    if os.name == 'mac':
+        import macostools
+        try:
+            macostools.copy(src, dst, 0, preserve_times)
+        except os.error as exc:
+            raise DistutilsFileError(
+                  "could not copy '%s' to '%s': %s" % (src, dst, exc.args[-1]))
+
     # If linking (hard or symbolic), use the appropriate system call
     # (Unix only, of course, but that's the caller's responsibility)
     elif link == 'hard':
@@ -159,12 +166,13 @@
 
 
 # XXX I suspect this is Unix-specific -- need porting help!
-def move_file(src, dst, verbose=1, dry_run=0):
-    """Move a file 'src' to 'dst'.
-
-    If 'dst' is a directory, the file will be moved into it with the same
-    name; otherwise, 'src' is just renamed to 'dst'.  Return the new
-    full name of the file.
+def move_file (src, dst,
+               verbose=1,
+               dry_run=0):
+
+    """Move a file 'src' to 'dst'.  If 'dst' is a directory, the file will
+    be moved into it with the same name; otherwise, 'src' is just renamed
+    to 'dst'.  Return the new full name of the file.
 
     Handles cross-device moves on Unix using 'copy_file()'.  What about
     other systems???
@@ -221,7 +229,7 @@
     return dst
 
 
-def write_file(filename, contents):
+def write_file (filename, contents):
     """Create a file with the specified name and write 'contents' (a
     sequence of strings without line terminators) to it.
     """

Modified: python/branches/py3k/Lib/distutils/filelist.py
==============================================================================
--- python/branches/py3k/Lib/distutils/filelist.py	(original)
+++ python/branches/py3k/Lib/distutils/filelist.py	Thu Jul 22 14:50:05 2010
@@ -108,7 +108,7 @@
         # defined: it's the first word of the line.  Which of the other
         # three are defined depends on the action; it'll be either
         # patterns, (dir and patterns), or (dir_pattern).
-        action, patterns, dir, dir_pattern = self._parse_template_line(line)
+        (action, patterns, dir, dir_pattern) = self._parse_template_line(line)
 
         # OK, now we know that the action is valid and we have the
         # right number of words on the line for that action -- so we
@@ -175,15 +175,15 @@
             raise DistutilsInternalError(
                   "this cannot happen: invalid action '%s'" % action)
 
+
     # -- Filtering/selection methods -----------------------------------
 
     def include_pattern(self, pattern, anchor=1, prefix=None, is_regex=0):
         """Select strings (presumably filenames) from 'self.files' that
-        match 'pattern', a Unix-style wildcard (glob) pattern.
-
-        Patterns are not quite the same as implemented by the 'fnmatch'
-        module: '*' and '?'  match non-special characters, where "special"
-        is platform-dependent: slash on Unix; colon, slash, and backslash on
+        match 'pattern', a Unix-style wildcard (glob) pattern.  Patterns
+        are not quite the same as implemented by the 'fnmatch' module: '*'
+        and '?'  match non-special characters, where "special" is platform-
+        dependent: slash on Unix; colon, slash, and backslash on
         DOS/Windows; and colon on Mac OS.
 
         If 'anchor' is true (the default), then the pattern match is more
@@ -220,13 +220,13 @@
         return files_found
 
 
-    def exclude_pattern(self, pattern, anchor=1, prefix=None, is_regex=0):
+    def exclude_pattern (self, pattern,
+                         anchor=1, prefix=None, is_regex=0):
         """Remove strings (presumably filenames) from 'files' that match
-        'pattern'.
-
-        Other parameters are the same as for 'include_pattern()', above.
-        The list 'self.files' is modified in place. Return 1 if files are
-        found.
+        'pattern'.  Other parameters are the same as for
+        'include_pattern()', above.
+        The list 'self.files' is modified in place.
+        Return True if files are found, False otherwise.
         """
         files_found = False
         pattern_re = translate_pattern(pattern, anchor, prefix, is_regex)
@@ -275,11 +275,10 @@
 
 
 def glob_to_re(pattern):
-    """Translate a shell-like glob pattern to a regular expression.
-
-    Return a string containing the regex.  Differs from
-    'fnmatch.translate()' in that '*' does not match "special characters"
-    (which are platform-specific).
+    """Translate a shell-like glob pattern to a regular expression; return
+    a string containing the regex.  Differs from 'fnmatch.translate()' in
+    that '*' does not match "special characters" (which are
+    platform-specific).
     """
     pattern_re = fnmatch.translate(pattern)
 
@@ -297,9 +296,7 @@
 
 def translate_pattern(pattern, anchor=1, prefix=None, is_regex=0):
     """Translate a shell-like wildcard pattern to a compiled regular
-    expression.
-
-    Return the compiled regex.  If 'is_regex' true,
+    expression.  Return the compiled regex.  If 'is_regex' true,
     then 'pattern' is directly compiled to a regex (if it's a string)
     or just returned as-is (assumes it's a regex object).
     """
@@ -317,7 +314,7 @@
     if prefix is not None:
         # ditch end of pattern character
         empty_pattern = glob_to_re('')
-        prefix_re = glob_to_re(prefix)[:-len(empty_pattern)]
+        prefix_re = (glob_to_re(prefix))[:-len(empty_pattern)]
         pattern_re = "^" + os.path.join(prefix_re, ".*" + pattern_re)
     else:                               # no prefix -- respect anchor flag
         if anchor:

Modified: python/branches/py3k/Lib/distutils/msvc9compiler.py
==============================================================================
--- python/branches/py3k/Lib/distutils/msvc9compiler.py	(original)
+++ python/branches/py3k/Lib/distutils/msvc9compiler.py	Thu Jul 22 14:50:05 2010
@@ -19,9 +19,10 @@
 import sys
 import re
 
-from distutils.errors import (DistutilsExecError, DistutilsPlatformError,
-                              CompileError, LibError, LinkError)
-from distutils.ccompiler import CCompiler, gen_lib_options
+from distutils.errors import DistutilsExecError, DistutilsPlatformError, \
+                             CompileError, LibError, LinkError
+from distutils.ccompiler import CCompiler, gen_preprocess_options, \
+                                gen_lib_options
 from distutils import log
 from distutils.util import get_platform
 

Modified: python/branches/py3k/Lib/distutils/msvccompiler.py
==============================================================================
--- python/branches/py3k/Lib/distutils/msvccompiler.py	(original)
+++ python/branches/py3k/Lib/distutils/msvccompiler.py	Thu Jul 22 14:50:05 2010
@@ -10,12 +10,12 @@
 
 __revision__ = "$Id$"
 
-import sys
-import os
-
-from distutils.errors import (DistutilsExecError, DistutilsPlatformError,
-                              CompileError, LibError, LinkError)
-from distutils.ccompiler import CCompiler, gen_lib_options
+import sys, os
+from distutils.errors import \
+     DistutilsExecError, DistutilsPlatformError, \
+     CompileError, LibError, LinkError
+from distutils.ccompiler import \
+     CCompiler, gen_preprocess_options, gen_lib_options
 from distutils import log
 
 _can_read_reg = False
@@ -124,7 +124,7 @@
                 self.set_macro("FrameworkSDKDir", net, "sdkinstallrootv1.1")
             else:
                 self.set_macro("FrameworkSDKDir", net, "sdkinstallroot")
-        except KeyError:
+        except KeyError as exc: #
             raise DistutilsPlatformError(
             """Python was built with Visual Studio 2003;
 extensions must be built with a compiler than can generate compatible binaries.

Modified: python/branches/py3k/Lib/distutils/sysconfig.py
==============================================================================
--- python/branches/py3k/Lib/distutils/sysconfig.py	(original)
+++ python/branches/py3k/Lib/distutils/sysconfig.py	Thu Jul 22 14:50:05 2010
@@ -7,9 +7,6 @@
 
 Written by:   Fred L. Drake, Jr.
 Email:        <fdrake at acm.org>
-
-**This module has been moved out of Distutils and will be removed from
-Python in the next version (3.3)**
 """
 
 __revision__ = "$Id$"
@@ -17,45 +14,51 @@
 import io
 import os
 import re
-from warnings import warn
-
-from distutils.errors import DistutilsPlatformError
+import sys
 
-# importing sysconfig from Lib
-# to avoid this module to shadow it
-_sysconfig = __import__('sysconfig')
-
-# names defined here to keep backward compatibility
-# for APIs that were relocated
-get_python_version = _sysconfig.get_python_version
-get_config_h_filename = _sysconfig.get_config_h_filename
-parse_config_h = _sysconfig.parse_config_h
-get_config_vars = _sysconfig.get_config_vars
-get_config_var = _sysconfig.get_config_var
-from distutils.ccompiler import customize_compiler
-
-_DEPRECATION_MSG = ("distutils.sysconfig.%s is deprecated. "
-                    "Use the APIs provided by the sysconfig module instead")
-
-def _get_project_base():
-    return _sysconfig._PROJECT_BASE
-
-project_base = _get_project_base()
-
-class _DeprecatedBool(int):
-    def __nonzero__(self):
-        warn(_DEPRECATION_MSG % 'get_python_version', DeprecationWarning)
-        return super(_DeprecatedBool, self).__nonzero__()
+from .errors import DistutilsPlatformError
 
+# These are needed in a couple of spots, so just compute them once.
+PREFIX = os.path.normpath(sys.prefix)
+EXEC_PREFIX = os.path.normpath(sys.exec_prefix)
+
+# Path to the base directory of the project. On Windows the binary may
+# live in project/PCBuild9.  If we're dealing with an x64 Windows build,
+# it'll live in project/PCbuild/amd64.
+project_base = os.path.dirname(os.path.abspath(sys.executable))
+if os.name == "nt" and "pcbuild" in project_base[-8:].lower():
+    project_base = os.path.abspath(os.path.join(project_base, os.path.pardir))
+# PC/VS7.1
+if os.name == "nt" and "\\pc\\v" in project_base[-10:].lower():
+    project_base = os.path.abspath(os.path.join(project_base, os.path.pardir,
+                                                os.path.pardir))
+# PC/AMD64
+if os.name == "nt" and "\\pcbuild\\amd64" in project_base[-14:].lower():
+    project_base = os.path.abspath(os.path.join(project_base, os.path.pardir,
+                                                os.path.pardir))
+
+# python_build: (Boolean) if true, we're either building Python or
+# building an extension with an un-installed Python, so we use
+# different (hard-wired) directories.
+# Setup.local is available for Makefile builds including VPATH builds,
+# Setup.dist is available on Windows
 def _python_build():
-    return _DeprecatedBool(_sysconfig.is_python_build())
-
+    for fn in ("Setup.dist", "Setup.local"):
+        if os.path.isfile(os.path.join(project_base, "Modules", fn)):
+            return True
+    return False
 python_build = _python_build()
 
-def get_python_inc(plat_specific=0, prefix=None):
-    """This function is deprecated.
+def get_python_version():
+    """Return a string containing the major and minor Python version,
+    leaving off the patchlevel.  Sample return values could be '1.5'
+    or '2.2'.
+    """
+    return sys.version[:3]
+
 
-    Return the directory containing installed Python header files.
+def get_python_inc(plat_specific=0, prefix=None):
+    """Return the directory containing installed Python header files.
 
     If 'plat_specific' is false (the default), this is the path to the
     non-platform-specific header files, i.e. Python.h and so on;
@@ -65,22 +68,39 @@
     If 'prefix' is supplied, use it instead of sys.prefix or
     sys.exec_prefix -- i.e., ignore 'plat_specific'.
     """
-    warn(_DEPRECATION_MSG % 'get_python_inc', DeprecationWarning)
-    get_path = _sysconfig.get_path
-
-    if prefix is not None:
-        vars = {'base': prefix}
-        return get_path('include', vars=vars)
-
-    if not plat_specific:
-        return get_path('include')
+    if prefix is None:
+        prefix = plat_specific and EXEC_PREFIX or PREFIX
+    if os.name == "posix":
+        if python_build:
+            # Assume the executable is in the build directory.  The
+            # pyconfig.h file should be in the same directory.  Since
+            # the build directory may not be the source directory, we
+            # must use "srcdir" from the makefile to find the "Include"
+            # directory.
+            base = os.path.dirname(os.path.abspath(sys.executable))
+            if plat_specific:
+                return base
+            else:
+                incdir = os.path.join(get_config_var('srcdir'), 'Include')
+                return os.path.normpath(incdir)
+        return os.path.join(prefix, "include", "python" + get_python_version())
+    elif os.name == "nt":
+        return os.path.join(prefix, "include")
+    elif os.name == "mac":
+        if plat_specific:
+            return os.path.join(prefix, "Mac", "Include")
+        else:
+            return os.path.join(prefix, "Include")
+    elif os.name == "os2":
+        return os.path.join(prefix, "Include")
     else:
-        return get_path('platinclude')
+        raise DistutilsPlatformError(
+            "I don't know where Python installs its C header files "
+            "on platform '%s'" % os.name)
 
-def get_python_lib(plat_specific=False, standard_lib=False, prefix=None):
-    """This function is deprecated.
 
-    Return the directory containing the Python library (standard or
+def get_python_lib(plat_specific=0, standard_lib=0, prefix=None):
+    """Return the directory containing the Python library (standard or
     site additions).
 
     If 'plat_specific' is true, return the directory containing
@@ -93,33 +113,149 @@
     If 'prefix' is supplied, use it instead of sys.prefix or
     sys.exec_prefix -- i.e., ignore 'plat_specific'.
     """
-    warn(_DEPRECATION_MSG % 'get_python_lib', DeprecationWarning)
-    vars = {}
-    get_path = _sysconfig.get_path
-    if prefix is not None:
-        if plat_specific:
-            vars['platbase'] = prefix
+    if prefix is None:
+        prefix = plat_specific and EXEC_PREFIX or PREFIX
+
+    if os.name == "posix":
+        libpython = os.path.join(prefix,
+                                 "lib", "python" + get_python_version())
+        if standard_lib:
+            return libpython
+        else:
+            return os.path.join(libpython, "site-packages")
+    elif os.name == "nt":
+        if standard_lib:
+            return os.path.join(prefix, "Lib")
         else:
-            vars['base'] = prefix
-    if standard_lib:
+            if get_python_version() < "2.2":
+                return prefix
+            else:
+                return os.path.join(prefix, "Lib", "site-packages")
+    elif os.name == "mac":
         if plat_specific:
-            return get_path('platstdlib', vars=vars)
+            if standard_lib:
+                return os.path.join(prefix, "Lib", "lib-dynload")
+            else:
+                return os.path.join(prefix, "Lib", "site-packages")
         else:
-            return get_path('stdlib', vars=vars)
+            if standard_lib:
+                return os.path.join(prefix, "Lib")
+            else:
+                return os.path.join(prefix, "Lib", "site-packages")
+    elif os.name == "os2":
+        if standard_lib:
+            return os.path.join(prefix, "Lib")
+        else:
+            return os.path.join(prefix, "Lib", "site-packages")
     else:
-        if plat_specific:
-            return get_path('platlib', vars=vars)
+        raise DistutilsPlatformError(
+            "I don't know where Python installs its library "
+            "on platform '%s'" % os.name)
+
+
+def customize_compiler(compiler):
+    """Do any platform-specific customization of a CCompiler instance.
+
+    Mainly needed on Unix, so we can plug in the information that
+    varies across Unices and is stored in Python's Makefile.
+    """
+    if compiler.compiler_type == "unix":
+        (cc, cxx, opt, cflags, ccshared, ldshared, so_ext, ar, ar_flags) = \
+            get_config_vars('CC', 'CXX', 'OPT', 'CFLAGS',
+                            'CCSHARED', 'LDSHARED', 'SO', 'AR', 'ARFLAGS')
+
+        if 'CC' in os.environ:
+            cc = os.environ['CC']
+        if 'CXX' in os.environ:
+            cxx = os.environ['CXX']
+        if 'LDSHARED' in os.environ:
+            ldshared = os.environ['LDSHARED']
+        if 'CPP' in os.environ:
+            cpp = os.environ['CPP']
         else:
-            return get_path('purelib', vars=vars)
+            cpp = cc + " -E"           # not always
+        if 'LDFLAGS' in os.environ:
+            ldshared = ldshared + ' ' + os.environ['LDFLAGS']
+        if 'CFLAGS' in os.environ:
+            cflags = opt + ' ' + os.environ['CFLAGS']
+            ldshared = ldshared + ' ' + os.environ['CFLAGS']
+        if 'CPPFLAGS' in os.environ:
+            cpp = cpp + ' ' + os.environ['CPPFLAGS']
+            cflags = cflags + ' ' + os.environ['CPPFLAGS']
+            ldshared = ldshared + ' ' + os.environ['CPPFLAGS']
+        if 'AR' in os.environ:
+            ar = os.environ['AR']
+        if 'ARFLAGS' in os.environ:
+            archiver = ar + ' ' + os.environ['ARFLAGS']
+        else:
+            archiver = ar + ' ' + ar_flags
+
+        cc_cmd = cc + ' ' + cflags
+        compiler.set_executables(
+            preprocessor=cpp,
+            compiler=cc_cmd,
+            compiler_so=cc_cmd + ' ' + ccshared,
+            compiler_cxx=cxx,
+            linker_so=ldshared,
+            linker_exe=cc,
+            archiver=archiver)
+
+        compiler.shared_lib_extension = so_ext
+
+
+def get_config_h_filename():
+    """Return full pathname of installed pyconfig.h file."""
+    if python_build:
+        if os.name == "nt":
+            inc_dir = os.path.join(project_base, "PC")
+        else:
+            inc_dir = project_base
+    else:
+        inc_dir = get_python_inc(plat_specific=1)
+    if get_python_version() < '2.2':
+        config_h = 'config.h'
+    else:
+        # The name of the config.h file changed in 2.2
+        config_h = 'pyconfig.h'
+    return os.path.join(inc_dir, config_h)
+
 
 def get_makefile_filename():
-    """This function is deprecated.
+    """Return full pathname of installed Makefile from the Python build."""
+    if python_build:
+        return os.path.join(os.path.dirname(sys.executable), "Makefile")
+    lib_dir = get_python_lib(plat_specific=1, standard_lib=1)
+    return os.path.join(lib_dir, "config", "Makefile")
+
+
+def parse_config_h(fp, g=None):
+    """Parse a config.h-style file.
 
-    Return full pathname of installed Makefile from the Python build.
+    A dictionary containing name/value pairs is returned.  If an
+    optional dictionary is passed in as the second argument, it is
+    used instead of a new dictionary.
     """
+    if g is None:
+        g = {}
+    define_rx = re.compile("#define ([A-Z][A-Za-z0-9_]+) (.*)\n")
+    undef_rx = re.compile("/[*] #undef ([A-Z][A-Za-z0-9_]+) [*]/\n")
+    #
+    while True:
+        line = fp.readline()
+        if not line:
+            break
+        m = define_rx.match(line)
+        if m:
+            n, v = m.group(1, 2)
+            try: v = int(v)
+            except ValueError: pass
+            g[n] = v
+        else:
+            m = undef_rx.match(line)
+            if m:
+                g[m.group(1)] = 0
+    return g
 
-    warn(_DEPRECATION_MSG % 'get_makefile_filename', DeprecationWarning)
-    return _sysconfig._get_makefile_filename()
 
 # Regexes needed for parsing Makefile (and similar syntaxes,
 # like old-style Setup files).
@@ -128,29 +264,91 @@
 _findvar2_rx = re.compile(r"\${([A-Za-z][A-Za-z0-9_]*)}")
 
 def parse_makefile(fn, g=None):
-    """This function is deprecated.
-
-    Parse a Makefile-style file.
+    """Parse a Makefile-style file.
 
     A dictionary containing name/value pairs is returned.  If an
     optional dictionary is passed in as the second argument, it is
     used instead of a new dictionary.
     """
-    warn(_DEPRECATION_MSG % 'parse_makefile', DeprecationWarning)
-    return _sysconfig._parse_makefile(fn, g)
+    from distutils.text_file import TextFile
+    fp = TextFile(fn, strip_comments=1, skip_blanks=1, join_lines=1)
+
+    if g is None:
+        g = {}
+    done = {}
+    notdone = {}
+
+    while True:
+        line = fp.readline()
+        if line is None: # eof
+            break
+        m = _variable_rx.match(line)
+        if m:
+            n, v = m.group(1, 2)
+            v = v.strip()
+            # `$$' is a literal `$' in make
+            tmpv = v.replace('$$', '')
+
+            if "$" in tmpv:
+                notdone[n] = v
+            else:
+                try:
+                    v = int(v)
+                except ValueError:
+                    # insert literal `$'
+                    done[n] = v.replace('$$', '$')
+                else:
+                    done[n] = v
+
+    # do variable interpolation here
+    while notdone:
+        for name in list(notdone):
+            value = notdone[name]
+            m = _findvar1_rx.search(value) or _findvar2_rx.search(value)
+            if m:
+                n = m.group(1)
+                found = True
+                if n in done:
+                    item = str(done[n])
+                elif n in notdone:
+                    # get it on a subsequent round
+                    found = False
+                elif n in os.environ:
+                    # do it like make: fall back to environment
+                    item = os.environ[n]
+                else:
+                    done[n] = item = ""
+                if found:
+                    after = value[m.end():]
+                    value = value[:m.start()] + item + after
+                    if "$" in after:
+                        notdone[name] = value
+                    else:
+                        try: value = int(value)
+                        except ValueError:
+                            done[name] = value.strip()
+                        else:
+                            done[name] = value
+                        del notdone[name]
+            else:
+                # bogus variable reference; just drop it since we can't deal
+                del notdone[name]
+
+    fp.close()
+
+    # save the results in the global dictionary
+    g.update(done)
+    return g
 
-def expand_makefile_vars(s, vars):
-    """This function is deprecated.
 
-    Expand Makefile-style variables -- "${foo}" or "$(foo)" -- in
+def expand_makefile_vars(s, vars):
+    """Expand Makefile-style variables -- "${foo}" or "$(foo)" -- in
     'string' according to 'vars' (a dictionary mapping variable names to
     values).  Variables not present in 'vars' are silently expanded to the
     empty string.  The variable values in 'vars' should not contain further
     variable expansions; if 'vars' is the output of 'parse_makefile()',
     you're fine.  Returns a variable-expanded version of 's'.
     """
-    warn('this function will be removed in then next version of Python',
-         DeprecationWarning)
 
     # This algorithm does multiple expansion, so if vars['foo'] contains
     # "${bar}", it will expand ${foo} to ${bar}, and then expand
@@ -166,3 +364,220 @@
         else:
             break
     return s
+
+
+_config_vars = None
+
+def _init_posix():
+    """Initialize the module as appropriate for POSIX systems."""
+    g = {}
+    # load the installed Makefile:
+    try:
+        filename = get_makefile_filename()
+        parse_makefile(filename, g)
+    except IOError as msg:
+        my_msg = "invalid Python installation: unable to open %s" % filename
+        if hasattr(msg, "strerror"):
+            my_msg = my_msg + " (%s)" % msg.strerror
+
+        raise DistutilsPlatformError(my_msg)
+
+    # load the installed pyconfig.h:
+    try:
+        filename = get_config_h_filename()
+        parse_config_h(io.open(filename), g)
+    except IOError as msg:
+        my_msg = "invalid Python installation: unable to open %s" % filename
+        if hasattr(msg, "strerror"):
+            my_msg = my_msg + " (%s)" % msg.strerror
+
+        raise DistutilsPlatformError(my_msg)
+
+    # On MacOSX we need to check the setting of the environment variable
+    # MACOSX_DEPLOYMENT_TARGET: configure bases some choices on it so
+    # it needs to be compatible.
+    # If it isn't set we set it to the configure-time value
+    if sys.platform == 'darwin' and 'MACOSX_DEPLOYMENT_TARGET' in g:
+        cfg_target = g['MACOSX_DEPLOYMENT_TARGET']
+        cur_target = os.getenv('MACOSX_DEPLOYMENT_TARGET', '')
+        if cur_target == '':
+            cur_target = cfg_target
+            os.putenv('MACOSX_DEPLOYMENT_TARGET', cfg_target)
+        elif [int(x) for x in cfg_target.split('.')] > [int(x) for x in cur_target.split('.')]:
+            my_msg = ('$MACOSX_DEPLOYMENT_TARGET mismatch: now "%s" but "%s" during configure'
+                % (cur_target, cfg_target))
+            raise DistutilsPlatformError(my_msg)
+
+    # On AIX, there are wrong paths to the linker scripts in the Makefile
+    # -- these paths are relative to the Python source, but when installed
+    # the scripts are in another directory.
+    if python_build:
+        g['LDSHARED'] = g['BLDSHARED']
+
+    elif get_python_version() < '2.1':
+        # The following two branches are for 1.5.2 compatibility.
+        if sys.platform == 'aix4':          # what about AIX 3.x ?
+            # Linker script is in the config directory, not in Modules as the
+            # Makefile says.
+            python_lib = get_python_lib(standard_lib=1)
+            ld_so_aix = os.path.join(python_lib, 'config', 'ld_so_aix')
+            python_exp = os.path.join(python_lib, 'config', 'python.exp')
+
+            g['LDSHARED'] = "%s %s -bI:%s" % (ld_so_aix, g['CC'], python_exp)
+
+    global _config_vars
+    _config_vars = g
+
+
+def _init_nt():
+    """Initialize the module as appropriate for NT"""
+    g = {}
+    # set basic install directories
+    g['LIBDEST'] = get_python_lib(plat_specific=0, standard_lib=1)
+    g['BINLIBDEST'] = get_python_lib(plat_specific=1, standard_lib=1)
+
+    # XXX hmmm.. a normal install puts include files here
+    g['INCLUDEPY'] = get_python_inc(plat_specific=0)
+
+    g['SO'] = '.pyd'
+    g['EXE'] = ".exe"
+    g['VERSION'] = get_python_version().replace(".", "")
+    g['BINDIR'] = os.path.dirname(os.path.abspath(sys.executable))
+
+    global _config_vars
+    _config_vars = g
+
+
+def _init_mac():
+    """Initialize the module as appropriate for Macintosh systems"""
+    g = {}
+    # set basic install directories
+    g['LIBDEST'] = get_python_lib(plat_specific=0, standard_lib=1)
+    g['BINLIBDEST'] = get_python_lib(plat_specific=1, standard_lib=1)
+
+    # XXX hmmm.. a normal install puts include files here
+    g['INCLUDEPY'] = get_python_inc(plat_specific=0)
+
+    import MacOS
+    if not hasattr(MacOS, 'runtimemodel'):
+        g['SO'] = '.ppc.slb'
+    else:
+        g['SO'] = '.%s.slb' % MacOS.runtimemodel
+
+    # XXX are these used anywhere?
+    g['install_lib'] = os.path.join(EXEC_PREFIX, "Lib")
+    g['install_platlib'] = os.path.join(EXEC_PREFIX, "Mac", "Lib")
+
+    # These are used by the extension module build
+    g['srcdir'] = ':'
+    global _config_vars
+    _config_vars = g
+
+
+def _init_os2():
+    """Initialize the module as appropriate for OS/2"""
+    g = {}
+    # set basic install directories
+    g['LIBDEST'] = get_python_lib(plat_specific=0, standard_lib=1)
+    g['BINLIBDEST'] = get_python_lib(plat_specific=1, standard_lib=1)
+
+    # XXX hmmm.. a normal install puts include files here
+    g['INCLUDEPY'] = get_python_inc(plat_specific=0)
+
+    g['SO'] = '.pyd'
+    g['EXE'] = ".exe"
+
+    global _config_vars
+    _config_vars = g
+
+
+def get_config_vars(*args):
+    """With no arguments, return a dictionary of all configuration
+    variables relevant for the current platform.  Generally this includes
+    everything needed to build extensions and install both pure modules and
+    extensions.  On Unix, this means every variable defined in Python's
+    installed Makefile; on Windows and Mac OS it's a much smaller set.
+
+    With arguments, return a list of values that result from looking up
+    each argument in the configuration variable dictionary.
+    """
+    global _config_vars
+    if _config_vars is None:
+        func = globals().get("_init_" + os.name)
+        if func:
+            func()
+        else:
+            _config_vars = {}
+
+        # Normalized versions of prefix and exec_prefix are handy to have;
+        # in fact, these are the standard versions used most places in the
+        # Distutils.
+        _config_vars['prefix'] = PREFIX
+        _config_vars['exec_prefix'] = EXEC_PREFIX
+
+        # Convert srcdir into an absolute path if it appears necessary.
+        # Normally it is relative to the build directory.  However, during
+        # testing, for example, we might be running a non-installed python
+        # from a different directory.
+        if python_build and os.name == "posix":
+            base = os.path.dirname(os.path.abspath(sys.executable))
+            if (not os.path.isabs(_config_vars['srcdir']) and
+                base != os.getcwd()):
+                # srcdir is relative and we are not in the same directory
+                # as the executable. Assume executable is in the build
+                # directory and make srcdir absolute.
+                srcdir = os.path.join(base, _config_vars['srcdir'])
+                _config_vars['srcdir'] = os.path.normpath(srcdir)
+
+        if sys.platform == 'darwin':
+            kernel_version = os.uname()[2] # Kernel version (8.4.3)
+            major_version = int(kernel_version.split('.')[0])
+
+            if major_version < 8:
+                # On Mac OS X before 10.4, check if -arch and -isysroot
+                # are in CFLAGS or LDFLAGS and remove them if they are.
+                # This is needed when building extensions on a 10.3 system
+                # using a universal build of python.
+                for key in ('LDFLAGS', 'BASECFLAGS',
+                        # a number of derived variables. These need to be
+                        # patched up as well.
+                        'CFLAGS', 'PY_CFLAGS', 'BLDSHARED'):
+                    flags = _config_vars[key]
+                    flags = re.sub('-arch\s+\w+\s', ' ', flags, re.ASCII)
+                    flags = re.sub('-isysroot [^ \t]*', ' ', flags)
+                    _config_vars[key] = flags
+
+            else:
+
+                # Allow the user to override the architecture flags using
+                # an environment variable.
+                # NOTE: This name was introduced by Apple in OSX 10.5 and
+                # is used by several scripting languages distributed with
+                # that OS release.
+
+                if 'ARCHFLAGS' in os.environ:
+                    arch = os.environ['ARCHFLAGS']
+                    for key in ('LDFLAGS', 'BASECFLAGS',
+                        # a number of derived variables. These need to be
+                        # patched up as well.
+                        'CFLAGS', 'PY_CFLAGS', 'BLDSHARED'):
+
+                        flags = _config_vars[key]
+                        flags = re.sub('-arch\s+\w+\s', ' ', flags)
+                        flags = flags + ' ' + arch
+                        _config_vars[key] = flags
+
+    if args:
+        vals = []
+        for name in args:
+            vals.append(_config_vars.get(name))
+        return vals
+    else:
+        return _config_vars
+
+def get_config_var(name):
+    """Return the value of a single variable using the dictionary
+    returned by 'get_config_vars()'.  Equivalent to
+    get_config_vars().get(name)
+    """
+    return get_config_vars().get(name)

Modified: python/branches/py3k/Lib/distutils/tests/support.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/support.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/support.py	Thu Jul 22 14:50:05 2010
@@ -3,19 +3,11 @@
 import shutil
 import tempfile
 from copy import deepcopy
-import warnings
 
 from distutils import log
 from distutils.log import DEBUG, INFO, WARN, ERROR, FATAL
 from distutils.core import Distribution
 
-def capture_warnings(func):
-    def _capture_warnings(*args, **kw):
-        with warnings.catch_warnings():
-            warnings.simplefilter("ignore")
-            return func(*args, **kw)
-    return _capture_warnings
-
 class LoggingSilencer(object):
 
     def setUp(self):
@@ -63,8 +55,6 @@
         super().tearDown()
         while self.tempdirs:
             d = self.tempdirs.pop()
-            if not os.path.exists(d):
-                continue
             shutil.rmtree(d, os.name in ('nt', 'cygwin'))
 
     def mkdtemp(self):

Modified: python/branches/py3k/Lib/distutils/tests/test_archive_util.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_archive_util.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/test_archive_util.py	Thu Jul 22 14:50:05 2010
@@ -15,30 +15,15 @@
 from test.support import check_warnings
 
 try:
-    import grp
-    import pwd
-    UID_GID_SUPPORT = True
-except ImportError:
-    UID_GID_SUPPORT = False
-
-try:
     import zipfile
     ZIP_SUPPORT = True
 except ImportError:
     ZIP_SUPPORT = find_executable('zip')
 
-# some tests will fail if zlib is not available
-try:
-    import zlib
-except ImportError:
-    zlib = None
-
-
 class ArchiveUtilTestCase(support.TempdirManager,
                           support.LoggingSilencer,
                           unittest.TestCase):
 
-    @unittest.skipUnless(zlib, "requires zlib")
     def test_make_tarball(self):
         # creating something to tar
         tmpdir = self.mkdtemp()
@@ -49,7 +34,7 @@
 
         tmpdir2 = self.mkdtemp()
         unittest.skipUnless(splitdrive(tmpdir)[0] == splitdrive(tmpdir2)[0],
-                            "source and target should be on same drive")
+                            "Source and target should be on same drive")
 
         base_name = os.path.join(tmpdir2, 'archive')
 
@@ -99,7 +84,6 @@
         base_name = os.path.join(tmpdir2, 'archive')
         return tmpdir, tmpdir2, base_name
 
-    @unittest.skipUnless(zlib, "Requires zlib")
     @unittest.skipUnless(find_executable('tar') and find_executable('gzip'),
                          'Need the tar command to run')
     def test_tarfile_vs_tar(self):
@@ -185,7 +169,6 @@
         self.assertTrue(not os.path.exists(tarball))
         self.assertEquals(len(w.warnings), 1)
 
-    @unittest.skipUnless(zlib, "Requires zlib")
     @unittest.skipUnless(ZIP_SUPPORT, 'Need zip support to run')
     def test_make_zipfile(self):
         # creating something to tar
@@ -210,59 +193,6 @@
         base_name = os.path.join(tmpdir, 'archive')
         self.assertRaises(ValueError, make_archive, base_name, 'xxx')
 
-    @unittest.skipUnless(zlib, "Requires zlib")
-    def test_make_archive_owner_group(self):
-        # testing make_archive with owner and group, with various combinations
-        # this works even if there's not gid/uid support
-        if UID_GID_SUPPORT:
-            group = grp.getgrgid(0)[0]
-            owner = pwd.getpwuid(0)[0]
-        else:
-            group = owner = 'root'
-
-        base_dir, root_dir, base_name =  self._create_files()
-        base_name = os.path.join(self.mkdtemp() , 'archive')
-        res = make_archive(base_name, 'zip', root_dir, base_dir, owner=owner,
-                           group=group)
-        self.assertTrue(os.path.exists(res))
-
-        res = make_archive(base_name, 'zip', root_dir, base_dir)
-        self.assertTrue(os.path.exists(res))
-
-        res = make_archive(base_name, 'tar', root_dir, base_dir,
-                           owner=owner, group=group)
-        self.assertTrue(os.path.exists(res))
-
-        res = make_archive(base_name, 'tar', root_dir, base_dir,
-                           owner='kjhkjhkjg', group='oihohoh')
-        self.assertTrue(os.path.exists(res))
-
-    @unittest.skipUnless(zlib, "Requires zlib")
-    @unittest.skipUnless(UID_GID_SUPPORT, "Requires grp and pwd support")
-    def test_tarfile_root_owner(self):
-        tmpdir, tmpdir2, base_name =  self._create_files()
-        old_dir = os.getcwd()
-        os.chdir(tmpdir)
-        group = grp.getgrgid(0)[0]
-        owner = pwd.getpwuid(0)[0]
-        try:
-            archive_name = make_tarball(base_name, 'dist', compress=None,
-                                        owner=owner, group=group)
-        finally:
-            os.chdir(old_dir)
-
-        # check if the compressed tarball was created
-        self.assertTrue(os.path.exists(archive_name))
-
-        # now checks the rights
-        archive = tarfile.open(archive_name)
-        try:
-            for member in archive.getmembers():
-                self.assertEquals(member.uid, 0)
-                self.assertEquals(member.gid, 0)
-        finally:
-            archive.close()
-
     def test_make_archive_cwd(self):
         current_dir = os.getcwd()
         def _breaks(*args, **kw):

Modified: python/branches/py3k/Lib/distutils/tests/test_bdist.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_bdist.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/test_bdist.py	Thu Jul 22 14:50:05 2010
@@ -5,8 +5,6 @@
 import tempfile
 import shutil
 
-from test.support import run_unittest
-
 from distutils.core import Distribution
 from distutils.command.bdist import bdist
 from distutils.tests import support
@@ -42,4 +40,4 @@
     return unittest.makeSuite(BuildTestCase)
 
 if __name__ == '__main__':
-    run_unittest(test_suite())
+    test_support.run_unittest(test_suite())

Modified: python/branches/py3k/Lib/distutils/tests/test_bdist_dumb.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_bdist_dumb.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/test_bdist_dumb.py	Thu Jul 22 14:50:05 2010
@@ -4,15 +4,6 @@
 import sys
 import os
 
-# zlib is not used here, but if it's not available
-# test_simple_built will fail
-try:
-    import zlib
-except ImportError:
-    zlib = None
-
-from test.support import run_unittest
-
 from distutils.core import Distribution
 from distutils.command.bdist_dumb import bdist_dumb
 from distutils.tests import support
@@ -42,7 +33,6 @@
         sys.argv[:] = self.old_sys_argv[1]
         super(BuildDumbTestCase, self).tearDown()
 
-    @unittest.skipUnless(zlib, "requires zlib")
     def test_simple_built(self):
 
         # let's create a simple package
@@ -83,23 +73,8 @@
         # now let's check what we have in the zip file
         # XXX to be done
 
-    def test_finalize_options(self):
-        pkg_dir, dist = self.create_dist()
-        os.chdir(pkg_dir)
-        cmd = bdist_dumb(dist)
-        self.assertEquals(cmd.bdist_dir, None)
-        cmd.finalize_options()
-
-        # bdist_dir is initialized to bdist_base/dumb if not set
-        base = cmd.get_finalized_command('bdist').bdist_base
-        self.assertEquals(cmd.bdist_dir, os.path.join(base, 'dumb'))
-
-        # the format is set to a default value depending on the os.name
-        default = cmd.default_format[os.name]
-        self.assertEquals(cmd.format, default)
-
 def test_suite():
     return unittest.makeSuite(BuildDumbTestCase)
 
 if __name__ == '__main__':
-    run_unittest(test_suite())
+    test_support.run_unittest(test_suite())

Modified: python/branches/py3k/Lib/distutils/tests/test_bdist_rpm.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_bdist_rpm.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/test_bdist_rpm.py	Thu Jul 22 14:50:05 2010
@@ -6,8 +6,6 @@
 import tempfile
 import shutil
 
-from test.support import run_unittest
-
 from distutils.core import Distribution
 from distutils.command.bdist_rpm import bdist_rpm
 from distutils.tests import support
@@ -124,4 +122,4 @@
     return unittest.makeSuite(BuildRpmTestCase)
 
 if __name__ == '__main__':
-    run_unittest(test_suite())
+    test_support.run_unittest(test_suite())

Modified: python/branches/py3k/Lib/distutils/tests/test_bdist_wininst.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_bdist_wininst.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/test_bdist_wininst.py	Thu Jul 22 14:50:05 2010
@@ -1,8 +1,6 @@
 """Tests for distutils.command.bdist_wininst."""
 import unittest
 
-from test.support import run_unittest
-
 from distutils.command.bdist_wininst import bdist_wininst
 from distutils.tests import support
 
@@ -29,4 +27,4 @@
     return unittest.makeSuite(BuildWinInstTestCase)
 
 if __name__ == '__main__':
-    run_unittest(test_suite())
+    test_support.run_unittest(test_suite())

Modified: python/branches/py3k/Lib/distutils/tests/test_build_clib.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_build_clib.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/test_build_clib.py	Thu Jul 22 14:50:05 2010
@@ -120,7 +120,8 @@
         # before we run the command, we want to make sure
         # all commands are present on the system
         # by creating a compiler and checking its executables
-        from distutils.ccompiler import new_compiler, customize_compiler
+        from distutils.ccompiler import new_compiler
+        from distutils.sysconfig import customize_compiler
 
         compiler = new_compiler()
         customize_compiler(compiler)

Modified: python/branches/py3k/Lib/distutils/tests/test_build_ext.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_build_ext.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/test_build_ext.py	Thu Jul 22 14:50:05 2010
@@ -3,13 +3,10 @@
 import tempfile
 import shutil
 from io import StringIO
-import warnings
-from test.support import check_warnings
-from test.support import captured_stdout
 
 from distutils.core import Extension, Distribution
 from distutils.command.build_ext import build_ext
-import sysconfig
+from distutils import sysconfig
 from distutils.tests.support import TempdirManager
 from distutils.tests.support import LoggingSilencer
 from distutils.extension import Extension
@@ -25,23 +22,19 @@
 
 def _get_source_filename():
     srcdir = sysconfig.get_config_var('srcdir')
-    if srcdir is None:
-        return os.path.join(sysconfig.project_base, 'Modules', 'xxmodule.c')
     return os.path.join(srcdir, 'Modules', 'xxmodule.c')
 
-_XX_MODULE_PATH = _get_source_filename()
-
-class BuildExtTestCase(TempdirManager, LoggingSilencer, unittest.TestCase):
-
+class BuildExtTestCase(TempdirManager,
+                       LoggingSilencer,
+                       unittest.TestCase):
     def setUp(self):
         # Create a simple test environment
         # Note that we're making changes to sys.path
         super(BuildExtTestCase, self).setUp()
         self.tmp_dir = self.mkdtemp()
-        if os.path.exists(_XX_MODULE_PATH):
-            self.sys_path = sys.path[:]
-            sys.path.append(self.tmp_dir)
-            shutil.copy(_XX_MODULE_PATH, self.tmp_dir)
+        self.sys_path = sys.path, sys.path[:]
+        sys.path.append(self.tmp_dir)
+        shutil.copy(_get_source_filename(), self.tmp_dir)
         if sys.version > "2.6":
             import site
             self.old_user_base = site.USER_BASE
@@ -49,19 +42,6 @@
             from distutils.command import build_ext
             build_ext.USER_BASE = site.USER_BASE
 
-    def tearDown(self):
-        # Get everything back to normal
-        if os.path.exists(_XX_MODULE_PATH):
-            support.unload('xx')
-            sys.path[:] = self.sys_path
-            # XXX on Windows the test leaves a directory
-            # with xx module in TEMP
-        shutil.rmtree(self.tmp_dir, os.name == 'nt' or
-                                    sys.platform == 'cygwin')
-        super(BuildExtTestCase, self).tearDown()
-
-    @unittest.skipIf(not os.path.exists(_XX_MODULE_PATH),
-                     'xxmodule.c not found')
     def test_build_ext(self):
         global ALREADY_TESTED
         xx_c = os.path.join(self.tmp_dir, 'xxmodule.c')
@@ -104,23 +84,35 @@
         self.assertTrue(isinstance(xx.Null(), xx.Null))
         self.assertTrue(isinstance(xx.Str(), xx.Str))
 
+    def tearDown(self):
+        # Get everything back to normal
+        support.unload('xx')
+        sys.path = self.sys_path[0]
+        sys.path[:] = self.sys_path[1]
+        if sys.version > "2.6":
+            import site
+            site.USER_BASE = self.old_user_base
+            from distutils.command import build_ext
+            build_ext.USER_BASE = self.old_user_base
+        super(BuildExtTestCase, self).tearDown()
+
     def test_solaris_enable_shared(self):
         dist = Distribution({'name': 'xx'})
         cmd = build_ext(dist)
         old = sys.platform
 
         sys.platform = 'sunos' # fooling finalize_options
-        from sysconfig import _CONFIG_VARS
-        old_var = _CONFIG_VARS.get('Py_ENABLE_SHARED')
-        _CONFIG_VARS['Py_ENABLE_SHARED'] = 1
+        from distutils.sysconfig import  _config_vars
+        old_var = _config_vars.get('Py_ENABLE_SHARED')
+        _config_vars['Py_ENABLE_SHARED'] = 1
         try:
             cmd.ensure_finalized()
         finally:
             sys.platform = old
             if old_var is None:
-                del _CONFIG_VARS['Py_ENABLE_SHARED']
+                del _config_vars['Py_ENABLE_SHARED']
             else:
-                _CONFIG_VARS['Py_ENABLE_SHARED'] = old_var
+                _config_vars['Py_ENABLE_SHARED'] = old_var
 
         # make sure we get some library dirs under solaris
         self.assertTrue(len(cmd.library_dirs) > 0)
@@ -182,10 +174,11 @@
         cmd = build_ext(dist)
         cmd.finalize_options()
 
-        py_include = sysconfig.get_path('include')
+        from distutils import sysconfig
+        py_include = sysconfig.get_python_inc()
         self.assertTrue(py_include in cmd.include_dirs)
 
-        plat_py_include = sysconfig.get_path('platinclude')
+        plat_py_include = sysconfig.get_python_inc(plat_specific=1)
         self.assertTrue(plat_py_include in cmd.include_dirs)
 
         # make sure cmd.libraries is turned into a list
@@ -336,6 +329,7 @@
         self.assertEquals(so_dir, other_tmp_dir)
 
         cmd.inplace = 0
+        cmd.compiler = None
         cmd.run()
         so_file = cmd.get_outputs()[0]
         self.assertTrue(os.path.exists(so_file))
@@ -404,26 +398,6 @@
         wanted = os.path.join(curdir, 'twisted', 'runner', 'portmap' + ext)
         self.assertEquals(wanted, path)
 
-    def test_compiler_deprecation_warning(self):
-        dist = Distribution()
-        cmd = build_ext(dist)
-
-        class MyCompiler(object):
-            def do_something(self):
-                pass
-
-        with check_warnings() as w:
-            warnings.simplefilter("always")
-            cmd.compiler = MyCompiler()
-            self.assertEquals(len(w.warnings), 1)
-            cmd.compile = 'unix'
-            self.assertEquals(len(w.warnings), 1)
-            cmd.compiler = MyCompiler()
-            cmd.compiler.do_something()
-            # two more warnings genereated by the get
-            # and the set
-            self.assertEquals(len(w.warnings), 3)
-
 def test_suite():
     src = _get_source_filename()
     if not os.path.exists(src):

Modified: python/branches/py3k/Lib/distutils/tests/test_build_py.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_build_py.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/test_build_py.py	Thu Jul 22 14:50:05 2010
@@ -16,7 +16,7 @@
                       support.LoggingSilencer,
                       unittest.TestCase):
 
-    def _setup_package_data(self):
+    def test_package_data(self):
         sources = self.mkdtemp()
         f = open(os.path.join(sources, "__init__.py"), "w")
         f.write("# Pretend this is a package.")
@@ -52,18 +52,9 @@
         self.assertEqual(len(cmd.get_outputs()), 3)
         pkgdest = os.path.join(destination, "pkg")
         files = os.listdir(pkgdest)
-        return files
-
-    def test_package_data(self):
-        files = self._setup_package_data()
         self.assertTrue("__init__.py" in files)
-        self.assertTrue("README.txt" in files)
-
-    @unittest.skipIf(sys.flags.optimize >= 2,
-                     "pyc files are not written with -O2 and above")
-    def test_package_data_pyc(self):
-        files = self._setup_package_data()
         self.assertTrue("__init__.pyc" in files)
+        self.assertTrue("README.txt" in files)
 
     def test_empty_package_dir (self):
         # See SF 1668596/1720897.

Modified: python/branches/py3k/Lib/distutils/tests/test_build_scripts.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_build_scripts.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/test_build_scripts.py	Thu Jul 22 14:50:05 2010
@@ -5,7 +5,7 @@
 
 from distutils.command.build_scripts import build_scripts
 from distutils.core import Distribution
-import sysconfig
+from distutils import sysconfig
 
 from distutils.tests import support
 
@@ -91,12 +91,12 @@
         # --with-suffix=3`, python is compiled okay but the build scripts
         # failed when writing the name of the executable
         old = sysconfig.get_config_vars().get('VERSION')
-        sysconfig._CONFIG_VARS['VERSION'] = 4
+        sysconfig._config_vars['VERSION'] = 4
         try:
             cmd.run()
         finally:
             if old is not None:
-                sysconfig._CONFIG_VARS['VERSION'] = old
+                sysconfig._config_vars['VERSION'] = old
 
         built = os.listdir(target)
         for name in expected:

Deleted: python/branches/py3k/Lib/distutils/tests/test_ccompiler.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_ccompiler.py	Thu Jul 22 14:50:05 2010
+++ (empty file)
@@ -1,81 +0,0 @@
-"""Tests for distutils.ccompiler."""
-import os
-import unittest
-from test.support import captured_stdout
-
-from distutils.ccompiler import (gen_lib_options, CCompiler,
-                                 get_default_compiler, customize_compiler)
-from distutils import debug
-from distutils.tests import support
-
-class FakeCompiler(object):
-    def library_dir_option(self, dir):
-        return "-L" + dir
-
-    def runtime_library_dir_option(self, dir):
-        return ["-cool", "-R" + dir]
-
-    def find_library_file(self, dirs, lib, debug=0):
-        return 'found'
-
-    def library_option(self, lib):
-        return "-l" + lib
-
-class CCompilerTestCase(support.EnvironGuard, unittest.TestCase):
-
-    def test_gen_lib_options(self):
-        compiler = FakeCompiler()
-        libdirs = ['lib1', 'lib2']
-        runlibdirs = ['runlib1']
-        libs = [os.path.join('dir', 'name'), 'name2']
-
-        opts = gen_lib_options(compiler, libdirs, runlibdirs, libs)
-        wanted = ['-Llib1', '-Llib2', '-cool', '-Rrunlib1', 'found',
-                  '-lname2']
-        self.assertEquals(opts, wanted)
-
-    def test_debug_print(self):
-
-        class MyCCompiler(CCompiler):
-            executables = {}
-
-        compiler = MyCCompiler()
-        with captured_stdout() as stdout:
-            compiler.debug_print('xxx')
-        stdout.seek(0)
-        self.assertEquals(stdout.read(), '')
-
-        debug.DEBUG = True
-        try:
-            with captured_stdout() as stdout:
-                compiler.debug_print('xxx')
-            stdout.seek(0)
-            self.assertEquals(stdout.read(), 'xxx\n')
-        finally:
-            debug.DEBUG = False
-
-    def test_customize_compiler(self):
-
-        # not testing if default compiler is not unix
-        if get_default_compiler() != 'unix':
-            return
-
-        os.environ['AR'] = 'my_ar'
-        os.environ['ARFLAGS'] = '-arflags'
-
-        # make sure AR gets caught
-        class compiler:
-            compiler_type = 'unix'
-
-            def set_executables(self, **kw):
-                self.exes = kw
-
-        comp = compiler()
-        customize_compiler(comp)
-        self.assertEquals(comp.exes['archiver'], 'my_ar -arflags')
-
-def test_suite():
-    return unittest.makeSuite(CCompilerTestCase)
-
-if __name__ == "__main__":
-    unittest.main(defaultTest="test_suite")

Modified: python/branches/py3k/Lib/distutils/tests/test_cmd.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_cmd.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/test_cmd.py	Thu Jul 22 14:50:05 2010
@@ -1,7 +1,7 @@
 """Tests for distutils.cmd."""
 import unittest
 import os
-from test.support import captured_stdout, run_unittest
+from test.support import captured_stdout
 
 from distutils.cmd import Command
 from distutils.dist import Distribution
@@ -124,4 +124,4 @@
     return unittest.makeSuite(CommandTestCase)
 
 if __name__ == '__main__':
-    run_unittest(test_suite())
+    test_support.run_unittest(test_suite())

Modified: python/branches/py3k/Lib/distutils/tests/test_cygwinccompiler.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_cygwinccompiler.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/test_cygwinccompiler.py	Thu Jul 22 14:50:05 2010
@@ -2,21 +2,29 @@
 import unittest
 import sys
 import os
+from io import BytesIO
 import subprocess
-import warnings
-import sysconfig
-
-from test.support import check_warnings, run_unittest
-from test.support import captured_stdout
 
 from distutils import cygwinccompiler
 from distutils.cygwinccompiler import (CygwinCCompiler, check_config_h,
                                        CONFIG_H_OK, CONFIG_H_NOTOK,
                                        CONFIG_H_UNCERTAIN, get_versions,
-                                       get_msvcr, RE_VERSION)
-from distutils.util import get_compiler_versions
+                                       get_msvcr)
 from distutils.tests import support
 
+class FakePopen(object):
+    test_class = None
+
+    def __init__(self, cmd, shell, stdout):
+        self.cmd = cmd.split()[0]
+        exes = self.test_class._exes
+        if self.cmd in exes:
+            # issue #6438 in Python 3.x, Popen returns bytes
+            self.stdout = BytesIO(exes[self.cmd])
+        else:
+            self.stdout = os.popen(cmd, 'r')
+
+
 class CygwinCCompilerTestCase(support.TempdirManager,
                               unittest.TestCase):
 
@@ -24,17 +32,32 @@
         super(CygwinCCompilerTestCase, self).setUp()
         self.version = sys.version
         self.python_h = os.path.join(self.mkdtemp(), 'python.h')
+        from distutils import sysconfig
         self.old_get_config_h_filename = sysconfig.get_config_h_filename
         sysconfig.get_config_h_filename = self._get_config_h_filename
+        self.old_find_executable = cygwinccompiler.find_executable
+        cygwinccompiler.find_executable = self._find_executable
+        self._exes = {}
+        self.old_popen = cygwinccompiler.Popen
+        FakePopen.test_class = self
+        cygwinccompiler.Popen = FakePopen
 
     def tearDown(self):
         sys.version = self.version
+        from distutils import sysconfig
         sysconfig.get_config_h_filename = self.old_get_config_h_filename
+        cygwinccompiler.find_executable = self.old_find_executable
+        cygwinccompiler.Popen = self.old_popen
         super(CygwinCCompilerTestCase, self).tearDown()
 
     def _get_config_h_filename(self):
         return self.python_h
 
+    def _find_executable(self, name):
+        if name in self._exes:
+            return name
+        return None
+
     def test_check_config_h(self):
 
         # check_config_h looks for "GCC" in sys.version first
@@ -58,6 +81,40 @@
         self.write_file(self.python_h, 'xxx __GNUC__ xxx')
         self.assertEquals(check_config_h()[0], CONFIG_H_OK)
 
+    def test_get_versions(self):
+
+        # get_versions calls distutils.spawn.find_executable on
+        # 'gcc', 'ld' and 'dllwrap'
+        self.assertEquals(get_versions(), (None, None, None))
+
+        # Let's fake we have 'gcc' and it returns '3.4.5'
+        self._exes['gcc'] = b'gcc (GCC) 3.4.5 (mingw special)\nFSF'
+        res = get_versions()
+        self.assertEquals(str(res[0]), '3.4.5')
+
+        # and let's see what happens when the version
+        # doesn't match the regular expression
+        # (\d+\.\d+(\.\d+)*)
+        self._exes['gcc'] = b'very strange output'
+        res = get_versions()
+        self.assertEquals(res[0], None)
+
+        # same thing for ld
+        self._exes['ld'] = b'GNU ld version 2.17.50 20060824'
+        res = get_versions()
+        self.assertEquals(str(res[1]), '2.17.50')
+        self._exes['ld'] = b'@(#)PROGRAM:ld  PROJECT:ld64-77'
+        res = get_versions()
+        self.assertEquals(res[1], None)
+
+        # and dllwrap
+        self._exes['dllwrap'] = b'GNU dllwrap 2.17.50 20060824\nFSF'
+        res = get_versions()
+        self.assertEquals(str(res[2]), '2.17.50')
+        self._exes['dllwrap'] = b'Cheese Wrap'
+        res = get_versions()
+        self.assertEquals(res[2], None)
+
     def test_get_msvcr(self):
 
         # none
@@ -90,23 +147,8 @@
                        '[MSC v.1999 32 bits (Intel)]')
         self.assertRaises(ValueError, get_msvcr)
 
-
-    def test_get_version_deprecated(self):
-        with check_warnings() as w:
-            warnings.simplefilter("always")
-            # make sure get_compiler_versions and get_versions
-            # returns the same thing
-            self.assertEquals(get_compiler_versions(), get_versions())
-            # make sure using get_version() generated a warning
-            self.assertEquals(len(w.warnings), 1)
-            # make sure any usage of RE_VERSION will also
-            # generate a warning, but till works
-            version = RE_VERSION.search('1.2').group(1)
-            self.assertEquals(version, '1.2')
-            self.assertEquals(len(w.warnings), 2)
-
 def test_suite():
     return unittest.makeSuite(CygwinCCompilerTestCase)
 
 if __name__ == '__main__':
-    run_unittest(test_suite())
+    test_support.run_unittest(test_suite())

Modified: python/branches/py3k/Lib/distutils/tests/test_dist.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_dist.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/test_dist.py	Thu Jul 22 14:50:05 2010
@@ -7,9 +7,8 @@
 import warnings
 import textwrap
 
-from distutils.dist import Distribution, fix_help_options, DistributionMetadata
+from distutils.dist import Distribution, fix_help_options
 from distutils.cmd import Command
-import distutils.dist
 
 from test.support import TESTFN, captured_stdout
 from distutils.tests import support
@@ -38,8 +37,7 @@
         return self._config_files
 
 
-class DistributionTestCase(support.TempdirManager,
-                           support.LoggingSilencer,
+class DistributionTestCase(support.LoggingSilencer,
                            support.EnvironGuard,
                            unittest.TestCase):
 
@@ -60,27 +58,6 @@
         d.parse_command_line()
         return d
 
-    def test_debug_mode(self):
-        with open(TESTFN, "w") as f:
-            f.write("[global]")
-            f.write("command_packages = foo.bar, splat")
-
-        files = [TESTFN]
-        sys.argv.append("build")
-
-        with captured_stdout() as stdout:
-            self.create_distribution(files)
-        stdout.seek(0)
-        self.assertEquals(stdout.read(), '')
-        distutils.dist.DEBUG = True
-        try:
-            with captured_stdout() as stdout:
-                self.create_distribution(files)
-            stdout.seek(0)
-            self.assertEquals(stdout.read(), '')
-        finally:
-            distutils.dist.DEBUG = False
-
     def test_command_packages_unspecified(self):
         sys.argv.append("build")
         d = self.create_distribution()
@@ -182,35 +159,6 @@
         kwargs = {'level': 'ok2'}
         self.assertRaises(ValueError, dist.announce, args, kwargs)
 
-    def test_find_config_files_disable(self):
-        # Ticket #1180: Allow user to disable their home config file.
-        temp_home = self.mkdtemp()
-        if os.name == 'posix':
-            user_filename = os.path.join(temp_home, ".pydistutils.cfg")
-        else:
-            user_filename = os.path.join(temp_home, "pydistutils.cfg")
-
-        with open(user_filename, 'w') as f:
-            f.write('[distutils]\n')
-
-        def _expander(path):
-            return temp_home
-
-        old_expander = os.path.expanduser
-        os.path.expanduser = _expander
-        try:
-            d = distutils.dist.Distribution()
-            all_files = d.find_config_files()
-
-            d = distutils.dist.Distribution(attrs={'script_args':
-                                            ['--no-user-cfg']})
-            files = d.find_config_files()
-        finally:
-            os.path.expanduser = old_expander
-
-        # make sure --no-user-cfg disables the user cfg file
-        self.assertEquals(len(all_files)-1, len(files))
-
 class MetadataTestCase(support.TempdirManager, support.EnvironGuard,
                        unittest.TestCase):
 
@@ -364,38 +312,11 @@
                  "version": "1.0",
                  "long_description": long_desc}
 
-        dist = distutils.dist.Distribution(attrs)
+        dist = Distribution(attrs)
         meta = self.format_metadata(dist)
         meta = meta.replace('\n' + 8 * ' ', '\n')
         self.assertTrue(long_desc in meta)
 
-    def test_read_metadata(self):
-        attrs = {"name": "package",
-                 "version": "1.0",
-                 "long_description": "desc",
-                 "description": "xxx",
-                 "download_url": "http://example.com",
-                 "keywords": ['one', 'two'],
-                 "requires": ['foo']}
-
-        dist = Distribution(attrs)
-        metadata = dist.metadata
-
-        # write it then reloads it
-        PKG_INFO = io.StringIO()
-        metadata.write_pkg_file(PKG_INFO)
-        PKG_INFO.seek(0)
-        metadata.read_pkg_file(PKG_INFO)
-
-        self.assertEquals(metadata.name, "package")
-        self.assertEquals(metadata.version, "1.0")
-        self.assertEquals(metadata.description, "xxx")
-        self.assertEquals(metadata.download_url, 'http://example.com')
-        self.assertEquals(metadata.keywords, ['one', 'two'])
-        self.assertEquals(metadata.platforms, ['UNKNOWN'])
-        self.assertEquals(metadata.obsoletes, None)
-        self.assertEquals(metadata.requires, ['foo'])
-
 def test_suite():
     suite = unittest.TestSuite()
     suite.addTest(unittest.makeSuite(DistributionTestCase))

Deleted: python/branches/py3k/Lib/distutils/tests/test_emxccompiler.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_emxccompiler.py	Thu Jul 22 14:50:05 2010
+++ (empty file)
@@ -1,33 +0,0 @@
-"""Tests for distutils.emxccompiler."""
-import unittest
-import sys
-import os
-import warnings
-
-from test.support import check_warnings, run_unittest
-from test.support import captured_stdout
-
-from distutils.emxccompiler import get_versions
-from distutils.util import get_compiler_versions
-from distutils.tests import support
-
-class EmxCCompilerTestCase(support.TempdirManager,
-                           unittest.TestCase):
-
-    def test_get_version_deprecated(self):
-        with check_warnings() as w:
-            warnings.simplefilter("always")
-            # make sure get_compiler_versions and get_versions
-            # returns the same gcc
-            gcc, ld, dllwrap = get_compiler_versions()
-            emx_gcc, emx_ld = get_versions()
-            self.assertEquals(gcc, emx_gcc)
-
-            # make sure using get_version() generated a warning
-            self.assertEquals(len(w.warnings), 1)
-
-def test_suite():
-    return unittest.makeSuite(EmxCCompilerTestCase)
-
-if __name__ == '__main__':
-    run_unittest(test_suite())

Modified: python/branches/py3k/Lib/distutils/tests/test_extension.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_extension.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/test_extension.py	Thu Jul 22 14:50:05 2010
@@ -1,16 +1,13 @@
 """Tests for distutils.extension."""
-import os
-import sys
 import unittest
+import os
 import warnings
 
 from test.support import check_warnings
 from distutils.extension import read_setup_file, Extension
-from distutils.tests.support import capture_warnings
 
 class ExtensionTestCase(unittest.TestCase):
 
-    @capture_warnings
     def test_read_setup_file(self):
         # trying to read a Setup file
         # (sample extracted from the PyGame project)
@@ -33,22 +30,16 @@
 
         self.assertEquals(names, wanted)
 
-    @unittest.skipIf(sys.flags.optimize >= 2,
-                     "Assertions are omitted with -O2 and above")
-    def test_extension_init_assertions(self):
-        # The first argument, which is the name, must be a string.
+    def test_extension_init(self):
+        # the first argument, which is the name, must be a string
         self.assertRaises(AssertionError, Extension, 1, [])
+        ext = Extension('name', [])
+        self.assertEquals(ext.name, 'name')
 
         # the second argument, which is the list of files, must
         # be a list of strings
         self.assertRaises(AssertionError, Extension, 'name', 'file')
         self.assertRaises(AssertionError, Extension, 'name', ['file', 1])
-
-    def test_extension_init(self):
-        ext = Extension('name', [])
-        self.assertEquals(ext.name, 'name')
-
-
         ext = Extension('name', ['file1', 'file2'])
         self.assertEquals(ext.sources, ['file1', 'file2'])
 

Modified: python/branches/py3k/Lib/distutils/tests/test_file_util.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_file_util.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/test_file_util.py	Thu Jul 22 14:50:05 2010
@@ -3,7 +3,7 @@
 import os
 import shutil
 
-from distutils.file_util import move_file, write_file, copy_file
+from distutils.file_util import move_file
 from distutils import log
 from distutils.tests import support
 
@@ -55,21 +55,6 @@
         wanted = ['moving %s -> %s' % (self.source, self.target_dir)]
         self.assertEquals(self._logs, wanted)
 
-    def test_write_file(self):
-        lines = ['a', 'b', 'c']
-        dir = self.mkdtemp()
-        foo = os.path.join(dir, 'foo')
-        write_file(foo, lines)
-        content = [line.strip() for line in open(foo).readlines()]
-        self.assertEquals(content, lines)
-
-    def test_copy_file(self):
-        src_dir = self.mkdtemp()
-        foo = os.path.join(src_dir, 'foo')
-        write_file(foo, 'content')
-        dst_dir = self.mkdtemp()
-        copy_file(foo, dst_dir)
-        self.assertTrue(os.path.exists(os.path.join(dst_dir, 'foo')))
 
 def test_suite():
     return unittest.makeSuite(FileUtilTestCase)

Modified: python/branches/py3k/Lib/distutils/tests/test_filelist.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_filelist.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/test_filelist.py	Thu Jul 22 14:50:05 2010
@@ -1,25 +1,10 @@
 """Tests for distutils.filelist."""
-from os.path import join
 import unittest
-from test.support import captured_stdout
 
 from distutils.filelist import glob_to_re, FileList
+from test.support import captured_stdout
 from distutils import debug
 
-MANIFEST_IN = """\
-include ok
-include xo
-exclude xo
-include foo.tmp
-global-include *.x
-global-include *.txt
-global-exclude *.tmp
-recursive-include f *.oo
-recursive-exclude global *.x
-graft dir
-prune dir3
-"""
-
 class FileListTestCase(unittest.TestCase):
 
     def test_glob_to_re(self):
@@ -34,34 +19,6 @@
         self.assertEquals(glob_to_re('foo????'), r'foo[^/][^/][^/][^/]\Z(?ms)')
         self.assertEquals(glob_to_re(r'foo\\??'), r'foo\\\\[^/][^/]\Z(?ms)')
 
-    def test_process_template_line(self):
-        # testing  all MANIFEST.in template patterns
-        file_list = FileList()
-
-        # simulated file list
-        file_list.allfiles = ['foo.tmp', 'ok', 'xo', 'four.txt',
-                              join('global', 'one.txt'),
-                              join('global', 'two.txt'),
-                              join('global', 'files.x'),
-                              join('global', 'here.tmp'),
-                              join('f', 'o', 'f.oo'),
-                              join('dir', 'graft-one'),
-                              join('dir', 'dir2', 'graft2'),
-                              join('dir3', 'ok'),
-                              join('dir3', 'sub', 'ok.txt')
-                              ]
-
-        for line in MANIFEST_IN.split('\n'):
-            if line.strip() == '':
-                continue
-            file_list.process_template_line(line)
-
-        wanted = ['ok', 'four.txt', join('global', 'one.txt'),
-                  join('global', 'two.txt'), join('f', 'o', 'f.oo'),
-                  join('dir', 'graft-one'), join('dir', 'dir2', 'graft2')]
-
-        self.assertEquals(file_list.files, wanted)
-
     def test_debug_print(self):
         file_list = FileList()
         with captured_stdout() as stdout:

Modified: python/branches/py3k/Lib/distutils/tests/test_install.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_install.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/test_install.py	Thu Jul 22 14:50:05 2010
@@ -5,14 +5,12 @@
 import sys
 import unittest
 import site
-import sysconfig
-from sysconfig import (get_scheme_names, _CONFIG_VARS, _INSTALL_SCHEMES,
-                       get_config_var, get_path)
 
 from test.support import captured_stdout
 
 from distutils.command.install import install
 from distutils.command import install as install_module
+from distutils.command.install import INSTALL_SCHEMES
 from distutils.core import Distribution
 from distutils.errors import DistutilsOptionError
 
@@ -38,23 +36,9 @@
             build_lib=os.path.join(builddir, "lib"),
             )
 
-
-
-        posix_prefix = _INSTALL_SCHEMES['posix_prefix']
-        old_posix_prefix = posix_prefix['platinclude']
-        posix_prefix['platinclude'] = \
-                '{platbase}/include/python{py_version_short}'
-
-        posix_home = _INSTALL_SCHEMES['posix_home']
-        old_posix_home = posix_home['platinclude']
-        posix_home['platinclude'] = '{base}/include/python'
-        try:
-            cmd = install(dist)
-            cmd.home = destination
-            cmd.ensure_finalized()
-        finally:
-            posix_home['platinclude'] = old_posix_home
-            posix_prefix['platinclude'] = old_posix_prefix
+        cmd = install(dist)
+        cmd.home = destination
+        cmd.ensure_finalized()
 
         self.assertEqual(cmd.install_base, destination)
         self.assertEqual(cmd.install_platbase, destination)
@@ -79,19 +63,18 @@
             return
 
         # preparing the environement for the test
-        self.old_user_base = get_config_var('userbase')
-        self.old_user_site = get_path('purelib', '%s_user' % os.name)
+        self.old_user_base = site.USER_BASE
+        self.old_user_site = site.USER_SITE
         self.tmpdir = self.mkdtemp()
         self.user_base = os.path.join(self.tmpdir, 'B')
         self.user_site = os.path.join(self.tmpdir, 'S')
-        _CONFIG_VARS['userbase'] = self.user_base
-        scheme = _INSTALL_SCHEMES['%s_user' % os.name]
-        scheme['purelib'] = self.user_site
+        site.USER_BASE = self.user_base
+        site.USER_SITE = self.user_site
+        install_module.USER_BASE = self.user_base
+        install_module.USER_SITE = self.user_site
 
         def _expanduser(path):
-            if path[0] == '~':
-                path = os.path.normpath(self.tmpdir) + path[1:]
-            return path
+            return self.tmpdir
         self.old_expand = os.path.expanduser
         os.path.expanduser = _expanduser
 
@@ -99,17 +82,19 @@
             # this is the actual test
             self._test_user_site()
         finally:
-            _CONFIG_VARS['userbase'] = self.old_user_base
-            scheme['purelib'] = self.old_user_site
+            site.USER_BASE = self.old_user_base
+            site.USER_SITE = self.old_user_site
+            install_module.USER_BASE = self.old_user_base
+            install_module.USER_SITE = self.old_user_site
             os.path.expanduser = self.old_expand
 
     def _test_user_site(self):
-        schemes = get_scheme_names()
-        for key in ('nt_user', 'posix_user', 'os2_home'):
-            self.assertTrue(key in schemes)
+        for key in ('nt_user', 'unix_user', 'os2_home'):
+            self.assertTrue(key in INSTALL_SCHEMES)
 
         dist = Distribution({'name': 'xx'})
         cmd = install(dist)
+
         # making sure the user option is there
         options = [name for name, short, lable in
                    cmd.user_options]
@@ -200,7 +185,7 @@
         with open(cmd.record) as f:
             self.assertEquals(len(f.readlines()), 1)
 
-    def _test_debug_mode(self):
+    def test_debug_mode(self):
         # this covers the code called when DEBUG is set
         old_logs_len = len(self.logs)
         install_module.DEBUG = True

Modified: python/branches/py3k/Lib/distutils/tests/test_install_lib.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_install_lib.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/test_install_lib.py	Thu Jul 22 14:50:05 2010
@@ -1,6 +1,6 @@
 """Tests for distutils.command.install_data."""
-import os
 import sys
+import os
 import unittest
 
 from distutils.command.install_lib import install_lib
@@ -31,7 +31,9 @@
         cmd.finalize_options()
         self.assertEquals(cmd.optimize, 2)
 
-    def _setup_byte_compile(self):
+    @unittest.skipUnless(not sys.dont_write_bytecode,
+                         'byte-compile not supported')
+    def test_byte_compile(self):
         pkg_dir, dist = self.create_dist()
         cmd = install_lib(dist)
         cmd.compile = cmd.optimize = 1
@@ -39,15 +41,8 @@
         f = os.path.join(pkg_dir, 'foo.py')
         self.write_file(f, '# python file')
         cmd.byte_compile([f])
-        return pkg_dir
-
-    @unittest.skipIf(sys.dont_write_bytecode, 'byte-compile not enabled')
-    def test_byte_compile(self):
-        pkg_dir = self._setup_byte_compile()
-        if sys.flags.optimize < 1:
-            self.assertTrue(os.path.exists(os.path.join(pkg_dir, 'foo.pyc')))
-        else:
-            self.assertTrue(os.path.exists(os.path.join(pkg_dir, 'foo.pyo')))
+        self.assertTrue(os.path.exists(os.path.join(pkg_dir, 'foo.pyc')))
+        self.assertTrue(os.path.exists(os.path.join(pkg_dir, 'foo.pyo')))
 
     def test_get_outputs(self):
         pkg_dir, dist = self.create_dist()

Modified: python/branches/py3k/Lib/distutils/tests/test_register.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_register.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/test_register.py	Thu Jul 22 14:50:05 2010
@@ -202,10 +202,10 @@
         self.assertRaises(DistutilsSetupError, cmd.run)
 
         # we don't test the reSt feature if docutils
-        # is not installed or we our on py3k
+        # is not installed
         try:
             import docutils
-        except Exception:
+        except ImportError:
             return
 
         # metadata are OK but long_description is broken

Modified: python/branches/py3k/Lib/distutils/tests/test_sdist.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_sdist.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/test_sdist.py	Thu Jul 22 14:50:05 2010
@@ -3,22 +3,6 @@
 import unittest
 import shutil
 import zipfile
-import tarfile
-
-# zlib is not used here, but if it's not available
-# the tests that use zipfile may fail
-try:
-    import zlib
-except ImportError:
-    zlib = None
-
-try:
-    import grp
-    import pwd
-    UID_GID_SUPPORT = True
-except ImportError:
-    UID_GID_SUPPORT = False
-
 from os.path import join
 import sys
 import tempfile
@@ -95,7 +79,6 @@
         cmd.warn = _warn
         return dist, cmd
 
-    @unittest.skipUnless(zlib, "requires zlib")
     def test_prune_file_list(self):
         # this test creates a package with some vcs dirs in it
         # and launch sdist to make sure they get pruned
@@ -137,7 +120,6 @@
         # making sure everything has been pruned correctly
         self.assertEquals(len(content), 4)
 
-    @unittest.skipUnless(zlib, "requires zlib")
     def test_make_distribution(self):
 
         # check if tar and gzip are installed
@@ -174,7 +156,6 @@
         self.assertEquals(result,
                 ['fake-1.0.tar', 'fake-1.0.tar.gz'])
 
-    @unittest.skipUnless(zlib, "requires zlib")
     def test_add_defaults(self):
 
         # http://bugs.python.org/issue2279
@@ -236,7 +217,6 @@
         manifest = open(join(self.tmp_dir, 'MANIFEST')).read()
         self.assertEquals(manifest, MANIFEST % {'sep': os.sep})
 
-    @unittest.skipUnless(zlib, "requires zlib")
     def test_metadata_check_option(self):
         # testing the `medata-check` option
         dist, cmd = self.get_cmd(metadata={})
@@ -296,57 +276,7 @@
         cmd.formats = 'supazipa'
         self.assertRaises(DistutilsOptionError, cmd.finalize_options)
 
-    @unittest.skipUnless(zlib, "requires zlib")
-    @unittest.skipUnless(UID_GID_SUPPORT, "Requires grp and pwd support")
-    def test_make_distribution_owner_group(self):
-
-        # check if tar and gzip are installed
-        if (find_executable('tar') is None or
-            find_executable('gzip') is None):
-            return
-
-        # now building a sdist
-        dist, cmd = self.get_cmd()
-
-        # creating a gztar and specifying the owner+group
-        cmd.formats = ['gztar']
-        cmd.owner = pwd.getpwuid(0)[0]
-        cmd.group = grp.getgrgid(0)[0]
-        cmd.ensure_finalized()
-        cmd.run()
-
-        # making sure we have the good rights
-        archive_name = join(self.tmp_dir, 'dist', 'fake-1.0.tar.gz')
-        archive = tarfile.open(archive_name)
-        try:
-            for member in archive.getmembers():
-                self.assertEquals(member.uid, 0)
-                self.assertEquals(member.gid, 0)
-        finally:
-            archive.close()
-
-        # building a sdist again
-        dist, cmd = self.get_cmd()
-
-        # creating a gztar
-        cmd.formats = ['gztar']
-        cmd.ensure_finalized()
-        cmd.run()
-
-        # making sure we have the good rights
-        archive_name = join(self.tmp_dir, 'dist', 'fake-1.0.tar.gz')
-        archive = tarfile.open(archive_name)
-
-        # note that we are not testing the group ownership here
-        # because, depending on the platforms and the container
-        # rights (see #7408)
-        try:
-            for member in archive.getmembers():
-                self.assertEquals(member.uid, os.getuid())
-        finally:
-            archive.close()
 
-    @unittest.skipUnless(zlib, "requires zlib")
     def test_get_file_list(self):
         # make sure MANIFEST is recalculated
         dist, cmd = self.get_cmd()

Modified: python/branches/py3k/Lib/distutils/tests/test_sysconfig.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_sysconfig.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/test_sysconfig.py	Thu Jul 22 14:50:05 2010
@@ -2,9 +2,9 @@
 import os
 import test
 import unittest
-import shutil
 
 from distutils import sysconfig
+from distutils.ccompiler import get_default_compiler
 from distutils.tests import support
 from test.support import TESTFN, run_unittest
 
@@ -26,7 +26,10 @@
         elif os.path.isdir(TESTFN):
             shutil.rmtree(TESTFN)
 
-    @support.capture_warnings
+    def test_get_config_h_filename(self):
+        config_h = sysconfig.get_config_h_filename()
+        self.assertTrue(os.path.isfile(config_h), config_h)
+
     def test_get_python_lib(self):
         lib_dir = sysconfig.get_python_lib()
         # XXX doesn't work on Linux when Python was never installed before
@@ -34,11 +37,7 @@
         # test for pythonxx.lib?
         self.assertNotEqual(sysconfig.get_python_lib(),
                             sysconfig.get_python_lib(prefix=TESTFN))
-        _sysconfig = __import__('sysconfig')
-        res = sysconfig.get_python_lib(True, True)
-        self.assertEquals(_sysconfig.get_path('platstdlib'), res)
 
-    @support.capture_warnings
     def test_get_python_inc(self):
         inc_dir = sysconfig.get_python_inc()
         # This is not much of a test.  We make sure Python.h exists
@@ -48,7 +47,31 @@
         python_h = os.path.join(inc_dir, "Python.h")
         self.assertTrue(os.path.isfile(python_h), python_h)
 
-    @support.capture_warnings
+    def test_get_config_vars(self):
+        cvars = sysconfig.get_config_vars()
+        self.assertTrue(isinstance(cvars, dict))
+        self.assertTrue(cvars)
+
+    def test_customize_compiler(self):
+
+        # not testing if default compiler is not unix
+        if get_default_compiler() != 'unix':
+            return
+
+        os.environ['AR'] = 'my_ar'
+        os.environ['ARFLAGS'] = '-arflags'
+
+        # make sure AR gets caught
+        class compiler:
+            compiler_type = 'unix'
+
+            def set_executables(self, **kw):
+                self.exes = kw
+
+        comp = compiler()
+        sysconfig.customize_compiler(comp)
+        self.assertEquals(comp.exes['archiver'], 'my_ar -arflags')
+
     def test_parse_makefile_base(self):
         self.makefile = TESTFN
         fd = open(self.makefile, 'w')

Modified: python/branches/py3k/Lib/distutils/tests/test_unixccompiler.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_unixccompiler.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/test_unixccompiler.py	Thu Jul 22 14:50:05 2010
@@ -1,8 +1,8 @@
 """Tests for distutils.unixccompiler."""
 import sys
 import unittest
-import sysconfig
 
+from distutils import sysconfig
 from distutils.unixccompiler import UnixCCompiler
 
 class UnixCCompilerTestCase(unittest.TestCase):
@@ -114,14 +114,6 @@
         sysconfig.get_config_var = gcv
         self.assertEqual(self.cc.rpath_foo(), '-R/foo')
 
-        # AIX C/C++ linker
-        sys.platform = 'aix'
-        def gcv(v):
-            return 'xxx'
-        sysconfig.get_config_var = gcv
-        self.assertEqual(self.cc.rpath_foo(), '-blibpath:/foo')
-
-
 def test_suite():
     return unittest.makeSuite(UnixCCompilerTestCase)
 

Modified: python/branches/py3k/Lib/distutils/tests/test_upload.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_upload.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/test_upload.py	Thu Jul 22 14:50:05 2010
@@ -2,8 +2,8 @@
 import sys
 import os
 import unittest
+import http.client as httpclient
 
-from distutils.command import upload as upload_mod
 from distutils.command.upload import upload
 from distutils.core import Distribution
 
@@ -38,37 +38,48 @@
 [server1]
 username:me
 """
+class Response(object):
+    def __init__(self, status=200, reason='OK'):
+        self.status = status
+        self.reason = reason
 
-class FakeOpen(object):
+class FakeConnection(object):
 
-    def __init__(self, url):
-        self.url = url
-        if not isinstance(url, str):
-            self.req = url
-        else:
-            self.req = None
-        self.msg = 'OK'
+    def __init__(self):
+        self.requests = []
+        self.headers = []
+        self.body = ''
 
-    def getcode(self):
-        return 200
+    def __call__(self, netloc):
+        return self
 
+    def connect(self):
+        pass
+    endheaders = connect
+
+    def putrequest(self, method, url):
+        self.requests.append((method, url))
+
+    def putheader(self, name, value):
+        self.headers.append((name, value))
+
+    def send(self, body):
+        self.body = body
+
+    def getresponse(self):
+        return Response()
 
 class uploadTestCase(PyPIRCCommandTestCase):
 
     def setUp(self):
         super(uploadTestCase, self).setUp()
-        self.old_open = upload_mod.urlopen
-        upload_mod.urlopen = self._urlopen
-        self.last_open = None
+        self.old_class = httpclient.HTTPConnection
+        self.conn = httpclient.HTTPConnection = FakeConnection()
 
     def tearDown(self):
-        upload_mod.urlopen = self.old_open
+        httpclient.HTTPConnection = self.old_class
         super(uploadTestCase, self).tearDown()
 
-    def _urlopen(self, url):
-        self.last_open = FakeOpen(url)
-        return self.last_open
-
     def test_finalize_options(self):
 
         # new format
@@ -113,15 +124,13 @@
         cmd.run()
 
         # what did we send ?
-        headers = dict(self.last_open.req.headers)
+        headers = dict(self.conn.headers)
         self.assertEquals(headers['Content-length'], '2087')
         self.assertTrue(headers['Content-type'].startswith('multipart/form-data'))
-        self.assertEquals(self.last_open.req.get_method(), 'POST')
-        self.assertEquals(self.last_open.req.get_full_url(),
-                          'http://pypi.python.org/pypi')
-        self.assertTrue(b'xxx' in self.last_open.req.data)
-        auth = self.last_open.req.headers['Authorization']
-        self.assertFalse('\n' in auth)
+        self.assertFalse('\n' in headers['Authorization'])
+
+        self.assertEquals(self.conn.requests, [('POST', '/pypi')])
+        self.assert_((b'xxx') in self.conn.body)
 
 def test_suite():
     return unittest.makeSuite(uploadTestCase)

Modified: python/branches/py3k/Lib/distutils/tests/test_util.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_util.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/test_util.py	Thu Jul 22 14:50:05 2010
@@ -3,33 +3,15 @@
 import sys
 import unittest
 from copy import copy
-from io import BytesIO
-import subprocess
 
-from sysconfig import get_config_vars, get_platform
 from distutils.errors import DistutilsPlatformError, DistutilsByteCompileError
-from distutils.util import (convert_path, change_root,
+from distutils.util import (get_platform, convert_path, change_root,
                             check_environ, split_quoted, strtobool,
-                            rfc822_escape, get_compiler_versions,
-                            _find_exe_version, _MAC_OS_X_LD_VERSION,
-                            byte_compile)
-from distutils import util
+                            rfc822_escape, byte_compile)
+from distutils import util # used to patch _environ_checked
+from distutils.sysconfig import get_config_vars
+from distutils import sysconfig
 from distutils.tests import support
-from distutils.version import LooseVersion
-
-class FakePopen(object):
-    test_class = None
-    def __init__(self, cmd, shell, stdout, stderr):
-        self.cmd = cmd.split()[0]
-        exes = self.test_class._exes
-        if self.cmd not in exes:
-            # we don't want to call the system, returning an empty
-            # output so it doesn't match
-            self.stdout = BytesIO()
-            self.stderr = BytesIO()
-        else:
-            self.stdout = BytesIO(exes[self.cmd])
-            self.stderr = BytesIO()
 
 class UtilTestCase(support.EnvironGuard, unittest.TestCase):
 
@@ -43,7 +25,7 @@
         self.join = os.path.join
         self.isabs = os.path.isabs
         self.splitdrive = os.path.splitdrive
-        #self._config_vars = copy(sysconfig._config_vars)
+        self._config_vars = copy(sysconfig._config_vars)
 
         # patching os.uname
         if hasattr(os, 'uname'):
@@ -52,17 +34,8 @@
         else:
             self.uname = None
             self._uname = None
-        os.uname = self._get_uname
 
-        # patching POpen
-        self.old_find_executable = util.find_executable
-        util.find_executable = self._find_executable
-        self._exes = {}
-        self.old_popen = subprocess.Popen
-        self.old_stdout  = sys.stdout
-        self.old_stderr = sys.stderr
-        FakePopen.test_class = self
-        subprocess.Popen = FakePopen
+        os.uname = self._get_uname
 
     def tearDown(self):
         # getting back the environment
@@ -77,11 +50,7 @@
             os.uname = self.uname
         else:
             del os.uname
-        #sysconfig._config_vars = copy(self._config_vars)
-        util.find_executable = self.old_find_executable
-        subprocess.Popen = self.old_popen
-        sys.old_stdout  = self.old_stdout
-        sys.old_stderr = self.old_stderr
+        sysconfig._config_vars = copy(self._config_vars)
         super(UtilTestCase, self).tearDown()
 
     def _set_uname(self, uname):
@@ -91,11 +60,103 @@
         return self._uname
 
     def test_get_platform(self):
-        platform = util.get_platform()
-        self.assertEquals(platform, get_platform())
-        util.set_platform('MyOwnPlatform')
-        self.assertEquals('MyOwnPlatform', util.get_platform())
-        util.set_platform(platform)
+
+        # windows XP, 32bits
+        os.name = 'nt'
+        sys.version = ('2.4.4 (#71, Oct 18 2006, 08:34:43) '
+                       '[MSC v.1310 32 bit (Intel)]')
+        sys.platform = 'win32'
+        self.assertEquals(get_platform(), 'win32')
+
+        # windows XP, amd64
+        os.name = 'nt'
+        sys.version = ('2.4.4 (#71, Oct 18 2006, 08:34:43) '
+                       '[MSC v.1310 32 bit (Amd64)]')
+        sys.platform = 'win32'
+        self.assertEquals(get_platform(), 'win-amd64')
+
+        # windows XP, itanium
+        os.name = 'nt'
+        sys.version = ('2.4.4 (#71, Oct 18 2006, 08:34:43) '
+                       '[MSC v.1310 32 bit (Itanium)]')
+        sys.platform = 'win32'
+        self.assertEquals(get_platform(), 'win-ia64')
+
+        # macbook
+        os.name = 'posix'
+        sys.version = ('2.5 (r25:51918, Sep 19 2006, 08:49:13) '
+                       '\n[GCC 4.0.1 (Apple Computer, Inc. build 5341)]')
+        sys.platform = 'darwin'
+        self._set_uname(('Darwin', 'macziade', '8.11.1',
+                   ('Darwin Kernel Version 8.11.1: '
+                    'Wed Oct 10 18:23:28 PDT 2007; '
+                    'root:xnu-792.25.20~1/RELEASE_I386'), 'i386'))
+        os.environ['MACOSX_DEPLOYMENT_TARGET'] = '10.3'
+
+        get_config_vars()['CFLAGS'] = ('-fno-strict-aliasing -DNDEBUG -g '
+                                       '-fwrapv -O3 -Wall -Wstrict-prototypes')
+
+        cursize = sys.maxsize
+        sys.maxsize = (2 ** 31)-1
+        try:
+            self.assertEquals(get_platform(), 'macosx-10.3-i386')
+        finally:
+            sys.maxsize = cursize
+
+        # macbook with fat binaries (fat, universal or fat64)
+        os.environ['MACOSX_DEPLOYMENT_TARGET'] = '10.4'
+        get_config_vars()['CFLAGS'] = ('-arch ppc -arch i386 -isysroot '
+                                       '/Developer/SDKs/MacOSX10.4u.sdk  '
+                                       '-fno-strict-aliasing -fno-common '
+                                       '-dynamic -DNDEBUG -g -O3')
+
+        self.assertEquals(get_platform(), 'macosx-10.4-fat')
+
+        get_config_vars()['CFLAGS'] = ('-arch x86_64 -arch i386 -isysroot '
+                                       '/Developer/SDKs/MacOSX10.4u.sdk  '
+                                       '-fno-strict-aliasing -fno-common '
+                                       '-dynamic -DNDEBUG -g -O3')
+
+        self.assertEquals(get_platform(), 'macosx-10.4-intel')
+
+        get_config_vars()['CFLAGS'] = ('-arch x86_64 -arch ppc -arch i386 -isysroot '
+                                       '/Developer/SDKs/MacOSX10.4u.sdk  '
+                                       '-fno-strict-aliasing -fno-common '
+                                       '-dynamic -DNDEBUG -g -O3')
+        self.assertEquals(get_platform(), 'macosx-10.4-fat3')
+
+        get_config_vars()['CFLAGS'] = ('-arch ppc64 -arch x86_64 -arch ppc -arch i386 -isysroot '
+                                       '/Developer/SDKs/MacOSX10.4u.sdk  '
+                                       '-fno-strict-aliasing -fno-common '
+                                       '-dynamic -DNDEBUG -g -O3')
+        self.assertEquals(get_platform(), 'macosx-10.4-universal')
+
+        get_config_vars()['CFLAGS'] = ('-arch x86_64 -arch ppc64 -isysroot '
+                                       '/Developer/SDKs/MacOSX10.4u.sdk  '
+                                       '-fno-strict-aliasing -fno-common '
+                                       '-dynamic -DNDEBUG -g -O3')
+
+        self.assertEquals(get_platform(), 'macosx-10.4-fat64')
+
+        for arch in ('ppc', 'i386', 'x86_64', 'ppc64'):
+            get_config_vars()['CFLAGS'] = ('-arch %s -isysroot '
+                                           '/Developer/SDKs/MacOSX10.4u.sdk  '
+                                           '-fno-strict-aliasing -fno-common '
+                                           '-dynamic -DNDEBUG -g -O3'%(arch,))
+
+            self.assertEquals(get_platform(), 'macosx-10.4-%s'%(arch,))
+
+        # linux debian sarge
+        os.name = 'posix'
+        sys.version = ('2.3.5 (#1, Jul  4 2007, 17:28:59) '
+                       '\n[GCC 4.1.2 20061115 (prerelease) (Debian 4.1.1-21)]')
+        sys.platform = 'linux2'
+        self._set_uname(('Linux', 'aglae', '2.6.21.1dedibox-r7',
+                    '#1 Mon Apr 30 17:25:38 CEST 2007', 'i686'))
+
+        self.assertEquals(get_platform(), 'linux-i686')
+
+        # XXX more platforms to tests here
 
     def test_convert_path(self):
         # linux/mac
@@ -199,70 +260,6 @@
                   'header%(8s)s') % {'8s': '\n'+8*' '}
         self.assertEquals(res, wanted)
 
-    def test_find_exe_version(self):
-        # the ld version scheme under MAC OS is:
-        #   ^@(#)PROGRAM:ld  PROJECT:ld64-VERSION
-        #
-        # where VERSION is a 2-digit number for major
-        # revisions. For instance under Leopard, it's
-        # currently 77
-        #
-        # Dots are used when branching is done.
-        #
-        # The SnowLeopard ld64 is currently 95.2.12
-
-        for output, version in ((b'@(#)PROGRAM:ld  PROJECT:ld64-77', '77'),
-                                (b'@(#)PROGRAM:ld  PROJECT:ld64-95.2.12',
-                                 '95.2.12')):
-            result = _MAC_OS_X_LD_VERSION.search(output)
-            self.assertEquals(result.group(1).decode(), version)
-
-    def _find_executable(self, name):
-        if name in self._exes:
-            return name
-        return None
-
-    def test_get_compiler_versions(self):
-        # get_versions calls distutils.spawn.find_executable on
-        # 'gcc', 'ld' and 'dllwrap'
-        self.assertEquals(get_compiler_versions(), (None, None, None))
-
-        # Let's fake we have 'gcc' and it returns '3.4.5'
-        self._exes['gcc'] = b'gcc (GCC) 3.4.5 (mingw special)\nFSF'
-        res = get_compiler_versions()
-        self.assertEquals(str(res[0]), '3.4.5')
-
-        # and let's see what happens when the version
-        # doesn't match the regular expression
-        # (\d+\.\d+(\.\d+)*)
-        self._exes['gcc'] = b'very strange output'
-        res = get_compiler_versions()
-        self.assertEquals(res[0], None)
-
-        # same thing for ld
-        if sys.platform != 'darwin':
-            self._exes['ld'] = b'GNU ld version 2.17.50 20060824'
-            res = get_compiler_versions()
-            self.assertEquals(str(res[1]), '2.17.50')
-            self._exes['ld'] = b'@(#)PROGRAM:ld  PROJECT:ld64-77'
-            res = get_compiler_versions()
-            self.assertEquals(res[1], None)
-        else:
-            self._exes['ld'] = b'GNU ld version 2.17.50 20060824'
-            res = get_compiler_versions()
-            self.assertEquals(res[1], None)
-            self._exes['ld'] = b'@(#)PROGRAM:ld  PROJECT:ld64-77'
-            res = get_compiler_versions()
-            self.assertEquals(str(res[1]), '77')
-
-        # and dllwrap
-        self._exes['dllwrap'] = b'GNU dllwrap 2.17.50 20060824\nFSF'
-        res = get_compiler_versions()
-        self.assertEquals(str(res[2]), '2.17.50')
-        self._exes['dllwrap'] = b'Cheese Wrap'
-        res = get_compiler_versions()
-        self.assertEquals(res[2], None)
-
     def test_dont_write_bytecode(self):
         # makes sure byte_compile raise a DistutilsError
         # if sys.dont_write_bytecode is True

Modified: python/branches/py3k/Lib/distutils/text_file.py
==============================================================================
--- python/branches/py3k/Lib/distutils/text_file.py	(original)
+++ python/branches/py3k/Lib/distutils/text_file.py	Thu Jul 22 14:50:05 2010
@@ -6,8 +6,8 @@
 
 __revision__ = "$Id$"
 
-import sys
-import io
+import sys, os, io
+
 
 class TextFile:
     """Provides a file-like object that takes care of all the things you

Modified: python/branches/py3k/Lib/distutils/unixccompiler.py
==============================================================================
--- python/branches/py3k/Lib/distutils/unixccompiler.py	(original)
+++ python/branches/py3k/Lib/distutils/unixccompiler.py	Thu Jul 22 14:50:05 2010
@@ -17,6 +17,7 @@
 
 import os, sys, re
 
+from distutils import sysconfig
 from distutils.dep_util import newer
 from distutils.ccompiler import \
      CCompiler, gen_preprocess_options, gen_lib_options
@@ -24,7 +25,6 @@
      DistutilsExecError, CompileError, LibError, LinkError
 from distutils import log
 
-
 # XXX Things not currently handled:
 #   * optimization/debug/warning flags; we just use whatever's in Python's
 #     Makefile and live with it.  Is this adequate?  If not, we might
@@ -74,7 +74,7 @@
 
     if 'ARCHFLAGS' in os.environ and not stripArch:
         # User specified different -arch flags in the environ,
-        # see also the sysconfig
+        # see also distutils.sysconfig
         compiler_so = compiler_so + os.environ['ARCHFLAGS'].split()
 
     if stripSysroot:
@@ -281,9 +281,7 @@
         # this time, there's no way to determine this information from
         # the configuration data stored in the Python installation, so
         # we use this hack.
-        _sysconfig = __import__('sysconfig')
-
-        compiler = os.path.basename(_sysconfig.get_config_var("CC"))
+        compiler = os.path.basename(sysconfig.get_config_var("CC"))
         if sys.platform[:6] == "darwin":
             # MacOSX's linker doesn't understand the -R flag at all
             return "-L" + dir
@@ -293,24 +291,23 @@
             return ["+s", "-L" + dir]
         elif sys.platform[:7] == "irix646" or sys.platform[:6] == "osf1V5":
             return ["-rpath", dir]
-        elif self._is_gcc(compiler):
-            # gcc on non-GNU systems does not need -Wl, but can
-            # use it anyway.  Since distutils has always passed in
-            # -Wl whenever gcc was used in the past it is probably
-            # safest to keep doing so.
-            if _sysconfig.get_config_var("GNULD") == "yes":
-                # GNU ld needs an extra option to get a RUNPATH
-                # instead of just an RPATH.
-                return "-Wl,--enable-new-dtags,-R" + dir
-            else:
-                return "-Wl,-R" + dir
-        elif sys.platform[:3] == "aix":
-            return "-blibpath:" + dir
         else:
-            # No idea how --enable-new-dtags would be passed on to
-            # ld if this system was using GNU ld.  Don't know if a
-            # system like this even exists.
-            return "-R" + dir
+            if self._is_gcc(compiler):
+                # gcc on non-GNU systems does not need -Wl, but can
+                # use it anyway.  Since distutils has always passed in
+                # -Wl whenever gcc was used in the past it is probably
+                # safest to keep doing so.
+                if sysconfig.get_config_var("GNULD") == "yes":
+                    # GNU ld needs an extra option to get a RUNPATH
+                    # instead of just an RPATH.
+                    return "-Wl,--enable-new-dtags,-R" + dir
+                else:
+                    return "-Wl,-R" + dir
+            else:
+                # No idea how --enable-new-dtags would be passed on to
+                # ld if this system was using GNU ld.  Don't know if a
+                # system like this even exists.
+                return "-R" + dir
 
     def library_option(self, lib):
         return "-l" + lib
@@ -324,8 +321,7 @@
             # On OSX users can specify an alternate SDK using
             # '-isysroot', calculate the SDK root if it is specified
             # (and use it further on)
-            _sysconfig = __import__('sysconfig')
-            cflags = _sysconfig.get_config_var('CFLAGS')
+            cflags = sysconfig.get_config_var('CFLAGS')
             m = re.search(r'-isysroot\s+(\S+)', cflags)
             if m is None:
                 sysroot = '/'

Modified: python/branches/py3k/Lib/distutils/util.py
==============================================================================
--- python/branches/py3k/Lib/distutils/util.py	(original)
+++ python/branches/py3k/Lib/distutils/util.py	Thu Jul 22 14:50:05 2010
@@ -7,40 +7,182 @@
 __revision__ = "$Id$"
 
 import sys, os, string, re
-
 from distutils.errors import DistutilsPlatformError
 from distutils.dep_util import newer
-from distutils.spawn import spawn, find_executable
+from distutils.spawn import spawn
 from distutils import log
-from distutils.version import LooseVersion
 from distutils.errors import DistutilsByteCompileError
 
-_sysconfig = __import__('sysconfig')
-_PLATFORM = None
+def get_platform ():
+    """Return a string that identifies the current platform.  This is used
+    mainly to distinguish platform-specific build directories and
+    platform-specific built distributions.  Typically includes the OS name
+    and version and the architecture (as supplied by 'os.uname()'),
+    although the exact information included depends on the OS; eg. for IRIX
+    the architecture isn't particularly important (IRIX only runs on SGI
+    hardware), but for Linux the kernel version isn't particularly
+    important.
+
+    Examples of returned values:
+       linux-i586
+       linux-alpha (?)
+       solaris-2.6-sun4u
+       irix-5.3
+       irix64-6.2
+
+    Windows will return one of:
+       win-amd64 (64bit Windows on AMD64 (aka x86_64, Intel64, EM64T, etc)
+       win-ia64 (64bit Windows on Itanium)
+       win32 (all others - specifically, sys.platform is returned)
+
+    For other non-POSIX platforms, currently just returns 'sys.platform'.
+    """
+    if os.name == 'nt':
+        # sniff sys.version for architecture.
+        prefix = " bit ("
+        i = sys.version.find(prefix)
+        if i == -1:
+            return sys.platform
+        j = sys.version.find(")", i)
+        look = sys.version[i+len(prefix):j].lower()
+        if look == 'amd64':
+            return 'win-amd64'
+        if look == 'itanium':
+            return 'win-ia64'
+        return sys.platform
+
+    if os.name != "posix" or not hasattr(os, 'uname'):
+        # XXX what about the architecture? NT is Intel or Alpha,
+        # Mac OS is M68k or PPC, etc.
+        return sys.platform
+
+    # Try to distinguish various flavours of Unix
+
+    (osname, host, release, version, machine) = os.uname()
+
+    # Convert the OS name to lowercase, remove '/' characters
+    # (to accommodate BSD/OS), and translate spaces (for "Power Macintosh")
+    osname = osname.lower().replace('/', '')
+    machine = machine.replace(' ', '_')
+    machine = machine.replace('/', '-')
+
+    if osname[:5] == "linux":
+        # At least on Linux/Intel, 'machine' is the processor --
+        # i386, etc.
+        # XXX what about Alpha, SPARC, etc?
+        return  "%s-%s" % (osname, machine)
+    elif osname[:5] == "sunos":
+        if release[0] >= "5":           # SunOS 5 == Solaris 2
+            osname = "solaris"
+            release = "%d.%s" % (int(release[0]) - 3, release[2:])
+        # fall through to standard osname-release-machine representation
+    elif osname[:4] == "irix":              # could be "irix64"!
+        return "%s-%s" % (osname, release)
+    elif osname[:3] == "aix":
+        return "%s-%s.%s" % (osname, version, release)
+    elif osname[:6] == "cygwin":
+        osname = "cygwin"
+        rel_re = re.compile (r'[\d.]+', re.ASCII)
+        m = rel_re.match(release)
+        if m:
+            release = m.group()
+    elif osname[:6] == "darwin":
+        #
+        # For our purposes, we'll assume that the system version from
+        # distutils' perspective is what MACOSX_DEPLOYMENT_TARGET is set
+        # to. This makes the compatibility story a bit more sane because the
+        # machine is going to compile and link as if it were
+        # MACOSX_DEPLOYMENT_TARGET.
+        from distutils.sysconfig import get_config_vars
+        cfgvars = get_config_vars()
+
+        macver = os.environ.get('MACOSX_DEPLOYMENT_TARGET')
+        if not macver:
+            macver = cfgvars.get('MACOSX_DEPLOYMENT_TARGET')
+
+        if 1:
+            # Always calculate the release of the running machine,
+            # needed to determine if we can build fat binaries or not.
+
+            macrelease = macver
+            # Get the system version. Reading this plist is a documented
+            # way to get the system version (see the documentation for
+            # the Gestalt Manager)
+            try:
+                f = open('/System/Library/CoreServices/SystemVersion.plist')
+            except IOError:
+                # We're on a plain darwin box, fall back to the default
+                # behaviour.
+                pass
+            else:
+                m = re.search(
+                        r'<key>ProductUserVisibleVersion</key>\s*' +
+                        r'<string>(.*?)</string>', f.read())
+                f.close()
+                if m is not None:
+                    macrelease = '.'.join(m.group(1).split('.')[:2])
+                # else: fall back to the default behaviour
+
+        if not macver:
+            macver = macrelease
+
+        if macver:
+            from distutils.sysconfig import get_config_vars
+            release = macver
+            osname = "macosx"
+
+            if (macrelease + '.') >= '10.4.' and \
+                    '-arch' in get_config_vars().get('CFLAGS', '').strip():
+                # The universal build will build fat binaries, but not on
+                # systems before 10.4
+                #
+                # Try to detect 4-way universal builds, those have machine-type
+                # 'universal' instead of 'fat'.
+
+                machine = 'fat'
+                cflags = get_config_vars().get('CFLAGS')
+
+                archs = re.findall('-arch\s+(\S+)', cflags)
+                archs = tuple(sorted(set(archs)))
+
+                if len(archs) == 1:
+                    machine = archs[0]
+                elif archs == ('i386', 'ppc'):
+                    machine = 'fat'
+                elif archs == ('i386', 'x86_64'):
+                    machine = 'intel'
+                elif archs == ('i386', 'ppc', 'x86_64'):
+                    machine = 'fat3'
+                elif archs == ('ppc64', 'x86_64'):
+                    machine = 'fat64'
+                elif archs == ('i386', 'ppc', 'ppc64', 'x86_64'):
+                    machine = 'universal'
+                else:
+                    raise ValueError(
+                       "Don't know machine value for archs=%r"%(archs,))
 
-def get_platform():
-    """Return a string that identifies the current platform.
+            elif machine == 'i386':
+                # On OSX the machine type returned by uname is always the
+                # 32-bit variant, even if the executable architecture is
+                # the 64-bit variant
+                if sys.maxsize >= 2**32:
+                    machine = 'x86_64'
 
-    By default, will return the value returned by sysconfig.get_platform(),
-    but it can be changed by calling set_platform().
-    """
-    global _PLATFORM
-    if _PLATFORM is None:
-        _PLATFORM = _sysconfig.get_platform()
-    return _PLATFORM
+            elif machine in ('PowerPC', 'Power_Macintosh'):
+                # Pick a sane name for the PPC architecture.
+                machine = 'ppc'
 
-def set_platform(identifier):
-    """Sets the platform string identifier returned by get_platform().
+                # See 'i386' case
+                if sys.maxsize >= 2**32:
+                    machine = 'ppc64'
 
-    Note that this change doesn't impact the value returned by
-    sysconfig.get_platform() and is local to Distutils
-    """
-    global _PLATFORM
-    _PLATFORM = identifier
+    return "%s-%s-%s" % (osname, release, machine)
+
+# get_platform ()
 
-def convert_path(pathname):
-    """Return 'pathname' as a name that will work on the native filesystem.
 
+def convert_path (pathname):
+    """Return 'pathname' as a name that will work on the native filesystem,
     i.e. split it on '/' and put it back together again using the current
     directory separator.  Needed because filenames in the setup script are
     always supplied in Unix style, and have to be converted to the local
@@ -64,12 +206,12 @@
         return os.curdir
     return os.path.join(*paths)
 
+# convert_path ()
 
-def change_root(new_root, pathname):
-    """Return 'pathname' with 'new_root' prepended.
 
-    If 'pathname' is relative, this is equivalent to
-    "os.path.join(new_root,pathname)".
+def change_root (new_root, pathname):
+    """Return 'pathname' with 'new_root' prepended.  If 'pathname' is
+    relative, this is equivalent to "os.path.join(new_root,pathname)".
     Otherwise, it requires making 'pathname' relative and then joining the
     two, which is tricky on DOS/Windows and Mac OS.
     """
@@ -91,16 +233,23 @@
             path = path[1:]
         return os.path.join(new_root, path)
 
-    else:
-        raise DistutilsPlatformError("nothing known about "
-                                     "platform '%s'" % os.name)
+    elif os.name == 'mac':
+        if not os.path.isabs(pathname):
+            return os.path.join(new_root, pathname)
+        else:
+            # Chop off volume name from start of path
+            elements = pathname.split(":", 1)
+            pathname = ":" + elements[1]
+            return os.path.join(new_root, pathname)
 
-_environ_checked = 0
+    else:
+        raise DistutilsPlatformError("nothing known about platform '%s'" % os.name)
 
-def check_environ():
-    """Ensure that 'os.environ' has all the environment variables needed.
 
-    We guarantee that users can use in config files, command-line options,
+_environ_checked = 0
+def check_environ ():
+    """Ensure that 'os.environ' has all the environment variables we
+    guarantee that users can use in config files, command-line options,
     etc.  Currently this includes:
       HOME - user's home directory (Unix only)
       PLAT - description of the current platform, including hardware
@@ -115,14 +264,14 @@
         os.environ['HOME'] = pwd.getpwuid(os.getuid())[5]
 
     if 'PLAT' not in os.environ:
-        os.environ['PLAT'] = _sysconfig.get_platform()
+        os.environ['PLAT'] = get_platform()
 
     _environ_checked = 1
 
-def subst_vars(s, local_vars):
-    """Perform shell/Perl-style variable substitution on 'string'.
 
-    Every occurrence of '$' followed by a name is considered a variable, and
+def subst_vars (s, local_vars):
+    """Perform shell/Perl-style variable substitution on 'string'.  Every
+    occurrence of '$' followed by a name is considered a variable, and
     variable is substituted by the value found in the 'local_vars'
     dictionary, or in 'os.environ' if it's not in 'local_vars'.
     'os.environ' is first checked/augmented to guarantee that it contains
@@ -142,11 +291,12 @@
     except KeyError as var:
         raise ValueError("invalid variable '$%s'" % var)
 
-def grok_environment_error(exc, prefix="error: "):
-    """Generate a useful error message from an EnvironmentError.
+# subst_vars ()
+
 
-    This will generate an IOError or an OSError exception object.
-    Handles Python 1.5.1 and 1.5.2 styles, and
+def grok_environment_error (exc, prefix="error: "):
+    """Generate a useful error message from an EnvironmentError (IOError or
+    OSError) exception object.  Handles Python 1.5.1 and 1.5.2 styles, and
     does what it can to deal with exception objects that don't have a
     filename (which happens when the error is due to a two-file operation,
     such as 'rename()' or 'link()'.  Returns the error message as a string
@@ -165,20 +315,18 @@
 
     return error
 
+
 # Needed by 'split_quoted()'
 _wordchars_re = _squote_re = _dquote_re = None
-
 def _init_regex():
     global _wordchars_re, _squote_re, _dquote_re
     _wordchars_re = re.compile(r'[^\\\'\"%s ]*' % string.whitespace)
     _squote_re = re.compile(r"'(?:[^'\\]|\\.)*'")
     _dquote_re = re.compile(r'"(?:[^"\\]|\\.)*"')
 
-def split_quoted(s):
+def split_quoted (s):
     """Split a string up according to Unix shell-like rules for quotes and
-    backslashes.
-
-    In short: words are delimited by spaces, as long as those
+    backslashes.  In short: words are delimited by spaces, as long as those
     spaces are not escaped by a backslash, or inside a quoted string.
     Single and double quotes are equivalent, and the quote characters can
     be backslash-escaped.  The backslash is stripped from any two-character
@@ -186,6 +334,7 @@
     characters are stripped from any quoted string.  Returns a list of
     words.
     """
+
     # This is a nice algorithm for splitting up a single string, since it
     # doesn't require character-by-character examination.  It was a little
     # bit of a brain-bender to get it working right, though...
@@ -233,12 +382,13 @@
 
     return words
 
+# split_quoted ()
 
-def execute(func, args, msg=None, verbose=0, dry_run=0):
-    """Perform some action that affects the outside world.
 
-    eg. by writing to the filesystem).  Such actions are special because
-    they are disabled by the 'dry_run' flag.  This method takes care of all
+def execute (func, args, msg=None, verbose=0, dry_run=0):
+    """Perform some action that affects the outside world (eg.  by
+    writing to the filesystem).  Such actions are special because they
+    are disabled by the 'dry_run' flag.  This method takes care of all
     that bureaucracy for you; all you have to do is supply the
     function to call and an argument tuple for it (to embody the
     "external action" being performed), and an optional message to
@@ -254,7 +404,7 @@
         func(*args)
 
 
-def strtobool(val):
+def strtobool (val):
     """Convert a string representation of truth to true (1) or false (0).
 
     True values are 'y', 'yes', 't', 'true', 'on', and '1'; false values
@@ -270,13 +420,15 @@
         raise ValueError("invalid truth value %r" % (val,))
 
 
-def byte_compile(py_files, optimize=0, force=0, prefix=None, base_dir=None,
-                  verbose=1, dry_run=0, direct=None):
+def byte_compile (py_files,
+                  optimize=0, force=0,
+                  prefix=None, base_dir=None,
+                  verbose=1, dry_run=0,
+                  direct=None):
     """Byte-compile a collection of Python source files to either .pyc
-    or .pyo files in the same directory.
-
-    'py_files' is a list of files to compile; any files that don't end in
-    ".py" are silently skipped. 'optimize' must be one of the following:
+    or .pyo files in the same directory.  'py_files' is a list of files
+    to compile; any files that don't end in ".py" are silently skipped.
+    'optimize' must be one of the following:
       0 - don't optimize (generate .pyc)
       1 - normal optimization (like "python -O")
       2 - extra optimization (like "python -OO")
@@ -392,8 +544,8 @@
             dfile = file
             if prefix:
                 if file[:len(prefix)] != prefix:
-                    raise ValueError("invalid prefix: filename %r doesn't "
-                                     "start with %r" % (file, prefix))
+                    raise ValueError("invalid prefix: filename %r doesn't start with %r"
+                           % (file, prefix))
                 dfile = dfile[len(prefix):]
             if base_dir:
                 dfile = os.path.join(base_dir, dfile)
@@ -408,8 +560,9 @@
                     log.debug("skipping byte-compilation of %s to %s",
                               file, cfile_base)
 
+# byte_compile ()
 
-def rfc822_escape(header):
+def rfc822_escape (header):
     """Return a version of the string escaped for inclusion in an
     RFC-822 header, by ensuring there are 8 spaces space after each newline.
     """
@@ -417,56 +570,6 @@
     sep = '\n' + 8 * ' '
     return sep.join(lines)
 
-_RE_VERSION = re.compile(b'(\d+\.\d+(\.\d+)*)')
-_MAC_OS_X_LD_VERSION = re.compile(b'^@\(#\)PROGRAM:ld  PROJECT:ld64-((\d+)(\.\d+)*)')
-
-def _find_ld_version():
-    """Finds the ld version. The version scheme differs under Mac OSX."""
-    if sys.platform == 'darwin':
-        return _find_exe_version('ld -v', _MAC_OS_X_LD_VERSION)
-    else:
-        return _find_exe_version('ld -v')
-
-def _find_exe_version(cmd, pattern=_RE_VERSION):
-    """Find the version of an executable by running `cmd` in the shell.
-
-    `pattern` is a compiled regular expression. If not provided, default
-    to _RE_VERSION. If the command is not found, or the output does not
-    match the mattern, returns None.
-    """
-    from subprocess import Popen, PIPE
-    executable = cmd.split()[0]
-    if find_executable(executable) is None:
-        return None
-    pipe = Popen(cmd, shell=True, stdout=PIPE, stderr=PIPE)
-    try:
-        stdout, stderr = pipe.stdout.read(), pipe.stderr.read()
-    finally:
-        pipe.stdout.close()
-        pipe.stderr.close()
-    # some commands like ld under MacOS X, will give the
-    # output in the stderr, rather than stdout.
-    if stdout != b'':
-        out_string = stdout
-    else:
-        out_string = stderr
-
-    result = pattern.search(out_string)
-    if result is None:
-        return None
-    return LooseVersion(result.group(1).decode())
-
-def get_compiler_versions():
-    """Returns a tuple providing the versions of gcc, ld and dllwrap
-
-    For each command, if a command is not found, None is returned.
-    Otherwise a LooseVersion instance is returned.
-    """
-    gcc = _find_exe_version('gcc -dumpversion')
-    ld = _find_ld_version()
-    dllwrap = _find_exe_version('dllwrap --version')
-    return gcc, ld, dllwrap
-
 # 2to3 support
 
 def run_2to3(files, fixer_names=None, options=None, explicit=None):

Modified: python/branches/py3k/setup.py
==============================================================================
--- python/branches/py3k/setup.py	(original)
+++ python/branches/py3k/setup.py	Thu Jul 22 14:50:05 2010
@@ -222,7 +222,7 @@
         if compiler is not None:
             (ccshared,cflags) = sysconfig.get_config_vars('CCSHARED','CFLAGS')
             args['compiler_so'] = compiler + ' ' + ccshared + ' ' + cflags
-        self.compiler_obj.set_executables(**args)
+        self.compiler.set_executables(**args)
 
         build_ext.build_extensions(self)
 
@@ -343,8 +343,8 @@
 
     def detect_modules(self):
         # Ensure that /usr/local is always used
-        add_dir_to_list(self.compiler_obj.library_dirs, '/usr/local/lib')
-        add_dir_to_list(self.compiler_obj.include_dirs, '/usr/local/include')
+        add_dir_to_list(self.compiler.library_dirs, '/usr/local/lib')
+        add_dir_to_list(self.compiler.include_dirs, '/usr/local/include')
 
         # Add paths specified in the environment variables LDFLAGS and
         # CPPFLAGS for header and library files.
@@ -353,9 +353,9 @@
         # the environment variable is not set even though the value were passed
         # into configure and stored in the Makefile (issue found on OS X 10.3).
         for env_var, arg_name, dir_list in (
-                ('LDFLAGS', '-R', self.compiler_obj.runtime_library_dirs),
-                ('LDFLAGS', '-L', self.compiler_obj.library_dirs),
-                ('CPPFLAGS', '-I', self.compiler_obj.include_dirs)):
+                ('LDFLAGS', '-R', self.compiler.runtime_library_dirs),
+                ('LDFLAGS', '-L', self.compiler.library_dirs),
+                ('CPPFLAGS', '-I', self.compiler.include_dirs)):
             env_val = sysconfig.get_config_var(env_var)
             if env_val:
                 # To prevent optparse from raising an exception about any
@@ -381,19 +381,19 @@
                         add_dir_to_list(dir_list, directory)
 
         if os.path.normpath(sys.prefix) != '/usr':
-            add_dir_to_list(self.compiler_obj.library_dirs,
+            add_dir_to_list(self.compiler.library_dirs,
                             sysconfig.get_config_var("LIBDIR"))
-            add_dir_to_list(self.compiler_obj.include_dirs,
+            add_dir_to_list(self.compiler.include_dirs,
                             sysconfig.get_config_var("INCLUDEDIR"))
 
         # lib_dirs and inc_dirs are used to search for files;
         # if a file is found in one of those directories, it can
         # be assumed that no additional -I,-L directives are needed.
-        lib_dirs = self.compiler_obj.library_dirs + [
+        lib_dirs = self.compiler.library_dirs + [
             '/lib64', '/usr/lib64',
             '/lib', '/usr/lib',
             ]
-        inc_dirs = self.compiler_obj.include_dirs + ['/usr/include']
+        inc_dirs = self.compiler.include_dirs + ['/usr/include']
         exts = []
         missing = []
 
@@ -525,7 +525,7 @@
         exts.append( Extension('audioop', ['audioop.c']) )
 
         # readline
-        do_readline = self.compiler_obj.find_library_file(lib_dirs, 'readline')
+        do_readline = self.compiler.find_library_file(lib_dirs, 'readline')
         readline_termcap_library = ""
         curses_library = ""
         # Determine if readline is already linked against curses or tinfo.
@@ -552,11 +552,11 @@
         # use the same library for the readline and curses modules.
         if 'curses' in readline_termcap_library:
             curses_library = readline_termcap_library
-        elif self.compiler_obj.find_library_file(lib_dirs, 'ncursesw'):
+        elif self.compiler.find_library_file(lib_dirs, 'ncursesw'):
             curses_library = 'ncursesw'
-        elif self.compiler_obj.find_library_file(lib_dirs, 'ncurses'):
+        elif self.compiler.find_library_file(lib_dirs, 'ncurses'):
             curses_library = 'ncurses'
-        elif self.compiler_obj.find_library_file(lib_dirs, 'curses'):
+        elif self.compiler.find_library_file(lib_dirs, 'curses'):
             curses_library = 'curses'
 
         if platform == 'darwin':
@@ -586,7 +586,7 @@
                 pass # Issue 7384: Already linked against curses or tinfo.
             elif curses_library:
                 readline_libs.append(curses_library)
-            elif self.compiler_obj.find_library_file(lib_dirs +
+            elif self.compiler.find_library_file(lib_dirs +
                                                      ['/usr/lib/termcap'],
                                                      'termcap'):
                 readline_libs.append('termcap')
@@ -599,7 +599,7 @@
 
         # crypt module.
 
-        if self.compiler_obj.find_library_file(lib_dirs, 'crypt'):
+        if self.compiler.find_library_file(lib_dirs, 'crypt'):
             libs = ['crypt']
         else:
             libs = []
@@ -627,7 +627,7 @@
                                ['/usr/kerberos/include'])
             if krb5_h:
                 ssl_incs += krb5_h
-        ssl_libs = find_library_file(self.compiler_obj, 'ssl',lib_dirs,
+        ssl_libs = find_library_file(self.compiler, 'ssl',lib_dirs,
                                      ['/usr/local/ssl/lib',
                                       '/usr/contrib/ssl/lib/'
                                      ] )
@@ -952,7 +952,7 @@
                 os.path.join(sqlite_incdir, '..', '..', 'lib64'),
                 os.path.join(sqlite_incdir, '..', '..', 'lib'),
             ]
-            sqlite_libfile = self.compiler_obj.find_library_file(
+            sqlite_libfile = self.compiler.find_library_file(
                                 sqlite_dirs_to_check + lib_dirs, 'sqlite3')
             if sqlite_libfile:
                 sqlite_libdir = [os.path.abspath(os.path.dirname(sqlite_libfile))]
@@ -1014,7 +1014,7 @@
                 if cand == "ndbm":
                     if find_file("ndbm.h", inc_dirs, []) is not None:
                         # Some systems have -lndbm, others don't
-                        if self.compiler_obj.find_library_file(lib_dirs,
+                        if self.compiler.find_library_file(lib_dirs,
                                                                'ndbm'):
                             ndbm_libs = ['ndbm']
                         else:
@@ -1028,9 +1028,9 @@
                         break
 
                 elif cand == "gdbm":
-                    if self.compiler_obj.find_library_file(lib_dirs, 'gdbm'):
+                    if self.compiler.find_library_file(lib_dirs, 'gdbm'):
                         gdbm_libs = ['gdbm']
-                        if self.compiler_obj.find_library_file(lib_dirs,
+                        if self.compiler.find_library_file(lib_dirs,
                                                                'gdbm_compat'):
                             gdbm_libs.append('gdbm_compat')
                         if find_file("gdbm/ndbm.h", inc_dirs, []) is not None:
@@ -1071,7 +1071,7 @@
 
         # Anthony Baxter's gdbm module.  GNU dbm(3) will require -lgdbm:
         if ('gdbm' in dbm_order and
-            self.compiler_obj.find_library_file(lib_dirs, 'gdbm')):
+            self.compiler.find_library_file(lib_dirs, 'gdbm')):
             exts.append( Extension('_gdbm', ['_gdbmmodule.c'],
                                    libraries = ['gdbm'] ) )
         else:
@@ -1087,7 +1087,7 @@
             # Sun yellow pages. Some systems have the functions in libc.
             if (platform not in ['cygwin', 'qnx6'] and
                 find_file('rpcsvc/yp_prot.h', inc_dirs, []) is not None):
-                if (self.compiler_obj.find_library_file(lib_dirs, 'nsl')):
+                if (self.compiler.find_library_file(lib_dirs, 'nsl')):
                     libs = ['nsl']
                 else:
                     libs = []
@@ -1112,9 +1112,9 @@
         elif curses_library == 'curses' and platform != 'darwin':
                 # OSX has an old Berkeley curses, not good enough for
                 # the _curses module.
-            if (self.compiler_obj.find_library_file(lib_dirs, 'terminfo')):
+            if (self.compiler.find_library_file(lib_dirs, 'terminfo')):
                 curses_libs = ['curses', 'terminfo']
-            elif (self.compiler_obj.find_library_file(lib_dirs, 'termcap')):
+            elif (self.compiler.find_library_file(lib_dirs, 'termcap')):
                 curses_libs = ['curses', 'termcap']
             else:
                 curses_libs = ['curses']
@@ -1126,7 +1126,7 @@
 
         # If the curses module is enabled, check for the panel module
         if (module_enabled(exts, '_curses') and
-            self.compiler_obj.find_library_file(lib_dirs, panel_library)):
+            self.compiler.find_library_file(lib_dirs, panel_library)):
             exts.append( Extension('_curses_panel', ['_curses_panel.c'],
                                    libraries = [panel_library] + curses_libs) )
         else:
@@ -1159,7 +1159,7 @@
                     version = line.split()[2]
                     break
             if version >= version_req:
-                if (self.compiler_obj.find_library_file(lib_dirs, 'z')):
+                if (self.compiler.find_library_file(lib_dirs, 'z')):
                     if sys.platform == "darwin":
                         zlib_extra_link_args = ('-Wl,-search_paths_first',)
                     else:
@@ -1191,7 +1191,7 @@
                                extra_link_args = extra_link_args) )
 
         # Gustavo Niemeyer's bz2 module.
-        if (self.compiler_obj.find_library_file(lib_dirs, 'bz2')):
+        if (self.compiler.find_library_file(lib_dirs, 'bz2')):
             if sys.platform == "darwin":
                 bz2_extra_link_args = ('-Wl,-search_paths_first',)
             else:
@@ -1452,9 +1452,9 @@
         tcllib = tklib = tcl_includes = tk_includes = None
         for version in ['8.6', '86', '8.5', '85', '8.4', '84', '8.3', '83',
                         '8.2', '82', '8.1', '81', '8.0', '80']:
-            tklib = self.compiler_obj.find_library_file(lib_dirs,
+            tklib = self.compiler.find_library_file(lib_dirs,
                                                         'tk' + version)
-            tcllib = self.compiler_obj.find_library_file(lib_dirs,
+            tcllib = self.compiler.find_library_file(lib_dirs,
                                                          'tcl' + version)
             if tklib and tcllib:
                 # Exit the loop when we've found the Tcl/Tk libraries
@@ -1513,11 +1513,11 @@
                 return
 
         # Check for BLT extension
-        if self.compiler_obj.find_library_file(lib_dirs + added_lib_dirs,
+        if self.compiler.find_library_file(lib_dirs + added_lib_dirs,
                                                'BLT8.0'):
             defs.append( ('WITH_BLT', 1) )
             libs.append('BLT8.0')
-        elif self.compiler_obj.find_library_file(lib_dirs + added_lib_dirs,
+        elif self.compiler.find_library_file(lib_dirs + added_lib_dirs,
                                                 'BLT'):
             defs.append( ('WITH_BLT', 1) )
             libs.append('BLT')
@@ -1572,7 +1572,7 @@
                              ]]
 
         # Add .S (preprocessed assembly) to C compiler source extensions.
-        self.compiler_obj.src_extensions.append('.S')
+        self.compiler.src_extensions.append('.S')
 
         include_dirs = [os.path.join(ffi_srcdir, 'include'),
                         os.path.join(ffi_srcdir, 'powerpc')]
@@ -1617,7 +1617,7 @@
                 exec(f.read(), globals(), fficonfig)
 
             # Add .S (preprocessed assembly) to C compiler source extensions.
-            self.compiler_obj.src_extensions.append('.S')
+            self.compiler.src_extensions.append('.S')
 
             include_dirs = [os.path.join(ffi_builddir, 'include'),
                             ffi_builddir,
@@ -1699,7 +1699,7 @@
         ffi_lib = None
         if ffi_inc is not None:
             for lib_name in ('ffi_convenience', 'ffi_pic', 'ffi'):
-                if (self.compiler_obj.find_library_file(lib_dirs, lib_name)):
+                if (self.compiler.find_library_file(lib_dirs, lib_name)):
                     ffi_lib = lib_name
                     break
 

From python-checkins at python.org  Thu Jul 22 14:58:55 2010
From: python-checkins at python.org (tarek.ziade)
Date: Thu, 22 Jul 2010 14:58:55 +0200 (CEST)
Subject: [Python-checkins] r83054 - python/branches/release31-maint
Message-ID: <20100722125855.2E795F411@mail.python.org>

Author: tarek.ziade
Date: Thu Jul 22 14:58:55 2010
New Revision: 83054

Log:
Blocked revisions 83053 via svnmerge

........
  r83053 | tarek.ziade | 2010-07-22 14:50:05 +0200 (jeu., 22 juil. 2010) | 1 line
  
  reverted distutils its 3.1 state. All new work is now happening in disutils2, and distutils is now feature-frozen.
........


Modified:
   python/branches/release31-maint/   (props changed)

From python-checkins at python.org  Thu Jul 22 15:01:58 2010
From: python-checkins at python.org (nick.coghlan)
Date: Thu, 22 Jul 2010 15:01:58 +0200 (CEST)
Subject: [Python-checkins] r83055 - peps/trunk/pep-3150.txt
Message-ID: <20100722130158.26F7BF4CE@mail.python.org>

Author: nick.coghlan
Date: Thu Jul 22 15:01:57 2010
New Revision: 83055

Log:
Rephrase the new TODO items based on python-ideas thread

Modified:
   peps/trunk/pep-3150.txt

Modified: peps/trunk/pep-3150.txt
==============================================================================
--- peps/trunk/pep-3150.txt	(original)
+++ peps/trunk/pep-3150.txt	Thu Jul 22 15:01:57 2010
@@ -392,10 +392,12 @@
 * Define the expected semantics of ``break``, ``continue``, ``return``
   and ``yield`` in a ``given`` clause (i.e. syntax errors at the clause
   level, but allowed inside the appropriate compound statements)
-* Define the expected semantics of ``nonlocal`` and ``global`` in the
-  ``given`` clause
-* Define the name lookup semantics for function definitions in a
-  ``given`` clause at function, class and module scope.
+* Describe the expected semantics of ``nonlocal`` and ``global`` in the
+  ``given`` clause.
+* Describe the name lookup semantics for function definitions in a
+  ``given`` clause at function, class and module scope. In particular,
+  note the early binding effect on loop variables or other variables
+  that are rebound after the ``given`` clause is complete.
 
 
 References

From ncoghlan at gmail.com  Thu Jul 22 15:03:03 2010
From: ncoghlan at gmail.com (Nick Coghlan)
Date: Thu, 22 Jul 2010 23:03:03 +1000
Subject: [Python-checkins] r83049 - peps/trunk/pep-3151.txt
In-Reply-To: <20100722114149.64D93F1D3@mail.python.org>
References: <20100722114149.64D93F1D3@mail.python.org>
Message-ID: <AANLkTimdzx3Y5BUfCHQncMvXAnlgPkev55xxTG4QgYKd@mail.gmail.com>

On Thu, Jul 22, 2010 at 9:41 PM, antoine.pitrou
<python-checkins at python.org> wrote:
> -This list assumes step 1 is accepted in full; the exception classes
> +This list assumes `Step 1`_ is accepted in full; the exception classes
> ?described above would all derive from the now unified exception type
> ?OSError. ?It will need reworking if a partial version of step 1 is accepted
> ?instead (again, see appendix A for the current distribution of errnos
> ?and exception types).

With the latest PEP changes, this paragraph should now refer to "IOError".

Cheers,
Nick.

-- 
Nick Coghlan?? |?? ncoghlan at gmail.com?? |?? Brisbane, Australia

From python-checkins at python.org  Thu Jul 22 15:13:03 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Thu, 22 Jul 2010 15:13:03 +0200 (CEST)
Subject: [Python-checkins] r83056 - peps/trunk/pep-3151.txt
Message-ID: <20100722131303.7B04BEDD9@mail.python.org>

Author: antoine.pitrou
Date: Thu Jul 22 15:13:03 2010
New Revision: 83056

Log:
Fixes



Modified:
   peps/trunk/pep-3151.txt

Modified: peps/trunk/pep-3151.txt
==============================================================================
--- peps/trunk/pep-3151.txt	(original)
+++ peps/trunk/pep-3151.txt	Thu Jul 22 15:13:03 2010
@@ -29,8 +29,8 @@
 
 This PEP proposes changes to a part of the exception hierarchy in
 order to better embody the qualities mentioned above: the errors
-related to operating system calls (OSError, IOError, select.error, and
-all their subclasses).
+related to operating system calls (OSError, IOError, mmap.error,
+select.error, and all their subclasses).
 
 
 Rationale
@@ -339,7 +339,7 @@
 
 This list assumes `Step 1`_ is accepted in full; the exception classes
 described above would all derive from the now unified exception type
-OSError.  It will need reworking if a partial version of step 1 is accepted
+IOError.  It will need reworking if a partial version of step 1 is accepted
 instead (again, see appendix A for the current distribution of errnos
 and exception types).
 

From solipsis at pitrou.net  Thu Jul 22 15:13:25 2010
From: solipsis at pitrou.net (Antoine Pitrou)
Date: Thu, 22 Jul 2010 15:13:25 +0200
Subject: [Python-checkins] r83049 - peps/trunk/pep-3151.txt
References: <20100722114149.64D93F1D3@mail.python.org>
	<AANLkTimdzx3Y5BUfCHQncMvXAnlgPkev55xxTG4QgYKd@mail.gmail.com>
Message-ID: <20100722151325.0d2a3970@pitrou.net>

On Thu, 22 Jul 2010 23:03:03 +1000
Nick Coghlan <ncoghlan at gmail.com> wrote:
> On Thu, Jul 22, 2010 at 9:41 PM, antoine.pitrou
> <python-checkins at python.org> wrote:
> > -This list assumes step 1 is accepted in full; the exception classes
> > +This list assumes `Step 1`_ is accepted in full; the exception classes
> > ?described above would all derive from the now unified exception type
> > ?OSError. ?It will need reworking if a partial version of step 1 is accepted
> > ?instead (again, see appendix A for the current distribution of errnos
> > ?and exception types).
> 
> With the latest PEP changes, this paragraph should now refer to "IOError".

Thank you!

Antoine.



From python-checkins at python.org  Thu Jul 22 16:18:44 2010
From: python-checkins at python.org (barry.warsaw)
Date: Thu, 22 Jul 2010 16:18:44 +0200 (CEST)
Subject: [Python-checkins] r83058 - peps/trunk/pep-3149.txt
Message-ID: <20100722141844.0B2BAF3BD@mail.python.org>

Author: barry.warsaw
Date: Thu Jul 22 16:18:43 2010
New Revision: 83058

Log:
Updates based on python-dev discussion.



Modified:
   peps/trunk/pep-3149.txt

Modified: peps/trunk/pep-3149.txt
==============================================================================
--- peps/trunk/pep-3149.txt	(original)
+++ peps/trunk/pep-3149.txt	Thu Jul 22 16:18:43 2010
@@ -8,7 +8,7 @@
 Content-Type: text/x-rst
 Created: 2010-07-09
 Python-Version: 3.2
-Post-History: 2010-07-14
+Post-History: 2010-07-14, 2010-07-22
 Resolution: TBD
 
 
@@ -49,10 +49,14 @@
 For packages with extension modules, a similar differentiation is
 needed for the module's .so files.  Extension modules compiled for
 different Python major versions are incompatible with each other due
-to changes in the ABI.  While PEP 384 [2]_ defines a stable ABI, it
-will minimize, but not eliminate extension module incompatibilities
-between Python major versions.  Thus a mechanism for discriminating
-extension module file names is proposed.
+to changes in the ABI.  Different configuration/compilation options
+for the same Python version can result in different ABIs
+(e.g. --with-wide-unicode).
+
+While PEP 384 [2]_ defines a stable ABI, it will minimize, but not
+eliminate extension module incompatibilities between Python builds or
+major versions.  Thus a mechanism for discriminating extension module
+file names is proposed.
 
 
 Rationale
@@ -77,51 +81,85 @@
 A similar situation arises with shared library extensions.  Because
 extension modules are typically named `foo.so` for a `foo` extension
 module, these would also name collide if `foo` was provided for more
-than one Python version.  There are several approaches that could be
-taken to avoid this, which will be explored below, but this PEP
-proposes a fairly simple compile-time option to allow extension
-modules to live in the same file system directory and avoid any name
-collisions.
+than one Python version.
+
+In addition, because different configuration/compilation options for
+the same Python version can cause different ABIs to be presented to
+extension modules.  On POSIX systems for example, the configure
+options ``--with-pydebug``, ``--with-pymalloc``, and
+``--with-wide-unicode`` all change the ABI.  This PEP proposes to
+encode build-time options in the file name of the ``.so`` extension
+module files.
+
+PyPy [5]_ can also benefit from this PEP, allowing it to avoid name
+collisions in extension modules built for its API, but with a
+different `.so` tag.
 
 
 Proposal
 ========
 
-A new configure option is added for building Python, called
-`--with-so-abi-tag`.  This takes as an argument a unique, but
-arbitrary string, e.g.::
-
-    ./configure --with-so-abi-tag=cpython-32
-
-This string is passed into the `Makefile` and affects two aspects of
-the Python build.  First, it is compiled into `Python/dynload_shlib.c`
-where it defines some additional `.so` file names to search for when
-importing extension modules.  Second, it modifies the `Makefile`'s
-`$SO` variable, which in turn controls the `distutils` module's default
-filename when compiling extension modules.
-
-When `--with-so-abi-tag` is not given to `configure` nothing changes
-in the way the Python executable is built, or acts.  Thus, this
-configure switch is completely optional and has no effect if not used.
-
-What this allows is for distributions that want to distinguish among
-extension modules built for different versions of Python, but shared
-in the same file system path, to arrange for `.so` names that are
-unique and non-colliding.
-
-For example, let's say Python 3.2 was built with::
-
-    ./configure --with-so-abi-tag=cpython-32
-
-and Python 3.3 was built with::
-
-    ./configure --with-so-abi-tag=cpython-33
+The configure/compilation options chosen at Python interpreter
+build-time will be encoded in the shared library file name for
+extension modules.  This "tag" will appear between the module base
+name and the operation file system extension for shared libraries.
+
+The following information *MUST* be included in the shared library
+file name:
+
+ * The Python implementation (e.g. cpython, pypy, jython, etc.)
+ * The interpreter's major and minor version numbers
+
+These two fields are separated by a hyphen and no dots are to appear
+between the major and minor version numbers.  E.g. ``cpython-32``.
+
+Python implementations *MAY* include additional flags in the file name
+tag as appropriate.  For example, on POSIX systems these flags will
+also contribute to the file name:
+
+ * ``--with-pydebug`` (flag: ``d``)
+ * ``--with-pymalloc`` (flag: ``m``)
+ * ``--with-wide-unicode`` (flag: ``u``)
+
+By default in Python 3.2, ``configure`` enables ``--with-pymalloc`` so
+shared library file names would appear as ``foo.cpython-32m.so``.
+When the other two flags are also enabled, the file names would be
+``foo.cpython-32dmu.so``.
+
+(This PEP only addresses build issues on POSIX systems that use the
+``configure`` script.  While Windows or other platform support is not
+explicitly disallowed under this PEP, platform expertise is needed in
+order to evaluate, describe, and implement support on such platforms.)
+
+The shared library file name tag is used unconditionally; it cannot be
+changed.  The tag and extension module suffix are available through
+the ``sysconfig`` modules via the following variables::
+
+    >>> sysconfig.get_config_var('SO')
+    '.cpython-32mu.so'
+    >>> sysconfig.get_config_var('SOABI')
+    'cpython-32mu'
+
+Note that ``$SOABI`` contains just the tag, while ``$SO`` includes the
+platform extension for shared library files, and is the exact suffix
+added to the extension module name.
 
 For an arbitrary package `foo`, you would see these files when the
 distribution package was installed::
 
-    /usr/share/pyshared/foo.cpython-32.so
-    /usr/share/pyshared/foo.cpython-33.so
+    /usr/share/pyshared/foo.cpython-32m.so
+    /usr/share/pyshared/foo.cpython-33m.so
+
+Python's dynamic module loader will recognize and import shared
+library extension modules with a tag that matches its build-time
+options.  For backward compatibility, Python will also continue to
+import untagged extension modules, e.g. ``foo.so``.
+
+This shared library tag would be used globally for all distutils-based
+extension modules, regardless of where on the file system they are
+built.  Extension modules built by means other than distutils would
+either have to calculate the tag manually, or fallback to the
+non-tagged `.so` file name.
 
 
 Proven approach
@@ -130,26 +168,22 @@
 The approach described here is already proven, in a sense, on Debian
 and Ubuntu system where different extensions are used for debug builds
 of Python and extension modules.  Debug builds on Windows also already
-use a different file extension for dynamic libraries.
+use a different file extension for dynamic libraries, and in fact
+encoded (in a different way than proposed in this PEP) the Python
+major and minor version in the `.dll` file name.
 
 
 PEP 384
 =======
 
-PEP 384 defines a stable ABI for extension modules.  Universal
-adoption of PEP 384 would eliminate the need for this PEP because all
-extension modules would be compatible with any Python version.  In
-practice of course, it will be impossible to achieve universal
-adoption.  Older extensions may not be ported to PEP 384, or an
-extension may require Python APIs outside of PEP 384 definition.
-Therefore there will always be a (hopefully diminishing, but never
-zero) need for ABI version tagged shared libraries.
-
-Further, it is anticipated that the stable ABI will evolve over time,
-meaning that existing PEP 384 compatible extension modules may be
-incompatible with future versions of Python.  While a complete
-specification is reserved for PEP 384, here is a discussion of the
-relevant issues.
+PEP 384 defines a stable ABI for extension modules.  In theory,
+universal adoption of PEP 384 would eliminate the need for this PEP
+because all extension modules could be compatible with any Python
+version.  In practice of course, it will be impossible to achieve
+universal adoption, and as described above, different built-time flags
+still affect the ABI.  Thus even with a stable ABI, this PEP may still
+be necessary.  While a complete specification is reserved for PEP 384,
+here is a discussion of the relevant issues.
 
 PEP 384 describes a change to ``PyModule_Create()`` where ``3`` is
 passed as the API version if the extension was complied with
@@ -160,15 +194,15 @@
 search for extension modules with the ``PYTHON_ABI_VERSION`` number in
 its name.  The prefix ``abi`` is reserved for Python's use.
 
-Thus for example, an initial implementation of PEP 384, compiled with
-`--with-so-abi-tag=cpython-xy` would search for the following file
+Thus, an initial implementation of PEP 384, when Python is configured
+with the default set of flags, would search for the following file
 names when extension module `foo` is imported (in this order)::
 
+    foo.cpython-XYm.so
     foo.abi3.so
-    foo.cpython-xy.so
     foo.so
 
-The distutils [7]_ ``build_ext`` command would also have to be
+The distutils [6]_ ``build_ext`` command would also have to be
 extended to compile to shared library files with the ``abi3`` tag,
 when the module author indicates that their extension supports that
 version of the ABI.  This could be done in a backward compatible way
@@ -180,7 +214,7 @@
 Alternatives
 ============
 
-In the initial python-dev thread [8]_ where this idea was first
+In the initial python-dev thread [7]_ where this idea was first
 introduced, several alternatives were suggested.  For completeness
 they are listed here, along with the reasons for not adopting them.
 
@@ -193,12 +227,12 @@
 version of Python.  Or the symlink trick eliminated in PEP 3147 could
 be retained for just shared libraries.  This approach is rejected
 because it propagates the essential complexity that PEP 3147 tries to
-avoid, and adds yet another directory to search for all modules, even
-when the number of extension modules is much fewer than the total
-number of Python packages.  It also makes for more robust management
-when all of a package's module files live in the same directory,
-because it allows systems such as `dpkg` to detect file conflicts
-between distribution packages.
+avoid, and adds potentially several additional directories to search
+for all modules, even when the number of extension modules is much
+fewer than the total number of Python packages.  For example, builds
+were made available both with and without wide unicode, with and
+without pydebug, and with and without pymalloc, the total number of
+directories search increases substantially.
 
 
 Don't share packages with extension modules
@@ -216,17 +250,17 @@
 Also, even though all extension shared libraries will be compiled and
 distributed once for every supported Python, there's a big difference
 between duplicating the `.so` files and duplicating all `.py` files.
-The extra space increases the download time for such packages, and
-more immediately, increases the space pressures on already constrained
+The extra size increases the download time for such packages, and more
+immediately, increases the space pressures on already constrained
 distribution CD-ROMs.
 
 
 Reference implementation
 ========================
 
-Work on this code is tracked in a Bazaar branch on Launchpad [5]_
+Work on this code is tracked in a Bazaar branch on Launchpad [8]_
 until it's ready for merge into Python 3.2.  The work-in-progress diff
-can also be viewed [6]_ and is updated automatically as new changes
+can also be viewed [9]_ and is updated automatically as new changes
 are uploaded.
 
 
@@ -241,13 +275,15 @@
 
 .. [4] Debian: <http://www.debian.org>
 
-.. [5] https://code.edge.launchpad.net/~barry/python/sovers
+.. [5] http://codespeak.net/pypy/dist/pypy/doc/
+
+.. [6] http://docs.python.org/py3k/distutils/index.html
 
-.. [6] https://code.edge.launchpad.net/~barry/python/sovers/+merge/29411
+.. [7] http://mail.python.org/pipermail/python-dev/2010-June/100998.html
 
-.. [7] http://docs.python.org/py3k/distutils/index.html
+.. [8] https://code.edge.launchpad.net/~barry/python/sovers
 
-.. [8] http://mail.python.org/pipermail/python-dev/2010-June/100998.html
+.. [9] https://code.edge.launchpad.net/~barry/python/sovers/+merge/29411
 
 
 Copyright

From python-checkins at python.org  Thu Jul 22 17:38:28 2010
From: python-checkins at python.org (brian.curtin)
Date: Thu, 22 Jul 2010 17:38:28 +0200 (CEST)
Subject: [Python-checkins] r83059 -
	python/branches/py3k/Lib/test/test_sysconfig.py
Message-ID: <20100722153828.80640F503@mail.python.org>

Author: brian.curtin
Date: Thu Jul 22 17:38:28 2010
New Revision: 83059

Log:
Skip this test as it doesn't apply to Windows. It was added for
#9189 for some GCC flags.


Modified:
   python/branches/py3k/Lib/test/test_sysconfig.py

Modified: python/branches/py3k/Lib/test/test_sysconfig.py
==============================================================================
--- python/branches/py3k/Lib/test/test_sysconfig.py	(original)
+++ python/branches/py3k/Lib/test/test_sysconfig.py	Thu Jul 22 17:38:28 2010
@@ -277,6 +277,7 @@
             _main()
         self.assertTrue(len(output.getvalue().split('\n')) > 0)
 
+    @unittest.skipIf(sys.platform == "win32", "Does not apply to Windows")
     def test_ldshared_value(self):
         ldflags = sysconfig.get_config_var('LDFLAGS')
         ldshared = sysconfig.get_config_var('LDSHARED')

From python-checkins at python.org  Thu Jul 22 18:29:05 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Thu, 22 Jul 2010 18:29:05 +0200 (CEST)
Subject: [Python-checkins] r83060 - peps/trunk/pep-0382.txt
Message-ID: <20100722162905.1D66DF303@mail.python.org>

Author: martin.v.loewis
Date: Thu Jul 22 18:29:04 2010
New Revision: 83060

Log:
Specify import hook support.


Modified:
   peps/trunk/pep-0382.txt

Modified: peps/trunk/pep-0382.txt
==============================================================================
--- peps/trunk/pep-0382.txt	(original)
+++ peps/trunk/pep-0382.txt	Thu Jul 22 18:29:04 2010
@@ -114,6 +114,12 @@
 prepended to the path. In addition, the (possibly dotted) names of all
 namespace packages are added to the set sys.namespace_packages.
 
+If the first directory found on the path only contains an __init__.py
+and no \*.pth file, searching the path stops; IOW, namespace packages
+must include at least one .pth file. If both \*.pth files and an
+__init__.py had been found, search continues looking for further
+.pth files.
+
 No other change to the importing mechanism is made; searching modules
 (including __init__.py) will continue to stop at the first module
 encountered. In summary, the process import a package foo works like
@@ -121,13 +127,48 @@
 
  1. sys.path is search for a directory foo, or a file foo.<ext>.
     If a file is found and no directory, it is treated as a module, and imported.
- 2. if it is a directory, it checks for \*.pth files. If it finds
-    any, a package is created, and its __path__ is extended.
- 3. The __init__ module is imported; this import will search the
-    __path__ that got computed already.
- 4. If neither a \*.pth file nor an __init__.py was found, the
+ 2. if it is a directory, it checks for both \*.pth and an __init__ file. If it finds
+    only \*.pth files, a package is created, and its __path__ is extended.
+ 3. If neither a \*.pth file nor an __init__.py was found, the
     directory is skipped, and search for the module/package
-    continues.
+    continues. If an __init__.py was found, further search only looks
+    for \*.pth files.
+ 4. If an __init__ module was found, it is imported, with __path__
+    being initialized to the path computed from the \*.pth files.
+
+Impact on Import Hooks
+----------------------
+
+Both loaders and finders as defined in PEP 302 will need to be changed
+to support namespace packages. Failure to comform to the protocol
+below might cause a package not being recognized as a namespace
+package; loaders and finders not supporting this protocol must raise
+AttributeError when the functions below get accessed.
+
+Finders need to support looking for \*.pth files in step 1 of above
+algorithm. To do so, a finder must support a method:
+
+   finder.find_path(fullname, path=None)
+
+This method will be called in the same manner as find_module, and it
+must return a list of strings, each representing the contents of one
+\*.pth file. If fullname is not found, is not a package, or does not
+have any \*.pth files, None must be returned.
+
+If any \*.pth files are found, but no loader was returned from
+find_module, a package is created and initialized with the path.
+
+If a loader was return, but no \*.pth files, load_module is called
+as defined in PEP 302.
+
+If both \*.pth files where found, and a loader was returned, a new
+method is called on the loader:
+
+   loader.load_module_with_path(load_module, path)
+
+where the path parameter is the list of strings containing the
+contents of all \*.pth files.
+
 
 Discussion
 ==========
@@ -159,6 +200,7 @@
 users of this feature are encouraged to comment on this particular
 question.
 
+
 Copyright
 =========
 

From python-checkins at python.org  Thu Jul 22 18:51:14 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Thu, 22 Jul 2010 18:51:14 +0200 (CEST)
Subject: [Python-checkins] r83061 - python/branches/pep-0384
Message-ID: <20100722165114.AC2A2CA5E@mail.python.org>

Author: martin.v.loewis
Date: Thu Jul 22 18:51:14 2010
New Revision: 83061

Log:
Drop branch to recreate it.

Removed:
   python/branches/pep-0384/

From python-checkins at python.org  Thu Jul 22 18:51:46 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Thu, 22 Jul 2010 18:51:46 +0200 (CEST)
Subject: [Python-checkins] r83062 - python/branches/pep-0384
Message-ID: <20100722165146.AE90ACA5E@mail.python.org>

Author: martin.v.loewis
Date: Thu Jul 22 18:51:46 2010
New Revision: 83062

Log:
Create branch for PEP 384

Added:
   python/branches/pep-0384/
      - copied from r83061, /python/branches/py3k/

From python-checkins at python.org  Thu Jul 22 18:53:27 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Thu, 22 Jul 2010 18:53:27 +0200 (CEST)
Subject: [Python-checkins] r83063 - python/branches/pep-0384
Message-ID: <20100722165327.4CEECF1D3@mail.python.org>

Author: martin.v.loewis
Date: Thu Jul 22 18:53:27 2010
New Revision: 83063

Log:
Initialized merge tracking via "svnmerge" with revisions "1-83061" from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k


Modified:
   python/branches/pep-0384/   (props changed)

From python-checkins at python.org  Fri Jul 23 00:11:32 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Fri, 23 Jul 2010 00:11:32 +0200 (CEST)
Subject: [Python-checkins] r83064 -
	tracker/instances/python-dev/lib/openid2rp.py
Message-ID: <20100722221132.999B8EE986@mail.python.org>

Author: martin.v.loewis
Date: Fri Jul 23 00:11:32 2010
New Revision: 83064

Log:
Incorporate bugfixes from upstream.


Modified:
   tracker/instances/python-dev/lib/openid2rp.py

Modified: tracker/instances/python-dev/lib/openid2rp.py
==============================================================================
--- tracker/instances/python-dev/lib/openid2rp.py	(original)
+++ tracker/instances/python-dev/lib/openid2rp.py	Fri Jul 23 00:11:32 2010
@@ -151,7 +151,7 @@
     def do_meta(self, attrs):
         attrs = dict(attrs)
         # Yadis 6.2.5 option 1: meta tag
-        if attrs['http-equiv'].lower() == 'x-xrds-location':
+        if attrs.get('http-equiv','').lower() == 'x-xrds-location':
             self.xrds_location = attrs['content']
 
 def discover(url):
@@ -340,6 +340,9 @@
     if 'error' in data:
         raise ValueError, "associate failed: "+data['error']
     if url.startswith('http:'):
+        enc_mac_key = data.get('enc_mac_key')
+        if not enc_mac_key:
+            raise ValueError, "Provider protocol error: not using DH-SHA1"
         enc_mac_key = base64.b64decode(data['enc_mac_key'])
         dh_server_public = unbtwoc(base64.b64decode(data['dh_server_public']))
         # shared secret: sha1(2^(server_priv*priv) mod prime) xor enc_mac_key

From solipsis at pitrou.net  Fri Jul 23 01:23:54 2010
From: solipsis at pitrou.net (solipsis at pitrou.net)
Date: Fri, 23 Jul 2010 01:23:54 +0200 (CEST)
Subject: [Python-checkins] Daily py3k reference leaks (r83059): sum=0
Message-ID: <20100722232354.A64D91771F@ns6635.ovh.net>

py3k results for svn r83059 (hg cset afdf9987bbe6)
--------------------------------------------------



Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/antoine/py3k/refleaks/reflogulsEhy', '-x']

From python-checkins at python.org  Fri Jul 23 10:46:35 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 23 Jul 2010 10:46:35 +0200 (CEST)
Subject: [Python-checkins] r83065 - python/branches/py3k/Lib/ntpath.py
Message-ID: <20100723084635.3CACAF42D@mail.python.org>

Author: georg.brandl
Date: Fri Jul 23 10:46:35 2010
New Revision: 83065

Log:
Use augassign.

Modified:
   python/branches/py3k/Lib/ntpath.py

Modified: python/branches/py3k/Lib/ntpath.py
==============================================================================
--- python/branches/py3k/Lib/ntpath.py	(original)
+++ python/branches/py3k/Lib/ntpath.py	Fri Jul 23 10:46:35 2010
@@ -275,7 +275,7 @@
     # set i to index beyond p's last slash
     i = len(p)
     while i and p[i-1] not in seps:
-        i = i - 1
+        i -= 1
     head, tail = p[:i], p[i:]  # now tail has no slashes
     # remove trailing slashes from head, unless it's all slashes
     head2 = head
@@ -366,7 +366,7 @@
         return path
     i, n = 1, len(path)
     while i < n and path[i] not in _get_bothseps(path):
-        i = i + 1
+        i += 1
 
     if 'HOME' in os.environ:
         userhome = os.environ['HOME']
@@ -435,21 +435,21 @@
             pathlen = len(path)
             try:
                 index = path.index(c)
-                res = res + c + path[:index + 1]
+                res += c + path[:index + 1]
             except ValueError:
-                res = res + path
+                res += path
                 index = pathlen - 1
         elif c == percent:  # variable or '%'
             if path[index + 1:index + 2] == percent:
-                res = res + c
-                index = index + 1
+                res += c
+                index += 1
             else:
                 path = path[index+1:]
                 pathlen = len(path)
                 try:
                     index = path.index(percent)
                 except ValueError:
-                    res = res + percent + path
+                    res += percent + path
                     index = pathlen - 1
                 else:
                     var = path[:index]
@@ -461,11 +461,11 @@
                         value = '%' + var + '%'
                     if isinstance(path, bytes):
                         value = value.encode('ascii')
-                    res = res + value
+                    res += value
         elif c == dollar:  # variable or '$$'
             if path[index + 1:index + 2] == dollar:
-                res = res + c
-                index = index + 1
+                res += c
+                index += 1
             elif path[index + 1:index + 2] == brace:
                 path = path[index+2:]
                 pathlen = len(path)
@@ -483,23 +483,23 @@
                         value = '${' + var + '}'
                     if isinstance(path, bytes):
                         value = value.encode('ascii')
-                    res = res + value
+                    res += value
                 except ValueError:
                     if isinstance(path, bytes):
-                        res = res + b'${' + path
+                        res += b'${' + path
                     else:
-                        res = res + '${' + path
+                        res += '${' + path
                     index = pathlen - 1
             else:
                 var = ''
-                index = index + 1
+                index += 1
                 c = path[index:index + 1]
                 while c and c in varchars:
                     if isinstance(path, bytes):
-                        var = var + c.decode('ascii')
+                        var += c.decode('ascii')
                     else:
-                        var = var + c
-                    index = index + 1
+                        var += c
+                    index += 1
                     c = path[index:index + 1]
                 if var in os.environ:
                     value = os.environ[var]
@@ -507,12 +507,12 @@
                     value = '$' + var
                 if isinstance(path, bytes):
                     value = value.encode('ascii')
-                res = res + value
+                res += value
                 if c:
-                    index = index - 1
+                    index -= 1
         else:
-            res = res + c
-        index = index + 1
+            res += c
+        index += 1
     return res
 
 
@@ -529,7 +529,7 @@
 
     # collapse initial backslashes
     if path.startswith(sep):
-        prefix = prefix + sep
+        prefix += sep
         path = path.lstrip(sep)
 
     comps = path.split(sep)

From python-checkins at python.org  Fri Jul 23 11:43:17 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Fri, 23 Jul 2010 11:43:17 +0200 (CEST)
Subject: [Python-checkins] r83066 - in python/branches/py3k/Lib/distutils:
	sysconfig.py tests/test_sysconfig.py
Message-ID: <20100723094317.83B80EE981@mail.python.org>

Author: ronald.oussoren
Date: Fri Jul 23 11:43:17 2010
New Revision: 83066

Log:
Ensure that the Makefile variable expansion
in distutils.sysconfig matches that in the 
toplevel sysconfig module.

Without this patch universal builds on OSX are
broken.

Als add a test that checks that the two version
of get_config_vars agree on important values.


Modified:
   python/branches/py3k/Lib/distutils/sysconfig.py
   python/branches/py3k/Lib/distutils/tests/test_sysconfig.py

Modified: python/branches/py3k/Lib/distutils/sysconfig.py
==============================================================================
--- python/branches/py3k/Lib/distutils/sysconfig.py	(original)
+++ python/branches/py3k/Lib/distutils/sysconfig.py	Fri Jul 23 11:43:17 2010
@@ -300,6 +300,12 @@
                 else:
                     done[n] = v
 
+    # Variables with a 'PY_' prefix in the makefile. These need to
+    # be made available without that prefix through sysconfig.
+    # Special care is needed to ensure that variable expansion works, even
+    # if the expansion uses the name without a prefix.
+    renamed_variables = ('CFLAGS', 'LDFLAGS', 'CPPFLAGS')
+
     # do variable interpolation here
     while notdone:
         for name in list(notdone):
@@ -316,6 +322,16 @@
                 elif n in os.environ:
                     # do it like make: fall back to environment
                     item = os.environ[n]
+
+                elif n in renamed_variables:
+                    if name.startswith('PY_') and name[3:] in renamed_variables:
+                        item = ""
+
+                    elif 'PY_' + n in notdone:
+                        found = False
+
+                    else:
+                        item = str(done['PY_' + n])
                 else:
                     done[n] = item = ""
                 if found:
@@ -330,6 +346,13 @@
                         else:
                             done[name] = value
                         del notdone[name]
+
+                        if name.startswith('PY_') \
+                            and name[3:] in renamed_variables:
+
+                            name = name[3:]
+                            if name not in done:
+                                done[name] = value
             else:
                 # bogus variable reference; just drop it since we can't deal
                 del notdone[name]

Modified: python/branches/py3k/Lib/distutils/tests/test_sysconfig.py
==============================================================================
--- python/branches/py3k/Lib/distutils/tests/test_sysconfig.py	(original)
+++ python/branches/py3k/Lib/distutils/tests/test_sysconfig.py	Fri Jul 23 11:43:17 2010
@@ -93,6 +93,15 @@
                               'OTHER': 'foo'})
 
 
+    def test_sysconfig_module(self):
+        import sysconfig as global_sysconfig
+        self.assertEquals(global_sysconfig.get_config_var('CFLAGS'), sysconfig.get_config_var('CFLAGS'))
+        self.assertEquals(global_sysconfig.get_config_var('LDFLAGS'), sysconfig.get_config_var('LDFLAGS'))
+        self.assertEquals(global_sysconfig.get_config_var('LDSHARED'),sysconfig.get_config_var('LDSHARED'))
+        self.assertEquals(global_sysconfig.get_config_var('CC'), sysconfig.get_config_var('CC'))
+
+
+
 def test_suite():
     suite = unittest.TestSuite()
     suite.addTest(unittest.makeSuite(SysconfigTestCase))

From python-checkins at python.org  Fri Jul 23 11:50:05 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Fri, 23 Jul 2010 11:50:05 +0200 (CEST)
Subject: [Python-checkins] r83067 -
	python/branches/py3k/Lib/test/test_subprocess.py
Message-ID: <20100723095005.C0971EE9FB@mail.python.org>

Author: ronald.oussoren
Date: Fri Jul 23 11:50:05 2010
New Revision: 83067

Log:
Workaround for issue 4047: in some configurations of
the Crash Reporter on OSX test_subprocess will trigger
the reporter. 

This patch prints a warning when the Crash Reporter will
get triggered intentionally, which should avoid confusing
people.


Modified:
   python/branches/py3k/Lib/test/test_subprocess.py

Modified: python/branches/py3k/Lib/test/test_subprocess.py
==============================================================================
--- python/branches/py3k/Lib/test/test_subprocess.py	(original)
+++ python/branches/py3k/Lib/test/test_subprocess.py	Fri Jul 23 11:50:05 2010
@@ -559,6 +559,21 @@
         except (ImportError, ValueError, resource.error):
             pass
 
+        if sys.platform == 'darwin':
+            # Check if the 'Crash Reporter' on OSX was configured
+            # in 'Developer' mode and warn that it will get triggered
+            # when it is.
+            #
+            # This assumes that this context manager is used in tests
+            # that might trigger the next manager.
+            value = subprocess.Popen(['/usr/bin/defaults', 'read',
+                    'com.apple.CrashReporter', 'DialogType'],
+                    stdout=subprocess.PIPE).communicate()[0]
+            if value.strip() == b'developer':
+                print("this tests triggers the Crash Reporter, "
+                      "that is intentional", end='')
+                sys.stdout.flush()
+
     def __exit__(self, *args):
         """Return core file behavior to default."""
         if self.old_limit is None:

From python-checkins at python.org  Fri Jul 23 12:32:46 2010
From: python-checkins at python.org (brett.cannon)
Date: Fri, 23 Jul 2010 12:32:46 +0200 (CEST)
Subject: [Python-checkins] r83068 - peps/trunk/pep-0360.txt
Message-ID: <20100723103246.7D766F36E@mail.python.org>

Author: brett.cannon
Date: Fri Jul 23 12:32:46 2010
New Revision: 83068

Log:
Drop sync history as keeping it updated is a pain and typically forgotten.

Modified:
   peps/trunk/pep-0360.txt

Modified: peps/trunk/pep-0360.txt
==============================================================================
--- peps/trunk/pep-0360.txt	(original)
+++ peps/trunk/pep-0360.txt	Fri Jul 23 12:32:46 2010
@@ -65,8 +65,6 @@
     xml.etree
 :Contact person:
     Fredrik Lundh
-:Synchronisation history:
-    * 1.2.6 [ElementTree] / 1.0.5 [cElementTree] (2.5)
 
 Patches should not be directly applied to Python HEAD, but instead
 reported to the Python tracker [#python-tracker]_ (critical bug fixes
@@ -84,11 +82,6 @@
     bindings)
 :Contact person:
     None
-:Synchronisation history:
-    * 1.95.8 (2.4)
-    * 1.95.7 (2.3)
-    * 2.0    (2.5)
-
 
 Optik
 -----
@@ -99,11 +92,8 @@
     optparse
 :Contact person:
     Greg Ward
-:Synchronisation history:
-    * 1.5.1 (2.5)
-    * 1.5a1 (2.4)
-    * 1.4   (2.3)
 
+External development seems to have ceased.
 
 wsgiref
 -------
@@ -113,11 +103,10 @@
     wsgiref
 :Contact Person:
     Phillip J. Eby
-:Synchronisation history:
-    * 0.1 (2.5)
 
 Bugs and patches should pass through the Web-SIG mailing list [#web-sig]_
-before being applied to HEAD.
+before being applied to HEAD. External maintenance seems to have
+ceased.
 
 
 References

From python-checkins at python.org  Fri Jul 23 12:35:20 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Fri, 23 Jul 2010 12:35:20 +0200 (CEST)
Subject: [Python-checkins] r83069 - in python/branches/release31-maint:
	Lib/test/test_subprocess.py
Message-ID: <20100723103520.7FCB6F5EF@mail.python.org>

Author: ronald.oussoren
Date: Fri Jul 23 12:35:20 2010
New Revision: 83069

Log:
Merged revisions 83067 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83067 | ronald.oussoren | 2010-07-23 10:50:05 +0100 (Fri, 23 Jul 2010) | 8 lines
  
  Workaround for issue 4047: in some configurations of
  the Crash Reporter on OSX test_subprocess will trigger
  the reporter. 
  
  This patch prints a warning when the Crash Reporter will
  get triggered intentionally, which should avoid confusing
  people.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/test/test_subprocess.py

Modified: python/branches/release31-maint/Lib/test/test_subprocess.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_subprocess.py	(original)
+++ python/branches/release31-maint/Lib/test/test_subprocess.py	Fri Jul 23 12:35:20 2010
@@ -568,6 +568,21 @@
             """Try to prevent core files from being created.
             Returns previous ulimit if successful, else None.
             """
+            if sys.platform == 'darwin':
+                # Check if the 'Crash Reporter' on OSX was configured
+                # in 'Developer' mode and warn that it will get triggered
+                # when it is.
+                #
+                # This assumes that this context manager is used in tests
+                # that might trigger the next manager.
+                value = subprocess.Popen(['/usr/bin/defaults', 'read',
+                    'com.apple.CrashReporter', 'DialogType'],
+                    stdout=subprocess.PIPE).communicate()[0]
+                if value.strip() == b'developer':
+                    print("this tests triggers the Crash Reporter, "
+                          "that is intentional", end='')
+                    sys.stdout.flush()
+
             try:
                 import resource
                 old_limit = resource.getrlimit(resource.RLIMIT_CORE)
@@ -576,6 +591,8 @@
             except (ImportError, ValueError, resource.error):
                 return None
 
+
+
         def _unsuppress_core_files(self, old_limit):
             """Return core file behavior to default."""
             if old_limit is None:

From python-checkins at python.org  Fri Jul 23 13:11:26 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Fri, 23 Jul 2010 13:11:26 +0200 (CEST)
Subject: [Python-checkins] r83070 -
	python/branches/release27-maint/Mac/BuildScript/build-installer.py
Message-ID: <20100723111126.CFC7DEEA05@mail.python.org>

Author: ronald.oussoren
Date: Fri Jul 23 13:11:26 2010
New Revision: 83070

Log:
Fix for issue 9275: the OSX installer no longer installs links in /usr/local

This was an unintentional change to the 2.7 installer, and confuses users.


Modified:
   python/branches/release27-maint/Mac/BuildScript/build-installer.py

Modified: python/branches/release27-maint/Mac/BuildScript/build-installer.py
==============================================================================
--- python/branches/release27-maint/Mac/BuildScript/build-installer.py	(original)
+++ python/branches/release27-maint/Mac/BuildScript/build-installer.py	Fri Jul 23 13:11:26 2010
@@ -275,7 +275,7 @@
                 is not necessary to use Python.
                 """,
             required=False,
-            selected='unselected',
+            selected='selected',
         ),
         dict(
             name="PythonDocumentation",

From python-checkins at python.org  Fri Jul 23 13:13:03 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Fri, 23 Jul 2010 13:13:03 +0200 (CEST)
Subject: [Python-checkins] r83071 - python/branches/release27-maint/Misc/NEWS
Message-ID: <20100723111303.921C1EEA20@mail.python.org>

Author: ronald.oussoren
Date: Fri Jul 23 13:13:03 2010
New Revision: 83071

Log:
As the installer change is user-visible its better to
add a line to the NEWS file


Modified:
   python/branches/release27-maint/Misc/NEWS

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Fri Jul 23 13:13:03 2010
@@ -54,6 +54,8 @@
 
 - Issue #9280: Make sharedinstall depend on sharedmods.
 
+- Issue #9275: The OSX installer once again installs links to binaries in ``/usr/local/bin``
+
 Tests
 -----
 

From python-checkins at python.org  Fri Jul 23 13:31:32 2010
From: python-checkins at python.org (brett.cannon)
Date: Fri, 23 Jul 2010 13:31:32 +0200 (CEST)
Subject: [Python-checkins] r83072 - in python/branches/py3k:
	Doc/library/test.rst Misc/NEWS
Message-ID: <20100723113132.14D99EE9CF@mail.python.org>

Author: brett.cannon
Date: Fri Jul 23 13:31:31 2010
New Revision: 83072

Log:
Document the fact that the 'test' package is meant only for use by Python
itself and not by others.

Closes issue 9255.


Modified:
   python/branches/py3k/Doc/library/test.rst
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Doc/library/test.rst
==============================================================================
--- python/branches/py3k/Doc/library/test.rst	(original)
+++ python/branches/py3k/Doc/library/test.rst	Fri Jul 23 13:31:31 2010
@@ -5,6 +5,13 @@
    :synopsis: Regression tests package containing the testing suite for Python.
 .. sectionauthor:: Brett Cannon <brett at python.org>
 
+.. note::
+    The :mod:`test` package is meant for internal use by Python only. It is
+    documented for the benefit of the core developers of Python. Any use of
+    this package outside of Python's standard library is discouraged as code
+    mentioned here can change or be removed without notice between releases of
+    Python.
+
 
 The :mod:`test` package contains all regression tests for Python as well as the
 modules :mod:`test.support` and :mod:`test.regrtest`.

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Fri Jul 23 13:31:31 2010
@@ -1655,6 +1655,9 @@
 Documentation
 ------------
 
+- Issue 9255: Document that the 'test' package is meant for internal Python use
+  only.
+
 - A small wsgi server was added as Tools/scripts/serve.py, and is used to
   implement a local documentation server via 'make serve' in the doc directory.
 

From ncoghlan at gmail.com  Fri Jul 23 13:36:15 2010
From: ncoghlan at gmail.com (Nick Coghlan)
Date: Fri, 23 Jul 2010 21:36:15 +1000
Subject: [Python-checkins] r83072 - in python/branches/py3k:
	Doc/library/test.rst Misc/NEWS
In-Reply-To: <20100723113132.14D99EE9CF@mail.python.org>
References: <20100723113132.14D99EE9CF@mail.python.org>
Message-ID: <AANLkTinO23fgNOLGN6g7FHf_Vyu3MbWtPkaUY-n-OSMk@mail.gmail.com>

On Fri, Jul 23, 2010 at 9:31 PM, brett.cannon
<python-checkins at python.org> wrote:
> Author: brett.cannon
> Date: Fri Jul 23 13:31:31 2010
> New Revision: 83072
>
> Log:
> Document the fact that the 'test' package is meant only for use by Python
> itself and not by others.

Perhaps add a pointer to unittest for testing infrastructure that is
covered by all the usual backwards compatibility conventions?

Cheers,
Nick.

-- 
Nick Coghlan?? |?? ncoghlan at gmail.com?? |?? Brisbane, Australia

From python-checkins at python.org  Fri Jul 23 13:48:36 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Fri, 23 Jul 2010 13:48:36 +0200 (CEST)
Subject: [Python-checkins] r83073 -
	python/branches/py3k/Lib/test/test_osx_env.py
Message-ID: <20100723114836.3BB89EE9C5@mail.python.org>

Author: ronald.oussoren
Date: Fri Jul 23 13:48:36 2010
New Revision: 83073

Log:
Minor tweak of test_osx_env to 
avoid failing when the framework is not
yet installed.


Modified:
   python/branches/py3k/Lib/test/test_osx_env.py

Modified: python/branches/py3k/Lib/test/test_osx_env.py
==============================================================================
--- python/branches/py3k/Lib/test/test_osx_env.py	(original)
+++ python/branches/py3k/Lib/test/test_osx_env.py	Fri Jul 23 13:48:36 2010
@@ -8,7 +8,7 @@
 import unittest
 
 class OSXEnvironmentVariableTestCase(unittest.TestCase):
-    def _check_sys(self, ev, cond, sv, val = '/some/path/to/python'):
+    def _check_sys(self, ev, cond, sv, val = sys.executable + 'dummy'):
         with EnvironmentVarGuard() as evg:
             subpc = [str(sys.executable), '-c',
                 'import sys; sys.exit(2 if "%s" %s %s else 3)' % (val, cond, sv)]

From python-checkins at python.org  Fri Jul 23 13:51:28 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Fri, 23 Jul 2010 13:51:28 +0200 (CEST)
Subject: [Python-checkins] r83074 - in python/branches/release31-maint:
	Lib/test/test_osx_env.py
Message-ID: <20100723115128.ED9CDF2AC@mail.python.org>

Author: ronald.oussoren
Date: Fri Jul 23 13:51:28 2010
New Revision: 83074

Log:
Merged revisions 83073 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83073 | ronald.oussoren | 2010-07-23 12:48:36 +0100 (Fri, 23 Jul 2010) | 4 lines
  
  Minor tweak of test_osx_env to 
  avoid failing when the framework is not
  yet installed.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/test/test_osx_env.py

Modified: python/branches/release31-maint/Lib/test/test_osx_env.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_osx_env.py	(original)
+++ python/branches/release31-maint/Lib/test/test_osx_env.py	Fri Jul 23 13:51:28 2010
@@ -8,7 +8,7 @@
 import unittest
 
 class OSXEnvironmentVariableTestCase(unittest.TestCase):
-    def _check_sys(self, ev, cond, sv, val = '/some/path/to/python'):
+    def _check_sys(self, ev, cond, sv, val = sys.executable + 'dummy'):
         with EnvironmentVarGuard() as evg:
             subpc = [str(sys.executable), '-c',
                 'import sys; sys.exit(2 if "%s" %s %s else 3)' % (val, cond, sv)]

From python-checkins at python.org  Fri Jul 23 13:54:59 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Fri, 23 Jul 2010 13:54:59 +0200 (CEST)
Subject: [Python-checkins] r83075 - in python/branches/py3k: Lib/platform.py
	Lib/test/test_platform.py Misc/NEWS
Message-ID: <20100723115459.666E0EE981@mail.python.org>

Author: ronald.oussoren
Date: Fri Jul 23 13:54:59 2010
New Revision: 83075

Log:
Fix for issue 7895. Avoid crashing the interpreter
when calling platform.mac_ver after calling os.fork by
reading from a system configuration file instead of
using OSX APIs.


Modified:
   python/branches/py3k/Lib/platform.py
   python/branches/py3k/Lib/test/test_platform.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/platform.py
==============================================================================
--- python/branches/py3k/Lib/platform.py	(original)
+++ python/branches/py3k/Lib/platform.py	Fri Jul 23 13:54:59 2010
@@ -724,27 +724,19 @@
 
     return hex(bcd)[2:]
 
-def mac_ver(release='',versioninfo=('','',''),machine=''):
-
-    """ Get MacOS version information and return it as tuple (release,
-        versioninfo, machine) with versioninfo being a tuple (version,
-        dev_stage, non_release_version).
-
-        Entries which cannot be determined are set to the paramter values
-        which default to ''. All tuple entries are strings.
-
+def _mac_ver_gestalt():
+    """
         Thanks to Mark R. Levinson for mailing documentation links and
         code examples for this function. Documentation for the
         gestalt() API is available online at:
 
            http://www.rgaros.nl/gestalt/
-
     """
     # Check whether the version info module is available
     try:
         import _gestalt
     except ImportError:
-        return release,versioninfo,machine
+        return None
     # Get the infos
     sysv, sysa = _mac_ver_lookup(('sysv','sysa'))
     # Decode the infos
@@ -768,6 +760,53 @@
         machine = {0x1: '68k',
                    0x2: 'PowerPC',
                    0xa: 'i386'}.get(sysa,'')
+
+    return release,versioninfo,machine
+
+def _mac_ver_xml():
+    fn = '/System/Library/CoreServices/SystemVersion.plist'
+    if not os.path.exists(fn):
+        return None
+
+    try:
+        import plistlib
+    except ImportError:
+        return None
+
+    pl = plistlib.readPlist(fn)
+    release = pl['ProductVersion']
+    versioninfo=('', '', '')
+    machine = os.uname()[4]
+    if machine == 'ppc':
+        # for compatibility with the gestalt based code
+        machine = 'PowerPC'
+
+    return release,versioninfo,machine
+
+
+def mac_ver(release='',versioninfo=('','',''),machine=''):
+
+    """ Get MacOS version information and return it as tuple (release,
+        versioninfo, machine) with versioninfo being a tuple (version,
+        dev_stage, non_release_version).
+
+        Entries which cannot be determined are set to the paramter values
+        which default to ''. All tuple entries are strings.
+    """
+
+    # First try reading the information from an XML file which should
+    # always be present
+    info = _mac_ver_xml()
+    if info is not None:
+        return info
+
+    # If that doesn't work for some reason fall back to reading the
+    # information using gestalt calls.
+    info = _mac_ver_gestalt()
+    if info is not None:
+        return info
+
+    # If that also doesn't work return the default values
     return release,versioninfo,machine
 
 def _java_getprop(name,default):

Modified: python/branches/py3k/Lib/test/test_platform.py
==============================================================================
--- python/branches/py3k/Lib/test/test_platform.py	(original)
+++ python/branches/py3k/Lib/test/test_platform.py	Fri Jul 23 13:54:59 2010
@@ -194,6 +194,25 @@
             else:
                 self.assertEquals(res[2], 'PowerPC')
 
+
+    @unittest.skipUnless(sys.platform == 'darwin', "OSX only test")
+    def test_mac_ver_with_fork(self):
+        # Issue7895: platform.mac_ver() crashes when using fork without exec
+        #
+        # This test checks that the fix for that issue works.
+        #
+        pid = os.fork()
+        if pid == 0:
+            # child
+            info = platform.mac_ver()
+            os._exit(0)
+
+        else:
+            # parent
+            cpid, sts = os.waitpid(pid, 0)
+            self.assertEquals(cpid, pid)
+            self.assertEquals(sts, 0)
+
     def test_dist(self):
         res = platform.dist()
 

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Fri Jul 23 13:54:59 2010
@@ -473,6 +473,8 @@
 Library
 -------
 
+- Issue #7895: platform.mac_ver() no longer crashes after calling os.fork()
+
 - Issue #9323: Fixed a bug in trace.py that resulted in loosing the
   name of the script being traced.  Patch by Eli Bendersky.
 

From python-checkins at python.org  Fri Jul 23 14:03:37 2010
From: python-checkins at python.org (brett.cannon)
Date: Fri, 23 Jul 2010 14:03:37 +0200 (CEST)
Subject: [Python-checkins] r83076 - in python/branches/release31-maint:
	Doc/library/test.rst Misc/NEWS
Message-ID: <20100723120337.E7234F5EF@mail.python.org>

Author: brett.cannon
Date: Fri Jul 23 14:03:37 2010
New Revision: 83076

Log:
Blocked revisions 83072 via svnmerge

........
  r83072 | brett.cannon | 2010-07-23 12:31:31 +0100 (Fri, 23 Jul 2010) | 5 lines
  
  Document the fact that the 'test' package is meant only for use by Python
  itself and not by others.
  
  Closes issue 9255.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Doc/library/test.rst
   python/branches/release31-maint/Misc/NEWS

Modified: python/branches/release31-maint/Doc/library/test.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/test.rst	(original)
+++ python/branches/release31-maint/Doc/library/test.rst	Fri Jul 23 14:03:37 2010
@@ -5,6 +5,13 @@
    :synopsis: Regression tests package containing the testing suite for Python.
 .. sectionauthor:: Brett Cannon <brett at python.org>
 
+.. note::
+    The :mod:`test` package is meant for internal use by Python only. It is
+    documented for the benefit of the core developers of Python. Any use of
+    this package outside of Python's standard library is discouraged as code
+    mentioned here can change or be removed without notice between releases of
+    Python.
+
 
 The :mod:`test` package contains all regression tests for Python as well as the
 modules :mod:`test.support` and :mod:`test.regrtest`.

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Fri Jul 23 14:03:37 2010
@@ -369,6 +369,9 @@
 Documentation
 -------------
 
+- Issue 9255: Document that the 'test' package is meant for interal Python use
+  only.
+
 - Issue #7829: Document in dis that bytecode is an implementation detail.
 
 

From python-checkins at python.org  Fri Jul 23 14:07:27 2010
From: python-checkins at python.org (brett.cannon)
Date: Fri, 23 Jul 2010 14:07:27 +0200 (CEST)
Subject: [Python-checkins] r83077 - in python/branches/release27-maint:
	Doc/library/test.rst Misc/NEWS
Message-ID: <20100723120727.6562EEE981@mail.python.org>

Author: brett.cannon
Date: Fri Jul 23 14:07:27 2010
New Revision: 83077

Log:
Merged revisions 83072 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83072 | brett.cannon | 2010-07-23 12:31:31 +0100 (Fri, 23 Jul 2010) | 5 lines
  
  Document the fact that the 'test' package is meant only for use by Python
  itself and not by others.
  
  Closes issue 9255.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Doc/library/test.rst
   python/branches/release27-maint/Misc/NEWS

Modified: python/branches/release27-maint/Doc/library/test.rst
==============================================================================
--- python/branches/release27-maint/Doc/library/test.rst	(original)
+++ python/branches/release27-maint/Doc/library/test.rst	Fri Jul 23 14:07:27 2010
@@ -6,6 +6,13 @@
    :synopsis: Regression tests package containing the testing suite for Python.
 .. sectionauthor:: Brett Cannon <brett at python.org>
 
+.. note::
+    The :mod:`test` package is meant for internal use by Python only. It is
+    documented for the benefit of the core developers of Python. Any use of
+    this package outside of Python's standard library is discouraged as code
+    mentioned here can change or be removed without notice between releases of
+    Python.
+
 
 The :mod:`test` package contains all regression tests for Python as well as the
 modules :mod:`test.test_support` and :mod:`test.regrtest`.

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Fri Jul 23 14:07:27 2010
@@ -64,6 +64,8 @@
 Documentation
 -------------
 
+- Issue 9255: Document that the 'test' package is for internal Python use only.
+
 - Issue 7829: Document in dis that bytecode is an implementation detail.
 
 

From python-checkins at python.org  Fri Jul 23 14:16:41 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Fri, 23 Jul 2010 14:16:41 +0200 (CEST)
Subject: [Python-checkins] r83078 - in python/branches/py3k:
	Doc/library/os.rst Lib/test/test_posix.py Misc/NEWS
	Modules/posixmodule.c
Message-ID: <20100723121641.8C479D7D7@mail.python.org>

Author: martin.v.loewis
Date: Fri Jul 23 14:16:41 2010
New Revision: 83078

Log:
Issue #6095: Make directory argument to os.listdir optional.
Patch by Virgil Dupras.


Modified:
   python/branches/py3k/Doc/library/os.rst
   python/branches/py3k/Lib/test/test_posix.py
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/Modules/posixmodule.c

Modified: python/branches/py3k/Doc/library/os.rst
==============================================================================
--- python/branches/py3k/Doc/library/os.rst	(original)
+++ python/branches/py3k/Doc/library/os.rst	Fri Jul 23 14:16:41 2010
@@ -1049,10 +1049,10 @@
    Availability: Unix.
 
 
-.. function:: listdir(path)
+.. function:: listdir([path])
 
    Return a list containing the names of the entries in the directory given by
-   *path*.  The list is in arbitrary order.  It does not include the special
+   *path* (default: '.').  The list is in arbitrary order.  It does not include the special
    entries ``'.'`` and ``'..'`` even if they are present in the directory.
 
    This function can be called with a bytes or string argument, and returns
@@ -1060,6 +1060,8 @@
 
    Availability: Unix, Windows.
 
+   .. versionchanged:: 3.2
+      The *path* parameter became optional.
 
 .. function:: lstat(path)
 

Modified: python/branches/py3k/Lib/test/test_posix.py
==============================================================================
--- python/branches/py3k/Lib/test/test_posix.py	(original)
+++ python/branches/py3k/Lib/test/test_posix.py	Fri Jul 23 14:16:41 2010
@@ -252,9 +252,14 @@
             posix.chdir(os.curdir)
             self.assertRaises(OSError, posix.chdir, support.TESTFN)
 
-    def test_lsdir(self):
-        if hasattr(posix, 'lsdir'):
-            self.assertIn(support.TESTFN, posix.lsdir(os.curdir))
+    def test_listdir(self):
+        if hasattr(posix, 'listdir'):
+            self.assertTrue(support.TESTFN in posix.listdir(os.curdir))
+
+    def test_listdir_default(self):
+        # When listdir is called without argument, it's the same as listdir(os.curdir)
+        if hasattr(posix, 'listdir'):
+            self.assertTrue(support.TESTFN in posix.listdir())
 
     def test_access(self):
         if hasattr(posix, 'access'):

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Fri Jul 23 14:16:41 2010
@@ -1440,6 +1440,8 @@
 Extension Modules
 -----------------
 
+- Issue #6095: Make directory argument to os.listdir optional.
+
 - Issue #9277: Fix bug in struct.pack for bools in standard mode
   (e.g., struct.pack('>?')):  if conversion to bool raised an exception
   then that exception wasn't properly propagated on machines where

Modified: python/branches/py3k/Modules/posixmodule.c
==============================================================================
--- python/branches/py3k/Modules/posixmodule.c	(original)
+++ python/branches/py3k/Modules/posixmodule.c	Fri Jul 23 14:16:41 2010
@@ -2291,10 +2291,10 @@
 
 
 PyDoc_STRVAR(posix_listdir__doc__,
-"listdir(path) -> list_of_strings\n\n\
+"listdir([path]) -> list_of_strings\n\n\
 Return a list containing the names of the entries in the directory.\n\
 \n\
-    path: path of directory to list\n\
+    path: path of directory to list (default: '.')\n\
 \n\
 The list is in arbitrary order.  It does not include the special\n\
 entries '.' and '..' even if they are present in the directory.");
@@ -2315,18 +2315,25 @@
     char *bufptr = namebuf;
     Py_ssize_t len = sizeof(namebuf)-5; /* only claim to have space for MAX_PATH */
 
-    PyObject *po;
-    if (PyArg_ParseTuple(args, "U:listdir", &po)) {
+    PyObject *po = NULL;
+    if (PyArg_ParseTuple(args, "|U:listdir", &po)) {
         WIN32_FIND_DATAW wFileData;
-        Py_UNICODE *wnamebuf;
+        Py_UNICODE *wnamebuf, *po_wchars;
+        
+        if (po == NULL) { // Default arg: "."
+            po_wchars = L".";
+            len = 1;
+        } else {
+            po_wchars = PyUnicode_AS_UNICODE(po);
+            len = PyUnicode_GET_SIZE(po);
+        }
         /* Overallocate for \\*.*\0 */
-        len = PyUnicode_GET_SIZE(po);
         wnamebuf = malloc((len + 5) * sizeof(wchar_t));
         if (!wnamebuf) {
             PyErr_NoMemory();
             return NULL;
         }
-        wcscpy(wnamebuf, PyUnicode_AS_UNICODE(po));
+        wcscpy(wnamebuf, po_wchars);
         if (len > 0) {
             Py_UNICODE wch = wnamebuf[len-1];
             if (wch != L'/' && wch != L'\\' && wch != L':')
@@ -2548,12 +2555,17 @@
     int arg_is_unicode = 1;
 
     errno = 0;
-    if (!PyArg_ParseTuple(args, "U:listdir", &v)) {
+    /* v is never read, so it does not need to be initialized yet. */
+    if (!PyArg_ParseTuple(args, "|U:listdir", &v)) {
         arg_is_unicode = 0;
         PyErr_Clear();
     }
-    if (!PyArg_ParseTuple(args, "O&:listdir", PyUnicode_FSConverter, &oname))
+    oname = NULL;
+    if (!PyArg_ParseTuple(args, "|O&:listdir", PyUnicode_FSConverter, &oname))
         return NULL;
+    if (oname == NULL) { // Default arg: "."
+      oname = PyBytes_FromString(".");
+    }
     name = PyBytes_AsString(oname);
     if ((dirp = opendir(name)) == NULL) {
         return posix_error_with_allocated_filename(oname);

From benjamin at python.org  Fri Jul 23 14:25:36 2010
From: benjamin at python.org (Benjamin Peterson)
Date: Fri, 23 Jul 2010 07:25:36 -0500
Subject: [Python-checkins] r83078 - in python/branches/py3k:
	Doc/library/os.rst Lib/test/test_posix.py Misc/NEWS
	Modules/posixmodule.c
In-Reply-To: <20100723121641.8C479D7D7@mail.python.org>
References: <20100723121641.8C479D7D7@mail.python.org>
Message-ID: <AANLkTi=5YMUnEFUemQ8+kR0dR+sai5cm67SubS83dcpt@mail.gmail.com>

2010/7/23 martin.v.loewis <python-checkins at python.org>:
> Author: martin.v.loewis
> Date: Fri Jul 23 14:16:41 2010
> New Revision: 83078
>
> Log:
> Issue #6095: Make directory argument to os.listdir optional.
> Patch by Virgil Dupras.
>
>
> Modified:
> ? python/branches/py3k/Doc/library/os.rst
> ? python/branches/py3k/Lib/test/test_posix.py
> ? python/branches/py3k/Misc/NEWS
> ? python/branches/py3k/Modules/posixmodule.c
>
> Modified: python/branches/py3k/Doc/library/os.rst
> ==============================================================================
> --- python/branches/py3k/Doc/library/os.rst ? ? (original)
> +++ python/branches/py3k/Doc/library/os.rst ? ? Fri Jul 23 14:16:41 2010
> @@ -1049,10 +1049,10 @@
> ? ?Availability: Unix.
>
>
> -.. function:: listdir(path)
> +.. function:: listdir([path])

You can use the new default argument convention in the docs:
listdir(path=curdir)


-- 
Regards,
Benjamin

From python-checkins at python.org  Fri Jul 23 14:26:30 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Fri, 23 Jul 2010 14:26:30 +0200 (CEST)
Subject: [Python-checkins] r83079 -
	python/branches/release27-maint/Lib/test/test_subprocess.py
Message-ID: <20100723122630.CDE1FEE981@mail.python.org>

Author: ronald.oussoren
Date: Fri Jul 23 14:26:30 2010
New Revision: 83079

Log:
Merged revisions 83067 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83067 | ronald.oussoren | 2010-07-23 10:50:05 +0100 (Fri, 23 Jul 2010) | 8 lines
  
  Workaround for issue 4047: in some configurations of
  the Crash Reporter on OSX test_subprocess will trigger
  the reporter. 
  
  This patch prints a warning when the Crash Reporter will
  get triggered intentionally, which should avoid confusing
  people.
........


Modified:
   python/branches/release27-maint/Lib/test/test_subprocess.py

Modified: python/branches/release27-maint/Lib/test/test_subprocess.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_subprocess.py	(original)
+++ python/branches/release27-maint/Lib/test/test_subprocess.py	Fri Jul 23 14:26:30 2010
@@ -548,6 +548,20 @@
         except (ImportError, ValueError, resource.error):
             pass
 
+        if sys.platform == 'darwin':
+            # Check if the 'Crash Reporter' on OSX was configured
+            # in 'Developer' mode and warn that it will get triggered
+            # when it is.
+            #
+            # This assumes that this context manager is used in tests
+            # that might trigger the next manager.
+            value = subprocess.Popen(['/usr/bin/defaults', 'read',
+                    'com.apple.CrashReporter', 'DialogType'],
+                    stdout=subprocess.PIPE).communicate()[0]
+            if value.strip() == b'developer':
+                print "this tests triggers the Crash Reporter, that is intentional"
+                sys.stdout.flush()
+
     def __exit__(self, *args):
         """Return core file behavior to default."""
         if self.old_limit is None:

From python-checkins at python.org  Fri Jul 23 14:26:35 2010
From: python-checkins at python.org (brett.cannon)
Date: Fri, 23 Jul 2010 14:26:35 +0200 (CEST)
Subject: [Python-checkins] r83080 -
	python/branches/py3k/Doc/library/threading.rst
Message-ID: <20100723122635.D8BB3EE9BD@mail.python.org>

Author: brett.cannon
Date: Fri Jul 23 14:26:35 2010
New Revision: 83080

Log:
Clarify the wording for threading.is_alive() to not suggest something is
"roughly" done.

Closes issue 9339. Thanks Brian Brazil for the patch.


Modified:
   python/branches/py3k/Doc/library/threading.rst

Modified: python/branches/py3k/Doc/library/threading.rst
==============================================================================
--- python/branches/py3k/Doc/library/threading.rst	(original)
+++ python/branches/py3k/Doc/library/threading.rst	Fri Jul 23 14:26:35 2010
@@ -310,8 +310,8 @@
 
       Return whether the thread is alive.
 
-      Roughly, a thread is alive from the moment the :meth:`start` method
-      returns until its :meth:`run` method terminates. The module function
+      This method returns ``True`` just before the :meth:`run` method starts
+      until just after the :meth:`run` method terminates.  The module function
       :func:`.enumerate` returns a list of all alive threads.
 
    .. attribute:: daemon

From python-checkins at python.org  Fri Jul 23 14:28:28 2010
From: python-checkins at python.org (brett.cannon)
Date: Fri, 23 Jul 2010 14:28:28 +0200 (CEST)
Subject: [Python-checkins] r83081 - in python/branches/release31-maint:
	Doc/library/threading.rst
Message-ID: <20100723122828.BC4D0F42F@mail.python.org>

Author: brett.cannon
Date: Fri Jul 23 14:28:28 2010
New Revision: 83081

Log:
Merged revisions 83080 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83080 | brett.cannon | 2010-07-23 13:26:35 +0100 (Fri, 23 Jul 2010) | 5 lines
  
  Clarify the wording for threading.is_alive() to not suggest something is
  "roughly" done.
  
  Closes issue 9339. Thanks Brian Brazil for the patch.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Doc/library/threading.rst

Modified: python/branches/release31-maint/Doc/library/threading.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/threading.rst	(original)
+++ python/branches/release31-maint/Doc/library/threading.rst	Fri Jul 23 14:28:28 2010
@@ -299,8 +299,8 @@
 
       Return whether the thread is alive.
 
-      Roughly, a thread is alive from the moment the :meth:`start` method
-      returns until its :meth:`run` method terminates. The module function
+      This method returns ``True`` just before the :meth:`run` method starts
+      until just after the :meth:`run` method terminates.  The module function
       :func:`.enumerate` returns a list of all alive threads.
 
    .. attribute:: daemon

From python-checkins at python.org  Fri Jul 23 14:30:10 2010
From: python-checkins at python.org (brett.cannon)
Date: Fri, 23 Jul 2010 14:30:10 +0200 (CEST)
Subject: [Python-checkins] r83082 - in python/branches/release27-maint:
	Doc/library/threading.rst
Message-ID: <20100723123010.602DAF469@mail.python.org>

Author: brett.cannon
Date: Fri Jul 23 14:30:10 2010
New Revision: 83082

Log:
Merged revisions 83080 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83080 | brett.cannon | 2010-07-23 13:26:35 +0100 (Fri, 23 Jul 2010) | 5 lines
  
  Clarify the wording for threading.is_alive() to not suggest something is
  "roughly" done.
  
  Closes issue 9339. Thanks Brian Brazil for the patch.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Doc/library/threading.rst

Modified: python/branches/release27-maint/Doc/library/threading.rst
==============================================================================
--- python/branches/release27-maint/Doc/library/threading.rst	(original)
+++ python/branches/release27-maint/Doc/library/threading.rst	Fri Jul 23 14:30:10 2010
@@ -319,8 +319,8 @@
 
       Return whether the thread is alive.
 
-      Roughly, a thread is alive from the moment the :meth:`start` method
-      returns until its :meth:`run` method terminates. The module function
+      This method returns ``True`` just before the :meth:`run` method starts
+      until just after the :meth:`run` method terminates.  The module function
       :func:`.enumerate` returns a list of all alive threads.
 
    .. method:: isDaemon()

From python-checkins at python.org  Fri Jul 23 14:32:08 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Fri, 23 Jul 2010 14:32:08 +0200 (CEST)
Subject: [Python-checkins] r83083 - in python/branches/release26-maint:
	Lib/test/test_subprocess.py
Message-ID: <20100723123208.2A337F469@mail.python.org>

Author: ronald.oussoren
Date: Fri Jul 23 14:32:07 2010
New Revision: 83083

Log:
Merged revisions 83079 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

................
  r83079 | ronald.oussoren | 2010-07-23 13:26:30 +0100 (Fri, 23 Jul 2010) | 15 lines
  
  Merged revisions 83067 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k
  
  ........
    r83067 | ronald.oussoren | 2010-07-23 10:50:05 +0100 (Fri, 23 Jul 2010) | 8 lines
    
    Workaround for issue 4047: in some configurations of
    the Crash Reporter on OSX test_subprocess will trigger
    the reporter. 
    
    This patch prints a warning when the Crash Reporter will
    get triggered intentionally, which should avoid confusing
    people.
  ........
................


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Lib/test/test_subprocess.py

Modified: python/branches/release26-maint/Lib/test/test_subprocess.py
==============================================================================
--- python/branches/release26-maint/Lib/test/test_subprocess.py	(original)
+++ python/branches/release26-maint/Lib/test/test_subprocess.py	Fri Jul 23 14:32:07 2010
@@ -521,6 +521,20 @@
             """Try to prevent core files from being created.
             Returns previous ulimit if successful, else None.
             """
+            if sys.platform == 'darwin':
+                # Check if the 'Crash Reporter' on OSX was configured
+                # in 'Developer' mode and warn that it will get triggered
+                # when it is.
+                #
+                # This assumes that this context manager is used in tests
+                # that might trigger the next manager.
+                value = subprocess.Popen(['/usr/bin/defaults', 'read',
+                    'com.apple.CrashReporter', 'DialogType'],
+                    stdout=subprocess.PIPE).communicate()[0]
+                if value.strip() == b'developer':
+                    print "this tests triggers the Crash Reporter, that is intentional"
+                    sys.stdout.flush()
+
             try:
                 import resource
                 old_limit = resource.getrlimit(resource.RLIMIT_CORE)

From python-checkins at python.org  Fri Jul 23 14:36:59 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Fri, 23 Jul 2010 14:36:59 +0200 (CEST)
Subject: [Python-checkins] r83084 - in python/branches/release31-maint:
	Lib/platform.py Lib/test/test_platform.py Misc/NEWS
Message-ID: <20100723123659.4D698F4E7@mail.python.org>

Author: ronald.oussoren
Date: Fri Jul 23 14:36:59 2010
New Revision: 83084

Log:
Merged revisions 83075 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83075 | ronald.oussoren | 2010-07-23 12:54:59 +0100 (Fri, 23 Jul 2010) | 5 lines
  
  Fix for issue 7895. Avoid crashing the interpreter
  when calling platform.mac_ver after calling os.fork by
  reading from a system configuration file instead of
  using OSX APIs.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/platform.py
   python/branches/release31-maint/Lib/test/test_platform.py
   python/branches/release31-maint/Misc/NEWS

Modified: python/branches/release31-maint/Lib/platform.py
==============================================================================
--- python/branches/release31-maint/Lib/platform.py	(original)
+++ python/branches/release31-maint/Lib/platform.py	Fri Jul 23 14:36:59 2010
@@ -696,27 +696,19 @@
 
     return hex(bcd)[2:]
 
-def mac_ver(release='',versioninfo=('','',''),machine=''):
-
-    """ Get MacOS version information and return it as tuple (release,
-        versioninfo, machine) with versioninfo being a tuple (version,
-        dev_stage, non_release_version).
-
-        Entries which cannot be determined are set to the paramter values
-        which default to ''. All tuple entries are strings.
-
+def _mac_ver_gestalt():
+    """
         Thanks to Mark R. Levinson for mailing documentation links and
         code examples for this function. Documentation for the
         gestalt() API is available online at:
 
            http://www.rgaros.nl/gestalt/
-
     """
     # Check whether the version info module is available
     try:
         import _gestalt
     except ImportError:
-        return release,versioninfo,machine
+        return None
     # Get the infos
     sysv, sysa = _mac_ver_lookup(('sysv','sysa'))
     # Decode the infos
@@ -740,6 +732,53 @@
         machine = {0x1: '68k',
                    0x2: 'PowerPC',
                    0xa: 'i386'}.get(sysa,'')
+
+    return release,versioninfo,machine
+
+def _mac_ver_xml():
+    fn = '/System/Library/CoreServices/SystemVersion.plist'
+    if not os.path.exists(fn):
+        return None
+
+    try:
+        import plistlib
+    except ImportError:
+        return None
+
+    pl = plistlib.readPlist(fn)
+    release = pl['ProductVersion']
+    versioninfo=('', '', '')
+    machine = os.uname()[4]
+    if machine == 'ppc':
+        # for compatibility with the gestalt based code
+        machine = 'PowerPC'
+
+    return release,versioninfo,machine
+
+
+def mac_ver(release='',versioninfo=('','',''),machine=''):
+
+    """ Get MacOS version information and return it as tuple (release,
+        versioninfo, machine) with versioninfo being a tuple (version,
+        dev_stage, non_release_version).
+
+        Entries which cannot be determined are set to the paramter values
+        which default to ''. All tuple entries are strings.
+    """
+
+    # First try reading the information from an XML file which should
+    # always be present
+    info = _mac_ver_xml()
+    if info is not None:
+        return info
+
+    # If that doesn't work for some reason fall back to reading the
+    # information using gestalt calls.
+    info = _mac_ver_gestalt()
+    if info is not None:
+        return info
+
+    # If that also doesn't work return the default values
     return release,versioninfo,machine
 
 def _java_getprop(name,default):

Modified: python/branches/release31-maint/Lib/test/test_platform.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_platform.py	(original)
+++ python/branches/release31-maint/Lib/test/test_platform.py	Fri Jul 23 14:36:59 2010
@@ -167,6 +167,25 @@
             else:
                 self.assertEquals(res[2], 'PowerPC')
 
+
+    @unittest.skipUnless(sys.platform == 'darwin', "OSX only test")
+    def test_mac_ver_with_fork(self):
+        # Issue7895: platform.mac_ver() crashes when using fork without exec
+        #
+        # This test checks that the fix for that issue works.
+        #
+        pid = os.fork()
+        if pid == 0:
+            # child
+            info = platform.mac_ver()
+            os._exit(0)
+
+        else:
+            # parent
+            cpid, sts = os.waitpid(pid, 0)
+            self.assertEquals(cpid, pid)
+            self.assertEquals(sts, 0)
+
     def test_dist(self):
         res = platform.dist()
 

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Fri Jul 23 14:36:59 2010
@@ -78,6 +78,8 @@
 Library
 -------
 
+- Issue #7895: platform.mac_ver() no longer crashes after calling os.fork()
+
 - Issue #9323: Fixed a bug in trace.py that resulted in loosing the
   name of the script being traced.  Patch by Eli Bendersky.
 

From python-checkins at python.org  Fri Jul 23 14:41:00 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Fri, 23 Jul 2010 14:41:00 +0200 (CEST)
Subject: [Python-checkins] r83085 - in python/branches/release27-maint:
	Lib/platform.py Lib/test/test_platform.py Misc/NEWS
Message-ID: <20100723124100.5455DF42F@mail.python.org>

Author: ronald.oussoren
Date: Fri Jul 23 14:41:00 2010
New Revision: 83085

Log:
Merged revisions 83075 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83075 | ronald.oussoren | 2010-07-23 12:54:59 +0100 (Fri, 23 Jul 2010) | 5 lines
  
  Fix for issue 7895. Avoid crashing the interpreter
  when calling platform.mac_ver after calling os.fork by
  reading from a system configuration file instead of
  using OSX APIs.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/platform.py
   python/branches/release27-maint/Lib/test/test_platform.py
   python/branches/release27-maint/Misc/NEWS

Modified: python/branches/release27-maint/Lib/platform.py
==============================================================================
--- python/branches/release27-maint/Lib/platform.py	(original)
+++ python/branches/release27-maint/Lib/platform.py	Fri Jul 23 14:41:00 2010
@@ -727,28 +727,20 @@
 
     return hex(bcd)[2:]
 
-def mac_ver(release='',versioninfo=('','',''),machine=''):
-
-    """ Get MacOS version information and return it as tuple (release,
-        versioninfo, machine) with versioninfo being a tuple (version,
-        dev_stage, non_release_version).
-
-        Entries which cannot be determined are set to the paramter values
-        which default to ''. All tuple entries are strings.
-
+def _mac_ver_gestalt():
+    """
         Thanks to Mark R. Levinson for mailing documentation links and
         code examples for this function. Documentation for the
         gestalt() API is available online at:
 
            http://www.rgaros.nl/gestalt/
-
     """
     # Check whether the version info module is available
     try:
         import gestalt
         import MacOS
     except ImportError:
-        return release,versioninfo,machine
+        return None
     # Get the infos
     sysv,sysa = _mac_ver_lookup(('sysv','sysa'))
     # Decode the infos
@@ -772,6 +764,53 @@
         machine = {0x1: '68k',
                    0x2: 'PowerPC',
                    0xa: 'i386'}.get(sysa,'')
+
+    return release,versioninfo,machine
+
+def _mac_ver_xml():
+    fn = '/System/Library/CoreServices/SystemVersion.plist'
+    if not os.path.exists(fn):
+        return None
+
+    try:
+        import plistlib
+    except ImportError:
+        return None
+
+    pl = plistlib.readPlist(fn)
+    release = pl['ProductVersion']
+    versioninfo=('', '', '')
+    machine = os.uname()[4]
+    if machine == 'ppc':
+        # for compatibility with the gestalt based code
+        machine = 'PowerPC'
+
+    return release,versioninfo,machine
+
+
+def mac_ver(release='',versioninfo=('','',''),machine=''):
+
+    """ Get MacOS version information and return it as tuple (release,
+        versioninfo, machine) with versioninfo being a tuple (version,
+        dev_stage, non_release_version).
+
+        Entries which cannot be determined are set to the paramter values
+        which default to ''. All tuple entries are strings.
+    """
+
+    # First try reading the information from an XML file which should
+    # always be present
+    info = _mac_ver_xml()
+    if info is not None:
+        return info
+
+    # If that doesn't work for some reason fall back to reading the
+    # information using gestalt calls.
+    info = _mac_ver_gestalt()
+    if info is not None:
+        return info
+
+    # If that also doesn't work return the default values
     return release,versioninfo,machine
 
 def _java_getprop(name,default):

Modified: python/branches/release27-maint/Lib/test/test_platform.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_platform.py	(original)
+++ python/branches/release27-maint/Lib/test/test_platform.py	Fri Jul 23 14:41:00 2010
@@ -195,6 +195,25 @@
             else:
                 self.assertEquals(res[2], 'PowerPC')
 
+
+    @unittest.skipUnless(sys.platform == 'darwin', "OSX only test")
+    def test_mac_ver_with_fork(self):
+        # Issue7895: platform.mac_ver() crashes when using fork without exec
+        #
+        # This test checks that the fix for that issue works.
+        #
+        pid = os.fork()
+        if pid == 0:
+            # child
+            info = platform.mac_ver()
+            os._exit(0)
+
+        else:
+            # parent
+            cpid, sts = os.waitpid(pid, 0)
+            self.assertEquals(cpid, pid)
+            self.assertEquals(sts, 0)
+
     def test_dist(self):
         res = platform.dist()
 

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Fri Jul 23 14:41:00 2010
@@ -18,6 +18,8 @@
 Library
 -------
 
+- Issue #7895: platform.mac_ver() no longer crashes after calling os.fork()
+
 - Issue #5395: array.fromfile() would raise a spurious EOFError when an
   I/O error occurred.  Now an IOError is raised instead.  Patch by chuck
   (Jan Hosang).

From python-checkins at python.org  Fri Jul 23 14:45:54 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Fri, 23 Jul 2010 14:45:54 +0200 (CEST)
Subject: [Python-checkins] r83086 - in python/branches/release26-maint:
	Lib/platform.py Lib/test/test_platform.py Misc/NEWS
Message-ID: <20100723124554.14EB2EE983@mail.python.org>

Author: ronald.oussoren
Date: Fri Jul 23 14:45:53 2010
New Revision: 83086

Log:
Merged revisions 83085 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

................
  r83085 | ronald.oussoren | 2010-07-23 13:41:00 +0100 (Fri, 23 Jul 2010) | 12 lines
  
  Merged revisions 83075 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k
  
  ........
    r83075 | ronald.oussoren | 2010-07-23 12:54:59 +0100 (Fri, 23 Jul 2010) | 5 lines
    
    Fix for issue 7895. Avoid crashing the interpreter
    when calling platform.mac_ver after calling os.fork by
    reading from a system configuration file instead of
    using OSX APIs.
  ........
................


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Lib/platform.py
   python/branches/release26-maint/Lib/test/test_platform.py
   python/branches/release26-maint/Misc/NEWS

Modified: python/branches/release26-maint/Lib/platform.py
==============================================================================
--- python/branches/release26-maint/Lib/platform.py	(original)
+++ python/branches/release26-maint/Lib/platform.py	Fri Jul 23 14:45:53 2010
@@ -717,28 +717,20 @@
 
     return hex(bcd)[2:]
 
-def mac_ver(release='',versioninfo=('','',''),machine=''):
-
-    """ Get MacOS version information and return it as tuple (release,
-        versioninfo, machine) with versioninfo being a tuple (version,
-        dev_stage, non_release_version).
-
-        Entries which cannot be determined are set to the paramter values
-        which default to ''. All tuple entries are strings.
-
+def _mac_ver_gestalt():
+    """
         Thanks to Mark R. Levinson for mailing documentation links and
         code examples for this function. Documentation for the
         gestalt() API is available online at:
 
            http://www.rgaros.nl/gestalt/
-
     """
     # Check whether the version info module is available
     try:
         import gestalt
         import MacOS
     except ImportError:
-        return release,versioninfo,machine
+        return None
     # Get the infos
     sysv,sysa = _mac_ver_lookup(('sysv','sysa'))
     # Decode the infos
@@ -762,6 +754,53 @@
         machine = {0x1: '68k',
                    0x2: 'PowerPC',
                    0xa: 'i386'}.get(sysa,'')
+
+    return release,versioninfo,machine
+
+def _mac_ver_xml():
+    fn = '/System/Library/CoreServices/SystemVersion.plist'
+    if not os.path.exists(fn):
+        return None
+
+    try:
+        import plistlib
+    except ImportError:
+        return None
+
+    pl = plistlib.readPlist(fn)
+    release = pl['ProductVersion']
+    versioninfo=('', '', '')
+    machine = os.uname()[4]
+    if machine == 'ppc':
+        # for compatibility with the gestalt based code
+        machine = 'PowerPC'
+
+    return release,versioninfo,machine
+
+
+def mac_ver(release='',versioninfo=('','',''),machine=''):
+
+    """ Get MacOS version information and return it as tuple (release,
+        versioninfo, machine) with versioninfo being a tuple (version,
+        dev_stage, non_release_version).
+
+        Entries which cannot be determined are set to the paramter values
+        which default to ''. All tuple entries are strings.
+    """
+
+    # First try reading the information from an XML file which should
+    # always be present
+    info = _mac_ver_xml()
+    if info is not None:
+        return info
+
+    # If that doesn't work for some reason fall back to reading the
+    # information using gestalt calls.
+    info = _mac_ver_gestalt()
+    if info is not None:
+        return info
+
+    # If that also doesn't work return the default values
     return release,versioninfo,machine
 
 def _java_getprop(name,default):

Modified: python/branches/release26-maint/Lib/test/test_platform.py
==============================================================================
--- python/branches/release26-maint/Lib/test/test_platform.py	(original)
+++ python/branches/release26-maint/Lib/test/test_platform.py	Fri Jul 23 14:45:53 2010
@@ -120,6 +120,25 @@
             else:
                 self.assertEquals(res[2], 'PowerPC')
 
+
+    if sys.platform == 'darwin':
+        def test_mac_ver_with_fork(self):
+            # Issue7895: platform.mac_ver() crashes when using fork without exec
+            #
+            # This test checks that the fix for that issue works.
+            #
+            pid = os.fork()
+            if pid == 0:
+                # child
+                info = platform.mac_ver()
+                os._exit(0)
+
+            else:
+                # parent
+                cpid, sts = os.waitpid(pid, 0)
+                self.assertEquals(cpid, pid)
+                self.assertEquals(sts, 0)
+
     def test_dist(self):
         res = platform.dist()
 

Modified: python/branches/release26-maint/Misc/NEWS
==============================================================================
--- python/branches/release26-maint/Misc/NEWS	(original)
+++ python/branches/release26-maint/Misc/NEWS	Fri Jul 23 14:45:53 2010
@@ -81,6 +81,8 @@
 Library
 -------
 
+- Issue #7895: platform.mac_ver() no longer crashes after calling os.fork()
+
 - Issue #5395: array.fromfile() would raise a spurious EOFError when an
   I/O error occurred.  Now an IOError is raised instead.  Patch by chuck
   (Jan Hosang).

From python-checkins at python.org  Fri Jul 23 15:53:51 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Fri, 23 Jul 2010 15:53:51 +0200 (CEST)
Subject: [Python-checkins] r83088 - in python/branches/py3k:
	Lib/test/test_posix.py Misc/NEWS Modules/posixmodule.c
Message-ID: <20100723135351.A6B46F651@mail.python.org>

Author: ronald.oussoren
Date: Fri Jul 23 15:53:51 2010
New Revision: 83088

Log:
This fixes issue7900 by adding code that deals
with the fact that getgroups(2) might return
more that MAX_GROUPS on OSX.

See the issue (and python-dev archives) for the
gory details. Summarized: OSX behaves rather oddly
and Apple says this is intentional.


Modified:
   python/branches/py3k/Lib/test/test_posix.py
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/Modules/posixmodule.c

Modified: python/branches/py3k/Lib/test/test_posix.py
==============================================================================
--- python/branches/py3k/Lib/test/test_posix.py	(original)
+++ python/branches/py3k/Lib/test/test_posix.py	Fri Jul 23 15:53:51 2010
@@ -6,6 +6,7 @@
 posix = support.import_module('posix')
 
 import errno
+import sys
 import time
 import os
 import pwd
@@ -349,9 +350,59 @@
                 os.chdir(curdir)
                 support.rmtree(base_path)
 
+    def test_getgroups(self):
+        with os.popen('id -G') as idg:
+            groups = idg.read().strip()
+
+        if not groups:
+            raise unittest.SkipTest("need working 'id -G'")
+
+        self.assertEqual([int(x) for x in groups.split()], posix.getgroups())
+
+class PosixGroupsTester(unittest.TestCase):
+
+    def setUp(self):
+        if posix.getuid() != 0:
+            raise unittest.SkipTest("not enough privileges")
+        if not hasattr(posix, 'getgroups'):
+            raise unittest.SkipTest("need posix.getgroups")
+        if sys.platform == 'darwin':
+            raise unittest.SkipTest("getgroups(2) is broken on OSX")
+        self.saved_groups = posix.getgroups()
+
+    def tearDown(self):
+        if hasattr(posix, 'setgroups'):
+            posix.setgroups(self.saved_groups)
+        elif hasattr(posix, 'initgroups'):
+            name = pwd.getpwuid(posix.getuid()).pw_name
+            posix.initgroups(name, self.saved_groups[0])
+
+    @unittest.skipUnless(hasattr(posix, 'initgroups'),
+                         "test needs posix.initgroups()")
+    def test_initgroups(self):
+        # find missing group
+
+        groups = sorted(self.saved_groups)
+        for g1,g2 in zip(groups[:-1], groups[1:]):
+            g = g1 + 1
+            if g < g2:
+                break
+        else:
+            g = g2 + 1
+        name = pwd.getpwuid(posix.getuid()).pw_name
+        posix.initgroups(name, g)
+        self.assertIn(g, posix.getgroups())
+
+    @unittest.skipUnless(hasattr(posix, 'setgroups'),
+                         "test needs posix.setgroups()")
+    def test_setgroups(self):
+        for groups in [[0], range(16)]:
+            posix.setgroups(groups)
+            self.assertListEqual(groups, posix.getgroups())
+
 
 def test_main():
-    support.run_unittest(PosixTester)
+    support.run_unittest(PosixTester, PosixGroupsTester)
 
 if __name__ == '__main__':
     test_main()

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Fri Jul 23 15:53:51 2010
@@ -1440,6 +1440,13 @@
 Extension Modules
 -----------------
 
+- Issue #7900: The getgroups(2) system call on MacOSX behaves rather oddly
+  compared to other unix systems. In particular, os.getgroups() does
+  not reflect any changes made using os.setgroups() but basicly always
+  returns the same information as the id command.
+
+  os.getgroups() can now return more than 16 groups on MacOSX.
+
 - Issue #6095: Make directory argument to os.listdir optional.
 
 - Issue #9277: Fix bug in struct.pack for bools in standard mode

Modified: python/branches/py3k/Modules/posixmodule.c
==============================================================================
--- python/branches/py3k/Modules/posixmodule.c	(original)
+++ python/branches/py3k/Modules/posixmodule.c	Fri Jul 23 15:53:51 2010
@@ -4166,17 +4166,49 @@
 #define MAX_GROUPS 64
 #endif
     gid_t grouplist[MAX_GROUPS];
+
+    /* On MacOSX getgroups(2) can return more than MAX_GROUPS results 
+     * This is a helper variable to store the intermediate result when
+     * that happens.
+     *
+     * To keep the code readable the OSX behaviour is unconditional,
+     * according to the POSIX spec this should be safe on all unix-y
+     * systems.
+     */
+    gid_t* alt_grouplist = grouplist;
     int n;
 
     n = getgroups(MAX_GROUPS, grouplist);
-    if (n < 0)
-        posix_error();
-    else {
-        result = PyList_New(n);
-        if (result != NULL) {
+    if (n < 0) {
+        if (errno == EINVAL) {
+            n = getgroups(0, NULL);
+            if (n == -1) {
+                return posix_error();
+            }
+            if (n == 0) {
+                /* Avoid malloc(0) */
+                alt_grouplist = grouplist;
+            } else {
+                alt_grouplist = PyMem_Malloc(n * sizeof(gid_t));
+                if (alt_grouplist == NULL) {
+                    errno = EINVAL;
+                    return posix_error();
+                }
+                n = getgroups(n, alt_grouplist);
+                if (n == -1) {
+                    PyMem_Free(alt_grouplist);
+                    return posix_error();
+                }
+            }
+        } else {
+            return posix_error();
+        }
+    }
+    result = PyList_New(n);
+    if (result != NULL) {
         int i;
         for (i = 0; i < n; ++i) {
-            PyObject *o = PyLong_FromLong((long)grouplist[i]);
+            PyObject *o = PyLong_FromLong((long)alt_grouplist[i]);
             if (o == NULL) {
             Py_DECREF(result);
             result = NULL;
@@ -4184,7 +4216,10 @@
             }
             PyList_SET_ITEM(result, i, o);
         }
-        }
+    }
+
+    if (alt_grouplist != grouplist) {
+        PyMem_Free(alt_grouplist);
     }
 
     return result;

From python-checkins at python.org  Fri Jul 23 15:54:15 2010
From: python-checkins at python.org (brett.cannon)
Date: Fri, 23 Jul 2010 15:54:15 +0200 (CEST)
Subject: [Python-checkins] r83089 - in python/branches/py3k:
	Lib/test/test_calendar.py Misc/ACKS
Message-ID: <20100723135415.0660BF651@mail.python.org>

Author: brett.cannon
Date: Fri Jul 23 15:54:14 2010
New Revision: 83089

Log:
Test calendar.monthrange.

Closes issue 9342. Thanks John Chandler for the patch.


Modified:
   python/branches/py3k/Lib/test/test_calendar.py
   python/branches/py3k/Misc/ACKS

Modified: python/branches/py3k/Lib/test/test_calendar.py
==============================================================================
--- python/branches/py3k/Lib/test/test_calendar.py	(original)
+++ python/branches/py3k/Lib/test/test_calendar.py	Fri Jul 23 15:54:14 2010
@@ -389,6 +389,34 @@
             tuple = time.gmtime(secs)
             self.assertEqual(secs, calendar.timegm(tuple))
 
+class MonthRangeTestCase(unittest.TestCase):
+    def test_january(self):
+        # Tests valid lower boundary case.
+        self.assertEqual(calendar.monthrange(2004,1), (3,31))
+
+    def test_february_leap(self):
+        # Tests February during leap year.
+        self.assertEqual(calendar.monthrange(2004,2), (6,29))
+
+    def test_february_nonleap(self):
+        # Tests February in non-leap year.
+        self.assertEqual(calendar.monthrange(2010,2), (0,28))
+
+    def test_december(self):
+        # Tests valid upper boundary case.
+        self.assertEqual(calendar.monthrange(2004,12), (2,31))
+
+    def test_zeroth_month(self):
+        # Tests low invalid boundary case.
+        with self.assertRaises(calendar.IllegalMonthError):
+            calendar.monthrange(2004, 0)
+
+    def test_thirteenth_month(self):
+        # Tests high invalid boundary case.
+        with self.assertRaises(calendar.IllegalMonthError):
+            calendar.monthrange(2004, 13)
+
+
 def test_main():
     support.run_unittest(
         OutputTestCase,
@@ -396,6 +424,7 @@
         MondayTestCase,
         SundayTestCase,
         TimegmTestCase,
+        MonthRangeTestCase,
     )
 
 

Modified: python/branches/py3k/Misc/ACKS
==============================================================================
--- python/branches/py3k/Misc/ACKS	(original)
+++ python/branches/py3k/Misc/ACKS	Fri Jul 23 15:54:14 2010
@@ -129,6 +129,7 @@
 Per Cederqvist
 Octavian Cerna
 Pascal Chambon
+John Chandler
 Hye-Shik Chang
 Jeffrey Chang
 Mitch Chapman

From python-checkins at python.org  Fri Jul 23 16:03:17 2010
From: python-checkins at python.org (brett.cannon)
Date: Fri, 23 Jul 2010 16:03:17 +0200 (CEST)
Subject: [Python-checkins] r83090 -
	python/branches/py3k/Lib/test/test_import.py
Message-ID: <20100723140317.119C4EE9EF@mail.python.org>

Author: brett.cannon
Date: Fri Jul 23 16:03:16 2010
New Revision: 83090

Log:
Explicitly test relative imports by reusing importlib tests.

Closes issue 8392. Thanks Virgil Dupras for the initial patch.


Modified:
   python/branches/py3k/Lib/test/test_import.py

Modified: python/branches/py3k/Lib/test/test_import.py
==============================================================================
--- python/branches/py3k/Lib/test/test_import.py	(original)
+++ python/branches/py3k/Lib/test/test_import.py	Fri Jul 23 16:03:16 2010
@@ -1,6 +1,8 @@
 import builtins
 import errno
 import imp
+from importlib.test.import_ import test_relative_imports
+from importlib.test.import_ import util as importlib_util
 import marshal
 import os
 import py_compile
@@ -633,10 +635,21 @@
                          os.path.join(os.curdir, foo_pyc))
 
 
+class RelativeImportTests(test_relative_imports.RelativeImports):
+
+    def setUp(self):
+        self._importlib_util_flag = importlib_util.using___import__
+        importlib_util.using___import__ = True
+
+    def tearDown(self):
+        importlib_util.using___import__ = self._importlib_util_flag
+
+
 def test_main(verbose=None):
     run_unittest(ImportTests, PycacheTests,
                  PycRewritingTests, PathsTests, RelativeImportTests,
-                 OverridingImportBuiltinTests)
+                 OverridingImportBuiltinTests,
+                 RelativeImportTests)
 
 
 if __name__ == '__main__':

From python-checkins at python.org  Fri Jul 23 16:45:19 2010
From: python-checkins at python.org (brett.cannon)
Date: Fri, 23 Jul 2010 16:45:19 +0200 (CEST)
Subject: [Python-checkins] r83091 -
	python/branches/py3k/Lib/test/test_import.py
Message-ID: <20100723144519.69FD3EE991@mail.python.org>

Author: brett.cannon
Date: Fri Jul 23 16:45:19 2010
New Revision: 83091

Log:
Stop shadowing a test class.

Modified:
   python/branches/py3k/Lib/test/test_import.py

Modified: python/branches/py3k/Lib/test/test_import.py
==============================================================================
--- python/branches/py3k/Lib/test/test_import.py	(original)
+++ python/branches/py3k/Lib/test/test_import.py	Fri Jul 23 16:45:19 2010
@@ -635,7 +635,7 @@
                          os.path.join(os.curdir, foo_pyc))
 
 
-class RelativeImportTests(test_relative_imports.RelativeImports):
+class RelativeImportFromImportlibTests(test_relative_imports.RelativeImports):
 
     def setUp(self):
         self._importlib_util_flag = importlib_util.using___import__
@@ -649,7 +649,7 @@
     run_unittest(ImportTests, PycacheTests,
                  PycRewritingTests, PathsTests, RelativeImportTests,
                  OverridingImportBuiltinTests,
-                 RelativeImportTests)
+                 RelativeImportFromImportlibTests)
 
 
 if __name__ == '__main__':

From python-checkins at python.org  Fri Jul 23 17:43:14 2010
From: python-checkins at python.org (brett.cannon)
Date: Fri, 23 Jul 2010 17:43:14 +0200 (CEST)
Subject: [Python-checkins] r83092 - python/branches/py3k/Modules/sha1module.c
Message-ID: <20100723154314.5C3EEEE9E0@mail.python.org>

Author: brett.cannon
Date: Fri Jul 23 17:43:14 2010
New Revision: 83092

Log:
Rename some macros in the sha1 module to no longer conflict with termios.h.


Modified:
   python/branches/py3k/Modules/sha1module.c

Modified: python/branches/py3k/Modules/sha1module.c
==============================================================================
--- python/branches/py3k/Modules/sha1module.c	(original)
+++ python/branches/py3k/Modules/sha1module.c	Fri Jul 23 17:43:14 2010
@@ -127,50 +127,50 @@
 
     /* compress */
     /* round one */
-    #define FF0(a,b,c,d,e,i) e = (ROLc(a, 5) + F0(b,c,d) + e + W[i] + 0x5a827999UL); b = ROLc(b, 30);
-    #define FF1(a,b,c,d,e,i) e = (ROLc(a, 5) + F1(b,c,d) + e + W[i] + 0x6ed9eba1UL); b = ROLc(b, 30);
-    #define FF2(a,b,c,d,e,i) e = (ROLc(a, 5) + F2(b,c,d) + e + W[i] + 0x8f1bbcdcUL); b = ROLc(b, 30);
-    #define FF3(a,b,c,d,e,i) e = (ROLc(a, 5) + F3(b,c,d) + e + W[i] + 0xca62c1d6UL); b = ROLc(b, 30);
+    #define FF_0(a,b,c,d,e,i) e = (ROLc(a, 5) + F0(b,c,d) + e + W[i] + 0x5a827999UL); b = ROLc(b, 30);
+    #define FF_1(a,b,c,d,e,i) e = (ROLc(a, 5) + F1(b,c,d) + e + W[i] + 0x6ed9eba1UL); b = ROLc(b, 30);
+    #define FF_2(a,b,c,d,e,i) e = (ROLc(a, 5) + F2(b,c,d) + e + W[i] + 0x8f1bbcdcUL); b = ROLc(b, 30);
+    #define FF_3(a,b,c,d,e,i) e = (ROLc(a, 5) + F3(b,c,d) + e + W[i] + 0xca62c1d6UL); b = ROLc(b, 30);
 
     for (i = 0; i < 20; ) {
-       FF0(a,b,c,d,e,i++);
-       FF0(e,a,b,c,d,i++);
-       FF0(d,e,a,b,c,i++);
-       FF0(c,d,e,a,b,i++);
-       FF0(b,c,d,e,a,i++);
+       FF_0(a,b,c,d,e,i++);
+       FF_0(e,a,b,c,d,i++);
+       FF_0(d,e,a,b,c,i++);
+       FF_0(c,d,e,a,b,i++);
+       FF_0(b,c,d,e,a,i++);
     }
 
     /* round two */
     for (; i < 40; )  {
-       FF1(a,b,c,d,e,i++);
-       FF1(e,a,b,c,d,i++);
-       FF1(d,e,a,b,c,i++);
-       FF1(c,d,e,a,b,i++);
-       FF1(b,c,d,e,a,i++);
+       FF_1(a,b,c,d,e,i++);
+       FF_1(e,a,b,c,d,i++);
+       FF_1(d,e,a,b,c,i++);
+       FF_1(c,d,e,a,b,i++);
+       FF_1(b,c,d,e,a,i++);
     }
 
     /* round three */
     for (; i < 60; )  {
-       FF2(a,b,c,d,e,i++);
-       FF2(e,a,b,c,d,i++);
-       FF2(d,e,a,b,c,i++);
-       FF2(c,d,e,a,b,i++);
-       FF2(b,c,d,e,a,i++);
+       FF_2(a,b,c,d,e,i++);
+       FF_2(e,a,b,c,d,i++);
+       FF_2(d,e,a,b,c,i++);
+       FF_2(c,d,e,a,b,i++);
+       FF_2(b,c,d,e,a,i++);
     }
 
     /* round four */
     for (; i < 80; )  {
-       FF3(a,b,c,d,e,i++);
-       FF3(e,a,b,c,d,i++);
-       FF3(d,e,a,b,c,i++);
-       FF3(c,d,e,a,b,i++);
-       FF3(b,c,d,e,a,i++);
+       FF_3(a,b,c,d,e,i++);
+       FF_3(e,a,b,c,d,i++);
+       FF_3(d,e,a,b,c,i++);
+       FF_3(c,d,e,a,b,i++);
+       FF_3(b,c,d,e,a,i++);
     }
 
-    #undef FF0
-    #undef FF1
-    #undef FF2
-    #undef FF3
+    #undef FF_0
+    #undef FF_1
+    #undef FF_2
+    #undef FF_3
 
     /* store */
     sha1->state[0] = sha1->state[0] + a;

From python-checkins at python.org  Fri Jul 23 17:46:03 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Fri, 23 Jul 2010 17:46:03 +0200 (CEST)
Subject: [Python-checkins] r83093 - in python/branches/release31-maint:
	Lib/test/test_posix.py Misc/NEWS Modules/posixmodule.c
Message-ID: <20100723154603.BCBF8EEA18@mail.python.org>

Author: ronald.oussoren
Date: Fri Jul 23 17:46:03 2010
New Revision: 83093

Log:
This fixes issue7900 by adding code that deals
with the fact that getgroups(2) might return
more that MAX_GROUPS on OSX.

See the issue (and python-dev archives) for the
gory details. Summarized: OSX behaves rather oddly
and Apple says this is intentional.


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/test/test_posix.py
   python/branches/release31-maint/Misc/NEWS
   python/branches/release31-maint/Modules/posixmodule.c

Modified: python/branches/release31-maint/Lib/test/test_posix.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_posix.py	(original)
+++ python/branches/release31-maint/Lib/test/test_posix.py	Fri Jul 23 17:46:03 2010
@@ -5,6 +5,7 @@
 # Skip these tests if there is no posix module.
 posix = support.import_module('posix')
 
+import sys
 import time
 import os
 import pwd
@@ -273,9 +274,59 @@
                 os.chdir(curdir)
                 support.rmtree(base_path)
 
+    def test_getgroups(self):
+        with os.popen('id -G') as idg:
+            groups = idg.read().strip()
+
+        if not groups:
+            raise unittest.SkipTest("need working 'id -G'")
+
+        self.assertEqual([int(x) for x in groups.split()], posix.getgroups())
+
+class PosixGroupsTester(unittest.TestCase):
+
+    def setUp(self):
+        if posix.getuid() != 0:
+            raise unittest.SkipTest("not enough privileges")
+        if not hasattr(posix, 'getgroups'):
+            raise unittest.SkipTest("need posix.getgroups")
+        if sys.platform == 'darwin':
+            raise unittest.SkipTest("getgroups(2) is broken on OSX")
+        self.saved_groups = posix.getgroups()
+
+    def tearDown(self):
+        if hasattr(posix, 'setgroups'):
+            posix.setgroups(self.saved_groups)
+        elif hasattr(posix, 'initgroups'):
+            name = pwd.getpwuid(posix.getuid()).pw_name
+            posix.initgroups(name, self.saved_groups[0])
+
+    @unittest.skipUnless(hasattr(posix, 'initgroups'),
+                         "test needs posix.initgroups()")
+    def test_initgroups(self):
+        # find missing group
+
+        groups = sorted(self.saved_groups)
+        for g1,g2 in zip(groups[:-1], groups[1:]):
+            g = g1 + 1
+            if g < g2:
+                break
+        else:
+            g = g2 + 1
+        name = pwd.getpwuid(posix.getuid()).pw_name
+        posix.initgroups(name, g)
+        self.assertIn(g, posix.getgroups())
+
+    @unittest.skipUnless(hasattr(posix, 'setgroups'),
+                         "test needs posix.setgroups()")
+    def test_setgroups(self):
+        for groups in [[0], range(16)]:
+            posix.setgroups(groups)
+            self.assertListEqual(groups, posix.getgroups())
+
 
 def test_main():
-    support.run_unittest(PosixTester)
+    support.run_unittest(PosixTester, PosixGroupsTester)
 
 if __name__ == '__main__':
     test_main()

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Fri Jul 23 17:46:03 2010
@@ -316,6 +316,13 @@
 Extension Modules
 -----------------
 
+- Issue #7900: The getgroups(2) system call on MacOSX behaves rather oddly
+  compared to other unix systems. In particular, os.getgroups() does
+  not reflect any changes made using os.setgroups() but basicly always
+  returns the same information as the id command.
+
+  os.getgroups() can now return more than 16 groups on MacOSX.
+
 - Issue #7384: If the system readline library is linked against
   ncurses, do not link the readline module against ncursesw. The
   additional restriction of linking the readline and curses modules

Modified: python/branches/release31-maint/Modules/posixmodule.c
==============================================================================
--- python/branches/release31-maint/Modules/posixmodule.c	(original)
+++ python/branches/release31-maint/Modules/posixmodule.c	Fri Jul 23 17:46:03 2010
@@ -3995,17 +3995,49 @@
 #define MAX_GROUPS 64
 #endif
     gid_t grouplist[MAX_GROUPS];
+
+    /* On MacOSX getgroups(2) can return more than MAX_GROUPS results 
+     * This is a helper variable to store the intermediate result when
+     * that happens.
+     *
+     * To keep the code readable the OSX behaviour is unconditional,
+     * according to the POSIX spec this should be safe on all unix-y
+     * systems.
+     */
+    gid_t* alt_grouplist = grouplist;
     int n;
 
     n = getgroups(MAX_GROUPS, grouplist);
-    if (n < 0)
-        posix_error();
-    else {
-        result = PyList_New(n);
-        if (result != NULL) {
+    if (n < 0) {
+        if (errno == EINVAL) {
+            n = getgroups(0, NULL);
+            if (n == -1) {
+                return posix_error();
+            }
+            if (n == 0) {
+                /* Avoid malloc(0) */
+                alt_grouplist = grouplist;
+            } else {
+                alt_grouplist = PyMem_Malloc(n * sizeof(gid_t));
+                if (alt_grouplist == NULL) {
+                    errno = EINVAL;
+                    return posix_error();
+                }
+                n = getgroups(n, alt_grouplist);
+                if (n == -1) {
+                    PyMem_Free(alt_grouplist);
+                    return posix_error();
+                }
+            }
+        } else {
+            return posix_error();
+        }
+    }
+    result = PyList_New(n);
+    if (result != NULL) {
         int i;
         for (i = 0; i < n; ++i) {
-            PyObject *o = PyLong_FromLong((long)grouplist[i]);
+            PyObject *o = PyLong_FromLong((long)alt_grouplist[i]);
             if (o == NULL) {
             Py_DECREF(result);
             result = NULL;
@@ -4013,7 +4045,10 @@
             }
             PyList_SET_ITEM(result, i, o);
         }
-        }
+    }
+
+    if (alt_grouplist != grouplist) {
+        PyMem_Free(alt_grouplist);
     }
 
     return result;

From python-checkins at python.org  Fri Jul 23 17:50:02 2010
From: python-checkins at python.org (brett.cannon)
Date: Fri, 23 Jul 2010 17:50:02 +0200 (CEST)
Subject: [Python-checkins] r83094 - python/branches/py3k/Lib/threading.py
Message-ID: <20100723155002.E7D3FEE9F8@mail.python.org>

Author: brett.cannon
Date: Fri Jul 23 17:50:02 2010
New Revision: 83094

Log:
Rip out old testing code that was inlined in threading.

Partially closes issue 9346. Thanks Brian Brazil for the patch.


Modified:
   python/branches/py3k/Lib/threading.py

Modified: python/branches/py3k/Lib/threading.py
==============================================================================
--- python/branches/py3k/Lib/threading.py	(original)
+++ python/branches/py3k/Lib/threading.py	Fri Jul 23 17:50:02 2010
@@ -871,91 +871,3 @@
         _active.clear()
         _active.update(new_active)
         assert len(_active) == 1
-
-
-# Self-test code
-
-def _test():
-
-    class BoundedQueue(_Verbose):
-
-        def __init__(self, limit):
-            _Verbose.__init__(self)
-            self.mon = RLock()
-            self.rc = Condition(self.mon)
-            self.wc = Condition(self.mon)
-            self.limit = limit
-            self.queue = deque()
-
-        def put(self, item):
-            self.mon.acquire()
-            while len(self.queue) >= self.limit:
-                self._note("put(%s): queue full", item)
-                self.wc.wait()
-            self.queue.append(item)
-            self._note("put(%s): appended, length now %d",
-                       item, len(self.queue))
-            self.rc.notify()
-            self.mon.release()
-
-        def get(self):
-            self.mon.acquire()
-            while not self.queue:
-                self._note("get(): queue empty")
-                self.rc.wait()
-            item = self.queue.popleft()
-            self._note("get(): got %s, %d left", item, len(self.queue))
-            self.wc.notify()
-            self.mon.release()
-            return item
-
-    class ProducerThread(Thread):
-
-        def __init__(self, queue, quota):
-            Thread.__init__(self, name="Producer")
-            self.queue = queue
-            self.quota = quota
-
-        def run(self):
-            from random import random
-            counter = 0
-            while counter < self.quota:
-                counter = counter + 1
-                self.queue.put("%s.%d" % (self.name, counter))
-                _sleep(random() * 0.00001)
-
-
-    class ConsumerThread(Thread):
-
-        def __init__(self, queue, count):
-            Thread.__init__(self, name="Consumer")
-            self.queue = queue
-            self.count = count
-
-        def run(self):
-            while self.count > 0:
-                item = self.queue.get()
-                print(item)
-                self.count = self.count - 1
-
-    NP = 3
-    QL = 4
-    NI = 5
-
-    Q = BoundedQueue(QL)
-    P = []
-    for i in range(NP):
-        t = ProducerThread(Q, NI)
-        t.name = "Producer-%d" % (i+1)
-        P.append(t)
-    C = ConsumerThread(Q, NI*NP)
-    for t in P:
-        t.start()
-        _sleep(0.000001)
-    C.start()
-    for t in P:
-        t.join()
-    C.join()
-
-if __name__ == '__main__':
-    _test()

From python-checkins at python.org  Fri Jul 23 17:50:52 2010
From: python-checkins at python.org (brett.cannon)
Date: Fri, 23 Jul 2010 17:50:52 +0200 (CEST)
Subject: [Python-checkins] r83095 -
	python/branches/py3k/Lib/test/test_threading.py
Message-ID: <20100723155052.4D468EE9F8@mail.python.org>

Author: brett.cannon
Date: Fri Jul 23 17:50:52 2010
New Revision: 83095

Log:
Add more tests for the threading.Thread.repr.

Partially closes issue 9346. Thanks to Brian Brazil for the patch.


Modified:
   python/branches/py3k/Lib/test/test_threading.py

Modified: python/branches/py3k/Lib/test/test_threading.py
==============================================================================
--- python/branches/py3k/Lib/test/test_threading.py	(original)
+++ python/branches/py3k/Lib/test/test_threading.py	Fri Jul 23 17:50:52 2010
@@ -97,7 +97,8 @@
             self.assertTrue(not t.is_alive())
             self.assertNotEqual(t.ident, 0)
             self.assertFalse(t.ident is None)
-            self.assertTrue(re.match('<TestThread\(.*, \w+ -?\d+\)>', repr(t)))
+            self.assertTrue(re.match('<TestThread\(.*, stopped -?\d+\)>',
+                                     repr(t)))
         if verbose:
             print('all tasks done')
         self.assertEqual(numrunning.get(), 0)
@@ -413,6 +414,12 @@
         e.isSet()
         threading.activeCount()
 
+def test_repr_daemon(self):
+    t = threading.Thread()
+    self.assertFalse('daemon' in repr(t))
+    t.daemon = True
+    self.assertTrue('daemon' in repr(t))
+
 
 class ThreadJoinOnShutdown(BaseTestCase):
 

From python-checkins at python.org  Fri Jul 23 18:05:35 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Fri, 23 Jul 2010 18:05:35 +0200 (CEST)
Subject: [Python-checkins] r83096 - python/branches/py3k/Modules/getpath.c
Message-ID: <20100723160535.5B184F496@mail.python.org>

Author: ronald.oussoren
Date: Fri Jul 23 18:05:35 2010
New Revision: 83096

Log:
Ensure that sys.prefix can reliably be found
on OSX. This fixes a small issue that was exposed
by running test_subprocess through regrtest (and
hence in a subdirectory).

Without this patch running python.exe from the
build tree will fail when these tree conditions
are true:
1) the CWD is not the root of build tree
2) python.exe is found through $PATH
3) the framework is not yet installed



Modified:
   python/branches/py3k/Modules/getpath.c

Modified: python/branches/py3k/Modules/getpath.c
==============================================================================
--- python/branches/py3k/Modules/getpath.c	(original)
+++ python/branches/py3k/Modules/getpath.c	Fri Jul 23 18:05:35 2010
@@ -552,7 +552,7 @@
         if (!ismodule(argv0_path)) {
                 /* We are in the build directory so use the name of the
                    executable - we know that the absolute path is passed */
-                wcsncpy(argv0_path, prog, MAXPATHLEN);
+                wcsncpy(argv0_path, progpath, MAXPATHLEN);
         }
         else {
                 /* Use the location of the library as the progpath */

From python-checkins at python.org  Fri Jul 23 18:09:28 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Fri, 23 Jul 2010 18:09:28 +0200 (CEST)
Subject: [Python-checkins] r83097 - in python/branches/release31-maint:
	Modules/getpath.c
Message-ID: <20100723160928.6C016EEA20@mail.python.org>

Author: ronald.oussoren
Date: Fri Jul 23 18:09:28 2010
New Revision: 83097

Log:
Merged revisions 83096 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83096 | ronald.oussoren | 2010-07-23 17:05:35 +0100 (Fri, 23 Jul 2010) | 13 lines
  
  Ensure that sys.prefix can reliably be found
  on OSX. This fixes a small issue that was exposed
  by running test_subprocess through regrtest (and
  hence in a subdirectory).
  
  Without this patch running python.exe from the
  build tree will fail when these tree conditions
  are true:
  1) the CWD is not the root of build tree
  2) python.exe is found through $PATH
  3) the framework is not yet installed
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Modules/getpath.c

Modified: python/branches/release31-maint/Modules/getpath.c
==============================================================================
--- python/branches/release31-maint/Modules/getpath.c	(original)
+++ python/branches/release31-maint/Modules/getpath.c	Fri Jul 23 18:09:28 2010
@@ -552,7 +552,7 @@
         if (!ismodule(argv0_path)) {
                 /* We are in the build directory so use the name of the
                    executable - we know that the absolute path is passed */
-                wcsncpy(argv0_path, prog, MAXPATHLEN);
+                wcsncpy(argv0_path, progpath, MAXPATHLEN);
         }
         else {
                 /* Use the location of the library as the progpath */

From python-checkins at python.org  Fri Jul 23 18:12:28 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Fri, 23 Jul 2010 18:12:28 +0200 (CEST)
Subject: [Python-checkins] r83098 - in python/branches/release27-maint:
	Modules/getpath.c
Message-ID: <20100723161228.D8466EE9E4@mail.python.org>

Author: ronald.oussoren
Date: Fri Jul 23 18:12:28 2010
New Revision: 83098

Log:
Merged revisions 83096 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83096 | ronald.oussoren | 2010-07-23 17:05:35 +0100 (Fri, 23 Jul 2010) | 13 lines
  
  Ensure that sys.prefix can reliably be found
  on OSX. This fixes a small issue that was exposed
  by running test_subprocess through regrtest (and
  hence in a subdirectory).
  
  Without this patch running python.exe from the
  build tree will fail when these tree conditions
  are true:
  1) the CWD is not the root of build tree
  2) python.exe is found through $PATH
  3) the framework is not yet installed
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Modules/getpath.c

Modified: python/branches/release27-maint/Modules/getpath.c
==============================================================================
--- python/branches/release27-maint/Modules/getpath.c	(original)
+++ python/branches/release27-maint/Modules/getpath.c	Fri Jul 23 18:12:28 2010
@@ -471,7 +471,7 @@
         if (!ismodule(argv0_path)) {
                 /* We are in the build directory so use the name of the
                    executable - we know that the absolute path is passed */
-                strncpy(argv0_path, prog, MAXPATHLEN);
+                strncpy(argv0_path, progpath, MAXPATHLEN);
         }
         else {
                 /* Use the location of the library as the progpath */

From python-checkins at python.org  Fri Jul 23 18:20:41 2010
From: python-checkins at python.org (richard.jones)
Date: Fri, 23 Jul 2010 18:20:41 +0200 (CEST)
Subject: [Python-checkins] r83099 -
	python/branches/py3k/Lib/test/test_smtpd.py
Message-ID: <20100723162041.091F9EEA06@mail.python.org>

Author: richard.jones
Date: Fri Jul 23 18:20:40 2010
New Revision: 83099

Log:
New tests for smtpd module. Has full coverage for SMTPChannel and SMTPServer.


Added:
   python/branches/py3k/Lib/test/test_smtpd.py

Added: python/branches/py3k/Lib/test/test_smtpd.py
==============================================================================
--- (empty file)
+++ python/branches/py3k/Lib/test/test_smtpd.py	Fri Jul 23 18:20:40 2010
@@ -0,0 +1,221 @@
+import asynchat
+from unittest import TestCase
+import socket
+from test import support
+import asyncore
+import io
+import smtpd
+
+# mock-ish socket to sit underneath asyncore
+class DummySocket:
+    def __init__(self):
+        self.output = []
+        self.queue = []
+        self.conn = None
+    def queue_recv(self, line):
+        self.queue.append(line)
+    def recv(self, *args):
+        data = self.queue.pop(0) + b'\r\n'
+        return data
+    def fileno(self):
+        return 0
+    def setsockopt(self, *args):
+        pass
+    def getsockopt(self, *args):
+        return 0
+    def bind(self, *args):
+        pass
+    def accept(self):
+        self.conn = DummySocket()
+        return self.conn, 'c'
+    def listen(self, *args):
+        pass
+    def setblocking(self, *args):
+        pass
+    def send(self, data):
+        self.last = data
+        self.output.append(data)
+        return len(data)
+    def getpeername(self):
+        return 'peer'
+    def close(self):
+        pass
+
+class DummyServer(smtpd.SMTPServer):
+    messages = []
+    def create_socket(self, family, type):
+        self.family_and_type = (socket.AF_INET, socket.SOCK_STREAM)
+        self.set_socket(DummySocket())
+    def process_message(self, peer, mailfrom, rcpttos, data):
+        self.messages.append((peer, mailfrom, rcpttos, data))
+        if data == 'return status':
+            return '250 Okish'
+
+class DummyDispatcherBroken(Exception):
+    pass
+
+class BrokenDummyServer(DummyServer):
+    def listen(self, num):
+        raise DummyDispatcherBroken()
+
+class SMTPDChannelTest(TestCase):
+    def setUp(self):
+        self.debug = smtpd.DEBUGSTREAM = io.StringIO()
+        self.server = DummyServer('a', 'b')
+        conn, addr = self.server.accept()
+        self.channel = smtpd.SMTPChannel(self.server, conn, addr)
+
+    def write_line(self, line):
+        self.channel.socket.queue_recv(line)
+        self.channel.handle_read()
+
+    def test_broken_connect(self):
+        self.assertRaises(DummyDispatcherBroken, BrokenDummyServer, 'a', 'b')
+
+    def test_server_accept(self):
+        self.server.handle_accept()
+
+    def test_missing_data(self):
+        self.write_line(b'')
+        self.assertEqual(self.channel.last,
+                         b'500 Error: bad syntax\r\n')
+
+    def test_EHLO_not_implemented(self):
+        self.write_line(b'EHLO test.example')
+        self.assertEqual(self.channel.last,
+                         b'502 Error: command "EHLO" not implemented\r\n')
+
+    def test_HELO(self):
+        name = socket.getfqdn()
+        self.write_line(b'HELO test.example')
+        self.assertEqual(self.channel.last,
+                         '250 {}\r\n'.format(name).encode('ascii'))
+
+    def test_HELO_bad_syntax(self):
+        self.write_line(b'HELO')
+        self.assertEqual(self.channel.last,
+                         b'501 Syntax: HELO hostname\r\n')
+
+    def test_HELO_duplicate(self):
+        self.write_line(b'HELO test.example')
+        self.write_line(b'HELO test.example')
+        self.assertEqual(self.channel.last,
+                         b'503 Duplicate HELO/EHLO\r\n')
+
+    def test_NOOP(self):
+        self.write_line(b'NOOP')
+        self.assertEqual(self.channel.last, b'250 Ok\r\n')
+
+    def test_NOOP_bad_syntax(self):
+        self.write_line(b'NOOP hi')
+        self.assertEqual(self.channel.last,
+                         b'501 Syntax: NOOP\r\n')
+
+    def test_QUIT(self):
+        self.write_line(b'QUIT')
+        self.assertEqual(self.channel.last, b'221 Bye\r\n')
+
+    def test_QUIT_arg_ignored(self):
+        self.write_line(b'QUIT bye bye')
+        self.assertEqual(self.channel.last, b'221 Bye\r\n')
+
+    def test_bad_state(self):
+        self.channel._SMTPChannel__state = 'BAD STATE'
+        self.write_line(b'HELO')
+        self.assertEqual(self.channel.last, b'451 Internal confusion\r\n')
+
+    def test_need_MAIL(self):
+        self.write_line(b'RCPT to:spam at example')
+        self.assertEqual(self.channel.last,
+            b'503 Error: need MAIL command\r\n')
+
+    def test_MAIL_syntax(self):
+        self.write_line(b'MAIL from eggs at example')
+        self.assertEqual(self.channel.last,
+            b'501 Syntax: MAIL FROM:<address>\r\n')
+
+    def test_MAIL_missing_from(self):
+        self.write_line(b'MAIL from:')
+        self.assertEqual(self.channel.last,
+            b'501 Syntax: MAIL FROM:<address>\r\n')
+
+    def test_MAIL_chevrons(self):
+        self.write_line(b'MAIL from:<eggs at example>')
+        self.assertEqual(self.channel.last, b'250 Ok\r\n')
+
+    def test_nested_MAIL(self):
+        self.write_line(b'MAIL from:eggs at example')
+        self.write_line(b'MAIL from:spam at example')
+        self.assertEqual(self.channel.last,
+            b'503 Error: nested MAIL command\r\n')
+
+    def test_need_RCPT(self):
+        self.write_line(b'MAIL From:eggs at example')
+        self.write_line(b'DATA')
+        self.assertEqual(self.channel.last,
+            b'503 Error: need RCPT command\r\n')
+
+    def test_RCPT_syntax(self):
+        self.write_line(b'MAIL From:eggs at example')
+        self.write_line(b'RCPT to eggs at example')
+        self.assertEqual(self.channel.last,
+            b'501 Syntax: RCPT TO: <address>\r\n')
+
+    def test_data_dialog(self):
+        self.write_line(b'MAIL From:eggs at example')
+        self.assertEqual(self.channel.last, b'250 Ok\r\n')
+        self.write_line(b'RCPT To:spam at example')
+        self.assertEqual(self.channel.last, b'250 Ok\r\n')
+
+        self.write_line(b'DATA')
+        self.assertEqual(self.channel.last,
+            b'354 End data with <CR><LF>.<CR><LF>\r\n')
+        self.write_line(b'data\r\nmore\r\n.')
+        self.assertEqual(self.channel.last, b'250 Ok\r\n')
+        self.assertEqual(self.server.messages[-1],
+            ('peer', 'eggs at example', ['spam at example'], 'data\nmore'))
+
+    def test_DATA_syntax(self):
+        self.write_line(b'MAIL From:eggs at example')
+        self.write_line(b'RCPT To:spam at example')
+        self.write_line(b'DATA spam')
+        self.assertEqual(self.channel.last, b'501 Syntax: DATA\r\n')
+
+    def test_multiple_RCPT(self):
+        self.write_line(b'MAIL From:eggs at example')
+        self.write_line(b'RCPT To:spam at example')
+        self.write_line(b'RCPT To:ham at example')
+        self.write_line(b'DATA')
+        self.write_line(b'data\r\n.')
+        self.assertEqual(self.server.messages[-1],
+            ('peer', 'eggs at example', ['spam at example','ham at example'], 'data'))
+
+    def test_manual_status(self):
+        self.write_line(b'MAIL From:eggs at example')
+        self.write_line(b'RCPT To:spam at example')
+        self.write_line(b'DATA')
+        self.write_line(b'return status\r\n.')
+        self.assertEqual(self.channel.last, b'250 Okish\r\n')
+
+    def test_RSET(self):
+        self.write_line(b'MAIL From:eggs at example')
+        self.write_line(b'RCPT To:spam at example')
+        self.write_line(b'RSET')
+        self.assertEqual(self.channel.last, b'250 Ok\r\n')
+        self.write_line(b'MAIL From:foo at example')
+        self.write_line(b'RCPT To:eggs at example')
+        self.write_line(b'DATA')
+        self.write_line(b'data\r\n.')
+        self.assertEqual(self.server.messages[0],
+            ('peer', 'foo at example', ['eggs at example'], 'data'))
+
+    def test_RSET_syntax(self):
+        self.write_line(b'RSET hi')
+        self.assertEqual(self.channel.last, b'501 Syntax: RSET\r\n')
+
+
+def test_main():
+    support.run_unittest(SMTPDChannelTest)
+
+if __name__ == "__main__":
+    test_main()

From python-checkins at python.org  Fri Jul 23 18:22:25 2010
From: python-checkins at python.org (brett.cannon)
Date: Fri, 23 Jul 2010 18:22:25 +0200 (CEST)
Subject: [Python-checkins] r83100 - python/branches/py3k/Lib/fnmatch.py
Message-ID: <20100723162225.69534EBF2@mail.python.org>

Author: brett.cannon
Date: Fri Jul 23 18:22:25 2010
New Revision: 83100

Log:
Make fnmatch be more PEP 8 compliant.

Partially closes issue 9356. Thanks to Brian Brazil for the patch.


Modified:
   python/branches/py3k/Lib/fnmatch.py

Modified: python/branches/py3k/Lib/fnmatch.py
==============================================================================
--- python/branches/py3k/Lib/fnmatch.py	(original)
+++ python/branches/py3k/Lib/fnmatch.py	Fri Jul 23 18:22:25 2010
@@ -9,20 +9,23 @@
 The function translate(PATTERN) returns a regular expression
 corresponding to PATTERN.  (It does not compile it.)
 """
-
+import os
+import posixpath
 import re
 
 __all__ = ["filter", "fnmatch", "fnmatchcase", "purge", "translate"]
 
 _cache = {}  # Maps text patterns to compiled regexen.
 _cacheb = {}  # Ditto for bytes patterns.
-_MAXCACHE = 100 # Maximum size of caches
+_MAXCACHE = 100  # Maximum size of caches.
+
 
 def purge():
-    """Clear the pattern cache"""
+    """Clear the pattern cache."""
     _cache.clear()
     _cacheb.clear()
 
+
 def fnmatch(name, pat):
     """Test whether FILENAME matches PATTERN.
 
@@ -38,12 +41,11 @@
     if the operating system requires it.
     If you don't want this, use fnmatchcase(FILENAME, PATTERN).
     """
-
-    import os
     name = os.path.normcase(name)
     pat = os.path.normcase(pat)
     return fnmatchcase(name, pat)
 
+
 def _compile_pattern(pat):
     cache = _cacheb if isinstance(pat, bytes) else _cache
     regex = cache.get(pat)
@@ -59,9 +61,9 @@
         cache[pat] = regex = re.compile(res)
     return regex.match
 
+
 def filter(names, pat):
-    """Return the subset of the list NAMES that match PAT"""
-    import os,posixpath
+    """Return the subset of the list NAMES that match PAT."""
     result = []
     pat = os.path.normcase(pat)
     match = _compile_pattern(pat)
@@ -76,16 +78,17 @@
                 result.append(name)
     return result
 
+
 def fnmatchcase(name, pat):
     """Test whether FILENAME matches PATTERN, including case.
 
     This is a version of fnmatch() which doesn't case-normalize
     its arguments.
     """
-
     match = _compile_pattern(pat)
     return match(name) is not None
 
+
 def translate(pat):
     """Translate a shell PATTERN to a regular expression.
 

From python-checkins at python.org  Fri Jul 23 18:23:13 2010
From: python-checkins at python.org (brett.cannon)
Date: Fri, 23 Jul 2010 18:23:13 +0200 (CEST)
Subject: [Python-checkins] r83101 -
	python/branches/py3k/Lib/test/test_fnmatch.py
Message-ID: <20100723162313.3BEF0EEA04@mail.python.org>

Author: brett.cannon
Date: Fri Jul 23 18:23:13 2010
New Revision: 83101

Log:
Add tests for fnmatch.filter and translate.

Partially closes issue 9356. Thanks to Brian Brazil for the patch.


Modified:
   python/branches/py3k/Lib/test/test_fnmatch.py

Modified: python/branches/py3k/Lib/test/test_fnmatch.py
==============================================================================
--- python/branches/py3k/Lib/test/test_fnmatch.py	(original)
+++ python/branches/py3k/Lib/test/test_fnmatch.py	Fri Jul 23 18:23:13 2010
@@ -3,7 +3,8 @@
 from test import support
 import unittest
 
-from fnmatch import fnmatch, fnmatchcase, _MAXCACHE, _cache, _cacheb, purge
+from fnmatch import (fnmatch, fnmatchcase, _MAXCACHE, _cache, _cacheb, purge,
+                        translate, filter)
 
 
 class FnmatchTestCase(unittest.TestCase):
@@ -80,8 +81,29 @@
         self.assertLessEqual(len(_cacheb), _MAXCACHE)
 
 
+class TranslateTestCase(unittest.TestCase):
+
+    def test_translate(self):
+        self.assertEqual(translate('*'), '.*\Z(?ms)')
+        self.assertEqual(translate('?'), '.\Z(?ms)')
+        self.assertEqual(translate('a?b*'), 'a.b.*\Z(?ms)')
+        self.assertEqual(translate('[abc]'), '[abc]\Z(?ms)')
+        self.assertEqual(translate('[]]'), '[]]\Z(?ms)')
+        self.assertEqual(translate('[!x]'), '[^x]\Z(?ms)')
+        self.assertEqual(translate('[^x]'), '[\\^x]\Z(?ms)')
+        self.assertEqual(translate('[x'), '\\[x\Z(?ms)')
+
+
+class FilterTestCase(unittest.TestCase):
+
+    def test_filter(self):
+        self.assertEqual(filter(['a', 'b'], 'a'), ['a'])
+
+
 def test_main():
-    support.run_unittest(FnmatchTestCase)
+    support.run_unittest(FnmatchTestCase,
+                         TranslateTestCase,
+                         FilterTestCase)
 
 
 if __name__ == "__main__":

From python-checkins at python.org  Fri Jul 23 18:30:10 2010
From: python-checkins at python.org (brian.curtin)
Date: Fri, 23 Jul 2010 18:30:10 +0200 (CEST)
Subject: [Python-checkins] r83102 -
	python/branches/py3k/Lib/test/test_threading.py
Message-ID: <20100723163010.B449CEEA17@mail.python.org>

Author: brian.curtin
Date: Fri Jul 23 18:30:10 2010
New Revision: 83102

Log:
Tab test_repr_daemon over so it's included in ThreadTests.
Noticed by Amaury.


Modified:
   python/branches/py3k/Lib/test/test_threading.py

Modified: python/branches/py3k/Lib/test/test_threading.py
==============================================================================
--- python/branches/py3k/Lib/test/test_threading.py	(original)
+++ python/branches/py3k/Lib/test/test_threading.py	Fri Jul 23 18:30:10 2010
@@ -414,11 +414,11 @@
         e.isSet()
         threading.activeCount()
 
-def test_repr_daemon(self):
-    t = threading.Thread()
-    self.assertFalse('daemon' in repr(t))
-    t.daemon = True
-    self.assertTrue('daemon' in repr(t))
+    def test_repr_daemon(self):
+        t = threading.Thread()
+        self.assertFalse('daemon' in repr(t))
+        t.daemon = True
+        self.assertTrue('daemon' in repr(t))
 
 
 class ThreadJoinOnShutdown(BaseTestCase):

From brian.curtin at gmail.com  Fri Jul 23 18:32:18 2010
From: brian.curtin at gmail.com (Brian Curtin)
Date: Fri, 23 Jul 2010 11:32:18 -0500
Subject: [Python-checkins] r83095 -
	python/branches/py3k/Lib/test/test_threading.py
In-Reply-To: <20100723155052.4D468EE9F8@mail.python.org>
References: <20100723155052.4D468EE9F8@mail.python.org>
Message-ID: <AANLkTinRnX2H=pQp4GgcVEU3E2pF3Wcwq0mM+SrgrYTK@mail.gmail.com>

On Fri, Jul 23, 2010 at 10:50, brett.cannon <python-checkins at python.org>wrote:

> Author: brett.cannon
> Date: Fri Jul 23 17:50:52 2010
> New Revision: 83095
>
> Log:
> Add more tests for the threading.Thread.repr.
>
> Partially closes issue 9346. Thanks to Brian Brazil for the patch.
>
>
> Modified:
>   python/branches/py3k/Lib/test/test_threading.py
>
> Modified: python/branches/py3k/Lib/test/test_threading.py
>
> ==============================================================================
> --- python/branches/py3k/Lib/test/test_threading.py     (original)
> +++ python/branches/py3k/Lib/test/test_threading.py     Fri Jul 23 17:50:52
> 2010
> @@ -97,7 +97,8 @@
>             self.assertTrue(not t.is_alive())
>             self.assertNotEqual(t.ident, 0)
>             self.assertFalse(t.ident is None)
> -            self.assertTrue(re.match('<TestThread\(.*, \w+ -?\d+\)>',
> repr(t)))
> +            self.assertTrue(re.match('<TestThread\(.*, stopped -?\d+\)>',
> +                                     repr(t)))
>         if verbose:
>             print('all tasks done')
>         self.assertEqual(numrunning.get(), 0)
> @@ -413,6 +414,12 @@
>         e.isSet()
>         threading.activeCount()
>
> +def test_repr_daemon(self):
> +    t = threading.Thread()
> +    self.assertFalse('daemon' in repr(t))
> +    t.daemon = True
> +    self.assertTrue('daemon' in repr(t))
> +
>
>  class ThreadJoinOnShutdown(BaseTestCase):


Amaury noticed that the test wasn't tabbed over so I fixed it in r83102.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-checkins/attachments/20100723/7af5f60e/attachment.html>

From brett at python.org  Fri Jul 23 18:39:17 2010
From: brett at python.org (Brett Cannon)
Date: Fri, 23 Jul 2010 17:39:17 +0100
Subject: [Python-checkins] r83072 - in python/branches/py3k:
	Doc/library/test.rst Misc/NEWS
In-Reply-To: <AANLkTinO23fgNOLGN6g7FHf_Vyu3MbWtPkaUY-n-OSMk@mail.gmail.com>
References: <20100723113132.14D99EE9CF@mail.python.org>
	<AANLkTinO23fgNOLGN6g7FHf_Vyu3MbWtPkaUY-n-OSMk@mail.gmail.com>
Message-ID: <AANLkTinYreJKeWz7yixDopxQnZnRpwY78f7Wiqk-5rWH@mail.gmail.com>

On Fri, Jul 23, 2010 at 12:36, Nick Coghlan <ncoghlan at gmail.com> wrote:

> On Fri, Jul 23, 2010 at 9:31 PM, brett.cannon
> <python-checkins at python.org> wrote:
> > Author: brett.cannon
> > Date: Fri Jul 23 13:31:31 2010
> > New Revision: 83072
> >
> > Log:
> > Document the fact that the 'test' package is meant only for use by Python
> > itself and not by others.
>
> Perhaps add a pointer to unittest for testing infrastructure that is
> covered by all the usual backwards compatibility conventions?
>

Eh, if you want. Busy with other stuff at the moment.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-checkins/attachments/20100723/6c3dbd5f/attachment.html>

From python-checkins at python.org  Fri Jul 23 18:48:22 2010
From: python-checkins at python.org (ezio.melotti)
Date: Fri, 23 Jul 2010 18:48:22 +0200 (CEST)
Subject: [Python-checkins] r83103 -
	python/branches/py3k/Lib/idlelib/PyShell.py
Message-ID: <20100723164822.DA59EF43F@mail.python.org>

Author: ezio.melotti
Date: Fri Jul 23 18:48:22 2010
New Revision: 83103

Log:
#9359: fix typo. Thanks to Piotr Kasprzyk for the patch.

Modified:
   python/branches/py3k/Lib/idlelib/PyShell.py

Modified: python/branches/py3k/Lib/idlelib/PyShell.py
==============================================================================
--- python/branches/py3k/Lib/idlelib/PyShell.py	(original)
+++ python/branches/py3k/Lib/idlelib/PyShell.py	Fri Jul 23 18:48:22 2010
@@ -548,7 +548,7 @@
         This method is called from the subprocess, and by returning from this
         method we allow the subprocess to unblock.  After a bit the shell
         requests the subprocess to open the remote stack viewer which returns a
-        static object looking at the last exceptiopn.  It is queried through
+        static object looking at the last exception.  It is queried through
         the RPC mechanism.
 
         """

From python-checkins at python.org  Fri Jul 23 18:50:08 2010
From: python-checkins at python.org (ezio.melotti)
Date: Fri, 23 Jul 2010 18:50:08 +0200 (CEST)
Subject: [Python-checkins] r83104 - in python/branches/release31-maint:
	Lib/idlelib/PyShell.py
Message-ID: <20100723165008.D91E5EE9B9@mail.python.org>

Author: ezio.melotti
Date: Fri Jul 23 18:50:08 2010
New Revision: 83104

Log:
Merged revisions 83103 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83103 | ezio.melotti | 2010-07-23 19:48:22 +0300 (Fri, 23 Jul 2010) | 1 line
  
  #9359: fix typo. Thanks to Piotr Kasprzyk for the patch.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/idlelib/PyShell.py

Modified: python/branches/release31-maint/Lib/idlelib/PyShell.py
==============================================================================
--- python/branches/release31-maint/Lib/idlelib/PyShell.py	(original)
+++ python/branches/release31-maint/Lib/idlelib/PyShell.py	Fri Jul 23 18:50:08 2010
@@ -548,7 +548,7 @@
         This method is called from the subprocess, and by returning from this
         method we allow the subprocess to unblock.  After a bit the shell
         requests the subprocess to open the remote stack viewer which returns a
-        static object looking at the last exceptiopn.  It is queried through
+        static object looking at the last exception.  It is queried through
         the RPC mechanism.
 
         """

From python-checkins at python.org  Fri Jul 23 18:55:21 2010
From: python-checkins at python.org (ezio.melotti)
Date: Fri, 23 Jul 2010 18:55:21 +0200 (CEST)
Subject: [Python-checkins] r83105 - in python/branches/release27-maint:
	Lib/idlelib/PyShell.py
Message-ID: <20100723165521.D1AC5EEA27@mail.python.org>

Author: ezio.melotti
Date: Fri Jul 23 18:55:21 2010
New Revision: 83105

Log:
Merged revisions 83103 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83103 | ezio.melotti | 2010-07-23 19:48:22 +0300 (Fri, 23 Jul 2010) | 1 line
  
  #9359: fix typo. Thanks to Piotr Kasprzyk for the patch.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/idlelib/PyShell.py

Modified: python/branches/release27-maint/Lib/idlelib/PyShell.py
==============================================================================
--- python/branches/release27-maint/Lib/idlelib/PyShell.py	(original)
+++ python/branches/release27-maint/Lib/idlelib/PyShell.py	Fri Jul 23 18:55:21 2010
@@ -552,7 +552,7 @@
         This method is called from the subprocess, and by returning from this
         method we allow the subprocess to unblock.  After a bit the shell
         requests the subprocess to open the remote stack viewer which returns a
-        static object looking at the last exceptiopn.  It is queried through
+        static object looking at the last exception.  It is queried through
         the RPC mechanism.
 
         """

From python-checkins at python.org  Fri Jul 23 18:55:26 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 23 Jul 2010 18:55:26 +0200 (CEST)
Subject: [Python-checkins] r83106 - in python/branches/py3k/Doc:
	library/pdb.rst library/urllib.request.rst whatsnew/2.0.rst
	whatsnew/2.5.rst
Message-ID: <20100723165526.96662EEA1F@mail.python.org>

Author: georg.brandl
Date: Fri Jul 23 18:55:26 2010
New Revision: 83106

Log:
Fix some markup glitches.

Modified:
   python/branches/py3k/Doc/library/pdb.rst
   python/branches/py3k/Doc/library/urllib.request.rst
   python/branches/py3k/Doc/whatsnew/2.0.rst
   python/branches/py3k/Doc/whatsnew/2.5.rst

Modified: python/branches/py3k/Doc/library/pdb.rst
==============================================================================
--- python/branches/py3k/Doc/library/pdb.rst	(original)
+++ python/branches/py3k/Doc/library/pdb.rst	Fri Jul 23 18:55:26 2010
@@ -347,7 +347,8 @@
 .. pdbcommand:: l(ist) [first[, last]]
 
    List source code for the current file.  Without arguments, list 11 lines
-   around the current line or continue the previous listing.  With one argument,
+   around the current line or continue the previous listing.  With ``.`` as
+   argument, list 11 lines around the current line.  With one argument,
    list 11 lines around at that line.  With two arguments, list the given range;
    if the second argument is less than the first, it is interpreted as a count.
 

Modified: python/branches/py3k/Doc/library/urllib.request.rst
==============================================================================
--- python/branches/py3k/Doc/library/urllib.request.rst	(original)
+++ python/branches/py3k/Doc/library/urllib.request.rst	Fri Jul 23 18:55:26 2010
@@ -638,7 +638,8 @@
    :meth:`unknown_open`.
 
    Note that the implementation of these methods may involve calls of the parent
-   :class:`OpenerDirector` instance's :meth:`.open` and :meth:`.error` methods.
+   :class:`OpenerDirector` instance's :meth:`~OpenerDirector.open` and
+   :meth:`~OpenerDirector.error` methods.
 
 #. Every handler with a method named like :meth:`protocol_response` has that
    method called to post-process the response.

Modified: python/branches/py3k/Doc/whatsnew/2.0.rst
==============================================================================
--- python/branches/py3k/Doc/whatsnew/2.0.rst	(original)
+++ python/branches/py3k/Doc/whatsnew/2.0.rst	Fri Jul 23 18:55:26 2010
@@ -656,7 +656,7 @@
 The change which will probably break the most code is tightening up the
 arguments accepted by some methods.  Some methods would take multiple arguments
 and treat them as a tuple, particularly various list methods such as
-:meth:`.append` and :meth:`.insert`. In earlier versions of Python, if ``L`` is
+:meth:`append` and :meth:`insert`. In earlier versions of Python, if ``L`` is
 a list, ``L.append( 1,2 )`` appends the tuple ``(1,2)`` to the list.  In Python
 2.0 this causes a :exc:`TypeError` exception to be raised, with the message:
 'append requires exactly 1 argument; 2 given'.  The fix is to simply add an

Modified: python/branches/py3k/Doc/whatsnew/2.5.rst
==============================================================================
--- python/branches/py3k/Doc/whatsnew/2.5.rst	(original)
+++ python/branches/py3k/Doc/whatsnew/2.5.rst	Fri Jul 23 18:55:26 2010
@@ -1765,7 +1765,7 @@
 http://effbot.org/zone/element-index.htm.
 
 ElementTree represents an XML document as a tree of element nodes. The text
-content of the document is stored as the :attr:`.text` and :attr:`.tail`
+content of the document is stored as the :attr:`text` and :attr:`tail`
 attributes of  (This is one of the major differences between ElementTree and
 the Document Object Model; in the DOM there are many different types of node,
 including :class:`TextNode`.)

From python-checkins at python.org  Fri Jul 23 18:55:42 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 23 Jul 2010 18:55:42 +0200 (CEST)
Subject: [Python-checkins] r83107 - python/branches/py3k/Doc/Makefile
Message-ID: <20100723165542.87D27EEA24@mail.python.org>

Author: georg.brandl
Date: Fri Jul 23 18:55:42 2010
New Revision: 83107

Log:
Update to 1.0.

Modified:
   python/branches/py3k/Doc/Makefile

Modified: python/branches/py3k/Doc/Makefile
==============================================================================
--- python/branches/py3k/Doc/Makefile	(original)
+++ python/branches/py3k/Doc/Makefile	Fri Jul 23 18:55:42 2010
@@ -40,7 +40,7 @@
 checkout:
 	@if [ ! -d tools/sphinx ]; then \
 	  echo "Checking out Sphinx..."; \
-	  svn checkout $(SVNROOT)/external/Sphinx-0.6.5/sphinx tools/sphinx; \
+	  svn checkout $(SVNROOT)/external/Sphinx-1.0/sphinx tools/sphinx; \
 	fi
 	@if [ ! -d tools/docutils ]; then \
 	  echo "Checking out Docutils..."; \

From python-checkins at python.org  Fri Jul 23 18:56:21 2010
From: python-checkins at python.org (brett.cannon)
Date: Fri, 23 Jul 2010 18:56:21 +0200 (CEST)
Subject: [Python-checkins] r83108 -
	python/branches/py3k/Lib/test/test_queue.py
Message-ID: <20100723165621.C7442EEA2B@mail.python.org>

Author: brett.cannon
Date: Fri Jul 23 18:56:21 2010
New Revision: 83108

Log:
Add queue tests for empty, full, put_nowait, and get_nowait.

Closes issue 9357. Thanks to Brian Brazil for the patch.


Modified:
   python/branches/py3k/Lib/test/test_queue.py

Modified: python/branches/py3k/Lib/test/test_queue.py
==============================================================================
--- python/branches/py3k/Lib/test/test_queue.py	(original)
+++ python/branches/py3k/Lib/test/test_queue.py	Fri Jul 23 18:56:21 2010
@@ -90,6 +90,8 @@
     def simple_queue_test(self, q):
         if q.qsize():
             raise RuntimeError("Call this function with an empty queue")
+        self.assertTrue(q.empty())
+        self.assertFalse(q.full())
         # I guess we better check things actually queue correctly a little :)
         q.put(111)
         q.put(333)
@@ -108,6 +110,8 @@
         full = 3 * 2 * QUEUE_SIZE
         q.put(last)
         self.assertTrue(qfull(q), "Queue should be full")
+        self.assertFalse(q.empty())
+        self.assertTrue(q.full())
         try:
             q.put(full, block=0)
             self.fail("Didn't appear to block with a full queue")
@@ -193,6 +197,25 @@
         self.simple_queue_test(q)
         self.simple_queue_test(q)
 
+    def test_negative_timeout_raises_exception(self):
+        q = self.type2test(QUEUE_SIZE)
+        with self.assertRaises(ValueError):
+            q.put(1, timeout=-1)
+        with self.assertRaises(ValueError):
+            q.get(1, timeout=-1)
+
+    def test_nowait(self):
+        q = self.type2test(QUEUE_SIZE)
+        for i in range(QUEUE_SIZE):
+            q.put_nowait(1)
+        with self.assertRaises(queue.Full):
+            q.put_nowait(1)
+
+        for i in range(QUEUE_SIZE):
+            q.get_nowait()
+        with self.assertRaises(queue.Empty):
+            q.get_nowait()
+
 
 class QueueTest(BaseQueueTest):
     type2test = queue.Queue

From python-checkins at python.org  Fri Jul 23 18:58:21 2010
From: python-checkins at python.org (brett.cannon)
Date: Fri, 23 Jul 2010 18:58:21 +0200 (CEST)
Subject: [Python-checkins] r83109 -
	python/branches/py3k/Doc/library/fnmatch.rst
Message-ID: <20100723165821.28C47EEA24@mail.python.org>

Author: brett.cannon
Date: Fri Jul 23 18:58:21 2010
New Revision: 83109

Log:
Mention in the fnmatch docs that meta-characters in translate cannot be quoted.

Closes issue 9358. Thanks to Brian Brazil for the patch.


Modified:
   python/branches/py3k/Doc/library/fnmatch.rst

Modified: python/branches/py3k/Doc/library/fnmatch.rst
==============================================================================
--- python/branches/py3k/Doc/library/fnmatch.rst	(original)
+++ python/branches/py3k/Doc/library/fnmatch.rst	Fri Jul 23 18:58:21 2010
@@ -70,6 +70,8 @@
 
    Return the shell-style *pattern* converted to a regular expression.
 
+   Be aware there is no way to quote meta-characters.
+
    Example:
 
       >>> import fnmatch, re

From python-checkins at python.org  Fri Jul 23 18:58:35 2010
From: python-checkins at python.org (ezio.melotti)
Date: Fri, 23 Jul 2010 18:58:35 +0200 (CEST)
Subject: [Python-checkins] r83110 - in python/branches/release26-maint:
	Lib/idlelib/PyShell.py
Message-ID: <20100723165835.D8046EEA21@mail.python.org>

Author: ezio.melotti
Date: Fri Jul 23 18:58:35 2010
New Revision: 83110

Log:
Merged revisions 83105 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

................
  r83105 | ezio.melotti | 2010-07-23 19:55:21 +0300 (Fri, 23 Jul 2010) | 9 lines
  
  Merged revisions 83103 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k
  
  ........
    r83103 | ezio.melotti | 2010-07-23 19:48:22 +0300 (Fri, 23 Jul 2010) | 1 line
    
    #9359: fix typo. Thanks to Piotr Kasprzyk for the patch.
  ........
................


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Lib/idlelib/PyShell.py

Modified: python/branches/release26-maint/Lib/idlelib/PyShell.py
==============================================================================
--- python/branches/release26-maint/Lib/idlelib/PyShell.py	(original)
+++ python/branches/release26-maint/Lib/idlelib/PyShell.py	Fri Jul 23 18:58:35 2010
@@ -539,7 +539,7 @@
         This method is called from the subprocess, and by returning from this
         method we allow the subprocess to unblock.  After a bit the shell
         requests the subprocess to open the remote stack viewer which returns a
-        static object looking at the last exceptiopn.  It is queried through
+        static object looking at the last exception.  It is queried through
         the RPC mechanism.
 
         """

From python-checkins at python.org  Fri Jul 23 18:58:57 2010
From: python-checkins at python.org (brett.cannon)
Date: Fri, 23 Jul 2010 18:58:57 +0200 (CEST)
Subject: [Python-checkins] r83111 - python/branches/py3k/Misc/ACKS
Message-ID: <20100723165857.8E975EEA1B@mail.python.org>

Author: brett.cannon
Date: Fri Jul 23 18:58:57 2010
New Revision: 83111

Log:
Add Brian Brazil.


Modified:
   python/branches/py3k/Misc/ACKS

Modified: python/branches/py3k/Misc/ACKS
==============================================================================
--- python/branches/py3k/Misc/ACKS	(original)
+++ python/branches/py3k/Misc/ACKS	Fri Jul 23 18:58:57 2010
@@ -98,6 +98,7 @@
 Georg Brandl
 Christopher Brannon
 Terrence Brannon
+Brian Brazil
 Dave Brennan
 Tom Bridgman
 Richard Brodie

From python-checkins at python.org  Fri Jul 23 21:25:48 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Fri, 23 Jul 2010 21:25:48 +0200 (CEST)
Subject: [Python-checkins] r83112 - in python/branches/py3k: Lib/datetime.py
	Lib/test/datetimetester.py Lib/test/test_datetime.py
	Misc/NEWS Modules/Setup.dist Modules/_datetimemodule.c
	Modules/datetimemodule.c PC/config.c
	PCbuild/pythoncore.vcproj setup.py
Message-ID: <20100723192548.2D930EEA1B@mail.python.org>

Author: alexander.belopolsky
Date: Fri Jul 23 21:25:47 2010
New Revision: 83112

Log:
Issue #7989: Added pure python implementation of the datetime module.


Added:
   python/branches/py3k/Lib/datetime.py
      - copied, changed from r83064, /sandbox/branches/py3k-datetime/datetime.py
   python/branches/py3k/Lib/test/datetimetester.py
      - copied, changed from r83064, /python/branches/py3k/Lib/test/test_datetime.py
   python/branches/py3k/Modules/_datetimemodule.c
      - copied, changed from r83064, /python/branches/py3k/Modules/datetimemodule.c
Removed:
   python/branches/py3k/Modules/datetimemodule.c
Modified:
   python/branches/py3k/Lib/test/test_datetime.py
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/Modules/Setup.dist
   python/branches/py3k/PC/config.c
   python/branches/py3k/PCbuild/pythoncore.vcproj
   python/branches/py3k/setup.py

Copied: python/branches/py3k/Lib/datetime.py (from r83064, /sandbox/branches/py3k-datetime/datetime.py)
==============================================================================
--- /sandbox/branches/py3k-datetime/datetime.py	(original)
+++ python/branches/py3k/Lib/datetime.py	Fri Jul 23 21:25:47 2010
@@ -494,11 +494,11 @@
         return NotImplemented
 
     def __neg__(self):
-            # for CPython compatibility, we cannot use
-            # our __class__ here, but need a real timedelta
-            return timedelta(-self._days,
-                             -self._seconds,
-                             -self._microseconds)
+        # for CPython compatibility, we cannot use
+        # our __class__ here, but need a real timedelta
+        return timedelta(-self._days,
+                         -self._seconds,
+                         -self._microseconds)
 
     def __pos__(self):
         return self
@@ -2071,3 +2071,17 @@
 perverse time zone returns a negative dst()).  So a breaking case must be
 pretty bizarre, and a tzinfo subclass can override fromutc() if it is.
 """
+try:
+    from _datetime import *
+except ImportError:
+    pass
+else:
+    # Clean up unused names
+    del (_DAYNAMES, _DAYS_BEFORE_MONTH, _DAYS_IN_MONTH,
+         _DI100Y, _DI400Y, _DI4Y, _MAXORDINAL, _MONTHNAMES,
+         _build_struct_time, _call_tzinfo_method, _check_date_fields,
+         _check_time_fields, _check_tzinfo_arg, _check_tzname,
+         _check_utc_offset, _cmp, _cmperror, _date_class, _days_before_month,
+         _days_before_year, _days_in_month, _format_time, _is_leap,
+         _isoweek1monday, _math, _ord2ymd, _time, _time_class, _tzinfo_class,
+         _wrap_strftime, _ymd2ord)

Copied: python/branches/py3k/Lib/test/datetimetester.py (from r83064, /python/branches/py3k/Lib/test/test_datetime.py)
==============================================================================
--- /python/branches/py3k/Lib/test/test_datetime.py	(original)
+++ python/branches/py3k/Lib/test/datetimetester.py	Fri Jul 23 21:25:47 2010
@@ -11,6 +11,7 @@
 
 from test import support
 
+import datetime as datetime_module
 from datetime import MINYEAR, MAXYEAR
 from datetime import timedelta
 from datetime import tzinfo
@@ -19,6 +20,11 @@
 from datetime import date, datetime
 import time as _time
 
+# Needed by test_datetime
+import _strptime
+#
+
+
 pickle_choices = [(pickle, pickle, proto)
                   for proto in range(pickle.HIGHEST_PROTOCOL + 1)]
 assert len(pickle_choices) == pickle.HIGHEST_PROTOCOL + 1
@@ -44,7 +50,7 @@
 class TestModule(unittest.TestCase):
 
     def test_constants(self):
-        import datetime
+        datetime = datetime_module
         self.assertEqual(datetime.MINYEAR, 1)
         self.assertEqual(datetime.MAXYEAR, 9999)
 
@@ -154,7 +160,7 @@
             self.assertEqual(str(tz), tz.tzname(None))
 
     def test_repr(self):
-        import datetime
+        datetime = datetime_module
         for tz in [self.ACDT, self.EST, timezone.utc,
                    timezone.min, timezone.max]:
             # test round-trip

Modified: python/branches/py3k/Lib/test/test_datetime.py
==============================================================================
--- python/branches/py3k/Lib/test/test_datetime.py	(original)
+++ python/branches/py3k/Lib/test/test_datetime.py	Fri Jul 23 21:25:47 2010
@@ -1,3668 +1,42 @@
-"""Test date/time type.
-
-See http://www.zope.org/Members/fdrake/DateTimeWiki/TestCases
-"""
-
-import sys
-import pickle
 import unittest
-
-from operator import lt, le, gt, ge, eq, ne, truediv, floordiv, mod
-
-from test import support
-
-from datetime import MINYEAR, MAXYEAR
-from datetime import timedelta
-from datetime import tzinfo
-from datetime import time
-from datetime import timezone
-from datetime import date, datetime
-import time as _time
-
-pickle_choices = [(pickle, pickle, proto)
-                  for proto in range(pickle.HIGHEST_PROTOCOL + 1)]
-assert len(pickle_choices) == pickle.HIGHEST_PROTOCOL + 1
-
-# An arbitrary collection of objects of non-datetime types, for testing
-# mixed-type comparisons.
-OTHERSTUFF = (10, 34.5, "abc", {}, [], ())
-
-
-# XXX Copied from test_float.
-INF = float("inf")
-NAN = float("nan")
-
-# decorator for skipping tests on non-IEEE 754 platforms
-requires_IEEE_754 = unittest.skipUnless(
-    float.__getformat__("double").startswith("IEEE"),
-    "test requires IEEE 754 doubles")
-
-
-#############################################################################
-# module tests
-
-class TestModule(unittest.TestCase):
-
-    def test_constants(self):
-        import datetime
-        self.assertEqual(datetime.MINYEAR, 1)
-        self.assertEqual(datetime.MAXYEAR, 9999)
-
-#############################################################################
-# tzinfo tests
-
-class FixedOffset(tzinfo):
-
-    def __init__(self, offset, name, dstoffset=42):
-        if isinstance(offset, int):
-            offset = timedelta(minutes=offset)
-        if isinstance(dstoffset, int):
-            dstoffset = timedelta(minutes=dstoffset)
-        self.__offset = offset
-        self.__name = name
-        self.__dstoffset = dstoffset
-    def __repr__(self):
-        return self.__name.lower()
-    def utcoffset(self, dt):
-        return self.__offset
-    def tzname(self, dt):
-        return self.__name
-    def dst(self, dt):
-        return self.__dstoffset
-
-class PicklableFixedOffset(FixedOffset):
-
-    def __init__(self, offset=None, name=None, dstoffset=None):
-        FixedOffset.__init__(self, offset, name, dstoffset)
-
-class TestTZInfo(unittest.TestCase):
-
-    def test_non_abstractness(self):
-        # In order to allow subclasses to get pickled, the C implementation
-        # wasn't able to get away with having __init__ raise
-        # NotImplementedError.
-        useless = tzinfo()
-        dt = datetime.max
-        self.assertRaises(NotImplementedError, useless.tzname, dt)
-        self.assertRaises(NotImplementedError, useless.utcoffset, dt)
-        self.assertRaises(NotImplementedError, useless.dst, dt)
-
-    def test_subclass_must_override(self):
-        class NotEnough(tzinfo):
-            def __init__(self, offset, name):
-                self.__offset = offset
-                self.__name = name
-        self.assertTrue(issubclass(NotEnough, tzinfo))
-        ne = NotEnough(3, "NotByALongShot")
-        self.assertIsInstance(ne, tzinfo)
-
-        dt = datetime.now()
-        self.assertRaises(NotImplementedError, ne.tzname, dt)
-        self.assertRaises(NotImplementedError, ne.utcoffset, dt)
-        self.assertRaises(NotImplementedError, ne.dst, dt)
-
-    def test_normal(self):
-        fo = FixedOffset(3, "Three")
-        self.assertIsInstance(fo, tzinfo)
-        for dt in datetime.now(), None:
-            self.assertEqual(fo.utcoffset(dt), timedelta(minutes=3))
-            self.assertEqual(fo.tzname(dt), "Three")
-            self.assertEqual(fo.dst(dt), timedelta(minutes=42))
-
-    def test_pickling_base(self):
-        # There's no point to pickling tzinfo objects on their own (they
-        # carry no data), but they need to be picklable anyway else
-        # concrete subclasses can't be pickled.
-        orig = tzinfo.__new__(tzinfo)
-        self.assertTrue(type(orig) is tzinfo)
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertTrue(type(derived) is tzinfo)
-
-    def test_pickling_subclass(self):
-        # Make sure we can pickle/unpickle an instance of a subclass.
-        offset = timedelta(minutes=-300)
-        for otype, args in [
-            (PicklableFixedOffset, (offset, 'cookie')),
-            (timezone, (offset,)),
-            (timezone, (offset, "EST"))]:
-            orig = otype(*args)
-            oname = orig.tzname(None)
-            self.assertIsInstance(orig, tzinfo)
-            self.assertIs(type(orig), otype)
-            self.assertEqual(orig.utcoffset(None), offset)
-            self.assertEqual(orig.tzname(None), oname)
-            for pickler, unpickler, proto in pickle_choices:
-                green = pickler.dumps(orig, proto)
-                derived = unpickler.loads(green)
-                self.assertIsInstance(derived, tzinfo)
-                self.assertIs(type(derived), otype)
-                self.assertEqual(derived.utcoffset(None), offset)
-                self.assertEqual(derived.tzname(None), oname)
-
-class TestTimeZone(unittest.TestCase):
-
-    def setUp(self):
-        self.ACDT = timezone(timedelta(hours=9.5), 'ACDT')
-        self.EST = timezone(-timedelta(hours=5), 'EST')
-        self.DT = datetime(2010, 1, 1)
-
-    def test_str(self):
-        for tz in [self.ACDT, self.EST, timezone.utc,
-                   timezone.min, timezone.max]:
-            self.assertEqual(str(tz), tz.tzname(None))
-
-    def test_repr(self):
-        import datetime
-        for tz in [self.ACDT, self.EST, timezone.utc,
-                   timezone.min, timezone.max]:
-            # test round-trip
-            tzrep = repr(tz)
-            self.assertEqual(tz, eval(tzrep))
-
-
-    def test_class_members(self):
-        limit = timedelta(hours=23, minutes=59)
-        self.assertEqual(timezone.utc.utcoffset(None), ZERO)
-        self.assertEqual(timezone.min.utcoffset(None), -limit)
-        self.assertEqual(timezone.max.utcoffset(None), limit)
-
-
-    def test_constructor(self):
-        self.assertEqual(timezone.utc, timezone(timedelta(0)))
-        # invalid offsets
-        for invalid in [timedelta(microseconds=1), timedelta(1, 1),
-                        timedelta(seconds=1), timedelta(1), -timedelta(1)]:
-            self.assertRaises(ValueError, timezone, invalid)
-            self.assertRaises(ValueError, timezone, -invalid)
-
-        with self.assertRaises(TypeError): timezone(None)
-        with self.assertRaises(TypeError): timezone(42)
-        with self.assertRaises(TypeError): timezone(ZERO, None)
-        with self.assertRaises(TypeError): timezone(ZERO, 42)
-        with self.assertRaises(TypeError): timezone(ZERO, 'ABC', 'extra')
-
-    def test_inheritance(self):
-        self.assertIsInstance(timezone.utc, tzinfo)
-        self.assertIsInstance(self.EST, tzinfo)
-
-    def test_utcoffset(self):
-        dummy = self.DT
-        for h in [0, 1.5, 12]:
-            offset = h * HOUR
-            self.assertEqual(offset, timezone(offset).utcoffset(dummy))
-            self.assertEqual(-offset, timezone(-offset).utcoffset(dummy))
-
-        with self.assertRaises(TypeError): self.EST.utcoffset('')
-        with self.assertRaises(TypeError): self.EST.utcoffset(5)
-
-
-    def test_dst(self):
-        self.assertEqual(None, timezone.utc.dst(self.DT))
-
-        with self.assertRaises(TypeError): self.EST.dst('')
-        with self.assertRaises(TypeError): self.EST.dst(5)
-
-    def test_tzname(self):
-        self.assertEqual('UTC+00:00', timezone(ZERO).tzname(None))
-        self.assertEqual('UTC-05:00', timezone(-5 * HOUR).tzname(None))
-        self.assertEqual('UTC+09:30', timezone(9.5 * HOUR).tzname(None))
-        self.assertEqual('UTC-00:01', timezone(timedelta(minutes=-1)).tzname(None))
-        self.assertEqual('XYZ', timezone(-5 * HOUR, 'XYZ').tzname(None))
-
-        with self.assertRaises(TypeError): self.EST.tzname('')
-        with self.assertRaises(TypeError): self.EST.tzname(5)
-
-    def test_fromutc(self):
-        with self.assertRaises(ValueError):
-            timezone.utc.fromutc(self.DT)
-        with self.assertRaises(TypeError):
-            timezone.utc.fromutc('not datetime')
-        for tz in [self.EST, self.ACDT, Eastern]:
-            utctime = self.DT.replace(tzinfo=tz)
-            local = tz.fromutc(utctime)
-            self.assertEqual(local - utctime, tz.utcoffset(local))
-            self.assertEqual(local,
-                             self.DT.replace(tzinfo=timezone.utc))
-
-    def test_comparison(self):
-        self.assertNotEqual(timezone(ZERO), timezone(HOUR))
-        self.assertEqual(timezone(HOUR), timezone(HOUR))
-        self.assertEqual(timezone(-5 * HOUR), timezone(-5 * HOUR, 'EST'))
-        with self.assertRaises(TypeError): timezone(ZERO) < timezone(ZERO)
-        self.assertIn(timezone(ZERO), {timezone(ZERO)})
-
-    def test_aware_datetime(self):
-        # test that timezone instances can be used by datetime
-        t = datetime(1, 1, 1)
-        for tz in [timezone.min, timezone.max, timezone.utc]:
-            self.assertEqual(tz.tzname(t),
-                             t.replace(tzinfo=tz).tzname())
-            self.assertEqual(tz.utcoffset(t),
-                             t.replace(tzinfo=tz).utcoffset())
-            self.assertEqual(tz.dst(t),
-                             t.replace(tzinfo=tz).dst())
-
-#############################################################################
-# Base clase for testing a particular aspect of timedelta, time, date and
-# datetime comparisons.
-
-class HarmlessMixedComparison:
-    # Test that __eq__ and __ne__ don't complain for mixed-type comparisons.
-
-    # Subclasses must define 'theclass', and theclass(1, 1, 1) must be a
-    # legit constructor.
-
-    def test_harmless_mixed_comparison(self):
-        me = self.theclass(1, 1, 1)
-
-        self.assertFalse(me == ())
-        self.assertTrue(me != ())
-        self.assertFalse(() == me)
-        self.assertTrue(() != me)
-
-        self.assertIn(me, [1, 20, [], me])
-        self.assertIn([], [me, 1, 20, []])
-
-    def test_harmful_mixed_comparison(self):
-        me = self.theclass(1, 1, 1)
-
-        self.assertRaises(TypeError, lambda: me < ())
-        self.assertRaises(TypeError, lambda: me <= ())
-        self.assertRaises(TypeError, lambda: me > ())
-        self.assertRaises(TypeError, lambda: me >= ())
-
-        self.assertRaises(TypeError, lambda: () < me)
-        self.assertRaises(TypeError, lambda: () <= me)
-        self.assertRaises(TypeError, lambda: () > me)
-        self.assertRaises(TypeError, lambda: () >= me)
-
-#############################################################################
-# timedelta tests
-
-class TestTimeDelta(HarmlessMixedComparison, unittest.TestCase):
-
-    theclass = timedelta
-
-    def test_constructor(self):
-        eq = self.assertEqual
-        td = timedelta
-
-        # Check keyword args to constructor
-        eq(td(), td(weeks=0, days=0, hours=0, minutes=0, seconds=0,
-                    milliseconds=0, microseconds=0))
-        eq(td(1), td(days=1))
-        eq(td(0, 1), td(seconds=1))
-        eq(td(0, 0, 1), td(microseconds=1))
-        eq(td(weeks=1), td(days=7))
-        eq(td(days=1), td(hours=24))
-        eq(td(hours=1), td(minutes=60))
-        eq(td(minutes=1), td(seconds=60))
-        eq(td(seconds=1), td(milliseconds=1000))
-        eq(td(milliseconds=1), td(microseconds=1000))
-
-        # Check float args to constructor
-        eq(td(weeks=1.0/7), td(days=1))
-        eq(td(days=1.0/24), td(hours=1))
-        eq(td(hours=1.0/60), td(minutes=1))
-        eq(td(minutes=1.0/60), td(seconds=1))
-        eq(td(seconds=0.001), td(milliseconds=1))
-        eq(td(milliseconds=0.001), td(microseconds=1))
-
-    def test_computations(self):
-        eq = self.assertEqual
-        td = timedelta
-
-        a = td(7) # One week
-        b = td(0, 60) # One minute
-        c = td(0, 0, 1000) # One millisecond
-        eq(a+b+c, td(7, 60, 1000))
-        eq(a-b, td(6, 24*3600 - 60))
-        eq(b.__rsub__(a), td(6, 24*3600 - 60))
-        eq(-a, td(-7))
-        eq(+a, td(7))
-        eq(-b, td(-1, 24*3600 - 60))
-        eq(-c, td(-1, 24*3600 - 1, 999000))
-        eq(abs(a), a)
-        eq(abs(-a), a)
-        eq(td(6, 24*3600), a)
-        eq(td(0, 0, 60*1000000), b)
-        eq(a*10, td(70))
-        eq(a*10, 10*a)
-        eq(a*10, 10*a)
-        eq(b*10, td(0, 600))
-        eq(10*b, td(0, 600))
-        eq(b*10, td(0, 600))
-        eq(c*10, td(0, 0, 10000))
-        eq(10*c, td(0, 0, 10000))
-        eq(c*10, td(0, 0, 10000))
-        eq(a*-1, -a)
-        eq(b*-2, -b-b)
-        eq(c*-2, -c+-c)
-        eq(b*(60*24), (b*60)*24)
-        eq(b*(60*24), (60*b)*24)
-        eq(c*1000, td(0, 1))
-        eq(1000*c, td(0, 1))
-        eq(a//7, td(1))
-        eq(b//10, td(0, 6))
-        eq(c//1000, td(0, 0, 1))
-        eq(a//10, td(0, 7*24*360))
-        eq(a//3600000, td(0, 0, 7*24*1000))
-        eq(a/0.5, td(14))
-        eq(b/0.5, td(0, 120))
-        eq(a/7, td(1))
-        eq(b/10, td(0, 6))
-        eq(c/1000, td(0, 0, 1))
-        eq(a/10, td(0, 7*24*360))
-        eq(a/3600000, td(0, 0, 7*24*1000))
-
-        # Multiplication by float
-        us = td(microseconds=1)
-        eq((3*us) * 0.5, 2*us)
-        eq((5*us) * 0.5, 2*us)
-        eq(0.5 * (3*us), 2*us)
-        eq(0.5 * (5*us), 2*us)
-        eq((-3*us) * 0.5, -2*us)
-        eq((-5*us) * 0.5, -2*us)
-
-        # Division by int and float
-        eq((3*us) / 2, 2*us)
-        eq((5*us) / 2, 2*us)
-        eq((-3*us) / 2.0, -2*us)
-        eq((-5*us) / 2.0, -2*us)
-        eq((3*us) / -2, -2*us)
-        eq((5*us) / -2, -2*us)
-        eq((3*us) / -2.0, -2*us)
-        eq((5*us) / -2.0, -2*us)
-        for i in range(-10, 10):
-            eq((i*us/3)//us, round(i/3))
-        for i in range(-10, 10):
-            eq((i*us/-3)//us, round(i/-3))
-
-    def test_disallowed_computations(self):
-        a = timedelta(42)
-
-        # Add/sub ints or floats should be illegal
-        for i in 1, 1.0:
-            self.assertRaises(TypeError, lambda: a+i)
-            self.assertRaises(TypeError, lambda: a-i)
-            self.assertRaises(TypeError, lambda: i+a)
-            self.assertRaises(TypeError, lambda: i-a)
-
-        # Division of int by timedelta doesn't make sense.
-        # Division by zero doesn't make sense.
-        zero = 0
-        self.assertRaises(TypeError, lambda: zero // a)
-        self.assertRaises(ZeroDivisionError, lambda: a // zero)
-        self.assertRaises(ZeroDivisionError, lambda: a / zero)
-        self.assertRaises(ZeroDivisionError, lambda: a / 0.0)
-        self.assertRaises(TypeError, lambda: a / '')
-
-    @requires_IEEE_754
-    def test_disallowed_special(self):
-        a = timedelta(42)
-        self.assertRaises(ValueError, a.__mul__, NAN)
-        self.assertRaises(ValueError, a.__truediv__, NAN)
-
-    def test_basic_attributes(self):
-        days, seconds, us = 1, 7, 31
-        td = timedelta(days, seconds, us)
-        self.assertEqual(td.days, days)
-        self.assertEqual(td.seconds, seconds)
-        self.assertEqual(td.microseconds, us)
-
-    def test_total_seconds(self):
-        td = timedelta(days=365)
-        self.assertEqual(td.total_seconds(), 31536000.0)
-        for total_seconds in [123456.789012, -123456.789012, 0.123456, 0, 1e6]:
-            td = timedelta(seconds=total_seconds)
-            self.assertEqual(td.total_seconds(), total_seconds)
-        # Issue8644: Test that td.total_seconds() has the same
-        # accuracy as td / timedelta(seconds=1).
-        for ms in [-1, -2, -123]:
-            td = timedelta(microseconds=ms)
-            self.assertEqual(td.total_seconds(), td / timedelta(seconds=1))
-
-    def test_carries(self):
-        t1 = timedelta(days=100,
-                       weeks=-7,
-                       hours=-24*(100-49),
-                       minutes=-3,
-                       seconds=12,
-                       microseconds=(3*60 - 12) * 1e6 + 1)
-        t2 = timedelta(microseconds=1)
-        self.assertEqual(t1, t2)
-
-    def test_hash_equality(self):
-        t1 = timedelta(days=100,
-                       weeks=-7,
-                       hours=-24*(100-49),
-                       minutes=-3,
-                       seconds=12,
-                       microseconds=(3*60 - 12) * 1000000)
-        t2 = timedelta()
-        self.assertEqual(hash(t1), hash(t2))
-
-        t1 += timedelta(weeks=7)
-        t2 += timedelta(days=7*7)
-        self.assertEqual(t1, t2)
-        self.assertEqual(hash(t1), hash(t2))
-
-        d = {t1: 1}
-        d[t2] = 2
-        self.assertEqual(len(d), 1)
-        self.assertEqual(d[t1], 2)
-
-    def test_pickling(self):
-        args = 12, 34, 56
-        orig = timedelta(*args)
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertEqual(orig, derived)
-
-    def test_compare(self):
-        t1 = timedelta(2, 3, 4)
-        t2 = timedelta(2, 3, 4)
-        self.assertEqual(t1, t2)
-        self.assertTrue(t1 <= t2)
-        self.assertTrue(t1 >= t2)
-        self.assertTrue(not t1 != t2)
-        self.assertTrue(not t1 < t2)
-        self.assertTrue(not t1 > t2)
-
-        for args in (3, 3, 3), (2, 4, 4), (2, 3, 5):
-            t2 = timedelta(*args)   # this is larger than t1
-            self.assertTrue(t1 < t2)
-            self.assertTrue(t2 > t1)
-            self.assertTrue(t1 <= t2)
-            self.assertTrue(t2 >= t1)
-            self.assertTrue(t1 != t2)
-            self.assertTrue(t2 != t1)
-            self.assertTrue(not t1 == t2)
-            self.assertTrue(not t2 == t1)
-            self.assertTrue(not t1 > t2)
-            self.assertTrue(not t2 < t1)
-            self.assertTrue(not t1 >= t2)
-            self.assertTrue(not t2 <= t1)
-
-        for badarg in OTHERSTUFF:
-            self.assertEqual(t1 == badarg, False)
-            self.assertEqual(t1 != badarg, True)
-            self.assertEqual(badarg == t1, False)
-            self.assertEqual(badarg != t1, True)
-
-            self.assertRaises(TypeError, lambda: t1 <= badarg)
-            self.assertRaises(TypeError, lambda: t1 < badarg)
-            self.assertRaises(TypeError, lambda: t1 > badarg)
-            self.assertRaises(TypeError, lambda: t1 >= badarg)
-            self.assertRaises(TypeError, lambda: badarg <= t1)
-            self.assertRaises(TypeError, lambda: badarg < t1)
-            self.assertRaises(TypeError, lambda: badarg > t1)
-            self.assertRaises(TypeError, lambda: badarg >= t1)
-
-    def test_str(self):
-        td = timedelta
-        eq = self.assertEqual
-
-        eq(str(td(1)), "1 day, 0:00:00")
-        eq(str(td(-1)), "-1 day, 0:00:00")
-        eq(str(td(2)), "2 days, 0:00:00")
-        eq(str(td(-2)), "-2 days, 0:00:00")
-
-        eq(str(td(hours=12, minutes=58, seconds=59)), "12:58:59")
-        eq(str(td(hours=2, minutes=3, seconds=4)), "2:03:04")
-        eq(str(td(weeks=-30, hours=23, minutes=12, seconds=34)),
-           "-210 days, 23:12:34")
-
-        eq(str(td(milliseconds=1)), "0:00:00.001000")
-        eq(str(td(microseconds=3)), "0:00:00.000003")
-
-        eq(str(td(days=999999999, hours=23, minutes=59, seconds=59,
-                   microseconds=999999)),
-           "999999999 days, 23:59:59.999999")
-
-    def test_repr(self):
-        name = 'datetime.' + self.theclass.__name__
-        self.assertEqual(repr(self.theclass(1)),
-                         "%s(1)" % name)
-        self.assertEqual(repr(self.theclass(10, 2)),
-                         "%s(10, 2)" % name)
-        self.assertEqual(repr(self.theclass(-10, 2, 400000)),
-                         "%s(-10, 2, 400000)" % name)
-
-    def test_roundtrip(self):
-        for td in (timedelta(days=999999999, hours=23, minutes=59,
-                             seconds=59, microseconds=999999),
-                   timedelta(days=-999999999),
-                   timedelta(days=-999999999, seconds=1),
-                   timedelta(days=1, seconds=2, microseconds=3)):
-
-            # Verify td -> string -> td identity.
-            s = repr(td)
-            self.assertTrue(s.startswith('datetime.'))
-            s = s[9:]
-            td2 = eval(s)
-            self.assertEqual(td, td2)
-
-            # Verify identity via reconstructing from pieces.
-            td2 = timedelta(td.days, td.seconds, td.microseconds)
-            self.assertEqual(td, td2)
-
-    def test_resolution_info(self):
-        self.assertIsInstance(timedelta.min, timedelta)
-        self.assertIsInstance(timedelta.max, timedelta)
-        self.assertIsInstance(timedelta.resolution, timedelta)
-        self.assertTrue(timedelta.max > timedelta.min)
-        self.assertEqual(timedelta.min, timedelta(-999999999))
-        self.assertEqual(timedelta.max, timedelta(999999999, 24*3600-1, 1e6-1))
-        self.assertEqual(timedelta.resolution, timedelta(0, 0, 1))
-
-    def test_overflow(self):
-        tiny = timedelta.resolution
-
-        td = timedelta.min + tiny
-        td -= tiny  # no problem
-        self.assertRaises(OverflowError, td.__sub__, tiny)
-        self.assertRaises(OverflowError, td.__add__, -tiny)
-
-        td = timedelta.max - tiny
-        td += tiny  # no problem
-        self.assertRaises(OverflowError, td.__add__, tiny)
-        self.assertRaises(OverflowError, td.__sub__, -tiny)
-
-        self.assertRaises(OverflowError, lambda: -timedelta.max)
-
-        day = timedelta(1)
-        self.assertRaises(OverflowError, day.__mul__, 10**9)
-        self.assertRaises(OverflowError, day.__mul__, 1e9)
-        self.assertRaises(OverflowError, day.__truediv__, 1e-20)
-        self.assertRaises(OverflowError, day.__truediv__, 1e-10)
-        self.assertRaises(OverflowError, day.__truediv__, 9e-10)
-
-    @requires_IEEE_754
-    def _test_overflow_special(self):
-        day = timedelta(1)
-        self.assertRaises(OverflowError, day.__mul__, INF)
-        self.assertRaises(OverflowError, day.__mul__, -INF)
-
-    def test_microsecond_rounding(self):
-        td = timedelta
-        eq = self.assertEqual
-
-        # Single-field rounding.
-        eq(td(milliseconds=0.4/1000), td(0))    # rounds to 0
-        eq(td(milliseconds=-0.4/1000), td(0))    # rounds to 0
-        eq(td(milliseconds=0.6/1000), td(microseconds=1))
-        eq(td(milliseconds=-0.6/1000), td(microseconds=-1))
-
-        # Rounding due to contributions from more than one field.
-        us_per_hour = 3600e6
-        us_per_day = us_per_hour * 24
-        eq(td(days=.4/us_per_day), td(0))
-        eq(td(hours=.2/us_per_hour), td(0))
-        eq(td(days=.4/us_per_day, hours=.2/us_per_hour), td(microseconds=1))
-
-        eq(td(days=-.4/us_per_day), td(0))
-        eq(td(hours=-.2/us_per_hour), td(0))
-        eq(td(days=-.4/us_per_day, hours=-.2/us_per_hour), td(microseconds=-1))
-
-    def test_massive_normalization(self):
-        td = timedelta(microseconds=-1)
-        self.assertEqual((td.days, td.seconds, td.microseconds),
-                         (-1, 24*3600-1, 999999))
-
-    def test_bool(self):
-        self.assertTrue(timedelta(1))
-        self.assertTrue(timedelta(0, 1))
-        self.assertTrue(timedelta(0, 0, 1))
-        self.assertTrue(timedelta(microseconds=1))
-        self.assertTrue(not timedelta(0))
-
-    def test_subclass_timedelta(self):
-
-        class T(timedelta):
-            @staticmethod
-            def from_td(td):
-                return T(td.days, td.seconds, td.microseconds)
-
-            def as_hours(self):
-                sum = (self.days * 24 +
-                       self.seconds / 3600.0 +
-                       self.microseconds / 3600e6)
-                return round(sum)
-
-        t1 = T(days=1)
-        self.assertTrue(type(t1) is T)
-        self.assertEqual(t1.as_hours(), 24)
-
-        t2 = T(days=-1, seconds=-3600)
-        self.assertTrue(type(t2) is T)
-        self.assertEqual(t2.as_hours(), -25)
-
-        t3 = t1 + t2
-        self.assertTrue(type(t3) is timedelta)
-        t4 = T.from_td(t3)
-        self.assertTrue(type(t4) is T)
-        self.assertEqual(t3.days, t4.days)
-        self.assertEqual(t3.seconds, t4.seconds)
-        self.assertEqual(t3.microseconds, t4.microseconds)
-        self.assertEqual(str(t3), str(t4))
-        self.assertEqual(t4.as_hours(), -1)
-
-    def test_division(self):
-        t = timedelta(hours=1, minutes=24, seconds=19)
-        second = timedelta(seconds=1)
-        self.assertEqual(t / second, 5059.0)
-        self.assertEqual(t // second, 5059)
-
-        t = timedelta(minutes=2, seconds=30)
-        minute = timedelta(minutes=1)
-        self.assertEqual(t / minute, 2.5)
-        self.assertEqual(t // minute, 2)
-
-        zerotd = timedelta(0)
-        self.assertRaises(ZeroDivisionError, truediv, t, zerotd)
-        self.assertRaises(ZeroDivisionError, floordiv, t, zerotd)
-
-        # self.assertRaises(TypeError, truediv, t, 2)
-        # note: floor division of a timedelta by an integer *is*
-        # currently permitted.
-
-    def test_remainder(self):
-        t = timedelta(minutes=2, seconds=30)
-        minute = timedelta(minutes=1)
-        r = t % minute
-        self.assertEqual(r, timedelta(seconds=30))
-
-        t = timedelta(minutes=-2, seconds=30)
-        r = t %  minute
-        self.assertEqual(r, timedelta(seconds=30))
-
-        zerotd = timedelta(0)
-        self.assertRaises(ZeroDivisionError, mod, t, zerotd)
-
-        self.assertRaises(TypeError, mod, t, 10)
-
-    def test_divmod(self):
-        t = timedelta(minutes=2, seconds=30)
-        minute = timedelta(minutes=1)
-        q, r = divmod(t, minute)
-        self.assertEqual(q, 2)
-        self.assertEqual(r, timedelta(seconds=30))
-
-        t = timedelta(minutes=-2, seconds=30)
-        q, r = divmod(t, minute)
-        self.assertEqual(q, -2)
-        self.assertEqual(r, timedelta(seconds=30))
-
-        zerotd = timedelta(0)
-        self.assertRaises(ZeroDivisionError, divmod, t, zerotd)
-
-        self.assertRaises(TypeError, divmod, t, 10)
-
-
-#############################################################################
-# date tests
-
-class TestDateOnly(unittest.TestCase):
-    # Tests here won't pass if also run on datetime objects, so don't
-    # subclass this to test datetimes too.
-
-    def test_delta_non_days_ignored(self):
-        dt = date(2000, 1, 2)
-        delta = timedelta(days=1, hours=2, minutes=3, seconds=4,
-                          microseconds=5)
-        days = timedelta(delta.days)
-        self.assertEqual(days, timedelta(1))
-
-        dt2 = dt + delta
-        self.assertEqual(dt2, dt + days)
-
-        dt2 = delta + dt
-        self.assertEqual(dt2, dt + days)
-
-        dt2 = dt - delta
-        self.assertEqual(dt2, dt - days)
-
-        delta = -delta
-        days = timedelta(delta.days)
-        self.assertEqual(days, timedelta(-2))
-
-        dt2 = dt + delta
-        self.assertEqual(dt2, dt + days)
-
-        dt2 = delta + dt
-        self.assertEqual(dt2, dt + days)
-
-        dt2 = dt - delta
-        self.assertEqual(dt2, dt - days)
-
-class SubclassDate(date):
-    sub_var = 1
-
-class TestDate(HarmlessMixedComparison, unittest.TestCase):
-    # Tests here should pass for both dates and datetimes, except for a
-    # few tests that TestDateTime overrides.
-
-    theclass = date
-
-    def test_basic_attributes(self):
-        dt = self.theclass(2002, 3, 1)
-        self.assertEqual(dt.year, 2002)
-        self.assertEqual(dt.month, 3)
-        self.assertEqual(dt.day, 1)
-
-    def test_roundtrip(self):
-        for dt in (self.theclass(1, 2, 3),
-                   self.theclass.today()):
-            # Verify dt -> string -> date identity.
-            s = repr(dt)
-            self.assertTrue(s.startswith('datetime.'))
-            s = s[9:]
-            dt2 = eval(s)
-            self.assertEqual(dt, dt2)
-
-            # Verify identity via reconstructing from pieces.
-            dt2 = self.theclass(dt.year, dt.month, dt.day)
-            self.assertEqual(dt, dt2)
-
-    def test_ordinal_conversions(self):
-        # Check some fixed values.
-        for y, m, d, n in [(1, 1, 1, 1),      # calendar origin
-                           (1, 12, 31, 365),
-                           (2, 1, 1, 366),
-                           # first example from "Calendrical Calculations"
-                           (1945, 11, 12, 710347)]:
-            d = self.theclass(y, m, d)
-            self.assertEqual(n, d.toordinal())
-            fromord = self.theclass.fromordinal(n)
-            self.assertEqual(d, fromord)
-            if hasattr(fromord, "hour"):
-            # if we're checking something fancier than a date, verify
-            # the extra fields have been zeroed out
-                self.assertEqual(fromord.hour, 0)
-                self.assertEqual(fromord.minute, 0)
-                self.assertEqual(fromord.second, 0)
-                self.assertEqual(fromord.microsecond, 0)
-
-        # Check first and last days of year spottily across the whole
-        # range of years supported.
-        for year in range(MINYEAR, MAXYEAR+1, 7):
-            # Verify (year, 1, 1) -> ordinal -> y, m, d is identity.
-            d = self.theclass(year, 1, 1)
-            n = d.toordinal()
-            d2 = self.theclass.fromordinal(n)
-            self.assertEqual(d, d2)
-            # Verify that moving back a day gets to the end of year-1.
-            if year > 1:
-                d = self.theclass.fromordinal(n-1)
-                d2 = self.theclass(year-1, 12, 31)
-                self.assertEqual(d, d2)
-                self.assertEqual(d2.toordinal(), n-1)
-
-        # Test every day in a leap-year and a non-leap year.
-        dim = [31, 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31]
-        for year, isleap in (2000, True), (2002, False):
-            n = self.theclass(year, 1, 1).toordinal()
-            for month, maxday in zip(range(1, 13), dim):
-                if month == 2 and isleap:
-                    maxday += 1
-                for day in range(1, maxday+1):
-                    d = self.theclass(year, month, day)
-                    self.assertEqual(d.toordinal(), n)
-                    self.assertEqual(d, self.theclass.fromordinal(n))
-                    n += 1
-
-    def test_extreme_ordinals(self):
-        a = self.theclass.min
-        a = self.theclass(a.year, a.month, a.day)  # get rid of time parts
-        aord = a.toordinal()
-        b = a.fromordinal(aord)
-        self.assertEqual(a, b)
-
-        self.assertRaises(ValueError, lambda: a.fromordinal(aord - 1))
-
-        b = a + timedelta(days=1)
-        self.assertEqual(b.toordinal(), aord + 1)
-        self.assertEqual(b, self.theclass.fromordinal(aord + 1))
-
-        a = self.theclass.max
-        a = self.theclass(a.year, a.month, a.day)  # get rid of time parts
-        aord = a.toordinal()
-        b = a.fromordinal(aord)
-        self.assertEqual(a, b)
-
-        self.assertRaises(ValueError, lambda: a.fromordinal(aord + 1))
-
-        b = a - timedelta(days=1)
-        self.assertEqual(b.toordinal(), aord - 1)
-        self.assertEqual(b, self.theclass.fromordinal(aord - 1))
-
-    def test_bad_constructor_arguments(self):
-        # bad years
-        self.theclass(MINYEAR, 1, 1)  # no exception
-        self.theclass(MAXYEAR, 1, 1)  # no exception
-        self.assertRaises(ValueError, self.theclass, MINYEAR-1, 1, 1)
-        self.assertRaises(ValueError, self.theclass, MAXYEAR+1, 1, 1)
-        # bad months
-        self.theclass(2000, 1, 1)    # no exception
-        self.theclass(2000, 12, 1)   # no exception
-        self.assertRaises(ValueError, self.theclass, 2000, 0, 1)
-        self.assertRaises(ValueError, self.theclass, 2000, 13, 1)
-        # bad days
-        self.theclass(2000, 2, 29)   # no exception
-        self.theclass(2004, 2, 29)   # no exception
-        self.theclass(2400, 2, 29)   # no exception
-        self.assertRaises(ValueError, self.theclass, 2000, 2, 30)
-        self.assertRaises(ValueError, self.theclass, 2001, 2, 29)
-        self.assertRaises(ValueError, self.theclass, 2100, 2, 29)
-        self.assertRaises(ValueError, self.theclass, 1900, 2, 29)
-        self.assertRaises(ValueError, self.theclass, 2000, 1, 0)
-        self.assertRaises(ValueError, self.theclass, 2000, 1, 32)
-
-    def test_hash_equality(self):
-        d = self.theclass(2000, 12, 31)
-        # same thing
-        e = self.theclass(2000, 12, 31)
-        self.assertEqual(d, e)
-        self.assertEqual(hash(d), hash(e))
-
-        dic = {d: 1}
-        dic[e] = 2
-        self.assertEqual(len(dic), 1)
-        self.assertEqual(dic[d], 2)
-        self.assertEqual(dic[e], 2)
-
-        d = self.theclass(2001,  1,  1)
-        # same thing
-        e = self.theclass(2001,  1,  1)
-        self.assertEqual(d, e)
-        self.assertEqual(hash(d), hash(e))
-
-        dic = {d: 1}
-        dic[e] = 2
-        self.assertEqual(len(dic), 1)
-        self.assertEqual(dic[d], 2)
-        self.assertEqual(dic[e], 2)
-
-    def test_computations(self):
-        a = self.theclass(2002, 1, 31)
-        b = self.theclass(1956, 1, 31)
-        c = self.theclass(2001,2,1)
-
-        diff = a-b
-        self.assertEqual(diff.days, 46*365 + len(range(1956, 2002, 4)))
-        self.assertEqual(diff.seconds, 0)
-        self.assertEqual(diff.microseconds, 0)
-
-        day = timedelta(1)
-        week = timedelta(7)
-        a = self.theclass(2002, 3, 2)
-        self.assertEqual(a + day, self.theclass(2002, 3, 3))
-        self.assertEqual(day + a, self.theclass(2002, 3, 3))
-        self.assertEqual(a - day, self.theclass(2002, 3, 1))
-        self.assertEqual(-day + a, self.theclass(2002, 3, 1))
-        self.assertEqual(a + week, self.theclass(2002, 3, 9))
-        self.assertEqual(a - week, self.theclass(2002, 2, 23))
-        self.assertEqual(a + 52*week, self.theclass(2003, 3, 1))
-        self.assertEqual(a - 52*week, self.theclass(2001, 3, 3))
-        self.assertEqual((a + week) - a, week)
-        self.assertEqual((a + day) - a, day)
-        self.assertEqual((a - week) - a, -week)
-        self.assertEqual((a - day) - a, -day)
-        self.assertEqual(a - (a + week), -week)
-        self.assertEqual(a - (a + day), -day)
-        self.assertEqual(a - (a - week), week)
-        self.assertEqual(a - (a - day), day)
-        self.assertEqual(c - (c - day), day)
-
-        # Add/sub ints or floats should be illegal
-        for i in 1, 1.0:
-            self.assertRaises(TypeError, lambda: a+i)
-            self.assertRaises(TypeError, lambda: a-i)
-            self.assertRaises(TypeError, lambda: i+a)
-            self.assertRaises(TypeError, lambda: i-a)
-
-        # delta - date is senseless.
-        self.assertRaises(TypeError, lambda: day - a)
-        # mixing date and (delta or date) via * or // is senseless
-        self.assertRaises(TypeError, lambda: day * a)
-        self.assertRaises(TypeError, lambda: a * day)
-        self.assertRaises(TypeError, lambda: day // a)
-        self.assertRaises(TypeError, lambda: a // day)
-        self.assertRaises(TypeError, lambda: a * a)
-        self.assertRaises(TypeError, lambda: a // a)
-        # date + date is senseless
-        self.assertRaises(TypeError, lambda: a + a)
-
-    def test_overflow(self):
-        tiny = self.theclass.resolution
-
-        for delta in [tiny, timedelta(1), timedelta(2)]:
-            dt = self.theclass.min + delta
-            dt -= delta  # no problem
-            self.assertRaises(OverflowError, dt.__sub__, delta)
-            self.assertRaises(OverflowError, dt.__add__, -delta)
-
-            dt = self.theclass.max - delta
-            dt += delta  # no problem
-            self.assertRaises(OverflowError, dt.__add__, delta)
-            self.assertRaises(OverflowError, dt.__sub__, -delta)
-
-    def test_fromtimestamp(self):
-        import time
-
-        # Try an arbitrary fixed value.
-        year, month, day = 1999, 9, 19
-        ts = time.mktime((year, month, day, 0, 0, 0, 0, 0, -1))
-        d = self.theclass.fromtimestamp(ts)
-        self.assertEqual(d.year, year)
-        self.assertEqual(d.month, month)
-        self.assertEqual(d.day, day)
-
-    def test_insane_fromtimestamp(self):
-        # It's possible that some platform maps time_t to double,
-        # and that this test will fail there.  This test should
-        # exempt such platforms (provided they return reasonable
-        # results!).
-        for insane in -1e200, 1e200:
-            self.assertRaises(ValueError, self.theclass.fromtimestamp,
-                              insane)
-
-    def test_today(self):
-        import time
-
-        # We claim that today() is like fromtimestamp(time.time()), so
-        # prove it.
-        for dummy in range(3):
-            today = self.theclass.today()
-            ts = time.time()
-            todayagain = self.theclass.fromtimestamp(ts)
-            if today == todayagain:
-                break
-            # There are several legit reasons that could fail:
-            # 1. It recently became midnight, between the today() and the
-            #    time() calls.
-            # 2. The platform time() has such fine resolution that we'll
-            #    never get the same value twice.
-            # 3. The platform time() has poor resolution, and we just
-            #    happened to call today() right before a resolution quantum
-            #    boundary.
-            # 4. The system clock got fiddled between calls.
-            # In any case, wait a little while and try again.
-            time.sleep(0.1)
-
-        # It worked or it didn't.  If it didn't, assume it's reason #2, and
-        # let the test pass if they're within half a second of each other.
-        self.assertTrue(today == todayagain or
-                        abs(todayagain - today) < timedelta(seconds=0.5))
-
-    def test_weekday(self):
-        for i in range(7):
-            # March 4, 2002 is a Monday
-            self.assertEqual(self.theclass(2002, 3, 4+i).weekday(), i)
-            self.assertEqual(self.theclass(2002, 3, 4+i).isoweekday(), i+1)
-            # January 2, 1956 is a Monday
-            self.assertEqual(self.theclass(1956, 1, 2+i).weekday(), i)
-            self.assertEqual(self.theclass(1956, 1, 2+i).isoweekday(), i+1)
-
-    def test_isocalendar(self):
-        # Check examples from
-        # http://www.phys.uu.nl/~vgent/calendar/isocalendar.htm
-        for i in range(7):
-            d = self.theclass(2003, 12, 22+i)
-            self.assertEqual(d.isocalendar(), (2003, 52, i+1))
-            d = self.theclass(2003, 12, 29) + timedelta(i)
-            self.assertEqual(d.isocalendar(), (2004, 1, i+1))
-            d = self.theclass(2004, 1, 5+i)
-            self.assertEqual(d.isocalendar(), (2004, 2, i+1))
-            d = self.theclass(2009, 12, 21+i)
-            self.assertEqual(d.isocalendar(), (2009, 52, i+1))
-            d = self.theclass(2009, 12, 28) + timedelta(i)
-            self.assertEqual(d.isocalendar(), (2009, 53, i+1))
-            d = self.theclass(2010, 1, 4+i)
-            self.assertEqual(d.isocalendar(), (2010, 1, i+1))
-
-    def test_iso_long_years(self):
-        # Calculate long ISO years and compare to table from
-        # http://www.phys.uu.nl/~vgent/calendar/isocalendar.htm
-        ISO_LONG_YEARS_TABLE = """
-              4   32   60   88
-              9   37   65   93
-             15   43   71   99
-             20   48   76
-             26   54   82
-
-            105  133  161  189
-            111  139  167  195
-            116  144  172
-            122  150  178
-            128  156  184
-
-            201  229  257  285
-            207  235  263  291
-            212  240  268  296
-            218  246  274
-            224  252  280
-
-            303  331  359  387
-            308  336  364  392
-            314  342  370  398
-            320  348  376
-            325  353  381
-        """
-        iso_long_years = sorted(map(int, ISO_LONG_YEARS_TABLE.split()))
-        L = []
-        for i in range(400):
-            d = self.theclass(2000+i, 12, 31)
-            d1 = self.theclass(1600+i, 12, 31)
-            self.assertEqual(d.isocalendar()[1:], d1.isocalendar()[1:])
-            if d.isocalendar()[1] == 53:
-                L.append(i)
-        self.assertEqual(L, iso_long_years)
-
-    def test_isoformat(self):
-        t = self.theclass(2, 3, 2)
-        self.assertEqual(t.isoformat(), "0002-03-02")
-
-    def test_ctime(self):
-        t = self.theclass(2002, 3, 2)
-        self.assertEqual(t.ctime(), "Sat Mar  2 00:00:00 2002")
-
-    def test_strftime(self):
-        t = self.theclass(2005, 3, 2)
-        self.assertEqual(t.strftime("m:%m d:%d y:%y"), "m:03 d:02 y:05")
-        self.assertEqual(t.strftime(""), "") # SF bug #761337
-        self.assertEqual(t.strftime('x'*1000), 'x'*1000) # SF bug #1556784
-
-        self.assertRaises(TypeError, t.strftime) # needs an arg
-        self.assertRaises(TypeError, t.strftime, "one", "two") # too many args
-        self.assertRaises(TypeError, t.strftime, 42) # arg wrong type
-
-        # test that unicode input is allowed (issue 2782)
-        self.assertEqual(t.strftime("%m"), "03")
-
-        # A naive object replaces %z and %Z w/ empty strings.
-        self.assertEqual(t.strftime("'%z' '%Z'"), "'' ''")
-
-        #make sure that invalid format specifiers are handled correctly
-        #self.assertRaises(ValueError, t.strftime, "%e")
-        #self.assertRaises(ValueError, t.strftime, "%")
-        #self.assertRaises(ValueError, t.strftime, "%#")
-
-        #oh well, some systems just ignore those invalid ones.
-        #at least, excercise them to make sure that no crashes
-        #are generated
-        for f in ["%e", "%", "%#"]:
-            try:
-                t.strftime(f)
-            except ValueError:
-                pass
-
-        #check that this standard extension works
-        t.strftime("%f")
-
-
-    def test_format(self):
-        dt = self.theclass(2007, 9, 10)
-        self.assertEqual(dt.__format__(''), str(dt))
-
-        # check that a derived class's __str__() gets called
-        class A(self.theclass):
-            def __str__(self):
-                return 'A'
-        a = A(2007, 9, 10)
-        self.assertEqual(a.__format__(''), 'A')
-
-        # check that a derived class's strftime gets called
-        class B(self.theclass):
-            def strftime(self, format_spec):
-                return 'B'
-        b = B(2007, 9, 10)
-        self.assertEqual(b.__format__(''), str(dt))
-
-        for fmt in ["m:%m d:%d y:%y",
-                    "m:%m d:%d y:%y H:%H M:%M S:%S",
-                    "%z %Z",
-                    ]:
-            self.assertEqual(dt.__format__(fmt), dt.strftime(fmt))
-            self.assertEqual(a.__format__(fmt), dt.strftime(fmt))
-            self.assertEqual(b.__format__(fmt), 'B')
-
-    def test_resolution_info(self):
-        # XXX: Should min and max respect subclassing?
-        if issubclass(self.theclass, datetime):
-            expected_class = datetime
-        else:
-            expected_class = date
-        self.assertIsInstance(self.theclass.min, expected_class)
-        self.assertIsInstance(self.theclass.max, expected_class)
-        self.assertIsInstance(self.theclass.resolution, timedelta)
-        self.assertTrue(self.theclass.max > self.theclass.min)
-
-    def test_extreme_timedelta(self):
-        big = self.theclass.max - self.theclass.min
-        # 3652058 days, 23 hours, 59 minutes, 59 seconds, 999999 microseconds
-        n = (big.days*24*3600 + big.seconds)*1000000 + big.microseconds
-        # n == 315537897599999999 ~= 2**58.13
-        justasbig = timedelta(0, 0, n)
-        self.assertEqual(big, justasbig)
-        self.assertEqual(self.theclass.min + big, self.theclass.max)
-        self.assertEqual(self.theclass.max - big, self.theclass.min)
-
-    def test_timetuple(self):
-        for i in range(7):
-            # January 2, 1956 is a Monday (0)
-            d = self.theclass(1956, 1, 2+i)
-            t = d.timetuple()
-            self.assertEqual(t, (1956, 1, 2+i, 0, 0, 0, i, 2+i, -1))
-            # February 1, 1956 is a Wednesday (2)
-            d = self.theclass(1956, 2, 1+i)
-            t = d.timetuple()
-            self.assertEqual(t, (1956, 2, 1+i, 0, 0, 0, (2+i)%7, 32+i, -1))
-            # March 1, 1956 is a Thursday (3), and is the 31+29+1 = 61st day
-            # of the year.
-            d = self.theclass(1956, 3, 1+i)
-            t = d.timetuple()
-            self.assertEqual(t, (1956, 3, 1+i, 0, 0, 0, (3+i)%7, 61+i, -1))
-            self.assertEqual(t.tm_year, 1956)
-            self.assertEqual(t.tm_mon, 3)
-            self.assertEqual(t.tm_mday, 1+i)
-            self.assertEqual(t.tm_hour, 0)
-            self.assertEqual(t.tm_min, 0)
-            self.assertEqual(t.tm_sec, 0)
-            self.assertEqual(t.tm_wday, (3+i)%7)
-            self.assertEqual(t.tm_yday, 61+i)
-            self.assertEqual(t.tm_isdst, -1)
-
-    def test_pickling(self):
-        args = 6, 7, 23
-        orig = self.theclass(*args)
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertEqual(orig, derived)
-
-    def test_compare(self):
-        t1 = self.theclass(2, 3, 4)
-        t2 = self.theclass(2, 3, 4)
-        self.assertEqual(t1, t2)
-        self.assertTrue(t1 <= t2)
-        self.assertTrue(t1 >= t2)
-        self.assertTrue(not t1 != t2)
-        self.assertTrue(not t1 < t2)
-        self.assertTrue(not t1 > t2)
-
-        for args in (3, 3, 3), (2, 4, 4), (2, 3, 5):
-            t2 = self.theclass(*args)   # this is larger than t1
-            self.assertTrue(t1 < t2)
-            self.assertTrue(t2 > t1)
-            self.assertTrue(t1 <= t2)
-            self.assertTrue(t2 >= t1)
-            self.assertTrue(t1 != t2)
-            self.assertTrue(t2 != t1)
-            self.assertTrue(not t1 == t2)
-            self.assertTrue(not t2 == t1)
-            self.assertTrue(not t1 > t2)
-            self.assertTrue(not t2 < t1)
-            self.assertTrue(not t1 >= t2)
-            self.assertTrue(not t2 <= t1)
-
-        for badarg in OTHERSTUFF:
-            self.assertEqual(t1 == badarg, False)
-            self.assertEqual(t1 != badarg, True)
-            self.assertEqual(badarg == t1, False)
-            self.assertEqual(badarg != t1, True)
-
-            self.assertRaises(TypeError, lambda: t1 < badarg)
-            self.assertRaises(TypeError, lambda: t1 > badarg)
-            self.assertRaises(TypeError, lambda: t1 >= badarg)
-            self.assertRaises(TypeError, lambda: badarg <= t1)
-            self.assertRaises(TypeError, lambda: badarg < t1)
-            self.assertRaises(TypeError, lambda: badarg > t1)
-            self.assertRaises(TypeError, lambda: badarg >= t1)
-
-    def test_mixed_compare(self):
-        our = self.theclass(2000, 4, 5)
-
-        # Our class can be compared for equality to other classes
-        self.assertEqual(our == 1, False)
-        self.assertEqual(1 == our, False)
-        self.assertEqual(our != 1, True)
-        self.assertEqual(1 != our, True)
-
-        # But the ordering is undefined
-        self.assertRaises(TypeError, lambda: our < 1)
-        self.assertRaises(TypeError, lambda: 1 < our)
-
-        # Repeat those tests with a different class
-
-        class SomeClass:
-            pass
-
-        their = SomeClass()
-        self.assertEqual(our == their, False)
-        self.assertEqual(their == our, False)
-        self.assertEqual(our != their, True)
-        self.assertEqual(their != our, True)
-        self.assertRaises(TypeError, lambda: our < their)
-        self.assertRaises(TypeError, lambda: their < our)
-
-        # However, if the other class explicitly defines ordering
-        # relative to our class, it is allowed to do so
-
-        class LargerThanAnything:
-            def __lt__(self, other):
-                return False
-            def __le__(self, other):
-                return isinstance(other, LargerThanAnything)
-            def __eq__(self, other):
-                return isinstance(other, LargerThanAnything)
-            def __ne__(self, other):
-                return not isinstance(other, LargerThanAnything)
-            def __gt__(self, other):
-                return not isinstance(other, LargerThanAnything)
-            def __ge__(self, other):
-                return True
-
-        their = LargerThanAnything()
-        self.assertEqual(our == their, False)
-        self.assertEqual(their == our, False)
-        self.assertEqual(our != their, True)
-        self.assertEqual(their != our, True)
-        self.assertEqual(our < their, True)
-        self.assertEqual(their < our, False)
-
-    def test_bool(self):
-        # All dates are considered true.
-        self.assertTrue(self.theclass.min)
-        self.assertTrue(self.theclass.max)
-
-    def test_strftime_out_of_range(self):
-        # For nasty technical reasons, we can't handle years before 1900.
-        cls = self.theclass
-        self.assertEqual(cls(1900, 1, 1).strftime("%Y"), "1900")
-        for y in 1, 49, 51, 99, 100, 1000, 1899:
-            self.assertRaises(ValueError, cls(y, 1, 1).strftime, "%Y")
-
-    def test_replace(self):
-        cls = self.theclass
-        args = [1, 2, 3]
-        base = cls(*args)
-        self.assertEqual(base, base.replace())
-
-        i = 0
-        for name, newval in (("year", 2),
-                             ("month", 3),
-                             ("day", 4)):
-            newargs = args[:]
-            newargs[i] = newval
-            expected = cls(*newargs)
-            got = base.replace(**{name: newval})
-            self.assertEqual(expected, got)
-            i += 1
-
-        # Out of bounds.
-        base = cls(2000, 2, 29)
-        self.assertRaises(ValueError, base.replace, year=2001)
-
-    def test_subclass_date(self):
-
-        class C(self.theclass):
-            theAnswer = 42
-
-            def __new__(cls, *args, **kws):
-                temp = kws.copy()
-                extra = temp.pop('extra')
-                result = self.theclass.__new__(cls, *args, **temp)
-                result.extra = extra
-                return result
-
-            def newmeth(self, start):
-                return start + self.year + self.month
-
-        args = 2003, 4, 14
-
-        dt1 = self.theclass(*args)
-        dt2 = C(*args, **{'extra': 7})
-
-        self.assertEqual(dt2.__class__, C)
-        self.assertEqual(dt2.theAnswer, 42)
-        self.assertEqual(dt2.extra, 7)
-        self.assertEqual(dt1.toordinal(), dt2.toordinal())
-        self.assertEqual(dt2.newmeth(-7), dt1.year + dt1.month - 7)
-
-    def test_pickling_subclass_date(self):
-
-        args = 6, 7, 23
-        orig = SubclassDate(*args)
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertEqual(orig, derived)
-
-    def test_backdoor_resistance(self):
-        # For fast unpickling, the constructor accepts a pickle byte string.
-        # This is a low-overhead backdoor.  A user can (by intent or
-        # mistake) pass a string directly, which (if it's the right length)
-        # will get treated like a pickle, and bypass the normal sanity
-        # checks in the constructor.  This can create insane objects.
-        # The constructor doesn't want to burn the time to validate all
-        # fields, but does check the month field.  This stops, e.g.,
-        # datetime.datetime('1995-03-25') from yielding an insane object.
-        base = b'1995-03-25'
-        if not issubclass(self.theclass, datetime):
-            base = base[:4]
-        for month_byte in b'9', b'\0', b'\r', b'\xff':
-            self.assertRaises(TypeError, self.theclass,
-                                         base[:2] + month_byte + base[3:])
-        # Good bytes, but bad tzinfo:
-        self.assertRaises(TypeError, self.theclass,
-                          bytes([1] * len(base)), 'EST')
-
-        for ord_byte in range(1, 13):
-            # This shouldn't blow up because of the month byte alone.  If
-            # the implementation changes to do more-careful checking, it may
-            # blow up because other fields are insane.
-            self.theclass(base[:2] + bytes([ord_byte]) + base[3:])
-
-#############################################################################
-# datetime tests
-
-class SubclassDatetime(datetime):
-    sub_var = 1
-
-class TestDateTime(TestDate):
-
-    theclass = datetime
-
-    def test_basic_attributes(self):
-        dt = self.theclass(2002, 3, 1, 12, 0)
-        self.assertEqual(dt.year, 2002)
-        self.assertEqual(dt.month, 3)
-        self.assertEqual(dt.day, 1)
-        self.assertEqual(dt.hour, 12)
-        self.assertEqual(dt.minute, 0)
-        self.assertEqual(dt.second, 0)
-        self.assertEqual(dt.microsecond, 0)
-
-    def test_basic_attributes_nonzero(self):
-        # Make sure all attributes are non-zero so bugs in
-        # bit-shifting access show up.
-        dt = self.theclass(2002, 3, 1, 12, 59, 59, 8000)
-        self.assertEqual(dt.year, 2002)
-        self.assertEqual(dt.month, 3)
-        self.assertEqual(dt.day, 1)
-        self.assertEqual(dt.hour, 12)
-        self.assertEqual(dt.minute, 59)
-        self.assertEqual(dt.second, 59)
-        self.assertEqual(dt.microsecond, 8000)
-
-    def test_roundtrip(self):
-        for dt in (self.theclass(1, 2, 3, 4, 5, 6, 7),
-                   self.theclass.now()):
-            # Verify dt -> string -> datetime identity.
-            s = repr(dt)
-            self.assertTrue(s.startswith('datetime.'))
-            s = s[9:]
-            dt2 = eval(s)
-            self.assertEqual(dt, dt2)
-
-            # Verify identity via reconstructing from pieces.
-            dt2 = self.theclass(dt.year, dt.month, dt.day,
-                                dt.hour, dt.minute, dt.second,
-                                dt.microsecond)
-            self.assertEqual(dt, dt2)
-
-    def test_isoformat(self):
-        t = self.theclass(2, 3, 2, 4, 5, 1, 123)
-        self.assertEqual(t.isoformat(),    "0002-03-02T04:05:01.000123")
-        self.assertEqual(t.isoformat('T'), "0002-03-02T04:05:01.000123")
-        self.assertEqual(t.isoformat(' '), "0002-03-02 04:05:01.000123")
-        self.assertEqual(t.isoformat('\x00'), "0002-03-02\x0004:05:01.000123")
-        # str is ISO format with the separator forced to a blank.
-        self.assertEqual(str(t), "0002-03-02 04:05:01.000123")
-
-        t = self.theclass(2, 3, 2)
-        self.assertEqual(t.isoformat(),    "0002-03-02T00:00:00")
-        self.assertEqual(t.isoformat('T'), "0002-03-02T00:00:00")
-        self.assertEqual(t.isoformat(' '), "0002-03-02 00:00:00")
-        # str is ISO format with the separator forced to a blank.
-        self.assertEqual(str(t), "0002-03-02 00:00:00")
-
-    def test_format(self):
-        dt = self.theclass(2007, 9, 10, 4, 5, 1, 123)
-        self.assertEqual(dt.__format__(''), str(dt))
-
-        # check that a derived class's __str__() gets called
-        class A(self.theclass):
-            def __str__(self):
-                return 'A'
-        a = A(2007, 9, 10, 4, 5, 1, 123)
-        self.assertEqual(a.__format__(''), 'A')
-
-        # check that a derived class's strftime gets called
-        class B(self.theclass):
-            def strftime(self, format_spec):
-                return 'B'
-        b = B(2007, 9, 10, 4, 5, 1, 123)
-        self.assertEqual(b.__format__(''), str(dt))
-
-        for fmt in ["m:%m d:%d y:%y",
-                    "m:%m d:%d y:%y H:%H M:%M S:%S",
-                    "%z %Z",
-                    ]:
-            self.assertEqual(dt.__format__(fmt), dt.strftime(fmt))
-            self.assertEqual(a.__format__(fmt), dt.strftime(fmt))
-            self.assertEqual(b.__format__(fmt), 'B')
-
-    def test_more_ctime(self):
-        # Test fields that TestDate doesn't touch.
-        import time
-
-        t = self.theclass(2002, 3, 2, 18, 3, 5, 123)
-        self.assertEqual(t.ctime(), "Sat Mar  2 18:03:05 2002")
-        # Oops!  The next line fails on Win2K under MSVC 6, so it's commented
-        # out.  The difference is that t.ctime() produces " 2" for the day,
-        # but platform ctime() produces "02" for the day.  According to
-        # C99, t.ctime() is correct here.
-        # self.assertEqual(t.ctime(), time.ctime(time.mktime(t.timetuple())))
-
-        # So test a case where that difference doesn't matter.
-        t = self.theclass(2002, 3, 22, 18, 3, 5, 123)
-        self.assertEqual(t.ctime(), time.ctime(time.mktime(t.timetuple())))
-
-    def test_tz_independent_comparing(self):
-        dt1 = self.theclass(2002, 3, 1, 9, 0, 0)
-        dt2 = self.theclass(2002, 3, 1, 10, 0, 0)
-        dt3 = self.theclass(2002, 3, 1, 9, 0, 0)
-        self.assertEqual(dt1, dt3)
-        self.assertTrue(dt2 > dt3)
-
-        # Make sure comparison doesn't forget microseconds, and isn't done
-        # via comparing a float timestamp (an IEEE double doesn't have enough
-        # precision to span microsecond resolution across years 1 thru 9999,
-        # so comparing via timestamp necessarily calls some distinct values
-        # equal).
-        dt1 = self.theclass(MAXYEAR, 12, 31, 23, 59, 59, 999998)
-        us = timedelta(microseconds=1)
-        dt2 = dt1 + us
-        self.assertEqual(dt2 - dt1, us)
-        self.assertTrue(dt1 < dt2)
-
-    def test_strftime_with_bad_tzname_replace(self):
-        # verify ok if tzinfo.tzname().replace() returns a non-string
-        class MyTzInfo(FixedOffset):
-            def tzname(self, dt):
-                class MyStr(str):
-                    def replace(self, *args):
-                        return None
-                return MyStr('name')
-        t = self.theclass(2005, 3, 2, 0, 0, 0, 0, MyTzInfo(3, 'name'))
-        self.assertRaises(TypeError, t.strftime, '%Z')
-
-    def test_bad_constructor_arguments(self):
-        # bad years
-        self.theclass(MINYEAR, 1, 1)  # no exception
-        self.theclass(MAXYEAR, 1, 1)  # no exception
-        self.assertRaises(ValueError, self.theclass, MINYEAR-1, 1, 1)
-        self.assertRaises(ValueError, self.theclass, MAXYEAR+1, 1, 1)
-        # bad months
-        self.theclass(2000, 1, 1)    # no exception
-        self.theclass(2000, 12, 1)   # no exception
-        self.assertRaises(ValueError, self.theclass, 2000, 0, 1)
-        self.assertRaises(ValueError, self.theclass, 2000, 13, 1)
-        # bad days
-        self.theclass(2000, 2, 29)   # no exception
-        self.theclass(2004, 2, 29)   # no exception
-        self.theclass(2400, 2, 29)   # no exception
-        self.assertRaises(ValueError, self.theclass, 2000, 2, 30)
-        self.assertRaises(ValueError, self.theclass, 2001, 2, 29)
-        self.assertRaises(ValueError, self.theclass, 2100, 2, 29)
-        self.assertRaises(ValueError, self.theclass, 1900, 2, 29)
-        self.assertRaises(ValueError, self.theclass, 2000, 1, 0)
-        self.assertRaises(ValueError, self.theclass, 2000, 1, 32)
-        # bad hours
-        self.theclass(2000, 1, 31, 0)    # no exception
-        self.theclass(2000, 1, 31, 23)   # no exception
-        self.assertRaises(ValueError, self.theclass, 2000, 1, 31, -1)
-        self.assertRaises(ValueError, self.theclass, 2000, 1, 31, 24)
-        # bad minutes
-        self.theclass(2000, 1, 31, 23, 0)    # no exception
-        self.theclass(2000, 1, 31, 23, 59)   # no exception
-        self.assertRaises(ValueError, self.theclass, 2000, 1, 31, 23, -1)
-        self.assertRaises(ValueError, self.theclass, 2000, 1, 31, 23, 60)
-        # bad seconds
-        self.theclass(2000, 1, 31, 23, 59, 0)    # no exception
-        self.theclass(2000, 1, 31, 23, 59, 59)   # no exception
-        self.assertRaises(ValueError, self.theclass, 2000, 1, 31, 23, 59, -1)
-        self.assertRaises(ValueError, self.theclass, 2000, 1, 31, 23, 59, 60)
-        # bad microseconds
-        self.theclass(2000, 1, 31, 23, 59, 59, 0)    # no exception
-        self.theclass(2000, 1, 31, 23, 59, 59, 999999)   # no exception
-        self.assertRaises(ValueError, self.theclass,
-                          2000, 1, 31, 23, 59, 59, -1)
-        self.assertRaises(ValueError, self.theclass,
-                          2000, 1, 31, 23, 59, 59,
-                          1000000)
-
-    def test_hash_equality(self):
-        d = self.theclass(2000, 12, 31, 23, 30, 17)
-        e = self.theclass(2000, 12, 31, 23, 30, 17)
-        self.assertEqual(d, e)
-        self.assertEqual(hash(d), hash(e))
-
-        dic = {d: 1}
-        dic[e] = 2
-        self.assertEqual(len(dic), 1)
-        self.assertEqual(dic[d], 2)
-        self.assertEqual(dic[e], 2)
-
-        d = self.theclass(2001,  1,  1,  0,  5, 17)
-        e = self.theclass(2001,  1,  1,  0,  5, 17)
-        self.assertEqual(d, e)
-        self.assertEqual(hash(d), hash(e))
-
-        dic = {d: 1}
-        dic[e] = 2
-        self.assertEqual(len(dic), 1)
-        self.assertEqual(dic[d], 2)
-        self.assertEqual(dic[e], 2)
-
-    def test_computations(self):
-        a = self.theclass(2002, 1, 31)
-        b = self.theclass(1956, 1, 31)
-        diff = a-b
-        self.assertEqual(diff.days, 46*365 + len(range(1956, 2002, 4)))
-        self.assertEqual(diff.seconds, 0)
-        self.assertEqual(diff.microseconds, 0)
-        a = self.theclass(2002, 3, 2, 17, 6)
-        millisec = timedelta(0, 0, 1000)
-        hour = timedelta(0, 3600)
-        day = timedelta(1)
-        week = timedelta(7)
-        self.assertEqual(a + hour, self.theclass(2002, 3, 2, 18, 6))
-        self.assertEqual(hour + a, self.theclass(2002, 3, 2, 18, 6))
-        self.assertEqual(a + 10*hour, self.theclass(2002, 3, 3, 3, 6))
-        self.assertEqual(a - hour, self.theclass(2002, 3, 2, 16, 6))
-        self.assertEqual(-hour + a, self.theclass(2002, 3, 2, 16, 6))
-        self.assertEqual(a - hour, a + -hour)
-        self.assertEqual(a - 20*hour, self.theclass(2002, 3, 1, 21, 6))
-        self.assertEqual(a + day, self.theclass(2002, 3, 3, 17, 6))
-        self.assertEqual(a - day, self.theclass(2002, 3, 1, 17, 6))
-        self.assertEqual(a + week, self.theclass(2002, 3, 9, 17, 6))
-        self.assertEqual(a - week, self.theclass(2002, 2, 23, 17, 6))
-        self.assertEqual(a + 52*week, self.theclass(2003, 3, 1, 17, 6))
-        self.assertEqual(a - 52*week, self.theclass(2001, 3, 3, 17, 6))
-        self.assertEqual((a + week) - a, week)
-        self.assertEqual((a + day) - a, day)
-        self.assertEqual((a + hour) - a, hour)
-        self.assertEqual((a + millisec) - a, millisec)
-        self.assertEqual((a - week) - a, -week)
-        self.assertEqual((a - day) - a, -day)
-        self.assertEqual((a - hour) - a, -hour)
-        self.assertEqual((a - millisec) - a, -millisec)
-        self.assertEqual(a - (a + week), -week)
-        self.assertEqual(a - (a + day), -day)
-        self.assertEqual(a - (a + hour), -hour)
-        self.assertEqual(a - (a + millisec), -millisec)
-        self.assertEqual(a - (a - week), week)
-        self.assertEqual(a - (a - day), day)
-        self.assertEqual(a - (a - hour), hour)
-        self.assertEqual(a - (a - millisec), millisec)
-        self.assertEqual(a + (week + day + hour + millisec),
-                         self.theclass(2002, 3, 10, 18, 6, 0, 1000))
-        self.assertEqual(a + (week + day + hour + millisec),
-                         (((a + week) + day) + hour) + millisec)
-        self.assertEqual(a - (week + day + hour + millisec),
-                         self.theclass(2002, 2, 22, 16, 5, 59, 999000))
-        self.assertEqual(a - (week + day + hour + millisec),
-                         (((a - week) - day) - hour) - millisec)
-        # Add/sub ints or floats should be illegal
-        for i in 1, 1.0:
-            self.assertRaises(TypeError, lambda: a+i)
-            self.assertRaises(TypeError, lambda: a-i)
-            self.assertRaises(TypeError, lambda: i+a)
-            self.assertRaises(TypeError, lambda: i-a)
-
-        # delta - datetime is senseless.
-        self.assertRaises(TypeError, lambda: day - a)
-        # mixing datetime and (delta or datetime) via * or // is senseless
-        self.assertRaises(TypeError, lambda: day * a)
-        self.assertRaises(TypeError, lambda: a * day)
-        self.assertRaises(TypeError, lambda: day // a)
-        self.assertRaises(TypeError, lambda: a // day)
-        self.assertRaises(TypeError, lambda: a * a)
-        self.assertRaises(TypeError, lambda: a // a)
-        # datetime + datetime is senseless
-        self.assertRaises(TypeError, lambda: a + a)
-
-    def test_pickling(self):
-        args = 6, 7, 23, 20, 59, 1, 64**2
-        orig = self.theclass(*args)
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertEqual(orig, derived)
-
-    def test_more_pickling(self):
-        a = self.theclass(2003, 2, 7, 16, 48, 37, 444116)
-        s = pickle.dumps(a)
-        b = pickle.loads(s)
-        self.assertEqual(b.year, 2003)
-        self.assertEqual(b.month, 2)
-        self.assertEqual(b.day, 7)
-
-    def test_pickling_subclass_datetime(self):
-        args = 6, 7, 23, 20, 59, 1, 64**2
-        orig = SubclassDatetime(*args)
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertEqual(orig, derived)
-
-    def test_more_compare(self):
-        # The test_compare() inherited from TestDate covers the error cases.
-        # We just want to test lexicographic ordering on the members datetime
-        # has that date lacks.
-        args = [2000, 11, 29, 20, 58, 16, 999998]
-        t1 = self.theclass(*args)
-        t2 = self.theclass(*args)
-        self.assertEqual(t1, t2)
-        self.assertTrue(t1 <= t2)
-        self.assertTrue(t1 >= t2)
-        self.assertTrue(not t1 != t2)
-        self.assertTrue(not t1 < t2)
-        self.assertTrue(not t1 > t2)
-
-        for i in range(len(args)):
-            newargs = args[:]
-            newargs[i] = args[i] + 1
-            t2 = self.theclass(*newargs)   # this is larger than t1
-            self.assertTrue(t1 < t2)
-            self.assertTrue(t2 > t1)
-            self.assertTrue(t1 <= t2)
-            self.assertTrue(t2 >= t1)
-            self.assertTrue(t1 != t2)
-            self.assertTrue(t2 != t1)
-            self.assertTrue(not t1 == t2)
-            self.assertTrue(not t2 == t1)
-            self.assertTrue(not t1 > t2)
-            self.assertTrue(not t2 < t1)
-            self.assertTrue(not t1 >= t2)
-            self.assertTrue(not t2 <= t1)
-
-
-    # A helper for timestamp constructor tests.
-    def verify_field_equality(self, expected, got):
-        self.assertEqual(expected.tm_year, got.year)
-        self.assertEqual(expected.tm_mon, got.month)
-        self.assertEqual(expected.tm_mday, got.day)
-        self.assertEqual(expected.tm_hour, got.hour)
-        self.assertEqual(expected.tm_min, got.minute)
-        self.assertEqual(expected.tm_sec, got.second)
-
-    def test_fromtimestamp(self):
-        import time
-
-        ts = time.time()
-        expected = time.localtime(ts)
-        got = self.theclass.fromtimestamp(ts)
-        self.verify_field_equality(expected, got)
-
-    def test_utcfromtimestamp(self):
-        import time
-
-        ts = time.time()
-        expected = time.gmtime(ts)
-        got = self.theclass.utcfromtimestamp(ts)
-        self.verify_field_equality(expected, got)
-
-    def test_microsecond_rounding(self):
-        # Test whether fromtimestamp "rounds up" floats that are less
-        # than one microsecond smaller than an integer.
-        self.assertEqual(self.theclass.fromtimestamp(0.9999999),
-                         self.theclass.fromtimestamp(1))
-
-    def test_insane_fromtimestamp(self):
-        # It's possible that some platform maps time_t to double,
-        # and that this test will fail there.  This test should
-        # exempt such platforms (provided they return reasonable
-        # results!).
-        for insane in -1e200, 1e200:
-            self.assertRaises(ValueError, self.theclass.fromtimestamp,
-                              insane)
-
-    def test_insane_utcfromtimestamp(self):
-        # It's possible that some platform maps time_t to double,
-        # and that this test will fail there.  This test should
-        # exempt such platforms (provided they return reasonable
-        # results!).
-        for insane in -1e200, 1e200:
-            self.assertRaises(ValueError, self.theclass.utcfromtimestamp,
-                              insane)
-    @unittest.skipIf(sys.platform == "win32", "Windows doesn't accept negative timestamps")
-    def test_negative_float_fromtimestamp(self):
-        # The result is tz-dependent; at least test that this doesn't
-        # fail (like it did before bug 1646728 was fixed).
-        self.theclass.fromtimestamp(-1.05)
-
-    @unittest.skipIf(sys.platform == "win32", "Windows doesn't accept negative timestamps")
-    def test_negative_float_utcfromtimestamp(self):
-        d = self.theclass.utcfromtimestamp(-1.05)
-        self.assertEqual(d, self.theclass(1969, 12, 31, 23, 59, 58, 950000))
-
-    def test_utcnow(self):
-        import time
-
-        # Call it a success if utcnow() and utcfromtimestamp() are within
-        # a second of each other.
-        tolerance = timedelta(seconds=1)
-        for dummy in range(3):
-            from_now = self.theclass.utcnow()
-            from_timestamp = self.theclass.utcfromtimestamp(time.time())
-            if abs(from_timestamp - from_now) <= tolerance:
-                break
-            # Else try again a few times.
-        self.assertTrue(abs(from_timestamp - from_now) <= tolerance)
-
-    def test_strptime(self):
-        import _strptime
-
-        string = '2004-12-01 13:02:47.197'
-        format = '%Y-%m-%d %H:%M:%S.%f'
-        expected = _strptime._strptime_datetime(self.theclass, string, format)
-        got = self.theclass.strptime(string, format)
-        self.assertEqual(expected, got)
-        self.assertIs(type(expected), self.theclass)
-        self.assertIs(type(got), self.theclass)
-
-        strptime = self.theclass.strptime
-        self.assertEqual(strptime("+0002", "%z").utcoffset(), 2 * MINUTE)
-        self.assertEqual(strptime("-0002", "%z").utcoffset(), -2 * MINUTE)
-        # Only local timezone and UTC are supported
-        for tzseconds, tzname in ((0, 'UTC'), (0, 'GMT'),
-                                 (-_time.timezone, _time.tzname[0])):
-            if tzseconds < 0:
-                sign = '-'
-                seconds = -tzseconds
-            else:
-                sign ='+'
-                seconds = tzseconds
-            hours, minutes = divmod(seconds//60, 60)
-            dtstr = "{}{:02d}{:02d} {}".format(sign, hours, minutes, tzname)
-            dt = strptime(dtstr, "%z %Z")
-            self.assertEqual(dt.utcoffset(), timedelta(seconds=tzseconds))
-            self.assertEqual(dt.tzname(), tzname)
-        # Can produce inconsistent datetime
-        dtstr, fmt = "+1234 UTC", "%z %Z"
-        dt = strptime(dtstr, fmt)
-        self.assertEqual(dt.utcoffset(), 12 * HOUR + 34 * MINUTE)
-        self.assertEqual(dt.tzname(), 'UTC')
-        # yet will roundtrip
-        self.assertEqual(dt.strftime(fmt), dtstr)
-
-        # Produce naive datetime if no %z is provided
-        self.assertEqual(strptime("UTC", "%Z").tzinfo, None)
-
-        with self.assertRaises(ValueError): strptime("-2400", "%z")
-        with self.assertRaises(ValueError): strptime("-000", "%z")
-
-    def test_more_timetuple(self):
-        # This tests fields beyond those tested by the TestDate.test_timetuple.
-        t = self.theclass(2004, 12, 31, 6, 22, 33)
-        self.assertEqual(t.timetuple(), (2004, 12, 31, 6, 22, 33, 4, 366, -1))
-        self.assertEqual(t.timetuple(),
-                         (t.year, t.month, t.day,
-                          t.hour, t.minute, t.second,
-                          t.weekday(),
-                          t.toordinal() - date(t.year, 1, 1).toordinal() + 1,
-                          -1))
-        tt = t.timetuple()
-        self.assertEqual(tt.tm_year, t.year)
-        self.assertEqual(tt.tm_mon, t.month)
-        self.assertEqual(tt.tm_mday, t.day)
-        self.assertEqual(tt.tm_hour, t.hour)
-        self.assertEqual(tt.tm_min, t.minute)
-        self.assertEqual(tt.tm_sec, t.second)
-        self.assertEqual(tt.tm_wday, t.weekday())
-        self.assertEqual(tt.tm_yday, t.toordinal() -
-                                     date(t.year, 1, 1).toordinal() + 1)
-        self.assertEqual(tt.tm_isdst, -1)
-
-    def test_more_strftime(self):
-        # This tests fields beyond those tested by the TestDate.test_strftime.
-        t = self.theclass(2004, 12, 31, 6, 22, 33, 47)
-        self.assertEqual(t.strftime("%m %d %y %f %S %M %H %j"),
-                                    "12 31 04 000047 33 22 06 366")
-
-    def test_extract(self):
-        dt = self.theclass(2002, 3, 4, 18, 45, 3, 1234)
-        self.assertEqual(dt.date(), date(2002, 3, 4))
-        self.assertEqual(dt.time(), time(18, 45, 3, 1234))
-
-    def test_combine(self):
-        d = date(2002, 3, 4)
-        t = time(18, 45, 3, 1234)
-        expected = self.theclass(2002, 3, 4, 18, 45, 3, 1234)
-        combine = self.theclass.combine
-        dt = combine(d, t)
-        self.assertEqual(dt, expected)
-
-        dt = combine(time=t, date=d)
-        self.assertEqual(dt, expected)
-
-        self.assertEqual(d, dt.date())
-        self.assertEqual(t, dt.time())
-        self.assertEqual(dt, combine(dt.date(), dt.time()))
-
-        self.assertRaises(TypeError, combine) # need an arg
-        self.assertRaises(TypeError, combine, d) # need two args
-        self.assertRaises(TypeError, combine, t, d) # args reversed
-        self.assertRaises(TypeError, combine, d, t, 1) # too many args
-        self.assertRaises(TypeError, combine, "date", "time") # wrong types
-        self.assertRaises(TypeError, combine, d, "time") # wrong type
-        self.assertRaises(TypeError, combine, "date", t) # wrong type
-
-    def test_replace(self):
-        cls = self.theclass
-        args = [1, 2, 3, 4, 5, 6, 7]
-        base = cls(*args)
-        self.assertEqual(base, base.replace())
-
-        i = 0
-        for name, newval in (("year", 2),
-                             ("month", 3),
-                             ("day", 4),
-                             ("hour", 5),
-                             ("minute", 6),
-                             ("second", 7),
-                             ("microsecond", 8)):
-            newargs = args[:]
-            newargs[i] = newval
-            expected = cls(*newargs)
-            got = base.replace(**{name: newval})
-            self.assertEqual(expected, got)
-            i += 1
-
-        # Out of bounds.
-        base = cls(2000, 2, 29)
-        self.assertRaises(ValueError, base.replace, year=2001)
-
-    def test_astimezone(self):
-        # Pretty boring!  The TZ test is more interesting here.  astimezone()
-        # simply can't be applied to a naive object.
-        dt = self.theclass.now()
-        f = FixedOffset(44, "")
-        self.assertRaises(TypeError, dt.astimezone) # not enough args
-        self.assertRaises(TypeError, dt.astimezone, f, f) # too many args
-        self.assertRaises(TypeError, dt.astimezone, dt) # arg wrong type
-        self.assertRaises(ValueError, dt.astimezone, f) # naive
-        self.assertRaises(ValueError, dt.astimezone, tz=f)  # naive
-
-        class Bogus(tzinfo):
-            def utcoffset(self, dt): return None
-            def dst(self, dt): return timedelta(0)
-        bog = Bogus()
-        self.assertRaises(ValueError, dt.astimezone, bog)   # naive
-        self.assertRaises(ValueError,
-                          dt.replace(tzinfo=bog).astimezone, f)
-
-        class AlsoBogus(tzinfo):
-            def utcoffset(self, dt): return timedelta(0)
-            def dst(self, dt): return None
-        alsobog = AlsoBogus()
-        self.assertRaises(ValueError, dt.astimezone, alsobog) # also naive
-
-    def test_subclass_datetime(self):
-
-        class C(self.theclass):
-            theAnswer = 42
-
-            def __new__(cls, *args, **kws):
-                temp = kws.copy()
-                extra = temp.pop('extra')
-                result = self.theclass.__new__(cls, *args, **temp)
-                result.extra = extra
-                return result
-
-            def newmeth(self, start):
-                return start + self.year + self.month + self.second
-
-        args = 2003, 4, 14, 12, 13, 41
-
-        dt1 = self.theclass(*args)
-        dt2 = C(*args, **{'extra': 7})
-
-        self.assertEqual(dt2.__class__, C)
-        self.assertEqual(dt2.theAnswer, 42)
-        self.assertEqual(dt2.extra, 7)
-        self.assertEqual(dt1.toordinal(), dt2.toordinal())
-        self.assertEqual(dt2.newmeth(-7), dt1.year + dt1.month +
-                                          dt1.second - 7)
-
-class TestSubclassDateTime(TestDateTime):
-    theclass = SubclassDatetime
-    # Override tests not designed for subclass
-    def test_roundtrip(self):
-        pass
-
-class SubclassTime(time):
-    sub_var = 1
-
-class TestTime(HarmlessMixedComparison, unittest.TestCase):
-
-    theclass = time
-
-    def test_basic_attributes(self):
-        t = self.theclass(12, 0)
-        self.assertEqual(t.hour, 12)
-        self.assertEqual(t.minute, 0)
-        self.assertEqual(t.second, 0)
-        self.assertEqual(t.microsecond, 0)
-
-    def test_basic_attributes_nonzero(self):
-        # Make sure all attributes are non-zero so bugs in
-        # bit-shifting access show up.
-        t = self.theclass(12, 59, 59, 8000)
-        self.assertEqual(t.hour, 12)
-        self.assertEqual(t.minute, 59)
-        self.assertEqual(t.second, 59)
-        self.assertEqual(t.microsecond, 8000)
-
-    def test_roundtrip(self):
-        t = self.theclass(1, 2, 3, 4)
-
-        # Verify t -> string -> time identity.
-        s = repr(t)
-        self.assertTrue(s.startswith('datetime.'))
-        s = s[9:]
-        t2 = eval(s)
-        self.assertEqual(t, t2)
-
-        # Verify identity via reconstructing from pieces.
-        t2 = self.theclass(t.hour, t.minute, t.second,
-                           t.microsecond)
-        self.assertEqual(t, t2)
-
-    def test_comparing(self):
-        args = [1, 2, 3, 4]
-        t1 = self.theclass(*args)
-        t2 = self.theclass(*args)
-        self.assertEqual(t1, t2)
-        self.assertTrue(t1 <= t2)
-        self.assertTrue(t1 >= t2)
-        self.assertTrue(not t1 != t2)
-        self.assertTrue(not t1 < t2)
-        self.assertTrue(not t1 > t2)
-
-        for i in range(len(args)):
-            newargs = args[:]
-            newargs[i] = args[i] + 1
-            t2 = self.theclass(*newargs)   # this is larger than t1
-            self.assertTrue(t1 < t2)
-            self.assertTrue(t2 > t1)
-            self.assertTrue(t1 <= t2)
-            self.assertTrue(t2 >= t1)
-            self.assertTrue(t1 != t2)
-            self.assertTrue(t2 != t1)
-            self.assertTrue(not t1 == t2)
-            self.assertTrue(not t2 == t1)
-            self.assertTrue(not t1 > t2)
-            self.assertTrue(not t2 < t1)
-            self.assertTrue(not t1 >= t2)
-            self.assertTrue(not t2 <= t1)
-
-        for badarg in OTHERSTUFF:
-            self.assertEqual(t1 == badarg, False)
-            self.assertEqual(t1 != badarg, True)
-            self.assertEqual(badarg == t1, False)
-            self.assertEqual(badarg != t1, True)
-
-            self.assertRaises(TypeError, lambda: t1 <= badarg)
-            self.assertRaises(TypeError, lambda: t1 < badarg)
-            self.assertRaises(TypeError, lambda: t1 > badarg)
-            self.assertRaises(TypeError, lambda: t1 >= badarg)
-            self.assertRaises(TypeError, lambda: badarg <= t1)
-            self.assertRaises(TypeError, lambda: badarg < t1)
-            self.assertRaises(TypeError, lambda: badarg > t1)
-            self.assertRaises(TypeError, lambda: badarg >= t1)
-
-    def test_bad_constructor_arguments(self):
-        # bad hours
-        self.theclass(0, 0)    # no exception
-        self.theclass(23, 0)   # no exception
-        self.assertRaises(ValueError, self.theclass, -1, 0)
-        self.assertRaises(ValueError, self.theclass, 24, 0)
-        # bad minutes
-        self.theclass(23, 0)    # no exception
-        self.theclass(23, 59)   # no exception
-        self.assertRaises(ValueError, self.theclass, 23, -1)
-        self.assertRaises(ValueError, self.theclass, 23, 60)
-        # bad seconds
-        self.theclass(23, 59, 0)    # no exception
-        self.theclass(23, 59, 59)   # no exception
-        self.assertRaises(ValueError, self.theclass, 23, 59, -1)
-        self.assertRaises(ValueError, self.theclass, 23, 59, 60)
-        # bad microseconds
-        self.theclass(23, 59, 59, 0)        # no exception
-        self.theclass(23, 59, 59, 999999)   # no exception
-        self.assertRaises(ValueError, self.theclass, 23, 59, 59, -1)
-        self.assertRaises(ValueError, self.theclass, 23, 59, 59, 1000000)
-
-    def test_hash_equality(self):
-        d = self.theclass(23, 30, 17)
-        e = self.theclass(23, 30, 17)
-        self.assertEqual(d, e)
-        self.assertEqual(hash(d), hash(e))
-
-        dic = {d: 1}
-        dic[e] = 2
-        self.assertEqual(len(dic), 1)
-        self.assertEqual(dic[d], 2)
-        self.assertEqual(dic[e], 2)
-
-        d = self.theclass(0,  5, 17)
-        e = self.theclass(0,  5, 17)
-        self.assertEqual(d, e)
-        self.assertEqual(hash(d), hash(e))
-
-        dic = {d: 1}
-        dic[e] = 2
-        self.assertEqual(len(dic), 1)
-        self.assertEqual(dic[d], 2)
-        self.assertEqual(dic[e], 2)
-
-    def test_isoformat(self):
-        t = self.theclass(4, 5, 1, 123)
-        self.assertEqual(t.isoformat(), "04:05:01.000123")
-        self.assertEqual(t.isoformat(), str(t))
-
-        t = self.theclass()
-        self.assertEqual(t.isoformat(), "00:00:00")
-        self.assertEqual(t.isoformat(), str(t))
-
-        t = self.theclass(microsecond=1)
-        self.assertEqual(t.isoformat(), "00:00:00.000001")
-        self.assertEqual(t.isoformat(), str(t))
-
-        t = self.theclass(microsecond=10)
-        self.assertEqual(t.isoformat(), "00:00:00.000010")
-        self.assertEqual(t.isoformat(), str(t))
-
-        t = self.theclass(microsecond=100)
-        self.assertEqual(t.isoformat(), "00:00:00.000100")
-        self.assertEqual(t.isoformat(), str(t))
-
-        t = self.theclass(microsecond=1000)
-        self.assertEqual(t.isoformat(), "00:00:00.001000")
-        self.assertEqual(t.isoformat(), str(t))
-
-        t = self.theclass(microsecond=10000)
-        self.assertEqual(t.isoformat(), "00:00:00.010000")
-        self.assertEqual(t.isoformat(), str(t))
-
-        t = self.theclass(microsecond=100000)
-        self.assertEqual(t.isoformat(), "00:00:00.100000")
-        self.assertEqual(t.isoformat(), str(t))
-
-    def test_1653736(self):
-        # verify it doesn't accept extra keyword arguments
-        t = self.theclass(second=1)
-        self.assertRaises(TypeError, t.isoformat, foo=3)
-
-    def test_strftime(self):
-        t = self.theclass(1, 2, 3, 4)
-        self.assertEqual(t.strftime('%H %M %S %f'), "01 02 03 000004")
-        # A naive object replaces %z and %Z with empty strings.
-        self.assertEqual(t.strftime("'%z' '%Z'"), "'' ''")
-
-    def test_format(self):
-        t = self.theclass(1, 2, 3, 4)
-        self.assertEqual(t.__format__(''), str(t))
-
-        # check that a derived class's __str__() gets called
-        class A(self.theclass):
-            def __str__(self):
-                return 'A'
-        a = A(1, 2, 3, 4)
-        self.assertEqual(a.__format__(''), 'A')
-
-        # check that a derived class's strftime gets called
-        class B(self.theclass):
-            def strftime(self, format_spec):
-                return 'B'
-        b = B(1, 2, 3, 4)
-        self.assertEqual(b.__format__(''), str(t))
-
-        for fmt in ['%H %M %S',
-                    ]:
-            self.assertEqual(t.__format__(fmt), t.strftime(fmt))
-            self.assertEqual(a.__format__(fmt), t.strftime(fmt))
-            self.assertEqual(b.__format__(fmt), 'B')
-
-    def test_str(self):
-        self.assertEqual(str(self.theclass(1, 2, 3, 4)), "01:02:03.000004")
-        self.assertEqual(str(self.theclass(10, 2, 3, 4000)), "10:02:03.004000")
-        self.assertEqual(str(self.theclass(0, 2, 3, 400000)), "00:02:03.400000")
-        self.assertEqual(str(self.theclass(12, 2, 3, 0)), "12:02:03")
-        self.assertEqual(str(self.theclass(23, 15, 0, 0)), "23:15:00")
-
-    def test_repr(self):
-        name = 'datetime.' + self.theclass.__name__
-        self.assertEqual(repr(self.theclass(1, 2, 3, 4)),
-                         "%s(1, 2, 3, 4)" % name)
-        self.assertEqual(repr(self.theclass(10, 2, 3, 4000)),
-                         "%s(10, 2, 3, 4000)" % name)
-        self.assertEqual(repr(self.theclass(0, 2, 3, 400000)),
-                         "%s(0, 2, 3, 400000)" % name)
-        self.assertEqual(repr(self.theclass(12, 2, 3, 0)),
-                         "%s(12, 2, 3)" % name)
-        self.assertEqual(repr(self.theclass(23, 15, 0, 0)),
-                         "%s(23, 15)" % name)
-
-    def test_resolution_info(self):
-        self.assertIsInstance(self.theclass.min, self.theclass)
-        self.assertIsInstance(self.theclass.max, self.theclass)
-        self.assertIsInstance(self.theclass.resolution, timedelta)
-        self.assertTrue(self.theclass.max > self.theclass.min)
-
-    def test_pickling(self):
-        args = 20, 59, 16, 64**2
-        orig = self.theclass(*args)
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertEqual(orig, derived)
-
-    def test_pickling_subclass_time(self):
-        args = 20, 59, 16, 64**2
-        orig = SubclassTime(*args)
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertEqual(orig, derived)
-
-    def test_bool(self):
-        cls = self.theclass
-        self.assertTrue(cls(1))
-        self.assertTrue(cls(0, 1))
-        self.assertTrue(cls(0, 0, 1))
-        self.assertTrue(cls(0, 0, 0, 1))
-        self.assertTrue(not cls(0))
-        self.assertTrue(not cls())
-
-    def test_replace(self):
-        cls = self.theclass
-        args = [1, 2, 3, 4]
-        base = cls(*args)
-        self.assertEqual(base, base.replace())
-
-        i = 0
-        for name, newval in (("hour", 5),
-                             ("minute", 6),
-                             ("second", 7),
-                             ("microsecond", 8)):
-            newargs = args[:]
-            newargs[i] = newval
-            expected = cls(*newargs)
-            got = base.replace(**{name: newval})
-            self.assertEqual(expected, got)
-            i += 1
-
-        # Out of bounds.
-        base = cls(1)
-        self.assertRaises(ValueError, base.replace, hour=24)
-        self.assertRaises(ValueError, base.replace, minute=-1)
-        self.assertRaises(ValueError, base.replace, second=100)
-        self.assertRaises(ValueError, base.replace, microsecond=1000000)
-
-    def test_subclass_time(self):
-
-        class C(self.theclass):
-            theAnswer = 42
-
-            def __new__(cls, *args, **kws):
-                temp = kws.copy()
-                extra = temp.pop('extra')
-                result = self.theclass.__new__(cls, *args, **temp)
-                result.extra = extra
-                return result
-
-            def newmeth(self, start):
-                return start + self.hour + self.second
-
-        args = 4, 5, 6
-
-        dt1 = self.theclass(*args)
-        dt2 = C(*args, **{'extra': 7})
-
-        self.assertEqual(dt2.__class__, C)
-        self.assertEqual(dt2.theAnswer, 42)
-        self.assertEqual(dt2.extra, 7)
-        self.assertEqual(dt1.isoformat(), dt2.isoformat())
-        self.assertEqual(dt2.newmeth(-7), dt1.hour + dt1.second - 7)
-
-    def test_backdoor_resistance(self):
-        # see TestDate.test_backdoor_resistance().
-        base = '2:59.0'
-        for hour_byte in ' ', '9', chr(24), '\xff':
-            self.assertRaises(TypeError, self.theclass,
-                                         hour_byte + base[1:])
-
-# A mixin for classes with a tzinfo= argument.  Subclasses must define
-# theclass as a class atribute, and theclass(1, 1, 1, tzinfo=whatever)
-# must be legit (which is true for time and datetime).
-class TZInfoBase:
-
-    def test_argument_passing(self):
-        cls = self.theclass
-        # A datetime passes itself on, a time passes None.
-        class introspective(tzinfo):
-            def tzname(self, dt):    return dt and "real" or "none"
-            def utcoffset(self, dt):
-                return timedelta(minutes = dt and 42 or -42)
-            dst = utcoffset
-
-        obj = cls(1, 2, 3, tzinfo=introspective())
-
-        expected = cls is time and "none" or "real"
-        self.assertEqual(obj.tzname(), expected)
-
-        expected = timedelta(minutes=(cls is time and -42 or 42))
-        self.assertEqual(obj.utcoffset(), expected)
-        self.assertEqual(obj.dst(), expected)
-
-    def test_bad_tzinfo_classes(self):
-        cls = self.theclass
-        self.assertRaises(TypeError, cls, 1, 1, 1, tzinfo=12)
-
-        class NiceTry(object):
-            def __init__(self): pass
-            def utcoffset(self, dt): pass
-        self.assertRaises(TypeError, cls, 1, 1, 1, tzinfo=NiceTry)
-
-        class BetterTry(tzinfo):
-            def __init__(self): pass
-            def utcoffset(self, dt): pass
-        b = BetterTry()
-        t = cls(1, 1, 1, tzinfo=b)
-        self.assertTrue(t.tzinfo is b)
-
-    def test_utc_offset_out_of_bounds(self):
-        class Edgy(tzinfo):
-            def __init__(self, offset):
-                self.offset = timedelta(minutes=offset)
-            def utcoffset(self, dt):
-                return self.offset
-
-        cls = self.theclass
-        for offset, legit in ((-1440, False),
-                              (-1439, True),
-                              (1439, True),
-                              (1440, False)):
-            if cls is time:
-                t = cls(1, 2, 3, tzinfo=Edgy(offset))
-            elif cls is datetime:
-                t = cls(6, 6, 6, 1, 2, 3, tzinfo=Edgy(offset))
-            else:
-                assert 0, "impossible"
-            if legit:
-                aofs = abs(offset)
-                h, m = divmod(aofs, 60)
-                tag = "%c%02d:%02d" % (offset < 0 and '-' or '+', h, m)
-                if isinstance(t, datetime):
-                    t = t.timetz()
-                self.assertEqual(str(t), "01:02:03" + tag)
-            else:
-                self.assertRaises(ValueError, str, t)
-
-    def test_tzinfo_classes(self):
-        cls = self.theclass
-        class C1(tzinfo):
-            def utcoffset(self, dt): return None
-            def dst(self, dt): return None
-            def tzname(self, dt): return None
-        for t in (cls(1, 1, 1),
-                  cls(1, 1, 1, tzinfo=None),
-                  cls(1, 1, 1, tzinfo=C1())):
-            self.assertTrue(t.utcoffset() is None)
-            self.assertTrue(t.dst() is None)
-            self.assertTrue(t.tzname() is None)
-
-        class C3(tzinfo):
-            def utcoffset(self, dt): return timedelta(minutes=-1439)
-            def dst(self, dt): return timedelta(minutes=1439)
-            def tzname(self, dt): return "aname"
-        t = cls(1, 1, 1, tzinfo=C3())
-        self.assertEqual(t.utcoffset(), timedelta(minutes=-1439))
-        self.assertEqual(t.dst(), timedelta(minutes=1439))
-        self.assertEqual(t.tzname(), "aname")
-
-        # Wrong types.
-        class C4(tzinfo):
-            def utcoffset(self, dt): return "aname"
-            def dst(self, dt): return 7
-            def tzname(self, dt): return 0
-        t = cls(1, 1, 1, tzinfo=C4())
-        self.assertRaises(TypeError, t.utcoffset)
-        self.assertRaises(TypeError, t.dst)
-        self.assertRaises(TypeError, t.tzname)
-
-        # Offset out of range.
-        class C6(tzinfo):
-            def utcoffset(self, dt): return timedelta(hours=-24)
-            def dst(self, dt): return timedelta(hours=24)
-        t = cls(1, 1, 1, tzinfo=C6())
-        self.assertRaises(ValueError, t.utcoffset)
-        self.assertRaises(ValueError, t.dst)
-
-        # Not a whole number of minutes.
-        class C7(tzinfo):
-            def utcoffset(self, dt): return timedelta(seconds=61)
-            def dst(self, dt): return timedelta(microseconds=-81)
-        t = cls(1, 1, 1, tzinfo=C7())
-        self.assertRaises(ValueError, t.utcoffset)
-        self.assertRaises(ValueError, t.dst)
-
-    def test_aware_compare(self):
-        cls = self.theclass
-
-        # Ensure that utcoffset() gets ignored if the comparands have
-        # the same tzinfo member.
-        class OperandDependentOffset(tzinfo):
-            def utcoffset(self, t):
-                if t.minute < 10:
-                    # d0 and d1 equal after adjustment
-                    return timedelta(minutes=t.minute)
-                else:
-                    # d2 off in the weeds
-                    return timedelta(minutes=59)
-
-        base = cls(8, 9, 10, tzinfo=OperandDependentOffset())
-        d0 = base.replace(minute=3)
-        d1 = base.replace(minute=9)
-        d2 = base.replace(minute=11)
-        for x in d0, d1, d2:
-            for y in d0, d1, d2:
-                for op in lt, le, gt, ge, eq, ne:
-                    got = op(x, y)
-                    expected = op(x.minute, y.minute)
-                    self.assertEqual(got, expected)
-
-        # However, if they're different members, uctoffset is not ignored.
-        # Note that a time can't actually have an operand-depedent offset,
-        # though (and time.utcoffset() passes None to tzinfo.utcoffset()),
-        # so skip this test for time.
-        if cls is not time:
-            d0 = base.replace(minute=3, tzinfo=OperandDependentOffset())
-            d1 = base.replace(minute=9, tzinfo=OperandDependentOffset())
-            d2 = base.replace(minute=11, tzinfo=OperandDependentOffset())
-            for x in d0, d1, d2:
-                for y in d0, d1, d2:
-                    got = (x > y) - (x < y)
-                    if (x is d0 or x is d1) and (y is d0 or y is d1):
-                        expected = 0
-                    elif x is y is d2:
-                        expected = 0
-                    elif x is d2:
-                        expected = -1
-                    else:
-                        assert y is d2
-                        expected = 1
-                    self.assertEqual(got, expected)
-
-
-# Testing time objects with a non-None tzinfo.
-class TestTimeTZ(TestTime, TZInfoBase, unittest.TestCase):
-    theclass = time
-
-    def test_empty(self):
-        t = self.theclass()
-        self.assertEqual(t.hour, 0)
-        self.assertEqual(t.minute, 0)
-        self.assertEqual(t.second, 0)
-        self.assertEqual(t.microsecond, 0)
-        self.assertTrue(t.tzinfo is None)
-
-    def test_zones(self):
-        est = FixedOffset(-300, "EST", 1)
-        utc = FixedOffset(0, "UTC", -2)
-        met = FixedOffset(60, "MET", 3)
-        t1 = time( 7, 47, tzinfo=est)
-        t2 = time(12, 47, tzinfo=utc)
-        t3 = time(13, 47, tzinfo=met)
-        t4 = time(microsecond=40)
-        t5 = time(microsecond=40, tzinfo=utc)
-
-        self.assertEqual(t1.tzinfo, est)
-        self.assertEqual(t2.tzinfo, utc)
-        self.assertEqual(t3.tzinfo, met)
-        self.assertTrue(t4.tzinfo is None)
-        self.assertEqual(t5.tzinfo, utc)
-
-        self.assertEqual(t1.utcoffset(), timedelta(minutes=-300))
-        self.assertEqual(t2.utcoffset(), timedelta(minutes=0))
-        self.assertEqual(t3.utcoffset(), timedelta(minutes=60))
-        self.assertTrue(t4.utcoffset() is None)
-        self.assertRaises(TypeError, t1.utcoffset, "no args")
-
-        self.assertEqual(t1.tzname(), "EST")
-        self.assertEqual(t2.tzname(), "UTC")
-        self.assertEqual(t3.tzname(), "MET")
-        self.assertTrue(t4.tzname() is None)
-        self.assertRaises(TypeError, t1.tzname, "no args")
-
-        self.assertEqual(t1.dst(), timedelta(minutes=1))
-        self.assertEqual(t2.dst(), timedelta(minutes=-2))
-        self.assertEqual(t3.dst(), timedelta(minutes=3))
-        self.assertTrue(t4.dst() is None)
-        self.assertRaises(TypeError, t1.dst, "no args")
-
-        self.assertEqual(hash(t1), hash(t2))
-        self.assertEqual(hash(t1), hash(t3))
-        self.assertEqual(hash(t2), hash(t3))
-
-        self.assertEqual(t1, t2)
-        self.assertEqual(t1, t3)
-        self.assertEqual(t2, t3)
-        self.assertRaises(TypeError, lambda: t4 == t5) # mixed tz-aware & naive
-        self.assertRaises(TypeError, lambda: t4 < t5) # mixed tz-aware & naive
-        self.assertRaises(TypeError, lambda: t5 < t4) # mixed tz-aware & naive
-
-        self.assertEqual(str(t1), "07:47:00-05:00")
-        self.assertEqual(str(t2), "12:47:00+00:00")
-        self.assertEqual(str(t3), "13:47:00+01:00")
-        self.assertEqual(str(t4), "00:00:00.000040")
-        self.assertEqual(str(t5), "00:00:00.000040+00:00")
-
-        self.assertEqual(t1.isoformat(), "07:47:00-05:00")
-        self.assertEqual(t2.isoformat(), "12:47:00+00:00")
-        self.assertEqual(t3.isoformat(), "13:47:00+01:00")
-        self.assertEqual(t4.isoformat(), "00:00:00.000040")
-        self.assertEqual(t5.isoformat(), "00:00:00.000040+00:00")
-
-        d = 'datetime.time'
-        self.assertEqual(repr(t1), d + "(7, 47, tzinfo=est)")
-        self.assertEqual(repr(t2), d + "(12, 47, tzinfo=utc)")
-        self.assertEqual(repr(t3), d + "(13, 47, tzinfo=met)")
-        self.assertEqual(repr(t4), d + "(0, 0, 0, 40)")
-        self.assertEqual(repr(t5), d + "(0, 0, 0, 40, tzinfo=utc)")
-
-        self.assertEqual(t1.strftime("%H:%M:%S %%Z=%Z %%z=%z"),
-                                     "07:47:00 %Z=EST %z=-0500")
-        self.assertEqual(t2.strftime("%H:%M:%S %Z %z"), "12:47:00 UTC +0000")
-        self.assertEqual(t3.strftime("%H:%M:%S %Z %z"), "13:47:00 MET +0100")
-
-        yuck = FixedOffset(-1439, "%z %Z %%z%%Z")
-        t1 = time(23, 59, tzinfo=yuck)
-        self.assertEqual(t1.strftime("%H:%M %%Z='%Z' %%z='%z'"),
-                                     "23:59 %Z='%z %Z %%z%%Z' %z='-2359'")
-
-        # Check that an invalid tzname result raises an exception.
-        class Badtzname(tzinfo):
-            def tzname(self, dt): return 42
-        t = time(2, 3, 4, tzinfo=Badtzname())
-        self.assertEqual(t.strftime("%H:%M:%S"), "02:03:04")
-        self.assertRaises(TypeError, t.strftime, "%Z")
-
-    def test_hash_edge_cases(self):
-        # Offsets that overflow a basic time.
-        t1 = self.theclass(0, 1, 2, 3, tzinfo=FixedOffset(1439, ""))
-        t2 = self.theclass(0, 0, 2, 3, tzinfo=FixedOffset(1438, ""))
-        self.assertEqual(hash(t1), hash(t2))
-
-        t1 = self.theclass(23, 58, 6, 100, tzinfo=FixedOffset(-1000, ""))
-        t2 = self.theclass(23, 48, 6, 100, tzinfo=FixedOffset(-1010, ""))
-        self.assertEqual(hash(t1), hash(t2))
-
-    def test_pickling(self):
-        # Try one without a tzinfo.
-        args = 20, 59, 16, 64**2
-        orig = self.theclass(*args)
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertEqual(orig, derived)
-
-        # Try one with a tzinfo.
-        tinfo = PicklableFixedOffset(-300, 'cookie')
-        orig = self.theclass(5, 6, 7, tzinfo=tinfo)
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertEqual(orig, derived)
-            self.assertIsInstance(derived.tzinfo, PicklableFixedOffset)
-            self.assertEqual(derived.utcoffset(), timedelta(minutes=-300))
-            self.assertEqual(derived.tzname(), 'cookie')
-
-    def test_more_bool(self):
-        # Test cases with non-None tzinfo.
-        cls = self.theclass
-
-        t = cls(0, tzinfo=FixedOffset(-300, ""))
-        self.assertTrue(t)
-
-        t = cls(5, tzinfo=FixedOffset(-300, ""))
-        self.assertTrue(t)
-
-        t = cls(5, tzinfo=FixedOffset(300, ""))
-        self.assertTrue(not t)
-
-        t = cls(23, 59, tzinfo=FixedOffset(23*60 + 59, ""))
-        self.assertTrue(not t)
-
-        # Mostly ensuring this doesn't overflow internally.
-        t = cls(0, tzinfo=FixedOffset(23*60 + 59, ""))
-        self.assertTrue(t)
-
-        # But this should yield a value error -- the utcoffset is bogus.
-        t = cls(0, tzinfo=FixedOffset(24*60, ""))
-        self.assertRaises(ValueError, lambda: bool(t))
-
-        # Likewise.
-        t = cls(0, tzinfo=FixedOffset(-24*60, ""))
-        self.assertRaises(ValueError, lambda: bool(t))
-
-    def test_replace(self):
-        cls = self.theclass
-        z100 = FixedOffset(100, "+100")
-        zm200 = FixedOffset(timedelta(minutes=-200), "-200")
-        args = [1, 2, 3, 4, z100]
-        base = cls(*args)
-        self.assertEqual(base, base.replace())
-
-        i = 0
-        for name, newval in (("hour", 5),
-                             ("minute", 6),
-                             ("second", 7),
-                             ("microsecond", 8),
-                             ("tzinfo", zm200)):
-            newargs = args[:]
-            newargs[i] = newval
-            expected = cls(*newargs)
-            got = base.replace(**{name: newval})
-            self.assertEqual(expected, got)
-            i += 1
-
-        # Ensure we can get rid of a tzinfo.
-        self.assertEqual(base.tzname(), "+100")
-        base2 = base.replace(tzinfo=None)
-        self.assertTrue(base2.tzinfo is None)
-        self.assertTrue(base2.tzname() is None)
-
-        # Ensure we can add one.
-        base3 = base2.replace(tzinfo=z100)
-        self.assertEqual(base, base3)
-        self.assertTrue(base.tzinfo is base3.tzinfo)
-
-        # Out of bounds.
-        base = cls(1)
-        self.assertRaises(ValueError, base.replace, hour=24)
-        self.assertRaises(ValueError, base.replace, minute=-1)
-        self.assertRaises(ValueError, base.replace, second=100)
-        self.assertRaises(ValueError, base.replace, microsecond=1000000)
-
-    def test_mixed_compare(self):
-        t1 = time(1, 2, 3)
-        t2 = time(1, 2, 3)
-        self.assertEqual(t1, t2)
-        t2 = t2.replace(tzinfo=None)
-        self.assertEqual(t1, t2)
-        t2 = t2.replace(tzinfo=FixedOffset(None, ""))
-        self.assertEqual(t1, t2)
-        t2 = t2.replace(tzinfo=FixedOffset(0, ""))
-        self.assertRaises(TypeError, lambda: t1 == t2)
-
-        # In time w/ identical tzinfo objects, utcoffset is ignored.
-        class Varies(tzinfo):
-            def __init__(self):
-                self.offset = timedelta(minutes=22)
-            def utcoffset(self, t):
-                self.offset += timedelta(minutes=1)
-                return self.offset
-
-        v = Varies()
-        t1 = t2.replace(tzinfo=v)
-        t2 = t2.replace(tzinfo=v)
-        self.assertEqual(t1.utcoffset(), timedelta(minutes=23))
-        self.assertEqual(t2.utcoffset(), timedelta(minutes=24))
-        self.assertEqual(t1, t2)
-
-        # But if they're not identical, it isn't ignored.
-        t2 = t2.replace(tzinfo=Varies())
-        self.assertTrue(t1 < t2)  # t1's offset counter still going up
-
-    def test_subclass_timetz(self):
-
-        class C(self.theclass):
-            theAnswer = 42
-
-            def __new__(cls, *args, **kws):
-                temp = kws.copy()
-                extra = temp.pop('extra')
-                result = self.theclass.__new__(cls, *args, **temp)
-                result.extra = extra
-                return result
-
-            def newmeth(self, start):
-                return start + self.hour + self.second
-
-        args = 4, 5, 6, 500, FixedOffset(-300, "EST", 1)
-
-        dt1 = self.theclass(*args)
-        dt2 = C(*args, **{'extra': 7})
-
-        self.assertEqual(dt2.__class__, C)
-        self.assertEqual(dt2.theAnswer, 42)
-        self.assertEqual(dt2.extra, 7)
-        self.assertEqual(dt1.utcoffset(), dt2.utcoffset())
-        self.assertEqual(dt2.newmeth(-7), dt1.hour + dt1.second - 7)
-
-
-# Testing datetime objects with a non-None tzinfo.
-
-class TestDateTimeTZ(TestDateTime, TZInfoBase, unittest.TestCase):
-    theclass = datetime
-
-    def test_trivial(self):
-        dt = self.theclass(1, 2, 3, 4, 5, 6, 7)
-        self.assertEqual(dt.year, 1)
-        self.assertEqual(dt.month, 2)
-        self.assertEqual(dt.day, 3)
-        self.assertEqual(dt.hour, 4)
-        self.assertEqual(dt.minute, 5)
-        self.assertEqual(dt.second, 6)
-        self.assertEqual(dt.microsecond, 7)
-        self.assertEqual(dt.tzinfo, None)
-
-    def test_even_more_compare(self):
-        # The test_compare() and test_more_compare() inherited from TestDate
-        # and TestDateTime covered non-tzinfo cases.
-
-        # Smallest possible after UTC adjustment.
-        t1 = self.theclass(1, 1, 1, tzinfo=FixedOffset(1439, ""))
-        # Largest possible after UTC adjustment.
-        t2 = self.theclass(MAXYEAR, 12, 31, 23, 59, 59, 999999,
-                           tzinfo=FixedOffset(-1439, ""))
-
-        # Make sure those compare correctly, and w/o overflow.
-        self.assertTrue(t1 < t2)
-        self.assertTrue(t1 != t2)
-        self.assertTrue(t2 > t1)
-
-        self.assertEqual(t1, t1)
-        self.assertEqual(t2, t2)
-
-        # Equal afer adjustment.
-        t1 = self.theclass(1, 12, 31, 23, 59, tzinfo=FixedOffset(1, ""))
-        t2 = self.theclass(2, 1, 1, 3, 13, tzinfo=FixedOffset(3*60+13+2, ""))
-        self.assertEqual(t1, t2)
-
-        # Change t1 not to subtract a minute, and t1 should be larger.
-        t1 = self.theclass(1, 12, 31, 23, 59, tzinfo=FixedOffset(0, ""))
-        self.assertTrue(t1 > t2)
-
-        # Change t1 to subtract 2 minutes, and t1 should be smaller.
-        t1 = self.theclass(1, 12, 31, 23, 59, tzinfo=FixedOffset(2, ""))
-        self.assertTrue(t1 < t2)
-
-        # Back to the original t1, but make seconds resolve it.
-        t1 = self.theclass(1, 12, 31, 23, 59, tzinfo=FixedOffset(1, ""),
-                           second=1)
-        self.assertTrue(t1 > t2)
-
-        # Likewise, but make microseconds resolve it.
-        t1 = self.theclass(1, 12, 31, 23, 59, tzinfo=FixedOffset(1, ""),
-                           microsecond=1)
-        self.assertTrue(t1 > t2)
-
-        # Make t2 naive and it should fail.
-        t2 = self.theclass.min
-        self.assertRaises(TypeError, lambda: t1 == t2)
-        self.assertEqual(t2, t2)
-
-        # It's also naive if it has tzinfo but tzinfo.utcoffset() is None.
-        class Naive(tzinfo):
-            def utcoffset(self, dt): return None
-        t2 = self.theclass(5, 6, 7, tzinfo=Naive())
-        self.assertRaises(TypeError, lambda: t1 == t2)
-        self.assertEqual(t2, t2)
-
-        # OTOH, it's OK to compare two of these mixing the two ways of being
-        # naive.
-        t1 = self.theclass(5, 6, 7)
-        self.assertEqual(t1, t2)
-
-        # Try a bogus uctoffset.
-        class Bogus(tzinfo):
-            def utcoffset(self, dt):
-                return timedelta(minutes=1440) # out of bounds
-        t1 = self.theclass(2, 2, 2, tzinfo=Bogus())
-        t2 = self.theclass(2, 2, 2, tzinfo=FixedOffset(0, ""))
-        self.assertRaises(ValueError, lambda: t1 == t2)
-
-    def test_pickling(self):
-        # Try one without a tzinfo.
-        args = 6, 7, 23, 20, 59, 1, 64**2
-        orig = self.theclass(*args)
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertEqual(orig, derived)
-
-        # Try one with a tzinfo.
-        tinfo = PicklableFixedOffset(-300, 'cookie')
-        orig = self.theclass(*args, **{'tzinfo': tinfo})
-        derived = self.theclass(1, 1, 1, tzinfo=FixedOffset(0, "", 0))
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertEqual(orig, derived)
-            self.assertIsInstance(derived.tzinfo, PicklableFixedOffset)
-            self.assertEqual(derived.utcoffset(), timedelta(minutes=-300))
-            self.assertEqual(derived.tzname(), 'cookie')
-
-    def test_extreme_hashes(self):
-        # If an attempt is made to hash these via subtracting the offset
-        # then hashing a datetime object, OverflowError results.  The
-        # Python implementation used to blow up here.
-        t = self.theclass(1, 1, 1, tzinfo=FixedOffset(1439, ""))
-        hash(t)
-        t = self.theclass(MAXYEAR, 12, 31, 23, 59, 59, 999999,
-                          tzinfo=FixedOffset(-1439, ""))
-        hash(t)
-
-        # OTOH, an OOB offset should blow up.
-        t = self.theclass(5, 5, 5, tzinfo=FixedOffset(-1440, ""))
-        self.assertRaises(ValueError, hash, t)
-
-    def test_zones(self):
-        est = FixedOffset(-300, "EST")
-        utc = FixedOffset(0, "UTC")
-        met = FixedOffset(60, "MET")
-        t1 = datetime(2002, 3, 19,  7, 47, tzinfo=est)
-        t2 = datetime(2002, 3, 19, 12, 47, tzinfo=utc)
-        t3 = datetime(2002, 3, 19, 13, 47, tzinfo=met)
-        self.assertEqual(t1.tzinfo, est)
-        self.assertEqual(t2.tzinfo, utc)
-        self.assertEqual(t3.tzinfo, met)
-        self.assertEqual(t1.utcoffset(), timedelta(minutes=-300))
-        self.assertEqual(t2.utcoffset(), timedelta(minutes=0))
-        self.assertEqual(t3.utcoffset(), timedelta(minutes=60))
-        self.assertEqual(t1.tzname(), "EST")
-        self.assertEqual(t2.tzname(), "UTC")
-        self.assertEqual(t3.tzname(), "MET")
-        self.assertEqual(hash(t1), hash(t2))
-        self.assertEqual(hash(t1), hash(t3))
-        self.assertEqual(hash(t2), hash(t3))
-        self.assertEqual(t1, t2)
-        self.assertEqual(t1, t3)
-        self.assertEqual(t2, t3)
-        self.assertEqual(str(t1), "2002-03-19 07:47:00-05:00")
-        self.assertEqual(str(t2), "2002-03-19 12:47:00+00:00")
-        self.assertEqual(str(t3), "2002-03-19 13:47:00+01:00")
-        d = 'datetime.datetime(2002, 3, 19, '
-        self.assertEqual(repr(t1), d + "7, 47, tzinfo=est)")
-        self.assertEqual(repr(t2), d + "12, 47, tzinfo=utc)")
-        self.assertEqual(repr(t3), d + "13, 47, tzinfo=met)")
-
-    def test_combine(self):
-        met = FixedOffset(60, "MET")
-        d = date(2002, 3, 4)
-        tz = time(18, 45, 3, 1234, tzinfo=met)
-        dt = datetime.combine(d, tz)
-        self.assertEqual(dt, datetime(2002, 3, 4, 18, 45, 3, 1234,
-                                        tzinfo=met))
-
-    def test_extract(self):
-        met = FixedOffset(60, "MET")
-        dt = self.theclass(2002, 3, 4, 18, 45, 3, 1234, tzinfo=met)
-        self.assertEqual(dt.date(), date(2002, 3, 4))
-        self.assertEqual(dt.time(), time(18, 45, 3, 1234))
-        self.assertEqual(dt.timetz(), time(18, 45, 3, 1234, tzinfo=met))
-
-    def test_tz_aware_arithmetic(self):
-        import random
-
-        now = self.theclass.now()
-        tz55 = FixedOffset(-330, "west 5:30")
-        timeaware = now.time().replace(tzinfo=tz55)
-        nowaware = self.theclass.combine(now.date(), timeaware)
-        self.assertTrue(nowaware.tzinfo is tz55)
-        self.assertEqual(nowaware.timetz(), timeaware)
-
-        # Can't mix aware and non-aware.
-        self.assertRaises(TypeError, lambda: now - nowaware)
-        self.assertRaises(TypeError, lambda: nowaware - now)
-
-        # And adding datetime's doesn't make sense, aware or not.
-        self.assertRaises(TypeError, lambda: now + nowaware)
-        self.assertRaises(TypeError, lambda: nowaware + now)
-        self.assertRaises(TypeError, lambda: nowaware + nowaware)
-
-        # Subtracting should yield 0.
-        self.assertEqual(now - now, timedelta(0))
-        self.assertEqual(nowaware - nowaware, timedelta(0))
-
-        # Adding a delta should preserve tzinfo.
-        delta = timedelta(weeks=1, minutes=12, microseconds=5678)
-        nowawareplus = nowaware + delta
-        self.assertTrue(nowaware.tzinfo is tz55)
-        nowawareplus2 = delta + nowaware
-        self.assertTrue(nowawareplus2.tzinfo is tz55)
-        self.assertEqual(nowawareplus, nowawareplus2)
-
-        # that - delta should be what we started with, and that - what we
-        # started with should be delta.
-        diff = nowawareplus - delta
-        self.assertTrue(diff.tzinfo is tz55)
-        self.assertEqual(nowaware, diff)
-        self.assertRaises(TypeError, lambda: delta - nowawareplus)
-        self.assertEqual(nowawareplus - nowaware, delta)
-
-        # Make up a random timezone.
-        tzr = FixedOffset(random.randrange(-1439, 1440), "randomtimezone")
-        # Attach it to nowawareplus.
-        nowawareplus = nowawareplus.replace(tzinfo=tzr)
-        self.assertTrue(nowawareplus.tzinfo is tzr)
-        # Make sure the difference takes the timezone adjustments into account.
-        got = nowaware - nowawareplus
-        # Expected:  (nowaware base - nowaware offset) -
-        #            (nowawareplus base - nowawareplus offset) =
-        #            (nowaware base - nowawareplus base) +
-        #            (nowawareplus offset - nowaware offset) =
-        #            -delta + nowawareplus offset - nowaware offset
-        expected = nowawareplus.utcoffset() - nowaware.utcoffset() - delta
-        self.assertEqual(got, expected)
-
-        # Try max possible difference.
-        min = self.theclass(1, 1, 1, tzinfo=FixedOffset(1439, "min"))
-        max = self.theclass(MAXYEAR, 12, 31, 23, 59, 59, 999999,
-                            tzinfo=FixedOffset(-1439, "max"))
-        maxdiff = max - min
-        self.assertEqual(maxdiff, self.theclass.max - self.theclass.min +
-                                  timedelta(minutes=2*1439))
-        # Different tzinfo, but the same offset
-        tza = timezone(HOUR, 'A')
-        tzb = timezone(HOUR, 'B')
-        delta = min.replace(tzinfo=tza) - max.replace(tzinfo=tzb)
-        self.assertEqual(delta, self.theclass.min - self.theclass.max)
-
-    def test_tzinfo_now(self):
-        meth = self.theclass.now
-        # Ensure it doesn't require tzinfo (i.e., that this doesn't blow up).
-        base = meth()
-        # Try with and without naming the keyword.
-        off42 = FixedOffset(42, "42")
-        another = meth(off42)
-        again = meth(tz=off42)
-        self.assertTrue(another.tzinfo is again.tzinfo)
-        self.assertEqual(another.utcoffset(), timedelta(minutes=42))
-        # Bad argument with and w/o naming the keyword.
-        self.assertRaises(TypeError, meth, 16)
-        self.assertRaises(TypeError, meth, tzinfo=16)
-        # Bad keyword name.
-        self.assertRaises(TypeError, meth, tinfo=off42)
-        # Too many args.
-        self.assertRaises(TypeError, meth, off42, off42)
-
-        # We don't know which time zone we're in, and don't have a tzinfo
-        # class to represent it, so seeing whether a tz argument actually
-        # does a conversion is tricky.
-        utc = FixedOffset(0, "utc", 0)
-        for weirdtz in [FixedOffset(timedelta(hours=15, minutes=58), "weirdtz", 0),
-                        timezone(timedelta(hours=15, minutes=58), "weirdtz"),]:
-            for dummy in range(3):
-                now = datetime.now(weirdtz)
-                self.assertTrue(now.tzinfo is weirdtz)
-                utcnow = datetime.utcnow().replace(tzinfo=utc)
-                now2 = utcnow.astimezone(weirdtz)
-                if abs(now - now2) < timedelta(seconds=30):
-                    break
-                # Else the code is broken, or more than 30 seconds passed between
-                # calls; assuming the latter, just try again.
-            else:
-                # Three strikes and we're out.
-                self.fail("utcnow(), now(tz), or astimezone() may be broken")
-
-    def test_tzinfo_fromtimestamp(self):
-        import time
-        meth = self.theclass.fromtimestamp
-        ts = time.time()
-        # Ensure it doesn't require tzinfo (i.e., that this doesn't blow up).
-        base = meth(ts)
-        # Try with and without naming the keyword.
-        off42 = FixedOffset(42, "42")
-        another = meth(ts, off42)
-        again = meth(ts, tz=off42)
-        self.assertTrue(another.tzinfo is again.tzinfo)
-        self.assertEqual(another.utcoffset(), timedelta(minutes=42))
-        # Bad argument with and w/o naming the keyword.
-        self.assertRaises(TypeError, meth, ts, 16)
-        self.assertRaises(TypeError, meth, ts, tzinfo=16)
-        # Bad keyword name.
-        self.assertRaises(TypeError, meth, ts, tinfo=off42)
-        # Too many args.
-        self.assertRaises(TypeError, meth, ts, off42, off42)
-        # Too few args.
-        self.assertRaises(TypeError, meth)
-
-        # Try to make sure tz= actually does some conversion.
-        timestamp = 1000000000
-        utcdatetime = datetime.utcfromtimestamp(timestamp)
-        # In POSIX (epoch 1970), that's 2001-09-09 01:46:40 UTC, give or take.
-        # But on some flavor of Mac, it's nowhere near that.  So we can't have
-        # any idea here what time that actually is, we can only test that
-        # relative changes match.
-        utcoffset = timedelta(hours=-15, minutes=39) # arbitrary, but not zero
-        tz = FixedOffset(utcoffset, "tz", 0)
-        expected = utcdatetime + utcoffset
-        got = datetime.fromtimestamp(timestamp, tz)
-        self.assertEqual(expected, got.replace(tzinfo=None))
-
-    def test_tzinfo_utcnow(self):
-        meth = self.theclass.utcnow
-        # Ensure it doesn't require tzinfo (i.e., that this doesn't blow up).
-        base = meth()
-        # Try with and without naming the keyword; for whatever reason,
-        # utcnow() doesn't accept a tzinfo argument.
-        off42 = FixedOffset(42, "42")
-        self.assertRaises(TypeError, meth, off42)
-        self.assertRaises(TypeError, meth, tzinfo=off42)
-
-    def test_tzinfo_utcfromtimestamp(self):
-        import time
-        meth = self.theclass.utcfromtimestamp
-        ts = time.time()
-        # Ensure it doesn't require tzinfo (i.e., that this doesn't blow up).
-        base = meth(ts)
-        # Try with and without naming the keyword; for whatever reason,
-        # utcfromtimestamp() doesn't accept a tzinfo argument.
-        off42 = FixedOffset(42, "42")
-        self.assertRaises(TypeError, meth, ts, off42)
-        self.assertRaises(TypeError, meth, ts, tzinfo=off42)
-
-    def test_tzinfo_timetuple(self):
-        # TestDateTime tested most of this.  datetime adds a twist to the
-        # DST flag.
-        class DST(tzinfo):
-            def __init__(self, dstvalue):
-                if isinstance(dstvalue, int):
-                    dstvalue = timedelta(minutes=dstvalue)
-                self.dstvalue = dstvalue
-            def dst(self, dt):
-                return self.dstvalue
-
-        cls = self.theclass
-        for dstvalue, flag in (-33, 1), (33, 1), (0, 0), (None, -1):
-            d = cls(1, 1, 1, 10, 20, 30, 40, tzinfo=DST(dstvalue))
-            t = d.timetuple()
-            self.assertEqual(1, t.tm_year)
-            self.assertEqual(1, t.tm_mon)
-            self.assertEqual(1, t.tm_mday)
-            self.assertEqual(10, t.tm_hour)
-            self.assertEqual(20, t.tm_min)
-            self.assertEqual(30, t.tm_sec)
-            self.assertEqual(0, t.tm_wday)
-            self.assertEqual(1, t.tm_yday)
-            self.assertEqual(flag, t.tm_isdst)
-
-        # dst() returns wrong type.
-        self.assertRaises(TypeError, cls(1, 1, 1, tzinfo=DST("x")).timetuple)
-
-        # dst() at the edge.
-        self.assertEqual(cls(1,1,1, tzinfo=DST(1439)).timetuple().tm_isdst, 1)
-        self.assertEqual(cls(1,1,1, tzinfo=DST(-1439)).timetuple().tm_isdst, 1)
-
-        # dst() out of range.
-        self.assertRaises(ValueError, cls(1,1,1, tzinfo=DST(1440)).timetuple)
-        self.assertRaises(ValueError, cls(1,1,1, tzinfo=DST(-1440)).timetuple)
-
-    def test_utctimetuple(self):
-        class DST(tzinfo):
-            def __init__(self, dstvalue=0):
-                if isinstance(dstvalue, int):
-                    dstvalue = timedelta(minutes=dstvalue)
-                self.dstvalue = dstvalue
-            def dst(self, dt):
-                return self.dstvalue
-
-        cls = self.theclass
-        # This can't work:  DST didn't implement utcoffset.
-        self.assertRaises(NotImplementedError,
-                          cls(1, 1, 1, tzinfo=DST(0)).utcoffset)
-
-        class UOFS(DST):
-            def __init__(self, uofs, dofs=None):
-                DST.__init__(self, dofs)
-                self.uofs = timedelta(minutes=uofs)
-            def utcoffset(self, dt):
-                return self.uofs
-
-        for dstvalue in -33, 33, 0, None:
-            d = cls(1, 2, 3, 10, 20, 30, 40, tzinfo=UOFS(-53, dstvalue))
-            t = d.utctimetuple()
-            self.assertEqual(d.year, t.tm_year)
-            self.assertEqual(d.month, t.tm_mon)
-            self.assertEqual(d.day, t.tm_mday)
-            self.assertEqual(11, t.tm_hour) # 20mm + 53mm = 1hn + 13mm
-            self.assertEqual(13, t.tm_min)
-            self.assertEqual(d.second, t.tm_sec)
-            self.assertEqual(d.weekday(), t.tm_wday)
-            self.assertEqual(d.toordinal() - date(1, 1, 1).toordinal() + 1,
-                             t.tm_yday)
-            # Ensure tm_isdst is 0 regardless of what dst() says: DST
-            # is never in effect for a UTC time.
-            self.assertEqual(0, t.tm_isdst)
-
-        # For naive datetime, utctimetuple == timetuple except for isdst
-        d = cls(1, 2, 3, 10, 20, 30, 40)
-        t = d.utctimetuple()
-        self.assertEqual(t[:-1], d.timetuple()[:-1])
-        self.assertEqual(0, t.tm_isdst)
-        # Same if utcoffset is None
-        class NOFS(DST):
-            def utcoffset(self, dt):
-                return None
-        d = cls(1, 2, 3, 10, 20, 30, 40, tzinfo=NOFS())
-        t = d.utctimetuple()
-        self.assertEqual(t[:-1], d.timetuple()[:-1])
-        self.assertEqual(0, t.tm_isdst)
-        # Check that bad tzinfo is detected
-        class BOFS(DST):
-            def utcoffset(self, dt):
-                return "EST"
-        d = cls(1, 2, 3, 10, 20, 30, 40, tzinfo=BOFS())
-        self.assertRaises(TypeError, d.utctimetuple)
-
-        # Check that utctimetuple() is the same as
-        # astimezone(utc).timetuple()
-        d = cls(2010, 11, 13, 14, 15, 16, 171819)
-        for tz in [timezone.min, timezone.utc, timezone.max]:
-            dtz = d.replace(tzinfo=tz)
-            self.assertEqual(dtz.utctimetuple()[:-1],
-                             dtz.astimezone(timezone.utc).timetuple()[:-1])
-        # At the edges, UTC adjustment can produce years out-of-range
-        # for a datetime object.  Ensure that an OverflowError is
-        # raised.
-        tiny = cls(MINYEAR, 1, 1, 0, 0, 37, tzinfo=UOFS(1439))
-        # That goes back 1 minute less than a full day.
-        self.assertRaises(OverflowError, tiny.utctimetuple)
-
-        huge = cls(MAXYEAR, 12, 31, 23, 59, 37, 999999, tzinfo=UOFS(-1439))
-        # That goes forward 1 minute less than a full day.
-        self.assertRaises(OverflowError, huge.utctimetuple)
-        # More overflow cases
-        tiny = cls.min.replace(tzinfo=timezone(MINUTE))
-        self.assertRaises(OverflowError, tiny.utctimetuple)
-        huge = cls.max.replace(tzinfo=timezone(-MINUTE))
-        self.assertRaises(OverflowError, huge.utctimetuple)
-
-    def test_tzinfo_isoformat(self):
-        zero = FixedOffset(0, "+00:00")
-        plus = FixedOffset(220, "+03:40")
-        minus = FixedOffset(-231, "-03:51")
-        unknown = FixedOffset(None, "")
-
-        cls = self.theclass
-        datestr = '0001-02-03'
-        for ofs in None, zero, plus, minus, unknown:
-            for us in 0, 987001:
-                d = cls(1, 2, 3, 4, 5, 59, us, tzinfo=ofs)
-                timestr = '04:05:59' + (us and '.987001' or '')
-                ofsstr = ofs is not None and d.tzname() or ''
-                tailstr = timestr + ofsstr
-                iso = d.isoformat()
-                self.assertEqual(iso, datestr + 'T' + tailstr)
-                self.assertEqual(iso, d.isoformat('T'))
-                self.assertEqual(d.isoformat('k'), datestr + 'k' + tailstr)
-                self.assertEqual(d.isoformat('\u1234'), datestr + '\u1234' + tailstr)
-                self.assertEqual(str(d), datestr + ' ' + tailstr)
-
-    def test_replace(self):
-        cls = self.theclass
-        z100 = FixedOffset(100, "+100")
-        zm200 = FixedOffset(timedelta(minutes=-200), "-200")
-        args = [1, 2, 3, 4, 5, 6, 7, z100]
-        base = cls(*args)
-        self.assertEqual(base, base.replace())
-
-        i = 0
-        for name, newval in (("year", 2),
-                             ("month", 3),
-                             ("day", 4),
-                             ("hour", 5),
-                             ("minute", 6),
-                             ("second", 7),
-                             ("microsecond", 8),
-                             ("tzinfo", zm200)):
-            newargs = args[:]
-            newargs[i] = newval
-            expected = cls(*newargs)
-            got = base.replace(**{name: newval})
-            self.assertEqual(expected, got)
-            i += 1
-
-        # Ensure we can get rid of a tzinfo.
-        self.assertEqual(base.tzname(), "+100")
-        base2 = base.replace(tzinfo=None)
-        self.assertTrue(base2.tzinfo is None)
-        self.assertTrue(base2.tzname() is None)
-
-        # Ensure we can add one.
-        base3 = base2.replace(tzinfo=z100)
-        self.assertEqual(base, base3)
-        self.assertTrue(base.tzinfo is base3.tzinfo)
-
-        # Out of bounds.
-        base = cls(2000, 2, 29)
-        self.assertRaises(ValueError, base.replace, year=2001)
-
-    def test_more_astimezone(self):
-        # The inherited test_astimezone covered some trivial and error cases.
-        fnone = FixedOffset(None, "None")
-        f44m = FixedOffset(44, "44")
-        fm5h = FixedOffset(-timedelta(hours=5), "m300")
-
-        dt = self.theclass.now(tz=f44m)
-        self.assertTrue(dt.tzinfo is f44m)
-        # Replacing with degenerate tzinfo raises an exception.
-        self.assertRaises(ValueError, dt.astimezone, fnone)
-        # Ditto with None tz.
-        self.assertRaises(TypeError, dt.astimezone, None)
-        # Replacing with same tzinfo makes no change.
-        x = dt.astimezone(dt.tzinfo)
-        self.assertTrue(x.tzinfo is f44m)
-        self.assertEqual(x.date(), dt.date())
-        self.assertEqual(x.time(), dt.time())
-
-        # Replacing with different tzinfo does adjust.
-        got = dt.astimezone(fm5h)
-        self.assertTrue(got.tzinfo is fm5h)
-        self.assertEqual(got.utcoffset(), timedelta(hours=-5))
-        expected = dt - dt.utcoffset()  # in effect, convert to UTC
-        expected += fm5h.utcoffset(dt)  # and from there to local time
-        expected = expected.replace(tzinfo=fm5h) # and attach new tzinfo
-        self.assertEqual(got.date(), expected.date())
-        self.assertEqual(got.time(), expected.time())
-        self.assertEqual(got.timetz(), expected.timetz())
-        self.assertTrue(got.tzinfo is expected.tzinfo)
-        self.assertEqual(got, expected)
-
-    def test_aware_subtract(self):
-        cls = self.theclass
-
-        # Ensure that utcoffset() is ignored when the operands have the
-        # same tzinfo member.
-        class OperandDependentOffset(tzinfo):
-            def utcoffset(self, t):
-                if t.minute < 10:
-                    # d0 and d1 equal after adjustment
-                    return timedelta(minutes=t.minute)
-                else:
-                    # d2 off in the weeds
-                    return timedelta(minutes=59)
-
-        base = cls(8, 9, 10, 11, 12, 13, 14, tzinfo=OperandDependentOffset())
-        d0 = base.replace(minute=3)
-        d1 = base.replace(minute=9)
-        d2 = base.replace(minute=11)
-        for x in d0, d1, d2:
-            for y in d0, d1, d2:
-                got = x - y
-                expected = timedelta(minutes=x.minute - y.minute)
-                self.assertEqual(got, expected)
-
-        # OTOH, if the tzinfo members are distinct, utcoffsets aren't
-        # ignored.
-        base = cls(8, 9, 10, 11, 12, 13, 14)
-        d0 = base.replace(minute=3, tzinfo=OperandDependentOffset())
-        d1 = base.replace(minute=9, tzinfo=OperandDependentOffset())
-        d2 = base.replace(minute=11, tzinfo=OperandDependentOffset())
-        for x in d0, d1, d2:
-            for y in d0, d1, d2:
-                got = x - y
-                if (x is d0 or x is d1) and (y is d0 or y is d1):
-                    expected = timedelta(0)
-                elif x is y is d2:
-                    expected = timedelta(0)
-                elif x is d2:
-                    expected = timedelta(minutes=(11-59)-0)
-                else:
-                    assert y is d2
-                    expected = timedelta(minutes=0-(11-59))
-                self.assertEqual(got, expected)
-
-    def test_mixed_compare(self):
-        t1 = datetime(1, 2, 3, 4, 5, 6, 7)
-        t2 = datetime(1, 2, 3, 4, 5, 6, 7)
-        self.assertEqual(t1, t2)
-        t2 = t2.replace(tzinfo=None)
-        self.assertEqual(t1, t2)
-        t2 = t2.replace(tzinfo=FixedOffset(None, ""))
-        self.assertEqual(t1, t2)
-        t2 = t2.replace(tzinfo=FixedOffset(0, ""))
-        self.assertRaises(TypeError, lambda: t1 == t2)
-
-        # In datetime w/ identical tzinfo objects, utcoffset is ignored.
-        class Varies(tzinfo):
-            def __init__(self):
-                self.offset = timedelta(minutes=22)
-            def utcoffset(self, t):
-                self.offset += timedelta(minutes=1)
-                return self.offset
-
-        v = Varies()
-        t1 = t2.replace(tzinfo=v)
-        t2 = t2.replace(tzinfo=v)
-        self.assertEqual(t1.utcoffset(), timedelta(minutes=23))
-        self.assertEqual(t2.utcoffset(), timedelta(minutes=24))
-        self.assertEqual(t1, t2)
-
-        # But if they're not identical, it isn't ignored.
-        t2 = t2.replace(tzinfo=Varies())
-        self.assertTrue(t1 < t2)  # t1's offset counter still going up
-
-    def test_subclass_datetimetz(self):
-
-        class C(self.theclass):
-            theAnswer = 42
-
-            def __new__(cls, *args, **kws):
-                temp = kws.copy()
-                extra = temp.pop('extra')
-                result = self.theclass.__new__(cls, *args, **temp)
-                result.extra = extra
-                return result
-
-            def newmeth(self, start):
-                return start + self.hour + self.year
-
-        args = 2002, 12, 31, 4, 5, 6, 500, FixedOffset(-300, "EST", 1)
-
-        dt1 = self.theclass(*args)
-        dt2 = C(*args, **{'extra': 7})
-
-        self.assertEqual(dt2.__class__, C)
-        self.assertEqual(dt2.theAnswer, 42)
-        self.assertEqual(dt2.extra, 7)
-        self.assertEqual(dt1.utcoffset(), dt2.utcoffset())
-        self.assertEqual(dt2.newmeth(-7), dt1.hour + dt1.year - 7)
-
-# Pain to set up DST-aware tzinfo classes.
-
-def first_sunday_on_or_after(dt):
-    days_to_go = 6 - dt.weekday()
-    if days_to_go:
-        dt += timedelta(days_to_go)
-    return dt
-
-ZERO = timedelta(0)
-MINUTE = timedelta(minutes=1)
-HOUR = timedelta(hours=1)
-DAY = timedelta(days=1)
-# In the US, DST starts at 2am (standard time) on the first Sunday in April.
-DSTSTART = datetime(1, 4, 1, 2)
-# and ends at 2am (DST time; 1am standard time) on the last Sunday of Oct,
-# which is the first Sunday on or after Oct 25.  Because we view 1:MM as
-# being standard time on that day, there is no spelling in local time of
-# the last hour of DST (that's 1:MM DST, but 1:MM is taken as standard time).
-DSTEND = datetime(1, 10, 25, 1)
-
-class USTimeZone(tzinfo):
-
-    def __init__(self, hours, reprname, stdname, dstname):
-        self.stdoffset = timedelta(hours=hours)
-        self.reprname = reprname
-        self.stdname = stdname
-        self.dstname = dstname
-
-    def __repr__(self):
-        return self.reprname
-
-    def tzname(self, dt):
-        if self.dst(dt):
-            return self.dstname
-        else:
-            return self.stdname
-
-    def utcoffset(self, dt):
-        return self.stdoffset + self.dst(dt)
-
-    def dst(self, dt):
-        if dt is None or dt.tzinfo is None:
-            # An exception instead may be sensible here, in one or more of
-            # the cases.
-            return ZERO
-        assert dt.tzinfo is self
-
-        # Find first Sunday in April.
-        start = first_sunday_on_or_after(DSTSTART.replace(year=dt.year))
-        assert start.weekday() == 6 and start.month == 4 and start.day <= 7
-
-        # Find last Sunday in October.
-        end = first_sunday_on_or_after(DSTEND.replace(year=dt.year))
-        assert end.weekday() == 6 and end.month == 10 and end.day >= 25
-
-        # Can't compare naive to aware objects, so strip the timezone from
-        # dt first.
-        if start <= dt.replace(tzinfo=None) < end:
-            return HOUR
-        else:
-            return ZERO
-
-Eastern  = USTimeZone(-5, "Eastern",  "EST", "EDT")
-Central  = USTimeZone(-6, "Central",  "CST", "CDT")
-Mountain = USTimeZone(-7, "Mountain", "MST", "MDT")
-Pacific  = USTimeZone(-8, "Pacific",  "PST", "PDT")
-utc_real = FixedOffset(0, "UTC", 0)
-# For better test coverage, we want another flavor of UTC that's west of
-# the Eastern and Pacific timezones.
-utc_fake = FixedOffset(-12*60, "UTCfake", 0)
-
-class TestTimezoneConversions(unittest.TestCase):
-    # The DST switch times for 2002, in std time.
-    dston = datetime(2002, 4, 7, 2)
-    dstoff = datetime(2002, 10, 27, 1)
-
-    theclass = datetime
-
-    # Check a time that's inside DST.
-    def checkinside(self, dt, tz, utc, dston, dstoff):
-        self.assertEqual(dt.dst(), HOUR)
-
-        # Conversion to our own timezone is always an identity.
-        self.assertEqual(dt.astimezone(tz), dt)
-
-        asutc = dt.astimezone(utc)
-        there_and_back = asutc.astimezone(tz)
-
-        # Conversion to UTC and back isn't always an identity here,
-        # because there are redundant spellings (in local time) of
-        # UTC time when DST begins:  the clock jumps from 1:59:59
-        # to 3:00:00, and a local time of 2:MM:SS doesn't really
-        # make sense then.  The classes above treat 2:MM:SS as
-        # daylight time then (it's "after 2am"), really an alias
-        # for 1:MM:SS standard time.  The latter form is what
-        # conversion back from UTC produces.
-        if dt.date() == dston.date() and dt.hour == 2:
-            # We're in the redundant hour, and coming back from
-            # UTC gives the 1:MM:SS standard-time spelling.
-            self.assertEqual(there_and_back + HOUR, dt)
-            # Although during was considered to be in daylight
-            # time, there_and_back is not.
-            self.assertEqual(there_and_back.dst(), ZERO)
-            # They're the same times in UTC.
-            self.assertEqual(there_and_back.astimezone(utc),
-                             dt.astimezone(utc))
-        else:
-            # We're not in the redundant hour.
-            self.assertEqual(dt, there_and_back)
-
-        # Because we have a redundant spelling when DST begins, there is
-        # (unforunately) an hour when DST ends that can't be spelled at all in
-        # local time.  When DST ends, the clock jumps from 1:59 back to 1:00
-        # again.  The hour 1:MM DST has no spelling then:  1:MM is taken to be
-        # standard time.  1:MM DST == 0:MM EST, but 0:MM is taken to be
-        # daylight time.  The hour 1:MM daylight == 0:MM standard can't be
-        # expressed in local time.  Nevertheless, we want conversion back
-        # from UTC to mimic the local clock's "repeat an hour" behavior.
-        nexthour_utc = asutc + HOUR
-        nexthour_tz = nexthour_utc.astimezone(tz)
-        if dt.date() == dstoff.date() and dt.hour == 0:
-            # We're in the hour before the last DST hour.  The last DST hour
-            # is ineffable.  We want the conversion back to repeat 1:MM.
-            self.assertEqual(nexthour_tz, dt.replace(hour=1))
-            nexthour_utc += HOUR
-            nexthour_tz = nexthour_utc.astimezone(tz)
-            self.assertEqual(nexthour_tz, dt.replace(hour=1))
-        else:
-            self.assertEqual(nexthour_tz - dt, HOUR)
-
-    # Check a time that's outside DST.
-    def checkoutside(self, dt, tz, utc):
-        self.assertEqual(dt.dst(), ZERO)
-
-        # Conversion to our own timezone is always an identity.
-        self.assertEqual(dt.astimezone(tz), dt)
-
-        # Converting to UTC and back is an identity too.
-        asutc = dt.astimezone(utc)
-        there_and_back = asutc.astimezone(tz)
-        self.assertEqual(dt, there_and_back)
-
-    def convert_between_tz_and_utc(self, tz, utc):
-        dston = self.dston.replace(tzinfo=tz)
-        # Because 1:MM on the day DST ends is taken as being standard time,
-        # there is no spelling in tz for the last hour of daylight time.
-        # For purposes of the test, the last hour of DST is 0:MM, which is
-        # taken as being daylight time (and 1:MM is taken as being standard
-        # time).
-        dstoff = self.dstoff.replace(tzinfo=tz)
-        for delta in (timedelta(weeks=13),
-                      DAY,
-                      HOUR,
-                      timedelta(minutes=1),
-                      timedelta(microseconds=1)):
-
-            self.checkinside(dston, tz, utc, dston, dstoff)
-            for during in dston + delta, dstoff - delta:
-                self.checkinside(during, tz, utc, dston, dstoff)
-
-            self.checkoutside(dstoff, tz, utc)
-            for outside in dston - delta, dstoff + delta:
-                self.checkoutside(outside, tz, utc)
-
-    def test_easy(self):
-        # Despite the name of this test, the endcases are excruciating.
-        self.convert_between_tz_and_utc(Eastern, utc_real)
-        self.convert_between_tz_and_utc(Pacific, utc_real)
-        self.convert_between_tz_and_utc(Eastern, utc_fake)
-        self.convert_between_tz_and_utc(Pacific, utc_fake)
-        # The next is really dancing near the edge.  It works because
-        # Pacific and Eastern are far enough apart that their "problem
-        # hours" don't overlap.
-        self.convert_between_tz_and_utc(Eastern, Pacific)
-        self.convert_between_tz_and_utc(Pacific, Eastern)
-        # OTOH, these fail!  Don't enable them.  The difficulty is that
-        # the edge case tests assume that every hour is representable in
-        # the "utc" class.  This is always true for a fixed-offset tzinfo
-        # class (lke utc_real and utc_fake), but not for Eastern or Central.
-        # For these adjacent DST-aware time zones, the range of time offsets
-        # tested ends up creating hours in the one that aren't representable
-        # in the other.  For the same reason, we would see failures in the
-        # Eastern vs Pacific tests too if we added 3*HOUR to the list of
-        # offset deltas in convert_between_tz_and_utc().
-        #
-        # self.convert_between_tz_and_utc(Eastern, Central)  # can't work
-        # self.convert_between_tz_and_utc(Central, Eastern)  # can't work
-
-    def test_tricky(self):
-        # 22:00 on day before daylight starts.
-        fourback = self.dston - timedelta(hours=4)
-        ninewest = FixedOffset(-9*60, "-0900", 0)
-        fourback = fourback.replace(tzinfo=ninewest)
-        # 22:00-0900 is 7:00 UTC == 2:00 EST == 3:00 DST.  Since it's "after
-        # 2", we should get the 3 spelling.
-        # If we plug 22:00 the day before into Eastern, it "looks like std
-        # time", so its offset is returned as -5, and -5 - -9 = 4.  Adding 4
-        # to 22:00 lands on 2:00, which makes no sense in local time (the
-        # local clock jumps from 1 to 3).  The point here is to make sure we
-        # get the 3 spelling.
-        expected = self.dston.replace(hour=3)
-        got = fourback.astimezone(Eastern).replace(tzinfo=None)
-        self.assertEqual(expected, got)
-
-        # Similar, but map to 6:00 UTC == 1:00 EST == 2:00 DST.  In that
-        # case we want the 1:00 spelling.
-        sixutc = self.dston.replace(hour=6, tzinfo=utc_real)
-        # Now 6:00 "looks like daylight", so the offset wrt Eastern is -4,
-        # and adding -4-0 == -4 gives the 2:00 spelling.  We want the 1:00 EST
-        # spelling.
-        expected = self.dston.replace(hour=1)
-        got = sixutc.astimezone(Eastern).replace(tzinfo=None)
-        self.assertEqual(expected, got)
-
-        # Now on the day DST ends, we want "repeat an hour" behavior.
-        #  UTC  4:MM  5:MM  6:MM  7:MM  checking these
-        #  EST 23:MM  0:MM  1:MM  2:MM
-        #  EDT  0:MM  1:MM  2:MM  3:MM
-        # wall  0:MM  1:MM  1:MM  2:MM  against these
-        for utc in utc_real, utc_fake:
-            for tz in Eastern, Pacific:
-                first_std_hour = self.dstoff - timedelta(hours=2) # 23:MM
-                # Convert that to UTC.
-                first_std_hour -= tz.utcoffset(None)
-                # Adjust for possibly fake UTC.
-                asutc = first_std_hour + utc.utcoffset(None)
-                # First UTC hour to convert; this is 4:00 when utc=utc_real &
-                # tz=Eastern.
-                asutcbase = asutc.replace(tzinfo=utc)
-                for tzhour in (0, 1, 1, 2):
-                    expectedbase = self.dstoff.replace(hour=tzhour)
-                    for minute in 0, 30, 59:
-                        expected = expectedbase.replace(minute=minute)
-                        asutc = asutcbase.replace(minute=minute)
-                        astz = asutc.astimezone(tz)
-                        self.assertEqual(astz.replace(tzinfo=None), expected)
-                    asutcbase += HOUR
-
-
-    def test_bogus_dst(self):
-        class ok(tzinfo):
-            def utcoffset(self, dt): return HOUR
-            def dst(self, dt): return HOUR
-
-        now = self.theclass.now().replace(tzinfo=utc_real)
-        # Doesn't blow up.
-        now.astimezone(ok())
-
-        # Does blow up.
-        class notok(ok):
-            def dst(self, dt): return None
-        self.assertRaises(ValueError, now.astimezone, notok())
-
-        # Sometimes blow up. In the following, tzinfo.dst()
-        # implementation may return None or not None depending on
-        # whether DST is assumed to be in effect.  In this situation,
-        # a ValueError should be raised by astimezone().
-        class tricky_notok(ok):
-            def dst(self, dt):
-                if dt.year == 2000:
-                    return None
-                else:
-                    return 10*HOUR
-        dt = self.theclass(2001, 1, 1).replace(tzinfo=utc_real)
-        self.assertRaises(ValueError, dt.astimezone, tricky_notok())
-
-    def test_fromutc(self):
-        self.assertRaises(TypeError, Eastern.fromutc)   # not enough args
-        now = datetime.utcnow().replace(tzinfo=utc_real)
-        self.assertRaises(ValueError, Eastern.fromutc, now) # wrong tzinfo
-        now = now.replace(tzinfo=Eastern)   # insert correct tzinfo
-        enow = Eastern.fromutc(now)         # doesn't blow up
-        self.assertEqual(enow.tzinfo, Eastern) # has right tzinfo member
-        self.assertRaises(TypeError, Eastern.fromutc, now, now) # too many args
-        self.assertRaises(TypeError, Eastern.fromutc, date.today()) # wrong type
-
-        # Always converts UTC to standard time.
-        class FauxUSTimeZone(USTimeZone):
-            def fromutc(self, dt):
-                return dt + self.stdoffset
-        FEastern  = FauxUSTimeZone(-5, "FEastern",  "FEST", "FEDT")
-
-        #  UTC  4:MM  5:MM  6:MM  7:MM  8:MM  9:MM
-        #  EST 23:MM  0:MM  1:MM  2:MM  3:MM  4:MM
-        #  EDT  0:MM  1:MM  2:MM  3:MM  4:MM  5:MM
-
-        # Check around DST start.
-        start = self.dston.replace(hour=4, tzinfo=Eastern)
-        fstart = start.replace(tzinfo=FEastern)
-        for wall in 23, 0, 1, 3, 4, 5:
-            expected = start.replace(hour=wall)
-            if wall == 23:
-                expected -= timedelta(days=1)
-            got = Eastern.fromutc(start)
-            self.assertEqual(expected, got)
-
-            expected = fstart + FEastern.stdoffset
-            got = FEastern.fromutc(fstart)
-            self.assertEqual(expected, got)
-
-            # Ensure astimezone() calls fromutc() too.
-            got = fstart.replace(tzinfo=utc_real).astimezone(FEastern)
-            self.assertEqual(expected, got)
-
-            start += HOUR
-            fstart += HOUR
-
-        # Check around DST end.
-        start = self.dstoff.replace(hour=4, tzinfo=Eastern)
-        fstart = start.replace(tzinfo=FEastern)
-        for wall in 0, 1, 1, 2, 3, 4:
-            expected = start.replace(hour=wall)
-            got = Eastern.fromutc(start)
-            self.assertEqual(expected, got)
-
-            expected = fstart + FEastern.stdoffset
-            got = FEastern.fromutc(fstart)
-            self.assertEqual(expected, got)
-
-            # Ensure astimezone() calls fromutc() too.
-            got = fstart.replace(tzinfo=utc_real).astimezone(FEastern)
-            self.assertEqual(expected, got)
-
-            start += HOUR
-            fstart += HOUR
-
-
-#############################################################################
-# oddballs
-
-class Oddballs(unittest.TestCase):
-
-    def test_bug_1028306(self):
-        # Trying to compare a date to a datetime should act like a mixed-
-        # type comparison, despite that datetime is a subclass of date.
-        as_date = date.today()
-        as_datetime = datetime.combine(as_date, time())
-        self.assertTrue(as_date != as_datetime)
-        self.assertTrue(as_datetime != as_date)
-        self.assertTrue(not as_date == as_datetime)
-        self.assertTrue(not as_datetime == as_date)
-        self.assertRaises(TypeError, lambda: as_date < as_datetime)
-        self.assertRaises(TypeError, lambda: as_datetime < as_date)
-        self.assertRaises(TypeError, lambda: as_date <= as_datetime)
-        self.assertRaises(TypeError, lambda: as_datetime <= as_date)
-        self.assertRaises(TypeError, lambda: as_date > as_datetime)
-        self.assertRaises(TypeError, lambda: as_datetime > as_date)
-        self.assertRaises(TypeError, lambda: as_date >= as_datetime)
-        self.assertRaises(TypeError, lambda: as_datetime >= as_date)
-
-        # Neverthelss, comparison should work with the base-class (date)
-        # projection if use of a date method is forced.
-        self.assertEqual(as_date.__eq__(as_datetime), True)
-        different_day = (as_date.day + 1) % 20 + 1
-        as_different = as_datetime.replace(day= different_day)
-        self.assertEqual(as_date.__eq__(as_different), False)
-
-        # And date should compare with other subclasses of date.  If a
-        # subclass wants to stop this, it's up to the subclass to do so.
-        date_sc = SubclassDate(as_date.year, as_date.month, as_date.day)
-        self.assertEqual(as_date, date_sc)
-        self.assertEqual(date_sc, as_date)
-
-        # Ditto for datetimes.
-        datetime_sc = SubclassDatetime(as_datetime.year, as_datetime.month,
-                                       as_date.day, 0, 0, 0)
-        self.assertEqual(as_datetime, datetime_sc)
-        self.assertEqual(datetime_sc, as_datetime)
+import sys
+from test.support import import_fresh_module, run_unittest
+TESTS = 'test.datetimetester'
+# XXX: import_fresh_module() is supposed to leave sys.module cache untouched,
+# XXX: but it does not, so we have to save and restore it ourselves.
+save_sys_modules = sys.modules.copy()
+try:
+    pure_tests = import_fresh_module(TESTS, fresh=['datetime', '_strptime'],
+                                     blocked=['_datetime'])
+    fast_tests = import_fresh_module(TESTS, fresh=['datetime',
+                                                   '_datetime', '_strptime'])
+finally:
+    sys.modules.clear()
+    sys.modules.update(save_sys_modules)
+test_modules = [pure_tests, fast_tests]
+test_suffixes = ["_Pure", "_Fast"]
+
+for module, suffix in zip(test_modules, test_suffixes):
+    for name, cls in module.__dict__.items():
+        if isinstance(cls, type) and issubclass(cls, unittest.TestCase):
+            name += suffix
+            cls.__name__ = name
+            globals()[name] = cls
+            def setUp(self, module=module, setup=cls.setUp):
+                self._save_sys_modules = sys.modules.copy()
+                sys.modules[TESTS] = module
+                sys.modules['datetime'] = module.datetime_module
+                sys.modules['_strptime'] = module._strptime
+                setup(self)
+            def tearDown(self, teardown=cls.tearDown):
+                teardown(self)
+                sys.modules.clear()
+                sys.modules.update(self._save_sys_modules)
+            cls.setUp = setUp
+            cls.tearDown = tearDown
 
 def test_main():
-    support.run_unittest(__name__)
+    run_unittest(__name__)
 
 if __name__ == "__main__":
     test_main()

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Fri Jul 23 21:25:47 2010
@@ -473,6 +473,14 @@
 Library
 -------
 
+- Issue #7989: Added pure python implementation of the `datetime`
+  module.  The C module is renamed to `_datetime` and if available,
+  overrides all classes defined in datetime with fast C impementation.
+  Python implementation is based on the original python prototype for
+  the datetime module by Tim Peters with minor modifications by the
+  PyPy project.  The test suite now tests `datetime` module with and
+  without `_datetime` acceleration using the same test cases.
+
 - Issue #7895: platform.mac_ver() no longer crashes after calling os.fork()
 
 - Issue #9323: Fixed a bug in trace.py that resulted in loosing the

Modified: python/branches/py3k/Modules/Setup.dist
==============================================================================
--- python/branches/py3k/Modules/Setup.dist	(original)
+++ python/branches/py3k/Modules/Setup.dist	Fri Jul 23 21:25:47 2010
@@ -170,7 +170,7 @@
 #atexit atexitmodule.c      # Register functions to be run at interpreter-shutdown
 #_elementtree -I$(srcdir)/Modules/expat -DHAVE_EXPAT_CONFIG_H -DUSE_PYEXPAT_CAPI _elementtree.c	# elementtree accelerator
 #_pickle _pickle.c	# pickle accelerator
-#datetime datetimemodule.c	# date/time type
+#_datetime _datetimemodule.c	# datetime accelerator
 #_bisect _bisectmodule.c	# Bisection algorithms
 #_heapq _heapqmodule.c	# Heap queue algorithm
 

Copied: python/branches/py3k/Modules/_datetimemodule.c (from r83064, /python/branches/py3k/Modules/datetimemodule.c)
==============================================================================
--- /python/branches/py3k/Modules/datetimemodule.c	(original)
+++ python/branches/py3k/Modules/_datetimemodule.c	Fri Jul 23 21:25:47 2010
@@ -25,7 +25,7 @@
  * final result fits in a C int (this can be an issue on 64-bit boxes).
  */
 #if SIZEOF_INT < 4
-#       error "datetime.c requires that C int have at least 32 bits"
+#       error "_datetime.c requires that C int have at least 32 bits"
 #endif
 
 #define MINYEAR 1
@@ -5086,7 +5086,7 @@
 
 static struct PyModuleDef datetimemodule = {
     PyModuleDef_HEAD_INIT,
-    "datetime",
+    "_datetime",
     "Fast implementation of the datetime type.",
     -1,
     module_methods,
@@ -5097,7 +5097,7 @@
 };
 
 PyMODINIT_FUNC
-PyInit_datetime(void)
+PyInit__datetime(void)
 {
     PyObject *m;        /* a module object */
     PyObject *d;        /* its dict */

Deleted: python/branches/py3k/Modules/datetimemodule.c
==============================================================================
--- python/branches/py3k/Modules/datetimemodule.c	Fri Jul 23 21:25:47 2010
+++ (empty file)
@@ -1,5488 +0,0 @@
-/*  C implementation for the date/time type documented at
- *  http://www.zope.org/Members/fdrake/DateTimeWiki/FrontPage
- */
-
-#include "Python.h"
-#include "modsupport.h"
-#include "structmember.h"
-
-#include <time.h>
-
-#include "timefuncs.h"
-
-/* Differentiate between building the core module and building extension
- * modules.
- */
-#ifndef Py_BUILD_CORE
-#define Py_BUILD_CORE
-#endif
-#include "datetime.h"
-#undef Py_BUILD_CORE
-
-/* We require that C int be at least 32 bits, and use int virtually
- * everywhere.  In just a few cases we use a temp long, where a Python
- * API returns a C long.  In such cases, we have to ensure that the
- * final result fits in a C int (this can be an issue on 64-bit boxes).
- */
-#if SIZEOF_INT < 4
-#       error "datetime.c requires that C int have at least 32 bits"
-#endif
-
-#define MINYEAR 1
-#define MAXYEAR 9999
-#define MAXORDINAL 3652059 /* date(9999,12,31).toordinal() */
-
-/* Nine decimal digits is easy to communicate, and leaves enough room
- * so that two delta days can be added w/o fear of overflowing a signed
- * 32-bit int, and with plenty of room left over to absorb any possible
- * carries from adding seconds.
- */
-#define MAX_DELTA_DAYS 999999999
-
-/* Rename the long macros in datetime.h to more reasonable short names. */
-#define GET_YEAR                PyDateTime_GET_YEAR
-#define GET_MONTH               PyDateTime_GET_MONTH
-#define GET_DAY                 PyDateTime_GET_DAY
-#define DATE_GET_HOUR           PyDateTime_DATE_GET_HOUR
-#define DATE_GET_MINUTE         PyDateTime_DATE_GET_MINUTE
-#define DATE_GET_SECOND         PyDateTime_DATE_GET_SECOND
-#define DATE_GET_MICROSECOND    PyDateTime_DATE_GET_MICROSECOND
-
-/* Date accessors for date and datetime. */
-#define SET_YEAR(o, v)          (((o)->data[0] = ((v) & 0xff00) >> 8), \
-                 ((o)->data[1] = ((v) & 0x00ff)))
-#define SET_MONTH(o, v)         (PyDateTime_GET_MONTH(o) = (v))
-#define SET_DAY(o, v)           (PyDateTime_GET_DAY(o) = (v))
-
-/* Date/Time accessors for datetime. */
-#define DATE_SET_HOUR(o, v)     (PyDateTime_DATE_GET_HOUR(o) = (v))
-#define DATE_SET_MINUTE(o, v)   (PyDateTime_DATE_GET_MINUTE(o) = (v))
-#define DATE_SET_SECOND(o, v)   (PyDateTime_DATE_GET_SECOND(o) = (v))
-#define DATE_SET_MICROSECOND(o, v)      \
-    (((o)->data[7] = ((v) & 0xff0000) >> 16), \
-     ((o)->data[8] = ((v) & 0x00ff00) >> 8), \
-     ((o)->data[9] = ((v) & 0x0000ff)))
-
-/* Time accessors for time. */
-#define TIME_GET_HOUR           PyDateTime_TIME_GET_HOUR
-#define TIME_GET_MINUTE         PyDateTime_TIME_GET_MINUTE
-#define TIME_GET_SECOND         PyDateTime_TIME_GET_SECOND
-#define TIME_GET_MICROSECOND    PyDateTime_TIME_GET_MICROSECOND
-#define TIME_SET_HOUR(o, v)     (PyDateTime_TIME_GET_HOUR(o) = (v))
-#define TIME_SET_MINUTE(o, v)   (PyDateTime_TIME_GET_MINUTE(o) = (v))
-#define TIME_SET_SECOND(o, v)   (PyDateTime_TIME_GET_SECOND(o) = (v))
-#define TIME_SET_MICROSECOND(o, v)      \
-    (((o)->data[3] = ((v) & 0xff0000) >> 16), \
-     ((o)->data[4] = ((v) & 0x00ff00) >> 8), \
-     ((o)->data[5] = ((v) & 0x0000ff)))
-
-/* Delta accessors for timedelta. */
-#define GET_TD_DAYS(o)          (((PyDateTime_Delta *)(o))->days)
-#define GET_TD_SECONDS(o)       (((PyDateTime_Delta *)(o))->seconds)
-#define GET_TD_MICROSECONDS(o)  (((PyDateTime_Delta *)(o))->microseconds)
-
-#define SET_TD_DAYS(o, v)       ((o)->days = (v))
-#define SET_TD_SECONDS(o, v)    ((o)->seconds = (v))
-#define SET_TD_MICROSECONDS(o, v) ((o)->microseconds = (v))
-
-/* p is a pointer to a time or a datetime object; HASTZINFO(p) returns
- * p->hastzinfo.
- */
-#define HASTZINFO(p) (((_PyDateTime_BaseTZInfo *)(p))->hastzinfo)
-#define GET_TIME_TZINFO(p) (HASTZINFO(p) ? \
-                            ((PyDateTime_Time *)(p))->tzinfo : Py_None)
-#define GET_DT_TZINFO(p) (HASTZINFO(p) ? \
-                          ((PyDateTime_DateTime *)(p))->tzinfo : Py_None)
-/* M is a char or int claiming to be a valid month.  The macro is equivalent
- * to the two-sided Python test
- *      1 <= M <= 12
- */
-#define MONTH_IS_SANE(M) ((unsigned int)(M) - 1 < 12)
-
-/* Forward declarations. */
-static PyTypeObject PyDateTime_DateType;
-static PyTypeObject PyDateTime_DateTimeType;
-static PyTypeObject PyDateTime_DeltaType;
-static PyTypeObject PyDateTime_TimeType;
-static PyTypeObject PyDateTime_TZInfoType;
-static PyTypeObject PyDateTime_TimeZoneType;
-
-/* ---------------------------------------------------------------------------
- * Math utilities.
- */
-
-/* k = i+j overflows iff k differs in sign from both inputs,
- * iff k^i has sign bit set and k^j has sign bit set,
- * iff (k^i)&(k^j) has sign bit set.
- */
-#define SIGNED_ADD_OVERFLOWED(RESULT, I, J) \
-    ((((RESULT) ^ (I)) & ((RESULT) ^ (J))) < 0)
-
-/* Compute Python divmod(x, y), returning the quotient and storing the
- * remainder into *r.  The quotient is the floor of x/y, and that's
- * the real point of this.  C will probably truncate instead (C99
- * requires truncation; C89 left it implementation-defined).
- * Simplification:  we *require* that y > 0 here.  That's appropriate
- * for all the uses made of it.  This simplifies the code and makes
- * the overflow case impossible (divmod(LONG_MIN, -1) is the only
- * overflow case).
- */
-static int
-divmod(int x, int y, int *r)
-{
-    int quo;
-
-    assert(y > 0);
-    quo = x / y;
-    *r = x - quo * y;
-    if (*r < 0) {
-        --quo;
-        *r += y;
-    }
-    assert(0 <= *r && *r < y);
-    return quo;
-}
-
-/* Round a double to the nearest long.  |x| must be small enough to fit
- * in a C long; this is not checked.
- */
-static long
-round_to_long(double x)
-{
-    if (x >= 0.0)
-        x = floor(x + 0.5);
-    else
-        x = ceil(x - 0.5);
-    return (long)x;
-}
-
-/* Nearest integer to m / n for integers m and n. Half-integer results
- * are rounded to even.
- */
-static PyObject *
-divide_nearest(PyObject *m, PyObject *n)
-{
-    PyObject *result;
-    PyObject *temp;
-
-    temp = _PyLong_DivmodNear(m, n);
-    if (temp == NULL)
-        return NULL;
-    result = PyTuple_GET_ITEM(temp, 0);
-    Py_INCREF(result);
-    Py_DECREF(temp);
-
-    return result;
-}
-
-/* ---------------------------------------------------------------------------
- * General calendrical helper functions
- */
-
-/* For each month ordinal in 1..12, the number of days in that month,
- * and the number of days before that month in the same year.  These
- * are correct for non-leap years only.
- */
-static int _days_in_month[] = {
-    0, /* unused; this vector uses 1-based indexing */
-    31, 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31
-};
-
-static int _days_before_month[] = {
-    0, /* unused; this vector uses 1-based indexing */
-    0, 31, 59, 90, 120, 151, 181, 212, 243, 273, 304, 334
-};
-
-/* year -> 1 if leap year, else 0. */
-static int
-is_leap(int year)
-{
-    /* Cast year to unsigned.  The result is the same either way, but
-     * C can generate faster code for unsigned mod than for signed
-     * mod (especially for % 4 -- a good compiler should just grab
-     * the last 2 bits when the LHS is unsigned).
-     */
-    const unsigned int ayear = (unsigned int)year;
-    return ayear % 4 == 0 && (ayear % 100 != 0 || ayear % 400 == 0);
-}
-
-/* year, month -> number of days in that month in that year */
-static int
-days_in_month(int year, int month)
-{
-    assert(month >= 1);
-    assert(month <= 12);
-    if (month == 2 && is_leap(year))
-        return 29;
-    else
-        return _days_in_month[month];
-}
-
-/* year, month -> number of days in year preceeding first day of month */
-static int
-days_before_month(int year, int month)
-{
-    int days;
-
-    assert(month >= 1);
-    assert(month <= 12);
-    days = _days_before_month[month];
-    if (month > 2 && is_leap(year))
-        ++days;
-    return days;
-}
-
-/* year -> number of days before January 1st of year.  Remember that we
- * start with year 1, so days_before_year(1) == 0.
- */
-static int
-days_before_year(int year)
-{
-    int y = year - 1;
-    /* This is incorrect if year <= 0; we really want the floor
-     * here.  But so long as MINYEAR is 1, the smallest year this
-     * can see is 0 (this can happen in some normalization endcases),
-     * so we'll just special-case that.
-     */
-    assert (year >= 0);
-    if (y >= 0)
-        return y*365 + y/4 - y/100 + y/400;
-    else {
-        assert(y == -1);
-        return -366;
-    }
-}
-
-/* Number of days in 4, 100, and 400 year cycles.  That these have
- * the correct values is asserted in the module init function.
- */
-#define DI4Y    1461    /* days_before_year(5); days in 4 years */
-#define DI100Y  36524   /* days_before_year(101); days in 100 years */
-#define DI400Y  146097  /* days_before_year(401); days in 400 years  */
-
-/* ordinal -> year, month, day, considering 01-Jan-0001 as day 1. */
-static void
-ord_to_ymd(int ordinal, int *year, int *month, int *day)
-{
-    int n, n1, n4, n100, n400, leapyear, preceding;
-
-    /* ordinal is a 1-based index, starting at 1-Jan-1.  The pattern of
-     * leap years repeats exactly every 400 years.  The basic strategy is
-     * to find the closest 400-year boundary at or before ordinal, then
-     * work with the offset from that boundary to ordinal.  Life is much
-     * clearer if we subtract 1 from ordinal first -- then the values
-     * of ordinal at 400-year boundaries are exactly those divisible
-     * by DI400Y:
-     *
-     *    D  M   Y            n              n-1
-     *    -- --- ----        ----------     ----------------
-     *    31 Dec -400        -DI400Y       -DI400Y -1
-     *     1 Jan -399         -DI400Y +1   -DI400Y      400-year boundary
-     *    ...
-     *    30 Dec  000        -1             -2
-     *    31 Dec  000         0             -1
-     *     1 Jan  001         1              0          400-year boundary
-     *     2 Jan  001         2              1
-     *     3 Jan  001         3              2
-     *    ...
-     *    31 Dec  400         DI400Y        DI400Y -1
-     *     1 Jan  401         DI400Y +1     DI400Y      400-year boundary
-     */
-    assert(ordinal >= 1);
-    --ordinal;
-    n400 = ordinal / DI400Y;
-    n = ordinal % DI400Y;
-    *year = n400 * 400 + 1;
-
-    /* Now n is the (non-negative) offset, in days, from January 1 of
-     * year, to the desired date.  Now compute how many 100-year cycles
-     * precede n.
-     * Note that it's possible for n100 to equal 4!  In that case 4 full
-     * 100-year cycles precede the desired day, which implies the
-     * desired day is December 31 at the end of a 400-year cycle.
-     */
-    n100 = n / DI100Y;
-    n = n % DI100Y;
-
-    /* Now compute how many 4-year cycles precede it. */
-    n4 = n / DI4Y;
-    n = n % DI4Y;
-
-    /* And now how many single years.  Again n1 can be 4, and again
-     * meaning that the desired day is December 31 at the end of the
-     * 4-year cycle.
-     */
-    n1 = n / 365;
-    n = n % 365;
-
-    *year += n100 * 100 + n4 * 4 + n1;
-    if (n1 == 4 || n100 == 4) {
-        assert(n == 0);
-        *year -= 1;
-        *month = 12;
-        *day = 31;
-        return;
-    }
-
-    /* Now the year is correct, and n is the offset from January 1.  We
-     * find the month via an estimate that's either exact or one too
-     * large.
-     */
-    leapyear = n1 == 3 && (n4 != 24 || n100 == 3);
-    assert(leapyear == is_leap(*year));
-    *month = (n + 50) >> 5;
-    preceding = (_days_before_month[*month] + (*month > 2 && leapyear));
-    if (preceding > n) {
-        /* estimate is too large */
-        *month -= 1;
-        preceding -= days_in_month(*year, *month);
-    }
-    n -= preceding;
-    assert(0 <= n);
-    assert(n < days_in_month(*year, *month));
-
-    *day = n + 1;
-}
-
-/* year, month, day -> ordinal, considering 01-Jan-0001 as day 1. */
-static int
-ymd_to_ord(int year, int month, int day)
-{
-    return days_before_year(year) + days_before_month(year, month) + day;
-}
-
-/* Day of week, where Monday==0, ..., Sunday==6.  1/1/1 was a Monday. */
-static int
-weekday(int year, int month, int day)
-{
-    return (ymd_to_ord(year, month, day) + 6) % 7;
-}
-
-/* Ordinal of the Monday starting week 1 of the ISO year.  Week 1 is the
- * first calendar week containing a Thursday.
- */
-static int
-iso_week1_monday(int year)
-{
-    int first_day = ymd_to_ord(year, 1, 1);     /* ord of 1/1 */
-    /* 0 if 1/1 is a Monday, 1 if a Tue, etc. */
-    int first_weekday = (first_day + 6) % 7;
-    /* ordinal of closest Monday at or before 1/1 */
-    int week1_monday  = first_day - first_weekday;
-
-    if (first_weekday > 3)      /* if 1/1 was Fri, Sat, Sun */
-        week1_monday += 7;
-    return week1_monday;
-}
-
-/* ---------------------------------------------------------------------------
- * Range checkers.
- */
-
-/* Check that -MAX_DELTA_DAYS <= days <= MAX_DELTA_DAYS.  If so, return 0.
- * If not, raise OverflowError and return -1.
- */
-static int
-check_delta_day_range(int days)
-{
-    if (-MAX_DELTA_DAYS <= days && days <= MAX_DELTA_DAYS)
-        return 0;
-    PyErr_Format(PyExc_OverflowError,
-                 "days=%d; must have magnitude <= %d",
-                 days, MAX_DELTA_DAYS);
-    return -1;
-}
-
-/* Check that date arguments are in range.  Return 0 if they are.  If they
- * aren't, raise ValueError and return -1.
- */
-static int
-check_date_args(int year, int month, int day)
-{
-
-    if (year < MINYEAR || year > MAXYEAR) {
-        PyErr_SetString(PyExc_ValueError,
-                        "year is out of range");
-        return -1;
-    }
-    if (month < 1 || month > 12) {
-        PyErr_SetString(PyExc_ValueError,
-                        "month must be in 1..12");
-        return -1;
-    }
-    if (day < 1 || day > days_in_month(year, month)) {
-        PyErr_SetString(PyExc_ValueError,
-                        "day is out of range for month");
-        return -1;
-    }
-    return 0;
-}
-
-/* Check that time arguments are in range.  Return 0 if they are.  If they
- * aren't, raise ValueError and return -1.
- */
-static int
-check_time_args(int h, int m, int s, int us)
-{
-    if (h < 0 || h > 23) {
-        PyErr_SetString(PyExc_ValueError,
-                        "hour must be in 0..23");
-        return -1;
-    }
-    if (m < 0 || m > 59) {
-        PyErr_SetString(PyExc_ValueError,
-                        "minute must be in 0..59");
-        return -1;
-    }
-    if (s < 0 || s > 59) {
-        PyErr_SetString(PyExc_ValueError,
-                        "second must be in 0..59");
-        return -1;
-    }
-    if (us < 0 || us > 999999) {
-        PyErr_SetString(PyExc_ValueError,
-                        "microsecond must be in 0..999999");
-        return -1;
-    }
-    return 0;
-}
-
-/* ---------------------------------------------------------------------------
- * Normalization utilities.
- */
-
-/* One step of a mixed-radix conversion.  A "hi" unit is equivalent to
- * factor "lo" units.  factor must be > 0.  If *lo is less than 0, or
- * at least factor, enough of *lo is converted into "hi" units so that
- * 0 <= *lo < factor.  The input values must be such that int overflow
- * is impossible.
- */
-static void
-normalize_pair(int *hi, int *lo, int factor)
-{
-    assert(factor > 0);
-    assert(lo != hi);
-    if (*lo < 0 || *lo >= factor) {
-        const int num_hi = divmod(*lo, factor, lo);
-        const int new_hi = *hi + num_hi;
-        assert(! SIGNED_ADD_OVERFLOWED(new_hi, *hi, num_hi));
-        *hi = new_hi;
-    }
-    assert(0 <= *lo && *lo < factor);
-}
-
-/* Fiddle days (d), seconds (s), and microseconds (us) so that
- *      0 <= *s < 24*3600
- *      0 <= *us < 1000000
- * The input values must be such that the internals don't overflow.
- * The way this routine is used, we don't get close.
- */
-static void
-normalize_d_s_us(int *d, int *s, int *us)
-{
-    if (*us < 0 || *us >= 1000000) {
-        normalize_pair(s, us, 1000000);
-        /* |s| can't be bigger than about
-         * |original s| + |original us|/1000000 now.
-         */
-
-    }
-    if (*s < 0 || *s >= 24*3600) {
-        normalize_pair(d, s, 24*3600);
-        /* |d| can't be bigger than about
-         * |original d| +
-         * (|original s| + |original us|/1000000) / (24*3600) now.
-         */
-    }
-    assert(0 <= *s && *s < 24*3600);
-    assert(0 <= *us && *us < 1000000);
-}
-
-/* Fiddle years (y), months (m), and days (d) so that
- *      1 <= *m <= 12
- *      1 <= *d <= days_in_month(*y, *m)
- * The input values must be such that the internals don't overflow.
- * The way this routine is used, we don't get close.
- */
-static int
-normalize_y_m_d(int *y, int *m, int *d)
-{
-    int dim;            /* # of days in month */
-
-    /* This gets muddy:  the proper range for day can't be determined
-     * without knowing the correct month and year, but if day is, e.g.,
-     * plus or minus a million, the current month and year values make
-     * no sense (and may also be out of bounds themselves).
-     * Saying 12 months == 1 year should be non-controversial.
-     */
-    if (*m < 1 || *m > 12) {
-        --*m;
-        normalize_pair(y, m, 12);
-        ++*m;
-        /* |y| can't be bigger than about
-         * |original y| + |original m|/12 now.
-         */
-    }
-    assert(1 <= *m && *m <= 12);
-
-    /* Now only day can be out of bounds (year may also be out of bounds
-     * for a datetime object, but we don't care about that here).
-     * If day is out of bounds, what to do is arguable, but at least the
-     * method here is principled and explainable.
-     */
-    dim = days_in_month(*y, *m);
-    if (*d < 1 || *d > dim) {
-        /* Move day-1 days from the first of the month.  First try to
-         * get off cheap if we're only one day out of range
-         * (adjustments for timezone alone can't be worse than that).
-         */
-        if (*d == 0) {
-            --*m;
-            if (*m > 0)
-                *d = days_in_month(*y, *m);
-            else {
-                --*y;
-                *m = 12;
-                *d = 31;
-            }
-        }
-        else if (*d == dim + 1) {
-            /* move forward a day */
-            ++*m;
-            *d = 1;
-            if (*m > 12) {
-                *m = 1;
-                ++*y;
-            }
-        }
-        else {
-            int ordinal = ymd_to_ord(*y, *m, 1) +
-                                      *d - 1;
-            if (ordinal < 1 || ordinal > MAXORDINAL) {
-                goto error;
-            } else {
-                ord_to_ymd(ordinal, y, m, d);
-                return 0;
-            }
-        }
-    }
-    assert(*m > 0);
-    assert(*d > 0);
-    if (MINYEAR <= *y && *y <= MAXYEAR)
-        return 0;
- error:
-    PyErr_SetString(PyExc_OverflowError,
-            "date value out of range");
-    return -1;
-
-}
-
-/* Fiddle out-of-bounds months and days so that the result makes some kind
- * of sense.  The parameters are both inputs and outputs.  Returns < 0 on
- * failure, where failure means the adjusted year is out of bounds.
- */
-static int
-normalize_date(int *year, int *month, int *day)
-{
-    return normalize_y_m_d(year, month, day);
-}
-
-/* Force all the datetime fields into range.  The parameters are both
- * inputs and outputs.  Returns < 0 on error.
- */
-static int
-normalize_datetime(int *year, int *month, int *day,
-                   int *hour, int *minute, int *second,
-                   int *microsecond)
-{
-    normalize_pair(second, microsecond, 1000000);
-    normalize_pair(minute, second, 60);
-    normalize_pair(hour, minute, 60);
-    normalize_pair(day, hour, 24);
-    return normalize_date(year, month, day);
-}
-
-/* ---------------------------------------------------------------------------
- * Basic object allocation:  tp_alloc implementations.  These allocate
- * Python objects of the right size and type, and do the Python object-
- * initialization bit.  If there's not enough memory, they return NULL after
- * setting MemoryError.  All data members remain uninitialized trash.
- *
- * We abuse the tp_alloc "nitems" argument to communicate whether a tzinfo
- * member is needed.  This is ugly, imprecise, and possibly insecure.
- * tp_basicsize for the time and datetime types is set to the size of the
- * struct that has room for the tzinfo member, so subclasses in Python will
- * allocate enough space for a tzinfo member whether or not one is actually
- * needed.  That's the "ugly and imprecise" parts.  The "possibly insecure"
- * part is that PyType_GenericAlloc() (which subclasses in Python end up
- * using) just happens today to effectively ignore the nitems argument
- * when tp_itemsize is 0, which it is for these type objects.  If that
- * changes, perhaps the callers of tp_alloc slots in this file should
- * be changed to force a 0 nitems argument unless the type being allocated
- * is a base type implemented in this file (so that tp_alloc is time_alloc
- * or datetime_alloc below, which know about the nitems abuse).
- */
-
-static PyObject *
-time_alloc(PyTypeObject *type, Py_ssize_t aware)
-{
-    PyObject *self;
-
-    self = (PyObject *)
-        PyObject_MALLOC(aware ?
-                        sizeof(PyDateTime_Time) :
-                sizeof(_PyDateTime_BaseTime));
-    if (self == NULL)
-        return (PyObject *)PyErr_NoMemory();
-    PyObject_INIT(self, type);
-    return self;
-}
-
-static PyObject *
-datetime_alloc(PyTypeObject *type, Py_ssize_t aware)
-{
-    PyObject *self;
-
-    self = (PyObject *)
-        PyObject_MALLOC(aware ?
-                        sizeof(PyDateTime_DateTime) :
-                sizeof(_PyDateTime_BaseDateTime));
-    if (self == NULL)
-        return (PyObject *)PyErr_NoMemory();
-    PyObject_INIT(self, type);
-    return self;
-}
-
-/* ---------------------------------------------------------------------------
- * Helpers for setting object fields.  These work on pointers to the
- * appropriate base class.
- */
-
-/* For date and datetime. */
-static void
-set_date_fields(PyDateTime_Date *self, int y, int m, int d)
-{
-    self->hashcode = -1;
-    SET_YEAR(self, y);
-    SET_MONTH(self, m);
-    SET_DAY(self, d);
-}
-
-/* ---------------------------------------------------------------------------
- * Create various objects, mostly without range checking.
- */
-
-/* Create a date instance with no range checking. */
-static PyObject *
-new_date_ex(int year, int month, int day, PyTypeObject *type)
-{
-    PyDateTime_Date *self;
-
-    self = (PyDateTime_Date *) (type->tp_alloc(type, 0));
-    if (self != NULL)
-        set_date_fields(self, year, month, day);
-    return (PyObject *) self;
-}
-
-#define new_date(year, month, day) \
-    new_date_ex(year, month, day, &PyDateTime_DateType)
-
-/* Create a datetime instance with no range checking. */
-static PyObject *
-new_datetime_ex(int year, int month, int day, int hour, int minute,
-             int second, int usecond, PyObject *tzinfo, PyTypeObject *type)
-{
-    PyDateTime_DateTime *self;
-    char aware = tzinfo != Py_None;
-
-    self = (PyDateTime_DateTime *) (type->tp_alloc(type, aware));
-    if (self != NULL) {
-        self->hastzinfo = aware;
-        set_date_fields((PyDateTime_Date *)self, year, month, day);
-        DATE_SET_HOUR(self, hour);
-        DATE_SET_MINUTE(self, minute);
-        DATE_SET_SECOND(self, second);
-        DATE_SET_MICROSECOND(self, usecond);
-        if (aware) {
-            Py_INCREF(tzinfo);
-            self->tzinfo = tzinfo;
-        }
-    }
-    return (PyObject *)self;
-}
-
-#define new_datetime(y, m, d, hh, mm, ss, us, tzinfo)           \
-    new_datetime_ex(y, m, d, hh, mm, ss, us, tzinfo,            \
-                    &PyDateTime_DateTimeType)
-
-/* Create a time instance with no range checking. */
-static PyObject *
-new_time_ex(int hour, int minute, int second, int usecond,
-            PyObject *tzinfo, PyTypeObject *type)
-{
-    PyDateTime_Time *self;
-    char aware = tzinfo != Py_None;
-
-    self = (PyDateTime_Time *) (type->tp_alloc(type, aware));
-    if (self != NULL) {
-        self->hastzinfo = aware;
-        self->hashcode = -1;
-        TIME_SET_HOUR(self, hour);
-        TIME_SET_MINUTE(self, minute);
-        TIME_SET_SECOND(self, second);
-        TIME_SET_MICROSECOND(self, usecond);
-        if (aware) {
-            Py_INCREF(tzinfo);
-            self->tzinfo = tzinfo;
-        }
-    }
-    return (PyObject *)self;
-}
-
-#define new_time(hh, mm, ss, us, tzinfo)                \
-    new_time_ex(hh, mm, ss, us, tzinfo, &PyDateTime_TimeType)
-
-/* Create a timedelta instance.  Normalize the members iff normalize is
- * true.  Passing false is a speed optimization, if you know for sure
- * that seconds and microseconds are already in their proper ranges.  In any
- * case, raises OverflowError and returns NULL if the normalized days is out
- * of range).
- */
-static PyObject *
-new_delta_ex(int days, int seconds, int microseconds, int normalize,
-             PyTypeObject *type)
-{
-    PyDateTime_Delta *self;
-
-    if (normalize)
-        normalize_d_s_us(&days, &seconds, &microseconds);
-    assert(0 <= seconds && seconds < 24*3600);
-    assert(0 <= microseconds && microseconds < 1000000);
-
-    if (check_delta_day_range(days) < 0)
-        return NULL;
-
-    self = (PyDateTime_Delta *) (type->tp_alloc(type, 0));
-    if (self != NULL) {
-        self->hashcode = -1;
-        SET_TD_DAYS(self, days);
-        SET_TD_SECONDS(self, seconds);
-        SET_TD_MICROSECONDS(self, microseconds);
-    }
-    return (PyObject *) self;
-}
-
-#define new_delta(d, s, us, normalize)  \
-    new_delta_ex(d, s, us, normalize, &PyDateTime_DeltaType)
-
-
-typedef struct
-{
-    PyObject_HEAD
-    PyObject *offset;
-    PyObject *name;
-} PyDateTime_TimeZone;
-
-PyObject *PyDateTime_TimeZone_UTC;
-
-/* Create new timezone instance checking offset range.  This
-   function does not check the name argument.  Caller must assure
-   that offset is a timedelta instance and name is either NULL
-   or a unicode object. */
-static PyObject *
-new_timezone(PyObject *offset, PyObject *name)
-{
-    PyDateTime_TimeZone *self;
-    PyTypeObject *type = &PyDateTime_TimeZoneType;
-
-    assert(offset != NULL);
-    assert(PyDelta_Check(offset));
-    assert(name == NULL || PyUnicode_Check(name));
-
-    if (GET_TD_MICROSECONDS(offset) != 0 || GET_TD_SECONDS(offset) % 60 != 0) {
-        PyErr_Format(PyExc_ValueError, "offset must be a timedelta"
-                     " representing a whole number of minutes");
-        return NULL;
-    }
-    if ((GET_TD_DAYS(offset) == -1 && GET_TD_SECONDS(offset) == 0) ||
-        GET_TD_DAYS(offset) < -1 || GET_TD_DAYS(offset) >= 1) {
-        PyErr_Format(PyExc_ValueError, "offset must be a timedelta"
-                     " strictly between -timedelta(hours=24) and"
-                     " timedelta(hours=24).");
-        return NULL;
-    }
-
-    self = (PyDateTime_TimeZone *)(type->tp_alloc(type, 0));
-    if (self == NULL) {
-        return NULL;
-    }
-    Py_INCREF(offset);
-    self->offset = offset;
-    Py_XINCREF(name);
-    self->name = name;
-    return (PyObject *)self;
-}
-
-/* ---------------------------------------------------------------------------
- * tzinfo helpers.
- */
-
-/* Ensure that p is None or of a tzinfo subclass.  Return 0 if OK; if not
- * raise TypeError and return -1.
- */
-static int
-check_tzinfo_subclass(PyObject *p)
-{
-    if (p == Py_None || PyTZInfo_Check(p))
-        return 0;
-    PyErr_Format(PyExc_TypeError,
-                 "tzinfo argument must be None or of a tzinfo subclass, "
-                 "not type '%s'",
-                 Py_TYPE(p)->tp_name);
-    return -1;
-}
-
-/* If self has a tzinfo member, return a BORROWED reference to it.  Else
- * return NULL, which is NOT AN ERROR.  There are no error returns here,
- * and the caller must not decref the result.
- */
-static PyObject *
-get_tzinfo_member(PyObject *self)
-{
-    PyObject *tzinfo = NULL;
-
-    if (PyDateTime_Check(self) && HASTZINFO(self))
-        tzinfo = ((PyDateTime_DateTime *)self)->tzinfo;
-    else if (PyTime_Check(self) && HASTZINFO(self))
-        tzinfo = ((PyDateTime_Time *)self)->tzinfo;
-
-    return tzinfo;
-}
-
-/* Call getattr(tzinfo, name)(tzinfoarg), and check the result.  tzinfo must
- * be an instance of the tzinfo class.  If the method returns None, this
- * returns None.  If the method doesn't return None or timedelta, TypeError is
- * raised and this returns NULL.  If it returns a timedelta and the value is
- * out of range or isn't a whole number of minutes, ValueError is raised and
- * this returns NULL.  Else result is returned.
- */
-static PyObject *
-call_tzinfo_method(PyObject *tzinfo, char *name, PyObject *tzinfoarg)
-{
-    PyObject *offset;
-
-    assert(tzinfo != NULL);
-    assert(PyTZInfo_Check(tzinfo) || tzinfo == Py_None);
-    assert(tzinfoarg != NULL);
-
-    if (tzinfo == Py_None)
-        Py_RETURN_NONE;
-    offset = PyObject_CallMethod(tzinfo, name, "O", tzinfoarg);
-    if (offset == Py_None || offset == NULL)
-        return offset;
-    if (PyDelta_Check(offset)) {
-        if (GET_TD_MICROSECONDS(offset) != 0 || GET_TD_SECONDS(offset) % 60 != 0) {
-            Py_DECREF(offset);
-            PyErr_Format(PyExc_ValueError, "offset must be a timedelta"
-                         " representing a whole number of minutes");
-            return NULL;
-        }
-        if ((GET_TD_DAYS(offset) == -1 && GET_TD_SECONDS(offset) == 0) ||
-            GET_TD_DAYS(offset) < -1 || GET_TD_DAYS(offset) >= 1) {
-            Py_DECREF(offset);
-            PyErr_Format(PyExc_ValueError, "offset must be a timedelta"
-                         " strictly between -timedelta(hours=24) and"
-                         " timedelta(hours=24).");
-            return NULL;
-        }
-    }
-    else {
-        Py_DECREF(offset);
-        PyErr_Format(PyExc_TypeError,
-                     "tzinfo.%s() must return None or "
-                     "timedelta, not '%.200s'",
-                     name, Py_TYPE(offset)->tp_name);
-        return NULL;
-    }
-
-    return offset;
-}
-
-/* Call tzinfo.utcoffset(tzinfoarg), and extract an integer from the
- * result.  tzinfo must be an instance of the tzinfo class.  If utcoffset()
- * returns None, call_utcoffset returns 0 and sets *none to 1.  If uctoffset()
- * doesn't return None or timedelta, TypeError is raised and this returns -1.
- * If utcoffset() returns an invalid timedelta (out of range, or not a whole
- * # of minutes), ValueError is raised and this returns -1.  Else *none is
- * set to 0 and the offset is returned (as int # of minutes east of UTC).
- */
-static PyObject *
-call_utcoffset(PyObject *tzinfo, PyObject *tzinfoarg)
-{
-    return call_tzinfo_method(tzinfo, "utcoffset", tzinfoarg);
-}
-
-/* Call tzinfo.dst(tzinfoarg), and extract an integer from the
- * result.  tzinfo must be an instance of the tzinfo class.  If dst()
- * returns None, call_dst returns 0 and sets *none to 1.  If dst()
- & doesn't return None or timedelta, TypeError is raised and this
- * returns -1.  If dst() returns an invalid timedelta for a UTC offset,
- * ValueError is raised and this returns -1.  Else *none is set to 0 and
- * the offset is returned (as an int # of minutes east of UTC).
- */
-static PyObject *
-call_dst(PyObject *tzinfo, PyObject *tzinfoarg)
-{
-    return call_tzinfo_method(tzinfo, "dst", tzinfoarg);
-}
-
-/* Call tzinfo.tzname(tzinfoarg), and return the result.  tzinfo must be
- * an instance of the tzinfo class or None.  If tzinfo isn't None, and
- * tzname() doesn't return None or a string, TypeError is raised and this
- * returns NULL.  If the result is a string, we ensure it is a Unicode
- * string.
- */
-static PyObject *
-call_tzname(PyObject *tzinfo, PyObject *tzinfoarg)
-{
-    PyObject *result;
-
-    assert(tzinfo != NULL);
-    assert(check_tzinfo_subclass(tzinfo) >= 0);
-    assert(tzinfoarg != NULL);
-
-    if (tzinfo == Py_None)
-        Py_RETURN_NONE;
-
-    result = PyObject_CallMethod(tzinfo, "tzname", "O", tzinfoarg);
-
-    if (result == NULL || result == Py_None)
-        return result;
-
-    if (!PyUnicode_Check(result)) {
-        PyErr_Format(PyExc_TypeError, "tzinfo.tzname() must "
-                     "return None or a string, not '%s'",
-                     Py_TYPE(result)->tp_name);
-        Py_DECREF(result);
-        result = NULL;
-    }
-
-    return result;
-}
-
-/* repr is like "someclass(arg1, arg2)".  If tzinfo isn't None,
- * stuff
- *     ", tzinfo=" + repr(tzinfo)
- * before the closing ")".
- */
-static PyObject *
-append_keyword_tzinfo(PyObject *repr, PyObject *tzinfo)
-{
-    PyObject *temp;
-
-    assert(PyUnicode_Check(repr));
-    assert(tzinfo);
-    if (tzinfo == Py_None)
-        return repr;
-    /* Get rid of the trailing ')'. */
-    assert(PyUnicode_AS_UNICODE(repr)[PyUnicode_GET_SIZE(repr)-1] == ')');
-    temp = PyUnicode_FromUnicode(PyUnicode_AS_UNICODE(repr),
-                                      PyUnicode_GET_SIZE(repr) - 1);
-    Py_DECREF(repr);
-    if (temp == NULL)
-        return NULL;
-    repr = PyUnicode_FromFormat("%U, tzinfo=%R)", temp, tzinfo);
-    Py_DECREF(temp);
-    return repr;
-}
-
-/* ---------------------------------------------------------------------------
- * String format helpers.
- */
-
-static PyObject *
-format_ctime(PyDateTime_Date *date, int hours, int minutes, int seconds)
-{
-    static const char *DayNames[] = {
-        "Mon", "Tue", "Wed", "Thu", "Fri", "Sat", "Sun"
-    };
-    static const char *MonthNames[] = {
-        "Jan", "Feb", "Mar", "Apr", "May", "Jun",
-        "Jul", "Aug", "Sep", "Oct", "Nov", "Dec"
-    };
-
-    int wday = weekday(GET_YEAR(date), GET_MONTH(date), GET_DAY(date));
-
-    return PyUnicode_FromFormat("%s %s %2d %02d:%02d:%02d %04d",
-                                DayNames[wday], MonthNames[GET_MONTH(date)-1],
-                                GET_DAY(date), hours, minutes, seconds,
-                                GET_YEAR(date));
-}
-
-static PyObject *delta_negative(PyDateTime_Delta *self);
-
-/* Add an hours & minutes UTC offset string to buf.  buf has no more than
- * buflen bytes remaining.  The UTC offset is gotten by calling
- * tzinfo.uctoffset(tzinfoarg).  If that returns None, \0 is stored into
- * *buf, and that's all.  Else the returned value is checked for sanity (an
- * integer in range), and if that's OK it's converted to an hours & minutes
- * string of the form
- *   sign HH sep MM
- * Returns 0 if everything is OK.  If the return value from utcoffset() is
- * bogus, an appropriate exception is set and -1 is returned.
- */
-static int
-format_utcoffset(char *buf, size_t buflen, const char *sep,
-                PyObject *tzinfo, PyObject *tzinfoarg)
-{
-    PyObject *offset;
-    int hours, minutes, seconds;
-    char sign;
-
-    assert(buflen >= 1);
-
-    offset = call_utcoffset(tzinfo, tzinfoarg);
-    if (offset == NULL)
-        return -1;
-    if (offset == Py_None) {
-        Py_DECREF(offset);
-        *buf = '\0';
-        return 0;
-    }
-    /* Offset is normalized, so it is negative if days < 0 */
-    if (GET_TD_DAYS(offset) < 0) {
-        PyObject *temp = offset;
-        sign = '-';
-        offset = delta_negative((PyDateTime_Delta *)offset);
-        Py_DECREF(temp);
-        if (offset == NULL)
-            return -1;
-    }
-    else {
-        sign = '+';
-    }
-    /* Offset is not negative here. */
-    seconds = GET_TD_SECONDS(offset);
-    Py_DECREF(offset);
-    minutes = divmod(seconds, 60, &seconds);
-    hours = divmod(minutes, 60, &minutes);
-    assert(seconds == 0);
-    /* XXX ignore sub-minute data, curently not allowed. */
-    PyOS_snprintf(buf, buflen, "%c%02d%s%02d", sign, hours, sep, minutes);
-
-    return 0;
-}
-
-static PyObject *
-make_Zreplacement(PyObject *object, PyObject *tzinfoarg)
-{
-    PyObject *temp;
-    PyObject *tzinfo = get_tzinfo_member(object);
-    PyObject *Zreplacement = PyUnicode_FromStringAndSize(NULL, 0);
-    if (Zreplacement == NULL)
-        return NULL;
-    if (tzinfo == Py_None || tzinfo == NULL)
-        return Zreplacement;
-
-    assert(tzinfoarg != NULL);
-    temp = call_tzname(tzinfo, tzinfoarg);
-    if (temp == NULL)
-        goto Error;
-    if (temp == Py_None) {
-        Py_DECREF(temp);
-        return Zreplacement;
-    }
-
-    assert(PyUnicode_Check(temp));
-    /* Since the tzname is getting stuffed into the
-     * format, we have to double any % signs so that
-     * strftime doesn't treat them as format codes.
-     */
-    Py_DECREF(Zreplacement);
-    Zreplacement = PyObject_CallMethod(temp, "replace", "ss", "%", "%%");
-    Py_DECREF(temp);
-    if (Zreplacement == NULL)
-        return NULL;
-    if (!PyUnicode_Check(Zreplacement)) {
-        PyErr_SetString(PyExc_TypeError,
-                        "tzname.replace() did not return a string");
-        goto Error;
-    }
-    return Zreplacement;
-
-  Error:
-    Py_DECREF(Zreplacement);
-    return NULL;
-}
-
-static PyObject *
-make_freplacement(PyObject *object)
-{
-    char freplacement[64];
-    if (PyTime_Check(object))
-        sprintf(freplacement, "%06d", TIME_GET_MICROSECOND(object));
-    else if (PyDateTime_Check(object))
-        sprintf(freplacement, "%06d", DATE_GET_MICROSECOND(object));
-    else
-        sprintf(freplacement, "%06d", 0);
-
-    return PyBytes_FromStringAndSize(freplacement, strlen(freplacement));
-}
-
-/* I sure don't want to reproduce the strftime code from the time module,
- * so this imports the module and calls it.  All the hair is due to
- * giving special meanings to the %z, %Z and %f format codes via a
- * preprocessing step on the format string.
- * tzinfoarg is the argument to pass to the object's tzinfo method, if
- * needed.
- */
-static PyObject *
-wrap_strftime(PyObject *object, PyObject *format, PyObject *timetuple,
-              PyObject *tzinfoarg)
-{
-    PyObject *result = NULL;            /* guilty until proved innocent */
-
-    PyObject *zreplacement = NULL;      /* py string, replacement for %z */
-    PyObject *Zreplacement = NULL;      /* py string, replacement for %Z */
-    PyObject *freplacement = NULL;      /* py string, replacement for %f */
-
-    const char *pin;            /* pointer to next char in input format */
-    Py_ssize_t flen;            /* length of input format */
-    char ch;                    /* next char in input format */
-
-    PyObject *newfmt = NULL;            /* py string, the output format */
-    char *pnew;         /* pointer to available byte in output format */
-    size_t totalnew;            /* number bytes total in output format buffer,
-                               exclusive of trailing \0 */
-    size_t usednew;     /* number bytes used so far in output format buffer */
-
-    const char *ptoappend;      /* ptr to string to append to output buffer */
-    Py_ssize_t ntoappend;       /* # of bytes to append to output buffer */
-
-    assert(object && format && timetuple);
-    assert(PyUnicode_Check(format));
-    /* Convert the input format to a C string and size */
-    pin = _PyUnicode_AsStringAndSize(format, &flen);
-    if (!pin)
-        return NULL;
-
-    /* Give up if the year is before 1900.
-     * Python strftime() plays games with the year, and different
-     * games depending on whether envar PYTHON2K is set.  This makes
-     * years before 1900 a nightmare, even if the platform strftime
-     * supports them (and not all do).
-     * We could get a lot farther here by avoiding Python's strftime
-     * wrapper and calling the C strftime() directly, but that isn't
-     * an option in the Python implementation of this module.
-     */
-    {
-        long year;
-        PyObject *pyyear = PySequence_GetItem(timetuple, 0);
-        if (pyyear == NULL) return NULL;
-        assert(PyLong_Check(pyyear));
-        year = PyLong_AsLong(pyyear);
-        Py_DECREF(pyyear);
-        if (year < 1900) {
-            PyErr_Format(PyExc_ValueError, "year=%ld is before "
-                         "1900; the datetime strftime() "
-                         "methods require year >= 1900",
-                         year);
-            return NULL;
-        }
-    }
-
-    /* Scan the input format, looking for %z/%Z/%f escapes, building
-     * a new format.  Since computing the replacements for those codes
-     * is expensive, don't unless they're actually used.
-     */
-    if (flen > INT_MAX - 1) {
-        PyErr_NoMemory();
-        goto Done;
-    }
-
-    totalnew = flen + 1;        /* realistic if no %z/%Z */
-    newfmt = PyBytes_FromStringAndSize(NULL, totalnew);
-    if (newfmt == NULL) goto Done;
-    pnew = PyBytes_AsString(newfmt);
-    usednew = 0;
-
-    while ((ch = *pin++) != '\0') {
-        if (ch != '%') {
-            ptoappend = pin - 1;
-            ntoappend = 1;
-        }
-        else if ((ch = *pin++) == '\0') {
-            /* There's a lone trailing %; doesn't make sense. */
-            PyErr_SetString(PyExc_ValueError, "strftime format "
-                            "ends with raw %");
-            goto Done;
-        }
-        /* A % has been seen and ch is the character after it. */
-        else if (ch == 'z') {
-            if (zreplacement == NULL) {
-                /* format utcoffset */
-                char buf[100];
-                PyObject *tzinfo = get_tzinfo_member(object);
-                zreplacement = PyBytes_FromStringAndSize("", 0);
-                if (zreplacement == NULL) goto Done;
-                if (tzinfo != Py_None && tzinfo != NULL) {
-                    assert(tzinfoarg != NULL);
-                    if (format_utcoffset(buf,
-                                         sizeof(buf),
-                                         "",
-                                         tzinfo,
-                                         tzinfoarg) < 0)
-                        goto Done;
-                    Py_DECREF(zreplacement);
-                    zreplacement =
-                      PyBytes_FromStringAndSize(buf,
-                                               strlen(buf));
-                    if (zreplacement == NULL)
-                        goto Done;
-                }
-            }
-            assert(zreplacement != NULL);
-            ptoappend = PyBytes_AS_STRING(zreplacement);
-            ntoappend = PyBytes_GET_SIZE(zreplacement);
-        }
-        else if (ch == 'Z') {
-            /* format tzname */
-            if (Zreplacement == NULL) {
-                Zreplacement = make_Zreplacement(object,
-                                                 tzinfoarg);
-                if (Zreplacement == NULL)
-                    goto Done;
-            }
-            assert(Zreplacement != NULL);
-            assert(PyUnicode_Check(Zreplacement));
-            ptoappend = _PyUnicode_AsStringAndSize(Zreplacement,
-                                                  &ntoappend);
-            ntoappend = Py_SIZE(Zreplacement);
-        }
-        else if (ch == 'f') {
-            /* format microseconds */
-            if (freplacement == NULL) {
-                freplacement = make_freplacement(object);
-                if (freplacement == NULL)
-                    goto Done;
-            }
-            assert(freplacement != NULL);
-            assert(PyBytes_Check(freplacement));
-            ptoappend = PyBytes_AS_STRING(freplacement);
-            ntoappend = PyBytes_GET_SIZE(freplacement);
-        }
-        else {
-            /* percent followed by neither z nor Z */
-            ptoappend = pin - 2;
-            ntoappend = 2;
-        }
-
-        /* Append the ntoappend chars starting at ptoappend to
-         * the new format.
-         */
-        if (ntoappend == 0)
-            continue;
-        assert(ptoappend != NULL);
-        assert(ntoappend > 0);
-        while (usednew + ntoappend > totalnew) {
-            size_t bigger = totalnew << 1;
-            if ((bigger >> 1) != totalnew) { /* overflow */
-                PyErr_NoMemory();
-                goto Done;
-            }
-            if (_PyBytes_Resize(&newfmt, bigger) < 0)
-                goto Done;
-            totalnew = bigger;
-            pnew = PyBytes_AsString(newfmt) + usednew;
-        }
-        memcpy(pnew, ptoappend, ntoappend);
-        pnew += ntoappend;
-        usednew += ntoappend;
-        assert(usednew <= totalnew);
-    }  /* end while() */
-
-    if (_PyBytes_Resize(&newfmt, usednew) < 0)
-        goto Done;
-    {
-        PyObject *format;
-        PyObject *time = PyImport_ImportModuleNoBlock("time");
-        if (time == NULL)
-            goto Done;
-        format = PyUnicode_FromString(PyBytes_AS_STRING(newfmt));
-        if (format != NULL) {
-            result = PyObject_CallMethod(time, "strftime", "OO",
-                                         format, timetuple, NULL);
-            Py_DECREF(format);
-        }
-        Py_DECREF(time);
-    }
- Done:
-    Py_XDECREF(freplacement);
-    Py_XDECREF(zreplacement);
-    Py_XDECREF(Zreplacement);
-    Py_XDECREF(newfmt);
-    return result;
-}
-
-/* ---------------------------------------------------------------------------
- * Wrap functions from the time module.  These aren't directly available
- * from C.  Perhaps they should be.
- */
-
-/* Call time.time() and return its result (a Python float). */
-static PyObject *
-time_time(void)
-{
-    PyObject *result = NULL;
-    PyObject *time = PyImport_ImportModuleNoBlock("time");
-
-    if (time != NULL) {
-        result = PyObject_CallMethod(time, "time", "()");
-        Py_DECREF(time);
-    }
-    return result;
-}
-
-/* Build a time.struct_time.  The weekday and day number are automatically
- * computed from the y,m,d args.
- */
-static PyObject *
-build_struct_time(int y, int m, int d, int hh, int mm, int ss, int dstflag)
-{
-    PyObject *time;
-    PyObject *result = NULL;
-
-    time = PyImport_ImportModuleNoBlock("time");
-    if (time != NULL) {
-        result = PyObject_CallMethod(time, "struct_time",
-                                     "((iiiiiiiii))",
-                                     y, m, d,
-                                     hh, mm, ss,
-                                     weekday(y, m, d),
-                                     days_before_month(y, m) + d,
-                                     dstflag);
-        Py_DECREF(time);
-    }
-    return result;
-}
-
-/* ---------------------------------------------------------------------------
- * Miscellaneous helpers.
- */
-
-/* For various reasons, we need to use tp_richcompare instead of tp_reserved.
- * The comparisons here all most naturally compute a cmp()-like result.
- * This little helper turns that into a bool result for rich comparisons.
- */
-static PyObject *
-diff_to_bool(int diff, int op)
-{
-    PyObject *result;
-    int istrue;
-
-    switch (op) {
-        case Py_EQ: istrue = diff == 0; break;
-        case Py_NE: istrue = diff != 0; break;
-        case Py_LE: istrue = diff <= 0; break;
-        case Py_GE: istrue = diff >= 0; break;
-        case Py_LT: istrue = diff < 0; break;
-        case Py_GT: istrue = diff > 0; break;
-        default:
-            assert(! "op unknown");
-            istrue = 0; /* To shut up compiler */
-    }
-    result = istrue ? Py_True : Py_False;
-    Py_INCREF(result);
-    return result;
-}
-
-/* Raises a "can't compare" TypeError and returns NULL. */
-static PyObject *
-cmperror(PyObject *a, PyObject *b)
-{
-    PyErr_Format(PyExc_TypeError,
-                 "can't compare %s to %s",
-                 Py_TYPE(a)->tp_name, Py_TYPE(b)->tp_name);
-    return NULL;
-}
-
-/* ---------------------------------------------------------------------------
- * Cached Python objects; these are set by the module init function.
- */
-
-/* Conversion factors. */
-static PyObject *us_per_us = NULL;      /* 1 */
-static PyObject *us_per_ms = NULL;      /* 1000 */
-static PyObject *us_per_second = NULL;  /* 1000000 */
-static PyObject *us_per_minute = NULL;  /* 1e6 * 60 as Python int */
-static PyObject *us_per_hour = NULL;    /* 1e6 * 3600 as Python long */
-static PyObject *us_per_day = NULL;     /* 1e6 * 3600 * 24 as Python long */
-static PyObject *us_per_week = NULL;    /* 1e6*3600*24*7 as Python long */
-static PyObject *seconds_per_day = NULL; /* 3600*24 as Python int */
-
-/* ---------------------------------------------------------------------------
- * Class implementations.
- */
-
-/*
- * PyDateTime_Delta implementation.
- */
-
-/* Convert a timedelta to a number of us,
- *      (24*3600*self.days + self.seconds)*1000000 + self.microseconds
- * as a Python int or long.
- * Doing mixed-radix arithmetic by hand instead is excruciating in C,
- * due to ubiquitous overflow possibilities.
- */
-static PyObject *
-delta_to_microseconds(PyDateTime_Delta *self)
-{
-    PyObject *x1 = NULL;
-    PyObject *x2 = NULL;
-    PyObject *x3 = NULL;
-    PyObject *result = NULL;
-
-    x1 = PyLong_FromLong(GET_TD_DAYS(self));
-    if (x1 == NULL)
-        goto Done;
-    x2 = PyNumber_Multiply(x1, seconds_per_day);        /* days in seconds */
-    if (x2 == NULL)
-        goto Done;
-    Py_DECREF(x1);
-    x1 = NULL;
-
-    /* x2 has days in seconds */
-    x1 = PyLong_FromLong(GET_TD_SECONDS(self));         /* seconds */
-    if (x1 == NULL)
-        goto Done;
-    x3 = PyNumber_Add(x1, x2);          /* days and seconds in seconds */
-    if (x3 == NULL)
-        goto Done;
-    Py_DECREF(x1);
-    Py_DECREF(x2);
-    x1 = x2 = NULL;
-
-    /* x3 has days+seconds in seconds */
-    x1 = PyNumber_Multiply(x3, us_per_second);          /* us */
-    if (x1 == NULL)
-        goto Done;
-    Py_DECREF(x3);
-    x3 = NULL;
-
-    /* x1 has days+seconds in us */
-    x2 = PyLong_FromLong(GET_TD_MICROSECONDS(self));
-    if (x2 == NULL)
-        goto Done;
-    result = PyNumber_Add(x1, x2);
-
-Done:
-    Py_XDECREF(x1);
-    Py_XDECREF(x2);
-    Py_XDECREF(x3);
-    return result;
-}
-
-/* Convert a number of us (as a Python int or long) to a timedelta.
- */
-static PyObject *
-microseconds_to_delta_ex(PyObject *pyus, PyTypeObject *type)
-{
-    int us;
-    int s;
-    int d;
-    long temp;
-
-    PyObject *tuple = NULL;
-    PyObject *num = NULL;
-    PyObject *result = NULL;
-
-    tuple = PyNumber_Divmod(pyus, us_per_second);
-    if (tuple == NULL)
-        goto Done;
-
-    num = PyTuple_GetItem(tuple, 1);            /* us */
-    if (num == NULL)
-        goto Done;
-    temp = PyLong_AsLong(num);
-    num = NULL;
-    if (temp == -1 && PyErr_Occurred())
-        goto Done;
-    assert(0 <= temp && temp < 1000000);
-    us = (int)temp;
-    if (us < 0) {
-        /* The divisor was positive, so this must be an error. */
-        assert(PyErr_Occurred());
-        goto Done;
-    }
-
-    num = PyTuple_GetItem(tuple, 0);            /* leftover seconds */
-    if (num == NULL)
-        goto Done;
-    Py_INCREF(num);
-    Py_DECREF(tuple);
-
-    tuple = PyNumber_Divmod(num, seconds_per_day);
-    if (tuple == NULL)
-        goto Done;
-    Py_DECREF(num);
-
-    num = PyTuple_GetItem(tuple, 1);            /* seconds */
-    if (num == NULL)
-        goto Done;
-    temp = PyLong_AsLong(num);
-    num = NULL;
-    if (temp == -1 && PyErr_Occurred())
-        goto Done;
-    assert(0 <= temp && temp < 24*3600);
-    s = (int)temp;
-
-    if (s < 0) {
-        /* The divisor was positive, so this must be an error. */
-        assert(PyErr_Occurred());
-        goto Done;
-    }
-
-    num = PyTuple_GetItem(tuple, 0);            /* leftover days */
-    if (num == NULL)
-        goto Done;
-    Py_INCREF(num);
-    temp = PyLong_AsLong(num);
-    if (temp == -1 && PyErr_Occurred())
-        goto Done;
-    d = (int)temp;
-    if ((long)d != temp) {
-        PyErr_SetString(PyExc_OverflowError, "normalized days too "
-                        "large to fit in a C int");
-        goto Done;
-    }
-    result = new_delta_ex(d, s, us, 0, type);
-
-Done:
-    Py_XDECREF(tuple);
-    Py_XDECREF(num);
-    return result;
-}
-
-#define microseconds_to_delta(pymicros) \
-    microseconds_to_delta_ex(pymicros, &PyDateTime_DeltaType)
-
-static PyObject *
-multiply_int_timedelta(PyObject *intobj, PyDateTime_Delta *delta)
-{
-    PyObject *pyus_in;
-    PyObject *pyus_out;
-    PyObject *result;
-
-    pyus_in = delta_to_microseconds(delta);
-    if (pyus_in == NULL)
-        return NULL;
-
-    pyus_out = PyNumber_Multiply(pyus_in, intobj);
-    Py_DECREF(pyus_in);
-    if (pyus_out == NULL)
-        return NULL;
-
-    result = microseconds_to_delta(pyus_out);
-    Py_DECREF(pyus_out);
-    return result;
-}
-
-static PyObject *
-multiply_float_timedelta(PyObject *floatobj, PyDateTime_Delta *delta)
-{
-    PyObject *result = NULL;
-    PyObject *pyus_in = NULL, *temp, *pyus_out;
-    PyObject *ratio = NULL;
-
-    pyus_in = delta_to_microseconds(delta);
-    if (pyus_in == NULL)
-        return NULL;
-    ratio = PyObject_CallMethod(floatobj, "as_integer_ratio", NULL);
-    if (ratio == NULL)
-        goto error;
-    temp = PyNumber_Multiply(pyus_in, PyTuple_GET_ITEM(ratio, 0));
-    Py_DECREF(pyus_in);
-    pyus_in = NULL;
-    if (temp == NULL)
-        goto error;
-    pyus_out = divide_nearest(temp, PyTuple_GET_ITEM(ratio, 1));
-    Py_DECREF(temp);
-    if (pyus_out == NULL)
-        goto error;
-    result = microseconds_to_delta(pyus_out);
-    Py_DECREF(pyus_out);
- error:
-    Py_XDECREF(pyus_in);
-    Py_XDECREF(ratio);
-
-    return result;
-}
-
-static PyObject *
-divide_timedelta_int(PyDateTime_Delta *delta, PyObject *intobj)
-{
-    PyObject *pyus_in;
-    PyObject *pyus_out;
-    PyObject *result;
-
-    pyus_in = delta_to_microseconds(delta);
-    if (pyus_in == NULL)
-        return NULL;
-
-    pyus_out = PyNumber_FloorDivide(pyus_in, intobj);
-    Py_DECREF(pyus_in);
-    if (pyus_out == NULL)
-        return NULL;
-
-    result = microseconds_to_delta(pyus_out);
-    Py_DECREF(pyus_out);
-    return result;
-}
-
-static PyObject *
-divide_timedelta_timedelta(PyDateTime_Delta *left, PyDateTime_Delta *right)
-{
-    PyObject *pyus_left;
-    PyObject *pyus_right;
-    PyObject *result;
-
-    pyus_left = delta_to_microseconds(left);
-    if (pyus_left == NULL)
-        return NULL;
-
-    pyus_right = delta_to_microseconds(right);
-    if (pyus_right == NULL)     {
-        Py_DECREF(pyus_left);
-        return NULL;
-    }
-
-    result = PyNumber_FloorDivide(pyus_left, pyus_right);
-    Py_DECREF(pyus_left);
-    Py_DECREF(pyus_right);
-    return result;
-}
-
-static PyObject *
-truedivide_timedelta_timedelta(PyDateTime_Delta *left, PyDateTime_Delta *right)
-{
-    PyObject *pyus_left;
-    PyObject *pyus_right;
-    PyObject *result;
-
-    pyus_left = delta_to_microseconds(left);
-    if (pyus_left == NULL)
-        return NULL;
-
-    pyus_right = delta_to_microseconds(right);
-    if (pyus_right == NULL)     {
-        Py_DECREF(pyus_left);
-        return NULL;
-    }
-
-    result = PyNumber_TrueDivide(pyus_left, pyus_right);
-    Py_DECREF(pyus_left);
-    Py_DECREF(pyus_right);
-    return result;
-}
-
-static PyObject *
-truedivide_timedelta_float(PyDateTime_Delta *delta, PyObject *f)
-{
-    PyObject *result = NULL;
-    PyObject *pyus_in = NULL, *temp, *pyus_out;
-    PyObject *ratio = NULL;
-
-    pyus_in = delta_to_microseconds(delta);
-    if (pyus_in == NULL)
-        return NULL;
-    ratio = PyObject_CallMethod(f, "as_integer_ratio", NULL);
-    if (ratio == NULL)
-        goto error;
-    temp = PyNumber_Multiply(pyus_in, PyTuple_GET_ITEM(ratio, 1));
-    Py_DECREF(pyus_in);
-    pyus_in = NULL;
-    if (temp == NULL)
-        goto error;
-    pyus_out = divide_nearest(temp, PyTuple_GET_ITEM(ratio, 0));
-    Py_DECREF(temp);
-    if (pyus_out == NULL)
-        goto error;
-    result = microseconds_to_delta(pyus_out);
-    Py_DECREF(pyus_out);
- error:
-    Py_XDECREF(pyus_in);
-    Py_XDECREF(ratio);
-
-    return result;
-}
-
-static PyObject *
-truedivide_timedelta_int(PyDateTime_Delta *delta, PyObject *i)
-{
-    PyObject *result;
-    PyObject *pyus_in, *pyus_out;
-    pyus_in = delta_to_microseconds(delta);
-    if (pyus_in == NULL)
-        return NULL;
-    pyus_out = divide_nearest(pyus_in, i);
-    Py_DECREF(pyus_in);
-    if (pyus_out == NULL)
-        return NULL;
-    result = microseconds_to_delta(pyus_out);
-    Py_DECREF(pyus_out);
-
-    return result;
-}
-
-static PyObject *
-delta_add(PyObject *left, PyObject *right)
-{
-    PyObject *result = Py_NotImplemented;
-
-    if (PyDelta_Check(left) && PyDelta_Check(right)) {
-        /* delta + delta */
-        /* The C-level additions can't overflow because of the
-         * invariant bounds.
-         */
-        int days = GET_TD_DAYS(left) + GET_TD_DAYS(right);
-        int seconds = GET_TD_SECONDS(left) + GET_TD_SECONDS(right);
-        int microseconds = GET_TD_MICROSECONDS(left) +
-                           GET_TD_MICROSECONDS(right);
-        result = new_delta(days, seconds, microseconds, 1);
-    }
-
-    if (result == Py_NotImplemented)
-        Py_INCREF(result);
-    return result;
-}
-
-static PyObject *
-delta_negative(PyDateTime_Delta *self)
-{
-    return new_delta(-GET_TD_DAYS(self),
-                     -GET_TD_SECONDS(self),
-                     -GET_TD_MICROSECONDS(self),
-                     1);
-}
-
-static PyObject *
-delta_positive(PyDateTime_Delta *self)
-{
-    /* Could optimize this (by returning self) if this isn't a
-     * subclass -- but who uses unary + ?  Approximately nobody.
-     */
-    return new_delta(GET_TD_DAYS(self),
-                     GET_TD_SECONDS(self),
-                     GET_TD_MICROSECONDS(self),
-                     0);
-}
-
-static PyObject *
-delta_abs(PyDateTime_Delta *self)
-{
-    PyObject *result;
-
-    assert(GET_TD_MICROSECONDS(self) >= 0);
-    assert(GET_TD_SECONDS(self) >= 0);
-
-    if (GET_TD_DAYS(self) < 0)
-        result = delta_negative(self);
-    else
-        result = delta_positive(self);
-
-    return result;
-}
-
-static PyObject *
-delta_subtract(PyObject *left, PyObject *right)
-{
-    PyObject *result = Py_NotImplemented;
-
-    if (PyDelta_Check(left) && PyDelta_Check(right)) {
-        /* delta - delta */
-        PyObject *minus_right = PyNumber_Negative(right);
-        if (minus_right) {
-            result = delta_add(left, minus_right);
-            Py_DECREF(minus_right);
-        }
-        else
-            result = NULL;
-    }
-
-    if (result == Py_NotImplemented)
-        Py_INCREF(result);
-    return result;
-}
-
-static int
-delta_cmp(PyObject *self, PyObject *other)
-{
-    int diff = GET_TD_DAYS(self) - GET_TD_DAYS(other);
-    if (diff == 0) {
-        diff = GET_TD_SECONDS(self) - GET_TD_SECONDS(other);
-        if (diff == 0)
-            diff = GET_TD_MICROSECONDS(self) -
-                GET_TD_MICROSECONDS(other);
-    }
-    return diff;
-}
-
-static PyObject *
-delta_richcompare(PyObject *self, PyObject *other, int op)
-{
-    if (PyDelta_Check(other)) {
-        int diff = delta_cmp(self, other);
-        return diff_to_bool(diff, op);
-    }
-    else {
-        Py_INCREF(Py_NotImplemented);
-        return Py_NotImplemented;
-    }
-}
-
-static PyObject *delta_getstate(PyDateTime_Delta *self);
-
-static long
-delta_hash(PyDateTime_Delta *self)
-{
-    if (self->hashcode == -1) {
-        PyObject *temp = delta_getstate(self);
-        if (temp != NULL) {
-            self->hashcode = PyObject_Hash(temp);
-            Py_DECREF(temp);
-        }
-    }
-    return self->hashcode;
-}
-
-static PyObject *
-delta_multiply(PyObject *left, PyObject *right)
-{
-    PyObject *result = Py_NotImplemented;
-
-    if (PyDelta_Check(left)) {
-        /* delta * ??? */
-        if (PyLong_Check(right))
-            result = multiply_int_timedelta(right,
-                            (PyDateTime_Delta *) left);
-        else if (PyFloat_Check(right))
-            result = multiply_float_timedelta(right,
-                            (PyDateTime_Delta *) left);
-    }
-    else if (PyLong_Check(left))
-        result = multiply_int_timedelta(left,
-                        (PyDateTime_Delta *) right);
-    else if (PyFloat_Check(left))
-        result = multiply_float_timedelta(left,
-                        (PyDateTime_Delta *) right);
-
-    if (result == Py_NotImplemented)
-        Py_INCREF(result);
-    return result;
-}
-
-static PyObject *
-delta_divide(PyObject *left, PyObject *right)
-{
-    PyObject *result = Py_NotImplemented;
-
-    if (PyDelta_Check(left)) {
-        /* delta * ??? */
-        if (PyLong_Check(right))
-            result = divide_timedelta_int(
-                            (PyDateTime_Delta *)left,
-                            right);
-        else if (PyDelta_Check(right))
-            result = divide_timedelta_timedelta(
-                            (PyDateTime_Delta *)left,
-                            (PyDateTime_Delta *)right);
-    }
-
-    if (result == Py_NotImplemented)
-        Py_INCREF(result);
-    return result;
-}
-
-static PyObject *
-delta_truedivide(PyObject *left, PyObject *right)
-{
-    PyObject *result = Py_NotImplemented;
-
-    if (PyDelta_Check(left)) {
-        if (PyDelta_Check(right))
-            result = truedivide_timedelta_timedelta(
-                            (PyDateTime_Delta *)left,
-                            (PyDateTime_Delta *)right);
-        else if (PyFloat_Check(right))
-            result = truedivide_timedelta_float(
-                            (PyDateTime_Delta *)left, right);
-        else if (PyLong_Check(right))
-            result = truedivide_timedelta_int(
-                            (PyDateTime_Delta *)left, right);
-    }
-
-    if (result == Py_NotImplemented)
-        Py_INCREF(result);
-    return result;
-}
-
-static PyObject *
-delta_remainder(PyObject *left, PyObject *right)
-{
-    PyObject *pyus_left;
-    PyObject *pyus_right;
-    PyObject *pyus_remainder;
-    PyObject *remainder;
-
-    if (!PyDelta_Check(left) || !PyDelta_Check(right)) {
-        Py_INCREF(Py_NotImplemented);
-        return Py_NotImplemented;
-    }
-
-    pyus_left = delta_to_microseconds((PyDateTime_Delta *)left);
-    if (pyus_left == NULL)
-        return NULL;
-
-    pyus_right = delta_to_microseconds((PyDateTime_Delta *)right);
-    if (pyus_right == NULL) {
-        Py_DECREF(pyus_left);
-        return NULL;
-    }
-
-    pyus_remainder = PyNumber_Remainder(pyus_left, pyus_right);
-    Py_DECREF(pyus_left);
-    Py_DECREF(pyus_right);
-    if (pyus_remainder == NULL)
-        return NULL;
-
-    remainder = microseconds_to_delta(pyus_remainder);
-    Py_DECREF(pyus_remainder);
-    if (remainder == NULL)
-        return NULL;
-
-    return remainder;
-}
-
-static PyObject *
-delta_divmod(PyObject *left, PyObject *right)
-{
-    PyObject *pyus_left;
-    PyObject *pyus_right;
-    PyObject *divmod;
-    PyObject *delta;
-    PyObject *result;
-
-    if (!PyDelta_Check(left) || !PyDelta_Check(right)) {
-        Py_INCREF(Py_NotImplemented);
-        return Py_NotImplemented;
-    }
-
-    pyus_left = delta_to_microseconds((PyDateTime_Delta *)left);
-    if (pyus_left == NULL)
-        return NULL;
-
-    pyus_right = delta_to_microseconds((PyDateTime_Delta *)right);
-    if (pyus_right == NULL) {
-        Py_DECREF(pyus_left);
-        return NULL;
-    }
-
-    divmod = PyNumber_Divmod(pyus_left, pyus_right);
-    Py_DECREF(pyus_left);
-    Py_DECREF(pyus_right);
-    if (divmod == NULL)
-        return NULL;
-
-    assert(PyTuple_Size(divmod) == 2);
-    delta = microseconds_to_delta(PyTuple_GET_ITEM(divmod, 1));
-    if (delta == NULL) {
-        Py_DECREF(divmod);
-        return NULL;
-    }
-    result = PyTuple_Pack(2, PyTuple_GET_ITEM(divmod, 0), delta);
-    Py_DECREF(delta);
-    Py_DECREF(divmod);
-    return result;
-}
-
-/* Fold in the value of the tag ("seconds", "weeks", etc) component of a
- * timedelta constructor.  sofar is the # of microseconds accounted for
- * so far, and there are factor microseconds per current unit, the number
- * of which is given by num.  num * factor is added to sofar in a
- * numerically careful way, and that's the result.  Any fractional
- * microseconds left over (this can happen if num is a float type) are
- * added into *leftover.
- * Note that there are many ways this can give an error (NULL) return.
- */
-static PyObject *
-accum(const char* tag, PyObject *sofar, PyObject *num, PyObject *factor,
-      double *leftover)
-{
-    PyObject *prod;
-    PyObject *sum;
-
-    assert(num != NULL);
-
-    if (PyLong_Check(num)) {
-        prod = PyNumber_Multiply(num, factor);
-        if (prod == NULL)
-            return NULL;
-        sum = PyNumber_Add(sofar, prod);
-        Py_DECREF(prod);
-        return sum;
-    }
-
-    if (PyFloat_Check(num)) {
-        double dnum;
-        double fracpart;
-        double intpart;
-        PyObject *x;
-        PyObject *y;
-
-        /* The Plan:  decompose num into an integer part and a
-         * fractional part, num = intpart + fracpart.
-         * Then num * factor ==
-         *      intpart * factor + fracpart * factor
-         * and the LHS can be computed exactly in long arithmetic.
-         * The RHS is again broken into an int part and frac part.
-         * and the frac part is added into *leftover.
-         */
-        dnum = PyFloat_AsDouble(num);
-        if (dnum == -1.0 && PyErr_Occurred())
-            return NULL;
-        fracpart = modf(dnum, &intpart);
-        x = PyLong_FromDouble(intpart);
-        if (x == NULL)
-            return NULL;
-
-        prod = PyNumber_Multiply(x, factor);
-        Py_DECREF(x);
-        if (prod == NULL)
-            return NULL;
-
-        sum = PyNumber_Add(sofar, prod);
-        Py_DECREF(prod);
-        if (sum == NULL)
-            return NULL;
-
-        if (fracpart == 0.0)
-            return sum;
-        /* So far we've lost no information.  Dealing with the
-         * fractional part requires float arithmetic, and may
-         * lose a little info.
-         */
-        assert(PyLong_Check(factor));
-        dnum = PyLong_AsDouble(factor);
-
-        dnum *= fracpart;
-        fracpart = modf(dnum, &intpart);
-        x = PyLong_FromDouble(intpart);
-        if (x == NULL) {
-            Py_DECREF(sum);
-            return NULL;
-        }
-
-        y = PyNumber_Add(sum, x);
-        Py_DECREF(sum);
-        Py_DECREF(x);
-        *leftover += fracpart;
-        return y;
-    }
-
-    PyErr_Format(PyExc_TypeError,
-                 "unsupported type for timedelta %s component: %s",
-                 tag, Py_TYPE(num)->tp_name);
-    return NULL;
-}
-
-static PyObject *
-delta_new(PyTypeObject *type, PyObject *args, PyObject *kw)
-{
-    PyObject *self = NULL;
-
-    /* Argument objects. */
-    PyObject *day = NULL;
-    PyObject *second = NULL;
-    PyObject *us = NULL;
-    PyObject *ms = NULL;
-    PyObject *minute = NULL;
-    PyObject *hour = NULL;
-    PyObject *week = NULL;
-
-    PyObject *x = NULL;         /* running sum of microseconds */
-    PyObject *y = NULL;         /* temp sum of microseconds */
-    double leftover_us = 0.0;
-
-    static char *keywords[] = {
-        "days", "seconds", "microseconds", "milliseconds",
-        "minutes", "hours", "weeks", NULL
-    };
-
-    if (PyArg_ParseTupleAndKeywords(args, kw, "|OOOOOOO:__new__",
-                                    keywords,
-                                    &day, &second, &us,
-                                    &ms, &minute, &hour, &week) == 0)
-        goto Done;
-
-    x = PyLong_FromLong(0);
-    if (x == NULL)
-        goto Done;
-
-#define CLEANUP         \
-    Py_DECREF(x);       \
-    x = y;              \
-    if (x == NULL)      \
-        goto Done
-
-    if (us) {
-        y = accum("microseconds", x, us, us_per_us, &leftover_us);
-        CLEANUP;
-    }
-    if (ms) {
-        y = accum("milliseconds", x, ms, us_per_ms, &leftover_us);
-        CLEANUP;
-    }
-    if (second) {
-        y = accum("seconds", x, second, us_per_second, &leftover_us);
-        CLEANUP;
-    }
-    if (minute) {
-        y = accum("minutes", x, minute, us_per_minute, &leftover_us);
-        CLEANUP;
-    }
-    if (hour) {
-        y = accum("hours", x, hour, us_per_hour, &leftover_us);
-        CLEANUP;
-    }
-    if (day) {
-        y = accum("days", x, day, us_per_day, &leftover_us);
-        CLEANUP;
-    }
-    if (week) {
-        y = accum("weeks", x, week, us_per_week, &leftover_us);
-        CLEANUP;
-    }
-    if (leftover_us) {
-        /* Round to nearest whole # of us, and add into x. */
-        PyObject *temp = PyLong_FromLong(round_to_long(leftover_us));
-        if (temp == NULL) {
-            Py_DECREF(x);
-            goto Done;
-        }
-        y = PyNumber_Add(x, temp);
-        Py_DECREF(temp);
-        CLEANUP;
-    }
-
-    self = microseconds_to_delta_ex(x, type);
-    Py_DECREF(x);
-Done:
-    return self;
-
-#undef CLEANUP
-}
-
-static int
-delta_bool(PyDateTime_Delta *self)
-{
-    return (GET_TD_DAYS(self) != 0
-        || GET_TD_SECONDS(self) != 0
-        || GET_TD_MICROSECONDS(self) != 0);
-}
-
-static PyObject *
-delta_repr(PyDateTime_Delta *self)
-{
-    if (GET_TD_MICROSECONDS(self) != 0)
-        return PyUnicode_FromFormat("%s(%d, %d, %d)",
-                                    Py_TYPE(self)->tp_name,
-                                    GET_TD_DAYS(self),
-                                    GET_TD_SECONDS(self),
-                                    GET_TD_MICROSECONDS(self));
-    if (GET_TD_SECONDS(self) != 0)
-        return PyUnicode_FromFormat("%s(%d, %d)",
-                                    Py_TYPE(self)->tp_name,
-                                    GET_TD_DAYS(self),
-                                    GET_TD_SECONDS(self));
-
-    return PyUnicode_FromFormat("%s(%d)",
-                                Py_TYPE(self)->tp_name,
-                                GET_TD_DAYS(self));
-}
-
-static PyObject *
-delta_str(PyDateTime_Delta *self)
-{
-    int us = GET_TD_MICROSECONDS(self);
-    int seconds = GET_TD_SECONDS(self);
-    int minutes = divmod(seconds, 60, &seconds);
-    int hours = divmod(minutes, 60, &minutes);
-    int days = GET_TD_DAYS(self);
-
-    if (days) {
-        if (us)
-            return PyUnicode_FromFormat("%d day%s, %d:%02d:%02d.%06d",
-                                        days, (days == 1 || days == -1) ? "" : "s",
-                                        hours, minutes, seconds, us);
-        else
-            return PyUnicode_FromFormat("%d day%s, %d:%02d:%02d",
-                                        days, (days == 1 || days == -1) ? "" : "s",
-                                        hours, minutes, seconds);
-    } else {
-        if (us)
-            return PyUnicode_FromFormat("%d:%02d:%02d.%06d",
-                                        hours, minutes, seconds, us);
-        else
-            return PyUnicode_FromFormat("%d:%02d:%02d",
-                                        hours, minutes, seconds);
-    }
-
-}
-
-/* Pickle support, a simple use of __reduce__. */
-
-/* __getstate__ isn't exposed */
-static PyObject *
-delta_getstate(PyDateTime_Delta *self)
-{
-    return Py_BuildValue("iii", GET_TD_DAYS(self),
-                                GET_TD_SECONDS(self),
-                                GET_TD_MICROSECONDS(self));
-}
-
-static PyObject *
-delta_total_seconds(PyObject *self)
-{
-    PyObject *total_seconds;
-    PyObject *total_microseconds;
-    PyObject *one_million;
-
-    total_microseconds = delta_to_microseconds((PyDateTime_Delta *)self);
-    if (total_microseconds == NULL)
-        return NULL;
-
-    one_million = PyLong_FromLong(1000000L);
-    if (one_million == NULL) {
-        Py_DECREF(total_microseconds);
-        return NULL;
-    }
-
-    total_seconds = PyNumber_TrueDivide(total_microseconds, one_million);
-
-    Py_DECREF(total_microseconds);
-    Py_DECREF(one_million);
-    return total_seconds;
-}
-
-static PyObject *
-delta_reduce(PyDateTime_Delta* self)
-{
-    return Py_BuildValue("ON", Py_TYPE(self), delta_getstate(self));
-}
-
-#define OFFSET(field)  offsetof(PyDateTime_Delta, field)
-
-static PyMemberDef delta_members[] = {
-
-    {"days",         T_INT, OFFSET(days),         READONLY,
-     PyDoc_STR("Number of days.")},
-
-    {"seconds",      T_INT, OFFSET(seconds),      READONLY,
-     PyDoc_STR("Number of seconds (>= 0 and less than 1 day).")},
-
-    {"microseconds", T_INT, OFFSET(microseconds), READONLY,
-     PyDoc_STR("Number of microseconds (>= 0 and less than 1 second).")},
-    {NULL}
-};
-
-static PyMethodDef delta_methods[] = {
-    {"total_seconds", (PyCFunction)delta_total_seconds, METH_NOARGS,
-     PyDoc_STR("Total seconds in the duration.")},
-
-    {"__reduce__", (PyCFunction)delta_reduce, METH_NOARGS,
-     PyDoc_STR("__reduce__() -> (cls, state)")},
-
-    {NULL,      NULL},
-};
-
-static char delta_doc[] =
-PyDoc_STR("Difference between two datetime values.");
-
-static PyNumberMethods delta_as_number = {
-    delta_add,                                  /* nb_add */
-    delta_subtract,                             /* nb_subtract */
-    delta_multiply,                             /* nb_multiply */
-    delta_remainder,                            /* nb_remainder */
-    delta_divmod,                               /* nb_divmod */
-    0,                                          /* nb_power */
-    (unaryfunc)delta_negative,                  /* nb_negative */
-    (unaryfunc)delta_positive,                  /* nb_positive */
-    (unaryfunc)delta_abs,                       /* nb_absolute */
-    (inquiry)delta_bool,                        /* nb_bool */
-    0,                                          /*nb_invert*/
-    0,                                          /*nb_lshift*/
-    0,                                          /*nb_rshift*/
-    0,                                          /*nb_and*/
-    0,                                          /*nb_xor*/
-    0,                                          /*nb_or*/
-    0,                                          /*nb_int*/
-    0,                                          /*nb_reserved*/
-    0,                                          /*nb_float*/
-    0,                                          /*nb_inplace_add*/
-    0,                                          /*nb_inplace_subtract*/
-    0,                                          /*nb_inplace_multiply*/
-    0,                                          /*nb_inplace_remainder*/
-    0,                                          /*nb_inplace_power*/
-    0,                                          /*nb_inplace_lshift*/
-    0,                                          /*nb_inplace_rshift*/
-    0,                                          /*nb_inplace_and*/
-    0,                                          /*nb_inplace_xor*/
-    0,                                          /*nb_inplace_or*/
-    delta_divide,                               /* nb_floor_divide */
-    delta_truedivide,                           /* nb_true_divide */
-    0,                                          /* nb_inplace_floor_divide */
-    0,                                          /* nb_inplace_true_divide */
-};
-
-static PyTypeObject PyDateTime_DeltaType = {
-    PyVarObject_HEAD_INIT(NULL, 0)
-    "datetime.timedelta",                               /* tp_name */
-    sizeof(PyDateTime_Delta),                           /* tp_basicsize */
-    0,                                                  /* tp_itemsize */
-    0,                                                  /* tp_dealloc */
-    0,                                                  /* tp_print */
-    0,                                                  /* tp_getattr */
-    0,                                                  /* tp_setattr */
-    0,                                                  /* tp_reserved */
-    (reprfunc)delta_repr,                               /* tp_repr */
-    &delta_as_number,                                   /* tp_as_number */
-    0,                                                  /* tp_as_sequence */
-    0,                                                  /* tp_as_mapping */
-    (hashfunc)delta_hash,                               /* tp_hash */
-    0,                                                  /* tp_call */
-    (reprfunc)delta_str,                                /* tp_str */
-    PyObject_GenericGetAttr,                            /* tp_getattro */
-    0,                                                  /* tp_setattro */
-    0,                                                  /* tp_as_buffer */
-    Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE,           /* tp_flags */
-    delta_doc,                                          /* tp_doc */
-    0,                                                  /* tp_traverse */
-    0,                                                  /* tp_clear */
-    delta_richcompare,                                  /* tp_richcompare */
-    0,                                                  /* tp_weaklistoffset */
-    0,                                                  /* tp_iter */
-    0,                                                  /* tp_iternext */
-    delta_methods,                                      /* tp_methods */
-    delta_members,                                      /* tp_members */
-    0,                                                  /* tp_getset */
-    0,                                                  /* tp_base */
-    0,                                                  /* tp_dict */
-    0,                                                  /* tp_descr_get */
-    0,                                                  /* tp_descr_set */
-    0,                                                  /* tp_dictoffset */
-    0,                                                  /* tp_init */
-    0,                                                  /* tp_alloc */
-    delta_new,                                          /* tp_new */
-    0,                                                  /* tp_free */
-};
-
-/*
- * PyDateTime_Date implementation.
- */
-
-/* Accessor properties. */
-
-static PyObject *
-date_year(PyDateTime_Date *self, void *unused)
-{
-    return PyLong_FromLong(GET_YEAR(self));
-}
-
-static PyObject *
-date_month(PyDateTime_Date *self, void *unused)
-{
-    return PyLong_FromLong(GET_MONTH(self));
-}
-
-static PyObject *
-date_day(PyDateTime_Date *self, void *unused)
-{
-    return PyLong_FromLong(GET_DAY(self));
-}
-
-static PyGetSetDef date_getset[] = {
-    {"year",        (getter)date_year},
-    {"month",       (getter)date_month},
-    {"day",         (getter)date_day},
-    {NULL}
-};
-
-/* Constructors. */
-
-static char *date_kws[] = {"year", "month", "day", NULL};
-
-static PyObject *
-date_new(PyTypeObject *type, PyObject *args, PyObject *kw)
-{
-    PyObject *self = NULL;
-    PyObject *state;
-    int year;
-    int month;
-    int day;
-
-    /* Check for invocation from pickle with __getstate__ state */
-    if (PyTuple_GET_SIZE(args) == 1 &&
-        PyBytes_Check(state = PyTuple_GET_ITEM(args, 0)) &&
-        PyBytes_GET_SIZE(state) == _PyDateTime_DATE_DATASIZE &&
-        MONTH_IS_SANE(PyBytes_AS_STRING(state)[2]))
-    {
-        PyDateTime_Date *me;
-
-        me = (PyDateTime_Date *) (type->tp_alloc(type, 0));
-        if (me != NULL) {
-            char *pdata = PyBytes_AS_STRING(state);
-            memcpy(me->data, pdata, _PyDateTime_DATE_DATASIZE);
-            me->hashcode = -1;
-        }
-        return (PyObject *)me;
-    }
-
-    if (PyArg_ParseTupleAndKeywords(args, kw, "iii", date_kws,
-                                    &year, &month, &day)) {
-        if (check_date_args(year, month, day) < 0)
-            return NULL;
-        self = new_date_ex(year, month, day, type);
-    }
-    return self;
-}
-
-/* Return new date from localtime(t). */
-static PyObject *
-date_local_from_time_t(PyObject *cls, double ts)
-{
-    struct tm *tm;
-    time_t t;
-    PyObject *result = NULL;
-
-    t = _PyTime_DoubleToTimet(ts);
-    if (t == (time_t)-1 && PyErr_Occurred())
-        return NULL;
-    tm = localtime(&t);
-    if (tm)
-        result = PyObject_CallFunction(cls, "iii",
-                                       tm->tm_year + 1900,
-                                       tm->tm_mon + 1,
-                                       tm->tm_mday);
-    else
-        PyErr_SetString(PyExc_ValueError,
-                        "timestamp out of range for "
-                        "platform localtime() function");
-    return result;
-}
-
-/* Return new date from current time.
- * We say this is equivalent to fromtimestamp(time.time()), and the
- * only way to be sure of that is to *call* time.time().  That's not
- * generally the same as calling C's time.
- */
-static PyObject *
-date_today(PyObject *cls, PyObject *dummy)
-{
-    PyObject *time;
-    PyObject *result;
-
-    time = time_time();
-    if (time == NULL)
-        return NULL;
-
-    /* Note well:  today() is a class method, so this may not call
-     * date.fromtimestamp.  For example, it may call
-     * datetime.fromtimestamp.  That's why we need all the accuracy
-     * time.time() delivers; if someone were gonzo about optimization,
-     * date.today() could get away with plain C time().
-     */
-    result = PyObject_CallMethod(cls, "fromtimestamp", "O", time);
-    Py_DECREF(time);
-    return result;
-}
-
-/* Return new date from given timestamp (Python timestamp -- a double). */
-static PyObject *
-date_fromtimestamp(PyObject *cls, PyObject *args)
-{
-    double timestamp;
-    PyObject *result = NULL;
-
-    if (PyArg_ParseTuple(args, "d:fromtimestamp", &timestamp))
-        result = date_local_from_time_t(cls, timestamp);
-    return result;
-}
-
-/* Return new date from proleptic Gregorian ordinal.  Raises ValueError if
- * the ordinal is out of range.
- */
-static PyObject *
-date_fromordinal(PyObject *cls, PyObject *args)
-{
-    PyObject *result = NULL;
-    int ordinal;
-
-    if (PyArg_ParseTuple(args, "i:fromordinal", &ordinal)) {
-        int year;
-        int month;
-        int day;
-
-        if (ordinal < 1)
-            PyErr_SetString(PyExc_ValueError, "ordinal must be "
-                                              ">= 1");
-        else {
-            ord_to_ymd(ordinal, &year, &month, &day);
-            result = PyObject_CallFunction(cls, "iii",
-                                           year, month, day);
-        }
-    }
-    return result;
-}
-
-/*
- * Date arithmetic.
- */
-
-/* date + timedelta -> date.  If arg negate is true, subtract the timedelta
- * instead.
- */
-static PyObject *
-add_date_timedelta(PyDateTime_Date *date, PyDateTime_Delta *delta, int negate)
-{
-    PyObject *result = NULL;
-    int year = GET_YEAR(date);
-    int month = GET_MONTH(date);
-    int deltadays = GET_TD_DAYS(delta);
-    /* C-level overflow is impossible because |deltadays| < 1e9. */
-    int day = GET_DAY(date) + (negate ? -deltadays : deltadays);
-
-    if (normalize_date(&year, &month, &day) >= 0)
-        result = new_date(year, month, day);
-    return result;
-}
-
-static PyObject *
-date_add(PyObject *left, PyObject *right)
-{
-    if (PyDateTime_Check(left) || PyDateTime_Check(right)) {
-        Py_INCREF(Py_NotImplemented);
-        return Py_NotImplemented;
-    }
-    if (PyDate_Check(left)) {
-        /* date + ??? */
-        if (PyDelta_Check(right))
-            /* date + delta */
-            return add_date_timedelta((PyDateTime_Date *) left,
-                                      (PyDateTime_Delta *) right,
-                                      0);
-    }
-    else {
-        /* ??? + date
-         * 'right' must be one of us, or we wouldn't have been called
-         */
-        if (PyDelta_Check(left))
-            /* delta + date */
-            return add_date_timedelta((PyDateTime_Date *) right,
-                                      (PyDateTime_Delta *) left,
-                                      0);
-    }
-    Py_INCREF(Py_NotImplemented);
-    return Py_NotImplemented;
-}
-
-static PyObject *
-date_subtract(PyObject *left, PyObject *right)
-{
-    if (PyDateTime_Check(left) || PyDateTime_Check(right)) {
-        Py_INCREF(Py_NotImplemented);
-        return Py_NotImplemented;
-    }
-    if (PyDate_Check(left)) {
-        if (PyDate_Check(right)) {
-            /* date - date */
-            int left_ord = ymd_to_ord(GET_YEAR(left),
-                                      GET_MONTH(left),
-                                      GET_DAY(left));
-            int right_ord = ymd_to_ord(GET_YEAR(right),
-                                       GET_MONTH(right),
-                                       GET_DAY(right));
-            return new_delta(left_ord - right_ord, 0, 0, 0);
-        }
-        if (PyDelta_Check(right)) {
-            /* date - delta */
-            return add_date_timedelta((PyDateTime_Date *) left,
-                                      (PyDateTime_Delta *) right,
-                                      1);
-        }
-    }
-    Py_INCREF(Py_NotImplemented);
-    return Py_NotImplemented;
-}
-
-
-/* Various ways to turn a date into a string. */
-
-static PyObject *
-date_repr(PyDateTime_Date *self)
-{
-    return PyUnicode_FromFormat("%s(%d, %d, %d)",
-                                Py_TYPE(self)->tp_name,
-                                GET_YEAR(self), GET_MONTH(self), GET_DAY(self));
-}
-
-static PyObject *
-date_isoformat(PyDateTime_Date *self)
-{
-    return PyUnicode_FromFormat("%04d-%02d-%02d",
-                                GET_YEAR(self), GET_MONTH(self), GET_DAY(self));
-}
-
-/* str() calls the appropriate isoformat() method. */
-static PyObject *
-date_str(PyDateTime_Date *self)
-{
-    return PyObject_CallMethod((PyObject *)self, "isoformat", "()");
-}
-
-
-static PyObject *
-date_ctime(PyDateTime_Date *self)
-{
-    return format_ctime(self, 0, 0, 0);
-}
-
-static PyObject *
-date_strftime(PyDateTime_Date *self, PyObject *args, PyObject *kw)
-{
-    /* This method can be inherited, and needs to call the
-     * timetuple() method appropriate to self's class.
-     */
-    PyObject *result;
-    PyObject *tuple;
-    PyObject *format;
-    static char *keywords[] = {"format", NULL};
-
-    if (! PyArg_ParseTupleAndKeywords(args, kw, "U:strftime", keywords,
-                                      &format))
-        return NULL;
-
-    tuple = PyObject_CallMethod((PyObject *)self, "timetuple", "()");
-    if (tuple == NULL)
-        return NULL;
-    result = wrap_strftime((PyObject *)self, format, tuple,
-                           (PyObject *)self);
-    Py_DECREF(tuple);
-    return result;
-}
-
-static PyObject *
-date_format(PyDateTime_Date *self, PyObject *args)
-{
-    PyObject *format;
-
-    if (!PyArg_ParseTuple(args, "U:__format__", &format))
-        return NULL;
-
-    /* if the format is zero length, return str(self) */
-    if (PyUnicode_GetSize(format) == 0)
-        return PyObject_Str((PyObject *)self);
-
-    return PyObject_CallMethod((PyObject *)self, "strftime", "O", format);
-}
-
-/* ISO methods. */
-
-static PyObject *
-date_isoweekday(PyDateTime_Date *self)
-{
-    int dow = weekday(GET_YEAR(self), GET_MONTH(self), GET_DAY(self));
-
-    return PyLong_FromLong(dow + 1);
-}
-
-static PyObject *
-date_isocalendar(PyDateTime_Date *self)
-{
-    int  year         = GET_YEAR(self);
-    int  week1_monday = iso_week1_monday(year);
-    int today         = ymd_to_ord(year, GET_MONTH(self), GET_DAY(self));
-    int  week;
-    int  day;
-
-    week = divmod(today - week1_monday, 7, &day);
-    if (week < 0) {
-        --year;
-        week1_monday = iso_week1_monday(year);
-        week = divmod(today - week1_monday, 7, &day);
-    }
-    else if (week >= 52 && today >= iso_week1_monday(year + 1)) {
-        ++year;
-        week = 0;
-    }
-    return Py_BuildValue("iii", year, week + 1, day + 1);
-}
-
-/* Miscellaneous methods. */
-
-static PyObject *
-date_richcompare(PyObject *self, PyObject *other, int op)
-{
-    if (PyDate_Check(other)) {
-        int diff = memcmp(((PyDateTime_Date *)self)->data,
-                          ((PyDateTime_Date *)other)->data,
-                          _PyDateTime_DATE_DATASIZE);
-        return diff_to_bool(diff, op);
-    }
-    else {
-        Py_INCREF(Py_NotImplemented);
-        return Py_NotImplemented;
-    }
-}
-
-static PyObject *
-date_timetuple(PyDateTime_Date *self)
-{
-    return build_struct_time(GET_YEAR(self),
-                             GET_MONTH(self),
-                             GET_DAY(self),
-                             0, 0, 0, -1);
-}
-
-static PyObject *
-date_replace(PyDateTime_Date *self, PyObject *args, PyObject *kw)
-{
-    PyObject *clone;
-    PyObject *tuple;
-    int year = GET_YEAR(self);
-    int month = GET_MONTH(self);
-    int day = GET_DAY(self);
-
-    if (! PyArg_ParseTupleAndKeywords(args, kw, "|iii:replace", date_kws,
-                                      &year, &month, &day))
-        return NULL;
-    tuple = Py_BuildValue("iii", year, month, day);
-    if (tuple == NULL)
-        return NULL;
-    clone = date_new(Py_TYPE(self), tuple, NULL);
-    Py_DECREF(tuple);
-    return clone;
-}
-
-/*
-    Borrowed from stringobject.c, originally it was string_hash()
-*/
-static long
-generic_hash(unsigned char *data, int len)
-{
-    register unsigned char *p;
-    register long x;
-
-    p = (unsigned char *) data;
-    x = *p << 7;
-    while (--len >= 0)
-        x = (1000003*x) ^ *p++;
-    x ^= len;
-    if (x == -1)
-        x = -2;
-
-    return x;
-}
-
-
-static PyObject *date_getstate(PyDateTime_Date *self);
-
-static long
-date_hash(PyDateTime_Date *self)
-{
-    if (self->hashcode == -1)
-        self->hashcode = generic_hash(
-            (unsigned char *)self->data, _PyDateTime_DATE_DATASIZE);
-
-    return self->hashcode;
-}
-
-static PyObject *
-date_toordinal(PyDateTime_Date *self)
-{
-    return PyLong_FromLong(ymd_to_ord(GET_YEAR(self), GET_MONTH(self),
-                                     GET_DAY(self)));
-}
-
-static PyObject *
-date_weekday(PyDateTime_Date *self)
-{
-    int dow = weekday(GET_YEAR(self), GET_MONTH(self), GET_DAY(self));
-
-    return PyLong_FromLong(dow);
-}
-
-/* Pickle support, a simple use of __reduce__. */
-
-/* __getstate__ isn't exposed */
-static PyObject *
-date_getstate(PyDateTime_Date *self)
-{
-    PyObject* field;
-    field = PyBytes_FromStringAndSize((char*)self->data,
-                                       _PyDateTime_DATE_DATASIZE);
-    return Py_BuildValue("(N)", field);
-}
-
-static PyObject *
-date_reduce(PyDateTime_Date *self, PyObject *arg)
-{
-    return Py_BuildValue("(ON)", Py_TYPE(self), date_getstate(self));
-}
-
-static PyMethodDef date_methods[] = {
-
-    /* Class methods: */
-
-    {"fromtimestamp", (PyCFunction)date_fromtimestamp, METH_VARARGS |
-                                                       METH_CLASS,
-     PyDoc_STR("timestamp -> local date from a POSIX timestamp (like "
-               "time.time()).")},
-
-    {"fromordinal", (PyCFunction)date_fromordinal,      METH_VARARGS |
-                                                    METH_CLASS,
-     PyDoc_STR("int -> date corresponding to a proleptic Gregorian "
-               "ordinal.")},
-
-    {"today",         (PyCFunction)date_today,   METH_NOARGS | METH_CLASS,
-     PyDoc_STR("Current date or datetime:  same as "
-               "self.__class__.fromtimestamp(time.time()).")},
-
-    /* Instance methods: */
-
-    {"ctime",       (PyCFunction)date_ctime,        METH_NOARGS,
-     PyDoc_STR("Return ctime() style string.")},
-
-    {"strftime",        (PyCFunction)date_strftime,     METH_VARARGS | METH_KEYWORDS,
-     PyDoc_STR("format -> strftime() style string.")},
-
-    {"__format__",      (PyCFunction)date_format,       METH_VARARGS,
-     PyDoc_STR("Formats self with strftime.")},
-
-    {"timetuple",   (PyCFunction)date_timetuple,    METH_NOARGS,
-     PyDoc_STR("Return time tuple, compatible with time.localtime().")},
-
-    {"isocalendar", (PyCFunction)date_isocalendar,  METH_NOARGS,
-     PyDoc_STR("Return a 3-tuple containing ISO year, week number, and "
-               "weekday.")},
-
-    {"isoformat",   (PyCFunction)date_isoformat,        METH_NOARGS,
-     PyDoc_STR("Return string in ISO 8601 format, YYYY-MM-DD.")},
-
-    {"isoweekday",  (PyCFunction)date_isoweekday,   METH_NOARGS,
-     PyDoc_STR("Return the day of the week represented by the date.\n"
-               "Monday == 1 ... Sunday == 7")},
-
-    {"toordinal",   (PyCFunction)date_toordinal,    METH_NOARGS,
-     PyDoc_STR("Return proleptic Gregorian ordinal.  January 1 of year "
-               "1 is day 1.")},
-
-    {"weekday",     (PyCFunction)date_weekday,      METH_NOARGS,
-     PyDoc_STR("Return the day of the week represented by the date.\n"
-               "Monday == 0 ... Sunday == 6")},
-
-    {"replace",     (PyCFunction)date_replace,      METH_VARARGS | METH_KEYWORDS,
-     PyDoc_STR("Return date with new specified fields.")},
-
-    {"__reduce__", (PyCFunction)date_reduce,        METH_NOARGS,
-     PyDoc_STR("__reduce__() -> (cls, state)")},
-
-    {NULL,      NULL}
-};
-
-static char date_doc[] =
-PyDoc_STR("date(year, month, day) --> date object");
-
-static PyNumberMethods date_as_number = {
-    date_add,                                           /* nb_add */
-    date_subtract,                                      /* nb_subtract */
-    0,                                                  /* nb_multiply */
-    0,                                                  /* nb_remainder */
-    0,                                                  /* nb_divmod */
-    0,                                                  /* nb_power */
-    0,                                                  /* nb_negative */
-    0,                                                  /* nb_positive */
-    0,                                                  /* nb_absolute */
-    0,                                                  /* nb_bool */
-};
-
-static PyTypeObject PyDateTime_DateType = {
-    PyVarObject_HEAD_INIT(NULL, 0)
-    "datetime.date",                                    /* tp_name */
-    sizeof(PyDateTime_Date),                            /* tp_basicsize */
-    0,                                                  /* tp_itemsize */
-    0,                                                  /* tp_dealloc */
-    0,                                                  /* tp_print */
-    0,                                                  /* tp_getattr */
-    0,                                                  /* tp_setattr */
-    0,                                                  /* tp_reserved */
-    (reprfunc)date_repr,                                /* tp_repr */
-    &date_as_number,                                    /* tp_as_number */
-    0,                                                  /* tp_as_sequence */
-    0,                                                  /* tp_as_mapping */
-    (hashfunc)date_hash,                                /* tp_hash */
-    0,                                                  /* tp_call */
-    (reprfunc)date_str,                                 /* tp_str */
-    PyObject_GenericGetAttr,                            /* tp_getattro */
-    0,                                                  /* tp_setattro */
-    0,                                                  /* tp_as_buffer */
-    Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE,           /* tp_flags */
-    date_doc,                                           /* tp_doc */
-    0,                                                  /* tp_traverse */
-    0,                                                  /* tp_clear */
-    date_richcompare,                                   /* tp_richcompare */
-    0,                                                  /* tp_weaklistoffset */
-    0,                                                  /* tp_iter */
-    0,                                                  /* tp_iternext */
-    date_methods,                                       /* tp_methods */
-    0,                                                  /* tp_members */
-    date_getset,                                        /* tp_getset */
-    0,                                                  /* tp_base */
-    0,                                                  /* tp_dict */
-    0,                                                  /* tp_descr_get */
-    0,                                                  /* tp_descr_set */
-    0,                                                  /* tp_dictoffset */
-    0,                                                  /* tp_init */
-    0,                                                  /* tp_alloc */
-    date_new,                                           /* tp_new */
-    0,                                                  /* tp_free */
-};
-
-/*
- * PyDateTime_TZInfo implementation.
- */
-
-/* This is a pure abstract base class, so doesn't do anything beyond
- * raising NotImplemented exceptions.  Real tzinfo classes need
- * to derive from this.  This is mostly for clarity, and for efficiency in
- * datetime and time constructors (their tzinfo arguments need to
- * be subclasses of this tzinfo class, which is easy and quick to check).
- *
- * Note:  For reasons having to do with pickling of subclasses, we have
- * to allow tzinfo objects to be instantiated.  This wasn't an issue
- * in the Python implementation (__init__() could raise NotImplementedError
- * there without ill effect), but doing so in the C implementation hit a
- * brick wall.
- */
-
-static PyObject *
-tzinfo_nogo(const char* methodname)
-{
-    PyErr_Format(PyExc_NotImplementedError,
-                 "a tzinfo subclass must implement %s()",
-                 methodname);
-    return NULL;
-}
-
-/* Methods.  A subclass must implement these. */
-
-static PyObject *
-tzinfo_tzname(PyDateTime_TZInfo *self, PyObject *dt)
-{
-    return tzinfo_nogo("tzname");
-}
-
-static PyObject *
-tzinfo_utcoffset(PyDateTime_TZInfo *self, PyObject *dt)
-{
-    return tzinfo_nogo("utcoffset");
-}
-
-static PyObject *
-tzinfo_dst(PyDateTime_TZInfo *self, PyObject *dt)
-{
-    return tzinfo_nogo("dst");
-}
-
-
-static PyObject *add_datetime_timedelta(PyDateTime_DateTime *date,
-                                        PyDateTime_Delta *delta,
-                                        int factor);
-static PyObject *datetime_utcoffset(PyObject *self, PyObject *);
-static PyObject *datetime_dst(PyObject *self, PyObject *);
-
-static PyObject *
-tzinfo_fromutc(PyDateTime_TZInfo *self, PyObject *dt)
-{
-    PyObject *result = NULL;
-    PyObject *off = NULL, *dst = NULL;
-    PyDateTime_Delta *delta = NULL;
-
-    if (!PyDateTime_Check(dt)) {
-        PyErr_SetString(PyExc_TypeError,
-                        "fromutc: argument must be a datetime");
-        return NULL;
-    }
-    if (GET_DT_TZINFO(dt) != (PyObject *)self) {
-        PyErr_SetString(PyExc_ValueError, "fromutc: dt.tzinfo "
-                        "is not self");
-        return NULL;
-    }
-
-    off = datetime_utcoffset(dt, NULL);
-    if (off == NULL)
-        return NULL;
-    if (off == Py_None) {
-        PyErr_SetString(PyExc_ValueError, "fromutc: non-None "
-                        "utcoffset() result required");
-        goto Fail;
-    }
-
-    dst = datetime_dst(dt, NULL);
-    if (dst == NULL)
-        goto Fail;
-    if (dst == Py_None) {
-        PyErr_SetString(PyExc_ValueError, "fromutc: non-None "
-                        "dst() result required");
-        goto Fail;
-    }
-
-    delta = (PyDateTime_Delta *)delta_subtract(off, dst);
-    if (delta == NULL)
-        goto Fail;
-    result = add_datetime_timedelta((PyDateTime_DateTime *)dt, delta, 1);
-    if (result == NULL)
-        goto Fail;
-
-    Py_DECREF(dst);
-    dst = call_dst(GET_DT_TZINFO(dt), result);
-    if (dst == NULL)
-        goto Fail;
-    if (dst == Py_None)
-        goto Inconsistent;
-    if (delta_bool(delta) != 0) {
-        PyObject *temp = result;
-        result = add_datetime_timedelta((PyDateTime_DateTime *)result,
-                                        (PyDateTime_Delta *)dst, 1);
-        Py_DECREF(temp);
-        if (result == NULL)
-            goto Fail;
-    }
-    Py_DECREF(delta);
-    Py_DECREF(dst);
-    Py_DECREF(off);
-    return result;
-
-Inconsistent:
-    PyErr_SetString(PyExc_ValueError, "fromutc: tz.dst() gave"
-                    "inconsistent results; cannot convert");
-
-    /* fall thru to failure */
-Fail:
-    Py_XDECREF(off);
-    Py_XDECREF(dst);
-    Py_XDECREF(delta);
-    Py_XDECREF(result);
-    return NULL;
-}
-
-/*
- * Pickle support.  This is solely so that tzinfo subclasses can use
- * pickling -- tzinfo itself is supposed to be uninstantiable.
- */
-
-static PyObject *
-tzinfo_reduce(PyObject *self)
-{
-    PyObject *args, *state, *tmp;
-    PyObject *getinitargs, *getstate;
-
-    tmp = PyTuple_New(0);
-    if (tmp == NULL)
-        return NULL;
-
-    getinitargs = PyObject_GetAttrString(self, "__getinitargs__");
-    if (getinitargs != NULL) {
-        args = PyObject_CallObject(getinitargs, tmp);
-        Py_DECREF(getinitargs);
-        if (args == NULL) {
-            Py_DECREF(tmp);
-            return NULL;
-        }
-    }
-    else {
-        PyErr_Clear();
-        args = tmp;
-        Py_INCREF(args);
-    }
-
-    getstate = PyObject_GetAttrString(self, "__getstate__");
-    if (getstate != NULL) {
-        state = PyObject_CallObject(getstate, tmp);
-        Py_DECREF(getstate);
-        if (state == NULL) {
-            Py_DECREF(args);
-            Py_DECREF(tmp);
-            return NULL;
-        }
-    }
-    else {
-        PyObject **dictptr;
-        PyErr_Clear();
-        state = Py_None;
-        dictptr = _PyObject_GetDictPtr(self);
-        if (dictptr && *dictptr && PyDict_Size(*dictptr))
-            state = *dictptr;
-        Py_INCREF(state);
-    }
-
-    Py_DECREF(tmp);
-
-    if (state == Py_None) {
-        Py_DECREF(state);
-        return Py_BuildValue("(ON)", Py_TYPE(self), args);
-    }
-    else
-        return Py_BuildValue("(ONN)", Py_TYPE(self), args, state);
-}
-
-static PyMethodDef tzinfo_methods[] = {
-
-    {"tzname",          (PyCFunction)tzinfo_tzname,             METH_O,
-     PyDoc_STR("datetime -> string name of time zone.")},
-
-    {"utcoffset",       (PyCFunction)tzinfo_utcoffset,          METH_O,
-     PyDoc_STR("datetime -> timedelta showing offset from UTC, negative "
-           "values indicating West of UTC")},
-
-    {"dst",             (PyCFunction)tzinfo_dst,                METH_O,
-     PyDoc_STR("datetime -> DST offset in minutes east of UTC.")},
-
-    {"fromutc",         (PyCFunction)tzinfo_fromutc,            METH_O,
-     PyDoc_STR("datetime in UTC -> datetime in local time.")},
-
-    {"__reduce__",  (PyCFunction)tzinfo_reduce,             METH_NOARGS,
-     PyDoc_STR("-> (cls, state)")},
-
-    {NULL, NULL}
-};
-
-static char tzinfo_doc[] =
-PyDoc_STR("Abstract base class for time zone info objects.");
-
-static PyTypeObject PyDateTime_TZInfoType = {
-    PyVarObject_HEAD_INIT(NULL, 0)
-    "datetime.tzinfo",                          /* tp_name */
-    sizeof(PyDateTime_TZInfo),                  /* tp_basicsize */
-    0,                                          /* tp_itemsize */
-    0,                                          /* tp_dealloc */
-    0,                                          /* tp_print */
-    0,                                          /* tp_getattr */
-    0,                                          /* tp_setattr */
-    0,                                          /* tp_reserved */
-    0,                                          /* tp_repr */
-    0,                                          /* tp_as_number */
-    0,                                          /* tp_as_sequence */
-    0,                                          /* tp_as_mapping */
-    0,                                          /* tp_hash */
-    0,                                          /* tp_call */
-    0,                                          /* tp_str */
-    PyObject_GenericGetAttr,                    /* tp_getattro */
-    0,                                          /* tp_setattro */
-    0,                                          /* tp_as_buffer */
-    Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE,   /* tp_flags */
-    tzinfo_doc,                                 /* tp_doc */
-    0,                                          /* tp_traverse */
-    0,                                          /* tp_clear */
-    0,                                          /* tp_richcompare */
-    0,                                          /* tp_weaklistoffset */
-    0,                                          /* tp_iter */
-    0,                                          /* tp_iternext */
-    tzinfo_methods,                             /* tp_methods */
-    0,                                          /* tp_members */
-    0,                                          /* tp_getset */
-    0,                                          /* tp_base */
-    0,                                          /* tp_dict */
-    0,                                          /* tp_descr_get */
-    0,                                          /* tp_descr_set */
-    0,                                          /* tp_dictoffset */
-    0,                                          /* tp_init */
-    0,                                          /* tp_alloc */
-    PyType_GenericNew,                          /* tp_new */
-    0,                                          /* tp_free */
-};
-
-static char *timezone_kws[] = {"offset", "name", NULL};
-
-static PyObject *
-timezone_new(PyTypeObject *type, PyObject *args, PyObject *kw)
-{
-    PyObject *offset;
-    PyObject *name = NULL;
-    if (PyArg_ParseTupleAndKeywords(args, kw, "O!|O!:timezone", timezone_kws,
-                                    &PyDateTime_DeltaType, &offset,
-                                    &PyUnicode_Type, &name))
-        return new_timezone(offset, name);
-
-    return NULL;
-}
-
-static void
-timezone_dealloc(PyDateTime_TimeZone *self)
-{
-    Py_CLEAR(self->offset);
-    Py_CLEAR(self->name);
-    Py_TYPE(self)->tp_free((PyObject *)self);
-}
-
-static PyObject *
-timezone_richcompare(PyDateTime_TimeZone *self,
-                     PyDateTime_TimeZone *other, int op)
-{
-    if (op != Py_EQ && op != Py_NE) {
-        Py_INCREF(Py_NotImplemented);
-        return Py_NotImplemented;
-    }
-    return delta_richcompare(self->offset, other->offset, op);
-}
-
-static long
-timezone_hash(PyDateTime_TimeZone *self)
-{
-    return delta_hash((PyDateTime_Delta *)self->offset);
-}
-
-/* Check argument type passed to tzname, utcoffset, or dst methods.
-   Returns 0 for good argument.  Returns -1 and sets exception info
-   otherwise.
- */
-static int
-_timezone_check_argument(PyObject *dt, const char *meth)
-{
-    if (dt == Py_None || PyDateTime_Check(dt))
-        return 0;
-    PyErr_Format(PyExc_TypeError, "%s(dt) argument must be a datetime instance"
-                 " or None, not %.200s", meth, Py_TYPE(dt)->tp_name);
-    return -1;
-}
-
-static PyObject *
-timezone_repr(PyDateTime_TimeZone *self)
-{
-    /* Note that although timezone is not subclassable, it is convenient
-       to use Py_TYPE(self)->tp_name here. */
-    const char *type_name = Py_TYPE(self)->tp_name;
-
-    if (((PyObject *)self) == PyDateTime_TimeZone_UTC)
-        return PyUnicode_FromFormat("%s.utc", type_name);
-
-    if (self->name == NULL)
-        return PyUnicode_FromFormat("%s(%R)", type_name, self->offset);
-
-    return PyUnicode_FromFormat("%s(%R, %R)", type_name, self->offset,
-                                self->name);
-}
-
-
-static PyObject *
-timezone_str(PyDateTime_TimeZone *self)
-{
-    char buf[10];
-    int hours, minutes, seconds;
-    PyObject *offset;
-    char sign;
-
-    if (self->name != NULL) {
-        Py_INCREF(self->name);
-        return self->name;
-    }
-    /* Offset is normalized, so it is negative if days < 0 */
-    if (GET_TD_DAYS(self->offset) < 0) {
-        sign = '-';
-        offset = delta_negative((PyDateTime_Delta *)self->offset);
-        if (offset == NULL)
-            return NULL;
-    }
-    else {
-        sign = '+';
-        offset = self->offset;
-        Py_INCREF(offset);
-    }
-    /* Offset is not negative here. */
-    seconds = GET_TD_SECONDS(offset);
-    Py_DECREF(offset);
-    minutes = divmod(seconds, 60, &seconds);
-    hours = divmod(minutes, 60, &minutes);
-    assert(seconds == 0);
-    /* XXX ignore sub-minute data, curently not allowed. */
-    PyOS_snprintf(buf, sizeof(buf), "UTC%c%02d:%02d", sign, hours, minutes);
-
-    return PyUnicode_FromString(buf);
-}
-
-static PyObject *
-timezone_tzname(PyDateTime_TimeZone *self, PyObject *dt)
-{
-    if (_timezone_check_argument(dt, "tzname") == -1)
-        return NULL;
-
-    return timezone_str(self);
-}
-
-static PyObject *
-timezone_utcoffset(PyDateTime_TimeZone *self, PyObject *dt)
-{
-    if (_timezone_check_argument(dt, "utcoffset") == -1)
-        return NULL;
-
-    Py_INCREF(self->offset);
-    return self->offset;
-}
-
-static PyObject *
-timezone_dst(PyObject *self, PyObject *dt)
-{
-    if (_timezone_check_argument(dt, "dst") == -1)
-        return NULL;
-
-    Py_RETURN_NONE;
-}
-
-static PyObject *
-timezone_fromutc(PyDateTime_TimeZone *self, PyDateTime_DateTime *dt)
-{
-    if (!PyDateTime_Check(dt)) {
-        PyErr_SetString(PyExc_TypeError,
-                        "fromutc: argument must be a datetime");
-        return NULL;
-    }
-    if (!HASTZINFO(dt) || dt->tzinfo != (PyObject *)self) {
-        PyErr_SetString(PyExc_ValueError, "fromutc: dt.tzinfo "
-                        "is not self");
-        return NULL;
-    }
-
-    return add_datetime_timedelta(dt, (PyDateTime_Delta *)self->offset, 1);
-}
-
-static PyObject *
-timezone_getinitargs(PyDateTime_TimeZone *self)
-{
-    if (self->name == NULL)
-        return Py_BuildValue("(O)", self->offset);
-    return Py_BuildValue("(OO)", self->offset, self->name);
-}
-
-static PyMethodDef timezone_methods[] = {
-    {"tzname", (PyCFunction)timezone_tzname, METH_O,
-     PyDoc_STR("If name is specified when timezone is created, returns the name."
-               "  Otherwise returns offset as 'UTC(+|-)HH:MM'.")},
-
-    {"utcoffset", (PyCFunction)timezone_utcoffset, METH_O,
-     PyDoc_STR("Return fixed offset.")},
-
-    {"dst", (PyCFunction)timezone_dst, METH_O,
-     PyDoc_STR("Return None.")},
-
-    {"fromutc", (PyCFunction)timezone_fromutc, METH_O,
-     PyDoc_STR("datetime in UTC -> datetime in local time.")},
-
-    {"__getinitargs__", (PyCFunction)timezone_getinitargs, METH_NOARGS,
-     PyDoc_STR("pickle support")},
-
-    {NULL, NULL}
-};
-
-static char timezone_doc[] =
-PyDoc_STR("Fixed offset from UTC implementation of tzinfo.");
-
-static PyTypeObject PyDateTime_TimeZoneType = {
-    PyVarObject_HEAD_INIT(NULL, 0)
-    "datetime.timezone",              /* tp_name */
-    sizeof(PyDateTime_TimeZone),      /* tp_basicsize */
-    0,                                /* tp_itemsize */
-    (destructor)timezone_dealloc,     /* tp_dealloc */
-    0,                                /* tp_print */
-    0,                                /* tp_getattr */
-    0,                                /* tp_setattr */
-    0,                                /* tp_reserved */
-    (reprfunc)timezone_repr,          /* tp_repr */
-    0,                                /* tp_as_number */
-    0,                                /* tp_as_sequence */
-    0,                                /* tp_as_mapping */
-    (hashfunc)timezone_hash,          /* tp_hash */
-    0,                                /* tp_call */
-    (reprfunc)timezone_str,           /* tp_str */
-    0,                                /* tp_getattro */
-    0,                                /* tp_setattro */
-    0,                                /* tp_as_buffer */
-    Py_TPFLAGS_DEFAULT,               /* tp_flags */
-    timezone_doc,                     /* tp_doc */
-    0,                                /* tp_traverse */
-    0,                                /* tp_clear */
-    (richcmpfunc)timezone_richcompare,/* tp_richcompare */
-    0,                                /* tp_weaklistoffset */
-    0,                                /* tp_iter */
-    0,                                /* tp_iternext */
-    timezone_methods,                 /* tp_methods */
-    0,                                /* tp_members */
-    0,                                /* tp_getset */
-    &PyDateTime_TZInfoType,           /* tp_base */
-    0,                                /* tp_dict */
-    0,                                /* tp_descr_get */
-    0,                                /* tp_descr_set */
-    0,                                /* tp_dictoffset */
-    0,                                /* tp_init */
-    0,                                /* tp_alloc */
-    timezone_new,                     /* tp_new */
-};
-
-/*
- * PyDateTime_Time implementation.
- */
-
-/* Accessor properties.
- */
-
-static PyObject *
-time_hour(PyDateTime_Time *self, void *unused)
-{
-    return PyLong_FromLong(TIME_GET_HOUR(self));
-}
-
-static PyObject *
-time_minute(PyDateTime_Time *self, void *unused)
-{
-    return PyLong_FromLong(TIME_GET_MINUTE(self));
-}
-
-/* The name time_second conflicted with some platform header file. */
-static PyObject *
-py_time_second(PyDateTime_Time *self, void *unused)
-{
-    return PyLong_FromLong(TIME_GET_SECOND(self));
-}
-
-static PyObject *
-time_microsecond(PyDateTime_Time *self, void *unused)
-{
-    return PyLong_FromLong(TIME_GET_MICROSECOND(self));
-}
-
-static PyObject *
-time_tzinfo(PyDateTime_Time *self, void *unused)
-{
-    PyObject *result = HASTZINFO(self) ? self->tzinfo : Py_None;
-    Py_INCREF(result);
-    return result;
-}
-
-static PyGetSetDef time_getset[] = {
-    {"hour",        (getter)time_hour},
-    {"minute",      (getter)time_minute},
-    {"second",      (getter)py_time_second},
-    {"microsecond", (getter)time_microsecond},
-    {"tzinfo",          (getter)time_tzinfo},
-    {NULL}
-};
-
-/*
- * Constructors.
- */
-
-static char *time_kws[] = {"hour", "minute", "second", "microsecond",
-                           "tzinfo", NULL};
-
-static PyObject *
-time_new(PyTypeObject *type, PyObject *args, PyObject *kw)
-{
-    PyObject *self = NULL;
-    PyObject *state;
-    int hour = 0;
-    int minute = 0;
-    int second = 0;
-    int usecond = 0;
-    PyObject *tzinfo = Py_None;
-
-    /* Check for invocation from pickle with __getstate__ state */
-    if (PyTuple_GET_SIZE(args) >= 1 &&
-        PyTuple_GET_SIZE(args) <= 2 &&
-        PyBytes_Check(state = PyTuple_GET_ITEM(args, 0)) &&
-        PyBytes_GET_SIZE(state) == _PyDateTime_TIME_DATASIZE &&
-        ((unsigned char) (PyBytes_AS_STRING(state)[0])) < 24)
-    {
-        PyDateTime_Time *me;
-        char aware;
-
-        if (PyTuple_GET_SIZE(args) == 2) {
-            tzinfo = PyTuple_GET_ITEM(args, 1);
-            if (check_tzinfo_subclass(tzinfo) < 0) {
-                PyErr_SetString(PyExc_TypeError, "bad "
-                    "tzinfo state arg");
-                return NULL;
-            }
-        }
-        aware = (char)(tzinfo != Py_None);
-        me = (PyDateTime_Time *) (type->tp_alloc(type, aware));
-        if (me != NULL) {
-            char *pdata = PyBytes_AS_STRING(state);
-
-            memcpy(me->data, pdata, _PyDateTime_TIME_DATASIZE);
-            me->hashcode = -1;
-            me->hastzinfo = aware;
-            if (aware) {
-                Py_INCREF(tzinfo);
-                me->tzinfo = tzinfo;
-            }
-        }
-        return (PyObject *)me;
-    }
-
-    if (PyArg_ParseTupleAndKeywords(args, kw, "|iiiiO", time_kws,
-                                    &hour, &minute, &second, &usecond,
-                                    &tzinfo)) {
-        if (check_time_args(hour, minute, second, usecond) < 0)
-            return NULL;
-        if (check_tzinfo_subclass(tzinfo) < 0)
-            return NULL;
-        self = new_time_ex(hour, minute, second, usecond, tzinfo,
-                           type);
-    }
-    return self;
-}
-
-/*
- * Destructor.
- */
-
-static void
-time_dealloc(PyDateTime_Time *self)
-{
-    if (HASTZINFO(self)) {
-        Py_XDECREF(self->tzinfo);
-    }
-    Py_TYPE(self)->tp_free((PyObject *)self);
-}
-
-/*
- * Indirect access to tzinfo methods.
- */
-
-/* These are all METH_NOARGS, so don't need to check the arglist. */
-static PyObject *
-time_utcoffset(PyObject *self, PyObject *unused) {
-    return call_utcoffset(GET_TIME_TZINFO(self), Py_None);
-}
-
-static PyObject *
-time_dst(PyObject *self, PyObject *unused) {
-    return call_dst(GET_TIME_TZINFO(self), Py_None);
-}
-
-static PyObject *
-time_tzname(PyDateTime_Time *self, PyObject *unused) {
-    return call_tzname(GET_TIME_TZINFO(self), Py_None);
-}
-
-/*
- * Various ways to turn a time into a string.
- */
-
-static PyObject *
-time_repr(PyDateTime_Time *self)
-{
-    const char *type_name = Py_TYPE(self)->tp_name;
-    int h = TIME_GET_HOUR(self);
-    int m = TIME_GET_MINUTE(self);
-    int s = TIME_GET_SECOND(self);
-    int us = TIME_GET_MICROSECOND(self);
-    PyObject *result = NULL;
-
-    if (us)
-        result = PyUnicode_FromFormat("%s(%d, %d, %d, %d)",
-                                      type_name, h, m, s, us);
-    else if (s)
-        result = PyUnicode_FromFormat("%s(%d, %d, %d)",
-                                      type_name, h, m, s);
-    else
-        result = PyUnicode_FromFormat("%s(%d, %d)", type_name, h, m);
-    if (result != NULL && HASTZINFO(self))
-        result = append_keyword_tzinfo(result, self->tzinfo);
-    return result;
-}
-
-static PyObject *
-time_str(PyDateTime_Time *self)
-{
-    return PyObject_CallMethod((PyObject *)self, "isoformat", "()");
-}
-
-static PyObject *
-time_isoformat(PyDateTime_Time *self, PyObject *unused)
-{
-    char buf[100];
-    PyObject *result;
-    int us = TIME_GET_MICROSECOND(self);;
-
-    if (us)
-        result = PyUnicode_FromFormat("%02d:%02d:%02d.%06d",
-                                      TIME_GET_HOUR(self),
-                                      TIME_GET_MINUTE(self),
-                                      TIME_GET_SECOND(self),
-                                      us);
-    else
-        result = PyUnicode_FromFormat("%02d:%02d:%02d",
-                                      TIME_GET_HOUR(self),
-                                      TIME_GET_MINUTE(self),
-                                      TIME_GET_SECOND(self));
-
-    if (result == NULL || !HASTZINFO(self) || self->tzinfo == Py_None)
-        return result;
-
-    /* We need to append the UTC offset. */
-    if (format_utcoffset(buf, sizeof(buf), ":", self->tzinfo,
-                         Py_None) < 0) {
-        Py_DECREF(result);
-        return NULL;
-    }
-    PyUnicode_AppendAndDel(&result, PyUnicode_FromString(buf));
-    return result;
-}
-
-static PyObject *
-time_strftime(PyDateTime_Time *self, PyObject *args, PyObject *kw)
-{
-    PyObject *result;
-    PyObject *tuple;
-    PyObject *format;
-    static char *keywords[] = {"format", NULL};
-
-    if (! PyArg_ParseTupleAndKeywords(args, kw, "U:strftime", keywords,
-                                      &format))
-        return NULL;
-
-    /* Python's strftime does insane things with the year part of the
-     * timetuple.  The year is forced to (the otherwise nonsensical)
-     * 1900 to worm around that.
-     */
-    tuple = Py_BuildValue("iiiiiiiii",
-                          1900, 1, 1, /* year, month, day */
-                  TIME_GET_HOUR(self),
-                  TIME_GET_MINUTE(self),
-                  TIME_GET_SECOND(self),
-                  0, 1, -1); /* weekday, daynum, dst */
-    if (tuple == NULL)
-        return NULL;
-    assert(PyTuple_Size(tuple) == 9);
-    result = wrap_strftime((PyObject *)self, format, tuple,
-                           Py_None);
-    Py_DECREF(tuple);
-    return result;
-}
-
-/*
- * Miscellaneous methods.
- */
-
-static PyObject *
-time_richcompare(PyObject *self, PyObject *other, int op)
-{
-    PyObject *result = NULL;
-    PyObject *offset1, *offset2;
-    int diff;
-
-    if (! PyTime_Check(other)) {
-        Py_INCREF(Py_NotImplemented);
-        return Py_NotImplemented;
-    }
-
-    if (GET_TIME_TZINFO(self) == GET_TIME_TZINFO(other)) {
-        diff = memcmp(((PyDateTime_Time *)self)->data,
-                      ((PyDateTime_Time *)other)->data,
-                      _PyDateTime_TIME_DATASIZE);
-        return diff_to_bool(diff, op);
-    }
-    offset1 = time_utcoffset(self, NULL);
-    if (offset1 == NULL)
-        return NULL;
-    offset2 = time_utcoffset(other, NULL);
-    if (offset2 == NULL)
-        goto done;
-    /* If they're both naive, or both aware and have the same offsets,
-     * we get off cheap.  Note that if they're both naive, offset1 ==
-     * offset2 == Py_None at this point.
-     */
-    if ((offset1 == offset2) ||
-        (PyDelta_Check(offset1) && PyDelta_Check(offset2) &&
-         delta_cmp(offset1, offset2) == 0)) {
-        diff = memcmp(((PyDateTime_Time *)self)->data,
-                      ((PyDateTime_Time *)other)->data,
-                      _PyDateTime_TIME_DATASIZE);
-        result = diff_to_bool(diff, op);
-    }
-    /* The hard case: both aware with different UTC offsets */
-    else if (offset1 != Py_None && offset2 != Py_None) {
-        int offsecs1, offsecs2;
-        assert(offset1 != offset2); /* else last "if" handled it */
-        offsecs1 = TIME_GET_HOUR(self) * 3600 +
-                   TIME_GET_MINUTE(self) * 60 +
-                   TIME_GET_SECOND(self) -
-                   GET_TD_DAYS(offset1) * 86400 -
-                   GET_TD_SECONDS(offset1);
-        offsecs2 = TIME_GET_HOUR(other) * 3600 +
-                   TIME_GET_MINUTE(other) * 60 +
-                   TIME_GET_SECOND(other) -
-                   GET_TD_DAYS(offset2) * 86400 -
-                   GET_TD_SECONDS(offset2);
-        diff = offsecs1 - offsecs2;
-        if (diff == 0)
-            diff = TIME_GET_MICROSECOND(self) -
-                   TIME_GET_MICROSECOND(other);
-        result = diff_to_bool(diff, op);
-    }
-    else {
-        PyErr_SetString(PyExc_TypeError,
-                        "can't compare offset-naive and "
-                        "offset-aware times");
-    }
- done:
-    Py_DECREF(offset1);
-    Py_XDECREF(offset2);
-    return result;
-}
-
-static long
-time_hash(PyDateTime_Time *self)
-{
-    if (self->hashcode == -1) {
-        PyObject *offset;
-
-        offset = time_utcoffset((PyObject *)self, NULL);
-
-        if (offset == NULL)
-            return -1;
-
-        /* Reduce this to a hash of another object. */
-        if (offset == Py_None)
-            self->hashcode = generic_hash(
-                (unsigned char *)self->data, _PyDateTime_TIME_DATASIZE);
-        else {
-            PyObject *temp1, *temp2;
-            int seconds, microseconds;
-            assert(HASTZINFO(self));
-            seconds = TIME_GET_HOUR(self) * 3600 +
-                      TIME_GET_MINUTE(self) * 60 +
-                      TIME_GET_SECOND(self);
-            microseconds = TIME_GET_MICROSECOND(self);
-            temp1 = new_delta(0, seconds, microseconds, 1);
-            if (temp1 == NULL) {
-                Py_DECREF(offset);
-                return -1;
-            }
-            temp2 = delta_subtract(temp1, offset);
-            Py_DECREF(temp1);
-            if (temp2 == NULL) {
-                Py_DECREF(offset);
-                return -1;
-            }
-            self->hashcode = PyObject_Hash(temp2);
-            Py_DECREF(temp2);
-        }
-        Py_DECREF(offset);
-    }
-    return self->hashcode;
-}
-
-static PyObject *
-time_replace(PyDateTime_Time *self, PyObject *args, PyObject *kw)
-{
-    PyObject *clone;
-    PyObject *tuple;
-    int hh = TIME_GET_HOUR(self);
-    int mm = TIME_GET_MINUTE(self);
-    int ss = TIME_GET_SECOND(self);
-    int us = TIME_GET_MICROSECOND(self);
-    PyObject *tzinfo = HASTZINFO(self) ? self->tzinfo : Py_None;
-
-    if (! PyArg_ParseTupleAndKeywords(args, kw, "|iiiiO:replace",
-                                      time_kws,
-                                      &hh, &mm, &ss, &us, &tzinfo))
-        return NULL;
-    tuple = Py_BuildValue("iiiiO", hh, mm, ss, us, tzinfo);
-    if (tuple == NULL)
-        return NULL;
-    clone = time_new(Py_TYPE(self), tuple, NULL);
-    Py_DECREF(tuple);
-    return clone;
-}
-
-static int
-time_bool(PyObject *self)
-{
-    PyObject *offset, *tzinfo;
-    int offsecs = 0;
-
-    if (TIME_GET_SECOND(self) || TIME_GET_MICROSECOND(self)) {
-        /* Since utcoffset is in whole minutes, nothing can
-         * alter the conclusion that this is nonzero.
-         */
-        return 1;
-    }
-    tzinfo = GET_TIME_TZINFO(self);
-    if (tzinfo != Py_None) {
-        offset = call_utcoffset(tzinfo, Py_None);
-        if (offset == NULL)
-            return -1;
-        offsecs = GET_TD_DAYS(offset)*86400 + GET_TD_SECONDS(offset);
-        Py_DECREF(offset);
-    }
-    return (TIME_GET_MINUTE(self)*60 - offsecs + TIME_GET_HOUR(self)*3600) != 0;
-}
-
-/* Pickle support, a simple use of __reduce__. */
-
-/* Let basestate be the non-tzinfo data string.
- * If tzinfo is None, this returns (basestate,), else (basestate, tzinfo).
- * So it's a tuple in any (non-error) case.
- * __getstate__ isn't exposed.
- */
-static PyObject *
-time_getstate(PyDateTime_Time *self)
-{
-    PyObject *basestate;
-    PyObject *result = NULL;
-
-    basestate =  PyBytes_FromStringAndSize((char *)self->data,
-                                            _PyDateTime_TIME_DATASIZE);
-    if (basestate != NULL) {
-        if (! HASTZINFO(self) || self->tzinfo == Py_None)
-            result = PyTuple_Pack(1, basestate);
-        else
-            result = PyTuple_Pack(2, basestate, self->tzinfo);
-        Py_DECREF(basestate);
-    }
-    return result;
-}
-
-static PyObject *
-time_reduce(PyDateTime_Time *self, PyObject *arg)
-{
-    return Py_BuildValue("(ON)", Py_TYPE(self), time_getstate(self));
-}
-
-static PyMethodDef time_methods[] = {
-
-    {"isoformat",   (PyCFunction)time_isoformat,        METH_NOARGS,
-     PyDoc_STR("Return string in ISO 8601 format, HH:MM:SS[.mmmmmm]"
-               "[+HH:MM].")},
-
-    {"strftime",        (PyCFunction)time_strftime,     METH_VARARGS | METH_KEYWORDS,
-     PyDoc_STR("format -> strftime() style string.")},
-
-    {"__format__",      (PyCFunction)date_format,       METH_VARARGS,
-     PyDoc_STR("Formats self with strftime.")},
-
-    {"utcoffset",       (PyCFunction)time_utcoffset,    METH_NOARGS,
-     PyDoc_STR("Return self.tzinfo.utcoffset(self).")},
-
-    {"tzname",          (PyCFunction)time_tzname,       METH_NOARGS,
-     PyDoc_STR("Return self.tzinfo.tzname(self).")},
-
-    {"dst",             (PyCFunction)time_dst,          METH_NOARGS,
-     PyDoc_STR("Return self.tzinfo.dst(self).")},
-
-    {"replace",     (PyCFunction)time_replace,          METH_VARARGS | METH_KEYWORDS,
-     PyDoc_STR("Return time with new specified fields.")},
-
-    {"__reduce__", (PyCFunction)time_reduce,        METH_NOARGS,
-     PyDoc_STR("__reduce__() -> (cls, state)")},
-
-    {NULL,      NULL}
-};
-
-static char time_doc[] =
-PyDoc_STR("time([hour[, minute[, second[, microsecond[, tzinfo]]]]]) --> a time object\n\
-\n\
-All arguments are optional. tzinfo may be None, or an instance of\n\
-a tzinfo subclass. The remaining arguments may be ints or longs.\n");
-
-static PyNumberMethods time_as_number = {
-    0,                                          /* nb_add */
-    0,                                          /* nb_subtract */
-    0,                                          /* nb_multiply */
-    0,                                          /* nb_remainder */
-    0,                                          /* nb_divmod */
-    0,                                          /* nb_power */
-    0,                                          /* nb_negative */
-    0,                                          /* nb_positive */
-    0,                                          /* nb_absolute */
-    (inquiry)time_bool,                         /* nb_bool */
-};
-
-static PyTypeObject PyDateTime_TimeType = {
-    PyVarObject_HEAD_INIT(NULL, 0)
-    "datetime.time",                            /* tp_name */
-    sizeof(PyDateTime_Time),                    /* tp_basicsize */
-    0,                                          /* tp_itemsize */
-    (destructor)time_dealloc,                   /* tp_dealloc */
-    0,                                          /* tp_print */
-    0,                                          /* tp_getattr */
-    0,                                          /* tp_setattr */
-    0,                                          /* tp_reserved */
-    (reprfunc)time_repr,                        /* tp_repr */
-    &time_as_number,                            /* tp_as_number */
-    0,                                          /* tp_as_sequence */
-    0,                                          /* tp_as_mapping */
-    (hashfunc)time_hash,                        /* tp_hash */
-    0,                                          /* tp_call */
-    (reprfunc)time_str,                         /* tp_str */
-    PyObject_GenericGetAttr,                    /* tp_getattro */
-    0,                                          /* tp_setattro */
-    0,                                          /* tp_as_buffer */
-    Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE, /* tp_flags */
-    time_doc,                                   /* tp_doc */
-    0,                                          /* tp_traverse */
-    0,                                          /* tp_clear */
-    time_richcompare,                           /* tp_richcompare */
-    0,                                          /* tp_weaklistoffset */
-    0,                                          /* tp_iter */
-    0,                                          /* tp_iternext */
-    time_methods,                               /* tp_methods */
-    0,                                          /* tp_members */
-    time_getset,                                /* tp_getset */
-    0,                                          /* tp_base */
-    0,                                          /* tp_dict */
-    0,                                          /* tp_descr_get */
-    0,                                          /* tp_descr_set */
-    0,                                          /* tp_dictoffset */
-    0,                                          /* tp_init */
-    time_alloc,                                 /* tp_alloc */
-    time_new,                                   /* tp_new */
-    0,                                          /* tp_free */
-};
-
-/*
- * PyDateTime_DateTime implementation.
- */
-
-/* Accessor properties.  Properties for day, month, and year are inherited
- * from date.
- */
-
-static PyObject *
-datetime_hour(PyDateTime_DateTime *self, void *unused)
-{
-    return PyLong_FromLong(DATE_GET_HOUR(self));
-}
-
-static PyObject *
-datetime_minute(PyDateTime_DateTime *self, void *unused)
-{
-    return PyLong_FromLong(DATE_GET_MINUTE(self));
-}
-
-static PyObject *
-datetime_second(PyDateTime_DateTime *self, void *unused)
-{
-    return PyLong_FromLong(DATE_GET_SECOND(self));
-}
-
-static PyObject *
-datetime_microsecond(PyDateTime_DateTime *self, void *unused)
-{
-    return PyLong_FromLong(DATE_GET_MICROSECOND(self));
-}
-
-static PyObject *
-datetime_tzinfo(PyDateTime_DateTime *self, void *unused)
-{
-    PyObject *result = HASTZINFO(self) ? self->tzinfo : Py_None;
-    Py_INCREF(result);
-    return result;
-}
-
-static PyGetSetDef datetime_getset[] = {
-    {"hour",        (getter)datetime_hour},
-    {"minute",      (getter)datetime_minute},
-    {"second",      (getter)datetime_second},
-    {"microsecond", (getter)datetime_microsecond},
-    {"tzinfo",          (getter)datetime_tzinfo},
-    {NULL}
-};
-
-/*
- * Constructors.
- */
-
-static char *datetime_kws[] = {
-    "year", "month", "day", "hour", "minute", "second",
-    "microsecond", "tzinfo", NULL
-};
-
-static PyObject *
-datetime_new(PyTypeObject *type, PyObject *args, PyObject *kw)
-{
-    PyObject *self = NULL;
-    PyObject *state;
-    int year;
-    int month;
-    int day;
-    int hour = 0;
-    int minute = 0;
-    int second = 0;
-    int usecond = 0;
-    PyObject *tzinfo = Py_None;
-
-    /* Check for invocation from pickle with __getstate__ state */
-    if (PyTuple_GET_SIZE(args) >= 1 &&
-        PyTuple_GET_SIZE(args) <= 2 &&
-        PyBytes_Check(state = PyTuple_GET_ITEM(args, 0)) &&
-        PyBytes_GET_SIZE(state) == _PyDateTime_DATETIME_DATASIZE &&
-        MONTH_IS_SANE(PyBytes_AS_STRING(state)[2]))
-    {
-        PyDateTime_DateTime *me;
-        char aware;
-
-        if (PyTuple_GET_SIZE(args) == 2) {
-            tzinfo = PyTuple_GET_ITEM(args, 1);
-            if (check_tzinfo_subclass(tzinfo) < 0) {
-                PyErr_SetString(PyExc_TypeError, "bad "
-                    "tzinfo state arg");
-                return NULL;
-            }
-        }
-        aware = (char)(tzinfo != Py_None);
-        me = (PyDateTime_DateTime *) (type->tp_alloc(type , aware));
-        if (me != NULL) {
-            char *pdata = PyBytes_AS_STRING(state);
-
-            memcpy(me->data, pdata, _PyDateTime_DATETIME_DATASIZE);
-            me->hashcode = -1;
-            me->hastzinfo = aware;
-            if (aware) {
-                Py_INCREF(tzinfo);
-                me->tzinfo = tzinfo;
-            }
-        }
-        return (PyObject *)me;
-    }
-
-    if (PyArg_ParseTupleAndKeywords(args, kw, "iii|iiiiO", datetime_kws,
-                                    &year, &month, &day, &hour, &minute,
-                                    &second, &usecond, &tzinfo)) {
-        if (check_date_args(year, month, day) < 0)
-            return NULL;
-        if (check_time_args(hour, minute, second, usecond) < 0)
-            return NULL;
-        if (check_tzinfo_subclass(tzinfo) < 0)
-            return NULL;
-        self = new_datetime_ex(year, month, day,
-                                hour, minute, second, usecond,
-                                tzinfo, type);
-    }
-    return self;
-}
-
-/* TM_FUNC is the shared type of localtime() and gmtime(). */
-typedef struct tm *(*TM_FUNC)(const time_t *timer);
-
-/* Internal helper.
- * Build datetime from a time_t and a distinct count of microseconds.
- * Pass localtime or gmtime for f, to control the interpretation of timet.
- */
-static PyObject *
-datetime_from_timet_and_us(PyObject *cls, TM_FUNC f, time_t timet, int us,
-                           PyObject *tzinfo)
-{
-    struct tm *tm;
-    PyObject *result = NULL;
-
-    tm = f(&timet);
-    if (tm) {
-        /* The platform localtime/gmtime may insert leap seconds,
-         * indicated by tm->tm_sec > 59.  We don't care about them,
-         * except to the extent that passing them on to the datetime
-         * constructor would raise ValueError for a reason that
-         * made no sense to the user.
-         */
-        if (tm->tm_sec > 59)
-            tm->tm_sec = 59;
-        result = PyObject_CallFunction(cls, "iiiiiiiO",
-                                       tm->tm_year + 1900,
-                                       tm->tm_mon + 1,
-                                       tm->tm_mday,
-                                       tm->tm_hour,
-                                       tm->tm_min,
-                                       tm->tm_sec,
-                                       us,
-                                       tzinfo);
-    }
-    else
-        PyErr_SetString(PyExc_ValueError,
-                        "timestamp out of range for "
-                        "platform localtime()/gmtime() function");
-    return result;
-}
-
-/* Internal helper.
- * Build datetime from a Python timestamp.  Pass localtime or gmtime for f,
- * to control the interpretation of the timestamp.  Since a double doesn't
- * have enough bits to cover a datetime's full range of precision, it's
- * better to call datetime_from_timet_and_us provided you have a way
- * to get that much precision (e.g., C time() isn't good enough).
- */
-static PyObject *
-datetime_from_timestamp(PyObject *cls, TM_FUNC f, double timestamp,
-                        PyObject *tzinfo)
-{
-    time_t timet;
-    double fraction;
-    int us;
-
-    timet = _PyTime_DoubleToTimet(timestamp);
-    if (timet == (time_t)-1 && PyErr_Occurred())
-        return NULL;
-    fraction = timestamp - (double)timet;
-    us = (int)round_to_long(fraction * 1e6);
-    if (us < 0) {
-        /* Truncation towards zero is not what we wanted
-           for negative numbers (Python's mod semantics) */
-        timet -= 1;
-        us += 1000000;
-    }
-    /* If timestamp is less than one microsecond smaller than a
-     * full second, round up. Otherwise, ValueErrors are raised
-     * for some floats. */
-    if (us == 1000000) {
-        timet += 1;
-        us = 0;
-    }
-    return datetime_from_timet_and_us(cls, f, timet, us, tzinfo);
-}
-
-/* Internal helper.
- * Build most accurate possible datetime for current time.  Pass localtime or
- * gmtime for f as appropriate.
- */
-static PyObject *
-datetime_best_possible(PyObject *cls, TM_FUNC f, PyObject *tzinfo)
-{
-#ifdef HAVE_GETTIMEOFDAY
-    struct timeval t;
-
-#ifdef GETTIMEOFDAY_NO_TZ
-    gettimeofday(&t);
-#else
-    gettimeofday(&t, (struct timezone *)NULL);
-#endif
-    return datetime_from_timet_and_us(cls, f, t.tv_sec, (int)t.tv_usec,
-                                      tzinfo);
-
-#else   /* ! HAVE_GETTIMEOFDAY */
-    /* No flavor of gettimeofday exists on this platform.  Python's
-     * time.time() does a lot of other platform tricks to get the
-     * best time it can on the platform, and we're not going to do
-     * better than that (if we could, the better code would belong
-     * in time.time()!)  We're limited by the precision of a double,
-     * though.
-     */
-    PyObject *time;
-    double dtime;
-
-    time = time_time();
-    if (time == NULL)
-        return NULL;
-    dtime = PyFloat_AsDouble(time);
-    Py_DECREF(time);
-    if (dtime == -1.0 && PyErr_Occurred())
-        return NULL;
-    return datetime_from_timestamp(cls, f, dtime, tzinfo);
-#endif  /* ! HAVE_GETTIMEOFDAY */
-}
-
-/* Return best possible local time -- this isn't constrained by the
- * precision of a timestamp.
- */
-static PyObject *
-datetime_now(PyObject *cls, PyObject *args, PyObject *kw)
-{
-    PyObject *self;
-    PyObject *tzinfo = Py_None;
-    static char *keywords[] = {"tz", NULL};
-
-    if (! PyArg_ParseTupleAndKeywords(args, kw, "|O:now", keywords,
-                                      &tzinfo))
-        return NULL;
-    if (check_tzinfo_subclass(tzinfo) < 0)
-        return NULL;
-
-    self = datetime_best_possible(cls,
-                                  tzinfo == Py_None ? localtime : gmtime,
-                                  tzinfo);
-    if (self != NULL && tzinfo != Py_None) {
-        /* Convert UTC to tzinfo's zone. */
-        PyObject *temp = self;
-        self = PyObject_CallMethod(tzinfo, "fromutc", "O", self);
-        Py_DECREF(temp);
-    }
-    return self;
-}
-
-/* Return best possible UTC time -- this isn't constrained by the
- * precision of a timestamp.
- */
-static PyObject *
-datetime_utcnow(PyObject *cls, PyObject *dummy)
-{
-    return datetime_best_possible(cls, gmtime, Py_None);
-}
-
-/* Return new local datetime from timestamp (Python timestamp -- a double). */
-static PyObject *
-datetime_fromtimestamp(PyObject *cls, PyObject *args, PyObject *kw)
-{
-    PyObject *self;
-    double timestamp;
-    PyObject *tzinfo = Py_None;
-    static char *keywords[] = {"timestamp", "tz", NULL};
-
-    if (! PyArg_ParseTupleAndKeywords(args, kw, "d|O:fromtimestamp",
-                                      keywords, &timestamp, &tzinfo))
-        return NULL;
-    if (check_tzinfo_subclass(tzinfo) < 0)
-        return NULL;
-
-    self = datetime_from_timestamp(cls,
-                                   tzinfo == Py_None ? localtime : gmtime,
-                                   timestamp,
-                                   tzinfo);
-    if (self != NULL && tzinfo != Py_None) {
-        /* Convert UTC to tzinfo's zone. */
-        PyObject *temp = self;
-        self = PyObject_CallMethod(tzinfo, "fromutc", "O", self);
-        Py_DECREF(temp);
-    }
-    return self;
-}
-
-/* Return new UTC datetime from timestamp (Python timestamp -- a double). */
-static PyObject *
-datetime_utcfromtimestamp(PyObject *cls, PyObject *args)
-{
-    double timestamp;
-    PyObject *result = NULL;
-
-    if (PyArg_ParseTuple(args, "d:utcfromtimestamp", &timestamp))
-        result = datetime_from_timestamp(cls, gmtime, timestamp,
-                                         Py_None);
-    return result;
-}
-
-/* Return new datetime from _strptime.strptime_datetime(). */
-static PyObject *
-datetime_strptime(PyObject *cls, PyObject *args)
-{
-    static PyObject *module = NULL;
-    const Py_UNICODE *string, *format;
-
-    if (!PyArg_ParseTuple(args, "uu:strptime", &string, &format))
-        return NULL;
-
-    if (module == NULL) {
-        module = PyImport_ImportModuleNoBlock("_strptime");
-        if (module == NULL)
-            return NULL;
-    }
-    return PyObject_CallMethod(module, "_strptime_datetime", "Ouu",
-                               cls, string, format);
-}
-
-/* Return new datetime from date/datetime and time arguments. */
-static PyObject *
-datetime_combine(PyObject *cls, PyObject *args, PyObject *kw)
-{
-    static char *keywords[] = {"date", "time", NULL};
-    PyObject *date;
-    PyObject *time;
-    PyObject *result = NULL;
-
-    if (PyArg_ParseTupleAndKeywords(args, kw, "O!O!:combine", keywords,
-                                    &PyDateTime_DateType, &date,
-                                    &PyDateTime_TimeType, &time)) {
-        PyObject *tzinfo = Py_None;
-
-        if (HASTZINFO(time))
-            tzinfo = ((PyDateTime_Time *)time)->tzinfo;
-        result = PyObject_CallFunction(cls, "iiiiiiiO",
-                                        GET_YEAR(date),
-                                        GET_MONTH(date),
-                                        GET_DAY(date),
-                                        TIME_GET_HOUR(time),
-                                        TIME_GET_MINUTE(time),
-                                        TIME_GET_SECOND(time),
-                                        TIME_GET_MICROSECOND(time),
-                                        tzinfo);
-    }
-    return result;
-}
-
-/*
- * Destructor.
- */
-
-static void
-datetime_dealloc(PyDateTime_DateTime *self)
-{
-    if (HASTZINFO(self)) {
-        Py_XDECREF(self->tzinfo);
-    }
-    Py_TYPE(self)->tp_free((PyObject *)self);
-}
-
-/*
- * Indirect access to tzinfo methods.
- */
-
-/* These are all METH_NOARGS, so don't need to check the arglist. */
-static PyObject *
-datetime_utcoffset(PyObject *self, PyObject *unused) {
-    return call_utcoffset(GET_DT_TZINFO(self), self);
-}
-
-static PyObject *
-datetime_dst(PyObject *self, PyObject *unused) {
-    return call_dst(GET_DT_TZINFO(self), self);
-}
-
-static PyObject *
-datetime_tzname(PyObject *self, PyObject *unused) {
-    return call_tzname(GET_DT_TZINFO(self), self);
-}
-
-/*
- * datetime arithmetic.
- */
-
-/* factor must be 1 (to add) or -1 (to subtract).  The result inherits
- * the tzinfo state of date.
- */
-static PyObject *
-add_datetime_timedelta(PyDateTime_DateTime *date, PyDateTime_Delta *delta,
-                       int factor)
-{
-    /* Note that the C-level additions can't overflow, because of
-     * invariant bounds on the member values.
-     */
-    int year = GET_YEAR(date);
-    int month = GET_MONTH(date);
-    int day = GET_DAY(date) + GET_TD_DAYS(delta) * factor;
-    int hour = DATE_GET_HOUR(date);
-    int minute = DATE_GET_MINUTE(date);
-    int second = DATE_GET_SECOND(date) + GET_TD_SECONDS(delta) * factor;
-    int microsecond = DATE_GET_MICROSECOND(date) +
-                      GET_TD_MICROSECONDS(delta) * factor;
-
-    assert(factor == 1 || factor == -1);
-    if (normalize_datetime(&year, &month, &day,
-                           &hour, &minute, &second, &microsecond) < 0)
-        return NULL;
-    else
-        return new_datetime(year, month, day,
-                            hour, minute, second, microsecond,
-                            HASTZINFO(date) ? date->tzinfo : Py_None);
-}
-
-static PyObject *
-datetime_add(PyObject *left, PyObject *right)
-{
-    if (PyDateTime_Check(left)) {
-        /* datetime + ??? */
-        if (PyDelta_Check(right))
-            /* datetime + delta */
-            return add_datetime_timedelta(
-                            (PyDateTime_DateTime *)left,
-                            (PyDateTime_Delta *)right,
-                            1);
-    }
-    else if (PyDelta_Check(left)) {
-        /* delta + datetime */
-        return add_datetime_timedelta((PyDateTime_DateTime *) right,
-                                      (PyDateTime_Delta *) left,
-                                      1);
-    }
-    Py_INCREF(Py_NotImplemented);
-    return Py_NotImplemented;
-}
-
-static PyObject *
-datetime_subtract(PyObject *left, PyObject *right)
-{
-    PyObject *result = Py_NotImplemented;
-
-    if (PyDateTime_Check(left)) {
-        /* datetime - ??? */
-        if (PyDateTime_Check(right)) {
-            /* datetime - datetime */
-            PyObject *offset1, *offset2, *offdiff = NULL;
-            int delta_d, delta_s, delta_us;
-
-            if (GET_DT_TZINFO(left) == GET_DT_TZINFO(right)) {
-                offset2 = offset1 = Py_None;
-                Py_INCREF(offset1);
-                Py_INCREF(offset2);
-            }
-            else {
-                offset1 = datetime_utcoffset(left, NULL);
-                if (offset1 == NULL)
-                    return NULL;
-                offset2 = datetime_utcoffset(right, NULL);
-                if (offset2 == NULL) {
-                    Py_DECREF(offset1);
-                    return NULL;
-                }
-                if ((offset1 != Py_None) != (offset2 != Py_None)) {
-                    PyErr_SetString(PyExc_TypeError,
-                                    "can't subtract offset-naive and "
-                                    "offset-aware datetimes");
-                    Py_DECREF(offset1);
-                    Py_DECREF(offset2);
-                    return NULL;
-                }
-            }
-            if ((offset1 != offset2) &&
-                delta_cmp(offset1, offset2) != 0) {
-                offdiff = delta_subtract(offset1, offset2);
-                if (offdiff == NULL) {
-                    Py_DECREF(offset1);
-                    Py_DECREF(offset2);
-                    return NULL;
-                }
-            }
-            Py_DECREF(offset1);
-            Py_DECREF(offset2);
-            delta_d = ymd_to_ord(GET_YEAR(left),
-                                 GET_MONTH(left),
-                                 GET_DAY(left)) -
-                      ymd_to_ord(GET_YEAR(right),
-                                 GET_MONTH(right),
-                                 GET_DAY(right));
-            /* These can't overflow, since the values are
-             * normalized.  At most this gives the number of
-             * seconds in one day.
-             */
-            delta_s = (DATE_GET_HOUR(left) -
-                       DATE_GET_HOUR(right)) * 3600 +
-                      (DATE_GET_MINUTE(left) -
-                       DATE_GET_MINUTE(right)) * 60 +
-                      (DATE_GET_SECOND(left) -
-                       DATE_GET_SECOND(right));
-            delta_us = DATE_GET_MICROSECOND(left) -
-                       DATE_GET_MICROSECOND(right);
-            result = new_delta(delta_d, delta_s, delta_us, 1);
-            if (offdiff != NULL) {
-                PyObject *temp = result;
-                result = delta_subtract(result, offdiff);
-                Py_DECREF(temp);
-                Py_DECREF(offdiff);
-            }
-        }
-        else if (PyDelta_Check(right)) {
-            /* datetime - delta */
-            result = add_datetime_timedelta(
-                            (PyDateTime_DateTime *)left,
-                            (PyDateTime_Delta *)right,
-                            -1);
-        }
-    }
-
-    if (result == Py_NotImplemented)
-        Py_INCREF(result);
-    return result;
-}
-
-/* Various ways to turn a datetime into a string. */
-
-static PyObject *
-datetime_repr(PyDateTime_DateTime *self)
-{
-    const char *type_name = Py_TYPE(self)->tp_name;
-    PyObject *baserepr;
-
-    if (DATE_GET_MICROSECOND(self)) {
-        baserepr = PyUnicode_FromFormat(
-                      "%s(%d, %d, %d, %d, %d, %d, %d)",
-                      type_name,
-                      GET_YEAR(self), GET_MONTH(self), GET_DAY(self),
-                      DATE_GET_HOUR(self), DATE_GET_MINUTE(self),
-                      DATE_GET_SECOND(self),
-                      DATE_GET_MICROSECOND(self));
-    }
-    else if (DATE_GET_SECOND(self)) {
-        baserepr = PyUnicode_FromFormat(
-                      "%s(%d, %d, %d, %d, %d, %d)",
-                      type_name,
-                      GET_YEAR(self), GET_MONTH(self), GET_DAY(self),
-                      DATE_GET_HOUR(self), DATE_GET_MINUTE(self),
-                      DATE_GET_SECOND(self));
-    }
-    else {
-        baserepr = PyUnicode_FromFormat(
-                      "%s(%d, %d, %d, %d, %d)",
-                      type_name,
-                      GET_YEAR(self), GET_MONTH(self), GET_DAY(self),
-                      DATE_GET_HOUR(self), DATE_GET_MINUTE(self));
-    }
-    if (baserepr == NULL || ! HASTZINFO(self))
-        return baserepr;
-    return append_keyword_tzinfo(baserepr, self->tzinfo);
-}
-
-static PyObject *
-datetime_str(PyDateTime_DateTime *self)
-{
-    return PyObject_CallMethod((PyObject *)self, "isoformat", "(s)", " ");
-}
-
-static PyObject *
-datetime_isoformat(PyDateTime_DateTime *self, PyObject *args, PyObject *kw)
-{
-    int sep = 'T';
-    static char *keywords[] = {"sep", NULL};
-    char buffer[100];
-    PyObject *result;
-    int us = DATE_GET_MICROSECOND(self);
-
-    if (!PyArg_ParseTupleAndKeywords(args, kw, "|C:isoformat", keywords, &sep))
-        return NULL;
-    if (us)
-        result = PyUnicode_FromFormat("%04d-%02d-%02d%c%02d:%02d:%02d.%06d",
-                                      GET_YEAR(self), GET_MONTH(self),
-                                      GET_DAY(self), (int)sep,
-                                      DATE_GET_HOUR(self), DATE_GET_MINUTE(self),
-                                      DATE_GET_SECOND(self), us);
-    else
-        result = PyUnicode_FromFormat("%04d-%02d-%02d%c%02d:%02d:%02d",
-                                      GET_YEAR(self), GET_MONTH(self),
-                                      GET_DAY(self), (int)sep,
-                                      DATE_GET_HOUR(self), DATE_GET_MINUTE(self),
-                                      DATE_GET_SECOND(self));
-
-    if (!result || !HASTZINFO(self))
-        return result;
-
-    /* We need to append the UTC offset. */
-    if (format_utcoffset(buffer, sizeof(buffer), ":", self->tzinfo,
-                         (PyObject *)self) < 0) {
-        Py_DECREF(result);
-        return NULL;
-    }
-    PyUnicode_AppendAndDel(&result, PyUnicode_FromString(buffer));
-    return result;
-}
-
-static PyObject *
-datetime_ctime(PyDateTime_DateTime *self)
-{
-    return format_ctime((PyDateTime_Date *)self,
-                        DATE_GET_HOUR(self),
-                        DATE_GET_MINUTE(self),
-                        DATE_GET_SECOND(self));
-}
-
-/* Miscellaneous methods. */
-
-static PyObject *
-datetime_richcompare(PyObject *self, PyObject *other, int op)
-{
-    PyObject *result = NULL;
-    PyObject *offset1, *offset2;
-    int diff;
-
-    if (! PyDateTime_Check(other)) {
-        if (PyDate_Check(other)) {
-            /* Prevent invocation of date_richcompare.  We want to
-               return NotImplemented here to give the other object
-               a chance.  But since DateTime is a subclass of
-               Date, if the other object is a Date, it would
-               compute an ordering based on the date part alone,
-               and we don't want that.  So force unequal or
-               uncomparable here in that case. */
-            if (op == Py_EQ)
-                Py_RETURN_FALSE;
-            if (op == Py_NE)
-                Py_RETURN_TRUE;
-            return cmperror(self, other);
-        }
-        Py_INCREF(Py_NotImplemented);
-        return Py_NotImplemented;
-    }
-
-    if (GET_DT_TZINFO(self) == GET_DT_TZINFO(other)) {
-        diff = memcmp(((PyDateTime_DateTime *)self)->data,
-                      ((PyDateTime_DateTime *)other)->data,
-                      _PyDateTime_DATETIME_DATASIZE);
-        return diff_to_bool(diff, op);
-    }
-    offset1 = datetime_utcoffset(self, NULL);
-    if (offset1 == NULL)
-        return NULL;
-    offset2 = datetime_utcoffset(other, NULL);
-    if (offset2 == NULL)
-        goto done;
-    /* If they're both naive, or both aware and have the same offsets,
-     * we get off cheap.  Note that if they're both naive, offset1 ==
-     * offset2 == Py_None at this point.
-     */
-    if ((offset1 == offset2) ||
-        (PyDelta_Check(offset1) && PyDelta_Check(offset2) &&
-         delta_cmp(offset1, offset2) == 0)) {
-        diff = memcmp(((PyDateTime_DateTime *)self)->data,
-                      ((PyDateTime_DateTime *)other)->data,
-                      _PyDateTime_DATETIME_DATASIZE);
-        result = diff_to_bool(diff, op);
-    }
-    else if (offset1 != Py_None && offset2 != Py_None) {
-        PyDateTime_Delta *delta;
-
-        assert(offset1 != offset2); /* else last "if" handled it */
-        delta = (PyDateTime_Delta *)datetime_subtract((PyObject *)self,
-                                                       other);
-        if (delta == NULL)
-            goto done;
-        diff = GET_TD_DAYS(delta);
-        if (diff == 0)
-            diff = GET_TD_SECONDS(delta) |
-                   GET_TD_MICROSECONDS(delta);
-        Py_DECREF(delta);
-        result = diff_to_bool(diff, op);
-    }
-    else {
-        PyErr_SetString(PyExc_TypeError,
-                        "can't compare offset-naive and "
-                        "offset-aware datetimes");
-    }
- done:
-    Py_DECREF(offset1);
-    Py_XDECREF(offset2);
-    return result;
-}
-
-static long
-datetime_hash(PyDateTime_DateTime *self)
-{
-    if (self->hashcode == -1) {
-        PyObject *offset;
-
-        offset = datetime_utcoffset((PyObject *)self, NULL);
-
-        if (offset == NULL)
-            return -1;
-
-        /* Reduce this to a hash of another object. */
-        if (offset == Py_None)
-            self->hashcode = generic_hash(
-                (unsigned char *)self->data, _PyDateTime_DATETIME_DATASIZE);
-        else {
-            PyObject *temp1, *temp2;
-            int days, seconds;
-
-            assert(HASTZINFO(self));
-            days = ymd_to_ord(GET_YEAR(self),
-                              GET_MONTH(self),
-                              GET_DAY(self));
-            seconds = DATE_GET_HOUR(self) * 3600 +
-                      DATE_GET_MINUTE(self) * 60 +
-                      DATE_GET_SECOND(self);
-            temp1 = new_delta(days, seconds,
-                              DATE_GET_MICROSECOND(self),
-                              1);
-            if (temp1 == NULL) {
-                Py_DECREF(offset);
-                return -1;
-            }
-            temp2 = delta_subtract(temp1, offset);
-            Py_DECREF(temp1);
-            if (temp2 == NULL) {
-                Py_DECREF(offset);
-                return -1;
-            }
-            self->hashcode = PyObject_Hash(temp2);
-            Py_DECREF(temp2);
-        }
-        Py_DECREF(offset);
-    }
-    return self->hashcode;
-}
-
-static PyObject *
-datetime_replace(PyDateTime_DateTime *self, PyObject *args, PyObject *kw)
-{
-    PyObject *clone;
-    PyObject *tuple;
-    int y = GET_YEAR(self);
-    int m = GET_MONTH(self);
-    int d = GET_DAY(self);
-    int hh = DATE_GET_HOUR(self);
-    int mm = DATE_GET_MINUTE(self);
-    int ss = DATE_GET_SECOND(self);
-    int us = DATE_GET_MICROSECOND(self);
-    PyObject *tzinfo = HASTZINFO(self) ? self->tzinfo : Py_None;
-
-    if (! PyArg_ParseTupleAndKeywords(args, kw, "|iiiiiiiO:replace",
-                                      datetime_kws,
-                                      &y, &m, &d, &hh, &mm, &ss, &us,
-                                      &tzinfo))
-        return NULL;
-    tuple = Py_BuildValue("iiiiiiiO", y, m, d, hh, mm, ss, us, tzinfo);
-    if (tuple == NULL)
-        return NULL;
-    clone = datetime_new(Py_TYPE(self), tuple, NULL);
-    Py_DECREF(tuple);
-    return clone;
-}
-
-static PyObject *
-datetime_astimezone(PyDateTime_DateTime *self, PyObject *args, PyObject *kw)
-{
-    PyObject *result;
-    PyObject *offset;
-    PyObject *temp;
-    PyObject *tzinfo;
-    static char *keywords[] = {"tz", NULL};
-
-    if (! PyArg_ParseTupleAndKeywords(args, kw, "O!:astimezone", keywords,
-                                      &PyDateTime_TZInfoType, &tzinfo))
-        return NULL;
-
-    if (!HASTZINFO(self) || self->tzinfo == Py_None)
-        goto NeedAware;
-
-    /* Conversion to self's own time zone is a NOP. */
-    if (self->tzinfo == tzinfo) {
-        Py_INCREF(self);
-        return (PyObject *)self;
-    }
-
-    /* Convert self to UTC. */
-    offset = datetime_utcoffset((PyObject *)self, NULL);
-    if (offset == NULL)
-        return NULL;
-    if (offset == Py_None) {
-        Py_DECREF(offset);
-      NeedAware:
-        PyErr_SetString(PyExc_ValueError, "astimezone() cannot be applied to "
-                        "a naive datetime");
-        return NULL;
-    }
-
-    /* result = self - offset */
-    result = add_datetime_timedelta(self,
-                (PyDateTime_Delta *)offset, -1);
-    Py_DECREF(offset);
-    if (result == NULL)
-        return NULL;
-
-    /* Attach new tzinfo and let fromutc() do the rest. */
-    temp = ((PyDateTime_DateTime *)result)->tzinfo;
-    ((PyDateTime_DateTime *)result)->tzinfo = tzinfo;
-    Py_INCREF(tzinfo);
-    Py_DECREF(temp);
-
-    temp = result;
-    result = PyObject_CallMethod(tzinfo, "fromutc", "O", temp);
-    Py_DECREF(temp);
-
-    return result;
-}
-
-static PyObject *
-datetime_timetuple(PyDateTime_DateTime *self)
-{
-    int dstflag = -1;
-
-    if (HASTZINFO(self) && self->tzinfo != Py_None) {
-        PyObject * dst;
-
-        dst = call_dst(self->tzinfo, (PyObject *)self);
-        if (dst == NULL)
-            return NULL;
-
-        if (dst != Py_None)
-            dstflag = delta_bool((PyDateTime_Delta *)dst);
-        Py_DECREF(dst);
-    }
-    return build_struct_time(GET_YEAR(self),
-                             GET_MONTH(self),
-                             GET_DAY(self),
-                             DATE_GET_HOUR(self),
-                             DATE_GET_MINUTE(self),
-                             DATE_GET_SECOND(self),
-                             dstflag);
-}
-
-static PyObject *
-datetime_getdate(PyDateTime_DateTime *self)
-{
-    return new_date(GET_YEAR(self),
-                    GET_MONTH(self),
-                    GET_DAY(self));
-}
-
-static PyObject *
-datetime_gettime(PyDateTime_DateTime *self)
-{
-    return new_time(DATE_GET_HOUR(self),
-                    DATE_GET_MINUTE(self),
-                    DATE_GET_SECOND(self),
-                    DATE_GET_MICROSECOND(self),
-                    Py_None);
-}
-
-static PyObject *
-datetime_gettimetz(PyDateTime_DateTime *self)
-{
-    return new_time(DATE_GET_HOUR(self),
-                    DATE_GET_MINUTE(self),
-                    DATE_GET_SECOND(self),
-                    DATE_GET_MICROSECOND(self),
-                    GET_DT_TZINFO(self));
-}
-
-static PyObject *
-datetime_utctimetuple(PyDateTime_DateTime *self)
-{
-    int y, m, d, hh, mm, ss;
-    PyObject *tzinfo;
-    PyDateTime_DateTime *utcself;
-
-    tzinfo = GET_DT_TZINFO(self);
-    if (tzinfo == Py_None) {
-        utcself = self;
-        Py_INCREF(utcself);
-    }
-    else {
-        PyObject *offset;
-        offset = call_utcoffset(tzinfo, (PyObject *)self);
-        if (offset == NULL)
-            return NULL;
-        if (offset == Py_None) {
-            Py_DECREF(offset);
-            utcself = self;
-            Py_INCREF(utcself);
-        }
-        else {
-            utcself = (PyDateTime_DateTime *)add_datetime_timedelta(self,
-                                                (PyDateTime_Delta *)offset, -1);
-            Py_DECREF(offset);
-            if (utcself == NULL)
-                return NULL;
-        }
-    }
-    y = GET_YEAR(utcself);
-    m = GET_MONTH(utcself);
-    d = GET_DAY(utcself);
-    hh = DATE_GET_HOUR(utcself);
-    mm = DATE_GET_MINUTE(utcself);
-    ss = DATE_GET_SECOND(utcself);
-
-    Py_DECREF(utcself);
-    return build_struct_time(y, m, d, hh, mm, ss, 0);
-}
-
-/* Pickle support, a simple use of __reduce__. */
-
-/* Let basestate be the non-tzinfo data string.
- * If tzinfo is None, this returns (basestate,), else (basestate, tzinfo).
- * So it's a tuple in any (non-error) case.
- * __getstate__ isn't exposed.
- */
-static PyObject *
-datetime_getstate(PyDateTime_DateTime *self)
-{
-    PyObject *basestate;
-    PyObject *result = NULL;
-
-    basestate = PyBytes_FromStringAndSize((char *)self->data,
-                                           _PyDateTime_DATETIME_DATASIZE);
-    if (basestate != NULL) {
-        if (! HASTZINFO(self) || self->tzinfo == Py_None)
-            result = PyTuple_Pack(1, basestate);
-        else
-            result = PyTuple_Pack(2, basestate, self->tzinfo);
-        Py_DECREF(basestate);
-    }
-    return result;
-}
-
-static PyObject *
-datetime_reduce(PyDateTime_DateTime *self, PyObject *arg)
-{
-    return Py_BuildValue("(ON)", Py_TYPE(self), datetime_getstate(self));
-}
-
-static PyMethodDef datetime_methods[] = {
-
-    /* Class methods: */
-
-    {"now",         (PyCFunction)datetime_now,
-     METH_VARARGS | METH_KEYWORDS | METH_CLASS,
-     PyDoc_STR("[tz] -> new datetime with tz's local day and time.")},
-
-    {"utcnow",         (PyCFunction)datetime_utcnow,
-     METH_NOARGS | METH_CLASS,
-     PyDoc_STR("Return a new datetime representing UTC day and time.")},
-
-    {"fromtimestamp", (PyCFunction)datetime_fromtimestamp,
-     METH_VARARGS | METH_KEYWORDS | METH_CLASS,
-     PyDoc_STR("timestamp[, tz] -> tz's local time from POSIX timestamp.")},
-
-    {"utcfromtimestamp", (PyCFunction)datetime_utcfromtimestamp,
-     METH_VARARGS | METH_CLASS,
-     PyDoc_STR("timestamp -> UTC datetime from a POSIX timestamp "
-               "(like time.time()).")},
-
-    {"strptime", (PyCFunction)datetime_strptime,
-     METH_VARARGS | METH_CLASS,
-     PyDoc_STR("string, format -> new datetime parsed from a string "
-               "(like time.strptime()).")},
-
-    {"combine", (PyCFunction)datetime_combine,
-     METH_VARARGS | METH_KEYWORDS | METH_CLASS,
-     PyDoc_STR("date, time -> datetime with same date and time fields")},
-
-    /* Instance methods: */
-
-    {"date",   (PyCFunction)datetime_getdate, METH_NOARGS,
-     PyDoc_STR("Return date object with same year, month and day.")},
-
-    {"time",   (PyCFunction)datetime_gettime, METH_NOARGS,
-     PyDoc_STR("Return time object with same time but with tzinfo=None.")},
-
-    {"timetz",   (PyCFunction)datetime_gettimetz, METH_NOARGS,
-     PyDoc_STR("Return time object with same time and tzinfo.")},
-
-    {"ctime",       (PyCFunction)datetime_ctime,        METH_NOARGS,
-     PyDoc_STR("Return ctime() style string.")},
-
-    {"timetuple",   (PyCFunction)datetime_timetuple, METH_NOARGS,
-     PyDoc_STR("Return time tuple, compatible with time.localtime().")},
-
-    {"utctimetuple",   (PyCFunction)datetime_utctimetuple, METH_NOARGS,
-     PyDoc_STR("Return UTC time tuple, compatible with time.localtime().")},
-
-    {"isoformat",   (PyCFunction)datetime_isoformat, METH_VARARGS | METH_KEYWORDS,
-     PyDoc_STR("[sep] -> string in ISO 8601 format, "
-               "YYYY-MM-DDTHH:MM:SS[.mmmmmm][+HH:MM].\n\n"
-               "sep is used to separate the year from the time, and "
-               "defaults to 'T'.")},
-
-    {"utcoffset",       (PyCFunction)datetime_utcoffset, METH_NOARGS,
-     PyDoc_STR("Return self.tzinfo.utcoffset(self).")},
-
-    {"tzname",          (PyCFunction)datetime_tzname,   METH_NOARGS,
-     PyDoc_STR("Return self.tzinfo.tzname(self).")},
-
-    {"dst",             (PyCFunction)datetime_dst, METH_NOARGS,
-     PyDoc_STR("Return self.tzinfo.dst(self).")},
-
-    {"replace",     (PyCFunction)datetime_replace,      METH_VARARGS | METH_KEYWORDS,
-     PyDoc_STR("Return datetime with new specified fields.")},
-
-    {"astimezone",  (PyCFunction)datetime_astimezone, METH_VARARGS | METH_KEYWORDS,
-     PyDoc_STR("tz -> convert to local time in new timezone tz\n")},
-
-    {"__reduce__", (PyCFunction)datetime_reduce,     METH_NOARGS,
-     PyDoc_STR("__reduce__() -> (cls, state)")},
-
-    {NULL,      NULL}
-};
-
-static char datetime_doc[] =
-PyDoc_STR("datetime(year, month, day[, hour[, minute[, second[, microsecond[,tzinfo]]]]])\n\
-\n\
-The year, month and day arguments are required. tzinfo may be None, or an\n\
-instance of a tzinfo subclass. The remaining arguments may be ints or longs.\n");
-
-static PyNumberMethods datetime_as_number = {
-    datetime_add,                               /* nb_add */
-    datetime_subtract,                          /* nb_subtract */
-    0,                                          /* nb_multiply */
-    0,                                          /* nb_remainder */
-    0,                                          /* nb_divmod */
-    0,                                          /* nb_power */
-    0,                                          /* nb_negative */
-    0,                                          /* nb_positive */
-    0,                                          /* nb_absolute */
-    0,                                          /* nb_bool */
-};
-
-static PyTypeObject PyDateTime_DateTimeType = {
-    PyVarObject_HEAD_INIT(NULL, 0)
-    "datetime.datetime",                        /* tp_name */
-    sizeof(PyDateTime_DateTime),                /* tp_basicsize */
-    0,                                          /* tp_itemsize */
-    (destructor)datetime_dealloc,               /* tp_dealloc */
-    0,                                          /* tp_print */
-    0,                                          /* tp_getattr */
-    0,                                          /* tp_setattr */
-    0,                                          /* tp_reserved */
-    (reprfunc)datetime_repr,                    /* tp_repr */
-    &datetime_as_number,                        /* tp_as_number */
-    0,                                          /* tp_as_sequence */
-    0,                                          /* tp_as_mapping */
-    (hashfunc)datetime_hash,                    /* tp_hash */
-    0,                                          /* tp_call */
-    (reprfunc)datetime_str,                     /* tp_str */
-    PyObject_GenericGetAttr,                    /* tp_getattro */
-    0,                                          /* tp_setattro */
-    0,                                          /* tp_as_buffer */
-    Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE, /* tp_flags */
-    datetime_doc,                               /* tp_doc */
-    0,                                          /* tp_traverse */
-    0,                                          /* tp_clear */
-    datetime_richcompare,                       /* tp_richcompare */
-    0,                                          /* tp_weaklistoffset */
-    0,                                          /* tp_iter */
-    0,                                          /* tp_iternext */
-    datetime_methods,                           /* tp_methods */
-    0,                                          /* tp_members */
-    datetime_getset,                            /* tp_getset */
-    &PyDateTime_DateType,                       /* tp_base */
-    0,                                          /* tp_dict */
-    0,                                          /* tp_descr_get */
-    0,                                          /* tp_descr_set */
-    0,                                          /* tp_dictoffset */
-    0,                                          /* tp_init */
-    datetime_alloc,                             /* tp_alloc */
-    datetime_new,                               /* tp_new */
-    0,                                          /* tp_free */
-};
-
-/* ---------------------------------------------------------------------------
- * Module methods and initialization.
- */
-
-static PyMethodDef module_methods[] = {
-    {NULL, NULL}
-};
-
-/* C API.  Clients get at this via PyDateTime_IMPORT, defined in
- * datetime.h.
- */
-static PyDateTime_CAPI CAPI = {
-    &PyDateTime_DateType,
-    &PyDateTime_DateTimeType,
-    &PyDateTime_TimeType,
-    &PyDateTime_DeltaType,
-    &PyDateTime_TZInfoType,
-    new_date_ex,
-    new_datetime_ex,
-    new_time_ex,
-    new_delta_ex,
-    datetime_fromtimestamp,
-    date_fromtimestamp
-};
-
-
-
-static struct PyModuleDef datetimemodule = {
-    PyModuleDef_HEAD_INIT,
-    "datetime",
-    "Fast implementation of the datetime type.",
-    -1,
-    module_methods,
-    NULL,
-    NULL,
-    NULL,
-    NULL
-};
-
-PyMODINIT_FUNC
-PyInit_datetime(void)
-{
-    PyObject *m;        /* a module object */
-    PyObject *d;        /* its dict */
-    PyObject *x;
-    PyObject *delta;
-
-    m = PyModule_Create(&datetimemodule);
-    if (m == NULL)
-        return NULL;
-
-    if (PyType_Ready(&PyDateTime_DateType) < 0)
-        return NULL;
-    if (PyType_Ready(&PyDateTime_DateTimeType) < 0)
-        return NULL;
-    if (PyType_Ready(&PyDateTime_DeltaType) < 0)
-        return NULL;
-    if (PyType_Ready(&PyDateTime_TimeType) < 0)
-        return NULL;
-    if (PyType_Ready(&PyDateTime_TZInfoType) < 0)
-        return NULL;
-    if (PyType_Ready(&PyDateTime_TimeZoneType) < 0)
-        return NULL;
-
-    /* timedelta values */
-    d = PyDateTime_DeltaType.tp_dict;
-
-    x = new_delta(0, 0, 1, 0);
-    if (x == NULL || PyDict_SetItemString(d, "resolution", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    x = new_delta(-MAX_DELTA_DAYS, 0, 0, 0);
-    if (x == NULL || PyDict_SetItemString(d, "min", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    x = new_delta(MAX_DELTA_DAYS, 24*3600-1, 1000000-1, 0);
-    if (x == NULL || PyDict_SetItemString(d, "max", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    /* date values */
-    d = PyDateTime_DateType.tp_dict;
-
-    x = new_date(1, 1, 1);
-    if (x == NULL || PyDict_SetItemString(d, "min", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    x = new_date(MAXYEAR, 12, 31);
-    if (x == NULL || PyDict_SetItemString(d, "max", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    x = new_delta(1, 0, 0, 0);
-    if (x == NULL || PyDict_SetItemString(d, "resolution", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    /* time values */
-    d = PyDateTime_TimeType.tp_dict;
-
-    x = new_time(0, 0, 0, 0, Py_None);
-    if (x == NULL || PyDict_SetItemString(d, "min", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    x = new_time(23, 59, 59, 999999, Py_None);
-    if (x == NULL || PyDict_SetItemString(d, "max", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    x = new_delta(0, 0, 1, 0);
-    if (x == NULL || PyDict_SetItemString(d, "resolution", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    /* datetime values */
-    d = PyDateTime_DateTimeType.tp_dict;
-
-    x = new_datetime(1, 1, 1, 0, 0, 0, 0, Py_None);
-    if (x == NULL || PyDict_SetItemString(d, "min", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    x = new_datetime(MAXYEAR, 12, 31, 23, 59, 59, 999999, Py_None);
-    if (x == NULL || PyDict_SetItemString(d, "max", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    x = new_delta(0, 0, 1, 0);
-    if (x == NULL || PyDict_SetItemString(d, "resolution", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    /* timezone values */
-    d = PyDateTime_TimeZoneType.tp_dict;
-
-    delta = new_delta(0, 0, 0, 0);
-    if (delta == NULL)
-        return NULL;
-    x = new_timezone(delta, NULL);
-    Py_DECREF(delta);
-    if (x == NULL || PyDict_SetItemString(d, "utc", x) < 0)
-        return NULL;
-    PyDateTime_TimeZone_UTC = x;
-
-    delta = new_delta(-1, 60, 0, 1); /* -23:59 */
-    if (delta == NULL)
-        return NULL;
-    x = new_timezone(delta, NULL);
-    Py_DECREF(delta);
-    if (x == NULL || PyDict_SetItemString(d, "min", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    delta = new_delta(0, (23 * 60 + 59) * 60, 0, 0); /* +23:59 */
-    if (delta == NULL)
-        return NULL;
-    x = new_timezone(delta, NULL);
-    Py_DECREF(delta);
-    if (x == NULL || PyDict_SetItemString(d, "max", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    /* module initialization */
-    PyModule_AddIntConstant(m, "MINYEAR", MINYEAR);
-    PyModule_AddIntConstant(m, "MAXYEAR", MAXYEAR);
-
-    Py_INCREF(&PyDateTime_DateType);
-    PyModule_AddObject(m, "date", (PyObject *) &PyDateTime_DateType);
-
-    Py_INCREF(&PyDateTime_DateTimeType);
-    PyModule_AddObject(m, "datetime",
-                       (PyObject *)&PyDateTime_DateTimeType);
-
-    Py_INCREF(&PyDateTime_TimeType);
-    PyModule_AddObject(m, "time", (PyObject *) &PyDateTime_TimeType);
-
-    Py_INCREF(&PyDateTime_DeltaType);
-    PyModule_AddObject(m, "timedelta", (PyObject *) &PyDateTime_DeltaType);
-
-    Py_INCREF(&PyDateTime_TZInfoType);
-    PyModule_AddObject(m, "tzinfo", (PyObject *) &PyDateTime_TZInfoType);
-
-    Py_INCREF(&PyDateTime_TimeZoneType);
-    PyModule_AddObject(m, "timezone", (PyObject *) &PyDateTime_TimeZoneType);
-
-    x = PyCapsule_New(&CAPI, PyDateTime_CAPSULE_NAME, NULL);
-    if (x == NULL)
-        return NULL;
-    PyModule_AddObject(m, "datetime_CAPI", x);
-
-    /* A 4-year cycle has an extra leap day over what we'd get from
-     * pasting together 4 single years.
-     */
-    assert(DI4Y == 4 * 365 + 1);
-    assert(DI4Y == days_before_year(4+1));
-
-    /* Similarly, a 400-year cycle has an extra leap day over what we'd
-     * get from pasting together 4 100-year cycles.
-     */
-    assert(DI400Y == 4 * DI100Y + 1);
-    assert(DI400Y == days_before_year(400+1));
-
-    /* OTOH, a 100-year cycle has one fewer leap day than we'd get from
-     * pasting together 25 4-year cycles.
-     */
-    assert(DI100Y == 25 * DI4Y - 1);
-    assert(DI100Y == days_before_year(100+1));
-
-    us_per_us = PyLong_FromLong(1);
-    us_per_ms = PyLong_FromLong(1000);
-    us_per_second = PyLong_FromLong(1000000);
-    us_per_minute = PyLong_FromLong(60000000);
-    seconds_per_day = PyLong_FromLong(24 * 3600);
-    if (us_per_us == NULL || us_per_ms == NULL || us_per_second == NULL ||
-        us_per_minute == NULL || seconds_per_day == NULL)
-        return NULL;
-
-    /* The rest are too big for 32-bit ints, but even
-     * us_per_week fits in 40 bits, so doubles should be exact.
-     */
-    us_per_hour = PyLong_FromDouble(3600000000.0);
-    us_per_day = PyLong_FromDouble(86400000000.0);
-    us_per_week = PyLong_FromDouble(604800000000.0);
-    if (us_per_hour == NULL || us_per_day == NULL || us_per_week == NULL)
-        return NULL;
-    return m;
-}
-
-/* ---------------------------------------------------------------------------
-Some time zone algebra.  For a datetime x, let
-    x.n = x stripped of its timezone -- its naive time.
-    x.o = x.utcoffset(), and assuming that doesn't raise an exception or
-      return None
-    x.d = x.dst(), and assuming that doesn't raise an exception or
-      return None
-    x.s = x's standard offset, x.o - x.d
-
-Now some derived rules, where k is a duration (timedelta).
-
-1. x.o = x.s + x.d
-   This follows from the definition of x.s.
-
-2. If x and y have the same tzinfo member, x.s = y.s.
-   This is actually a requirement, an assumption we need to make about
-   sane tzinfo classes.
-
-3. The naive UTC time corresponding to x is x.n - x.o.
-   This is again a requirement for a sane tzinfo class.
-
-4. (x+k).s = x.s
-   This follows from #2, and that datimetimetz+timedelta preserves tzinfo.
-
-5. (x+k).n = x.n + k
-   Again follows from how arithmetic is defined.
-
-Now we can explain tz.fromutc(x).  Let's assume it's an interesting case
-(meaning that the various tzinfo methods exist, and don't blow up or return
-None when called).
-
-The function wants to return a datetime y with timezone tz, equivalent to x.
-x is already in UTC.
-
-By #3, we want
-
-    y.n - y.o = x.n                             [1]
-
-The algorithm starts by attaching tz to x.n, and calling that y.  So
-x.n = y.n at the start.  Then it wants to add a duration k to y, so that [1]
-becomes true; in effect, we want to solve [2] for k:
-
-   (y+k).n - (y+k).o = x.n                      [2]
-
-By #1, this is the same as
-
-   (y+k).n - ((y+k).s + (y+k).d) = x.n          [3]
-
-By #5, (y+k).n = y.n + k, which equals x.n + k because x.n=y.n at the start.
-Substituting that into [3],
-
-   x.n + k - (y+k).s - (y+k).d = x.n; the x.n terms cancel, leaving
-   k - (y+k).s - (y+k).d = 0; rearranging,
-   k = (y+k).s - (y+k).d; by #4, (y+k).s == y.s, so
-   k = y.s - (y+k).d
-
-On the RHS, (y+k).d can't be computed directly, but y.s can be, and we
-approximate k by ignoring the (y+k).d term at first.  Note that k can't be
-very large, since all offset-returning methods return a duration of magnitude
-less than 24 hours.  For that reason, if y is firmly in std time, (y+k).d must
-be 0, so ignoring it has no consequence then.
-
-In any case, the new value is
-
-    z = y + y.s                                 [4]
-
-It's helpful to step back at look at [4] from a higher level:  it's simply
-mapping from UTC to tz's standard time.
-
-At this point, if
-
-    z.n - z.o = x.n                             [5]
-
-we have an equivalent time, and are almost done.  The insecurity here is
-at the start of daylight time.  Picture US Eastern for concreteness.  The wall
-time jumps from 1:59 to 3:00, and wall hours of the form 2:MM don't make good
-sense then.  The docs ask that an Eastern tzinfo class consider such a time to
-be EDT (because it's "after 2"), which is a redundant spelling of 1:MM EST
-on the day DST starts.  We want to return the 1:MM EST spelling because that's
-the only spelling that makes sense on the local wall clock.
-
-In fact, if [5] holds at this point, we do have the standard-time spelling,
-but that takes a bit of proof.  We first prove a stronger result.  What's the
-difference between the LHS and RHS of [5]?  Let
-
-    diff = x.n - (z.n - z.o)                    [6]
-
-Now
-    z.n =                       by [4]
-    (y + y.s).n =               by #5
-    y.n + y.s =                 since y.n = x.n
-    x.n + y.s =                 since z and y are have the same tzinfo member,
-                                    y.s = z.s by #2
-    x.n + z.s
-
-Plugging that back into [6] gives
-
-    diff =
-    x.n - ((x.n + z.s) - z.o) =     expanding
-    x.n - x.n - z.s + z.o =         cancelling
-    - z.s + z.o =                   by #2
-    z.d
-
-So diff = z.d.
-
-If [5] is true now, diff = 0, so z.d = 0 too, and we have the standard-time
-spelling we wanted in the endcase described above.  We're done.  Contrarily,
-if z.d = 0, then we have a UTC equivalent, and are also done.
-
-If [5] is not true now, diff = z.d != 0, and z.d is the offset we need to
-add to z (in effect, z is in tz's standard time, and we need to shift the
-local clock into tz's daylight time).
-
-Let
-
-    z' = z + z.d = z + diff                     [7]
-
-and we can again ask whether
-
-    z'.n - z'.o = x.n                           [8]
-
-If so, we're done.  If not, the tzinfo class is insane, according to the
-assumptions we've made.  This also requires a bit of proof.  As before, let's
-compute the difference between the LHS and RHS of [8] (and skipping some of
-the justifications for the kinds of substitutions we've done several times
-already):
-
-    diff' = x.n - (z'.n - z'.o) =           replacing z'.n via [7]
-        x.n  - (z.n + diff - z'.o) =    replacing diff via [6]
-        x.n - (z.n + x.n - (z.n - z.o) - z'.o) =
-        x.n - z.n - x.n + z.n - z.o + z'.o =    cancel x.n
-        - z.n + z.n - z.o + z'.o =              cancel z.n
-        - z.o + z'.o =                      #1 twice
-        -z.s - z.d + z'.s + z'.d =          z and z' have same tzinfo
-        z'.d - z.d
-
-So z' is UTC-equivalent to x iff z'.d = z.d at this point.  If they are equal,
-we've found the UTC-equivalent so are done.  In fact, we stop with [7] and
-return z', not bothering to compute z'.d.
-
-How could z.d and z'd differ?  z' = z + z.d [7], so merely moving z' by
-a dst() offset, and starting *from* a time already in DST (we know z.d != 0),
-would have to change the result dst() returns:  we start in DST, and moving
-a little further into it takes us out of DST.
-
-There isn't a sane case where this can happen.  The closest it gets is at
-the end of DST, where there's an hour in UTC with no spelling in a hybrid
-tzinfo class.  In US Eastern, that's 5:MM UTC = 0:MM EST = 1:MM EDT.  During
-that hour, on an Eastern clock 1:MM is taken as being in standard time (6:MM
-UTC) because the docs insist on that, but 0:MM is taken as being in daylight
-time (4:MM UTC).  There is no local time mapping to 5:MM UTC.  The local
-clock jumps from 1:59 back to 1:00 again, and repeats the 1:MM hour in
-standard time.  Since that's what the local clock *does*, we want to map both
-UTC hours 5:MM and 6:MM to 1:MM Eastern.  The result is ambiguous
-in local time, but so it goes -- it's the way the local clock works.
-
-When x = 5:MM UTC is the input to this algorithm, x.o=0, y.o=-5 and y.d=0,
-so z=0:MM.  z.d=60 (minutes) then, so [5] doesn't hold and we keep going.
-z' = z + z.d = 1:MM then, and z'.d=0, and z'.d - z.d = -60 != 0 so [8]
-(correctly) concludes that z' is not UTC-equivalent to x.
-
-Because we know z.d said z was in daylight time (else [5] would have held and
-we would have stopped then), and we know z.d != z'.d (else [8] would have held
-and we would have stopped then), and there are only 2 possible values dst() can
-return in Eastern, it follows that z'.d must be 0 (which it is in the example,
-but the reasoning doesn't depend on the example -- it depends on there being
-two possible dst() outcomes, one zero and the other non-zero).  Therefore
-z' must be in standard time, and is the spelling we want in this case.
-
-Note again that z' is not UTC-equivalent as far as the hybrid tzinfo class is
-concerned (because it takes z' as being in standard time rather than the
-daylight time we intend here), but returning it gives the real-life "local
-clock repeats an hour" behavior when mapping the "unspellable" UTC hour into
-tz.
-
-When the input is 6:MM, z=1:MM and z.d=0, and we stop at once, again with
-the 1:MM standard time spelling we want.
-
-So how can this break?  One of the assumptions must be violated.  Two
-possibilities:
-
-1) [2] effectively says that y.s is invariant across all y belong to a given
-   time zone.  This isn't true if, for political reasons or continental drift,
-   a region decides to change its base offset from UTC.
-
-2) There may be versions of "double daylight" time where the tail end of
-   the analysis gives up a step too early.  I haven't thought about that
-   enough to say.
-
-In any case, it's clear that the default fromutc() is strong enough to handle
-"almost all" time zones:  so long as the standard offset is invariant, it
-doesn't matter if daylight time transition points change from year to year, or
-if daylight time is skipped in some years; it doesn't matter how large or
-small dst() may get within its bounds; and it doesn't even matter if some
-perverse time zone returns a negative dst()).  So a breaking case must be
-pretty bizarre, and a tzinfo subclass can override fromutc() if it is.
---------------------------------------------------------------------------- */

Modified: python/branches/py3k/PC/config.c
==============================================================================
--- python/branches/py3k/PC/config.c	(original)
+++ python/branches/py3k/PC/config.c	Fri Jul 23 21:25:47 2010
@@ -43,7 +43,7 @@
 extern PyObject* PyInit_parser(void);
 extern PyObject* PyInit_winreg(void);
 extern PyObject* PyInit__struct(void);
-extern PyObject* PyInit_datetime(void);
+extern PyObject* PyInit__datetime(void);
 extern PyObject* PyInit__functools(void);
 extern PyObject* PyInit__json(void);
 extern PyObject* PyInit_zlib(void);
@@ -116,7 +116,7 @@
     {"parser", PyInit_parser},
     {"winreg", PyInit_winreg},
     {"_struct", PyInit__struct},
-    {"datetime", PyInit_datetime},
+    {"_datetime", PyInit__datetime},
     {"_functools", PyInit__functools},
     {"_json", PyInit__json},
 

Modified: python/branches/py3k/PCbuild/pythoncore.vcproj
==============================================================================
--- python/branches/py3k/PCbuild/pythoncore.vcproj	(original)
+++ python/branches/py3k/PCbuild/pythoncore.vcproj	Fri Jul 23 21:25:47 2010
@@ -1068,7 +1068,7 @@
 				>
 			</File>
 			<File
-				RelativePath="..\Modules\datetimemodule.c"
+				RelativePath="..\Modules\_datetimemodule.c"
 				>
 			</File>
 			<File

Modified: python/branches/py3k/setup.py
==============================================================================
--- python/branches/py3k/setup.py	(original)
+++ python/branches/py3k/setup.py	Fri Jul 23 21:25:47 2010
@@ -452,7 +452,7 @@
         # time operations and variables
         exts.append( Extension('time', ['timemodule.c', '_time.c'],
                                libraries=math_libs) )
-        exts.append( Extension('datetime', ['datetimemodule.c', '_time.c'],
+        exts.append( Extension('_datetime', ['_datetimemodule.c', '_time.c'],
                                libraries=math_libs) )
         # fast iterator tools implemented in C
         exts.append( Extension("itertools", ["itertoolsmodule.c"]) )

From python-checkins at python.org  Fri Jul 23 21:39:30 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Fri, 23 Jul 2010 21:39:30 +0200 (CEST)
Subject: [Python-checkins] r83113 - python/branches/release31-maint
Message-ID: <20100723193930.051A9EE982@mail.python.org>

Author: alexander.belopolsky
Date: Fri Jul 23 21:39:29 2010
New Revision: 83113

Log:
Blocked revisions 83112 via svnmerge

........
  r83112 | alexander.belopolsky | 2010-07-23 15:25:47 -0400 (Fri, 23 Jul 2010) | 2 lines
  
  Issue #7989: Added pure python implementation of the datetime module.
........


Modified:
   python/branches/release31-maint/   (props changed)

From python-checkins at python.org  Fri Jul 23 21:44:14 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Fri, 23 Jul 2010 21:44:14 +0200 (CEST)
Subject: [Python-checkins] r83114 - python/branches/release27-maint
Message-ID: <20100723194414.78049EE982@mail.python.org>

Author: alexander.belopolsky
Date: Fri Jul 23 21:44:14 2010
New Revision: 83114

Log:
Blocked revisions 83112 via svnmerge

........
  r83112 | alexander.belopolsky | 2010-07-23 15:25:47 -0400 (Fri, 23 Jul 2010) | 2 lines
  
  Issue #7989: Added pure python implementation of the datetime module.
........


Modified:
   python/branches/release27-maint/   (props changed)

From python-checkins at python.org  Fri Jul 23 22:03:53 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Fri, 23 Jul 2010 22:03:53 +0200 (CEST)
Subject: [Python-checkins] r83115 - python/branches/py3k/Lib/datetime.py
Message-ID: <20100723200353.EEAFDEE99B@mail.python.org>

Author: alexander.belopolsky
Date: Fri Jul 23 22:03:53 2010
New Revision: 83115

Log:
Use _datetime docstring if _datetime is available.

Modified:
   python/branches/py3k/Lib/datetime.py

Modified: python/branches/py3k/Lib/datetime.py
==============================================================================
--- python/branches/py3k/Lib/datetime.py	(original)
+++ python/branches/py3k/Lib/datetime.py	Fri Jul 23 22:03:53 2010
@@ -2085,3 +2085,8 @@
          _days_before_year, _days_in_month, _format_time, _is_leap,
          _isoweek1monday, _math, _ord2ymd, _time, _time_class, _tzinfo_class,
          _wrap_strftime, _ymd2ord)
+    # XXX Since import * above excludes names that start with _,
+    # docstring does not get overwritten. In the future, it may be
+    # appropriate to maintain a single module level docstring and
+    # remove the following line.
+    from _datetime import __doc__

From python-checkins at python.org  Sat Jul 24 02:49:21 2010
From: python-checkins at python.org (victor.stinner)
Date: Sat, 24 Jul 2010 02:49:21 +0200 (CEST)
Subject: [Python-checkins] r83116 - in python/branches/py3k: Lib/getopt.py
	Lib/test/test_getopt.py Misc/NEWS
Message-ID: <20100724004921.019C6EE9D8@mail.python.org>

Author: victor.stinner
Date: Sat Jul 24 02:49:20 2010
New Revision: 83116

Log:
Issue #4629: getopt raises an error if an argument ends with = whereas getopt
doesn't except a value (eg. --help= is rejected if getopt uses ['help='] long
options).


Modified:
   python/branches/py3k/Lib/getopt.py
   python/branches/py3k/Lib/test/test_getopt.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/getopt.py
==============================================================================
--- python/branches/py3k/Lib/getopt.py	(original)
+++ python/branches/py3k/Lib/getopt.py	Sat Jul 24 02:49:20 2010
@@ -156,7 +156,7 @@
             if not args:
                 raise GetoptError('option --%s requires argument' % opt, opt)
             optarg, args = args[0], args[1:]
-    elif optarg:
+    elif optarg is not None:
         raise GetoptError('option --%s must not have an argument' % opt, opt)
     opts.append(('--' + opt, optarg or ''))
     return opts, args

Modified: python/branches/py3k/Lib/test/test_getopt.py
==============================================================================
--- python/branches/py3k/Lib/test/test_getopt.py	(original)
+++ python/branches/py3k/Lib/test/test_getopt.py	Sat Jul 24 02:49:20 2010
@@ -173,6 +173,12 @@
         m = types.ModuleType("libreftest", s)
         run_doctest(m, verbose)
 
+    def test_issue4629(self):
+        longopts, shortopts = getopt.getopt(['--help='], '', ['help='])
+        self.assertEquals(longopts, [('--help', '')])
+        longopts, shortopts = getopt.getopt(['--help=x'], '', ['help='])
+        self.assertEquals(longopts, [('--help', 'x')])
+        self.assertRaises(getopt.GetoptError, getopt.getopt, ['--help='], '', ['help'])
 
 def test_main():
     run_unittest(GetoptTests)

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sat Jul 24 02:49:20 2010
@@ -473,6 +473,10 @@
 Library
 -------
 
+- Issue #4629: getopt raises an error if an argument ends with = whereas getopt
+  doesn't except a value (eg. --help= is rejected if getopt uses ['help='] long
+  options).
+
 - Issue #7989: Added pure python implementation of the `datetime`
   module.  The C module is renamed to `_datetime` and if available,
   overrides all classes defined in datetime with fast C impementation.

From python-checkins at python.org  Sat Jul 24 03:07:52 2010
From: python-checkins at python.org (victor.stinner)
Date: Sat, 24 Jul 2010 03:07:52 +0200 (CEST)
Subject: [Python-checkins] r83117 - in python/branches/release27-maint:
	Lib/getopt.py Lib/test/test_getopt.py Misc/NEWS
Message-ID: <20100724010752.5F10AEEA8F@mail.python.org>

Author: victor.stinner
Date: Sat Jul 24 03:07:52 2010
New Revision: 83117

Log:
Merged revisions 83116 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83116 | victor.stinner | 2010-07-24 02:49:20 +0200 (sam., 24 juil. 2010) | 4 lines
  
  Issue #4629: getopt raises an error if an argument ends with = whereas getopt
  doesn't except a value (eg. --help= is rejected if getopt uses ['help='] long
  options).
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/getopt.py
   python/branches/release27-maint/Lib/test/test_getopt.py
   python/branches/release27-maint/Misc/NEWS

Modified: python/branches/release27-maint/Lib/getopt.py
==============================================================================
--- python/branches/release27-maint/Lib/getopt.py	(original)
+++ python/branches/release27-maint/Lib/getopt.py	Sat Jul 24 03:07:52 2010
@@ -155,7 +155,7 @@
             if not args:
                 raise GetoptError('option --%s requires argument' % opt, opt)
             optarg, args = args[0], args[1:]
-    elif optarg:
+    elif optarg is not None:
         raise GetoptError('option --%s must not have an argument' % opt, opt)
     opts.append(('--' + opt, optarg or ''))
     return opts, args

Modified: python/branches/release27-maint/Lib/test/test_getopt.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_getopt.py	(original)
+++ python/branches/release27-maint/Lib/test/test_getopt.py	Sat Jul 24 03:07:52 2010
@@ -173,6 +173,12 @@
         m = types.ModuleType("libreftest", s)
         run_doctest(m, verbose)
 
+    def test_issue4629(self):
+        longopts, shortopts = getopt.getopt(['--help='], '', ['help='])
+        self.assertEquals(longopts, [('--help', '')])
+        longopts, shortopts = getopt.getopt(['--help=x'], '', ['help='])
+        self.assertEquals(longopts, [('--help', 'x')])
+        self.assertRaises(getopt.GetoptError, getopt.getopt, ['--help='], '', ['help'])
 
 def test_main():
     run_unittest(GetoptTests)

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Sat Jul 24 03:07:52 2010
@@ -18,6 +18,10 @@
 Library
 -------
 
+- Issue #4629: getopt raises an error if an argument ends with = whereas getopt
+  doesn't except a value (eg. --help= is rejected if getopt uses ['help='] long
+  options).
+
 - Issue #7895: platform.mac_ver() no longer crashes after calling os.fork()
 
 - Issue #5395: array.fromfile() would raise a spurious EOFError when an

From python-checkins at python.org  Sat Jul 24 03:11:58 2010
From: python-checkins at python.org (victor.stinner)
Date: Sat, 24 Jul 2010 03:11:58 +0200 (CEST)
Subject: [Python-checkins] r83118 - in python/branches/release31-maint:
	Lib/getopt.py Lib/test/test_getopt.py Misc/NEWS
Message-ID: <20100724011158.B0987EEA7B@mail.python.org>

Author: victor.stinner
Date: Sat Jul 24 03:11:58 2010
New Revision: 83118

Log:
Merged revisions 83116 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83116 | victor.stinner | 2010-07-24 02:49:20 +0200 (sam., 24 juil. 2010) | 4 lines
  
  Issue #4629: getopt raises an error if an argument ends with = whereas getopt
  doesn't except a value (eg. --help= is rejected if getopt uses ['help='] long
  options).
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/getopt.py
   python/branches/release31-maint/Lib/test/test_getopt.py
   python/branches/release31-maint/Misc/NEWS

Modified: python/branches/release31-maint/Lib/getopt.py
==============================================================================
--- python/branches/release31-maint/Lib/getopt.py	(original)
+++ python/branches/release31-maint/Lib/getopt.py	Sat Jul 24 03:11:58 2010
@@ -156,7 +156,7 @@
             if not args:
                 raise GetoptError('option --%s requires argument' % opt, opt)
             optarg, args = args[0], args[1:]
-    elif optarg:
+    elif optarg is not None:
         raise GetoptError('option --%s must not have an argument' % opt, opt)
     opts.append(('--' + opt, optarg or ''))
     return opts, args

Modified: python/branches/release31-maint/Lib/test/test_getopt.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_getopt.py	(original)
+++ python/branches/release31-maint/Lib/test/test_getopt.py	Sat Jul 24 03:11:58 2010
@@ -174,6 +174,12 @@
         m = types.ModuleType("libreftest", s)
         run_doctest(m, verbose)
 
+    def test_issue4629(self):
+        longopts, shortopts = getopt.getopt(['--help='], '', ['help='])
+        self.assertEquals(longopts, [('--help', '')])
+        longopts, shortopts = getopt.getopt(['--help=x'], '', ['help='])
+        self.assertEquals(longopts, [('--help', 'x')])
+        self.assertRaises(getopt.GetoptError, getopt.getopt, ['--help='], '', ['help'])
 
 def test_main():
     run_unittest(GetoptTests)

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Sat Jul 24 03:11:58 2010
@@ -78,6 +78,10 @@
 Library
 -------
 
+- Issue #4629: getopt raises an error if an argument ends with = whereas getopt
+  doesn't except a value (eg. --help= is rejected if getopt uses ['help='] long
+  options).
+
 - Issue #7895: platform.mac_ver() no longer crashes after calling os.fork()
 
 - Issue #9323: Fixed a bug in trace.py that resulted in loosing the

From python-checkins at python.org  Sat Jul 24 03:12:54 2010
From: python-checkins at python.org (victor.stinner)
Date: Sat, 24 Jul 2010 03:12:54 +0200 (CEST)
Subject: [Python-checkins] r83119 - in python/branches/release26-maint:
	Lib/getopt.py Lib/test/test_getopt.py Misc/NEWS
Message-ID: <20100724011254.EE070F081@mail.python.org>

Author: victor.stinner
Date: Sat Jul 24 03:12:54 2010
New Revision: 83119

Log:
Merged revisions 83117 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

................
  r83117 | victor.stinner | 2010-07-24 03:07:52 +0200 (sam., 24 juil. 2010) | 11 lines
  
  Merged revisions 83116 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k
  
  ........
    r83116 | victor.stinner | 2010-07-24 02:49:20 +0200 (sam., 24 juil. 2010) | 4 lines
    
    Issue #4629: getopt raises an error if an argument ends with = whereas getopt
    doesn't except a value (eg. --help= is rejected if getopt uses ['help='] long
    options).
  ........
................


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Lib/getopt.py
   python/branches/release26-maint/Lib/test/test_getopt.py
   python/branches/release26-maint/Misc/NEWS

Modified: python/branches/release26-maint/Lib/getopt.py
==============================================================================
--- python/branches/release26-maint/Lib/getopt.py	(original)
+++ python/branches/release26-maint/Lib/getopt.py	Sat Jul 24 03:12:54 2010
@@ -156,7 +156,7 @@
             if not args:
                 raise GetoptError('option --%s requires argument' % opt, opt)
             optarg, args = args[0], args[1:]
-    elif optarg:
+    elif optarg is not None:
         raise GetoptError('option --%s must not have an argument' % opt, opt)
     opts.append(('--' + opt, optarg or ''))
     return opts, args

Modified: python/branches/release26-maint/Lib/test/test_getopt.py
==============================================================================
--- python/branches/release26-maint/Lib/test/test_getopt.py	(original)
+++ python/branches/release26-maint/Lib/test/test_getopt.py	Sat Jul 24 03:12:54 2010
@@ -171,6 +171,12 @@
         m = types.ModuleType("libreftest", s)
         run_doctest(m, verbose)
 
+    def test_issue4629(self):
+        longopts, shortopts = getopt.getopt(['--help='], '', ['help='])
+        self.assertEquals(longopts, [('--help', '')])
+        longopts, shortopts = getopt.getopt(['--help=x'], '', ['help='])
+        self.assertEquals(longopts, [('--help', 'x')])
+        self.assertRaises(getopt.GetoptError, getopt.getopt, ['--help='], '', ['help'])
 
 def test_main():
     run_unittest(GetoptTests)

Modified: python/branches/release26-maint/Misc/NEWS
==============================================================================
--- python/branches/release26-maint/Misc/NEWS	(original)
+++ python/branches/release26-maint/Misc/NEWS	Sat Jul 24 03:12:54 2010
@@ -81,6 +81,10 @@
 Library
 -------
 
+- Issue #4629: getopt raises an error if an argument ends with = whereas getopt
+  doesn't except a value (eg. --help= is rejected if getopt uses ['help='] long
+  options).
+
 - Issue #7895: platform.mac_ver() no longer crashes after calling os.fork()
 
 - Issue #5395: array.fromfile() would raise a spurious EOFError when an

From python-checkins at python.org  Sat Jul 24 04:24:56 2010
From: python-checkins at python.org (victor.stinner)
Date: Sat, 24 Jul 2010 04:24:56 +0200 (CEST)
Subject: [Python-checkins] r83120 - in python/branches/py3k:
	Lib/http/client.py Lib/xmlrpc/client.py Misc/NEWS
Message-ID: <20100724022456.25F88EEA6D@mail.python.org>

Author: victor.stinner
Date: Sat Jul 24 04:24:55 2010
New Revision: 83120

Log:
#9032: XML-RPC client: Transport.request() retries on EPIPE error

The EPIPE error occurs when the server closes the socket and the client sends a
"big" XML-RPC request (I don't know exactly the size threshold).

request() just have to ignore the error because single_request() closes the
socket on error, and so the next call to single_request() will open a new
socket.

Remove also a comment in the HTTP client because it's now wrong: see r70643
and issue #5542.


Modified:
   python/branches/py3k/Lib/http/client.py
   python/branches/py3k/Lib/xmlrpc/client.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/http/client.py
==============================================================================
--- python/branches/py3k/Lib/http/client.py	(original)
+++ python/branches/py3k/Lib/http/client.py	Sat Jul 24 04:24:55 2010
@@ -734,11 +734,6 @@
             else:
                 raise NotConnected()
 
-        # send the data to the server. if we get a broken pipe, then close
-        # the socket. we want to reconnect when somebody tries to send again.
-        #
-        # NOTE: we DO propagate the error, though, because we cannot simply
-        #       ignore the error... the caller will know if they can retry.
         if self.debuglevel > 0:
             print("send:", repr(str))
         blocksize = 8192

Modified: python/branches/py3k/Lib/xmlrpc/client.py
==============================================================================
--- python/branches/py3k/Lib/xmlrpc/client.py	(original)
+++ python/branches/py3k/Lib/xmlrpc/client.py	Sat Jul 24 04:24:55 2010
@@ -1135,7 +1135,7 @@
             try:
                 return self.single_request(host, handler, request_body, verbose)
             except socket.error as e:
-                if i or e.errno not in (errno.ECONNRESET, errno.ECONNABORTED):
+                if i or e.errno not in (errno.ECONNRESET, errno.ECONNABORTED, errno.EPIPE):
                     raise
             except http.client.BadStatusLine: #close after we sent request
                 if i:

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sat Jul 24 04:24:55 2010
@@ -473,6 +473,10 @@
 Library
 -------
 
+- Issue #9032: XML-RPC client retries the request on EPIPE error. The EPIPE
+  error occurs when the server closes the socket and the client sends a big
+  XML-RPC request.
+
 - Issue #4629: getopt raises an error if an argument ends with = whereas getopt
   doesn't except a value (eg. --help= is rejected if getopt uses ['help='] long
   options).

From python-checkins at python.org  Sat Jul 24 04:28:08 2010
From: python-checkins at python.org (victor.stinner)
Date: Sat, 24 Jul 2010 04:28:08 +0200 (CEST)
Subject: [Python-checkins] r83121 - python/branches/release31-maint
Message-ID: <20100724022808.C09E5EEA2A@mail.python.org>

Author: victor.stinner
Date: Sat Jul 24 04:28:08 2010
New Revision: 83121

Log:
Blocked revisions 83120 via svnmerge

........
  r83120 | victor.stinner | 2010-07-24 04:24:55 +0200 (sam., 24 juil. 2010) | 12 lines

  #9032: XML-RPC client: Transport.request() retries on EPIPE error

  The EPIPE error occurs when the server closes the socket and the client sends a
  "big" XML-RPC request (I don't know exactly the size threshold).

  request() just have to ignore the error because single_request() closes the
  socket on error, and so the next call to single_request() will open a new
  socket.

  Remove also a comment in the HTTP client because it's now wrong: see r70643
  and issue #5542.
........

In Python 3.1, Transport.request() doesn't retry on socket error.


Modified:
   python/branches/release31-maint/   (props changed)

From python-checkins at python.org  Sat Jul 24 04:46:16 2010
From: python-checkins at python.org (victor.stinner)
Date: Sat, 24 Jul 2010 04:46:16 +0200 (CEST)
Subject: [Python-checkins] r83122 - in python/branches/release27-maint:
	Lib/httplib.py Lib/test/test_httplib.py Misc/NEWS
Message-ID: <20100724024617.002A5EEA87@mail.python.org>

Author: victor.stinner
Date: Sat Jul 24 04:46:16 2010
New Revision: 83122

Log:
Forward port r70643 (#5542) + part of r83120 (just remove the comment)

Remove special logic that closes HTTPConnection socket on EPIPE.

If the socket is closed, the client has no chance to read the response
from the server.  EPIPE means that it isn't possible to write more
data from the socket, but not that it is impossible to read.


Modified:
   python/branches/release27-maint/Lib/httplib.py
   python/branches/release27-maint/Lib/test/test_httplib.py
   python/branches/release27-maint/Misc/NEWS

Modified: python/branches/release27-maint/Lib/httplib.py
==============================================================================
--- python/branches/release27-maint/Lib/httplib.py	(original)
+++ python/branches/release27-maint/Lib/httplib.py	Sat Jul 24 04:46:16 2010
@@ -67,6 +67,7 @@
 """
 
 from array import array
+import os
 import socket
 from sys import py3kwarning
 from urlparse import urlsplit
@@ -756,27 +757,17 @@
             else:
                 raise NotConnected()
 
-        # send the data to the server. if we get a broken pipe, then close
-        # the socket. we want to reconnect when somebody tries to send again.
-        #
-        # NOTE: we DO propagate the error, though, because we cannot simply
-        #       ignore the error... the caller will know if they can retry.
         if self.debuglevel > 0:
             print "send:", repr(str)
-        try:
-            blocksize=8192
-            if hasattr(str,'read') and not isinstance(str, array):
-                if self.debuglevel > 0: print "sendIng a read()able"
-                data=str.read(blocksize)
-                while data:
-                    self.sock.sendall(data)
-                    data=str.read(blocksize)
-            else:
-                self.sock.sendall(str)
-        except socket.error, v:
-            if v.args[0] == 32:      # Broken pipe
-                self.close()
-            raise
+        blocksize = 8192
+        if hasattr(str,'read') and not isinstance(str, array):
+            if self.debuglevel > 0: print "sendIng a read()able"
+            data = str.read(blocksize)
+            while data:
+                self.sock.sendall(data)
+                data = str.read(blocksize)
+        else:
+            self.sock.sendall(str)
 
     def _output(self, s):
         """Add a line of output to the current request buffer.
@@ -941,15 +932,7 @@
 
     def request(self, method, url, body=None, headers={}):
         """Send a complete request to the server."""
-
-        try:
-            self._send_request(method, url, body, headers)
-        except socket.error, v:
-            # trap 'Broken pipe' if we're allowed to automatically reconnect
-            if v.args[0] != 32 or not self.auto_open:
-                raise
-            # try one more time
-            self._send_request(method, url, body, headers)
+        self._send_request(method, url, body, headers)
 
     def _set_content_length(self, body):
         # Set the content-length based on the body.
@@ -959,7 +942,6 @@
         except TypeError, te:
             # If this is a file-like object, try to
             # fstat its file descriptor
-            import os
             try:
                 thelen = str(os.fstat(body.fileno()).st_size)
             except (AttributeError, OSError):
@@ -970,7 +952,7 @@
             self.putheader('Content-Length', thelen)
 
     def _send_request(self, method, url, body, headers):
-        # honour explicitly requested Host: and Accept-Encoding headers
+        # Honor explicitly requested Host: and Accept-Encoding: headers.
         header_names = dict.fromkeys([k.lower() for k in headers])
         skips = {}
         if 'host' in header_names:

Modified: python/branches/release27-maint/Lib/test/test_httplib.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_httplib.py	(original)
+++ python/branches/release27-maint/Lib/test/test_httplib.py	Sat Jul 24 04:46:16 2010
@@ -2,6 +2,7 @@
 import httplib
 import StringIO
 import socket
+import errno
 
 import unittest
 TestCase = unittest.TestCase
@@ -24,6 +25,21 @@
             raise httplib.UnimplementedFileMode()
         return self.fileclass(self.text)
 
+class EPipeSocket(FakeSocket):
+
+    def __init__(self, text, pipe_trigger):
+        # When sendall() is called with pipe_trigger, raise EPIPE.
+        FakeSocket.__init__(self, text)
+        self.pipe_trigger = pipe_trigger
+
+    def sendall(self, data):
+        if self.pipe_trigger in data:
+            raise socket.error(errno.EPIPE, "gotcha")
+        self.data += data
+
+    def close(self):
+        pass
+
 class NoEOFStringIO(StringIO.StringIO):
     """Like StringIO, but raises AssertionError on EOF.
 
@@ -254,6 +270,21 @@
         finally:
             resp.close()
 
+    def test_epipe(self):
+        sock = EPipeSocket(
+            "HTTP/1.0 401 Authorization Required\r\n"
+            "Content-type: text/html\r\n"
+            "WWW-Authenticate: Basic realm=\"example\"\r\n",
+            b"Content-Length")
+        conn = httplib.HTTPConnection("example.com")
+        conn.sock = sock
+        self.assertRaises(socket.error,
+                          lambda: conn.request("PUT", "/url", "body"))
+        resp = conn.getresponse()
+        self.assertEqual(401, resp.status)
+        self.assertEqual("Basic realm=\"example\"",
+                         resp.getheader("www-authenticate"))
+
 
 class OfflineTest(TestCase):
     def test_responses(self):

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Sat Jul 24 04:46:16 2010
@@ -18,6 +18,8 @@
 Library
 -------
 
+- Issue #5542: Remove special logic that closes HTTPConnection socket on EPIPE.
+
 - Issue #4629: getopt raises an error if an argument ends with = whereas getopt
   doesn't except a value (eg. --help= is rejected if getopt uses ['help='] long
   options).

From python-checkins at python.org  Sat Jul 24 04:51:49 2010
From: python-checkins at python.org (victor.stinner)
Date: Sat, 24 Jul 2010 04:51:49 +0200 (CEST)
Subject: [Python-checkins] r83123 - in python/branches/release27-maint:
	Lib/xmlrpclib.py Misc/NEWS
Message-ID: <20100724025149.AF610EEA67@mail.python.org>

Author: victor.stinner
Date: Sat Jul 24 04:51:49 2010
New Revision: 83123

Log:
Backport of 83120 (#9032)

XML-RPC client: Transport.request() retries on EPIPE error

The EPIPE error occurs when the server closes the socket and the client sends a
"big" XML-RPC request (I don't know exactly the size threshold).

request() just have to ignore the error because single_request() closes the
socket on error, and so the next call to single_request() will open a new
socket.

Remove also a comment in the HTTP client because it's now wrong: see r70643 and
issue #5542.


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/xmlrpclib.py
   python/branches/release27-maint/Misc/NEWS

Modified: python/branches/release27-maint/Lib/xmlrpclib.py
==============================================================================
--- python/branches/release27-maint/Lib/xmlrpclib.py	(original)
+++ python/branches/release27-maint/Lib/xmlrpclib.py	Sat Jul 24 04:51:49 2010
@@ -1263,7 +1263,7 @@
             try:
                 return self.single_request(host, handler, request_body, verbose)
             except socket.error, e:
-                if i or e.errno not in (errno.ECONNRESET, errno.ECONNABORTED):
+                if i or e.errno not in (errno.ECONNRESET, errno.ECONNABORTED, errno.EPIPE):
                     raise
             except httplib.BadStatusLine: #close after we sent request
                 if i:

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Sat Jul 24 04:51:49 2010
@@ -18,6 +18,10 @@
 Library
 -------
 
+- Issue #9032: XML-RPC client retries the request on EPIPE error. The EPIPE
+  error occurs when the server closes the socket and the client sends a big
+  XML-RPC request.
+
 - Issue #5542: Remove special logic that closes HTTPConnection socket on EPIPE.
 
 - Issue #4629: getopt raises an error if an argument ends with = whereas getopt

From python-checkins at python.org  Sat Jul 24 11:46:42 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Sat, 24 Jul 2010 11:46:42 +0200 (CEST)
Subject: [Python-checkins] r83124 - in python/branches/release27-maint:
	Lib/test/test_posix.py Misc/NEWS Modules/posixmodule.c
Message-ID: <20100724094642.0C2DFEE990@mail.python.org>

Author: ronald.oussoren
Date: Sat Jul 24 11:46:41 2010
New Revision: 83124

Log:
Merged revisions 83088 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83088 | ronald.oussoren | 2010-07-23 14:53:51 +0100 (Fri, 23 Jul 2010) | 8 lines
  
  This fixes issue7900 by adding code that deals
  with the fact that getgroups(2) might return
  more that MAX_GROUPS on OSX.
  
  See the issue (and python-dev archives) for the
  gory details. Summarized: OSX behaves rather oddly
  and Apple says this is intentional.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/test/test_posix.py
   python/branches/release27-maint/Misc/NEWS
   python/branches/release27-maint/Modules/posixmodule.c

Modified: python/branches/release27-maint/Lib/test/test_posix.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_posix.py	(original)
+++ python/branches/release27-maint/Lib/test/test_posix.py	Sat Jul 24 11:46:41 2010
@@ -6,6 +6,7 @@
 posix = test_support.import_module('posix')
 
 import errno
+import sys
 import time
 import os
 import pwd
@@ -363,9 +364,59 @@
                 os.chdir(curdir)
                 shutil.rmtree(base_path)
 
+    def test_getgroups(self):
+        with os.popen('id -G') as idg:
+            groups = idg.read().strip()
+
+        if not groups:
+            raise unittest.SkipTest("need working 'id -G'")
+
+        self.assertEqual([int(x) for x in groups.split()], posix.getgroups())
+
+class PosixGroupsTester(unittest.TestCase):
+
+    def setUp(self):
+        if posix.getuid() != 0:
+            raise unittest.SkipTest("not enough privileges")
+        if not hasattr(posix, 'getgroups'):
+            raise unittest.SkipTest("need posix.getgroups")
+        if sys.platform == 'darwin':
+            raise unittest.SkipTest("getgroups(2) is broken on OSX")
+        self.saved_groups = posix.getgroups()
+
+    def tearDown(self):
+        if hasattr(posix, 'setgroups'):
+            posix.setgroups(self.saved_groups)
+        elif hasattr(posix, 'initgroups'):
+            name = pwd.getpwuid(posix.getuid()).pw_name
+            posix.initgroups(name, self.saved_groups[0])
+
+    @unittest.skipUnless(hasattr(posix, 'initgroups'),
+                         "test needs posix.initgroups()")
+    def test_initgroups(self):
+        # find missing group
+
+        groups = sorted(self.saved_groups)
+        for g1,g2 in zip(groups[:-1], groups[1:]):
+            g = g1 + 1
+            if g < g2:
+                break
+        else:
+            g = g2 + 1
+        name = pwd.getpwuid(posix.getuid()).pw_name
+        posix.initgroups(name, g)
+        self.assertIn(g, posix.getgroups())
+
+    @unittest.skipUnless(hasattr(posix, 'setgroups'),
+                         "test needs posix.setgroups()")
+    def test_setgroups(self):
+        for groups in [[0], range(16)]:
+            posix.setgroups(groups)
+            self.assertListEqual(groups, posix.getgroups())
+
 
 def test_main():
-    test_support.run_unittest(PosixTester)
+    test_support.run_unittest(PosixTester, PosixGroupsTester)
 
 if __name__ == '__main__':
     test_main()

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Sat Jul 24 11:46:41 2010
@@ -56,6 +56,13 @@
 Extension Modules
 -----------------
 
+- Issue #7900: The getgroups(2) system call on MacOSX behaves rather oddly
+  compared to other unix systems. In particular, os.getgroups() does
+  not reflect any changes made using os.setgroups() but basicly always
+  returns the same information as the id command.
+
+  os.getgroups() can now return more than 16 groups on MacOSX.
+
 - Issue #9277: Fix bug in struct.pack for bools in standard mode
   (e.g., struct.pack('>?')):  if conversion to bool raised an exception
   then that exception wasn't properly propagated on machines where

Modified: python/branches/release27-maint/Modules/posixmodule.c
==============================================================================
--- python/branches/release27-maint/Modules/posixmodule.c	(original)
+++ python/branches/release27-maint/Modules/posixmodule.c	Sat Jul 24 11:46:41 2010
@@ -3877,17 +3877,49 @@
 #define MAX_GROUPS 64
 #endif
     gid_t grouplist[MAX_GROUPS];
+
+    /* On MacOSX getgroups(2) can return more than MAX_GROUPS results 
+     * This is a helper variable to store the intermediate result when
+     * that happens.
+     *
+     * To keep the code readable the OSX behaviour is unconditional,
+     * according to the POSIX spec this should be safe on all unix-y
+     * systems.
+     */
+    gid_t* alt_grouplist = grouplist;
     int n;
 
     n = getgroups(MAX_GROUPS, grouplist);
-    if (n < 0)
-        posix_error();
-    else {
-        result = PyList_New(n);
-        if (result != NULL) {
+    if (n < 0) {
+        if (errno == EINVAL) {
+            n = getgroups(0, NULL);
+            if (n == -1) {
+                return posix_error();
+            }
+            if (n == 0) {
+                /* Avoid malloc(0) */
+                alt_grouplist = grouplist;
+            } else {
+                alt_grouplist = PyMem_Malloc(n * sizeof(gid_t));
+                if (alt_grouplist == NULL) {
+                    errno = EINVAL;
+                    return posix_error();
+                }
+                n = getgroups(n, alt_grouplist);
+                if (n == -1) {
+                    PyMem_Free(alt_grouplist);
+                    return posix_error();
+                }
+            }
+        } else {
+            return posix_error();
+        }
+    }
+    result = PyList_New(n);
+    if (result != NULL) {
         int i;
         for (i = 0; i < n; ++i) {
-            PyObject *o = PyInt_FromLong((long)grouplist[i]);
+            PyObject *o = PyInt_FromLong((long)alt_grouplist[i]);
             if (o == NULL) {
             Py_DECREF(result);
             result = NULL;
@@ -3895,7 +3927,10 @@
             }
             PyList_SET_ITEM(result, i, o);
         }
-        }
+    }
+
+    if (alt_grouplist != grouplist) {
+        PyMem_Free(alt_grouplist);
     }
 
     return result;

From python-checkins at python.org  Sat Jul 24 11:51:40 2010
From: python-checkins at python.org (richard.jones)
Date: Sat, 24 Jul 2010 11:51:40 +0200 (CEST)
Subject: [Python-checkins] r83125 - in python/branches/py3k:
	Doc/library/smtpd.rst Lib/smtpd.py
Message-ID: <20100724095140.94B4EEE990@mail.python.org>

Author: richard.jones
Date: Sat Jul 24 11:51:40 2010
New Revision: 83125

Log:
Implementation for issue 4184

Changes the previously private attributes to make them public, increasing the potential for extending the library in user code. Backward-compatible and documented.






Modified:
   python/branches/py3k/Doc/library/smtpd.rst
   python/branches/py3k/Lib/smtpd.py

Modified: python/branches/py3k/Doc/library/smtpd.rst
==============================================================================
--- python/branches/py3k/Doc/library/smtpd.rst	(original)
+++ python/branches/py3k/Doc/library/smtpd.rst	Sat Jul 24 11:51:40 2010
@@ -10,10 +10,14 @@
 
 
 
-This module offers several classes to implement SMTP servers.  One is a generic
+This module offers several classes to implement SMTP (email) servers.
+
+Several server implementations are present; one is a generic
 do-nothing implementation, which can be overridden, while the other two offer
 specific mail-sending strategies.
 
+Additionally the SMTPChannel may be extended to implement very specific
+interaction behaviour with SMTP clients.
 
 SMTPServer Objects
 ------------------
@@ -26,7 +30,6 @@
    inherits from :class:`asyncore.dispatcher`, and so will insert itself into
    :mod:`asyncore`'s event loop on instantiation.
 
-
    .. method:: process_message(peer, mailfrom, rcpttos, data)
 
       Raise :exc:`NotImplementedError` exception. Override this in subclasses to
@@ -37,6 +40,11 @@
       containing the contents of the e-mail (which should be in :rfc:`2822`
       format).
 
+   .. attribute:: channel_class
+
+      Override this in subclasses to use a custom :class:`SMTPChannel` for
+      managing SMTP clients.
+
 
 DebuggingServer Objects
 -----------------------
@@ -71,3 +79,91 @@
    running this has a good chance to make you into an open relay, so please be
    careful.
 
+SMTPChannel Objects
+-------------------
+
+.. class:: SMTPChannel(server, conn, addr)
+
+   Create a new :class:`SMTPChannel` object which manages the communication
+   between the server and a single SMTP client.
+
+   To use a custom SMTPChannel implementation you need to override the
+   :attr:`SMTPServer.channel_class` of your :class:`SMTPServer`.
+
+   The :class:`SMTPChannel` has the following instance variables:
+
+   .. attribute:: smtp_server
+
+      Holds the :class:`SMTPServer` that spawned this channel.
+
+   .. attribute:: conn
+
+      Holds the socket object connecting to the client.
+
+   .. attribute:: addr
+
+      Holds the address of the client, the second value returned by
+      socket.accept()
+
+   .. attribute:: received_lines
+
+      Holds a list of the line strings (decoded using UTF-8) received from
+      the client. The lines have their "\r\n" line ending translated to "\n".
+
+   .. attribute:: smtp_state
+
+      Holds the current state of the channel. This will be either
+      :attr:`COMMAND` initially and then :attr:`DATA` after the client sends
+      a "DATA" line.
+
+   .. attribute:: seen_greeting
+
+      Holds a string containing the greeting sent by the client in its "HELO".
+
+   .. attribute:: mailfrom
+
+      Holds a string containing the address identified in the "MAIL FROM:" line
+      from the client.
+
+   .. attribute:: rcpttos
+
+      Holds a list of strings containing the addresses identified in the
+      "RCPT TO:" lines from the client.
+
+   .. attribute:: received_data
+
+      Holds a string containing all of the data sent by the client during the
+      DATA state, up to but not including the terminating "\r\n.\r\n".
+
+   .. attribute:: fqdn
+
+      Holds the fully-qualified domain name of the server as returned by
+      ``socket.getfqdn()``.
+
+   .. attribute:: peer
+
+      Holds the name of the client peer as returned by ``conn.getpeername()``
+      where ``conn`` is :attr:`conn`.
+
+   The :class:`SMTPChannel` operates by invoking methods named ``smtp_<command>``
+   upon reception of a command line from the client. Built into the base
+   :class:`SMTPChannel` class are methods for handling the following commands
+   (and responding to them appropriately):
+
+   ======== ===================================================================
+   Command  Action taken
+   ======== ===================================================================
+   HELO     Accepts the greeting from the client and stores it in
+            :attr:`seen_greeting`.
+   NOOP     Takes no action.
+   QUIT     Closes the connection cleanly.
+   MAIL     Accepts the "MAIL FROM:" syntax and stores the supplied address as
+            :attr:`mailfrom`.
+   RCPT     Accepts the "RCPT TO:" syntax and stores the supplied addresses in
+            the :attr:`rcpttos` list.
+   RSET     Resets the :attr:`mailfrom`, :attr:`rcpttos`, and
+            :attr:`received_data`, but not the greeting.
+   DATA     Sets the internal state to :attr:`DATA` and stores remaining lines
+            from the client in :attr:`received_data` until the terminator
+            "\r\n.\r\n" is received.
+   ======== ===================================================================
\ No newline at end of file

Modified: python/branches/py3k/Lib/smtpd.py
==============================================================================
--- python/branches/py3k/Lib/smtpd.py	(original)
+++ python/branches/py3k/Lib/smtpd.py	Sat Jul 24 11:51:40 2010
@@ -78,6 +78,7 @@
 import socket
 import asyncore
 import asynchat
+from warnings import warn
 
 __all__ = ["SMTPServer","DebuggingServer","PureProxy","MailmanProxy"]
 
@@ -111,35 +112,157 @@
 
     def __init__(self, server, conn, addr):
         asynchat.async_chat.__init__(self, conn)
-        self.__server = server
-        self.__conn = conn
-        self.__addr = addr
-        self.__line = []
-        self.__state = self.COMMAND
-        self.__greeting = 0
-        self.__mailfrom = None
-        self.__rcpttos = []
-        self.__data = ''
-        self.__fqdn = socket.getfqdn()
-        self.__peer = conn.getpeername()
-        print('Peer:', repr(self.__peer), file=DEBUGSTREAM)
-        self.push('220 %s %s' % (self.__fqdn, __version__))
+        self.smtp_server = server
+        self.conn = conn
+        self.addr = addr
+        self.received_lines = []
+        self.smtp_state = self.COMMAND
+        self.seen_greeting = ''
+        self.mailfrom = None
+        self.rcpttos = []
+        self.received_data = ''
+        self.fqdn = socket.getfqdn()
+        self.peer = conn.getpeername()
+        print('Peer:', repr(self.peer), file=DEBUGSTREAM)
+        self.push('220 %s %s' % (self.fqdn, __version__))
         self.set_terminator(b'\r\n')
 
+    # properties for backwards-compatibility
+    @property
+    def __server(self):
+        warn("Access to __server attribute on SMTPChannel is deprecated, "
+            "use 'smtp_server' instead", PendingDeprecationWarning, 2)
+        return self.smtp_server
+    @__server.setter
+    def __server(self, value):
+        warn("Setting __server attribute on SMTPChannel is deprecated, "
+            "set 'smtp_server' instead", PendingDeprecationWarning, 2)
+        self.smtp_server = value
+
+    @property
+    def __line(self):
+        warn("Access to __line attribute on SMTPChannel is deprecated, "
+            "use 'received_lines' instead", PendingDeprecationWarning, 2)
+        return self.received_lines
+    @__line.setter
+    def __line(self, value):
+        warn("Setting __line attribute on SMTPChannel is deprecated, "
+            "set 'received_lines' instead", PendingDeprecationWarning, 2)
+        self.received_lines = value
+
+    @property
+    def __state(self):
+        warn("Access to __state attribute on SMTPChannel is deprecated, "
+            "use 'smtp_state' instead", PendingDeprecationWarning, 2)
+        return self.smtp_state
+    @__state.setter
+    def __state(self, value):
+        warn("Setting __state attribute on SMTPChannel is deprecated, "
+            "set 'smtp_state' instead", PendingDeprecationWarning, 2)
+        self.smtp_state = value
+
+    @property
+    def __greeting(self):
+        warn("Access to __greeting attribute on SMTPChannel is deprecated, "
+            "use 'seen_greeting' instead", PendingDeprecationWarning, 2)
+        return self.seen_greeting
+    @__greeting.setter
+    def __greeting(self, value):
+        warn("Setting __greeting attribute on SMTPChannel is deprecated, "
+            "set 'seen_greeting' instead", PendingDeprecationWarning, 2)
+        self.seen_greeting = value
+
+    @property
+    def __mailfrom(self):
+        warn("Access to __mailfrom attribute on SMTPChannel is deprecated, "
+            "use 'mailfrom' instead", PendingDeprecationWarning, 2)
+        return self.mailfrom
+    @__mailfrom.setter
+    def __mailfrom(self, value):
+        warn("Setting __mailfrom attribute on SMTPChannel is deprecated, "
+            "set 'mailfrom' instead", PendingDeprecationWarning, 2)
+        self.mailfrom = value
+
+    @property
+    def __rcpttos(self):
+        warn("Access to __rcpttos attribute on SMTPChannel is deprecated, "
+            "use 'rcpttos' instead", PendingDeprecationWarning, 2)
+        return self.rcpttos
+    @__rcpttos.setter
+    def __rcpttos(self, value):
+        warn("Setting __rcpttos attribute on SMTPChannel is deprecated, "
+            "set 'rcpttos' instead", PendingDeprecationWarning, 2)
+        self.rcpttos = value
+
+    @property
+    def __data(self):
+        warn("Access to __data attribute on SMTPChannel is deprecated, "
+            "use 'received_data' instead", PendingDeprecationWarning, 2)
+        return self.received_data
+    @__data.setter
+    def __data(self, value):
+        warn("Setting __data attribute on SMTPChannel is deprecated, "
+            "set 'received_data' instead", PendingDeprecationWarning, 2)
+        self.received_data = value
+
+    @property
+    def __fqdn(self):
+        warn("Access to __fqdn attribute on SMTPChannel is deprecated, "
+            "use 'fqdn' instead", PendingDeprecationWarning, 2)
+        return self.fqdn
+    @__fqdn.setter
+    def __fqdn(self, value):
+        warn("Setting __fqdn attribute on SMTPChannel is deprecated, "
+            "set 'fqdn' instead", PendingDeprecationWarning, 2)
+        self.fqdn = value
+
+    @property
+    def __peer(self):
+        warn("Access to __peer attribute on SMTPChannel is deprecated, "
+            "use 'peer' instead", PendingDeprecationWarning, 2)
+        return self.peer
+    @__peer.setter
+    def __peer(self, value):
+        warn("Setting __peer attribute on SMTPChannel is deprecated, "
+            "set 'peer' instead", PendingDeprecationWarning, 2)
+        self.peer = value
+
+    @property
+    def __conn(self):
+        warn("Access to __conn attribute on SMTPChannel is deprecated, "
+            "use 'conn' instead", PendingDeprecationWarning, 2)
+        return self.conn
+    @__conn.setter
+    def __conn(self, value):
+        warn("Setting __conn attribute on SMTPChannel is deprecated, "
+            "set 'conn' instead", PendingDeprecationWarning, 2)
+        self.conn = value
+
+    @property
+    def __addr(self):
+        warn("Access to __addr attribute on SMTPChannel is deprecated, "
+            "use 'addr' instead", PendingDeprecationWarning, 2)
+        return self.addr
+    @__addr.setter
+    def __addr(self, value):
+        warn("Setting __addr attribute on SMTPChannel is deprecated, "
+            "set 'addr' instead", PendingDeprecationWarning, 2)
+        self.addr = value
+
     # Overrides base class for convenience
     def push(self, msg):
         asynchat.async_chat.push(self, bytes(msg + '\r\n', 'ascii'))
 
     # Implementation of base class abstract method
     def collect_incoming_data(self, data):
-        self.__line.append(str(data, "utf8"))
+        self.received_lines.append(str(data, "utf8"))
 
     # Implementation of base class abstract method
     def found_terminator(self):
-        line = EMPTYSTRING.join(self.__line)
+        line = EMPTYSTRING.join(self.received_lines)
         print('Data:', repr(line), file=DEBUGSTREAM)
-        self.__line = []
-        if self.__state == self.COMMAND:
+        self.received_lines = []
+        if self.smtp_state == self.COMMAND:
             if not line:
                 self.push('500 Error: bad syntax')
                 return
@@ -158,7 +281,7 @@
             method(arg)
             return
         else:
-            if self.__state != self.DATA:
+            if self.smtp_state != self.DATA:
                 self.push('451 Internal confusion')
                 return
             # Remove extraneous carriage returns and de-transparency according
@@ -169,14 +292,14 @@
                     data.append(text[1:])
                 else:
                     data.append(text)
-            self.__data = NEWLINE.join(data)
-            status = self.__server.process_message(self.__peer,
-                                                   self.__mailfrom,
-                                                   self.__rcpttos,
-                                                   self.__data)
-            self.__rcpttos = []
-            self.__mailfrom = None
-            self.__state = self.COMMAND
+            self.received_data = NEWLINE.join(data)
+            status = self.__server.process_message(self.peer,
+                                                   self.mailfrom,
+                                                   self.rcpttos,
+                                                   self.received_data)
+            self.rcpttos = []
+            self.mailfrom = None
+            self.smtp_state = self.COMMAND
             self.set_terminator(b'\r\n')
             if not status:
                 self.push('250 Ok')
@@ -188,11 +311,11 @@
         if not arg:
             self.push('501 Syntax: HELO hostname')
             return
-        if self.__greeting:
+        if self.seen_greeting:
             self.push('503 Duplicate HELO/EHLO')
         else:
-            self.__greeting = arg
-            self.push('250 %s' % self.__fqdn)
+            self.seen_greeting = arg
+            self.push('250 %s' % self.fqdn)
 
     def smtp_NOOP(self, arg):
         if arg:
@@ -225,24 +348,24 @@
         if not address:
             self.push('501 Syntax: MAIL FROM:<address>')
             return
-        if self.__mailfrom:
+        if self.mailfrom:
             self.push('503 Error: nested MAIL command')
             return
-        self.__mailfrom = address
-        print('sender:', self.__mailfrom, file=DEBUGSTREAM)
+        self.mailfrom = address
+        print('sender:', self.mailfrom, file=DEBUGSTREAM)
         self.push('250 Ok')
 
     def smtp_RCPT(self, arg):
         print('===> RCPT', arg, file=DEBUGSTREAM)
-        if not self.__mailfrom:
+        if not self.mailfrom:
             self.push('503 Error: need MAIL command')
             return
         address = self.__getaddr('TO:', arg) if arg else None
         if not address:
             self.push('501 Syntax: RCPT TO: <address>')
             return
-        self.__rcpttos.append(address)
-        print('recips:', self.__rcpttos, file=DEBUGSTREAM)
+        self.rcpttos.append(address)
+        print('recips:', self.rcpttos, file=DEBUGSTREAM)
         self.push('250 Ok')
 
     def smtp_RSET(self, arg):
@@ -250,26 +373,29 @@
             self.push('501 Syntax: RSET')
             return
         # Resets the sender, recipients, and data, but not the greeting
-        self.__mailfrom = None
-        self.__rcpttos = []
-        self.__data = ''
-        self.__state = self.COMMAND
+        self.mailfrom = None
+        self.rcpttos = []
+        self.received_data = ''
+        self.smtp_state = self.COMMAND
         self.push('250 Ok')
 
     def smtp_DATA(self, arg):
-        if not self.__rcpttos:
+        if not self.rcpttos:
             self.push('503 Error: need RCPT command')
             return
         if arg:
             self.push('501 Syntax: DATA')
             return
-        self.__state = self.DATA
+        self.smtp_state = self.DATA
         self.set_terminator(b'\r\n.\r\n')
         self.push('354 End data with <CR><LF>.<CR><LF>')
 
 
 
 class SMTPServer(asyncore.dispatcher):
+    # SMTPChannel class to use for managing client connections
+    channel_class = SMTPChannel
+
     def __init__(self, localaddr, remoteaddr):
         self._localaddr = localaddr
         self._remoteaddr = remoteaddr
@@ -291,7 +417,7 @@
     def handle_accept(self):
         conn, addr = self.accept()
         print('Incoming connection from %s' % repr(addr), file=DEBUGSTREAM)
-        channel = SMTPChannel(self, conn, addr)
+        channel = self.channel_class(self, conn, addr)
 
     # API for "doing something useful with the message"
     def process_message(self, peer, mailfrom, rcpttos, data):

From python-checkins at python.org  Sat Jul 24 12:05:19 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Sat, 24 Jul 2010 12:05:19 +0200 (CEST)
Subject: [Python-checkins] r83126 - in python/branches/release26-maint:
	Lib/test/test_posix.py Misc/NEWS Modules/posixmodule.c
Message-ID: <20100724100519.E320CEE9CA@mail.python.org>

Author: ronald.oussoren
Date: Sat Jul 24 12:05:19 2010
New Revision: 83126

Log:
Merged revisions 83124 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

................
  r83124 | ronald.oussoren | 2010-07-24 10:46:41 +0100 (Sat, 24 Jul 2010) | 15 lines
  
  Merged revisions 83088 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k
  
  ........
    r83088 | ronald.oussoren | 2010-07-23 14:53:51 +0100 (Fri, 23 Jul 2010) | 8 lines
    
    This fixes issue7900 by adding code that deals
    with the fact that getgroups(2) might return
    more that MAX_GROUPS on OSX.
    
    See the issue (and python-dev archives) for the
    gory details. Summarized: OSX behaves rather oddly
    and Apple says this is intentional.
  ........
................


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Lib/test/test_posix.py
   python/branches/release26-maint/Misc/NEWS
   python/branches/release26-maint/Modules/posixmodule.c

Modified: python/branches/release26-maint/Lib/test/test_posix.py
==============================================================================
--- python/branches/release26-maint/Lib/test/test_posix.py	(original)
+++ python/branches/release26-maint/Lib/test/test_posix.py	Sat Jul 24 12:05:19 2010
@@ -8,6 +8,7 @@
     raise test_support.TestSkipped, "posix is not available"
 
 import errno
+import sys
 import time
 import os
 import pwd
@@ -306,9 +307,52 @@
                 os.chdir(curdir)
                 shutil.rmtree(base_path)
 
+    def test_getgroups(self):
+        with os.popen('id -G') as idg:
+            groups = idg.read().strip()
+
+        if not groups:
+            raise unittest.SkipTest("need working 'id -G'")
+
+        self.assertEqual([int(x) for x in groups.split()], posix.getgroups())
+
+class PosixGroupsTester(unittest.TestCase):
+    if posix.getuid() == 0 and hasattr(posix, 'getgroups') and sys.platform != 'darwin':
+
+        def setUp(self):
+            self.saved_groups = posix.getgroups()
+
+        def tearDown(self):
+            if hasattr(posix, 'setgroups'):
+                posix.setgroups(self.saved_groups)
+            elif hasattr(posix, 'initgroups'):
+                name = pwd.getpwuid(posix.getuid()).pw_name
+                posix.initgroups(name, self.saved_groups[0])
+
+        if hasattr(posix, 'initgroups'):
+            def test_initgroups(self):
+                # find missing group
+
+                groups = sorted(self.saved_groups)
+                for g1,g2 in zip(groups[:-1], groups[1:]):
+                    g = g1 + 1
+                    if g < g2:
+                        break
+                else:
+                    g = g2 + 1
+                name = pwd.getpwuid(posix.getuid()).pw_name
+                posix.initgroups(name, g)
+                self.assertIn(g, posix.getgroups())
+
+        if hasattr(posix, 'setgroups'):
+            def test_setgroups(self):
+                for groups in [[0], range(16)]:
+                    posix.setgroups(groups)
+                    self.assertListEqual(groups, posix.getgroups())
+
 
 def test_main():
-    test_support.run_unittest(PosixTester)
+    test_support.run_unittest(PosixTester, PosixGroupsTester)
 
 if __name__ == '__main__':
     test_main()

Modified: python/branches/release26-maint/Misc/NEWS
==============================================================================
--- python/branches/release26-maint/Misc/NEWS	(original)
+++ python/branches/release26-maint/Misc/NEWS	Sat Jul 24 12:05:19 2010
@@ -312,6 +312,13 @@
 Extension Modules
 -----------------
 
+- Issue #7900: The getgroups(2) system call on MacOSX behaves rather oddly
+  compared to other unix systems. In particular, os.getgroups() does
+  not reflect any changes made using os.setgroups() but basicly always
+  returns the same information as the id command.
+
+  os.getgroups() can now return more than 16 groups on MacOSX.
+
 - Issue #9277: Fix bug in struct.pack for bools in standard mode
   (e.g., struct.pack('>?')):  if conversion to bool raised an exception
   then that exception wasn't properly propagated on machines where

Modified: python/branches/release26-maint/Modules/posixmodule.c
==============================================================================
--- python/branches/release26-maint/Modules/posixmodule.c	(original)
+++ python/branches/release26-maint/Modules/posixmodule.c	Sat Jul 24 12:05:19 2010
@@ -3895,17 +3895,49 @@
 #define MAX_GROUPS 64
 #endif
     gid_t grouplist[MAX_GROUPS];
+
+    /* On MacOSX getgroups(2) can return more than MAX_GROUPS results 
+     * This is a helper variable to store the intermediate result when
+     * that happens.
+     *
+     * To keep the code readable the OSX behaviour is unconditional,
+     * according to the POSIX spec this should be safe on all unix-y
+     * systems.
+     */
+    gid_t* alt_grouplist = grouplist;
     int n;
 
     n = getgroups(MAX_GROUPS, grouplist);
-    if (n < 0)
-        posix_error();
-    else {
-        result = PyList_New(n);
-        if (result != NULL) {
+    if (n < 0) {
+        if (errno == EINVAL) {
+            n = getgroups(0, NULL);
+            if (n == -1) {
+                return posix_error();
+            }
+            if (n == 0) {
+                /* Avoid malloc(0) */
+                alt_grouplist = grouplist;
+            } else {
+                alt_grouplist = PyMem_Malloc(n * sizeof(gid_t));
+                if (alt_grouplist == NULL) {
+                    errno = EINVAL;
+                    return posix_error();
+                }
+                n = getgroups(n, alt_grouplist);
+                if (n == -1) {
+                    PyMem_Free(alt_grouplist);
+                    return posix_error();
+                }
+            }
+        } else {
+            return posix_error();
+        }
+    }
+    result = PyList_New(n);
+    if (result != NULL) {
         int i;
         for (i = 0; i < n; ++i) {
-            PyObject *o = PyInt_FromLong((long)grouplist[i]);
+            PyObject *o = PyInt_FromLong((long)alt_grouplist[i]);
             if (o == NULL) {
             Py_DECREF(result);
             result = NULL;
@@ -3913,7 +3945,10 @@
             }
             PyList_SET_ITEM(result, i, o);
         }
-        }
+    }
+
+    if (alt_grouplist != grouplist) {
+        PyMem_Free(alt_grouplist);
     }
 
     return result;

From python-checkins at python.org  Sat Jul 24 12:09:12 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Sat, 24 Jul 2010 12:09:12 +0200 (CEST)
Subject: [Python-checkins] r83127 - python/branches/py3k/Doc/library/os.rst
Message-ID: <20100724100912.0197AEE98B@mail.python.org>

Author: martin.v.loewis
Date: Sat Jul 24 12:09:11 2010
New Revision: 83127

Log:
Put listdir default argument into declaration.


Modified:
   python/branches/py3k/Doc/library/os.rst

Modified: python/branches/py3k/Doc/library/os.rst
==============================================================================
--- python/branches/py3k/Doc/library/os.rst	(original)
+++ python/branches/py3k/Doc/library/os.rst	Sat Jul 24 12:09:11 2010
@@ -1049,10 +1049,10 @@
    Availability: Unix.
 
 
-.. function:: listdir([path])
+.. function:: listdir(path='.')
 
    Return a list containing the names of the entries in the directory given by
-   *path* (default: '.').  The list is in arbitrary order.  It does not include the special
+   *path* (default: ``'.'``).  The list is in arbitrary order.  It does not include the special
    entries ``'.'`` and ``'..'`` even if they are present in the directory.
 
    This function can be called with a bytes or string argument, and returns

From python-checkins at python.org  Sat Jul 24 12:09:37 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Sat, 24 Jul 2010 12:09:37 +0200 (CEST)
Subject: [Python-checkins] r83128 - in python/branches/release26-maint:
	Modules/getpath.c
Message-ID: <20100724100937.0B3D0EE990@mail.python.org>

Author: ronald.oussoren
Date: Sat Jul 24 12:09:36 2010
New Revision: 83128

Log:
Merged revisions 83098 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

................
  r83098 | ronald.oussoren | 2010-07-23 17:12:28 +0100 (Fri, 23 Jul 2010) | 19 lines
  
  Merged revisions 83096 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k
  
  ........
    r83096 | ronald.oussoren | 2010-07-23 17:05:35 +0100 (Fri, 23 Jul 2010) | 13 lines
    
    Ensure that sys.prefix can reliably be found
    on OSX. This fixes a small issue that was exposed
    by running test_subprocess through regrtest (and
    hence in a subdirectory).
    
    Without this patch running python.exe from the
    build tree will fail when these tree conditions
    are true:
    1) the CWD is not the root of build tree
    2) python.exe is found through $PATH
    3) the framework is not yet installed
  ........
................


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Modules/getpath.c

Modified: python/branches/release26-maint/Modules/getpath.c
==============================================================================
--- python/branches/release26-maint/Modules/getpath.c	(original)
+++ python/branches/release26-maint/Modules/getpath.c	Sat Jul 24 12:09:36 2010
@@ -471,7 +471,7 @@
         if (!ismodule(argv0_path)) {
                 /* We are in the build directory so use the name of the
                    executable - we know that the absolute path is passed */
-                strncpy(argv0_path, prog, MAXPATHLEN);
+                strncpy(argv0_path, progpath, MAXPATHLEN);
         }
         else {
                 /* Use the location of the library as the progpath */

From martin at v.loewis.de  Sat Jul 24 12:09:58 2010
From: martin at v.loewis.de (=?UTF-8?B?Ik1hcnRpbiB2LiBMw7Z3aXMi?=)
Date: Sat, 24 Jul 2010 11:09:58 +0100
Subject: [Python-checkins] r83078 - in
 python/branches/py3k:	Doc/library/os.rst Lib/test/test_posix.py
 Misc/NEWS	Modules/posixmodule.c
In-Reply-To: <AANLkTi=5YMUnEFUemQ8+kR0dR+sai5cm67SubS83dcpt@mail.gmail.com>
References: <20100723121641.8C479D7D7@mail.python.org>
	<AANLkTi=5YMUnEFUemQ8+kR0dR+sai5cm67SubS83dcpt@mail.gmail.com>
Message-ID: <4C4ABBF6.5090301@v.loewis.de>

> You can use the new default argument convention in the docs:
> listdir(path=curdir)

Done!

Martin

From python-checkins at python.org  Sat Jul 24 14:29:51 2010
From: python-checkins at python.org (ezio.melotti)
Date: Sat, 24 Jul 2010 14:29:51 +0200 (CEST)
Subject: [Python-checkins] r83129 - in tracker:
	instances/python-dev/extensions/local_replace.py
	instances/python-dev/extensions/test
	instances/python-dev/extensions/test/test_local_replace.py
	instances/python-dev/html/issue.item.html
	roundup-src/roundup/cgi/templating.py
Message-ID: <20100724122951.4F6DDEE99D@mail.python.org>

Author: ezio.melotti
Date: Sat Jul 24 14:29:51 2010
New Revision: 83129

Log:
#285: add a tooltip that shows status and title of linked issues on the tracker. Also avoid linking to unexisting issues.

Added:
   tracker/instances/python-dev/extensions/test/
   tracker/instances/python-dev/extensions/test/test_local_replace.py
Modified:
   tracker/instances/python-dev/extensions/local_replace.py
   tracker/instances/python-dev/html/issue.item.html
   tracker/roundup-src/roundup/cgi/templating.py

Modified: tracker/instances/python-dev/extensions/local_replace.py
==============================================================================
--- tracker/instances/python-dev/extensions/local_replace.py	(original)
+++ tracker/instances/python-dev/extensions/local_replace.py	Sat Jul 24 14:29:51 2010
@@ -1,34 +1,86 @@
 import re
+import cgi
+from roundup import hyperdb
+from roundup.cgi.templating import register_propclass, StringHTMLProperty
+
 
 substitutions = [
-    #  #1234, # 1234
-    (re.compile(r'\#(?P<ws>\s*)(?P<id>1?\d{4,6})\b'),
-     r'<a href="issue\g<id>">#\g<ws>\g<id></a>'),
     #  r12345, r 12345, rev12345, rev 12345, revision12345, revision 12345
     (re.compile(r'\b(?<!/)(?P<revstr>r(ev(ision)?)?\s*)(?P<revision>\d+)'),
-     r'<a href="http://svn.python.org/view?rev=\g<revision>&view=rev">\g<revstr>\g<revision></a>'),
+     r'<a href="http://svn.python.org/view?rev=\g<revision>'
+     r'&view=rev">\g<revstr>\g<revision></a>'),
+
     # Lib/somefile.py, Modules/somemodule.c, Doc/somedocfile.rst, ...
-    (re.compile(r'(?P<sep>(?<!\w/)|(?<!\w)/)(?P<path>(?:Demo|Doc|Grammar|Include|Lib|Mac|Misc|'
-                r'Modules|Parser|PC|PCbuild|Python|RISCOS|Tools|Objects)/[-.a-zA-Z0-9_/]+[a-zA-Z0-9]/?)'),
-     r'<a href="http://svn.python.org/view/python/trunk/\g<path>">\g<sep>\g<path></a>'),
+    (re.compile(r'(?P<sep>(?<!\w/)|(?<!\w)/)(?P<path>(?:Demo|Doc|Grammar|'
+                r'Include|Lib|Mac|Misc|Modules|Parser|PC|PCbuild|Python|'
+                'RISCOS|Tools|Objects)/[-.a-zA-Z0-9_/]+[a-zA-Z0-9]/?)'),
+     r'<a href="http://svn.python.org/view/python/'
+     r'trunk/\g<path>">\g<sep>\g<path></a>'), # XXX trunk is dead
 ]
 
-def localReplace(message):
-    """
-    Turn the following strings in HTML links:
-    #1234, # 1234  ->  <a href="issue1234">#1234</a>
-    r12345, r 12345, rev12345, rev 12345, revision12345, revision 12345
-        ->  <a href="http://svn.python.org/view?rev=12345&view=rev">r12345</a>
-    Lib/somefile.py, Modules/somemodule.c, Doc/somedocfile.rst, ...
-        ->  <a href="http://svn.python.org/view/python/trunk/Dir/filename.ext">Dir/filename.ext</a>
-    """
-    for cre, replacement in substitutions:
-        message = cre.sub(replacement, message)
 
-    return message
+# if the issue number is too big the db will explode -- limit it to 7 digits
+issue_re = re.compile(r'(?P<text>(\#|issue)\s*(?P<id>1?\d{1,6}))\b', re.I)
+
+
+class PyDevStringHTMLProperty(StringHTMLProperty):
+    def _hyper_repl(self, match):
+        """
+        Override the original method and change it to still linkify URLs and
+        emails but avoid linkification of issues and other items
+        (except messages and files).
+        """
+        if match.group('url'):
+            return self._hyper_repl_url(match, '<a href="%s">%s</a>%s')
+        elif match.group('email'):
+            return self._hyper_repl_email(match, '<a href="mailto:%s">%s</a>')
+        elif (len(match.group('id')) < 10 and
+              match.group('class') and
+              match.group('class').lower() in ('msg', 'file')):
+            # linkify msgs but not issues and other things
+            return self._hyper_repl_item(match,
+                '<a href="%(cls)s%(id)s">%(item)s</a>')
+        else:
+            # just return the matched text
+            return match.group(0)
+
+    def pydev_hyperlinked(self):
+        """Create python-dev-specific links."""
+        # first do the normal linkification (without linkify the issues)
+        message =  self.plain(hyperlink=1)
+        # then add links for revision numbers and paths
+        for cre, replacement in substitutions:
+            message = cre.sub(replacement, message)
+        # finally add links for issues
+        message = issue_re.sub(self._linkify_issue, message)
+        return message
+
+    def _linkify_issue(self, match):
+        """Turn an issue (e.g. 'issue 1234' or '#1234') to an HTML link"""
+        template = ('<a class="%(status)s" title="[%(status)s] %(title)s" '
+                    'href="issue%(issue_id)s">%(text)s</a>')
+        issue_id = match.group('id')
+        text = match.group('text')
+        cl = self._db.issue
+        # check if the issue exists
+        if not cl.hasnode(issue_id):
+            return text
+        # get the title
+        title = cgi.escape(cl.get(issue_id, 'title').replace('"', "'"))
+        status_id = cl.get(issue_id, 'status')
+        # get the name of the status
+        if status_id is not None:
+            status = self._db.status.get(status_id, 'name')
+        else:
+            status = 'none'
+        return template % dict(issue_id=issue_id, title=title,
+                               status=status, text=text)
 
+
+# XXX there should be only one regex for this
 noise_change = re.compile('(nosy_count|message_count)\: \d+\.0 -> \d+\.0')
 noise_init = re.compile('(nosy_count|message_count)\: \d+\.0')
+# XXX this shouldn't be a regex
 br = re.compile('<td><br />')
 
 def clean_count(history):
@@ -38,73 +90,5 @@
     return history
 
 def init(instance):
-    instance.registerUtil('localReplace', localReplace)
+    register_propclass(hyperdb.String, PyDevStringHTMLProperty)
     instance.registerUtil('clean_count', clean_count)
-
-if "__main__" == __name__:
-    import unittest
-
-    class TestLocalReplace(unittest.TestCase):
-        def test_localReplace(self):
-            test_strings = [
-                ## r12345, r 12345, rev12345, rev 12345, revision12345, revision 12345
-                ('r222 ',
-                 '<a href="http://svn.python.org/view?rev=222&view=rev">r222</a> '),
-                (' r222 ',
-                 ' <a href="http://svn.python.org/view?rev=222&view=rev">r222</a> '),
-                (' r 222 ',
-                 ' <a href="http://svn.python.org/view?rev=222&view=rev">r 222</a> '),
-                (' rev222 ',
-                 ' <a href="http://svn.python.org/view?rev=222&view=rev">rev222</a> '),
-                (' rev  222 ',
-                 ' <a href="http://svn.python.org/view?rev=222&view=rev">rev  222</a> '),
-                (' revision222 ',
-                 ' <a href="http://svn.python.org/view?rev=222&view=rev">revision222</a> '),
-                (' revision 222 ',
-                 ' <a href="http://svn.python.org/view?rev=222&view=rev">revision 222</a> '),
-                ('wordthatendswithr 222',
-                 'wordthatendswithr 222'),
-                ('http://svn.python.org/view/python/tags/r265 ',
-                 'http://svn.python.org/view/python/tags/r265 '),
-
-                ##  #1234, # 1234
-                ## the lowest issue id is #1000, the highest is #1779871
-                (' #1 ', ' #1 '),
-                (' #10 ', ' #10 '),
-                (' #999 ', ' #999 '),
-                (' # 999 ', ' # 999 '),
-                ('#1000 ', '<a href="issue1000">#1000</a> '),
-                (' #1000 ', ' <a href="issue1000">#1000</a> '),
-                (' # 1000 ', ' <a href="issue1000"># 1000</a> '),
-                (' #5555 ', ' <a href="issue5555">#5555</a> '),
-                (' # 5555 ', ' <a href="issue5555"># 5555</a> '),
-                (' #555555 ', ' <a href="issue555555">#555555</a> '),
-                (' # 555555 ', ' <a href="issue555555"># 555555</a> '),
-                (' #1999999 ', ' <a href="issue1999999">#1999999</a> '),
-                (' # 1999999 ', ' <a href="issue1999999"># 1999999</a> '),
-                (' #2000000 ', ' #2000000 '),
-                (' # 2000000 ', ' # 2000000 '),
-
-                ## Lib/somefile.py, Modules/somemodule.c, Doc/somedocfile.rst, ...
-                ('Lib/cgi.py',
-                 '<a href="http://svn.python.org/view/python/trunk/Lib/cgi.py">Lib/cgi.py</a>'),
-                ('/Lib/cgi.py',
-                 '<a href="http://svn.python.org/view/python/trunk/Lib/cgi.py">/Lib/cgi.py</a>'),
-                ('the bug is in Lib/cgi.py.',
-                 'the bug is in <a href="http://svn.python.org/view/python/trunk/Lib/cgi.py">Lib/cgi.py</a>.'),
-                ('the bug is in /Lib/cgi.py.',
-                 'the bug is in <a href="http://svn.python.org/view/python/trunk/Lib/cgi.py">/Lib/cgi.py</a>.'),
-                ('(Lib/cgi.py)',
-                 '(<a href="http://svn.python.org/view/python/trunk/Lib/cgi.py">Lib/cgi.py</a>)'),
-                ('(/Lib/cgi.py)',
-                 '(<a href="http://svn.python.org/view/python/trunk/Lib/cgi.py">/Lib/cgi.py</a>)'),
-                ('http://svn.python.org/view/python/tags/r265/Lib/cgi.py',
-                 'http://svn.python.org/view/python/tags/r265/Lib/cgi.py'),
-                ('http://svn.python.org/view/python/tags/r265/Lib/cgi.py?view=markup',
-                 'http://svn.python.org/view/python/tags/r265/Lib/cgi.py?view=markup'),
-            ]
-            for text, expected_result in test_strings:
-                self.assertEqual(expected_result, localReplace(text))
-
-    # run the tests when executed directly
-    unittest.main()

Added: tracker/instances/python-dev/extensions/test/test_local_replace.py
==============================================================================
--- (empty file)
+++ tracker/instances/python-dev/extensions/test/test_local_replace.py	Sat Jul 24 14:29:51 2010
@@ -0,0 +1,150 @@
+import sys
+import unittest
+import os.path
+
+if len(sys.argv) != 2:
+    sys.exit('Error: You have to provide the path of Roundup in order to run '
+             'the tests (e.g. /opt/tracker-roundup/lib/python2.6/site-packages/).')
+# add to sys.path the dir where roundup is installed (local_replace will try
+# to import it)
+sys.path.append(sys.argv.pop())
+
+dirs = os.path.dirname(os.path.abspath(__file__)).split(os.path.sep)
+# add the dir where local_replace is (i.e. one level up)
+sys.path.append(os.path.sep.join(dirs[:-1]))
+# add the dir where the roundup tests are
+sys.path.append(os.path.sep.join(dirs[:-4] + ['roundup-src', 'test']))
+
+
+from local_replace import PyDevStringHTMLProperty
+from test_templating import MockDatabase, TemplatingTestCase
+
+
+class MockDBItem(object):
+    def __init__(self, values):
+        # *values* is a list of ids that are supposed to exist in the db
+        # (note that there's no difference here between issues, msg, etc.)
+        self.ids = map(str, values)
+
+    def hasnode(self, id):
+        return id in self.ids
+
+    def get(self, id, value):
+        if value == 'title':
+            return 'Mock title'
+        if value == 'status':
+            return 1
+        if value == 'name':
+            return 'open'
+
+class PyDevMockDatabase(MockDatabase):
+    def __init__(self, values):
+        self.issue = self.msg = self.status = MockDBItem(values)
+    def getclass(self, cls):
+        return self.issue
+
+
+test_strings = [
+    ## r12345, r 12345, rev12345, rev 12345, revision12345, revision 12345
+    ('r222 ',
+     '<a href="http://svn.python.org/view?rev=222&view=rev">r222</a> '),
+    (' r222 ',
+     ' <a href="http://svn.python.org/view?rev=222&view=rev">r222</a> '),
+    (' r 222 ',
+     ' <a href="http://svn.python.org/view?rev=222&view=rev">r 222</a> '),
+    (' rev222 ',
+     ' <a href="http://svn.python.org/view?rev=222&view=rev">rev222</a> '),
+    (' rev  222 ',
+     ' <a href="http://svn.python.org/view?rev=222&view=rev">rev  222</a> '),
+    (' revision222 ',
+     ' <a href="http://svn.python.org/view?rev=222&view=rev">revision222</a> '),
+    (' revision 222 ',
+     ' <a href="http://svn.python.org/view?rev=222&view=rev">revision 222</a> '),
+    ('wordthatendswithr 222',
+     'wordthatendswithr 222'),
+
+
+    ##  #1234, # 1234
+    ## the lowest issue id on bugs.python.org is #1000, the highest is #1779871
+    (' #1 ', ' #1 '),
+    (' #10 ', ' #10 '),
+    (' #999 ', ' #999 '),
+    (' # 999 ', ' # 999 '),
+    ('#1000 ',
+     '<a class="open" title="[open] Mock title" href="issue1000">#1000</a> '),
+    (' #1000 ',
+     ' <a class="open" title="[open] Mock title" href="issue1000">#1000</a> '),
+    (' # 1000 ',
+     ' <a class="open" title="[open] Mock title" href="issue1000"># 1000</a> '),
+    (' #5555 ',
+     ' <a class="open" title="[open] Mock title" href="issue5555">#5555</a> '),
+    (' # 5555 ',
+     ' <a class="open" title="[open] Mock title" href="issue5555"># 5555</a> '),
+    (' #555555 ',
+     ' <a class="open" title="[open] Mock title" href="issue555555">#555555</a> '),
+    (' # 555555 ',
+     ' <a class="open" title="[open] Mock title" href="issue555555"># 555555</a> '),
+    (' #1999999 ',
+     ' <a class="open" title="[open] Mock title" href="issue1999999">#1999999</a> '),
+    (' # 1999999 ',
+     ' <a class="open" title="[open] Mock title" href="issue1999999"># 1999999</a> '),
+    (' #1020 ', ' #1020 '),
+    (' #2000000 ', ' #2000000 '),
+    (' # 2000000 ', ' # 2000000 '),
+    (' #1234567890123  ', ' #1234567890123  '),
+
+    ## Lib/somefile.py, Modules/somemodule.c, Doc/somedocfile.rst, ...
+    ('Lib/cgi.py',
+     '<a href="http://svn.python.org/view/python/trunk/Lib/cgi.py">Lib/cgi.py</a>'),
+    ('/Lib/cgi.py',
+     '<a href="http://svn.python.org/view/python/trunk/Lib/cgi.py">/Lib/cgi.py</a>'),
+    ('see Lib/cgi.py.',
+     'see <a href="http://svn.python.org/view/python/trunk/Lib/cgi.py">Lib/cgi.py</a>.'),
+    ('see /Lib/cgi.py.',
+     'see <a href="http://svn.python.org/view/python/trunk/Lib/cgi.py">/Lib/cgi.py</a>.'),
+    ('(Lib/cgi.py)',
+     '(<a href="http://svn.python.org/view/python/trunk/Lib/cgi.py">Lib/cgi.py</a>)'),
+    ('(/Lib/cgi.py)',
+     '(<a href="http://svn.python.org/view/python/trunk/Lib/cgi.py">/Lib/cgi.py</a>)'),
+
+    ## URLs
+    ('http://svn.python.org/view/python/tags/r265 ',
+     ('<a href="http://svn.python.org/view/python/tags/r265">'
+      'http://svn.python.org/view/python/tags/r265</a> ')),
+    ('http://svn.python.org/view/python/tags/r265/Lib/cgi.py',
+     ('<a href="http://svn.python.org/view/python/tags/r265/Lib/cgi.py">'
+      'http://svn.python.org/view/python/tags/r265/Lib/cgi.py</a>')),
+    ('http://svn.python.org/view/python/tags/r265/Lib/cgi.py?view=markup',
+     ('<a href="http://svn.python.org/view/python/tags/r265/Lib/cgi.py?view=markup">'
+      'http://svn.python.org/view/python/tags/r265/Lib/cgi.py?view=markup</a>')),
+
+    ## emails
+    ('fixed at europython.eu',
+     '<a href="mailto:fixed at europython.eu">fixed at europython.eu</a>'),
+    ('<fixed at europython.eu>',
+     '&lt;<a href="mailto:fixed at europython.eu">fixed at europython.eu</a>&gt;'),
+
+    ## items
+    ('see msg1000.', 'see <a href="msg1000">msg1000</a>.'),
+    ('see msg 1000.', 'see <a href="msg1000">msg 1000</a>.'),
+    ('see file1000.', 'see <a href="file1000">file1000</a>.'),
+    ('see file 1000.', 'see <a href="file1000">file 1000</a>.'),
+    # only msgs and files are linkified
+    ('see version1000.', 'see version1000.'),
+    ('see version 1000.', 'see version 1000.'),
+]
+
+
+class TestPyDevStringHTMLProperty(TemplatingTestCase):
+    def test_replacement(self):
+        # create a db with a few issue/msg/file ids
+        self.client.db = self._db = PyDevMockDatabase(
+                [1000, 5555, 555555, 1999999, 2000000, 1234567890123])
+        self.client.db.security.hasPermission = lambda *args, **kw: True
+        for text, expected_result in test_strings:
+            p = PyDevStringHTMLProperty(self.client, 'test', '1',
+                                        None, 'test', text)
+            self.assertEqual(expected_result, p.pydev_hyperlinked())
+
+# run the tests
+unittest.main()

Modified: tracker/instances/python-dev/html/issue.item.html
==============================================================================
--- tracker/instances/python-dev/html/issue.item.html	(original)
+++ tracker/instances/python-dev/html/issue.item.html	Sat Jul 24 14:29:51 2010
@@ -266,7 +266,7 @@
        Message has been classified as spam.
     </p>
     <pre tal:condition="python:msg.content.is_view_ok()"
-         tal:content="structure python:utils.localReplace(msg.content.hyperlinked())">content</pre>
+         tal:content="structure msg/content/pydev_hyperlinked">content</pre>
    </td>
   </tr>
  </tal:block>

Modified: tracker/roundup-src/roundup/cgi/templating.py
==============================================================================
--- tracker/roundup-src/roundup/cgi/templating.py	(original)
+++ tracker/roundup-src/roundup/cgi/templating.py	Sat Jul 24 14:29:51 2010
@@ -1325,7 +1325,42 @@
     )''', re.X | re.I)
     protocol_re = re.compile('^(ht|f)tp(s?)://', re.I)
 
-    def _hyper_repl_item(self,match,replacement):
+
+
+    def _hyper_repl(self, match):
+        if match.group('url'):
+            return self._hyper_repl_url(match, '<a href="%s">%s</a>%s')
+        elif match.group('email'):
+            return self._hyper_repl_email(match, '<a href="mailto:%s">%s</a>')
+        elif len(match.group('id')) < 10:
+            return self._hyper_repl_item(match,
+                '<a href="%(cls)s%(id)s">%(item)s</a>')
+        else:
+            # just return the matched text
+            return match.group(0)
+
+    def _hyper_repl_url(self, match, replacement):
+        u = s = match.group('url')
+        if not self.protocol_re.search(s):
+            u = 'http://' + s
+        if s.endswith('&gt;'):
+            # catch an escaped ">" at the end of the URL
+            u = s = s[:-4]
+            e = '&gt;'
+        elif s.count('(') != s.count(')'):
+            # don't include extraneous ')' in the link
+            pos = s.rfind(')')
+            e = s[pos:]
+            u = s = s[:pos]
+        else:
+            e = ''
+        return replacement % (u, s, e)
+
+    def _hyper_repl_email(self, match, replacement):
+        s = match.group('email')
+        return replacement % (s, s)
+
+    def _hyper_repl_item(self, match, replacement):
         item = match.group('item')
         cls = match.group('class').lower()
         id = match.group('id')
@@ -1338,32 +1373,6 @@
         except KeyError:
             return item
 
-    def _hyper_repl(self, match):
-        if match.group('url'):
-            u = s = match.group('url')
-            if not self.protocol_re.search(s):
-                u = 'http://' + s
-            if s.endswith('&gt;'):
-                # catch an escaped ">" at the end of the URL
-                u = s = s[:-4]
-                e = '&gt;'
-            elif s.count('(') != s.count(')'):
-                # don't include extraneous ')' in the link
-                pos = s.rfind(')')
-                e = s[pos:]
-                u = s = s[:pos]
-            else:
-                e = ''
-            return '<a href="%s">%s</a>%s' % (u, s, e)
-        elif match.group('email'):
-            s = match.group('email')
-            return '<a href="mailto:%s">%s</a>' % (s, s)
-        elif len(match.group('id')) < 10:
-            return self._hyper_repl_item(match,
-                '<a href="%(cls)s%(id)s">%(item)s</a>')
-        else:
-            # just return the matched text
-            return match.group(0)
 
     def _hyper_repl_rst(self, match):
         if match.group('url'):
@@ -2259,8 +2268,9 @@
         l.append('</select>')
         return '\n'.join(l)
 
+
 # set the propclasses for HTMLItem
-propclasses = (
+propclasses = [
     (hyperdb.String, StringHTMLProperty),
     (hyperdb.Number, NumberHTMLProperty),
     (hyperdb.Boolean, BooleanHTMLProperty),
@@ -2269,7 +2279,17 @@
     (hyperdb.Password, PasswordHTMLProperty),
     (hyperdb.Link, LinkHTMLProperty),
     (hyperdb.Multilink, MultilinkHTMLProperty),
-)
+]
+
+def register_propclass(prop, cls):
+    for index,propclass in enumerate(propclasses):
+        p, c = propclass
+        if prop == p:
+            propclasses[index] = (prop, cls)
+            break
+    else:
+        propclasses.append((prop, cls))
+
 
 def make_sort_function(db, classname, sort_on=None):
     """Make a sort function for a given class

From python-checkins at python.org  Sat Jul 24 14:49:10 2010
From: python-checkins at python.org (ezio.melotti)
Date: Sat, 24 Jul 2010 14:49:10 +0200 (CEST)
Subject: [Python-checkins] r83130 - in
	tracker/instances/python-dev/extensions: local_replace.py
	test/test_local_replace.py
Message-ID: <20100724124910.A89F8EBD3@mail.python.org>

Author: ezio.melotti
Date: Sat Jul 24 14:49:09 2010
New Revision: 83130

Log:
Tweak regex to use a word boundary anchor + negative lookbehind to avoid linking issues in URLs.

Modified:
   tracker/instances/python-dev/extensions/local_replace.py
   tracker/instances/python-dev/extensions/test/test_local_replace.py

Modified: tracker/instances/python-dev/extensions/local_replace.py
==============================================================================
--- tracker/instances/python-dev/extensions/local_replace.py	(original)
+++ tracker/instances/python-dev/extensions/local_replace.py	Sat Jul 24 14:49:09 2010
@@ -20,7 +20,7 @@
 
 
 # if the issue number is too big the db will explode -- limit it to 7 digits
-issue_re = re.compile(r'(?P<text>(\#|issue)\s*(?P<id>1?\d{1,6}))\b', re.I)
+issue_re = re.compile(r'(?P<text>(\#|\b(?<!/)issue)\s*(?P<id>1?\d{1,6}))\b', re.I)
 
 
 class PyDevStringHTMLProperty(StringHTMLProperty):

Modified: tracker/instances/python-dev/extensions/test/test_local_replace.py
==============================================================================
--- tracker/instances/python-dev/extensions/test/test_local_replace.py	(original)
+++ tracker/instances/python-dev/extensions/test/test_local_replace.py	Sat Jul 24 14:49:09 2010
@@ -92,6 +92,7 @@
     (' #2000000 ', ' #2000000 '),
     (' # 2000000 ', ' # 2000000 '),
     (' #1234567890123  ', ' #1234567890123  '),
+    ('pyissue1000', 'pyissue1000'),
 
     ## Lib/somefile.py, Modules/somemodule.c, Doc/somedocfile.rst, ...
     ('Lib/cgi.py',
@@ -117,6 +118,8 @@
     ('http://svn.python.org/view/python/tags/r265/Lib/cgi.py?view=markup',
      ('<a href="http://svn.python.org/view/python/tags/r265/Lib/cgi.py?view=markup">'
       'http://svn.python.org/view/python/tags/r265/Lib/cgi.py?view=markup</a>')),
+    ('http://bugs.python.org/issue1000',
+     '<a href="http://bugs.python.org/issue1000">http://bugs.python.org/issue1000</a>'),
 
     ## emails
     ('fixed at europython.eu',

From python-checkins at python.org  Sat Jul 24 14:54:09 2010
From: python-checkins at python.org (ezio.melotti)
Date: Sat, 24 Jul 2010 14:54:09 +0200 (CEST)
Subject: [Python-checkins] r83131 -
	tracker/instances/python-dev/html/msg.item.html
Message-ID: <20100724125409.1CB1FF42D@mail.python.org>

Author: ezio.melotti
Date: Sat Jul 24 14:54:08 2010
New Revision: 83131

Log:
Use pydev_hyperlinked in msg.item.html too.

Modified:
   tracker/instances/python-dev/html/msg.item.html

Modified: tracker/instances/python-dev/html/msg.item.html
==============================================================================
--- tracker/instances/python-dev/html/msg.item.html	(original)
+++ tracker/instances/python-dev/html/msg.item.html	Sat Jul 24 14:54:08 2010
@@ -95,7 +95,7 @@
   <td class="content" colspan=2
       tal:condition="python:context.content.is_view_ok()">
       <pre
-      tal:content="structure python:utils.localReplace(context.content.hyperlinked())"></pre></td>
+      tal:content="structure msg/content/pydev_hyperlinked"></pre></td>
   <td class="content" colspan=2
       tal:condition="python:not context.content.is_view_ok()">
             Message has been classified as spam and is therefore not

From python-checkins at python.org  Sat Jul 24 15:00:54 2010
From: python-checkins at python.org (ezio.melotti)
Date: Sat, 24 Jul 2010 15:00:54 +0200 (CEST)
Subject: [Python-checkins] r83132 -
	tracker/instances/python-dev/html/msg.item.html
Message-ID: <20100724130054.141F3EE982@mail.python.org>

Author: ezio.melotti
Date: Sat Jul 24 15:00:53 2010
New Revision: 83132

Log:
Use pydev_hyperlinked on the right object.

Modified:
   tracker/instances/python-dev/html/msg.item.html

Modified: tracker/instances/python-dev/html/msg.item.html
==============================================================================
--- tracker/instances/python-dev/html/msg.item.html	(original)
+++ tracker/instances/python-dev/html/msg.item.html	Sat Jul 24 15:00:53 2010
@@ -95,7 +95,7 @@
   <td class="content" colspan=2
       tal:condition="python:context.content.is_view_ok()">
       <pre
-      tal:content="structure msg/content/pydev_hyperlinked"></pre></td>
+      tal:content="structure context/content/pydev_hyperlinked"></pre></td>
   <td class="content" colspan=2
       tal:condition="python:not context.content.is_view_ok()">
             Message has been classified as spam and is therefore not

From python-checkins at python.org  Sat Jul 24 16:15:20 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Sat, 24 Jul 2010 16:15:20 +0200 (CEST)
Subject: [Python-checkins] r83133 -
	python/branches/py3k/Lib/test/test_posix.py
Message-ID: <20100724141520.10BB9EE9D6@mail.python.org>

Author: ronald.oussoren
Date: Sat Jul 24 16:15:19 2010
New Revision: 83133

Log:
Fix for issue 9367: the test code for os.getgroups
assumes that the result of getgroups and the output
of the id(1) command return groups in the same 
order.  That assumption is both fragile and false.


Modified:
   python/branches/py3k/Lib/test/test_posix.py

Modified: python/branches/py3k/Lib/test/test_posix.py
==============================================================================
--- python/branches/py3k/Lib/test/test_posix.py	(original)
+++ python/branches/py3k/Lib/test/test_posix.py	Sat Jul 24 16:15:19 2010
@@ -357,7 +357,11 @@
         if not groups:
             raise unittest.SkipTest("need working 'id -G'")
 
-        self.assertEqual([int(x) for x in groups.split()], posix.getgroups())
+        # The order of groups isn't important, hence the calls
+        # to sorted.
+        self.assertEqual(
+                list(sorted([int(x) for x in groups.split()])),
+                list(sorted(posix.getgroups())))
 
 class PosixGroupsTester(unittest.TestCase):
 

From python-checkins at python.org  Sat Jul 24 16:19:26 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Sat, 24 Jul 2010 16:19:26 +0200 (CEST)
Subject: [Python-checkins] r83134 - in python/branches/release31-maint:
	Lib/test/test_posix.py
Message-ID: <20100724141926.D1AC2EE990@mail.python.org>

Author: ronald.oussoren
Date: Sat Jul 24 16:19:26 2010
New Revision: 83134

Log:
Merged revisions 83133 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83133 | ronald.oussoren | 2010-07-24 15:15:19 +0100 (Sat, 24 Jul 2010) | 5 lines
  
  Fix for issue 9367: the test code for os.getgroups
  assumes that the result of getgroups and the output
  of the id(1) command return groups in the same 
  order.  That assumption is both fragile and false.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/test/test_posix.py

Modified: python/branches/release31-maint/Lib/test/test_posix.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_posix.py	(original)
+++ python/branches/release31-maint/Lib/test/test_posix.py	Sat Jul 24 16:19:26 2010
@@ -281,7 +281,11 @@
         if not groups:
             raise unittest.SkipTest("need working 'id -G'")
 
-        self.assertEqual([int(x) for x in groups.split()], posix.getgroups())
+        # The order of groups isn't important, hence the calls
+        # to sorted.
+        self.assertEqual(
+                list(sorted([int(x) for x in groups.split()])),
+                list(sorted(posix.getgroups())))
 
 class PosixGroupsTester(unittest.TestCase):
 

From python-checkins at python.org  Sat Jul 24 16:21:29 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Sat, 24 Jul 2010 16:21:29 +0200 (CEST)
Subject: [Python-checkins] r83135 - in python/branches/release27-maint:
	Lib/test/test_posix.py
Message-ID: <20100724142129.8A0EFEE9D7@mail.python.org>

Author: ronald.oussoren
Date: Sat Jul 24 16:21:29 2010
New Revision: 83135

Log:
Merged revisions 83133 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83133 | ronald.oussoren | 2010-07-24 15:15:19 +0100 (Sat, 24 Jul 2010) | 5 lines
  
  Fix for issue 9367: the test code for os.getgroups
  assumes that the result of getgroups and the output
  of the id(1) command return groups in the same 
  order.  That assumption is both fragile and false.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/test/test_posix.py

Modified: python/branches/release27-maint/Lib/test/test_posix.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_posix.py	(original)
+++ python/branches/release27-maint/Lib/test/test_posix.py	Sat Jul 24 16:21:29 2010
@@ -371,7 +371,11 @@
         if not groups:
             raise unittest.SkipTest("need working 'id -G'")
 
-        self.assertEqual([int(x) for x in groups.split()], posix.getgroups())
+        # The order of groups isn't important, hence the calls
+        # to sorted.
+        self.assertEqual(
+                list(sorted([int(x) for x in groups.split()])),
+                list(sorted(posix.getgroups())))
 
 class PosixGroupsTester(unittest.TestCase):
 

From python-checkins at python.org  Sat Jul 24 16:23:23 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Sat, 24 Jul 2010 16:23:23 +0200 (CEST)
Subject: [Python-checkins] r83136 - in python/branches/release26-maint:
	Lib/test/test_posix.py
Message-ID: <20100724142323.61D1EEE9A3@mail.python.org>

Author: ronald.oussoren
Date: Sat Jul 24 16:23:23 2010
New Revision: 83136

Log:
Merged revisions 83135 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

................
  r83135 | ronald.oussoren | 2010-07-24 15:21:29 +0100 (Sat, 24 Jul 2010) | 12 lines
  
  Merged revisions 83133 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k
  
  ........
    r83133 | ronald.oussoren | 2010-07-24 15:15:19 +0100 (Sat, 24 Jul 2010) | 5 lines
    
    Fix for issue 9367: the test code for os.getgroups
    assumes that the result of getgroups and the output
    of the id(1) command return groups in the same 
    order.  That assumption is both fragile and false.
  ........
................


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Lib/test/test_posix.py

Modified: python/branches/release26-maint/Lib/test/test_posix.py
==============================================================================
--- python/branches/release26-maint/Lib/test/test_posix.py	(original)
+++ python/branches/release26-maint/Lib/test/test_posix.py	Sat Jul 24 16:23:23 2010
@@ -314,7 +314,11 @@
         if not groups:
             raise unittest.SkipTest("need working 'id -G'")
 
-        self.assertEqual([int(x) for x in groups.split()], posix.getgroups())
+        # The order of groups isn't important, hence the calls
+        # to sorted.
+        self.assertEqual(
+                list(sorted([int(x) for x in groups.split()])),
+                list(sorted(posix.getgroups())))
 
 class PosixGroupsTester(unittest.TestCase):
     if posix.getuid() == 0 and hasattr(posix, 'getgroups') and sys.platform != 'darwin':

From python-checkins at python.org  Sat Jul 24 17:14:39 2010
From: python-checkins at python.org (ezio.melotti)
Date: Sat, 24 Jul 2010 17:14:39 +0200 (CEST)
Subject: [Python-checkins] r83137 -
	tracker/instances/python-dev/extensions/local_replace.py
Message-ID: <20100724151439.08F00D5C5@mail.python.org>

Author: ezio.melotti
Date: Sat Jul 24 17:14:38 2010
New Revision: 83137

Log:
Fix a couple of XXX.

Modified:
   tracker/instances/python-dev/extensions/local_replace.py

Modified: tracker/instances/python-dev/extensions/local_replace.py
==============================================================================
--- tracker/instances/python-dev/extensions/local_replace.py	(original)
+++ tracker/instances/python-dev/extensions/local_replace.py	Sat Jul 24 17:14:38 2010
@@ -77,16 +77,10 @@
                                status=status, text=text)
 
 
-# XXX there should be only one regex for this
-noise_change = re.compile('(nosy_count|message_count)\: \d+\.0 -> \d+\.0')
-noise_init = re.compile('(nosy_count|message_count)\: \d+\.0')
-# XXX this shouldn't be a regex
-br = re.compile('<td><br />')
+noise_changes = re.compile('(nosy_count|message_count)\: \d+\.0( -> \d+\.0)?')
 
 def clean_count(history):
-    history = noise_change.sub('', history)
-    history = noise_init.sub('', history)
-    history = br.sub('<td>', history)
+    history = noise_changes.sub('', history).replace('<td><br />', '<td>')
     return history
 
 def init(instance):

From python-checkins at python.org  Sat Jul 24 17:40:54 2010
From: python-checkins at python.org (ezio.melotti)
Date: Sat, 24 Jul 2010 17:40:54 +0200 (CEST)
Subject: [Python-checkins] r83138 - in tracker/instances:
	meta/detectors/patches.py setuptools/detectors/patches.py
Message-ID: <20100724154054.A7064EE9E1@mail.python.org>

Author: ezio.melotti
Date: Sat Jul 24 17:40:54 2010
New Revision: 83138

Log:
#280: save diff/patch/py attachement to issues in the meta and setuptools trackers as text/plain.

Added:
   tracker/instances/meta/detectors/patches.py
   tracker/instances/setuptools/detectors/patches.py

Added: tracker/instances/meta/detectors/patches.py
==============================================================================
--- (empty file)
+++ tracker/instances/meta/detectors/patches.py	Sat Jul 24 17:40:54 2010
@@ -0,0 +1,17 @@
+# Auditor for patch files
+# Patches should be declared as text/plain (also .py files),
+# independent of what the browser says
+
+import posixpath
+
+sourcetypes = ('.diff', '.patch', '.py')
+
+def ispatch(file, types):
+    return posixpath.splitext(file)[1] in types
+
+def patches_text_plain(db, cl, nodeid, newvalues):
+    if ispatch(newvalues['name'], sourcetypes):
+        newvalues['type'] = 'text/plain'
+
+def init(db):
+    db.file.audit('create', patches_text_plain)

Added: tracker/instances/setuptools/detectors/patches.py
==============================================================================
--- (empty file)
+++ tracker/instances/setuptools/detectors/patches.py	Sat Jul 24 17:40:54 2010
@@ -0,0 +1,17 @@
+# Auditor for patch files
+# Patches should be declared as text/plain (also .py files),
+# independent of what the browser says
+
+import posixpath
+
+sourcetypes = ('.diff', '.patch', '.py')
+
+def ispatch(file, types):
+    return posixpath.splitext(file)[1] in types
+
+def patches_text_plain(db, cl, nodeid, newvalues):
+    if ispatch(newvalues['name'], sourcetypes):
+        newvalues['type'] = 'text/plain'
+
+def init(db):
+    db.file.audit('create', patches_text_plain)

From python-checkins at python.org  Sat Jul 24 23:45:28 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 24 Jul 2010 23:45:28 +0200
Subject: [Python-checkins] distutils2: Automated merge
	with	https://bitbucket.org/mtlpython/distutils2
Message-ID: <E1OcmXA-0006OV-6U@dinsdale.python.org>

tarek.ziade pushed 331aa3da3e67 to distutils2:

http://hg.python.org/distutils2/rev/331aa3da3e67
changeset:   370:331aa3da3e67
parent:      368:fd0608261a21
parent:      369:82ffe3b5a13b
user:        Nicolas Cadou <ncadou at cadou.ca>
date:        Mon Jul 05 23:55:32 2010 -0400
summary:     Automated merge with https://bitbucket.org/mtlpython/distutils2
files:       

diff --git a/src/distutils2/tests/test_version.py b/src/distutils2/tests/test_version.py
--- a/src/distutils2/tests/test_version.py
+++ b/src/distutils2/tests/test_version.py
@@ -3,7 +3,7 @@
 import os
 
 from distutils2.version import NormalizedVersion as V
-from distutils2.version import IrrationalVersionError
+from distutils2.version import HugeMajorVersionNumError, IrrationalVersionError
 from distutils2.version import suggest_normalized_version as suggest
 from distutils2.version import VersionPredicate
 from distutils2.tests.support import unittest
@@ -22,6 +22,10 @@
                 (V('1.0.dev345'), '1.0.dev345'),
                 (V('1.0.post456.dev623'), '1.0.post456.dev623'))
 
+    def test_repr(self):
+
+        self.assertEqual(repr(V('1.0')), "NormalizedVersion('1.0')")
+
     def test_basic_versions(self):
 
         for v, s in self.versions:
@@ -44,6 +48,12 @@
         for s in irrational:
             self.assertRaises(IrrationalVersionError, V, s)
 
+    def test_huge_version(self):
+
+        self.assertEquals(str(V('1980.0')), '1980.0')
+        self.assertRaises(HugeMajorVersionNumError, V, '1981.0')
+        self.assertEquals(str(V('1981.0', error_on_huge_major_num=False)), '1981.0')
+
     def test_comparison(self):
         r"""
         >>> V('1.2.0') == '1.2'
@@ -51,12 +61,33 @@
         ...
         TypeError: cannot compare NormalizedVersion and str
 
+        >>> V('1.2') < '1.3'
+        Traceback (most recent call last):
+        ...
+        TypeError: cannot compare NormalizedVersion and str
+
         >>> V('1.2.0') == V('1.2')
         True
         >>> V('1.2.0') == V('1.2.3')
         False
+        >>> V('1.2.0') != V('1.2.3')
+        True
         >>> V('1.2.0') < V('1.2.3')
         True
+        >>> V('1.2.0') < V('1.2.0')
+        False
+        >>> V('1.2.0') <= V('1.2.0')
+        True
+        >>> V('1.2.0') <= V('1.2.3')
+        True
+        >>> V('1.2.3') <= V('1.2.0')
+        False
+        >>> V('1.2.0') >= V('1.2.0')
+        True
+        >>> V('1.2.3') >= V('1.2.0')
+        True
+        >>> V('1.2.0') >= V('1.2.3')
+        False
         >>> (V('1.0') > V('1.0b2'))
         True
         >>> (V('1.0') > V('1.0c2') > V('1.0c1') > V('1.0b2') > V('1.0b1')
@@ -101,6 +132,7 @@
         self.assertEqual(suggest('1.0c2'), '1.0c2')
         self.assertEqual(suggest('walla walla washington'), None)
         self.assertEqual(suggest('2.4c1'), '2.4c1')
+        self.assertEqual(suggest('v1.0'), '1.0')
 
         # from setuptools
         self.assertEqual(suggest('0.4a1.r10'), '0.4a1.post10')
@@ -151,6 +183,8 @@
         self.assertFalse(VersionPredicate('Hey (<=2.5)').match('2.6.0'))
         self.assertTrue(VersionPredicate('Hey (>=2.5)').match('2.5.1'))
 
+        self.assertRaises(ValueError, VersionPredicate, '')
+
         # XXX need to silent the micro version in this case
         #assert not VersionPredicate('Ho (<3.0,!=2.6)').match('2.6.3')
 
@@ -164,6 +198,11 @@
         for version in other_versions:
             self.assertFalse(V(version).is_final)
 
+    def test_parse_numdots(self):
+        # For code coverage completeness
+        self.assertEquals(V('1.0')._parse_numdots('1.0', '1.0', pad_zeros_length=3),
+                          [1, 0, 0])
+
 
 def test_suite():
     #README = os.path.join(os.path.dirname(__file__), 'README.txt')

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul 24 23:45:28 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 24 Jul 2010 23:45:28 +0200
Subject: [Python-checkins] distutils2: added a version of the test runner
	with built-in support	for producing a
Message-ID: <E1OcmXA-0006OZ-6x@dinsdale.python.org>

tarek.ziade pushed f9b7d091dd66 to distutils2:

http://hg.python.org/distutils2/rev/f9b7d091dd66
changeset:   371:f9b7d091dd66
user:        Yannick Gingras <ygingras at ygingras.net>
date:        Tue Jul 06 06:18:11 2010 -0400
summary:     added a version of the test runner with built-in support for producing a coverage report
files:       src/runtests-cov.py

diff --git a/src/runtests-cov.py b/src/runtests-cov.py
new file mode 100755
--- /dev/null
+++ b/src/runtests-cov.py
@@ -0,0 +1,105 @@
+#!/usr/bin/env python
+"""Tests for distutils2.
+
+The tests for distutils2 are defined in the distutils2.tests package.
+"""
+
+# TODO:
+
+# The coverage report is only accurate when ran inside a virtualenv
+# created with the --no-site-packages option.  When it's not the case,
+# the built-in ignore list is not accurate and third party packages
+# show-up in the report, lowering the overall coverage.  
+
+# One particular problem it docutils on Ubuntu which has a __file__
+# starting with /usr/lib/python2.6 while the path in the coverage
+# report starts with /usr/share/pyshared.
+
+import sys
+from os.path import dirname
+from optparse import OptionParser
+
+def parse_opts():
+    parser = OptionParser(usage="%prog [OPTIONS]", 
+                          description="run the distutils2 unittests")
+    
+    parser.add_option("-q", "--quiet", help="do not print verbose messages", 
+                      action="store_true", default=False)
+    parser.add_option("-c", "--coverage", action="store_true", default=False,
+                      help="produce a coverage report at the end of the run")
+    parser.add_option("-r", "--report", action="store_true", default=False,
+                      help="produce a coverage report from the last test run")
+    parser.add_option("-m", "--show-missing", action="store_true", 
+                      default=False,
+                      help=("Show line numbers of statements in each module "
+                            "that weren't executed."))
+    
+    opts, args = parser.parse_args()
+    return opts, args
+
+def coverage_report(opts):
+    import coverage
+    import unittest2
+    import docutils
+    cov = coverage.coverage()
+    cov.load()
+
+    cov.report(omit_prefixes=["distutils2/tests", 
+                              "runtests", 
+                              "distutils2/_backport", 
+                              dirname(unittest2.__file__),
+                              dirname(dirname(docutils.__file__))], 
+               show_missing=opts.show_missing)
+
+
+def test_main():
+    opts, args = parse_opts()
+    verbose = not opts.quiet
+    ret = 0
+    
+    if opts.coverage or opts.report:
+        import coverage
+        
+    if opts.coverage:
+        cov = coverage.coverage()
+        cov.erase()
+        cov.start()
+    if not opts.report:
+        ret = run_tests(verbose)
+    if opts.coverage:
+        cov.stop()
+        cov.save()
+
+    if opts.report or opts.coverage:
+        coverage_report(opts)
+    
+    return ret
+    
+def run_tests(verbose):
+    import distutils2.tests
+    from distutils2.tests import run_unittest, reap_children, TestFailed
+    from distutils2._backport.tests import test_suite as btest_suite
+    # XXX just supporting -q right now to enable detailed/quiet output
+    if len(sys.argv) > 1:
+        verbose = sys.argv[-1] != '-q'
+    else:
+        verbose = 1
+    try:
+        try:
+            run_unittest([distutils2.tests.test_suite(), btest_suite()],
+                         verbose_=verbose)
+            return 0
+        except TestFailed:
+            return 1
+    finally:
+        reap_children()
+
+if __name__ == "__main__":
+    try:
+        from distutils2.tests.support import unittest
+    except ImportError:
+        sys.stderr.write('Error: You have to install unittest2')
+        sys.exit(1)
+
+    sys.exit(test_main())
+

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul 24 23:45:28 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 24 Jul 2010 23:45:28 +0200
Subject: [Python-checkins] distutils2: Split a white-box test into its own
	class
Message-ID: <E1OcmXA-0006Ol-8n@dinsdale.python.org>

tarek.ziade pushed 05ef666f3870 to distutils2:

http://hg.python.org/distutils2/rev/05ef666f3870
changeset:   374:05ef666f3870
user:        Nicolas Cadou <ncadou at cadou.ca>
date:        Thu Jul 15 20:06:29 2010 -0400
summary:     Split a white-box test into its own class
files:       src/distutils2/tests/test_version.py

diff --git a/src/distutils2/tests/test_version.py b/src/distutils2/tests/test_version.py
--- a/src/distutils2/tests/test_version.py
+++ b/src/distutils2/tests/test_version.py
@@ -198,16 +198,21 @@
         for version in other_versions:
             self.assertFalse(V(version).is_final)
 
+class VersionWhiteBoxTestCase(unittest.TestCase):
+
     def test_parse_numdots(self):
-        # For code coverage completeness
-        self.assertEquals(V('1.0')._parse_numdots('1.0', '1.0', pad_zeros_length=3),
+        # For code coverage completeness, as pad_zeros_length can't be set or
+        # influenced from the public interface
+        self.assertEquals(V('1.0')._parse_numdots('1.0', '1.0',
+                                                  pad_zeros_length=3),
                           [1, 0, 0])
 
 
 def test_suite():
     #README = os.path.join(os.path.dirname(__file__), 'README.txt')
     #suite = [doctest.DocFileSuite(README), unittest.makeSuite(VersionTestCase)]
-    suite = [unittest.makeSuite(VersionTestCase)]
+    suite = [unittest.makeSuite(VersionTestCase),
+             unittest.makeSuite(VersionWhiteBoxTestCase)]
     return unittest.TestSuite(suite)
 
 if __name__ == "__main__":

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul 24 23:45:28 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 24 Jul 2010 23:45:28 +0200
Subject: [Python-checkins] distutils2: Removed impossible code paths
Message-ID: <E1OcmXA-0006Oh-8E@dinsdale.python.org>

tarek.ziade pushed f3724d596ce4 to distutils2:

http://hg.python.org/distutils2/rev/f3724d596ce4
changeset:   373:f3724d596ce4
user:        Nicolas Cadou <ncadou at cadou.ca>
date:        Thu Jul 15 20:02:57 2010 -0400
summary:     Removed impossible code paths
files:       src/distutils2/version.py

diff --git a/src/distutils2/version.py b/src/distutils2/version.py
--- a/src/distutils2/version.py
+++ b/src/distutils2/version.py
@@ -379,8 +379,6 @@
     def __init__(self, predicate):
         predicate = predicate.strip()
         match = _PLAIN_VERSIONS.match(predicate)
-        if match is None:
-            raise ValueError('Bad predicate "%s"' % predicate)
         self.name = None
         predicates = match.groups()[0]
         self.predicates = [_split_predicate(pred.strip())
@@ -391,8 +389,6 @@
     def __init__(self, predicate):
         predicate = predicate.strip()
         match = _PLAIN_VERSIONS.match(predicate)
-        if match is None:
-            raise ValueError('Bad predicate "%s"' % predicate)
         self.name = None
         self.predicates = _split_predicate(match.groups()[0])
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul 24 23:45:28 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 24 Jul 2010 23:45:28 +0200
Subject: [Python-checkins] distutils2: Improved test coverage
Message-ID: <E1OcmXA-0006Ou-9p@dinsdale.python.org>

tarek.ziade pushed e781043ca958 to distutils2:

http://hg.python.org/distutils2/rev/e781043ca958
changeset:   376:e781043ca958
user:        Nicolas Cadou <ncadou at cadou.ca>
date:        Fri Jul 16 00:12:41 2010 -0400
summary:     Improved test coverage
files:       src/distutils2/tests/test_metadata.py

diff --git a/src/distutils2/tests/test_metadata.py b/src/distutils2/tests/test_metadata.py
--- a/src/distutils2/tests/test_metadata.py
+++ b/src/distutils2/tests/test_metadata.py
@@ -6,6 +6,8 @@
 from distutils2.metadata import (DistributionMetadata, _interpret,
                                  PKG_INFO_PREFERRED_VERSION)
 from distutils2.tests.support import unittest, LoggingSilencer
+from distutils2.errors import (MetadataConflictError,
+                               MetadataUnrecognizedVersionError)
 
 class DistributionMetadataTestCase(LoggingSilencer, unittest.TestCase):
 
@@ -125,6 +127,10 @@
         metadata['Obsoletes-Dist'] = 'ok'
         self.assertEqual(metadata['Metadata-Version'], '1.2')
 
+        self.assertRaises(MetadataConflictError, metadata.set,
+                          'Obsoletes', 'ok')
+
+        del metadata['Obsoletes']
         del metadata['Obsoletes-Dist']
         metadata['Version'] = '1'
         self.assertEqual(metadata['Metadata-Version'], '1.0')
@@ -139,6 +145,9 @@
         metadata.read_file(StringIO(open(PKG_INFO).read()))
         self.assertEqual(metadata['Metadata-Version'], '1.1')
 
+        metadata.version = '1.618'
+        self.assertRaises(MetadataUnrecognizedVersionError, metadata.keys)
+
     def test_warnings(self):
         metadata = DistributionMetadata()
 

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul 24 23:45:28 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 24 Jul 2010 23:45:28 +0200
Subject: [Python-checkins] distutils2: Add missing names to acks,
	fix another two
Message-ID: <E1OcmXA-0006PA-C9@dinsdale.python.org>

tarek.ziade pushed feceafae5855 to distutils2:

http://hg.python.org/distutils2/rev/feceafae5855
changeset:   380:feceafae5855
user:        ?ric Araujo <merwok at netwok.org>
date:        Sun Jul 18 01:38:36 2010 +0200
summary:     Add missing names to acks, fix another two
files:       src/CONTRIBUTORS.txt

diff --git a/src/CONTRIBUTORS.txt b/src/CONTRIBUTORS.txt
--- a/src/CONTRIBUTORS.txt
+++ b/src/CONTRIBUTORS.txt
@@ -13,15 +13,18 @@
 - Pior Bastida
 - Titus Brown
 - Nicolas Cadou
+- Konrad Delong
 - Josip Djolonga
 - Yannick Gringas
 - Jeremy Kloth
 - Martin von L??wis
 - Carl Meyer
-- Alexis Metaireau
+- Alexis M??taireau
+- Zubin Mithra
 - Michael Mulich
-- George Peris
+- George Peristerakis
 - Sean Reifschneider
+- Luis Rojas
 - Erik Rose
 - Brian Rosner
 - Alexandre Vassalotti

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul 24 23:45:28 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 24 Jul 2010 23:45:28 +0200
Subject: [Python-checkins] distutils2: Improved test coverage
Message-ID: <E1OcmXA-0006Op-9F@dinsdale.python.org>

tarek.ziade pushed fdc0f44dc5ec to distutils2:

http://hg.python.org/distutils2/rev/fdc0f44dc5ec
changeset:   375:fdc0f44dc5ec
user:        Luis Rojas <luis.r.rojas at gmail.com>
date:        Thu Jul 15 21:30:43 2010 -0400
summary:     Improved test coverage
files:       src/distutils2/tests/test_util.py

diff --git a/src/distutils2/tests/test_util.py b/src/distutils2/tests/test_util.py
--- a/src/distutils2/tests/test_util.py
+++ b/src/distutils2/tests/test_util.py
@@ -4,6 +4,8 @@
 from copy import copy
 from StringIO import StringIO
 import subprocess
+import tempfile
+import time
 
 from distutils2.errors import (DistutilsPlatformError,
                                DistutilsByteCompileError,
@@ -257,7 +259,10 @@
 
     def test_newer(self):
         self.assertRaises(DistutilsFileError, util.newer, 'xxx', 'xxx')
-
+        self.newer_f1 = tempfile.NamedTemporaryFile()
+        time.sleep(1)
+        self.newer_f2 = tempfile.NamedTemporaryFile()
+        self.assertTrue(util.newer(self.newer_f2.name, self.newer_f1.name))
 
     def test_find_packages(self):
         # let's create a structure we want to scan:

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul 24 23:45:28 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 24 Jul 2010 23:45:28 +0200
Subject: [Python-checkins] distutils2: Use unittest[2] import kludge in
	runtest-cov; kill trailing whitespace
Message-ID: <E1OcmXA-0006PE-Ck@dinsdale.python.org>

tarek.ziade pushed 2a04121546a9 to distutils2:

http://hg.python.org/distutils2/rev/2a04121546a9
changeset:   381:2a04121546a9
user:        ?ric Araujo <merwok at netwok.org>
date:        Sun Jul 18 03:27:26 2010 +0200
summary:     Use unittest[2] import kludge in runtest-cov; kill trailing whitespace
files:       src/runtests-cov.py

diff --git a/src/runtests-cov.py b/src/runtests-cov.py
--- a/src/runtests-cov.py
+++ b/src/runtests-cov.py
@@ -23,16 +23,16 @@
     return dirnames
 
 def parse_opts():
-    parser = OptionParser(usage="%prog [OPTIONS]", 
+    parser = OptionParser(usage="%prog [OPTIONS]",
                           description="run the distutils2 unittests")
     
-    parser.add_option("-q", "--quiet", help="do not print verbose messages", 
+    parser.add_option("-q", "--quiet", help="do not print verbose messages",
                       action="store_true", default=False)
     parser.add_option("-c", "--coverage", action="store_true", default=False,
                       help="produce a coverage report at the end of the run")
     parser.add_option("-r", "--report", action="store_true", default=False,
                       help="produce a coverage report from the last test run")
-    parser.add_option("-m", "--show-missing", action="store_true", 
+    parser.add_option("-m", "--show-missing", action="store_true",
                       default=False,
                       help=("Show line numbers of statements in each module "
                             "that weren't executed."))
@@ -42,12 +42,12 @@
 
 def coverage_report(opts):
     import coverage
-    import unittest2
+    from distutils2.tests.support import unittest
     cov = coverage.coverage()
     cov.load()
 
     prefixes = ["runtests", "distutils2/tests", "distutils2/_backport"]
-    prefixes += ignore_prefixes(unittest2)
+    prefixes += ignore_prefixes(unittest)
 
     try:
         import docutils
@@ -70,10 +70,10 @@
     opts, args = parse_opts()
     verbose = not opts.quiet
     ret = 0
-    
+
     if opts.coverage or opts.report:
         import coverage
-        
+
     if opts.coverage:
         cov = coverage.coverage()
         cov.erase()
@@ -86,9 +86,9 @@
 
     if opts.report or opts.coverage:
         coverage_report(opts)
-    
+
     return ret
-    
+
 def run_tests(verbose):
     import distutils2.tests
     from distutils2.tests import run_unittest, reap_children, TestFailed

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul 24 23:45:28 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 24 Jul 2010 23:45:28 +0200
Subject: [Python-checkins] distutils2: Improved code coverage in
	distutils2.version
Message-ID: <E1OcmXA-0006OS-5n@dinsdale.python.org>

tarek.ziade pushed 82ffe3b5a13b to distutils2:

http://hg.python.org/distutils2/rev/82ffe3b5a13b
changeset:   369:82ffe3b5a13b
parent:      350:8b73f08d1d8d
user:        Nicolas Cadou <ncadou at cadou.ca>
date:        Mon Jul 05 23:54:27 2010 -0400
summary:     Improved code coverage in distutils2.version
files:       src/distutils2/tests/test_version.py

diff --git a/src/distutils2/tests/test_version.py b/src/distutils2/tests/test_version.py
--- a/src/distutils2/tests/test_version.py
+++ b/src/distutils2/tests/test_version.py
@@ -3,7 +3,7 @@
 import os
 
 from distutils2.version import NormalizedVersion as V
-from distutils2.version import IrrationalVersionError
+from distutils2.version import HugeMajorVersionNumError, IrrationalVersionError
 from distutils2.version import suggest_normalized_version as suggest
 from distutils2.version import VersionPredicate
 from distutils2.tests.support import unittest
@@ -22,6 +22,10 @@
                 (V('1.0.dev345'), '1.0.dev345'),
                 (V('1.0.post456.dev623'), '1.0.post456.dev623'))
 
+    def test_repr(self):
+
+        self.assertEqual(repr(V('1.0')), "NormalizedVersion('1.0')")
+
     def test_basic_versions(self):
 
         for v, s in self.versions:
@@ -44,6 +48,12 @@
         for s in irrational:
             self.assertRaises(IrrationalVersionError, V, s)
 
+    def test_huge_version(self):
+
+        self.assertEquals(str(V('1980.0')), '1980.0')
+        self.assertRaises(HugeMajorVersionNumError, V, '1981.0')
+        self.assertEquals(str(V('1981.0', error_on_huge_major_num=False)), '1981.0')
+
     def test_comparison(self):
         r"""
         >>> V('1.2.0') == '1.2'
@@ -51,12 +61,33 @@
         ...
         TypeError: cannot compare NormalizedVersion and str
 
+        >>> V('1.2') < '1.3'
+        Traceback (most recent call last):
+        ...
+        TypeError: cannot compare NormalizedVersion and str
+
         >>> V('1.2.0') == V('1.2')
         True
         >>> V('1.2.0') == V('1.2.3')
         False
+        >>> V('1.2.0') != V('1.2.3')
+        True
         >>> V('1.2.0') < V('1.2.3')
         True
+        >>> V('1.2.0') < V('1.2.0')
+        False
+        >>> V('1.2.0') <= V('1.2.0')
+        True
+        >>> V('1.2.0') <= V('1.2.3')
+        True
+        >>> V('1.2.3') <= V('1.2.0')
+        False
+        >>> V('1.2.0') >= V('1.2.0')
+        True
+        >>> V('1.2.3') >= V('1.2.0')
+        True
+        >>> V('1.2.0') >= V('1.2.3')
+        False
         >>> (V('1.0') > V('1.0b2'))
         True
         >>> (V('1.0') > V('1.0c2') > V('1.0c1') > V('1.0b2') > V('1.0b1')
@@ -101,6 +132,7 @@
         self.assertEqual(suggest('1.0c2'), '1.0c2')
         self.assertEqual(suggest('walla walla washington'), None)
         self.assertEqual(suggest('2.4c1'), '2.4c1')
+        self.assertEqual(suggest('v1.0'), '1.0')
 
         # from setuptools
         self.assertEqual(suggest('0.4a1.r10'), '0.4a1.post10')
@@ -151,6 +183,8 @@
         self.assertFalse(VersionPredicate('Hey (<=2.5)').match('2.6.0'))
         self.assertTrue(VersionPredicate('Hey (>=2.5)').match('2.5.1'))
 
+        self.assertRaises(ValueError, VersionPredicate, '')
+
         # XXX need to silent the micro version in this case
         #assert not VersionPredicate('Ho (<3.0,!=2.6)').match('2.6.3')
 
@@ -164,6 +198,11 @@
         for version in other_versions:
             self.assertFalse(V(version).is_final)
 
+    def test_parse_numdots(self):
+        # For code coverage completeness
+        self.assertEquals(V('1.0')._parse_numdots('1.0', '1.0', pad_zeros_length=3),
+                          [1, 0, 0])
+
 
 def test_suite():
     #README = os.path.join(os.path.dirname(__file__), 'README.txt')

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul 24 23:45:28 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 24 Jul 2010 23:45:28 +0200
Subject: [Python-checkins] distutils2: Add support of imports fixer for
	multi packages import	and related unittest
Message-ID: <E1OcmXA-0006OH-58@dinsdale.python.org>

tarek.ziade pushed fd0608261a21 to distutils2:

http://hg.python.org/distutils2/rev/fd0608261a21
changeset:   368:fd0608261a21
parent:      350:8b73f08d1d8d
user:        Pior Bastida <pior at pbastida.net>
date:        Mon Jul 05 23:00:28 2010 -0400
summary:     Add support of imports fixer for multi packages import and related unittest
files:       src/distutils2/converter/fixers/fix_imports.py, src/distutils2/tests/conversions/05_after.py, src/distutils2/tests/conversions/05_before.py

diff --git a/src/distutils2/converter/fixers/fix_imports.py b/src/distutils2/converter/fixers/fix_imports.py
--- a/src/distutils2/converter/fixers/fix_imports.py
+++ b/src/distutils2/converter/fixers/fix_imports.py
@@ -36,11 +36,16 @@
             pattern = []
             next = imp.next_sibling
             while next is not None:
+                # Get the first child if we have a Node
+                if not hasattr(next, "value"):
+                    next = next.children[0]
                 pattern.append(next.value)
                 if not hasattr(next, "next_sibling"):
                     next.next_sibling = next.get_next_sibling()
                 next = next.next_sibling
-            if pattern == ['import', 'setup']:
+            
+            if set(pattern).issubset(set(
+                    ['import', ',', 'setup', 'find_packages'])):
                 imp.value = 'distutils2.core'
                 imp.changed()
 
diff --git a/src/distutils2/tests/conversions/05_after.py b/src/distutils2/tests/conversions/05_after.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/conversions/05_after.py
@@ -0,0 +1,137 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+#
+# Copyright (C) 2003-2009 Edgewall Software
+# All rights reserved.
+#
+# This software is licensed as described in the file COPYING, which
+# you should have received as part of this distribution. The terms
+# are also available at http://trac.edgewall.org/wiki/TracLicense.
+#
+# This software consists of voluntary contributions made by many
+# individuals. For the exact contribution history, see the revision
+# history and logs, available at http://trac.edgewall.org/log/.
+
+from distutils2.core import setup, find_packages
+
+extra = {}
+
+try:
+    import babel
+    
+    extractors = [
+        ('**.py',                'python', None),
+        ('**/templates/**.html', 'genshi', None),
+        ('**/templates/**.txt',  'genshi',
+         {'template_class': 'genshi.template:NewTextTemplate'}),
+    ]
+    extra['message_extractors'] = {
+        'trac': extractors,
+        'tracopt': extractors,
+    }
+
+    from trac.util.dist import get_l10n_js_cmdclass
+    extra['cmdclass'] = get_l10n_js_cmdclass()
+
+except ImportError, e:
+    pass
+
+setup(
+    name = 'Trac',
+    version = '0.12.1',
+    summary = 'Integrated SCM, wiki, issue tracker and project environment',
+    description = """
+Trac is a minimalistic web-based software project management and bug/issue
+tracking system. It provides an interface to the Subversion revision control
+systems, an integrated wiki, flexible issue tracking and convenient report
+facilities.
+""",
+    author = 'Edgewall Software',
+    author_email = 'info at edgewall.com',
+    license = 'BSD',
+    home_page = 'http://trac.edgewall.org/',
+    download_url = 'http://trac.edgewall.org/wiki/TracDownload',
+    classifiers = [
+        'Environment :: Web Environment',
+        'Framework :: Trac',
+        'Intended Audience :: Developers',
+        'License :: OSI Approved :: BSD License',
+        'Operating System :: OS Independent',
+        'Programming Language :: Python',
+        'Topic :: Software Development :: Bug Tracking',
+        'Topic :: Software Development :: Version Control',
+    ],
+
+    packages = find_packages(exclude=['*.tests']),
+    package_data = {
+        '': ['templates/*'],
+        'trac': ['htdocs/*.*', 'htdocs/README', 'htdocs/js/*.*',
+                 'htdocs/js/messages/*.*', 'htdocs/css/*.*',
+                 'htdocs/guide/*', 'locale/*/LC_MESSAGES/messages.mo'],
+        'trac.wiki': ['default-pages/*'],
+        'trac.ticket': ['workflows/*.ini'],
+    },
+
+    test_suite = 'trac.test.suite',
+    zip_safe = True,
+
+    requires_dist = [
+        'setuptools>=0.6b1',
+        'Genshi>=0.6',
+    ],
+    extras_require = {
+        'Babel': ['Babel>=0.9.5'],
+        'Pygments': ['Pygments>=0.6'],
+        'reST': ['docutils>=0.3'],
+        'SilverCity': ['SilverCity>=0.9.4'],
+        'Textile': ['textile>=2.0'],
+    },
+
+    entry_points = """
+        [console_scripts]
+        trac-admin = trac.admin.console:run
+        tracd = trac.web.standalone:main
+
+        [trac.plugins]
+        trac.about = trac.about
+        trac.admin.console = trac.admin.console
+        trac.admin.web_ui = trac.admin.web_ui
+        trac.attachment = trac.attachment
+        trac.db.mysql = trac.db.mysql_backend
+        trac.db.postgres = trac.db.postgres_backend
+        trac.db.sqlite = trac.db.sqlite_backend
+        trac.mimeview.patch = trac.mimeview.patch
+        trac.mimeview.pygments = trac.mimeview.pygments[Pygments]
+        trac.mimeview.rst = trac.mimeview.rst[reST]
+        trac.mimeview.silvercity = trac.mimeview.silvercity[SilverCity]
+        trac.mimeview.txtl = trac.mimeview.txtl[Textile]
+        trac.prefs = trac.prefs.web_ui
+        trac.search = trac.search.web_ui
+        trac.ticket.admin = trac.ticket.admin
+        trac.ticket.query = trac.ticket.query
+        trac.ticket.report = trac.ticket.report
+        trac.ticket.roadmap = trac.ticket.roadmap
+        trac.ticket.web_ui = trac.ticket.web_ui
+        trac.timeline = trac.timeline.web_ui
+        trac.versioncontrol.admin = trac.versioncontrol.admin
+        trac.versioncontrol.svn_authz = trac.versioncontrol.svn_authz
+        trac.versioncontrol.svn_fs = trac.versioncontrol.svn_fs
+        trac.versioncontrol.svn_prop = trac.versioncontrol.svn_prop
+        trac.versioncontrol.web_ui = trac.versioncontrol.web_ui
+        trac.web.auth = trac.web.auth
+        trac.web.session = trac.web.session
+        trac.wiki.admin = trac.wiki.admin
+        trac.wiki.interwiki = trac.wiki.interwiki
+        trac.wiki.macros = trac.wiki.macros
+        trac.wiki.web_ui = trac.wiki.web_ui
+        trac.wiki.web_api = trac.wiki.web_api
+        tracopt.mimeview.enscript = tracopt.mimeview.enscript
+        tracopt.mimeview.php = tracopt.mimeview.php
+        tracopt.perm.authz_policy = tracopt.perm.authz_policy
+        tracopt.perm.config_perm_provider = tracopt.perm.config_perm_provider
+        tracopt.ticket.commit_updater = tracopt.ticket.commit_updater
+        tracopt.ticket.deleter = tracopt.ticket.deleter
+    """,
+
+    **extra
+)
diff --git a/src/distutils2/tests/conversions/05_before.py b/src/distutils2/tests/conversions/05_before.py
new file mode 100755
--- /dev/null
+++ b/src/distutils2/tests/conversions/05_before.py
@@ -0,0 +1,137 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+#
+# Copyright (C) 2003-2009 Edgewall Software
+# All rights reserved.
+#
+# This software is licensed as described in the file COPYING, which
+# you should have received as part of this distribution. The terms
+# are also available at http://trac.edgewall.org/wiki/TracLicense.
+#
+# This software consists of voluntary contributions made by many
+# individuals. For the exact contribution history, see the revision
+# history and logs, available at http://trac.edgewall.org/log/.
+
+from setuptools import setup, find_packages
+
+extra = {}
+
+try:
+    import babel
+    
+    extractors = [
+        ('**.py',                'python', None),
+        ('**/templates/**.html', 'genshi', None),
+        ('**/templates/**.txt',  'genshi',
+         {'template_class': 'genshi.template:NewTextTemplate'}),
+    ]
+    extra['message_extractors'] = {
+        'trac': extractors,
+        'tracopt': extractors,
+    }
+
+    from trac.util.dist import get_l10n_js_cmdclass
+    extra['cmdclass'] = get_l10n_js_cmdclass()
+
+except ImportError, e:
+    pass
+
+setup(
+    name = 'Trac',
+    version = '0.12.1',
+    description = 'Integrated SCM, wiki, issue tracker and project environment',
+    long_description = """
+Trac is a minimalistic web-based software project management and bug/issue
+tracking system. It provides an interface to the Subversion revision control
+systems, an integrated wiki, flexible issue tracking and convenient report
+facilities.
+""",
+    author = 'Edgewall Software',
+    author_email = 'info at edgewall.com',
+    license = 'BSD',
+    url = 'http://trac.edgewall.org/',
+    download_url = 'http://trac.edgewall.org/wiki/TracDownload',
+    classifiers = [
+        'Environment :: Web Environment',
+        'Framework :: Trac',
+        'Intended Audience :: Developers',
+        'License :: OSI Approved :: BSD License',
+        'Operating System :: OS Independent',
+        'Programming Language :: Python',
+        'Topic :: Software Development :: Bug Tracking',
+        'Topic :: Software Development :: Version Control',
+    ],
+
+    packages = find_packages(exclude=['*.tests']),
+    package_data = {
+        '': ['templates/*'],
+        'trac': ['htdocs/*.*', 'htdocs/README', 'htdocs/js/*.*',
+                 'htdocs/js/messages/*.*', 'htdocs/css/*.*',
+                 'htdocs/guide/*', 'locale/*/LC_MESSAGES/messages.mo'],
+        'trac.wiki': ['default-pages/*'],
+        'trac.ticket': ['workflows/*.ini'],
+    },
+
+    test_suite = 'trac.test.suite',
+    zip_safe = True,
+
+    install_requires = [
+        'setuptools>=0.6b1',
+        'Genshi>=0.6',
+    ],
+    extras_require = {
+        'Babel': ['Babel>=0.9.5'],
+        'Pygments': ['Pygments>=0.6'],
+        'reST': ['docutils>=0.3'],
+        'SilverCity': ['SilverCity>=0.9.4'],
+        'Textile': ['textile>=2.0'],
+    },
+
+    entry_points = """
+        [console_scripts]
+        trac-admin = trac.admin.console:run
+        tracd = trac.web.standalone:main
+
+        [trac.plugins]
+        trac.about = trac.about
+        trac.admin.console = trac.admin.console
+        trac.admin.web_ui = trac.admin.web_ui
+        trac.attachment = trac.attachment
+        trac.db.mysql = trac.db.mysql_backend
+        trac.db.postgres = trac.db.postgres_backend
+        trac.db.sqlite = trac.db.sqlite_backend
+        trac.mimeview.patch = trac.mimeview.patch
+        trac.mimeview.pygments = trac.mimeview.pygments[Pygments]
+        trac.mimeview.rst = trac.mimeview.rst[reST]
+        trac.mimeview.silvercity = trac.mimeview.silvercity[SilverCity]
+        trac.mimeview.txtl = trac.mimeview.txtl[Textile]
+        trac.prefs = trac.prefs.web_ui
+        trac.search = trac.search.web_ui
+        trac.ticket.admin = trac.ticket.admin
+        trac.ticket.query = trac.ticket.query
+        trac.ticket.report = trac.ticket.report
+        trac.ticket.roadmap = trac.ticket.roadmap
+        trac.ticket.web_ui = trac.ticket.web_ui
+        trac.timeline = trac.timeline.web_ui
+        trac.versioncontrol.admin = trac.versioncontrol.admin
+        trac.versioncontrol.svn_authz = trac.versioncontrol.svn_authz
+        trac.versioncontrol.svn_fs = trac.versioncontrol.svn_fs
+        trac.versioncontrol.svn_prop = trac.versioncontrol.svn_prop
+        trac.versioncontrol.web_ui = trac.versioncontrol.web_ui
+        trac.web.auth = trac.web.auth
+        trac.web.session = trac.web.session
+        trac.wiki.admin = trac.wiki.admin
+        trac.wiki.interwiki = trac.wiki.interwiki
+        trac.wiki.macros = trac.wiki.macros
+        trac.wiki.web_ui = trac.wiki.web_ui
+        trac.wiki.web_api = trac.wiki.web_api
+        tracopt.mimeview.enscript = tracopt.mimeview.enscript
+        tracopt.mimeview.php = tracopt.mimeview.php
+        tracopt.perm.authz_policy = tracopt.perm.authz_policy
+        tracopt.perm.config_perm_provider = tracopt.perm.config_perm_provider
+        tracopt.ticket.commit_updater = tracopt.ticket.commit_updater
+        tracopt.ticket.deleter = tracopt.ticket.deleter
+    """,
+
+    **extra
+)

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul 24 23:45:28 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 24 Jul 2010 23:45:28 +0200
Subject: [Python-checkins] distutils2: Automated merge
	with	ssh://hg@bitbucket.org/tarek/distutils2
Message-ID: <E1OcmXA-0006Oe-7g@dinsdale.python.org>

tarek.ziade pushed 6684bd704207 to distutils2:

http://hg.python.org/distutils2/rev/6684bd704207
changeset:   372:6684bd704207
parent:      367:70b98f177413
parent:      371:f9b7d091dd66
user:        Yannick Gingras <ygingras at ygingras.net>
date:        Thu Jul 15 18:45:05 2010 -0400
summary:     Automated merge with ssh://hg at bitbucket.org/tarek/distutils2
files:       

diff --git a/src/distutils2/converter/fixers/fix_imports.py b/src/distutils2/converter/fixers/fix_imports.py
--- a/src/distutils2/converter/fixers/fix_imports.py
+++ b/src/distutils2/converter/fixers/fix_imports.py
@@ -36,11 +36,16 @@
             pattern = []
             next = imp.next_sibling
             while next is not None:
+                # Get the first child if we have a Node
+                if not hasattr(next, "value"):
+                    next = next.children[0]
                 pattern.append(next.value)
                 if not hasattr(next, "next_sibling"):
                     next.next_sibling = next.get_next_sibling()
                 next = next.next_sibling
-            if pattern == ['import', 'setup']:
+            
+            if set(pattern).issubset(set(
+                    ['import', ',', 'setup', 'find_packages'])):
                 imp.value = 'distutils2.core'
                 imp.changed()
 
diff --git a/src/distutils2/tests/conversions/05_after.py b/src/distutils2/tests/conversions/05_after.py
new file mode 100644
--- /dev/null
+++ b/src/distutils2/tests/conversions/05_after.py
@@ -0,0 +1,137 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+#
+# Copyright (C) 2003-2009 Edgewall Software
+# All rights reserved.
+#
+# This software is licensed as described in the file COPYING, which
+# you should have received as part of this distribution. The terms
+# are also available at http://trac.edgewall.org/wiki/TracLicense.
+#
+# This software consists of voluntary contributions made by many
+# individuals. For the exact contribution history, see the revision
+# history and logs, available at http://trac.edgewall.org/log/.
+
+from distutils2.core import setup, find_packages
+
+extra = {}
+
+try:
+    import babel
+    
+    extractors = [
+        ('**.py',                'python', None),
+        ('**/templates/**.html', 'genshi', None),
+        ('**/templates/**.txt',  'genshi',
+         {'template_class': 'genshi.template:NewTextTemplate'}),
+    ]
+    extra['message_extractors'] = {
+        'trac': extractors,
+        'tracopt': extractors,
+    }
+
+    from trac.util.dist import get_l10n_js_cmdclass
+    extra['cmdclass'] = get_l10n_js_cmdclass()
+
+except ImportError, e:
+    pass
+
+setup(
+    name = 'Trac',
+    version = '0.12.1',
+    summary = 'Integrated SCM, wiki, issue tracker and project environment',
+    description = """
+Trac is a minimalistic web-based software project management and bug/issue
+tracking system. It provides an interface to the Subversion revision control
+systems, an integrated wiki, flexible issue tracking and convenient report
+facilities.
+""",
+    author = 'Edgewall Software',
+    author_email = 'info at edgewall.com',
+    license = 'BSD',
+    home_page = 'http://trac.edgewall.org/',
+    download_url = 'http://trac.edgewall.org/wiki/TracDownload',
+    classifiers = [
+        'Environment :: Web Environment',
+        'Framework :: Trac',
+        'Intended Audience :: Developers',
+        'License :: OSI Approved :: BSD License',
+        'Operating System :: OS Independent',
+        'Programming Language :: Python',
+        'Topic :: Software Development :: Bug Tracking',
+        'Topic :: Software Development :: Version Control',
+    ],
+
+    packages = find_packages(exclude=['*.tests']),
+    package_data = {
+        '': ['templates/*'],
+        'trac': ['htdocs/*.*', 'htdocs/README', 'htdocs/js/*.*',
+                 'htdocs/js/messages/*.*', 'htdocs/css/*.*',
+                 'htdocs/guide/*', 'locale/*/LC_MESSAGES/messages.mo'],
+        'trac.wiki': ['default-pages/*'],
+        'trac.ticket': ['workflows/*.ini'],
+    },
+
+    test_suite = 'trac.test.suite',
+    zip_safe = True,
+
+    requires_dist = [
+        'setuptools>=0.6b1',
+        'Genshi>=0.6',
+    ],
+    extras_require = {
+        'Babel': ['Babel>=0.9.5'],
+        'Pygments': ['Pygments>=0.6'],
+        'reST': ['docutils>=0.3'],
+        'SilverCity': ['SilverCity>=0.9.4'],
+        'Textile': ['textile>=2.0'],
+    },
+
+    entry_points = """
+        [console_scripts]
+        trac-admin = trac.admin.console:run
+        tracd = trac.web.standalone:main
+
+        [trac.plugins]
+        trac.about = trac.about
+        trac.admin.console = trac.admin.console
+        trac.admin.web_ui = trac.admin.web_ui
+        trac.attachment = trac.attachment
+        trac.db.mysql = trac.db.mysql_backend
+        trac.db.postgres = trac.db.postgres_backend
+        trac.db.sqlite = trac.db.sqlite_backend
+        trac.mimeview.patch = trac.mimeview.patch
+        trac.mimeview.pygments = trac.mimeview.pygments[Pygments]
+        trac.mimeview.rst = trac.mimeview.rst[reST]
+        trac.mimeview.silvercity = trac.mimeview.silvercity[SilverCity]
+        trac.mimeview.txtl = trac.mimeview.txtl[Textile]
+        trac.prefs = trac.prefs.web_ui
+        trac.search = trac.search.web_ui
+        trac.ticket.admin = trac.ticket.admin
+        trac.ticket.query = trac.ticket.query
+        trac.ticket.report = trac.ticket.report
+        trac.ticket.roadmap = trac.ticket.roadmap
+        trac.ticket.web_ui = trac.ticket.web_ui
+        trac.timeline = trac.timeline.web_ui
+        trac.versioncontrol.admin = trac.versioncontrol.admin
+        trac.versioncontrol.svn_authz = trac.versioncontrol.svn_authz
+        trac.versioncontrol.svn_fs = trac.versioncontrol.svn_fs
+        trac.versioncontrol.svn_prop = trac.versioncontrol.svn_prop
+        trac.versioncontrol.web_ui = trac.versioncontrol.web_ui
+        trac.web.auth = trac.web.auth
+        trac.web.session = trac.web.session
+        trac.wiki.admin = trac.wiki.admin
+        trac.wiki.interwiki = trac.wiki.interwiki
+        trac.wiki.macros = trac.wiki.macros
+        trac.wiki.web_ui = trac.wiki.web_ui
+        trac.wiki.web_api = trac.wiki.web_api
+        tracopt.mimeview.enscript = tracopt.mimeview.enscript
+        tracopt.mimeview.php = tracopt.mimeview.php
+        tracopt.perm.authz_policy = tracopt.perm.authz_policy
+        tracopt.perm.config_perm_provider = tracopt.perm.config_perm_provider
+        tracopt.ticket.commit_updater = tracopt.ticket.commit_updater
+        tracopt.ticket.deleter = tracopt.ticket.deleter
+    """,
+
+    **extra
+)
diff --git a/src/distutils2/tests/conversions/05_before.py b/src/distutils2/tests/conversions/05_before.py
new file mode 100755
--- /dev/null
+++ b/src/distutils2/tests/conversions/05_before.py
@@ -0,0 +1,137 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+#
+# Copyright (C) 2003-2009 Edgewall Software
+# All rights reserved.
+#
+# This software is licensed as described in the file COPYING, which
+# you should have received as part of this distribution. The terms
+# are also available at http://trac.edgewall.org/wiki/TracLicense.
+#
+# This software consists of voluntary contributions made by many
+# individuals. For the exact contribution history, see the revision
+# history and logs, available at http://trac.edgewall.org/log/.
+
+from setuptools import setup, find_packages
+
+extra = {}
+
+try:
+    import babel
+    
+    extractors = [
+        ('**.py',                'python', None),
+        ('**/templates/**.html', 'genshi', None),
+        ('**/templates/**.txt',  'genshi',
+         {'template_class': 'genshi.template:NewTextTemplate'}),
+    ]
+    extra['message_extractors'] = {
+        'trac': extractors,
+        'tracopt': extractors,
+    }
+
+    from trac.util.dist import get_l10n_js_cmdclass
+    extra['cmdclass'] = get_l10n_js_cmdclass()
+
+except ImportError, e:
+    pass
+
+setup(
+    name = 'Trac',
+    version = '0.12.1',
+    description = 'Integrated SCM, wiki, issue tracker and project environment',
+    long_description = """
+Trac is a minimalistic web-based software project management and bug/issue
+tracking system. It provides an interface to the Subversion revision control
+systems, an integrated wiki, flexible issue tracking and convenient report
+facilities.
+""",
+    author = 'Edgewall Software',
+    author_email = 'info at edgewall.com',
+    license = 'BSD',
+    url = 'http://trac.edgewall.org/',
+    download_url = 'http://trac.edgewall.org/wiki/TracDownload',
+    classifiers = [
+        'Environment :: Web Environment',
+        'Framework :: Trac',
+        'Intended Audience :: Developers',
+        'License :: OSI Approved :: BSD License',
+        'Operating System :: OS Independent',
+        'Programming Language :: Python',
+        'Topic :: Software Development :: Bug Tracking',
+        'Topic :: Software Development :: Version Control',
+    ],
+
+    packages = find_packages(exclude=['*.tests']),
+    package_data = {
+        '': ['templates/*'],
+        'trac': ['htdocs/*.*', 'htdocs/README', 'htdocs/js/*.*',
+                 'htdocs/js/messages/*.*', 'htdocs/css/*.*',
+                 'htdocs/guide/*', 'locale/*/LC_MESSAGES/messages.mo'],
+        'trac.wiki': ['default-pages/*'],
+        'trac.ticket': ['workflows/*.ini'],
+    },
+
+    test_suite = 'trac.test.suite',
+    zip_safe = True,
+
+    install_requires = [
+        'setuptools>=0.6b1',
+        'Genshi>=0.6',
+    ],
+    extras_require = {
+        'Babel': ['Babel>=0.9.5'],
+        'Pygments': ['Pygments>=0.6'],
+        'reST': ['docutils>=0.3'],
+        'SilverCity': ['SilverCity>=0.9.4'],
+        'Textile': ['textile>=2.0'],
+    },
+
+    entry_points = """
+        [console_scripts]
+        trac-admin = trac.admin.console:run
+        tracd = trac.web.standalone:main
+
+        [trac.plugins]
+        trac.about = trac.about
+        trac.admin.console = trac.admin.console
+        trac.admin.web_ui = trac.admin.web_ui
+        trac.attachment = trac.attachment
+        trac.db.mysql = trac.db.mysql_backend
+        trac.db.postgres = trac.db.postgres_backend
+        trac.db.sqlite = trac.db.sqlite_backend
+        trac.mimeview.patch = trac.mimeview.patch
+        trac.mimeview.pygments = trac.mimeview.pygments[Pygments]
+        trac.mimeview.rst = trac.mimeview.rst[reST]
+        trac.mimeview.silvercity = trac.mimeview.silvercity[SilverCity]
+        trac.mimeview.txtl = trac.mimeview.txtl[Textile]
+        trac.prefs = trac.prefs.web_ui
+        trac.search = trac.search.web_ui
+        trac.ticket.admin = trac.ticket.admin
+        trac.ticket.query = trac.ticket.query
+        trac.ticket.report = trac.ticket.report
+        trac.ticket.roadmap = trac.ticket.roadmap
+        trac.ticket.web_ui = trac.ticket.web_ui
+        trac.timeline = trac.timeline.web_ui
+        trac.versioncontrol.admin = trac.versioncontrol.admin
+        trac.versioncontrol.svn_authz = trac.versioncontrol.svn_authz
+        trac.versioncontrol.svn_fs = trac.versioncontrol.svn_fs
+        trac.versioncontrol.svn_prop = trac.versioncontrol.svn_prop
+        trac.versioncontrol.web_ui = trac.versioncontrol.web_ui
+        trac.web.auth = trac.web.auth
+        trac.web.session = trac.web.session
+        trac.wiki.admin = trac.wiki.admin
+        trac.wiki.interwiki = trac.wiki.interwiki
+        trac.wiki.macros = trac.wiki.macros
+        trac.wiki.web_ui = trac.wiki.web_ui
+        trac.wiki.web_api = trac.wiki.web_api
+        tracopt.mimeview.enscript = tracopt.mimeview.enscript
+        tracopt.mimeview.php = tracopt.mimeview.php
+        tracopt.perm.authz_policy = tracopt.perm.authz_policy
+        tracopt.perm.config_perm_provider = tracopt.perm.config_perm_provider
+        tracopt.ticket.commit_updater = tracopt.ticket.commit_updater
+        tracopt.ticket.deleter = tracopt.ticket.deleter
+    """,
+
+    **extra
+)
diff --git a/src/distutils2/tests/test_version.py b/src/distutils2/tests/test_version.py
--- a/src/distutils2/tests/test_version.py
+++ b/src/distutils2/tests/test_version.py
@@ -3,7 +3,7 @@
 import os
 
 from distutils2.version import NormalizedVersion as V
-from distutils2.version import IrrationalVersionError
+from distutils2.version import HugeMajorVersionNumError, IrrationalVersionError
 from distutils2.version import suggest_normalized_version as suggest
 from distutils2.version import VersionPredicate
 from distutils2.tests.support import unittest
@@ -22,6 +22,10 @@
                 (V('1.0.dev345'), '1.0.dev345'),
                 (V('1.0.post456.dev623'), '1.0.post456.dev623'))
 
+    def test_repr(self):
+
+        self.assertEqual(repr(V('1.0')), "NormalizedVersion('1.0')")
+
     def test_basic_versions(self):
 
         for v, s in self.versions:
@@ -44,6 +48,12 @@
         for s in irrational:
             self.assertRaises(IrrationalVersionError, V, s)
 
+    def test_huge_version(self):
+
+        self.assertEquals(str(V('1980.0')), '1980.0')
+        self.assertRaises(HugeMajorVersionNumError, V, '1981.0')
+        self.assertEquals(str(V('1981.0', error_on_huge_major_num=False)), '1981.0')
+
     def test_comparison(self):
         r"""
         >>> V('1.2.0') == '1.2'
@@ -51,12 +61,33 @@
         ...
         TypeError: cannot compare NormalizedVersion and str
 
+        >>> V('1.2') < '1.3'
+        Traceback (most recent call last):
+        ...
+        TypeError: cannot compare NormalizedVersion and str
+
         >>> V('1.2.0') == V('1.2')
         True
         >>> V('1.2.0') == V('1.2.3')
         False
+        >>> V('1.2.0') != V('1.2.3')
+        True
         >>> V('1.2.0') < V('1.2.3')
         True
+        >>> V('1.2.0') < V('1.2.0')
+        False
+        >>> V('1.2.0') <= V('1.2.0')
+        True
+        >>> V('1.2.0') <= V('1.2.3')
+        True
+        >>> V('1.2.3') <= V('1.2.0')
+        False
+        >>> V('1.2.0') >= V('1.2.0')
+        True
+        >>> V('1.2.3') >= V('1.2.0')
+        True
+        >>> V('1.2.0') >= V('1.2.3')
+        False
         >>> (V('1.0') > V('1.0b2'))
         True
         >>> (V('1.0') > V('1.0c2') > V('1.0c1') > V('1.0b2') > V('1.0b1')
@@ -101,6 +132,7 @@
         self.assertEqual(suggest('1.0c2'), '1.0c2')
         self.assertEqual(suggest('walla walla washington'), None)
         self.assertEqual(suggest('2.4c1'), '2.4c1')
+        self.assertEqual(suggest('v1.0'), '1.0')
 
         # from setuptools
         self.assertEqual(suggest('0.4a1.r10'), '0.4a1.post10')
@@ -151,6 +183,8 @@
         self.assertFalse(VersionPredicate('Hey (<=2.5)').match('2.6.0'))
         self.assertTrue(VersionPredicate('Hey (>=2.5)').match('2.5.1'))
 
+        self.assertRaises(ValueError, VersionPredicate, '')
+
         # XXX need to silent the micro version in this case
         #assert not VersionPredicate('Ho (<3.0,!=2.6)').match('2.6.3')
 
@@ -164,6 +198,11 @@
         for version in other_versions:
             self.assertFalse(V(version).is_final)
 
+    def test_parse_numdots(self):
+        # For code coverage completeness
+        self.assertEquals(V('1.0')._parse_numdots('1.0', '1.0', pad_zeros_length=3),
+                          [1, 0, 0])
+
 
 def test_suite():
     #README = os.path.join(os.path.dirname(__file__), 'README.txt')
diff --git a/src/runtests-cov.py b/src/runtests-cov.py
new file mode 100755
--- /dev/null
+++ b/src/runtests-cov.py
@@ -0,0 +1,105 @@
+#!/usr/bin/env python
+"""Tests for distutils2.
+
+The tests for distutils2 are defined in the distutils2.tests package.
+"""
+
+# TODO:
+
+# The coverage report is only accurate when ran inside a virtualenv
+# created with the --no-site-packages option.  When it's not the case,
+# the built-in ignore list is not accurate and third party packages
+# show-up in the report, lowering the overall coverage.  
+
+# One particular problem it docutils on Ubuntu which has a __file__
+# starting with /usr/lib/python2.6 while the path in the coverage
+# report starts with /usr/share/pyshared.
+
+import sys
+from os.path import dirname
+from optparse import OptionParser
+
+def parse_opts():
+    parser = OptionParser(usage="%prog [OPTIONS]", 
+                          description="run the distutils2 unittests")
+    
+    parser.add_option("-q", "--quiet", help="do not print verbose messages", 
+                      action="store_true", default=False)
+    parser.add_option("-c", "--coverage", action="store_true", default=False,
+                      help="produce a coverage report at the end of the run")
+    parser.add_option("-r", "--report", action="store_true", default=False,
+                      help="produce a coverage report from the last test run")
+    parser.add_option("-m", "--show-missing", action="store_true", 
+                      default=False,
+                      help=("Show line numbers of statements in each module "
+                            "that weren't executed."))
+    
+    opts, args = parser.parse_args()
+    return opts, args
+
+def coverage_report(opts):
+    import coverage
+    import unittest2
+    import docutils
+    cov = coverage.coverage()
+    cov.load()
+
+    cov.report(omit_prefixes=["distutils2/tests", 
+                              "runtests", 
+                              "distutils2/_backport", 
+                              dirname(unittest2.__file__),
+                              dirname(dirname(docutils.__file__))], 
+               show_missing=opts.show_missing)
+
+
+def test_main():
+    opts, args = parse_opts()
+    verbose = not opts.quiet
+    ret = 0
+    
+    if opts.coverage or opts.report:
+        import coverage
+        
+    if opts.coverage:
+        cov = coverage.coverage()
+        cov.erase()
+        cov.start()
+    if not opts.report:
+        ret = run_tests(verbose)
+    if opts.coverage:
+        cov.stop()
+        cov.save()
+
+    if opts.report or opts.coverage:
+        coverage_report(opts)
+    
+    return ret
+    
+def run_tests(verbose):
+    import distutils2.tests
+    from distutils2.tests import run_unittest, reap_children, TestFailed
+    from distutils2._backport.tests import test_suite as btest_suite
+    # XXX just supporting -q right now to enable detailed/quiet output
+    if len(sys.argv) > 1:
+        verbose = sys.argv[-1] != '-q'
+    else:
+        verbose = 1
+    try:
+        try:
+            run_unittest([distutils2.tests.test_suite(), btest_suite()],
+                         verbose_=verbose)
+            return 0
+        except TestFailed:
+            return 1
+    finally:
+        reap_children()
+
+if __name__ == "__main__":
+    try:
+        from distutils2.tests.support import unittest
+    except ImportError:
+        sys.stderr.write('Error: You have to install unittest2')
+        sys.exit(1)
+
+    sys.exit(test_main())
+

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul 24 23:45:28 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 24 Jul 2010 23:45:28 +0200
Subject: [Python-checkins] distutils2: Automated merge
	with	ssh://hg@bitbucket.org/mtlpython/distutils2
Message-ID: <E1OcmXA-0006P6-BL@dinsdale.python.org>

tarek.ziade pushed da7378fa803a to distutils2:

http://hg.python.org/distutils2/rev/da7378fa803a
changeset:   379:da7378fa803a
parent:      376:e781043ca958
parent:      378:444578131d2e
user:        Yannick Gingras <ygingras at ygingras.net>
date:        Fri Jul 16 01:38:37 2010 -0400
summary:     Automated merge with ssh://hg at bitbucket.org/mtlpython/distutils2
files:       

diff --git a/.hgignore b/.hgignore
--- a/.hgignore
+++ b/.hgignore
@@ -6,3 +6,4 @@
 MANIFEST
 dist/
 *.swp
+.coverage
\ No newline at end of file
diff --git a/src/runtests-cov.py b/src/runtests-cov.py
--- a/src/runtests-cov.py
+++ b/src/runtests-cov.py
@@ -4,20 +4,23 @@
 The tests for distutils2 are defined in the distutils2.tests package.
 """
 
-# TODO:
+import sys
+from os.path import dirname, islink, realpath
+from optparse import OptionParser
 
-# The coverage report is only accurate when ran inside a virtualenv
-# created with the --no-site-packages option.  When it's not the case,
-# the built-in ignore list is not accurate and third party packages
-# show-up in the report, lowering the overall coverage.  
-
-# One particular problem it docutils on Ubuntu which has a __file__
-# starting with /usr/lib/python2.6 while the path in the coverage
-# report starts with /usr/share/pyshared.
-
-import sys
-from os.path import dirname
-from optparse import OptionParser
+def ignore_prefixes(module):
+    """ Return a list of prefixes to ignore in the coverage report if
+    we want to completely skip `module`.
+    """
+    # A function like that is needed because some GNU/Linux
+    # distributions, such a Ubuntu, really like to build link farm in
+    # /usr/lib in order to save a few bytes on the disk.
+    dirnames = [dirname(module.__file__)]
+    
+    pymod = module.__file__.rstrip("c")
+    if islink(pymod):
+        dirnames.append(dirname(realpath(pymod)))
+    return dirnames
 
 def parse_opts():
     parser = OptionParser(usage="%prog [OPTIONS]", 
@@ -40,16 +43,27 @@
 def coverage_report(opts):
     import coverage
     import unittest2
-    import docutils
     cov = coverage.coverage()
     cov.load()
 
-    cov.report(omit_prefixes=["distutils2/tests", 
-                              "runtests", 
-                              "distutils2/_backport", 
-                              dirname(unittest2.__file__),
-                              dirname(dirname(docutils.__file__))], 
-               show_missing=opts.show_missing)
+    prefixes = ["runtests", "distutils2/tests", "distutils2/_backport"]
+    prefixes += ignore_prefixes(unittest2)
+
+    try:
+        import docutils
+        prefixes += ignore_prefixes(docutils)
+    except ImportError:
+        # that module is completely optional
+        pass
+
+    try:
+        import roman
+        prefixes += ignore_prefixes(roman)
+    except ImportError:
+        # that module is also completely optional
+        pass
+
+    cov.report(omit_prefixes=prefixes, show_missing=opts.show_missing)
 
 
 def test_main():

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul 24 23:45:28 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 24 Jul 2010 23:45:28 +0200
Subject: [Python-checkins] distutils2: made sure that the coverage report is
	accurate even when	ran outsite of a venv
Message-ID: <E1OcmXA-0006Oy-AM@dinsdale.python.org>

tarek.ziade pushed 39220974a1e4 to distutils2:

http://hg.python.org/distutils2/rev/39220974a1e4
changeset:   377:39220974a1e4
parent:      375:fdc0f44dc5ec
user:        Yannick Gingras <ygingras at ygingras.net>
date:        Fri Jul 16 01:33:03 2010 -0400
summary:     made sure that the coverage report is accurate even when ran outsite of a venv
files:       src/runtests-cov.py

diff --git a/src/runtests-cov.py b/src/runtests-cov.py
--- a/src/runtests-cov.py
+++ b/src/runtests-cov.py
@@ -4,20 +4,23 @@
 The tests for distutils2 are defined in the distutils2.tests package.
 """
 
-# TODO:
+import sys
+from os.path import dirname, islink, realpath
+from optparse import OptionParser
 
-# The coverage report is only accurate when ran inside a virtualenv
-# created with the --no-site-packages option.  When it's not the case,
-# the built-in ignore list is not accurate and third party packages
-# show-up in the report, lowering the overall coverage.  
-
-# One particular problem it docutils on Ubuntu which has a __file__
-# starting with /usr/lib/python2.6 while the path in the coverage
-# report starts with /usr/share/pyshared.
-
-import sys
-from os.path import dirname
-from optparse import OptionParser
+def ignore_prefixes(module):
+    """ Return a list of prefixes to ignore in the coverage report if
+    we want to completely skip `module`.
+    """
+    # A function like that is needed because some GNU/Linux
+    # distributions, such a Ubuntu, really like to build link farm in
+    # /usr/lib in order to save a few bytes on the disk.
+    dirnames = [dirname(module.__file__)]
+    
+    pymod = module.__file__.rstrip("c")
+    if islink(pymod):
+        dirnames.append(dirname(realpath(pymod)))
+    return dirnames
 
 def parse_opts():
     parser = OptionParser(usage="%prog [OPTIONS]", 
@@ -40,16 +43,27 @@
 def coverage_report(opts):
     import coverage
     import unittest2
-    import docutils
     cov = coverage.coverage()
     cov.load()
 
-    cov.report(omit_prefixes=["distutils2/tests", 
-                              "runtests", 
-                              "distutils2/_backport", 
-                              dirname(unittest2.__file__),
-                              dirname(dirname(docutils.__file__))], 
-               show_missing=opts.show_missing)
+    prefixes = ["runtests", "distutils2/tests", "distutils2/_backport"]
+    prefixes += ignore_prefixes(unittest2)
+
+    try:
+        import docutils
+        prefixes += ignore_prefixes(docutils)
+    except ImportError:
+        # that module is completely optional
+        pass
+
+    try:
+        import roman
+        prefixes += ignore_prefixes(roman)
+    except ImportError:
+        # that module is also completely optional
+        pass
+
+    cov.report(omit_prefixes=prefixes, show_missing=opts.show_missing)
 
 
 def test_main():

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul 24 23:45:28 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 24 Jul 2010 23:45:28 +0200
Subject: [Python-checkins] distutils2: improved tests.sh so .so files are
	removed on failure
Message-ID: <E1OcmXA-0006PM-EU@dinsdale.python.org>

tarek.ziade pushed edf4d7ededd8 to distutils2:

http://hg.python.org/distutils2/rev/edf4d7ededd8
changeset:   383:edf4d7ededd8
tag:         tip
user:        Tarek Ziade <tarek at ziade.org>
date:        Sat Jul 24 23:43:31 2010 +0200
summary:     improved tests.sh so .so files are removed on failure
files:       src/tests.sh

diff --git a/src/tests.sh b/src/tests.sh
--- a/src/tests.sh
+++ b/src/tests.sh
@@ -1,8 +1,9 @@
 #!/bin/sh
 echo -n "Running tests for Python 2.4... "
-rm -rf _hashlib.so
+rm -rf *.so
 python2.4 setup.py build_ext -i -q 2> /dev/null > /dev/null
 python2.4 -Wd runtests.py -q 2> /dev/null
+rm -rf *.so
 if [ $? -ne 0 ];then
     echo "Failed"
     exit 1
@@ -11,9 +12,9 @@
 fi
 
 echo -n "Running tests for Python 2.5... "
-rm -rf _hashlib.so
 python2.5 setup.py build_ext -i -q 2> /dev/null > /dev/null
 python2.5 -Wd runtests.py -q 2> /dev/null
+rm -rf *.so
 if [ $? -ne 0 ];then
     echo "Failed"
     exit 1

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul 24 23:45:28 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 24 Jul 2010 23:45:28 +0200
Subject: [Python-checkins] distutils2: Fix import errors found by pyflakes
Message-ID: <E1OcmXA-0006PI-Dj@dinsdale.python.org>

tarek.ziade pushed 733c3bdaa25e to distutils2:

http://hg.python.org/distutils2/rev/733c3bdaa25e
changeset:   382:733c3bdaa25e
user:        ?ric Araujo <merwok at netwok.org>
date:        Sun Jul 18 03:30:10 2010 +0200
summary:     Fix import errors found by pyflakes
files:       src/distutils2/command/bdist_dumb.py, src/distutils2/command/bdist_wininst.py, src/distutils2/command/build_py.py, src/distutils2/command/cmd.py, src/distutils2/command/sdist.py, src/distutils2/command/upload_docs.py, src/distutils2/extension.py, src/distutils2/tests/test_build_ext.py, src/distutils2/tests/test_depgraph.py, src/distutils2/tests/test_manifest.py, src/distutils2/tests/test_metadata.py, src/distutils2/tests/test_pypi_versions.py, src/distutils2/version.py, src/setup.py

diff --git a/src/distutils2/command/bdist_dumb.py b/src/distutils2/command/bdist_dumb.py
--- a/src/distutils2/command/bdist_dumb.py
+++ b/src/distutils2/command/bdist_dumb.py
@@ -115,8 +115,9 @@
                        % (repr(install.install_base),
                           repr(install.install_platbase)))
             else:
-                archive_root = os.path.join(self.bdist_dir,
-                                   ensure_relative(install.install_base))
+                archive_root = os.path.join(
+                    self.bdist_dir,
+                    self._ensure_relative(install.install_base))
 
         # Make the archive
         filename = self.make_archive(pseudoinstall_root,
@@ -135,3 +136,9 @@
             else:
                 rmtree(self.bdist_dir)
 
+    def _ensure_relative(self, path):
+        # copied from dir_util, deleted
+        drive, path = os.path.splitdrive(path)
+        if path[0:1] == os.sep:
+            path = drive + path[1:]
+        return path
diff --git a/src/distutils2/command/bdist_wininst.py b/src/distutils2/command/bdist_wininst.py
--- a/src/distutils2/command/bdist_wininst.py
+++ b/src/distutils2/command/bdist_wininst.py
@@ -8,6 +8,7 @@
 import sys
 import os
 import string
+from shutil import rmtree
 try:
     from sysconfig import get_python_version
 except ImportError:
diff --git a/src/distutils2/command/build_py.py b/src/distutils2/command/build_py.py
--- a/src/distutils2/command/build_py.py
+++ b/src/distutils2/command/build_py.py
@@ -9,6 +9,7 @@
 import logging
 from glob import glob
 
+import distutils2
 from distutils2.core import Command
 from distutils2.errors import DistutilsOptionError, DistutilsFileError
 from distutils2.util import convert_path
diff --git a/src/distutils2/command/cmd.py b/src/distutils2/command/cmd.py
--- a/src/distutils2/command/cmd.py
+++ b/src/distutils2/command/cmd.py
@@ -6,7 +6,7 @@
 
 __revision__ = "$Id: cmd.py 75192 2009-10-02 23:49:48Z tarek.ziade $"
 
-import sys, os, re
+import os, re
 from distutils2.errors import DistutilsOptionError
 from distutils2 import util
 from distutils2 import log
@@ -447,7 +447,7 @@
         # If 'outfile' must be regenerated (either because it doesn't
         # exist, is out-of-date, or the 'force' flag is true) then
         # perform the action that presumably regenerates it
-        if self.force or dep_util.newer_group(infiles, outfile):
+        if self.force or util.newer_group(infiles, outfile):
             self.execute(func, args, exec_msg, level)
 
         # Otherwise, print the "skip" message
diff --git a/src/distutils2/command/sdist.py b/src/distutils2/command/sdist.py
--- a/src/distutils2/command/sdist.py
+++ b/src/distutils2/command/sdist.py
@@ -18,12 +18,11 @@
     from distutils2._backport.shutil import get_archive_formats
 
 from distutils2.core import Command
-from distutils2 import util
 from distutils2.errors import (DistutilsPlatformError, DistutilsOptionError,
                                DistutilsTemplateError)
 from distutils2.manifest import Manifest
 from distutils2 import log
-from distutils2.util import convert_path, newer
+from distutils2.util import convert_path
 
 def show_formats():
     """Print all possible values for the 'formats' option (used by
diff --git a/src/distutils2/command/upload_docs.py b/src/distutils2/command/upload_docs.py
--- a/src/distutils2/command/upload_docs.py
+++ b/src/distutils2/command/upload_docs.py
@@ -1,4 +1,4 @@
-import base64, httplib, os.path, socket, tempfile, urlparse, zipfile
+import base64, httplib, os.path, socket, urlparse, zipfile
 from cStringIO import StringIO
 from distutils2 import log
 from distutils2.command.upload import upload
@@ -81,7 +81,6 @@
             raise DistutilsFileError(mesg % upload_dir)
 
     def run(self):
-        tmp_dir = tempfile.mkdtemp()
         name = self.distribution.metadata['Name']
         zip_file = zip_dir(self.upload_dir)
 
@@ -124,7 +123,7 @@
         elif r.status == 301:
             location = r.getheader('Location')
             if location is None:
-                location = 'http://packages.python.org/%s/' % meta.get_name()
+                location = 'http://packages.python.org/%s/' % name
             self.announce('Upload successful. Visit %s' % location,
                           log.INFO)
         else:
diff --git a/src/distutils2/extension.py b/src/distutils2/extension.py
--- a/src/distutils2/extension.py
+++ b/src/distutils2/extension.py
@@ -5,14 +5,8 @@
 
 __revision__ = "$Id: extension.py 77704 2010-01-23 09:23:15Z tarek.ziade $"
 
-import os
 import warnings
 
-try:
-    import sysconfig
-except ImportError:
-    from distutils2._backport import sysconfig
-
 # This class is really only used by the "build_ext" command, so it might
 # make sense to put it in distutils.command.build_ext.  However, that
 # module is already big enough, and I want to make this class a bit more
diff --git a/src/distutils2/tests/test_build_ext.py b/src/distutils2/tests/test_build_ext.py
--- a/src/distutils2/tests/test_build_ext.py
+++ b/src/distutils2/tests/test_build_ext.py
@@ -368,4 +368,4 @@
     else: return unittest.makeSuite(BuildExtTestCase)
 
 if __name__ == '__main__':
-    distsutils2.tests.run_unittest(test_suite())
+    distutils2.tests.run_unittest(test_suite())
diff --git a/src/distutils2/tests/test_depgraph.py b/src/distutils2/tests/test_depgraph.py
--- a/src/distutils2/tests/test_depgraph.py
+++ b/src/distutils2/tests/test_depgraph.py
@@ -10,7 +10,7 @@
 import re
 try:
     import cStringIO as StringIO
-except ImportErorr:
+except ImportError:
     import StringIO
 
 class DepGraphTestCase(support.LoggingSilencer,
diff --git a/src/distutils2/tests/test_manifest.py b/src/distutils2/tests/test_manifest.py
--- a/src/distutils2/tests/test_manifest.py
+++ b/src/distutils2/tests/test_manifest.py
@@ -3,6 +3,7 @@
 import sys
 import logging
 
+from distutils2.tests import run_unittest
 from distutils2.tests import support
 from distutils2.tests.support import unittest
 from distutils2.manifest import Manifest
diff --git a/src/distutils2/tests/test_metadata.py b/src/distutils2/tests/test_metadata.py
--- a/src/distutils2/tests/test_metadata.py
+++ b/src/distutils2/tests/test_metadata.py
@@ -5,6 +5,7 @@
 
 from distutils2.metadata import (DistributionMetadata, _interpret,
                                  PKG_INFO_PREFERRED_VERSION)
+from distutils2.tests import run_unittest
 from distutils2.tests.support import unittest, LoggingSilencer
 from distutils2.errors import (MetadataConflictError,
                                MetadataUnrecognizedVersionError)
diff --git a/src/distutils2/tests/test_pypi_versions.py b/src/distutils2/tests/test_pypi_versions.py
--- a/src/distutils2/tests/test_pypi_versions.py
+++ b/src/distutils2/tests/test_pypi_versions.py
@@ -19,6 +19,7 @@
 import os.path
 
 from distutils2.version import suggest_normalized_version
+from distutils2.tests import run_unittest
 from distutils2.tests.support import unittest
 
 def test_pypi():
@@ -52,7 +53,7 @@
         print "Saving package info..."
         f = open(INDEX_PICKLE_FILE, 'wb')
         try:
-            pickle.dump(package_info, o)
+            pickle.dump(package_info, f)
         finally:
             f.close()
 
diff --git a/src/distutils2/version.py b/src/distutils2/version.py
--- a/src/distutils2/version.py
+++ b/src/distutils2/version.py
@@ -1,4 +1,3 @@
-import sys
 import re
 
 from distutils2.errors import IrrationalVersionError, HugeMajorVersionNumError
diff --git a/src/setup.py b/src/setup.py
--- a/src/setup.py
+++ b/src/setup.py
@@ -11,7 +11,6 @@
 from distutils2.compiler.ccompiler import new_compiler
 from distutils2.command.sdist import sdist
 from distutils2.command.install import install
-from distutils2 import __version__ as VERSION
 from distutils2.util import find_packages
 
 f = open('README.txt')

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul 24 23:45:28 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 24 Jul 2010 23:45:28 +0200
Subject: [Python-checkins] distutils2: added the coverage instrumentation
	log to .hgignore
Message-ID: <E1OcmXA-0006P1-Ak@dinsdale.python.org>

tarek.ziade pushed 444578131d2e to distutils2:

http://hg.python.org/distutils2/rev/444578131d2e
changeset:   378:444578131d2e
user:        Yannick Gingras <ygingras at ygingras.net>
date:        Fri Jul 16 01:33:36 2010 -0400
summary:     added the coverage instrumentation log to .hgignore
files:       .hgignore

diff --git a/.hgignore b/.hgignore
--- a/.hgignore
+++ b/.hgignore
@@ -6,3 +6,4 @@
 MANIFEST
 dist/
 *.swp
+.coverage
\ No newline at end of file

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sat Jul 24 23:47:54 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 24 Jul 2010 23:47:54 +0200
Subject: [Python-checkins] distutils2: added a simple pep8/flake checker
Message-ID: <E1OcmZW-0006Sv-3i@dinsdale.python.org>

tarek.ziade pushed dfe11edcb88d to distutils2:

http://hg.python.org/distutils2/rev/dfe11edcb88d
changeset:   384:dfe11edcb88d
tag:         tip
user:        Tarek Ziade <tarek at ziade.org>
date:        Sat Jul 24 23:47:46 2010 +0200
summary:     added a simple pep8/flake checker
files:       src/check.sh

diff --git a/src/check.sh b/src/check.sh
new file mode 100755
--- /dev/null
+++ b/src/check.sh
@@ -0,0 +1,2 @@
+pep8 distutils2
+pyflakes distutils2

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul 25 00:01:03 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 25 Jul 2010 00:01:03 +0200
Subject: [Python-checkins] distutils2: merged spawn into util
Message-ID: <E1OcmmF-0000hh-H6@dinsdale.python.org>

tarek.ziade pushed 23d074c6978d to distutils2:

http://hg.python.org/distutils2/rev/23d074c6978d
changeset:   385:23d074c6978d
tag:         tip
user:        Tarek Ziade <tarek at ziade.org>
date:        Sun Jul 25 00:00:54 2010 +0200
summary:     merged spawn into util
files:       src/distutils2/command/cmd.py, src/distutils2/command/upload.py, src/distutils2/compiler/ccompiler.py, src/distutils2/spawn.py, src/distutils2/tests/test_bdist.py, src/distutils2/tests/test_build_clib.py, src/distutils2/tests/test_sdist.py, src/distutils2/tests/test_spawn.py, src/distutils2/tests/test_util.py, src/distutils2/util.py

diff --git a/src/distutils2/command/cmd.py b/src/distutils2/command/cmd.py
--- a/src/distutils2/command/cmd.py
+++ b/src/distutils2/command/cmd.py
@@ -411,7 +411,7 @@
 
     def spawn(self, cmd, search_path=1, level=1):
         """Spawn an external command respecting dry-run flag."""
-        from distutils2.spawn import spawn
+        from distutils2.util import spawn
         spawn(cmd, search_path, dry_run= self.dry_run)
 
     def make_archive(self, base_name, format, root_dir=None, base_dir=None,
diff --git a/src/distutils2/command/upload.py b/src/distutils2/command/upload.py
--- a/src/distutils2/command/upload.py
+++ b/src/distutils2/command/upload.py
@@ -15,7 +15,7 @@
 
 from distutils2.errors import DistutilsOptionError
 from distutils2.core import PyPIRCCommand
-from distutils2.spawn import spawn
+from distutils2.util import spawn
 from distutils2 import log
 
 class upload(PyPIRCCommand):
diff --git a/src/distutils2/compiler/ccompiler.py b/src/distutils2/compiler/ccompiler.py
--- a/src/distutils2/compiler/ccompiler.py
+++ b/src/distutils2/compiler/ccompiler.py
@@ -11,8 +11,7 @@
 
 from distutils2.errors import (CompileError, LinkError, UnknownFileError,
                                DistutilsPlatformError, DistutilsModuleError)
-from distutils2.spawn import spawn
-from distutils2.util import split_quoted, execute, newer_group
+from distutils2.util import split_quoted, execute, newer_group, spawn
 from distutils2 import log
 from shutil import move
 
diff --git a/src/distutils2/spawn.py b/src/distutils2/spawn.py
deleted file mode 100644
--- a/src/distutils2/spawn.py
+++ /dev/null
@@ -1,192 +0,0 @@
-"""distutils.spawn
-
-Provides the 'spawn()' function, a front-end to various platform-
-specific functions for launching another program in a sub-process.
-Also provides the 'find_executable()' to search the path for a given
-executable name.
-"""
-
-__revision__ = "$Id: spawn.py 73147 2009-06-02 15:58:43Z tarek.ziade $"
-
-import sys
-import os
-
-from distutils2.errors import DistutilsPlatformError, DistutilsExecError
-from distutils2 import log
-
-def spawn(cmd, search_path=1, verbose=0, dry_run=0, env=None):
-    """Run another program, specified as a command list 'cmd', in a new process.
-
-    'cmd' is just the argument list for the new process, ie.
-    cmd[0] is the program to run and cmd[1:] are the rest of its arguments.
-    There is no way to run a program with a name different from that of its
-    executable.
-
-    If 'search_path' is true (the default), the system's executable
-    search path will be used to find the program; otherwise, cmd[0]
-    must be the exact path to the executable.  If 'dry_run' is true,
-    the command will not actually be run.
-
-    If 'env' is given, it's a environment dictionary used for the execution
-    environment.
-
-    Raise DistutilsExecError if running the program fails in any way; just
-    return on success.
-    """
-    if os.name == 'posix':
-        _spawn_posix(cmd, search_path, dry_run=dry_run, env=env)
-    elif os.name == 'nt':
-        _spawn_nt(cmd, search_path, dry_run=dry_run, env=env)
-    elif os.name == 'os2':
-        _spawn_os2(cmd, search_path, dry_run=dry_run, env=env)
-    else:
-        raise DistutilsPlatformError, \
-              "don't know how to spawn programs on platform '%s'" % os.name
-
-def _nt_quote_args(args):
-    """Quote command-line arguments for DOS/Windows conventions.
-
-    Just wraps every argument which contains blanks in double quotes, and
-    returns a new argument list.
-    """
-    # XXX this doesn't seem very robust to me -- but if the Windows guys
-    # say it'll work, I guess I'll have to accept it.  (What if an arg
-    # contains quotes?  What other magic characters, other than spaces,
-    # have to be escaped?  Is there an escaping mechanism other than
-    # quoting?)
-    for i, arg in enumerate(args):
-        if ' ' in arg:
-            args[i] = '"%s"' % arg
-    return args
-
-def _spawn_nt(cmd, search_path=1, verbose=0, dry_run=0, env=None):
-    executable = cmd[0]
-    cmd = _nt_quote_args(cmd)
-    if search_path:
-        # either we find one or it stays the same
-        executable = find_executable(executable) or executable
-    log.info(' '.join([executable] + cmd[1:]))
-    if not dry_run:
-        # spawn for NT requires a full path to the .exe
-        try:
-            if env is None:
-                rc = os.spawnv(os.P_WAIT, executable, cmd)
-            else:
-                rc = os.spawnve(os.P_WAIT, executable, cmd, env)
-
-        except OSError, exc:
-            # this seems to happen when the command isn't found
-            raise DistutilsExecError, \
-                  "command '%s' failed: %s" % (cmd[0], exc[-1])
-        if rc != 0:
-            # and this reflects the command running but failing
-            raise DistutilsExecError, \
-                  "command '%s' failed with exit status %d" % (cmd[0], rc)
-
-def _spawn_os2(cmd, search_path=1, verbose=0, dry_run=0, env=None):
-    executable = cmd[0]
-    if search_path:
-        # either we find one or it stays the same
-        executable = find_executable(executable) or executable
-    log.info(' '.join([executable] + cmd[1:]))
-    if not dry_run:
-        # spawnv for OS/2 EMX requires a full path to the .exe
-        try:
-            if env is None:
-                rc = os.spawnv(os.P_WAIT, executable, cmd)
-            else:
-                rc = os.spawnve(os.P_WAIT, executable, cmd, env)
-
-        except OSError, exc:
-            # this seems to happen when the command isn't found
-            raise DistutilsExecError, \
-                  "command '%s' failed: %s" % (cmd[0], exc[-1])
-        if rc != 0:
-            # and this reflects the command running but failing
-            log.debug("command '%s' failed with exit status %d" % (cmd[0], rc))
-            raise DistutilsExecError, \
-                  "command '%s' failed with exit status %d" % (cmd[0], rc)
-
-
-def _spawn_posix(cmd, search_path=1, verbose=0, dry_run=0, env=None):
-    log.info(' '.join(cmd))
-    if dry_run:
-        return
-
-    if env is None:
-        exec_fn = search_path and os.execvp or os.execv
-    else:
-        exec_fn = search_path and os.execvpe or os.execve
-
-    pid = os.fork()
-
-    if pid == 0:  # in the child
-        try:
-            if env is None:
-                exec_fn(cmd[0], cmd)
-            else:
-                exec_fn(cmd[0], cmd, env)
-        except OSError, e:
-            sys.stderr.write("unable to execute %s: %s\n" %
-                             (cmd[0], e.strerror))
-            os._exit(1)
-
-        sys.stderr.write("unable to execute %s for unknown reasons" % cmd[0])
-        os._exit(1)
-    else:   # in the parent
-        # Loop until the child either exits or is terminated by a signal
-        # (ie. keep waiting if it's merely stopped)
-        while 1:
-            try:
-                pid, status = os.waitpid(pid, 0)
-            except OSError, exc:
-                import errno
-                if exc.errno == errno.EINTR:
-                    continue
-                raise DistutilsExecError, \
-                      "command '%s' failed: %s" % (cmd[0], exc[-1])
-            if os.WIFSIGNALED(status):
-                raise DistutilsExecError, \
-                      "command '%s' terminated by signal %d" % \
-                      (cmd[0], os.WTERMSIG(status))
-
-            elif os.WIFEXITED(status):
-                exit_status = os.WEXITSTATUS(status)
-                if exit_status == 0:
-                    return   # hey, it succeeded!
-                else:
-                    raise DistutilsExecError, \
-                          "command '%s' failed with exit status %d" % \
-                          (cmd[0], exit_status)
-
-            elif os.WIFSTOPPED(status):
-                continue
-
-            else:
-                raise DistutilsExecError, \
-                      "unknown error executing '%s': termination status %d" % \
-                      (cmd[0], status)
-
-def find_executable(executable, path=None):
-    """Tries to find 'executable' in the directories listed in 'path'.
-
-    A string listing directories separated by 'os.pathsep'; defaults to
-    os.environ['PATH'].  Returns the complete filename or None if not found.
-    """
-    if path is None:
-        path = os.environ['PATH']
-    paths = path.split(os.pathsep)
-    base, ext = os.path.splitext(executable)
-
-    if (sys.platform == 'win32' or os.name == 'os2') and (ext != '.exe'):
-        executable = executable + '.exe'
-
-    if not os.path.isfile(executable):
-        for p in paths:
-            f = os.path.join(p, executable)
-            if os.path.isfile(f):
-                # the file exists, we have a shot at spawn working
-                return f
-        return None
-    else:
-        return executable
diff --git a/src/distutils2/tests/test_bdist.py b/src/distutils2/tests/test_bdist.py
--- a/src/distutils2/tests/test_bdist.py
+++ b/src/distutils2/tests/test_bdist.py
@@ -8,8 +8,7 @@
 from distutils2.command.bdist import bdist
 from distutils2.tests import support
 from distutils2.tests.support import unittest
-from distutils2.spawn import find_executable
-from distutils2 import spawn
+from distutils2.util import find_executable
 from distutils2.errors import DistutilsExecError
 
 class BuildTestCase(support.TempdirManager,
diff --git a/src/distutils2/tests/test_build_clib.py b/src/distutils2/tests/test_build_clib.py
--- a/src/distutils2/tests/test_build_clib.py
+++ b/src/distutils2/tests/test_build_clib.py
@@ -5,7 +5,7 @@
 from distutils2.command.build_clib import build_clib
 from distutils2.errors import DistutilsSetupError
 from distutils2.tests import support
-from distutils2.spawn import find_executable
+from distutils2.util import find_executable
 from distutils2.tests.support import unittest
 
 class BuildCLibTestCase(support.TempdirManager,
diff --git a/src/distutils2/tests/test_sdist.py b/src/distutils2/tests/test_sdist.py
--- a/src/distutils2/tests/test_sdist.py
+++ b/src/distutils2/tests/test_sdist.py
@@ -30,7 +30,7 @@
 from distutils2.tests.support import unittest
 from distutils2.tests.test_config import PyPIRCCommandTestCase
 from distutils2.errors import DistutilsExecError, DistutilsOptionError
-from distutils2.spawn import find_executable
+from distutils2.util import find_executable
 from distutils2.tests import support
 from distutils2.log import WARN
 try:
diff --git a/src/distutils2/tests/test_spawn.py b/src/distutils2/tests/test_spawn.py
deleted file mode 100644
--- a/src/distutils2/tests/test_spawn.py
+++ /dev/null
@@ -1,60 +0,0 @@
-"""Tests for distutils.spawn."""
-import os
-import time
-from distutils2.tests import captured_stdout
-
-from distutils2.spawn import _nt_quote_args
-from distutils2.spawn import spawn, find_executable
-from distutils2.errors import DistutilsExecError
-from distutils2.tests import support
-from distutils2.tests.support import unittest
-
-class SpawnTestCase(support.TempdirManager,
-                    support.LoggingSilencer,
-                    unittest.TestCase):
-
-    def test_nt_quote_args(self):
-
-        for (args, wanted) in ((['with space', 'nospace'],
-                                ['"with space"', 'nospace']),
-                               (['nochange', 'nospace'],
-                                ['nochange', 'nospace'])):
-            res = _nt_quote_args(args)
-            self.assertEqual(res, wanted)
-
-
-    @unittest.skipUnless(os.name in ('nt', 'posix'),
-                         'Runs only under posix or nt')
-    def test_spawn(self):
-        tmpdir = self.mkdtemp()
-
-        # creating something executable
-        # through the shell that returns 1
-        if os.name == 'posix':
-            exe = os.path.join(tmpdir, 'foo.sh')
-            self.write_file(exe, '#!/bin/sh\nexit 1')
-            os.chmod(exe, 0777)
-        else:
-            exe = os.path.join(tmpdir, 'foo.bat')
-            self.write_file(exe, 'exit 1')
-
-        os.chmod(exe, 0777)
-        self.assertRaises(DistutilsExecError, spawn, [exe])
-
-        # now something that works
-        if os.name == 'posix':
-            exe = os.path.join(tmpdir, 'foo.sh')
-            self.write_file(exe, '#!/bin/sh\nexit 0')
-            os.chmod(exe, 0777)
-        else:
-            exe = os.path.join(tmpdir, 'foo.bat')
-            self.write_file(exe, 'exit 0')
-
-        os.chmod(exe, 0777)
-        spawn([exe])  # should work without any error
-
-def test_suite():
-    return unittest.makeSuite(SpawnTestCase)
-
-if __name__ == "__main__":
-    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_util.py b/src/distutils2/tests/test_util.py
--- a/src/distutils2/tests/test_util.py
+++ b/src/distutils2/tests/test_util.py
@@ -7,15 +7,18 @@
 import tempfile
 import time
 
+from distutils2.tests import captured_stdout
+from distutils2.tests.support import unittest
 from distutils2.errors import (DistutilsPlatformError,
                                DistutilsByteCompileError,
-                               DistutilsFileError)
-
+                               DistutilsFileError,
+                               DistutilsExecError)
 from distutils2.util import (convert_path, change_root,
-                            check_environ, split_quoted, strtobool,
-                            rfc822_escape, get_compiler_versions,
-                            _find_exe_version, _MAC_OS_X_LD_VERSION,
-                            byte_compile, find_packages)
+                             check_environ, split_quoted, strtobool,
+                             rfc822_escape, get_compiler_versions,
+                             _find_exe_version, _MAC_OS_X_LD_VERSION,
+                             byte_compile, find_packages, spawn, find_executable,
+                             _nt_quote_args)
 from distutils2 import util
 from distutils2.tests import support
 from distutils2.tests.support import unittest
@@ -332,6 +335,46 @@
         file_handle.close()
         self.assertEquals(new_content, converted_content)
 
+    def test_nt_quote_args(self):
+
+        for (args, wanted) in ((['with space', 'nospace'],
+                                ['"with space"', 'nospace']),
+                               (['nochange', 'nospace'],
+                                ['nochange', 'nospace'])):
+            res = _nt_quote_args(args)
+            self.assertEqual(res, wanted)
+
+
+    @unittest.skipUnless(os.name in ('nt', 'posix'),
+                         'Runs only under posix or nt')
+    def test_spawn(self):
+        tmpdir = self.mkdtemp()
+
+        # creating something executable
+        # through the shell that returns 1
+        if os.name == 'posix':
+            exe = os.path.join(tmpdir, 'foo.sh')
+            self.write_file(exe, '#!/bin/sh\nexit 1')
+            os.chmod(exe, 0777)
+        else:
+            exe = os.path.join(tmpdir, 'foo.bat')
+            self.write_file(exe, 'exit 1')
+
+        os.chmod(exe, 0777)
+        self.assertRaises(DistutilsExecError, spawn, [exe])
+
+        # now something that works
+        if os.name == 'posix':
+            exe = os.path.join(tmpdir, 'foo.sh')
+            self.write_file(exe, '#!/bin/sh\nexit 0')
+            os.chmod(exe, 0777)
+        else:
+            exe = os.path.join(tmpdir, 'foo.bat')
+            self.write_file(exe, 'exit 0')
+
+        os.chmod(exe, 0777)
+        spawn([exe])  # should work without any error
+
 
 def test_suite():
     return unittest.makeSuite(UtilTestCase)
diff --git a/src/distutils2/util.py b/src/distutils2/util.py
--- a/src/distutils2/util.py
+++ b/src/distutils2/util.py
@@ -13,8 +13,7 @@
 from fnmatch import fnmatchcase
 
 from distutils2.errors import (DistutilsPlatformError, DistutilsFileError,
-                               DistutilsByteCompileError)
-from distutils2.spawn import spawn, find_executable
+                               DistutilsByteCompileError, DistutilsExecError)
 from distutils2 import log
 from distutils2._backport import sysconfig as _sysconfig
 
@@ -688,3 +687,181 @@
         """ Issues a call to util.run_2to3. """
         return run_2to3(files, doctests_only, self.fixer_names,
                         self.options, self.explicit)
+
+
+def spawn(cmd, search_path=1, verbose=0, dry_run=0, env=None):
+    """Run another program, specified as a command list 'cmd', in a new process.
+
+    'cmd' is just the argument list for the new process, ie.
+    cmd[0] is the program to run and cmd[1:] are the rest of its arguments.
+    There is no way to run a program with a name different from that of its
+    executable.
+
+    If 'search_path' is true (the default), the system's executable
+    search path will be used to find the program; otherwise, cmd[0]
+    must be the exact path to the executable.  If 'dry_run' is true,
+    the command will not actually be run.
+
+    If 'env' is given, it's a environment dictionary used for the execution
+    environment.
+
+    Raise DistutilsExecError if running the program fails in any way; just
+    return on success.
+    """
+    if os.name == 'posix':
+        _spawn_posix(cmd, search_path, dry_run=dry_run, env=env)
+    elif os.name == 'nt':
+        _spawn_nt(cmd, search_path, dry_run=dry_run, env=env)
+    elif os.name == 'os2':
+        _spawn_os2(cmd, search_path, dry_run=dry_run, env=env)
+    else:
+        raise DistutilsPlatformError, \
+              "don't know how to spawn programs on platform '%s'" % os.name
+
+def _nt_quote_args(args):
+    """Quote command-line arguments for DOS/Windows conventions.
+
+    Just wraps every argument which contains blanks in double quotes, and
+    returns a new argument list.
+    """
+    # XXX this doesn't seem very robust to me -- but if the Windows guys
+    # say it'll work, I guess I'll have to accept it.  (What if an arg
+    # contains quotes?  What other magic characters, other than spaces,
+    # have to be escaped?  Is there an escaping mechanism other than
+    # quoting?)
+    for i, arg in enumerate(args):
+        if ' ' in arg:
+            args[i] = '"%s"' % arg
+    return args
+
+def _spawn_nt(cmd, search_path=1, verbose=0, dry_run=0, env=None):
+    executable = cmd[0]
+    cmd = _nt_quote_args(cmd)
+    if search_path:
+        # either we find one or it stays the same
+        executable = find_executable(executable) or executable
+    log.info(' '.join([executable] + cmd[1:]))
+    if not dry_run:
+        # spawn for NT requires a full path to the .exe
+        try:
+            if env is None:
+                rc = os.spawnv(os.P_WAIT, executable, cmd)
+            else:
+                rc = os.spawnve(os.P_WAIT, executable, cmd, env)
+
+        except OSError, exc:
+            # this seems to happen when the command isn't found
+            raise DistutilsExecError, \
+                  "command '%s' failed: %s" % (cmd[0], exc[-1])
+        if rc != 0:
+            # and this reflects the command running but failing
+            raise DistutilsExecError, \
+                  "command '%s' failed with exit status %d" % (cmd[0], rc)
+
+def _spawn_os2(cmd, search_path=1, verbose=0, dry_run=0, env=None):
+    executable = cmd[0]
+    if search_path:
+        # either we find one or it stays the same
+        executable = find_executable(executable) or executable
+    log.info(' '.join([executable] + cmd[1:]))
+    if not dry_run:
+        # spawnv for OS/2 EMX requires a full path to the .exe
+        try:
+            if env is None:
+                rc = os.spawnv(os.P_WAIT, executable, cmd)
+            else:
+                rc = os.spawnve(os.P_WAIT, executable, cmd, env)
+
+        except OSError, exc:
+            # this seems to happen when the command isn't found
+            raise DistutilsExecError, \
+                  "command '%s' failed: %s" % (cmd[0], exc[-1])
+        if rc != 0:
+            # and this reflects the command running but failing
+            log.debug("command '%s' failed with exit status %d" % (cmd[0], rc))
+            raise DistutilsExecError, \
+                  "command '%s' failed with exit status %d" % (cmd[0], rc)
+
+
+def _spawn_posix(cmd, search_path=1, verbose=0, dry_run=0, env=None):
+    log.info(' '.join(cmd))
+    if dry_run:
+        return
+
+    if env is None:
+        exec_fn = search_path and os.execvp or os.execv
+    else:
+        exec_fn = search_path and os.execvpe or os.execve
+
+    pid = os.fork()
+
+    if pid == 0:  # in the child
+        try:
+            if env is None:
+                exec_fn(cmd[0], cmd)
+            else:
+                exec_fn(cmd[0], cmd, env)
+        except OSError, e:
+            sys.stderr.write("unable to execute %s: %s\n" %
+                             (cmd[0], e.strerror))
+            os._exit(1)
+
+        sys.stderr.write("unable to execute %s for unknown reasons" % cmd[0])
+        os._exit(1)
+    else:   # in the parent
+        # Loop until the child either exits or is terminated by a signal
+        # (ie. keep waiting if it's merely stopped)
+        while 1:
+            try:
+                pid, status = os.waitpid(pid, 0)
+            except OSError, exc:
+                import errno
+                if exc.errno == errno.EINTR:
+                    continue
+                raise DistutilsExecError, \
+                      "command '%s' failed: %s" % (cmd[0], exc[-1])
+            if os.WIFSIGNALED(status):
+                raise DistutilsExecError, \
+                      "command '%s' terminated by signal %d" % \
+                      (cmd[0], os.WTERMSIG(status))
+
+            elif os.WIFEXITED(status):
+                exit_status = os.WEXITSTATUS(status)
+                if exit_status == 0:
+                    return   # hey, it succeeded!
+                else:
+                    raise DistutilsExecError, \
+                          "command '%s' failed with exit status %d" % \
+                          (cmd[0], exit_status)
+
+            elif os.WIFSTOPPED(status):
+                continue
+
+            else:
+                raise DistutilsExecError, \
+                      "unknown error executing '%s': termination status %d" % \
+                      (cmd[0], status)
+
+def find_executable(executable, path=None):
+    """Tries to find 'executable' in the directories listed in 'path'.
+
+    A string listing directories separated by 'os.pathsep'; defaults to
+    os.environ['PATH'].  Returns the complete filename or None if not found.
+    """
+    if path is None:
+        path = os.environ['PATH']
+    paths = path.split(os.pathsep)
+    base, ext = os.path.splitext(executable)
+
+    if (sys.platform == 'win32' or os.name == 'os2') and (ext != '.exe'):
+        executable = executable + '.exe'
+
+    if not os.path.isfile(executable):
+        for p in paths:
+            f = os.path.join(p, executable)
+            if os.path.isfile(f):
+                # the file exists, we have a shot at spawn working
+                return f
+        return None
+    else:
+        return executable

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Sun Jul 25 00:05:25 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sun, 25 Jul 2010 00:05:25 +0200
Subject: [Python-checkins] distutils2: pep8-fy util.py
Message-ID: <E1OcmqT-0000rG-Ax@dinsdale.python.org>

tarek.ziade pushed 04327e2e55e5 to distutils2:

http://hg.python.org/distutils2/rev/04327e2e55e5
changeset:   386:04327e2e55e5
tag:         tip
user:        Tarek Ziade <tarek at ziade.org>
date:        Sun Jul 25 00:05:15 2010 +0200
summary:     pep8-fy util.py
files:       src/distutils2/util.py

diff --git a/src/distutils2/util.py b/src/distutils2/util.py
--- a/src/distutils2/util.py
+++ b/src/distutils2/util.py
@@ -690,7 +690,7 @@
 
 
 def spawn(cmd, search_path=1, verbose=0, dry_run=0, env=None):
-    """Run another program, specified as a command list 'cmd', in a new process.
+    """Run another program specified as a command list 'cmd' in a new process.
 
     'cmd' is just the argument list for the new process, ie.
     cmd[0] is the program to run and cmd[1:] are the rest of its arguments.
@@ -715,8 +715,9 @@
     elif os.name == 'os2':
         _spawn_os2(cmd, search_path, dry_run=dry_run, env=env)
     else:
-        raise DistutilsPlatformError, \
-              "don't know how to spawn programs on platform '%s'" % os.name
+        raise DistutilsPlatformError(
+              "don't know how to spawn programs on platform '%s'" % os.name)
+
 
 def _nt_quote_args(args):
     """Quote command-line arguments for DOS/Windows conventions.
@@ -734,6 +735,7 @@
             args[i] = '"%s"' % arg
     return args
 
+
 def _spawn_nt(cmd, search_path=1, verbose=0, dry_run=0, env=None):
     executable = cmd[0]
     cmd = _nt_quote_args(cmd)
@@ -751,12 +753,13 @@
 
         except OSError, exc:
             # this seems to happen when the command isn't found
-            raise DistutilsExecError, \
-                  "command '%s' failed: %s" % (cmd[0], exc[-1])
+            raise DistutilsExecError(
+                  "command '%s' failed: %s" % (cmd[0], exc[-1]))
         if rc != 0:
             # and this reflects the command running but failing
-            raise DistutilsExecError, \
-                  "command '%s' failed with exit status %d" % (cmd[0], rc)
+            raise DistutilsExecError(
+                  "command '%s' failed with exit status %d" % (cmd[0], rc))
+
 
 def _spawn_os2(cmd, search_path=1, verbose=0, dry_run=0, env=None):
     executable = cmd[0]
@@ -774,13 +777,13 @@
 
         except OSError, exc:
             # this seems to happen when the command isn't found
-            raise DistutilsExecError, \
-                  "command '%s' failed: %s" % (cmd[0], exc[-1])
+            raise DistutilsExecError(
+                  "command '%s' failed: %s" % (cmd[0], exc[-1]))
         if rc != 0:
             # and this reflects the command running but failing
             log.debug("command '%s' failed with exit status %d" % (cmd[0], rc))
-            raise DistutilsExecError, \
-                  "command '%s' failed with exit status %d" % (cmd[0], rc)
+            raise DistutilsExecError(
+                  "command '%s' failed with exit status %d" % (cmd[0], rc))
 
 
 def _spawn_posix(cmd, search_path=1, verbose=0, dry_run=0, env=None):
@@ -818,29 +821,30 @@
                 import errno
                 if exc.errno == errno.EINTR:
                     continue
-                raise DistutilsExecError, \
-                      "command '%s' failed: %s" % (cmd[0], exc[-1])
+                raise DistutilsExecError(
+                      "command '%s' failed: %s" % (cmd[0], exc[-1]))
             if os.WIFSIGNALED(status):
-                raise DistutilsExecError, \
+                raise DistutilsExecError(
                       "command '%s' terminated by signal %d" % \
-                      (cmd[0], os.WTERMSIG(status))
+                      (cmd[0], os.WTERMSIG(status)))
 
             elif os.WIFEXITED(status):
                 exit_status = os.WEXITSTATUS(status)
                 if exit_status == 0:
                     return   # hey, it succeeded!
                 else:
-                    raise DistutilsExecError, \
+                    raise DistutilsExecError(
                           "command '%s' failed with exit status %d" % \
-                          (cmd[0], exit_status)
+                          (cmd[0], exit_status))
 
             elif os.WIFSTOPPED(status):
                 continue
 
             else:
-                raise DistutilsExecError, \
+                raise DistutilsExecError(
                       "unknown error executing '%s': termination status %d" % \
-                      (cmd[0], status)
+                      (cmd[0], status))
+
 
 def find_executable(executable, path=None):
     """Tries to find 'executable' in the directories listed in 'path'.

--
Repository URL: http://hg.python.org/distutils2

From amk at amk.ca  Sun Jul 25 00:47:03 2010
From: amk at amk.ca (A.M. Kuchling)
Date: Sat, 24 Jul 2010 18:47:03 -0400
Subject: [Python-checkins] r83123 - in
	python/branches/release27-maint:	Lib/xmlrpclib.py Misc/NEWS
In-Reply-To: <20100724025149.AF610EEA67@mail.python.org>
References: <20100724025149.AF610EEA67@mail.python.org>
Message-ID: <20100724224703.GA45123@andrew-kuchlings-macbook.local>

On Sat, Jul 24, 2010 at 04:51:49AM +0200, victor.stinner wrote:
> New Revision: 83123
> 
> Log:
> Backport of 83120 (#9032)
> 
> XML-RPC client: Transport.request() retries on EPIPE error
  ...
> Remove also a comment in the HTTP client because it's now wrong: see r70643 and
> issue #5542.

I think this commit message is wrong on the 2.7-maint branch since it
doesn't look like you backported the comment removal; I suggest
editing the log message on this revision to take out the misleading
paragraph.

--amk


> 
> 
> Modified:
>    python/branches/release27-maint/   (props changed)
>    python/branches/release27-maint/Lib/xmlrpclib.py
>    python/branches/release27-maint/Misc/NEWS
> 
> Modified: python/branches/release27-maint/Lib/xmlrpclib.py
> ==============================================================================
> --- python/branches/release27-maint/Lib/xmlrpclib.py	(original)
> +++ python/branches/release27-maint/Lib/xmlrpclib.py	Sat Jul 24 04:51:49 2010
> @@ -1263,7 +1263,7 @@
>              try:
>                  return self.single_request(host, handler, request_body, verbose)
>              except socket.error, e:
> -                if i or e.errno not in (errno.ECONNRESET, errno.ECONNABORTED):
> +                if i or e.errno not in (errno.ECONNRESET, errno.ECONNABORTED, errno.EPIPE):
>                      raise
>              except httplib.BadStatusLine: #close after we sent request
>                  if i:
> 
> Modified: python/branches/release27-maint/Misc/NEWS
> ==============================================================================
> --- python/branches/release27-maint/Misc/NEWS	(original)
> +++ python/branches/release27-maint/Misc/NEWS	Sat Jul 24 04:51:49 2010
> @@ -18,6 +18,10 @@
>  Library
>  -------
>  
> +- Issue #9032: XML-RPC client retries the request on EPIPE error. The EPIPE
> +  error occurs when the server closes the socket and the client sends a big
> +  XML-RPC request.
> +
>  - Issue #5542: Remove special logic that closes HTTPConnection socket on EPIPE.
>  
>  - Issue #4629: getopt raises an error if an argument ends with = whereas getopt
> _______________________________________________
> Python-checkins mailing list
> Python-checkins at python.org
> http://mail.python.org/mailman/listinfo/python-checkins

From python-checkins at python.org  Sun Jul 25 16:58:54 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Sun, 25 Jul 2010 16:58:54 +0200 (CEST)
Subject: [Python-checkins] r83139 - in python/branches/py3k/Lib/test:
	test_line_tracing.py test_trace.py
Message-ID: <20100725145854.5401BF328@mail.python.org>

Author: alexander.belopolsky
Date: Sun Jul 25 16:58:54 2010
New Revision: 83139

Log:
Issue #9315: Revert r83005 before renaming test_trace to
test_sys_settrace and test_profilehooks to test_sys_setprofile in all
three branches.



Added:
   python/branches/py3k/Lib/test/test_trace.py
      - copied unchanged from r83138, /python/branches/py3k/Lib/test/test_line_tracing.py
Removed:
   python/branches/py3k/Lib/test/test_line_tracing.py

Deleted: python/branches/py3k/Lib/test/test_line_tracing.py
==============================================================================
--- python/branches/py3k/Lib/test/test_line_tracing.py	Sun Jul 25 16:58:54 2010
+++ (empty file)
@@ -1,790 +0,0 @@
-# Testing the line trace facility.
-
-from test import support
-import unittest
-import sys
-import difflib
-import gc
-
-# A very basic example.  If this fails, we're in deep trouble.
-def basic():
-    return 1
-
-basic.events = [(0, 'call'),
-                (1, 'line'),
-                (1, 'return')]
-
-# Many of the tests below are tricky because they involve pass statements.
-# If there is implicit control flow around a pass statement (in an except
-# clause or else caluse) under what conditions do you set a line number
-# following that clause?
-
-
-# The entire "while 0:" statement is optimized away.  No code
-# exists for it, so the line numbers skip directly from "del x"
-# to "x = 1".
-def arigo_example():
-    x = 1
-    del x
-    while 0:
-        pass
-    x = 1
-
-arigo_example.events = [(0, 'call'),
-                        (1, 'line'),
-                        (2, 'line'),
-                        (5, 'line'),
-                        (5, 'return')]
-
-# check that lines consisting of just one instruction get traced:
-def one_instr_line():
-    x = 1
-    del x
-    x = 1
-
-one_instr_line.events = [(0, 'call'),
-                         (1, 'line'),
-                         (2, 'line'),
-                         (3, 'line'),
-                         (3, 'return')]
-
-def no_pop_tops():      # 0
-    x = 1               # 1
-    for a in range(2):  # 2
-        if a:           # 3
-            x = 1       # 4
-        else:           # 5
-            x = 1       # 6
-
-no_pop_tops.events = [(0, 'call'),
-                      (1, 'line'),
-                      (2, 'line'),
-                      (3, 'line'),
-                      (6, 'line'),
-                      (2, 'line'),
-                      (3, 'line'),
-                      (4, 'line'),
-                      (2, 'line'),
-                      (2, 'return')]
-
-def no_pop_blocks():
-    y = 1
-    while not y:
-        bla
-    x = 1
-
-no_pop_blocks.events = [(0, 'call'),
-                        (1, 'line'),
-                        (2, 'line'),
-                        (4, 'line'),
-                        (4, 'return')]
-
-def called(): # line -3
-    x = 1
-
-def call():   # line 0
-    called()
-
-call.events = [(0, 'call'),
-               (1, 'line'),
-               (-3, 'call'),
-               (-2, 'line'),
-               (-2, 'return'),
-               (1, 'return')]
-
-def raises():
-    raise Exception
-
-def test_raise():
-    try:
-        raises()
-    except Exception as exc:
-        x = 1
-
-test_raise.events = [(0, 'call'),
-                     (1, 'line'),
-                     (2, 'line'),
-                     (-3, 'call'),
-                     (-2, 'line'),
-                     (-2, 'exception'),
-                     (-2, 'return'),
-                     (2, 'exception'),
-                     (3, 'line'),
-                     (4, 'line'),
-                     (4, 'return')]
-
-def _settrace_and_return(tracefunc):
-    sys.settrace(tracefunc)
-    sys._getframe().f_back.f_trace = tracefunc
-def settrace_and_return(tracefunc):
-    _settrace_and_return(tracefunc)
-
-settrace_and_return.events = [(1, 'return')]
-
-def _settrace_and_raise(tracefunc):
-    sys.settrace(tracefunc)
-    sys._getframe().f_back.f_trace = tracefunc
-    raise RuntimeError
-def settrace_and_raise(tracefunc):
-    try:
-        _settrace_and_raise(tracefunc)
-    except RuntimeError as exc:
-        pass
-
-settrace_and_raise.events = [(2, 'exception'),
-                             (3, 'line'),
-                             (4, 'line'),
-                             (4, 'return')]
-
-# implicit return example
-# This test is interesting because of the else: pass
-# part of the code.  The code generate for the true
-# part of the if contains a jump past the else branch.
-# The compiler then generates an implicit "return None"
-# Internally, the compiler visits the pass statement
-# and stores its line number for use on the next instruction.
-# The next instruction is the implicit return None.
-def ireturn_example():
-    a = 5
-    b = 5
-    if a == b:
-        b = a+1
-    else:
-        pass
-
-ireturn_example.events = [(0, 'call'),
-                          (1, 'line'),
-                          (2, 'line'),
-                          (3, 'line'),
-                          (4, 'line'),
-                          (6, 'line'),
-                          (6, 'return')]
-
-# Tight loop with while(1) example (SF #765624)
-def tightloop_example():
-    items = range(0, 3)
-    try:
-        i = 0
-        while 1:
-            b = items[i]; i+=1
-    except IndexError:
-        pass
-
-tightloop_example.events = [(0, 'call'),
-                            (1, 'line'),
-                            (2, 'line'),
-                            (3, 'line'),
-                            (4, 'line'),
-                            (5, 'line'),
-                            (5, 'line'),
-                            (5, 'line'),
-                            (5, 'line'),
-                            (5, 'exception'),
-                            (6, 'line'),
-                            (7, 'line'),
-                            (7, 'return')]
-
-def tighterloop_example():
-    items = range(1, 4)
-    try:
-        i = 0
-        while 1: i = items[i]
-    except IndexError:
-        pass
-
-tighterloop_example.events = [(0, 'call'),
-                            (1, 'line'),
-                            (2, 'line'),
-                            (3, 'line'),
-                            (4, 'line'),
-                            (4, 'line'),
-                            (4, 'line'),
-                            (4, 'line'),
-                            (4, 'exception'),
-                            (5, 'line'),
-                            (6, 'line'),
-                            (6, 'return')]
-
-def generator_function():
-    try:
-        yield True
-        "continued"
-    finally:
-        "finally"
-def generator_example():
-    # any() will leave the generator before its end
-    x = any(generator_function())
-
-    # the following lines were not traced
-    for x in range(10):
-        y = x
-
-generator_example.events = ([(0, 'call'),
-                             (2, 'line'),
-                             (-6, 'call'),
-                             (-5, 'line'),
-                             (-4, 'line'),
-                             (-4, 'return'),
-                             (-4, 'call'),
-                             (-4, 'exception'),
-                             (-1, 'line'),
-                             (-1, 'return')] +
-                            [(5, 'line'), (6, 'line')] * 10 +
-                            [(5, 'line'), (5, 'return')])
-
-
-class Tracer:
-    def __init__(self):
-        self.events = []
-    def trace(self, frame, event, arg):
-        self.events.append((frame.f_lineno, event))
-        return self.trace
-    def traceWithGenexp(self, frame, event, arg):
-        (o for o in [1])
-        self.events.append((frame.f_lineno, event))
-        return self.trace
-
-class TraceTestCase(unittest.TestCase):
-
-    # Disable gc collection when tracing, otherwise the
-    # deallocators may be traced as well.
-    def setUp(self):
-        self.using_gc = gc.isenabled()
-        gc.disable()
-
-    def tearDown(self):
-        if self.using_gc:
-            gc.enable()
-
-    def compare_events(self, line_offset, events, expected_events):
-        events = [(l - line_offset, e) for (l, e) in events]
-        if events != expected_events:
-            self.fail(
-                "events did not match expectation:\n" +
-                "\n".join(difflib.ndiff([str(x) for x in expected_events],
-                                        [str(x) for x in events])))
-
-    def run_and_compare(self, func, events):
-        tracer = Tracer()
-        sys.settrace(tracer.trace)
-        func()
-        sys.settrace(None)
-        self.compare_events(func.__code__.co_firstlineno,
-                            tracer.events, events)
-
-    def run_test(self, func):
-        self.run_and_compare(func, func.events)
-
-    def run_test2(self, func):
-        tracer = Tracer()
-        func(tracer.trace)
-        sys.settrace(None)
-        self.compare_events(func.__code__.co_firstlineno,
-                            tracer.events, func.events)
-
-    def set_and_retrieve_none(self):
-        sys.settrace(None)
-        assert sys.gettrace() is None
-
-    def set_and_retrieve_func(self):
-        def fn(*args):
-            pass
-
-        sys.settrace(fn)
-        try:
-            assert sys.gettrace() is fn
-        finally:
-            sys.settrace(None)
-
-    def test_01_basic(self):
-        self.run_test(basic)
-    def test_02_arigo(self):
-        self.run_test(arigo_example)
-    def test_03_one_instr(self):
-        self.run_test(one_instr_line)
-    def test_04_no_pop_blocks(self):
-        self.run_test(no_pop_blocks)
-    def test_05_no_pop_tops(self):
-        self.run_test(no_pop_tops)
-    def test_06_call(self):
-        self.run_test(call)
-    def test_07_raise(self):
-        self.run_test(test_raise)
-
-    def test_08_settrace_and_return(self):
-        self.run_test2(settrace_and_return)
-    def test_09_settrace_and_raise(self):
-        self.run_test2(settrace_and_raise)
-    def test_10_ireturn(self):
-        self.run_test(ireturn_example)
-    def test_11_tightloop(self):
-        self.run_test(tightloop_example)
-    def test_12_tighterloop(self):
-        self.run_test(tighterloop_example)
-
-    def test_13_genexp(self):
-        self.run_test(generator_example)
-        # issue1265: if the trace function contains a generator,
-        # and if the traced function contains another generator
-        # that is not completely exhausted, the trace stopped.
-        # Worse: the 'finally' clause was not invoked.
-        tracer = Tracer()
-        sys.settrace(tracer.traceWithGenexp)
-        generator_example()
-        sys.settrace(None)
-        self.compare_events(generator_example.__code__.co_firstlineno,
-                            tracer.events, generator_example.events)
-
-    def test_14_onliner_if(self):
-        def onliners():
-            if True: False
-            else: True
-            return 0
-        self.run_and_compare(
-            onliners,
-            [(0, 'call'),
-             (1, 'line'),
-             (3, 'line'),
-             (3, 'return')])
-
-    def test_15_loops(self):
-        # issue1750076: "while" expression is skipped by debugger
-        def for_example():
-            for x in range(2):
-                pass
-        self.run_and_compare(
-            for_example,
-            [(0, 'call'),
-             (1, 'line'),
-             (2, 'line'),
-             (1, 'line'),
-             (2, 'line'),
-             (1, 'line'),
-             (1, 'return')])
-
-        def while_example():
-            # While expression should be traced on every loop
-            x = 2
-            while x > 0:
-                x -= 1
-        self.run_and_compare(
-            while_example,
-            [(0, 'call'),
-             (2, 'line'),
-             (3, 'line'),
-             (4, 'line'),
-             (3, 'line'),
-             (4, 'line'),
-             (3, 'line'),
-             (3, 'return')])
-
-    def test_16_blank_lines(self):
-        namespace = {}
-        exec("def f():\n" + "\n" * 256 + "    pass", namespace)
-        self.run_and_compare(
-            namespace["f"],
-            [(0, 'call'),
-             (257, 'line'),
-             (257, 'return')])
-
-
-class RaisingTraceFuncTestCase(unittest.TestCase):
-    def trace(self, frame, event, arg):
-        """A trace function that raises an exception in response to a
-        specific trace event."""
-        if event == self.raiseOnEvent:
-            raise ValueError # just something that isn't RuntimeError
-        else:
-            return self.trace
-
-    def f(self):
-        """The function to trace; raises an exception if that's the case
-        we're testing, so that the 'exception' trace event fires."""
-        if self.raiseOnEvent == 'exception':
-            x = 0
-            y = 1/x
-        else:
-            return 1
-
-    def run_test_for_event(self, event):
-        """Tests that an exception raised in response to the given event is
-        handled OK."""
-        self.raiseOnEvent = event
-        try:
-            for i in range(sys.getrecursionlimit() + 1):
-                sys.settrace(self.trace)
-                try:
-                    self.f()
-                except ValueError:
-                    pass
-                else:
-                    self.fail("exception not thrown!")
-        except RuntimeError:
-            self.fail("recursion counter not reset")
-
-    # Test the handling of exceptions raised by each kind of trace event.
-    def test_call(self):
-        self.run_test_for_event('call')
-    def test_line(self):
-        self.run_test_for_event('line')
-    def test_return(self):
-        self.run_test_for_event('return')
-    def test_exception(self):
-        self.run_test_for_event('exception')
-
-    def test_trash_stack(self):
-        def f():
-            for i in range(5):
-                print(i)  # line tracing will raise an exception at this line
-
-        def g(frame, why, extra):
-            if (why == 'line' and
-                frame.f_lineno == f.__code__.co_firstlineno + 2):
-                raise RuntimeError("i am crashing")
-            return g
-
-        sys.settrace(g)
-        try:
-            f()
-        except RuntimeError:
-            # the test is really that this doesn't segfault:
-            import gc
-            gc.collect()
-        else:
-            self.fail("exception not propagated")
-
-
-# 'Jump' tests: assigning to frame.f_lineno within a trace function
-# moves the execution position - it's how debuggers implement a Jump
-# command (aka. "Set next statement").
-
-class JumpTracer:
-    """Defines a trace function that jumps from one place to another,
-    with the source and destination lines of the jump being defined by
-    the 'jump' property of the function under test."""
-
-    def __init__(self, function):
-        self.function = function
-        self.jumpFrom = function.jump[0]
-        self.jumpTo = function.jump[1]
-        self.done = False
-
-    def trace(self, frame, event, arg):
-        if not self.done and frame.f_code == self.function.__code__:
-            firstLine = frame.f_code.co_firstlineno
-            if event == 'line' and frame.f_lineno == firstLine + self.jumpFrom:
-                # Cope with non-integer self.jumpTo (because of
-                # no_jump_to_non_integers below).
-                try:
-                    frame.f_lineno = firstLine + self.jumpTo
-                except TypeError:
-                    frame.f_lineno = self.jumpTo
-                self.done = True
-        return self.trace
-
-# The first set of 'jump' tests are for things that are allowed:
-
-def jump_simple_forwards(output):
-    output.append(1)
-    output.append(2)
-    output.append(3)
-
-jump_simple_forwards.jump = (1, 3)
-jump_simple_forwards.output = [3]
-
-def jump_simple_backwards(output):
-    output.append(1)
-    output.append(2)
-
-jump_simple_backwards.jump = (2, 1)
-jump_simple_backwards.output = [1, 1, 2]
-
-def jump_out_of_block_forwards(output):
-    for i in 1, 2:
-        output.append(2)
-        for j in [3]:  # Also tests jumping over a block
-            output.append(4)
-    output.append(5)
-
-jump_out_of_block_forwards.jump = (3, 5)
-jump_out_of_block_forwards.output = [2, 5]
-
-def jump_out_of_block_backwards(output):
-    output.append(1)
-    for i in [1]:
-        output.append(3)
-        for j in [2]:  # Also tests jumping over a block
-            output.append(5)
-        output.append(6)
-    output.append(7)
-
-jump_out_of_block_backwards.jump = (6, 1)
-jump_out_of_block_backwards.output = [1, 3, 5, 1, 3, 5, 6, 7]
-
-def jump_to_codeless_line(output):
-    output.append(1)
-    # Jumping to this line should skip to the next one.
-    output.append(3)
-
-jump_to_codeless_line.jump = (1, 2)
-jump_to_codeless_line.output = [3]
-
-def jump_to_same_line(output):
-    output.append(1)
-    output.append(2)
-    output.append(3)
-
-jump_to_same_line.jump = (2, 2)
-jump_to_same_line.output = [1, 2, 3]
-
-# Tests jumping within a finally block, and over one.
-def jump_in_nested_finally(output):
-    try:
-        output.append(2)
-    finally:
-        output.append(4)
-        try:
-            output.append(6)
-        finally:
-            output.append(8)
-        output.append(9)
-
-jump_in_nested_finally.jump = (4, 9)
-jump_in_nested_finally.output = [2, 9]
-
-# The second set of 'jump' tests are for things that are not allowed:
-
-def no_jump_too_far_forwards(output):
-    try:
-        output.append(2)
-        output.append(3)
-    except ValueError as e:
-        output.append('after' in str(e))
-
-no_jump_too_far_forwards.jump = (3, 6)
-no_jump_too_far_forwards.output = [2, True]
-
-def no_jump_too_far_backwards(output):
-    try:
-        output.append(2)
-        output.append(3)
-    except ValueError as e:
-        output.append('before' in str(e))
-
-no_jump_too_far_backwards.jump = (3, -1)
-no_jump_too_far_backwards.output = [2, True]
-
-# Test each kind of 'except' line.
-def no_jump_to_except_1(output):
-    try:
-        output.append(2)
-    except:
-        e = sys.exc_info()[1]
-        output.append('except' in str(e))
-
-no_jump_to_except_1.jump = (2, 3)
-no_jump_to_except_1.output = [True]
-
-def no_jump_to_except_2(output):
-    try:
-        output.append(2)
-    except ValueError:
-        e = sys.exc_info()[1]
-        output.append('except' in str(e))
-
-no_jump_to_except_2.jump = (2, 3)
-no_jump_to_except_2.output = [True]
-
-def no_jump_to_except_3(output):
-    try:
-        output.append(2)
-    except ValueError as e:
-        output.append('except' in str(e))
-
-no_jump_to_except_3.jump = (2, 3)
-no_jump_to_except_3.output = [True]
-
-def no_jump_to_except_4(output):
-    try:
-        output.append(2)
-    except (ValueError, RuntimeError) as e:
-        output.append('except' in str(e))
-
-no_jump_to_except_4.jump = (2, 3)
-no_jump_to_except_4.output = [True]
-
-def no_jump_forwards_into_block(output):
-    try:
-        output.append(2)
-        for i in 1, 2:
-            output.append(4)
-    except ValueError as e:
-        output.append('into' in str(e))
-
-no_jump_forwards_into_block.jump = (2, 4)
-no_jump_forwards_into_block.output = [True]
-
-def no_jump_backwards_into_block(output):
-    try:
-        for i in 1, 2:
-            output.append(3)
-        output.append(4)
-    except ValueError as e:
-        output.append('into' in str(e))
-
-no_jump_backwards_into_block.jump = (4, 3)
-no_jump_backwards_into_block.output = [3, 3, True]
-
-def no_jump_into_finally_block(output):
-    try:
-        try:
-            output.append(3)
-            x = 1
-        finally:
-            output.append(6)
-    except ValueError as e:
-        output.append('finally' in str(e))
-
-no_jump_into_finally_block.jump = (4, 6)
-no_jump_into_finally_block.output = [3, 6, True]  # The 'finally' still runs
-
-def no_jump_out_of_finally_block(output):
-    try:
-        try:
-            output.append(3)
-        finally:
-            output.append(5)
-            output.append(6)
-    except ValueError as e:
-        output.append('finally' in str(e))
-
-no_jump_out_of_finally_block.jump = (5, 1)
-no_jump_out_of_finally_block.output = [3, True]
-
-# This verifies the line-numbers-must-be-integers rule.
-def no_jump_to_non_integers(output):
-    try:
-        output.append(2)
-    except ValueError as e:
-        output.append('integer' in str(e))
-
-no_jump_to_non_integers.jump = (2, "Spam")
-no_jump_to_non_integers.output = [True]
-
-# This verifies that you can't set f_lineno via _getframe or similar
-# trickery.
-def no_jump_without_trace_function():
-    try:
-        previous_frame = sys._getframe().f_back
-        previous_frame.f_lineno = previous_frame.f_lineno
-    except ValueError as e:
-        # This is the exception we wanted; make sure the error message
-        # talks about trace functions.
-        if 'trace' not in str(e):
-            raise
-    else:
-        # Something's wrong - the expected exception wasn't raised.
-        raise RuntimeError("Trace-function-less jump failed to fail")
-
-
-class JumpTestCase(unittest.TestCase):
-    def compare_jump_output(self, expected, received):
-        if received != expected:
-            self.fail( "Outputs don't match:\n" +
-                       "Expected: " + repr(expected) + "\n" +
-                       "Received: " + repr(received))
-
-    def run_test(self, func):
-        tracer = JumpTracer(func)
-        sys.settrace(tracer.trace)
-        output = []
-        func(output)
-        sys.settrace(None)
-        self.compare_jump_output(func.output, output)
-
-    def test_01_jump_simple_forwards(self):
-        self.run_test(jump_simple_forwards)
-    def test_02_jump_simple_backwards(self):
-        self.run_test(jump_simple_backwards)
-    def test_03_jump_out_of_block_forwards(self):
-        self.run_test(jump_out_of_block_forwards)
-    def test_04_jump_out_of_block_backwards(self):
-        self.run_test(jump_out_of_block_backwards)
-    def test_05_jump_to_codeless_line(self):
-        self.run_test(jump_to_codeless_line)
-    def test_06_jump_to_same_line(self):
-        self.run_test(jump_to_same_line)
-    def test_07_jump_in_nested_finally(self):
-        self.run_test(jump_in_nested_finally)
-    def test_08_no_jump_too_far_forwards(self):
-        self.run_test(no_jump_too_far_forwards)
-    def test_09_no_jump_too_far_backwards(self):
-        self.run_test(no_jump_too_far_backwards)
-    def test_10_no_jump_to_except_1(self):
-        self.run_test(no_jump_to_except_1)
-    def test_11_no_jump_to_except_2(self):
-        self.run_test(no_jump_to_except_2)
-    def test_12_no_jump_to_except_3(self):
-        self.run_test(no_jump_to_except_3)
-    def test_13_no_jump_to_except_4(self):
-        self.run_test(no_jump_to_except_4)
-    def test_14_no_jump_forwards_into_block(self):
-        self.run_test(no_jump_forwards_into_block)
-    def test_15_no_jump_backwards_into_block(self):
-        self.run_test(no_jump_backwards_into_block)
-    def test_16_no_jump_into_finally_block(self):
-        self.run_test(no_jump_into_finally_block)
-    def test_17_no_jump_out_of_finally_block(self):
-        self.run_test(no_jump_out_of_finally_block)
-    def test_18_no_jump_to_non_integers(self):
-        self.run_test(no_jump_to_non_integers)
-    def test_19_no_jump_without_trace_function(self):
-        no_jump_without_trace_function()
-
-    def test_20_large_function(self):
-        d = {}
-        exec("""def f(output):        # line 0
-            x = 0                     # line 1
-            y = 1                     # line 2
-            '''                       # line 3
-            %s                        # lines 4-1004
-            '''                       # line 1005
-            x += 1                    # line 1006
-            output.append(x)          # line 1007
-            return""" % ('\n' * 1000,), d)
-        f = d['f']
-
-        f.jump = (2, 1007)
-        f.output = [0]
-        self.run_test(f)
-
-    def test_jump_to_firstlineno(self):
-        # This tests that PDB can jump back to the first line in a
-        # file.  See issue #1689458.  It can only be triggered in a
-        # function call if the function is defined on a single line.
-        code = compile("""
-# Comments don't count.
-output.append(2)  # firstlineno is here.
-output.append(3)
-output.append(4)
-""", "<fake module>", "exec")
-        class fake_function:
-            __code__ = code
-            jump = (2, 0)
-        tracer = JumpTracer(fake_function)
-        sys.settrace(tracer.trace)
-        namespace = {"output": []}
-        exec(code, namespace)
-        sys.settrace(None)
-        self.compare_jump_output([2, 3, 2, 3, 4], namespace["output"])
-
-
-def test_main():
-    support.run_unittest(
-        TraceTestCase,
-        RaisingTraceFuncTestCase,
-        JumpTestCase
-    )
-
-if __name__ == "__main__":
-    test_main()

From python-checkins at python.org  Sun Jul 25 17:02:56 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Sun, 25 Jul 2010 17:02:56 +0200 (CEST)
Subject: [Python-checkins] r83140 - in python/branches/py3k/Lib/test:
	test_profilehooks.py test_sys_setprofile.py
	test_sys_settrace.py test_trace.py
Message-ID: <20100725150256.394D4F430@mail.python.org>

Author: alexander.belopolsky
Date: Sun Jul 25 17:02:55 2010
New Revision: 83140

Log:
Issue #9315: Renamed test_trace to test_sys_settrace and
test_profilehooks to test_sys_setprofile so that test_trace can be
used for testing the trace module and for naming consistency.



Added:
   python/branches/py3k/Lib/test/test_sys_setprofile.py
      - copied unchanged from r83138, /python/branches/py3k/Lib/test/test_profilehooks.py
   python/branches/py3k/Lib/test/test_sys_settrace.py
      - copied unchanged from r83139, /python/branches/py3k/Lib/test/test_trace.py
Removed:
   python/branches/py3k/Lib/test/test_profilehooks.py
   python/branches/py3k/Lib/test/test_trace.py

Deleted: python/branches/py3k/Lib/test/test_profilehooks.py
==============================================================================
--- python/branches/py3k/Lib/test/test_profilehooks.py	Sun Jul 25 17:02:55 2010
+++ (empty file)
@@ -1,385 +0,0 @@
-import pprint
-import sys
-import unittest
-
-from test import support
-
-class TestGetProfile(unittest.TestCase):
-    def setUp(self):
-        sys.setprofile(None)
-
-    def tearDown(self):
-        sys.setprofile(None)
-
-    def test_empty(self):
-        assert sys.getprofile() is None
-
-    def test_setget(self):
-        def fn(*args):
-            pass
-
-        sys.setprofile(fn)
-        assert sys.getprofile() == fn
-
-class HookWatcher:
-    def __init__(self):
-        self.frames = []
-        self.events = []
-
-    def callback(self, frame, event, arg):
-        if (event == "call"
-            or event == "return"
-            or event == "exception"):
-            self.add_event(event, frame)
-
-    def add_event(self, event, frame=None):
-        """Add an event to the log."""
-        if frame is None:
-            frame = sys._getframe(1)
-
-        try:
-            frameno = self.frames.index(frame)
-        except ValueError:
-            frameno = len(self.frames)
-            self.frames.append(frame)
-
-        self.events.append((frameno, event, ident(frame)))
-
-    def get_events(self):
-        """Remove calls to add_event()."""
-        disallowed = [ident(self.add_event.__func__), ident(ident)]
-        self.frames = None
-
-        return [item for item in self.events if item[2] not in disallowed]
-
-
-class ProfileSimulator(HookWatcher):
-    def __init__(self, testcase):
-        self.testcase = testcase
-        self.stack = []
-        HookWatcher.__init__(self)
-
-    def callback(self, frame, event, arg):
-        # Callback registered with sys.setprofile()/sys.settrace()
-        self.dispatch[event](self, frame)
-
-    def trace_call(self, frame):
-        self.add_event('call', frame)
-        self.stack.append(frame)
-
-    def trace_return(self, frame):
-        self.add_event('return', frame)
-        self.stack.pop()
-
-    def trace_exception(self, frame):
-        self.testcase.fail(
-            "the profiler should never receive exception events")
-
-    def trace_pass(self, frame):
-        pass
-
-    dispatch = {
-        'call': trace_call,
-        'exception': trace_exception,
-        'return': trace_return,
-        'c_call': trace_pass,
-        'c_return': trace_pass,
-        'c_exception': trace_pass,
-        }
-
-
-class TestCaseBase(unittest.TestCase):
-    def check_events(self, callable, expected):
-        events = capture_events(callable, self.new_watcher())
-        if events != expected:
-            self.fail("Expected events:\n%s\nReceived events:\n%s"
-                      % (pprint.pformat(expected), pprint.pformat(events)))
-
-
-class ProfileHookTestCase(TestCaseBase):
-    def new_watcher(self):
-        return HookWatcher()
-
-    def test_simple(self):
-        def f(p):
-            pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_exception(self):
-        def f(p):
-            1/0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_caught_exception(self):
-        def f(p):
-            try: 1/0
-            except: pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_caught_nested_exception(self):
-        def f(p):
-            try: 1/0
-            except: pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_nested_exception(self):
-        def f(p):
-            1/0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              # This isn't what I expected:
-                              # (0, 'exception', protect_ident),
-                              # I expected this again:
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_exception_in_except_clause(self):
-        def f(p):
-            1/0
-        def g(p):
-            try:
-                f(p)
-            except:
-                try: f(p)
-                except: pass
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (3, 'call', f_ident),
-                              (3, 'return', f_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-    def test_exception_propogation(self):
-        def f(p):
-            1/0
-        def g(p):
-            try: f(p)
-            finally: p.add_event("falling through")
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (1, 'falling through', g_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-    def test_raise_twice(self):
-        def f(p):
-            try: 1/0
-            except: 1/0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_raise_reraise(self):
-        def f(p):
-            try: 1/0
-            except: raise
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_raise(self):
-        def f(p):
-            raise Exception()
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_distant_exception(self):
-        def f():
-            1/0
-        def g():
-            f()
-        def h():
-            g()
-        def i():
-            h()
-        def j(p):
-            i()
-        f_ident = ident(f)
-        g_ident = ident(g)
-        h_ident = ident(h)
-        i_ident = ident(i)
-        j_ident = ident(j)
-        self.check_events(j, [(1, 'call', j_ident),
-                              (2, 'call', i_ident),
-                              (3, 'call', h_ident),
-                              (4, 'call', g_ident),
-                              (5, 'call', f_ident),
-                              (5, 'return', f_ident),
-                              (4, 'return', g_ident),
-                              (3, 'return', h_ident),
-                              (2, 'return', i_ident),
-                              (1, 'return', j_ident),
-                              ])
-
-    def test_generator(self):
-        def f():
-            for i in range(2):
-                yield i
-        def g(p):
-            for i in f():
-                pass
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              # call the iterator twice to generate values
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              # once more; returns end-of-iteration with
-                              # actually raising an exception
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-    def test_stop_iteration(self):
-        def f():
-            for i in range(2):
-                yield i
-            raise StopIteration
-        def g(p):
-            for i in f():
-                pass
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              # call the iterator twice to generate values
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              # once more to hit the raise:
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-
-class ProfileSimulatorTestCase(TestCaseBase):
-    def new_watcher(self):
-        return ProfileSimulator(self)
-
-    def test_simple(self):
-        def f(p):
-            pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_basic_exception(self):
-        def f(p):
-            1/0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_caught_exception(self):
-        def f(p):
-            try: 1/0
-            except: pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_distant_exception(self):
-        def f():
-            1/0
-        def g():
-            f()
-        def h():
-            g()
-        def i():
-            h()
-        def j(p):
-            i()
-        f_ident = ident(f)
-        g_ident = ident(g)
-        h_ident = ident(h)
-        i_ident = ident(i)
-        j_ident = ident(j)
-        self.check_events(j, [(1, 'call', j_ident),
-                              (2, 'call', i_ident),
-                              (3, 'call', h_ident),
-                              (4, 'call', g_ident),
-                              (5, 'call', f_ident),
-                              (5, 'return', f_ident),
-                              (4, 'return', g_ident),
-                              (3, 'return', h_ident),
-                              (2, 'return', i_ident),
-                              (1, 'return', j_ident),
-                              ])
-
-
-def ident(function):
-    if hasattr(function, "f_code"):
-        code = function.f_code
-    else:
-        code = function.__code__
-    return code.co_firstlineno, code.co_name
-
-
-def protect(f, p):
-    try: f(p)
-    except: pass
-
-protect_ident = ident(protect)
-
-
-def capture_events(callable, p=None):
-    try:
-        sys.setprofile()
-    except TypeError:
-        pass
-    else:
-        raise support.TestFailed(
-            'sys.setprofile() did not raise TypeError')
-
-    if p is None:
-        p = HookWatcher()
-    sys.setprofile(p.callback)
-    protect(callable, p)
-    sys.setprofile(None)
-    return p.get_events()[1:-1]
-
-
-def show_events(callable):
-    import pprint
-    pprint.pprint(capture_events(callable))
-
-
-def test_main():
-    support.run_unittest(
-        TestGetProfile,
-        ProfileHookTestCase,
-        ProfileSimulatorTestCase
-    )
-
-
-if __name__ == "__main__":
-    test_main()

Deleted: python/branches/py3k/Lib/test/test_trace.py
==============================================================================
--- python/branches/py3k/Lib/test/test_trace.py	Sun Jul 25 17:02:55 2010
+++ (empty file)
@@ -1,790 +0,0 @@
-# Testing the line trace facility.
-
-from test import support
-import unittest
-import sys
-import difflib
-import gc
-
-# A very basic example.  If this fails, we're in deep trouble.
-def basic():
-    return 1
-
-basic.events = [(0, 'call'),
-                (1, 'line'),
-                (1, 'return')]
-
-# Many of the tests below are tricky because they involve pass statements.
-# If there is implicit control flow around a pass statement (in an except
-# clause or else caluse) under what conditions do you set a line number
-# following that clause?
-
-
-# The entire "while 0:" statement is optimized away.  No code
-# exists for it, so the line numbers skip directly from "del x"
-# to "x = 1".
-def arigo_example():
-    x = 1
-    del x
-    while 0:
-        pass
-    x = 1
-
-arigo_example.events = [(0, 'call'),
-                        (1, 'line'),
-                        (2, 'line'),
-                        (5, 'line'),
-                        (5, 'return')]
-
-# check that lines consisting of just one instruction get traced:
-def one_instr_line():
-    x = 1
-    del x
-    x = 1
-
-one_instr_line.events = [(0, 'call'),
-                         (1, 'line'),
-                         (2, 'line'),
-                         (3, 'line'),
-                         (3, 'return')]
-
-def no_pop_tops():      # 0
-    x = 1               # 1
-    for a in range(2):  # 2
-        if a:           # 3
-            x = 1       # 4
-        else:           # 5
-            x = 1       # 6
-
-no_pop_tops.events = [(0, 'call'),
-                      (1, 'line'),
-                      (2, 'line'),
-                      (3, 'line'),
-                      (6, 'line'),
-                      (2, 'line'),
-                      (3, 'line'),
-                      (4, 'line'),
-                      (2, 'line'),
-                      (2, 'return')]
-
-def no_pop_blocks():
-    y = 1
-    while not y:
-        bla
-    x = 1
-
-no_pop_blocks.events = [(0, 'call'),
-                        (1, 'line'),
-                        (2, 'line'),
-                        (4, 'line'),
-                        (4, 'return')]
-
-def called(): # line -3
-    x = 1
-
-def call():   # line 0
-    called()
-
-call.events = [(0, 'call'),
-               (1, 'line'),
-               (-3, 'call'),
-               (-2, 'line'),
-               (-2, 'return'),
-               (1, 'return')]
-
-def raises():
-    raise Exception
-
-def test_raise():
-    try:
-        raises()
-    except Exception as exc:
-        x = 1
-
-test_raise.events = [(0, 'call'),
-                     (1, 'line'),
-                     (2, 'line'),
-                     (-3, 'call'),
-                     (-2, 'line'),
-                     (-2, 'exception'),
-                     (-2, 'return'),
-                     (2, 'exception'),
-                     (3, 'line'),
-                     (4, 'line'),
-                     (4, 'return')]
-
-def _settrace_and_return(tracefunc):
-    sys.settrace(tracefunc)
-    sys._getframe().f_back.f_trace = tracefunc
-def settrace_and_return(tracefunc):
-    _settrace_and_return(tracefunc)
-
-settrace_and_return.events = [(1, 'return')]
-
-def _settrace_and_raise(tracefunc):
-    sys.settrace(tracefunc)
-    sys._getframe().f_back.f_trace = tracefunc
-    raise RuntimeError
-def settrace_and_raise(tracefunc):
-    try:
-        _settrace_and_raise(tracefunc)
-    except RuntimeError as exc:
-        pass
-
-settrace_and_raise.events = [(2, 'exception'),
-                             (3, 'line'),
-                             (4, 'line'),
-                             (4, 'return')]
-
-# implicit return example
-# This test is interesting because of the else: pass
-# part of the code.  The code generate for the true
-# part of the if contains a jump past the else branch.
-# The compiler then generates an implicit "return None"
-# Internally, the compiler visits the pass statement
-# and stores its line number for use on the next instruction.
-# The next instruction is the implicit return None.
-def ireturn_example():
-    a = 5
-    b = 5
-    if a == b:
-        b = a+1
-    else:
-        pass
-
-ireturn_example.events = [(0, 'call'),
-                          (1, 'line'),
-                          (2, 'line'),
-                          (3, 'line'),
-                          (4, 'line'),
-                          (6, 'line'),
-                          (6, 'return')]
-
-# Tight loop with while(1) example (SF #765624)
-def tightloop_example():
-    items = range(0, 3)
-    try:
-        i = 0
-        while 1:
-            b = items[i]; i+=1
-    except IndexError:
-        pass
-
-tightloop_example.events = [(0, 'call'),
-                            (1, 'line'),
-                            (2, 'line'),
-                            (3, 'line'),
-                            (4, 'line'),
-                            (5, 'line'),
-                            (5, 'line'),
-                            (5, 'line'),
-                            (5, 'line'),
-                            (5, 'exception'),
-                            (6, 'line'),
-                            (7, 'line'),
-                            (7, 'return')]
-
-def tighterloop_example():
-    items = range(1, 4)
-    try:
-        i = 0
-        while 1: i = items[i]
-    except IndexError:
-        pass
-
-tighterloop_example.events = [(0, 'call'),
-                            (1, 'line'),
-                            (2, 'line'),
-                            (3, 'line'),
-                            (4, 'line'),
-                            (4, 'line'),
-                            (4, 'line'),
-                            (4, 'line'),
-                            (4, 'exception'),
-                            (5, 'line'),
-                            (6, 'line'),
-                            (6, 'return')]
-
-def generator_function():
-    try:
-        yield True
-        "continued"
-    finally:
-        "finally"
-def generator_example():
-    # any() will leave the generator before its end
-    x = any(generator_function())
-
-    # the following lines were not traced
-    for x in range(10):
-        y = x
-
-generator_example.events = ([(0, 'call'),
-                             (2, 'line'),
-                             (-6, 'call'),
-                             (-5, 'line'),
-                             (-4, 'line'),
-                             (-4, 'return'),
-                             (-4, 'call'),
-                             (-4, 'exception'),
-                             (-1, 'line'),
-                             (-1, 'return')] +
-                            [(5, 'line'), (6, 'line')] * 10 +
-                            [(5, 'line'), (5, 'return')])
-
-
-class Tracer:
-    def __init__(self):
-        self.events = []
-    def trace(self, frame, event, arg):
-        self.events.append((frame.f_lineno, event))
-        return self.trace
-    def traceWithGenexp(self, frame, event, arg):
-        (o for o in [1])
-        self.events.append((frame.f_lineno, event))
-        return self.trace
-
-class TraceTestCase(unittest.TestCase):
-
-    # Disable gc collection when tracing, otherwise the
-    # deallocators may be traced as well.
-    def setUp(self):
-        self.using_gc = gc.isenabled()
-        gc.disable()
-
-    def tearDown(self):
-        if self.using_gc:
-            gc.enable()
-
-    def compare_events(self, line_offset, events, expected_events):
-        events = [(l - line_offset, e) for (l, e) in events]
-        if events != expected_events:
-            self.fail(
-                "events did not match expectation:\n" +
-                "\n".join(difflib.ndiff([str(x) for x in expected_events],
-                                        [str(x) for x in events])))
-
-    def run_and_compare(self, func, events):
-        tracer = Tracer()
-        sys.settrace(tracer.trace)
-        func()
-        sys.settrace(None)
-        self.compare_events(func.__code__.co_firstlineno,
-                            tracer.events, events)
-
-    def run_test(self, func):
-        self.run_and_compare(func, func.events)
-
-    def run_test2(self, func):
-        tracer = Tracer()
-        func(tracer.trace)
-        sys.settrace(None)
-        self.compare_events(func.__code__.co_firstlineno,
-                            tracer.events, func.events)
-
-    def set_and_retrieve_none(self):
-        sys.settrace(None)
-        assert sys.gettrace() is None
-
-    def set_and_retrieve_func(self):
-        def fn(*args):
-            pass
-
-        sys.settrace(fn)
-        try:
-            assert sys.gettrace() is fn
-        finally:
-            sys.settrace(None)
-
-    def test_01_basic(self):
-        self.run_test(basic)
-    def test_02_arigo(self):
-        self.run_test(arigo_example)
-    def test_03_one_instr(self):
-        self.run_test(one_instr_line)
-    def test_04_no_pop_blocks(self):
-        self.run_test(no_pop_blocks)
-    def test_05_no_pop_tops(self):
-        self.run_test(no_pop_tops)
-    def test_06_call(self):
-        self.run_test(call)
-    def test_07_raise(self):
-        self.run_test(test_raise)
-
-    def test_08_settrace_and_return(self):
-        self.run_test2(settrace_and_return)
-    def test_09_settrace_and_raise(self):
-        self.run_test2(settrace_and_raise)
-    def test_10_ireturn(self):
-        self.run_test(ireturn_example)
-    def test_11_tightloop(self):
-        self.run_test(tightloop_example)
-    def test_12_tighterloop(self):
-        self.run_test(tighterloop_example)
-
-    def test_13_genexp(self):
-        self.run_test(generator_example)
-        # issue1265: if the trace function contains a generator,
-        # and if the traced function contains another generator
-        # that is not completely exhausted, the trace stopped.
-        # Worse: the 'finally' clause was not invoked.
-        tracer = Tracer()
-        sys.settrace(tracer.traceWithGenexp)
-        generator_example()
-        sys.settrace(None)
-        self.compare_events(generator_example.__code__.co_firstlineno,
-                            tracer.events, generator_example.events)
-
-    def test_14_onliner_if(self):
-        def onliners():
-            if True: False
-            else: True
-            return 0
-        self.run_and_compare(
-            onliners,
-            [(0, 'call'),
-             (1, 'line'),
-             (3, 'line'),
-             (3, 'return')])
-
-    def test_15_loops(self):
-        # issue1750076: "while" expression is skipped by debugger
-        def for_example():
-            for x in range(2):
-                pass
-        self.run_and_compare(
-            for_example,
-            [(0, 'call'),
-             (1, 'line'),
-             (2, 'line'),
-             (1, 'line'),
-             (2, 'line'),
-             (1, 'line'),
-             (1, 'return')])
-
-        def while_example():
-            # While expression should be traced on every loop
-            x = 2
-            while x > 0:
-                x -= 1
-        self.run_and_compare(
-            while_example,
-            [(0, 'call'),
-             (2, 'line'),
-             (3, 'line'),
-             (4, 'line'),
-             (3, 'line'),
-             (4, 'line'),
-             (3, 'line'),
-             (3, 'return')])
-
-    def test_16_blank_lines(self):
-        namespace = {}
-        exec("def f():\n" + "\n" * 256 + "    pass", namespace)
-        self.run_and_compare(
-            namespace["f"],
-            [(0, 'call'),
-             (257, 'line'),
-             (257, 'return')])
-
-
-class RaisingTraceFuncTestCase(unittest.TestCase):
-    def trace(self, frame, event, arg):
-        """A trace function that raises an exception in response to a
-        specific trace event."""
-        if event == self.raiseOnEvent:
-            raise ValueError # just something that isn't RuntimeError
-        else:
-            return self.trace
-
-    def f(self):
-        """The function to trace; raises an exception if that's the case
-        we're testing, so that the 'exception' trace event fires."""
-        if self.raiseOnEvent == 'exception':
-            x = 0
-            y = 1/x
-        else:
-            return 1
-
-    def run_test_for_event(self, event):
-        """Tests that an exception raised in response to the given event is
-        handled OK."""
-        self.raiseOnEvent = event
-        try:
-            for i in range(sys.getrecursionlimit() + 1):
-                sys.settrace(self.trace)
-                try:
-                    self.f()
-                except ValueError:
-                    pass
-                else:
-                    self.fail("exception not thrown!")
-        except RuntimeError:
-            self.fail("recursion counter not reset")
-
-    # Test the handling of exceptions raised by each kind of trace event.
-    def test_call(self):
-        self.run_test_for_event('call')
-    def test_line(self):
-        self.run_test_for_event('line')
-    def test_return(self):
-        self.run_test_for_event('return')
-    def test_exception(self):
-        self.run_test_for_event('exception')
-
-    def test_trash_stack(self):
-        def f():
-            for i in range(5):
-                print(i)  # line tracing will raise an exception at this line
-
-        def g(frame, why, extra):
-            if (why == 'line' and
-                frame.f_lineno == f.__code__.co_firstlineno + 2):
-                raise RuntimeError("i am crashing")
-            return g
-
-        sys.settrace(g)
-        try:
-            f()
-        except RuntimeError:
-            # the test is really that this doesn't segfault:
-            import gc
-            gc.collect()
-        else:
-            self.fail("exception not propagated")
-
-
-# 'Jump' tests: assigning to frame.f_lineno within a trace function
-# moves the execution position - it's how debuggers implement a Jump
-# command (aka. "Set next statement").
-
-class JumpTracer:
-    """Defines a trace function that jumps from one place to another,
-    with the source and destination lines of the jump being defined by
-    the 'jump' property of the function under test."""
-
-    def __init__(self, function):
-        self.function = function
-        self.jumpFrom = function.jump[0]
-        self.jumpTo = function.jump[1]
-        self.done = False
-
-    def trace(self, frame, event, arg):
-        if not self.done and frame.f_code == self.function.__code__:
-            firstLine = frame.f_code.co_firstlineno
-            if event == 'line' and frame.f_lineno == firstLine + self.jumpFrom:
-                # Cope with non-integer self.jumpTo (because of
-                # no_jump_to_non_integers below).
-                try:
-                    frame.f_lineno = firstLine + self.jumpTo
-                except TypeError:
-                    frame.f_lineno = self.jumpTo
-                self.done = True
-        return self.trace
-
-# The first set of 'jump' tests are for things that are allowed:
-
-def jump_simple_forwards(output):
-    output.append(1)
-    output.append(2)
-    output.append(3)
-
-jump_simple_forwards.jump = (1, 3)
-jump_simple_forwards.output = [3]
-
-def jump_simple_backwards(output):
-    output.append(1)
-    output.append(2)
-
-jump_simple_backwards.jump = (2, 1)
-jump_simple_backwards.output = [1, 1, 2]
-
-def jump_out_of_block_forwards(output):
-    for i in 1, 2:
-        output.append(2)
-        for j in [3]:  # Also tests jumping over a block
-            output.append(4)
-    output.append(5)
-
-jump_out_of_block_forwards.jump = (3, 5)
-jump_out_of_block_forwards.output = [2, 5]
-
-def jump_out_of_block_backwards(output):
-    output.append(1)
-    for i in [1]:
-        output.append(3)
-        for j in [2]:  # Also tests jumping over a block
-            output.append(5)
-        output.append(6)
-    output.append(7)
-
-jump_out_of_block_backwards.jump = (6, 1)
-jump_out_of_block_backwards.output = [1, 3, 5, 1, 3, 5, 6, 7]
-
-def jump_to_codeless_line(output):
-    output.append(1)
-    # Jumping to this line should skip to the next one.
-    output.append(3)
-
-jump_to_codeless_line.jump = (1, 2)
-jump_to_codeless_line.output = [3]
-
-def jump_to_same_line(output):
-    output.append(1)
-    output.append(2)
-    output.append(3)
-
-jump_to_same_line.jump = (2, 2)
-jump_to_same_line.output = [1, 2, 3]
-
-# Tests jumping within a finally block, and over one.
-def jump_in_nested_finally(output):
-    try:
-        output.append(2)
-    finally:
-        output.append(4)
-        try:
-            output.append(6)
-        finally:
-            output.append(8)
-        output.append(9)
-
-jump_in_nested_finally.jump = (4, 9)
-jump_in_nested_finally.output = [2, 9]
-
-# The second set of 'jump' tests are for things that are not allowed:
-
-def no_jump_too_far_forwards(output):
-    try:
-        output.append(2)
-        output.append(3)
-    except ValueError as e:
-        output.append('after' in str(e))
-
-no_jump_too_far_forwards.jump = (3, 6)
-no_jump_too_far_forwards.output = [2, True]
-
-def no_jump_too_far_backwards(output):
-    try:
-        output.append(2)
-        output.append(3)
-    except ValueError as e:
-        output.append('before' in str(e))
-
-no_jump_too_far_backwards.jump = (3, -1)
-no_jump_too_far_backwards.output = [2, True]
-
-# Test each kind of 'except' line.
-def no_jump_to_except_1(output):
-    try:
-        output.append(2)
-    except:
-        e = sys.exc_info()[1]
-        output.append('except' in str(e))
-
-no_jump_to_except_1.jump = (2, 3)
-no_jump_to_except_1.output = [True]
-
-def no_jump_to_except_2(output):
-    try:
-        output.append(2)
-    except ValueError:
-        e = sys.exc_info()[1]
-        output.append('except' in str(e))
-
-no_jump_to_except_2.jump = (2, 3)
-no_jump_to_except_2.output = [True]
-
-def no_jump_to_except_3(output):
-    try:
-        output.append(2)
-    except ValueError as e:
-        output.append('except' in str(e))
-
-no_jump_to_except_3.jump = (2, 3)
-no_jump_to_except_3.output = [True]
-
-def no_jump_to_except_4(output):
-    try:
-        output.append(2)
-    except (ValueError, RuntimeError) as e:
-        output.append('except' in str(e))
-
-no_jump_to_except_4.jump = (2, 3)
-no_jump_to_except_4.output = [True]
-
-def no_jump_forwards_into_block(output):
-    try:
-        output.append(2)
-        for i in 1, 2:
-            output.append(4)
-    except ValueError as e:
-        output.append('into' in str(e))
-
-no_jump_forwards_into_block.jump = (2, 4)
-no_jump_forwards_into_block.output = [True]
-
-def no_jump_backwards_into_block(output):
-    try:
-        for i in 1, 2:
-            output.append(3)
-        output.append(4)
-    except ValueError as e:
-        output.append('into' in str(e))
-
-no_jump_backwards_into_block.jump = (4, 3)
-no_jump_backwards_into_block.output = [3, 3, True]
-
-def no_jump_into_finally_block(output):
-    try:
-        try:
-            output.append(3)
-            x = 1
-        finally:
-            output.append(6)
-    except ValueError as e:
-        output.append('finally' in str(e))
-
-no_jump_into_finally_block.jump = (4, 6)
-no_jump_into_finally_block.output = [3, 6, True]  # The 'finally' still runs
-
-def no_jump_out_of_finally_block(output):
-    try:
-        try:
-            output.append(3)
-        finally:
-            output.append(5)
-            output.append(6)
-    except ValueError as e:
-        output.append('finally' in str(e))
-
-no_jump_out_of_finally_block.jump = (5, 1)
-no_jump_out_of_finally_block.output = [3, True]
-
-# This verifies the line-numbers-must-be-integers rule.
-def no_jump_to_non_integers(output):
-    try:
-        output.append(2)
-    except ValueError as e:
-        output.append('integer' in str(e))
-
-no_jump_to_non_integers.jump = (2, "Spam")
-no_jump_to_non_integers.output = [True]
-
-# This verifies that you can't set f_lineno via _getframe or similar
-# trickery.
-def no_jump_without_trace_function():
-    try:
-        previous_frame = sys._getframe().f_back
-        previous_frame.f_lineno = previous_frame.f_lineno
-    except ValueError as e:
-        # This is the exception we wanted; make sure the error message
-        # talks about trace functions.
-        if 'trace' not in str(e):
-            raise
-    else:
-        # Something's wrong - the expected exception wasn't raised.
-        raise RuntimeError("Trace-function-less jump failed to fail")
-
-
-class JumpTestCase(unittest.TestCase):
-    def compare_jump_output(self, expected, received):
-        if received != expected:
-            self.fail( "Outputs don't match:\n" +
-                       "Expected: " + repr(expected) + "\n" +
-                       "Received: " + repr(received))
-
-    def run_test(self, func):
-        tracer = JumpTracer(func)
-        sys.settrace(tracer.trace)
-        output = []
-        func(output)
-        sys.settrace(None)
-        self.compare_jump_output(func.output, output)
-
-    def test_01_jump_simple_forwards(self):
-        self.run_test(jump_simple_forwards)
-    def test_02_jump_simple_backwards(self):
-        self.run_test(jump_simple_backwards)
-    def test_03_jump_out_of_block_forwards(self):
-        self.run_test(jump_out_of_block_forwards)
-    def test_04_jump_out_of_block_backwards(self):
-        self.run_test(jump_out_of_block_backwards)
-    def test_05_jump_to_codeless_line(self):
-        self.run_test(jump_to_codeless_line)
-    def test_06_jump_to_same_line(self):
-        self.run_test(jump_to_same_line)
-    def test_07_jump_in_nested_finally(self):
-        self.run_test(jump_in_nested_finally)
-    def test_08_no_jump_too_far_forwards(self):
-        self.run_test(no_jump_too_far_forwards)
-    def test_09_no_jump_too_far_backwards(self):
-        self.run_test(no_jump_too_far_backwards)
-    def test_10_no_jump_to_except_1(self):
-        self.run_test(no_jump_to_except_1)
-    def test_11_no_jump_to_except_2(self):
-        self.run_test(no_jump_to_except_2)
-    def test_12_no_jump_to_except_3(self):
-        self.run_test(no_jump_to_except_3)
-    def test_13_no_jump_to_except_4(self):
-        self.run_test(no_jump_to_except_4)
-    def test_14_no_jump_forwards_into_block(self):
-        self.run_test(no_jump_forwards_into_block)
-    def test_15_no_jump_backwards_into_block(self):
-        self.run_test(no_jump_backwards_into_block)
-    def test_16_no_jump_into_finally_block(self):
-        self.run_test(no_jump_into_finally_block)
-    def test_17_no_jump_out_of_finally_block(self):
-        self.run_test(no_jump_out_of_finally_block)
-    def test_18_no_jump_to_non_integers(self):
-        self.run_test(no_jump_to_non_integers)
-    def test_19_no_jump_without_trace_function(self):
-        no_jump_without_trace_function()
-
-    def test_20_large_function(self):
-        d = {}
-        exec("""def f(output):        # line 0
-            x = 0                     # line 1
-            y = 1                     # line 2
-            '''                       # line 3
-            %s                        # lines 4-1004
-            '''                       # line 1005
-            x += 1                    # line 1006
-            output.append(x)          # line 1007
-            return""" % ('\n' * 1000,), d)
-        f = d['f']
-
-        f.jump = (2, 1007)
-        f.output = [0]
-        self.run_test(f)
-
-    def test_jump_to_firstlineno(self):
-        # This tests that PDB can jump back to the first line in a
-        # file.  See issue #1689458.  It can only be triggered in a
-        # function call if the function is defined on a single line.
-        code = compile("""
-# Comments don't count.
-output.append(2)  # firstlineno is here.
-output.append(3)
-output.append(4)
-""", "<fake module>", "exec")
-        class fake_function:
-            __code__ = code
-            jump = (2, 0)
-        tracer = JumpTracer(fake_function)
-        sys.settrace(tracer.trace)
-        namespace = {"output": []}
-        exec(code, namespace)
-        sys.settrace(None)
-        self.compare_jump_output([2, 3, 2, 3, 4], namespace["output"])
-
-
-def test_main():
-    support.run_unittest(
-        TraceTestCase,
-        RaisingTraceFuncTestCase,
-        JumpTestCase
-    )
-
-if __name__ == "__main__":
-    test_main()

From python-checkins at python.org  Sun Jul 25 17:05:42 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Sun, 25 Jul 2010 17:05:42 +0200 (CEST)
Subject: [Python-checkins] r83141 - python/branches/py3k/Lib/test/test_sys.py
Message-ID: <20100725150542.376CFEE999@mail.python.org>

Author: alexander.belopolsky
Date: Sun Jul 25 17:05:42 2010
New Revision: 83141

Log:
Corrected comments on where settrace and setprofile are tested.

Modified:
   python/branches/py3k/Lib/test/test_sys.py

Modified: python/branches/py3k/Lib/test/test_sys.py
==============================================================================
--- python/branches/py3k/Lib/test/test_sys.py	(original)
+++ python/branches/py3k/Lib/test/test_sys.py	Sun Jul 25 17:05:42 2010
@@ -179,8 +179,8 @@
         # can't check more than the type, as the user might have changed it
         self.assertIsInstance(sys.getdefaultencoding(), str)
 
-    # testing sys.settrace() is done in test_trace.py
-    # testing sys.setprofile() is done in test_profile.py
+    # testing sys.settrace() is done in test_sys_settrace.py
+    # testing sys.setprofile() is done in test_sys_setprofile.py
 
     def test_setcheckinterval(self):
         with warnings.catch_warnings():

From python-checkins at python.org  Sun Jul 25 17:07:37 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Sun, 25 Jul 2010 17:07:37 +0200 (CEST)
Subject: [Python-checkins] r83142 - in python/branches/release27-maint:
	Lib/test/test_profilehooks.py Lib/test/test_sys.py
	Lib/test/test_sys_setprofile.py Lib/test/test_sys_settrace.py
	Lib/test/test_trace.py
Message-ID: <20100725150737.263D1EE9BE@mail.python.org>

Author: alexander.belopolsky
Date: Sun Jul 25 17:07:36 2010
New Revision: 83142

Log:
Merged revisions 83140-83141 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83140 | alexander.belopolsky | 2010-07-25 11:02:55 -0400 (Sun, 25 Jul 2010) | 5 lines
  
  Issue #9315: Renamed test_trace to test_sys_settrace and
  test_profilehooks to test_sys_setprofile so that test_trace can be
  used for testing the trace module and for naming consistency.
........
  r83141 | alexander.belopolsky | 2010-07-25 11:05:42 -0400 (Sun, 25 Jul 2010) | 1 line
  
  Corrected comments on where settrace and setprofile are tested.
........


Added:
   python/branches/release27-maint/Lib/test/test_sys_setprofile.py
      - copied unchanged from r83141, /python/branches/py3k/Lib/test/test_sys_setprofile.py
   python/branches/release27-maint/Lib/test/test_sys_settrace.py
      - copied unchanged from r83141, /python/branches/py3k/Lib/test/test_sys_settrace.py
Removed:
   python/branches/release27-maint/Lib/test/test_profilehooks.py
   python/branches/release27-maint/Lib/test/test_trace.py
Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/test/test_sys.py

Deleted: python/branches/release27-maint/Lib/test/test_profilehooks.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_profilehooks.py	Sun Jul 25 17:07:36 2010
+++ (empty file)
@@ -1,385 +0,0 @@
-import pprint
-import sys
-import unittest
-
-from test import test_support
-
-class TestGetProfile(unittest.TestCase):
-    def setUp(self):
-        sys.setprofile(None)
-
-    def tearDown(self):
-        sys.setprofile(None)
-
-    def test_empty(self):
-        assert sys.getprofile() is None
-
-    def test_setget(self):
-        def fn(*args):
-            pass
-
-        sys.setprofile(fn)
-        assert sys.getprofile() == fn
-
-class HookWatcher:
-    def __init__(self):
-        self.frames = []
-        self.events = []
-
-    def callback(self, frame, event, arg):
-        if (event == "call"
-            or event == "return"
-            or event == "exception"):
-            self.add_event(event, frame)
-
-    def add_event(self, event, frame=None):
-        """Add an event to the log."""
-        if frame is None:
-            frame = sys._getframe(1)
-
-        try:
-            frameno = self.frames.index(frame)
-        except ValueError:
-            frameno = len(self.frames)
-            self.frames.append(frame)
-
-        self.events.append((frameno, event, ident(frame)))
-
-    def get_events(self):
-        """Remove calls to add_event()."""
-        disallowed = [ident(self.add_event.im_func), ident(ident)]
-        self.frames = None
-
-        return [item for item in self.events if item[2] not in disallowed]
-
-
-class ProfileSimulator(HookWatcher):
-    def __init__(self, testcase):
-        self.testcase = testcase
-        self.stack = []
-        HookWatcher.__init__(self)
-
-    def callback(self, frame, event, arg):
-        # Callback registered with sys.setprofile()/sys.settrace()
-        self.dispatch[event](self, frame)
-
-    def trace_call(self, frame):
-        self.add_event('call', frame)
-        self.stack.append(frame)
-
-    def trace_return(self, frame):
-        self.add_event('return', frame)
-        self.stack.pop()
-
-    def trace_exception(self, frame):
-        self.testcase.fail(
-            "the profiler should never receive exception events")
-
-    def trace_pass(self, frame):
-        pass
-
-    dispatch = {
-        'call': trace_call,
-        'exception': trace_exception,
-        'return': trace_return,
-        'c_call': trace_pass,
-        'c_return': trace_pass,
-        'c_exception': trace_pass,
-        }
-
-
-class TestCaseBase(unittest.TestCase):
-    def check_events(self, callable, expected):
-        events = capture_events(callable, self.new_watcher())
-        if events != expected:
-            self.fail("Expected events:\n%s\nReceived events:\n%s"
-                      % (pprint.pformat(expected), pprint.pformat(events)))
-
-
-class ProfileHookTestCase(TestCaseBase):
-    def new_watcher(self):
-        return HookWatcher()
-
-    def test_simple(self):
-        def f(p):
-            pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_exception(self):
-        def f(p):
-            1./0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_caught_exception(self):
-        def f(p):
-            try: 1./0
-            except: pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_caught_nested_exception(self):
-        def f(p):
-            try: 1./0
-            except: pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_nested_exception(self):
-        def f(p):
-            1./0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              # This isn't what I expected:
-                              # (0, 'exception', protect_ident),
-                              # I expected this again:
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_exception_in_except_clause(self):
-        def f(p):
-            1./0
-        def g(p):
-            try:
-                f(p)
-            except:
-                try: f(p)
-                except: pass
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (3, 'call', f_ident),
-                              (3, 'return', f_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-    def test_exception_propogation(self):
-        def f(p):
-            1./0
-        def g(p):
-            try: f(p)
-            finally: p.add_event("falling through")
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (1, 'falling through', g_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-    def test_raise_twice(self):
-        def f(p):
-            try: 1./0
-            except: 1./0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_raise_reraise(self):
-        def f(p):
-            try: 1./0
-            except: raise
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_raise(self):
-        def f(p):
-            raise Exception()
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_distant_exception(self):
-        def f():
-            1./0
-        def g():
-            f()
-        def h():
-            g()
-        def i():
-            h()
-        def j(p):
-            i()
-        f_ident = ident(f)
-        g_ident = ident(g)
-        h_ident = ident(h)
-        i_ident = ident(i)
-        j_ident = ident(j)
-        self.check_events(j, [(1, 'call', j_ident),
-                              (2, 'call', i_ident),
-                              (3, 'call', h_ident),
-                              (4, 'call', g_ident),
-                              (5, 'call', f_ident),
-                              (5, 'return', f_ident),
-                              (4, 'return', g_ident),
-                              (3, 'return', h_ident),
-                              (2, 'return', i_ident),
-                              (1, 'return', j_ident),
-                              ])
-
-    def test_generator(self):
-        def f():
-            for i in range(2):
-                yield i
-        def g(p):
-            for i in f():
-                pass
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              # call the iterator twice to generate values
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              # once more; returns end-of-iteration with
-                              # actually raising an exception
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-    def test_stop_iteration(self):
-        def f():
-            for i in range(2):
-                yield i
-            raise StopIteration
-        def g(p):
-            for i in f():
-                pass
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              # call the iterator twice to generate values
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              # once more to hit the raise:
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-
-class ProfileSimulatorTestCase(TestCaseBase):
-    def new_watcher(self):
-        return ProfileSimulator(self)
-
-    def test_simple(self):
-        def f(p):
-            pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_basic_exception(self):
-        def f(p):
-            1./0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_caught_exception(self):
-        def f(p):
-            try: 1./0
-            except: pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_distant_exception(self):
-        def f():
-            1./0
-        def g():
-            f()
-        def h():
-            g()
-        def i():
-            h()
-        def j(p):
-            i()
-        f_ident = ident(f)
-        g_ident = ident(g)
-        h_ident = ident(h)
-        i_ident = ident(i)
-        j_ident = ident(j)
-        self.check_events(j, [(1, 'call', j_ident),
-                              (2, 'call', i_ident),
-                              (3, 'call', h_ident),
-                              (4, 'call', g_ident),
-                              (5, 'call', f_ident),
-                              (5, 'return', f_ident),
-                              (4, 'return', g_ident),
-                              (3, 'return', h_ident),
-                              (2, 'return', i_ident),
-                              (1, 'return', j_ident),
-                              ])
-
-
-def ident(function):
-    if hasattr(function, "f_code"):
-        code = function.f_code
-    else:
-        code = function.func_code
-    return code.co_firstlineno, code.co_name
-
-
-def protect(f, p):
-    try: f(p)
-    except: pass
-
-protect_ident = ident(protect)
-
-
-def capture_events(callable, p=None):
-    try:
-        sys.setprofile()
-    except TypeError:
-        pass
-    else:
-        raise test_support.TestFailed(
-            'sys.setprofile() did not raise TypeError')
-
-    if p is None:
-        p = HookWatcher()
-    sys.setprofile(p.callback)
-    protect(callable, p)
-    sys.setprofile(None)
-    return p.get_events()[1:-1]
-
-
-def show_events(callable):
-    import pprint
-    pprint.pprint(capture_events(callable))
-
-
-def test_main():
-    test_support.run_unittest(
-        TestGetProfile,
-        ProfileHookTestCase,
-        ProfileSimulatorTestCase
-    )
-
-
-if __name__ == "__main__":
-    test_main()

Modified: python/branches/release27-maint/Lib/test/test_sys.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_sys.py	(original)
+++ python/branches/release27-maint/Lib/test/test_sys.py	Sun Jul 25 17:07:36 2010
@@ -205,8 +205,8 @@
             # can't check more than the type, as the user might have changed it
             self.assertIsInstance(sys.getdefaultencoding(), str)
 
-    # testing sys.settrace() is done in test_trace.py
-    # testing sys.setprofile() is done in test_profile.py
+    # testing sys.settrace() is done in test_sys_settrace.py
+    # testing sys.setprofile() is done in test_sys_setprofile.py
 
     def test_setcheckinterval(self):
         self.assertRaises(TypeError, sys.setcheckinterval)

Deleted: python/branches/release27-maint/Lib/test/test_trace.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_trace.py	Sun Jul 25 17:07:36 2010
+++ (empty file)
@@ -1,789 +0,0 @@
-# Testing the line trace facility.
-
-from test import test_support
-import unittest
-import sys
-import difflib
-import gc
-
-# A very basic example.  If this fails, we're in deep trouble.
-def basic():
-    return 1
-
-basic.events = [(0, 'call'),
-                (1, 'line'),
-                (1, 'return')]
-
-# Many of the tests below are tricky because they involve pass statements.
-# If there is implicit control flow around a pass statement (in an except
-# clause or else caluse) under what conditions do you set a line number
-# following that clause?
-
-
-# The entire "while 0:" statement is optimized away.  No code
-# exists for it, so the line numbers skip directly from "del x"
-# to "x = 1".
-def arigo_example():
-    x = 1
-    del x
-    while 0:
-        pass
-    x = 1
-
-arigo_example.events = [(0, 'call'),
-                        (1, 'line'),
-                        (2, 'line'),
-                        (5, 'line'),
-                        (5, 'return')]
-
-# check that lines consisting of just one instruction get traced:
-def one_instr_line():
-    x = 1
-    del x
-    x = 1
-
-one_instr_line.events = [(0, 'call'),
-                         (1, 'line'),
-                         (2, 'line'),
-                         (3, 'line'),
-                         (3, 'return')]
-
-def no_pop_tops():      # 0
-    x = 1               # 1
-    for a in range(2):  # 2
-        if a:           # 3
-            x = 1       # 4
-        else:           # 5
-            x = 1       # 6
-
-no_pop_tops.events = [(0, 'call'),
-                      (1, 'line'),
-                      (2, 'line'),
-                      (3, 'line'),
-                      (6, 'line'),
-                      (2, 'line'),
-                      (3, 'line'),
-                      (4, 'line'),
-                      (2, 'line'),
-                      (2, 'return')]
-
-def no_pop_blocks():
-    y = 1
-    while not y:
-        bla
-    x = 1
-
-no_pop_blocks.events = [(0, 'call'),
-                        (1, 'line'),
-                        (2, 'line'),
-                        (4, 'line'),
-                        (4, 'return')]
-
-def called(): # line -3
-    x = 1
-
-def call():   # line 0
-    called()
-
-call.events = [(0, 'call'),
-               (1, 'line'),
-               (-3, 'call'),
-               (-2, 'line'),
-               (-2, 'return'),
-               (1, 'return')]
-
-def raises():
-    raise Exception
-
-def test_raise():
-    try:
-        raises()
-    except Exception, exc:
-        x = 1
-
-test_raise.events = [(0, 'call'),
-                     (1, 'line'),
-                     (2, 'line'),
-                     (-3, 'call'),
-                     (-2, 'line'),
-                     (-2, 'exception'),
-                     (-2, 'return'),
-                     (2, 'exception'),
-                     (3, 'line'),
-                     (4, 'line'),
-                     (4, 'return')]
-
-def _settrace_and_return(tracefunc):
-    sys.settrace(tracefunc)
-    sys._getframe().f_back.f_trace = tracefunc
-def settrace_and_return(tracefunc):
-    _settrace_and_return(tracefunc)
-
-settrace_and_return.events = [(1, 'return')]
-
-def _settrace_and_raise(tracefunc):
-    sys.settrace(tracefunc)
-    sys._getframe().f_back.f_trace = tracefunc
-    raise RuntimeError
-def settrace_and_raise(tracefunc):
-    try:
-        _settrace_and_raise(tracefunc)
-    except RuntimeError, exc:
-        pass
-
-settrace_and_raise.events = [(2, 'exception'),
-                             (3, 'line'),
-                             (4, 'line'),
-                             (4, 'return')]
-
-# implicit return example
-# This test is interesting because of the else: pass
-# part of the code.  The code generate for the true
-# part of the if contains a jump past the else branch.
-# The compiler then generates an implicit "return None"
-# Internally, the compiler visits the pass statement
-# and stores its line number for use on the next instruction.
-# The next instruction is the implicit return None.
-def ireturn_example():
-    a = 5
-    b = 5
-    if a == b:
-        b = a+1
-    else:
-        pass
-
-ireturn_example.events = [(0, 'call'),
-                          (1, 'line'),
-                          (2, 'line'),
-                          (3, 'line'),
-                          (4, 'line'),
-                          (6, 'line'),
-                          (6, 'return')]
-
-# Tight loop with while(1) example (SF #765624)
-def tightloop_example():
-    items = range(0, 3)
-    try:
-        i = 0
-        while 1:
-            b = items[i]; i+=1
-    except IndexError:
-        pass
-
-tightloop_example.events = [(0, 'call'),
-                            (1, 'line'),
-                            (2, 'line'),
-                            (3, 'line'),
-                            (4, 'line'),
-                            (5, 'line'),
-                            (5, 'line'),
-                            (5, 'line'),
-                            (5, 'line'),
-                            (5, 'exception'),
-                            (6, 'line'),
-                            (7, 'line'),
-                            (7, 'return')]
-
-def tighterloop_example():
-    items = range(1, 4)
-    try:
-        i = 0
-        while 1: i = items[i]
-    except IndexError:
-        pass
-
-tighterloop_example.events = [(0, 'call'),
-                            (1, 'line'),
-                            (2, 'line'),
-                            (3, 'line'),
-                            (4, 'line'),
-                            (4, 'line'),
-                            (4, 'line'),
-                            (4, 'line'),
-                            (4, 'exception'),
-                            (5, 'line'),
-                            (6, 'line'),
-                            (6, 'return')]
-
-def generator_function():
-    try:
-        yield True
-        "continued"
-    finally:
-        "finally"
-def generator_example():
-    # any() will leave the generator before its end
-    x = any(generator_function())
-
-    # the following lines were not traced
-    for x in range(10):
-        y = x
-
-generator_example.events = ([(0, 'call'),
-                             (2, 'line'),
-                             (-6, 'call'),
-                             (-5, 'line'),
-                             (-4, 'line'),
-                             (-4, 'return'),
-                             (-4, 'call'),
-                             (-4, 'exception'),
-                             (-1, 'line'),
-                             (-1, 'return')] +
-                            [(5, 'line'), (6, 'line')] * 10 +
-                            [(5, 'line'), (5, 'return')])
-
-
-class Tracer:
-    def __init__(self):
-        self.events = []
-    def trace(self, frame, event, arg):
-        self.events.append((frame.f_lineno, event))
-        return self.trace
-    def traceWithGenexp(self, frame, event, arg):
-        (o for o in [1])
-        self.events.append((frame.f_lineno, event))
-        return self.trace
-
-class TraceTestCase(unittest.TestCase):
-
-    # Disable gc collection when tracing, otherwise the
-    # deallocators may be traced as well.
-    def setUp(self):
-        self.using_gc = gc.isenabled()
-        gc.disable()
-
-    def tearDown(self):
-        if self.using_gc:
-            gc.enable()
-
-    def compare_events(self, line_offset, events, expected_events):
-        events = [(l - line_offset, e) for (l, e) in events]
-        if events != expected_events:
-            self.fail(
-                "events did not match expectation:\n" +
-                "\n".join(difflib.ndiff([str(x) for x in expected_events],
-                                        [str(x) for x in events])))
-
-    def run_and_compare(self, func, events):
-        tracer = Tracer()
-        sys.settrace(tracer.trace)
-        func()
-        sys.settrace(None)
-        self.compare_events(func.func_code.co_firstlineno,
-                            tracer.events, events)
-
-    def run_test(self, func):
-        self.run_and_compare(func, func.events)
-
-    def run_test2(self, func):
-        tracer = Tracer()
-        func(tracer.trace)
-        sys.settrace(None)
-        self.compare_events(func.func_code.co_firstlineno,
-                            tracer.events, func.events)
-
-    def set_and_retrieve_none(self):
-        sys.settrace(None)
-        assert sys.gettrace() is None
-
-    def set_and_retrieve_func(self):
-        def fn(*args):
-            pass
-
-        sys.settrace(fn)
-        try:
-            assert sys.gettrace() is fn
-        finally:
-            sys.settrace(None)
-
-    def test_01_basic(self):
-        self.run_test(basic)
-    def test_02_arigo(self):
-        self.run_test(arigo_example)
-    def test_03_one_instr(self):
-        self.run_test(one_instr_line)
-    def test_04_no_pop_blocks(self):
-        self.run_test(no_pop_blocks)
-    def test_05_no_pop_tops(self):
-        self.run_test(no_pop_tops)
-    def test_06_call(self):
-        self.run_test(call)
-    def test_07_raise(self):
-        self.run_test(test_raise)
-
-    def test_08_settrace_and_return(self):
-        self.run_test2(settrace_and_return)
-    def test_09_settrace_and_raise(self):
-        self.run_test2(settrace_and_raise)
-    def test_10_ireturn(self):
-        self.run_test(ireturn_example)
-    def test_11_tightloop(self):
-        self.run_test(tightloop_example)
-    def test_12_tighterloop(self):
-        self.run_test(tighterloop_example)
-
-    def test_13_genexp(self):
-        self.run_test(generator_example)
-        # issue1265: if the trace function contains a generator,
-        # and if the traced function contains another generator
-        # that is not completely exhausted, the trace stopped.
-        # Worse: the 'finally' clause was not invoked.
-        tracer = Tracer()
-        sys.settrace(tracer.traceWithGenexp)
-        generator_example()
-        sys.settrace(None)
-        self.compare_events(generator_example.__code__.co_firstlineno,
-                            tracer.events, generator_example.events)
-
-    def test_14_onliner_if(self):
-        def onliners():
-            if True: False
-            else: True
-            return 0
-        self.run_and_compare(
-            onliners,
-            [(0, 'call'),
-             (1, 'line'),
-             (3, 'line'),
-             (3, 'return')])
-
-    def test_15_loops(self):
-        # issue1750076: "while" expression is skipped by debugger
-        def for_example():
-            for x in range(2):
-                pass
-        self.run_and_compare(
-            for_example,
-            [(0, 'call'),
-             (1, 'line'),
-             (2, 'line'),
-             (1, 'line'),
-             (2, 'line'),
-             (1, 'line'),
-             (1, 'return')])
-
-        def while_example():
-            # While expression should be traced on every loop
-            x = 2
-            while x > 0:
-                x -= 1
-        self.run_and_compare(
-            while_example,
-            [(0, 'call'),
-             (2, 'line'),
-             (3, 'line'),
-             (4, 'line'),
-             (3, 'line'),
-             (4, 'line'),
-             (3, 'line'),
-             (3, 'return')])
-
-    def test_16_blank_lines(self):
-        exec("def f():\n" + "\n" * 256 + "    pass")
-        self.run_and_compare(
-            f,
-            [(0, 'call'),
-             (257, 'line'),
-             (257, 'return')])
-
-
-class RaisingTraceFuncTestCase(unittest.TestCase):
-    def trace(self, frame, event, arg):
-        """A trace function that raises an exception in response to a
-        specific trace event."""
-        if event == self.raiseOnEvent:
-            raise ValueError # just something that isn't RuntimeError
-        else:
-            return self.trace
-
-    def f(self):
-        """The function to trace; raises an exception if that's the case
-        we're testing, so that the 'exception' trace event fires."""
-        if self.raiseOnEvent == 'exception':
-            x = 0
-            y = 1 // x
-        else:
-            return 1
-
-    def run_test_for_event(self, event):
-        """Tests that an exception raised in response to the given event is
-        handled OK."""
-        self.raiseOnEvent = event
-        try:
-            for i in xrange(sys.getrecursionlimit() + 1):
-                sys.settrace(self.trace)
-                try:
-                    self.f()
-                except ValueError:
-                    pass
-                else:
-                    self.fail("exception not thrown!")
-        except RuntimeError:
-            self.fail("recursion counter not reset")
-
-    # Test the handling of exceptions raised by each kind of trace event.
-    def test_call(self):
-        self.run_test_for_event('call')
-    def test_line(self):
-        self.run_test_for_event('line')
-    def test_return(self):
-        self.run_test_for_event('return')
-    def test_exception(self):
-        self.run_test_for_event('exception')
-
-    def test_trash_stack(self):
-        def f():
-            for i in range(5):
-                print i  # line tracing will raise an exception at this line
-
-        def g(frame, why, extra):
-            if (why == 'line' and
-                frame.f_lineno == f.func_code.co_firstlineno + 2):
-                raise RuntimeError, "i am crashing"
-            return g
-
-        sys.settrace(g)
-        try:
-            f()
-        except RuntimeError:
-            # the test is really that this doesn't segfault:
-            import gc
-            gc.collect()
-        else:
-            self.fail("exception not propagated")
-
-
-# 'Jump' tests: assigning to frame.f_lineno within a trace function
-# moves the execution position - it's how debuggers implement a Jump
-# command (aka. "Set next statement").
-
-class JumpTracer:
-    """Defines a trace function that jumps from one place to another,
-    with the source and destination lines of the jump being defined by
-    the 'jump' property of the function under test."""
-
-    def __init__(self, function):
-        self.function = function
-        self.jumpFrom = function.jump[0]
-        self.jumpTo = function.jump[1]
-        self.done = False
-
-    def trace(self, frame, event, arg):
-        if not self.done and frame.f_code == self.function.func_code:
-            firstLine = frame.f_code.co_firstlineno
-            if event == 'line' and frame.f_lineno == firstLine + self.jumpFrom:
-                # Cope with non-integer self.jumpTo (because of
-                # no_jump_to_non_integers below).
-                try:
-                    frame.f_lineno = firstLine + self.jumpTo
-                except TypeError:
-                    frame.f_lineno = self.jumpTo
-                self.done = True
-        return self.trace
-
-# The first set of 'jump' tests are for things that are allowed:
-
-def jump_simple_forwards(output):
-    output.append(1)
-    output.append(2)
-    output.append(3)
-
-jump_simple_forwards.jump = (1, 3)
-jump_simple_forwards.output = [3]
-
-def jump_simple_backwards(output):
-    output.append(1)
-    output.append(2)
-
-jump_simple_backwards.jump = (2, 1)
-jump_simple_backwards.output = [1, 1, 2]
-
-def jump_out_of_block_forwards(output):
-    for i in 1, 2:
-        output.append(2)
-        for j in [3]:  # Also tests jumping over a block
-            output.append(4)
-    output.append(5)
-
-jump_out_of_block_forwards.jump = (3, 5)
-jump_out_of_block_forwards.output = [2, 5]
-
-def jump_out_of_block_backwards(output):
-    output.append(1)
-    for i in [1]:
-        output.append(3)
-        for j in [2]:  # Also tests jumping over a block
-            output.append(5)
-        output.append(6)
-    output.append(7)
-
-jump_out_of_block_backwards.jump = (6, 1)
-jump_out_of_block_backwards.output = [1, 3, 5, 1, 3, 5, 6, 7]
-
-def jump_to_codeless_line(output):
-    output.append(1)
-    # Jumping to this line should skip to the next one.
-    output.append(3)
-
-jump_to_codeless_line.jump = (1, 2)
-jump_to_codeless_line.output = [3]
-
-def jump_to_same_line(output):
-    output.append(1)
-    output.append(2)
-    output.append(3)
-
-jump_to_same_line.jump = (2, 2)
-jump_to_same_line.output = [1, 2, 3]
-
-# Tests jumping within a finally block, and over one.
-def jump_in_nested_finally(output):
-    try:
-        output.append(2)
-    finally:
-        output.append(4)
-        try:
-            output.append(6)
-        finally:
-            output.append(8)
-        output.append(9)
-
-jump_in_nested_finally.jump = (4, 9)
-jump_in_nested_finally.output = [2, 9]
-
-# The second set of 'jump' tests are for things that are not allowed:
-
-def no_jump_too_far_forwards(output):
-    try:
-        output.append(2)
-        output.append(3)
-    except ValueError, e:
-        output.append('after' in str(e))
-
-no_jump_too_far_forwards.jump = (3, 6)
-no_jump_too_far_forwards.output = [2, True]
-
-def no_jump_too_far_backwards(output):
-    try:
-        output.append(2)
-        output.append(3)
-    except ValueError, e:
-        output.append('before' in str(e))
-
-no_jump_too_far_backwards.jump = (3, -1)
-no_jump_too_far_backwards.output = [2, True]
-
-# Test each kind of 'except' line.
-def no_jump_to_except_1(output):
-    try:
-        output.append(2)
-    except:
-        e = sys.exc_info()[1]
-        output.append('except' in str(e))
-
-no_jump_to_except_1.jump = (2, 3)
-no_jump_to_except_1.output = [True]
-
-def no_jump_to_except_2(output):
-    try:
-        output.append(2)
-    except ValueError:
-        e = sys.exc_info()[1]
-        output.append('except' in str(e))
-
-no_jump_to_except_2.jump = (2, 3)
-no_jump_to_except_2.output = [True]
-
-def no_jump_to_except_3(output):
-    try:
-        output.append(2)
-    except ValueError, e:
-        output.append('except' in str(e))
-
-no_jump_to_except_3.jump = (2, 3)
-no_jump_to_except_3.output = [True]
-
-def no_jump_to_except_4(output):
-    try:
-        output.append(2)
-    except (ValueError, RuntimeError), e:
-        output.append('except' in str(e))
-
-no_jump_to_except_4.jump = (2, 3)
-no_jump_to_except_4.output = [True]
-
-def no_jump_forwards_into_block(output):
-    try:
-        output.append(2)
-        for i in 1, 2:
-            output.append(4)
-    except ValueError, e:
-        output.append('into' in str(e))
-
-no_jump_forwards_into_block.jump = (2, 4)
-no_jump_forwards_into_block.output = [True]
-
-def no_jump_backwards_into_block(output):
-    try:
-        for i in 1, 2:
-            output.append(3)
-        output.append(4)
-    except ValueError, e:
-        output.append('into' in str(e))
-
-no_jump_backwards_into_block.jump = (4, 3)
-no_jump_backwards_into_block.output = [3, 3, True]
-
-def no_jump_into_finally_block(output):
-    try:
-        try:
-            output.append(3)
-            x = 1
-        finally:
-            output.append(6)
-    except ValueError, e:
-        output.append('finally' in str(e))
-
-no_jump_into_finally_block.jump = (4, 6)
-no_jump_into_finally_block.output = [3, 6, True]  # The 'finally' still runs
-
-def no_jump_out_of_finally_block(output):
-    try:
-        try:
-            output.append(3)
-        finally:
-            output.append(5)
-            output.append(6)
-    except ValueError, e:
-        output.append('finally' in str(e))
-
-no_jump_out_of_finally_block.jump = (5, 1)
-no_jump_out_of_finally_block.output = [3, True]
-
-# This verifies the line-numbers-must-be-integers rule.
-def no_jump_to_non_integers(output):
-    try:
-        output.append(2)
-    except ValueError, e:
-        output.append('integer' in str(e))
-
-no_jump_to_non_integers.jump = (2, "Spam")
-no_jump_to_non_integers.output = [True]
-
-# This verifies that you can't set f_lineno via _getframe or similar
-# trickery.
-def no_jump_without_trace_function():
-    try:
-        previous_frame = sys._getframe().f_back
-        previous_frame.f_lineno = previous_frame.f_lineno
-    except ValueError, e:
-        # This is the exception we wanted; make sure the error message
-        # talks about trace functions.
-        if 'trace' not in str(e):
-            raise
-    else:
-        # Something's wrong - the expected exception wasn't raised.
-        raise RuntimeError, "Trace-function-less jump failed to fail"
-
-
-class JumpTestCase(unittest.TestCase):
-    def compare_jump_output(self, expected, received):
-        if received != expected:
-            self.fail( "Outputs don't match:\n" +
-                       "Expected: " + repr(expected) + "\n" +
-                       "Received: " + repr(received))
-
-    def run_test(self, func):
-        tracer = JumpTracer(func)
-        sys.settrace(tracer.trace)
-        output = []
-        func(output)
-        sys.settrace(None)
-        self.compare_jump_output(func.output, output)
-
-    def test_01_jump_simple_forwards(self):
-        self.run_test(jump_simple_forwards)
-    def test_02_jump_simple_backwards(self):
-        self.run_test(jump_simple_backwards)
-    def test_03_jump_out_of_block_forwards(self):
-        self.run_test(jump_out_of_block_forwards)
-    def test_04_jump_out_of_block_backwards(self):
-        self.run_test(jump_out_of_block_backwards)
-    def test_05_jump_to_codeless_line(self):
-        self.run_test(jump_to_codeless_line)
-    def test_06_jump_to_same_line(self):
-        self.run_test(jump_to_same_line)
-    def test_07_jump_in_nested_finally(self):
-        self.run_test(jump_in_nested_finally)
-    def test_08_no_jump_too_far_forwards(self):
-        self.run_test(no_jump_too_far_forwards)
-    def test_09_no_jump_too_far_backwards(self):
-        self.run_test(no_jump_too_far_backwards)
-    def test_10_no_jump_to_except_1(self):
-        self.run_test(no_jump_to_except_1)
-    def test_11_no_jump_to_except_2(self):
-        self.run_test(no_jump_to_except_2)
-    def test_12_no_jump_to_except_3(self):
-        self.run_test(no_jump_to_except_3)
-    def test_13_no_jump_to_except_4(self):
-        self.run_test(no_jump_to_except_4)
-    def test_14_no_jump_forwards_into_block(self):
-        self.run_test(no_jump_forwards_into_block)
-    def test_15_no_jump_backwards_into_block(self):
-        self.run_test(no_jump_backwards_into_block)
-    def test_16_no_jump_into_finally_block(self):
-        self.run_test(no_jump_into_finally_block)
-    def test_17_no_jump_out_of_finally_block(self):
-        self.run_test(no_jump_out_of_finally_block)
-    def test_18_no_jump_to_non_integers(self):
-        self.run_test(no_jump_to_non_integers)
-    def test_19_no_jump_without_trace_function(self):
-        no_jump_without_trace_function()
-
-    def test_20_large_function(self):
-        d = {}
-        exec("""def f(output):        # line 0
-            x = 0                     # line 1
-            y = 1                     # line 2
-            '''                       # line 3
-            %s                        # lines 4-1004
-            '''                       # line 1005
-            x += 1                    # line 1006
-            output.append(x)          # line 1007
-            return""" % ('\n' * 1000,), d)
-        f = d['f']
-
-        f.jump = (2, 1007)
-        f.output = [0]
-        self.run_test(f)
-
-    def test_jump_to_firstlineno(self):
-        # This tests that PDB can jump back to the first line in a
-        # file.  See issue #1689458.  It can only be triggered in a
-        # function call if the function is defined on a single line.
-        code = compile("""
-# Comments don't count.
-output.append(2)  # firstlineno is here.
-output.append(3)
-output.append(4)
-""", "<fake module>", "exec")
-        class fake_function:
-            func_code = code
-            jump = (2, 0)
-        tracer = JumpTracer(fake_function)
-        sys.settrace(tracer.trace)
-        namespace = {"output": []}
-        exec code in namespace
-        sys.settrace(None)
-        self.compare_jump_output([2, 3, 2, 3, 4], namespace["output"])
-
-
-def test_main():
-    test_support.run_unittest(
-        TraceTestCase,
-        RaisingTraceFuncTestCase,
-        JumpTestCase
-    )
-
-if __name__ == "__main__":
-    test_main()

From python-checkins at python.org  Sun Jul 25 17:13:42 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Sun, 25 Jul 2010 17:13:42 +0200 (CEST)
Subject: [Python-checkins] r83143 - in python/branches/release31-maint:
	Lib/test/test_profilehooks.py Lib/test/test_sys.py
	Lib/test/test_trace.py
Message-ID: <20100725151342.D947FF29A@mail.python.org>

Author: alexander.belopolsky
Date: Sun Jul 25 17:13:42 2010
New Revision: 83143

Log:
Merged revisions 83140-83141 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83140 | alexander.belopolsky | 2010-07-25 11:02:55 -0400 (Sun, 25 Jul 2010) | 5 lines
  
  Issue #9315: Renamed test_trace to test_sys_settrace and
  test_profilehooks to test_sys_setprofile so that test_trace can be
  used for testing the trace module and for naming consistency.
........
  r83141 | alexander.belopolsky | 2010-07-25 11:05:42 -0400 (Sun, 25 Jul 2010) | 1 line
  
  Corrected comments on where settrace and setprofile are tested.
........


Removed:
   python/branches/release31-maint/Lib/test/test_profilehooks.py
   python/branches/release31-maint/Lib/test/test_trace.py
Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/test/test_sys.py

Deleted: python/branches/release31-maint/Lib/test/test_profilehooks.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_profilehooks.py	Sun Jul 25 17:13:42 2010
+++ (empty file)
@@ -1,385 +0,0 @@
-import pprint
-import sys
-import unittest
-
-from test import support
-
-class TestGetProfile(unittest.TestCase):
-    def setUp(self):
-        sys.setprofile(None)
-
-    def tearDown(self):
-        sys.setprofile(None)
-
-    def test_empty(self):
-        assert sys.getprofile() == None
-
-    def test_setget(self):
-        def fn(*args):
-            pass
-
-        sys.setprofile(fn)
-        assert sys.getprofile() == fn
-
-class HookWatcher:
-    def __init__(self):
-        self.frames = []
-        self.events = []
-
-    def callback(self, frame, event, arg):
-        if (event == "call"
-            or event == "return"
-            or event == "exception"):
-            self.add_event(event, frame)
-
-    def add_event(self, event, frame=None):
-        """Add an event to the log."""
-        if frame is None:
-            frame = sys._getframe(1)
-
-        try:
-            frameno = self.frames.index(frame)
-        except ValueError:
-            frameno = len(self.frames)
-            self.frames.append(frame)
-
-        self.events.append((frameno, event, ident(frame)))
-
-    def get_events(self):
-        """Remove calls to add_event()."""
-        disallowed = [ident(self.add_event.__func__), ident(ident)]
-        self.frames = None
-
-        return [item for item in self.events if item[2] not in disallowed]
-
-
-class ProfileSimulator(HookWatcher):
-    def __init__(self, testcase):
-        self.testcase = testcase
-        self.stack = []
-        HookWatcher.__init__(self)
-
-    def callback(self, frame, event, arg):
-        # Callback registered with sys.setprofile()/sys.settrace()
-        self.dispatch[event](self, frame)
-
-    def trace_call(self, frame):
-        self.add_event('call', frame)
-        self.stack.append(frame)
-
-    def trace_return(self, frame):
-        self.add_event('return', frame)
-        self.stack.pop()
-
-    def trace_exception(self, frame):
-        self.testcase.fail(
-            "the profiler should never receive exception events")
-
-    def trace_pass(self, frame):
-        pass
-
-    dispatch = {
-        'call': trace_call,
-        'exception': trace_exception,
-        'return': trace_return,
-        'c_call': trace_pass,
-        'c_return': trace_pass,
-        'c_exception': trace_pass,
-        }
-
-
-class TestCaseBase(unittest.TestCase):
-    def check_events(self, callable, expected):
-        events = capture_events(callable, self.new_watcher())
-        if events != expected:
-            self.fail("Expected events:\n%s\nReceived events:\n%s"
-                      % (pprint.pformat(expected), pprint.pformat(events)))
-
-
-class ProfileHookTestCase(TestCaseBase):
-    def new_watcher(self):
-        return HookWatcher()
-
-    def test_simple(self):
-        def f(p):
-            pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_exception(self):
-        def f(p):
-            1/0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_caught_exception(self):
-        def f(p):
-            try: 1/0
-            except: pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_caught_nested_exception(self):
-        def f(p):
-            try: 1/0
-            except: pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_nested_exception(self):
-        def f(p):
-            1/0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              # This isn't what I expected:
-                              # (0, 'exception', protect_ident),
-                              # I expected this again:
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_exception_in_except_clause(self):
-        def f(p):
-            1/0
-        def g(p):
-            try:
-                f(p)
-            except:
-                try: f(p)
-                except: pass
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (3, 'call', f_ident),
-                              (3, 'return', f_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-    def test_exception_propogation(self):
-        def f(p):
-            1/0
-        def g(p):
-            try: f(p)
-            finally: p.add_event("falling through")
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (1, 'falling through', g_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-    def test_raise_twice(self):
-        def f(p):
-            try: 1/0
-            except: 1/0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_raise_reraise(self):
-        def f(p):
-            try: 1/0
-            except: raise
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_raise(self):
-        def f(p):
-            raise Exception()
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_distant_exception(self):
-        def f():
-            1/0
-        def g():
-            f()
-        def h():
-            g()
-        def i():
-            h()
-        def j(p):
-            i()
-        f_ident = ident(f)
-        g_ident = ident(g)
-        h_ident = ident(h)
-        i_ident = ident(i)
-        j_ident = ident(j)
-        self.check_events(j, [(1, 'call', j_ident),
-                              (2, 'call', i_ident),
-                              (3, 'call', h_ident),
-                              (4, 'call', g_ident),
-                              (5, 'call', f_ident),
-                              (5, 'return', f_ident),
-                              (4, 'return', g_ident),
-                              (3, 'return', h_ident),
-                              (2, 'return', i_ident),
-                              (1, 'return', j_ident),
-                              ])
-
-    def test_generator(self):
-        def f():
-            for i in range(2):
-                yield i
-        def g(p):
-            for i in f():
-                pass
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              # call the iterator twice to generate values
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              # once more; returns end-of-iteration with
-                              # actually raising an exception
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-    def test_stop_iteration(self):
-        def f():
-            for i in range(2):
-                yield i
-            raise StopIteration
-        def g(p):
-            for i in f():
-                pass
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              # call the iterator twice to generate values
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              # once more to hit the raise:
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-
-class ProfileSimulatorTestCase(TestCaseBase):
-    def new_watcher(self):
-        return ProfileSimulator(self)
-
-    def test_simple(self):
-        def f(p):
-            pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_basic_exception(self):
-        def f(p):
-            1/0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_caught_exception(self):
-        def f(p):
-            try: 1/0
-            except: pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_distant_exception(self):
-        def f():
-            1/0
-        def g():
-            f()
-        def h():
-            g()
-        def i():
-            h()
-        def j(p):
-            i()
-        f_ident = ident(f)
-        g_ident = ident(g)
-        h_ident = ident(h)
-        i_ident = ident(i)
-        j_ident = ident(j)
-        self.check_events(j, [(1, 'call', j_ident),
-                              (2, 'call', i_ident),
-                              (3, 'call', h_ident),
-                              (4, 'call', g_ident),
-                              (5, 'call', f_ident),
-                              (5, 'return', f_ident),
-                              (4, 'return', g_ident),
-                              (3, 'return', h_ident),
-                              (2, 'return', i_ident),
-                              (1, 'return', j_ident),
-                              ])
-
-
-def ident(function):
-    if hasattr(function, "f_code"):
-        code = function.f_code
-    else:
-        code = function.__code__
-    return code.co_firstlineno, code.co_name
-
-
-def protect(f, p):
-    try: f(p)
-    except: pass
-
-protect_ident = ident(protect)
-
-
-def capture_events(callable, p=None):
-    try:
-        sys.setprofile()
-    except TypeError:
-        pass
-    else:
-        raise support.TestFailed(
-            'sys.setprofile() did not raise TypeError')
-
-    if p is None:
-        p = HookWatcher()
-    sys.setprofile(p.callback)
-    protect(callable, p)
-    sys.setprofile(None)
-    return p.get_events()[1:-1]
-
-
-def show_events(callable):
-    import pprint
-    pprint.pprint(capture_events(callable))
-
-
-def test_main():
-    support.run_unittest(
-        TestGetProfile,
-        ProfileHookTestCase,
-        ProfileSimulatorTestCase
-    )
-
-
-if __name__ == "__main__":
-    test_main()

Modified: python/branches/release31-maint/Lib/test/test_sys.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_sys.py	(original)
+++ python/branches/release31-maint/Lib/test/test_sys.py	Sun Jul 25 17:13:42 2010
@@ -164,8 +164,8 @@
         # can't check more than the type, as the user might have changed it
         self.assertTrue(isinstance(sys.getdefaultencoding(), str))
 
-    # testing sys.settrace() is done in test_trace.py
-    # testing sys.setprofile() is done in test_profile.py
+    # testing sys.settrace() is done in test_sys_settrace.py
+    # testing sys.setprofile() is done in test_sys_setprofile.py
 
     def test_setcheckinterval(self):
         self.assertRaises(TypeError, sys.setcheckinterval)

Deleted: python/branches/release31-maint/Lib/test/test_trace.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_trace.py	Sun Jul 25 17:13:42 2010
+++ (empty file)
@@ -1,790 +0,0 @@
-# Testing the line trace facility.
-
-from test import support
-import unittest
-import sys
-import difflib
-import gc
-
-# A very basic example.  If this fails, we're in deep trouble.
-def basic():
-    return 1
-
-basic.events = [(0, 'call'),
-                (1, 'line'),
-                (1, 'return')]
-
-# Many of the tests below are tricky because they involve pass statements.
-# If there is implicit control flow around a pass statement (in an except
-# clause or else caluse) under what conditions do you set a line number
-# following that clause?
-
-
-# The entire "while 0:" statement is optimized away.  No code
-# exists for it, so the line numbers skip directly from "del x"
-# to "x = 1".
-def arigo_example():
-    x = 1
-    del x
-    while 0:
-        pass
-    x = 1
-
-arigo_example.events = [(0, 'call'),
-                        (1, 'line'),
-                        (2, 'line'),
-                        (5, 'line'),
-                        (5, 'return')]
-
-# check that lines consisting of just one instruction get traced:
-def one_instr_line():
-    x = 1
-    del x
-    x = 1
-
-one_instr_line.events = [(0, 'call'),
-                         (1, 'line'),
-                         (2, 'line'),
-                         (3, 'line'),
-                         (3, 'return')]
-
-def no_pop_tops():      # 0
-    x = 1               # 1
-    for a in range(2):  # 2
-        if a:           # 3
-            x = 1       # 4
-        else:           # 5
-            x = 1       # 6
-
-no_pop_tops.events = [(0, 'call'),
-                      (1, 'line'),
-                      (2, 'line'),
-                      (3, 'line'),
-                      (6, 'line'),
-                      (2, 'line'),
-                      (3, 'line'),
-                      (4, 'line'),
-                      (2, 'line'),
-                      (2, 'return')]
-
-def no_pop_blocks():
-    y = 1
-    while not y:
-        bla
-    x = 1
-
-no_pop_blocks.events = [(0, 'call'),
-                        (1, 'line'),
-                        (2, 'line'),
-                        (4, 'line'),
-                        (4, 'return')]
-
-def called(): # line -3
-    x = 1
-
-def call():   # line 0
-    called()
-
-call.events = [(0, 'call'),
-               (1, 'line'),
-               (-3, 'call'),
-               (-2, 'line'),
-               (-2, 'return'),
-               (1, 'return')]
-
-def raises():
-    raise Exception
-
-def test_raise():
-    try:
-        raises()
-    except Exception as exc:
-        x = 1
-
-test_raise.events = [(0, 'call'),
-                     (1, 'line'),
-                     (2, 'line'),
-                     (-3, 'call'),
-                     (-2, 'line'),
-                     (-2, 'exception'),
-                     (-2, 'return'),
-                     (2, 'exception'),
-                     (3, 'line'),
-                     (4, 'line'),
-                     (4, 'return')]
-
-def _settrace_and_return(tracefunc):
-    sys.settrace(tracefunc)
-    sys._getframe().f_back.f_trace = tracefunc
-def settrace_and_return(tracefunc):
-    _settrace_and_return(tracefunc)
-
-settrace_and_return.events = [(1, 'return')]
-
-def _settrace_and_raise(tracefunc):
-    sys.settrace(tracefunc)
-    sys._getframe().f_back.f_trace = tracefunc
-    raise RuntimeError
-def settrace_and_raise(tracefunc):
-    try:
-        _settrace_and_raise(tracefunc)
-    except RuntimeError as exc:
-        pass
-
-settrace_and_raise.events = [(2, 'exception'),
-                             (3, 'line'),
-                             (4, 'line'),
-                             (4, 'return')]
-
-# implicit return example
-# This test is interesting because of the else: pass
-# part of the code.  The code generate for the true
-# part of the if contains a jump past the else branch.
-# The compiler then generates an implicit "return None"
-# Internally, the compiler visits the pass statement
-# and stores its line number for use on the next instruction.
-# The next instruction is the implicit return None.
-def ireturn_example():
-    a = 5
-    b = 5
-    if a == b:
-        b = a+1
-    else:
-        pass
-
-ireturn_example.events = [(0, 'call'),
-                          (1, 'line'),
-                          (2, 'line'),
-                          (3, 'line'),
-                          (4, 'line'),
-                          (6, 'line'),
-                          (6, 'return')]
-
-# Tight loop with while(1) example (SF #765624)
-def tightloop_example():
-    items = range(0, 3)
-    try:
-        i = 0
-        while 1:
-            b = items[i]; i+=1
-    except IndexError:
-        pass
-
-tightloop_example.events = [(0, 'call'),
-                            (1, 'line'),
-                            (2, 'line'),
-                            (3, 'line'),
-                            (4, 'line'),
-                            (5, 'line'),
-                            (5, 'line'),
-                            (5, 'line'),
-                            (5, 'line'),
-                            (5, 'exception'),
-                            (6, 'line'),
-                            (7, 'line'),
-                            (7, 'return')]
-
-def tighterloop_example():
-    items = range(1, 4)
-    try:
-        i = 0
-        while 1: i = items[i]
-    except IndexError:
-        pass
-
-tighterloop_example.events = [(0, 'call'),
-                            (1, 'line'),
-                            (2, 'line'),
-                            (3, 'line'),
-                            (4, 'line'),
-                            (4, 'line'),
-                            (4, 'line'),
-                            (4, 'line'),
-                            (4, 'exception'),
-                            (5, 'line'),
-                            (6, 'line'),
-                            (6, 'return')]
-
-def generator_function():
-    try:
-        yield True
-        "continued"
-    finally:
-        "finally"
-def generator_example():
-    # any() will leave the generator before its end
-    x = any(generator_function())
-
-    # the following lines were not traced
-    for x in range(10):
-        y = x
-
-generator_example.events = ([(0, 'call'),
-                             (2, 'line'),
-                             (-6, 'call'),
-                             (-5, 'line'),
-                             (-4, 'line'),
-                             (-4, 'return'),
-                             (-4, 'call'),
-                             (-4, 'exception'),
-                             (-1, 'line'),
-                             (-1, 'return')] +
-                            [(5, 'line'), (6, 'line')] * 10 +
-                            [(5, 'line'), (5, 'return')])
-
-
-class Tracer:
-    def __init__(self):
-        self.events = []
-    def trace(self, frame, event, arg):
-        self.events.append((frame.f_lineno, event))
-        return self.trace
-    def traceWithGenexp(self, frame, event, arg):
-        (o for o in [1])
-        self.events.append((frame.f_lineno, event))
-        return self.trace
-
-class TraceTestCase(unittest.TestCase):
-
-    # Disable gc collection when tracing, otherwise the
-    # deallocators may be traced as well.
-    def setUp(self):
-        self.using_gc = gc.isenabled()
-        gc.disable()
-
-    def tearDown(self):
-        if self.using_gc:
-            gc.enable()
-
-    def compare_events(self, line_offset, events, expected_events):
-        events = [(l - line_offset, e) for (l, e) in events]
-        if events != expected_events:
-            self.fail(
-                "events did not match expectation:\n" +
-                "\n".join(difflib.ndiff([str(x) for x in expected_events],
-                                        [str(x) for x in events])))
-
-    def run_and_compare(self, func, events):
-        tracer = Tracer()
-        sys.settrace(tracer.trace)
-        func()
-        sys.settrace(None)
-        self.compare_events(func.__code__.co_firstlineno,
-                            tracer.events, events)
-
-    def run_test(self, func):
-        self.run_and_compare(func, func.events)
-
-    def run_test2(self, func):
-        tracer = Tracer()
-        func(tracer.trace)
-        sys.settrace(None)
-        self.compare_events(func.__code__.co_firstlineno,
-                            tracer.events, func.events)
-
-    def set_and_retrieve_none(self):
-        sys.settrace(None)
-        assert sys.gettrace() is None
-
-    def set_and_retrieve_func(self):
-        def fn(*args):
-            pass
-
-        sys.settrace(fn)
-        try:
-            assert sys.gettrace() is fn
-        finally:
-            sys.settrace(None)
-
-    def test_01_basic(self):
-        self.run_test(basic)
-    def test_02_arigo(self):
-        self.run_test(arigo_example)
-    def test_03_one_instr(self):
-        self.run_test(one_instr_line)
-    def test_04_no_pop_blocks(self):
-        self.run_test(no_pop_blocks)
-    def test_05_no_pop_tops(self):
-        self.run_test(no_pop_tops)
-    def test_06_call(self):
-        self.run_test(call)
-    def test_07_raise(self):
-        self.run_test(test_raise)
-
-    def test_08_settrace_and_return(self):
-        self.run_test2(settrace_and_return)
-    def test_09_settrace_and_raise(self):
-        self.run_test2(settrace_and_raise)
-    def test_10_ireturn(self):
-        self.run_test(ireturn_example)
-    def test_11_tightloop(self):
-        self.run_test(tightloop_example)
-    def test_12_tighterloop(self):
-        self.run_test(tighterloop_example)
-
-    def test_13_genexp(self):
-        self.run_test(generator_example)
-        # issue1265: if the trace function contains a generator,
-        # and if the traced function contains another generator
-        # that is not completely exhausted, the trace stopped.
-        # Worse: the 'finally' clause was not invoked.
-        tracer = Tracer()
-        sys.settrace(tracer.traceWithGenexp)
-        generator_example()
-        sys.settrace(None)
-        self.compare_events(generator_example.__code__.co_firstlineno,
-                            tracer.events, generator_example.events)
-
-    def test_14_onliner_if(self):
-        def onliners():
-            if True: False
-            else: True
-            return 0
-        self.run_and_compare(
-            onliners,
-            [(0, 'call'),
-             (1, 'line'),
-             (3, 'line'),
-             (3, 'return')])
-
-    def test_15_loops(self):
-        # issue1750076: "while" expression is skipped by debugger
-        def for_example():
-            for x in range(2):
-                pass
-        self.run_and_compare(
-            for_example,
-            [(0, 'call'),
-             (1, 'line'),
-             (2, 'line'),
-             (1, 'line'),
-             (2, 'line'),
-             (1, 'line'),
-             (1, 'return')])
-
-        def while_example():
-            # While expression should be traced on every loop
-            x = 2
-            while x > 0:
-                x -= 1
-        self.run_and_compare(
-            while_example,
-            [(0, 'call'),
-             (2, 'line'),
-             (3, 'line'),
-             (4, 'line'),
-             (3, 'line'),
-             (4, 'line'),
-             (3, 'line'),
-             (3, 'return')])
-
-    def test_16_blank_lines(self):
-        namespace = {}
-        exec("def f():\n" + "\n" * 256 + "    pass", namespace)
-        self.run_and_compare(
-            namespace["f"],
-            [(0, 'call'),
-             (257, 'line'),
-             (257, 'return')])
-
-
-class RaisingTraceFuncTestCase(unittest.TestCase):
-    def trace(self, frame, event, arg):
-        """A trace function that raises an exception in response to a
-        specific trace event."""
-        if event == self.raiseOnEvent:
-            raise ValueError # just something that isn't RuntimeError
-        else:
-            return self.trace
-
-    def f(self):
-        """The function to trace; raises an exception if that's the case
-        we're testing, so that the 'exception' trace event fires."""
-        if self.raiseOnEvent == 'exception':
-            x = 0
-            y = 1/x
-        else:
-            return 1
-
-    def run_test_for_event(self, event):
-        """Tests that an exception raised in response to the given event is
-        handled OK."""
-        self.raiseOnEvent = event
-        try:
-            for i in range(sys.getrecursionlimit() + 1):
-                sys.settrace(self.trace)
-                try:
-                    self.f()
-                except ValueError:
-                    pass
-                else:
-                    self.fail("exception not thrown!")
-        except RuntimeError:
-            self.fail("recursion counter not reset")
-
-    # Test the handling of exceptions raised by each kind of trace event.
-    def test_call(self):
-        self.run_test_for_event('call')
-    def test_line(self):
-        self.run_test_for_event('line')
-    def test_return(self):
-        self.run_test_for_event('return')
-    def test_exception(self):
-        self.run_test_for_event('exception')
-
-    def test_trash_stack(self):
-        def f():
-            for i in range(5):
-                print(i)  # line tracing will raise an exception at this line
-
-        def g(frame, why, extra):
-            if (why == 'line' and
-                frame.f_lineno == f.__code__.co_firstlineno + 2):
-                raise RuntimeError("i am crashing")
-            return g
-
-        sys.settrace(g)
-        try:
-            f()
-        except RuntimeError:
-            # the test is really that this doesn't segfault:
-            import gc
-            gc.collect()
-        else:
-            self.fail("exception not propagated")
-
-
-# 'Jump' tests: assigning to frame.f_lineno within a trace function
-# moves the execution position - it's how debuggers implement a Jump
-# command (aka. "Set next statement").
-
-class JumpTracer:
-    """Defines a trace function that jumps from one place to another,
-    with the source and destination lines of the jump being defined by
-    the 'jump' property of the function under test."""
-
-    def __init__(self, function):
-        self.function = function
-        self.jumpFrom = function.jump[0]
-        self.jumpTo = function.jump[1]
-        self.done = False
-
-    def trace(self, frame, event, arg):
-        if not self.done and frame.f_code == self.function.__code__:
-            firstLine = frame.f_code.co_firstlineno
-            if event == 'line' and frame.f_lineno == firstLine + self.jumpFrom:
-                # Cope with non-integer self.jumpTo (because of
-                # no_jump_to_non_integers below).
-                try:
-                    frame.f_lineno = firstLine + self.jumpTo
-                except TypeError:
-                    frame.f_lineno = self.jumpTo
-                self.done = True
-        return self.trace
-
-# The first set of 'jump' tests are for things that are allowed:
-
-def jump_simple_forwards(output):
-    output.append(1)
-    output.append(2)
-    output.append(3)
-
-jump_simple_forwards.jump = (1, 3)
-jump_simple_forwards.output = [3]
-
-def jump_simple_backwards(output):
-    output.append(1)
-    output.append(2)
-
-jump_simple_backwards.jump = (2, 1)
-jump_simple_backwards.output = [1, 1, 2]
-
-def jump_out_of_block_forwards(output):
-    for i in 1, 2:
-        output.append(2)
-        for j in [3]:  # Also tests jumping over a block
-            output.append(4)
-    output.append(5)
-
-jump_out_of_block_forwards.jump = (3, 5)
-jump_out_of_block_forwards.output = [2, 5]
-
-def jump_out_of_block_backwards(output):
-    output.append(1)
-    for i in [1]:
-        output.append(3)
-        for j in [2]:  # Also tests jumping over a block
-            output.append(5)
-        output.append(6)
-    output.append(7)
-
-jump_out_of_block_backwards.jump = (6, 1)
-jump_out_of_block_backwards.output = [1, 3, 5, 1, 3, 5, 6, 7]
-
-def jump_to_codeless_line(output):
-    output.append(1)
-    # Jumping to this line should skip to the next one.
-    output.append(3)
-
-jump_to_codeless_line.jump = (1, 2)
-jump_to_codeless_line.output = [3]
-
-def jump_to_same_line(output):
-    output.append(1)
-    output.append(2)
-    output.append(3)
-
-jump_to_same_line.jump = (2, 2)
-jump_to_same_line.output = [1, 2, 3]
-
-# Tests jumping within a finally block, and over one.
-def jump_in_nested_finally(output):
-    try:
-        output.append(2)
-    finally:
-        output.append(4)
-        try:
-            output.append(6)
-        finally:
-            output.append(8)
-        output.append(9)
-
-jump_in_nested_finally.jump = (4, 9)
-jump_in_nested_finally.output = [2, 9]
-
-# The second set of 'jump' tests are for things that are not allowed:
-
-def no_jump_too_far_forwards(output):
-    try:
-        output.append(2)
-        output.append(3)
-    except ValueError as e:
-        output.append('after' in str(e))
-
-no_jump_too_far_forwards.jump = (3, 6)
-no_jump_too_far_forwards.output = [2, True]
-
-def no_jump_too_far_backwards(output):
-    try:
-        output.append(2)
-        output.append(3)
-    except ValueError as e:
-        output.append('before' in str(e))
-
-no_jump_too_far_backwards.jump = (3, -1)
-no_jump_too_far_backwards.output = [2, True]
-
-# Test each kind of 'except' line.
-def no_jump_to_except_1(output):
-    try:
-        output.append(2)
-    except:
-        e = sys.exc_info()[1]
-        output.append('except' in str(e))
-
-no_jump_to_except_1.jump = (2, 3)
-no_jump_to_except_1.output = [True]
-
-def no_jump_to_except_2(output):
-    try:
-        output.append(2)
-    except ValueError:
-        e = sys.exc_info()[1]
-        output.append('except' in str(e))
-
-no_jump_to_except_2.jump = (2, 3)
-no_jump_to_except_2.output = [True]
-
-def no_jump_to_except_3(output):
-    try:
-        output.append(2)
-    except ValueError as e:
-        output.append('except' in str(e))
-
-no_jump_to_except_3.jump = (2, 3)
-no_jump_to_except_3.output = [True]
-
-def no_jump_to_except_4(output):
-    try:
-        output.append(2)
-    except (ValueError, RuntimeError) as e:
-        output.append('except' in str(e))
-
-no_jump_to_except_4.jump = (2, 3)
-no_jump_to_except_4.output = [True]
-
-def no_jump_forwards_into_block(output):
-    try:
-        output.append(2)
-        for i in 1, 2:
-            output.append(4)
-    except ValueError as e:
-        output.append('into' in str(e))
-
-no_jump_forwards_into_block.jump = (2, 4)
-no_jump_forwards_into_block.output = [True]
-
-def no_jump_backwards_into_block(output):
-    try:
-        for i in 1, 2:
-            output.append(3)
-        output.append(4)
-    except ValueError as e:
-        output.append('into' in str(e))
-
-no_jump_backwards_into_block.jump = (4, 3)
-no_jump_backwards_into_block.output = [3, 3, True]
-
-def no_jump_into_finally_block(output):
-    try:
-        try:
-            output.append(3)
-            x = 1
-        finally:
-            output.append(6)
-    except ValueError as e:
-        output.append('finally' in str(e))
-
-no_jump_into_finally_block.jump = (4, 6)
-no_jump_into_finally_block.output = [3, 6, True]  # The 'finally' still runs
-
-def no_jump_out_of_finally_block(output):
-    try:
-        try:
-            output.append(3)
-        finally:
-            output.append(5)
-            output.append(6)
-    except ValueError as e:
-        output.append('finally' in str(e))
-
-no_jump_out_of_finally_block.jump = (5, 1)
-no_jump_out_of_finally_block.output = [3, True]
-
-# This verifies the line-numbers-must-be-integers rule.
-def no_jump_to_non_integers(output):
-    try:
-        output.append(2)
-    except ValueError as e:
-        output.append('integer' in str(e))
-
-no_jump_to_non_integers.jump = (2, "Spam")
-no_jump_to_non_integers.output = [True]
-
-# This verifies that you can't set f_lineno via _getframe or similar
-# trickery.
-def no_jump_without_trace_function():
-    try:
-        previous_frame = sys._getframe().f_back
-        previous_frame.f_lineno = previous_frame.f_lineno
-    except ValueError as e:
-        # This is the exception we wanted; make sure the error message
-        # talks about trace functions.
-        if 'trace' not in str(e):
-            raise
-    else:
-        # Something's wrong - the expected exception wasn't raised.
-        raise RuntimeError("Trace-function-less jump failed to fail")
-
-
-class JumpTestCase(unittest.TestCase):
-    def compare_jump_output(self, expected, received):
-        if received != expected:
-            self.fail( "Outputs don't match:\n" +
-                       "Expected: " + repr(expected) + "\n" +
-                       "Received: " + repr(received))
-
-    def run_test(self, func):
-        tracer = JumpTracer(func)
-        sys.settrace(tracer.trace)
-        output = []
-        func(output)
-        sys.settrace(None)
-        self.compare_jump_output(func.output, output)
-
-    def test_01_jump_simple_forwards(self):
-        self.run_test(jump_simple_forwards)
-    def test_02_jump_simple_backwards(self):
-        self.run_test(jump_simple_backwards)
-    def test_03_jump_out_of_block_forwards(self):
-        self.run_test(jump_out_of_block_forwards)
-    def test_04_jump_out_of_block_backwards(self):
-        self.run_test(jump_out_of_block_backwards)
-    def test_05_jump_to_codeless_line(self):
-        self.run_test(jump_to_codeless_line)
-    def test_06_jump_to_same_line(self):
-        self.run_test(jump_to_same_line)
-    def test_07_jump_in_nested_finally(self):
-        self.run_test(jump_in_nested_finally)
-    def test_08_no_jump_too_far_forwards(self):
-        self.run_test(no_jump_too_far_forwards)
-    def test_09_no_jump_too_far_backwards(self):
-        self.run_test(no_jump_too_far_backwards)
-    def test_10_no_jump_to_except_1(self):
-        self.run_test(no_jump_to_except_1)
-    def test_11_no_jump_to_except_2(self):
-        self.run_test(no_jump_to_except_2)
-    def test_12_no_jump_to_except_3(self):
-        self.run_test(no_jump_to_except_3)
-    def test_13_no_jump_to_except_4(self):
-        self.run_test(no_jump_to_except_4)
-    def test_14_no_jump_forwards_into_block(self):
-        self.run_test(no_jump_forwards_into_block)
-    def test_15_no_jump_backwards_into_block(self):
-        self.run_test(no_jump_backwards_into_block)
-    def test_16_no_jump_into_finally_block(self):
-        self.run_test(no_jump_into_finally_block)
-    def test_17_no_jump_out_of_finally_block(self):
-        self.run_test(no_jump_out_of_finally_block)
-    def test_18_no_jump_to_non_integers(self):
-        self.run_test(no_jump_to_non_integers)
-    def test_19_no_jump_without_trace_function(self):
-        no_jump_without_trace_function()
-
-    def test_20_large_function(self):
-        d = {}
-        exec("""def f(output):        # line 0
-            x = 0                     # line 1
-            y = 1                     # line 2
-            '''                       # line 3
-            %s                        # lines 4-1004
-            '''                       # line 1005
-            x += 1                    # line 1006
-            output.append(x)          # line 1007
-            return""" % ('\n' * 1000,), d)
-        f = d['f']
-
-        f.jump = (2, 1007)
-        f.output = [0]
-        self.run_test(f)
-
-    def test_jump_to_firstlineno(self):
-        # This tests that PDB can jump back to the first line in a
-        # file.  See issue #1689458.  It can only be triggered in a
-        # function call if the function is defined on a single line.
-        code = compile("""
-# Comments don't count.
-output.append(2)  # firstlineno is here.
-output.append(3)
-output.append(4)
-""", "<fake module>", "exec")
-        class fake_function:
-            __code__ = code
-            jump = (2, 0)
-        tracer = JumpTracer(fake_function)
-        sys.settrace(tracer.trace)
-        namespace = {"output": []}
-        exec(code, namespace)
-        sys.settrace(None)
-        self.compare_jump_output([2, 3, 2, 3, 4], namespace["output"])
-
-
-def test_main():
-    support.run_unittest(
-        TraceTestCase,
-        RaisingTraceFuncTestCase,
-        JumpTestCase
-    )
-
-if __name__ == "__main__":
-    test_main()

From python-checkins at python.org  Sun Jul 25 17:23:24 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Sun, 25 Jul 2010 17:23:24 +0200 (CEST)
Subject: [Python-checkins] r83144 - in python/branches/release31-maint:
	Lib/test/test_profilehooks.py Lib/test/test_sys.py
	Lib/test/test_trace.py
Message-ID: <20100725152324.A9C26C3F6@mail.python.org>

Author: alexander.belopolsky
Date: Sun Jul 25 17:23:24 2010
New Revision: 83144

Log:
Reverted r83143: svnmerge failed to add new names

Added:
   python/branches/release31-maint/Lib/test/test_profilehooks.py
      - copied unchanged from r83142, /python/branches/release31-maint/Lib/test/test_profilehooks.py
   python/branches/release31-maint/Lib/test/test_trace.py
      - copied unchanged from r83142, /python/branches/release31-maint/Lib/test/test_trace.py
Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/test/test_sys.py

Modified: python/branches/release31-maint/Lib/test/test_sys.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_sys.py	(original)
+++ python/branches/release31-maint/Lib/test/test_sys.py	Sun Jul 25 17:23:24 2010
@@ -164,8 +164,8 @@
         # can't check more than the type, as the user might have changed it
         self.assertTrue(isinstance(sys.getdefaultencoding(), str))
 
-    # testing sys.settrace() is done in test_sys_settrace.py
-    # testing sys.setprofile() is done in test_sys_setprofile.py
+    # testing sys.settrace() is done in test_trace.py
+    # testing sys.setprofile() is done in test_profile.py
 
     def test_setcheckinterval(self):
         self.assertRaises(TypeError, sys.setcheckinterval)

From alexander.belopolsky at gmail.com  Sun Jul 25 17:33:58 2010
From: alexander.belopolsky at gmail.com (Alexander Belopolsky)
Date: Sun, 25 Jul 2010 11:33:58 -0400
Subject: [Python-checkins] r79102 not in release31-maint
Message-ID: <AANLkTins4qd_z=wsVreS6-UE0SZx5LYAws4VhRcJBsDG@mail.gmail.com>

Does anyone know why r79102 is not in release31-maint?  It looks like
this prevents me from merging r83140.

------------------------------------------------------------------------
r79102 | florent.xicluna | 2010-03-19 15:00:44 -0400 (Fri, 19 Mar
2010) | 9 lines

Merged revisions 79100 via svnmerge from
svn+ssh://pythondev at svn.python.org/python/trunk

........
  r79100 | florent.xicluna | 2010-03-19 19:34:55 +0100 (ven, 19 mar
2010) | 2 lines

  Various tests cleanup: check_warnings/check_py3k_warnings,
unittest.assert* and setUp/tearDown.
........

------------------------------------------------------------------------

From python-checkins at python.org  Sun Jul 25 21:11:36 2010
From: python-checkins at python.org (gregory.p.smith)
Date: Sun, 25 Jul 2010 21:11:36 +0200 (CEST)
Subject: [Python-checkins] r83145 - in python/branches/release27-maint:
	Lib/cookielib.py Lib/test/test_cookielib.py Misc/NEWS
Message-ID: <20100725191136.E3786F3BD@mail.python.org>

Author: gregory.p.smith
Date: Sun Jul 25 21:11:36 2010
New Revision: 83145

Log:
Fixes issue #3704: cookielib was not properly handling URLs with a / in the
parameters.


Modified:
   python/branches/release27-maint/Lib/cookielib.py
   python/branches/release27-maint/Lib/test/test_cookielib.py
   python/branches/release27-maint/Misc/NEWS

Modified: python/branches/release27-maint/Lib/cookielib.py
==============================================================================
--- python/branches/release27-maint/Lib/cookielib.py	(original)
+++ python/branches/release27-maint/Lib/cookielib.py	Sun Jul 25 21:11:36 2010
@@ -607,19 +607,14 @@
     return req_host, erhn
 
 def request_path(request):
-    """request-URI, as defined by RFC 2965."""
+    """Path component of request-URI, as defined by RFC 2965."""
     url = request.get_full_url()
-    #scheme, netloc, path, parameters, query, frag = urlparse.urlparse(url)
-    #req_path = escape_path("".join(urlparse.urlparse(url)[2:]))
-    path, parameters, query, frag = urlparse.urlparse(url)[2:]
-    if parameters:
-        path = "%s;%s" % (path, parameters)
-    path = escape_path(path)
-    req_path = urlparse.urlunparse(("", "", path, "", query, frag))
-    if not req_path.startswith("/"):
+    parts = urlparse.urlsplit(url)
+    path = escape_path(parts.path)
+    if not path.startswith("/"):
         # fix bad RFC 2396 absoluteURI
-        req_path = "/"+req_path
-    return req_path
+        path = "/" + path
+    return path
 
 def request_port(request):
     host = request.get_host()

Modified: python/branches/release27-maint/Lib/test/test_cookielib.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_cookielib.py	(original)
+++ python/branches/release27-maint/Lib/test/test_cookielib.py	Sun Jul 25 21:11:36 2010
@@ -1,11 +1,16 @@
 # -*- coding: latin-1 -*-
 """Tests for cookielib.py."""
 
-import re, os, time
+import cookielib
+import os
+import re
+import time
+
 from unittest import TestCase
 
 from test import test_support
 
+
 class DateTimeTests(TestCase):
 
     def test_time2isoz(self):
@@ -563,6 +568,16 @@
         interact_netscape(c, "http://www.acme.com/blah/rhubarb/", 'eggs="bar"')
         self.assertIn("/blah/rhubarb", c._cookies["www.acme.com"])
 
+    def test_default_path_with_query(self):
+        cj = cookielib.CookieJar()
+        uri = "http://example.com/?spam/eggs"
+        value = 'eggs="bar"'
+        interact_netscape(cj, uri, value)
+        # default path does not include query, so is "/", not "/?spam"
+        self.assertIn("/", cj._cookies["example.com"])
+        # cookie is sent back to the same URI
+        self.assertEquals(interact_netscape(cj, uri), value)
+
     def test_escape_path(self):
         from cookielib import escape_path
         cases = [
@@ -591,15 +606,14 @@
         from urllib2 import Request
         from cookielib import request_path
         # with parameters
-        req = Request("http://www.example.com/rheum/rhaponicum;"
+        req = Request("http://www.example.com/rheum/rhaponticum;"
                       "foo=bar;sing=song?apples=pears&spam=eggs#ni")
-        self.assertEquals(request_path(req), "/rheum/rhaponicum;"
-                     "foo=bar;sing=song?apples=pears&spam=eggs#ni")
+        self.assertEquals(request_path(req),
+                          "/rheum/rhaponticum;foo=bar;sing=song")
         # without parameters
-        req = Request("http://www.example.com/rheum/rhaponicum?"
+        req = Request("http://www.example.com/rheum/rhaponticum?"
                       "apples=pears&spam=eggs#ni")
-        self.assertEquals(request_path(req), "/rheum/rhaponicum?"
-                     "apples=pears&spam=eggs#ni")
+        self.assertEquals(request_path(req), "/rheum/rhaponticum")
         # missing final slash
         req = Request("http://www.example.com")
         self.assertEquals(request_path(req), "/")

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Sun Jul 25 21:11:36 2010
@@ -18,6 +18,9 @@
 Library
 -------
 
+- Issue #3704: cookielib was not properly handling URLs with a / in the
+  parameters.
+
 - Issue #9032: XML-RPC client retries the request on EPIPE error. The EPIPE
   error occurs when the server closes the socket and the client sends a big
   XML-RPC request.

From python-checkins at python.org  Sun Jul 25 21:53:20 2010
From: python-checkins at python.org (gregory.p.smith)
Date: Sun, 25 Jul 2010 21:53:20 +0200 (CEST)
Subject: [Python-checkins] r83146 - in python/branches/release31-maint:
	Lib/http/cookiejar.py Lib/test/test_http_cookiejar.py Misc/NEWS
Message-ID: <20100725195320.A06B8F2CB@mail.python.org>

Author: gregory.p.smith
Date: Sun Jul 25 21:53:20 2010
New Revision: 83146

Log:
Merged revisions 82985 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82985 | gregory.p.smith | 2010-07-19 16:17:22 -0700 (Mon, 19 Jul 2010) | 3 lines
  
  Fixes Issue #3704: http.cookiejar was not properly handling URLs with a / in
  the parameters.  (This is jjlee's issue3704.patch ported to py3k)
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/http/cookiejar.py
   python/branches/release31-maint/Lib/test/test_http_cookiejar.py
   python/branches/release31-maint/Misc/NEWS

Modified: python/branches/release31-maint/Lib/http/cookiejar.py
==============================================================================
--- python/branches/release31-maint/Lib/http/cookiejar.py	(original)
+++ python/branches/release31-maint/Lib/http/cookiejar.py	Sun Jul 25 21:53:20 2010
@@ -603,17 +603,14 @@
     return req_host, erhn
 
 def request_path(request):
-    """request-URI, as defined by RFC 2965."""
+    """Path component of request-URI, as defined by RFC 2965."""
     url = request.get_full_url()
-    path, parameters, query, frag = urllib.parse.urlparse(url)[2:]
-    if parameters:
-        path = "%s;%s" % (path, parameters)
-    path = escape_path(path)
-    req_path = urllib.parse.urlunparse(("", "", path, "", query, frag))
-    if not req_path.startswith("/"):
+    parts = urllib.parse.urlsplit(url)
+    path = escape_path(parts.path)
+    if not path.startswith("/"):
         # fix bad RFC 2396 absoluteURI
-        req_path = "/"+req_path
-    return req_path
+        path = "/" + path
+    return path
 
 def request_port(request):
     host = request.get_host()

Modified: python/branches/release31-maint/Lib/test/test_http_cookiejar.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_http_cookiejar.py	(original)
+++ python/branches/release31-maint/Lib/test/test_http_cookiejar.py	Sun Jul 25 21:53:20 2010
@@ -1,18 +1,21 @@
 """Tests for http/cookiejar.py."""
 
-import re, os, time, urllib.request
-from unittest import TestCase
+import os
+import re
+import test.support
+import time
+import unittest
+import urllib.request
+
+from http.cookiejar import (time2isoz, http2time, time2netscape,
+     parse_ns_headers, join_header_words, split_header_words, Cookie,
+     CookieJar, DefaultCookiePolicy, LWPCookieJar, MozillaCookieJar,
+     LoadError, lwp_cookie_str, DEFAULT_HTTP_PORT, escape_path,
+     reach, is_HDN, domain_match, user_domain_match, request_path,
+     request_port, request_host)
 
-from test import support
 
-from http.cookiejar import time2isoz, http2time, time2netscape, \
-     parse_ns_headers, join_header_words, split_header_words, Cookie, \
-     CookieJar, DefaultCookiePolicy, LWPCookieJar, MozillaCookieJar, \
-     LoadError, lwp_cookie_str, DEFAULT_HTTP_PORT, escape_path, \
-     reach, is_HDN, domain_match, user_domain_match, request_path, \
-     request_port, request_host
-
-class DateTimeTests(TestCase):
+class DateTimeTests(unittest.TestCase):
 
     def test_time2isoz(self):
         base = 1019227000
@@ -96,7 +99,7 @@
                          )
 
 
-class HeaderTests(TestCase):
+class HeaderTests(unittest.TestCase):
     def test_parse_ns_headers(self):
         # quotes should be stripped
         expected = [[('foo', 'bar'), ('expires', 2209069412), ('version', '0')]]
@@ -217,10 +220,10 @@
     return cookie_hdr
 
 
-class FileCookieJarTests(TestCase):
+class FileCookieJarTests(unittest.TestCase):
     def test_lwp_valueless_cookie(self):
         # cookies with no value should be saved and loaded consistently
-        filename = support.TESTFN
+        filename = test.support.TESTFN
         c = LWPCookieJar()
         interact_netscape(c, "http://www.acme.com/", 'boo')
         self.assertEqual(c._cookies["www.acme.com"]["/"]["boo"].value, None)
@@ -235,7 +238,7 @@
 
     def test_bad_magic(self):
         # IOErrors (eg. file doesn't exist) are allowed to propagate
-        filename = support.TESTFN
+        filename = test.support.TESTFN
         for cookiejar_class in LWPCookieJar, MozillaCookieJar:
             c = cookiejar_class()
             try:
@@ -258,7 +261,7 @@
             try: os.unlink(filename)
             except OSError: pass
 
-class CookieTests(TestCase):
+class CookieTests(unittest.TestCase):
     # XXX
     # Get rid of string comparisons where not actually testing str / repr.
     # .clear() etc.
@@ -338,7 +341,7 @@
     def test_missing_value(self):
         # missing = sign in Cookie: header is regarded by Mozilla as a missing
         # name, and by http.cookiejar as a missing value
-        filename = support.TESTFN
+        filename = test.support.TESTFN
         c = MozillaCookieJar(filename)
         interact_netscape(c, "http://www.acme.com/", 'eggs')
         interact_netscape(c, "http://www.acme.com/", '"spam"; path=/foo/')
@@ -522,6 +525,16 @@
         interact_netscape(c, "http://www.acme.com/blah/rhubarb/", 'eggs="bar"')
         self.assertTrue("/blah/rhubarb" in c._cookies["www.acme.com"])
 
+    def test_default_path_with_query(self):
+        cj = CookieJar()
+        uri = "http://example.com/?spam/eggs"
+        value = 'eggs="bar"'
+        interact_netscape(cj, uri, value)
+        # Default path does not include query, so is "/", not "/?spam".
+        self.assertIn("/", cj._cookies["example.com"])
+        # Cookie is sent back to the same URI.
+        self.assertEquals(interact_netscape(cj, uri), value)
+
     def test_escape_path(self):
         cases = [
             # quoted safe
@@ -550,16 +563,15 @@
     def test_request_path(self):
         # with parameters
         req = urllib.request.Request(
-            "http://www.example.com/rheum/rhaponicum;"
+            "http://www.example.com/rheum/rhaponticum;"
             "foo=bar;sing=song?apples=pears&spam=eggs#ni")
-        self.assertEquals(request_path(req), "/rheum/rhaponicum;"
-                     "foo=bar;sing=song?apples=pears&spam=eggs#ni")
+        self.assertEquals(request_path(req),
+                          "/rheum/rhaponticum;foo=bar;sing=song")
         # without parameters
         req = urllib.request.Request(
-            "http://www.example.com/rheum/rhaponicum?"
+            "http://www.example.com/rheum/rhaponticum?"
             "apples=pears&spam=eggs#ni")
-        self.assertEquals(request_path(req), "/rheum/rhaponicum?"
-                     "apples=pears&spam=eggs#ni")
+        self.assertEquals(request_path(req), "/rheum/rhaponticum")
         # missing final slash
         req = urllib.request.Request("http://www.example.com")
         self.assertEquals(request_path(req), "/")
@@ -1033,7 +1045,7 @@
         self.assertTrue(cookie.expires is None)
 
 
-class LWPCookieTests(TestCase):
+class LWPCookieTests(unittest.TestCase):
     # Tests taken from libwww-perl, with a few modifications and additions.
 
     def test_netscape_example_1(self):
@@ -1423,7 +1435,7 @@
         self.assertEquals(len(c), 6)
 
         # save and restore
-        filename = support.TESTFN
+        filename = test.support.TESTFN
 
         try:
             c.save(filename, ignore_discard=True)
@@ -1462,7 +1474,7 @@
         # Save / load Mozilla/Netscape cookie file format.
         year_plus_one = time.localtime()[0] + 1
 
-        filename = support.TESTFN
+        filename = test.support.TESTFN
 
         c = MozillaCookieJar(filename,
                              policy=DefaultCookiePolicy(rfc2965=True))
@@ -1624,7 +1636,7 @@
 
 
 def test_main(verbose=None):
-    support.run_unittest(
+    test.support.run_unittest(
         DateTimeTests,
         HeaderTests,
         CookieTests,

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Sun Jul 25 21:53:20 2010
@@ -78,6 +78,9 @@
 Library
 -------
 
+- Issue #3704: http.cookiejar was not properly handling URLs with a / in the
+  parameters.
+
 - Issue #4629: getopt raises an error if an argument ends with = whereas getopt
   doesn't except a value (eg. --help= is rejected if getopt uses ['help='] long
   options).

From python-checkins at python.org  Sun Jul 25 22:00:02 2010
From: python-checkins at python.org (gregory.p.smith)
Date: Sun, 25 Jul 2010 22:00:02 +0200 (CEST)
Subject: [Python-checkins] r83147 - in python/branches/release26-maint:
	Lib/cookielib.py Lib/test/test_cookielib.py Misc/NEWS
Message-ID: <20100725200002.A1A39EE981@mail.python.org>

Author: gregory.p.smith
Date: Sun Jul 25 22:00:02 2010
New Revision: 83147

Log:
Merged revisions 83145 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

........
  r83145 | gregory.p.smith | 2010-07-25 12:11:36 -0700 (Sun, 25 Jul 2010) | 3 lines
  
  Fixes issue #3704: cookielib was not properly handling URLs with a / in the
  parameters.
........


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Lib/cookielib.py
   python/branches/release26-maint/Lib/test/test_cookielib.py
   python/branches/release26-maint/Misc/NEWS

Modified: python/branches/release26-maint/Lib/cookielib.py
==============================================================================
--- python/branches/release26-maint/Lib/cookielib.py	(original)
+++ python/branches/release26-maint/Lib/cookielib.py	Sun Jul 25 22:00:02 2010
@@ -607,19 +607,14 @@
     return req_host, erhn
 
 def request_path(request):
-    """request-URI, as defined by RFC 2965."""
+    """Path component of request-URI, as defined by RFC 2965."""
     url = request.get_full_url()
-    #scheme, netloc, path, parameters, query, frag = urlparse.urlparse(url)
-    #req_path = escape_path("".join(urlparse.urlparse(url)[2:]))
-    path, parameters, query, frag = urlparse.urlparse(url)[2:]
-    if parameters:
-        path = "%s;%s" % (path, parameters)
-    path = escape_path(path)
-    req_path = urlparse.urlunparse(("", "", path, "", query, frag))
-    if not req_path.startswith("/"):
+    parts = urlparse.urlsplit(url)
+    path = escape_path(parts.path)
+    if not path.startswith("/"):
         # fix bad RFC 2396 absoluteURI
-        req_path = "/"+req_path
-    return req_path
+        path = "/" + path
+    return path
 
 def request_port(request):
     host = request.get_host()

Modified: python/branches/release26-maint/Lib/test/test_cookielib.py
==============================================================================
--- python/branches/release26-maint/Lib/test/test_cookielib.py	(original)
+++ python/branches/release26-maint/Lib/test/test_cookielib.py	Sun Jul 25 22:00:02 2010
@@ -1,11 +1,16 @@
 # -*- coding: latin-1 -*-
 """Tests for cookielib.py."""
 
-import re, os, time
+import cookielib
+import os
+import re
+import time
+
 from unittest import TestCase
 
 from test import test_support
 
+
 class DateTimeTests(TestCase):
 
     def test_time2isoz(self):
@@ -563,6 +568,16 @@
         interact_netscape(c, "http://www.acme.com/blah/rhubarb/", 'eggs="bar"')
         self.assert_("/blah/rhubarb" in c._cookies["www.acme.com"])
 
+    def test_default_path_with_query(self):
+        cj = cookielib.CookieJar()
+        uri = "http://example.com/?spam/eggs"
+        value = 'eggs="bar"'
+        interact_netscape(cj, uri, value)
+        # default path does not include query, so is "/", not "/?spam"
+        self.assert_("/" in cj._cookies["example.com"])
+        # cookie is sent back to the same URI
+        self.assertEquals(interact_netscape(cj, uri), value)
+
     def test_escape_path(self):
         from cookielib import escape_path
         cases = [
@@ -591,15 +606,14 @@
         from urllib2 import Request
         from cookielib import request_path
         # with parameters
-        req = Request("http://www.example.com/rheum/rhaponicum;"
+        req = Request("http://www.example.com/rheum/rhaponticum;"
                       "foo=bar;sing=song?apples=pears&spam=eggs#ni")
-        self.assertEquals(request_path(req), "/rheum/rhaponicum;"
-                     "foo=bar;sing=song?apples=pears&spam=eggs#ni")
+        self.assertEquals(request_path(req),
+                          "/rheum/rhaponticum;foo=bar;sing=song")
         # without parameters
-        req = Request("http://www.example.com/rheum/rhaponicum?"
+        req = Request("http://www.example.com/rheum/rhaponticum?"
                       "apples=pears&spam=eggs#ni")
-        self.assertEquals(request_path(req), "/rheum/rhaponicum?"
-                     "apples=pears&spam=eggs#ni")
+        self.assertEquals(request_path(req), "/rheum/rhaponticum")
         # missing final slash
         req = Request("http://www.example.com")
         self.assertEquals(request_path(req), "/")

Modified: python/branches/release26-maint/Misc/NEWS
==============================================================================
--- python/branches/release26-maint/Misc/NEWS	(original)
+++ python/branches/release26-maint/Misc/NEWS	Sun Jul 25 22:00:02 2010
@@ -81,6 +81,9 @@
 Library
 -------
 
+- Issue #3704: cookielib was not properly handling URLs with a / in the
+  parameters.
+
 - Issue #4629: getopt raises an error if an argument ends with = whereas getopt
   doesn't except a value (eg. --help= is rejected if getopt uses ['help='] long
   options).

From python-checkins at python.org  Mon Jul 26 00:15:16 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Mon, 26 Jul 2010 00:15:16 +0200 (CEST)
Subject: [Python-checkins] r83148 - in python/branches/release31-maint:
	Lib/test/test_profilehooks.py Lib/test/test_sys.py
	Lib/test/test_sys_setprofile.py Lib/test/test_sys_settrace.py
	Lib/test/test_trace.py
Message-ID: <20100725221516.3EE28EE99C@mail.python.org>

Author: alexander.belopolsky
Date: Mon Jul 26 00:15:16 2010
New Revision: 83148

Log:
Merged revisions 83140-83141 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83140 | alexander.belopolsky | 2010-07-25 11:02:55 -0400 (Sun, 25 Jul 2010) | 5 lines
  
  Issue #9315: Renamed test_trace to test_sys_settrace and
  test_profilehooks to test_sys_setprofile so that test_trace can be
  used for testing the trace module and for naming consistency.
........
  r83141 | alexander.belopolsky | 2010-07-25 11:05:42 -0400 (Sun, 25 Jul 2010) | 1 line
  
  Corrected comments on where settrace and setprofile are tested.
........


Added:
   python/branches/release31-maint/Lib/test/test_sys_setprofile.py
      - copied, changed from r83147, /python/branches/release31-maint/Lib/test/test_profilehooks.py
   python/branches/release31-maint/Lib/test/test_sys_settrace.py
      - copied unchanged from r83147, /python/branches/release31-maint/Lib/test/test_trace.py
Removed:
   python/branches/release31-maint/Lib/test/test_profilehooks.py
   python/branches/release31-maint/Lib/test/test_trace.py
Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/test/test_sys.py

Deleted: python/branches/release31-maint/Lib/test/test_profilehooks.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_profilehooks.py	Mon Jul 26 00:15:16 2010
+++ (empty file)
@@ -1,385 +0,0 @@
-import pprint
-import sys
-import unittest
-
-from test import support
-
-class TestGetProfile(unittest.TestCase):
-    def setUp(self):
-        sys.setprofile(None)
-
-    def tearDown(self):
-        sys.setprofile(None)
-
-    def test_empty(self):
-        assert sys.getprofile() == None
-
-    def test_setget(self):
-        def fn(*args):
-            pass
-
-        sys.setprofile(fn)
-        assert sys.getprofile() == fn
-
-class HookWatcher:
-    def __init__(self):
-        self.frames = []
-        self.events = []
-
-    def callback(self, frame, event, arg):
-        if (event == "call"
-            or event == "return"
-            or event == "exception"):
-            self.add_event(event, frame)
-
-    def add_event(self, event, frame=None):
-        """Add an event to the log."""
-        if frame is None:
-            frame = sys._getframe(1)
-
-        try:
-            frameno = self.frames.index(frame)
-        except ValueError:
-            frameno = len(self.frames)
-            self.frames.append(frame)
-
-        self.events.append((frameno, event, ident(frame)))
-
-    def get_events(self):
-        """Remove calls to add_event()."""
-        disallowed = [ident(self.add_event.__func__), ident(ident)]
-        self.frames = None
-
-        return [item for item in self.events if item[2] not in disallowed]
-
-
-class ProfileSimulator(HookWatcher):
-    def __init__(self, testcase):
-        self.testcase = testcase
-        self.stack = []
-        HookWatcher.__init__(self)
-
-    def callback(self, frame, event, arg):
-        # Callback registered with sys.setprofile()/sys.settrace()
-        self.dispatch[event](self, frame)
-
-    def trace_call(self, frame):
-        self.add_event('call', frame)
-        self.stack.append(frame)
-
-    def trace_return(self, frame):
-        self.add_event('return', frame)
-        self.stack.pop()
-
-    def trace_exception(self, frame):
-        self.testcase.fail(
-            "the profiler should never receive exception events")
-
-    def trace_pass(self, frame):
-        pass
-
-    dispatch = {
-        'call': trace_call,
-        'exception': trace_exception,
-        'return': trace_return,
-        'c_call': trace_pass,
-        'c_return': trace_pass,
-        'c_exception': trace_pass,
-        }
-
-
-class TestCaseBase(unittest.TestCase):
-    def check_events(self, callable, expected):
-        events = capture_events(callable, self.new_watcher())
-        if events != expected:
-            self.fail("Expected events:\n%s\nReceived events:\n%s"
-                      % (pprint.pformat(expected), pprint.pformat(events)))
-
-
-class ProfileHookTestCase(TestCaseBase):
-    def new_watcher(self):
-        return HookWatcher()
-
-    def test_simple(self):
-        def f(p):
-            pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_exception(self):
-        def f(p):
-            1/0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_caught_exception(self):
-        def f(p):
-            try: 1/0
-            except: pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_caught_nested_exception(self):
-        def f(p):
-            try: 1/0
-            except: pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_nested_exception(self):
-        def f(p):
-            1/0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              # This isn't what I expected:
-                              # (0, 'exception', protect_ident),
-                              # I expected this again:
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_exception_in_except_clause(self):
-        def f(p):
-            1/0
-        def g(p):
-            try:
-                f(p)
-            except:
-                try: f(p)
-                except: pass
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (3, 'call', f_ident),
-                              (3, 'return', f_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-    def test_exception_propogation(self):
-        def f(p):
-            1/0
-        def g(p):
-            try: f(p)
-            finally: p.add_event("falling through")
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (1, 'falling through', g_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-    def test_raise_twice(self):
-        def f(p):
-            try: 1/0
-            except: 1/0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_raise_reraise(self):
-        def f(p):
-            try: 1/0
-            except: raise
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_raise(self):
-        def f(p):
-            raise Exception()
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_distant_exception(self):
-        def f():
-            1/0
-        def g():
-            f()
-        def h():
-            g()
-        def i():
-            h()
-        def j(p):
-            i()
-        f_ident = ident(f)
-        g_ident = ident(g)
-        h_ident = ident(h)
-        i_ident = ident(i)
-        j_ident = ident(j)
-        self.check_events(j, [(1, 'call', j_ident),
-                              (2, 'call', i_ident),
-                              (3, 'call', h_ident),
-                              (4, 'call', g_ident),
-                              (5, 'call', f_ident),
-                              (5, 'return', f_ident),
-                              (4, 'return', g_ident),
-                              (3, 'return', h_ident),
-                              (2, 'return', i_ident),
-                              (1, 'return', j_ident),
-                              ])
-
-    def test_generator(self):
-        def f():
-            for i in range(2):
-                yield i
-        def g(p):
-            for i in f():
-                pass
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              # call the iterator twice to generate values
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              # once more; returns end-of-iteration with
-                              # actually raising an exception
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-    def test_stop_iteration(self):
-        def f():
-            for i in range(2):
-                yield i
-            raise StopIteration
-        def g(p):
-            for i in f():
-                pass
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              # call the iterator twice to generate values
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              # once more to hit the raise:
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-
-class ProfileSimulatorTestCase(TestCaseBase):
-    def new_watcher(self):
-        return ProfileSimulator(self)
-
-    def test_simple(self):
-        def f(p):
-            pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_basic_exception(self):
-        def f(p):
-            1/0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_caught_exception(self):
-        def f(p):
-            try: 1/0
-            except: pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_distant_exception(self):
-        def f():
-            1/0
-        def g():
-            f()
-        def h():
-            g()
-        def i():
-            h()
-        def j(p):
-            i()
-        f_ident = ident(f)
-        g_ident = ident(g)
-        h_ident = ident(h)
-        i_ident = ident(i)
-        j_ident = ident(j)
-        self.check_events(j, [(1, 'call', j_ident),
-                              (2, 'call', i_ident),
-                              (3, 'call', h_ident),
-                              (4, 'call', g_ident),
-                              (5, 'call', f_ident),
-                              (5, 'return', f_ident),
-                              (4, 'return', g_ident),
-                              (3, 'return', h_ident),
-                              (2, 'return', i_ident),
-                              (1, 'return', j_ident),
-                              ])
-
-
-def ident(function):
-    if hasattr(function, "f_code"):
-        code = function.f_code
-    else:
-        code = function.__code__
-    return code.co_firstlineno, code.co_name
-
-
-def protect(f, p):
-    try: f(p)
-    except: pass
-
-protect_ident = ident(protect)
-
-
-def capture_events(callable, p=None):
-    try:
-        sys.setprofile()
-    except TypeError:
-        pass
-    else:
-        raise support.TestFailed(
-            'sys.setprofile() did not raise TypeError')
-
-    if p is None:
-        p = HookWatcher()
-    sys.setprofile(p.callback)
-    protect(callable, p)
-    sys.setprofile(None)
-    return p.get_events()[1:-1]
-
-
-def show_events(callable):
-    import pprint
-    pprint.pprint(capture_events(callable))
-
-
-def test_main():
-    support.run_unittest(
-        TestGetProfile,
-        ProfileHookTestCase,
-        ProfileSimulatorTestCase
-    )
-
-
-if __name__ == "__main__":
-    test_main()

Modified: python/branches/release31-maint/Lib/test/test_sys.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_sys.py	(original)
+++ python/branches/release31-maint/Lib/test/test_sys.py	Mon Jul 26 00:15:16 2010
@@ -164,8 +164,8 @@
         # can't check more than the type, as the user might have changed it
         self.assertTrue(isinstance(sys.getdefaultencoding(), str))
 
-    # testing sys.settrace() is done in test_trace.py
-    # testing sys.setprofile() is done in test_profile.py
+    # testing sys.settrace() is done in test_sys_settrace.py
+    # testing sys.setprofile() is done in test_sys_setprofile.py
 
     def test_setcheckinterval(self):
         self.assertRaises(TypeError, sys.setcheckinterval)

Copied: python/branches/release31-maint/Lib/test/test_sys_setprofile.py (from r83147, /python/branches/release31-maint/Lib/test/test_profilehooks.py)
==============================================================================
--- /python/branches/release31-maint/Lib/test/test_profilehooks.py	(original)
+++ python/branches/release31-maint/Lib/test/test_sys_setprofile.py	Mon Jul 26 00:15:16 2010
@@ -12,7 +12,7 @@
         sys.setprofile(None)
 
     def test_empty(self):
-        assert sys.getprofile() == None
+        assert sys.getprofile() is None
 
     def test_setget(self):
         def fn(*args):

Deleted: python/branches/release31-maint/Lib/test/test_trace.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_trace.py	Mon Jul 26 00:15:16 2010
+++ (empty file)
@@ -1,790 +0,0 @@
-# Testing the line trace facility.
-
-from test import support
-import unittest
-import sys
-import difflib
-import gc
-
-# A very basic example.  If this fails, we're in deep trouble.
-def basic():
-    return 1
-
-basic.events = [(0, 'call'),
-                (1, 'line'),
-                (1, 'return')]
-
-# Many of the tests below are tricky because they involve pass statements.
-# If there is implicit control flow around a pass statement (in an except
-# clause or else caluse) under what conditions do you set a line number
-# following that clause?
-
-
-# The entire "while 0:" statement is optimized away.  No code
-# exists for it, so the line numbers skip directly from "del x"
-# to "x = 1".
-def arigo_example():
-    x = 1
-    del x
-    while 0:
-        pass
-    x = 1
-
-arigo_example.events = [(0, 'call'),
-                        (1, 'line'),
-                        (2, 'line'),
-                        (5, 'line'),
-                        (5, 'return')]
-
-# check that lines consisting of just one instruction get traced:
-def one_instr_line():
-    x = 1
-    del x
-    x = 1
-
-one_instr_line.events = [(0, 'call'),
-                         (1, 'line'),
-                         (2, 'line'),
-                         (3, 'line'),
-                         (3, 'return')]
-
-def no_pop_tops():      # 0
-    x = 1               # 1
-    for a in range(2):  # 2
-        if a:           # 3
-            x = 1       # 4
-        else:           # 5
-            x = 1       # 6
-
-no_pop_tops.events = [(0, 'call'),
-                      (1, 'line'),
-                      (2, 'line'),
-                      (3, 'line'),
-                      (6, 'line'),
-                      (2, 'line'),
-                      (3, 'line'),
-                      (4, 'line'),
-                      (2, 'line'),
-                      (2, 'return')]
-
-def no_pop_blocks():
-    y = 1
-    while not y:
-        bla
-    x = 1
-
-no_pop_blocks.events = [(0, 'call'),
-                        (1, 'line'),
-                        (2, 'line'),
-                        (4, 'line'),
-                        (4, 'return')]
-
-def called(): # line -3
-    x = 1
-
-def call():   # line 0
-    called()
-
-call.events = [(0, 'call'),
-               (1, 'line'),
-               (-3, 'call'),
-               (-2, 'line'),
-               (-2, 'return'),
-               (1, 'return')]
-
-def raises():
-    raise Exception
-
-def test_raise():
-    try:
-        raises()
-    except Exception as exc:
-        x = 1
-
-test_raise.events = [(0, 'call'),
-                     (1, 'line'),
-                     (2, 'line'),
-                     (-3, 'call'),
-                     (-2, 'line'),
-                     (-2, 'exception'),
-                     (-2, 'return'),
-                     (2, 'exception'),
-                     (3, 'line'),
-                     (4, 'line'),
-                     (4, 'return')]
-
-def _settrace_and_return(tracefunc):
-    sys.settrace(tracefunc)
-    sys._getframe().f_back.f_trace = tracefunc
-def settrace_and_return(tracefunc):
-    _settrace_and_return(tracefunc)
-
-settrace_and_return.events = [(1, 'return')]
-
-def _settrace_and_raise(tracefunc):
-    sys.settrace(tracefunc)
-    sys._getframe().f_back.f_trace = tracefunc
-    raise RuntimeError
-def settrace_and_raise(tracefunc):
-    try:
-        _settrace_and_raise(tracefunc)
-    except RuntimeError as exc:
-        pass
-
-settrace_and_raise.events = [(2, 'exception'),
-                             (3, 'line'),
-                             (4, 'line'),
-                             (4, 'return')]
-
-# implicit return example
-# This test is interesting because of the else: pass
-# part of the code.  The code generate for the true
-# part of the if contains a jump past the else branch.
-# The compiler then generates an implicit "return None"
-# Internally, the compiler visits the pass statement
-# and stores its line number for use on the next instruction.
-# The next instruction is the implicit return None.
-def ireturn_example():
-    a = 5
-    b = 5
-    if a == b:
-        b = a+1
-    else:
-        pass
-
-ireturn_example.events = [(0, 'call'),
-                          (1, 'line'),
-                          (2, 'line'),
-                          (3, 'line'),
-                          (4, 'line'),
-                          (6, 'line'),
-                          (6, 'return')]
-
-# Tight loop with while(1) example (SF #765624)
-def tightloop_example():
-    items = range(0, 3)
-    try:
-        i = 0
-        while 1:
-            b = items[i]; i+=1
-    except IndexError:
-        pass
-
-tightloop_example.events = [(0, 'call'),
-                            (1, 'line'),
-                            (2, 'line'),
-                            (3, 'line'),
-                            (4, 'line'),
-                            (5, 'line'),
-                            (5, 'line'),
-                            (5, 'line'),
-                            (5, 'line'),
-                            (5, 'exception'),
-                            (6, 'line'),
-                            (7, 'line'),
-                            (7, 'return')]
-
-def tighterloop_example():
-    items = range(1, 4)
-    try:
-        i = 0
-        while 1: i = items[i]
-    except IndexError:
-        pass
-
-tighterloop_example.events = [(0, 'call'),
-                            (1, 'line'),
-                            (2, 'line'),
-                            (3, 'line'),
-                            (4, 'line'),
-                            (4, 'line'),
-                            (4, 'line'),
-                            (4, 'line'),
-                            (4, 'exception'),
-                            (5, 'line'),
-                            (6, 'line'),
-                            (6, 'return')]
-
-def generator_function():
-    try:
-        yield True
-        "continued"
-    finally:
-        "finally"
-def generator_example():
-    # any() will leave the generator before its end
-    x = any(generator_function())
-
-    # the following lines were not traced
-    for x in range(10):
-        y = x
-
-generator_example.events = ([(0, 'call'),
-                             (2, 'line'),
-                             (-6, 'call'),
-                             (-5, 'line'),
-                             (-4, 'line'),
-                             (-4, 'return'),
-                             (-4, 'call'),
-                             (-4, 'exception'),
-                             (-1, 'line'),
-                             (-1, 'return')] +
-                            [(5, 'line'), (6, 'line')] * 10 +
-                            [(5, 'line'), (5, 'return')])
-
-
-class Tracer:
-    def __init__(self):
-        self.events = []
-    def trace(self, frame, event, arg):
-        self.events.append((frame.f_lineno, event))
-        return self.trace
-    def traceWithGenexp(self, frame, event, arg):
-        (o for o in [1])
-        self.events.append((frame.f_lineno, event))
-        return self.trace
-
-class TraceTestCase(unittest.TestCase):
-
-    # Disable gc collection when tracing, otherwise the
-    # deallocators may be traced as well.
-    def setUp(self):
-        self.using_gc = gc.isenabled()
-        gc.disable()
-
-    def tearDown(self):
-        if self.using_gc:
-            gc.enable()
-
-    def compare_events(self, line_offset, events, expected_events):
-        events = [(l - line_offset, e) for (l, e) in events]
-        if events != expected_events:
-            self.fail(
-                "events did not match expectation:\n" +
-                "\n".join(difflib.ndiff([str(x) for x in expected_events],
-                                        [str(x) for x in events])))
-
-    def run_and_compare(self, func, events):
-        tracer = Tracer()
-        sys.settrace(tracer.trace)
-        func()
-        sys.settrace(None)
-        self.compare_events(func.__code__.co_firstlineno,
-                            tracer.events, events)
-
-    def run_test(self, func):
-        self.run_and_compare(func, func.events)
-
-    def run_test2(self, func):
-        tracer = Tracer()
-        func(tracer.trace)
-        sys.settrace(None)
-        self.compare_events(func.__code__.co_firstlineno,
-                            tracer.events, func.events)
-
-    def set_and_retrieve_none(self):
-        sys.settrace(None)
-        assert sys.gettrace() is None
-
-    def set_and_retrieve_func(self):
-        def fn(*args):
-            pass
-
-        sys.settrace(fn)
-        try:
-            assert sys.gettrace() is fn
-        finally:
-            sys.settrace(None)
-
-    def test_01_basic(self):
-        self.run_test(basic)
-    def test_02_arigo(self):
-        self.run_test(arigo_example)
-    def test_03_one_instr(self):
-        self.run_test(one_instr_line)
-    def test_04_no_pop_blocks(self):
-        self.run_test(no_pop_blocks)
-    def test_05_no_pop_tops(self):
-        self.run_test(no_pop_tops)
-    def test_06_call(self):
-        self.run_test(call)
-    def test_07_raise(self):
-        self.run_test(test_raise)
-
-    def test_08_settrace_and_return(self):
-        self.run_test2(settrace_and_return)
-    def test_09_settrace_and_raise(self):
-        self.run_test2(settrace_and_raise)
-    def test_10_ireturn(self):
-        self.run_test(ireturn_example)
-    def test_11_tightloop(self):
-        self.run_test(tightloop_example)
-    def test_12_tighterloop(self):
-        self.run_test(tighterloop_example)
-
-    def test_13_genexp(self):
-        self.run_test(generator_example)
-        # issue1265: if the trace function contains a generator,
-        # and if the traced function contains another generator
-        # that is not completely exhausted, the trace stopped.
-        # Worse: the 'finally' clause was not invoked.
-        tracer = Tracer()
-        sys.settrace(tracer.traceWithGenexp)
-        generator_example()
-        sys.settrace(None)
-        self.compare_events(generator_example.__code__.co_firstlineno,
-                            tracer.events, generator_example.events)
-
-    def test_14_onliner_if(self):
-        def onliners():
-            if True: False
-            else: True
-            return 0
-        self.run_and_compare(
-            onliners,
-            [(0, 'call'),
-             (1, 'line'),
-             (3, 'line'),
-             (3, 'return')])
-
-    def test_15_loops(self):
-        # issue1750076: "while" expression is skipped by debugger
-        def for_example():
-            for x in range(2):
-                pass
-        self.run_and_compare(
-            for_example,
-            [(0, 'call'),
-             (1, 'line'),
-             (2, 'line'),
-             (1, 'line'),
-             (2, 'line'),
-             (1, 'line'),
-             (1, 'return')])
-
-        def while_example():
-            # While expression should be traced on every loop
-            x = 2
-            while x > 0:
-                x -= 1
-        self.run_and_compare(
-            while_example,
-            [(0, 'call'),
-             (2, 'line'),
-             (3, 'line'),
-             (4, 'line'),
-             (3, 'line'),
-             (4, 'line'),
-             (3, 'line'),
-             (3, 'return')])
-
-    def test_16_blank_lines(self):
-        namespace = {}
-        exec("def f():\n" + "\n" * 256 + "    pass", namespace)
-        self.run_and_compare(
-            namespace["f"],
-            [(0, 'call'),
-             (257, 'line'),
-             (257, 'return')])
-
-
-class RaisingTraceFuncTestCase(unittest.TestCase):
-    def trace(self, frame, event, arg):
-        """A trace function that raises an exception in response to a
-        specific trace event."""
-        if event == self.raiseOnEvent:
-            raise ValueError # just something that isn't RuntimeError
-        else:
-            return self.trace
-
-    def f(self):
-        """The function to trace; raises an exception if that's the case
-        we're testing, so that the 'exception' trace event fires."""
-        if self.raiseOnEvent == 'exception':
-            x = 0
-            y = 1/x
-        else:
-            return 1
-
-    def run_test_for_event(self, event):
-        """Tests that an exception raised in response to the given event is
-        handled OK."""
-        self.raiseOnEvent = event
-        try:
-            for i in range(sys.getrecursionlimit() + 1):
-                sys.settrace(self.trace)
-                try:
-                    self.f()
-                except ValueError:
-                    pass
-                else:
-                    self.fail("exception not thrown!")
-        except RuntimeError:
-            self.fail("recursion counter not reset")
-
-    # Test the handling of exceptions raised by each kind of trace event.
-    def test_call(self):
-        self.run_test_for_event('call')
-    def test_line(self):
-        self.run_test_for_event('line')
-    def test_return(self):
-        self.run_test_for_event('return')
-    def test_exception(self):
-        self.run_test_for_event('exception')
-
-    def test_trash_stack(self):
-        def f():
-            for i in range(5):
-                print(i)  # line tracing will raise an exception at this line
-
-        def g(frame, why, extra):
-            if (why == 'line' and
-                frame.f_lineno == f.__code__.co_firstlineno + 2):
-                raise RuntimeError("i am crashing")
-            return g
-
-        sys.settrace(g)
-        try:
-            f()
-        except RuntimeError:
-            # the test is really that this doesn't segfault:
-            import gc
-            gc.collect()
-        else:
-            self.fail("exception not propagated")
-
-
-# 'Jump' tests: assigning to frame.f_lineno within a trace function
-# moves the execution position - it's how debuggers implement a Jump
-# command (aka. "Set next statement").
-
-class JumpTracer:
-    """Defines a trace function that jumps from one place to another,
-    with the source and destination lines of the jump being defined by
-    the 'jump' property of the function under test."""
-
-    def __init__(self, function):
-        self.function = function
-        self.jumpFrom = function.jump[0]
-        self.jumpTo = function.jump[1]
-        self.done = False
-
-    def trace(self, frame, event, arg):
-        if not self.done and frame.f_code == self.function.__code__:
-            firstLine = frame.f_code.co_firstlineno
-            if event == 'line' and frame.f_lineno == firstLine + self.jumpFrom:
-                # Cope with non-integer self.jumpTo (because of
-                # no_jump_to_non_integers below).
-                try:
-                    frame.f_lineno = firstLine + self.jumpTo
-                except TypeError:
-                    frame.f_lineno = self.jumpTo
-                self.done = True
-        return self.trace
-
-# The first set of 'jump' tests are for things that are allowed:
-
-def jump_simple_forwards(output):
-    output.append(1)
-    output.append(2)
-    output.append(3)
-
-jump_simple_forwards.jump = (1, 3)
-jump_simple_forwards.output = [3]
-
-def jump_simple_backwards(output):
-    output.append(1)
-    output.append(2)
-
-jump_simple_backwards.jump = (2, 1)
-jump_simple_backwards.output = [1, 1, 2]
-
-def jump_out_of_block_forwards(output):
-    for i in 1, 2:
-        output.append(2)
-        for j in [3]:  # Also tests jumping over a block
-            output.append(4)
-    output.append(5)
-
-jump_out_of_block_forwards.jump = (3, 5)
-jump_out_of_block_forwards.output = [2, 5]
-
-def jump_out_of_block_backwards(output):
-    output.append(1)
-    for i in [1]:
-        output.append(3)
-        for j in [2]:  # Also tests jumping over a block
-            output.append(5)
-        output.append(6)
-    output.append(7)
-
-jump_out_of_block_backwards.jump = (6, 1)
-jump_out_of_block_backwards.output = [1, 3, 5, 1, 3, 5, 6, 7]
-
-def jump_to_codeless_line(output):
-    output.append(1)
-    # Jumping to this line should skip to the next one.
-    output.append(3)
-
-jump_to_codeless_line.jump = (1, 2)
-jump_to_codeless_line.output = [3]
-
-def jump_to_same_line(output):
-    output.append(1)
-    output.append(2)
-    output.append(3)
-
-jump_to_same_line.jump = (2, 2)
-jump_to_same_line.output = [1, 2, 3]
-
-# Tests jumping within a finally block, and over one.
-def jump_in_nested_finally(output):
-    try:
-        output.append(2)
-    finally:
-        output.append(4)
-        try:
-            output.append(6)
-        finally:
-            output.append(8)
-        output.append(9)
-
-jump_in_nested_finally.jump = (4, 9)
-jump_in_nested_finally.output = [2, 9]
-
-# The second set of 'jump' tests are for things that are not allowed:
-
-def no_jump_too_far_forwards(output):
-    try:
-        output.append(2)
-        output.append(3)
-    except ValueError as e:
-        output.append('after' in str(e))
-
-no_jump_too_far_forwards.jump = (3, 6)
-no_jump_too_far_forwards.output = [2, True]
-
-def no_jump_too_far_backwards(output):
-    try:
-        output.append(2)
-        output.append(3)
-    except ValueError as e:
-        output.append('before' in str(e))
-
-no_jump_too_far_backwards.jump = (3, -1)
-no_jump_too_far_backwards.output = [2, True]
-
-# Test each kind of 'except' line.
-def no_jump_to_except_1(output):
-    try:
-        output.append(2)
-    except:
-        e = sys.exc_info()[1]
-        output.append('except' in str(e))
-
-no_jump_to_except_1.jump = (2, 3)
-no_jump_to_except_1.output = [True]
-
-def no_jump_to_except_2(output):
-    try:
-        output.append(2)
-    except ValueError:
-        e = sys.exc_info()[1]
-        output.append('except' in str(e))
-
-no_jump_to_except_2.jump = (2, 3)
-no_jump_to_except_2.output = [True]
-
-def no_jump_to_except_3(output):
-    try:
-        output.append(2)
-    except ValueError as e:
-        output.append('except' in str(e))
-
-no_jump_to_except_3.jump = (2, 3)
-no_jump_to_except_3.output = [True]
-
-def no_jump_to_except_4(output):
-    try:
-        output.append(2)
-    except (ValueError, RuntimeError) as e:
-        output.append('except' in str(e))
-
-no_jump_to_except_4.jump = (2, 3)
-no_jump_to_except_4.output = [True]
-
-def no_jump_forwards_into_block(output):
-    try:
-        output.append(2)
-        for i in 1, 2:
-            output.append(4)
-    except ValueError as e:
-        output.append('into' in str(e))
-
-no_jump_forwards_into_block.jump = (2, 4)
-no_jump_forwards_into_block.output = [True]
-
-def no_jump_backwards_into_block(output):
-    try:
-        for i in 1, 2:
-            output.append(3)
-        output.append(4)
-    except ValueError as e:
-        output.append('into' in str(e))
-
-no_jump_backwards_into_block.jump = (4, 3)
-no_jump_backwards_into_block.output = [3, 3, True]
-
-def no_jump_into_finally_block(output):
-    try:
-        try:
-            output.append(3)
-            x = 1
-        finally:
-            output.append(6)
-    except ValueError as e:
-        output.append('finally' in str(e))
-
-no_jump_into_finally_block.jump = (4, 6)
-no_jump_into_finally_block.output = [3, 6, True]  # The 'finally' still runs
-
-def no_jump_out_of_finally_block(output):
-    try:
-        try:
-            output.append(3)
-        finally:
-            output.append(5)
-            output.append(6)
-    except ValueError as e:
-        output.append('finally' in str(e))
-
-no_jump_out_of_finally_block.jump = (5, 1)
-no_jump_out_of_finally_block.output = [3, True]
-
-# This verifies the line-numbers-must-be-integers rule.
-def no_jump_to_non_integers(output):
-    try:
-        output.append(2)
-    except ValueError as e:
-        output.append('integer' in str(e))
-
-no_jump_to_non_integers.jump = (2, "Spam")
-no_jump_to_non_integers.output = [True]
-
-# This verifies that you can't set f_lineno via _getframe or similar
-# trickery.
-def no_jump_without_trace_function():
-    try:
-        previous_frame = sys._getframe().f_back
-        previous_frame.f_lineno = previous_frame.f_lineno
-    except ValueError as e:
-        # This is the exception we wanted; make sure the error message
-        # talks about trace functions.
-        if 'trace' not in str(e):
-            raise
-    else:
-        # Something's wrong - the expected exception wasn't raised.
-        raise RuntimeError("Trace-function-less jump failed to fail")
-
-
-class JumpTestCase(unittest.TestCase):
-    def compare_jump_output(self, expected, received):
-        if received != expected:
-            self.fail( "Outputs don't match:\n" +
-                       "Expected: " + repr(expected) + "\n" +
-                       "Received: " + repr(received))
-
-    def run_test(self, func):
-        tracer = JumpTracer(func)
-        sys.settrace(tracer.trace)
-        output = []
-        func(output)
-        sys.settrace(None)
-        self.compare_jump_output(func.output, output)
-
-    def test_01_jump_simple_forwards(self):
-        self.run_test(jump_simple_forwards)
-    def test_02_jump_simple_backwards(self):
-        self.run_test(jump_simple_backwards)
-    def test_03_jump_out_of_block_forwards(self):
-        self.run_test(jump_out_of_block_forwards)
-    def test_04_jump_out_of_block_backwards(self):
-        self.run_test(jump_out_of_block_backwards)
-    def test_05_jump_to_codeless_line(self):
-        self.run_test(jump_to_codeless_line)
-    def test_06_jump_to_same_line(self):
-        self.run_test(jump_to_same_line)
-    def test_07_jump_in_nested_finally(self):
-        self.run_test(jump_in_nested_finally)
-    def test_08_no_jump_too_far_forwards(self):
-        self.run_test(no_jump_too_far_forwards)
-    def test_09_no_jump_too_far_backwards(self):
-        self.run_test(no_jump_too_far_backwards)
-    def test_10_no_jump_to_except_1(self):
-        self.run_test(no_jump_to_except_1)
-    def test_11_no_jump_to_except_2(self):
-        self.run_test(no_jump_to_except_2)
-    def test_12_no_jump_to_except_3(self):
-        self.run_test(no_jump_to_except_3)
-    def test_13_no_jump_to_except_4(self):
-        self.run_test(no_jump_to_except_4)
-    def test_14_no_jump_forwards_into_block(self):
-        self.run_test(no_jump_forwards_into_block)
-    def test_15_no_jump_backwards_into_block(self):
-        self.run_test(no_jump_backwards_into_block)
-    def test_16_no_jump_into_finally_block(self):
-        self.run_test(no_jump_into_finally_block)
-    def test_17_no_jump_out_of_finally_block(self):
-        self.run_test(no_jump_out_of_finally_block)
-    def test_18_no_jump_to_non_integers(self):
-        self.run_test(no_jump_to_non_integers)
-    def test_19_no_jump_without_trace_function(self):
-        no_jump_without_trace_function()
-
-    def test_20_large_function(self):
-        d = {}
-        exec("""def f(output):        # line 0
-            x = 0                     # line 1
-            y = 1                     # line 2
-            '''                       # line 3
-            %s                        # lines 4-1004
-            '''                       # line 1005
-            x += 1                    # line 1006
-            output.append(x)          # line 1007
-            return""" % ('\n' * 1000,), d)
-        f = d['f']
-
-        f.jump = (2, 1007)
-        f.output = [0]
-        self.run_test(f)
-
-    def test_jump_to_firstlineno(self):
-        # This tests that PDB can jump back to the first line in a
-        # file.  See issue #1689458.  It can only be triggered in a
-        # function call if the function is defined on a single line.
-        code = compile("""
-# Comments don't count.
-output.append(2)  # firstlineno is here.
-output.append(3)
-output.append(4)
-""", "<fake module>", "exec")
-        class fake_function:
-            __code__ = code
-            jump = (2, 0)
-        tracer = JumpTracer(fake_function)
-        sys.settrace(tracer.trace)
-        namespace = {"output": []}
-        exec(code, namespace)
-        sys.settrace(None)
-        self.compare_jump_output([2, 3, 2, 3, 4], namespace["output"])
-
-
-def test_main():
-    support.run_unittest(
-        TraceTestCase,
-        RaisingTraceFuncTestCase,
-        JumpTestCase
-    )
-
-if __name__ == "__main__":
-    test_main()

From python-checkins at python.org  Mon Jul 26 00:57:12 2010
From: python-checkins at python.org (andrew.kuchling)
Date: Mon, 26 Jul 2010 00:57:12 +0200 (CEST)
Subject: [Python-checkins] r83149 -
	python/branches/py3k/Doc/library/xml.dom.rst
Message-ID: <20100725225712.2A210EE9D8@mail.python.org>

Author: andrew.kuchling
Date: Mon Jul 26 00:57:12 2010
New Revision: 83149

Log:
#1495229: update the type names used by the XML DOM mapping

Modified:
   python/branches/py3k/Doc/library/xml.dom.rst

Modified: python/branches/py3k/Doc/library/xml.dom.rst
==============================================================================
--- python/branches/py3k/Doc/library/xml.dom.rst	(original)
+++ python/branches/py3k/Doc/library/xml.dom.rst	Mon Jul 26 00:57:12 2010
@@ -976,29 +976,24 @@
 Type Mapping
 ^^^^^^^^^^^^
 
-The primitive IDL types used in the DOM specification are mapped to Python types
+The IDL types used in the DOM specification are mapped to Python types
 according to the following table.
 
 +------------------+-------------------------------------------+
 | IDL Type         | Python Type                               |
 +==================+===========================================+
-| ``boolean``      | ``IntegerType`` (with a value of ``0`` or |
-|                  | ``1``)                                    |
+| ``boolean``      | ``bool`` or ``int``                       |
 +------------------+-------------------------------------------+
-| ``int``          | ``IntegerType``                           |
+| ``int``          | ``int``                                   |
 +------------------+-------------------------------------------+
-| ``long int``     | ``IntegerType``                           |
+| ``long int``     | ``int``                                   |
 +------------------+-------------------------------------------+
-| ``unsigned int`` | ``IntegerType``                           |
+| ``unsigned int`` | ``int``                                   |
++------------------+-------------------------------------------+
+| ``DOMString``    | ``str`` or ``bytes``                      |
++------------------+-------------------------------------------+
+| ``null``         | ``None``                                  |
 +------------------+-------------------------------------------+
-
-Additionally, the :class:`DOMString` defined in the recommendation is mapped to
-a bytes or string object.  Applications should be able to handle
-Unicode whenever a string is returned from the DOM.
-
-The IDL ``null`` value is mapped to ``None``, which may be accepted or
-provided by the implementation whenever ``null`` is allowed by the API.
-
 
 .. _dom-accessor-methods:
 

From python-checkins at python.org  Mon Jul 26 01:09:25 2010
From: python-checkins at python.org (michael.foord)
Date: Mon, 26 Jul 2010 01:09:25 +0200 (CEST)
Subject: [Python-checkins] r83150 - in python/branches/py3k:
	Lib/configparser.py Lib/test/test_cfgparser.py Misc/python-wing.wpr
Message-ID: <20100725230925.EBCB0EE98C@mail.python.org>

Author: michael.foord
Date: Mon Jul 26 01:09:25 2010
New Revision: 83150

Log:
Issue #4686 - add .args to exceptions in the configparsermodule

Modified:
   python/branches/py3k/Lib/configparser.py
   python/branches/py3k/Lib/test/test_cfgparser.py
   python/branches/py3k/Misc/python-wing.wpr

Modified: python/branches/py3k/Lib/configparser.py
==============================================================================
--- python/branches/py3k/Lib/configparser.py	(original)
+++ python/branches/py3k/Lib/configparser.py	Mon Jul 26 01:09:25 2010
@@ -142,6 +142,7 @@
     def __init__(self, section):
         Error.__init__(self, 'No section: %r' % (section,))
         self.section = section
+        self.args = (section, )
 
 class DuplicateSectionError(Error):
     """Raised when a section is multiply-created."""
@@ -149,6 +150,7 @@
     def __init__(self, section):
         Error.__init__(self, "Section %r already exists" % section)
         self.section = section
+        self.args = (section, )
 
 class NoOptionError(Error):
     """A requested option was not found."""
@@ -158,6 +160,7 @@
                        (option, section))
         self.option = option
         self.section = section
+        self.args = (option, section)
 
 class InterpolationError(Error):
     """Base class for interpolation-related exceptions."""
@@ -166,6 +169,7 @@
         Error.__init__(self, msg)
         self.option = option
         self.section = section
+        self.args = (option, section, msg)
 
 class InterpolationMissingOptionError(InterpolationError):
     """A string substitution required a setting which was not available."""
@@ -179,6 +183,7 @@
                % (section, option, reference, rawval))
         InterpolationError.__init__(self, option, section, msg)
         self.reference = reference
+        self.args = (option, section, rawval, reference)
 
 class InterpolationSyntaxError(InterpolationError):
     """Raised when the source text into which substitutions are made
@@ -194,6 +199,7 @@
                "\trawval : %s\n"
                % (section, option, rawval))
         InterpolationError.__init__(self, option, section, msg)
+        self.args = (option, section, rawval)
 
 class ParsingError(Error):
     """Raised when a configuration file does not follow legal syntax."""
@@ -202,6 +208,7 @@
         Error.__init__(self, 'File contains parsing errors: %s' % filename)
         self.filename = filename
         self.errors = []
+        self.args = (filename, )
 
     def append(self, lineno, line):
         self.errors.append((lineno, line))
@@ -218,7 +225,7 @@
         self.filename = filename
         self.lineno = lineno
         self.line = line
-
+        self.args = (filename, lineno, line)
 
 class RawConfigParser:
     def __init__(self, defaults=None, dict_type=_default_dict,

Modified: python/branches/py3k/Lib/test/test_cfgparser.py
==============================================================================
--- python/branches/py3k/Lib/test/test_cfgparser.py	(original)
+++ python/branches/py3k/Lib/test/test_cfgparser.py	Mon Jul 26 01:09:25 2010
@@ -107,8 +107,9 @@
                     "remove_option() failed to report non-existence of option"
                     " that was removed")
 
-        self.assertRaises(configparser.NoSectionError,
-                          cf.remove_option, 'No Such Section', 'foo')
+        with self.assertRaises(configparser.NoSectionError) as cm:
+            cf.remove_option('No Such Section', 'foo')
+        self.assertEqual(cm.exception.args, ('No Such Section',))
 
         eq(cf.get('Long Line', 'foo'),
            'this line is much, much longer than my editor\nlikes it.')
@@ -160,20 +161,24 @@
 
     def test_parse_errors(self):
         self.newconfig()
-        self.parse_error(configparser.ParsingError,
+        e = self.parse_error(configparser.ParsingError,
                          "[Foo]\n  extra-spaces: splat\n")
+        self.assertEqual(e.args, ('<???>',))
         self.parse_error(configparser.ParsingError,
                          "[Foo]\n  extra-spaces= splat\n")
         self.parse_error(configparser.ParsingError,
                          "[Foo]\n:value-without-option-name\n")
         self.parse_error(configparser.ParsingError,
                          "[Foo]\n=value-without-option-name\n")
-        self.parse_error(configparser.MissingSectionHeaderError,
+        e = self.parse_error(configparser.MissingSectionHeaderError,
                          "No Section!\n")
+        self.assertEqual(e.args, ('<???>', 1, "No Section!\n"))
 
     def parse_error(self, exc, src):
         sio = io.StringIO(src)
-        self.assertRaises(exc, self.cf.readfp, sio)
+        with self.assertRaises(exc) as cm:
+            self.cf.readfp(sio)
+        return cm.exception
 
     def test_query_errors(self):
         cf = self.newconfig()
@@ -181,13 +186,15 @@
                          "new ConfigParser should have no defined sections")
         self.assertFalse(cf.has_section("Foo"),
                     "new ConfigParser should have no acknowledged sections")
-        self.assertRaises(configparser.NoSectionError,
-                          cf.options, "Foo")
-        self.assertRaises(configparser.NoSectionError,
-                          cf.set, "foo", "bar", "value")
-        self.get_error(configparser.NoSectionError, "foo", "bar")
+        with self.assertRaises(configparser.NoSectionError) as cm:
+            cf.options("Foo")
+        with self.assertRaises(configparser.NoSectionError) as cm:
+            cf.set("foo", "bar", "value")
+        e = self.get_error(configparser.NoSectionError, "foo", "bar")
+        self.assertEqual(e.args, ("foo",))
         cf.add_section("foo")
-        self.get_error(configparser.NoOptionError, "foo", "bar")
+        e = self.get_error(configparser.NoOptionError, "foo", "bar")
+        self.assertEqual(e.args, ("bar", "foo"))
 
     def get_error(self, exc, section, option):
         try:
@@ -226,8 +233,9 @@
     def test_weird_errors(self):
         cf = self.newconfig()
         cf.add_section("Foo")
-        self.assertRaises(configparser.DuplicateSectionError,
-                          cf.add_section, "Foo")
+        with self.assertRaises(configparser.DuplicateSectionError) as cm:
+            cf.add_section("Foo")
+        self.assertEqual(cm.exception.args, ("Foo",))
 
     def test_write(self):
         config_string = (
@@ -346,6 +354,11 @@
     config_class = configparser.ConfigParser
 
     def test_interpolation(self):
+        rawval = {
+            configparser.ConfigParser: "something %(with11)s "\
+                                           "lots of interpolation (11 steps)",
+            configparser.SafeConfigParser: "%(with1)s",
+        }
         cf = self.get_interpolation_config()
         eq = self.assertEqual
         eq(cf.get("Foo", "getname"), "Foo")
@@ -354,15 +367,22 @@
            "something with lots of interpolation (9 steps)")
         eq(cf.get("Foo", "bar10"),
            "something with lots of interpolation (10 steps)")
-        self.get_error(configparser.InterpolationDepthError, "Foo", "bar11")
+        e = self.get_error(configparser.InterpolationDepthError, "Foo", "bar11")
+        self.assertEqual(e.args, ("bar11", "Foo", rawval[self.config_class]))
 
     def test_interpolation_missing_value(self):
+        rawval = {
+            configparser.ConfigParser: '%(reference)s',
+            configparser.SafeConfigParser: '',
+        }
         self.get_interpolation_config()
-        e = self.get_error(configparser.InterpolationError,
+        e = self.get_error(configparser.InterpolationMissingOptionError,
                            "Interpolation Error", "name")
         self.assertEqual(e.reference, "reference")
         self.assertEqual(e.section, "Interpolation Error")
         self.assertEqual(e.option, "name")
+        self.assertEqual(e.args, ('name', 'Interpolation Error',
+                                  rawval[self.config_class], 'reference'))
 
     def test_items(self):
         self.check_items_config([('default', '<default>'),

Modified: python/branches/py3k/Misc/python-wing.wpr
==============================================================================
--- python/branches/py3k/Misc/python-wing.wpr	(original)
+++ python/branches/py3k/Misc/python-wing.wpr	Mon Jul 26 01:09:25 2010
@@ -5,7 +5,7 @@
 ##################################################################
 [project attributes]
 proj.directory-list = [{'dirloc': loc('..'),
-                        'excludes': (),
+                        'excludes': [u'Lib/__pycache__'],
                         'filter': '*',
                         'include_hidden': False,
                         'recursive': True,

From python-checkins at python.org  Mon Jul 26 01:23:31 2010
From: python-checkins at python.org (andrew.kuchling)
Date: Mon, 26 Jul 2010 01:23:31 +0200 (CEST)
Subject: [Python-checkins] r83151 -
	python/branches/py3k/Doc/library/xml.dom.minidom.rst
Message-ID: <20100725232331.15301EDD9@mail.python.org>

Author: andrew.kuchling
Date: Mon Jul 26 01:23:30 2010
New Revision: 83151

Log:
#7637: clarify description of encoding parameter

Modified:
   python/branches/py3k/Doc/library/xml.dom.minidom.rst

Modified: python/branches/py3k/Doc/library/xml.dom.minidom.rst
==============================================================================
--- python/branches/py3k/Doc/library/xml.dom.minidom.rst	(original)
+++ python/branches/py3k/Doc/library/xml.dom.minidom.rst	Mon Jul 26 01:23:30 2010
@@ -136,18 +136,20 @@
 
 .. method:: Node.toxml(encoding=None)
 
-   Return the XML that the DOM represents as a string.
-
-   With no argument, the XML header does not specify an encoding, and the result is
-   Unicode string if the default encoding cannot represent all characters in the
-   document. Encoding this string in an encoding other than UTF-8 is likely
-   incorrect, since UTF-8 is the default encoding of XML.
-
-   With an explicit *encoding* [1]_ argument, the result is a byte string in the
-   specified encoding. It is recommended that this argument is always specified. To
-   avoid :exc:`UnicodeError` exceptions in case of unrepresentable text data, the
-   encoding argument should be specified as "utf-8".
+   Return a string or byte string containing the XML represented by
+   the DOM node.
 
+   With an explicit *encoding* [1]_ argument, the result is a byte
+   string in the specified encoding.  It is recommended that you
+   always specify an encoding; you may use any encoding you like, but
+   an argument of "utf-8" is the most common, avoid
+   :exc:`UnicodeError` exceptions in case of unrepresentable text
+   data.
+
+   With no *encoding* argument, the result is a Unicode string, and the
+   XML declaration in the resulting string does not specify an
+   encoding. Encoding this string in an encoding other than UTF-8 is
+   likely incorrect, since UTF-8 is the default encoding of XML.
 
 .. method:: Node.toprettyxml(indent="", newl="", encoding="")
 
@@ -155,7 +157,8 @@
    indentation string and defaults to a tabulator; *newl* specifies the string
    emitted at the end of each line and defaults to ``\n``.
 
-   There's also an *encoding* argument; see :meth:`toxml`.
+   There's also an *encoding* argument, that behaves like the corresponding
+   argument of :meth:`toxml`.
 
 
 .. _dom-example:
@@ -240,7 +243,9 @@
 
 .. rubric:: Footnotes
 
-.. [#] The encoding string included in XML output should conform to the
-   appropriate standards. For example, "UTF-8" is valid, but "UTF8" is
-   not. See http://www.w3.org/TR/2006/REC-xml11-20060816/#NT-EncodingDecl
+.. [#] The encoding name included in the XML output should conform to
+   the appropriate standards. For example, "UTF-8" is valid, but
+   "UTF8" is not valid in an XML document's declaration, even though
+   Python accepts it as an encoding name.
+   See http://www.w3.org/TR/2006/REC-xml11-20060816/#NT-EncodingDecl
    and http://www.iana.org/assignments/character-sets .

From python-checkins at python.org  Mon Jul 26 01:38:47 2010
From: python-checkins at python.org (andrew.kuchling)
Date: Mon, 26 Jul 2010 01:38:47 +0200 (CEST)
Subject: [Python-checkins] r83152 - in python/branches/py3k/Lib:
	test/test_minidom.py xml/dom/minidom.py
Message-ID: <20100725233847.963C9EE99E@mail.python.org>

Author: andrew.kuchling
Date: Mon Jul 26 01:38:47 2010
New Revision: 83152

Log:
#777884: make .normalize() do nothing for childless nodes, instead of raising an exception

Modified:
   python/branches/py3k/Lib/test/test_minidom.py
   python/branches/py3k/Lib/xml/dom/minidom.py

Modified: python/branches/py3k/Lib/test/test_minidom.py
==============================================================================
--- python/branches/py3k/Lib/test/test_minidom.py	(original)
+++ python/branches/py3k/Lib/test/test_minidom.py	Mon Jul 26 01:38:47 2010
@@ -951,6 +951,14 @@
         doc.unlink()
 
 
+    def testBug0777884(self):
+        doc = parseString("<o>text</o>")
+        text = doc.documentElement.childNodes[0]
+        self.assertEquals(text.nodeType, Node.TEXT_NODE)
+        # Should run quietly, doing nothing.
+        text.normalize()
+        doc.unlink()
+
     def testBug1433694(self):
         doc = parseString("<o><i/>t</o>")
         node = doc.documentElement

Modified: python/branches/py3k/Lib/xml/dom/minidom.py
==============================================================================
--- python/branches/py3k/Lib/xml/dom/minidom.py	(original)
+++ python/branches/py3k/Lib/xml/dom/minidom.py	Mon Jul 26 01:38:47 2010
@@ -920,6 +920,10 @@
         raise xml.dom.NotFoundErr(
             self.nodeName + " nodes do not have children")
 
+    def normalize(self):
+        # For childless nodes, normalize() has nothing to do.
+        pass
+
     def replaceChild(self, newChild, oldChild):
         raise xml.dom.HierarchyRequestErr(
             self.nodeName + " nodes do not have children")

From python-checkins at python.org  Mon Jul 26 01:49:57 2010
From: python-checkins at python.org (andrew.kuchling)
Date: Mon, 26 Jul 2010 01:49:57 +0200 (CEST)
Subject: [Python-checkins] r83153 - in python/branches/release27-maint:
	Lib/test/test_minidom.py Lib/xml/dom/minidom.py
Message-ID: <20100725234957.55659EE98C@mail.python.org>

Author: andrew.kuchling
Date: Mon Jul 26 01:49:57 2010
New Revision: 83153

Log:
Merged revisions 83152 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83152 | andrew.kuchling | 2010-07-25 19:38:47 -0400 (Sun, 25 Jul 2010) | 1 line
  
  #777884: make .normalize() do nothing for childless nodes, instead of raising an exception
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/test/test_minidom.py
   python/branches/release27-maint/Lib/xml/dom/minidom.py

Modified: python/branches/release27-maint/Lib/test/test_minidom.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_minidom.py	(original)
+++ python/branches/release27-maint/Lib/test/test_minidom.py	Mon Jul 26 01:49:57 2010
@@ -945,6 +945,14 @@
         doc.unlink()
 
 
+    def testBug0777884(self):
+        doc = parseString("<o>text</o>")
+        text = doc.documentElement.childNodes[0]
+        self.assertEquals(text.nodeType, Node.TEXT_NODE)
+        # Should run quietly, doing nothing.
+        text.normalize()
+        doc.unlink()
+
     def testBug1433694(self):
         doc = parseString("<o><i/>t</o>")
         node = doc.documentElement

Modified: python/branches/release27-maint/Lib/xml/dom/minidom.py
==============================================================================
--- python/branches/release27-maint/Lib/xml/dom/minidom.py	(original)
+++ python/branches/release27-maint/Lib/xml/dom/minidom.py	Mon Jul 26 01:49:57 2010
@@ -890,6 +890,10 @@
         raise xml.dom.NotFoundErr(
             self.nodeName + " nodes do not have children")
 
+    def normalize(self):
+        # For childless nodes, normalize() has nothing to do.
+        pass
+
     def replaceChild(self, newChild, oldChild):
         raise xml.dom.HierarchyRequestErr(
             self.nodeName + " nodes do not have children")

From python-checkins at python.org  Mon Jul 26 02:27:10 2010
From: python-checkins at python.org (brian.curtin)
Date: Mon, 26 Jul 2010 02:27:10 +0200 (CEST)
Subject: [Python-checkins] r83154 - in python/branches/py3k/Lib:
	configparser.py test/test_cfgparser.py
Message-ID: <20100726002710.DD3DAEEA0E@mail.python.org>

Author: brian.curtin
Date: Mon Jul 26 02:27:10 2010
New Revision: 83154

Log:
Fix #7113. Patch by ?ukasz Langa.

Changes include using a list of lines instead of patching together using
string interpolation, and a multi-line value test cases.


Modified:
   python/branches/py3k/Lib/configparser.py
   python/branches/py3k/Lib/test/test_cfgparser.py

Modified: python/branches/py3k/Lib/configparser.py
==============================================================================
--- python/branches/py3k/Lib/configparser.py	(original)
+++ python/branches/py3k/Lib/configparser.py	Mon Jul 26 02:27:10 2010
@@ -405,12 +405,11 @@
         for section in self._sections:
             fp.write("[%s]\n" % section)
             for (key, value) in self._sections[section].items():
-                if key != "__name__":
-                    if value is None:
-                        fp.write("%s\n" % (key))
-                    else:
-                        fp.write("%s = %s\n" %
-                                 (key, str(value).replace('\n', '\n\t')))
+                if key == "__name__":
+                    continue
+                if value is not None:
+                    key = " = ".join((key, str(value).replace('\n', '\n\t')))
+                fp.write("%s\n" % (key))
             fp.write("\n")
 
     def remove_option(self, section, option):
@@ -471,10 +470,10 @@
         leading whitespace.  Blank lines, lines beginning with a '#',
         and just about everything else are ignored.
         """
-        cursect = None                            # None, or a dictionary
+        cursect = None                        # None, or a dictionary
         optname = None
         lineno = 0
-        e = None                                  # None, or an exception
+        e = None                              # None, or an exception
         while True:
             line = fp.readline()
             if not line:
@@ -490,7 +489,7 @@
             if line[0].isspace() and cursect is not None and optname:
                 value = line.strip()
                 if value:
-                    cursect[optname] = "%s\n%s" % (cursect[optname], value)
+                    cursect[optname].append(value)
             # a section header or option header?
             else:
                 # is it a section header?
@@ -515,6 +514,7 @@
                     mo = self._optcre.match(line)
                     if mo:
                         optname, vi, optval = mo.group('option', 'vi', 'value')
+                        optname = self.optionxform(optname.rstrip())
                         # This check is fine because the OPTCRE cannot
                         # match if it would set optval to None
                         if optval is not None:
@@ -525,11 +525,13 @@
                                 if pos != -1 and optval[pos-1].isspace():
                                     optval = optval[:pos]
                             optval = optval.strip()
-                        # allow empty values
-                        if optval == '""':
-                            optval = ''
-                        optname = self.optionxform(optname.rstrip())
-                        cursect[optname] = optval
+                            # allow empty values
+                            if optval == '""':
+                                optval = ''
+                            cursect[optname] = [optval]
+                        else:
+                            # valueless option handling
+                            cursect[optname] = optval
                     else:
                         # a non-fatal parsing error occurred.  set up the
                         # exception but keep going. the exception will be
@@ -542,6 +544,13 @@
         if e:
             raise e
 
+        # join the multi-line values collected while reading
+        all_sections = [self._defaults]
+        all_sections.extend(self._sections.values())
+        for options in all_sections:
+            for name, val in options.items():
+                if isinstance(val, list):
+                    options[name] = '\n'.join(val)
 
 class ConfigParser(RawConfigParser):
 

Modified: python/branches/py3k/Lib/test/test_cfgparser.py
==============================================================================
--- python/branches/py3k/Lib/test/test_cfgparser.py	(original)
+++ python/branches/py3k/Lib/test/test_cfgparser.py	Mon Jul 26 02:27:10 2010
@@ -1,7 +1,8 @@
+import collections
 import configparser
 import io
+import os
 import unittest
-import collections
 
 from test import support
 
@@ -162,7 +163,7 @@
     def test_parse_errors(self):
         self.newconfig()
         e = self.parse_error(configparser.ParsingError,
-                         "[Foo]\n  extra-spaces: splat\n")
+                             "[Foo]\n  extra-spaces: splat\n")
         self.assertEqual(e.args, ('<???>',))
         self.parse_error(configparser.ParsingError,
                          "[Foo]\n  extra-spaces= splat\n")
@@ -171,7 +172,7 @@
         self.parse_error(configparser.ParsingError,
                          "[Foo]\n=value-without-option-name\n")
         e = self.parse_error(configparser.MissingSectionHeaderError,
-                         "No Section!\n")
+                             "No Section!\n")
         self.assertEqual(e.args, ('<???>', 1, "No Section!\n"))
 
     def parse_error(self, exc, src):
@@ -185,7 +186,8 @@
         self.assertEqual(cf.sections(), [],
                          "new ConfigParser should have no defined sections")
         self.assertFalse(cf.has_section("Foo"),
-                    "new ConfigParser should have no acknowledged sections")
+                         "new ConfigParser should have no acknowledged "
+                         "sections")
         with self.assertRaises(configparser.NoSectionError) as cm:
             cf.options("Foo")
         with self.assertRaises(configparser.NoSectionError) as cm:
@@ -355,8 +357,8 @@
 
     def test_interpolation(self):
         rawval = {
-            configparser.ConfigParser: "something %(with11)s "\
-                                           "lots of interpolation (11 steps)",
+            configparser.ConfigParser: ("something %(with11)s "
+                                        "lots of interpolation (11 steps)"),
             configparser.SafeConfigParser: "%(with1)s",
         }
         cf = self.get_interpolation_config()
@@ -412,6 +414,33 @@
         self.assertRaises(ValueError, cf.get, 'non-string',
                           'string_with_interpolation', raw=False)
 
+class MultilineValuesTestCase(TestCaseBase):
+    config_class = configparser.ConfigParser
+    wonderful_spam = ("I'm having spam spam spam spam "
+                      "spam spam spam beaked beans spam "
+                      "spam spam and spam!").replace(' ', '\t\n')
+
+    def setUp(self):
+        cf = self.newconfig()
+        for i in range(100):
+            s = 'section{}'.format(i)
+            cf.add_section(s)
+            for j in range(10):
+                cf.set(s, 'lovely_spam{}'.format(j), self.wonderful_spam)
+        with open(support.TESTFN, 'w') as f:
+            cf.write(f)
+
+    def tearDown(self):
+        os.unlink(support.TESTFN)
+
+    def test_dominating_multiline_values(self):
+        # We're reading from file because this is where the code changed
+        # during performance updates in Python 3.2
+        cf_from_file = self.newconfig()
+        with open(support.TESTFN) as f:
+            cf_from_file.readfp(f)
+        self.assertEqual(cf_from_file.get('section8', 'lovely_spam4'),
+                         self.wonderful_spam.replace('\t\n', '\n'))
 
 class RawConfigParserTestCase(TestCaseBase):
     config_class = configparser.RawConfigParser
@@ -530,10 +559,11 @@
 def test_main():
     support.run_unittest(
         ConfigParserTestCase,
+        MultilineValuesTestCase,
         RawConfigParserTestCase,
         SafeConfigParserTestCase,
-        SortedTestCase,
         SafeConfigParserTestCaseNoValue,
+        SortedTestCase,
         )
 
 

From python-checkins at python.org  Mon Jul 26 04:30:15 2010
From: python-checkins at python.org (brian.curtin)
Date: Mon, 26 Jul 2010 04:30:15 +0200 (CEST)
Subject: [Python-checkins] r83155 - in python/branches/release27-maint/Lib:
	ConfigParser.py test/test_cfgparser.py
Message-ID: <20100726023015.CAAC9EE99B@mail.python.org>

Author: brian.curtin
Date: Mon Jul 26 04:30:15 2010
New Revision: 83155

Log:
Fix #7113. Patch by ?ukasz Langa.

Changes include using a list of lines instead of patching together using
string interpolation, and a multi-line value test cases.


Modified:
   python/branches/release27-maint/Lib/ConfigParser.py
   python/branches/release27-maint/Lib/test/test_cfgparser.py

Modified: python/branches/release27-maint/Lib/ConfigParser.py
==============================================================================
--- python/branches/release27-maint/Lib/ConfigParser.py	(original)
+++ python/branches/release27-maint/Lib/ConfigParser.py	Mon Jul 26 04:30:15 2010
@@ -398,12 +398,11 @@
         for section in self._sections:
             fp.write("[%s]\n" % section)
             for (key, value) in self._sections[section].items():
-                if key != "__name__":
-                    if value is None:
-                        fp.write("%s\n" % (key))
-                    else:
-                        fp.write("%s = %s\n" %
-                                 (key, str(value).replace('\n', '\n\t')))
+                if key == "__name__":
+                    continue
+                if value is not None:
+                    key = " = ".join((key, str(value).replace('\n', '\n\t')))
+                fp.write("%s\n" % (key))
             fp.write("\n")
 
     def remove_option(self, section, option):
@@ -464,10 +463,10 @@
         leading whitespace.  Blank lines, lines beginning with a '#',
         and just about everything else are ignored.
         """
-        cursect = None                            # None, or a dictionary
+        cursect = None                        # None, or a dictionary
         optname = None
         lineno = 0
-        e = None                                  # None, or an exception
+        e = None                              # None, or an exception
         while True:
             line = fp.readline()
             if not line:
@@ -483,7 +482,7 @@
             if line[0].isspace() and cursect is not None and optname:
                 value = line.strip()
                 if value:
-                    cursect[optname] = "%s\n%s" % (cursect[optname], value)
+                    cursect[optname].append(value)
             # a section header or option header?
             else:
                 # is it a section header?
@@ -508,6 +507,7 @@
                     mo = self._optcre.match(line)
                     if mo:
                         optname, vi, optval = mo.group('option', 'vi', 'value')
+                        optname = self.optionxform(optname.rstrip())
                         # This check is fine because the OPTCRE cannot
                         # match if it would set optval to None
                         if optval is not None:
@@ -518,11 +518,13 @@
                                 if pos != -1 and optval[pos-1].isspace():
                                     optval = optval[:pos]
                             optval = optval.strip()
-                        # allow empty values
-                        if optval == '""':
-                            optval = ''
-                        optname = self.optionxform(optname.rstrip())
-                        cursect[optname] = optval
+                            # allow empty values
+                            if optval == '""':
+                                optval = ''
+                            cursect[optname] = [optval]
+                        else:
+                            # valueless option handling
+                            cursect[optname] = optval
                     else:
                         # a non-fatal parsing error occurred.  set up the
                         # exception but keep going. the exception will be
@@ -535,6 +537,13 @@
         if e:
             raise e
 
+        # join the multi-line values collected while reading
+        all_sections = [self._defaults]
+        all_sections.extend(self._sections.values())
+        for options in all_sections:
+            for name, val in options.items():
+                if isinstance(val, list):
+                    options[name] = '\n'.join(val)
 
 class ConfigParser(RawConfigParser):
 

Modified: python/branches/release27-maint/Lib/test/test_cfgparser.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_cfgparser.py	(original)
+++ python/branches/release27-maint/Lib/test/test_cfgparser.py	Mon Jul 26 04:30:15 2010
@@ -1,5 +1,6 @@
 import ConfigParser
 import StringIO
+import os
 import unittest
 import UserDict
 
@@ -186,7 +187,8 @@
         self.assertEqual(cf.sections(), [],
                          "new ConfigParser should have no defined sections")
         self.assertFalse(cf.has_section("Foo"),
-                    "new ConfigParser should have no acknowledged sections")
+                         "new ConfigParser should have no acknowledged "
+                         "sections")
         self.assertRaises(ConfigParser.NoSectionError,
                           cf.options, "Foo")
         self.assertRaises(ConfigParser.NoSectionError,
@@ -357,6 +359,11 @@
     config_class = ConfigParser.ConfigParser
 
     def test_interpolation(self):
+        rawval = {
+            ConfigParser.ConfigParser: ("something %(with11)s "
+                                        "lots of interpolation (11 steps)"),
+            ConfigParser.SafeConfigParser: "%(with1)s",
+        }
         cf = self.get_interpolation_config()
         eq = self.assertEqual
         eq(cf.get("Foo", "getname"), "Foo")
@@ -403,6 +410,33 @@
         self.assertRaises(ValueError, cf.get, 'non-string',
                           'string_with_interpolation', raw=False)
 
+class MultilineValuesTestCase(TestCaseBase):
+    config_class = ConfigParser.ConfigParser
+    wonderful_spam = ("I'm having spam spam spam spam "
+                      "spam spam spam beaked beans spam "
+                      "spam spam and spam!").replace(' ', '\t\n')
+
+    def setUp(self):
+        cf = self.newconfig()
+        for i in range(100):
+            s = 'section{}'.format(i)
+            cf.add_section(s)
+            for j in range(10):
+                cf.set(s, 'lovely_spam{}'.format(j), self.wonderful_spam)
+        with open(test_support.TESTFN, 'w') as f:
+            cf.write(f)
+
+    def tearDown(self):
+        os.unlink(test_support.TESTFN)
+
+    def test_dominating_multiline_values(self):
+        # we're reading from file because this is where the code changed
+        # during performance updates in Python 3.2
+        cf_from_file = self.newconfig()
+        with open(test_support.TESTFN) as f:
+            cf_from_file.readfp(f)
+        self.assertEqual(cf_from_file.get('section8', 'lovely_spam4'),
+                         self.wonderful_spam.replace('\t\n', '\n'))
 
 class RawConfigParserTestCase(TestCaseBase):
     config_class = ConfigParser.RawConfigParser
@@ -521,10 +555,11 @@
 def test_main():
     test_support.run_unittest(
         ConfigParserTestCase,
+        MultilineValuesTestCase,
         RawConfigParserTestCase,
         SafeConfigParserTestCase,
-        SortedTestCase,
         SafeConfigParserTestCaseNoValue,
+        SortedTestCase,
         )
 
 

From python-checkins at python.org  Mon Jul 26 04:36:32 2010
From: python-checkins at python.org (brian.curtin)
Date: Mon, 26 Jul 2010 04:36:32 +0200 (CEST)
Subject: [Python-checkins] r83156 - in python/branches/py3k/Misc: ACKS NEWS
Message-ID: <20100726023632.9BE18EE994@mail.python.org>

Author: brian.curtin
Date: Mon Jul 26 04:36:32 2010
New Revision: 83156

Log:
Add note about #7113 and add ?ukasz Langa to ACKS


Modified:
   python/branches/py3k/Misc/ACKS
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Misc/ACKS
==============================================================================
--- python/branches/py3k/Misc/ACKS	(original)
+++ python/branches/py3k/Misc/ACKS	Mon Jul 26 04:36:32 2010
@@ -449,6 +449,7 @@
 Cameron Laird
 Torsten Landschoff
 Tino Lange
+?ukasz Langa
 Andrew Langmead
 Detlef Lannert
 Soren Larsen

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Mon Jul 26 04:36:32 2010
@@ -473,6 +473,8 @@
 Library
 -------
 
+- Issue #7113: Speed up loading in configparser. Patch by ?ukasz Langa.
+
 - Issue #9032: XML-RPC client retries the request on EPIPE error. The EPIPE
   error occurs when the server closes the socket and the client sends a big
   XML-RPC request.

From python-checkins at python.org  Mon Jul 26 04:36:41 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Mon, 26 Jul 2010 04:36:41 +0200 (CEST)
Subject: [Python-checkins] r83157 - in python/branches/py3k/Lib: datetime.py
	test/datetimetester.py
Message-ID: <20100726023641.ADB81EEA1E@mail.python.org>

Author: alexander.belopolsky
Date: Mon Jul 26 04:36:41 2010
New Revision: 83157

Log:
Make python version of fromtimestamp behave more like C.

Modified:
   python/branches/py3k/Lib/datetime.py
   python/branches/py3k/Lib/test/datetimetester.py

Modified: python/branches/py3k/Lib/datetime.py
==============================================================================
--- python/branches/py3k/Lib/datetime.py	(original)
+++ python/branches/py3k/Lib/datetime.py	Mon Jul 26 04:36:41 2010
@@ -1356,16 +1356,20 @@
         """
 
         _check_tzinfo_arg(tz)
-        if tz is None:
-            converter = _time.localtime
-        else:
-            converter = _time.gmtime
-        if 1 - (t % 1.0) < 0.000001:
-            t = float(int(t)) + 1
-        if t < 0:
-            t -= 1
+
+        converter = _time.localtime if tz is None else _time.gmtime
+
+        t, frac = divmod(t, 1.0)
+        us = round(frac * 1e6)
+
+        # If timestamp is less than one microsecond smaller than a
+        # full second, us can be rounded up to 1000000.  In this case,
+        # roll over to seconds, otherwise, ValueError is raised
+        # by the constructor.
+        if us == 1000000:
+            t += 1
+            us = 0
         y, m, d, hh, mm, ss, weekday, jday, dst = converter(t)
-        us = int((t % 1.0) * 1000000)
         ss = min(ss, 59)    # clamp out leap seconds if the platform has them
         result = cls(y, m, d, hh, mm, ss, us, tz)
         if tz is not None:

Modified: python/branches/py3k/Lib/test/datetimetester.py
==============================================================================
--- python/branches/py3k/Lib/test/datetimetester.py	(original)
+++ python/branches/py3k/Lib/test/datetimetester.py	Mon Jul 26 04:36:41 2010
@@ -1728,9 +1728,15 @@
 
     def test_microsecond_rounding(self):
         # Test whether fromtimestamp "rounds up" floats that are less
-        # than one microsecond smaller than an integer.
+        # than 1/2 microsecond smaller than an integer.
         self.assertEqual(self.theclass.fromtimestamp(0.9999999),
                          self.theclass.fromtimestamp(1))
+        self.assertEqual(self.theclass.fromtimestamp(0.99999949).microsecond,
+                         999999)
+        # XXX Arguably incorrect behavior.  Since round(0.6112295, 6)
+        # returns 0.611229, we should see 611229 us below, not 611230
+        self.assertEqual(self.theclass.fromtimestamp(0.6112295).microsecond,
+                         611230)
 
     def test_insane_fromtimestamp(self):
         # It's possible that some platform maps time_t to double,

From python-checkins at python.org  Mon Jul 26 04:40:17 2010
From: python-checkins at python.org (brian.curtin)
Date: Mon, 26 Jul 2010 04:40:17 +0200 (CEST)
Subject: [Python-checkins] r83158 - in python/branches/release27-maint:
	Misc/ACKS Misc/NEWS
Message-ID: <20100726024017.83941EE98C@mail.python.org>

Author: brian.curtin
Date: Mon Jul 26 04:40:17 2010
New Revision: 83158

Log:
Merged revisions 83156 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83156 | brian.curtin | 2010-07-25 21:36:32 -0500 (Sun, 25 Jul 2010) | 2 lines
  
  Add note about #7113 and add ?ukasz Langa to ACKS
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Misc/ACKS
   python/branches/release27-maint/Misc/NEWS

Modified: python/branches/release27-maint/Misc/ACKS
==============================================================================
--- python/branches/release27-maint/Misc/ACKS	(original)
+++ python/branches/release27-maint/Misc/ACKS	Mon Jul 26 04:40:17 2010
@@ -438,6 +438,7 @@
 Vladimir Kushnir
 Cameron Laird
 Tino Lange
+?ukasz Langa
 Andrew Langmead
 Detlef Lannert
 Soren Larsen

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Mon Jul 26 04:40:17 2010
@@ -18,6 +18,8 @@
 Library
 -------
 
+- Issue #7113: Speed up loading in ConfigParser. Patch by ?ukasz Langa.
+
 - Issue #3704: cookielib was not properly handling URLs with a / in the
   parameters.
 

From python-checkins at python.org  Mon Jul 26 04:43:43 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Mon, 26 Jul 2010 04:43:43 +0200 (CEST)
Subject: [Python-checkins] r83159 - python/branches/release31-maint
Message-ID: <20100726024343.6083DEEA21@mail.python.org>

Author: alexander.belopolsky
Date: Mon Jul 26 04:43:43 2010
New Revision: 83159

Log:
Blocked revisions 83157 via svnmerge

........
  r83157 | alexander.belopolsky | 2010-07-25 22:36:41 -0400 (Sun, 25 Jul 2010) | 1 line
  
  Make python version of fromtimestamp behave more like C.
........


Modified:
   python/branches/release31-maint/   (props changed)

From python-checkins at python.org  Mon Jul 26 10:51:42 2010
From: python-checkins at python.org (georg.brandl)
Date: Mon, 26 Jul 2010 10:51:42 +0200 (CEST)
Subject: [Python-checkins] r83160 -
	python/branches/py3k/Doc/c-api/unicode.rst
Message-ID: <20100726085142.48373EEA4D@mail.python.org>

Author: georg.brandl
Date: Mon Jul 26 10:51:42 2010
New Revision: 83160

Log:
#9381: fix markup.

Modified:
   python/branches/py3k/Doc/c-api/unicode.rst

Modified: python/branches/py3k/Doc/c-api/unicode.rst
==============================================================================
--- python/branches/py3k/Doc/c-api/unicode.rst	(original)
+++ python/branches/py3k/Doc/c-api/unicode.rst	Mon Jul 26 10:51:42 2010
@@ -376,7 +376,7 @@
 :cdata:`Py_FileSystemEncoding` should be used as the encoding, and
 ``"surrogateescape"`` should be used as the error handler (:pep:`383`). To
 encode file names during argument parsing, the ``"O&"`` converter should be
-used, passsing :func:PyUnicode_FSConverter as the conversion function:
+used, passsing :func:`PyUnicode_FSConverter` as the conversion function:
 
 .. cfunction:: int PyUnicode_FSConverter(PyObject* obj, void* result)
 

From python-checkins at python.org  Mon Jul 26 11:33:13 2010
From: python-checkins at python.org (georg.brandl)
Date: Mon, 26 Jul 2010 11:33:13 +0200 (CEST)
Subject: [Python-checkins] r83161 - python/branches/py3k/Misc/developers.txt
Message-ID: <20100726093313.10960EE9B4@mail.python.org>

Author: georg.brandl
Date: Mon Jul 26 11:33:12 2010
New Revision: 83161

Log:
Add Brian Quinlan.

Modified:
   python/branches/py3k/Misc/developers.txt

Modified: python/branches/py3k/Misc/developers.txt
==============================================================================
--- python/branches/py3k/Misc/developers.txt	(original)
+++ python/branches/py3k/Misc/developers.txt	Mon Jul 26 11:33:12 2010
@@ -20,6 +20,9 @@
 Permissions History
 -------------------
 
+- Brian Quinlan was given commit access on Jul 26 2010 by GFB,
+  for work related to PEP 3148.
+
 - Alexander Belopolsky was given commit access on May 25 2010
   by MvL at suggestion of Mark Dickinson.
 

From python-checkins at python.org  Mon Jul 26 14:54:02 2010
From: python-checkins at python.org (andrew.kuchling)
Date: Mon, 26 Jul 2010 14:54:02 +0200 (CEST)
Subject: [Python-checkins] r83162 -
	python/branches/py3k/Doc/library/xml.dom.minidom.rst
Message-ID: <20100726125402.6EB08EE9DB@mail.python.org>

Author: andrew.kuchling
Date: Mon Jul 26 14:54:02 2010
New Revision: 83162

Log:
#7637: fix a grammar error; simplify a sentence

Modified:
   python/branches/py3k/Doc/library/xml.dom.minidom.rst

Modified: python/branches/py3k/Doc/library/xml.dom.minidom.rst
==============================================================================
--- python/branches/py3k/Doc/library/xml.dom.minidom.rst	(original)
+++ python/branches/py3k/Doc/library/xml.dom.minidom.rst	Mon Jul 26 14:54:02 2010
@@ -142,7 +142,7 @@
    With an explicit *encoding* [1]_ argument, the result is a byte
    string in the specified encoding.  It is recommended that you
    always specify an encoding; you may use any encoding you like, but
-   an argument of "utf-8" is the most common, avoid
+   an argument of "utf-8" is the most common choice, avoiding
    :exc:`UnicodeError` exceptions in case of unrepresentable text
    data.
 
@@ -157,8 +157,8 @@
    indentation string and defaults to a tabulator; *newl* specifies the string
    emitted at the end of each line and defaults to ``\n``.
 
-   There's also an *encoding* argument, that behaves like the corresponding
-   argument of :meth:`toxml`.
+   The *encoding* argument behaves like the corresponding argument of
+   :meth:`toxml`.
 
 
 .. _dom-example:

From python-checkins at python.org  Mon Jul 26 15:08:58 2010
From: python-checkins at python.org (andrew.kuchling)
Date: Mon, 26 Jul 2010 15:08:58 +0200 (CEST)
Subject: [Python-checkins] r83163 - python/branches/py3k/Doc/library/os.rst
Message-ID: <20100726130858.EA895F4E7@mail.python.org>

Author: andrew.kuchling
Date: Mon Jul 26 15:08:58 2010
New Revision: 83163

Log:
Reword paragraph

Modified:
   python/branches/py3k/Doc/library/os.rst

Modified: python/branches/py3k/Doc/library/os.rst
==============================================================================
--- python/branches/py3k/Doc/library/os.rst	(original)
+++ python/branches/py3k/Doc/library/os.rst	Mon Jul 26 15:08:58 2010
@@ -1923,8 +1923,9 @@
 
    The :mod:`subprocess` module provides more powerful facilities for spawning new
    processes and retrieving their results; using that module is preferable to using
-   this function.  Use the :mod:`subprocess` module.  Check especially the
-   :ref:`subprocess-replacements` section.
+   this function.  See the
+   :ref:`subprocess-replacements` section in the :mod:`subprocess` documentation
+   for some helpful recipes.
 
    Availability: Unix, Windows.
 

From python-checkins at python.org  Mon Jul 26 15:42:35 2010
From: python-checkins at python.org (andrew.kuchling)
Date: Mon, 26 Jul 2010 15:42:35 +0200 (CEST)
Subject: [Python-checkins] r83164 - in python/branches/release27-maint:
	Doc/library/os.rst
Message-ID: <20100726134235.CB152EE9C8@mail.python.org>

Author: andrew.kuchling
Date: Mon Jul 26 15:42:35 2010
New Revision: 83164

Log:
Merged revisions 83163 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83163 | andrew.kuchling | 2010-07-26 09:08:58 -0400 (Mon, 26 Jul 2010) | 1 line
  
  Reword paragraph
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Doc/library/os.rst

Modified: python/branches/release27-maint/Doc/library/os.rst
==============================================================================
--- python/branches/release27-maint/Doc/library/os.rst	(original)
+++ python/branches/release27-maint/Doc/library/os.rst	Mon Jul 26 15:42:35 2010
@@ -2096,8 +2096,9 @@
 
    The :mod:`subprocess` module provides more powerful facilities for spawning new
    processes and retrieving their results; using that module is preferable to using
-   this function.  Use the :mod:`subprocess` module.  Check especially the
-   :ref:`subprocess-replacements` section.
+   this function.  See the
+   :ref:`subprocess-replacements` section in the :mod:`subprocess` documentation
+   for some helpful recipes.
 
    Availability: Unix, Windows.
 

From python-checkins at python.org  Mon Jul 26 17:11:49 2010
From: python-checkins at python.org (georg.brandl)
Date: Mon, 26 Jul 2010 17:11:49 +0200 (CEST)
Subject: [Python-checkins] r83166 -
	python/branches/py3k/Doc/install/index.rst
Message-ID: <20100726151149.99398EEA26@mail.python.org>

Author: georg.brandl
Date: Mon Jul 26 17:11:49 2010
New Revision: 83166

Log:
Fix grammar.

Modified:
   python/branches/py3k/Doc/install/index.rst

Modified: python/branches/py3k/Doc/install/index.rst
==============================================================================
--- python/branches/py3k/Doc/install/index.rst	(original)
+++ python/branches/py3k/Doc/install/index.rst	Mon Jul 26 17:11:49 2010
@@ -314,8 +314,8 @@
 stash of Python modules.  This scheme's name is derived from the idea of a
 "home" directory on Unix, since it's not unusual for a Unix user to make their
 home directory have a layout similar to :file:`/usr/` or :file:`/usr/local/`.
-This scheme can be used by anyone, regardless of the operating system their
-installing for.
+This scheme can be used by anyone, regardless of the operating system they
+are installing for.
 
 Installing a new module distribution is as simple as ::
 

From python-checkins at python.org  Mon Jul 26 17:38:01 2010
From: python-checkins at python.org (ezio.melotti)
Date: Mon, 26 Jul 2010 17:38:01 +0200 (CEST)
Subject: [Python-checkins] r83167 - in
	tracker/instances/python-dev/extensions: local_replace.py
	test/test_local_replace.py
Message-ID: <20100726153801.F0ED2EE981@mail.python.org>

Author: ezio.melotti
Date: Mon Jul 26 17:38:01 2010
New Revision: 83167

Log:
Trunk is dead, link to py3k.

Modified:
   tracker/instances/python-dev/extensions/local_replace.py
   tracker/instances/python-dev/extensions/test/test_local_replace.py

Modified: tracker/instances/python-dev/extensions/local_replace.py
==============================================================================
--- tracker/instances/python-dev/extensions/local_replace.py	(original)
+++ tracker/instances/python-dev/extensions/local_replace.py	Mon Jul 26 17:38:01 2010
@@ -14,8 +14,8 @@
     (re.compile(r'(?P<sep>(?<!\w/)|(?<!\w)/)(?P<path>(?:Demo|Doc|Grammar|'
                 r'Include|Lib|Mac|Misc|Modules|Parser|PC|PCbuild|Python|'
                 'RISCOS|Tools|Objects)/[-.a-zA-Z0-9_/]+[a-zA-Z0-9]/?)'),
-     r'<a href="http://svn.python.org/view/python/'
-     r'trunk/\g<path>">\g<sep>\g<path></a>'), # XXX trunk is dead
+     r'<a href="http://svn.python.org/view/python/branches/'
+     r'py3k/\g<path>">\g<sep>\g<path></a>'),
 ]
 
 

Modified: tracker/instances/python-dev/extensions/test/test_local_replace.py
==============================================================================
--- tracker/instances/python-dev/extensions/test/test_local_replace.py	(original)
+++ tracker/instances/python-dev/extensions/test/test_local_replace.py	Mon Jul 26 17:38:01 2010
@@ -96,17 +96,17 @@
 
     ## Lib/somefile.py, Modules/somemodule.c, Doc/somedocfile.rst, ...
     ('Lib/cgi.py',
-     '<a href="http://svn.python.org/view/python/trunk/Lib/cgi.py">Lib/cgi.py</a>'),
+     '<a href="http://svn.python.org/view/python/branches/py3k/Lib/cgi.py">Lib/cgi.py</a>'),
     ('/Lib/cgi.py',
-     '<a href="http://svn.python.org/view/python/trunk/Lib/cgi.py">/Lib/cgi.py</a>'),
+     '<a href="http://svn.python.org/view/python/branches/py3k/Lib/cgi.py">/Lib/cgi.py</a>'),
     ('see Lib/cgi.py.',
-     'see <a href="http://svn.python.org/view/python/trunk/Lib/cgi.py">Lib/cgi.py</a>.'),
+     'see <a href="http://svn.python.org/view/python/branches/py3k/Lib/cgi.py">Lib/cgi.py</a>.'),
     ('see /Lib/cgi.py.',
-     'see <a href="http://svn.python.org/view/python/trunk/Lib/cgi.py">/Lib/cgi.py</a>.'),
+     'see <a href="http://svn.python.org/view/python/branches/py3k/Lib/cgi.py">/Lib/cgi.py</a>.'),
     ('(Lib/cgi.py)',
-     '(<a href="http://svn.python.org/view/python/trunk/Lib/cgi.py">Lib/cgi.py</a>)'),
+     '(<a href="http://svn.python.org/view/python/branches/py3k/Lib/cgi.py">Lib/cgi.py</a>)'),
     ('(/Lib/cgi.py)',
-     '(<a href="http://svn.python.org/view/python/trunk/Lib/cgi.py">/Lib/cgi.py</a>)'),
+     '(<a href="http://svn.python.org/view/python/branches/py3k/Lib/cgi.py">/Lib/cgi.py</a>)'),
 
     ## URLs
     ('http://svn.python.org/view/python/tags/r265 ',

From python-checkins at python.org  Mon Jul 26 19:00:20 2010
From: python-checkins at python.org (georg.brandl)
Date: Mon, 26 Jul 2010 19:00:20 +0200 (CEST)
Subject: [Python-checkins] r83168 -
	python/branches/py3k/Doc/library/argparse.rst
Message-ID: <20100726170020.39CC0EE984@mail.python.org>

Author: georg.brandl
Date: Mon Jul 26 19:00:20 2010
New Revision: 83168

Log:
Fix indentation in example.

Modified:
   python/branches/py3k/Doc/library/argparse.rst

Modified: python/branches/py3k/Doc/library/argparse.rst
==============================================================================
--- python/branches/py3k/Doc/library/argparse.rst	(original)
+++ python/branches/py3k/Doc/library/argparse.rst	Mon Jul 26 19:00:20 2010
@@ -698,8 +698,8 @@
 
    >>> class FooAction(argparse.Action):
    ...     def __call__(self, parser, namespace, values, option_string=None):
-   ...     print('%r %r %r' % (namespace, values, option_string))
-   ...     setattr(namespace, self.dest, values)
+   ...         print('%r %r %r' % (namespace, values, option_string))
+   ...         setattr(namespace, self.dest, values)
    ...
    >>> parser = argparse.ArgumentParser()
    >>> parser.add_argument('--foo', action=FooAction)

From python-checkins at python.org  Mon Jul 26 19:09:33 2010
From: python-checkins at python.org (georg.brandl)
Date: Mon, 26 Jul 2010 19:09:33 +0200 (CEST)
Subject: [Python-checkins] r83169 - python/branches/py3k/Misc/developers.txt
Message-ID: <20100726170933.1CBD2F5D6@mail.python.org>

Author: georg.brandl
Date: Mon Jul 26 19:09:32 2010
New Revision: 83169

Log:
Add Reid.

Modified:
   python/branches/py3k/Misc/developers.txt

Modified: python/branches/py3k/Misc/developers.txt
==============================================================================
--- python/branches/py3k/Misc/developers.txt	(original)
+++ python/branches/py3k/Misc/developers.txt	Mon Jul 26 19:09:32 2010
@@ -23,6 +23,10 @@
 - Brian Quinlan was given commit access on Jul 26 2010 by GFB,
   for work related to PEP 3148.
 
+- Reid Kleckner was given commit access on Jul 11 2010 by GFB,
+  for work on the py3k-jit branch, at suggestion of the Unladen
+  Swallow team.
+
 - Alexander Belopolsky was given commit access on May 25 2010
   by MvL at suggestion of Mark Dickinson.
 

From python-checkins at python.org  Mon Jul 26 20:27:49 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Mon, 26 Jul 2010 20:27:49 +0200 (CEST)
Subject: [Python-checkins] r83170 -
	python/branches/py3k/Doc/library/pickletools.rst
Message-ID: <20100726182749.5581BEE987@mail.python.org>

Author: alexander.belopolsky
Date: Mon Jul 26 20:27:49 2010
New Revision: 83170

Log:
Added versionadded entry for the annotate argument.

Modified:
   python/branches/py3k/Doc/library/pickletools.rst

Modified: python/branches/py3k/Doc/library/pickletools.rst
==============================================================================
--- python/branches/py3k/Doc/library/pickletools.rst	(original)
+++ python/branches/py3k/Doc/library/pickletools.rst	Mon Jul 26 20:27:49 2010
@@ -26,6 +26,8 @@
    a short description.  The value of *annotate* is used as a hint for
    the column where annotation should start.
 
+  .. versionadded:: 3.2 the *annotate* argument.
+
 .. function:: genops(pickle)
 
    Provides an :term:`iterator` over all of the opcodes in a pickle, returning a

From python-checkins at python.org  Mon Jul 26 23:12:14 2010
From: python-checkins at python.org (georg.brandl)
Date: Mon, 26 Jul 2010 23:12:14 +0200 (CEST)
Subject: [Python-checkins] r83171 -
	python/branches/py3k/Doc/documenting/markup.rst
Message-ID: <20100726211214.08D84EE9EC@mail.python.org>

Author: georg.brandl
Date: Mon Jul 26 23:12:13 2010
New Revision: 83171

Log:
Clarify.

Modified:
   python/branches/py3k/Doc/documenting/markup.rst

Modified: python/branches/py3k/Doc/documenting/markup.rst
==============================================================================
--- python/branches/py3k/Doc/documenting/markup.rst	(original)
+++ python/branches/py3k/Doc/documenting/markup.rst	Mon Jul 26 23:12:13 2010
@@ -200,8 +200,8 @@
 
 .. describe:: cmdoption
 
-   Describes a command line option or switch.  Option argument names should be
-   enclosed in angle brackets.  Example::
+   Describes a Python command line option or switch.  Option argument names
+   should be enclosed in angle brackets.  Example::
 
       .. cmdoption:: -m <module>
 

From python-checkins at python.org  Tue Jul 27 00:20:53 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Tue, 27 Jul 2010 00:20:53 +0200 (CEST)
Subject: [Python-checkins] r83172 - peps/trunk/pep-3151.txt
Message-ID: <20100726222053.25E47EE9F3@mail.python.org>

Author: antoine.pitrou
Date: Tue Jul 27 00:20:52 2010
New Revision: 83172

Log:
Typo fixes by ?ric Araujo.



Modified:
   peps/trunk/pep-3151.txt

Modified: peps/trunk/pep-3151.txt
==============================================================================
--- peps/trunk/pep-3151.txt	(original)
+++ peps/trunk/pep-3151.txt	Tue Jul 27 00:20:52 2010
@@ -40,7 +40,7 @@
 --------------------------------------
 
 OS-related (or system call-related) exceptions are currently a diversity
-of classes, arranged in the following subhierarchies::
+of classes, arranged in the following sub-hierarchies::
 
     +-- EnvironmentError
         +-- IOError
@@ -95,7 +95,7 @@
 anyway.
 
 .. note::
-    `Appendix B`_ surveys the use of the various exception types accross
+    `Appendix B`_ surveys the use of the various exception types across
     the interpreter and the standard library.
 
 
@@ -178,7 +178,7 @@
 
 * *useful compatibility* doesn't alter the behaviour of *careful*
   exception-catching code.  Given the following kind of snippet, the same
-  errors should be silenced or reraised, regardless of whether this PEP
+  errors should be silenced or re-raised, regardless of whether this PEP
   has been implemented or not::
 
       try:
@@ -213,9 +213,9 @@
 Each of these changes doesn't preserve exact compatibility, but it does
 preserve *useful compatibility* (see "compatibility" section above).
 
-Each of these changes can be accepted or refused invididually, but of course
+Each of these changes can be accepted or refused individually, but of course
 it is considered that the greatest impact can be achieved if this first step
-is accepted in full.  In this case, the IO exception subhierarchy would
+is accepted in full.  In this case, the IO exception sub-hierarchy would
 become::
 
     +-- IOError   (replacing OSError, WindowsError, EnvironmentError, etc.)
@@ -667,7 +667,7 @@
 mmap
 ''''
 
-Undex Unix, raises its own ``mmap.error`` (inheriting from EnvironmentError)
+Under Unix, raises its own ``mmap.error`` (inheriting from EnvironmentError)
 throughout::
 
     >>> mmap.mmap(123, 10)
@@ -711,7 +711,7 @@
 ''''''''''
 
 The ``os`` (or ``posix``) module raises OSError throughout, except under
-Windows where WindosError can be raised instead.
+Windows where WindowsError can be raised instead.
 
 ossaudiodev
 '''''''''''

From python-checkins at python.org  Tue Jul 27 07:31:29 2010
From: python-checkins at python.org (gregory.p.smith)
Date: Tue, 27 Jul 2010 07:31:29 +0200 (CEST)
Subject: [Python-checkins] r83173 - in python/branches/py3k: Lib/re.py
	Lib/test/test_re.py Misc/NEWS
Message-ID: <20100727053129.BE8F3EE987@mail.python.org>

Author: gregory.p.smith
Date: Tue Jul 27 07:31:29 2010
New Revision: 83173

Log:
The default size of the re module's compiled regular expression cache has
been increased from 100 to 500 and the cache replacement policy has changed
from simply clearing the entire cache on overflow to randomly forgetting 20%
of the existing cached compiled regular expressions.  This is a performance
win for applications that use a lot of regular expressions and limits the
impact of the performance hit anytime the cache is exceeded.


Modified:
   python/branches/py3k/Lib/re.py
   python/branches/py3k/Lib/test/test_re.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/re.py
==============================================================================
--- python/branches/py3k/Lib/re.py	(original)
+++ python/branches/py3k/Lib/re.py	Tue Jul 27 07:31:29 2010
@@ -254,7 +254,40 @@
 
 _pattern_type = type(sre_compile.compile("", 0))
 
-_MAXCACHE = 100
+_MAXCACHE = 500
+
+def _shrink_cache(cache_dict, max_length, divisor=5):
+    """Make room in the given cache.
+
+    Args:
+        cache_dict: The cache dictionary to modify.
+        max_length: Maximum # of entries in cache_dict before it is shrunk.
+        divisor: Cache will shrink to max_length - 1/divisor*max_length items.
+    """
+    # Toss out a fraction of the entries at random to make room for new ones.
+    # A random algorithm was chosen as opposed to simply cache_dict.popitem()
+    # as popitem could penalize the same regular expression repeatedly based
+    # on its internal hash value.  Being random should spread the cache miss
+    # love around.
+    cache_keys = tuple(cache_dict.keys())
+    overage = len(cache_keys) - max_length
+    if overage < 0:
+        # Cache is already within limits.  Normally this should not happen
+        # but it could due to multithreading.
+        return
+    number_to_toss = max_length // divisor + overage
+    # The import is done here to avoid a circular depencency.
+    import random
+    if not hasattr(random, 'sample'):
+        # Do nothing while resolving the circular dependency:
+        #  re->random->warnings->tokenize->string->re
+        return
+    for doomed_key in random.sample(cache_keys, number_to_toss):
+        try:
+            del cache_dict[doomed_key]
+        except KeyError:
+            # Ignore problems if the cache changed from another thread.
+            pass
 
 def _compile(*key):
     # internal: compile pattern
@@ -272,7 +305,7 @@
         raise TypeError("first argument must be string or compiled pattern")
     p = sre_compile.compile(pattern, flags)
     if len(_cache) >= _MAXCACHE:
-        _cache.clear()
+        _shrink_cache(_cache, _MAXCACHE)
     _cache[cachekey] = p
     return p
 
@@ -284,7 +317,7 @@
     repl, pattern = key
     p = sre_parse.parse_template(repl, pattern)
     if len(_cache_repl) >= _MAXCACHE:
-        _cache_repl.clear()
+        _shrink_cache(_cache_repl, _MAXCACHE)
     _cache_repl[key] = p
     return p
 

Modified: python/branches/py3k/Lib/test/test_re.py
==============================================================================
--- python/branches/py3k/Lib/test/test_re.py	(original)
+++ python/branches/py3k/Lib/test/test_re.py	Tue Jul 27 07:31:29 2010
@@ -874,8 +874,71 @@
                 if result is None:
                     print('=== Fails on unicode-sensitive match', t)
 
+
+class ReCacheTests(unittest.TestCase):
+    """These tests are specific to the re._shrink_cache implementation."""
+
+    def setUp(self):
+        self._orig_maxcache = re._MAXCACHE
+
+    def tearDown(self):
+        re._MAXCACHE = self._orig_maxcache
+
+    def test_compile_cache_overflow(self):
+        # NOTE: If a profiler or debugger is tracing code and compiling
+        # regular expressions while tracing through this test... expect
+        # the test to fail.  This test is not concurrency safe.
+
+        # Explicitly fill the caches.
+        re._MAXCACHE = 20
+        max_cache = re._MAXCACHE
+        unique_chars = tuple(chr(char_num) for char_num in
+                             range(b'a'[0], b'a'[0]+max_cache))
+        re._cache.clear()
+        for char in unique_chars:
+            re._compile(char, 0)
+        self.assertEqual(max_cache, len(re._cache))
+        re._cache_repl.clear()
+        for char in unique_chars:
+            re._compile_repl(char*2, char)
+        self.assertEqual(max_cache, len(re._cache_repl))
+
+        # Overflow both caches and make sure they have extra room left
+        # afterwards as well as having more than a single entry.
+        re._compile('A', 0)
+        self.assertLess(len(re._cache), max_cache)
+        self.assertGreater(len(re._cache), 1)
+        re._compile_repl('A', 'A')
+        self.assertLess(len(re._cache_repl), max_cache)
+        self.assertGreater(len(re._cache_repl), 1)
+
+    def test_shrink_cache_at_limit(self):
+        cache = dict(zip(range(6), range(6)))
+        re._shrink_cache(cache, 6, divisor=3)
+        self.assertEqual(4, len(cache))
+
+    def test_shrink_cache_empty(self):
+        cache = {}
+        re._shrink_cache(cache, 6, divisor=3)
+        # Cache was empty, make sure we didn't raise an exception.
+        self.assertEqual(0, len(cache))
+
+    def test_shrink_cache_overflowing(self):
+        cache = dict(zip(range(6), range(6)))
+        re._shrink_cache(cache, 4, divisor=2)
+        # Cache was larger than the maximum, be sure we shrunk to smaller.
+        self.assertEqual(2, len(cache))
+
+    def test_shrink_cache_underflow(self):
+        cache = dict(zip(range(6), range(6)))
+        # No shrinking to do.
+        re._shrink_cache(cache, 9, divisor=3)
+        self.assertEqual(6, len(cache))
+
+
 def test_main():
     run_unittest(ReTests)
+    run_unittest(ReCacheTests)
     run_re_tests()
 
 if __name__ == "__main__":

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Tue Jul 27 07:31:29 2010
@@ -473,6 +473,13 @@
 Library
 -------
 
+- The default size of the re module's compiled regular expression cache has
+  been increased from 100 to 500 and the cache replacement policy has changed
+  from simply clearing the entire cache on overflow to randomly forgetting 20%
+  of the existing cached compiled regular expressions.  This is a performance
+  win for applications that use a lot of regular expressions and limits the
+  impact of the performance hit anytime the cache is exceeded.
+
 - Issue #7113: Speed up loading in configparser. Patch by ?ukasz Langa.
 
 - Issue #9032: XML-RPC client retries the request on EPIPE error. The EPIPE

From python-checkins at python.org  Tue Jul 27 10:40:58 2010
From: python-checkins at python.org (georg.brandl)
Date: Tue, 27 Jul 2010 10:40:58 +0200 (CEST)
Subject: [Python-checkins] r83174 - python/branches/release27-maint
Message-ID: <20100727084058.0F0E1EE99E@mail.python.org>

Author: georg.brandl
Date: Tue Jul 27 10:40:57 2010
New Revision: 83174

Log:
Blocked revisions 82526,82593 via svnmerge

........
  r82526 | georg.brandl | 2010-07-04 09:25:18 +0200 (So, 04 Jul 2010) | 1 line
  
  Remove old and unsafe Dbm demo class.
........
  r82593 | georg.brandl | 2010-07-05 22:13:06 +0200 (Mo, 05 Jul 2010) | 1 line
  
  Remove Dbm.
........


Modified:
   python/branches/release27-maint/   (props changed)

From python-checkins at python.org  Tue Jul 27 10:45:36 2010
From: python-checkins at python.org (georg.brandl)
Date: Tue, 27 Jul 2010 10:45:36 +0200 (CEST)
Subject: [Python-checkins] r83175 - in python/branches/release27-maint:
	Demo/classes/Vec.py
Message-ID: <20100727084536.0C07EEEA96@mail.python.org>

Author: georg.brandl
Date: Tue Jul 27 10:45:35 2010
New Revision: 83175

Log:
Merged revisions 82594,82599 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r82594 | georg.brandl | 2010-07-05 22:13:41 +0200 (Mo, 05 Jul 2010) | 1 line
  
  Update Vec class constructor, remove indirection via function, use operator module.
........
  r82599 | alexander.belopolsky | 2010-07-05 23:44:05 +0200 (Mo, 05 Jul 2010) | 1 line
  
  "Modernized" the demo a little.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Demo/classes/Vec.py

Modified: python/branches/release27-maint/Demo/classes/Vec.py
==============================================================================
--- python/branches/release27-maint/Demo/classes/Vec.py	(original)
+++ python/branches/release27-maint/Demo/classes/Vec.py	Tue Jul 27 10:45:35 2010
@@ -1,23 +1,41 @@
-# A simple vector class
+class Vec:
+    """ A simple vector class
 
+    Instances of the Vec class  can be constructed from numbers
 
-def vec(*v):
-    return Vec(*v)
+    >>> a = Vec(1, 2, 3)
+    >>> b = Vec(3, 2, 1)
 
+    added
+    >>> a + b
+    Vec(4, 4, 4)
 
-class Vec:
+    subtracted
+    >>> a - b
+    Vec(-2, 0, 2)
+
+    and multiplied by a scalar on the left
+    >>> 3.0 * a
+    Vec(3.0, 6.0, 9.0)
 
+    or on the right
+    >>> a * 3.0
+    Vec(3.0, 6.0, 9.0)
+    """
     def __init__(self, *v):
         self.v = list(v)
 
-    def fromlist(self, v):
+    @classmethod
+    def fromlist(cls, v):
         if not isinstance(v, list):
             raise TypeError
-        self.v = v[:]
-        return self
+        inst = cls()
+        inst.v = v
+        return inst
 
     def __repr__(self):
-        return 'vec(' + repr(self.v)[1:-1] + ')'
+        args = ', '.join(repr(x) for x in self.v)
+        return 'Vec({0})'.format(args)
 
     def __len__(self):
         return len(self.v)
@@ -27,28 +45,24 @@
 
     def __add__(self, other):
         # Element-wise addition
-        v = map(lambda x, y: x+y, self, other)
-        return Vec().fromlist(v)
+        v = [x + y for x, y in zip(self.v, other.v)]
+        return Vec.fromlist(v)
 
     def __sub__(self, other):
         # Element-wise subtraction
-        v = map(lambda x, y: x-y, self, other)
-        return Vec().fromlist(v)
+        v = [x - y for x, y in zip(self.v, other.v)]
+        return Vec.fromlist(v)
 
     def __mul__(self, scalar):
         # Multiply by scalar
-        v = map(lambda x: x*scalar, self.v)
-        return Vec().fromlist(v)
+        v = [x * scalar for x in self.v]
+        return Vec.fromlist(v)
 
+    __rmul__ = __mul__
 
 
 def test():
-    a = vec(1, 2, 3)
-    b = vec(3, 2, 1)
-    print a
-    print b
-    print a+b
-    print a-b
-    print a*3.0
+    import doctest
+    doctest.testmod()
 
 test()

From python-checkins at python.org  Tue Jul 27 10:46:45 2010
From: python-checkins at python.org (georg.brandl)
Date: Tue, 27 Jul 2010 10:46:45 +0200 (CEST)
Subject: [Python-checkins] r83176 - python/branches/release27-maint
Message-ID: <20100727084645.10760EEA23@mail.python.org>

Author: georg.brandl
Date: Tue Jul 27 10:46:44 2010
New Revision: 83176

Log:
Blocked revisions 82759 via svnmerge

........
  r82759 | georg.brandl | 2010-07-10 12:32:36 +0200 (Sa, 10 Jul 2010) | 1 line
  
  #3071: tell how many values were expected when unpacking too many.
........


Modified:
   python/branches/release27-maint/   (props changed)

From python-checkins at python.org  Tue Jul 27 16:16:36 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Tue, 27 Jul 2010 16:16:36 +0200 (CEST)
Subject: [Python-checkins] r83177 - in python/branches/py3k:
	Doc/library/tkinter.rst Lib/tkinter/__main__.py Misc/NEWS
Message-ID: <20100727141636.1AB2AEEA95@mail.python.org>

Author: alexander.belopolsky
Date: Tue Jul 27 16:16:32 2010
New Revision: 83177

Log:
Issue #9384: python -m tkinter will now display a simple demo applet.

Added:
   python/branches/py3k/Lib/tkinter/__main__.py
Modified:
   python/branches/py3k/Doc/library/tkinter.rst
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Doc/library/tkinter.rst
==============================================================================
--- python/branches/py3k/Doc/library/tkinter.rst	(original)
+++ python/branches/py3k/Doc/library/tkinter.rst	Tue Jul 27 16:16:32 2010
@@ -9,7 +9,9 @@
 The :mod:`tkinter` package ("Tk interface") is the standard Python interface to
 the Tk GUI toolkit.  Both Tk and :mod:`tkinter` are available on most Unix
 platforms, as well as on Windows systems.  (Tk itself is not part of Python; it
-is maintained at ActiveState.)
+is maintained at ActiveState.) You can check that :mod:`tkinter` is properly
+installed on your system by running ``python -m tkinter`` from the command line;
+this should open a window demonstrating a simple Tk interface.
 
 .. seealso::
 

Added: python/branches/py3k/Lib/tkinter/__main__.py
==============================================================================
--- (empty file)
+++ python/branches/py3k/Lib/tkinter/__main__.py	Tue Jul 27 16:16:32 2010
@@ -0,0 +1,7 @@
+"""Main entry point"""
+
+import sys
+if sys.argv[0].endswith("__main__.py"):
+    sys.argv[0] = "python -m tkinter"
+from . import _test as main
+main()

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Tue Jul 27 16:16:32 2010
@@ -473,6 +473,8 @@
 Library
 -------
 
+- Issue #9384: python -m tkinter will now display a simple demo applet.
+
 - The default size of the re module's compiled regular expression cache has
   been increased from 100 to 500 and the cache replacement policy has changed
   from simply clearing the entire cache on overflow to randomly forgetting 20%

From python-checkins at python.org  Tue Jul 27 16:19:57 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Tue, 27 Jul 2010 16:19:57 +0200 (CEST)
Subject: [Python-checkins] r83178 - python/branches/release31-maint
Message-ID: <20100727141957.E1517EEAA2@mail.python.org>

Author: alexander.belopolsky
Date: Tue Jul 27 16:19:57 2010
New Revision: 83178

Log:
Blocked revisions 83177 via svnmerge

........
  r83177 | alexander.belopolsky | 2010-07-27 10:16:32 -0400 (Tue, 27 Jul 2010) | 1 line
  
  Issue #9384: python -m tkinter will now display a simple demo applet.
........


Modified:
   python/branches/release31-maint/   (props changed)

From python-checkins at python.org  Tue Jul 27 16:22:00 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Tue, 27 Jul 2010 16:22:00 +0200 (CEST)
Subject: [Python-checkins] r83179 - python/branches/release27-maint
Message-ID: <20100727142200.A1AD9EEAE4@mail.python.org>

Author: alexander.belopolsky
Date: Tue Jul 27 16:22:00 2010
New Revision: 83179

Log:
Blocked revisions 83157,83177 via svnmerge

........
  r83157 | alexander.belopolsky | 2010-07-25 22:36:41 -0400 (Sun, 25 Jul 2010) | 1 line
  
  Make python version of fromtimestamp behave more like C.
........
  r83177 | alexander.belopolsky | 2010-07-27 10:16:32 -0400 (Tue, 27 Jul 2010) | 1 line
  
  Issue #9384: python -m tkinter will now display a simple demo applet.
........


Modified:
   python/branches/release27-maint/   (props changed)

From python-checkins at python.org  Tue Jul 27 20:19:21 2010
From: python-checkins at python.org (georg.brandl)
Date: Tue, 27 Jul 2010 20:19:21 +0200 (CEST)
Subject: [Python-checkins] r83181 - in python/branches/py3k/Doc: Makefile
	conf.py
Message-ID: <20100727181921.404CBEEA49@mail.python.org>

Author: georg.brandl
Date: Tue Jul 27 20:19:21 2010
New Revision: 83181

Log:
Update Sphinx to 1.0.1.

Modified:
   python/branches/py3k/Doc/Makefile
   python/branches/py3k/Doc/conf.py

Modified: python/branches/py3k/Doc/Makefile
==============================================================================
--- python/branches/py3k/Doc/Makefile	(original)
+++ python/branches/py3k/Doc/Makefile	Tue Jul 27 20:19:21 2010
@@ -40,7 +40,7 @@
 checkout:
 	@if [ ! -d tools/sphinx ]; then \
 	  echo "Checking out Sphinx..."; \
-	  svn checkout $(SVNROOT)/external/Sphinx-1.0/sphinx tools/sphinx; \
+	  svn checkout $(SVNROOT)/external/Sphinx-1.0.1/sphinx tools/sphinx; \
 	fi
 	@if [ ! -d tools/docutils ]; then \
 	  echo "Checking out Docutils..."; \

Modified: python/branches/py3k/Doc/conf.py
==============================================================================
--- python/branches/py3k/Doc/conf.py	(original)
+++ python/branches/py3k/Doc/conf.py	Tue Jul 27 20:19:21 2010
@@ -152,7 +152,7 @@
 latex_appendices = ['glossary', 'about', 'license', 'copyright']
 
 # Get LaTeX to handle Unicode correctly
-latex_elements = {'inputenc': r'\usepackage[utf8x]{inputenc}'}
+latex_elements = {'inputenc': r'\usepackage[utf8x]{inputenc}', 'utf8extra': ''}
 
 # Options for the coverage checker
 # --------------------------------

From python-checkins at python.org  Tue Jul 27 23:20:15 2010
From: python-checkins at python.org (florent.xicluna)
Date: Tue, 27 Jul 2010 23:20:15 +0200 (CEST)
Subject: [Python-checkins] r83182 - in python/branches/py3k:
	Doc/library/binascii.rst Lib/email/base64mime.py
	Lib/email/message.py Lib/test/test_binascii.py
	Lib/test/test_struct.py Misc/NEWS Modules/binascii.c
Message-ID: <20100727212015.DEF1BD39B@mail.python.org>

Author: florent.xicluna
Date: Tue Jul 27 23:20:15 2010
New Revision: 83182

Log:
Issue #4770: Restrict binascii module to accept only bytes (as specified).
And fix the email package to encode to ASCII instead of ``raw-unicode-escape`` before ASCII-to-binary decoding.


Modified:
   python/branches/py3k/Doc/library/binascii.rst
   python/branches/py3k/Lib/email/base64mime.py
   python/branches/py3k/Lib/email/message.py
   python/branches/py3k/Lib/test/test_binascii.py
   python/branches/py3k/Lib/test/test_struct.py
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/Modules/binascii.c

Modified: python/branches/py3k/Doc/library/binascii.rst
==============================================================================
--- python/branches/py3k/Doc/library/binascii.rst	(original)
+++ python/branches/py3k/Doc/library/binascii.rst	Tue Jul 27 23:20:15 2010
@@ -18,6 +18,11 @@
 low-level functions written in C for greater speed that are used by the
 higher-level modules.
 
+.. note::
+
+   Encoding and decoding functions do not accept Unicode strings.  Only bytestring
+   and bytearray objects can be processed.
+
 The :mod:`binascii` module defines the following functions:
 
 
@@ -54,6 +59,9 @@
    data. More than one line may be passed at a time. If the optional argument
    *header* is present and true, underscores will be decoded as spaces.
 
+   .. versionchanged:: 3.2
+      accept only bytestring or bytearray object as input.
+
 
 .. function:: b2a_qp(data, quotetabs=False, istext=True, header=False)
 
@@ -83,6 +91,9 @@
    decompressed data, unless data input data ends in an orphaned repeat indicator,
    in which case the :exc:`Incomplete` exception is raised.
 
+   .. versionchanged:: 3.2
+      accept only bytestring or bytearray object as input.
+
 
 .. function:: rlecode_hqx(data)
 
@@ -139,6 +150,9 @@
    of hexadecimal digits (which can be upper or lower case), otherwise a
    :exc:`TypeError` is raised.
 
+   .. versionchanged:: 3.2
+      accept only bytestring or bytearray object as input.
+
 
 .. exception:: Error
 
@@ -164,4 +178,3 @@
 
    Module :mod:`quopri`
       Support for quoted-printable encoding used in MIME email messages.
-

Modified: python/branches/py3k/Lib/email/base64mime.py
==============================================================================
--- python/branches/py3k/Lib/email/base64mime.py	(original)
+++ python/branches/py3k/Lib/email/base64mime.py	Tue Jul 27 23:20:15 2010
@@ -74,12 +74,12 @@
 
 
 def body_encode(s, maxlinelen=76, eol=NL):
-    """Encode a string with base64.
+    r"""Encode a string with base64.
 
     Each line will be wrapped at, at most, maxlinelen characters (defaults to
     76 characters).
 
-    Each line of encoded text will end with eol, which defaults to "\\n".  Set
+    Each line of encoded text will end with eol, which defaults to "\n".  Set
     this to "\r\n" if you will be using the result of this function directly
     in an email.
     """

Modified: python/branches/py3k/Lib/email/message.py
==============================================================================
--- python/branches/py3k/Lib/email/message.py	(original)
+++ python/branches/py3k/Lib/email/message.py	Tue Jul 27 23:20:15 2010
@@ -198,17 +198,19 @@
             return None
         cte = self.get('content-transfer-encoding', '').lower()
         if cte == 'quoted-printable':
+            if isinstance(payload, str):
+                payload = payload.encode('ascii')
             return utils._qdecode(payload)
         elif cte == 'base64':
             try:
                 if isinstance(payload, str):
-                    payload = payload.encode('raw-unicode-escape')
+                    payload = payload.encode('ascii')
                 return base64.b64decode(payload)
             except binascii.Error:
                 # Incorrect padding
                 pass
         elif cte in ('x-uuencode', 'uuencode', 'uue', 'x-uue'):
-            in_file = BytesIO(payload.encode('raw-unicode-escape'))
+            in_file = BytesIO(payload.encode('ascii'))
             out_file = BytesIO()
             try:
                 uu.decode(in_file, out_file, quiet=True)

Modified: python/branches/py3k/Lib/test/test_binascii.py
==============================================================================
--- python/branches/py3k/Lib/test/test_binascii.py	(original)
+++ python/branches/py3k/Lib/test/test_binascii.py	Tue Jul 27 23:20:15 2010
@@ -55,7 +55,7 @@
                              "{!r} != {!r}".format(fb, fa, res, raw))
             self.assertIsInstance(res, bytes)
             self.assertIsInstance(a, bytes)
-            self.assertLess(max(c for c in a), 128)
+            self.assertLess(max(a), 128)
         self.assertIsInstance(binascii.crc_hqx(raw, 0), int)
         self.assertIsInstance(binascii.crc32(raw), int)
 
@@ -167,7 +167,7 @@
     def test_qp(self):
         # A test for SF bug 534347 (segfaults without the proper fix)
         try:
-            binascii.a2b_qp("", **{1:1})
+            binascii.a2b_qp(b"", **{1:1})
         except TypeError:
             pass
         else:
@@ -179,12 +179,10 @@
         self.assertEqual(binascii.a2b_qp(b"=00\r\n=00"), b"\x00\r\n\x00")
         self.assertEqual(
             binascii.b2a_qp(b"\xff\r\n\xff\n\xff"),
-            b"=FF\r\n=FF\r\n=FF"
-        )
+            b"=FF\r\n=FF\r\n=FF")
         self.assertEqual(
             binascii.b2a_qp(b"0"*75+b"\xff\r\n\xff\r\n\xff"),
-            b"0"*75+b"=\r\n=FF\r\n=FF\r\n=FF"
-        )
+            b"0"*75+b"=\r\n=FF\r\n=FF\r\n=FF")
 
         self.assertEqual(binascii.b2a_qp(b'\0\n'), b'=00\n')
         self.assertEqual(binascii.b2a_qp(b'\0\n', quotetabs=True), b'=00\n')
@@ -210,13 +208,15 @@
             except Exception as err:
                 self.fail("{}({!r}) raises {!r}".format(func, empty, err))
 
-    def test_no_binary_strings(self):
-        # b2a_ must not accept strings
-        for f in (binascii.b2a_uu, binascii.b2a_base64,
-                  binascii.b2a_hqx, binascii.b2a_qp,
-                  binascii.hexlify, binascii.rlecode_hqx,
-                  binascii.crc_hqx, binascii.crc32):
-            self.assertRaises(TypeError, f, "test")
+    def test_unicode_strings(self):
+        # Unicode strings are not accepted.
+        for func in all_functions:
+            try:
+                self.assertRaises(TypeError, getattr(binascii, func), "test")
+            except Exception as err:
+                self.fail('{}("test") raises {!r}'.format(func, err))
+        # crc_hqx needs 2 arguments
+        self.assertRaises(TypeError, binascii.crc_hqx, "test", 0)
 
 
 class ArrayBinASCIITest(BinASCIITest):

Modified: python/branches/py3k/Lib/test/test_struct.py
==============================================================================
--- python/branches/py3k/Lib/test/test_struct.py	(original)
+++ python/branches/py3k/Lib/test/test_struct.py	Tue Jul 27 23:20:15 2010
@@ -213,6 +213,7 @@
                     expected = '%x' % expected
                     if len(expected) & 1:
                         expected = "0" + expected
+                    expected = expected.encode('ascii')
                     expected = unhexlify(expected)
                     expected = (b"\x00" * (self.bytesize - len(expected)) +
                                 expected)

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Tue Jul 27 23:20:15 2010
@@ -473,6 +473,10 @@
 Library
 -------
 
+- Issue #4770: Restrict binascii module to accept only bytes (as specified).
+  And fix the email package to encode to ASCII instead of
+  ``raw-unicode-escape`` before ASCII-to-binary decoding.
+
 - Issue #9384: python -m tkinter will now display a simple demo applet.
 
 - The default size of the re module's compiled regular expression cache has

Modified: python/branches/py3k/Modules/binascii.c
==============================================================================
--- python/branches/py3k/Modules/binascii.c	(original)
+++ python/branches/py3k/Modules/binascii.c	Tue Jul 27 23:20:15 2010
@@ -546,7 +546,7 @@
     Py_ssize_t len;
     int done = 0;
 
-    if ( !PyArg_ParseTuple(args, "s*:a2b_hqx", &pascii) )
+    if ( !PyArg_ParseTuple(args, "y*:a2b_hqx", &pascii) )
         return NULL;
     ascii_data = pascii.buf;
     len = pascii.len;
@@ -750,7 +750,7 @@
     PyObject *rv;
     Py_ssize_t in_len, out_len, out_len_left;
 
-    if ( !PyArg_ParseTuple(args, "s*:rledecode_hqx", &pin) )
+    if ( !PyArg_ParseTuple(args, "y*:rledecode_hqx", &pin) )
         return NULL;
     in_data = pin.buf;
     in_len = pin.len;
@@ -1121,7 +1121,7 @@
     char* retbuf;
     Py_ssize_t i, j;
 
-    if (!PyArg_ParseTuple(args, "s*:a2b_hex", &parg))
+    if (!PyArg_ParseTuple(args, "y*:a2b_hex", &parg))
         return NULL;
     argbuf = parg.buf;
     arglen = parg.len;
@@ -1199,7 +1199,7 @@
     static char *kwlist[] = {"data", "header", NULL};
     int header = 0;
 
-    if (!PyArg_ParseTupleAndKeywords(args, kwargs, "s*|i", kwlist, &pdata,
+    if (!PyArg_ParseTupleAndKeywords(args, kwargs, "y*|i", kwlist, &pdata,
           &header))
         return NULL;
     data = pdata.buf;

From python-checkins at python.org  Wed Jul 28 00:03:33 2010
From: python-checkins at python.org (ezio.melotti)
Date: Wed, 28 Jul 2010 00:03:33 +0200 (CEST)
Subject: [Python-checkins] r83183 -
	python/branches/py3k/Lib/test/test_asyncore.py
Message-ID: <20100727220333.8F4D8EE99B@mail.python.org>

Author: ezio.melotti
Date: Wed Jul 28 00:03:33 2010
New Revision: 83183

Log:
Use proper skips and assert* methods in test_asyncore.

Modified:
   python/branches/py3k/Lib/test/test_asyncore.py

Modified: python/branches/py3k/Lib/test/test_asyncore.py
==============================================================================
--- python/branches/py3k/Lib/test/test_asyncore.py	(original)
+++ python/branches/py3k/Lib/test/test_asyncore.py	Wed Jul 28 00:03:33 2010
@@ -118,65 +118,65 @@
     # http://mail.python.org/pipermail/python-list/2001-October/109973.html)
     # These constants should be present as long as poll is available
 
-    if hasattr(select, 'poll'):
-        def test_readwrite(self):
-            # Check that correct methods are called by readwrite()
+    @unittest.skipUnless(hasattr(select, 'poll'), 'select.poll required')
+    def test_readwrite(self):
+        # Check that correct methods are called by readwrite()
+
+        attributes = ('read', 'expt', 'write', 'closed', 'error_handled')
+
+        expected = (
+            (select.POLLIN, 'read'),
+            (select.POLLPRI, 'expt'),
+            (select.POLLOUT, 'write'),
+            (select.POLLERR, 'closed'),
+            (select.POLLHUP, 'closed'),
+            (select.POLLNVAL, 'closed'),
+            )
+
+        class testobj:
+            def __init__(self):
+                self.read = False
+                self.write = False
+                self.closed = False
+                self.expt = False
+                self.error_handled = False
+
+            def handle_read_event(self):
+                self.read = True
 
-            attributes = ('read', 'expt', 'write', 'closed', 'error_handled')
+            def handle_write_event(self):
+                self.write = True
 
-            expected = (
-                (select.POLLIN, 'read'),
-                (select.POLLPRI, 'expt'),
-                (select.POLLOUT, 'write'),
-                (select.POLLERR, 'closed'),
-                (select.POLLHUP, 'closed'),
-                (select.POLLNVAL, 'closed'),
-                )
-
-            class testobj:
-                def __init__(self):
-                    self.read = False
-                    self.write = False
-                    self.closed = False
-                    self.expt = False
-                    self.error_handled = False
-
-                def handle_read_event(self):
-                    self.read = True
-
-                def handle_write_event(self):
-                    self.write = True
-
-                def handle_close(self):
-                    self.closed = True
-
-                def handle_expt_event(self):
-                    self.expt = True
-
-                def handle_error(self):
-                    self.error_handled = True
-
-            for flag, expectedattr in expected:
-                tobj = testobj()
-                self.assertEqual(getattr(tobj, expectedattr), False)
-                asyncore.readwrite(tobj, flag)
-
-                # Only the attribute modified by the routine we expect to be
-                # called should be True.
-                for attr in attributes:
-                    self.assertEqual(getattr(tobj, attr), attr==expectedattr)
-
-                # check that ExitNow exceptions in the object handler method
-                # bubbles all the way up through asyncore readwrite call
-                tr1 = exitingdummy()
-                self.assertRaises(asyncore.ExitNow, asyncore.readwrite, tr1, flag)
-
-                # check that an exception other than ExitNow in the object handler
-                # method causes the handle_error method to get called
-                tr2 = crashingdummy()
-                self.assertEqual(tr2.error_handled, False)
-                asyncore.readwrite(tr2, flag)
-                self.assertEqual(tr2.error_handled, True)
+            def handle_close(self):
+                self.closed = True
+
+            def handle_expt_event(self):
+                self.expt = True
+
+            def handle_error(self):
+                self.error_handled = True
+
+        for flag, expectedattr in expected:
+            tobj = testobj()
+            self.assertEqual(getattr(tobj, expectedattr), False)
+            asyncore.readwrite(tobj, flag)
+
+            # Only the attribute modified by the routine we expect to be
+            # called should be True.
+            for attr in attributes:
+                self.assertEqual(getattr(tobj, attr), attr==expectedattr)
+
+            # check that ExitNow exceptions in the object handler method
+            # bubbles all the way up through asyncore readwrite call
+            tr1 = exitingdummy()
+            self.assertRaises(asyncore.ExitNow, asyncore.readwrite, tr1, flag)
+
+            # check that an exception other than ExitNow in the object handler
+            # method causes the handle_error method to get called
+            tr2 = crashingdummy()
+            self.assertEqual(tr2.error_handled, False)
+            asyncore.readwrite(tr2, flag)
+            self.assertEqual(tr2.error_handled, True)
 
     def test_closeall(self):
         self.closeall_check(False)
@@ -259,7 +259,7 @@
             sys.stderr = stderr
 
         lines = fp.getvalue().splitlines()
-        self.assertEquals(lines, ['log: %s' % l1, 'log: %s' % l2])
+        self.assertEqual(lines, ['log: %s' % l1, 'log: %s' % l2])
 
     def test_log_info(self):
         d = asyncore.dispatcher()
@@ -281,7 +281,7 @@
         lines = fp.getvalue().splitlines()
         expected = ['EGGS: %s' % l1, 'info: %s' % l2, 'SPAM: %s' % l3]
 
-        self.assertEquals(lines, expected)
+        self.assertEqual(lines, expected)
 
     def test_unhandled(self):
         d = asyncore.dispatcher()
@@ -306,7 +306,7 @@
                     'warning: unhandled write event',
                     'warning: unhandled connect event',
                     'warning: unhandled accept event']
-        self.assertEquals(lines, expected)
+        self.assertEqual(lines, expected)
 
     def test_issue_8594(self):
         # XXX - this test is supposed to be removed in next major Python
@@ -322,7 +322,7 @@
             warnings.simplefilter("always")
             family = d.family
             self.assertEqual(family, socket.AF_INET)
-            self.assertTrue(len(w) == 1)
+            self.assertEqual(len(w), 1)
             self.assertTrue(issubclass(w[0].category, DeprecationWarning))
 
     def test_strerror(self):
@@ -331,7 +331,7 @@
         if hasattr(os, 'strerror'):
             self.assertEqual(err, os.strerror(errno.EPERM))
         err = asyncore._strerror(-1)
-        self.assertTrue("unknown error" in err.lower())
+        self.assertIn("unknown error", err.lower())
 
 
 class dispatcherwithsend_noread(asyncore.dispatcher_with_send):
@@ -394,38 +394,39 @@
 class DispatcherWithSendTests_UsePoll(DispatcherWithSendTests):
     usepoll = True
 
-if hasattr(asyncore, 'file_wrapper'):
-    class FileWrapperTest(unittest.TestCase):
-        def setUp(self):
-            self.d = b"It's not dead, it's sleeping!"
-            open(TESTFN, 'wb').write(self.d)
-
-        def tearDown(self):
-            unlink(TESTFN)
-
-        def test_recv(self):
-            fd = os.open(TESTFN, os.O_RDONLY)
-            w = asyncore.file_wrapper(fd)
-            os.close(fd)
-
-            self.assertNotEqual(w.fd, fd)
-            self.assertNotEqual(w.fileno(), fd)
-            self.assertEqual(w.recv(13), b"It's not dead")
-            self.assertEqual(w.read(6), b", it's")
-            w.close()
-            self.assertRaises(OSError, w.read, 1)
-
-        def test_send(self):
-            d1 = b"Come again?"
-            d2 = b"I want to buy some cheese."
-            fd = os.open(TESTFN, os.O_WRONLY | os.O_APPEND)
-            w = asyncore.file_wrapper(fd)
-            os.close(fd)
-
-            w.write(d1)
-            w.send(d2)
-            w.close()
-            self.assertEqual(open(TESTFN, 'rb').read(), self.d + d1 + d2)
+ at unittest.skipUnless(hasattr(asyncore, 'file_wrapper'),
+                     'asyncore.file_wrapper required')
+class FileWrapperTest(unittest.TestCase):
+    def setUp(self):
+        self.d = b"It's not dead, it's sleeping!"
+        open(TESTFN, 'wb').write(self.d)
+
+    def tearDown(self):
+        unlink(TESTFN)
+
+    def test_recv(self):
+        fd = os.open(TESTFN, os.O_RDONLY)
+        w = asyncore.file_wrapper(fd)
+        os.close(fd)
+
+        self.assertNotEqual(w.fd, fd)
+        self.assertNotEqual(w.fileno(), fd)
+        self.assertEqual(w.recv(13), b"It's not dead")
+        self.assertEqual(w.read(6), b", it's")
+        w.close()
+        self.assertRaises(OSError, w.read, 1)
+
+    def test_send(self):
+        d1 = b"Come again?"
+        d2 = b"I want to buy some cheese."
+        fd = os.open(TESTFN, os.O_WRONLY | os.O_APPEND)
+        w = asyncore.file_wrapper(fd)
+        os.close(fd)
+
+        w.write(d1)
+        w.send(d2)
+        w.close()
+        self.assertEqual(open(TESTFN, 'rb').read(), self.d + d1 + d2)
 
 
 class BaseTestHandler(asyncore.dispatcher):

From python-checkins at python.org  Wed Jul 28 00:08:27 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 28 Jul 2010 00:08:27 +0200 (CEST)
Subject: [Python-checkins] r83184 - in python/branches/py3k: Misc/ACKS
	Objects/object.c
Message-ID: <20100727220827.EA066C402@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 28 00:08:27 2010
New Revision: 83184

Log:
Issue #9294: remove dead code in Objects/object.c.  Patch by Grant Limberg.



Modified:
   python/branches/py3k/Misc/ACKS
   python/branches/py3k/Objects/object.c

Modified: python/branches/py3k/Misc/ACKS
==============================================================================
--- python/branches/py3k/Misc/ACKS	(original)
+++ python/branches/py3k/Misc/ACKS	Wed Jul 28 00:08:27 2010
@@ -479,6 +479,7 @@
 Ross Light
 Shawn Ligocki
 Martin Ligr
+Grant Limberg
 Christopher Lindblad
 Bjorn Lindqvist
 Per Lindqvist

Modified: python/branches/py3k/Objects/object.c
==============================================================================
--- python/branches/py3k/Objects/object.c	(original)
+++ python/branches/py3k/Objects/object.c	Wed Jul 28 00:08:27 2010
@@ -258,15 +258,10 @@
     return PyObject_INIT_VAR(op, tp, nitems);
 }
 
-/* Implementation of PyObject_Print with recursion checking */
-static int
-internal_print(PyObject *op, FILE *fp, int flags, int nesting)
+int
+PyObject_Print(PyObject *op, FILE *fp, int flags)
 {
     int ret = 0;
-    if (nesting > 10) {
-        PyErr_SetString(PyExc_RuntimeError, "print recursion");
-        return -1;
-    }
     if (PyErr_CheckSignals())
         return -1;
 #ifdef USE_STACKCHECK
@@ -333,12 +328,6 @@
     return ret;
 }
 
-int
-PyObject_Print(PyObject *op, FILE *fp, int flags)
-{
-    return internal_print(op, fp, flags, 0);
-}
-
 /* For debugging convenience.  Set a breakpoint here and call it from your DLL */
 void
 _Py_BreakPoint(void)

From python-checkins at python.org  Wed Jul 28 00:09:59 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Wed, 28 Jul 2010 00:09:59 +0200 (CEST)
Subject: [Python-checkins] r83185 - in python/branches/release31-maint:
	Misc/ACKS Objects/object.c
Message-ID: <20100727220959.BDB3CEEA91@mail.python.org>

Author: antoine.pitrou
Date: Wed Jul 28 00:09:59 2010
New Revision: 83185

Log:
Merged revisions 83184 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83184 | antoine.pitrou | 2010-07-28 00:08:27 +0200 (mer., 28 juil. 2010) | 3 lines
  
  Issue #9294: remove dead code in Objects/object.c.  Patch by Grant Limberg.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Misc/ACKS
   python/branches/release31-maint/Objects/object.c

Modified: python/branches/release31-maint/Misc/ACKS
==============================================================================
--- python/branches/release31-maint/Misc/ACKS	(original)
+++ python/branches/release31-maint/Misc/ACKS	Wed Jul 28 00:09:59 2010
@@ -454,6 +454,7 @@
 Ross Light
 Shawn Ligocki
 Martin Ligr
+Grant Limberg
 Christopher Lindblad
 Bjorn Lindqvist
 Per Lindqvist

Modified: python/branches/release31-maint/Objects/object.c
==============================================================================
--- python/branches/release31-maint/Objects/object.c	(original)
+++ python/branches/release31-maint/Objects/object.c	Wed Jul 28 00:09:59 2010
@@ -258,15 +258,10 @@
     return PyObject_INIT_VAR(op, tp, nitems);
 }
 
-/* Implementation of PyObject_Print with recursion checking */
-static int
-internal_print(PyObject *op, FILE *fp, int flags, int nesting)
+int
+PyObject_Print(PyObject *op, FILE *fp, int flags)
 {
     int ret = 0;
-    if (nesting > 10) {
-        PyErr_SetString(PyExc_RuntimeError, "print recursion");
-        return -1;
-    }
     if (PyErr_CheckSignals())
         return -1;
 #ifdef USE_STACKCHECK
@@ -333,12 +328,6 @@
     return ret;
 }
 
-int
-PyObject_Print(PyObject *op, FILE *fp, int flags)
-{
-    return internal_print(op, fp, flags, 0);
-}
-
 /* For debugging convenience.  Set a breakpoint here and call it from your DLL */
 void
 _Py_BreakPoint(void)

From python-checkins at python.org  Wed Jul 28 00:24:13 2010
From: python-checkins at python.org (ezio.melotti)
Date: Wed, 28 Jul 2010 00:24:13 +0200 (CEST)
Subject: [Python-checkins] r83186 -
	python/branches/py3k/Lib/test/test_asyncore.py
Message-ID: <20100727222413.94498EEA9F@mail.python.org>

Author: ezio.melotti
Date: Wed Jul 28 00:24:13 2010
New Revision: 83186

Log:
With skipUnless there is no need to add test classes conditionally.

Modified:
   python/branches/py3k/Lib/test/test_asyncore.py

Modified: python/branches/py3k/Lib/test/test_asyncore.py
==============================================================================
--- python/branches/py3k/Lib/test/test_asyncore.py	(original)
+++ python/branches/py3k/Lib/test/test_asyncore.py	Wed Jul 28 00:24:13 2010
@@ -693,18 +693,15 @@
 class TestAPI_UseSelect(BaseTestAPI):
     use_poll = False
 
+ at unittest.skipUnless(hasattr(select, 'poll'), 'select.poll required')
 class TestAPI_UsePoll(BaseTestAPI):
     use_poll = True
 
 
 def test_main():
     tests = [HelperFunctionTests, DispatcherTests, DispatcherWithSendTests,
-             DispatcherWithSendTests_UsePoll, TestAPI_UseSelect]
-    if hasattr(asyncore, 'file_wrapper'):
-        tests.append(FileWrapperTest)
-    if hasattr(select, 'poll'):
-        tests.append(TestAPI_UsePoll)
-
+             DispatcherWithSendTests_UsePoll, TestAPI_UseSelect,
+             TestAPI_UsePoll, FileWrapperTest]
     run_unittest(*tests)
 
 if __name__ == "__main__":

From florent.xicluna at m4x.org  Wed Jul 28 00:28:43 2010
From: florent.xicluna at m4x.org (Florent XICLUNA)
Date: Wed, 28 Jul 2010 00:28:43 +0200
Subject: [Python-checkins] r79102 not in release31-maint
In-Reply-To: <AANLkTins4qd_z=wsVreS6-UE0SZx5LYAws4VhRcJBsDG@mail.gmail.com>
References: <AANLkTins4qd_z=wsVreS6-UE0SZx5LYAws4VhRcJBsDG@mail.gmail.com>
Message-ID: <AANLkTin5vzJJYrVv4VKuFzPi_0-1bJL3zXZodA=NBXeJ@mail.gmail.com>

This is a cosmetic patch, it is not strictly a bug fix.
So I didn't see any reason to backport it.

But if it helps you for other merges, you can port it. It seems quite safe.

Regards,

-- 
Florent

2010/7/25 Alexander Belopolsky

> Does anyone know why r79102 is not in release31-maint?  It looks like
> this prevents me from merging r83140.
>
> ------------------------------------------------------------------------
> r79102 | florent.xicluna | 2010-03-19 15:00:44 -0400 (Fri, 19 Mar
> 2010) | 9 lines
>
> Merged revisions 79100 via svnmerge from
> svn+ssh://pythondev at svn.python.org/python/trunk
>
> ........
>  r79100 | florent.xicluna | 2010-03-19 19:34:55 +0100 (ven, 19 mar
> 2010) | 2 lines
>
>  Various tests cleanup: check_warnings/check_py3k_warnings,
> unittest.assert* and setUp/tearDown.
> ........
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-checkins/attachments/20100728/434c3471/attachment-0001.html>

From python-checkins at python.org  Wed Jul 28 00:48:26 2010
From: python-checkins at python.org (florent.xicluna)
Date: Wed, 28 Jul 2010 00:48:26 +0200 (CEST)
Subject: [Python-checkins] r83187 -
	python/branches/release27-maint/Lib/test/test_threadsignals.py
Message-ID: <20100727224826.E77BCEEAA9@mail.python.org>

Author: florent.xicluna
Date: Wed Jul 28 00:48:26 2010
New Revision: 83187

Log:
Issue #7825: fix transient refleak in test_threadsignals.


Modified:
   python/branches/release27-maint/Lib/test/test_threadsignals.py

Modified: python/branches/release27-maint/Lib/test/test_threadsignals.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_threadsignals.py	(original)
+++ python/branches/release27-maint/Lib/test/test_threadsignals.py	Wed Jul 28 00:48:26 2010
@@ -4,7 +4,7 @@
 import signal
 import os
 import sys
-from test.test_support import run_unittest, import_module
+from test.test_support import run_unittest, import_module, reap_threads
 thread = import_module('thread')
 
 if sys.platform[:3] in ('win', 'os2') or sys.platform=='riscos':
@@ -39,6 +39,7 @@
        wait for it to finish. Check that we got both signals
        and that they were run by the main thread.
     """
+    @reap_threads
     def test_signals(self):
         signalled_all.acquire()
         self.spawnSignallingThread()

From python-checkins at python.org  Wed Jul 28 01:02:38 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Wed, 28 Jul 2010 01:02:38 +0200 (CEST)
Subject: [Python-checkins] r83188 - in python/branches/py3k: Lib/pickle.py
	Misc/NEWS
Message-ID: <20100727230238.E0EDFEEAA7@mail.python.org>

Author: alexander.belopolsky
Date: Wed Jul 28 01:02:38 2010
New Revision: 83188

Log:
Issue #9378: python -m pickle <pickle file> will now load and display
the first object in the pickle file.


Modified:
   python/branches/py3k/Lib/pickle.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/pickle.py
==============================================================================
--- python/branches/py3k/Lib/pickle.py	(original)
+++ python/branches/py3k/Lib/pickle.py	Wed Jul 28 01:02:38 2010
@@ -1322,4 +1322,26 @@
     return doctest.testmod()
 
 if __name__ == "__main__":
-    _test()
+    import sys, argparse
+    parser = argparse.ArgumentParser(
+        description='display contents of the pickle files')
+    parser.add_argument(
+        'pickle_file', type=argparse.FileType('br'),
+        nargs='*', help='the pickle file')
+    parser.add_argument(
+        '-t', '--test', action='store_true',
+        help='run self-test suite')
+    parser.add_argument(
+        '-v', action='store_true',
+        help='run verbosely; only affects self-test run')
+    args = parser.parse_args()
+    if args.test:
+        _test()
+    else:
+        if not args.pickle_file:
+            parser.print_help()
+        else:
+            import pprint
+            for f in args.pickle_file:
+                obj = load(f)
+                pprint.pprint(obj)

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Wed Jul 28 01:02:38 2010
@@ -473,6 +473,9 @@
 Library
 -------
 
+- Issue #9378: python -m pickle <pickle file> will now load and
+  display the first object in the pickle file.
+
 - Issue #4770: Restrict binascii module to accept only bytes (as specified).
   And fix the email package to encode to ASCII instead of
   ``raw-unicode-escape`` before ASCII-to-binary decoding.

From python-checkins at python.org  Wed Jul 28 01:04:30 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Wed, 28 Jul 2010 01:04:30 +0200 (CEST)
Subject: [Python-checkins] r83189 - python/branches/release31-maint
Message-ID: <20100727230430.C9308EEA6C@mail.python.org>

Author: alexander.belopolsky
Date: Wed Jul 28 01:04:30 2010
New Revision: 83189

Log:
Blocked revisions 83188 via svnmerge

........
  r83188 | alexander.belopolsky | 2010-07-27 19:02:38 -0400 (Tue, 27 Jul 2010) | 3 lines
  
  Issue #9378: python -m pickle <pickle file> will now load and display
  the first object in the pickle file.
........


Modified:
   python/branches/release31-maint/   (props changed)

From python-checkins at python.org  Wed Jul 28 01:05:31 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Wed, 28 Jul 2010 01:05:31 +0200 (CEST)
Subject: [Python-checkins] r83190 - python/branches/release27-maint
Message-ID: <20100727230531.D8777EEA86@mail.python.org>

Author: alexander.belopolsky
Date: Wed Jul 28 01:05:31 2010
New Revision: 83190

Log:
Blocked revisions 83188 via svnmerge

........
  r83188 | alexander.belopolsky | 2010-07-27 19:02:38 -0400 (Tue, 27 Jul 2010) | 3 lines
  
  Issue #9378: python -m pickle <pickle file> will now load and display
  the first object in the pickle file.
........


Modified:
   python/branches/release27-maint/   (props changed)

From python-checkins at python.org  Wed Jul 28 01:36:41 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 28 Jul 2010 01:36:41 +0200 (CEST)
Subject: [Python-checkins] r83191 - in python/branches/py3k/Lib/ctypes/test:
	test_arrays.py test_bitfields.py test_buffers.py
	test_bytes.py test_cast.py test_cfuncs.py test_errno.py
	test_internals.py test_keeprefs.py test_libc.py
	test_objects.py test_parameters.py test_prototypes.py
	test_python_api.py test_random_things.py test_repr.py
	test_returnfuncptrs.py test_stringptr.py test_strings.py
	test_structures.py
Message-ID: <20100727233641.7EAA1EEA9E@mail.python.org>

Author: victor.stinner
Date: Wed Jul 28 01:36:41 2010
New Revision: 83191

Log:
Fix ctypes tests to avoid implicit bytes-unicode conversion


Modified:
   python/branches/py3k/Lib/ctypes/test/test_arrays.py
   python/branches/py3k/Lib/ctypes/test/test_bitfields.py
   python/branches/py3k/Lib/ctypes/test/test_buffers.py
   python/branches/py3k/Lib/ctypes/test/test_bytes.py
   python/branches/py3k/Lib/ctypes/test/test_cast.py
   python/branches/py3k/Lib/ctypes/test/test_cfuncs.py
   python/branches/py3k/Lib/ctypes/test/test_errno.py
   python/branches/py3k/Lib/ctypes/test/test_internals.py
   python/branches/py3k/Lib/ctypes/test/test_keeprefs.py
   python/branches/py3k/Lib/ctypes/test/test_libc.py
   python/branches/py3k/Lib/ctypes/test/test_objects.py
   python/branches/py3k/Lib/ctypes/test/test_parameters.py
   python/branches/py3k/Lib/ctypes/test/test_prototypes.py
   python/branches/py3k/Lib/ctypes/test/test_python_api.py
   python/branches/py3k/Lib/ctypes/test/test_random_things.py
   python/branches/py3k/Lib/ctypes/test/test_repr.py
   python/branches/py3k/Lib/ctypes/test/test_returnfuncptrs.py
   python/branches/py3k/Lib/ctypes/test/test_stringptr.py
   python/branches/py3k/Lib/ctypes/test/test_strings.py
   python/branches/py3k/Lib/ctypes/test/test_structures.py

Modified: python/branches/py3k/Lib/ctypes/test/test_arrays.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_arrays.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_arrays.py	Wed Jul 28 01:36:41 2010
@@ -42,7 +42,7 @@
 
         CharArray = ARRAY(c_char, 3)
 
-        ca = CharArray("a", "b", "c")
+        ca = CharArray(b"a", b"b", b"c")
 
         # Should this work? It doesn't:
         # CharArray("abc")
@@ -89,7 +89,7 @@
 
     def test_from_address(self):
         # Failed with 0.9.8, reported by JUrner
-        p = create_string_buffer("foo")
+        p = create_string_buffer(b"foo")
         sz = (c_char * 3).from_address(addressof(p))
         self.assertEqual(sz[:], b"foo")
         self.assertEqual(sz[::], b"foo")

Modified: python/branches/py3k/Lib/ctypes/test/test_bitfields.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_bitfields.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_bitfields.py	Wed Jul 28 01:36:41 2010
@@ -37,14 +37,14 @@
             for name in "ABCDEFGHI":
                 b = BITS()
                 setattr(b, name, i)
-                self.assertEqual((name, i, getattr(b, name)), (name, i, func(byref(b), name)))
+                self.assertEqual(getattr(b, name), func(byref(b), name.encode('ascii')))
 
     def test_shorts(self):
         for i in range(256):
             for name in "MNOPQRS":
                 b = BITS()
                 setattr(b, name, i)
-                self.assertEqual((name, i, getattr(b, name)), (name, i, func(byref(b), name)))
+                self.assertEqual(getattr(b, name), func(byref(b), name.encode('ascii')))
 
 signed_int_types = (c_byte, c_short, c_int, c_long, c_longlong)
 unsigned_int_types = (c_ubyte, c_ushort, c_uint, c_ulong, c_ulonglong)

Modified: python/branches/py3k/Lib/ctypes/test/test_buffers.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_buffers.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_buffers.py	Wed Jul 28 01:36:41 2010
@@ -9,7 +9,7 @@
         self.assertEqual(sizeof(b), 32 * sizeof(c_char))
         self.assertTrue(type(b[0]) is bytes)
 
-        b = create_string_buffer("abc")
+        b = create_string_buffer(b"abc")
         self.assertEqual(len(b), 4) # trailing nul char
         self.assertEqual(sizeof(b), 4 * sizeof(c_char))
         self.assertTrue(type(b[0]) is bytes)

Modified: python/branches/py3k/Lib/ctypes/test/test_bytes.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_bytes.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_bytes.py	Wed Jul 28 01:36:41 2010
@@ -11,10 +11,10 @@
         (c_char * 3)(b"a", b"b", b"c")
 
     def test_c_wchar(self):
-        x = c_wchar(b"x")
-        x.value = b"y"
-        c_wchar.from_param(b"x")
-        (c_wchar * 3)(b"a", b"b", b"c")
+        x = c_wchar("x")
+        x.value = "y"
+        c_wchar.from_param("x")
+        (c_wchar * 3)("a", "b", "c")
 
     def test_c_char_p(self):
         c_char_p("foo bar")
@@ -37,8 +37,8 @@
         class X(Structure):
             _fields_ = [("a", c_wchar * 3)]
 
-        X("abc")
-        x = X(b"abc")
+        X(b"abc")
+        x = X("abc")
         self.assertEqual(x.a, "abc")
         self.assertEqual(type(x.a), str)
 

Modified: python/branches/py3k/Lib/ctypes/test/test_cast.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_cast.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_cast.py	Wed Jul 28 01:36:41 2010
@@ -33,17 +33,17 @@
     def test_p2a_objects(self):
         array = (c_char_p * 5)()
         self.assertEqual(array._objects, None)
-        array[0] = "foo bar"
+        array[0] = b"foo bar"
         self.assertEqual(array._objects, {'0': b"foo bar"})
 
         p = cast(array, POINTER(c_char_p))
         # array and p share a common _objects attribute
         self.assertTrue(p._objects is array._objects)
         self.assertEqual(array._objects, {'0': b"foo bar", id(array): array})
-        p[0] = "spam spam"
+        p[0] = b"spam spam"
         self.assertEqual(p._objects, {'0': b"spam spam", id(array): array})
         self.assertTrue(array._objects is p._objects)
-        p[1] = "foo bar"
+        p[1] = b"foo bar"
         self.assertEqual(p._objects, {'1': b'foo bar', '0': b"spam spam", id(array): array})
         self.assertTrue(array._objects is p._objects)
 
@@ -71,7 +71,7 @@
 
     def test_char_p(self):
         # This didn't work: bad argument to internal function
-        s = c_char_p("hiho")
+        s = c_char_p(b"hiho")
         self.assertEqual(cast(cast(s, c_void_p), c_char_p).value,
                              b"hiho")
 

Modified: python/branches/py3k/Lib/ctypes/test/test_cfuncs.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_cfuncs.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_cfuncs.py	Wed Jul 28 01:36:41 2010
@@ -107,7 +107,7 @@
     def test_ulong_plus(self):
         self._dll.tf_bL.restype = c_ulong
         self._dll.tf_bL.argtypes = (c_char, c_ulong)
-        self.assertEqual(self._dll.tf_bL(' ', 4294967295), 1431655765)
+        self.assertEqual(self._dll.tf_bL(b' ', 4294967295), 1431655765)
         self.assertEqual(self.U(), 4294967295)
 
     def test_longlong(self):

Modified: python/branches/py3k/Lib/ctypes/test/test_errno.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_errno.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_errno.py	Wed Jul 28 01:36:41 2010
@@ -19,7 +19,7 @@
 
         libc_open.argtypes = c_char_p, c_int
 
-        self.assertEqual(libc_open("", 0), -1)
+        self.assertEqual(libc_open(b"", 0), -1)
         self.assertEqual(get_errno(), errno.ENOENT)
 
         self.assertEqual(set_errno(32), errno.ENOENT)
@@ -35,7 +35,7 @@
                 else:
                     libc_open = libc.open
                 libc_open.argtypes = c_char_p, c_int
-                self.assertEqual(libc_open("", 0), -1)
+                self.assertEqual(libc_open(b"", 0), -1)
                 self.assertEqual(get_errno(), 0)
 
             t = threading.Thread(target=_worker)

Modified: python/branches/py3k/Lib/ctypes/test/test_internals.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_internals.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_internals.py	Wed Jul 28 01:36:41 2010
@@ -70,19 +70,17 @@
         class Y(Structure):
             _fields_ = [("x", X), ("y", X)]
 
-        s1 = "Hello, World"
-        s2 = "Hallo, Welt"
+        s1 = b"Hello, World"
+        s2 = b"Hallo, Welt"
 
         x = X()
         x.a = s1
         x.b = s2
-        self.assertEqual(x._objects, {"0": bytes(s1, "ascii"),
-                                          "1": bytes(s2, "ascii")})
+        self.assertEqual(x._objects, {"0": s1, "1": s2})
 
         y = Y()
         y.x = x
-        self.assertEqual(y._objects, {"0": {"0": bytes(s1, "ascii"),
-                                                "1": bytes(s2, "ascii")}})
+        self.assertEqual(y._objects, {"0": {"0": s1, "1": s2}})
 ##        x = y.x
 ##        del y
 ##        print x._b_base_._objects

Modified: python/branches/py3k/Lib/ctypes/test/test_keeprefs.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_keeprefs.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_keeprefs.py	Wed Jul 28 01:36:41 2010
@@ -13,9 +13,9 @@
     def test_ccharp(self):
         x = c_char_p()
         self.assertEquals(x._objects, None)
-        x.value = "abc"
+        x.value = b"abc"
         self.assertEquals(x._objects, b"abc")
-        x = c_char_p("spam")
+        x = c_char_p(b"spam")
         self.assertEquals(x._objects, b"spam")
 
 class StructureTestCase(unittest.TestCase):
@@ -37,8 +37,8 @@
         x = X()
         self.assertEquals(x._objects, None)
 
-        x.a = "spam"
-        x.b = "foo"
+        x.a = b"spam"
+        x.b = b"foo"
         self.assertEquals(x._objects, {"0": b"spam", "1": b"foo"})
 
     def test_struct_struct(self):

Modified: python/branches/py3k/Lib/ctypes/test/test_libc.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_libc.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_libc.py	Wed Jul 28 01:36:41 2010
@@ -25,7 +25,7 @@
         def sort(a, b):
             return three_way_cmp(a[0], b[0])
 
-        chars = create_string_buffer("spam, spam, and spam")
+        chars = create_string_buffer(b"spam, spam, and spam")
         lib.my_qsort(chars, len(chars)-1, sizeof(c_char), comparefunc(sort))
         self.assertEqual(chars.raw, b"   ,,aaaadmmmnpppsss\x00")
 

Modified: python/branches/py3k/Lib/ctypes/test/test_objects.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_objects.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_objects.py	Wed Jul 28 01:36:41 2010
@@ -20,7 +20,7 @@
 The memory block stores pointers to strings, and the strings itself
 assigned from Python must be kept.
 
->>> array[4] = 'foo bar'
+>>> array[4] = b'foo bar'
 >>> array._objects
 {'4': b'foo bar'}
 >>> array[4]
@@ -45,7 +45,7 @@
 <ctypes.test.test_objects.X object at 0x...>
 >>>
 
->>> x.array[0] = 'spam spam spam'
+>>> x.array[0] = b'spam spam spam'
 >>> x._objects
 {'0:2': b'spam spam spam'}
 >>> x.array._b_base_._objects

Modified: python/branches/py3k/Lib/ctypes/test/test_parameters.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_parameters.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_parameters.py	Wed Jul 28 01:36:41 2010
@@ -19,7 +19,6 @@
         else:
             set_conversion_mode(*self.prev_conv_mode)
 
-
     def test_subclasses(self):
         from ctypes import c_void_p, c_char_p
         # ctypes 0.9.5 and before did overwrite from_param in SimpleType_new
@@ -60,12 +59,11 @@
         # new in 0.9.1: convert (encode) unicode to ascii
         self.assertEqual(c_char_p.from_param("123")._obj, b"123")
         self.assertRaises(UnicodeEncodeError, c_char_p.from_param, "123\377")
-
         self.assertRaises(TypeError, c_char_p.from_param, 42)
 
         # calling c_char_p.from_param with a c_char_p instance
         # returns the argument itself:
-        a = c_char_p("123")
+        a = c_char_p(b"123")
         self.assertTrue(c_char_p.from_param(a) is a)
 
     def test_cw_strings(self):

Modified: python/branches/py3k/Lib/ctypes/test/test_prototypes.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_prototypes.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_prototypes.py	Wed Jul 28 01:36:41 2010
@@ -127,7 +127,7 @@
         self.assertEqual(None, func(c_char_p(None)))
 
         self.assertEqual(b"123", func(c_buffer(b"123")))
-        ca = c_char("a")
+        ca = c_char(b"a")
         self.assertEqual(ord(b"a"), func(pointer(ca))[0])
         self.assertEqual(ord(b"a"), func(byref(ca))[0])
 

Modified: python/branches/py3k/Lib/ctypes/test/test_python_api.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_python_api.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_python_api.py	Wed Jul 28 01:36:41 2010
@@ -72,10 +72,10 @@
         PyOS_snprintf.argtypes = POINTER(c_char), c_size_t, c_char_p
 
         buf = c_buffer(256)
-        PyOS_snprintf(buf, sizeof(buf), "Hello from %s", b"ctypes")
+        PyOS_snprintf(buf, sizeof(buf), b"Hello from %s", b"ctypes")
         self.assertEqual(buf.value, b"Hello from ctypes")
 
-        PyOS_snprintf(buf, sizeof(buf), "Hello from %s (%d, %d, %d)", b"ctypes", 1, 2, 3)
+        PyOS_snprintf(buf, sizeof(buf), b"Hello from %s (%d, %d, %d)", b"ctypes", 1, 2, 3)
         self.assertEqual(buf.value, b"Hello from ctypes (1, 2, 3)")
 
         # not enough arguments

Modified: python/branches/py3k/Lib/ctypes/test/test_random_things.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_random_things.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_random_things.py	Wed Jul 28 01:36:41 2010
@@ -66,7 +66,7 @@
 
     def test_TypeErrorDivisionError(self):
         cb = CFUNCTYPE(c_int, c_char_p)(callback_func)
-        out = self.capture_stderr(cb, "spam")
+        out = self.capture_stderr(cb, b"spam")
         self.assertEqual(out.splitlines()[-1],
                              "TypeError: "
                              "unsupported operand type(s) for /: 'int' and 'bytes'")

Modified: python/branches/py3k/Lib/ctypes/test/test_repr.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_repr.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_repr.py	Wed Jul 28 01:36:41 2010
@@ -22,8 +22,8 @@
             self.assertEqual("<X object at", repr(typ(42))[:12])
 
     def test_char(self):
-        self.assertEqual("c_char(b'x')", repr(c_char('x')))
-        self.assertEqual("<X object at", repr(X('x'))[:12])
+        self.assertEqual("c_char(b'x')", repr(c_char(b'x')))
+        self.assertEqual("<X object at", repr(X(b'x'))[:12])
 
 if __name__ == "__main__":
     unittest.main()

Modified: python/branches/py3k/Lib/ctypes/test/test_returnfuncptrs.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_returnfuncptrs.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_returnfuncptrs.py	Wed Jul 28 01:36:41 2010
@@ -28,10 +28,10 @@
         # _CFuncPtr instances are now callable with an integer argument
         # which denotes a function address:
         strchr = CFUNCTYPE(c_char_p, c_char_p, c_char)(addr)
-        self.assertTrue(strchr("abcdef", "b"), "bcdef")
-        self.assertEqual(strchr("abcdef", "x"), None)
-        self.assertRaises(ArgumentError, strchr, "abcdef", 3.0)
-        self.assertRaises(TypeError, strchr, "abcdef")
+        self.assertTrue(strchr(b"abcdef", b"b"), "bcdef")
+        self.assertEqual(strchr(b"abcdef", b"x"), None)
+        self.assertRaises(ArgumentError, strchr, b"abcdef", 3.0)
+        self.assertRaises(TypeError, strchr, b"abcdef")
 
 if __name__ == "__main__":
     unittest.main()

Modified: python/branches/py3k/Lib/ctypes/test/test_stringptr.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_stringptr.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_stringptr.py	Wed Jul 28 01:36:41 2010
@@ -14,7 +14,7 @@
 
         # NULL pointer access
         self.assertRaises(ValueError, getattr, x.str, "contents")
-        b = c_buffer("Hello, World")
+        b = c_buffer(b"Hello, World")
         from sys import getrefcount as grc
         self.assertEqual(grc(b), 2)
         x.str = b
@@ -63,8 +63,8 @@
         # So we must keep a reference to buf separately
 
         strchr.restype = POINTER(c_char)
-        buf = c_buffer("abcdef")
-        r = strchr(buf, "c")
+        buf = c_buffer(b"abcdef")
+        r = strchr(buf, b"c")
         x = r[0], r[1], r[2], r[3], r[4]
         self.assertEqual(x, (b"c", b"d", b"e", b"f", b"\000"))
         del buf

Modified: python/branches/py3k/Lib/ctypes/test/test_strings.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_strings.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_strings.py	Wed Jul 28 01:36:41 2010
@@ -5,23 +5,23 @@
     def test(self):
         BUF = c_char * 4
 
-        buf = BUF("a", "b", "c")
+        buf = BUF(b"a", b"b", b"c")
         self.assertEqual(buf.value, b"abc")
         self.assertEqual(buf.raw, b"abc\000")
 
-        buf.value = "ABCD"
+        buf.value = b"ABCD"
         self.assertEqual(buf.value, b"ABCD")
         self.assertEqual(buf.raw, b"ABCD")
 
-        buf.value = "x"
+        buf.value = b"x"
         self.assertEqual(buf.value, b"x")
         self.assertEqual(buf.raw, b"x\000CD")
 
-        buf[1] = "Z"
+        buf[1] = b"Z"
         self.assertEqual(buf.value, b"xZCD")
         self.assertEqual(buf.raw, b"xZCD")
 
-        self.assertRaises(ValueError, setattr, buf, "value", "aaaaaaaa")
+        self.assertRaises(ValueError, setattr, buf, "value", b"aaaaaaaa")
         self.assertRaises(TypeError, setattr, buf, "value", 42)
 
     def test_c_buffer_value(self):

Modified: python/branches/py3k/Lib/ctypes/test/test_structures.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_structures.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_structures.py	Wed Jul 28 01:36:41 2010
@@ -205,15 +205,15 @@
                         ("age", c_int)]
 
         self.assertRaises(TypeError, Person, 42)
-        self.assertRaises(ValueError, Person, "asldkjaslkdjaslkdj")
+        self.assertRaises(ValueError, Person, b"asldkjaslkdjaslkdj")
         self.assertRaises(TypeError, Person, "Name", "HI")
 
         # short enough
-        self.assertEqual(Person("12345", 5).name, b"12345")
+        self.assertEqual(Person(b"12345", 5).name, b"12345")
         # exact fit
-        self.assertEqual(Person("123456", 5).name, b"123456")
+        self.assertEqual(Person(b"123456", 5).name, b"123456")
         # too long
-        self.assertRaises(ValueError, Person, "1234567", 5)
+        self.assertRaises(ValueError, Person, b"1234567", 5)
 
     def test_conflicting_initializers(self):
         class POINT(Structure):
@@ -267,7 +267,7 @@
                         ("phone", Phone),
                         ("age", c_int)]
 
-        p = Person("Someone", ("1234", "5678"), 5)
+        p = Person(b"Someone", (b"1234", b"5678"), 5)
 
         self.assertEqual(p.name, b"Someone")
         self.assertEqual(p.phone.areacode, b"1234")
@@ -284,8 +284,8 @@
             _fields_ = [("name", c_wchar * 12),
                         ("age", c_int)]
 
-        p = PersonW("Someone")
-        self.assertEqual(p.name, "Someone")
+        p = PersonW("Someone \xe9")
+        self.assertEqual(p.name, "Someone \xe9")
 
         self.assertEqual(PersonW("1234567890").name, "1234567890")
         self.assertEqual(PersonW("12345678901").name, "12345678901")
@@ -304,13 +304,13 @@
                         ("phone", Phone),
                         ("age", c_int)]
 
-        cls, msg = self.get_except(Person, "Someone", (1, 2))
+        cls, msg = self.get_except(Person, b"Someone", (1, 2))
         self.assertEqual(cls, RuntimeError)
         self.assertEqual(msg,
                              "(Phone) <class 'TypeError'>: "
                              "expected string, int found")
 
-        cls, msg = self.get_except(Person, "Someone", ("a", "b", "c"))
+        cls, msg = self.get_except(Person, b"Someone", (b"a", b"b", b"c"))
         self.assertEqual(cls, RuntimeError)
         if issubclass(Exception, object):
             self.assertEqual(msg,

From python-checkins at python.org  Wed Jul 28 01:45:05 2010
From: python-checkins at python.org (ezio.melotti)
Date: Wed, 28 Jul 2010 01:45:05 +0200 (CEST)
Subject: [Python-checkins] r83192 - in python/branches/release27-maint:
	Lib/test/test_asyncore.py
Message-ID: <20100727234505.AD74AEEA86@mail.python.org>

Author: ezio.melotti
Date: Wed Jul 28 01:45:05 2010
New Revision: 83192

Log:
Merged revisions 83183,83186 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83183 | ezio.melotti | 2010-07-28 01:03:33 +0300 (Wed, 28 Jul 2010) | 1 line
  
  Use proper skips and assert* methods in test_asyncore.
........
  r83186 | ezio.melotti | 2010-07-28 01:24:13 +0300 (Wed, 28 Jul 2010) | 1 line
  
  With skipUnless there is no need to add test classes conditionally.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/test/test_asyncore.py

Modified: python/branches/release27-maint/Lib/test/test_asyncore.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_asyncore.py	(original)
+++ python/branches/release27-maint/Lib/test/test_asyncore.py	Wed Jul 28 01:45:05 2010
@@ -117,65 +117,65 @@
     # http://mail.python.org/pipermail/python-list/2001-October/109973.html)
     # These constants should be present as long as poll is available
 
-    if hasattr(select, 'poll'):
-        def test_readwrite(self):
-            # Check that correct methods are called by readwrite()
-
-            attributes = ('read', 'expt', 'write', 'closed', 'error_handled')
-
-            expected = (
-                (select.POLLIN, 'read'),
-                (select.POLLPRI, 'expt'),
-                (select.POLLOUT, 'write'),
-                (select.POLLERR, 'closed'),
-                (select.POLLHUP, 'closed'),
-                (select.POLLNVAL, 'closed'),
-                )
-
-            class testobj:
-                def __init__(self):
-                    self.read = False
-                    self.write = False
-                    self.closed = False
-                    self.expt = False
-                    self.error_handled = False
-
-                def handle_read_event(self):
-                    self.read = True
-
-                def handle_write_event(self):
-                    self.write = True
-
-                def handle_close(self):
-                    self.closed = True
-
-                def handle_expt_event(self):
-                    self.expt = True
-
-                def handle_error(self):
-                    self.error_handled = True
-
-            for flag, expectedattr in expected:
-                tobj = testobj()
-                self.assertEqual(getattr(tobj, expectedattr), False)
-                asyncore.readwrite(tobj, flag)
-
-                # Only the attribute modified by the routine we expect to be
-                # called should be True.
-                for attr in attributes:
-                    self.assertEqual(getattr(tobj, attr), attr==expectedattr)
-
-                # check that ExitNow exceptions in the object handler method
-                # bubbles all the way up through asyncore readwrite call
-                tr1 = exitingdummy()
-                self.assertRaises(asyncore.ExitNow, asyncore.readwrite, tr1, flag)
-
-                # check that an exception other than ExitNow in the object handler
-                # method causes the handle_error method to get called
-                tr2 = crashingdummy()
-                self.assertEqual(tr2.error_handled, False)
-                asyncore.readwrite(tr2, flag)
-                self.assertEqual(tr2.error_handled, True)
+    @unittest.skipUnless(hasattr(select, 'poll'), 'select.poll required')
+    def test_readwrite(self):
+        # Check that correct methods are called by readwrite()
+
+        attributes = ('read', 'expt', 'write', 'closed', 'error_handled')
+
+        expected = (
+            (select.POLLIN, 'read'),
+            (select.POLLPRI, 'expt'),
+            (select.POLLOUT, 'write'),
+            (select.POLLERR, 'closed'),
+            (select.POLLHUP, 'closed'),
+            (select.POLLNVAL, 'closed'),
+            )
+
+        class testobj:
+            def __init__(self):
+                self.read = False
+                self.write = False
+                self.closed = False
+                self.expt = False
+                self.error_handled = False
+
+            def handle_read_event(self):
+                self.read = True
+
+            def handle_write_event(self):
+                self.write = True
+
+            def handle_close(self):
+                self.closed = True
+
+            def handle_expt_event(self):
+                self.expt = True
+
+            def handle_error(self):
+                self.error_handled = True
+
+        for flag, expectedattr in expected:
+            tobj = testobj()
+            self.assertEqual(getattr(tobj, expectedattr), False)
+            asyncore.readwrite(tobj, flag)
+
+            # Only the attribute modified by the routine we expect to be
+            # called should be True.
+            for attr in attributes:
+                self.assertEqual(getattr(tobj, attr), attr==expectedattr)
+
+            # check that ExitNow exceptions in the object handler method
+            # bubbles all the way up through asyncore readwrite call
+            tr1 = exitingdummy()
+            self.assertRaises(asyncore.ExitNow, asyncore.readwrite, tr1, flag)
+
+            # check that an exception other than ExitNow in the object handler
+            # method causes the handle_error method to get called
+            tr2 = crashingdummy()
+            self.assertEqual(tr2.error_handled, False)
+            asyncore.readwrite(tr2, flag)
+            self.assertEqual(tr2.error_handled, True)
 
     def test_closeall(self):
         self.closeall_check(False)
@@ -258,7 +258,7 @@
             sys.stderr = stderr
 
         lines = fp.getvalue().splitlines()
-        self.assertEquals(lines, ['log: %s' % l1, 'log: %s' % l2])
+        self.assertEqual(lines, ['log: %s' % l1, 'log: %s' % l2])
 
     def test_log_info(self):
         d = asyncore.dispatcher()
@@ -280,7 +280,7 @@
         lines = fp.getvalue().splitlines()
         expected = ['EGGS: %s' % l1, 'info: %s' % l2, 'SPAM: %s' % l3]
 
-        self.assertEquals(lines, expected)
+        self.assertEqual(lines, expected)
 
     def test_unhandled(self):
         d = asyncore.dispatcher()
@@ -305,7 +305,7 @@
                     'warning: unhandled write event',
                     'warning: unhandled connect event',
                     'warning: unhandled accept event']
-        self.assertEquals(lines, expected)
+        self.assertEqual(lines, expected)
 
     def test_issue_8594(self):
         # XXX - this test is supposed to be removed in next major Python
@@ -321,7 +321,7 @@
             warnings.simplefilter("always")
             family = d.family
             self.assertEqual(family, socket.AF_INET)
-            self.assertTrue(len(w) == 1)
+            self.assertEqual(len(w), 1)
             self.assertTrue(issubclass(w[0].category, DeprecationWarning))
 
     def test_strerror(self):
@@ -330,7 +330,7 @@
         if hasattr(os, 'strerror'):
             self.assertEqual(err, os.strerror(errno.EPERM))
         err = asyncore._strerror(-1)
-        self.assertTrue("unknown error" in err.lower())
+        self.assertIn("unknown error", err.lower())
 
 
 class dispatcherwithsend_noread(asyncore.dispatcher_with_send):
@@ -393,38 +393,40 @@
 class DispatcherWithSendTests_UsePoll(DispatcherWithSendTests):
     usepoll = True
 
-if hasattr(asyncore, 'file_wrapper'):
-    class FileWrapperTest(unittest.TestCase):
-        def setUp(self):
-            self.d = "It's not dead, it's sleeping!"
-            file(TESTFN, 'w').write(self.d)
-
-        def tearDown(self):
-            unlink(TESTFN)
-
-        def test_recv(self):
-            fd = os.open(TESTFN, os.O_RDONLY)
-            w = asyncore.file_wrapper(fd)
-            os.close(fd)
-
-            self.assertNotEqual(w.fd, fd)
-            self.assertNotEqual(w.fileno(), fd)
-            self.assertEqual(w.recv(13), "It's not dead")
-            self.assertEqual(w.read(6), ", it's")
-            w.close()
-            self.assertRaises(OSError, w.read, 1)
-
-        def test_send(self):
-            d1 = "Come again?"
-            d2 = "I want to buy some cheese."
-            fd = os.open(TESTFN, os.O_WRONLY | os.O_APPEND)
-            w = asyncore.file_wrapper(fd)
-            os.close(fd)
-
-            w.write(d1)
-            w.send(d2)
-            w.close()
-            self.assertEqual(file(TESTFN).read(), self.d + d1 + d2)
+ at unittest.skipUnless(hasattr(asyncore, 'file_wrapper'),
+                     'asyncore.file_wrapper required')
+class FileWrapperTest(unittest.TestCase):
+    def setUp(self):
+        self.d = "It's not dead, it's sleeping!"
+        file(TESTFN, 'w').write(self.d)
+
+    def tearDown(self):
+        unlink(TESTFN)
+
+    def test_recv(self):
+        fd = os.open(TESTFN, os.O_RDONLY)
+        w = asyncore.file_wrapper(fd)
+        os.close(fd)
+
+        self.assertNotEqual(w.fd, fd)
+        self.assertNotEqual(w.fileno(), fd)
+        self.assertEqual(w.recv(13), "It's not dead")
+        self.assertEqual(w.read(6), ", it's")
+        w.close()
+        self.assertRaises(OSError, w.read, 1)
+
+
+    def test_send(self):
+        d1 = "Come again?"
+        d2 = "I want to buy some cheese."
+        fd = os.open(TESTFN, os.O_WRONLY | os.O_APPEND)
+        w = asyncore.file_wrapper(fd)
+        os.close(fd)
+
+        w.write(d1)
+        w.send(d2)
+        w.close()
+        self.assertEqual(file(TESTFN).read(), self.d + d1 + d2)
 
 
 class BaseTestHandler(asyncore.dispatcher):
@@ -691,18 +693,15 @@
 class TestAPI_UseSelect(BaseTestAPI):
     use_poll = False
 
+ at unittest.skipUnless(hasattr(select, 'poll'), 'select.poll required')
 class TestAPI_UsePoll(BaseTestAPI):
     use_poll = True
 
 
 def test_main():
     tests = [HelperFunctionTests, DispatcherTests, DispatcherWithSendTests,
-             DispatcherWithSendTests_UsePoll, TestAPI_UseSelect]
-    if hasattr(asyncore, 'file_wrapper'):
-        tests.append(FileWrapperTest)
-    if hasattr(select, 'poll'):
-        tests.append(TestAPI_UsePoll)
-
+             DispatcherWithSendTests_UsePoll, TestAPI_UseSelect,
+             TestAPI_UsePoll, FileWrapperTest]
     run_unittest(*tests)
 
 if __name__ == "__main__":

From python-checkins at python.org  Wed Jul 28 01:47:57 2010
From: python-checkins at python.org (ezio.melotti)
Date: Wed, 28 Jul 2010 01:47:57 +0200 (CEST)
Subject: [Python-checkins] r83193 - python/branches/release31-maint
Message-ID: <20100727234757.8A9C0EE9B5@mail.python.org>

Author: ezio.melotti
Date: Wed Jul 28 01:47:57 2010
New Revision: 83193

Log:
Blocked revisions 83183,83186 via svnmerge

........
  r83183 | ezio.melotti | 2010-07-28 01:03:33 +0300 (Wed, 28 Jul 2010) | 1 line
  
  Use proper skips and assert* methods in test_asyncore.
........
  r83186 | ezio.melotti | 2010-07-28 01:24:13 +0300 (Wed, 28 Jul 2010) | 1 line
  
  With skipUnless there is no need to add test classes conditionally.
........


Modified:
   python/branches/release31-maint/   (props changed)

From python-checkins at python.org  Wed Jul 28 01:51:04 2010
From: python-checkins at python.org (ezio.melotti)
Date: Wed, 28 Jul 2010 01:51:04 +0200 (CEST)
Subject: [Python-checkins] r83194 - python/branches/release26-maint
Message-ID: <20100727235104.9F186EE9B5@mail.python.org>

Author: ezio.melotti
Date: Wed Jul 28 01:51:04 2010
New Revision: 83194

Log:
Blocked revisions 83192 via svnmerge

................
  r83192 | ezio.melotti | 2010-07-28 02:45:05 +0300 (Wed, 28 Jul 2010) | 13 lines
  
  Merged revisions 83183,83186 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k
  
  ........
    r83183 | ezio.melotti | 2010-07-28 01:03:33 +0300 (Wed, 28 Jul 2010) | 1 line
    
    Use proper skips and assert* methods in test_asyncore.
  ........
    r83186 | ezio.melotti | 2010-07-28 01:24:13 +0300 (Wed, 28 Jul 2010) | 1 line
    
    With skipUnless there is no need to add test classes conditionally.
  ........
................


Modified:
   python/branches/release26-maint/   (props changed)

From python-checkins at python.org  Wed Jul 28 02:15:04 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 28 Jul 2010 02:15:04 +0200 (CEST)
Subject: [Python-checkins] r83195 - in python/branches/py3k:
	Lib/ctypes/__init__.py Lib/ctypes/test/test_buffers.py
	Lib/ctypes/test/test_bytes.py Lib/ctypes/test/test_parameters.py
	Lib/ctypes/test/test_unicode.py Misc/NEWS
	Modules/_ctypes/_ctypes.c Modules/_ctypes/callproc.c
	Modules/_ctypes/cfield.c
Message-ID: <20100728001504.1364AEE9F8@mail.python.org>

Author: victor.stinner
Date: Wed Jul 28 02:15:03 2010
New Revision: 83195

Log:
Issue #8966: ctypes: Remove implicit bytes-unicode conversion


Modified:
   python/branches/py3k/Lib/ctypes/__init__.py
   python/branches/py3k/Lib/ctypes/test/test_buffers.py
   python/branches/py3k/Lib/ctypes/test/test_bytes.py
   python/branches/py3k/Lib/ctypes/test/test_parameters.py
   python/branches/py3k/Lib/ctypes/test/test_unicode.py
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/Modules/_ctypes/_ctypes.c
   python/branches/py3k/Modules/_ctypes/callproc.c
   python/branches/py3k/Modules/_ctypes/cfield.c

Modified: python/branches/py3k/Lib/ctypes/__init__.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/__init__.py	(original)
+++ python/branches/py3k/Lib/ctypes/__init__.py	Wed Jul 28 02:15:03 2010
@@ -259,41 +259,31 @@
 
 from _ctypes import POINTER, pointer, _pointer_type_cache
 
-try:
-    from _ctypes import set_conversion_mode
-except ImportError:
-    pass
-else:
-    if _os.name in ("nt", "ce"):
-        set_conversion_mode("mbcs", "strict")
-    else:
-        set_conversion_mode("ascii", "strict")
-
-    class c_wchar_p(_SimpleCData):
-        _type_ = "Z"
+class c_wchar_p(_SimpleCData):
+    _type_ = "Z"
 
-    class c_wchar(_SimpleCData):
-        _type_ = "u"
+class c_wchar(_SimpleCData):
+    _type_ = "u"
 
-    POINTER(c_wchar).from_param = c_wchar_p.from_param #_SimpleCData.c_wchar_p_from_param
+POINTER(c_wchar).from_param = c_wchar_p.from_param #_SimpleCData.c_wchar_p_from_param
 
-    def create_unicode_buffer(init, size=None):
-        """create_unicode_buffer(aString) -> character array
-        create_unicode_buffer(anInteger) -> character array
-        create_unicode_buffer(aString, anInteger) -> character array
-        """
-        if isinstance(init, (str, bytes)):
-            if size is None:
-                size = len(init)+1
-            buftype = c_wchar * size
-            buf = buftype()
-            buf.value = init
-            return buf
-        elif isinstance(init, int):
-            buftype = c_wchar * init
-            buf = buftype()
-            return buf
-        raise TypeError(init)
+def create_unicode_buffer(init, size=None):
+    """create_unicode_buffer(aString) -> character array
+    create_unicode_buffer(anInteger) -> character array
+    create_unicode_buffer(aString, anInteger) -> character array
+    """
+    if isinstance(init, (str, bytes)):
+        if size is None:
+            size = len(init)+1
+        buftype = c_wchar * size
+        buf = buftype()
+        buf.value = init
+        return buf
+    elif isinstance(init, int):
+        buftype = c_wchar * init
+        buf = buftype()
+        return buf
+    raise TypeError(init)
 
 POINTER(c_char).from_param = c_char_p.from_param #_SimpleCData.c_char_p_from_param
 

Modified: python/branches/py3k/Lib/ctypes/test/test_buffers.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_buffers.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_buffers.py	Wed Jul 28 02:15:03 2010
@@ -20,18 +20,6 @@
         self.assertEqual(b[::2], b"ac")
         self.assertEqual(b[::5], b"a")
 
-    def test_string_conversion(self):
-        b = create_string_buffer("abc")
-        self.assertEqual(len(b), 4) # trailing nul char
-        self.assertEqual(sizeof(b), 4 * sizeof(c_char))
-        self.assertTrue(type(b[0]) is bytes)
-        self.assertEqual(b[0], b"a")
-        self.assertEqual(b[:], b"abc\0")
-        self.assertEqual(b[::], b"abc\0")
-        self.assertEqual(b[::-1], b"\0cba")
-        self.assertEqual(b[::2], b"ac")
-        self.assertEqual(b[::5], b"a")
-
     try:
         c_wchar
     except NameError:

Modified: python/branches/py3k/Lib/ctypes/test/test_bytes.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_bytes.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_bytes.py	Wed Jul 28 02:15:03 2010
@@ -17,18 +17,15 @@
         (c_wchar * 3)("a", "b", "c")
 
     def test_c_char_p(self):
-        c_char_p("foo bar")
         c_char_p(b"foo bar")
 
     def test_c_wchar_p(self):
         c_wchar_p("foo bar")
-        c_wchar_p(b"foo bar")
 
     def test_struct(self):
         class X(Structure):
             _fields_ = [("a", c_char * 3)]
 
-        X("abc")
         x = X(b"abc")
         self.assertEqual(x.a, b"abc")
         self.assertEqual(type(x.a), bytes)
@@ -37,7 +34,6 @@
         class X(Structure):
             _fields_ = [("a", c_wchar * 3)]
 
-        X(b"abc")
         x = X("abc")
         self.assertEqual(x.a, "abc")
         self.assertEqual(type(x.a), str)

Modified: python/branches/py3k/Lib/ctypes/test/test_parameters.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_parameters.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_parameters.py	Wed Jul 28 02:15:03 2010
@@ -57,8 +57,8 @@
         self.assertTrue(c_char_p.from_param(s)._obj is s)
 
         # new in 0.9.1: convert (encode) unicode to ascii
-        self.assertEqual(c_char_p.from_param("123")._obj, b"123")
-        self.assertRaises(UnicodeEncodeError, c_char_p.from_param, "123\377")
+        self.assertEqual(c_char_p.from_param(b"123")._obj, b"123")
+        self.assertRaises(TypeError, c_char_p.from_param, "123\377")
         self.assertRaises(TypeError, c_char_p.from_param, 42)
 
         # calling c_char_p.from_param with a c_char_p instance
@@ -80,7 +80,7 @@
 
             # new in 0.9.1: convert (decode) ascii to unicode
             self.assertEqual(c_wchar_p.from_param("123")._obj, "123")
-        self.assertRaises(UnicodeDecodeError, c_wchar_p.from_param, b"123\377")
+        self.assertRaises(TypeError, c_wchar_p.from_param, b"123\377")
 
         pa = c_wchar_p.from_param(c_wchar_p("123"))
         self.assertEqual(type(pa), c_wchar_p)

Modified: python/branches/py3k/Lib/ctypes/test/test_unicode.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_unicode.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_unicode.py	Wed Jul 28 02:15:03 2010
@@ -7,122 +7,53 @@
     pass
 else:
     import _ctypes_test
-    dll = ctypes.CDLL(_ctypes_test.__file__)
-    wcslen = dll.my_wcslen
-    wcslen.argtypes = [ctypes.c_wchar_p]
-
 
     class UnicodeTestCase(unittest.TestCase):
-        def setUp(self):
-            self.prev_conv_mode = ctypes.set_conversion_mode("ascii", "strict")
+        def test_wcslen(self):
+            dll = ctypes.CDLL(_ctypes_test.__file__)
+            wcslen = dll.my_wcslen
+            wcslen.argtypes = [ctypes.c_wchar_p]
 
-        def tearDown(self):
-            ctypes.set_conversion_mode(*self.prev_conv_mode)
-
-        def test_ascii_strict(self):
-            ctypes.set_conversion_mode("ascii", "strict")
-            # no conversions take place with unicode arguments
             self.assertEqual(wcslen("abc"), 3)
             self.assertEqual(wcslen("ab\u2070"), 3)
-            # string args are converted
-            self.assertEqual(wcslen("abc"), 3)
             self.assertRaises(ctypes.ArgumentError, wcslen, b"ab\xe4")
 
-        def test_ascii_replace(self):
-            ctypes.set_conversion_mode("ascii", "replace")
-            self.assertEqual(wcslen("abc"), 3)
-            self.assertEqual(wcslen("ab\u2070"), 3)
-            self.assertEqual(wcslen("abc"), 3)
-            self.assertEqual(wcslen("ab\xe4"), 3)
-
-        def test_ascii_ignore(self):
-            ctypes.set_conversion_mode("ascii", "ignore")
-            self.assertEqual(wcslen("abc"), 3)
-            self.assertEqual(wcslen("ab\u2070"), 3)
-            # ignore error mode skips non-ascii characters
-            self.assertEqual(wcslen("abc"), 3)
-            self.assertEqual(wcslen(b"\xe4\xf6\xfc\xdf"), 0)
-
-        def test_latin1_strict(self):
-            ctypes.set_conversion_mode("latin-1", "strict")
-            self.assertEqual(wcslen("abc"), 3)
-            self.assertEqual(wcslen("ab\u2070"), 3)
-            self.assertEqual(wcslen("abc"), 3)
-            self.assertEqual(wcslen("\xe4\xf6\xfc\xdf"), 4)
-
         def test_buffers(self):
-            ctypes.set_conversion_mode("ascii", "strict")
             buf = ctypes.create_unicode_buffer("abc")
             self.assertEqual(len(buf), 3+1)
 
-            ctypes.set_conversion_mode("ascii", "replace")
-            buf = ctypes.create_unicode_buffer(b"ab\xe4\xf6\xfc")
-            self.assertEqual(buf[:], "ab\uFFFD\uFFFD\uFFFD\0")
-            self.assertEqual(buf[::], "ab\uFFFD\uFFFD\uFFFD\0")
-            self.assertEqual(buf[::-1], "\0\uFFFD\uFFFD\uFFFDba")
-            self.assertEqual(buf[::2], "a\uFFFD\uFFFD")
+            buf = ctypes.create_unicode_buffer("ab\xe4\xf6\xfc")
+            self.assertEqual(buf[:], "ab\xe4\xf6\xfc\0")
+            self.assertEqual(buf[::], "ab\xe4\xf6\xfc\0")
+            self.assertEqual(buf[::-1], '\x00\xfc\xf6\xe4ba')
+            self.assertEqual(buf[::2], 'a\xe4\xfc')
             self.assertEqual(buf[6:5:-1], "")
 
-            ctypes.set_conversion_mode("ascii", "ignore")
-            buf = ctypes.create_unicode_buffer(b"ab\xe4\xf6\xfc")
-            # is that correct? not sure.  But with 'ignore', you get what you pay for..
-            self.assertEqual(buf[:], "ab\0\0\0\0")
-            self.assertEqual(buf[::], "ab\0\0\0\0")
-            self.assertEqual(buf[::-1], "\0\0\0\0ba")
-            self.assertEqual(buf[::2], "a\0\0")
-            self.assertEqual(buf[6:5:-1], "")
-
-    import _ctypes_test
     func = ctypes.CDLL(_ctypes_test.__file__)._testfunc_p_p
 
     class StringTestCase(UnicodeTestCase):
         def setUp(self):
-            self.prev_conv_mode = ctypes.set_conversion_mode("ascii", "strict")
             func.argtypes = [ctypes.c_char_p]
             func.restype = ctypes.c_char_p
 
         def tearDown(self):
-            ctypes.set_conversion_mode(*self.prev_conv_mode)
             func.argtypes = None
             func.restype = ctypes.c_int
 
-        def test_ascii_replace(self):
-            ctypes.set_conversion_mode("ascii", "strict")
-            self.assertEqual(func("abc"), "abc")
-            self.assertEqual(func("abc"), "abc")
-            self.assertRaises(ctypes.ArgumentError, func, "ab\xe4")
-
-        def test_ascii_ignore(self):
-            ctypes.set_conversion_mode("ascii", "ignore")
-            self.assertEqual(func("abc"), b"abc")
-            self.assertEqual(func("abc"), b"abc")
-            self.assertEqual(func("\xe4\xf6\xfc\xdf"), b"")
-
-        def test_ascii_replace(self):
-            ctypes.set_conversion_mode("ascii", "replace")
-            self.assertEqual(func("abc"), b"abc")
-            self.assertEqual(func("abc"), b"abc")
-            self.assertEqual(func("\xe4\xf6\xfc\xdf"), b"????")
+        def test_func(self):
+            self.assertEqual(func(b"abc\xe4"), b"abc\xe4")
 
         def test_buffers(self):
-            ctypes.set_conversion_mode("ascii", "strict")
-            buf = ctypes.create_string_buffer("abc")
+            buf = ctypes.create_string_buffer(b"abc")
             self.assertEqual(len(buf), 3+1)
 
-            ctypes.set_conversion_mode("ascii", "replace")
-            buf = ctypes.create_string_buffer("ab\xe4\xf6\xfc")
-            self.assertEqual(buf[:], b"ab???\0")
-            self.assertEqual(buf[::], b"ab???\0")
-            self.assertEqual(buf[::-1], b"\0???ba")
-            self.assertEqual(buf[::2], b"a??")
+            buf = ctypes.create_string_buffer(b"ab\xe4\xf6\xfc")
+            self.assertEqual(buf[:], b"ab\xe4\xf6\xfc\0")
+            self.assertEqual(buf[::], b"ab\xe4\xf6\xfc\0")
+            self.assertEqual(buf[::-1], b'\x00\xfc\xf6\xe4ba')
+            self.assertEqual(buf[::2], b'a\xe4\xfc')
             self.assertEqual(buf[6:5:-1], b"")
 
-            ctypes.set_conversion_mode("ascii", "ignore")
-            buf = ctypes.create_string_buffer("ab\xe4\xf6\xfc")
-            # is that correct? not sure.  But with 'ignore', you get what you pay for..
-            self.assertEqual(buf[:], b"ab\0\0\0\0")
-            self.assertEqual(buf[::], b"ab\0\0\0\0")
-            self.assertEqual(buf[::-1], b"\0\0\0\0ba")
 
 if __name__ == '__main__':
     unittest.main()

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Wed Jul 28 02:15:03 2010
@@ -473,6 +473,8 @@
 Library
 -------
 
+- Issue #8966: ctypes: Remove implicit bytes-unicode conversion.
+
 - Issue #9378: python -m pickle <pickle file> will now load and
   display the first object in the pickle file.
 

Modified: python/branches/py3k/Modules/_ctypes/_ctypes.c
==============================================================================
--- python/branches/py3k/Modules/_ctypes/_ctypes.c	(original)
+++ python/branches/py3k/Modules/_ctypes/_ctypes.c	Wed Jul 28 02:15:03 2010
@@ -132,8 +132,6 @@
 /* a callable object used for unpickling */
 static PyObject *_unpickle;
 
-char *_ctypes_conversion_encoding = NULL;
-char *_ctypes_conversion_errors = NULL;
 
 
 /****************************************************************/
@@ -1090,13 +1088,7 @@
         return -1;
     }
 
-    if (PyUnicode_Check(value)) {
-        value = PyUnicode_AsEncodedString(value,
-                                          _ctypes_conversion_encoding,
-                                          _ctypes_conversion_errors);
-        if (!value)
-            return -1;
-    } else if (!PyBytes_Check(value)) {
+    if (!PyBytes_Check(value)) {
         PyErr_Format(PyExc_TypeError,
                      "str/bytes expected instead of %s instance",
                      Py_TYPE(value)->tp_name);
@@ -1150,13 +1142,7 @@
                         "can't delete attribute");
         return -1;
     }
-    if (PyBytes_Check(value)) {
-        value = PyUnicode_FromEncodedObject(value,
-                                            _ctypes_conversion_encoding,
-                                            _ctypes_conversion_errors);
-        if (!value)
-            return -1;
-    } else if (!PyUnicode_Check(value)) {
+    if (!PyUnicode_Check(value)) {
         PyErr_Format(PyExc_TypeError,
                         "unicode string expected instead of %s instance",
                         Py_TYPE(value)->tp_name);
@@ -1510,7 +1496,7 @@
         Py_INCREF(Py_None);
         return Py_None;
     }
-    if (PyBytes_Check(value) || PyUnicode_Check(value)) {
+    if (PyBytes_Check(value)) {
         PyCArgObject *parg;
         struct fielddesc *fd = _ctypes_get_fielddesc("z");
 

Modified: python/branches/py3k/Modules/_ctypes/callproc.c
==============================================================================
--- python/branches/py3k/Modules/_ctypes/callproc.c	(original)
+++ python/branches/py3k/Modules/_ctypes/callproc.c	Wed Jul 28 02:15:03 2010
@@ -1641,37 +1641,6 @@
     return arg;
 }
 
-#ifdef CTYPES_UNICODE
-
-static char set_conversion_mode_doc[] =
-"set_conversion_mode(encoding, errors) -> (previous-encoding, previous-errors)\n\
-\n\
-Set the encoding and error handling ctypes uses when converting\n\
-between unicode and strings.  Returns the previous values.\n";
-
-static PyObject *
-set_conversion_mode(PyObject *self, PyObject *args)
-{
-    char *coding, *mode;
-    PyObject *result;
-
-    if (!PyArg_ParseTuple(args, "zs:set_conversion_mode", &coding, &mode))
-        return NULL;
-    result = Py_BuildValue("(zz)", _ctypes_conversion_encoding, _ctypes_conversion_errors);
-    if (coding) {
-        PyMem_Free(_ctypes_conversion_encoding);
-        _ctypes_conversion_encoding = PyMem_Malloc(strlen(coding) + 1);
-        strcpy(_ctypes_conversion_encoding, coding);
-    } else {
-        _ctypes_conversion_encoding = NULL;
-    }
-    PyMem_Free(_ctypes_conversion_errors);
-    _ctypes_conversion_errors = PyMem_Malloc(strlen(mode) + 1);
-    strcpy(_ctypes_conversion_errors, mode);
-    return result;
-}
-#endif
-
 static PyObject *
 resize(PyObject *self, PyObject *args)
 {
@@ -1852,9 +1821,6 @@
     {"_unpickle", unpickle, METH_VARARGS },
     {"buffer_info", buffer_info, METH_O, "Return buffer interface information"},
     {"resize", resize, METH_VARARGS, "Resize the memory buffer of a ctypes instance"},
-#ifdef CTYPES_UNICODE
-    {"set_conversion_mode", set_conversion_mode, METH_VARARGS, set_conversion_mode_doc},
-#endif
 #ifdef MS_WIN32
     {"get_last_error", get_last_error, METH_NOARGS},
     {"set_last_error", set_last_error, METH_VARARGS},

Modified: python/branches/py3k/Modules/_ctypes/cfield.c
==============================================================================
--- python/branches/py3k/Modules/_ctypes/cfield.c	(original)
+++ python/branches/py3k/Modules/_ctypes/cfield.c	Wed Jul 28 02:15:03 2010
@@ -1168,20 +1168,6 @@
 static PyObject *
 c_set(void *ptr, PyObject *value, Py_ssize_t size)
 {
-    if (PyUnicode_Check(value)) {
-        value = PyUnicode_AsEncodedString(value,
-                                          _ctypes_conversion_encoding,
-                                          _ctypes_conversion_errors);
-        if (value == NULL)
-            return NULL;
-        if (PyBytes_GET_SIZE(value) != 1) {
-            Py_DECREF(value);
-            goto error;
-        }
-        *(char *)ptr = PyBytes_AS_STRING(value)[0];
-        Py_DECREF(value);
-        _RET(value);
-    }
     if (PyBytes_Check(value) && PyBytes_GET_SIZE(value) == 1) {
         *(char *)ptr = PyBytes_AS_STRING(value)[0];
         _RET(value);
@@ -1217,13 +1203,7 @@
 u_set(void *ptr, PyObject *value, Py_ssize_t size)
 {
     Py_ssize_t len;
-    if (PyBytes_Check(value)) {
-        value = PyUnicode_FromEncodedObject(value,
-                                            _ctypes_conversion_encoding,
-                                            _ctypes_conversion_errors);
-        if (!value)
-            return NULL;
-    } else if (!PyUnicode_Check(value)) {
+    if (!PyUnicode_Check(value)) {
         PyErr_Format(PyExc_TypeError,
                         "unicode string expected instead of %s instance",
                         value->ob_type->tp_name);
@@ -1292,13 +1272,7 @@
     /* It's easier to calculate in characters than in bytes */
     length /= sizeof(wchar_t);
 
-    if (PyBytes_Check(value)) {
-        value = PyUnicode_FromEncodedObject(value,
-                                            _ctypes_conversion_encoding,
-                                            _ctypes_conversion_errors);
-        if (!value)
-            return NULL;
-    } else if (!PyUnicode_Check(value)) {
+    if (!PyUnicode_Check(value)) {
         PyErr_Format(PyExc_TypeError,
                         "unicode string expected instead of %s instance",
                         value->ob_type->tp_name);
@@ -1342,14 +1316,7 @@
     char *data;
     Py_ssize_t size;
 
-    if (PyUnicode_Check(value)) {
-        value = PyUnicode_AsEncodedString(value,
-                                                                          _ctypes_conversion_encoding,
-                                                                          _ctypes_conversion_errors);
-        if (value == NULL)
-            return NULL;
-        assert(PyBytes_Check(value));
-    } else if(PyBytes_Check(value)) {
+    if(PyBytes_Check(value)) {
         Py_INCREF(value);
     } else {
         PyErr_Format(PyExc_TypeError,
@@ -1393,14 +1360,6 @@
         *(char **)ptr = PyBytes_AsString(value);
         Py_INCREF(value);
         return value;
-    } else if (PyUnicode_Check(value)) {
-        PyObject *str = PyUnicode_AsEncodedString(value,
-                                                  _ctypes_conversion_encoding,
-                                                  _ctypes_conversion_errors);
-        if (str == NULL)
-            return NULL;
-        *(char **)ptr = PyBytes_AS_STRING(str);
-        return str;
     } else if (PyLong_Check(value)) {
 #if SIZEOF_VOID_P == SIZEOF_LONG_LONG
         *(char **)ptr = (char *)PyLong_AsUnsignedLongLongMask(value);
@@ -1454,13 +1413,7 @@
         Py_INCREF(Py_None);
         return Py_None;
     }
-    if (PyBytes_Check(value)) {
-        value = PyUnicode_FromEncodedObject(value,
-                                            _ctypes_conversion_encoding,
-                                            _ctypes_conversion_errors);
-        if (!value)
-            return NULL;
-    } else if (!PyUnicode_Check(value)) {
+    if (!PyUnicode_Check(value)) {
         PyErr_Format(PyExc_TypeError,
                      "unicode string or integer address expected instead of %s instance",
                      value->ob_type->tp_name);
@@ -1540,12 +1493,6 @@
     /* convert value into a PyUnicodeObject or NULL */
     if (Py_None == value) {
         value = NULL;
-    } else if (PyBytes_Check(value)) {
-        value = PyUnicode_FromEncodedObject(value,
-                                            _ctypes_conversion_encoding,
-                                            _ctypes_conversion_errors);
-        if (!value)
-            return NULL;
     } else if (PyUnicode_Check(value)) {
         Py_INCREF(value); /* for the descref below */
     } else {

From python-checkins at python.org  Wed Jul 28 02:23:21 2010
From: python-checkins at python.org (ezio.melotti)
Date: Wed, 28 Jul 2010 02:23:21 +0200 (CEST)
Subject: [Python-checkins] r83196 - python/branches/py3k/Lib/base64.py
Message-ID: <20100728002321.CD03AEE9F3@mail.python.org>

Author: ezio.melotti
Date: Wed Jul 28 02:23:21 2010
New Revision: 83196

Log:
Fix failure introduced in r83182.

Modified:
   python/branches/py3k/Lib/base64.py

Modified: python/branches/py3k/Lib/base64.py
==============================================================================
--- python/branches/py3k/Lib/base64.py	(original)
+++ python/branches/py3k/Lib/base64.py	Wed Jul 28 02:23:21 2010
@@ -241,7 +241,7 @@
         acc += _b32rev[c] << shift
         shift -= 5
         if shift < 0:
-            parts.append(binascii.unhexlify('%010x' % acc))
+            parts.append(binascii.unhexlify(bytes('%010x' % acc, "ascii")))
             acc = 0
             shift = 35
     # Process the last, partial quanta

From python-checkins at python.org  Wed Jul 28 02:40:58 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 28 Jul 2010 02:40:58 +0200 (CEST)
Subject: [Python-checkins] r83197 - in python/branches/py3k: Misc/NEWS
	Python/getargs.c
Message-ID: <20100728004058.960E6EE982@mail.python.org>

Author: victor.stinner
Date: Wed Jul 28 02:40:58 2010
New Revision: 83197

Log:
Issue #8991: convertbuffer() rejects discontigious buffers


Modified:
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/Python/getargs.c

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Wed Jul 28 02:40:58 2010
@@ -12,6 +12,8 @@
 Core and Builtins
 -----------------
 
+- Issue #8991: convertbuffer() rejects discontigious buffers.
+
 - Issue #7616: Fix copying of overlapping memoryview slices with the Intel
   compiler.
 

Modified: python/branches/py3k/Python/getargs.c
==============================================================================
--- python/branches/py3k/Python/getargs.c	(original)
+++ python/branches/py3k/Python/getargs.c	Wed Jul 28 02:40:58 2010
@@ -1246,13 +1246,15 @@
             PyErr_Clear();
             return converterr("read-write buffer", arg, msgbuf, bufsize);
         }
+        if (!PyBuffer_IsContiguous((Py_buffer*)p, 'C')) {
+            PyBuffer_Release((Py_buffer*)p);
+            return converterr("contiguous buffer", arg, msgbuf, bufsize);
+        }
         if (addcleanup(p, freelist, cleanup_buffer)) {
             return converterr(
                 "(cleanup problem)",
                 arg, msgbuf, bufsize);
         }
-        if (!PyBuffer_IsContiguous((Py_buffer*)p, 'C'))
-            return converterr("contiguous buffer", arg, msgbuf, bufsize);
         break;
     }
 
@@ -1274,41 +1276,26 @@
 
     *errmsg = NULL;
     *p = NULL;
-    if (pb == NULL ||
-        pb->bf_getbuffer == NULL ||
-        pb->bf_releasebuffer != NULL) {
-        *errmsg = "bytes or read-only buffer";
+    if (pb != NULL && pb->bf_releasebuffer != NULL) {
+        *errmsg = "read-only pinned buffer";
         return -1;
     }
 
-    if (PyObject_GetBuffer(arg, &view, PyBUF_SIMPLE) != 0) {
-        *errmsg = "bytes or single-segment read-only buffer";
+    if (getbuffer(arg, &view, errmsg) < 0)
         return -1;
-    }
     count = view.len;
     *p = view.buf;
     PyBuffer_Release(&view);
     return count;
 }
 
-/* XXX for 3.x, getbuffer and convertbuffer can probably
-   be merged again. */
 static int
 getbuffer(PyObject *arg, Py_buffer *view, char **errmsg)
 {
-    PyBufferProcs *pb = Py_TYPE(arg)->tp_as_buffer;
-    if (pb == NULL) {
+    if (PyObject_GetBuffer(arg, view, PyBUF_SIMPLE) != 0) {
         *errmsg = "bytes or buffer";
         return -1;
     }
-    if (pb->bf_getbuffer == NULL) {
-        *errmsg = "convertible to a buffer";
-        return -1;
-    }
-    if (PyObject_GetBuffer(arg, view, 0) < 0) {
-        *errmsg = "convertible to a buffer";
-        return -1;
-    }
     if (!PyBuffer_IsContiguous(view, 'C')) {
         PyBuffer_Release(view);
         *errmsg = "contiguous buffer";

From python-checkins at python.org  Wed Jul 28 03:39:46 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 28 Jul 2010 03:39:46 +0200 (CEST)
Subject: [Python-checkins] r83198 - in python/branches/release27-maint:
	Lib/encodings/utf_16.py Lib/encodings/utf_8_sig.py
	Lib/test/test_io.py Misc/NEWS
Message-ID: <20100728013946.21A2CEE982@mail.python.org>

Author: victor.stinner
Date: Wed Jul 28 03:39:45 2010
New Revision: 83198

Log:
Issue #6213: Implement getstate() and setstate() methods of utf-8-sig and
utf-16 incremental encoders.


Modified:
   python/branches/release27-maint/Lib/encodings/utf_16.py
   python/branches/release27-maint/Lib/encodings/utf_8_sig.py
   python/branches/release27-maint/Lib/test/test_io.py
   python/branches/release27-maint/Misc/NEWS

Modified: python/branches/release27-maint/Lib/encodings/utf_16.py
==============================================================================
--- python/branches/release27-maint/Lib/encodings/utf_16.py	(original)
+++ python/branches/release27-maint/Lib/encodings/utf_16.py	Wed Jul 28 03:39:45 2010
@@ -34,6 +34,22 @@
         codecs.IncrementalEncoder.reset(self)
         self.encoder = None
 
+    def getstate(self):
+        # state info we return to the caller:
+        # 0: stream is in natural order for this platform
+        # 2: endianness hasn't been determined yet
+        # (we're never writing in unnatural order)
+        return (2 if self.encoder is None else 0)
+
+    def setstate(self, state):
+        if state:
+            self.encoder = None
+        else:
+            if sys.byteorder == 'little':
+                self.encoder = codecs.utf_16_le_encode
+            else:
+                self.encoder = codecs.utf_16_be_encode
+
 class IncrementalDecoder(codecs.BufferedIncrementalDecoder):
     def __init__(self, errors='strict'):
         codecs.BufferedIncrementalDecoder.__init__(self, errors)

Modified: python/branches/release27-maint/Lib/encodings/utf_8_sig.py
==============================================================================
--- python/branches/release27-maint/Lib/encodings/utf_8_sig.py	(original)
+++ python/branches/release27-maint/Lib/encodings/utf_8_sig.py	Wed Jul 28 03:39:45 2010
@@ -25,18 +25,24 @@
 class IncrementalEncoder(codecs.IncrementalEncoder):
     def __init__(self, errors='strict'):
         codecs.IncrementalEncoder.__init__(self, errors)
-        self.first = True
+        self.first = 1
 
     def encode(self, input, final=False):
         if self.first:
-            self.first = False
+            self.first = 0
             return codecs.BOM_UTF8 + codecs.utf_8_encode(input, self.errors)[0]
         else:
             return codecs.utf_8_encode(input, self.errors)[0]
 
     def reset(self):
         codecs.IncrementalEncoder.reset(self)
-        self.first = True
+        self.first = 1
+
+    def getstate(self):
+        return self.first
+
+    def setstate(self, state):
+        self.first = state
 
 class IncrementalDecoder(codecs.BufferedIncrementalDecoder):
     def __init__(self, errors='strict'):

Modified: python/branches/release27-maint/Lib/test/test_io.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_io.py	(original)
+++ python/branches/release27-maint/Lib/test/test_io.py	Wed Jul 28 03:39:45 2010
@@ -2087,7 +2087,6 @@
 
         self.assertEqual(buffer.seekable(), txt.seekable())
 
-    @unittest.skip("Issue #6213 with incremental encoders")
     def test_append_bom(self):
         # The BOM is not written again when appending to a non-empty file
         filename = support.TESTFN
@@ -2103,7 +2102,6 @@
             with self.open(filename, 'rb') as f:
                 self.assertEquals(f.read(), 'aaaxxx'.encode(charset))
 
-    @unittest.skip("Issue #6213 with incremental encoders")
     def test_seek_bom(self):
         # Same test, but when seeking manually
         filename = support.TESTFN

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Wed Jul 28 03:39:45 2010
@@ -18,6 +18,9 @@
 Library
 -------
 
+- Issue #6213: Implement getstate() and setstate() methods of utf-8-sig and
+  utf-16 incremental encoders.
+
 - Issue #7113: Speed up loading in ConfigParser. Patch by ?ukasz Langa.
 
 - Issue #3704: cookielib was not properly handling URLs with a / in the

From python-checkins at python.org  Wed Jul 28 03:55:43 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 28 Jul 2010 03:55:43 +0200 (CEST)
Subject: [Python-checkins] r83199 - in python/branches/release26-maint:
	Lib/encodings/utf_16.py Lib/encodings/utf_8_sig.py Misc/NEWS
Message-ID: <20100728015543.F384CEE99B@mail.python.org>

Author: victor.stinner
Date: Wed Jul 28 03:55:43 2010
New Revision: 83199

Log:
Merged revisions 83198 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

........
  r83198 | victor.stinner | 2010-07-28 03:39:45 +0200 (mer., 28 juil. 2010) | 3 lines
  
  Issue #6213: Implement getstate() and setstate() methods of utf-8-sig and
  utf-16 incremental encoders.
........


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Lib/encodings/utf_16.py
   python/branches/release26-maint/Lib/encodings/utf_8_sig.py
   python/branches/release26-maint/Misc/NEWS

Modified: python/branches/release26-maint/Lib/encodings/utf_16.py
==============================================================================
--- python/branches/release26-maint/Lib/encodings/utf_16.py	(original)
+++ python/branches/release26-maint/Lib/encodings/utf_16.py	Wed Jul 28 03:55:43 2010
@@ -34,6 +34,22 @@
         codecs.IncrementalEncoder.reset(self)
         self.encoder = None
 
+    def getstate(self):
+        # state info we return to the caller:
+        # 0: stream is in natural order for this platform
+        # 2: endianness hasn't been determined yet
+        # (we're never writing in unnatural order)
+        return (2 if self.encoder is None else 0)
+
+    def setstate(self, state):
+        if state:
+            self.encoder = None
+        else:
+            if sys.byteorder == 'little':
+                self.encoder = codecs.utf_16_le_encode
+            else:
+                self.encoder = codecs.utf_16_be_encode
+
 class IncrementalDecoder(codecs.BufferedIncrementalDecoder):
     def __init__(self, errors='strict'):
         codecs.BufferedIncrementalDecoder.__init__(self, errors)

Modified: python/branches/release26-maint/Lib/encodings/utf_8_sig.py
==============================================================================
--- python/branches/release26-maint/Lib/encodings/utf_8_sig.py	(original)
+++ python/branches/release26-maint/Lib/encodings/utf_8_sig.py	Wed Jul 28 03:55:43 2010
@@ -25,18 +25,24 @@
 class IncrementalEncoder(codecs.IncrementalEncoder):
     def __init__(self, errors='strict'):
         codecs.IncrementalEncoder.__init__(self, errors)
-        self.first = True
+        self.first = 1
 
     def encode(self, input, final=False):
         if self.first:
-            self.first = False
+            self.first = 0
             return codecs.BOM_UTF8 + codecs.utf_8_encode(input, self.errors)[0]
         else:
             return codecs.utf_8_encode(input, self.errors)[0]
 
     def reset(self):
         codecs.IncrementalEncoder.reset(self)
-        self.first = True
+        self.first = 1
+
+    def getstate(self):
+        return self.first
+
+    def setstate(self, state):
+        self.first = state
 
 class IncrementalDecoder(codecs.BufferedIncrementalDecoder):
     def __init__(self, errors='strict'):

Modified: python/branches/release26-maint/Misc/NEWS
==============================================================================
--- python/branches/release26-maint/Misc/NEWS	(original)
+++ python/branches/release26-maint/Misc/NEWS	Wed Jul 28 03:55:43 2010
@@ -12,6 +12,9 @@
 Core and Builtins
 -----------------
 
+- Issue #6213: Implement getstate() and setstate() methods of utf-8-sig and
+  utf-16 incremental encoders.
+
 - Issue #8271: during the decoding of an invalid UTF-8 byte sequence, only the
   start byte and the continuation byte(s) are now considered invalid, instead
   of the number of bytes specified by the start byte.

From python-checkins at python.org  Wed Jul 28 03:58:41 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 28 Jul 2010 03:58:41 +0200 (CEST)
Subject: [Python-checkins] r83200 - in python/branches/release26-maint:
	Lib/io.py Lib/test/test_io.py Misc/NEWS
Message-ID: <20100728015841.72E96EE99B@mail.python.org>

Author: victor.stinner
Date: Wed Jul 28 03:58:41 2010
New Revision: 83200

Log:
Issue #5006: Better handling of unicode byte-order marks (BOM) in the io
library. This means, for example, that opening an UTF-16 text file in append
mode doesn't add a BOM at the end of the file if the file isn't empty.


Modified:
   python/branches/release26-maint/Lib/io.py
   python/branches/release26-maint/Lib/test/test_io.py
   python/branches/release26-maint/Misc/NEWS

Modified: python/branches/release26-maint/Lib/io.py
==============================================================================
--- python/branches/release26-maint/Lib/io.py	(original)
+++ python/branches/release26-maint/Lib/io.py	Wed Jul 28 03:58:41 2010
@@ -1440,6 +1440,15 @@
         self._snapshot = None  # info for reconstructing decoder state
         self._seekable = self._telling = self.buffer.seekable()
 
+        if self._seekable and self.writable():
+            position = self.buffer.tell()
+            if position != 0:
+                try:
+                    self._get_encoder().setstate(0)
+                except LookupError:
+                    # Sometimes the encoder doesn't exist
+                    pass
+
     # self._snapshot is either None, or a tuple (dec_flags, next_input)
     # where dec_flags is the second (integer) item of the decoder state
     # and next_input is the chunk of input bytes that comes next after the
@@ -1726,6 +1735,17 @@
                 raise IOError("can't restore logical file position")
             self._decoded_chars_used = chars_to_skip
 
+        # Finally, reset the encoder (merely useful for proper BOM handling)
+        try:
+            encoder = self._encoder or self._get_encoder()
+        except LookupError:
+            # Sometimes the encoder doesn't exist
+            pass
+        else:
+            if cookie != 0:
+                encoder.setstate(0)
+            else:
+                encoder.reset()
         return cookie
 
     def read(self, n=None):

Modified: python/branches/release26-maint/Lib/test/test_io.py
==============================================================================
--- python/branches/release26-maint/Lib/test/test_io.py	(original)
+++ python/branches/release26-maint/Lib/test/test_io.py	Wed Jul 28 03:58:41 2010
@@ -799,6 +799,37 @@
         self.assertEquals(d.decode(b'oiabcd'), '')
         self.assertEquals(d.decode(b'', 1), 'abcd.')
 
+    def test_append_bom(self):
+        # The BOM is not written again when appending to a non-empty file
+        filename = test_support.TESTFN
+        for charset in ('utf-8-sig', 'utf-16', 'utf-32'):
+            with io.open(filename, 'w', encoding=charset) as f:
+                f.write('aaa')
+                pos = f.tell()
+            with io.open(filename, 'rb') as f:
+                self.assertEquals(f.read(), 'aaa'.encode(charset))
+
+            with io.open(filename, 'a', encoding=charset) as f:
+                f.write('xxx')
+            with io.open(filename, 'rb') as f:
+                self.assertEquals(f.read(), 'aaaxxx'.encode(charset))
+
+    def test_seek_bom(self):
+        # Same test, but when seeking manually
+        filename = test_support.TESTFN
+        for charset in ('utf-8-sig', 'utf-16', 'utf-32'):
+            with io.open(filename, 'w', encoding=charset) as f:
+                f.write('aaa')
+                pos = f.tell()
+            with io.open(filename, 'r+', encoding=charset) as f:
+                f.seek(pos)
+                f.write('zzz')
+                f.seek(0)
+                f.write('bbb')
+            with io.open(filename, 'rb') as f:
+                self.assertEquals(f.read(), 'bbbzzz'.encode(charset))
+
+
 class TextIOWrapperTest(unittest.TestCase):
 
     def setUp(self):

Modified: python/branches/release26-maint/Misc/NEWS
==============================================================================
--- python/branches/release26-maint/Misc/NEWS	(original)
+++ python/branches/release26-maint/Misc/NEWS	Wed Jul 28 03:58:41 2010
@@ -84,6 +84,10 @@
 Library
 -------
 
+- Issue #5006: Better handling of unicode byte-order marks (BOM) in the io
+  library. This means, for example, that opening an UTF-16 text file in append
+  mode doesn't add a BOM at the end of the file if the file isn't empty.
+
 - Issue #3704: cookielib was not properly handling URLs with a / in the
   parameters.
 

From python-checkins at python.org  Wed Jul 28 10:19:37 2010
From: python-checkins at python.org (georg.brandl)
Date: Wed, 28 Jul 2010 10:19:37 +0200 (CEST)
Subject: [Python-checkins] r83201 - in python/branches/py3k: Lib/asyncore.py
	Lib/test/test_asyncore.py Misc/ACKS Misc/NEWS
Message-ID: <20100728081937.4D370F44C@mail.python.org>

Author: georg.brandl
Date: Wed Jul 28 10:19:35 2010
New Revision: 83201

Log:
#9354: Provide getsockopt() in asyncore file_wrapper().  Patch by Lukas Langa.

Modified:
   python/branches/py3k/Lib/asyncore.py
   python/branches/py3k/Lib/test/test_asyncore.py
   python/branches/py3k/Misc/ACKS
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/asyncore.py
==============================================================================
--- python/branches/py3k/Lib/asyncore.py	(original)
+++ python/branches/py3k/Lib/asyncore.py	Wed Jul 28 10:19:35 2010
@@ -607,6 +607,14 @@
         def send(self, *args):
             return os.write(self.fd, *args)
 
+        def getsockopt(self, level, optname, buflen=None):
+            if (level == socket.SOL_SOCKET and
+                optname == socket.SO_ERROR and
+                not buflen):
+                return 0
+            raise NotImplementedError("Only asyncore specific behaviour "
+                                      "implemented.")
+
         read = recv
         write = send
 

Modified: python/branches/py3k/Lib/test/test_asyncore.py
==============================================================================
--- python/branches/py3k/Lib/test/test_asyncore.py	(original)
+++ python/branches/py3k/Lib/test/test_asyncore.py	Wed Jul 28 10:19:35 2010
@@ -428,6 +428,19 @@
         w.close()
         self.assertEqual(open(TESTFN, 'rb').read(), self.d + d1 + d2)
 
+    @unittest.skipUnless(hasattr(asyncore, 'file_dispatcher'),
+                         'asyncore.file_dispatcher required')
+    def test_dispatcher(self):
+        fd = os.open(TESTFN, os.O_RDONLY)
+        data = []
+        class FileDispatcher(asyncore.file_dispatcher):
+            def handle_read(self):
+                data.append(self.recv(29))
+        s = FileDispatcher(fd)
+        os.close(fd)
+        asyncore.loop(timeout=0.01, use_poll=True, count=2)
+        self.assertEqual(b"".join(data), self.d)
+
 
 class BaseTestHandler(asyncore.dispatcher):
 

Modified: python/branches/py3k/Misc/ACKS
==============================================================================
--- python/branches/py3k/Misc/ACKS	(original)
+++ python/branches/py3k/Misc/ACKS	Wed Jul 28 10:19:35 2010
@@ -448,8 +448,8 @@
 Vladimir Kushnir
 Cameron Laird
 Torsten Landschoff
-Tino Lange
 ?ukasz Langa
+Tino Lange
 Andrew Langmead
 Detlef Lannert
 Soren Larsen

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Wed Jul 28 10:19:35 2010
@@ -475,6 +475,8 @@
 Library
 -------
 
+- Issue #9354: Provide getsockopt() in asyncore's file_wrapper.
+
 - Issue #8966: ctypes: Remove implicit bytes-unicode conversion.
 
 - Issue #9378: python -m pickle <pickle file> will now load and

From python-checkins at python.org  Wed Jul 28 15:13:47 2010
From: python-checkins at python.org (georg.brandl)
Date: Wed, 28 Jul 2010 15:13:47 +0200 (CEST)
Subject: [Python-checkins] r83202 - in python/branches/py3k:
	Doc/library/configparser.rst Lib/configparser.py
	Lib/test/test_cfgparser.py Misc/NEWS
Message-ID: <20100728131347.3B686EE98A@mail.python.org>

Author: georg.brandl
Date: Wed Jul 28 15:13:46 2010
New Revision: 83202

Log:
#1682942: add some ConfigParser features: alternate delimiters, alternate comments, empty lines in values.  Also enhance the docs with more examples and mention SafeConfigParser before ConfigParser.  Patch by Lukas Langa, review by myself, Eric and Ezio.

Modified:
   python/branches/py3k/Doc/library/configparser.rst
   python/branches/py3k/Lib/configparser.py
   python/branches/py3k/Lib/test/test_cfgparser.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Doc/library/configparser.rst
==============================================================================
--- python/branches/py3k/Doc/library/configparser.rst	(original)
+++ python/branches/py3k/Doc/library/configparser.rst	Wed Jul 28 15:13:46 2010
@@ -15,105 +15,168 @@
    single: ini file
    single: Windows ini file
 
-This module defines the class :class:`ConfigParser`.   The :class:`ConfigParser`
-class implements a basic configuration file parser language which provides a
-structure similar to what you would find on Microsoft Windows INI files.  You
-can use this to write Python programs which can be customized by end users
-easily.
+This module provides the classes :class:`RawConfigParser` and
+:class:`SafeConfigParser`.  They implement a basic configuration file parser
+language which provides a structure similar to what you would find in Microsoft
+Windows INI files.  You can use this to write Python programs which can be
+customized by end users easily.
 
 .. note::
 
    This library does *not* interpret or write the value-type prefixes used in
    the Windows Registry extended version of INI syntax.
 
-The configuration file consists of sections, led by a ``[section]`` header and
-followed by ``name: value`` entries, with continuations in the style of
-:rfc:`822` (see section 3.1.1, "LONG HEADER FIELDS"); ``name=value`` is also
-accepted.  Note that leading whitespace is removed from values. The optional
-values can contain format strings which refer to other values in the same
-section, or values in a special ``DEFAULT`` section.  Additional defaults can be
-provided on initialization and retrieval.  Lines beginning with ``'#'`` or
-``';'`` are ignored and may be used to provide comments.
+A configuration file consists of sections, each led by a ``[section]`` header,
+followed by name/value entries separated by a specific string (``=`` or ``:`` by
+default).  Note that leading whitespace is removed from values.  Values can be
+ommitted, in which case the key/value delimiter may also be left out.  Values
+can also span multiple lines, as long as they are indented deeper than the first
+line of the value.  Depending on the parser's mode, blank lines may be treated
+as parts of multiline values or ignored.
+
+Configuration files may include comments, prefixed by specific characters (``#``
+and ``;`` by default).  Comments may appear on their own in an otherwise empty
+line, or may be entered in lines holding values or spection names.  In the
+latter case, they need to be preceded by a whitespace character to be recognized
+as a comment.  (For backwards compatibility, by default only ``;`` starts an
+inline comment, while ``#`` does not.)
+
+On top of the core functionality, :class:`SafeConfigParser` supports
+interpolation.  This means values can contain format strings which refer to
+other values in the same section, or values in a special ``DEFAULT`` section.
+Additional defaults can be provided on initialization and retrieval.
 
 For example::
 
-   [My Section]
-   foodir: %(dir)s/whatever
-   dir=frob
-   long: this value continues
-      in the next line
-
-would resolve the ``%(dir)s`` to the value of ``dir`` (``frob`` in this case).
-All reference expansions are done on demand.
-
-Default values can be specified by passing them into the :class:`ConfigParser`
-constructor as a dictionary.  Additional defaults  may be passed into the
-:meth:`get` method which will override all others.
-
-Sections are normally stored in a built-in dictionary. An alternative dictionary
-type can be passed to the :class:`ConfigParser` constructor. For example, if a
-dictionary type is passed that sorts its keys, the sections will be sorted on
-write-back, as will be the keys within each section.
+   [Paths]
+   home_dir: /Users
+   my_dir: %(home_dir)s/lumberjack
+   my_pictures: %(my_dir)s/Pictures
+
+   [Multiline Values]
+   chorus: I'm a lumberjack, and I'm okay
+      I sleep all night and I work all day
+
+   [No Values]
+   key_without_value
+   empty string value here =
+
+   [You can use comments] ; after a useful line
+   ; in an empty line
+   after: a_value ; here's another comment
+   inside: a         ;comment
+           multiline ;comment
+           value!    ;comment
+
+      [Sections Can Be Indented]
+         can_values_be_as_well = True
+         does_that_mean_anything_special = False
+         purpose = formatting for readability
+         multiline_values = are
+            handled just fine as
+            long as they are indented
+            deeper than the first line
+            of a value
+         # Did I mention we can indent comments, too?
+
+
+In the example above, :class:`SafeConfigParser` would resolve ``%(home_dir)s``
+to the value of ``home_dir`` (``/Users`` in this case).  ``%(my_dir)s`` in
+effect would resolve to ``/Users/lumberjack``.  All interpolations are done on
+demand so keys used in the chain of references do not have to be specified in
+any specific order in the configuration file.
+
+:class:`RawConfigParser` would simply return ``%(my_dir)s/Pictures`` as the
+value of ``my_pictures`` and ``%(home_dir)s/lumberjack`` as the value of
+``my_dir``.  Other features presented in the example are handled in the same
+manner by both parsers.
+
+Default values can be specified by passing them as a dictionary when
+constructing the :class:`SafeConfigParser`.  Additional defaults may be passed
+to the :meth:`get` method which will override all others.
+
+Sections are normally stored in an :class:`collections.OrderedDict` which
+maintains the order of all keys.  An alternative dictionary type can be passed
+to the :meth:`__init__` method.  For example, if a dictionary type is passed
+that sorts its keys, the sections will be sorted on write-back, as will be the
+keys within each section.
 
 
-.. class:: RawConfigParser(defaults=None, dict_type=collections.OrderedDict,
-                           allow_no_value=False)
+.. class:: RawConfigParser(defaults=None, dict_type=collections.OrderedDict, delimiters=('=', ':'), comment_prefixes=_COMPATIBLE, empty_lines_in_values=True, allow_no_value=False)
 
    The basic configuration object.  When *defaults* is given, it is initialized
-   into the dictionary of intrinsic defaults.  When *dict_type* is given, it will
-   be used to create the dictionary objects for the list of sections, for the
-   options within a section, and for the default values.  When *allow_no_value*
-   is true (default: ``False``), options without values are accepted; the value
+   into the dictionary of intrinsic defaults.  When *dict_type* is given, it
+   will be used to create the dictionary objects for the list of sections, for
+   the options within a section, and for the default values.
+
+   When *delimiters* is given, it will be used as the set of substrings that
+   divide keys from values.  When *comment_prefixes* is given, it will be used
+   as the set of substrings that prefix comments in a line, both for the whole
+   line and inline comments.  For backwards compatibility, the default value for
+   *comment_prefixes* is a special value that indicates that ``;`` and ``#`` can
+   start whole line comments while only ``;`` can start inline comments.
+
+   When *empty_lines_in_values* is ``False`` (default: ``True``), each empty
+   line marks the end of an option.  Otherwise, internal empty lines of a
+   multiline option are kept as part of the value.  When *allow_no_value* is
+   true (default: ``False``), options without values are accepted; the value
    presented for these is ``None``.
 
-   This class does not
-   support the magical interpolation behavior.
+   This class does not support the magical interpolation behavior.
 
    .. versionchanged:: 3.1
       The default *dict_type* is :class:`collections.OrderedDict`.
 
    .. versionchanged:: 3.2
-      *allow_no_value* was added.
+      *delimiters*, *comment_prefixes*, *empty_lines_in_values* and
+      *allow_no_value* were added.
 
 
-.. class:: ConfigParser(defaults=None, dict_type=collections.OrderedDict,
-                        allow_no_value=False)
+.. class:: SafeConfigParser(defaults=None, dict_type=collections.OrderedDict, delimiters=('=', ':'), comment_prefixes=('#', ';'), empty_lines_in_values=True, allow_no_value=False)
 
-   Derived class of :class:`RawConfigParser` that implements the magical
-   interpolation feature and adds optional arguments to the :meth:`get` and
-   :meth:`items` methods.  The values in *defaults* must be appropriate for the
-   ``%()s`` string interpolation.  Note that *__name__* is an intrinsic default;
-   its value is the section name, and will override any value provided in
-   *defaults*.
+   Derived class of :class:`ConfigParser` that implements a sane variant of the
+   magical interpolation feature.  This implementation is more predictable as it
+   validates the interpolation syntax used within a configuration file.  This
+   class also enables escaping the interpolation character (e.g. a key can have
+   ``%`` as part of the value by specifying ``%%`` in the file).
 
-   All option names used in interpolation will be passed through the
-   :meth:`optionxform` method just like any other option name reference.  For
-   example, using the default implementation of :meth:`optionxform` (which converts
-   option names to lower case), the values ``foo %(bar)s`` and ``foo %(BAR)s`` are
-   equivalent.
+   Applications that don't require interpolation should use
+   :class:`RawConfigParser`, otherwise :class:`SafeConfigParser` is the best
+   option.
 
    .. versionchanged:: 3.1
       The default *dict_type* is :class:`collections.OrderedDict`.
 
    .. versionchanged:: 3.2
-      *allow_no_value* was added.
+      *delimiters*, *comment_prefixes*, *empty_lines_in_values* and
+      *allow_no_value* were added.
 
 
-.. class:: SafeConfigParser(defaults=None, dict_type=collections.OrderedDict,
-                            allow_no_value=False)
+.. class:: ConfigParser(defaults=None, dict_type=collections.OrderedDict, delimiters=('=', ':'), comment_prefixes=('#', ';'), empty_lines_in_values=True, allow_no_value=False)
 
-   Derived class of :class:`ConfigParser` that implements a more-sane variant of
-   the magical interpolation feature.  This implementation is more predictable as
-   well. New applications should prefer this version if they don't need to be
-   compatible with older versions of Python.
+   Derived class of :class:`RawConfigParser` that implements the magical
+   interpolation feature and adds optional arguments to the :meth:`get` and
+   :meth:`items` methods.
 
-   .. XXX Need to explain what's safer/more predictable about it.
+   :class:`SafeConfigParser` is generally recommended over this class if you
+   need interpolation.
+
+   The values in *defaults* must be appropriate for the ``%()s`` string
+   interpolation.  Note that *__name__* is an intrinsic default; its value is
+   the section name, and will override any value provided in *defaults*.
+
+   All option names used in interpolation will be passed through the
+   :meth:`optionxform` method just like any other option name reference.  For
+   example, using the default implementation of :meth:`optionxform` (which
+   converts option names to lower case), the values ``foo %(bar)s`` and ``foo
+   %(BAR)s`` are equivalent.
 
    .. versionchanged:: 3.1
       The default *dict_type* is :class:`collections.OrderedDict`.
 
    .. versionchanged:: 3.2
-      *allow_no_value* was added.
+      *delimiters*, *comment_prefixes*, *empty_lines_in_values* and
+      *allow_no_value* were added.
 
 
 .. exception:: NoSectionError
@@ -295,11 +358,13 @@
    interpolation and output to files) can only be achieved using string values.
 
 
-.. method:: RawConfigParser.write(fileobject)
+.. method:: RawConfigParser.write(fileobject, space_around_delimiters=True)
 
    Write a representation of the configuration to the specified file object,
    which must be opened in text mode (accepting strings).  This representation
-   can be parsed by a future :meth:`read` call.
+   can be parsed by a future :meth:`read` call. If ``space_around_delimiters``
+   is ``True`` (the default), delimiters between keys and values are surrounded
+   by spaces.
 
 
 .. method:: RawConfigParser.remove_option(section, option)
@@ -342,21 +407,24 @@
 --------------------
 
 The :class:`ConfigParser` class extends some methods of the
-:class:`RawConfigParser` interface, adding some optional arguments.
+:class:`RawConfigParser` interface, adding some optional arguments. Whenever you
+can, consider using :class:`SafeConfigParser` which adds validation and escaping
+for the interpolation.
 
 
 .. method:: ConfigParser.get(section, option, raw=False, vars=None)
 
-   Get an *option* value for the named *section*.  All the ``'%'`` interpolations
-   are expanded in the return values, based on the defaults passed into the
-   constructor, as well as the options *vars* provided, unless the *raw* argument
-   is true.
+   Get an *option* value for the named *section*.  All the ``'%'``
+   interpolations are expanded in the return values, based on the defaults
+   passed into the :meth:`__init__` method, as well as the options *vars*
+   provided, unless the *raw* argument is true.
 
 
 .. method:: ConfigParser.items(section, raw=False, vars=None)
 
-   Return a list of ``(name, value)`` pairs for each option in the given *section*.
-   Optional arguments have the same meaning as for the :meth:`get` method.
+   Return a list of ``(name, value)`` pairs for each option in the given
+   *section*.  Optional arguments have the same meaning as for the :meth:`get`
+   method.
 
 
 .. _safeconfigparser-objects:
@@ -466,8 +534,8 @@
 
 Some configuration files are known to include settings without values, but which
 otherwise conform to the syntax supported by :mod:`configparser`.  The
-*allow_no_value* parameter to the constructor can be used to indicate that such
-values should be accepted:
+*allow_no_value* parameter to the :meth:`__init__` method can be used to
+indicate that such values should be accepted:
 
 .. doctest::
 
@@ -476,12 +544,12 @@
 
    >>> sample_config = """
    ... [mysqld]
-   ... user = mysql
-   ... pid-file = /var/run/mysqld/mysqld.pid
-   ... skip-external-locking
-   ... old_passwords = 1
-   ... skip-bdb
-   ... skip-innodb
+   ...   user = mysql
+   ...   pid-file = /var/run/mysqld/mysqld.pid
+   ...   skip-external-locking
+   ...   old_passwords = 1
+   ...   skip-bdb
+   ...   skip-innodb # we don't need ACID today
    ... """
    >>> config = configparser.RawConfigParser(allow_no_value=True)
    >>> config.readfp(io.BytesIO(sample_config))

Modified: python/branches/py3k/Lib/configparser.py
==============================================================================
--- python/branches/py3k/Lib/configparser.py	(original)
+++ python/branches/py3k/Lib/configparser.py	Wed Jul 28 15:13:46 2010
@@ -1,6 +1,6 @@
 """Configuration file parser.
 
-A setup file consists of sections, lead by a "[section]" header,
+A configuration file consists of sections, lead by a "[section]" header,
 and followed by "name: value" entries, with continuations and such in
 the style of RFC 822.
 
@@ -24,67 +24,88 @@
 
     methods:
 
-    __init__(defaults=None)
-        create the parser and specify a dictionary of intrinsic defaults.  The
-        keys must be strings, the values must be appropriate for %()s string
-        interpolation.  Note that `__name__' is always an intrinsic default;
-        its value is the section's name.
+    __init__(defaults=None, dict_type=_default_dict,
+             delimiters=('=', ':'), comment_prefixes=('#', ';'),
+             empty_lines_in_values=True, allow_no_value=False):
+        Create the parser. When `defaults' is given, it is initialized into the
+        dictionary or intrinsic defaults. The keys must be strings, the values
+        must be appropriate for %()s string interpolation.  Note that `__name__'
+        is always an intrinsic default; its value is the section's name.
+
+        When `dict_type' is given, it will be used to create the dictionary
+        objects for the list of sections, for the options within a section, and
+        for the default values.
+
+        When `delimiters' is given, it will be used as the set of substrings
+        that divide keys from values.
+
+        When `comment_prefixes' is given, it will be used as the set of
+        substrings that prefix comments in a line.
+
+        When `empty_lines_in_values' is False (default: True), each empty line
+        marks the end of an option. Otherwise, internal empty lines of
+        a multiline option are kept as part of the value.
+
+        When `allow_no_value' is True (default: False), options without
+        values are accepted; the value presented for these is None.
 
     sections()
-        return all the configuration section names, sans DEFAULT
+        Return all the configuration section names, sans DEFAULT.
 
     has_section(section)
-        return whether the given section exists
+        Return whether the given section exists.
 
     has_option(section, option)
-        return whether the given option exists in the given section
+        Return whether the given option exists in the given section.
 
     options(section)
-        return list of configuration options for the named section
+        Return list of configuration options for the named section.
 
     read(filenames)
-        read and parse the list of named configuration files, given by
+        Read and parse the list of named configuration files, given by
         name.  A single filename is also allowed.  Non-existing files
         are ignored.  Return list of successfully read files.
 
     readfp(fp, filename=None)
-        read and parse one configuration file, given as a file object.
+        Read and parse one configuration file, given as a file object.
         The filename defaults to fp.name; it is only used in error
         messages (if fp has no `name' attribute, the string `<???>' is used).
 
     get(section, option, raw=False, vars=None)
-        return a string value for the named option.  All % interpolations are
+        Return a string value for the named option.  All % interpolations are
         expanded in the return values, based on the defaults passed into the
         constructor and the DEFAULT section.  Additional substitutions may be
         provided using the `vars' argument, which must be a dictionary whose
         contents override any pre-existing defaults.
 
     getint(section, options)
-        like get(), but convert value to an integer
+        Like get(), but convert value to an integer.
 
     getfloat(section, options)
-        like get(), but convert value to a float
+        Like get(), but convert value to a float.
 
     getboolean(section, options)
-        like get(), but convert value to a boolean (currently case
+        Like get(), but convert value to a boolean (currently case
         insensitively defined as 0, false, no, off for False, and 1, true,
         yes, on for True).  Returns False or True.
 
     items(section, raw=False, vars=None)
-        return a list of tuples with (name, value) for each option
+        Return a list of tuples with (name, value) for each option
         in the section.
 
     remove_section(section)
-        remove the given file section and all its options
+        Remove the given file section and all its options.
 
     remove_option(section, option)
-        remove the given option from the given section
+        Remove the given option from the given section.
 
     set(section, option, value)
-        set the given option
+        Set the given option.
 
-    write(fp)
-        write the configuration state in .ini format
+    write(fp, space_around_delimiters=True)
+        Write the configuration state in .ini format. If
+        `space_around_delimiters' is True (the default), delimiters
+        between keys and values are surrounded by spaces.
 """
 
 try:
@@ -94,6 +115,7 @@
     _default_dict = dict
 
 import re
+import sys
 
 __all__ = ["NoSectionError", "DuplicateSectionError", "NoOptionError",
            "InterpolationError", "InterpolationDepthError",
@@ -114,17 +136,19 @@
 
     def _get_message(self):
         """Getter for 'message'; needed only to override deprecation in
-        BaseException."""
+        BaseException.
+        """
         return self.__message
 
     def _set_message(self, value):
         """Setter for 'message'; needed only to override deprecation in
-        BaseException."""
+        BaseException.
+        """
         self.__message = value
 
     # BaseException.message has been deprecated since Python 2.6.  To prevent
-    # DeprecationWarning from popping up over this pre-existing attribute, use
-    # a new property that takes lookup precedence.
+    # DeprecationWarning from popping up over this pre-existing attribute, use a
+    # new property that takes lookup precedence.
     message = property(_get_message, _set_message)
 
     def __init__(self, msg=''):
@@ -136,6 +160,7 @@
 
     __str__ = __repr__
 
+
 class NoSectionError(Error):
     """Raised when no section matches a requested option."""
 
@@ -144,6 +169,7 @@
         self.section = section
         self.args = (section, )
 
+
 class DuplicateSectionError(Error):
     """Raised when a section is multiply-created."""
 
@@ -152,6 +178,7 @@
         self.section = section
         self.args = (section, )
 
+
 class NoOptionError(Error):
     """A requested option was not found."""
 
@@ -162,6 +189,7 @@
         self.section = section
         self.args = (option, section)
 
+
 class InterpolationError(Error):
     """Base class for interpolation-related exceptions."""
 
@@ -171,6 +199,7 @@
         self.section = section
         self.args = (option, section, msg)
 
+
 class InterpolationMissingOptionError(InterpolationError):
     """A string substitution required a setting which was not available."""
 
@@ -185,10 +214,12 @@
         self.reference = reference
         self.args = (option, section, rawval, reference)
 
+
 class InterpolationSyntaxError(InterpolationError):
     """Raised when the source text into which substitutions are made
     does not conform to the required syntax."""
 
+
 class InterpolationDepthError(InterpolationError):
     """Raised when substitutions are nested too deeply."""
 
@@ -201,6 +232,7 @@
         InterpolationError.__init__(self, option, section, msg)
         self.args = (option, section, rawval)
 
+
 class ParsingError(Error):
     """Raised when a configuration file does not follow legal syntax."""
 
@@ -214,6 +246,7 @@
         self.errors.append((lineno, line))
         self.message += '\n\t[line %2d]: %s' % (lineno, line)
 
+
 class MissingSectionHeaderError(ParsingError):
     """Raised when a key-value pair is found before any section header."""
 
@@ -227,19 +260,74 @@
         self.line = line
         self.args = (filename, lineno, line)
 
+
 class RawConfigParser:
+    """ConfigParser that does not do interpolation."""
+
+    # Regular expressions for parsing section headers and options
+    _SECT_TMPL = r"""
+        \[                                 # [
+        (?P<header>[^]]+)                  # very permissive!
+        \]                                 # ]
+        """
+    _OPT_TMPL = r"""
+        (?P<option>.*?)                    # very permissive!
+        \s*(?P<vi>{delim})\s*              # any number of space/tab,
+                                           # followed by any of the
+                                           # allowed delimiters,
+                                           # followed by any space/tab
+        (?P<value>.*)$                     # everything up to eol
+        """
+    _OPT_NV_TMPL = r"""
+        (?P<option>.*?)                    # very permissive!
+        \s*(?:                             # any number of space/tab,
+        (?P<vi>{delim})\s*                 # optionally followed by
+                                           # any of the allowed
+                                           # delimiters, followed by any
+                                           # space/tab
+        (?P<value>.*))?$                   # everything up to eol
+        """
+
+    # Compiled regular expression for matching sections
+    SECTCRE = re.compile(_SECT_TMPL, re.VERBOSE)
+    # Compiled regular expression for matching options with typical separators
+    OPTCRE = re.compile(_OPT_TMPL.format(delim="=|:"), re.VERBOSE)
+    # Compiled regular expression for matching options with optional values
+    # delimited using typical separators
+    OPTCRE_NV = re.compile(_OPT_NV_TMPL.format(delim="=|:"), re.VERBOSE)
+    # Compiled regular expression for matching leading whitespace in a line
+    NONSPACECRE = re.compile(r"\S")
+    # Select backwards-compatible inline comment character behavior
+    # (; and # are comments at the start of a line, but ; only inline)
+    _COMPATIBLE = object()
+
     def __init__(self, defaults=None, dict_type=_default_dict,
-                 allow_no_value=False):
+                 delimiters=('=', ':'), comment_prefixes=_COMPATIBLE,
+                 empty_lines_in_values=True, allow_no_value=False):
         self._dict = dict_type
         self._sections = self._dict()
         self._defaults = self._dict()
-        if allow_no_value:
-            self._optcre = self.OPTCRE_NV
-        else:
-            self._optcre = self.OPTCRE
         if defaults:
             for key, value in defaults.items():
                 self._defaults[self.optionxform(key)] = value
+        self._delimiters = tuple(delimiters)
+        if delimiters == ('=', ':'):
+            self._optcre = self.OPTCRE_NV if allow_no_value else self.OPTCRE
+        else:
+            delim = "|".join(re.escape(d) for d in delimiters)
+            if allow_no_value:
+                self._optcre = re.compile(self._OPT_NV_TMPL.format(delim=delim),
+                                          re.VERBOSE)
+            else:
+                self._optcre = re.compile(self._OPT_TMPL.format(delim=delim),
+                                          re.VERBOSE)
+        if comment_prefixes is self._COMPATIBLE:
+            self._startonly_comment_prefixes = ('#',)
+            self._comment_prefixes = (';',)
+        else:
+            self._startonly_comment_prefixes = ()
+            self._comment_prefixes = tuple(comment_prefixes or ())
+        self._empty_lines_in_values = empty_lines_in_values
 
     def defaults(self):
         return self._defaults
@@ -313,7 +401,6 @@
         second argument is the `filename', which if not given, is
         taken from fp.name.  If fp has no `name' attribute, `<???>' is
         used.
-
         """
         if filename is None:
             try:
@@ -374,6 +461,7 @@
 
     def has_option(self, section, option):
         """Check for the existence of a given option in a given section."""
+
         if not section or section == DEFAULTSECT:
             option = self.optionxform(option)
             return option in self._defaults
@@ -386,6 +474,7 @@
 
     def set(self, section, option, value=None):
         """Set an option."""
+
         if not section or section == DEFAULTSECT:
             sectdict = self._defaults
         else:
@@ -395,22 +484,34 @@
                 raise NoSectionError(section)
         sectdict[self.optionxform(option)] = value
 
-    def write(self, fp):
-        """Write an .ini-format representation of the configuration state."""
+    def write(self, fp, space_around_delimiters=True):
+        """Write an .ini-format representation of the configuration state.
+
+        If `space_around_delimiters' is True (the default), delimiters
+        between keys and values are surrounded by spaces.
+        """
+        if space_around_delimiters:
+            d = " {} ".format(self._delimiters[0])
+        else:
+            d = self._delimiters[0]
         if self._defaults:
-            fp.write("[%s]\n" % DEFAULTSECT)
-            for (key, value) in self._defaults.items():
-                fp.write("%s = %s\n" % (key, str(value).replace('\n', '\n\t')))
-            fp.write("\n")
+            self._write_section(fp, DEFAULTSECT, self._defaults.items(), d)
         for section in self._sections:
-            fp.write("[%s]\n" % section)
-            for (key, value) in self._sections[section].items():
-                if key == "__name__":
-                    continue
-                if value is not None:
-                    key = " = ".join((key, str(value).replace('\n', '\n\t')))
-                fp.write("%s\n" % (key))
-            fp.write("\n")
+            self._write_section(fp, section,
+                                self._sections[section].items(), d)
+
+    def _write_section(self, fp, section_name, section_items, delimiter):
+        """Write a single section to the specified `fp'."""
+        fp.write("[{}]\n".format(section_name))
+        for key, value in section_items:
+            if key == "__name__":
+                continue
+            if value is not None:
+                value = delimiter + str(value).replace('\n', '\n\t')
+            else:
+                value = ""
+            fp.write("{}{}\n".format(key, value))
+        fp.write("\n")
 
     def remove_option(self, section, option):
         """Remove an option."""
@@ -434,66 +535,63 @@
             del self._sections[section]
         return existed
 
-    #
-    # Regular expressions for parsing section headers and options.
-    #
-    SECTCRE = re.compile(
-        r'\['                                 # [
-        r'(?P<header>[^]]+)'                  # very permissive!
-        r'\]'                                 # ]
-        )
-    OPTCRE = re.compile(
-        r'(?P<option>[^:=\s][^:=]*)'          # very permissive!
-        r'\s*(?P<vi>[:=])\s*'                 # any number of space/tab,
-                                              # followed by separator
-                                              # (either : or =), followed
-                                              # by any # space/tab
-        r'(?P<value>.*)$'                     # everything up to eol
-        )
-    OPTCRE_NV = re.compile(
-        r'(?P<option>[^:=\s][^:=]*)'          # very permissive!
-        r'\s*(?:'                             # any number of space/tab,
-        r'(?P<vi>[:=])\s*'                    # optionally followed by
-                                              # separator (either : or
-                                              # =), followed by any #
-                                              # space/tab
-        r'(?P<value>.*))?$'                   # everything up to eol
-        )
-
     def _read(self, fp, fpname):
-        """Parse a sectioned setup file.
+        """Parse a sectioned configuration file.
 
-        The sections in setup file contains a title line at the top,
-        indicated by a name in square brackets (`[]'), plus key/value
-        options lines, indicated by `name: value' format lines.
-        Continuations are represented by an embedded newline then
-        leading whitespace.  Blank lines, lines beginning with a '#',
-        and just about everything else are ignored.
+        Each section in a configuration file contains a header, indicated by a
+        name in square brackets (`[]'), plus key/value options, indicated by
+        `name' and `value' delimited with a specific substring (`=' or `:' by
+        default).
+
+        Values can span multiple lines, as long as they are indented deeper than
+        the first line of the value. Depending on the parser's mode, blank lines
+        may be treated as parts of multiline values or ignored.
+
+        Configuration files may include comments, prefixed by specific
+        characters (`#' and `;' by default). Comments may appear on their own in
+        an otherwise empty line or may be entered in lines holding values or
+        section names.
         """
         cursect = None                        # None, or a dictionary
         optname = None
         lineno = 0
+        indent_level = 0
         e = None                              # None, or an exception
-        while True:
-            line = fp.readline()
-            if not line:
-                break
-            lineno = lineno + 1
-            # comment or blank line?
-            if line.strip() == '' or line[0] in '#;':
-                continue
-            if line.split(None, 1)[0].lower() == 'rem' and line[0] in "rR":
-                # no leading whitespace
+        for lineno, line in enumerate(fp, start=1):
+            # strip prefix-only comments
+            comment_start = None
+            for prefix in self._startonly_comment_prefixes:
+                if line.strip().startswith(prefix):
+                    comment_start = 0
+                    break
+            # strip inline comments
+            for prefix in self._comment_prefixes:
+                index = line.find(prefix)
+                if index == 0 or (index > 0 and line[index-1].isspace()):
+                    comment_start = index
+                    break
+            value = line[:comment_start].strip()
+            if not value:
+                if self._empty_lines_in_values and comment_start is None:
+                    # add empty line to the value, but only if there was no
+                    # comment on the line
+                    if cursect is not None and optname:
+                        cursect[optname].append('\n')
+                else:
+                    # empty line marks end of value
+                    indent_level = sys.maxsize
                 continue
             # continuation line?
-            if line[0].isspace() and cursect is not None and optname:
-                value = line.strip()
-                if value:
-                    cursect[optname].append(value)
+            first_nonspace = self.NONSPACECRE.search(line)
+            cur_indent_level = first_nonspace.start() if first_nonspace else 0
+            if (cursect is not None and optname and
+                cur_indent_level > indent_level):
+                cursect[optname].append(value)
             # a section header or option header?
             else:
+                indent_level = cur_indent_level
                 # is it a section header?
-                mo = self.SECTCRE.match(line)
+                mo = self.SECTCRE.match(value)
                 if mo:
                     sectname = mo.group('header')
                     if sectname in self._sections:
@@ -511,19 +609,15 @@
                     raise MissingSectionHeaderError(fpname, lineno, line)
                 # an option line?
                 else:
-                    mo = self._optcre.match(line)
+                    mo = self._optcre.match(value)
                     if mo:
                         optname, vi, optval = mo.group('option', 'vi', 'value')
+                        if not optname:
+                            e = self._handle_error(e, fpname, lineno, line)
                         optname = self.optionxform(optname.rstrip())
                         # This check is fine because the OPTCRE cannot
                         # match if it would set optval to None
                         if optval is not None:
-                            if vi in ('=', ':') and ';' in optval:
-                                # ';' is a comment delimiter only if it follows
-                                # a spacing character
-                                pos = optval.find(';')
-                                if pos != -1 and optval[pos-1].isspace():
-                                    optval = optval[:pos]
                             optval = optval.strip()
                             # allow empty values
                             if optval == '""':
@@ -533,26 +627,35 @@
                             # valueless option handling
                             cursect[optname] = optval
                     else:
-                        # a non-fatal parsing error occurred.  set up the
+                        # a non-fatal parsing error occurred. set up the
                         # exception but keep going. the exception will be
                         # raised at the end of the file and will contain a
                         # list of all bogus lines
-                        if not e:
-                            e = ParsingError(fpname)
-                        e.append(lineno, repr(line))
+                        e = self._handle_error(e, fpname, lineno, line)
         # if any parsing errors occurred, raise an exception
         if e:
             raise e
+        self._join_multiline_values()
 
-        # join the multi-line values collected while reading
+    def _join_multiline_values(self):
         all_sections = [self._defaults]
         all_sections.extend(self._sections.values())
         for options in all_sections:
             for name, val in options.items():
                 if isinstance(val, list):
+                    if val[-1] == '\n':
+                        val = val[:-1]
                     options[name] = '\n'.join(val)
 
+    def _handle_error(self, exc, fpname, lineno, line):
+        if not exc:
+            exc = ParsingError(fpname)
+        exc.append(lineno, repr(line))
+        return exc
+
+
 class ConfigParser(RawConfigParser):
+    """ConfigParser implementing interpolation."""
 
     def get(self, section, option, raw=False, vars=None):
         """Get an option value for a given section.
@@ -648,6 +751,7 @@
 
 
 class SafeConfigParser(ConfigParser):
+    """ConfigParser implementing sane interpolation."""
 
     def _interpolate(self, section, option, rawval, vars):
         # do the string interpolation

Modified: python/branches/py3k/Lib/test/test_cfgparser.py
==============================================================================
--- python/branches/py3k/Lib/test/test_cfgparser.py	(original)
+++ python/branches/py3k/Lib/test/test_cfgparser.py	Wed Jul 28 15:13:46 2010
@@ -3,6 +3,7 @@
 import io
 import os
 import unittest
+import textwrap
 
 from test import support
 
@@ -23,15 +24,26 @@
     def itervalues(self): return iter(self.values())
 
 
-class TestCaseBase(unittest.TestCase):
+class CfgParserTestCaseClass(unittest.TestCase):
     allow_no_value = False
+    delimiters = ('=', ':')
+    comment_prefixes = (';', '#')
+    empty_lines_in_values = True
+    dict_type = configparser._default_dict
 
     def newconfig(self, defaults=None):
+        arguments = dict(
+            allow_no_value=self.allow_no_value,
+            delimiters=self.delimiters,
+            comment_prefixes=self.comment_prefixes,
+            empty_lines_in_values=self.empty_lines_in_values,
+            dict_type=self.dict_type,
+        )
         if defaults is None:
-            self.cf = self.config_class(allow_no_value=self.allow_no_value)
+            self.cf = self.config_class(**arguments)
         else:
             self.cf = self.config_class(defaults,
-                                        allow_no_value=self.allow_no_value)
+                                        **arguments)
         return self.cf
 
     def fromstring(self, string, defaults=None):
@@ -40,27 +52,33 @@
         cf.readfp(sio)
         return cf
 
+class BasicTestCase(CfgParserTestCaseClass):
+
     def test_basic(self):
-        config_string = (
-            "[Foo Bar]\n"
-            "foo=bar\n"
-            "[Spacey Bar]\n"
-            "foo = bar\n"
-            "[Commented Bar]\n"
-            "foo: bar ; comment\n"
-            "[Long Line]\n"
-            "foo: this line is much, much longer than my editor\n"
-            "   likes it.\n"
-            "[Section\\with$weird%characters[\t]\n"
-            "[Internationalized Stuff]\n"
-            "foo[bg]: Bulgarian\n"
-            "foo=Default\n"
-            "foo[en]=English\n"
-            "foo[de]=Deutsch\n"
-            "[Spaces]\n"
-            "key with spaces : value\n"
-            "another with spaces = splat!\n"
-            )
+        config_string = """\
+[Foo Bar]
+foo{0[0]}bar
+[Spacey Bar]
+foo {0[0]} bar
+[Spacey Bar From The Beginning]
+  foo {0[0]} bar
+  baz {0[0]} qwe
+[Commented Bar]
+foo{0[1]} bar {1[1]} comment
+baz{0[0]}qwe {1[0]}another one
+[Long Line]
+foo{0[1]} this line is much, much longer than my editor
+   likes it.
+[Section\\with$weird%characters[\t]
+[Internationalized Stuff]
+foo[bg]{0[1]} Bulgarian
+foo{0[0]}Default
+foo[en]{0[0]}English
+foo[de]{0[0]}Deutsch
+[Spaces]
+key with spaces {0[1]} value
+another with spaces {0[0]} splat!
+""".format(self.delimiters, self.comment_prefixes)
         if self.allow_no_value:
             config_string += (
                 "[NoValue]\n"
@@ -70,13 +88,14 @@
         cf = self.fromstring(config_string)
         L = cf.sections()
         L.sort()
-        E = [r'Commented Bar',
-             r'Foo Bar',
-             r'Internationalized Stuff',
-             r'Long Line',
-             r'Section\with$weird%characters[' '\t',
-             r'Spaces',
-             r'Spacey Bar',
+        E = ['Commented Bar',
+             'Foo Bar',
+             'Internationalized Stuff',
+             'Long Line',
+             'Section\\with$weird%characters[\t',
+             'Spaces',
+             'Spacey Bar',
+             'Spacey Bar From The Beginning',
              ]
         if self.allow_no_value:
             E.append(r'NoValue')
@@ -89,7 +108,10 @@
         # http://www.python.org/sf/583248
         eq(cf.get('Foo Bar', 'foo'), 'bar')
         eq(cf.get('Spacey Bar', 'foo'), 'bar')
+        eq(cf.get('Spacey Bar From The Beginning', 'foo'), 'bar')
+        eq(cf.get('Spacey Bar From The Beginning', 'baz'), 'qwe')
         eq(cf.get('Commented Bar', 'foo'), 'bar')
+        eq(cf.get('Commented Bar', 'baz'), 'qwe')
         eq(cf.get('Spaces', 'key with spaces'), 'value')
         eq(cf.get('Spaces', 'another with spaces'), 'splat!')
         if self.allow_no_value:
@@ -140,12 +162,14 @@
 
         # SF bug #432369:
         cf = self.fromstring(
-            "[MySection]\nOption: first line\n\tsecond line\n")
+            "[MySection]\nOption{} first line\n\tsecond line\n".format(
+                self.delimiters[0]))
         eq(cf.options("MySection"), ["option"])
         eq(cf.get("MySection", "Option"), "first line\nsecond line")
 
         # SF bug #561822:
-        cf = self.fromstring("[section]\nnekey=nevalue\n",
+        cf = self.fromstring("[section]\n"
+                             "nekey{}nevalue\n".format(self.delimiters[0]),
                              defaults={"key":"value"})
         self.assertTrue(cf.has_option("section", "Key"))
 
@@ -162,18 +186,19 @@
 
     def test_parse_errors(self):
         self.newconfig()
-        e = self.parse_error(configparser.ParsingError,
-                             "[Foo]\n  extra-spaces: splat\n")
-        self.assertEqual(e.args, ('<???>',))
-        self.parse_error(configparser.ParsingError,
-                         "[Foo]\n  extra-spaces= splat\n")
         self.parse_error(configparser.ParsingError,
-                         "[Foo]\n:value-without-option-name\n")
+                         "[Foo]\n"
+                         "{}val-without-opt-name\n".format(self.delimiters[0]))
         self.parse_error(configparser.ParsingError,
-                         "[Foo]\n=value-without-option-name\n")
+                         "[Foo]\n"
+                         "{}val-without-opt-name\n".format(self.delimiters[1]))
         e = self.parse_error(configparser.MissingSectionHeaderError,
                              "No Section!\n")
         self.assertEqual(e.args, ('<???>', 1, "No Section!\n"))
+        if not self.allow_no_value:
+            e = self.parse_error(configparser.ParsingError,
+                                "[Foo]\n  wrong-indent\n")
+            self.assertEqual(e.args, ('<???>',))
 
     def parse_error(self, exc, src):
         sio = io.StringIO(src)
@@ -188,9 +213,9 @@
         self.assertFalse(cf.has_section("Foo"),
                          "new ConfigParser should have no acknowledged "
                          "sections")
-        with self.assertRaises(configparser.NoSectionError) as cm:
+        with self.assertRaises(configparser.NoSectionError):
             cf.options("Foo")
-        with self.assertRaises(configparser.NoSectionError) as cm:
+        with self.assertRaises(configparser.NoSectionError):
             cf.set("foo", "bar", "value")
         e = self.get_error(configparser.NoSectionError, "foo", "bar")
         self.assertEqual(e.args, ("foo",))
@@ -210,21 +235,21 @@
     def test_boolean(self):
         cf = self.fromstring(
             "[BOOLTEST]\n"
-            "T1=1\n"
-            "T2=TRUE\n"
-            "T3=True\n"
-            "T4=oN\n"
-            "T5=yes\n"
-            "F1=0\n"
-            "F2=FALSE\n"
-            "F3=False\n"
-            "F4=oFF\n"
-            "F5=nO\n"
-            "E1=2\n"
-            "E2=foo\n"
-            "E3=-1\n"
-            "E4=0.1\n"
-            "E5=FALSE AND MORE"
+            "T1{equals}1\n"
+            "T2{equals}TRUE\n"
+            "T3{equals}True\n"
+            "T4{equals}oN\n"
+            "T5{equals}yes\n"
+            "F1{equals}0\n"
+            "F2{equals}FALSE\n"
+            "F3{equals}False\n"
+            "F4{equals}oFF\n"
+            "F5{equals}nO\n"
+            "E1{equals}2\n"
+            "E2{equals}foo\n"
+            "E3{equals}-1\n"
+            "E4{equals}0.1\n"
+            "E5{equals}FALSE AND MORE".format(equals=self.delimiters[0])
             )
         for x in range(1, 5):
             self.assertTrue(cf.getboolean('BOOLTEST', 't%d' % x))
@@ -242,11 +267,17 @@
     def test_write(self):
         config_string = (
             "[Long Line]\n"
-            "foo: this line is much, much longer than my editor\n"
+            "foo{0[0]} this line is much, much longer than my editor\n"
             "   likes it.\n"
             "[DEFAULT]\n"
-            "foo: another very\n"
+            "foo{0[1]} another very\n"
             " long line\n"
+            "[Long Line - With Comments!]\n"
+            "test {0[1]} we        {comment} can\n"
+            "            also      {comment} place\n"
+            "            comments  {comment} in\n"
+            "            multiline {comment} values"
+            "\n".format(self.delimiters, comment=self.comment_prefixes[0])
             )
         if self.allow_no_value:
             config_string += (
@@ -259,13 +290,19 @@
         cf.write(output)
         expect_string = (
             "[DEFAULT]\n"
-            "foo = another very\n"
+            "foo {equals} another very\n"
             "\tlong line\n"
             "\n"
             "[Long Line]\n"
-            "foo = this line is much, much longer than my editor\n"
+            "foo {equals} this line is much, much longer than my editor\n"
             "\tlikes it.\n"
             "\n"
+            "[Long Line - With Comments!]\n"
+            "test {equals} we\n"
+            "\talso\n"
+            "\tcomments\n"
+            "\tmultiline\n"
+            "\n".format(equals=self.delimiters[0])
             )
         if self.allow_no_value:
             expect_string += (
@@ -277,7 +314,7 @@
 
     def test_set_string_types(self):
         cf = self.fromstring("[sect]\n"
-                             "option1=foo\n")
+                             "option1{eq}foo\n".format(eq=self.delimiters[0]))
         # Check that we don't get an exception when setting values in
         # an existing section using strings:
         class mystr(str):
@@ -290,6 +327,9 @@
         cf.set("sect", "option2", "splat")
 
     def test_read_returns_file_list(self):
+        if self.delimiters[0] != '=':
+            # skip reading the file if we're using an incompatible format
+            return
         file1 = support.findfile("cfgparser.1")
         # check when we pass a mix of readable and non-readable files:
         cf = self.newconfig()
@@ -314,45 +354,45 @@
     def get_interpolation_config(self):
         return self.fromstring(
             "[Foo]\n"
-            "bar=something %(with1)s interpolation (1 step)\n"
-            "bar9=something %(with9)s lots of interpolation (9 steps)\n"
-            "bar10=something %(with10)s lots of interpolation (10 steps)\n"
-            "bar11=something %(with11)s lots of interpolation (11 steps)\n"
-            "with11=%(with10)s\n"
-            "with10=%(with9)s\n"
-            "with9=%(with8)s\n"
-            "with8=%(With7)s\n"
-            "with7=%(WITH6)s\n"
-            "with6=%(with5)s\n"
-            "With5=%(with4)s\n"
-            "WITH4=%(with3)s\n"
-            "with3=%(with2)s\n"
-            "with2=%(with1)s\n"
-            "with1=with\n"
+            "bar{equals}something %(with1)s interpolation (1 step)\n"
+            "bar9{equals}something %(with9)s lots of interpolation (9 steps)\n"
+            "bar10{equals}something %(with10)s lots of interpolation (10 steps)\n"
+            "bar11{equals}something %(with11)s lots of interpolation (11 steps)\n"
+            "with11{equals}%(with10)s\n"
+            "with10{equals}%(with9)s\n"
+            "with9{equals}%(with8)s\n"
+            "with8{equals}%(With7)s\n"
+            "with7{equals}%(WITH6)s\n"
+            "with6{equals}%(with5)s\n"
+            "With5{equals}%(with4)s\n"
+            "WITH4{equals}%(with3)s\n"
+            "with3{equals}%(with2)s\n"
+            "with2{equals}%(with1)s\n"
+            "with1{equals}with\n"
             "\n"
             "[Mutual Recursion]\n"
-            "foo=%(bar)s\n"
-            "bar=%(foo)s\n"
+            "foo{equals}%(bar)s\n"
+            "bar{equals}%(foo)s\n"
             "\n"
             "[Interpolation Error]\n"
-            "name=%(reference)s\n",
+            "name{equals}%(reference)s\n".format(equals=self.delimiters[0]),
             # no definition for 'reference'
             defaults={"getname": "%(__name__)s"})
 
     def check_items_config(self, expected):
         cf = self.fromstring(
             "[section]\n"
-            "name = value\n"
-            "key: |%(name)s| \n"
-            "getdefault: |%(default)s|\n"
-            "getname: |%(__name__)s|",
+            "name {0[0]} value\n"
+            "key{0[1]} |%(name)s| \n"
+            "getdefault{0[1]} |%(default)s|\n"
+            "getname{0[1]} |%(__name__)s|".format(self.delimiters),
             defaults={"default": "<default>"})
         L = list(cf.items("section"))
         L.sort()
         self.assertEqual(L, expected)
 
 
-class ConfigParserTestCase(TestCaseBase):
+class ConfigParserTestCase(BasicTestCase):
     config_class = configparser.ConfigParser
 
     def test_interpolation(self):
@@ -414,7 +454,11 @@
         self.assertRaises(ValueError, cf.get, 'non-string',
                           'string_with_interpolation', raw=False)
 
-class MultilineValuesTestCase(TestCaseBase):
+class ConfigParserTestCaseNonStandardDelimiters(ConfigParserTestCase):
+    delimiters = (':=', '$')
+    comment_prefixes = ('//', '"')
+
+class MultilineValuesTestCase(BasicTestCase):
     config_class = configparser.ConfigParser
     wonderful_spam = ("I'm having spam spam spam spam "
                       "spam spam spam beaked beans spam "
@@ -442,7 +486,7 @@
         self.assertEqual(cf_from_file.get('section8', 'lovely_spam4'),
                          self.wonderful_spam.replace('\t\n', '\n'))
 
-class RawConfigParserTestCase(TestCaseBase):
+class RawConfigParserTestCase(BasicTestCase):
     config_class = configparser.RawConfigParser
 
     def test_interpolation(self):
@@ -476,6 +520,28 @@
                          [0, 1, 1, 2, 3, 5, 8, 13])
         self.assertEqual(cf.get('non-string', 'dict'), {'pi': 3.14159})
 
+class RawConfigParserTestCaseNonStandardDelimiters(RawConfigParserTestCase):
+    delimiters = (':=', '$')
+    comment_prefixes = ('//', '"')
+
+class RawConfigParserTestSambaConf(BasicTestCase):
+    config_class = configparser.RawConfigParser
+    comment_prefixes = ('#', ';', '//', '----')
+    empty_lines_in_values = False
+
+    def test_reading(self):
+        smbconf = support.findfile("cfgparser.2")
+        # check when we pass a mix of readable and non-readable files:
+        cf = self.newconfig()
+        parsed_files = cf.read([smbconf, "nonexistent-file"])
+        self.assertEqual(parsed_files, [smbconf])
+        sections = ['global', 'homes', 'printers',
+                    'print$', 'pdf-generator', 'tmp', 'Agustin']
+        self.assertEqual(cf.sections(), sections)
+        self.assertEqual(cf.get("global", "workgroup"), "MDKGROUP")
+        self.assertEqual(cf.getint("global", "max log size"), 50)
+        self.assertEqual(cf.get("global", "hosts allow"), "127.")
+        self.assertEqual(cf.get("tmp", "echo command"), "cat %s; rm %s")
 
 class SafeConfigParserTestCase(ConfigParserTestCase):
     config_class = configparser.SafeConfigParser
@@ -483,16 +549,17 @@
     def test_safe_interpolation(self):
         # See http://www.python.org/sf/511737
         cf = self.fromstring("[section]\n"
-                             "option1=xxx\n"
-                             "option2=%(option1)s/xxx\n"
-                             "ok=%(option1)s/%%s\n"
-                             "not_ok=%(option2)s/%%s")
+                             "option1{eq}xxx\n"
+                             "option2{eq}%(option1)s/xxx\n"
+                             "ok{eq}%(option1)s/%%s\n"
+                             "not_ok{eq}%(option2)s/%%s".format(
+                                 eq=self.delimiters[0]))
         self.assertEqual(cf.get("section", "ok"), "xxx/%s")
         self.assertEqual(cf.get("section", "not_ok"), "xxx/xxx/%s")
 
     def test_set_malformatted_interpolation(self):
         cf = self.fromstring("[sect]\n"
-                             "option1=foo\n")
+                             "option1{eq}foo\n".format(eq=self.delimiters[0]))
 
         self.assertEqual(cf.get('sect', "option1"), "foo")
 
@@ -508,7 +575,7 @@
 
     def test_set_nonstring_types(self):
         cf = self.fromstring("[sect]\n"
-                             "option1=foo\n")
+                             "option1{eq}foo\n".format(eq=self.delimiters[0]))
         # Check that we get a TypeError when setting non-string values
         # in an existing section:
         self.assertRaises(TypeError, cf.set, "sect", "option1", 1)
@@ -526,15 +593,16 @@
         cf = self.newconfig()
         self.assertRaises(ValueError, cf.add_section, "DEFAULT")
 
+class SafeConfigParserTestCaseNonStandardDelimiters(SafeConfigParserTestCase):
+    delimiters = (':=', '$')
+    comment_prefixes = ('//', '"')
 
 class SafeConfigParserTestCaseNoValue(SafeConfigParserTestCase):
     allow_no_value = True
 
 
 class SortedTestCase(RawConfigParserTestCase):
-    def newconfig(self, defaults=None):
-        self.cf = self.config_class(defaults=defaults, dict_type=SortedDict)
-        return self.cf
+    dict_type = SortedDict
 
     def test_sorted(self):
         self.fromstring("[b]\n"
@@ -556,14 +624,36 @@
                           "o4 = 1\n\n")
 
 
+class CompatibleTestCase(CfgParserTestCaseClass):
+    config_class = configparser.RawConfigParser
+    comment_prefixes = configparser.RawConfigParser._COMPATIBLE
+
+    def test_comment_handling(self):
+        config_string = textwrap.dedent("""\
+        [Commented Bar]
+        baz=qwe ; a comment
+        foo: bar # not a comment!
+        # but this is a comment
+        ; another comment
+        """)
+        cf = self.fromstring(config_string)
+        self.assertEqual(cf.get('Commented Bar', 'foo'), 'bar # not a comment!')
+        self.assertEqual(cf.get('Commented Bar', 'baz'), 'qwe')
+
+
 def test_main():
     support.run_unittest(
         ConfigParserTestCase,
+        ConfigParserTestCaseNonStandardDelimiters,
         MultilineValuesTestCase,
         RawConfigParserTestCase,
+        RawConfigParserTestCaseNonStandardDelimiters,
+        RawConfigParserTestSambaConf,
         SafeConfigParserTestCase,
+        SafeConfigParserTestCaseNonStandardDelimiters,
         SafeConfigParserTestCaseNoValue,
         SortedTestCase,
+        CompatibleTestCase,
         )
 
 

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Wed Jul 28 15:13:46 2010
@@ -475,6 +475,9 @@
 Library
 -------
 
+- Issue #1682942: Improvements to configparser: support alternate
+  delimiters, alternate comment prefixes and empty lines in values.
+
 - Issue #9354: Provide getsockopt() in asyncore's file_wrapper.
 
 - Issue #8966: ctypes: Remove implicit bytes-unicode conversion.

From python-checkins at python.org  Wed Jul 28 16:26:11 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Wed, 28 Jul 2010 16:26:11 +0200 (CEST)
Subject: [Python-checkins] r83204 - in
	python/branches/release27-maint/Lib/test:
	test_profilehooks.py test_sys.py test_sys_setprofile.py
	test_sys_settrace.py
Message-ID: <20100728142611.1C5D0EEAD5@mail.python.org>

Author: alexander.belopolsky
Date: Wed Jul 28 16:26:06 2010
New Revision: 83204

Log:
Reverted r83142: bad svnmerge

Added:
   python/branches/release27-maint/Lib/test/test_profilehooks.py
      - copied unchanged from r83141, /python/branches/release27-maint/Lib/test/test_profilehooks.py
Removed:
   python/branches/release27-maint/Lib/test/test_sys_setprofile.py
   python/branches/release27-maint/Lib/test/test_sys_settrace.py
Modified:
   python/branches/release27-maint/Lib/test/test_sys.py

Modified: python/branches/release27-maint/Lib/test/test_sys.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_sys.py	(original)
+++ python/branches/release27-maint/Lib/test/test_sys.py	Wed Jul 28 16:26:06 2010
@@ -205,8 +205,8 @@
             # can't check more than the type, as the user might have changed it
             self.assertIsInstance(sys.getdefaultencoding(), str)
 
-    # testing sys.settrace() is done in test_sys_settrace.py
-    # testing sys.setprofile() is done in test_sys_setprofile.py
+    # testing sys.settrace() is done in test_trace.py
+    # testing sys.setprofile() is done in test_profile.py
 
     def test_setcheckinterval(self):
         self.assertRaises(TypeError, sys.setcheckinterval)

Deleted: python/branches/release27-maint/Lib/test/test_sys_setprofile.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_sys_setprofile.py	Wed Jul 28 16:26:06 2010
+++ (empty file)
@@ -1,385 +0,0 @@
-import pprint
-import sys
-import unittest
-
-from test import support
-
-class TestGetProfile(unittest.TestCase):
-    def setUp(self):
-        sys.setprofile(None)
-
-    def tearDown(self):
-        sys.setprofile(None)
-
-    def test_empty(self):
-        assert sys.getprofile() is None
-
-    def test_setget(self):
-        def fn(*args):
-            pass
-
-        sys.setprofile(fn)
-        assert sys.getprofile() == fn
-
-class HookWatcher:
-    def __init__(self):
-        self.frames = []
-        self.events = []
-
-    def callback(self, frame, event, arg):
-        if (event == "call"
-            or event == "return"
-            or event == "exception"):
-            self.add_event(event, frame)
-
-    def add_event(self, event, frame=None):
-        """Add an event to the log."""
-        if frame is None:
-            frame = sys._getframe(1)
-
-        try:
-            frameno = self.frames.index(frame)
-        except ValueError:
-            frameno = len(self.frames)
-            self.frames.append(frame)
-
-        self.events.append((frameno, event, ident(frame)))
-
-    def get_events(self):
-        """Remove calls to add_event()."""
-        disallowed = [ident(self.add_event.__func__), ident(ident)]
-        self.frames = None
-
-        return [item for item in self.events if item[2] not in disallowed]
-
-
-class ProfileSimulator(HookWatcher):
-    def __init__(self, testcase):
-        self.testcase = testcase
-        self.stack = []
-        HookWatcher.__init__(self)
-
-    def callback(self, frame, event, arg):
-        # Callback registered with sys.setprofile()/sys.settrace()
-        self.dispatch[event](self, frame)
-
-    def trace_call(self, frame):
-        self.add_event('call', frame)
-        self.stack.append(frame)
-
-    def trace_return(self, frame):
-        self.add_event('return', frame)
-        self.stack.pop()
-
-    def trace_exception(self, frame):
-        self.testcase.fail(
-            "the profiler should never receive exception events")
-
-    def trace_pass(self, frame):
-        pass
-
-    dispatch = {
-        'call': trace_call,
-        'exception': trace_exception,
-        'return': trace_return,
-        'c_call': trace_pass,
-        'c_return': trace_pass,
-        'c_exception': trace_pass,
-        }
-
-
-class TestCaseBase(unittest.TestCase):
-    def check_events(self, callable, expected):
-        events = capture_events(callable, self.new_watcher())
-        if events != expected:
-            self.fail("Expected events:\n%s\nReceived events:\n%s"
-                      % (pprint.pformat(expected), pprint.pformat(events)))
-
-
-class ProfileHookTestCase(TestCaseBase):
-    def new_watcher(self):
-        return HookWatcher()
-
-    def test_simple(self):
-        def f(p):
-            pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_exception(self):
-        def f(p):
-            1/0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_caught_exception(self):
-        def f(p):
-            try: 1/0
-            except: pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_caught_nested_exception(self):
-        def f(p):
-            try: 1/0
-            except: pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_nested_exception(self):
-        def f(p):
-            1/0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              # This isn't what I expected:
-                              # (0, 'exception', protect_ident),
-                              # I expected this again:
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_exception_in_except_clause(self):
-        def f(p):
-            1/0
-        def g(p):
-            try:
-                f(p)
-            except:
-                try: f(p)
-                except: pass
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (3, 'call', f_ident),
-                              (3, 'return', f_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-    def test_exception_propogation(self):
-        def f(p):
-            1/0
-        def g(p):
-            try: f(p)
-            finally: p.add_event("falling through")
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (1, 'falling through', g_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-    def test_raise_twice(self):
-        def f(p):
-            try: 1/0
-            except: 1/0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_raise_reraise(self):
-        def f(p):
-            try: 1/0
-            except: raise
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_raise(self):
-        def f(p):
-            raise Exception()
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_distant_exception(self):
-        def f():
-            1/0
-        def g():
-            f()
-        def h():
-            g()
-        def i():
-            h()
-        def j(p):
-            i()
-        f_ident = ident(f)
-        g_ident = ident(g)
-        h_ident = ident(h)
-        i_ident = ident(i)
-        j_ident = ident(j)
-        self.check_events(j, [(1, 'call', j_ident),
-                              (2, 'call', i_ident),
-                              (3, 'call', h_ident),
-                              (4, 'call', g_ident),
-                              (5, 'call', f_ident),
-                              (5, 'return', f_ident),
-                              (4, 'return', g_ident),
-                              (3, 'return', h_ident),
-                              (2, 'return', i_ident),
-                              (1, 'return', j_ident),
-                              ])
-
-    def test_generator(self):
-        def f():
-            for i in range(2):
-                yield i
-        def g(p):
-            for i in f():
-                pass
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              # call the iterator twice to generate values
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              # once more; returns end-of-iteration with
-                              # actually raising an exception
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-    def test_stop_iteration(self):
-        def f():
-            for i in range(2):
-                yield i
-            raise StopIteration
-        def g(p):
-            for i in f():
-                pass
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              # call the iterator twice to generate values
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              # once more to hit the raise:
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-
-class ProfileSimulatorTestCase(TestCaseBase):
-    def new_watcher(self):
-        return ProfileSimulator(self)
-
-    def test_simple(self):
-        def f(p):
-            pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_basic_exception(self):
-        def f(p):
-            1/0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_caught_exception(self):
-        def f(p):
-            try: 1/0
-            except: pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_distant_exception(self):
-        def f():
-            1/0
-        def g():
-            f()
-        def h():
-            g()
-        def i():
-            h()
-        def j(p):
-            i()
-        f_ident = ident(f)
-        g_ident = ident(g)
-        h_ident = ident(h)
-        i_ident = ident(i)
-        j_ident = ident(j)
-        self.check_events(j, [(1, 'call', j_ident),
-                              (2, 'call', i_ident),
-                              (3, 'call', h_ident),
-                              (4, 'call', g_ident),
-                              (5, 'call', f_ident),
-                              (5, 'return', f_ident),
-                              (4, 'return', g_ident),
-                              (3, 'return', h_ident),
-                              (2, 'return', i_ident),
-                              (1, 'return', j_ident),
-                              ])
-
-
-def ident(function):
-    if hasattr(function, "f_code"):
-        code = function.f_code
-    else:
-        code = function.__code__
-    return code.co_firstlineno, code.co_name
-
-
-def protect(f, p):
-    try: f(p)
-    except: pass
-
-protect_ident = ident(protect)
-
-
-def capture_events(callable, p=None):
-    try:
-        sys.setprofile()
-    except TypeError:
-        pass
-    else:
-        raise support.TestFailed(
-            'sys.setprofile() did not raise TypeError')
-
-    if p is None:
-        p = HookWatcher()
-    sys.setprofile(p.callback)
-    protect(callable, p)
-    sys.setprofile(None)
-    return p.get_events()[1:-1]
-
-
-def show_events(callable):
-    import pprint
-    pprint.pprint(capture_events(callable))
-
-
-def test_main():
-    support.run_unittest(
-        TestGetProfile,
-        ProfileHookTestCase,
-        ProfileSimulatorTestCase
-    )
-
-
-if __name__ == "__main__":
-    test_main()

Deleted: python/branches/release27-maint/Lib/test/test_sys_settrace.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_sys_settrace.py	Wed Jul 28 16:26:06 2010
+++ (empty file)
@@ -1,790 +0,0 @@
-# Testing the line trace facility.
-
-from test import support
-import unittest
-import sys
-import difflib
-import gc
-
-# A very basic example.  If this fails, we're in deep trouble.
-def basic():
-    return 1
-
-basic.events = [(0, 'call'),
-                (1, 'line'),
-                (1, 'return')]
-
-# Many of the tests below are tricky because they involve pass statements.
-# If there is implicit control flow around a pass statement (in an except
-# clause or else caluse) under what conditions do you set a line number
-# following that clause?
-
-
-# The entire "while 0:" statement is optimized away.  No code
-# exists for it, so the line numbers skip directly from "del x"
-# to "x = 1".
-def arigo_example():
-    x = 1
-    del x
-    while 0:
-        pass
-    x = 1
-
-arigo_example.events = [(0, 'call'),
-                        (1, 'line'),
-                        (2, 'line'),
-                        (5, 'line'),
-                        (5, 'return')]
-
-# check that lines consisting of just one instruction get traced:
-def one_instr_line():
-    x = 1
-    del x
-    x = 1
-
-one_instr_line.events = [(0, 'call'),
-                         (1, 'line'),
-                         (2, 'line'),
-                         (3, 'line'),
-                         (3, 'return')]
-
-def no_pop_tops():      # 0
-    x = 1               # 1
-    for a in range(2):  # 2
-        if a:           # 3
-            x = 1       # 4
-        else:           # 5
-            x = 1       # 6
-
-no_pop_tops.events = [(0, 'call'),
-                      (1, 'line'),
-                      (2, 'line'),
-                      (3, 'line'),
-                      (6, 'line'),
-                      (2, 'line'),
-                      (3, 'line'),
-                      (4, 'line'),
-                      (2, 'line'),
-                      (2, 'return')]
-
-def no_pop_blocks():
-    y = 1
-    while not y:
-        bla
-    x = 1
-
-no_pop_blocks.events = [(0, 'call'),
-                        (1, 'line'),
-                        (2, 'line'),
-                        (4, 'line'),
-                        (4, 'return')]
-
-def called(): # line -3
-    x = 1
-
-def call():   # line 0
-    called()
-
-call.events = [(0, 'call'),
-               (1, 'line'),
-               (-3, 'call'),
-               (-2, 'line'),
-               (-2, 'return'),
-               (1, 'return')]
-
-def raises():
-    raise Exception
-
-def test_raise():
-    try:
-        raises()
-    except Exception as exc:
-        x = 1
-
-test_raise.events = [(0, 'call'),
-                     (1, 'line'),
-                     (2, 'line'),
-                     (-3, 'call'),
-                     (-2, 'line'),
-                     (-2, 'exception'),
-                     (-2, 'return'),
-                     (2, 'exception'),
-                     (3, 'line'),
-                     (4, 'line'),
-                     (4, 'return')]
-
-def _settrace_and_return(tracefunc):
-    sys.settrace(tracefunc)
-    sys._getframe().f_back.f_trace = tracefunc
-def settrace_and_return(tracefunc):
-    _settrace_and_return(tracefunc)
-
-settrace_and_return.events = [(1, 'return')]
-
-def _settrace_and_raise(tracefunc):
-    sys.settrace(tracefunc)
-    sys._getframe().f_back.f_trace = tracefunc
-    raise RuntimeError
-def settrace_and_raise(tracefunc):
-    try:
-        _settrace_and_raise(tracefunc)
-    except RuntimeError as exc:
-        pass
-
-settrace_and_raise.events = [(2, 'exception'),
-                             (3, 'line'),
-                             (4, 'line'),
-                             (4, 'return')]
-
-# implicit return example
-# This test is interesting because of the else: pass
-# part of the code.  The code generate for the true
-# part of the if contains a jump past the else branch.
-# The compiler then generates an implicit "return None"
-# Internally, the compiler visits the pass statement
-# and stores its line number for use on the next instruction.
-# The next instruction is the implicit return None.
-def ireturn_example():
-    a = 5
-    b = 5
-    if a == b:
-        b = a+1
-    else:
-        pass
-
-ireturn_example.events = [(0, 'call'),
-                          (1, 'line'),
-                          (2, 'line'),
-                          (3, 'line'),
-                          (4, 'line'),
-                          (6, 'line'),
-                          (6, 'return')]
-
-# Tight loop with while(1) example (SF #765624)
-def tightloop_example():
-    items = range(0, 3)
-    try:
-        i = 0
-        while 1:
-            b = items[i]; i+=1
-    except IndexError:
-        pass
-
-tightloop_example.events = [(0, 'call'),
-                            (1, 'line'),
-                            (2, 'line'),
-                            (3, 'line'),
-                            (4, 'line'),
-                            (5, 'line'),
-                            (5, 'line'),
-                            (5, 'line'),
-                            (5, 'line'),
-                            (5, 'exception'),
-                            (6, 'line'),
-                            (7, 'line'),
-                            (7, 'return')]
-
-def tighterloop_example():
-    items = range(1, 4)
-    try:
-        i = 0
-        while 1: i = items[i]
-    except IndexError:
-        pass
-
-tighterloop_example.events = [(0, 'call'),
-                            (1, 'line'),
-                            (2, 'line'),
-                            (3, 'line'),
-                            (4, 'line'),
-                            (4, 'line'),
-                            (4, 'line'),
-                            (4, 'line'),
-                            (4, 'exception'),
-                            (5, 'line'),
-                            (6, 'line'),
-                            (6, 'return')]
-
-def generator_function():
-    try:
-        yield True
-        "continued"
-    finally:
-        "finally"
-def generator_example():
-    # any() will leave the generator before its end
-    x = any(generator_function())
-
-    # the following lines were not traced
-    for x in range(10):
-        y = x
-
-generator_example.events = ([(0, 'call'),
-                             (2, 'line'),
-                             (-6, 'call'),
-                             (-5, 'line'),
-                             (-4, 'line'),
-                             (-4, 'return'),
-                             (-4, 'call'),
-                             (-4, 'exception'),
-                             (-1, 'line'),
-                             (-1, 'return')] +
-                            [(5, 'line'), (6, 'line')] * 10 +
-                            [(5, 'line'), (5, 'return')])
-
-
-class Tracer:
-    def __init__(self):
-        self.events = []
-    def trace(self, frame, event, arg):
-        self.events.append((frame.f_lineno, event))
-        return self.trace
-    def traceWithGenexp(self, frame, event, arg):
-        (o for o in [1])
-        self.events.append((frame.f_lineno, event))
-        return self.trace
-
-class TraceTestCase(unittest.TestCase):
-
-    # Disable gc collection when tracing, otherwise the
-    # deallocators may be traced as well.
-    def setUp(self):
-        self.using_gc = gc.isenabled()
-        gc.disable()
-
-    def tearDown(self):
-        if self.using_gc:
-            gc.enable()
-
-    def compare_events(self, line_offset, events, expected_events):
-        events = [(l - line_offset, e) for (l, e) in events]
-        if events != expected_events:
-            self.fail(
-                "events did not match expectation:\n" +
-                "\n".join(difflib.ndiff([str(x) for x in expected_events],
-                                        [str(x) for x in events])))
-
-    def run_and_compare(self, func, events):
-        tracer = Tracer()
-        sys.settrace(tracer.trace)
-        func()
-        sys.settrace(None)
-        self.compare_events(func.__code__.co_firstlineno,
-                            tracer.events, events)
-
-    def run_test(self, func):
-        self.run_and_compare(func, func.events)
-
-    def run_test2(self, func):
-        tracer = Tracer()
-        func(tracer.trace)
-        sys.settrace(None)
-        self.compare_events(func.__code__.co_firstlineno,
-                            tracer.events, func.events)
-
-    def set_and_retrieve_none(self):
-        sys.settrace(None)
-        assert sys.gettrace() is None
-
-    def set_and_retrieve_func(self):
-        def fn(*args):
-            pass
-
-        sys.settrace(fn)
-        try:
-            assert sys.gettrace() is fn
-        finally:
-            sys.settrace(None)
-
-    def test_01_basic(self):
-        self.run_test(basic)
-    def test_02_arigo(self):
-        self.run_test(arigo_example)
-    def test_03_one_instr(self):
-        self.run_test(one_instr_line)
-    def test_04_no_pop_blocks(self):
-        self.run_test(no_pop_blocks)
-    def test_05_no_pop_tops(self):
-        self.run_test(no_pop_tops)
-    def test_06_call(self):
-        self.run_test(call)
-    def test_07_raise(self):
-        self.run_test(test_raise)
-
-    def test_08_settrace_and_return(self):
-        self.run_test2(settrace_and_return)
-    def test_09_settrace_and_raise(self):
-        self.run_test2(settrace_and_raise)
-    def test_10_ireturn(self):
-        self.run_test(ireturn_example)
-    def test_11_tightloop(self):
-        self.run_test(tightloop_example)
-    def test_12_tighterloop(self):
-        self.run_test(tighterloop_example)
-
-    def test_13_genexp(self):
-        self.run_test(generator_example)
-        # issue1265: if the trace function contains a generator,
-        # and if the traced function contains another generator
-        # that is not completely exhausted, the trace stopped.
-        # Worse: the 'finally' clause was not invoked.
-        tracer = Tracer()
-        sys.settrace(tracer.traceWithGenexp)
-        generator_example()
-        sys.settrace(None)
-        self.compare_events(generator_example.__code__.co_firstlineno,
-                            tracer.events, generator_example.events)
-
-    def test_14_onliner_if(self):
-        def onliners():
-            if True: False
-            else: True
-            return 0
-        self.run_and_compare(
-            onliners,
-            [(0, 'call'),
-             (1, 'line'),
-             (3, 'line'),
-             (3, 'return')])
-
-    def test_15_loops(self):
-        # issue1750076: "while" expression is skipped by debugger
-        def for_example():
-            for x in range(2):
-                pass
-        self.run_and_compare(
-            for_example,
-            [(0, 'call'),
-             (1, 'line'),
-             (2, 'line'),
-             (1, 'line'),
-             (2, 'line'),
-             (1, 'line'),
-             (1, 'return')])
-
-        def while_example():
-            # While expression should be traced on every loop
-            x = 2
-            while x > 0:
-                x -= 1
-        self.run_and_compare(
-            while_example,
-            [(0, 'call'),
-             (2, 'line'),
-             (3, 'line'),
-             (4, 'line'),
-             (3, 'line'),
-             (4, 'line'),
-             (3, 'line'),
-             (3, 'return')])
-
-    def test_16_blank_lines(self):
-        namespace = {}
-        exec("def f():\n" + "\n" * 256 + "    pass", namespace)
-        self.run_and_compare(
-            namespace["f"],
-            [(0, 'call'),
-             (257, 'line'),
-             (257, 'return')])
-
-
-class RaisingTraceFuncTestCase(unittest.TestCase):
-    def trace(self, frame, event, arg):
-        """A trace function that raises an exception in response to a
-        specific trace event."""
-        if event == self.raiseOnEvent:
-            raise ValueError # just something that isn't RuntimeError
-        else:
-            return self.trace
-
-    def f(self):
-        """The function to trace; raises an exception if that's the case
-        we're testing, so that the 'exception' trace event fires."""
-        if self.raiseOnEvent == 'exception':
-            x = 0
-            y = 1/x
-        else:
-            return 1
-
-    def run_test_for_event(self, event):
-        """Tests that an exception raised in response to the given event is
-        handled OK."""
-        self.raiseOnEvent = event
-        try:
-            for i in range(sys.getrecursionlimit() + 1):
-                sys.settrace(self.trace)
-                try:
-                    self.f()
-                except ValueError:
-                    pass
-                else:
-                    self.fail("exception not thrown!")
-        except RuntimeError:
-            self.fail("recursion counter not reset")
-
-    # Test the handling of exceptions raised by each kind of trace event.
-    def test_call(self):
-        self.run_test_for_event('call')
-    def test_line(self):
-        self.run_test_for_event('line')
-    def test_return(self):
-        self.run_test_for_event('return')
-    def test_exception(self):
-        self.run_test_for_event('exception')
-
-    def test_trash_stack(self):
-        def f():
-            for i in range(5):
-                print(i)  # line tracing will raise an exception at this line
-
-        def g(frame, why, extra):
-            if (why == 'line' and
-                frame.f_lineno == f.__code__.co_firstlineno + 2):
-                raise RuntimeError("i am crashing")
-            return g
-
-        sys.settrace(g)
-        try:
-            f()
-        except RuntimeError:
-            # the test is really that this doesn't segfault:
-            import gc
-            gc.collect()
-        else:
-            self.fail("exception not propagated")
-
-
-# 'Jump' tests: assigning to frame.f_lineno within a trace function
-# moves the execution position - it's how debuggers implement a Jump
-# command (aka. "Set next statement").
-
-class JumpTracer:
-    """Defines a trace function that jumps from one place to another,
-    with the source and destination lines of the jump being defined by
-    the 'jump' property of the function under test."""
-
-    def __init__(self, function):
-        self.function = function
-        self.jumpFrom = function.jump[0]
-        self.jumpTo = function.jump[1]
-        self.done = False
-
-    def trace(self, frame, event, arg):
-        if not self.done and frame.f_code == self.function.__code__:
-            firstLine = frame.f_code.co_firstlineno
-            if event == 'line' and frame.f_lineno == firstLine + self.jumpFrom:
-                # Cope with non-integer self.jumpTo (because of
-                # no_jump_to_non_integers below).
-                try:
-                    frame.f_lineno = firstLine + self.jumpTo
-                except TypeError:
-                    frame.f_lineno = self.jumpTo
-                self.done = True
-        return self.trace
-
-# The first set of 'jump' tests are for things that are allowed:
-
-def jump_simple_forwards(output):
-    output.append(1)
-    output.append(2)
-    output.append(3)
-
-jump_simple_forwards.jump = (1, 3)
-jump_simple_forwards.output = [3]
-
-def jump_simple_backwards(output):
-    output.append(1)
-    output.append(2)
-
-jump_simple_backwards.jump = (2, 1)
-jump_simple_backwards.output = [1, 1, 2]
-
-def jump_out_of_block_forwards(output):
-    for i in 1, 2:
-        output.append(2)
-        for j in [3]:  # Also tests jumping over a block
-            output.append(4)
-    output.append(5)
-
-jump_out_of_block_forwards.jump = (3, 5)
-jump_out_of_block_forwards.output = [2, 5]
-
-def jump_out_of_block_backwards(output):
-    output.append(1)
-    for i in [1]:
-        output.append(3)
-        for j in [2]:  # Also tests jumping over a block
-            output.append(5)
-        output.append(6)
-    output.append(7)
-
-jump_out_of_block_backwards.jump = (6, 1)
-jump_out_of_block_backwards.output = [1, 3, 5, 1, 3, 5, 6, 7]
-
-def jump_to_codeless_line(output):
-    output.append(1)
-    # Jumping to this line should skip to the next one.
-    output.append(3)
-
-jump_to_codeless_line.jump = (1, 2)
-jump_to_codeless_line.output = [3]
-
-def jump_to_same_line(output):
-    output.append(1)
-    output.append(2)
-    output.append(3)
-
-jump_to_same_line.jump = (2, 2)
-jump_to_same_line.output = [1, 2, 3]
-
-# Tests jumping within a finally block, and over one.
-def jump_in_nested_finally(output):
-    try:
-        output.append(2)
-    finally:
-        output.append(4)
-        try:
-            output.append(6)
-        finally:
-            output.append(8)
-        output.append(9)
-
-jump_in_nested_finally.jump = (4, 9)
-jump_in_nested_finally.output = [2, 9]
-
-# The second set of 'jump' tests are for things that are not allowed:
-
-def no_jump_too_far_forwards(output):
-    try:
-        output.append(2)
-        output.append(3)
-    except ValueError as e:
-        output.append('after' in str(e))
-
-no_jump_too_far_forwards.jump = (3, 6)
-no_jump_too_far_forwards.output = [2, True]
-
-def no_jump_too_far_backwards(output):
-    try:
-        output.append(2)
-        output.append(3)
-    except ValueError as e:
-        output.append('before' in str(e))
-
-no_jump_too_far_backwards.jump = (3, -1)
-no_jump_too_far_backwards.output = [2, True]
-
-# Test each kind of 'except' line.
-def no_jump_to_except_1(output):
-    try:
-        output.append(2)
-    except:
-        e = sys.exc_info()[1]
-        output.append('except' in str(e))
-
-no_jump_to_except_1.jump = (2, 3)
-no_jump_to_except_1.output = [True]
-
-def no_jump_to_except_2(output):
-    try:
-        output.append(2)
-    except ValueError:
-        e = sys.exc_info()[1]
-        output.append('except' in str(e))
-
-no_jump_to_except_2.jump = (2, 3)
-no_jump_to_except_2.output = [True]
-
-def no_jump_to_except_3(output):
-    try:
-        output.append(2)
-    except ValueError as e:
-        output.append('except' in str(e))
-
-no_jump_to_except_3.jump = (2, 3)
-no_jump_to_except_3.output = [True]
-
-def no_jump_to_except_4(output):
-    try:
-        output.append(2)
-    except (ValueError, RuntimeError) as e:
-        output.append('except' in str(e))
-
-no_jump_to_except_4.jump = (2, 3)
-no_jump_to_except_4.output = [True]
-
-def no_jump_forwards_into_block(output):
-    try:
-        output.append(2)
-        for i in 1, 2:
-            output.append(4)
-    except ValueError as e:
-        output.append('into' in str(e))
-
-no_jump_forwards_into_block.jump = (2, 4)
-no_jump_forwards_into_block.output = [True]
-
-def no_jump_backwards_into_block(output):
-    try:
-        for i in 1, 2:
-            output.append(3)
-        output.append(4)
-    except ValueError as e:
-        output.append('into' in str(e))
-
-no_jump_backwards_into_block.jump = (4, 3)
-no_jump_backwards_into_block.output = [3, 3, True]
-
-def no_jump_into_finally_block(output):
-    try:
-        try:
-            output.append(3)
-            x = 1
-        finally:
-            output.append(6)
-    except ValueError as e:
-        output.append('finally' in str(e))
-
-no_jump_into_finally_block.jump = (4, 6)
-no_jump_into_finally_block.output = [3, 6, True]  # The 'finally' still runs
-
-def no_jump_out_of_finally_block(output):
-    try:
-        try:
-            output.append(3)
-        finally:
-            output.append(5)
-            output.append(6)
-    except ValueError as e:
-        output.append('finally' in str(e))
-
-no_jump_out_of_finally_block.jump = (5, 1)
-no_jump_out_of_finally_block.output = [3, True]
-
-# This verifies the line-numbers-must-be-integers rule.
-def no_jump_to_non_integers(output):
-    try:
-        output.append(2)
-    except ValueError as e:
-        output.append('integer' in str(e))
-
-no_jump_to_non_integers.jump = (2, "Spam")
-no_jump_to_non_integers.output = [True]
-
-# This verifies that you can't set f_lineno via _getframe or similar
-# trickery.
-def no_jump_without_trace_function():
-    try:
-        previous_frame = sys._getframe().f_back
-        previous_frame.f_lineno = previous_frame.f_lineno
-    except ValueError as e:
-        # This is the exception we wanted; make sure the error message
-        # talks about trace functions.
-        if 'trace' not in str(e):
-            raise
-    else:
-        # Something's wrong - the expected exception wasn't raised.
-        raise RuntimeError("Trace-function-less jump failed to fail")
-
-
-class JumpTestCase(unittest.TestCase):
-    def compare_jump_output(self, expected, received):
-        if received != expected:
-            self.fail( "Outputs don't match:\n" +
-                       "Expected: " + repr(expected) + "\n" +
-                       "Received: " + repr(received))
-
-    def run_test(self, func):
-        tracer = JumpTracer(func)
-        sys.settrace(tracer.trace)
-        output = []
-        func(output)
-        sys.settrace(None)
-        self.compare_jump_output(func.output, output)
-
-    def test_01_jump_simple_forwards(self):
-        self.run_test(jump_simple_forwards)
-    def test_02_jump_simple_backwards(self):
-        self.run_test(jump_simple_backwards)
-    def test_03_jump_out_of_block_forwards(self):
-        self.run_test(jump_out_of_block_forwards)
-    def test_04_jump_out_of_block_backwards(self):
-        self.run_test(jump_out_of_block_backwards)
-    def test_05_jump_to_codeless_line(self):
-        self.run_test(jump_to_codeless_line)
-    def test_06_jump_to_same_line(self):
-        self.run_test(jump_to_same_line)
-    def test_07_jump_in_nested_finally(self):
-        self.run_test(jump_in_nested_finally)
-    def test_08_no_jump_too_far_forwards(self):
-        self.run_test(no_jump_too_far_forwards)
-    def test_09_no_jump_too_far_backwards(self):
-        self.run_test(no_jump_too_far_backwards)
-    def test_10_no_jump_to_except_1(self):
-        self.run_test(no_jump_to_except_1)
-    def test_11_no_jump_to_except_2(self):
-        self.run_test(no_jump_to_except_2)
-    def test_12_no_jump_to_except_3(self):
-        self.run_test(no_jump_to_except_3)
-    def test_13_no_jump_to_except_4(self):
-        self.run_test(no_jump_to_except_4)
-    def test_14_no_jump_forwards_into_block(self):
-        self.run_test(no_jump_forwards_into_block)
-    def test_15_no_jump_backwards_into_block(self):
-        self.run_test(no_jump_backwards_into_block)
-    def test_16_no_jump_into_finally_block(self):
-        self.run_test(no_jump_into_finally_block)
-    def test_17_no_jump_out_of_finally_block(self):
-        self.run_test(no_jump_out_of_finally_block)
-    def test_18_no_jump_to_non_integers(self):
-        self.run_test(no_jump_to_non_integers)
-    def test_19_no_jump_without_trace_function(self):
-        no_jump_without_trace_function()
-
-    def test_20_large_function(self):
-        d = {}
-        exec("""def f(output):        # line 0
-            x = 0                     # line 1
-            y = 1                     # line 2
-            '''                       # line 3
-            %s                        # lines 4-1004
-            '''                       # line 1005
-            x += 1                    # line 1006
-            output.append(x)          # line 1007
-            return""" % ('\n' * 1000,), d)
-        f = d['f']
-
-        f.jump = (2, 1007)
-        f.output = [0]
-        self.run_test(f)
-
-    def test_jump_to_firstlineno(self):
-        # This tests that PDB can jump back to the first line in a
-        # file.  See issue #1689458.  It can only be triggered in a
-        # function call if the function is defined on a single line.
-        code = compile("""
-# Comments don't count.
-output.append(2)  # firstlineno is here.
-output.append(3)
-output.append(4)
-""", "<fake module>", "exec")
-        class fake_function:
-            __code__ = code
-            jump = (2, 0)
-        tracer = JumpTracer(fake_function)
-        sys.settrace(tracer.trace)
-        namespace = {"output": []}
-        exec(code, namespace)
-        sys.settrace(None)
-        self.compare_jump_output([2, 3, 2, 3, 4], namespace["output"])
-
-
-def test_main():
-    support.run_unittest(
-        TraceTestCase,
-        RaisingTraceFuncTestCase,
-        JumpTestCase
-    )
-
-if __name__ == "__main__":
-    test_main()

From python-checkins at python.org  Wed Jul 28 16:29:34 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Wed, 28 Jul 2010 16:29:34 +0200 (CEST)
Subject: [Python-checkins] r83205 -
	python/branches/release27-maint/Lib/test/test_trace.py
Message-ID: <20100728142934.3B5D1EEB1D@mail.python.org>

Author: alexander.belopolsky
Date: Wed Jul 28 16:29:34 2010
New Revision: 83205

Log:
Reverted r83142: add Lib/test/test_trace.py back

Added:
   python/branches/release27-maint/Lib/test/test_trace.py

Added: python/branches/release27-maint/Lib/test/test_trace.py
==============================================================================
--- (empty file)
+++ python/branches/release27-maint/Lib/test/test_trace.py	Wed Jul 28 16:29:34 2010
@@ -0,0 +1,214 @@
+import imp
+import os
+import sys
+from test import test_support
+import unittest
+
+import trace
+from trace import CoverageResults, Trace
+
+
+#------------------------------- Utilities -----------------------------------#
+
+def make_fake_module():
+    """Creates a fake module named 'fakemodule'.
+
+    The new module has a single function named 'foo', and it's placed in
+    sys.modules
+    The file this fake module "comes from" is fakefile.py
+
+    """
+
+    # Prepare the function to import from the fake module
+    #
+    _fake_foo_src = r'''
+def foo(a_):
+    b = a_ + 1
+    return b + 2
+'''.lstrip()
+
+    _fake_foo = compile(_fake_foo_src, 'fakefile.py', 'exec')
+
+    # Create a new module, place the function into it and add it to sys.modules
+    #
+    fakemodule = imp.new_module('fakemodule')
+    exec _fake_foo in fakemodule.__dict__
+    fakemodule.__file__ = 'fakefile.py'
+    sys.modules['fakemodule'] = fakemodule
+
+
+def modname(filename):
+    """Infer a module name from a containing file name"""
+    base = os.path.basename(filename)
+    mod, ext = os.path.splitext(base)
+    return mod
+
+
+def my_file_and_modname():
+    """The file and module name of this file (__file__)"""
+    return __file__, modname(__file__)
+
+
+#-------------------- Target functions for tracing ---------------------------#
+
+def _traced_func_linear(a_, b_):
+    a = a_
+    b = b_
+    c = a + b
+    return c
+
+def _traced_func_loop(a_, b_):
+    c = a_
+    for i in range(5):
+        c += b_
+    return c
+
+# Expects the 'fakemodule' module to exist and have a 'foo' function in it
+#
+def _traced_func_importing(a_, b_):
+    from fakemodule import foo
+    return a_ + b_ + foo(1)
+
+def _traced_func_simple_caller(a_):
+    c = _traced_func_linear(a_, a_)
+    return c + a_
+
+def _traced_func_importing_caller(a_):
+    k = _traced_func_simple_caller(a_)
+    k += _traced_func_importing(k, a_)
+    return k
+
+
+#------------------------------ Test cases -----------------------------------#
+
+
+class TestLineCounts(unittest.TestCase):
+    """White-box testing of line-counting, via runfunc"""
+    def setUp(self):
+        self.tr = Trace(count=1, trace=0, countfuncs=0, countcallers=0)
+
+    def test_traced_func_linear(self):
+        result = self.tr.runfunc(_traced_func_linear, 2, 5)
+        self.assertEqual(result, 7)
+
+        # all lines are executed once
+        expected = {}
+        firstlineno = _traced_func_linear.__code__.co_firstlineno
+        for i in range(1, 5):
+            expected[(__file__, firstlineno +  i)] = 1
+
+        self.assertEqual(self.tr.results().counts, expected)
+
+    def test_traced_func_loop(self):
+        self.tr.runfunc(_traced_func_loop, 2, 3)
+
+        firstlineno = _traced_func_loop.__code__.co_firstlineno
+        expected = {
+            (__file__, firstlineno + 1): 1,
+            (__file__, firstlineno + 2): 6,
+            (__file__, firstlineno + 3): 5,
+            (__file__, firstlineno + 4): 1,
+        }
+        self.assertEqual(self.tr.results().counts, expected)
+
+    def test_traced_func_importing(self):
+        make_fake_module()
+        self.tr.runfunc(_traced_func_importing, 2, 5)
+
+        firstlineno = _traced_func_importing.__code__.co_firstlineno
+        expected = {
+            (__file__, firstlineno + 1): 1,
+            (__file__, firstlineno + 2): 1,
+            ('fakefile.py', 2): 1,
+            ('fakefile.py', 3): 1,
+        }
+        self.assertEqual(self.tr.results().counts, expected)
+
+
+class TestRunExecCounts(unittest.TestCase):
+    """A simple sanity test of line-counting, via run (exec)"""
+    def test_tt(self):
+        self.tr = Trace(count=1, trace=0, countfuncs=0, countcallers=0)
+        code = r'''_traced_func_loop(2, 5)'''
+        code = compile(code, __file__, 'exec')
+        self.tr.run(code)
+
+        firstlineno = _traced_func_loop.__code__.co_firstlineno
+        expected = {
+            (__file__, firstlineno + 1): 1,
+            (__file__, firstlineno + 2): 6,
+            (__file__, firstlineno + 3): 5,
+            (__file__, firstlineno + 4): 1,
+        }
+
+        # When used through 'run', some other spurios counts are produced, like
+        # the settrace of threading, which we ignore, just making sure that the
+        # counts fo _traced_func_loop were right.
+        #
+        for k in expected.keys():
+            self.assertEqual(self.tr.results().counts[k], expected[k])
+
+
+class TestFuncs(unittest.TestCase):
+    """White-box testing of funcs tracing"""
+    def setUp(self):
+        self.tr = Trace(count=0, trace=0, countfuncs=1)
+        self.filemod = my_file_and_modname()
+
+    def test_simple_caller(self):
+        self.tr.runfunc(_traced_func_simple_caller, 1)
+
+        expected = {
+            self.filemod + ('_traced_func_simple_caller',): 1,
+            self.filemod + ('_traced_func_linear',): 1,
+        }
+        self.assertEqual(self.tr.results().calledfuncs, expected)
+
+    def test_loop_caller_importing(self):
+        make_fake_module()
+        self.tr.runfunc(_traced_func_importing_caller, 1)
+
+        expected = {
+            self.filemod + ('_traced_func_simple_caller',): 1,
+            self.filemod + ('_traced_func_linear',): 1,
+            self.filemod + ('_traced_func_importing_caller',): 1,
+            self.filemod + ('_traced_func_importing',): 1,
+            ('fakefile.py', 'fakefile', 'foo'): 1,
+        }
+        self.assertEqual(self.tr.results().calledfuncs, expected)
+
+
+class TestCallers(unittest.TestCase):
+    """White-box testing of callers tracing"""
+    def setUp(self):
+        self.tr = Trace(count=0, trace=0, countcallers=1)
+        self.filemod = my_file_and_modname()
+
+    def test_loop_caller_importing(self):
+        make_fake_module()
+        self.tr.runfunc(_traced_func_importing_caller, 1)
+
+        expected = {
+            ((os.path.splitext(trace.__file__)[0] + '.py', 'trace', 'Trace.runfunc'),
+                (self.filemod + ('_traced_func_importing_caller',))): 1,
+            ((self.filemod + ('_traced_func_simple_caller',)),
+                (self.filemod + ('_traced_func_linear',))): 1,
+            ((self.filemod + ('_traced_func_importing_caller',)),
+                (self.filemod + ('_traced_func_simple_caller',))): 1,
+            ((self.filemod + ('_traced_func_importing_caller',)),
+                (self.filemod + ('_traced_func_importing',))): 1,
+            ((self.filemod + ('_traced_func_importing',)),
+                ('fakefile.py', 'fakefile', 'foo')): 1,
+        }
+        self.assertEqual(self.tr.results().callers, expected)
+
+
+#------------------------------ Driver ---------------------------------------#
+
+def test_main():
+    print(__name__, type(__name__))
+    test_support.run_unittest(__name__)
+
+
+if __name__ == '__main__':
+    test_main()

From python-checkins at python.org  Wed Jul 28 16:33:17 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Wed, 28 Jul 2010 16:33:17 +0200 (CEST)
Subject: [Python-checkins] r83206 - in
	python/branches/release27-maint/Lib/test:
	test_profilehooks.py test_sys.py test_sys_setprofile.py
	test_sys_settrace.py test_trace.py
Message-ID: <20100728143317.EB333EE9BC@mail.python.org>

Author: alexander.belopolsky
Date: Wed Jul 28 16:33:17 2010
New Revision: 83206

Log:
Redo r83142 merge manually. QOTD: svnmerge isn't really your best friend

Added:
   python/branches/release27-maint/Lib/test/test_sys_setprofile.py
      - copied unchanged from r83204, /python/branches/release27-maint/Lib/test/test_profilehooks.py
   python/branches/release27-maint/Lib/test/test_sys_settrace.py
      - copied unchanged from r83205, /python/branches/release27-maint/Lib/test/test_trace.py
Removed:
   python/branches/release27-maint/Lib/test/test_profilehooks.py
   python/branches/release27-maint/Lib/test/test_trace.py
Modified:
   python/branches/release27-maint/Lib/test/test_sys.py

Deleted: python/branches/release27-maint/Lib/test/test_profilehooks.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_profilehooks.py	Wed Jul 28 16:33:17 2010
+++ (empty file)
@@ -1,385 +0,0 @@
-import pprint
-import sys
-import unittest
-
-from test import test_support
-
-class TestGetProfile(unittest.TestCase):
-    def setUp(self):
-        sys.setprofile(None)
-
-    def tearDown(self):
-        sys.setprofile(None)
-
-    def test_empty(self):
-        assert sys.getprofile() is None
-
-    def test_setget(self):
-        def fn(*args):
-            pass
-
-        sys.setprofile(fn)
-        assert sys.getprofile() == fn
-
-class HookWatcher:
-    def __init__(self):
-        self.frames = []
-        self.events = []
-
-    def callback(self, frame, event, arg):
-        if (event == "call"
-            or event == "return"
-            or event == "exception"):
-            self.add_event(event, frame)
-
-    def add_event(self, event, frame=None):
-        """Add an event to the log."""
-        if frame is None:
-            frame = sys._getframe(1)
-
-        try:
-            frameno = self.frames.index(frame)
-        except ValueError:
-            frameno = len(self.frames)
-            self.frames.append(frame)
-
-        self.events.append((frameno, event, ident(frame)))
-
-    def get_events(self):
-        """Remove calls to add_event()."""
-        disallowed = [ident(self.add_event.im_func), ident(ident)]
-        self.frames = None
-
-        return [item for item in self.events if item[2] not in disallowed]
-
-
-class ProfileSimulator(HookWatcher):
-    def __init__(self, testcase):
-        self.testcase = testcase
-        self.stack = []
-        HookWatcher.__init__(self)
-
-    def callback(self, frame, event, arg):
-        # Callback registered with sys.setprofile()/sys.settrace()
-        self.dispatch[event](self, frame)
-
-    def trace_call(self, frame):
-        self.add_event('call', frame)
-        self.stack.append(frame)
-
-    def trace_return(self, frame):
-        self.add_event('return', frame)
-        self.stack.pop()
-
-    def trace_exception(self, frame):
-        self.testcase.fail(
-            "the profiler should never receive exception events")
-
-    def trace_pass(self, frame):
-        pass
-
-    dispatch = {
-        'call': trace_call,
-        'exception': trace_exception,
-        'return': trace_return,
-        'c_call': trace_pass,
-        'c_return': trace_pass,
-        'c_exception': trace_pass,
-        }
-
-
-class TestCaseBase(unittest.TestCase):
-    def check_events(self, callable, expected):
-        events = capture_events(callable, self.new_watcher())
-        if events != expected:
-            self.fail("Expected events:\n%s\nReceived events:\n%s"
-                      % (pprint.pformat(expected), pprint.pformat(events)))
-
-
-class ProfileHookTestCase(TestCaseBase):
-    def new_watcher(self):
-        return HookWatcher()
-
-    def test_simple(self):
-        def f(p):
-            pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_exception(self):
-        def f(p):
-            1./0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_caught_exception(self):
-        def f(p):
-            try: 1./0
-            except: pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_caught_nested_exception(self):
-        def f(p):
-            try: 1./0
-            except: pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_nested_exception(self):
-        def f(p):
-            1./0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              # This isn't what I expected:
-                              # (0, 'exception', protect_ident),
-                              # I expected this again:
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_exception_in_except_clause(self):
-        def f(p):
-            1./0
-        def g(p):
-            try:
-                f(p)
-            except:
-                try: f(p)
-                except: pass
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (3, 'call', f_ident),
-                              (3, 'return', f_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-    def test_exception_propogation(self):
-        def f(p):
-            1./0
-        def g(p):
-            try: f(p)
-            finally: p.add_event("falling through")
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (1, 'falling through', g_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-    def test_raise_twice(self):
-        def f(p):
-            try: 1./0
-            except: 1./0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_raise_reraise(self):
-        def f(p):
-            try: 1./0
-            except: raise
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_raise(self):
-        def f(p):
-            raise Exception()
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_distant_exception(self):
-        def f():
-            1./0
-        def g():
-            f()
-        def h():
-            g()
-        def i():
-            h()
-        def j(p):
-            i()
-        f_ident = ident(f)
-        g_ident = ident(g)
-        h_ident = ident(h)
-        i_ident = ident(i)
-        j_ident = ident(j)
-        self.check_events(j, [(1, 'call', j_ident),
-                              (2, 'call', i_ident),
-                              (3, 'call', h_ident),
-                              (4, 'call', g_ident),
-                              (5, 'call', f_ident),
-                              (5, 'return', f_ident),
-                              (4, 'return', g_ident),
-                              (3, 'return', h_ident),
-                              (2, 'return', i_ident),
-                              (1, 'return', j_ident),
-                              ])
-
-    def test_generator(self):
-        def f():
-            for i in range(2):
-                yield i
-        def g(p):
-            for i in f():
-                pass
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              # call the iterator twice to generate values
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              # once more; returns end-of-iteration with
-                              # actually raising an exception
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-    def test_stop_iteration(self):
-        def f():
-            for i in range(2):
-                yield i
-            raise StopIteration
-        def g(p):
-            for i in f():
-                pass
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              # call the iterator twice to generate values
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              # once more to hit the raise:
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-
-class ProfileSimulatorTestCase(TestCaseBase):
-    def new_watcher(self):
-        return ProfileSimulator(self)
-
-    def test_simple(self):
-        def f(p):
-            pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_basic_exception(self):
-        def f(p):
-            1./0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_caught_exception(self):
-        def f(p):
-            try: 1./0
-            except: pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_distant_exception(self):
-        def f():
-            1./0
-        def g():
-            f()
-        def h():
-            g()
-        def i():
-            h()
-        def j(p):
-            i()
-        f_ident = ident(f)
-        g_ident = ident(g)
-        h_ident = ident(h)
-        i_ident = ident(i)
-        j_ident = ident(j)
-        self.check_events(j, [(1, 'call', j_ident),
-                              (2, 'call', i_ident),
-                              (3, 'call', h_ident),
-                              (4, 'call', g_ident),
-                              (5, 'call', f_ident),
-                              (5, 'return', f_ident),
-                              (4, 'return', g_ident),
-                              (3, 'return', h_ident),
-                              (2, 'return', i_ident),
-                              (1, 'return', j_ident),
-                              ])
-
-
-def ident(function):
-    if hasattr(function, "f_code"):
-        code = function.f_code
-    else:
-        code = function.func_code
-    return code.co_firstlineno, code.co_name
-
-
-def protect(f, p):
-    try: f(p)
-    except: pass
-
-protect_ident = ident(protect)
-
-
-def capture_events(callable, p=None):
-    try:
-        sys.setprofile()
-    except TypeError:
-        pass
-    else:
-        raise test_support.TestFailed(
-            'sys.setprofile() did not raise TypeError')
-
-    if p is None:
-        p = HookWatcher()
-    sys.setprofile(p.callback)
-    protect(callable, p)
-    sys.setprofile(None)
-    return p.get_events()[1:-1]
-
-
-def show_events(callable):
-    import pprint
-    pprint.pprint(capture_events(callable))
-
-
-def test_main():
-    test_support.run_unittest(
-        TestGetProfile,
-        ProfileHookTestCase,
-        ProfileSimulatorTestCase
-    )
-
-
-if __name__ == "__main__":
-    test_main()

Modified: python/branches/release27-maint/Lib/test/test_sys.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_sys.py	(original)
+++ python/branches/release27-maint/Lib/test/test_sys.py	Wed Jul 28 16:33:17 2010
@@ -205,8 +205,8 @@
             # can't check more than the type, as the user might have changed it
             self.assertIsInstance(sys.getdefaultencoding(), str)
 
-    # testing sys.settrace() is done in test_trace.py
-    # testing sys.setprofile() is done in test_profile.py
+    # testing sys.settrace() is done in test_sys_settrace.py
+    # testing sys.setprofile() is done in test_sys_setprofile.py
 
     def test_setcheckinterval(self):
         self.assertRaises(TypeError, sys.setcheckinterval)

Deleted: python/branches/release27-maint/Lib/test/test_trace.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_trace.py	Wed Jul 28 16:33:17 2010
+++ (empty file)
@@ -1,214 +0,0 @@
-import imp
-import os
-import sys
-from test import test_support
-import unittest
-
-import trace
-from trace import CoverageResults, Trace
-
-
-#------------------------------- Utilities -----------------------------------#
-
-def make_fake_module():
-    """Creates a fake module named 'fakemodule'.
-
-    The new module has a single function named 'foo', and it's placed in
-    sys.modules
-    The file this fake module "comes from" is fakefile.py
-
-    """
-
-    # Prepare the function to import from the fake module
-    #
-    _fake_foo_src = r'''
-def foo(a_):
-    b = a_ + 1
-    return b + 2
-'''.lstrip()
-
-    _fake_foo = compile(_fake_foo_src, 'fakefile.py', 'exec')
-
-    # Create a new module, place the function into it and add it to sys.modules
-    #
-    fakemodule = imp.new_module('fakemodule')
-    exec _fake_foo in fakemodule.__dict__
-    fakemodule.__file__ = 'fakefile.py'
-    sys.modules['fakemodule'] = fakemodule
-
-
-def modname(filename):
-    """Infer a module name from a containing file name"""
-    base = os.path.basename(filename)
-    mod, ext = os.path.splitext(base)
-    return mod
-
-
-def my_file_and_modname():
-    """The file and module name of this file (__file__)"""
-    return __file__, modname(__file__)
-
-
-#-------------------- Target functions for tracing ---------------------------#
-
-def _traced_func_linear(a_, b_):
-    a = a_
-    b = b_
-    c = a + b
-    return c
-
-def _traced_func_loop(a_, b_):
-    c = a_
-    for i in range(5):
-        c += b_
-    return c
-
-# Expects the 'fakemodule' module to exist and have a 'foo' function in it
-#
-def _traced_func_importing(a_, b_):
-    from fakemodule import foo
-    return a_ + b_ + foo(1)
-
-def _traced_func_simple_caller(a_):
-    c = _traced_func_linear(a_, a_)
-    return c + a_
-
-def _traced_func_importing_caller(a_):
-    k = _traced_func_simple_caller(a_)
-    k += _traced_func_importing(k, a_)
-    return k
-
-
-#------------------------------ Test cases -----------------------------------#
-
-
-class TestLineCounts(unittest.TestCase):
-    """White-box testing of line-counting, via runfunc"""
-    def setUp(self):
-        self.tr = Trace(count=1, trace=0, countfuncs=0, countcallers=0)
-
-    def test_traced_func_linear(self):
-        result = self.tr.runfunc(_traced_func_linear, 2, 5)
-        self.assertEqual(result, 7)
-
-        # all lines are executed once
-        expected = {}
-        firstlineno = _traced_func_linear.__code__.co_firstlineno
-        for i in range(1, 5):
-            expected[(__file__, firstlineno +  i)] = 1
-
-        self.assertEqual(self.tr.results().counts, expected)
-
-    def test_traced_func_loop(self):
-        self.tr.runfunc(_traced_func_loop, 2, 3)
-
-        firstlineno = _traced_func_loop.__code__.co_firstlineno
-        expected = {
-            (__file__, firstlineno + 1): 1,
-            (__file__, firstlineno + 2): 6,
-            (__file__, firstlineno + 3): 5,
-            (__file__, firstlineno + 4): 1,
-        }
-        self.assertEqual(self.tr.results().counts, expected)
-
-    def test_traced_func_importing(self):
-        make_fake_module()
-        self.tr.runfunc(_traced_func_importing, 2, 5)
-
-        firstlineno = _traced_func_importing.__code__.co_firstlineno
-        expected = {
-            (__file__, firstlineno + 1): 1,
-            (__file__, firstlineno + 2): 1,
-            ('fakefile.py', 2): 1,
-            ('fakefile.py', 3): 1,
-        }
-        self.assertEqual(self.tr.results().counts, expected)
-
-
-class TestRunExecCounts(unittest.TestCase):
-    """A simple sanity test of line-counting, via run (exec)"""
-    def test_tt(self):
-        self.tr = Trace(count=1, trace=0, countfuncs=0, countcallers=0)
-        code = r'''_traced_func_loop(2, 5)'''
-        code = compile(code, __file__, 'exec')
-        self.tr.run(code)
-
-        firstlineno = _traced_func_loop.__code__.co_firstlineno
-        expected = {
-            (__file__, firstlineno + 1): 1,
-            (__file__, firstlineno + 2): 6,
-            (__file__, firstlineno + 3): 5,
-            (__file__, firstlineno + 4): 1,
-        }
-
-        # When used through 'run', some other spurios counts are produced, like
-        # the settrace of threading, which we ignore, just making sure that the
-        # counts fo _traced_func_loop were right.
-        #
-        for k in expected.keys():
-            self.assertEqual(self.tr.results().counts[k], expected[k])
-
-
-class TestFuncs(unittest.TestCase):
-    """White-box testing of funcs tracing"""
-    def setUp(self):
-        self.tr = Trace(count=0, trace=0, countfuncs=1)
-        self.filemod = my_file_and_modname()
-
-    def test_simple_caller(self):
-        self.tr.runfunc(_traced_func_simple_caller, 1)
-
-        expected = {
-            self.filemod + ('_traced_func_simple_caller',): 1,
-            self.filemod + ('_traced_func_linear',): 1,
-        }
-        self.assertEqual(self.tr.results().calledfuncs, expected)
-
-    def test_loop_caller_importing(self):
-        make_fake_module()
-        self.tr.runfunc(_traced_func_importing_caller, 1)
-
-        expected = {
-            self.filemod + ('_traced_func_simple_caller',): 1,
-            self.filemod + ('_traced_func_linear',): 1,
-            self.filemod + ('_traced_func_importing_caller',): 1,
-            self.filemod + ('_traced_func_importing',): 1,
-            ('fakefile.py', 'fakefile', 'foo'): 1,
-        }
-        self.assertEqual(self.tr.results().calledfuncs, expected)
-
-
-class TestCallers(unittest.TestCase):
-    """White-box testing of callers tracing"""
-    def setUp(self):
-        self.tr = Trace(count=0, trace=0, countcallers=1)
-        self.filemod = my_file_and_modname()
-
-    def test_loop_caller_importing(self):
-        make_fake_module()
-        self.tr.runfunc(_traced_func_importing_caller, 1)
-
-        expected = {
-            ((os.path.splitext(trace.__file__)[0] + '.py', 'trace', 'Trace.runfunc'),
-                (self.filemod + ('_traced_func_importing_caller',))): 1,
-            ((self.filemod + ('_traced_func_simple_caller',)),
-                (self.filemod + ('_traced_func_linear',))): 1,
-            ((self.filemod + ('_traced_func_importing_caller',)),
-                (self.filemod + ('_traced_func_simple_caller',))): 1,
-            ((self.filemod + ('_traced_func_importing_caller',)),
-                (self.filemod + ('_traced_func_importing',))): 1,
-            ((self.filemod + ('_traced_func_importing',)),
-                ('fakefile.py', 'fakefile', 'foo')): 1,
-        }
-        self.assertEqual(self.tr.results().callers, expected)
-
-
-#------------------------------ Driver ---------------------------------------#
-
-def test_main():
-    print(__name__, type(__name__))
-    test_support.run_unittest(__name__)
-
-
-if __name__ == '__main__':
-    test_main()

From python-checkins at python.org  Wed Jul 28 17:57:38 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Wed, 28 Jul 2010 17:57:38 +0200 (CEST)
Subject: [Python-checkins] r83207 -
	python/branches/release27-maint/Lib/test/test_sys_settrace.py
Message-ID: <20100728155738.CFDBCEEAB7@mail.python.org>

Author: alexander.belopolsky
Date: Wed Jul 28 17:57:38 2010
New Revision: 83207

Log:
Adding this file lost history

Removed:
   python/branches/release27-maint/Lib/test/test_sys_settrace.py

Deleted: python/branches/release27-maint/Lib/test/test_sys_settrace.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_sys_settrace.py	Wed Jul 28 17:57:38 2010
+++ (empty file)
@@ -1,214 +0,0 @@
-import imp
-import os
-import sys
-from test import test_support
-import unittest
-
-import trace
-from trace import CoverageResults, Trace
-
-
-#------------------------------- Utilities -----------------------------------#
-
-def make_fake_module():
-    """Creates a fake module named 'fakemodule'.
-
-    The new module has a single function named 'foo', and it's placed in
-    sys.modules
-    The file this fake module "comes from" is fakefile.py
-
-    """
-
-    # Prepare the function to import from the fake module
-    #
-    _fake_foo_src = r'''
-def foo(a_):
-    b = a_ + 1
-    return b + 2
-'''.lstrip()
-
-    _fake_foo = compile(_fake_foo_src, 'fakefile.py', 'exec')
-
-    # Create a new module, place the function into it and add it to sys.modules
-    #
-    fakemodule = imp.new_module('fakemodule')
-    exec _fake_foo in fakemodule.__dict__
-    fakemodule.__file__ = 'fakefile.py'
-    sys.modules['fakemodule'] = fakemodule
-
-
-def modname(filename):
-    """Infer a module name from a containing file name"""
-    base = os.path.basename(filename)
-    mod, ext = os.path.splitext(base)
-    return mod
-
-
-def my_file_and_modname():
-    """The file and module name of this file (__file__)"""
-    return __file__, modname(__file__)
-
-
-#-------------------- Target functions for tracing ---------------------------#
-
-def _traced_func_linear(a_, b_):
-    a = a_
-    b = b_
-    c = a + b
-    return c
-
-def _traced_func_loop(a_, b_):
-    c = a_
-    for i in range(5):
-        c += b_
-    return c
-
-# Expects the 'fakemodule' module to exist and have a 'foo' function in it
-#
-def _traced_func_importing(a_, b_):
-    from fakemodule import foo
-    return a_ + b_ + foo(1)
-
-def _traced_func_simple_caller(a_):
-    c = _traced_func_linear(a_, a_)
-    return c + a_
-
-def _traced_func_importing_caller(a_):
-    k = _traced_func_simple_caller(a_)
-    k += _traced_func_importing(k, a_)
-    return k
-
-
-#------------------------------ Test cases -----------------------------------#
-
-
-class TestLineCounts(unittest.TestCase):
-    """White-box testing of line-counting, via runfunc"""
-    def setUp(self):
-        self.tr = Trace(count=1, trace=0, countfuncs=0, countcallers=0)
-
-    def test_traced_func_linear(self):
-        result = self.tr.runfunc(_traced_func_linear, 2, 5)
-        self.assertEqual(result, 7)
-
-        # all lines are executed once
-        expected = {}
-        firstlineno = _traced_func_linear.__code__.co_firstlineno
-        for i in range(1, 5):
-            expected[(__file__, firstlineno +  i)] = 1
-
-        self.assertEqual(self.tr.results().counts, expected)
-
-    def test_traced_func_loop(self):
-        self.tr.runfunc(_traced_func_loop, 2, 3)
-
-        firstlineno = _traced_func_loop.__code__.co_firstlineno
-        expected = {
-            (__file__, firstlineno + 1): 1,
-            (__file__, firstlineno + 2): 6,
-            (__file__, firstlineno + 3): 5,
-            (__file__, firstlineno + 4): 1,
-        }
-        self.assertEqual(self.tr.results().counts, expected)
-
-    def test_traced_func_importing(self):
-        make_fake_module()
-        self.tr.runfunc(_traced_func_importing, 2, 5)
-
-        firstlineno = _traced_func_importing.__code__.co_firstlineno
-        expected = {
-            (__file__, firstlineno + 1): 1,
-            (__file__, firstlineno + 2): 1,
-            ('fakefile.py', 2): 1,
-            ('fakefile.py', 3): 1,
-        }
-        self.assertEqual(self.tr.results().counts, expected)
-
-
-class TestRunExecCounts(unittest.TestCase):
-    """A simple sanity test of line-counting, via run (exec)"""
-    def test_tt(self):
-        self.tr = Trace(count=1, trace=0, countfuncs=0, countcallers=0)
-        code = r'''_traced_func_loop(2, 5)'''
-        code = compile(code, __file__, 'exec')
-        self.tr.run(code)
-
-        firstlineno = _traced_func_loop.__code__.co_firstlineno
-        expected = {
-            (__file__, firstlineno + 1): 1,
-            (__file__, firstlineno + 2): 6,
-            (__file__, firstlineno + 3): 5,
-            (__file__, firstlineno + 4): 1,
-        }
-
-        # When used through 'run', some other spurios counts are produced, like
-        # the settrace of threading, which we ignore, just making sure that the
-        # counts fo _traced_func_loop were right.
-        #
-        for k in expected.keys():
-            self.assertEqual(self.tr.results().counts[k], expected[k])
-
-
-class TestFuncs(unittest.TestCase):
-    """White-box testing of funcs tracing"""
-    def setUp(self):
-        self.tr = Trace(count=0, trace=0, countfuncs=1)
-        self.filemod = my_file_and_modname()
-
-    def test_simple_caller(self):
-        self.tr.runfunc(_traced_func_simple_caller, 1)
-
-        expected = {
-            self.filemod + ('_traced_func_simple_caller',): 1,
-            self.filemod + ('_traced_func_linear',): 1,
-        }
-        self.assertEqual(self.tr.results().calledfuncs, expected)
-
-    def test_loop_caller_importing(self):
-        make_fake_module()
-        self.tr.runfunc(_traced_func_importing_caller, 1)
-
-        expected = {
-            self.filemod + ('_traced_func_simple_caller',): 1,
-            self.filemod + ('_traced_func_linear',): 1,
-            self.filemod + ('_traced_func_importing_caller',): 1,
-            self.filemod + ('_traced_func_importing',): 1,
-            ('fakefile.py', 'fakefile', 'foo'): 1,
-        }
-        self.assertEqual(self.tr.results().calledfuncs, expected)
-
-
-class TestCallers(unittest.TestCase):
-    """White-box testing of callers tracing"""
-    def setUp(self):
-        self.tr = Trace(count=0, trace=0, countcallers=1)
-        self.filemod = my_file_and_modname()
-
-    def test_loop_caller_importing(self):
-        make_fake_module()
-        self.tr.runfunc(_traced_func_importing_caller, 1)
-
-        expected = {
-            ((os.path.splitext(trace.__file__)[0] + '.py', 'trace', 'Trace.runfunc'),
-                (self.filemod + ('_traced_func_importing_caller',))): 1,
-            ((self.filemod + ('_traced_func_simple_caller',)),
-                (self.filemod + ('_traced_func_linear',))): 1,
-            ((self.filemod + ('_traced_func_importing_caller',)),
-                (self.filemod + ('_traced_func_simple_caller',))): 1,
-            ((self.filemod + ('_traced_func_importing_caller',)),
-                (self.filemod + ('_traced_func_importing',))): 1,
-            ((self.filemod + ('_traced_func_importing',)),
-                ('fakefile.py', 'fakefile', 'foo')): 1,
-        }
-        self.assertEqual(self.tr.results().callers, expected)
-
-
-#------------------------------ Driver ---------------------------------------#
-
-def test_main():
-    print(__name__, type(__name__))
-    test_support.run_unittest(__name__)
-
-
-if __name__ == '__main__':
-    test_main()

From python-checkins at python.org  Wed Jul 28 18:00:40 2010
From: python-checkins at python.org (alexander.belopolsky)
Date: Wed, 28 Jul 2010 18:00:40 +0200 (CEST)
Subject: [Python-checkins] r83208 -
	python/branches/release27-maint/Lib/test/test_sys_settrace.py
Message-ID: <20100728160040.991F2EE985@mail.python.org>

Author: alexander.belopolsky
Date: Wed Jul 28 18:00:40 2010
New Revision: 83208

Log:
Fix merge of r83140 again.

Added:
   python/branches/release27-maint/Lib/test/test_sys_settrace.py
      - copied unchanged from r82503, /python/branches/release27-maint/Lib/test/test_trace.py

From python-checkins at python.org  Wed Jul 28 18:27:57 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Wed, 28 Jul 2010 18:27:57 +0200 (CEST)
Subject: [Python-checkins] r83209 - in python/branches/py3k/Lib:
	test/test_robotparser.py urllib/robotparser.py
Message-ID: <20100728162757.0267DEE982@mail.python.org>

Author: senthil.kumaran
Date: Wed Jul 28 18:27:56 2010
New Revision: 83209

Log:
Fix Issue6325 - robotparse to honor urls with query strings.



Modified:
   python/branches/py3k/Lib/test/test_robotparser.py
   python/branches/py3k/Lib/urllib/robotparser.py

Modified: python/branches/py3k/Lib/test/test_robotparser.py
==============================================================================
--- python/branches/py3k/Lib/test/test_robotparser.py	(original)
+++ python/branches/py3k/Lib/test/test_robotparser.py	Wed Jul 28 18:27:56 2010
@@ -205,6 +205,17 @@
 RobotTest(13, doc, good, bad, agent="googlebot")
 
 
+# 14. For issue #6325 (query string support)
+doc = """
+User-agent: *
+Disallow: /some/path?name=value
+"""
+
+good = ['/some/path']
+bad = ['/some/path?name=value']
+
+RobotTest(14, doc, good, bad)
+
 
 class NetworkTestCase(unittest.TestCase):
 

Modified: python/branches/py3k/Lib/urllib/robotparser.py
==============================================================================
--- python/branches/py3k/Lib/urllib/robotparser.py	(original)
+++ python/branches/py3k/Lib/urllib/robotparser.py	Wed Jul 28 18:27:56 2010
@@ -129,8 +129,10 @@
             return True
         # search for given user agent matches
         # the first match counts
-        url = urllib.parse.quote(
-            urllib.parse.urlparse(urllib.parse.unquote(url))[2])
+        parsed_url = urllib.parse.urlparse(urllib.parse.unquote(url))
+        url = urllib.parse.urlunparse(('','',parsed_url.path,
+            parsed_url.params,parsed_url.query, parsed_url.fragment))
+        url = urllib.parse.quote(url)
         if not url:
             url = "/"
         for entry in self.entries:

From python-checkins at python.org  Wed Jul 28 18:30:46 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Wed, 28 Jul 2010 18:30:46 +0200 (CEST)
Subject: [Python-checkins] r83210 - in python/branches/release31-maint:
	Lib/test/test_robotparser.py Lib/urllib/robotparser.py
Message-ID: <20100728163046.F2F84F2F6@mail.python.org>

Author: senthil.kumaran
Date: Wed Jul 28 18:30:46 2010
New Revision: 83210

Log:
Merged revisions 83209 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83209 | senthil.kumaran | 2010-07-28 21:57:56 +0530 (Wed, 28 Jul 2010) | 3 lines
  
  Fix Issue6325 - robotparse to honor urls with query strings.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Lib/test/test_robotparser.py
   python/branches/release31-maint/Lib/urllib/robotparser.py

Modified: python/branches/release31-maint/Lib/test/test_robotparser.py
==============================================================================
--- python/branches/release31-maint/Lib/test/test_robotparser.py	(original)
+++ python/branches/release31-maint/Lib/test/test_robotparser.py	Wed Jul 28 18:30:46 2010
@@ -205,6 +205,17 @@
 RobotTest(13, doc, good, bad, agent="googlebot")
 
 
+# 14. For issue #6325 (query string support)
+doc = """
+User-agent: *
+Disallow: /some/path?name=value
+"""
+
+good = ['/some/path']
+bad = ['/some/path?name=value']
+
+RobotTest(14, doc, good, bad)
+
 
 class NetworkTestCase(unittest.TestCase):
 

Modified: python/branches/release31-maint/Lib/urllib/robotparser.py
==============================================================================
--- python/branches/release31-maint/Lib/urllib/robotparser.py	(original)
+++ python/branches/release31-maint/Lib/urllib/robotparser.py	Wed Jul 28 18:30:46 2010
@@ -129,8 +129,10 @@
             return True
         # search for given user agent matches
         # the first match counts
-        url = urllib.parse.quote(
-            urllib.parse.urlparse(urllib.parse.unquote(url))[2])
+        parsed_url = urllib.parse.urlparse(urllib.parse.unquote(url))
+        url = urllib.parse.urlunparse(('','',parsed_url.path,
+            parsed_url.params,parsed_url.query, parsed_url.fragment))
+        url = urllib.parse.quote(url)
         if not url:
             url = "/"
         for entry in self.entries:

From python-checkins at python.org  Wed Jul 28 18:35:35 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Wed, 28 Jul 2010 18:35:35 +0200 (CEST)
Subject: [Python-checkins] r83211 - in python/branches/release27-maint:
	Lib/robotparser.py Lib/test/test_robotparser.py
Message-ID: <20100728163535.CABC6EE9C8@mail.python.org>

Author: senthil.kumaran
Date: Wed Jul 28 18:35:35 2010
New Revision: 83211

Log:
Merged revisions 83209 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83209 | senthil.kumaran | 2010-07-28 21:57:56 +0530 (Wed, 28 Jul 2010) | 3 lines
  
  Fix Issue6325 - robotparse to honor urls with query strings.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Lib/robotparser.py
   python/branches/release27-maint/Lib/test/test_robotparser.py

Modified: python/branches/release27-maint/Lib/robotparser.py
==============================================================================
--- python/branches/release27-maint/Lib/robotparser.py	(original)
+++ python/branches/release27-maint/Lib/robotparser.py	Wed Jul 28 18:35:35 2010
@@ -131,7 +131,12 @@
             return True
         # search for given user agent matches
         # the first match counts
-        url = urllib.quote(urlparse.urlparse(urllib.unquote(url))[2]) or "/"
+        parsed_url = urlparse.urlparse(urllib.unquote(url))
+        url = urlparse.urlunparse(('', '', parsed_url.path,
+            parsed_url.params, parsed_url.query, parsed_url.fragment))
+        url = urllib.quote(url)
+        if not url:
+            url = "/"
         for entry in self.entries:
             if entry.applies_to(useragent):
                 return entry.allowance(url)

Modified: python/branches/release27-maint/Lib/test/test_robotparser.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_robotparser.py	(original)
+++ python/branches/release27-maint/Lib/test/test_robotparser.py	Wed Jul 28 18:35:35 2010
@@ -202,6 +202,17 @@
 RobotTest(13, doc, good, bad, agent="googlebot")
 
 
+# 14. For issue #6325 (query string support)
+doc = """
+User-agent: *
+Disallow: /some/path?name=value
+"""
+
+good = ['/some/path']
+bad = ['/some/path?name=value']
+
+RobotTest(14, doc, good, bad)
+
 
 class NetworkTestCase(unittest.TestCase):
 

From python-checkins at python.org  Wed Jul 28 18:39:41 2010
From: python-checkins at python.org (florent.xicluna)
Date: Wed, 28 Jul 2010 18:39:41 +0200 (CEST)
Subject: [Python-checkins] r83212 - in python/branches/py3k/Lib:
	ctypes/test/test_callbacks.py json/encoder.py pickletools.py
	test/script_helper.py test/test_binop.py test/test_pow.py
	test/test_sys.py
Message-ID: <20100728163941.DF948EEA2C@mail.python.org>

Author: florent.xicluna
Date: Wed Jul 28 18:39:41 2010
New Revision: 83212

Log:
Syntax cleanup.


Modified:
   python/branches/py3k/Lib/ctypes/test/test_callbacks.py
   python/branches/py3k/Lib/json/encoder.py
   python/branches/py3k/Lib/pickletools.py
   python/branches/py3k/Lib/test/script_helper.py
   python/branches/py3k/Lib/test/test_binop.py
   python/branches/py3k/Lib/test/test_pow.py
   python/branches/py3k/Lib/test/test_sys.py

Modified: python/branches/py3k/Lib/ctypes/test/test_callbacks.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_callbacks.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_callbacks.py	Wed Jul 28 18:39:41 2010
@@ -164,7 +164,7 @@
         result = integrate(0.0, 1.0, CALLBACK(func), 10)
         diff = abs(result - 1./3.)
 
-        self.assertTrue(diff < 0.01, "%s not less than 0.01" % diff)
+        self.assertLess(diff, 0.01, "%s not less than 0.01" % diff)
 
     def test_issue_8959_a(self):
         from ctypes.util import find_library

Modified: python/branches/py3k/Lib/json/encoder.py
==============================================================================
--- python/branches/py3k/Lib/json/encoder.py	(original)
+++ python/branches/py3k/Lib/json/encoder.py	Wed Jul 28 18:39:41 2010
@@ -397,7 +397,7 @@
             yield 'true'
         elif o is False:
             yield 'false'
-        elif isinstance(o, (int, int)):
+        elif isinstance(o, int):
             yield str(o)
         elif isinstance(o, float):
             yield _floatstr(o)

Modified: python/branches/py3k/Lib/pickletools.py
==============================================================================
--- python/branches/py3k/Lib/pickletools.py	(original)
+++ python/branches/py3k/Lib/pickletools.py	Wed Jul 28 18:39:41 2010
@@ -733,7 +733,7 @@
 
 pyinteger_or_bool = StackObject(
                         name='int_or_bool',
-                        obtype=(int, int, bool),
+                        obtype=(int, bool),
                         doc="A Python integer object (short or long), or "
                             "a Python bool.")
 

Modified: python/branches/py3k/Lib/test/script_helper.py
==============================================================================
--- python/branches/py3k/Lib/test/script_helper.py	(original)
+++ python/branches/py3k/Lib/test/script_helper.py	Wed Jul 28 18:39:41 2010
@@ -79,7 +79,7 @@
             name_in_zip = os.path.basename(script_name)
     zip_file.write(script_name, name_in_zip)
     zip_file.close()
-    #if test.test_support.verbose:
+    #if test.support.verbose:
     #    zip_file = zipfile.ZipFile(zip_name, 'r')
     #    print 'Contents of %r:' % zip_name
     #    zip_file.printdir()
@@ -114,7 +114,7 @@
     zip_file.close()
     for name in unlink:
         os.unlink(name)
-    #if test.test_support.verbose:
+    #if test.support.verbose:
     #    zip_file = zipfile.ZipFile(zip_name, 'r')
     #    print 'Contents of %r:' % zip_name
     #    zip_file.printdir()

Modified: python/branches/py3k/Lib/test/test_binop.py
==============================================================================
--- python/branches/py3k/Lib/test/test_binop.py	(original)
+++ python/branches/py3k/Lib/test/test_binop.py	Wed Jul 28 18:39:41 2010
@@ -16,7 +16,7 @@
 
 def isnum(x):
     """Test whether an object is an instance of a built-in numeric type."""
-    for T in int, int, float, complex:
+    for T in int, float, complex:
         if isinstance(x, T):
             return 1
     return 0

Modified: python/branches/py3k/Lib/test/test_pow.py
==============================================================================
--- python/branches/py3k/Lib/test/test_pow.py	(original)
+++ python/branches/py3k/Lib/test/test_pow.py	Wed Jul 28 18:39:41 2010
@@ -18,14 +18,14 @@
                 self.assertEquals(pow(2, i), pow2)
                 if i != 30 : pow2 = pow2*2
 
-            for othertype in int, int:
+            for othertype in (int,):
                 for i in list(range(-10, 0)) + list(range(1, 10)):
                     ii = type(i)
                     for j in range(1, 11):
                         jj = -othertype(j)
                         pow(ii, jj)
 
-        for othertype in int, int, float:
+        for othertype in int, float:
             for i in range(1, 100):
                 zero = type(0)
                 exp = -othertype(i/10.0)

Modified: python/branches/py3k/Lib/test/test_sys.py
==============================================================================
--- python/branches/py3k/Lib/test/test_sys.py	(original)
+++ python/branches/py3k/Lib/test/test_sys.py	Wed Jul 28 18:39:41 2010
@@ -86,7 +86,6 @@
     # Python/pythonrun.c::PyErr_PrintEx() is tricky.
 
     def test_exit(self):
-        import subprocess
 
         self.assertRaises(TypeError, sys.exit, 42, 42)
 
@@ -529,7 +528,6 @@
         sys._clear_type_cache()
 
     def test_ioencoding(self):
-        import subprocess
         env = dict(os.environ)
 
         # Test character: cent sign, encoded as 0x4A (ASCII J) in CP424,
@@ -551,7 +549,7 @@
         # Issue #7774: Ensure that sys.executable is an empty string if argv[0]
         # has been set to an non existent program name and Python is unable to
         # retrieve the real program name
-        import subprocess
+
         # For a normal installation, it should work without 'cwd'
         # argument. For test runs in the build directory, see #7774.
         python_dir = os.path.dirname(os.path.realpath(sys.executable))

From python-checkins at python.org  Wed Jul 28 19:37:28 2010
From: python-checkins at python.org (georg.brandl)
Date: Wed, 28 Jul 2010 19:37:28 +0200 (CEST)
Subject: [Python-checkins] r83213 - python/branches/py3k/Lib/test/cfgparser.2
Message-ID: <20100728173728.10A75EEA59@mail.python.org>

Author: georg.brandl
Date: Wed Jul 28 19:37:27 2010
New Revision: 83213

Log:
Add missing file.

Added:
   python/branches/py3k/Lib/test/cfgparser.2

Added: python/branches/py3k/Lib/test/cfgparser.2
==============================================================================
--- (empty file)
+++ python/branches/py3k/Lib/test/cfgparser.2	Wed Jul 28 19:37:27 2010
@@ -0,0 +1,537 @@
+# This is the main Samba configuration file. You should read the
+# smb.conf(5) manual page in order to understand the options listed
+# here. Samba has a huge number of configurable options (perhaps too
+# many!) most of which are not shown in this example
+#
+# Any line which starts with a ; (semi-colon) or a # (hash) 
+# is a comment and is ignored. In this example we will use a #
+# for commentry and a ; for parts of the config file that you
+# may wish to enable
+#
+# NOTE: Whenever you modify this file you should run the command #"testparm" # to check that you have not made any basic syntactic #errors. 
+#
+#======================= Global Settings =====================================
+[global]
+
+# 1. Server Naming Options:
+# workgroup = NT-Domain-Name or Workgroup-Name
+   
+  workgroup = MDKGROUP
+
+# netbios name is the name you will see in "Network Neighbourhood",
+# but defaults to your hostname
+
+;  netbios name = <name_of_this_server>
+
+# server string is the equivalent of the NT Description field
+   
+  server string = Samba Server %v
+
+# Message command is run by samba when a "popup" message is sent to it.
+# The example below is for use with LinPopUp:
+; message command = /usr/bin/linpopup "%f" "%m" %s; rm %s
+
+# 2. Printing Options:
+# CHANGES TO ENABLE PRINTING ON ALL CUPS PRINTERS IN THE NETWORK
+# (as cups is now used in linux-mandrake 7.2 by default)
+# if you want to automatically load your printer list rather
+# than setting them up individually then you'll need this
+   
+   printcap name = lpstat
+   load printers = yes
+
+# It should not be necessary to spell out the print system type unless
+# yours is non-standard. Currently supported print systems include:
+# bsd, sysv, plp, lprng, aix, hpux, qnx, cups
+   
+  printing = cups
+
+# Samba 2.2 supports the Windows NT-style point-and-print feature. To
+# use this, you need to be able to upload print drivers to the samba
+# server. The printer admins (or root) may install drivers onto samba.
+# Note that this feature uses the print$ share, so you will need to 
+# enable it below.
+# This parameter works like domain admin group:
+# printer admin = @<group> <user>
+;   printer admin = @adm
+# This should work well for winbind:
+;   printer admin = @"Domain Admins"
+
+# 3. Logging Options:
+# this tells Samba to use a separate log file for each machine
+# that connects
+
+   log file = /var/log/samba/log.%m
+
+# Put a capping on the size of the log files (in Kb).
+   max log size = 50
+
+# Set the log (verbosity) level (0 <= log level <= 10)
+; log level = 3
+
+# 4. Security and Domain Membership Options:
+# This option is important for security. It allows you to restrict
+# connections to machines which are on your local network. The
+# following example restricts access to two C class networks and
+# the "loopback" interface. For more examples of the syntax see
+# the smb.conf man page. Do not enable this if (tcp/ip) name resolution #does
+# not work for all the hosts in your network.
+;   hosts allow = 192.168.1. 192.168.2. 127.
+
+  hosts allow = 127.  //note this is only my private IP address
+
+# Uncomment this if you want a guest account, you must add this to
+# /etc/passwd
+# otherwise the user "nobody" is used
+;  guest account = pcguest
+
+# Security mode. Most people will want user level security. See
+# security_level.txt for details.
+
+   security = user
+
+# Use password server option only with security = server or security = # domain
+# When using security = domain, you should use password server = *
+;   password server = 
+;   password server = *
+
+# Password Level allows matching of _n_ characters of the password for
+# all combinations of upper and lower case.
+
+  password level = 8
+
+;  username level = 8
+
+# You may wish to use password encryption. Please read
+# ENCRYPTION.txt, Win95.txt and WinNT.txt in the Samba documentation.
+# Do not enable this option unless you have read those documents
+# Encrypted passwords are required for any use of samba in a Windows NT #domain
+# The smbpasswd file is only required by a server doing authentication, #thus members of a domain do not need one.
+
+  encrypt passwords = yes
+  smb passwd file = /etc/samba/smbpasswd
+
+# The following are needed to allow password changing from Windows to
+# also update the Linux system password.
+# NOTE: Use these with 'encrypt passwords' and 'smb passwd file' above.
+# NOTE2: You do NOT need these to allow workstations to change only
+#        the encrypted SMB passwords. They allow the Unix password
+#        to be kept in sync with the SMB password.
+;  unix password sync = Yes
+# You either need to setup a passwd program and passwd chat, or
+# enable pam password change
+;  pam password change = yes
+;  passwd program = /usr/bin/passwd %u
+;  passwd chat = *New*UNIX*password* %n\n *ReType*new*UNIX*password* 
+# %n\n
+;*passwd:*all*authentication*tokens*updated*successfully*
+
+# Unix users can map to different SMB User names
+;  username map = /etc/samba/smbusers
+
+# Using the following line enables you to customise your configuration
+# on a per machine basis. The %m gets replaced with the netbios name
+# of the machine that is connecting
+;   include = /etc/samba/smb.conf.%m
+
+# Options for using winbind. Winbind allows you to do all account and
+# authentication from a Windows or samba domain controller, creating
+# accounts on the fly, and maintaining a mapping of Windows RIDs to
+# unix uid's 
+# and gid's. winbind uid and winbind gid are the only required
+# parameters.
+#
+# winbind uid is the range of uid's winbind can use when mapping RIDs #to uid's
+;  winbind uid = 10000-20000
+#
+# winbind gid is the range of uid's winbind can use when mapping RIDs
+# to gid's
+;  winbind gid = 10000-20000
+#
+# winbind separator is the character a user must use between their
+# domain name and username, defaults to "\"
+;  winbind separator = +
+#
+# winbind use default domain allows you to have winbind return
+# usernames in the form user instead of DOMAIN+user for the domain
+# listed in the workgroup parameter.
+;  winbind use default domain = yes
+#
+# template homedir determines the home directory for winbind users,
+# with %D expanding to their domain name and %U expanding to their
+# username:
+;  template homedir = /home/%D/%U
+
+# When using winbind, you may want to have samba create home
+# directories on the fly for authenticated users. Ensure that 
+# /etc/pam.d/samba is using 'service=system-auth-winbind' in pam_stack 
+# modules, and then enable obedience of pam restrictions below:
+;  obey pam restrictions = yes
+
+#
+# template shell determines the shell users authenticated by winbind #get
+;  template shell = /bin/bash
+
+# 5. Browser Control and Networking Options:
+# Most people will find that this option gives better performance.
+# See speed.txt and the manual pages for details
+
+   socket options = TCP_NODELAY SO_RCVBUF=8192 SO_SNDBUF=8192
+
+# Configure Samba to use multiple interfaces
+# If you have multiple network interfaces then you must list them
+# here. See the man page for details.
+;   interfaces = 192.168.12.2/24 192.168.13.2/24 
+
+# Configure remote browse list synchronisation here
+#  request announcement to, or browse list sync from:
+#       a specific host or from / to a whole subnet (see below)
+;   remote browse sync = 192.168.3.25 192.168.5.255
+# Cause this host to announce itself to local subnets here
+;   remote announce = 192.168.1.255 192.168.2.44
+
+# set local master to no if you don't want Samba to become a master
+# browser on your network. Otherwise the normal election rules apply
+;   local master = no
+
+# OS Level determines the precedence of this server in master browser
+# elections. The default value should be reasonable
+;   os level = 33
+
+# Domain Master specifies Samba to be the Domain Master Browser. This
+# allows Samba to collate browse lists between subnets. Don't use this
+# if you already have a Windows NT domain controller doing this job
+;   domain master = yes 
+
+# Preferred Master causes Samba to force a local browser election on
+# startup and gives it a slightly higher chance of winning the election
+;   preferred master = yes
+
+# 6. Domain Control Options:
+# Enable this if you want Samba to be a domain logon server for 
+# Windows95 workstations or Primary Domain Controller for WinNT and
+# Win2k
+
+;  domain logons = yes
+
+
+# if you enable domain logons then you may want a per-machine or
+# per user logon script
+# run a specific logon batch file per workstation (machine)
+;   logon script = %m.bat
+# run a specific logon batch file per username
+;   logon script = %U.bat
+
+# Where to store roaming profiles for WinNT and Win2k
+#        %L substitutes for this servers netbios name, %U is username
+#        You must uncomment the [Profiles] share below
+;   logon path = \\%L\Profiles\%U
+
+# Where to store roaming profiles for Win9x. Be careful with this as it
+# also impacts where Win2k finds it's /HOME share
+; logon home = \\%L\%U\.profile
+
+# The add user script is used by a domain member to add local user
+# accounts that have been authenticated by the domain controller, or by 
+# the domain controller to add local machine accounts when adding 
+# machines to the domain.
+# The script must work from the command line when replacing the macros,
+# or the operation will fail. Check that groups exist if forcing a 
+# group.
+# Script for domain controller for adding machines:
+; add user script = /usr/sbin/useradd -d /dev/null -g machines ?c
+# 'Machine Account' -s /bin/false -M %u
+# Script for domain controller with LDAP backend for adding machines 
+#(please
+# configure in /etc/samba/smbldap_conf.pm first):
+; add user script = /usr/share/samba/scripts/smbldap-useradd.pl -w ?d
+# /dev/null -g machines -c 'Machine Account' -s /bin/false %u
+# Script for domain member for adding local accounts for authenticated
+# users:
+; add user script = /usr/sbin/useradd -s /bin/false %u
+
+# Domain groups:
+# domain admin group is a list of unix users or groups who are made
+# members
+# of the Domain Admin group
+; domain admin group = root @wheel
+#
+# domain guest groups is a list of unix users or groups who are made
+# members
+# of the Domain Guests group
+; domain guest group = nobody @guest
+
+# LDAP configuration for Domain Controlling:
+# The account (dn) that samba uses to access the LDAP server
+# This account needs to have write access to the LDAP tree
+# You will need to give samba the password for this dn, by 
+# running 'smbpasswd -w mypassword'
+; ldap admin dn = cn=root,dc=mydomain,dc=com
+; ldap ssl = start_tls
+# start_tls should run on 389, but samba defaults incorrectly to 636
+; ldap port = 389
+; ldap suffix = dc=mydomain,dc=com
+; ldap server = ldap.mydomain.com
+
+
+# 7. Name Resolution Options:
+# All NetBIOS names must be resolved to IP Addresses
+# 'Name Resolve Order' allows the named resolution mechanism to be
+# specified the default order is "host lmhosts wins bcast". "host" 
+# means use the unix system gethostbyname() function call that will use 
+# either /etc/hosts OR DNS or NIS depending on the settings of 
+# /etc/host.config, /etc/nsswitch.conf
+# and the /etc/resolv.conf file. "host" therefore is system 
+# configuration dependant. This parameter is most often of use to 
+# prevent DNS lookups
+# in order to resolve NetBIOS names to IP Addresses. Use with care!
+# The example below excludes use of name resolution for machines that
+# are NOT on the local network segment  - OR - are not deliberately to 
+# be known via lmhosts or via WINS.
+; name resolve order = wins lmhosts bcast
+
+# Windows Internet Name Serving Support Section:
+# WINS Support - Tells the NMBD component of Samba to enable it's WINS
+# Server
+;   wins support = yes
+
+# WINS Server - Tells the NMBD components of Samba to be a WINS Client
+#       Note: Samba can be either a WINS Server, or a WINS Client, but 
+# NOT both
+;   wins server = w.x.y.z
+
+# WINS Proxy - Tells Samba to answer name resolution queries on
+# behalf of a non WINS capable client, for this to work there must be
+# at least one  WINS Server on the network. The default is NO.
+;   wins proxy = yes
+
+# DNS Proxy - tells Samba whether or not to try to resolve NetBIOS 
+# names  via DNS nslookups. The built-in default for versions 1.9.17 is 
+# yes, this has been changed in version 1.9.18 to no.
+
+   dns proxy = no 
+
+# 8. File Naming Options:
+# Case Preservation can be handy - system default is _no_
+# NOTE: These can be set on a per share basis
+;  preserve case = no
+;  short preserve case = no
+# Default case is normally upper case for all DOS files
+;  default case = lower
+# Be very careful with case sensitivity - it can break things!
+;  case sensitive = no
+
+# Enabling internationalization:
+# you can match a Windows code page with a UNIX character set.
+# Windows: 437 (US), 737 (GREEK), 850 (Latin1 - Western European),
+# 852 (Eastern Eu.), 861 (Icelandic), 932 (Cyrillic - Russian),
+# 936 (Japanese - Shift-JIS), 936 (Simpl. Chinese), 949 (Korean 
+# Hangul),
+# 950 (Trad. Chin.).
+# UNIX: ISO8859-1 (Western European), ISO8859-2 (Eastern Eu.),
+# ISO8859-5 (Russian Cyrillic), KOI8-R (Alt-Russ. Cyril.)
+# This is an example for french users:
+;   client code page = 850
+;   character set = ISO8859-1
+
+#============================ Share Definitions ==============================
+
+[homes]
+   comment = Home Directories
+   browseable = no
+   writable = yes
+
+# You can enable VFS recycle bin on a per share basis:
+# Uncomment the next 2 lines (make sure you create a
+# .recycle folder in the base of the share and ensure
+# all users will have write access to it. See
+# examples/VFS/recycle/REAME in samba-doc for details
+;   vfs object = /usr/lib/samba/vfs/recycle.so
+;   vfs options= /etc/samba/recycle.conf
+
+# Un-comment the following and create the netlogon directory for Domain
+# Logons
+; [netlogon]
+;   comment = Network Logon Service
+;   path = /var/lib/samba/netlogon
+;   guest ok = yes
+;   writable = no
+
+#Uncomment the following 2 lines if you would like your login scripts
+# to be created dynamically by ntlogon (check that you have it in the
+# correct location (the default of the ntlogon rpm available in
+# contribs)
+
+;root preexec = /usr/bin/ntlogon -u %U -g %G -o %a -d /var/lib/samba/netlogon
+;root postexec = rm -f /var/lib/samba/netlogon/%U.bat
+
+# Un-comment the following to provide a specific roving profile share
+# the default is to use the user's home directory
+;[Profiles]
+;    path = /var/lib/samba/profiles
+;    browseable = no
+;    guest ok = yes
+
+
+# NOTE: If you have a CUPS print system there is no need to 
+# specifically define each individual printer.
+# You must configure the samba printers with the appropriate Windows
+# drivers on your Windows clients. On the Samba server no filtering is
+# done. If you wish that the server provides the driver and the clients
+# send PostScript ("Generic PostScript Printer" under Windows), you
+# have to swap the 'print command' line below with the commented one.
+
+[printers]
+   comment = All Printers
+   path = /var/spool/samba
+   browseable = no
+# to allow user 'guest account' to print.
+   guest ok = yes
+   writable = no
+   printable = yes
+   create mode = 0700
+
+# =====================================
+# print command: see above for details.
+# =====================================
+
+   print command = lpr-cups -P %p -o raw %s -r 
+# using client side printer drivers.
+;  print command = lpr-cups -P %p %s 
+# using cups own drivers (use generic PostScript on clients).
+# The following two commands are the samba defaults for printing=cups
+# change them only if you need different options:
+;   lpq command = lpq -P %p
+;   lprm command = cancel %p-%j
+
+# This share is used for Windows NT-style point-and-print support.
+# To be able to install drivers, you need to be either root, or listed
+# in the printer admin parameter above. Note that you also need write 
+# access to the directory and share definition to be able to upload the 
+# drivers.
+# For more information on this, please see the Printing Support Section
+# of  /usr/share/doc/samba-/docs/Samba-HOWTO-Collection.pdf 
+
+[print$]
+   path = /var/lib/samba/printers
+   browseable = yes
+   read only = yes
+   write list = @adm root
+
+# A useful application of samba is to make a PDF-generation service
+# To streamline this, install windows postscript drivers (preferably 
+# colour)on the samba server, so that clients can automatically install
+# them.
+
+[pdf-generator]
+   path = /var/tmp
+   guest ok = No
+   printable = Yes
+   comment = PDF Generator (only valid users)
+   #print command = /usr/share/samba/scripts/print-pdf file path win_path recipient IP &
+   print command = /usr/share/samba/scripts/print-pdf %s ~%u \\\\\\\\%L\\\\%u %m %I &
+
+# This one is useful for people to share files
+[tmp]
+   comment = Temporary file space
+   path = /tmp
+   read only = no
+   public = yes
+   echo command = cat %s; rm %s
+
+# A publicly accessible directory, but read only, except for people in
+# the "staff" group
+
+
+
+
+;[public]
+;   comment = Public Stuff
+;   path = /home/samba/public
+;   public = yes
+;   writable = no
+;   write list = @staff
+# Audited directory through experimental VFS audit.so module:
+# Uncomment next line.
+;   vfs object = /usr/lib/samba/vfs/audit.so
+
+# Other examples. 
+#
+# A private printer, usable only by Fred. Spool data will be placed in
+# Fred's
+# home directory. Note that fred must have write access to the spool 
+# directory,
+# wherever it is.
+;[fredsprn]
+;   comment = Fred's Printer
+;   valid users = fred
+;   path = /homes/fred
+;   printer = freds_printer
+;   public = no
+;   writable = no
+;   printable = yes
+
+
+-----------------------------------------------------------
+# A private directory, usable only by Fred. Note that Fred requires 
+# write access to the directory.
+
+;[fredsdir]
+
+    [Agustin]
+;   comment = Fred's Service
+    comment = Agustin Private Files
+;   path = /usr/somewhere/private
+    path = /home/agustin/Documents
+;   valid users = fred
+    valid users = agustin
+;   public = no
+;   writable = yes
+    writable = yes
+;   printable = no
+
+
+-----------------------------------------------------------
+
+# a service which has a different directory for each machine that 
+# connects this allows you to tailor configurations to incoming 
+# machines. You could also use the %u option to tailor it by user name.
+# The %m gets replaced with the machine name that is connecting.
+;[pchome]
+;  comment = PC Directories
+;  path = /usr/pc/%m
+;  public = no
+;  writable = yes
+
+
+-----------------------------------------------------------
+# A publicly accessible directory, read/write to all users. Note that
+# all files created in the directory by users will be owned by the 
+# default user, so any user with access can delete any other user's 
+# files. Obviously this directory must be writable by the default user. 
+# Another user could of course be specified, in which case all files 
+# would be owned by that user instead.
+
+;[public]
+;   path = /usr/somewhere/else/public
+;   public = yes
+;   only guest = yes
+;   writable = yes
+;   printable = no
+
+-----------------------------------------------------------
+
+# The following two entries demonstrate how to share a directory so 
+# that two users can place files there that will be owned by the 
+# specific users. In this setup, the directory should be writable by 
+# both users and should have the sticky bit set on it to prevent abuse. 
+# Obviously this could be extended to as many users as required.
+
+;[myshare]
+;   comment = Mary's and Fred's stuff
+;   path = /usr/somewhere/shared
+;   valid users = mary fred
+;   public = no
+;   writable = yes
+;   printable = no
+;   create mask = 0765

From python-checkins at python.org  Wed Jul 28 23:23:23 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 28 Jul 2010 23:23:23 +0200 (CEST)
Subject: [Python-checkins] r83214 - python/branches/py3k/Lib/os.py
Message-ID: <20100728212323.43B2EEE9A7@mail.python.org>

Author: victor.stinner
Date: Wed Jul 28 23:23:23 2010
New Revision: 83214

Log:
#9283: Fix repr(os.environ), display unicode keys and values on POSIX systems


Modified:
   python/branches/py3k/Lib/os.py

Modified: python/branches/py3k/Lib/os.py
==============================================================================
--- python/branches/py3k/Lib/os.py	(original)
+++ python/branches/py3k/Lib/os.py	Wed Jul 28 23:23:23 2010
@@ -446,7 +446,9 @@
         return len(self.data)
 
     def __repr__(self):
-        return 'environ({!r})'.format(self.data)
+        return 'environ({})'.format(', '.join(
+            ('{!r}: {!r}'.format(self.decodekey(key), self.decodevalue(value))
+            for key, value in self.data.items())))
 
     def copy(self):
         return dict(self)

From python-checkins at python.org  Wed Jul 28 23:25:42 2010
From: python-checkins at python.org (victor.stinner)
Date: Wed, 28 Jul 2010 23:25:42 +0200 (CEST)
Subject: [Python-checkins] r83215 - python/branches/py3k/Lib/os.py
Message-ID: <20100728212542.5B8A4EE98D@mail.python.org>

Author: victor.stinner
Date: Wed Jul 28 23:25:42 2010
New Revision: 83215

Log:
Issue #9283: Oops, add missing { and } to repr(os.environ)


Modified:
   python/branches/py3k/Lib/os.py

Modified: python/branches/py3k/Lib/os.py
==============================================================================
--- python/branches/py3k/Lib/os.py	(original)
+++ python/branches/py3k/Lib/os.py	Wed Jul 28 23:25:42 2010
@@ -446,7 +446,7 @@
         return len(self.data)
 
     def __repr__(self):
-        return 'environ({})'.format(', '.join(
+        return 'environ({{{}}})'.format(', '.join(
             ('{!r}: {!r}'.format(self.decodekey(key), self.decodevalue(value))
             for key, value in self.data.items())))
 

From python-checkins at python.org  Thu Jul 29 02:29:00 2010
From: python-checkins at python.org (victor.stinner)
Date: Thu, 29 Jul 2010 02:29:00 +0200 (CEST)
Subject: [Python-checkins] r83216 - python/branches/py3k/Lib/test/test_os.py
Message-ID: <20100729002900.9489BEE992@mail.python.org>

Author: victor.stinner
Date: Thu Jul 29 02:29:00 2010
New Revision: 83216

Log:
Update test_os.py according to my last changes on _Environ.__repr__()


Modified:
   python/branches/py3k/Lib/test/test_os.py

Modified: python/branches/py3k/Lib/test/test_os.py
==============================================================================
--- python/branches/py3k/Lib/test/test_os.py	(original)
+++ python/branches/py3k/Lib/test/test_os.py	Thu Jul 29 02:29:00 2010
@@ -423,7 +423,9 @@
         """Check that the repr() of os.environ looks like environ({...})."""
         env = os.environ
         self.assertTrue(isinstance(env.data, dict))
-        self.assertEqual(repr(env), 'environ({!r})'.format(env.data))
+        self.assertEqual(repr(env), 'environ({{{}}})'.format(', '.join(
+            '{!r}: {!r}'.format(key, value)
+            for key, value in env.items())))
 
     def test_get_exec_path(self):
         defpath_list = os.defpath.split(os.pathsep)

From solipsis at pitrou.net  Thu Jul 29 04:58:09 2010
From: solipsis at pitrou.net (solipsis at pitrou.net)
Date: Thu, 29 Jul 2010 04:58:09 +0200
Subject: [Python-checkins] Daily py3k reference leaks (r83216): sum=75
Message-ID: <E1OeJJx-0002H1-EG@ap.vmr.nerim.net>

py3k results for svn r83216 (hg cset 071f9fa8f0a4)
--------------------------------------------------

test_smtpd leaked [25, 25, 25] references, sum=75


Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/antoine/py3k/refleaks/refloga4fWih', '-x']

From python-checkins at python.org  Thu Jul 29 11:11:19 2010
From: python-checkins at python.org (tarek.ziade)
Date: Thu, 29 Jul 2010 11:11:19 +0200
Subject: [Python-checkins] distutils2: removed the config module
Message-ID: <E1OeP95-0000R2-Hl@dinsdale.python.org>

tarek.ziade pushed ba37b166e4f9 to distutils2:

http://hg.python.org/distutils2/rev/ba37b166e4f9
changeset:   387:ba37b166e4f9
tag:         tip
user:        Tarek Ziade <tarek at ziade.org>
date:        Thu Jul 29 11:10:45 2010 +0200
summary:     removed the config module
files:       src/distutils2/command/register.py, src/distutils2/command/upload.py, src/distutils2/command/upload_docs.py, src/distutils2/config.py, src/distutils2/core.py, src/distutils2/tests/test_config.py, src/distutils2/tests/test_register.py, src/distutils2/tests/test_sdist.py, src/distutils2/tests/test_upload.py, src/distutils2/tests/test_upload_docs.py, src/distutils2/tests/test_util.py, src/distutils2/util.py

diff --git a/src/distutils2/command/register.py b/src/distutils2/command/register.py
--- a/src/distutils2/command/register.py
+++ b/src/distutils2/command/register.py
@@ -13,28 +13,40 @@
 import StringIO
 from warnings import warn
 
-from distutils2.core import PyPIRCCommand
+from distutils2.command.cmd import Command
 from distutils2 import log
+from distutils2.util import (metadata_to_dict, read_pypirc, generate_pypirc,
+                             DEFAULT_REPOSITORY, DEFAULT_REALM,
+                             get_pypirc_path)
 
-class register(PyPIRCCommand):
+class register(Command):
 
     description = ("register the distribution with the Python package index")
-    user_options = PyPIRCCommand.user_options + [
+    user_options = [('repository=', 'r',
+         "url of repository [default: %s]" % DEFAULT_REPOSITORY),
+        ('show-response', None,
+         'display full response text from server'),
         ('list-classifiers', None,
          'list the valid Trove classifiers'),
         ('strict', None ,
          'Will stop the registering if the meta-data are not fully compliant')
         ]
-    boolean_options = PyPIRCCommand.boolean_options + [
-        'verify', 'list-classifiers', 'strict']
+
+    boolean_options = ['show-response', 'verify', 'list-classifiers',
+                       'strict']
 
     def initialize_options(self):
-        PyPIRCCommand.initialize_options(self)
+        self.repository = None
+        self.realm = None
+        self.show_response = 0
         self.list_classifiers = 0
         self.strict = 0
 
     def finalize_options(self):
-        PyPIRCCommand.finalize_options(self)
+        if self.repository is None:
+            self.repository = DEFAULT_REPOSITORY
+        if self.realm is None:
+            self.realm = DEFAULT_REALM
         # setting options for the `check` subcommand
         check_options = {'strict': ('register', self.strict),
                          'restructuredtext': ('register', 1)}
@@ -67,7 +79,7 @@
     def _set_config(self):
         ''' Reads the configuration file and set attributes.
         '''
-        config = self._read_pypirc()
+        config = read_pypirc(self.repository, self.realm)
         if config != {}:
             self.username = config['username']
             self.password = config['password']
@@ -75,10 +87,10 @@
             self.realm = config['realm']
             self.has_config = True
         else:
-            if self.repository not in ('pypi', self.DEFAULT_REPOSITORY):
+            if self.repository not in ('pypi', DEFAULT_REPOSITORY):
                 raise ValueError('%s not found in .pypirc' % self.repository)
             if self.repository == 'pypi':
-                self.repository = self.DEFAULT_REPOSITORY
+                self.repository = DEFAULT_REPOSITORY
             self.has_config = False
 
     def classifiers(self):
@@ -177,14 +189,14 @@
                     self.announce(('I can store your PyPI login so future '
                                    'submissions will be faster.'), log.INFO)
                     self.announce('(the login will be stored in %s)' % \
-                                  self._get_rc_file(), log.INFO)
+                                  get_pypirc_path(), log.INFO)
                     choice = 'X'
                     while choice.lower() not in 'yn':
                         choice = raw_input('Save your login (y/N)?')
                         if not choice:
                             choice = 'n'
                     if choice.lower() == 'y':
-                        self._store_pypirc(username, password)
+                        generate_pypirc(username, password)
 
         elif choice == '2':
             data = {':action': 'user'}
@@ -221,7 +233,7 @@
     def build_post_data(self, action):
         # figure the data to send - the metadata plus some additional
         # information used by the package server
-        data = self._metadata_to_pypy_dict()
+        data = metadata_to_dict(self.distribution.metadata)
         data[':action'] = action
         return data
 
diff --git a/src/distutils2/command/upload.py b/src/distutils2/command/upload.py
--- a/src/distutils2/command/upload.py
+++ b/src/distutils2/command/upload.py
@@ -14,24 +14,34 @@
     from md5 import md5
 
 from distutils2.errors import DistutilsOptionError
-from distutils2.core import PyPIRCCommand
 from distutils2.util import spawn
 from distutils2 import log
+from distutils2.command.cmd import Command
+from distutils2 import log
+from distutils2.util import (metadata_to_dict, read_pypirc,
+                             DEFAULT_REPOSITORY, DEFAULT_REALM)
 
-class upload(PyPIRCCommand):
+
+class upload(Command):
 
     description = "upload binary package to PyPI"
 
-    user_options = PyPIRCCommand.user_options + [
+    user_options = [('repository=', 'r',
+         "url of repository [default: %s]" % \
+            DEFAULT_REPOSITORY),
+        ('show-response', None,
+         'display full response text from server'),
         ('sign', 's',
          'sign files to upload using gpg'),
         ('identity=', 'i', 'GPG identity used to sign files'),
         ]
 
-    boolean_options = PyPIRCCommand.boolean_options + ['sign']
+    boolean_options = ['show-response', 'sign']
 
     def initialize_options(self):
-        PyPIRCCommand.initialize_options(self)
+        self.repository = None
+        self.realm = None
+        self.show_response = 0
         self.username = ''
         self.password = ''
         self.show_response = 0
@@ -39,12 +49,15 @@
         self.identity = None
 
     def finalize_options(self):
-        PyPIRCCommand.finalize_options(self)
+        if self.repository is None:
+            self.repository = DEFAULT_REPOSITORY
+        if self.realm is None:
+            self.realm = DEFAULT_REALM
         if self.identity and not self.sign:
             raise DistutilsOptionError(
                 "Must use --sign for --identity to have meaning"
             )
-        config = self._read_pypirc()
+        config = read_pypirc(self.repository, self.realm)
         if config != {}:
             self.username = config['username']
             self.password = config['password']
@@ -85,7 +98,7 @@
         # register a new release
         content = open(filename,'rb').read()
 
-        data = self._metadata_to_pypy_dict()
+        data = metadata_to_dict(self.distribution.metadata)
 
         # extra upload infos
         data[':action'] = 'file_upload'
diff --git a/src/distutils2/command/upload_docs.py b/src/distutils2/command/upload_docs.py
--- a/src/distutils2/command/upload_docs.py
+++ b/src/distutils2/command/upload_docs.py
@@ -1,9 +1,11 @@
 import base64, httplib, os.path, socket, urlparse, zipfile
 from cStringIO import StringIO
+
 from distutils2 import log
 from distutils2.command.upload import upload
-from distutils2.core import PyPIRCCommand
+from distutils2.command.cmd import Command
 from distutils2.errors import DistutilsFileError
+from distutils2.util import read_pypirc, DEFAULT_REPOSITORY, DEFAULT_REALM
 
 def zip_dir(directory):
     """Compresses recursively contents of directory into a StringIO object"""
@@ -47,26 +49,33 @@
     content_type = 'multipart/form-data; boundary=%s' % boundary
     return content_type, body
 
-class upload_docs(PyPIRCCommand):
+class upload_docs(Command):
 
     user_options = [
-        ('repository=', 'r', "url of repository [default: %s]" % upload.DEFAULT_REPOSITORY),
+        ('repository=', 'r', "url of repository [default: %s]" % DEFAULT_REPOSITORY),
         ('show-response', None, 'display full response text from server'),
         ('upload-dir=', None, 'directory to upload'),
         ]
 
     def initialize_options(self):
-        PyPIRCCommand.initialize_options(self)
+        self.repository = None
+        self.realm = None
+        self.show_response = 0
         self.upload_dir = "build/docs"
+        self.username = ''
+        self.password = ''
 
     def finalize_options(self):
-        PyPIRCCommand.finalize_options(self)
+        if self.repository is None:
+            self.repository = DEFAULT_REPOSITORY
+        if self.realm is None:
+            self.realm = DEFAULT_REALM
         if self.upload_dir == None:
             build = self.get_finalized_command('build')
             self.upload_dir = os.path.join(build.build_base, "docs")
         self.announce('Using upload directory %s' % self.upload_dir)
         self.verify_upload_dir(self.upload_dir)
-        config = self._read_pypirc()
+        config = read_pypirc(self.repository, self.realm)
         if config != {}:
             self.username = config['username']
             self.password = config['password']
diff --git a/src/distutils2/config.py b/src/distutils2/config.py
deleted file mode 100644
--- a/src/distutils2/config.py
+++ /dev/null
@@ -1,155 +0,0 @@
-"""distutils.pypirc
-
-Provides the PyPIRCCommand class, the base class for the command classes
-that uses .pypirc in the distutils.command package.
-"""
-import os
-from ConfigParser import RawConfigParser
-
-from distutils2.command.cmd import Command
-
-DEFAULT_PYPIRC = """\
-[distutils]
-index-servers =
-    pypi
-
-[pypi]
-username:%s
-password:%s
-"""
-
-class PyPIRCCommand(Command):
-    """Base command that knows how to handle the .pypirc file
-    """
-    DEFAULT_REPOSITORY = 'http://pypi.python.org/pypi'
-    DEFAULT_REALM = 'pypi'
-    repository = None
-    realm = None
-
-    user_options = [
-        ('repository=', 'r',
-         "url of repository [default: %s]" % \
-            DEFAULT_REPOSITORY),
-        ('show-response', None,
-         'display full response text from server')]
-
-    boolean_options = ['show-response']
-
-    def _get_rc_file(self):
-        """Returns rc file path."""
-        return os.path.join(os.path.expanduser('~'), '.pypirc')
-
-    def _store_pypirc(self, username, password):
-        """Creates a default .pypirc file."""
-        rc = self._get_rc_file()
-        f = open(rc, 'w')
-        try:
-            f.write(DEFAULT_PYPIRC % (username, password))
-        finally:
-            f.close()
-        try:
-            os.chmod(rc, 0600)
-        except OSError:
-            # should do something better here
-            pass
-
-    def _read_pypirc(self):
-        """Reads the .pypirc file."""
-        rc = self._get_rc_file()
-        if os.path.exists(rc):
-            self.announce('Using PyPI login from %s' % rc)
-            repository = self.repository or self.DEFAULT_REPOSITORY
-            config = RawConfigParser()
-            config.read(rc)
-            sections = config.sections()
-            if 'distutils' in sections:
-                # let's get the list of servers
-                index_servers = config.get('distutils', 'index-servers')
-                _servers = [server.strip() for server in
-                            index_servers.split('\n')
-                            if server.strip() != '']
-                if _servers == []:
-                    # nothing set, let's try to get the default pypi
-                    if 'pypi' in sections:
-                        _servers = ['pypi']
-                    else:
-                        # the file is not properly defined, returning
-                        # an empty dict
-                        return {}
-                for server in _servers:
-                    current = {'server': server}
-                    current['username'] = config.get(server, 'username')
-
-                    # optional params
-                    for key, default in (('repository',
-                                          self.DEFAULT_REPOSITORY),
-                                         ('realm', self.DEFAULT_REALM),
-                                         ('password', None)):
-                        if config.has_option(server, key):
-                            current[key] = config.get(server, key)
-                        else:
-                            current[key] = default
-                    if (current['server'] == repository or
-                        current['repository'] == repository):
-                        return current
-            elif 'server-login' in sections:
-                # old format
-                server = 'server-login'
-                if config.has_option(server, 'repository'):
-                    repository = config.get(server, 'repository')
-                else:
-                    repository = self.DEFAULT_REPOSITORY
-                return {'username': config.get(server, 'username'),
-                        'password': config.get(server, 'password'),
-                        'repository': repository,
-                        'server': server,
-                        'realm': self.DEFAULT_REALM}
-
-        return {}
-
-    def _metadata_to_pypy_dict(self):
-        meta = self.distribution.metadata
-        data = {
-            'metadata_version' : meta.version,
-            'name': meta['Name'],
-            'version': meta['Version'],
-            'summary': meta['Summary'],
-            'home_page': meta['Home-page'],
-            'author': meta['Author'],
-            'author_email': meta['Author-email'],
-            'license': meta['License'],
-            'description': meta['Description'],
-            'keywords': meta['Keywords'],
-            'platform': meta['Platform'],
-            'classifier': meta['Classifier'],
-            'download_url': meta['Download-URL'],
-        }
-
-        if meta.version == '1.2':
-            data['requires_dist'] = meta['Requires-Dist']
-            data['requires_python'] = meta['Requires-Python']
-            data['requires_external'] = meta['Requires-External']
-            data['provides_dist'] = meta['Provides-Dist']
-            data['obsoletes_dist'] = meta['Obsoletes-Dist']
-            data['project_url'] = [','.join(url) for url in
-                                   meta['Project-URL']]
-
-        elif meta.version == '1.1':
-            data['provides'] = meta['Provides']
-            data['requires'] = meta['Requires']
-            data['obsoletes'] = meta['Obsoletes']
-
-        return data
-
-    def initialize_options(self):
-        """Initialize options."""
-        self.repository = None
-        self.realm = None
-        self.show_response = 0
-
-    def finalize_options(self):
-        """Finalizes options."""
-        if self.repository is None:
-            self.repository = self.DEFAULT_REPOSITORY
-        if self.realm is None:
-            self.realm = self.DEFAULT_REALM
diff --git a/src/distutils2/core.py b/src/distutils2/core.py
--- a/src/distutils2/core.py
+++ b/src/distutils2/core.py
@@ -19,7 +19,6 @@
 # Mainly import these so setup scripts can "from distutils2.core import" them.
 from distutils2.dist import Distribution
 from distutils2.command.cmd import Command
-from distutils2.config import PyPIRCCommand
 from distutils2.extension import Extension
 from distutils2.util import find_packages
 
diff --git a/src/distutils2/tests/test_config.py b/src/distutils2/tests/test_config.py
deleted file mode 100644
--- a/src/distutils2/tests/test_config.py
+++ /dev/null
@@ -1,117 +0,0 @@
-"""Tests for distutils.pypirc.pypirc."""
-import sys
-import os
-import shutil
-
-from distutils2.core import PyPIRCCommand
-from distutils2.core import Distribution
-from distutils2.log import set_threshold
-from distutils2.log import WARN
-
-from distutils2.tests import support
-from distutils2.tests.support import unittest
-
-PYPIRC = """\
-[distutils]
-
-index-servers =
-    server1
-    server2
-
-[server1]
-username:me
-password:secret
-
-[server2]
-username:meagain
-password: secret
-realm:acme
-repository:http://another.pypi/
-"""
-
-PYPIRC_OLD = """\
-[server-login]
-username:tarek
-password:secret
-"""
-
-WANTED = """\
-[distutils]
-index-servers =
-    pypi
-
-[pypi]
-username:tarek
-password:xxx
-"""
-
-
-class PyPIRCCommandTestCase(support.TempdirManager,
-                            support.LoggingSilencer,
-                            support.EnvironGuard,
-                            unittest.TestCase):
-
-    def setUp(self):
-        """Patches the environment."""
-        super(PyPIRCCommandTestCase, self).setUp()
-        self.tmp_dir = self.mkdtemp()
-        os.environ['HOME'] = self.tmp_dir
-        self.rc = os.path.join(self.tmp_dir, '.pypirc')
-        self.dist = Distribution()
-
-        class command(PyPIRCCommand):
-            def __init__(self, dist):
-                PyPIRCCommand.__init__(self, dist)
-            def initialize_options(self):
-                pass
-            finalize_options = initialize_options
-
-        self._cmd = command
-        self.old_threshold = set_threshold(WARN)
-
-    def tearDown(self):
-        """Removes the patch."""
-        set_threshold(self.old_threshold)
-        super(PyPIRCCommandTestCase, self).tearDown()
-
-    def test_server_registration(self):
-        # This test makes sure PyPIRCCommand knows how to:
-        # 1. handle several sections in .pypirc
-        # 2. handle the old format
-
-        # new format
-        self.write_file(self.rc, PYPIRC)
-        cmd = self._cmd(self.dist)
-        config = cmd._read_pypirc()
-
-        config = config.items()
-        config.sort()
-        expected = [('password', 'secret'), ('realm', 'pypi'),
-                    ('repository', 'http://pypi.python.org/pypi'),
-                    ('server', 'server1'), ('username', 'me')]
-        self.assertEqual(config, expected)
-
-        # old format
-        self.write_file(self.rc, PYPIRC_OLD)
-        config = cmd._read_pypirc()
-        config = config.items()
-        config.sort()
-        expected = [('password', 'secret'), ('realm', 'pypi'),
-                    ('repository', 'http://pypi.python.org/pypi'),
-                    ('server', 'server-login'), ('username', 'tarek')]
-        self.assertEqual(config, expected)
-
-    def test_server_empty_registration(self):
-        cmd = self._cmd(self.dist)
-        rc = cmd._get_rc_file()
-        self.assertTrue(not os.path.exists(rc))
-        cmd._store_pypirc('tarek', 'xxx')
-        self.assertTrue(os.path.exists(rc))
-        content = open(rc).read()
-        self.assertEqual(content, WANTED)
-
-def test_suite():
-    return unittest.makeSuite(PyPIRCCommandTestCase)
-
-if __name__ == "__main__":
-    unittest.main(defaultTest="test_suite")
diff --git a/src/distutils2/tests/test_register.py b/src/distutils2/tests/test_register.py
--- a/src/distutils2/tests/test_register.py
+++ b/src/distutils2/tests/test_register.py
@@ -19,7 +19,7 @@
 
 from distutils2.tests import support
 from distutils2.tests.support import unittest
-from distutils2.tests.test_config import PYPIRC, PyPIRCCommandTestCase
+
 
 PYPIRC_NOPASSWORD = """\
 [distutils]
@@ -68,10 +68,15 @@
     def read(self):
         return 'xxx'
 
-class RegisterTestCase(PyPIRCCommandTestCase):
+class RegisterTestCase(support.TempdirManager, support.EnvironGuard,
+                       unittest.TestCase):
 
     def setUp(self):
         super(RegisterTestCase, self).setUp()
+        self.tmp_dir = self.mkdtemp()
+        self.rc = os.path.join(self.tmp_dir, '.pypirc')
+        os.environ['HOME'] = self.tmp_dir
+
         # patching the password prompt
         self._old_getpass = getpass.getpass
         def _getpass(prompt):
@@ -148,8 +153,8 @@
 
         self.write_file(self.rc, PYPIRC_NOPASSWORD)
         cmd = self._get_cmd()
+        cmd.finalize_options()
         cmd._set_config()
-        cmd.finalize_options()
         cmd.send_metadata()
 
         # dist.password should be set
diff --git a/src/distutils2/tests/test_sdist.py b/src/distutils2/tests/test_sdist.py
--- a/src/distutils2/tests/test_sdist.py
+++ b/src/distutils2/tests/test_sdist.py
@@ -28,7 +28,6 @@
 from distutils2.command.sdist import show_formats
 from distutils2.core import Distribution
 from distutils2.tests.support import unittest
-from distutils2.tests.test_config import PyPIRCCommandTestCase
 from distutils2.errors import DistutilsExecError, DistutilsOptionError
 from distutils2.util import find_executable
 from distutils2.tests import support
@@ -58,12 +57,15 @@
 somecode%(sep)sdoc.txt
 """
 
-class SDistTestCase(PyPIRCCommandTestCase):
+class SDistTestCase(support.TempdirManager, support.LoggingSilencer,
+                    support.EnvironGuard, unittest.TestCase):
 
     def setUp(self):
         # PyPIRCCommandTestCase creates a temp dir already
         # and put it in self.tmp_dir
         super(SDistTestCase, self).setUp()
+        self.tmp_dir = self.mkdtemp()
+        os.environ['HOME'] = self.tmp_dir
         # setting up an environment
         self.old_path = os.getcwd()
         os.mkdir(join(self.tmp_dir, 'somecode'))
diff --git a/src/distutils2/tests/test_upload.py b/src/distutils2/tests/test_upload.py
--- a/src/distutils2/tests/test_upload.py
+++ b/src/distutils2/tests/test_upload.py
@@ -9,7 +9,6 @@
 from distutils2.tests import support
 from distutils2.tests.pypi_server import PyPIServer, PyPIServerTestCase
 from distutils2.tests.support import unittest
-from distutils2.tests.test_config import PYPIRC, PyPIRCCommandTestCase
 
 
 PYPIRC_NOPASSWORD = """\
@@ -22,7 +21,33 @@
 username:me
 """
 
-class UploadTestCase(PyPIServerTestCase, PyPIRCCommandTestCase):
+PYPIRC = """\
+[distutils]
+
+index-servers =
+    server1
+    server2
+
+[server1]
+username:me
+password:secret
+
+[server2]
+username:meagain
+password: secret
+realm:acme
+repository:http://another.pypi/
+"""
+
+
+class UploadTestCase(support.TempdirManager, support.EnvironGuard,
+                     PyPIServerTestCase):
+
+    def setUp(self):
+        super(UploadTestCase, self).setUp()
+        self.tmp_dir = self.mkdtemp()
+        self.rc = os.path.join(self.tmp_dir, '.pypirc')
+        os.environ['HOME'] = self.tmp_dir
 
     def test_finalize_options(self):
         # new format
diff --git a/src/distutils2/tests/test_upload_docs.py b/src/distutils2/tests/test_upload_docs.py
--- a/src/distutils2/tests/test_upload_docs.py
+++ b/src/distutils2/tests/test_upload_docs.py
@@ -12,7 +12,6 @@
 
 from distutils2.tests import support
 from distutils2.tests.pypi_server import PyPIServer, PyPIServerTestCase
-from distutils2.tests.test_config import PyPIRCCommandTestCase
 from distutils2.tests.support import unittest
 
 
@@ -47,10 +46,14 @@
 password = long_island
 """
 
-class UploadDocsTestCase(PyPIServerTestCase, PyPIRCCommandTestCase):
+class UploadDocsTestCase(support.TempdirManager, support.EnvironGuard,
+                         PyPIServerTestCase):
 
     def setUp(self):
         super(UploadDocsTestCase, self).setUp()
+        self.tmp_dir = self.mkdtemp()
+        self.rc = os.path.join(self.tmp_dir, '.pypirc')
+        os.environ['HOME'] = self.tmp_dir
         self.dist = Distribution()
         self.dist.metadata['Name'] = "distr-name"
         self.cmd = upload_docs(self.dist)
@@ -178,9 +181,13 @@
     def test_show_response(self):
         orig_stdout = sys.stdout
         write_args = []
+
         class MockStdIn(object):
             def write(self, arg):
                 write_args.append(arg)
+            def flush(self):
+                pass
+
         sys.stdout = MockStdIn()
         try:
             self.prepare_command()
@@ -188,8 +195,10 @@
             self.cmd.run()
         finally:
             sys.stdout = orig_stdout
+
         self.assertTrue(write_args[0], "should report the response")
-        self.assertIn(self.pypi.default_response_data + "\n", write_args[0])
+        self.assertIn(self.pypi.default_response_data + "\n",
+                      '\n'.join(write_args))
 
 def test_suite():
     return unittest.makeSuite(UploadDocsTestCase)
diff --git a/src/distutils2/tests/test_util.py b/src/distutils2/tests/test_util.py
--- a/src/distutils2/tests/test_util.py
+++ b/src/distutils2/tests/test_util.py
@@ -18,11 +18,47 @@
                              rfc822_escape, get_compiler_versions,
                              _find_exe_version, _MAC_OS_X_LD_VERSION,
                              byte_compile, find_packages, spawn, find_executable,
-                             _nt_quote_args)
+                             _nt_quote_args, get_pypirc_path, generate_pypirc,
+                             read_pypirc)
+
 from distutils2 import util
 from distutils2.tests import support
 from distutils2.tests.support import unittest
 
+
+PYPIRC = """\
+[distutils]
+index-servers =
+    pypi
+    server1
+
+[pypi]
+username:me
+password:xxxx
+
+[server1]
+repository:http://example.com
+username:tarek
+password:secret
+"""
+
+PYPIRC_OLD = """\
+[server-login]
+username:tarek
+password:secret
+"""
+
+WANTED = """\
+[distutils]
+index-servers =
+    pypi
+
+[pypi]
+username:tarek
+password:xxx
+"""
+
+
 class FakePopen(object):
     test_class = None
     def __init__(self, cmd, shell, stdout, stderr):
@@ -43,6 +79,9 @@
 
     def setUp(self):
         super(UtilTestCase, self).setUp()
+        self.tmp_dir = self.mkdtemp()
+        self.rc = os.path.join(self.tmp_dir, '.pypirc')
+        os.environ['HOME'] = self.tmp_dir
         # saving the environment
         self.name = os.name
         self.platform = sys.platform
@@ -375,6 +414,40 @@
         os.chmod(exe, 0777)
         spawn([exe])  # should work without any error
 
+    def test_server_registration(self):
+        # This test makes sure we know how to:
+        # 1. handle several sections in .pypirc
+        # 2. handle the old format
+
+        # new format
+        self.write_file(self.rc, PYPIRC)
+        config = read_pypirc()
+
+        config = config.items()
+        config.sort()
+        expected = [('password', 'xxxx'), ('realm', 'pypi'),
+                    ('repository', 'http://pypi.python.org/pypi'),
+                    ('server', 'pypi'), ('username', 'me')]
+        self.assertEqual(config, expected)
+
+        # old format
+        self.write_file(self.rc, PYPIRC_OLD)
+        config = read_pypirc()
+        config = config.items()
+        config.sort()
+        expected = [('password', 'secret'), ('realm', 'pypi'),
+                    ('repository', 'http://pypi.python.org/pypi'),
+                    ('server', 'server-login'), ('username', 'tarek')]
+        self.assertEqual(config, expected)
+
+    def test_server_empty_registration(self):
+        rc = get_pypirc_path()
+        self.assertTrue(not os.path.exists(rc))
+        generate_pypirc('tarek', 'xxx')
+        self.assertTrue(os.path.exists(rc))
+        content = open(rc).read()
+        self.assertEqual(content, WANTED)
+
 
 def test_suite():
     return unittest.makeSuite(UtilTestCase)
diff --git a/src/distutils2/util.py b/src/distutils2/util.py
--- a/src/distutils2/util.py
+++ b/src/distutils2/util.py
@@ -11,6 +11,7 @@
 import re
 from copy import copy
 from fnmatch import fnmatchcase
+from ConfigParser import RawConfigParser
 
 from distutils2.errors import (DistutilsPlatformError, DistutilsFileError,
                                DistutilsByteCompileError, DistutilsExecError)
@@ -869,3 +870,124 @@
         return None
     else:
         return executable
+
+
+DEFAULT_REPOSITORY = 'http://pypi.python.org/pypi'
+DEFAULT_REALM = 'pypi'
+DEFAULT_PYPIRC = """\
+[distutils]
+index-servers =
+    pypi
+
+[pypi]
+username:%s
+password:%s
+"""
+
+def get_pypirc_path():
+    """Returns rc file path."""
+    return os.path.join(os.path.expanduser('~'), '.pypirc')
+
+
+def generate_pypirc(username, password):
+    """Creates a default .pypirc file."""
+    rc = get_pypirc_path()
+    f = open(rc, 'w')
+    try:
+        f.write(DEFAULT_PYPIRC % (username, password))
+    finally:
+        f.close()
+    try:
+        os.chmod(rc, 0600)
+    except OSError:
+        # should do something better here
+        pass
+
+
+def read_pypirc(repository=DEFAULT_REPOSITORY, realm=DEFAULT_REALM):
+    """Reads the .pypirc file."""
+    rc = get_pypirc_path()
+    if os.path.exists(rc):
+        config = RawConfigParser()
+        config.read(rc)
+        sections = config.sections()
+        if 'distutils' in sections:
+            # let's get the list of servers
+            index_servers = config.get('distutils', 'index-servers')
+            _servers = [server.strip() for server in
+                        index_servers.split('\n')
+                        if server.strip() != '']
+            if _servers == []:
+                # nothing set, let's try to get the default pypi
+                if 'pypi' in sections:
+                    _servers = ['pypi']
+                else:
+                    # the file is not properly defined, returning
+                    # an empty dict
+                    return {}
+            for server in _servers:
+                current = {'server': server}
+                current['username'] = config.get(server, 'username')
+
+                # optional params
+                for key, default in (('repository',
+                                       DEFAULT_REPOSITORY),
+                                     ('realm', DEFAULT_REALM),
+                                     ('password', None)):
+                    if config.has_option(server, key):
+                        current[key] = config.get(server, key)
+                    else:
+                        current[key] = default
+                if (current['server'] == repository or
+                    current['repository'] == repository):
+                    return current
+        elif 'server-login' in sections:
+            # old format
+            server = 'server-login'
+            if config.has_option(server, 'repository'):
+                repository = config.get(server, 'repository')
+            else:
+                repository = DEFAULT_REPOSITORY
+
+            return {'username': config.get(server, 'username'),
+                    'password': config.get(server, 'password'),
+                    'repository': repository,
+                    'server': server,
+                    'realm': DEFAULT_REALM}
+
+    return {}
+
+
+def metadata_to_dict(meta):
+    """XXX might want to move it to the Metadata class."""
+    data = {
+        'metadata_version' : meta.version,
+        'name': meta['Name'],
+        'version': meta['Version'],
+        'summary': meta['Summary'],
+        'home_page': meta['Home-page'],
+        'author': meta['Author'],
+        'author_email': meta['Author-email'],
+        'license': meta['License'],
+        'description': meta['Description'],
+        'keywords': meta['Keywords'],
+        'platform': meta['Platform'],
+        'classifier': meta['Classifier'],
+        'download_url': meta['Download-URL'],
+    }
+
+    if meta.version == '1.2':
+        data['requires_dist'] = meta['Requires-Dist']
+        data['requires_python'] = meta['Requires-Python']
+        data['requires_external'] = meta['Requires-External']
+        data['provides_dist'] = meta['Provides-Dist']
+        data['obsoletes_dist'] = meta['Obsoletes-Dist']
+        data['project_url'] = [','.join(url) for url in
+                                meta['Project-URL']]
+
+    elif meta.version == '1.1':
+        data['provides'] = meta['Provides']
+        data['requires'] = meta['Requires']
+        data['obsoletes'] = meta['Obsoletes']
+
+    return data

--
Repository URL: http://hg.python.org/distutils2

From python-checkins at python.org  Thu Jul 29 13:15:36 2010
From: python-checkins at python.org (georg.brandl)
Date: Thu, 29 Jul 2010 13:15:36 +0200 (CEST)
Subject: [Python-checkins] r83217 - python/branches/py3k/Doc/library/re.rst
Message-ID: <20100729111536.7FE50F09C@mail.python.org>

Author: georg.brandl
Date: Thu Jul 29 13:15:36 2010
New Revision: 83217

Log:
Remove Python 1.5 compatibility note.

Modified:
   python/branches/py3k/Doc/library/re.rst

Modified: python/branches/py3k/Doc/library/re.rst
==============================================================================
--- python/branches/py3k/Doc/library/re.rst	(original)
+++ python/branches/py3k/Doc/library/re.rst	Thu Jul 29 13:15:36 2010
@@ -485,7 +485,7 @@
 
    Note that for backward compatibility, the :const:`re.U` flag still
    exists (as well as its synonym :const:`re.UNICODE` and its embedded
-   counterpart ``(?u)``), but these are redundant in Python 3.0 since
+   counterpart ``(?u)``), but these are redundant in Python 3 since
    matches are Unicode by default for strings (and Unicode matching
    isn't allowed for bytes).
 
@@ -504,7 +504,7 @@
    Make ``\w``, ``\W``, ``\b``, ``\B``, ``\s`` and ``\S`` dependent on the
    current locale. The use of this flag is discouraged as the locale mechanism
    is very unreliable, and it only handles one "culture" at a time anyway;
-   you should use Unicode matching instead, which is the default in Python 3.0
+   you should use Unicode matching instead, which is the default in Python 3
    for Unicode (str) patterns.
 
 
@@ -889,10 +889,7 @@
 
       Return a tuple containing all the subgroups of the match, from 1 up to however
       many groups are in the pattern.  The *default* argument is used for groups that
-      did not participate in the match; it defaults to ``None``.  (Incompatibility
-      note: in the original Python 1.5 release, if the tuple was one element long, a
-      string would be returned instead.  In later versions (from 1.5.1 on), a
-      singleton tuple is returned in such cases.)
+      did not participate in the match; it defaults to ``None``.
 
       For example:
 
@@ -1114,7 +1111,7 @@
    >>> re.match('Begin (\w| )*? end', s).end()
    Traceback (most recent call last):
      File "<stdin>", line 1, in ?
-     File "/usr/local/lib/python2.5/re.py", line 132, in match
+     File "/usr/local/lib/python3.2/re.py", line 132, in match
        return _compile(pattern, flags).match(string)
    RuntimeError: maximum recursion limit exceeded
 

From python-checkins at python.org  Thu Jul 29 13:49:05 2010
From: python-checkins at python.org (georg.brandl)
Date: Thu, 29 Jul 2010 13:49:05 +0200 (CEST)
Subject: [Python-checkins] r83218 - python/branches/py3k/Doc/library/re.rst
Message-ID: <20100729114905.96E23C3C3@mail.python.org>

Author: georg.brandl
Date: Thu Jul 29 13:49:05 2010
New Revision: 83218

Log:
#6538: fix regex documentation again -- use fictional class names "regex" and "match" but do not document them as classes, remove 1.5 compat info and use new default argument syntax where possible.

Modified:
   python/branches/py3k/Doc/library/re.rst

Modified: python/branches/py3k/Doc/library/re.rst
==============================================================================
--- python/branches/py3k/Doc/library/re.rst	(original)
+++ python/branches/py3k/Doc/library/re.rst	Thu Jul 29 13:49:05 2010
@@ -33,8 +33,9 @@
 string notation.
 
 It is important to note that most regular expression operations are available as
-module-level functions and :class:`RegexObject` methods.  The functions are
-shortcuts that don't require you to compile a regex object first, but miss some
+module-level functions and methods on
+:ref:`compiled regular expressions <re-objects>`.  The functions are shortcuts
+that don't require you to compile a regex object first, but miss some
 fine-tuning parameters.
 
 .. seealso::
@@ -546,21 +547,21 @@
 
 
 
-.. function:: search(pattern, string[, flags])
+.. function:: search(pattern, string, flags=0)
 
    Scan through *string* looking for a location where the regular expression
-   *pattern* produces a match, and return a corresponding :class:`MatchObject`
-   instance. Return ``None`` if no position in the string matches the pattern; note
-   that this is different from finding a zero-length match at some point in the
-   string.
+   *pattern* produces a match, and return a corresponding :ref:`match object
+   <match-objects>`.  Return ``None`` if no position in the string matches the
+   pattern; note that this is different from finding a zero-length match at some
+   point in the string.
 
 
 .. function:: match(pattern, string, flags=0)
 
    If zero or more characters at the beginning of *string* match the regular
-   expression *pattern*, return a corresponding :class:`MatchObject` instance.
-   Return ``None`` if the string does not match the pattern; note that this is
-   different from a zero-length match.
+   expression *pattern*, return a corresponding :ref:`match object
+   <match-objects>`.  Return ``None`` if the string does not match the pattern;
+   note that this is different from a zero-length match.
 
    .. note::
 
@@ -620,9 +621,9 @@
 
 .. function:: finditer(pattern, string, flags=0)
 
-   Return an :term:`iterator` yielding :class:`MatchObject` instances over all
-   non-overlapping matches for the RE *pattern* in *string*.  The *string* is
-   scanned left-to-right, and matches are returned in the order found.  Empty
+   Return an :term:`iterator` yielding :ref:`match objects <match-objects>` over
+   all non-overlapping matches for the RE *pattern* in *string*.  The *string*
+   is scanned left-to-right, and matches are returned in the order found.  Empty
    matches are included in the result unless they touch the beginning of another
    match.
 
@@ -710,107 +711,107 @@
 Regular Expression Objects
 --------------------------
 
-.. class:: RegexObject
-
-   The :class:`RegexObject` class supports the following methods and attributes:
-
-   .. method:: RegexObject.search(string[, pos[, endpos]])
+Compiled regular expression objects support the following methods and
+attributes.
 
-      Scan through *string* looking for a location where this regular expression
-      produces a match, and return a corresponding :class:`MatchObject` instance.
-      Return ``None`` if no position in the string matches the pattern; note that this
-      is different from finding a zero-length match at some point in the string.
+.. method:: regex.search(string[, pos[, endpos]])
 
-      The optional second parameter *pos* gives an index in the string where the
-      search is to start; it defaults to ``0``.  This is not completely equivalent to
-      slicing the string; the ``'^'`` pattern character matches at the real beginning
-      of the string and at positions just after a newline, but not necessarily at the
-      index where the search is to start.
+   Scan through *string* looking for a location where this regular expression
+   produces a match, and return a corresponding :ref:`match object
+   <match-objects>`.  Return ``None`` if no position in the string matches the
+   pattern; note that this is different from finding a zero-length match at some
+   point in the string.
+
+   The optional second parameter *pos* gives an index in the string where the
+   search is to start; it defaults to ``0``.  This is not completely equivalent to
+   slicing the string; the ``'^'`` pattern character matches at the real beginning
+   of the string and at positions just after a newline, but not necessarily at the
+   index where the search is to start.
+
+   The optional parameter *endpos* limits how far the string will be searched; it
+   will be as if the string is *endpos* characters long, so only the characters
+   from *pos* to ``endpos - 1`` will be searched for a match.  If *endpos* is less
+   than *pos*, no match will be found, otherwise, if *rx* is a compiled regular
+   expression object, ``rx.search(string, 0, 50)`` is equivalent to
+   ``rx.search(string[:50], 0)``.
 
-      The optional parameter *endpos* limits how far the string will be searched; it
-      will be as if the string is *endpos* characters long, so only the characters
-      from *pos* to ``endpos - 1`` will be searched for a match.  If *endpos* is less
-      than *pos*, no match will be found, otherwise, if *rx* is a compiled regular
-      expression object, ``rx.search(string, 0, 50)`` is equivalent to
-      ``rx.search(string[:50], 0)``.
-
-      >>> pattern = re.compile("d")
-      >>> pattern.search("dog")     # Match at index 0
-      <_sre.SRE_Match object at ...>
-      >>> pattern.search("dog", 1)  # No match; search doesn't include the "d"
+   >>> pattern = re.compile("d")
+   >>> pattern.search("dog")     # Match at index 0
+   <_sre.SRE_Match object at ...>
+   >>> pattern.search("dog", 1)  # No match; search doesn't include the "d"
 
 
-   .. method:: RegexObject.match(string[, pos[, endpos]])
+.. method:: regex.match(string[, pos[, endpos]])
 
-      If zero or more characters at the *beginning* of *string* match this regular
-      expression, return a corresponding :class:`MatchObject` instance.  Return
-      ``None`` if the string does not match the pattern; note that this is different
-      from a zero-length match.
+   If zero or more characters at the *beginning* of *string* match this regular
+   expression, return a corresponding :ref:`match object <match-objects>`.
+   Return ``None`` if the string does not match the pattern; note that this is
+   different from a zero-length match.
 
-      The optional *pos* and *endpos* parameters have the same meaning as for the
-      :meth:`~RegexObject.search` method.
+   The optional *pos* and *endpos* parameters have the same meaning as for the
+   :meth:`~regex.search` method.
 
-      .. note::
+   .. note::
 
-         If you want to locate a match anywhere in *string*, use
-         :meth:`~RegexObject.search` instead.
+      If you want to locate a match anywhere in *string*, use
+      :meth:`~regex.search` instead.
 
-      >>> pattern = re.compile("o")
-      >>> pattern.match("dog")      # No match as "o" is not at the start of "dog".
-      >>> pattern.match("dog", 1)   # Match as "o" is the 2nd character of "dog".
-      <_sre.SRE_Match object at ...>
+   >>> pattern = re.compile("o")
+   >>> pattern.match("dog")      # No match as "o" is not at the start of "dog".
+   >>> pattern.match("dog", 1)   # Match as "o" is the 2nd character of "dog".
+   <_sre.SRE_Match object at ...>
 
 
-   .. method:: RegexObject.split(string[, maxsplit=0])
+.. method:: regex.split(string, maxsplit=0)
 
-      Identical to the :func:`split` function, using the compiled pattern.
+   Identical to the :func:`split` function, using the compiled pattern.
 
 
-   .. method:: RegexObject.findall(string[, pos[, endpos]])
+.. method:: regex.findall(string[, pos[, endpos]])
 
-      Similar to the :func:`findall` function, using the compiled pattern, but
-      also accepts optional *pos* and *endpos* parameters that limit the search
-      region like for :meth:`match`.
+   Similar to the :func:`findall` function, using the compiled pattern, but
+   also accepts optional *pos* and *endpos* parameters that limit the search
+   region like for :meth:`match`.
 
 
-   .. method:: RegexObject.finditer(string[, pos[, endpos]])
+.. method:: regex.finditer(string[, pos[, endpos]])
 
-      Similar to the :func:`finditer` function, using the compiled pattern, but
-      also accepts optional *pos* and *endpos* parameters that limit the search
-      region like for :meth:`match`.
+   Similar to the :func:`finditer` function, using the compiled pattern, but
+   also accepts optional *pos* and *endpos* parameters that limit the search
+   region like for :meth:`match`.
 
 
-   .. method:: RegexObject.sub(repl, string[, count=0])
+.. method:: regex.sub(repl, string, count=0)
 
-      Identical to the :func:`sub` function, using the compiled pattern.
+   Identical to the :func:`sub` function, using the compiled pattern.
 
 
-   .. method:: RegexObject.subn(repl, string[, count=0])
+.. method:: regex.subn(repl, string, count=0)
 
-      Identical to the :func:`subn` function, using the compiled pattern.
+   Identical to the :func:`subn` function, using the compiled pattern.
 
 
-   .. attribute:: RegexObject.flags
+.. attribute:: regex.flags
 
-      The flags argument used when the RE object was compiled, or ``0`` if no flags
-      were provided.
+   The flags argument used when the RE object was compiled, or ``0`` if no flags
+   were provided.
 
 
-   .. attribute:: RegexObject.groups
+.. attribute:: regex.groups
 
-      The number of capturing groups in the pattern.
+   The number of capturing groups in the pattern.
 
 
-   .. attribute:: RegexObject.groupindex
+.. attribute:: regex.groupindex
 
-      A dictionary mapping any symbolic group names defined by ``(?P<id>)`` to group
-      numbers.  The dictionary is empty if no symbolic groups were used in the
-      pattern.
+   A dictionary mapping any symbolic group names defined by ``(?P<id>)`` to group
+   numbers.  The dictionary is empty if no symbolic groups were used in the
+   pattern.
 
 
-   .. attribute:: RegexObject.pattern
+.. attribute:: regex.pattern
 
-      The pattern string from which the RE object was compiled.
+   The pattern string from which the RE object was compiled.
 
 
 .. _match-objects:
@@ -818,178 +819,176 @@
 Match Objects
 -------------
 
-.. class:: MatchObject
-
-   Match Objects always have a boolean value of :const:`True`, so that you can test
-   whether e.g. :func:`match` resulted in a match with a simple if statement.  They
-   support the following methods and attributes:
-
-
-   .. method:: MatchObject.expand(template)
-
-      Return the string obtained by doing backslash substitution on the template
-      string *template*, as done by the :meth:`~RegexObject.sub` method.  Escapes
-      such as ``\n`` are converted to the appropriate characters, and numeric
-      backreferences (``\1``, ``\2``) and named backreferences (``\g<1>``,
-      ``\g<name>``) are replaced by the contents of the corresponding group.
+Match objects always have a boolean value of :const:`True`, so that you can test
+whether e.g. :func:`match` resulted in a match with a simple if statement.  They
+support the following methods and attributes:
+
+
+.. method:: match.expand(template)
+
+   Return the string obtained by doing backslash substitution on the template
+   string *template*, as done by the :meth:`~regex.sub` method.
+   Escapes such as ``\n`` are converted to the appropriate characters,
+   and numeric backreferences (``\1``, ``\2``) and named backreferences
+   (``\g<1>``, ``\g<name>``) are replaced by the contents of the
+   corresponding group.
+
+
+.. method:: match.group([group1, ...])
+
+   Returns one or more subgroups of the match.  If there is a single argument, the
+   result is a single string; if there are multiple arguments, the result is a
+   tuple with one item per argument. Without arguments, *group1* defaults to zero
+   (the whole match is returned). If a *groupN* argument is zero, the corresponding
+   return value is the entire matching string; if it is in the inclusive range
+   [1..99], it is the string matching the corresponding parenthesized group.  If a
+   group number is negative or larger than the number of groups defined in the
+   pattern, an :exc:`IndexError` exception is raised. If a group is contained in a
+   part of the pattern that did not match, the corresponding result is ``None``.
+   If a group is contained in a part of the pattern that matched multiple times,
+   the last match is returned.
+
+      >>> m = re.match(r"(\w+) (\w+)", "Isaac Newton, physicist")
+      >>> m.group(0)       # The entire match
+      'Isaac Newton'
+      >>> m.group(1)       # The first parenthesized subgroup.
+      'Isaac'
+      >>> m.group(2)       # The second parenthesized subgroup.
+      'Newton'
+      >>> m.group(1, 2)    # Multiple arguments give us a tuple.
+      ('Isaac', 'Newton')
+
+   If the regular expression uses the ``(?P<name>...)`` syntax, the *groupN*
+   arguments may also be strings identifying groups by their group name.  If a
+   string argument is not used as a group name in the pattern, an :exc:`IndexError`
+   exception is raised.
+
+   A moderately complicated example:
+
+      >>> m = re.match(r"(?P<first_name>\w+) (?P<last_name>\w+)", "Malcolm Reynolds")
+      >>> m.group('first_name')
+      'Malcolm'
+      >>> m.group('last_name')
+      'Reynolds'
+
+   Named groups can also be referred to by their index:
+
+      >>> m.group(1)
+      'Malcolm'
+      >>> m.group(2)
+      'Reynolds'
+
+   If a group matches multiple times, only the last match is accessible:
+
+      >>> m = re.match(r"(..)+", "a1b2c3")  # Matches 3 times.
+      >>> m.group(1)                        # Returns only the last match.
+      'c3'
+
+
+.. method:: match.groups(default=None)
+
+   Return a tuple containing all the subgroups of the match, from 1 up to however
+   many groups are in the pattern.  The *default* argument is used for groups that
+   did not participate in the match; it defaults to ``None``.
 
+   For example:
 
-   .. method:: MatchObject.group([group1, ...])
-
-      Returns one or more subgroups of the match.  If there is a single argument, the
-      result is a single string; if there are multiple arguments, the result is a
-      tuple with one item per argument. Without arguments, *group1* defaults to zero
-      (the whole match is returned). If a *groupN* argument is zero, the corresponding
-      return value is the entire matching string; if it is in the inclusive range
-      [1..99], it is the string matching the corresponding parenthesized group.  If a
-      group number is negative or larger than the number of groups defined in the
-      pattern, an :exc:`IndexError` exception is raised. If a group is contained in a
-      part of the pattern that did not match, the corresponding result is ``None``.
-      If a group is contained in a part of the pattern that matched multiple times,
-      the last match is returned.
-
-         >>> m = re.match(r"(\w+) (\w+)", "Isaac Newton, physicist")
-         >>> m.group(0)       # The entire match
-         'Isaac Newton'
-         >>> m.group(1)       # The first parenthesized subgroup.
-         'Isaac'
-         >>> m.group(2)       # The second parenthesized subgroup.
-         'Newton'
-         >>> m.group(1, 2)    # Multiple arguments give us a tuple.
-         ('Isaac', 'Newton')
-
-      If the regular expression uses the ``(?P<name>...)`` syntax, the *groupN*
-      arguments may also be strings identifying groups by their group name.  If a
-      string argument is not used as a group name in the pattern, an :exc:`IndexError`
-      exception is raised.
-
-      A moderately complicated example:
-
-         >>> m = re.match(r"(?P<first_name>\w+) (?P<last_name>\w+)", "Malcolm Reynolds")
-         >>> m.group('first_name')
-         'Malcolm'
-         >>> m.group('last_name')
-         'Reynolds'
-
-      Named groups can also be referred to by their index:
-
-         >>> m.group(1)
-         'Malcolm'
-         >>> m.group(2)
-         'Reynolds'
-
-      If a group matches multiple times, only the last match is accessible:
-
-         >>> m = re.match(r"(..)+", "a1b2c3")  # Matches 3 times.
-         >>> m.group(1)                        # Returns only the last match.
-         'c3'
-
-
-   .. method:: MatchObject.groups(default=None)
-
-      Return a tuple containing all the subgroups of the match, from 1 up to however
-      many groups are in the pattern.  The *default* argument is used for groups that
-      did not participate in the match; it defaults to ``None``.
-
-      For example:
-
-         >>> m = re.match(r"(\d+)\.(\d+)", "24.1632")
-         >>> m.groups()
-         ('24', '1632')
+      >>> m = re.match(r"(\d+)\.(\d+)", "24.1632")
+      >>> m.groups()
+      ('24', '1632')
 
-      If we make the decimal place and everything after it optional, not all groups
-      might participate in the match.  These groups will default to ``None`` unless
-      the *default* argument is given:
+   If we make the decimal place and everything after it optional, not all groups
+   might participate in the match.  These groups will default to ``None`` unless
+   the *default* argument is given:
 
-         >>> m = re.match(r"(\d+)\.?(\d+)?", "24")
-         >>> m.groups()      # Second group defaults to None.
-         ('24', None)
-         >>> m.groups('0')   # Now, the second group defaults to '0'.
-         ('24', '0')
+      >>> m = re.match(r"(\d+)\.?(\d+)?", "24")
+      >>> m.groups()      # Second group defaults to None.
+      ('24', None)
+      >>> m.groups('0')   # Now, the second group defaults to '0'.
+      ('24', '0')
 
 
-   .. method:: MatchObject.groupdict([default])
+.. method:: match.groupdict(default=None)
 
-      Return a dictionary containing all the *named* subgroups of the match, keyed by
-      the subgroup name.  The *default* argument is used for groups that did not
-      participate in the match; it defaults to ``None``.  For example:
+   Return a dictionary containing all the *named* subgroups of the match, keyed by
+   the subgroup name.  The *default* argument is used for groups that did not
+   participate in the match; it defaults to ``None``.  For example:
 
-         >>> m = re.match(r"(?P<first_name>\w+) (?P<last_name>\w+)", "Malcolm Reynolds")
-         >>> m.groupdict()
-         {'first_name': 'Malcolm', 'last_name': 'Reynolds'}
+      >>> m = re.match(r"(?P<first_name>\w+) (?P<last_name>\w+)", "Malcolm Reynolds")
+      >>> m.groupdict()
+      {'first_name': 'Malcolm', 'last_name': 'Reynolds'}
 
 
-   .. method:: MatchObject.start([group])
-               MatchObject.end([group])
+.. method:: match.start([group])
+            match.end([group])
 
-      Return the indices of the start and end of the substring matched by *group*;
-      *group* defaults to zero (meaning the whole matched substring). Return ``-1`` if
-      *group* exists but did not contribute to the match.  For a match object *m*, and
-      a group *g* that did contribute to the match, the substring matched by group *g*
-      (equivalent to ``m.group(g)``) is ::
+   Return the indices of the start and end of the substring matched by *group*;
+   *group* defaults to zero (meaning the whole matched substring). Return ``-1`` if
+   *group* exists but did not contribute to the match.  For a match object *m*, and
+   a group *g* that did contribute to the match, the substring matched by group *g*
+   (equivalent to ``m.group(g)``) is ::
 
-         m.string[m.start(g):m.end(g)]
+      m.string[m.start(g):m.end(g)]
 
-      Note that ``m.start(group)`` will equal ``m.end(group)`` if *group* matched a
-      null string.  For example, after ``m = re.search('b(c?)', 'cba')``,
-      ``m.start(0)`` is 1, ``m.end(0)`` is 2, ``m.start(1)`` and ``m.end(1)`` are both
-      2, and ``m.start(2)`` raises an :exc:`IndexError` exception.
+   Note that ``m.start(group)`` will equal ``m.end(group)`` if *group* matched a
+   null string.  For example, after ``m = re.search('b(c?)', 'cba')``,
+   ``m.start(0)`` is 1, ``m.end(0)`` is 2, ``m.start(1)`` and ``m.end(1)`` are both
+   2, and ``m.start(2)`` raises an :exc:`IndexError` exception.
 
-      An example that will remove *remove_this* from email addresses:
+   An example that will remove *remove_this* from email addresses:
 
-         >>> email = "tony at tiremove_thisger.net"
-         >>> m = re.search("remove_this", email)
-         >>> email[:m.start()] + email[m.end():]
-         'tony at tiger.net'
+      >>> email = "tony at tiremove_thisger.net"
+      >>> m = re.search("remove_this", email)
+      >>> email[:m.start()] + email[m.end():]
+      'tony at tiger.net'
 
 
-   .. method:: MatchObject.span([group])
+.. method:: match.span([group])
 
-      For :class:`MatchObject` *m*, return the 2-tuple ``(m.start(group),
-      m.end(group))``. Note that if *group* did not contribute to the match, this is
-      ``(-1, -1)``.  *group* defaults to zero, the entire match.
+   For a match *m*, return the 2-tuple ``(m.start(group), m.end(group))``. Note
+   that if *group* did not contribute to the match, this is ``(-1, -1)``.
+   *group* defaults to zero, the entire match.
 
 
-   .. attribute:: MatchObject.pos
+.. attribute:: match.pos
 
-      The value of *pos* which was passed to the :meth:`~RegexObject.search` or
-      :meth:`~RegexObject.match` method of the :class:`RegexObject`.  This is the
-      index into the string at which the RE engine started looking for a match.
+   The value of *pos* which was passed to the :meth:`~regex.search` or
+   :meth:`~regex.match` method of a :ref:`match object <match-objects>`.  This
+   is the index into the string at which the RE engine started looking for a
+   match.
 
 
-   .. attribute:: MatchObject.endpos
+.. attribute:: match.endpos
 
-      The value of *endpos* which was passed to the :meth:`~RegexObject.search` or
-      :meth:`~RegexObject.match` method of the :class:`RegexObject`.  This is the
-      index into the string beyond which the RE engine will not go.
+   The value of *endpos* which was passed to the :meth:`~regex.search` or
+   :meth:`~regex.match` method of a :ref:`match object <match-objects>`.  This
+   is the index into the string beyond which the RE engine will not go.
 
 
-   .. attribute:: MatchObject.lastindex
+.. attribute:: match.lastindex
 
-      The integer index of the last matched capturing group, or ``None`` if no group
-      was matched at all. For example, the expressions ``(a)b``, ``((a)(b))``, and
-      ``((ab))`` will have ``lastindex == 1`` if applied to the string ``'ab'``, while
-      the expression ``(a)(b)`` will have ``lastindex == 2``, if applied to the same
-      string.
+   The integer index of the last matched capturing group, or ``None`` if no group
+   was matched at all. For example, the expressions ``(a)b``, ``((a)(b))``, and
+   ``((ab))`` will have ``lastindex == 1`` if applied to the string ``'ab'``, while
+   the expression ``(a)(b)`` will have ``lastindex == 2``, if applied to the same
+   string.
 
 
-   .. attribute:: MatchObject.lastgroup
+.. attribute:: match.lastgroup
 
-      The name of the last matched capturing group, or ``None`` if the group didn't
-      have a name, or if no group was matched at all.
+   The name of the last matched capturing group, or ``None`` if the group didn't
+   have a name, or if no group was matched at all.
 
 
-   .. attribute:: MatchObject.re
+.. attribute:: match.re
 
-      The regular expression object whose :meth:`~RegexObject.match` or
-      :meth:`~RegexObject.search` method produced this :class:`MatchObject`
-      instance.
+   The regular expression object whose :meth:`~regex.match` or
+   :meth:`~regex.search` method produced this match instance.
 
 
-   .. attribute:: MatchObject.string
+.. attribute:: match.string
 
-      The string passed to :meth:`~RegexObject.match` or
-      :meth:`~RegexObject.search`.
+   The string passed to :meth:`~regex.match` or :meth:`~regex.search`.
 
 
 Examples
@@ -1035,8 +1034,7 @@
    "<Match: '354aa', groups=('a',)>"
 
 To find out what card the pair consists of, one could use the
-:meth:`~MatchObject.group` method of :class:`MatchObject` in the following
-manner:
+:meth:`~match.group` method of the match object in the following manner:
 
 .. doctest::
 
@@ -1250,10 +1248,10 @@
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
 If one wants more information about all matches of a pattern than the matched
-text, :func:`finditer` is useful as it provides instances of
-:class:`MatchObject` instead of strings.  Continuing with the previous example,
-if one was a writer who wanted to find all of the adverbs *and their positions*
-in some text, he or she would use :func:`finditer` in the following manner:
+text, :func:`finditer` is useful as it provides :ref:`match objects
+<match-objects>` instead of strings.  Continuing with the previous example, if
+one was a writer who wanted to find all of the adverbs *and their positions* in
+some text, he or she would use :func:`finditer` in the following manner:
 
    >>> text = "He was carefully disguised but captured quickly by police."
    >>> for m in re.finditer(r"\w+ly", text):

From python-checkins at python.org  Thu Jul 29 13:56:20 2010
From: python-checkins at python.org (georg.brandl)
Date: Thu, 29 Jul 2010 13:56:20 +0200 (CEST)
Subject: [Python-checkins] r83219 - python/branches/py3k/Lib/configparser.py
Message-ID: <20100729115620.BDCF7EE99B@mail.python.org>

Author: georg.brandl
Date: Thu Jul 29 13:56:20 2010
New Revision: 83219

Log:
Fix for r83202: improve the handling of empty lines.

Modified:
   python/branches/py3k/Lib/configparser.py

Modified: python/branches/py3k/Lib/configparser.py
==============================================================================
--- python/branches/py3k/Lib/configparser.py	(original)
+++ python/branches/py3k/Lib/configparser.py	Thu Jul 29 13:56:20 2010
@@ -558,7 +558,7 @@
         indent_level = 0
         e = None                              # None, or an exception
         for lineno, line in enumerate(fp, start=1):
-            # strip prefix-only comments
+            # strip full line comments
             comment_start = None
             for prefix in self._startonly_comment_prefixes:
                 if line.strip().startswith(prefix):
@@ -572,11 +572,14 @@
                     break
             value = line[:comment_start].strip()
             if not value:
-                if self._empty_lines_in_values and comment_start is None:
+                if self._empty_lines_in_values:
                     # add empty line to the value, but only if there was no
                     # comment on the line
-                    if cursect is not None and optname:
-                        cursect[optname].append('\n')
+                    if (comment_start is None and
+                        cursect is not None and
+                        optname and
+                        cursect[optname] is not None):
+                        cursect[optname].append('') # newlines added at join
                 else:
                     # empty line marks end of value
                     indent_level = sys.maxsize
@@ -643,9 +646,7 @@
         for options in all_sections:
             for name, val in options.items():
                 if isinstance(val, list):
-                    if val[-1] == '\n':
-                        val = val[:-1]
-                    options[name] = '\n'.join(val)
+                    options[name] = '\n'.join(val).rstrip()
 
     def _handle_error(self, exc, fpname, lineno, line):
         if not exc:

From python-checkins at python.org  Thu Jul 29 14:17:40 2010
From: python-checkins at python.org (georg.brandl)
Date: Thu, 29 Jul 2010 14:17:40 +0200 (CEST)
Subject: [Python-checkins] r83220 - in python/branches/py3k:
	Doc/library/configparser.rst Lib/configparser.py
	Lib/test/cfgparser.3 Lib/test/test_cfgparser.py Misc/NEWS
Message-ID: <20100729121740.C185AF4AB@mail.python.org>

Author: georg.brandl
Date: Thu Jul 29 14:17:40 2010
New Revision: 83220

Log:
#9411: allow selecting an encoding for configparser files.  Also adds a new test config file to test special cases.

Added:
   python/branches/py3k/Lib/test/cfgparser.3
Modified:
   python/branches/py3k/Doc/library/configparser.rst
   python/branches/py3k/Lib/configparser.py
   python/branches/py3k/Lib/test/test_cfgparser.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Doc/library/configparser.rst
==============================================================================
--- python/branches/py3k/Doc/library/configparser.rst	(original)
+++ python/branches/py3k/Doc/library/configparser.rst	Thu Jul 29 14:17:40 2010
@@ -286,25 +286,29 @@
    :const:`True`; otherwise return :const:`False`.
 
 
-.. method:: RawConfigParser.read(filenames)
+.. method:: RawConfigParser.read(filenames, encoding=None)
 
    Attempt to read and parse a list of filenames, returning a list of filenames
-   which were successfully parsed.  If *filenames* is a string,
-   it is treated as a single filename. If a file named in *filenames* cannot be
-   opened, that file will be ignored.  This is designed so that you can specify a
-   list of potential configuration file locations (for example, the current
-   directory, the user's home directory, and some system-wide directory), and all
-   existing configuration files in the list will be read.  If none of the named
-   files exist, the :class:`ConfigParser` instance will contain an empty dataset.
-   An application which requires initial values to be loaded from a file should
-   load the required file or files using :meth:`readfp` before calling :meth:`read`
-   for any optional files::
+   which were successfully parsed.  If *filenames* is a string, it is treated as
+   a single filename.  If a file named in *filenames* cannot be opened, that
+   file will be ignored.  This is designed so that you can specify a list of
+   potential configuration file locations (for example, the current directory,
+   the user's home directory, and some system-wide directory), and all existing
+   configuration files in the list will be read.  If none of the named files
+   exist, the :class:`ConfigParser` instance will contain an empty dataset.  An
+   application which requires initial values to be loaded from a file should
+   load the required file or files using :meth:`readfp` before calling
+   :meth:`read` for any optional files::
 
       import configparser, os
 
       config = configparser.ConfigParser()
       config.readfp(open('defaults.cfg'))
-      config.read(['site.cfg', os.path.expanduser('~/.myapp.cfg')])
+      config.read(['site.cfg', os.path.expanduser('~/.myapp.cfg')], encoding='cp1250')
+
+   .. versionadded:: 3.2
+      The *encoding* parameter.  Previously, all files were read using the
+      default encoding for :func:`open`.
 
 
 .. method:: RawConfigParser.readfp(fp, filename=None)

Modified: python/branches/py3k/Lib/configparser.py
==============================================================================
--- python/branches/py3k/Lib/configparser.py	(original)
+++ python/branches/py3k/Lib/configparser.py	Thu Jul 29 14:17:40 2010
@@ -61,7 +61,7 @@
     options(section)
         Return list of configuration options for the named section.
 
-    read(filenames)
+    read(filenames, encoding=None)
         Read and parse the list of named configuration files, given by
         name.  A single filename is also allowed.  Non-existing files
         are ignored.  Return list of successfully read files.
@@ -369,7 +369,7 @@
             del opts['__name__']
         return list(opts.keys())
 
-    def read(self, filenames):
+    def read(self, filenames, encoding=None):
         """Read and parse a filename or a list of filenames.
 
         Files that cannot be opened are silently ignored; this is
@@ -386,7 +386,7 @@
         read_ok = []
         for filename in filenames:
             try:
-                fp = open(filename)
+                fp = open(filename, encoding=encoding)
             except IOError:
                 continue
             self._read(fp, filename)

Added: python/branches/py3k/Lib/test/cfgparser.3
==============================================================================
--- (empty file)
+++ python/branches/py3k/Lib/test/cfgparser.3	Thu Jul 29 14:17:40 2010
@@ -0,0 +1,69 @@
+  # INI with as many tricky parts as possible
+  # Most of them could not be used before 3.2
+     
+  # This will be parsed with the following options
+    # delimiters = {'='}
+    # comment_prefixes = {'#'}
+    # allow_no_value = True
+
+[DEFAULT]
+go = %(interpolate)s
+
+[strange]
+  values = that are indented # and end with hash comments
+  other = that do continue
+    in     # and still have
+    other  # comments mixed
+    lines  # with the values
+
+
+
+
+
+[corruption]
+  value = that is 
+
+
+    actually still here
+
+
+      and holds all these weird newlines
+
+
+        # but not for the lines that are comments
+        nor the indentation
+
+  another value = # empty string
+  yet another # None!
+
+  [yeah, sections can be indented as well]
+  and that does not mean = anything
+  are they subsections = False
+  if you want subsections = use XML
+  lets use some Unicode = ???
+
+  [another one!]
+ even if values are indented like this = seriously
+yes, this still applies to = section "another one!"
+this too = are there people with configurations broken as this? 
+ beware, this is going to be a continuation
+ of the value for
+ key "this too"
+ even if it has a = character
+ this is still the continuation
+ your editor probably highlights it wrong
+ but that's life
+# let's set this value so there is no error
+# when getting all items for this section:
+interpolate = anything will do
+
+[no values here]
+# but there's this `go` in DEFAULT
+
+    [tricky interpolation]
+      interpolate = do this
+      lets = %(go)s
+      
+    [more interpolation]
+      interpolate = go shopping
+      lets = %(go)s

Modified: python/branches/py3k/Lib/test/test_cfgparser.py
==============================================================================
--- python/branches/py3k/Lib/test/test_cfgparser.py	(original)
+++ python/branches/py3k/Lib/test/test_cfgparser.py	Thu Jul 29 14:17:40 2010
@@ -533,7 +533,7 @@
         smbconf = support.findfile("cfgparser.2")
         # check when we pass a mix of readable and non-readable files:
         cf = self.newconfig()
-        parsed_files = cf.read([smbconf, "nonexistent-file"])
+        parsed_files = cf.read([smbconf, "nonexistent-file"], encoding='utf-8')
         self.assertEqual(parsed_files, [smbconf])
         sections = ['global', 'homes', 'printers',
                     'print$', 'pdf-generator', 'tmp', 'Agustin']
@@ -600,6 +600,46 @@
 class SafeConfigParserTestCaseNoValue(SafeConfigParserTestCase):
     allow_no_value = True
 
+class SafeConfigParserTestCaseTrickyFile(CfgParserTestCaseClass):
+    config_class = configparser.SafeConfigParser
+    delimiters = {'='}
+    comment_prefixes = {'#'}
+    allow_no_value = True
+
+    def test_cfgparser_dot_3(self):
+        tricky = support.findfile("cfgparser.3")
+        cf = self.newconfig()
+        self.assertEqual(len(cf.read(tricky, encoding='utf-8')), 1)
+        self.assertEqual(cf.sections(), ['strange',
+                                         'corruption',
+                                         'yeah, sections can be '
+                                         'indented as well',
+                                         'another one!',
+                                         'no values here',
+                                         'tricky interpolation',
+                                         'more interpolation'])
+        #self.assertEqual(cf.getint('DEFAULT', 'go', vars={'interpolate': '-1'}),
+        #                 -1)
+        self.assertEqual(len(cf.get('strange', 'other').split('\n')), 4)
+        self.assertEqual(len(cf.get('corruption', 'value').split('\n')), 10)
+        longname = 'yeah, sections can be indented as well'
+        self.assertFalse(cf.getboolean(longname, 'are they subsections'))
+        self.assertEquals(cf.get(longname, 'lets use some Unicode'),
+                                           '???')
+        self.assertEqual(len(cf.items('another one!')), 5) # 4 in section and
+                                                           # `go` from DEFAULT
+        with self.assertRaises(configparser.InterpolationMissingOptionError):
+            cf.items('no values here')
+        self.assertEqual(cf.get('tricky interpolation', 'lets'), 'do this')
+        self.assertEqual(cf.get('tricky interpolation', 'lets'),
+                         cf.get('tricky interpolation', 'go'))
+        self.assertEqual(cf.get('more interpolation', 'lets'), 'go shopping')
+
+    def test_unicode_failure(self):
+        tricky = support.findfile("cfgparser.3")
+        cf = self.newconfig()
+        with self.assertRaises(UnicodeDecodeError):
+            cf.read(tricky, encoding='ascii')
 
 class SortedTestCase(RawConfigParserTestCase):
     dict_type = SortedDict
@@ -635,10 +675,13 @@
         foo: bar # not a comment!
         # but this is a comment
         ; another comment
+        quirk: this;is not a comment
+        ; a space must precede a comment character
         """)
         cf = self.fromstring(config_string)
         self.assertEqual(cf.get('Commented Bar', 'foo'), 'bar # not a comment!')
         self.assertEqual(cf.get('Commented Bar', 'baz'), 'qwe')
+        self.assertEqual(cf.get('Commented Bar', 'quirk'), 'this;is not a comment')
 
 
 def test_main():
@@ -652,6 +695,7 @@
         SafeConfigParserTestCase,
         SafeConfigParserTestCaseNonStandardDelimiters,
         SafeConfigParserTestCaseNoValue,
+        SafeConfigParserTestCaseTrickyFile,
         SortedTestCase,
         CompatibleTestCase,
         )

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Thu Jul 29 14:17:40 2010
@@ -475,6 +475,9 @@
 Library
 -------
 
+- Issue #9411: Allow specifying an encoding for config files in the
+  configparser module.
+
 - Issue #1682942: Improvements to configparser: support alternate
   delimiters, alternate comment prefixes and empty lines in values.
 

From python-checkins at python.org  Thu Jul 29 15:18:19 2010
From: python-checkins at python.org (ronald.oussoren)
Date: Thu, 29 Jul 2010 15:18:19 +0200 (CEST)
Subject: [Python-checkins] r83221 - in python/branches/release27-maint:
	Mac/Makefile.in Misc/NEWS
Message-ID: <20100729131819.93E81EE9A0@mail.python.org>

Author: ronald.oussoren
Date: Thu Jul 29 15:18:19 2010
New Revision: 83221

Log:
Fix for issue 9392: without this patch a framework build will not install
2to3-2.7, while a versioned copy is installed of other tools and a versioned copy of2to3 is installed by python 2.6, 3.1 and the 3.2 alpha.


Modified:
   python/branches/release27-maint/Mac/Makefile.in
   python/branches/release27-maint/Misc/NEWS

Modified: python/branches/release27-maint/Mac/Makefile.in
==============================================================================
--- python/branches/release27-maint/Mac/Makefile.in	(original)
+++ python/branches/release27-maint/Mac/Makefile.in	Thu Jul 29 15:18:19 2010
@@ -70,9 +70,10 @@
 	if [ ! -d "$(DESTDIR)$(FRAMEWORKUNIXTOOLSPREFIX)/bin" ]; then  \
 		$(INSTALL) -d -m $(DIRMODE) "$(DESTDIR)$(FRAMEWORKUNIXTOOLSPREFIX)/bin" ;\
 	fi
-	for fn in python pythonw idle pydoc python-config smtpd.py   \
+	for fn in python pythonw idle pydoc python-config smtpd.py 2to3 \
 		  python$(VERSION) pythonw$(VERSION) idle$(VERSION) \
-		  pydoc$(VERSION) python$(VERSION)-config smtpd$(VERSION).py ;\
+		  pydoc$(VERSION) python$(VERSION)-config smtpd$(VERSION).py \
+		  2to3-$(VERSION) ;\
 	do \
 		ln -fs "$(prefix)/bin/$${fn}" "$(DESTDIR)$(FRAMEWORKUNIXTOOLSPREFIX)/bin/$${fn}" ;\
 	done
@@ -90,6 +91,7 @@
 	do \
 		ln -fs "$(prefix)/bin/$${fn}" "$(DESTDIR)$(FRAMEWORKUNIXTOOLSPREFIX)/bin/$${fn}" ;\
 	done
+	ln -fs "$(prefix)/bin/2to3-$(VERSION)" "$(DESTDIR)$(FRAMEWORKUNIXTOOLSPREFIX)/bin/2to3-$(VERSION)" ;\
 
 # By default most tools are installed without a version in their basename, to
 # make it easier to install (and use) several python versions side-by-side move
@@ -112,6 +114,10 @@
 		mv "$(DESTDIR)$(prefix)/bin/smtpd.py" "$(DESTDIR)$(prefix)/bin/smtpd$(VERSION).py"  ;\
 		ln -sf "smtpd$(VERSION).py" "$(DESTDIR)$(prefix)/bin/smtpd.py" ;\
 	fi
+	if [ ! -h "$(DESTDIR)$(prefix)/bin/2to3" ]; then \
+		mv "$(DESTDIR)$(prefix)/bin/2to3" "$(DESTDIR)$(prefix)/bin/2to3-$(VERSION)"  ;\
+		ln -sf "2to3-$(VERSION)" "$(DESTDIR)$(prefix)/bin/2to3" ;\
+	fi
 
 
 pythonw: $(srcdir)/Tools/pythonw.c Makefile

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Thu Jul 29 15:18:19 2010
@@ -83,6 +83,8 @@
 
 - Issue #9275: The OSX installer once again installs links to binaries in ``/usr/local/bin``
 
+- Issue #9392: A framework build on OSX  will once again use a versioned name of the ``2to3`` tool, that is you can use ``2to3-2.7`` to select the Python 2.7 edition of 2to3.
+
 Tests
 -----
 

From python-checkins at python.org  Thu Jul 29 15:19:42 2010
From: python-checkins at python.org (georg.brandl)
Date: Thu, 29 Jul 2010 15:19:42 +0200 (CEST)
Subject: [Python-checkins] r83222 -
	python/branches/py3k/Lib/test/test_smtpd.py
Message-ID: <20100729131942.72F77EE9D8@mail.python.org>

Author: georg.brandl
Date: Thu Jul 29 15:19:42 2010
New Revision: 83222

Log:
Fix #9412: make list of messages an instance attribute instead of class attribute.

Modified:
   python/branches/py3k/Lib/test/test_smtpd.py

Modified: python/branches/py3k/Lib/test/test_smtpd.py
==============================================================================
--- python/branches/py3k/Lib/test/test_smtpd.py	(original)
+++ python/branches/py3k/Lib/test/test_smtpd.py	Thu Jul 29 15:19:42 2010
@@ -42,7 +42,9 @@
         pass
 
 class DummyServer(smtpd.SMTPServer):
-    messages = []
+    def __init__(self, *args):
+        smtpd.SMTPServer.__init__(self, *args)
+        self.messages = []
     def create_socket(self, family, type):
         self.family_and_type = (socket.AF_INET, socket.SOCK_STREAM)
         self.set_socket(DummySocket())

From python-checkins at python.org  Thu Jul 29 15:38:37 2010
From: python-checkins at python.org (georg.brandl)
Date: Thu, 29 Jul 2010 15:38:37 +0200 (CEST)
Subject: [Python-checkins] r83223 -
	python/branches/py3k/Doc/library/html.parser.rst
Message-ID: <20100729133837.94457EE9CE@mail.python.org>

Author: georg.brandl
Date: Thu Jul 29 15:38:37 2010
New Revision: 83223

Log:
#3874: document HTMLParser.unknown_decl().

Modified:
   python/branches/py3k/Doc/library/html.parser.rst

Modified: python/branches/py3k/Doc/library/html.parser.rst
==============================================================================
--- python/branches/py3k/Doc/library/html.parser.rst	(original)
+++ python/branches/py3k/Doc/library/html.parser.rst	Thu Jul 29 15:38:37 2010
@@ -134,10 +134,18 @@
 
 .. method:: HTMLParser.handle_decl(decl)
 
-   Method called when an SGML declaration is read by the parser.  The *decl*
-   parameter will be the entire contents of the declaration inside the ``<!``...\
-   ``>`` markup.  It is intended to be overridden by a derived class; the base
-   class implementation does nothing.
+   Method called when an SGML ``doctype`` declaration is read by the parser.
+   The *decl* parameter will be the entire contents of the declaration inside
+   the ``<!...>`` markup.  It is intended to be overridden by a derived class;
+   the base class implementation does nothing.
+
+
+.. method:: HTMLParser.unknown_decl(data)
+
+   Method called when an unrecognized SGML declaration is read by the parser.
+   The *data* parameter will be the entire contents of the declaration inside
+   the ``<!...>`` markup.  It is sometimes useful to be be overridden by a
+   derived class; the base class implementation throws an :exc:`HTMLParseError`.
 
 
 .. method:: HTMLParser.handle_pi(data)

From python-checkins at python.org  Thu Jul 29 15:56:56 2010
From: python-checkins at python.org (mark.dickinson)
Date: Thu, 29 Jul 2010 15:56:56 +0200 (CEST)
Subject: [Python-checkins] r83224 -
	python/branches/py3k/Doc/tutorial/floatingpoint.rst
Message-ID: <20100729135656.E0D2BED53@mail.python.org>

Author: mark.dickinson
Date: Thu Jul 29 15:56:56 2010
New Revision: 83224

Log:
Fix typo.

Modified:
   python/branches/py3k/Doc/tutorial/floatingpoint.rst

Modified: python/branches/py3k/Doc/tutorial/floatingpoint.rst
==============================================================================
--- python/branches/py3k/Doc/tutorial/floatingpoint.rst	(original)
+++ python/branches/py3k/Doc/tutorial/floatingpoint.rst	Thu Jul 29 15:56:56 2010
@@ -81,7 +81,7 @@
 values share the same approximation, any one of them could be displayed
 while still preserving the invariant ``eval(repr(x)) == x``.
 
-Historically, the Python prompt and built-in :func:`repr` function would chose
+Historically, the Python prompt and built-in :func:`repr` function would choose
 the one with 17 significant digits, ``0.10000000000000001``.   Starting with
 Python 3.1, Python (on most systems) is now able to choose the shortest of
 these and simply display ``0.1``.

From python-checkins at python.org  Thu Jul 29 15:59:01 2010
From: python-checkins at python.org (mark.dickinson)
Date: Thu, 29 Jul 2010 15:59:01 +0200 (CEST)
Subject: [Python-checkins] r83225 - in python/branches/release31-maint:
	Doc/tutorial/floatingpoint.rst
Message-ID: <20100729135901.D54CBED53@mail.python.org>

Author: mark.dickinson
Date: Thu Jul 29 15:59:01 2010
New Revision: 83225

Log:
Merged revisions 83224 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83224 | mark.dickinson | 2010-07-29 14:56:56 +0100 (Thu, 29 Jul 2010) | 1 line
  
  Fix typo.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Doc/tutorial/floatingpoint.rst

Modified: python/branches/release31-maint/Doc/tutorial/floatingpoint.rst
==============================================================================
--- python/branches/release31-maint/Doc/tutorial/floatingpoint.rst	(original)
+++ python/branches/release31-maint/Doc/tutorial/floatingpoint.rst	Thu Jul 29 15:59:01 2010
@@ -81,7 +81,7 @@
 values share the same approximation, any one of them could be displayed
 while still preserving the invariant ``eval(repr(x)) == x``.
 
-Historically, the Python prompt and built-in :func:`repr` function would chose
+Historically, the Python prompt and built-in :func:`repr` function would choose
 the one with 17 significant digits, ``0.10000000000000001``.   Starting with
 Python 3.1, Python (on most systems) is now able to choose the shortest of
 these and simply display ``0.1``.

From python-checkins at python.org  Thu Jul 29 16:17:12 2010
From: python-checkins at python.org (georg.brandl)
Date: Thu, 29 Jul 2010 16:17:12 +0200 (CEST)
Subject: [Python-checkins] r83226 - in python/branches/py3k:
	Doc/library/configparser.rst Lib/configparser.py
Message-ID: <20100729141712.EF4CED66A@mail.python.org>

Author: georg.brandl
Date: Thu Jul 29 16:17:12 2010
New Revision: 83226

Log:
#1090076: explain the behavior of *vars* in get() better.

Modified:
   python/branches/py3k/Doc/library/configparser.rst
   python/branches/py3k/Lib/configparser.py

Modified: python/branches/py3k/Doc/library/configparser.rst
==============================================================================
--- python/branches/py3k/Doc/library/configparser.rst	(original)
+++ python/branches/py3k/Doc/library/configparser.rst	Thu Jul 29 16:17:12 2010
@@ -44,7 +44,7 @@
 On top of the core functionality, :class:`SafeConfigParser` supports
 interpolation.  This means values can contain format strings which refer to
 other values in the same section, or values in a special ``DEFAULT`` section.
-Additional defaults can be provided on initialization and retrieval.
+Additional defaults can be provided on initialization.
 
 For example::
 
@@ -92,8 +92,7 @@
 manner by both parsers.
 
 Default values can be specified by passing them as a dictionary when
-constructing the :class:`SafeConfigParser`.  Additional defaults may be passed
-to the :meth:`get` method which will override all others.
+constructing the :class:`SafeConfigParser`.
 
 Sections are normally stored in an :class:`collections.OrderedDict` which
 maintains the order of all keys.  An alternative dictionary type can be passed
@@ -418,10 +417,13 @@
 
 .. method:: ConfigParser.get(section, option, raw=False, vars=None)
 
-   Get an *option* value for the named *section*.  All the ``'%'``
-   interpolations are expanded in the return values, based on the defaults
-   passed into the :meth:`__init__` method, as well as the options *vars*
-   provided, unless the *raw* argument is true.
+   Get an *option* value for the named *section*.  If *vars* is provided, it
+   must be a dictionary.  The *option* is looked up in *vars* (if provided),
+   *section*, and in *defaults* in that order.
+
+   All the ``'%'`` interpolations are expanded in the return values, unless the
+   *raw* argument is true.  Values for interpolation keys are looked up in the
+   same manner as the option.
 
 
 .. method:: ConfigParser.items(section, raw=False, vars=None)

Modified: python/branches/py3k/Lib/configparser.py
==============================================================================
--- python/branches/py3k/Lib/configparser.py	(original)
+++ python/branches/py3k/Lib/configparser.py	Thu Jul 29 16:17:12 2010
@@ -661,11 +661,12 @@
     def get(self, section, option, raw=False, vars=None):
         """Get an option value for a given section.
 
-        All % interpolations are expanded in the return values, based on the
-        defaults passed into the constructor, unless the optional argument
-        `raw' is true.  Additional substitutions may be provided using the
-        `vars' argument, which must be a dictionary whose contents overrides
-        any pre-existing defaults.
+        If `vars' is provided, it must be a dictionary. The option is looked up
+        in `vars' (if provided), `section', and in `defaults' in that order.
+
+        All % interpolations are expanded in the return values, unless the
+        optional argument `raw' is true.  Values for interpolation keys are
+        looked up in the same manner as the option.
 
         The section DEFAULT is special.
         """

From python-checkins at python.org  Thu Jul 29 16:23:06 2010
From: python-checkins at python.org (georg.brandl)
Date: Thu, 29 Jul 2010 16:23:06 +0200 (CEST)
Subject: [Python-checkins] r83227 -
	python/branches/py3k/Objects/unicodeobject.c
Message-ID: <20100729142306.EF921EE985@mail.python.org>

Author: georg.brandl
Date: Thu Jul 29 16:23:06 2010
New Revision: 83227

Log:
Use Py_CLEAR().

Modified:
   python/branches/py3k/Objects/unicodeobject.c

Modified: python/branches/py3k/Objects/unicodeobject.c
==============================================================================
--- python/branches/py3k/Objects/unicodeobject.c	(original)
+++ python/branches/py3k/Objects/unicodeobject.c	Thu Jul 29 16:23:06 2010
@@ -307,8 +307,7 @@
   reset:
     /* Reset the object caches */
     if (unicode->defenc) {
-        Py_DECREF(unicode->defenc);
-        unicode->defenc = NULL;
+        Py_CLEAR(unicode->defenc);
     }
     unicode->hash = -1;
 
@@ -427,8 +426,7 @@
             unicode->length = 0;
         }
         if (unicode->defenc) {
-            Py_DECREF(unicode->defenc);
-            unicode->defenc = NULL;
+            Py_CLEAR(unicode->defenc);
         }
         /* Add to free list */
         *(PyUnicodeObject **)unicode = free_list;

From python-checkins at python.org  Thu Jul 29 16:32:27 2010
From: python-checkins at python.org (georg.brandl)
Date: Thu, 29 Jul 2010 16:32:27 +0200 (CEST)
Subject: [Python-checkins] r83229 -
	python/branches/py3k/Doc/library/configparser.rst
Message-ID: <20100729143227.29ADCEEA2@mail.python.org>

Author: georg.brandl
Date: Thu Jul 29 16:32:22 2010
New Revision: 83229

Log:
#9407: document configparser.Error.

Modified:
   python/branches/py3k/Doc/library/configparser.rst

Modified: python/branches/py3k/Doc/library/configparser.rst
==============================================================================
--- python/branches/py3k/Doc/library/configparser.rst	(original)
+++ python/branches/py3k/Doc/library/configparser.rst	Thu Jul 29 16:32:22 2010
@@ -178,6 +178,11 @@
       *allow_no_value* were added.
 
 
+.. exception:: Error
+
+   Base class for all other configparser exceptions.
+
+
 .. exception:: NoSectionError
 
    Exception raised when a specified section is not found.

From python-checkins at python.org  Thu Jul 29 16:36:12 2010
From: python-checkins at python.org (georg.brandl)
Date: Thu, 29 Jul 2010 16:36:12 +0200 (CEST)
Subject: [Python-checkins] r83230 -
	python/branches/py3k/Doc/library/itertools.rst
Message-ID: <20100729143612.217A9EE982@mail.python.org>

Author: georg.brandl
Date: Thu Jul 29 16:36:11 2010
New Revision: 83230

Log:
Use correct directive and name.

Modified:
   python/branches/py3k/Doc/library/itertools.rst

Modified: python/branches/py3k/Doc/library/itertools.rst
==============================================================================
--- python/branches/py3k/Doc/library/itertools.rst	(original)
+++ python/branches/py3k/Doc/library/itertools.rst	Thu Jul 29 16:36:11 2010
@@ -99,7 +99,7 @@
                   yield element
 
 
-.. function:: itertools.chain.from_iterable(iterable)
+.. classmethod:: chain.from_iterable(iterable)
 
    Alternate constructor for :func:`chain`.  Gets chained inputs from a
    single iterable argument that is evaluated lazily.  Equivalent to::

From python-checkins at python.org  Thu Jul 29 16:46:08 2010
From: python-checkins at python.org (georg.brandl)
Date: Thu, 29 Jul 2010 16:46:08 +0200 (CEST)
Subject: [Python-checkins] r83231 - python/branches/py3k/Doc/library/dbm.rst
Message-ID: <20100729144608.1DAC6EEA24@mail.python.org>

Author: georg.brandl
Date: Thu Jul 29 16:46:07 2010
New Revision: 83231

Log:
#9397: remove mention of dbm.bsd which does not exist anymore.

Modified:
   python/branches/py3k/Doc/library/dbm.rst

Modified: python/branches/py3k/Doc/library/dbm.rst
==============================================================================
--- python/branches/py3k/Doc/library/dbm.rst	(original)
+++ python/branches/py3k/Doc/library/dbm.rst	Thu Jul 29 16:46:07 2010
@@ -5,10 +5,10 @@
    :synopsis: Interfaces to various Unix "database" formats.
 
 :mod:`dbm` is a generic interface to variants of the DBM database ---
- :mod:`dbm.gnu` or :mod:`dbm.ndbm`.  If none of these modules is installed, the
- slow-but-simple implementation in module :mod:`dbm.dumb` will be used.  There
- is a `third party interface <http://www.jcea.es/programacion/pybsddb.htm>`_ to
- the Oracle Berkely DB.
+:mod:`dbm.gnu` or :mod:`dbm.ndbm`.  If none of these modules is installed, the
+slow-but-simple implementation in module :mod:`dbm.dumb` will be used.  There
+is a `third party interface <http://www.jcea.es/programacion/pybsddb.htm>`_ to
+the Oracle Berkely DB.
 
 
 .. exception:: error
@@ -21,8 +21,8 @@
 .. function:: whichdb(filename)
 
    This functionattempts to guess which of the several simple database modules
-   available --- :mod:`dbm.bsd`, :mod:`dbm.gnu`, :mod:`dbm.ndbm` or
-   :mod:`dbm.dumb` --- should be used to open a given file.
+   available --- :mod:`dbm.gnu`, :mod:`dbm.ndbm` or :mod:`dbm.dumb` --- should
+   be used to open a given file.
 
    Returns one of the following values: ``None`` if the file can't be opened
    because it's unreadable or doesn't exist; the empty string (``''``) if the
@@ -227,10 +227,9 @@
 always stored as bytes. Printing a ``dbm`` object doesn't print the keys and
 values, and the :meth:`items` and :meth:`values` methods are not supported.
 
-This module can be used with the "classic" ndbm interface, the BSD DB
-compatibility interface, or the GNU GDBM compatibility interface. On Unix, the
-:program:`configure` script will attempt to locate the appropriate header file
-to simplify building this module.
+This module can be used with the "classic" ndbm interface or the GNU GDBM
+compatibility interface. On Unix, the :program:`configure` script will attempt
+to locate the appropriate header file to simplify building this module.
 
 .. exception:: error
 
@@ -246,9 +245,7 @@
 .. function:: open(filename[, flag[, mode]])
 
    Open a dbm database and return a ``dbm`` object.  The *filename* argument is the
-   name of the database file (without the :file:`.dir` or :file:`.pag` extensions;
-   note that the BSD DB implementation of the interface will append the extension
-   :file:`.db` and only create one file).
+   name of the database file (without the :file:`.dir` or :file:`.pag` extensions).
 
    The optional *flag* argument must be one of these values:
 

From python-checkins at python.org  Thu Jul 29 16:49:08 2010
From: python-checkins at python.org (georg.brandl)
Date: Thu, 29 Jul 2010 16:49:08 +0200 (CEST)
Subject: [Python-checkins] r83232 -
	python/branches/py3k/Doc/library/locale.rst
Message-ID: <20100729144908.CA374F399@mail.python.org>

Author: georg.brandl
Date: Thu Jul 29 16:49:08 2010
New Revision: 83232

Log:
#9388: remove ERA_YEAR which is never defined in the source code.

Modified:
   python/branches/py3k/Doc/library/locale.rst

Modified: python/branches/py3k/Doc/library/locale.rst
==============================================================================
--- python/branches/py3k/Doc/library/locale.rst	(original)
+++ python/branches/py3k/Doc/library/locale.rst	Thu Jul 29 16:49:08 2010
@@ -244,10 +244,6 @@
       specified, and therefore you should not assume knowledge of it on different
       systems.
 
-   .. data:: ERA_YEAR
-
-      Get the year in the relevant era of the locale.
-
    .. data:: ERA_D_T_FMT
 
       Get a format string for :func:`strftime` to represent dates and times in a

From python-checkins at python.org  Thu Jul 29 18:01:11 2010
From: python-checkins at python.org (georg.brandl)
Date: Thu, 29 Jul 2010 18:01:11 +0200 (CEST)
Subject: [Python-checkins] r83234 - in python/branches/py3k/Doc:
	documenting/markup.rst library/abc.rst library/contextlib.rst
	library/functools.rst library/importlib.rst
	library/unittest.rst tools/sphinxext/pyspecific.py
Message-ID: <20100729160111.6BA5DEE983@mail.python.org>

Author: georg.brandl
Date: Thu Jul 29 18:01:11 2010
New Revision: 83234

Log:
#6522: add a "decorator" directive to explicitly document decorators, and use it in a few places.

Modified:
   python/branches/py3k/Doc/documenting/markup.rst
   python/branches/py3k/Doc/library/abc.rst
   python/branches/py3k/Doc/library/contextlib.rst
   python/branches/py3k/Doc/library/functools.rst
   python/branches/py3k/Doc/library/importlib.rst
   python/branches/py3k/Doc/library/unittest.rst
   python/branches/py3k/Doc/tools/sphinxext/pyspecific.py

Modified: python/branches/py3k/Doc/documenting/markup.rst
==============================================================================
--- python/branches/py3k/Doc/documenting/markup.rst	(original)
+++ python/branches/py3k/Doc/documenting/markup.rst	Thu Jul 29 18:01:11 2010
@@ -177,6 +177,34 @@
    are modified), side effects, and possible exceptions.  A small example may be
    provided.
 
+.. describe:: decorator
+
+   Describes a decorator function.  The signature should *not* represent the
+   signature of the actual function, but the usage as a decorator.  For example,
+   given the functions
+
+   .. code-block:: python
+
+      def removename(func):
+          func.__name__ = ''
+          return func
+
+      def setnewname(name):
+          def decorator(func):
+              func.__name__ = name
+              return func
+          return decorator
+
+   the descriptions should look like this::
+
+      .. decorator:: removename
+
+         Remove name of the decorated function.
+
+      .. decorator:: setnewname(name)
+
+         Set name of the decorated function to *name*.
+
 .. describe:: class
 
    Describes a class.  The signature can include parentheses with parameters
@@ -194,6 +222,10 @@
    parameter.  The description should include similar information to that
    described for ``function``.
 
+.. describe:: decoratormethod
+
+   Same as ``decorator``, but for decorators that are methods.
+
 .. describe:: opcode
 
    Describes a Python :term:`bytecode` instruction.

Modified: python/branches/py3k/Doc/library/abc.rst
==============================================================================
--- python/branches/py3k/Doc/library/abc.rst	(original)
+++ python/branches/py3k/Doc/library/abc.rst	Thu Jul 29 18:01:11 2010
@@ -122,7 +122,7 @@
 
 It also provides the following decorators:
 
-.. function:: abstractmethod(function)
+.. decorator:: abstractmethod(function)
 
    A decorator indicating abstract methods.
 

Modified: python/branches/py3k/Doc/library/contextlib.rst
==============================================================================
--- python/branches/py3k/Doc/library/contextlib.rst	(original)
+++ python/branches/py3k/Doc/library/contextlib.rst	Thu Jul 29 18:01:11 2010
@@ -12,7 +12,7 @@
 Functions provided:
 
 
-.. function:: contextmanager(func)
+.. decorator:: contextmanager
 
    This function is a :term:`decorator` that can be used to define a factory
    function for :keyword:`with` statement context managers, without needing to

Modified: python/branches/py3k/Doc/library/functools.rst
==============================================================================
--- python/branches/py3k/Doc/library/functools.rst	(original)
+++ python/branches/py3k/Doc/library/functools.rst	Thu Jul 29 18:01:11 2010
@@ -37,7 +37,7 @@
 
    .. versionadded:: 3.2
 
-.. function:: total_ordering(cls)
+.. decorator:: total_ordering
 
    Given a class defining one or more rich comparison ordering methods, this
    class decorator supplies the rest.  This simplifies the effort involved
@@ -122,7 +122,7 @@
    than helpful.
 
 
-.. function:: wraps(wrapped, assigned=WRAPPER_ASSIGNMENTS, updated=WRAPPER_UPDATES)
+.. decorator:: wraps(wrapped, assigned=WRAPPER_ASSIGNMENTS, updated=WRAPPER_UPDATES)
 
    This is a convenience function for invoking ``partial(update_wrapper,
    wrapped=wrapped, assigned=assigned, updated=updated)`` as a function decorator

Modified: python/branches/py3k/Doc/library/importlib.rst
==============================================================================
--- python/branches/py3k/Doc/library/importlib.rst	(original)
+++ python/branches/py3k/Doc/library/importlib.rst	Thu Jul 29 18:01:11 2010
@@ -469,7 +469,7 @@
 This module contains the various objects that help in the construction of
 an :term:`importer`.
 
-.. function:: module_for_loader(method)
+.. decorator:: module_for_loader
 
     A :term:`decorator` for a :term:`loader` method,
     to handle selecting the proper
@@ -494,7 +494,7 @@
     Use of this decorator handles all the details of which module object a
     loader should initialize as specified by :pep:`302`.
 
-.. function:: set_loader(fxn)
+.. decorator:: set_loader
 
     A :term:`decorator` for a :term:`loader` method,
     to set the :attr:`__loader__`
@@ -502,7 +502,7 @@
     does nothing. It is assumed that the first positional argument to the
     wrapped method is what :attr:`__loader__` should be set to.
 
-.. function:: set_package(fxn)
+.. decorator:: set_package
 
     A :term:`decorator` for a :term:`loader` to set the :attr:`__package__`
     attribute on the module returned by the loader. If :attr:`__package__` is

Modified: python/branches/py3k/Doc/library/unittest.rst
==============================================================================
--- python/branches/py3k/Doc/library/unittest.rst	(original)
+++ python/branches/py3k/Doc/library/unittest.rst	Thu Jul 29 18:01:11 2010
@@ -621,20 +621,20 @@
 
 The following decorators implement test skipping and expected failures:
 
-.. function:: skip(reason)
+.. decorator:: skip(reason)
 
    Unconditionally skip the decorated test.  *reason* should describe why the
    test is being skipped.
 
-.. function:: skipIf(condition, reason)
+.. decorator:: skipIf(condition, reason)
 
    Skip the decorated test if *condition* is true.
 
-.. function:: skipUnless(condition, reason)
+.. decorator:: skipUnless(condition, reason)
 
    Skip the decoratored test unless *condition* is true.
 
-.. function:: expectedFailure
+.. decorator:: expectedFailure
 
    Mark the test as an expected failure.  If the test fails when run, the test
    is not counted as a failure.
@@ -1048,11 +1048,11 @@
       :attr:`exception` attribute.  This can be useful if the intention
       is to perform additional checks on the exception raised::
 
-        with self.assertRaises(SomeException) as cm:
-            do_something()
+         with self.assertRaises(SomeException) as cm:
+             do_something()
 
-        the_exception = cm.exception
-        self.assertEqual(the_exception.error_code, 3)
+         the_exception = cm.exception
+         self.assertEqual(the_exception.error_code, 3)
 
       .. versionchanged:: 3.1
          Added the ability to use :meth:`assertRaises` as a context manager.

Modified: python/branches/py3k/Doc/tools/sphinxext/pyspecific.py
==============================================================================
--- python/branches/py3k/Doc/tools/sphinxext/pyspecific.py	(original)
+++ python/branches/py3k/Doc/tools/sphinxext/pyspecific.py	Thu Jul 29 18:01:11 2010
@@ -72,6 +72,32 @@
         return [pnode]
 
 
+# Support for documenting decorators
+
+from sphinx import addnodes
+from sphinx.domains.python import PyModulelevel, PyClassmember
+
+class PyDecoratorMixin(object):
+    def handle_signature(self, sig, signode):
+        ret = super(PyDecoratorMixin, self).handle_signature(sig, signode)
+        signode.insert(0, addnodes.desc_addname('@', '@'))
+        return ret
+
+    def needs_arglist(self):
+        return False
+
+class PyDecoratorFunction(PyDecoratorMixin, PyModulelevel):
+    def run(self):
+        # a decorator function is a function after all
+        self.name = 'py:function'
+        return PyModulelevel.run(self)
+
+class PyDecoratorMethod(PyDecoratorMixin, PyClassmember):
+    def run(self):
+        self.name = 'py:method'
+        return PyClassmember.run(self)
+
+
 # Support for building "topic help" for pydoc
 
 pydoc_topic_labels = [
@@ -147,7 +173,6 @@
 # Support for documenting Opcodes
 
 import re
-from sphinx import addnodes
 
 opcode_sig_re = re.compile(r'(\w+(?:\+\d)?)(?:\s*\((.*)\))?')
 
@@ -197,3 +222,5 @@
     app.add_description_unit('pdbcommand', 'pdbcmd', '%s (pdb command)',
                              parse_pdb_command)
     app.add_description_unit('2to3fixer', '2to3fixer', '%s (2to3 fixer)')
+    app.add_directive_to_domain('py', 'decorator', PyDecoratorFunction)
+    app.add_directive_to_domain('py', 'decoratormethod', PyDecoratorMethod)

From python-checkins at python.org  Thu Jul 29 18:26:56 2010
From: python-checkins at python.org (victor.stinner)
Date: Thu, 29 Jul 2010 18:26:56 +0200 (CEST)
Subject: [Python-checkins] r83235 - python/branches/py3k/Lib/dbm/__init__.py
Message-ID: <20100729162656.C6D8EF503@mail.python.org>

Author: victor.stinner
Date: Thu Jul 29 18:26:56 2010
New Revision: 83235

Log:
#9397: remove mention of dbm.bsd which does not exist anymore.


Modified:
   python/branches/py3k/Lib/dbm/__init__.py

Modified: python/branches/py3k/Lib/dbm/__init__.py
==============================================================================
--- python/branches/py3k/Lib/dbm/__init__.py	(original)
+++ python/branches/py3k/Lib/dbm/__init__.py	Thu Jul 29 18:26:56 2010
@@ -5,12 +5,11 @@
         import dbm
         d = dbm.open(file, 'w', 0o666)
 
-The returned object is a dbm.bsd, dbm.gnu, dbm.ndbm or dbm.dumb
-object, dependent on the type of database being opened (determined by
-the whichdb function) in the case of an existing dbm.  If the dbm does
-not exist and the create or new flag ('c' or 'n') was specified, the
-dbm type will be determined by the availability of the modules (tested
-in the above order).
+The returned object is a dbm.gnu, dbm.ndbm or dbm.dumb object, dependent on the
+type of database being opened (determined by the whichdb function) in the case
+of an existing dbm. If the dbm does not exist and the create or new flag ('c'
+or 'n') was specified, the dbm type will be determined by the availability of
+the modules (tested in the above order).
 
 It has the following interface (key and data are strings):
 
@@ -47,7 +46,7 @@
 class error(Exception):
     pass
 
-_names = ['dbm.bsd', 'dbm.gnu', 'dbm.ndbm', 'dbm.dumb']
+_names = ['dbm.gnu', 'dbm.ndbm', 'dbm.dumb']
 _defaultmod = None
 _modules = {}
 
@@ -168,10 +167,6 @@
     if magic == 0x13579ace:
         return "dbm.gnu"
 
-    ## Check for old Berkeley db hash file format v2
-    #if magic in (0x00061561, 0x61150600):
-    #    return "bsddb185" # not supported anymore
-
     # Later versions of Berkeley db hash file have a 12-byte pad in
     # front of the file type
     try:
@@ -179,10 +174,6 @@
     except struct.error:
         return ""
 
-    ## Check for BSD hash
-    #if magic in (0x00061561, 0x61150600):
-    #    return "dbm.bsd"
-
     # Unknown
     return ""
 

From python-checkins at python.org  Thu Jul 29 19:16:10 2010
From: python-checkins at python.org (georg.brandl)
Date: Thu, 29 Jul 2010 19:16:10 +0200 (CEST)
Subject: [Python-checkins] r83236 - in python/branches/py3k:
	Doc/library/string.rst Lib/string.py Misc/NEWS
Message-ID: <20100729171610.BE251F13F@mail.python.org>

Author: georg.brandl
Date: Thu Jul 29 19:16:10 2010
New Revision: 83236

Log:
#6630: allow customizing flags for compiling string.Template.idpattern.

Modified:
   python/branches/py3k/Doc/library/string.rst
   python/branches/py3k/Lib/string.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Doc/library/string.rst
==============================================================================
--- python/branches/py3k/Doc/library/string.rst	(original)
+++ python/branches/py3k/Doc/library/string.rst	Thu Jul 29 19:16:10 2010
@@ -710,6 +710,14 @@
   appropriate).  The default value is the regular expression
   ``[_a-z][_a-z0-9]*``.
 
+* *flags* -- The regular expression flags that will be applied when compiling
+  the regular expression used for recognizing substitutions.  The default value
+  is ``re.IGNORECASE``.  Note that ``re.VERBOSE`` will always be added to the
+  flags, so custom *idpattern*\ s must follow conventions for verbose regular
+  expressions.
+
+  .. versionadded:: 3.2
+
 Alternatively, you can provide the entire regular expression pattern by
 overriding the class attribute *pattern*.  If you do this, the value must be a
 regular expression object with four named capturing groups.  The capturing

Modified: python/branches/py3k/Lib/string.py
==============================================================================
--- python/branches/py3k/Lib/string.py	(original)
+++ python/branches/py3k/Lib/string.py	Thu Jul 29 19:16:10 2010
@@ -81,7 +81,7 @@
                 'delim' : _re.escape(cls.delimiter),
                 'id'    : cls.idpattern,
                 }
-        cls.pattern = _re.compile(pattern, _re.IGNORECASE | _re.VERBOSE)
+        cls.pattern = _re.compile(pattern, cls.flags | _re.VERBOSE)
 
 
 class Template(metaclass=_TemplateMetaclass):
@@ -89,6 +89,7 @@
 
     delimiter = '$'
     idpattern = r'[_a-z][_a-z0-9]*'
+    flags = _re.IGNORECASE
 
     def __init__(self, template):
         self.template = template

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Thu Jul 29 19:16:10 2010
@@ -475,6 +475,9 @@
 Library
 -------
 
+- Issue #6630: Allow customizing regex flags when subclassing the
+  string.Template class.
+
 - Issue #9411: Allow specifying an encoding for config files in the
   configparser module.
 

From python-checkins at python.org  Thu Jul 29 19:19:38 2010
From: python-checkins at python.org (victor.stinner)
Date: Thu, 29 Jul 2010 19:19:38 +0200 (CEST)
Subject: [Python-checkins] r83237 - python/branches/py3k/Lib/os.py
Message-ID: <20100729171938.6A8CAF54A@mail.python.org>

Author: victor.stinner
Date: Thu Jul 29 19:19:38 2010
New Revision: 83237

Log:
#8603: Add environb to os.__all__


Modified:
   python/branches/py3k/Lib/os.py

Modified: python/branches/py3k/Lib/os.py
==============================================================================
--- python/branches/py3k/Lib/os.py	(original)
+++ python/branches/py3k/Lib/os.py	Thu Jul 29 19:19:38 2010
@@ -533,7 +533,8 @@
         The optional second argument can specify an alternate default.
         key, default and the result are bytes."""
         return environb.get(key, default)
-    __all__.append("getenvb")
+
+    __all__.extend(("environb", "getenvb"))
 
 def fsencode(value):
     """Encode value for use in the file system, environment variables

From python-checkins at python.org  Thu Jul 29 19:55:01 2010
From: python-checkins at python.org (georg.brandl)
Date: Thu, 29 Jul 2010 19:55:01 +0200 (CEST)
Subject: [Python-checkins] r83238 - in python/branches/py3k:
	Lib/test/test_robotparser.py Lib/urllib/robotparser.py Misc/NEWS
Message-ID: <20100729175501.6A8A4D2A1@mail.python.org>

Author: georg.brandl
Date: Thu Jul 29 19:55:01 2010
New Revision: 83238

Log:
#4108: the first default entry (User-agent: *) wins.

Modified:
   python/branches/py3k/Lib/test/test_robotparser.py
   python/branches/py3k/Lib/urllib/robotparser.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/test/test_robotparser.py
==============================================================================
--- python/branches/py3k/Lib/test/test_robotparser.py	(original)
+++ python/branches/py3k/Lib/test/test_robotparser.py	Thu Jul 29 19:55:01 2010
@@ -216,6 +216,20 @@
 
 RobotTest(14, doc, good, bad)
 
+# 15. For issue #4108 (obey first * entry)
+doc = """
+User-agent: *
+Disallow: /some/path
+
+User-agent: *
+Disallow: /another/path
+"""
+
+good = ['/another/path']
+bad = ['/some/path']
+
+RobotTest(15, doc, good, bad)
+
 
 class NetworkTestCase(unittest.TestCase):
 

Modified: python/branches/py3k/Lib/urllib/robotparser.py
==============================================================================
--- python/branches/py3k/Lib/urllib/robotparser.py	(original)
+++ python/branches/py3k/Lib/urllib/robotparser.py	Thu Jul 29 19:55:01 2010
@@ -66,7 +66,9 @@
     def _add_entry(self, entry):
         if "*" in entry.useragents:
             # the default entry is considered last
-            self.default_entry = entry
+            if self.default_entry is None:
+                # the first default entry wins
+                self.default_entry = entry
         else:
             self.entries.append(entry)
 
@@ -118,7 +120,7 @@
                         entry.rulelines.append(RuleLine(line[1], True))
                         state = 2
         if state == 2:
-            self.entries.append(entry)
+            self._add_entry(entry)
 
 
     def can_fetch(self, useragent, url):

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Thu Jul 29 19:55:01 2010
@@ -475,6 +475,9 @@
 Library
 -------
 
+- Issue #4108: In urllib.robotparser, if there are multiple 'User-agent: *'
+  entries, consider the first one.
+
 - Issue #6630: Allow customizing regex flags when subclassing the
   string.Template class.
 

From python-checkins at python.org  Thu Jul 29 23:42:00 2010
From: python-checkins at python.org (mark.dickinson)
Date: Thu, 29 Jul 2010 23:42:00 +0200 (CEST)
Subject: [Python-checkins] r83239 - in python/branches/py3k: Misc/NEWS
	Modules/_struct.c
Message-ID: <20100729214200.11698EE982@mail.python.org>

Author: mark.dickinson
Date: Thu Jul 29 23:41:59 2010
New Revision: 83239

Log:
Issue #9422:  Fix memory leak when re-initializing a struct.Struct object.


Modified:
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/Modules/_struct.c

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Thu Jul 29 23:41:59 2010
@@ -1492,6 +1492,8 @@
 Extension Modules
 -----------------
 
+- Issue #9422: Fix memory leak when re-initializing a struct.Struct object.
+
 - Issue #7900: The getgroups(2) system call on MacOSX behaves rather oddly
   compared to other unix systems. In particular, os.getgroups() does
   not reflect any changes made using os.setgroups() but basicly always

Modified: python/branches/py3k/Modules/_struct.c
==============================================================================
--- python/branches/py3k/Modules/_struct.c	(original)
+++ python/branches/py3k/Modules/_struct.c	Thu Jul 29 23:41:59 2010
@@ -1247,6 +1247,9 @@
         PyErr_NoMemory();
         return -1;
     }
+    /* Free any s_codes value left over from a previous initialization. */
+    if (self->s_codes != NULL)
+        PyMem_FREE(self->s_codes);
     self->s_codes = codes;
 
     s = fmt;

From python-checkins at python.org  Thu Jul 29 23:43:24 2010
From: python-checkins at python.org (mark.dickinson)
Date: Thu, 29 Jul 2010 23:43:24 +0200 (CEST)
Subject: [Python-checkins] r83240 - in python/branches/release31-maint:
	Misc/NEWS Modules/_struct.c
Message-ID: <20100729214324.78FD4EE986@mail.python.org>

Author: mark.dickinson
Date: Thu Jul 29 23:43:24 2010
New Revision: 83240

Log:
Merged revisions 83239 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83239 | mark.dickinson | 2010-07-29 22:41:59 +0100 (Thu, 29 Jul 2010) | 2 lines
  
  Issue #9422:  Fix memory leak when re-initializing a struct.Struct object.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Misc/NEWS
   python/branches/release31-maint/Modules/_struct.c

Modified: python/branches/release31-maint/Misc/NEWS
==============================================================================
--- python/branches/release31-maint/Misc/NEWS	(original)
+++ python/branches/release31-maint/Misc/NEWS	Thu Jul 29 23:43:24 2010
@@ -323,6 +323,8 @@
 Extension Modules
 -----------------
 
+- Issue #9422: Fix memory leak when re-initializing a struct.Struct object.
+
 - Issue #7900: The getgroups(2) system call on MacOSX behaves rather oddly
   compared to other unix systems. In particular, os.getgroups() does
   not reflect any changes made using os.setgroups() but basicly always

Modified: python/branches/release31-maint/Modules/_struct.c
==============================================================================
--- python/branches/release31-maint/Modules/_struct.c	(original)
+++ python/branches/release31-maint/Modules/_struct.c	Thu Jul 29 23:43:24 2010
@@ -1233,6 +1233,9 @@
         PyErr_NoMemory();
         return -1;
     }
+    /* Free any s_codes value left over from a previous initialization. */
+    if (self->s_codes != NULL)
+        PyMem_FREE(self->s_codes);
     self->s_codes = codes;
 
     s = fmt;

From python-checkins at python.org  Thu Jul 29 23:44:47 2010
From: python-checkins at python.org (mark.dickinson)
Date: Thu, 29 Jul 2010 23:44:47 +0200 (CEST)
Subject: [Python-checkins] r83241 - in python/branches/release27-maint:
	Misc/NEWS Modules/_struct.c
Message-ID: <20100729214447.9105DEE986@mail.python.org>

Author: mark.dickinson
Date: Thu Jul 29 23:44:47 2010
New Revision: 83241

Log:
Merged revisions 83239 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83239 | mark.dickinson | 2010-07-29 22:41:59 +0100 (Thu, 29 Jul 2010) | 2 lines
  
  Issue #9422:  Fix memory leak when re-initializing a struct.Struct object.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Misc/NEWS
   python/branches/release27-maint/Modules/_struct.c

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Thu Jul 29 23:44:47 2010
@@ -64,6 +64,8 @@
 Extension Modules
 -----------------
 
+- Issue #9422: Fix memory leak when re-initializing a struct.Struct object.
+
 - Issue #7900: The getgroups(2) system call on MacOSX behaves rather oddly
   compared to other unix systems. In particular, os.getgroups() does
   not reflect any changes made using os.setgroups() but basicly always

Modified: python/branches/release27-maint/Modules/_struct.c
==============================================================================
--- python/branches/release27-maint/Modules/_struct.c	(original)
+++ python/branches/release27-maint/Modules/_struct.c	Thu Jul 29 23:44:47 2010
@@ -1289,6 +1289,9 @@
         PyErr_NoMemory();
         return -1;
     }
+    /* Free any s_codes value left over from a previous initialization. */
+    if (self->s_codes != NULL)
+        PyMem_FREE(self->s_codes);
     self->s_codes = codes;
 
     s = fmt;

From python-checkins at python.org  Thu Jul 29 23:47:28 2010
From: python-checkins at python.org (mark.dickinson)
Date: Thu, 29 Jul 2010 23:47:28 +0200 (CEST)
Subject: [Python-checkins] r83242 - in python/branches/release26-maint:
	Misc/NEWS Modules/_struct.c
Message-ID: <20100729214728.A3A3FEE986@mail.python.org>

Author: mark.dickinson
Date: Thu Jul 29 23:47:28 2010
New Revision: 83242

Log:
Merged revisions 83241 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/release27-maint

................
  r83241 | mark.dickinson | 2010-07-29 22:44:47 +0100 (Thu, 29 Jul 2010) | 9 lines
  
  Merged revisions 83239 via svnmerge from 
  svn+ssh://pythondev at svn.python.org/python/branches/py3k
  
  ........
    r83239 | mark.dickinson | 2010-07-29 22:41:59 +0100 (Thu, 29 Jul 2010) | 2 lines
    
    Issue #9422:  Fix memory leak when re-initializing a struct.Struct object.
  ........
................


Modified:
   python/branches/release26-maint/   (props changed)
   python/branches/release26-maint/Misc/NEWS
   python/branches/release26-maint/Modules/_struct.c

Modified: python/branches/release26-maint/Misc/NEWS
==============================================================================
--- python/branches/release26-maint/Misc/NEWS	(original)
+++ python/branches/release26-maint/Misc/NEWS	Thu Jul 29 23:47:28 2010
@@ -322,6 +322,8 @@
 Extension Modules
 -----------------
 
+- Issue #9422: Fix memory leak when re-initializing a struct.Struct object.
+
 - Issue #7900: The getgroups(2) system call on MacOSX behaves rather oddly
   compared to other unix systems. In particular, os.getgroups() does
   not reflect any changes made using os.setgroups() but basicly always

Modified: python/branches/release26-maint/Modules/_struct.c
==============================================================================
--- python/branches/release26-maint/Modules/_struct.c	(original)
+++ python/branches/release26-maint/Modules/_struct.c	Thu Jul 29 23:47:28 2010
@@ -1404,6 +1404,9 @@
         PyErr_NoMemory();
         return -1;
     }
+    /* Free any s_codes value left over from a previous initialization. */
+    if (self->s_codes != NULL)
+        PyMem_FREE(self->s_codes);
     self->s_codes = codes;
 
     s = fmt;

From python-checkins at python.org  Fri Jul 30 00:51:20 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 00:51:20 +0200 (CEST)
Subject: [Python-checkins] r83243 -
	python/branches/import_unicode/Python/errors.c
Message-ID: <20100729225120.137E0EE98A@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 00:51:19 2010
New Revision: 83243

Log:
PyErr_SyntaxLocation() uses PyUnicode_DecodeFSDefault()

to decode the filename

Modified:
   python/branches/import_unicode/Python/errors.c

Modified: python/branches/import_unicode/Python/errors.c
==============================================================================
--- python/branches/import_unicode/Python/errors.c	(original)
+++ python/branches/import_unicode/Python/errors.c	Fri Jul 30 00:51:19 2010
@@ -803,7 +803,7 @@
         Py_DECREF(tmp);
     }
     if (filename != NULL) {
-        tmp = PyUnicode_FromString(filename);
+        tmp = PyUnicode_DecodeFSDefault(filename);
         if (tmp == NULL)
             PyErr_Clear();
         else {

From python-checkins at python.org  Fri Jul 30 00:51:21 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 00:51:21 +0200 (CEST)
Subject: [Python-checkins] r83244 -
	python/branches/import_unicode/Python/compile.c
Message-ID: <20100729225121.9ECBBEE9A0@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 00:51:21 2010
New Revision: 83244

Log:
compiler_error() uses PyUnicode_DecodeFSDefault()

to decode the filename

Modified:
   python/branches/import_unicode/Python/compile.c

Modified: python/branches/import_unicode/Python/compile.c
==============================================================================
--- python/branches/import_unicode/Python/compile.c	(original)
+++ python/branches/import_unicode/Python/compile.c	Fri Jul 30 00:51:21 2010
@@ -3363,13 +3363,23 @@
 {
     PyObject *loc;
     PyObject *u = NULL, *v = NULL;
+    PyObject *filename;
 
     loc = PyErr_ProgramText(c->c_filename, c->u->u_lineno);
     if (!loc) {
         Py_INCREF(Py_None);
         loc = Py_None;
     }
-    u = Py_BuildValue("(ziOO)", c->c_filename, c->u->u_lineno,
+    if (c->c_filename != NULL) {
+        filename = PyUnicode_DecodeFSDefault(c->c_filename);
+        if (filename == NULL)
+            goto exit;
+    }
+    else {
+        Py_INCREF(Py_None);
+        filename = Py_None;
+    }
+    u = Py_BuildValue("(NiOO)", filename, c->u->u_lineno,
                       Py_None, loc);
     if (!u)
         goto exit;

From python-checkins at python.org  Fri Jul 30 00:57:24 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 00:57:24 +0200 (CEST)
Subject: [Python-checkins] r83245 -
	python/branches/import_unicode/Python/pythonrun.c
Message-ID: <20100729225724.1480DEE998@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 00:57:23 2010
New Revision: 83245

Log:
err_input() uses PyUnicode_DecodeFSDefault()

to decode the input filename

Modified:
   python/branches/import_unicode/Python/pythonrun.c

Modified: python/branches/import_unicode/Python/pythonrun.c
==============================================================================
--- python/branches/import_unicode/Python/pythonrun.c	(original)
+++ python/branches/import_unicode/Python/pythonrun.c	Fri Jul 30 00:57:23 2010
@@ -2088,6 +2088,8 @@
     PyObject *v, *w, *errtype, *errtext;
     PyObject *msg_obj = NULL;
     char *msg = NULL;
+    PyObject *filename;
+
     errtype = PyExc_SyntaxError;
     switch (err->error) {
     case E_ERROR:
@@ -2171,8 +2173,17 @@
         errtext = PyUnicode_DecodeUTF8(err->text, strlen(err->text),
                                        "replace");
     }
-    v = Py_BuildValue("(ziiN)", err->filename,
-                      err->lineno, err->offset, errtext);
+    if (err->filename != NULL)
+        filename = PyUnicode_DecodeFSDefault(err->filename);
+    else {
+        Py_INCREF(Py_None);
+        filename = Py_None;
+    }
+    if (filename != NULL)
+        v = Py_BuildValue("(NiiN)", filename,
+                          err->lineno, err->offset, errtext);
+    else
+        v = NULL;
     if (v != NULL) {
         if (msg_obj)
             w = Py_BuildValue("(OO)", msg_obj, v);

From python-checkins at python.org  Fri Jul 30 01:03:50 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 01:03:50 +0200 (CEST)
Subject: [Python-checkins] r83246 -
	python/branches/import_unicode/Python/ast.c
Message-ID: <20100729230350.A30E4F399@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 01:03:50 2010
New Revision: 83246

Log:
ast_error_finish() uses PyUnicode_DecodeFSDefault()

to decode the input filename

Modified:
   python/branches/import_unicode/Python/ast.c

Modified: python/branches/import_unicode/Python/ast.c
==============================================================================
--- python/branches/import_unicode/Python/ast.c	(original)
+++ python/branches/import_unicode/Python/ast.c	Fri Jul 30 01:03:50 2010
@@ -103,6 +103,7 @@
 {
     PyObject *type, *value, *tback, *errstr, *loc, *tmp;
     long lineno;
+    PyObject *filename_obj;
 
     assert(PyErr_Occurred());
     if (!PyErr_ExceptionMatches(PyExc_SyntaxError))
@@ -125,8 +126,19 @@
         Py_INCREF(Py_None);
         loc = Py_None;
     }
-    tmp = Py_BuildValue("(zlOO)", filename, lineno, Py_None, loc);
-    Py_DECREF(loc);
+    if (filename != NULL) {
+        filename_obj = PyUnicode_DecodeFSDefault(filename);
+        if (filename_obj == NULL) {
+            Py_DECREF(loc);
+            Py_DECREF(errstr);
+            return;
+        }
+    }
+    else {
+        Py_INCREF(Py_None);
+        filename_obj = Py_None;
+    }
+    tmp = Py_BuildValue("(NlON)", filename_obj, lineno, Py_None, loc);
     if (!tmp) {
         Py_DECREF(errstr);
         return;

From python-checkins at python.org  Fri Jul 30 01:19:40 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 01:19:40 +0200 (CEST)
Subject: [Python-checkins] r83247 - in python/branches/import_unicode:
	Doc/Makefile Doc/c-api/unicode.rst Doc/conf.py
	Doc/distutils/apiref.rst Doc/distutils/extending.rst
	Doc/documenting/markup.rst Doc/faq/extending.rst
	Doc/glossary.rst Doc/install/index.rst Doc/library/abc.rst
	Doc/library/argparse.rst Doc/library/ast.rst
	Doc/library/binascii.rst Doc/library/cmath.rst
	Doc/library/collections.rst Doc/library/configparser.rst
	Doc/library/contextlib.rst Doc/library/dbm.rst
	Doc/library/doctest.rst Doc/library/fnmatch.rst
	Doc/library/functions.rst Doc/library/functools.rst
	Doc/library/html.parser.rst Doc/library/http.cookiejar.rst
	Doc/library/importlib.rst Doc/library/io.rst
	Doc/library/itertools.rst Doc/library/locale.rst
	Doc/library/math.rst Doc/library/multiprocessing.rst
	Doc/library/os.path.rst Doc/library/os.rst
	Doc/library/pdb.rst Doc/library/pickletools.rst
	Doc/library/re.rst Doc/library/select.rst
	Doc/library/smtpd.rst Doc/library/socket.rst
	Doc/library/sqlite3.rst Doc/library/stdtypes.rst
	Doc/library/string.rst Doc/library/struct.rst
	Doc/library/subprocess.rst Doc/library/test.rst
	Doc/library/threading.rst Doc/library/tkinter.rst
	Doc/library/unittest.rst Doc/library/urllib.request.rst
	Doc/library/xml.dom.minidom.rst Doc/library/xml.dom.rst
	Doc/reference/expressions.rst Doc/tools/sphinxext/indexsidebar.html
	Doc/tools/sphinxext/pyspecific.py Doc/tutorial/classes.rst
	Doc/tutorial/floatingpoint.rst Doc/whatsnew/2.0.rst
	Doc/whatsnew/2.5.rst Doc/whatsnew/2.7.rst Grammar/Grammar
	Include/Python.h Lib/_abcoll.py Lib/_strptime.py Lib/ast.py
	Lib/asyncore.py Lib/base64.py Lib/configparser.py
	Lib/ctypes/__init__.py Lib/ctypes/test/test_arrays.py
	Lib/ctypes/test/test_bitfields.py
	Lib/ctypes/test/test_buffers.py Lib/ctypes/test/test_bytes.py
	Lib/ctypes/test/test_callbacks.py
	Lib/ctypes/test/test_cast.py Lib/ctypes/test/test_cfuncs.py
	Lib/ctypes/test/test_errno.py Lib/ctypes/test/test_internals.py
	Lib/ctypes/test/test_keeprefs.py Lib/ctypes/test/test_libc.py
	Lib/ctypes/test/test_objects.py Lib/ctypes/test/test_parameters.py
	Lib/ctypes/test/test_prototypes.py
	Lib/ctypes/test/test_python_api.py
	Lib/ctypes/test/test_random_things.py Lib/ctypes/test/test_repr.py
	Lib/ctypes/test/test_returnfuncptrs.py
	Lib/ctypes/test/test_stringptr.py Lib/ctypes/test/test_strings.py
	Lib/ctypes/test/test_structures.py
	Lib/ctypes/test/test_unicode.py Lib/datetime.py
	Lib/dbm/__init__.py Lib/distutils/archive_util.py
	Lib/distutils/bcppcompiler.py Lib/distutils/ccompiler.py
	Lib/distutils/cmd.py Lib/distutils/command/bdist.py
	Lib/distutils/command/bdist_dumb.py
	Lib/distutils/command/bdist_msi.py
	Lib/distutils/command/bdist_rpm.py
	Lib/distutils/command/bdist_wininst.py
	Lib/distutils/command/build.py Lib/distutils/command/build_clib.py
	Lib/distutils/command/build_ext.py Lib/distutils/command/build_py.py
	Lib/distutils/command/build_scripts.py
	Lib/distutils/command/config.py Lib/distutils/command/install.py
	Lib/distutils/command/register.py Lib/distutils/command/sdist.py
	Lib/distutils/command/upload.py Lib/distutils/config.py
	Lib/distutils/core.py Lib/distutils/cygwinccompiler.py
	Lib/distutils/dep_util.py Lib/distutils/dir_util.py
	Lib/distutils/dist.py Lib/distutils/emxccompiler.py
	Lib/distutils/errors.py Lib/distutils/extension.py
	Lib/distutils/fancy_getopt.py Lib/distutils/file_util.py
	Lib/distutils/filelist.py Lib/distutils/msvc9compiler.py
	Lib/distutils/msvccompiler.py Lib/distutils/sysconfig.py
	Lib/distutils/tests/support.py
	Lib/distutils/tests/test_archive_util.py
	Lib/distutils/tests/test_bdist.py
	Lib/distutils/tests/test_bdist_dumb.py
	Lib/distutils/tests/test_bdist_rpm.py
	Lib/distutils/tests/test_bdist_wininst.py
	Lib/distutils/tests/test_build_clib.py
	Lib/distutils/tests/test_build_ext.py
	Lib/distutils/tests/test_build_py.py
	Lib/distutils/tests/test_build_scripts.py
	Lib/distutils/tests/test_ccompiler.py
	Lib/distutils/tests/test_cmd.py
	Lib/distutils/tests/test_cygwinccompiler.py
	Lib/distutils/tests/test_dist.py
	Lib/distutils/tests/test_emxccompiler.py
	Lib/distutils/tests/test_extension.py
	Lib/distutils/tests/test_file_util.py
	Lib/distutils/tests/test_filelist.py
	Lib/distutils/tests/test_install.py
	Lib/distutils/tests/test_install_lib.py
	Lib/distutils/tests/test_register.py
	Lib/distutils/tests/test_sdist.py
	Lib/distutils/tests/test_sysconfig.py
	Lib/distutils/tests/test_unixccompiler.py
	Lib/distutils/tests/test_upload.py
	Lib/distutils/tests/test_util.py Lib/distutils/text_file.py
	Lib/distutils/unixccompiler.py Lib/distutils/util.py
	Lib/email/base64mime.py Lib/email/feedparser.py
	Lib/email/message.py Lib/email/test/test_email.py
	Lib/fnmatch.py Lib/getopt.py Lib/http/client.py
	Lib/http/cookiejar.py Lib/idlelib/PyShell.py
	Lib/importlib/test/benchmark.py Lib/json/encoder.py
	Lib/json/tests/test_fail.py Lib/ntpath.py Lib/os.py
	Lib/pdb.doc Lib/pdb.py Lib/pickle.py Lib/pickletools.py
	Lib/platform.py Lib/re.py Lib/smtpd.py Lib/sndhdr.py
	Lib/socket.py Lib/string.py Lib/subprocess.py
	Lib/sysconfig.py Lib/tarfile.py Lib/test/cfgparser.2
	Lib/test/cfgparser.3 Lib/test/datetimetester.py
	Lib/test/pickletester.py Lib/test/script_helper.py
	Lib/test/sndhdrdata Lib/test/support.py
	Lib/test/symlink_support.py Lib/test/test_array.py
	Lib/test/test_ast.py Lib/test/test_asyncore.py
	Lib/test/test_binascii.py Lib/test/test_binop.py
	Lib/test/test_calendar.py Lib/test/test_cfgparser.py
	Lib/test/test_cmath.py Lib/test/test_collections.py
	Lib/test/test_datetime.py Lib/test/test_fnmatch.py
	Lib/test/test_getopt.py Lib/test/test_http_cookiejar.py
	Lib/test/test_import.py Lib/test/test_math.py
	Lib/test/test_minidom.py Lib/test/test_os.py
	Lib/test/test_osx_env.py Lib/test/test_platform.py
	Lib/test/test_posix.py Lib/test/test_posixpath.py
	Lib/test/test_pow.py Lib/test/test_profilehooks.py
	Lib/test/test_queue.py Lib/test/test_re.py
	Lib/test/test_robotparser.py Lib/test/test_smtpd.py
	Lib/test/test_sndhdr.py Lib/test/test_struct.py
	Lib/test/test_structseq.py Lib/test/test_subprocess.py
	Lib/test/test_sys.py Lib/test/test_sys_setprofile.py
	Lib/test/test_sys_settrace.py Lib/test/test_sysconfig.py
	Lib/test/test_tarfile.py Lib/test/test_threaded_import.py
	Lib/test/test_threading.py Lib/test/test_trace.py
	Lib/test/test_unicode.py Lib/test/test_unpack.py
	Lib/test/test_urllib.py Lib/test/test_urllib2.py
	Lib/test/test_urlparse.py Lib/threading.py
	Lib/tkinter/__main__.py Lib/trace.py Lib/unittest/case.py
	Lib/unittest/test/test_case.py Lib/urllib/parse.py
	Lib/urllib/request.py Lib/urllib/robotparser.py
	Lib/xml/dom/expatbuilder.py Lib/xml/dom/minidom.py
	Lib/xmlrpc/client.py Makefile.pre.in Misc/ACKS Misc/NEWS
	Misc/developers.txt Misc/maintainers.rst Misc/python-wing.wpr
	Modules/Setup.dist Modules/_ctypes/_ctypes.c
	Modules/_ctypes/callproc.c Modules/_ctypes/cfield.c
	Modules/_datetimemodule.c Modules/_io/_iomodule.c
	Modules/_json.c Modules/_pickle.c Modules/_struct.c
	Modules/binascii.c Modules/cmathmodule.c
	Modules/datetimemodule.c Modules/getpath.c Modules/makesetup
	Modules/mathmodule.c Modules/posixmodule.c
	Modules/sha1module.c Modules/zipimport.c
	Objects/bytearrayobject.c Objects/memoryobject.c
	Objects/object.c Objects/stringlib/formatter.h
	Objects/unicodeobject.c PC/config.c PCbuild/pythoncore.vcproj
	Python/ceval.c Python/getargs.c Python/graminit.c
	Tools/gdb/libpython.py Tools/i18n/msgfmt.py configure
	configure.in setup.py
Message-ID: <20100729231940.B642AEE986@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 01:19:37 2010
New Revision: 83247

Log:
Merge import_unicode with py3k (r83246)


Added:
   python/branches/import_unicode/Lib/datetime.py
      - copied unchanged from r83246, /python/branches/py3k/Lib/datetime.py
   python/branches/import_unicode/Lib/test/cfgparser.2
      - copied unchanged from r83246, /python/branches/py3k/Lib/test/cfgparser.2
   python/branches/import_unicode/Lib/test/cfgparser.3
      - copied unchanged from r83246, /python/branches/py3k/Lib/test/cfgparser.3
   python/branches/import_unicode/Lib/test/datetimetester.py
      - copied unchanged from r83246, /python/branches/py3k/Lib/test/datetimetester.py
   python/branches/import_unicode/Lib/test/sndhdrdata/
      - copied from r83246, /python/branches/py3k/Lib/test/sndhdrdata/
   python/branches/import_unicode/Lib/test/test_smtpd.py
      - copied unchanged from r83246, /python/branches/py3k/Lib/test/test_smtpd.py
   python/branches/import_unicode/Lib/test/test_sndhdr.py
      - copied unchanged from r83246, /python/branches/py3k/Lib/test/test_sndhdr.py
   python/branches/import_unicode/Lib/test/test_sys_setprofile.py
      - copied unchanged from r83246, /python/branches/py3k/Lib/test/test_sys_setprofile.py
   python/branches/import_unicode/Lib/test/test_sys_settrace.py
      - copied unchanged from r83246, /python/branches/py3k/Lib/test/test_sys_settrace.py
   python/branches/import_unicode/Lib/tkinter/__main__.py
      - copied unchanged from r83246, /python/branches/py3k/Lib/tkinter/__main__.py
   python/branches/import_unicode/Modules/_datetimemodule.c
      - copied unchanged from r83246, /python/branches/py3k/Modules/_datetimemodule.c
Removed:
   python/branches/import_unicode/Lib/distutils/tests/test_ccompiler.py
   python/branches/import_unicode/Lib/distutils/tests/test_emxccompiler.py
   python/branches/import_unicode/Lib/pdb.doc
   python/branches/import_unicode/Lib/test/test_profilehooks.py
   python/branches/import_unicode/Lib/test/test_trace.py
   python/branches/import_unicode/Modules/datetimemodule.c
Modified:
   python/branches/import_unicode/Doc/Makefile
   python/branches/import_unicode/Doc/c-api/unicode.rst
   python/branches/import_unicode/Doc/conf.py
   python/branches/import_unicode/Doc/distutils/apiref.rst
   python/branches/import_unicode/Doc/distutils/extending.rst
   python/branches/import_unicode/Doc/documenting/markup.rst
   python/branches/import_unicode/Doc/faq/extending.rst
   python/branches/import_unicode/Doc/glossary.rst
   python/branches/import_unicode/Doc/install/index.rst
   python/branches/import_unicode/Doc/library/abc.rst
   python/branches/import_unicode/Doc/library/argparse.rst
   python/branches/import_unicode/Doc/library/ast.rst
   python/branches/import_unicode/Doc/library/binascii.rst
   python/branches/import_unicode/Doc/library/cmath.rst
   python/branches/import_unicode/Doc/library/collections.rst
   python/branches/import_unicode/Doc/library/configparser.rst
   python/branches/import_unicode/Doc/library/contextlib.rst
   python/branches/import_unicode/Doc/library/dbm.rst
   python/branches/import_unicode/Doc/library/doctest.rst
   python/branches/import_unicode/Doc/library/fnmatch.rst
   python/branches/import_unicode/Doc/library/functions.rst
   python/branches/import_unicode/Doc/library/functools.rst
   python/branches/import_unicode/Doc/library/html.parser.rst
   python/branches/import_unicode/Doc/library/http.cookiejar.rst
   python/branches/import_unicode/Doc/library/importlib.rst
   python/branches/import_unicode/Doc/library/io.rst
   python/branches/import_unicode/Doc/library/itertools.rst
   python/branches/import_unicode/Doc/library/locale.rst
   python/branches/import_unicode/Doc/library/math.rst
   python/branches/import_unicode/Doc/library/multiprocessing.rst
   python/branches/import_unicode/Doc/library/os.path.rst
   python/branches/import_unicode/Doc/library/os.rst
   python/branches/import_unicode/Doc/library/pdb.rst
   python/branches/import_unicode/Doc/library/pickletools.rst
   python/branches/import_unicode/Doc/library/re.rst
   python/branches/import_unicode/Doc/library/select.rst
   python/branches/import_unicode/Doc/library/smtpd.rst
   python/branches/import_unicode/Doc/library/socket.rst
   python/branches/import_unicode/Doc/library/sqlite3.rst
   python/branches/import_unicode/Doc/library/stdtypes.rst
   python/branches/import_unicode/Doc/library/string.rst
   python/branches/import_unicode/Doc/library/struct.rst
   python/branches/import_unicode/Doc/library/subprocess.rst
   python/branches/import_unicode/Doc/library/test.rst
   python/branches/import_unicode/Doc/library/threading.rst
   python/branches/import_unicode/Doc/library/tkinter.rst
   python/branches/import_unicode/Doc/library/unittest.rst
   python/branches/import_unicode/Doc/library/urllib.request.rst
   python/branches/import_unicode/Doc/library/xml.dom.minidom.rst
   python/branches/import_unicode/Doc/library/xml.dom.rst
   python/branches/import_unicode/Doc/reference/expressions.rst
   python/branches/import_unicode/Doc/tools/sphinxext/indexsidebar.html
   python/branches/import_unicode/Doc/tools/sphinxext/pyspecific.py
   python/branches/import_unicode/Doc/tutorial/classes.rst
   python/branches/import_unicode/Doc/tutorial/floatingpoint.rst
   python/branches/import_unicode/Doc/whatsnew/2.0.rst
   python/branches/import_unicode/Doc/whatsnew/2.5.rst
   python/branches/import_unicode/Doc/whatsnew/2.7.rst
   python/branches/import_unicode/Grammar/Grammar
   python/branches/import_unicode/Include/Python.h
   python/branches/import_unicode/Lib/_abcoll.py
   python/branches/import_unicode/Lib/_strptime.py
   python/branches/import_unicode/Lib/ast.py
   python/branches/import_unicode/Lib/asyncore.py
   python/branches/import_unicode/Lib/base64.py
   python/branches/import_unicode/Lib/configparser.py
   python/branches/import_unicode/Lib/ctypes/__init__.py
   python/branches/import_unicode/Lib/ctypes/test/test_arrays.py
   python/branches/import_unicode/Lib/ctypes/test/test_bitfields.py
   python/branches/import_unicode/Lib/ctypes/test/test_buffers.py
   python/branches/import_unicode/Lib/ctypes/test/test_bytes.py
   python/branches/import_unicode/Lib/ctypes/test/test_callbacks.py
   python/branches/import_unicode/Lib/ctypes/test/test_cast.py
   python/branches/import_unicode/Lib/ctypes/test/test_cfuncs.py
   python/branches/import_unicode/Lib/ctypes/test/test_errno.py
   python/branches/import_unicode/Lib/ctypes/test/test_internals.py
   python/branches/import_unicode/Lib/ctypes/test/test_keeprefs.py
   python/branches/import_unicode/Lib/ctypes/test/test_libc.py
   python/branches/import_unicode/Lib/ctypes/test/test_objects.py
   python/branches/import_unicode/Lib/ctypes/test/test_parameters.py
   python/branches/import_unicode/Lib/ctypes/test/test_prototypes.py
   python/branches/import_unicode/Lib/ctypes/test/test_python_api.py
   python/branches/import_unicode/Lib/ctypes/test/test_random_things.py
   python/branches/import_unicode/Lib/ctypes/test/test_repr.py
   python/branches/import_unicode/Lib/ctypes/test/test_returnfuncptrs.py
   python/branches/import_unicode/Lib/ctypes/test/test_stringptr.py
   python/branches/import_unicode/Lib/ctypes/test/test_strings.py
   python/branches/import_unicode/Lib/ctypes/test/test_structures.py
   python/branches/import_unicode/Lib/ctypes/test/test_unicode.py
   python/branches/import_unicode/Lib/dbm/__init__.py
   python/branches/import_unicode/Lib/distutils/archive_util.py
   python/branches/import_unicode/Lib/distutils/bcppcompiler.py
   python/branches/import_unicode/Lib/distutils/ccompiler.py
   python/branches/import_unicode/Lib/distutils/cmd.py
   python/branches/import_unicode/Lib/distutils/command/bdist.py
   python/branches/import_unicode/Lib/distutils/command/bdist_dumb.py
   python/branches/import_unicode/Lib/distutils/command/bdist_msi.py
   python/branches/import_unicode/Lib/distutils/command/bdist_rpm.py
   python/branches/import_unicode/Lib/distutils/command/bdist_wininst.py
   python/branches/import_unicode/Lib/distutils/command/build.py
   python/branches/import_unicode/Lib/distutils/command/build_clib.py
   python/branches/import_unicode/Lib/distutils/command/build_ext.py
   python/branches/import_unicode/Lib/distutils/command/build_py.py
   python/branches/import_unicode/Lib/distutils/command/build_scripts.py
   python/branches/import_unicode/Lib/distutils/command/config.py
   python/branches/import_unicode/Lib/distutils/command/install.py
   python/branches/import_unicode/Lib/distutils/command/register.py
   python/branches/import_unicode/Lib/distutils/command/sdist.py
   python/branches/import_unicode/Lib/distutils/command/upload.py
   python/branches/import_unicode/Lib/distutils/config.py
   python/branches/import_unicode/Lib/distutils/core.py
   python/branches/import_unicode/Lib/distutils/cygwinccompiler.py
   python/branches/import_unicode/Lib/distutils/dep_util.py
   python/branches/import_unicode/Lib/distutils/dir_util.py
   python/branches/import_unicode/Lib/distutils/dist.py
   python/branches/import_unicode/Lib/distutils/emxccompiler.py
   python/branches/import_unicode/Lib/distutils/errors.py
   python/branches/import_unicode/Lib/distutils/extension.py
   python/branches/import_unicode/Lib/distutils/fancy_getopt.py
   python/branches/import_unicode/Lib/distutils/file_util.py
   python/branches/import_unicode/Lib/distutils/filelist.py
   python/branches/import_unicode/Lib/distutils/msvc9compiler.py
   python/branches/import_unicode/Lib/distutils/msvccompiler.py
   python/branches/import_unicode/Lib/distutils/sysconfig.py
   python/branches/import_unicode/Lib/distutils/tests/support.py
   python/branches/import_unicode/Lib/distutils/tests/test_archive_util.py
   python/branches/import_unicode/Lib/distutils/tests/test_bdist.py
   python/branches/import_unicode/Lib/distutils/tests/test_bdist_dumb.py
   python/branches/import_unicode/Lib/distutils/tests/test_bdist_rpm.py
   python/branches/import_unicode/Lib/distutils/tests/test_bdist_wininst.py
   python/branches/import_unicode/Lib/distutils/tests/test_build_clib.py
   python/branches/import_unicode/Lib/distutils/tests/test_build_ext.py
   python/branches/import_unicode/Lib/distutils/tests/test_build_py.py
   python/branches/import_unicode/Lib/distutils/tests/test_build_scripts.py
   python/branches/import_unicode/Lib/distutils/tests/test_cmd.py
   python/branches/import_unicode/Lib/distutils/tests/test_cygwinccompiler.py
   python/branches/import_unicode/Lib/distutils/tests/test_dist.py
   python/branches/import_unicode/Lib/distutils/tests/test_extension.py
   python/branches/import_unicode/Lib/distutils/tests/test_file_util.py
   python/branches/import_unicode/Lib/distutils/tests/test_filelist.py
   python/branches/import_unicode/Lib/distutils/tests/test_install.py
   python/branches/import_unicode/Lib/distutils/tests/test_install_lib.py
   python/branches/import_unicode/Lib/distutils/tests/test_register.py
   python/branches/import_unicode/Lib/distutils/tests/test_sdist.py
   python/branches/import_unicode/Lib/distutils/tests/test_sysconfig.py
   python/branches/import_unicode/Lib/distutils/tests/test_unixccompiler.py
   python/branches/import_unicode/Lib/distutils/tests/test_upload.py
   python/branches/import_unicode/Lib/distutils/tests/test_util.py
   python/branches/import_unicode/Lib/distutils/text_file.py
   python/branches/import_unicode/Lib/distutils/unixccompiler.py
   python/branches/import_unicode/Lib/distutils/util.py
   python/branches/import_unicode/Lib/email/base64mime.py
   python/branches/import_unicode/Lib/email/feedparser.py
   python/branches/import_unicode/Lib/email/message.py
   python/branches/import_unicode/Lib/email/test/test_email.py
   python/branches/import_unicode/Lib/fnmatch.py
   python/branches/import_unicode/Lib/getopt.py
   python/branches/import_unicode/Lib/http/client.py
   python/branches/import_unicode/Lib/http/cookiejar.py
   python/branches/import_unicode/Lib/idlelib/PyShell.py
   python/branches/import_unicode/Lib/importlib/test/benchmark.py
   python/branches/import_unicode/Lib/json/encoder.py
   python/branches/import_unicode/Lib/json/tests/test_fail.py
   python/branches/import_unicode/Lib/ntpath.py
   python/branches/import_unicode/Lib/os.py
   python/branches/import_unicode/Lib/pdb.py
   python/branches/import_unicode/Lib/pickle.py
   python/branches/import_unicode/Lib/pickletools.py
   python/branches/import_unicode/Lib/platform.py
   python/branches/import_unicode/Lib/re.py
   python/branches/import_unicode/Lib/smtpd.py
   python/branches/import_unicode/Lib/sndhdr.py
   python/branches/import_unicode/Lib/socket.py
   python/branches/import_unicode/Lib/string.py
   python/branches/import_unicode/Lib/subprocess.py
   python/branches/import_unicode/Lib/sysconfig.py
   python/branches/import_unicode/Lib/tarfile.py
   python/branches/import_unicode/Lib/test/pickletester.py
   python/branches/import_unicode/Lib/test/script_helper.py
   python/branches/import_unicode/Lib/test/support.py
   python/branches/import_unicode/Lib/test/symlink_support.py
   python/branches/import_unicode/Lib/test/test_array.py
   python/branches/import_unicode/Lib/test/test_ast.py
   python/branches/import_unicode/Lib/test/test_asyncore.py
   python/branches/import_unicode/Lib/test/test_binascii.py
   python/branches/import_unicode/Lib/test/test_binop.py
   python/branches/import_unicode/Lib/test/test_calendar.py
   python/branches/import_unicode/Lib/test/test_cfgparser.py
   python/branches/import_unicode/Lib/test/test_cmath.py
   python/branches/import_unicode/Lib/test/test_collections.py
   python/branches/import_unicode/Lib/test/test_datetime.py
   python/branches/import_unicode/Lib/test/test_fnmatch.py
   python/branches/import_unicode/Lib/test/test_getopt.py
   python/branches/import_unicode/Lib/test/test_http_cookiejar.py
   python/branches/import_unicode/Lib/test/test_import.py
   python/branches/import_unicode/Lib/test/test_math.py
   python/branches/import_unicode/Lib/test/test_minidom.py
   python/branches/import_unicode/Lib/test/test_os.py
   python/branches/import_unicode/Lib/test/test_osx_env.py
   python/branches/import_unicode/Lib/test/test_platform.py
   python/branches/import_unicode/Lib/test/test_posix.py
   python/branches/import_unicode/Lib/test/test_posixpath.py
   python/branches/import_unicode/Lib/test/test_pow.py
   python/branches/import_unicode/Lib/test/test_queue.py
   python/branches/import_unicode/Lib/test/test_re.py
   python/branches/import_unicode/Lib/test/test_robotparser.py
   python/branches/import_unicode/Lib/test/test_struct.py
   python/branches/import_unicode/Lib/test/test_structseq.py
   python/branches/import_unicode/Lib/test/test_subprocess.py
   python/branches/import_unicode/Lib/test/test_sys.py
   python/branches/import_unicode/Lib/test/test_sysconfig.py
   python/branches/import_unicode/Lib/test/test_tarfile.py
   python/branches/import_unicode/Lib/test/test_threaded_import.py
   python/branches/import_unicode/Lib/test/test_threading.py
   python/branches/import_unicode/Lib/test/test_unicode.py
   python/branches/import_unicode/Lib/test/test_unpack.py
   python/branches/import_unicode/Lib/test/test_urllib.py
   python/branches/import_unicode/Lib/test/test_urllib2.py
   python/branches/import_unicode/Lib/test/test_urlparse.py
   python/branches/import_unicode/Lib/threading.py
   python/branches/import_unicode/Lib/trace.py
   python/branches/import_unicode/Lib/unittest/case.py
   python/branches/import_unicode/Lib/unittest/test/test_case.py
   python/branches/import_unicode/Lib/urllib/parse.py
   python/branches/import_unicode/Lib/urllib/request.py
   python/branches/import_unicode/Lib/urllib/robotparser.py
   python/branches/import_unicode/Lib/xml/dom/expatbuilder.py
   python/branches/import_unicode/Lib/xml/dom/minidom.py
   python/branches/import_unicode/Lib/xmlrpc/client.py
   python/branches/import_unicode/Makefile.pre.in
   python/branches/import_unicode/Misc/ACKS
   python/branches/import_unicode/Misc/NEWS
   python/branches/import_unicode/Misc/developers.txt
   python/branches/import_unicode/Misc/maintainers.rst
   python/branches/import_unicode/Misc/python-wing.wpr
   python/branches/import_unicode/Modules/Setup.dist
   python/branches/import_unicode/Modules/_ctypes/_ctypes.c
   python/branches/import_unicode/Modules/_ctypes/callproc.c
   python/branches/import_unicode/Modules/_ctypes/cfield.c
   python/branches/import_unicode/Modules/_io/_iomodule.c
   python/branches/import_unicode/Modules/_json.c
   python/branches/import_unicode/Modules/_pickle.c
   python/branches/import_unicode/Modules/_struct.c
   python/branches/import_unicode/Modules/binascii.c
   python/branches/import_unicode/Modules/cmathmodule.c
   python/branches/import_unicode/Modules/getpath.c
   python/branches/import_unicode/Modules/makesetup
   python/branches/import_unicode/Modules/mathmodule.c
   python/branches/import_unicode/Modules/posixmodule.c
   python/branches/import_unicode/Modules/sha1module.c
   python/branches/import_unicode/Modules/zipimport.c
   python/branches/import_unicode/Objects/bytearrayobject.c
   python/branches/import_unicode/Objects/memoryobject.c
   python/branches/import_unicode/Objects/object.c
   python/branches/import_unicode/Objects/stringlib/formatter.h
   python/branches/import_unicode/Objects/unicodeobject.c
   python/branches/import_unicode/PC/config.c
   python/branches/import_unicode/PCbuild/pythoncore.vcproj
   python/branches/import_unicode/Python/ceval.c
   python/branches/import_unicode/Python/getargs.c
   python/branches/import_unicode/Python/graminit.c
   python/branches/import_unicode/Tools/gdb/libpython.py
   python/branches/import_unicode/Tools/i18n/msgfmt.py
   python/branches/import_unicode/configure
   python/branches/import_unicode/configure.in
   python/branches/import_unicode/setup.py

Modified: python/branches/import_unicode/Doc/Makefile
==============================================================================
--- python/branches/import_unicode/Doc/Makefile	(original)
+++ python/branches/import_unicode/Doc/Makefile	Fri Jul 30 01:19:37 2010
@@ -40,7 +40,7 @@
 checkout:
 	@if [ ! -d tools/sphinx ]; then \
 	  echo "Checking out Sphinx..."; \
-	  svn checkout $(SVNROOT)/external/Sphinx-0.6.5/sphinx tools/sphinx; \
+	  svn checkout $(SVNROOT)/external/Sphinx-1.0.1/sphinx tools/sphinx; \
 	fi
 	@if [ ! -d tools/docutils ]; then \
 	  echo "Checking out Docutils..."; \
@@ -144,6 +144,7 @@
 	# archive the A4 latex
 	-rm -r build/latex
 	make latex PAPER=a4
+	-sed -i 's/makeindex/makeindex -q/' build/latex/Makefile
 	(cd build/latex; make clean && make all-pdf && make FMT=pdf zip bz2)
 	cp build/latex/docs-pdf.zip dist/python-$(DISTVERSION)-docs-pdf-a4.zip
 	cp build/latex/docs-pdf.tar.bz2 dist/python-$(DISTVERSION)-docs-pdf-a4.tar.bz2
@@ -151,6 +152,7 @@
 	# archive the letter latex
 	rm -r build/latex
 	make latex PAPER=letter
+	-sed -i 's/makeindex/makeindex -q/' build/latex/Makefile
 	(cd build/latex; make clean && make all-pdf && make FMT=pdf zip bz2)
 	cp build/latex/docs-pdf.zip dist/python-$(DISTVERSION)-docs-pdf-letter.zip
 	cp build/latex/docs-pdf.tar.bz2 dist/python-$(DISTVERSION)-docs-pdf-letter.tar.bz2

Modified: python/branches/import_unicode/Doc/c-api/unicode.rst
==============================================================================
--- python/branches/import_unicode/Doc/c-api/unicode.rst	(original)
+++ python/branches/import_unicode/Doc/c-api/unicode.rst	Fri Jul 30 01:19:37 2010
@@ -376,7 +376,7 @@
 :cdata:`Py_FileSystemEncoding` should be used as the encoding, and
 ``"surrogateescape"`` should be used as the error handler (:pep:`383`). To
 encode file names during argument parsing, the ``"O&"`` converter should be
-used, passsing :func:PyUnicode_FSConverter as the conversion function:
+used, passsing :func:`PyUnicode_FSConverter` as the conversion function:
 
 .. cfunction:: int PyUnicode_FSConverter(PyObject* obj, void* result)
 

Modified: python/branches/import_unicode/Doc/conf.py
==============================================================================
--- python/branches/import_unicode/Doc/conf.py	(original)
+++ python/branches/import_unicode/Doc/conf.py	Fri Jul 30 01:19:37 2010
@@ -152,7 +152,7 @@
 latex_appendices = ['glossary', 'about', 'license', 'copyright']
 
 # Get LaTeX to handle Unicode correctly
-latex_elements = {'inputenc': r'\usepackage[utf8x]{inputenc}'}
+latex_elements = {'inputenc': r'\usepackage[utf8x]{inputenc}', 'utf8extra': ''}
 
 # Options for the coverage checker
 # --------------------------------

Modified: python/branches/import_unicode/Doc/distutils/apiref.rst
==============================================================================
--- python/branches/import_unicode/Doc/distutils/apiref.rst	(original)
+++ python/branches/import_unicode/Doc/distutils/apiref.rst	Fri Jul 30 01:19:37 2010
@@ -21,7 +21,9 @@
 .. function:: setup(arguments)
 
    The basic do-everything function that does most everything you could ever ask
-   for from a Distutils method. See XXXXX
+   for from a Distutils method.
+
+   .. See XXXXX
 
    The setup function takes a large number of arguments. These are laid out in the
    following table.
@@ -147,11 +149,11 @@
 In addition, the :mod:`distutils.core` module exposed a number of  classes that
 live elsewhere.
 
-* :class:`Extension` from :mod:`distutils.extension`
+* :class:`~distutils.extension.Extension` from :mod:`distutils.extension`
 
-* :class:`Command` from :mod:`distutils.cmd`
+* :class:`~distutils.cmd.Command` from :mod:`distutils.cmd`
 
-* :class:`Distribution` from :mod:`distutils.dist`
+* :class:`~distutils.dist.Distribution` from :mod:`distutils.dist`
 
 A short description of each of these follows, but see the relevant module for
 the full reference.
@@ -1679,8 +1681,8 @@
 ===================================================================
 
 .. module:: distutils.cmd
-   :synopsis: This module provides the abstract base class Command. This class is subclassed
-              by the modules in the distutils.command  subpackage.
+   :synopsis: This module provides the abstract base class Command. This class
+              is subclassed by the modules in the distutils.command subpackage.
 
 
 This module supplies the abstract base class :class:`Command`.
@@ -1690,20 +1692,84 @@
 
    Abstract base class for defining command classes, the "worker bees" of the
    Distutils.  A useful analogy for command classes is to think of them as
-   subroutines with local variables called *options*.  The options are declared in
-   :meth:`initialize_options` and defined (given their final values) in
-   :meth:`finalize_options`, both of which must be defined by every command class.
-   The distinction between the two is necessary because option values might come
-   from the outside world (command line, config file, ...), and any options
-   dependent on other options must be computed after these outside influences have
-   been processed --- hence :meth:`finalize_options`.  The body of the subroutine,
-   where it does all its work based on the values of its options, is the
-   :meth:`run` method, which must also be implemented by every command class.
+   subroutines with local variables called *options*.  The options are declared
+   in :meth:`initialize_options` and defined (given their final values) in
+   :meth:`finalize_options`, both of which must be defined by every command
+   class.  The distinction between the two is necessary because option values
+   might come from the outside world (command line, config file, ...), and any
+   options dependent on other options must be computed after these outside
+   influences have been processed --- hence :meth:`finalize_options`.  The body
+   of the subroutine, where it does all its work based on the values of its
+   options, is the :meth:`run` method, which must also be implemented by every
+   command class.
 
-   The class constructor takes a single argument *dist*, a  :class:`Distribution`
+   The class constructor takes a single argument *dist*, a :class:`Distribution`
    instance.
 
 
+Creating a new Distutils command
+================================
+
+This section outlines the steps to create a new Distutils command.
+
+A new command lives in a module in the :mod:`distutils.command` package. There
+is a sample template in that directory called :file:`command_template`.  Copy
+this file to a new module with the same name as the new command you're
+implementing.  This module should implement a class with the same name as the
+module (and the command).  So, for instance, to create the command
+``peel_banana`` (so that users can run ``setup.py peel_banana``), you'd copy
+:file:`command_template` to :file:`distutils/command/peel_banana.py`, then edit
+it so that it's implementing the class :class:`peel_banana`, a subclass of
+:class:`distutils.cmd.Command`.
+
+Subclasses of :class:`Command` must define the following methods.
+
+.. method:: Command.initialize_options()
+
+   Set default values for all the options that this command supports.  Note that
+   these defaults may be overridden by other commands, by the setup script, by
+   config files, or by the command-line.  Thus, this is not the place to code
+   dependencies between options; generally, :meth:`initialize_options`
+   implementations are just a bunch of ``self.foo = None`` assignments.
+
+
+.. method:: Command.finalize_options()
+
+   Set final values for all the options that this command supports. This is
+   always called as late as possible, ie.  after any option assignments from the
+   command-line or from other commands have been done.  Thus, this is the place
+   to to code option dependencies: if *foo* depends on *bar*, then it is safe to
+   set *foo* from *bar* as long as *foo* still has the same value it was
+   assigned in :meth:`initialize_options`.
+
+
+.. method:: Command.run()
+
+   A command's raison d'etre: carry out the action it exists to perform,
+   controlled by the options initialized in :meth:`initialize_options`,
+   customized by other commands, the setup script, the command-line, and config
+   files, and finalized in :meth:`finalize_options`.  All terminal output and
+   filesystem interaction should be done by :meth:`run`.
+
+
+.. attribute:: Command.sub_commands
+
+   *sub_commands* formalizes the notion of a "family" of commands,
+   e.g. ``install`` as the parent with sub-commands ``install_lib``,
+   ``install_headers``, etc.  The parent of a family of commands defines
+   *sub_commands* as a class attribute; it's a list of 2-tuples ``(command_name,
+   predicate)``, with *command_name* a string and *predicate* a function, a
+   string or ``None``.  *predicate* is a method of the parent command that
+   determines whether the corresponding command is applicable in the current
+   situation.  (E.g. we ``install_headers`` is only applicable if we have any C
+   header files to install.)  If *predicate* is ``None``, that command is always
+   applicable.
+
+   *sub_commands* is usually defined at the *end* of a class, because
+   predicates can be methods of the class, so they must already have been
+   defined.  The canonical example is the :command:`install` command.
+
+
 :mod:`distutils.command` --- Individual Distutils commands
 ==========================================================
 
@@ -1942,63 +2008,3 @@
 This is described in more detail in :pep:`301`.
 
 .. % todo
-
-
-Creating a new Distutils command
-================================
-
-This section outlines the steps to create a new Distutils command.
-
-A new command lives in a module in the :mod:`distutils.command` package. There
-is a sample template in that directory called  :file:`command_template`. Copy
-this file to a new module with the same name as the new command you're
-implementing. This module should implement a class with the same name as the
-module (and the command). So, for instance, to create the command
-``peel_banana`` (so that users can run ``setup.py peel_banana``), you'd copy
-:file:`command_template`  to :file:`distutils/command/peel_banana.py`, then edit
-it so that it's implementing the class :class:`peel_banana`, a subclass of
-:class:`distutils.cmd.Command`.
-
-Subclasses of :class:`Command` must define the following methods.
-
-
-.. method:: Command.initialize_options()
-
-   Set default values for all the options that this command supports.  Note that
-   these defaults may be overridden by other commands, by the setup script, by
-   config files, or by the command-line.  Thus, this is not the place to code
-   dependencies between options; generally, :meth:`initialize_options`
-   implementations are just a bunch of ``self.foo = None`` assignments.
-
-
-.. method:: Command.finalize_options()
-
-   Set final values for all the options that this command supports. This is
-   always called as late as possible, ie.  after any option assignments from the
-   command-line or from other commands have been done.  Thus, this is the place
-   to to code option dependencies: if *foo* depends on *bar*, then it is safe to
-   set *foo* from *bar* as long as *foo* still has the same value it was
-   assigned in :meth:`initialize_options`.
-
-
-.. method:: Command.run()
-
-   A command's raison d'etre: carry out the action it exists to perform, controlled
-   by the options initialized in :meth:`initialize_options`, customized by other
-   commands, the setup script, the command-line, and config files, and finalized in
-   :meth:`finalize_options`.  All terminal output and filesystem interaction should
-   be done by :meth:`run`.
-
-*sub_commands* formalizes the notion of a "family" of commands, eg. ``install``
-as the parent with sub-commands ``install_lib``, ``install_headers``, etc.  The
-parent of a family of commands defines *sub_commands* as a class attribute; it's
-a list of 2-tuples ``(command_name, predicate)``, with *command_name* a string
-and *predicate* a function, a string or None. *predicate* is a method of
-the parent command that determines whether the corresponding command is
-applicable in the current situation.  (Eg. we ``install_headers`` is only
-applicable if we have any C header files to install.)  If *predicate* is None,
-that command is always applicable.
-
-*sub_commands* is usually defined at the \*end\* of a class, because predicates
-can be methods of the class, so they must already have been defined.  The
-canonical example is the :command:`install` command.

Modified: python/branches/import_unicode/Doc/distutils/extending.rst
==============================================================================
--- python/branches/import_unicode/Doc/distutils/extending.rst	(original)
+++ python/branches/import_unicode/Doc/distutils/extending.rst	Fri Jul 30 01:19:37 2010
@@ -15,8 +15,8 @@
 should be copied into packages in addition to :file:`.py` files as a
 convenience.
 
-Most distutils command implementations are subclasses of the :class:`Command`
-class from :mod:`distutils.cmd`.  New commands may directly inherit from
+Most distutils command implementations are subclasses of the
+:class:`distutils.cmd.Command` class.  New commands may directly inherit from
 :class:`Command`, while replacements often derive from :class:`Command`
 indirectly, directly subclassing the command they are replacing.  Commands are
 required to derive from :class:`Command`.

Modified: python/branches/import_unicode/Doc/documenting/markup.rst
==============================================================================
--- python/branches/import_unicode/Doc/documenting/markup.rst	(original)
+++ python/branches/import_unicode/Doc/documenting/markup.rst	Fri Jul 30 01:19:37 2010
@@ -177,6 +177,34 @@
    are modified), side effects, and possible exceptions.  A small example may be
    provided.
 
+.. describe:: decorator
+
+   Describes a decorator function.  The signature should *not* represent the
+   signature of the actual function, but the usage as a decorator.  For example,
+   given the functions
+
+   .. code-block:: python
+
+      def removename(func):
+          func.__name__ = ''
+          return func
+
+      def setnewname(name):
+          def decorator(func):
+              func.__name__ = name
+              return func
+          return decorator
+
+   the descriptions should look like this::
+
+      .. decorator:: removename
+
+         Remove name of the decorated function.
+
+      .. decorator:: setnewname(name)
+
+         Set name of the decorated function to *name*.
+
 .. describe:: class
 
    Describes a class.  The signature can include parentheses with parameters
@@ -194,14 +222,18 @@
    parameter.  The description should include similar information to that
    described for ``function``.
 
+.. describe:: decoratormethod
+
+   Same as ``decorator``, but for decorators that are methods.
+
 .. describe:: opcode
 
    Describes a Python :term:`bytecode` instruction.
 
 .. describe:: cmdoption
 
-   Describes a command line option or switch.  Option argument names should be
-   enclosed in angle brackets.  Example::
+   Describes a Python command line option or switch.  Option argument names
+   should be enclosed in angle brackets.  Example::
 
       .. cmdoption:: -m <module>
 
@@ -502,8 +534,9 @@
 
 .. describe:: option
 
-   A command-line option to an executable program.  The leading hyphen(s) must
-   be included.
+   A command-line option of Python.  The leading hyphen(s) must be included.
+   If a matching ``cmdoption`` directive exists, it is linked to.  For options
+   of other programs or scripts, use simple ````code```` markup.
 
 .. describe:: program
 

Modified: python/branches/import_unicode/Doc/faq/extending.rst
==============================================================================
--- python/branches/import_unicode/Doc/faq/extending.rst	(original)
+++ python/branches/import_unicode/Doc/faq/extending.rst	Fri Jul 30 01:19:37 2010
@@ -29,6 +29,8 @@
 C++ objects with constructors are probably not a good idea.
 
 
+.. _c-wrapper-software:
+
 Writing C is hard; are there any alternatives?
 ----------------------------------------------
 
@@ -200,11 +202,7 @@
 whole lot of difference between C and C++ -- so the strategy of building a new
 Python type around a C structure (pointer) type will also work for C++ objects.
 
-For C++ libraries, you can look at `SIP
-<http://www.riverbankcomputing.co.uk/software/sip/>`_, `CXX
-<http://cxx.sourceforge.net/>`_, `Boost
-<http://www.boost.org/libs/python/doc/index.html>`_, `Weave
-<http://www.scipy.org/Weave>`_ or `SWIG <http://www.swig.org>`_
+For C++ libraries, see :ref:`c-wrapper-software`.
 
 
 I added a module using the Setup file and the make fails; why?

Modified: python/branches/import_unicode/Doc/glossary.rst
==============================================================================
--- python/branches/import_unicode/Doc/glossary.rst	(original)
+++ python/branches/import_unicode/Doc/glossary.rst	Fri Jul 30 01:19:37 2010
@@ -154,15 +154,15 @@
       object.
 
    duck-typing
-      A pythonic programming style which determines an object's type by inspection
-      of its method or attribute signature rather than by explicit relationship
-      to some type object ("If it looks like a duck and quacks like a duck, it
+      A programming style which does not look at an object's type to determine
+      if it has the right interface; instead, the method or attribute is simply
+      called or used ("If it looks like a duck and quacks like a duck, it
       must be a duck.")  By emphasizing interfaces rather than specific types,
       well-designed code improves its flexibility by allowing polymorphic
       substitution.  Duck-typing avoids tests using :func:`type` or
-      :func:`isinstance`. (Note, however, that duck-typing can be complemented
-      with abstract base classes.) Instead, it typically employs :func:`hasattr`
-      tests or :term:`EAFP` programming.
+      :func:`isinstance`.  (Note, however, that duck-typing can be complemented
+      with :term:`abstract base class`\ es.)  Instead, it typically employs
+      :func:`hasattr` tests or :term:`EAFP` programming.
 
    EAFP
       Easier to ask for forgiveness than permission.  This common Python coding

Modified: python/branches/import_unicode/Doc/install/index.rst
==============================================================================
--- python/branches/import_unicode/Doc/install/index.rst	(original)
+++ python/branches/import_unicode/Doc/install/index.rst	Fri Jul 30 01:19:37 2010
@@ -314,8 +314,8 @@
 stash of Python modules.  This scheme's name is derived from the idea of a
 "home" directory on Unix, since it's not unusual for a Unix user to make their
 home directory have a layout similar to :file:`/usr/` or :file:`/usr/local/`.
-This scheme can be used by anyone, regardless of the operating system their
-installing for.
+This scheme can be used by anyone, regardless of the operating system they
+are installing for.
 
 Installing a new module distribution is as simple as ::
 

Modified: python/branches/import_unicode/Doc/library/abc.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/abc.rst	(original)
+++ python/branches/import_unicode/Doc/library/abc.rst	Fri Jul 30 01:19:37 2010
@@ -122,7 +122,7 @@
 
 It also provides the following decorators:
 
-.. function:: abstractmethod(function)
+.. decorator:: abstractmethod(function)
 
    A decorator indicating abstract methods.
 

Modified: python/branches/import_unicode/Doc/library/argparse.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/argparse.rst	(original)
+++ python/branches/import_unicode/Doc/library/argparse.rst	Fri Jul 30 01:19:37 2010
@@ -698,8 +698,8 @@
 
    >>> class FooAction(argparse.Action):
    ...     def __call__(self, parser, namespace, values, option_string=None):
-   ...     print('%r %r %r' % (namespace, values, option_string))
-   ...     setattr(namespace, self.dest, values)
+   ...         print('%r %r %r' % (namespace, values, option_string))
+   ...         setattr(namespace, self.dest, values)
    ...
    >>> parser = argparse.ArgumentParser()
    >>> parser.add_argument('--foo', action=FooAction)

Modified: python/branches/import_unicode/Doc/library/ast.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/ast.rst	(original)
+++ python/branches/import_unicode/Doc/library/ast.rst	Fri Jul 30 01:19:37 2010
@@ -119,12 +119,15 @@
 
    Safely evaluate an expression node or a string containing a Python
    expression.  The string or node provided may only consist of the following
-   Python literal structures: strings, numbers, tuples, lists, dicts, booleans,
-   and ``None``.
+   Python literal structures: strings, bytes, numbers, tuples, lists, dicts,
+   sets, booleans, and ``None``.
 
    This can be used for safely evaluating strings containing Python expressions
    from untrusted sources without the need to parse the values oneself.
 
+   .. versionchanged:: 3.2
+      Now allows bytes and set literals.
+
 
 .. function:: get_docstring(node, clean=True)
 

Modified: python/branches/import_unicode/Doc/library/binascii.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/binascii.rst	(original)
+++ python/branches/import_unicode/Doc/library/binascii.rst	Fri Jul 30 01:19:37 2010
@@ -18,6 +18,11 @@
 low-level functions written in C for greater speed that are used by the
 higher-level modules.
 
+.. note::
+
+   Encoding and decoding functions do not accept Unicode strings.  Only bytestring
+   and bytearray objects can be processed.
+
 The :mod:`binascii` module defines the following functions:
 
 
@@ -54,6 +59,9 @@
    data. More than one line may be passed at a time. If the optional argument
    *header* is present and true, underscores will be decoded as spaces.
 
+   .. versionchanged:: 3.2
+      accept only bytestring or bytearray object as input.
+
 
 .. function:: b2a_qp(data, quotetabs=False, istext=True, header=False)
 
@@ -83,6 +91,9 @@
    decompressed data, unless data input data ends in an orphaned repeat indicator,
    in which case the :exc:`Incomplete` exception is raised.
 
+   .. versionchanged:: 3.2
+      accept only bytestring or bytearray object as input.
+
 
 .. function:: rlecode_hqx(data)
 
@@ -139,6 +150,9 @@
    of hexadecimal digits (which can be upper or lower case), otherwise a
    :exc:`TypeError` is raised.
 
+   .. versionchanged:: 3.2
+      accept only bytestring or bytearray object as input.
+
 
 .. exception:: Error
 
@@ -164,4 +178,3 @@
 
    Module :mod:`quopri`
       Support for quoted-printable encoding used in MIME email messages.
-

Modified: python/branches/import_unicode/Doc/library/cmath.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/cmath.rst	(original)
+++ python/branches/import_unicode/Doc/library/cmath.rst	Fri Jul 30 01:19:37 2010
@@ -187,15 +187,24 @@
 Classification functions
 ------------------------
 
+.. function:: isfinite(x)
+
+   Return ``True`` if both the real and imaginary parts of *x* are finite, and
+   ``False`` otherwise.
+
+   .. versionadded:: 3.2
+
+
 .. function:: isinf(x)
 
-   Return *True* if the real or the imaginary part of x is positive
-   or negative infinity.
+   Return ``True`` if either the real or the imaginary part of *x* is an
+   infinity, and ``False`` otherwise.
 
 
 .. function:: isnan(x)
 
-   Return *True* if the real or imaginary part of x is not a number (NaN).
+   Return ``True`` if either the real or the imaginary part of *x* is a NaN,
+   and ``False`` otherwise.
 
 
 Constants

Modified: python/branches/import_unicode/Doc/library/collections.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/collections.rst	(original)
+++ python/branches/import_unicode/Doc/library/collections.rst	Fri Jul 30 01:19:37 2010
@@ -551,7 +551,7 @@
    :class:`defaultdict` objects support the following method in addition to the
    standard :class:`dict` operations:
 
-   .. method:: defaultdict.__missing__(key)
+   .. method:: __missing__(key)
 
       If the :attr:`default_factory` attribute is ``None``, this raises a
       :exc:`KeyError` exception with the *key* as argument.
@@ -571,7 +571,7 @@
    :class:`defaultdict` objects support the following instance variable:
 
 
-   .. attribute:: defaultdict.default_factory
+   .. attribute:: default_factory
 
       This attribute is used by the :meth:`__missing__` method; it is
       initialized from the first argument to the constructor, if present, or to
@@ -763,7 +763,7 @@
 three additional methods and one attribute.  To prevent conflicts with
 field names, the method and attribute names start with an underscore.
 
-.. method:: somenamedtuple._make(iterable)
+.. classmethod:: somenamedtuple._make(iterable)
 
    Class method that makes a new instance from an existing sequence or iterable.
 
@@ -891,11 +891,11 @@
 
    .. versionadded:: 3.1
 
-.. method:: OrderedDict.popitem(last=True)
+   .. method:: popitem(last=True)
 
-   The :meth:`popitem` method for ordered dictionaries returns and removes
-   a (key, value) pair.  The pairs are returned in LIFO order if *last* is
-   true or FIFO order if false.
+      The :meth:`popitem` method for ordered dictionaries returns and removes a
+      (key, value) pair.  The pairs are returned in LIFO order if *last* is true
+      or FIFO order if false.
 
 In addition to the usual mapping methods, ordered dictionaries also support
 reverse iteration using :func:`reversed`.
@@ -956,12 +956,13 @@
    initialized with its contents; note that a reference to *initialdata* will not
    be kept, allowing it be used for other purposes.
 
-In addition to supporting the methods and operations of mappings,
-:class:`UserDict` instances provide the following attribute:
+   In addition to supporting the methods and operations of mappings,
+   :class:`UserDict` instances provide the following attribute:
 
-.. attribute:: UserDict.data
+   .. attribute:: data
 
-   A real dictionary used to store the contents of the :class:`UserDict` class.
+      A real dictionary used to store the contents of the :class:`UserDict`
+      class.
 
 
 
@@ -985,13 +986,13 @@
    defaulting to the empty list ``[]``.  *list* can be any iterable, for
    example a real Python list or a :class:`UserList` object.
 
-In addition to supporting the methods and operations of mutable sequences,
-:class:`UserList` instances provide the following attribute:
+   In addition to supporting the methods and operations of mutable sequences,
+   :class:`UserList` instances provide the following attribute:
 
-.. attribute:: UserList.data
+   .. attribute:: data
 
-   A real :class:`list` object used to store the contents of the
-   :class:`UserList` class.
+      A real :class:`list` object used to store the contents of the
+      :class:`UserList` class.
 
 **Subclassing requirements:** Subclasses of :class:`UserList` are expect to
 offer a constructor which can be called with either no arguments or one

Modified: python/branches/import_unicode/Doc/library/configparser.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/configparser.rst	(original)
+++ python/branches/import_unicode/Doc/library/configparser.rst	Fri Jul 30 01:19:37 2010
@@ -15,105 +15,172 @@
    single: ini file
    single: Windows ini file
 
-This module defines the class :class:`ConfigParser`.   The :class:`ConfigParser`
-class implements a basic configuration file parser language which provides a
-structure similar to what you would find on Microsoft Windows INI files.  You
-can use this to write Python programs which can be customized by end users
-easily.
+This module provides the classes :class:`RawConfigParser` and
+:class:`SafeConfigParser`.  They implement a basic configuration file parser
+language which provides a structure similar to what you would find in Microsoft
+Windows INI files.  You can use this to write Python programs which can be
+customized by end users easily.
 
 .. note::
 
    This library does *not* interpret or write the value-type prefixes used in
    the Windows Registry extended version of INI syntax.
 
-The configuration file consists of sections, led by a ``[section]`` header and
-followed by ``name: value`` entries, with continuations in the style of
-:rfc:`822` (see section 3.1.1, "LONG HEADER FIELDS"); ``name=value`` is also
-accepted.  Note that leading whitespace is removed from values. The optional
-values can contain format strings which refer to other values in the same
-section, or values in a special ``DEFAULT`` section.  Additional defaults can be
-provided on initialization and retrieval.  Lines beginning with ``'#'`` or
-``';'`` are ignored and may be used to provide comments.
+A configuration file consists of sections, each led by a ``[section]`` header,
+followed by name/value entries separated by a specific string (``=`` or ``:`` by
+default).  Note that leading whitespace is removed from values.  Values can be
+ommitted, in which case the key/value delimiter may also be left out.  Values
+can also span multiple lines, as long as they are indented deeper than the first
+line of the value.  Depending on the parser's mode, blank lines may be treated
+as parts of multiline values or ignored.
+
+Configuration files may include comments, prefixed by specific characters (``#``
+and ``;`` by default).  Comments may appear on their own in an otherwise empty
+line, or may be entered in lines holding values or spection names.  In the
+latter case, they need to be preceded by a whitespace character to be recognized
+as a comment.  (For backwards compatibility, by default only ``;`` starts an
+inline comment, while ``#`` does not.)
+
+On top of the core functionality, :class:`SafeConfigParser` supports
+interpolation.  This means values can contain format strings which refer to
+other values in the same section, or values in a special ``DEFAULT`` section.
+Additional defaults can be provided on initialization.
 
 For example::
 
-   [My Section]
-   foodir: %(dir)s/whatever
-   dir=frob
-   long: this value continues
-      in the next line
-
-would resolve the ``%(dir)s`` to the value of ``dir`` (``frob`` in this case).
-All reference expansions are done on demand.
-
-Default values can be specified by passing them into the :class:`ConfigParser`
-constructor as a dictionary.  Additional defaults  may be passed into the
-:meth:`get` method which will override all others.
-
-Sections are normally stored in a built-in dictionary. An alternative dictionary
-type can be passed to the :class:`ConfigParser` constructor. For example, if a
-dictionary type is passed that sorts its keys, the sections will be sorted on
-write-back, as will be the keys within each section.
+   [Paths]
+   home_dir: /Users
+   my_dir: %(home_dir)s/lumberjack
+   my_pictures: %(my_dir)s/Pictures
+
+   [Multiline Values]
+   chorus: I'm a lumberjack, and I'm okay
+      I sleep all night and I work all day
+
+   [No Values]
+   key_without_value
+   empty string value here =
+
+   [You can use comments] ; after a useful line
+   ; in an empty line
+   after: a_value ; here's another comment
+   inside: a         ;comment
+           multiline ;comment
+           value!    ;comment
+
+      [Sections Can Be Indented]
+         can_values_be_as_well = True
+         does_that_mean_anything_special = False
+         purpose = formatting for readability
+         multiline_values = are
+            handled just fine as
+            long as they are indented
+            deeper than the first line
+            of a value
+         # Did I mention we can indent comments, too?
+
+
+In the example above, :class:`SafeConfigParser` would resolve ``%(home_dir)s``
+to the value of ``home_dir`` (``/Users`` in this case).  ``%(my_dir)s`` in
+effect would resolve to ``/Users/lumberjack``.  All interpolations are done on
+demand so keys used in the chain of references do not have to be specified in
+any specific order in the configuration file.
+
+:class:`RawConfigParser` would simply return ``%(my_dir)s/Pictures`` as the
+value of ``my_pictures`` and ``%(home_dir)s/lumberjack`` as the value of
+``my_dir``.  Other features presented in the example are handled in the same
+manner by both parsers.
+
+Default values can be specified by passing them as a dictionary when
+constructing the :class:`SafeConfigParser`.
+
+Sections are normally stored in an :class:`collections.OrderedDict` which
+maintains the order of all keys.  An alternative dictionary type can be passed
+to the :meth:`__init__` method.  For example, if a dictionary type is passed
+that sorts its keys, the sections will be sorted on write-back, as will be the
+keys within each section.
 
 
-.. class:: RawConfigParser(defaults=None, dict_type=collections.OrderedDict,
-                           allow_no_value=False)
+.. class:: RawConfigParser(defaults=None, dict_type=collections.OrderedDict, delimiters=('=', ':'), comment_prefixes=_COMPATIBLE, empty_lines_in_values=True, allow_no_value=False)
 
    The basic configuration object.  When *defaults* is given, it is initialized
-   into the dictionary of intrinsic defaults.  When *dict_type* is given, it will
-   be used to create the dictionary objects for the list of sections, for the
-   options within a section, and for the default values.  When *allow_no_value*
-   is true (default: ``False``), options without values are accepted; the value
+   into the dictionary of intrinsic defaults.  When *dict_type* is given, it
+   will be used to create the dictionary objects for the list of sections, for
+   the options within a section, and for the default values.
+
+   When *delimiters* is given, it will be used as the set of substrings that
+   divide keys from values.  When *comment_prefixes* is given, it will be used
+   as the set of substrings that prefix comments in a line, both for the whole
+   line and inline comments.  For backwards compatibility, the default value for
+   *comment_prefixes* is a special value that indicates that ``;`` and ``#`` can
+   start whole line comments while only ``;`` can start inline comments.
+
+   When *empty_lines_in_values* is ``False`` (default: ``True``), each empty
+   line marks the end of an option.  Otherwise, internal empty lines of a
+   multiline option are kept as part of the value.  When *allow_no_value* is
+   true (default: ``False``), options without values are accepted; the value
    presented for these is ``None``.
 
-   This class does not
-   support the magical interpolation behavior.
+   This class does not support the magical interpolation behavior.
 
    .. versionchanged:: 3.1
       The default *dict_type* is :class:`collections.OrderedDict`.
 
    .. versionchanged:: 3.2
-      *allow_no_value* was added.
+      *delimiters*, *comment_prefixes*, *empty_lines_in_values* and
+      *allow_no_value* were added.
 
 
-.. class:: ConfigParser(defaults=None, dict_type=collections.OrderedDict,
-                        allow_no_value=False)
+.. class:: SafeConfigParser(defaults=None, dict_type=collections.OrderedDict, delimiters=('=', ':'), comment_prefixes=('#', ';'), empty_lines_in_values=True, allow_no_value=False)
 
-   Derived class of :class:`RawConfigParser` that implements the magical
-   interpolation feature and adds optional arguments to the :meth:`get` and
-   :meth:`items` methods.  The values in *defaults* must be appropriate for the
-   ``%()s`` string interpolation.  Note that *__name__* is an intrinsic default;
-   its value is the section name, and will override any value provided in
-   *defaults*.
+   Derived class of :class:`ConfigParser` that implements a sane variant of the
+   magical interpolation feature.  This implementation is more predictable as it
+   validates the interpolation syntax used within a configuration file.  This
+   class also enables escaping the interpolation character (e.g. a key can have
+   ``%`` as part of the value by specifying ``%%`` in the file).
 
-   All option names used in interpolation will be passed through the
-   :meth:`optionxform` method just like any other option name reference.  For
-   example, using the default implementation of :meth:`optionxform` (which converts
-   option names to lower case), the values ``foo %(bar)s`` and ``foo %(BAR)s`` are
-   equivalent.
+   Applications that don't require interpolation should use
+   :class:`RawConfigParser`, otherwise :class:`SafeConfigParser` is the best
+   option.
 
    .. versionchanged:: 3.1
       The default *dict_type* is :class:`collections.OrderedDict`.
 
    .. versionchanged:: 3.2
-      *allow_no_value* was added.
+      *delimiters*, *comment_prefixes*, *empty_lines_in_values* and
+      *allow_no_value* were added.
 
 
-.. class:: SafeConfigParser(defaults=None, dict_type=collections.OrderedDict,
-                            allow_no_value=False)
+.. class:: ConfigParser(defaults=None, dict_type=collections.OrderedDict, delimiters=('=', ':'), comment_prefixes=('#', ';'), empty_lines_in_values=True, allow_no_value=False)
 
-   Derived class of :class:`ConfigParser` that implements a more-sane variant of
-   the magical interpolation feature.  This implementation is more predictable as
-   well. New applications should prefer this version if they don't need to be
-   compatible with older versions of Python.
+   Derived class of :class:`RawConfigParser` that implements the magical
+   interpolation feature and adds optional arguments to the :meth:`get` and
+   :meth:`items` methods.
 
-   .. XXX Need to explain what's safer/more predictable about it.
+   :class:`SafeConfigParser` is generally recommended over this class if you
+   need interpolation.
+
+   The values in *defaults* must be appropriate for the ``%()s`` string
+   interpolation.  Note that *__name__* is an intrinsic default; its value is
+   the section name, and will override any value provided in *defaults*.
+
+   All option names used in interpolation will be passed through the
+   :meth:`optionxform` method just like any other option name reference.  For
+   example, using the default implementation of :meth:`optionxform` (which
+   converts option names to lower case), the values ``foo %(bar)s`` and ``foo
+   %(BAR)s`` are equivalent.
 
    .. versionchanged:: 3.1
       The default *dict_type* is :class:`collections.OrderedDict`.
 
    .. versionchanged:: 3.2
-      *allow_no_value* was added.
+      *delimiters*, *comment_prefixes*, *empty_lines_in_values* and
+      *allow_no_value* were added.
+
+
+.. exception:: Error
+
+   Base class for all other configparser exceptions.
 
 
 .. exception:: NoSectionError
@@ -223,25 +290,29 @@
    :const:`True`; otherwise return :const:`False`.
 
 
-.. method:: RawConfigParser.read(filenames)
+.. method:: RawConfigParser.read(filenames, encoding=None)
 
    Attempt to read and parse a list of filenames, returning a list of filenames
-   which were successfully parsed.  If *filenames* is a string,
-   it is treated as a single filename. If a file named in *filenames* cannot be
-   opened, that file will be ignored.  This is designed so that you can specify a
-   list of potential configuration file locations (for example, the current
-   directory, the user's home directory, and some system-wide directory), and all
-   existing configuration files in the list will be read.  If none of the named
-   files exist, the :class:`ConfigParser` instance will contain an empty dataset.
-   An application which requires initial values to be loaded from a file should
-   load the required file or files using :meth:`readfp` before calling :meth:`read`
-   for any optional files::
+   which were successfully parsed.  If *filenames* is a string, it is treated as
+   a single filename.  If a file named in *filenames* cannot be opened, that
+   file will be ignored.  This is designed so that you can specify a list of
+   potential configuration file locations (for example, the current directory,
+   the user's home directory, and some system-wide directory), and all existing
+   configuration files in the list will be read.  If none of the named files
+   exist, the :class:`ConfigParser` instance will contain an empty dataset.  An
+   application which requires initial values to be loaded from a file should
+   load the required file or files using :meth:`readfp` before calling
+   :meth:`read` for any optional files::
 
       import configparser, os
 
       config = configparser.ConfigParser()
       config.readfp(open('defaults.cfg'))
-      config.read(['site.cfg', os.path.expanduser('~/.myapp.cfg')])
+      config.read(['site.cfg', os.path.expanduser('~/.myapp.cfg')], encoding='cp1250')
+
+   .. versionadded:: 3.2
+      The *encoding* parameter.  Previously, all files were read using the
+      default encoding for :func:`open`.
 
 
 .. method:: RawConfigParser.readfp(fp, filename=None)
@@ -295,11 +366,13 @@
    interpolation and output to files) can only be achieved using string values.
 
 
-.. method:: RawConfigParser.write(fileobject)
+.. method:: RawConfigParser.write(fileobject, space_around_delimiters=True)
 
    Write a representation of the configuration to the specified file object,
    which must be opened in text mode (accepting strings).  This representation
-   can be parsed by a future :meth:`read` call.
+   can be parsed by a future :meth:`read` call. If ``space_around_delimiters``
+   is ``True`` (the default), delimiters between keys and values are surrounded
+   by spaces.
 
 
 .. method:: RawConfigParser.remove_option(section, option)
@@ -342,21 +415,27 @@
 --------------------
 
 The :class:`ConfigParser` class extends some methods of the
-:class:`RawConfigParser` interface, adding some optional arguments.
+:class:`RawConfigParser` interface, adding some optional arguments. Whenever you
+can, consider using :class:`SafeConfigParser` which adds validation and escaping
+for the interpolation.
 
 
 .. method:: ConfigParser.get(section, option, raw=False, vars=None)
 
-   Get an *option* value for the named *section*.  All the ``'%'`` interpolations
-   are expanded in the return values, based on the defaults passed into the
-   constructor, as well as the options *vars* provided, unless the *raw* argument
-   is true.
+   Get an *option* value for the named *section*.  If *vars* is provided, it
+   must be a dictionary.  The *option* is looked up in *vars* (if provided),
+   *section*, and in *defaults* in that order.
+
+   All the ``'%'`` interpolations are expanded in the return values, unless the
+   *raw* argument is true.  Values for interpolation keys are looked up in the
+   same manner as the option.
 
 
 .. method:: ConfigParser.items(section, raw=False, vars=None)
 
-   Return a list of ``(name, value)`` pairs for each option in the given *section*.
-   Optional arguments have the same meaning as for the :meth:`get` method.
+   Return a list of ``(name, value)`` pairs for each option in the given
+   *section*.  Optional arguments have the same meaning as for the :meth:`get`
+   method.
 
 
 .. _safeconfigparser-objects:
@@ -466,8 +545,8 @@
 
 Some configuration files are known to include settings without values, but which
 otherwise conform to the syntax supported by :mod:`configparser`.  The
-*allow_no_value* parameter to the constructor can be used to indicate that such
-values should be accepted:
+*allow_no_value* parameter to the :meth:`__init__` method can be used to
+indicate that such values should be accepted:
 
 .. doctest::
 
@@ -476,12 +555,12 @@
 
    >>> sample_config = """
    ... [mysqld]
-   ... user = mysql
-   ... pid-file = /var/run/mysqld/mysqld.pid
-   ... skip-external-locking
-   ... old_passwords = 1
-   ... skip-bdb
-   ... skip-innodb
+   ...   user = mysql
+   ...   pid-file = /var/run/mysqld/mysqld.pid
+   ...   skip-external-locking
+   ...   old_passwords = 1
+   ...   skip-bdb
+   ...   skip-innodb # we don't need ACID today
    ... """
    >>> config = configparser.RawConfigParser(allow_no_value=True)
    >>> config.readfp(io.BytesIO(sample_config))

Modified: python/branches/import_unicode/Doc/library/contextlib.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/contextlib.rst	(original)
+++ python/branches/import_unicode/Doc/library/contextlib.rst	Fri Jul 30 01:19:37 2010
@@ -12,7 +12,7 @@
 Functions provided:
 
 
-.. function:: contextmanager(func)
+.. decorator:: contextmanager
 
    This function is a :term:`decorator` that can be used to define a factory
    function for :keyword:`with` statement context managers, without needing to
@@ -57,6 +57,7 @@
    .. versionchanged:: 3.2
       Use of :class:`ContextDecorator`.
 
+
 .. function:: closing(thing)
 
    Return a context manager that closes *thing* upon completion of the block.  This
@@ -92,22 +93,25 @@
    ``__enter__`` and ``__exit__`` as normal. ``__exit__`` retains its optional
    exception handling even when used as a decorator.
 
-   Example::
+   ``ContextDecorator`` is used by :func:`contextmanager`, so you get this
+   functionality automatically.
+
+   Example of ``ContextDecorator``::
 
       from contextlib import ContextDecorator
 
       class mycontext(ContextDecorator):
-         def __enter__(self):
-            print('Starting')
-            return self
-
-         def __exit__(self, *exc):
-            print('Finishing')
-            return False
+          def __enter__(self):
+              print('Starting')
+              return self
+
+          def __exit__(self, *exc):
+              print('Finishing')
+              return False
 
       >>> @mycontext()
       ... def function():
-      ...    print('The bit in the middle')
+      ...     print('The bit in the middle')
       ...
       >>> function()
       Starting
@@ -115,23 +119,38 @@
       Finishing
 
       >>> with mycontext():
-      ...    print('The bit in the middle')
+      ...     print('The bit in the middle')
       ...
       Starting
       The bit in the middle
       Finishing
 
+   This change is just syntactic sugar for any construct of the following form::
+
+      def f():
+          with cm():
+              # Do stuff
+
+   ``ContextDecorator`` lets you instead write::
+
+      @cm()
+      def f():
+          # Do stuff
+
+   It makes it clear that the ``cm`` applies to the whole function, rather than
+   just a piece of it (and saving an indentation level is nice, too).
+
    Existing context managers that already have a base class can be extended by
    using ``ContextDecorator`` as a mixin class::
 
       from contextlib import ContextDecorator
 
       class mycontext(ContextBaseClass, ContextDecorator):
-         def __enter__(self):
-            return self
+          def __enter__(self):
+              return self
 
-         def __exit__(self, *exc):
-            return False
+          def __exit__(self, *exc):
+              return False
 
    .. versionadded:: 3.2
 

Modified: python/branches/import_unicode/Doc/library/dbm.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/dbm.rst	(original)
+++ python/branches/import_unicode/Doc/library/dbm.rst	Fri Jul 30 01:19:37 2010
@@ -5,10 +5,10 @@
    :synopsis: Interfaces to various Unix "database" formats.
 
 :mod:`dbm` is a generic interface to variants of the DBM database ---
- :mod:`dbm.gnu` or :mod:`dbm.ndbm`.  If none of these modules is installed, the
- slow-but-simple implementation in module :mod:`dbm.dumb` will be used.  There
- is a `third party interface <http://www.jcea.es/programacion/pybsddb.htm>`_ to
- the Oracle Berkely DB.
+:mod:`dbm.gnu` or :mod:`dbm.ndbm`.  If none of these modules is installed, the
+slow-but-simple implementation in module :mod:`dbm.dumb` will be used.  There
+is a `third party interface <http://www.jcea.es/programacion/pybsddb.htm>`_ to
+the Oracle Berkely DB.
 
 
 .. exception:: error
@@ -21,8 +21,8 @@
 .. function:: whichdb(filename)
 
    This functionattempts to guess which of the several simple database modules
-   available --- :mod:`dbm.bsd`, :mod:`dbm.gnu`, :mod:`dbm.ndbm` or
-   :mod:`dbm.dumb` --- should be used to open a given file.
+   available --- :mod:`dbm.gnu`, :mod:`dbm.ndbm` or :mod:`dbm.dumb` --- should
+   be used to open a given file.
 
    Returns one of the following values: ``None`` if the file can't be opened
    because it's unreadable or doesn't exist; the empty string (``''``) if the
@@ -227,10 +227,9 @@
 always stored as bytes. Printing a ``dbm`` object doesn't print the keys and
 values, and the :meth:`items` and :meth:`values` methods are not supported.
 
-This module can be used with the "classic" ndbm interface, the BSD DB
-compatibility interface, or the GNU GDBM compatibility interface. On Unix, the
-:program:`configure` script will attempt to locate the appropriate header file
-to simplify building this module.
+This module can be used with the "classic" ndbm interface or the GNU GDBM
+compatibility interface. On Unix, the :program:`configure` script will attempt
+to locate the appropriate header file to simplify building this module.
 
 .. exception:: error
 
@@ -246,9 +245,7 @@
 .. function:: open(filename[, flag[, mode]])
 
    Open a dbm database and return a ``dbm`` object.  The *filename* argument is the
-   name of the database file (without the :file:`.dir` or :file:`.pag` extensions;
-   note that the BSD DB implementation of the interface will append the extension
-   :file:`.db` and only create one file).
+   name of the database file (without the :file:`.dir` or :file:`.pag` extensions).
 
    The optional *flag* argument must be one of these values:
 

Modified: python/branches/import_unicode/Doc/library/doctest.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/doctest.rst	(original)
+++ python/branches/import_unicode/Doc/library/doctest.rst	Fri Jul 30 01:19:37 2010
@@ -913,18 +913,16 @@
 As your collection of doctest'ed modules grows, you'll want a way to run all
 their doctests systematically.  :mod:`doctest` provides two functions that can
 be used to create :mod:`unittest` test suites from modules and text files
-containing doctests.  These test suites can then be run using :mod:`unittest`
-test runners::
+containing doctests.  To integrate with :mod:`unittest` test discovery, include
+a :func:`load_tests` function in your test module::
 
    import unittest
    import doctest
-   import my_module_with_doctests, and_another
+   import my_module_with_doctests
 
-   suite = unittest.TestSuite()
-   for mod in my_module_with_doctests, and_another:
-       suite.addTest(doctest.DocTestSuite(mod))
-   runner = unittest.TextTestRunner()
-   runner.run(suite)
+   def load_tests(loader, tests, ignore):
+       tests.addTests(doctest.DocTestSuite(my_module_with_doctests))
+       return test
 
 There are two main functions for creating :class:`unittest.TestSuite` instances
 from text files and modules with doctests:

Modified: python/branches/import_unicode/Doc/library/fnmatch.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/fnmatch.rst	(original)
+++ python/branches/import_unicode/Doc/library/fnmatch.rst	Fri Jul 30 01:19:37 2010
@@ -70,6 +70,8 @@
 
    Return the shell-style *pattern* converted to a regular expression.
 
+   Be aware there is no way to quote meta-characters.
+
    Example:
 
       >>> import fnmatch, re
@@ -82,6 +84,13 @@
       <_sre.SRE_Match object at 0x...>
 
 
+.. function:: purge()
+
+   Clear the internal pattern cache.
+
+   .. versionadded:: 3.2
+
+
 .. seealso::
 
    Module :mod:`glob`

Modified: python/branches/import_unicode/Doc/library/functions.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/functions.rst	(original)
+++ python/branches/import_unicode/Doc/library/functions.rst	Fri Jul 30 01:19:37 2010
@@ -335,6 +335,9 @@
    returns the current global and local dictionary, respectively, which may be
    useful to pass around for use by :func:`eval` or :func:`exec`.
 
+   See :func:`ast.literal_eval` for a function that can safely evaluate strings
+   with expressions containing only literals.
+
 
 .. function:: exec(object[, globals[, locals]])
 
@@ -676,7 +679,7 @@
    :meth:`__index__` method that returns an integer.
 
 
-.. function:: open(file, mode='r', buffering=None, encoding=None, errors=None, newline=None, closefd=True)
+.. function:: open(file, mode='r', buffering=-1, encoding=None, errors=None, newline=None, closefd=True)
 
    Open *file* and return a corresponding stream.  If the file cannot be opened,
    an :exc:`IOError` is raised.

Modified: python/branches/import_unicode/Doc/library/functools.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/functools.rst	(original)
+++ python/branches/import_unicode/Doc/library/functools.rst	Fri Jul 30 01:19:37 2010
@@ -37,7 +37,7 @@
 
    .. versionadded:: 3.2
 
-.. function:: total_ordering(cls)
+.. decorator:: total_ordering
 
    Given a class defining one or more rich comparison ordering methods, this
    class decorator supplies the rest.  This simplifies the effort involved
@@ -122,7 +122,7 @@
    than helpful.
 
 
-.. function:: wraps(wrapped, assigned=WRAPPER_ASSIGNMENTS, updated=WRAPPER_UPDATES)
+.. decorator:: wraps(wrapped, assigned=WRAPPER_ASSIGNMENTS, updated=WRAPPER_UPDATES)
 
    This is a convenience function for invoking ``partial(update_wrapper,
    wrapped=wrapped, assigned=assigned, updated=updated)`` as a function decorator

Modified: python/branches/import_unicode/Doc/library/html.parser.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/html.parser.rst	(original)
+++ python/branches/import_unicode/Doc/library/html.parser.rst	Fri Jul 30 01:19:37 2010
@@ -134,10 +134,18 @@
 
 .. method:: HTMLParser.handle_decl(decl)
 
-   Method called when an SGML declaration is read by the parser.  The *decl*
-   parameter will be the entire contents of the declaration inside the ``<!``...\
-   ``>`` markup.  It is intended to be overridden by a derived class; the base
-   class implementation does nothing.
+   Method called when an SGML ``doctype`` declaration is read by the parser.
+   The *decl* parameter will be the entire contents of the declaration inside
+   the ``<!...>`` markup.  It is intended to be overridden by a derived class;
+   the base class implementation does nothing.
+
+
+.. method:: HTMLParser.unknown_decl(data)
+
+   Method called when an unrecognized SGML declaration is read by the parser.
+   The *data* parameter will be the entire contents of the declaration inside
+   the ``<!...>`` markup.  It is sometimes useful to be be overridden by a
+   derived class; the base class implementation throws an :exc:`HTMLParseError`.
 
 
 .. method:: HTMLParser.handle_pi(data)

Modified: python/branches/import_unicode/Doc/library/http.cookiejar.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/http.cookiejar.rst	(original)
+++ python/branches/import_unicode/Doc/library/http.cookiejar.rst	Fri Jul 30 01:19:37 2010
@@ -108,10 +108,6 @@
       :mod:`http.cookiejar` and :mod:`http.cookies` modules do not depend on each
       other.
 
-   http://wwwsearch.sourceforge.net/mechanize/
-      Extensions to this module, including a class for reading Microsoft Internet
-      Explorer cookies on Windows.
-
    http://wp.netscape.com/newsref/std/cookie_spec.html
       The specification of the original Netscape cookie protocol.  Though this is
       still the dominant protocol, the 'Netscape cookie protocol' implemented by all
@@ -292,11 +288,8 @@
 FileCookieJar subclasses and co-operation with web browsers
 -----------------------------------------------------------
 
-The following :class:`CookieJar` subclasses are provided for reading and writing
-.  Further :class:`CookieJar` subclasses, including one that reads Microsoft
-Internet Explorer cookies, are available at
-http://wwwsearch.sourceforge.net/mechanize/.
-
+The following :class:`CookieJar` subclasses are provided for reading and
+writing .
 
 .. class:: MozillaCookieJar(filename, delayload=None, policy=None)
 

Modified: python/branches/import_unicode/Doc/library/importlib.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/importlib.rst	(original)
+++ python/branches/import_unicode/Doc/library/importlib.rst	Fri Jul 30 01:19:37 2010
@@ -228,7 +228,7 @@
 
     * :meth:`ResourceLoader.get_data`
     * :meth:`ExecutionLoader.get_filename`
-          Implement to only return the path to the source file; sourceless
+          Should only return the path to the source file; sourceless
           loading is not supported.
 
     The abstract methods defined by this class are to add optional bytecode
@@ -469,7 +469,7 @@
 This module contains the various objects that help in the construction of
 an :term:`importer`.
 
-.. function:: module_for_loader(method)
+.. decorator:: module_for_loader
 
     A :term:`decorator` for a :term:`loader` method,
     to handle selecting the proper
@@ -494,7 +494,7 @@
     Use of this decorator handles all the details of which module object a
     loader should initialize as specified by :pep:`302`.
 
-.. function:: set_loader(fxn)
+.. decorator:: set_loader
 
     A :term:`decorator` for a :term:`loader` method,
     to set the :attr:`__loader__`
@@ -502,7 +502,7 @@
     does nothing. It is assumed that the first positional argument to the
     wrapped method is what :attr:`__loader__` should be set to.
 
-.. function:: set_package(fxn)
+.. decorator:: set_package
 
     A :term:`decorator` for a :term:`loader` to set the :attr:`__package__`
     attribute on the module returned by the loader. If :attr:`__package__` is

Modified: python/branches/import_unicode/Doc/library/io.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/io.rst	(original)
+++ python/branches/import_unicode/Doc/library/io.rst	Fri Jul 30 01:19:37 2010
@@ -61,8 +61,8 @@
    Open *file* and return a corresponding stream.  If the file cannot be opened,
    an :exc:`IOError` is raised.
 
-   *file* is either a string or bytes object giving the name (and the path if
-   the file isn't in the current working directory) of the file to be opened or
+   *file* is either a string or bytes object giving the pathname (absolute or
+   relative to the current working directory) of the file to be opened or
    an integer file descriptor of the file to be wrapped.  (If a file descriptor
    is given, it is closed when the returned I/O object is closed, unless
    *closefd* is set to ``False``.)

Modified: python/branches/import_unicode/Doc/library/itertools.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/itertools.rst	(original)
+++ python/branches/import_unicode/Doc/library/itertools.rst	Fri Jul 30 01:19:37 2010
@@ -99,7 +99,7 @@
                   yield element
 
 
-.. function:: itertools.chain.from_iterable(iterable)
+.. classmethod:: chain.from_iterable(iterable)
 
    Alternate constructor for :func:`chain`.  Gets chained inputs from a
    single iterable argument that is evaluated lazily.  Equivalent to::

Modified: python/branches/import_unicode/Doc/library/locale.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/locale.rst	(original)
+++ python/branches/import_unicode/Doc/library/locale.rst	Fri Jul 30 01:19:37 2010
@@ -244,10 +244,6 @@
       specified, and therefore you should not assume knowledge of it on different
       systems.
 
-   .. data:: ERA_YEAR
-
-      Get the year in the relevant era of the locale.
-
    .. data:: ERA_D_T_FMT
 
       Get a format string for :func:`strftime` to represent dates and times in a

Modified: python/branches/import_unicode/Doc/library/math.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/math.rst	(original)
+++ python/branches/import_unicode/Doc/library/math.rst	Fri Jul 30 01:19:37 2010
@@ -97,15 +97,23 @@
    <http://code.activestate.com/recipes/393090/>`_\.
 
 
+.. function:: isfinite(x)
+
+   Return ``True`` if *x* is neither an infinity nor a NaN, and
+   ``False`` otherwise.  (Note that ``0.0`` *is* considered finite.)
+
+   .. versionadded:: 3.2
+
+
 .. function:: isinf(x)
 
-   Check if the float *x* is positive or negative infinity.
+   Return ``True`` if *x* is a positive or negative infinity, and
+   ``False`` otherwise.
 
 
 .. function:: isnan(x)
 
-   Check if the float *x* is a NaN (not a number).  For more information
-   on NaNs, see the IEEE 754 standards.
+   Return ``True`` if *x* is a NaN (not a number), and ``False`` otherwise.
 
 
 .. function:: ldexp(x, i)

Modified: python/branches/import_unicode/Doc/library/multiprocessing.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/multiprocessing.rst	(original)
+++ python/branches/import_unicode/Doc/library/multiprocessing.rst	Fri Jul 30 01:19:37 2010
@@ -2228,8 +2228,8 @@
 
 
 An example of how a pool of worker processes can each run a
-:class:`SimpleHTTPServer.HttpServer` instance while sharing a single listening
-socket.
+:class:`~http.server.SimpleHTTPRequestHandler` instance while sharing a single
+listening socket.
 
 .. literalinclude:: ../includes/mp_webserver.py
 

Modified: python/branches/import_unicode/Doc/library/os.path.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/os.path.rst	(original)
+++ python/branches/import_unicode/Doc/library/os.path.rst	Fri Jul 30 01:19:37 2010
@@ -236,11 +236,14 @@
    exception if a :func:`os.stat` call on either pathname fails.
 
    On Windows, two files are the same if they resolve to the same final path
-   name using the Windows API call GetFinalPathNameByHandle and this function
+   name using the Windows API call GetFinalPathNameByHandle. This function
    raises an exception if handles cannot be obtained to either file.
 
    Availability: Windows, Unix.
 
+   .. versionchanged:: 3.2
+      Added Windows support.
+
 
 .. function:: sameopenfile(fp1, fp2)
 

Modified: python/branches/import_unicode/Doc/library/os.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/os.rst	(original)
+++ python/branches/import_unicode/Doc/library/os.rst	Fri Jul 30 01:19:37 2010
@@ -723,7 +723,7 @@
 
       This function is intended for low-level I/O.  For normal usage, use the
       built-in function :func:`open`, which returns a "file object" with
-      :meth:`~file.read` and :meth:`~file.wprite` methods (and many more).  To
+      :meth:`~file.read` and :meth:`~file.write` methods (and many more).  To
       wrap a file descriptor in a "file object", use :func:`fdopen`.
 
 
@@ -1049,10 +1049,10 @@
    Availability: Unix.
 
 
-.. function:: listdir(path)
+.. function:: listdir(path='.')
 
    Return a list containing the names of the entries in the directory given by
-   *path*.  The list is in arbitrary order.  It does not include the special
+   *path* (default: ``'.'``).  The list is in arbitrary order.  It does not include the special
    entries ``'.'`` and ``'..'`` even if they are present in the directory.
 
    This function can be called with a bytes or string argument, and returns
@@ -1060,12 +1060,16 @@
 
    Availability: Unix, Windows.
 
+   .. versionchanged:: 3.2
+      The *path* parameter became optional.
 
 .. function:: lstat(path)
 
    Like :func:`stat`, but do not follow symbolic links.  This is an alias for
-   :func:`stat` on platforms that do not support symbolic links, such as
-   Windows prior to 6.0 (Vista).
+   :func:`stat` on platforms that do not support symbolic links.
+
+   .. versionchanged:: 3.2
+      Added support for Windows 6.0 (Vista) symbolic links.
 
 
 .. function:: mkfifo(path[, mode])
@@ -1181,7 +1185,10 @@
    and the call may raise an UnicodeDecodeError. If the *path* is a bytes
    object, the result will be a bytes object.
 
-   Availability: Unix, Windows.
+   Availability: Unix, Windows
+
+   .. versionchanged:: 3.2
+      Added support for Windows 6.0 (Vista) symbolic links.
 
 
 .. function:: remove(path)
@@ -1341,8 +1348,9 @@
 
 .. function:: symlink(source, link_name)
 
-   Create a symbolic link pointing to *source* named *link_name*.  On Windows,
-   symlink version takes an additional, optional parameter,
+   Create a symbolic link pointing to *source* named *link_name*.
+
+   On Windows, symlink version takes an additional, optional parameter,
    *target_is_directory*, which defaults to False.
 
    symlink(source, link_name, target_is_directory=False)
@@ -1359,7 +1367,10 @@
    will raise a NotImplementedError on Windows versions earlier than 6.0. The
    SeCreateSymbolicLinkPrivilege is required in order to create symlinks.
 
-   Availability:  Unix, Windows 6.0.
+   Availability:  Unix, Windows
+
+   .. versionchanged:: 3.2
+      Added support for Windows 6.0 (Vista) symbolic links.
 
 
 .. function:: unlink(path)
@@ -1912,8 +1923,9 @@
 
    The :mod:`subprocess` module provides more powerful facilities for spawning new
    processes and retrieving their results; using that module is preferable to using
-   this function.  Use the :mod:`subprocess` module.  Check especially the
-   :ref:`subprocess-replacements` section.
+   this function.  See the
+   :ref:`subprocess-replacements` section in the :mod:`subprocess` documentation
+   for some helpful recipes.
 
    Availability: Unix, Windows.
 

Modified: python/branches/import_unicode/Doc/library/pdb.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/pdb.rst	(original)
+++ python/branches/import_unicode/Doc/library/pdb.rst	Fri Jul 30 01:19:37 2010
@@ -20,7 +20,7 @@
    module: bdb
    module: cmd
 
-The debugger is extensible --- it is actually defined as the class :class:`Pdb`.
+The debugger is extensible -- it is actually defined as the class :class:`Pdb`.
 This is currently undocumented but easily understood by reading the source.  The
 extension interface uses the modules :mod:`bdb` and :mod:`cmd`.
 
@@ -55,8 +55,8 @@
    import pdb; pdb.set_trace()
 
 at the location you want to break into the debugger.  You can then step through
-the code following this statement, and continue running without the debugger using
-the ``c`` command.
+the code following this statement, and continue running without the debugger
+using the :pdbcmd:`continue` command.
 
 The typical usage to inspect a crashed program is::
 
@@ -82,26 +82,28 @@
 .. function:: run(statement, globals=None, locals=None)
 
    Execute the *statement* (given as a string) under debugger control.  The
-   debugger prompt appears before any code is executed; you can set breakpoints and
-   type ``continue``, or you can step through the statement using ``step`` or
-   ``next`` (all these commands are explained below).  The optional *globals* and
-   *locals* arguments specify the environment in which the code is executed; by
-   default the dictionary of the module :mod:`__main__` is used.  (See the
-   explanation of the built-in :func:`exec` or :func:`eval` functions.)
+   debugger prompt appears before any code is executed; you can set breakpoints
+   and type :pdbcmd:`continue`, or you can step through the statement using
+   :pdbcmd:`step` or :pdbcmd:`next` (all these commands are explained below).
+   The optional *globals* and *locals* arguments specify the environment in
+   which the code is executed; by default the dictionary of the module
+   :mod:`__main__` is used.  (See the explanation of the built-in :func:`exec`
+   or :func:`eval` functions.)
 
 
 .. function:: runeval(expression, globals=None, locals=None)
 
    Evaluate the *expression* (given as a string) under debugger control.  When
-   :func:`runeval` returns, it returns the value of the expression.  Otherwise this
-   function is similar to :func:`run`.
+   :func:`runeval` returns, it returns the value of the expression.  Otherwise
+   this function is similar to :func:`run`.
 
 
 .. function:: runcall(function, *args, **kwds)
 
-   Call the *function* (a function or method object, not a string) with the given
-   arguments.  When :func:`runcall` returns, it returns whatever the function call
-   returned.  The debugger prompt appears as soon as the function is entered.
+   Call the *function* (a function or method object, not a string) with the
+   given arguments.  When :func:`runcall` returns, it returns whatever the
+   function call returned.  The debugger prompt appears as soon as the function
+   is entered.
 
 
 .. function:: set_trace()
@@ -160,16 +162,17 @@
 Debugger Commands
 =================
 
-The debugger recognizes the following commands.  Most commands can be
-abbreviated to one or two letters; e.g. ``h(elp)`` means that either ``h`` or
-``help`` can be used to enter the help command (but not ``he`` or ``hel``, nor
-``H`` or ``Help`` or ``HELP``).  Arguments to commands must be separated by
-whitespace (spaces or tabs).  Optional arguments are enclosed in square brackets
-(``[]``) in the command syntax; the square brackets must not be typed.
-Alternatives in the command syntax are separated by a vertical bar (``|``).
+The commands recognized by the debugger are listed below.  Most commands can be
+abbreviated to one or two letters as indicated; e.g. ``h(elp)`` means that
+either ``h`` or ``help`` can be used to enter the help command (but not ``he``
+or ``hel``, nor ``H`` or ``Help`` or ``HELP``).  Arguments to commands must be
+separated by whitespace (spaces or tabs).  Optional arguments are enclosed in
+square brackets (``[]``) in the command syntax; the square brackets must not be
+typed.  Alternatives in the command syntax are separated by a vertical bar
+(``|``).
 
 Entering a blank line repeats the last command entered.  Exception: if the last
-command was a ``list`` command, the next 11 lines are listed.
+command was a :pdbcmd:`list` command, the next 11 lines are listed.
 
 Commands that the debugger doesn't recognize are assumed to be Python statements
 and are executed in the context of the program being debugged.  Python
@@ -179,93 +182,106 @@
 statement, the exception name is printed but the debugger's state is not
 changed.
 
+The debugger supports :ref:`aliases <debugger-aliases>`.  Aliases can have
+parameters which allows one a certain level of adaptability to the context under
+examination.
+
 Multiple commands may be entered on a single line, separated by ``;;``.  (A
 single ``;`` is not used as it is the separator for multiple commands in a line
-that is passed to the Python parser.) No intelligence is applied to separating
+that is passed to the Python parser.)  No intelligence is applied to separating
 the commands; the input is split at the first ``;;`` pair, even if it is in the
 middle of a quoted string.
 
-The debugger supports aliases.  Aliases can have parameters which allows one a
-certain level of adaptability to the context under examination.
-
 .. index::
    pair: .pdbrc; file
    triple: debugger; configuration; file
 
-If a file :file:`.pdbrc`  exists in the user's home directory or in the current
+If a file :file:`.pdbrc` exists in the user's home directory or in the current
 directory, it is read in and executed as if it had been typed at the debugger
-prompt. This is particularly useful for aliases.  If both files exist, the one
+prompt.  This is particularly useful for aliases.  If both files exist, the one
 in the home directory is read first and aliases defined there can be overridden
 by the local file.
 
-h(elp) [*command*]
+
+.. pdbcommand:: h(elp) [command]
+
    Without argument, print the list of available commands.  With a *command* as
    argument, print help about that command.  ``help pdb`` displays the full
-   documentation file; if the environment variable :envvar:`PAGER` is defined, the
-   file is piped through that command instead.  Since the *command* argument must
-   be an identifier, ``help exec`` must be entered to get help on the ``!``
-   command.
+   documentation (the docstring of the :mod:`pdb` module).  Since the *command*
+   argument must be an identifier, ``help exec`` must be entered to get help on
+   the ``!`` command.
+
+.. pdbcommand:: w(here)
 
-w(here)
    Print a stack trace, with the most recent frame at the bottom.  An arrow
    indicates the current frame, which determines the context of most commands.
 
-d(own) [*count*]
+.. pdbcommand:: d(own) [count]
+
    Move the current frame *count* (default one) levels down in the stack trace
    (to a newer frame).
 
-u(p) [*count*]
-   Move the current frame *count* (default one) levels up in the stack trace
-   (to an older frame).
+.. pdbcommand:: u(p) [count]
+
+   Move the current frame *count* (default one) levels up in the stack trace (to
+   an older frame).
+
+.. pdbcommand:: b(reak) [([filename:]lineno | function) [, condition]]
 
-b(reak) [[*filename*:]\ *lineno* | *function*\ [, *condition*]]
    With a *lineno* argument, set a break there in the current file.  With a
-   *function* argument, set a break at the first executable statement within that
-   function. The line number may be prefixed with a filename and a colon, to
-   specify a breakpoint in another file (probably one that hasn't been loaded yet).
-   The file is searched on ``sys.path``. Note that each breakpoint is assigned a
-   number to which all the other breakpoint commands refer.
-
-   If a second argument is present, it is an expression which must evaluate to true
-   before the breakpoint is honored.
-
-   Without argument, list all breaks, including for each breakpoint, the number of
-   times that breakpoint has been hit, the current ignore count, and the associated
-   condition if any.
-
-tbreak [[*filename*:]\ *lineno* | *function*\ [, *condition*]]
-   Temporary breakpoint, which is removed automatically when it is first hit.  The
-   arguments are the same as break.
+   *function* argument, set a break at the first executable statement within
+   that function.  The line number may be prefixed with a filename and a colon,
+   to specify a breakpoint in another file (probably one that hasn't been loaded
+   yet).  The file is searched on :data:`sys.path`.  Note that each breakpoint
+   is assigned a number to which all the other breakpoint commands refer.
+
+   If a second argument is present, it is an expression which must evaluate to
+   true before the breakpoint is honored.
+
+   Without argument, list all breaks, including for each breakpoint, the number
+   of times that breakpoint has been hit, the current ignore count, and the
+   associated condition if any.
+
+.. pdbcommand:: tbreak [([filename:]lineno | function) [, condition]]
+
+   Temporary breakpoint, which is removed automatically when it is first hit.
+   The arguments are the same as for :pdbcmd:`break`.
+
+.. pdbcommand:: cl(ear) [bpnumber [bpnumber ...]]
 
-cl(ear) [*bpnumber* [*bpnumber ...*]]
    With a space separated list of breakpoint numbers, clear those breakpoints.
    Without argument, clear all breaks (but first ask confirmation).
 
-disable [*bpnumber* [*bpnumber ...*]]
-   Disables the breakpoints given as a space separated list of breakpoint numbers.
-   Disabling a breakpoint means it cannot cause the program to stop execution, but
-   unlike clearing a breakpoint, it remains in the list of breakpoints and can be
-   (re-)enabled.
-
-enable [*bpnumber* [*bpnumber ...*]]
-   Enables the breakpoints specified.
-
-ignore *bpnumber* [*count*]
-   Sets the ignore count for the given breakpoint number.  If count is omitted, the
-   ignore count is set to 0.  A breakpoint becomes active when the ignore count is
-   zero.  When non-zero, the count is decremented each time the breakpoint is
-   reached and the breakpoint is not disabled and any associated condition
-   evaluates to true.
-
-condition *bpnumber* [*condition*]
-   Condition is an expression which must evaluate to true before the breakpoint is
-   honored.  If condition is absent, any existing condition is removed; i.e., the
-   breakpoint is made unconditional.
+.. pdbcommand:: disable [bpnumber [bpnumber ...]]
+
+   Disable the breakpoints given as a space separated list of breakpoint
+   numbers.  Disabling a breakpoint means it cannot cause the program to stop
+   execution, but unlike clearing a breakpoint, it remains in the list of
+   breakpoints and can be (re-)enabled.
+
+.. pdbcommand:: enable [bpnumber [bpnumber ...]]
+
+   Enable the breakpoints specified.
+
+.. pdbcommand:: ignore bpnumber [count]
+
+   Set the ignore count for the given breakpoint number.  If count is omitted,
+   the ignore count is set to 0.  A breakpoint becomes active when the ignore
+   count is zero.  When non-zero, the count is decremented each time the
+   breakpoint is reached and the breakpoint is not disabled and any associated
+   condition evaluates to true.
+
+.. pdbcommand:: condition bpnumber [condition]
+
+   Set a new *condition* for the breakpoint, an expression which must evaluate
+   to true before the breakpoint is honored.  If *condition* is absent, any
+   existing condition is removed; i.e., the breakpoint is made unconditional.
+
+.. pdbcommand:: commands [bpnumber]
 
-commands [*bpnumber*]
    Specify a list of commands for breakpoint number *bpnumber*.  The commands
-   themselves appear on the following lines.  Type a line containing just 'end' to
-   terminate the commands. An example::
+   themselves appear on the following lines.  Type a line containing just
+   ``end`` to terminate the commands. An example::
 
       (Pdb) commands 1
       (com) print some_variable
@@ -273,12 +289,12 @@
       (Pdb)
 
    To remove all commands from a breakpoint, type commands and follow it
-   immediately with  end; that is, give no commands.
+   immediately with ``end``; that is, give no commands.
 
    With no *bpnumber* argument, commands refers to the last breakpoint set.
 
-   You can use breakpoint commands to start your program up again. Simply use the
-   continue command, or step, or any other command that resumes execution.
+   You can use breakpoint commands to start your program up again.  Simply use
+   the continue command, or step, or any other command that resumes execution.
 
    Specifying any command resuming execution (currently continue, step, next,
    return, jump, quit and their abbreviations) terminates the command list (as if
@@ -292,63 +308,82 @@
    that are to print a specific message and then continue.  If none of the other
    commands print anything, you see no sign that the breakpoint was reached.
 
-s(tep)
+.. pdbcommand:: s(tep)
+
    Execute the current line, stop at the first possible occasion (either in a
    function that is called or on the next line in the current function).
 
-n(ext)
-   Continue execution until the next line in the current function is reached or it
-   returns.  (The difference between ``next`` and ``step`` is that ``step`` stops
-   inside a called function, while ``next`` executes called functions at (nearly)
-   full speed, only stopping at the next line in the current function.)
+.. pdbcommand:: n(ext)
+
+   Continue execution until the next line in the current function is reached or
+   it returns.  (The difference between :pdbcmd:`next` and :pdbcmd:`step` is
+   that :pdbcmd:`step` stops inside a called function, while :pdbcmd:`next`
+   executes called functions at (nearly) full speed, only stopping at the next
+   line in the current function.)
+
+.. pdbcommand:: unt(il)
 
-unt(il)
    Continue execution until the line with the line number greater than the
    current one is reached or when returning from current frame.
 
-r(eturn)
+.. pdbcommand:: r(eturn)
+
    Continue execution until the current function returns.
 
-c(ont(inue))
+.. pdbcommand:: c(ont(inue))
+
    Continue execution, only stop when a breakpoint is encountered.
 
-j(ump) *lineno*
+.. pdbcommand:: j(ump) lineno
+
    Set the next line that will be executed.  Only available in the bottom-most
-   frame.  This lets you jump back and execute code again, or jump forward to skip
-   code that you don't want to run.
+   frame.  This lets you jump back and execute code again, or jump forward to
+   skip code that you don't want to run.
 
-   It should be noted that not all jumps are allowed --- for instance it is not
+   It should be noted that not all jumps are allowed -- for instance it is not
    possible to jump into the middle of a :keyword:`for` loop or out of a
    :keyword:`finally` clause.
 
-l(ist) [*first*\ [, *last*]]
-   List source code for the current file.  Without arguments, list 11 lines around
-   the current line or continue the previous listing.  With one argument, list 11
-   lines around at that line.  With two arguments, list the given range; if the
-   second argument is less than the first, it is interpreted as a count.
+.. pdbcommand:: l(ist) [first[, last]]
+
+   List source code for the current file.  Without arguments, list 11 lines
+   around the current line or continue the previous listing.  With ``.`` as
+   argument, list 11 lines around the current line.  With one argument,
+   list 11 lines around at that line.  With two arguments, list the given range;
+   if the second argument is less than the first, it is interpreted as a count.
+
+.. pdbcommand:: a(rgs)
 
-a(rgs)
    Print the argument list of the current function.
 
-p(rint) *expression*
+.. pdbcommand:: p(rint) expression
+
    Evaluate the *expression* in the current context and print its value.
 
-pp *expression*
-   Like the ``p`` command, except the value of the expression is pretty-printed
-   using the :mod:`pprint` module.
-
-alias [*name* [command]]
-   Creates an alias called *name* that executes *command*.  The command must *not*
-   be enclosed in quotes.  Replaceable parameters can be indicated by ``%1``,
-   ``%2``, and so on, while ``%*`` is replaced by all the parameters.  If no
-   command is given, the current alias for *name* is shown. If no arguments are
-   given, all aliases are listed.
-
-   Aliases may be nested and can contain anything that can be legally typed at the
-   pdb prompt.  Note that internal pdb commands *can* be overridden by aliases.
-   Such a command is then hidden until the alias is removed.  Aliasing is
-   recursively applied to the first word of the command line; all other words in
-   the line are left alone.
+.. pdbcommand:: pp expression
+
+   Like the :pdbcmd:`print` command, except the value of the expression is
+   pretty-printed using the :mod:`pprint` module.
+
+.. pdbcommand:: whatis expression
+
+   Print the type of the *expression*.
+
+.. _debugger-aliases:
+
+.. pdbcommand:: alias [name [command]]
+
+   Create an alias called *name* that executes *command*.  The command must
+   *not* be enclosed in quotes.  Replaceable parameters can be indicated by
+   ``%1``, ``%2``, and so on, while ``%*`` is replaced by all the parameters.
+   If no command is given, the current alias for *name* is shown. If no
+   arguments are given, all aliases are listed.
+
+   Aliases may be nested and can contain anything that can be legally typed at
+   the pdb prompt.  Note that internal pdb commands *can* be overridden by
+   aliases.  Such a command is then hidden until the alias is removed.  Aliasing
+   is recursively applied to the first word of the command line; all other words
+   in the line are left alone.
 
    As an example, here are two useful aliases (especially when placed in the
    :file:`.pdbrc` file)::
@@ -358,25 +393,32 @@
       #Print instance variables in self
       alias ps pi self
 
-unalias *name*
-   Deletes the specified alias.
+.. pdbcommand:: unalias name
+
+   Delete the specified alias.
+
+.. pdbcommand:: ! statement
 
-[!]\ *statement*
    Execute the (one-line) *statement* in the context of the current stack frame.
    The exclamation point can be omitted unless the first word of the statement
-   resembles a debugger command. To set a global variable, you can prefix the
-   assignment command with a ``global`` command on the same line, e.g.::
+   resembles a debugger command.  To set a global variable, you can prefix the
+   assignment command with a :keyword:`global` statement on the same line,
+   e.g.::
 
       (Pdb) global list_options; list_options = ['-l']
       (Pdb)
 
-run [*args* ...]
-   Restart the debugged Python program. If an argument is supplied, it is split
-   with "shlex" and the result is used as the new sys.argv. History, breakpoints,
-   actions and debugger options are preserved. "restart" is an alias for "run".
+.. pdbcommand:: run [args ...]
+                restart [args ...]
+
+   Restart the debugged Python program.  If an argument is supplied, it is split
+   with :mod:`shlex` and the result is used as the new :data:`sys.argv`.
+   History, breakpoints, actions and debugger options are preserved.
+   :pdbcmd:`restart` is an alias for :pdbcmd:`run`.
+
+.. pdbcommand:: q(uit)
 
-q(uit)
-   Quit from the debugger. The program being executed is aborted.
+   Quit from the debugger.  The program being executed is aborted.
 
 
 .. rubric:: Footnotes

Modified: python/branches/import_unicode/Doc/library/pickletools.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/pickletools.rst	(original)
+++ python/branches/import_unicode/Doc/library/pickletools.rst	Fri Jul 30 01:19:37 2010
@@ -13,14 +13,20 @@
 :mod:`pickletools` module relevant.
 
 
-.. function:: dis(pickle, out=None, memo=None, indentlevel=4)
+.. function:: dis(pickle, out=None, memo=None, indentlevel=4, annotate=0)
 
-   Outputs a symbolic disassembly of the pickle to the file-like object *out*,
-   defaulting to ``sys.stdout``.  *pickle* can be a string or a file-like object.
-   *memo* can be a Python dictionary that will be used as the pickle's memo; it can
-   be used to perform disassemblies across multiple pickles created by the same
-   pickler. Successive levels, indicated by ``MARK`` opcodes in the stream, are
-   indented by *indentlevel* spaces.
+   Outputs a symbolic disassembly of the pickle to the file-like
+   object *out*, defaulting to ``sys.stdout``.  *pickle* can be a
+   string or a file-like object.  *memo* can be a Python dictionary
+   that will be used as the pickle's memo; it can be used to perform
+   disassemblies across multiple pickles created by the same
+   pickler. Successive levels, indicated by ``MARK`` opcodes in the
+   stream, are indented by *indentlevel* spaces.  If a nonzero value
+   is given to *annotate*, each opcode in the output is annotated with
+   a short description.  The value of *annotate* is used as a hint for
+   the column where annotation should start.
+
+  .. versionadded:: 3.2 the *annotate* argument.
 
 .. function:: genops(pickle)
 

Modified: python/branches/import_unicode/Doc/library/re.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/re.rst	(original)
+++ python/branches/import_unicode/Doc/library/re.rst	Fri Jul 30 01:19:37 2010
@@ -33,8 +33,9 @@
 string notation.
 
 It is important to note that most regular expression operations are available as
-module-level functions and :class:`RegexObject` methods.  The functions are
-shortcuts that don't require you to compile a regex object first, but miss some
+module-level functions and methods on
+:ref:`compiled regular expressions <re-objects>`.  The functions are shortcuts
+that don't require you to compile a regex object first, but miss some
 fine-tuning parameters.
 
 .. seealso::
@@ -485,7 +486,7 @@
 
    Note that for backward compatibility, the :const:`re.U` flag still
    exists (as well as its synonym :const:`re.UNICODE` and its embedded
-   counterpart ``(?u)``), but these are redundant in Python 3.0 since
+   counterpart ``(?u)``), but these are redundant in Python 3 since
    matches are Unicode by default for strings (and Unicode matching
    isn't allowed for bytes).
 
@@ -504,7 +505,7 @@
    Make ``\w``, ``\W``, ``\b``, ``\B``, ``\s`` and ``\S`` dependent on the
    current locale. The use of this flag is discouraged as the locale mechanism
    is very unreliable, and it only handles one "culture" at a time anyway;
-   you should use Unicode matching instead, which is the default in Python 3.0
+   you should use Unicode matching instead, which is the default in Python 3
    for Unicode (str) patterns.
 
 
@@ -546,21 +547,21 @@
 
 
 
-.. function:: search(pattern, string[, flags])
+.. function:: search(pattern, string, flags=0)
 
    Scan through *string* looking for a location where the regular expression
-   *pattern* produces a match, and return a corresponding :class:`MatchObject`
-   instance. Return ``None`` if no position in the string matches the pattern; note
-   that this is different from finding a zero-length match at some point in the
-   string.
+   *pattern* produces a match, and return a corresponding :ref:`match object
+   <match-objects>`.  Return ``None`` if no position in the string matches the
+   pattern; note that this is different from finding a zero-length match at some
+   point in the string.
 
 
 .. function:: match(pattern, string, flags=0)
 
    If zero or more characters at the beginning of *string* match the regular
-   expression *pattern*, return a corresponding :class:`MatchObject` instance.
-   Return ``None`` if the string does not match the pattern; note that this is
-   different from a zero-length match.
+   expression *pattern*, return a corresponding :ref:`match object
+   <match-objects>`.  Return ``None`` if the string does not match the pattern;
+   note that this is different from a zero-length match.
 
    .. note::
 
@@ -620,9 +621,9 @@
 
 .. function:: finditer(pattern, string, flags=0)
 
-   Return an :term:`iterator` yielding :class:`MatchObject` instances over all
-   non-overlapping matches for the RE *pattern* in *string*.  The *string* is
-   scanned left-to-right, and matches are returned in the order found.  Empty
+   Return an :term:`iterator` yielding :ref:`match objects <match-objects>` over
+   all non-overlapping matches for the RE *pattern* in *string*.  The *string*
+   is scanned left-to-right, and matches are returned in the order found.  Empty
    matches are included in the result unless they touch the beginning of another
    match.
 
@@ -692,6 +693,11 @@
    metacharacters in it.
 
 
+.. function:: purge()
+
+   Clear the regular expression cache.
+
+
 .. exception:: error
 
    Exception raised when a string passed to one of the functions here is not a
@@ -705,107 +711,107 @@
 Regular Expression Objects
 --------------------------
 
-.. class:: RegexObject
-
-   The :class:`RegexObject` class supports the following methods and attributes:
-
-   .. method:: RegexObject.search(string[, pos[, endpos]])
+Compiled regular expression objects support the following methods and
+attributes.
 
-      Scan through *string* looking for a location where this regular expression
-      produces a match, and return a corresponding :class:`MatchObject` instance.
-      Return ``None`` if no position in the string matches the pattern; note that this
-      is different from finding a zero-length match at some point in the string.
+.. method:: regex.search(string[, pos[, endpos]])
 
-      The optional second parameter *pos* gives an index in the string where the
-      search is to start; it defaults to ``0``.  This is not completely equivalent to
-      slicing the string; the ``'^'`` pattern character matches at the real beginning
-      of the string and at positions just after a newline, but not necessarily at the
-      index where the search is to start.
+   Scan through *string* looking for a location where this regular expression
+   produces a match, and return a corresponding :ref:`match object
+   <match-objects>`.  Return ``None`` if no position in the string matches the
+   pattern; note that this is different from finding a zero-length match at some
+   point in the string.
+
+   The optional second parameter *pos* gives an index in the string where the
+   search is to start; it defaults to ``0``.  This is not completely equivalent to
+   slicing the string; the ``'^'`` pattern character matches at the real beginning
+   of the string and at positions just after a newline, but not necessarily at the
+   index where the search is to start.
+
+   The optional parameter *endpos* limits how far the string will be searched; it
+   will be as if the string is *endpos* characters long, so only the characters
+   from *pos* to ``endpos - 1`` will be searched for a match.  If *endpos* is less
+   than *pos*, no match will be found, otherwise, if *rx* is a compiled regular
+   expression object, ``rx.search(string, 0, 50)`` is equivalent to
+   ``rx.search(string[:50], 0)``.
 
-      The optional parameter *endpos* limits how far the string will be searched; it
-      will be as if the string is *endpos* characters long, so only the characters
-      from *pos* to ``endpos - 1`` will be searched for a match.  If *endpos* is less
-      than *pos*, no match will be found, otherwise, if *rx* is a compiled regular
-      expression object, ``rx.search(string, 0, 50)`` is equivalent to
-      ``rx.search(string[:50], 0)``.
-
-      >>> pattern = re.compile("d")
-      >>> pattern.search("dog")     # Match at index 0
-      <_sre.SRE_Match object at ...>
-      >>> pattern.search("dog", 1)  # No match; search doesn't include the "d"
+   >>> pattern = re.compile("d")
+   >>> pattern.search("dog")     # Match at index 0
+   <_sre.SRE_Match object at ...>
+   >>> pattern.search("dog", 1)  # No match; search doesn't include the "d"
 
 
-   .. method:: RegexObject.match(string[, pos[, endpos]])
+.. method:: regex.match(string[, pos[, endpos]])
 
-      If zero or more characters at the *beginning* of *string* match this regular
-      expression, return a corresponding :class:`MatchObject` instance.  Return
-      ``None`` if the string does not match the pattern; note that this is different
-      from a zero-length match.
+   If zero or more characters at the *beginning* of *string* match this regular
+   expression, return a corresponding :ref:`match object <match-objects>`.
+   Return ``None`` if the string does not match the pattern; note that this is
+   different from a zero-length match.
 
-      The optional *pos* and *endpos* parameters have the same meaning as for the
-      :meth:`~RegexObject.search` method.
+   The optional *pos* and *endpos* parameters have the same meaning as for the
+   :meth:`~regex.search` method.
 
-      .. note::
+   .. note::
 
-         If you want to locate a match anywhere in *string*, use
-         :meth:`~RegexObject.search` instead.
+      If you want to locate a match anywhere in *string*, use
+      :meth:`~regex.search` instead.
 
-      >>> pattern = re.compile("o")
-      >>> pattern.match("dog")      # No match as "o" is not at the start of "dog".
-      >>> pattern.match("dog", 1)   # Match as "o" is the 2nd character of "dog".
-      <_sre.SRE_Match object at ...>
+   >>> pattern = re.compile("o")
+   >>> pattern.match("dog")      # No match as "o" is not at the start of "dog".
+   >>> pattern.match("dog", 1)   # Match as "o" is the 2nd character of "dog".
+   <_sre.SRE_Match object at ...>
 
 
-   .. method:: RegexObject.split(string[, maxsplit=0])
+.. method:: regex.split(string, maxsplit=0)
 
-      Identical to the :func:`split` function, using the compiled pattern.
+   Identical to the :func:`split` function, using the compiled pattern.
 
 
-   .. method:: RegexObject.findall(string[, pos[, endpos]])
+.. method:: regex.findall(string[, pos[, endpos]])
 
-      Similar to the :func:`findall` function, using the compiled pattern, but
-      also accepts optional *pos* and *endpos* parameters that limit the search
-      region like for :meth:`match`.
+   Similar to the :func:`findall` function, using the compiled pattern, but
+   also accepts optional *pos* and *endpos* parameters that limit the search
+   region like for :meth:`match`.
 
 
-   .. method:: RegexObject.finditer(string[, pos[, endpos]])
+.. method:: regex.finditer(string[, pos[, endpos]])
 
-      Similar to the :func:`finditer` function, using the compiled pattern, but
-      also accepts optional *pos* and *endpos* parameters that limit the search
-      region like for :meth:`match`.
+   Similar to the :func:`finditer` function, using the compiled pattern, but
+   also accepts optional *pos* and *endpos* parameters that limit the search
+   region like for :meth:`match`.
 
 
-   .. method:: RegexObject.sub(repl, string[, count=0])
+.. method:: regex.sub(repl, string, count=0)
 
-      Identical to the :func:`sub` function, using the compiled pattern.
+   Identical to the :func:`sub` function, using the compiled pattern.
 
 
-   .. method:: RegexObject.subn(repl, string[, count=0])
+.. method:: regex.subn(repl, string, count=0)
 
-      Identical to the :func:`subn` function, using the compiled pattern.
+   Identical to the :func:`subn` function, using the compiled pattern.
 
 
-   .. attribute:: RegexObject.flags
+.. attribute:: regex.flags
 
-      The flags argument used when the RE object was compiled, or ``0`` if no flags
-      were provided.
+   The flags argument used when the RE object was compiled, or ``0`` if no flags
+   were provided.
 
 
-   .. attribute:: RegexObject.groups
+.. attribute:: regex.groups
 
-      The number of capturing groups in the pattern.
+   The number of capturing groups in the pattern.
 
 
-   .. attribute:: RegexObject.groupindex
+.. attribute:: regex.groupindex
 
-      A dictionary mapping any symbolic group names defined by ``(?P<id>)`` to group
-      numbers.  The dictionary is empty if no symbolic groups were used in the
-      pattern.
+   A dictionary mapping any symbolic group names defined by ``(?P<id>)`` to group
+   numbers.  The dictionary is empty if no symbolic groups were used in the
+   pattern.
 
 
-   .. attribute:: RegexObject.pattern
+.. attribute:: regex.pattern
 
-      The pattern string from which the RE object was compiled.
+   The pattern string from which the RE object was compiled.
 
 
 .. _match-objects:
@@ -813,181 +819,176 @@
 Match Objects
 -------------
 
-.. class:: MatchObject
-
-   Match Objects always have a boolean value of :const:`True`, so that you can test
-   whether e.g. :func:`match` resulted in a match with a simple if statement.  They
-   support the following methods and attributes:
-
-
-   .. method:: MatchObject.expand(template)
+Match objects always have a boolean value of :const:`True`, so that you can test
+whether e.g. :func:`match` resulted in a match with a simple if statement.  They
+support the following methods and attributes:
+
+
+.. method:: match.expand(template)
+
+   Return the string obtained by doing backslash substitution on the template
+   string *template*, as done by the :meth:`~regex.sub` method.
+   Escapes such as ``\n`` are converted to the appropriate characters,
+   and numeric backreferences (``\1``, ``\2``) and named backreferences
+   (``\g<1>``, ``\g<name>``) are replaced by the contents of the
+   corresponding group.
+
+
+.. method:: match.group([group1, ...])
+
+   Returns one or more subgroups of the match.  If there is a single argument, the
+   result is a single string; if there are multiple arguments, the result is a
+   tuple with one item per argument. Without arguments, *group1* defaults to zero
+   (the whole match is returned). If a *groupN* argument is zero, the corresponding
+   return value is the entire matching string; if it is in the inclusive range
+   [1..99], it is the string matching the corresponding parenthesized group.  If a
+   group number is negative or larger than the number of groups defined in the
+   pattern, an :exc:`IndexError` exception is raised. If a group is contained in a
+   part of the pattern that did not match, the corresponding result is ``None``.
+   If a group is contained in a part of the pattern that matched multiple times,
+   the last match is returned.
+
+      >>> m = re.match(r"(\w+) (\w+)", "Isaac Newton, physicist")
+      >>> m.group(0)       # The entire match
+      'Isaac Newton'
+      >>> m.group(1)       # The first parenthesized subgroup.
+      'Isaac'
+      >>> m.group(2)       # The second parenthesized subgroup.
+      'Newton'
+      >>> m.group(1, 2)    # Multiple arguments give us a tuple.
+      ('Isaac', 'Newton')
+
+   If the regular expression uses the ``(?P<name>...)`` syntax, the *groupN*
+   arguments may also be strings identifying groups by their group name.  If a
+   string argument is not used as a group name in the pattern, an :exc:`IndexError`
+   exception is raised.
+
+   A moderately complicated example:
+
+      >>> m = re.match(r"(?P<first_name>\w+) (?P<last_name>\w+)", "Malcolm Reynolds")
+      >>> m.group('first_name')
+      'Malcolm'
+      >>> m.group('last_name')
+      'Reynolds'
+
+   Named groups can also be referred to by their index:
+
+      >>> m.group(1)
+      'Malcolm'
+      >>> m.group(2)
+      'Reynolds'
+
+   If a group matches multiple times, only the last match is accessible:
+
+      >>> m = re.match(r"(..)+", "a1b2c3")  # Matches 3 times.
+      >>> m.group(1)                        # Returns only the last match.
+      'c3'
+
+
+.. method:: match.groups(default=None)
+
+   Return a tuple containing all the subgroups of the match, from 1 up to however
+   many groups are in the pattern.  The *default* argument is used for groups that
+   did not participate in the match; it defaults to ``None``.
 
-      Return the string obtained by doing backslash substitution on the template
-      string *template*, as done by the :meth:`~RegexObject.sub` method.  Escapes
-      such as ``\n`` are converted to the appropriate characters, and numeric
-      backreferences (``\1``, ``\2``) and named backreferences (``\g<1>``,
-      ``\g<name>``) are replaced by the contents of the corresponding group.
-
-
-   .. method:: MatchObject.group([group1, ...])
-
-      Returns one or more subgroups of the match.  If there is a single argument, the
-      result is a single string; if there are multiple arguments, the result is a
-      tuple with one item per argument. Without arguments, *group1* defaults to zero
-      (the whole match is returned). If a *groupN* argument is zero, the corresponding
-      return value is the entire matching string; if it is in the inclusive range
-      [1..99], it is the string matching the corresponding parenthesized group.  If a
-      group number is negative or larger than the number of groups defined in the
-      pattern, an :exc:`IndexError` exception is raised. If a group is contained in a
-      part of the pattern that did not match, the corresponding result is ``None``.
-      If a group is contained in a part of the pattern that matched multiple times,
-      the last match is returned.
-
-         >>> m = re.match(r"(\w+) (\w+)", "Isaac Newton, physicist")
-         >>> m.group(0)       # The entire match
-         'Isaac Newton'
-         >>> m.group(1)       # The first parenthesized subgroup.
-         'Isaac'
-         >>> m.group(2)       # The second parenthesized subgroup.
-         'Newton'
-         >>> m.group(1, 2)    # Multiple arguments give us a tuple.
-         ('Isaac', 'Newton')
-
-      If the regular expression uses the ``(?P<name>...)`` syntax, the *groupN*
-      arguments may also be strings identifying groups by their group name.  If a
-      string argument is not used as a group name in the pattern, an :exc:`IndexError`
-      exception is raised.
-
-      A moderately complicated example:
-
-         >>> m = re.match(r"(?P<first_name>\w+) (?P<last_name>\w+)", "Malcolm Reynolds")
-         >>> m.group('first_name')
-         'Malcolm'
-         >>> m.group('last_name')
-         'Reynolds'
-
-      Named groups can also be referred to by their index:
-
-         >>> m.group(1)
-         'Malcolm'
-         >>> m.group(2)
-         'Reynolds'
-
-      If a group matches multiple times, only the last match is accessible:
-
-         >>> m = re.match(r"(..)+", "a1b2c3")  # Matches 3 times.
-         >>> m.group(1)                        # Returns only the last match.
-         'c3'
-
-
-   .. method:: MatchObject.groups(default=None)
-
-      Return a tuple containing all the subgroups of the match, from 1 up to however
-      many groups are in the pattern.  The *default* argument is used for groups that
-      did not participate in the match; it defaults to ``None``.  (Incompatibility
-      note: in the original Python 1.5 release, if the tuple was one element long, a
-      string would be returned instead.  In later versions (from 1.5.1 on), a
-      singleton tuple is returned in such cases.)
-
-      For example:
+   For example:
 
-         >>> m = re.match(r"(\d+)\.(\d+)", "24.1632")
-         >>> m.groups()
-         ('24', '1632')
+      >>> m = re.match(r"(\d+)\.(\d+)", "24.1632")
+      >>> m.groups()
+      ('24', '1632')
 
-      If we make the decimal place and everything after it optional, not all groups
-      might participate in the match.  These groups will default to ``None`` unless
-      the *default* argument is given:
+   If we make the decimal place and everything after it optional, not all groups
+   might participate in the match.  These groups will default to ``None`` unless
+   the *default* argument is given:
 
-         >>> m = re.match(r"(\d+)\.?(\d+)?", "24")
-         >>> m.groups()      # Second group defaults to None.
-         ('24', None)
-         >>> m.groups('0')   # Now, the second group defaults to '0'.
-         ('24', '0')
+      >>> m = re.match(r"(\d+)\.?(\d+)?", "24")
+      >>> m.groups()      # Second group defaults to None.
+      ('24', None)
+      >>> m.groups('0')   # Now, the second group defaults to '0'.
+      ('24', '0')
 
 
-   .. method:: MatchObject.groupdict([default])
+.. method:: match.groupdict(default=None)
 
-      Return a dictionary containing all the *named* subgroups of the match, keyed by
-      the subgroup name.  The *default* argument is used for groups that did not
-      participate in the match; it defaults to ``None``.  For example:
+   Return a dictionary containing all the *named* subgroups of the match, keyed by
+   the subgroup name.  The *default* argument is used for groups that did not
+   participate in the match; it defaults to ``None``.  For example:
 
-         >>> m = re.match(r"(?P<first_name>\w+) (?P<last_name>\w+)", "Malcolm Reynolds")
-         >>> m.groupdict()
-         {'first_name': 'Malcolm', 'last_name': 'Reynolds'}
+      >>> m = re.match(r"(?P<first_name>\w+) (?P<last_name>\w+)", "Malcolm Reynolds")
+      >>> m.groupdict()
+      {'first_name': 'Malcolm', 'last_name': 'Reynolds'}
 
 
-   .. method:: MatchObject.start([group])
-               MatchObject.end([group])
+.. method:: match.start([group])
+            match.end([group])
 
-      Return the indices of the start and end of the substring matched by *group*;
-      *group* defaults to zero (meaning the whole matched substring). Return ``-1`` if
-      *group* exists but did not contribute to the match.  For a match object *m*, and
-      a group *g* that did contribute to the match, the substring matched by group *g*
-      (equivalent to ``m.group(g)``) is ::
+   Return the indices of the start and end of the substring matched by *group*;
+   *group* defaults to zero (meaning the whole matched substring). Return ``-1`` if
+   *group* exists but did not contribute to the match.  For a match object *m*, and
+   a group *g* that did contribute to the match, the substring matched by group *g*
+   (equivalent to ``m.group(g)``) is ::
 
-         m.string[m.start(g):m.end(g)]
+      m.string[m.start(g):m.end(g)]
 
-      Note that ``m.start(group)`` will equal ``m.end(group)`` if *group* matched a
-      null string.  For example, after ``m = re.search('b(c?)', 'cba')``,
-      ``m.start(0)`` is 1, ``m.end(0)`` is 2, ``m.start(1)`` and ``m.end(1)`` are both
-      2, and ``m.start(2)`` raises an :exc:`IndexError` exception.
+   Note that ``m.start(group)`` will equal ``m.end(group)`` if *group* matched a
+   null string.  For example, after ``m = re.search('b(c?)', 'cba')``,
+   ``m.start(0)`` is 1, ``m.end(0)`` is 2, ``m.start(1)`` and ``m.end(1)`` are both
+   2, and ``m.start(2)`` raises an :exc:`IndexError` exception.
 
-      An example that will remove *remove_this* from email addresses:
+   An example that will remove *remove_this* from email addresses:
 
-         >>> email = "tony at tiremove_thisger.net"
-         >>> m = re.search("remove_this", email)
-         >>> email[:m.start()] + email[m.end():]
-         'tony at tiger.net'
+      >>> email = "tony at tiremove_thisger.net"
+      >>> m = re.search("remove_this", email)
+      >>> email[:m.start()] + email[m.end():]
+      'tony at tiger.net'
 
 
-   .. method:: MatchObject.span([group])
+.. method:: match.span([group])
 
-      For :class:`MatchObject` *m*, return the 2-tuple ``(m.start(group),
-      m.end(group))``. Note that if *group* did not contribute to the match, this is
-      ``(-1, -1)``.  *group* defaults to zero, the entire match.
+   For a match *m*, return the 2-tuple ``(m.start(group), m.end(group))``. Note
+   that if *group* did not contribute to the match, this is ``(-1, -1)``.
+   *group* defaults to zero, the entire match.
 
 
-   .. attribute:: MatchObject.pos
+.. attribute:: match.pos
 
-      The value of *pos* which was passed to the :meth:`~RegexObject.search` or
-      :meth:`~RegexObject.match` method of the :class:`RegexObject`.  This is the
-      index into the string at which the RE engine started looking for a match.
+   The value of *pos* which was passed to the :meth:`~regex.search` or
+   :meth:`~regex.match` method of a :ref:`match object <match-objects>`.  This
+   is the index into the string at which the RE engine started looking for a
+   match.
 
 
-   .. attribute:: MatchObject.endpos
+.. attribute:: match.endpos
 
-      The value of *endpos* which was passed to the :meth:`~RegexObject.search` or
-      :meth:`~RegexObject.match` method of the :class:`RegexObject`.  This is the
-      index into the string beyond which the RE engine will not go.
+   The value of *endpos* which was passed to the :meth:`~regex.search` or
+   :meth:`~regex.match` method of a :ref:`match object <match-objects>`.  This
+   is the index into the string beyond which the RE engine will not go.
 
 
-   .. attribute:: MatchObject.lastindex
+.. attribute:: match.lastindex
 
-      The integer index of the last matched capturing group, or ``None`` if no group
-      was matched at all. For example, the expressions ``(a)b``, ``((a)(b))``, and
-      ``((ab))`` will have ``lastindex == 1`` if applied to the string ``'ab'``, while
-      the expression ``(a)(b)`` will have ``lastindex == 2``, if applied to the same
-      string.
+   The integer index of the last matched capturing group, or ``None`` if no group
+   was matched at all. For example, the expressions ``(a)b``, ``((a)(b))``, and
+   ``((ab))`` will have ``lastindex == 1`` if applied to the string ``'ab'``, while
+   the expression ``(a)(b)`` will have ``lastindex == 2``, if applied to the same
+   string.
 
 
-   .. attribute:: MatchObject.lastgroup
+.. attribute:: match.lastgroup
 
-      The name of the last matched capturing group, or ``None`` if the group didn't
-      have a name, or if no group was matched at all.
+   The name of the last matched capturing group, or ``None`` if the group didn't
+   have a name, or if no group was matched at all.
 
 
-   .. attribute:: MatchObject.re
+.. attribute:: match.re
 
-      The regular expression object whose :meth:`~RegexObject.match` or
-      :meth:`~RegexObject.search` method produced this :class:`MatchObject`
-      instance.
+   The regular expression object whose :meth:`~regex.match` or
+   :meth:`~regex.search` method produced this match instance.
 
 
-   .. attribute:: MatchObject.string
+.. attribute:: match.string
 
-      The string passed to :meth:`~RegexObject.match` or
-      :meth:`~RegexObject.search`.
+   The string passed to :meth:`~regex.match` or :meth:`~regex.search`.
 
 
 Examples
@@ -1033,8 +1034,7 @@
    "<Match: '354aa', groups=('a',)>"
 
 To find out what card the pair consists of, one could use the
-:meth:`~MatchObject.group` method of :class:`MatchObject` in the following
-manner:
+:meth:`~match.group` method of the match object in the following manner:
 
 .. doctest::
 
@@ -1109,7 +1109,7 @@
    >>> re.match('Begin (\w| )*? end', s).end()
    Traceback (most recent call last):
      File "<stdin>", line 1, in ?
-     File "/usr/local/lib/python2.5/re.py", line 132, in match
+     File "/usr/local/lib/python3.2/re.py", line 132, in match
        return _compile(pattern, flags).match(string)
    RuntimeError: maximum recursion limit exceeded
 
@@ -1225,9 +1225,9 @@
    ...   random.shuffle(inner_word)
    ...   return m.group(1) + "".join(inner_word) + m.group(3)
    >>> text = "Professor Abdolmalek, please report your absences promptly."
-   >>> re.sub("(\w)(\w+)(\w)", repl, text)
+   >>> re.sub(r"(\w)(\w+)(\w)", repl, text)
    'Poefsrosr Aealmlobdk, pslaee reorpt your abnseces plmrptoy.'
-   >>> re.sub("(\w)(\w+)(\w)", repl, text)
+   >>> re.sub(r"(\w)(\w+)(\w)", repl, text)
    'Pofsroser Aodlambelk, plasee reoprt yuor asnebces potlmrpy.'
 
 
@@ -1248,10 +1248,10 @@
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
 If one wants more information about all matches of a pattern than the matched
-text, :func:`finditer` is useful as it provides instances of
-:class:`MatchObject` instead of strings.  Continuing with the previous example,
-if one was a writer who wanted to find all of the adverbs *and their positions*
-in some text, he or she would use :func:`finditer` in the following manner:
+text, :func:`finditer` is useful as it provides :ref:`match objects
+<match-objects>` instead of strings.  Continuing with the previous example, if
+one was a writer who wanted to find all of the adverbs *and their positions* in
+some text, he or she would use :func:`finditer` in the following manner:
 
    >>> text = "He was carefully disguised but captured quickly by police."
    >>> for m in re.finditer(r"\w+ly", text):

Modified: python/branches/import_unicode/Doc/library/select.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/select.rst	(original)
+++ python/branches/import_unicode/Doc/library/select.rst	Fri Jul 30 01:19:37 2010
@@ -41,14 +41,14 @@
 
 .. function:: kqueue()
 
-   (Only supported on BSD.)  Returns a kernel queue object object; see section
+   (Only supported on BSD.)  Returns a kernel queue object; see section
    :ref:`kqueue-objects` below for the methods supported by kqueue objects.
 
 
 .. function:: kevent(ident, filter=KQ_FILTER_READ, flags=KQ_EV_ADD, fflags=0, data=0, udata=0)
 
-   (Only supported on BSD.)  Returns a kernel event object object; see section
-   :ref:`kevent-objects` below for the methods supported by kqueue objects.
+   (Only supported on BSD.)  Returns a kernel event object; see section
+   :ref:`kevent-objects` below for the methods supported by kevent objects.
 
 
 .. function:: select(rlist, wlist, xlist[, timeout])
@@ -131,15 +131,15 @@
    | :const:`EPOLLONESHOT` | Set one-shot behavior. After one event is     |
    |                       | pulled out, the fd is internally disabled     |
    +-----------------------+-----------------------------------------------+
-   | :const:`EPOLLRDNORM`  | ???                                           |
+   | :const:`EPOLLRDNORM`  | Equivalent to :const:`EPOLLIN`                |
    +-----------------------+-----------------------------------------------+
-   | :const:`EPOLLRDBAND`  | ???                                           |
+   | :const:`EPOLLRDBAND`  | Priority data band can be read.               |
    +-----------------------+-----------------------------------------------+
-   | :const:`EPOLLWRNORM`  | ???                                           |
+   | :const:`EPOLLWRNORM`  | Equivalent to :const:`EPOLLOUT`               |
    +-----------------------+-----------------------------------------------+
-   | :const:`EPOLLWRBAND`  | ???                                           |
+   | :const:`EPOLLWRBAND`  | Priority data may be written.                 |
    +-----------------------+-----------------------------------------------+
-   | :const:`EPOLLMSG`     | ???                                           |
+   | :const:`EPOLLMSG`     | Ignored.                                      |
    +-----------------------+-----------------------------------------------+
 
 
@@ -233,7 +233,7 @@
 .. method:: poll.modify(fd, eventmask)
 
    Modifies an already registered fd. This has the same effect as
-   :meth:`register(fd, eventmask)`.  Attempting to modify a file descriptor
+   ``register(fd, eventmask)``.  Attempting to modify a file descriptor
    that was never registered causes an :exc:`IOError` exception with errno
    :const:`ENOENT` to be raised.
 

Modified: python/branches/import_unicode/Doc/library/smtpd.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/smtpd.rst	(original)
+++ python/branches/import_unicode/Doc/library/smtpd.rst	Fri Jul 30 01:19:37 2010
@@ -10,10 +10,14 @@
 
 
 
-This module offers several classes to implement SMTP servers.  One is a generic
+This module offers several classes to implement SMTP (email) servers.
+
+Several server implementations are present; one is a generic
 do-nothing implementation, which can be overridden, while the other two offer
 specific mail-sending strategies.
 
+Additionally the SMTPChannel may be extended to implement very specific
+interaction behaviour with SMTP clients.
 
 SMTPServer Objects
 ------------------
@@ -26,7 +30,6 @@
    inherits from :class:`asyncore.dispatcher`, and so will insert itself into
    :mod:`asyncore`'s event loop on instantiation.
 
-
    .. method:: process_message(peer, mailfrom, rcpttos, data)
 
       Raise :exc:`NotImplementedError` exception. Override this in subclasses to
@@ -37,6 +40,11 @@
       containing the contents of the e-mail (which should be in :rfc:`2822`
       format).
 
+   .. attribute:: channel_class
+
+      Override this in subclasses to use a custom :class:`SMTPChannel` for
+      managing SMTP clients.
+
 
 DebuggingServer Objects
 -----------------------
@@ -71,3 +79,91 @@
    running this has a good chance to make you into an open relay, so please be
    careful.
 
+SMTPChannel Objects
+-------------------
+
+.. class:: SMTPChannel(server, conn, addr)
+
+   Create a new :class:`SMTPChannel` object which manages the communication
+   between the server and a single SMTP client.
+
+   To use a custom SMTPChannel implementation you need to override the
+   :attr:`SMTPServer.channel_class` of your :class:`SMTPServer`.
+
+   The :class:`SMTPChannel` has the following instance variables:
+
+   .. attribute:: smtp_server
+
+      Holds the :class:`SMTPServer` that spawned this channel.
+
+   .. attribute:: conn
+
+      Holds the socket object connecting to the client.
+
+   .. attribute:: addr
+
+      Holds the address of the client, the second value returned by
+      socket.accept()
+
+   .. attribute:: received_lines
+
+      Holds a list of the line strings (decoded using UTF-8) received from
+      the client. The lines have their "\r\n" line ending translated to "\n".
+
+   .. attribute:: smtp_state
+
+      Holds the current state of the channel. This will be either
+      :attr:`COMMAND` initially and then :attr:`DATA` after the client sends
+      a "DATA" line.
+
+   .. attribute:: seen_greeting
+
+      Holds a string containing the greeting sent by the client in its "HELO".
+
+   .. attribute:: mailfrom
+
+      Holds a string containing the address identified in the "MAIL FROM:" line
+      from the client.
+
+   .. attribute:: rcpttos
+
+      Holds a list of strings containing the addresses identified in the
+      "RCPT TO:" lines from the client.
+
+   .. attribute:: received_data
+
+      Holds a string containing all of the data sent by the client during the
+      DATA state, up to but not including the terminating "\r\n.\r\n".
+
+   .. attribute:: fqdn
+
+      Holds the fully-qualified domain name of the server as returned by
+      ``socket.getfqdn()``.
+
+   .. attribute:: peer
+
+      Holds the name of the client peer as returned by ``conn.getpeername()``
+      where ``conn`` is :attr:`conn`.
+
+   The :class:`SMTPChannel` operates by invoking methods named ``smtp_<command>``
+   upon reception of a command line from the client. Built into the base
+   :class:`SMTPChannel` class are methods for handling the following commands
+   (and responding to them appropriately):
+
+   ======== ===================================================================
+   Command  Action taken
+   ======== ===================================================================
+   HELO     Accepts the greeting from the client and stores it in
+            :attr:`seen_greeting`.
+   NOOP     Takes no action.
+   QUIT     Closes the connection cleanly.
+   MAIL     Accepts the "MAIL FROM:" syntax and stores the supplied address as
+            :attr:`mailfrom`.
+   RCPT     Accepts the "RCPT TO:" syntax and stores the supplied addresses in
+            the :attr:`rcpttos` list.
+   RSET     Resets the :attr:`mailfrom`, :attr:`rcpttos`, and
+            :attr:`received_data`, but not the greeting.
+   DATA     Sets the internal state to :attr:`DATA` and stores remaining lines
+            from the client in :attr:`received_data` until the terminator
+            "\r\n.\r\n" is received.
+   ======== ===================================================================
\ No newline at end of file

Modified: python/branches/import_unicode/Doc/library/socket.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/socket.rst	(original)
+++ python/branches/import_unicode/Doc/library/socket.rst	Fri Jul 30 01:19:37 2010
@@ -748,7 +748,9 @@
    Shut down one or both halves of the connection.  If *how* is :const:`SHUT_RD`,
    further receives are disallowed.  If *how* is :const:`SHUT_WR`, further sends
    are disallowed.  If *how* is :const:`SHUT_RDWR`, further sends and receives are
-   disallowed.
+   disallowed.  Depending on the platform, shutting down one half of the connection
+   can also close the opposite half (e.g. on Mac OS X, ``shutdown(SHUT_WR)`` does
+   not allow further reads on the other end of the connection).
 
 Note that there are no methods :meth:`read` or :meth:`write`; use
 :meth:`~socket.recv` and :meth:`~socket.send` without *flags* argument instead.

Modified: python/branches/import_unicode/Doc/library/sqlite3.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/sqlite3.rst	(original)
+++ python/branches/import_unicode/Doc/library/sqlite3.rst	Fri Jul 30 01:19:37 2010
@@ -136,7 +136,7 @@
    first blank for the column name: the column name would simply be "x".
 
 
-.. function:: connect(database[, timeout, isolation_level, detect_types, factory])
+.. function:: connect(database[, timeout, detect_types, isolation_level, check_same_thread, factory, cached_statements])
 
    Opens a connection to the SQLite database file *database*. You can use
    ``":memory:"`` to open a database connection to a database that resides in RAM

Modified: python/branches/import_unicode/Doc/library/stdtypes.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/stdtypes.rst	(original)
+++ python/branches/import_unicode/Doc/library/stdtypes.rst	Fri Jul 30 01:19:37 2010
@@ -2264,10 +2264,19 @@
    buffer protocol.  Builtin objects that support the buffer protocol include
    :class:`bytes` and :class:`bytearray`.
 
-   ``len(view)`` returns the total number of bytes in the memoryview, *view*.
+   A :class:`memoryview` has the notion of an *element*, which is the
+   atomic memory unit handled by the originating object *obj*.  For many
+   simple types such as :class:`bytes` and :class:`bytearray`, an element
+   is a single byte, but other types such as :class:`array.array` may have
+   bigger elements.
+
+   ``len(view)`` returns the total number of elements in the memoryview,
+   *view*.  The :class:`~memoryview.itemsize` attribute will give you the
+   number of bytes in a single element.
 
    A :class:`memoryview` supports slicing to expose its data.  Taking a single
-   index will return a single byte.  Full slicing will result in a subview::
+   index will return a single element as a :class:`bytes` object.  Full
+   slicing will result in a subview::
 
       >>> v = memoryview(b'abcefg')
       >>> v[1]
@@ -2278,11 +2287,8 @@
       <memory at 0x77ab28>
       >>> bytes(v[1:4])
       b'bce'
-      >>> v[3:-1]
-      <memory at 0x744f18>
-      >>> bytes(v[4:-1])
 
-   If the object the memory view is over supports changing its data, the
+   If the object the memoryview is over supports changing its data, the
    memoryview supports slice assignment::
 
       >>> data = bytearray(b'abcefg')
@@ -2302,12 +2308,18 @@
 
    Notice how the size of the memoryview object cannot be changed.
 
-
    :class:`memoryview` has two methods:
 
    .. method:: tobytes()
 
-      Return the data in the buffer as a bytestring.
+      Return the data in the buffer as a bytestring.  This is equivalent to
+      calling the :class:`bytes` constructor on the memoryview. ::
+
+         >>> m = memoryview(b"abc")
+         >>> m.tobytes()
+         b'abc'
+         >>> bytes(m)
+         b'abc'
 
    .. method:: tolist()
 
@@ -2325,7 +2337,15 @@
 
    .. attribute:: itemsize
 
-      The size in bytes of each element of the memoryview.
+      The size in bytes of each element of the memoryview::
+
+         >>> m = memoryview(array.array('H', [1,2,3]))
+         >>> m.itemsize
+         2
+         >>> m[0]
+         b'\x01\x00'
+         >>> len(m[0]) == m.itemsize
+         True
 
    .. attribute:: shape
 

Modified: python/branches/import_unicode/Doc/library/string.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/string.rst	(original)
+++ python/branches/import_unicode/Doc/library/string.rst	Fri Jul 30 01:19:37 2010
@@ -7,13 +7,13 @@
 
 .. index:: module: re
 
-The :mod:`string` module contains a number of useful constants and classes, as
-well as some deprecated legacy functions that are also available as methods on
-strings. In addition, Python's built-in string classes support the sequence type
-methods described in the :ref:`typesseq` section, and also the string-specific
-methods described in the :ref:`string-methods` section. To output formatted
-strings, see the :ref:`string-formatting` section. Also, see the :mod:`re`
-module for string functions based on regular expressions.
+The :mod:`string` module contains a number of useful constants and classes
+for string formatting.  In addition, Python's built-in string classes
+support the sequence type methods described in the :ref:`typesseq`
+section, and also the string-specific methods described in the
+:ref:`string-methods` section.  To output formatted strings, see the
+:ref:`string-formatting` section.  Also, see the :mod:`re` module for
+string functions based on regular expressions.
 
 
 String constants
@@ -710,6 +710,14 @@
   appropriate).  The default value is the regular expression
   ``[_a-z][_a-z0-9]*``.
 
+* *flags* -- The regular expression flags that will be applied when compiling
+  the regular expression used for recognizing substitutions.  The default value
+  is ``re.IGNORECASE``.  Note that ``re.VERBOSE`` will always be added to the
+  flags, so custom *idpattern*\ s must follow conventions for verbose regular
+  expressions.
+
+  .. versionadded:: 3.2
+
 Alternatively, you can provide the entire regular expression pattern by
 overriding the class attribute *pattern*.  If you do this, the value must be a
 regular expression object with four named capturing groups.  The capturing

Modified: python/branches/import_unicode/Doc/library/struct.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/struct.rst	(original)
+++ python/branches/import_unicode/Doc/library/struct.rst	Fri Jul 30 01:19:37 2010
@@ -259,18 +259,16 @@
    In 3.0, some of the integer formats wrapped out-of-range values and
    raised :exc:`DeprecationWarning` instead of :exc:`struct.error`.
 
-
 The ``'p'`` format character encodes a "Pascal string", meaning a short
-variable-length string stored in a fixed number of bytes. The count is the total
-number of bytes stored.  The first byte stored is the length of the string, or
-255, whichever is smaller.  The bytes of the string follow.  If the string
-passed in to :func:`pack` is too long (longer than the count minus 1), only the
-leading count-1 bytes of the string are stored.  If the string is shorter than
-count-1, it is padded with null bytes so that exactly count bytes in all are
-used.  Note that for :func:`unpack`, the ``'p'`` format character consumes count
-bytes, but that the string returned can never contain more than 255 bytes.
-
-
+variable-length string stored in a *fixed number of bytes*, given by the count.
+The first byte stored is the length of the string, or 255, whichever is
+smaller.  The bytes of the string follow.  If the string passed in to
+:func:`pack` is too long (longer than the count minus 1), only the leading
+``count-1`` bytes of the string are stored.  If the string is shorter than
+``count-1``, it is padded with null bytes so that exactly count bytes in all
+are used.  Note that for :func:`unpack`, the ``'p'`` format character consumes
+``count`` bytes, but that the string returned can never contain more than 255
+bytes.
 
 For the ``'?'`` format character, the return value is either :const:`True` or
 :const:`False`. When packing, the truth value of the argument object is used.

Modified: python/branches/import_unicode/Doc/library/subprocess.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/subprocess.rst	(original)
+++ python/branches/import_unicode/Doc/library/subprocess.rst	Fri Jul 30 01:19:37 2010
@@ -575,7 +575,7 @@
    pipe = os.popen(cmd, 'w')
    ...
    rc = pipe.close()
-   if rc is not None and rc % 256:
+   if rc is not None and rc >> 8:
        print("There were some errors")
    ==>
    process = Popen(cmd, 'w', stdin=PIPE)

Modified: python/branches/import_unicode/Doc/library/test.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/test.rst	(original)
+++ python/branches/import_unicode/Doc/library/test.rst	Fri Jul 30 01:19:37 2010
@@ -5,6 +5,13 @@
    :synopsis: Regression tests package containing the testing suite for Python.
 .. sectionauthor:: Brett Cannon <brett at python.org>
 
+.. note::
+    The :mod:`test` package is meant for internal use by Python only. It is
+    documented for the benefit of the core developers of Python. Any use of
+    this package outside of Python's standard library is discouraged as code
+    mentioned here can change or be removed without notice between releases of
+    Python.
+
 
 The :mod:`test` package contains all regression tests for Python as well as the
 modules :mod:`test.support` and :mod:`test.regrtest`.

Modified: python/branches/import_unicode/Doc/library/threading.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/threading.rst	(original)
+++ python/branches/import_unicode/Doc/library/threading.rst	Fri Jul 30 01:19:37 2010
@@ -310,8 +310,8 @@
 
       Return whether the thread is alive.
 
-      Roughly, a thread is alive from the moment the :meth:`start` method
-      returns until its :meth:`run` method terminates. The module function
+      This method returns ``True`` just before the :meth:`run` method starts
+      until just after the :meth:`run` method terminates.  The module function
       :func:`.enumerate` returns a list of all alive threads.
 
    .. attribute:: daemon

Modified: python/branches/import_unicode/Doc/library/tkinter.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/tkinter.rst	(original)
+++ python/branches/import_unicode/Doc/library/tkinter.rst	Fri Jul 30 01:19:37 2010
@@ -9,7 +9,9 @@
 The :mod:`tkinter` package ("Tk interface") is the standard Python interface to
 the Tk GUI toolkit.  Both Tk and :mod:`tkinter` are available on most Unix
 platforms, as well as on Windows systems.  (Tk itself is not part of Python; it
-is maintained at ActiveState.)
+is maintained at ActiveState.) You can check that :mod:`tkinter` is properly
+installed on your system by running ``python -m tkinter`` from the command line;
+this should open a window demonstrating a simple Tk interface.
 
 .. seealso::
 

Modified: python/branches/import_unicode/Doc/library/unittest.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/unittest.rst	(original)
+++ python/branches/import_unicode/Doc/library/unittest.rst	Fri Jul 30 01:19:37 2010
@@ -621,20 +621,20 @@
 
 The following decorators implement test skipping and expected failures:
 
-.. function:: skip(reason)
+.. decorator:: skip(reason)
 
    Unconditionally skip the decorated test.  *reason* should describe why the
    test is being skipped.
 
-.. function:: skipIf(condition, reason)
+.. decorator:: skipIf(condition, reason)
 
    Skip the decorated test if *condition* is true.
 
-.. function:: skipUnless(condition, reason)
+.. decorator:: skipUnless(condition, reason)
 
    Skip the decoratored test unless *condition* is true.
 
-.. function:: expectedFailure
+.. decorator:: expectedFailure
 
    Mark the test as an expected failure.  If the test fails when run, the test
    is not counted as a failure.
@@ -1048,11 +1048,11 @@
       :attr:`exception` attribute.  This can be useful if the intention
       is to perform additional checks on the exception raised::
 
-        with self.assertRaises(SomeException) as cm:
-            do_something()
+         with self.assertRaises(SomeException) as cm:
+             do_something()
 
-        the_exception = cm.exception
-        self.assertEqual(the_exception.error_code, 3)
+         the_exception = cm.exception
+         self.assertEqual(the_exception.error_code, 3)
 
       .. versionchanged:: 3.1
          Added the ability to use :meth:`assertRaises` as a context manager.

Modified: python/branches/import_unicode/Doc/library/urllib.request.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/urllib.request.rst	(original)
+++ python/branches/import_unicode/Doc/library/urllib.request.rst	Fri Jul 30 01:19:37 2010
@@ -638,7 +638,8 @@
    :meth:`unknown_open`.
 
    Note that the implementation of these methods may involve calls of the parent
-   :class:`OpenerDirector` instance's :meth:`.open` and :meth:`.error` methods.
+   :class:`OpenerDirector` instance's :meth:`~OpenerDirector.open` and
+   :meth:`~OpenerDirector.error` methods.
 
 #. Every handler with a method named like :meth:`protocol_response` has that
    method called to post-process the response.

Modified: python/branches/import_unicode/Doc/library/xml.dom.minidom.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/xml.dom.minidom.rst	(original)
+++ python/branches/import_unicode/Doc/library/xml.dom.minidom.rst	Fri Jul 30 01:19:37 2010
@@ -136,18 +136,20 @@
 
 .. method:: Node.toxml(encoding=None)
 
-   Return the XML that the DOM represents as a string.
-
-   With no argument, the XML header does not specify an encoding, and the result is
-   Unicode string if the default encoding cannot represent all characters in the
-   document. Encoding this string in an encoding other than UTF-8 is likely
-   incorrect, since UTF-8 is the default encoding of XML.
-
-   With an explicit *encoding* [1]_ argument, the result is a byte string in the
-   specified encoding. It is recommended that this argument is always specified. To
-   avoid :exc:`UnicodeError` exceptions in case of unrepresentable text data, the
-   encoding argument should be specified as "utf-8".
+   Return a string or byte string containing the XML represented by
+   the DOM node.
 
+   With an explicit *encoding* [1]_ argument, the result is a byte
+   string in the specified encoding.  It is recommended that you
+   always specify an encoding; you may use any encoding you like, but
+   an argument of "utf-8" is the most common choice, avoiding
+   :exc:`UnicodeError` exceptions in case of unrepresentable text
+   data.
+
+   With no *encoding* argument, the result is a Unicode string, and the
+   XML declaration in the resulting string does not specify an
+   encoding. Encoding this string in an encoding other than UTF-8 is
+   likely incorrect, since UTF-8 is the default encoding of XML.
 
 .. method:: Node.toprettyxml(indent="", newl="", encoding="")
 
@@ -155,7 +157,8 @@
    indentation string and defaults to a tabulator; *newl* specifies the string
    emitted at the end of each line and defaults to ``\n``.
 
-   There's also an *encoding* argument; see :meth:`toxml`.
+   The *encoding* argument behaves like the corresponding argument of
+   :meth:`toxml`.
 
 
 .. _dom-example:
@@ -240,7 +243,9 @@
 
 .. rubric:: Footnotes
 
-.. [#] The encoding string included in XML output should conform to the
-   appropriate standards. For example, "UTF-8" is valid, but "UTF8" is
-   not. See http://www.w3.org/TR/2006/REC-xml11-20060816/#NT-EncodingDecl
+.. [#] The encoding name included in the XML output should conform to
+   the appropriate standards. For example, "UTF-8" is valid, but
+   "UTF8" is not valid in an XML document's declaration, even though
+   Python accepts it as an encoding name.
+   See http://www.w3.org/TR/2006/REC-xml11-20060816/#NT-EncodingDecl
    and http://www.iana.org/assignments/character-sets .

Modified: python/branches/import_unicode/Doc/library/xml.dom.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/xml.dom.rst	(original)
+++ python/branches/import_unicode/Doc/library/xml.dom.rst	Fri Jul 30 01:19:37 2010
@@ -976,29 +976,24 @@
 Type Mapping
 ^^^^^^^^^^^^
 
-The primitive IDL types used in the DOM specification are mapped to Python types
+The IDL types used in the DOM specification are mapped to Python types
 according to the following table.
 
 +------------------+-------------------------------------------+
 | IDL Type         | Python Type                               |
 +==================+===========================================+
-| ``boolean``      | ``IntegerType`` (with a value of ``0`` or |
-|                  | ``1``)                                    |
+| ``boolean``      | ``bool`` or ``int``                       |
 +------------------+-------------------------------------------+
-| ``int``          | ``IntegerType``                           |
+| ``int``          | ``int``                                   |
 +------------------+-------------------------------------------+
-| ``long int``     | ``IntegerType``                           |
+| ``long int``     | ``int``                                   |
 +------------------+-------------------------------------------+
-| ``unsigned int`` | ``IntegerType``                           |
+| ``unsigned int`` | ``int``                                   |
++------------------+-------------------------------------------+
+| ``DOMString``    | ``str`` or ``bytes``                      |
++------------------+-------------------------------------------+
+| ``null``         | ``None``                                  |
 +------------------+-------------------------------------------+
-
-Additionally, the :class:`DOMString` defined in the recommendation is mapped to
-a bytes or string object.  Applications should be able to handle
-Unicode whenever a string is returned from the DOM.
-
-The IDL ``null`` value is mapped to ``None``, which may be accepted or
-provided by the implementation whenever ``null`` is allowed by the API.
-
 
 .. _dom-accessor-methods:
 

Modified: python/branches/import_unicode/Doc/reference/expressions.rst
==============================================================================
--- python/branches/import_unicode/Doc/reference/expressions.rst	(original)
+++ python/branches/import_unicode/Doc/reference/expressions.rst	Fri Jul 30 01:19:37 2010
@@ -1031,9 +1031,9 @@
   ``x <= y``.  If the corresponding element does not exist, the shorter
   sequence is ordered first (for example, ``[1,2] < [1,2,3]``).
 
-* Mappings (dictionaries) compare equal if and only if their sorted ``(key,
-  value)`` lists compare equal. [#]_ Outcomes other than equality are resolved
-  consistently, but are not otherwise defined. [#]_
+* Mappings (dictionaries) compare equal if and only if they have the same
+  ``(key, value)`` pairs. Order comparisons ``('<', '<=', '>=', '>')``
+  raise :exc:`TypeError`.
 
 * Sets and frozensets define comparison operators to mean subset and superset
   tests.  Those relations do not define total orderings (the two sets ``{1,2}``
@@ -1330,15 +1330,6 @@
    strings in a human recognizable way, compare using
    :func:`unicodedata.normalize`.
 
-.. [#] The implementation computes this efficiently, without constructing lists
-   or sorting.
-
-.. [#] Earlier versions of Python used lexicographic comparison of the sorted (key,
-   value) lists, but this was very expensive for the common case of comparing
-   for equality.  An even earlier version of Python compared dictionaries by
-   identity only, but this caused surprises because people expected to be able
-   to test a dictionary for emptiness by comparing it to ``{}``.
-
 .. [#] Due to automatic garbage-collection, free lists, and the dynamic nature of
    descriptors, you may notice seemingly unusual behaviour in certain uses of
    the :keyword:`is` operator, like those involving comparisons between instance

Modified: python/branches/import_unicode/Doc/tools/sphinxext/indexsidebar.html
==============================================================================
--- python/branches/import_unicode/Doc/tools/sphinxext/indexsidebar.html	(original)
+++ python/branches/import_unicode/Doc/tools/sphinxext/indexsidebar.html	Fri Jul 30 01:19:37 2010
@@ -2,8 +2,7 @@
             <p><a href="{{ pathto('download') }}">Download these documents</a></p>
 	    <h3>Docs for other versions</h3>
 	    <ul>
-	      <li><a href="http://docs.python.org/2.6/">Python 2.6 (stable)</a></li>
-	      <li><a href="http://docs.python.org/2.7/">Python 2.7 (in development)</a></li>
+	      <li><a href="http://docs.python.org/2.7/">Python 2.7 (stable)</a></li>
 	      <li><a href="http://docs.python.org/3.1/">Python 3.1 (stable)</a></li>
               <li><a href="http://www.python.org/doc/versions/">Old versions</a></li>
             </ul>

Modified: python/branches/import_unicode/Doc/tools/sphinxext/pyspecific.py
==============================================================================
--- python/branches/import_unicode/Doc/tools/sphinxext/pyspecific.py	(original)
+++ python/branches/import_unicode/Doc/tools/sphinxext/pyspecific.py	Fri Jul 30 01:19:37 2010
@@ -72,6 +72,32 @@
         return [pnode]
 
 
+# Support for documenting decorators
+
+from sphinx import addnodes
+from sphinx.domains.python import PyModulelevel, PyClassmember
+
+class PyDecoratorMixin(object):
+    def handle_signature(self, sig, signode):
+        ret = super(PyDecoratorMixin, self).handle_signature(sig, signode)
+        signode.insert(0, addnodes.desc_addname('@', '@'))
+        return ret
+
+    def needs_arglist(self):
+        return False
+
+class PyDecoratorFunction(PyDecoratorMixin, PyModulelevel):
+    def run(self):
+        # a decorator function is a function after all
+        self.name = 'py:function'
+        return PyModulelevel.run(self)
+
+class PyDecoratorMethod(PyDecoratorMixin, PyClassmember):
+    def run(self):
+        self.name = 'py:method'
+        return PyClassmember.run(self)
+
+
 # Support for building "topic help" for pydoc
 
 pydoc_topic_labels = [
@@ -147,7 +173,6 @@
 # Support for documenting Opcodes
 
 import re
-from sphinx import addnodes
 
 opcode_sig_re = re.compile(r'(\w+(?:\+\d)?)(?:\s*\((.*)\))?')
 
@@ -165,6 +190,28 @@
     return opname.strip()
 
 
+pdbcmd_sig_re = re.compile(r'([a-z()!]+)\s*(.*)')
+
+# later...
+#pdbargs_tokens_re = re.compile(r'''[a-zA-Z]+  |  # identifiers
+#                                   [.,:]+     |  # punctuation
+#                                   [\[\]()]   |  # parens
+#                                   \s+           # whitespace
+#                                   ''', re.X)
+
+def parse_pdb_command(env, sig, signode):
+    """Transform a pdb command signature into RST nodes."""
+    m = pdbcmd_sig_re.match(sig)
+    if m is None:
+        raise ValueError
+    name, args = m.groups()
+    fullname = name.replace('(', '').replace(')', '')
+    signode += addnodes.desc_name(name, name)
+    if args:
+        signode += addnodes.desc_addname(' '+args, ' '+args)
+    return fullname
+
+
 def setup(app):
     app.add_role('issue', issue_role)
     app.add_directive('impl-detail', ImplementationDetail)
@@ -172,4 +219,8 @@
     app.add_builder(suspicious.CheckSuspiciousMarkupBuilder)
     app.add_description_unit('opcode', 'opcode', '%s (opcode)',
                              parse_opcode_signature)
+    app.add_description_unit('pdbcommand', 'pdbcmd', '%s (pdb command)',
+                             parse_pdb_command)
     app.add_description_unit('2to3fixer', '2to3fixer', '%s (2to3 fixer)')
+    app.add_directive_to_domain('py', 'decorator', PyDecoratorFunction)
+    app.add_directive_to_domain('py', 'decoratormethod', PyDecoratorMethod)

Modified: python/branches/import_unicode/Doc/tutorial/classes.rst
==============================================================================
--- python/branches/import_unicode/Doc/tutorial/classes.rst	(original)
+++ python/branches/import_unicode/Doc/tutorial/classes.rst	Fri Jul 30 01:19:37 2010
@@ -599,12 +599,12 @@
 possible to access or modify a variable that is considered private.  This can
 even be useful in special circumstances, such as in the debugger.
 
-Notice that code passed to ``exec()``, ``eval()`` or ``execfile()`` does not
-consider the classname of the invoking  class to be the current class; this is
-similar to the effect of the  ``global`` statement, the effect of which is
-likewise restricted to  code that is byte-compiled together.  The same
-restriction applies to ``getattr()``, ``setattr()`` and ``delattr()``, as well
-as when referencing ``__dict__`` directly.
+Notice that code passed to ``exec()`` or ``eval()`` does not consider the
+classname of the invoking class to be the current class; this is similar to the
+effect of the ``global`` statement, the effect of which is likewise restricted
+to code that is byte-compiled together.  The same restriction applies to
+``getattr()``, ``setattr()`` and ``delattr()``, as well as when referencing
+``__dict__`` directly.
 
 
 .. _tut-odds:
@@ -737,7 +737,7 @@
    StopIteration
 
 Having seen the mechanics behind the iterator protocol, it is easy to add
-iterator behavior to your classes.  Define a :meth:`__iter__` method which
+iterator behavior to your classes.  Define an :meth:`__iter__` method which
 returns an object with a :meth:`__next__` method.  If the class defines
 :meth:`__next__`, then :meth:`__iter__` can just return ``self``::
 
@@ -754,7 +754,10 @@
            self.index = self.index - 1
            return self.data[self.index]
 
-   >>> for char in Reverse('spam'):
+   >>> rev = Reverse('spam')
+   >>> iter(rev)
+   <__main__.Reverse object at 0x00A1DB50>
+   >>> for char in rev:
    ...     print(char)
    ...
    m

Modified: python/branches/import_unicode/Doc/tutorial/floatingpoint.rst
==============================================================================
--- python/branches/import_unicode/Doc/tutorial/floatingpoint.rst	(original)
+++ python/branches/import_unicode/Doc/tutorial/floatingpoint.rst	Fri Jul 30 01:19:37 2010
@@ -81,7 +81,7 @@
 values share the same approximation, any one of them could be displayed
 while still preserving the invariant ``eval(repr(x)) == x``.
 
-Historically, the Python prompt and built-in :func:`repr` function would chose
+Historically, the Python prompt and built-in :func:`repr` function would choose
 the one with 17 significant digits, ``0.10000000000000001``.   Starting with
 Python 3.1, Python (on most systems) is now able to choose the shortest of
 these and simply display ``0.1``.

Modified: python/branches/import_unicode/Doc/whatsnew/2.0.rst
==============================================================================
--- python/branches/import_unicode/Doc/whatsnew/2.0.rst	(original)
+++ python/branches/import_unicode/Doc/whatsnew/2.0.rst	Fri Jul 30 01:19:37 2010
@@ -656,7 +656,7 @@
 The change which will probably break the most code is tightening up the
 arguments accepted by some methods.  Some methods would take multiple arguments
 and treat them as a tuple, particularly various list methods such as
-:meth:`.append` and :meth:`.insert`. In earlier versions of Python, if ``L`` is
+:meth:`append` and :meth:`insert`. In earlier versions of Python, if ``L`` is
 a list, ``L.append( 1,2 )`` appends the tuple ``(1,2)`` to the list.  In Python
 2.0 this causes a :exc:`TypeError` exception to be raised, with the message:
 'append requires exactly 1 argument; 2 given'.  The fix is to simply add an

Modified: python/branches/import_unicode/Doc/whatsnew/2.5.rst
==============================================================================
--- python/branches/import_unicode/Doc/whatsnew/2.5.rst	(original)
+++ python/branches/import_unicode/Doc/whatsnew/2.5.rst	Fri Jul 30 01:19:37 2010
@@ -1765,7 +1765,7 @@
 http://effbot.org/zone/element-index.htm.
 
 ElementTree represents an XML document as a tree of element nodes. The text
-content of the document is stored as the :attr:`.text` and :attr:`.tail`
+content of the document is stored as the :attr:`text` and :attr:`tail`
 attributes of  (This is one of the major differences between ElementTree and
 the Document Object Model; in the DOM there are many different types of node,
 including :class:`TextNode`.)

Modified: python/branches/import_unicode/Doc/whatsnew/2.7.rst
==============================================================================
--- python/branches/import_unicode/Doc/whatsnew/2.7.rst	(original)
+++ python/branches/import_unicode/Doc/whatsnew/2.7.rst	Fri Jul 30 01:19:37 2010
@@ -146,7 +146,7 @@
 * Float-to-string and string-to-float conversions now round their
   results more correctly, and :func:`repr` of a floating-point
   number *x* returns a result that's guaranteed to round back to the
-  same number when converted back to a string.
+  same number when converted back to a float.
 * The :ctype:`PyCapsule` type, used to provide a C API for extension modules.
 * The :cfunc:`PyLong_AsLongAndOverflow` C API function.
 

Modified: python/branches/import_unicode/Grammar/Grammar
==============================================================================
--- python/branches/import_unicode/Grammar/Grammar	(original)
+++ python/branches/import_unicode/Grammar/Grammar	Fri Jul 30 01:19:37 2010
@@ -24,13 +24,13 @@
 decorated: decorators (classdef | funcdef)
 funcdef: 'def' NAME parameters ['->' test] ':' suite
 parameters: '(' [typedargslist] ')'
-typedargslist: ((tfpdef ['=' test] ',')*
-                ('*' [tfpdef] (',' tfpdef ['=' test])* [',' '**' tfpdef] | '**' tfpdef)
-                | tfpdef ['=' test] (',' tfpdef ['=' test])* [','])
+typedargslist: (tfpdef ['=' test] (',' tfpdef ['=' test])* [','
+       ['*' [tfpdef] (',' tfpdef ['=' test])* [',' '**' tfpdef] | '**' tfpdef]]
+     |  '*' [tfpdef] (',' tfpdef ['=' test])* [',' '**' tfpdef] | '**' tfpdef)
 tfpdef: NAME [':' test]
-varargslist: ((vfpdef ['=' test] ',')*
-              ('*' [vfpdef] (',' vfpdef ['=' test])* [',' '**' vfpdef] | '**' vfpdef)
-              | vfpdef ['=' test] (',' vfpdef ['=' test])* [','])
+varargslist: (vfpdef ['=' test] (',' vfpdef ['=' test])* [','
+       ['*' [vfpdef] (',' vfpdef ['=' test])* [',' '**' vfpdef] | '**' vfpdef]]
+     |  '*' [vfpdef] (',' vfpdef ['=' test])* [',' '**' vfpdef] | '**' vfpdef)
 vfpdef: NAME
 
 stmt: simple_stmt | compound_stmt

Modified: python/branches/import_unicode/Include/Python.h
==============================================================================
--- python/branches/import_unicode/Include/Python.h	(original)
+++ python/branches/import_unicode/Include/Python.h	Fri Jul 30 01:19:37 2010
@@ -136,13 +136,8 @@
 }
 #endif
 
-/* Convert a possibly signed character to a nonnegative int */
-/* XXX This assumes characters are 8 bits wide */
-#ifdef __CHAR_UNSIGNED__
-#define Py_CHARMASK(c)		(c)
-#else
+/* Argument must be a char or an int in [-128, 127] or [0, 255]. */
 #define Py_CHARMASK(c)		((unsigned char)((c) & 0xff))
-#endif
 
 #include "pyfpe.h"
 

Modified: python/branches/import_unicode/Lib/_abcoll.py
==============================================================================
--- python/branches/import_unicode/Lib/_abcoll.py	(original)
+++ python/branches/import_unicode/Lib/_abcoll.py	Fri Jul 30 01:19:37 2010
@@ -480,7 +480,15 @@
         except KeyError:
             pass
 
-    def update(self, other=(), **kwds):
+    def update(*args, **kwds):
+        if len(args) > 2:
+            raise TypeError("update() takes at most 2 positional "
+                            "arguments ({} given)".format(len(args)))
+        elif not args:
+            raise TypeError("update() takes at least 1 argument (0 given)")
+        self = args[0]
+        other = args[1] if len(args) >= 2 else ()
+
         if isinstance(other, Mapping):
             for key in other:
                 self[key] = other[key]

Modified: python/branches/import_unicode/Lib/_strptime.py
==============================================================================
--- python/branches/import_unicode/Lib/_strptime.py	(original)
+++ python/branches/import_unicode/Lib/_strptime.py	Fri Jul 30 01:19:37 2010
@@ -482,8 +482,8 @@
     tt = _strptime(data_string, format)[0]
     return time.struct_time(tt[:9])
 
-def _strptime_datetime(class_, data_string, format="%a %b %d %H:%M:%S %Y"):
-    """Return a class_ instance based on the input string and the
+def _strptime_datetime(cls, data_string, format="%a %b %d %H:%M:%S %Y"):
+    """Return a class cls instance based on the input string and the
     format string."""
     tt, fraction = _strptime(data_string, format)
     gmtoff, tzname = tt[-2:]
@@ -496,4 +496,4 @@
             tz = datetime_timezone(tzdelta)
         args += (tz,)
 
-    return class_(*args)
+    return cls(*args)

Modified: python/branches/import_unicode/Lib/ast.py
==============================================================================
--- python/branches/import_unicode/Lib/ast.py	(original)
+++ python/branches/import_unicode/Lib/ast.py	Fri Jul 30 01:19:37 2010
@@ -50,7 +50,7 @@
     if isinstance(node_or_string, Expression):
         node_or_string = node_or_string.body
     def _convert(node):
-        if isinstance(node, Str):
+        if isinstance(node, (Str, Bytes)):
             return node.s
         elif isinstance(node, Num):
             return node.n
@@ -58,6 +58,8 @@
             return tuple(map(_convert, node.elts))
         elif isinstance(node, List):
             return list(map(_convert, node.elts))
+        elif isinstance(node, Set):
+            return set(map(_convert, node.elts))
         elif isinstance(node, Dict):
             return dict((_convert(k), _convert(v)) for k, v
                         in zip(node.keys, node.values))

Modified: python/branches/import_unicode/Lib/asyncore.py
==============================================================================
--- python/branches/import_unicode/Lib/asyncore.py	(original)
+++ python/branches/import_unicode/Lib/asyncore.py	Fri Jul 30 01:19:37 2010
@@ -607,6 +607,14 @@
         def send(self, *args):
             return os.write(self.fd, *args)
 
+        def getsockopt(self, level, optname, buflen=None):
+            if (level == socket.SOL_SOCKET and
+                optname == socket.SO_ERROR and
+                not buflen):
+                return 0
+            raise NotImplementedError("Only asyncore specific behaviour "
+                                      "implemented.")
+
         read = recv
         write = send
 

Modified: python/branches/import_unicode/Lib/base64.py
==============================================================================
--- python/branches/import_unicode/Lib/base64.py	(original)
+++ python/branches/import_unicode/Lib/base64.py	Fri Jul 30 01:19:37 2010
@@ -241,7 +241,7 @@
         acc += _b32rev[c] << shift
         shift -= 5
         if shift < 0:
-            parts.append(binascii.unhexlify('%010x' % acc))
+            parts.append(binascii.unhexlify(bytes('%010x' % acc, "ascii")))
             acc = 0
             shift = 35
     # Process the last, partial quanta

Modified: python/branches/import_unicode/Lib/configparser.py
==============================================================================
--- python/branches/import_unicode/Lib/configparser.py	(original)
+++ python/branches/import_unicode/Lib/configparser.py	Fri Jul 30 01:19:37 2010
@@ -1,6 +1,6 @@
 """Configuration file parser.
 
-A setup file consists of sections, lead by a "[section]" header,
+A configuration file consists of sections, lead by a "[section]" header,
 and followed by "name: value" entries, with continuations and such in
 the style of RFC 822.
 
@@ -24,67 +24,88 @@
 
     methods:
 
-    __init__(defaults=None)
-        create the parser and specify a dictionary of intrinsic defaults.  The
-        keys must be strings, the values must be appropriate for %()s string
-        interpolation.  Note that `__name__' is always an intrinsic default;
-        its value is the section's name.
+    __init__(defaults=None, dict_type=_default_dict,
+             delimiters=('=', ':'), comment_prefixes=('#', ';'),
+             empty_lines_in_values=True, allow_no_value=False):
+        Create the parser. When `defaults' is given, it is initialized into the
+        dictionary or intrinsic defaults. The keys must be strings, the values
+        must be appropriate for %()s string interpolation.  Note that `__name__'
+        is always an intrinsic default; its value is the section's name.
+
+        When `dict_type' is given, it will be used to create the dictionary
+        objects for the list of sections, for the options within a section, and
+        for the default values.
+
+        When `delimiters' is given, it will be used as the set of substrings
+        that divide keys from values.
+
+        When `comment_prefixes' is given, it will be used as the set of
+        substrings that prefix comments in a line.
+
+        When `empty_lines_in_values' is False (default: True), each empty line
+        marks the end of an option. Otherwise, internal empty lines of
+        a multiline option are kept as part of the value.
+
+        When `allow_no_value' is True (default: False), options without
+        values are accepted; the value presented for these is None.
 
     sections()
-        return all the configuration section names, sans DEFAULT
+        Return all the configuration section names, sans DEFAULT.
 
     has_section(section)
-        return whether the given section exists
+        Return whether the given section exists.
 
     has_option(section, option)
-        return whether the given option exists in the given section
+        Return whether the given option exists in the given section.
 
     options(section)
-        return list of configuration options for the named section
+        Return list of configuration options for the named section.
 
-    read(filenames)
-        read and parse the list of named configuration files, given by
+    read(filenames, encoding=None)
+        Read and parse the list of named configuration files, given by
         name.  A single filename is also allowed.  Non-existing files
         are ignored.  Return list of successfully read files.
 
     readfp(fp, filename=None)
-        read and parse one configuration file, given as a file object.
+        Read and parse one configuration file, given as a file object.
         The filename defaults to fp.name; it is only used in error
         messages (if fp has no `name' attribute, the string `<???>' is used).
 
     get(section, option, raw=False, vars=None)
-        return a string value for the named option.  All % interpolations are
+        Return a string value for the named option.  All % interpolations are
         expanded in the return values, based on the defaults passed into the
         constructor and the DEFAULT section.  Additional substitutions may be
         provided using the `vars' argument, which must be a dictionary whose
         contents override any pre-existing defaults.
 
     getint(section, options)
-        like get(), but convert value to an integer
+        Like get(), but convert value to an integer.
 
     getfloat(section, options)
-        like get(), but convert value to a float
+        Like get(), but convert value to a float.
 
     getboolean(section, options)
-        like get(), but convert value to a boolean (currently case
+        Like get(), but convert value to a boolean (currently case
         insensitively defined as 0, false, no, off for False, and 1, true,
         yes, on for True).  Returns False or True.
 
     items(section, raw=False, vars=None)
-        return a list of tuples with (name, value) for each option
+        Return a list of tuples with (name, value) for each option
         in the section.
 
     remove_section(section)
-        remove the given file section and all its options
+        Remove the given file section and all its options.
 
     remove_option(section, option)
-        remove the given option from the given section
+        Remove the given option from the given section.
 
     set(section, option, value)
-        set the given option
+        Set the given option.
 
-    write(fp)
-        write the configuration state in .ini format
+    write(fp, space_around_delimiters=True)
+        Write the configuration state in .ini format. If
+        `space_around_delimiters' is True (the default), delimiters
+        between keys and values are surrounded by spaces.
 """
 
 try:
@@ -94,6 +115,7 @@
     _default_dict = dict
 
 import re
+import sys
 
 __all__ = ["NoSectionError", "DuplicateSectionError", "NoOptionError",
            "InterpolationError", "InterpolationDepthError",
@@ -114,17 +136,19 @@
 
     def _get_message(self):
         """Getter for 'message'; needed only to override deprecation in
-        BaseException."""
+        BaseException.
+        """
         return self.__message
 
     def _set_message(self, value):
         """Setter for 'message'; needed only to override deprecation in
-        BaseException."""
+        BaseException.
+        """
         self.__message = value
 
     # BaseException.message has been deprecated since Python 2.6.  To prevent
-    # DeprecationWarning from popping up over this pre-existing attribute, use
-    # a new property that takes lookup precedence.
+    # DeprecationWarning from popping up over this pre-existing attribute, use a
+    # new property that takes lookup precedence.
     message = property(_get_message, _set_message)
 
     def __init__(self, msg=''):
@@ -136,12 +160,15 @@
 
     __str__ = __repr__
 
+
 class NoSectionError(Error):
     """Raised when no section matches a requested option."""
 
     def __init__(self, section):
         Error.__init__(self, 'No section: %r' % (section,))
         self.section = section
+        self.args = (section, )
+
 
 class DuplicateSectionError(Error):
     """Raised when a section is multiply-created."""
@@ -149,6 +176,8 @@
     def __init__(self, section):
         Error.__init__(self, "Section %r already exists" % section)
         self.section = section
+        self.args = (section, )
+
 
 class NoOptionError(Error):
     """A requested option was not found."""
@@ -158,6 +187,8 @@
                        (option, section))
         self.option = option
         self.section = section
+        self.args = (option, section)
+
 
 class InterpolationError(Error):
     """Base class for interpolation-related exceptions."""
@@ -166,6 +197,8 @@
         Error.__init__(self, msg)
         self.option = option
         self.section = section
+        self.args = (option, section, msg)
+
 
 class InterpolationMissingOptionError(InterpolationError):
     """A string substitution required a setting which was not available."""
@@ -179,11 +212,14 @@
                % (section, option, reference, rawval))
         InterpolationError.__init__(self, option, section, msg)
         self.reference = reference
+        self.args = (option, section, rawval, reference)
+
 
 class InterpolationSyntaxError(InterpolationError):
     """Raised when the source text into which substitutions are made
     does not conform to the required syntax."""
 
+
 class InterpolationDepthError(InterpolationError):
     """Raised when substitutions are nested too deeply."""
 
@@ -194,6 +230,8 @@
                "\trawval : %s\n"
                % (section, option, rawval))
         InterpolationError.__init__(self, option, section, msg)
+        self.args = (option, section, rawval)
+
 
 class ParsingError(Error):
     """Raised when a configuration file does not follow legal syntax."""
@@ -202,11 +240,13 @@
         Error.__init__(self, 'File contains parsing errors: %s' % filename)
         self.filename = filename
         self.errors = []
+        self.args = (filename, )
 
     def append(self, lineno, line):
         self.errors.append((lineno, line))
         self.message += '\n\t[line %2d]: %s' % (lineno, line)
 
+
 class MissingSectionHeaderError(ParsingError):
     """Raised when a key-value pair is found before any section header."""
 
@@ -218,21 +258,76 @@
         self.filename = filename
         self.lineno = lineno
         self.line = line
+        self.args = (filename, lineno, line)
 
 
 class RawConfigParser:
+    """ConfigParser that does not do interpolation."""
+
+    # Regular expressions for parsing section headers and options
+    _SECT_TMPL = r"""
+        \[                                 # [
+        (?P<header>[^]]+)                  # very permissive!
+        \]                                 # ]
+        """
+    _OPT_TMPL = r"""
+        (?P<option>.*?)                    # very permissive!
+        \s*(?P<vi>{delim})\s*              # any number of space/tab,
+                                           # followed by any of the
+                                           # allowed delimiters,
+                                           # followed by any space/tab
+        (?P<value>.*)$                     # everything up to eol
+        """
+    _OPT_NV_TMPL = r"""
+        (?P<option>.*?)                    # very permissive!
+        \s*(?:                             # any number of space/tab,
+        (?P<vi>{delim})\s*                 # optionally followed by
+                                           # any of the allowed
+                                           # delimiters, followed by any
+                                           # space/tab
+        (?P<value>.*))?$                   # everything up to eol
+        """
+
+    # Compiled regular expression for matching sections
+    SECTCRE = re.compile(_SECT_TMPL, re.VERBOSE)
+    # Compiled regular expression for matching options with typical separators
+    OPTCRE = re.compile(_OPT_TMPL.format(delim="=|:"), re.VERBOSE)
+    # Compiled regular expression for matching options with optional values
+    # delimited using typical separators
+    OPTCRE_NV = re.compile(_OPT_NV_TMPL.format(delim="=|:"), re.VERBOSE)
+    # Compiled regular expression for matching leading whitespace in a line
+    NONSPACECRE = re.compile(r"\S")
+    # Select backwards-compatible inline comment character behavior
+    # (; and # are comments at the start of a line, but ; only inline)
+    _COMPATIBLE = object()
+
     def __init__(self, defaults=None, dict_type=_default_dict,
-                 allow_no_value=False):
+                 delimiters=('=', ':'), comment_prefixes=_COMPATIBLE,
+                 empty_lines_in_values=True, allow_no_value=False):
         self._dict = dict_type
         self._sections = self._dict()
         self._defaults = self._dict()
-        if allow_no_value:
-            self._optcre = self.OPTCRE_NV
-        else:
-            self._optcre = self.OPTCRE
         if defaults:
             for key, value in defaults.items():
                 self._defaults[self.optionxform(key)] = value
+        self._delimiters = tuple(delimiters)
+        if delimiters == ('=', ':'):
+            self._optcre = self.OPTCRE_NV if allow_no_value else self.OPTCRE
+        else:
+            delim = "|".join(re.escape(d) for d in delimiters)
+            if allow_no_value:
+                self._optcre = re.compile(self._OPT_NV_TMPL.format(delim=delim),
+                                          re.VERBOSE)
+            else:
+                self._optcre = re.compile(self._OPT_TMPL.format(delim=delim),
+                                          re.VERBOSE)
+        if comment_prefixes is self._COMPATIBLE:
+            self._startonly_comment_prefixes = ('#',)
+            self._comment_prefixes = (';',)
+        else:
+            self._startonly_comment_prefixes = ()
+            self._comment_prefixes = tuple(comment_prefixes or ())
+        self._empty_lines_in_values = empty_lines_in_values
 
     def defaults(self):
         return self._defaults
@@ -274,7 +369,7 @@
             del opts['__name__']
         return list(opts.keys())
 
-    def read(self, filenames):
+    def read(self, filenames, encoding=None):
         """Read and parse a filename or a list of filenames.
 
         Files that cannot be opened are silently ignored; this is
@@ -291,7 +386,7 @@
         read_ok = []
         for filename in filenames:
             try:
-                fp = open(filename)
+                fp = open(filename, encoding=encoding)
             except IOError:
                 continue
             self._read(fp, filename)
@@ -306,7 +401,6 @@
         second argument is the `filename', which if not given, is
         taken from fp.name.  If fp has no `name' attribute, `<???>' is
         used.
-
         """
         if filename is None:
             try:
@@ -367,6 +461,7 @@
 
     def has_option(self, section, option):
         """Check for the existence of a given option in a given section."""
+
         if not section or section == DEFAULTSECT:
             option = self.optionxform(option)
             return option in self._defaults
@@ -379,6 +474,7 @@
 
     def set(self, section, option, value=None):
         """Set an option."""
+
         if not section or section == DEFAULTSECT:
             sectdict = self._defaults
         else:
@@ -388,23 +484,34 @@
                 raise NoSectionError(section)
         sectdict[self.optionxform(option)] = value
 
-    def write(self, fp):
-        """Write an .ini-format representation of the configuration state."""
+    def write(self, fp, space_around_delimiters=True):
+        """Write an .ini-format representation of the configuration state.
+
+        If `space_around_delimiters' is True (the default), delimiters
+        between keys and values are surrounded by spaces.
+        """
+        if space_around_delimiters:
+            d = " {} ".format(self._delimiters[0])
+        else:
+            d = self._delimiters[0]
         if self._defaults:
-            fp.write("[%s]\n" % DEFAULTSECT)
-            for (key, value) in self._defaults.items():
-                fp.write("%s = %s\n" % (key, str(value).replace('\n', '\n\t')))
-            fp.write("\n")
+            self._write_section(fp, DEFAULTSECT, self._defaults.items(), d)
         for section in self._sections:
-            fp.write("[%s]\n" % section)
-            for (key, value) in self._sections[section].items():
-                if key != "__name__":
-                    if value is None:
-                        fp.write("%s\n" % (key))
-                    else:
-                        fp.write("%s = %s\n" %
-                                 (key, str(value).replace('\n', '\n\t')))
-            fp.write("\n")
+            self._write_section(fp, section,
+                                self._sections[section].items(), d)
+
+    def _write_section(self, fp, section_name, section_items, delimiter):
+        """Write a single section to the specified `fp'."""
+        fp.write("[{}]\n".format(section_name))
+        for key, value in section_items:
+            if key == "__name__":
+                continue
+            if value is not None:
+                value = delimiter + str(value).replace('\n', '\n\t')
+            else:
+                value = ""
+            fp.write("{}{}\n".format(key, value))
+        fp.write("\n")
 
     def remove_option(self, section, option):
         """Remove an option."""
@@ -428,66 +535,66 @@
             del self._sections[section]
         return existed
 
-    #
-    # Regular expressions for parsing section headers and options.
-    #
-    SECTCRE = re.compile(
-        r'\['                                 # [
-        r'(?P<header>[^]]+)'                  # very permissive!
-        r'\]'                                 # ]
-        )
-    OPTCRE = re.compile(
-        r'(?P<option>[^:=\s][^:=]*)'          # very permissive!
-        r'\s*(?P<vi>[:=])\s*'                 # any number of space/tab,
-                                              # followed by separator
-                                              # (either : or =), followed
-                                              # by any # space/tab
-        r'(?P<value>.*)$'                     # everything up to eol
-        )
-    OPTCRE_NV = re.compile(
-        r'(?P<option>[^:=\s][^:=]*)'          # very permissive!
-        r'\s*(?:'                             # any number of space/tab,
-        r'(?P<vi>[:=])\s*'                    # optionally followed by
-                                              # separator (either : or
-                                              # =), followed by any #
-                                              # space/tab
-        r'(?P<value>.*))?$'                   # everything up to eol
-        )
-
     def _read(self, fp, fpname):
-        """Parse a sectioned setup file.
+        """Parse a sectioned configuration file.
 
-        The sections in setup file contains a title line at the top,
-        indicated by a name in square brackets (`[]'), plus key/value
-        options lines, indicated by `name: value' format lines.
-        Continuations are represented by an embedded newline then
-        leading whitespace.  Blank lines, lines beginning with a '#',
-        and just about everything else are ignored.
+        Each section in a configuration file contains a header, indicated by a
+        name in square brackets (`[]'), plus key/value options, indicated by
+        `name' and `value' delimited with a specific substring (`=' or `:' by
+        default).
+
+        Values can span multiple lines, as long as they are indented deeper than
+        the first line of the value. Depending on the parser's mode, blank lines
+        may be treated as parts of multiline values or ignored.
+
+        Configuration files may include comments, prefixed by specific
+        characters (`#' and `;' by default). Comments may appear on their own in
+        an otherwise empty line or may be entered in lines holding values or
+        section names.
         """
-        cursect = None                            # None, or a dictionary
+        cursect = None                        # None, or a dictionary
         optname = None
         lineno = 0
-        e = None                                  # None, or an exception
-        while True:
-            line = fp.readline()
-            if not line:
-                break
-            lineno = lineno + 1
-            # comment or blank line?
-            if line.strip() == '' or line[0] in '#;':
-                continue
-            if line.split(None, 1)[0].lower() == 'rem' and line[0] in "rR":
-                # no leading whitespace
+        indent_level = 0
+        e = None                              # None, or an exception
+        for lineno, line in enumerate(fp, start=1):
+            # strip full line comments
+            comment_start = None
+            for prefix in self._startonly_comment_prefixes:
+                if line.strip().startswith(prefix):
+                    comment_start = 0
+                    break
+            # strip inline comments
+            for prefix in self._comment_prefixes:
+                index = line.find(prefix)
+                if index == 0 or (index > 0 and line[index-1].isspace()):
+                    comment_start = index
+                    break
+            value = line[:comment_start].strip()
+            if not value:
+                if self._empty_lines_in_values:
+                    # add empty line to the value, but only if there was no
+                    # comment on the line
+                    if (comment_start is None and
+                        cursect is not None and
+                        optname and
+                        cursect[optname] is not None):
+                        cursect[optname].append('') # newlines added at join
+                else:
+                    # empty line marks end of value
+                    indent_level = sys.maxsize
                 continue
             # continuation line?
-            if line[0].isspace() and cursect is not None and optname:
-                value = line.strip()
-                if value:
-                    cursect[optname] = "%s\n%s" % (cursect[optname], value)
+            first_nonspace = self.NONSPACECRE.search(line)
+            cur_indent_level = first_nonspace.start() if first_nonspace else 0
+            if (cursect is not None and optname and
+                cur_indent_level > indent_level):
+                cursect[optname].append(value)
             # a section header or option header?
             else:
+                indent_level = cur_indent_level
                 # is it a section header?
-                mo = self.SECTCRE.match(line)
+                mo = self.SECTCRE.match(value)
                 if mo:
                     sectname = mo.group('header')
                     if sectname in self._sections:
@@ -505,47 +612,61 @@
                     raise MissingSectionHeaderError(fpname, lineno, line)
                 # an option line?
                 else:
-                    mo = self._optcre.match(line)
+                    mo = self._optcre.match(value)
                     if mo:
                         optname, vi, optval = mo.group('option', 'vi', 'value')
+                        if not optname:
+                            e = self._handle_error(e, fpname, lineno, line)
+                        optname = self.optionxform(optname.rstrip())
                         # This check is fine because the OPTCRE cannot
                         # match if it would set optval to None
                         if optval is not None:
-                            if vi in ('=', ':') and ';' in optval:
-                                # ';' is a comment delimiter only if it follows
-                                # a spacing character
-                                pos = optval.find(';')
-                                if pos != -1 and optval[pos-1].isspace():
-                                    optval = optval[:pos]
                             optval = optval.strip()
-                        # allow empty values
-                        if optval == '""':
-                            optval = ''
-                        optname = self.optionxform(optname.rstrip())
-                        cursect[optname] = optval
+                            # allow empty values
+                            if optval == '""':
+                                optval = ''
+                            cursect[optname] = [optval]
+                        else:
+                            # valueless option handling
+                            cursect[optname] = optval
                     else:
-                        # a non-fatal parsing error occurred.  set up the
+                        # a non-fatal parsing error occurred. set up the
                         # exception but keep going. the exception will be
                         # raised at the end of the file and will contain a
                         # list of all bogus lines
-                        if not e:
-                            e = ParsingError(fpname)
-                        e.append(lineno, repr(line))
+                        e = self._handle_error(e, fpname, lineno, line)
         # if any parsing errors occurred, raise an exception
         if e:
             raise e
+        self._join_multiline_values()
+
+    def _join_multiline_values(self):
+        all_sections = [self._defaults]
+        all_sections.extend(self._sections.values())
+        for options in all_sections:
+            for name, val in options.items():
+                if isinstance(val, list):
+                    options[name] = '\n'.join(val).rstrip()
+
+    def _handle_error(self, exc, fpname, lineno, line):
+        if not exc:
+            exc = ParsingError(fpname)
+        exc.append(lineno, repr(line))
+        return exc
 
 
 class ConfigParser(RawConfigParser):
+    """ConfigParser implementing interpolation."""
 
     def get(self, section, option, raw=False, vars=None):
         """Get an option value for a given section.
 
-        All % interpolations are expanded in the return values, based on the
-        defaults passed into the constructor, unless the optional argument
-        `raw' is true.  Additional substitutions may be provided using the
-        `vars' argument, which must be a dictionary whose contents overrides
-        any pre-existing defaults.
+        If `vars' is provided, it must be a dictionary. The option is looked up
+        in `vars' (if provided), `section', and in `defaults' in that order.
+
+        All % interpolations are expanded in the return values, unless the
+        optional argument `raw' is true.  Values for interpolation keys are
+        looked up in the same manner as the option.
 
         The section DEFAULT is special.
         """
@@ -632,6 +753,7 @@
 
 
 class SafeConfigParser(ConfigParser):
+    """ConfigParser implementing sane interpolation."""
 
     def _interpolate(self, section, option, rawval, vars):
         # do the string interpolation

Modified: python/branches/import_unicode/Lib/ctypes/__init__.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/__init__.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/__init__.py	Fri Jul 30 01:19:37 2010
@@ -259,41 +259,31 @@
 
 from _ctypes import POINTER, pointer, _pointer_type_cache
 
-try:
-    from _ctypes import set_conversion_mode
-except ImportError:
-    pass
-else:
-    if _os.name in ("nt", "ce"):
-        set_conversion_mode("mbcs", "strict")
-    else:
-        set_conversion_mode("ascii", "strict")
-
-    class c_wchar_p(_SimpleCData):
-        _type_ = "Z"
+class c_wchar_p(_SimpleCData):
+    _type_ = "Z"
 
-    class c_wchar(_SimpleCData):
-        _type_ = "u"
+class c_wchar(_SimpleCData):
+    _type_ = "u"
 
-    POINTER(c_wchar).from_param = c_wchar_p.from_param #_SimpleCData.c_wchar_p_from_param
+POINTER(c_wchar).from_param = c_wchar_p.from_param #_SimpleCData.c_wchar_p_from_param
 
-    def create_unicode_buffer(init, size=None):
-        """create_unicode_buffer(aString) -> character array
-        create_unicode_buffer(anInteger) -> character array
-        create_unicode_buffer(aString, anInteger) -> character array
-        """
-        if isinstance(init, (str, bytes)):
-            if size is None:
-                size = len(init)+1
-            buftype = c_wchar * size
-            buf = buftype()
-            buf.value = init
-            return buf
-        elif isinstance(init, int):
-            buftype = c_wchar * init
-            buf = buftype()
-            return buf
-        raise TypeError(init)
+def create_unicode_buffer(init, size=None):
+    """create_unicode_buffer(aString) -> character array
+    create_unicode_buffer(anInteger) -> character array
+    create_unicode_buffer(aString, anInteger) -> character array
+    """
+    if isinstance(init, (str, bytes)):
+        if size is None:
+            size = len(init)+1
+        buftype = c_wchar * size
+        buf = buftype()
+        buf.value = init
+        return buf
+    elif isinstance(init, int):
+        buftype = c_wchar * init
+        buf = buftype()
+        return buf
+    raise TypeError(init)
 
 POINTER(c_char).from_param = c_char_p.from_param #_SimpleCData.c_char_p_from_param
 

Modified: python/branches/import_unicode/Lib/ctypes/test/test_arrays.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/test/test_arrays.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/test/test_arrays.py	Fri Jul 30 01:19:37 2010
@@ -42,7 +42,7 @@
 
         CharArray = ARRAY(c_char, 3)
 
-        ca = CharArray("a", "b", "c")
+        ca = CharArray(b"a", b"b", b"c")
 
         # Should this work? It doesn't:
         # CharArray("abc")
@@ -89,7 +89,7 @@
 
     def test_from_address(self):
         # Failed with 0.9.8, reported by JUrner
-        p = create_string_buffer("foo")
+        p = create_string_buffer(b"foo")
         sz = (c_char * 3).from_address(addressof(p))
         self.assertEqual(sz[:], b"foo")
         self.assertEqual(sz[::], b"foo")

Modified: python/branches/import_unicode/Lib/ctypes/test/test_bitfields.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/test/test_bitfields.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/test/test_bitfields.py	Fri Jul 30 01:19:37 2010
@@ -37,14 +37,14 @@
             for name in "ABCDEFGHI":
                 b = BITS()
                 setattr(b, name, i)
-                self.assertEqual((name, i, getattr(b, name)), (name, i, func(byref(b), name)))
+                self.assertEqual(getattr(b, name), func(byref(b), name.encode('ascii')))
 
     def test_shorts(self):
         for i in range(256):
             for name in "MNOPQRS":
                 b = BITS()
                 setattr(b, name, i)
-                self.assertEqual((name, i, getattr(b, name)), (name, i, func(byref(b), name)))
+                self.assertEqual(getattr(b, name), func(byref(b), name.encode('ascii')))
 
 signed_int_types = (c_byte, c_short, c_int, c_long, c_longlong)
 unsigned_int_types = (c_ubyte, c_ushort, c_uint, c_ulong, c_ulonglong)

Modified: python/branches/import_unicode/Lib/ctypes/test/test_buffers.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/test/test_buffers.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/test/test_buffers.py	Fri Jul 30 01:19:37 2010
@@ -9,19 +9,7 @@
         self.assertEqual(sizeof(b), 32 * sizeof(c_char))
         self.assertTrue(type(b[0]) is bytes)
 
-        b = create_string_buffer("abc")
-        self.assertEqual(len(b), 4) # trailing nul char
-        self.assertEqual(sizeof(b), 4 * sizeof(c_char))
-        self.assertTrue(type(b[0]) is bytes)
-        self.assertEqual(b[0], b"a")
-        self.assertEqual(b[:], b"abc\0")
-        self.assertEqual(b[::], b"abc\0")
-        self.assertEqual(b[::-1], b"\0cba")
-        self.assertEqual(b[::2], b"ac")
-        self.assertEqual(b[::5], b"a")
-
-    def test_string_conversion(self):
-        b = create_string_buffer("abc")
+        b = create_string_buffer(b"abc")
         self.assertEqual(len(b), 4) # trailing nul char
         self.assertEqual(sizeof(b), 4 * sizeof(c_char))
         self.assertTrue(type(b[0]) is bytes)

Modified: python/branches/import_unicode/Lib/ctypes/test/test_bytes.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/test/test_bytes.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/test/test_bytes.py	Fri Jul 30 01:19:37 2010
@@ -11,24 +11,21 @@
         (c_char * 3)(b"a", b"b", b"c")
 
     def test_c_wchar(self):
-        x = c_wchar(b"x")
-        x.value = b"y"
-        c_wchar.from_param(b"x")
-        (c_wchar * 3)(b"a", b"b", b"c")
+        x = c_wchar("x")
+        x.value = "y"
+        c_wchar.from_param("x")
+        (c_wchar * 3)("a", "b", "c")
 
     def test_c_char_p(self):
-        c_char_p("foo bar")
         c_char_p(b"foo bar")
 
     def test_c_wchar_p(self):
         c_wchar_p("foo bar")
-        c_wchar_p(b"foo bar")
 
     def test_struct(self):
         class X(Structure):
             _fields_ = [("a", c_char * 3)]
 
-        X("abc")
         x = X(b"abc")
         self.assertEqual(x.a, b"abc")
         self.assertEqual(type(x.a), bytes)
@@ -37,8 +34,7 @@
         class X(Structure):
             _fields_ = [("a", c_wchar * 3)]
 
-        X("abc")
-        x = X(b"abc")
+        x = X("abc")
         self.assertEqual(x.a, "abc")
         self.assertEqual(type(x.a), str)
 

Modified: python/branches/import_unicode/Lib/ctypes/test/test_callbacks.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/test/test_callbacks.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/test/test_callbacks.py	Fri Jul 30 01:19:37 2010
@@ -164,7 +164,7 @@
         result = integrate(0.0, 1.0, CALLBACK(func), 10)
         diff = abs(result - 1./3.)
 
-        self.assertTrue(diff < 0.01, "%s not less than 0.01" % diff)
+        self.assertLess(diff, 0.01, "%s not less than 0.01" % diff)
 
     def test_issue_8959_a(self):
         from ctypes.util import find_library

Modified: python/branches/import_unicode/Lib/ctypes/test/test_cast.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/test/test_cast.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/test/test_cast.py	Fri Jul 30 01:19:37 2010
@@ -33,17 +33,17 @@
     def test_p2a_objects(self):
         array = (c_char_p * 5)()
         self.assertEqual(array._objects, None)
-        array[0] = "foo bar"
+        array[0] = b"foo bar"
         self.assertEqual(array._objects, {'0': b"foo bar"})
 
         p = cast(array, POINTER(c_char_p))
         # array and p share a common _objects attribute
         self.assertTrue(p._objects is array._objects)
         self.assertEqual(array._objects, {'0': b"foo bar", id(array): array})
-        p[0] = "spam spam"
+        p[0] = b"spam spam"
         self.assertEqual(p._objects, {'0': b"spam spam", id(array): array})
         self.assertTrue(array._objects is p._objects)
-        p[1] = "foo bar"
+        p[1] = b"foo bar"
         self.assertEqual(p._objects, {'1': b'foo bar', '0': b"spam spam", id(array): array})
         self.assertTrue(array._objects is p._objects)
 
@@ -71,7 +71,7 @@
 
     def test_char_p(self):
         # This didn't work: bad argument to internal function
-        s = c_char_p("hiho")
+        s = c_char_p(b"hiho")
         self.assertEqual(cast(cast(s, c_void_p), c_char_p).value,
                              b"hiho")
 

Modified: python/branches/import_unicode/Lib/ctypes/test/test_cfuncs.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/test/test_cfuncs.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/test/test_cfuncs.py	Fri Jul 30 01:19:37 2010
@@ -107,7 +107,7 @@
     def test_ulong_plus(self):
         self._dll.tf_bL.restype = c_ulong
         self._dll.tf_bL.argtypes = (c_char, c_ulong)
-        self.assertEqual(self._dll.tf_bL(' ', 4294967295), 1431655765)
+        self.assertEqual(self._dll.tf_bL(b' ', 4294967295), 1431655765)
         self.assertEqual(self.U(), 4294967295)
 
     def test_longlong(self):

Modified: python/branches/import_unicode/Lib/ctypes/test/test_errno.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/test/test_errno.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/test/test_errno.py	Fri Jul 30 01:19:37 2010
@@ -19,7 +19,7 @@
 
         libc_open.argtypes = c_char_p, c_int
 
-        self.assertEqual(libc_open("", 0), -1)
+        self.assertEqual(libc_open(b"", 0), -1)
         self.assertEqual(get_errno(), errno.ENOENT)
 
         self.assertEqual(set_errno(32), errno.ENOENT)
@@ -35,7 +35,7 @@
                 else:
                     libc_open = libc.open
                 libc_open.argtypes = c_char_p, c_int
-                self.assertEqual(libc_open("", 0), -1)
+                self.assertEqual(libc_open(b"", 0), -1)
                 self.assertEqual(get_errno(), 0)
 
             t = threading.Thread(target=_worker)

Modified: python/branches/import_unicode/Lib/ctypes/test/test_internals.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/test/test_internals.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/test/test_internals.py	Fri Jul 30 01:19:37 2010
@@ -70,19 +70,17 @@
         class Y(Structure):
             _fields_ = [("x", X), ("y", X)]
 
-        s1 = "Hello, World"
-        s2 = "Hallo, Welt"
+        s1 = b"Hello, World"
+        s2 = b"Hallo, Welt"
 
         x = X()
         x.a = s1
         x.b = s2
-        self.assertEqual(x._objects, {"0": bytes(s1, "ascii"),
-                                          "1": bytes(s2, "ascii")})
+        self.assertEqual(x._objects, {"0": s1, "1": s2})
 
         y = Y()
         y.x = x
-        self.assertEqual(y._objects, {"0": {"0": bytes(s1, "ascii"),
-                                                "1": bytes(s2, "ascii")}})
+        self.assertEqual(y._objects, {"0": {"0": s1, "1": s2}})
 ##        x = y.x
 ##        del y
 ##        print x._b_base_._objects

Modified: python/branches/import_unicode/Lib/ctypes/test/test_keeprefs.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/test/test_keeprefs.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/test/test_keeprefs.py	Fri Jul 30 01:19:37 2010
@@ -13,9 +13,9 @@
     def test_ccharp(self):
         x = c_char_p()
         self.assertEquals(x._objects, None)
-        x.value = "abc"
+        x.value = b"abc"
         self.assertEquals(x._objects, b"abc")
-        x = c_char_p("spam")
+        x = c_char_p(b"spam")
         self.assertEquals(x._objects, b"spam")
 
 class StructureTestCase(unittest.TestCase):
@@ -37,8 +37,8 @@
         x = X()
         self.assertEquals(x._objects, None)
 
-        x.a = "spam"
-        x.b = "foo"
+        x.a = b"spam"
+        x.b = b"foo"
         self.assertEquals(x._objects, {"0": b"spam", "1": b"foo"})
 
     def test_struct_struct(self):

Modified: python/branches/import_unicode/Lib/ctypes/test/test_libc.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/test/test_libc.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/test/test_libc.py	Fri Jul 30 01:19:37 2010
@@ -25,7 +25,7 @@
         def sort(a, b):
             return three_way_cmp(a[0], b[0])
 
-        chars = create_string_buffer("spam, spam, and spam")
+        chars = create_string_buffer(b"spam, spam, and spam")
         lib.my_qsort(chars, len(chars)-1, sizeof(c_char), comparefunc(sort))
         self.assertEqual(chars.raw, b"   ,,aaaadmmmnpppsss\x00")
 

Modified: python/branches/import_unicode/Lib/ctypes/test/test_objects.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/test/test_objects.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/test/test_objects.py	Fri Jul 30 01:19:37 2010
@@ -20,7 +20,7 @@
 The memory block stores pointers to strings, and the strings itself
 assigned from Python must be kept.
 
->>> array[4] = 'foo bar'
+>>> array[4] = b'foo bar'
 >>> array._objects
 {'4': b'foo bar'}
 >>> array[4]
@@ -45,7 +45,7 @@
 <ctypes.test.test_objects.X object at 0x...>
 >>>
 
->>> x.array[0] = 'spam spam spam'
+>>> x.array[0] = b'spam spam spam'
 >>> x._objects
 {'0:2': b'spam spam spam'}
 >>> x.array._b_base_._objects

Modified: python/branches/import_unicode/Lib/ctypes/test/test_parameters.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/test/test_parameters.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/test/test_parameters.py	Fri Jul 30 01:19:37 2010
@@ -19,7 +19,6 @@
         else:
             set_conversion_mode(*self.prev_conv_mode)
 
-
     def test_subclasses(self):
         from ctypes import c_void_p, c_char_p
         # ctypes 0.9.5 and before did overwrite from_param in SimpleType_new
@@ -58,14 +57,13 @@
         self.assertTrue(c_char_p.from_param(s)._obj is s)
 
         # new in 0.9.1: convert (encode) unicode to ascii
-        self.assertEqual(c_char_p.from_param("123")._obj, b"123")
-        self.assertRaises(UnicodeEncodeError, c_char_p.from_param, "123\377")
-
+        self.assertEqual(c_char_p.from_param(b"123")._obj, b"123")
+        self.assertRaises(TypeError, c_char_p.from_param, "123\377")
         self.assertRaises(TypeError, c_char_p.from_param, 42)
 
         # calling c_char_p.from_param with a c_char_p instance
         # returns the argument itself:
-        a = c_char_p("123")
+        a = c_char_p(b"123")
         self.assertTrue(c_char_p.from_param(a) is a)
 
     def test_cw_strings(self):
@@ -82,7 +80,7 @@
 
             # new in 0.9.1: convert (decode) ascii to unicode
             self.assertEqual(c_wchar_p.from_param("123")._obj, "123")
-        self.assertRaises(UnicodeDecodeError, c_wchar_p.from_param, b"123\377")
+        self.assertRaises(TypeError, c_wchar_p.from_param, b"123\377")
 
         pa = c_wchar_p.from_param(c_wchar_p("123"))
         self.assertEqual(type(pa), c_wchar_p)

Modified: python/branches/import_unicode/Lib/ctypes/test/test_prototypes.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/test/test_prototypes.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/test/test_prototypes.py	Fri Jul 30 01:19:37 2010
@@ -127,7 +127,7 @@
         self.assertEqual(None, func(c_char_p(None)))
 
         self.assertEqual(b"123", func(c_buffer(b"123")))
-        ca = c_char("a")
+        ca = c_char(b"a")
         self.assertEqual(ord(b"a"), func(pointer(ca))[0])
         self.assertEqual(ord(b"a"), func(byref(ca))[0])
 

Modified: python/branches/import_unicode/Lib/ctypes/test/test_python_api.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/test/test_python_api.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/test/test_python_api.py	Fri Jul 30 01:19:37 2010
@@ -72,10 +72,10 @@
         PyOS_snprintf.argtypes = POINTER(c_char), c_size_t, c_char_p
 
         buf = c_buffer(256)
-        PyOS_snprintf(buf, sizeof(buf), "Hello from %s", b"ctypes")
+        PyOS_snprintf(buf, sizeof(buf), b"Hello from %s", b"ctypes")
         self.assertEqual(buf.value, b"Hello from ctypes")
 
-        PyOS_snprintf(buf, sizeof(buf), "Hello from %s (%d, %d, %d)", b"ctypes", 1, 2, 3)
+        PyOS_snprintf(buf, sizeof(buf), b"Hello from %s (%d, %d, %d)", b"ctypes", 1, 2, 3)
         self.assertEqual(buf.value, b"Hello from ctypes (1, 2, 3)")
 
         # not enough arguments

Modified: python/branches/import_unicode/Lib/ctypes/test/test_random_things.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/test/test_random_things.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/test/test_random_things.py	Fri Jul 30 01:19:37 2010
@@ -66,7 +66,7 @@
 
     def test_TypeErrorDivisionError(self):
         cb = CFUNCTYPE(c_int, c_char_p)(callback_func)
-        out = self.capture_stderr(cb, "spam")
+        out = self.capture_stderr(cb, b"spam")
         self.assertEqual(out.splitlines()[-1],
                              "TypeError: "
                              "unsupported operand type(s) for /: 'int' and 'bytes'")

Modified: python/branches/import_unicode/Lib/ctypes/test/test_repr.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/test/test_repr.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/test/test_repr.py	Fri Jul 30 01:19:37 2010
@@ -22,8 +22,8 @@
             self.assertEqual("<X object at", repr(typ(42))[:12])
 
     def test_char(self):
-        self.assertEqual("c_char(b'x')", repr(c_char('x')))
-        self.assertEqual("<X object at", repr(X('x'))[:12])
+        self.assertEqual("c_char(b'x')", repr(c_char(b'x')))
+        self.assertEqual("<X object at", repr(X(b'x'))[:12])
 
 if __name__ == "__main__":
     unittest.main()

Modified: python/branches/import_unicode/Lib/ctypes/test/test_returnfuncptrs.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/test/test_returnfuncptrs.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/test/test_returnfuncptrs.py	Fri Jul 30 01:19:37 2010
@@ -28,10 +28,10 @@
         # _CFuncPtr instances are now callable with an integer argument
         # which denotes a function address:
         strchr = CFUNCTYPE(c_char_p, c_char_p, c_char)(addr)
-        self.assertTrue(strchr("abcdef", "b"), "bcdef")
-        self.assertEqual(strchr("abcdef", "x"), None)
-        self.assertRaises(ArgumentError, strchr, "abcdef", 3.0)
-        self.assertRaises(TypeError, strchr, "abcdef")
+        self.assertTrue(strchr(b"abcdef", b"b"), "bcdef")
+        self.assertEqual(strchr(b"abcdef", b"x"), None)
+        self.assertRaises(ArgumentError, strchr, b"abcdef", 3.0)
+        self.assertRaises(TypeError, strchr, b"abcdef")
 
 if __name__ == "__main__":
     unittest.main()

Modified: python/branches/import_unicode/Lib/ctypes/test/test_stringptr.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/test/test_stringptr.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/test/test_stringptr.py	Fri Jul 30 01:19:37 2010
@@ -14,7 +14,7 @@
 
         # NULL pointer access
         self.assertRaises(ValueError, getattr, x.str, "contents")
-        b = c_buffer("Hello, World")
+        b = c_buffer(b"Hello, World")
         from sys import getrefcount as grc
         self.assertEqual(grc(b), 2)
         x.str = b
@@ -63,8 +63,8 @@
         # So we must keep a reference to buf separately
 
         strchr.restype = POINTER(c_char)
-        buf = c_buffer("abcdef")
-        r = strchr(buf, "c")
+        buf = c_buffer(b"abcdef")
+        r = strchr(buf, b"c")
         x = r[0], r[1], r[2], r[3], r[4]
         self.assertEqual(x, (b"c", b"d", b"e", b"f", b"\000"))
         del buf

Modified: python/branches/import_unicode/Lib/ctypes/test/test_strings.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/test/test_strings.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/test/test_strings.py	Fri Jul 30 01:19:37 2010
@@ -5,23 +5,23 @@
     def test(self):
         BUF = c_char * 4
 
-        buf = BUF("a", "b", "c")
+        buf = BUF(b"a", b"b", b"c")
         self.assertEqual(buf.value, b"abc")
         self.assertEqual(buf.raw, b"abc\000")
 
-        buf.value = "ABCD"
+        buf.value = b"ABCD"
         self.assertEqual(buf.value, b"ABCD")
         self.assertEqual(buf.raw, b"ABCD")
 
-        buf.value = "x"
+        buf.value = b"x"
         self.assertEqual(buf.value, b"x")
         self.assertEqual(buf.raw, b"x\000CD")
 
-        buf[1] = "Z"
+        buf[1] = b"Z"
         self.assertEqual(buf.value, b"xZCD")
         self.assertEqual(buf.raw, b"xZCD")
 
-        self.assertRaises(ValueError, setattr, buf, "value", "aaaaaaaa")
+        self.assertRaises(ValueError, setattr, buf, "value", b"aaaaaaaa")
         self.assertRaises(TypeError, setattr, buf, "value", 42)
 
     def test_c_buffer_value(self):

Modified: python/branches/import_unicode/Lib/ctypes/test/test_structures.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/test/test_structures.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/test/test_structures.py	Fri Jul 30 01:19:37 2010
@@ -205,15 +205,15 @@
                         ("age", c_int)]
 
         self.assertRaises(TypeError, Person, 42)
-        self.assertRaises(ValueError, Person, "asldkjaslkdjaslkdj")
+        self.assertRaises(ValueError, Person, b"asldkjaslkdjaslkdj")
         self.assertRaises(TypeError, Person, "Name", "HI")
 
         # short enough
-        self.assertEqual(Person("12345", 5).name, b"12345")
+        self.assertEqual(Person(b"12345", 5).name, b"12345")
         # exact fit
-        self.assertEqual(Person("123456", 5).name, b"123456")
+        self.assertEqual(Person(b"123456", 5).name, b"123456")
         # too long
-        self.assertRaises(ValueError, Person, "1234567", 5)
+        self.assertRaises(ValueError, Person, b"1234567", 5)
 
     def test_conflicting_initializers(self):
         class POINT(Structure):
@@ -267,7 +267,7 @@
                         ("phone", Phone),
                         ("age", c_int)]
 
-        p = Person("Someone", ("1234", "5678"), 5)
+        p = Person(b"Someone", (b"1234", b"5678"), 5)
 
         self.assertEqual(p.name, b"Someone")
         self.assertEqual(p.phone.areacode, b"1234")
@@ -284,8 +284,8 @@
             _fields_ = [("name", c_wchar * 12),
                         ("age", c_int)]
 
-        p = PersonW("Someone")
-        self.assertEqual(p.name, "Someone")
+        p = PersonW("Someone \xe9")
+        self.assertEqual(p.name, "Someone \xe9")
 
         self.assertEqual(PersonW("1234567890").name, "1234567890")
         self.assertEqual(PersonW("12345678901").name, "12345678901")
@@ -304,13 +304,13 @@
                         ("phone", Phone),
                         ("age", c_int)]
 
-        cls, msg = self.get_except(Person, "Someone", (1, 2))
+        cls, msg = self.get_except(Person, b"Someone", (1, 2))
         self.assertEqual(cls, RuntimeError)
         self.assertEqual(msg,
                              "(Phone) <class 'TypeError'>: "
                              "expected string, int found")
 
-        cls, msg = self.get_except(Person, "Someone", ("a", "b", "c"))
+        cls, msg = self.get_except(Person, b"Someone", (b"a", b"b", b"c"))
         self.assertEqual(cls, RuntimeError)
         if issubclass(Exception, object):
             self.assertEqual(msg,

Modified: python/branches/import_unicode/Lib/ctypes/test/test_unicode.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/test/test_unicode.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/test/test_unicode.py	Fri Jul 30 01:19:37 2010
@@ -7,122 +7,53 @@
     pass
 else:
     import _ctypes_test
-    dll = ctypes.CDLL(_ctypes_test.__file__)
-    wcslen = dll.my_wcslen
-    wcslen.argtypes = [ctypes.c_wchar_p]
-
 
     class UnicodeTestCase(unittest.TestCase):
-        def setUp(self):
-            self.prev_conv_mode = ctypes.set_conversion_mode("ascii", "strict")
+        def test_wcslen(self):
+            dll = ctypes.CDLL(_ctypes_test.__file__)
+            wcslen = dll.my_wcslen
+            wcslen.argtypes = [ctypes.c_wchar_p]
 
-        def tearDown(self):
-            ctypes.set_conversion_mode(*self.prev_conv_mode)
-
-        def test_ascii_strict(self):
-            ctypes.set_conversion_mode("ascii", "strict")
-            # no conversions take place with unicode arguments
             self.assertEqual(wcslen("abc"), 3)
             self.assertEqual(wcslen("ab\u2070"), 3)
-            # string args are converted
-            self.assertEqual(wcslen("abc"), 3)
             self.assertRaises(ctypes.ArgumentError, wcslen, b"ab\xe4")
 
-        def test_ascii_replace(self):
-            ctypes.set_conversion_mode("ascii", "replace")
-            self.assertEqual(wcslen("abc"), 3)
-            self.assertEqual(wcslen("ab\u2070"), 3)
-            self.assertEqual(wcslen("abc"), 3)
-            self.assertEqual(wcslen("ab\xe4"), 3)
-
-        def test_ascii_ignore(self):
-            ctypes.set_conversion_mode("ascii", "ignore")
-            self.assertEqual(wcslen("abc"), 3)
-            self.assertEqual(wcslen("ab\u2070"), 3)
-            # ignore error mode skips non-ascii characters
-            self.assertEqual(wcslen("abc"), 3)
-            self.assertEqual(wcslen(b"\xe4\xf6\xfc\xdf"), 0)
-
-        def test_latin1_strict(self):
-            ctypes.set_conversion_mode("latin-1", "strict")
-            self.assertEqual(wcslen("abc"), 3)
-            self.assertEqual(wcslen("ab\u2070"), 3)
-            self.assertEqual(wcslen("abc"), 3)
-            self.assertEqual(wcslen("\xe4\xf6\xfc\xdf"), 4)
-
         def test_buffers(self):
-            ctypes.set_conversion_mode("ascii", "strict")
             buf = ctypes.create_unicode_buffer("abc")
             self.assertEqual(len(buf), 3+1)
 
-            ctypes.set_conversion_mode("ascii", "replace")
-            buf = ctypes.create_unicode_buffer(b"ab\xe4\xf6\xfc")
-            self.assertEqual(buf[:], "ab\uFFFD\uFFFD\uFFFD\0")
-            self.assertEqual(buf[::], "ab\uFFFD\uFFFD\uFFFD\0")
-            self.assertEqual(buf[::-1], "\0\uFFFD\uFFFD\uFFFDba")
-            self.assertEqual(buf[::2], "a\uFFFD\uFFFD")
+            buf = ctypes.create_unicode_buffer("ab\xe4\xf6\xfc")
+            self.assertEqual(buf[:], "ab\xe4\xf6\xfc\0")
+            self.assertEqual(buf[::], "ab\xe4\xf6\xfc\0")
+            self.assertEqual(buf[::-1], '\x00\xfc\xf6\xe4ba')
+            self.assertEqual(buf[::2], 'a\xe4\xfc')
             self.assertEqual(buf[6:5:-1], "")
 
-            ctypes.set_conversion_mode("ascii", "ignore")
-            buf = ctypes.create_unicode_buffer(b"ab\xe4\xf6\xfc")
-            # is that correct? not sure.  But with 'ignore', you get what you pay for..
-            self.assertEqual(buf[:], "ab\0\0\0\0")
-            self.assertEqual(buf[::], "ab\0\0\0\0")
-            self.assertEqual(buf[::-1], "\0\0\0\0ba")
-            self.assertEqual(buf[::2], "a\0\0")
-            self.assertEqual(buf[6:5:-1], "")
-
-    import _ctypes_test
     func = ctypes.CDLL(_ctypes_test.__file__)._testfunc_p_p
 
     class StringTestCase(UnicodeTestCase):
         def setUp(self):
-            self.prev_conv_mode = ctypes.set_conversion_mode("ascii", "strict")
             func.argtypes = [ctypes.c_char_p]
             func.restype = ctypes.c_char_p
 
         def tearDown(self):
-            ctypes.set_conversion_mode(*self.prev_conv_mode)
             func.argtypes = None
             func.restype = ctypes.c_int
 
-        def test_ascii_replace(self):
-            ctypes.set_conversion_mode("ascii", "strict")
-            self.assertEqual(func("abc"), "abc")
-            self.assertEqual(func("abc"), "abc")
-            self.assertRaises(ctypes.ArgumentError, func, "ab\xe4")
-
-        def test_ascii_ignore(self):
-            ctypes.set_conversion_mode("ascii", "ignore")
-            self.assertEqual(func("abc"), b"abc")
-            self.assertEqual(func("abc"), b"abc")
-            self.assertEqual(func("\xe4\xf6\xfc\xdf"), b"")
-
-        def test_ascii_replace(self):
-            ctypes.set_conversion_mode("ascii", "replace")
-            self.assertEqual(func("abc"), b"abc")
-            self.assertEqual(func("abc"), b"abc")
-            self.assertEqual(func("\xe4\xf6\xfc\xdf"), b"????")
+        def test_func(self):
+            self.assertEqual(func(b"abc\xe4"), b"abc\xe4")
 
         def test_buffers(self):
-            ctypes.set_conversion_mode("ascii", "strict")
-            buf = ctypes.create_string_buffer("abc")
+            buf = ctypes.create_string_buffer(b"abc")
             self.assertEqual(len(buf), 3+1)
 
-            ctypes.set_conversion_mode("ascii", "replace")
-            buf = ctypes.create_string_buffer("ab\xe4\xf6\xfc")
-            self.assertEqual(buf[:], b"ab???\0")
-            self.assertEqual(buf[::], b"ab???\0")
-            self.assertEqual(buf[::-1], b"\0???ba")
-            self.assertEqual(buf[::2], b"a??")
+            buf = ctypes.create_string_buffer(b"ab\xe4\xf6\xfc")
+            self.assertEqual(buf[:], b"ab\xe4\xf6\xfc\0")
+            self.assertEqual(buf[::], b"ab\xe4\xf6\xfc\0")
+            self.assertEqual(buf[::-1], b'\x00\xfc\xf6\xe4ba')
+            self.assertEqual(buf[::2], b'a\xe4\xfc')
             self.assertEqual(buf[6:5:-1], b"")
 
-            ctypes.set_conversion_mode("ascii", "ignore")
-            buf = ctypes.create_string_buffer("ab\xe4\xf6\xfc")
-            # is that correct? not sure.  But with 'ignore', you get what you pay for..
-            self.assertEqual(buf[:], b"ab\0\0\0\0")
-            self.assertEqual(buf[::], b"ab\0\0\0\0")
-            self.assertEqual(buf[::-1], b"\0\0\0\0ba")
 
 if __name__ == '__main__':
     unittest.main()

Modified: python/branches/import_unicode/Lib/dbm/__init__.py
==============================================================================
--- python/branches/import_unicode/Lib/dbm/__init__.py	(original)
+++ python/branches/import_unicode/Lib/dbm/__init__.py	Fri Jul 30 01:19:37 2010
@@ -5,12 +5,11 @@
         import dbm
         d = dbm.open(file, 'w', 0o666)
 
-The returned object is a dbm.bsd, dbm.gnu, dbm.ndbm or dbm.dumb
-object, dependent on the type of database being opened (determined by
-the whichdb function) in the case of an existing dbm.  If the dbm does
-not exist and the create or new flag ('c' or 'n') was specified, the
-dbm type will be determined by the availability of the modules (tested
-in the above order).
+The returned object is a dbm.gnu, dbm.ndbm or dbm.dumb object, dependent on the
+type of database being opened (determined by the whichdb function) in the case
+of an existing dbm. If the dbm does not exist and the create or new flag ('c'
+or 'n') was specified, the dbm type will be determined by the availability of
+the modules (tested in the above order).
 
 It has the following interface (key and data are strings):
 
@@ -47,7 +46,7 @@
 class error(Exception):
     pass
 
-_names = ['dbm.bsd', 'dbm.gnu', 'dbm.ndbm', 'dbm.dumb']
+_names = ['dbm.gnu', 'dbm.ndbm', 'dbm.dumb']
 _defaultmod = None
 _modules = {}
 
@@ -168,10 +167,6 @@
     if magic == 0x13579ace:
         return "dbm.gnu"
 
-    ## Check for old Berkeley db hash file format v2
-    #if magic in (0x00061561, 0x61150600):
-    #    return "bsddb185" # not supported anymore
-
     # Later versions of Berkeley db hash file have a 12-byte pad in
     # front of the file type
     try:
@@ -179,10 +174,6 @@
     except struct.error:
         return ""
 
-    ## Check for BSD hash
-    #if magic in (0x00061561, 0x61150600):
-    #    return "dbm.bsd"
-
     # Unknown
     return ""
 

Modified: python/branches/import_unicode/Lib/distutils/archive_util.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/archive_util.py	(original)
+++ python/branches/import_unicode/Lib/distutils/archive_util.py	Fri Jul 30 01:19:37 2010
@@ -14,55 +14,15 @@
 from distutils.dir_util import mkpath
 from distutils import log
 
-try:
-    from pwd import getpwnam
-except ImportError:
-    getpwnam = None
-
-try:
-    from grp import getgrnam
-except ImportError:
-    getgrnam = None
-
-def _get_gid(name):
-    """Returns a gid, given a group name."""
-    if getgrnam is None or name is None:
-        return None
-    try:
-        result = getgrnam(name)
-    except KeyError:
-        result = None
-    if result is not None:
-        return result[2]
-    return None
-
-def _get_uid(name):
-    """Returns an uid, given a user name."""
-    if getpwnam is None or name is None:
-        return None
-    try:
-        result = getpwnam(name)
-    except KeyError:
-        result = None
-    if result is not None:
-        return result[2]
-    return None
-
-def make_tarball(base_name, base_dir, compress="gzip", verbose=0, dry_run=0,
-                 owner=None, group=None):
+def make_tarball(base_name, base_dir, compress="gzip", verbose=0, dry_run=0):
     """Create a (possibly compressed) tar file from all the files under
     'base_dir'.
 
     'compress' must be "gzip" (the default), "compress", "bzip2", or None.
-    (compress will be deprecated in Python 3.2)
-
-    'owner' and 'group' can be used to define an owner and a group for the
-    archive that is being built. If not provided, the current owner and group
-    will be used.
-
+    Both "tar" and the compression utility named by 'compress' must be on
+    the default program search path, so this is probably Unix-specific.
     The output tar file will be named 'base_dir' +  ".tar", possibly plus
     the appropriate compression extension (".gz", ".bz2" or ".Z").
-
     Returns the output filename.
     """
     tar_compression = {'gzip': 'gz', 'bzip2': 'bz2', None: '', 'compress': ''}
@@ -84,23 +44,10 @@
     import tarfile  # late import so Python build itself doesn't break
 
     log.info('Creating tar archive')
-
-    uid = _get_uid(owner)
-    gid = _get_gid(group)
-
-    def _set_uid_gid(tarinfo):
-        if gid is not None:
-            tarinfo.gid = gid
-            tarinfo.gname = group
-        if uid is not None:
-            tarinfo.uid = uid
-            tarinfo.uname = owner
-        return tarinfo
-
     if not dry_run:
         tar = tarfile.open(archive_name, 'w|%s' % tar_compression[compress])
         try:
-            tar.add(base_dir, filter=_set_uid_gid)
+            tar.add(base_dir)
         finally:
             tar.close()
 
@@ -190,7 +137,7 @@
     return None
 
 def make_archive(base_name, format, root_dir=None, base_dir=None, verbose=0,
-                 dry_run=0, owner=None, group=None):
+                 dry_run=0):
     """Create an archive file (eg. zip or tar).
 
     'base_name' is the name of the file to create, minus any format-specific
@@ -203,9 +150,6 @@
     ie. 'base_dir' will be the common prefix of all files and
     directories in the archive.  'root_dir' and 'base_dir' both default
     to the current directory.  Returns the name of the archive file.
-
-    'owner' and 'group' are used when creating a tar archive. By default,
-    uses the current owner and group.
     """
     save_cwd = os.getcwd()
     if root_dir is not None:
@@ -227,11 +171,6 @@
     func = format_info[0]
     for arg, val in format_info[1]:
         kwargs[arg] = val
-
-    if format != 'zip':
-        kwargs['owner'] = owner
-        kwargs['group'] = group
-
     try:
         filename = func(base_name, base_dir, **kwargs)
     finally:

Modified: python/branches/import_unicode/Lib/distutils/bcppcompiler.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/bcppcompiler.py	(original)
+++ python/branches/import_unicode/Lib/distutils/bcppcompiler.py	Fri Jul 30 01:19:37 2010
@@ -13,11 +13,13 @@
 
 __revision__ = "$Id$"
 
-import os
 
-from distutils.errors import (DistutilsExecError, CompileError, LibError,
-                              LinkError, UnknownFileError)
-from distutils.ccompiler import CCompiler, gen_preprocess_options
+import os
+from distutils.errors import \
+     DistutilsExecError, DistutilsPlatformError, \
+     CompileError, LibError, LinkError, UnknownFileError
+from distutils.ccompiler import \
+     CCompiler, gen_preprocess_options, gen_lib_options
 from distutils.file_util import write_file
 from distutils.dep_util import newer
 from distutils import log

Modified: python/branches/import_unicode/Lib/distutils/ccompiler.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/ccompiler.py	(original)
+++ python/branches/import_unicode/Lib/distutils/ccompiler.py	Fri Jul 30 01:19:37 2010
@@ -5,71 +5,15 @@
 
 __revision__ = "$Id$"
 
-import sys
-import os
-import re
-
-from distutils.errors import (CompileError, LinkError, UnknownFileError,
-                              DistutilsPlatformError, DistutilsModuleError)
+import sys, os, re
+from distutils.errors import *
 from distutils.spawn import spawn
 from distutils.file_util import move_file
 from distutils.dir_util import mkpath
-from distutils.dep_util import newer_group
+from distutils.dep_util import newer_pairwise, newer_group
 from distutils.util import split_quoted, execute
 from distutils import log
 
-_sysconfig = __import__('sysconfig')
-
-def customize_compiler(compiler):
-    """Do any platform-specific customization of a CCompiler instance.
-
-    Mainly needed on Unix, so we can plug in the information that
-    varies across Unices and is stored in Python's Makefile.
-    """
-    if compiler.compiler_type == "unix":
-        (cc, cxx, opt, cflags, ccshared, ldshared, so_ext, ar, ar_flags) = \
-            _sysconfig.get_config_vars('CC', 'CXX', 'OPT', 'CFLAGS',
-                                       'CCSHARED', 'LDSHARED', 'SO', 'AR',
-                                       'ARFLAGS')
-
-        if 'CC' in os.environ:
-            cc = os.environ['CC']
-        if 'CXX' in os.environ:
-            cxx = os.environ['CXX']
-        if 'LDSHARED' in os.environ:
-            ldshared = os.environ['LDSHARED']
-        if 'CPP' in os.environ:
-            cpp = os.environ['CPP']
-        else:
-            cpp = cc + " -E"           # not always
-        if 'LDFLAGS' in os.environ:
-            ldshared = ldshared + ' ' + os.environ['LDFLAGS']
-        if 'CFLAGS' in os.environ:
-            cflags = opt + ' ' + os.environ['CFLAGS']
-            ldshared = ldshared + ' ' + os.environ['CFLAGS']
-        if 'CPPFLAGS' in os.environ:
-            cpp = cpp + ' ' + os.environ['CPPFLAGS']
-            cflags = cflags + ' ' + os.environ['CPPFLAGS']
-            ldshared = ldshared + ' ' + os.environ['CPPFLAGS']
-        if 'AR' in os.environ:
-            ar = os.environ['AR']
-        if 'ARFLAGS' in os.environ:
-            archiver = ar + ' ' + os.environ['ARFLAGS']
-        else:
-            archiver = ar + ' ' + ar_flags
-
-        cc_cmd = cc + ' ' + cflags
-        compiler.set_executables(
-            preprocessor=cpp,
-            compiler=cc_cmd,
-            compiler_so=cc_cmd + ' ' + ccshared,
-            compiler_cxx=cxx,
-            linker_so=ldshared,
-            linker_exe=cc,
-            archiver=archiver)
-
-        compiler.shared_lib_extension = so_ext
-
 class CCompiler:
     """Abstract base class to define the interface that must be implemented
     by real compiler classes.  Also has some utility methods used by
@@ -449,6 +393,22 @@
 
         return output_dir, macros, include_dirs
 
+    def _prep_compile(self, sources, output_dir, depends=None):
+        """Decide which souce files must be recompiled.
+
+        Determine the list of object files corresponding to 'sources',
+        and figure out which ones really need to be recompiled.
+        Return a list of all object files and a dictionary telling
+        which source files can be skipped.
+        """
+        # Get the list of expected output (object) files
+        objects = self.object_filenames(sources, output_dir=output_dir)
+        assert len(objects) == len(sources)
+
+        # Return an empty dict for the "which source files can be skipped"
+        # return value to preserve API compatibility.
+        return objects, {}
+
     def _fix_object_args(self, objects, output_dir):
         """Typecheck and fix up some arguments supplied to various methods.
         Specifically: ensure that 'objects' is a list; if output_dir is
@@ -650,15 +610,26 @@
         """
         pass
 
+
     # values for target_desc parameter in link()
     SHARED_OBJECT = "shared_object"
     SHARED_LIBRARY = "shared_library"
     EXECUTABLE = "executable"
 
-    def link(self, target_desc, objects, output_filename, output_dir=None,
-             libraries=None, library_dirs=None, runtime_library_dirs=None,
-             export_symbols=None, debug=0, extra_preargs=None,
-             extra_postargs=None, build_temp=None, target_lang=None):
+    def link(self,
+             target_desc,
+             objects,
+             output_filename,
+             output_dir=None,
+             libraries=None,
+             library_dirs=None,
+             runtime_library_dirs=None,
+             export_symbols=None,
+             debug=0,
+             extra_preargs=None,
+             extra_postargs=None,
+             build_temp=None,
+             target_lang=None):
         """Link a bunch of stuff together to create an executable or
         shared library file.
 
@@ -707,11 +678,19 @@
 
     # Old 'link_*()' methods, rewritten to use the new 'link()' method.
 
-    def link_shared_lib(self, objects, output_libname, output_dir=None,
-                        libraries=None, library_dirs=None,
-                        runtime_library_dirs=None, export_symbols=None,
-                        debug=0, extra_preargs=None, extra_postargs=None,
-                        build_temp=None, target_lang=None):
+    def link_shared_lib(self,
+                        objects,
+                        output_libname,
+                        output_dir=None,
+                        libraries=None,
+                        library_dirs=None,
+                        runtime_library_dirs=None,
+                        export_symbols=None,
+                        debug=0,
+                        extra_preargs=None,
+                        extra_postargs=None,
+                        build_temp=None,
+                        target_lang=None):
         self.link(CCompiler.SHARED_LIBRARY, objects,
                   self.library_filename(output_libname, lib_type='shared'),
                   output_dir,
@@ -720,11 +699,19 @@
                   extra_preargs, extra_postargs, build_temp, target_lang)
 
 
-    def link_shared_object(self, objects, output_filename, output_dir=None,
-                           libraries=None, library_dirs=None,
-                           runtime_library_dirs=None, export_symbols=None,
-                           debug=0, extra_preargs=None, extra_postargs=None,
-                           build_temp=None, target_lang=None):
+    def link_shared_object(self,
+                           objects,
+                           output_filename,
+                           output_dir=None,
+                           libraries=None,
+                           library_dirs=None,
+                           runtime_library_dirs=None,
+                           export_symbols=None,
+                           debug=0,
+                           extra_preargs=None,
+                           extra_postargs=None,
+                           build_temp=None,
+                           target_lang=None):
         self.link(CCompiler.SHARED_OBJECT, objects,
                   output_filename, output_dir,
                   libraries, library_dirs, runtime_library_dirs,
@@ -732,10 +719,17 @@
                   extra_preargs, extra_postargs, build_temp, target_lang)
 
 
-    def link_executable(self, objects, output_progname, output_dir=None,
-                        libraries=None, library_dirs=None,
-                        runtime_library_dirs=None, debug=0, extra_preargs=None,
-                        extra_postargs=None, target_lang=None):
+    def link_executable(self,
+                        objects,
+                        output_progname,
+                        output_dir=None,
+                        libraries=None,
+                        library_dirs=None,
+                        runtime_library_dirs=None,
+                        debug=0,
+                        extra_preargs=None,
+                        extra_postargs=None,
+                        target_lang=None):
         self.link(CCompiler.EXECUTABLE, objects,
                   self.executable_filename(output_progname), output_dir,
                   libraries, library_dirs, runtime_library_dirs, None,
@@ -917,7 +911,7 @@
     def move_file(self, src, dst):
         return move_file(src, dst, dry_run=self.dry_run)
 
-    def mkpath(self, name, mode=0o777):
+    def mkpath (self, name, mode=0o777):
         mkpath(name, mode, dry_run=self.dry_run)
 
 
@@ -1085,14 +1079,12 @@
     return pp_opts
 
 
-def gen_lib_options(compiler, library_dirs, runtime_library_dirs, libraries):
+def gen_lib_options (compiler, library_dirs, runtime_library_dirs, libraries):
     """Generate linker options for searching library directories and
-    linking with specific libraries.
-
-    'libraries' and 'library_dirs' are, respectively, lists of library names
-    (not filenames!) and search directories.  Returns a list of command-line
-    options suitable for use with some compiler (depending on the two format
-    strings passed in).
+    linking with specific libraries.  'libraries' and 'library_dirs' are,
+    respectively, lists of library names (not filenames!) and search
+    directories.  Returns a list of command-line options suitable for use
+    with some compiler (depending on the two format strings passed in).
     """
     lib_opts = []
 
@@ -1102,7 +1094,7 @@
     for dir in runtime_library_dirs:
         opt = compiler.runtime_library_dir_option(dir)
         if isinstance(opt, list):
-            lib_opts.extend(opt)
+            lib_opts = lib_opts + opt
         else:
             lib_opts.append(opt)
 
@@ -1113,14 +1105,14 @@
     # pretty nasty way to arrange your C code.
 
     for lib in libraries:
-        lib_dir, lib_name = os.path.split(lib)
-        if lib_dir != '':
+        (lib_dir, lib_name) = os.path.split(lib)
+        if lib_dir:
             lib_file = compiler.find_library_file([lib_dir], lib_name)
-            if lib_file is not None:
+            if lib_file:
                 lib_opts.append(lib_file)
             else:
                 compiler.warn("no library file corresponding to "
                               "'%s' found (skipping)" % lib)
         else:
-            lib_opts.append(compiler.library_option(lib))
+            lib_opts.append(compiler.library_option (lib))
     return lib_opts

Modified: python/branches/import_unicode/Lib/distutils/cmd.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/cmd.py	(original)
+++ python/branches/import_unicode/Lib/distutils/cmd.py	Fri Jul 30 01:19:37 2010
@@ -367,11 +367,9 @@
         from distutils.spawn import spawn
         spawn(cmd, search_path, dry_run=self.dry_run)
 
-    def make_archive(self, base_name, format, root_dir=None, base_dir=None,
-                     owner=None, group=None):
-        return archive_util.make_archive(base_name, format, root_dir,
-                                         base_dir, dry_run=self.dry_run,
-                                         owner=owner, group=group)
+    def make_archive(self, base_name, format, root_dir=None, base_dir=None):
+        return archive_util.make_archive(base_name, format, root_dir, base_dir,
+                                         dry_run=self.dry_run)
 
     def make_file(self, infiles, outfile, func, args,
                   exec_msg=None, skip_msg=None, level=1):

Modified: python/branches/import_unicode/Lib/distutils/command/bdist.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/command/bdist.py	(original)
+++ python/branches/import_unicode/Lib/distutils/command/bdist.py	Fri Jul 30 01:19:37 2010
@@ -6,10 +6,9 @@
 __revision__ = "$Id$"
 
 import os
-
-from distutils.util import get_platform
 from distutils.core import Command
-from distutils.errors import DistutilsPlatformError, DistutilsOptionError
+from distutils.errors import *
+from distutils.util import get_platform
 
 
 def show_formats():
@@ -40,12 +39,6 @@
                      "[default: dist]"),
                     ('skip-build', None,
                      "skip rebuilding everything (for testing/debugging)"),
-                    ('owner=', 'u',
-                     "Owner name used when creating a tar file"
-                     " [default: current user]"),
-                    ('group=', 'g',
-                     "Group name used when creating a tar file"
-                     " [default: current group]"),
                    ]
 
     boolean_options = ['skip-build']
@@ -87,8 +80,6 @@
         self.formats = None
         self.dist_dir = None
         self.skip_build = 0
-        self.group = None
-        self.owner = None
 
     def finalize_options(self):
         # have to finalize 'plat_name' before 'bdist_base'
@@ -134,11 +125,6 @@
             if cmd_name not in self.no_format_option:
                 sub_cmd.format = self.formats[i]
 
-            # passing the owner and group names for tar archiving
-            if cmd_name == 'bdist_dumb':
-                sub_cmd.owner = self.owner
-                sub_cmd.group = self.group
-
             # If we're going to need to run this command again, tell it to
             # keep its temporary files around so subsequent runs go faster.
             if cmd_name in commands[i+1:]:

Modified: python/branches/import_unicode/Lib/distutils/command/bdist_dumb.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/command/bdist_dumb.py	(original)
+++ python/branches/import_unicode/Lib/distutils/command/bdist_dumb.py	Fri Jul 30 01:19:37 2010
@@ -7,18 +7,16 @@
 __revision__ = "$Id$"
 
 import os
-
-from sysconfig import get_python_version
-
-from distutils.util import get_platform
 from distutils.core import Command
+from distutils.util import get_platform
 from distutils.dir_util import remove_tree, ensure_relative
-from distutils.errors import DistutilsPlatformError
+from distutils.errors import *
+from distutils.sysconfig import get_python_version
 from distutils import log
 
 class bdist_dumb(Command):
 
-    description = 'create a "dumb" built distribution'
+    description = "create a \"dumb\" built distribution"
 
     user_options = [('bdist-dir=', 'd',
                      "temporary directory for creating the distribution"),
@@ -37,12 +35,6 @@
                     ('relative', None,
                      "build the archive using relative paths"
                      "(default: false)"),
-                    ('owner=', 'u',
-                     "Owner name used when creating a tar file"
-                     " [default: current user]"),
-                    ('group=', 'g',
-                     "Group name used when creating a tar file"
-                     " [default: current group]"),
                    ]
 
     boolean_options = ['keep-temp', 'skip-build', 'relative']
@@ -59,8 +51,6 @@
         self.dist_dir = None
         self.skip_build = 0
         self.relative = 0
-        self.owner = None
-        self.group = None
 
     def finalize_options(self):
         if self.bdist_dir is None:
@@ -118,8 +108,7 @@
 
         # Make the archive
         filename = self.make_archive(pseudoinstall_root,
-                                     self.format, root_dir=archive_root,
-                                     owner=self.owner, group=self.group)
+                                     self.format, root_dir=archive_root)
         if self.distribution.has_ext_modules():
             pyversion = get_python_version()
         else:

Modified: python/branches/import_unicode/Lib/distutils/command/bdist_msi.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/command/bdist_msi.py	(original)
+++ python/branches/import_unicode/Lib/distutils/command/bdist_msi.py	Fri Jul 30 01:19:37 2010
@@ -6,15 +6,15 @@
 """
 Implements the bdist_msi command.
 """
-import sys, os
-from sysconfig import get_python_version, get_platform
 
+import sys, os
 from distutils.core import Command
 from distutils.dir_util import remove_tree
+from distutils.sysconfig import get_python_version
 from distutils.version import StrictVersion
 from distutils.errors import DistutilsOptionError
+from distutils.util import get_platform
 from distutils import log
-
 import msilib
 from msilib import schema, sequence, text
 from msilib import Directory, Feature, Dialog, add_data
@@ -28,6 +28,7 @@
         default, cancel, bitmap=true)"""
         Dialog.__init__(self, *args)
         ruler = self.h - 36
+        bmwidth = 152*ruler/328
         #if kw.get("bitmap", True):
         #    self.bitmap("Bitmap", 0, 0, bmwidth, ruler, "PythonWin")
         self.line("BottomLine", 0, ruler, self.w, 0)
@@ -418,6 +419,7 @@
         # see "Dialog Style Bits"
         modal = 3      # visible | modal
         modeless = 1   # visible
+        track_disk_space = 32
 
         # UI customization properties
         add_data(db, "Property",

Modified: python/branches/import_unicode/Lib/distutils/command/bdist_rpm.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/command/bdist_rpm.py	(original)
+++ python/branches/import_unicode/Lib/distutils/command/bdist_rpm.py	Fri Jul 30 01:19:37 2010
@@ -5,14 +5,13 @@
 
 __revision__ = "$Id$"
 
-import sys
-import os
-
+import sys, os
 from distutils.core import Command
 from distutils.debug import DEBUG
+from distutils.util import get_platform
 from distutils.file_util import write_file
-from distutils.errors import (DistutilsOptionError, DistutilsPlatformError,
-                              DistutilsFileError, DistutilsExecError)
+from distutils.errors import *
+from distutils.sysconfig import get_python_version
 from distutils import log
 
 class bdist_rpm(Command):

Modified: python/branches/import_unicode/Lib/distutils/command/bdist_wininst.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/command/bdist_wininst.py	(original)
+++ python/branches/import_unicode/Lib/distutils/command/bdist_wininst.py	Fri Jul 30 01:19:37 2010
@@ -5,16 +5,13 @@
 
 __revision__ = "$Id$"
 
-import sys
-import os
-
-from sysconfig import get_python_version
-
+import sys, os
 from distutils.core import Command
-from distutils.dir_util import remove_tree
-from distutils.errors import DistutilsOptionError, DistutilsPlatformError
-from distutils import log
 from distutils.util import get_platform
+from distutils.dir_util import create_tree, remove_tree
+from distutils.errors import *
+from distutils.sysconfig import get_python_version
+from distutils import log
 
 class bdist_wininst(Command):
 

Modified: python/branches/import_unicode/Lib/distutils/command/build.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/command/build.py	(original)
+++ python/branches/import_unicode/Lib/distutils/command/build.py	Fri Jul 30 01:19:37 2010
@@ -5,15 +5,16 @@
 __revision__ = "$Id$"
 
 import sys, os
-
-from distutils.util import get_platform
 from distutils.core import Command
 from distutils.errors import DistutilsOptionError
+from distutils.util import get_platform
+
 
 def show_compilers():
     from distutils.ccompiler import show_compilers
     show_compilers()
 
+
 class build(Command):
 
     description = "build everything needed to install"
@@ -126,6 +127,7 @@
         for cmd_name in self.get_sub_commands():
             self.run_command(cmd_name)
 
+
     # -- Predicates for the sub-command list ---------------------------
 
     def has_pure_modules(self):
@@ -140,6 +142,7 @@
     def has_scripts(self):
         return self.distribution.has_scripts()
 
+
     sub_commands = [('build_py',      has_pure_modules),
                     ('build_clib',    has_c_libraries),
                     ('build_ext',     has_ext_modules),

Modified: python/branches/import_unicode/Lib/distutils/command/build_clib.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/command/build_clib.py	(original)
+++ python/branches/import_unicode/Lib/distutils/command/build_clib.py	Fri Jul 30 01:19:37 2010
@@ -18,8 +18,8 @@
 
 import os
 from distutils.core import Command
-from distutils.errors import DistutilsSetupError
-from distutils.ccompiler import customize_compiler
+from distutils.errors import *
+from distutils.sysconfig import customize_compiler
 from distutils import log
 
 def show_compilers():

Modified: python/branches/import_unicode/Lib/distutils/command/build_ext.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/command/build_ext.py	(original)
+++ python/branches/import_unicode/Lib/distutils/command/build_ext.py	Fri Jul 30 01:19:37 2010
@@ -7,14 +7,12 @@
 __revision__ = "$Id$"
 
 import sys, os, re
-from warnings import warn
-
-from distutils.util import get_platform
 from distutils.core import Command
 from distutils.errors import *
-from distutils.ccompiler import customize_compiler
+from distutils.sysconfig import customize_compiler, get_python_version
 from distutils.dep_util import newer_group
 from distutils.extension import Extension
+from distutils.util import get_platform
 from distutils import log
 
 # this keeps compatibility from 2.3 to 2.5
@@ -114,39 +112,6 @@
          "list available compilers", show_compilers),
         ]
 
-    # making 'compiler' a property to deprecate
-    # its usage as something else than a compiler type
-    # e.g. like a compiler instance
-    def __init__(self, dist):
-        self._compiler = None
-        Command.__init__(self, dist)
-
-    def __setattr__(self, name, value):
-        # need this to make sure setattr() (used in distutils)
-        # doesn't kill our property
-        if name == 'compiler':
-            self._set_compiler(value)
-        else:
-            self.__dict__[name] = value
-
-    def _set_compiler(self, compiler):
-        if not isinstance(compiler, str) and compiler is not None:
-            # we don't want to allow that anymore in the future
-            warn("'compiler' specifies the compiler type in build_ext. "
-                 "If you want to get the compiler object itself, "
-                 "use 'compiler_obj'", DeprecationWarning)
-        self._compiler = compiler
-
-    def _get_compiler(self):
-        if not isinstance(self._compiler, str) and self._compiler is not None:
-            # we don't want to allow that anymore in the future
-            warn("'compiler' specifies the compiler type in build_ext. "
-                 "If you want to get the compiler object itself, "
-                 "use 'compiler_obj'", DeprecationWarning)
-        return self._compiler
-
-    compiler = property(_get_compiler, _set_compiler)
-
     def initialize_options(self):
         self.extensions = None
         self.build_lib = None
@@ -171,7 +136,8 @@
         self.user = None
 
     def finalize_options(self):
-        _sysconfig = __import__('sysconfig')
+        from distutils import sysconfig
+
         self.set_undefined_options('build',
                                    ('build_lib', 'build_lib'),
                                    ('build_temp', 'build_temp'),
@@ -188,8 +154,8 @@
 
         # Make sure Python's include directories (for Python.h, pyconfig.h,
         # etc.) are in the include search path.
-        py_include = _sysconfig.get_path('include')
-        plat_py_include = _sysconfig.get_path('platinclude')
+        py_include = sysconfig.get_python_inc()
+        plat_py_include = sysconfig.get_python_inc(plat_specific=1)
         if self.include_dirs is None:
             self.include_dirs = self.distribution.include_dirs or []
         if isinstance(self.include_dirs, str):
@@ -261,13 +227,13 @@
         if os.name == 'os2':
             self.library_dirs.append(os.path.join(sys.exec_prefix, 'Config'))
 
-        # for extensions under Cygwin Python's library directory must be
+        # for extensions under Cygwin and AtheOS Python's library directory must be
         # appended to library_dirs
-        if sys.platform[:6] == 'cygwin':
+        if sys.platform[:6] == 'cygwin' or sys.platform[:6] == 'atheos':
             if sys.executable.startswith(os.path.join(sys.exec_prefix, "bin")):
                 # building third party extensions
                 self.library_dirs.append(os.path.join(sys.prefix, "lib",
-                                  "python" + _sysconfig.get_python_version(),
+                                                      "python" + get_python_version(),
                                                       "config"))
             else:
                 # building python standard extensions
@@ -275,13 +241,13 @@
 
         # for extensions under Linux or Solaris with a shared Python library,
         # Python's library directory must be appended to library_dirs
-        _sysconfig.get_config_var('Py_ENABLE_SHARED')
+        sysconfig.get_config_var('Py_ENABLE_SHARED')
         if ((sys.platform.startswith('linux') or sys.platform.startswith('gnu')
              or sys.platform.startswith('sunos'))
-            and _sysconfig.get_config_var('Py_ENABLE_SHARED')):
+            and sysconfig.get_config_var('Py_ENABLE_SHARED')):
             if sys.executable.startswith(os.path.join(sys.exec_prefix, "bin")):
                 # building third party extensions
-                self.library_dirs.append(_sysconfig.get_config_var('LIBDIR'))
+                self.library_dirs.append(sysconfig.get_config_var('LIBDIR'))
             else:
                 # building python standard extensions
                 self.library_dirs.append('.')
@@ -344,50 +310,38 @@
 
         # Setup the CCompiler object that we'll use to do all the
         # compiling and linking
-
-        # used to prevent the usage of an existing compiler for the
-        # compiler option when calling new_compiler()
-        # this will be removed in 3.3 and 2.8
-        if not isinstance(self._compiler, str):
-            self._compiler = None
-
-        self.compiler_obj = new_compiler(compiler=self._compiler,
-                                         verbose=self.verbose,
-                                         dry_run=self.dry_run,
-                                         force=self.force)
-
-        # used to keep the compiler object reachable with
-        # "self.compiler". this will be removed in 3.3 and 2.8
-        self._compiler = self.compiler_obj
-
-        customize_compiler(self.compiler_obj)
+        self.compiler = new_compiler(compiler=self.compiler,
+                                     verbose=self.verbose,
+                                     dry_run=self.dry_run,
+                                     force=self.force)
+        customize_compiler(self.compiler)
         # If we are cross-compiling, init the compiler now (if we are not
         # cross-compiling, init would not hurt, but people may rely on
         # late initialization of compiler even if they shouldn't...)
         if os.name == 'nt' and self.plat_name != get_platform():
-            self.compiler_obj.initialize(self.plat_name)
+            self.compiler.initialize(self.plat_name)
 
         # And make sure that any compile/link-related options (which might
         # come from the command-line or from the setup script) are set in
         # that CCompiler object -- that way, they automatically apply to
         # all compiling and linking done here.
         if self.include_dirs is not None:
-            self.compiler_obj.set_include_dirs(self.include_dirs)
+            self.compiler.set_include_dirs(self.include_dirs)
         if self.define is not None:
             # 'define' option is a list of (name,value) tuples
             for (name, value) in self.define:
-                self.compiler_obj.define_macro(name, value)
+                self.compiler.define_macro(name, value)
         if self.undef is not None:
             for macro in self.undef:
-                self.compiler_obj.undefine_macro(macro)
+                self.compiler.undefine_macro(macro)
         if self.libraries is not None:
-            self.compiler_obj.set_libraries(self.libraries)
+            self.compiler.set_libraries(self.libraries)
         if self.library_dirs is not None:
-            self.compiler_obj.set_library_dirs(self.library_dirs)
+            self.compiler.set_library_dirs(self.library_dirs)
         if self.rpath is not None:
-            self.compiler_obj.set_runtime_library_dirs(self.rpath)
+            self.compiler.set_runtime_library_dirs(self.rpath)
         if self.link_objects is not None:
-            self.compiler_obj.set_link_objects(self.link_objects)
+            self.compiler.set_link_objects(self.link_objects)
 
         # Now actually compile and link everything.
         self.build_extensions()
@@ -548,13 +502,13 @@
         for undef in ext.undef_macros:
             macros.append((undef,))
 
-        objects = self.compiler_obj.compile(sources,
-                                            output_dir=self.build_temp,
-                                            macros=macros,
-                                            include_dirs=ext.include_dirs,
-                                            debug=self.debug,
-                                            extra_postargs=extra_args,
-                                            depends=ext.depends)
+        objects = self.compiler.compile(sources,
+                                         output_dir=self.build_temp,
+                                         macros=macros,
+                                         include_dirs=ext.include_dirs,
+                                         debug=self.debug,
+                                         extra_postargs=extra_args,
+                                         depends=ext.depends)
 
         # XXX -- this is a Vile HACK!
         #
@@ -575,9 +529,9 @@
         extra_args = ext.extra_link_args or []
 
         # Detect target language, if not provided
-        language = ext.language or self.compiler_obj.detect_language(sources)
+        language = ext.language or self.compiler.detect_language(sources)
 
-        self.compiler_obj.link_shared_object(
+        self.compiler.link_shared_object(
             objects, ext_path,
             libraries=self.get_libraries(ext),
             library_dirs=ext.library_dirs,
@@ -710,13 +664,13 @@
         of the file from which it will be loaded (eg. "foo/bar.so", or
         "foo\bar.pyd").
         """
-        _sysconfig = __import__('sysconfig')
+        from distutils.sysconfig import get_config_var
         ext_path = ext_name.split('.')
         # OS/2 has an 8 character module (extension) limit :-(
         if os.name == "os2":
             ext_path[len(ext_path) - 1] = ext_path[len(ext_path) - 1][:8]
         # extensions in debug_mode are named 'module_d.pyd' under windows
-        so_ext = _sysconfig.get_config_var('SO')
+        so_ext = get_config_var('SO')
         if os.name == 'nt' and self.debug:
             return os.path.join(*ext_path) + '_d' + so_ext
         return os.path.join(*ext_path) + so_ext
@@ -744,7 +698,7 @@
         # Append '_d' to the python import library on debug builds.
         if sys.platform == "win32":
             from distutils.msvccompiler import MSVCCompiler
-            if not isinstance(self.compiler_obj, MSVCCompiler):
+            if not isinstance(self.compiler, MSVCCompiler):
                 template = "python%d%d"
                 if self.debug:
                     template = template + '_d'
@@ -775,12 +729,28 @@
             # don't extend ext.libraries, it may be shared with other
             # extensions, it is a reference to the original list
             return ext.libraries + [pythonlib]
+        elif sys.platform[:6] == "atheos":
+            from distutils import sysconfig
+
+            template = "python%d.%d"
+            pythonlib = (template %
+                   (sys.hexversion >> 24, (sys.hexversion >> 16) & 0xff))
+            # Get SHLIBS from Makefile
+            extra = []
+            for lib in sysconfig.get_config_var('SHLIBS').split():
+                if lib.startswith('-l'):
+                    extra.append(lib[2:])
+                else:
+                    extra.append(lib)
+            # don't extend ext.libraries, it may be shared with other
+            # extensions, it is a reference to the original list
+            return ext.libraries + [pythonlib, "m"] + extra
         elif sys.platform == 'darwin':
             # Don't use the default code below
             return ext.libraries
         else:
-            _sysconfig = __import__('sysconfig')
-            if _sysconfig.get_config_var('Py_ENABLE_SHARED'):
+            from distutils import sysconfig
+            if sysconfig.get_config_var('Py_ENABLE_SHARED'):
                 template = "python%d.%d"
                 pythonlib = (template %
                              (sys.hexversion >> 24, (sys.hexversion >> 16) & 0xff))

Modified: python/branches/import_unicode/Lib/distutils/command/build_py.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/command/build_py.py	(original)
+++ python/branches/import_unicode/Lib/distutils/command/build_py.py	Fri Jul 30 01:19:37 2010
@@ -4,16 +4,16 @@
 
 __revision__ = "$Id$"
 
-import os
+import sys, os
 import sys
 from glob import glob
 
 from distutils.core import Command
-from distutils.errors import DistutilsOptionError, DistutilsFileError
+from distutils.errors import *
 from distutils.util import convert_path, Mixin2to3
 from distutils import log
 
-class build_py(Command):
+class build_py (Command):
 
     description = "\"build\" pure Python modules (copy to build directory)"
 
@@ -133,6 +133,7 @@
 
     def build_package_data(self):
         """Copy data files into build directory"""
+        lastdir = None
         for package, src_dir, build_dir, filenames in self.data_files:
             for filename in filenames:
                 target = os.path.join(build_dir, filename)

Modified: python/branches/import_unicode/Lib/distutils/command/build_scripts.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/command/build_scripts.py	(original)
+++ python/branches/import_unicode/Lib/distutils/command/build_scripts.py	Fri Jul 30 01:19:37 2010
@@ -6,6 +6,7 @@
 
 import os, re
 from stat import ST_MODE
+from distutils import sysconfig
 from distutils.core import Command
 from distutils.dep_util import newer
 from distutils.util import convert_path, Mixin2to3
@@ -56,7 +57,6 @@
         ie. starts with "\#!" and contains "python"), then adjust the first
         line to refer to the current Python interpreter as we copy.
         """
-        _sysconfig = __import__('sysconfig')
         self.mkpath(self.build_dir)
         outfiles = []
         updated_files = []
@@ -96,16 +96,16 @@
                 updated_files.append(outfile)
                 if not self.dry_run:
                     outf = open(outfile, "w")
-                    if not _sysconfig.is_python_build():
+                    if not sysconfig.python_build:
                         outf.write("#!%s%s\n" %
                                    (self.executable,
                                     post_interp))
                     else:
                         outf.write("#!%s%s\n" %
                                    (os.path.join(
-                            _sysconfig.get_config_var("BINDIR"),
-                           "python%s%s" % (_sysconfig.get_config_var("VERSION"),
-                                           _sysconfig.get_config_var("EXE"))),
+                            sysconfig.get_config_var("BINDIR"),
+                           "python%s%s" % (sysconfig.get_config_var("VERSION"),
+                                           sysconfig.get_config_var("EXE"))),
                                     post_interp))
                     outf.writelines(f.readlines())
                     outf.close()

Modified: python/branches/import_unicode/Lib/distutils/command/config.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/command/config.py	(original)
+++ python/branches/import_unicode/Lib/distutils/command/config.py	Fri Jul 30 01:19:37 2010
@@ -11,12 +11,11 @@
 
 __revision__ = "$Id$"
 
-import os
-import re
+import sys, os, re
 
 from distutils.core import Command
 from distutils.errors import DistutilsExecError
-from distutils.ccompiler import customize_compiler
+from distutils.sysconfig import customize_compiler
 from distutils import log
 
 LANG_EXT = {"c": ".c", "c++": ".cxx"}

Modified: python/branches/import_unicode/Lib/distutils/command/install.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/command/install.py	(original)
+++ python/branches/import_unicode/Lib/distutils/command/install.py	Fri Jul 30 01:19:37 2010
@@ -7,17 +7,26 @@
 import sys
 import os
 
-from sysconfig import get_config_vars, get_paths, get_path, get_config_var
-
 from distutils import log
 from distutils.core import Command
 from distutils.debug import DEBUG
+from distutils.sysconfig import get_config_vars
 from distutils.errors import DistutilsPlatformError
 from distutils.file_util import write_file
-from distutils.util import convert_path, change_root, get_platform
+from distutils.util import convert_path, subst_vars, change_root
+from distutils.util import get_platform
 from distutils.errors import DistutilsOptionError
 
-# kept for backward compat, will be removed in 3.2
+# this keeps compatibility from 2.3 to 2.5
+if sys.version < "2.6":
+    USER_BASE = None
+    USER_SITE = None
+    HAS_USER_SITE = False
+else:
+    from site import USER_BASE
+    from site import USER_SITE
+    HAS_USER_SITE = True
+
 if sys.version < "2.2":
     WINDOWS_SCHEME = {
         'purelib': '$base',
@@ -50,21 +59,15 @@
         'scripts': '$base/bin',
         'data'   : '$base',
         },
-    'unix_user': {
-        'purelib': '$usersite',
-        'platlib': '$usersite',
-        'headers': '$userbase/include/python$py_version_short/$dist_name',
-        'scripts': '$userbase/bin',
-        'data'   : '$userbase',
-        },
     'nt': WINDOWS_SCHEME,
-    'nt_user': {
-        'purelib': '$usersite',
-        'platlib': '$usersite',
-        'headers': '$userbase/Python$py_version_nodot/Include/$dist_name',
-        'scripts': '$userbase/Scripts',
-        'data'   : '$userbase',
+    'mac': {
+        'purelib': '$base/Lib/site-packages',
+        'platlib': '$base/Lib/site-packages',
+        'headers': '$base/Include/$dist_name',
+        'scripts': '$base/Scripts',
+        'data'   : '$base',
         },
+
     'os2': {
         'purelib': '$base/Lib/site-packages',
         'platlib': '$base/Lib/site-packages',
@@ -72,26 +75,47 @@
         'scripts': '$base/Scripts',
         'data'   : '$base',
         },
-    'os2_home': {
+    }
+
+# user site schemes
+if HAS_USER_SITE:
+    INSTALL_SCHEMES['nt_user'] = {
+        'purelib': '$usersite',
+        'platlib': '$usersite',
+        'headers': '$userbase/Python$py_version_nodot/Include/$dist_name',
+        'scripts': '$userbase/Scripts',
+        'data'   : '$userbase',
+        }
+
+    INSTALL_SCHEMES['unix_user'] = {
         'purelib': '$usersite',
         'platlib': '$usersite',
         'headers': '$userbase/include/python$py_version_short/$dist_name',
         'scripts': '$userbase/bin',
         'data'   : '$userbase',
-        },
-    }
+        }
+
+    INSTALL_SCHEMES['mac_user'] = {
+        'purelib': '$usersite',
+        'platlib': '$usersite',
+        'headers': '$userbase/$py_version_short/include/$dist_name',
+        'scripts': '$userbase/bin',
+        'data'   : '$userbase',
+        }
 
+    INSTALL_SCHEMES['os2_home'] = {
+        'purelib': '$usersite',
+        'platlib': '$usersite',
+        'headers': '$userbase/include/python$py_version_short/$dist_name',
+        'scripts': '$userbase/bin',
+        'data'   : '$userbase',
+        }
+
+# The keys to an installation scheme; if any new types of files are to be
+# installed, be sure to add an entry to every installation scheme above,
+# and to SCHEME_KEYS here.
 SCHEME_KEYS = ('purelib', 'platlib', 'headers', 'scripts', 'data')
-# end of backward compat
 
-def _subst_vars(s, local_vars):
-    try:
-        return s.format(**local_vars)
-    except KeyError:
-        try:
-            return s.format(**os.environ)
-        except KeyError as var:
-            raise AttributeError('{%s}' % var)
 
 class install(Command):
 
@@ -158,10 +182,11 @@
 
     boolean_options = ['compile', 'force', 'skip-build']
 
-    user_options.append(('user', None,
-                        "install in user site-package '%s'" % \
-                            get_path('purelib', '%s_user' % os.name)))
-    boolean_options.append('user')
+    if HAS_USER_SITE:
+        user_options.append(('user', None,
+                             "install in user site-package '%s'" % USER_SITE))
+        boolean_options.append('user')
+
     negative_opt = {'no-compile' : 'compile'}
 
 
@@ -191,8 +216,8 @@
         self.install_lib = None         # set to either purelib or platlib
         self.install_scripts = None
         self.install_data = None
-        self.install_userbase = get_config_var('userbase')
-        self.install_usersite = get_path('purelib', '%s_user' % os.name)
+        self.install_userbase = USER_BASE
+        self.install_usersite = USER_SITE
 
         self.compile = None
         self.optimize = None
@@ -302,9 +327,7 @@
         # about needing recursive variable expansion (shudder).
 
         py_version = sys.version.split()[0]
-        prefix, exec_prefix, srcdir, projectbase = get_config_vars('prefix', 'exec_prefix',
-                                                      'srcdir', 'projectbase')
-
+        (prefix, exec_prefix) = get_config_vars('prefix', 'exec_prefix')
         self.config_vars = {'dist_name': self.distribution.get_name(),
                             'dist_version': self.distribution.get_version(),
                             'dist_fullname': self.distribution.get_fullname(),
@@ -315,12 +338,12 @@
                             'prefix': prefix,
                             'sys_exec_prefix': exec_prefix,
                             'exec_prefix': exec_prefix,
-                            'srcdir': srcdir,
-                            'projectbase': projectbase,
                            }
 
-        self.config_vars['userbase'] = self.install_userbase
-        self.config_vars['usersite'] = self.install_usersite
+        if HAS_USER_SITE:
+            self.config_vars['userbase'] = self.install_userbase
+            self.config_vars['usersite'] = self.install_usersite
+
         self.expand_basedirs()
 
         self.dump_dirs("post-expand_basedirs()")
@@ -424,10 +447,10 @@
                 raise DistutilsPlatformError(
                     "User base directory is not specified")
             self.install_base = self.install_platbase = self.install_userbase
-            self.select_scheme("posix_user")
+            self.select_scheme("unix_user")
         elif self.home is not None:
             self.install_base = self.install_platbase = self.home
-            self.select_scheme("posix_home")
+            self.select_scheme("unix_home")
         else:
             if self.prefix is None:
                 if self.exec_prefix is not None:
@@ -443,7 +466,7 @@
 
             self.install_base = self.prefix
             self.install_platbase = self.exec_prefix
-            self.select_scheme("posix_prefix")
+            self.select_scheme("unix_prefix")
 
     def finalize_other(self):
         """Finalizes options for non-posix platforms"""
@@ -455,7 +478,7 @@
             self.select_scheme(os.name + "_user")
         elif self.home is not None:
             self.install_base = self.install_platbase = self.home
-            self.select_scheme("posix_home")
+            self.select_scheme("unix_home")
         else:
             if self.prefix is None:
                 self.prefix = os.path.normpath(sys.prefix)
@@ -470,15 +493,11 @@
     def select_scheme(self, name):
         """Sets the install directories by applying the install schemes."""
         # it's the caller's problem if they supply a bad name!
-        scheme = get_paths(name, expand=False)
-        for key, value in scheme.items():
-            if key == 'platinclude':
-                key = 'headers'
-                value = os.path.join(value, self.distribution.get_name())
+        scheme = INSTALL_SCHEMES[name]
+        for key in SCHEME_KEYS:
             attrname = 'install_' + key
-            if hasattr(self, attrname):
-                if getattr(self, attrname) is None:
-                    setattr(self, attrname, value)
+            if getattr(self, attrname) is None:
+                setattr(self, attrname, scheme[key])
 
     def _expand_attrs(self, attrs):
         for attr in attrs:
@@ -486,7 +505,7 @@
             if val is not None:
                 if os.name == 'posix' or os.name == 'nt':
                     val = os.path.expanduser(val)
-                val = _subst_vars(val, self.config_vars)
+                val = subst_vars(val, self.config_vars)
                 setattr(self, attr, val)
 
     def expand_basedirs(self):

Modified: python/branches/import_unicode/Lib/distutils/command/register.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/command/register.py	(original)
+++ python/branches/import_unicode/Lib/distutils/command/register.py	Fri Jul 30 01:19:37 2010
@@ -7,15 +7,13 @@
 
 __revision__ = "$Id$"
 
-import os
-import string
-import getpass
+import os, string, getpass
 import io
-import urllib.parse
-import urllib.request
+import urllib.parse, urllib.request
 from warnings import warn
 
 from distutils.core import PyPIRCCommand
+from distutils.errors import *
 from distutils import log
 
 class register(PyPIRCCommand):

Modified: python/branches/import_unicode/Lib/distutils/command/sdist.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/command/sdist.py	(original)
+++ python/branches/import_unicode/Lib/distutils/command/sdist.py	Fri Jul 30 01:19:37 2010
@@ -7,14 +7,14 @@
 import os
 import string
 import sys
+from types import *
 from glob import glob
 from warnings import warn
 
 from distutils.core import Command
 from distutils import dir_util, dep_util, file_util, archive_util
 from distutils.text_file import TextFile
-from distutils.errors import (DistutilsPlatformError, DistutilsOptionError,
-                              DistutilsTemplateError)
+from distutils.errors import *
 from distutils.filelist import FileList
 from distutils import log
 from distutils.util import convert_path
@@ -76,10 +76,6 @@
         ('medata-check', None,
          "Ensure that all required elements of meta-data "
          "are supplied. Warn if any missing. [default]"),
-        ('owner=', 'u',
-         "Owner name used when creating a tar file [default: current user]"),
-        ('group=', 'g',
-         "Group name used when creating a tar file [default: current group]"),
         ]
 
     boolean_options = ['use-defaults', 'prune',
@@ -119,8 +115,6 @@
 
         self.archive_files = None
         self.metadata_check = 1
-        self.owner = None
-        self.group = None
 
     def finalize_options(self):
         if self.manifest is None:
@@ -424,8 +418,7 @@
             self.formats.append(self.formats.pop(self.formats.index('tar')))
 
         for fmt in self.formats:
-            file = self.make_archive(base_name, fmt, base_dir=base_dir,
-                                     owner=self.owner, group=self.group)
+            file = self.make_archive(base_name, fmt, base_dir=base_dir)
             archive_files.append(file)
             self.distribution.dist_files.append(('sdist', '', file))
 

Modified: python/branches/import_unicode/Lib/distutils/command/upload.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/command/upload.py	(original)
+++ python/branches/import_unicode/Lib/distutils/command/upload.py	Fri Jul 30 01:19:37 2010
@@ -1,19 +1,25 @@
 """distutils.command.upload
 
 Implements the Distutils 'upload' subcommand (upload package to PyPI)."""
-import os
-import io
-import socket
-import platform
-from urllib.request import urlopen, Request, HTTPError
-from base64 import standard_b64encode
-from urllib.parse import urlparse
-from hashlib import md5
 
-from distutils.errors import DistutilsOptionError
+from distutils.errors import *
 from distutils.core import PyPIRCCommand
 from distutils.spawn import spawn
 from distutils import log
+import sys
+import os, io
+import socket
+import platform
+import configparser
+import http.client as httpclient
+from base64 import standard_b64encode
+import urllib.parse
+
+# this keeps compatibility for 2.3 and 2.4
+if sys.version < "2.5":
+    from md5 import md5
+else:
+    from hashlib import md5
 
 class upload(PyPIRCCommand):
 
@@ -60,15 +66,6 @@
             self.upload_file(command, pyversion, filename)
 
     def upload_file(self, command, pyversion, filename):
-        # Makes sure the repository URL is compliant
-        schema, netloc, url, params, query, fragments = \
-            urlparse(self.repository)
-        if params or query or fragments:
-            raise AssertionError("Incompatible url %s" % self.repository)
-
-        if schema not in ('http', 'https'):
-            raise AssertionError("unsupported schema " + schema)
-
         # Sign if requested
         if self.sign:
             gpg_args = ["gpg", "--detach-sign", "-a", filename]
@@ -140,10 +137,10 @@
         for key, value in data.items():
             title = '\nContent-Disposition: form-data; name="%s"' % key
             # handle multiple entries for the same name
-            if not isinstance(value, list):
+            if type(value) != type([]):
                 value = [value]
             for value in value:
-                if isinstance(value, tuple):
+                if type(value) is tuple:
                     title += '; filename="%s"' % value[0]
                     value = value[1]
                 else:
@@ -161,30 +158,40 @@
         self.announce("Submitting %s to %s" % (filename, self.repository), log.INFO)
 
         # build the Request
-        headers = {'Content-type':
-                        'multipart/form-data; boundary=%s' % boundary,
-                   'Content-length': str(len(body)),
-                   'Authorization': auth}
-
-        request = Request(self.repository, data=body,
-                          headers=headers)
-        # send the data
+        # We can't use urllib since we need to send the Basic
+        # auth right with the first request
+        # TODO(jhylton): Can we fix urllib?
+        schema, netloc, url, params, query, fragments = \
+            urllib.parse.urlparse(self.repository)
+        assert not params and not query and not fragments
+        if schema == 'http':
+            http = httpclient.HTTPConnection(netloc)
+        elif schema == 'https':
+            http = httpclient.HTTPSConnection(netloc)
+        else:
+            raise AssertionError("unsupported schema "+schema)
+
+        data = ''
+        loglevel = log.INFO
         try:
-            result = urlopen(request)
-            status = result.getcode()
-            reason = result.msg
+            http.connect()
+            http.putrequest("POST", url)
+            http.putheader('Content-type',
+                           'multipart/form-data; boundary=%s'%boundary)
+            http.putheader('Content-length', str(len(body)))
+            http.putheader('Authorization', auth)
+            http.endheaders()
+            http.send(body)
         except socket.error as e:
             self.announce(str(e), log.ERROR)
             return
-        except HTTPError as e:
-            status = e.code
-            reason = e.msg
 
-        if status == 200:
-            self.announce('Server response (%s): %s' % (status, reason),
+        r = http.getresponse()
+        if r.status == 200:
+            self.announce('Server response (%s): %s' % (r.status, r.reason),
                           log.INFO)
         else:
-            self.announce('Upload failed (%s): %s' % (status, reason),
+            self.announce('Upload failed (%s): %s' % (r.status, r.reason),
                           log.ERROR)
         if self.show_response:
-            self.announce('-'*75, result.read(), '-'*75)
+            self.announce('-'*75, r.read(), '-'*75)

Modified: python/branches/import_unicode/Lib/distutils/config.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/config.py	(original)
+++ python/branches/import_unicode/Lib/distutils/config.py	Fri Jul 30 01:19:37 2010
@@ -4,6 +4,7 @@
 that uses .pypirc in the distutils.command package.
 """
 import os
+import sys
 from configparser import ConfigParser
 
 from distutils.cmd import Command
@@ -59,6 +60,8 @@
         if os.path.exists(rc):
             self.announce('Using PyPI login from %s' % rc)
             repository = self.repository or self.DEFAULT_REPOSITORY
+            realm = self.realm or self.DEFAULT_REALM
+
             config = ConfigParser()
             config.read(rc)
             sections = config.sections()

Modified: python/branches/import_unicode/Lib/distutils/core.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/core.py	(original)
+++ python/branches/import_unicode/Lib/distutils/core.py	Fri Jul 30 01:19:37 2010
@@ -8,12 +8,10 @@
 
 __revision__ = "$Id$"
 
-import sys
-import os
+import sys, os
 
 from distutils.debug import DEBUG
-from distutils.errors import (DistutilsSetupError, DistutilsArgError,
-                              DistutilsError, CCompilerError)
+from distutils.errors import *
 from distutils.util import grok_environment_error
 
 # Mainly import these so setup scripts can "from distutils.core import" them.
@@ -33,9 +31,9 @@
    or: %(script)s cmd --help
 """
 
-def gen_usage(script_name):
+def gen_usage (script_name):
     script = os.path.basename(script_name)
-    return USAGE % {'script': script}
+    return USAGE % vars()
 
 
 # Some mild magic to control the behaviour of 'setup()' from 'run_setup()'.
@@ -58,7 +56,7 @@
                       'extra_objects', 'extra_compile_args', 'extra_link_args',
                       'swig_opts', 'export_symbols', 'depends', 'language')
 
-def setup(**attrs):
+def setup (**attrs):
     """The gateway to the Distutils: do everything your setup script needs
     to do, in a highly flexible and user-driven way.  Briefly: create a
     Distribution instance; find and parse config files; parse the command
@@ -131,9 +129,8 @@
     if _setup_stop_after == "config":
         return dist
 
-    # Parse the command line and override config files; any
-    # command-line errors are the end user's fault, so turn them into
-    # SystemExit to suppress tracebacks.
+    # Parse the command line; any command-line errors are the end user's
+    # fault, so turn them into SystemExit to suppress tracebacks.
     try:
         ok = dist.parse_command_line()
     except DistutilsArgError as msg:
@@ -170,8 +167,10 @@
 
     return dist
 
+# setup ()
 
-def run_setup(script_name, script_args=None, stop_after="run"):
+
+def run_setup (script_name, script_args=None, stop_after="run"):
     """Run a setup script in a somewhat controlled environment, and
     return the Distribution instance that drives things.  This is useful
     if you need to find out the distribution meta-data (passed as
@@ -234,4 +233,7 @@
 
     # I wonder if the setup script's namespace -- g and l -- would be of
     # any interest to callers?
+    #print "_setup_distribution:", _setup_distribution
     return _setup_distribution
+
+# run_setup ()

Modified: python/branches/import_unicode/Lib/distutils/cygwinccompiler.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/cygwinccompiler.py	(original)
+++ python/branches/import_unicode/Lib/distutils/cygwinccompiler.py	Fri Jul 30 01:19:37 2010
@@ -50,13 +50,16 @@
 import os
 import sys
 import copy
+from subprocess import Popen, PIPE
 import re
-from warnings import warn
 
+from distutils.ccompiler import gen_preprocess_options, gen_lib_options
 from distutils.unixccompiler import UnixCCompiler
 from distutils.file_util import write_file
 from distutils.errors import DistutilsExecError, CompileError, UnknownFileError
-from distutils.util import get_compiler_versions
+from distutils import log
+from distutils.version import LooseVersion
+from distutils.spawn import find_executable
 
 def get_msvcr():
     """Include the appropriate MSVC runtime library if Python was built
@@ -107,7 +110,7 @@
                 % details)
 
         self.gcc_version, self.ld_version, self.dllwrap_version = \
-            get_compiler_versions()
+            get_versions()
         self.debug_print(self.compiler_type + ": gcc %s, ld %s, dllwrap %s\n" %
                          (self.gcc_version,
                           self.ld_version,
@@ -337,7 +340,7 @@
     # XXX since this function also checks sys.version, it's not strictly a
     # "pyconfig.h" check -- should probably be renamed...
 
-    _sysconfig = __import__('sysconfig')
+    from distutils import sysconfig
 
     # if sys.version contains GCC then python was compiled with GCC, and the
     # pyconfig.h file should be OK
@@ -345,7 +348,7 @@
         return CONFIG_H_OK, "sys.version mentions 'GCC'"
 
     # let's see if __GNUC__ is mentioned in python.h
-    fn = _sysconfig.get_config_h_filename()
+    fn = sysconfig.get_config_h_filename()
     try:
         with open(fn) as config_h:
             if "__GNUC__" in config_h.read():
@@ -356,27 +359,33 @@
         return (CONFIG_H_UNCERTAIN,
                 "couldn't read '%s': %s" % (fn, exc.strerror))
 
-class _Deprecated_SRE_Pattern(object):
-    def __init__(self, pattern):
-        self.pattern = pattern
-
-    def __getattr__(self, name):
-        if name in ('findall', 'finditer', 'match', 'scanner', 'search',
-                    'split', 'sub', 'subn'):
-            warn("'distutils.cygwinccompiler.RE_VERSION' is deprecated "
-                 "and will be removed in the next version", DeprecationWarning)
-        return getattr(self.pattern, name)
+RE_VERSION = re.compile(b'(\d+\.\d+(\.\d+)*)')
 
+def _find_exe_version(cmd):
+    """Find the version of an executable by running `cmd` in the shell.
 
-RE_VERSION = _Deprecated_SRE_Pattern(re.compile('(\d+\.\d+(\.\d+)*)'))
+    If the command is not found, or the output does not match
+    `RE_VERSION`, returns None.
+    """
+    executable = cmd.split()[0]
+    if find_executable(executable) is None:
+        return None
+    out = Popen(cmd, shell=True, stdout=PIPE).stdout
+    try:
+        out_string = out.read()
+    finally:
+        out.close()
+    result = RE_VERSION.search(out_string)
+    if result is None:
+        return None
+    # LooseVersion works with strings
+    # so we need to decode our bytes
+    return LooseVersion(result.group(1).decode())
 
 def get_versions():
     """ Try to find out the versions of gcc, ld and dllwrap.
 
     If not possible it returns None for it.
     """
-    warn("'distutils.cygwinccompiler.get_versions' is deprecated "
-         "use 'distutils.util.get_compiler_versions' instead",
-         DeprecationWarning)
-
-    return get_compiler_versions()
+    commands = ['gcc -dumpversion', 'ld -v', 'dllwrap --version']
+    return tuple([_find_exe_version(cmd) for cmd in commands])

Modified: python/branches/import_unicode/Lib/distutils/dep_util.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/dep_util.py	(original)
+++ python/branches/import_unicode/Lib/distutils/dep_util.py	Fri Jul 30 01:19:37 2010
@@ -9,27 +9,29 @@
 import os
 from distutils.errors import DistutilsFileError
 
-def newer(source, target):
-    """Tells if the target is newer than the source.
 
-    Return true if 'source' exists and is more recently modified than
-    'target', or if 'source' exists and 'target' doesn't.
-
-    Return false if both exist and 'target' is the same age or younger
-    than 'source'. Raise DistutilsFileError if 'source' does not exist.
-
-    Note that this test is not very accurate: files created in the same second
-    will have the same "age".
+def newer (source, target):
+    """Return true if 'source' exists and is more recently modified than
+    'target', or if 'source' exists and 'target' doesn't.  Return false if
+    both exist and 'target' is the same age or younger than 'source'.
+    Raise DistutilsFileError if 'source' does not exist.
     """
     if not os.path.exists(source):
         raise DistutilsFileError("file '%s' does not exist" %
                                  os.path.abspath(source))
     if not os.path.exists(target):
-        return True
+        return 1
+
+    from stat import ST_MTIME
+    mtime1 = os.stat(source)[ST_MTIME]
+    mtime2 = os.stat(target)[ST_MTIME]
+
+    return mtime1 > mtime2
+
+# newer ()
 
-    return os.stat(source).st_mtime > os.stat(target).st_mtime
 
-def newer_pairwise(sources, targets):
+def newer_pairwise (sources, targets):
     """Walk two filename lists in parallel, testing if each source is newer
     than its corresponding target.  Return a pair of lists (sources,
     targets) where source is newer than target, according to the semantics
@@ -41,18 +43,19 @@
     # build a pair of lists (sources, targets) where  source is newer
     n_sources = []
     n_targets = []
-    for source, target in zip(sources, targets):
-        if newer(source, target):
-            n_sources.append(source)
-            n_targets.append(target)
+    for i in range(len(sources)):
+        if newer(sources[i], targets[i]):
+            n_sources.append(sources[i])
+            n_targets.append(targets[i])
 
-    return n_sources, n_targets
+    return (n_sources, n_targets)
+
+# newer_pairwise ()
 
-def newer_group(sources, target, missing='error'):
-    """Return true if 'target' is out-of-date with respect to any file
-    listed in 'sources'.
 
-    In other words, if 'target' exists and is newer
+def newer_group (sources, target, missing='error'):
+    """Return true if 'target' is out-of-date with respect to any file
+    listed in 'sources'.  In other words, if 'target' exists and is newer
     than every file in 'sources', return false; otherwise return true.
     'missing' controls what we do when a source file is missing; the
     default ("error") is to blow up with an OSError from inside 'stat()';
@@ -65,14 +68,14 @@
     """
     # If the target doesn't even exist, then it's definitely out-of-date.
     if not os.path.exists(target):
-        return True
+        return 1
 
     # Otherwise we have to find out the hard way: if *any* source file
     # is more recent than 'target', then 'target' is out-of-date and
     # we can immediately return true.  If we fall through to the end
     # of the loop, then 'target' is up-to-date and we return false.
-    target_mtime = os.stat(target).st_mtime
-
+    from stat import ST_MTIME
+    target_mtime = os.stat(target)[ST_MTIME]
     for source in sources:
         if not os.path.exists(source):
             if missing == 'error':      # blow up when we stat() the file
@@ -80,9 +83,12 @@
             elif missing == 'ignore':   # missing source dropped from
                 continue                #  target's dependency list
             elif missing == 'newer':    # missing source means target is
-                return True             #  out-of-date
+                return 1                #  out-of-date
 
-        if os.stat(source).st_mtime > target_mtime:
-            return True
+        source_mtime = os.stat(source)[ST_MTIME]
+        if source_mtime > target_mtime:
+            return 1
+    else:
+        return 0
 
-    return False
+# newer_group ()

Modified: python/branches/import_unicode/Lib/distutils/dir_util.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/dir_util.py	(original)
+++ python/branches/import_unicode/Lib/distutils/dir_util.py	Fri Jul 30 01:19:37 2010
@@ -4,7 +4,7 @@
 
 __revision__ = "$Id$"
 
-import os
+import os, sys
 from distutils.errors import DistutilsFileError, DistutilsInternalError
 from distutils import log
 

Modified: python/branches/import_unicode/Lib/distutils/dist.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/dist.py	(original)
+++ python/branches/import_unicode/Lib/distutils/dist.py	Fri Jul 30 01:19:37 2010
@@ -7,15 +7,13 @@
 __revision__ = "$Id$"
 
 import sys, os, re
-from email import message_from_file
 
 try:
     import warnings
 except ImportError:
     warnings = None
 
-from distutils.errors import (DistutilsOptionError, DistutilsArgError,
-                              DistutilsModuleError, DistutilsClassError)
+from distutils.errors import *
 from distutils.fancy_getopt import FancyGetopt, translate_longopt
 from distutils.util import check_environ, strtobool, rfc822_escape
 from distutils import log
@@ -55,9 +53,7 @@
                       ('quiet', 'q', "run quietly (turns verbosity off)"),
                       ('dry-run', 'n', "don't actually do anything"),
                       ('help', 'h', "show detailed help message"),
-                      ('no-user-cfg', None,
-                       'ignore pydistutils.cfg in your home directory'),
-    ]
+                     ]
 
     # 'common_usage' is a short (2-3 line) string describing the common
     # usage of the setup script.
@@ -264,22 +260,6 @@
                     else:
                         sys.stderr.write(msg + "\n")
 
-        # no-user-cfg is handled before other command line args
-        # because other args override the config files, and this
-        # one is needed before we can load the config files.
-        # If attrs['script_args'] wasn't passed, assume false.
-        #
-        # This also make sure we just look at the global options
-        self.want_user_cfg = True
-
-        if self.script_args is not None:
-            for arg in self.script_args:
-                if not arg.startswith('-'):
-                    break
-                if arg == '--no-user-cfg':
-                    self.want_user_cfg = False
-                    break
-
         self.finalize_options()
 
     def get_option_dict(self, command):
@@ -331,10 +311,7 @@
         Distutils installation directory (ie. where the top-level
         Distutils __inst__.py file lives), a file in the user's home
         directory named .pydistutils.cfg on Unix and pydistutils.cfg
-        on Windows/Mac; and setup.cfg in the current directory.
-
-        The file in the user's home directory can be disabled with the
-        --no-user-cfg option.
+        on Windows/Mac, and setup.cfg in the current directory.
         """
         files = []
         check_environ()
@@ -354,19 +331,15 @@
             user_filename = "pydistutils.cfg"
 
         # And look for the user config file
-        if self.want_user_cfg:
-            user_file = os.path.join(os.path.expanduser('~'), user_filename)
-            if os.path.isfile(user_file):
-                files.append(user_file)
+        user_file = os.path.join(os.path.expanduser('~'), user_filename)
+        if os.path.isfile(user_file):
+            files.append(user_file)
 
         # All platforms support local setup.cfg
         local_file = "setup.cfg"
         if os.path.isfile(local_file):
             files.append(local_file)
 
-        if DEBUG:
-            self.announce("using config files: %s" % ', '.join(files))
-
         return files
 
     def parse_config_files(self, filenames=None):
@@ -1016,80 +989,25 @@
                          "provides", "requires", "obsoletes",
                          )
 
-    def __init__(self, path=None):
-        if path is not None:
-            self.read_pkg_file(open(path))
-        else:
-            self.name = None
-            self.version = None
-            self.author = None
-            self.author_email = None
-            self.maintainer = None
-            self.maintainer_email = None
-            self.url = None
-            self.license = None
-            self.description = None
-            self.long_description = None
-            self.keywords = None
-            self.platforms = None
-            self.classifiers = None
-            self.download_url = None
-            # PEP 314
-            self.provides = None
-            self.requires = None
-            self.obsoletes = None
-
-    def read_pkg_file(self, file):
-        """Reads the metadata values from a file object."""
-        msg = message_from_file(file)
-
-        def _read_field(name):
-            value = msg[name]
-            if value == 'UNKNOWN':
-                return None
-            return value
-
-        def _read_list(name):
-            values = msg.get_all(name, None)
-            if values == []:
-                return None
-            return values
-
-        metadata_version = msg['metadata-version']
-        self.name = _read_field('name')
-        self.version = _read_field('version')
-        self.description = _read_field('summary')
-        # we are filling author only.
-        self.author = _read_field('author')
+    def __init__ (self):
+        self.name = None
+        self.version = None
+        self.author = None
+        self.author_email = None
         self.maintainer = None
-        self.author_email = _read_field('author-email')
         self.maintainer_email = None
-        self.url = _read_field('home-page')
-        self.license = _read_field('license')
-
-        if 'download-url' in msg:
-            self.download_url = _read_field('download-url')
-        else:
-            self.download_url = None
-
-        self.long_description = _read_field('description')
-        self.description = _read_field('summary')
-
-        if 'keywords' in msg:
-            self.keywords = _read_field('keywords').split(',')
-
-        self.platforms = _read_list('platform')
-        self.classifiers = _read_list('classifier')
-
-        # PEP 314 - these fields only exist in 1.1
-        if metadata_version == '1.1':
-            self.requires = _read_list('requires')
-            self.provides = _read_list('provides')
-            self.obsoletes = _read_list('obsoletes')
-        else:
-            self.requires = None
-            self.provides = None
-            self.obsoletes = None
+        self.url = None
+        self.license = None
+        self.description = None
+        self.long_description = None
+        self.keywords = None
+        self.platforms = None
+        self.classifiers = None
+        self.download_url = None
+        # PEP 314
+        self.provides = None
+        self.requires = None
+        self.obsoletes = None
 
     def write_pkg_info(self, base_dir):
         """Write the PKG-INFO file into the release tree.

Modified: python/branches/import_unicode/Lib/distutils/emxccompiler.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/emxccompiler.py	(original)
+++ python/branches/import_unicode/Lib/distutils/emxccompiler.py	Fri Jul 30 01:19:37 2010
@@ -21,13 +21,12 @@
 
 __revision__ = "$Id$"
 
-import os, sys, copy
-from warnings import warn
-
+import os,sys,copy
+from distutils.ccompiler import gen_preprocess_options, gen_lib_options
 from distutils.unixccompiler import UnixCCompiler
 from distutils.file_util import write_file
 from distutils.errors import DistutilsExecError, CompileError, UnknownFileError
-from distutils.util import get_compiler_versions
+from distutils import log
 
 class EMXCCompiler (UnixCCompiler):
 
@@ -56,8 +55,8 @@
                 ("Reason: %s." % details) +
                 "Compiling may fail because of undefined preprocessor macros.")
 
-        gcc_version, ld_version, dllwrap_version = get_compiler_versions()
-        self.gcc_version, self.ld_version = gcc_version, ld_version
+        (self.gcc_version, self.ld_version) = \
+            get_versions()
         self.debug_print(self.compiler_type + ": gcc %s, ld %s\n" %
                          (self.gcc_version,
                           self.ld_version) )
@@ -292,11 +291,23 @@
     """ Try to find out the versions of gcc and ld.
         If not possible it returns None for it.
     """
-    warn("'distutils.emxccompiler.get_versions' is deprecated "
-         "use 'distutils.util.get_compiler_versions' instead",
-         DeprecationWarning)
-
+    from distutils.version import StrictVersion
+    from distutils.spawn import find_executable
+    import re
+
+    gcc_exe = find_executable('gcc')
+    if gcc_exe:
+        out = os.popen(gcc_exe + ' -dumpversion','r')
+        out_string = out.read()
+        out.close()
+        result = re.search('(\d+\.\d+\.\d+)', out_string, re.ASCII)
+        if result:
+            gcc_version = StrictVersion(result.group(1))
+        else:
+            gcc_version = None
+    else:
+        gcc_version = None
     # EMX ld has no way of reporting version number, and we use GCC
     # anyway - so we can link OMF DLLs
-    gcc_version, ld_version, dllwrap_version = get_compiler_versions()
-    return gcc_version, None
+    ld_version = None
+    return (gcc_version, ld_version)

Modified: python/branches/import_unicode/Lib/distutils/errors.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/errors.py	(original)
+++ python/branches/import_unicode/Lib/distutils/errors.py	Fri Jul 30 01:19:37 2010
@@ -10,79 +10,90 @@
 
 __revision__ = "$Id$"
 
-class DistutilsError(Exception):
+class DistutilsError (Exception):
     """The root of all Distutils evil."""
+    pass
 
-class DistutilsModuleError(DistutilsError):
+class DistutilsModuleError (DistutilsError):
     """Unable to load an expected module, or to find an expected class
     within some module (in particular, command modules and classes)."""
+    pass
 
-class DistutilsClassError(DistutilsError):
+class DistutilsClassError (DistutilsError):
     """Some command class (or possibly distribution class, if anyone
     feels a need to subclass Distribution) is found not to be holding
     up its end of the bargain, ie. implementing some part of the
     "command "interface."""
+    pass
 
-class DistutilsGetoptError(DistutilsError):
+class DistutilsGetoptError (DistutilsError):
     """The option table provided to 'fancy_getopt()' is bogus."""
+    pass
 
-class DistutilsArgError(DistutilsError):
+class DistutilsArgError (DistutilsError):
     """Raised by fancy_getopt in response to getopt.error -- ie. an
     error in the command line usage."""
+    pass
 
-class DistutilsFileError(DistutilsError):
+class DistutilsFileError (DistutilsError):
     """Any problems in the filesystem: expected file not found, etc.
     Typically this is for problems that we detect before IOError or
     OSError could be raised."""
+    pass
 
-class DistutilsOptionError(DistutilsError):
+class DistutilsOptionError (DistutilsError):
     """Syntactic/semantic errors in command options, such as use of
     mutually conflicting options, or inconsistent options,
     badly-spelled values, etc.  No distinction is made between option
     values originating in the setup script, the command line, config
     files, or what-have-you -- but if we *know* something originated in
     the setup script, we'll raise DistutilsSetupError instead."""
+    pass
 
-class DistutilsSetupError(DistutilsError):
+class DistutilsSetupError (DistutilsError):
     """For errors that can be definitely blamed on the setup script,
     such as invalid keyword arguments to 'setup()'."""
+    pass
 
-class DistutilsPlatformError(DistutilsError):
+class DistutilsPlatformError (DistutilsError):
     """We don't know how to do something on the current platform (but
     we do know how to do it on some platform) -- eg. trying to compile
     C files on a platform not supported by a CCompiler subclass."""
+    pass
 
-class DistutilsExecError(DistutilsError):
+class DistutilsExecError (DistutilsError):
     """Any problems executing an external program (such as the C
     compiler, when compiling C files)."""
+    pass
 
-class DistutilsInternalError(DistutilsError):
+class DistutilsInternalError (DistutilsError):
     """Internal inconsistencies or impossibilities (obviously, this
     should never be seen if the code is working!)."""
+    pass
 
-class DistutilsTemplateError(DistutilsError):
+class DistutilsTemplateError (DistutilsError):
     """Syntax error in a file list template."""
 
 class DistutilsByteCompileError(DistutilsError):
     """Byte compile error."""
 
 # Exception classes used by the CCompiler implementation classes
-class CCompilerError(Exception):
+class CCompilerError (Exception):
     """Some compile/link operation failed."""
 
-class PreprocessError(CCompilerError):
+class PreprocessError (CCompilerError):
     """Failure to preprocess one or more C/C++ files."""
 
-class CompileError(CCompilerError):
+class CompileError (CCompilerError):
     """Failure to compile one or more C/C++ source files."""
 
-class LibError(CCompilerError):
+class LibError (CCompilerError):
     """Failure to create a static library from one or more C/C++ object
     files."""
 
-class LinkError(CCompilerError):
+class LinkError (CCompilerError):
     """Failure to link one or more C/C++ object files into an executable
     or shared library file."""
 
-class UnknownFileError(CCompilerError):
+class UnknownFileError (CCompilerError):
     """Attempt to process an unknown file type."""

Modified: python/branches/import_unicode/Lib/distutils/extension.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/extension.py	(original)
+++ python/branches/import_unicode/Lib/distutils/extension.py	Fri Jul 30 01:19:37 2010
@@ -6,6 +6,7 @@
 __revision__ = "$Id$"
 
 import os
+import sys
 import warnings
 
 # This class is really only used by the "build_ext" command, so it might
@@ -134,17 +135,14 @@
 
 def read_setup_file(filename):
     """Reads a Setup file and returns Extension instances."""
-    warnings.warn('distutils.extensions.read_setup_file is deprecated. '
-                  'It will be removed in the next Python release.')
-    _sysconfig = __import__('sysconfig')
-    from distutils.sysconfig import (expand_makefile_vars,
+    from distutils.sysconfig import (parse_makefile, expand_makefile_vars,
                                      _variable_rx)
 
     from distutils.text_file import TextFile
     from distutils.util import split_quoted
 
     # First pass over the file to gather "VAR = VALUE" assignments.
-    vars = _sysconfig._parse_makefile(filename)
+    vars = parse_makefile(filename)
 
     # Second pass to gobble up the real content: lines of the form
     #   <module> ... [<sourcefile> ...] [<cpparg> ...] [<library> ...]
@@ -164,10 +162,7 @@
             file.warn("'%s' lines not handled yet" % line)
             continue
 
-        with warnings.catch_warnings():
-            warnings.simplefilter("ignore")
-            line = expand_makefile_vars(line, vars)
-
+        line = expand_makefile_vars(line, vars)
         words = split_quoted(line)
 
         # NB. this parses a slightly different syntax than the old

Modified: python/branches/import_unicode/Lib/distutils/fancy_getopt.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/fancy_getopt.py	(original)
+++ python/branches/import_unicode/Lib/distutils/fancy_getopt.py	Fri Jul 30 01:19:37 2010
@@ -10,11 +10,9 @@
 
 __revision__ = "$Id$"
 
-import sys
-import string
-import re
+import sys, string, re
 import getopt
-from distutils.errors import DistutilsGetoptError, DistutilsArgError
+from distutils.errors import *
 
 # Much like command_re in distutils.core, this is close to but not quite
 # the same as a Python NAME -- except, in the spirit of most GNU
@@ -446,3 +444,16 @@
         'options' will be initialized to None."""
         for opt in options:
             setattr(self, opt, None)
+
+
+if __name__ == "__main__":
+    text = """\
+Tra-la-la, supercalifragilisticexpialidocious.
+How *do* you spell that odd word, anyways?
+(Someone ask Mary -- she'll know [or she'll
+say, "How should I know?"].)"""
+
+    for w in (10, 20, 30, 40):
+        print("width: %d" % w)
+        print("\n".join(wrap_text(text, w)))
+        print()

Modified: python/branches/import_unicode/Lib/distutils/file_util.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/file_util.py	(original)
+++ python/branches/import_unicode/Lib/distutils/file_util.py	Fri Jul 30 01:19:37 2010
@@ -10,18 +10,17 @@
 from distutils import log
 
 # for generating verbose output in 'copy_file()'
-_copy_action = {None: 'copying',
-                'hard': 'hard linking',
-                'sym': 'symbolically linking'}
+_copy_action = { None:   'copying',
+                 'hard': 'hard linking',
+                 'sym':  'symbolically linking' }
 
 
 def _copy_file_contents(src, dst, buffer_size=16*1024):
-    """Copy the file 'src' to 'dst'.
-
-    Both must be filenames. Any error opening either file, reading from
-    'src', or writing to 'dst', raises DistutilsFileError.  Data is
-    read/written in chunks of 'buffer_size' bytes (default 16k).  No attempt
-    is made to handle anything apart from regular files.
+    """Copy the file 'src' to 'dst'; both must be filenames.  Any error
+    opening either file, reading from 'src', or writing to 'dst', raises
+    DistutilsFileError.  Data is read/written in chunks of 'buffer_size'
+    bytes (default 16k).  No attempt is made to handle anything apart from
+    regular files.
     """
     # Stolen from shutil module in the standard library, but with
     # custom error-handling added.
@@ -69,16 +68,15 @@
 
 def copy_file(src, dst, preserve_mode=1, preserve_times=1, update=0,
               link=None, verbose=1, dry_run=0):
-    """Copy a file 'src' to 'dst'.
-
-    If 'dst' is a directory, then 'src' is copied there with the same name;
-    otherwise, it must be a filename.  (If the file exists, it will be
-    ruthlessly clobbered.)  If 'preserve_mode' is true (the default),
-    the file's mode (type and permission bits, or whatever is analogous on
-    the current platform) is copied.  If 'preserve_times' is true (the
-    default), the last-modified and last-access times are copied as well.
-    If 'update' is true, 'src' will only be copied if 'dst' does not exist,
-    or if 'dst' does exist but is older than 'src'.
+    """Copy a file 'src' to 'dst'.  If 'dst' is a directory, then 'src' is
+    copied there with the same name; otherwise, it must be a filename.  (If
+    the file exists, it will be ruthlessly clobbered.)  If 'preserve_mode'
+    is true (the default), the file's mode (type and permission bits, or
+    whatever is analogous on the current platform) is copied.  If
+    'preserve_times' is true (the default), the last-modified and
+    last-access times are copied as well.  If 'update' is true, 'src' will
+    only be copied if 'dst' does not exist, or if 'dst' does exist but is
+    older than 'src'.
 
     'link' allows you to make hard links (os.link) or symbolic links
     (os.symlink) instead of copying: set it to "hard" or "sym"; if it is
@@ -132,6 +130,15 @@
     if dry_run:
         return (dst, 1)
 
+    # On Mac OS, use the native file copy routine
+    if os.name == 'mac':
+        import macostools
+        try:
+            macostools.copy(src, dst, 0, preserve_times)
+        except os.error as exc:
+            raise DistutilsFileError(
+                  "could not copy '%s' to '%s': %s" % (src, dst, exc.args[-1]))
+
     # If linking (hard or symbolic), use the appropriate system call
     # (Unix only, of course, but that's the caller's responsibility)
     elif link == 'hard':
@@ -159,12 +166,13 @@
 
 
 # XXX I suspect this is Unix-specific -- need porting help!
-def move_file(src, dst, verbose=1, dry_run=0):
-    """Move a file 'src' to 'dst'.
-
-    If 'dst' is a directory, the file will be moved into it with the same
-    name; otherwise, 'src' is just renamed to 'dst'.  Return the new
-    full name of the file.
+def move_file (src, dst,
+               verbose=1,
+               dry_run=0):
+
+    """Move a file 'src' to 'dst'.  If 'dst' is a directory, the file will
+    be moved into it with the same name; otherwise, 'src' is just renamed
+    to 'dst'.  Return the new full name of the file.
 
     Handles cross-device moves on Unix using 'copy_file()'.  What about
     other systems???
@@ -221,7 +229,7 @@
     return dst
 
 
-def write_file(filename, contents):
+def write_file (filename, contents):
     """Create a file with the specified name and write 'contents' (a
     sequence of strings without line terminators) to it.
     """

Modified: python/branches/import_unicode/Lib/distutils/filelist.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/filelist.py	(original)
+++ python/branches/import_unicode/Lib/distutils/filelist.py	Fri Jul 30 01:19:37 2010
@@ -108,7 +108,7 @@
         # defined: it's the first word of the line.  Which of the other
         # three are defined depends on the action; it'll be either
         # patterns, (dir and patterns), or (dir_pattern).
-        action, patterns, dir, dir_pattern = self._parse_template_line(line)
+        (action, patterns, dir, dir_pattern) = self._parse_template_line(line)
 
         # OK, now we know that the action is valid and we have the
         # right number of words on the line for that action -- so we
@@ -175,15 +175,15 @@
             raise DistutilsInternalError(
                   "this cannot happen: invalid action '%s'" % action)
 
+
     # -- Filtering/selection methods -----------------------------------
 
     def include_pattern(self, pattern, anchor=1, prefix=None, is_regex=0):
         """Select strings (presumably filenames) from 'self.files' that
-        match 'pattern', a Unix-style wildcard (glob) pattern.
-
-        Patterns are not quite the same as implemented by the 'fnmatch'
-        module: '*' and '?'  match non-special characters, where "special"
-        is platform-dependent: slash on Unix; colon, slash, and backslash on
+        match 'pattern', a Unix-style wildcard (glob) pattern.  Patterns
+        are not quite the same as implemented by the 'fnmatch' module: '*'
+        and '?'  match non-special characters, where "special" is platform-
+        dependent: slash on Unix; colon, slash, and backslash on
         DOS/Windows; and colon on Mac OS.
 
         If 'anchor' is true (the default), then the pattern match is more
@@ -220,13 +220,13 @@
         return files_found
 
 
-    def exclude_pattern(self, pattern, anchor=1, prefix=None, is_regex=0):
+    def exclude_pattern (self, pattern,
+                         anchor=1, prefix=None, is_regex=0):
         """Remove strings (presumably filenames) from 'files' that match
-        'pattern'.
-
-        Other parameters are the same as for 'include_pattern()', above.
-        The list 'self.files' is modified in place. Return 1 if files are
-        found.
+        'pattern'.  Other parameters are the same as for
+        'include_pattern()', above.
+        The list 'self.files' is modified in place.
+        Return True if files are found, False otherwise.
         """
         files_found = False
         pattern_re = translate_pattern(pattern, anchor, prefix, is_regex)
@@ -275,11 +275,10 @@
 
 
 def glob_to_re(pattern):
-    """Translate a shell-like glob pattern to a regular expression.
-
-    Return a string containing the regex.  Differs from
-    'fnmatch.translate()' in that '*' does not match "special characters"
-    (which are platform-specific).
+    """Translate a shell-like glob pattern to a regular expression; return
+    a string containing the regex.  Differs from 'fnmatch.translate()' in
+    that '*' does not match "special characters" (which are
+    platform-specific).
     """
     pattern_re = fnmatch.translate(pattern)
 
@@ -297,9 +296,7 @@
 
 def translate_pattern(pattern, anchor=1, prefix=None, is_regex=0):
     """Translate a shell-like wildcard pattern to a compiled regular
-    expression.
-
-    Return the compiled regex.  If 'is_regex' true,
+    expression.  Return the compiled regex.  If 'is_regex' true,
     then 'pattern' is directly compiled to a regex (if it's a string)
     or just returned as-is (assumes it's a regex object).
     """
@@ -317,7 +314,7 @@
     if prefix is not None:
         # ditch end of pattern character
         empty_pattern = glob_to_re('')
-        prefix_re = glob_to_re(prefix)[:-len(empty_pattern)]
+        prefix_re = (glob_to_re(prefix))[:-len(empty_pattern)]
         pattern_re = "^" + os.path.join(prefix_re, ".*" + pattern_re)
     else:                               # no prefix -- respect anchor flag
         if anchor:

Modified: python/branches/import_unicode/Lib/distutils/msvc9compiler.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/msvc9compiler.py	(original)
+++ python/branches/import_unicode/Lib/distutils/msvc9compiler.py	Fri Jul 30 01:19:37 2010
@@ -19,9 +19,10 @@
 import sys
 import re
 
-from distutils.errors import (DistutilsExecError, DistutilsPlatformError,
-                              CompileError, LibError, LinkError)
-from distutils.ccompiler import CCompiler, gen_lib_options
+from distutils.errors import DistutilsExecError, DistutilsPlatformError, \
+                             CompileError, LibError, LinkError
+from distutils.ccompiler import CCompiler, gen_preprocess_options, \
+                                gen_lib_options
 from distutils import log
 from distutils.util import get_platform
 

Modified: python/branches/import_unicode/Lib/distutils/msvccompiler.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/msvccompiler.py	(original)
+++ python/branches/import_unicode/Lib/distutils/msvccompiler.py	Fri Jul 30 01:19:37 2010
@@ -10,12 +10,12 @@
 
 __revision__ = "$Id$"
 
-import sys
-import os
-
-from distutils.errors import (DistutilsExecError, DistutilsPlatformError,
-                              CompileError, LibError, LinkError)
-from distutils.ccompiler import CCompiler, gen_lib_options
+import sys, os
+from distutils.errors import \
+     DistutilsExecError, DistutilsPlatformError, \
+     CompileError, LibError, LinkError
+from distutils.ccompiler import \
+     CCompiler, gen_preprocess_options, gen_lib_options
 from distutils import log
 
 _can_read_reg = False
@@ -124,7 +124,7 @@
                 self.set_macro("FrameworkSDKDir", net, "sdkinstallrootv1.1")
             else:
                 self.set_macro("FrameworkSDKDir", net, "sdkinstallroot")
-        except KeyError:
+        except KeyError as exc: #
             raise DistutilsPlatformError(
             """Python was built with Visual Studio 2003;
 extensions must be built with a compiler than can generate compatible binaries.

Modified: python/branches/import_unicode/Lib/distutils/sysconfig.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/sysconfig.py	(original)
+++ python/branches/import_unicode/Lib/distutils/sysconfig.py	Fri Jul 30 01:19:37 2010
@@ -7,9 +7,6 @@
 
 Written by:   Fred L. Drake, Jr.
 Email:        <fdrake at acm.org>
-
-**This module has been moved out of Distutils and will be removed from
-Python in the next version (3.3)**
 """
 
 __revision__ = "$Id$"
@@ -17,45 +14,51 @@
 import io
 import os
 import re
-from warnings import warn
-
-from distutils.errors import DistutilsPlatformError
+import sys
 
-# importing sysconfig from Lib
-# to avoid this module to shadow it
-_sysconfig = __import__('sysconfig')
-
-# names defined here to keep backward compatibility
-# for APIs that were relocated
-get_python_version = _sysconfig.get_python_version
-get_config_h_filename = _sysconfig.get_config_h_filename
-parse_config_h = _sysconfig.parse_config_h
-get_config_vars = _sysconfig.get_config_vars
-get_config_var = _sysconfig.get_config_var
-from distutils.ccompiler import customize_compiler
-
-_DEPRECATION_MSG = ("distutils.sysconfig.%s is deprecated. "
-                    "Use the APIs provided by the sysconfig module instead")
-
-def _get_project_base():
-    return _sysconfig._PROJECT_BASE
-
-project_base = _get_project_base()
-
-class _DeprecatedBool(int):
-    def __nonzero__(self):
-        warn(_DEPRECATION_MSG % 'get_python_version', DeprecationWarning)
-        return super(_DeprecatedBool, self).__nonzero__()
+from .errors import DistutilsPlatformError
 
+# These are needed in a couple of spots, so just compute them once.
+PREFIX = os.path.normpath(sys.prefix)
+EXEC_PREFIX = os.path.normpath(sys.exec_prefix)
+
+# Path to the base directory of the project. On Windows the binary may
+# live in project/PCBuild9.  If we're dealing with an x64 Windows build,
+# it'll live in project/PCbuild/amd64.
+project_base = os.path.dirname(os.path.abspath(sys.executable))
+if os.name == "nt" and "pcbuild" in project_base[-8:].lower():
+    project_base = os.path.abspath(os.path.join(project_base, os.path.pardir))
+# PC/VS7.1
+if os.name == "nt" and "\\pc\\v" in project_base[-10:].lower():
+    project_base = os.path.abspath(os.path.join(project_base, os.path.pardir,
+                                                os.path.pardir))
+# PC/AMD64
+if os.name == "nt" and "\\pcbuild\\amd64" in project_base[-14:].lower():
+    project_base = os.path.abspath(os.path.join(project_base, os.path.pardir,
+                                                os.path.pardir))
+
+# python_build: (Boolean) if true, we're either building Python or
+# building an extension with an un-installed Python, so we use
+# different (hard-wired) directories.
+# Setup.local is available for Makefile builds including VPATH builds,
+# Setup.dist is available on Windows
 def _python_build():
-    return _DeprecatedBool(_sysconfig.is_python_build())
-
+    for fn in ("Setup.dist", "Setup.local"):
+        if os.path.isfile(os.path.join(project_base, "Modules", fn)):
+            return True
+    return False
 python_build = _python_build()
 
-def get_python_inc(plat_specific=0, prefix=None):
-    """This function is deprecated.
+def get_python_version():
+    """Return a string containing the major and minor Python version,
+    leaving off the patchlevel.  Sample return values could be '1.5'
+    or '2.2'.
+    """
+    return sys.version[:3]
+
 
-    Return the directory containing installed Python header files.
+def get_python_inc(plat_specific=0, prefix=None):
+    """Return the directory containing installed Python header files.
 
     If 'plat_specific' is false (the default), this is the path to the
     non-platform-specific header files, i.e. Python.h and so on;
@@ -65,22 +68,39 @@
     If 'prefix' is supplied, use it instead of sys.prefix or
     sys.exec_prefix -- i.e., ignore 'plat_specific'.
     """
-    warn(_DEPRECATION_MSG % 'get_python_inc', DeprecationWarning)
-    get_path = _sysconfig.get_path
-
-    if prefix is not None:
-        vars = {'base': prefix}
-        return get_path('include', vars=vars)
-
-    if not plat_specific:
-        return get_path('include')
+    if prefix is None:
+        prefix = plat_specific and EXEC_PREFIX or PREFIX
+    if os.name == "posix":
+        if python_build:
+            # Assume the executable is in the build directory.  The
+            # pyconfig.h file should be in the same directory.  Since
+            # the build directory may not be the source directory, we
+            # must use "srcdir" from the makefile to find the "Include"
+            # directory.
+            base = os.path.dirname(os.path.abspath(sys.executable))
+            if plat_specific:
+                return base
+            else:
+                incdir = os.path.join(get_config_var('srcdir'), 'Include')
+                return os.path.normpath(incdir)
+        return os.path.join(prefix, "include", "python" + get_python_version())
+    elif os.name == "nt":
+        return os.path.join(prefix, "include")
+    elif os.name == "mac":
+        if plat_specific:
+            return os.path.join(prefix, "Mac", "Include")
+        else:
+            return os.path.join(prefix, "Include")
+    elif os.name == "os2":
+        return os.path.join(prefix, "Include")
     else:
-        return get_path('platinclude')
+        raise DistutilsPlatformError(
+            "I don't know where Python installs its C header files "
+            "on platform '%s'" % os.name)
 
-def get_python_lib(plat_specific=False, standard_lib=False, prefix=None):
-    """This function is deprecated.
 
-    Return the directory containing the Python library (standard or
+def get_python_lib(plat_specific=0, standard_lib=0, prefix=None):
+    """Return the directory containing the Python library (standard or
     site additions).
 
     If 'plat_specific' is true, return the directory containing
@@ -93,33 +113,149 @@
     If 'prefix' is supplied, use it instead of sys.prefix or
     sys.exec_prefix -- i.e., ignore 'plat_specific'.
     """
-    warn(_DEPRECATION_MSG % 'get_python_lib', DeprecationWarning)
-    vars = {}
-    get_path = _sysconfig.get_path
-    if prefix is not None:
-        if plat_specific:
-            vars['platbase'] = prefix
+    if prefix is None:
+        prefix = plat_specific and EXEC_PREFIX or PREFIX
+
+    if os.name == "posix":
+        libpython = os.path.join(prefix,
+                                 "lib", "python" + get_python_version())
+        if standard_lib:
+            return libpython
+        else:
+            return os.path.join(libpython, "site-packages")
+    elif os.name == "nt":
+        if standard_lib:
+            return os.path.join(prefix, "Lib")
         else:
-            vars['base'] = prefix
-    if standard_lib:
+            if get_python_version() < "2.2":
+                return prefix
+            else:
+                return os.path.join(prefix, "Lib", "site-packages")
+    elif os.name == "mac":
         if plat_specific:
-            return get_path('platstdlib', vars=vars)
+            if standard_lib:
+                return os.path.join(prefix, "Lib", "lib-dynload")
+            else:
+                return os.path.join(prefix, "Lib", "site-packages")
         else:
-            return get_path('stdlib', vars=vars)
+            if standard_lib:
+                return os.path.join(prefix, "Lib")
+            else:
+                return os.path.join(prefix, "Lib", "site-packages")
+    elif os.name == "os2":
+        if standard_lib:
+            return os.path.join(prefix, "Lib")
+        else:
+            return os.path.join(prefix, "Lib", "site-packages")
     else:
-        if plat_specific:
-            return get_path('platlib', vars=vars)
+        raise DistutilsPlatformError(
+            "I don't know where Python installs its library "
+            "on platform '%s'" % os.name)
+
+
+def customize_compiler(compiler):
+    """Do any platform-specific customization of a CCompiler instance.
+
+    Mainly needed on Unix, so we can plug in the information that
+    varies across Unices and is stored in Python's Makefile.
+    """
+    if compiler.compiler_type == "unix":
+        (cc, cxx, opt, cflags, ccshared, ldshared, so_ext, ar, ar_flags) = \
+            get_config_vars('CC', 'CXX', 'OPT', 'CFLAGS',
+                            'CCSHARED', 'LDSHARED', 'SO', 'AR', 'ARFLAGS')
+
+        if 'CC' in os.environ:
+            cc = os.environ['CC']
+        if 'CXX' in os.environ:
+            cxx = os.environ['CXX']
+        if 'LDSHARED' in os.environ:
+            ldshared = os.environ['LDSHARED']
+        if 'CPP' in os.environ:
+            cpp = os.environ['CPP']
         else:
-            return get_path('purelib', vars=vars)
+            cpp = cc + " -E"           # not always
+        if 'LDFLAGS' in os.environ:
+            ldshared = ldshared + ' ' + os.environ['LDFLAGS']
+        if 'CFLAGS' in os.environ:
+            cflags = opt + ' ' + os.environ['CFLAGS']
+            ldshared = ldshared + ' ' + os.environ['CFLAGS']
+        if 'CPPFLAGS' in os.environ:
+            cpp = cpp + ' ' + os.environ['CPPFLAGS']
+            cflags = cflags + ' ' + os.environ['CPPFLAGS']
+            ldshared = ldshared + ' ' + os.environ['CPPFLAGS']
+        if 'AR' in os.environ:
+            ar = os.environ['AR']
+        if 'ARFLAGS' in os.environ:
+            archiver = ar + ' ' + os.environ['ARFLAGS']
+        else:
+            archiver = ar + ' ' + ar_flags
+
+        cc_cmd = cc + ' ' + cflags
+        compiler.set_executables(
+            preprocessor=cpp,
+            compiler=cc_cmd,
+            compiler_so=cc_cmd + ' ' + ccshared,
+            compiler_cxx=cxx,
+            linker_so=ldshared,
+            linker_exe=cc,
+            archiver=archiver)
+
+        compiler.shared_lib_extension = so_ext
+
+
+def get_config_h_filename():
+    """Return full pathname of installed pyconfig.h file."""
+    if python_build:
+        if os.name == "nt":
+            inc_dir = os.path.join(project_base, "PC")
+        else:
+            inc_dir = project_base
+    else:
+        inc_dir = get_python_inc(plat_specific=1)
+    if get_python_version() < '2.2':
+        config_h = 'config.h'
+    else:
+        # The name of the config.h file changed in 2.2
+        config_h = 'pyconfig.h'
+    return os.path.join(inc_dir, config_h)
+
 
 def get_makefile_filename():
-    """This function is deprecated.
+    """Return full pathname of installed Makefile from the Python build."""
+    if python_build:
+        return os.path.join(os.path.dirname(sys.executable), "Makefile")
+    lib_dir = get_python_lib(plat_specific=1, standard_lib=1)
+    return os.path.join(lib_dir, "config", "Makefile")
+
+
+def parse_config_h(fp, g=None):
+    """Parse a config.h-style file.
 
-    Return full pathname of installed Makefile from the Python build.
+    A dictionary containing name/value pairs is returned.  If an
+    optional dictionary is passed in as the second argument, it is
+    used instead of a new dictionary.
     """
+    if g is None:
+        g = {}
+    define_rx = re.compile("#define ([A-Z][A-Za-z0-9_]+) (.*)\n")
+    undef_rx = re.compile("/[*] #undef ([A-Z][A-Za-z0-9_]+) [*]/\n")
+    #
+    while True:
+        line = fp.readline()
+        if not line:
+            break
+        m = define_rx.match(line)
+        if m:
+            n, v = m.group(1, 2)
+            try: v = int(v)
+            except ValueError: pass
+            g[n] = v
+        else:
+            m = undef_rx.match(line)
+            if m:
+                g[m.group(1)] = 0
+    return g
 
-    warn(_DEPRECATION_MSG % 'get_makefile_filename', DeprecationWarning)
-    return _sysconfig._get_makefile_filename()
 
 # Regexes needed for parsing Makefile (and similar syntaxes,
 # like old-style Setup files).
@@ -128,29 +264,114 @@
 _findvar2_rx = re.compile(r"\${([A-Za-z][A-Za-z0-9_]*)}")
 
 def parse_makefile(fn, g=None):
-    """This function is deprecated.
-
-    Parse a Makefile-style file.
+    """Parse a Makefile-style file.
 
     A dictionary containing name/value pairs is returned.  If an
     optional dictionary is passed in as the second argument, it is
     used instead of a new dictionary.
     """
-    warn(_DEPRECATION_MSG % 'parse_makefile', DeprecationWarning)
-    return _sysconfig._parse_makefile(fn, g)
+    from distutils.text_file import TextFile
+    fp = TextFile(fn, strip_comments=1, skip_blanks=1, join_lines=1)
+
+    if g is None:
+        g = {}
+    done = {}
+    notdone = {}
+
+    while True:
+        line = fp.readline()
+        if line is None: # eof
+            break
+        m = _variable_rx.match(line)
+        if m:
+            n, v = m.group(1, 2)
+            v = v.strip()
+            # `$$' is a literal `$' in make
+            tmpv = v.replace('$$', '')
+
+            if "$" in tmpv:
+                notdone[n] = v
+            else:
+                try:
+                    v = int(v)
+                except ValueError:
+                    # insert literal `$'
+                    done[n] = v.replace('$$', '$')
+                else:
+                    done[n] = v
+
+    # Variables with a 'PY_' prefix in the makefile. These need to
+    # be made available without that prefix through sysconfig.
+    # Special care is needed to ensure that variable expansion works, even
+    # if the expansion uses the name without a prefix.
+    renamed_variables = ('CFLAGS', 'LDFLAGS', 'CPPFLAGS')
+
+    # do variable interpolation here
+    while notdone:
+        for name in list(notdone):
+            value = notdone[name]
+            m = _findvar1_rx.search(value) or _findvar2_rx.search(value)
+            if m:
+                n = m.group(1)
+                found = True
+                if n in done:
+                    item = str(done[n])
+                elif n in notdone:
+                    # get it on a subsequent round
+                    found = False
+                elif n in os.environ:
+                    # do it like make: fall back to environment
+                    item = os.environ[n]
+
+                elif n in renamed_variables:
+                    if name.startswith('PY_') and name[3:] in renamed_variables:
+                        item = ""
+
+                    elif 'PY_' + n in notdone:
+                        found = False
+
+                    else:
+                        item = str(done['PY_' + n])
+                else:
+                    done[n] = item = ""
+                if found:
+                    after = value[m.end():]
+                    value = value[:m.start()] + item + after
+                    if "$" in after:
+                        notdone[name] = value
+                    else:
+                        try: value = int(value)
+                        except ValueError:
+                            done[name] = value.strip()
+                        else:
+                            done[name] = value
+                        del notdone[name]
+
+                        if name.startswith('PY_') \
+                            and name[3:] in renamed_variables:
+
+                            name = name[3:]
+                            if name not in done:
+                                done[name] = value
+            else:
+                # bogus variable reference; just drop it since we can't deal
+                del notdone[name]
+
+    fp.close()
+
+    # save the results in the global dictionary
+    g.update(done)
+    return g
 
-def expand_makefile_vars(s, vars):
-    """This function is deprecated.
 
-    Expand Makefile-style variables -- "${foo}" or "$(foo)" -- in
+def expand_makefile_vars(s, vars):
+    """Expand Makefile-style variables -- "${foo}" or "$(foo)" -- in
     'string' according to 'vars' (a dictionary mapping variable names to
     values).  Variables not present in 'vars' are silently expanded to the
     empty string.  The variable values in 'vars' should not contain further
     variable expansions; if 'vars' is the output of 'parse_makefile()',
     you're fine.  Returns a variable-expanded version of 's'.
     """
-    warn('this function will be removed in then next version of Python',
-         DeprecationWarning)
 
     # This algorithm does multiple expansion, so if vars['foo'] contains
     # "${bar}", it will expand ${foo} to ${bar}, and then expand
@@ -166,3 +387,220 @@
         else:
             break
     return s
+
+
+_config_vars = None
+
+def _init_posix():
+    """Initialize the module as appropriate for POSIX systems."""
+    g = {}
+    # load the installed Makefile:
+    try:
+        filename = get_makefile_filename()
+        parse_makefile(filename, g)
+    except IOError as msg:
+        my_msg = "invalid Python installation: unable to open %s" % filename
+        if hasattr(msg, "strerror"):
+            my_msg = my_msg + " (%s)" % msg.strerror
+
+        raise DistutilsPlatformError(my_msg)
+
+    # load the installed pyconfig.h:
+    try:
+        filename = get_config_h_filename()
+        parse_config_h(io.open(filename), g)
+    except IOError as msg:
+        my_msg = "invalid Python installation: unable to open %s" % filename
+        if hasattr(msg, "strerror"):
+            my_msg = my_msg + " (%s)" % msg.strerror
+
+        raise DistutilsPlatformError(my_msg)
+
+    # On MacOSX we need to check the setting of the environment variable
+    # MACOSX_DEPLOYMENT_TARGET: configure bases some choices on it so
+    # it needs to be compatible.
+    # If it isn't set we set it to the configure-time value
+    if sys.platform == 'darwin' and 'MACOSX_DEPLOYMENT_TARGET' in g:
+        cfg_target = g['MACOSX_DEPLOYMENT_TARGET']
+        cur_target = os.getenv('MACOSX_DEPLOYMENT_TARGET', '')
+        if cur_target == '':
+            cur_target = cfg_target
+            os.putenv('MACOSX_DEPLOYMENT_TARGET', cfg_target)
+        elif [int(x) for x in cfg_target.split('.')] > [int(x) for x in cur_target.split('.')]:
+            my_msg = ('$MACOSX_DEPLOYMENT_TARGET mismatch: now "%s" but "%s" during configure'
+                % (cur_target, cfg_target))
+            raise DistutilsPlatformError(my_msg)
+
+    # On AIX, there are wrong paths to the linker scripts in the Makefile
+    # -- these paths are relative to the Python source, but when installed
+    # the scripts are in another directory.
+    if python_build:
+        g['LDSHARED'] = g['BLDSHARED']
+
+    elif get_python_version() < '2.1':
+        # The following two branches are for 1.5.2 compatibility.
+        if sys.platform == 'aix4':          # what about AIX 3.x ?
+            # Linker script is in the config directory, not in Modules as the
+            # Makefile says.
+            python_lib = get_python_lib(standard_lib=1)
+            ld_so_aix = os.path.join(python_lib, 'config', 'ld_so_aix')
+            python_exp = os.path.join(python_lib, 'config', 'python.exp')
+
+            g['LDSHARED'] = "%s %s -bI:%s" % (ld_so_aix, g['CC'], python_exp)
+
+    global _config_vars
+    _config_vars = g
+
+
+def _init_nt():
+    """Initialize the module as appropriate for NT"""
+    g = {}
+    # set basic install directories
+    g['LIBDEST'] = get_python_lib(plat_specific=0, standard_lib=1)
+    g['BINLIBDEST'] = get_python_lib(plat_specific=1, standard_lib=1)
+
+    # XXX hmmm.. a normal install puts include files here
+    g['INCLUDEPY'] = get_python_inc(plat_specific=0)
+
+    g['SO'] = '.pyd'
+    g['EXE'] = ".exe"
+    g['VERSION'] = get_python_version().replace(".", "")
+    g['BINDIR'] = os.path.dirname(os.path.abspath(sys.executable))
+
+    global _config_vars
+    _config_vars = g
+
+
+def _init_mac():
+    """Initialize the module as appropriate for Macintosh systems"""
+    g = {}
+    # set basic install directories
+    g['LIBDEST'] = get_python_lib(plat_specific=0, standard_lib=1)
+    g['BINLIBDEST'] = get_python_lib(plat_specific=1, standard_lib=1)
+
+    # XXX hmmm.. a normal install puts include files here
+    g['INCLUDEPY'] = get_python_inc(plat_specific=0)
+
+    import MacOS
+    if not hasattr(MacOS, 'runtimemodel'):
+        g['SO'] = '.ppc.slb'
+    else:
+        g['SO'] = '.%s.slb' % MacOS.runtimemodel
+
+    # XXX are these used anywhere?
+    g['install_lib'] = os.path.join(EXEC_PREFIX, "Lib")
+    g['install_platlib'] = os.path.join(EXEC_PREFIX, "Mac", "Lib")
+
+    # These are used by the extension module build
+    g['srcdir'] = ':'
+    global _config_vars
+    _config_vars = g
+
+
+def _init_os2():
+    """Initialize the module as appropriate for OS/2"""
+    g = {}
+    # set basic install directories
+    g['LIBDEST'] = get_python_lib(plat_specific=0, standard_lib=1)
+    g['BINLIBDEST'] = get_python_lib(plat_specific=1, standard_lib=1)
+
+    # XXX hmmm.. a normal install puts include files here
+    g['INCLUDEPY'] = get_python_inc(plat_specific=0)
+
+    g['SO'] = '.pyd'
+    g['EXE'] = ".exe"
+
+    global _config_vars
+    _config_vars = g
+
+
+def get_config_vars(*args):
+    """With no arguments, return a dictionary of all configuration
+    variables relevant for the current platform.  Generally this includes
+    everything needed to build extensions and install both pure modules and
+    extensions.  On Unix, this means every variable defined in Python's
+    installed Makefile; on Windows and Mac OS it's a much smaller set.
+
+    With arguments, return a list of values that result from looking up
+    each argument in the configuration variable dictionary.
+    """
+    global _config_vars
+    if _config_vars is None:
+        func = globals().get("_init_" + os.name)
+        if func:
+            func()
+        else:
+            _config_vars = {}
+
+        # Normalized versions of prefix and exec_prefix are handy to have;
+        # in fact, these are the standard versions used most places in the
+        # Distutils.
+        _config_vars['prefix'] = PREFIX
+        _config_vars['exec_prefix'] = EXEC_PREFIX
+
+        # Convert srcdir into an absolute path if it appears necessary.
+        # Normally it is relative to the build directory.  However, during
+        # testing, for example, we might be running a non-installed python
+        # from a different directory.
+        if python_build and os.name == "posix":
+            base = os.path.dirname(os.path.abspath(sys.executable))
+            if (not os.path.isabs(_config_vars['srcdir']) and
+                base != os.getcwd()):
+                # srcdir is relative and we are not in the same directory
+                # as the executable. Assume executable is in the build
+                # directory and make srcdir absolute.
+                srcdir = os.path.join(base, _config_vars['srcdir'])
+                _config_vars['srcdir'] = os.path.normpath(srcdir)
+
+        if sys.platform == 'darwin':
+            kernel_version = os.uname()[2] # Kernel version (8.4.3)
+            major_version = int(kernel_version.split('.')[0])
+
+            if major_version < 8:
+                # On Mac OS X before 10.4, check if -arch and -isysroot
+                # are in CFLAGS or LDFLAGS and remove them if they are.
+                # This is needed when building extensions on a 10.3 system
+                # using a universal build of python.
+                for key in ('LDFLAGS', 'BASECFLAGS',
+                        # a number of derived variables. These need to be
+                        # patched up as well.
+                        'CFLAGS', 'PY_CFLAGS', 'BLDSHARED'):
+                    flags = _config_vars[key]
+                    flags = re.sub('-arch\s+\w+\s', ' ', flags, re.ASCII)
+                    flags = re.sub('-isysroot [^ \t]*', ' ', flags)
+                    _config_vars[key] = flags
+
+            else:
+
+                # Allow the user to override the architecture flags using
+                # an environment variable.
+                # NOTE: This name was introduced by Apple in OSX 10.5 and
+                # is used by several scripting languages distributed with
+                # that OS release.
+
+                if 'ARCHFLAGS' in os.environ:
+                    arch = os.environ['ARCHFLAGS']
+                    for key in ('LDFLAGS', 'BASECFLAGS',
+                        # a number of derived variables. These need to be
+                        # patched up as well.
+                        'CFLAGS', 'PY_CFLAGS', 'BLDSHARED'):
+
+                        flags = _config_vars[key]
+                        flags = re.sub('-arch\s+\w+\s', ' ', flags)
+                        flags = flags + ' ' + arch
+                        _config_vars[key] = flags
+
+    if args:
+        vals = []
+        for name in args:
+            vals.append(_config_vars.get(name))
+        return vals
+    else:
+        return _config_vars
+
+def get_config_var(name):
+    """Return the value of a single variable using the dictionary
+    returned by 'get_config_vars()'.  Equivalent to
+    get_config_vars().get(name)
+    """
+    return get_config_vars().get(name)

Modified: python/branches/import_unicode/Lib/distutils/tests/support.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/support.py	(original)
+++ python/branches/import_unicode/Lib/distutils/tests/support.py	Fri Jul 30 01:19:37 2010
@@ -3,19 +3,11 @@
 import shutil
 import tempfile
 from copy import deepcopy
-import warnings
 
 from distutils import log
 from distutils.log import DEBUG, INFO, WARN, ERROR, FATAL
 from distutils.core import Distribution
 
-def capture_warnings(func):
-    def _capture_warnings(*args, **kw):
-        with warnings.catch_warnings():
-            warnings.simplefilter("ignore")
-            return func(*args, **kw)
-    return _capture_warnings
-
 class LoggingSilencer(object):
 
     def setUp(self):
@@ -63,8 +55,6 @@
         super().tearDown()
         while self.tempdirs:
             d = self.tempdirs.pop()
-            if not os.path.exists(d):
-                continue
             shutil.rmtree(d, os.name in ('nt', 'cygwin'))
 
     def mkdtemp(self):

Modified: python/branches/import_unicode/Lib/distutils/tests/test_archive_util.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/test_archive_util.py	(original)
+++ python/branches/import_unicode/Lib/distutils/tests/test_archive_util.py	Fri Jul 30 01:19:37 2010
@@ -15,30 +15,15 @@
 from test.support import check_warnings
 
 try:
-    import grp
-    import pwd
-    UID_GID_SUPPORT = True
-except ImportError:
-    UID_GID_SUPPORT = False
-
-try:
     import zipfile
     ZIP_SUPPORT = True
 except ImportError:
     ZIP_SUPPORT = find_executable('zip')
 
-# some tests will fail if zlib is not available
-try:
-    import zlib
-except ImportError:
-    zlib = None
-
-
 class ArchiveUtilTestCase(support.TempdirManager,
                           support.LoggingSilencer,
                           unittest.TestCase):
 
-    @unittest.skipUnless(zlib, "requires zlib")
     def test_make_tarball(self):
         # creating something to tar
         tmpdir = self.mkdtemp()
@@ -49,7 +34,7 @@
 
         tmpdir2 = self.mkdtemp()
         unittest.skipUnless(splitdrive(tmpdir)[0] == splitdrive(tmpdir2)[0],
-                            "source and target should be on same drive")
+                            "Source and target should be on same drive")
 
         base_name = os.path.join(tmpdir2, 'archive')
 
@@ -99,7 +84,6 @@
         base_name = os.path.join(tmpdir2, 'archive')
         return tmpdir, tmpdir2, base_name
 
-    @unittest.skipUnless(zlib, "Requires zlib")
     @unittest.skipUnless(find_executable('tar') and find_executable('gzip'),
                          'Need the tar command to run')
     def test_tarfile_vs_tar(self):
@@ -185,7 +169,6 @@
         self.assertTrue(not os.path.exists(tarball))
         self.assertEquals(len(w.warnings), 1)
 
-    @unittest.skipUnless(zlib, "Requires zlib")
     @unittest.skipUnless(ZIP_SUPPORT, 'Need zip support to run')
     def test_make_zipfile(self):
         # creating something to tar
@@ -210,59 +193,6 @@
         base_name = os.path.join(tmpdir, 'archive')
         self.assertRaises(ValueError, make_archive, base_name, 'xxx')
 
-    @unittest.skipUnless(zlib, "Requires zlib")
-    def test_make_archive_owner_group(self):
-        # testing make_archive with owner and group, with various combinations
-        # this works even if there's not gid/uid support
-        if UID_GID_SUPPORT:
-            group = grp.getgrgid(0)[0]
-            owner = pwd.getpwuid(0)[0]
-        else:
-            group = owner = 'root'
-
-        base_dir, root_dir, base_name =  self._create_files()
-        base_name = os.path.join(self.mkdtemp() , 'archive')
-        res = make_archive(base_name, 'zip', root_dir, base_dir, owner=owner,
-                           group=group)
-        self.assertTrue(os.path.exists(res))
-
-        res = make_archive(base_name, 'zip', root_dir, base_dir)
-        self.assertTrue(os.path.exists(res))
-
-        res = make_archive(base_name, 'tar', root_dir, base_dir,
-                           owner=owner, group=group)
-        self.assertTrue(os.path.exists(res))
-
-        res = make_archive(base_name, 'tar', root_dir, base_dir,
-                           owner='kjhkjhkjg', group='oihohoh')
-        self.assertTrue(os.path.exists(res))
-
-    @unittest.skipUnless(zlib, "Requires zlib")
-    @unittest.skipUnless(UID_GID_SUPPORT, "Requires grp and pwd support")
-    def test_tarfile_root_owner(self):
-        tmpdir, tmpdir2, base_name =  self._create_files()
-        old_dir = os.getcwd()
-        os.chdir(tmpdir)
-        group = grp.getgrgid(0)[0]
-        owner = pwd.getpwuid(0)[0]
-        try:
-            archive_name = make_tarball(base_name, 'dist', compress=None,
-                                        owner=owner, group=group)
-        finally:
-            os.chdir(old_dir)
-
-        # check if the compressed tarball was created
-        self.assertTrue(os.path.exists(archive_name))
-
-        # now checks the rights
-        archive = tarfile.open(archive_name)
-        try:
-            for member in archive.getmembers():
-                self.assertEquals(member.uid, 0)
-                self.assertEquals(member.gid, 0)
-        finally:
-            archive.close()
-
     def test_make_archive_cwd(self):
         current_dir = os.getcwd()
         def _breaks(*args, **kw):

Modified: python/branches/import_unicode/Lib/distutils/tests/test_bdist.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/test_bdist.py	(original)
+++ python/branches/import_unicode/Lib/distutils/tests/test_bdist.py	Fri Jul 30 01:19:37 2010
@@ -5,8 +5,6 @@
 import tempfile
 import shutil
 
-from test.support import run_unittest
-
 from distutils.core import Distribution
 from distutils.command.bdist import bdist
 from distutils.tests import support
@@ -42,4 +40,4 @@
     return unittest.makeSuite(BuildTestCase)
 
 if __name__ == '__main__':
-    run_unittest(test_suite())
+    test_support.run_unittest(test_suite())

Modified: python/branches/import_unicode/Lib/distutils/tests/test_bdist_dumb.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/test_bdist_dumb.py	(original)
+++ python/branches/import_unicode/Lib/distutils/tests/test_bdist_dumb.py	Fri Jul 30 01:19:37 2010
@@ -4,15 +4,6 @@
 import sys
 import os
 
-# zlib is not used here, but if it's not available
-# test_simple_built will fail
-try:
-    import zlib
-except ImportError:
-    zlib = None
-
-from test.support import run_unittest
-
 from distutils.core import Distribution
 from distutils.command.bdist_dumb import bdist_dumb
 from distutils.tests import support
@@ -42,7 +33,6 @@
         sys.argv[:] = self.old_sys_argv[1]
         super(BuildDumbTestCase, self).tearDown()
 
-    @unittest.skipUnless(zlib, "requires zlib")
     def test_simple_built(self):
 
         # let's create a simple package
@@ -83,23 +73,8 @@
         # now let's check what we have in the zip file
         # XXX to be done
 
-    def test_finalize_options(self):
-        pkg_dir, dist = self.create_dist()
-        os.chdir(pkg_dir)
-        cmd = bdist_dumb(dist)
-        self.assertEquals(cmd.bdist_dir, None)
-        cmd.finalize_options()
-
-        # bdist_dir is initialized to bdist_base/dumb if not set
-        base = cmd.get_finalized_command('bdist').bdist_base
-        self.assertEquals(cmd.bdist_dir, os.path.join(base, 'dumb'))
-
-        # the format is set to a default value depending on the os.name
-        default = cmd.default_format[os.name]
-        self.assertEquals(cmd.format, default)
-
 def test_suite():
     return unittest.makeSuite(BuildDumbTestCase)
 
 if __name__ == '__main__':
-    run_unittest(test_suite())
+    test_support.run_unittest(test_suite())

Modified: python/branches/import_unicode/Lib/distutils/tests/test_bdist_rpm.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/test_bdist_rpm.py	(original)
+++ python/branches/import_unicode/Lib/distutils/tests/test_bdist_rpm.py	Fri Jul 30 01:19:37 2010
@@ -6,8 +6,6 @@
 import tempfile
 import shutil
 
-from test.support import run_unittest
-
 from distutils.core import Distribution
 from distutils.command.bdist_rpm import bdist_rpm
 from distutils.tests import support
@@ -124,4 +122,4 @@
     return unittest.makeSuite(BuildRpmTestCase)
 
 if __name__ == '__main__':
-    run_unittest(test_suite())
+    test_support.run_unittest(test_suite())

Modified: python/branches/import_unicode/Lib/distutils/tests/test_bdist_wininst.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/test_bdist_wininst.py	(original)
+++ python/branches/import_unicode/Lib/distutils/tests/test_bdist_wininst.py	Fri Jul 30 01:19:37 2010
@@ -1,8 +1,6 @@
 """Tests for distutils.command.bdist_wininst."""
 import unittest
 
-from test.support import run_unittest
-
 from distutils.command.bdist_wininst import bdist_wininst
 from distutils.tests import support
 
@@ -29,4 +27,4 @@
     return unittest.makeSuite(BuildWinInstTestCase)
 
 if __name__ == '__main__':
-    run_unittest(test_suite())
+    test_support.run_unittest(test_suite())

Modified: python/branches/import_unicode/Lib/distutils/tests/test_build_clib.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/test_build_clib.py	(original)
+++ python/branches/import_unicode/Lib/distutils/tests/test_build_clib.py	Fri Jul 30 01:19:37 2010
@@ -120,7 +120,8 @@
         # before we run the command, we want to make sure
         # all commands are present on the system
         # by creating a compiler and checking its executables
-        from distutils.ccompiler import new_compiler, customize_compiler
+        from distutils.ccompiler import new_compiler
+        from distutils.sysconfig import customize_compiler
 
         compiler = new_compiler()
         customize_compiler(compiler)

Modified: python/branches/import_unicode/Lib/distutils/tests/test_build_ext.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/test_build_ext.py	(original)
+++ python/branches/import_unicode/Lib/distutils/tests/test_build_ext.py	Fri Jul 30 01:19:37 2010
@@ -3,13 +3,10 @@
 import tempfile
 import shutil
 from io import StringIO
-import warnings
-from test.support import check_warnings
-from test.support import captured_stdout
 
 from distutils.core import Extension, Distribution
 from distutils.command.build_ext import build_ext
-import sysconfig
+from distutils import sysconfig
 from distutils.tests.support import TempdirManager
 from distutils.tests.support import LoggingSilencer
 from distutils.extension import Extension
@@ -25,23 +22,19 @@
 
 def _get_source_filename():
     srcdir = sysconfig.get_config_var('srcdir')
-    if srcdir is None:
-        return os.path.join(sysconfig.project_base, 'Modules', 'xxmodule.c')
     return os.path.join(srcdir, 'Modules', 'xxmodule.c')
 
-_XX_MODULE_PATH = _get_source_filename()
-
-class BuildExtTestCase(TempdirManager, LoggingSilencer, unittest.TestCase):
-
+class BuildExtTestCase(TempdirManager,
+                       LoggingSilencer,
+                       unittest.TestCase):
     def setUp(self):
         # Create a simple test environment
         # Note that we're making changes to sys.path
         super(BuildExtTestCase, self).setUp()
         self.tmp_dir = self.mkdtemp()
-        if os.path.exists(_XX_MODULE_PATH):
-            self.sys_path = sys.path[:]
-            sys.path.append(self.tmp_dir)
-            shutil.copy(_XX_MODULE_PATH, self.tmp_dir)
+        self.sys_path = sys.path, sys.path[:]
+        sys.path.append(self.tmp_dir)
+        shutil.copy(_get_source_filename(), self.tmp_dir)
         if sys.version > "2.6":
             import site
             self.old_user_base = site.USER_BASE
@@ -49,19 +42,6 @@
             from distutils.command import build_ext
             build_ext.USER_BASE = site.USER_BASE
 
-    def tearDown(self):
-        # Get everything back to normal
-        if os.path.exists(_XX_MODULE_PATH):
-            support.unload('xx')
-            sys.path[:] = self.sys_path
-            # XXX on Windows the test leaves a directory
-            # with xx module in TEMP
-        shutil.rmtree(self.tmp_dir, os.name == 'nt' or
-                                    sys.platform == 'cygwin')
-        super(BuildExtTestCase, self).tearDown()
-
-    @unittest.skipIf(not os.path.exists(_XX_MODULE_PATH),
-                     'xxmodule.c not found')
     def test_build_ext(self):
         global ALREADY_TESTED
         xx_c = os.path.join(self.tmp_dir, 'xxmodule.c')
@@ -104,23 +84,35 @@
         self.assertTrue(isinstance(xx.Null(), xx.Null))
         self.assertTrue(isinstance(xx.Str(), xx.Str))
 
+    def tearDown(self):
+        # Get everything back to normal
+        support.unload('xx')
+        sys.path = self.sys_path[0]
+        sys.path[:] = self.sys_path[1]
+        if sys.version > "2.6":
+            import site
+            site.USER_BASE = self.old_user_base
+            from distutils.command import build_ext
+            build_ext.USER_BASE = self.old_user_base
+        super(BuildExtTestCase, self).tearDown()
+
     def test_solaris_enable_shared(self):
         dist = Distribution({'name': 'xx'})
         cmd = build_ext(dist)
         old = sys.platform
 
         sys.platform = 'sunos' # fooling finalize_options
-        from sysconfig import _CONFIG_VARS
-        old_var = _CONFIG_VARS.get('Py_ENABLE_SHARED')
-        _CONFIG_VARS['Py_ENABLE_SHARED'] = 1
+        from distutils.sysconfig import  _config_vars
+        old_var = _config_vars.get('Py_ENABLE_SHARED')
+        _config_vars['Py_ENABLE_SHARED'] = 1
         try:
             cmd.ensure_finalized()
         finally:
             sys.platform = old
             if old_var is None:
-                del _CONFIG_VARS['Py_ENABLE_SHARED']
+                del _config_vars['Py_ENABLE_SHARED']
             else:
-                _CONFIG_VARS['Py_ENABLE_SHARED'] = old_var
+                _config_vars['Py_ENABLE_SHARED'] = old_var
 
         # make sure we get some library dirs under solaris
         self.assertTrue(len(cmd.library_dirs) > 0)
@@ -182,10 +174,11 @@
         cmd = build_ext(dist)
         cmd.finalize_options()
 
-        py_include = sysconfig.get_path('include')
+        from distutils import sysconfig
+        py_include = sysconfig.get_python_inc()
         self.assertTrue(py_include in cmd.include_dirs)
 
-        plat_py_include = sysconfig.get_path('platinclude')
+        plat_py_include = sysconfig.get_python_inc(plat_specific=1)
         self.assertTrue(plat_py_include in cmd.include_dirs)
 
         # make sure cmd.libraries is turned into a list
@@ -336,6 +329,7 @@
         self.assertEquals(so_dir, other_tmp_dir)
 
         cmd.inplace = 0
+        cmd.compiler = None
         cmd.run()
         so_file = cmd.get_outputs()[0]
         self.assertTrue(os.path.exists(so_file))
@@ -404,26 +398,6 @@
         wanted = os.path.join(curdir, 'twisted', 'runner', 'portmap' + ext)
         self.assertEquals(wanted, path)
 
-    def test_compiler_deprecation_warning(self):
-        dist = Distribution()
-        cmd = build_ext(dist)
-
-        class MyCompiler(object):
-            def do_something(self):
-                pass
-
-        with check_warnings() as w:
-            warnings.simplefilter("always")
-            cmd.compiler = MyCompiler()
-            self.assertEquals(len(w.warnings), 1)
-            cmd.compile = 'unix'
-            self.assertEquals(len(w.warnings), 1)
-            cmd.compiler = MyCompiler()
-            cmd.compiler.do_something()
-            # two more warnings genereated by the get
-            # and the set
-            self.assertEquals(len(w.warnings), 3)
-
 def test_suite():
     src = _get_source_filename()
     if not os.path.exists(src):

Modified: python/branches/import_unicode/Lib/distutils/tests/test_build_py.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/test_build_py.py	(original)
+++ python/branches/import_unicode/Lib/distutils/tests/test_build_py.py	Fri Jul 30 01:19:37 2010
@@ -16,7 +16,7 @@
                       support.LoggingSilencer,
                       unittest.TestCase):
 
-    def _setup_package_data(self):
+    def test_package_data(self):
         sources = self.mkdtemp()
         f = open(os.path.join(sources, "__init__.py"), "w")
         f.write("# Pretend this is a package.")
@@ -52,18 +52,9 @@
         self.assertEqual(len(cmd.get_outputs()), 3)
         pkgdest = os.path.join(destination, "pkg")
         files = os.listdir(pkgdest)
-        return files
-
-    def test_package_data(self):
-        files = self._setup_package_data()
         self.assertTrue("__init__.py" in files)
-        self.assertTrue("README.txt" in files)
-
-    @unittest.skipIf(sys.flags.optimize >= 2,
-                     "pyc files are not written with -O2 and above")
-    def test_package_data_pyc(self):
-        files = self._setup_package_data()
         self.assertTrue("__init__.pyc" in files)
+        self.assertTrue("README.txt" in files)
 
     def test_empty_package_dir (self):
         # See SF 1668596/1720897.

Modified: python/branches/import_unicode/Lib/distutils/tests/test_build_scripts.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/test_build_scripts.py	(original)
+++ python/branches/import_unicode/Lib/distutils/tests/test_build_scripts.py	Fri Jul 30 01:19:37 2010
@@ -5,7 +5,7 @@
 
 from distutils.command.build_scripts import build_scripts
 from distutils.core import Distribution
-import sysconfig
+from distutils import sysconfig
 
 from distutils.tests import support
 
@@ -91,12 +91,12 @@
         # --with-suffix=3`, python is compiled okay but the build scripts
         # failed when writing the name of the executable
         old = sysconfig.get_config_vars().get('VERSION')
-        sysconfig._CONFIG_VARS['VERSION'] = 4
+        sysconfig._config_vars['VERSION'] = 4
         try:
             cmd.run()
         finally:
             if old is not None:
-                sysconfig._CONFIG_VARS['VERSION'] = old
+                sysconfig._config_vars['VERSION'] = old
 
         built = os.listdir(target)
         for name in expected:

Deleted: python/branches/import_unicode/Lib/distutils/tests/test_ccompiler.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/test_ccompiler.py	Fri Jul 30 01:19:37 2010
+++ (empty file)
@@ -1,81 +0,0 @@
-"""Tests for distutils.ccompiler."""
-import os
-import unittest
-from test.support import captured_stdout
-
-from distutils.ccompiler import (gen_lib_options, CCompiler,
-                                 get_default_compiler, customize_compiler)
-from distutils import debug
-from distutils.tests import support
-
-class FakeCompiler(object):
-    def library_dir_option(self, dir):
-        return "-L" + dir
-
-    def runtime_library_dir_option(self, dir):
-        return ["-cool", "-R" + dir]
-
-    def find_library_file(self, dirs, lib, debug=0):
-        return 'found'
-
-    def library_option(self, lib):
-        return "-l" + lib
-
-class CCompilerTestCase(support.EnvironGuard, unittest.TestCase):
-
-    def test_gen_lib_options(self):
-        compiler = FakeCompiler()
-        libdirs = ['lib1', 'lib2']
-        runlibdirs = ['runlib1']
-        libs = [os.path.join('dir', 'name'), 'name2']
-
-        opts = gen_lib_options(compiler, libdirs, runlibdirs, libs)
-        wanted = ['-Llib1', '-Llib2', '-cool', '-Rrunlib1', 'found',
-                  '-lname2']
-        self.assertEquals(opts, wanted)
-
-    def test_debug_print(self):
-
-        class MyCCompiler(CCompiler):
-            executables = {}
-
-        compiler = MyCCompiler()
-        with captured_stdout() as stdout:
-            compiler.debug_print('xxx')
-        stdout.seek(0)
-        self.assertEquals(stdout.read(), '')
-
-        debug.DEBUG = True
-        try:
-            with captured_stdout() as stdout:
-                compiler.debug_print('xxx')
-            stdout.seek(0)
-            self.assertEquals(stdout.read(), 'xxx\n')
-        finally:
-            debug.DEBUG = False
-
-    def test_customize_compiler(self):
-
-        # not testing if default compiler is not unix
-        if get_default_compiler() != 'unix':
-            return
-
-        os.environ['AR'] = 'my_ar'
-        os.environ['ARFLAGS'] = '-arflags'
-
-        # make sure AR gets caught
-        class compiler:
-            compiler_type = 'unix'
-
-            def set_executables(self, **kw):
-                self.exes = kw
-
-        comp = compiler()
-        customize_compiler(comp)
-        self.assertEquals(comp.exes['archiver'], 'my_ar -arflags')
-
-def test_suite():
-    return unittest.makeSuite(CCompilerTestCase)
-
-if __name__ == "__main__":
-    unittest.main(defaultTest="test_suite")

Modified: python/branches/import_unicode/Lib/distutils/tests/test_cmd.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/test_cmd.py	(original)
+++ python/branches/import_unicode/Lib/distutils/tests/test_cmd.py	Fri Jul 30 01:19:37 2010
@@ -1,7 +1,7 @@
 """Tests for distutils.cmd."""
 import unittest
 import os
-from test.support import captured_stdout, run_unittest
+from test.support import captured_stdout
 
 from distutils.cmd import Command
 from distutils.dist import Distribution
@@ -124,4 +124,4 @@
     return unittest.makeSuite(CommandTestCase)
 
 if __name__ == '__main__':
-    run_unittest(test_suite())
+    test_support.run_unittest(test_suite())

Modified: python/branches/import_unicode/Lib/distutils/tests/test_cygwinccompiler.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/test_cygwinccompiler.py	(original)
+++ python/branches/import_unicode/Lib/distutils/tests/test_cygwinccompiler.py	Fri Jul 30 01:19:37 2010
@@ -2,21 +2,29 @@
 import unittest
 import sys
 import os
+from io import BytesIO
 import subprocess
-import warnings
-import sysconfig
-
-from test.support import check_warnings, run_unittest
-from test.support import captured_stdout
 
 from distutils import cygwinccompiler
 from distutils.cygwinccompiler import (CygwinCCompiler, check_config_h,
                                        CONFIG_H_OK, CONFIG_H_NOTOK,
                                        CONFIG_H_UNCERTAIN, get_versions,
-                                       get_msvcr, RE_VERSION)
-from distutils.util import get_compiler_versions
+                                       get_msvcr)
 from distutils.tests import support
 
+class FakePopen(object):
+    test_class = None
+
+    def __init__(self, cmd, shell, stdout):
+        self.cmd = cmd.split()[0]
+        exes = self.test_class._exes
+        if self.cmd in exes:
+            # issue #6438 in Python 3.x, Popen returns bytes
+            self.stdout = BytesIO(exes[self.cmd])
+        else:
+            self.stdout = os.popen(cmd, 'r')
+
+
 class CygwinCCompilerTestCase(support.TempdirManager,
                               unittest.TestCase):
 
@@ -24,17 +32,32 @@
         super(CygwinCCompilerTestCase, self).setUp()
         self.version = sys.version
         self.python_h = os.path.join(self.mkdtemp(), 'python.h')
+        from distutils import sysconfig
         self.old_get_config_h_filename = sysconfig.get_config_h_filename
         sysconfig.get_config_h_filename = self._get_config_h_filename
+        self.old_find_executable = cygwinccompiler.find_executable
+        cygwinccompiler.find_executable = self._find_executable
+        self._exes = {}
+        self.old_popen = cygwinccompiler.Popen
+        FakePopen.test_class = self
+        cygwinccompiler.Popen = FakePopen
 
     def tearDown(self):
         sys.version = self.version
+        from distutils import sysconfig
         sysconfig.get_config_h_filename = self.old_get_config_h_filename
+        cygwinccompiler.find_executable = self.old_find_executable
+        cygwinccompiler.Popen = self.old_popen
         super(CygwinCCompilerTestCase, self).tearDown()
 
     def _get_config_h_filename(self):
         return self.python_h
 
+    def _find_executable(self, name):
+        if name in self._exes:
+            return name
+        return None
+
     def test_check_config_h(self):
 
         # check_config_h looks for "GCC" in sys.version first
@@ -58,6 +81,40 @@
         self.write_file(self.python_h, 'xxx __GNUC__ xxx')
         self.assertEquals(check_config_h()[0], CONFIG_H_OK)
 
+    def test_get_versions(self):
+
+        # get_versions calls distutils.spawn.find_executable on
+        # 'gcc', 'ld' and 'dllwrap'
+        self.assertEquals(get_versions(), (None, None, None))
+
+        # Let's fake we have 'gcc' and it returns '3.4.5'
+        self._exes['gcc'] = b'gcc (GCC) 3.4.5 (mingw special)\nFSF'
+        res = get_versions()
+        self.assertEquals(str(res[0]), '3.4.5')
+
+        # and let's see what happens when the version
+        # doesn't match the regular expression
+        # (\d+\.\d+(\.\d+)*)
+        self._exes['gcc'] = b'very strange output'
+        res = get_versions()
+        self.assertEquals(res[0], None)
+
+        # same thing for ld
+        self._exes['ld'] = b'GNU ld version 2.17.50 20060824'
+        res = get_versions()
+        self.assertEquals(str(res[1]), '2.17.50')
+        self._exes['ld'] = b'@(#)PROGRAM:ld  PROJECT:ld64-77'
+        res = get_versions()
+        self.assertEquals(res[1], None)
+
+        # and dllwrap
+        self._exes['dllwrap'] = b'GNU dllwrap 2.17.50 20060824\nFSF'
+        res = get_versions()
+        self.assertEquals(str(res[2]), '2.17.50')
+        self._exes['dllwrap'] = b'Cheese Wrap'
+        res = get_versions()
+        self.assertEquals(res[2], None)
+
     def test_get_msvcr(self):
 
         # none
@@ -90,23 +147,8 @@
                        '[MSC v.1999 32 bits (Intel)]')
         self.assertRaises(ValueError, get_msvcr)
 
-
-    def test_get_version_deprecated(self):
-        with check_warnings() as w:
-            warnings.simplefilter("always")
-            # make sure get_compiler_versions and get_versions
-            # returns the same thing
-            self.assertEquals(get_compiler_versions(), get_versions())
-            # make sure using get_version() generated a warning
-            self.assertEquals(len(w.warnings), 1)
-            # make sure any usage of RE_VERSION will also
-            # generate a warning, but till works
-            version = RE_VERSION.search('1.2').group(1)
-            self.assertEquals(version, '1.2')
-            self.assertEquals(len(w.warnings), 2)
-
 def test_suite():
     return unittest.makeSuite(CygwinCCompilerTestCase)
 
 if __name__ == '__main__':
-    run_unittest(test_suite())
+    test_support.run_unittest(test_suite())

Modified: python/branches/import_unicode/Lib/distutils/tests/test_dist.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/test_dist.py	(original)
+++ python/branches/import_unicode/Lib/distutils/tests/test_dist.py	Fri Jul 30 01:19:37 2010
@@ -7,9 +7,8 @@
 import warnings
 import textwrap
 
-from distutils.dist import Distribution, fix_help_options, DistributionMetadata
+from distutils.dist import Distribution, fix_help_options
 from distutils.cmd import Command
-import distutils.dist
 
 from test.support import TESTFN, captured_stdout
 from distutils.tests import support
@@ -38,8 +37,7 @@
         return self._config_files
 
 
-class DistributionTestCase(support.TempdirManager,
-                           support.LoggingSilencer,
+class DistributionTestCase(support.LoggingSilencer,
                            support.EnvironGuard,
                            unittest.TestCase):
 
@@ -60,27 +58,6 @@
         d.parse_command_line()
         return d
 
-    def test_debug_mode(self):
-        with open(TESTFN, "w") as f:
-            f.write("[global]")
-            f.write("command_packages = foo.bar, splat")
-
-        files = [TESTFN]
-        sys.argv.append("build")
-
-        with captured_stdout() as stdout:
-            self.create_distribution(files)
-        stdout.seek(0)
-        self.assertEquals(stdout.read(), '')
-        distutils.dist.DEBUG = True
-        try:
-            with captured_stdout() as stdout:
-                self.create_distribution(files)
-            stdout.seek(0)
-            self.assertEquals(stdout.read(), '')
-        finally:
-            distutils.dist.DEBUG = False
-
     def test_command_packages_unspecified(self):
         sys.argv.append("build")
         d = self.create_distribution()
@@ -182,35 +159,6 @@
         kwargs = {'level': 'ok2'}
         self.assertRaises(ValueError, dist.announce, args, kwargs)
 
-    def test_find_config_files_disable(self):
-        # Ticket #1180: Allow user to disable their home config file.
-        temp_home = self.mkdtemp()
-        if os.name == 'posix':
-            user_filename = os.path.join(temp_home, ".pydistutils.cfg")
-        else:
-            user_filename = os.path.join(temp_home, "pydistutils.cfg")
-
-        with open(user_filename, 'w') as f:
-            f.write('[distutils]\n')
-
-        def _expander(path):
-            return temp_home
-
-        old_expander = os.path.expanduser
-        os.path.expanduser = _expander
-        try:
-            d = distutils.dist.Distribution()
-            all_files = d.find_config_files()
-
-            d = distutils.dist.Distribution(attrs={'script_args':
-                                            ['--no-user-cfg']})
-            files = d.find_config_files()
-        finally:
-            os.path.expanduser = old_expander
-
-        # make sure --no-user-cfg disables the user cfg file
-        self.assertEquals(len(all_files)-1, len(files))
-
 class MetadataTestCase(support.TempdirManager, support.EnvironGuard,
                        unittest.TestCase):
 
@@ -364,38 +312,11 @@
                  "version": "1.0",
                  "long_description": long_desc}
 
-        dist = distutils.dist.Distribution(attrs)
+        dist = Distribution(attrs)
         meta = self.format_metadata(dist)
         meta = meta.replace('\n' + 8 * ' ', '\n')
         self.assertTrue(long_desc in meta)
 
-    def test_read_metadata(self):
-        attrs = {"name": "package",
-                 "version": "1.0",
-                 "long_description": "desc",
-                 "description": "xxx",
-                 "download_url": "http://example.com",
-                 "keywords": ['one', 'two'],
-                 "requires": ['foo']}
-
-        dist = Distribution(attrs)
-        metadata = dist.metadata
-
-        # write it then reloads it
-        PKG_INFO = io.StringIO()
-        metadata.write_pkg_file(PKG_INFO)
-        PKG_INFO.seek(0)
-        metadata.read_pkg_file(PKG_INFO)
-
-        self.assertEquals(metadata.name, "package")
-        self.assertEquals(metadata.version, "1.0")
-        self.assertEquals(metadata.description, "xxx")
-        self.assertEquals(metadata.download_url, 'http://example.com')
-        self.assertEquals(metadata.keywords, ['one', 'two'])
-        self.assertEquals(metadata.platforms, ['UNKNOWN'])
-        self.assertEquals(metadata.obsoletes, None)
-        self.assertEquals(metadata.requires, ['foo'])
-
 def test_suite():
     suite = unittest.TestSuite()
     suite.addTest(unittest.makeSuite(DistributionTestCase))

Deleted: python/branches/import_unicode/Lib/distutils/tests/test_emxccompiler.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/test_emxccompiler.py	Fri Jul 30 01:19:37 2010
+++ (empty file)
@@ -1,33 +0,0 @@
-"""Tests for distutils.emxccompiler."""
-import unittest
-import sys
-import os
-import warnings
-
-from test.support import check_warnings, run_unittest
-from test.support import captured_stdout
-
-from distutils.emxccompiler import get_versions
-from distutils.util import get_compiler_versions
-from distutils.tests import support
-
-class EmxCCompilerTestCase(support.TempdirManager,
-                           unittest.TestCase):
-
-    def test_get_version_deprecated(self):
-        with check_warnings() as w:
-            warnings.simplefilter("always")
-            # make sure get_compiler_versions and get_versions
-            # returns the same gcc
-            gcc, ld, dllwrap = get_compiler_versions()
-            emx_gcc, emx_ld = get_versions()
-            self.assertEquals(gcc, emx_gcc)
-
-            # make sure using get_version() generated a warning
-            self.assertEquals(len(w.warnings), 1)
-
-def test_suite():
-    return unittest.makeSuite(EmxCCompilerTestCase)
-
-if __name__ == '__main__':
-    run_unittest(test_suite())

Modified: python/branches/import_unicode/Lib/distutils/tests/test_extension.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/test_extension.py	(original)
+++ python/branches/import_unicode/Lib/distutils/tests/test_extension.py	Fri Jul 30 01:19:37 2010
@@ -1,16 +1,13 @@
 """Tests for distutils.extension."""
-import os
-import sys
 import unittest
+import os
 import warnings
 
 from test.support import check_warnings
 from distutils.extension import read_setup_file, Extension
-from distutils.tests.support import capture_warnings
 
 class ExtensionTestCase(unittest.TestCase):
 
-    @capture_warnings
     def test_read_setup_file(self):
         # trying to read a Setup file
         # (sample extracted from the PyGame project)
@@ -33,22 +30,16 @@
 
         self.assertEquals(names, wanted)
 
-    @unittest.skipIf(sys.flags.optimize >= 2,
-                     "Assertions are omitted with -O2 and above")
-    def test_extension_init_assertions(self):
-        # The first argument, which is the name, must be a string.
+    def test_extension_init(self):
+        # the first argument, which is the name, must be a string
         self.assertRaises(AssertionError, Extension, 1, [])
+        ext = Extension('name', [])
+        self.assertEquals(ext.name, 'name')
 
         # the second argument, which is the list of files, must
         # be a list of strings
         self.assertRaises(AssertionError, Extension, 'name', 'file')
         self.assertRaises(AssertionError, Extension, 'name', ['file', 1])
-
-    def test_extension_init(self):
-        ext = Extension('name', [])
-        self.assertEquals(ext.name, 'name')
-
-
         ext = Extension('name', ['file1', 'file2'])
         self.assertEquals(ext.sources, ['file1', 'file2'])
 

Modified: python/branches/import_unicode/Lib/distutils/tests/test_file_util.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/test_file_util.py	(original)
+++ python/branches/import_unicode/Lib/distutils/tests/test_file_util.py	Fri Jul 30 01:19:37 2010
@@ -3,7 +3,7 @@
 import os
 import shutil
 
-from distutils.file_util import move_file, write_file, copy_file
+from distutils.file_util import move_file
 from distutils import log
 from distutils.tests import support
 
@@ -55,21 +55,6 @@
         wanted = ['moving %s -> %s' % (self.source, self.target_dir)]
         self.assertEquals(self._logs, wanted)
 
-    def test_write_file(self):
-        lines = ['a', 'b', 'c']
-        dir = self.mkdtemp()
-        foo = os.path.join(dir, 'foo')
-        write_file(foo, lines)
-        content = [line.strip() for line in open(foo).readlines()]
-        self.assertEquals(content, lines)
-
-    def test_copy_file(self):
-        src_dir = self.mkdtemp()
-        foo = os.path.join(src_dir, 'foo')
-        write_file(foo, 'content')
-        dst_dir = self.mkdtemp()
-        copy_file(foo, dst_dir)
-        self.assertTrue(os.path.exists(os.path.join(dst_dir, 'foo')))
 
 def test_suite():
     return unittest.makeSuite(FileUtilTestCase)

Modified: python/branches/import_unicode/Lib/distutils/tests/test_filelist.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/test_filelist.py	(original)
+++ python/branches/import_unicode/Lib/distutils/tests/test_filelist.py	Fri Jul 30 01:19:37 2010
@@ -1,25 +1,10 @@
 """Tests for distutils.filelist."""
-from os.path import join
 import unittest
-from test.support import captured_stdout
 
 from distutils.filelist import glob_to_re, FileList
+from test.support import captured_stdout
 from distutils import debug
 
-MANIFEST_IN = """\
-include ok
-include xo
-exclude xo
-include foo.tmp
-global-include *.x
-global-include *.txt
-global-exclude *.tmp
-recursive-include f *.oo
-recursive-exclude global *.x
-graft dir
-prune dir3
-"""
-
 class FileListTestCase(unittest.TestCase):
 
     def test_glob_to_re(self):
@@ -34,34 +19,6 @@
         self.assertEquals(glob_to_re('foo????'), r'foo[^/][^/][^/][^/]\Z(?ms)')
         self.assertEquals(glob_to_re(r'foo\\??'), r'foo\\\\[^/][^/]\Z(?ms)')
 
-    def test_process_template_line(self):
-        # testing  all MANIFEST.in template patterns
-        file_list = FileList()
-
-        # simulated file list
-        file_list.allfiles = ['foo.tmp', 'ok', 'xo', 'four.txt',
-                              join('global', 'one.txt'),
-                              join('global', 'two.txt'),
-                              join('global', 'files.x'),
-                              join('global', 'here.tmp'),
-                              join('f', 'o', 'f.oo'),
-                              join('dir', 'graft-one'),
-                              join('dir', 'dir2', 'graft2'),
-                              join('dir3', 'ok'),
-                              join('dir3', 'sub', 'ok.txt')
-                              ]
-
-        for line in MANIFEST_IN.split('\n'):
-            if line.strip() == '':
-                continue
-            file_list.process_template_line(line)
-
-        wanted = ['ok', 'four.txt', join('global', 'one.txt'),
-                  join('global', 'two.txt'), join('f', 'o', 'f.oo'),
-                  join('dir', 'graft-one'), join('dir', 'dir2', 'graft2')]
-
-        self.assertEquals(file_list.files, wanted)
-
     def test_debug_print(self):
         file_list = FileList()
         with captured_stdout() as stdout:

Modified: python/branches/import_unicode/Lib/distutils/tests/test_install.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/test_install.py	(original)
+++ python/branches/import_unicode/Lib/distutils/tests/test_install.py	Fri Jul 30 01:19:37 2010
@@ -5,14 +5,12 @@
 import sys
 import unittest
 import site
-import sysconfig
-from sysconfig import (get_scheme_names, _CONFIG_VARS, _INSTALL_SCHEMES,
-                       get_config_var, get_path)
 
 from test.support import captured_stdout
 
 from distutils.command.install import install
 from distutils.command import install as install_module
+from distutils.command.install import INSTALL_SCHEMES
 from distutils.core import Distribution
 from distutils.errors import DistutilsOptionError
 
@@ -38,23 +36,9 @@
             build_lib=os.path.join(builddir, "lib"),
             )
 
-
-
-        posix_prefix = _INSTALL_SCHEMES['posix_prefix']
-        old_posix_prefix = posix_prefix['platinclude']
-        posix_prefix['platinclude'] = \
-                '{platbase}/include/python{py_version_short}'
-
-        posix_home = _INSTALL_SCHEMES['posix_home']
-        old_posix_home = posix_home['platinclude']
-        posix_home['platinclude'] = '{base}/include/python'
-        try:
-            cmd = install(dist)
-            cmd.home = destination
-            cmd.ensure_finalized()
-        finally:
-            posix_home['platinclude'] = old_posix_home
-            posix_prefix['platinclude'] = old_posix_prefix
+        cmd = install(dist)
+        cmd.home = destination
+        cmd.ensure_finalized()
 
         self.assertEqual(cmd.install_base, destination)
         self.assertEqual(cmd.install_platbase, destination)
@@ -79,19 +63,18 @@
             return
 
         # preparing the environement for the test
-        self.old_user_base = get_config_var('userbase')
-        self.old_user_site = get_path('purelib', '%s_user' % os.name)
+        self.old_user_base = site.USER_BASE
+        self.old_user_site = site.USER_SITE
         self.tmpdir = self.mkdtemp()
         self.user_base = os.path.join(self.tmpdir, 'B')
         self.user_site = os.path.join(self.tmpdir, 'S')
-        _CONFIG_VARS['userbase'] = self.user_base
-        scheme = _INSTALL_SCHEMES['%s_user' % os.name]
-        scheme['purelib'] = self.user_site
+        site.USER_BASE = self.user_base
+        site.USER_SITE = self.user_site
+        install_module.USER_BASE = self.user_base
+        install_module.USER_SITE = self.user_site
 
         def _expanduser(path):
-            if path[0] == '~':
-                path = os.path.normpath(self.tmpdir) + path[1:]
-            return path
+            return self.tmpdir
         self.old_expand = os.path.expanduser
         os.path.expanduser = _expanduser
 
@@ -99,17 +82,19 @@
             # this is the actual test
             self._test_user_site()
         finally:
-            _CONFIG_VARS['userbase'] = self.old_user_base
-            scheme['purelib'] = self.old_user_site
+            site.USER_BASE = self.old_user_base
+            site.USER_SITE = self.old_user_site
+            install_module.USER_BASE = self.old_user_base
+            install_module.USER_SITE = self.old_user_site
             os.path.expanduser = self.old_expand
 
     def _test_user_site(self):
-        schemes = get_scheme_names()
-        for key in ('nt_user', 'posix_user', 'os2_home'):
-            self.assertTrue(key in schemes)
+        for key in ('nt_user', 'unix_user', 'os2_home'):
+            self.assertTrue(key in INSTALL_SCHEMES)
 
         dist = Distribution({'name': 'xx'})
         cmd = install(dist)
+
         # making sure the user option is there
         options = [name for name, short, lable in
                    cmd.user_options]
@@ -200,7 +185,7 @@
         with open(cmd.record) as f:
             self.assertEquals(len(f.readlines()), 1)
 
-    def _test_debug_mode(self):
+    def test_debug_mode(self):
         # this covers the code called when DEBUG is set
         old_logs_len = len(self.logs)
         install_module.DEBUG = True

Modified: python/branches/import_unicode/Lib/distutils/tests/test_install_lib.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/test_install_lib.py	(original)
+++ python/branches/import_unicode/Lib/distutils/tests/test_install_lib.py	Fri Jul 30 01:19:37 2010
@@ -1,6 +1,6 @@
 """Tests for distutils.command.install_data."""
-import os
 import sys
+import os
 import unittest
 
 from distutils.command.install_lib import install_lib
@@ -31,7 +31,9 @@
         cmd.finalize_options()
         self.assertEquals(cmd.optimize, 2)
 
-    def _setup_byte_compile(self):
+    @unittest.skipUnless(not sys.dont_write_bytecode,
+                         'byte-compile not supported')
+    def test_byte_compile(self):
         pkg_dir, dist = self.create_dist()
         cmd = install_lib(dist)
         cmd.compile = cmd.optimize = 1
@@ -39,15 +41,8 @@
         f = os.path.join(pkg_dir, 'foo.py')
         self.write_file(f, '# python file')
         cmd.byte_compile([f])
-        return pkg_dir
-
-    @unittest.skipIf(sys.dont_write_bytecode, 'byte-compile not enabled')
-    def test_byte_compile(self):
-        pkg_dir = self._setup_byte_compile()
-        if sys.flags.optimize < 1:
-            self.assertTrue(os.path.exists(os.path.join(pkg_dir, 'foo.pyc')))
-        else:
-            self.assertTrue(os.path.exists(os.path.join(pkg_dir, 'foo.pyo')))
+        self.assertTrue(os.path.exists(os.path.join(pkg_dir, 'foo.pyc')))
+        self.assertTrue(os.path.exists(os.path.join(pkg_dir, 'foo.pyo')))
 
     def test_get_outputs(self):
         pkg_dir, dist = self.create_dist()

Modified: python/branches/import_unicode/Lib/distutils/tests/test_register.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/test_register.py	(original)
+++ python/branches/import_unicode/Lib/distutils/tests/test_register.py	Fri Jul 30 01:19:37 2010
@@ -202,10 +202,10 @@
         self.assertRaises(DistutilsSetupError, cmd.run)
 
         # we don't test the reSt feature if docutils
-        # is not installed or we our on py3k
+        # is not installed
         try:
             import docutils
-        except Exception:
+        except ImportError:
             return
 
         # metadata are OK but long_description is broken

Modified: python/branches/import_unicode/Lib/distutils/tests/test_sdist.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/test_sdist.py	(original)
+++ python/branches/import_unicode/Lib/distutils/tests/test_sdist.py	Fri Jul 30 01:19:37 2010
@@ -3,22 +3,6 @@
 import unittest
 import shutil
 import zipfile
-import tarfile
-
-# zlib is not used here, but if it's not available
-# the tests that use zipfile may fail
-try:
-    import zlib
-except ImportError:
-    zlib = None
-
-try:
-    import grp
-    import pwd
-    UID_GID_SUPPORT = True
-except ImportError:
-    UID_GID_SUPPORT = False
-
 from os.path import join
 import sys
 import tempfile
@@ -95,7 +79,6 @@
         cmd.warn = _warn
         return dist, cmd
 
-    @unittest.skipUnless(zlib, "requires zlib")
     def test_prune_file_list(self):
         # this test creates a package with some vcs dirs in it
         # and launch sdist to make sure they get pruned
@@ -137,7 +120,6 @@
         # making sure everything has been pruned correctly
         self.assertEquals(len(content), 4)
 
-    @unittest.skipUnless(zlib, "requires zlib")
     def test_make_distribution(self):
 
         # check if tar and gzip are installed
@@ -174,7 +156,6 @@
         self.assertEquals(result,
                 ['fake-1.0.tar', 'fake-1.0.tar.gz'])
 
-    @unittest.skipUnless(zlib, "requires zlib")
     def test_add_defaults(self):
 
         # http://bugs.python.org/issue2279
@@ -236,7 +217,6 @@
         manifest = open(join(self.tmp_dir, 'MANIFEST')).read()
         self.assertEquals(manifest, MANIFEST % {'sep': os.sep})
 
-    @unittest.skipUnless(zlib, "requires zlib")
     def test_metadata_check_option(self):
         # testing the `medata-check` option
         dist, cmd = self.get_cmd(metadata={})
@@ -296,55 +276,6 @@
         cmd.formats = 'supazipa'
         self.assertRaises(DistutilsOptionError, cmd.finalize_options)
 
-    @unittest.skipUnless(zlib, "requires zlib")
-    @unittest.skipUnless(UID_GID_SUPPORT, "Requires grp and pwd support")
-    def test_make_distribution_owner_group(self):
-
-        # check if tar and gzip are installed
-        if (find_executable('tar') is None or
-            find_executable('gzip') is None):
-            return
-
-        # now building a sdist
-        dist, cmd = self.get_cmd()
-
-        # creating a gztar and specifying the owner+group
-        cmd.formats = ['gztar']
-        cmd.owner = pwd.getpwuid(0)[0]
-        cmd.group = grp.getgrgid(0)[0]
-        cmd.ensure_finalized()
-        cmd.run()
-
-        # making sure we have the good rights
-        archive_name = join(self.tmp_dir, 'dist', 'fake-1.0.tar.gz')
-        archive = tarfile.open(archive_name)
-        try:
-            for member in archive.getmembers():
-                self.assertEquals(member.uid, 0)
-                self.assertEquals(member.gid, 0)
-        finally:
-            archive.close()
-
-        # building a sdist again
-        dist, cmd = self.get_cmd()
-
-        # creating a gztar
-        cmd.formats = ['gztar']
-        cmd.ensure_finalized()
-        cmd.run()
-
-        # making sure we have the good rights
-        archive_name = join(self.tmp_dir, 'dist', 'fake-1.0.tar.gz')
-        archive = tarfile.open(archive_name)
-
-        # note that we are not testing the group ownership here
-        # because, depending on the platforms and the container
-        # rights (see #7408)
-        try:
-            for member in archive.getmembers():
-                self.assertEquals(member.uid, os.getuid())
-        finally:
-            archive.close()
 
     def test_get_file_list(self):
         # make sure MANIFEST is recalculated

Modified: python/branches/import_unicode/Lib/distutils/tests/test_sysconfig.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/test_sysconfig.py	(original)
+++ python/branches/import_unicode/Lib/distutils/tests/test_sysconfig.py	Fri Jul 30 01:19:37 2010
@@ -2,9 +2,9 @@
 import os
 import test
 import unittest
-import shutil
 
 from distutils import sysconfig
+from distutils.ccompiler import get_default_compiler
 from distutils.tests import support
 from test.support import TESTFN, run_unittest
 
@@ -26,7 +26,10 @@
         elif os.path.isdir(TESTFN):
             shutil.rmtree(TESTFN)
 
-    @support.capture_warnings
+    def test_get_config_h_filename(self):
+        config_h = sysconfig.get_config_h_filename()
+        self.assertTrue(os.path.isfile(config_h), config_h)
+
     def test_get_python_lib(self):
         lib_dir = sysconfig.get_python_lib()
         # XXX doesn't work on Linux when Python was never installed before
@@ -34,11 +37,7 @@
         # test for pythonxx.lib?
         self.assertNotEqual(sysconfig.get_python_lib(),
                             sysconfig.get_python_lib(prefix=TESTFN))
-        _sysconfig = __import__('sysconfig')
-        res = sysconfig.get_python_lib(True, True)
-        self.assertEquals(_sysconfig.get_path('platstdlib'), res)
 
-    @support.capture_warnings
     def test_get_python_inc(self):
         inc_dir = sysconfig.get_python_inc()
         # This is not much of a test.  We make sure Python.h exists
@@ -48,7 +47,31 @@
         python_h = os.path.join(inc_dir, "Python.h")
         self.assertTrue(os.path.isfile(python_h), python_h)
 
-    @support.capture_warnings
+    def test_get_config_vars(self):
+        cvars = sysconfig.get_config_vars()
+        self.assertTrue(isinstance(cvars, dict))
+        self.assertTrue(cvars)
+
+    def test_customize_compiler(self):
+
+        # not testing if default compiler is not unix
+        if get_default_compiler() != 'unix':
+            return
+
+        os.environ['AR'] = 'my_ar'
+        os.environ['ARFLAGS'] = '-arflags'
+
+        # make sure AR gets caught
+        class compiler:
+            compiler_type = 'unix'
+
+            def set_executables(self, **kw):
+                self.exes = kw
+
+        comp = compiler()
+        sysconfig.customize_compiler(comp)
+        self.assertEquals(comp.exes['archiver'], 'my_ar -arflags')
+
     def test_parse_makefile_base(self):
         self.makefile = TESTFN
         fd = open(self.makefile, 'w')
@@ -70,6 +93,15 @@
                               'OTHER': 'foo'})
 
 
+    def test_sysconfig_module(self):
+        import sysconfig as global_sysconfig
+        self.assertEquals(global_sysconfig.get_config_var('CFLAGS'), sysconfig.get_config_var('CFLAGS'))
+        self.assertEquals(global_sysconfig.get_config_var('LDFLAGS'), sysconfig.get_config_var('LDFLAGS'))
+        self.assertEquals(global_sysconfig.get_config_var('LDSHARED'),sysconfig.get_config_var('LDSHARED'))
+        self.assertEquals(global_sysconfig.get_config_var('CC'), sysconfig.get_config_var('CC'))
+
+
+
 def test_suite():
     suite = unittest.TestSuite()
     suite.addTest(unittest.makeSuite(SysconfigTestCase))

Modified: python/branches/import_unicode/Lib/distutils/tests/test_unixccompiler.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/test_unixccompiler.py	(original)
+++ python/branches/import_unicode/Lib/distutils/tests/test_unixccompiler.py	Fri Jul 30 01:19:37 2010
@@ -1,8 +1,8 @@
 """Tests for distutils.unixccompiler."""
 import sys
 import unittest
-import sysconfig
 
+from distutils import sysconfig
 from distutils.unixccompiler import UnixCCompiler
 
 class UnixCCompilerTestCase(unittest.TestCase):
@@ -114,14 +114,6 @@
         sysconfig.get_config_var = gcv
         self.assertEqual(self.cc.rpath_foo(), '-R/foo')
 
-        # AIX C/C++ linker
-        sys.platform = 'aix'
-        def gcv(v):
-            return 'xxx'
-        sysconfig.get_config_var = gcv
-        self.assertEqual(self.cc.rpath_foo(), '-blibpath:/foo')
-
-
 def test_suite():
     return unittest.makeSuite(UnixCCompilerTestCase)
 

Modified: python/branches/import_unicode/Lib/distutils/tests/test_upload.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/test_upload.py	(original)
+++ python/branches/import_unicode/Lib/distutils/tests/test_upload.py	Fri Jul 30 01:19:37 2010
@@ -2,8 +2,8 @@
 import sys
 import os
 import unittest
+import http.client as httpclient
 
-from distutils.command import upload as upload_mod
 from distutils.command.upload import upload
 from distutils.core import Distribution
 
@@ -38,37 +38,48 @@
 [server1]
 username:me
 """
+class Response(object):
+    def __init__(self, status=200, reason='OK'):
+        self.status = status
+        self.reason = reason
 
-class FakeOpen(object):
+class FakeConnection(object):
 
-    def __init__(self, url):
-        self.url = url
-        if not isinstance(url, str):
-            self.req = url
-        else:
-            self.req = None
-        self.msg = 'OK'
+    def __init__(self):
+        self.requests = []
+        self.headers = []
+        self.body = ''
 
-    def getcode(self):
-        return 200
+    def __call__(self, netloc):
+        return self
 
+    def connect(self):
+        pass
+    endheaders = connect
+
+    def putrequest(self, method, url):
+        self.requests.append((method, url))
+
+    def putheader(self, name, value):
+        self.headers.append((name, value))
+
+    def send(self, body):
+        self.body = body
+
+    def getresponse(self):
+        return Response()
 
 class uploadTestCase(PyPIRCCommandTestCase):
 
     def setUp(self):
         super(uploadTestCase, self).setUp()
-        self.old_open = upload_mod.urlopen
-        upload_mod.urlopen = self._urlopen
-        self.last_open = None
+        self.old_class = httpclient.HTTPConnection
+        self.conn = httpclient.HTTPConnection = FakeConnection()
 
     def tearDown(self):
-        upload_mod.urlopen = self.old_open
+        httpclient.HTTPConnection = self.old_class
         super(uploadTestCase, self).tearDown()
 
-    def _urlopen(self, url):
-        self.last_open = FakeOpen(url)
-        return self.last_open
-
     def test_finalize_options(self):
 
         # new format
@@ -113,15 +124,13 @@
         cmd.run()
 
         # what did we send ?
-        headers = dict(self.last_open.req.headers)
+        headers = dict(self.conn.headers)
         self.assertEquals(headers['Content-length'], '2087')
         self.assertTrue(headers['Content-type'].startswith('multipart/form-data'))
-        self.assertEquals(self.last_open.req.get_method(), 'POST')
-        self.assertEquals(self.last_open.req.get_full_url(),
-                          'http://pypi.python.org/pypi')
-        self.assertTrue(b'xxx' in self.last_open.req.data)
-        auth = self.last_open.req.headers['Authorization']
-        self.assertFalse('\n' in auth)
+        self.assertFalse('\n' in headers['Authorization'])
+
+        self.assertEquals(self.conn.requests, [('POST', '/pypi')])
+        self.assert_((b'xxx') in self.conn.body)
 
 def test_suite():
     return unittest.makeSuite(uploadTestCase)

Modified: python/branches/import_unicode/Lib/distutils/tests/test_util.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/tests/test_util.py	(original)
+++ python/branches/import_unicode/Lib/distutils/tests/test_util.py	Fri Jul 30 01:19:37 2010
@@ -3,33 +3,15 @@
 import sys
 import unittest
 from copy import copy
-from io import BytesIO
-import subprocess
 
-from sysconfig import get_config_vars, get_platform
 from distutils.errors import DistutilsPlatformError, DistutilsByteCompileError
-from distutils.util import (convert_path, change_root,
+from distutils.util import (get_platform, convert_path, change_root,
                             check_environ, split_quoted, strtobool,
-                            rfc822_escape, get_compiler_versions,
-                            _find_exe_version, _MAC_OS_X_LD_VERSION,
-                            byte_compile)
-from distutils import util
+                            rfc822_escape, byte_compile)
+from distutils import util # used to patch _environ_checked
+from distutils.sysconfig import get_config_vars
+from distutils import sysconfig
 from distutils.tests import support
-from distutils.version import LooseVersion
-
-class FakePopen(object):
-    test_class = None
-    def __init__(self, cmd, shell, stdout, stderr):
-        self.cmd = cmd.split()[0]
-        exes = self.test_class._exes
-        if self.cmd not in exes:
-            # we don't want to call the system, returning an empty
-            # output so it doesn't match
-            self.stdout = BytesIO()
-            self.stderr = BytesIO()
-        else:
-            self.stdout = BytesIO(exes[self.cmd])
-            self.stderr = BytesIO()
 
 class UtilTestCase(support.EnvironGuard, unittest.TestCase):
 
@@ -43,7 +25,7 @@
         self.join = os.path.join
         self.isabs = os.path.isabs
         self.splitdrive = os.path.splitdrive
-        #self._config_vars = copy(sysconfig._config_vars)
+        self._config_vars = copy(sysconfig._config_vars)
 
         # patching os.uname
         if hasattr(os, 'uname'):
@@ -52,17 +34,8 @@
         else:
             self.uname = None
             self._uname = None
-        os.uname = self._get_uname
 
-        # patching POpen
-        self.old_find_executable = util.find_executable
-        util.find_executable = self._find_executable
-        self._exes = {}
-        self.old_popen = subprocess.Popen
-        self.old_stdout  = sys.stdout
-        self.old_stderr = sys.stderr
-        FakePopen.test_class = self
-        subprocess.Popen = FakePopen
+        os.uname = self._get_uname
 
     def tearDown(self):
         # getting back the environment
@@ -77,11 +50,7 @@
             os.uname = self.uname
         else:
             del os.uname
-        #sysconfig._config_vars = copy(self._config_vars)
-        util.find_executable = self.old_find_executable
-        subprocess.Popen = self.old_popen
-        sys.old_stdout  = self.old_stdout
-        sys.old_stderr = self.old_stderr
+        sysconfig._config_vars = copy(self._config_vars)
         super(UtilTestCase, self).tearDown()
 
     def _set_uname(self, uname):
@@ -91,11 +60,103 @@
         return self._uname
 
     def test_get_platform(self):
-        platform = util.get_platform()
-        self.assertEquals(platform, get_platform())
-        util.set_platform('MyOwnPlatform')
-        self.assertEquals('MyOwnPlatform', util.get_platform())
-        util.set_platform(platform)
+
+        # windows XP, 32bits
+        os.name = 'nt'
+        sys.version = ('2.4.4 (#71, Oct 18 2006, 08:34:43) '
+                       '[MSC v.1310 32 bit (Intel)]')
+        sys.platform = 'win32'
+        self.assertEquals(get_platform(), 'win32')
+
+        # windows XP, amd64
+        os.name = 'nt'
+        sys.version = ('2.4.4 (#71, Oct 18 2006, 08:34:43) '
+                       '[MSC v.1310 32 bit (Amd64)]')
+        sys.platform = 'win32'
+        self.assertEquals(get_platform(), 'win-amd64')
+
+        # windows XP, itanium
+        os.name = 'nt'
+        sys.version = ('2.4.4 (#71, Oct 18 2006, 08:34:43) '
+                       '[MSC v.1310 32 bit (Itanium)]')
+        sys.platform = 'win32'
+        self.assertEquals(get_platform(), 'win-ia64')
+
+        # macbook
+        os.name = 'posix'
+        sys.version = ('2.5 (r25:51918, Sep 19 2006, 08:49:13) '
+                       '\n[GCC 4.0.1 (Apple Computer, Inc. build 5341)]')
+        sys.platform = 'darwin'
+        self._set_uname(('Darwin', 'macziade', '8.11.1',
+                   ('Darwin Kernel Version 8.11.1: '
+                    'Wed Oct 10 18:23:28 PDT 2007; '
+                    'root:xnu-792.25.20~1/RELEASE_I386'), 'i386'))
+        os.environ['MACOSX_DEPLOYMENT_TARGET'] = '10.3'
+
+        get_config_vars()['CFLAGS'] = ('-fno-strict-aliasing -DNDEBUG -g '
+                                       '-fwrapv -O3 -Wall -Wstrict-prototypes')
+
+        cursize = sys.maxsize
+        sys.maxsize = (2 ** 31)-1
+        try:
+            self.assertEquals(get_platform(), 'macosx-10.3-i386')
+        finally:
+            sys.maxsize = cursize
+
+        # macbook with fat binaries (fat, universal or fat64)
+        os.environ['MACOSX_DEPLOYMENT_TARGET'] = '10.4'
+        get_config_vars()['CFLAGS'] = ('-arch ppc -arch i386 -isysroot '
+                                       '/Developer/SDKs/MacOSX10.4u.sdk  '
+                                       '-fno-strict-aliasing -fno-common '
+                                       '-dynamic -DNDEBUG -g -O3')
+
+        self.assertEquals(get_platform(), 'macosx-10.4-fat')
+
+        get_config_vars()['CFLAGS'] = ('-arch x86_64 -arch i386 -isysroot '
+                                       '/Developer/SDKs/MacOSX10.4u.sdk  '
+                                       '-fno-strict-aliasing -fno-common '
+                                       '-dynamic -DNDEBUG -g -O3')
+
+        self.assertEquals(get_platform(), 'macosx-10.4-intel')
+
+        get_config_vars()['CFLAGS'] = ('-arch x86_64 -arch ppc -arch i386 -isysroot '
+                                       '/Developer/SDKs/MacOSX10.4u.sdk  '
+                                       '-fno-strict-aliasing -fno-common '
+                                       '-dynamic -DNDEBUG -g -O3')
+        self.assertEquals(get_platform(), 'macosx-10.4-fat3')
+
+        get_config_vars()['CFLAGS'] = ('-arch ppc64 -arch x86_64 -arch ppc -arch i386 -isysroot '
+                                       '/Developer/SDKs/MacOSX10.4u.sdk  '
+                                       '-fno-strict-aliasing -fno-common '
+                                       '-dynamic -DNDEBUG -g -O3')
+        self.assertEquals(get_platform(), 'macosx-10.4-universal')
+
+        get_config_vars()['CFLAGS'] = ('-arch x86_64 -arch ppc64 -isysroot '
+                                       '/Developer/SDKs/MacOSX10.4u.sdk  '
+                                       '-fno-strict-aliasing -fno-common '
+                                       '-dynamic -DNDEBUG -g -O3')
+
+        self.assertEquals(get_platform(), 'macosx-10.4-fat64')
+
+        for arch in ('ppc', 'i386', 'x86_64', 'ppc64'):
+            get_config_vars()['CFLAGS'] = ('-arch %s -isysroot '
+                                           '/Developer/SDKs/MacOSX10.4u.sdk  '
+                                           '-fno-strict-aliasing -fno-common '
+                                           '-dynamic -DNDEBUG -g -O3'%(arch,))
+
+            self.assertEquals(get_platform(), 'macosx-10.4-%s'%(arch,))
+
+        # linux debian sarge
+        os.name = 'posix'
+        sys.version = ('2.3.5 (#1, Jul  4 2007, 17:28:59) '
+                       '\n[GCC 4.1.2 20061115 (prerelease) (Debian 4.1.1-21)]')
+        sys.platform = 'linux2'
+        self._set_uname(('Linux', 'aglae', '2.6.21.1dedibox-r7',
+                    '#1 Mon Apr 30 17:25:38 CEST 2007', 'i686'))
+
+        self.assertEquals(get_platform(), 'linux-i686')
+
+        # XXX more platforms to tests here
 
     def test_convert_path(self):
         # linux/mac
@@ -199,70 +260,6 @@
                   'header%(8s)s') % {'8s': '\n'+8*' '}
         self.assertEquals(res, wanted)
 
-    def test_find_exe_version(self):
-        # the ld version scheme under MAC OS is:
-        #   ^@(#)PROGRAM:ld  PROJECT:ld64-VERSION
-        #
-        # where VERSION is a 2-digit number for major
-        # revisions. For instance under Leopard, it's
-        # currently 77
-        #
-        # Dots are used when branching is done.
-        #
-        # The SnowLeopard ld64 is currently 95.2.12
-
-        for output, version in ((b'@(#)PROGRAM:ld  PROJECT:ld64-77', '77'),
-                                (b'@(#)PROGRAM:ld  PROJECT:ld64-95.2.12',
-                                 '95.2.12')):
-            result = _MAC_OS_X_LD_VERSION.search(output)
-            self.assertEquals(result.group(1).decode(), version)
-
-    def _find_executable(self, name):
-        if name in self._exes:
-            return name
-        return None
-
-    def test_get_compiler_versions(self):
-        # get_versions calls distutils.spawn.find_executable on
-        # 'gcc', 'ld' and 'dllwrap'
-        self.assertEquals(get_compiler_versions(), (None, None, None))
-
-        # Let's fake we have 'gcc' and it returns '3.4.5'
-        self._exes['gcc'] = b'gcc (GCC) 3.4.5 (mingw special)\nFSF'
-        res = get_compiler_versions()
-        self.assertEquals(str(res[0]), '3.4.5')
-
-        # and let's see what happens when the version
-        # doesn't match the regular expression
-        # (\d+\.\d+(\.\d+)*)
-        self._exes['gcc'] = b'very strange output'
-        res = get_compiler_versions()
-        self.assertEquals(res[0], None)
-
-        # same thing for ld
-        if sys.platform != 'darwin':
-            self._exes['ld'] = b'GNU ld version 2.17.50 20060824'
-            res = get_compiler_versions()
-            self.assertEquals(str(res[1]), '2.17.50')
-            self._exes['ld'] = b'@(#)PROGRAM:ld  PROJECT:ld64-77'
-            res = get_compiler_versions()
-            self.assertEquals(res[1], None)
-        else:
-            self._exes['ld'] = b'GNU ld version 2.17.50 20060824'
-            res = get_compiler_versions()
-            self.assertEquals(res[1], None)
-            self._exes['ld'] = b'@(#)PROGRAM:ld  PROJECT:ld64-77'
-            res = get_compiler_versions()
-            self.assertEquals(str(res[1]), '77')
-
-        # and dllwrap
-        self._exes['dllwrap'] = b'GNU dllwrap 2.17.50 20060824\nFSF'
-        res = get_compiler_versions()
-        self.assertEquals(str(res[2]), '2.17.50')
-        self._exes['dllwrap'] = b'Cheese Wrap'
-        res = get_compiler_versions()
-        self.assertEquals(res[2], None)
-
     def test_dont_write_bytecode(self):
         # makes sure byte_compile raise a DistutilsError
         # if sys.dont_write_bytecode is True

Modified: python/branches/import_unicode/Lib/distutils/text_file.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/text_file.py	(original)
+++ python/branches/import_unicode/Lib/distutils/text_file.py	Fri Jul 30 01:19:37 2010
@@ -6,8 +6,8 @@
 
 __revision__ = "$Id$"
 
-import sys
-import io
+import sys, os, io
+
 
 class TextFile:
     """Provides a file-like object that takes care of all the things you

Modified: python/branches/import_unicode/Lib/distutils/unixccompiler.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/unixccompiler.py	(original)
+++ python/branches/import_unicode/Lib/distutils/unixccompiler.py	Fri Jul 30 01:19:37 2010
@@ -17,6 +17,7 @@
 
 import os, sys, re
 
+from distutils import sysconfig
 from distutils.dep_util import newer
 from distutils.ccompiler import \
      CCompiler, gen_preprocess_options, gen_lib_options
@@ -24,7 +25,6 @@
      DistutilsExecError, CompileError, LibError, LinkError
 from distutils import log
 
-
 # XXX Things not currently handled:
 #   * optimization/debug/warning flags; we just use whatever's in Python's
 #     Makefile and live with it.  Is this adequate?  If not, we might
@@ -74,7 +74,7 @@
 
     if 'ARCHFLAGS' in os.environ and not stripArch:
         # User specified different -arch flags in the environ,
-        # see also the sysconfig
+        # see also distutils.sysconfig
         compiler_so = compiler_so + os.environ['ARCHFLAGS'].split()
 
     if stripSysroot:
@@ -281,9 +281,7 @@
         # this time, there's no way to determine this information from
         # the configuration data stored in the Python installation, so
         # we use this hack.
-        _sysconfig = __import__('sysconfig')
-
-        compiler = os.path.basename(_sysconfig.get_config_var("CC"))
+        compiler = os.path.basename(sysconfig.get_config_var("CC"))
         if sys.platform[:6] == "darwin":
             # MacOSX's linker doesn't understand the -R flag at all
             return "-L" + dir
@@ -293,24 +291,23 @@
             return ["+s", "-L" + dir]
         elif sys.platform[:7] == "irix646" or sys.platform[:6] == "osf1V5":
             return ["-rpath", dir]
-        elif self._is_gcc(compiler):
-            # gcc on non-GNU systems does not need -Wl, but can
-            # use it anyway.  Since distutils has always passed in
-            # -Wl whenever gcc was used in the past it is probably
-            # safest to keep doing so.
-            if _sysconfig.get_config_var("GNULD") == "yes":
-                # GNU ld needs an extra option to get a RUNPATH
-                # instead of just an RPATH.
-                return "-Wl,--enable-new-dtags,-R" + dir
-            else:
-                return "-Wl,-R" + dir
-        elif sys.platform[:3] == "aix":
-            return "-blibpath:" + dir
         else:
-            # No idea how --enable-new-dtags would be passed on to
-            # ld if this system was using GNU ld.  Don't know if a
-            # system like this even exists.
-            return "-R" + dir
+            if self._is_gcc(compiler):
+                # gcc on non-GNU systems does not need -Wl, but can
+                # use it anyway.  Since distutils has always passed in
+                # -Wl whenever gcc was used in the past it is probably
+                # safest to keep doing so.
+                if sysconfig.get_config_var("GNULD") == "yes":
+                    # GNU ld needs an extra option to get a RUNPATH
+                    # instead of just an RPATH.
+                    return "-Wl,--enable-new-dtags,-R" + dir
+                else:
+                    return "-Wl,-R" + dir
+            else:
+                # No idea how --enable-new-dtags would be passed on to
+                # ld if this system was using GNU ld.  Don't know if a
+                # system like this even exists.
+                return "-R" + dir
 
     def library_option(self, lib):
         return "-l" + lib
@@ -324,8 +321,7 @@
             # On OSX users can specify an alternate SDK using
             # '-isysroot', calculate the SDK root if it is specified
             # (and use it further on)
-            _sysconfig = __import__('sysconfig')
-            cflags = _sysconfig.get_config_var('CFLAGS')
+            cflags = sysconfig.get_config_var('CFLAGS')
             m = re.search(r'-isysroot\s+(\S+)', cflags)
             if m is None:
                 sysroot = '/'

Modified: python/branches/import_unicode/Lib/distutils/util.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/util.py	(original)
+++ python/branches/import_unicode/Lib/distutils/util.py	Fri Jul 30 01:19:37 2010
@@ -7,40 +7,182 @@
 __revision__ = "$Id$"
 
 import sys, os, string, re
-
 from distutils.errors import DistutilsPlatformError
 from distutils.dep_util import newer
-from distutils.spawn import spawn, find_executable
+from distutils.spawn import spawn
 from distutils import log
-from distutils.version import LooseVersion
 from distutils.errors import DistutilsByteCompileError
 
-_sysconfig = __import__('sysconfig')
-_PLATFORM = None
+def get_platform ():
+    """Return a string that identifies the current platform.  This is used
+    mainly to distinguish platform-specific build directories and
+    platform-specific built distributions.  Typically includes the OS name
+    and version and the architecture (as supplied by 'os.uname()'),
+    although the exact information included depends on the OS; eg. for IRIX
+    the architecture isn't particularly important (IRIX only runs on SGI
+    hardware), but for Linux the kernel version isn't particularly
+    important.
+
+    Examples of returned values:
+       linux-i586
+       linux-alpha (?)
+       solaris-2.6-sun4u
+       irix-5.3
+       irix64-6.2
+
+    Windows will return one of:
+       win-amd64 (64bit Windows on AMD64 (aka x86_64, Intel64, EM64T, etc)
+       win-ia64 (64bit Windows on Itanium)
+       win32 (all others - specifically, sys.platform is returned)
+
+    For other non-POSIX platforms, currently just returns 'sys.platform'.
+    """
+    if os.name == 'nt':
+        # sniff sys.version for architecture.
+        prefix = " bit ("
+        i = sys.version.find(prefix)
+        if i == -1:
+            return sys.platform
+        j = sys.version.find(")", i)
+        look = sys.version[i+len(prefix):j].lower()
+        if look == 'amd64':
+            return 'win-amd64'
+        if look == 'itanium':
+            return 'win-ia64'
+        return sys.platform
+
+    if os.name != "posix" or not hasattr(os, 'uname'):
+        # XXX what about the architecture? NT is Intel or Alpha,
+        # Mac OS is M68k or PPC, etc.
+        return sys.platform
+
+    # Try to distinguish various flavours of Unix
+
+    (osname, host, release, version, machine) = os.uname()
+
+    # Convert the OS name to lowercase, remove '/' characters
+    # (to accommodate BSD/OS), and translate spaces (for "Power Macintosh")
+    osname = osname.lower().replace('/', '')
+    machine = machine.replace(' ', '_')
+    machine = machine.replace('/', '-')
+
+    if osname[:5] == "linux":
+        # At least on Linux/Intel, 'machine' is the processor --
+        # i386, etc.
+        # XXX what about Alpha, SPARC, etc?
+        return  "%s-%s" % (osname, machine)
+    elif osname[:5] == "sunos":
+        if release[0] >= "5":           # SunOS 5 == Solaris 2
+            osname = "solaris"
+            release = "%d.%s" % (int(release[0]) - 3, release[2:])
+        # fall through to standard osname-release-machine representation
+    elif osname[:4] == "irix":              # could be "irix64"!
+        return "%s-%s" % (osname, release)
+    elif osname[:3] == "aix":
+        return "%s-%s.%s" % (osname, version, release)
+    elif osname[:6] == "cygwin":
+        osname = "cygwin"
+        rel_re = re.compile (r'[\d.]+', re.ASCII)
+        m = rel_re.match(release)
+        if m:
+            release = m.group()
+    elif osname[:6] == "darwin":
+        #
+        # For our purposes, we'll assume that the system version from
+        # distutils' perspective is what MACOSX_DEPLOYMENT_TARGET is set
+        # to. This makes the compatibility story a bit more sane because the
+        # machine is going to compile and link as if it were
+        # MACOSX_DEPLOYMENT_TARGET.
+        from distutils.sysconfig import get_config_vars
+        cfgvars = get_config_vars()
+
+        macver = os.environ.get('MACOSX_DEPLOYMENT_TARGET')
+        if not macver:
+            macver = cfgvars.get('MACOSX_DEPLOYMENT_TARGET')
+
+        if 1:
+            # Always calculate the release of the running machine,
+            # needed to determine if we can build fat binaries or not.
+
+            macrelease = macver
+            # Get the system version. Reading this plist is a documented
+            # way to get the system version (see the documentation for
+            # the Gestalt Manager)
+            try:
+                f = open('/System/Library/CoreServices/SystemVersion.plist')
+            except IOError:
+                # We're on a plain darwin box, fall back to the default
+                # behaviour.
+                pass
+            else:
+                m = re.search(
+                        r'<key>ProductUserVisibleVersion</key>\s*' +
+                        r'<string>(.*?)</string>', f.read())
+                f.close()
+                if m is not None:
+                    macrelease = '.'.join(m.group(1).split('.')[:2])
+                # else: fall back to the default behaviour
+
+        if not macver:
+            macver = macrelease
+
+        if macver:
+            from distutils.sysconfig import get_config_vars
+            release = macver
+            osname = "macosx"
+
+            if (macrelease + '.') >= '10.4.' and \
+                    '-arch' in get_config_vars().get('CFLAGS', '').strip():
+                # The universal build will build fat binaries, but not on
+                # systems before 10.4
+                #
+                # Try to detect 4-way universal builds, those have machine-type
+                # 'universal' instead of 'fat'.
+
+                machine = 'fat'
+                cflags = get_config_vars().get('CFLAGS')
+
+                archs = re.findall('-arch\s+(\S+)', cflags)
+                archs = tuple(sorted(set(archs)))
+
+                if len(archs) == 1:
+                    machine = archs[0]
+                elif archs == ('i386', 'ppc'):
+                    machine = 'fat'
+                elif archs == ('i386', 'x86_64'):
+                    machine = 'intel'
+                elif archs == ('i386', 'ppc', 'x86_64'):
+                    machine = 'fat3'
+                elif archs == ('ppc64', 'x86_64'):
+                    machine = 'fat64'
+                elif archs == ('i386', 'ppc', 'ppc64', 'x86_64'):
+                    machine = 'universal'
+                else:
+                    raise ValueError(
+                       "Don't know machine value for archs=%r"%(archs,))
 
-def get_platform():
-    """Return a string that identifies the current platform.
+            elif machine == 'i386':
+                # On OSX the machine type returned by uname is always the
+                # 32-bit variant, even if the executable architecture is
+                # the 64-bit variant
+                if sys.maxsize >= 2**32:
+                    machine = 'x86_64'
 
-    By default, will return the value returned by sysconfig.get_platform(),
-    but it can be changed by calling set_platform().
-    """
-    global _PLATFORM
-    if _PLATFORM is None:
-        _PLATFORM = _sysconfig.get_platform()
-    return _PLATFORM
+            elif machine in ('PowerPC', 'Power_Macintosh'):
+                # Pick a sane name for the PPC architecture.
+                machine = 'ppc'
 
-def set_platform(identifier):
-    """Sets the platform string identifier returned by get_platform().
+                # See 'i386' case
+                if sys.maxsize >= 2**32:
+                    machine = 'ppc64'
 
-    Note that this change doesn't impact the value returned by
-    sysconfig.get_platform() and is local to Distutils
-    """
-    global _PLATFORM
-    _PLATFORM = identifier
+    return "%s-%s-%s" % (osname, release, machine)
+
+# get_platform ()
 
-def convert_path(pathname):
-    """Return 'pathname' as a name that will work on the native filesystem.
 
+def convert_path (pathname):
+    """Return 'pathname' as a name that will work on the native filesystem,
     i.e. split it on '/' and put it back together again using the current
     directory separator.  Needed because filenames in the setup script are
     always supplied in Unix style, and have to be converted to the local
@@ -64,12 +206,12 @@
         return os.curdir
     return os.path.join(*paths)
 
+# convert_path ()
 
-def change_root(new_root, pathname):
-    """Return 'pathname' with 'new_root' prepended.
 
-    If 'pathname' is relative, this is equivalent to
-    "os.path.join(new_root,pathname)".
+def change_root (new_root, pathname):
+    """Return 'pathname' with 'new_root' prepended.  If 'pathname' is
+    relative, this is equivalent to "os.path.join(new_root,pathname)".
     Otherwise, it requires making 'pathname' relative and then joining the
     two, which is tricky on DOS/Windows and Mac OS.
     """
@@ -91,16 +233,23 @@
             path = path[1:]
         return os.path.join(new_root, path)
 
-    else:
-        raise DistutilsPlatformError("nothing known about "
-                                     "platform '%s'" % os.name)
+    elif os.name == 'mac':
+        if not os.path.isabs(pathname):
+            return os.path.join(new_root, pathname)
+        else:
+            # Chop off volume name from start of path
+            elements = pathname.split(":", 1)
+            pathname = ":" + elements[1]
+            return os.path.join(new_root, pathname)
 
-_environ_checked = 0
+    else:
+        raise DistutilsPlatformError("nothing known about platform '%s'" % os.name)
 
-def check_environ():
-    """Ensure that 'os.environ' has all the environment variables needed.
 
-    We guarantee that users can use in config files, command-line options,
+_environ_checked = 0
+def check_environ ():
+    """Ensure that 'os.environ' has all the environment variables we
+    guarantee that users can use in config files, command-line options,
     etc.  Currently this includes:
       HOME - user's home directory (Unix only)
       PLAT - description of the current platform, including hardware
@@ -115,14 +264,14 @@
         os.environ['HOME'] = pwd.getpwuid(os.getuid())[5]
 
     if 'PLAT' not in os.environ:
-        os.environ['PLAT'] = _sysconfig.get_platform()
+        os.environ['PLAT'] = get_platform()
 
     _environ_checked = 1
 
-def subst_vars(s, local_vars):
-    """Perform shell/Perl-style variable substitution on 'string'.
 
-    Every occurrence of '$' followed by a name is considered a variable, and
+def subst_vars (s, local_vars):
+    """Perform shell/Perl-style variable substitution on 'string'.  Every
+    occurrence of '$' followed by a name is considered a variable, and
     variable is substituted by the value found in the 'local_vars'
     dictionary, or in 'os.environ' if it's not in 'local_vars'.
     'os.environ' is first checked/augmented to guarantee that it contains
@@ -142,11 +291,12 @@
     except KeyError as var:
         raise ValueError("invalid variable '$%s'" % var)
 
-def grok_environment_error(exc, prefix="error: "):
-    """Generate a useful error message from an EnvironmentError.
+# subst_vars ()
+
 
-    This will generate an IOError or an OSError exception object.
-    Handles Python 1.5.1 and 1.5.2 styles, and
+def grok_environment_error (exc, prefix="error: "):
+    """Generate a useful error message from an EnvironmentError (IOError or
+    OSError) exception object.  Handles Python 1.5.1 and 1.5.2 styles, and
     does what it can to deal with exception objects that don't have a
     filename (which happens when the error is due to a two-file operation,
     such as 'rename()' or 'link()'.  Returns the error message as a string
@@ -165,20 +315,18 @@
 
     return error
 
+
 # Needed by 'split_quoted()'
 _wordchars_re = _squote_re = _dquote_re = None
-
 def _init_regex():
     global _wordchars_re, _squote_re, _dquote_re
     _wordchars_re = re.compile(r'[^\\\'\"%s ]*' % string.whitespace)
     _squote_re = re.compile(r"'(?:[^'\\]|\\.)*'")
     _dquote_re = re.compile(r'"(?:[^"\\]|\\.)*"')
 
-def split_quoted(s):
+def split_quoted (s):
     """Split a string up according to Unix shell-like rules for quotes and
-    backslashes.
-
-    In short: words are delimited by spaces, as long as those
+    backslashes.  In short: words are delimited by spaces, as long as those
     spaces are not escaped by a backslash, or inside a quoted string.
     Single and double quotes are equivalent, and the quote characters can
     be backslash-escaped.  The backslash is stripped from any two-character
@@ -186,6 +334,7 @@
     characters are stripped from any quoted string.  Returns a list of
     words.
     """
+
     # This is a nice algorithm for splitting up a single string, since it
     # doesn't require character-by-character examination.  It was a little
     # bit of a brain-bender to get it working right, though...
@@ -233,12 +382,13 @@
 
     return words
 
+# split_quoted ()
 
-def execute(func, args, msg=None, verbose=0, dry_run=0):
-    """Perform some action that affects the outside world.
 
-    eg. by writing to the filesystem).  Such actions are special because
-    they are disabled by the 'dry_run' flag.  This method takes care of all
+def execute (func, args, msg=None, verbose=0, dry_run=0):
+    """Perform some action that affects the outside world (eg.  by
+    writing to the filesystem).  Such actions are special because they
+    are disabled by the 'dry_run' flag.  This method takes care of all
     that bureaucracy for you; all you have to do is supply the
     function to call and an argument tuple for it (to embody the
     "external action" being performed), and an optional message to
@@ -254,7 +404,7 @@
         func(*args)
 
 
-def strtobool(val):
+def strtobool (val):
     """Convert a string representation of truth to true (1) or false (0).
 
     True values are 'y', 'yes', 't', 'true', 'on', and '1'; false values
@@ -270,13 +420,15 @@
         raise ValueError("invalid truth value %r" % (val,))
 
 
-def byte_compile(py_files, optimize=0, force=0, prefix=None, base_dir=None,
-                  verbose=1, dry_run=0, direct=None):
+def byte_compile (py_files,
+                  optimize=0, force=0,
+                  prefix=None, base_dir=None,
+                  verbose=1, dry_run=0,
+                  direct=None):
     """Byte-compile a collection of Python source files to either .pyc
-    or .pyo files in the same directory.
-
-    'py_files' is a list of files to compile; any files that don't end in
-    ".py" are silently skipped. 'optimize' must be one of the following:
+    or .pyo files in the same directory.  'py_files' is a list of files
+    to compile; any files that don't end in ".py" are silently skipped.
+    'optimize' must be one of the following:
       0 - don't optimize (generate .pyc)
       1 - normal optimization (like "python -O")
       2 - extra optimization (like "python -OO")
@@ -392,8 +544,8 @@
             dfile = file
             if prefix:
                 if file[:len(prefix)] != prefix:
-                    raise ValueError("invalid prefix: filename %r doesn't "
-                                     "start with %r" % (file, prefix))
+                    raise ValueError("invalid prefix: filename %r doesn't start with %r"
+                           % (file, prefix))
                 dfile = dfile[len(prefix):]
             if base_dir:
                 dfile = os.path.join(base_dir, dfile)
@@ -408,8 +560,9 @@
                     log.debug("skipping byte-compilation of %s to %s",
                               file, cfile_base)
 
+# byte_compile ()
 
-def rfc822_escape(header):
+def rfc822_escape (header):
     """Return a version of the string escaped for inclusion in an
     RFC-822 header, by ensuring there are 8 spaces space after each newline.
     """
@@ -417,56 +570,6 @@
     sep = '\n' + 8 * ' '
     return sep.join(lines)
 
-_RE_VERSION = re.compile(b'(\d+\.\d+(\.\d+)*)')
-_MAC_OS_X_LD_VERSION = re.compile(b'^@\(#\)PROGRAM:ld  PROJECT:ld64-((\d+)(\.\d+)*)')
-
-def _find_ld_version():
-    """Finds the ld version. The version scheme differs under Mac OSX."""
-    if sys.platform == 'darwin':
-        return _find_exe_version('ld -v', _MAC_OS_X_LD_VERSION)
-    else:
-        return _find_exe_version('ld -v')
-
-def _find_exe_version(cmd, pattern=_RE_VERSION):
-    """Find the version of an executable by running `cmd` in the shell.
-
-    `pattern` is a compiled regular expression. If not provided, default
-    to _RE_VERSION. If the command is not found, or the output does not
-    match the mattern, returns None.
-    """
-    from subprocess import Popen, PIPE
-    executable = cmd.split()[0]
-    if find_executable(executable) is None:
-        return None
-    pipe = Popen(cmd, shell=True, stdout=PIPE, stderr=PIPE)
-    try:
-        stdout, stderr = pipe.stdout.read(), pipe.stderr.read()
-    finally:
-        pipe.stdout.close()
-        pipe.stderr.close()
-    # some commands like ld under MacOS X, will give the
-    # output in the stderr, rather than stdout.
-    if stdout != b'':
-        out_string = stdout
-    else:
-        out_string = stderr
-
-    result = pattern.search(out_string)
-    if result is None:
-        return None
-    return LooseVersion(result.group(1).decode())
-
-def get_compiler_versions():
-    """Returns a tuple providing the versions of gcc, ld and dllwrap
-
-    For each command, if a command is not found, None is returned.
-    Otherwise a LooseVersion instance is returned.
-    """
-    gcc = _find_exe_version('gcc -dumpversion')
-    ld = _find_ld_version()
-    dllwrap = _find_exe_version('dllwrap --version')
-    return gcc, ld, dllwrap
-
 # 2to3 support
 
 def run_2to3(files, fixer_names=None, options=None, explicit=None):

Modified: python/branches/import_unicode/Lib/email/base64mime.py
==============================================================================
--- python/branches/import_unicode/Lib/email/base64mime.py	(original)
+++ python/branches/import_unicode/Lib/email/base64mime.py	Fri Jul 30 01:19:37 2010
@@ -74,12 +74,12 @@
 
 
 def body_encode(s, maxlinelen=76, eol=NL):
-    """Encode a string with base64.
+    r"""Encode a string with base64.
 
     Each line will be wrapped at, at most, maxlinelen characters (defaults to
     76 characters).
 
-    Each line of encoded text will end with eol, which defaults to "\\n".  Set
+    Each line of encoded text will end with eol, which defaults to "\n".  Set
     this to "\r\n" if you will be using the result of this function directly
     in an email.
     """

Modified: python/branches/import_unicode/Lib/email/feedparser.py
==============================================================================
--- python/branches/import_unicode/Lib/email/feedparser.py	(original)
+++ python/branches/import_unicode/Lib/email/feedparser.py	Fri Jul 30 01:19:37 2010
@@ -104,6 +104,10 @@
         # data after the final RE.  In the case of a NL/CR terminated string,
         # this is the empty string.
         self._partial = parts.pop()
+        #GAN 29Mar09  bugs 1555570, 1721862  Confusion at 8K boundary ending with \r:
+        # is there a \n to follow later?
+        if not self._partial and parts and parts[-1].endswith('\r'):
+            self._partial = parts.pop(-2)+parts.pop()
         # parts is a list of strings, alternating between the line contents
         # and the eol character(s).  Gather up a list of lines after
         # re-attaching the newlines.

Modified: python/branches/import_unicode/Lib/email/message.py
==============================================================================
--- python/branches/import_unicode/Lib/email/message.py	(original)
+++ python/branches/import_unicode/Lib/email/message.py	Fri Jul 30 01:19:37 2010
@@ -198,17 +198,19 @@
             return None
         cte = self.get('content-transfer-encoding', '').lower()
         if cte == 'quoted-printable':
+            if isinstance(payload, str):
+                payload = payload.encode('ascii')
             return utils._qdecode(payload)
         elif cte == 'base64':
             try:
                 if isinstance(payload, str):
-                    payload = payload.encode('raw-unicode-escape')
+                    payload = payload.encode('ascii')
                 return base64.b64decode(payload)
             except binascii.Error:
                 # Incorrect padding
                 pass
         elif cte in ('x-uuencode', 'uuencode', 'uue', 'x-uue'):
-            in_file = BytesIO(payload.encode('raw-unicode-escape'))
+            in_file = BytesIO(payload.encode('ascii'))
             out_file = BytesIO()
             try:
                 uu.decode(in_file, out_file, quiet=True)

Modified: python/branches/import_unicode/Lib/email/test/test_email.py
==============================================================================
--- python/branches/import_unicode/Lib/email/test/test_email.py	(original)
+++ python/branches/import_unicode/Lib/email/test/test_email.py	Fri Jul 30 01:19:37 2010
@@ -2454,6 +2454,39 @@
 -Me
 """)
 
+    def test_pushCR_LF(self):
+        '''FeedParser BufferedSubFile.push() assumed it received complete
+           line endings.  A CR ending one push() followed by a LF starting
+           the next push() added an empty line.
+        '''
+        imt = [
+            ("a\r \n",  2),
+            ("b",       0),
+            ("c\n",     1),
+            ("",        0),
+            ("d\r\n",   1),
+            ("e\r",     0),
+            ("\nf",     1),
+            ("\r\n",    1),
+          ]
+        from email.feedparser import BufferedSubFile, NeedMoreData
+        bsf = BufferedSubFile()
+        om = []
+        nt = 0
+        for il, n in imt:
+            bsf.push(il)
+            nt += n
+            n1 = 0
+            while True:
+                ol = bsf.readline()
+                if ol == NeedMoreData:
+                    break
+                om.append(ol)
+                n1 += 1
+            self.assertTrue(n == n1)
+        self.assertTrue(len(om) == nt)
+        self.assertTrue(''.join([il for il, n in imt]) == ''.join(om))
+
 
 
 class TestParsers(TestEmailBase):

Modified: python/branches/import_unicode/Lib/fnmatch.py
==============================================================================
--- python/branches/import_unicode/Lib/fnmatch.py	(original)
+++ python/branches/import_unicode/Lib/fnmatch.py	Fri Jul 30 01:19:37 2010
@@ -9,13 +9,22 @@
 The function translate(PATTERN) returns a regular expression
 corresponding to PATTERN.  (It does not compile it.)
 """
-
+import os
+import posixpath
 import re
 
-__all__ = ["filter", "fnmatch","fnmatchcase","translate"]
+__all__ = ["filter", "fnmatch", "fnmatchcase", "purge", "translate"]
 
 _cache = {}  # Maps text patterns to compiled regexen.
 _cacheb = {}  # Ditto for bytes patterns.
+_MAXCACHE = 100  # Maximum size of caches.
+
+
+def purge():
+    """Clear the pattern cache."""
+    _cache.clear()
+    _cacheb.clear()
+
 
 def fnmatch(name, pat):
     """Test whether FILENAME matches PATTERN.
@@ -32,12 +41,11 @@
     if the operating system requires it.
     If you don't want this, use fnmatchcase(FILENAME, PATTERN).
     """
-
-    import os
     name = os.path.normcase(name)
     pat = os.path.normcase(pat)
     return fnmatchcase(name, pat)
 
+
 def _compile_pattern(pat):
     cache = _cacheb if isinstance(pat, bytes) else _cache
     regex = cache.get(pat)
@@ -48,12 +56,14 @@
             res = bytes(res_str, 'ISO-8859-1')
         else:
             res = translate(pat)
+        if len(cache) >= _MAXCACHE:
+            cache.clear()
         cache[pat] = regex = re.compile(res)
     return regex.match
 
+
 def filter(names, pat):
-    """Return the subset of the list NAMES that match PAT"""
-    import os,posixpath
+    """Return the subset of the list NAMES that match PAT."""
     result = []
     pat = os.path.normcase(pat)
     match = _compile_pattern(pat)
@@ -68,16 +78,17 @@
                 result.append(name)
     return result
 
+
 def fnmatchcase(name, pat):
     """Test whether FILENAME matches PATTERN, including case.
 
     This is a version of fnmatch() which doesn't case-normalize
     its arguments.
     """
-
     match = _compile_pattern(pat)
     return match(name) is not None
 
+
 def translate(pat):
     """Translate a shell PATTERN to a regular expression.
 

Modified: python/branches/import_unicode/Lib/getopt.py
==============================================================================
--- python/branches/import_unicode/Lib/getopt.py	(original)
+++ python/branches/import_unicode/Lib/getopt.py	Fri Jul 30 01:19:37 2010
@@ -156,7 +156,7 @@
             if not args:
                 raise GetoptError('option --%s requires argument' % opt, opt)
             optarg, args = args[0], args[1:]
-    elif optarg:
+    elif optarg is not None:
         raise GetoptError('option --%s must not have an argument' % opt, opt)
     opts.append(('--' + opt, optarg or ''))
     return opts, args

Modified: python/branches/import_unicode/Lib/http/client.py
==============================================================================
--- python/branches/import_unicode/Lib/http/client.py	(original)
+++ python/branches/import_unicode/Lib/http/client.py	Fri Jul 30 01:19:37 2010
@@ -734,11 +734,6 @@
             else:
                 raise NotConnected()
 
-        # send the data to the server. if we get a broken pipe, then close
-        # the socket. we want to reconnect when somebody tries to send again.
-        #
-        # NOTE: we DO propagate the error, though, because we cannot simply
-        #       ignore the error... the caller will know if they can retry.
         if self.debuglevel > 0:
             print("send:", repr(str))
         blocksize = 8192

Modified: python/branches/import_unicode/Lib/http/cookiejar.py
==============================================================================
--- python/branches/import_unicode/Lib/http/cookiejar.py	(original)
+++ python/branches/import_unicode/Lib/http/cookiejar.py	Fri Jul 30 01:19:37 2010
@@ -609,17 +609,14 @@
     return req_host, erhn
 
 def request_path(request):
-    """request-URI, as defined by RFC 2965."""
+    """Path component of request-URI, as defined by RFC 2965."""
     url = request.get_full_url()
-    path, parameters, query, frag = urllib.parse.urlparse(url)[2:]
-    if parameters:
-        path = "%s;%s" % (path, parameters)
-    path = escape_path(path)
-    req_path = urllib.parse.urlunparse(("", "", path, "", query, frag))
-    if not req_path.startswith("/"):
+    parts = urllib.parse.urlsplit(url)
+    path = escape_path(parts.path)
+    if not path.startswith("/"):
         # fix bad RFC 2396 absoluteURI
-        req_path = "/"+req_path
-    return req_path
+        path = "/" + path
+    return path
 
 def request_port(request):
     host = request.get_host()

Modified: python/branches/import_unicode/Lib/idlelib/PyShell.py
==============================================================================
--- python/branches/import_unicode/Lib/idlelib/PyShell.py	(original)
+++ python/branches/import_unicode/Lib/idlelib/PyShell.py	Fri Jul 30 01:19:37 2010
@@ -548,7 +548,7 @@
         This method is called from the subprocess, and by returning from this
         method we allow the subprocess to unblock.  After a bit the shell
         requests the subprocess to open the remote stack viewer which returns a
-        static object looking at the last exceptiopn.  It is queried through
+        static object looking at the last exception.  It is queried through
         the RPC mechanism.
 
         """

Modified: python/branches/import_unicode/Lib/importlib/test/benchmark.py
==============================================================================
--- python/branches/import_unicode/Lib/importlib/test/benchmark.py	(original)
+++ python/branches/import_unicode/Lib/importlib/test/benchmark.py	Fri Jul 30 01:19:37 2010
@@ -1,13 +1,16 @@
-"""Benchmark some basic import use-cases."""
-# XXX
-#    - Bench from source (turn off bytecode generation)
-#    - Bench from bytecode (remove existence of source)
-#    - Bench bytecode generation
-#    - Bench extensions
+"""Benchmark some basic import use-cases.
+
+The assumption is made that this benchmark is run in a fresh interpreter and
+thus has no external changes made to import-related attributes in sys.
+
+"""
 from . import util
 from .source import util as source_util
+import decimal
 import imp
 import importlib
+import os
+import py_compile
 import sys
 import timeit
 
@@ -30,9 +33,9 @@
             # One execution too far
             if total_time > seconds:
                 count -= 1
-        yield count
+        yield count // seconds
 
-def from_cache(repeat):
+def from_cache(seconds, repeat):
     """sys.modules"""
     name = '<benchmark import>'
     module = imp.new_module(name)
@@ -40,31 +43,117 @@
     module.__package__ = ''
     with util.uncache(name):
         sys.modules[name] = module
-        for result in bench(name, repeat=repeat):
+        for result in bench(name, repeat=repeat, seconds=seconds):
             yield result
 
 
-def builtin_mod(repeat):
+def builtin_mod(seconds, repeat):
     """Built-in module"""
     name = 'errno'
     if name in sys.modules:
         del sys.modules[name]
-    for result in bench(name, lambda: sys.modules.pop(name), repeat=repeat):
+    # Relying on built-in importer being implicit.
+    for result in bench(name, lambda: sys.modules.pop(name), repeat=repeat,
+                        seconds=seconds):
+        yield result
+
+
+def source_wo_bytecode(seconds, repeat):
+    """Source w/o bytecode: simple"""
+    sys.dont_write_bytecode = True
+    try:
+        name = '__importlib_test_benchmark__'
+        # Clears out sys.modules and puts an entry at the front of sys.path.
+        with source_util.create_modules(name) as mapping:
+            assert not os.path.exists(imp.cache_from_source(mapping[name]))
+            for result in bench(name, lambda: sys.modules.pop(name), repeat=repeat,
+                                seconds=seconds):
+                yield result
+    finally:
+        sys.dont_write_bytecode = False
+
+
+def decimal_wo_bytecode(seconds, repeat):
+    """Source w/o bytecode: decimal"""
+    name = 'decimal'
+    decimal_bytecode = imp.cache_from_source(decimal.__file__)
+    if os.path.exists(decimal_bytecode):
+        os.unlink(decimal_bytecode)
+    sys.dont_write_bytecode = True
+    try:
+        for result in bench(name, lambda: sys.modules.pop(name), repeat=repeat,
+                            seconds=seconds):
+            yield result
+    finally:
+        sys.dont_write_bytecode = False
+
+
+def source_writing_bytecode(seconds, repeat):
+    """Source writing bytecode: simple"""
+    assert not sys.dont_write_bytecode
+    name = '__importlib_test_benchmark__'
+    with source_util.create_modules(name) as mapping:
+        def cleanup():
+            sys.modules.pop(name)
+            os.unlink(imp.cache_from_source(mapping[name]))
+        for result in bench(name, cleanup, repeat=repeat, seconds=seconds):
+            assert not os.path.exists(imp.cache_from_source(mapping[name]))
+            yield result
+
+
+def decimal_writing_bytecode(seconds, repeat):
+    """Source writing bytecode: decimal"""
+    assert not sys.dont_write_bytecode
+    name = 'decimal'
+    def cleanup():
+        sys.modules.pop(name)
+        os.unlink(imp.cache_from_source(decimal.__file__))
+    for result in bench(name, cleanup, repeat=repeat, seconds=seconds):
+        yield result
+
+
+def source_using_bytecode(seconds, repeat):
+    """Bytecode w/ source: simple"""
+    name = '__importlib_test_benchmark__'
+    with source_util.create_modules(name) as mapping:
+        py_compile.compile(mapping[name])
+        assert os.path.exists(imp.cache_from_source(mapping[name]))
+        for result in bench(name, lambda: sys.modules.pop(name), repeat=repeat,
+                            seconds=seconds):
+            yield result
+
+
+def decimal_using_bytecode(seconds, repeat):
+    """Bytecode w/ source: decimal"""
+    name = 'decimal'
+    py_compile.compile(decimal.__file__)
+    for result in bench(name, lambda: sys.modules.pop(name), repeat=repeat,
+                        seconds=seconds):
         yield result
 
 
-def main(import_, repeat=3):
+def main(import_):
     __builtins__.__import__ = import_
-    benchmarks = from_cache, builtin_mod
+    benchmarks = (from_cache, builtin_mod,
+                  source_using_bytecode, source_wo_bytecode,
+                  source_writing_bytecode,
+                  decimal_using_bytecode, decimal_writing_bytecode,
+                  decimal_wo_bytecode,)
+    seconds = 1
+    seconds_plural = 's' if seconds > 1 else ''
+    repeat = 3
+    header = "Measuring imports/second over {} second{}, best out of {}\n"
+    print(header.format(seconds, seconds_plural, repeat))
     for benchmark in benchmarks:
         print(benchmark.__doc__, "[", end=' ')
         sys.stdout.flush()
         results = []
-        for result in benchmark(repeat):
+        for result in benchmark(seconds=seconds, repeat=repeat):
             results.append(result)
             print(result, end=' ')
             sys.stdout.flush()
-        print("]", "best is", max(results))
+        assert not sys.dont_write_bytecode
+        print("]", "best is", format(max(results), ',d'))
 
 
 if __name__ == '__main__':
@@ -75,7 +164,7 @@
                         default=False, help="use the built-in __import__")
     options, args = parser.parse_args()
     if args:
-        raise RuntimeError("unrecognized args: {0}".format(args))
+        raise RuntimeError("unrecognized args: {}".format(args))
     import_ = __import__
     if not options.builtin:
         import_ = importlib.__import__

Modified: python/branches/import_unicode/Lib/json/encoder.py
==============================================================================
--- python/branches/import_unicode/Lib/json/encoder.py	(original)
+++ python/branches/import_unicode/Lib/json/encoder.py	Fri Jul 30 01:19:37 2010
@@ -397,7 +397,7 @@
             yield 'true'
         elif o is False:
             yield 'false'
-        elif isinstance(o, (int, int)):
+        elif isinstance(o, int):
             yield str(o)
         elif isinstance(o, float):
             yield _floatstr(o)

Modified: python/branches/import_unicode/Lib/json/tests/test_fail.py
==============================================================================
--- python/branches/import_unicode/Lib/json/tests/test_fail.py	(original)
+++ python/branches/import_unicode/Lib/json/tests/test_fail.py	Fri Jul 30 01:19:37 2010
@@ -74,3 +74,12 @@
                 pass
             else:
                 self.fail("Expected failure for fail{0}.json: {1!r}".format(idx, doc))
+
+    def test_non_string_keys_dict(self):
+        data = {'a' : 1, (1, 2) : 2}
+
+        #This is for c encoder
+        self.assertRaises(TypeError, json.dumps, data)
+
+        #This is for python encoder
+        self.assertRaises(TypeError, json.dumps, data, indent=True)

Modified: python/branches/import_unicode/Lib/ntpath.py
==============================================================================
--- python/branches/import_unicode/Lib/ntpath.py	(original)
+++ python/branches/import_unicode/Lib/ntpath.py	Fri Jul 30 01:19:37 2010
@@ -275,7 +275,7 @@
     # set i to index beyond p's last slash
     i = len(p)
     while i and p[i-1] not in seps:
-        i = i - 1
+        i -= 1
     head, tail = p[:i], p[i:]  # now tail has no slashes
     # remove trailing slashes from head, unless it's all slashes
     head2 = head
@@ -366,7 +366,7 @@
         return path
     i, n = 1, len(path)
     while i < n and path[i] not in _get_bothseps(path):
-        i = i + 1
+        i += 1
 
     if 'HOME' in os.environ:
         userhome = os.environ['HOME']
@@ -435,21 +435,21 @@
             pathlen = len(path)
             try:
                 index = path.index(c)
-                res = res + c + path[:index + 1]
+                res += c + path[:index + 1]
             except ValueError:
-                res = res + path
+                res += path
                 index = pathlen - 1
         elif c == percent:  # variable or '%'
             if path[index + 1:index + 2] == percent:
-                res = res + c
-                index = index + 1
+                res += c
+                index += 1
             else:
                 path = path[index+1:]
                 pathlen = len(path)
                 try:
                     index = path.index(percent)
                 except ValueError:
-                    res = res + percent + path
+                    res += percent + path
                     index = pathlen - 1
                 else:
                     var = path[:index]
@@ -461,11 +461,11 @@
                         value = '%' + var + '%'
                     if isinstance(path, bytes):
                         value = value.encode('ascii')
-                    res = res + value
+                    res += value
         elif c == dollar:  # variable or '$$'
             if path[index + 1:index + 2] == dollar:
-                res = res + c
-                index = index + 1
+                res += c
+                index += 1
             elif path[index + 1:index + 2] == brace:
                 path = path[index+2:]
                 pathlen = len(path)
@@ -483,23 +483,23 @@
                         value = '${' + var + '}'
                     if isinstance(path, bytes):
                         value = value.encode('ascii')
-                    res = res + value
+                    res += value
                 except ValueError:
                     if isinstance(path, bytes):
-                        res = res + b'${' + path
+                        res += b'${' + path
                     else:
-                        res = res + '${' + path
+                        res += '${' + path
                     index = pathlen - 1
             else:
                 var = ''
-                index = index + 1
+                index += 1
                 c = path[index:index + 1]
                 while c and c in varchars:
                     if isinstance(path, bytes):
-                        var = var + c.decode('ascii')
+                        var += c.decode('ascii')
                     else:
-                        var = var + c
-                    index = index + 1
+                        var += c
+                    index += 1
                     c = path[index:index + 1]
                 if var in os.environ:
                     value = os.environ[var]
@@ -507,12 +507,12 @@
                     value = '$' + var
                 if isinstance(path, bytes):
                     value = value.encode('ascii')
-                res = res + value
+                res += value
                 if c:
-                    index = index - 1
+                    index -= 1
         else:
-            res = res + c
-        index = index + 1
+            res += c
+        index += 1
     return res
 
 
@@ -529,7 +529,7 @@
 
     # collapse initial backslashes
     if path.startswith(sep):
-        prefix = prefix + sep
+        prefix += sep
         path = path.lstrip(sep)
 
     comps = path.split(sep)

Modified: python/branches/import_unicode/Lib/os.py
==============================================================================
--- python/branches/import_unicode/Lib/os.py	(original)
+++ python/branches/import_unicode/Lib/os.py	Fri Jul 30 01:19:37 2010
@@ -446,7 +446,9 @@
         return len(self.data)
 
     def __repr__(self):
-        return 'environ({!r})'.format(self.data)
+        return 'environ({{{}}})'.format(', '.join(
+            ('{!r}: {!r}'.format(self.decodekey(key), self.decodevalue(value))
+            for key, value in self.data.items())))
 
     def copy(self):
         return dict(self)
@@ -531,7 +533,8 @@
         The optional second argument can specify an alternate default.
         key, default and the result are bytes."""
         return environb.get(key, default)
-    __all__.append("getenvb")
+
+    __all__.extend(("environb", "getenvb"))
 
 def fsencode(value):
     """Encode value for use in the file system, environment variables

Deleted: python/branches/import_unicode/Lib/pdb.doc
==============================================================================
--- python/branches/import_unicode/Lib/pdb.doc	Fri Jul 30 01:19:37 2010
+++ (empty file)
@@ -1,202 +0,0 @@
-The Python Debugger Pdb
-=======================
-
-To use the debugger in its simplest form:
-
-        >>> import pdb
-        >>> pdb.run('<a statement>')
-
-The debugger's prompt is '(Pdb) '.  This will stop in the first
-function call in <a statement>.
-
-Alternatively, if a statement terminated with an unhandled exception,
-you can use pdb's post-mortem facility to inspect the contents of the
-traceback:
-
-        >>> <a statement>
-        <exception traceback>
-        >>> import pdb
-        >>> pdb.pm()
-
-The commands recognized by the debugger are listed in the next
-section.  Most can be abbreviated as indicated; e.g., h(elp) means
-that 'help' can be typed as 'h' or 'help' (but not as 'he' or 'hel',
-nor as 'H' or 'Help' or 'HELP').  Optional arguments are enclosed in
-square brackets.
-
-A blank line repeats the previous command literally, except for
-'list', where it lists the next 11 lines.
-
-Commands that the debugger doesn't recognize are assumed to be Python
-statements and are executed in the context of the program being
-debugged.  Python statements can also be prefixed with an exclamation
-point ('!').  This is a powerful way to inspect the program being
-debugged; it is even possible to change variables.  When an exception
-occurs in such a statement, the exception name is printed but the
-debugger's state is not changed.
-
-The debugger supports aliases, which can save typing.  And aliases can
-have parameters (see the alias help entry) which allows one a certain
-level of adaptability to the context under examination.
-
-Multiple commands may be entered on a single line, separated by the
-pair ';;'.  No intelligence is applied to separating the commands; the
-input is split at the first ';;', even if it is in the middle of a
-quoted string.
-
-If a file ".pdbrc" exists in your home directory or in the current
-directory, it is read in and executed as if it had been typed at the
-debugger prompt.  This is particularly useful for aliases.  If both
-files exist, the one in the home directory is read first and aliases
-defined there can be overriden by the local file.
-
-Aside from aliases, the debugger is not directly programmable; but it
-is implemented as a class from which you can derive your own debugger
-class, which you can make as fancy as you like.
-
-
-Debugger commands
-=================
-
-h(elp)
-        Without argument, print the list of available commands.  With
-        a command name as argument, print help about that command
-        (this is currently not implemented).
-
-w(here)
-        Print a stack trace, with the most recent frame at the bottom.
-        An arrow indicates the "current frame", which determines the
-        context of most commands.
-
-d(own) [ count ]
-        Move the current frame count (default one) levels down in the
-        stack trace (to a newer frame).
-
-u(p) [ count ]
-        Move the current frame count (default one) levels up in the
-        stack trace (to an older frame).
-
-b(reak) [ ([filename:]lineno | function) [, condition] ]
-        With a filename:line number argument, set a break there.  If
-        filename is omitted, use the current file.  With a function
-        name, set a break at the first executable line of that
-        function.  Without argument, list all breaks.  Each breakpoint
-        is assigned a number to which all the other breakpoint
-        commands refer.
-
-        The condition argument, if present, is a string which must
-        evaluate to true in order for the breakpoint to be honored.
-
-tbreak [ ([filename:]lineno | function) [, condition] ]
-        Temporary breakpoint, which is removed automatically when it
-        is first hit.  The arguments are the same as break.
-
-cl(ear) [bpnumber [bpnumber ...] ]
-        With a space separated list of breakpoint numbers, clear those
-        breakpoints.  Without argument, clear all breaks (but first
-        ask confirmation).
-
-disable bpnumber [bpnumber ...]
-        Disables the breakpoints given as a space separated list of
-        breakpoint numbers.  Disabling a breakpoint means it cannot
-        cause the program to stop execution, but unlike clearing a
-        breakpoint, it remains in the list of breakpoints and can be
-        (re-)enabled.
-
-enable bpnumber [bpnumber ...]
-        Enables the breakpoints specified.
-
-ignore bpnumber count
-        Sets the ignore count for the given breakpoint number.  If
-        count is omitted, the ignore count is set to 0.  A breakpoint
-        becomes active when the ignore count is zero.  When non-zero,
-        the count is decremented each time the breakpoint is reached
-        and the breakpoint is not disabled and any associated
-        condition evaluates to true.
-
-condition bpnumber condition
-        condition is an expression which must evaluate to true before
-        the breakpoint is honored.  If condition is absent, any
-        existing condition is removed; i.e., the breakpoint is made
-        unconditional.
-
-s(tep)
-        Execute the current line, stop at the first possible occasion
-        (either in a function that is called or in the current function).
-
-n(ext)
-        Continue execution until the next line in the current function
-        is reached or it returns.
-
-unt(il)
-        Continue execution until the line with a number greater than the 
-        current one is reached or until the current frame returns.
-
-r(eturn)
-        Continue execution until the current function returns.
-
-run [args...]
-        Restart the debugged python program. If a string is supplied it is
-        splitted with "shlex", and the result is used as the new sys.argv.
-        History, breakpoints, actions and debugger options are preserved.
-        "restart" is an alias for "run".
-
-c(ont(inue))
-        Continue execution, only stop when a breakpoint is encountered.
-
-l(ist) [first [,last]]
-        List source code for the current file.
-        Without arguments, list 11 lines around the current line
-        or continue the previous listing.
-        With one argument, list 11 lines starting at that line.
-        With two arguments, list the given range;
-        if the second argument is less than the first, it is a count.
-
-a(rgs)
-        Print the argument list of the current function.
-
-p expression
-        Print the value of the expression.
-
-(!) statement
-        Execute the (one-line) statement in the context of the current
-        stack frame.  The exclamation point can be omitted unless the
-        first word of the statement resembles a debugger command.  To
-        assign to a global variable you must always prefix the command
-        with a 'global' command, e.g.:
-        (Pdb) global list_options; list_options = ['-l']
-        (Pdb)
-
-
-whatis arg
-         Prints the type of the argument.
-
-alias [name [command]]
-        Creates an alias called 'name' that executes 'command'.  The
-        command must *not* be enclosed in quotes.  Replaceable
-        parameters can be indicated by %1, %2, and so on, while %* is
-        replaced by all the parameters.  If no command is given, the
-        current alias for name is shown. If no name is given, all
-        aliases are listed.
-
-        Aliases may be nested and can contain anything that can be
-        legally typed at the pdb prompt.  Note!  You *can* override
-        internal pdb commands with aliases!  Those internal commands
-        are then hidden until the alias is removed.  Aliasing is
-        recursively applied to the first word of the command line; all
-        other words in the line are left alone.
-
-        As an example, here are two useful aliases (especially when
-        placed in the .pdbrc file):
-
-        #Print instance variables (usage "pi classInst")
-        alias pi for k in %1.__dict__.keys(): print "%1.",k,"=",%1.__dict__[k]
-        #Print instance variables in self
-        alias ps pi self
-                
-unalias name
-        Deletes the specified alias.
-
-q(uit)
-        Quit from the debugger.
-        The program being executed is aborted.

Modified: python/branches/import_unicode/Lib/pdb.py
==============================================================================
--- python/branches/import_unicode/Lib/pdb.py	(original)
+++ python/branches/import_unicode/Lib/pdb.py	Fri Jul 30 01:19:37 2010
@@ -1,8 +1,217 @@
 #! /usr/bin/env python3
 
-"""A Python debugger."""
-
-# (See pdb.doc for documentation.)
+"""
+The Python Debugger Pdb
+=======================
+
+To use the debugger in its simplest form:
+
+        >>> import pdb
+        >>> pdb.run('<a statement>')
+
+The debugger's prompt is '(Pdb) '.  This will stop in the first
+function call in <a statement>.
+
+Alternatively, if a statement terminated with an unhandled exception,
+you can use pdb's post-mortem facility to inspect the contents of the
+traceback:
+
+        >>> <a statement>
+        <exception traceback>
+        >>> import pdb
+        >>> pdb.pm()
+
+The commands recognized by the debugger are listed in the next
+section.  Most can be abbreviated as indicated; e.g., h(elp) means
+that 'help' can be typed as 'h' or 'help' (but not as 'he' or 'hel',
+nor as 'H' or 'Help' or 'HELP').  Optional arguments are enclosed in
+square brackets.  Alternatives in the command syntax are separated
+by a vertical bar (|).
+
+A blank line repeats the previous command literally, except for
+'list', where it lists the next 11 lines.
+
+Commands that the debugger doesn't recognize are assumed to be Python
+statements and are executed in the context of the program being
+debugged.  Python statements can also be prefixed with an exclamation
+point ('!').  This is a powerful way to inspect the program being
+debugged; it is even possible to change variables or call functions.
+When an exception occurs in such a statement, the exception name is
+printed but the debugger's state is not changed.
+
+The debugger supports aliases, which can save typing.  And aliases can
+have parameters (see the alias help entry) which allows one a certain
+level of adaptability to the context under examination.
+
+Multiple commands may be entered on a single line, separated by the
+pair ';;'.  No intelligence is applied to separating the commands; the
+input is split at the first ';;', even if it is in the middle of a
+quoted string.
+
+If a file ".pdbrc" exists in your home directory or in the current
+directory, it is read in and executed as if it had been typed at the
+debugger prompt.  This is particularly useful for aliases.  If both
+files exist, the one in the home directory is read first and aliases
+defined there can be overriden by the local file.
+
+Aside from aliases, the debugger is not directly programmable; but it
+is implemented as a class from which you can derive your own debugger
+class, which you can make as fancy as you like.
+
+
+Debugger commands
+=================
+
+h(elp)
+        Without argument, print the list of available commands.  With
+        a command name as argument, print help about that command.
+
+w(here)
+        Print a stack trace, with the most recent frame at the bottom.
+        An arrow indicates the "current frame", which determines the
+        context of most commands.
+
+d(own) [ count ]
+        Move the current frame count (default one) levels down in the
+        stack trace (to a newer frame).
+
+u(p) [ count ]
+        Move the current frame count (default one) levels up in the
+        stack trace (to an older frame).
+
+b(reak) [ ([filename:]lineno | function) [, condition] ]
+        With a filename:lineno argument, set a break there.  If
+        filename is omitted, use the current file.  With a function
+        name, set a break at the first executable line of that
+        function.  Without argument, list all breaks.  Each breakpoint
+        is assigned a number to which all the other breakpoint
+        commands refer.
+
+        The condition argument, if present, is a string which must
+        evaluate to true in order for the breakpoint to be honored.
+
+tbreak [ ([filename:]lineno | function) [, condition] ]
+        Temporary breakpoint, which is removed automatically when it
+        is first hit.  The arguments are the same as for break.
+
+cl(ear) [bpnumber [bpnumber ...] ]
+        With a space separated list of breakpoint numbers, clear those
+        breakpoints.  Without argument, clear all breaks (but first
+        ask confirmation).
+
+disable bpnumber [bpnumber ...]
+        Disable the breakpoints given as a space separated list of
+        breakpoint numbers.  Disabling a breakpoint means it cannot
+        cause the program to stop execution, but unlike clearing a
+        breakpoint, it remains in the list of breakpoints and can be
+        (re-)enabled.
+
+enable bpnumber [bpnumber ...]
+        Enable the breakpoints specified.
+
+ignore bpnumber [count]
+        Set the ignore count for the given breakpoint number.  If
+        count is omitted, the ignore count is set to 0.  A breakpoint
+        becomes active when the ignore count is zero.  When non-zero,
+        the count is decremented each time the breakpoint is reached
+        and the breakpoint is not disabled and any associated
+        condition evaluates to true.
+
+condition bpnumber [condition]
+        Set a new condition for the breakpoint, an expression which
+        must evaluate to true before the breakpoint is honored.  If
+        condition is absent, any existing condition is removed; i.e.,
+        the breakpoint is made unconditional.
+
+commands [bpnumber]
+        Specify a list of commands for the breakpoint.  Type a line
+        containing just 'end' to terminate the commands.  The commands
+        are executed when the breakpoint is hit.
+
+        With no breakpoint number argument, refers to the last
+        breakpoint set.
+
+s(tep)
+        Execute the current line, stop at the first possible occasion
+        (either in a function that is called or in the current
+        function).
+
+n(ext)
+        Continue execution until the next line in the current function
+        is reached or it returns.
+
+unt(il)
+        Continue execution until the line with a number greater than
+        the current one is reached or until the current frame returns.
+
+r(eturn)
+        Continue execution until the current function returns.
+
+run [args...]
+        Restart the debugged python program. If a string is supplied
+        it is splitted with "shlex", and the result is used as the new
+        sys.argv.  History, breakpoints, actions and debugger options
+        are preserved.  "restart" is an alias for "run".
+
+c(ont(inue))
+        Continue execution, only stop when a breakpoint is encountered.
+
+l(ist) [first [,last]]
+        List source code for the current file.
+        Without arguments, list 11 lines around the current line
+        or continue the previous listing.
+        With one argument, list 11 lines starting at that line.
+        With two arguments, list the given range;
+        if the second argument is less than the first, it is a count.
+
+a(rgs)
+        Print the argument list of the current function.
+
+p expression
+        Print the value of the expression.
+
+(!) statement
+        Execute the (one-line) statement in the context of the current
+        stack frame.  The exclamation point can be omitted unless the
+        first word of the statement resembles a debugger command.  To
+        assign to a global variable you must always prefix the command
+        with a 'global' command, e.g.:
+        (Pdb) global list_options; list_options = ['-l']
+        (Pdb)
+
+
+whatis arg
+        Print the type of the argument.
+
+alias [name [command]]
+        Creates an alias called 'name' that executes 'command'.  The
+        command must *not* be enclosed in quotes.  Replaceable
+        parameters can be indicated by %1, %2, and so on, while %* is
+        replaced by all the parameters.  If no command is given, the
+        current alias for name is shown. If no name is given, all
+        aliases are listed.
+
+        Aliases may be nested and can contain anything that can be
+        legally typed at the pdb prompt.  Note!  You *can* override
+        internal pdb commands with aliases!  Those internal commands
+        are then hidden until the alias is removed.  Aliasing is
+        recursively applied to the first word of the command line; all
+        other words in the line are left alone.
+
+        As an example, here are two useful aliases (especially when
+        placed in the .pdbrc file):
+
+        # Print instance variables (usage "pi classInst")
+        alias pi for k in %1.__dict__.keys(): print "%1.",k,"=",%1.__dict__[k]
+        # Print instance variables in self
+        alias ps pi self
+
+unalias name
+        Delete the specified alias.
+
+q(uit)
+        Quit from the debugger. The program being executed is aborted.
+"""
 
 import sys
 import linecache
@@ -905,7 +1114,7 @@
                                                      prompt_prefix), file=self.stdout)
 
 
-    # Help methods (derived from pdb.doc)
+    # Help methods (derived from docstring)
 
     def help_help(self):
         self.help_h()
@@ -914,7 +1123,7 @@
         print("""h(elp)
 Without argument, print the list of available commands.
 With a command name as argument, print help about that command
-"help pdb" pipes the full documentation file to the $PAGER
+"help pdb" shows the full pdb documentation
 "help exec" gives help on the ! command""", file=self.stdout)
 
     def help_where(self):
@@ -1281,15 +1490,8 @@
 
 # print help
 def help():
-    for dirname in sys.path:
-        fullname = os.path.join(dirname, 'pdb.doc')
-        if os.path.exists(fullname):
-            sts = os.system('${PAGER-more} '+fullname)
-            if sts: print('*** Pager exit status:', sts)
-            break
-    else:
-        print('Sorry, can\'t find the help file "pdb.doc"', end=' ')
-        print('along the Python search path')
+    import pydoc
+    pydoc.pager(__doc__)
 
 def main():
     if not sys.argv[1:] or sys.argv[1] in ("--help", "-h"):

Modified: python/branches/import_unicode/Lib/pickle.py
==============================================================================
--- python/branches/import_unicode/Lib/pickle.py	(original)
+++ python/branches/import_unicode/Lib/pickle.py	Fri Jul 30 01:19:37 2010
@@ -1034,19 +1034,15 @@
     def _instantiate(self, klass, k):
         args = tuple(self.stack[k+1:])
         del self.stack[k:]
-        instantiated = False
-        if (not args and
-                isinstance(klass, type) and
-                not hasattr(klass, "__getinitargs__")):
-            value = _EmptyClass()
-            value.__class__ = klass
-            instantiated = True
-        if not instantiated:
+        if (args or not isinstance(klass, type) or
+            hasattr(klass, "__getinitargs__")):
             try:
                 value = klass(*args)
             except TypeError as err:
                 raise TypeError("in constructor for %s: %s" %
                                 (klass.__name__, str(err)), sys.exc_info()[2])
+        else:
+            value = klass.__new__(klass)
         self.append(value)
 
     def load_inst(self):
@@ -1239,14 +1235,7 @@
         raise _Stop(value)
     dispatch[STOP[0]] = load_stop
 
-# Helper class for load_inst/load_obj
-
-class _EmptyClass:
-    pass
-
-# Encode/decode longs in linear time.
-
-import binascii as _binascii
+# Encode/decode longs.
 
 def encode_long(x):
     r"""Encode a long to a two's complement little-endian binary string.
@@ -1333,4 +1322,26 @@
     return doctest.testmod()
 
 if __name__ == "__main__":
-    _test()
+    import sys, argparse
+    parser = argparse.ArgumentParser(
+        description='display contents of the pickle files')
+    parser.add_argument(
+        'pickle_file', type=argparse.FileType('br'),
+        nargs='*', help='the pickle file')
+    parser.add_argument(
+        '-t', '--test', action='store_true',
+        help='run self-test suite')
+    parser.add_argument(
+        '-v', action='store_true',
+        help='run verbosely; only affects self-test run')
+    args = parser.parse_args()
+    if args.test:
+        _test()
+    else:
+        if not args.pickle_file:
+            parser.print_help()
+        else:
+            import pprint
+            for f in args.pickle_file:
+                obj = load(f)
+                pprint.pprint(obj)

Modified: python/branches/import_unicode/Lib/pickletools.py
==============================================================================
--- python/branches/import_unicode/Lib/pickletools.py	(original)
+++ python/branches/import_unicode/Lib/pickletools.py	Fri Jul 30 01:19:37 2010
@@ -733,7 +733,7 @@
 
 pyinteger_or_bool = StackObject(
                         name='int_or_bool',
-                        obtype=(int, int, bool),
+                        obtype=(int, bool),
                         doc="A Python integer object (short or long), or "
                             "a Python bool.")
 
@@ -1250,10 +1250,11 @@
       stack_before=[anyobject],
       stack_after=[pytuple],
       proto=2,
-      doc="""One-tuple.
+      doc="""Build a one-tuple out of the topmost item on the stack.
 
       This code pops one value off the stack and pushes a tuple of
-      length 1 whose one item is that value back onto it.  IOW:
+      length 1 whose one item is that value back onto it.  In other
+      words:
 
           stack[-1] = tuple(stack[-1:])
       """),
@@ -1264,10 +1265,11 @@
       stack_before=[anyobject, anyobject],
       stack_after=[pytuple],
       proto=2,
-      doc="""One-tuple.
+      doc="""Build a two-tuple out of the top two items on the stack.
 
-      This code pops two values off the stack and pushes a tuple
-      of length 2 whose items are those values back onto it.  IOW:
+      This code pops two values off the stack and pushes a tuple of
+      length 2 whose items are those values back onto it.  In other
+      words:
 
           stack[-2:] = [tuple(stack[-2:])]
       """),
@@ -1278,10 +1280,11 @@
       stack_before=[anyobject, anyobject, anyobject],
       stack_after=[pytuple],
       proto=2,
-      doc="""One-tuple.
+      doc="""Build a three-tuple out of the top three items on the stack.
 
-      This code pops three values off the stack and pushes a tuple
-      of length 3 whose items are those values back onto it.  IOW:
+      This code pops three values off the stack and pushes a tuple of
+      length 3 whose items are those values back onto it.  In other
+      words:
 
           stack[-3:] = [tuple(stack[-3:])]
       """),
@@ -1904,7 +1907,7 @@
 ##############################################################################
 # A symbolic pickle disassembler.
 
-def dis(pickle, out=None, memo=None, indentlevel=4):
+def dis(pickle, out=None, memo=None, indentlevel=4, annotate=0):
     """Produce a symbolic disassembly of a pickle.
 
     'pickle' is a file-like object, or string, containing a (at least one)
@@ -1920,9 +1923,15 @@
     to proceed across multiple pickles that were all created by the same
     pickler with the same memo.  Ordinarily you don't need to worry about this.
 
-    Optional arg indentlevel is the number of blanks by which to indent
+    Optional arg 'indentlevel' is the number of blanks by which to indent
     a new MARK level.  It defaults to 4.
 
+    Optional arg 'annotate' if nonzero instructs dis() to add short
+    description of the opcode on each line of disassembled output.
+    The value given to 'annotate' must be an integer and is used as a
+    hint for the column where annotation should start.  The default
+    value is 0, meaning no annotations.
+
     In addition to printing the disassembly, some sanity checks are made:
 
     + All embedded opcode arguments "make sense".
@@ -1950,6 +1959,7 @@
     markstack = []      # bytecode positions of MARK opcodes
     indentchunk = ' ' * indentlevel
     errormsg = None
+    annocol = annotate  # columnt hint for annotations
     for opcode, arg, pos in genops(pickle):
         if pos is not None:
             print("%5d:" % pos, end=' ', file=out)
@@ -2017,6 +2027,13 @@
                 line += ' ' + repr(arg)
             if markmsg:
                 line += ' ' + markmsg
+        if annotate:
+            line += ' ' * (annocol - len(line))
+            # make a mild effort to align annotations
+            annocol = len(line)
+            if annocol > 50:
+                annocol = annotate
+            line += ' ' + opcode.doc.split('\n', 1)[0]
         print(line, file=out)
 
         if errormsg:
@@ -2290,6 +2307,22 @@
    12: h    BINGET     1
    14: .    STOP
 highest protocol among opcodes = 2
+
+Try protocol 3 with annotations:
+
+>>> dis(pickle.dumps(T, 3), annotate=1)
+    0: \x80 PROTO      3 Protocol version indicator.
+    2: ]    EMPTY_LIST   Push an empty list.
+    3: q    BINPUT     0 Store the stack top into the memo.  The stack is not popped.
+    5: h    BINGET     0 Read an object from the memo and push it on the stack.
+    7: \x85 TUPLE1       Build a one-tuple out of the topmost item on the stack.
+    8: q    BINPUT     1 Store the stack top into the memo.  The stack is not popped.
+   10: a    APPEND       Append an object to a list.
+   11: 0    POP          Discard the top stack item, shrinking the stack by one item.
+   12: h    BINGET     1 Read an object from the memo and push it on the stack.
+   14: .    STOP         Stop the unpickling machine.
+highest protocol among opcodes = 2
+
 """
 
 _memo_test = r"""
@@ -2346,6 +2379,9 @@
         '-l', '--indentlevel', default=4, type=int,
         help='the number of blanks by which to indent a new MARK level')
     parser.add_argument(
+        '-a', '--annotate',  action='store_true',
+        help='annotate each line with a short opcode description')
+    parser.add_argument(
         '-p', '--preamble', default="==> {name} <==",
         help='if more than one pickle file is specified, print this before'
         ' each disassembly')
@@ -2359,14 +2395,15 @@
     if args.test:
         _test()
     else:
+        annotate = 30 if args.annotate else 0
         if not args.pickle_file:
             parser.print_help()
         elif len(args.pickle_file) == 1:
-            dis(args.pickle_file[0], args.output,
-                indentlevel=args.indentlevel)
+            dis(args.pickle_file[0], args.output, None,
+                args.indentlevel, annotate)
         else:
             memo = {} if args.memo else None
             for f in args.pickle_file:
                 preamble = args.preamble.format(name=f.name)
                 args.output.write(preamble + '\n')
-                dis(f, args.output, memo, args.indentlevel)
+                dis(f, args.output, memo, args.indentlevel, annotate)

Modified: python/branches/import_unicode/Lib/platform.py
==============================================================================
--- python/branches/import_unicode/Lib/platform.py	(original)
+++ python/branches/import_unicode/Lib/platform.py	Fri Jul 30 01:19:37 2010
@@ -724,27 +724,19 @@
 
     return hex(bcd)[2:]
 
-def mac_ver(release='',versioninfo=('','',''),machine=''):
-
-    """ Get MacOS version information and return it as tuple (release,
-        versioninfo, machine) with versioninfo being a tuple (version,
-        dev_stage, non_release_version).
-
-        Entries which cannot be determined are set to the paramter values
-        which default to ''. All tuple entries are strings.
-
+def _mac_ver_gestalt():
+    """
         Thanks to Mark R. Levinson for mailing documentation links and
         code examples for this function. Documentation for the
         gestalt() API is available online at:
 
            http://www.rgaros.nl/gestalt/
-
     """
     # Check whether the version info module is available
     try:
         import _gestalt
     except ImportError:
-        return release,versioninfo,machine
+        return None
     # Get the infos
     sysv, sysa = _mac_ver_lookup(('sysv','sysa'))
     # Decode the infos
@@ -768,6 +760,53 @@
         machine = {0x1: '68k',
                    0x2: 'PowerPC',
                    0xa: 'i386'}.get(sysa,'')
+
+    return release,versioninfo,machine
+
+def _mac_ver_xml():
+    fn = '/System/Library/CoreServices/SystemVersion.plist'
+    if not os.path.exists(fn):
+        return None
+
+    try:
+        import plistlib
+    except ImportError:
+        return None
+
+    pl = plistlib.readPlist(fn)
+    release = pl['ProductVersion']
+    versioninfo=('', '', '')
+    machine = os.uname()[4]
+    if machine == 'ppc':
+        # for compatibility with the gestalt based code
+        machine = 'PowerPC'
+
+    return release,versioninfo,machine
+
+
+def mac_ver(release='',versioninfo=('','',''),machine=''):
+
+    """ Get MacOS version information and return it as tuple (release,
+        versioninfo, machine) with versioninfo being a tuple (version,
+        dev_stage, non_release_version).
+
+        Entries which cannot be determined are set to the paramter values
+        which default to ''. All tuple entries are strings.
+    """
+
+    # First try reading the information from an XML file which should
+    # always be present
+    info = _mac_ver_xml()
+    if info is not None:
+        return info
+
+    # If that doesn't work for some reason fall back to reading the
+    # information using gestalt calls.
+    info = _mac_ver_gestalt()
+    if info is not None:
+        return info
+
+    # If that also doesn't work return the default values
     return release,versioninfo,machine
 
 def _java_getprop(name,default):

Modified: python/branches/import_unicode/Lib/re.py
==============================================================================
--- python/branches/import_unicode/Lib/re.py	(original)
+++ python/branches/import_unicode/Lib/re.py	Fri Jul 30 01:19:37 2010
@@ -254,7 +254,40 @@
 
 _pattern_type = type(sre_compile.compile("", 0))
 
-_MAXCACHE = 100
+_MAXCACHE = 500
+
+def _shrink_cache(cache_dict, max_length, divisor=5):
+    """Make room in the given cache.
+
+    Args:
+        cache_dict: The cache dictionary to modify.
+        max_length: Maximum # of entries in cache_dict before it is shrunk.
+        divisor: Cache will shrink to max_length - 1/divisor*max_length items.
+    """
+    # Toss out a fraction of the entries at random to make room for new ones.
+    # A random algorithm was chosen as opposed to simply cache_dict.popitem()
+    # as popitem could penalize the same regular expression repeatedly based
+    # on its internal hash value.  Being random should spread the cache miss
+    # love around.
+    cache_keys = tuple(cache_dict.keys())
+    overage = len(cache_keys) - max_length
+    if overage < 0:
+        # Cache is already within limits.  Normally this should not happen
+        # but it could due to multithreading.
+        return
+    number_to_toss = max_length // divisor + overage
+    # The import is done here to avoid a circular depencency.
+    import random
+    if not hasattr(random, 'sample'):
+        # Do nothing while resolving the circular dependency:
+        #  re->random->warnings->tokenize->string->re
+        return
+    for doomed_key in random.sample(cache_keys, number_to_toss):
+        try:
+            del cache_dict[doomed_key]
+        except KeyError:
+            # Ignore problems if the cache changed from another thread.
+            pass
 
 def _compile(*key):
     # internal: compile pattern
@@ -272,7 +305,7 @@
         raise TypeError("first argument must be string or compiled pattern")
     p = sre_compile.compile(pattern, flags)
     if len(_cache) >= _MAXCACHE:
-        _cache.clear()
+        _shrink_cache(_cache, _MAXCACHE)
     _cache[cachekey] = p
     return p
 
@@ -284,7 +317,7 @@
     repl, pattern = key
     p = sre_parse.parse_template(repl, pattern)
     if len(_cache_repl) >= _MAXCACHE:
-        _cache_repl.clear()
+        _shrink_cache(_cache_repl, _MAXCACHE)
     _cache_repl[key] = p
     return p
 

Modified: python/branches/import_unicode/Lib/smtpd.py
==============================================================================
--- python/branches/import_unicode/Lib/smtpd.py	(original)
+++ python/branches/import_unicode/Lib/smtpd.py	Fri Jul 30 01:19:37 2010
@@ -78,6 +78,7 @@
 import socket
 import asyncore
 import asynchat
+from warnings import warn
 
 __all__ = ["SMTPServer","DebuggingServer","PureProxy","MailmanProxy"]
 
@@ -111,35 +112,157 @@
 
     def __init__(self, server, conn, addr):
         asynchat.async_chat.__init__(self, conn)
-        self.__server = server
-        self.__conn = conn
-        self.__addr = addr
-        self.__line = []
-        self.__state = self.COMMAND
-        self.__greeting = 0
-        self.__mailfrom = None
-        self.__rcpttos = []
-        self.__data = ''
-        self.__fqdn = socket.getfqdn()
-        self.__peer = conn.getpeername()
-        print('Peer:', repr(self.__peer), file=DEBUGSTREAM)
-        self.push('220 %s %s' % (self.__fqdn, __version__))
+        self.smtp_server = server
+        self.conn = conn
+        self.addr = addr
+        self.received_lines = []
+        self.smtp_state = self.COMMAND
+        self.seen_greeting = ''
+        self.mailfrom = None
+        self.rcpttos = []
+        self.received_data = ''
+        self.fqdn = socket.getfqdn()
+        self.peer = conn.getpeername()
+        print('Peer:', repr(self.peer), file=DEBUGSTREAM)
+        self.push('220 %s %s' % (self.fqdn, __version__))
         self.set_terminator(b'\r\n')
 
+    # properties for backwards-compatibility
+    @property
+    def __server(self):
+        warn("Access to __server attribute on SMTPChannel is deprecated, "
+            "use 'smtp_server' instead", PendingDeprecationWarning, 2)
+        return self.smtp_server
+    @__server.setter
+    def __server(self, value):
+        warn("Setting __server attribute on SMTPChannel is deprecated, "
+            "set 'smtp_server' instead", PendingDeprecationWarning, 2)
+        self.smtp_server = value
+
+    @property
+    def __line(self):
+        warn("Access to __line attribute on SMTPChannel is deprecated, "
+            "use 'received_lines' instead", PendingDeprecationWarning, 2)
+        return self.received_lines
+    @__line.setter
+    def __line(self, value):
+        warn("Setting __line attribute on SMTPChannel is deprecated, "
+            "set 'received_lines' instead", PendingDeprecationWarning, 2)
+        self.received_lines = value
+
+    @property
+    def __state(self):
+        warn("Access to __state attribute on SMTPChannel is deprecated, "
+            "use 'smtp_state' instead", PendingDeprecationWarning, 2)
+        return self.smtp_state
+    @__state.setter
+    def __state(self, value):
+        warn("Setting __state attribute on SMTPChannel is deprecated, "
+            "set 'smtp_state' instead", PendingDeprecationWarning, 2)
+        self.smtp_state = value
+
+    @property
+    def __greeting(self):
+        warn("Access to __greeting attribute on SMTPChannel is deprecated, "
+            "use 'seen_greeting' instead", PendingDeprecationWarning, 2)
+        return self.seen_greeting
+    @__greeting.setter
+    def __greeting(self, value):
+        warn("Setting __greeting attribute on SMTPChannel is deprecated, "
+            "set 'seen_greeting' instead", PendingDeprecationWarning, 2)
+        self.seen_greeting = value
+
+    @property
+    def __mailfrom(self):
+        warn("Access to __mailfrom attribute on SMTPChannel is deprecated, "
+            "use 'mailfrom' instead", PendingDeprecationWarning, 2)
+        return self.mailfrom
+    @__mailfrom.setter
+    def __mailfrom(self, value):
+        warn("Setting __mailfrom attribute on SMTPChannel is deprecated, "
+            "set 'mailfrom' instead", PendingDeprecationWarning, 2)
+        self.mailfrom = value
+
+    @property
+    def __rcpttos(self):
+        warn("Access to __rcpttos attribute on SMTPChannel is deprecated, "
+            "use 'rcpttos' instead", PendingDeprecationWarning, 2)
+        return self.rcpttos
+    @__rcpttos.setter
+    def __rcpttos(self, value):
+        warn("Setting __rcpttos attribute on SMTPChannel is deprecated, "
+            "set 'rcpttos' instead", PendingDeprecationWarning, 2)
+        self.rcpttos = value
+
+    @property
+    def __data(self):
+        warn("Access to __data attribute on SMTPChannel is deprecated, "
+            "use 'received_data' instead", PendingDeprecationWarning, 2)
+        return self.received_data
+    @__data.setter
+    def __data(self, value):
+        warn("Setting __data attribute on SMTPChannel is deprecated, "
+            "set 'received_data' instead", PendingDeprecationWarning, 2)
+        self.received_data = value
+
+    @property
+    def __fqdn(self):
+        warn("Access to __fqdn attribute on SMTPChannel is deprecated, "
+            "use 'fqdn' instead", PendingDeprecationWarning, 2)
+        return self.fqdn
+    @__fqdn.setter
+    def __fqdn(self, value):
+        warn("Setting __fqdn attribute on SMTPChannel is deprecated, "
+            "set 'fqdn' instead", PendingDeprecationWarning, 2)
+        self.fqdn = value
+
+    @property
+    def __peer(self):
+        warn("Access to __peer attribute on SMTPChannel is deprecated, "
+            "use 'peer' instead", PendingDeprecationWarning, 2)
+        return self.peer
+    @__peer.setter
+    def __peer(self, value):
+        warn("Setting __peer attribute on SMTPChannel is deprecated, "
+            "set 'peer' instead", PendingDeprecationWarning, 2)
+        self.peer = value
+
+    @property
+    def __conn(self):
+        warn("Access to __conn attribute on SMTPChannel is deprecated, "
+            "use 'conn' instead", PendingDeprecationWarning, 2)
+        return self.conn
+    @__conn.setter
+    def __conn(self, value):
+        warn("Setting __conn attribute on SMTPChannel is deprecated, "
+            "set 'conn' instead", PendingDeprecationWarning, 2)
+        self.conn = value
+
+    @property
+    def __addr(self):
+        warn("Access to __addr attribute on SMTPChannel is deprecated, "
+            "use 'addr' instead", PendingDeprecationWarning, 2)
+        return self.addr
+    @__addr.setter
+    def __addr(self, value):
+        warn("Setting __addr attribute on SMTPChannel is deprecated, "
+            "set 'addr' instead", PendingDeprecationWarning, 2)
+        self.addr = value
+
     # Overrides base class for convenience
     def push(self, msg):
         asynchat.async_chat.push(self, bytes(msg + '\r\n', 'ascii'))
 
     # Implementation of base class abstract method
     def collect_incoming_data(self, data):
-        self.__line.append(str(data, "utf8"))
+        self.received_lines.append(str(data, "utf8"))
 
     # Implementation of base class abstract method
     def found_terminator(self):
-        line = EMPTYSTRING.join(self.__line)
+        line = EMPTYSTRING.join(self.received_lines)
         print('Data:', repr(line), file=DEBUGSTREAM)
-        self.__line = []
-        if self.__state == self.COMMAND:
+        self.received_lines = []
+        if self.smtp_state == self.COMMAND:
             if not line:
                 self.push('500 Error: bad syntax')
                 return
@@ -158,7 +281,7 @@
             method(arg)
             return
         else:
-            if self.__state != self.DATA:
+            if self.smtp_state != self.DATA:
                 self.push('451 Internal confusion')
                 return
             # Remove extraneous carriage returns and de-transparency according
@@ -169,14 +292,14 @@
                     data.append(text[1:])
                 else:
                     data.append(text)
-            self.__data = NEWLINE.join(data)
-            status = self.__server.process_message(self.__peer,
-                                                   self.__mailfrom,
-                                                   self.__rcpttos,
-                                                   self.__data)
-            self.__rcpttos = []
-            self.__mailfrom = None
-            self.__state = self.COMMAND
+            self.received_data = NEWLINE.join(data)
+            status = self.__server.process_message(self.peer,
+                                                   self.mailfrom,
+                                                   self.rcpttos,
+                                                   self.received_data)
+            self.rcpttos = []
+            self.mailfrom = None
+            self.smtp_state = self.COMMAND
             self.set_terminator(b'\r\n')
             if not status:
                 self.push('250 Ok')
@@ -188,11 +311,11 @@
         if not arg:
             self.push('501 Syntax: HELO hostname')
             return
-        if self.__greeting:
+        if self.seen_greeting:
             self.push('503 Duplicate HELO/EHLO')
         else:
-            self.__greeting = arg
-            self.push('250 %s' % self.__fqdn)
+            self.seen_greeting = arg
+            self.push('250 %s' % self.fqdn)
 
     def smtp_NOOP(self, arg):
         if arg:
@@ -225,24 +348,24 @@
         if not address:
             self.push('501 Syntax: MAIL FROM:<address>')
             return
-        if self.__mailfrom:
+        if self.mailfrom:
             self.push('503 Error: nested MAIL command')
             return
-        self.__mailfrom = address
-        print('sender:', self.__mailfrom, file=DEBUGSTREAM)
+        self.mailfrom = address
+        print('sender:', self.mailfrom, file=DEBUGSTREAM)
         self.push('250 Ok')
 
     def smtp_RCPT(self, arg):
         print('===> RCPT', arg, file=DEBUGSTREAM)
-        if not self.__mailfrom:
+        if not self.mailfrom:
             self.push('503 Error: need MAIL command')
             return
         address = self.__getaddr('TO:', arg) if arg else None
         if not address:
             self.push('501 Syntax: RCPT TO: <address>')
             return
-        self.__rcpttos.append(address)
-        print('recips:', self.__rcpttos, file=DEBUGSTREAM)
+        self.rcpttos.append(address)
+        print('recips:', self.rcpttos, file=DEBUGSTREAM)
         self.push('250 Ok')
 
     def smtp_RSET(self, arg):
@@ -250,26 +373,29 @@
             self.push('501 Syntax: RSET')
             return
         # Resets the sender, recipients, and data, but not the greeting
-        self.__mailfrom = None
-        self.__rcpttos = []
-        self.__data = ''
-        self.__state = self.COMMAND
+        self.mailfrom = None
+        self.rcpttos = []
+        self.received_data = ''
+        self.smtp_state = self.COMMAND
         self.push('250 Ok')
 
     def smtp_DATA(self, arg):
-        if not self.__rcpttos:
+        if not self.rcpttos:
             self.push('503 Error: need RCPT command')
             return
         if arg:
             self.push('501 Syntax: DATA')
             return
-        self.__state = self.DATA
+        self.smtp_state = self.DATA
         self.set_terminator(b'\r\n.\r\n')
         self.push('354 End data with <CR><LF>.<CR><LF>')
 
 
 
 class SMTPServer(asyncore.dispatcher):
+    # SMTPChannel class to use for managing client connections
+    channel_class = SMTPChannel
+
     def __init__(self, localaddr, remoteaddr):
         self._localaddr = localaddr
         self._remoteaddr = remoteaddr
@@ -291,7 +417,7 @@
     def handle_accept(self):
         conn, addr = self.accept()
         print('Incoming connection from %s' % repr(addr), file=DEBUGSTREAM)
-        channel = SMTPChannel(self, conn, addr)
+        channel = self.channel_class(self, conn, addr)
 
     # API for "doing something useful with the message"
     def process_message(self, peer, mailfrom, rcpttos, data):

Modified: python/branches/import_unicode/Lib/sndhdr.py
==============================================================================
--- python/branches/import_unicode/Lib/sndhdr.py	(original)
+++ python/branches/import_unicode/Lib/sndhdr.py	Fri Jul 30 01:19:37 2010
@@ -57,12 +57,12 @@
 
 def test_aifc(h, f):
     import aifc
-    if h.startswith(b'FORM'):
+    if not h.startswith(b'FORM'):
         return None
     if h[8:12] == b'AIFC':
         fmt = 'aifc'
     elif h[8:12] == b'AIFF':
-        fmt = b'aiff'
+        fmt = 'aiff'
     else:
         return None
     f.seek(0)
@@ -123,7 +123,7 @@
 
 
 def test_voc(h, f):
-    if h.startswith(b'Creative Voice File\032'):
+    if not h.startswith(b'Creative Voice File\032'):
         return None
     sbseek = get_short_le(h[20:22])
     rate = 0
@@ -150,7 +150,7 @@
 
 
 def test_8svx(h, f):
-    if h.startswith(b'FORM') or h[8:12] != b'8SVX':
+    if not h.startswith(b'FORM') or h[8:12] != b'8SVX':
         return None
     # Should decode it to get #channels -- assume always 1
     return '8svx', 0, 1, 0, 8

Modified: python/branches/import_unicode/Lib/socket.py
==============================================================================
--- python/branches/import_unicode/Lib/socket.py	(original)
+++ python/branches/import_unicode/Lib/socket.py	Fri Jul 30 01:19:37 2010
@@ -252,10 +252,6 @@
         self._sock._decref_socketios()
         self._sock = None
 
-    def __del__(self):
-        if not self.closed:
-            self._sock._decref_socketios()
-
 
 def getfqdn(name=''):
     """Get fully qualified domain name from name.

Modified: python/branches/import_unicode/Lib/string.py
==============================================================================
--- python/branches/import_unicode/Lib/string.py	(original)
+++ python/branches/import_unicode/Lib/string.py	Fri Jul 30 01:19:37 2010
@@ -81,7 +81,7 @@
                 'delim' : _re.escape(cls.delimiter),
                 'id'    : cls.idpattern,
                 }
-        cls.pattern = _re.compile(pattern, _re.IGNORECASE | _re.VERBOSE)
+        cls.pattern = _re.compile(pattern, cls.flags | _re.VERBOSE)
 
 
 class Template(metaclass=_TemplateMetaclass):
@@ -89,6 +89,7 @@
 
     delimiter = '$'
     idpattern = r'[_a-z][_a-z0-9]*'
+    flags = _re.IGNORECASE
 
     def __init__(self, template):
         self.template = template

Modified: python/branches/import_unicode/Lib/subprocess.py
==============================================================================
--- python/branches/import_unicode/Lib/subprocess.py	(original)
+++ python/branches/import_unicode/Lib/subprocess.py	Fri Jul 30 01:19:37 2010
@@ -1073,6 +1073,8 @@
 
             if shell:
                 args = ["/bin/sh", "-c"] + args
+                if executable:
+                    args[0] = executable
 
             if executable is None:
                 executable = args[0]

Modified: python/branches/import_unicode/Lib/sysconfig.py
==============================================================================
--- python/branches/import_unicode/Lib/sysconfig.py	(original)
+++ python/branches/import_unicode/Lib/sysconfig.py	Fri Jul 30 01:19:37 2010
@@ -225,6 +225,12 @@
     # do variable interpolation here
     variables = list(notdone.keys())
 
+    # Variables with a 'PY_' prefix in the makefile. These need to
+    # be made available without that prefix through sysconfig.
+    # Special care is needed to ensure that variable expansion works, even
+    # if the expansion uses the name without a prefix.
+    renamed_variables = ('CFLAGS', 'LDFLAGS', 'CPPFLAGS')
+
     while len(variables) > 0:
         for name in tuple(variables):
             value = notdone[name]
@@ -240,8 +246,20 @@
                 elif n in os.environ:
                     # do it like make: fall back to environment
                     item = os.environ[n]
+
+                elif n in renamed_variables:
+                    if name.startswith('PY_') and name[3:] in renamed_variables:
+                        item = ""
+
+                    elif 'PY_' + n in notdone:
+                        found = False
+
+                    else:
+                        item = str(done['PY_' + n])
+
                 else:
                     done[n] = item = ""
+
                 if found:
                     after = value[m.end():]
                     value = value[:m.start()] + item + after
@@ -255,6 +273,15 @@
                         else:
                             done[name] = value
                         variables.remove(name)
+
+                        if name.startswith('PY_') \
+                                and name[3:] in renamed_variables:
+
+                            name = name[3:]
+                            if name not in done:
+                                done[name] = value
+
+
             else:
                 # bogus variable reference; just drop it since we can't deal
                 variables.remove(name)
@@ -653,8 +680,7 @@
                 cflags = get_config_vars().get('CFLAGS')
 
                 archs = re.findall('-arch\s+(\S+)', cflags)
-                archs.sort()
-                archs = tuple(archs)
+                archs = tuple(sorted(set(archs)))
 
                 if len(archs) == 1:
                     machine = archs[0]

Modified: python/branches/import_unicode/Lib/tarfile.py
==============================================================================
--- python/branches/import_unicode/Lib/tarfile.py	(original)
+++ python/branches/import_unicode/Lib/tarfile.py	Fri Jul 30 01:19:37 2010
@@ -55,6 +55,15 @@
 except ImportError:
     grp = pwd = None
 
+# os.symlink on Windows prior to 6.0 raises NotImplementedError
+symlink_exception = (AttributeError, NotImplementedError)
+try:
+    # WindowsError (1314) will be raised if the caller does not hold the
+    # SeCreateSymbolicLinkPrivilege privilege
+    symlink_exception += (WindowsError,)
+except NameError:
+    pass
+
 # from tarfile import *
 __all__ = ["TarFile", "TarInfo", "is_tarfile", "TarError"]
 
@@ -2283,17 +2292,16 @@
                     os.link(tarinfo._link_target, targetpath)
                 else:
                     self._extract_mem
-        except (AttributeError, NotImplementedError, WindowsError):
-            # AttributeError if no os.symlink
-            # NotImplementedError if on Windows XP
-            # WindowsError (1314) if the required privilege is not held by the client
+        except symlink_exception:
             if tarinfo.issym():
-                linkpath = os.path.join(os.path.dirname(tarinfo.name),tarinfo.linkname)
+                linkpath = os.path.join(os.path.dirname(tarinfo.name),
+                                        tarinfo.linkname)
             else:
                 linkpath = tarinfo.linkname
         else:
             try:
-                self._extract_member(self._find_link_target(tarinfo), targetpath)
+                self._extract_member(self._find_link_target(tarinfo),
+                                     targetpath)
             except KeyError:
                 raise ExtractError("unable to resolve link inside archive")
 

Modified: python/branches/import_unicode/Lib/test/pickletester.py
==============================================================================
--- python/branches/import_unicode/Lib/test/pickletester.py	(original)
+++ python/branches/import_unicode/Lib/test/pickletester.py	Fri Jul 30 01:19:37 2010
@@ -65,9 +65,21 @@
     def __eq__(self, other):
         return self.__dict__ == other.__dict__
 
+class D(C):
+    def __init__(self, arg):
+        pass
+
+class E(C):
+    def __getinitargs__(self):
+        return ()
+
 import __main__
 __main__.C = C
 C.__module__ = "__main__"
+__main__.D = D
+D.__module__ = "__main__"
+__main__.E = E
+E.__module__ = "__main__"
 
 class myint(int):
     def __init__(self, x):
@@ -425,6 +437,63 @@
     def test_load_from_data2(self):
         self.assertEqual(self._testdata, self.loads(DATA2))
 
+    def test_load_classic_instance(self):
+        # See issue5180.  Test loading 2.x pickles that
+        # contain an instance of old style class.
+        for X, args in [(C, ()), (D, ('x',)), (E, ())]:
+            xname = X.__name__.encode('ascii')
+            # Protocol 0 (text mode pickle):
+            """
+            0: (    MARK
+            1: i        INST       '__main__ X' (MARK at 0)
+            15: p    PUT        0
+            18: (    MARK
+            19: d        DICT       (MARK at 18)
+            20: p    PUT        1
+            23: b    BUILD
+            24: .    STOP
+            """
+            pickle0 = (b"(i__main__\n"
+                       b"X\n"
+                       b"p0\n"
+                       b"(dp1\nb.").replace(b'X', xname)
+            self.assertEqual(X(*args), self.loads(pickle0))
+
+            # Protocol 1 (binary mode pickle)
+            """
+            0: (    MARK
+            1: c        GLOBAL     '__main__ X'
+            15: q        BINPUT     0
+            17: o        OBJ        (MARK at 0)
+            18: q    BINPUT     1
+            20: }    EMPTY_DICT
+            21: q    BINPUT     2
+            23: b    BUILD
+            24: .    STOP
+            """
+            pickle1 = (b'(c__main__\n'
+                       b'X\n'
+                       b'q\x00oq\x01}q\x02b.').replace(b'X', xname)
+            self.assertEqual(X(*args), self.loads(pickle1))
+
+            # Protocol 2 (pickle2 = b'\x80\x02' + pickle1)
+            """
+            0: \x80 PROTO      2
+            2: (    MARK
+            3: c        GLOBAL     '__main__ X'
+            17: q        BINPUT     0
+            19: o        OBJ        (MARK at 2)
+            20: q    BINPUT     1
+            22: }    EMPTY_DICT
+            23: q    BINPUT     2
+            25: b    BUILD
+            26: .    STOP
+            """
+            pickle2 = (b'\x80\x02(c__main__\n'
+                       b'X\n'
+                       b'q\x00oq\x01}q\x02b.').replace(b'X', xname)
+            self.assertEqual(X(*args), self.loads(pickle2))
+
     # There are gratuitous differences between pickles produced by
     # pickle and cPickle, largely because cPickle starts PUT indices at
     # 1 and pickle starts them at 0.  See XXX comment in cPickle's put2() --

Modified: python/branches/import_unicode/Lib/test/script_helper.py
==============================================================================
--- python/branches/import_unicode/Lib/test/script_helper.py	(original)
+++ python/branches/import_unicode/Lib/test/script_helper.py	Fri Jul 30 01:19:37 2010
@@ -79,7 +79,7 @@
             name_in_zip = os.path.basename(script_name)
     zip_file.write(script_name, name_in_zip)
     zip_file.close()
-    #if test.test_support.verbose:
+    #if test.support.verbose:
     #    zip_file = zipfile.ZipFile(zip_name, 'r')
     #    print 'Contents of %r:' % zip_name
     #    zip_file.printdir()
@@ -114,7 +114,7 @@
     zip_file.close()
     for name in unlink:
         os.unlink(name)
-    #if test.test_support.verbose:
+    #if test.support.verbose:
     #    zip_file = zipfile.ZipFile(zip_name, 'r')
     #    print 'Contents of %r:' % zip_name
     #    zip_file.printdir()

Modified: python/branches/import_unicode/Lib/test/support.py
==============================================================================
--- python/branches/import_unicode/Lib/test/support.py	(original)
+++ python/branches/import_unicode/Lib/test/support.py	Fri Jul 30 01:19:37 2010
@@ -109,7 +109,7 @@
         orig_modules[name] = sys.modules[name]
     except KeyError:
         saved = False
-    sys.modules[name] = 0
+    sys.modules[name] = None
     return saved
 
 

Modified: python/branches/import_unicode/Lib/test/symlink_support.py
==============================================================================
--- python/branches/import_unicode/Lib/test/symlink_support.py	(original)
+++ python/branches/import_unicode/Lib/test/symlink_support.py	Fri Jul 30 01:19:37 2010
@@ -12,7 +12,8 @@
 GetCurrentProcess = ctypes.windll.kernel32.GetCurrentProcess
 GetCurrentProcess.restype = wintypes.HANDLE
 OpenProcessToken = ctypes.windll.advapi32.OpenProcessToken
-OpenProcessToken.argtypes = (wintypes.HANDLE, wintypes.DWORD, ctypes.POINTER(wintypes.HANDLE))
+OpenProcessToken.argtypes = (wintypes.HANDLE, wintypes.DWORD,
+                             ctypes.POINTER(wintypes.HANDLE))
 OpenProcessToken.restype = wintypes.BOOL
 
 class LUID(ctypes.Structure):
@@ -91,7 +92,8 @@
 
     def get_array(self):
         array_type = LUID_AND_ATTRIBUTES*self.count
-        privileges = ctypes.cast(self.privileges, ctypes.POINTER(array_type)).contents
+        privileges = ctypes.cast(self.privileges,
+                                 ctypes.POINTER(array_type)).contents
         return privileges
 
     def __iter__(self):
@@ -133,7 +135,8 @@
 def get_symlink_luid():
     "Get the LUID for the SeCreateSymbolicLinkPrivilege"
     symlink_luid = LUID()
-    res = LookupPrivilegeValue(None, "SeCreateSymbolicLinkPrivilege", symlink_luid)
+    res = LookupPrivilegeValue(None, "SeCreateSymbolicLinkPrivilege",
+                               symlink_luid)
     if not res > 0:
         raise RuntimeError("Couldn't lookup privilege value")
     return symlink_luid

Modified: python/branches/import_unicode/Lib/test/test_array.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_array.py	(original)
+++ python/branches/import_unicode/Lib/test/test_array.py	Fri Jul 30 01:19:37 2010
@@ -326,6 +326,17 @@
                 f.close()
             support.unlink(support.TESTFN)
 
+    def test_fromfile_ioerror(self):
+        # Issue #5395: Check if fromfile raises a proper IOError
+        # instead of EOFError.
+        a = array.array(self.typecode)
+        f = open(support.TESTFN, 'wb')
+        try:
+            self.assertRaises(IOError, a.fromfile, f, len(self.example))
+        finally:
+            f.close()
+            support.unlink(support.TESTFN)
+
     def test_filewrite(self):
         a = array.array(self.typecode, 2*self.example)
         f = open(support.TESTFN, 'wb')

Modified: python/branches/import_unicode/Lib/test/test_ast.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_ast.py	(original)
+++ python/branches/import_unicode/Lib/test/test_ast.py	Fri Jul 30 01:19:37 2010
@@ -285,6 +285,8 @@
         self.assertEqual(ast.literal_eval('[1, 2, 3]'), [1, 2, 3])
         self.assertEqual(ast.literal_eval('{"foo": 42}'), {"foo": 42})
         self.assertEqual(ast.literal_eval('(True, False, None)'), (True, False, None))
+        self.assertEqual(ast.literal_eval('{1, 2, 3}'), {1, 2, 3})
+        self.assertEqual(ast.literal_eval('b"hi"'), b"hi")
         self.assertRaises(ValueError, ast.literal_eval, 'foo()')
 
     def test_literal_eval_issue4907(self):

Modified: python/branches/import_unicode/Lib/test/test_asyncore.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_asyncore.py	(original)
+++ python/branches/import_unicode/Lib/test/test_asyncore.py	Fri Jul 30 01:19:37 2010
@@ -118,65 +118,65 @@
     # http://mail.python.org/pipermail/python-list/2001-October/109973.html)
     # These constants should be present as long as poll is available
 
-    if hasattr(select, 'poll'):
-        def test_readwrite(self):
-            # Check that correct methods are called by readwrite()
-
-            attributes = ('read', 'expt', 'write', 'closed', 'error_handled')
-
-            expected = (
-                (select.POLLIN, 'read'),
-                (select.POLLPRI, 'expt'),
-                (select.POLLOUT, 'write'),
-                (select.POLLERR, 'closed'),
-                (select.POLLHUP, 'closed'),
-                (select.POLLNVAL, 'closed'),
-                )
-
-            class testobj:
-                def __init__(self):
-                    self.read = False
-                    self.write = False
-                    self.closed = False
-                    self.expt = False
-                    self.error_handled = False
-
-                def handle_read_event(self):
-                    self.read = True
-
-                def handle_write_event(self):
-                    self.write = True
-
-                def handle_close(self):
-                    self.closed = True
-
-                def handle_expt_event(self):
-                    self.expt = True
-
-                def handle_error(self):
-                    self.error_handled = True
-
-            for flag, expectedattr in expected:
-                tobj = testobj()
-                self.assertEqual(getattr(tobj, expectedattr), False)
-                asyncore.readwrite(tobj, flag)
-
-                # Only the attribute modified by the routine we expect to be
-                # called should be True.
-                for attr in attributes:
-                    self.assertEqual(getattr(tobj, attr), attr==expectedattr)
-
-                # check that ExitNow exceptions in the object handler method
-                # bubbles all the way up through asyncore readwrite call
-                tr1 = exitingdummy()
-                self.assertRaises(asyncore.ExitNow, asyncore.readwrite, tr1, flag)
-
-                # check that an exception other than ExitNow in the object handler
-                # method causes the handle_error method to get called
-                tr2 = crashingdummy()
-                self.assertEqual(tr2.error_handled, False)
-                asyncore.readwrite(tr2, flag)
-                self.assertEqual(tr2.error_handled, True)
+    @unittest.skipUnless(hasattr(select, 'poll'), 'select.poll required')
+    def test_readwrite(self):
+        # Check that correct methods are called by readwrite()
+
+        attributes = ('read', 'expt', 'write', 'closed', 'error_handled')
+
+        expected = (
+            (select.POLLIN, 'read'),
+            (select.POLLPRI, 'expt'),
+            (select.POLLOUT, 'write'),
+            (select.POLLERR, 'closed'),
+            (select.POLLHUP, 'closed'),
+            (select.POLLNVAL, 'closed'),
+            )
+
+        class testobj:
+            def __init__(self):
+                self.read = False
+                self.write = False
+                self.closed = False
+                self.expt = False
+                self.error_handled = False
+
+            def handle_read_event(self):
+                self.read = True
+
+            def handle_write_event(self):
+                self.write = True
+
+            def handle_close(self):
+                self.closed = True
+
+            def handle_expt_event(self):
+                self.expt = True
+
+            def handle_error(self):
+                self.error_handled = True
+
+        for flag, expectedattr in expected:
+            tobj = testobj()
+            self.assertEqual(getattr(tobj, expectedattr), False)
+            asyncore.readwrite(tobj, flag)
+
+            # Only the attribute modified by the routine we expect to be
+            # called should be True.
+            for attr in attributes:
+                self.assertEqual(getattr(tobj, attr), attr==expectedattr)
+
+            # check that ExitNow exceptions in the object handler method
+            # bubbles all the way up through asyncore readwrite call
+            tr1 = exitingdummy()
+            self.assertRaises(asyncore.ExitNow, asyncore.readwrite, tr1, flag)
+
+            # check that an exception other than ExitNow in the object handler
+            # method causes the handle_error method to get called
+            tr2 = crashingdummy()
+            self.assertEqual(tr2.error_handled, False)
+            asyncore.readwrite(tr2, flag)
+            self.assertEqual(tr2.error_handled, True)
 
     def test_closeall(self):
         self.closeall_check(False)
@@ -259,7 +259,7 @@
             sys.stderr = stderr
 
         lines = fp.getvalue().splitlines()
-        self.assertEquals(lines, ['log: %s' % l1, 'log: %s' % l2])
+        self.assertEqual(lines, ['log: %s' % l1, 'log: %s' % l2])
 
     def test_log_info(self):
         d = asyncore.dispatcher()
@@ -281,7 +281,7 @@
         lines = fp.getvalue().splitlines()
         expected = ['EGGS: %s' % l1, 'info: %s' % l2, 'SPAM: %s' % l3]
 
-        self.assertEquals(lines, expected)
+        self.assertEqual(lines, expected)
 
     def test_unhandled(self):
         d = asyncore.dispatcher()
@@ -306,7 +306,7 @@
                     'warning: unhandled write event',
                     'warning: unhandled connect event',
                     'warning: unhandled accept event']
-        self.assertEquals(lines, expected)
+        self.assertEqual(lines, expected)
 
     def test_issue_8594(self):
         # XXX - this test is supposed to be removed in next major Python
@@ -322,7 +322,7 @@
             warnings.simplefilter("always")
             family = d.family
             self.assertEqual(family, socket.AF_INET)
-            self.assertTrue(len(w) == 1)
+            self.assertEqual(len(w), 1)
             self.assertTrue(issubclass(w[0].category, DeprecationWarning))
 
     def test_strerror(self):
@@ -331,7 +331,7 @@
         if hasattr(os, 'strerror'):
             self.assertEqual(err, os.strerror(errno.EPERM))
         err = asyncore._strerror(-1)
-        self.assertTrue("unknown error" in err.lower())
+        self.assertIn("unknown error", err.lower())
 
 
 class dispatcherwithsend_noread(asyncore.dispatcher_with_send):
@@ -394,38 +394,52 @@
 class DispatcherWithSendTests_UsePoll(DispatcherWithSendTests):
     usepoll = True
 
-if hasattr(asyncore, 'file_wrapper'):
-    class FileWrapperTest(unittest.TestCase):
-        def setUp(self):
-            self.d = b"It's not dead, it's sleeping!"
-            open(TESTFN, 'wb').write(self.d)
-
-        def tearDown(self):
-            unlink(TESTFN)
-
-        def test_recv(self):
-            fd = os.open(TESTFN, os.O_RDONLY)
-            w = asyncore.file_wrapper(fd)
-            os.close(fd)
-
-            self.assertNotEqual(w.fd, fd)
-            self.assertNotEqual(w.fileno(), fd)
-            self.assertEqual(w.recv(13), b"It's not dead")
-            self.assertEqual(w.read(6), b", it's")
-            w.close()
-            self.assertRaises(OSError, w.read, 1)
-
-        def test_send(self):
-            d1 = b"Come again?"
-            d2 = b"I want to buy some cheese."
-            fd = os.open(TESTFN, os.O_WRONLY | os.O_APPEND)
-            w = asyncore.file_wrapper(fd)
-            os.close(fd)
-
-            w.write(d1)
-            w.send(d2)
-            w.close()
-            self.assertEqual(open(TESTFN, 'rb').read(), self.d + d1 + d2)
+ at unittest.skipUnless(hasattr(asyncore, 'file_wrapper'),
+                     'asyncore.file_wrapper required')
+class FileWrapperTest(unittest.TestCase):
+    def setUp(self):
+        self.d = b"It's not dead, it's sleeping!"
+        open(TESTFN, 'wb').write(self.d)
+
+    def tearDown(self):
+        unlink(TESTFN)
+
+    def test_recv(self):
+        fd = os.open(TESTFN, os.O_RDONLY)
+        w = asyncore.file_wrapper(fd)
+        os.close(fd)
+
+        self.assertNotEqual(w.fd, fd)
+        self.assertNotEqual(w.fileno(), fd)
+        self.assertEqual(w.recv(13), b"It's not dead")
+        self.assertEqual(w.read(6), b", it's")
+        w.close()
+        self.assertRaises(OSError, w.read, 1)
+
+    def test_send(self):
+        d1 = b"Come again?"
+        d2 = b"I want to buy some cheese."
+        fd = os.open(TESTFN, os.O_WRONLY | os.O_APPEND)
+        w = asyncore.file_wrapper(fd)
+        os.close(fd)
+
+        w.write(d1)
+        w.send(d2)
+        w.close()
+        self.assertEqual(open(TESTFN, 'rb').read(), self.d + d1 + d2)
+
+    @unittest.skipUnless(hasattr(asyncore, 'file_dispatcher'),
+                         'asyncore.file_dispatcher required')
+    def test_dispatcher(self):
+        fd = os.open(TESTFN, os.O_RDONLY)
+        data = []
+        class FileDispatcher(asyncore.file_dispatcher):
+            def handle_read(self):
+                data.append(self.recv(29))
+        s = FileDispatcher(fd)
+        os.close(fd)
+        asyncore.loop(timeout=0.01, use_poll=True, count=2)
+        self.assertEqual(b"".join(data), self.d)
 
 
 class BaseTestHandler(asyncore.dispatcher):
@@ -692,18 +706,15 @@
 class TestAPI_UseSelect(BaseTestAPI):
     use_poll = False
 
+ at unittest.skipUnless(hasattr(select, 'poll'), 'select.poll required')
 class TestAPI_UsePoll(BaseTestAPI):
     use_poll = True
 
 
 def test_main():
     tests = [HelperFunctionTests, DispatcherTests, DispatcherWithSendTests,
-             DispatcherWithSendTests_UsePoll, TestAPI_UseSelect]
-    if hasattr(asyncore, 'file_wrapper'):
-        tests.append(FileWrapperTest)
-    if hasattr(select, 'poll'):
-        tests.append(TestAPI_UsePoll)
-
+             DispatcherWithSendTests_UsePoll, TestAPI_UseSelect,
+             TestAPI_UsePoll, FileWrapperTest]
     run_unittest(*tests)
 
 if __name__ == "__main__":

Modified: python/branches/import_unicode/Lib/test/test_binascii.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_binascii.py	(original)
+++ python/branches/import_unicode/Lib/test/test_binascii.py	Fri Jul 30 01:19:37 2010
@@ -55,7 +55,7 @@
                              "{!r} != {!r}".format(fb, fa, res, raw))
             self.assertIsInstance(res, bytes)
             self.assertIsInstance(a, bytes)
-            self.assertLess(max(c for c in a), 128)
+            self.assertLess(max(a), 128)
         self.assertIsInstance(binascii.crc_hqx(raw, 0), int)
         self.assertIsInstance(binascii.crc32(raw), int)
 
@@ -167,7 +167,7 @@
     def test_qp(self):
         # A test for SF bug 534347 (segfaults without the proper fix)
         try:
-            binascii.a2b_qp("", **{1:1})
+            binascii.a2b_qp(b"", **{1:1})
         except TypeError:
             pass
         else:
@@ -179,12 +179,10 @@
         self.assertEqual(binascii.a2b_qp(b"=00\r\n=00"), b"\x00\r\n\x00")
         self.assertEqual(
             binascii.b2a_qp(b"\xff\r\n\xff\n\xff"),
-            b"=FF\r\n=FF\r\n=FF"
-        )
+            b"=FF\r\n=FF\r\n=FF")
         self.assertEqual(
             binascii.b2a_qp(b"0"*75+b"\xff\r\n\xff\r\n\xff"),
-            b"0"*75+b"=\r\n=FF\r\n=FF\r\n=FF"
-        )
+            b"0"*75+b"=\r\n=FF\r\n=FF\r\n=FF")
 
         self.assertEqual(binascii.b2a_qp(b'\0\n'), b'=00\n')
         self.assertEqual(binascii.b2a_qp(b'\0\n', quotetabs=True), b'=00\n')
@@ -210,13 +208,15 @@
             except Exception as err:
                 self.fail("{}({!r}) raises {!r}".format(func, empty, err))
 
-    def test_no_binary_strings(self):
-        # b2a_ must not accept strings
-        for f in (binascii.b2a_uu, binascii.b2a_base64,
-                  binascii.b2a_hqx, binascii.b2a_qp,
-                  binascii.hexlify, binascii.rlecode_hqx,
-                  binascii.crc_hqx, binascii.crc32):
-            self.assertRaises(TypeError, f, "test")
+    def test_unicode_strings(self):
+        # Unicode strings are not accepted.
+        for func in all_functions:
+            try:
+                self.assertRaises(TypeError, getattr(binascii, func), "test")
+            except Exception as err:
+                self.fail('{}("test") raises {!r}'.format(func, err))
+        # crc_hqx needs 2 arguments
+        self.assertRaises(TypeError, binascii.crc_hqx, "test", 0)
 
 
 class ArrayBinASCIITest(BinASCIITest):

Modified: python/branches/import_unicode/Lib/test/test_binop.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_binop.py	(original)
+++ python/branches/import_unicode/Lib/test/test_binop.py	Fri Jul 30 01:19:37 2010
@@ -16,7 +16,7 @@
 
 def isnum(x):
     """Test whether an object is an instance of a built-in numeric type."""
-    for T in int, int, float, complex:
+    for T in int, float, complex:
         if isinstance(x, T):
             return 1
     return 0

Modified: python/branches/import_unicode/Lib/test/test_calendar.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_calendar.py	(original)
+++ python/branches/import_unicode/Lib/test/test_calendar.py	Fri Jul 30 01:19:37 2010
@@ -389,6 +389,34 @@
             tuple = time.gmtime(secs)
             self.assertEqual(secs, calendar.timegm(tuple))
 
+class MonthRangeTestCase(unittest.TestCase):
+    def test_january(self):
+        # Tests valid lower boundary case.
+        self.assertEqual(calendar.monthrange(2004,1), (3,31))
+
+    def test_february_leap(self):
+        # Tests February during leap year.
+        self.assertEqual(calendar.monthrange(2004,2), (6,29))
+
+    def test_february_nonleap(self):
+        # Tests February in non-leap year.
+        self.assertEqual(calendar.monthrange(2010,2), (0,28))
+
+    def test_december(self):
+        # Tests valid upper boundary case.
+        self.assertEqual(calendar.monthrange(2004,12), (2,31))
+
+    def test_zeroth_month(self):
+        # Tests low invalid boundary case.
+        with self.assertRaises(calendar.IllegalMonthError):
+            calendar.monthrange(2004, 0)
+
+    def test_thirteenth_month(self):
+        # Tests high invalid boundary case.
+        with self.assertRaises(calendar.IllegalMonthError):
+            calendar.monthrange(2004, 13)
+
+
 def test_main():
     support.run_unittest(
         OutputTestCase,
@@ -396,6 +424,7 @@
         MondayTestCase,
         SundayTestCase,
         TimegmTestCase,
+        MonthRangeTestCase,
     )
 
 

Modified: python/branches/import_unicode/Lib/test/test_cfgparser.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_cfgparser.py	(original)
+++ python/branches/import_unicode/Lib/test/test_cfgparser.py	Fri Jul 30 01:19:37 2010
@@ -1,7 +1,9 @@
+import collections
 import configparser
 import io
+import os
 import unittest
-import collections
+import textwrap
 
 from test import support
 
@@ -22,15 +24,26 @@
     def itervalues(self): return iter(self.values())
 
 
-class TestCaseBase(unittest.TestCase):
+class CfgParserTestCaseClass(unittest.TestCase):
     allow_no_value = False
+    delimiters = ('=', ':')
+    comment_prefixes = (';', '#')
+    empty_lines_in_values = True
+    dict_type = configparser._default_dict
 
     def newconfig(self, defaults=None):
+        arguments = dict(
+            allow_no_value=self.allow_no_value,
+            delimiters=self.delimiters,
+            comment_prefixes=self.comment_prefixes,
+            empty_lines_in_values=self.empty_lines_in_values,
+            dict_type=self.dict_type,
+        )
         if defaults is None:
-            self.cf = self.config_class(allow_no_value=self.allow_no_value)
+            self.cf = self.config_class(**arguments)
         else:
             self.cf = self.config_class(defaults,
-                                        allow_no_value=self.allow_no_value)
+                                        **arguments)
         return self.cf
 
     def fromstring(self, string, defaults=None):
@@ -39,27 +52,33 @@
         cf.readfp(sio)
         return cf
 
+class BasicTestCase(CfgParserTestCaseClass):
+
     def test_basic(self):
-        config_string = (
-            "[Foo Bar]\n"
-            "foo=bar\n"
-            "[Spacey Bar]\n"
-            "foo = bar\n"
-            "[Commented Bar]\n"
-            "foo: bar ; comment\n"
-            "[Long Line]\n"
-            "foo: this line is much, much longer than my editor\n"
-            "   likes it.\n"
-            "[Section\\with$weird%characters[\t]\n"
-            "[Internationalized Stuff]\n"
-            "foo[bg]: Bulgarian\n"
-            "foo=Default\n"
-            "foo[en]=English\n"
-            "foo[de]=Deutsch\n"
-            "[Spaces]\n"
-            "key with spaces : value\n"
-            "another with spaces = splat!\n"
-            )
+        config_string = """\
+[Foo Bar]
+foo{0[0]}bar
+[Spacey Bar]
+foo {0[0]} bar
+[Spacey Bar From The Beginning]
+  foo {0[0]} bar
+  baz {0[0]} qwe
+[Commented Bar]
+foo{0[1]} bar {1[1]} comment
+baz{0[0]}qwe {1[0]}another one
+[Long Line]
+foo{0[1]} this line is much, much longer than my editor
+   likes it.
+[Section\\with$weird%characters[\t]
+[Internationalized Stuff]
+foo[bg]{0[1]} Bulgarian
+foo{0[0]}Default
+foo[en]{0[0]}English
+foo[de]{0[0]}Deutsch
+[Spaces]
+key with spaces {0[1]} value
+another with spaces {0[0]} splat!
+""".format(self.delimiters, self.comment_prefixes)
         if self.allow_no_value:
             config_string += (
                 "[NoValue]\n"
@@ -69,13 +88,14 @@
         cf = self.fromstring(config_string)
         L = cf.sections()
         L.sort()
-        E = [r'Commented Bar',
-             r'Foo Bar',
-             r'Internationalized Stuff',
-             r'Long Line',
-             r'Section\with$weird%characters[' '\t',
-             r'Spaces',
-             r'Spacey Bar',
+        E = ['Commented Bar',
+             'Foo Bar',
+             'Internationalized Stuff',
+             'Long Line',
+             'Section\\with$weird%characters[\t',
+             'Spaces',
+             'Spacey Bar',
+             'Spacey Bar From The Beginning',
              ]
         if self.allow_no_value:
             E.append(r'NoValue')
@@ -88,7 +108,10 @@
         # http://www.python.org/sf/583248
         eq(cf.get('Foo Bar', 'foo'), 'bar')
         eq(cf.get('Spacey Bar', 'foo'), 'bar')
+        eq(cf.get('Spacey Bar From The Beginning', 'foo'), 'bar')
+        eq(cf.get('Spacey Bar From The Beginning', 'baz'), 'qwe')
         eq(cf.get('Commented Bar', 'foo'), 'bar')
+        eq(cf.get('Commented Bar', 'baz'), 'qwe')
         eq(cf.get('Spaces', 'key with spaces'), 'value')
         eq(cf.get('Spaces', 'another with spaces'), 'splat!')
         if self.allow_no_value:
@@ -107,8 +130,9 @@
                     "remove_option() failed to report non-existence of option"
                     " that was removed")
 
-        self.assertRaises(configparser.NoSectionError,
-                          cf.remove_option, 'No Such Section', 'foo')
+        with self.assertRaises(configparser.NoSectionError) as cm:
+            cf.remove_option('No Such Section', 'foo')
+        self.assertEqual(cm.exception.args, ('No Such Section',))
 
         eq(cf.get('Long Line', 'foo'),
            'this line is much, much longer than my editor\nlikes it.')
@@ -138,12 +162,14 @@
 
         # SF bug #432369:
         cf = self.fromstring(
-            "[MySection]\nOption: first line\n\tsecond line\n")
+            "[MySection]\nOption{} first line\n\tsecond line\n".format(
+                self.delimiters[0]))
         eq(cf.options("MySection"), ["option"])
         eq(cf.get("MySection", "Option"), "first line\nsecond line")
 
         # SF bug #561822:
-        cf = self.fromstring("[section]\nnekey=nevalue\n",
+        cf = self.fromstring("[section]\n"
+                             "nekey{}nevalue\n".format(self.delimiters[0]),
                              defaults={"key":"value"})
         self.assertTrue(cf.has_option("section", "Key"))
 
@@ -161,33 +187,41 @@
     def test_parse_errors(self):
         self.newconfig()
         self.parse_error(configparser.ParsingError,
-                         "[Foo]\n  extra-spaces: splat\n")
+                         "[Foo]\n"
+                         "{}val-without-opt-name\n".format(self.delimiters[0]))
         self.parse_error(configparser.ParsingError,
-                         "[Foo]\n  extra-spaces= splat\n")
-        self.parse_error(configparser.ParsingError,
-                         "[Foo]\n:value-without-option-name\n")
-        self.parse_error(configparser.ParsingError,
-                         "[Foo]\n=value-without-option-name\n")
-        self.parse_error(configparser.MissingSectionHeaderError,
-                         "No Section!\n")
+                         "[Foo]\n"
+                         "{}val-without-opt-name\n".format(self.delimiters[1]))
+        e = self.parse_error(configparser.MissingSectionHeaderError,
+                             "No Section!\n")
+        self.assertEqual(e.args, ('<???>', 1, "No Section!\n"))
+        if not self.allow_no_value:
+            e = self.parse_error(configparser.ParsingError,
+                                "[Foo]\n  wrong-indent\n")
+            self.assertEqual(e.args, ('<???>',))
 
     def parse_error(self, exc, src):
         sio = io.StringIO(src)
-        self.assertRaises(exc, self.cf.readfp, sio)
+        with self.assertRaises(exc) as cm:
+            self.cf.readfp(sio)
+        return cm.exception
 
     def test_query_errors(self):
         cf = self.newconfig()
         self.assertEqual(cf.sections(), [],
                          "new ConfigParser should have no defined sections")
         self.assertFalse(cf.has_section("Foo"),
-                    "new ConfigParser should have no acknowledged sections")
-        self.assertRaises(configparser.NoSectionError,
-                          cf.options, "Foo")
-        self.assertRaises(configparser.NoSectionError,
-                          cf.set, "foo", "bar", "value")
-        self.get_error(configparser.NoSectionError, "foo", "bar")
+                         "new ConfigParser should have no acknowledged "
+                         "sections")
+        with self.assertRaises(configparser.NoSectionError):
+            cf.options("Foo")
+        with self.assertRaises(configparser.NoSectionError):
+            cf.set("foo", "bar", "value")
+        e = self.get_error(configparser.NoSectionError, "foo", "bar")
+        self.assertEqual(e.args, ("foo",))
         cf.add_section("foo")
-        self.get_error(configparser.NoOptionError, "foo", "bar")
+        e = self.get_error(configparser.NoOptionError, "foo", "bar")
+        self.assertEqual(e.args, ("bar", "foo"))
 
     def get_error(self, exc, section, option):
         try:
@@ -201,21 +235,21 @@
     def test_boolean(self):
         cf = self.fromstring(
             "[BOOLTEST]\n"
-            "T1=1\n"
-            "T2=TRUE\n"
-            "T3=True\n"
-            "T4=oN\n"
-            "T5=yes\n"
-            "F1=0\n"
-            "F2=FALSE\n"
-            "F3=False\n"
-            "F4=oFF\n"
-            "F5=nO\n"
-            "E1=2\n"
-            "E2=foo\n"
-            "E3=-1\n"
-            "E4=0.1\n"
-            "E5=FALSE AND MORE"
+            "T1{equals}1\n"
+            "T2{equals}TRUE\n"
+            "T3{equals}True\n"
+            "T4{equals}oN\n"
+            "T5{equals}yes\n"
+            "F1{equals}0\n"
+            "F2{equals}FALSE\n"
+            "F3{equals}False\n"
+            "F4{equals}oFF\n"
+            "F5{equals}nO\n"
+            "E1{equals}2\n"
+            "E2{equals}foo\n"
+            "E3{equals}-1\n"
+            "E4{equals}0.1\n"
+            "E5{equals}FALSE AND MORE".format(equals=self.delimiters[0])
             )
         for x in range(1, 5):
             self.assertTrue(cf.getboolean('BOOLTEST', 't%d' % x))
@@ -226,17 +260,24 @@
     def test_weird_errors(self):
         cf = self.newconfig()
         cf.add_section("Foo")
-        self.assertRaises(configparser.DuplicateSectionError,
-                          cf.add_section, "Foo")
+        with self.assertRaises(configparser.DuplicateSectionError) as cm:
+            cf.add_section("Foo")
+        self.assertEqual(cm.exception.args, ("Foo",))
 
     def test_write(self):
         config_string = (
             "[Long Line]\n"
-            "foo: this line is much, much longer than my editor\n"
+            "foo{0[0]} this line is much, much longer than my editor\n"
             "   likes it.\n"
             "[DEFAULT]\n"
-            "foo: another very\n"
+            "foo{0[1]} another very\n"
             " long line\n"
+            "[Long Line - With Comments!]\n"
+            "test {0[1]} we        {comment} can\n"
+            "            also      {comment} place\n"
+            "            comments  {comment} in\n"
+            "            multiline {comment} values"
+            "\n".format(self.delimiters, comment=self.comment_prefixes[0])
             )
         if self.allow_no_value:
             config_string += (
@@ -249,13 +290,19 @@
         cf.write(output)
         expect_string = (
             "[DEFAULT]\n"
-            "foo = another very\n"
+            "foo {equals} another very\n"
             "\tlong line\n"
             "\n"
             "[Long Line]\n"
-            "foo = this line is much, much longer than my editor\n"
+            "foo {equals} this line is much, much longer than my editor\n"
             "\tlikes it.\n"
             "\n"
+            "[Long Line - With Comments!]\n"
+            "test {equals} we\n"
+            "\talso\n"
+            "\tcomments\n"
+            "\tmultiline\n"
+            "\n".format(equals=self.delimiters[0])
             )
         if self.allow_no_value:
             expect_string += (
@@ -267,7 +314,7 @@
 
     def test_set_string_types(self):
         cf = self.fromstring("[sect]\n"
-                             "option1=foo\n")
+                             "option1{eq}foo\n".format(eq=self.delimiters[0]))
         # Check that we don't get an exception when setting values in
         # an existing section using strings:
         class mystr(str):
@@ -280,6 +327,9 @@
         cf.set("sect", "option2", "splat")
 
     def test_read_returns_file_list(self):
+        if self.delimiters[0] != '=':
+            # skip reading the file if we're using an incompatible format
+            return
         file1 = support.findfile("cfgparser.1")
         # check when we pass a mix of readable and non-readable files:
         cf = self.newconfig()
@@ -304,48 +354,53 @@
     def get_interpolation_config(self):
         return self.fromstring(
             "[Foo]\n"
-            "bar=something %(with1)s interpolation (1 step)\n"
-            "bar9=something %(with9)s lots of interpolation (9 steps)\n"
-            "bar10=something %(with10)s lots of interpolation (10 steps)\n"
-            "bar11=something %(with11)s lots of interpolation (11 steps)\n"
-            "with11=%(with10)s\n"
-            "with10=%(with9)s\n"
-            "with9=%(with8)s\n"
-            "with8=%(With7)s\n"
-            "with7=%(WITH6)s\n"
-            "with6=%(with5)s\n"
-            "With5=%(with4)s\n"
-            "WITH4=%(with3)s\n"
-            "with3=%(with2)s\n"
-            "with2=%(with1)s\n"
-            "with1=with\n"
+            "bar{equals}something %(with1)s interpolation (1 step)\n"
+            "bar9{equals}something %(with9)s lots of interpolation (9 steps)\n"
+            "bar10{equals}something %(with10)s lots of interpolation (10 steps)\n"
+            "bar11{equals}something %(with11)s lots of interpolation (11 steps)\n"
+            "with11{equals}%(with10)s\n"
+            "with10{equals}%(with9)s\n"
+            "with9{equals}%(with8)s\n"
+            "with8{equals}%(With7)s\n"
+            "with7{equals}%(WITH6)s\n"
+            "with6{equals}%(with5)s\n"
+            "With5{equals}%(with4)s\n"
+            "WITH4{equals}%(with3)s\n"
+            "with3{equals}%(with2)s\n"
+            "with2{equals}%(with1)s\n"
+            "with1{equals}with\n"
             "\n"
             "[Mutual Recursion]\n"
-            "foo=%(bar)s\n"
-            "bar=%(foo)s\n"
+            "foo{equals}%(bar)s\n"
+            "bar{equals}%(foo)s\n"
             "\n"
             "[Interpolation Error]\n"
-            "name=%(reference)s\n",
+            "name{equals}%(reference)s\n".format(equals=self.delimiters[0]),
             # no definition for 'reference'
             defaults={"getname": "%(__name__)s"})
 
     def check_items_config(self, expected):
         cf = self.fromstring(
             "[section]\n"
-            "name = value\n"
-            "key: |%(name)s| \n"
-            "getdefault: |%(default)s|\n"
-            "getname: |%(__name__)s|",
+            "name {0[0]} value\n"
+            "key{0[1]} |%(name)s| \n"
+            "getdefault{0[1]} |%(default)s|\n"
+            "getname{0[1]} |%(__name__)s|".format(self.delimiters),
             defaults={"default": "<default>"})
         L = list(cf.items("section"))
         L.sort()
         self.assertEqual(L, expected)
 
 
-class ConfigParserTestCase(TestCaseBase):
+class ConfigParserTestCase(BasicTestCase):
     config_class = configparser.ConfigParser
 
     def test_interpolation(self):
+        rawval = {
+            configparser.ConfigParser: ("something %(with11)s "
+                                        "lots of interpolation (11 steps)"),
+            configparser.SafeConfigParser: "%(with1)s",
+        }
         cf = self.get_interpolation_config()
         eq = self.assertEqual
         eq(cf.get("Foo", "getname"), "Foo")
@@ -354,15 +409,22 @@
            "something with lots of interpolation (9 steps)")
         eq(cf.get("Foo", "bar10"),
            "something with lots of interpolation (10 steps)")
-        self.get_error(configparser.InterpolationDepthError, "Foo", "bar11")
+        e = self.get_error(configparser.InterpolationDepthError, "Foo", "bar11")
+        self.assertEqual(e.args, ("bar11", "Foo", rawval[self.config_class]))
 
     def test_interpolation_missing_value(self):
+        rawval = {
+            configparser.ConfigParser: '%(reference)s',
+            configparser.SafeConfigParser: '',
+        }
         self.get_interpolation_config()
-        e = self.get_error(configparser.InterpolationError,
+        e = self.get_error(configparser.InterpolationMissingOptionError,
                            "Interpolation Error", "name")
         self.assertEqual(e.reference, "reference")
         self.assertEqual(e.section, "Interpolation Error")
         self.assertEqual(e.option, "name")
+        self.assertEqual(e.args, ('name', 'Interpolation Error',
+                                  rawval[self.config_class], 'reference'))
 
     def test_items(self):
         self.check_items_config([('default', '<default>'),
@@ -392,8 +454,39 @@
         self.assertRaises(ValueError, cf.get, 'non-string',
                           'string_with_interpolation', raw=False)
 
+class ConfigParserTestCaseNonStandardDelimiters(ConfigParserTestCase):
+    delimiters = (':=', '$')
+    comment_prefixes = ('//', '"')
+
+class MultilineValuesTestCase(BasicTestCase):
+    config_class = configparser.ConfigParser
+    wonderful_spam = ("I'm having spam spam spam spam "
+                      "spam spam spam beaked beans spam "
+                      "spam spam and spam!").replace(' ', '\t\n')
+
+    def setUp(self):
+        cf = self.newconfig()
+        for i in range(100):
+            s = 'section{}'.format(i)
+            cf.add_section(s)
+            for j in range(10):
+                cf.set(s, 'lovely_spam{}'.format(j), self.wonderful_spam)
+        with open(support.TESTFN, 'w') as f:
+            cf.write(f)
+
+    def tearDown(self):
+        os.unlink(support.TESTFN)
+
+    def test_dominating_multiline_values(self):
+        # We're reading from file because this is where the code changed
+        # during performance updates in Python 3.2
+        cf_from_file = self.newconfig()
+        with open(support.TESTFN) as f:
+            cf_from_file.readfp(f)
+        self.assertEqual(cf_from_file.get('section8', 'lovely_spam4'),
+                         self.wonderful_spam.replace('\t\n', '\n'))
 
-class RawConfigParserTestCase(TestCaseBase):
+class RawConfigParserTestCase(BasicTestCase):
     config_class = configparser.RawConfigParser
 
     def test_interpolation(self):
@@ -427,6 +520,28 @@
                          [0, 1, 1, 2, 3, 5, 8, 13])
         self.assertEqual(cf.get('non-string', 'dict'), {'pi': 3.14159})
 
+class RawConfigParserTestCaseNonStandardDelimiters(RawConfigParserTestCase):
+    delimiters = (':=', '$')
+    comment_prefixes = ('//', '"')
+
+class RawConfigParserTestSambaConf(BasicTestCase):
+    config_class = configparser.RawConfigParser
+    comment_prefixes = ('#', ';', '//', '----')
+    empty_lines_in_values = False
+
+    def test_reading(self):
+        smbconf = support.findfile("cfgparser.2")
+        # check when we pass a mix of readable and non-readable files:
+        cf = self.newconfig()
+        parsed_files = cf.read([smbconf, "nonexistent-file"], encoding='utf-8')
+        self.assertEqual(parsed_files, [smbconf])
+        sections = ['global', 'homes', 'printers',
+                    'print$', 'pdf-generator', 'tmp', 'Agustin']
+        self.assertEqual(cf.sections(), sections)
+        self.assertEqual(cf.get("global", "workgroup"), "MDKGROUP")
+        self.assertEqual(cf.getint("global", "max log size"), 50)
+        self.assertEqual(cf.get("global", "hosts allow"), "127.")
+        self.assertEqual(cf.get("tmp", "echo command"), "cat %s; rm %s")
 
 class SafeConfigParserTestCase(ConfigParserTestCase):
     config_class = configparser.SafeConfigParser
@@ -434,16 +549,17 @@
     def test_safe_interpolation(self):
         # See http://www.python.org/sf/511737
         cf = self.fromstring("[section]\n"
-                             "option1=xxx\n"
-                             "option2=%(option1)s/xxx\n"
-                             "ok=%(option1)s/%%s\n"
-                             "not_ok=%(option2)s/%%s")
+                             "option1{eq}xxx\n"
+                             "option2{eq}%(option1)s/xxx\n"
+                             "ok{eq}%(option1)s/%%s\n"
+                             "not_ok{eq}%(option2)s/%%s".format(
+                                 eq=self.delimiters[0]))
         self.assertEqual(cf.get("section", "ok"), "xxx/%s")
         self.assertEqual(cf.get("section", "not_ok"), "xxx/xxx/%s")
 
     def test_set_malformatted_interpolation(self):
         cf = self.fromstring("[sect]\n"
-                             "option1=foo\n")
+                             "option1{eq}foo\n".format(eq=self.delimiters[0]))
 
         self.assertEqual(cf.get('sect', "option1"), "foo")
 
@@ -459,7 +575,7 @@
 
     def test_set_nonstring_types(self):
         cf = self.fromstring("[sect]\n"
-                             "option1=foo\n")
+                             "option1{eq}foo\n".format(eq=self.delimiters[0]))
         # Check that we get a TypeError when setting non-string values
         # in an existing section:
         self.assertRaises(TypeError, cf.set, "sect", "option1", 1)
@@ -477,15 +593,56 @@
         cf = self.newconfig()
         self.assertRaises(ValueError, cf.add_section, "DEFAULT")
 
+class SafeConfigParserTestCaseNonStandardDelimiters(SafeConfigParserTestCase):
+    delimiters = (':=', '$')
+    comment_prefixes = ('//', '"')
 
 class SafeConfigParserTestCaseNoValue(SafeConfigParserTestCase):
     allow_no_value = True
 
+class SafeConfigParserTestCaseTrickyFile(CfgParserTestCaseClass):
+    config_class = configparser.SafeConfigParser
+    delimiters = {'='}
+    comment_prefixes = {'#'}
+    allow_no_value = True
+
+    def test_cfgparser_dot_3(self):
+        tricky = support.findfile("cfgparser.3")
+        cf = self.newconfig()
+        self.assertEqual(len(cf.read(tricky, encoding='utf-8')), 1)
+        self.assertEqual(cf.sections(), ['strange',
+                                         'corruption',
+                                         'yeah, sections can be '
+                                         'indented as well',
+                                         'another one!',
+                                         'no values here',
+                                         'tricky interpolation',
+                                         'more interpolation'])
+        #self.assertEqual(cf.getint('DEFAULT', 'go', vars={'interpolate': '-1'}),
+        #                 -1)
+        self.assertEqual(len(cf.get('strange', 'other').split('\n')), 4)
+        self.assertEqual(len(cf.get('corruption', 'value').split('\n')), 10)
+        longname = 'yeah, sections can be indented as well'
+        self.assertFalse(cf.getboolean(longname, 'are they subsections'))
+        self.assertEquals(cf.get(longname, 'lets use some Unicode'),
+                                           '???')
+        self.assertEqual(len(cf.items('another one!')), 5) # 4 in section and
+                                                           # `go` from DEFAULT
+        with self.assertRaises(configparser.InterpolationMissingOptionError):
+            cf.items('no values here')
+        self.assertEqual(cf.get('tricky interpolation', 'lets'), 'do this')
+        self.assertEqual(cf.get('tricky interpolation', 'lets'),
+                         cf.get('tricky interpolation', 'go'))
+        self.assertEqual(cf.get('more interpolation', 'lets'), 'go shopping')
+
+    def test_unicode_failure(self):
+        tricky = support.findfile("cfgparser.3")
+        cf = self.newconfig()
+        with self.assertRaises(UnicodeDecodeError):
+            cf.read(tricky, encoding='ascii')
 
 class SortedTestCase(RawConfigParserTestCase):
-    def newconfig(self, defaults=None):
-        self.cf = self.config_class(defaults=defaults, dict_type=SortedDict)
-        return self.cf
+    dict_type = SortedDict
 
     def test_sorted(self):
         self.fromstring("[b]\n"
@@ -507,13 +664,40 @@
                           "o4 = 1\n\n")
 
 
+class CompatibleTestCase(CfgParserTestCaseClass):
+    config_class = configparser.RawConfigParser
+    comment_prefixes = configparser.RawConfigParser._COMPATIBLE
+
+    def test_comment_handling(self):
+        config_string = textwrap.dedent("""\
+        [Commented Bar]
+        baz=qwe ; a comment
+        foo: bar # not a comment!
+        # but this is a comment
+        ; another comment
+        quirk: this;is not a comment
+        ; a space must precede a comment character
+        """)
+        cf = self.fromstring(config_string)
+        self.assertEqual(cf.get('Commented Bar', 'foo'), 'bar # not a comment!')
+        self.assertEqual(cf.get('Commented Bar', 'baz'), 'qwe')
+        self.assertEqual(cf.get('Commented Bar', 'quirk'), 'this;is not a comment')
+
+
 def test_main():
     support.run_unittest(
         ConfigParserTestCase,
+        ConfigParserTestCaseNonStandardDelimiters,
+        MultilineValuesTestCase,
         RawConfigParserTestCase,
+        RawConfigParserTestCaseNonStandardDelimiters,
+        RawConfigParserTestSambaConf,
         SafeConfigParserTestCase,
-        SortedTestCase,
+        SafeConfigParserTestCaseNonStandardDelimiters,
         SafeConfigParserTestCaseNoValue,
+        SafeConfigParserTestCaseTrickyFile,
+        SortedTestCase,
+        CompatibleTestCase,
         )
 
 

Modified: python/branches/import_unicode/Lib/test/test_cmath.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_cmath.py	(original)
+++ python/branches/import_unicode/Lib/test/test_cmath.py	Fri Jul 30 01:19:37 2010
@@ -442,6 +442,15 @@
         self.assertCEqual(rect(1, pi/2), (0, 1.))
         self.assertCEqual(rect(1, -pi/2), (0, -1.))
 
+    def test_isfinite(self):
+        real_vals = [float('-inf'), -2.3, -0.0,
+                     0.0, 2.3, float('inf'), float('nan')]
+        for x in real_vals:
+            for y in real_vals:
+                z = complex(x, y)
+                self.assertEqual(cmath.isfinite(z),
+                                  math.isfinite(x) and math.isfinite(y))
+
     def test_isnan(self):
         self.assertFalse(cmath.isnan(1))
         self.assertFalse(cmath.isnan(1j))

Modified: python/branches/import_unicode/Lib/test/test_collections.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_collections.py	(original)
+++ python/branches/import_unicode/Lib/test/test_collections.py	Fri Jul 30 01:19:37 2010
@@ -758,6 +758,19 @@
         od.update([('a', 1), ('b', 2), ('c', 9), ('d', 4)], c=3, e=5)
         self.assertEqual(list(od.items()), pairs)                                   # mixed input
 
+        # Issue 9137: Named argument called 'other' or 'self'
+        # shouldn't be treated specially.
+        od = OrderedDict()
+        od.update(self=23)
+        self.assertEqual(list(od.items()), [('self', 23)])
+        od = OrderedDict()
+        od.update(other={})
+        self.assertEqual(list(od.items()), [('other', {})])
+        od = OrderedDict()
+        od.update(red=5, blue=6, other=7, self=8)
+        self.assertEqual(sorted(list(od.items())),
+                         [('blue', 6), ('other', 7), ('red', 5), ('self', 8)])
+
         # Make sure that direct calls to update do not clear previous contents
         # add that updates items are not moved to the end
         d = OrderedDict([('a', 1), ('b', 2), ('c', 3), ('d', 44), ('e', 55)])

Modified: python/branches/import_unicode/Lib/test/test_datetime.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_datetime.py	(original)
+++ python/branches/import_unicode/Lib/test/test_datetime.py	Fri Jul 30 01:19:37 2010
@@ -1,3668 +1,42 @@
-"""Test date/time type.
-
-See http://www.zope.org/Members/fdrake/DateTimeWiki/TestCases
-"""
-
-import sys
-import pickle
 import unittest
-
-from operator import lt, le, gt, ge, eq, ne, truediv, floordiv, mod
-
-from test import support
-
-from datetime import MINYEAR, MAXYEAR
-from datetime import timedelta
-from datetime import tzinfo
-from datetime import time
-from datetime import timezone
-from datetime import date, datetime
-import time as _time
-
-pickle_choices = [(pickle, pickle, proto)
-                  for proto in range(pickle.HIGHEST_PROTOCOL + 1)]
-assert len(pickle_choices) == pickle.HIGHEST_PROTOCOL + 1
-
-# An arbitrary collection of objects of non-datetime types, for testing
-# mixed-type comparisons.
-OTHERSTUFF = (10, 34.5, "abc", {}, [], ())
-
-
-# XXX Copied from test_float.
-INF = float("inf")
-NAN = float("nan")
-
-# decorator for skipping tests on non-IEEE 754 platforms
-requires_IEEE_754 = unittest.skipUnless(
-    float.__getformat__("double").startswith("IEEE"),
-    "test requires IEEE 754 doubles")
-
-
-#############################################################################
-# module tests
-
-class TestModule(unittest.TestCase):
-
-    def test_constants(self):
-        import datetime
-        self.assertEqual(datetime.MINYEAR, 1)
-        self.assertEqual(datetime.MAXYEAR, 9999)
-
-#############################################################################
-# tzinfo tests
-
-class FixedOffset(tzinfo):
-
-    def __init__(self, offset, name, dstoffset=42):
-        if isinstance(offset, int):
-            offset = timedelta(minutes=offset)
-        if isinstance(dstoffset, int):
-            dstoffset = timedelta(minutes=dstoffset)
-        self.__offset = offset
-        self.__name = name
-        self.__dstoffset = dstoffset
-    def __repr__(self):
-        return self.__name.lower()
-    def utcoffset(self, dt):
-        return self.__offset
-    def tzname(self, dt):
-        return self.__name
-    def dst(self, dt):
-        return self.__dstoffset
-
-class PicklableFixedOffset(FixedOffset):
-
-    def __init__(self, offset=None, name=None, dstoffset=None):
-        FixedOffset.__init__(self, offset, name, dstoffset)
-
-class TestTZInfo(unittest.TestCase):
-
-    def test_non_abstractness(self):
-        # In order to allow subclasses to get pickled, the C implementation
-        # wasn't able to get away with having __init__ raise
-        # NotImplementedError.
-        useless = tzinfo()
-        dt = datetime.max
-        self.assertRaises(NotImplementedError, useless.tzname, dt)
-        self.assertRaises(NotImplementedError, useless.utcoffset, dt)
-        self.assertRaises(NotImplementedError, useless.dst, dt)
-
-    def test_subclass_must_override(self):
-        class NotEnough(tzinfo):
-            def __init__(self, offset, name):
-                self.__offset = offset
-                self.__name = name
-        self.assertTrue(issubclass(NotEnough, tzinfo))
-        ne = NotEnough(3, "NotByALongShot")
-        self.assertIsInstance(ne, tzinfo)
-
-        dt = datetime.now()
-        self.assertRaises(NotImplementedError, ne.tzname, dt)
-        self.assertRaises(NotImplementedError, ne.utcoffset, dt)
-        self.assertRaises(NotImplementedError, ne.dst, dt)
-
-    def test_normal(self):
-        fo = FixedOffset(3, "Three")
-        self.assertIsInstance(fo, tzinfo)
-        for dt in datetime.now(), None:
-            self.assertEqual(fo.utcoffset(dt), timedelta(minutes=3))
-            self.assertEqual(fo.tzname(dt), "Three")
-            self.assertEqual(fo.dst(dt), timedelta(minutes=42))
-
-    def test_pickling_base(self):
-        # There's no point to pickling tzinfo objects on their own (they
-        # carry no data), but they need to be picklable anyway else
-        # concrete subclasses can't be pickled.
-        orig = tzinfo.__new__(tzinfo)
-        self.assertTrue(type(orig) is tzinfo)
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertTrue(type(derived) is tzinfo)
-
-    def test_pickling_subclass(self):
-        # Make sure we can pickle/unpickle an instance of a subclass.
-        offset = timedelta(minutes=-300)
-        for otype, args in [
-            (PicklableFixedOffset, (offset, 'cookie')),
-            (timezone, (offset,)),
-            (timezone, (offset, "EST"))]:
-            orig = otype(*args)
-            oname = orig.tzname(None)
-            self.assertIsInstance(orig, tzinfo)
-            self.assertIs(type(orig), otype)
-            self.assertEqual(orig.utcoffset(None), offset)
-            self.assertEqual(orig.tzname(None), oname)
-            for pickler, unpickler, proto in pickle_choices:
-                green = pickler.dumps(orig, proto)
-                derived = unpickler.loads(green)
-                self.assertIsInstance(derived, tzinfo)
-                self.assertIs(type(derived), otype)
-                self.assertEqual(derived.utcoffset(None), offset)
-                self.assertEqual(derived.tzname(None), oname)
-
-class TestTimeZone(unittest.TestCase):
-
-    def setUp(self):
-        self.ACDT = timezone(timedelta(hours=9.5), 'ACDT')
-        self.EST = timezone(-timedelta(hours=5), 'EST')
-        self.DT = datetime(2010, 1, 1)
-
-    def test_str(self):
-        for tz in [self.ACDT, self.EST, timezone.utc,
-                   timezone.min, timezone.max]:
-            self.assertEqual(str(tz), tz.tzname(None))
-
-    def test_repr(self):
-        import datetime
-        for tz in [self.ACDT, self.EST, timezone.utc,
-                   timezone.min, timezone.max]:
-            # test round-trip
-            tzrep = repr(tz)
-            self.assertEqual(tz, eval(tzrep))
-
-
-    def test_class_members(self):
-        limit = timedelta(hours=23, minutes=59)
-        self.assertEqual(timezone.utc.utcoffset(None), ZERO)
-        self.assertEqual(timezone.min.utcoffset(None), -limit)
-        self.assertEqual(timezone.max.utcoffset(None), limit)
-
-
-    def test_constructor(self):
-        self.assertEqual(timezone.utc, timezone(timedelta(0)))
-        # invalid offsets
-        for invalid in [timedelta(microseconds=1), timedelta(1, 1),
-                        timedelta(seconds=1), timedelta(1), -timedelta(1)]:
-            self.assertRaises(ValueError, timezone, invalid)
-            self.assertRaises(ValueError, timezone, -invalid)
-
-        with self.assertRaises(TypeError): timezone(None)
-        with self.assertRaises(TypeError): timezone(42)
-        with self.assertRaises(TypeError): timezone(ZERO, None)
-        with self.assertRaises(TypeError): timezone(ZERO, 42)
-        with self.assertRaises(TypeError): timezone(ZERO, 'ABC', 'extra')
-
-    def test_inheritance(self):
-        self.assertIsInstance(timezone.utc, tzinfo)
-        self.assertIsInstance(self.EST, tzinfo)
-
-    def test_utcoffset(self):
-        dummy = self.DT
-        for h in [0, 1.5, 12]:
-            offset = h * HOUR
-            self.assertEqual(offset, timezone(offset).utcoffset(dummy))
-            self.assertEqual(-offset, timezone(-offset).utcoffset(dummy))
-
-        with self.assertRaises(TypeError): self.EST.utcoffset('')
-        with self.assertRaises(TypeError): self.EST.utcoffset(5)
-
-
-    def test_dst(self):
-        self.assertEqual(None, timezone.utc.dst(self.DT))
-
-        with self.assertRaises(TypeError): self.EST.dst('')
-        with self.assertRaises(TypeError): self.EST.dst(5)
-
-    def test_tzname(self):
-        self.assertEqual('UTC+00:00', timezone(ZERO).tzname(None))
-        self.assertEqual('UTC-05:00', timezone(-5 * HOUR).tzname(None))
-        self.assertEqual('UTC+09:30', timezone(9.5 * HOUR).tzname(None))
-        self.assertEqual('UTC-00:01', timezone(timedelta(minutes=-1)).tzname(None))
-        self.assertEqual('XYZ', timezone(-5 * HOUR, 'XYZ').tzname(None))
-
-        with self.assertRaises(TypeError): self.EST.tzname('')
-        with self.assertRaises(TypeError): self.EST.tzname(5)
-
-    def test_fromutc(self):
-        with self.assertRaises(ValueError):
-            timezone.utc.fromutc(self.DT)
-        with self.assertRaises(TypeError):
-            timezone.utc.fromutc('not datetime')
-        for tz in [self.EST, self.ACDT, Eastern]:
-            utctime = self.DT.replace(tzinfo=tz)
-            local = tz.fromutc(utctime)
-            self.assertEqual(local - utctime, tz.utcoffset(local))
-            self.assertEqual(local,
-                             self.DT.replace(tzinfo=timezone.utc))
-
-    def test_comparison(self):
-        self.assertNotEqual(timezone(ZERO), timezone(HOUR))
-        self.assertEqual(timezone(HOUR), timezone(HOUR))
-        self.assertEqual(timezone(-5 * HOUR), timezone(-5 * HOUR, 'EST'))
-        with self.assertRaises(TypeError): timezone(ZERO) < timezone(ZERO)
-        self.assertIn(timezone(ZERO), {timezone(ZERO)})
-
-    def test_aware_datetime(self):
-        # test that timezone instances can be used by datetime
-        t = datetime(1, 1, 1)
-        for tz in [timezone.min, timezone.max, timezone.utc]:
-            self.assertEqual(tz.tzname(t),
-                             t.replace(tzinfo=tz).tzname())
-            self.assertEqual(tz.utcoffset(t),
-                             t.replace(tzinfo=tz).utcoffset())
-            self.assertEqual(tz.dst(t),
-                             t.replace(tzinfo=tz).dst())
-
-#############################################################################
-# Base clase for testing a particular aspect of timedelta, time, date and
-# datetime comparisons.
-
-class HarmlessMixedComparison:
-    # Test that __eq__ and __ne__ don't complain for mixed-type comparisons.
-
-    # Subclasses must define 'theclass', and theclass(1, 1, 1) must be a
-    # legit constructor.
-
-    def test_harmless_mixed_comparison(self):
-        me = self.theclass(1, 1, 1)
-
-        self.assertFalse(me == ())
-        self.assertTrue(me != ())
-        self.assertFalse(() == me)
-        self.assertTrue(() != me)
-
-        self.assertIn(me, [1, 20, [], me])
-        self.assertIn([], [me, 1, 20, []])
-
-    def test_harmful_mixed_comparison(self):
-        me = self.theclass(1, 1, 1)
-
-        self.assertRaises(TypeError, lambda: me < ())
-        self.assertRaises(TypeError, lambda: me <= ())
-        self.assertRaises(TypeError, lambda: me > ())
-        self.assertRaises(TypeError, lambda: me >= ())
-
-        self.assertRaises(TypeError, lambda: () < me)
-        self.assertRaises(TypeError, lambda: () <= me)
-        self.assertRaises(TypeError, lambda: () > me)
-        self.assertRaises(TypeError, lambda: () >= me)
-
-#############################################################################
-# timedelta tests
-
-class TestTimeDelta(HarmlessMixedComparison, unittest.TestCase):
-
-    theclass = timedelta
-
-    def test_constructor(self):
-        eq = self.assertEqual
-        td = timedelta
-
-        # Check keyword args to constructor
-        eq(td(), td(weeks=0, days=0, hours=0, minutes=0, seconds=0,
-                    milliseconds=0, microseconds=0))
-        eq(td(1), td(days=1))
-        eq(td(0, 1), td(seconds=1))
-        eq(td(0, 0, 1), td(microseconds=1))
-        eq(td(weeks=1), td(days=7))
-        eq(td(days=1), td(hours=24))
-        eq(td(hours=1), td(minutes=60))
-        eq(td(minutes=1), td(seconds=60))
-        eq(td(seconds=1), td(milliseconds=1000))
-        eq(td(milliseconds=1), td(microseconds=1000))
-
-        # Check float args to constructor
-        eq(td(weeks=1.0/7), td(days=1))
-        eq(td(days=1.0/24), td(hours=1))
-        eq(td(hours=1.0/60), td(minutes=1))
-        eq(td(minutes=1.0/60), td(seconds=1))
-        eq(td(seconds=0.001), td(milliseconds=1))
-        eq(td(milliseconds=0.001), td(microseconds=1))
-
-    def test_computations(self):
-        eq = self.assertEqual
-        td = timedelta
-
-        a = td(7) # One week
-        b = td(0, 60) # One minute
-        c = td(0, 0, 1000) # One millisecond
-        eq(a+b+c, td(7, 60, 1000))
-        eq(a-b, td(6, 24*3600 - 60))
-        eq(b.__rsub__(a), td(6, 24*3600 - 60))
-        eq(-a, td(-7))
-        eq(+a, td(7))
-        eq(-b, td(-1, 24*3600 - 60))
-        eq(-c, td(-1, 24*3600 - 1, 999000))
-        eq(abs(a), a)
-        eq(abs(-a), a)
-        eq(td(6, 24*3600), a)
-        eq(td(0, 0, 60*1000000), b)
-        eq(a*10, td(70))
-        eq(a*10, 10*a)
-        eq(a*10, 10*a)
-        eq(b*10, td(0, 600))
-        eq(10*b, td(0, 600))
-        eq(b*10, td(0, 600))
-        eq(c*10, td(0, 0, 10000))
-        eq(10*c, td(0, 0, 10000))
-        eq(c*10, td(0, 0, 10000))
-        eq(a*-1, -a)
-        eq(b*-2, -b-b)
-        eq(c*-2, -c+-c)
-        eq(b*(60*24), (b*60)*24)
-        eq(b*(60*24), (60*b)*24)
-        eq(c*1000, td(0, 1))
-        eq(1000*c, td(0, 1))
-        eq(a//7, td(1))
-        eq(b//10, td(0, 6))
-        eq(c//1000, td(0, 0, 1))
-        eq(a//10, td(0, 7*24*360))
-        eq(a//3600000, td(0, 0, 7*24*1000))
-        eq(a/0.5, td(14))
-        eq(b/0.5, td(0, 120))
-        eq(a/7, td(1))
-        eq(b/10, td(0, 6))
-        eq(c/1000, td(0, 0, 1))
-        eq(a/10, td(0, 7*24*360))
-        eq(a/3600000, td(0, 0, 7*24*1000))
-
-        # Multiplication by float
-        us = td(microseconds=1)
-        eq((3*us) * 0.5, 2*us)
-        eq((5*us) * 0.5, 2*us)
-        eq(0.5 * (3*us), 2*us)
-        eq(0.5 * (5*us), 2*us)
-        eq((-3*us) * 0.5, -2*us)
-        eq((-5*us) * 0.5, -2*us)
-
-        # Division by int and float
-        eq((3*us) / 2, 2*us)
-        eq((5*us) / 2, 2*us)
-        eq((-3*us) / 2.0, -2*us)
-        eq((-5*us) / 2.0, -2*us)
-        eq((3*us) / -2, -2*us)
-        eq((5*us) / -2, -2*us)
-        eq((3*us) / -2.0, -2*us)
-        eq((5*us) / -2.0, -2*us)
-        for i in range(-10, 10):
-            eq((i*us/3)//us, round(i/3))
-        for i in range(-10, 10):
-            eq((i*us/-3)//us, round(i/-3))
-
-    def test_disallowed_computations(self):
-        a = timedelta(42)
-
-        # Add/sub ints or floats should be illegal
-        for i in 1, 1.0:
-            self.assertRaises(TypeError, lambda: a+i)
-            self.assertRaises(TypeError, lambda: a-i)
-            self.assertRaises(TypeError, lambda: i+a)
-            self.assertRaises(TypeError, lambda: i-a)
-
-        # Division of int by timedelta doesn't make sense.
-        # Division by zero doesn't make sense.
-        zero = 0
-        self.assertRaises(TypeError, lambda: zero // a)
-        self.assertRaises(ZeroDivisionError, lambda: a // zero)
-        self.assertRaises(ZeroDivisionError, lambda: a / zero)
-        self.assertRaises(ZeroDivisionError, lambda: a / 0.0)
-        self.assertRaises(TypeError, lambda: a / '')
-
-    @requires_IEEE_754
-    def test_disallowed_special(self):
-        a = timedelta(42)
-        self.assertRaises(ValueError, a.__mul__, NAN)
-        self.assertRaises(ValueError, a.__truediv__, NAN)
-
-    def test_basic_attributes(self):
-        days, seconds, us = 1, 7, 31
-        td = timedelta(days, seconds, us)
-        self.assertEqual(td.days, days)
-        self.assertEqual(td.seconds, seconds)
-        self.assertEqual(td.microseconds, us)
-
-    def test_total_seconds(self):
-        td = timedelta(days=365)
-        self.assertEqual(td.total_seconds(), 31536000.0)
-        for total_seconds in [123456.789012, -123456.789012, 0.123456, 0, 1e6]:
-            td = timedelta(seconds=total_seconds)
-            self.assertEqual(td.total_seconds(), total_seconds)
-        # Issue8644: Test that td.total_seconds() has the same
-        # accuracy as td / timedelta(seconds=1).
-        for ms in [-1, -2, -123]:
-            td = timedelta(microseconds=ms)
-            self.assertEqual(td.total_seconds(), td / timedelta(seconds=1))
-
-    def test_carries(self):
-        t1 = timedelta(days=100,
-                       weeks=-7,
-                       hours=-24*(100-49),
-                       minutes=-3,
-                       seconds=12,
-                       microseconds=(3*60 - 12) * 1e6 + 1)
-        t2 = timedelta(microseconds=1)
-        self.assertEqual(t1, t2)
-
-    def test_hash_equality(self):
-        t1 = timedelta(days=100,
-                       weeks=-7,
-                       hours=-24*(100-49),
-                       minutes=-3,
-                       seconds=12,
-                       microseconds=(3*60 - 12) * 1000000)
-        t2 = timedelta()
-        self.assertEqual(hash(t1), hash(t2))
-
-        t1 += timedelta(weeks=7)
-        t2 += timedelta(days=7*7)
-        self.assertEqual(t1, t2)
-        self.assertEqual(hash(t1), hash(t2))
-
-        d = {t1: 1}
-        d[t2] = 2
-        self.assertEqual(len(d), 1)
-        self.assertEqual(d[t1], 2)
-
-    def test_pickling(self):
-        args = 12, 34, 56
-        orig = timedelta(*args)
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertEqual(orig, derived)
-
-    def test_compare(self):
-        t1 = timedelta(2, 3, 4)
-        t2 = timedelta(2, 3, 4)
-        self.assertEqual(t1, t2)
-        self.assertTrue(t1 <= t2)
-        self.assertTrue(t1 >= t2)
-        self.assertTrue(not t1 != t2)
-        self.assertTrue(not t1 < t2)
-        self.assertTrue(not t1 > t2)
-
-        for args in (3, 3, 3), (2, 4, 4), (2, 3, 5):
-            t2 = timedelta(*args)   # this is larger than t1
-            self.assertTrue(t1 < t2)
-            self.assertTrue(t2 > t1)
-            self.assertTrue(t1 <= t2)
-            self.assertTrue(t2 >= t1)
-            self.assertTrue(t1 != t2)
-            self.assertTrue(t2 != t1)
-            self.assertTrue(not t1 == t2)
-            self.assertTrue(not t2 == t1)
-            self.assertTrue(not t1 > t2)
-            self.assertTrue(not t2 < t1)
-            self.assertTrue(not t1 >= t2)
-            self.assertTrue(not t2 <= t1)
-
-        for badarg in OTHERSTUFF:
-            self.assertEqual(t1 == badarg, False)
-            self.assertEqual(t1 != badarg, True)
-            self.assertEqual(badarg == t1, False)
-            self.assertEqual(badarg != t1, True)
-
-            self.assertRaises(TypeError, lambda: t1 <= badarg)
-            self.assertRaises(TypeError, lambda: t1 < badarg)
-            self.assertRaises(TypeError, lambda: t1 > badarg)
-            self.assertRaises(TypeError, lambda: t1 >= badarg)
-            self.assertRaises(TypeError, lambda: badarg <= t1)
-            self.assertRaises(TypeError, lambda: badarg < t1)
-            self.assertRaises(TypeError, lambda: badarg > t1)
-            self.assertRaises(TypeError, lambda: badarg >= t1)
-
-    def test_str(self):
-        td = timedelta
-        eq = self.assertEqual
-
-        eq(str(td(1)), "1 day, 0:00:00")
-        eq(str(td(-1)), "-1 day, 0:00:00")
-        eq(str(td(2)), "2 days, 0:00:00")
-        eq(str(td(-2)), "-2 days, 0:00:00")
-
-        eq(str(td(hours=12, minutes=58, seconds=59)), "12:58:59")
-        eq(str(td(hours=2, minutes=3, seconds=4)), "2:03:04")
-        eq(str(td(weeks=-30, hours=23, minutes=12, seconds=34)),
-           "-210 days, 23:12:34")
-
-        eq(str(td(milliseconds=1)), "0:00:00.001000")
-        eq(str(td(microseconds=3)), "0:00:00.000003")
-
-        eq(str(td(days=999999999, hours=23, minutes=59, seconds=59,
-                   microseconds=999999)),
-           "999999999 days, 23:59:59.999999")
-
-    def test_repr(self):
-        name = 'datetime.' + self.theclass.__name__
-        self.assertEqual(repr(self.theclass(1)),
-                         "%s(1)" % name)
-        self.assertEqual(repr(self.theclass(10, 2)),
-                         "%s(10, 2)" % name)
-        self.assertEqual(repr(self.theclass(-10, 2, 400000)),
-                         "%s(-10, 2, 400000)" % name)
-
-    def test_roundtrip(self):
-        for td in (timedelta(days=999999999, hours=23, minutes=59,
-                             seconds=59, microseconds=999999),
-                   timedelta(days=-999999999),
-                   timedelta(days=-999999999, seconds=1),
-                   timedelta(days=1, seconds=2, microseconds=3)):
-
-            # Verify td -> string -> td identity.
-            s = repr(td)
-            self.assertTrue(s.startswith('datetime.'))
-            s = s[9:]
-            td2 = eval(s)
-            self.assertEqual(td, td2)
-
-            # Verify identity via reconstructing from pieces.
-            td2 = timedelta(td.days, td.seconds, td.microseconds)
-            self.assertEqual(td, td2)
-
-    def test_resolution_info(self):
-        self.assertIsInstance(timedelta.min, timedelta)
-        self.assertIsInstance(timedelta.max, timedelta)
-        self.assertIsInstance(timedelta.resolution, timedelta)
-        self.assertTrue(timedelta.max > timedelta.min)
-        self.assertEqual(timedelta.min, timedelta(-999999999))
-        self.assertEqual(timedelta.max, timedelta(999999999, 24*3600-1, 1e6-1))
-        self.assertEqual(timedelta.resolution, timedelta(0, 0, 1))
-
-    def test_overflow(self):
-        tiny = timedelta.resolution
-
-        td = timedelta.min + tiny
-        td -= tiny  # no problem
-        self.assertRaises(OverflowError, td.__sub__, tiny)
-        self.assertRaises(OverflowError, td.__add__, -tiny)
-
-        td = timedelta.max - tiny
-        td += tiny  # no problem
-        self.assertRaises(OverflowError, td.__add__, tiny)
-        self.assertRaises(OverflowError, td.__sub__, -tiny)
-
-        self.assertRaises(OverflowError, lambda: -timedelta.max)
-
-        day = timedelta(1)
-        self.assertRaises(OverflowError, day.__mul__, 10**9)
-        self.assertRaises(OverflowError, day.__mul__, 1e9)
-        self.assertRaises(OverflowError, day.__truediv__, 1e-20)
-        self.assertRaises(OverflowError, day.__truediv__, 1e-10)
-        self.assertRaises(OverflowError, day.__truediv__, 9e-10)
-
-    @requires_IEEE_754
-    def _test_overflow_special(self):
-        day = timedelta(1)
-        self.assertRaises(OverflowError, day.__mul__, INF)
-        self.assertRaises(OverflowError, day.__mul__, -INF)
-
-    def test_microsecond_rounding(self):
-        td = timedelta
-        eq = self.assertEqual
-
-        # Single-field rounding.
-        eq(td(milliseconds=0.4/1000), td(0))    # rounds to 0
-        eq(td(milliseconds=-0.4/1000), td(0))    # rounds to 0
-        eq(td(milliseconds=0.6/1000), td(microseconds=1))
-        eq(td(milliseconds=-0.6/1000), td(microseconds=-1))
-
-        # Rounding due to contributions from more than one field.
-        us_per_hour = 3600e6
-        us_per_day = us_per_hour * 24
-        eq(td(days=.4/us_per_day), td(0))
-        eq(td(hours=.2/us_per_hour), td(0))
-        eq(td(days=.4/us_per_day, hours=.2/us_per_hour), td(microseconds=1))
-
-        eq(td(days=-.4/us_per_day), td(0))
-        eq(td(hours=-.2/us_per_hour), td(0))
-        eq(td(days=-.4/us_per_day, hours=-.2/us_per_hour), td(microseconds=-1))
-
-    def test_massive_normalization(self):
-        td = timedelta(microseconds=-1)
-        self.assertEqual((td.days, td.seconds, td.microseconds),
-                         (-1, 24*3600-1, 999999))
-
-    def test_bool(self):
-        self.assertTrue(timedelta(1))
-        self.assertTrue(timedelta(0, 1))
-        self.assertTrue(timedelta(0, 0, 1))
-        self.assertTrue(timedelta(microseconds=1))
-        self.assertTrue(not timedelta(0))
-
-    def test_subclass_timedelta(self):
-
-        class T(timedelta):
-            @staticmethod
-            def from_td(td):
-                return T(td.days, td.seconds, td.microseconds)
-
-            def as_hours(self):
-                sum = (self.days * 24 +
-                       self.seconds / 3600.0 +
-                       self.microseconds / 3600e6)
-                return round(sum)
-
-        t1 = T(days=1)
-        self.assertTrue(type(t1) is T)
-        self.assertEqual(t1.as_hours(), 24)
-
-        t2 = T(days=-1, seconds=-3600)
-        self.assertTrue(type(t2) is T)
-        self.assertEqual(t2.as_hours(), -25)
-
-        t3 = t1 + t2
-        self.assertTrue(type(t3) is timedelta)
-        t4 = T.from_td(t3)
-        self.assertTrue(type(t4) is T)
-        self.assertEqual(t3.days, t4.days)
-        self.assertEqual(t3.seconds, t4.seconds)
-        self.assertEqual(t3.microseconds, t4.microseconds)
-        self.assertEqual(str(t3), str(t4))
-        self.assertEqual(t4.as_hours(), -1)
-
-    def test_division(self):
-        t = timedelta(hours=1, minutes=24, seconds=19)
-        second = timedelta(seconds=1)
-        self.assertEqual(t / second, 5059.0)
-        self.assertEqual(t // second, 5059)
-
-        t = timedelta(minutes=2, seconds=30)
-        minute = timedelta(minutes=1)
-        self.assertEqual(t / minute, 2.5)
-        self.assertEqual(t // minute, 2)
-
-        zerotd = timedelta(0)
-        self.assertRaises(ZeroDivisionError, truediv, t, zerotd)
-        self.assertRaises(ZeroDivisionError, floordiv, t, zerotd)
-
-        # self.assertRaises(TypeError, truediv, t, 2)
-        # note: floor division of a timedelta by an integer *is*
-        # currently permitted.
-
-    def test_remainder(self):
-        t = timedelta(minutes=2, seconds=30)
-        minute = timedelta(minutes=1)
-        r = t % minute
-        self.assertEqual(r, timedelta(seconds=30))
-
-        t = timedelta(minutes=-2, seconds=30)
-        r = t %  minute
-        self.assertEqual(r, timedelta(seconds=30))
-
-        zerotd = timedelta(0)
-        self.assertRaises(ZeroDivisionError, mod, t, zerotd)
-
-        self.assertRaises(TypeError, mod, t, 10)
-
-    def test_divmod(self):
-        t = timedelta(minutes=2, seconds=30)
-        minute = timedelta(minutes=1)
-        q, r = divmod(t, minute)
-        self.assertEqual(q, 2)
-        self.assertEqual(r, timedelta(seconds=30))
-
-        t = timedelta(minutes=-2, seconds=30)
-        q, r = divmod(t, minute)
-        self.assertEqual(q, -2)
-        self.assertEqual(r, timedelta(seconds=30))
-
-        zerotd = timedelta(0)
-        self.assertRaises(ZeroDivisionError, divmod, t, zerotd)
-
-        self.assertRaises(TypeError, divmod, t, 10)
-
-
-#############################################################################
-# date tests
-
-class TestDateOnly(unittest.TestCase):
-    # Tests here won't pass if also run on datetime objects, so don't
-    # subclass this to test datetimes too.
-
-    def test_delta_non_days_ignored(self):
-        dt = date(2000, 1, 2)
-        delta = timedelta(days=1, hours=2, minutes=3, seconds=4,
-                          microseconds=5)
-        days = timedelta(delta.days)
-        self.assertEqual(days, timedelta(1))
-
-        dt2 = dt + delta
-        self.assertEqual(dt2, dt + days)
-
-        dt2 = delta + dt
-        self.assertEqual(dt2, dt + days)
-
-        dt2 = dt - delta
-        self.assertEqual(dt2, dt - days)
-
-        delta = -delta
-        days = timedelta(delta.days)
-        self.assertEqual(days, timedelta(-2))
-
-        dt2 = dt + delta
-        self.assertEqual(dt2, dt + days)
-
-        dt2 = delta + dt
-        self.assertEqual(dt2, dt + days)
-
-        dt2 = dt - delta
-        self.assertEqual(dt2, dt - days)
-
-class SubclassDate(date):
-    sub_var = 1
-
-class TestDate(HarmlessMixedComparison, unittest.TestCase):
-    # Tests here should pass for both dates and datetimes, except for a
-    # few tests that TestDateTime overrides.
-
-    theclass = date
-
-    def test_basic_attributes(self):
-        dt = self.theclass(2002, 3, 1)
-        self.assertEqual(dt.year, 2002)
-        self.assertEqual(dt.month, 3)
-        self.assertEqual(dt.day, 1)
-
-    def test_roundtrip(self):
-        for dt in (self.theclass(1, 2, 3),
-                   self.theclass.today()):
-            # Verify dt -> string -> date identity.
-            s = repr(dt)
-            self.assertTrue(s.startswith('datetime.'))
-            s = s[9:]
-            dt2 = eval(s)
-            self.assertEqual(dt, dt2)
-
-            # Verify identity via reconstructing from pieces.
-            dt2 = self.theclass(dt.year, dt.month, dt.day)
-            self.assertEqual(dt, dt2)
-
-    def test_ordinal_conversions(self):
-        # Check some fixed values.
-        for y, m, d, n in [(1, 1, 1, 1),      # calendar origin
-                           (1, 12, 31, 365),
-                           (2, 1, 1, 366),
-                           # first example from "Calendrical Calculations"
-                           (1945, 11, 12, 710347)]:
-            d = self.theclass(y, m, d)
-            self.assertEqual(n, d.toordinal())
-            fromord = self.theclass.fromordinal(n)
-            self.assertEqual(d, fromord)
-            if hasattr(fromord, "hour"):
-            # if we're checking something fancier than a date, verify
-            # the extra fields have been zeroed out
-                self.assertEqual(fromord.hour, 0)
-                self.assertEqual(fromord.minute, 0)
-                self.assertEqual(fromord.second, 0)
-                self.assertEqual(fromord.microsecond, 0)
-
-        # Check first and last days of year spottily across the whole
-        # range of years supported.
-        for year in range(MINYEAR, MAXYEAR+1, 7):
-            # Verify (year, 1, 1) -> ordinal -> y, m, d is identity.
-            d = self.theclass(year, 1, 1)
-            n = d.toordinal()
-            d2 = self.theclass.fromordinal(n)
-            self.assertEqual(d, d2)
-            # Verify that moving back a day gets to the end of year-1.
-            if year > 1:
-                d = self.theclass.fromordinal(n-1)
-                d2 = self.theclass(year-1, 12, 31)
-                self.assertEqual(d, d2)
-                self.assertEqual(d2.toordinal(), n-1)
-
-        # Test every day in a leap-year and a non-leap year.
-        dim = [31, 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31]
-        for year, isleap in (2000, True), (2002, False):
-            n = self.theclass(year, 1, 1).toordinal()
-            for month, maxday in zip(range(1, 13), dim):
-                if month == 2 and isleap:
-                    maxday += 1
-                for day in range(1, maxday+1):
-                    d = self.theclass(year, month, day)
-                    self.assertEqual(d.toordinal(), n)
-                    self.assertEqual(d, self.theclass.fromordinal(n))
-                    n += 1
-
-    def test_extreme_ordinals(self):
-        a = self.theclass.min
-        a = self.theclass(a.year, a.month, a.day)  # get rid of time parts
-        aord = a.toordinal()
-        b = a.fromordinal(aord)
-        self.assertEqual(a, b)
-
-        self.assertRaises(ValueError, lambda: a.fromordinal(aord - 1))
-
-        b = a + timedelta(days=1)
-        self.assertEqual(b.toordinal(), aord + 1)
-        self.assertEqual(b, self.theclass.fromordinal(aord + 1))
-
-        a = self.theclass.max
-        a = self.theclass(a.year, a.month, a.day)  # get rid of time parts
-        aord = a.toordinal()
-        b = a.fromordinal(aord)
-        self.assertEqual(a, b)
-
-        self.assertRaises(ValueError, lambda: a.fromordinal(aord + 1))
-
-        b = a - timedelta(days=1)
-        self.assertEqual(b.toordinal(), aord - 1)
-        self.assertEqual(b, self.theclass.fromordinal(aord - 1))
-
-    def test_bad_constructor_arguments(self):
-        # bad years
-        self.theclass(MINYEAR, 1, 1)  # no exception
-        self.theclass(MAXYEAR, 1, 1)  # no exception
-        self.assertRaises(ValueError, self.theclass, MINYEAR-1, 1, 1)
-        self.assertRaises(ValueError, self.theclass, MAXYEAR+1, 1, 1)
-        # bad months
-        self.theclass(2000, 1, 1)    # no exception
-        self.theclass(2000, 12, 1)   # no exception
-        self.assertRaises(ValueError, self.theclass, 2000, 0, 1)
-        self.assertRaises(ValueError, self.theclass, 2000, 13, 1)
-        # bad days
-        self.theclass(2000, 2, 29)   # no exception
-        self.theclass(2004, 2, 29)   # no exception
-        self.theclass(2400, 2, 29)   # no exception
-        self.assertRaises(ValueError, self.theclass, 2000, 2, 30)
-        self.assertRaises(ValueError, self.theclass, 2001, 2, 29)
-        self.assertRaises(ValueError, self.theclass, 2100, 2, 29)
-        self.assertRaises(ValueError, self.theclass, 1900, 2, 29)
-        self.assertRaises(ValueError, self.theclass, 2000, 1, 0)
-        self.assertRaises(ValueError, self.theclass, 2000, 1, 32)
-
-    def test_hash_equality(self):
-        d = self.theclass(2000, 12, 31)
-        # same thing
-        e = self.theclass(2000, 12, 31)
-        self.assertEqual(d, e)
-        self.assertEqual(hash(d), hash(e))
-
-        dic = {d: 1}
-        dic[e] = 2
-        self.assertEqual(len(dic), 1)
-        self.assertEqual(dic[d], 2)
-        self.assertEqual(dic[e], 2)
-
-        d = self.theclass(2001,  1,  1)
-        # same thing
-        e = self.theclass(2001,  1,  1)
-        self.assertEqual(d, e)
-        self.assertEqual(hash(d), hash(e))
-
-        dic = {d: 1}
-        dic[e] = 2
-        self.assertEqual(len(dic), 1)
-        self.assertEqual(dic[d], 2)
-        self.assertEqual(dic[e], 2)
-
-    def test_computations(self):
-        a = self.theclass(2002, 1, 31)
-        b = self.theclass(1956, 1, 31)
-        c = self.theclass(2001,2,1)
-
-        diff = a-b
-        self.assertEqual(diff.days, 46*365 + len(range(1956, 2002, 4)))
-        self.assertEqual(diff.seconds, 0)
-        self.assertEqual(diff.microseconds, 0)
-
-        day = timedelta(1)
-        week = timedelta(7)
-        a = self.theclass(2002, 3, 2)
-        self.assertEqual(a + day, self.theclass(2002, 3, 3))
-        self.assertEqual(day + a, self.theclass(2002, 3, 3))
-        self.assertEqual(a - day, self.theclass(2002, 3, 1))
-        self.assertEqual(-day + a, self.theclass(2002, 3, 1))
-        self.assertEqual(a + week, self.theclass(2002, 3, 9))
-        self.assertEqual(a - week, self.theclass(2002, 2, 23))
-        self.assertEqual(a + 52*week, self.theclass(2003, 3, 1))
-        self.assertEqual(a - 52*week, self.theclass(2001, 3, 3))
-        self.assertEqual((a + week) - a, week)
-        self.assertEqual((a + day) - a, day)
-        self.assertEqual((a - week) - a, -week)
-        self.assertEqual((a - day) - a, -day)
-        self.assertEqual(a - (a + week), -week)
-        self.assertEqual(a - (a + day), -day)
-        self.assertEqual(a - (a - week), week)
-        self.assertEqual(a - (a - day), day)
-        self.assertEqual(c - (c - day), day)
-
-        # Add/sub ints or floats should be illegal
-        for i in 1, 1.0:
-            self.assertRaises(TypeError, lambda: a+i)
-            self.assertRaises(TypeError, lambda: a-i)
-            self.assertRaises(TypeError, lambda: i+a)
-            self.assertRaises(TypeError, lambda: i-a)
-
-        # delta - date is senseless.
-        self.assertRaises(TypeError, lambda: day - a)
-        # mixing date and (delta or date) via * or // is senseless
-        self.assertRaises(TypeError, lambda: day * a)
-        self.assertRaises(TypeError, lambda: a * day)
-        self.assertRaises(TypeError, lambda: day // a)
-        self.assertRaises(TypeError, lambda: a // day)
-        self.assertRaises(TypeError, lambda: a * a)
-        self.assertRaises(TypeError, lambda: a // a)
-        # date + date is senseless
-        self.assertRaises(TypeError, lambda: a + a)
-
-    def test_overflow(self):
-        tiny = self.theclass.resolution
-
-        for delta in [tiny, timedelta(1), timedelta(2)]:
-            dt = self.theclass.min + delta
-            dt -= delta  # no problem
-            self.assertRaises(OverflowError, dt.__sub__, delta)
-            self.assertRaises(OverflowError, dt.__add__, -delta)
-
-            dt = self.theclass.max - delta
-            dt += delta  # no problem
-            self.assertRaises(OverflowError, dt.__add__, delta)
-            self.assertRaises(OverflowError, dt.__sub__, -delta)
-
-    def test_fromtimestamp(self):
-        import time
-
-        # Try an arbitrary fixed value.
-        year, month, day = 1999, 9, 19
-        ts = time.mktime((year, month, day, 0, 0, 0, 0, 0, -1))
-        d = self.theclass.fromtimestamp(ts)
-        self.assertEqual(d.year, year)
-        self.assertEqual(d.month, month)
-        self.assertEqual(d.day, day)
-
-    def test_insane_fromtimestamp(self):
-        # It's possible that some platform maps time_t to double,
-        # and that this test will fail there.  This test should
-        # exempt such platforms (provided they return reasonable
-        # results!).
-        for insane in -1e200, 1e200:
-            self.assertRaises(ValueError, self.theclass.fromtimestamp,
-                              insane)
-
-    def test_today(self):
-        import time
-
-        # We claim that today() is like fromtimestamp(time.time()), so
-        # prove it.
-        for dummy in range(3):
-            today = self.theclass.today()
-            ts = time.time()
-            todayagain = self.theclass.fromtimestamp(ts)
-            if today == todayagain:
-                break
-            # There are several legit reasons that could fail:
-            # 1. It recently became midnight, between the today() and the
-            #    time() calls.
-            # 2. The platform time() has such fine resolution that we'll
-            #    never get the same value twice.
-            # 3. The platform time() has poor resolution, and we just
-            #    happened to call today() right before a resolution quantum
-            #    boundary.
-            # 4. The system clock got fiddled between calls.
-            # In any case, wait a little while and try again.
-            time.sleep(0.1)
-
-        # It worked or it didn't.  If it didn't, assume it's reason #2, and
-        # let the test pass if they're within half a second of each other.
-        self.assertTrue(today == todayagain or
-                        abs(todayagain - today) < timedelta(seconds=0.5))
-
-    def test_weekday(self):
-        for i in range(7):
-            # March 4, 2002 is a Monday
-            self.assertEqual(self.theclass(2002, 3, 4+i).weekday(), i)
-            self.assertEqual(self.theclass(2002, 3, 4+i).isoweekday(), i+1)
-            # January 2, 1956 is a Monday
-            self.assertEqual(self.theclass(1956, 1, 2+i).weekday(), i)
-            self.assertEqual(self.theclass(1956, 1, 2+i).isoweekday(), i+1)
-
-    def test_isocalendar(self):
-        # Check examples from
-        # http://www.phys.uu.nl/~vgent/calendar/isocalendar.htm
-        for i in range(7):
-            d = self.theclass(2003, 12, 22+i)
-            self.assertEqual(d.isocalendar(), (2003, 52, i+1))
-            d = self.theclass(2003, 12, 29) + timedelta(i)
-            self.assertEqual(d.isocalendar(), (2004, 1, i+1))
-            d = self.theclass(2004, 1, 5+i)
-            self.assertEqual(d.isocalendar(), (2004, 2, i+1))
-            d = self.theclass(2009, 12, 21+i)
-            self.assertEqual(d.isocalendar(), (2009, 52, i+1))
-            d = self.theclass(2009, 12, 28) + timedelta(i)
-            self.assertEqual(d.isocalendar(), (2009, 53, i+1))
-            d = self.theclass(2010, 1, 4+i)
-            self.assertEqual(d.isocalendar(), (2010, 1, i+1))
-
-    def test_iso_long_years(self):
-        # Calculate long ISO years and compare to table from
-        # http://www.phys.uu.nl/~vgent/calendar/isocalendar.htm
-        ISO_LONG_YEARS_TABLE = """
-              4   32   60   88
-              9   37   65   93
-             15   43   71   99
-             20   48   76
-             26   54   82
-
-            105  133  161  189
-            111  139  167  195
-            116  144  172
-            122  150  178
-            128  156  184
-
-            201  229  257  285
-            207  235  263  291
-            212  240  268  296
-            218  246  274
-            224  252  280
-
-            303  331  359  387
-            308  336  364  392
-            314  342  370  398
-            320  348  376
-            325  353  381
-        """
-        iso_long_years = sorted(map(int, ISO_LONG_YEARS_TABLE.split()))
-        L = []
-        for i in range(400):
-            d = self.theclass(2000+i, 12, 31)
-            d1 = self.theclass(1600+i, 12, 31)
-            self.assertEqual(d.isocalendar()[1:], d1.isocalendar()[1:])
-            if d.isocalendar()[1] == 53:
-                L.append(i)
-        self.assertEqual(L, iso_long_years)
-
-    def test_isoformat(self):
-        t = self.theclass(2, 3, 2)
-        self.assertEqual(t.isoformat(), "0002-03-02")
-
-    def test_ctime(self):
-        t = self.theclass(2002, 3, 2)
-        self.assertEqual(t.ctime(), "Sat Mar  2 00:00:00 2002")
-
-    def test_strftime(self):
-        t = self.theclass(2005, 3, 2)
-        self.assertEqual(t.strftime("m:%m d:%d y:%y"), "m:03 d:02 y:05")
-        self.assertEqual(t.strftime(""), "") # SF bug #761337
-        self.assertEqual(t.strftime('x'*1000), 'x'*1000) # SF bug #1556784
-
-        self.assertRaises(TypeError, t.strftime) # needs an arg
-        self.assertRaises(TypeError, t.strftime, "one", "two") # too many args
-        self.assertRaises(TypeError, t.strftime, 42) # arg wrong type
-
-        # test that unicode input is allowed (issue 2782)
-        self.assertEqual(t.strftime("%m"), "03")
-
-        # A naive object replaces %z and %Z w/ empty strings.
-        self.assertEqual(t.strftime("'%z' '%Z'"), "'' ''")
-
-        #make sure that invalid format specifiers are handled correctly
-        #self.assertRaises(ValueError, t.strftime, "%e")
-        #self.assertRaises(ValueError, t.strftime, "%")
-        #self.assertRaises(ValueError, t.strftime, "%#")
-
-        #oh well, some systems just ignore those invalid ones.
-        #at least, excercise them to make sure that no crashes
-        #are generated
-        for f in ["%e", "%", "%#"]:
-            try:
-                t.strftime(f)
-            except ValueError:
-                pass
-
-        #check that this standard extension works
-        t.strftime("%f")
-
-
-    def test_format(self):
-        dt = self.theclass(2007, 9, 10)
-        self.assertEqual(dt.__format__(''), str(dt))
-
-        # check that a derived class's __str__() gets called
-        class A(self.theclass):
-            def __str__(self):
-                return 'A'
-        a = A(2007, 9, 10)
-        self.assertEqual(a.__format__(''), 'A')
-
-        # check that a derived class's strftime gets called
-        class B(self.theclass):
-            def strftime(self, format_spec):
-                return 'B'
-        b = B(2007, 9, 10)
-        self.assertEqual(b.__format__(''), str(dt))
-
-        for fmt in ["m:%m d:%d y:%y",
-                    "m:%m d:%d y:%y H:%H M:%M S:%S",
-                    "%z %Z",
-                    ]:
-            self.assertEqual(dt.__format__(fmt), dt.strftime(fmt))
-            self.assertEqual(a.__format__(fmt), dt.strftime(fmt))
-            self.assertEqual(b.__format__(fmt), 'B')
-
-    def test_resolution_info(self):
-        # XXX: Should min and max respect subclassing?
-        if issubclass(self.theclass, datetime):
-            expected_class = datetime
-        else:
-            expected_class = date
-        self.assertIsInstance(self.theclass.min, expected_class)
-        self.assertIsInstance(self.theclass.max, expected_class)
-        self.assertIsInstance(self.theclass.resolution, timedelta)
-        self.assertTrue(self.theclass.max > self.theclass.min)
-
-    def test_extreme_timedelta(self):
-        big = self.theclass.max - self.theclass.min
-        # 3652058 days, 23 hours, 59 minutes, 59 seconds, 999999 microseconds
-        n = (big.days*24*3600 + big.seconds)*1000000 + big.microseconds
-        # n == 315537897599999999 ~= 2**58.13
-        justasbig = timedelta(0, 0, n)
-        self.assertEqual(big, justasbig)
-        self.assertEqual(self.theclass.min + big, self.theclass.max)
-        self.assertEqual(self.theclass.max - big, self.theclass.min)
-
-    def test_timetuple(self):
-        for i in range(7):
-            # January 2, 1956 is a Monday (0)
-            d = self.theclass(1956, 1, 2+i)
-            t = d.timetuple()
-            self.assertEqual(t, (1956, 1, 2+i, 0, 0, 0, i, 2+i, -1))
-            # February 1, 1956 is a Wednesday (2)
-            d = self.theclass(1956, 2, 1+i)
-            t = d.timetuple()
-            self.assertEqual(t, (1956, 2, 1+i, 0, 0, 0, (2+i)%7, 32+i, -1))
-            # March 1, 1956 is a Thursday (3), and is the 31+29+1 = 61st day
-            # of the year.
-            d = self.theclass(1956, 3, 1+i)
-            t = d.timetuple()
-            self.assertEqual(t, (1956, 3, 1+i, 0, 0, 0, (3+i)%7, 61+i, -1))
-            self.assertEqual(t.tm_year, 1956)
-            self.assertEqual(t.tm_mon, 3)
-            self.assertEqual(t.tm_mday, 1+i)
-            self.assertEqual(t.tm_hour, 0)
-            self.assertEqual(t.tm_min, 0)
-            self.assertEqual(t.tm_sec, 0)
-            self.assertEqual(t.tm_wday, (3+i)%7)
-            self.assertEqual(t.tm_yday, 61+i)
-            self.assertEqual(t.tm_isdst, -1)
-
-    def test_pickling(self):
-        args = 6, 7, 23
-        orig = self.theclass(*args)
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertEqual(orig, derived)
-
-    def test_compare(self):
-        t1 = self.theclass(2, 3, 4)
-        t2 = self.theclass(2, 3, 4)
-        self.assertEqual(t1, t2)
-        self.assertTrue(t1 <= t2)
-        self.assertTrue(t1 >= t2)
-        self.assertTrue(not t1 != t2)
-        self.assertTrue(not t1 < t2)
-        self.assertTrue(not t1 > t2)
-
-        for args in (3, 3, 3), (2, 4, 4), (2, 3, 5):
-            t2 = self.theclass(*args)   # this is larger than t1
-            self.assertTrue(t1 < t2)
-            self.assertTrue(t2 > t1)
-            self.assertTrue(t1 <= t2)
-            self.assertTrue(t2 >= t1)
-            self.assertTrue(t1 != t2)
-            self.assertTrue(t2 != t1)
-            self.assertTrue(not t1 == t2)
-            self.assertTrue(not t2 == t1)
-            self.assertTrue(not t1 > t2)
-            self.assertTrue(not t2 < t1)
-            self.assertTrue(not t1 >= t2)
-            self.assertTrue(not t2 <= t1)
-
-        for badarg in OTHERSTUFF:
-            self.assertEqual(t1 == badarg, False)
-            self.assertEqual(t1 != badarg, True)
-            self.assertEqual(badarg == t1, False)
-            self.assertEqual(badarg != t1, True)
-
-            self.assertRaises(TypeError, lambda: t1 < badarg)
-            self.assertRaises(TypeError, lambda: t1 > badarg)
-            self.assertRaises(TypeError, lambda: t1 >= badarg)
-            self.assertRaises(TypeError, lambda: badarg <= t1)
-            self.assertRaises(TypeError, lambda: badarg < t1)
-            self.assertRaises(TypeError, lambda: badarg > t1)
-            self.assertRaises(TypeError, lambda: badarg >= t1)
-
-    def test_mixed_compare(self):
-        our = self.theclass(2000, 4, 5)
-
-        # Our class can be compared for equality to other classes
-        self.assertEqual(our == 1, False)
-        self.assertEqual(1 == our, False)
-        self.assertEqual(our != 1, True)
-        self.assertEqual(1 != our, True)
-
-        # But the ordering is undefined
-        self.assertRaises(TypeError, lambda: our < 1)
-        self.assertRaises(TypeError, lambda: 1 < our)
-
-        # Repeat those tests with a different class
-
-        class SomeClass:
-            pass
-
-        their = SomeClass()
-        self.assertEqual(our == their, False)
-        self.assertEqual(their == our, False)
-        self.assertEqual(our != their, True)
-        self.assertEqual(their != our, True)
-        self.assertRaises(TypeError, lambda: our < their)
-        self.assertRaises(TypeError, lambda: their < our)
-
-        # However, if the other class explicitly defines ordering
-        # relative to our class, it is allowed to do so
-
-        class LargerThanAnything:
-            def __lt__(self, other):
-                return False
-            def __le__(self, other):
-                return isinstance(other, LargerThanAnything)
-            def __eq__(self, other):
-                return isinstance(other, LargerThanAnything)
-            def __ne__(self, other):
-                return not isinstance(other, LargerThanAnything)
-            def __gt__(self, other):
-                return not isinstance(other, LargerThanAnything)
-            def __ge__(self, other):
-                return True
-
-        their = LargerThanAnything()
-        self.assertEqual(our == their, False)
-        self.assertEqual(their == our, False)
-        self.assertEqual(our != their, True)
-        self.assertEqual(their != our, True)
-        self.assertEqual(our < their, True)
-        self.assertEqual(their < our, False)
-
-    def test_bool(self):
-        # All dates are considered true.
-        self.assertTrue(self.theclass.min)
-        self.assertTrue(self.theclass.max)
-
-    def test_strftime_out_of_range(self):
-        # For nasty technical reasons, we can't handle years before 1900.
-        cls = self.theclass
-        self.assertEqual(cls(1900, 1, 1).strftime("%Y"), "1900")
-        for y in 1, 49, 51, 99, 100, 1000, 1899:
-            self.assertRaises(ValueError, cls(y, 1, 1).strftime, "%Y")
-
-    def test_replace(self):
-        cls = self.theclass
-        args = [1, 2, 3]
-        base = cls(*args)
-        self.assertEqual(base, base.replace())
-
-        i = 0
-        for name, newval in (("year", 2),
-                             ("month", 3),
-                             ("day", 4)):
-            newargs = args[:]
-            newargs[i] = newval
-            expected = cls(*newargs)
-            got = base.replace(**{name: newval})
-            self.assertEqual(expected, got)
-            i += 1
-
-        # Out of bounds.
-        base = cls(2000, 2, 29)
-        self.assertRaises(ValueError, base.replace, year=2001)
-
-    def test_subclass_date(self):
-
-        class C(self.theclass):
-            theAnswer = 42
-
-            def __new__(cls, *args, **kws):
-                temp = kws.copy()
-                extra = temp.pop('extra')
-                result = self.theclass.__new__(cls, *args, **temp)
-                result.extra = extra
-                return result
-
-            def newmeth(self, start):
-                return start + self.year + self.month
-
-        args = 2003, 4, 14
-
-        dt1 = self.theclass(*args)
-        dt2 = C(*args, **{'extra': 7})
-
-        self.assertEqual(dt2.__class__, C)
-        self.assertEqual(dt2.theAnswer, 42)
-        self.assertEqual(dt2.extra, 7)
-        self.assertEqual(dt1.toordinal(), dt2.toordinal())
-        self.assertEqual(dt2.newmeth(-7), dt1.year + dt1.month - 7)
-
-    def test_pickling_subclass_date(self):
-
-        args = 6, 7, 23
-        orig = SubclassDate(*args)
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertEqual(orig, derived)
-
-    def test_backdoor_resistance(self):
-        # For fast unpickling, the constructor accepts a pickle byte string.
-        # This is a low-overhead backdoor.  A user can (by intent or
-        # mistake) pass a string directly, which (if it's the right length)
-        # will get treated like a pickle, and bypass the normal sanity
-        # checks in the constructor.  This can create insane objects.
-        # The constructor doesn't want to burn the time to validate all
-        # fields, but does check the month field.  This stops, e.g.,
-        # datetime.datetime('1995-03-25') from yielding an insane object.
-        base = b'1995-03-25'
-        if not issubclass(self.theclass, datetime):
-            base = base[:4]
-        for month_byte in b'9', b'\0', b'\r', b'\xff':
-            self.assertRaises(TypeError, self.theclass,
-                                         base[:2] + month_byte + base[3:])
-        # Good bytes, but bad tzinfo:
-        self.assertRaises(TypeError, self.theclass,
-                          bytes([1] * len(base)), 'EST')
-
-        for ord_byte in range(1, 13):
-            # This shouldn't blow up because of the month byte alone.  If
-            # the implementation changes to do more-careful checking, it may
-            # blow up because other fields are insane.
-            self.theclass(base[:2] + bytes([ord_byte]) + base[3:])
-
-#############################################################################
-# datetime tests
-
-class SubclassDatetime(datetime):
-    sub_var = 1
-
-class TestDateTime(TestDate):
-
-    theclass = datetime
-
-    def test_basic_attributes(self):
-        dt = self.theclass(2002, 3, 1, 12, 0)
-        self.assertEqual(dt.year, 2002)
-        self.assertEqual(dt.month, 3)
-        self.assertEqual(dt.day, 1)
-        self.assertEqual(dt.hour, 12)
-        self.assertEqual(dt.minute, 0)
-        self.assertEqual(dt.second, 0)
-        self.assertEqual(dt.microsecond, 0)
-
-    def test_basic_attributes_nonzero(self):
-        # Make sure all attributes are non-zero so bugs in
-        # bit-shifting access show up.
-        dt = self.theclass(2002, 3, 1, 12, 59, 59, 8000)
-        self.assertEqual(dt.year, 2002)
-        self.assertEqual(dt.month, 3)
-        self.assertEqual(dt.day, 1)
-        self.assertEqual(dt.hour, 12)
-        self.assertEqual(dt.minute, 59)
-        self.assertEqual(dt.second, 59)
-        self.assertEqual(dt.microsecond, 8000)
-
-    def test_roundtrip(self):
-        for dt in (self.theclass(1, 2, 3, 4, 5, 6, 7),
-                   self.theclass.now()):
-            # Verify dt -> string -> datetime identity.
-            s = repr(dt)
-            self.assertTrue(s.startswith('datetime.'))
-            s = s[9:]
-            dt2 = eval(s)
-            self.assertEqual(dt, dt2)
-
-            # Verify identity via reconstructing from pieces.
-            dt2 = self.theclass(dt.year, dt.month, dt.day,
-                                dt.hour, dt.minute, dt.second,
-                                dt.microsecond)
-            self.assertEqual(dt, dt2)
-
-    def test_isoformat(self):
-        t = self.theclass(2, 3, 2, 4, 5, 1, 123)
-        self.assertEqual(t.isoformat(),    "0002-03-02T04:05:01.000123")
-        self.assertEqual(t.isoformat('T'), "0002-03-02T04:05:01.000123")
-        self.assertEqual(t.isoformat(' '), "0002-03-02 04:05:01.000123")
-        self.assertEqual(t.isoformat('\x00'), "0002-03-02\x0004:05:01.000123")
-        # str is ISO format with the separator forced to a blank.
-        self.assertEqual(str(t), "0002-03-02 04:05:01.000123")
-
-        t = self.theclass(2, 3, 2)
-        self.assertEqual(t.isoformat(),    "0002-03-02T00:00:00")
-        self.assertEqual(t.isoformat('T'), "0002-03-02T00:00:00")
-        self.assertEqual(t.isoformat(' '), "0002-03-02 00:00:00")
-        # str is ISO format with the separator forced to a blank.
-        self.assertEqual(str(t), "0002-03-02 00:00:00")
-
-    def test_format(self):
-        dt = self.theclass(2007, 9, 10, 4, 5, 1, 123)
-        self.assertEqual(dt.__format__(''), str(dt))
-
-        # check that a derived class's __str__() gets called
-        class A(self.theclass):
-            def __str__(self):
-                return 'A'
-        a = A(2007, 9, 10, 4, 5, 1, 123)
-        self.assertEqual(a.__format__(''), 'A')
-
-        # check that a derived class's strftime gets called
-        class B(self.theclass):
-            def strftime(self, format_spec):
-                return 'B'
-        b = B(2007, 9, 10, 4, 5, 1, 123)
-        self.assertEqual(b.__format__(''), str(dt))
-
-        for fmt in ["m:%m d:%d y:%y",
-                    "m:%m d:%d y:%y H:%H M:%M S:%S",
-                    "%z %Z",
-                    ]:
-            self.assertEqual(dt.__format__(fmt), dt.strftime(fmt))
-            self.assertEqual(a.__format__(fmt), dt.strftime(fmt))
-            self.assertEqual(b.__format__(fmt), 'B')
-
-    def test_more_ctime(self):
-        # Test fields that TestDate doesn't touch.
-        import time
-
-        t = self.theclass(2002, 3, 2, 18, 3, 5, 123)
-        self.assertEqual(t.ctime(), "Sat Mar  2 18:03:05 2002")
-        # Oops!  The next line fails on Win2K under MSVC 6, so it's commented
-        # out.  The difference is that t.ctime() produces " 2" for the day,
-        # but platform ctime() produces "02" for the day.  According to
-        # C99, t.ctime() is correct here.
-        # self.assertEqual(t.ctime(), time.ctime(time.mktime(t.timetuple())))
-
-        # So test a case where that difference doesn't matter.
-        t = self.theclass(2002, 3, 22, 18, 3, 5, 123)
-        self.assertEqual(t.ctime(), time.ctime(time.mktime(t.timetuple())))
-
-    def test_tz_independent_comparing(self):
-        dt1 = self.theclass(2002, 3, 1, 9, 0, 0)
-        dt2 = self.theclass(2002, 3, 1, 10, 0, 0)
-        dt3 = self.theclass(2002, 3, 1, 9, 0, 0)
-        self.assertEqual(dt1, dt3)
-        self.assertTrue(dt2 > dt3)
-
-        # Make sure comparison doesn't forget microseconds, and isn't done
-        # via comparing a float timestamp (an IEEE double doesn't have enough
-        # precision to span microsecond resolution across years 1 thru 9999,
-        # so comparing via timestamp necessarily calls some distinct values
-        # equal).
-        dt1 = self.theclass(MAXYEAR, 12, 31, 23, 59, 59, 999998)
-        us = timedelta(microseconds=1)
-        dt2 = dt1 + us
-        self.assertEqual(dt2 - dt1, us)
-        self.assertTrue(dt1 < dt2)
-
-    def test_strftime_with_bad_tzname_replace(self):
-        # verify ok if tzinfo.tzname().replace() returns a non-string
-        class MyTzInfo(FixedOffset):
-            def tzname(self, dt):
-                class MyStr(str):
-                    def replace(self, *args):
-                        return None
-                return MyStr('name')
-        t = self.theclass(2005, 3, 2, 0, 0, 0, 0, MyTzInfo(3, 'name'))
-        self.assertRaises(TypeError, t.strftime, '%Z')
-
-    def test_bad_constructor_arguments(self):
-        # bad years
-        self.theclass(MINYEAR, 1, 1)  # no exception
-        self.theclass(MAXYEAR, 1, 1)  # no exception
-        self.assertRaises(ValueError, self.theclass, MINYEAR-1, 1, 1)
-        self.assertRaises(ValueError, self.theclass, MAXYEAR+1, 1, 1)
-        # bad months
-        self.theclass(2000, 1, 1)    # no exception
-        self.theclass(2000, 12, 1)   # no exception
-        self.assertRaises(ValueError, self.theclass, 2000, 0, 1)
-        self.assertRaises(ValueError, self.theclass, 2000, 13, 1)
-        # bad days
-        self.theclass(2000, 2, 29)   # no exception
-        self.theclass(2004, 2, 29)   # no exception
-        self.theclass(2400, 2, 29)   # no exception
-        self.assertRaises(ValueError, self.theclass, 2000, 2, 30)
-        self.assertRaises(ValueError, self.theclass, 2001, 2, 29)
-        self.assertRaises(ValueError, self.theclass, 2100, 2, 29)
-        self.assertRaises(ValueError, self.theclass, 1900, 2, 29)
-        self.assertRaises(ValueError, self.theclass, 2000, 1, 0)
-        self.assertRaises(ValueError, self.theclass, 2000, 1, 32)
-        # bad hours
-        self.theclass(2000, 1, 31, 0)    # no exception
-        self.theclass(2000, 1, 31, 23)   # no exception
-        self.assertRaises(ValueError, self.theclass, 2000, 1, 31, -1)
-        self.assertRaises(ValueError, self.theclass, 2000, 1, 31, 24)
-        # bad minutes
-        self.theclass(2000, 1, 31, 23, 0)    # no exception
-        self.theclass(2000, 1, 31, 23, 59)   # no exception
-        self.assertRaises(ValueError, self.theclass, 2000, 1, 31, 23, -1)
-        self.assertRaises(ValueError, self.theclass, 2000, 1, 31, 23, 60)
-        # bad seconds
-        self.theclass(2000, 1, 31, 23, 59, 0)    # no exception
-        self.theclass(2000, 1, 31, 23, 59, 59)   # no exception
-        self.assertRaises(ValueError, self.theclass, 2000, 1, 31, 23, 59, -1)
-        self.assertRaises(ValueError, self.theclass, 2000, 1, 31, 23, 59, 60)
-        # bad microseconds
-        self.theclass(2000, 1, 31, 23, 59, 59, 0)    # no exception
-        self.theclass(2000, 1, 31, 23, 59, 59, 999999)   # no exception
-        self.assertRaises(ValueError, self.theclass,
-                          2000, 1, 31, 23, 59, 59, -1)
-        self.assertRaises(ValueError, self.theclass,
-                          2000, 1, 31, 23, 59, 59,
-                          1000000)
-
-    def test_hash_equality(self):
-        d = self.theclass(2000, 12, 31, 23, 30, 17)
-        e = self.theclass(2000, 12, 31, 23, 30, 17)
-        self.assertEqual(d, e)
-        self.assertEqual(hash(d), hash(e))
-
-        dic = {d: 1}
-        dic[e] = 2
-        self.assertEqual(len(dic), 1)
-        self.assertEqual(dic[d], 2)
-        self.assertEqual(dic[e], 2)
-
-        d = self.theclass(2001,  1,  1,  0,  5, 17)
-        e = self.theclass(2001,  1,  1,  0,  5, 17)
-        self.assertEqual(d, e)
-        self.assertEqual(hash(d), hash(e))
-
-        dic = {d: 1}
-        dic[e] = 2
-        self.assertEqual(len(dic), 1)
-        self.assertEqual(dic[d], 2)
-        self.assertEqual(dic[e], 2)
-
-    def test_computations(self):
-        a = self.theclass(2002, 1, 31)
-        b = self.theclass(1956, 1, 31)
-        diff = a-b
-        self.assertEqual(diff.days, 46*365 + len(range(1956, 2002, 4)))
-        self.assertEqual(diff.seconds, 0)
-        self.assertEqual(diff.microseconds, 0)
-        a = self.theclass(2002, 3, 2, 17, 6)
-        millisec = timedelta(0, 0, 1000)
-        hour = timedelta(0, 3600)
-        day = timedelta(1)
-        week = timedelta(7)
-        self.assertEqual(a + hour, self.theclass(2002, 3, 2, 18, 6))
-        self.assertEqual(hour + a, self.theclass(2002, 3, 2, 18, 6))
-        self.assertEqual(a + 10*hour, self.theclass(2002, 3, 3, 3, 6))
-        self.assertEqual(a - hour, self.theclass(2002, 3, 2, 16, 6))
-        self.assertEqual(-hour + a, self.theclass(2002, 3, 2, 16, 6))
-        self.assertEqual(a - hour, a + -hour)
-        self.assertEqual(a - 20*hour, self.theclass(2002, 3, 1, 21, 6))
-        self.assertEqual(a + day, self.theclass(2002, 3, 3, 17, 6))
-        self.assertEqual(a - day, self.theclass(2002, 3, 1, 17, 6))
-        self.assertEqual(a + week, self.theclass(2002, 3, 9, 17, 6))
-        self.assertEqual(a - week, self.theclass(2002, 2, 23, 17, 6))
-        self.assertEqual(a + 52*week, self.theclass(2003, 3, 1, 17, 6))
-        self.assertEqual(a - 52*week, self.theclass(2001, 3, 3, 17, 6))
-        self.assertEqual((a + week) - a, week)
-        self.assertEqual((a + day) - a, day)
-        self.assertEqual((a + hour) - a, hour)
-        self.assertEqual((a + millisec) - a, millisec)
-        self.assertEqual((a - week) - a, -week)
-        self.assertEqual((a - day) - a, -day)
-        self.assertEqual((a - hour) - a, -hour)
-        self.assertEqual((a - millisec) - a, -millisec)
-        self.assertEqual(a - (a + week), -week)
-        self.assertEqual(a - (a + day), -day)
-        self.assertEqual(a - (a + hour), -hour)
-        self.assertEqual(a - (a + millisec), -millisec)
-        self.assertEqual(a - (a - week), week)
-        self.assertEqual(a - (a - day), day)
-        self.assertEqual(a - (a - hour), hour)
-        self.assertEqual(a - (a - millisec), millisec)
-        self.assertEqual(a + (week + day + hour + millisec),
-                         self.theclass(2002, 3, 10, 18, 6, 0, 1000))
-        self.assertEqual(a + (week + day + hour + millisec),
-                         (((a + week) + day) + hour) + millisec)
-        self.assertEqual(a - (week + day + hour + millisec),
-                         self.theclass(2002, 2, 22, 16, 5, 59, 999000))
-        self.assertEqual(a - (week + day + hour + millisec),
-                         (((a - week) - day) - hour) - millisec)
-        # Add/sub ints or floats should be illegal
-        for i in 1, 1.0:
-            self.assertRaises(TypeError, lambda: a+i)
-            self.assertRaises(TypeError, lambda: a-i)
-            self.assertRaises(TypeError, lambda: i+a)
-            self.assertRaises(TypeError, lambda: i-a)
-
-        # delta - datetime is senseless.
-        self.assertRaises(TypeError, lambda: day - a)
-        # mixing datetime and (delta or datetime) via * or // is senseless
-        self.assertRaises(TypeError, lambda: day * a)
-        self.assertRaises(TypeError, lambda: a * day)
-        self.assertRaises(TypeError, lambda: day // a)
-        self.assertRaises(TypeError, lambda: a // day)
-        self.assertRaises(TypeError, lambda: a * a)
-        self.assertRaises(TypeError, lambda: a // a)
-        # datetime + datetime is senseless
-        self.assertRaises(TypeError, lambda: a + a)
-
-    def test_pickling(self):
-        args = 6, 7, 23, 20, 59, 1, 64**2
-        orig = self.theclass(*args)
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertEqual(orig, derived)
-
-    def test_more_pickling(self):
-        a = self.theclass(2003, 2, 7, 16, 48, 37, 444116)
-        s = pickle.dumps(a)
-        b = pickle.loads(s)
-        self.assertEqual(b.year, 2003)
-        self.assertEqual(b.month, 2)
-        self.assertEqual(b.day, 7)
-
-    def test_pickling_subclass_datetime(self):
-        args = 6, 7, 23, 20, 59, 1, 64**2
-        orig = SubclassDatetime(*args)
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertEqual(orig, derived)
-
-    def test_more_compare(self):
-        # The test_compare() inherited from TestDate covers the error cases.
-        # We just want to test lexicographic ordering on the members datetime
-        # has that date lacks.
-        args = [2000, 11, 29, 20, 58, 16, 999998]
-        t1 = self.theclass(*args)
-        t2 = self.theclass(*args)
-        self.assertEqual(t1, t2)
-        self.assertTrue(t1 <= t2)
-        self.assertTrue(t1 >= t2)
-        self.assertTrue(not t1 != t2)
-        self.assertTrue(not t1 < t2)
-        self.assertTrue(not t1 > t2)
-
-        for i in range(len(args)):
-            newargs = args[:]
-            newargs[i] = args[i] + 1
-            t2 = self.theclass(*newargs)   # this is larger than t1
-            self.assertTrue(t1 < t2)
-            self.assertTrue(t2 > t1)
-            self.assertTrue(t1 <= t2)
-            self.assertTrue(t2 >= t1)
-            self.assertTrue(t1 != t2)
-            self.assertTrue(t2 != t1)
-            self.assertTrue(not t1 == t2)
-            self.assertTrue(not t2 == t1)
-            self.assertTrue(not t1 > t2)
-            self.assertTrue(not t2 < t1)
-            self.assertTrue(not t1 >= t2)
-            self.assertTrue(not t2 <= t1)
-
-
-    # A helper for timestamp constructor tests.
-    def verify_field_equality(self, expected, got):
-        self.assertEqual(expected.tm_year, got.year)
-        self.assertEqual(expected.tm_mon, got.month)
-        self.assertEqual(expected.tm_mday, got.day)
-        self.assertEqual(expected.tm_hour, got.hour)
-        self.assertEqual(expected.tm_min, got.minute)
-        self.assertEqual(expected.tm_sec, got.second)
-
-    def test_fromtimestamp(self):
-        import time
-
-        ts = time.time()
-        expected = time.localtime(ts)
-        got = self.theclass.fromtimestamp(ts)
-        self.verify_field_equality(expected, got)
-
-    def test_utcfromtimestamp(self):
-        import time
-
-        ts = time.time()
-        expected = time.gmtime(ts)
-        got = self.theclass.utcfromtimestamp(ts)
-        self.verify_field_equality(expected, got)
-
-    def test_microsecond_rounding(self):
-        # Test whether fromtimestamp "rounds up" floats that are less
-        # than one microsecond smaller than an integer.
-        self.assertEqual(self.theclass.fromtimestamp(0.9999999),
-                         self.theclass.fromtimestamp(1))
-
-    def test_insane_fromtimestamp(self):
-        # It's possible that some platform maps time_t to double,
-        # and that this test will fail there.  This test should
-        # exempt such platforms (provided they return reasonable
-        # results!).
-        for insane in -1e200, 1e200:
-            self.assertRaises(ValueError, self.theclass.fromtimestamp,
-                              insane)
-
-    def test_insane_utcfromtimestamp(self):
-        # It's possible that some platform maps time_t to double,
-        # and that this test will fail there.  This test should
-        # exempt such platforms (provided they return reasonable
-        # results!).
-        for insane in -1e200, 1e200:
-            self.assertRaises(ValueError, self.theclass.utcfromtimestamp,
-                              insane)
-    @unittest.skipIf(sys.platform == "win32", "Windows doesn't accept negative timestamps")
-    def test_negative_float_fromtimestamp(self):
-        # The result is tz-dependent; at least test that this doesn't
-        # fail (like it did before bug 1646728 was fixed).
-        self.theclass.fromtimestamp(-1.05)
-
-    @unittest.skipIf(sys.platform == "win32", "Windows doesn't accept negative timestamps")
-    def test_negative_float_utcfromtimestamp(self):
-        d = self.theclass.utcfromtimestamp(-1.05)
-        self.assertEqual(d, self.theclass(1969, 12, 31, 23, 59, 58, 950000))
-
-    def test_utcnow(self):
-        import time
-
-        # Call it a success if utcnow() and utcfromtimestamp() are within
-        # a second of each other.
-        tolerance = timedelta(seconds=1)
-        for dummy in range(3):
-            from_now = self.theclass.utcnow()
-            from_timestamp = self.theclass.utcfromtimestamp(time.time())
-            if abs(from_timestamp - from_now) <= tolerance:
-                break
-            # Else try again a few times.
-        self.assertTrue(abs(from_timestamp - from_now) <= tolerance)
-
-    def test_strptime(self):
-        import _strptime
-
-        string = '2004-12-01 13:02:47.197'
-        format = '%Y-%m-%d %H:%M:%S.%f'
-        expected = _strptime._strptime_datetime(self.theclass, string, format)
-        got = self.theclass.strptime(string, format)
-        self.assertEqual(expected, got)
-        self.assertIs(type(expected), self.theclass)
-        self.assertIs(type(got), self.theclass)
-
-        strptime = self.theclass.strptime
-        self.assertEqual(strptime("+0002", "%z").utcoffset(), 2 * MINUTE)
-        self.assertEqual(strptime("-0002", "%z").utcoffset(), -2 * MINUTE)
-        # Only local timezone and UTC are supported
-        for tzseconds, tzname in ((0, 'UTC'), (0, 'GMT'),
-                                 (-_time.timezone, _time.tzname[0])):
-            if tzseconds < 0:
-                sign = '-'
-                seconds = -tzseconds
-            else:
-                sign ='+'
-                seconds = tzseconds
-            hours, minutes = divmod(seconds//60, 60)
-            dtstr = "{}{:02d}{:02d} {}".format(sign, hours, minutes, tzname)
-            dt = strptime(dtstr, "%z %Z")
-            self.assertEqual(dt.utcoffset(), timedelta(seconds=tzseconds))
-            self.assertEqual(dt.tzname(), tzname)
-        # Can produce inconsistent datetime
-        dtstr, fmt = "+1234 UTC", "%z %Z"
-        dt = strptime(dtstr, fmt)
-        self.assertEqual(dt.utcoffset(), 12 * HOUR + 34 * MINUTE)
-        self.assertEqual(dt.tzname(), 'UTC')
-        # yet will roundtrip
-        self.assertEqual(dt.strftime(fmt), dtstr)
-
-        # Produce naive datetime if no %z is provided
-        self.assertEqual(strptime("UTC", "%Z").tzinfo, None)
-
-        with self.assertRaises(ValueError): strptime("-2400", "%z")
-        with self.assertRaises(ValueError): strptime("-000", "%z")
-
-    def test_more_timetuple(self):
-        # This tests fields beyond those tested by the TestDate.test_timetuple.
-        t = self.theclass(2004, 12, 31, 6, 22, 33)
-        self.assertEqual(t.timetuple(), (2004, 12, 31, 6, 22, 33, 4, 366, -1))
-        self.assertEqual(t.timetuple(),
-                         (t.year, t.month, t.day,
-                          t.hour, t.minute, t.second,
-                          t.weekday(),
-                          t.toordinal() - date(t.year, 1, 1).toordinal() + 1,
-                          -1))
-        tt = t.timetuple()
-        self.assertEqual(tt.tm_year, t.year)
-        self.assertEqual(tt.tm_mon, t.month)
-        self.assertEqual(tt.tm_mday, t.day)
-        self.assertEqual(tt.tm_hour, t.hour)
-        self.assertEqual(tt.tm_min, t.minute)
-        self.assertEqual(tt.tm_sec, t.second)
-        self.assertEqual(tt.tm_wday, t.weekday())
-        self.assertEqual(tt.tm_yday, t.toordinal() -
-                                     date(t.year, 1, 1).toordinal() + 1)
-        self.assertEqual(tt.tm_isdst, -1)
-
-    def test_more_strftime(self):
-        # This tests fields beyond those tested by the TestDate.test_strftime.
-        t = self.theclass(2004, 12, 31, 6, 22, 33, 47)
-        self.assertEqual(t.strftime("%m %d %y %f %S %M %H %j"),
-                                    "12 31 04 000047 33 22 06 366")
-
-    def test_extract(self):
-        dt = self.theclass(2002, 3, 4, 18, 45, 3, 1234)
-        self.assertEqual(dt.date(), date(2002, 3, 4))
-        self.assertEqual(dt.time(), time(18, 45, 3, 1234))
-
-    def test_combine(self):
-        d = date(2002, 3, 4)
-        t = time(18, 45, 3, 1234)
-        expected = self.theclass(2002, 3, 4, 18, 45, 3, 1234)
-        combine = self.theclass.combine
-        dt = combine(d, t)
-        self.assertEqual(dt, expected)
-
-        dt = combine(time=t, date=d)
-        self.assertEqual(dt, expected)
-
-        self.assertEqual(d, dt.date())
-        self.assertEqual(t, dt.time())
-        self.assertEqual(dt, combine(dt.date(), dt.time()))
-
-        self.assertRaises(TypeError, combine) # need an arg
-        self.assertRaises(TypeError, combine, d) # need two args
-        self.assertRaises(TypeError, combine, t, d) # args reversed
-        self.assertRaises(TypeError, combine, d, t, 1) # too many args
-        self.assertRaises(TypeError, combine, "date", "time") # wrong types
-        self.assertRaises(TypeError, combine, d, "time") # wrong type
-        self.assertRaises(TypeError, combine, "date", t) # wrong type
-
-    def test_replace(self):
-        cls = self.theclass
-        args = [1, 2, 3, 4, 5, 6, 7]
-        base = cls(*args)
-        self.assertEqual(base, base.replace())
-
-        i = 0
-        for name, newval in (("year", 2),
-                             ("month", 3),
-                             ("day", 4),
-                             ("hour", 5),
-                             ("minute", 6),
-                             ("second", 7),
-                             ("microsecond", 8)):
-            newargs = args[:]
-            newargs[i] = newval
-            expected = cls(*newargs)
-            got = base.replace(**{name: newval})
-            self.assertEqual(expected, got)
-            i += 1
-
-        # Out of bounds.
-        base = cls(2000, 2, 29)
-        self.assertRaises(ValueError, base.replace, year=2001)
-
-    def test_astimezone(self):
-        # Pretty boring!  The TZ test is more interesting here.  astimezone()
-        # simply can't be applied to a naive object.
-        dt = self.theclass.now()
-        f = FixedOffset(44, "")
-        self.assertRaises(TypeError, dt.astimezone) # not enough args
-        self.assertRaises(TypeError, dt.astimezone, f, f) # too many args
-        self.assertRaises(TypeError, dt.astimezone, dt) # arg wrong type
-        self.assertRaises(ValueError, dt.astimezone, f) # naive
-        self.assertRaises(ValueError, dt.astimezone, tz=f)  # naive
-
-        class Bogus(tzinfo):
-            def utcoffset(self, dt): return None
-            def dst(self, dt): return timedelta(0)
-        bog = Bogus()
-        self.assertRaises(ValueError, dt.astimezone, bog)   # naive
-        self.assertRaises(ValueError,
-                          dt.replace(tzinfo=bog).astimezone, f)
-
-        class AlsoBogus(tzinfo):
-            def utcoffset(self, dt): return timedelta(0)
-            def dst(self, dt): return None
-        alsobog = AlsoBogus()
-        self.assertRaises(ValueError, dt.astimezone, alsobog) # also naive
-
-    def test_subclass_datetime(self):
-
-        class C(self.theclass):
-            theAnswer = 42
-
-            def __new__(cls, *args, **kws):
-                temp = kws.copy()
-                extra = temp.pop('extra')
-                result = self.theclass.__new__(cls, *args, **temp)
-                result.extra = extra
-                return result
-
-            def newmeth(self, start):
-                return start + self.year + self.month + self.second
-
-        args = 2003, 4, 14, 12, 13, 41
-
-        dt1 = self.theclass(*args)
-        dt2 = C(*args, **{'extra': 7})
-
-        self.assertEqual(dt2.__class__, C)
-        self.assertEqual(dt2.theAnswer, 42)
-        self.assertEqual(dt2.extra, 7)
-        self.assertEqual(dt1.toordinal(), dt2.toordinal())
-        self.assertEqual(dt2.newmeth(-7), dt1.year + dt1.month +
-                                          dt1.second - 7)
-
-class TestSubclassDateTime(TestDateTime):
-    theclass = SubclassDatetime
-    # Override tests not designed for subclass
-    def test_roundtrip(self):
-        pass
-
-class SubclassTime(time):
-    sub_var = 1
-
-class TestTime(HarmlessMixedComparison, unittest.TestCase):
-
-    theclass = time
-
-    def test_basic_attributes(self):
-        t = self.theclass(12, 0)
-        self.assertEqual(t.hour, 12)
-        self.assertEqual(t.minute, 0)
-        self.assertEqual(t.second, 0)
-        self.assertEqual(t.microsecond, 0)
-
-    def test_basic_attributes_nonzero(self):
-        # Make sure all attributes are non-zero so bugs in
-        # bit-shifting access show up.
-        t = self.theclass(12, 59, 59, 8000)
-        self.assertEqual(t.hour, 12)
-        self.assertEqual(t.minute, 59)
-        self.assertEqual(t.second, 59)
-        self.assertEqual(t.microsecond, 8000)
-
-    def test_roundtrip(self):
-        t = self.theclass(1, 2, 3, 4)
-
-        # Verify t -> string -> time identity.
-        s = repr(t)
-        self.assertTrue(s.startswith('datetime.'))
-        s = s[9:]
-        t2 = eval(s)
-        self.assertEqual(t, t2)
-
-        # Verify identity via reconstructing from pieces.
-        t2 = self.theclass(t.hour, t.minute, t.second,
-                           t.microsecond)
-        self.assertEqual(t, t2)
-
-    def test_comparing(self):
-        args = [1, 2, 3, 4]
-        t1 = self.theclass(*args)
-        t2 = self.theclass(*args)
-        self.assertEqual(t1, t2)
-        self.assertTrue(t1 <= t2)
-        self.assertTrue(t1 >= t2)
-        self.assertTrue(not t1 != t2)
-        self.assertTrue(not t1 < t2)
-        self.assertTrue(not t1 > t2)
-
-        for i in range(len(args)):
-            newargs = args[:]
-            newargs[i] = args[i] + 1
-            t2 = self.theclass(*newargs)   # this is larger than t1
-            self.assertTrue(t1 < t2)
-            self.assertTrue(t2 > t1)
-            self.assertTrue(t1 <= t2)
-            self.assertTrue(t2 >= t1)
-            self.assertTrue(t1 != t2)
-            self.assertTrue(t2 != t1)
-            self.assertTrue(not t1 == t2)
-            self.assertTrue(not t2 == t1)
-            self.assertTrue(not t1 > t2)
-            self.assertTrue(not t2 < t1)
-            self.assertTrue(not t1 >= t2)
-            self.assertTrue(not t2 <= t1)
-
-        for badarg in OTHERSTUFF:
-            self.assertEqual(t1 == badarg, False)
-            self.assertEqual(t1 != badarg, True)
-            self.assertEqual(badarg == t1, False)
-            self.assertEqual(badarg != t1, True)
-
-            self.assertRaises(TypeError, lambda: t1 <= badarg)
-            self.assertRaises(TypeError, lambda: t1 < badarg)
-            self.assertRaises(TypeError, lambda: t1 > badarg)
-            self.assertRaises(TypeError, lambda: t1 >= badarg)
-            self.assertRaises(TypeError, lambda: badarg <= t1)
-            self.assertRaises(TypeError, lambda: badarg < t1)
-            self.assertRaises(TypeError, lambda: badarg > t1)
-            self.assertRaises(TypeError, lambda: badarg >= t1)
-
-    def test_bad_constructor_arguments(self):
-        # bad hours
-        self.theclass(0, 0)    # no exception
-        self.theclass(23, 0)   # no exception
-        self.assertRaises(ValueError, self.theclass, -1, 0)
-        self.assertRaises(ValueError, self.theclass, 24, 0)
-        # bad minutes
-        self.theclass(23, 0)    # no exception
-        self.theclass(23, 59)   # no exception
-        self.assertRaises(ValueError, self.theclass, 23, -1)
-        self.assertRaises(ValueError, self.theclass, 23, 60)
-        # bad seconds
-        self.theclass(23, 59, 0)    # no exception
-        self.theclass(23, 59, 59)   # no exception
-        self.assertRaises(ValueError, self.theclass, 23, 59, -1)
-        self.assertRaises(ValueError, self.theclass, 23, 59, 60)
-        # bad microseconds
-        self.theclass(23, 59, 59, 0)        # no exception
-        self.theclass(23, 59, 59, 999999)   # no exception
-        self.assertRaises(ValueError, self.theclass, 23, 59, 59, -1)
-        self.assertRaises(ValueError, self.theclass, 23, 59, 59, 1000000)
-
-    def test_hash_equality(self):
-        d = self.theclass(23, 30, 17)
-        e = self.theclass(23, 30, 17)
-        self.assertEqual(d, e)
-        self.assertEqual(hash(d), hash(e))
-
-        dic = {d: 1}
-        dic[e] = 2
-        self.assertEqual(len(dic), 1)
-        self.assertEqual(dic[d], 2)
-        self.assertEqual(dic[e], 2)
-
-        d = self.theclass(0,  5, 17)
-        e = self.theclass(0,  5, 17)
-        self.assertEqual(d, e)
-        self.assertEqual(hash(d), hash(e))
-
-        dic = {d: 1}
-        dic[e] = 2
-        self.assertEqual(len(dic), 1)
-        self.assertEqual(dic[d], 2)
-        self.assertEqual(dic[e], 2)
-
-    def test_isoformat(self):
-        t = self.theclass(4, 5, 1, 123)
-        self.assertEqual(t.isoformat(), "04:05:01.000123")
-        self.assertEqual(t.isoformat(), str(t))
-
-        t = self.theclass()
-        self.assertEqual(t.isoformat(), "00:00:00")
-        self.assertEqual(t.isoformat(), str(t))
-
-        t = self.theclass(microsecond=1)
-        self.assertEqual(t.isoformat(), "00:00:00.000001")
-        self.assertEqual(t.isoformat(), str(t))
-
-        t = self.theclass(microsecond=10)
-        self.assertEqual(t.isoformat(), "00:00:00.000010")
-        self.assertEqual(t.isoformat(), str(t))
-
-        t = self.theclass(microsecond=100)
-        self.assertEqual(t.isoformat(), "00:00:00.000100")
-        self.assertEqual(t.isoformat(), str(t))
-
-        t = self.theclass(microsecond=1000)
-        self.assertEqual(t.isoformat(), "00:00:00.001000")
-        self.assertEqual(t.isoformat(), str(t))
-
-        t = self.theclass(microsecond=10000)
-        self.assertEqual(t.isoformat(), "00:00:00.010000")
-        self.assertEqual(t.isoformat(), str(t))
-
-        t = self.theclass(microsecond=100000)
-        self.assertEqual(t.isoformat(), "00:00:00.100000")
-        self.assertEqual(t.isoformat(), str(t))
-
-    def test_1653736(self):
-        # verify it doesn't accept extra keyword arguments
-        t = self.theclass(second=1)
-        self.assertRaises(TypeError, t.isoformat, foo=3)
-
-    def test_strftime(self):
-        t = self.theclass(1, 2, 3, 4)
-        self.assertEqual(t.strftime('%H %M %S %f'), "01 02 03 000004")
-        # A naive object replaces %z and %Z with empty strings.
-        self.assertEqual(t.strftime("'%z' '%Z'"), "'' ''")
-
-    def test_format(self):
-        t = self.theclass(1, 2, 3, 4)
-        self.assertEqual(t.__format__(''), str(t))
-
-        # check that a derived class's __str__() gets called
-        class A(self.theclass):
-            def __str__(self):
-                return 'A'
-        a = A(1, 2, 3, 4)
-        self.assertEqual(a.__format__(''), 'A')
-
-        # check that a derived class's strftime gets called
-        class B(self.theclass):
-            def strftime(self, format_spec):
-                return 'B'
-        b = B(1, 2, 3, 4)
-        self.assertEqual(b.__format__(''), str(t))
-
-        for fmt in ['%H %M %S',
-                    ]:
-            self.assertEqual(t.__format__(fmt), t.strftime(fmt))
-            self.assertEqual(a.__format__(fmt), t.strftime(fmt))
-            self.assertEqual(b.__format__(fmt), 'B')
-
-    def test_str(self):
-        self.assertEqual(str(self.theclass(1, 2, 3, 4)), "01:02:03.000004")
-        self.assertEqual(str(self.theclass(10, 2, 3, 4000)), "10:02:03.004000")
-        self.assertEqual(str(self.theclass(0, 2, 3, 400000)), "00:02:03.400000")
-        self.assertEqual(str(self.theclass(12, 2, 3, 0)), "12:02:03")
-        self.assertEqual(str(self.theclass(23, 15, 0, 0)), "23:15:00")
-
-    def test_repr(self):
-        name = 'datetime.' + self.theclass.__name__
-        self.assertEqual(repr(self.theclass(1, 2, 3, 4)),
-                         "%s(1, 2, 3, 4)" % name)
-        self.assertEqual(repr(self.theclass(10, 2, 3, 4000)),
-                         "%s(10, 2, 3, 4000)" % name)
-        self.assertEqual(repr(self.theclass(0, 2, 3, 400000)),
-                         "%s(0, 2, 3, 400000)" % name)
-        self.assertEqual(repr(self.theclass(12, 2, 3, 0)),
-                         "%s(12, 2, 3)" % name)
-        self.assertEqual(repr(self.theclass(23, 15, 0, 0)),
-                         "%s(23, 15)" % name)
-
-    def test_resolution_info(self):
-        self.assertIsInstance(self.theclass.min, self.theclass)
-        self.assertIsInstance(self.theclass.max, self.theclass)
-        self.assertIsInstance(self.theclass.resolution, timedelta)
-        self.assertTrue(self.theclass.max > self.theclass.min)
-
-    def test_pickling(self):
-        args = 20, 59, 16, 64**2
-        orig = self.theclass(*args)
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertEqual(orig, derived)
-
-    def test_pickling_subclass_time(self):
-        args = 20, 59, 16, 64**2
-        orig = SubclassTime(*args)
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertEqual(orig, derived)
-
-    def test_bool(self):
-        cls = self.theclass
-        self.assertTrue(cls(1))
-        self.assertTrue(cls(0, 1))
-        self.assertTrue(cls(0, 0, 1))
-        self.assertTrue(cls(0, 0, 0, 1))
-        self.assertTrue(not cls(0))
-        self.assertTrue(not cls())
-
-    def test_replace(self):
-        cls = self.theclass
-        args = [1, 2, 3, 4]
-        base = cls(*args)
-        self.assertEqual(base, base.replace())
-
-        i = 0
-        for name, newval in (("hour", 5),
-                             ("minute", 6),
-                             ("second", 7),
-                             ("microsecond", 8)):
-            newargs = args[:]
-            newargs[i] = newval
-            expected = cls(*newargs)
-            got = base.replace(**{name: newval})
-            self.assertEqual(expected, got)
-            i += 1
-
-        # Out of bounds.
-        base = cls(1)
-        self.assertRaises(ValueError, base.replace, hour=24)
-        self.assertRaises(ValueError, base.replace, minute=-1)
-        self.assertRaises(ValueError, base.replace, second=100)
-        self.assertRaises(ValueError, base.replace, microsecond=1000000)
-
-    def test_subclass_time(self):
-
-        class C(self.theclass):
-            theAnswer = 42
-
-            def __new__(cls, *args, **kws):
-                temp = kws.copy()
-                extra = temp.pop('extra')
-                result = self.theclass.__new__(cls, *args, **temp)
-                result.extra = extra
-                return result
-
-            def newmeth(self, start):
-                return start + self.hour + self.second
-
-        args = 4, 5, 6
-
-        dt1 = self.theclass(*args)
-        dt2 = C(*args, **{'extra': 7})
-
-        self.assertEqual(dt2.__class__, C)
-        self.assertEqual(dt2.theAnswer, 42)
-        self.assertEqual(dt2.extra, 7)
-        self.assertEqual(dt1.isoformat(), dt2.isoformat())
-        self.assertEqual(dt2.newmeth(-7), dt1.hour + dt1.second - 7)
-
-    def test_backdoor_resistance(self):
-        # see TestDate.test_backdoor_resistance().
-        base = '2:59.0'
-        for hour_byte in ' ', '9', chr(24), '\xff':
-            self.assertRaises(TypeError, self.theclass,
-                                         hour_byte + base[1:])
-
-# A mixin for classes with a tzinfo= argument.  Subclasses must define
-# theclass as a class atribute, and theclass(1, 1, 1, tzinfo=whatever)
-# must be legit (which is true for time and datetime).
-class TZInfoBase:
-
-    def test_argument_passing(self):
-        cls = self.theclass
-        # A datetime passes itself on, a time passes None.
-        class introspective(tzinfo):
-            def tzname(self, dt):    return dt and "real" or "none"
-            def utcoffset(self, dt):
-                return timedelta(minutes = dt and 42 or -42)
-            dst = utcoffset
-
-        obj = cls(1, 2, 3, tzinfo=introspective())
-
-        expected = cls is time and "none" or "real"
-        self.assertEqual(obj.tzname(), expected)
-
-        expected = timedelta(minutes=(cls is time and -42 or 42))
-        self.assertEqual(obj.utcoffset(), expected)
-        self.assertEqual(obj.dst(), expected)
-
-    def test_bad_tzinfo_classes(self):
-        cls = self.theclass
-        self.assertRaises(TypeError, cls, 1, 1, 1, tzinfo=12)
-
-        class NiceTry(object):
-            def __init__(self): pass
-            def utcoffset(self, dt): pass
-        self.assertRaises(TypeError, cls, 1, 1, 1, tzinfo=NiceTry)
-
-        class BetterTry(tzinfo):
-            def __init__(self): pass
-            def utcoffset(self, dt): pass
-        b = BetterTry()
-        t = cls(1, 1, 1, tzinfo=b)
-        self.assertTrue(t.tzinfo is b)
-
-    def test_utc_offset_out_of_bounds(self):
-        class Edgy(tzinfo):
-            def __init__(self, offset):
-                self.offset = timedelta(minutes=offset)
-            def utcoffset(self, dt):
-                return self.offset
-
-        cls = self.theclass
-        for offset, legit in ((-1440, False),
-                              (-1439, True),
-                              (1439, True),
-                              (1440, False)):
-            if cls is time:
-                t = cls(1, 2, 3, tzinfo=Edgy(offset))
-            elif cls is datetime:
-                t = cls(6, 6, 6, 1, 2, 3, tzinfo=Edgy(offset))
-            else:
-                assert 0, "impossible"
-            if legit:
-                aofs = abs(offset)
-                h, m = divmod(aofs, 60)
-                tag = "%c%02d:%02d" % (offset < 0 and '-' or '+', h, m)
-                if isinstance(t, datetime):
-                    t = t.timetz()
-                self.assertEqual(str(t), "01:02:03" + tag)
-            else:
-                self.assertRaises(ValueError, str, t)
-
-    def test_tzinfo_classes(self):
-        cls = self.theclass
-        class C1(tzinfo):
-            def utcoffset(self, dt): return None
-            def dst(self, dt): return None
-            def tzname(self, dt): return None
-        for t in (cls(1, 1, 1),
-                  cls(1, 1, 1, tzinfo=None),
-                  cls(1, 1, 1, tzinfo=C1())):
-            self.assertTrue(t.utcoffset() is None)
-            self.assertTrue(t.dst() is None)
-            self.assertTrue(t.tzname() is None)
-
-        class C3(tzinfo):
-            def utcoffset(self, dt): return timedelta(minutes=-1439)
-            def dst(self, dt): return timedelta(minutes=1439)
-            def tzname(self, dt): return "aname"
-        t = cls(1, 1, 1, tzinfo=C3())
-        self.assertEqual(t.utcoffset(), timedelta(minutes=-1439))
-        self.assertEqual(t.dst(), timedelta(minutes=1439))
-        self.assertEqual(t.tzname(), "aname")
-
-        # Wrong types.
-        class C4(tzinfo):
-            def utcoffset(self, dt): return "aname"
-            def dst(self, dt): return 7
-            def tzname(self, dt): return 0
-        t = cls(1, 1, 1, tzinfo=C4())
-        self.assertRaises(TypeError, t.utcoffset)
-        self.assertRaises(TypeError, t.dst)
-        self.assertRaises(TypeError, t.tzname)
-
-        # Offset out of range.
-        class C6(tzinfo):
-            def utcoffset(self, dt): return timedelta(hours=-24)
-            def dst(self, dt): return timedelta(hours=24)
-        t = cls(1, 1, 1, tzinfo=C6())
-        self.assertRaises(ValueError, t.utcoffset)
-        self.assertRaises(ValueError, t.dst)
-
-        # Not a whole number of minutes.
-        class C7(tzinfo):
-            def utcoffset(self, dt): return timedelta(seconds=61)
-            def dst(self, dt): return timedelta(microseconds=-81)
-        t = cls(1, 1, 1, tzinfo=C7())
-        self.assertRaises(ValueError, t.utcoffset)
-        self.assertRaises(ValueError, t.dst)
-
-    def test_aware_compare(self):
-        cls = self.theclass
-
-        # Ensure that utcoffset() gets ignored if the comparands have
-        # the same tzinfo member.
-        class OperandDependentOffset(tzinfo):
-            def utcoffset(self, t):
-                if t.minute < 10:
-                    # d0 and d1 equal after adjustment
-                    return timedelta(minutes=t.minute)
-                else:
-                    # d2 off in the weeds
-                    return timedelta(minutes=59)
-
-        base = cls(8, 9, 10, tzinfo=OperandDependentOffset())
-        d0 = base.replace(minute=3)
-        d1 = base.replace(minute=9)
-        d2 = base.replace(minute=11)
-        for x in d0, d1, d2:
-            for y in d0, d1, d2:
-                for op in lt, le, gt, ge, eq, ne:
-                    got = op(x, y)
-                    expected = op(x.minute, y.minute)
-                    self.assertEqual(got, expected)
-
-        # However, if they're different members, uctoffset is not ignored.
-        # Note that a time can't actually have an operand-depedent offset,
-        # though (and time.utcoffset() passes None to tzinfo.utcoffset()),
-        # so skip this test for time.
-        if cls is not time:
-            d0 = base.replace(minute=3, tzinfo=OperandDependentOffset())
-            d1 = base.replace(minute=9, tzinfo=OperandDependentOffset())
-            d2 = base.replace(minute=11, tzinfo=OperandDependentOffset())
-            for x in d0, d1, d2:
-                for y in d0, d1, d2:
-                    got = (x > y) - (x < y)
-                    if (x is d0 or x is d1) and (y is d0 or y is d1):
-                        expected = 0
-                    elif x is y is d2:
-                        expected = 0
-                    elif x is d2:
-                        expected = -1
-                    else:
-                        assert y is d2
-                        expected = 1
-                    self.assertEqual(got, expected)
-
-
-# Testing time objects with a non-None tzinfo.
-class TestTimeTZ(TestTime, TZInfoBase, unittest.TestCase):
-    theclass = time
-
-    def test_empty(self):
-        t = self.theclass()
-        self.assertEqual(t.hour, 0)
-        self.assertEqual(t.minute, 0)
-        self.assertEqual(t.second, 0)
-        self.assertEqual(t.microsecond, 0)
-        self.assertTrue(t.tzinfo is None)
-
-    def test_zones(self):
-        est = FixedOffset(-300, "EST", 1)
-        utc = FixedOffset(0, "UTC", -2)
-        met = FixedOffset(60, "MET", 3)
-        t1 = time( 7, 47, tzinfo=est)
-        t2 = time(12, 47, tzinfo=utc)
-        t3 = time(13, 47, tzinfo=met)
-        t4 = time(microsecond=40)
-        t5 = time(microsecond=40, tzinfo=utc)
-
-        self.assertEqual(t1.tzinfo, est)
-        self.assertEqual(t2.tzinfo, utc)
-        self.assertEqual(t3.tzinfo, met)
-        self.assertTrue(t4.tzinfo is None)
-        self.assertEqual(t5.tzinfo, utc)
-
-        self.assertEqual(t1.utcoffset(), timedelta(minutes=-300))
-        self.assertEqual(t2.utcoffset(), timedelta(minutes=0))
-        self.assertEqual(t3.utcoffset(), timedelta(minutes=60))
-        self.assertTrue(t4.utcoffset() is None)
-        self.assertRaises(TypeError, t1.utcoffset, "no args")
-
-        self.assertEqual(t1.tzname(), "EST")
-        self.assertEqual(t2.tzname(), "UTC")
-        self.assertEqual(t3.tzname(), "MET")
-        self.assertTrue(t4.tzname() is None)
-        self.assertRaises(TypeError, t1.tzname, "no args")
-
-        self.assertEqual(t1.dst(), timedelta(minutes=1))
-        self.assertEqual(t2.dst(), timedelta(minutes=-2))
-        self.assertEqual(t3.dst(), timedelta(minutes=3))
-        self.assertTrue(t4.dst() is None)
-        self.assertRaises(TypeError, t1.dst, "no args")
-
-        self.assertEqual(hash(t1), hash(t2))
-        self.assertEqual(hash(t1), hash(t3))
-        self.assertEqual(hash(t2), hash(t3))
-
-        self.assertEqual(t1, t2)
-        self.assertEqual(t1, t3)
-        self.assertEqual(t2, t3)
-        self.assertRaises(TypeError, lambda: t4 == t5) # mixed tz-aware & naive
-        self.assertRaises(TypeError, lambda: t4 < t5) # mixed tz-aware & naive
-        self.assertRaises(TypeError, lambda: t5 < t4) # mixed tz-aware & naive
-
-        self.assertEqual(str(t1), "07:47:00-05:00")
-        self.assertEqual(str(t2), "12:47:00+00:00")
-        self.assertEqual(str(t3), "13:47:00+01:00")
-        self.assertEqual(str(t4), "00:00:00.000040")
-        self.assertEqual(str(t5), "00:00:00.000040+00:00")
-
-        self.assertEqual(t1.isoformat(), "07:47:00-05:00")
-        self.assertEqual(t2.isoformat(), "12:47:00+00:00")
-        self.assertEqual(t3.isoformat(), "13:47:00+01:00")
-        self.assertEqual(t4.isoformat(), "00:00:00.000040")
-        self.assertEqual(t5.isoformat(), "00:00:00.000040+00:00")
-
-        d = 'datetime.time'
-        self.assertEqual(repr(t1), d + "(7, 47, tzinfo=est)")
-        self.assertEqual(repr(t2), d + "(12, 47, tzinfo=utc)")
-        self.assertEqual(repr(t3), d + "(13, 47, tzinfo=met)")
-        self.assertEqual(repr(t4), d + "(0, 0, 0, 40)")
-        self.assertEqual(repr(t5), d + "(0, 0, 0, 40, tzinfo=utc)")
-
-        self.assertEqual(t1.strftime("%H:%M:%S %%Z=%Z %%z=%z"),
-                                     "07:47:00 %Z=EST %z=-0500")
-        self.assertEqual(t2.strftime("%H:%M:%S %Z %z"), "12:47:00 UTC +0000")
-        self.assertEqual(t3.strftime("%H:%M:%S %Z %z"), "13:47:00 MET +0100")
-
-        yuck = FixedOffset(-1439, "%z %Z %%z%%Z")
-        t1 = time(23, 59, tzinfo=yuck)
-        self.assertEqual(t1.strftime("%H:%M %%Z='%Z' %%z='%z'"),
-                                     "23:59 %Z='%z %Z %%z%%Z' %z='-2359'")
-
-        # Check that an invalid tzname result raises an exception.
-        class Badtzname(tzinfo):
-            def tzname(self, dt): return 42
-        t = time(2, 3, 4, tzinfo=Badtzname())
-        self.assertEqual(t.strftime("%H:%M:%S"), "02:03:04")
-        self.assertRaises(TypeError, t.strftime, "%Z")
-
-    def test_hash_edge_cases(self):
-        # Offsets that overflow a basic time.
-        t1 = self.theclass(0, 1, 2, 3, tzinfo=FixedOffset(1439, ""))
-        t2 = self.theclass(0, 0, 2, 3, tzinfo=FixedOffset(1438, ""))
-        self.assertEqual(hash(t1), hash(t2))
-
-        t1 = self.theclass(23, 58, 6, 100, tzinfo=FixedOffset(-1000, ""))
-        t2 = self.theclass(23, 48, 6, 100, tzinfo=FixedOffset(-1010, ""))
-        self.assertEqual(hash(t1), hash(t2))
-
-    def test_pickling(self):
-        # Try one without a tzinfo.
-        args = 20, 59, 16, 64**2
-        orig = self.theclass(*args)
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertEqual(orig, derived)
-
-        # Try one with a tzinfo.
-        tinfo = PicklableFixedOffset(-300, 'cookie')
-        orig = self.theclass(5, 6, 7, tzinfo=tinfo)
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertEqual(orig, derived)
-            self.assertIsInstance(derived.tzinfo, PicklableFixedOffset)
-            self.assertEqual(derived.utcoffset(), timedelta(minutes=-300))
-            self.assertEqual(derived.tzname(), 'cookie')
-
-    def test_more_bool(self):
-        # Test cases with non-None tzinfo.
-        cls = self.theclass
-
-        t = cls(0, tzinfo=FixedOffset(-300, ""))
-        self.assertTrue(t)
-
-        t = cls(5, tzinfo=FixedOffset(-300, ""))
-        self.assertTrue(t)
-
-        t = cls(5, tzinfo=FixedOffset(300, ""))
-        self.assertTrue(not t)
-
-        t = cls(23, 59, tzinfo=FixedOffset(23*60 + 59, ""))
-        self.assertTrue(not t)
-
-        # Mostly ensuring this doesn't overflow internally.
-        t = cls(0, tzinfo=FixedOffset(23*60 + 59, ""))
-        self.assertTrue(t)
-
-        # But this should yield a value error -- the utcoffset is bogus.
-        t = cls(0, tzinfo=FixedOffset(24*60, ""))
-        self.assertRaises(ValueError, lambda: bool(t))
-
-        # Likewise.
-        t = cls(0, tzinfo=FixedOffset(-24*60, ""))
-        self.assertRaises(ValueError, lambda: bool(t))
-
-    def test_replace(self):
-        cls = self.theclass
-        z100 = FixedOffset(100, "+100")
-        zm200 = FixedOffset(timedelta(minutes=-200), "-200")
-        args = [1, 2, 3, 4, z100]
-        base = cls(*args)
-        self.assertEqual(base, base.replace())
-
-        i = 0
-        for name, newval in (("hour", 5),
-                             ("minute", 6),
-                             ("second", 7),
-                             ("microsecond", 8),
-                             ("tzinfo", zm200)):
-            newargs = args[:]
-            newargs[i] = newval
-            expected = cls(*newargs)
-            got = base.replace(**{name: newval})
-            self.assertEqual(expected, got)
-            i += 1
-
-        # Ensure we can get rid of a tzinfo.
-        self.assertEqual(base.tzname(), "+100")
-        base2 = base.replace(tzinfo=None)
-        self.assertTrue(base2.tzinfo is None)
-        self.assertTrue(base2.tzname() is None)
-
-        # Ensure we can add one.
-        base3 = base2.replace(tzinfo=z100)
-        self.assertEqual(base, base3)
-        self.assertTrue(base.tzinfo is base3.tzinfo)
-
-        # Out of bounds.
-        base = cls(1)
-        self.assertRaises(ValueError, base.replace, hour=24)
-        self.assertRaises(ValueError, base.replace, minute=-1)
-        self.assertRaises(ValueError, base.replace, second=100)
-        self.assertRaises(ValueError, base.replace, microsecond=1000000)
-
-    def test_mixed_compare(self):
-        t1 = time(1, 2, 3)
-        t2 = time(1, 2, 3)
-        self.assertEqual(t1, t2)
-        t2 = t2.replace(tzinfo=None)
-        self.assertEqual(t1, t2)
-        t2 = t2.replace(tzinfo=FixedOffset(None, ""))
-        self.assertEqual(t1, t2)
-        t2 = t2.replace(tzinfo=FixedOffset(0, ""))
-        self.assertRaises(TypeError, lambda: t1 == t2)
-
-        # In time w/ identical tzinfo objects, utcoffset is ignored.
-        class Varies(tzinfo):
-            def __init__(self):
-                self.offset = timedelta(minutes=22)
-            def utcoffset(self, t):
-                self.offset += timedelta(minutes=1)
-                return self.offset
-
-        v = Varies()
-        t1 = t2.replace(tzinfo=v)
-        t2 = t2.replace(tzinfo=v)
-        self.assertEqual(t1.utcoffset(), timedelta(minutes=23))
-        self.assertEqual(t2.utcoffset(), timedelta(minutes=24))
-        self.assertEqual(t1, t2)
-
-        # But if they're not identical, it isn't ignored.
-        t2 = t2.replace(tzinfo=Varies())
-        self.assertTrue(t1 < t2)  # t1's offset counter still going up
-
-    def test_subclass_timetz(self):
-
-        class C(self.theclass):
-            theAnswer = 42
-
-            def __new__(cls, *args, **kws):
-                temp = kws.copy()
-                extra = temp.pop('extra')
-                result = self.theclass.__new__(cls, *args, **temp)
-                result.extra = extra
-                return result
-
-            def newmeth(self, start):
-                return start + self.hour + self.second
-
-        args = 4, 5, 6, 500, FixedOffset(-300, "EST", 1)
-
-        dt1 = self.theclass(*args)
-        dt2 = C(*args, **{'extra': 7})
-
-        self.assertEqual(dt2.__class__, C)
-        self.assertEqual(dt2.theAnswer, 42)
-        self.assertEqual(dt2.extra, 7)
-        self.assertEqual(dt1.utcoffset(), dt2.utcoffset())
-        self.assertEqual(dt2.newmeth(-7), dt1.hour + dt1.second - 7)
-
-
-# Testing datetime objects with a non-None tzinfo.
-
-class TestDateTimeTZ(TestDateTime, TZInfoBase, unittest.TestCase):
-    theclass = datetime
-
-    def test_trivial(self):
-        dt = self.theclass(1, 2, 3, 4, 5, 6, 7)
-        self.assertEqual(dt.year, 1)
-        self.assertEqual(dt.month, 2)
-        self.assertEqual(dt.day, 3)
-        self.assertEqual(dt.hour, 4)
-        self.assertEqual(dt.minute, 5)
-        self.assertEqual(dt.second, 6)
-        self.assertEqual(dt.microsecond, 7)
-        self.assertEqual(dt.tzinfo, None)
-
-    def test_even_more_compare(self):
-        # The test_compare() and test_more_compare() inherited from TestDate
-        # and TestDateTime covered non-tzinfo cases.
-
-        # Smallest possible after UTC adjustment.
-        t1 = self.theclass(1, 1, 1, tzinfo=FixedOffset(1439, ""))
-        # Largest possible after UTC adjustment.
-        t2 = self.theclass(MAXYEAR, 12, 31, 23, 59, 59, 999999,
-                           tzinfo=FixedOffset(-1439, ""))
-
-        # Make sure those compare correctly, and w/o overflow.
-        self.assertTrue(t1 < t2)
-        self.assertTrue(t1 != t2)
-        self.assertTrue(t2 > t1)
-
-        self.assertEqual(t1, t1)
-        self.assertEqual(t2, t2)
-
-        # Equal afer adjustment.
-        t1 = self.theclass(1, 12, 31, 23, 59, tzinfo=FixedOffset(1, ""))
-        t2 = self.theclass(2, 1, 1, 3, 13, tzinfo=FixedOffset(3*60+13+2, ""))
-        self.assertEqual(t1, t2)
-
-        # Change t1 not to subtract a minute, and t1 should be larger.
-        t1 = self.theclass(1, 12, 31, 23, 59, tzinfo=FixedOffset(0, ""))
-        self.assertTrue(t1 > t2)
-
-        # Change t1 to subtract 2 minutes, and t1 should be smaller.
-        t1 = self.theclass(1, 12, 31, 23, 59, tzinfo=FixedOffset(2, ""))
-        self.assertTrue(t1 < t2)
-
-        # Back to the original t1, but make seconds resolve it.
-        t1 = self.theclass(1, 12, 31, 23, 59, tzinfo=FixedOffset(1, ""),
-                           second=1)
-        self.assertTrue(t1 > t2)
-
-        # Likewise, but make microseconds resolve it.
-        t1 = self.theclass(1, 12, 31, 23, 59, tzinfo=FixedOffset(1, ""),
-                           microsecond=1)
-        self.assertTrue(t1 > t2)
-
-        # Make t2 naive and it should fail.
-        t2 = self.theclass.min
-        self.assertRaises(TypeError, lambda: t1 == t2)
-        self.assertEqual(t2, t2)
-
-        # It's also naive if it has tzinfo but tzinfo.utcoffset() is None.
-        class Naive(tzinfo):
-            def utcoffset(self, dt): return None
-        t2 = self.theclass(5, 6, 7, tzinfo=Naive())
-        self.assertRaises(TypeError, lambda: t1 == t2)
-        self.assertEqual(t2, t2)
-
-        # OTOH, it's OK to compare two of these mixing the two ways of being
-        # naive.
-        t1 = self.theclass(5, 6, 7)
-        self.assertEqual(t1, t2)
-
-        # Try a bogus uctoffset.
-        class Bogus(tzinfo):
-            def utcoffset(self, dt):
-                return timedelta(minutes=1440) # out of bounds
-        t1 = self.theclass(2, 2, 2, tzinfo=Bogus())
-        t2 = self.theclass(2, 2, 2, tzinfo=FixedOffset(0, ""))
-        self.assertRaises(ValueError, lambda: t1 == t2)
-
-    def test_pickling(self):
-        # Try one without a tzinfo.
-        args = 6, 7, 23, 20, 59, 1, 64**2
-        orig = self.theclass(*args)
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertEqual(orig, derived)
-
-        # Try one with a tzinfo.
-        tinfo = PicklableFixedOffset(-300, 'cookie')
-        orig = self.theclass(*args, **{'tzinfo': tinfo})
-        derived = self.theclass(1, 1, 1, tzinfo=FixedOffset(0, "", 0))
-        for pickler, unpickler, proto in pickle_choices:
-            green = pickler.dumps(orig, proto)
-            derived = unpickler.loads(green)
-            self.assertEqual(orig, derived)
-            self.assertIsInstance(derived.tzinfo, PicklableFixedOffset)
-            self.assertEqual(derived.utcoffset(), timedelta(minutes=-300))
-            self.assertEqual(derived.tzname(), 'cookie')
-
-    def test_extreme_hashes(self):
-        # If an attempt is made to hash these via subtracting the offset
-        # then hashing a datetime object, OverflowError results.  The
-        # Python implementation used to blow up here.
-        t = self.theclass(1, 1, 1, tzinfo=FixedOffset(1439, ""))
-        hash(t)
-        t = self.theclass(MAXYEAR, 12, 31, 23, 59, 59, 999999,
-                          tzinfo=FixedOffset(-1439, ""))
-        hash(t)
-
-        # OTOH, an OOB offset should blow up.
-        t = self.theclass(5, 5, 5, tzinfo=FixedOffset(-1440, ""))
-        self.assertRaises(ValueError, hash, t)
-
-    def test_zones(self):
-        est = FixedOffset(-300, "EST")
-        utc = FixedOffset(0, "UTC")
-        met = FixedOffset(60, "MET")
-        t1 = datetime(2002, 3, 19,  7, 47, tzinfo=est)
-        t2 = datetime(2002, 3, 19, 12, 47, tzinfo=utc)
-        t3 = datetime(2002, 3, 19, 13, 47, tzinfo=met)
-        self.assertEqual(t1.tzinfo, est)
-        self.assertEqual(t2.tzinfo, utc)
-        self.assertEqual(t3.tzinfo, met)
-        self.assertEqual(t1.utcoffset(), timedelta(minutes=-300))
-        self.assertEqual(t2.utcoffset(), timedelta(minutes=0))
-        self.assertEqual(t3.utcoffset(), timedelta(minutes=60))
-        self.assertEqual(t1.tzname(), "EST")
-        self.assertEqual(t2.tzname(), "UTC")
-        self.assertEqual(t3.tzname(), "MET")
-        self.assertEqual(hash(t1), hash(t2))
-        self.assertEqual(hash(t1), hash(t3))
-        self.assertEqual(hash(t2), hash(t3))
-        self.assertEqual(t1, t2)
-        self.assertEqual(t1, t3)
-        self.assertEqual(t2, t3)
-        self.assertEqual(str(t1), "2002-03-19 07:47:00-05:00")
-        self.assertEqual(str(t2), "2002-03-19 12:47:00+00:00")
-        self.assertEqual(str(t3), "2002-03-19 13:47:00+01:00")
-        d = 'datetime.datetime(2002, 3, 19, '
-        self.assertEqual(repr(t1), d + "7, 47, tzinfo=est)")
-        self.assertEqual(repr(t2), d + "12, 47, tzinfo=utc)")
-        self.assertEqual(repr(t3), d + "13, 47, tzinfo=met)")
-
-    def test_combine(self):
-        met = FixedOffset(60, "MET")
-        d = date(2002, 3, 4)
-        tz = time(18, 45, 3, 1234, tzinfo=met)
-        dt = datetime.combine(d, tz)
-        self.assertEqual(dt, datetime(2002, 3, 4, 18, 45, 3, 1234,
-                                        tzinfo=met))
-
-    def test_extract(self):
-        met = FixedOffset(60, "MET")
-        dt = self.theclass(2002, 3, 4, 18, 45, 3, 1234, tzinfo=met)
-        self.assertEqual(dt.date(), date(2002, 3, 4))
-        self.assertEqual(dt.time(), time(18, 45, 3, 1234))
-        self.assertEqual(dt.timetz(), time(18, 45, 3, 1234, tzinfo=met))
-
-    def test_tz_aware_arithmetic(self):
-        import random
-
-        now = self.theclass.now()
-        tz55 = FixedOffset(-330, "west 5:30")
-        timeaware = now.time().replace(tzinfo=tz55)
-        nowaware = self.theclass.combine(now.date(), timeaware)
-        self.assertTrue(nowaware.tzinfo is tz55)
-        self.assertEqual(nowaware.timetz(), timeaware)
-
-        # Can't mix aware and non-aware.
-        self.assertRaises(TypeError, lambda: now - nowaware)
-        self.assertRaises(TypeError, lambda: nowaware - now)
-
-        # And adding datetime's doesn't make sense, aware or not.
-        self.assertRaises(TypeError, lambda: now + nowaware)
-        self.assertRaises(TypeError, lambda: nowaware + now)
-        self.assertRaises(TypeError, lambda: nowaware + nowaware)
-
-        # Subtracting should yield 0.
-        self.assertEqual(now - now, timedelta(0))
-        self.assertEqual(nowaware - nowaware, timedelta(0))
-
-        # Adding a delta should preserve tzinfo.
-        delta = timedelta(weeks=1, minutes=12, microseconds=5678)
-        nowawareplus = nowaware + delta
-        self.assertTrue(nowaware.tzinfo is tz55)
-        nowawareplus2 = delta + nowaware
-        self.assertTrue(nowawareplus2.tzinfo is tz55)
-        self.assertEqual(nowawareplus, nowawareplus2)
-
-        # that - delta should be what we started with, and that - what we
-        # started with should be delta.
-        diff = nowawareplus - delta
-        self.assertTrue(diff.tzinfo is tz55)
-        self.assertEqual(nowaware, diff)
-        self.assertRaises(TypeError, lambda: delta - nowawareplus)
-        self.assertEqual(nowawareplus - nowaware, delta)
-
-        # Make up a random timezone.
-        tzr = FixedOffset(random.randrange(-1439, 1440), "randomtimezone")
-        # Attach it to nowawareplus.
-        nowawareplus = nowawareplus.replace(tzinfo=tzr)
-        self.assertTrue(nowawareplus.tzinfo is tzr)
-        # Make sure the difference takes the timezone adjustments into account.
-        got = nowaware - nowawareplus
-        # Expected:  (nowaware base - nowaware offset) -
-        #            (nowawareplus base - nowawareplus offset) =
-        #            (nowaware base - nowawareplus base) +
-        #            (nowawareplus offset - nowaware offset) =
-        #            -delta + nowawareplus offset - nowaware offset
-        expected = nowawareplus.utcoffset() - nowaware.utcoffset() - delta
-        self.assertEqual(got, expected)
-
-        # Try max possible difference.
-        min = self.theclass(1, 1, 1, tzinfo=FixedOffset(1439, "min"))
-        max = self.theclass(MAXYEAR, 12, 31, 23, 59, 59, 999999,
-                            tzinfo=FixedOffset(-1439, "max"))
-        maxdiff = max - min
-        self.assertEqual(maxdiff, self.theclass.max - self.theclass.min +
-                                  timedelta(minutes=2*1439))
-        # Different tzinfo, but the same offset
-        tza = timezone(HOUR, 'A')
-        tzb = timezone(HOUR, 'B')
-        delta = min.replace(tzinfo=tza) - max.replace(tzinfo=tzb)
-        self.assertEqual(delta, self.theclass.min - self.theclass.max)
-
-    def test_tzinfo_now(self):
-        meth = self.theclass.now
-        # Ensure it doesn't require tzinfo (i.e., that this doesn't blow up).
-        base = meth()
-        # Try with and without naming the keyword.
-        off42 = FixedOffset(42, "42")
-        another = meth(off42)
-        again = meth(tz=off42)
-        self.assertTrue(another.tzinfo is again.tzinfo)
-        self.assertEqual(another.utcoffset(), timedelta(minutes=42))
-        # Bad argument with and w/o naming the keyword.
-        self.assertRaises(TypeError, meth, 16)
-        self.assertRaises(TypeError, meth, tzinfo=16)
-        # Bad keyword name.
-        self.assertRaises(TypeError, meth, tinfo=off42)
-        # Too many args.
-        self.assertRaises(TypeError, meth, off42, off42)
-
-        # We don't know which time zone we're in, and don't have a tzinfo
-        # class to represent it, so seeing whether a tz argument actually
-        # does a conversion is tricky.
-        utc = FixedOffset(0, "utc", 0)
-        for weirdtz in [FixedOffset(timedelta(hours=15, minutes=58), "weirdtz", 0),
-                        timezone(timedelta(hours=15, minutes=58), "weirdtz"),]:
-            for dummy in range(3):
-                now = datetime.now(weirdtz)
-                self.assertTrue(now.tzinfo is weirdtz)
-                utcnow = datetime.utcnow().replace(tzinfo=utc)
-                now2 = utcnow.astimezone(weirdtz)
-                if abs(now - now2) < timedelta(seconds=30):
-                    break
-                # Else the code is broken, or more than 30 seconds passed between
-                # calls; assuming the latter, just try again.
-            else:
-                # Three strikes and we're out.
-                self.fail("utcnow(), now(tz), or astimezone() may be broken")
-
-    def test_tzinfo_fromtimestamp(self):
-        import time
-        meth = self.theclass.fromtimestamp
-        ts = time.time()
-        # Ensure it doesn't require tzinfo (i.e., that this doesn't blow up).
-        base = meth(ts)
-        # Try with and without naming the keyword.
-        off42 = FixedOffset(42, "42")
-        another = meth(ts, off42)
-        again = meth(ts, tz=off42)
-        self.assertTrue(another.tzinfo is again.tzinfo)
-        self.assertEqual(another.utcoffset(), timedelta(minutes=42))
-        # Bad argument with and w/o naming the keyword.
-        self.assertRaises(TypeError, meth, ts, 16)
-        self.assertRaises(TypeError, meth, ts, tzinfo=16)
-        # Bad keyword name.
-        self.assertRaises(TypeError, meth, ts, tinfo=off42)
-        # Too many args.
-        self.assertRaises(TypeError, meth, ts, off42, off42)
-        # Too few args.
-        self.assertRaises(TypeError, meth)
-
-        # Try to make sure tz= actually does some conversion.
-        timestamp = 1000000000
-        utcdatetime = datetime.utcfromtimestamp(timestamp)
-        # In POSIX (epoch 1970), that's 2001-09-09 01:46:40 UTC, give or take.
-        # But on some flavor of Mac, it's nowhere near that.  So we can't have
-        # any idea here what time that actually is, we can only test that
-        # relative changes match.
-        utcoffset = timedelta(hours=-15, minutes=39) # arbitrary, but not zero
-        tz = FixedOffset(utcoffset, "tz", 0)
-        expected = utcdatetime + utcoffset
-        got = datetime.fromtimestamp(timestamp, tz)
-        self.assertEqual(expected, got.replace(tzinfo=None))
-
-    def test_tzinfo_utcnow(self):
-        meth = self.theclass.utcnow
-        # Ensure it doesn't require tzinfo (i.e., that this doesn't blow up).
-        base = meth()
-        # Try with and without naming the keyword; for whatever reason,
-        # utcnow() doesn't accept a tzinfo argument.
-        off42 = FixedOffset(42, "42")
-        self.assertRaises(TypeError, meth, off42)
-        self.assertRaises(TypeError, meth, tzinfo=off42)
-
-    def test_tzinfo_utcfromtimestamp(self):
-        import time
-        meth = self.theclass.utcfromtimestamp
-        ts = time.time()
-        # Ensure it doesn't require tzinfo (i.e., that this doesn't blow up).
-        base = meth(ts)
-        # Try with and without naming the keyword; for whatever reason,
-        # utcfromtimestamp() doesn't accept a tzinfo argument.
-        off42 = FixedOffset(42, "42")
-        self.assertRaises(TypeError, meth, ts, off42)
-        self.assertRaises(TypeError, meth, ts, tzinfo=off42)
-
-    def test_tzinfo_timetuple(self):
-        # TestDateTime tested most of this.  datetime adds a twist to the
-        # DST flag.
-        class DST(tzinfo):
-            def __init__(self, dstvalue):
-                if isinstance(dstvalue, int):
-                    dstvalue = timedelta(minutes=dstvalue)
-                self.dstvalue = dstvalue
-            def dst(self, dt):
-                return self.dstvalue
-
-        cls = self.theclass
-        for dstvalue, flag in (-33, 1), (33, 1), (0, 0), (None, -1):
-            d = cls(1, 1, 1, 10, 20, 30, 40, tzinfo=DST(dstvalue))
-            t = d.timetuple()
-            self.assertEqual(1, t.tm_year)
-            self.assertEqual(1, t.tm_mon)
-            self.assertEqual(1, t.tm_mday)
-            self.assertEqual(10, t.tm_hour)
-            self.assertEqual(20, t.tm_min)
-            self.assertEqual(30, t.tm_sec)
-            self.assertEqual(0, t.tm_wday)
-            self.assertEqual(1, t.tm_yday)
-            self.assertEqual(flag, t.tm_isdst)
-
-        # dst() returns wrong type.
-        self.assertRaises(TypeError, cls(1, 1, 1, tzinfo=DST("x")).timetuple)
-
-        # dst() at the edge.
-        self.assertEqual(cls(1,1,1, tzinfo=DST(1439)).timetuple().tm_isdst, 1)
-        self.assertEqual(cls(1,1,1, tzinfo=DST(-1439)).timetuple().tm_isdst, 1)
-
-        # dst() out of range.
-        self.assertRaises(ValueError, cls(1,1,1, tzinfo=DST(1440)).timetuple)
-        self.assertRaises(ValueError, cls(1,1,1, tzinfo=DST(-1440)).timetuple)
-
-    def test_utctimetuple(self):
-        class DST(tzinfo):
-            def __init__(self, dstvalue=0):
-                if isinstance(dstvalue, int):
-                    dstvalue = timedelta(minutes=dstvalue)
-                self.dstvalue = dstvalue
-            def dst(self, dt):
-                return self.dstvalue
-
-        cls = self.theclass
-        # This can't work:  DST didn't implement utcoffset.
-        self.assertRaises(NotImplementedError,
-                          cls(1, 1, 1, tzinfo=DST(0)).utcoffset)
-
-        class UOFS(DST):
-            def __init__(self, uofs, dofs=None):
-                DST.__init__(self, dofs)
-                self.uofs = timedelta(minutes=uofs)
-            def utcoffset(self, dt):
-                return self.uofs
-
-        for dstvalue in -33, 33, 0, None:
-            d = cls(1, 2, 3, 10, 20, 30, 40, tzinfo=UOFS(-53, dstvalue))
-            t = d.utctimetuple()
-            self.assertEqual(d.year, t.tm_year)
-            self.assertEqual(d.month, t.tm_mon)
-            self.assertEqual(d.day, t.tm_mday)
-            self.assertEqual(11, t.tm_hour) # 20mm + 53mm = 1hn + 13mm
-            self.assertEqual(13, t.tm_min)
-            self.assertEqual(d.second, t.tm_sec)
-            self.assertEqual(d.weekday(), t.tm_wday)
-            self.assertEqual(d.toordinal() - date(1, 1, 1).toordinal() + 1,
-                             t.tm_yday)
-            # Ensure tm_isdst is 0 regardless of what dst() says: DST
-            # is never in effect for a UTC time.
-            self.assertEqual(0, t.tm_isdst)
-
-        # For naive datetime, utctimetuple == timetuple except for isdst
-        d = cls(1, 2, 3, 10, 20, 30, 40)
-        t = d.utctimetuple()
-        self.assertEqual(t[:-1], d.timetuple()[:-1])
-        self.assertEqual(0, t.tm_isdst)
-        # Same if utcoffset is None
-        class NOFS(DST):
-            def utcoffset(self, dt):
-                return None
-        d = cls(1, 2, 3, 10, 20, 30, 40, tzinfo=NOFS())
-        t = d.utctimetuple()
-        self.assertEqual(t[:-1], d.timetuple()[:-1])
-        self.assertEqual(0, t.tm_isdst)
-        # Check that bad tzinfo is detected
-        class BOFS(DST):
-            def utcoffset(self, dt):
-                return "EST"
-        d = cls(1, 2, 3, 10, 20, 30, 40, tzinfo=BOFS())
-        self.assertRaises(TypeError, d.utctimetuple)
-
-        # Check that utctimetuple() is the same as
-        # astimezone(utc).timetuple()
-        d = cls(2010, 11, 13, 14, 15, 16, 171819)
-        for tz in [timezone.min, timezone.utc, timezone.max]:
-            dtz = d.replace(tzinfo=tz)
-            self.assertEqual(dtz.utctimetuple()[:-1],
-                             dtz.astimezone(timezone.utc).timetuple()[:-1])
-        # At the edges, UTC adjustment can produce years out-of-range
-        # for a datetime object.  Ensure that an OverflowError is
-        # raised.
-        tiny = cls(MINYEAR, 1, 1, 0, 0, 37, tzinfo=UOFS(1439))
-        # That goes back 1 minute less than a full day.
-        self.assertRaises(OverflowError, tiny.utctimetuple)
-
-        huge = cls(MAXYEAR, 12, 31, 23, 59, 37, 999999, tzinfo=UOFS(-1439))
-        # That goes forward 1 minute less than a full day.
-        self.assertRaises(OverflowError, huge.utctimetuple)
-        # More overflow cases
-        tiny = cls.min.replace(tzinfo=timezone(MINUTE))
-        self.assertRaises(OverflowError, tiny.utctimetuple)
-        huge = cls.max.replace(tzinfo=timezone(-MINUTE))
-        self.assertRaises(OverflowError, huge.utctimetuple)
-
-    def test_tzinfo_isoformat(self):
-        zero = FixedOffset(0, "+00:00")
-        plus = FixedOffset(220, "+03:40")
-        minus = FixedOffset(-231, "-03:51")
-        unknown = FixedOffset(None, "")
-
-        cls = self.theclass
-        datestr = '0001-02-03'
-        for ofs in None, zero, plus, minus, unknown:
-            for us in 0, 987001:
-                d = cls(1, 2, 3, 4, 5, 59, us, tzinfo=ofs)
-                timestr = '04:05:59' + (us and '.987001' or '')
-                ofsstr = ofs is not None and d.tzname() or ''
-                tailstr = timestr + ofsstr
-                iso = d.isoformat()
-                self.assertEqual(iso, datestr + 'T' + tailstr)
-                self.assertEqual(iso, d.isoformat('T'))
-                self.assertEqual(d.isoformat('k'), datestr + 'k' + tailstr)
-                self.assertEqual(d.isoformat('\u1234'), datestr + '\u1234' + tailstr)
-                self.assertEqual(str(d), datestr + ' ' + tailstr)
-
-    def test_replace(self):
-        cls = self.theclass
-        z100 = FixedOffset(100, "+100")
-        zm200 = FixedOffset(timedelta(minutes=-200), "-200")
-        args = [1, 2, 3, 4, 5, 6, 7, z100]
-        base = cls(*args)
-        self.assertEqual(base, base.replace())
-
-        i = 0
-        for name, newval in (("year", 2),
-                             ("month", 3),
-                             ("day", 4),
-                             ("hour", 5),
-                             ("minute", 6),
-                             ("second", 7),
-                             ("microsecond", 8),
-                             ("tzinfo", zm200)):
-            newargs = args[:]
-            newargs[i] = newval
-            expected = cls(*newargs)
-            got = base.replace(**{name: newval})
-            self.assertEqual(expected, got)
-            i += 1
-
-        # Ensure we can get rid of a tzinfo.
-        self.assertEqual(base.tzname(), "+100")
-        base2 = base.replace(tzinfo=None)
-        self.assertTrue(base2.tzinfo is None)
-        self.assertTrue(base2.tzname() is None)
-
-        # Ensure we can add one.
-        base3 = base2.replace(tzinfo=z100)
-        self.assertEqual(base, base3)
-        self.assertTrue(base.tzinfo is base3.tzinfo)
-
-        # Out of bounds.
-        base = cls(2000, 2, 29)
-        self.assertRaises(ValueError, base.replace, year=2001)
-
-    def test_more_astimezone(self):
-        # The inherited test_astimezone covered some trivial and error cases.
-        fnone = FixedOffset(None, "None")
-        f44m = FixedOffset(44, "44")
-        fm5h = FixedOffset(-timedelta(hours=5), "m300")
-
-        dt = self.theclass.now(tz=f44m)
-        self.assertTrue(dt.tzinfo is f44m)
-        # Replacing with degenerate tzinfo raises an exception.
-        self.assertRaises(ValueError, dt.astimezone, fnone)
-        # Ditto with None tz.
-        self.assertRaises(TypeError, dt.astimezone, None)
-        # Replacing with same tzinfo makes no change.
-        x = dt.astimezone(dt.tzinfo)
-        self.assertTrue(x.tzinfo is f44m)
-        self.assertEqual(x.date(), dt.date())
-        self.assertEqual(x.time(), dt.time())
-
-        # Replacing with different tzinfo does adjust.
-        got = dt.astimezone(fm5h)
-        self.assertTrue(got.tzinfo is fm5h)
-        self.assertEqual(got.utcoffset(), timedelta(hours=-5))
-        expected = dt - dt.utcoffset()  # in effect, convert to UTC
-        expected += fm5h.utcoffset(dt)  # and from there to local time
-        expected = expected.replace(tzinfo=fm5h) # and attach new tzinfo
-        self.assertEqual(got.date(), expected.date())
-        self.assertEqual(got.time(), expected.time())
-        self.assertEqual(got.timetz(), expected.timetz())
-        self.assertTrue(got.tzinfo is expected.tzinfo)
-        self.assertEqual(got, expected)
-
-    def test_aware_subtract(self):
-        cls = self.theclass
-
-        # Ensure that utcoffset() is ignored when the operands have the
-        # same tzinfo member.
-        class OperandDependentOffset(tzinfo):
-            def utcoffset(self, t):
-                if t.minute < 10:
-                    # d0 and d1 equal after adjustment
-                    return timedelta(minutes=t.minute)
-                else:
-                    # d2 off in the weeds
-                    return timedelta(minutes=59)
-
-        base = cls(8, 9, 10, 11, 12, 13, 14, tzinfo=OperandDependentOffset())
-        d0 = base.replace(minute=3)
-        d1 = base.replace(minute=9)
-        d2 = base.replace(minute=11)
-        for x in d0, d1, d2:
-            for y in d0, d1, d2:
-                got = x - y
-                expected = timedelta(minutes=x.minute - y.minute)
-                self.assertEqual(got, expected)
-
-        # OTOH, if the tzinfo members are distinct, utcoffsets aren't
-        # ignored.
-        base = cls(8, 9, 10, 11, 12, 13, 14)
-        d0 = base.replace(minute=3, tzinfo=OperandDependentOffset())
-        d1 = base.replace(minute=9, tzinfo=OperandDependentOffset())
-        d2 = base.replace(minute=11, tzinfo=OperandDependentOffset())
-        for x in d0, d1, d2:
-            for y in d0, d1, d2:
-                got = x - y
-                if (x is d0 or x is d1) and (y is d0 or y is d1):
-                    expected = timedelta(0)
-                elif x is y is d2:
-                    expected = timedelta(0)
-                elif x is d2:
-                    expected = timedelta(minutes=(11-59)-0)
-                else:
-                    assert y is d2
-                    expected = timedelta(minutes=0-(11-59))
-                self.assertEqual(got, expected)
-
-    def test_mixed_compare(self):
-        t1 = datetime(1, 2, 3, 4, 5, 6, 7)
-        t2 = datetime(1, 2, 3, 4, 5, 6, 7)
-        self.assertEqual(t1, t2)
-        t2 = t2.replace(tzinfo=None)
-        self.assertEqual(t1, t2)
-        t2 = t2.replace(tzinfo=FixedOffset(None, ""))
-        self.assertEqual(t1, t2)
-        t2 = t2.replace(tzinfo=FixedOffset(0, ""))
-        self.assertRaises(TypeError, lambda: t1 == t2)
-
-        # In datetime w/ identical tzinfo objects, utcoffset is ignored.
-        class Varies(tzinfo):
-            def __init__(self):
-                self.offset = timedelta(minutes=22)
-            def utcoffset(self, t):
-                self.offset += timedelta(minutes=1)
-                return self.offset
-
-        v = Varies()
-        t1 = t2.replace(tzinfo=v)
-        t2 = t2.replace(tzinfo=v)
-        self.assertEqual(t1.utcoffset(), timedelta(minutes=23))
-        self.assertEqual(t2.utcoffset(), timedelta(minutes=24))
-        self.assertEqual(t1, t2)
-
-        # But if they're not identical, it isn't ignored.
-        t2 = t2.replace(tzinfo=Varies())
-        self.assertTrue(t1 < t2)  # t1's offset counter still going up
-
-    def test_subclass_datetimetz(self):
-
-        class C(self.theclass):
-            theAnswer = 42
-
-            def __new__(cls, *args, **kws):
-                temp = kws.copy()
-                extra = temp.pop('extra')
-                result = self.theclass.__new__(cls, *args, **temp)
-                result.extra = extra
-                return result
-
-            def newmeth(self, start):
-                return start + self.hour + self.year
-
-        args = 2002, 12, 31, 4, 5, 6, 500, FixedOffset(-300, "EST", 1)
-
-        dt1 = self.theclass(*args)
-        dt2 = C(*args, **{'extra': 7})
-
-        self.assertEqual(dt2.__class__, C)
-        self.assertEqual(dt2.theAnswer, 42)
-        self.assertEqual(dt2.extra, 7)
-        self.assertEqual(dt1.utcoffset(), dt2.utcoffset())
-        self.assertEqual(dt2.newmeth(-7), dt1.hour + dt1.year - 7)
-
-# Pain to set up DST-aware tzinfo classes.
-
-def first_sunday_on_or_after(dt):
-    days_to_go = 6 - dt.weekday()
-    if days_to_go:
-        dt += timedelta(days_to_go)
-    return dt
-
-ZERO = timedelta(0)
-MINUTE = timedelta(minutes=1)
-HOUR = timedelta(hours=1)
-DAY = timedelta(days=1)
-# In the US, DST starts at 2am (standard time) on the first Sunday in April.
-DSTSTART = datetime(1, 4, 1, 2)
-# and ends at 2am (DST time; 1am standard time) on the last Sunday of Oct,
-# which is the first Sunday on or after Oct 25.  Because we view 1:MM as
-# being standard time on that day, there is no spelling in local time of
-# the last hour of DST (that's 1:MM DST, but 1:MM is taken as standard time).
-DSTEND = datetime(1, 10, 25, 1)
-
-class USTimeZone(tzinfo):
-
-    def __init__(self, hours, reprname, stdname, dstname):
-        self.stdoffset = timedelta(hours=hours)
-        self.reprname = reprname
-        self.stdname = stdname
-        self.dstname = dstname
-
-    def __repr__(self):
-        return self.reprname
-
-    def tzname(self, dt):
-        if self.dst(dt):
-            return self.dstname
-        else:
-            return self.stdname
-
-    def utcoffset(self, dt):
-        return self.stdoffset + self.dst(dt)
-
-    def dst(self, dt):
-        if dt is None or dt.tzinfo is None:
-            # An exception instead may be sensible here, in one or more of
-            # the cases.
-            return ZERO
-        assert dt.tzinfo is self
-
-        # Find first Sunday in April.
-        start = first_sunday_on_or_after(DSTSTART.replace(year=dt.year))
-        assert start.weekday() == 6 and start.month == 4 and start.day <= 7
-
-        # Find last Sunday in October.
-        end = first_sunday_on_or_after(DSTEND.replace(year=dt.year))
-        assert end.weekday() == 6 and end.month == 10 and end.day >= 25
-
-        # Can't compare naive to aware objects, so strip the timezone from
-        # dt first.
-        if start <= dt.replace(tzinfo=None) < end:
-            return HOUR
-        else:
-            return ZERO
-
-Eastern  = USTimeZone(-5, "Eastern",  "EST", "EDT")
-Central  = USTimeZone(-6, "Central",  "CST", "CDT")
-Mountain = USTimeZone(-7, "Mountain", "MST", "MDT")
-Pacific  = USTimeZone(-8, "Pacific",  "PST", "PDT")
-utc_real = FixedOffset(0, "UTC", 0)
-# For better test coverage, we want another flavor of UTC that's west of
-# the Eastern and Pacific timezones.
-utc_fake = FixedOffset(-12*60, "UTCfake", 0)
-
-class TestTimezoneConversions(unittest.TestCase):
-    # The DST switch times for 2002, in std time.
-    dston = datetime(2002, 4, 7, 2)
-    dstoff = datetime(2002, 10, 27, 1)
-
-    theclass = datetime
-
-    # Check a time that's inside DST.
-    def checkinside(self, dt, tz, utc, dston, dstoff):
-        self.assertEqual(dt.dst(), HOUR)
-
-        # Conversion to our own timezone is always an identity.
-        self.assertEqual(dt.astimezone(tz), dt)
-
-        asutc = dt.astimezone(utc)
-        there_and_back = asutc.astimezone(tz)
-
-        # Conversion to UTC and back isn't always an identity here,
-        # because there are redundant spellings (in local time) of
-        # UTC time when DST begins:  the clock jumps from 1:59:59
-        # to 3:00:00, and a local time of 2:MM:SS doesn't really
-        # make sense then.  The classes above treat 2:MM:SS as
-        # daylight time then (it's "after 2am"), really an alias
-        # for 1:MM:SS standard time.  The latter form is what
-        # conversion back from UTC produces.
-        if dt.date() == dston.date() and dt.hour == 2:
-            # We're in the redundant hour, and coming back from
-            # UTC gives the 1:MM:SS standard-time spelling.
-            self.assertEqual(there_and_back + HOUR, dt)
-            # Although during was considered to be in daylight
-            # time, there_and_back is not.
-            self.assertEqual(there_and_back.dst(), ZERO)
-            # They're the same times in UTC.
-            self.assertEqual(there_and_back.astimezone(utc),
-                             dt.astimezone(utc))
-        else:
-            # We're not in the redundant hour.
-            self.assertEqual(dt, there_and_back)
-
-        # Because we have a redundant spelling when DST begins, there is
-        # (unforunately) an hour when DST ends that can't be spelled at all in
-        # local time.  When DST ends, the clock jumps from 1:59 back to 1:00
-        # again.  The hour 1:MM DST has no spelling then:  1:MM is taken to be
-        # standard time.  1:MM DST == 0:MM EST, but 0:MM is taken to be
-        # daylight time.  The hour 1:MM daylight == 0:MM standard can't be
-        # expressed in local time.  Nevertheless, we want conversion back
-        # from UTC to mimic the local clock's "repeat an hour" behavior.
-        nexthour_utc = asutc + HOUR
-        nexthour_tz = nexthour_utc.astimezone(tz)
-        if dt.date() == dstoff.date() and dt.hour == 0:
-            # We're in the hour before the last DST hour.  The last DST hour
-            # is ineffable.  We want the conversion back to repeat 1:MM.
-            self.assertEqual(nexthour_tz, dt.replace(hour=1))
-            nexthour_utc += HOUR
-            nexthour_tz = nexthour_utc.astimezone(tz)
-            self.assertEqual(nexthour_tz, dt.replace(hour=1))
-        else:
-            self.assertEqual(nexthour_tz - dt, HOUR)
-
-    # Check a time that's outside DST.
-    def checkoutside(self, dt, tz, utc):
-        self.assertEqual(dt.dst(), ZERO)
-
-        # Conversion to our own timezone is always an identity.
-        self.assertEqual(dt.astimezone(tz), dt)
-
-        # Converting to UTC and back is an identity too.
-        asutc = dt.astimezone(utc)
-        there_and_back = asutc.astimezone(tz)
-        self.assertEqual(dt, there_and_back)
-
-    def convert_between_tz_and_utc(self, tz, utc):
-        dston = self.dston.replace(tzinfo=tz)
-        # Because 1:MM on the day DST ends is taken as being standard time,
-        # there is no spelling in tz for the last hour of daylight time.
-        # For purposes of the test, the last hour of DST is 0:MM, which is
-        # taken as being daylight time (and 1:MM is taken as being standard
-        # time).
-        dstoff = self.dstoff.replace(tzinfo=tz)
-        for delta in (timedelta(weeks=13),
-                      DAY,
-                      HOUR,
-                      timedelta(minutes=1),
-                      timedelta(microseconds=1)):
-
-            self.checkinside(dston, tz, utc, dston, dstoff)
-            for during in dston + delta, dstoff - delta:
-                self.checkinside(during, tz, utc, dston, dstoff)
-
-            self.checkoutside(dstoff, tz, utc)
-            for outside in dston - delta, dstoff + delta:
-                self.checkoutside(outside, tz, utc)
-
-    def test_easy(self):
-        # Despite the name of this test, the endcases are excruciating.
-        self.convert_between_tz_and_utc(Eastern, utc_real)
-        self.convert_between_tz_and_utc(Pacific, utc_real)
-        self.convert_between_tz_and_utc(Eastern, utc_fake)
-        self.convert_between_tz_and_utc(Pacific, utc_fake)
-        # The next is really dancing near the edge.  It works because
-        # Pacific and Eastern are far enough apart that their "problem
-        # hours" don't overlap.
-        self.convert_between_tz_and_utc(Eastern, Pacific)
-        self.convert_between_tz_and_utc(Pacific, Eastern)
-        # OTOH, these fail!  Don't enable them.  The difficulty is that
-        # the edge case tests assume that every hour is representable in
-        # the "utc" class.  This is always true for a fixed-offset tzinfo
-        # class (lke utc_real and utc_fake), but not for Eastern or Central.
-        # For these adjacent DST-aware time zones, the range of time offsets
-        # tested ends up creating hours in the one that aren't representable
-        # in the other.  For the same reason, we would see failures in the
-        # Eastern vs Pacific tests too if we added 3*HOUR to the list of
-        # offset deltas in convert_between_tz_and_utc().
-        #
-        # self.convert_between_tz_and_utc(Eastern, Central)  # can't work
-        # self.convert_between_tz_and_utc(Central, Eastern)  # can't work
-
-    def test_tricky(self):
-        # 22:00 on day before daylight starts.
-        fourback = self.dston - timedelta(hours=4)
-        ninewest = FixedOffset(-9*60, "-0900", 0)
-        fourback = fourback.replace(tzinfo=ninewest)
-        # 22:00-0900 is 7:00 UTC == 2:00 EST == 3:00 DST.  Since it's "after
-        # 2", we should get the 3 spelling.
-        # If we plug 22:00 the day before into Eastern, it "looks like std
-        # time", so its offset is returned as -5, and -5 - -9 = 4.  Adding 4
-        # to 22:00 lands on 2:00, which makes no sense in local time (the
-        # local clock jumps from 1 to 3).  The point here is to make sure we
-        # get the 3 spelling.
-        expected = self.dston.replace(hour=3)
-        got = fourback.astimezone(Eastern).replace(tzinfo=None)
-        self.assertEqual(expected, got)
-
-        # Similar, but map to 6:00 UTC == 1:00 EST == 2:00 DST.  In that
-        # case we want the 1:00 spelling.
-        sixutc = self.dston.replace(hour=6, tzinfo=utc_real)
-        # Now 6:00 "looks like daylight", so the offset wrt Eastern is -4,
-        # and adding -4-0 == -4 gives the 2:00 spelling.  We want the 1:00 EST
-        # spelling.
-        expected = self.dston.replace(hour=1)
-        got = sixutc.astimezone(Eastern).replace(tzinfo=None)
-        self.assertEqual(expected, got)
-
-        # Now on the day DST ends, we want "repeat an hour" behavior.
-        #  UTC  4:MM  5:MM  6:MM  7:MM  checking these
-        #  EST 23:MM  0:MM  1:MM  2:MM
-        #  EDT  0:MM  1:MM  2:MM  3:MM
-        # wall  0:MM  1:MM  1:MM  2:MM  against these
-        for utc in utc_real, utc_fake:
-            for tz in Eastern, Pacific:
-                first_std_hour = self.dstoff - timedelta(hours=2) # 23:MM
-                # Convert that to UTC.
-                first_std_hour -= tz.utcoffset(None)
-                # Adjust for possibly fake UTC.
-                asutc = first_std_hour + utc.utcoffset(None)
-                # First UTC hour to convert; this is 4:00 when utc=utc_real &
-                # tz=Eastern.
-                asutcbase = asutc.replace(tzinfo=utc)
-                for tzhour in (0, 1, 1, 2):
-                    expectedbase = self.dstoff.replace(hour=tzhour)
-                    for minute in 0, 30, 59:
-                        expected = expectedbase.replace(minute=minute)
-                        asutc = asutcbase.replace(minute=minute)
-                        astz = asutc.astimezone(tz)
-                        self.assertEqual(astz.replace(tzinfo=None), expected)
-                    asutcbase += HOUR
-
-
-    def test_bogus_dst(self):
-        class ok(tzinfo):
-            def utcoffset(self, dt): return HOUR
-            def dst(self, dt): return HOUR
-
-        now = self.theclass.now().replace(tzinfo=utc_real)
-        # Doesn't blow up.
-        now.astimezone(ok())
-
-        # Does blow up.
-        class notok(ok):
-            def dst(self, dt): return None
-        self.assertRaises(ValueError, now.astimezone, notok())
-
-        # Sometimes blow up. In the following, tzinfo.dst()
-        # implementation may return None or not None depending on
-        # whether DST is assumed to be in effect.  In this situation,
-        # a ValueError should be raised by astimezone().
-        class tricky_notok(ok):
-            def dst(self, dt):
-                if dt.year == 2000:
-                    return None
-                else:
-                    return 10*HOUR
-        dt = self.theclass(2001, 1, 1).replace(tzinfo=utc_real)
-        self.assertRaises(ValueError, dt.astimezone, tricky_notok())
-
-    def test_fromutc(self):
-        self.assertRaises(TypeError, Eastern.fromutc)   # not enough args
-        now = datetime.utcnow().replace(tzinfo=utc_real)
-        self.assertRaises(ValueError, Eastern.fromutc, now) # wrong tzinfo
-        now = now.replace(tzinfo=Eastern)   # insert correct tzinfo
-        enow = Eastern.fromutc(now)         # doesn't blow up
-        self.assertEqual(enow.tzinfo, Eastern) # has right tzinfo member
-        self.assertRaises(TypeError, Eastern.fromutc, now, now) # too many args
-        self.assertRaises(TypeError, Eastern.fromutc, date.today()) # wrong type
-
-        # Always converts UTC to standard time.
-        class FauxUSTimeZone(USTimeZone):
-            def fromutc(self, dt):
-                return dt + self.stdoffset
-        FEastern  = FauxUSTimeZone(-5, "FEastern",  "FEST", "FEDT")
-
-        #  UTC  4:MM  5:MM  6:MM  7:MM  8:MM  9:MM
-        #  EST 23:MM  0:MM  1:MM  2:MM  3:MM  4:MM
-        #  EDT  0:MM  1:MM  2:MM  3:MM  4:MM  5:MM
-
-        # Check around DST start.
-        start = self.dston.replace(hour=4, tzinfo=Eastern)
-        fstart = start.replace(tzinfo=FEastern)
-        for wall in 23, 0, 1, 3, 4, 5:
-            expected = start.replace(hour=wall)
-            if wall == 23:
-                expected -= timedelta(days=1)
-            got = Eastern.fromutc(start)
-            self.assertEqual(expected, got)
-
-            expected = fstart + FEastern.stdoffset
-            got = FEastern.fromutc(fstart)
-            self.assertEqual(expected, got)
-
-            # Ensure astimezone() calls fromutc() too.
-            got = fstart.replace(tzinfo=utc_real).astimezone(FEastern)
-            self.assertEqual(expected, got)
-
-            start += HOUR
-            fstart += HOUR
-
-        # Check around DST end.
-        start = self.dstoff.replace(hour=4, tzinfo=Eastern)
-        fstart = start.replace(tzinfo=FEastern)
-        for wall in 0, 1, 1, 2, 3, 4:
-            expected = start.replace(hour=wall)
-            got = Eastern.fromutc(start)
-            self.assertEqual(expected, got)
-
-            expected = fstart + FEastern.stdoffset
-            got = FEastern.fromutc(fstart)
-            self.assertEqual(expected, got)
-
-            # Ensure astimezone() calls fromutc() too.
-            got = fstart.replace(tzinfo=utc_real).astimezone(FEastern)
-            self.assertEqual(expected, got)
-
-            start += HOUR
-            fstart += HOUR
-
-
-#############################################################################
-# oddballs
-
-class Oddballs(unittest.TestCase):
-
-    def test_bug_1028306(self):
-        # Trying to compare a date to a datetime should act like a mixed-
-        # type comparison, despite that datetime is a subclass of date.
-        as_date = date.today()
-        as_datetime = datetime.combine(as_date, time())
-        self.assertTrue(as_date != as_datetime)
-        self.assertTrue(as_datetime != as_date)
-        self.assertTrue(not as_date == as_datetime)
-        self.assertTrue(not as_datetime == as_date)
-        self.assertRaises(TypeError, lambda: as_date < as_datetime)
-        self.assertRaises(TypeError, lambda: as_datetime < as_date)
-        self.assertRaises(TypeError, lambda: as_date <= as_datetime)
-        self.assertRaises(TypeError, lambda: as_datetime <= as_date)
-        self.assertRaises(TypeError, lambda: as_date > as_datetime)
-        self.assertRaises(TypeError, lambda: as_datetime > as_date)
-        self.assertRaises(TypeError, lambda: as_date >= as_datetime)
-        self.assertRaises(TypeError, lambda: as_datetime >= as_date)
-
-        # Neverthelss, comparison should work with the base-class (date)
-        # projection if use of a date method is forced.
-        self.assertEqual(as_date.__eq__(as_datetime), True)
-        different_day = (as_date.day + 1) % 20 + 1
-        as_different = as_datetime.replace(day= different_day)
-        self.assertEqual(as_date.__eq__(as_different), False)
-
-        # And date should compare with other subclasses of date.  If a
-        # subclass wants to stop this, it's up to the subclass to do so.
-        date_sc = SubclassDate(as_date.year, as_date.month, as_date.day)
-        self.assertEqual(as_date, date_sc)
-        self.assertEqual(date_sc, as_date)
-
-        # Ditto for datetimes.
-        datetime_sc = SubclassDatetime(as_datetime.year, as_datetime.month,
-                                       as_date.day, 0, 0, 0)
-        self.assertEqual(as_datetime, datetime_sc)
-        self.assertEqual(datetime_sc, as_datetime)
+import sys
+from test.support import import_fresh_module, run_unittest
+TESTS = 'test.datetimetester'
+# XXX: import_fresh_module() is supposed to leave sys.module cache untouched,
+# XXX: but it does not, so we have to save and restore it ourselves.
+save_sys_modules = sys.modules.copy()
+try:
+    pure_tests = import_fresh_module(TESTS, fresh=['datetime', '_strptime'],
+                                     blocked=['_datetime'])
+    fast_tests = import_fresh_module(TESTS, fresh=['datetime',
+                                                   '_datetime', '_strptime'])
+finally:
+    sys.modules.clear()
+    sys.modules.update(save_sys_modules)
+test_modules = [pure_tests, fast_tests]
+test_suffixes = ["_Pure", "_Fast"]
+
+for module, suffix in zip(test_modules, test_suffixes):
+    for name, cls in module.__dict__.items():
+        if isinstance(cls, type) and issubclass(cls, unittest.TestCase):
+            name += suffix
+            cls.__name__ = name
+            globals()[name] = cls
+            def setUp(self, module=module, setup=cls.setUp):
+                self._save_sys_modules = sys.modules.copy()
+                sys.modules[TESTS] = module
+                sys.modules['datetime'] = module.datetime_module
+                sys.modules['_strptime'] = module._strptime
+                setup(self)
+            def tearDown(self, teardown=cls.tearDown):
+                teardown(self)
+                sys.modules.clear()
+                sys.modules.update(self._save_sys_modules)
+            cls.setUp = setUp
+            cls.tearDown = tearDown
 
 def test_main():
-    support.run_unittest(__name__)
+    run_unittest(__name__)
 
 if __name__ == "__main__":
     test_main()

Modified: python/branches/import_unicode/Lib/test/test_fnmatch.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_fnmatch.py	(original)
+++ python/branches/import_unicode/Lib/test/test_fnmatch.py	Fri Jul 30 01:19:37 2010
@@ -3,10 +3,15 @@
 from test import support
 import unittest
 
-from fnmatch import fnmatch, fnmatchcase
+from fnmatch import (fnmatch, fnmatchcase, _MAXCACHE, _cache, _cacheb, purge,
+                        translate, filter)
 
 
 class FnmatchTestCase(unittest.TestCase):
+
+    def tearDown(self):
+        purge()
+
     def check_match(self, filename, pattern, should_match=1, fn=fnmatch):
         if should_match:
             self.assertTrue(fn(filename, pattern),
@@ -60,9 +65,45 @@
         self.check_match(b'test\xff', b'te*\xff')
         self.check_match(b'foo\nbar', b'foo*')
 
+    def test_cache_clearing(self):
+        # check that caches do not grow too large
+        # http://bugs.python.org/issue7846
+
+        # string pattern cache
+        for i in range(_MAXCACHE + 1):
+            fnmatch('foo', '?' * i)
+
+        self.assertLessEqual(len(_cache), _MAXCACHE)
+
+        # bytes pattern cache
+        for i in range(_MAXCACHE + 1):
+            fnmatch(b'foo', b'?' * i)
+        self.assertLessEqual(len(_cacheb), _MAXCACHE)
+
+
+class TranslateTestCase(unittest.TestCase):
+
+    def test_translate(self):
+        self.assertEqual(translate('*'), '.*\Z(?ms)')
+        self.assertEqual(translate('?'), '.\Z(?ms)')
+        self.assertEqual(translate('a?b*'), 'a.b.*\Z(?ms)')
+        self.assertEqual(translate('[abc]'), '[abc]\Z(?ms)')
+        self.assertEqual(translate('[]]'), '[]]\Z(?ms)')
+        self.assertEqual(translate('[!x]'), '[^x]\Z(?ms)')
+        self.assertEqual(translate('[^x]'), '[\\^x]\Z(?ms)')
+        self.assertEqual(translate('[x'), '\\[x\Z(?ms)')
+
+
+class FilterTestCase(unittest.TestCase):
+
+    def test_filter(self):
+        self.assertEqual(filter(['a', 'b'], 'a'), ['a'])
+
 
 def test_main():
-    support.run_unittest(FnmatchTestCase)
+    support.run_unittest(FnmatchTestCase,
+                         TranslateTestCase,
+                         FilterTestCase)
 
 
 if __name__ == "__main__":

Modified: python/branches/import_unicode/Lib/test/test_getopt.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_getopt.py	(original)
+++ python/branches/import_unicode/Lib/test/test_getopt.py	Fri Jul 30 01:19:37 2010
@@ -173,6 +173,12 @@
         m = types.ModuleType("libreftest", s)
         run_doctest(m, verbose)
 
+    def test_issue4629(self):
+        longopts, shortopts = getopt.getopt(['--help='], '', ['help='])
+        self.assertEquals(longopts, [('--help', '')])
+        longopts, shortopts = getopt.getopt(['--help=x'], '', ['help='])
+        self.assertEquals(longopts, [('--help', 'x')])
+        self.assertRaises(getopt.GetoptError, getopt.getopt, ['--help='], '', ['help'])
 
 def test_main():
     run_unittest(GetoptTests)

Modified: python/branches/import_unicode/Lib/test/test_http_cookiejar.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_http_cookiejar.py	(original)
+++ python/branches/import_unicode/Lib/test/test_http_cookiejar.py	Fri Jul 30 01:19:37 2010
@@ -1,18 +1,21 @@
 """Tests for http/cookiejar.py."""
 
-import re, os, time, urllib.request
-from unittest import TestCase
+import os
+import re
+import test.support
+import time
+import unittest
+import urllib.request
+
+from http.cookiejar import (time2isoz, http2time, time2netscape,
+     parse_ns_headers, join_header_words, split_header_words, Cookie,
+     CookieJar, DefaultCookiePolicy, LWPCookieJar, MozillaCookieJar,
+     LoadError, lwp_cookie_str, DEFAULT_HTTP_PORT, escape_path,
+     reach, is_HDN, domain_match, user_domain_match, request_path,
+     request_port, request_host)
 
-from test import support
 
-from http.cookiejar import time2isoz, http2time, time2netscape, \
-     parse_ns_headers, join_header_words, split_header_words, Cookie, \
-     CookieJar, DefaultCookiePolicy, LWPCookieJar, MozillaCookieJar, \
-     LoadError, lwp_cookie_str, DEFAULT_HTTP_PORT, escape_path, \
-     reach, is_HDN, domain_match, user_domain_match, request_path, \
-     request_port, request_host
-
-class DateTimeTests(TestCase):
+class DateTimeTests(unittest.TestCase):
 
     def test_time2isoz(self):
         base = 1019227000
@@ -96,7 +99,7 @@
                          )
 
 
-class HeaderTests(TestCase):
+class HeaderTests(unittest.TestCase):
 
     def test_parse_ns_headers(self):
         # quotes should be stripped
@@ -228,10 +231,10 @@
     return cookie_hdr
 
 
-class FileCookieJarTests(TestCase):
+class FileCookieJarTests(unittest.TestCase):
     def test_lwp_valueless_cookie(self):
         # cookies with no value should be saved and loaded consistently
-        filename = support.TESTFN
+        filename = test.support.TESTFN
         c = LWPCookieJar()
         interact_netscape(c, "http://www.acme.com/", 'boo')
         self.assertEqual(c._cookies["www.acme.com"]["/"]["boo"].value, None)
@@ -246,7 +249,7 @@
 
     def test_bad_magic(self):
         # IOErrors (eg. file doesn't exist) are allowed to propagate
-        filename = support.TESTFN
+        filename = test.support.TESTFN
         for cookiejar_class in LWPCookieJar, MozillaCookieJar:
             c = cookiejar_class()
             try:
@@ -269,7 +272,7 @@
             try: os.unlink(filename)
             except OSError: pass
 
-class CookieTests(TestCase):
+class CookieTests(unittest.TestCase):
     # XXX
     # Get rid of string comparisons where not actually testing str / repr.
     # .clear() etc.
@@ -349,7 +352,7 @@
     def test_missing_value(self):
         # missing = sign in Cookie: header is regarded by Mozilla as a missing
         # name, and by http.cookiejar as a missing value
-        filename = support.TESTFN
+        filename = test.support.TESTFN
         c = MozillaCookieJar(filename)
         interact_netscape(c, "http://www.acme.com/", 'eggs')
         interact_netscape(c, "http://www.acme.com/", '"spam"; path=/foo/')
@@ -534,6 +537,16 @@
         interact_netscape(c, "http://www.acme.com/blah/rhubarb/", 'eggs="bar"')
         self.assertIn("/blah/rhubarb", c._cookies["www.acme.com"])
 
+    def test_default_path_with_query(self):
+        cj = CookieJar()
+        uri = "http://example.com/?spam/eggs"
+        value = 'eggs="bar"'
+        interact_netscape(cj, uri, value)
+        # Default path does not include query, so is "/", not "/?spam".
+        self.assertIn("/", cj._cookies["example.com"])
+        # Cookie is sent back to the same URI.
+        self.assertEquals(interact_netscape(cj, uri), value)
+
     def test_escape_path(self):
         cases = [
             # quoted safe
@@ -562,16 +575,15 @@
     def test_request_path(self):
         # with parameters
         req = urllib.request.Request(
-            "http://www.example.com/rheum/rhaponicum;"
+            "http://www.example.com/rheum/rhaponticum;"
             "foo=bar;sing=song?apples=pears&spam=eggs#ni")
-        self.assertEquals(request_path(req), "/rheum/rhaponicum;"
-                     "foo=bar;sing=song?apples=pears&spam=eggs#ni")
+        self.assertEquals(request_path(req),
+                          "/rheum/rhaponticum;foo=bar;sing=song")
         # without parameters
         req = urllib.request.Request(
-            "http://www.example.com/rheum/rhaponicum?"
+            "http://www.example.com/rheum/rhaponticum?"
             "apples=pears&spam=eggs#ni")
-        self.assertEquals(request_path(req), "/rheum/rhaponicum?"
-                     "apples=pears&spam=eggs#ni")
+        self.assertEquals(request_path(req), "/rheum/rhaponticum")
         # missing final slash
         req = urllib.request.Request("http://www.example.com")
         self.assertEquals(request_path(req), "/")
@@ -1045,7 +1057,7 @@
         self.assertTrue(cookie.expires is None)
 
 
-class LWPCookieTests(TestCase):
+class LWPCookieTests(unittest.TestCase):
     # Tests taken from libwww-perl, with a few modifications and additions.
 
     def test_netscape_example_1(self):
@@ -1435,7 +1447,7 @@
         self.assertEquals(len(c), 6)
 
         # save and restore
-        filename = support.TESTFN
+        filename = test.support.TESTFN
 
         try:
             c.save(filename, ignore_discard=True)
@@ -1475,7 +1487,7 @@
         # Save / load Mozilla/Netscape cookie file format.
         year_plus_one = time.localtime()[0] + 1
 
-        filename = support.TESTFN
+        filename = test.support.TESTFN
 
         c = MozillaCookieJar(filename,
                              policy=DefaultCookiePolicy(rfc2965=True))
@@ -1637,7 +1649,7 @@
 
 
 def test_main(verbose=None):
-    support.run_unittest(
+    test.support.run_unittest(
         DateTimeTests,
         HeaderTests,
         CookieTests,

Modified: python/branches/import_unicode/Lib/test/test_import.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_import.py	(original)
+++ python/branches/import_unicode/Lib/test/test_import.py	Fri Jul 30 01:19:37 2010
@@ -1,6 +1,8 @@
 import builtins
 import errno
 import imp
+from importlib.test.import_ import test_relative_imports
+from importlib.test.import_ import util as importlib_util
 import marshal
 import os
 import py_compile
@@ -640,10 +642,21 @@
                          os.path.join(os.curdir, foo_pyc))
 
 
+class RelativeImportFromImportlibTests(test_relative_imports.RelativeImports):
+
+    def setUp(self):
+        self._importlib_util_flag = importlib_util.using___import__
+        importlib_util.using___import__ = True
+
+    def tearDown(self):
+        importlib_util.using___import__ = self._importlib_util_flag
+
+
 def test_main(verbose=None):
     run_unittest(ImportTests, PycacheTests,
                  PycRewritingTests, PathsTests, RelativeImportTests,
-                 OverridingImportBuiltinTests)
+                 OverridingImportBuiltinTests,
+                 RelativeImportFromImportlibTests)
 
 
 if __name__ == '__main__':

Modified: python/branches/import_unicode/Lib/test/test_math.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_math.py	(original)
+++ python/branches/import_unicode/Lib/test/test_math.py	Fri Jul 30 01:19:37 2010
@@ -915,6 +915,15 @@
         self.assertRaises(TypeError, math.trunc, 1, 2)
         self.assertRaises(TypeError, math.trunc, TestNoTrunc())
 
+    def testIsfinite(self):
+        self.assertTrue(math.isfinite(0.0))
+        self.assertTrue(math.isfinite(-0.0))
+        self.assertTrue(math.isfinite(1.0))
+        self.assertTrue(math.isfinite(-1.0))
+        self.assertFalse(math.isfinite(float("nan")))
+        self.assertFalse(math.isfinite(float("inf")))
+        self.assertFalse(math.isfinite(float("-inf")))
+
     def testIsnan(self):
         self.assertTrue(math.isnan(float("nan")))
         self.assertTrue(math.isnan(float("inf")* 0.))

Modified: python/branches/import_unicode/Lib/test/test_minidom.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_minidom.py	(original)
+++ python/branches/import_unicode/Lib/test/test_minidom.py	Fri Jul 30 01:19:37 2010
@@ -951,6 +951,14 @@
         doc.unlink()
 
 
+    def testBug0777884(self):
+        doc = parseString("<o>text</o>")
+        text = doc.documentElement.childNodes[0]
+        self.assertEquals(text.nodeType, Node.TEXT_NODE)
+        # Should run quietly, doing nothing.
+        text.normalize()
+        doc.unlink()
+
     def testBug1433694(self):
         doc = parseString("<o><i/>t</o>")
         node = doc.documentElement

Modified: python/branches/import_unicode/Lib/test/test_os.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_os.py	(original)
+++ python/branches/import_unicode/Lib/test/test_os.py	Fri Jul 30 01:19:37 2010
@@ -423,7 +423,9 @@
         """Check that the repr() of os.environ looks like environ({...})."""
         env = os.environ
         self.assertTrue(isinstance(env.data, dict))
-        self.assertEqual(repr(env), 'environ({!r})'.format(env.data))
+        self.assertEqual(repr(env), 'environ({{{}}})'.format(', '.join(
+            '{!r}: {!r}'.format(key, value)
+            for key, value in env.items())))
 
     def test_get_exec_path(self):
         defpath_list = os.defpath.split(os.pathsep)
@@ -1034,7 +1036,8 @@
 
 
 def skipUnlessWindows6(test):
-    if hasattr(sys, 'getwindowsversion') and sys.getwindowsversion().major >= 6:
+    if (hasattr(sys, 'getwindowsversion')
+        and sys.getwindowsversion().major >= 6):
         return test
     return unittest.skip("Requires Windows Vista or later")(test)
 

Modified: python/branches/import_unicode/Lib/test/test_osx_env.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_osx_env.py	(original)
+++ python/branches/import_unicode/Lib/test/test_osx_env.py	Fri Jul 30 01:19:37 2010
@@ -8,7 +8,7 @@
 import unittest
 
 class OSXEnvironmentVariableTestCase(unittest.TestCase):
-    def _check_sys(self, ev, cond, sv, val = '/some/path/to/python'):
+    def _check_sys(self, ev, cond, sv, val = sys.executable + 'dummy'):
         with EnvironmentVarGuard() as evg:
             subpc = [str(sys.executable), '-c',
                 'import sys; sys.exit(2 if "%s" %s %s else 3)' % (val, cond, sv)]

Modified: python/branches/import_unicode/Lib/test/test_platform.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_platform.py	(original)
+++ python/branches/import_unicode/Lib/test/test_platform.py	Fri Jul 30 01:19:37 2010
@@ -15,8 +15,8 @@
         # On Windows, the EXE needs to know where pythonXY.dll is at so we have
         # to add the directory to the path.
         if sys.platform == "win32":
-            os.environ["Path"] = "{};{}".format(os.path.dirname(sys.executable),
-                                                os.environ["Path"])
+            os.environ["Path"] = "{};{}".format(
+                os.path.dirname(sys.executable), os.environ["Path"])
 
         def get(python):
             cmd = [python, '-c',
@@ -194,6 +194,25 @@
             else:
                 self.assertEquals(res[2], 'PowerPC')
 
+
+    @unittest.skipUnless(sys.platform == 'darwin', "OSX only test")
+    def test_mac_ver_with_fork(self):
+        # Issue7895: platform.mac_ver() crashes when using fork without exec
+        #
+        # This test checks that the fix for that issue works.
+        #
+        pid = os.fork()
+        if pid == 0:
+            # child
+            info = platform.mac_ver()
+            os._exit(0)
+
+        else:
+            # parent
+            cpid, sts = os.waitpid(pid, 0)
+            self.assertEquals(cpid, pid)
+            self.assertEquals(sts, 0)
+
     def test_dist(self):
         res = platform.dist()
 

Modified: python/branches/import_unicode/Lib/test/test_posix.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_posix.py	(original)
+++ python/branches/import_unicode/Lib/test/test_posix.py	Fri Jul 30 01:19:37 2010
@@ -6,6 +6,7 @@
 posix = support.import_module('posix')
 
 import errno
+import sys
 import time
 import os
 import pwd
@@ -252,9 +253,14 @@
             posix.chdir(os.curdir)
             self.assertRaises(OSError, posix.chdir, support.TESTFN)
 
-    def test_lsdir(self):
-        if hasattr(posix, 'lsdir'):
-            self.assertIn(support.TESTFN, posix.lsdir(os.curdir))
+    def test_listdir(self):
+        if hasattr(posix, 'listdir'):
+            self.assertTrue(support.TESTFN in posix.listdir(os.curdir))
+
+    def test_listdir_default(self):
+        # When listdir is called without argument, it's the same as listdir(os.curdir)
+        if hasattr(posix, 'listdir'):
+            self.assertTrue(support.TESTFN in posix.listdir())
 
     def test_access(self):
         if hasattr(posix, 'access'):
@@ -344,9 +350,63 @@
                 os.chdir(curdir)
                 support.rmtree(base_path)
 
+    def test_getgroups(self):
+        with os.popen('id -G') as idg:
+            groups = idg.read().strip()
+
+        if not groups:
+            raise unittest.SkipTest("need working 'id -G'")
+
+        # The order of groups isn't important, hence the calls
+        # to sorted.
+        self.assertEqual(
+                list(sorted([int(x) for x in groups.split()])),
+                list(sorted(posix.getgroups())))
+
+class PosixGroupsTester(unittest.TestCase):
+
+    def setUp(self):
+        if posix.getuid() != 0:
+            raise unittest.SkipTest("not enough privileges")
+        if not hasattr(posix, 'getgroups'):
+            raise unittest.SkipTest("need posix.getgroups")
+        if sys.platform == 'darwin':
+            raise unittest.SkipTest("getgroups(2) is broken on OSX")
+        self.saved_groups = posix.getgroups()
+
+    def tearDown(self):
+        if hasattr(posix, 'setgroups'):
+            posix.setgroups(self.saved_groups)
+        elif hasattr(posix, 'initgroups'):
+            name = pwd.getpwuid(posix.getuid()).pw_name
+            posix.initgroups(name, self.saved_groups[0])
+
+    @unittest.skipUnless(hasattr(posix, 'initgroups'),
+                         "test needs posix.initgroups()")
+    def test_initgroups(self):
+        # find missing group
+
+        groups = sorted(self.saved_groups)
+        for g1,g2 in zip(groups[:-1], groups[1:]):
+            g = g1 + 1
+            if g < g2:
+                break
+        else:
+            g = g2 + 1
+        name = pwd.getpwuid(posix.getuid()).pw_name
+        posix.initgroups(name, g)
+        self.assertIn(g, posix.getgroups())
+
+    @unittest.skipUnless(hasattr(posix, 'setgroups'),
+                         "test needs posix.setgroups()")
+    def test_setgroups(self):
+        for groups in [[0], range(16)]:
+            posix.setgroups(groups)
+            self.assertListEqual(groups, posix.getgroups())
+
 
 def test_main():
-    support.run_unittest(PosixTester)
+    support.run_unittest(PosixTester, PosixGroupsTester)
 
 if __name__ == '__main__':
     test_main()

Modified: python/branches/import_unicode/Lib/test/test_posixpath.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_posixpath.py	(original)
+++ python/branches/import_unicode/Lib/test/test_posixpath.py	Fri Jul 30 01:19:37 2010
@@ -208,6 +208,7 @@
     def test_samestat_on_links(self):
         test_fn1 = support.TESTFN + "1"
         test_fn2 = support.TESTFN + "2"
+        self._create_file(test_fn1)
         test_fns = (test_fn1, test_fn2)
         os.symlink(*test_fns)
         stats = map(os.stat, test_fns)

Modified: python/branches/import_unicode/Lib/test/test_pow.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_pow.py	(original)
+++ python/branches/import_unicode/Lib/test/test_pow.py	Fri Jul 30 01:19:37 2010
@@ -18,14 +18,14 @@
                 self.assertEquals(pow(2, i), pow2)
                 if i != 30 : pow2 = pow2*2
 
-            for othertype in int, int:
+            for othertype in (int,):
                 for i in list(range(-10, 0)) + list(range(1, 10)):
                     ii = type(i)
                     for j in range(1, 11):
                         jj = -othertype(j)
                         pow(ii, jj)
 
-        for othertype in int, int, float:
+        for othertype in int, float:
             for i in range(1, 100):
                 zero = type(0)
                 exp = -othertype(i/10.0)

Deleted: python/branches/import_unicode/Lib/test/test_profilehooks.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_profilehooks.py	Fri Jul 30 01:19:37 2010
+++ (empty file)
@@ -1,385 +0,0 @@
-import pprint
-import sys
-import unittest
-
-from test import support
-
-class TestGetProfile(unittest.TestCase):
-    def setUp(self):
-        sys.setprofile(None)
-
-    def tearDown(self):
-        sys.setprofile(None)
-
-    def test_empty(self):
-        assert sys.getprofile() is None
-
-    def test_setget(self):
-        def fn(*args):
-            pass
-
-        sys.setprofile(fn)
-        assert sys.getprofile() == fn
-
-class HookWatcher:
-    def __init__(self):
-        self.frames = []
-        self.events = []
-
-    def callback(self, frame, event, arg):
-        if (event == "call"
-            or event == "return"
-            or event == "exception"):
-            self.add_event(event, frame)
-
-    def add_event(self, event, frame=None):
-        """Add an event to the log."""
-        if frame is None:
-            frame = sys._getframe(1)
-
-        try:
-            frameno = self.frames.index(frame)
-        except ValueError:
-            frameno = len(self.frames)
-            self.frames.append(frame)
-
-        self.events.append((frameno, event, ident(frame)))
-
-    def get_events(self):
-        """Remove calls to add_event()."""
-        disallowed = [ident(self.add_event.__func__), ident(ident)]
-        self.frames = None
-
-        return [item for item in self.events if item[2] not in disallowed]
-
-
-class ProfileSimulator(HookWatcher):
-    def __init__(self, testcase):
-        self.testcase = testcase
-        self.stack = []
-        HookWatcher.__init__(self)
-
-    def callback(self, frame, event, arg):
-        # Callback registered with sys.setprofile()/sys.settrace()
-        self.dispatch[event](self, frame)
-
-    def trace_call(self, frame):
-        self.add_event('call', frame)
-        self.stack.append(frame)
-
-    def trace_return(self, frame):
-        self.add_event('return', frame)
-        self.stack.pop()
-
-    def trace_exception(self, frame):
-        self.testcase.fail(
-            "the profiler should never receive exception events")
-
-    def trace_pass(self, frame):
-        pass
-
-    dispatch = {
-        'call': trace_call,
-        'exception': trace_exception,
-        'return': trace_return,
-        'c_call': trace_pass,
-        'c_return': trace_pass,
-        'c_exception': trace_pass,
-        }
-
-
-class TestCaseBase(unittest.TestCase):
-    def check_events(self, callable, expected):
-        events = capture_events(callable, self.new_watcher())
-        if events != expected:
-            self.fail("Expected events:\n%s\nReceived events:\n%s"
-                      % (pprint.pformat(expected), pprint.pformat(events)))
-
-
-class ProfileHookTestCase(TestCaseBase):
-    def new_watcher(self):
-        return HookWatcher()
-
-    def test_simple(self):
-        def f(p):
-            pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_exception(self):
-        def f(p):
-            1/0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_caught_exception(self):
-        def f(p):
-            try: 1/0
-            except: pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_caught_nested_exception(self):
-        def f(p):
-            try: 1/0
-            except: pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_nested_exception(self):
-        def f(p):
-            1/0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              # This isn't what I expected:
-                              # (0, 'exception', protect_ident),
-                              # I expected this again:
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_exception_in_except_clause(self):
-        def f(p):
-            1/0
-        def g(p):
-            try:
-                f(p)
-            except:
-                try: f(p)
-                except: pass
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (3, 'call', f_ident),
-                              (3, 'return', f_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-    def test_exception_propogation(self):
-        def f(p):
-            1/0
-        def g(p):
-            try: f(p)
-            finally: p.add_event("falling through")
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (1, 'falling through', g_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-    def test_raise_twice(self):
-        def f(p):
-            try: 1/0
-            except: 1/0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_raise_reraise(self):
-        def f(p):
-            try: 1/0
-            except: raise
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_raise(self):
-        def f(p):
-            raise Exception()
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_distant_exception(self):
-        def f():
-            1/0
-        def g():
-            f()
-        def h():
-            g()
-        def i():
-            h()
-        def j(p):
-            i()
-        f_ident = ident(f)
-        g_ident = ident(g)
-        h_ident = ident(h)
-        i_ident = ident(i)
-        j_ident = ident(j)
-        self.check_events(j, [(1, 'call', j_ident),
-                              (2, 'call', i_ident),
-                              (3, 'call', h_ident),
-                              (4, 'call', g_ident),
-                              (5, 'call', f_ident),
-                              (5, 'return', f_ident),
-                              (4, 'return', g_ident),
-                              (3, 'return', h_ident),
-                              (2, 'return', i_ident),
-                              (1, 'return', j_ident),
-                              ])
-
-    def test_generator(self):
-        def f():
-            for i in range(2):
-                yield i
-        def g(p):
-            for i in f():
-                pass
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              # call the iterator twice to generate values
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              # once more; returns end-of-iteration with
-                              # actually raising an exception
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-    def test_stop_iteration(self):
-        def f():
-            for i in range(2):
-                yield i
-            raise StopIteration
-        def g(p):
-            for i in f():
-                pass
-        f_ident = ident(f)
-        g_ident = ident(g)
-        self.check_events(g, [(1, 'call', g_ident),
-                              # call the iterator twice to generate values
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              # once more to hit the raise:
-                              (2, 'call', f_ident),
-                              (2, 'return', f_ident),
-                              (1, 'return', g_ident),
-                              ])
-
-
-class ProfileSimulatorTestCase(TestCaseBase):
-    def new_watcher(self):
-        return ProfileSimulator(self)
-
-    def test_simple(self):
-        def f(p):
-            pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_basic_exception(self):
-        def f(p):
-            1/0
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_caught_exception(self):
-        def f(p):
-            try: 1/0
-            except: pass
-        f_ident = ident(f)
-        self.check_events(f, [(1, 'call', f_ident),
-                              (1, 'return', f_ident),
-                              ])
-
-    def test_distant_exception(self):
-        def f():
-            1/0
-        def g():
-            f()
-        def h():
-            g()
-        def i():
-            h()
-        def j(p):
-            i()
-        f_ident = ident(f)
-        g_ident = ident(g)
-        h_ident = ident(h)
-        i_ident = ident(i)
-        j_ident = ident(j)
-        self.check_events(j, [(1, 'call', j_ident),
-                              (2, 'call', i_ident),
-                              (3, 'call', h_ident),
-                              (4, 'call', g_ident),
-                              (5, 'call', f_ident),
-                              (5, 'return', f_ident),
-                              (4, 'return', g_ident),
-                              (3, 'return', h_ident),
-                              (2, 'return', i_ident),
-                              (1, 'return', j_ident),
-                              ])
-
-
-def ident(function):
-    if hasattr(function, "f_code"):
-        code = function.f_code
-    else:
-        code = function.__code__
-    return code.co_firstlineno, code.co_name
-
-
-def protect(f, p):
-    try: f(p)
-    except: pass
-
-protect_ident = ident(protect)
-
-
-def capture_events(callable, p=None):
-    try:
-        sys.setprofile()
-    except TypeError:
-        pass
-    else:
-        raise support.TestFailed(
-            'sys.setprofile() did not raise TypeError')
-
-    if p is None:
-        p = HookWatcher()
-    sys.setprofile(p.callback)
-    protect(callable, p)
-    sys.setprofile(None)
-    return p.get_events()[1:-1]
-
-
-def show_events(callable):
-    import pprint
-    pprint.pprint(capture_events(callable))
-
-
-def test_main():
-    support.run_unittest(
-        TestGetProfile,
-        ProfileHookTestCase,
-        ProfileSimulatorTestCase
-    )
-
-
-if __name__ == "__main__":
-    test_main()

Modified: python/branches/import_unicode/Lib/test/test_queue.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_queue.py	(original)
+++ python/branches/import_unicode/Lib/test/test_queue.py	Fri Jul 30 01:19:37 2010
@@ -90,6 +90,8 @@
     def simple_queue_test(self, q):
         if q.qsize():
             raise RuntimeError("Call this function with an empty queue")
+        self.assertTrue(q.empty())
+        self.assertFalse(q.full())
         # I guess we better check things actually queue correctly a little :)
         q.put(111)
         q.put(333)
@@ -108,6 +110,8 @@
         full = 3 * 2 * QUEUE_SIZE
         q.put(last)
         self.assertTrue(qfull(q), "Queue should be full")
+        self.assertFalse(q.empty())
+        self.assertTrue(q.full())
         try:
             q.put(full, block=0)
             self.fail("Didn't appear to block with a full queue")
@@ -193,6 +197,25 @@
         self.simple_queue_test(q)
         self.simple_queue_test(q)
 
+    def test_negative_timeout_raises_exception(self):
+        q = self.type2test(QUEUE_SIZE)
+        with self.assertRaises(ValueError):
+            q.put(1, timeout=-1)
+        with self.assertRaises(ValueError):
+            q.get(1, timeout=-1)
+
+    def test_nowait(self):
+        q = self.type2test(QUEUE_SIZE)
+        for i in range(QUEUE_SIZE):
+            q.put_nowait(1)
+        with self.assertRaises(queue.Full):
+            q.put_nowait(1)
+
+        for i in range(QUEUE_SIZE):
+            q.get_nowait()
+        with self.assertRaises(queue.Empty):
+            q.get_nowait()
+
 
 class QueueTest(BaseQueueTest):
     type2test = queue.Queue

Modified: python/branches/import_unicode/Lib/test/test_re.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_re.py	(original)
+++ python/branches/import_unicode/Lib/test/test_re.py	Fri Jul 30 01:19:37 2010
@@ -874,8 +874,71 @@
                 if result is None:
                     print('=== Fails on unicode-sensitive match', t)
 
+
+class ReCacheTests(unittest.TestCase):
+    """These tests are specific to the re._shrink_cache implementation."""
+
+    def setUp(self):
+        self._orig_maxcache = re._MAXCACHE
+
+    def tearDown(self):
+        re._MAXCACHE = self._orig_maxcache
+
+    def test_compile_cache_overflow(self):
+        # NOTE: If a profiler or debugger is tracing code and compiling
+        # regular expressions while tracing through this test... expect
+        # the test to fail.  This test is not concurrency safe.
+
+        # Explicitly fill the caches.
+        re._MAXCACHE = 20
+        max_cache = re._MAXCACHE
+        unique_chars = tuple(chr(char_num) for char_num in
+                             range(b'a'[0], b'a'[0]+max_cache))
+        re._cache.clear()
+        for char in unique_chars:
+            re._compile(char, 0)
+        self.assertEqual(max_cache, len(re._cache))
+        re._cache_repl.clear()
+        for char in unique_chars:
+            re._compile_repl(char*2, char)
+        self.assertEqual(max_cache, len(re._cache_repl))
+
+        # Overflow both caches and make sure they have extra room left
+        # afterwards as well as having more than a single entry.
+        re._compile('A', 0)
+        self.assertLess(len(re._cache), max_cache)
+        self.assertGreater(len(re._cache), 1)
+        re._compile_repl('A', 'A')
+        self.assertLess(len(re._cache_repl), max_cache)
+        self.assertGreater(len(re._cache_repl), 1)
+
+    def test_shrink_cache_at_limit(self):
+        cache = dict(zip(range(6), range(6)))
+        re._shrink_cache(cache, 6, divisor=3)
+        self.assertEqual(4, len(cache))
+
+    def test_shrink_cache_empty(self):
+        cache = {}
+        re._shrink_cache(cache, 6, divisor=3)
+        # Cache was empty, make sure we didn't raise an exception.
+        self.assertEqual(0, len(cache))
+
+    def test_shrink_cache_overflowing(self):
+        cache = dict(zip(range(6), range(6)))
+        re._shrink_cache(cache, 4, divisor=2)
+        # Cache was larger than the maximum, be sure we shrunk to smaller.
+        self.assertEqual(2, len(cache))
+
+    def test_shrink_cache_underflow(self):
+        cache = dict(zip(range(6), range(6)))
+        # No shrinking to do.
+        re._shrink_cache(cache, 9, divisor=3)
+        self.assertEqual(6, len(cache))
+
+
 def test_main():
     run_unittest(ReTests)
+    run_unittest(ReCacheTests)
     run_re_tests()
 
 if __name__ == "__main__":

Modified: python/branches/import_unicode/Lib/test/test_robotparser.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_robotparser.py	(original)
+++ python/branches/import_unicode/Lib/test/test_robotparser.py	Fri Jul 30 01:19:37 2010
@@ -205,6 +205,31 @@
 RobotTest(13, doc, good, bad, agent="googlebot")
 
 
+# 14. For issue #6325 (query string support)
+doc = """
+User-agent: *
+Disallow: /some/path?name=value
+"""
+
+good = ['/some/path']
+bad = ['/some/path?name=value']
+
+RobotTest(14, doc, good, bad)
+
+# 15. For issue #4108 (obey first * entry)
+doc = """
+User-agent: *
+Disallow: /some/path
+
+User-agent: *
+Disallow: /another/path
+"""
+
+good = ['/another/path']
+bad = ['/some/path']
+
+RobotTest(15, doc, good, bad)
+
 
 class NetworkTestCase(unittest.TestCase):
 

Modified: python/branches/import_unicode/Lib/test/test_struct.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_struct.py	(original)
+++ python/branches/import_unicode/Lib/test/test_struct.py	Fri Jul 30 01:19:37 2010
@@ -47,8 +47,8 @@
 
         self.assertRaises(struct.error, struct.pack, 'iii', 3)
         self.assertRaises(struct.error, struct.pack, 'i', 3, 3, 3)
-        self.assertRaises(struct.error, struct.pack, 'i', 'foo')
-        self.assertRaises(struct.error, struct.pack, 'P', 'foo')
+        self.assertRaises((TypeError, struct.error), struct.pack, 'i', 'foo')
+        self.assertRaises((TypeError, struct.error), struct.pack, 'P', 'foo')
         self.assertRaises(struct.error, struct.unpack, 'd', b'flap')
         s = struct.pack('ii', 1, 2)
         self.assertRaises(struct.error, struct.unpack, 'iii', s)
@@ -213,6 +213,7 @@
                     expected = '%x' % expected
                     if len(expected) & 1:
                         expected = "0" + expected
+                    expected = expected.encode('ascii')
                     expected = unhexlify(expected)
                     expected = (b"\x00" * (self.bytesize - len(expected)) +
                                 expected)
@@ -234,7 +235,8 @@
                                                                  b'\x01' + got)
                 else:
                     # x is out of range -- verify pack realizes that.
-                    self.assertRaises(struct.error, pack, format, x)
+                    self.assertRaises((OverflowError, ValueError, struct.error),
+                                      pack, format, x)
 
             def run(self):
                 from random import randrange
@@ -445,7 +447,8 @@
 
         # Test bogus offset (issue 3694)
         sb = small_buf
-        self.assertRaises(TypeError, struct.pack_into, b'', sb, None)
+        self.assertRaises((TypeError, struct.error), struct.pack_into, b'', sb,
+                          None)
 
     def test_pack_into_fn(self):
         test_string = b'Reykjavik rocks, eow!'
@@ -510,8 +513,13 @@
                 self.assertFalse(prefix, msg='encoded bool is not one byte: %r'
                                              %packed)
 
-            self.assertRaises(IOError, struct.pack, prefix + '?',
-                              ExplodingBool())
+            try:
+                struct.pack(prefix + '?', ExplodingBool())
+            except IOError:
+                pass
+            else:
+                self.fail("Expected IOError: struct.pack(%r, "
+                          "ExplodingBool())" % (prefix + '?'))
 
         for c in [b'\x01', b'\x7f', b'\xff', b'\x0f', b'\xf0']:
             self.assertTrue(struct.unpack('>?', c)[0])

Modified: python/branches/import_unicode/Lib/test/test_structseq.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_structseq.py	(original)
+++ python/branches/import_unicode/Lib/test/test_structseq.py	Fri Jul 30 01:19:37 2010
@@ -8,7 +8,7 @@
 
     def test_tuple(self):
         t = time.gmtime()
-        assert isinstance(t, tuple)
+        self.assertIsInstance(t, tuple)
         astuple = tuple(t)
         self.assertEqual(len(t), len(astuple))
         self.assertEqual(t, astuple)
@@ -38,7 +38,7 @@
         # os.stat() gives a complicated struct sequence.
         st = os.stat(__file__)
         rep = repr(st)
-        self.assertTrue(rep.startswith("posix.stat_result"))
+        self.assertTrue(rep.startswith(os.name + ".stat_result"))
         self.assertIn("st_mode=", rep)
         self.assertIn("st_ino=", rep)
         self.assertIn("st_dev=", rep)

Modified: python/branches/import_unicode/Lib/test/test_subprocess.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_subprocess.py	(original)
+++ python/branches/import_unicode/Lib/test/test_subprocess.py	Fri Jul 30 01:19:37 2010
@@ -559,6 +559,21 @@
         except (ImportError, ValueError, resource.error):
             pass
 
+        if sys.platform == 'darwin':
+            # Check if the 'Crash Reporter' on OSX was configured
+            # in 'Developer' mode and warn that it will get triggered
+            # when it is.
+            #
+            # This assumes that this context manager is used in tests
+            # that might trigger the next manager.
+            value = subprocess.Popen(['/usr/bin/defaults', 'read',
+                    'com.apple.CrashReporter', 'DialogType'],
+                    stdout=subprocess.PIPE).communicate()[0]
+            if value.strip() == b'developer':
+                print("this tests triggers the Crash Reporter, "
+                      "that is intentional", end='')
+                sys.stdout.flush()
+
     def __exit__(self, *args):
         """Return core file behavior to default."""
         if self.old_limit is None:
@@ -750,6 +765,25 @@
         os.remove(fname)
         self.assertEqual(rc, 47)
 
+    def test_specific_shell(self):
+        # Issue #9265: Incorrect name passed as arg[0].
+        shells = []
+        for prefix in ['/bin', '/usr/bin/', '/usr/local/bin']:
+            for name in ['bash', 'ksh']:
+                sh = os.path.join(prefix, name)
+                if os.path.isfile(sh):
+                    shells.append(sh)
+        if not shells: # Will probably work for any shell but csh.
+            self.skipTest("bash or ksh required for this test")
+        sh = '/bin/sh'
+        if os.path.isfile(sh) and not os.path.islink(sh):
+            # Test will fail if /bin/sh is a symlink to csh.
+            shells.append(sh)
+        for sh in shells:
+            p = subprocess.Popen("echo $0", executable=sh, shell=True,
+                                 stdout=subprocess.PIPE)
+            self.assertEqual(p.stdout.read().strip(), bytes(sh, 'ascii'))
+
     def _kill_process(self, method, *args):
         # Do not inherit file handles from the parent.
         # It should fix failures on some platforms.

Modified: python/branches/import_unicode/Lib/test/test_sys.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_sys.py	(original)
+++ python/branches/import_unicode/Lib/test/test_sys.py	Fri Jul 30 01:19:37 2010
@@ -86,7 +86,6 @@
     # Python/pythonrun.c::PyErr_PrintEx() is tricky.
 
     def test_exit(self):
-        import subprocess
 
         self.assertRaises(TypeError, sys.exit, 42, 42)
 
@@ -179,8 +178,8 @@
         # can't check more than the type, as the user might have changed it
         self.assertIsInstance(sys.getdefaultencoding(), str)
 
-    # testing sys.settrace() is done in test_trace.py
-    # testing sys.setprofile() is done in test_profile.py
+    # testing sys.settrace() is done in test_sys_settrace.py
+    # testing sys.setprofile() is done in test_sys_setprofile.py
 
     def test_setcheckinterval(self):
         with warnings.catch_warnings():
@@ -530,7 +529,6 @@
         sys._clear_type_cache()
 
     def test_ioencoding(self):
-        import subprocess
         env = dict(os.environ)
 
         # Test character: cent sign, encoded as 0x4A (ASCII J) in CP424,
@@ -552,7 +550,7 @@
         # Issue #7774: Ensure that sys.executable is an empty string if argv[0]
         # has been set to an non existent program name and Python is unable to
         # retrieve the real program name
-        import subprocess
+
         # For a normal installation, it should work without 'cwd'
         # argument. For test runs in the build directory, see #7774.
         python_dir = os.path.dirname(os.path.realpath(sys.executable))

Modified: python/branches/import_unicode/Lib/test/test_sysconfig.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_sysconfig.py	(original)
+++ python/branches/import_unicode/Lib/test/test_sysconfig.py	Fri Jul 30 01:19:37 2010
@@ -244,8 +244,8 @@
         # On Windows, the EXE needs to know where pythonXY.dll is at so we have
         # to add the directory to the path.
         if sys.platform == "win32":
-            os.environ["Path"] = "{};{}".format(os.path.dirname(sys.executable),
-                                                os.environ["Path"])
+            os.environ["Path"] = "{};{}".format(
+                os.path.dirname(sys.executable), os.environ["Path"])
 
         # Issue 7880
         def get(python):
@@ -277,6 +277,13 @@
             _main()
         self.assertTrue(len(output.getvalue().split('\n')) > 0)
 
+    @unittest.skipIf(sys.platform == "win32", "Does not apply to Windows")
+    def test_ldshared_value(self):
+        ldflags = sysconfig.get_config_var('LDFLAGS')
+        ldshared = sysconfig.get_config_var('LDSHARED')
+
+        self.assertIn(ldflags, ldshared)
+
 
 def test_main():
     run_unittest(TestSysConfig)

Modified: python/branches/import_unicode/Lib/test/test_tarfile.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_tarfile.py	(original)
+++ python/branches/import_unicode/Lib/test/test_tarfile.py	Fri Jul 30 01:19:37 2010
@@ -291,7 +291,8 @@
         self.assertTrue(self.tar.getmembers()[-1].name == "misc/eof",
                 "could not find all members")
 
-    @unittest.skipUnless(hasattr(os, "link"), "Missing hardlink implementation")
+    @unittest.skipUnless(hasattr(os, "link"),
+                         "Missing hardlink implementation")
     @support.skip_unless_symlink
     def test_extract_hardlink(self):
         # Test hardlink extraction (e.g. bug #857297).
@@ -1423,11 +1424,13 @@
     def test_hardlink_extraction2(self):
         self._test_link_extraction("./ustar/linktest2/lnktype")
 
-    @unittest.skipIf(hasattr(os, "symlink"), "Skip emulation if symlink exists")
+    @unittest.skipIf(hasattr(os, "symlink"),
+                     "Skip emulation if symlink exists")
     def test_symlink_extraction1(self):
         self._test_link_extraction("ustar/symtype")
 
-    @unittest.skipIf(hasattr(os, "symlink"), "Skip emulation if symlink exists")
+    @unittest.skipIf(hasattr(os, "symlink"),
+                     "Skip emulation if symlink exists")
     def test_symlink_extraction2(self):
         self._test_link_extraction("./ustar/linktest2/symtype")
 

Modified: python/branches/import_unicode/Lib/test/test_threaded_import.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_threaded_import.py	(original)
+++ python/branches/import_unicode/Lib/test/test_threaded_import.py	Fri Jul 30 01:19:37 2010
@@ -5,72 +5,77 @@
 # complains several times about module random having no attribute
 # randrange, and then Python hangs.
 
+import imp
+import sys
 import unittest
-from test.support import verbose, TestFailed, import_module
+from test.support import verbose, TestFailed, import_module, run_unittest
 thread = import_module('_thread')
 
-critical_section = thread.allocate_lock()
-done = thread.allocate_lock()
+def task(N, done, done_tasks, errors):
+    try:
+        import random
+        # This will fail if random is not completely initialized
+        x = random.randrange(1, 3)
+    except Exception as e:
+        errors.append(e.with_traceback(None))
+    finally:
+        done_tasks.append(thread.get_ident())
+        finished = len(done_tasks) == N
+        if finished:
+            done.release()
+
+
+class ThreadedImportTests(unittest.TestCase):
+
+    def setUp(self):
+        self.old_random = sys.modules.pop('random', None)
+
+    def tearDown(self):
+        # If the `random` module was already initialized, we restore the
+        # old module at the end so that pickling tests don't fail.
+        # See http://bugs.python.org/issue3657#msg110461
+        if self.old_random is not None:
+            sys.modules['random'] = self.old_random
+
+    def test_parallel_module_init(self):
+        if imp.lock_held():
+            # This triggers on, e.g., from test import autotest.
+            raise unittest.SkipTest("can't run when import lock is held")
 
-def task():
-    global N, critical_section, done
-    import random
-    x = random.randrange(1, 3)
-    critical_section.acquire()
-    N -= 1
-    # Must release critical_section before releasing done, else the main
-    # thread can exit and set critical_section to None as part of global
-    # teardown; then critical_section.release() raises AttributeError.
-    finished = N == 0
-    critical_section.release()
-    if finished:
+        done = thread.allocate_lock()
+        done.acquire()
+        for N in (20, 50) * 3:
+            if verbose:
+                print("Trying", N, "threads ...", end=' ')
+            # Make sure that random gets reimported freshly
+            try:
+                del sys.modules['random']
+            except KeyError:
+                pass
+            errors = []
+            done_tasks = []
+            for i in range(N):
+                thread.start_new_thread(task, (N, done, done_tasks, errors,))
+            done.acquire()
+            self.assertFalse(errors)
+            if verbose:
+                print("OK.")
         done.release()
 
-def test_import_hangers():
-    import sys
-    if verbose:
-        print("testing import hangers ...", end=' ')
-
-    import test.threaded_import_hangers
-    try:
-        if test.threaded_import_hangers.errors:
-            raise TestFailed(test.threaded_import_hangers.errors)
-        elif verbose:
-            print("OK.")
-    finally:
+    def test_import_hangers(self):
         # In case this test is run again, make sure the helper module
         # gets loaded from scratch again.
-        del sys.modules['test.threaded_import_hangers']
+        try:
+            del sys.modules['test.threaded_import_hangers']
+        except KeyError:
+            pass
+        import test.threaded_import_hangers
+        self.assertFalse(test.threaded_import_hangers.errors)
 
-# Tricky:  When regrtest imports this module, the thread running regrtest
-# grabs the import lock and won't let go of it until this module returns.
-# All other threads attempting an import hang for the duration.  Since
-# this test spawns threads that do little *but* import, we can't do that
-# successfully until after this module finishes importing and regrtest
-# regains control.  To make this work, a special case was added to
-# regrtest to invoke a module's "test_main" function (if any) after
-# importing it.
-
-def test_main():        # magic name!  see above
-    global N, done
-
-    import imp
-    if imp.lock_held():
-        # This triggers on, e.g., from test import autotest.
-        raise unittest.SkipTest("can't run when import lock is held")
-
-    done.acquire()
-    for N in (20, 50) * 3:
-        if verbose:
-            print("Trying", N, "threads ...", end=' ')
-        for i in range(N):
-            thread.start_new_thread(task, ())
-        done.acquire()
-        if verbose:
-            print("OK.")
-    done.release()
 
-    test_import_hangers()
+def test_main():
+    run_unittest(ThreadedImportTests)
+
 
 if __name__ == "__main__":
     test_main()

Modified: python/branches/import_unicode/Lib/test/test_threading.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_threading.py	(original)
+++ python/branches/import_unicode/Lib/test/test_threading.py	Fri Jul 30 01:19:37 2010
@@ -97,7 +97,8 @@
             self.assertTrue(not t.is_alive())
             self.assertNotEqual(t.ident, 0)
             self.assertFalse(t.ident is None)
-            self.assertTrue(re.match('<TestThread\(.*, \w+ -?\d+\)>', repr(t)))
+            self.assertTrue(re.match('<TestThread\(.*, stopped -?\d+\)>',
+                                     repr(t)))
         if verbose:
             print('all tasks done')
         self.assertEqual(numrunning.get(), 0)
@@ -413,6 +414,12 @@
         e.isSet()
         threading.activeCount()
 
+    def test_repr_daemon(self):
+        t = threading.Thread()
+        self.assertFalse('daemon' in repr(t))
+        t.daemon = True
+        self.assertTrue('daemon' in repr(t))
+
 
 class ThreadJoinOnShutdown(BaseTestCase):
 

Deleted: python/branches/import_unicode/Lib/test/test_trace.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_trace.py	Fri Jul 30 01:19:37 2010
+++ (empty file)
@@ -1,790 +0,0 @@
-# Testing the line trace facility.
-
-from test import support
-import unittest
-import sys
-import difflib
-import gc
-
-# A very basic example.  If this fails, we're in deep trouble.
-def basic():
-    return 1
-
-basic.events = [(0, 'call'),
-                (1, 'line'),
-                (1, 'return')]
-
-# Many of the tests below are tricky because they involve pass statements.
-# If there is implicit control flow around a pass statement (in an except
-# clause or else caluse) under what conditions do you set a line number
-# following that clause?
-
-
-# The entire "while 0:" statement is optimized away.  No code
-# exists for it, so the line numbers skip directly from "del x"
-# to "x = 1".
-def arigo_example():
-    x = 1
-    del x
-    while 0:
-        pass
-    x = 1
-
-arigo_example.events = [(0, 'call'),
-                        (1, 'line'),
-                        (2, 'line'),
-                        (5, 'line'),
-                        (5, 'return')]
-
-# check that lines consisting of just one instruction get traced:
-def one_instr_line():
-    x = 1
-    del x
-    x = 1
-
-one_instr_line.events = [(0, 'call'),
-                         (1, 'line'),
-                         (2, 'line'),
-                         (3, 'line'),
-                         (3, 'return')]
-
-def no_pop_tops():      # 0
-    x = 1               # 1
-    for a in range(2):  # 2
-        if a:           # 3
-            x = 1       # 4
-        else:           # 5
-            x = 1       # 6
-
-no_pop_tops.events = [(0, 'call'),
-                      (1, 'line'),
-                      (2, 'line'),
-                      (3, 'line'),
-                      (6, 'line'),
-                      (2, 'line'),
-                      (3, 'line'),
-                      (4, 'line'),
-                      (2, 'line'),
-                      (2, 'return')]
-
-def no_pop_blocks():
-    y = 1
-    while not y:
-        bla
-    x = 1
-
-no_pop_blocks.events = [(0, 'call'),
-                        (1, 'line'),
-                        (2, 'line'),
-                        (4, 'line'),
-                        (4, 'return')]
-
-def called(): # line -3
-    x = 1
-
-def call():   # line 0
-    called()
-
-call.events = [(0, 'call'),
-               (1, 'line'),
-               (-3, 'call'),
-               (-2, 'line'),
-               (-2, 'return'),
-               (1, 'return')]
-
-def raises():
-    raise Exception
-
-def test_raise():
-    try:
-        raises()
-    except Exception as exc:
-        x = 1
-
-test_raise.events = [(0, 'call'),
-                     (1, 'line'),
-                     (2, 'line'),
-                     (-3, 'call'),
-                     (-2, 'line'),
-                     (-2, 'exception'),
-                     (-2, 'return'),
-                     (2, 'exception'),
-                     (3, 'line'),
-                     (4, 'line'),
-                     (4, 'return')]
-
-def _settrace_and_return(tracefunc):
-    sys.settrace(tracefunc)
-    sys._getframe().f_back.f_trace = tracefunc
-def settrace_and_return(tracefunc):
-    _settrace_and_return(tracefunc)
-
-settrace_and_return.events = [(1, 'return')]
-
-def _settrace_and_raise(tracefunc):
-    sys.settrace(tracefunc)
-    sys._getframe().f_back.f_trace = tracefunc
-    raise RuntimeError
-def settrace_and_raise(tracefunc):
-    try:
-        _settrace_and_raise(tracefunc)
-    except RuntimeError as exc:
-        pass
-
-settrace_and_raise.events = [(2, 'exception'),
-                             (3, 'line'),
-                             (4, 'line'),
-                             (4, 'return')]
-
-# implicit return example
-# This test is interesting because of the else: pass
-# part of the code.  The code generate for the true
-# part of the if contains a jump past the else branch.
-# The compiler then generates an implicit "return None"
-# Internally, the compiler visits the pass statement
-# and stores its line number for use on the next instruction.
-# The next instruction is the implicit return None.
-def ireturn_example():
-    a = 5
-    b = 5
-    if a == b:
-        b = a+1
-    else:
-        pass
-
-ireturn_example.events = [(0, 'call'),
-                          (1, 'line'),
-                          (2, 'line'),
-                          (3, 'line'),
-                          (4, 'line'),
-                          (6, 'line'),
-                          (6, 'return')]
-
-# Tight loop with while(1) example (SF #765624)
-def tightloop_example():
-    items = range(0, 3)
-    try:
-        i = 0
-        while 1:
-            b = items[i]; i+=1
-    except IndexError:
-        pass
-
-tightloop_example.events = [(0, 'call'),
-                            (1, 'line'),
-                            (2, 'line'),
-                            (3, 'line'),
-                            (4, 'line'),
-                            (5, 'line'),
-                            (5, 'line'),
-                            (5, 'line'),
-                            (5, 'line'),
-                            (5, 'exception'),
-                            (6, 'line'),
-                            (7, 'line'),
-                            (7, 'return')]
-
-def tighterloop_example():
-    items = range(1, 4)
-    try:
-        i = 0
-        while 1: i = items[i]
-    except IndexError:
-        pass
-
-tighterloop_example.events = [(0, 'call'),
-                            (1, 'line'),
-                            (2, 'line'),
-                            (3, 'line'),
-                            (4, 'line'),
-                            (4, 'line'),
-                            (4, 'line'),
-                            (4, 'line'),
-                            (4, 'exception'),
-                            (5, 'line'),
-                            (6, 'line'),
-                            (6, 'return')]
-
-def generator_function():
-    try:
-        yield True
-        "continued"
-    finally:
-        "finally"
-def generator_example():
-    # any() will leave the generator before its end
-    x = any(generator_function())
-
-    # the following lines were not traced
-    for x in range(10):
-        y = x
-
-generator_example.events = ([(0, 'call'),
-                             (2, 'line'),
-                             (-6, 'call'),
-                             (-5, 'line'),
-                             (-4, 'line'),
-                             (-4, 'return'),
-                             (-4, 'call'),
-                             (-4, 'exception'),
-                             (-1, 'line'),
-                             (-1, 'return')] +
-                            [(5, 'line'), (6, 'line')] * 10 +
-                            [(5, 'line'), (5, 'return')])
-
-
-class Tracer:
-    def __init__(self):
-        self.events = []
-    def trace(self, frame, event, arg):
-        self.events.append((frame.f_lineno, event))
-        return self.trace
-    def traceWithGenexp(self, frame, event, arg):
-        (o for o in [1])
-        self.events.append((frame.f_lineno, event))
-        return self.trace
-
-class TraceTestCase(unittest.TestCase):
-
-    # Disable gc collection when tracing, otherwise the
-    # deallocators may be traced as well.
-    def setUp(self):
-        self.using_gc = gc.isenabled()
-        gc.disable()
-
-    def tearDown(self):
-        if self.using_gc:
-            gc.enable()
-
-    def compare_events(self, line_offset, events, expected_events):
-        events = [(l - line_offset, e) for (l, e) in events]
-        if events != expected_events:
-            self.fail(
-                "events did not match expectation:\n" +
-                "\n".join(difflib.ndiff([str(x) for x in expected_events],
-                                        [str(x) for x in events])))
-
-    def run_and_compare(self, func, events):
-        tracer = Tracer()
-        sys.settrace(tracer.trace)
-        func()
-        sys.settrace(None)
-        self.compare_events(func.__code__.co_firstlineno,
-                            tracer.events, events)
-
-    def run_test(self, func):
-        self.run_and_compare(func, func.events)
-
-    def run_test2(self, func):
-        tracer = Tracer()
-        func(tracer.trace)
-        sys.settrace(None)
-        self.compare_events(func.__code__.co_firstlineno,
-                            tracer.events, func.events)
-
-    def set_and_retrieve_none(self):
-        sys.settrace(None)
-        assert sys.gettrace() is None
-
-    def set_and_retrieve_func(self):
-        def fn(*args):
-            pass
-
-        sys.settrace(fn)
-        try:
-            assert sys.gettrace() is fn
-        finally:
-            sys.settrace(None)
-
-    def test_01_basic(self):
-        self.run_test(basic)
-    def test_02_arigo(self):
-        self.run_test(arigo_example)
-    def test_03_one_instr(self):
-        self.run_test(one_instr_line)
-    def test_04_no_pop_blocks(self):
-        self.run_test(no_pop_blocks)
-    def test_05_no_pop_tops(self):
-        self.run_test(no_pop_tops)
-    def test_06_call(self):
-        self.run_test(call)
-    def test_07_raise(self):
-        self.run_test(test_raise)
-
-    def test_08_settrace_and_return(self):
-        self.run_test2(settrace_and_return)
-    def test_09_settrace_and_raise(self):
-        self.run_test2(settrace_and_raise)
-    def test_10_ireturn(self):
-        self.run_test(ireturn_example)
-    def test_11_tightloop(self):
-        self.run_test(tightloop_example)
-    def test_12_tighterloop(self):
-        self.run_test(tighterloop_example)
-
-    def test_13_genexp(self):
-        self.run_test(generator_example)
-        # issue1265: if the trace function contains a generator,
-        # and if the traced function contains another generator
-        # that is not completely exhausted, the trace stopped.
-        # Worse: the 'finally' clause was not invoked.
-        tracer = Tracer()
-        sys.settrace(tracer.traceWithGenexp)
-        generator_example()
-        sys.settrace(None)
-        self.compare_events(generator_example.__code__.co_firstlineno,
-                            tracer.events, generator_example.events)
-
-    def test_14_onliner_if(self):
-        def onliners():
-            if True: False
-            else: True
-            return 0
-        self.run_and_compare(
-            onliners,
-            [(0, 'call'),
-             (1, 'line'),
-             (3, 'line'),
-             (3, 'return')])
-
-    def test_15_loops(self):
-        # issue1750076: "while" expression is skipped by debugger
-        def for_example():
-            for x in range(2):
-                pass
-        self.run_and_compare(
-            for_example,
-            [(0, 'call'),
-             (1, 'line'),
-             (2, 'line'),
-             (1, 'line'),
-             (2, 'line'),
-             (1, 'line'),
-             (1, 'return')])
-
-        def while_example():
-            # While expression should be traced on every loop
-            x = 2
-            while x > 0:
-                x -= 1
-        self.run_and_compare(
-            while_example,
-            [(0, 'call'),
-             (2, 'line'),
-             (3, 'line'),
-             (4, 'line'),
-             (3, 'line'),
-             (4, 'line'),
-             (3, 'line'),
-             (3, 'return')])
-
-    def test_16_blank_lines(self):
-        namespace = {}
-        exec("def f():\n" + "\n" * 256 + "    pass", namespace)
-        self.run_and_compare(
-            namespace["f"],
-            [(0, 'call'),
-             (257, 'line'),
-             (257, 'return')])
-
-
-class RaisingTraceFuncTestCase(unittest.TestCase):
-    def trace(self, frame, event, arg):
-        """A trace function that raises an exception in response to a
-        specific trace event."""
-        if event == self.raiseOnEvent:
-            raise ValueError # just something that isn't RuntimeError
-        else:
-            return self.trace
-
-    def f(self):
-        """The function to trace; raises an exception if that's the case
-        we're testing, so that the 'exception' trace event fires."""
-        if self.raiseOnEvent == 'exception':
-            x = 0
-            y = 1/x
-        else:
-            return 1
-
-    def run_test_for_event(self, event):
-        """Tests that an exception raised in response to the given event is
-        handled OK."""
-        self.raiseOnEvent = event
-        try:
-            for i in range(sys.getrecursionlimit() + 1):
-                sys.settrace(self.trace)
-                try:
-                    self.f()
-                except ValueError:
-                    pass
-                else:
-                    self.fail("exception not thrown!")
-        except RuntimeError:
-            self.fail("recursion counter not reset")
-
-    # Test the handling of exceptions raised by each kind of trace event.
-    def test_call(self):
-        self.run_test_for_event('call')
-    def test_line(self):
-        self.run_test_for_event('line')
-    def test_return(self):
-        self.run_test_for_event('return')
-    def test_exception(self):
-        self.run_test_for_event('exception')
-
-    def test_trash_stack(self):
-        def f():
-            for i in range(5):
-                print(i)  # line tracing will raise an exception at this line
-
-        def g(frame, why, extra):
-            if (why == 'line' and
-                frame.f_lineno == f.__code__.co_firstlineno + 2):
-                raise RuntimeError("i am crashing")
-            return g
-
-        sys.settrace(g)
-        try:
-            f()
-        except RuntimeError:
-            # the test is really that this doesn't segfault:
-            import gc
-            gc.collect()
-        else:
-            self.fail("exception not propagated")
-
-
-# 'Jump' tests: assigning to frame.f_lineno within a trace function
-# moves the execution position - it's how debuggers implement a Jump
-# command (aka. "Set next statement").
-
-class JumpTracer:
-    """Defines a trace function that jumps from one place to another,
-    with the source and destination lines of the jump being defined by
-    the 'jump' property of the function under test."""
-
-    def __init__(self, function):
-        self.function = function
-        self.jumpFrom = function.jump[0]
-        self.jumpTo = function.jump[1]
-        self.done = False
-
-    def trace(self, frame, event, arg):
-        if not self.done and frame.f_code == self.function.__code__:
-            firstLine = frame.f_code.co_firstlineno
-            if event == 'line' and frame.f_lineno == firstLine + self.jumpFrom:
-                # Cope with non-integer self.jumpTo (because of
-                # no_jump_to_non_integers below).
-                try:
-                    frame.f_lineno = firstLine + self.jumpTo
-                except TypeError:
-                    frame.f_lineno = self.jumpTo
-                self.done = True
-        return self.trace
-
-# The first set of 'jump' tests are for things that are allowed:
-
-def jump_simple_forwards(output):
-    output.append(1)
-    output.append(2)
-    output.append(3)
-
-jump_simple_forwards.jump = (1, 3)
-jump_simple_forwards.output = [3]
-
-def jump_simple_backwards(output):
-    output.append(1)
-    output.append(2)
-
-jump_simple_backwards.jump = (2, 1)
-jump_simple_backwards.output = [1, 1, 2]
-
-def jump_out_of_block_forwards(output):
-    for i in 1, 2:
-        output.append(2)
-        for j in [3]:  # Also tests jumping over a block
-            output.append(4)
-    output.append(5)
-
-jump_out_of_block_forwards.jump = (3, 5)
-jump_out_of_block_forwards.output = [2, 5]
-
-def jump_out_of_block_backwards(output):
-    output.append(1)
-    for i in [1]:
-        output.append(3)
-        for j in [2]:  # Also tests jumping over a block
-            output.append(5)
-        output.append(6)
-    output.append(7)
-
-jump_out_of_block_backwards.jump = (6, 1)
-jump_out_of_block_backwards.output = [1, 3, 5, 1, 3, 5, 6, 7]
-
-def jump_to_codeless_line(output):
-    output.append(1)
-    # Jumping to this line should skip to the next one.
-    output.append(3)
-
-jump_to_codeless_line.jump = (1, 2)
-jump_to_codeless_line.output = [3]
-
-def jump_to_same_line(output):
-    output.append(1)
-    output.append(2)
-    output.append(3)
-
-jump_to_same_line.jump = (2, 2)
-jump_to_same_line.output = [1, 2, 3]
-
-# Tests jumping within a finally block, and over one.
-def jump_in_nested_finally(output):
-    try:
-        output.append(2)
-    finally:
-        output.append(4)
-        try:
-            output.append(6)
-        finally:
-            output.append(8)
-        output.append(9)
-
-jump_in_nested_finally.jump = (4, 9)
-jump_in_nested_finally.output = [2, 9]
-
-# The second set of 'jump' tests are for things that are not allowed:
-
-def no_jump_too_far_forwards(output):
-    try:
-        output.append(2)
-        output.append(3)
-    except ValueError as e:
-        output.append('after' in str(e))
-
-no_jump_too_far_forwards.jump = (3, 6)
-no_jump_too_far_forwards.output = [2, True]
-
-def no_jump_too_far_backwards(output):
-    try:
-        output.append(2)
-        output.append(3)
-    except ValueError as e:
-        output.append('before' in str(e))
-
-no_jump_too_far_backwards.jump = (3, -1)
-no_jump_too_far_backwards.output = [2, True]
-
-# Test each kind of 'except' line.
-def no_jump_to_except_1(output):
-    try:
-        output.append(2)
-    except:
-        e = sys.exc_info()[1]
-        output.append('except' in str(e))
-
-no_jump_to_except_1.jump = (2, 3)
-no_jump_to_except_1.output = [True]
-
-def no_jump_to_except_2(output):
-    try:
-        output.append(2)
-    except ValueError:
-        e = sys.exc_info()[1]
-        output.append('except' in str(e))
-
-no_jump_to_except_2.jump = (2, 3)
-no_jump_to_except_2.output = [True]
-
-def no_jump_to_except_3(output):
-    try:
-        output.append(2)
-    except ValueError as e:
-        output.append('except' in str(e))
-
-no_jump_to_except_3.jump = (2, 3)
-no_jump_to_except_3.output = [True]
-
-def no_jump_to_except_4(output):
-    try:
-        output.append(2)
-    except (ValueError, RuntimeError) as e:
-        output.append('except' in str(e))
-
-no_jump_to_except_4.jump = (2, 3)
-no_jump_to_except_4.output = [True]
-
-def no_jump_forwards_into_block(output):
-    try:
-        output.append(2)
-        for i in 1, 2:
-            output.append(4)
-    except ValueError as e:
-        output.append('into' in str(e))
-
-no_jump_forwards_into_block.jump = (2, 4)
-no_jump_forwards_into_block.output = [True]
-
-def no_jump_backwards_into_block(output):
-    try:
-        for i in 1, 2:
-            output.append(3)
-        output.append(4)
-    except ValueError as e:
-        output.append('into' in str(e))
-
-no_jump_backwards_into_block.jump = (4, 3)
-no_jump_backwards_into_block.output = [3, 3, True]
-
-def no_jump_into_finally_block(output):
-    try:
-        try:
-            output.append(3)
-            x = 1
-        finally:
-            output.append(6)
-    except ValueError as e:
-        output.append('finally' in str(e))
-
-no_jump_into_finally_block.jump = (4, 6)
-no_jump_into_finally_block.output = [3, 6, True]  # The 'finally' still runs
-
-def no_jump_out_of_finally_block(output):
-    try:
-        try:
-            output.append(3)
-        finally:
-            output.append(5)
-            output.append(6)
-    except ValueError as e:
-        output.append('finally' in str(e))
-
-no_jump_out_of_finally_block.jump = (5, 1)
-no_jump_out_of_finally_block.output = [3, True]
-
-# This verifies the line-numbers-must-be-integers rule.
-def no_jump_to_non_integers(output):
-    try:
-        output.append(2)
-    except ValueError as e:
-        output.append('integer' in str(e))
-
-no_jump_to_non_integers.jump = (2, "Spam")
-no_jump_to_non_integers.output = [True]
-
-# This verifies that you can't set f_lineno via _getframe or similar
-# trickery.
-def no_jump_without_trace_function():
-    try:
-        previous_frame = sys._getframe().f_back
-        previous_frame.f_lineno = previous_frame.f_lineno
-    except ValueError as e:
-        # This is the exception we wanted; make sure the error message
-        # talks about trace functions.
-        if 'trace' not in str(e):
-            raise
-    else:
-        # Something's wrong - the expected exception wasn't raised.
-        raise RuntimeError("Trace-function-less jump failed to fail")
-
-
-class JumpTestCase(unittest.TestCase):
-    def compare_jump_output(self, expected, received):
-        if received != expected:
-            self.fail( "Outputs don't match:\n" +
-                       "Expected: " + repr(expected) + "\n" +
-                       "Received: " + repr(received))
-
-    def run_test(self, func):
-        tracer = JumpTracer(func)
-        sys.settrace(tracer.trace)
-        output = []
-        func(output)
-        sys.settrace(None)
-        self.compare_jump_output(func.output, output)
-
-    def test_01_jump_simple_forwards(self):
-        self.run_test(jump_simple_forwards)
-    def test_02_jump_simple_backwards(self):
-        self.run_test(jump_simple_backwards)
-    def test_03_jump_out_of_block_forwards(self):
-        self.run_test(jump_out_of_block_forwards)
-    def test_04_jump_out_of_block_backwards(self):
-        self.run_test(jump_out_of_block_backwards)
-    def test_05_jump_to_codeless_line(self):
-        self.run_test(jump_to_codeless_line)
-    def test_06_jump_to_same_line(self):
-        self.run_test(jump_to_same_line)
-    def test_07_jump_in_nested_finally(self):
-        self.run_test(jump_in_nested_finally)
-    def test_08_no_jump_too_far_forwards(self):
-        self.run_test(no_jump_too_far_forwards)
-    def test_09_no_jump_too_far_backwards(self):
-        self.run_test(no_jump_too_far_backwards)
-    def test_10_no_jump_to_except_1(self):
-        self.run_test(no_jump_to_except_1)
-    def test_11_no_jump_to_except_2(self):
-        self.run_test(no_jump_to_except_2)
-    def test_12_no_jump_to_except_3(self):
-        self.run_test(no_jump_to_except_3)
-    def test_13_no_jump_to_except_4(self):
-        self.run_test(no_jump_to_except_4)
-    def test_14_no_jump_forwards_into_block(self):
-        self.run_test(no_jump_forwards_into_block)
-    def test_15_no_jump_backwards_into_block(self):
-        self.run_test(no_jump_backwards_into_block)
-    def test_16_no_jump_into_finally_block(self):
-        self.run_test(no_jump_into_finally_block)
-    def test_17_no_jump_out_of_finally_block(self):
-        self.run_test(no_jump_out_of_finally_block)
-    def test_18_no_jump_to_non_integers(self):
-        self.run_test(no_jump_to_non_integers)
-    def test_19_no_jump_without_trace_function(self):
-        no_jump_without_trace_function()
-
-    def test_20_large_function(self):
-        d = {}
-        exec("""def f(output):        # line 0
-            x = 0                     # line 1
-            y = 1                     # line 2
-            '''                       # line 3
-            %s                        # lines 4-1004
-            '''                       # line 1005
-            x += 1                    # line 1006
-            output.append(x)          # line 1007
-            return""" % ('\n' * 1000,), d)
-        f = d['f']
-
-        f.jump = (2, 1007)
-        f.output = [0]
-        self.run_test(f)
-
-    def test_jump_to_firstlineno(self):
-        # This tests that PDB can jump back to the first line in a
-        # file.  See issue #1689458.  It can only be triggered in a
-        # function call if the function is defined on a single line.
-        code = compile("""
-# Comments don't count.
-output.append(2)  # firstlineno is here.
-output.append(3)
-output.append(4)
-""", "<fake module>", "exec")
-        class fake_function:
-            __code__ = code
-            jump = (2, 0)
-        tracer = JumpTracer(fake_function)
-        sys.settrace(tracer.trace)
-        namespace = {"output": []}
-        exec(code, namespace)
-        sys.settrace(None)
-        self.compare_jump_output([2, 3, 2, 3, 4], namespace["output"])
-
-
-def test_main():
-    support.run_unittest(
-        TraceTestCase,
-        RaisingTraceFuncTestCase,
-        JumpTestCase
-    )
-
-if __name__ == "__main__":
-    test_main()

Modified: python/branches/import_unicode/Lib/test/test_unicode.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_unicode.py	(original)
+++ python/branches/import_unicode/Lib/test/test_unicode.py	Fri Jul 30 01:19:37 2010
@@ -757,6 +757,7 @@
         self.assertRaises(OverflowError, "%c".__mod__, (0x110000,))
         self.assertEqual('%c' % '\U00021483', '\U00021483')
         self.assertRaises(TypeError, "%c".__mod__, "aa")
+        self.assertRaises(ValueError, "%.1\u1032f".__mod__, (1.0/3))
 
         # formatting jobs delegated from the string implementation:
         self.assertEqual('...%(foo)s...' % {'foo':"abc"}, '...abc...')

Modified: python/branches/import_unicode/Lib/test/test_unpack.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_unpack.py	(original)
+++ python/branches/import_unicode/Lib/test/test_unpack.py	Fri Jul 30 01:19:37 2010
@@ -62,14 +62,14 @@
     >>> a, b = t
     Traceback (most recent call last):
       ...
-    ValueError: too many values to unpack
+    ValueError: too many values to unpack (expected 2)
 
 Unpacking tuple of wrong size
 
     >>> a, b = l
     Traceback (most recent call last):
       ...
-    ValueError: too many values to unpack
+    ValueError: too many values to unpack (expected 2)
 
 Unpacking sequence too short
 
@@ -83,7 +83,7 @@
     >>> a, b = Seq()
     Traceback (most recent call last):
       ...
-    ValueError: too many values to unpack
+    ValueError: too many values to unpack (expected 2)
 
 Unpacking a sequence where the test for too long raises a different kind of
 error

Modified: python/branches/import_unicode/Lib/test/test_urllib.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_urllib.py	(original)
+++ python/branches/import_unicode/Lib/test/test_urllib.py	Fri Jul 30 01:19:37 2010
@@ -558,6 +558,7 @@
         self.assertEqual(result.count('%'), 1,
                          "using unquote(): not all characters escaped: "
                          "%s" % result)
+        self.assertRaises(TypeError, urllib.parse.unquote, None)
 
     def test_unquoting_badpercent(self):
         # Test unquoting on bad percent-escapes
@@ -593,6 +594,8 @@
         self.assertEqual(expect, result, "using unquote_to_bytes(): %r != %r"
                          % (expect, result))
 
+        self.assertRaises(TypeError, urllib.parse.unquote_to_bytes, None)
+
     def test_unquoting_mixed_case(self):
         # Test unquoting on mixed-case hex digits in the percent-escapes
         given = '%Ab%eA'

Modified: python/branches/import_unicode/Lib/test/test_urllib2.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_urllib2.py	(original)
+++ python/branches/import_unicode/Lib/test/test_urllib2.py	Fri Jul 30 01:19:37 2010
@@ -735,6 +735,8 @@
             ("file://ftp.example.com///foo.txt", False),
 # XXXX bug: fails with OSError, should be URLError
             ("file://ftp.example.com/foo.txt", False),
+            ("file://somehost//foo/something.txt", True),
+            ("file://localhost//foo/something.txt", False),
             ]:
             req = Request(url)
             try:
@@ -745,6 +747,7 @@
             else:
                 self.assertTrue(o.req is req)
                 self.assertEqual(req.type, "ftp")
+            self.assertEqual(req.type is "ftp", ftp)
 
     def test_http(self):
 

Modified: python/branches/import_unicode/Lib/test/test_urlparse.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_urlparse.py	(original)
+++ python/branches/import_unicode/Lib/test/test_urlparse.py	Fri Jul 30 01:19:37 2010
@@ -7,6 +7,7 @@
 RFC1808_BASE = "http://a/b/c/d;p?q#f"
 RFC2396_BASE = "http://a/b/c/d;p?q"
 RFC3986_BASE = 'http://a/b/c/d;p?q'
+SIMPLE_BASE  = 'http://a/b/c/d'
 
 # A list of test cases.  Each test case is a a two-tuple that contains
 # a string with the query and a dictionary with the expected result.
@@ -295,6 +296,37 @@
         #self.checkJoin(RFC3986_BASE, 'http:g','http:g') # strict parser
         self.checkJoin(RFC3986_BASE, 'http:g','http://a/b/c/g') #relaxed parser
 
+    def test_urljoins(self):
+        self.checkJoin(SIMPLE_BASE, 'g:h','g:h')
+        self.checkJoin(SIMPLE_BASE, 'http:g','http://a/b/c/g')
+        self.checkJoin(SIMPLE_BASE, 'http:','http://a/b/c/d')
+        self.checkJoin(SIMPLE_BASE, 'g','http://a/b/c/g')
+        self.checkJoin(SIMPLE_BASE, './g','http://a/b/c/g')
+        self.checkJoin(SIMPLE_BASE, 'g/','http://a/b/c/g/')
+        self.checkJoin(SIMPLE_BASE, '/g','http://a/g')
+        self.checkJoin(SIMPLE_BASE, '//g','http://g')
+        self.checkJoin(SIMPLE_BASE, '?y','http://a/b/c/d?y')
+        self.checkJoin(SIMPLE_BASE, 'g?y','http://a/b/c/g?y')
+        self.checkJoin(SIMPLE_BASE, 'g?y/./x','http://a/b/c/g?y/./x')
+        self.checkJoin(SIMPLE_BASE, '.','http://a/b/c/')
+        self.checkJoin(SIMPLE_BASE, './','http://a/b/c/')
+        self.checkJoin(SIMPLE_BASE, '..','http://a/b/')
+        self.checkJoin(SIMPLE_BASE, '../','http://a/b/')
+        self.checkJoin(SIMPLE_BASE, '../g','http://a/b/g')
+        self.checkJoin(SIMPLE_BASE, '../..','http://a/')
+        self.checkJoin(SIMPLE_BASE, '../../g','http://a/g')
+        self.checkJoin(SIMPLE_BASE, '../../../g','http://a/../g')
+        self.checkJoin(SIMPLE_BASE, './../g','http://a/b/g')
+        self.checkJoin(SIMPLE_BASE, './g/.','http://a/b/c/g/')
+        self.checkJoin(SIMPLE_BASE, '/./g','http://a/./g')
+        self.checkJoin(SIMPLE_BASE, 'g/./h','http://a/b/c/g/h')
+        self.checkJoin(SIMPLE_BASE, 'g/../h','http://a/b/c/h')
+        self.checkJoin(SIMPLE_BASE, 'http:g','http://a/b/c/g')
+        self.checkJoin(SIMPLE_BASE, 'http:','http://a/b/c/d')
+        self.checkJoin(SIMPLE_BASE, 'http:?y','http://a/b/c/d?y')
+        self.checkJoin(SIMPLE_BASE, 'http:g?y','http://a/b/c/g?y')
+        self.checkJoin(SIMPLE_BASE, 'http:g?y/./x','http://a/b/c/g?y/./x')
+
     def test_RFC2732(self):
         for url, hostname, port in [
             ('http://Test.python.org:5432/foo/', 'test.python.org', 5432),

Modified: python/branches/import_unicode/Lib/threading.py
==============================================================================
--- python/branches/import_unicode/Lib/threading.py	(original)
+++ python/branches/import_unicode/Lib/threading.py	Fri Jul 30 01:19:37 2010
@@ -871,91 +871,3 @@
         _active.clear()
         _active.update(new_active)
         assert len(_active) == 1
-
-
-# Self-test code
-
-def _test():
-
-    class BoundedQueue(_Verbose):
-
-        def __init__(self, limit):
-            _Verbose.__init__(self)
-            self.mon = RLock()
-            self.rc = Condition(self.mon)
-            self.wc = Condition(self.mon)
-            self.limit = limit
-            self.queue = deque()
-
-        def put(self, item):
-            self.mon.acquire()
-            while len(self.queue) >= self.limit:
-                self._note("put(%s): queue full", item)
-                self.wc.wait()
-            self.queue.append(item)
-            self._note("put(%s): appended, length now %d",
-                       item, len(self.queue))
-            self.rc.notify()
-            self.mon.release()
-
-        def get(self):
-            self.mon.acquire()
-            while not self.queue:
-                self._note("get(): queue empty")
-                self.rc.wait()
-            item = self.queue.popleft()
-            self._note("get(): got %s, %d left", item, len(self.queue))
-            self.wc.notify()
-            self.mon.release()
-            return item
-
-    class ProducerThread(Thread):
-
-        def __init__(self, queue, quota):
-            Thread.__init__(self, name="Producer")
-            self.queue = queue
-            self.quota = quota
-
-        def run(self):
-            from random import random
-            counter = 0
-            while counter < self.quota:
-                counter = counter + 1
-                self.queue.put("%s.%d" % (self.name, counter))
-                _sleep(random() * 0.00001)
-
-
-    class ConsumerThread(Thread):
-
-        def __init__(self, queue, count):
-            Thread.__init__(self, name="Consumer")
-            self.queue = queue
-            self.count = count
-
-        def run(self):
-            while self.count > 0:
-                item = self.queue.get()
-                print(item)
-                self.count = self.count - 1
-
-    NP = 3
-    QL = 4
-    NI = 5
-
-    Q = BoundedQueue(QL)
-    P = []
-    for i in range(NP):
-        t = ProducerThread(Q, NI)
-        t.name = "Producer-%d" % (i+1)
-        P.append(t)
-    C = ConsumerThread(Q, NI*NP)
-    for t in P:
-        t.start()
-        _sleep(0.000001)
-    C.start()
-    for t in P:
-        t.join()
-    C.join()
-
-if __name__ == '__main__':
-    _test()

Modified: python/branches/import_unicode/Lib/trace.py
==============================================================================
--- python/branches/import_unicode/Lib/trace.py	(original)
+++ python/branches/import_unicode/Lib/trace.py	Fri Jul 30 01:19:37 2010
@@ -257,7 +257,8 @@
         if self.calledfuncs:
             print()
             print("functions called:")
-            for filename, modulename, funcname in sorted(calls.keys()):
+            calls = self.calledfuncs.keys()
+            for filename, modulename, funcname in sorted(calls):
                 print(("filename: %s, modulename: %s, funcname: %s"
                        % (filename, modulename, funcname)))
 
@@ -796,12 +797,9 @@
                   ignoredirs=ignore_dirs, infile=counts_file,
                   outfile=counts_file, timing=timing)
         try:
-            fp = open(progname)
-            try:
-                script = fp.read()
-            finally:
-                fp.close()
-            t.run('exec(%r)' % (script,))
+            with open(progname) as fp:
+                code = compile(fp.read(), progname, 'exec')
+            t.run(code)
         except IOError as err:
             _err_exit("Cannot run file %r because: %s" % (sys.argv[0], err))
         except SystemExit:

Modified: python/branches/import_unicode/Lib/unittest/case.py
==============================================================================
--- python/branches/import_unicode/Lib/unittest/case.py	(original)
+++ python/branches/import_unicode/Lib/unittest/case.py	Fri Jul 30 01:19:37 2010
@@ -940,9 +940,14 @@
                 'Second argument is not a string'))
 
         if first != second:
-            standardMsg = '%s != %s' % (safe_repr(first, True), safe_repr(second, True))
-            diff = '\n' + ''.join(difflib.ndiff(first.splitlines(True),
-                                                       second.splitlines(True)))
+            firstlines = first.splitlines(True)
+            secondlines = second.splitlines(True)
+            if len(firstlines) == 1 and first.strip('\r\n') == first:
+                firstlines = [first + '\n']
+                secondlines = [second + '\n']
+            standardMsg = '%s != %s' % (safe_repr(first, True),
+                                        safe_repr(second, True))
+            diff = '\n' + ''.join(difflib.ndiff(firstlines, secondlines))
             standardMsg = self._truncateMessage(standardMsg, diff)
             self.fail(self._formatMessage(msg, standardMsg))
 

Modified: python/branches/import_unicode/Lib/unittest/test/test_case.py
==============================================================================
--- python/branches/import_unicode/Lib/unittest/test/test_case.py	(original)
+++ python/branches/import_unicode/Lib/unittest/test/test_case.py	Fri Jul 30 01:19:37 2010
@@ -841,6 +841,21 @@
             # so can't use assertEqual either. Just use assertTrue.
             self.assertTrue(sample_text_error == error)
 
+    def testAsertEqualSingleLine(self):
+        sample_text = "laden swallows fly slowly"
+        revised_sample_text = "unladen swallows fly quickly"
+        sample_text_error = """\
+- laden swallows fly slowly
+?                    ^^^^
++ unladen swallows fly quickly
+? ++                   ^^^^^
+"""
+        try:
+            self.assertEqual(sample_text, revised_sample_text)
+        except self.failureException as e:
+            error = str(e).split('\n', 1)[1]
+            self.assertTrue(sample_text_error == error)
+
     def testAssertIsNone(self):
         self.assertIsNone(None)
         self.assertRaises(self.failureException, self.assertIsNone, False)

Modified: python/branches/import_unicode/Lib/urllib/parse.py
==============================================================================
--- python/branches/import_unicode/Lib/urllib/parse.py	(original)
+++ python/branches/import_unicode/Lib/urllib/parse.py	Fri Jul 30 01:19:37 2010
@@ -314,6 +314,8 @@
     # Note: strings are encoded as UTF-8. This is only an issue if it contains
     # unescaped non-ASCII characters, which URIs should not.
     if not string:
+        if string is None:
+            raise TypeError('None object is invalid for unquote_to_bytes()')
         return b''
     if isinstance(string, str):
         string = string.encode('utf-8')
@@ -339,6 +341,8 @@
     unquote('abc%20def') -> 'abc def'.
     """
     if not string:
+        if string is None:
+            raise TypeError('None object is invalid for unquote() function.')
         return string
     res = string.split('%')
     if len(res) == 1:
@@ -799,67 +803,3 @@
     match = _valueprog.match(attr)
     if match: return match.group(1, 2)
     return attr, None
-
-test_input = """
-      http://a/b/c/d
-
-      g:h        = <URL:g:h>
-      http:g     = <URL:http://a/b/c/g>
-      http:      = <URL:http://a/b/c/d>
-      g          = <URL:http://a/b/c/g>
-      ./g        = <URL:http://a/b/c/g>
-      g/         = <URL:http://a/b/c/g/>
-      /g         = <URL:http://a/g>
-      //g        = <URL:http://g>
-      ?y         = <URL:http://a/b/c/d?y>
-      g?y        = <URL:http://a/b/c/g?y>
-      g?y/./x    = <URL:http://a/b/c/g?y/./x>
-      .          = <URL:http://a/b/c/>
-      ./         = <URL:http://a/b/c/>
-      ..         = <URL:http://a/b/>
-      ../        = <URL:http://a/b/>
-      ../g       = <URL:http://a/b/g>
-      ../..      = <URL:http://a/>
-      ../../g    = <URL:http://a/g>
-      ../../../g = <URL:http://a/../g>
-      ./../g     = <URL:http://a/b/g>
-      ./g/.      = <URL:http://a/b/c/g/>
-      /./g       = <URL:http://a/./g>
-      g/./h      = <URL:http://a/b/c/g/h>
-      g/../h     = <URL:http://a/b/c/h>
-      http:g     = <URL:http://a/b/c/g>
-      http:      = <URL:http://a/b/c/d>
-      http:?y         = <URL:http://a/b/c/d?y>
-      http:g?y        = <URL:http://a/b/c/g?y>
-      http:g?y/./x    = <URL:http://a/b/c/g?y/./x>
-"""
-
-def test():
-    base = ''
-    if sys.argv[1:]:
-        fn = sys.argv[1]
-        if fn == '-':
-            fp = sys.stdin
-        else:
-            fp = open(fn)
-    else:
-        from io import StringIO
-        fp = StringIO(test_input)
-    for line in fp:
-        words = line.split()
-        if not words:
-            continue
-        url = words[0]
-        parts = urlparse(url)
-        print('%-10s : %s' % (url, parts))
-        abs = urljoin(base, url)
-        if not base:
-            base = abs
-        wrapped = '<URL:%s>' % abs
-        print('%-10s = %s' % (url, wrapped))
-        if len(words) == 3 and words[1] == '=':
-            if wrapped != words[2]:
-                print('EXPECTED', words[2], '!!!!!!!!!!')
-
-if __name__ == '__main__':
-    test()

Modified: python/branches/import_unicode/Lib/urllib/request.py
==============================================================================
--- python/branches/import_unicode/Lib/urllib/request.py	(original)
+++ python/branches/import_unicode/Lib/urllib/request.py	Fri Jul 30 01:19:37 2010
@@ -1188,7 +1188,8 @@
     # Use local file or FTP depending on form of URL
     def file_open(self, req):
         url = req.selector
-        if url[:2] == '//' and url[2:3] != '/':
+        if url[:2] == '//' and url[2:3] != '/' and (req.host and
+                req.host != 'localhost'):
             req.type = 'ftp'
             return self.parent.open(req)
         else:
@@ -2278,6 +2279,7 @@
                         proxies['http'] = proxyServer
                     else:
                         proxies['http'] = 'http://%s' % proxyServer
+                        proxies['https'] = 'https://%s' % proxyServer
                         proxies['ftp'] = 'ftp://%s' % proxyServer
             internetSettings.Close()
         except (WindowsError, ValueError, TypeError):

Modified: python/branches/import_unicode/Lib/urllib/robotparser.py
==============================================================================
--- python/branches/import_unicode/Lib/urllib/robotparser.py	(original)
+++ python/branches/import_unicode/Lib/urllib/robotparser.py	Fri Jul 30 01:19:37 2010
@@ -66,7 +66,9 @@
     def _add_entry(self, entry):
         if "*" in entry.useragents:
             # the default entry is considered last
-            self.default_entry = entry
+            if self.default_entry is None:
+                # the first default entry wins
+                self.default_entry = entry
         else:
             self.entries.append(entry)
 
@@ -118,7 +120,7 @@
                         entry.rulelines.append(RuleLine(line[1], True))
                         state = 2
         if state == 2:
-            self.entries.append(entry)
+            self._add_entry(entry)
 
 
     def can_fetch(self, useragent, url):
@@ -129,8 +131,10 @@
             return True
         # search for given user agent matches
         # the first match counts
-        url = urllib.parse.quote(
-            urllib.parse.urlparse(urllib.parse.unquote(url))[2])
+        parsed_url = urllib.parse.urlparse(urllib.parse.unquote(url))
+        url = urllib.parse.urlunparse(('','',parsed_url.path,
+            parsed_url.params,parsed_url.query, parsed_url.fragment))
+        url = urllib.parse.quote(url)
         if not url:
             url = "/"
         for entry in self.entries:

Modified: python/branches/import_unicode/Lib/xml/dom/expatbuilder.py
==============================================================================
--- python/branches/import_unicode/Lib/xml/dom/expatbuilder.py	(original)
+++ python/branches/import_unicode/Lib/xml/dom/expatbuilder.py	Fri Jul 30 01:19:37 2010
@@ -242,7 +242,7 @@
         doctype = self.document.implementation.createDocumentType(
             doctypeName, publicId, systemId)
         doctype.ownerDocument = self.document
-        self.document.childNodes.append(doctype)
+        _append_child(self.document, doctype)
         self.document.doctype = doctype
         if self._filter and self._filter.acceptNode(doctype) == FILTER_REJECT:
             self.document.doctype = None

Modified: python/branches/import_unicode/Lib/xml/dom/minidom.py
==============================================================================
--- python/branches/import_unicode/Lib/xml/dom/minidom.py	(original)
+++ python/branches/import_unicode/Lib/xml/dom/minidom.py	Fri Jul 30 01:19:37 2010
@@ -920,6 +920,10 @@
         raise xml.dom.NotFoundErr(
             self.nodeName + " nodes do not have children")
 
+    def normalize(self):
+        # For childless nodes, normalize() has nothing to do.
+        pass
+
     def replaceChild(self, newChild, oldChild):
         raise xml.dom.HierarchyRequestErr(
             self.nodeName + " nodes do not have children")

Modified: python/branches/import_unicode/Lib/xmlrpc/client.py
==============================================================================
--- python/branches/import_unicode/Lib/xmlrpc/client.py	(original)
+++ python/branches/import_unicode/Lib/xmlrpc/client.py	Fri Jul 30 01:19:37 2010
@@ -1135,7 +1135,7 @@
             try:
                 return self.single_request(host, handler, request_body, verbose)
             except socket.error as e:
-                if i or e.errno not in (errno.ECONNRESET, errno.ECONNABORTED):
+                if i or e.errno not in (errno.ECONNRESET, errno.ECONNABORTED, errno.EPIPE):
                     raise
             except http.client.BadStatusLine: #close after we sent request
                 if i:

Modified: python/branches/import_unicode/Makefile.pre.in
==============================================================================
--- python/branches/import_unicode/Makefile.pre.in	(original)
+++ python/branches/import_unicode/Makefile.pre.in	Fri Jul 30 01:19:37 2010
@@ -59,12 +59,18 @@
 # Compiler options
 OPT=		@OPT@
 BASECFLAGS=	@BASECFLAGS@
-CFLAGS=		$(BASECFLAGS) @CFLAGS@ $(OPT) $(EXTRA_CFLAGS)
+CONFIGURE_CFLAGS=	@CFLAGS@
+CONFIGURE_CPPFLAGS=	@CPPFLAGS@
+CONFIGURE_LDFLAGS=	@LDFLAGS@
+# Avoid assigning CFLAGS, LDFLAGS, etc. so users can use them on the
+# command line to append to these values without stomping the pre-set
+# values.
+PY_CFLAGS=	$(BASECFLAGS) $(OPT) $(CONFIGURE_CFLAGS) $(CFLAGS) $(EXTRA_CFLAGS)
 # Both CPPFLAGS and LDFLAGS need to contain the shell's value for setup.py to
 # be able to build extension modules using the directories specified in the
 # environment variables
-CPPFLAGS=	-I. -IInclude -I$(srcdir)/Include @CPPFLAGS@
-LDFLAGS=	@LDFLAGS@
+PY_CPPFLAGS=	-I. -IInclude -I$(srcdir)/Include $(CONFIGURE_CPPFLAGS) $(CPPFLAGS)
+PY_LDFLAGS=	$(CONFIGURE_LDFLAGS) $(LDFLAGS)
 LDLAST=		@LDLAST@
 SGI_ABI=	@SGI_ABI@
 CCSHARED=	@CCSHARED@
@@ -73,7 +79,7 @@
 # Extra C flags added for building the interpreter object files.
 CFLAGSFORSHARED=@CFLAGSFORSHARED@
 # C flags used for building the interpreter object files
-PY_CFLAGS=	$(CFLAGS) $(CPPFLAGS) $(CFLAGSFORSHARED) -DPy_BUILD_CORE
+PY_CORE_CFLAGS=	$(PY_CFLAGS) $(PY_CPPFLAGS) $(CFLAGSFORSHARED) -DPy_BUILD_CORE
 
 
 # Machine-dependent subdirectories
@@ -411,7 +417,7 @@
 
 # Build the interpreter
 $(BUILDPYTHON):	Modules/python.o $(LIBRARY) $(LDLIBRARY)
-		$(LINKCC) $(LDFLAGS) $(LINKFORSHARED) -o $@ \
+		$(LINKCC) $(PY_LDFLAGS) $(LINKFORSHARED) -o $@ \
 			Modules/python.o \
 			$(BLDLIBRARY) $(LIBS) $(MODLIBS) $(SYSLIBS) $(LDLAST)
 
@@ -422,8 +428,8 @@
 # Build the shared modules
 sharedmods: $(BUILDPYTHON)
 	@case $$MAKEFLAGS in \
-	*s*) $(RUNSHARED) CC='$(CC)' LDSHARED='$(BLDSHARED)' LDFLAGS='$(LDFLAGS)' OPT='$(OPT)' ./$(BUILDPYTHON) -E $(srcdir)/setup.py -q build;; \
-	*) $(RUNSHARED) CC='$(CC)' LDSHARED='$(BLDSHARED)' LDFLAGS='$(LDFLAGS)' OPT='$(OPT)' ./$(BUILDPYTHON) -E $(srcdir)/setup.py build;; \
+	*s*) $(RUNSHARED) CC='$(CC)' LDSHARED='$(BLDSHARED)' LDFLAGS='$(PY_LDFLAGS)' OPT='$(OPT)' ./$(BUILDPYTHON) -E $(srcdir)/setup.py -q build;; \
+	*) $(RUNSHARED) CC='$(CC)' LDSHARED='$(BLDSHARED)' LDFLAGS='$(PY_LDFLAGS)' OPT='$(OPT)' ./$(BUILDPYTHON) -E $(srcdir)/setup.py build;; \
 	esac
 
 # Build static library
@@ -440,18 +446,18 @@
 
 libpython$(VERSION).so: $(LIBRARY_OBJS)
 	if test $(INSTSONAME) != $(LDLIBRARY); then \
-		$(LDSHARED) $(LDFLAGS) -Wl,-h$(INSTSONAME) -o $(INSTSONAME) $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
+		$(LDSHARED) $(PY_LDFLAGS) -Wl,-h$(INSTSONAME) -o $(INSTSONAME) $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
 		$(LN) -f $(INSTSONAME) $@; \
 	else \
-		$(LDSHARED) $(LDFLAGS) -o $@ $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
+		$(LDSHARED) $(PY_LDFLAGS) -o $@ $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
 	fi
 
 libpython$(VERSION).dylib: $(LIBRARY_OBJS)
-	 $(CC) -dynamiclib -Wl,-single_module $(LDFLAGS) -undefined dynamic_lookup -Wl,-install_name,$(prefix)/lib/libpython$(VERSION).dylib -Wl,-compatibility_version,$(VERSION) -Wl,-current_version,$(VERSION) -o $@ $(LIBRARY_OBJS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
+	 $(CC) -dynamiclib -Wl,-single_module $(PY_LDFLAGS) -undefined dynamic_lookup -Wl,-install_name,$(prefix)/lib/libpython$(VERSION).dylib -Wl,-compatibility_version,$(VERSION) -Wl,-current_version,$(VERSION) -o $@ $(LIBRARY_OBJS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST); \
 		 
 
 libpython$(VERSION).sl: $(LIBRARY_OBJS)
-	$(LDSHARED) $(LDFLAGS) -o $@ $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST)
+	$(LDSHARED) $(PY_LDFLAGS) -o $@ $(LIBRARY_OBJS) $(MODLIBS) $(SHLIBS) $(LIBC) $(LIBM) $(LDLAST)
 
 # Copy up the gdb python hooks into a position where they can be automatically
 # loaded by gdb during Lib/test/test_gdb.py
@@ -497,7 +503,7 @@
 # for a shared core library; otherwise, this rule is a noop.
 $(DLLLIBRARY) libpython$(VERSION).dll.a: $(LIBRARY_OBJS)
 	if test -n "$(DLLLIBRARY)"; then \
-		$(LDSHARED) $(LDFLAGS) -Wl,--out-implib=$@ -o $(DLLLIBRARY) $^ \
+		$(LDSHARED) $(PY_LDFLAGS) -Wl,--out-implib=$@ -o $(DLLLIBRARY) $^ \
 			$(LIBS) $(MODLIBS) $(SYSLIBS) $(LDLAST); \
 	else true; \
 	fi
@@ -541,10 +547,10 @@
 		$(SIGNAL_OBJS) \
 		$(MODOBJS) \
 		$(srcdir)/Modules/getbuildinfo.c
-	$(CC) -c $(PY_CFLAGS) -DSVNVERSION="\"`LC_ALL=C $(SVNVERSION)`\"" -o $@ $(srcdir)/Modules/getbuildinfo.c
+	$(CC) -c $(PY_CORE_CFLAGS) -DSVNVERSION="\"`LC_ALL=C $(SVNVERSION)`\"" -o $@ $(srcdir)/Modules/getbuildinfo.c
 
 Modules/getpath.o: $(srcdir)/Modules/getpath.c Makefile
-	$(CC) -c $(PY_CFLAGS) -DPYTHONPATH='"$(PYTHONPATH)"' \
+	$(CC) -c $(PY_CORE_CFLAGS) -DPYTHONPATH='"$(PYTHONPATH)"' \
 		-DPREFIX='"$(prefix)"' \
 		-DEXEC_PREFIX='"$(exec_prefix)"' \
 		-DVERSION='"$(VERSION)"' \
@@ -552,7 +558,7 @@
 		-o $@ $(srcdir)/Modules/getpath.c
 
 Modules/python.o: $(srcdir)/Modules/python.c
-	$(MAINCC) -c $(PY_CFLAGS) -o $@ $(srcdir)/Modules/python.c
+	$(MAINCC) -c $(PY_CORE_CFLAGS) -o $@ $(srcdir)/Modules/python.c
 
 $(IO_OBJS): $(IO_H)
 
@@ -561,7 +567,7 @@
 		-$(PGEN) $(GRAMMAR_INPUT) $(GRAMMAR_H) $(GRAMMAR_C)
 
 $(PGEN):	$(PGENOBJS)
-		$(CC) $(OPT) $(LDFLAGS) $(PGENOBJS) $(LIBS) -o $(PGEN)
+		$(CC) $(OPT) $(PY_LDFLAGS) $(PGENOBJS) $(LIBS) -o $(PGEN)
 
 Parser/grammar.o:	$(srcdir)/Parser/grammar.c \
 				$(srcdir)/Include/token.h \
@@ -581,10 +587,10 @@
 Python/compile.o Python/symtable.o Python/ast.o: $(GRAMMAR_H) $(AST_H)
 
 Python/getplatform.o: $(srcdir)/Python/getplatform.c
-		$(CC) -c $(PY_CFLAGS) -DPLATFORM='"$(MACHDEP)"' -o $@ $(srcdir)/Python/getplatform.c
+		$(CC) -c $(PY_CORE_CFLAGS) -DPLATFORM='"$(MACHDEP)"' -o $@ $(srcdir)/Python/getplatform.c
 
 Python/importdl.o: $(srcdir)/Python/importdl.c
-		$(CC) -c $(PY_CFLAGS) -I$(DLINCLDIR) -o $@ $(srcdir)/Python/importdl.c
+		$(CC) -c $(PY_CORE_CFLAGS) -I$(DLINCLDIR) -o $@ $(srcdir)/Python/importdl.c
 
 Objects/unicodectype.o:	$(srcdir)/Objects/unicodectype.c \
 				$(srcdir)/Objects/unicodetype_db.h
@@ -1039,7 +1045,7 @@
 
 # Install the dynamically loadable modules
 # This goes into $(exec_prefix)
-sharedinstall:
+sharedinstall: sharedmods
 	$(RUNSHARED) ./$(BUILDPYTHON) -E $(srcdir)/setup.py install \
 	   	--prefix=$(prefix) \
 		--install-scripts=$(BINDIR) \
@@ -1130,7 +1136,7 @@
 
 # Some make's put the object file in the current directory
 .c.o:
-	$(CC) -c $(PY_CFLAGS) -o $@ $<
+	$(CC) -c $(PY_CORE_CFLAGS) -o $@ $<
 
 # Run reindent on the library
 reindent:

Modified: python/branches/import_unicode/Misc/ACKS
==============================================================================
--- python/branches/import_unicode/Misc/ACKS	(original)
+++ python/branches/import_unicode/Misc/ACKS	Fri Jul 30 01:19:37 2010
@@ -63,6 +63,7 @@
 Ben Bell
 Thomas Bellman
 Alexander Belopolsky
+Eli Bendersky
 Andrew Bennetts
 Andy Bensky
 Michel Van den Bergh
@@ -97,6 +98,7 @@
 Georg Brandl
 Christopher Brannon
 Terrence Brannon
+Brian Brazil
 Dave Brennan
 Tom Bridgman
 Richard Brodie
@@ -128,6 +130,7 @@
 Per Cederqvist
 Octavian Cerna
 Pascal Chambon
+John Chandler
 Hye-Shik Chang
 Jeffrey Chang
 Mitch Chapman
@@ -145,6 +148,7 @@
 Vadim Chugunov
 David Cinege
 Mike Clarkson
+Andrew Clegg
 Brad Clements
 Steve Clift
 Nick Coghlan
@@ -154,11 +158,11 @@
 Jeffery Collins
 Robert Collins
 Paul Colomiets
-Jason R. Coombs
 Geremy Condra
 Juan Jos? Conti
 Matt Conway
 David M. Cooke
+Jason R. Coombs
 Greg Copeland
 Aldo Cortesi
 David Costanzo
@@ -444,6 +448,7 @@
 Vladimir Kushnir
 Cameron Laird
 Torsten Landschoff
+?ukasz Langa
 Tino Lange
 Andrew Langmead
 Detlef Lannert
@@ -453,6 +458,7 @@
 Simon Law
 Chris Lawrence
 Brian Leair
+James Lee
 John J. Lee
 Inyeol Lee
 Thomas Lee
@@ -473,6 +479,7 @@
 Ross Light
 Shawn Ligocki
 Martin Ligr
+Grant Limberg
 Christopher Lindblad
 Bjorn Lindqvist
 Per Lindqvist

Modified: python/branches/import_unicode/Misc/NEWS
==============================================================================
--- python/branches/import_unicode/Misc/NEWS	(original)
+++ python/branches/import_unicode/Misc/NEWS	Fri Jul 30 01:19:37 2010
@@ -12,6 +12,11 @@
 Core and Builtins
 -----------------
 
+- Issue #8991: convertbuffer() rejects discontigious buffers.
+
+- Issue #7616: Fix copying of overlapping memoryview slices with the Intel
+  compiler.
+
 - Issue #8413: structsequence now subclasses tuple.
 
 - Issue #8271: during the decoding of an invalid UTF-8 byte sequence, only the
@@ -470,6 +475,87 @@
 Library
 -------
 
+- Issue #4108: In urllib.robotparser, if there are multiple 'User-agent: *'
+  entries, consider the first one.
+
+- Issue #6630: Allow customizing regex flags when subclassing the
+  string.Template class.
+
+- Issue #9411: Allow specifying an encoding for config files in the
+  configparser module.
+
+- Issue #1682942: Improvements to configparser: support alternate
+  delimiters, alternate comment prefixes and empty lines in values.
+
+- Issue #9354: Provide getsockopt() in asyncore's file_wrapper.
+
+- Issue #8966: ctypes: Remove implicit bytes-unicode conversion.
+
+- Issue #9378: python -m pickle <pickle file> will now load and
+  display the first object in the pickle file.
+
+- Issue #4770: Restrict binascii module to accept only bytes (as specified).
+  And fix the email package to encode to ASCII instead of
+  ``raw-unicode-escape`` before ASCII-to-binary decoding.
+
+- Issue #9384: python -m tkinter will now display a simple demo applet.
+
+- The default size of the re module's compiled regular expression cache has
+  been increased from 100 to 500 and the cache replacement policy has changed
+  from simply clearing the entire cache on overflow to randomly forgetting 20%
+  of the existing cached compiled regular expressions.  This is a performance
+  win for applications that use a lot of regular expressions and limits the
+  impact of the performance hit anytime the cache is exceeded.
+
+- Issue #7113: Speed up loading in configparser. Patch by ?ukasz Langa.
+
+- Issue #9032: XML-RPC client retries the request on EPIPE error. The EPIPE
+  error occurs when the server closes the socket and the client sends a big
+  XML-RPC request.
+
+- Issue #4629: getopt raises an error if an argument ends with = whereas getopt
+  doesn't except a value (eg. --help= is rejected if getopt uses ['help='] long
+  options).
+
+- Issue #7989: Added pure python implementation of the `datetime`
+  module.  The C module is renamed to `_datetime` and if available,
+  overrides all classes defined in datetime with fast C impementation.
+  Python implementation is based on the original python prototype for
+  the datetime module by Tim Peters with minor modifications by the
+  PyPy project.  The test suite now tests `datetime` module with and
+  without `_datetime` acceleration using the same test cases.
+
+- Issue #7895: platform.mac_ver() no longer crashes after calling os.fork()
+
+- Issue #9323: Fixed a bug in trace.py that resulted in loosing the
+  name of the script being traced.  Patch by Eli Bendersky.
+
+- Issue #9282: Fixed --listfuncs option of trace.py.  Thanks Eli
+  Bendersky for the patch.
+
+- Issue #3704: http.cookiejar was not properly handling URLs with a / in the
+  parameters.
+
+- Issue #9268: ``pickletools.dis()`` now has an optional *annotate*
+  argument which controls printing of opcode descriptions in ``dis()``
+  output.
+
+- Issue #1555570: email no longer inserts extra blank lines when a \r\n
+  combo crosses an 8192 byte boundary.
+
+- Issue #9243: Fix sndhdr module and add unit tests, contributed by James Lee.
+
+- ``ast.literal_eval()`` now allows byte literals.
+
+- Issue #9137: Fix issue in MutableMapping.update, which incorrectly
+  treated keyword arguments called 'self' or 'other' specially.
+
+- ``ast.literal_eval()`` now allows set literals.
+
+- Issue #9164: Ensure that sysconfig handles duplicate -arch flags in CFLAGS.
+
+- Issue #7646: The fnmatch pattern cache no longer grows without bound.
+
 - Issue #9136: Fix 'dictionary changed size during iteration'
   RuntimeError produced when profiling the decimal module.  This was
   due to a dangerous iteration over 'locals()' in Context.__init__.
@@ -1406,6 +1492,28 @@
 Extension Modules
 -----------------
 
+- Issue #9422: Fix memory leak when re-initializing a struct.Struct object.
+
+- Issue #7900: The getgroups(2) system call on MacOSX behaves rather oddly
+  compared to other unix systems. In particular, os.getgroups() does
+  not reflect any changes made using os.setgroups() but basicly always
+  returns the same information as the id command.
+
+  os.getgroups() can now return more than 16 groups on MacOSX.
+
+- Issue #6095: Make directory argument to os.listdir optional.
+
+- Issue #9277: Fix bug in struct.pack for bools in standard mode
+  (e.g., struct.pack('>?')):  if conversion to bool raised an exception
+  then that exception wasn't properly propagated on machines where
+  char is unsigned.
+
+- Issue #5180: Fixed a bug that prevented loading 2.x pickles in 3.x
+  python when they contain instances of old-style classes.
+
+- Issue #9165: Add new functions math.isfinite and cmath.isfinite, to
+  accompany existing isinf and isnan functions.
+
 - Issue #1578269: Implement os.symlink for Windows 6.0+. Patch by
   Jason R. Coombs
 
@@ -1528,6 +1636,14 @@
 Build
 -----
 
+- Issue #9280: Make sharedinstall depend on sharedmods.
+
+- Issue #9189: Make a user-specified CFLAGS, CPPFLAGS, or LDFLAGS
+  setting override the configure and makefile defaults, without
+  deleting options the user didn't intend to override.  Developers
+  should no longer need to specify OPT or EXTRA_CFLAGS, although those
+  variables are still present for backward-compatibility.
+
 - Issue #8854: Fix finding Visual Studio 2008 on Windows x64.
 
 - Issue #1759169, #8864: Drop _XOPEN_SOURCE on Solaris, define it for
@@ -1604,6 +1720,9 @@
 Documentation
 ------------
 
+- Issue 9255: Document that the 'test' package is meant for internal Python use
+  only.
+
 - A small wsgi server was added as Tools/scripts/serve.py, and is used to
   implement a local documentation server via 'make serve' in the doc directory.
 
@@ -1625,6 +1744,9 @@
 Tests
 -----
 
+- Issue #9251: test_threaded_import didn't fail when run through regrtest
+  if the import lock was disabled.
+
 - Issue #8605: Skip test_gdb if Python is compiled with optimizations.
 
 - Issue #7449: Skip test_socketserver if threading support is disabled
@@ -1755,7 +1877,7 @@
 Tools/Demos
 -----------
 
-- Issue #5464: Implement plural forms in msgfmt.py.
+- Issue #5464, #8974: Implement plural forms in msgfmt.py.
 
 - iobench (a file I/O benchmark) and ccbench (a concurrency benchmark) were
   added to the `Tools/` directory.  They were previously living in the

Modified: python/branches/import_unicode/Misc/developers.txt
==============================================================================
--- python/branches/import_unicode/Misc/developers.txt	(original)
+++ python/branches/import_unicode/Misc/developers.txt	Fri Jul 30 01:19:37 2010
@@ -20,6 +20,13 @@
 Permissions History
 -------------------
 
+- Brian Quinlan was given commit access on Jul 26 2010 by GFB,
+  for work related to PEP 3148.
+
+- Reid Kleckner was given commit access on Jul 11 2010 by GFB,
+  for work on the py3k-jit branch, at suggestion of the Unladen
+  Swallow team.
+
 - Alexander Belopolsky was given commit access on May 25 2010
   by MvL at suggestion of Mark Dickinson.
 

Modified: python/branches/import_unicode/Misc/maintainers.rst
==============================================================================
--- python/branches/import_unicode/Misc/maintainers.rst	(original)
+++ python/branches/import_unicode/Misc/maintainers.rst	Fri Jul 30 01:19:37 2010
@@ -89,7 +89,7 @@
 doctest             tim_one (inactive)
 dummy_threading     brett.cannon
 email               barry, r.david.murray
-encodings           lemburg, loewis
+encodings           lemburg
 errno
 exceptions
 fcntl
@@ -104,7 +104,7 @@
 gc                  pitrou
 getopt
 getpass
-gettext             loewis
+gettext
 glob
 grp
 gzip
@@ -125,7 +125,7 @@
 keyword
 lib2to3             benjamin.peterson
 linecache
-locale              loewis, lemburg
+locale              lemburg
 logging             vsajip
 macpath
 mailbox             andrew.kuchling
@@ -135,7 +135,7 @@
 mimetypes
 mmap
 modulefinder        theller, jvr
-msilib              loewis
+msilib
 msvcrt
 multiprocessing     jnoller
 netrc
@@ -144,7 +144,7 @@
 numbers
 operator
 optparse            aronacher
-os                  loewis
+os
 ossaudiodev
 parser
 pdb
@@ -212,12 +212,12 @@
 tkinter             gpolo
 token               georg.brandl
 tokenize
-trace
+trace               alexander.belopolsky
 traceback           georg.brandl
 tty
 turtle              gregorlingl
 types
-unicodedata         loewis, lemburg, ezio.melotti
+unicodedata         lemburg, ezio.melotti
 unittest            michael.foord
 urllib              orsenthil
 uu
@@ -226,13 +226,13 @@
 wave
 weakref             fdrake
 webbrowser          georg.brandl
-winreg
+winreg              brian.curtin
 winsound            effbot (inactive)
 wsgiref             pje
 xdrlib
-xml                 loewis
+xml
 xml.etree           effbot (inactive)
-xmlrpc              loewis
+xmlrpc
 zipfile
 zipimport
 zlib
@@ -277,13 +277,13 @@
 i18n                lemburg
 import machinery    brett.cannon, ncoghlan
 io                  pitrou, benjamin.peterson
-locale              lemburg, loewis
+locale              lemburg
 mathematics         mark.dickinson, eric.smith, lemburg
 memory management   tim_one, lemburg
 networking          giampaolo.rodola
 packaging           tarek, lemburg
 py3 transition      benjamin.peterson
-release management  tarek, lemburg, benjamin.peterson, barry, loewis,
+release management  tarek, lemburg, benjamin.peterson, barry,
                     gvanrossum, anthonybaxter
 str.format          eric.smith
 time and dates      lemburg

Modified: python/branches/import_unicode/Misc/python-wing.wpr
==============================================================================
--- python/branches/import_unicode/Misc/python-wing.wpr	(original)
+++ python/branches/import_unicode/Misc/python-wing.wpr	Fri Jul 30 01:19:37 2010
@@ -5,7 +5,7 @@
 ##################################################################
 [project attributes]
 proj.directory-list = [{'dirloc': loc('..'),
-                        'excludes': (),
+                        'excludes': [u'Lib/__pycache__'],
                         'filter': '*',
                         'include_hidden': False,
                         'recursive': True,

Modified: python/branches/import_unicode/Modules/Setup.dist
==============================================================================
--- python/branches/import_unicode/Modules/Setup.dist	(original)
+++ python/branches/import_unicode/Modules/Setup.dist	Fri Jul 30 01:19:37 2010
@@ -170,7 +170,7 @@
 #atexit atexitmodule.c      # Register functions to be run at interpreter-shutdown
 #_elementtree -I$(srcdir)/Modules/expat -DHAVE_EXPAT_CONFIG_H -DUSE_PYEXPAT_CAPI _elementtree.c	# elementtree accelerator
 #_pickle _pickle.c	# pickle accelerator
-#datetime datetimemodule.c	# date/time type
+#_datetime _datetimemodule.c	# datetime accelerator
 #_bisect _bisectmodule.c	# Bisection algorithms
 #_heapq _heapqmodule.c	# Heap queue algorithm
 

Modified: python/branches/import_unicode/Modules/_ctypes/_ctypes.c
==============================================================================
--- python/branches/import_unicode/Modules/_ctypes/_ctypes.c	(original)
+++ python/branches/import_unicode/Modules/_ctypes/_ctypes.c	Fri Jul 30 01:19:37 2010
@@ -132,8 +132,6 @@
 /* a callable object used for unpickling */
 static PyObject *_unpickle;
 
-char *_ctypes_conversion_encoding = NULL;
-char *_ctypes_conversion_errors = NULL;
 
 
 /****************************************************************/
@@ -1090,13 +1088,7 @@
         return -1;
     }
 
-    if (PyUnicode_Check(value)) {
-        value = PyUnicode_AsEncodedString(value,
-                                          _ctypes_conversion_encoding,
-                                          _ctypes_conversion_errors);
-        if (!value)
-            return -1;
-    } else if (!PyBytes_Check(value)) {
+    if (!PyBytes_Check(value)) {
         PyErr_Format(PyExc_TypeError,
                      "str/bytes expected instead of %s instance",
                      Py_TYPE(value)->tp_name);
@@ -1150,13 +1142,7 @@
                         "can't delete attribute");
         return -1;
     }
-    if (PyBytes_Check(value)) {
-        value = PyUnicode_FromEncodedObject(value,
-                                            _ctypes_conversion_encoding,
-                                            _ctypes_conversion_errors);
-        if (!value)
-            return -1;
-    } else if (!PyUnicode_Check(value)) {
+    if (!PyUnicode_Check(value)) {
         PyErr_Format(PyExc_TypeError,
                         "unicode string expected instead of %s instance",
                         Py_TYPE(value)->tp_name);
@@ -1510,7 +1496,7 @@
         Py_INCREF(Py_None);
         return Py_None;
     }
-    if (PyBytes_Check(value) || PyUnicode_Check(value)) {
+    if (PyBytes_Check(value)) {
         PyCArgObject *parg;
         struct fielddesc *fd = _ctypes_get_fielddesc("z");
 

Modified: python/branches/import_unicode/Modules/_ctypes/callproc.c
==============================================================================
--- python/branches/import_unicode/Modules/_ctypes/callproc.c	(original)
+++ python/branches/import_unicode/Modules/_ctypes/callproc.c	Fri Jul 30 01:19:37 2010
@@ -1641,37 +1641,6 @@
     return arg;
 }
 
-#ifdef CTYPES_UNICODE
-
-static char set_conversion_mode_doc[] =
-"set_conversion_mode(encoding, errors) -> (previous-encoding, previous-errors)\n\
-\n\
-Set the encoding and error handling ctypes uses when converting\n\
-between unicode and strings.  Returns the previous values.\n";
-
-static PyObject *
-set_conversion_mode(PyObject *self, PyObject *args)
-{
-    char *coding, *mode;
-    PyObject *result;
-
-    if (!PyArg_ParseTuple(args, "zs:set_conversion_mode", &coding, &mode))
-        return NULL;
-    result = Py_BuildValue("(zz)", _ctypes_conversion_encoding, _ctypes_conversion_errors);
-    if (coding) {
-        PyMem_Free(_ctypes_conversion_encoding);
-        _ctypes_conversion_encoding = PyMem_Malloc(strlen(coding) + 1);
-        strcpy(_ctypes_conversion_encoding, coding);
-    } else {
-        _ctypes_conversion_encoding = NULL;
-    }
-    PyMem_Free(_ctypes_conversion_errors);
-    _ctypes_conversion_errors = PyMem_Malloc(strlen(mode) + 1);
-    strcpy(_ctypes_conversion_errors, mode);
-    return result;
-}
-#endif
-
 static PyObject *
 resize(PyObject *self, PyObject *args)
 {
@@ -1852,9 +1821,6 @@
     {"_unpickle", unpickle, METH_VARARGS },
     {"buffer_info", buffer_info, METH_O, "Return buffer interface information"},
     {"resize", resize, METH_VARARGS, "Resize the memory buffer of a ctypes instance"},
-#ifdef CTYPES_UNICODE
-    {"set_conversion_mode", set_conversion_mode, METH_VARARGS, set_conversion_mode_doc},
-#endif
 #ifdef MS_WIN32
     {"get_last_error", get_last_error, METH_NOARGS},
     {"set_last_error", set_last_error, METH_VARARGS},

Modified: python/branches/import_unicode/Modules/_ctypes/cfield.c
==============================================================================
--- python/branches/import_unicode/Modules/_ctypes/cfield.c	(original)
+++ python/branches/import_unicode/Modules/_ctypes/cfield.c	Fri Jul 30 01:19:37 2010
@@ -1168,20 +1168,6 @@
 static PyObject *
 c_set(void *ptr, PyObject *value, Py_ssize_t size)
 {
-    if (PyUnicode_Check(value)) {
-        value = PyUnicode_AsEncodedString(value,
-                                          _ctypes_conversion_encoding,
-                                          _ctypes_conversion_errors);
-        if (value == NULL)
-            return NULL;
-        if (PyBytes_GET_SIZE(value) != 1) {
-            Py_DECREF(value);
-            goto error;
-        }
-        *(char *)ptr = PyBytes_AS_STRING(value)[0];
-        Py_DECREF(value);
-        _RET(value);
-    }
     if (PyBytes_Check(value) && PyBytes_GET_SIZE(value) == 1) {
         *(char *)ptr = PyBytes_AS_STRING(value)[0];
         _RET(value);
@@ -1217,13 +1203,7 @@
 u_set(void *ptr, PyObject *value, Py_ssize_t size)
 {
     Py_ssize_t len;
-    if (PyBytes_Check(value)) {
-        value = PyUnicode_FromEncodedObject(value,
-                                            _ctypes_conversion_encoding,
-                                            _ctypes_conversion_errors);
-        if (!value)
-            return NULL;
-    } else if (!PyUnicode_Check(value)) {
+    if (!PyUnicode_Check(value)) {
         PyErr_Format(PyExc_TypeError,
                         "unicode string expected instead of %s instance",
                         value->ob_type->tp_name);
@@ -1292,13 +1272,7 @@
     /* It's easier to calculate in characters than in bytes */
     length /= sizeof(wchar_t);
 
-    if (PyBytes_Check(value)) {
-        value = PyUnicode_FromEncodedObject(value,
-                                            _ctypes_conversion_encoding,
-                                            _ctypes_conversion_errors);
-        if (!value)
-            return NULL;
-    } else if (!PyUnicode_Check(value)) {
+    if (!PyUnicode_Check(value)) {
         PyErr_Format(PyExc_TypeError,
                         "unicode string expected instead of %s instance",
                         value->ob_type->tp_name);
@@ -1342,14 +1316,7 @@
     char *data;
     Py_ssize_t size;
 
-    if (PyUnicode_Check(value)) {
-        value = PyUnicode_AsEncodedString(value,
-                                                                          _ctypes_conversion_encoding,
-                                                                          _ctypes_conversion_errors);
-        if (value == NULL)
-            return NULL;
-        assert(PyBytes_Check(value));
-    } else if(PyBytes_Check(value)) {
+    if(PyBytes_Check(value)) {
         Py_INCREF(value);
     } else {
         PyErr_Format(PyExc_TypeError,
@@ -1393,14 +1360,6 @@
         *(char **)ptr = PyBytes_AsString(value);
         Py_INCREF(value);
         return value;
-    } else if (PyUnicode_Check(value)) {
-        PyObject *str = PyUnicode_AsEncodedString(value,
-                                                  _ctypes_conversion_encoding,
-                                                  _ctypes_conversion_errors);
-        if (str == NULL)
-            return NULL;
-        *(char **)ptr = PyBytes_AS_STRING(str);
-        return str;
     } else if (PyLong_Check(value)) {
 #if SIZEOF_VOID_P == SIZEOF_LONG_LONG
         *(char **)ptr = (char *)PyLong_AsUnsignedLongLongMask(value);
@@ -1454,13 +1413,7 @@
         Py_INCREF(Py_None);
         return Py_None;
     }
-    if (PyBytes_Check(value)) {
-        value = PyUnicode_FromEncodedObject(value,
-                                            _ctypes_conversion_encoding,
-                                            _ctypes_conversion_errors);
-        if (!value)
-            return NULL;
-    } else if (!PyUnicode_Check(value)) {
+    if (!PyUnicode_Check(value)) {
         PyErr_Format(PyExc_TypeError,
                      "unicode string or integer address expected instead of %s instance",
                      value->ob_type->tp_name);
@@ -1540,12 +1493,6 @@
     /* convert value into a PyUnicodeObject or NULL */
     if (Py_None == value) {
         value = NULL;
-    } else if (PyBytes_Check(value)) {
-        value = PyUnicode_FromEncodedObject(value,
-                                            _ctypes_conversion_encoding,
-                                            _ctypes_conversion_errors);
-        if (!value)
-            return NULL;
     } else if (PyUnicode_Check(value)) {
         Py_INCREF(value); /* for the descref below */
     } else {

Modified: python/branches/import_unicode/Modules/_io/_iomodule.c
==============================================================================
--- python/branches/import_unicode/Modules/_io/_iomodule.c	(original)
+++ python/branches/import_unicode/Modules/_io/_iomodule.c	Fri Jul 30 01:19:37 2010
@@ -176,7 +176,7 @@
  * The main open() function
  */
 PyDoc_STRVAR(open_doc,
-"open(file, mode='r', buffering=None, encoding=None,\n"
+"open(file, mode='r', buffering=-1, encoding=None,\n"
 "     errors=None, newline=None, closefd=True) -> file object\n"
 "\n"
 "Open file and return a stream.  Raise IOError upon failure.\n"

Modified: python/branches/import_unicode/Modules/_json.c
==============================================================================
--- python/branches/import_unicode/Modules/_json.c	(original)
+++ python/branches/import_unicode/Modules/_json.c	Fri Jul 30 01:19:37 2010
@@ -1454,7 +1454,7 @@
         }
         else {
             /* TODO: include repr of key */
-            PyErr_SetString(PyExc_ValueError, "keys must be a string");
+            PyErr_SetString(PyExc_TypeError, "keys must be a string");
             goto bail;
         }
 

Modified: python/branches/import_unicode/Modules/_pickle.c
==============================================================================
--- python/branches/import_unicode/Modules/_pickle.c	(original)
+++ python/branches/import_unicode/Modules/_pickle.c	Fri Jul 30 01:19:37 2010
@@ -3466,29 +3466,19 @@
 static PyObject *
 instantiate(PyObject *cls, PyObject *args)
 {
-    PyObject *r = NULL;
-
-    /* XXX: The pickle.py module does not create instances this way when the
-       args tuple is empty. See Unpickler._instantiate(). */
-    if ((r = PyObject_CallObject(cls, args)))
-        return r;
-
-    /* XXX: Is this still nescessary? */
-    {
-        PyObject *tp, *v, *tb, *tmp_value;
-
-        PyErr_Fetch(&tp, &v, &tb);
-        tmp_value = v;
-        /* NULL occurs when there was a KeyboardInterrupt */
-        if (tmp_value == NULL)
-            tmp_value = Py_None;
-        if ((r = PyTuple_Pack(3, tmp_value, cls, args))) {
-            Py_XDECREF(v);
-            v = r;
-        }
-        PyErr_Restore(tp, v, tb);
+    PyObject *result = NULL;
+    /* Caller must assure args are a tuple.  Normally, args come from
+       Pdata_poptuple which packs objects from the top of the stack
+       into a newly created tuple. */
+    assert(PyTuple_Check(args));
+    if (Py_SIZE(args) > 0 || !PyType_Check(cls) ||
+        PyObject_HasAttrString(cls, "__getinitargs__")) {
+        result = PyObject_CallObject(cls, args);
+    }
+    else {
+        result = PyObject_CallMethod(cls, "__new__", "O", cls);
     }
-    return NULL;
+    return result;
 }
 
 static int
@@ -3547,7 +3537,7 @@
         if (len < 2)
             return bad_readline();
         class_name = PyUnicode_DecodeASCII(s, len - 1, "strict");
-        if (class_name == NULL) {
+        if (class_name != NULL) {
             cls = find_class(self, module_name, class_name);
             Py_DECREF(class_name);
         }
@@ -4276,7 +4266,7 @@
         return -1;
     PDATA_POP(self->stack, callable);
     if (callable) {
-        obj = instantiate(callable, argtup);
+        obj = PyObject_CallObject(callable, argtup);
         Py_DECREF(callable);
     }
     Py_DECREF(argtup);

Modified: python/branches/import_unicode/Modules/_struct.c
==============================================================================
--- python/branches/import_unicode/Modules/_struct.c	(original)
+++ python/branches/import_unicode/Modules/_struct.c	Fri Jul 30 01:19:37 2010
@@ -867,11 +867,11 @@
 static int
 bp_bool(char *p, PyObject *v, const formatdef *f)
 {
-    char y;
+    int y;
     y = PyObject_IsTrue(v);
     if (y < 0)
         return -1;
-    memcpy(p, (char *)&y, sizeof y);
+    *p = (char)y;
     return 0;
 }
 
@@ -1247,6 +1247,9 @@
         PyErr_NoMemory();
         return -1;
     }
+    /* Free any s_codes value left over from a previous initialization. */
+    if (self->s_codes != NULL)
+        PyMem_FREE(self->s_codes);
     self->s_codes = codes;
 
     s = fmt;

Modified: python/branches/import_unicode/Modules/binascii.c
==============================================================================
--- python/branches/import_unicode/Modules/binascii.c	(original)
+++ python/branches/import_unicode/Modules/binascii.c	Fri Jul 30 01:19:37 2010
@@ -546,7 +546,7 @@
     Py_ssize_t len;
     int done = 0;
 
-    if ( !PyArg_ParseTuple(args, "s*:a2b_hqx", &pascii) )
+    if ( !PyArg_ParseTuple(args, "y*:a2b_hqx", &pascii) )
         return NULL;
     ascii_data = pascii.buf;
     len = pascii.len;
@@ -750,7 +750,7 @@
     PyObject *rv;
     Py_ssize_t in_len, out_len, out_len_left;
 
-    if ( !PyArg_ParseTuple(args, "s*:rledecode_hqx", &pin) )
+    if ( !PyArg_ParseTuple(args, "y*:rledecode_hqx", &pin) )
         return NULL;
     in_data = pin.buf;
     in_len = pin.len;
@@ -1121,7 +1121,7 @@
     char* retbuf;
     Py_ssize_t i, j;
 
-    if (!PyArg_ParseTuple(args, "s*:a2b_hex", &parg))
+    if (!PyArg_ParseTuple(args, "y*:a2b_hex", &parg))
         return NULL;
     argbuf = parg.buf;
     arglen = parg.len;
@@ -1199,7 +1199,7 @@
     static char *kwlist[] = {"data", "header", NULL};
     int header = 0;
 
-    if (!PyArg_ParseTupleAndKeywords(args, kwargs, "s*|i", kwlist, &pdata,
+    if (!PyArg_ParseTupleAndKeywords(args, kwargs, "y*|i", kwlist, &pdata,
           &header))
         return NULL;
     data = pdata.buf;

Modified: python/branches/import_unicode/Modules/cmathmodule.c
==============================================================================
--- python/branches/import_unicode/Modules/cmathmodule.c	(original)
+++ python/branches/import_unicode/Modules/cmathmodule.c	Fri Jul 30 01:19:37 2010
@@ -1024,6 +1024,19 @@
 Convert from polar coordinates to rectangular coordinates.");
 
 static PyObject *
+cmath_isfinite(PyObject *self, PyObject *args)
+{
+    Py_complex z;
+    if (!PyArg_ParseTuple(args, "D:isfinite", &z))
+        return NULL;
+    return PyBool_FromLong(Py_IS_FINITE(z.real) && Py_IS_FINITE(z.imag));
+}
+
+PyDoc_STRVAR(cmath_isfinite_doc,
+"isfinite(z) -> bool\n\
+Return True if both the real and imaginary parts of z are finite, else False.");
+
+static PyObject *
 cmath_isnan(PyObject *self, PyObject *args)
 {
     Py_complex z;
@@ -1065,6 +1078,7 @@
     {"cos",    cmath_cos,   METH_VARARGS, c_cos_doc},
     {"cosh",   cmath_cosh,  METH_VARARGS, c_cosh_doc},
     {"exp",    cmath_exp,   METH_VARARGS, c_exp_doc},
+    {"isfinite", cmath_isfinite, METH_VARARGS, cmath_isfinite_doc},
     {"isinf",  cmath_isinf, METH_VARARGS, cmath_isinf_doc},
     {"isnan",  cmath_isnan, METH_VARARGS, cmath_isnan_doc},
     {"log",    cmath_log,   METH_VARARGS, cmath_log_doc},

Deleted: python/branches/import_unicode/Modules/datetimemodule.c
==============================================================================
--- python/branches/import_unicode/Modules/datetimemodule.c	Fri Jul 30 01:19:37 2010
+++ (empty file)
@@ -1,5488 +0,0 @@
-/*  C implementation for the date/time type documented at
- *  http://www.zope.org/Members/fdrake/DateTimeWiki/FrontPage
- */
-
-#include "Python.h"
-#include "modsupport.h"
-#include "structmember.h"
-
-#include <time.h>
-
-#include "timefuncs.h"
-
-/* Differentiate between building the core module and building extension
- * modules.
- */
-#ifndef Py_BUILD_CORE
-#define Py_BUILD_CORE
-#endif
-#include "datetime.h"
-#undef Py_BUILD_CORE
-
-/* We require that C int be at least 32 bits, and use int virtually
- * everywhere.  In just a few cases we use a temp long, where a Python
- * API returns a C long.  In such cases, we have to ensure that the
- * final result fits in a C int (this can be an issue on 64-bit boxes).
- */
-#if SIZEOF_INT < 4
-#       error "datetime.c requires that C int have at least 32 bits"
-#endif
-
-#define MINYEAR 1
-#define MAXYEAR 9999
-#define MAXORDINAL 3652059 /* date(9999,12,31).toordinal() */
-
-/* Nine decimal digits is easy to communicate, and leaves enough room
- * so that two delta days can be added w/o fear of overflowing a signed
- * 32-bit int, and with plenty of room left over to absorb any possible
- * carries from adding seconds.
- */
-#define MAX_DELTA_DAYS 999999999
-
-/* Rename the long macros in datetime.h to more reasonable short names. */
-#define GET_YEAR                PyDateTime_GET_YEAR
-#define GET_MONTH               PyDateTime_GET_MONTH
-#define GET_DAY                 PyDateTime_GET_DAY
-#define DATE_GET_HOUR           PyDateTime_DATE_GET_HOUR
-#define DATE_GET_MINUTE         PyDateTime_DATE_GET_MINUTE
-#define DATE_GET_SECOND         PyDateTime_DATE_GET_SECOND
-#define DATE_GET_MICROSECOND    PyDateTime_DATE_GET_MICROSECOND
-
-/* Date accessors for date and datetime. */
-#define SET_YEAR(o, v)          (((o)->data[0] = ((v) & 0xff00) >> 8), \
-                 ((o)->data[1] = ((v) & 0x00ff)))
-#define SET_MONTH(o, v)         (PyDateTime_GET_MONTH(o) = (v))
-#define SET_DAY(o, v)           (PyDateTime_GET_DAY(o) = (v))
-
-/* Date/Time accessors for datetime. */
-#define DATE_SET_HOUR(o, v)     (PyDateTime_DATE_GET_HOUR(o) = (v))
-#define DATE_SET_MINUTE(o, v)   (PyDateTime_DATE_GET_MINUTE(o) = (v))
-#define DATE_SET_SECOND(o, v)   (PyDateTime_DATE_GET_SECOND(o) = (v))
-#define DATE_SET_MICROSECOND(o, v)      \
-    (((o)->data[7] = ((v) & 0xff0000) >> 16), \
-     ((o)->data[8] = ((v) & 0x00ff00) >> 8), \
-     ((o)->data[9] = ((v) & 0x0000ff)))
-
-/* Time accessors for time. */
-#define TIME_GET_HOUR           PyDateTime_TIME_GET_HOUR
-#define TIME_GET_MINUTE         PyDateTime_TIME_GET_MINUTE
-#define TIME_GET_SECOND         PyDateTime_TIME_GET_SECOND
-#define TIME_GET_MICROSECOND    PyDateTime_TIME_GET_MICROSECOND
-#define TIME_SET_HOUR(o, v)     (PyDateTime_TIME_GET_HOUR(o) = (v))
-#define TIME_SET_MINUTE(o, v)   (PyDateTime_TIME_GET_MINUTE(o) = (v))
-#define TIME_SET_SECOND(o, v)   (PyDateTime_TIME_GET_SECOND(o) = (v))
-#define TIME_SET_MICROSECOND(o, v)      \
-    (((o)->data[3] = ((v) & 0xff0000) >> 16), \
-     ((o)->data[4] = ((v) & 0x00ff00) >> 8), \
-     ((o)->data[5] = ((v) & 0x0000ff)))
-
-/* Delta accessors for timedelta. */
-#define GET_TD_DAYS(o)          (((PyDateTime_Delta *)(o))->days)
-#define GET_TD_SECONDS(o)       (((PyDateTime_Delta *)(o))->seconds)
-#define GET_TD_MICROSECONDS(o)  (((PyDateTime_Delta *)(o))->microseconds)
-
-#define SET_TD_DAYS(o, v)       ((o)->days = (v))
-#define SET_TD_SECONDS(o, v)    ((o)->seconds = (v))
-#define SET_TD_MICROSECONDS(o, v) ((o)->microseconds = (v))
-
-/* p is a pointer to a time or a datetime object; HASTZINFO(p) returns
- * p->hastzinfo.
- */
-#define HASTZINFO(p) (((_PyDateTime_BaseTZInfo *)(p))->hastzinfo)
-#define GET_TIME_TZINFO(p) (HASTZINFO(p) ? \
-                            ((PyDateTime_Time *)(p))->tzinfo : Py_None)
-#define GET_DT_TZINFO(p) (HASTZINFO(p) ? \
-                          ((PyDateTime_DateTime *)(p))->tzinfo : Py_None)
-/* M is a char or int claiming to be a valid month.  The macro is equivalent
- * to the two-sided Python test
- *      1 <= M <= 12
- */
-#define MONTH_IS_SANE(M) ((unsigned int)(M) - 1 < 12)
-
-/* Forward declarations. */
-static PyTypeObject PyDateTime_DateType;
-static PyTypeObject PyDateTime_DateTimeType;
-static PyTypeObject PyDateTime_DeltaType;
-static PyTypeObject PyDateTime_TimeType;
-static PyTypeObject PyDateTime_TZInfoType;
-static PyTypeObject PyDateTime_TimeZoneType;
-
-/* ---------------------------------------------------------------------------
- * Math utilities.
- */
-
-/* k = i+j overflows iff k differs in sign from both inputs,
- * iff k^i has sign bit set and k^j has sign bit set,
- * iff (k^i)&(k^j) has sign bit set.
- */
-#define SIGNED_ADD_OVERFLOWED(RESULT, I, J) \
-    ((((RESULT) ^ (I)) & ((RESULT) ^ (J))) < 0)
-
-/* Compute Python divmod(x, y), returning the quotient and storing the
- * remainder into *r.  The quotient is the floor of x/y, and that's
- * the real point of this.  C will probably truncate instead (C99
- * requires truncation; C89 left it implementation-defined).
- * Simplification:  we *require* that y > 0 here.  That's appropriate
- * for all the uses made of it.  This simplifies the code and makes
- * the overflow case impossible (divmod(LONG_MIN, -1) is the only
- * overflow case).
- */
-static int
-divmod(int x, int y, int *r)
-{
-    int quo;
-
-    assert(y > 0);
-    quo = x / y;
-    *r = x - quo * y;
-    if (*r < 0) {
-        --quo;
-        *r += y;
-    }
-    assert(0 <= *r && *r < y);
-    return quo;
-}
-
-/* Round a double to the nearest long.  |x| must be small enough to fit
- * in a C long; this is not checked.
- */
-static long
-round_to_long(double x)
-{
-    if (x >= 0.0)
-        x = floor(x + 0.5);
-    else
-        x = ceil(x - 0.5);
-    return (long)x;
-}
-
-/* Nearest integer to m / n for integers m and n. Half-integer results
- * are rounded to even.
- */
-static PyObject *
-divide_nearest(PyObject *m, PyObject *n)
-{
-    PyObject *result;
-    PyObject *temp;
-
-    temp = _PyLong_DivmodNear(m, n);
-    if (temp == NULL)
-        return NULL;
-    result = PyTuple_GET_ITEM(temp, 0);
-    Py_INCREF(result);
-    Py_DECREF(temp);
-
-    return result;
-}
-
-/* ---------------------------------------------------------------------------
- * General calendrical helper functions
- */
-
-/* For each month ordinal in 1..12, the number of days in that month,
- * and the number of days before that month in the same year.  These
- * are correct for non-leap years only.
- */
-static int _days_in_month[] = {
-    0, /* unused; this vector uses 1-based indexing */
-    31, 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31
-};
-
-static int _days_before_month[] = {
-    0, /* unused; this vector uses 1-based indexing */
-    0, 31, 59, 90, 120, 151, 181, 212, 243, 273, 304, 334
-};
-
-/* year -> 1 if leap year, else 0. */
-static int
-is_leap(int year)
-{
-    /* Cast year to unsigned.  The result is the same either way, but
-     * C can generate faster code for unsigned mod than for signed
-     * mod (especially for % 4 -- a good compiler should just grab
-     * the last 2 bits when the LHS is unsigned).
-     */
-    const unsigned int ayear = (unsigned int)year;
-    return ayear % 4 == 0 && (ayear % 100 != 0 || ayear % 400 == 0);
-}
-
-/* year, month -> number of days in that month in that year */
-static int
-days_in_month(int year, int month)
-{
-    assert(month >= 1);
-    assert(month <= 12);
-    if (month == 2 && is_leap(year))
-        return 29;
-    else
-        return _days_in_month[month];
-}
-
-/* year, month -> number of days in year preceeding first day of month */
-static int
-days_before_month(int year, int month)
-{
-    int days;
-
-    assert(month >= 1);
-    assert(month <= 12);
-    days = _days_before_month[month];
-    if (month > 2 && is_leap(year))
-        ++days;
-    return days;
-}
-
-/* year -> number of days before January 1st of year.  Remember that we
- * start with year 1, so days_before_year(1) == 0.
- */
-static int
-days_before_year(int year)
-{
-    int y = year - 1;
-    /* This is incorrect if year <= 0; we really want the floor
-     * here.  But so long as MINYEAR is 1, the smallest year this
-     * can see is 0 (this can happen in some normalization endcases),
-     * so we'll just special-case that.
-     */
-    assert (year >= 0);
-    if (y >= 0)
-        return y*365 + y/4 - y/100 + y/400;
-    else {
-        assert(y == -1);
-        return -366;
-    }
-}
-
-/* Number of days in 4, 100, and 400 year cycles.  That these have
- * the correct values is asserted in the module init function.
- */
-#define DI4Y    1461    /* days_before_year(5); days in 4 years */
-#define DI100Y  36524   /* days_before_year(101); days in 100 years */
-#define DI400Y  146097  /* days_before_year(401); days in 400 years  */
-
-/* ordinal -> year, month, day, considering 01-Jan-0001 as day 1. */
-static void
-ord_to_ymd(int ordinal, int *year, int *month, int *day)
-{
-    int n, n1, n4, n100, n400, leapyear, preceding;
-
-    /* ordinal is a 1-based index, starting at 1-Jan-1.  The pattern of
-     * leap years repeats exactly every 400 years.  The basic strategy is
-     * to find the closest 400-year boundary at or before ordinal, then
-     * work with the offset from that boundary to ordinal.  Life is much
-     * clearer if we subtract 1 from ordinal first -- then the values
-     * of ordinal at 400-year boundaries are exactly those divisible
-     * by DI400Y:
-     *
-     *    D  M   Y            n              n-1
-     *    -- --- ----        ----------     ----------------
-     *    31 Dec -400        -DI400Y       -DI400Y -1
-     *     1 Jan -399         -DI400Y +1   -DI400Y      400-year boundary
-     *    ...
-     *    30 Dec  000        -1             -2
-     *    31 Dec  000         0             -1
-     *     1 Jan  001         1              0          400-year boundary
-     *     2 Jan  001         2              1
-     *     3 Jan  001         3              2
-     *    ...
-     *    31 Dec  400         DI400Y        DI400Y -1
-     *     1 Jan  401         DI400Y +1     DI400Y      400-year boundary
-     */
-    assert(ordinal >= 1);
-    --ordinal;
-    n400 = ordinal / DI400Y;
-    n = ordinal % DI400Y;
-    *year = n400 * 400 + 1;
-
-    /* Now n is the (non-negative) offset, in days, from January 1 of
-     * year, to the desired date.  Now compute how many 100-year cycles
-     * precede n.
-     * Note that it's possible for n100 to equal 4!  In that case 4 full
-     * 100-year cycles precede the desired day, which implies the
-     * desired day is December 31 at the end of a 400-year cycle.
-     */
-    n100 = n / DI100Y;
-    n = n % DI100Y;
-
-    /* Now compute how many 4-year cycles precede it. */
-    n4 = n / DI4Y;
-    n = n % DI4Y;
-
-    /* And now how many single years.  Again n1 can be 4, and again
-     * meaning that the desired day is December 31 at the end of the
-     * 4-year cycle.
-     */
-    n1 = n / 365;
-    n = n % 365;
-
-    *year += n100 * 100 + n4 * 4 + n1;
-    if (n1 == 4 || n100 == 4) {
-        assert(n == 0);
-        *year -= 1;
-        *month = 12;
-        *day = 31;
-        return;
-    }
-
-    /* Now the year is correct, and n is the offset from January 1.  We
-     * find the month via an estimate that's either exact or one too
-     * large.
-     */
-    leapyear = n1 == 3 && (n4 != 24 || n100 == 3);
-    assert(leapyear == is_leap(*year));
-    *month = (n + 50) >> 5;
-    preceding = (_days_before_month[*month] + (*month > 2 && leapyear));
-    if (preceding > n) {
-        /* estimate is too large */
-        *month -= 1;
-        preceding -= days_in_month(*year, *month);
-    }
-    n -= preceding;
-    assert(0 <= n);
-    assert(n < days_in_month(*year, *month));
-
-    *day = n + 1;
-}
-
-/* year, month, day -> ordinal, considering 01-Jan-0001 as day 1. */
-static int
-ymd_to_ord(int year, int month, int day)
-{
-    return days_before_year(year) + days_before_month(year, month) + day;
-}
-
-/* Day of week, where Monday==0, ..., Sunday==6.  1/1/1 was a Monday. */
-static int
-weekday(int year, int month, int day)
-{
-    return (ymd_to_ord(year, month, day) + 6) % 7;
-}
-
-/* Ordinal of the Monday starting week 1 of the ISO year.  Week 1 is the
- * first calendar week containing a Thursday.
- */
-static int
-iso_week1_monday(int year)
-{
-    int first_day = ymd_to_ord(year, 1, 1);     /* ord of 1/1 */
-    /* 0 if 1/1 is a Monday, 1 if a Tue, etc. */
-    int first_weekday = (first_day + 6) % 7;
-    /* ordinal of closest Monday at or before 1/1 */
-    int week1_monday  = first_day - first_weekday;
-
-    if (first_weekday > 3)      /* if 1/1 was Fri, Sat, Sun */
-        week1_monday += 7;
-    return week1_monday;
-}
-
-/* ---------------------------------------------------------------------------
- * Range checkers.
- */
-
-/* Check that -MAX_DELTA_DAYS <= days <= MAX_DELTA_DAYS.  If so, return 0.
- * If not, raise OverflowError and return -1.
- */
-static int
-check_delta_day_range(int days)
-{
-    if (-MAX_DELTA_DAYS <= days && days <= MAX_DELTA_DAYS)
-        return 0;
-    PyErr_Format(PyExc_OverflowError,
-                 "days=%d; must have magnitude <= %d",
-                 days, MAX_DELTA_DAYS);
-    return -1;
-}
-
-/* Check that date arguments are in range.  Return 0 if they are.  If they
- * aren't, raise ValueError and return -1.
- */
-static int
-check_date_args(int year, int month, int day)
-{
-
-    if (year < MINYEAR || year > MAXYEAR) {
-        PyErr_SetString(PyExc_ValueError,
-                        "year is out of range");
-        return -1;
-    }
-    if (month < 1 || month > 12) {
-        PyErr_SetString(PyExc_ValueError,
-                        "month must be in 1..12");
-        return -1;
-    }
-    if (day < 1 || day > days_in_month(year, month)) {
-        PyErr_SetString(PyExc_ValueError,
-                        "day is out of range for month");
-        return -1;
-    }
-    return 0;
-}
-
-/* Check that time arguments are in range.  Return 0 if they are.  If they
- * aren't, raise ValueError and return -1.
- */
-static int
-check_time_args(int h, int m, int s, int us)
-{
-    if (h < 0 || h > 23) {
-        PyErr_SetString(PyExc_ValueError,
-                        "hour must be in 0..23");
-        return -1;
-    }
-    if (m < 0 || m > 59) {
-        PyErr_SetString(PyExc_ValueError,
-                        "minute must be in 0..59");
-        return -1;
-    }
-    if (s < 0 || s > 59) {
-        PyErr_SetString(PyExc_ValueError,
-                        "second must be in 0..59");
-        return -1;
-    }
-    if (us < 0 || us > 999999) {
-        PyErr_SetString(PyExc_ValueError,
-                        "microsecond must be in 0..999999");
-        return -1;
-    }
-    return 0;
-}
-
-/* ---------------------------------------------------------------------------
- * Normalization utilities.
- */
-
-/* One step of a mixed-radix conversion.  A "hi" unit is equivalent to
- * factor "lo" units.  factor must be > 0.  If *lo is less than 0, or
- * at least factor, enough of *lo is converted into "hi" units so that
- * 0 <= *lo < factor.  The input values must be such that int overflow
- * is impossible.
- */
-static void
-normalize_pair(int *hi, int *lo, int factor)
-{
-    assert(factor > 0);
-    assert(lo != hi);
-    if (*lo < 0 || *lo >= factor) {
-        const int num_hi = divmod(*lo, factor, lo);
-        const int new_hi = *hi + num_hi;
-        assert(! SIGNED_ADD_OVERFLOWED(new_hi, *hi, num_hi));
-        *hi = new_hi;
-    }
-    assert(0 <= *lo && *lo < factor);
-}
-
-/* Fiddle days (d), seconds (s), and microseconds (us) so that
- *      0 <= *s < 24*3600
- *      0 <= *us < 1000000
- * The input values must be such that the internals don't overflow.
- * The way this routine is used, we don't get close.
- */
-static void
-normalize_d_s_us(int *d, int *s, int *us)
-{
-    if (*us < 0 || *us >= 1000000) {
-        normalize_pair(s, us, 1000000);
-        /* |s| can't be bigger than about
-         * |original s| + |original us|/1000000 now.
-         */
-
-    }
-    if (*s < 0 || *s >= 24*3600) {
-        normalize_pair(d, s, 24*3600);
-        /* |d| can't be bigger than about
-         * |original d| +
-         * (|original s| + |original us|/1000000) / (24*3600) now.
-         */
-    }
-    assert(0 <= *s && *s < 24*3600);
-    assert(0 <= *us && *us < 1000000);
-}
-
-/* Fiddle years (y), months (m), and days (d) so that
- *      1 <= *m <= 12
- *      1 <= *d <= days_in_month(*y, *m)
- * The input values must be such that the internals don't overflow.
- * The way this routine is used, we don't get close.
- */
-static int
-normalize_y_m_d(int *y, int *m, int *d)
-{
-    int dim;            /* # of days in month */
-
-    /* This gets muddy:  the proper range for day can't be determined
-     * without knowing the correct month and year, but if day is, e.g.,
-     * plus or minus a million, the current month and year values make
-     * no sense (and may also be out of bounds themselves).
-     * Saying 12 months == 1 year should be non-controversial.
-     */
-    if (*m < 1 || *m > 12) {
-        --*m;
-        normalize_pair(y, m, 12);
-        ++*m;
-        /* |y| can't be bigger than about
-         * |original y| + |original m|/12 now.
-         */
-    }
-    assert(1 <= *m && *m <= 12);
-
-    /* Now only day can be out of bounds (year may also be out of bounds
-     * for a datetime object, but we don't care about that here).
-     * If day is out of bounds, what to do is arguable, but at least the
-     * method here is principled and explainable.
-     */
-    dim = days_in_month(*y, *m);
-    if (*d < 1 || *d > dim) {
-        /* Move day-1 days from the first of the month.  First try to
-         * get off cheap if we're only one day out of range
-         * (adjustments for timezone alone can't be worse than that).
-         */
-        if (*d == 0) {
-            --*m;
-            if (*m > 0)
-                *d = days_in_month(*y, *m);
-            else {
-                --*y;
-                *m = 12;
-                *d = 31;
-            }
-        }
-        else if (*d == dim + 1) {
-            /* move forward a day */
-            ++*m;
-            *d = 1;
-            if (*m > 12) {
-                *m = 1;
-                ++*y;
-            }
-        }
-        else {
-            int ordinal = ymd_to_ord(*y, *m, 1) +
-                                      *d - 1;
-            if (ordinal < 1 || ordinal > MAXORDINAL) {
-                goto error;
-            } else {
-                ord_to_ymd(ordinal, y, m, d);
-                return 0;
-            }
-        }
-    }
-    assert(*m > 0);
-    assert(*d > 0);
-    if (MINYEAR <= *y && *y <= MAXYEAR)
-        return 0;
- error:
-    PyErr_SetString(PyExc_OverflowError,
-            "date value out of range");
-    return -1;
-
-}
-
-/* Fiddle out-of-bounds months and days so that the result makes some kind
- * of sense.  The parameters are both inputs and outputs.  Returns < 0 on
- * failure, where failure means the adjusted year is out of bounds.
- */
-static int
-normalize_date(int *year, int *month, int *day)
-{
-    return normalize_y_m_d(year, month, day);
-}
-
-/* Force all the datetime fields into range.  The parameters are both
- * inputs and outputs.  Returns < 0 on error.
- */
-static int
-normalize_datetime(int *year, int *month, int *day,
-                   int *hour, int *minute, int *second,
-                   int *microsecond)
-{
-    normalize_pair(second, microsecond, 1000000);
-    normalize_pair(minute, second, 60);
-    normalize_pair(hour, minute, 60);
-    normalize_pair(day, hour, 24);
-    return normalize_date(year, month, day);
-}
-
-/* ---------------------------------------------------------------------------
- * Basic object allocation:  tp_alloc implementations.  These allocate
- * Python objects of the right size and type, and do the Python object-
- * initialization bit.  If there's not enough memory, they return NULL after
- * setting MemoryError.  All data members remain uninitialized trash.
- *
- * We abuse the tp_alloc "nitems" argument to communicate whether a tzinfo
- * member is needed.  This is ugly, imprecise, and possibly insecure.
- * tp_basicsize for the time and datetime types is set to the size of the
- * struct that has room for the tzinfo member, so subclasses in Python will
- * allocate enough space for a tzinfo member whether or not one is actually
- * needed.  That's the "ugly and imprecise" parts.  The "possibly insecure"
- * part is that PyType_GenericAlloc() (which subclasses in Python end up
- * using) just happens today to effectively ignore the nitems argument
- * when tp_itemsize is 0, which it is for these type objects.  If that
- * changes, perhaps the callers of tp_alloc slots in this file should
- * be changed to force a 0 nitems argument unless the type being allocated
- * is a base type implemented in this file (so that tp_alloc is time_alloc
- * or datetime_alloc below, which know about the nitems abuse).
- */
-
-static PyObject *
-time_alloc(PyTypeObject *type, Py_ssize_t aware)
-{
-    PyObject *self;
-
-    self = (PyObject *)
-        PyObject_MALLOC(aware ?
-                        sizeof(PyDateTime_Time) :
-                sizeof(_PyDateTime_BaseTime));
-    if (self == NULL)
-        return (PyObject *)PyErr_NoMemory();
-    PyObject_INIT(self, type);
-    return self;
-}
-
-static PyObject *
-datetime_alloc(PyTypeObject *type, Py_ssize_t aware)
-{
-    PyObject *self;
-
-    self = (PyObject *)
-        PyObject_MALLOC(aware ?
-                        sizeof(PyDateTime_DateTime) :
-                sizeof(_PyDateTime_BaseDateTime));
-    if (self == NULL)
-        return (PyObject *)PyErr_NoMemory();
-    PyObject_INIT(self, type);
-    return self;
-}
-
-/* ---------------------------------------------------------------------------
- * Helpers for setting object fields.  These work on pointers to the
- * appropriate base class.
- */
-
-/* For date and datetime. */
-static void
-set_date_fields(PyDateTime_Date *self, int y, int m, int d)
-{
-    self->hashcode = -1;
-    SET_YEAR(self, y);
-    SET_MONTH(self, m);
-    SET_DAY(self, d);
-}
-
-/* ---------------------------------------------------------------------------
- * Create various objects, mostly without range checking.
- */
-
-/* Create a date instance with no range checking. */
-static PyObject *
-new_date_ex(int year, int month, int day, PyTypeObject *type)
-{
-    PyDateTime_Date *self;
-
-    self = (PyDateTime_Date *) (type->tp_alloc(type, 0));
-    if (self != NULL)
-        set_date_fields(self, year, month, day);
-    return (PyObject *) self;
-}
-
-#define new_date(year, month, day) \
-    new_date_ex(year, month, day, &PyDateTime_DateType)
-
-/* Create a datetime instance with no range checking. */
-static PyObject *
-new_datetime_ex(int year, int month, int day, int hour, int minute,
-             int second, int usecond, PyObject *tzinfo, PyTypeObject *type)
-{
-    PyDateTime_DateTime *self;
-    char aware = tzinfo != Py_None;
-
-    self = (PyDateTime_DateTime *) (type->tp_alloc(type, aware));
-    if (self != NULL) {
-        self->hastzinfo = aware;
-        set_date_fields((PyDateTime_Date *)self, year, month, day);
-        DATE_SET_HOUR(self, hour);
-        DATE_SET_MINUTE(self, minute);
-        DATE_SET_SECOND(self, second);
-        DATE_SET_MICROSECOND(self, usecond);
-        if (aware) {
-            Py_INCREF(tzinfo);
-            self->tzinfo = tzinfo;
-        }
-    }
-    return (PyObject *)self;
-}
-
-#define new_datetime(y, m, d, hh, mm, ss, us, tzinfo)           \
-    new_datetime_ex(y, m, d, hh, mm, ss, us, tzinfo,            \
-                    &PyDateTime_DateTimeType)
-
-/* Create a time instance with no range checking. */
-static PyObject *
-new_time_ex(int hour, int minute, int second, int usecond,
-            PyObject *tzinfo, PyTypeObject *type)
-{
-    PyDateTime_Time *self;
-    char aware = tzinfo != Py_None;
-
-    self = (PyDateTime_Time *) (type->tp_alloc(type, aware));
-    if (self != NULL) {
-        self->hastzinfo = aware;
-        self->hashcode = -1;
-        TIME_SET_HOUR(self, hour);
-        TIME_SET_MINUTE(self, minute);
-        TIME_SET_SECOND(self, second);
-        TIME_SET_MICROSECOND(self, usecond);
-        if (aware) {
-            Py_INCREF(tzinfo);
-            self->tzinfo = tzinfo;
-        }
-    }
-    return (PyObject *)self;
-}
-
-#define new_time(hh, mm, ss, us, tzinfo)                \
-    new_time_ex(hh, mm, ss, us, tzinfo, &PyDateTime_TimeType)
-
-/* Create a timedelta instance.  Normalize the members iff normalize is
- * true.  Passing false is a speed optimization, if you know for sure
- * that seconds and microseconds are already in their proper ranges.  In any
- * case, raises OverflowError and returns NULL if the normalized days is out
- * of range).
- */
-static PyObject *
-new_delta_ex(int days, int seconds, int microseconds, int normalize,
-             PyTypeObject *type)
-{
-    PyDateTime_Delta *self;
-
-    if (normalize)
-        normalize_d_s_us(&days, &seconds, &microseconds);
-    assert(0 <= seconds && seconds < 24*3600);
-    assert(0 <= microseconds && microseconds < 1000000);
-
-    if (check_delta_day_range(days) < 0)
-        return NULL;
-
-    self = (PyDateTime_Delta *) (type->tp_alloc(type, 0));
-    if (self != NULL) {
-        self->hashcode = -1;
-        SET_TD_DAYS(self, days);
-        SET_TD_SECONDS(self, seconds);
-        SET_TD_MICROSECONDS(self, microseconds);
-    }
-    return (PyObject *) self;
-}
-
-#define new_delta(d, s, us, normalize)  \
-    new_delta_ex(d, s, us, normalize, &PyDateTime_DeltaType)
-
-
-typedef struct
-{
-    PyObject_HEAD
-    PyObject *offset;
-    PyObject *name;
-} PyDateTime_TimeZone;
-
-PyObject *PyDateTime_TimeZone_UTC;
-
-/* Create new timezone instance checking offset range.  This
-   function does not check the name argument.  Caller must assure
-   that offset is a timedelta instance and name is either NULL
-   or a unicode object. */
-static PyObject *
-new_timezone(PyObject *offset, PyObject *name)
-{
-    PyDateTime_TimeZone *self;
-    PyTypeObject *type = &PyDateTime_TimeZoneType;
-
-    assert(offset != NULL);
-    assert(PyDelta_Check(offset));
-    assert(name == NULL || PyUnicode_Check(name));
-
-    if (GET_TD_MICROSECONDS(offset) != 0 || GET_TD_SECONDS(offset) % 60 != 0) {
-        PyErr_Format(PyExc_ValueError, "offset must be a timedelta"
-                     " representing a whole number of minutes");
-        return NULL;
-    }
-    if ((GET_TD_DAYS(offset) == -1 && GET_TD_SECONDS(offset) == 0) ||
-        GET_TD_DAYS(offset) < -1 || GET_TD_DAYS(offset) >= 1) {
-        PyErr_Format(PyExc_ValueError, "offset must be a timedelta"
-                     " strictly between -timedelta(hours=24) and"
-                     " timedelta(hours=24).");
-        return NULL;
-    }
-
-    self = (PyDateTime_TimeZone *)(type->tp_alloc(type, 0));
-    if (self == NULL) {
-        return NULL;
-    }
-    Py_INCREF(offset);
-    self->offset = offset;
-    Py_XINCREF(name);
-    self->name = name;
-    return (PyObject *)self;
-}
-
-/* ---------------------------------------------------------------------------
- * tzinfo helpers.
- */
-
-/* Ensure that p is None or of a tzinfo subclass.  Return 0 if OK; if not
- * raise TypeError and return -1.
- */
-static int
-check_tzinfo_subclass(PyObject *p)
-{
-    if (p == Py_None || PyTZInfo_Check(p))
-        return 0;
-    PyErr_Format(PyExc_TypeError,
-                 "tzinfo argument must be None or of a tzinfo subclass, "
-                 "not type '%s'",
-                 Py_TYPE(p)->tp_name);
-    return -1;
-}
-
-/* If self has a tzinfo member, return a BORROWED reference to it.  Else
- * return NULL, which is NOT AN ERROR.  There are no error returns here,
- * and the caller must not decref the result.
- */
-static PyObject *
-get_tzinfo_member(PyObject *self)
-{
-    PyObject *tzinfo = NULL;
-
-    if (PyDateTime_Check(self) && HASTZINFO(self))
-        tzinfo = ((PyDateTime_DateTime *)self)->tzinfo;
-    else if (PyTime_Check(self) && HASTZINFO(self))
-        tzinfo = ((PyDateTime_Time *)self)->tzinfo;
-
-    return tzinfo;
-}
-
-/* Call getattr(tzinfo, name)(tzinfoarg), and check the result.  tzinfo must
- * be an instance of the tzinfo class.  If the method returns None, this
- * returns None.  If the method doesn't return None or timedelta, TypeError is
- * raised and this returns NULL.  If it returns a timedelta and the value is
- * out of range or isn't a whole number of minutes, ValueError is raised and
- * this returns NULL.  Else result is returned.
- */
-static PyObject *
-call_tzinfo_method(PyObject *tzinfo, char *name, PyObject *tzinfoarg)
-{
-    PyObject *offset;
-
-    assert(tzinfo != NULL);
-    assert(PyTZInfo_Check(tzinfo) || tzinfo == Py_None);
-    assert(tzinfoarg != NULL);
-
-    if (tzinfo == Py_None)
-        Py_RETURN_NONE;
-    offset = PyObject_CallMethod(tzinfo, name, "O", tzinfoarg);
-    if (offset == Py_None || offset == NULL)
-        return offset;
-    if (PyDelta_Check(offset)) {
-        if (GET_TD_MICROSECONDS(offset) != 0 || GET_TD_SECONDS(offset) % 60 != 0) {
-            Py_DECREF(offset);
-            PyErr_Format(PyExc_ValueError, "offset must be a timedelta"
-                         " representing a whole number of minutes");
-            return NULL;
-        }
-        if ((GET_TD_DAYS(offset) == -1 && GET_TD_SECONDS(offset) == 0) ||
-            GET_TD_DAYS(offset) < -1 || GET_TD_DAYS(offset) >= 1) {
-            Py_DECREF(offset);
-            PyErr_Format(PyExc_ValueError, "offset must be a timedelta"
-                         " strictly between -timedelta(hours=24) and"
-                         " timedelta(hours=24).");
-            return NULL;
-        }
-    }
-    else {
-        Py_DECREF(offset);
-        PyErr_Format(PyExc_TypeError,
-                     "tzinfo.%s() must return None or "
-                     "timedelta, not '%.200s'",
-                     name, Py_TYPE(offset)->tp_name);
-        return NULL;
-    }
-
-    return offset;
-}
-
-/* Call tzinfo.utcoffset(tzinfoarg), and extract an integer from the
- * result.  tzinfo must be an instance of the tzinfo class.  If utcoffset()
- * returns None, call_utcoffset returns 0 and sets *none to 1.  If uctoffset()
- * doesn't return None or timedelta, TypeError is raised and this returns -1.
- * If utcoffset() returns an invalid timedelta (out of range, or not a whole
- * # of minutes), ValueError is raised and this returns -1.  Else *none is
- * set to 0 and the offset is returned (as int # of minutes east of UTC).
- */
-static PyObject *
-call_utcoffset(PyObject *tzinfo, PyObject *tzinfoarg)
-{
-    return call_tzinfo_method(tzinfo, "utcoffset", tzinfoarg);
-}
-
-/* Call tzinfo.dst(tzinfoarg), and extract an integer from the
- * result.  tzinfo must be an instance of the tzinfo class.  If dst()
- * returns None, call_dst returns 0 and sets *none to 1.  If dst()
- & doesn't return None or timedelta, TypeError is raised and this
- * returns -1.  If dst() returns an invalid timedelta for a UTC offset,
- * ValueError is raised and this returns -1.  Else *none is set to 0 and
- * the offset is returned (as an int # of minutes east of UTC).
- */
-static PyObject *
-call_dst(PyObject *tzinfo, PyObject *tzinfoarg)
-{
-    return call_tzinfo_method(tzinfo, "dst", tzinfoarg);
-}
-
-/* Call tzinfo.tzname(tzinfoarg), and return the result.  tzinfo must be
- * an instance of the tzinfo class or None.  If tzinfo isn't None, and
- * tzname() doesn't return None or a string, TypeError is raised and this
- * returns NULL.  If the result is a string, we ensure it is a Unicode
- * string.
- */
-static PyObject *
-call_tzname(PyObject *tzinfo, PyObject *tzinfoarg)
-{
-    PyObject *result;
-
-    assert(tzinfo != NULL);
-    assert(check_tzinfo_subclass(tzinfo) >= 0);
-    assert(tzinfoarg != NULL);
-
-    if (tzinfo == Py_None)
-        Py_RETURN_NONE;
-
-    result = PyObject_CallMethod(tzinfo, "tzname", "O", tzinfoarg);
-
-    if (result == NULL || result == Py_None)
-        return result;
-
-    if (!PyUnicode_Check(result)) {
-        PyErr_Format(PyExc_TypeError, "tzinfo.tzname() must "
-                     "return None or a string, not '%s'",
-                     Py_TYPE(result)->tp_name);
-        Py_DECREF(result);
-        result = NULL;
-    }
-
-    return result;
-}
-
-/* repr is like "someclass(arg1, arg2)".  If tzinfo isn't None,
- * stuff
- *     ", tzinfo=" + repr(tzinfo)
- * before the closing ")".
- */
-static PyObject *
-append_keyword_tzinfo(PyObject *repr, PyObject *tzinfo)
-{
-    PyObject *temp;
-
-    assert(PyUnicode_Check(repr));
-    assert(tzinfo);
-    if (tzinfo == Py_None)
-        return repr;
-    /* Get rid of the trailing ')'. */
-    assert(PyUnicode_AS_UNICODE(repr)[PyUnicode_GET_SIZE(repr)-1] == ')');
-    temp = PyUnicode_FromUnicode(PyUnicode_AS_UNICODE(repr),
-                                      PyUnicode_GET_SIZE(repr) - 1);
-    Py_DECREF(repr);
-    if (temp == NULL)
-        return NULL;
-    repr = PyUnicode_FromFormat("%U, tzinfo=%R)", temp, tzinfo);
-    Py_DECREF(temp);
-    return repr;
-}
-
-/* ---------------------------------------------------------------------------
- * String format helpers.
- */
-
-static PyObject *
-format_ctime(PyDateTime_Date *date, int hours, int minutes, int seconds)
-{
-    static const char *DayNames[] = {
-        "Mon", "Tue", "Wed", "Thu", "Fri", "Sat", "Sun"
-    };
-    static const char *MonthNames[] = {
-        "Jan", "Feb", "Mar", "Apr", "May", "Jun",
-        "Jul", "Aug", "Sep", "Oct", "Nov", "Dec"
-    };
-
-    int wday = weekday(GET_YEAR(date), GET_MONTH(date), GET_DAY(date));
-
-    return PyUnicode_FromFormat("%s %s %2d %02d:%02d:%02d %04d",
-                                DayNames[wday], MonthNames[GET_MONTH(date)-1],
-                                GET_DAY(date), hours, minutes, seconds,
-                                GET_YEAR(date));
-}
-
-static PyObject *delta_negative(PyDateTime_Delta *self);
-
-/* Add an hours & minutes UTC offset string to buf.  buf has no more than
- * buflen bytes remaining.  The UTC offset is gotten by calling
- * tzinfo.uctoffset(tzinfoarg).  If that returns None, \0 is stored into
- * *buf, and that's all.  Else the returned value is checked for sanity (an
- * integer in range), and if that's OK it's converted to an hours & minutes
- * string of the form
- *   sign HH sep MM
- * Returns 0 if everything is OK.  If the return value from utcoffset() is
- * bogus, an appropriate exception is set and -1 is returned.
- */
-static int
-format_utcoffset(char *buf, size_t buflen, const char *sep,
-                PyObject *tzinfo, PyObject *tzinfoarg)
-{
-    PyObject *offset;
-    int hours, minutes, seconds;
-    char sign;
-
-    assert(buflen >= 1);
-
-    offset = call_utcoffset(tzinfo, tzinfoarg);
-    if (offset == NULL)
-        return -1;
-    if (offset == Py_None) {
-        Py_DECREF(offset);
-        *buf = '\0';
-        return 0;
-    }
-    /* Offset is normalized, so it is negative if days < 0 */
-    if (GET_TD_DAYS(offset) < 0) {
-        PyObject *temp = offset;
-        sign = '-';
-        offset = delta_negative((PyDateTime_Delta *)offset);
-        Py_DECREF(temp);
-        if (offset == NULL)
-            return -1;
-    }
-    else {
-        sign = '+';
-    }
-    /* Offset is not negative here. */
-    seconds = GET_TD_SECONDS(offset);
-    Py_DECREF(offset);
-    minutes = divmod(seconds, 60, &seconds);
-    hours = divmod(minutes, 60, &minutes);
-    assert(seconds == 0);
-    /* XXX ignore sub-minute data, curently not allowed. */
-    PyOS_snprintf(buf, buflen, "%c%02d%s%02d", sign, hours, sep, minutes);
-
-    return 0;
-}
-
-static PyObject *
-make_Zreplacement(PyObject *object, PyObject *tzinfoarg)
-{
-    PyObject *temp;
-    PyObject *tzinfo = get_tzinfo_member(object);
-    PyObject *Zreplacement = PyUnicode_FromStringAndSize(NULL, 0);
-    if (Zreplacement == NULL)
-        return NULL;
-    if (tzinfo == Py_None || tzinfo == NULL)
-        return Zreplacement;
-
-    assert(tzinfoarg != NULL);
-    temp = call_tzname(tzinfo, tzinfoarg);
-    if (temp == NULL)
-        goto Error;
-    if (temp == Py_None) {
-        Py_DECREF(temp);
-        return Zreplacement;
-    }
-
-    assert(PyUnicode_Check(temp));
-    /* Since the tzname is getting stuffed into the
-     * format, we have to double any % signs so that
-     * strftime doesn't treat them as format codes.
-     */
-    Py_DECREF(Zreplacement);
-    Zreplacement = PyObject_CallMethod(temp, "replace", "ss", "%", "%%");
-    Py_DECREF(temp);
-    if (Zreplacement == NULL)
-        return NULL;
-    if (!PyUnicode_Check(Zreplacement)) {
-        PyErr_SetString(PyExc_TypeError,
-                        "tzname.replace() did not return a string");
-        goto Error;
-    }
-    return Zreplacement;
-
-  Error:
-    Py_DECREF(Zreplacement);
-    return NULL;
-}
-
-static PyObject *
-make_freplacement(PyObject *object)
-{
-    char freplacement[64];
-    if (PyTime_Check(object))
-        sprintf(freplacement, "%06d", TIME_GET_MICROSECOND(object));
-    else if (PyDateTime_Check(object))
-        sprintf(freplacement, "%06d", DATE_GET_MICROSECOND(object));
-    else
-        sprintf(freplacement, "%06d", 0);
-
-    return PyBytes_FromStringAndSize(freplacement, strlen(freplacement));
-}
-
-/* I sure don't want to reproduce the strftime code from the time module,
- * so this imports the module and calls it.  All the hair is due to
- * giving special meanings to the %z, %Z and %f format codes via a
- * preprocessing step on the format string.
- * tzinfoarg is the argument to pass to the object's tzinfo method, if
- * needed.
- */
-static PyObject *
-wrap_strftime(PyObject *object, PyObject *format, PyObject *timetuple,
-              PyObject *tzinfoarg)
-{
-    PyObject *result = NULL;            /* guilty until proved innocent */
-
-    PyObject *zreplacement = NULL;      /* py string, replacement for %z */
-    PyObject *Zreplacement = NULL;      /* py string, replacement for %Z */
-    PyObject *freplacement = NULL;      /* py string, replacement for %f */
-
-    const char *pin;            /* pointer to next char in input format */
-    Py_ssize_t flen;            /* length of input format */
-    char ch;                    /* next char in input format */
-
-    PyObject *newfmt = NULL;            /* py string, the output format */
-    char *pnew;         /* pointer to available byte in output format */
-    size_t totalnew;            /* number bytes total in output format buffer,
-                               exclusive of trailing \0 */
-    size_t usednew;     /* number bytes used so far in output format buffer */
-
-    const char *ptoappend;      /* ptr to string to append to output buffer */
-    Py_ssize_t ntoappend;       /* # of bytes to append to output buffer */
-
-    assert(object && format && timetuple);
-    assert(PyUnicode_Check(format));
-    /* Convert the input format to a C string and size */
-    pin = _PyUnicode_AsStringAndSize(format, &flen);
-    if (!pin)
-        return NULL;
-
-    /* Give up if the year is before 1900.
-     * Python strftime() plays games with the year, and different
-     * games depending on whether envar PYTHON2K is set.  This makes
-     * years before 1900 a nightmare, even if the platform strftime
-     * supports them (and not all do).
-     * We could get a lot farther here by avoiding Python's strftime
-     * wrapper and calling the C strftime() directly, but that isn't
-     * an option in the Python implementation of this module.
-     */
-    {
-        long year;
-        PyObject *pyyear = PySequence_GetItem(timetuple, 0);
-        if (pyyear == NULL) return NULL;
-        assert(PyLong_Check(pyyear));
-        year = PyLong_AsLong(pyyear);
-        Py_DECREF(pyyear);
-        if (year < 1900) {
-            PyErr_Format(PyExc_ValueError, "year=%ld is before "
-                         "1900; the datetime strftime() "
-                         "methods require year >= 1900",
-                         year);
-            return NULL;
-        }
-    }
-
-    /* Scan the input format, looking for %z/%Z/%f escapes, building
-     * a new format.  Since computing the replacements for those codes
-     * is expensive, don't unless they're actually used.
-     */
-    if (flen > INT_MAX - 1) {
-        PyErr_NoMemory();
-        goto Done;
-    }
-
-    totalnew = flen + 1;        /* realistic if no %z/%Z */
-    newfmt = PyBytes_FromStringAndSize(NULL, totalnew);
-    if (newfmt == NULL) goto Done;
-    pnew = PyBytes_AsString(newfmt);
-    usednew = 0;
-
-    while ((ch = *pin++) != '\0') {
-        if (ch != '%') {
-            ptoappend = pin - 1;
-            ntoappend = 1;
-        }
-        else if ((ch = *pin++) == '\0') {
-            /* There's a lone trailing %; doesn't make sense. */
-            PyErr_SetString(PyExc_ValueError, "strftime format "
-                            "ends with raw %");
-            goto Done;
-        }
-        /* A % has been seen and ch is the character after it. */
-        else if (ch == 'z') {
-            if (zreplacement == NULL) {
-                /* format utcoffset */
-                char buf[100];
-                PyObject *tzinfo = get_tzinfo_member(object);
-                zreplacement = PyBytes_FromStringAndSize("", 0);
-                if (zreplacement == NULL) goto Done;
-                if (tzinfo != Py_None && tzinfo != NULL) {
-                    assert(tzinfoarg != NULL);
-                    if (format_utcoffset(buf,
-                                         sizeof(buf),
-                                         "",
-                                         tzinfo,
-                                         tzinfoarg) < 0)
-                        goto Done;
-                    Py_DECREF(zreplacement);
-                    zreplacement =
-                      PyBytes_FromStringAndSize(buf,
-                                               strlen(buf));
-                    if (zreplacement == NULL)
-                        goto Done;
-                }
-            }
-            assert(zreplacement != NULL);
-            ptoappend = PyBytes_AS_STRING(zreplacement);
-            ntoappend = PyBytes_GET_SIZE(zreplacement);
-        }
-        else if (ch == 'Z') {
-            /* format tzname */
-            if (Zreplacement == NULL) {
-                Zreplacement = make_Zreplacement(object,
-                                                 tzinfoarg);
-                if (Zreplacement == NULL)
-                    goto Done;
-            }
-            assert(Zreplacement != NULL);
-            assert(PyUnicode_Check(Zreplacement));
-            ptoappend = _PyUnicode_AsStringAndSize(Zreplacement,
-                                                  &ntoappend);
-            ntoappend = Py_SIZE(Zreplacement);
-        }
-        else if (ch == 'f') {
-            /* format microseconds */
-            if (freplacement == NULL) {
-                freplacement = make_freplacement(object);
-                if (freplacement == NULL)
-                    goto Done;
-            }
-            assert(freplacement != NULL);
-            assert(PyBytes_Check(freplacement));
-            ptoappend = PyBytes_AS_STRING(freplacement);
-            ntoappend = PyBytes_GET_SIZE(freplacement);
-        }
-        else {
-            /* percent followed by neither z nor Z */
-            ptoappend = pin - 2;
-            ntoappend = 2;
-        }
-
-        /* Append the ntoappend chars starting at ptoappend to
-         * the new format.
-         */
-        if (ntoappend == 0)
-            continue;
-        assert(ptoappend != NULL);
-        assert(ntoappend > 0);
-        while (usednew + ntoappend > totalnew) {
-            size_t bigger = totalnew << 1;
-            if ((bigger >> 1) != totalnew) { /* overflow */
-                PyErr_NoMemory();
-                goto Done;
-            }
-            if (_PyBytes_Resize(&newfmt, bigger) < 0)
-                goto Done;
-            totalnew = bigger;
-            pnew = PyBytes_AsString(newfmt) + usednew;
-        }
-        memcpy(pnew, ptoappend, ntoappend);
-        pnew += ntoappend;
-        usednew += ntoappend;
-        assert(usednew <= totalnew);
-    }  /* end while() */
-
-    if (_PyBytes_Resize(&newfmt, usednew) < 0)
-        goto Done;
-    {
-        PyObject *format;
-        PyObject *time = PyImport_ImportModuleNoBlock("time");
-        if (time == NULL)
-            goto Done;
-        format = PyUnicode_FromString(PyBytes_AS_STRING(newfmt));
-        if (format != NULL) {
-            result = PyObject_CallMethod(time, "strftime", "OO",
-                                         format, timetuple, NULL);
-            Py_DECREF(format);
-        }
-        Py_DECREF(time);
-    }
- Done:
-    Py_XDECREF(freplacement);
-    Py_XDECREF(zreplacement);
-    Py_XDECREF(Zreplacement);
-    Py_XDECREF(newfmt);
-    return result;
-}
-
-/* ---------------------------------------------------------------------------
- * Wrap functions from the time module.  These aren't directly available
- * from C.  Perhaps they should be.
- */
-
-/* Call time.time() and return its result (a Python float). */
-static PyObject *
-time_time(void)
-{
-    PyObject *result = NULL;
-    PyObject *time = PyImport_ImportModuleNoBlock("time");
-
-    if (time != NULL) {
-        result = PyObject_CallMethod(time, "time", "()");
-        Py_DECREF(time);
-    }
-    return result;
-}
-
-/* Build a time.struct_time.  The weekday and day number are automatically
- * computed from the y,m,d args.
- */
-static PyObject *
-build_struct_time(int y, int m, int d, int hh, int mm, int ss, int dstflag)
-{
-    PyObject *time;
-    PyObject *result = NULL;
-
-    time = PyImport_ImportModuleNoBlock("time");
-    if (time != NULL) {
-        result = PyObject_CallMethod(time, "struct_time",
-                                     "((iiiiiiiii))",
-                                     y, m, d,
-                                     hh, mm, ss,
-                                     weekday(y, m, d),
-                                     days_before_month(y, m) + d,
-                                     dstflag);
-        Py_DECREF(time);
-    }
-    return result;
-}
-
-/* ---------------------------------------------------------------------------
- * Miscellaneous helpers.
- */
-
-/* For various reasons, we need to use tp_richcompare instead of tp_reserved.
- * The comparisons here all most naturally compute a cmp()-like result.
- * This little helper turns that into a bool result for rich comparisons.
- */
-static PyObject *
-diff_to_bool(int diff, int op)
-{
-    PyObject *result;
-    int istrue;
-
-    switch (op) {
-        case Py_EQ: istrue = diff == 0; break;
-        case Py_NE: istrue = diff != 0; break;
-        case Py_LE: istrue = diff <= 0; break;
-        case Py_GE: istrue = diff >= 0; break;
-        case Py_LT: istrue = diff < 0; break;
-        case Py_GT: istrue = diff > 0; break;
-        default:
-            assert(! "op unknown");
-            istrue = 0; /* To shut up compiler */
-    }
-    result = istrue ? Py_True : Py_False;
-    Py_INCREF(result);
-    return result;
-}
-
-/* Raises a "can't compare" TypeError and returns NULL. */
-static PyObject *
-cmperror(PyObject *a, PyObject *b)
-{
-    PyErr_Format(PyExc_TypeError,
-                 "can't compare %s to %s",
-                 Py_TYPE(a)->tp_name, Py_TYPE(b)->tp_name);
-    return NULL;
-}
-
-/* ---------------------------------------------------------------------------
- * Cached Python objects; these are set by the module init function.
- */
-
-/* Conversion factors. */
-static PyObject *us_per_us = NULL;      /* 1 */
-static PyObject *us_per_ms = NULL;      /* 1000 */
-static PyObject *us_per_second = NULL;  /* 1000000 */
-static PyObject *us_per_minute = NULL;  /* 1e6 * 60 as Python int */
-static PyObject *us_per_hour = NULL;    /* 1e6 * 3600 as Python long */
-static PyObject *us_per_day = NULL;     /* 1e6 * 3600 * 24 as Python long */
-static PyObject *us_per_week = NULL;    /* 1e6*3600*24*7 as Python long */
-static PyObject *seconds_per_day = NULL; /* 3600*24 as Python int */
-
-/* ---------------------------------------------------------------------------
- * Class implementations.
- */
-
-/*
- * PyDateTime_Delta implementation.
- */
-
-/* Convert a timedelta to a number of us,
- *      (24*3600*self.days + self.seconds)*1000000 + self.microseconds
- * as a Python int or long.
- * Doing mixed-radix arithmetic by hand instead is excruciating in C,
- * due to ubiquitous overflow possibilities.
- */
-static PyObject *
-delta_to_microseconds(PyDateTime_Delta *self)
-{
-    PyObject *x1 = NULL;
-    PyObject *x2 = NULL;
-    PyObject *x3 = NULL;
-    PyObject *result = NULL;
-
-    x1 = PyLong_FromLong(GET_TD_DAYS(self));
-    if (x1 == NULL)
-        goto Done;
-    x2 = PyNumber_Multiply(x1, seconds_per_day);        /* days in seconds */
-    if (x2 == NULL)
-        goto Done;
-    Py_DECREF(x1);
-    x1 = NULL;
-
-    /* x2 has days in seconds */
-    x1 = PyLong_FromLong(GET_TD_SECONDS(self));         /* seconds */
-    if (x1 == NULL)
-        goto Done;
-    x3 = PyNumber_Add(x1, x2);          /* days and seconds in seconds */
-    if (x3 == NULL)
-        goto Done;
-    Py_DECREF(x1);
-    Py_DECREF(x2);
-    x1 = x2 = NULL;
-
-    /* x3 has days+seconds in seconds */
-    x1 = PyNumber_Multiply(x3, us_per_second);          /* us */
-    if (x1 == NULL)
-        goto Done;
-    Py_DECREF(x3);
-    x3 = NULL;
-
-    /* x1 has days+seconds in us */
-    x2 = PyLong_FromLong(GET_TD_MICROSECONDS(self));
-    if (x2 == NULL)
-        goto Done;
-    result = PyNumber_Add(x1, x2);
-
-Done:
-    Py_XDECREF(x1);
-    Py_XDECREF(x2);
-    Py_XDECREF(x3);
-    return result;
-}
-
-/* Convert a number of us (as a Python int or long) to a timedelta.
- */
-static PyObject *
-microseconds_to_delta_ex(PyObject *pyus, PyTypeObject *type)
-{
-    int us;
-    int s;
-    int d;
-    long temp;
-
-    PyObject *tuple = NULL;
-    PyObject *num = NULL;
-    PyObject *result = NULL;
-
-    tuple = PyNumber_Divmod(pyus, us_per_second);
-    if (tuple == NULL)
-        goto Done;
-
-    num = PyTuple_GetItem(tuple, 1);            /* us */
-    if (num == NULL)
-        goto Done;
-    temp = PyLong_AsLong(num);
-    num = NULL;
-    if (temp == -1 && PyErr_Occurred())
-        goto Done;
-    assert(0 <= temp && temp < 1000000);
-    us = (int)temp;
-    if (us < 0) {
-        /* The divisor was positive, so this must be an error. */
-        assert(PyErr_Occurred());
-        goto Done;
-    }
-
-    num = PyTuple_GetItem(tuple, 0);            /* leftover seconds */
-    if (num == NULL)
-        goto Done;
-    Py_INCREF(num);
-    Py_DECREF(tuple);
-
-    tuple = PyNumber_Divmod(num, seconds_per_day);
-    if (tuple == NULL)
-        goto Done;
-    Py_DECREF(num);
-
-    num = PyTuple_GetItem(tuple, 1);            /* seconds */
-    if (num == NULL)
-        goto Done;
-    temp = PyLong_AsLong(num);
-    num = NULL;
-    if (temp == -1 && PyErr_Occurred())
-        goto Done;
-    assert(0 <= temp && temp < 24*3600);
-    s = (int)temp;
-
-    if (s < 0) {
-        /* The divisor was positive, so this must be an error. */
-        assert(PyErr_Occurred());
-        goto Done;
-    }
-
-    num = PyTuple_GetItem(tuple, 0);            /* leftover days */
-    if (num == NULL)
-        goto Done;
-    Py_INCREF(num);
-    temp = PyLong_AsLong(num);
-    if (temp == -1 && PyErr_Occurred())
-        goto Done;
-    d = (int)temp;
-    if ((long)d != temp) {
-        PyErr_SetString(PyExc_OverflowError, "normalized days too "
-                        "large to fit in a C int");
-        goto Done;
-    }
-    result = new_delta_ex(d, s, us, 0, type);
-
-Done:
-    Py_XDECREF(tuple);
-    Py_XDECREF(num);
-    return result;
-}
-
-#define microseconds_to_delta(pymicros) \
-    microseconds_to_delta_ex(pymicros, &PyDateTime_DeltaType)
-
-static PyObject *
-multiply_int_timedelta(PyObject *intobj, PyDateTime_Delta *delta)
-{
-    PyObject *pyus_in;
-    PyObject *pyus_out;
-    PyObject *result;
-
-    pyus_in = delta_to_microseconds(delta);
-    if (pyus_in == NULL)
-        return NULL;
-
-    pyus_out = PyNumber_Multiply(pyus_in, intobj);
-    Py_DECREF(pyus_in);
-    if (pyus_out == NULL)
-        return NULL;
-
-    result = microseconds_to_delta(pyus_out);
-    Py_DECREF(pyus_out);
-    return result;
-}
-
-static PyObject *
-multiply_float_timedelta(PyObject *floatobj, PyDateTime_Delta *delta)
-{
-    PyObject *result = NULL;
-    PyObject *pyus_in = NULL, *temp, *pyus_out;
-    PyObject *ratio = NULL;
-
-    pyus_in = delta_to_microseconds(delta);
-    if (pyus_in == NULL)
-        return NULL;
-    ratio = PyObject_CallMethod(floatobj, "as_integer_ratio", NULL);
-    if (ratio == NULL)
-        goto error;
-    temp = PyNumber_Multiply(pyus_in, PyTuple_GET_ITEM(ratio, 0));
-    Py_DECREF(pyus_in);
-    pyus_in = NULL;
-    if (temp == NULL)
-        goto error;
-    pyus_out = divide_nearest(temp, PyTuple_GET_ITEM(ratio, 1));
-    Py_DECREF(temp);
-    if (pyus_out == NULL)
-        goto error;
-    result = microseconds_to_delta(pyus_out);
-    Py_DECREF(pyus_out);
- error:
-    Py_XDECREF(pyus_in);
-    Py_XDECREF(ratio);
-
-    return result;
-}
-
-static PyObject *
-divide_timedelta_int(PyDateTime_Delta *delta, PyObject *intobj)
-{
-    PyObject *pyus_in;
-    PyObject *pyus_out;
-    PyObject *result;
-
-    pyus_in = delta_to_microseconds(delta);
-    if (pyus_in == NULL)
-        return NULL;
-
-    pyus_out = PyNumber_FloorDivide(pyus_in, intobj);
-    Py_DECREF(pyus_in);
-    if (pyus_out == NULL)
-        return NULL;
-
-    result = microseconds_to_delta(pyus_out);
-    Py_DECREF(pyus_out);
-    return result;
-}
-
-static PyObject *
-divide_timedelta_timedelta(PyDateTime_Delta *left, PyDateTime_Delta *right)
-{
-    PyObject *pyus_left;
-    PyObject *pyus_right;
-    PyObject *result;
-
-    pyus_left = delta_to_microseconds(left);
-    if (pyus_left == NULL)
-        return NULL;
-
-    pyus_right = delta_to_microseconds(right);
-    if (pyus_right == NULL)     {
-        Py_DECREF(pyus_left);
-        return NULL;
-    }
-
-    result = PyNumber_FloorDivide(pyus_left, pyus_right);
-    Py_DECREF(pyus_left);
-    Py_DECREF(pyus_right);
-    return result;
-}
-
-static PyObject *
-truedivide_timedelta_timedelta(PyDateTime_Delta *left, PyDateTime_Delta *right)
-{
-    PyObject *pyus_left;
-    PyObject *pyus_right;
-    PyObject *result;
-
-    pyus_left = delta_to_microseconds(left);
-    if (pyus_left == NULL)
-        return NULL;
-
-    pyus_right = delta_to_microseconds(right);
-    if (pyus_right == NULL)     {
-        Py_DECREF(pyus_left);
-        return NULL;
-    }
-
-    result = PyNumber_TrueDivide(pyus_left, pyus_right);
-    Py_DECREF(pyus_left);
-    Py_DECREF(pyus_right);
-    return result;
-}
-
-static PyObject *
-truedivide_timedelta_float(PyDateTime_Delta *delta, PyObject *f)
-{
-    PyObject *result = NULL;
-    PyObject *pyus_in = NULL, *temp, *pyus_out;
-    PyObject *ratio = NULL;
-
-    pyus_in = delta_to_microseconds(delta);
-    if (pyus_in == NULL)
-        return NULL;
-    ratio = PyObject_CallMethod(f, "as_integer_ratio", NULL);
-    if (ratio == NULL)
-        goto error;
-    temp = PyNumber_Multiply(pyus_in, PyTuple_GET_ITEM(ratio, 1));
-    Py_DECREF(pyus_in);
-    pyus_in = NULL;
-    if (temp == NULL)
-        goto error;
-    pyus_out = divide_nearest(temp, PyTuple_GET_ITEM(ratio, 0));
-    Py_DECREF(temp);
-    if (pyus_out == NULL)
-        goto error;
-    result = microseconds_to_delta(pyus_out);
-    Py_DECREF(pyus_out);
- error:
-    Py_XDECREF(pyus_in);
-    Py_XDECREF(ratio);
-
-    return result;
-}
-
-static PyObject *
-truedivide_timedelta_int(PyDateTime_Delta *delta, PyObject *i)
-{
-    PyObject *result;
-    PyObject *pyus_in, *pyus_out;
-    pyus_in = delta_to_microseconds(delta);
-    if (pyus_in == NULL)
-        return NULL;
-    pyus_out = divide_nearest(pyus_in, i);
-    Py_DECREF(pyus_in);
-    if (pyus_out == NULL)
-        return NULL;
-    result = microseconds_to_delta(pyus_out);
-    Py_DECREF(pyus_out);
-
-    return result;
-}
-
-static PyObject *
-delta_add(PyObject *left, PyObject *right)
-{
-    PyObject *result = Py_NotImplemented;
-
-    if (PyDelta_Check(left) && PyDelta_Check(right)) {
-        /* delta + delta */
-        /* The C-level additions can't overflow because of the
-         * invariant bounds.
-         */
-        int days = GET_TD_DAYS(left) + GET_TD_DAYS(right);
-        int seconds = GET_TD_SECONDS(left) + GET_TD_SECONDS(right);
-        int microseconds = GET_TD_MICROSECONDS(left) +
-                           GET_TD_MICROSECONDS(right);
-        result = new_delta(days, seconds, microseconds, 1);
-    }
-
-    if (result == Py_NotImplemented)
-        Py_INCREF(result);
-    return result;
-}
-
-static PyObject *
-delta_negative(PyDateTime_Delta *self)
-{
-    return new_delta(-GET_TD_DAYS(self),
-                     -GET_TD_SECONDS(self),
-                     -GET_TD_MICROSECONDS(self),
-                     1);
-}
-
-static PyObject *
-delta_positive(PyDateTime_Delta *self)
-{
-    /* Could optimize this (by returning self) if this isn't a
-     * subclass -- but who uses unary + ?  Approximately nobody.
-     */
-    return new_delta(GET_TD_DAYS(self),
-                     GET_TD_SECONDS(self),
-                     GET_TD_MICROSECONDS(self),
-                     0);
-}
-
-static PyObject *
-delta_abs(PyDateTime_Delta *self)
-{
-    PyObject *result;
-
-    assert(GET_TD_MICROSECONDS(self) >= 0);
-    assert(GET_TD_SECONDS(self) >= 0);
-
-    if (GET_TD_DAYS(self) < 0)
-        result = delta_negative(self);
-    else
-        result = delta_positive(self);
-
-    return result;
-}
-
-static PyObject *
-delta_subtract(PyObject *left, PyObject *right)
-{
-    PyObject *result = Py_NotImplemented;
-
-    if (PyDelta_Check(left) && PyDelta_Check(right)) {
-        /* delta - delta */
-        PyObject *minus_right = PyNumber_Negative(right);
-        if (minus_right) {
-            result = delta_add(left, minus_right);
-            Py_DECREF(minus_right);
-        }
-        else
-            result = NULL;
-    }
-
-    if (result == Py_NotImplemented)
-        Py_INCREF(result);
-    return result;
-}
-
-static int
-delta_cmp(PyObject *self, PyObject *other)
-{
-    int diff = GET_TD_DAYS(self) - GET_TD_DAYS(other);
-    if (diff == 0) {
-        diff = GET_TD_SECONDS(self) - GET_TD_SECONDS(other);
-        if (diff == 0)
-            diff = GET_TD_MICROSECONDS(self) -
-                GET_TD_MICROSECONDS(other);
-    }
-    return diff;
-}
-
-static PyObject *
-delta_richcompare(PyObject *self, PyObject *other, int op)
-{
-    if (PyDelta_Check(other)) {
-        int diff = delta_cmp(self, other);
-        return diff_to_bool(diff, op);
-    }
-    else {
-        Py_INCREF(Py_NotImplemented);
-        return Py_NotImplemented;
-    }
-}
-
-static PyObject *delta_getstate(PyDateTime_Delta *self);
-
-static long
-delta_hash(PyDateTime_Delta *self)
-{
-    if (self->hashcode == -1) {
-        PyObject *temp = delta_getstate(self);
-        if (temp != NULL) {
-            self->hashcode = PyObject_Hash(temp);
-            Py_DECREF(temp);
-        }
-    }
-    return self->hashcode;
-}
-
-static PyObject *
-delta_multiply(PyObject *left, PyObject *right)
-{
-    PyObject *result = Py_NotImplemented;
-
-    if (PyDelta_Check(left)) {
-        /* delta * ??? */
-        if (PyLong_Check(right))
-            result = multiply_int_timedelta(right,
-                            (PyDateTime_Delta *) left);
-        else if (PyFloat_Check(right))
-            result = multiply_float_timedelta(right,
-                            (PyDateTime_Delta *) left);
-    }
-    else if (PyLong_Check(left))
-        result = multiply_int_timedelta(left,
-                        (PyDateTime_Delta *) right);
-    else if (PyFloat_Check(left))
-        result = multiply_float_timedelta(left,
-                        (PyDateTime_Delta *) right);
-
-    if (result == Py_NotImplemented)
-        Py_INCREF(result);
-    return result;
-}
-
-static PyObject *
-delta_divide(PyObject *left, PyObject *right)
-{
-    PyObject *result = Py_NotImplemented;
-
-    if (PyDelta_Check(left)) {
-        /* delta * ??? */
-        if (PyLong_Check(right))
-            result = divide_timedelta_int(
-                            (PyDateTime_Delta *)left,
-                            right);
-        else if (PyDelta_Check(right))
-            result = divide_timedelta_timedelta(
-                            (PyDateTime_Delta *)left,
-                            (PyDateTime_Delta *)right);
-    }
-
-    if (result == Py_NotImplemented)
-        Py_INCREF(result);
-    return result;
-}
-
-static PyObject *
-delta_truedivide(PyObject *left, PyObject *right)
-{
-    PyObject *result = Py_NotImplemented;
-
-    if (PyDelta_Check(left)) {
-        if (PyDelta_Check(right))
-            result = truedivide_timedelta_timedelta(
-                            (PyDateTime_Delta *)left,
-                            (PyDateTime_Delta *)right);
-        else if (PyFloat_Check(right))
-            result = truedivide_timedelta_float(
-                            (PyDateTime_Delta *)left, right);
-        else if (PyLong_Check(right))
-            result = truedivide_timedelta_int(
-                            (PyDateTime_Delta *)left, right);
-    }
-
-    if (result == Py_NotImplemented)
-        Py_INCREF(result);
-    return result;
-}
-
-static PyObject *
-delta_remainder(PyObject *left, PyObject *right)
-{
-    PyObject *pyus_left;
-    PyObject *pyus_right;
-    PyObject *pyus_remainder;
-    PyObject *remainder;
-
-    if (!PyDelta_Check(left) || !PyDelta_Check(right)) {
-        Py_INCREF(Py_NotImplemented);
-        return Py_NotImplemented;
-    }
-
-    pyus_left = delta_to_microseconds((PyDateTime_Delta *)left);
-    if (pyus_left == NULL)
-        return NULL;
-
-    pyus_right = delta_to_microseconds((PyDateTime_Delta *)right);
-    if (pyus_right == NULL) {
-        Py_DECREF(pyus_left);
-        return NULL;
-    }
-
-    pyus_remainder = PyNumber_Remainder(pyus_left, pyus_right);
-    Py_DECREF(pyus_left);
-    Py_DECREF(pyus_right);
-    if (pyus_remainder == NULL)
-        return NULL;
-
-    remainder = microseconds_to_delta(pyus_remainder);
-    Py_DECREF(pyus_remainder);
-    if (remainder == NULL)
-        return NULL;
-
-    return remainder;
-}
-
-static PyObject *
-delta_divmod(PyObject *left, PyObject *right)
-{
-    PyObject *pyus_left;
-    PyObject *pyus_right;
-    PyObject *divmod;
-    PyObject *delta;
-    PyObject *result;
-
-    if (!PyDelta_Check(left) || !PyDelta_Check(right)) {
-        Py_INCREF(Py_NotImplemented);
-        return Py_NotImplemented;
-    }
-
-    pyus_left = delta_to_microseconds((PyDateTime_Delta *)left);
-    if (pyus_left == NULL)
-        return NULL;
-
-    pyus_right = delta_to_microseconds((PyDateTime_Delta *)right);
-    if (pyus_right == NULL) {
-        Py_DECREF(pyus_left);
-        return NULL;
-    }
-
-    divmod = PyNumber_Divmod(pyus_left, pyus_right);
-    Py_DECREF(pyus_left);
-    Py_DECREF(pyus_right);
-    if (divmod == NULL)
-        return NULL;
-
-    assert(PyTuple_Size(divmod) == 2);
-    delta = microseconds_to_delta(PyTuple_GET_ITEM(divmod, 1));
-    if (delta == NULL) {
-        Py_DECREF(divmod);
-        return NULL;
-    }
-    result = PyTuple_Pack(2, PyTuple_GET_ITEM(divmod, 0), delta);
-    Py_DECREF(delta);
-    Py_DECREF(divmod);
-    return result;
-}
-
-/* Fold in the value of the tag ("seconds", "weeks", etc) component of a
- * timedelta constructor.  sofar is the # of microseconds accounted for
- * so far, and there are factor microseconds per current unit, the number
- * of which is given by num.  num * factor is added to sofar in a
- * numerically careful way, and that's the result.  Any fractional
- * microseconds left over (this can happen if num is a float type) are
- * added into *leftover.
- * Note that there are many ways this can give an error (NULL) return.
- */
-static PyObject *
-accum(const char* tag, PyObject *sofar, PyObject *num, PyObject *factor,
-      double *leftover)
-{
-    PyObject *prod;
-    PyObject *sum;
-
-    assert(num != NULL);
-
-    if (PyLong_Check(num)) {
-        prod = PyNumber_Multiply(num, factor);
-        if (prod == NULL)
-            return NULL;
-        sum = PyNumber_Add(sofar, prod);
-        Py_DECREF(prod);
-        return sum;
-    }
-
-    if (PyFloat_Check(num)) {
-        double dnum;
-        double fracpart;
-        double intpart;
-        PyObject *x;
-        PyObject *y;
-
-        /* The Plan:  decompose num into an integer part and a
-         * fractional part, num = intpart + fracpart.
-         * Then num * factor ==
-         *      intpart * factor + fracpart * factor
-         * and the LHS can be computed exactly in long arithmetic.
-         * The RHS is again broken into an int part and frac part.
-         * and the frac part is added into *leftover.
-         */
-        dnum = PyFloat_AsDouble(num);
-        if (dnum == -1.0 && PyErr_Occurred())
-            return NULL;
-        fracpart = modf(dnum, &intpart);
-        x = PyLong_FromDouble(intpart);
-        if (x == NULL)
-            return NULL;
-
-        prod = PyNumber_Multiply(x, factor);
-        Py_DECREF(x);
-        if (prod == NULL)
-            return NULL;
-
-        sum = PyNumber_Add(sofar, prod);
-        Py_DECREF(prod);
-        if (sum == NULL)
-            return NULL;
-
-        if (fracpart == 0.0)
-            return sum;
-        /* So far we've lost no information.  Dealing with the
-         * fractional part requires float arithmetic, and may
-         * lose a little info.
-         */
-        assert(PyLong_Check(factor));
-        dnum = PyLong_AsDouble(factor);
-
-        dnum *= fracpart;
-        fracpart = modf(dnum, &intpart);
-        x = PyLong_FromDouble(intpart);
-        if (x == NULL) {
-            Py_DECREF(sum);
-            return NULL;
-        }
-
-        y = PyNumber_Add(sum, x);
-        Py_DECREF(sum);
-        Py_DECREF(x);
-        *leftover += fracpart;
-        return y;
-    }
-
-    PyErr_Format(PyExc_TypeError,
-                 "unsupported type for timedelta %s component: %s",
-                 tag, Py_TYPE(num)->tp_name);
-    return NULL;
-}
-
-static PyObject *
-delta_new(PyTypeObject *type, PyObject *args, PyObject *kw)
-{
-    PyObject *self = NULL;
-
-    /* Argument objects. */
-    PyObject *day = NULL;
-    PyObject *second = NULL;
-    PyObject *us = NULL;
-    PyObject *ms = NULL;
-    PyObject *minute = NULL;
-    PyObject *hour = NULL;
-    PyObject *week = NULL;
-
-    PyObject *x = NULL;         /* running sum of microseconds */
-    PyObject *y = NULL;         /* temp sum of microseconds */
-    double leftover_us = 0.0;
-
-    static char *keywords[] = {
-        "days", "seconds", "microseconds", "milliseconds",
-        "minutes", "hours", "weeks", NULL
-    };
-
-    if (PyArg_ParseTupleAndKeywords(args, kw, "|OOOOOOO:__new__",
-                                    keywords,
-                                    &day, &second, &us,
-                                    &ms, &minute, &hour, &week) == 0)
-        goto Done;
-
-    x = PyLong_FromLong(0);
-    if (x == NULL)
-        goto Done;
-
-#define CLEANUP         \
-    Py_DECREF(x);       \
-    x = y;              \
-    if (x == NULL)      \
-        goto Done
-
-    if (us) {
-        y = accum("microseconds", x, us, us_per_us, &leftover_us);
-        CLEANUP;
-    }
-    if (ms) {
-        y = accum("milliseconds", x, ms, us_per_ms, &leftover_us);
-        CLEANUP;
-    }
-    if (second) {
-        y = accum("seconds", x, second, us_per_second, &leftover_us);
-        CLEANUP;
-    }
-    if (minute) {
-        y = accum("minutes", x, minute, us_per_minute, &leftover_us);
-        CLEANUP;
-    }
-    if (hour) {
-        y = accum("hours", x, hour, us_per_hour, &leftover_us);
-        CLEANUP;
-    }
-    if (day) {
-        y = accum("days", x, day, us_per_day, &leftover_us);
-        CLEANUP;
-    }
-    if (week) {
-        y = accum("weeks", x, week, us_per_week, &leftover_us);
-        CLEANUP;
-    }
-    if (leftover_us) {
-        /* Round to nearest whole # of us, and add into x. */
-        PyObject *temp = PyLong_FromLong(round_to_long(leftover_us));
-        if (temp == NULL) {
-            Py_DECREF(x);
-            goto Done;
-        }
-        y = PyNumber_Add(x, temp);
-        Py_DECREF(temp);
-        CLEANUP;
-    }
-
-    self = microseconds_to_delta_ex(x, type);
-    Py_DECREF(x);
-Done:
-    return self;
-
-#undef CLEANUP
-}
-
-static int
-delta_bool(PyDateTime_Delta *self)
-{
-    return (GET_TD_DAYS(self) != 0
-        || GET_TD_SECONDS(self) != 0
-        || GET_TD_MICROSECONDS(self) != 0);
-}
-
-static PyObject *
-delta_repr(PyDateTime_Delta *self)
-{
-    if (GET_TD_MICROSECONDS(self) != 0)
-        return PyUnicode_FromFormat("%s(%d, %d, %d)",
-                                    Py_TYPE(self)->tp_name,
-                                    GET_TD_DAYS(self),
-                                    GET_TD_SECONDS(self),
-                                    GET_TD_MICROSECONDS(self));
-    if (GET_TD_SECONDS(self) != 0)
-        return PyUnicode_FromFormat("%s(%d, %d)",
-                                    Py_TYPE(self)->tp_name,
-                                    GET_TD_DAYS(self),
-                                    GET_TD_SECONDS(self));
-
-    return PyUnicode_FromFormat("%s(%d)",
-                                Py_TYPE(self)->tp_name,
-                                GET_TD_DAYS(self));
-}
-
-static PyObject *
-delta_str(PyDateTime_Delta *self)
-{
-    int us = GET_TD_MICROSECONDS(self);
-    int seconds = GET_TD_SECONDS(self);
-    int minutes = divmod(seconds, 60, &seconds);
-    int hours = divmod(minutes, 60, &minutes);
-    int days = GET_TD_DAYS(self);
-
-    if (days) {
-        if (us)
-            return PyUnicode_FromFormat("%d day%s, %d:%02d:%02d.%06d",
-                                        days, (days == 1 || days == -1) ? "" : "s",
-                                        hours, minutes, seconds, us);
-        else
-            return PyUnicode_FromFormat("%d day%s, %d:%02d:%02d",
-                                        days, (days == 1 || days == -1) ? "" : "s",
-                                        hours, minutes, seconds);
-    } else {
-        if (us)
-            return PyUnicode_FromFormat("%d:%02d:%02d.%06d",
-                                        hours, minutes, seconds, us);
-        else
-            return PyUnicode_FromFormat("%d:%02d:%02d",
-                                        hours, minutes, seconds);
-    }
-
-}
-
-/* Pickle support, a simple use of __reduce__. */
-
-/* __getstate__ isn't exposed */
-static PyObject *
-delta_getstate(PyDateTime_Delta *self)
-{
-    return Py_BuildValue("iii", GET_TD_DAYS(self),
-                                GET_TD_SECONDS(self),
-                                GET_TD_MICROSECONDS(self));
-}
-
-static PyObject *
-delta_total_seconds(PyObject *self)
-{
-    PyObject *total_seconds;
-    PyObject *total_microseconds;
-    PyObject *one_million;
-
-    total_microseconds = delta_to_microseconds((PyDateTime_Delta *)self);
-    if (total_microseconds == NULL)
-        return NULL;
-
-    one_million = PyLong_FromLong(1000000L);
-    if (one_million == NULL) {
-        Py_DECREF(total_microseconds);
-        return NULL;
-    }
-
-    total_seconds = PyNumber_TrueDivide(total_microseconds, one_million);
-
-    Py_DECREF(total_microseconds);
-    Py_DECREF(one_million);
-    return total_seconds;
-}
-
-static PyObject *
-delta_reduce(PyDateTime_Delta* self)
-{
-    return Py_BuildValue("ON", Py_TYPE(self), delta_getstate(self));
-}
-
-#define OFFSET(field)  offsetof(PyDateTime_Delta, field)
-
-static PyMemberDef delta_members[] = {
-
-    {"days",         T_INT, OFFSET(days),         READONLY,
-     PyDoc_STR("Number of days.")},
-
-    {"seconds",      T_INT, OFFSET(seconds),      READONLY,
-     PyDoc_STR("Number of seconds (>= 0 and less than 1 day).")},
-
-    {"microseconds", T_INT, OFFSET(microseconds), READONLY,
-     PyDoc_STR("Number of microseconds (>= 0 and less than 1 second).")},
-    {NULL}
-};
-
-static PyMethodDef delta_methods[] = {
-    {"total_seconds", (PyCFunction)delta_total_seconds, METH_NOARGS,
-     PyDoc_STR("Total seconds in the duration.")},
-
-    {"__reduce__", (PyCFunction)delta_reduce, METH_NOARGS,
-     PyDoc_STR("__reduce__() -> (cls, state)")},
-
-    {NULL,      NULL},
-};
-
-static char delta_doc[] =
-PyDoc_STR("Difference between two datetime values.");
-
-static PyNumberMethods delta_as_number = {
-    delta_add,                                  /* nb_add */
-    delta_subtract,                             /* nb_subtract */
-    delta_multiply,                             /* nb_multiply */
-    delta_remainder,                            /* nb_remainder */
-    delta_divmod,                               /* nb_divmod */
-    0,                                          /* nb_power */
-    (unaryfunc)delta_negative,                  /* nb_negative */
-    (unaryfunc)delta_positive,                  /* nb_positive */
-    (unaryfunc)delta_abs,                       /* nb_absolute */
-    (inquiry)delta_bool,                        /* nb_bool */
-    0,                                          /*nb_invert*/
-    0,                                          /*nb_lshift*/
-    0,                                          /*nb_rshift*/
-    0,                                          /*nb_and*/
-    0,                                          /*nb_xor*/
-    0,                                          /*nb_or*/
-    0,                                          /*nb_int*/
-    0,                                          /*nb_reserved*/
-    0,                                          /*nb_float*/
-    0,                                          /*nb_inplace_add*/
-    0,                                          /*nb_inplace_subtract*/
-    0,                                          /*nb_inplace_multiply*/
-    0,                                          /*nb_inplace_remainder*/
-    0,                                          /*nb_inplace_power*/
-    0,                                          /*nb_inplace_lshift*/
-    0,                                          /*nb_inplace_rshift*/
-    0,                                          /*nb_inplace_and*/
-    0,                                          /*nb_inplace_xor*/
-    0,                                          /*nb_inplace_or*/
-    delta_divide,                               /* nb_floor_divide */
-    delta_truedivide,                           /* nb_true_divide */
-    0,                                          /* nb_inplace_floor_divide */
-    0,                                          /* nb_inplace_true_divide */
-};
-
-static PyTypeObject PyDateTime_DeltaType = {
-    PyVarObject_HEAD_INIT(NULL, 0)
-    "datetime.timedelta",                               /* tp_name */
-    sizeof(PyDateTime_Delta),                           /* tp_basicsize */
-    0,                                                  /* tp_itemsize */
-    0,                                                  /* tp_dealloc */
-    0,                                                  /* tp_print */
-    0,                                                  /* tp_getattr */
-    0,                                                  /* tp_setattr */
-    0,                                                  /* tp_reserved */
-    (reprfunc)delta_repr,                               /* tp_repr */
-    &delta_as_number,                                   /* tp_as_number */
-    0,                                                  /* tp_as_sequence */
-    0,                                                  /* tp_as_mapping */
-    (hashfunc)delta_hash,                               /* tp_hash */
-    0,                                                  /* tp_call */
-    (reprfunc)delta_str,                                /* tp_str */
-    PyObject_GenericGetAttr,                            /* tp_getattro */
-    0,                                                  /* tp_setattro */
-    0,                                                  /* tp_as_buffer */
-    Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE,           /* tp_flags */
-    delta_doc,                                          /* tp_doc */
-    0,                                                  /* tp_traverse */
-    0,                                                  /* tp_clear */
-    delta_richcompare,                                  /* tp_richcompare */
-    0,                                                  /* tp_weaklistoffset */
-    0,                                                  /* tp_iter */
-    0,                                                  /* tp_iternext */
-    delta_methods,                                      /* tp_methods */
-    delta_members,                                      /* tp_members */
-    0,                                                  /* tp_getset */
-    0,                                                  /* tp_base */
-    0,                                                  /* tp_dict */
-    0,                                                  /* tp_descr_get */
-    0,                                                  /* tp_descr_set */
-    0,                                                  /* tp_dictoffset */
-    0,                                                  /* tp_init */
-    0,                                                  /* tp_alloc */
-    delta_new,                                          /* tp_new */
-    0,                                                  /* tp_free */
-};
-
-/*
- * PyDateTime_Date implementation.
- */
-
-/* Accessor properties. */
-
-static PyObject *
-date_year(PyDateTime_Date *self, void *unused)
-{
-    return PyLong_FromLong(GET_YEAR(self));
-}
-
-static PyObject *
-date_month(PyDateTime_Date *self, void *unused)
-{
-    return PyLong_FromLong(GET_MONTH(self));
-}
-
-static PyObject *
-date_day(PyDateTime_Date *self, void *unused)
-{
-    return PyLong_FromLong(GET_DAY(self));
-}
-
-static PyGetSetDef date_getset[] = {
-    {"year",        (getter)date_year},
-    {"month",       (getter)date_month},
-    {"day",         (getter)date_day},
-    {NULL}
-};
-
-/* Constructors. */
-
-static char *date_kws[] = {"year", "month", "day", NULL};
-
-static PyObject *
-date_new(PyTypeObject *type, PyObject *args, PyObject *kw)
-{
-    PyObject *self = NULL;
-    PyObject *state;
-    int year;
-    int month;
-    int day;
-
-    /* Check for invocation from pickle with __getstate__ state */
-    if (PyTuple_GET_SIZE(args) == 1 &&
-        PyBytes_Check(state = PyTuple_GET_ITEM(args, 0)) &&
-        PyBytes_GET_SIZE(state) == _PyDateTime_DATE_DATASIZE &&
-        MONTH_IS_SANE(PyBytes_AS_STRING(state)[2]))
-    {
-        PyDateTime_Date *me;
-
-        me = (PyDateTime_Date *) (type->tp_alloc(type, 0));
-        if (me != NULL) {
-            char *pdata = PyBytes_AS_STRING(state);
-            memcpy(me->data, pdata, _PyDateTime_DATE_DATASIZE);
-            me->hashcode = -1;
-        }
-        return (PyObject *)me;
-    }
-
-    if (PyArg_ParseTupleAndKeywords(args, kw, "iii", date_kws,
-                                    &year, &month, &day)) {
-        if (check_date_args(year, month, day) < 0)
-            return NULL;
-        self = new_date_ex(year, month, day, type);
-    }
-    return self;
-}
-
-/* Return new date from localtime(t). */
-static PyObject *
-date_local_from_time_t(PyObject *cls, double ts)
-{
-    struct tm *tm;
-    time_t t;
-    PyObject *result = NULL;
-
-    t = _PyTime_DoubleToTimet(ts);
-    if (t == (time_t)-1 && PyErr_Occurred())
-        return NULL;
-    tm = localtime(&t);
-    if (tm)
-        result = PyObject_CallFunction(cls, "iii",
-                                       tm->tm_year + 1900,
-                                       tm->tm_mon + 1,
-                                       tm->tm_mday);
-    else
-        PyErr_SetString(PyExc_ValueError,
-                        "timestamp out of range for "
-                        "platform localtime() function");
-    return result;
-}
-
-/* Return new date from current time.
- * We say this is equivalent to fromtimestamp(time.time()), and the
- * only way to be sure of that is to *call* time.time().  That's not
- * generally the same as calling C's time.
- */
-static PyObject *
-date_today(PyObject *cls, PyObject *dummy)
-{
-    PyObject *time;
-    PyObject *result;
-
-    time = time_time();
-    if (time == NULL)
-        return NULL;
-
-    /* Note well:  today() is a class method, so this may not call
-     * date.fromtimestamp.  For example, it may call
-     * datetime.fromtimestamp.  That's why we need all the accuracy
-     * time.time() delivers; if someone were gonzo about optimization,
-     * date.today() could get away with plain C time().
-     */
-    result = PyObject_CallMethod(cls, "fromtimestamp", "O", time);
-    Py_DECREF(time);
-    return result;
-}
-
-/* Return new date from given timestamp (Python timestamp -- a double). */
-static PyObject *
-date_fromtimestamp(PyObject *cls, PyObject *args)
-{
-    double timestamp;
-    PyObject *result = NULL;
-
-    if (PyArg_ParseTuple(args, "d:fromtimestamp", &timestamp))
-        result = date_local_from_time_t(cls, timestamp);
-    return result;
-}
-
-/* Return new date from proleptic Gregorian ordinal.  Raises ValueError if
- * the ordinal is out of range.
- */
-static PyObject *
-date_fromordinal(PyObject *cls, PyObject *args)
-{
-    PyObject *result = NULL;
-    int ordinal;
-
-    if (PyArg_ParseTuple(args, "i:fromordinal", &ordinal)) {
-        int year;
-        int month;
-        int day;
-
-        if (ordinal < 1)
-            PyErr_SetString(PyExc_ValueError, "ordinal must be "
-                                              ">= 1");
-        else {
-            ord_to_ymd(ordinal, &year, &month, &day);
-            result = PyObject_CallFunction(cls, "iii",
-                                           year, month, day);
-        }
-    }
-    return result;
-}
-
-/*
- * Date arithmetic.
- */
-
-/* date + timedelta -> date.  If arg negate is true, subtract the timedelta
- * instead.
- */
-static PyObject *
-add_date_timedelta(PyDateTime_Date *date, PyDateTime_Delta *delta, int negate)
-{
-    PyObject *result = NULL;
-    int year = GET_YEAR(date);
-    int month = GET_MONTH(date);
-    int deltadays = GET_TD_DAYS(delta);
-    /* C-level overflow is impossible because |deltadays| < 1e9. */
-    int day = GET_DAY(date) + (negate ? -deltadays : deltadays);
-
-    if (normalize_date(&year, &month, &day) >= 0)
-        result = new_date(year, month, day);
-    return result;
-}
-
-static PyObject *
-date_add(PyObject *left, PyObject *right)
-{
-    if (PyDateTime_Check(left) || PyDateTime_Check(right)) {
-        Py_INCREF(Py_NotImplemented);
-        return Py_NotImplemented;
-    }
-    if (PyDate_Check(left)) {
-        /* date + ??? */
-        if (PyDelta_Check(right))
-            /* date + delta */
-            return add_date_timedelta((PyDateTime_Date *) left,
-                                      (PyDateTime_Delta *) right,
-                                      0);
-    }
-    else {
-        /* ??? + date
-         * 'right' must be one of us, or we wouldn't have been called
-         */
-        if (PyDelta_Check(left))
-            /* delta + date */
-            return add_date_timedelta((PyDateTime_Date *) right,
-                                      (PyDateTime_Delta *) left,
-                                      0);
-    }
-    Py_INCREF(Py_NotImplemented);
-    return Py_NotImplemented;
-}
-
-static PyObject *
-date_subtract(PyObject *left, PyObject *right)
-{
-    if (PyDateTime_Check(left) || PyDateTime_Check(right)) {
-        Py_INCREF(Py_NotImplemented);
-        return Py_NotImplemented;
-    }
-    if (PyDate_Check(left)) {
-        if (PyDate_Check(right)) {
-            /* date - date */
-            int left_ord = ymd_to_ord(GET_YEAR(left),
-                                      GET_MONTH(left),
-                                      GET_DAY(left));
-            int right_ord = ymd_to_ord(GET_YEAR(right),
-                                       GET_MONTH(right),
-                                       GET_DAY(right));
-            return new_delta(left_ord - right_ord, 0, 0, 0);
-        }
-        if (PyDelta_Check(right)) {
-            /* date - delta */
-            return add_date_timedelta((PyDateTime_Date *) left,
-                                      (PyDateTime_Delta *) right,
-                                      1);
-        }
-    }
-    Py_INCREF(Py_NotImplemented);
-    return Py_NotImplemented;
-}
-
-
-/* Various ways to turn a date into a string. */
-
-static PyObject *
-date_repr(PyDateTime_Date *self)
-{
-    return PyUnicode_FromFormat("%s(%d, %d, %d)",
-                                Py_TYPE(self)->tp_name,
-                                GET_YEAR(self), GET_MONTH(self), GET_DAY(self));
-}
-
-static PyObject *
-date_isoformat(PyDateTime_Date *self)
-{
-    return PyUnicode_FromFormat("%04d-%02d-%02d",
-                                GET_YEAR(self), GET_MONTH(self), GET_DAY(self));
-}
-
-/* str() calls the appropriate isoformat() method. */
-static PyObject *
-date_str(PyDateTime_Date *self)
-{
-    return PyObject_CallMethod((PyObject *)self, "isoformat", "()");
-}
-
-
-static PyObject *
-date_ctime(PyDateTime_Date *self)
-{
-    return format_ctime(self, 0, 0, 0);
-}
-
-static PyObject *
-date_strftime(PyDateTime_Date *self, PyObject *args, PyObject *kw)
-{
-    /* This method can be inherited, and needs to call the
-     * timetuple() method appropriate to self's class.
-     */
-    PyObject *result;
-    PyObject *tuple;
-    PyObject *format;
-    static char *keywords[] = {"format", NULL};
-
-    if (! PyArg_ParseTupleAndKeywords(args, kw, "U:strftime", keywords,
-                                      &format))
-        return NULL;
-
-    tuple = PyObject_CallMethod((PyObject *)self, "timetuple", "()");
-    if (tuple == NULL)
-        return NULL;
-    result = wrap_strftime((PyObject *)self, format, tuple,
-                           (PyObject *)self);
-    Py_DECREF(tuple);
-    return result;
-}
-
-static PyObject *
-date_format(PyDateTime_Date *self, PyObject *args)
-{
-    PyObject *format;
-
-    if (!PyArg_ParseTuple(args, "U:__format__", &format))
-        return NULL;
-
-    /* if the format is zero length, return str(self) */
-    if (PyUnicode_GetSize(format) == 0)
-        return PyObject_Str((PyObject *)self);
-
-    return PyObject_CallMethod((PyObject *)self, "strftime", "O", format);
-}
-
-/* ISO methods. */
-
-static PyObject *
-date_isoweekday(PyDateTime_Date *self)
-{
-    int dow = weekday(GET_YEAR(self), GET_MONTH(self), GET_DAY(self));
-
-    return PyLong_FromLong(dow + 1);
-}
-
-static PyObject *
-date_isocalendar(PyDateTime_Date *self)
-{
-    int  year         = GET_YEAR(self);
-    int  week1_monday = iso_week1_monday(year);
-    int today         = ymd_to_ord(year, GET_MONTH(self), GET_DAY(self));
-    int  week;
-    int  day;
-
-    week = divmod(today - week1_monday, 7, &day);
-    if (week < 0) {
-        --year;
-        week1_monday = iso_week1_monday(year);
-        week = divmod(today - week1_monday, 7, &day);
-    }
-    else if (week >= 52 && today >= iso_week1_monday(year + 1)) {
-        ++year;
-        week = 0;
-    }
-    return Py_BuildValue("iii", year, week + 1, day + 1);
-}
-
-/* Miscellaneous methods. */
-
-static PyObject *
-date_richcompare(PyObject *self, PyObject *other, int op)
-{
-    if (PyDate_Check(other)) {
-        int diff = memcmp(((PyDateTime_Date *)self)->data,
-                          ((PyDateTime_Date *)other)->data,
-                          _PyDateTime_DATE_DATASIZE);
-        return diff_to_bool(diff, op);
-    }
-    else {
-        Py_INCREF(Py_NotImplemented);
-        return Py_NotImplemented;
-    }
-}
-
-static PyObject *
-date_timetuple(PyDateTime_Date *self)
-{
-    return build_struct_time(GET_YEAR(self),
-                             GET_MONTH(self),
-                             GET_DAY(self),
-                             0, 0, 0, -1);
-}
-
-static PyObject *
-date_replace(PyDateTime_Date *self, PyObject *args, PyObject *kw)
-{
-    PyObject *clone;
-    PyObject *tuple;
-    int year = GET_YEAR(self);
-    int month = GET_MONTH(self);
-    int day = GET_DAY(self);
-
-    if (! PyArg_ParseTupleAndKeywords(args, kw, "|iii:replace", date_kws,
-                                      &year, &month, &day))
-        return NULL;
-    tuple = Py_BuildValue("iii", year, month, day);
-    if (tuple == NULL)
-        return NULL;
-    clone = date_new(Py_TYPE(self), tuple, NULL);
-    Py_DECREF(tuple);
-    return clone;
-}
-
-/*
-    Borrowed from stringobject.c, originally it was string_hash()
-*/
-static long
-generic_hash(unsigned char *data, int len)
-{
-    register unsigned char *p;
-    register long x;
-
-    p = (unsigned char *) data;
-    x = *p << 7;
-    while (--len >= 0)
-        x = (1000003*x) ^ *p++;
-    x ^= len;
-    if (x == -1)
-        x = -2;
-
-    return x;
-}
-
-
-static PyObject *date_getstate(PyDateTime_Date *self);
-
-static long
-date_hash(PyDateTime_Date *self)
-{
-    if (self->hashcode == -1)
-        self->hashcode = generic_hash(
-            (unsigned char *)self->data, _PyDateTime_DATE_DATASIZE);
-
-    return self->hashcode;
-}
-
-static PyObject *
-date_toordinal(PyDateTime_Date *self)
-{
-    return PyLong_FromLong(ymd_to_ord(GET_YEAR(self), GET_MONTH(self),
-                                     GET_DAY(self)));
-}
-
-static PyObject *
-date_weekday(PyDateTime_Date *self)
-{
-    int dow = weekday(GET_YEAR(self), GET_MONTH(self), GET_DAY(self));
-
-    return PyLong_FromLong(dow);
-}
-
-/* Pickle support, a simple use of __reduce__. */
-
-/* __getstate__ isn't exposed */
-static PyObject *
-date_getstate(PyDateTime_Date *self)
-{
-    PyObject* field;
-    field = PyBytes_FromStringAndSize((char*)self->data,
-                                       _PyDateTime_DATE_DATASIZE);
-    return Py_BuildValue("(N)", field);
-}
-
-static PyObject *
-date_reduce(PyDateTime_Date *self, PyObject *arg)
-{
-    return Py_BuildValue("(ON)", Py_TYPE(self), date_getstate(self));
-}
-
-static PyMethodDef date_methods[] = {
-
-    /* Class methods: */
-
-    {"fromtimestamp", (PyCFunction)date_fromtimestamp, METH_VARARGS |
-                                                       METH_CLASS,
-     PyDoc_STR("timestamp -> local date from a POSIX timestamp (like "
-               "time.time()).")},
-
-    {"fromordinal", (PyCFunction)date_fromordinal,      METH_VARARGS |
-                                                    METH_CLASS,
-     PyDoc_STR("int -> date corresponding to a proleptic Gregorian "
-               "ordinal.")},
-
-    {"today",         (PyCFunction)date_today,   METH_NOARGS | METH_CLASS,
-     PyDoc_STR("Current date or datetime:  same as "
-               "self.__class__.fromtimestamp(time.time()).")},
-
-    /* Instance methods: */
-
-    {"ctime",       (PyCFunction)date_ctime,        METH_NOARGS,
-     PyDoc_STR("Return ctime() style string.")},
-
-    {"strftime",        (PyCFunction)date_strftime,     METH_VARARGS | METH_KEYWORDS,
-     PyDoc_STR("format -> strftime() style string.")},
-
-    {"__format__",      (PyCFunction)date_format,       METH_VARARGS,
-     PyDoc_STR("Formats self with strftime.")},
-
-    {"timetuple",   (PyCFunction)date_timetuple,    METH_NOARGS,
-     PyDoc_STR("Return time tuple, compatible with time.localtime().")},
-
-    {"isocalendar", (PyCFunction)date_isocalendar,  METH_NOARGS,
-     PyDoc_STR("Return a 3-tuple containing ISO year, week number, and "
-               "weekday.")},
-
-    {"isoformat",   (PyCFunction)date_isoformat,        METH_NOARGS,
-     PyDoc_STR("Return string in ISO 8601 format, YYYY-MM-DD.")},
-
-    {"isoweekday",  (PyCFunction)date_isoweekday,   METH_NOARGS,
-     PyDoc_STR("Return the day of the week represented by the date.\n"
-               "Monday == 1 ... Sunday == 7")},
-
-    {"toordinal",   (PyCFunction)date_toordinal,    METH_NOARGS,
-     PyDoc_STR("Return proleptic Gregorian ordinal.  January 1 of year "
-               "1 is day 1.")},
-
-    {"weekday",     (PyCFunction)date_weekday,      METH_NOARGS,
-     PyDoc_STR("Return the day of the week represented by the date.\n"
-               "Monday == 0 ... Sunday == 6")},
-
-    {"replace",     (PyCFunction)date_replace,      METH_VARARGS | METH_KEYWORDS,
-     PyDoc_STR("Return date with new specified fields.")},
-
-    {"__reduce__", (PyCFunction)date_reduce,        METH_NOARGS,
-     PyDoc_STR("__reduce__() -> (cls, state)")},
-
-    {NULL,      NULL}
-};
-
-static char date_doc[] =
-PyDoc_STR("date(year, month, day) --> date object");
-
-static PyNumberMethods date_as_number = {
-    date_add,                                           /* nb_add */
-    date_subtract,                                      /* nb_subtract */
-    0,                                                  /* nb_multiply */
-    0,                                                  /* nb_remainder */
-    0,                                                  /* nb_divmod */
-    0,                                                  /* nb_power */
-    0,                                                  /* nb_negative */
-    0,                                                  /* nb_positive */
-    0,                                                  /* nb_absolute */
-    0,                                                  /* nb_bool */
-};
-
-static PyTypeObject PyDateTime_DateType = {
-    PyVarObject_HEAD_INIT(NULL, 0)
-    "datetime.date",                                    /* tp_name */
-    sizeof(PyDateTime_Date),                            /* tp_basicsize */
-    0,                                                  /* tp_itemsize */
-    0,                                                  /* tp_dealloc */
-    0,                                                  /* tp_print */
-    0,                                                  /* tp_getattr */
-    0,                                                  /* tp_setattr */
-    0,                                                  /* tp_reserved */
-    (reprfunc)date_repr,                                /* tp_repr */
-    &date_as_number,                                    /* tp_as_number */
-    0,                                                  /* tp_as_sequence */
-    0,                                                  /* tp_as_mapping */
-    (hashfunc)date_hash,                                /* tp_hash */
-    0,                                                  /* tp_call */
-    (reprfunc)date_str,                                 /* tp_str */
-    PyObject_GenericGetAttr,                            /* tp_getattro */
-    0,                                                  /* tp_setattro */
-    0,                                                  /* tp_as_buffer */
-    Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE,           /* tp_flags */
-    date_doc,                                           /* tp_doc */
-    0,                                                  /* tp_traverse */
-    0,                                                  /* tp_clear */
-    date_richcompare,                                   /* tp_richcompare */
-    0,                                                  /* tp_weaklistoffset */
-    0,                                                  /* tp_iter */
-    0,                                                  /* tp_iternext */
-    date_methods,                                       /* tp_methods */
-    0,                                                  /* tp_members */
-    date_getset,                                        /* tp_getset */
-    0,                                                  /* tp_base */
-    0,                                                  /* tp_dict */
-    0,                                                  /* tp_descr_get */
-    0,                                                  /* tp_descr_set */
-    0,                                                  /* tp_dictoffset */
-    0,                                                  /* tp_init */
-    0,                                                  /* tp_alloc */
-    date_new,                                           /* tp_new */
-    0,                                                  /* tp_free */
-};
-
-/*
- * PyDateTime_TZInfo implementation.
- */
-
-/* This is a pure abstract base class, so doesn't do anything beyond
- * raising NotImplemented exceptions.  Real tzinfo classes need
- * to derive from this.  This is mostly for clarity, and for efficiency in
- * datetime and time constructors (their tzinfo arguments need to
- * be subclasses of this tzinfo class, which is easy and quick to check).
- *
- * Note:  For reasons having to do with pickling of subclasses, we have
- * to allow tzinfo objects to be instantiated.  This wasn't an issue
- * in the Python implementation (__init__() could raise NotImplementedError
- * there without ill effect), but doing so in the C implementation hit a
- * brick wall.
- */
-
-static PyObject *
-tzinfo_nogo(const char* methodname)
-{
-    PyErr_Format(PyExc_NotImplementedError,
-                 "a tzinfo subclass must implement %s()",
-                 methodname);
-    return NULL;
-}
-
-/* Methods.  A subclass must implement these. */
-
-static PyObject *
-tzinfo_tzname(PyDateTime_TZInfo *self, PyObject *dt)
-{
-    return tzinfo_nogo("tzname");
-}
-
-static PyObject *
-tzinfo_utcoffset(PyDateTime_TZInfo *self, PyObject *dt)
-{
-    return tzinfo_nogo("utcoffset");
-}
-
-static PyObject *
-tzinfo_dst(PyDateTime_TZInfo *self, PyObject *dt)
-{
-    return tzinfo_nogo("dst");
-}
-
-
-static PyObject *add_datetime_timedelta(PyDateTime_DateTime *date,
-                                        PyDateTime_Delta *delta,
-                                        int factor);
-static PyObject *datetime_utcoffset(PyObject *self, PyObject *);
-static PyObject *datetime_dst(PyObject *self, PyObject *);
-
-static PyObject *
-tzinfo_fromutc(PyDateTime_TZInfo *self, PyObject *dt)
-{
-    PyObject *result = NULL;
-    PyObject *off = NULL, *dst = NULL;
-    PyDateTime_Delta *delta = NULL;
-
-    if (!PyDateTime_Check(dt)) {
-        PyErr_SetString(PyExc_TypeError,
-                        "fromutc: argument must be a datetime");
-        return NULL;
-    }
-    if (GET_DT_TZINFO(dt) != (PyObject *)self) {
-        PyErr_SetString(PyExc_ValueError, "fromutc: dt.tzinfo "
-                        "is not self");
-        return NULL;
-    }
-
-    off = datetime_utcoffset(dt, NULL);
-    if (off == NULL)
-        return NULL;
-    if (off == Py_None) {
-        PyErr_SetString(PyExc_ValueError, "fromutc: non-None "
-                        "utcoffset() result required");
-        goto Fail;
-    }
-
-    dst = datetime_dst(dt, NULL);
-    if (dst == NULL)
-        goto Fail;
-    if (dst == Py_None) {
-        PyErr_SetString(PyExc_ValueError, "fromutc: non-None "
-                        "dst() result required");
-        goto Fail;
-    }
-
-    delta = (PyDateTime_Delta *)delta_subtract(off, dst);
-    if (delta == NULL)
-        goto Fail;
-    result = add_datetime_timedelta((PyDateTime_DateTime *)dt, delta, 1);
-    if (result == NULL)
-        goto Fail;
-
-    Py_DECREF(dst);
-    dst = call_dst(GET_DT_TZINFO(dt), result);
-    if (dst == NULL)
-        goto Fail;
-    if (dst == Py_None)
-        goto Inconsistent;
-    if (delta_bool(delta) != 0) {
-        PyObject *temp = result;
-        result = add_datetime_timedelta((PyDateTime_DateTime *)result,
-                                        (PyDateTime_Delta *)dst, 1);
-        Py_DECREF(temp);
-        if (result == NULL)
-            goto Fail;
-    }
-    Py_DECREF(delta);
-    Py_DECREF(dst);
-    Py_DECREF(off);
-    return result;
-
-Inconsistent:
-    PyErr_SetString(PyExc_ValueError, "fromutc: tz.dst() gave"
-                    "inconsistent results; cannot convert");
-
-    /* fall thru to failure */
-Fail:
-    Py_XDECREF(off);
-    Py_XDECREF(dst);
-    Py_XDECREF(delta);
-    Py_XDECREF(result);
-    return NULL;
-}
-
-/*
- * Pickle support.  This is solely so that tzinfo subclasses can use
- * pickling -- tzinfo itself is supposed to be uninstantiable.
- */
-
-static PyObject *
-tzinfo_reduce(PyObject *self)
-{
-    PyObject *args, *state, *tmp;
-    PyObject *getinitargs, *getstate;
-
-    tmp = PyTuple_New(0);
-    if (tmp == NULL)
-        return NULL;
-
-    getinitargs = PyObject_GetAttrString(self, "__getinitargs__");
-    if (getinitargs != NULL) {
-        args = PyObject_CallObject(getinitargs, tmp);
-        Py_DECREF(getinitargs);
-        if (args == NULL) {
-            Py_DECREF(tmp);
-            return NULL;
-        }
-    }
-    else {
-        PyErr_Clear();
-        args = tmp;
-        Py_INCREF(args);
-    }
-
-    getstate = PyObject_GetAttrString(self, "__getstate__");
-    if (getstate != NULL) {
-        state = PyObject_CallObject(getstate, tmp);
-        Py_DECREF(getstate);
-        if (state == NULL) {
-            Py_DECREF(args);
-            Py_DECREF(tmp);
-            return NULL;
-        }
-    }
-    else {
-        PyObject **dictptr;
-        PyErr_Clear();
-        state = Py_None;
-        dictptr = _PyObject_GetDictPtr(self);
-        if (dictptr && *dictptr && PyDict_Size(*dictptr))
-            state = *dictptr;
-        Py_INCREF(state);
-    }
-
-    Py_DECREF(tmp);
-
-    if (state == Py_None) {
-        Py_DECREF(state);
-        return Py_BuildValue("(ON)", Py_TYPE(self), args);
-    }
-    else
-        return Py_BuildValue("(ONN)", Py_TYPE(self), args, state);
-}
-
-static PyMethodDef tzinfo_methods[] = {
-
-    {"tzname",          (PyCFunction)tzinfo_tzname,             METH_O,
-     PyDoc_STR("datetime -> string name of time zone.")},
-
-    {"utcoffset",       (PyCFunction)tzinfo_utcoffset,          METH_O,
-     PyDoc_STR("datetime -> timedelta showing offset from UTC, negative "
-           "values indicating West of UTC")},
-
-    {"dst",             (PyCFunction)tzinfo_dst,                METH_O,
-     PyDoc_STR("datetime -> DST offset in minutes east of UTC.")},
-
-    {"fromutc",         (PyCFunction)tzinfo_fromutc,            METH_O,
-     PyDoc_STR("datetime in UTC -> datetime in local time.")},
-
-    {"__reduce__",  (PyCFunction)tzinfo_reduce,             METH_NOARGS,
-     PyDoc_STR("-> (cls, state)")},
-
-    {NULL, NULL}
-};
-
-static char tzinfo_doc[] =
-PyDoc_STR("Abstract base class for time zone info objects.");
-
-static PyTypeObject PyDateTime_TZInfoType = {
-    PyVarObject_HEAD_INIT(NULL, 0)
-    "datetime.tzinfo",                          /* tp_name */
-    sizeof(PyDateTime_TZInfo),                  /* tp_basicsize */
-    0,                                          /* tp_itemsize */
-    0,                                          /* tp_dealloc */
-    0,                                          /* tp_print */
-    0,                                          /* tp_getattr */
-    0,                                          /* tp_setattr */
-    0,                                          /* tp_reserved */
-    0,                                          /* tp_repr */
-    0,                                          /* tp_as_number */
-    0,                                          /* tp_as_sequence */
-    0,                                          /* tp_as_mapping */
-    0,                                          /* tp_hash */
-    0,                                          /* tp_call */
-    0,                                          /* tp_str */
-    PyObject_GenericGetAttr,                    /* tp_getattro */
-    0,                                          /* tp_setattro */
-    0,                                          /* tp_as_buffer */
-    Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE,   /* tp_flags */
-    tzinfo_doc,                                 /* tp_doc */
-    0,                                          /* tp_traverse */
-    0,                                          /* tp_clear */
-    0,                                          /* tp_richcompare */
-    0,                                          /* tp_weaklistoffset */
-    0,                                          /* tp_iter */
-    0,                                          /* tp_iternext */
-    tzinfo_methods,                             /* tp_methods */
-    0,                                          /* tp_members */
-    0,                                          /* tp_getset */
-    0,                                          /* tp_base */
-    0,                                          /* tp_dict */
-    0,                                          /* tp_descr_get */
-    0,                                          /* tp_descr_set */
-    0,                                          /* tp_dictoffset */
-    0,                                          /* tp_init */
-    0,                                          /* tp_alloc */
-    PyType_GenericNew,                          /* tp_new */
-    0,                                          /* tp_free */
-};
-
-static char *timezone_kws[] = {"offset", "name", NULL};
-
-static PyObject *
-timezone_new(PyTypeObject *type, PyObject *args, PyObject *kw)
-{
-    PyObject *offset;
-    PyObject *name = NULL;
-    if (PyArg_ParseTupleAndKeywords(args, kw, "O!|O!:timezone", timezone_kws,
-                                    &PyDateTime_DeltaType, &offset,
-                                    &PyUnicode_Type, &name))
-        return new_timezone(offset, name);
-
-    return NULL;
-}
-
-static void
-timezone_dealloc(PyDateTime_TimeZone *self)
-{
-    Py_CLEAR(self->offset);
-    Py_CLEAR(self->name);
-    Py_TYPE(self)->tp_free((PyObject *)self);
-}
-
-static PyObject *
-timezone_richcompare(PyDateTime_TimeZone *self,
-                     PyDateTime_TimeZone *other, int op)
-{
-    if (op != Py_EQ && op != Py_NE) {
-        Py_INCREF(Py_NotImplemented);
-        return Py_NotImplemented;
-    }
-    return delta_richcompare(self->offset, other->offset, op);
-}
-
-static long
-timezone_hash(PyDateTime_TimeZone *self)
-{
-    return delta_hash((PyDateTime_Delta *)self->offset);
-}
-
-/* Check argument type passed to tzname, utcoffset, or dst methods.
-   Returns 0 for good argument.  Returns -1 and sets exception info
-   otherwise.
- */
-static int
-_timezone_check_argument(PyObject *dt, const char *meth)
-{
-    if (dt == Py_None || PyDateTime_Check(dt))
-        return 0;
-    PyErr_Format(PyExc_TypeError, "%s(dt) argument must be a datetime instance"
-                 " or None, not %.200s", meth, Py_TYPE(dt)->tp_name);
-    return -1;
-}
-
-static PyObject *
-timezone_repr(PyDateTime_TimeZone *self)
-{
-    /* Note that although timezone is not subclassable, it is convenient
-       to use Py_TYPE(self)->tp_name here. */
-    const char *type_name = Py_TYPE(self)->tp_name;
-
-    if (((PyObject *)self) == PyDateTime_TimeZone_UTC)
-        return PyUnicode_FromFormat("%s.utc", type_name);
-
-    if (self->name == NULL)
-        return PyUnicode_FromFormat("%s(%R)", type_name, self->offset);
-
-    return PyUnicode_FromFormat("%s(%R, %R)", type_name, self->offset,
-                                self->name);
-}
-
-
-static PyObject *
-timezone_str(PyDateTime_TimeZone *self)
-{
-    char buf[10];
-    int hours, minutes, seconds;
-    PyObject *offset;
-    char sign;
-
-    if (self->name != NULL) {
-        Py_INCREF(self->name);
-        return self->name;
-    }
-    /* Offset is normalized, so it is negative if days < 0 */
-    if (GET_TD_DAYS(self->offset) < 0) {
-        sign = '-';
-        offset = delta_negative((PyDateTime_Delta *)self->offset);
-        if (offset == NULL)
-            return NULL;
-    }
-    else {
-        sign = '+';
-        offset = self->offset;
-        Py_INCREF(offset);
-    }
-    /* Offset is not negative here. */
-    seconds = GET_TD_SECONDS(offset);
-    Py_DECREF(offset);
-    minutes = divmod(seconds, 60, &seconds);
-    hours = divmod(minutes, 60, &minutes);
-    assert(seconds == 0);
-    /* XXX ignore sub-minute data, curently not allowed. */
-    PyOS_snprintf(buf, sizeof(buf), "UTC%c%02d:%02d", sign, hours, minutes);
-
-    return PyUnicode_FromString(buf);
-}
-
-static PyObject *
-timezone_tzname(PyDateTime_TimeZone *self, PyObject *dt)
-{
-    if (_timezone_check_argument(dt, "tzname") == -1)
-        return NULL;
-
-    return timezone_str(self);
-}
-
-static PyObject *
-timezone_utcoffset(PyDateTime_TimeZone *self, PyObject *dt)
-{
-    if (_timezone_check_argument(dt, "utcoffset") == -1)
-        return NULL;
-
-    Py_INCREF(self->offset);
-    return self->offset;
-}
-
-static PyObject *
-timezone_dst(PyObject *self, PyObject *dt)
-{
-    if (_timezone_check_argument(dt, "dst") == -1)
-        return NULL;
-
-    Py_RETURN_NONE;
-}
-
-static PyObject *
-timezone_fromutc(PyDateTime_TimeZone *self, PyDateTime_DateTime *dt)
-{
-    if (!PyDateTime_Check(dt)) {
-        PyErr_SetString(PyExc_TypeError,
-                        "fromutc: argument must be a datetime");
-        return NULL;
-    }
-    if (!HASTZINFO(dt) || dt->tzinfo != (PyObject *)self) {
-        PyErr_SetString(PyExc_ValueError, "fromutc: dt.tzinfo "
-                        "is not self");
-        return NULL;
-    }
-
-    return add_datetime_timedelta(dt, (PyDateTime_Delta *)self->offset, 1);
-}
-
-static PyObject *
-timezone_getinitargs(PyDateTime_TimeZone *self)
-{
-    if (self->name == NULL)
-        return Py_BuildValue("(O)", self->offset);
-    return Py_BuildValue("(OO)", self->offset, self->name);
-}
-
-static PyMethodDef timezone_methods[] = {
-    {"tzname", (PyCFunction)timezone_tzname, METH_O,
-     PyDoc_STR("If name is specified when timezone is created, returns the name."
-               "  Otherwise returns offset as 'UTC(+|-)HH:MM'.")},
-
-    {"utcoffset", (PyCFunction)timezone_utcoffset, METH_O,
-     PyDoc_STR("Return fixed offset.")},
-
-    {"dst", (PyCFunction)timezone_dst, METH_O,
-     PyDoc_STR("Return None.")},
-
-    {"fromutc", (PyCFunction)timezone_fromutc, METH_O,
-     PyDoc_STR("datetime in UTC -> datetime in local time.")},
-
-    {"__getinitargs__", (PyCFunction)timezone_getinitargs, METH_NOARGS,
-     PyDoc_STR("pickle support")},
-
-    {NULL, NULL}
-};
-
-static char timezone_doc[] =
-PyDoc_STR("Fixed offset from UTC implementation of tzinfo.");
-
-static PyTypeObject PyDateTime_TimeZoneType = {
-    PyVarObject_HEAD_INIT(NULL, 0)
-    "datetime.timezone",              /* tp_name */
-    sizeof(PyDateTime_TimeZone),      /* tp_basicsize */
-    0,                                /* tp_itemsize */
-    (destructor)timezone_dealloc,     /* tp_dealloc */
-    0,                                /* tp_print */
-    0,                                /* tp_getattr */
-    0,                                /* tp_setattr */
-    0,                                /* tp_reserved */
-    (reprfunc)timezone_repr,          /* tp_repr */
-    0,                                /* tp_as_number */
-    0,                                /* tp_as_sequence */
-    0,                                /* tp_as_mapping */
-    (hashfunc)timezone_hash,          /* tp_hash */
-    0,                                /* tp_call */
-    (reprfunc)timezone_str,           /* tp_str */
-    0,                                /* tp_getattro */
-    0,                                /* tp_setattro */
-    0,                                /* tp_as_buffer */
-    Py_TPFLAGS_DEFAULT,               /* tp_flags */
-    timezone_doc,                     /* tp_doc */
-    0,                                /* tp_traverse */
-    0,                                /* tp_clear */
-    (richcmpfunc)timezone_richcompare,/* tp_richcompare */
-    0,                                /* tp_weaklistoffset */
-    0,                                /* tp_iter */
-    0,                                /* tp_iternext */
-    timezone_methods,                 /* tp_methods */
-    0,                                /* tp_members */
-    0,                                /* tp_getset */
-    &PyDateTime_TZInfoType,           /* tp_base */
-    0,                                /* tp_dict */
-    0,                                /* tp_descr_get */
-    0,                                /* tp_descr_set */
-    0,                                /* tp_dictoffset */
-    0,                                /* tp_init */
-    0,                                /* tp_alloc */
-    timezone_new,                     /* tp_new */
-};
-
-/*
- * PyDateTime_Time implementation.
- */
-
-/* Accessor properties.
- */
-
-static PyObject *
-time_hour(PyDateTime_Time *self, void *unused)
-{
-    return PyLong_FromLong(TIME_GET_HOUR(self));
-}
-
-static PyObject *
-time_minute(PyDateTime_Time *self, void *unused)
-{
-    return PyLong_FromLong(TIME_GET_MINUTE(self));
-}
-
-/* The name time_second conflicted with some platform header file. */
-static PyObject *
-py_time_second(PyDateTime_Time *self, void *unused)
-{
-    return PyLong_FromLong(TIME_GET_SECOND(self));
-}
-
-static PyObject *
-time_microsecond(PyDateTime_Time *self, void *unused)
-{
-    return PyLong_FromLong(TIME_GET_MICROSECOND(self));
-}
-
-static PyObject *
-time_tzinfo(PyDateTime_Time *self, void *unused)
-{
-    PyObject *result = HASTZINFO(self) ? self->tzinfo : Py_None;
-    Py_INCREF(result);
-    return result;
-}
-
-static PyGetSetDef time_getset[] = {
-    {"hour",        (getter)time_hour},
-    {"minute",      (getter)time_minute},
-    {"second",      (getter)py_time_second},
-    {"microsecond", (getter)time_microsecond},
-    {"tzinfo",          (getter)time_tzinfo},
-    {NULL}
-};
-
-/*
- * Constructors.
- */
-
-static char *time_kws[] = {"hour", "minute", "second", "microsecond",
-                           "tzinfo", NULL};
-
-static PyObject *
-time_new(PyTypeObject *type, PyObject *args, PyObject *kw)
-{
-    PyObject *self = NULL;
-    PyObject *state;
-    int hour = 0;
-    int minute = 0;
-    int second = 0;
-    int usecond = 0;
-    PyObject *tzinfo = Py_None;
-
-    /* Check for invocation from pickle with __getstate__ state */
-    if (PyTuple_GET_SIZE(args) >= 1 &&
-        PyTuple_GET_SIZE(args) <= 2 &&
-        PyBytes_Check(state = PyTuple_GET_ITEM(args, 0)) &&
-        PyBytes_GET_SIZE(state) == _PyDateTime_TIME_DATASIZE &&
-        ((unsigned char) (PyBytes_AS_STRING(state)[0])) < 24)
-    {
-        PyDateTime_Time *me;
-        char aware;
-
-        if (PyTuple_GET_SIZE(args) == 2) {
-            tzinfo = PyTuple_GET_ITEM(args, 1);
-            if (check_tzinfo_subclass(tzinfo) < 0) {
-                PyErr_SetString(PyExc_TypeError, "bad "
-                    "tzinfo state arg");
-                return NULL;
-            }
-        }
-        aware = (char)(tzinfo != Py_None);
-        me = (PyDateTime_Time *) (type->tp_alloc(type, aware));
-        if (me != NULL) {
-            char *pdata = PyBytes_AS_STRING(state);
-
-            memcpy(me->data, pdata, _PyDateTime_TIME_DATASIZE);
-            me->hashcode = -1;
-            me->hastzinfo = aware;
-            if (aware) {
-                Py_INCREF(tzinfo);
-                me->tzinfo = tzinfo;
-            }
-        }
-        return (PyObject *)me;
-    }
-
-    if (PyArg_ParseTupleAndKeywords(args, kw, "|iiiiO", time_kws,
-                                    &hour, &minute, &second, &usecond,
-                                    &tzinfo)) {
-        if (check_time_args(hour, minute, second, usecond) < 0)
-            return NULL;
-        if (check_tzinfo_subclass(tzinfo) < 0)
-            return NULL;
-        self = new_time_ex(hour, minute, second, usecond, tzinfo,
-                           type);
-    }
-    return self;
-}
-
-/*
- * Destructor.
- */
-
-static void
-time_dealloc(PyDateTime_Time *self)
-{
-    if (HASTZINFO(self)) {
-        Py_XDECREF(self->tzinfo);
-    }
-    Py_TYPE(self)->tp_free((PyObject *)self);
-}
-
-/*
- * Indirect access to tzinfo methods.
- */
-
-/* These are all METH_NOARGS, so don't need to check the arglist. */
-static PyObject *
-time_utcoffset(PyObject *self, PyObject *unused) {
-    return call_utcoffset(GET_TIME_TZINFO(self), Py_None);
-}
-
-static PyObject *
-time_dst(PyObject *self, PyObject *unused) {
-    return call_dst(GET_TIME_TZINFO(self), Py_None);
-}
-
-static PyObject *
-time_tzname(PyDateTime_Time *self, PyObject *unused) {
-    return call_tzname(GET_TIME_TZINFO(self), Py_None);
-}
-
-/*
- * Various ways to turn a time into a string.
- */
-
-static PyObject *
-time_repr(PyDateTime_Time *self)
-{
-    const char *type_name = Py_TYPE(self)->tp_name;
-    int h = TIME_GET_HOUR(self);
-    int m = TIME_GET_MINUTE(self);
-    int s = TIME_GET_SECOND(self);
-    int us = TIME_GET_MICROSECOND(self);
-    PyObject *result = NULL;
-
-    if (us)
-        result = PyUnicode_FromFormat("%s(%d, %d, %d, %d)",
-                                      type_name, h, m, s, us);
-    else if (s)
-        result = PyUnicode_FromFormat("%s(%d, %d, %d)",
-                                      type_name, h, m, s);
-    else
-        result = PyUnicode_FromFormat("%s(%d, %d)", type_name, h, m);
-    if (result != NULL && HASTZINFO(self))
-        result = append_keyword_tzinfo(result, self->tzinfo);
-    return result;
-}
-
-static PyObject *
-time_str(PyDateTime_Time *self)
-{
-    return PyObject_CallMethod((PyObject *)self, "isoformat", "()");
-}
-
-static PyObject *
-time_isoformat(PyDateTime_Time *self, PyObject *unused)
-{
-    char buf[100];
-    PyObject *result;
-    int us = TIME_GET_MICROSECOND(self);;
-
-    if (us)
-        result = PyUnicode_FromFormat("%02d:%02d:%02d.%06d",
-                                      TIME_GET_HOUR(self),
-                                      TIME_GET_MINUTE(self),
-                                      TIME_GET_SECOND(self),
-                                      us);
-    else
-        result = PyUnicode_FromFormat("%02d:%02d:%02d",
-                                      TIME_GET_HOUR(self),
-                                      TIME_GET_MINUTE(self),
-                                      TIME_GET_SECOND(self));
-
-    if (result == NULL || !HASTZINFO(self) || self->tzinfo == Py_None)
-        return result;
-
-    /* We need to append the UTC offset. */
-    if (format_utcoffset(buf, sizeof(buf), ":", self->tzinfo,
-                         Py_None) < 0) {
-        Py_DECREF(result);
-        return NULL;
-    }
-    PyUnicode_AppendAndDel(&result, PyUnicode_FromString(buf));
-    return result;
-}
-
-static PyObject *
-time_strftime(PyDateTime_Time *self, PyObject *args, PyObject *kw)
-{
-    PyObject *result;
-    PyObject *tuple;
-    PyObject *format;
-    static char *keywords[] = {"format", NULL};
-
-    if (! PyArg_ParseTupleAndKeywords(args, kw, "U:strftime", keywords,
-                                      &format))
-        return NULL;
-
-    /* Python's strftime does insane things with the year part of the
-     * timetuple.  The year is forced to (the otherwise nonsensical)
-     * 1900 to worm around that.
-     */
-    tuple = Py_BuildValue("iiiiiiiii",
-                          1900, 1, 1, /* year, month, day */
-                  TIME_GET_HOUR(self),
-                  TIME_GET_MINUTE(self),
-                  TIME_GET_SECOND(self),
-                  0, 1, -1); /* weekday, daynum, dst */
-    if (tuple == NULL)
-        return NULL;
-    assert(PyTuple_Size(tuple) == 9);
-    result = wrap_strftime((PyObject *)self, format, tuple,
-                           Py_None);
-    Py_DECREF(tuple);
-    return result;
-}
-
-/*
- * Miscellaneous methods.
- */
-
-static PyObject *
-time_richcompare(PyObject *self, PyObject *other, int op)
-{
-    PyObject *result = NULL;
-    PyObject *offset1, *offset2;
-    int diff;
-
-    if (! PyTime_Check(other)) {
-        Py_INCREF(Py_NotImplemented);
-        return Py_NotImplemented;
-    }
-
-    if (GET_TIME_TZINFO(self) == GET_TIME_TZINFO(other)) {
-        diff = memcmp(((PyDateTime_Time *)self)->data,
-                      ((PyDateTime_Time *)other)->data,
-                      _PyDateTime_TIME_DATASIZE);
-        return diff_to_bool(diff, op);
-    }
-    offset1 = time_utcoffset(self, NULL);
-    if (offset1 == NULL)
-        return NULL;
-    offset2 = time_utcoffset(other, NULL);
-    if (offset2 == NULL)
-        goto done;
-    /* If they're both naive, or both aware and have the same offsets,
-     * we get off cheap.  Note that if they're both naive, offset1 ==
-     * offset2 == Py_None at this point.
-     */
-    if ((offset1 == offset2) ||
-        (PyDelta_Check(offset1) && PyDelta_Check(offset2) &&
-         delta_cmp(offset1, offset2) == 0)) {
-        diff = memcmp(((PyDateTime_Time *)self)->data,
-                      ((PyDateTime_Time *)other)->data,
-                      _PyDateTime_TIME_DATASIZE);
-        result = diff_to_bool(diff, op);
-    }
-    /* The hard case: both aware with different UTC offsets */
-    else if (offset1 != Py_None && offset2 != Py_None) {
-        int offsecs1, offsecs2;
-        assert(offset1 != offset2); /* else last "if" handled it */
-        offsecs1 = TIME_GET_HOUR(self) * 3600 +
-                   TIME_GET_MINUTE(self) * 60 +
-                   TIME_GET_SECOND(self) -
-                   GET_TD_DAYS(offset1) * 86400 -
-                   GET_TD_SECONDS(offset1);
-        offsecs2 = TIME_GET_HOUR(other) * 3600 +
-                   TIME_GET_MINUTE(other) * 60 +
-                   TIME_GET_SECOND(other) -
-                   GET_TD_DAYS(offset2) * 86400 -
-                   GET_TD_SECONDS(offset2);
-        diff = offsecs1 - offsecs2;
-        if (diff == 0)
-            diff = TIME_GET_MICROSECOND(self) -
-                   TIME_GET_MICROSECOND(other);
-        result = diff_to_bool(diff, op);
-    }
-    else {
-        PyErr_SetString(PyExc_TypeError,
-                        "can't compare offset-naive and "
-                        "offset-aware times");
-    }
- done:
-    Py_DECREF(offset1);
-    Py_XDECREF(offset2);
-    return result;
-}
-
-static long
-time_hash(PyDateTime_Time *self)
-{
-    if (self->hashcode == -1) {
-        PyObject *offset;
-
-        offset = time_utcoffset((PyObject *)self, NULL);
-
-        if (offset == NULL)
-            return -1;
-
-        /* Reduce this to a hash of another object. */
-        if (offset == Py_None)
-            self->hashcode = generic_hash(
-                (unsigned char *)self->data, _PyDateTime_TIME_DATASIZE);
-        else {
-            PyObject *temp1, *temp2;
-            int seconds, microseconds;
-            assert(HASTZINFO(self));
-            seconds = TIME_GET_HOUR(self) * 3600 +
-                      TIME_GET_MINUTE(self) * 60 +
-                      TIME_GET_SECOND(self);
-            microseconds = TIME_GET_MICROSECOND(self);
-            temp1 = new_delta(0, seconds, microseconds, 1);
-            if (temp1 == NULL) {
-                Py_DECREF(offset);
-                return -1;
-            }
-            temp2 = delta_subtract(temp1, offset);
-            Py_DECREF(temp1);
-            if (temp2 == NULL) {
-                Py_DECREF(offset);
-                return -1;
-            }
-            self->hashcode = PyObject_Hash(temp2);
-            Py_DECREF(temp2);
-        }
-        Py_DECREF(offset);
-    }
-    return self->hashcode;
-}
-
-static PyObject *
-time_replace(PyDateTime_Time *self, PyObject *args, PyObject *kw)
-{
-    PyObject *clone;
-    PyObject *tuple;
-    int hh = TIME_GET_HOUR(self);
-    int mm = TIME_GET_MINUTE(self);
-    int ss = TIME_GET_SECOND(self);
-    int us = TIME_GET_MICROSECOND(self);
-    PyObject *tzinfo = HASTZINFO(self) ? self->tzinfo : Py_None;
-
-    if (! PyArg_ParseTupleAndKeywords(args, kw, "|iiiiO:replace",
-                                      time_kws,
-                                      &hh, &mm, &ss, &us, &tzinfo))
-        return NULL;
-    tuple = Py_BuildValue("iiiiO", hh, mm, ss, us, tzinfo);
-    if (tuple == NULL)
-        return NULL;
-    clone = time_new(Py_TYPE(self), tuple, NULL);
-    Py_DECREF(tuple);
-    return clone;
-}
-
-static int
-time_bool(PyObject *self)
-{
-    PyObject *offset, *tzinfo;
-    int offsecs = 0;
-
-    if (TIME_GET_SECOND(self) || TIME_GET_MICROSECOND(self)) {
-        /* Since utcoffset is in whole minutes, nothing can
-         * alter the conclusion that this is nonzero.
-         */
-        return 1;
-    }
-    tzinfo = GET_TIME_TZINFO(self);
-    if (tzinfo != Py_None) {
-        offset = call_utcoffset(tzinfo, Py_None);
-        if (offset == NULL)
-            return -1;
-        offsecs = GET_TD_DAYS(offset)*86400 + GET_TD_SECONDS(offset);
-        Py_DECREF(offset);
-    }
-    return (TIME_GET_MINUTE(self)*60 - offsecs + TIME_GET_HOUR(self)*3600) != 0;
-}
-
-/* Pickle support, a simple use of __reduce__. */
-
-/* Let basestate be the non-tzinfo data string.
- * If tzinfo is None, this returns (basestate,), else (basestate, tzinfo).
- * So it's a tuple in any (non-error) case.
- * __getstate__ isn't exposed.
- */
-static PyObject *
-time_getstate(PyDateTime_Time *self)
-{
-    PyObject *basestate;
-    PyObject *result = NULL;
-
-    basestate =  PyBytes_FromStringAndSize((char *)self->data,
-                                            _PyDateTime_TIME_DATASIZE);
-    if (basestate != NULL) {
-        if (! HASTZINFO(self) || self->tzinfo == Py_None)
-            result = PyTuple_Pack(1, basestate);
-        else
-            result = PyTuple_Pack(2, basestate, self->tzinfo);
-        Py_DECREF(basestate);
-    }
-    return result;
-}
-
-static PyObject *
-time_reduce(PyDateTime_Time *self, PyObject *arg)
-{
-    return Py_BuildValue("(ON)", Py_TYPE(self), time_getstate(self));
-}
-
-static PyMethodDef time_methods[] = {
-
-    {"isoformat",   (PyCFunction)time_isoformat,        METH_NOARGS,
-     PyDoc_STR("Return string in ISO 8601 format, HH:MM:SS[.mmmmmm]"
-               "[+HH:MM].")},
-
-    {"strftime",        (PyCFunction)time_strftime,     METH_VARARGS | METH_KEYWORDS,
-     PyDoc_STR("format -> strftime() style string.")},
-
-    {"__format__",      (PyCFunction)date_format,       METH_VARARGS,
-     PyDoc_STR("Formats self with strftime.")},
-
-    {"utcoffset",       (PyCFunction)time_utcoffset,    METH_NOARGS,
-     PyDoc_STR("Return self.tzinfo.utcoffset(self).")},
-
-    {"tzname",          (PyCFunction)time_tzname,       METH_NOARGS,
-     PyDoc_STR("Return self.tzinfo.tzname(self).")},
-
-    {"dst",             (PyCFunction)time_dst,          METH_NOARGS,
-     PyDoc_STR("Return self.tzinfo.dst(self).")},
-
-    {"replace",     (PyCFunction)time_replace,          METH_VARARGS | METH_KEYWORDS,
-     PyDoc_STR("Return time with new specified fields.")},
-
-    {"__reduce__", (PyCFunction)time_reduce,        METH_NOARGS,
-     PyDoc_STR("__reduce__() -> (cls, state)")},
-
-    {NULL,      NULL}
-};
-
-static char time_doc[] =
-PyDoc_STR("time([hour[, minute[, second[, microsecond[, tzinfo]]]]]) --> a time object\n\
-\n\
-All arguments are optional. tzinfo may be None, or an instance of\n\
-a tzinfo subclass. The remaining arguments may be ints or longs.\n");
-
-static PyNumberMethods time_as_number = {
-    0,                                          /* nb_add */
-    0,                                          /* nb_subtract */
-    0,                                          /* nb_multiply */
-    0,                                          /* nb_remainder */
-    0,                                          /* nb_divmod */
-    0,                                          /* nb_power */
-    0,                                          /* nb_negative */
-    0,                                          /* nb_positive */
-    0,                                          /* nb_absolute */
-    (inquiry)time_bool,                         /* nb_bool */
-};
-
-static PyTypeObject PyDateTime_TimeType = {
-    PyVarObject_HEAD_INIT(NULL, 0)
-    "datetime.time",                            /* tp_name */
-    sizeof(PyDateTime_Time),                    /* tp_basicsize */
-    0,                                          /* tp_itemsize */
-    (destructor)time_dealloc,                   /* tp_dealloc */
-    0,                                          /* tp_print */
-    0,                                          /* tp_getattr */
-    0,                                          /* tp_setattr */
-    0,                                          /* tp_reserved */
-    (reprfunc)time_repr,                        /* tp_repr */
-    &time_as_number,                            /* tp_as_number */
-    0,                                          /* tp_as_sequence */
-    0,                                          /* tp_as_mapping */
-    (hashfunc)time_hash,                        /* tp_hash */
-    0,                                          /* tp_call */
-    (reprfunc)time_str,                         /* tp_str */
-    PyObject_GenericGetAttr,                    /* tp_getattro */
-    0,                                          /* tp_setattro */
-    0,                                          /* tp_as_buffer */
-    Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE, /* tp_flags */
-    time_doc,                                   /* tp_doc */
-    0,                                          /* tp_traverse */
-    0,                                          /* tp_clear */
-    time_richcompare,                           /* tp_richcompare */
-    0,                                          /* tp_weaklistoffset */
-    0,                                          /* tp_iter */
-    0,                                          /* tp_iternext */
-    time_methods,                               /* tp_methods */
-    0,                                          /* tp_members */
-    time_getset,                                /* tp_getset */
-    0,                                          /* tp_base */
-    0,                                          /* tp_dict */
-    0,                                          /* tp_descr_get */
-    0,                                          /* tp_descr_set */
-    0,                                          /* tp_dictoffset */
-    0,                                          /* tp_init */
-    time_alloc,                                 /* tp_alloc */
-    time_new,                                   /* tp_new */
-    0,                                          /* tp_free */
-};
-
-/*
- * PyDateTime_DateTime implementation.
- */
-
-/* Accessor properties.  Properties for day, month, and year are inherited
- * from date.
- */
-
-static PyObject *
-datetime_hour(PyDateTime_DateTime *self, void *unused)
-{
-    return PyLong_FromLong(DATE_GET_HOUR(self));
-}
-
-static PyObject *
-datetime_minute(PyDateTime_DateTime *self, void *unused)
-{
-    return PyLong_FromLong(DATE_GET_MINUTE(self));
-}
-
-static PyObject *
-datetime_second(PyDateTime_DateTime *self, void *unused)
-{
-    return PyLong_FromLong(DATE_GET_SECOND(self));
-}
-
-static PyObject *
-datetime_microsecond(PyDateTime_DateTime *self, void *unused)
-{
-    return PyLong_FromLong(DATE_GET_MICROSECOND(self));
-}
-
-static PyObject *
-datetime_tzinfo(PyDateTime_DateTime *self, void *unused)
-{
-    PyObject *result = HASTZINFO(self) ? self->tzinfo : Py_None;
-    Py_INCREF(result);
-    return result;
-}
-
-static PyGetSetDef datetime_getset[] = {
-    {"hour",        (getter)datetime_hour},
-    {"minute",      (getter)datetime_minute},
-    {"second",      (getter)datetime_second},
-    {"microsecond", (getter)datetime_microsecond},
-    {"tzinfo",          (getter)datetime_tzinfo},
-    {NULL}
-};
-
-/*
- * Constructors.
- */
-
-static char *datetime_kws[] = {
-    "year", "month", "day", "hour", "minute", "second",
-    "microsecond", "tzinfo", NULL
-};
-
-static PyObject *
-datetime_new(PyTypeObject *type, PyObject *args, PyObject *kw)
-{
-    PyObject *self = NULL;
-    PyObject *state;
-    int year;
-    int month;
-    int day;
-    int hour = 0;
-    int minute = 0;
-    int second = 0;
-    int usecond = 0;
-    PyObject *tzinfo = Py_None;
-
-    /* Check for invocation from pickle with __getstate__ state */
-    if (PyTuple_GET_SIZE(args) >= 1 &&
-        PyTuple_GET_SIZE(args) <= 2 &&
-        PyBytes_Check(state = PyTuple_GET_ITEM(args, 0)) &&
-        PyBytes_GET_SIZE(state) == _PyDateTime_DATETIME_DATASIZE &&
-        MONTH_IS_SANE(PyBytes_AS_STRING(state)[2]))
-    {
-        PyDateTime_DateTime *me;
-        char aware;
-
-        if (PyTuple_GET_SIZE(args) == 2) {
-            tzinfo = PyTuple_GET_ITEM(args, 1);
-            if (check_tzinfo_subclass(tzinfo) < 0) {
-                PyErr_SetString(PyExc_TypeError, "bad "
-                    "tzinfo state arg");
-                return NULL;
-            }
-        }
-        aware = (char)(tzinfo != Py_None);
-        me = (PyDateTime_DateTime *) (type->tp_alloc(type , aware));
-        if (me != NULL) {
-            char *pdata = PyBytes_AS_STRING(state);
-
-            memcpy(me->data, pdata, _PyDateTime_DATETIME_DATASIZE);
-            me->hashcode = -1;
-            me->hastzinfo = aware;
-            if (aware) {
-                Py_INCREF(tzinfo);
-                me->tzinfo = tzinfo;
-            }
-        }
-        return (PyObject *)me;
-    }
-
-    if (PyArg_ParseTupleAndKeywords(args, kw, "iii|iiiiO", datetime_kws,
-                                    &year, &month, &day, &hour, &minute,
-                                    &second, &usecond, &tzinfo)) {
-        if (check_date_args(year, month, day) < 0)
-            return NULL;
-        if (check_time_args(hour, minute, second, usecond) < 0)
-            return NULL;
-        if (check_tzinfo_subclass(tzinfo) < 0)
-            return NULL;
-        self = new_datetime_ex(year, month, day,
-                                hour, minute, second, usecond,
-                                tzinfo, type);
-    }
-    return self;
-}
-
-/* TM_FUNC is the shared type of localtime() and gmtime(). */
-typedef struct tm *(*TM_FUNC)(const time_t *timer);
-
-/* Internal helper.
- * Build datetime from a time_t and a distinct count of microseconds.
- * Pass localtime or gmtime for f, to control the interpretation of timet.
- */
-static PyObject *
-datetime_from_timet_and_us(PyObject *cls, TM_FUNC f, time_t timet, int us,
-                           PyObject *tzinfo)
-{
-    struct tm *tm;
-    PyObject *result = NULL;
-
-    tm = f(&timet);
-    if (tm) {
-        /* The platform localtime/gmtime may insert leap seconds,
-         * indicated by tm->tm_sec > 59.  We don't care about them,
-         * except to the extent that passing them on to the datetime
-         * constructor would raise ValueError for a reason that
-         * made no sense to the user.
-         */
-        if (tm->tm_sec > 59)
-            tm->tm_sec = 59;
-        result = PyObject_CallFunction(cls, "iiiiiiiO",
-                                       tm->tm_year + 1900,
-                                       tm->tm_mon + 1,
-                                       tm->tm_mday,
-                                       tm->tm_hour,
-                                       tm->tm_min,
-                                       tm->tm_sec,
-                                       us,
-                                       tzinfo);
-    }
-    else
-        PyErr_SetString(PyExc_ValueError,
-                        "timestamp out of range for "
-                        "platform localtime()/gmtime() function");
-    return result;
-}
-
-/* Internal helper.
- * Build datetime from a Python timestamp.  Pass localtime or gmtime for f,
- * to control the interpretation of the timestamp.  Since a double doesn't
- * have enough bits to cover a datetime's full range of precision, it's
- * better to call datetime_from_timet_and_us provided you have a way
- * to get that much precision (e.g., C time() isn't good enough).
- */
-static PyObject *
-datetime_from_timestamp(PyObject *cls, TM_FUNC f, double timestamp,
-                        PyObject *tzinfo)
-{
-    time_t timet;
-    double fraction;
-    int us;
-
-    timet = _PyTime_DoubleToTimet(timestamp);
-    if (timet == (time_t)-1 && PyErr_Occurred())
-        return NULL;
-    fraction = timestamp - (double)timet;
-    us = (int)round_to_long(fraction * 1e6);
-    if (us < 0) {
-        /* Truncation towards zero is not what we wanted
-           for negative numbers (Python's mod semantics) */
-        timet -= 1;
-        us += 1000000;
-    }
-    /* If timestamp is less than one microsecond smaller than a
-     * full second, round up. Otherwise, ValueErrors are raised
-     * for some floats. */
-    if (us == 1000000) {
-        timet += 1;
-        us = 0;
-    }
-    return datetime_from_timet_and_us(cls, f, timet, us, tzinfo);
-}
-
-/* Internal helper.
- * Build most accurate possible datetime for current time.  Pass localtime or
- * gmtime for f as appropriate.
- */
-static PyObject *
-datetime_best_possible(PyObject *cls, TM_FUNC f, PyObject *tzinfo)
-{
-#ifdef HAVE_GETTIMEOFDAY
-    struct timeval t;
-
-#ifdef GETTIMEOFDAY_NO_TZ
-    gettimeofday(&t);
-#else
-    gettimeofday(&t, (struct timezone *)NULL);
-#endif
-    return datetime_from_timet_and_us(cls, f, t.tv_sec, (int)t.tv_usec,
-                                      tzinfo);
-
-#else   /* ! HAVE_GETTIMEOFDAY */
-    /* No flavor of gettimeofday exists on this platform.  Python's
-     * time.time() does a lot of other platform tricks to get the
-     * best time it can on the platform, and we're not going to do
-     * better than that (if we could, the better code would belong
-     * in time.time()!)  We're limited by the precision of a double,
-     * though.
-     */
-    PyObject *time;
-    double dtime;
-
-    time = time_time();
-    if (time == NULL)
-        return NULL;
-    dtime = PyFloat_AsDouble(time);
-    Py_DECREF(time);
-    if (dtime == -1.0 && PyErr_Occurred())
-        return NULL;
-    return datetime_from_timestamp(cls, f, dtime, tzinfo);
-#endif  /* ! HAVE_GETTIMEOFDAY */
-}
-
-/* Return best possible local time -- this isn't constrained by the
- * precision of a timestamp.
- */
-static PyObject *
-datetime_now(PyObject *cls, PyObject *args, PyObject *kw)
-{
-    PyObject *self;
-    PyObject *tzinfo = Py_None;
-    static char *keywords[] = {"tz", NULL};
-
-    if (! PyArg_ParseTupleAndKeywords(args, kw, "|O:now", keywords,
-                                      &tzinfo))
-        return NULL;
-    if (check_tzinfo_subclass(tzinfo) < 0)
-        return NULL;
-
-    self = datetime_best_possible(cls,
-                                  tzinfo == Py_None ? localtime : gmtime,
-                                  tzinfo);
-    if (self != NULL && tzinfo != Py_None) {
-        /* Convert UTC to tzinfo's zone. */
-        PyObject *temp = self;
-        self = PyObject_CallMethod(tzinfo, "fromutc", "O", self);
-        Py_DECREF(temp);
-    }
-    return self;
-}
-
-/* Return best possible UTC time -- this isn't constrained by the
- * precision of a timestamp.
- */
-static PyObject *
-datetime_utcnow(PyObject *cls, PyObject *dummy)
-{
-    return datetime_best_possible(cls, gmtime, Py_None);
-}
-
-/* Return new local datetime from timestamp (Python timestamp -- a double). */
-static PyObject *
-datetime_fromtimestamp(PyObject *cls, PyObject *args, PyObject *kw)
-{
-    PyObject *self;
-    double timestamp;
-    PyObject *tzinfo = Py_None;
-    static char *keywords[] = {"timestamp", "tz", NULL};
-
-    if (! PyArg_ParseTupleAndKeywords(args, kw, "d|O:fromtimestamp",
-                                      keywords, &timestamp, &tzinfo))
-        return NULL;
-    if (check_tzinfo_subclass(tzinfo) < 0)
-        return NULL;
-
-    self = datetime_from_timestamp(cls,
-                                   tzinfo == Py_None ? localtime : gmtime,
-                                   timestamp,
-                                   tzinfo);
-    if (self != NULL && tzinfo != Py_None) {
-        /* Convert UTC to tzinfo's zone. */
-        PyObject *temp = self;
-        self = PyObject_CallMethod(tzinfo, "fromutc", "O", self);
-        Py_DECREF(temp);
-    }
-    return self;
-}
-
-/* Return new UTC datetime from timestamp (Python timestamp -- a double). */
-static PyObject *
-datetime_utcfromtimestamp(PyObject *cls, PyObject *args)
-{
-    double timestamp;
-    PyObject *result = NULL;
-
-    if (PyArg_ParseTuple(args, "d:utcfromtimestamp", &timestamp))
-        result = datetime_from_timestamp(cls, gmtime, timestamp,
-                                         Py_None);
-    return result;
-}
-
-/* Return new datetime from _strptime.strptime_datetime(). */
-static PyObject *
-datetime_strptime(PyObject *cls, PyObject *args)
-{
-    static PyObject *module = NULL;
-    const Py_UNICODE *string, *format;
-
-    if (!PyArg_ParseTuple(args, "uu:strptime", &string, &format))
-        return NULL;
-
-    if (module == NULL) {
-        module = PyImport_ImportModuleNoBlock("_strptime");
-        if (module == NULL)
-            return NULL;
-    }
-    return PyObject_CallMethod(module, "_strptime_datetime", "Ouu",
-                               cls, string, format);
-}
-
-/* Return new datetime from date/datetime and time arguments. */
-static PyObject *
-datetime_combine(PyObject *cls, PyObject *args, PyObject *kw)
-{
-    static char *keywords[] = {"date", "time", NULL};
-    PyObject *date;
-    PyObject *time;
-    PyObject *result = NULL;
-
-    if (PyArg_ParseTupleAndKeywords(args, kw, "O!O!:combine", keywords,
-                                    &PyDateTime_DateType, &date,
-                                    &PyDateTime_TimeType, &time)) {
-        PyObject *tzinfo = Py_None;
-
-        if (HASTZINFO(time))
-            tzinfo = ((PyDateTime_Time *)time)->tzinfo;
-        result = PyObject_CallFunction(cls, "iiiiiiiO",
-                                        GET_YEAR(date),
-                                        GET_MONTH(date),
-                                        GET_DAY(date),
-                                        TIME_GET_HOUR(time),
-                                        TIME_GET_MINUTE(time),
-                                        TIME_GET_SECOND(time),
-                                        TIME_GET_MICROSECOND(time),
-                                        tzinfo);
-    }
-    return result;
-}
-
-/*
- * Destructor.
- */
-
-static void
-datetime_dealloc(PyDateTime_DateTime *self)
-{
-    if (HASTZINFO(self)) {
-        Py_XDECREF(self->tzinfo);
-    }
-    Py_TYPE(self)->tp_free((PyObject *)self);
-}
-
-/*
- * Indirect access to tzinfo methods.
- */
-
-/* These are all METH_NOARGS, so don't need to check the arglist. */
-static PyObject *
-datetime_utcoffset(PyObject *self, PyObject *unused) {
-    return call_utcoffset(GET_DT_TZINFO(self), self);
-}
-
-static PyObject *
-datetime_dst(PyObject *self, PyObject *unused) {
-    return call_dst(GET_DT_TZINFO(self), self);
-}
-
-static PyObject *
-datetime_tzname(PyObject *self, PyObject *unused) {
-    return call_tzname(GET_DT_TZINFO(self), self);
-}
-
-/*
- * datetime arithmetic.
- */
-
-/* factor must be 1 (to add) or -1 (to subtract).  The result inherits
- * the tzinfo state of date.
- */
-static PyObject *
-add_datetime_timedelta(PyDateTime_DateTime *date, PyDateTime_Delta *delta,
-                       int factor)
-{
-    /* Note that the C-level additions can't overflow, because of
-     * invariant bounds on the member values.
-     */
-    int year = GET_YEAR(date);
-    int month = GET_MONTH(date);
-    int day = GET_DAY(date) + GET_TD_DAYS(delta) * factor;
-    int hour = DATE_GET_HOUR(date);
-    int minute = DATE_GET_MINUTE(date);
-    int second = DATE_GET_SECOND(date) + GET_TD_SECONDS(delta) * factor;
-    int microsecond = DATE_GET_MICROSECOND(date) +
-                      GET_TD_MICROSECONDS(delta) * factor;
-
-    assert(factor == 1 || factor == -1);
-    if (normalize_datetime(&year, &month, &day,
-                           &hour, &minute, &second, &microsecond) < 0)
-        return NULL;
-    else
-        return new_datetime(year, month, day,
-                            hour, minute, second, microsecond,
-                            HASTZINFO(date) ? date->tzinfo : Py_None);
-}
-
-static PyObject *
-datetime_add(PyObject *left, PyObject *right)
-{
-    if (PyDateTime_Check(left)) {
-        /* datetime + ??? */
-        if (PyDelta_Check(right))
-            /* datetime + delta */
-            return add_datetime_timedelta(
-                            (PyDateTime_DateTime *)left,
-                            (PyDateTime_Delta *)right,
-                            1);
-    }
-    else if (PyDelta_Check(left)) {
-        /* delta + datetime */
-        return add_datetime_timedelta((PyDateTime_DateTime *) right,
-                                      (PyDateTime_Delta *) left,
-                                      1);
-    }
-    Py_INCREF(Py_NotImplemented);
-    return Py_NotImplemented;
-}
-
-static PyObject *
-datetime_subtract(PyObject *left, PyObject *right)
-{
-    PyObject *result = Py_NotImplemented;
-
-    if (PyDateTime_Check(left)) {
-        /* datetime - ??? */
-        if (PyDateTime_Check(right)) {
-            /* datetime - datetime */
-            PyObject *offset1, *offset2, *offdiff = NULL;
-            int delta_d, delta_s, delta_us;
-
-            if (GET_DT_TZINFO(left) == GET_DT_TZINFO(right)) {
-                offset2 = offset1 = Py_None;
-                Py_INCREF(offset1);
-                Py_INCREF(offset2);
-            }
-            else {
-                offset1 = datetime_utcoffset(left, NULL);
-                if (offset1 == NULL)
-                    return NULL;
-                offset2 = datetime_utcoffset(right, NULL);
-                if (offset2 == NULL) {
-                    Py_DECREF(offset1);
-                    return NULL;
-                }
-                if ((offset1 != Py_None) != (offset2 != Py_None)) {
-                    PyErr_SetString(PyExc_TypeError,
-                                    "can't subtract offset-naive and "
-                                    "offset-aware datetimes");
-                    Py_DECREF(offset1);
-                    Py_DECREF(offset2);
-                    return NULL;
-                }
-            }
-            if ((offset1 != offset2) &&
-                delta_cmp(offset1, offset2) != 0) {
-                offdiff = delta_subtract(offset1, offset2);
-                if (offdiff == NULL) {
-                    Py_DECREF(offset1);
-                    Py_DECREF(offset2);
-                    return NULL;
-                }
-            }
-            Py_DECREF(offset1);
-            Py_DECREF(offset2);
-            delta_d = ymd_to_ord(GET_YEAR(left),
-                                 GET_MONTH(left),
-                                 GET_DAY(left)) -
-                      ymd_to_ord(GET_YEAR(right),
-                                 GET_MONTH(right),
-                                 GET_DAY(right));
-            /* These can't overflow, since the values are
-             * normalized.  At most this gives the number of
-             * seconds in one day.
-             */
-            delta_s = (DATE_GET_HOUR(left) -
-                       DATE_GET_HOUR(right)) * 3600 +
-                      (DATE_GET_MINUTE(left) -
-                       DATE_GET_MINUTE(right)) * 60 +
-                      (DATE_GET_SECOND(left) -
-                       DATE_GET_SECOND(right));
-            delta_us = DATE_GET_MICROSECOND(left) -
-                       DATE_GET_MICROSECOND(right);
-            result = new_delta(delta_d, delta_s, delta_us, 1);
-            if (offdiff != NULL) {
-                PyObject *temp = result;
-                result = delta_subtract(result, offdiff);
-                Py_DECREF(temp);
-                Py_DECREF(offdiff);
-            }
-        }
-        else if (PyDelta_Check(right)) {
-            /* datetime - delta */
-            result = add_datetime_timedelta(
-                            (PyDateTime_DateTime *)left,
-                            (PyDateTime_Delta *)right,
-                            -1);
-        }
-    }
-
-    if (result == Py_NotImplemented)
-        Py_INCREF(result);
-    return result;
-}
-
-/* Various ways to turn a datetime into a string. */
-
-static PyObject *
-datetime_repr(PyDateTime_DateTime *self)
-{
-    const char *type_name = Py_TYPE(self)->tp_name;
-    PyObject *baserepr;
-
-    if (DATE_GET_MICROSECOND(self)) {
-        baserepr = PyUnicode_FromFormat(
-                      "%s(%d, %d, %d, %d, %d, %d, %d)",
-                      type_name,
-                      GET_YEAR(self), GET_MONTH(self), GET_DAY(self),
-                      DATE_GET_HOUR(self), DATE_GET_MINUTE(self),
-                      DATE_GET_SECOND(self),
-                      DATE_GET_MICROSECOND(self));
-    }
-    else if (DATE_GET_SECOND(self)) {
-        baserepr = PyUnicode_FromFormat(
-                      "%s(%d, %d, %d, %d, %d, %d)",
-                      type_name,
-                      GET_YEAR(self), GET_MONTH(self), GET_DAY(self),
-                      DATE_GET_HOUR(self), DATE_GET_MINUTE(self),
-                      DATE_GET_SECOND(self));
-    }
-    else {
-        baserepr = PyUnicode_FromFormat(
-                      "%s(%d, %d, %d, %d, %d)",
-                      type_name,
-                      GET_YEAR(self), GET_MONTH(self), GET_DAY(self),
-                      DATE_GET_HOUR(self), DATE_GET_MINUTE(self));
-    }
-    if (baserepr == NULL || ! HASTZINFO(self))
-        return baserepr;
-    return append_keyword_tzinfo(baserepr, self->tzinfo);
-}
-
-static PyObject *
-datetime_str(PyDateTime_DateTime *self)
-{
-    return PyObject_CallMethod((PyObject *)self, "isoformat", "(s)", " ");
-}
-
-static PyObject *
-datetime_isoformat(PyDateTime_DateTime *self, PyObject *args, PyObject *kw)
-{
-    int sep = 'T';
-    static char *keywords[] = {"sep", NULL};
-    char buffer[100];
-    PyObject *result;
-    int us = DATE_GET_MICROSECOND(self);
-
-    if (!PyArg_ParseTupleAndKeywords(args, kw, "|C:isoformat", keywords, &sep))
-        return NULL;
-    if (us)
-        result = PyUnicode_FromFormat("%04d-%02d-%02d%c%02d:%02d:%02d.%06d",
-                                      GET_YEAR(self), GET_MONTH(self),
-                                      GET_DAY(self), (int)sep,
-                                      DATE_GET_HOUR(self), DATE_GET_MINUTE(self),
-                                      DATE_GET_SECOND(self), us);
-    else
-        result = PyUnicode_FromFormat("%04d-%02d-%02d%c%02d:%02d:%02d",
-                                      GET_YEAR(self), GET_MONTH(self),
-                                      GET_DAY(self), (int)sep,
-                                      DATE_GET_HOUR(self), DATE_GET_MINUTE(self),
-                                      DATE_GET_SECOND(self));
-
-    if (!result || !HASTZINFO(self))
-        return result;
-
-    /* We need to append the UTC offset. */
-    if (format_utcoffset(buffer, sizeof(buffer), ":", self->tzinfo,
-                         (PyObject *)self) < 0) {
-        Py_DECREF(result);
-        return NULL;
-    }
-    PyUnicode_AppendAndDel(&result, PyUnicode_FromString(buffer));
-    return result;
-}
-
-static PyObject *
-datetime_ctime(PyDateTime_DateTime *self)
-{
-    return format_ctime((PyDateTime_Date *)self,
-                        DATE_GET_HOUR(self),
-                        DATE_GET_MINUTE(self),
-                        DATE_GET_SECOND(self));
-}
-
-/* Miscellaneous methods. */
-
-static PyObject *
-datetime_richcompare(PyObject *self, PyObject *other, int op)
-{
-    PyObject *result = NULL;
-    PyObject *offset1, *offset2;
-    int diff;
-
-    if (! PyDateTime_Check(other)) {
-        if (PyDate_Check(other)) {
-            /* Prevent invocation of date_richcompare.  We want to
-               return NotImplemented here to give the other object
-               a chance.  But since DateTime is a subclass of
-               Date, if the other object is a Date, it would
-               compute an ordering based on the date part alone,
-               and we don't want that.  So force unequal or
-               uncomparable here in that case. */
-            if (op == Py_EQ)
-                Py_RETURN_FALSE;
-            if (op == Py_NE)
-                Py_RETURN_TRUE;
-            return cmperror(self, other);
-        }
-        Py_INCREF(Py_NotImplemented);
-        return Py_NotImplemented;
-    }
-
-    if (GET_DT_TZINFO(self) == GET_DT_TZINFO(other)) {
-        diff = memcmp(((PyDateTime_DateTime *)self)->data,
-                      ((PyDateTime_DateTime *)other)->data,
-                      _PyDateTime_DATETIME_DATASIZE);
-        return diff_to_bool(diff, op);
-    }
-    offset1 = datetime_utcoffset(self, NULL);
-    if (offset1 == NULL)
-        return NULL;
-    offset2 = datetime_utcoffset(other, NULL);
-    if (offset2 == NULL)
-        goto done;
-    /* If they're both naive, or both aware and have the same offsets,
-     * we get off cheap.  Note that if they're both naive, offset1 ==
-     * offset2 == Py_None at this point.
-     */
-    if ((offset1 == offset2) ||
-        (PyDelta_Check(offset1) && PyDelta_Check(offset2) &&
-         delta_cmp(offset1, offset2) == 0)) {
-        diff = memcmp(((PyDateTime_DateTime *)self)->data,
-                      ((PyDateTime_DateTime *)other)->data,
-                      _PyDateTime_DATETIME_DATASIZE);
-        result = diff_to_bool(diff, op);
-    }
-    else if (offset1 != Py_None && offset2 != Py_None) {
-        PyDateTime_Delta *delta;
-
-        assert(offset1 != offset2); /* else last "if" handled it */
-        delta = (PyDateTime_Delta *)datetime_subtract((PyObject *)self,
-                                                       other);
-        if (delta == NULL)
-            goto done;
-        diff = GET_TD_DAYS(delta);
-        if (diff == 0)
-            diff = GET_TD_SECONDS(delta) |
-                   GET_TD_MICROSECONDS(delta);
-        Py_DECREF(delta);
-        result = diff_to_bool(diff, op);
-    }
-    else {
-        PyErr_SetString(PyExc_TypeError,
-                        "can't compare offset-naive and "
-                        "offset-aware datetimes");
-    }
- done:
-    Py_DECREF(offset1);
-    Py_XDECREF(offset2);
-    return result;
-}
-
-static long
-datetime_hash(PyDateTime_DateTime *self)
-{
-    if (self->hashcode == -1) {
-        PyObject *offset;
-
-        offset = datetime_utcoffset((PyObject *)self, NULL);
-
-        if (offset == NULL)
-            return -1;
-
-        /* Reduce this to a hash of another object. */
-        if (offset == Py_None)
-            self->hashcode = generic_hash(
-                (unsigned char *)self->data, _PyDateTime_DATETIME_DATASIZE);
-        else {
-            PyObject *temp1, *temp2;
-            int days, seconds;
-
-            assert(HASTZINFO(self));
-            days = ymd_to_ord(GET_YEAR(self),
-                              GET_MONTH(self),
-                              GET_DAY(self));
-            seconds = DATE_GET_HOUR(self) * 3600 +
-                      DATE_GET_MINUTE(self) * 60 +
-                      DATE_GET_SECOND(self);
-            temp1 = new_delta(days, seconds,
-                              DATE_GET_MICROSECOND(self),
-                              1);
-            if (temp1 == NULL) {
-                Py_DECREF(offset);
-                return -1;
-            }
-            temp2 = delta_subtract(temp1, offset);
-            Py_DECREF(temp1);
-            if (temp2 == NULL) {
-                Py_DECREF(offset);
-                return -1;
-            }
-            self->hashcode = PyObject_Hash(temp2);
-            Py_DECREF(temp2);
-        }
-        Py_DECREF(offset);
-    }
-    return self->hashcode;
-}
-
-static PyObject *
-datetime_replace(PyDateTime_DateTime *self, PyObject *args, PyObject *kw)
-{
-    PyObject *clone;
-    PyObject *tuple;
-    int y = GET_YEAR(self);
-    int m = GET_MONTH(self);
-    int d = GET_DAY(self);
-    int hh = DATE_GET_HOUR(self);
-    int mm = DATE_GET_MINUTE(self);
-    int ss = DATE_GET_SECOND(self);
-    int us = DATE_GET_MICROSECOND(self);
-    PyObject *tzinfo = HASTZINFO(self) ? self->tzinfo : Py_None;
-
-    if (! PyArg_ParseTupleAndKeywords(args, kw, "|iiiiiiiO:replace",
-                                      datetime_kws,
-                                      &y, &m, &d, &hh, &mm, &ss, &us,
-                                      &tzinfo))
-        return NULL;
-    tuple = Py_BuildValue("iiiiiiiO", y, m, d, hh, mm, ss, us, tzinfo);
-    if (tuple == NULL)
-        return NULL;
-    clone = datetime_new(Py_TYPE(self), tuple, NULL);
-    Py_DECREF(tuple);
-    return clone;
-}
-
-static PyObject *
-datetime_astimezone(PyDateTime_DateTime *self, PyObject *args, PyObject *kw)
-{
-    PyObject *result;
-    PyObject *offset;
-    PyObject *temp;
-    PyObject *tzinfo;
-    static char *keywords[] = {"tz", NULL};
-
-    if (! PyArg_ParseTupleAndKeywords(args, kw, "O!:astimezone", keywords,
-                                      &PyDateTime_TZInfoType, &tzinfo))
-        return NULL;
-
-    if (!HASTZINFO(self) || self->tzinfo == Py_None)
-        goto NeedAware;
-
-    /* Conversion to self's own time zone is a NOP. */
-    if (self->tzinfo == tzinfo) {
-        Py_INCREF(self);
-        return (PyObject *)self;
-    }
-
-    /* Convert self to UTC. */
-    offset = datetime_utcoffset((PyObject *)self, NULL);
-    if (offset == NULL)
-        return NULL;
-    if (offset == Py_None) {
-        Py_DECREF(offset);
-      NeedAware:
-        PyErr_SetString(PyExc_ValueError, "astimezone() cannot be applied to "
-                        "a naive datetime");
-        return NULL;
-    }
-
-    /* result = self - offset */
-    result = add_datetime_timedelta(self,
-                (PyDateTime_Delta *)offset, -1);
-    Py_DECREF(offset);
-    if (result == NULL)
-        return NULL;
-
-    /* Attach new tzinfo and let fromutc() do the rest. */
-    temp = ((PyDateTime_DateTime *)result)->tzinfo;
-    ((PyDateTime_DateTime *)result)->tzinfo = tzinfo;
-    Py_INCREF(tzinfo);
-    Py_DECREF(temp);
-
-    temp = result;
-    result = PyObject_CallMethod(tzinfo, "fromutc", "O", temp);
-    Py_DECREF(temp);
-
-    return result;
-}
-
-static PyObject *
-datetime_timetuple(PyDateTime_DateTime *self)
-{
-    int dstflag = -1;
-
-    if (HASTZINFO(self) && self->tzinfo != Py_None) {
-        PyObject * dst;
-
-        dst = call_dst(self->tzinfo, (PyObject *)self);
-        if (dst == NULL)
-            return NULL;
-
-        if (dst != Py_None)
-            dstflag = delta_bool((PyDateTime_Delta *)dst);
-        Py_DECREF(dst);
-    }
-    return build_struct_time(GET_YEAR(self),
-                             GET_MONTH(self),
-                             GET_DAY(self),
-                             DATE_GET_HOUR(self),
-                             DATE_GET_MINUTE(self),
-                             DATE_GET_SECOND(self),
-                             dstflag);
-}
-
-static PyObject *
-datetime_getdate(PyDateTime_DateTime *self)
-{
-    return new_date(GET_YEAR(self),
-                    GET_MONTH(self),
-                    GET_DAY(self));
-}
-
-static PyObject *
-datetime_gettime(PyDateTime_DateTime *self)
-{
-    return new_time(DATE_GET_HOUR(self),
-                    DATE_GET_MINUTE(self),
-                    DATE_GET_SECOND(self),
-                    DATE_GET_MICROSECOND(self),
-                    Py_None);
-}
-
-static PyObject *
-datetime_gettimetz(PyDateTime_DateTime *self)
-{
-    return new_time(DATE_GET_HOUR(self),
-                    DATE_GET_MINUTE(self),
-                    DATE_GET_SECOND(self),
-                    DATE_GET_MICROSECOND(self),
-                    GET_DT_TZINFO(self));
-}
-
-static PyObject *
-datetime_utctimetuple(PyDateTime_DateTime *self)
-{
-    int y, m, d, hh, mm, ss;
-    PyObject *tzinfo;
-    PyDateTime_DateTime *utcself;
-
-    tzinfo = GET_DT_TZINFO(self);
-    if (tzinfo == Py_None) {
-        utcself = self;
-        Py_INCREF(utcself);
-    }
-    else {
-        PyObject *offset;
-        offset = call_utcoffset(tzinfo, (PyObject *)self);
-        if (offset == NULL)
-            return NULL;
-        if (offset == Py_None) {
-            Py_DECREF(offset);
-            utcself = self;
-            Py_INCREF(utcself);
-        }
-        else {
-            utcself = (PyDateTime_DateTime *)add_datetime_timedelta(self,
-                                                (PyDateTime_Delta *)offset, -1);
-            Py_DECREF(offset);
-            if (utcself == NULL)
-                return NULL;
-        }
-    }
-    y = GET_YEAR(utcself);
-    m = GET_MONTH(utcself);
-    d = GET_DAY(utcself);
-    hh = DATE_GET_HOUR(utcself);
-    mm = DATE_GET_MINUTE(utcself);
-    ss = DATE_GET_SECOND(utcself);
-
-    Py_DECREF(utcself);
-    return build_struct_time(y, m, d, hh, mm, ss, 0);
-}
-
-/* Pickle support, a simple use of __reduce__. */
-
-/* Let basestate be the non-tzinfo data string.
- * If tzinfo is None, this returns (basestate,), else (basestate, tzinfo).
- * So it's a tuple in any (non-error) case.
- * __getstate__ isn't exposed.
- */
-static PyObject *
-datetime_getstate(PyDateTime_DateTime *self)
-{
-    PyObject *basestate;
-    PyObject *result = NULL;
-
-    basestate = PyBytes_FromStringAndSize((char *)self->data,
-                                           _PyDateTime_DATETIME_DATASIZE);
-    if (basestate != NULL) {
-        if (! HASTZINFO(self) || self->tzinfo == Py_None)
-            result = PyTuple_Pack(1, basestate);
-        else
-            result = PyTuple_Pack(2, basestate, self->tzinfo);
-        Py_DECREF(basestate);
-    }
-    return result;
-}
-
-static PyObject *
-datetime_reduce(PyDateTime_DateTime *self, PyObject *arg)
-{
-    return Py_BuildValue("(ON)", Py_TYPE(self), datetime_getstate(self));
-}
-
-static PyMethodDef datetime_methods[] = {
-
-    /* Class methods: */
-
-    {"now",         (PyCFunction)datetime_now,
-     METH_VARARGS | METH_KEYWORDS | METH_CLASS,
-     PyDoc_STR("[tz] -> new datetime with tz's local day and time.")},
-
-    {"utcnow",         (PyCFunction)datetime_utcnow,
-     METH_NOARGS | METH_CLASS,
-     PyDoc_STR("Return a new datetime representing UTC day and time.")},
-
-    {"fromtimestamp", (PyCFunction)datetime_fromtimestamp,
-     METH_VARARGS | METH_KEYWORDS | METH_CLASS,
-     PyDoc_STR("timestamp[, tz] -> tz's local time from POSIX timestamp.")},
-
-    {"utcfromtimestamp", (PyCFunction)datetime_utcfromtimestamp,
-     METH_VARARGS | METH_CLASS,
-     PyDoc_STR("timestamp -> UTC datetime from a POSIX timestamp "
-               "(like time.time()).")},
-
-    {"strptime", (PyCFunction)datetime_strptime,
-     METH_VARARGS | METH_CLASS,
-     PyDoc_STR("string, format -> new datetime parsed from a string "
-               "(like time.strptime()).")},
-
-    {"combine", (PyCFunction)datetime_combine,
-     METH_VARARGS | METH_KEYWORDS | METH_CLASS,
-     PyDoc_STR("date, time -> datetime with same date and time fields")},
-
-    /* Instance methods: */
-
-    {"date",   (PyCFunction)datetime_getdate, METH_NOARGS,
-     PyDoc_STR("Return date object with same year, month and day.")},
-
-    {"time",   (PyCFunction)datetime_gettime, METH_NOARGS,
-     PyDoc_STR("Return time object with same time but with tzinfo=None.")},
-
-    {"timetz",   (PyCFunction)datetime_gettimetz, METH_NOARGS,
-     PyDoc_STR("Return time object with same time and tzinfo.")},
-
-    {"ctime",       (PyCFunction)datetime_ctime,        METH_NOARGS,
-     PyDoc_STR("Return ctime() style string.")},
-
-    {"timetuple",   (PyCFunction)datetime_timetuple, METH_NOARGS,
-     PyDoc_STR("Return time tuple, compatible with time.localtime().")},
-
-    {"utctimetuple",   (PyCFunction)datetime_utctimetuple, METH_NOARGS,
-     PyDoc_STR("Return UTC time tuple, compatible with time.localtime().")},
-
-    {"isoformat",   (PyCFunction)datetime_isoformat, METH_VARARGS | METH_KEYWORDS,
-     PyDoc_STR("[sep] -> string in ISO 8601 format, "
-               "YYYY-MM-DDTHH:MM:SS[.mmmmmm][+HH:MM].\n\n"
-               "sep is used to separate the year from the time, and "
-               "defaults to 'T'.")},
-
-    {"utcoffset",       (PyCFunction)datetime_utcoffset, METH_NOARGS,
-     PyDoc_STR("Return self.tzinfo.utcoffset(self).")},
-
-    {"tzname",          (PyCFunction)datetime_tzname,   METH_NOARGS,
-     PyDoc_STR("Return self.tzinfo.tzname(self).")},
-
-    {"dst",             (PyCFunction)datetime_dst, METH_NOARGS,
-     PyDoc_STR("Return self.tzinfo.dst(self).")},
-
-    {"replace",     (PyCFunction)datetime_replace,      METH_VARARGS | METH_KEYWORDS,
-     PyDoc_STR("Return datetime with new specified fields.")},
-
-    {"astimezone",  (PyCFunction)datetime_astimezone, METH_VARARGS | METH_KEYWORDS,
-     PyDoc_STR("tz -> convert to local time in new timezone tz\n")},
-
-    {"__reduce__", (PyCFunction)datetime_reduce,     METH_NOARGS,
-     PyDoc_STR("__reduce__() -> (cls, state)")},
-
-    {NULL,      NULL}
-};
-
-static char datetime_doc[] =
-PyDoc_STR("datetime(year, month, day[, hour[, minute[, second[, microsecond[,tzinfo]]]]])\n\
-\n\
-The year, month and day arguments are required. tzinfo may be None, or an\n\
-instance of a tzinfo subclass. The remaining arguments may be ints or longs.\n");
-
-static PyNumberMethods datetime_as_number = {
-    datetime_add,                               /* nb_add */
-    datetime_subtract,                          /* nb_subtract */
-    0,                                          /* nb_multiply */
-    0,                                          /* nb_remainder */
-    0,                                          /* nb_divmod */
-    0,                                          /* nb_power */
-    0,                                          /* nb_negative */
-    0,                                          /* nb_positive */
-    0,                                          /* nb_absolute */
-    0,                                          /* nb_bool */
-};
-
-static PyTypeObject PyDateTime_DateTimeType = {
-    PyVarObject_HEAD_INIT(NULL, 0)
-    "datetime.datetime",                        /* tp_name */
-    sizeof(PyDateTime_DateTime),                /* tp_basicsize */
-    0,                                          /* tp_itemsize */
-    (destructor)datetime_dealloc,               /* tp_dealloc */
-    0,                                          /* tp_print */
-    0,                                          /* tp_getattr */
-    0,                                          /* tp_setattr */
-    0,                                          /* tp_reserved */
-    (reprfunc)datetime_repr,                    /* tp_repr */
-    &datetime_as_number,                        /* tp_as_number */
-    0,                                          /* tp_as_sequence */
-    0,                                          /* tp_as_mapping */
-    (hashfunc)datetime_hash,                    /* tp_hash */
-    0,                                          /* tp_call */
-    (reprfunc)datetime_str,                     /* tp_str */
-    PyObject_GenericGetAttr,                    /* tp_getattro */
-    0,                                          /* tp_setattro */
-    0,                                          /* tp_as_buffer */
-    Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE, /* tp_flags */
-    datetime_doc,                               /* tp_doc */
-    0,                                          /* tp_traverse */
-    0,                                          /* tp_clear */
-    datetime_richcompare,                       /* tp_richcompare */
-    0,                                          /* tp_weaklistoffset */
-    0,                                          /* tp_iter */
-    0,                                          /* tp_iternext */
-    datetime_methods,                           /* tp_methods */
-    0,                                          /* tp_members */
-    datetime_getset,                            /* tp_getset */
-    &PyDateTime_DateType,                       /* tp_base */
-    0,                                          /* tp_dict */
-    0,                                          /* tp_descr_get */
-    0,                                          /* tp_descr_set */
-    0,                                          /* tp_dictoffset */
-    0,                                          /* tp_init */
-    datetime_alloc,                             /* tp_alloc */
-    datetime_new,                               /* tp_new */
-    0,                                          /* tp_free */
-};
-
-/* ---------------------------------------------------------------------------
- * Module methods and initialization.
- */
-
-static PyMethodDef module_methods[] = {
-    {NULL, NULL}
-};
-
-/* C API.  Clients get at this via PyDateTime_IMPORT, defined in
- * datetime.h.
- */
-static PyDateTime_CAPI CAPI = {
-    &PyDateTime_DateType,
-    &PyDateTime_DateTimeType,
-    &PyDateTime_TimeType,
-    &PyDateTime_DeltaType,
-    &PyDateTime_TZInfoType,
-    new_date_ex,
-    new_datetime_ex,
-    new_time_ex,
-    new_delta_ex,
-    datetime_fromtimestamp,
-    date_fromtimestamp
-};
-
-
-
-static struct PyModuleDef datetimemodule = {
-    PyModuleDef_HEAD_INIT,
-    "datetime",
-    "Fast implementation of the datetime type.",
-    -1,
-    module_methods,
-    NULL,
-    NULL,
-    NULL,
-    NULL
-};
-
-PyMODINIT_FUNC
-PyInit_datetime(void)
-{
-    PyObject *m;        /* a module object */
-    PyObject *d;        /* its dict */
-    PyObject *x;
-    PyObject *delta;
-
-    m = PyModule_Create(&datetimemodule);
-    if (m == NULL)
-        return NULL;
-
-    if (PyType_Ready(&PyDateTime_DateType) < 0)
-        return NULL;
-    if (PyType_Ready(&PyDateTime_DateTimeType) < 0)
-        return NULL;
-    if (PyType_Ready(&PyDateTime_DeltaType) < 0)
-        return NULL;
-    if (PyType_Ready(&PyDateTime_TimeType) < 0)
-        return NULL;
-    if (PyType_Ready(&PyDateTime_TZInfoType) < 0)
-        return NULL;
-    if (PyType_Ready(&PyDateTime_TimeZoneType) < 0)
-        return NULL;
-
-    /* timedelta values */
-    d = PyDateTime_DeltaType.tp_dict;
-
-    x = new_delta(0, 0, 1, 0);
-    if (x == NULL || PyDict_SetItemString(d, "resolution", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    x = new_delta(-MAX_DELTA_DAYS, 0, 0, 0);
-    if (x == NULL || PyDict_SetItemString(d, "min", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    x = new_delta(MAX_DELTA_DAYS, 24*3600-1, 1000000-1, 0);
-    if (x == NULL || PyDict_SetItemString(d, "max", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    /* date values */
-    d = PyDateTime_DateType.tp_dict;
-
-    x = new_date(1, 1, 1);
-    if (x == NULL || PyDict_SetItemString(d, "min", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    x = new_date(MAXYEAR, 12, 31);
-    if (x == NULL || PyDict_SetItemString(d, "max", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    x = new_delta(1, 0, 0, 0);
-    if (x == NULL || PyDict_SetItemString(d, "resolution", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    /* time values */
-    d = PyDateTime_TimeType.tp_dict;
-
-    x = new_time(0, 0, 0, 0, Py_None);
-    if (x == NULL || PyDict_SetItemString(d, "min", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    x = new_time(23, 59, 59, 999999, Py_None);
-    if (x == NULL || PyDict_SetItemString(d, "max", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    x = new_delta(0, 0, 1, 0);
-    if (x == NULL || PyDict_SetItemString(d, "resolution", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    /* datetime values */
-    d = PyDateTime_DateTimeType.tp_dict;
-
-    x = new_datetime(1, 1, 1, 0, 0, 0, 0, Py_None);
-    if (x == NULL || PyDict_SetItemString(d, "min", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    x = new_datetime(MAXYEAR, 12, 31, 23, 59, 59, 999999, Py_None);
-    if (x == NULL || PyDict_SetItemString(d, "max", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    x = new_delta(0, 0, 1, 0);
-    if (x == NULL || PyDict_SetItemString(d, "resolution", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    /* timezone values */
-    d = PyDateTime_TimeZoneType.tp_dict;
-
-    delta = new_delta(0, 0, 0, 0);
-    if (delta == NULL)
-        return NULL;
-    x = new_timezone(delta, NULL);
-    Py_DECREF(delta);
-    if (x == NULL || PyDict_SetItemString(d, "utc", x) < 0)
-        return NULL;
-    PyDateTime_TimeZone_UTC = x;
-
-    delta = new_delta(-1, 60, 0, 1); /* -23:59 */
-    if (delta == NULL)
-        return NULL;
-    x = new_timezone(delta, NULL);
-    Py_DECREF(delta);
-    if (x == NULL || PyDict_SetItemString(d, "min", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    delta = new_delta(0, (23 * 60 + 59) * 60, 0, 0); /* +23:59 */
-    if (delta == NULL)
-        return NULL;
-    x = new_timezone(delta, NULL);
-    Py_DECREF(delta);
-    if (x == NULL || PyDict_SetItemString(d, "max", x) < 0)
-        return NULL;
-    Py_DECREF(x);
-
-    /* module initialization */
-    PyModule_AddIntConstant(m, "MINYEAR", MINYEAR);
-    PyModule_AddIntConstant(m, "MAXYEAR", MAXYEAR);
-
-    Py_INCREF(&PyDateTime_DateType);
-    PyModule_AddObject(m, "date", (PyObject *) &PyDateTime_DateType);
-
-    Py_INCREF(&PyDateTime_DateTimeType);
-    PyModule_AddObject(m, "datetime",
-                       (PyObject *)&PyDateTime_DateTimeType);
-
-    Py_INCREF(&PyDateTime_TimeType);
-    PyModule_AddObject(m, "time", (PyObject *) &PyDateTime_TimeType);
-
-    Py_INCREF(&PyDateTime_DeltaType);
-    PyModule_AddObject(m, "timedelta", (PyObject *) &PyDateTime_DeltaType);
-
-    Py_INCREF(&PyDateTime_TZInfoType);
-    PyModule_AddObject(m, "tzinfo", (PyObject *) &PyDateTime_TZInfoType);
-
-    Py_INCREF(&PyDateTime_TimeZoneType);
-    PyModule_AddObject(m, "timezone", (PyObject *) &PyDateTime_TimeZoneType);
-
-    x = PyCapsule_New(&CAPI, PyDateTime_CAPSULE_NAME, NULL);
-    if (x == NULL)
-        return NULL;
-    PyModule_AddObject(m, "datetime_CAPI", x);
-
-    /* A 4-year cycle has an extra leap day over what we'd get from
-     * pasting together 4 single years.
-     */
-    assert(DI4Y == 4 * 365 + 1);
-    assert(DI4Y == days_before_year(4+1));
-
-    /* Similarly, a 400-year cycle has an extra leap day over what we'd
-     * get from pasting together 4 100-year cycles.
-     */
-    assert(DI400Y == 4 * DI100Y + 1);
-    assert(DI400Y == days_before_year(400+1));
-
-    /* OTOH, a 100-year cycle has one fewer leap day than we'd get from
-     * pasting together 25 4-year cycles.
-     */
-    assert(DI100Y == 25 * DI4Y - 1);
-    assert(DI100Y == days_before_year(100+1));
-
-    us_per_us = PyLong_FromLong(1);
-    us_per_ms = PyLong_FromLong(1000);
-    us_per_second = PyLong_FromLong(1000000);
-    us_per_minute = PyLong_FromLong(60000000);
-    seconds_per_day = PyLong_FromLong(24 * 3600);
-    if (us_per_us == NULL || us_per_ms == NULL || us_per_second == NULL ||
-        us_per_minute == NULL || seconds_per_day == NULL)
-        return NULL;
-
-    /* The rest are too big for 32-bit ints, but even
-     * us_per_week fits in 40 bits, so doubles should be exact.
-     */
-    us_per_hour = PyLong_FromDouble(3600000000.0);
-    us_per_day = PyLong_FromDouble(86400000000.0);
-    us_per_week = PyLong_FromDouble(604800000000.0);
-    if (us_per_hour == NULL || us_per_day == NULL || us_per_week == NULL)
-        return NULL;
-    return m;
-}
-
-/* ---------------------------------------------------------------------------
-Some time zone algebra.  For a datetime x, let
-    x.n = x stripped of its timezone -- its naive time.
-    x.o = x.utcoffset(), and assuming that doesn't raise an exception or
-      return None
-    x.d = x.dst(), and assuming that doesn't raise an exception or
-      return None
-    x.s = x's standard offset, x.o - x.d
-
-Now some derived rules, where k is a duration (timedelta).
-
-1. x.o = x.s + x.d
-   This follows from the definition of x.s.
-
-2. If x and y have the same tzinfo member, x.s = y.s.
-   This is actually a requirement, an assumption we need to make about
-   sane tzinfo classes.
-
-3. The naive UTC time corresponding to x is x.n - x.o.
-   This is again a requirement for a sane tzinfo class.
-
-4. (x+k).s = x.s
-   This follows from #2, and that datimetimetz+timedelta preserves tzinfo.
-
-5. (x+k).n = x.n + k
-   Again follows from how arithmetic is defined.
-
-Now we can explain tz.fromutc(x).  Let's assume it's an interesting case
-(meaning that the various tzinfo methods exist, and don't blow up or return
-None when called).
-
-The function wants to return a datetime y with timezone tz, equivalent to x.
-x is already in UTC.
-
-By #3, we want
-
-    y.n - y.o = x.n                             [1]
-
-The algorithm starts by attaching tz to x.n, and calling that y.  So
-x.n = y.n at the start.  Then it wants to add a duration k to y, so that [1]
-becomes true; in effect, we want to solve [2] for k:
-
-   (y+k).n - (y+k).o = x.n                      [2]
-
-By #1, this is the same as
-
-   (y+k).n - ((y+k).s + (y+k).d) = x.n          [3]
-
-By #5, (y+k).n = y.n + k, which equals x.n + k because x.n=y.n at the start.
-Substituting that into [3],
-
-   x.n + k - (y+k).s - (y+k).d = x.n; the x.n terms cancel, leaving
-   k - (y+k).s - (y+k).d = 0; rearranging,
-   k = (y+k).s - (y+k).d; by #4, (y+k).s == y.s, so
-   k = y.s - (y+k).d
-
-On the RHS, (y+k).d can't be computed directly, but y.s can be, and we
-approximate k by ignoring the (y+k).d term at first.  Note that k can't be
-very large, since all offset-returning methods return a duration of magnitude
-less than 24 hours.  For that reason, if y is firmly in std time, (y+k).d must
-be 0, so ignoring it has no consequence then.
-
-In any case, the new value is
-
-    z = y + y.s                                 [4]
-
-It's helpful to step back at look at [4] from a higher level:  it's simply
-mapping from UTC to tz's standard time.
-
-At this point, if
-
-    z.n - z.o = x.n                             [5]
-
-we have an equivalent time, and are almost done.  The insecurity here is
-at the start of daylight time.  Picture US Eastern for concreteness.  The wall
-time jumps from 1:59 to 3:00, and wall hours of the form 2:MM don't make good
-sense then.  The docs ask that an Eastern tzinfo class consider such a time to
-be EDT (because it's "after 2"), which is a redundant spelling of 1:MM EST
-on the day DST starts.  We want to return the 1:MM EST spelling because that's
-the only spelling that makes sense on the local wall clock.
-
-In fact, if [5] holds at this point, we do have the standard-time spelling,
-but that takes a bit of proof.  We first prove a stronger result.  What's the
-difference between the LHS and RHS of [5]?  Let
-
-    diff = x.n - (z.n - z.o)                    [6]
-
-Now
-    z.n =                       by [4]
-    (y + y.s).n =               by #5
-    y.n + y.s =                 since y.n = x.n
-    x.n + y.s =                 since z and y are have the same tzinfo member,
-                                    y.s = z.s by #2
-    x.n + z.s
-
-Plugging that back into [6] gives
-
-    diff =
-    x.n - ((x.n + z.s) - z.o) =     expanding
-    x.n - x.n - z.s + z.o =         cancelling
-    - z.s + z.o =                   by #2
-    z.d
-
-So diff = z.d.
-
-If [5] is true now, diff = 0, so z.d = 0 too, and we have the standard-time
-spelling we wanted in the endcase described above.  We're done.  Contrarily,
-if z.d = 0, then we have a UTC equivalent, and are also done.
-
-If [5] is not true now, diff = z.d != 0, and z.d is the offset we need to
-add to z (in effect, z is in tz's standard time, and we need to shift the
-local clock into tz's daylight time).
-
-Let
-
-    z' = z + z.d = z + diff                     [7]
-
-and we can again ask whether
-
-    z'.n - z'.o = x.n                           [8]
-
-If so, we're done.  If not, the tzinfo class is insane, according to the
-assumptions we've made.  This also requires a bit of proof.  As before, let's
-compute the difference between the LHS and RHS of [8] (and skipping some of
-the justifications for the kinds of substitutions we've done several times
-already):
-
-    diff' = x.n - (z'.n - z'.o) =           replacing z'.n via [7]
-        x.n  - (z.n + diff - z'.o) =    replacing diff via [6]
-        x.n - (z.n + x.n - (z.n - z.o) - z'.o) =
-        x.n - z.n - x.n + z.n - z.o + z'.o =    cancel x.n
-        - z.n + z.n - z.o + z'.o =              cancel z.n
-        - z.o + z'.o =                      #1 twice
-        -z.s - z.d + z'.s + z'.d =          z and z' have same tzinfo
-        z'.d - z.d
-
-So z' is UTC-equivalent to x iff z'.d = z.d at this point.  If they are equal,
-we've found the UTC-equivalent so are done.  In fact, we stop with [7] and
-return z', not bothering to compute z'.d.
-
-How could z.d and z'd differ?  z' = z + z.d [7], so merely moving z' by
-a dst() offset, and starting *from* a time already in DST (we know z.d != 0),
-would have to change the result dst() returns:  we start in DST, and moving
-a little further into it takes us out of DST.
-
-There isn't a sane case where this can happen.  The closest it gets is at
-the end of DST, where there's an hour in UTC with no spelling in a hybrid
-tzinfo class.  In US Eastern, that's 5:MM UTC = 0:MM EST = 1:MM EDT.  During
-that hour, on an Eastern clock 1:MM is taken as being in standard time (6:MM
-UTC) because the docs insist on that, but 0:MM is taken as being in daylight
-time (4:MM UTC).  There is no local time mapping to 5:MM UTC.  The local
-clock jumps from 1:59 back to 1:00 again, and repeats the 1:MM hour in
-standard time.  Since that's what the local clock *does*, we want to map both
-UTC hours 5:MM and 6:MM to 1:MM Eastern.  The result is ambiguous
-in local time, but so it goes -- it's the way the local clock works.
-
-When x = 5:MM UTC is the input to this algorithm, x.o=0, y.o=-5 and y.d=0,
-so z=0:MM.  z.d=60 (minutes) then, so [5] doesn't hold and we keep going.
-z' = z + z.d = 1:MM then, and z'.d=0, and z'.d - z.d = -60 != 0 so [8]
-(correctly) concludes that z' is not UTC-equivalent to x.
-
-Because we know z.d said z was in daylight time (else [5] would have held and
-we would have stopped then), and we know z.d != z'.d (else [8] would have held
-and we would have stopped then), and there are only 2 possible values dst() can
-return in Eastern, it follows that z'.d must be 0 (which it is in the example,
-but the reasoning doesn't depend on the example -- it depends on there being
-two possible dst() outcomes, one zero and the other non-zero).  Therefore
-z' must be in standard time, and is the spelling we want in this case.
-
-Note again that z' is not UTC-equivalent as far as the hybrid tzinfo class is
-concerned (because it takes z' as being in standard time rather than the
-daylight time we intend here), but returning it gives the real-life "local
-clock repeats an hour" behavior when mapping the "unspellable" UTC hour into
-tz.
-
-When the input is 6:MM, z=1:MM and z.d=0, and we stop at once, again with
-the 1:MM standard time spelling we want.
-
-So how can this break?  One of the assumptions must be violated.  Two
-possibilities:
-
-1) [2] effectively says that y.s is invariant across all y belong to a given
-   time zone.  This isn't true if, for political reasons or continental drift,
-   a region decides to change its base offset from UTC.
-
-2) There may be versions of "double daylight" time where the tail end of
-   the analysis gives up a step too early.  I haven't thought about that
-   enough to say.
-
-In any case, it's clear that the default fromutc() is strong enough to handle
-"almost all" time zones:  so long as the standard offset is invariant, it
-doesn't matter if daylight time transition points change from year to year, or
-if daylight time is skipped in some years; it doesn't matter how large or
-small dst() may get within its bounds; and it doesn't even matter if some
-perverse time zone returns a negative dst()).  So a breaking case must be
-pretty bizarre, and a tzinfo subclass can override fromutc() if it is.
---------------------------------------------------------------------------- */

Modified: python/branches/import_unicode/Modules/getpath.c
==============================================================================
--- python/branches/import_unicode/Modules/getpath.c	(original)
+++ python/branches/import_unicode/Modules/getpath.c	Fri Jul 30 01:19:37 2010
@@ -559,7 +559,7 @@
         if (!ismodule(argv0_path)) {
                 /* We are in the build directory so use the name of the
                    executable - we know that the absolute path is passed */
-                wcsncpy(argv0_path, prog, MAXPATHLEN);
+                wcsncpy(argv0_path, progpath, MAXPATHLEN);
         }
         else {
                 /* Use the location of the library as the progpath */

Modified: python/branches/import_unicode/Modules/makesetup
==============================================================================
--- python/branches/import_unicode/Modules/makesetup	(original)
+++ python/branches/import_unicode/Modules/makesetup	Fri Jul 30 01:19:37 2010
@@ -219,7 +219,7 @@
 			case $doconfig in
 			no)	cc="$cc \$(CCSHARED) \$(CFLAGS) \$(CPPFLAGS)";;
 			*)
-				cc="$cc \$(PY_CFLAGS)";;
+				cc="$cc \$(PY_CORE_CFLAGS)";;
 			esac
 			rule="$obj: $src; $cc $cpps -c $src -o $obj"
 			echo "$rule" >>$rulesf

Modified: python/branches/import_unicode/Modules/mathmodule.c
==============================================================================
--- python/branches/import_unicode/Modules/mathmodule.c	(original)
+++ python/branches/import_unicode/Modules/mathmodule.c	Fri Jul 30 01:19:37 2010
@@ -1818,6 +1818,19 @@
 Convert angle x from degrees to radians.");
 
 static PyObject *
+math_isfinite(PyObject *self, PyObject *arg)
+{
+    double x = PyFloat_AsDouble(arg);
+    if (x == -1.0 && PyErr_Occurred())
+        return NULL;
+    return PyBool_FromLong((long)Py_IS_FINITE(x));
+}
+
+PyDoc_STRVAR(math_isfinite_doc,
+"isfinite(x) -> bool\n\n\
+Return True if x is neither an infinity nor a NaN, and False otherwise.");
+
+static PyObject *
 math_isnan(PyObject *self, PyObject *arg)
 {
     double x = PyFloat_AsDouble(arg);
@@ -1828,7 +1841,7 @@
 
 PyDoc_STRVAR(math_isnan_doc,
 "isnan(x) -> bool\n\n\
-Check if float x is not a number (NaN).");
+Return True if x is a NaN (not a number), and False otherwise.");
 
 static PyObject *
 math_isinf(PyObject *self, PyObject *arg)
@@ -1841,7 +1854,7 @@
 
 PyDoc_STRVAR(math_isinf_doc,
 "isinf(x) -> bool\n\n\
-Check if float x is infinite (positive or negative).");
+Return True if x is a positive or negative infinity, and False otherwise.");
 
 static PyMethodDef math_methods[] = {
     {"acos",            math_acos,      METH_O,         math_acos_doc},
@@ -1868,6 +1881,7 @@
     {"fsum",            math_fsum,      METH_O,         math_fsum_doc},
     {"gamma",           math_gamma,     METH_O,         math_gamma_doc},
     {"hypot",           math_hypot,     METH_VARARGS,   math_hypot_doc},
+    {"isfinite",        math_isfinite,  METH_O,         math_isfinite_doc},
     {"isinf",           math_isinf,     METH_O,         math_isinf_doc},
     {"isnan",           math_isnan,     METH_O,         math_isnan_doc},
     {"ldexp",           math_ldexp,     METH_VARARGS,   math_ldexp_doc},

Modified: python/branches/import_unicode/Modules/posixmodule.c
==============================================================================
--- python/branches/import_unicode/Modules/posixmodule.c	(original)
+++ python/branches/import_unicode/Modules/posixmodule.c	Fri Jul 30 01:19:37 2010
@@ -1117,7 +1117,8 @@
                                                 "GetFinalPathNameByHandleA");
         *(FARPROC*)&Py_GetFinalPathNameByHandleW = GetProcAddress(hKernel32,
                                                 "GetFinalPathNameByHandleW");
-        has_GetFinalPathNameByHandle = Py_GetFinalPathNameByHandleA && Py_GetFinalPathNameByHandleW;
+        has_GetFinalPathNameByHandle = Py_GetFinalPathNameByHandleA &&
+                                       Py_GetFinalPathNameByHandleW;
     }
     return has_GetFinalPathNameByHandle;
 }
@@ -1310,9 +1311,12 @@
     /* similar to stat() */
     result->st_mode = attributes_to_mode(info.dwFileAttributes);
     result->st_size = (((__int64)info.nFileSizeHigh)<<32) + info.nFileSizeLow;
-    FILE_TIME_to_time_t_nsec(&info.ftCreationTime, &result->st_ctime, &result->st_ctime_nsec);
-    FILE_TIME_to_time_t_nsec(&info.ftLastWriteTime, &result->st_mtime, &result->st_mtime_nsec);
-    FILE_TIME_to_time_t_nsec(&info.ftLastAccessTime, &result->st_atime, &result->st_atime_nsec);
+    FILE_TIME_to_time_t_nsec(&info.ftCreationTime, &result->st_ctime,
+                             &result->st_ctime_nsec);
+    FILE_TIME_to_time_t_nsec(&info.ftLastWriteTime, &result->st_mtime,
+                             &result->st_mtime_nsec);
+    FILE_TIME_to_time_t_nsec(&info.ftLastAccessTime, &result->st_atime,
+                             &result->st_atime_nsec);
     /* specific to fstat() */
     result->st_nlink = info.nNumberOfLinks;
     result->st_ino = (((__int64)info.nFileIndexHigh)<<32) + info.nFileIndexLow;
@@ -2287,10 +2291,10 @@
 
 
 PyDoc_STRVAR(posix_listdir__doc__,
-"listdir(path) -> list_of_strings\n\n\
+"listdir([path]) -> list_of_strings\n\n\
 Return a list containing the names of the entries in the directory.\n\
 \n\
-    path: path of directory to list\n\
+    path: path of directory to list (default: '.')\n\
 \n\
 The list is in arbitrary order.  It does not include the special\n\
 entries '.' and '..' even if they are present in the directory.");
@@ -2311,18 +2315,25 @@
     char *bufptr = namebuf;
     Py_ssize_t len = sizeof(namebuf)-5; /* only claim to have space for MAX_PATH */
 
-    PyObject *po;
-    if (PyArg_ParseTuple(args, "U:listdir", &po)) {
+    PyObject *po = NULL;
+    if (PyArg_ParseTuple(args, "|U:listdir", &po)) {
         WIN32_FIND_DATAW wFileData;
-        Py_UNICODE *wnamebuf;
+        Py_UNICODE *wnamebuf, *po_wchars;
+        
+        if (po == NULL) { // Default arg: "."
+            po_wchars = L".";
+            len = 1;
+        } else {
+            po_wchars = PyUnicode_AS_UNICODE(po);
+            len = PyUnicode_GET_SIZE(po);
+        }
         /* Overallocate for \\*.*\0 */
-        len = PyUnicode_GET_SIZE(po);
         wnamebuf = malloc((len + 5) * sizeof(wchar_t));
         if (!wnamebuf) {
             PyErr_NoMemory();
             return NULL;
         }
-        wcscpy(wnamebuf, PyUnicode_AS_UNICODE(po));
+        wcscpy(wnamebuf, po_wchars);
         if (len > 0) {
             Py_UNICODE wch = wnamebuf[len-1];
             if (wch != L'/' && wch != L'\\' && wch != L':')
@@ -2544,12 +2555,17 @@
     int arg_is_unicode = 1;
 
     errno = 0;
-    if (!PyArg_ParseTuple(args, "U:listdir", &v)) {
+    /* v is never read, so it does not need to be initialized yet. */
+    if (!PyArg_ParseTuple(args, "|U:listdir", &v)) {
         arg_is_unicode = 0;
         PyErr_Clear();
     }
-    if (!PyArg_ParseTuple(args, "O&:listdir", PyUnicode_FSConverter, &oname))
+    oname = NULL;
+    if (!PyArg_ParseTuple(args, "|O&:listdir", PyUnicode_FSConverter, &oname))
         return NULL;
+    if (oname == NULL) { // Default arg: "."
+      oname = PyBytes_FromString(".");
+    }
     name = PyBytes_AsString(oname);
     if ((dirp = opendir(name)) == NULL) {
         return posix_error_with_allocated_filename(oname);
@@ -2691,7 +2707,8 @@
     
     if(hFile == INVALID_HANDLE_VALUE) {
         return win32_error_unicode("GetFinalPathNamyByHandle", path);
-        return PyErr_Format(PyExc_RuntimeError, "Could not get a handle to file.");
+        return PyErr_Format(PyExc_RuntimeError,
+                            "Could not get a handle to file.");
     }
 
     /* We have a good handle to the target, use it to determine the
@@ -3005,7 +3022,8 @@
 posix_unlink(PyObject *self, PyObject *args)
 {
 #ifdef MS_WINDOWS
-    return win32_1str(args, "remove", "y:remove", DeleteFileA, "U:remove", Py_DeleteFileW);
+    return win32_1str(args, "remove", "y:remove", DeleteFileA,
+                      "U:remove", Py_DeleteFileW);
 #else
     return posix_1str(args, "O&:remove", unlink);
 #endif
@@ -4148,17 +4166,49 @@
 #define MAX_GROUPS 64
 #endif
     gid_t grouplist[MAX_GROUPS];
+
+    /* On MacOSX getgroups(2) can return more than MAX_GROUPS results 
+     * This is a helper variable to store the intermediate result when
+     * that happens.
+     *
+     * To keep the code readable the OSX behaviour is unconditional,
+     * according to the POSIX spec this should be safe on all unix-y
+     * systems.
+     */
+    gid_t* alt_grouplist = grouplist;
     int n;
 
     n = getgroups(MAX_GROUPS, grouplist);
-    if (n < 0)
-        posix_error();
-    else {
-        result = PyList_New(n);
-        if (result != NULL) {
+    if (n < 0) {
+        if (errno == EINVAL) {
+            n = getgroups(0, NULL);
+            if (n == -1) {
+                return posix_error();
+            }
+            if (n == 0) {
+                /* Avoid malloc(0) */
+                alt_grouplist = grouplist;
+            } else {
+                alt_grouplist = PyMem_Malloc(n * sizeof(gid_t));
+                if (alt_grouplist == NULL) {
+                    errno = EINVAL;
+                    return posix_error();
+                }
+                n = getgroups(n, alt_grouplist);
+                if (n == -1) {
+                    PyMem_Free(alt_grouplist);
+                    return posix_error();
+                }
+            }
+        } else {
+            return posix_error();
+        }
+    }
+    result = PyList_New(n);
+    if (result != NULL) {
         int i;
         for (i = 0; i < n; ++i) {
-            PyObject *o = PyLong_FromLong((long)grouplist[i]);
+            PyObject *o = PyLong_FromLong((long)alt_grouplist[i]);
             if (o == NULL) {
             Py_DECREF(result);
             result = NULL;
@@ -4166,7 +4216,10 @@
             }
             PyList_SET_ITEM(result, i, o);
         }
-        }
+    }
+
+    if (alt_grouplist != grouplist) {
+        PyMem_Free(alt_grouplist);
     }
 
     return result;
@@ -4959,7 +5012,8 @@
     };
 } REPARSE_DATA_BUFFER, *PREPARSE_DATA_BUFFER;
 
-#define REPARSE_DATA_BUFFER_HEADER_SIZE  FIELD_OFFSET(REPARSE_DATA_BUFFER, GenericReparseBuffer)
+#define REPARSE_DATA_BUFFER_HEADER_SIZE  FIELD_OFFSET(REPARSE_DATA_BUFFER,\
+                                                      GenericReparseBuffer)
 
 #define MAXIMUM_REPARSE_DATA_BUFFER_SIZE  ( 16 * 1024 )
 
@@ -5023,8 +5077,11 @@
                 "not a symbolic link");
         return NULL;
     }
-    print_name = rdb->SymbolicLinkReparseBuffer.PathBuffer + rdb->SymbolicLinkReparseBuffer.PrintNameOffset;
-    result = PyUnicode_FromWideChar(print_name, rdb->SymbolicLinkReparseBuffer.PrintNameLength/2);
+    print_name = rdb->SymbolicLinkReparseBuffer.PathBuffer +
+                 rdb->SymbolicLinkReparseBuffer.PrintNameOffset;
+
+    result = PyUnicode_FromWideChar(print_name,
+                    rdb->SymbolicLinkReparseBuffer.PrintNameLength/2);
     return result;
 }
 
@@ -5043,7 +5100,8 @@
     if (has_CreateSymbolicLinkW)
         return has_CreateSymbolicLinkW;
     hKernel32 = GetModuleHandle("KERNEL32");
-    *(FARPROC*)&Py_CreateSymbolicLinkW = GetProcAddress(hKernel32, "CreateSymbolicLinkW");
+    *(FARPROC*)&Py_CreateSymbolicLinkW = GetProcAddress(hKernel32,
+                                                        "CreateSymbolicLinkW");
     if (Py_CreateSymbolicLinkW)
         has_CreateSymbolicLinkW = 1;
     return has_CreateSymbolicLinkW;
@@ -7586,7 +7644,8 @@
     {"symlink",         posix_symlink, METH_VARARGS, posix_symlink__doc__},
 #endif /* HAVE_SYMLINK */
 #if !defined(HAVE_SYMLINK) && defined(MS_WINDOWS)
-    {"symlink", (PyCFunction)win_symlink, METH_VARARGS | METH_KEYWORDS, win_symlink__doc__},
+    {"symlink", (PyCFunction)win_symlink, METH_VARARGS | METH_KEYWORDS,
+                win_symlink__doc__},
 #endif /* !defined(HAVE_SYMLINK) && defined(MS_WINDOWS) */
 #ifdef HAVE_SYSTEM
     {"system",          posix_system, METH_VARARGS, posix_system__doc__},

Modified: python/branches/import_unicode/Modules/sha1module.c
==============================================================================
--- python/branches/import_unicode/Modules/sha1module.c	(original)
+++ python/branches/import_unicode/Modules/sha1module.c	Fri Jul 30 01:19:37 2010
@@ -127,50 +127,50 @@
 
     /* compress */
     /* round one */
-    #define FF0(a,b,c,d,e,i) e = (ROLc(a, 5) + F0(b,c,d) + e + W[i] + 0x5a827999UL); b = ROLc(b, 30);
-    #define FF1(a,b,c,d,e,i) e = (ROLc(a, 5) + F1(b,c,d) + e + W[i] + 0x6ed9eba1UL); b = ROLc(b, 30);
-    #define FF2(a,b,c,d,e,i) e = (ROLc(a, 5) + F2(b,c,d) + e + W[i] + 0x8f1bbcdcUL); b = ROLc(b, 30);
-    #define FF3(a,b,c,d,e,i) e = (ROLc(a, 5) + F3(b,c,d) + e + W[i] + 0xca62c1d6UL); b = ROLc(b, 30);
+    #define FF_0(a,b,c,d,e,i) e = (ROLc(a, 5) + F0(b,c,d) + e + W[i] + 0x5a827999UL); b = ROLc(b, 30);
+    #define FF_1(a,b,c,d,e,i) e = (ROLc(a, 5) + F1(b,c,d) + e + W[i] + 0x6ed9eba1UL); b = ROLc(b, 30);
+    #define FF_2(a,b,c,d,e,i) e = (ROLc(a, 5) + F2(b,c,d) + e + W[i] + 0x8f1bbcdcUL); b = ROLc(b, 30);
+    #define FF_3(a,b,c,d,e,i) e = (ROLc(a, 5) + F3(b,c,d) + e + W[i] + 0xca62c1d6UL); b = ROLc(b, 30);
 
     for (i = 0; i < 20; ) {
-       FF0(a,b,c,d,e,i++);
-       FF0(e,a,b,c,d,i++);
-       FF0(d,e,a,b,c,i++);
-       FF0(c,d,e,a,b,i++);
-       FF0(b,c,d,e,a,i++);
+       FF_0(a,b,c,d,e,i++);
+       FF_0(e,a,b,c,d,i++);
+       FF_0(d,e,a,b,c,i++);
+       FF_0(c,d,e,a,b,i++);
+       FF_0(b,c,d,e,a,i++);
     }
 
     /* round two */
     for (; i < 40; )  {
-       FF1(a,b,c,d,e,i++);
-       FF1(e,a,b,c,d,i++);
-       FF1(d,e,a,b,c,i++);
-       FF1(c,d,e,a,b,i++);
-       FF1(b,c,d,e,a,i++);
+       FF_1(a,b,c,d,e,i++);
+       FF_1(e,a,b,c,d,i++);
+       FF_1(d,e,a,b,c,i++);
+       FF_1(c,d,e,a,b,i++);
+       FF_1(b,c,d,e,a,i++);
     }
 
     /* round three */
     for (; i < 60; )  {
-       FF2(a,b,c,d,e,i++);
-       FF2(e,a,b,c,d,i++);
-       FF2(d,e,a,b,c,i++);
-       FF2(c,d,e,a,b,i++);
-       FF2(b,c,d,e,a,i++);
+       FF_2(a,b,c,d,e,i++);
+       FF_2(e,a,b,c,d,i++);
+       FF_2(d,e,a,b,c,i++);
+       FF_2(c,d,e,a,b,i++);
+       FF_2(b,c,d,e,a,i++);
     }
 
     /* round four */
     for (; i < 80; )  {
-       FF3(a,b,c,d,e,i++);
-       FF3(e,a,b,c,d,i++);
-       FF3(d,e,a,b,c,i++);
-       FF3(c,d,e,a,b,i++);
-       FF3(b,c,d,e,a,i++);
+       FF_3(a,b,c,d,e,i++);
+       FF_3(e,a,b,c,d,i++);
+       FF_3(d,e,a,b,c,i++);
+       FF_3(c,d,e,a,b,i++);
+       FF_3(b,c,d,e,a,i++);
     }
 
-    #undef FF0
-    #undef FF1
-    #undef FF2
-    #undef FF3
+    #undef FF_0
+    #undef FF_1
+    #undef FF_2
+    #undef FF_3
 
     /* store */
     sha1->state[0] = sha1->state[0] + a;

Modified: python/branches/import_unicode/Modules/zipimport.c
==============================================================================
--- python/branches/import_unicode/Modules/zipimport.c	(original)
+++ python/branches/import_unicode/Modules/zipimport.c	Fri Jul 30 01:19:37 2010
@@ -553,19 +553,19 @@
 "is_package(fullname) -> bool.\n\
 \n\
 Return True if the module specified by fullname is a package.\n\
-Raise ZipImportError is the module couldn't be found.");
+Raise ZipImportError if the module couldn't be found.");
 
 PyDoc_STRVAR(doc_get_code,
 "get_code(fullname) -> code object.\n\
 \n\
 Return the code object for the specified module. Raise ZipImportError\n\
-is the module couldn't be found.");
+if the module couldn't be found.");
 
 PyDoc_STRVAR(doc_get_source,
 "get_source(fullname) -> source string.\n\
 \n\
 Return the source code for the specified module. Raise ZipImportError\n\
-is the module couldn't be found, return None if the archive does\n\
+if the module couldn't be found, return None if the archive does\n\
 contain the module, but has no source for it.");
 
 

Modified: python/branches/import_unicode/Objects/bytearrayobject.c
==============================================================================
--- python/branches/import_unicode/Objects/bytearrayobject.c	(original)
+++ python/branches/import_unicode/Objects/bytearrayobject.c	Fri Jul 30 01:19:37 2010
@@ -33,6 +33,7 @@
         PyObject *index = PyNumber_Index(arg);
         if (index == NULL) {
             PyErr_Format(PyExc_TypeError, "an integer is required");
+            *value = -1;
             return 0;
         }
         face_value = PyLong_AsLong(index);
@@ -42,6 +43,7 @@
     if (face_value < 0 || face_value >= 256) {
         /* this includes the OverflowError in case the long is too large */
         PyErr_SetString(PyExc_ValueError, "byte must be in range(0, 256)");
+        *value = -1;
         return 0;
     }
 

Modified: python/branches/import_unicode/Objects/memoryobject.c
==============================================================================
--- python/branches/import_unicode/Objects/memoryobject.c	(original)
+++ python/branches/import_unicode/Objects/memoryobject.c	Fri Jul 30 01:19:37 2010
@@ -624,7 +624,7 @@
 static int
 memory_ass_sub(PyMemoryViewObject *self, PyObject *key, PyObject *value)
 {
-    Py_ssize_t start, len, bytelen, i;
+    Py_ssize_t start, len, bytelen;
     Py_buffer srcview;
     Py_buffer *view = &(self->view);
     char *srcbuf, *destbuf;
@@ -694,16 +694,8 @@
     if (destbuf + bytelen < srcbuf || srcbuf + bytelen < destbuf)
         /* No overlapping */
         memcpy(destbuf, srcbuf, bytelen);
-    else if (destbuf < srcbuf) {
-        /* Copy in ascending order */
-        for (i = 0; i < bytelen; i++)
-            destbuf[i] = srcbuf[i];
-    }
-    else {
-        /* Copy in descencing order */
-        for (i = bytelen - 1; i >= 0; i--)
-            destbuf[i] = srcbuf[i];
-    }
+    else
+        memmove(destbuf, srcbuf, bytelen);
 
     PyBuffer_Release(&srcview);
     return 0;

Modified: python/branches/import_unicode/Objects/object.c
==============================================================================
--- python/branches/import_unicode/Objects/object.c	(original)
+++ python/branches/import_unicode/Objects/object.c	Fri Jul 30 01:19:37 2010
@@ -260,15 +260,10 @@
     return PyObject_INIT_VAR(op, tp, nitems);
 }
 
-/* Implementation of PyObject_Print with recursion checking */
-static int
-internal_print(PyObject *op, FILE *fp, int flags, int nesting)
+int
+PyObject_Print(PyObject *op, FILE *fp, int flags)
 {
     int ret = 0;
-    if (nesting > 10) {
-        PyErr_SetString(PyExc_RuntimeError, "print recursion");
-        return -1;
-    }
     if (PyErr_CheckSignals())
         return -1;
 #ifdef USE_STACKCHECK
@@ -335,12 +330,6 @@
     return ret;
 }
 
-int
-PyObject_Print(PyObject *op, FILE *fp, int flags)
-{
-    return internal_print(op, fp, flags, 0);
-}
-
 /* For debugging convenience.  Set a breakpoint here and call it from your DLL */
 void
 _Py_BreakPoint(void)

Modified: python/branches/import_unicode/Objects/stringlib/formatter.h
==============================================================================
--- python/branches/import_unicode/Objects/stringlib/formatter.h	(original)
+++ python/branches/import_unicode/Objects/stringlib/formatter.h	Fri Jul 30 01:19:37 2010
@@ -776,14 +776,6 @@
             goto done;
         }
 
-        /* Error to specify a comma. */
-        if (format->thousands_separators) {
-            PyErr_SetString(PyExc_ValueError,
-                            "Thousands separators not allowed with integer"
-                            " format specifier 'c'");
-            goto done;
-        }
-
         /* taken from unicodeobject.c formatchar() */
         /* Integer input truncated to a character */
 /* XXX: won't work for int */

Modified: python/branches/import_unicode/Objects/unicodeobject.c
==============================================================================
--- python/branches/import_unicode/Objects/unicodeobject.c	(original)
+++ python/branches/import_unicode/Objects/unicodeobject.c	Fri Jul 30 01:19:37 2010
@@ -307,8 +307,7 @@
   reset:
     /* Reset the object caches */
     if (unicode->defenc) {
-        Py_DECREF(unicode->defenc);
-        unicode->defenc = NULL;
+        Py_CLEAR(unicode->defenc);
     }
     unicode->hash = -1;
 
@@ -427,8 +426,7 @@
             unicode->length = 0;
         }
         if (unicode->defenc) {
-            Py_DECREF(unicode->defenc);
-            unicode->defenc = NULL;
+            Py_CLEAR(unicode->defenc);
         }
         /* Add to free list */
         *(PyUnicodeObject **)unicode = free_list;
@@ -9243,7 +9241,7 @@
                 else if (c >= '0' && c <= '9') {
                     prec = c - '0';
                     while (--fmtcnt >= 0) {
-                        c = Py_CHARMASK(*fmt++);
+                        c = *fmt++;
                         if (c < '0' || c > '9')
                             break;
                         if ((prec*10) / 10 != prec) {

Modified: python/branches/import_unicode/PC/config.c
==============================================================================
--- python/branches/import_unicode/PC/config.c	(original)
+++ python/branches/import_unicode/PC/config.c	Fri Jul 30 01:19:37 2010
@@ -43,7 +43,7 @@
 extern PyObject* PyInit_parser(void);
 extern PyObject* PyInit_winreg(void);
 extern PyObject* PyInit__struct(void);
-extern PyObject* PyInit_datetime(void);
+extern PyObject* PyInit__datetime(void);
 extern PyObject* PyInit__functools(void);
 extern PyObject* PyInit__json(void);
 extern PyObject* PyInit_zlib(void);
@@ -116,7 +116,7 @@
     {"parser", PyInit_parser},
     {"winreg", PyInit_winreg},
     {"_struct", PyInit__struct},
-    {"datetime", PyInit_datetime},
+    {"_datetime", PyInit__datetime},
     {"_functools", PyInit__functools},
     {"_json", PyInit__json},
 

Modified: python/branches/import_unicode/PCbuild/pythoncore.vcproj
==============================================================================
--- python/branches/import_unicode/PCbuild/pythoncore.vcproj	(original)
+++ python/branches/import_unicode/PCbuild/pythoncore.vcproj	Fri Jul 30 01:19:37 2010
@@ -1068,7 +1068,7 @@
 				>
 			</File>
 			<File
-				RelativePath="..\Modules\datetimemodule.c"
+				RelativePath="..\Modules\_datetimemodule.c"
 				>
 			</File>
 			<File

Modified: python/branches/import_unicode/Python/ceval.c
==============================================================================
--- python/branches/import_unicode/Python/ceval.c	(original)
+++ python/branches/import_unicode/Python/ceval.c	Fri Jul 30 01:19:37 2010
@@ -3457,7 +3457,8 @@
             return 1;
         }
         Py_DECREF(w);
-        PyErr_SetString(PyExc_ValueError, "too many values to unpack");
+        PyErr_Format(PyExc_ValueError, "too many values to unpack "
+                     "(expected %d)", argcnt);
         goto Error;
     }
 

Modified: python/branches/import_unicode/Python/getargs.c
==============================================================================
--- python/branches/import_unicode/Python/getargs.c	(original)
+++ python/branches/import_unicode/Python/getargs.c	Fri Jul 30 01:19:37 2010
@@ -1246,13 +1246,15 @@
             PyErr_Clear();
             return converterr("read-write buffer", arg, msgbuf, bufsize);
         }
+        if (!PyBuffer_IsContiguous((Py_buffer*)p, 'C')) {
+            PyBuffer_Release((Py_buffer*)p);
+            return converterr("contiguous buffer", arg, msgbuf, bufsize);
+        }
         if (addcleanup(p, freelist, cleanup_buffer)) {
             return converterr(
                 "(cleanup problem)",
                 arg, msgbuf, bufsize);
         }
-        if (!PyBuffer_IsContiguous((Py_buffer*)p, 'C'))
-            return converterr("contiguous buffer", arg, msgbuf, bufsize);
         break;
     }
 
@@ -1274,41 +1276,26 @@
 
     *errmsg = NULL;
     *p = NULL;
-    if (pb == NULL ||
-        pb->bf_getbuffer == NULL ||
-        pb->bf_releasebuffer != NULL) {
-        *errmsg = "bytes or read-only buffer";
+    if (pb != NULL && pb->bf_releasebuffer != NULL) {
+        *errmsg = "read-only pinned buffer";
         return -1;
     }
 
-    if (PyObject_GetBuffer(arg, &view, PyBUF_SIMPLE) != 0) {
-        *errmsg = "bytes or single-segment read-only buffer";
+    if (getbuffer(arg, &view, errmsg) < 0)
         return -1;
-    }
     count = view.len;
     *p = view.buf;
     PyBuffer_Release(&view);
     return count;
 }
 
-/* XXX for 3.x, getbuffer and convertbuffer can probably
-   be merged again. */
 static int
 getbuffer(PyObject *arg, Py_buffer *view, char **errmsg)
 {
-    PyBufferProcs *pb = Py_TYPE(arg)->tp_as_buffer;
-    if (pb == NULL) {
+    if (PyObject_GetBuffer(arg, view, PyBUF_SIMPLE) != 0) {
         *errmsg = "bytes or buffer";
         return -1;
     }
-    if (pb->bf_getbuffer == NULL) {
-        *errmsg = "convertible to a buffer";
-        return -1;
-    }
-    if (PyObject_GetBuffer(arg, view, 0) < 0) {
-        *errmsg = "convertible to a buffer";
-        return -1;
-    }
     if (!PyBuffer_IsContiguous(view, 'C')) {
         PyBuffer_Release(view);
         *errmsg = "contiguous buffer";

Modified: python/branches/import_unicode/Python/graminit.c
==============================================================================
--- python/branches/import_unicode/Python/graminit.c	(original)
+++ python/branches/import_unicode/Python/graminit.c	Fri Jul 30 01:19:37 2010
@@ -180,8 +180,8 @@
     {24, 9},
 };
 static arc arcs_8_5[4] = {
-    {28, 1},
-    {31, 2},
+    {28, 10},
+    {31, 11},
     {32, 3},
     {0, 5},
 };
@@ -190,7 +190,7 @@
     {0, 6},
 };
 static arc arcs_8_7[2] = {
-    {28, 10},
+    {28, 12},
     {32, 3},
 };
 static arc arcs_8_8[1] = {
@@ -201,14 +201,40 @@
     {0, 9},
 };
 static arc arcs_8_10[3] = {
-    {30, 7},
-    {29, 11},
+    {30, 5},
+    {29, 4},
     {0, 10},
 };
-static arc arcs_8_11[1] = {
+static arc arcs_8_11[3] = {
+    {28, 13},
+    {30, 14},
+    {0, 11},
+};
+static arc arcs_8_12[3] = {
+    {30, 7},
+    {29, 15},
+    {0, 12},
+};
+static arc arcs_8_13[2] = {
+    {30, 14},
+    {0, 13},
+};
+static arc arcs_8_14[2] = {
+    {28, 16},
+    {32, 3},
+};
+static arc arcs_8_15[1] = {
     {24, 6},
 };
-static state states_8[12] = {
+static arc arcs_8_16[3] = {
+    {30, 14},
+    {29, 17},
+    {0, 16},
+};
+static arc arcs_8_17[1] = {
+    {24, 13},
+};
+static state states_8[18] = {
     {3, arcs_8_0},
     {3, arcs_8_1},
     {3, arcs_8_2},
@@ -220,7 +246,13 @@
     {1, arcs_8_8},
     {2, arcs_8_9},
     {3, arcs_8_10},
-    {1, arcs_8_11},
+    {3, arcs_8_11},
+    {3, arcs_8_12},
+    {2, arcs_8_13},
+    {2, arcs_8_14},
+    {1, arcs_8_15},
+    {3, arcs_8_16},
+    {1, arcs_8_17},
 };
 static arc arcs_9_0[1] = {
     {21, 1},
@@ -263,8 +295,8 @@
     {24, 9},
 };
 static arc arcs_10_5[4] = {
-    {34, 1},
-    {31, 2},
+    {34, 10},
+    {31, 11},
     {32, 3},
     {0, 5},
 };
@@ -273,7 +305,7 @@
     {0, 6},
 };
 static arc arcs_10_7[2] = {
-    {34, 10},
+    {34, 12},
     {32, 3},
 };
 static arc arcs_10_8[1] = {
@@ -284,14 +316,40 @@
     {0, 9},
 };
 static arc arcs_10_10[3] = {
-    {30, 7},
-    {29, 11},
+    {30, 5},
+    {29, 4},
     {0, 10},
 };
-static arc arcs_10_11[1] = {
+static arc arcs_10_11[3] = {
+    {34, 13},
+    {30, 14},
+    {0, 11},
+};
+static arc arcs_10_12[3] = {
+    {30, 7},
+    {29, 15},
+    {0, 12},
+};
+static arc arcs_10_13[2] = {
+    {30, 14},
+    {0, 13},
+};
+static arc arcs_10_14[2] = {
+    {34, 16},
+    {32, 3},
+};
+static arc arcs_10_15[1] = {
     {24, 6},
 };
-static state states_10[12] = {
+static arc arcs_10_16[3] = {
+    {30, 14},
+    {29, 17},
+    {0, 16},
+};
+static arc arcs_10_17[1] = {
+    {24, 13},
+};
+static state states_10[18] = {
     {3, arcs_10_0},
     {3, arcs_10_1},
     {3, arcs_10_2},
@@ -303,7 +361,13 @@
     {1, arcs_10_8},
     {2, arcs_10_9},
     {3, arcs_10_10},
-    {1, arcs_10_11},
+    {3, arcs_10_11},
+    {3, arcs_10_12},
+    {2, arcs_10_13},
+    {2, arcs_10_14},
+    {1, arcs_10_15},
+    {3, arcs_10_16},
+    {1, arcs_10_17},
 };
 static arc arcs_11_0[1] = {
     {21, 1},
@@ -1755,11 +1819,11 @@
      "\000\000\020\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000"},
     {263, "parameters", 0, 4, states_7,
      "\000\040\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000"},
-    {264, "typedargslist", 0, 12, states_8,
+    {264, "typedargslist", 0, 18, states_8,
      "\000\000\040\200\001\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000"},
     {265, "tfpdef", 0, 4, states_9,
      "\000\000\040\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000"},
-    {266, "varargslist", 0, 12, states_10,
+    {266, "varargslist", 0, 18, states_10,
      "\000\000\040\200\001\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000"},
     {267, "vfpdef", 0, 2, states_11,
      "\000\000\040\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000"},

Modified: python/branches/import_unicode/Tools/gdb/libpython.py
==============================================================================
--- python/branches/import_unicode/Tools/gdb/libpython.py	(original)
+++ python/branches/import_unicode/Tools/gdb/libpython.py	Fri Jul 30 01:19:37 2010
@@ -43,6 +43,7 @@
 from __future__ import with_statement
 import gdb
 import locale
+import sys
 
 # Look up the gdb.Type for some standard types:
 _type_char_ptr = gdb.lookup_type('char').pointer() # char*
@@ -55,7 +56,6 @@
 
 Py_TPFLAGS_HEAPTYPE = (1L << 9)
 
-Py_TPFLAGS_INT_SUBCLASS      = (1L << 23)
 Py_TPFLAGS_LONG_SUBCLASS     = (1L << 24)
 Py_TPFLAGS_LIST_SUBCLASS     = (1L << 25)
 Py_TPFLAGS_TUPLE_SUBCLASS    = (1L << 26)
@@ -312,8 +312,6 @@
         if tp_flags & Py_TPFLAGS_HEAPTYPE:
             return HeapTypeObjectPtr
 
-        if tp_flags & Py_TPFLAGS_INT_SUBCLASS:
-            return PyIntObjectPtr
         if tp_flags & Py_TPFLAGS_LONG_SUBCLASS:
             return PyLongObjectPtr
         if tp_flags & Py_TPFLAGS_LIST_SUBCLASS:

Modified: python/branches/import_unicode/Tools/i18n/msgfmt.py
==============================================================================
--- python/branches/import_unicode/Tools/i18n/msgfmt.py	(original)
+++ python/branches/import_unicode/Tools/i18n/msgfmt.py	Fri Jul 30 01:19:37 2010
@@ -165,7 +165,7 @@
             section = STR
             if l.startswith('msgstr['):
                 if not is_plural:
-                    print(sys.stderr, 'plural without msgid_plural on %s:%d' % (infile, lno),
+                    print('plural without msgid_plural on %s:%d' % (infile, lno),
                           file=sys.stderr)
                     sys.exit(1)
                 l = l.split(']', 1)[1]
@@ -173,7 +173,7 @@
                     msgstr += b'\0' # Separator of the various plural forms
             else:
                 if is_plural:
-                    print(sys.stderr, 'indexed msgstr required for plural on  %s:%d' % (infile, lno),
+                    print('indexed msgstr required for plural on  %s:%d' % (infile, lno),
                           file=sys.stderr)
                     sys.exit(1)
                 l = l[6:]

Modified: python/branches/import_unicode/configure
==============================================================================
--- python/branches/import_unicode/configure	(original)
+++ python/branches/import_unicode/configure	Fri Jul 30 01:19:37 2010
@@ -1929,11 +1929,11 @@
        cat confdefs.h - <<_ACEOF >conftest.$ac_ext
 /* end confdefs.h.  */
 $ac_includes_default
+	     enum { N = $2 / 2 - 1 };
 int
 main ()
 {
-static int test_array [1 - 2 * !(enum { N = $2 / 2 - 1 };
-	     0 < ($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1))];
+static int test_array [1 - 2 * !(0 < ($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1))];
 test_array [0] = 0
 
   ;
@@ -1944,11 +1944,11 @@
   cat confdefs.h - <<_ACEOF >conftest.$ac_ext
 /* end confdefs.h.  */
 $ac_includes_default
+	        enum { N = $2 / 2 - 1 };
 int
 main ()
 {
-static int test_array [1 - 2 * !(enum { N = $2 / 2 - 1 };
-		($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1)
+static int test_array [1 - 2 * !(($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 1)
 		 < ($ac_type) ((((($ac_type) 1 << N) << N) - 1) * 2 + 2))];
 test_array [0] = 0
 
@@ -3158,9 +3158,12 @@
 (it is also a good idea to do 'make clean' before compiling)" "$LINENO" 5
 fi
 
-# If the user set CFLAGS, use this instead of the automatically
-# determined setting
-preset_cflags="$CFLAGS"
+# Don't let AC_PROG_CC set the default CFLAGS. It normally sets -g -O2
+# when the compiler supports them, but we don't always want -O2, and
+# we set -g later.
+if test -z "$CFLAGS"; then
+        CFLAGS=
+fi
 ac_ext=c
 ac_cpp='$CPP $CPPFLAGS'
 ac_compile='$CC -c $CFLAGS $CPPFLAGS conftest.$ac_ext >&5'
@@ -3952,10 +3955,6 @@
 ac_link='$CC -o conftest$ac_exeext $CFLAGS $CPPFLAGS $LDFLAGS conftest.$ac_ext $LIBS >&5'
 ac_compiler_gnu=$ac_cv_c_compiler_gnu
 
-if test ! -z "$preset_cflags"
-then
-	CFLAGS=$preset_cflags
-fi
 
 
 

Modified: python/branches/import_unicode/configure.in
==============================================================================
--- python/branches/import_unicode/configure.in	(original)
+++ python/branches/import_unicode/configure.in	Fri Jul 30 01:19:37 2010
@@ -461,14 +461,13 @@
 (it is also a good idea to do 'make clean' before compiling)])
 fi
 
-# If the user set CFLAGS, use this instead of the automatically
-# determined setting
-preset_cflags="$CFLAGS"
-AC_PROG_CC
-if test ! -z "$preset_cflags"
-then
-	CFLAGS=$preset_cflags
+# Don't let AC_PROG_CC set the default CFLAGS. It normally sets -g -O2
+# when the compiler supports them, but we don't always want -O2, and
+# we set -g later.
+if test -z "$CFLAGS"; then
+        CFLAGS=
 fi
+AC_PROG_CC
 
 AC_SUBST(CXX)
 AC_SUBST(MAINCC)

Modified: python/branches/import_unicode/setup.py
==============================================================================
--- python/branches/import_unicode/setup.py	(original)
+++ python/branches/import_unicode/setup.py	Fri Jul 30 01:19:37 2010
@@ -222,7 +222,7 @@
         if compiler is not None:
             (ccshared,cflags) = sysconfig.get_config_vars('CCSHARED','CFLAGS')
             args['compiler_so'] = compiler + ' ' + ccshared + ' ' + cflags
-        self.compiler_obj.set_executables(**args)
+        self.compiler.set_executables(**args)
 
         build_ext.build_extensions(self)
 
@@ -343,8 +343,8 @@
 
     def detect_modules(self):
         # Ensure that /usr/local is always used
-        add_dir_to_list(self.compiler_obj.library_dirs, '/usr/local/lib')
-        add_dir_to_list(self.compiler_obj.include_dirs, '/usr/local/include')
+        add_dir_to_list(self.compiler.library_dirs, '/usr/local/lib')
+        add_dir_to_list(self.compiler.include_dirs, '/usr/local/include')
 
         # Add paths specified in the environment variables LDFLAGS and
         # CPPFLAGS for header and library files.
@@ -353,9 +353,9 @@
         # the environment variable is not set even though the value were passed
         # into configure and stored in the Makefile (issue found on OS X 10.3).
         for env_var, arg_name, dir_list in (
-                ('LDFLAGS', '-R', self.compiler_obj.runtime_library_dirs),
-                ('LDFLAGS', '-L', self.compiler_obj.library_dirs),
-                ('CPPFLAGS', '-I', self.compiler_obj.include_dirs)):
+                ('LDFLAGS', '-R', self.compiler.runtime_library_dirs),
+                ('LDFLAGS', '-L', self.compiler.library_dirs),
+                ('CPPFLAGS', '-I', self.compiler.include_dirs)):
             env_val = sysconfig.get_config_var(env_var)
             if env_val:
                 # To prevent optparse from raising an exception about any
@@ -381,19 +381,19 @@
                         add_dir_to_list(dir_list, directory)
 
         if os.path.normpath(sys.prefix) != '/usr':
-            add_dir_to_list(self.compiler_obj.library_dirs,
+            add_dir_to_list(self.compiler.library_dirs,
                             sysconfig.get_config_var("LIBDIR"))
-            add_dir_to_list(self.compiler_obj.include_dirs,
+            add_dir_to_list(self.compiler.include_dirs,
                             sysconfig.get_config_var("INCLUDEDIR"))
 
         # lib_dirs and inc_dirs are used to search for files;
         # if a file is found in one of those directories, it can
         # be assumed that no additional -I,-L directives are needed.
-        lib_dirs = self.compiler_obj.library_dirs + [
+        lib_dirs = self.compiler.library_dirs + [
             '/lib64', '/usr/lib64',
             '/lib', '/usr/lib',
             ]
-        inc_dirs = self.compiler_obj.include_dirs + ['/usr/include']
+        inc_dirs = self.compiler.include_dirs + ['/usr/include']
         exts = []
         missing = []
 
@@ -452,7 +452,7 @@
         # time operations and variables
         exts.append( Extension('time', ['timemodule.c', '_time.c'],
                                libraries=math_libs) )
-        exts.append( Extension('datetime', ['datetimemodule.c', '_time.c'],
+        exts.append( Extension('_datetime', ['_datetimemodule.c', '_time.c'],
                                libraries=math_libs) )
         # fast iterator tools implemented in C
         exts.append( Extension("itertools", ["itertoolsmodule.c"]) )
@@ -525,7 +525,7 @@
         exts.append( Extension('audioop', ['audioop.c']) )
 
         # readline
-        do_readline = self.compiler_obj.find_library_file(lib_dirs, 'readline')
+        do_readline = self.compiler.find_library_file(lib_dirs, 'readline')
         readline_termcap_library = ""
         curses_library = ""
         # Determine if readline is already linked against curses or tinfo.
@@ -534,28 +534,29 @@
             tmpfile = os.path.join(self.build_temp, 'readline_termcap_lib')
             if not os.path.exists(self.build_temp):
                 os.makedirs(self.build_temp)
-            os.system("ldd %s > %s" % (do_readline, tmpfile))
-            fp = open(tmpfile)
-            for ln in fp:
-                if 'curses' in ln:
-                    readline_termcap_library = re.sub(
-                        r'.*lib(n?cursesw?)\.so.*', r'\1', ln
-                    ).rstrip()
-                    break
-                if 'tinfo' in ln: # termcap interface split out from ncurses
-                    readline_termcap_library = 'tinfo'
-                    break
-            fp.close()
+            ret = os.system("ldd %s > %s" % (do_readline, tmpfile))
+            if ret >> 8 == 0:
+                fp = open(tmpfile)
+                for ln in fp:
+                    if 'curses' in ln:
+                        readline_termcap_library = re.sub(
+                            r'.*lib(n?cursesw?)\.so.*', r'\1', ln
+                        ).rstrip()
+                        break
+                    if 'tinfo' in ln: # termcap interface split out from ncurses
+                        readline_termcap_library = 'tinfo'
+                        break
+                fp.close()
             os.unlink(tmpfile)
         # Issue 7384: If readline is already linked against curses,
         # use the same library for the readline and curses modules.
         if 'curses' in readline_termcap_library:
             curses_library = readline_termcap_library
-        elif self.compiler_obj.find_library_file(lib_dirs, 'ncursesw'):
+        elif self.compiler.find_library_file(lib_dirs, 'ncursesw'):
             curses_library = 'ncursesw'
-        elif self.compiler_obj.find_library_file(lib_dirs, 'ncurses'):
+        elif self.compiler.find_library_file(lib_dirs, 'ncurses'):
             curses_library = 'ncurses'
-        elif self.compiler_obj.find_library_file(lib_dirs, 'curses'):
+        elif self.compiler.find_library_file(lib_dirs, 'curses'):
             curses_library = 'curses'
 
         if platform == 'darwin':
@@ -585,7 +586,7 @@
                 pass # Issue 7384: Already linked against curses or tinfo.
             elif curses_library:
                 readline_libs.append(curses_library)
-            elif self.compiler_obj.find_library_file(lib_dirs +
+            elif self.compiler.find_library_file(lib_dirs +
                                                      ['/usr/lib/termcap'],
                                                      'termcap'):
                 readline_libs.append('termcap')
@@ -598,7 +599,7 @@
 
         # crypt module.
 
-        if self.compiler_obj.find_library_file(lib_dirs, 'crypt'):
+        if self.compiler.find_library_file(lib_dirs, 'crypt'):
             libs = ['crypt']
         else:
             libs = []
@@ -626,7 +627,7 @@
                                ['/usr/kerberos/include'])
             if krb5_h:
                 ssl_incs += krb5_h
-        ssl_libs = find_library_file(self.compiler_obj, 'ssl',lib_dirs,
+        ssl_libs = find_library_file(self.compiler, 'ssl',lib_dirs,
                                      ['/usr/local/ssl/lib',
                                       '/usr/contrib/ssl/lib/'
                                      ] )
@@ -951,7 +952,7 @@
                 os.path.join(sqlite_incdir, '..', '..', 'lib64'),
                 os.path.join(sqlite_incdir, '..', '..', 'lib'),
             ]
-            sqlite_libfile = self.compiler_obj.find_library_file(
+            sqlite_libfile = self.compiler.find_library_file(
                                 sqlite_dirs_to_check + lib_dirs, 'sqlite3')
             if sqlite_libfile:
                 sqlite_libdir = [os.path.abspath(os.path.dirname(sqlite_libfile))]
@@ -1013,7 +1014,7 @@
                 if cand == "ndbm":
                     if find_file("ndbm.h", inc_dirs, []) is not None:
                         # Some systems have -lndbm, others don't
-                        if self.compiler_obj.find_library_file(lib_dirs,
+                        if self.compiler.find_library_file(lib_dirs,
                                                                'ndbm'):
                             ndbm_libs = ['ndbm']
                         else:
@@ -1027,9 +1028,9 @@
                         break
 
                 elif cand == "gdbm":
-                    if self.compiler_obj.find_library_file(lib_dirs, 'gdbm'):
+                    if self.compiler.find_library_file(lib_dirs, 'gdbm'):
                         gdbm_libs = ['gdbm']
-                        if self.compiler_obj.find_library_file(lib_dirs,
+                        if self.compiler.find_library_file(lib_dirs,
                                                                'gdbm_compat'):
                             gdbm_libs.append('gdbm_compat')
                         if find_file("gdbm/ndbm.h", inc_dirs, []) is not None:
@@ -1070,7 +1071,7 @@
 
         # Anthony Baxter's gdbm module.  GNU dbm(3) will require -lgdbm:
         if ('gdbm' in dbm_order and
-            self.compiler_obj.find_library_file(lib_dirs, 'gdbm')):
+            self.compiler.find_library_file(lib_dirs, 'gdbm')):
             exts.append( Extension('_gdbm', ['_gdbmmodule.c'],
                                    libraries = ['gdbm'] ) )
         else:
@@ -1086,7 +1087,7 @@
             # Sun yellow pages. Some systems have the functions in libc.
             if (platform not in ['cygwin', 'qnx6'] and
                 find_file('rpcsvc/yp_prot.h', inc_dirs, []) is not None):
-                if (self.compiler_obj.find_library_file(lib_dirs, 'nsl')):
+                if (self.compiler.find_library_file(lib_dirs, 'nsl')):
                     libs = ['nsl']
                 else:
                     libs = []
@@ -1111,9 +1112,9 @@
         elif curses_library == 'curses' and platform != 'darwin':
                 # OSX has an old Berkeley curses, not good enough for
                 # the _curses module.
-            if (self.compiler_obj.find_library_file(lib_dirs, 'terminfo')):
+            if (self.compiler.find_library_file(lib_dirs, 'terminfo')):
                 curses_libs = ['curses', 'terminfo']
-            elif (self.compiler_obj.find_library_file(lib_dirs, 'termcap')):
+            elif (self.compiler.find_library_file(lib_dirs, 'termcap')):
                 curses_libs = ['curses', 'termcap']
             else:
                 curses_libs = ['curses']
@@ -1125,7 +1126,7 @@
 
         # If the curses module is enabled, check for the panel module
         if (module_enabled(exts, '_curses') and
-            self.compiler_obj.find_library_file(lib_dirs, panel_library)):
+            self.compiler.find_library_file(lib_dirs, panel_library)):
             exts.append( Extension('_curses_panel', ['_curses_panel.c'],
                                    libraries = [panel_library] + curses_libs) )
         else:
@@ -1158,7 +1159,7 @@
                     version = line.split()[2]
                     break
             if version >= version_req:
-                if (self.compiler_obj.find_library_file(lib_dirs, 'z')):
+                if (self.compiler.find_library_file(lib_dirs, 'z')):
                     if sys.platform == "darwin":
                         zlib_extra_link_args = ('-Wl,-search_paths_first',)
                     else:
@@ -1190,7 +1191,7 @@
                                extra_link_args = extra_link_args) )
 
         # Gustavo Niemeyer's bz2 module.
-        if (self.compiler_obj.find_library_file(lib_dirs, 'bz2')):
+        if (self.compiler.find_library_file(lib_dirs, 'bz2')):
             if sys.platform == "darwin":
                 bz2_extra_link_args = ('-Wl,-search_paths_first',)
             else:
@@ -1451,9 +1452,9 @@
         tcllib = tklib = tcl_includes = tk_includes = None
         for version in ['8.6', '86', '8.5', '85', '8.4', '84', '8.3', '83',
                         '8.2', '82', '8.1', '81', '8.0', '80']:
-            tklib = self.compiler_obj.find_library_file(lib_dirs,
+            tklib = self.compiler.find_library_file(lib_dirs,
                                                         'tk' + version)
-            tcllib = self.compiler_obj.find_library_file(lib_dirs,
+            tcllib = self.compiler.find_library_file(lib_dirs,
                                                          'tcl' + version)
             if tklib and tcllib:
                 # Exit the loop when we've found the Tcl/Tk libraries
@@ -1512,11 +1513,11 @@
                 return
 
         # Check for BLT extension
-        if self.compiler_obj.find_library_file(lib_dirs + added_lib_dirs,
+        if self.compiler.find_library_file(lib_dirs + added_lib_dirs,
                                                'BLT8.0'):
             defs.append( ('WITH_BLT', 1) )
             libs.append('BLT8.0')
-        elif self.compiler_obj.find_library_file(lib_dirs + added_lib_dirs,
+        elif self.compiler.find_library_file(lib_dirs + added_lib_dirs,
                                                 'BLT'):
             defs.append( ('WITH_BLT', 1) )
             libs.append('BLT')
@@ -1571,7 +1572,7 @@
                              ]]
 
         # Add .S (preprocessed assembly) to C compiler source extensions.
-        self.compiler_obj.src_extensions.append('.S')
+        self.compiler.src_extensions.append('.S')
 
         include_dirs = [os.path.join(ffi_srcdir, 'include'),
                         os.path.join(ffi_srcdir, 'powerpc')]
@@ -1616,7 +1617,7 @@
                 exec(f.read(), globals(), fficonfig)
 
             # Add .S (preprocessed assembly) to C compiler source extensions.
-            self.compiler_obj.src_extensions.append('.S')
+            self.compiler.src_extensions.append('.S')
 
             include_dirs = [os.path.join(ffi_builddir, 'include'),
                             ffi_builddir,
@@ -1698,7 +1699,7 @@
         ffi_lib = None
         if ffi_inc is not None:
             for lib_name in ('ffi_convenience', 'ffi_pic', 'ffi'):
-                if (self.compiler_obj.find_library_file(lib_dirs, lib_name)):
+                if (self.compiler.find_library_file(lib_dirs, lib_name)):
                     ffi_lib = lib_name
                     break
 

From python-checkins at python.org  Fri Jul 30 02:47:07 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 02:47:07 +0200 (CEST)
Subject: [Python-checkins] r83248 -
	python/branches/import_unicode/Modules/zipimport.c
Message-ID: <20100730004707.157C7EE9C3@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 02:47:06 2010
New Revision: 83248

Log:
zipimport: fix for Visual Studio

Define all variables at the beginning of the block to be ISO C89
compliant.

Modified:
   python/branches/import_unicode/Modules/zipimport.c

Modified: python/branches/import_unicode/Modules/zipimport.c
==============================================================================
--- python/branches/import_unicode/Modules/zipimport.c	(original)
+++ python/branches/import_unicode/Modules/zipimport.c	Fri Jul 30 02:47:06 2010
@@ -514,10 +514,11 @@
 
     toc_entry = PyDict_GetItemString(self->files, path);
     if (toc_entry != NULL) {
+        PyObject *res;
         PyObject *bytes = get_data(self->archive, toc_entry);
         if (bytes == NULL)
             return NULL;
-        PyObject *res = PyUnicode_DecodeFSDefault(PyBytes_AsString(bytes));
+        res = PyUnicode_DecodeFSDefault(PyBytes_AsString(bytes));
         Py_XDECREF(bytes);
         return res;
     }

From alexander.belopolsky at gmail.com  Fri Jul 30 03:21:56 2010
From: alexander.belopolsky at gmail.com (Alexander Belopolsky)
Date: Thu, 29 Jul 2010 21:21:56 -0400
Subject: [Python-checkins] r83248 -
	python/branches/import_unicode/Modules/zipimport.c
In-Reply-To: <20100730004707.157C7EE9C3@mail.python.org>
References: <20100730004707.157C7EE9C3@mail.python.org>
Message-ID: <AANLkTincXMRiZi+raK07BGoWF7=fBjUxAi7uT4-3NPN7@mail.gmail.com>

On Thu, Jul 29, 2010 at 8:47 PM, victor.stinner
<python-checkins at python.org> wrote:
..
> + ? ? ? ?res = PyUnicode_DecodeFSDefault(PyBytes_AsString(bytes));

Shouldn't you check return value of PyBytes_AsString before passing it
to PyUnicode_DecodeFSDefault?

From python-checkins at python.org  Fri Jul 30 03:41:57 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 03:41:57 +0200 (CEST)
Subject: [Python-checkins] r83249 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100730014157.DBF81EE98C@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 03:41:57 2010
New Revision: 83249

Log:
NullImporter constructor uses PyUnicode_AsWideChar()

instead of _PyUnicode_AsString(), and GetFileAttributesW() instead of
instead of GetFileAttributesA(), to be fully unicode compliant.

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul 30 03:41:57 2010
@@ -3828,14 +3828,16 @@
         }
 #else /* MS_WINDOWS */
         DWORD rv;
-        char *path;
-        /* FIXME: use PyUnicode_AsWideChar() and GetFileAttributesW() */
-        path = _PyUnicode_AsString(pathobj);
+        wchar_t path[MAXPATHLEN+1];
+        Py_ssize_t len;
+        len = PyUnicode_AsWideChar(pathobj, path, sizeof(path));
+        if (len == -1)
+            return -1;
         /* see issue1293 and issue3677:
          * stat() on Windows doesn't recognise paths like
          * "e:\\shared\\" and "\\\\whiterab-c2znlh\\shared" as dirs.
          */
-        rv = GetFileAttributesA(path);
+        rv = GetFileAttributesW(path);
         if (rv != INVALID_FILE_ATTRIBUTES) {
             /* it exists */
             if (rv & FILE_ATTRIBUTE_DIRECTORY) {

From python-checkins at python.org  Fri Jul 30 03:47:30 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 03:47:30 +0200 (CEST)
Subject: [Python-checkins] r83250 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100730014730.0C1F4EE982@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 03:47:29 2010
New Revision: 83250

Log:
fix a compiler warning in NullImporter_init()

PyUnicode_AsWideChar() first argument is a PyUnicodeObject*

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul 30 03:47:29 2010
@@ -3830,7 +3830,8 @@
         DWORD rv;
         wchar_t path[MAXPATHLEN+1];
         Py_ssize_t len;
-        len = PyUnicode_AsWideChar(pathobj, path, sizeof(path));
+        len = PyUnicode_AsWideChar((PyUnicodeObject*)pathobj,
+                                   path, sizeof(path));
         if (len == -1)
             return -1;
         /* see issue1293 and issue3677:

From python-checkins at python.org  Fri Jul 30 04:04:11 2010
From: python-checkins at python.org (skip.montanaro)
Date: Fri, 30 Jul 2010 04:04:11 +0200 (CEST)
Subject: [Python-checkins] r83251 -
	python/branches/release27-maint/Doc/library/csv.rst
Message-ID: <20100730020411.051FAEE98C@mail.python.org>

Author: skip.montanaro
Date: Fri Jul 30 04:04:10 2010
New Revision: 83251

Log:
Make sure all files are opened in binary mode.


Modified:
   python/branches/release27-maint/Doc/library/csv.rst

Modified: python/branches/release27-maint/Doc/library/csv.rst
==============================================================================
--- python/branches/release27-maint/Doc/library/csv.rst	(original)
+++ python/branches/release27-maint/Doc/library/csv.rst	Fri Jul 30 04:04:10 2010
@@ -78,7 +78,7 @@
    A short usage example::
 
       >>> import csv
-      >>> spamReader = csv.reader(open('eggs.csv'), delimiter=' ', quotechar='|')
+      >>> spamReader = csv.reader(open('eggs.csv', 'rb'), delimiter=' ', quotechar='|')
       >>> for row in spamReader:
       ...     print ', '.join(row)
       Spam, Spam, Spam, Spam, Spam, Baked Beans
@@ -116,7 +116,7 @@
    A short usage example::
 
       >>> import csv
-      >>> spamWriter = csv.writer(open('eggs.csv', 'w'), delimiter=' ',
+      >>> spamWriter = csv.writer(open('eggs.csv', 'wb'), delimiter=' ',
       ...                         quotechar='|', quoting=csv.QUOTE_MINIMAL)
       >>> spamWriter.writerow(['Spam'] * 5 + ['Baked Beans'])
       >>> spamWriter.writerow(['Spam', 'Lovely Spam', 'Wonderful Spam'])
@@ -234,7 +234,7 @@
 
 An example for :class:`Sniffer` use::
 
-   csvfile = open("example.csv")
+   csvfile = open("example.csv", "rb")
    dialect = csv.Sniffer().sniff(csvfile.read(1024))
    csvfile.seek(0)
    reader = csv.reader(csvfile, dialect)

From python-checkins at python.org  Fri Jul 30 04:36:27 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 04:36:27 +0200 (CEST)
Subject: [Python-checkins] r83252 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100730023627.CE1EAEE99D@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 04:36:27 2010
New Revision: 83252

Log:
case_ok() uses FindFirstFileW()

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul 30 04:36:27 2010
@@ -1991,17 +1991,20 @@
 
 /* MS_WINDOWS */
 #if defined(MS_WINDOWS)
-    char *fullpath;
-    WIN32_FIND_DATA data;
+    wchar_t buffer[MAXPATHLEN+1];
+    Py_ssize_t len;
+    DWORD usize;
+    WIN32_FIND_DATAW data;
     HANDLE h;
 
     if (Py_GETENV("PYTHONCASEOK") != NULL)
         return 1;
 
-    /* FIXME: use PyUnicode_AsWideChar() and FindFirstFileW() */
-    fullpath = _PyUnicode_AsString(fullpath_obj);
+    len = PyUnicode_AsWideChar((PyUnicodeObject*)fullpath_obj, buffer, sizeof(buffer));
+    if (len == -1)
+        return 0;
 
-    h = FindFirstFile(fullpath, &data);
+    h = FindFirstFileW(buffer, &data);
     if (h == INVALID_HANDLE_VALUE) {
         PyErr_Format(PyExc_NameError,
           "Can't find file for module %.100s\n(filename %U)",
@@ -2009,7 +2012,12 @@
         return 0;
     }
     FindClose(h);
-    return strncmp(data.cFileName, name, namelen) == 0;
+
+    /* FIXME: use maybe a dynamic buffer? (namelen) */
+    usize = MultiByteToWideChar(CP_ACP, 0, name, namelen, buffer, sizeof(buffer));
+    if (usize == 0)
+        return 0;
+    return wcsncmp(data.cFileName, buffer, namelen) == 0;
 
 /* DJGPP */
 #elif defined(DJGPP)

From solipsis at pitrou.net  Fri Jul 30 04:59:45 2010
From: solipsis at pitrou.net (solipsis at pitrou.net)
Date: Fri, 30 Jul 2010 04:59:45 +0200
Subject: [Python-checkins] Daily py3k reference leaks (r83239): sum=0
Message-ID: <E1Oefp3-0006id-Tp@ap.vmr.nerim.net>

py3k results for svn r83239 (hg cset a1e852cf1172)
--------------------------------------------------



Command line was: ['./python', '-m', 'test.regrtest', '-uall', '-R', '3:3:/home/antoine/py3k/refleaks/reflogKVzT4S', '-x']

From python-checkins at python.org  Fri Jul 30 05:51:47 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 05:51:47 +0200 (CEST)
Subject: [Python-checkins] r83253 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100730035147.19DF1EE990@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 05:51:46 2010
New Revision: 83253

Log:
use _wfopen() and _wstat() in _Py_fopen() and _Py_stat()

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul 30 05:51:46 2010
@@ -2130,7 +2130,22 @@
 FILE*
 _Py_fopen(PyObject *unicode, const char *mode)
 {
-    /* FIXME: use _wfopen() on Windows */
+#ifdef MS_WINDOWS
+    wchar_t path[MAXPATHLEN+1];
+    wchar_t wmode[10];
+    Py_ssize_t len;
+    int usize;
+
+    len = PyUnicode_AsWideChar((PyUnicodeObject*)unicode, path, sizeof(path));
+    if (len == -1)
+        return NULL;
+
+    usize = MultiByteToWideChar(CP_ACP, 0, mode, -1, wmode, sizeof(wmode));
+    if (usize == 0)
+        return NULL;
+
+    return _wfopen(path, wmode);
+#else
     FILE *f;
     PyObject *bytes = PyUnicode_EncodeFSDefault(unicode);
     if (bytes == NULL) {
@@ -2141,6 +2156,7 @@
     f = fopen(PyBytes_AS_STRING(bytes), mode);
     Py_DECREF(bytes);
     return f;
+#endif
 }
 
 
@@ -2148,6 +2164,22 @@
 int
 _Py_stat(PyObject *unicode, struct stat *statbuf)
 {
+#ifdef MS_WINDOWS
+    wchar_t path[MAXPATHLEN+1];
+    Py_ssize_t len;
+    int err;
+    struct _stat wstatbuf;
+
+    len = PyUnicode_AsWideChar((PyUnicodeObject*)unicode, path, sizeof(path));
+    if (len == -1)
+        return -1;
+    err = _wstat(path, &wstatbuf);
+    if (!err) {
+        /* FIXME: copy more attributes? */
+        statbuf->st_mode = wstatbuf.st_mode;
+    }
+    return err;
+#else
     int ret;
     PyObject *bytes = PyUnicode_EncodeFSDefault(unicode);
     if (bytes == NULL) {
@@ -2158,6 +2190,7 @@
     ret = stat(PyBytes_AS_STRING(bytes), statbuf);
     Py_DECREF(bytes);
     return ret;
+#endif
 }
 
 /* Helper to look for __init__.py or __init__.py[co] in potential package */

From python-checkins at python.org  Fri Jul 30 05:51:48 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 05:51:48 +0200 (CEST)
Subject: [Python-checkins] r83254 - in python/branches/import_unicode:
	Modules/main.c Objects/fileobject.c Parser/tokenizer.c
Message-ID: <20100730035148.CD711EE9AE@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 05:51:48 2010
New Revision: 83254

Log:
Coding style: newline between } and else

Modified:
   python/branches/import_unicode/Modules/main.c
   python/branches/import_unicode/Objects/fileobject.c
   python/branches/import_unicode/Parser/tokenizer.c

Modified: python/branches/import_unicode/Modules/main.c
==============================================================================
--- python/branches/import_unicode/Modules/main.c	(original)
+++ python/branches/import_unicode/Modules/main.c	Fri Jul 30 05:51:48 2010
@@ -254,7 +254,7 @@
     }
 }
 
-static int 
+static int
 run_command(wchar_t *command, PyCompilerFlags *cf)
 {
     int sts;
@@ -749,7 +749,8 @@
             *bytes++ = c - 0xdc00;
             size--;
             continue;
-        } else {
+        }
+        else {
             wchar_t buf[2];
             buf[0] = c;
             buf[1] = 0;

Modified: python/branches/import_unicode/Objects/fileobject.c
==============================================================================
--- python/branches/import_unicode/Objects/fileobject.c	(original)
+++ python/branches/import_unicode/Objects/fileobject.c	Fri Jul 30 05:51:48 2010
@@ -57,7 +57,8 @@
         nameobj = PyUnicode_DecodeFSDefault(name);
         if (nameobj == NULL)
             PyErr_Clear();
-    } else
+    }
+    else
         nameobj = NULL;
     file = _PyFile_FromFdUnicode(fd, nameobj, mode, buffering,
                                  encoding, errors, newline, closefd);

Modified: python/branches/import_unicode/Parser/tokenizer.c
==============================================================================
--- python/branches/import_unicode/Parser/tokenizer.c	(original)
+++ python/branches/import_unicode/Parser/tokenizer.c	Fri Jul 30 05:51:48 2010
@@ -474,7 +474,8 @@
         stream = PyObject_CallMethod(io, "open", "Osis",
                                      filename, "r", -1, enc);
         Py_DECREF(filename);
-    } else
+    }
+    else
         stream = PyObject_CallMethod(io, "open", "isisOOO",
                         fileno(tok->fp), "r", -1, enc, Py_None, Py_None, Py_False);
     if (stream == NULL)

From python-checkins at python.org  Fri Jul 30 05:51:50 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 05:51:50 +0200 (CEST)
Subject: [Python-checkins] r83255 -
	python/branches/import_unicode/Modules/main.c
Message-ID: <20100730035150.73176EE9C0@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 05:51:50 2010
New Revision: 83255

Log:
run_command(): Py_XDECREF() => Py_DECREF()

Modified:
   python/branches/import_unicode/Modules/main.c

Modified: python/branches/import_unicode/Modules/main.c
==============================================================================
--- python/branches/import_unicode/Modules/main.c	(original)
+++ python/branches/import_unicode/Modules/main.c	Fri Jul 30 05:51:50 2010
@@ -271,7 +271,7 @@
         goto error;
     commandStr = PyBytes_AsString(command_bytes);
     sts = PyRun_SimpleStringFlags(commandStr, cf) != 0;
-    Py_XDECREF(command_bytes);
+    Py_DECREF(command_bytes);
     return sts;
 
 error:

From python-checkins at python.org  Fri Jul 30 05:51:51 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 05:51:51 +0200 (CEST)
Subject: [Python-checkins] r83256 -
	python/branches/import_unicode/Lib/test/test_import.py
Message-ID: <20100730035151.EEB00EE9BD@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 05:51:51 2010
New Revision: 83256

Log:
Remove useless code

Modified:
   python/branches/import_unicode/Lib/test/test_import.py

Modified: python/branches/import_unicode/Lib/test/test_import.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_import.py	(original)
+++ python/branches/import_unicode/Lib/test/test_import.py	Fri Jul 30 05:51:51 2010
@@ -296,8 +296,6 @@
             path.encode(encoding)
         except UnicodeEncodeError:
             self.skipTest('path is not encodable to {}'.format(encoding))
-        else:
-            pass
         with self.assertRaises(ImportError) as c:
             __import__(path)
         self.assertEqual("Import by filename is not supported.",

From python-checkins at python.org  Fri Jul 30 05:51:53 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 05:51:53 +0200 (CEST)
Subject: [Python-checkins] r83257 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100730035153.6288DEE9C4@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 05:51:53 2010
New Revision: 83257

Log:
remove space

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul 30 05:51:53 2010
@@ -723,7 +723,7 @@
                       "sys.modules failed");
 }
 
-static PyObject * get_sourcefile(PyObject *file);
+static PyObject *get_sourcefile(PyObject *file);
 static PyObject *make_source_pathname(PyObject *pathname);
 static PyObject *make_compiled_pathname(PyObject *pathname, int debug);
 

From python-checkins at python.org  Fri Jul 30 05:51:55 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 05:51:55 +0200 (CEST)
Subject: [Python-checkins] r83258 -
	python/branches/import_unicode/Lib/test/test_sys.py
Message-ID: <20100730035155.15E47EE9BF@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 05:51:54 2010
New Revision: 83258

Log:
use .assertIn() instead of .assert_(... in ...)

Modified:
   python/branches/import_unicode/Lib/test/test_sys.py

Modified: python/branches/import_unicode/Lib/test/test_sys.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_sys.py	(original)
+++ python/branches/import_unicode/Lib/test/test_sys.py	Fri Jul 30 05:51:54 2010
@@ -509,9 +509,10 @@
         p = subprocess.Popen([sys.executable, "-c", code], stderr=subprocess.PIPE)
         stdout, stderr = p.communicate()
         self.assertEqual(p.returncode, 1)
-        pattern = b"SyntaxError: (unicode error) 'utf8' codec can't decode byte 0xff"
-        self.assert_(pattern in stderr,
-                     "%r not in %s" % (pattern, ascii(stderr)))
+        self.assertIn(
+            b"SyntaxError: (unicode error) " \
+            b"'utf8' codec can't decode byte 0xff",
+            stderr)
 
     def test_sys_flags(self):
         self.assertTrue(sys.flags)

From victor.stinner at haypocalc.com  Fri Jul 30 05:54:19 2010
From: victor.stinner at haypocalc.com (Victor Stinner)
Date: Fri, 30 Jul 2010 05:54:19 +0200
Subject: [Python-checkins] r83248 -
	python/branches/import_unicode/Modules/zipimport.c
In-Reply-To: <AANLkTincXMRiZi+raK07BGoWF7=fBjUxAi7uT4-3NPN7@mail.gmail.com>
References: <20100730004707.157C7EE9C3@mail.python.org>
	<AANLkTincXMRiZi+raK07BGoWF7=fBjUxAi7uT4-3NPN7@mail.gmail.com>
Message-ID: <201007300554.20205.victor.stinner@haypocalc.com>

Le vendredi 30 juillet 2010 03:21:56, Alexander Belopolsky a ?crit :
> On Thu, Jul 29, 2010 at 8:47 PM, victor.stinner
> <python-checkins at python.org> wrote:
> ..
> 
> > +        res = PyUnicode_DecodeFSDefault(PyBytes_AsString(bytes));
> 
> Shouldn't you check return value of PyBytes_AsString before passing it
> to PyUnicode_DecodeFSDefault?

I would be better, but it's not important here, because bytes should be a 
PyBytesObjects (result of get_data). I didn't wrote this code and I prefer to 
not change to much code if it's possible.

Please patch zipimport.c in py3k directly.

-- 
Victor Stinner
http://www.haypocalc.com/

From python-checkins at python.org  Fri Jul 30 09:03:39 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 09:03:39 +0200 (CEST)
Subject: [Python-checkins] r83259 -
	python/branches/py3k/Doc/library/ftplib.rst
Message-ID: <20100730070339.DB58AEE998@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 09:03:39 2010
New Revision: 83259

Log:
Clarification.

Modified:
   python/branches/py3k/Doc/library/ftplib.rst

Modified: python/branches/py3k/Doc/library/ftplib.rst
==============================================================================
--- python/branches/py3k/Doc/library/ftplib.rst	(original)
+++ python/branches/py3k/Doc/library/ftplib.rst	Fri Jul 30 09:03:39 2010
@@ -231,9 +231,9 @@
    Retrieve a file or directory listing in ASCII transfer mode.  *cmd*
    should be an appropriate ``RETR`` command (see :meth:`retrbinary`) or a
    command such as ``LIST``, ``NLST`` or ``MLSD`` (usually just the string
-   ``'LIST'``).  The *callback* function is called for each line, with the
-   trailing CRLF stripped.  The default *callback* prints the line to
-   ``sys.stdout``.
+   ``'LIST'``).  The *callback* function is called for each line with a
+   string argument containing the line with the trailing CRLF stripped.
+   The default *callback* prints the line to ``sys.stdout``.
 
 
 .. method:: FTP.set_pasv(boolean)

From python-checkins at python.org  Fri Jul 30 09:14:01 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 09:14:01 +0200 (CEST)
Subject: [Python-checkins] r83260 - in python/branches/py3k: Lib/pdb.py
	Misc/NEWS
Message-ID: <20100730071401.8F806EE981@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 09:14:01 2010
New Revision: 83260

Log:
#4179: In pdb, allow "list ." as a command to return to the currently debugged line.

Modified:
   python/branches/py3k/Lib/pdb.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/pdb.py
==============================================================================
--- python/branches/py3k/Lib/pdb.py	(original)
+++ python/branches/py3k/Lib/pdb.py	Fri Jul 30 09:14:01 2010
@@ -160,6 +160,7 @@
         List source code for the current file.
         Without arguments, list 11 lines around the current line
         or continue the previous listing.
+        With . as argument, list 11 lines around the current line.
         With one argument, list 11 lines starting at that line.
         With two arguments, list the given range;
         if the second argument is less than the first, it is a count.
@@ -997,7 +998,7 @@
     def do_list(self, arg):
         self.lastcmd = 'list'
         last = None
-        if arg:
+        if arg and arg != '.':
             try:
                 x = eval(arg, {}, {})
                 if type(x) == type(()):
@@ -1012,7 +1013,7 @@
             except:
                 print('*** Error in argument:', repr(arg), file=self.stdout)
                 return
-        elif self.lineno is None:
+        elif self.lineno is None or arg == '.':
             first = max(1, self.curframe.f_lineno - 5)
         else:
             first = self.lineno + 1

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Fri Jul 30 09:14:01 2010
@@ -475,6 +475,9 @@
 Library
 -------
 
+- Issue #4179: In pdb, allow "list ." as a command to return to the
+  currently debugged line.
+
 - Issue #4108: In urllib.robotparser, if there are multiple 'User-agent: *'
   entries, consider the first one.
 

From python-checkins at python.org  Fri Jul 30 09:21:26 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 09:21:26 +0200 (CEST)
Subject: [Python-checkins] r83261 - python/branches/py3k/Lib/pdb.py
Message-ID: <20100730072126.404AEEE981@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 09:21:26 2010
New Revision: 83261

Log:
#9230: allow Pdb.checkline() to be called without a current frame, for setting breakpoints before starting debugging.

Modified:
   python/branches/py3k/Lib/pdb.py

Modified: python/branches/py3k/Lib/pdb.py
==============================================================================
--- python/branches/py3k/Lib/pdb.py	(original)
+++ python/branches/py3k/Lib/pdb.py	Fri Jul 30 09:21:26 2010
@@ -675,7 +675,10 @@
         Return `lineno` if it is, 0 if not (e.g. a docstring, comment, blank
         line or EOF). Warning: testing is not comprehensive.
         """
-        line = linecache.getline(filename, lineno, self.curframe.f_globals)
+        # this method should be callable before starting debugging, so default
+        # to "no globals" if there is no current frame
+        globs = self.curframe.f_globals if hasattr(self, 'curframe') else None
+        line = linecache.getline(filename, lineno, globs)
         if not line:
             print('End of file', file=self.stdout)
             return 0
@@ -1514,7 +1517,7 @@
     # changed by the user from the command line. There is a "restart" command
     # which allows explicit specification of command line arguments.
     pdb = Pdb()
-    while 1:
+    while True:
         try:
             pdb._runscript(mainpyfile)
             if pdb._user_requested_quit:

From python-checkins at python.org  Fri Jul 30 10:29:39 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 10:29:39 +0200 (CEST)
Subject: [Python-checkins] r83262 - in python/branches/py3k:
	Doc/library/pdb.rst Lib/pdb.py Misc/NEWS
Message-ID: <20100730082939.55140EE9E1@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 10:29:39 2010
New Revision: 83262

Log:
#1437051: allow "continue"/"next"/etc. in .pdbrc, also add pdb -c option to give these commands.  This allows to run a script until an exception occurs.

Modified:
   python/branches/py3k/Doc/library/pdb.rst
   python/branches/py3k/Lib/pdb.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Doc/library/pdb.rst
==============================================================================
--- python/branches/py3k/Doc/library/pdb.rst	(original)
+++ python/branches/py3k/Doc/library/pdb.rst	Fri Jul 30 10:29:39 2010
@@ -44,11 +44,15 @@
    python3 -m pdb myscript.py
 
 When invoked as a script, pdb will automatically enter post-mortem debugging if
-the program being debugged exits abnormally. After post-mortem debugging (or
-after normal exit of the program), pdb will restart the program. Automatic
+the program being debugged exits abnormally.  After post-mortem debugging (or
+after normal exit of the program), pdb will restart the program.  Automatic
 restarting preserves pdb's state (such as breakpoints) and in most cases is more
 useful than quitting the debugger upon program's exit.
 
+.. versionadded:: 3.2
+   :file:`pdb.py` now accepts a ``-c`` option that executes commands as if given
+   in a :file:`.pdbrc` file, see :ref:`debugger-commands`.
+
 The typical usage to break into the debugger from a running program is to
 insert ::
 
@@ -202,6 +206,11 @@
 in the home directory is read first and aliases defined there can be overridden
 by the local file.
 
+.. versionchanged:: 3.2
+   :file:`.pdbrc` can now contain commands that continue debugging, such as
+   :pdbcmd:`continue` or :pdbcmd:`next`.  Previously, these commands had no
+   effect.
+
 
 .. pdbcommand:: h(elp) [command]
 
@@ -388,9 +397,9 @@
    As an example, here are two useful aliases (especially when placed in the
    :file:`.pdbrc` file)::
 
-      #Print instance variables (usage "pi classInst")
+      # Print instance variables (usage "pi classInst")
       alias pi for k in %1.__dict__.keys(): print("%1.",k,"=",%1.__dict__[k])
-      #Print instance variables in self
+      # Print instance variables in self
       alias ps pi self
 
 .. pdbcommand:: unalias name

Modified: python/branches/py3k/Lib/pdb.py
==============================================================================
--- python/branches/py3k/Lib/pdb.py	(original)
+++ python/branches/py3k/Lib/pdb.py	Fri Jul 30 10:29:39 2010
@@ -332,19 +332,26 @@
         # locals whenever the .f_locals accessor is called, so we
         # cache it here to ensure that modifications are not overwritten.
         self.curframe_locals = self.curframe.f_locals
-        self.execRcLines()
+        return self.execRcLines()
 
     # Can be executed earlier than 'setup' if desired
     def execRcLines(self):
-        if self.rcLines:
-            # Make local copy because of recursion
-            rcLines = self.rcLines
-            # executed only once
-            self.rcLines = []
-            for line in rcLines:
-                line = line[:-1]
-                if len(line) > 0 and line[0] != '#':
-                    self.onecmd(line)
+        if not self.rcLines:
+            return
+        # local copy because of recursion
+        rcLines = self.rcLines
+        rcLines.reverse()
+        # execute every line only once
+        self.rcLines = []
+        while rcLines:
+            line = rcLines.pop().strip()
+            if line and line[0] != '#':
+                if self.onecmd(line):
+                    # if onecmd returns True, the command wants to exit
+                    # from the interaction, save leftover rc lines
+                    # to execute before next interaction
+                    self.rcLines += reversed(rcLines)
+                    return True
 
     # Override Bdb methods
 
@@ -367,7 +374,7 @@
         if self.bp_commands(frame):
             self.interaction(frame, None)
 
-    def bp_commands(self,frame):
+    def bp_commands(self, frame):
         """Call every command that was set for the current active breakpoint
         (if there is one).
 
@@ -409,7 +416,11 @@
     # General interaction function
 
     def interaction(self, frame, traceback):
-        self.setup(frame, traceback)
+        if self.setup(frame, traceback):
+            # no interaction desired at this time (happens if .pdbrc contains
+            # a command like "continue")
+            self.forget()
+            return
         self.print_stack_entry(self.stack[self.curindex])
         self.cmdloop()
         self.forget()
@@ -1497,17 +1508,42 @@
     import pydoc
     pydoc.pager(__doc__)
 
+_usage = """\
+usage: pdb.py [-c command] ... pyfile [arg] ...
+
+Debug the Python program given by pyfile.
+
+Initial commands are read from .pdbrc files in your home directory
+and in the current directory, if they exist.  Commands supplied with
+-c are executed after commands from .pdbrc files.
+
+To let the script run until an exception occurs, use "-c continue".
+To let the script run until a given line X in the debugged file, use
+"-c 'break X' -c continue"."""
+
 def main():
-    if not sys.argv[1:] or sys.argv[1] in ("--help", "-h"):
-        print("usage: pdb.py scriptfile [arg] ...")
+    import getopt
+
+    opts, args = getopt.getopt(sys.argv[1:], 'hc:', ['--help', '--command='])
+
+    if not args:
+        print(_usage)
         sys.exit(2)
 
-    mainpyfile =  sys.argv[1]     # Get script filename
+    commands = []
+    for opt, optarg in opts:
+        if opt in ['-h', '--help']:
+            print(_usage)
+            sys.exit()
+        elif opt in ['-c', '--command']:
+            commands.append(optarg)
+
+    mainpyfile = args[0]     # Get script filename
     if not os.path.exists(mainpyfile):
         print('Error:', mainpyfile, 'does not exist')
         sys.exit(1)
 
-    del sys.argv[0]         # Hide "pdb.py" from argument list
+    sys.argv[:] = args      # Hide "pdb.py" and pdb options from argument list
 
     # Replace pdb's dir with script's dir in front of module search path.
     sys.path[0] = os.path.dirname(mainpyfile)
@@ -1517,6 +1553,7 @@
     # changed by the user from the command line. There is a "restart" command
     # which allows explicit specification of command line arguments.
     pdb = Pdb()
+    pdb.rcLines.extend(commands)
     while True:
         try:
             pdb._runscript(mainpyfile)
@@ -1525,10 +1562,10 @@
             print("The program finished and will be restarted")
         except Restart:
             print("Restarting", mainpyfile, "with arguments:")
-            print("\t" + " ".join(sys.argv[1:]))
+            print("\t" + " ".join(args))
         except SystemExit:
             # In most cases SystemExit does not warrant a post-mortem session.
-            print("The program exited via sys.exit(). Exit status: ", end=' ')
+            print("The program exited via sys.exit(). Exit status:", end=' ')
             print(sys.exc_info()[1])
         except:
             traceback.print_exc()

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Fri Jul 30 10:29:39 2010
@@ -475,6 +475,10 @@
 Library
 -------
 
+- Issue #1437051: For pdb, allow "continue" and related commands in
+  .pdbrc files.  Also, add a command-line option "-c" that runs a
+  command as if given in .pdbrc.
+
 - Issue #4179: In pdb, allow "list ." as a command to return to the
   currently debugged line.
 

From python-checkins at python.org  Fri Jul 30 10:43:33 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 10:43:33 +0200 (CEST)
Subject: [Python-checkins] r83263 - in python/branches/py3k:
	Doc/library/pdb.rst Lib/bdb.py Lib/pdb.py Misc/NEWS
Message-ID: <20100730084333.21B99EFE5@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 10:43:32 2010
New Revision: 83263

Log:
Allow giving an explicit line number to "until".

Modified:
   python/branches/py3k/Doc/library/pdb.rst
   python/branches/py3k/Lib/bdb.py
   python/branches/py3k/Lib/pdb.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Doc/library/pdb.rst
==============================================================================
--- python/branches/py3k/Doc/library/pdb.rst	(original)
+++ python/branches/py3k/Doc/library/pdb.rst	Fri Jul 30 10:43:32 2010
@@ -330,10 +330,14 @@
    executes called functions at (nearly) full speed, only stopping at the next
    line in the current function.)
 
-.. pdbcommand:: unt(il)
+.. pdbcommand:: unt(il) [lineno]
 
-   Continue execution until the line with the line number greater than the
-   current one is reached or when returning from current frame.
+   Without argument, continue execution until the line with a number greater
+   than the current one is reached.
+
+   With a line number, continue execution until a line with a number greater or
+   equal to that is reached.  In both cases, also stop when the current frame
+   returns.
 
 .. pdbcommand:: r(eturn)
 

Modified: python/branches/py3k/Lib/bdb.py
==============================================================================
--- python/branches/py3k/Lib/bdb.py	(original)
+++ python/branches/py3k/Lib/bdb.py	Fri Jul 30 10:43:32 2010
@@ -174,10 +174,13 @@
     # Derived classes and clients can call the following methods
     # to affect the stepping state.
 
-    def set_until(self, frame): #the name "until" is borrowed from gdb
+    def set_until(self, frame, lineno=None):
         """Stop when the line with the line no greater than the current one is
         reached or when returning from current frame"""
-        self._set_stopinfo(frame, frame, frame.f_lineno+1)
+        # the name "until" is borrowed from gdb
+        if lineno is None:
+            lineno = frame.f_lineno + 1
+        self._set_stopinfo(frame, frame, lineno)
 
     def set_step(self):
         """Stop after one line of code."""

Modified: python/branches/py3k/Lib/pdb.py
==============================================================================
--- python/branches/py3k/Lib/pdb.py	(original)
+++ python/branches/py3k/Lib/pdb.py	Fri Jul 30 10:43:32 2010
@@ -71,11 +71,11 @@
         An arrow indicates the "current frame", which determines the
         context of most commands.
 
-d(own) [ count ]
+d(own) [count]
         Move the current frame count (default one) levels down in the
         stack trace (to a newer frame).
 
-u(p) [ count ]
+u(p) [count]
         Move the current frame count (default one) levels up in the
         stack trace (to an older frame).
 
@@ -140,9 +140,12 @@
         Continue execution until the next line in the current function
         is reached or it returns.
 
-unt(il)
-        Continue execution until the line with a number greater than
-        the current one is reached or until the current frame returns.
+unt(il) [lineno]
+        Without argument, continue execution until the line with a
+        number greater than the current one is reached.  With a line
+        number, continue execution until a line with a number greater
+        or equal to that is reached.  In both cases, also stop when
+        the current frame returns.
 
 r(eturn)
         Continue execution until the current function returns.
@@ -883,7 +886,19 @@
     do_d = do_down
 
     def do_until(self, arg):
-        self.set_until(self.curframe)
+        if arg:
+            try:
+                lineno = int(arg)
+            except ValueError:
+                print('*** Error in argument:', repr(arg), file=self.stdout)
+                return
+            if lineno <= self.curframe.f_lineno:
+                print('*** "until" line number is smaller than current '
+                      'line number', file=self.stdout)
+                return
+        else:
+            lineno = None
+        self.set_until(self.curframe, lineno)
         return 1
     do_unt = do_until
 
@@ -1518,8 +1533,8 @@
 -c are executed after commands from .pdbrc files.
 
 To let the script run until an exception occurs, use "-c continue".
-To let the script run until a given line X in the debugged file, use
-"-c 'break X' -c continue"."""
+To let the script run up to a given line X in the debugged file, use
+"-c 'until X'"."""
 
 def main():
     import getopt

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Fri Jul 30 10:43:32 2010
@@ -475,6 +475,8 @@
 Library
 -------
 
+- In pdb, allow giving a line number to the "until" command.
+
 - Issue #1437051: For pdb, allow "continue" and related commands in
   .pdbrc files.  Also, add a command-line option "-c" that runs a
   command as if given in .pdbrc.

From python-checkins at python.org  Fri Jul 30 10:45:26 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 10:45:26 +0200 (CEST)
Subject: [Python-checkins] r83264 - in python/branches/py3k:
	Doc/library/pdb.rst Lib/pdb.py
Message-ID: <20100730084526.D85B6EE981@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 10:45:26 2010
New Revision: 83264

Log:
Document the "jump" command in pdb.__doc__, and add a version tag for "until X".

Modified:
   python/branches/py3k/Doc/library/pdb.rst
   python/branches/py3k/Lib/pdb.py

Modified: python/branches/py3k/Doc/library/pdb.rst
==============================================================================
--- python/branches/py3k/Doc/library/pdb.rst	(original)
+++ python/branches/py3k/Doc/library/pdb.rst	Fri Jul 30 10:45:26 2010
@@ -339,6 +339,9 @@
    equal to that is reached.  In both cases, also stop when the current frame
    returns.
 
+   .. versionchanged:: 3.2
+      Allow giving an explicit line number.
+
 .. pdbcommand:: r(eturn)
 
    Continue execution until the current function returns.

Modified: python/branches/py3k/Lib/pdb.py
==============================================================================
--- python/branches/py3k/Lib/pdb.py	(original)
+++ python/branches/py3k/Lib/pdb.py	Fri Jul 30 10:45:26 2010
@@ -147,6 +147,16 @@
         or equal to that is reached.  In both cases, also stop when
         the current frame returns.
 
+j(ump) lineno
+        Set the next line that will be executed.  Only available in
+        the bottom-most frame.  This lets you jump back and execute
+        code again, or jump forward to skip code that you don't want
+        to run.
+
+        It should be noted that not all jumps are allowed -- for
+        instance it is not possible to jump into the middle of a
+        for loop or out of a finally clause.
+
 r(eturn)
         Continue execution until the current function returns.
 

From python-checkins at python.org  Fri Jul 30 10:54:49 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 10:54:49 +0200 (CEST)
Subject: [Python-checkins] r83265 - in python/branches/py3k: Lib/pdb.py
	Misc/NEWS
Message-ID: <20100730085449.4EBC6D25E@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 10:54:49 2010
New Revision: 83265

Log:
#8015: fix crash when entering an empty line for breakpoint commands.  Also restore environment properly when an exception occurs during the definition of commands.

Modified:
   python/branches/py3k/Lib/pdb.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/pdb.py
==============================================================================
--- python/branches/py3k/Lib/pdb.py	(original)
+++ python/branches/py3k/Lib/pdb.py	Fri Jul 30 10:54:49 2010
@@ -509,8 +509,10 @@
             return self.handle_command_def(line)
 
     def handle_command_def(self,line):
-        """ Handles one command line during command list definition. """
+        """Handles one command line during command list definition."""
         cmd, arg, line = self.parseline(line)
+        if not cmd:
+            return
         if cmd == 'silent':
             self.commands_silent[self.commands_bnum] = True
             return # continue to handle other cmd def in the cmd list
@@ -518,7 +520,7 @@
             self.cmdqueue = []
             return 1 # end of cmd list
         cmdlist = self.commands[self.commands_bnum]
-        if (arg):
+        if arg:
             cmdlist.append(cmd+' '+arg)
         else:
             cmdlist.append(cmd)
@@ -561,9 +563,11 @@
         prompt_back = self.prompt
         self.prompt = '(com) '
         self.commands_defining = True
-        self.cmdloop()
-        self.commands_defining = False
-        self.prompt = prompt_back
+        try:
+            self.cmdloop()
+        finally:
+            self.commands_defining = False
+            self.prompt = prompt_back
 
     def do_break(self, arg, temporary = 0):
         # break [ ([filename:]lineno | function) [, "condition"] ]

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Fri Jul 30 10:54:49 2010
@@ -475,6 +475,9 @@
 Library
 -------
 
+- Issue #8015: In pdb, do not crash when an empty line is entered as
+  a breakpoint command.
+
 - In pdb, allow giving a line number to the "until" command.
 
 - Issue #1437051: For pdb, allow "continue" and related commands in

From python-checkins at python.org  Fri Jul 30 11:14:20 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 11:14:20 +0200 (CEST)
Subject: [Python-checkins] r83266 - in python/branches/py3k:
	Doc/library/pdb.rst Lib/bdb.py Lib/test/test_pdb.py
Message-ID: <20100730091420.B7D79C2C0@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 11:14:20 2010
New Revision: 83266

Log:
#1472251: remove addition of "\n" to code given to pdb.run[eval](), the bug in exec() that made this necessary has been fixed.  Also document that you can give code objects to run() and runeval(), and add some tests to test_pdb.

Modified:
   python/branches/py3k/Doc/library/pdb.rst
   python/branches/py3k/Lib/bdb.py
   python/branches/py3k/Lib/test/test_pdb.py

Modified: python/branches/py3k/Doc/library/pdb.rst
==============================================================================
--- python/branches/py3k/Doc/library/pdb.rst	(original)
+++ python/branches/py3k/Doc/library/pdb.rst	Fri Jul 30 11:14:20 2010
@@ -85,21 +85,21 @@
 
 .. function:: run(statement, globals=None, locals=None)
 
-   Execute the *statement* (given as a string) under debugger control.  The
-   debugger prompt appears before any code is executed; you can set breakpoints
-   and type :pdbcmd:`continue`, or you can step through the statement using
-   :pdbcmd:`step` or :pdbcmd:`next` (all these commands are explained below).
-   The optional *globals* and *locals* arguments specify the environment in
-   which the code is executed; by default the dictionary of the module
-   :mod:`__main__` is used.  (See the explanation of the built-in :func:`exec`
-   or :func:`eval` functions.)
+   Execute the *statement* (given as a string or a code object) under debugger
+   control.  The debugger prompt appears before any code is executed; you can
+   set breakpoints and type :pdbcmd:`continue`, or you can step through the
+   statement using :pdbcmd:`step` or :pdbcmd:`next` (all these commands are
+   explained below).  The optional *globals* and *locals* arguments specify the
+   environment in which the code is executed; by default the dictionary of the
+   module :mod:`__main__` is used.  (See the explanation of the built-in
+   :func:`exec` or :func:`eval` functions.)
 
 
 .. function:: runeval(expression, globals=None, locals=None)
 
-   Evaluate the *expression* (given as a string) under debugger control.  When
-   :func:`runeval` returns, it returns the value of the expression.  Otherwise
-   this function is similar to :func:`run`.
+   Evaluate the *expression* (given as a string or a code object) under debugger
+   control.  When :func:`runeval` returns, it returns the value of the
+   expression.  Otherwise this function is similar to :func:`run`.
 
 
 .. function:: runcall(function, *args, **kwds)

Modified: python/branches/py3k/Lib/bdb.py
==============================================================================
--- python/branches/py3k/Lib/bdb.py	(original)
+++ python/branches/py3k/Lib/bdb.py	Fri Jul 30 11:14:20 2010
@@ -364,8 +364,9 @@
         if line: s = s + lprefix + line.strip()
         return s
 
-    # The following two methods can be called by clients to use
-    # a debugger to debug a statement, given as a string.
+    # The following methods can be called by clients to use
+    # a debugger to debug a statement or an expression.
+    # Both can be given as a string, or a code object.
 
     def run(self, cmd, globals=None, locals=None):
         if globals is None:
@@ -375,8 +376,6 @@
             locals = globals
         self.reset()
         sys.settrace(self.trace_dispatch)
-        if not isinstance(cmd, types.CodeType):
-            cmd = cmd+'\n'
         try:
             exec(cmd, globals, locals)
         except BdbQuit:
@@ -393,8 +392,6 @@
             locals = globals
         self.reset()
         sys.settrace(self.trace_dispatch)
-        if not isinstance(expr, types.CodeType):
-            expr = expr+'\n'
         try:
             return eval(expr, globals, locals)
         except BdbQuit:

Modified: python/branches/py3k/Lib/test/test_pdb.py
==============================================================================
--- python/branches/py3k/Lib/test/test_pdb.py	(original)
+++ python/branches/py3k/Lib/test/test_pdb.py	Fri Jul 30 11:14:20 2010
@@ -1,5 +1,4 @@
-# A test suite for pdb; at the moment, this only validates skipping of
-# specified test modules (RFE #5142).
+# A test suite for pdb; not very comprehensive at the moment.
 
 import imp
 import sys
@@ -123,6 +122,50 @@
     """
 
 
+def pdb_invoke(method, arg):
+    """Run pdb.method(arg)."""
+    import pdb; getattr(pdb, method)(arg)
+
+
+def test_pdb_run_with_incorrect_argument():
+    """Testing run and runeval with incorrect first argument.
+
+    >>> pti = PdbTestInput(['continue',])
+    >>> with pti:
+    ...     pdb_invoke('run', lambda x: x)
+    Traceback (most recent call last):
+    TypeError: exec() arg 1 must be a string, bytes or code object
+
+    >>> with pti:
+    ...     pdb_invoke('runeval', lambda x: x)
+    Traceback (most recent call last):
+    TypeError: eval() arg 1 must be a string, bytes or code object
+    """
+
+
+def test_pdb_run_with_code_object():
+    """Testing run and runeval with code object as a first argument.
+
+    >>> with PdbTestInput(['step','x', 'continue']):
+    ...     pdb_invoke('run', compile('x=1', '<string>', 'exec'))
+    > <string>(1)<module>()
+    (Pdb) step
+    --Return--
+    > <string>(1)<module>()->None
+    (Pdb) x
+    1
+    (Pdb) continue
+
+    >>> with PdbTestInput(['x', 'continue']):
+    ...     x=0
+    ...     pdb_invoke('runeval', compile('x+1', '<string>', 'eval'))
+    > <string>(1)<module>()->None
+    (Pdb) x
+    1
+    (Pdb) continue
+    """
+
+
 def test_main():
     from test import test_pdb
     support.run_doctest(test_pdb, verbosity=True)

From python-checkins at python.org  Fri Jul 30 11:18:49 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 11:18:49 +0200 (CEST)
Subject: [Python-checkins] r83267 - in python/branches/release27-maint/Lib:
	pdb.py test/test_pdb.py
Message-ID: <20100730091849.EA7D4EE981@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 11:18:49 2010
New Revision: 83267

Log:
#7539: use _saferepr() for printing exceptions from pdb.

Modified:
   python/branches/release27-maint/Lib/pdb.py
   python/branches/release27-maint/Lib/test/test_pdb.py

Modified: python/branches/release27-maint/Lib/pdb.py
==============================================================================
--- python/branches/release27-maint/Lib/pdb.py	(original)
+++ python/branches/release27-maint/Lib/pdb.py	Fri Jul 30 11:18:49 2010
@@ -237,7 +237,7 @@
             if type(t) == type(''):
                 exc_type_name = t
             else: exc_type_name = t.__name__
-            print >>self.stdout, '***', exc_type_name + ':', v
+            print >>self.stdout, '***', exc_type_name + ':', _saferepr(v)
 
     def precmd(self, line):
         """Handle alias expansion and ';;' separator."""
@@ -753,7 +753,7 @@
             if isinstance(t, str):
                 exc_type_name = t
             else: exc_type_name = t.__name__
-            print >>self.stdout, '***', exc_type_name + ':', repr(v)
+            print >>self.stdout, '***', exc_type_name + ':', _saferepr(v)
             raise
 
     def do_p(self, arg):

Modified: python/branches/release27-maint/Lib/test/test_pdb.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_pdb.py	(original)
+++ python/branches/release27-maint/Lib/test/test_pdb.py	Fri Jul 30 11:18:49 2010
@@ -126,6 +126,33 @@
     """
 
 
+def test_pdb_unicode_exception():
+    r"""This tests exceptions that cannot be displayed due to Unicode issues.
+    http://bugs.python.org/issue7539
+
+    >>> def test_function():
+    ...     import pdb; pdb.Pdb().set_trace()
+    ...     pass
+
+    >>> def raising_function():
+    ...     raise ValueError(u"\xff")
+
+    >>> with PdbTestInput([
+    ...     'raising_function()',
+    ...     'p raising_function()',
+    ...     'continue',
+    ... ]):
+    ...     test_function()
+    > <doctest test.test_pdb.test_pdb_unicode_exception[0]>(3)test_function()
+    -> pass
+    (Pdb) raising_function()
+    *** ValueError: ValueError(u'\xff',)
+    (Pdb) p raising_function()
+    *** ValueError: ValueError(u'\xff',)
+    (Pdb) continue
+    """
+
+
 def test_main():
     from test import test_pdb
     test_support.run_doctest(test_pdb, verbosity=True)

From python-checkins at python.org  Fri Jul 30 11:23:23 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 11:23:23 +0200 (CEST)
Subject: [Python-checkins] r83268 - in python/branches/py3k: Lib/doctest.py
	Lib/test/test_doctest.py Misc/NEWS
Message-ID: <20100730092323.3C088EE981@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 11:23:23 2010
New Revision: 83268

Log:
Issue #8048: Prevent doctests from failing when sys.displayhook has
been reassigned.

Modified:
   python/branches/py3k/Lib/doctest.py
   python/branches/py3k/Lib/test/test_doctest.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/doctest.py
==============================================================================
--- python/branches/py3k/Lib/doctest.py	(original)
+++ python/branches/py3k/Lib/doctest.py	Fri Jul 30 11:23:23 2010
@@ -1379,12 +1379,17 @@
         self.save_linecache_getlines = linecache.getlines
         linecache.getlines = self.__patched_linecache_getlines
 
+        # Make sure sys.displayhook just prints the value to stdout
+        save_displayhook = sys.displayhook
+        sys.displayhook = sys.__displayhook__
+
         try:
             return self.__run(test, compileflags, out)
         finally:
             sys.stdout = save_stdout
             pdb.set_trace = save_set_trace
             linecache.getlines = self.save_linecache_getlines
+            sys.displayhook = save_displayhook
             if clear_globs:
                 test.globs.clear()
                 import builtins

Modified: python/branches/py3k/Lib/test/test_doctest.py
==============================================================================
--- python/branches/py3k/Lib/test/test_doctest.py	(original)
+++ python/branches/py3k/Lib/test/test_doctest.py	Fri Jul 30 11:23:23 2010
@@ -980,6 +980,35 @@
         ZeroDivisionError: integer division or modulo by zero
     TestResults(failed=1, attempted=1)
 """
+    def displayhook(): r"""
+Test that changing sys.displayhook doesn't matter for doctest.
+
+    >>> import sys
+    >>> orig_displayhook = sys.displayhook
+    >>> def my_displayhook(x):
+    ...     print('hi!')
+    >>> sys.displayhook = my_displayhook
+    >>> def f():
+    ...     '''
+    ...     >>> 3
+    ...     3
+    ...     '''
+    >>> test = doctest.DocTestFinder().find(f)[0]
+    >>> r = doctest.DocTestRunner(verbose=False).run(test)
+    >>> post_displayhook = sys.displayhook
+
+    We need to restore sys.displayhook now, so that we'll be able to test
+    results.
+
+    >>> sys.displayhook = orig_displayhook
+
+    Ok, now we can check that everything is ok.
+
+    >>> r
+    TestResults(failed=0, attempted=1)
+    >>> post_displayhook is my_displayhook
+    True
+"""
     def optionflags(): r"""
 Tests of `DocTestRunner`'s option flag handling.
 

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Fri Jul 30 11:23:23 2010
@@ -475,6 +475,9 @@
 Library
 -------
 
+- Issue #8048: Prevent doctests from failing when sys.displayhook has
+  been reassigned.
+
 - Issue #8015: In pdb, do not crash when an empty line is entered as
   a breakpoint command.
 

From python-checkins at python.org  Fri Jul 30 11:43:00 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 11:43:00 +0200 (CEST)
Subject: [Python-checkins] r83269 - in python/branches/py3k: Lib/pdb.py
	Misc/NEWS
Message-ID: <20100730094300.A4BC0EE9EB@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 11:43:00 2010
New Revision: 83269

Log:
#6719: In pdb, do not stop somewhere in the encodings machinery if the source file to be debugged is in a non-builtin encoding.

Modified:
   python/branches/py3k/Lib/pdb.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/pdb.py
==============================================================================
--- python/branches/py3k/Lib/pdb.py	(original)
+++ python/branches/py3k/Lib/pdb.py	Fri Jul 30 11:43:00 2010
@@ -413,6 +413,8 @@
 
     def user_return(self, frame, return_value):
         """This function is called when a return trap is set here."""
+        if self._wait_for_mainpyfile:
+            return
         frame.f_locals['__return__'] = return_value
         print('--Return--', file=self.stdout)
         self.interaction(frame, None)
@@ -420,6 +422,8 @@
     def user_exception(self, frame, exc_info):
         """This function is called if an exception occurs,
         but only if we are to stop at or just below this level."""
+        if self._wait_for_mainpyfile:
+            return
         exc_type, exc_value, exc_traceback = exc_info
         frame.f_locals['__exception__'] = exc_type, exc_value
         exc_type_name = exc_type.__name__

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Fri Jul 30 11:43:00 2010
@@ -475,6 +475,9 @@
 Library
 -------
 
+- Issue #6719: In pdb, do not stop somewhere in the encodings machinery
+  if the source file to be debugged is in a non-builtin encoding.
+
 - Issue #8048: Prevent doctests from failing when sys.displayhook has
   been reassigned.
 

From python-checkins at python.org  Fri Jul 30 11:54:45 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 11:54:45 +0200 (CEST)
Subject: [Python-checkins] r83270 - python/branches/py3k/Lib/cmd.py
Message-ID: <20100730095445.13727EE99C@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 11:54:44 2010
New Revision: 83270

Log:
Remove redundant import.

Modified:
   python/branches/py3k/Lib/cmd.py

Modified: python/branches/py3k/Lib/cmd.py
==============================================================================
--- python/branches/py3k/Lib/cmd.py	(original)
+++ python/branches/py3k/Lib/cmd.py	Fri Jul 30 11:54:44 2010
@@ -84,7 +84,6 @@
         sys.stdin and sys.stdout are used.
 
         """
-        import sys
         if stdin is not None:
             self.stdin = stdin
         else:

From python-checkins at python.org  Fri Jul 30 11:59:28 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 11:59:28 +0200 (CEST)
Subject: [Python-checkins] r83271 - in python/branches/py3k: Lib/doctest.py
	Misc/NEWS
Message-ID: <20100730095928.3EC61EE99C@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 11:59:28 2010
New Revision: 83271

Log:
#5727: Restore the ability to use readline when calling into pdb in doctests.

Modified:
   python/branches/py3k/Lib/doctest.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/doctest.py
==============================================================================
--- python/branches/py3k/Lib/doctest.py	(original)
+++ python/branches/py3k/Lib/doctest.py	Fri Jul 30 11:59:28 2010
@@ -318,6 +318,8 @@
         self.__out = out
         self.__debugger_used = False
         pdb.Pdb.__init__(self, stdout=out)
+        # still use input() to get user input
+        self.use_rawinput = 1
 
     def set_trace(self, frame=None):
         self.__debugger_used = True

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Fri Jul 30 11:59:28 2010
@@ -475,6 +475,9 @@
 Library
 -------
 
+- Issue #5727: Restore the ability to use readline when calling into pdb
+  in doctests.
+
 - Issue #6719: In pdb, do not stop somewhere in the encodings machinery
   if the source file to be debugged is in a non-builtin encoding.
 

From python-checkins at python.org  Fri Jul 30 12:29:19 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 12:29:19 +0200 (CEST)
Subject: [Python-checkins] r83272 - in python/branches/py3k: Lib/bdb.py
	Lib/test/test_pdb.py Misc/NEWS
Message-ID: <20100730102919.7CA09EE986@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 12:29:19 2010
New Revision: 83272

Log:
#5294: Fix the behavior of pdb "continue" command when called in the top-level debugged frame.

Modified:
   python/branches/py3k/Lib/bdb.py
   python/branches/py3k/Lib/test/test_pdb.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/bdb.py
==============================================================================
--- python/branches/py3k/Lib/bdb.py	(original)
+++ python/branches/py3k/Lib/bdb.py	Fri Jul 30 12:29:19 2010
@@ -109,6 +109,8 @@
                self.is_skipped_module(frame.f_globals.get('__name__')):
             return False
         if frame is self.stopframe:
+            if self.stoplineno == -1:
+                return False
             return frame.f_lineno >= self.stoplineno
         while frame is not None and frame is not self.stopframe:
             if frame is self.botframe:
@@ -165,10 +167,12 @@
         but only if we are to stop at or just below this level."""
         pass
 
-    def _set_stopinfo(self, stopframe, returnframe, stoplineno=-1):
+    def _set_stopinfo(self, stopframe, returnframe, stoplineno=0):
         self.stopframe = stopframe
         self.returnframe = returnframe
         self.quitting = 0
+        # stoplineno >= 0 means: stop at line >= the stoplineno
+        # stoplineno -1 means: don't stop at all
         self.stoplineno = stoplineno
 
     # Derived classes and clients can call the following methods
@@ -184,7 +188,7 @@
 
     def set_step(self):
         """Stop after one line of code."""
-        self._set_stopinfo(None,None)
+        self._set_stopinfo(None, None)
 
     def set_next(self, frame):
         """Stop on the next line in or below the given frame."""
@@ -211,7 +215,7 @@
 
     def set_continue(self):
         # Don't stop except at breakpoints or when finished
-        self._set_stopinfo(self.botframe, None)
+        self._set_stopinfo(self.botframe, None, -1)
         if not self.breaks:
             # no breakpoints; run without debugger overhead
             sys.settrace(None)

Modified: python/branches/py3k/Lib/test/test_pdb.py
==============================================================================
--- python/branches/py3k/Lib/test/test_pdb.py	(original)
+++ python/branches/py3k/Lib/test/test_pdb.py	Fri Jul 30 12:29:19 2010
@@ -122,6 +122,48 @@
     """
 
 
+def test_pdb_continue_in_bottomframe():
+    """Test that "continue" and "next" work properly in bottom frame (issue #5294).
+
+    >>> def test_function():
+    ...     import pdb, sys; inst = pdb.Pdb()
+    ...     inst.set_trace()
+    ...     inst.botframe = sys._getframe()  # hackery to get the right botframe
+    ...     print(1)
+    ...     print(2)
+    ...     print(3)
+    ...     print(4)
+
+    >>> with PdbTestInput([
+    ...     'next',
+    ...     'break 7',
+    ...     'continue',
+    ...     'next',
+    ...     'continue',
+    ...     'continue',
+    ... ]):
+    ...    test_function()
+    > <doctest test.test_pdb.test_pdb_continue_in_bottomframe[0]>(4)test_function()
+    -> inst.botframe = sys._getframe()  # hackery to get the right botframe
+    (Pdb) next
+    > <doctest test.test_pdb.test_pdb_continue_in_bottomframe[0]>(5)test_function()
+    -> print(1)
+    (Pdb) break 7
+    Breakpoint 1 at <doctest test.test_pdb.test_pdb_continue_in_bottomframe[0]>:7
+    (Pdb) continue
+    1
+    2
+    > <doctest test.test_pdb.test_pdb_continue_in_bottomframe[0]>(7)test_function()
+    -> print(3)
+    (Pdb) next
+    3
+    > <doctest test.test_pdb.test_pdb_continue_in_bottomframe[0]>(8)test_function()
+    -> print(4)
+    (Pdb) continue
+    4
+    """
+
+
 def pdb_invoke(method, arg):
     """Run pdb.method(arg)."""
     import pdb; getattr(pdb, method)(arg)

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Fri Jul 30 12:29:19 2010
@@ -475,6 +475,9 @@
 Library
 -------
 
+- Issue #5294: Fix the behavior of pdb's "continue" command when called
+  in the top-level debugged frame.
+
 - Issue #5727: Restore the ability to use readline when calling into pdb
   in doctests.
 

From python-checkins at python.org  Fri Jul 30 12:49:13 2010
From: python-checkins at python.org (brian.quinlan)
Date: Fri, 30 Jul 2010 12:49:13 +0200 (CEST)
Subject: [Python-checkins] r83273 - peps/trunk/pep-3148.txt
Message-ID: <20100730104913.3A9AEEE981@mail.python.org>

Author: brian.quinlan
Date: Fri Jul 30 12:49:13 2010
New Revision: 83273

Log:
Describes the behavior of add_done_callback when the callable raises.

Modified:
   peps/trunk/pep-3148.txt

Modified: peps/trunk/pep-3148.txt
==============================================================================
--- peps/trunk/pep-3148.txt	(original)
+++ peps/trunk/pep-3148.txt	Fri Jul 30 12:49:13 2010
@@ -228,9 +228,13 @@
 
     Attaches a callable *fn* to the future that will be called when
     the future is cancelled or finishes running.  *fn* will be called
-    with the future as its only argument.  Added callables are called
-    in the order that they were added and are always called in a
-    thread belonging to the process that added them.
+    with the future as its only argument.
+     
+    Added callables are called in the order that they were added and
+    are always called in a thread belonging to the process that added
+    them.  If the callable raises an `Exception` then it will be
+    logged and ignored.  If the callable raises another
+    `BaseException` then behavior is not defined.
      
     If the future has already completed or been cancelled then *fn*
     will be called immediately.

From python-checkins at python.org  Fri Jul 30 13:31:03 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 13:31:03 +0200 (CEST)
Subject: [Python-checkins] r83274 - python/branches/py3k/Doc/conf.py
Message-ID: <20100730113103.A864BEE9E6@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 13:31:03 2010
New Revision: 83274

Log:
#3143: enable "collapsible sidebar" feature of new Sphinx version.

Modified:
   python/branches/py3k/Doc/conf.py

Modified: python/branches/py3k/Doc/conf.py
==============================================================================
--- python/branches/py3k/Doc/conf.py	(original)
+++ python/branches/py3k/Doc/conf.py	Fri Jul 30 13:31:03 2010
@@ -66,6 +66,9 @@
 # Options for HTML output
 # -----------------------
 
+html_theme = 'default'
+html_theme_options = {'collapsiblesidebar': True}
+
 # If not '', a 'Last updated on:' timestamp is inserted at every page bottom,
 # using the given strftime format.
 html_last_updated_fmt = '%b %d, %Y'

From python-checkins at python.org  Fri Jul 30 14:01:20 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 14:01:20 +0200 (CEST)
Subject: [Python-checkins] r83275 - in python/branches/py3k:
	Doc/library/bdb.rst Lib/bdb.py Lib/pdb.py
	Lib/test/test_pdb.py Misc/NEWS
Message-ID: <20100730120120.8FB42EE9C4@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 14:01:20 2010
New Revision: 83275

Log:
#809887: improve pdb feedback for breakpoint-related actions.  Also add a functional test for these commands.

Modified:
   python/branches/py3k/Doc/library/bdb.rst
   python/branches/py3k/Lib/bdb.py
   python/branches/py3k/Lib/pdb.py
   python/branches/py3k/Lib/test/test_pdb.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Doc/library/bdb.rst
==============================================================================
--- python/branches/py3k/Doc/library/bdb.rst	(original)
+++ python/branches/py3k/Doc/library/bdb.rst	Fri Jul 30 14:01:20 2010
@@ -267,6 +267,15 @@
 
       Delete all existing breakpoints.
 
+   .. method:: get_bpbynumber(arg)
+
+      Return a breakpoint specified by the given number.  If *arg* is a string,
+      it will be converted to a number.  If *arg* is a non-numeric string, if
+      the given breakpoint never existed or has been deleted, a
+      :exc:`ValueError` is raised.
+
+      .. versionadded:: 3.2
+
    .. method:: get_break(filename, lineno)
 
       Check if there is a breakpoint for *lineno* of *filename*.

Modified: python/branches/py3k/Lib/bdb.py
==============================================================================
--- python/branches/py3k/Lib/bdb.py	(original)
+++ python/branches/py3k/Lib/bdb.py	Fri Jul 30 14:01:20 2010
@@ -270,15 +270,9 @@
 
     def clear_bpbynumber(self, arg):
         try:
-            number = int(arg)
-        except:
-            return 'Non-numeric breakpoint number (%s)' % arg
-        try:
-            bp = Breakpoint.bpbynumber[number]
-        except IndexError:
-            return 'Breakpoint number (%d) out of range' % number
-        if not bp:
-            return 'Breakpoint (%d) already deleted' % number
+            bp = self.get_bpbynumber(arg)
+        except ValueError as err:
+            return str(err)
         self.clear_break(bp.file, bp.line)
 
     def clear_all_file_breaks(self, filename):
@@ -299,6 +293,21 @@
                 bp.deleteMe()
         self.breaks = {}
 
+    def get_bpbynumber(self, arg):
+        if not arg:
+            raise ValueError('Breakpoint number expected')
+        try:
+            number = int(arg)
+        except ValueError:
+            raise ValueError('Non-numeric breakpoint number %s' % arg)
+        try:
+            bp = Breakpoint.bpbynumber[number]
+        except IndexError:
+            raise ValueError('Breakpoint number %d out of range' % number)
+        if bp is None:
+            raise ValueError('Breakpoint %d already deleted' % number)
+        return bp
+
     def get_break(self, filename, lineno):
         filename = self.canonic(filename)
         return filename in self.breaks and \
@@ -510,6 +519,9 @@
             print(('\tbreakpoint already hit %d time%s' %
                           (self.hits, ss)), file=out)
 
+    def __str__(self):
+        return 'breakpoint %s at %s:%s' % (self.number, self.file, self.line)
+
 # -----------end of Breakpoint class----------
 
 def checkfuncname(b, frame):

Modified: python/branches/py3k/Lib/pdb.py
==============================================================================
--- python/branches/py3k/Lib/pdb.py	(original)
+++ python/branches/py3k/Lib/pdb.py	Fri Jul 30 14:01:20 2010
@@ -552,12 +552,12 @@
         Those commands will be executed whenever the breakpoint causes
         the program to stop execution."""
         if not arg:
-            bnum = len(bdb.Breakpoint.bpbynumber)-1
+            bnum = len(bdb.Breakpoint.bpbynumber) - 1
         else:
             try:
                 bnum = int(arg)
             except:
-                print("Usage : commands [bnum]\n        ...\n        end",
+                print("Usage: commands [bnum]\n        ...\n        end",
                       file=self.stdout)
                 return
         self.commands_bnum = bnum
@@ -634,10 +634,9 @@
                     # last thing to try
                     (ok, filename, ln) = self.lineinfo(arg)
                     if not ok:
-                        print('*** The specified object', end=' ', file=self.stdout)
-                        print(repr(arg), end=' ', file=self.stdout)
-                        print('is not a function', file=self.stdout)
-                        print('or was not found along sys.path.', file=self.stdout)
+                        print('*** The specified object %r is not a function '
+                              'or was not found along sys.path.' % arg,
+                              file=self.stdout)
                         return
                     funcname = ok # ok contains a function name
                     lineno = int(ln)
@@ -726,81 +725,56 @@
         args = arg.split()
         for i in args:
             try:
-                i = int(i)
-            except ValueError:
-                print('Breakpoint index %r is not a number' % i, file=self.stdout)
-                continue
-
-            if not (0 <= i < len(bdb.Breakpoint.bpbynumber)):
-                print('No breakpoint numbered', i, file=self.stdout)
-                continue
-
-            bp = bdb.Breakpoint.bpbynumber[i]
-            if bp:
+                bp = self.get_bpbynumber(i)
+            except ValueError as err:
+                print('***', err, file=self.stdout)
+            else:
                 bp.enable()
+                print('Enabled %s' % bp, file=self.stdout)
 
     def do_disable(self, arg):
         args = arg.split()
         for i in args:
             try:
-                i = int(i)
-            except ValueError:
-                print('Breakpoint index %r is not a number' % i, file=self.stdout)
-                continue
-
-            if not (0 <= i < len(bdb.Breakpoint.bpbynumber)):
-                print('No breakpoint numbered', i, file=self.stdout)
-                continue
-
-            bp = bdb.Breakpoint.bpbynumber[i]
-            if bp:
+                bp = self.get_bpbynumber(i)
+            except ValueError as err:
+                print('***', err, file=self.stdout)
+            else:
                 bp.disable()
+                print('Disabled %s' % bp, file=self.stdout)
 
     def do_condition(self, arg):
         # arg is breakpoint number and condition
         args = arg.split(' ', 1)
         try:
-            bpnum = int(args[0].strip())
-        except ValueError:
-            # something went wrong
-            print('Breakpoint index %r is not a number' % args[0], file=self.stdout)
-            return
-        try:
             cond = args[1]
-        except:
+        except IndexError:
             cond = None
         try:
-            bp = bdb.Breakpoint.bpbynumber[bpnum]
-        except IndexError:
-            print('Breakpoint index %r is not valid' % args[0],
-                  file=self.stdout)
-            return
-        if bp:
+            bp = self.get_bpbynumber(args[0].strip())
+        except ValueError as err:
+            print('***', err, file=self.stdout)
+        else:
             bp.cond = cond
             if not cond:
-                print('Breakpoint', bpnum, end=' ', file=self.stdout)
-                print('is now unconditional.', file=self.stdout)
+                print('Breakpoint %d is now unconditional.' % bp.number,
+                      file=self.stdout)
+            else:
+                print('New condition set for breakpoint %d.' % bp.number,
+                      file=self.stdout)
 
-    def do_ignore(self,arg):
+    def do_ignore(self, arg):
         """arg is bp number followed by ignore count."""
         args = arg.split()
         try:
-            bpnum = int(args[0].strip())
-        except ValueError:
-            # something went wrong
-            print('Breakpoint index %r is not a number' % args[0], file=self.stdout)
-            return
-        try:
             count = int(args[1].strip())
         except:
             count = 0
         try:
-            bp = bdb.Breakpoint.bpbynumber[bpnum]
-        except IndexError:
-            print('Breakpoint index %r is not valid' % args[0],
-                  file=self.stdout)
-            return
-        if bp:
+            bp = self.get_bpbynumber(args[0].strip())
+        except ValueError as err:
+            print('***', err, file=self.stdout)
+        else:
             bp.ignore = count
             if count > 0:
                 reply = 'Will ignore next '
@@ -808,10 +782,10 @@
                     reply = reply + '%d crossings' % count
                 else:
                     reply = reply + '1 crossing'
-                print(reply + ' of breakpoint %d.' % bpnum, file=self.stdout)
+                print(reply + ' of breakpoint %d.' % bp.number, file=self.stdout)
             else:
-                print('Will stop next time breakpoint', end=' ', file=self.stdout)
-                print(bpnum, 'is reached.', file=self.stdout)
+                print('Will stop next time breakpoint %d is reached.'
+                      % bp.number, file=self.stdout)
 
     def do_clear(self, arg):
         """Three possibilities, tried in this order:
@@ -825,7 +799,10 @@
                 reply = 'no'
             reply = reply.strip().lower()
             if reply in ('y', 'yes'):
+                bplist = [bp for bp in bdb.Breakpoint.bpbynumber if bp]
                 self.clear_all_breaks()
+                for bp in bplist:
+                    print('Deleted %s' % bp, file=self.stdout)
             return
         if ':' in arg:
             # Make sure it works for "clear C:\foo\bar.py:12"
@@ -837,25 +814,23 @@
             except ValueError:
                 err = "Invalid line number (%s)" % arg
             else:
+                bplist = self.get_breaks(filename, lineno)
                 err = self.clear_break(filename, lineno)
-            if err: print('***', err, file=self.stdout)
+            if err:
+                print('***', err, file=self.stdout)
+            else:
+                for bp in bplist:
+                    print('Deleted %s' % bp, file=self.stdout)
             return
         numberlist = arg.split()
         for i in numberlist:
             try:
-                i = int(i)
-            except ValueError:
-                print('Breakpoint index %r is not a number' % i, file=self.stdout)
-                continue
-
-            if not (0 <= i < len(bdb.Breakpoint.bpbynumber)):
-                print('No breakpoint numbered', i, file=self.stdout)
-                continue
-            err = self.clear_bpbynumber(i)
-            if err:
+                bp = self.get_bpbynumber(i)
+            except ValueError as err:
                 print('***', err, file=self.stdout)
             else:
-                print('Deleted breakpoint', i, file=self.stdout)
+                self.clear_break(bp.file, bp.line)
+                print('Deleted %s' % bp, file=self.stdout)
     do_cl = do_clear # 'c' is already an abbreviation for 'continue'
 
     def do_where(self, arg):

Modified: python/branches/py3k/Lib/test/test_pdb.py
==============================================================================
--- python/branches/py3k/Lib/test/test_pdb.py	(original)
+++ python/branches/py3k/Lib/test/test_pdb.py	Fri Jul 30 14:01:20 2010
@@ -134,7 +134,7 @@
     ...     print(3)
     ...     print(4)
 
-    >>> with PdbTestInput([
+    >>> with PdbTestInput([  # doctest: +ELLIPSIS
     ...     'next',
     ...     'break 7',
     ...     'continue',
@@ -149,7 +149,7 @@
     > <doctest test.test_pdb.test_pdb_continue_in_bottomframe[0]>(5)test_function()
     -> print(1)
     (Pdb) break 7
-    Breakpoint 1 at <doctest test.test_pdb.test_pdb_continue_in_bottomframe[0]>:7
+    Breakpoint ... at <doctest test.test_pdb.test_pdb_continue_in_bottomframe[0]>:7
     (Pdb) continue
     1
     2
@@ -164,6 +164,84 @@
     """
 
 
+def test_pdb_breakpoints():
+    """Test handling of breakpoints.
+
+    >>> def test_function():
+    ...     import pdb; pdb.Pdb().set_trace()
+    ...     print(1)
+    ...     print(2)
+    ...     print(3)
+    ...     print(4)
+
+    First, need to clear bdb state that might be left over from previous tests.
+    Otherwise, the new breakpoints might get assigned different numbers.
+
+    >>> from bdb import Breakpoint
+    >>> Breakpoint.next = 1
+    >>> Breakpoint.bplist = {}
+    >>> Breakpoint.bpbynumber = [None]
+
+    Now test the breakpoint commands.  NORMALIZE_WHITESPACE is needed because
+    the breakpoint list outputs a tab for the "stop only" and "ignore next"
+    lines, which we don't want to put in here.
+
+    >>> with PdbTestInput([  # doctest: +NORMALIZE_WHITESPACE
+    ...     'break 3',
+    ...     'disable 1',
+    ...     'ignore 1 10',
+    ...     'condition 1 1 < 2',
+    ...     'break 4',
+    ...     'break',
+    ...     'condition 1',
+    ...     'enable 1',
+    ...     'clear 1',
+    ...     'commands 2',
+    ...     'print 42',
+    ...     'end',
+    ...     'continue',  # will stop at breakpoint 2
+    ...     'continue',
+    ... ]):
+    ...    test_function()
+    > <doctest test.test_pdb.test_pdb_breakpoints[0]>(3)test_function()
+    -> print(1)
+    (Pdb) break 3
+    Breakpoint 1 at <doctest test.test_pdb.test_pdb_breakpoints[0]>:3
+    (Pdb) disable 1
+    Disabled breakpoint 1 at <doctest test.test_pdb.test_pdb_breakpoints[0]>:3
+    (Pdb) ignore 1 10
+    Will ignore next 10 crossings of breakpoint 1.
+    (Pdb) condition 1 1 < 2
+    New condition set for breakpoint 1.
+    (Pdb) break 4
+    Breakpoint 2 at <doctest test.test_pdb.test_pdb_breakpoints[0]>:4
+    (Pdb) break
+    Num Type         Disp Enb   Where
+    1   breakpoint   keep no    at <doctest test.test_pdb.test_pdb_breakpoints[0]>:3
+            stop only if 1 < 2
+            ignore next 10 hits
+    2   breakpoint   keep yes   at <doctest test.test_pdb.test_pdb_breakpoints[0]>:4
+    (Pdb) condition 1
+    Breakpoint 1 is now unconditional.
+    (Pdb) enable 1
+    Enabled breakpoint 1 at <doctest test.test_pdb.test_pdb_breakpoints[0]>:3
+    (Pdb) clear 1
+    Deleted breakpoint 1 at <doctest test.test_pdb.test_pdb_breakpoints[0]>:3
+    (Pdb) commands 2
+    (com) print 42
+    (com) end
+    (Pdb) continue
+    1
+    42
+    > <doctest test.test_pdb.test_pdb_breakpoints[0]>(4)test_function()
+    -> print(2)
+    (Pdb) continue
+    2
+    3
+    4
+    """
+
+
 def pdb_invoke(method, arg):
     """Run pdb.method(arg)."""
     import pdb; getattr(pdb, method)(arg)

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Fri Jul 30 14:01:20 2010
@@ -475,6 +475,9 @@
 Library
 -------
 
+- Issue #809887: Make the output of pdb's breakpoint deletions more
+  consistent; emit a message when a breakpoint is enabled or disabled.
+
 - Issue #5294: Fix the behavior of pdb's "continue" command when called
   in the top-level debugged frame.
 

From python-checkins at python.org  Fri Jul 30 14:58:44 2010
From: python-checkins at python.org (mark.dickinson)
Date: Fri, 30 Jul 2010 14:58:44 +0200 (CEST)
Subject: [Python-checkins] r83276 -
	python/branches/release27-maint/Doc/tutorial/floatingpoint.rst
Message-ID: <20100730125844.A2CEDEE9E4@mail.python.org>

Author: mark.dickinson
Date: Fri Jul 30 14:58:44 2010
New Revision: 83276

Log:
Update the floating-point section of the tutorial for the short float repr.

Modified:
   python/branches/release27-maint/Doc/tutorial/floatingpoint.rst

Modified: python/branches/release27-maint/Doc/tutorial/floatingpoint.rst
==============================================================================
--- python/branches/release27-maint/Doc/tutorial/floatingpoint.rst	(original)
+++ python/branches/release27-maint/Doc/tutorial/floatingpoint.rst	Fri Jul 30 14:58:44 2010
@@ -48,32 +48,37 @@
 
    0.0001100110011001100110011001100110011001100110011...
 
-Stop at any finite number of bits, and you get an approximation.  This is why
-you see things like::
+Stop at any finite number of bits, and you get an approximation.  On a typical
+machine, there are 53 bits of precision available, so the value stored
+internally is the binary fraction ::
+
+   0.00011001100110011001100110011001100110011001100110011010
+
+which is close to, but not exactly equal to, 1/10.
+
+It's easy to forget that the stored value is an approximation to the original
+decimal fraction, because of the way that floats are displayed at the
+interpreter prompt.  Python only prints a decimal approximation to the true
+decimal value of the binary approximation stored by the machine.  If Python
+were to print the true decimal value of the binary approximation stored for
+0.1, it would have to display ::
 
    >>> 0.1
-   0.10000000000000001
+   0.1000000000000000055511151231257827021181583404541015625
 
-On most machines today, that is what you'll see if you enter 0.1 at a Python
-prompt.  You may not, though, because the number of bits used by the hardware to
-store floating-point values can vary across machines, and Python only prints a
-decimal approximation to the true decimal value of the binary approximation
-stored by the machine.  On most machines, if Python were to print the true
-decimal value of the binary approximation stored for 0.1, it would have to
-display ::
+That is more digits than most people find useful, so Python keeps the number
+of digits manageable by displaying a rounded value instead ::
 
    >>> 0.1
-   0.1000000000000000055511151231257827021181583404541015625
+   0.1
 
-instead!  The Python prompt uses the built-in :func:`repr` function to obtain a
-string version of everything it displays.  For floats, ``repr(float)`` rounds
-the true decimal value to 17 significant digits, giving ::
-
-   0.10000000000000001
-
-``repr(float)`` produces 17 significant digits because it turns out that's
-enough (on most machines) so that ``eval(repr(x)) == x`` exactly for all finite
-floats *x*, but rounding to 16 digits is not enough to make that true.
+It's important to realize that this is, in a real sense, an illusion: the value
+in the machine is not exactly 1/10, you're simply rounding the *display* of the
+true machine value.  This fact becomes apparent as soon as you try to do
+arithmetic with these values ::
+
+   >>> 0.1 + 0.2
+   0.30000000000000004
 
 Note that this is in the very nature of binary floating-point: this is not a bug
 in Python, and it is not a bug in your code either.  You'll see the same kind of
@@ -81,31 +86,32 @@
 (although some languages may not *display* the difference by default, or in all
 output modes).
 
-Python's built-in :func:`str` function produces only 12 significant digits, and
-you may wish to use that instead.  It's unusual for ``eval(str(x))`` to
-reproduce *x*, but the output may be more pleasant to look at::
-
-   >>> print str(0.1)
-   0.1
-
-It's important to realize that this is, in a real sense, an illusion: the value
-in the machine is not exactly 1/10, you're simply rounding the *display* of the
-true machine value.
-
-Other surprises follow from this one.  For example, after seeing ::
-
-   >>> 0.1
-   0.10000000000000001
+Other surprises follow from this one.  For example, if you try to round the value
+2.675 to two decimal places, you get this ::
 
-you may be tempted to use the :func:`round` function to chop it back to the
-single digit you expect.  But that makes no difference::
+   >>> round(2.675, 2)
+   2.67
 
-   >>> round(0.1, 1)
-   0.10000000000000001
-
-The problem is that the binary floating-point value stored for "0.1" was already
-the best possible binary approximation to 1/10, so trying to round it again
-can't make it better:  it was already as good as it gets.
+The documentation for the built-in :func:`round` function says that it rounds
+to the nearest value, rounding ties away from zero.  Since the decimal fraction
+2.675 is exactly halfway between 2.67 and 2.68, you might expect the result
+here to be (a binary approximation to) 2.68.  It's not, because when the
+decimal literal ``2.675`` is converted to a binary floating-point number, it's
+again replaced with a binary approximation, whose exact value is ::
+
+   2.67499999999999982236431605997495353221893310546875
+
+Since this approximation is slightly closer to 2.67 than to 2.68, it's rounded
+down.
+
+If you're in a situation where you care which way your decimal halfway-cases
+are rounded, you should consider using the :mod:`decimal` module.
+Incidentally, the :mod:`decimal` module also provides a nice way to "see" the
+exact value that's stored in any particular Python float ::
+
+   >>> from decimal import Decimal
+   >>> Decimal(2.675)
+   Decimal('2.67499999999999982236431605997495353221893310546875')
 
 Another consequence is that since 0.1 is not exactly 1/10, summing ten values of
 0.1 may not yield exactly 1.0, either::
@@ -150,15 +156,15 @@
 This is the chief reason why Python (or Perl, C, C++, Java, Fortran, and many
 others) often won't display the exact decimal number you expect::
 
-   >>> 0.1
-   0.10000000000000001
+   >>> 0.1 + 0.2
+   0.30000000000000004
 
-Why is that?  1/10 is not exactly representable as a binary fraction. Almost all
-machines today (November 2000) use IEEE-754 floating point arithmetic, and
-almost all platforms map Python floats to IEEE-754 "double precision".  754
-doubles contain 53 bits of precision, so on input the computer strives to
-convert 0.1 to the closest fraction it can of the form *J*/2**\ *N* where *J* is
-an integer containing exactly 53 bits.  Rewriting ::
+Why is that?  1/10 and 2/10 are not exactly representable as a binary
+fraction. Almost all machines today (July 2010) use IEEE-754 floating point
+arithmetic, and almost all platforms map Python floats to IEEE-754 "double
+precision".  754 doubles contain 53 bits of precision, so on input the computer
+strives to convert 0.1 to the closest fraction it can of the form *J*/2**\ *N*
+where *J* is an integer containing exactly 53 bits.  Rewriting ::
 
    1 / 10 ~= J / (2**N)
 
@@ -211,9 +217,8 @@
    100000000000000005551115123125L
 
 meaning that the exact number stored in the computer is approximately equal to
-the decimal value 0.100000000000000005551115123125.  Rounding that to 17
-significant digits gives the 0.10000000000000001 that Python displays (well,
-will display on any 754-conforming platform that does best-possible input and
-output conversions in its C library --- yours may not!).
-
-
+the decimal value 0.100000000000000005551115123125.  In versions prior to
+Python 2.7 and Python 3.1, Python rounded this value to 17 significant digits,
+giving '0.10000000000000001'.  In current versions, Python displays a value based
+on the shortest decimal fraction that rounds correctly back to the true binary
+value, resulting simply in '0.1'.

From python-checkins at python.org  Fri Jul 30 15:13:02 2010
From: python-checkins at python.org (mark.dickinson)
Date: Fri, 30 Jul 2010 15:13:02 +0200 (CEST)
Subject: [Python-checkins] r83277 -
	python/branches/py3k/Doc/library/functions.rst
Message-ID: <20100730131302.31197EEA03@mail.python.org>

Author: mark.dickinson
Date: Fri Jul 30 15:13:02 2010
New Revision: 83277

Log:
Add note about surprising behaviour from round function.

Modified:
   python/branches/py3k/Doc/library/functions.rst

Modified: python/branches/py3k/Doc/library/functions.rst
==============================================================================
--- python/branches/py3k/Doc/library/functions.rst	(original)
+++ python/branches/py3k/Doc/library/functions.rst	Fri Jul 30 15:13:02 2010
@@ -972,6 +972,13 @@
    The return value is an integer if called with one argument, otherwise of the
    same type as *x*.
 
+   .. note::
+
+      The behavior of :func:`round` for floats can be surprising: for example,
+      ``round(2.675, 2)`` gives ``2.67`` instead of the expected ``2.68``.
+      This is not a bug: it's a result of the fact that most decimal fractions
+      can't be represented exactly as a float.  See :ref:`tut-fp-issues` for
+      more information.
 
 .. function:: set([iterable])
    :noindex:

From python-checkins at python.org  Fri Jul 30 15:13:55 2010
From: python-checkins at python.org (mark.dickinson)
Date: Fri, 30 Jul 2010 15:13:55 +0200 (CEST)
Subject: [Python-checkins] r83278 - in python/branches/release31-maint:
	Doc/library/functions.rst
Message-ID: <20100730131355.03055EE9FC@mail.python.org>

Author: mark.dickinson
Date: Fri Jul 30 15:13:54 2010
New Revision: 83278

Log:
Merged revisions 83277 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83277 | mark.dickinson | 2010-07-30 14:13:02 +0100 (Fri, 30 Jul 2010) | 1 line
  
  Add note about surprising behaviour from round function.
........


Modified:
   python/branches/release31-maint/   (props changed)
   python/branches/release31-maint/Doc/library/functions.rst

Modified: python/branches/release31-maint/Doc/library/functions.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/functions.rst	(original)
+++ python/branches/release31-maint/Doc/library/functions.rst	Fri Jul 30 15:13:54 2010
@@ -961,6 +961,13 @@
    The return value is an integer if called with one argument, otherwise of the
    same type as *x*.
 
+   .. note::
+
+      The behavior of :func:`round` for floats can be surprising: for example,
+      ``round(2.675, 2)`` gives ``2.67`` instead of the expected ``2.68``.
+      This is not a bug: it's a result of the fact that most decimal fractions
+      can't be represented exactly as a float.  See :ref:`tut-fp-issues` for
+      more information.
 
 .. function:: set([iterable])
    :noindex:

From python-checkins at python.org  Fri Jul 30 15:16:08 2010
From: python-checkins at python.org (mark.dickinson)
Date: Fri, 30 Jul 2010 15:16:08 +0200 (CEST)
Subject: [Python-checkins] r83279 - in python/branches/release27-maint:
	Doc/library/functions.rst
Message-ID: <20100730131608.111E1EEA07@mail.python.org>

Author: mark.dickinson
Date: Fri Jul 30 15:16:07 2010
New Revision: 83279

Log:
Merged revisions 83277 via svnmerge from 
svn+ssh://pythondev at svn.python.org/python/branches/py3k

........
  r83277 | mark.dickinson | 2010-07-30 14:13:02 +0100 (Fri, 30 Jul 2010) | 1 line
  
  Add note about surprising behaviour from round function.
........


Modified:
   python/branches/release27-maint/   (props changed)
   python/branches/release27-maint/Doc/library/functions.rst

Modified: python/branches/release27-maint/Doc/library/functions.rst
==============================================================================
--- python/branches/release27-maint/Doc/library/functions.rst	(original)
+++ python/branches/release27-maint/Doc/library/functions.rst	Fri Jul 30 15:16:07 2010
@@ -1095,6 +1095,14 @@
    example, ``round(0.5)`` is ``1.0`` and ``round(-0.5)`` is ``-1.0``).
 
 
+   .. note::
+
+      The behavior of :func:`round` for floats can be surprising: for example,
+      ``round(2.675, 2)`` gives ``2.67`` instead of the expected ``2.68``.
+      This is not a bug: it's a result of the fact that most decimal fractions
+      can't be represented exactly as a float.  See :ref:`tut-fp-issues` for
+      more information.
+
 .. function:: set([iterable])
    :noindex:
 

From python-checkins at python.org  Fri Jul 30 15:28:06 2010
From: python-checkins at python.org (ezio.melotti)
Date: Fri, 30 Jul 2010 15:28:06 +0200 (CEST)
Subject: [Python-checkins] r83280 -
	tracker/instances/python-dev/detectors/sendmail.py
Message-ID: <20100730132806.89C95EEA07@mail.python.org>

Author: ezio.melotti
Date: Fri Jul 30 15:28:06 2010
New Revision: 83280

Log:
#279: Removing a message sends an empty changenote to nosy users.

Modified:
   tracker/instances/python-dev/detectors/sendmail.py

Modified: tracker/instances/python-dev/detectors/sendmail.py
==============================================================================
--- tracker/instances/python-dev/detectors/sendmail.py	(original)
+++ tracker/instances/python-dev/detectors/sendmail.py	Fri Jul 30 15:28:06 2010
@@ -61,7 +61,7 @@
         pass
 
     # New submission? 
-    if None == oldvalues:
+    if oldvalues == None:
         changenote = cl.generateCreateNote(nodeid)
         try:
             # Add triage addresses
@@ -71,7 +71,8 @@
         oldfiles = []
     else:
         changenote = cl.generateChangeNote(nodeid, oldvalues)
-        oldfiles = oldvalues.get('files', [])        
+        oldfiles = oldvalues.get('files', [])
+        oldmsglist = oldvalues.get('messages', [])
 
     # Silence nosy_count/message_count
     lines = changenote.splitlines()
@@ -89,9 +90,14 @@
             changenote+="\nAdded file: %s" % url
         for fid in removed:
             url = db.config.TRACKER_WEB + "file%s/%s" % \
-                  (fid, db.file.get(fid, "name"))            
+                  (fid, db.file.get(fid, "name"))
             changenote+="\nRemoved file: %s" % url
 
+    # detect if any of the messages has been removed
+    newmsglist = db.issue.get(nodeid, 'messages', [])
+    for msgid in set(oldmsglist)-set(newmsglist):
+        url = db.config.TRACKER_WEB + "msg%s" % msgid
+        changenote += "\nRemoved message: %s" % url
 
     authid = db.getuid()
 

From python-checkins at python.org  Fri Jul 30 15:36:43 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 15:36:43 +0200 (CEST)
Subject: [Python-checkins] r83281 - python/branches/py3k/Misc/maintainers.rst
Message-ID: <20100730133643.B0A6DEE9DA@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 15:36:43 2010
New Revision: 83281

Log:
Add myself for pdb.

Modified:
   python/branches/py3k/Misc/maintainers.rst

Modified: python/branches/py3k/Misc/maintainers.rst
==============================================================================
--- python/branches/py3k/Misc/maintainers.rst	(original)
+++ python/branches/py3k/Misc/maintainers.rst	Fri Jul 30 15:36:43 2010
@@ -147,7 +147,7 @@
 os
 ossaudiodev
 parser
-pdb
+pdb                 georg.brandl
 pickle              alexandre.vassalotti, pitrou
 pickletools         alexandre.vassalotti
 pipes

From python-checkins at python.org  Fri Jul 30 16:14:42 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 16:14:42 +0200 (CEST)
Subject: [Python-checkins] r83282 - in python/branches/release27-maint/Lib:
	pdb.py test/test_pdb.py
Message-ID: <20100730141442.87C8EEEA26@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 16:14:42 2010
New Revision: 83282

Log:
Revert r83267, as it breaks a few doctests and generally leads to ugly truncated output.

Modified:
   python/branches/release27-maint/Lib/pdb.py
   python/branches/release27-maint/Lib/test/test_pdb.py

Modified: python/branches/release27-maint/Lib/pdb.py
==============================================================================
--- python/branches/release27-maint/Lib/pdb.py	(original)
+++ python/branches/release27-maint/Lib/pdb.py	Fri Jul 30 16:14:42 2010
@@ -237,7 +237,7 @@
             if type(t) == type(''):
                 exc_type_name = t
             else: exc_type_name = t.__name__
-            print >>self.stdout, '***', exc_type_name + ':', _saferepr(v)
+            print >>self.stdout, '***', exc_type_name + ':', v
 
     def precmd(self, line):
         """Handle alias expansion and ';;' separator."""
@@ -753,7 +753,7 @@
             if isinstance(t, str):
                 exc_type_name = t
             else: exc_type_name = t.__name__
-            print >>self.stdout, '***', exc_type_name + ':', _saferepr(v)
+            print >>self.stdout, '***', exc_type_name + ':', repr(v)
             raise
 
     def do_p(self, arg):

Modified: python/branches/release27-maint/Lib/test/test_pdb.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_pdb.py	(original)
+++ python/branches/release27-maint/Lib/test/test_pdb.py	Fri Jul 30 16:14:42 2010
@@ -126,33 +126,6 @@
     """
 
 
-def test_pdb_unicode_exception():
-    r"""This tests exceptions that cannot be displayed due to Unicode issues.
-    http://bugs.python.org/issue7539
-
-    >>> def test_function():
-    ...     import pdb; pdb.Pdb().set_trace()
-    ...     pass
-
-    >>> def raising_function():
-    ...     raise ValueError(u"\xff")
-
-    >>> with PdbTestInput([
-    ...     'raising_function()',
-    ...     'p raising_function()',
-    ...     'continue',
-    ... ]):
-    ...     test_function()
-    > <doctest test.test_pdb.test_pdb_unicode_exception[0]>(3)test_function()
-    -> pass
-    (Pdb) raising_function()
-    *** ValueError: ValueError(u'\xff',)
-    (Pdb) p raising_function()
-    *** ValueError: ValueError(u'\xff',)
-    (Pdb) continue
-    """
-
-
 def test_main():
     from test import test_pdb
     test_support.run_doctest(test_pdb, verbosity=True)

From python-checkins at python.org  Fri Jul 30 16:16:43 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 16:16:43 +0200 (CEST)
Subject: [Python-checkins] r83283 - python/branches/py3k/Lib/test/test_pdb.py
Message-ID: <20100730141643.4E586EEA1F@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 16:16:43 2010
New Revision: 83283

Log:
#7964 followup: add test case to ensure issue remains fixed.

Modified:
   python/branches/py3k/Lib/test/test_pdb.py

Modified: python/branches/py3k/Lib/test/test_pdb.py
==============================================================================
--- python/branches/py3k/Lib/test/test_pdb.py	(original)
+++ python/branches/py3k/Lib/test/test_pdb.py	Fri Jul 30 16:16:43 2010
@@ -1,7 +1,10 @@
 # A test suite for pdb; not very comprehensive at the moment.
 
 import imp
+import pdb
 import sys
+import unittest
+import subprocess
 
 from test import support
 # This little helper class is essential for testing pdb under doctest.
@@ -286,9 +289,30 @@
     """
 
 
+class PdbTestCase(unittest.TestCase):
+
+    def test_issue7964(self):
+        # open the file as binary so we can force \r\n newline
+        with open(support.TESTFN, 'wb') as f:
+            f.write(b'print("testing my pdb")\r\n')
+        cmd = [sys.executable, '-m', 'pdb', support.TESTFN]
+        proc = subprocess.Popen(cmd,
+            stdout=subprocess.PIPE,
+            stdin=subprocess.PIPE,
+            stderr=subprocess.STDOUT,
+            )
+        stdout, stderr = proc.communicate(b'quit\n')
+        self.assertNotIn(b'SyntaxError', stdout,
+                         "Got a syntax error running test script under PDB")
+
+    def tearDown(self):
+        support.unlink(support.TESTFN)
+
+
 def test_main():
     from test import test_pdb
     support.run_doctest(test_pdb, verbosity=True)
+    support.run_unittest(PdbTestCase)
 
 
 if __name__ == '__main__':

From python-checkins at python.org  Fri Jul 30 17:01:27 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 17:01:27 +0200 (CEST)
Subject: [Python-checkins] r83284 - in python/branches/py3k:
	Doc/library/bdb.rst Lib/bdb.py
Message-ID: <20100730150127.53D1BEEA14@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 17:01:23 2010
New Revision: 83284

Log:
Add Breakpoint.bpformat(), which returns the info usually printed by bpprint().  Necessary for major refactoring of pdb output handling.

Modified:
   python/branches/py3k/Doc/library/bdb.rst
   python/branches/py3k/Lib/bdb.py

Modified: python/branches/py3k/Doc/library/bdb.rst
==============================================================================
--- python/branches/py3k/Doc/library/bdb.rst	(original)
+++ python/branches/py3k/Doc/library/bdb.rst	Fri Jul 30 17:01:23 2010
@@ -50,9 +50,10 @@
       Mark the breakpoint as disabled.
 
 
-   .. method:: bpprint(out=None)
+   .. method:: bpformat()
 
-      Print all the information about the breakpoint:
+      Return a string with all the information about the breakpoint, nicely
+      formatted:
 
       * The breakpoint number.
       * If it is temporary or not.
@@ -61,6 +62,13 @@
       * If it must be ignored the next N times.
       * The breakpoint hit count.
 
+      .. versionadded:: 3.2
+
+   .. method:: bpprint(out=None)
+
+      Print the output of :meth:`bpformat` to the file *out*, or if it is
+      ``None``, to standard output.
+
 
 .. class:: Bdb(skip=None)
 

Modified: python/branches/py3k/Lib/bdb.py
==============================================================================
--- python/branches/py3k/Lib/bdb.py	(original)
+++ python/branches/py3k/Lib/bdb.py	Fri Jul 30 17:01:23 2010
@@ -499,6 +499,9 @@
     def bpprint(self, out=None):
         if out is None:
             out = sys.stdout
+        print(self.bpformat(), file=out)
+
+    def bpformat(self):
         if self.temporary:
             disp = 'del  '
         else:
@@ -507,17 +510,19 @@
             disp = disp + 'yes  '
         else:
             disp = disp + 'no   '
-        print('%-4dbreakpoint   %s at %s:%d' % (self.number, disp,
-                                                       self.file, self.line), file=out)
+        ret = '%-4dbreakpoint   %s at %s:%d' % (self.number, disp,
+                                                self.file, self.line)
         if self.cond:
-            print('\tstop only if %s' % (self.cond,), file=out)
+            ret += '\n\tstop only if %s' % (self.cond,)
         if self.ignore:
-            print('\tignore next %d hits' % (self.ignore), file=out)
-        if (self.hits):
-            if (self.hits > 1): ss = 's'
-            else: ss = ''
-            print(('\tbreakpoint already hit %d time%s' %
-                          (self.hits, ss)), file=out)
+            ret += '\n\tignore next %d hits' % (self.ignore,)
+        if self.hits:
+            if self.hits > 1:
+                ss = 's'
+            else:
+                ss = ''
+            ret += '\n\tbreakpoint already hit %d time%s' % (self.hits, ss)
+        return ret
 
     def __str__(self):
         return 'breakpoint %s at %s:%s' % (self.number, self.file, self.line)

From python-checkins at python.org  Fri Jul 30 17:33:52 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 17:33:52 +0200 (CEST)
Subject: [Python-checkins] r83285 -
	python/branches/py3k/Lib/test/test_sundry.py
Message-ID: <20100730153352.78F8EEE9C4@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 17:33:52 2010
New Revision: 83285

Log:
pdb now has its own tests.

Modified:
   python/branches/py3k/Lib/test/test_sundry.py

Modified: python/branches/py3k/Lib/test/test_sundry.py
==============================================================================
--- python/branches/py3k/Lib/test/test_sundry.py	(original)
+++ python/branches/py3k/Lib/test/test_sundry.py	Fri Jul 30 17:33:52 2010
@@ -54,7 +54,6 @@
             import nturl2path
             import opcode
             import os2emxpath
-            import pdb
             import pstats
             import py_compile
             import rlcompleter

From python-checkins at python.org  Fri Jul 30 18:00:46 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 18:00:46 +0200 (CEST)
Subject: [Python-checkins] r83286 - in python/branches/py3k/Lib: pdb.py
	test/test_doctest.py test/test_pdb.py
Message-ID: <20100730160046.F39BDEE9C4@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 18:00:46 2010
New Revision: 83286

Log:
Several enhancements to pdb and its test suite.

* added basic test for basic commands
* removed duplication of command docs, and moved them to their implementation
* unified and useful display of exceptions
* output messages and errors using overridable methods (also fixes #1503502)


Modified:
   python/branches/py3k/Lib/pdb.py
   python/branches/py3k/Lib/test/test_doctest.py
   python/branches/py3k/Lib/test/test_pdb.py

Modified: python/branches/py3k/Lib/pdb.py
==============================================================================
--- python/branches/py3k/Lib/pdb.py	(original)
+++ python/branches/py3k/Lib/pdb.py	Fri Jul 30 18:00:46 2010
@@ -62,176 +62,14 @@
 Debugger commands
 =================
 
-h(elp)
-        Without argument, print the list of available commands.  With
-        a command name as argument, print help about that command.
-
-w(here)
-        Print a stack trace, with the most recent frame at the bottom.
-        An arrow indicates the "current frame", which determines the
-        context of most commands.
-
-d(own) [count]
-        Move the current frame count (default one) levels down in the
-        stack trace (to a newer frame).
-
-u(p) [count]
-        Move the current frame count (default one) levels up in the
-        stack trace (to an older frame).
-
-b(reak) [ ([filename:]lineno | function) [, condition] ]
-        With a filename:lineno argument, set a break there.  If
-        filename is omitted, use the current file.  With a function
-        name, set a break at the first executable line of that
-        function.  Without argument, list all breaks.  Each breakpoint
-        is assigned a number to which all the other breakpoint
-        commands refer.
-
-        The condition argument, if present, is a string which must
-        evaluate to true in order for the breakpoint to be honored.
-
-tbreak [ ([filename:]lineno | function) [, condition] ]
-        Temporary breakpoint, which is removed automatically when it
-        is first hit.  The arguments are the same as for break.
-
-cl(ear) [bpnumber [bpnumber ...] ]
-        With a space separated list of breakpoint numbers, clear those
-        breakpoints.  Without argument, clear all breaks (but first
-        ask confirmation).
-
-disable bpnumber [bpnumber ...]
-        Disable the breakpoints given as a space separated list of
-        breakpoint numbers.  Disabling a breakpoint means it cannot
-        cause the program to stop execution, but unlike clearing a
-        breakpoint, it remains in the list of breakpoints and can be
-        (re-)enabled.
-
-enable bpnumber [bpnumber ...]
-        Enable the breakpoints specified.
-
-ignore bpnumber [count]
-        Set the ignore count for the given breakpoint number.  If
-        count is omitted, the ignore count is set to 0.  A breakpoint
-        becomes active when the ignore count is zero.  When non-zero,
-        the count is decremented each time the breakpoint is reached
-        and the breakpoint is not disabled and any associated
-        condition evaluates to true.
-
-condition bpnumber [condition]
-        Set a new condition for the breakpoint, an expression which
-        must evaluate to true before the breakpoint is honored.  If
-        condition is absent, any existing condition is removed; i.e.,
-        the breakpoint is made unconditional.
-
-commands [bpnumber]
-        Specify a list of commands for the breakpoint.  Type a line
-        containing just 'end' to terminate the commands.  The commands
-        are executed when the breakpoint is hit.
-
-        With no breakpoint number argument, refers to the last
-        breakpoint set.
-
-s(tep)
-        Execute the current line, stop at the first possible occasion
-        (either in a function that is called or in the current
-        function).
-
-n(ext)
-        Continue execution until the next line in the current function
-        is reached or it returns.
-
-unt(il) [lineno]
-        Without argument, continue execution until the line with a
-        number greater than the current one is reached.  With a line
-        number, continue execution until a line with a number greater
-        or equal to that is reached.  In both cases, also stop when
-        the current frame returns.
-
-j(ump) lineno
-        Set the next line that will be executed.  Only available in
-        the bottom-most frame.  This lets you jump back and execute
-        code again, or jump forward to skip code that you don't want
-        to run.
-
-        It should be noted that not all jumps are allowed -- for
-        instance it is not possible to jump into the middle of a
-        for loop or out of a finally clause.
-
-r(eturn)
-        Continue execution until the current function returns.
-
-run [args...]
-        Restart the debugged python program. If a string is supplied
-        it is splitted with "shlex", and the result is used as the new
-        sys.argv.  History, breakpoints, actions and debugger options
-        are preserved.  "restart" is an alias for "run".
-
-c(ont(inue))
-        Continue execution, only stop when a breakpoint is encountered.
-
-l(ist) [first [,last]]
-        List source code for the current file.
-        Without arguments, list 11 lines around the current line
-        or continue the previous listing.
-        With . as argument, list 11 lines around the current line.
-        With one argument, list 11 lines starting at that line.
-        With two arguments, list the given range;
-        if the second argument is less than the first, it is a count.
-
-a(rgs)
-        Print the argument list of the current function.
-
-p expression
-        Print the value of the expression.
-
-(!) statement
-        Execute the (one-line) statement in the context of the current
-        stack frame.  The exclamation point can be omitted unless the
-        first word of the statement resembles a debugger command.  To
-        assign to a global variable you must always prefix the command
-        with a 'global' command, e.g.:
-        (Pdb) global list_options; list_options = ['-l']
-        (Pdb)
-
-
-whatis arg
-        Print the type of the argument.
-
-alias [name [command]]
-        Creates an alias called 'name' that executes 'command'.  The
-        command must *not* be enclosed in quotes.  Replaceable
-        parameters can be indicated by %1, %2, and so on, while %* is
-        replaced by all the parameters.  If no command is given, the
-        current alias for name is shown. If no name is given, all
-        aliases are listed.
-
-        Aliases may be nested and can contain anything that can be
-        legally typed at the pdb prompt.  Note!  You *can* override
-        internal pdb commands with aliases!  Those internal commands
-        are then hidden until the alias is removed.  Aliasing is
-        recursively applied to the first word of the command line; all
-        other words in the line are left alone.
-
-        As an example, here are two useful aliases (especially when
-        placed in the .pdbrc file):
-
-        # Print instance variables (usage "pi classInst")
-        alias pi for k in %1.__dict__.keys(): print "%1.",k,"=",%1.__dict__[k]
-        # Print instance variables in self
-        alias ps pi self
-
-unalias name
-        Delete the specified alias.
-
-q(uit)
-        Quit from the debugger. The program being executed is aborted.
 """
+# NOTE: the actual command documentation is collected from docstrings of the
+# commands and is appended to __doc__ after the class has been defined.
 
 import sys
 import linecache
 import cmd
 import bdb
-from reprlib import Repr
 import os
 import re
 import pprint
@@ -242,12 +80,6 @@
     """Causes a debugger to be restarted for the debugged python program."""
     pass
 
-# Create a custom safe Repr instance and increase its maxstring.
-# The default of 30 truncates error messages too easily.
-_repr = Repr()
-_repr.maxstring = 200
-_saferepr = _repr.repr
-
 __all__ = ["run", "pm", "Pdb", "runeval", "runctx", "runcall", "set_trace",
            "post_mortem", "help"]
 
@@ -374,7 +206,7 @@
         if self._wait_for_mainpyfile:
             return
         if self.stop_here(frame):
-            print('--Call--', file=self.stdout)
+            self.message('--Call--')
             self.interaction(frame, None)
 
     def user_line(self, frame):
@@ -416,7 +248,7 @@
         if self._wait_for_mainpyfile:
             return
         frame.f_locals['__return__'] = return_value
-        print('--Return--', file=self.stdout)
+        self.message('--Return--')
         self.interaction(frame, None)
 
     def user_exception(self, frame, exc_info):
@@ -426,8 +258,8 @@
             return
         exc_type, exc_value, exc_traceback = exc_info
         frame.f_locals['__exception__'] = exc_type, exc_value
-        exc_type_name = exc_type.__name__
-        print(exc_type_name + ':', _saferepr(exc_value), file=self.stdout)
+        self.message(traceback.format_exception_only(exc_type,
+                                                     exc_value)[-1].strip())
         self.interaction(frame, exc_traceback)
 
     # General interaction function
@@ -448,7 +280,7 @@
         """
         # reproduce the behavior of the standard displayhook, not printing None
         if obj is not None:
-            print(repr(obj))
+            self.message(repr(obj))
 
     def default(self, line):
         if line[:1] == '!': line = line[1:]
@@ -469,11 +301,8 @@
                 sys.stdin = save_stdin
                 sys.displayhook = save_displayhook
         except:
-            t, v = sys.exc_info()[:2]
-            if type(t) == type(''):
-                exc_type_name = t
-            else: exc_type_name = t.__name__
-            print('***', exc_type_name + ':', v, file=self.stdout)
+            exc_info = sys.exc_info()[:2]
+            self.error(traceback.format_exception_only(*exc_info)[-1].strip())
 
     def precmd(self, line):
         """Handle alias expansion and ';;' separator."""
@@ -540,25 +369,62 @@
             return 1
         return
 
+    # interface abstraction functions
+
+    def message(self, msg):
+        print(msg, file=self.stdout)
+
+    def error(self, msg):
+        print('***', msg, file=self.stdout)
+
     # Command definitions, called by cmdloop()
     # The argument is the remaining string on the command line
     # Return true to exit from the command loop
 
-    do_h = cmd.Cmd.do_help
-
     def do_commands(self, arg):
-        """Defines a list of commands associated to a breakpoint.
+        """commands [bpnumber]
+        (com) ...
+        (com) end
+        (Pdb)
+
+        Specify a list of commands for breakpoint number bpnumber.
+        The commands themselves are entered on the following lines.
+        Type a line containing just 'end' to terminate the commands.
+        The commands are executed when the breakpoint is hit.
+
+        To remove all commands from a breakpoint, type commands and
+        follow it immediately with end; that is, give no commands.
+
+        With no bpnumber argument, commands refers to the last
+        breakpoint set.
 
-        Those commands will be executed whenever the breakpoint causes
-        the program to stop execution."""
+        You can use breakpoint commands to start your program up
+        again.  Simply use the continue command, or step, or any other
+        command that resumes execution.
+
+        Specifying any command resuming execution (currently continue,
+        step, next, return, jump, quit and their abbreviations)
+        terminates the command list (as if that command was
+        immediately followed by end).  This is because any time you
+        resume execution (even with a simple next or step), you may
+        encounter another breakpoint -- which could have its own
+        command list, leading to ambiguities about which list to
+        execute.
+
+        If you use the 'silent' command in the command list, the usual
+        message about stopping at a breakpoint is not printed.  This
+        may be desirable for breakpoints that are to print a specific
+        message and then continue.  If none of the other commands
+        print anything, you will see no sign that the breakpoint was
+        reached.
+        """
         if not arg:
             bnum = len(bdb.Breakpoint.bpbynumber) - 1
         else:
             try:
                 bnum = int(arg)
             except:
-                print("Usage: commands [bnum]\n        ...\n        end",
-                      file=self.stdout)
+                self.error("Usage: commands [bnum]\n        ...\n        end")
                 return
         self.commands_bnum = bnum
         self.commands[bnum] = []
@@ -574,13 +440,26 @@
             self.prompt = prompt_back
 
     def do_break(self, arg, temporary = 0):
-        # break [ ([filename:]lineno | function) [, "condition"] ]
+        """b(reak) [ ([filename:]lineno | function) [, condition] ]
+        Without argument, list all breaks.
+
+        With a line number argument, set a break at this line in the
+        current file.  With a function name, set a break at the first
+        executable line of that function.  If a second argument is
+        present, it is a string specifying an expression which must
+        evaluate to true before the breakpoint is honored.
+
+        The line number may be prefixed with a filename and a colon,
+        to specify a breakpoint in another file (probably one that
+        hasn't been loaded yet).  The file is searched for on
+        sys.path; the .py suffix may be omitted.
+        """
         if not arg:
             if self.breaks:  # There's at least one
-                print("Num Type         Disp Enb   Where", file=self.stdout)
+                self.message("Num Type         Disp Enb   Where")
                 for bp in bdb.Breakpoint.bpbynumber:
                     if bp:
-                        bp.bpprint(self.stdout)
+                        self.message(bp.bpformat())
             return
         # parse arguments; comma has lowest precedence
         # and cannot occur in filename
@@ -599,8 +478,7 @@
             filename = arg[:colon].rstrip()
             f = self.lookupmodule(filename)
             if not f:
-                print('*** ', repr(filename), end=' ', file=self.stdout)
-                print('not found from sys.path', file=self.stdout)
+                self.error('%r not found from sys.path' % filename)
                 return
             else:
                 filename = f
@@ -608,7 +486,7 @@
             try:
                 lineno = int(arg)
             except ValueError as msg:
-                print('*** Bad lineno:', arg, file=self.stdout)
+                self.error('Bad lineno: %s' % arg)
                 return
         else:
             # no colon; can be lineno or function
@@ -634,9 +512,8 @@
                     # last thing to try
                     (ok, filename, ln) = self.lineinfo(arg)
                     if not ok:
-                        print('*** The specified object %r is not a function '
-                              'or was not found along sys.path.' % arg,
-                              file=self.stdout)
+                        self.error('The specified object %r is not a function '
+                                   'or was not found along sys.path.' % arg)
                         return
                     funcname = ok # ok contains a function name
                     lineno = int(ln)
@@ -647,12 +524,12 @@
         if line:
             # now set the break point
             err = self.set_break(filename, line, temporary, cond, funcname)
-            if err: print('***', err, file=self.stdout)
+            if err:
+                self.error(err, file=self.stdout)
             else:
                 bp = self.get_breaks(filename, line)[-1]
-                print("Breakpoint %d at %s:%d" % (bp.number,
-                                                                 bp.file,
-                                                                 bp.line), file=self.stdout)
+                self.message("Breakpoint %d at %s:%d" %
+                             (bp.number, bp.file, bp.line))
 
     # To be overridden in derived debuggers
     def defaultFile(self):
@@ -665,6 +542,10 @@
     do_b = do_break
 
     def do_tbreak(self, arg):
+        """tbreak [ ([filename:]lineno | function) [, condition] ]
+        Same arguments as break, but sets a temporary breakpoint: it
+        is automatically deleted when first hit.
+        """
         self.do_break(arg, 1)
 
     def lineinfo(self, identifier):
@@ -711,40 +592,56 @@
         globs = self.curframe.f_globals if hasattr(self, 'curframe') else None
         line = linecache.getline(filename, lineno, globs)
         if not line:
-            print('End of file', file=self.stdout)
+            self.message('End of file')
             return 0
         line = line.strip()
         # Don't allow setting breakpoint at a blank line
         if (not line or (line[0] == '#') or
              (line[:3] == '"""') or line[:3] == "'''"):
-            print('*** Blank or comment', file=self.stdout)
+            self.error('Blank or comment')
             return 0
         return lineno
 
     def do_enable(self, arg):
+        """enable bpnumber [bpnumber ...]
+        Enables the breakpoints given as a space separated list of
+        breakpoint numbers.
+        """
         args = arg.split()
         for i in args:
             try:
                 bp = self.get_bpbynumber(i)
             except ValueError as err:
-                print('***', err, file=self.stdout)
+                self.error(err)
             else:
                 bp.enable()
-                print('Enabled %s' % bp, file=self.stdout)
+                self.message('Enabled %s' % bp)
 
     def do_disable(self, arg):
+        """disable bpnumber [bpnumber ...]
+        Disables the breakpoints given as a space separated list of
+        breakpoint numbers.  Disabling a breakpoint means it cannot
+        cause the program to stop execution, but unlike clearing a
+        breakpoint, it remains in the list of breakpoints and can be
+        (re-)enabled.
+        """
         args = arg.split()
         for i in args:
             try:
                 bp = self.get_bpbynumber(i)
             except ValueError as err:
-                print('***', err, file=self.stdout)
+                self.error(err)
             else:
                 bp.disable()
-                print('Disabled %s' % bp, file=self.stdout)
+                self.message('Disabled %s' % bp)
 
     def do_condition(self, arg):
-        # arg is breakpoint number and condition
+        """condition bpnumber [condition]
+        Set a new condition for the breakpoint, an expression which
+        must evaluate to true before the breakpoint is honored.  If
+        condition is absent, any existing condition is removed; i.e.,
+        the breakpoint is made unconditional.
+        """
         args = arg.split(' ', 1)
         try:
             cond = args[1]
@@ -753,18 +650,23 @@
         try:
             bp = self.get_bpbynumber(args[0].strip())
         except ValueError as err:
-            print('***', err, file=self.stdout)
+            self.error(err)
         else:
             bp.cond = cond
             if not cond:
-                print('Breakpoint %d is now unconditional.' % bp.number,
-                      file=self.stdout)
+                self.message('Breakpoint %d is now unconditional.' % bp.number)
             else:
-                print('New condition set for breakpoint %d.' % bp.number,
-                      file=self.stdout)
+                self.message('New condition set for breakpoint %d.' % bp.number)
 
     def do_ignore(self, arg):
-        """arg is bp number followed by ignore count."""
+        """ignore bpnumber [count]
+        Set the ignore count for the given breakpoint number.  If
+        count is omitted, the ignore count is set to 0.  A breakpoint
+        becomes active when the ignore count is zero.  When non-zero,
+        the count is decremented each time the breakpoint is reached
+        and the breakpoint is not disabled and any associated
+        condition evaluates to true.
+        """
         args = arg.split()
         try:
             count = int(args[1].strip())
@@ -773,25 +675,27 @@
         try:
             bp = self.get_bpbynumber(args[0].strip())
         except ValueError as err:
-            print('***', err, file=self.stdout)
+            self.error(err)
         else:
             bp.ignore = count
             if count > 0:
-                reply = 'Will ignore next '
                 if count > 1:
-                    reply = reply + '%d crossings' % count
+                    countstr = '%d crossings' % count
                 else:
-                    reply = reply + '1 crossing'
-                print(reply + ' of breakpoint %d.' % bp.number, file=self.stdout)
+                    countstr = '1 crossing'
+                self.message('Will ignore next %s of breakpoint %d.' %
+                             (countstr, bp.number))
             else:
-                print('Will stop next time breakpoint %d is reached.'
-                      % bp.number, file=self.stdout)
+                self.message('Will stop next time breakpoint %d is reached.'
+                             % bp.number)
 
     def do_clear(self, arg):
-        """Three possibilities, tried in this order:
-        clear -> clear all breaks, ask for confirmation
-        clear file:lineno -> clear all breaks at file:lineno
-        clear bpno bpno ... -> clear breakpoints by number"""
+        """cl(ear) filename:lineno\ncl(ear) [bpnumber [bpnumber...]]
+        With a space separated list of breakpoint numbers, clear
+        those breakpoints.  Without argument, clear all breaks (but
+        first ask confirmation).  With a filename:lineno argument,
+        clear all breaks at that line in that file.
+        """
         if not arg:
             try:
                 reply = input('Clear all breaks? ')
@@ -802,7 +706,7 @@
                 bplist = [bp for bp in bdb.Breakpoint.bpbynumber if bp]
                 self.clear_all_breaks()
                 for bp in bplist:
-                    print('Deleted %s' % bp, file=self.stdout)
+                    self.message('Deleted %s' % bp)
             return
         if ':' in arg:
             # Make sure it works for "clear C:\foo\bar.py:12"
@@ -817,23 +721,28 @@
                 bplist = self.get_breaks(filename, lineno)
                 err = self.clear_break(filename, lineno)
             if err:
-                print('***', err, file=self.stdout)
+                self.error(err)
             else:
                 for bp in bplist:
-                    print('Deleted %s' % bp, file=self.stdout)
+                    self.message('Deleted %s' % bp)
             return
         numberlist = arg.split()
         for i in numberlist:
             try:
                 bp = self.get_bpbynumber(i)
             except ValueError as err:
-                print('***', err, file=self.stdout)
+                self.error(err)
             else:
                 self.clear_break(bp.file, bp.line)
-                print('Deleted %s' % bp, file=self.stdout)
+                self.message('Deleted %s' % bp)
     do_cl = do_clear # 'c' is already an abbreviation for 'continue'
 
     def do_where(self, arg):
+        """w(here)
+        Print a stack trace, with the most recent frame at the bottom.
+        An arrow indicates the "current frame", which determines the
+        context of most commands.  'bt' is an alias for this command.
+        """
         self.print_stack_trace()
     do_w = do_where
     do_bt = do_where
@@ -847,13 +756,17 @@
         self.lineno = None
 
     def do_up(self, arg):
+        """u(p) [count]
+        Move the current frame count (default one) levels up in the
+        stack trace (to an older frame).
+        """
         if self.curindex == 0:
-            print('*** Oldest frame', file=self.stdout)
+            self.error('Oldest frame')
             return
         try:
             count = int(arg or 1)
         except ValueError:
-            print('*** Invalid frame count (%s)' % arg, file=self.stdout)
+            self.error('Invalid frame count (%s)' % arg)
             return
         if count < 0:
             newframe = 0
@@ -863,13 +776,17 @@
     do_u = do_up
 
     def do_down(self, arg):
+        """d(own) [count]
+        Move the current frame count (default one) levels down in the
+        stack trace (to a newer frame).
+        """
         if self.curindex + 1 == len(self.stack):
-            print('*** Newest frame', file=self.stdout)
+            self.error('Newest frame')
             return
         try:
             count = int(arg or 1)
         except ValueError:
-            print('*** Invalid frame count (%s)' % arg, file=self.stdout)
+            self.error('Invalid frame count (%s)' % arg)
             return
         if count < 0:
             newframe = len(self.stack) - 1
@@ -879,15 +796,22 @@
     do_d = do_down
 
     def do_until(self, arg):
+        """unt(il) [lineno]
+        Without argument, continue execution until the line with a
+        number greater than the current one is reached.  With a line
+        number, continue execution until a line with a number greater
+        or equal to that is reached.  In both cases, also stop when
+        the current frame returns.
+        """
         if arg:
             try:
                 lineno = int(arg)
             except ValueError:
-                print('*** Error in argument:', repr(arg), file=self.stdout)
+                self.error('Error in argument: %r' % arg)
                 return
             if lineno <= self.curframe.f_lineno:
-                print('*** "until" line number is smaller than current '
-                      'line number', file=self.stdout)
+                self.error('"until" line number is smaller than current '
+                           'line number')
                 return
         else:
             lineno = None
@@ -896,45 +820,75 @@
     do_unt = do_until
 
     def do_step(self, arg):
+        """s(tep)
+        Execute the current line, stop at the first possible occasion
+        (either in a function that is called or in the current
+        function).
+        """
         self.set_step()
         return 1
     do_s = do_step
 
     def do_next(self, arg):
+        """n(ext)
+        Continue execution until the next line in the current function
+        is reached or it returns.
+        """
         self.set_next(self.curframe)
         return 1
     do_n = do_next
 
     def do_run(self, arg):
-        """Restart program by raising an exception to be caught in the main
-        debugger loop.  If arguments were given, set them in sys.argv."""
+        """run [args...]
+        Restart the debugged python program. If a string is supplied
+        it is splitted with "shlex", and the result is used as the new
+        sys.argv.  History, breakpoints, actions and debugger options
+        are preserved.  "restart" is an alias for "run".
+        """
         if arg:
             import shlex
             argv0 = sys.argv[0:1]
             sys.argv = shlex.split(arg)
             sys.argv[:0] = argv0
+        # this is caught in the main debugger loop
         raise Restart
 
     do_restart = do_run
 
     def do_return(self, arg):
+        """r(eturn)
+        Continue execution until the current function returns.
+        """
         self.set_return(self.curframe)
         return 1
     do_r = do_return
 
     def do_continue(self, arg):
+        """c(ont(inue))
+        Continue execution, only stop when a breakpoint is encountered.
+        """
         self.set_continue()
         return 1
     do_c = do_cont = do_continue
 
     def do_jump(self, arg):
+        """j(ump) lineno
+        Set the next line that will be executed.  Only available in
+        the bottom-most frame.  This lets you jump back and execute
+        code again, or jump forward to skip code that you don't want
+        to run.
+
+        It should be noted that not all jumps are allowed -- for
+        instance it is not possible to jump into the middle of a
+        for loop or out of a finally clause.
+        """
         if self.curindex + 1 != len(self.stack):
-            print("*** You can only jump within the bottom frame", file=self.stdout)
+            self.error('You can only jump within the bottom frame')
             return
         try:
             arg = int(arg)
         except ValueError:
-            print("*** The 'jump' command requires a line number.", file=self.stdout)
+            self.error("The 'jump' command requires a line number")
         else:
             try:
                 # Do the jump, fix up our copy of the stack, and display the
@@ -943,22 +897,30 @@
                 self.stack[self.curindex] = self.stack[self.curindex][0], arg
                 self.print_stack_entry(self.stack[self.curindex])
             except ValueError as e:
-                print('*** Jump failed:', e, file=self.stdout)
+                self.error('Jump failed: %s' % e)
     do_j = do_jump
 
     def do_debug(self, arg):
+        """debug code
+        Enter a recursive debugger that steps through the code
+        argument (which is an arbitrary expression or statement to be
+        executed in the current environment).
+        """
         sys.settrace(None)
         globals = self.curframe.f_globals
         locals = self.curframe_locals
         p = Pdb(self.completekey, self.stdin, self.stdout)
         p.prompt = "(%s) " % self.prompt.strip()
-        print("ENTERING RECURSIVE DEBUGGER", file=self.stdout)
+        self.message("ENTERING RECURSIVE DEBUGGER")
         sys.call_tracing(p.run, (arg, globals, locals))
-        print("LEAVING RECURSIVE DEBUGGER", file=self.stdout)
+        self.message("LEAVING RECURSIVE DEBUGGER")
         sys.settrace(self.trace_dispatch)
         self.lastcmd = p.lastcmd
 
     def do_quit(self, arg):
+        """q(uit)\nexit
+        Quit from the debugger. The program being executed is aborted.
+        """
         self._user_requested_quit = 1
         self.set_quit()
         return 1
@@ -967,12 +929,18 @@
     do_exit = do_quit
 
     def do_EOF(self, arg):
-        print(file=self.stdout)
+        """EOF
+        Handles the receipt of EOF as a command.
+        """
+        self.message('')
         self._user_requested_quit = 1
         self.set_quit()
         return 1
 
     def do_args(self, arg):
+        """a(rgs)
+        Print the argument list of the current function.
+        """
         co = self.curframe.f_code
         dict = self.curframe_locals
         n = co.co_argcount
@@ -980,60 +948,76 @@
         if co.co_flags & 8: n = n+1
         for i in range(n):
             name = co.co_varnames[i]
-            print(name, '=', end=' ', file=self.stdout)
-            if name in dict: print(dict[name], file=self.stdout)
-            else: print("*** undefined ***", file=self.stdout)
+            if name in dict:
+                self.message('%s = %r' % (name, dict[name]))
+            else:
+                self.message('%s = *** undefined ***' % (name,))
     do_a = do_args
 
     def do_retval(self, arg):
+        """retval
+        Print the return value for the last return of a function.
+        """
         if '__return__' in self.curframe_locals:
-            print(self.curframe_locals['__return__'], file=self.stdout)
+            self.message(repr(self.curframe_locals['__return__']))
         else:
-            print('*** Not yet returned!', file=self.stdout)
+            self.error('Not yet returned!')
     do_rv = do_retval
 
     def _getval(self, arg):
         try:
             return eval(arg, self.curframe.f_globals, self.curframe_locals)
         except:
-            t, v = sys.exc_info()[:2]
-            if isinstance(t, str):
-                exc_type_name = t
-            else: exc_type_name = t.__name__
-            print('***', exc_type_name + ':', repr(v), file=self.stdout)
+            exc_info = sys.exc_info()[:2]
+            self.error(traceback.format_exception_only(*exc_info)[-1].strip())
             raise
 
     def do_p(self, arg):
+        """p(rint) expression
+        Print the value of the expression.
+        """
         try:
-            print(repr(self._getval(arg)), file=self.stdout)
+            self.message(repr(self._getval(arg)))
         except:
             pass
     # make "print" an alias of "p" since print isn't a Python statement anymore
     do_print = do_p
 
     def do_pp(self, arg):
+        """pp expression
+        Pretty-print the value of the expression.
+        """
         try:
-            pprint.pprint(self._getval(arg), self.stdout)
+            self.message(pprint.pformat(self._getval(arg)))
         except:
             pass
 
     def do_list(self, arg):
+        """l(ist) [first [,last] | .]
+        List source code for the current file.
+        Without arguments, list 11 lines around the current line
+        or continue the previous listing.
+        With . as argument, list 11 lines around the current line.
+        With one argument, list 11 lines starting at that line.
+        With two arguments, list the given range;
+        if the second argument is less than the first, it is a count.
+        """
         self.lastcmd = 'list'
         last = None
         if arg and arg != '.':
             try:
-                x = eval(arg, {}, {})
-                if type(x) == type(()):
-                    first, last = x
-                    first = int(first)
-                    last = int(last)
+                if ',' in arg:
+                    first, last = arg.split(',')
+                    first = int(first.strip())
+                    last = int(last.strip())
                     if last < first:
-                        # Assume it's a count
+                        # assume it's a count
                         last = first + last
                 else:
-                    first = max(1, int(x) - 5)
-            except:
-                print('*** Error in argument:', repr(arg), file=self.stdout)
+                    first = int(arg.strip())
+                    first = max(1, first - 5)
+            except ValueError:
+                self.error('Error in argument: %r' % arg)
                 return
         elif self.lineno is None or arg == '.':
             first = max(1, self.curframe.f_lineno - 5)
@@ -1048,7 +1032,7 @@
                 line = linecache.getline(filename, lineno,
                                          self.curframe.f_globals)
                 if not line:
-                    print('[EOF]', file=self.stdout)
+                    self.message('[EOF]')
                     break
                 else:
                     s = repr(lineno).rjust(3)
@@ -1057,58 +1041,90 @@
                     else: s = s + ' '
                     if lineno == self.curframe.f_lineno:
                         s = s + '->'
-                    print(s + '\t' + line, end='', file=self.stdout)
+                    self.message(s + '\t' + line.rstrip())
                     self.lineno = lineno
         except KeyboardInterrupt:
             pass
     do_l = do_list
 
     def do_whatis(self, arg):
+        """whatis arg
+        Print the type of the argument.
+        """
         try:
-            value = eval(arg, self.curframe.f_globals,
-                            self.curframe_locals)
+            value = self._getval(arg)
         except:
-            t, v = sys.exc_info()[:2]
-            if type(t) == type(''):
-                exc_type_name = t
-            else: exc_type_name = t.__name__
-            print('***', exc_type_name + ':', repr(v), file=self.stdout)
+            # _getval() already printed the error
             return
         code = None
         # Is it a function?
-        try: code = value.__code__
-        except: pass
+        try:
+            code = value.__code__
+        except Exception:
+            pass
         if code:
-            print('Function', code.co_name, file=self.stdout)
+            self.message('Function %s' % code.co_name)
             return
         # Is it an instance method?
-        try: code = value.__func__.__code__
-        except: pass
+        try:
+            code = value.__func__.__code__
+        except Exception:
+            pass
         if code:
-            print('Method', code.co_name, file=self.stdout)
+            self.message('Method %s' % code.co_name)
+            return
+        # Is it a class?
+        if value.__class__ is type:
+            self.message('Class %s.%s' % (value.__module__, value.__name__))
             return
         # None of the above...
-        print(type(value), file=self.stdout)
+        self.message(type(value))
 
     def do_alias(self, arg):
+        """alias [name [command [parameter parameter ...] ]]
+        Create an alias called 'name' that executes 'command'.  The
+        command must *not* be enclosed in quotes.  Replaceable
+        parameters can be indicated by %1, %2, and so on, while %* is
+        replaced by all the parameters.  If no command is given, the
+        current alias for name is shown. If no name is given, all
+        aliases are listed.
+
+        Aliases may be nested and can contain anything that can be
+        legally typed at the pdb prompt.  Note!  You *can* override
+        internal pdb commands with aliases!  Those internal commands
+        are then hidden until the alias is removed.  Aliasing is
+        recursively applied to the first word of the command line; all
+        other words in the line are left alone.
+
+        As an example, here are two useful aliases (especially when
+        placed in the .pdbrc file):
+
+        # Print instance variables (usage "pi classInst")
+        alias pi for k in %1.__dict__.keys(): print "%1.",k,"=",%1.__dict__[k]
+        # Print instance variables in self
+        alias ps pi self
+        """
         args = arg.split()
         if len(args) == 0:
             keys = sorted(self.aliases.keys())
             for alias in keys:
-                print("%s = %s" % (alias, self.aliases[alias]), file=self.stdout)
+                self.message("%s = %s" % (alias, self.aliases[alias]))
             return
         if args[0] in self.aliases and len(args) == 1:
-            print("%s = %s" % (args[0], self.aliases[args[0]]), file=self.stdout)
+            self.message("%s = %s" % (args[0], self.aliases[args[0]]))
         else:
             self.aliases[args[0]] = ' '.join(args[1:])
 
     def do_unalias(self, arg):
+        """unalias name
+        Delete the specified alias.
+        """
         args = arg.split()
         if len(args) == 0: return
         if args[0] in self.aliases:
             del self.aliases[args[0]]
 
-    #list of all the commands making the program resume execution.
+    # List of all the commands making the program resume execution.
     commands_resuming = ['do_continue', 'do_step', 'do_next', 'do_return',
                          'do_quit', 'do_jump']
 
@@ -1130,292 +1146,53 @@
     def print_stack_entry(self, frame_lineno, prompt_prefix=line_prefix):
         frame, lineno = frame_lineno
         if frame is self.curframe:
-            print('>', end=' ', file=self.stdout)
+            prefix = '> '
         else:
-            print(' ', end=' ', file=self.stdout)
-        print(self.format_stack_entry(frame_lineno,
-                                                     prompt_prefix), file=self.stdout)
-
-
-    # Help methods (derived from docstring)
-
-    def help_help(self):
-        self.help_h()
-
-    def help_h(self):
-        print("""h(elp)
-Without argument, print the list of available commands.
-With a command name as argument, print help about that command
-"help pdb" shows the full pdb documentation
-"help exec" gives help on the ! command""", file=self.stdout)
-
-    def help_where(self):
-        self.help_w()
-
-    def help_w(self):
-        print("""w(here)
-Print a stack trace, with the most recent frame at the bottom.
-An arrow indicates the "current frame", which determines the
-context of most commands.  'bt' is an alias for this command.""", file=self.stdout)
-
-    help_bt = help_w
-
-    def help_down(self):
-        self.help_d()
-
-    def help_d(self):
-        print("""d(own)
-Move the current frame one level down in the stack trace
-(to a newer frame).""", file=self.stdout)
-
-    def help_up(self):
-        self.help_u()
-
-    def help_u(self):
-        print("""u(p)
-Move the current frame one level up in the stack trace
-(to an older frame).""", file=self.stdout)
-
-    def help_break(self):
-        self.help_b()
-
-    def help_b(self):
-        print("""b(reak) ([file:]lineno | function) [, condition]
-With a line number argument, set a break there in the current
-file.  With a function name, set a break at first executable line
-of that function.  Without argument, list all breaks.  If a second
-argument is present, it is a string specifying an expression
-which must evaluate to true before the breakpoint is honored.
-
-The line number may be prefixed with a filename and a colon,
-to specify a breakpoint in another file (probably one that
-hasn't been loaded yet).  The file is searched for on sys.path;
-the .py suffix may be omitted.""", file=self.stdout)
-
-    def help_clear(self):
-        self.help_cl()
-
-    def help_cl(self):
-        print("cl(ear) filename:lineno", file=self.stdout)
-        print("""cl(ear) [bpnumber [bpnumber...]]
-With a space separated list of breakpoint numbers, clear
-those breakpoints.  Without argument, clear all breaks (but
-first ask confirmation).  With a filename:lineno argument,
-clear all breaks at that line in that file.""", file=self.stdout)
-
-    def help_tbreak(self):
-        print("""tbreak  same arguments as break, but breakpoint is
-removed when first hit.""", file=self.stdout)
-
-    def help_enable(self):
-        print("""enable bpnumber [bpnumber ...]
-Enables the breakpoints given as a space separated list of
-bp numbers.""", file=self.stdout)
-
-    def help_disable(self):
-        print("""disable bpnumber [bpnumber ...]
-Disables the breakpoints given as a space separated list of
-bp numbers.""", file=self.stdout)
-
-    def help_ignore(self):
-        print("""ignore bpnumber count
-Sets the ignore count for the given breakpoint number.  A breakpoint
-becomes active when the ignore count is zero.  When non-zero, the
-count is decremented each time the breakpoint is reached and the
-breakpoint is not disabled and any associated condition evaluates
-to true.""", file=self.stdout)
-
-    def help_condition(self):
-        print("""condition bpnumber str_condition
-str_condition is a string specifying an expression which
-must evaluate to true before the breakpoint is honored.
-If str_condition is absent, any existing condition is removed;
-i.e., the breakpoint is made unconditional.""", file=self.stdout)
-
-    def help_step(self):
-        self.help_s()
-
-    def help_s(self):
-        print("""s(tep)
-Execute the current line, stop at the first possible occasion
-(either in a function that is called or in the current function).""", file=self.stdout)
-
-    def help_until(self):
-        self.help_unt()
-
-    def help_unt(self):
-        print("""unt(il)
-Continue execution until the line with a number greater than the current
-one is reached or until the current frame returns""")
-
-    def help_next(self):
-        self.help_n()
-
-    def help_n(self):
-        print("""n(ext)
-Continue execution until the next line in the current function
-is reached or it returns.""", file=self.stdout)
-
-    def help_return(self):
-        self.help_r()
-
-    def help_r(self):
-        print("""r(eturn)
-Continue execution until the current function returns.""", file=self.stdout)
-
-    def help_continue(self):
-        self.help_c()
-
-    def help_cont(self):
-        self.help_c()
-
-    def help_c(self):
-        print("""c(ont(inue))
-Continue execution, only stop when a breakpoint is encountered.""", file=self.stdout)
-
-    def help_jump(self):
-        self.help_j()
-
-    def help_j(self):
-        print("""j(ump) lineno
-Set the next line that will be executed.""", file=self.stdout)
-
-    def help_debug(self):
-        print("""debug code
-Enter a recursive debugger that steps through the code argument
-(which is an arbitrary expression or statement to be executed
-in the current environment).""", file=self.stdout)
-
-    def help_list(self):
-        self.help_l()
-
-    def help_l(self):
-        print("""l(ist) [first [,last]]
-List source code for the current file.
-Without arguments, list 11 lines around the current line
-or continue the previous listing.
-With one argument, list 11 lines starting at that line.
-With two arguments, list the given range;
-if the second argument is less than the first, it is a count.""", file=self.stdout)
-
-    def help_args(self):
-        self.help_a()
-
-    def help_a(self):
-        print("""a(rgs)
-Print the arguments of the current function.""", file=self.stdout)
-
-    def help_p(self):
-        print("""p(rint) expression
-Print the value of the expression.""", file=self.stdout)
-
-    def help_pp(self):
-        print("""pp expression
-Pretty-print the value of the expression.""", file=self.stdout)
+            prefix = '  '
+        self.message(prefix +
+                     self.format_stack_entry(frame_lineno, prompt_prefix))
+
+    # Provide help
+
+    def do_help(self, arg):
+        """h(elp)
+        Without argument, print the list of available commands.
+        With a command name as argument, print help about that command.
+        "help pdb" shows the full pdb documentation.
+        "help exec" gives help on the ! command.
+        """
+        if not arg:
+            return cmd.Cmd.do_help(self, arg)
+        try:
+            try:
+                topic = getattr(self, 'help_' + arg)
+                return topic()
+            except AttributeError:
+                command = getattr(self, 'do_' + arg)
+        except AttributeError:
+            self.error('No help for %r' % arg)
+        else:
+            self.message(command.__doc__.rstrip())
+
+    do_h = do_help
 
     def help_exec(self):
-        print("""(!) statement
-Execute the (one-line) statement in the context of
-the current stack frame.
-The exclamation point can be omitted unless the first word
-of the statement resembles a debugger command.
-To assign to a global variable you must always prefix the
-command with a 'global' command, e.g.:
-(Pdb) global list_options; list_options = ['-l']
-(Pdb)""", file=self.stdout)
-
-    def help_run(self):
-        print("""run [args...]
-Restart the debugged python program. If a string is supplied, it is
-splitted with "shlex" and the result is used as the new sys.argv.
-History, breakpoints, actions and debugger options are preserved.
-"restart" is an alias for "run".""")
-
-    help_restart = help_run
-
-    def help_quit(self):
-        self.help_q()
-
-    def help_q(self):
-        print("""q(uit) or exit - Quit from the debugger.
-The program being executed is aborted.""", file=self.stdout)
-
-    help_exit = help_q
-
-    def help_whatis(self):
-        print("""whatis arg
-Prints the type of the argument.""", file=self.stdout)
-
-    def help_EOF(self):
-        print("""EOF
-Handles the receipt of EOF as a command.""", file=self.stdout)
-
-    def help_alias(self):
-        print("""alias [name [command [parameter parameter ...] ]]
-Creates an alias called 'name' the executes 'command'.  The command
-must *not* be enclosed in quotes.  Replaceable parameters are
-indicated by %1, %2, and so on, while %* is replaced by all the
-parameters.  If no command is given, the current alias for name
-is shown. If no name is given, all aliases are listed.
-
-Aliases may be nested and can contain anything that can be
-legally typed at the pdb prompt.  Note!  You *can* override
-internal pdb commands with aliases!  Those internal commands
-are then hidden until the alias is removed.  Aliasing is recursively
-applied to the first word of the command line; all other words
-in the line are left alone.
-
-Some useful aliases (especially when placed in the .pdbrc file) are:
-
-#Print instance variables (usage "pi classInst")
-alias pi for k in %1.__dict__.keys(): print "%1.",k,"=",%1.__dict__[k]
-
-#Print instance variables in self
-alias ps pi self
-""", file=self.stdout)
-
-    def help_unalias(self):
-        print("""unalias name
-Deletes the specified alias.""", file=self.stdout)
-
-    def help_commands(self):
-        print("""commands [bpnumber]
-(com) ...
-(com) end
-(Pdb)
-
-Specify a list of commands for breakpoint number bpnumber.  The
-commands themselves appear on the following lines.  Type a line
-containing just 'end' to terminate the commands.
-
-To remove all commands from a breakpoint, type commands and
-follow it immediately with  end; that is, give no commands.
-
-With no bpnumber argument, commands refers to the last
-breakpoint set.
-
-You can use breakpoint commands to start your program up again.
-Simply use the continue command, or step, or any other
-command that resumes execution.
-
-Specifying any command resuming execution (currently continue,
-step, next, return, jump, quit and their abbreviations) terminates
-the command list (as if that command was immediately followed by end).
-This is because any time you resume execution
-(even with a simple next or step), you may encounter
-another breakpoint--which could have its own command list, leading to
-ambiguities about which list to execute.
-
-   If you use the 'silent' command in the command list, the
-usual message about stopping at a breakpoint is not printed.  This may
-be desirable for breakpoints that are to print a specific message and
-then continue.  If none of the other commands print anything, you
-see no sign that the breakpoint was reached.
-""", file=self.stdout)
+        """(!) statement
+        Execute the (one-line) statement in the context of the current
+        stack frame.  The exclamation point can be omitted unless the
+        first word of the statement resembles a debugger command.  To
+        assign to a global variable you must always prefix the command
+        with a 'global' command, e.g.:
+        (Pdb) global list_options; list_options = ['-l']
+        (Pdb)
+        """
+        self.message(self.help_exec.__doc__.strip())
 
     def help_pdb(self):
         help()
 
+    # other helper functions
+
     def lookupmodule(self, filename):
         """Helper function for break/clear parsing -- may be overridden.
 
@@ -1466,6 +1243,22 @@
                         (fp.read(), self.mainpyfile)
         self.run(statement)
 
+# Collect all command help into docstring
+
+# unfortunately we can't guess this order from the class definition
+_help_order = [
+    'help', 'where', 'down', 'up', 'break', 'tbreak', 'clear', 'disable',
+    'enable', 'ignore', 'condition', 'commands', 'step', 'next', 'until',
+    'jump', 'return', 'retval', 'run', 'continue', 'list', 'args', 'print',
+    'whatis', 'alias', 'unalias', 'quit',
+]
+
+for _command in _help_order:
+    __doc__ += getattr(Pdb, 'do_' + _command).__doc__.strip() + '\n\n'
+__doc__ += Pdb.help_exec.__doc__
+
+del _help_order, _command
+
 # Simplified interface
 
 def run(statement, globals=None, locals=None):
@@ -1492,9 +1285,9 @@
         # sys.exc_info() returns (type, value, traceback) if an exception is
         # being handled, otherwise it returns None
         t = sys.exc_info()[2]
-        if t is None:
-            raise ValueError("A valid traceback must be passed if no "
-                                               "exception is being handled")
+    if t is None:
+        raise ValueError("A valid traceback must be passed if no "
+                         "exception is being handled")
 
     p = Pdb()
     p.reset()

Modified: python/branches/py3k/Lib/test/test_doctest.py
==============================================================================
--- python/branches/py3k/Lib/test/test_doctest.py	(original)
+++ python/branches/py3k/Lib/test/test_doctest.py	Fri Jul 30 18:00:46 2010
@@ -1898,7 +1898,7 @@
     > <doctest foo[1]>(1)<module>()
     -> calls_set_trace()
     (Pdb) print(foo)
-    *** NameError: NameError("name 'foo' is not defined",)
+    *** NameError: name 'foo' is not defined
     (Pdb) continue
     TestResults(failed=0, attempted=2)
 """

Modified: python/branches/py3k/Lib/test/test_pdb.py
==============================================================================
--- python/branches/py3k/Lib/test/test_pdb.py	(original)
+++ python/branches/py3k/Lib/test/test_pdb.py	Fri Jul 30 18:00:46 2010
@@ -54,6 +54,209 @@
     """
 
 
+def test_pdb_basic_commands():
+    """Test the basic commands of pdb.
+
+    >>> def test_function_2(foo, bar='default'):
+    ...     print(foo)
+    ...     for i in range(5):
+    ...         print(i)
+    ...     print(bar)
+    ...     for i in range(10):
+    ...         never_executed
+    ...     print('after for')
+    ...     print('...')
+    ...     return foo.upper()
+
+    >>> def test_function():
+    ...     import pdb; pdb.Pdb().set_trace()
+    ...     ret = test_function_2('baz')
+    ...     print(ret)
+
+    >>> with PdbTestInput([  # doctest: +ELLIPSIS, +NORMALIZE_WHITESPACE
+    ...     'step',       # entering the function call
+    ...     'args',       # display function args
+    ...     'list',       # list function source
+    ...     'bt',         # display backtrace
+    ...     'up',         # step up to test_function()
+    ...     'down',       # step down to test_function_2() again
+    ...     'next',       # stepping to print(foo)
+    ...     'next',       # stepping to the for loop
+    ...     'step',       # stepping into the for loop
+    ...     'until',      # continuing until out of the for loop
+    ...     'next',       # executing the print(bar)
+    ...     'jump 8',     # jump over second for loop
+    ...     'return',     # return out of function
+    ...     'retval',     # display return value
+    ...     'continue',
+    ... ]):
+    ...    test_function()
+    > <doctest test.test_pdb.test_pdb_basic_commands[1]>(3)test_function()
+    -> ret = test_function_2('baz')
+    (Pdb) step
+    --Call--
+    > <doctest test.test_pdb.test_pdb_basic_commands[0]>(1)test_function_2()
+    -> def test_function_2(foo, bar='default'):
+    (Pdb) args
+    foo = 'baz'
+    bar = 'default'
+    (Pdb) list
+      1  ->     def test_function_2(foo, bar='default'):
+      2             print(foo)
+      3             for i in range(5):
+      4                 print(i)
+      5             print(bar)
+      6             for i in range(10):
+      7                 never_executed
+      8             print('after for')
+      9             print('...')
+     10             return foo.upper()
+    [EOF]
+    (Pdb) bt
+    ...
+      <doctest test.test_pdb.test_pdb_basic_commands[2]>(18)<module>()
+    -> test_function()
+      <doctest test.test_pdb.test_pdb_basic_commands[1]>(3)test_function()
+    -> ret = test_function_2('baz')
+    > <doctest test.test_pdb.test_pdb_basic_commands[0]>(1)test_function_2()
+    -> def test_function_2(foo, bar='default'):
+    (Pdb) up
+    > <doctest test.test_pdb.test_pdb_basic_commands[1]>(3)test_function()
+    -> ret = test_function_2('baz')
+    (Pdb) down
+    > <doctest test.test_pdb.test_pdb_basic_commands[0]>(1)test_function_2()
+    -> def test_function_2(foo, bar='default'):
+    (Pdb) next
+    > <doctest test.test_pdb.test_pdb_basic_commands[0]>(2)test_function_2()
+    -> print(foo)
+    (Pdb) next
+    baz
+    > <doctest test.test_pdb.test_pdb_basic_commands[0]>(3)test_function_2()
+    -> for i in range(5):
+    (Pdb) step
+    > <doctest test.test_pdb.test_pdb_basic_commands[0]>(4)test_function_2()
+    -> print(i)
+    (Pdb) until
+    0
+    1
+    2
+    3
+    4
+    > <doctest test.test_pdb.test_pdb_basic_commands[0]>(5)test_function_2()
+    -> print(bar)
+    (Pdb) next
+    default
+    > <doctest test.test_pdb.test_pdb_basic_commands[0]>(6)test_function_2()
+    -> for i in range(10):
+    (Pdb) jump 8
+    > <doctest test.test_pdb.test_pdb_basic_commands[0]>(8)test_function_2()
+    -> print('after for')
+    (Pdb) return
+    after for
+    ...
+    --Return--
+    > <doctest test.test_pdb.test_pdb_basic_commands[0]>(10)test_function_2()->'BAZ'
+    -> return foo.upper()
+    (Pdb) retval
+    'BAZ'
+    (Pdb) continue
+    BAZ
+    """
+
+
+def test_pdb_breakpoint_commands():
+    """Test basic commands related to breakpoints.
+
+    >>> def test_function():
+    ...     import pdb; pdb.Pdb().set_trace()
+    ...     print(1)
+    ...     print(2)
+    ...     print(3)
+    ...     print(4)
+
+    First, need to clear bdb state that might be left over from previous tests.
+    Otherwise, the new breakpoints might get assigned different numbers.
+
+    >>> from bdb import Breakpoint
+    >>> Breakpoint.next = 1
+    >>> Breakpoint.bplist = {}
+    >>> Breakpoint.bpbynumber = [None]
+
+    Now test the breakpoint commands.  NORMALIZE_WHITESPACE is needed because
+    the breakpoint list outputs a tab for the "stop only" and "ignore next"
+    lines, which we don't want to put in here.
+
+    >>> with PdbTestInput([  # doctest: +NORMALIZE_WHITESPACE
+    ...     'break 3',
+    ...     'disable 1',
+    ...     'ignore 1 10',
+    ...     'condition 1 1 < 2',
+    ...     'break 4',
+    ...     'break',
+    ...     'condition 1',
+    ...     'enable 1',
+    ...     'clear 1',
+    ...     'commands 2',
+    ...     'print 42',
+    ...     'end',
+    ...     'continue',  # will stop at breakpoint 2 (line 4)
+    ...     'clear',     # clear all!
+    ...     'y',
+    ...     'tbreak 5',
+    ...     'continue',  # will stop at temporary breakpoint
+    ...     'break',     # make sure breakpoint is gone
+    ...     'continue',
+    ... ]):
+    ...    test_function()
+    > <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>(3)test_function()
+    -> print(1)
+    (Pdb) break 3
+    Breakpoint 1 at <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>:3
+    (Pdb) disable 1
+    Disabled breakpoint 1 at <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>:3
+    (Pdb) ignore 1 10
+    Will ignore next 10 crossings of breakpoint 1.
+    (Pdb) condition 1 1 < 2
+    New condition set for breakpoint 1.
+    (Pdb) break 4
+    Breakpoint 2 at <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>:4
+    (Pdb) break
+    Num Type         Disp Enb   Where
+    1   breakpoint   keep no    at <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>:3
+            stop only if 1 < 2
+            ignore next 10 hits
+    2   breakpoint   keep yes   at <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>:4
+    (Pdb) condition 1
+    Breakpoint 1 is now unconditional.
+    (Pdb) enable 1
+    Enabled breakpoint 1 at <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>:3
+    (Pdb) clear 1
+    Deleted breakpoint 1 at <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>:3
+    (Pdb) commands 2
+    (com) print 42
+    (com) end
+    (Pdb) continue
+    1
+    42
+    > <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>(4)test_function()
+    -> print(2)
+    (Pdb) clear
+    Clear all breaks? y
+    Deleted breakpoint 2 at <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>:4
+    (Pdb) tbreak 5
+    Breakpoint 3 at <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>:5
+    (Pdb) continue
+    2
+    Deleted breakpoint 3 at <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>:5
+    > <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>(5)test_function()
+    -> print(3)
+    (Pdb) break
+    (Pdb) continue
+    3
+    4
+    """
+
+
 def test_pdb_skip_modules():
     """This illustrates the simple case of module skipping.
 
@@ -167,84 +370,6 @@
     """
 
 
-def test_pdb_breakpoints():
-    """Test handling of breakpoints.
-
-    >>> def test_function():
-    ...     import pdb; pdb.Pdb().set_trace()
-    ...     print(1)
-    ...     print(2)
-    ...     print(3)
-    ...     print(4)
-
-    First, need to clear bdb state that might be left over from previous tests.
-    Otherwise, the new breakpoints might get assigned different numbers.
-
-    >>> from bdb import Breakpoint
-    >>> Breakpoint.next = 1
-    >>> Breakpoint.bplist = {}
-    >>> Breakpoint.bpbynumber = [None]
-
-    Now test the breakpoint commands.  NORMALIZE_WHITESPACE is needed because
-    the breakpoint list outputs a tab for the "stop only" and "ignore next"
-    lines, which we don't want to put in here.
-
-    >>> with PdbTestInput([  # doctest: +NORMALIZE_WHITESPACE
-    ...     'break 3',
-    ...     'disable 1',
-    ...     'ignore 1 10',
-    ...     'condition 1 1 < 2',
-    ...     'break 4',
-    ...     'break',
-    ...     'condition 1',
-    ...     'enable 1',
-    ...     'clear 1',
-    ...     'commands 2',
-    ...     'print 42',
-    ...     'end',
-    ...     'continue',  # will stop at breakpoint 2
-    ...     'continue',
-    ... ]):
-    ...    test_function()
-    > <doctest test.test_pdb.test_pdb_breakpoints[0]>(3)test_function()
-    -> print(1)
-    (Pdb) break 3
-    Breakpoint 1 at <doctest test.test_pdb.test_pdb_breakpoints[0]>:3
-    (Pdb) disable 1
-    Disabled breakpoint 1 at <doctest test.test_pdb.test_pdb_breakpoints[0]>:3
-    (Pdb) ignore 1 10
-    Will ignore next 10 crossings of breakpoint 1.
-    (Pdb) condition 1 1 < 2
-    New condition set for breakpoint 1.
-    (Pdb) break 4
-    Breakpoint 2 at <doctest test.test_pdb.test_pdb_breakpoints[0]>:4
-    (Pdb) break
-    Num Type         Disp Enb   Where
-    1   breakpoint   keep no    at <doctest test.test_pdb.test_pdb_breakpoints[0]>:3
-            stop only if 1 < 2
-            ignore next 10 hits
-    2   breakpoint   keep yes   at <doctest test.test_pdb.test_pdb_breakpoints[0]>:4
-    (Pdb) condition 1
-    Breakpoint 1 is now unconditional.
-    (Pdb) enable 1
-    Enabled breakpoint 1 at <doctest test.test_pdb.test_pdb_breakpoints[0]>:3
-    (Pdb) clear 1
-    Deleted breakpoint 1 at <doctest test.test_pdb.test_pdb_breakpoints[0]>:3
-    (Pdb) commands 2
-    (com) print 42
-    (com) end
-    (Pdb) continue
-    1
-    42
-    > <doctest test.test_pdb.test_pdb_breakpoints[0]>(4)test_function()
-    -> print(2)
-    (Pdb) continue
-    2
-    3
-    4
-    """
-
-
 def pdb_invoke(method, arg):
     """Run pdb.method(arg)."""
     import pdb; getattr(pdb, method)(arg)

From python-checkins at python.org  Fri Jul 30 19:04:28 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 19:04:28 +0200 (CEST)
Subject: [Python-checkins] r83287 - in python/branches/py3k:
	Doc/library/pdb.rst Lib/pdb.py Lib/test/test_pdb.py Misc/ACKS
	Misc/NEWS
Message-ID: <20100730170428.54933EE9AC@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 19:04:28 2010
New Revision: 83287

Log:
Add "longlist" and "source" commands, ideas borrowed from pdb++ by Antonio Cuni.

Modified:
   python/branches/py3k/Doc/library/pdb.rst
   python/branches/py3k/Lib/pdb.py
   python/branches/py3k/Lib/test/test_pdb.py
   python/branches/py3k/Misc/ACKS
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Doc/library/pdb.rst
==============================================================================
--- python/branches/py3k/Doc/library/pdb.rst	(original)
+++ python/branches/py3k/Doc/library/pdb.rst	Fri Jul 30 19:04:28 2010
@@ -368,6 +368,12 @@
    list 11 lines around at that line.  With two arguments, list the given range;
    if the second argument is less than the first, it is interpreted as a count.
 
+.. pdbcommand:: ll | longlist
+
+   List all source code for the current function or frame.
+
+   .. versionadded:: 3.2
+
 .. pdbcommand:: a(rgs)
 
    Print the argument list of the current function.
@@ -385,6 +391,12 @@
 
    Print the type of the *expression*.
 
+.. pdbcommand:: source expression
+
+   Try to get source code for the given object and display it.
+
+   .. versionadded:: 3.2
+
 .. _debugger-aliases:
 
 .. pdbcommand:: alias [name [command]]

Modified: python/branches/py3k/Lib/pdb.py
==============================================================================
--- python/branches/py3k/Lib/pdb.py	(original)
+++ python/branches/py3k/Lib/pdb.py	Fri Jul 30 19:04:28 2010
@@ -74,6 +74,8 @@
 import re
 import pprint
 import traceback
+import inspect
+import types
 
 
 class Restart(Exception):
@@ -1028,25 +1030,62 @@
         filename = self.curframe.f_code.co_filename
         breaklist = self.get_file_breaks(filename)
         try:
-            for lineno in range(first, last+1):
-                line = linecache.getline(filename, lineno,
-                                         self.curframe.f_globals)
-                if not line:
-                    self.message('[EOF]')
-                    break
-                else:
-                    s = repr(lineno).rjust(3)
-                    if len(s) < 4: s = s + ' '
-                    if lineno in breaklist: s = s + 'B'
-                    else: s = s + ' '
-                    if lineno == self.curframe.f_lineno:
-                        s = s + '->'
-                    self.message(s + '\t' + line.rstrip())
-                    self.lineno = lineno
+            # XXX add tb_lineno feature
+            lines = linecache.getlines(filename, self.curframe.f_globals)
+            self._print_lines(lines[first-1:last], first, breaklist,
+                              self.curframe.f_lineno, -1)
+            self.lineno = min(last, len(lines))
+            if len(lines) < last:
+                self.message('[EOF]')
         except KeyboardInterrupt:
             pass
     do_l = do_list
 
+    def do_longlist(self, arg):
+        """longlist | ll
+        List the whole source code for the current function or frame.
+        """
+        filename = self.curframe.f_code.co_filename
+        breaklist = self.get_file_breaks(filename)
+        try:
+            lines, lineno = inspect.getsourcelines(self.curframe)
+        except IOError as err:
+            self.error(err)
+            return
+        self._print_lines(lines, lineno, breaklist, self.curframe.f_lineno, -1)
+    do_ll = do_longlist
+
+    def do_source(self, arg):
+        """source expression
+        Try to get source code for the given object and display it.
+        """
+        try:
+            obj = self._getval(arg)
+        except:
+            return
+        try:
+            lines, lineno = inspect.getsourcelines(obj)
+        except (IOError, TypeError) as err:
+            self.error(err)
+            return
+        self._print_lines(lines, lineno, [], -1, -1)
+
+    def _print_lines(self, lines, start, breaks, current, special):
+        """Print a range of lines."""
+        for lineno, line in enumerate(lines, start):
+            s = str(lineno).rjust(3)
+            if len(s) < 4:
+                s += ' '
+            if lineno in breaks:
+                s += 'B'
+            else:
+                s += ' '
+            if lineno == current:
+                s += '->'
+            elif lineno == special:
+                s += '>>'
+            self.message(s + '\t' + line.rstrip())
+
     def do_whatis(self, arg):
         """whatis arg
         Print the type of the argument.
@@ -1249,10 +1288,12 @@
 _help_order = [
     'help', 'where', 'down', 'up', 'break', 'tbreak', 'clear', 'disable',
     'enable', 'ignore', 'condition', 'commands', 'step', 'next', 'until',
-    'jump', 'return', 'retval', 'run', 'continue', 'list', 'args', 'print',
-    'whatis', 'alias', 'unalias', 'quit',
+    'jump', 'return', 'retval', 'run', 'continue', 'list', 'longlist',
+    'args', 'print', 'pp', 'whatis', 'source', 'alias', 'unalias',
+    'debug', 'quit',
 ]
 
+docs = set()
 for _command in _help_order:
     __doc__ += getattr(Pdb, 'do_' + _command).__doc__.strip() + '\n\n'
 __doc__ += Pdb.help_exec.__doc__

Modified: python/branches/py3k/Lib/test/test_pdb.py
==============================================================================
--- python/branches/py3k/Lib/test/test_pdb.py	(original)
+++ python/branches/py3k/Lib/test/test_pdb.py	Fri Jul 30 19:04:28 2010
@@ -257,6 +257,105 @@
     """
 
 
+def do_nothing():
+    pass
+
+def do_something():
+    print(42)
+
+def test_list_commands():
+    """Test the list and source commands of pdb.
+
+    >>> def test_function_2(foo):
+    ...     import test_pdb
+    ...     test_pdb.do_nothing()
+    ...     'some...'
+    ...     'more...'
+    ...     'code...'
+    ...     'to...'
+    ...     'make...'
+    ...     'a...'
+    ...     'long...'
+    ...     'listing...'
+    ...     'useful...'
+    ...     '...'
+    ...     '...'
+    ...     return foo
+
+    >>> def test_function():
+    ...     import pdb; pdb.Pdb().set_trace()
+    ...     ret = test_function_2('baz')
+
+    >>> with PdbTestInput([  # doctest: +ELLIPSIS, +NORMALIZE_WHITESPACE
+    ...     'list',      # list first function
+    ...     'step',      # step into second function
+    ...     'list',      # list second function
+    ...     'list',      # continue listing to EOF
+    ...     'list 1,3',  # list specific lines
+    ...     'list x',    # invalid argument
+    ...     'next',      # step to import
+    ...     'next',      # step over import
+    ...     'step',      # step into do_nothing
+    ...     'longlist',  # list all lines
+    ...     'source do_something',  # list all lines of function
+    ...     'continue',
+    ... ]):
+    ...    test_function()
+    > <doctest test.test_pdb.test_list_commands[1]>(3)test_function()
+    -> ret = test_function_2('baz')
+    (Pdb) list
+      1         def test_function():
+      2             import pdb; pdb.Pdb().set_trace()
+      3  ->         ret = test_function_2('baz')
+    [EOF]
+    (Pdb) step
+    --Call--
+    > <doctest test.test_pdb.test_list_commands[0]>(1)test_function_2()
+    -> def test_function_2(foo):
+    (Pdb) list
+      1  ->     def test_function_2(foo):
+      2             import test_pdb
+      3             test_pdb.do_nothing()
+      4             'some...'
+      5             'more...'
+      6             'code...'
+      7             'to...'
+      8             'make...'
+      9             'a...'
+     10             'long...'
+     11             'listing...'
+    (Pdb) list
+     12             'useful...'
+     13             '...'
+     14             '...'
+     15             return foo
+    [EOF]
+    (Pdb) list 1,3
+      1  ->     def test_function_2(foo):
+      2             import test_pdb
+      3             test_pdb.do_nothing()
+    (Pdb) list x
+    *** ...
+    (Pdb) next
+    > <doctest test.test_pdb.test_list_commands[0]>(2)test_function_2()
+    -> import test_pdb
+    (Pdb) next
+    > <doctest test.test_pdb.test_list_commands[0]>(3)test_function_2()
+    -> test_pdb.do_nothing()
+    (Pdb) step
+    --Call--
+    > /home/gbr/devel/python/Lib/test/test_pdb.py(260)do_nothing()
+    -> def do_nothing():
+    (Pdb) longlist
+    ...  ->     def do_nothing():
+    ...             pass
+    (Pdb) source do_something
+    ...         def do_something():
+    ...             print(42)
+    (Pdb) continue
+    """
+
+
 def test_pdb_skip_modules():
     """This illustrates the simple case of module skipping.
 

Modified: python/branches/py3k/Misc/ACKS
==============================================================================
--- python/branches/py3k/Misc/ACKS	(original)
+++ python/branches/py3k/Misc/ACKS	Fri Jul 30 19:04:28 2010
@@ -180,6 +180,7 @@
 Joaquin Cuenca Abela
 John Cugini
 Tom Culliton
+Antonio Cuni
 Brian Curtin
 Lisandro Dalcin
 Andrew Dalke

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Fri Jul 30 19:04:28 2010
@@ -475,6 +475,15 @@
 Library
 -------
 
+- The pdb command "source" has been added.  It displays the source
+  code for a given object, if possible.
+
+- The pdb command "longlist" has been added.  It displays the whole
+  source code for the current function.
+
+- Issue #1503502: Make pdb.Pdb easier to subclass by putting message
+  and error output into methods.
+
 - Issue #809887: Make the output of pdb's breakpoint deletions more
   consistent; emit a message when a breakpoint is enabled or disabled.
 

From python-checkins at python.org  Fri Jul 30 19:29:39 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Fri, 30 Jul 2010 19:29:39 +0200 (CEST)
Subject: [Python-checkins] r83288 - python/branches/py3k/PCbuild/build_ssl.py
Message-ID: <20100730172939.CA489EEA48@mail.python.org>

Author: martin.v.loewis
Date: Fri Jul 30 19:29:39 2010
New Revision: 83288

Log:
Drop 2.4 support. Add disabled algorithms to Configure options.

Modified:
   python/branches/py3k/PCbuild/build_ssl.py

Modified: python/branches/py3k/PCbuild/build_ssl.py
==============================================================================
--- python/branches/py3k/PCbuild/build_ssl.py	(original)
+++ python/branches/py3k/PCbuild/build_ssl.py	Fri Jul 30 19:29:39 2010
@@ -23,6 +23,7 @@
 # python.exe build_ssl.py Release x64
 # python.exe build_ssl.py Release Win32
 
+from __future__ import with_statement
 import os, sys, re, shutil
 
 # Find all "foo.exe" files on the PATH.
@@ -121,13 +122,10 @@
     """
     if not os.path.isfile(makefile):
         return
-    # 2.4 compatibility
     fin = open(makefile)
-    if 1: # with open(makefile) as fin:
+    with open(makefile) as fin:
         lines = fin.readlines()
-        fin.close()
-    fout = open(makefile, 'w')
-    if 1: # with open(makefile, 'w') as fout:
+    with open(makefile, 'w') as fout:
         for line in lines:
             if line.startswith("PERL="):
                 continue
@@ -143,11 +141,10 @@
                         line = line + noalgo
                 line = line + '\n'
             fout.write(line)
-    fout.close()
 
 def run_configure(configure, do_script):
-    print("perl Configure "+configure)
-    os.system("perl Configure "+configure)
+    print("perl Configure "+configure+" no-idea no-mdc2")
+    os.system("perl Configure "+configure+" no-idea no-mdc2")
     print(do_script)
     os.system(do_script)
 

From python-checkins at python.org  Fri Jul 30 19:30:52 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Fri, 30 Jul 2010 19:30:52 +0200 (CEST)
Subject: [Python-checkins] r83289 - in python/branches/py3k: Misc/NEWS
	PCbuild/pyproject.vsprops PCbuild/readme.txt
Message-ID: <20100730173052.42AE6EE9E7@mail.python.org>

Author: martin.v.loewis
Date: Fri Jul 30 19:30:51 2010
New Revision: 83289

Log:
Use OpenSSL 1.0.0a on Windows.


Modified:
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/PCbuild/pyproject.vsprops
   python/branches/py3k/PCbuild/readme.txt

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Fri Jul 30 19:30:51 2010
@@ -1672,6 +1672,8 @@
 Build
 -----
 
+- Use OpenSSL 1.0.0a on Windows.
+
 - Issue #9280: Make sharedinstall depend on sharedmods.
 
 - Issue #9189: Make a user-specified CFLAGS, CPPFLAGS, or LDFLAGS

Modified: python/branches/py3k/PCbuild/pyproject.vsprops
==============================================================================
--- python/branches/py3k/PCbuild/pyproject.vsprops	(original)
+++ python/branches/py3k/PCbuild/pyproject.vsprops	Fri Jul 30 19:30:51 2010
@@ -58,7 +58,7 @@
 	/>
 	<UserMacro
 		Name="opensslDir"
-		Value="$(externalsDir)\openssl-0.9.8l"
+		Value="$(externalsDir)\openssl-1.0.0a"
 	/>
 	<UserMacro
 		Name="tcltkDir"

Modified: python/branches/py3k/PCbuild/readme.txt
==============================================================================
--- python/branches/py3k/PCbuild/readme.txt	(original)
+++ python/branches/py3k/PCbuild/readme.txt	Fri Jul 30 19:30:51 2010
@@ -139,7 +139,7 @@
 
     Get the source code through
 
-    svn export http://svn.python.org/projects/external/openssl-0.9.8l
+    svn export http://svn.python.org/projects/external/openssl-1.0.0a
 
     ** NOTE: if you use the Tools\buildbot\external(-amd64).bat approach for
     obtaining external sources then you don't need to manually get the source

From python-checkins at python.org  Fri Jul 30 19:48:29 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 19:48:29 +0200 (CEST)
Subject: [Python-checkins] r83290 -
	python/branches/import_unicode/Include/Python.h
Message-ID: <20100730174829.04544EE9C4@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 19:48:28 2010
New Revision: 83290

Log:
typo: lives => live

Modified:
   python/branches/import_unicode/Include/Python.h

Modified: python/branches/import_unicode/Include/Python.h
==============================================================================
--- python/branches/import_unicode/Include/Python.h	(original)
+++ python/branches/import_unicode/Include/Python.h	Fri Jul 30 19:48:28 2010
@@ -128,7 +128,7 @@
 /* _Py_Mangle is defined in compile.c */
 PyAPI_FUNC(PyObject*) _Py_Mangle(PyObject *p, PyObject *name);
 
-/* _Py_char2wchar and _Py_wchar2char lives in main.c */
+/* _Py_char2wchar and _Py_wchar2char live in main.c */
 PyAPI_FUNC(wchar_t *) _Py_char2wchar(char *);
 PyAPI_FUNC(char*) _Py_wchar2char(const wchar_t *text);
 

From python-checkins at python.org  Fri Jul 30 20:08:12 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 20:08:12 +0200 (CEST)
Subject: [Python-checkins] r83291 - python/branches/py3k/Lib/pdb.py
Message-ID: <20100730180812.89C1EEEA35@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 20:08:12 2010
New Revision: 83291

Log:
Fix source finding if the given frame is a module-level frame.

Modified:
   python/branches/py3k/Lib/pdb.py

Modified: python/branches/py3k/Lib/pdb.py
==============================================================================
--- python/branches/py3k/Lib/pdb.py	(original)
+++ python/branches/py3k/Lib/pdb.py	Fri Jul 30 20:08:12 2010
@@ -105,6 +105,15 @@
     fp.close()
     return answer
 
+def getsourcelines(obj):
+    lines, lineno = inspect.findsource(obj)
+    if inspect.isframe(obj) and lineno == 0 and \
+           obj.f_globals is obj.f_locals:
+        # must be a module frame: do not try to cut a block out of it
+        return lines, 0
+    elif inspect.ismodule(obj):
+        return lines, 0
+    return inspect.getblock(lines[lineno:]), lineno+1
 
 # Interaction prompt line will separate file and call info from code
 # text using value of line_prefix string.  A newline and arrow may
@@ -1048,7 +1057,7 @@
         filename = self.curframe.f_code.co_filename
         breaklist = self.get_file_breaks(filename)
         try:
-            lines, lineno = inspect.getsourcelines(self.curframe)
+            lines, lineno = getsourcelines(self.curframe)
         except IOError as err:
             self.error(err)
             return
@@ -1064,7 +1073,7 @@
         except:
             return
         try:
-            lines, lineno = inspect.getsourcelines(obj)
+            lines, lineno = getsourcelines(obj)
         except (IOError, TypeError) as err:
             self.error(err)
             return

From python-checkins at python.org  Fri Jul 30 20:15:16 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 20:15:16 +0200 (CEST)
Subject: [Python-checkins] r83292 - python/branches/py3k/Lib/test/test_pdb.py
Message-ID: <20100730181516.8898BEEA30@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 20:15:16 2010
New Revision: 83292

Log:
Test that "source" with nonexisting things works as expected.

Modified:
   python/branches/py3k/Lib/test/test_pdb.py

Modified: python/branches/py3k/Lib/test/test_pdb.py
==============================================================================
--- python/branches/py3k/Lib/test/test_pdb.py	(original)
+++ python/branches/py3k/Lib/test/test_pdb.py	Fri Jul 30 20:15:16 2010
@@ -298,6 +298,7 @@
     ...     'step',      # step into do_nothing
     ...     'longlist',  # list all lines
     ...     'source do_something',  # list all lines of function
+    ...     'source fooxxx',        # something that doesn't exit
     ...     'continue',
     ... ]):
     ...    test_function()
@@ -352,6 +353,8 @@
     (Pdb) source do_something
     ...         def do_something():
     ...             print(42)
+    (Pdb) source fooxxx
+    *** ...
     (Pdb) continue
     """
 

From python-checkins at python.org  Fri Jul 30 20:46:38 2010
From: python-checkins at python.org (georg.brandl)
Date: Fri, 30 Jul 2010 20:46:38 +0200 (CEST)
Subject: [Python-checkins] r83293 - in python/branches/py3k:
	Doc/library/pdb.rst Lib/pdb.py Lib/test/test_pdb.py Misc/NEWS
Message-ID: <20100730184638.8DFB4EEA1E@mail.python.org>

Author: georg.brandl
Date: Fri Jul 30 20:46:38 2010
New Revision: 83293

Log:
Show the traceback line numbers as well as the current line numbers if an exception is being debugged.  Courtesy of pdb++ by Antonio Cuni.  Also document -> and >> markers for "list".

Modified:
   python/branches/py3k/Doc/library/pdb.rst
   python/branches/py3k/Lib/pdb.py
   python/branches/py3k/Lib/test/test_pdb.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Doc/library/pdb.rst
==============================================================================
--- python/branches/py3k/Doc/library/pdb.rst	(original)
+++ python/branches/py3k/Doc/library/pdb.rst	Fri Jul 30 20:46:38 2010
@@ -368,9 +368,18 @@
    list 11 lines around at that line.  With two arguments, list the given range;
    if the second argument is less than the first, it is interpreted as a count.
 
+   The current line in the current frame is indicated by ``->``.  If an
+   exception is being debugged, the line where the exception was originally
+   raised or propagated is indicated by ``>>``, if it differs from the current
+   line.
+
+   .. versionadded:: 3.2
+      The ``>>`` marker.
+
 .. pdbcommand:: ll | longlist
 
-   List all source code for the current function or frame.
+   List all source code for the current function or frame.  Interesting lines
+   are marked as for :pdbcmd:`list`.
 
    .. versionadded:: 3.2
 

Modified: python/branches/py3k/Lib/pdb.py
==============================================================================
--- python/branches/py3k/Lib/pdb.py	(original)
+++ python/branches/py3k/Lib/pdb.py	Fri Jul 30 20:46:38 2010
@@ -70,8 +70,10 @@
 import linecache
 import cmd
 import bdb
+import dis
 import os
 import re
+import code
 import pprint
 import traceback
 import inspect
@@ -107,14 +109,22 @@
 
 def getsourcelines(obj):
     lines, lineno = inspect.findsource(obj)
-    if inspect.isframe(obj) and lineno == 0 and \
-           obj.f_globals is obj.f_locals:
+    if inspect.isframe(obj) and obj.f_globals is obj.f_locals:
         # must be a module frame: do not try to cut a block out of it
-        return lines, 0
+        return lines, 1
     elif inspect.ismodule(obj):
-        return lines, 0
+        return lines, 1
     return inspect.getblock(lines[lineno:]), lineno+1
 
+def lasti2lineno(code, lasti):
+    linestarts = list(dis.findlinestarts(code))
+    linestarts.reverse()
+    for i, lineno in linestarts:
+        if lasti >= i:
+            return lineno
+    return 0
+
+
 # Interaction prompt line will separate file and call info from code
 # text using value of line_prefix string.  A newline and arrow may
 # be to your liking.  You can set it once pdb is imported using the
@@ -133,6 +143,7 @@
         self.aliases = {}
         self.mainpyfile = ''
         self._wait_for_mainpyfile = 0
+        self.tb_lineno = {}
         # Try to load readline if it exists
         try:
             import readline
@@ -179,10 +190,18 @@
         self.stack = []
         self.curindex = 0
         self.curframe = None
+        self.tb_lineno.clear()
 
-    def setup(self, f, t):
+    def setup(self, f, tb):
         self.forget()
-        self.stack, self.curindex = self.get_stack(f, t)
+        self.stack, self.curindex = self.get_stack(f, tb)
+        while tb:
+            # when setting up post-mortem debugging with a traceback, save all
+            # the original line numbers to be displayed along the current line
+            # numbers (which can be different, e.g. due to finally clauses)
+            lineno = lasti2lineno(tb.tb_frame.f_code, tb.tb_lasti)
+            self.tb_lineno[tb.tb_frame] = lineno
+            tb = tb.tb_next
         self.curframe = self.stack[self.curindex][0]
         # The f_locals dictionary is updated from the actual frame
         # locals whenever the .f_locals accessor is called, so we
@@ -1005,13 +1024,18 @@
 
     def do_list(self, arg):
         """l(ist) [first [,last] | .]
-        List source code for the current file.
-        Without arguments, list 11 lines around the current line
-        or continue the previous listing.
-        With . as argument, list 11 lines around the current line.
-        With one argument, list 11 lines starting at that line.
-        With two arguments, list the given range;
-        if the second argument is less than the first, it is a count.
+
+        List source code for the current file.  Without arguments,
+        list 11 lines around the current line or continue the previous
+        listing.  With . as argument, list 11 lines around the current
+        line.  With one argument, list 11 lines starting at that line.
+        With two arguments, list the given range; if the second
+        argument is less than the first, it is a count.
+
+        The current line in the current frame is indicated by "->".
+        If an exception is being debugged, the line where the
+        exception was originally raised or propagated is indicated by
+        ">>", if it differs from the current line.
         """
         self.lastcmd = 'list'
         last = None
@@ -1039,10 +1063,9 @@
         filename = self.curframe.f_code.co_filename
         breaklist = self.get_file_breaks(filename)
         try:
-            # XXX add tb_lineno feature
             lines = linecache.getlines(filename, self.curframe.f_globals)
             self._print_lines(lines[first-1:last], first, breaklist,
-                              self.curframe.f_lineno, -1)
+                              self.curframe)
             self.lineno = min(last, len(lines))
             if len(lines) < last:
                 self.message('[EOF]')
@@ -1061,7 +1084,7 @@
         except IOError as err:
             self.error(err)
             return
-        self._print_lines(lines, lineno, breaklist, self.curframe.f_lineno, -1)
+        self._print_lines(lines, lineno, breaklist, self.curframe)
     do_ll = do_longlist
 
     def do_source(self, arg):
@@ -1077,10 +1100,15 @@
         except (IOError, TypeError) as err:
             self.error(err)
             return
-        self._print_lines(lines, lineno, [], -1, -1)
+        self._print_lines(lines, lineno)
 
-    def _print_lines(self, lines, start, breaks, current, special):
+    def _print_lines(self, lines, start, breaks=(), frame=None):
         """Print a range of lines."""
+        if frame:
+            current_lineno = frame.f_lineno
+            exc_lineno = self.tb_lineno.get(frame, -1)
+        else:
+            current_lineno = exc_lineno = -1
         for lineno, line in enumerate(lines, start):
             s = str(lineno).rjust(3)
             if len(s) < 4:
@@ -1089,9 +1117,9 @@
                 s += 'B'
             else:
                 s += ' '
-            if lineno == current:
+            if lineno == current_lineno:
                 s += '->'
-            elif lineno == special:
+            elif lineno == exc_lineno:
                 s += '>>'
             self.message(s + '\t' + line.rstrip())
 

Modified: python/branches/py3k/Lib/test/test_pdb.py
==============================================================================
--- python/branches/py3k/Lib/test/test_pdb.py	(original)
+++ python/branches/py3k/Lib/test/test_pdb.py	Fri Jul 30 20:46:38 2010
@@ -359,6 +359,68 @@
     """
 
 
+def test_post_mortem():
+    """Test post mortem traceback debugging.
+
+    >>> def test_function_2():
+    ...     try:
+    ...         1/0
+    ...     finally:
+    ...         print('Exception!')
+
+    >>> def test_function():
+    ...     import pdb; pdb.Pdb().set_trace()
+    ...     test_function_2()
+    ...     print('Not reached.')
+
+    >>> with PdbTestInput([  # doctest: +ELLIPSIS, +NORMALIZE_WHITESPACE
+    ...     'next',      # step over exception-raising call
+    ...     'bt',        # get a backtrace
+    ...     'list',      # list code of test_function()
+    ...     'down',      # step into test_function_2()
+    ...     'list',      # list code of test_function_2()
+    ...     'continue',
+    ... ]):
+    ...    try:
+    ...        test_function()
+    ...    except ZeroDivisionError:
+    ...        print('Correctly reraised.')
+    > <doctest test.test_pdb.test_post_mortem[1]>(3)test_function()
+    -> test_function_2()
+    (Pdb) next
+    Exception!
+    ZeroDivisionError: division by zero
+    > <doctest test.test_pdb.test_post_mortem[1]>(3)test_function()
+    -> test_function_2()
+    (Pdb) bt
+    ...
+      <doctest test.test_pdb.test_post_mortem[2]>(10)<module>()
+    -> test_function()
+    > <doctest test.test_pdb.test_post_mortem[1]>(3)test_function()
+    -> test_function_2()
+      <doctest test.test_pdb.test_post_mortem[0]>(3)test_function_2()
+    -> 1/0
+    (Pdb) list
+      1         def test_function():
+      2             import pdb; pdb.Pdb().set_trace()
+      3  ->         test_function_2()
+      4             print('Not reached.')
+    [EOF]
+    (Pdb) down
+    > <doctest test.test_pdb.test_post_mortem[0]>(3)test_function_2()
+    -> 1/0
+    (Pdb) list
+      1         def test_function_2():
+      2             try:
+      3  >>             1/0
+      4             finally:
+      5  ->             print('Exception!')
+    [EOF]
+    (Pdb) continue
+    Correctly reraised.
+    """
+
+
 def test_pdb_skip_modules():
     """This illustrates the simple case of module skipping.
 

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Fri Jul 30 20:46:38 2010
@@ -475,6 +475,10 @@
 Library
 -------
 
+- For traceback debugging, the pdb listing now also shows the locations
+  where the exception was originally (re)raised, if it differs from the
+  last line executed (e.g. in case of finally clauses).
+
 - The pdb command "source" has been added.  It displays the source
   code for a given object, if possible.
 

From python-checkins at python.org  Fri Jul 30 21:34:37 2010
From: python-checkins at python.org (senthil.kumaran)
Date: Fri, 30 Jul 2010 21:34:37 +0200 (CEST)
Subject: [Python-checkins] r83294 - in python/branches/py3k/Lib:
	test/test_urllib.py urllib/parse.py
Message-ID: <20100730193437.07F72EE9D2@mail.python.org>

Author: senthil.kumaran
Date: Fri Jul 30 21:34:36 2010
New Revision: 83294

Log:
Fix issue9301 - handle unquote({}) kind of case.


Modified:
   python/branches/py3k/Lib/test/test_urllib.py
   python/branches/py3k/Lib/urllib/parse.py

Modified: python/branches/py3k/Lib/test/test_urllib.py
==============================================================================
--- python/branches/py3k/Lib/test/test_urllib.py	(original)
+++ python/branches/py3k/Lib/test/test_urllib.py	Fri Jul 30 21:34:36 2010
@@ -527,6 +527,7 @@
         self.assertEqual(expect, result,
                          "using quote_plus(): %r != %r" % (expect, result))
 
+
 class UnquotingTests(unittest.TestCase):
     """Tests for unquote() and unquote_plus()
 
@@ -555,6 +556,7 @@
                          "using unquote(): not all characters escaped: "
                          "%s" % result)
         self.assertRaises(TypeError, urllib.parse.unquote, None)
+        self.assertRaises(TypeError, urllib.parse.unquote, ())
 
     def test_unquoting_badpercent(self):
         # Test unquoting on bad percent-escapes
@@ -589,8 +591,8 @@
         result = urllib.parse.unquote_to_bytes(given)
         self.assertEqual(expect, result, "using unquote_to_bytes(): %r != %r"
                          % (expect, result))
-
         self.assertRaises(TypeError, urllib.parse.unquote_to_bytes, None)
+        self.assertRaises(TypeError, urllib.parse.unquote_to_bytes, ())
 
     def test_unquoting_mixed_case(self):
         # Test unquoting on mixed-case hex digits in the percent-escapes

Modified: python/branches/py3k/Lib/urllib/parse.py
==============================================================================
--- python/branches/py3k/Lib/urllib/parse.py	(original)
+++ python/branches/py3k/Lib/urllib/parse.py	Fri Jul 30 21:34:36 2010
@@ -313,9 +313,7 @@
     """unquote_to_bytes('abc%20def') -> b'abc def'."""
     # Note: strings are encoded as UTF-8. This is only an issue if it contains
     # unescaped non-ASCII characters, which URIs should not.
-    if not string:
-        if string is None:
-            raise TypeError('None object is invalid for unquote_to_bytes()')
+    if string in (b'', ''):
         return b''
     if isinstance(string, str):
         string = string.encode('utf-8')
@@ -340,9 +338,7 @@
 
     unquote('abc%20def') -> 'abc def'.
     """
-    if not string:
-        if string is None:
-            raise TypeError('None object is invalid for unquote() function.')
+    if string in (b'', ''):
         return string
     res = string.split('%')
     if len(res) == 1:

From python-checkins at python.org  Fri Jul 30 21:43:11 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 21:43:11 +0200 (CEST)
Subject: [Python-checkins] r83295 -
	python/branches/import_unicode/Lib/test/test_subprocess.py
Message-ID: <20100730194311.E9D8BEEA32@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 21:43:11 2010
New Revision: 83295

Log:
test_subprocess: don't skip test_bytes_program()

The test now pass in any case.

Merge also with py3k.

Modified:
   python/branches/import_unicode/Lib/test/test_subprocess.py

Modified: python/branches/import_unicode/Lib/test/test_subprocess.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_subprocess.py	(original)
+++ python/branches/import_unicode/Lib/test/test_subprocess.py	Fri Jul 30 21:43:11 2010
@@ -854,7 +854,7 @@
                 [sys.executable, "-c", script],
                 env=env)
             stdout = stdout.rstrip(b'\n\r')
-            self.assertEquals(stdout.decode('ascii'), repr(value))
+            self.assertEqual(stdout.decode('ascii'), repr(value))
 
             # test bytes
             key = key.encode("ascii", "surrogateescape")
@@ -866,28 +866,28 @@
                 [sys.executable, "-c", script],
                 env=env)
             stdout = stdout.rstrip(b'\n\r')
-            self.assertEquals(stdout.decode('ascii'), repr(value))
+            self.assertEqual(stdout.decode('ascii'), repr(value))
 
-    def test_absolute_bytes_program(self):
-        exitcode = subprocess.call([os.fsencode(sys.executable), "-c", "pass"])
-        self.assertEquals(exitcode, 0)
-
-    @unittest.skipIf(sysconfig.is_python_build(), "need an installed Python")
     def test_bytes_program(self):
+        abs_program = os.fsencode(sys.executable)
         path, program = os.path.split(sys.executable)
         program = os.fsencode(program)
 
+        # absolute bytes path
+        exitcode = subprocess.call([abs_program, "-c", "pass"])
+        self.assertEqual(exitcode, 0)
+
         # bytes program, unicode PATH
         env = os.environ.copy()
         env["PATH"] = path
         exitcode = subprocess.call([program, "-c", "pass"], env=env)
-        self.assertEquals(exitcode, 0)
+        self.assertEqual(exitcode, 0)
 
         # bytes program, bytes PATH
         envb = os.environb.copy()
         envb[b"PATH"] = os.fsencode(path)
         exitcode = subprocess.call([program, "-c", "pass"], env=envb)
-        self.assertEquals(exitcode, 0)
+        self.assertEqual(exitcode, 0)
 
 
 @unittest.skipUnless(mswindows, "Windows specific tests")

From python-checkins at python.org  Fri Jul 30 22:03:18 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Fri, 30 Jul 2010 22:03:18 +0200 (CEST)
Subject: [Python-checkins] r83296 - in python/branches/py3k/Modules/zlib:
	ChangeLog FAQ INDEX Makefile Makefile.in README adler32.c
	compress.c configure crc32.c deflate.c deflate.h example.c
	infback.c inffast.c inffast.h inflate.c inflate.h inftrees.c
	inftrees.h make_vms.com minigzip.c trees.c trees.h uncompr.c
	zconf.h zlib.3 zlib.h zutil.c zutil.h
Message-ID: <20100730200318.6DEDDEE9EB@mail.python.org>

Author: martin.v.loewis
Date: Fri Jul 30 22:03:17 2010
New Revision: 83296

Log:
Import files from zlib 1.0.5.


Modified:
   python/branches/py3k/Modules/zlib/ChangeLog
   python/branches/py3k/Modules/zlib/FAQ
   python/branches/py3k/Modules/zlib/INDEX
   python/branches/py3k/Modules/zlib/Makefile
   python/branches/py3k/Modules/zlib/Makefile.in
   python/branches/py3k/Modules/zlib/README
   python/branches/py3k/Modules/zlib/adler32.c
   python/branches/py3k/Modules/zlib/compress.c
   python/branches/py3k/Modules/zlib/configure
   python/branches/py3k/Modules/zlib/crc32.c
   python/branches/py3k/Modules/zlib/deflate.c
   python/branches/py3k/Modules/zlib/deflate.h
   python/branches/py3k/Modules/zlib/example.c
   python/branches/py3k/Modules/zlib/infback.c
   python/branches/py3k/Modules/zlib/inffast.c
   python/branches/py3k/Modules/zlib/inffast.h
   python/branches/py3k/Modules/zlib/inflate.c
   python/branches/py3k/Modules/zlib/inflate.h
   python/branches/py3k/Modules/zlib/inftrees.c
   python/branches/py3k/Modules/zlib/inftrees.h
   python/branches/py3k/Modules/zlib/make_vms.com
   python/branches/py3k/Modules/zlib/minigzip.c
   python/branches/py3k/Modules/zlib/trees.c
   python/branches/py3k/Modules/zlib/trees.h
   python/branches/py3k/Modules/zlib/uncompr.c
   python/branches/py3k/Modules/zlib/zconf.h
   python/branches/py3k/Modules/zlib/zlib.3
   python/branches/py3k/Modules/zlib/zlib.h
   python/branches/py3k/Modules/zlib/zutil.c
   python/branches/py3k/Modules/zlib/zutil.h

Modified: python/branches/py3k/Modules/zlib/ChangeLog
==============================================================================
--- python/branches/py3k/Modules/zlib/ChangeLog	(original)
+++ python/branches/py3k/Modules/zlib/ChangeLog	Fri Jul 30 22:03:17 2010
@@ -1,6 +1,359 @@
 
                 ChangeLog file for zlib
 
+Changes in 1.2.5 (19 Apr 2010)
+- Disable visibility attribute in win32/Makefile.gcc [Bar-Lev]
+- Default to libdir as sharedlibdir in configure [Nieder]
+- Update copyright dates on modified source files
+- Update trees.c to be able to generate modified trees.h
+- Exit configure for MinGW, suggesting win32/Makefile.gcc
+
+Changes in 1.2.4.5 (18 Apr 2010)
+- Set sharedlibdir in configure [Torok]
+- Set LDFLAGS in Makefile.in [Bar-Lev]
+- Avoid mkdir objs race condition in Makefile.in [Bowler]
+- Add ZLIB_INTERNAL in front of internal inter-module functions and arrays
+- Define ZLIB_INTERNAL to hide internal functions and arrays for GNU C
+- Don't use hidden attribute when it is a warning generator (e.g. Solaris)
+
+Changes in 1.2.4.4 (18 Apr 2010)
+- Fix CROSS_PREFIX executable testing, CHOST extract, mingw* [Torok]
+- Undefine _LARGEFILE64_SOURCE in zconf.h if it is zero, but not if empty
+- Try to use bash or ksh regardless of functionality of /bin/sh
+- Fix configure incompatibility with NetBSD sh
+- Remove attempt to run under bash or ksh since have better NetBSD fix
+- Fix win32/Makefile.gcc for MinGW [Bar-Lev]
+- Add diagnostic messages when using CROSS_PREFIX in configure
+- Added --sharedlibdir option to configure [Weigelt]
+- Use hidden visibility attribute when available [Frysinger]
+
+Changes in 1.2.4.3 (10 Apr 2010)
+- Only use CROSS_PREFIX in configure for ar and ranlib if they exist
+- Use CROSS_PREFIX for nm [Bar-Lev]
+- Assume _LARGEFILE64_SOURCE defined is equivalent to true
+- Avoid use of undefined symbols in #if with && and ||
+- Make *64 prototypes in gzguts.h consistent with functions
+- Add -shared load option for MinGW in configure [Bowler]
+- Move z_off64_t to public interface, use instead of off64_t
+- Remove ! from shell test in configure (not portable to Solaris)
+- Change +0 macro tests to -0 for possibly increased portability
+
+Changes in 1.2.4.2 (9 Apr 2010)
+- Add consistent carriage returns to readme.txt's in masmx86 and masmx64
+- Really provide prototypes for *64 functions when building without LFS
+- Only define unlink() in minigzip.c if unistd.h not included
+- Update README to point to contrib/vstudio project files
+- Move projects/vc6 to old/ and remove projects/
+- Include stdlib.h in minigzip.c for setmode() definition under WinCE
+- Clean up assembler builds in win32/Makefile.msc [Rowe]
+- Include sys/types.h for Microsoft for off_t definition
+- Fix memory leak on error in gz_open()
+- Symbolize nm as $NM in configure [Weigelt]
+- Use TEST_LDSHARED instead of LDSHARED to link test programs [Weigelt]
+- Add +0 to _FILE_OFFSET_BITS and _LFS64_LARGEFILE in case not defined
+- Fix bug in gzeof() to take into account unused input data
+- Avoid initialization of structures with variables in puff.c
+- Updated win32/README-WIN32.txt [Rowe]
+
+Changes in 1.2.4.1 (28 Mar 2010)
+- Remove the use of [a-z] constructs for sed in configure [gentoo 310225]
+- Remove $(SHAREDLIB) from LIBS in Makefile.in [Creech]
+- Restore "for debugging" comment on sprintf() in gzlib.c
+- Remove fdopen for MVS from gzguts.h
+- Put new README-WIN32.txt in win32 [Rowe]
+- Add check for shell to configure and invoke another shell if needed
+- Fix big fat stinking bug in gzseek() on uncompressed files
+- Remove vestigial F_OPEN64 define in zutil.h
+- Set and check the value of _LARGEFILE_SOURCE and _LARGEFILE64_SOURCE
+- Avoid errors on non-LFS systems when applications define LFS macros
+- Set EXE to ".exe" in configure for MINGW [Kahle]
+- Match crc32() in crc32.c exactly to the prototype in zlib.h [Sherrill]
+- Add prefix for cross-compilation in win32/makefile.gcc [Bar-Lev]
+- Add DLL install in win32/makefile.gcc [Bar-Lev]
+- Allow Linux* or linux* from uname in configure [Bar-Lev]
+- Allow ldconfig to be redefined in configure and Makefile.in [Bar-Lev]
+- Add cross-compilation prefixes to configure [Bar-Lev]
+- Match type exactly in gz_load() invocation in gzread.c
+- Match type exactly of zcalloc() in zutil.c to zlib.h alloc_func
+- Provide prototypes for *64 functions when building zlib without LFS
+- Don't use -lc when linking shared library on MinGW
+- Remove errno.h check in configure and vestigial errno code in zutil.h
+
+Changes in 1.2.4 (14 Mar 2010)
+- Fix VER3 extraction in configure for no fourth subversion
+- Update zlib.3, add docs to Makefile.in to make .pdf out of it
+- Add zlib.3.pdf to distribution
+- Don't set error code in gzerror() if passed pointer is NULL
+- Apply destination directory fixes to CMakeLists.txt [Lowman]
+- Move #cmakedefine's to a new zconf.in.cmakein
+- Restore zconf.h for builds that don't use configure or cmake
+- Add distclean to dummy Makefile for convenience
+- Update and improve INDEX, README, and FAQ
+- Update CMakeLists.txt for the return of zconf.h [Lowman]
+- Update contrib/vstudio/vc9 and vc10 [Vollant]
+- Change libz.dll.a back to libzdll.a in win32/Makefile.gcc
+- Apply license and readme changes to contrib/asm686 [Raiter]
+- Check file name lengths and add -c option in minigzip.c [Li]
+- Update contrib/amd64 and contrib/masmx86/ [Vollant]
+- Avoid use of "eof" parameter in trees.c to not shadow library variable
+- Update make_vms.com for removal of zlibdefs.h [Zinser]
+- Update assembler code and vstudio projects in contrib [Vollant]
+- Remove outdated assembler code contrib/masm686 and contrib/asm586
+- Remove old vc7 and vc8 from contrib/vstudio
+- Update win32/Makefile.msc, add ZLIB_VER_SUBREVISION [Rowe]
+- Fix memory leaks in gzclose_r() and gzclose_w(), file leak in gz_open()
+- Add contrib/gcc_gvmat64 for longest_match and inflate_fast [Vollant]
+- Remove *64 functions from win32/zlib.def (they're not 64-bit yet)
+- Fix bug in void-returning vsprintf() case in gzwrite.c
+- Fix name change from inflate.h in contrib/inflate86/inffas86.c
+- Check if temporary file exists before removing in make_vms.com [Zinser]
+- Fix make install and uninstall for --static option
+- Fix usage of _MSC_VER in gzguts.h and zutil.h [Truta]
+- Update readme.txt in contrib/masmx64 and masmx86 to assemble
+
+Changes in 1.2.3.9 (21 Feb 2010)
+- Expunge gzio.c
+- Move as400 build information to old
+- Fix updates in contrib/minizip and contrib/vstudio
+- Add const to vsnprintf test in configure to avoid warnings [Weigelt]
+- Delete zconf.h (made by configure) [Weigelt]
+- Change zconf.in.h to zconf.h.in per convention [Weigelt]
+- Check for NULL buf in gzgets()
+- Return empty string for gzgets() with len == 1 (like fgets())
+- Fix description of gzgets() in zlib.h for end-of-file, NULL return
+- Update minizip to 1.1 [Vollant]
+- Avoid MSVC loss of data warnings in gzread.c, gzwrite.c
+- Note in zlib.h that gzerror() should be used to distinguish from EOF
+- Remove use of snprintf() from gzlib.c
+- Fix bug in gzseek()
+- Update contrib/vstudio, adding vc9 and vc10 [Kuno, Vollant]
+- Fix zconf.h generation in CMakeLists.txt [Lowman]
+- Improve comments in zconf.h where modified by configure
+
+Changes in 1.2.3.8 (13 Feb 2010)
+- Clean up text files (tabs, trailing whitespace, etc.) [Oberhumer]
+- Use z_off64_t in gz_zero() and gz_skip() to match state->skip
+- Avoid comparison problem when sizeof(int) == sizeof(z_off64_t)
+- Revert to Makefile.in from 1.2.3.6 (live with the clutter)
+- Fix missing error return in gzflush(), add zlib.h note
+- Add *64 functions to zlib.map [Levin]
+- Fix signed/unsigned comparison in gz_comp()
+- Use SFLAGS when testing shared linking in configure
+- Add --64 option to ./configure to use -m64 with gcc
+- Fix ./configure --help to correctly name options
+- Have make fail if a test fails [Levin]
+- Avoid buffer overrun in contrib/masmx64/gvmat64.asm [Simpson]
+- Remove assembler object files from contrib
+
+Changes in 1.2.3.7 (24 Jan 2010)
+- Always gzopen() with O_LARGEFILE if available
+- Fix gzdirect() to work immediately after gzopen() or gzdopen()
+- Make gzdirect() more precise when the state changes while reading
+- Improve zlib.h documentation in many places
+- Catch memory allocation failure in gz_open()
+- Complete close operation if seek forward in gzclose_w() fails
+- Return Z_ERRNO from gzclose_r() if close() fails
+- Return Z_STREAM_ERROR instead of EOF for gzclose() being passed NULL
+- Return zero for gzwrite() errors to match zlib.h description
+- Return -1 on gzputs() error to match zlib.h description
+- Add zconf.in.h to allow recovery from configure modification [Weigelt]
+- Fix static library permissions in Makefile.in [Weigelt]
+- Avoid warnings in configure tests that hide functionality [Weigelt]
+- Add *BSD and DragonFly to Linux case in configure [gentoo 123571]
+- Change libzdll.a to libz.dll.a in win32/Makefile.gcc [gentoo 288212]
+- Avoid access of uninitialized data for first inflateReset2 call [Gomes]
+- Keep object files in subdirectories to reduce the clutter somewhat
+- Remove default Makefile and zlibdefs.h, add dummy Makefile
+- Add new external functions to Z_PREFIX, remove duplicates, z_z_ -> z_
+- Remove zlibdefs.h completely -- modify zconf.h instead
+
+Changes in 1.2.3.6 (17 Jan 2010)
+- Avoid void * arithmetic in gzread.c and gzwrite.c
+- Make compilers happier with const char * for gz_error message
+- Avoid unused parameter warning in inflate.c
+- Avoid signed-unsigned comparison warning in inflate.c
+- Indent #pragma's for traditional C
+- Fix usage of strwinerror() in glib.c, change to gz_strwinerror()
+- Correct email address in configure for system options
+- Update make_vms.com and add make_vms.com to contrib/minizip [Zinser]
+- Update zlib.map [Brown]
+- Fix Makefile.in for Solaris 10 make of example64 and minizip64 [Torok]
+- Apply various fixes to CMakeLists.txt [Lowman]
+- Add checks on len in gzread() and gzwrite()
+- Add error message for no more room for gzungetc()
+- Remove zlib version check in gzwrite()
+- Defer compression of gzprintf() result until need to
+- Use snprintf() in gzdopen() if available
+- Remove USE_MMAP configuration determination (only used by minigzip)
+- Remove examples/pigz.c (available separately)
+- Update examples/gun.c to 1.6
+
+Changes in 1.2.3.5 (8 Jan 2010)
+- Add space after #if in zutil.h for some compilers
+- Fix relatively harmless bug in deflate_fast() [Exarevsky]
+- Fix same problem in deflate_slow()
+- Add $(SHAREDLIBV) to LIBS in Makefile.in [Brown]
+- Add deflate_rle() for faster Z_RLE strategy run-length encoding
+- Add deflate_huff() for faster Z_HUFFMAN_ONLY encoding
+- Change name of "write" variable in inffast.c to avoid library collisions
+- Fix premature EOF from gzread() in gzio.c [Brown]
+- Use zlib header window size if windowBits is 0 in inflateInit2()
+- Remove compressBound() call in deflate.c to avoid linking compress.o
+- Replace use of errno in gz* with functions, support WinCE [Alves]
+- Provide alternative to perror() in minigzip.c for WinCE [Alves]
+- Don't use _vsnprintf on later versions of MSVC [Lowman]
+- Add CMake build script and input file [Lowman]
+- Update contrib/minizip to 1.1 [Svensson, Vollant]
+- Moved nintendods directory from contrib to .
+- Replace gzio.c with a new set of routines with the same functionality
+- Add gzbuffer(), gzoffset(), gzclose_r(), gzclose_w() as part of above
+- Update contrib/minizip to 1.1b
+- Change gzeof() to return 0 on error instead of -1 to agree with zlib.h
+
+Changes in 1.2.3.4 (21 Dec 2009)
+- Use old school .SUFFIXES in Makefile.in for FreeBSD compatibility
+- Update comments in configure and Makefile.in for default --shared
+- Fix test -z's in configure [Marquess]
+- Build examplesh and minigzipsh when not testing
+- Change NULL's to Z_NULL's in deflate.c and in comments in zlib.h
+- Import LDFLAGS from the environment in configure
+- Fix configure to populate SFLAGS with discovered CFLAGS options
+- Adapt make_vms.com to the new Makefile.in [Zinser]
+- Add zlib2ansi script for C++ compilation [Marquess]
+- Add _FILE_OFFSET_BITS=64 test to make test (when applicable)
+- Add AMD64 assembler code for longest match to contrib [Teterin]
+- Include options from $SFLAGS when doing $LDSHARED
+- Simplify 64-bit file support by introducing z_off64_t type
+- Make shared object files in objs directory to work around old Sun cc
+- Use only three-part version number for Darwin shared compiles
+- Add rc option to ar in Makefile.in for when ./configure not run
+- Add -WI,-rpath,. to LDFLAGS for OSF 1 V4*
+- Set LD_LIBRARYN32_PATH for SGI IRIX shared compile
+- Protect against _FILE_OFFSET_BITS being defined when compiling zlib
+- Rename Makefile.in targets allstatic to static and allshared to shared
+- Fix static and shared Makefile.in targets to be independent
+- Correct error return bug in gz_open() by setting state [Brown]
+- Put spaces before ;;'s in configure for better sh compatibility
+- Add pigz.c (parallel implementation of gzip) to examples/
+- Correct constant in crc32.c to UL [Leventhal]
+- Reject negative lengths in crc32_combine()
+- Add inflateReset2() function to work like inflateEnd()/inflateInit2()
+- Include sys/types.h for _LARGEFILE64_SOURCE [Brown]
+- Correct typo in doc/algorithm.txt [Janik]
+- Fix bug in adler32_combine() [Zhu]
+- Catch missing-end-of-block-code error in all inflates and in puff
+    Assures that random input to inflate eventually results in an error
+- Added enough.c (calculation of ENOUGH for inftrees.h) to examples/
+- Update ENOUGH and its usage to reflect discovered bounds
+- Fix gzerror() error report on empty input file [Brown]
+- Add ush casts in trees.c to avoid pedantic runtime errors
+- Fix typo in zlib.h uncompress() description [Reiss]
+- Correct inflate() comments with regard to automatic header detection
+- Remove deprecation comment on Z_PARTIAL_FLUSH (it stays)
+- Put new version of gzlog (2.0) in examples with interruption recovery
+- Add puff compile option to permit invalid distance-too-far streams
+- Add puff TEST command options, ability to read piped input
+- Prototype the *64 functions in zlib.h when _FILE_OFFSET_BITS == 64, but
+  _LARGEFILE64_SOURCE not defined
+- Fix Z_FULL_FLUSH to truly erase the past by resetting s->strstart
+- Fix deflateSetDictionary() to use all 32K for output consistency
+- Remove extraneous #define MIN_LOOKAHEAD in deflate.c (in deflate.h)
+- Clear bytes after deflate lookahead to avoid use of uninitialized data
+- Change a limit in inftrees.c to be more transparent to Coverity Prevent
+- Update win32/zlib.def with exported symbols from zlib.h
+- Correct spelling error in zlib.h [Willem]
+- Allow Z_BLOCK for deflate() to force a new block
+- Allow negative bits in inflatePrime() to delete existing bit buffer
+- Add Z_TREES flush option to inflate() to return at end of trees
+- Add inflateMark() to return current state information for random access
+- Add Makefile for NintendoDS to contrib [Costa]
+- Add -w in configure compile tests to avoid spurious warnings [Beucler]
+- Fix typos in zlib.h comments for deflateSetDictionary()
+- Fix EOF detection in transparent gzread() [Maier]
+
+Changes in 1.2.3.3 (2 October 2006)
+- Make --shared the default for configure, add a --static option
+- Add compile option to permit invalid distance-too-far streams
+- Add inflateUndermine() function which is required to enable above
+- Remove use of "this" variable name for C++ compatibility [Marquess]
+- Add testing of shared library in make test, if shared library built
+- Use ftello() and fseeko() if available instead of ftell() and fseek()
+- Provide two versions of all functions that use the z_off_t type for
+  binary compatibility -- a normal version and a 64-bit offset version,
+  per the Large File Support Extension when _LARGEFILE64_SOURCE is
+  defined; use the 64-bit versions by default when _FILE_OFFSET_BITS
+  is defined to be 64
+- Add a --uname= option to configure to perhaps help with cross-compiling
+
+Changes in 1.2.3.2 (3 September 2006)
+- Turn off silly Borland warnings [Hay]
+- Use off64_t and define _LARGEFILE64_SOURCE when present
+- Fix missing dependency on inffixed.h in Makefile.in
+- Rig configure --shared to build both shared and static [Teredesai, Truta]
+- Remove zconf.in.h and instead create a new zlibdefs.h file
+- Fix contrib/minizip/unzip.c non-encrypted after encrypted [Vollant]
+- Add treebuild.xml (see http://treebuild.metux.de/) [Weigelt]
+
+Changes in 1.2.3.1 (16 August 2006)
+- Add watcom directory with OpenWatcom make files [Daniel]
+- Remove #undef of FAR in zconf.in.h for MVS [Fedtke]
+- Update make_vms.com [Zinser]
+- Use -fPIC for shared build in configure [Teredesai, Nicholson]
+- Use only major version number for libz.so on IRIX and OSF1 [Reinholdtsen]
+- Use fdopen() (not _fdopen()) for Interix in zutil.h [B?ck]
+- Add some FAQ entries about the contrib directory
+- Update the MVS question in the FAQ
+- Avoid extraneous reads after EOF in gzio.c [Brown]
+- Correct spelling of "successfully" in gzio.c [Randers-Pehrson]
+- Add comments to zlib.h about gzerror() usage [Brown]
+- Set extra flags in gzip header in gzopen() like deflate() does
+- Make configure options more compatible with double-dash conventions
+  [Weigelt]
+- Clean up compilation under Solaris SunStudio cc [Rowe, Reinholdtsen]
+- Fix uninstall target in Makefile.in [Truta]
+- Add pkgconfig support [Weigelt]
+- Use $(DESTDIR) macro in Makefile.in [Reinholdtsen, Weigelt]
+- Replace set_data_type() with a more accurate detect_data_type() in
+  trees.c, according to the txtvsbin.txt document [Truta]
+- Swap the order of #include <stdio.h> and #include "zlib.h" in
+  gzio.c, example.c and minigzip.c [Truta]
+- Shut up annoying VS2005 warnings about standard C deprecation [Rowe,
+  Truta] (where?)
+- Fix target "clean" from win32/Makefile.bor [Truta]
+- Create .pdb and .manifest files in win32/makefile.msc [Ziegler, Rowe]
+- Update zlib www home address in win32/DLL_FAQ.txt [Truta]
+- Update contrib/masmx86/inffas32.asm for VS2005 [Vollant, Van Wassenhove]
+- Enable browse info in the "Debug" and "ASM Debug" configurations in
+  the Visual C++ 6 project, and set (non-ASM) "Debug" as default [Truta]
+- Add pkgconfig support [Weigelt]
+- Add ZLIB_VER_MAJOR, ZLIB_VER_MINOR and ZLIB_VER_REVISION in zlib.h,
+  for use in win32/zlib1.rc [Polushin, Rowe, Truta]
+- Add a document that explains the new text detection scheme to
+  doc/txtvsbin.txt [Truta]
+- Add rfc1950.txt, rfc1951.txt and rfc1952.txt to doc/ [Truta]
+- Move algorithm.txt into doc/ [Truta]
+- Synchronize FAQ with website
+- Fix compressBound(), was low for some pathological cases [Fearnley]
+- Take into account wrapper variations in deflateBound()
+- Set examples/zpipe.c input and output to binary mode for Windows
+- Update examples/zlib_how.html with new zpipe.c (also web site)
+- Fix some warnings in examples/gzlog.c and examples/zran.c (it seems
+  that gcc became pickier in 4.0)
+- Add zlib.map for Linux: "All symbols from zlib-1.1.4 remain
+  un-versioned, the patch adds versioning only for symbols introduced in
+  zlib-1.2.0 or later.  It also declares as local those symbols which are
+  not designed to be exported." [Levin]
+- Update Z_PREFIX list in zconf.in.h, add --zprefix option to configure
+- Do not initialize global static by default in trees.c, add a response
+  NO_INIT_GLOBAL_POINTERS to initialize them if needed [Marquess]
+- Don't use strerror() in gzio.c under WinCE [Yakimov]
+- Don't use errno.h in zutil.h under WinCE [Yakimov]
+- Move arguments for AR to its usage to allow replacing ar [Marot]
+- Add HAVE_VISIBILITY_PRAGMA in zconf.in.h for Mozilla [Randers-Pehrson]
+- Improve inflateInit() and inflateInit2() documentation
+- Fix structure size comment in inflate.h
+- Change configure help option from --h* to --help [Santos]
+
 Changes in 1.2.3 (18 July 2005)
 - Apply security vulnerability fixes to contrib/infback9 as well
 - Clean up some text files (carriage returns, trailing space)
@@ -13,7 +366,7 @@
   compile
 - Fix some spelling errors in comments [Betts]
 - Correct inflateInit2() error return documentation in zlib.h
-- Added zran.c example of compressed data random access to examples
+- Add zran.c example of compressed data random access to examples
   directory, shows use of inflatePrime()
 - Fix cast for assignments to strm->state in inflate.c and infback.c
 - Fix zlibCompileFlags() in zutil.c to use 1L for long shifts [Oberhumer]

Modified: python/branches/py3k/Modules/zlib/FAQ
==============================================================================
--- python/branches/py3k/Modules/zlib/FAQ	(original)
+++ python/branches/py3k/Modules/zlib/FAQ	Fri Jul 30 22:03:17 2010
@@ -3,8 +3,8 @@
 
 
 If your question is not there, please check the zlib home page
-http://www.zlib.org which may have more recent information.
-The lastest zlib FAQ is at http://www.gzip.org/zlib/zlib_faq.html
+http://zlib.net/ which may have more recent information.
+The lastest zlib FAQ is at http://zlib.net/zlib_faq.html
 
 
  1. Is zlib Y2K-compliant?
@@ -13,54 +13,51 @@
 
  2. Where can I get a Windows DLL version?
 
-    The zlib sources can be compiled without change to produce a DLL.
-    See the file win32/DLL_FAQ.txt in the zlib distribution.
-    Pointers to the precompiled DLL are found in the zlib web site at
-    http://www.zlib.org.
+    The zlib sources can be compiled without change to produce a DLL.  See the
+    file win32/DLL_FAQ.txt in the zlib distribution.  Pointers to the
+    precompiled DLL are found in the zlib web site at http://zlib.net/ .
 
  3. Where can I get a Visual Basic interface to zlib?
 
     See
-        * http://www.dogma.net/markn/articles/zlibtool/zlibtool.htm
-        * contrib/visual-basic.txt in the zlib distribution
+        * http://marknelson.us/1997/01/01/zlib-engine/
         * win32/DLL_FAQ.txt in the zlib distribution
 
  4. compress() returns Z_BUF_ERROR.
 
-    Make sure that before the call of compress, the length of the compressed
-    buffer is equal to the total size of the compressed buffer and not
-    zero. For Visual Basic, check that this parameter is passed by reference
+    Make sure that before the call of compress(), the length of the compressed
+    buffer is equal to the available size of the compressed buffer and not
+    zero.  For Visual Basic, check that this parameter is passed by reference
     ("as any"), not by value ("as long").
 
  5. deflate() or inflate() returns Z_BUF_ERROR.
 
-    Before making the call, make sure that avail_in and avail_out are not
-    zero. When setting the parameter flush equal to Z_FINISH, also make sure
-    that avail_out is big enough to allow processing all pending input.
-    Note that a Z_BUF_ERROR is not fatal--another call to deflate() or
-    inflate() can be made with more input or output space. A Z_BUF_ERROR
-    may in fact be unavoidable depending on how the functions are used, since
-    it is not possible to tell whether or not there is more output pending
-    when strm.avail_out returns with zero.
+    Before making the call, make sure that avail_in and avail_out are not zero.
+    When setting the parameter flush equal to Z_FINISH, also make sure that
+    avail_out is big enough to allow processing all pending input.  Note that a
+    Z_BUF_ERROR is not fatal--another call to deflate() or inflate() can be
+    made with more input or output space.  A Z_BUF_ERROR may in fact be
+    unavoidable depending on how the functions are used, since it is not
+    possible to tell whether or not there is more output pending when
+    strm.avail_out returns with zero.  See http://zlib.net/zlib_how.html for a
+    heavily annotated example.
 
  6. Where's the zlib documentation (man pages, etc.)?
 
-    It's in zlib.h for the moment, and Francis S. Lin has converted it to a
-    web page zlib.html. Volunteers to transform this to Unix-style man pages,
-    please contact us (zlib at gzip.org). Examples of zlib usage are in the files
-    example.c and minigzip.c.
+    It's in zlib.h .  Examples of zlib usage are in the files example.c and
+    minigzip.c, with more in examples/ .
 
  7. Why don't you use GNU autoconf or libtool or ...?
 
-    Because we would like to keep zlib as a very small and simple
-    package. zlib is rather portable and doesn't need much configuration.
+    Because we would like to keep zlib as a very small and simple package.
+    zlib is rather portable and doesn't need much configuration.
 
  8. I found a bug in zlib.
 
-    Most of the time, such problems are due to an incorrect usage of
-    zlib. Please try to reproduce the problem with a small program and send
-    the corresponding source to us at zlib at gzip.org . Do not send
-    multi-megabyte data files without prior agreement.
+    Most of the time, such problems are due to an incorrect usage of zlib.
+    Please try to reproduce the problem with a small program and send the
+    corresponding source to us at zlib at gzip.org .  Do not send multi-megabyte
+    data files without prior agreement.
 
  9. Why do I get "undefined reference to gzputc"?
 
@@ -82,7 +79,7 @@
 
 12. Can zlib handle .Z files?
 
-    No, sorry. You have to spawn an uncompress or gunzip subprocess, or adapt
+    No, sorry.  You have to spawn an uncompress or gunzip subprocess, or adapt
     the code of uncompress on your own.
 
 13. How can I make a Unix shared library?
@@ -99,8 +96,10 @@
 
     However, many flavors of Unix come with a shared zlib already installed.
     Before going to the trouble of compiling a shared version of zlib and
-    trying to install it, you may want to check if it's already there! If you
-    can #include <zlib.h>, it's there. The -lz option will probably link to it.
+    trying to install it, you may want to check if it's already there!  If you
+    can #include <zlib.h>, it's there.  The -lz option will probably link to
+    it.  You can check the version at the top of zlib.h or with the
+    ZLIB_VERSION symbol defined in zlib.h .
 
 15. I have a question about OttoPDF.
 
@@ -109,8 +108,8 @@
 
 16. Can zlib decode Flate data in an Adobe PDF file?
 
-    Yes. See http://www.fastio.com/ (ClibPDF), or http://www.pdflib.com/ .
-    To modify PDF forms, see http://sourceforge.net/projects/acroformtool/ .
+    Yes. See http://www.pdflib.com/ . To modify PDF forms, see
+    http://sourceforge.net/projects/acroformtool/ .
 
 17. Why am I getting this "register_frame_info not found" error on Solaris?
 
@@ -121,67 +120,67 @@
         symbol __register_frame_info: referenced symbol not found
 
     The symbol __register_frame_info is not part of zlib, it is generated by
-    the C compiler (cc or gcc). You must recompile applications using zlib
-    which have this problem. This problem is specific to Solaris. See
+    the C compiler (cc or gcc).  You must recompile applications using zlib
+    which have this problem.  This problem is specific to Solaris.  See
     http://www.sunfreeware.com for Solaris versions of zlib and applications
     using zlib.
 
 18. Why does gzip give an error on a file I make with compress/deflate?
 
     The compress and deflate functions produce data in the zlib format, which
-    is different and incompatible with the gzip format. The gz* functions in
-    zlib on the other hand use the gzip format. Both the zlib and gzip
-    formats use the same compressed data format internally, but have different
-    headers and trailers around the compressed data.
+    is different and incompatible with the gzip format.  The gz* functions in
+    zlib on the other hand use the gzip format.  Both the zlib and gzip formats
+    use the same compressed data format internally, but have different headers
+    and trailers around the compressed data.
 
 19. Ok, so why are there two different formats?
 
-    The gzip format was designed to retain the directory information about
-    a single file, such as the name and last modification date. The zlib
-    format on the other hand was designed for in-memory and communication
-    channel applications, and has a much more compact header and trailer and
-    uses a faster integrity check than gzip.
+    The gzip format was designed to retain the directory information about a
+    single file, such as the name and last modification date.  The zlib format
+    on the other hand was designed for in-memory and communication channel
+    applications, and has a much more compact header and trailer and uses a
+    faster integrity check than gzip.
 
 20. Well that's nice, but how do I make a gzip file in memory?
 
     You can request that deflate write the gzip format instead of the zlib
-    format using deflateInit2(). You can also request that inflate decode
-    the gzip format using inflateInit2(). Read zlib.h for more details.
+    format using deflateInit2().  You can also request that inflate decode the
+    gzip format using inflateInit2().  Read zlib.h for more details.
 
 21. Is zlib thread-safe?
 
-    Yes. However any library routines that zlib uses and any application-
-    provided memory allocation routines must also be thread-safe. zlib's gz*
+    Yes.  However any library routines that zlib uses and any application-
+    provided memory allocation routines must also be thread-safe.  zlib's gz*
     functions use stdio library routines, and most of zlib's functions use the
-    library memory allocation routines by default. zlib's Init functions allow
-    for the application to provide custom memory allocation routines.
+    library memory allocation routines by default.  zlib's *Init* functions
+    allow for the application to provide custom memory allocation routines.
 
     Of course, you should only operate on any given zlib or gzip stream from a
     single thread at a time.
 
 22. Can I use zlib in my commercial application?
 
-    Yes. Please read the license in zlib.h.
+    Yes.  Please read the license in zlib.h.
 
 23. Is zlib under the GNU license?
 
-    No. Please read the license in zlib.h.
+    No.  Please read the license in zlib.h.
 
 24. The license says that altered source versions must be "plainly marked". So
     what exactly do I need to do to meet that requirement?
 
-    You need to change the ZLIB_VERSION and ZLIB_VERNUM #defines in zlib.h. In
+    You need to change the ZLIB_VERSION and ZLIB_VERNUM #defines in zlib.h.  In
     particular, the final version number needs to be changed to "f", and an
-    identification string should be appended to ZLIB_VERSION. Version numbers
+    identification string should be appended to ZLIB_VERSION.  Version numbers
     x.x.x.f are reserved for modifications to zlib by others than the zlib
-    maintainers. For example, if the version of the base zlib you are altering
+    maintainers.  For example, if the version of the base zlib you are altering
     is "1.2.3.4", then in zlib.h you should change ZLIB_VERNUM to 0x123f, and
-    ZLIB_VERSION to something like "1.2.3.f-zachary-mods-v3". You can also
+    ZLIB_VERSION to something like "1.2.3.f-zachary-mods-v3".  You can also
     update the version strings in deflate.c and inftrees.c.
 
     For altered source distributions, you should also note the origin and
     nature of the changes in zlib.h, as well as in ChangeLog and README, along
-    with the dates of the alterations. The origin should include at least your
+    with the dates of the alterations.  The origin should include at least your
     name (or your company's name), and an email address to contact for help or
     issues with the library.
 
@@ -197,105 +196,112 @@
 
 26. Will zlib work on a 64-bit machine?
 
-    It should. It has been tested on 64-bit machines, and has no dependence
-    on any data types being limited to 32-bits in length. If you have any
+    Yes.  It has been tested on 64-bit machines, and has no dependence on any
+    data types being limited to 32-bits in length.  If you have any
     difficulties, please provide a complete problem report to zlib at gzip.org
 
 27. Will zlib decompress data from the PKWare Data Compression Library?
 
-    No. The PKWare DCL uses a completely different compressed data format
-    than does PKZIP and zlib. However, you can look in zlib's contrib/blast
+    No.  The PKWare DCL uses a completely different compressed data format than
+    does PKZIP and zlib.  However, you can look in zlib's contrib/blast
     directory for a possible solution to your problem.
 
 28. Can I access data randomly in a compressed stream?
 
-    No, not without some preparation. If when compressing you periodically
-    use Z_FULL_FLUSH, carefully write all the pending data at those points,
-    and keep an index of those locations, then you can start decompression
-    at those points. You have to be careful to not use Z_FULL_FLUSH too
-    often, since it can significantly degrade compression.
+    No, not without some preparation.  If when compressing you periodically use
+    Z_FULL_FLUSH, carefully write all the pending data at those points, and
+    keep an index of those locations, then you can start decompression at those
+    points.  You have to be careful to not use Z_FULL_FLUSH too often, since it
+    can significantly degrade compression.  Alternatively, you can scan a
+    deflate stream once to generate an index, and then use that index for
+    random access.  See examples/zran.c .
 
 29. Does zlib work on MVS, OS/390, CICS, etc.?
 
-    We don't know for sure. We have heard occasional reports of success on
-    these systems. If you do use it on one of these, please provide us with
-    a report, instructions, and patches that we can reference when we get
-    these questions. Thanks.
+    It has in the past, but we have not heard of any recent evidence.  There
+    were working ports of zlib 1.1.4 to MVS, but those links no longer work.
+    If you know of recent, successful applications of zlib on these operating
+    systems, please let us know.  Thanks.
 
-30. Is there some simpler, easier to read version of inflate I can look at
-    to understand the deflate format?
+30. Is there some simpler, easier to read version of inflate I can look at to
+    understand the deflate format?
 
-    First off, you should read RFC 1951. Second, yes. Look in zlib's
+    First off, you should read RFC 1951.  Second, yes.  Look in zlib's
     contrib/puff directory.
 
 31. Does zlib infringe on any patents?
 
-    As far as we know, no. In fact, that was originally the whole point behind
-    zlib. Look here for some more information:
+    As far as we know, no.  In fact, that was originally the whole point behind
+    zlib.  Look here for some more information:
 
     http://www.gzip.org/#faq11
 
 32. Can zlib work with greater than 4 GB of data?
 
-    Yes. inflate() and deflate() will process any amount of data correctly.
+    Yes.  inflate() and deflate() will process any amount of data correctly.
     Each call of inflate() or deflate() is limited to input and output chunks
     of the maximum value that can be stored in the compiler's "unsigned int"
-    type, but there is no limit to the number of chunks. Note however that the
-    strm.total_in and strm_total_out counters may be limited to 4 GB. These
+    type, but there is no limit to the number of chunks.  Note however that the
+    strm.total_in and strm_total_out counters may be limited to 4 GB.  These
     counters are provided as a convenience and are not used internally by
-    inflate() or deflate(). The application can easily set up its own counters
+    inflate() or deflate().  The application can easily set up its own counters
     updated after each call of inflate() or deflate() to count beyond 4 GB.
     compress() and uncompress() may be limited to 4 GB, since they operate in a
-    single call. gzseek() and gztell() may be limited to 4 GB depending on how
-    zlib is compiled. See the zlibCompileFlags() function in zlib.h.
+    single call.  gzseek() and gztell() may be limited to 4 GB depending on how
+    zlib is compiled.  See the zlibCompileFlags() function in zlib.h.
 
-    The word "may" appears several times above since there is a 4 GB limit
-    only if the compiler's "long" type is 32 bits. If the compiler's "long"
-    type is 64 bits, then the limit is 16 exabytes.
+    The word "may" appears several times above since there is a 4 GB limit only
+    if the compiler's "long" type is 32 bits.  If the compiler's "long" type is
+    64 bits, then the limit is 16 exabytes.
 
 33. Does zlib have any security vulnerabilities?
 
-    The only one that we are aware of is potentially in gzprintf(). If zlib
-    is compiled to use sprintf() or vsprintf(), then there is no protection
-    against a buffer overflow of a 4K string space, other than the caller of
-    gzprintf() assuring that the output will not exceed 4K. On the other
-    hand, if zlib is compiled to use snprintf() or vsnprintf(), which should
-    normally be the case, then there is no vulnerability. The ./configure
-    script will display warnings if an insecure variation of sprintf() will
-    be used by gzprintf(). Also the zlibCompileFlags() function will return
-    information on what variant of sprintf() is used by gzprintf().
+    The only one that we are aware of is potentially in gzprintf().  If zlib is
+    compiled to use sprintf() or vsprintf(), then there is no protection
+    against a buffer overflow of an 8K string space (or other value as set by
+    gzbuffer()), other than the caller of gzprintf() assuring that the output
+    will not exceed 8K.  On the other hand, if zlib is compiled to use
+    snprintf() or vsnprintf(), which should normally be the case, then there is
+    no vulnerability.  The ./configure script will display warnings if an
+    insecure variation of sprintf() will be used by gzprintf().  Also the
+    zlibCompileFlags() function will return information on what variant of
+    sprintf() is used by gzprintf().
 
     If you don't have snprintf() or vsnprintf() and would like one, you can
     find a portable implementation here:
 
         http://www.ijs.si/software/snprintf/
 
-    Note that you should be using the most recent version of zlib. Versions
-    1.1.3 and before were subject to a double-free vulnerability.
+    Note that you should be using the most recent version of zlib.  Versions
+    1.1.3 and before were subject to a double-free vulnerability, and versions
+    1.2.1 and 1.2.2 were subject to an access exception when decompressing
+    invalid compressed data.
 
 34. Is there a Java version of zlib?
 
     Probably what you want is to use zlib in Java. zlib is already included
     as part of the Java SDK in the java.util.zip package. If you really want
     a version of zlib written in the Java language, look on the zlib home
-    page for links: http://www.zlib.org/
+    page for links: http://zlib.net/ .
 
 35. I get this or that compiler or source-code scanner warning when I crank it
     up to maximally-pedantic. Can't you guys write proper code?
 
     Many years ago, we gave up attempting to avoid warnings on every compiler
-    in the universe. It just got to be a waste of time, and some compilers
-    were downright silly. So now, we simply make sure that the code always
-    works.
+    in the universe.  It just got to be a waste of time, and some compilers
+    were downright silly as well as contradicted each other.  So now, we simply
+    make sure that the code always works.
 
 36. Valgrind (or some similar memory access checker) says that deflate is
     performing a conditional jump that depends on an uninitialized value.
     Isn't that a bug?
 
-    No.  That is intentional for performance reasons, and the output of
-    deflate is not affected.  This only started showing up recently since
-    zlib 1.2.x uses malloc() by default for allocations, whereas earlier
-    versions used calloc(), which zeros out the allocated memory.
+    No.  That is intentional for performance reasons, and the output of deflate
+    is not affected.  This only started showing up recently since zlib 1.2.x
+    uses malloc() by default for allocations, whereas earlier versions used
+    calloc(), which zeros out the allocated memory.  Even though the code was
+    correct, versions 1.2.4 and later was changed to not stimulate these
+    checkers.
 
 37. Will zlib read the (insert any ancient or arcane format here) compressed
     data format?
@@ -305,20 +311,21 @@
 
 38. How can I encrypt/decrypt zip files with zlib?
 
-    zlib doesn't support encryption. The original PKZIP encryption is very weak
-    and can be broken with freely available programs. To get strong encryption,
-    use GnuPG, http://www.gnupg.org/ , which already includes zlib compression.
-    For PKZIP compatible "encryption", look at http://www.info-zip.org/
+    zlib doesn't support encryption.  The original PKZIP encryption is very
+    weak and can be broken with freely available programs.  To get strong
+    encryption, use GnuPG, http://www.gnupg.org/ , which already includes zlib
+    compression.  For PKZIP compatible "encryption", look at
+    http://www.info-zip.org/
 
 39. What's the difference between the "gzip" and "deflate" HTTP 1.1 encodings?
 
-    "gzip" is the gzip format, and "deflate" is the zlib format. They should
-    probably have called the second one "zlib" instead to avoid confusion
-    with the raw deflate compressed data format. While the HTTP 1.1 RFC 2616
+    "gzip" is the gzip format, and "deflate" is the zlib format.  They should
+    probably have called the second one "zlib" instead to avoid confusion with
+    the raw deflate compressed data format.  While the HTTP 1.1 RFC 2616
     correctly points to the zlib specification in RFC 1950 for the "deflate"
     transfer encoding, there have been reports of servers and browsers that
     incorrectly produce or expect raw deflate data per the deflate
-    specficiation in RFC 1951, most notably Microsoft. So even though the
+    specficiation in RFC 1951, most notably Microsoft.  So even though the
     "deflate" transfer encoding using the zlib format would be the more
     efficient approach (and in fact exactly what the zlib format was designed
     for), using the "gzip" transfer encoding is probably more reliable due to
@@ -328,12 +335,32 @@
 
 40. Does zlib support the new "Deflate64" format introduced by PKWare?
 
-    No. PKWare has apparently decided to keep that format proprietary, since
-    they have not documented it as they have previous compression formats.
-    In any case, the compression improvements are so modest compared to other
-    more modern approaches, that it's not worth the effort to implement.
+    No.  PKWare has apparently decided to keep that format proprietary, since
+    they have not documented it as they have previous compression formats.  In
+    any case, the compression improvements are so modest compared to other more
+    modern approaches, that it's not worth the effort to implement.
 
-41. Can you please sign these lengthy legal documents and fax them back to us
+41. I'm having a problem with the zip functions in zlib, can you help?
+
+    There are no zip functions in zlib.  You are probably using minizip by
+    Giles Vollant, which is found in the contrib directory of zlib.  It is not
+    part of zlib.  In fact none of the stuff in contrib is part of zlib.  The
+    files in there are not supported by the zlib authors.  You need to contact
+    the authors of the respective contribution for help.
+
+42. The match.asm code in contrib is under the GNU General Public License.
+    Since it's part of zlib, doesn't that mean that all of zlib falls under the
+    GNU GPL?
+
+    No.  The files in contrib are not part of zlib.  They were contributed by
+    other authors and are provided as a convenience to the user within the zlib
+    distribution.  Each item in contrib has its own license.
+
+43. Is zlib subject to export controls?  What is its ECCN?
+
+    zlib is not subject to export controls, and so is classified as EAR99.
+
+44. Can you please sign these lengthy legal documents and fax them back to us
     so that we can use your software in our product?
 
     No. Go away. Shoo.

Modified: python/branches/py3k/Modules/zlib/INDEX
==============================================================================
--- python/branches/py3k/Modules/zlib/INDEX	(original)
+++ python/branches/py3k/Modules/zlib/INDEX	Fri Jul 30 22:03:17 2010
@@ -1,23 +1,32 @@
+CMakeLists.txt  cmake build file
 ChangeLog       history of changes
 FAQ             Frequently Asked Questions about zlib
 INDEX           this file
-Makefile        makefile for Unix (generated by configure)
-Makefile.in     makefile for Unix (template for configure)
+Makefile        dummy Makefile that tells you to ./configure
+Makefile.in     template for Unix Makefile
 README          guess what
-algorithm.txt   description of the (de)compression algorithm
 configure       configure script for Unix
-zconf.in.h      template for zconf.h (used by configure)
+make_vms.com    makefile for VMS
+treebuild.xml   XML description of source file dependencies
+zconf.h.cmakein zconf.h template for cmake
+zconf.h.in      zconf.h template for configure
+zlib.3          Man page for zlib
+zlib.3.pdf      Man page in PDF format
+zlib.map        Linux symbol information
+zlib.pc.in      Template for pkg-config descriptor
+zlib2ansi       perl script to convert source files for C++ compilation
 
 amiga/          makefiles for Amiga SAS C
-as400/          makefiles for IBM AS/400
+doc/            documentation for formats and algorithms
 msdos/          makefiles for MSDOS
+nintendods/     makefile for Nintendo DS
 old/            makefiles for various architectures and zlib documentation
                 files that have not yet been updated for zlib 1.2.x
-projects/       projects for various Integrated Development Environments
 qnx/            makefiles for QNX
+watcom/         makefiles for OpenWatcom
 win32/          makefiles for Windows
 
-                zlib public header files (must be kept):
+                zlib public header files (required for library use):
 zconf.h
 zlib.h
 
@@ -28,7 +37,11 @@
 crc32.h
 deflate.c
 deflate.h
-gzio.c
+gzclose.c
+gzguts.h
+gzlib.c
+gzread.c
+gzwrite.c
 infback.c
 inffast.c
 inffast.h
@@ -46,6 +59,7 @@
                 source files for sample programs:
 example.c
 minigzip.c
+See examples/README.examples for more
 
                 unsupported contribution by third parties
 See contrib/README.contrib

Modified: python/branches/py3k/Modules/zlib/Makefile
==============================================================================
--- python/branches/py3k/Modules/zlib/Makefile	(original)
+++ python/branches/py3k/Modules/zlib/Makefile	Fri Jul 30 22:03:17 2010
@@ -1,154 +1,5 @@
-# Makefile for zlib
-# Copyright (C) 1995-2005 Jean-loup Gailly.
-# For conditions of distribution and use, see copyright notice in zlib.h
+all:
+	- at echo "Please use ./configure first.  Thank you."
 
-# To compile and test, type:
-#    ./configure; make test
-# The call of configure is optional if you don't have special requirements
-# If you wish to build zlib as a shared library, use: ./configure -s
-
-# To use the asm code, type:
-#    cp contrib/asm?86/match.S ./match.S
-#    make LOC=-DASMV OBJA=match.o
-
-# To install /usr/local/lib/libz.* and /usr/local/include/zlib.h, type:
-#    make install
-# To install in $HOME instead of /usr/local, use:
-#    make install prefix=$HOME
-
-CC=cc
-
-CFLAGS=-O
-#CFLAGS=-O -DMAX_WBITS=14 -DMAX_MEM_LEVEL=7
-#CFLAGS=-g -DDEBUG
-#CFLAGS=-O3 -Wall -Wwrite-strings -Wpointer-arith -Wconversion \
-#           -Wstrict-prototypes -Wmissing-prototypes
-
-LDFLAGS=libz.a
-LDSHARED=$(CC)
-CPP=$(CC) -E
-
-LIBS=libz.a
-SHAREDLIB=libz.so
-SHAREDLIBV=libz.so.1.2.3
-SHAREDLIBM=libz.so.1
-
-AR=ar rc
-RANLIB=ranlib
-TAR=tar
-SHELL=/bin/sh
-EXE=
-
-prefix = /usr/local
-exec_prefix = ${prefix}
-libdir = ${exec_prefix}/lib
-includedir = ${prefix}/include
-mandir = ${prefix}/share/man
-man3dir = ${mandir}/man3
-
-OBJS = adler32.o compress.o crc32.o gzio.o uncompr.o deflate.o trees.o \
-       zutil.o inflate.o infback.o inftrees.o inffast.o
-
-OBJA =
-# to use the asm code: make OBJA=match.o
-
-TEST_OBJS = example.o minigzip.o
-
-all: example$(EXE) minigzip$(EXE)
-
-check: test
-test: all
-	@LD_LIBRARY_PATH=.:$(LD_LIBRARY_PATH) ; export LD_LIBRARY_PATH; \
-	echo hello world | ./minigzip | ./minigzip -d || \
-	  echo '		*** minigzip test FAILED ***' ; \
-	if ./example; then \
-	  echo '		*** zlib test OK ***'; \
-	else \
-	  echo '		*** zlib test FAILED ***'; \
-	fi
-
-libz.a: $(OBJS) $(OBJA)
-	$(AR) $@ $(OBJS) $(OBJA)
-	-@ ($(RANLIB) $@ || true) >/dev/null 2>&1
-
-match.o: match.S
-	$(CPP) match.S > _match.s
-	$(CC) -c _match.s
-	mv _match.o match.o
-	rm -f _match.s
-
-$(SHAREDLIBV): $(OBJS)
-	$(LDSHARED) -o $@ $(OBJS)
-	rm -f $(SHAREDLIB) $(SHAREDLIBM)
-	ln -s $@ $(SHAREDLIB)
-	ln -s $@ $(SHAREDLIBM)
-
-example$(EXE): example.o $(LIBS)
-	$(CC) $(CFLAGS) -o $@ example.o $(LDFLAGS)
-
-minigzip$(EXE): minigzip.o $(LIBS)
-	$(CC) $(CFLAGS) -o $@ minigzip.o $(LDFLAGS)
-
-install: $(LIBS)
-	- at if [ ! -d $(exec_prefix) ]; then mkdir -p $(exec_prefix); fi
-	- at if [ ! -d $(includedir)  ]; then mkdir -p $(includedir); fi
-	- at if [ ! -d $(libdir)      ]; then mkdir -p $(libdir); fi
-	- at if [ ! -d $(man3dir)     ]; then mkdir -p $(man3dir); fi
-	cp zlib.h zconf.h $(includedir)
-	chmod 644 $(includedir)/zlib.h $(includedir)/zconf.h
-	cp $(LIBS) $(libdir)
-	cd $(libdir); chmod 755 $(LIBS)
-	-@(cd $(libdir); $(RANLIB) libz.a || true) >/dev/null 2>&1
-	cd $(libdir); if test -f $(SHAREDLIBV); then \
-	  rm -f $(SHAREDLIB) $(SHAREDLIBM); \
-	  ln -s $(SHAREDLIBV) $(SHAREDLIB); \
-	  ln -s $(SHAREDLIBV) $(SHAREDLIBM); \
-	  (ldconfig || true)  >/dev/null 2>&1; \
-	fi
-	cp zlib.3 $(man3dir)
-	chmod 644 $(man3dir)/zlib.3
-# The ranlib in install is needed on NeXTSTEP which checks file times
-# ldconfig is for Linux
-
-uninstall:
-	cd $(includedir); \
-	cd $(libdir); rm -f libz.a; \
-	if test -f $(SHAREDLIBV); then \
-	  rm -f $(SHAREDLIBV) $(SHAREDLIB) $(SHAREDLIBM); \
-	fi
-	cd $(man3dir); rm -f zlib.3
-
-mostlyclean: clean
-clean:
-	rm -f *.o *~ example$(EXE) minigzip$(EXE) \
-	   libz.* foo.gz so_locations \
-	   _match.s maketree contrib/infback9/*.o
-
-maintainer-clean: distclean
-distclean: clean
-	cp -p Makefile.in Makefile
-	cp -p zconf.in.h zconf.h
-	rm -f .DS_Store
-
-tags:
-	etags *.[ch]
-
-depend:
-	makedepend -- $(CFLAGS) -- *.[ch]
-
-# DO NOT DELETE THIS LINE -- make depend depends on it.
-
-adler32.o: zlib.h zconf.h
-compress.o: zlib.h zconf.h
-crc32.o: crc32.h zlib.h zconf.h
-deflate.o: deflate.h zutil.h zlib.h zconf.h
-example.o: zlib.h zconf.h
-gzio.o: zutil.h zlib.h zconf.h
-inffast.o: zutil.h zlib.h zconf.h inftrees.h inflate.h inffast.h
-inflate.o: zutil.h zlib.h zconf.h inftrees.h inflate.h inffast.h
-infback.o: zutil.h zlib.h zconf.h inftrees.h inflate.h inffast.h
-inftrees.o: zutil.h zlib.h zconf.h inftrees.h
-minigzip.o: zlib.h zconf.h
-trees.o: deflate.h zutil.h zlib.h zconf.h trees.h
-uncompr.o: zlib.h zconf.h
-zutil.o: zutil.h zlib.h zconf.h
+distclean:
+	make -f Makefile.in distclean

Modified: python/branches/py3k/Modules/zlib/Makefile.in
==============================================================================
--- python/branches/py3k/Modules/zlib/Makefile.in	(original)
+++ python/branches/py3k/Modules/zlib/Makefile.in	Fri Jul 30 22:03:17 2010
@@ -1,11 +1,11 @@
 # Makefile for zlib
-# Copyright (C) 1995-2005 Jean-loup Gailly.
+# Copyright (C) 1995-2010 Jean-loup Gailly.
 # For conditions of distribution and use, see copyright notice in zlib.h
 
 # To compile and test, type:
 #    ./configure; make test
-# The call of configure is optional if you don't have special requirements
-# If you wish to build zlib as a shared library, use: ./configure -s
+# Normally configure builds both a static and a shared library.
+# If you want to build just a static library, use: ./configure --static
 
 # To use the asm code, type:
 #    cp contrib/asm?86/match.S ./match.S
@@ -24,17 +24,22 @@
 #CFLAGS=-O3 -Wall -Wwrite-strings -Wpointer-arith -Wconversion \
 #           -Wstrict-prototypes -Wmissing-prototypes
 
-LDFLAGS=libz.a
+SFLAGS=-O
+LDFLAGS=
+TEST_LDFLAGS=-L. libz.a
 LDSHARED=$(CC)
 CPP=$(CC) -E
 
-LIBS=libz.a
+STATICLIB=libz.a
 SHAREDLIB=libz.so
-SHAREDLIBV=libz.so.1.2.3
+SHAREDLIBV=libz.so.1.2.5
 SHAREDLIBM=libz.so.1
+LIBS=$(STATICLIB) $(SHAREDLIBV)
 
 AR=ar rc
 RANLIB=ranlib
+LDCONFIG=ldconfig
+LDSHAREDLIBC=-lc
 TAR=tar
 SHELL=/bin/sh
 EXE=
@@ -42,33 +47,68 @@
 prefix = /usr/local
 exec_prefix = ${prefix}
 libdir = ${exec_prefix}/lib
+sharedlibdir = ${libdir}
 includedir = ${prefix}/include
 mandir = ${prefix}/share/man
 man3dir = ${mandir}/man3
+pkgconfigdir = ${libdir}/pkgconfig
 
-OBJS = adler32.o compress.o crc32.o gzio.o uncompr.o deflate.o trees.o \
-       zutil.o inflate.o infback.o inftrees.o inffast.o
+OBJC = adler32.o compress.o crc32.o deflate.o gzclose.o gzlib.o gzread.o \
+	gzwrite.o infback.o inffast.o inflate.o inftrees.o trees.o uncompr.o zutil.o
 
+PIC_OBJC = adler32.lo compress.lo crc32.lo deflate.lo gzclose.lo gzlib.lo gzread.lo \
+	gzwrite.lo infback.lo inffast.lo inflate.lo inftrees.lo trees.lo uncompr.lo zutil.lo
+
+# to use the asm code: make OBJA=match.o, PIC_OBJA=match.lo
 OBJA =
-# to use the asm code: make OBJA=match.o
+PIC_OBJA =
+
+OBJS = $(OBJC) $(OBJA)
+
+PIC_OBJS = $(PIC_OBJC) $(PIC_OBJA)
+
+all: static shared
 
-TEST_OBJS = example.o minigzip.o
+static: example$(EXE) minigzip$(EXE)
 
-all: example$(EXE) minigzip$(EXE)
+shared: examplesh$(EXE) minigzipsh$(EXE)
+
+all64: example64$(EXE) minigzip64$(EXE)
 
 check: test
-test: all
-	@LD_LIBRARY_PATH=.:$(LD_LIBRARY_PATH) ; export LD_LIBRARY_PATH; \
-	echo hello world | ./minigzip | ./minigzip -d || \
-	  echo '		*** minigzip test FAILED ***' ; \
-	if ./example; then \
+
+test: all teststatic testshared
+
+teststatic: static
+	@if echo hello world | ./minigzip | ./minigzip -d && ./example; then \
 	  echo '		*** zlib test OK ***'; \
 	else \
-	  echo '		*** zlib test FAILED ***'; \
+	  echo '		*** zlib test FAILED ***'; false; \
+	fi
+	- at rm -f foo.gz
+
+testshared: shared
+	@LD_LIBRARY_PATH=`pwd`:$(LD_LIBRARY_PATH) ; export LD_LIBRARY_PATH; \
+	LD_LIBRARYN32_PATH=`pwd`:$(LD_LIBRARYN32_PATH) ; export LD_LIBRARYN32_PATH; \
+	DYLD_LIBRARY_PATH=`pwd`:$(DYLD_LIBRARY_PATH) ; export DYLD_LIBRARY_PATH; \
+	SHLIB_PATH=`pwd`:$(SHLIB_PATH) ; export SHLIB_PATH; \
+	if echo hello world | ./minigzipsh | ./minigzipsh -d && ./examplesh; then \
+	  echo '		*** zlib shared test OK ***'; \
+	else \
+	  echo '		*** zlib shared test FAILED ***'; false; \
+	fi
+	- at rm -f foo.gz
+
+test64: all64
+	@if echo hello world | ./minigzip64 | ./minigzip64 -d && ./example64; then \
+	  echo '		*** zlib 64-bit test OK ***'; \
+	else \
+	  echo '		*** zlib 64-bit test FAILED ***'; false; \
 	fi
+	- at rm -f foo.gz
 
-libz.a: $(OBJS) $(OBJA)
-	$(AR) $@ $(OBJS) $(OBJA)
+libz.a: $(OBJS)
+	$(AR) $@ $(OBJS)
 	-@ ($(RANLIB) $@ || true) >/dev/null 2>&1
 
 match.o: match.S
@@ -77,58 +117,116 @@
 	mv _match.o match.o
 	rm -f _match.s
 
-$(SHAREDLIBV): $(OBJS)
-	$(LDSHARED) -o $@ $(OBJS)
+match.lo: match.S
+	$(CPP) match.S > _match.s
+	$(CC) -c -fPIC _match.s
+	mv _match.o match.lo
+	rm -f _match.s
+
+example64.o: example.c zlib.h zconf.h
+	$(CC) $(CFLAGS) -D_FILE_OFFSET_BITS=64 -c -o $@ example.c
+
+minigzip64.o: minigzip.c zlib.h zconf.h
+	$(CC) $(CFLAGS) -D_FILE_OFFSET_BITS=64 -c -o $@ minigzip.c
+
+.SUFFIXES: .lo
+
+.c.lo:
+	- at mkdir objs 2>/dev/null || test -d objs
+	$(CC) $(SFLAGS) -DPIC -c -o objs/$*.o $<
+	- at mv objs/$*.o $@
+
+$(SHAREDLIBV): $(PIC_OBJS)
+	$(LDSHARED) $(SFLAGS) -o $@ $(PIC_OBJS) $(LDSHAREDLIBC) $(LDFLAGS)
 	rm -f $(SHAREDLIB) $(SHAREDLIBM)
 	ln -s $@ $(SHAREDLIB)
 	ln -s $@ $(SHAREDLIBM)
+	- at rmdir objs
 
-example$(EXE): example.o $(LIBS)
-	$(CC) $(CFLAGS) -o $@ example.o $(LDFLAGS)
+example$(EXE): example.o $(STATICLIB)
+	$(CC) $(CFLAGS) -o $@ example.o $(TEST_LDFLAGS)
 
-minigzip$(EXE): minigzip.o $(LIBS)
-	$(CC) $(CFLAGS) -o $@ minigzip.o $(LDFLAGS)
+minigzip$(EXE): minigzip.o $(STATICLIB)
+	$(CC) $(CFLAGS) -o $@ minigzip.o $(TEST_LDFLAGS)
 
-install: $(LIBS)
-	- at if [ ! -d $(exec_prefix) ]; then mkdir -p $(exec_prefix); fi
-	- at if [ ! -d $(includedir)  ]; then mkdir -p $(includedir); fi
-	- at if [ ! -d $(libdir)      ]; then mkdir -p $(libdir); fi
-	- at if [ ! -d $(man3dir)     ]; then mkdir -p $(man3dir); fi
-	cp zlib.h zconf.h $(includedir)
-	chmod 644 $(includedir)/zlib.h $(includedir)/zconf.h
-	cp $(LIBS) $(libdir)
-	cd $(libdir); chmod 755 $(LIBS)
-	-@(cd $(libdir); $(RANLIB) libz.a || true) >/dev/null 2>&1
-	cd $(libdir); if test -f $(SHAREDLIBV); then \
+examplesh$(EXE): example.o $(SHAREDLIBV)
+	$(CC) $(CFLAGS) -o $@ example.o -L. $(SHAREDLIBV)
+
+minigzipsh$(EXE): minigzip.o $(SHAREDLIBV)
+	$(CC) $(CFLAGS) -o $@ minigzip.o -L. $(SHAREDLIBV)
+
+example64$(EXE): example64.o $(STATICLIB)
+	$(CC) $(CFLAGS) -o $@ example64.o $(TEST_LDFLAGS)
+
+minigzip64$(EXE): minigzip64.o $(STATICLIB)
+	$(CC) $(CFLAGS) -o $@ minigzip64.o $(TEST_LDFLAGS)
+
+install-libs: $(LIBS)
+	- at if [ ! -d $(DESTDIR)$(exec_prefix)  ]; then mkdir -p $(DESTDIR)$(exec_prefix); fi
+	- at if [ ! -d $(DESTDIR)$(libdir)       ]; then mkdir -p $(DESTDIR)$(libdir); fi
+	- at if [ ! -d $(DESTDIR)$(sharedlibdir) ]; then mkdir -p $(DESTDIR)$(sharedlibdir); fi
+	- at if [ ! -d $(DESTDIR)$(man3dir)      ]; then mkdir -p $(DESTDIR)$(man3dir); fi
+	- at if [ ! -d $(DESTDIR)$(pkgconfigdir) ]; then mkdir -p $(DESTDIR)$(pkgconfigdir); fi
+	cp $(STATICLIB) $(DESTDIR)$(libdir)
+	cp $(SHAREDLIBV) $(DESTDIR)$(sharedlibdir)
+	cd $(DESTDIR)$(libdir); chmod u=rw,go=r $(STATICLIB)
+	-@(cd $(DESTDIR)$(libdir); $(RANLIB) libz.a || true) >/dev/null 2>&1
+	- at cd $(DESTDIR)$(sharedlibdir); if test "$(SHAREDLIBV)" -a -f $(SHAREDLIBV); then \
+	  chmod 755 $(SHAREDLIBV); \
 	  rm -f $(SHAREDLIB) $(SHAREDLIBM); \
 	  ln -s $(SHAREDLIBV) $(SHAREDLIB); \
 	  ln -s $(SHAREDLIBV) $(SHAREDLIBM); \
-	  (ldconfig || true)  >/dev/null 2>&1; \
+	  ($(LDCONFIG) || true)  >/dev/null 2>&1; \
 	fi
-	cp zlib.3 $(man3dir)
-	chmod 644 $(man3dir)/zlib.3
+	cp zlib.3 $(DESTDIR)$(man3dir)
+	chmod 644 $(DESTDIR)$(man3dir)/zlib.3
+	cp zlib.pc $(DESTDIR)$(pkgconfigdir)
+	chmod 644 $(DESTDIR)$(pkgconfigdir)/zlib.pc
 # The ranlib in install is needed on NeXTSTEP which checks file times
 # ldconfig is for Linux
 
+install: install-libs
+	- at if [ ! -d $(DESTDIR)$(includedir)   ]; then mkdir -p $(DESTDIR)$(includedir); fi
+	cp zlib.h zconf.h $(DESTDIR)$(includedir)
+	chmod 644 $(DESTDIR)$(includedir)/zlib.h $(DESTDIR)$(includedir)/zconf.h
+
 uninstall:
-	cd $(includedir); \
-	cd $(libdir); rm -f libz.a; \
-	if test -f $(SHAREDLIBV); then \
+	cd $(DESTDIR)$(includedir); rm -f zlib.h zconf.h
+	cd $(DESTDIR)$(libdir); rm -f libz.a; \
+	if test "$(SHAREDLIBV)" -a -f $(SHAREDLIBV); then \
 	  rm -f $(SHAREDLIBV) $(SHAREDLIB) $(SHAREDLIBM); \
 	fi
-	cd $(man3dir); rm -f zlib.3
+	cd $(DESTDIR)$(man3dir); rm -f zlib.3
+	cd $(DESTDIR)$(pkgconfigdir); rm -f zlib.pc
+
+docs: zlib.3.pdf
+
+zlib.3.pdf: zlib.3
+	groff -mandoc -f H -T ps zlib.3 | ps2pdf - zlib.3.pdf
+
+zconf.h.in: zconf.h.cmakein
+	sed "/^#cmakedefine/D" < zconf.h.cmakein > zconf.h.in
+	touch -r zconf.h.cmakein zconf.h.in
+
+zconf: zconf.h.in
+	cp -p zconf.h.in zconf.h
 
 mostlyclean: clean
 clean:
-	rm -f *.o *~ example$(EXE) minigzip$(EXE) \
+	rm -f *.o *.lo *~ \
+	   example$(EXE) minigzip$(EXE) examplesh$(EXE) minigzipsh$(EXE) \
+	   example64$(EXE) minigzip64$(EXE) \
 	   libz.* foo.gz so_locations \
 	   _match.s maketree contrib/infback9/*.o
+	rm -rf objs
 
 maintainer-clean: distclean
-distclean: clean
-	cp -p Makefile.in Makefile
-	cp -p zconf.in.h zconf.h
-	rm -f .DS_Store
+distclean: clean zconf docs
+	rm -f Makefile zlib.pc
+	- at rm -f .DS_Store
+	- at printf 'all:\n\t- at echo "Please use ./configure first.  Thank you."\n' > Makefile
+	- at printf '\ndistclean:\n\tmake -f Makefile.in distclean\n' >> Makefile
+	- at touch -r Makefile.in Makefile
 
 tags:
 	etags *.[ch]
@@ -138,17 +236,22 @@
 
 # DO NOT DELETE THIS LINE -- make depend depends on it.
 
-adler32.o: zlib.h zconf.h
-compress.o: zlib.h zconf.h
-crc32.o: crc32.h zlib.h zconf.h
+adler32.o zutil.o: zutil.h zlib.h zconf.h
+gzclose.o gzlib.o gzread.o gzwrite.o: zlib.h zconf.h gzguts.h
+compress.o example.o minigzip.o uncompr.o: zlib.h zconf.h
+crc32.o: zutil.h zlib.h zconf.h crc32.h
 deflate.o: deflate.h zutil.h zlib.h zconf.h
-example.o: zlib.h zconf.h
-gzio.o: zutil.h zlib.h zconf.h
+infback.o inflate.o: zutil.h zlib.h zconf.h inftrees.h inflate.h inffast.h inffixed.h
 inffast.o: zutil.h zlib.h zconf.h inftrees.h inflate.h inffast.h
-inflate.o: zutil.h zlib.h zconf.h inftrees.h inflate.h inffast.h
-infback.o: zutil.h zlib.h zconf.h inftrees.h inflate.h inffast.h
 inftrees.o: zutil.h zlib.h zconf.h inftrees.h
-minigzip.o: zlib.h zconf.h
 trees.o: deflate.h zutil.h zlib.h zconf.h trees.h
-uncompr.o: zlib.h zconf.h
-zutil.o: zutil.h zlib.h zconf.h
+
+adler32.lo zutil.lo: zutil.h zlib.h zconf.h
+gzclose.lo gzlib.lo gzread.lo gzwrite.lo: zlib.h zconf.h gzguts.h
+compress.lo example.lo minigzip.lo uncompr.lo: zlib.h zconf.h
+crc32.lo: zutil.h zlib.h zconf.h crc32.h
+deflate.lo: deflate.h zutil.h zlib.h zconf.h
+infback.lo inflate.lo: zutil.h zlib.h zconf.h inftrees.h inflate.h inffast.h inffixed.h
+inffast.lo: zutil.h zlib.h zconf.h inftrees.h inflate.h inffast.h
+inftrees.lo: zutil.h zlib.h zconf.h inftrees.h
+trees.lo: deflate.h zutil.h zlib.h zconf.h trees.h

Modified: python/branches/py3k/Modules/zlib/README
==============================================================================
--- python/branches/py3k/Modules/zlib/README	(original)
+++ python/branches/py3k/Modules/zlib/README	Fri Jul 30 22:03:17 2010
@@ -1,56 +1,52 @@
 ZLIB DATA COMPRESSION LIBRARY
 
-zlib 1.2.3 is a general purpose data compression library.  All the code is
+zlib 1.2.5 is a general purpose data compression library.  All the code is
 thread safe.  The data format used by the zlib library is described by RFCs
 (Request for Comments) 1950 to 1952 in the files
 http://www.ietf.org/rfc/rfc1950.txt (zlib format), rfc1951.txt (deflate format)
-and rfc1952.txt (gzip format). These documents are also available in other
-formats from ftp://ftp.uu.net/graphics/png/documents/zlib/zdoc-index.html
+and rfc1952.txt (gzip format).
 
 All functions of the compression library are documented in the file zlib.h
-(volunteer to write man pages welcome, contact zlib at gzip.org). A usage example
+(volunteer to write man pages welcome, contact zlib at gzip.org).  A usage example
 of the library is given in the file example.c which also tests that the library
-is working correctly. Another example is given in the file minigzip.c. The
+is working correctly.  Another example is given in the file minigzip.c.  The
 compression library itself is composed of all source files except example.c and
 minigzip.c.
 
 To compile all files and run the test program, follow the instructions given at
-the top of Makefile. In short "make test; make install" should work for most
-machines. For Unix: "./configure; make test; make install". For MSDOS, use one
-of the special makefiles such as Makefile.msc. For VMS, use make_vms.com.
+the top of Makefile.in.  In short "./configure; make test", and if that goes
+well, "make install" should work for most flavors of Unix.  For Windows, use one
+of the special makefiles in win32/ or contrib/vstudio/ .  For VMS, use
+make_vms.com.
 
 Questions about zlib should be sent to <zlib at gzip.org>, or to Gilles Vollant
-<info at winimage.com> for the Windows DLL version. The zlib home page is
-http://www.zlib.org or http://www.gzip.org/zlib/ Before reporting a problem,
-please check this site to verify that you have the latest version of zlib;
-otherwise get the latest version and check whether the problem still exists or
-not.
-
-PLEASE read the zlib FAQ http://www.gzip.org/zlib/zlib_faq.html before asking
-for help.
-
-Mark Nelson <markn at ieee.org> wrote an article about zlib for the Jan. 1997
-issue of  Dr. Dobb's Journal; a copy of the article is available in
-http://dogma.net/markn/articles/zlibtool/zlibtool.htm
-
-The changes made in version 1.2.3 are documented in the file ChangeLog.
-
-Unsupported third party contributions are provided in directory "contrib".
-
-A Java implementation of zlib is available in the Java Development Kit
-http://java.sun.com/j2se/1.4.2/docs/api/java/util/zip/package-summary.html
-See the zlib home page http://www.zlib.org for details.
-
-A Perl interface to zlib written by Paul Marquess <pmqs at cpan.org> is in the
-CPAN (Comprehensive Perl Archive Network) sites
-http://www.cpan.org/modules/by-module/Compress/
+<info at winimage.com> for the Windows DLL version.  The zlib home page is
+http://zlib.net/ .  Before reporting a problem, please check this site to
+verify that you have the latest version of zlib; otherwise get the latest
+version and check whether the problem still exists or not.
+
+PLEASE read the zlib FAQ http://zlib.net/zlib_faq.html before asking for help.
+
+Mark Nelson <markn at ieee.org> wrote an article about zlib for the Jan.  1997
+issue of Dr.  Dobb's Journal; a copy of the article is available at
+http://marknelson.us/1997/01/01/zlib-engine/ .
+
+The changes made in version 1.2.5 are documented in the file ChangeLog.
+
+Unsupported third party contributions are provided in directory contrib/ .
+
+zlib is available in Java using the java.util.zip package, documented at
+http://java.sun.com/developer/technicalArticles/Programming/compression/ .
+
+A Perl interface to zlib written by Paul Marquess <pmqs at cpan.org> is available
+at CPAN (Comprehensive Perl Archive Network) sites, including
+http://search.cpan.org/~pmqs/IO-Compress-Zlib/ .
 
 A Python interface to zlib written by A.M. Kuchling <amk at amk.ca> is
 available in Python 1.5 and later versions, see
-http://www.python.org/doc/lib/module-zlib.html
+http://www.python.org/doc/lib/module-zlib.html .
 
-A zlib binding for TCL written by Andreas Kupries <a.kupries at westend.com> is
-availlable at http://www.oche.de/~akupries/soft/trf/trf_zip.html
+zlib is built into tcl: http://wiki.tcl.tk/4610 .
 
 An experimental package to read and write files in .zip format, written on top
 of zlib by Gilles Vollant <info at winimage.com>, is available in the
@@ -74,25 +70,21 @@
 - zlib doesn't work on HP-UX 9.05 with some versions of /bin/cc. It works with
   other compilers. Use "make test" to check your compiler.
 
-- gzdopen is not supported on BEOS and by some Mac compilers.
+- gzdopen is not supported on RISCOS or BEOS.
 
 - For PalmOs, see http://palmzlib.sourceforge.net/
 
-- When building a shared, i.e. dynamic library on Mac OS X, the library must be
-  installed before testing (do "make install" before "make test"), since the
-  library location is specified in the library.
-
 
 Acknowledgments:
 
-  The deflate format used by zlib was defined by Phil Katz. The deflate
-  and zlib specifications were written by L. Peter Deutsch. Thanks to all the
-  people who reported problems and suggested various improvements in zlib;
-  they are too numerous to cite here.
+  The deflate format used by zlib was defined by Phil Katz.  The deflate and
+  zlib specifications were written by L.  Peter Deutsch.  Thanks to all the
+  people who reported problems and suggested various improvements in zlib; they
+  are too numerous to cite here.
 
 Copyright notice:
 
- (C) 1995-2004 Jean-loup Gailly and Mark Adler
+ (C) 1995-2010 Jean-loup Gailly and Mark Adler
 
   This software is provided 'as-is', without any express or implied
   warranty.  In no event will the authors be held liable for any damages
@@ -113,13 +105,11 @@
   Jean-loup Gailly        Mark Adler
   jloup at gzip.org          madler at alumni.caltech.edu
 
-If you use the zlib library in a product, we would appreciate *not*
-receiving lengthy legal documents to sign. The sources are provided
-for free but without warranty of any kind.  The library has been
-entirely written by Jean-loup Gailly and Mark Adler; it does not
-include third-party code.
-
-If you redistribute modified sources, we would appreciate that you include
-in the file ChangeLog history information documenting your changes. Please
-read the FAQ for more information on the distribution of modified source
-versions.
+If you use the zlib library in a product, we would appreciate *not* receiving
+lengthy legal documents to sign.  The sources are provided for free but without
+warranty of any kind.  The library has been entirely written by Jean-loup
+Gailly and Mark Adler; it does not include third-party code.
+
+If you redistribute modified sources, we would appreciate that you include in
+the file ChangeLog history information documenting your changes.  Please read
+the FAQ for more information on the distribution of modified source versions.

Modified: python/branches/py3k/Modules/zlib/adler32.c
==============================================================================
--- python/branches/py3k/Modules/zlib/adler32.c	(original)
+++ python/branches/py3k/Modules/zlib/adler32.c	Fri Jul 30 22:03:17 2010
@@ -1,12 +1,15 @@
 /* adler32.c -- compute the Adler-32 checksum of a data stream
- * Copyright (C) 1995-2004 Mark Adler
+ * Copyright (C) 1995-2007 Mark Adler
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
 /* @(#) $Id$ */
 
-#define ZLIB_INTERNAL
-#include "zlib.h"
+#include "zutil.h"
+
+#define local static
+
+local uLong adler32_combine_(uLong adler1, uLong adler2, z_off64_t len2);
 
 #define BASE 65521UL    /* largest prime smaller than 65536 */
 #define NMAX 5552
@@ -125,10 +128,10 @@
 }
 
 /* ========================================================================= */
-uLong ZEXPORT adler32_combine(adler1, adler2, len2)
+local uLong adler32_combine_(adler1, adler2, len2)
     uLong adler1;
     uLong adler2;
-    z_off_t len2;
+    z_off64_t len2;
 {
     unsigned long sum1;
     unsigned long sum2;
@@ -141,9 +144,26 @@
     MOD(sum2);
     sum1 += (adler2 & 0xffff) + BASE - 1;
     sum2 += ((adler1 >> 16) & 0xffff) + ((adler2 >> 16) & 0xffff) + BASE - rem;
-    if (sum1 > BASE) sum1 -= BASE;
-    if (sum1 > BASE) sum1 -= BASE;
-    if (sum2 > (BASE << 1)) sum2 -= (BASE << 1);
-    if (sum2 > BASE) sum2 -= BASE;
+    if (sum1 >= BASE) sum1 -= BASE;
+    if (sum1 >= BASE) sum1 -= BASE;
+    if (sum2 >= (BASE << 1)) sum2 -= (BASE << 1);
+    if (sum2 >= BASE) sum2 -= BASE;
     return sum1 | (sum2 << 16);
 }
+
+/* ========================================================================= */
+uLong ZEXPORT adler32_combine(adler1, adler2, len2)
+    uLong adler1;
+    uLong adler2;
+    z_off_t len2;
+{
+    return adler32_combine_(adler1, adler2, len2);
+}
+
+uLong ZEXPORT adler32_combine64(adler1, adler2, len2)
+    uLong adler1;
+    uLong adler2;
+    z_off64_t len2;
+{
+    return adler32_combine_(adler1, adler2, len2);
+}

Modified: python/branches/py3k/Modules/zlib/compress.c
==============================================================================
--- python/branches/py3k/Modules/zlib/compress.c	(original)
+++ python/branches/py3k/Modules/zlib/compress.c	Fri Jul 30 22:03:17 2010
@@ -1,5 +1,5 @@
 /* compress.c -- compress a memory buffer
- * Copyright (C) 1995-2003 Jean-loup Gailly.
+ * Copyright (C) 1995-2005 Jean-loup Gailly.
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -75,5 +75,6 @@
 uLong ZEXPORT compressBound (sourceLen)
     uLong sourceLen;
 {
-    return sourceLen + (sourceLen >> 12) + (sourceLen >> 14) + 11;
+    return sourceLen + (sourceLen >> 12) + (sourceLen >> 14) +
+           (sourceLen >> 25) + 13;
 }

Modified: python/branches/py3k/Modules/zlib/configure
==============================================================================
--- python/branches/py3k/Modules/zlib/configure	(original)
+++ python/branches/py3k/Modules/zlib/configure	Fri Jul 30 22:03:17 2010
@@ -1,37 +1,61 @@
 #!/bin/sh
-# configure script for zlib. This script is needed only if
-# you wish to build a shared library and your system supports them,
-# of if you need special compiler, flags or install directory.
-# Otherwise, you can just use directly "make test; make install"
+# configure script for zlib.
 #
-# To create a shared library, use "configure --shared"; by default a static
-# library is created. If the primitive shared library support provided here
-# does not work, use ftp://prep.ai.mit.edu/pub/gnu/libtool-*.tar.gz
+# Normally configure builds both a static and a shared library.
+# If you want to build just a static library, use: ./configure --static
 #
 # To impose specific compiler or flags or install directory, use for example:
 #    prefix=$HOME CC=cc CFLAGS="-O4" ./configure
 # or for csh/tcsh users:
 #    (setenv prefix $HOME; setenv CC cc; setenv CFLAGS "-O4"; ./configure)
-# LDSHARED is the command to be used to create a shared library
 
 # Incorrect settings of CC or CFLAGS may prevent creating a shared library.
 # If you have problems, try without defining CC and CFLAGS before reporting
 # an error.
 
-LIBS=libz.a
-LDFLAGS="-L. ${LIBS}"
+if [ -n "${CHOST}" ]; then
+    uname="$(echo "${CHOST}" | sed -e 's/^[^-]*-\([^-]*\)$/\1/' -e 's/^[^-]*-[^-]*-\([^-]*\)$/\1/' -e 's/^[^-]*-[^-]*-\([^-]*\)-.*$/\1/')"
+    CROSS_PREFIX="${CHOST}-"
+fi
+
+STATICLIB=libz.a
+LDFLAGS="${LDFLAGS} -L. ${STATICLIB}"
 VER=`sed -n -e '/VERSION "/s/.*"\(.*\)".*/\1/p' < zlib.h`
+VER3=`sed -n -e '/VERSION "/s/.*"\([0-9]*\\.[0-9]*\\.[0-9]*\).*/\1/p' < zlib.h`
 VER2=`sed -n -e '/VERSION "/s/.*"\([0-9]*\\.[0-9]*\)\\..*/\1/p' < zlib.h`
 VER1=`sed -n -e '/VERSION "/s/.*"\([0-9]*\)\\..*/\1/p' < zlib.h`
-AR=${AR-"ar rc"}
-RANLIB=${RANLIB-"ranlib"}
+if "${CROSS_PREFIX}ar" --version >/dev/null 2>/dev/null || test $? -lt 126; then
+    AR=${AR-"${CROSS_PREFIX}ar"}
+    test -n "${CROSS_PREFIX}" && echo Using ${AR}
+else
+    AR=${AR-"ar"}
+    test -n "${CROSS_PREFIX}" && echo Using ${AR}
+fi
+AR_RC="${AR} rc"
+if "${CROSS_PREFIX}ranlib" --version >/dev/null 2>/dev/null || test $? -lt 126; then
+    RANLIB=${RANLIB-"${CROSS_PREFIX}ranlib"}
+    test -n "${CROSS_PREFIX}" && echo Using ${RANLIB}
+else
+    RANLIB=${RANLIB-"ranlib"}
+fi
+if "${CROSS_PREFIX}nm" --version >/dev/null 2>/dev/null || test $? -lt 126; then
+    NM=${NM-"${CROSS_PREFIX}nm"}
+    test -n "${CROSS_PREFIX}" && echo Using ${NM}
+else
+    NM=${NM-"nm"}
+fi
+LDCONFIG=${LDCONFIG-"ldconfig"}
+LDSHAREDLIBC="${LDSHAREDLIBC--lc}"
 prefix=${prefix-/usr/local}
 exec_prefix=${exec_prefix-'${prefix}'}
 libdir=${libdir-'${exec_prefix}/lib'}
+sharedlibdir=${sharedlibdir-'${libdir}'}
 includedir=${includedir-'${prefix}/include'}
 mandir=${mandir-'${prefix}/share/man'}
 shared_ext='.so'
-shared=0
+shared=1
+zprefix=0
+build64=0
 gcc=0
 old_cc="$CC"
 old_cflags="$CFLAGS"
@@ -39,21 +63,29 @@
 while test $# -ge 1
 do
 case "$1" in
-    -h* | --h*)
+    -h* | --help)
       echo 'usage:'
-      echo '  configure [--shared] [--prefix=PREFIX]  [--exec_prefix=EXPREFIX]'
-      echo '     [--libdir=LIBDIR] [--includedir=INCLUDEDIR]'
-        exit 0;;
-    -p*=* | --p*=*) prefix=`echo $1 | sed 's/[-a-z_]*=//'`; shift;;
-    -e*=* | --e*=*) exec_prefix=`echo $1 | sed 's/[-a-z_]*=//'`; shift;;
-    -l*=* | --libdir=*) libdir=`echo $1 | sed 's/[-a-z_]*=//'`; shift;;
-    -i*=* | --includedir=*) includedir=`echo $1 | sed 's/[-a-z_]*=//'`;shift;;
-    -p* | --p*) prefix="$2"; shift; shift;;
-    -e* | --e*) exec_prefix="$2"; shift; shift;;
-    -l* | --l*) libdir="$2"; shift; shift;;
-    -i* | --i*) includedir="$2"; shift; shift;;
-    -s* | --s*) shared=1; shift;;
-    *) echo "unknown option: $1"; echo "$0 --help for help"; exit 1;;
+      echo '  configure [--zprefix] [--prefix=PREFIX]  [--eprefix=EXPREFIX]'
+      echo '    [--static] [--64] [--libdir=LIBDIR] [--sharedlibdir=LIBDIR]'
+      echo '    [--includedir=INCLUDEDIR]'
+        exit 0 ;;
+    -p*=* | --prefix=*) prefix=`echo $1 | sed 's/.*=//'`; shift ;;
+    -e*=* | --eprefix=*) exec_prefix=`echo $1 | sed 's/.*=//'`; shift ;;
+    -l*=* | --libdir=*) libdir=`echo $1 | sed 's/.*=//'`; shift ;;
+    --sharedlibdir=*) sharedlibdir=`echo $1 | sed 's/.*=//'`; shift ;;
+    -i*=* | --includedir=*) includedir=`echo $1 | sed 's/.*=//'`;shift ;;
+    -u*=* | --uname=*) uname=`echo $1 | sed 's/.*=//'`;shift ;;
+    -p* | --prefix) prefix="$2"; shift; shift ;;
+    -e* | --eprefix) exec_prefix="$2"; shift; shift ;;
+    -l* | --libdir) libdir="$2"; shift; shift ;;
+    -i* | --includedir) includedir="$2"; shift; shift ;;
+    -s* | --shared | --enable-shared) shared=1; shift ;;
+    -t | --static) shared=0; shift ;;
+    -z* | --zprefix) zprefix=1; shift ;;
+    -6* | --64) build64=1; shift ;;
+    --sysconfdir=*) echo "ignored option: --sysconfdir"; shift ;;
+    --localstatedir=*) echo "ignored option: --localstatedir"; shift ;;
+    *) echo "unknown option: $1"; echo "$0 --help for help"; exit 1 ;;
     esac
 done
 
@@ -63,46 +95,68 @@
 int hello() {return getchar();}
 EOF
 
-test -z "$CC" && echo Checking for gcc...
-cc=${CC-gcc}
+test -z "$CC" && echo Checking for ${CROSS_PREFIX}gcc...
+cc=${CC-${CROSS_PREFIX}gcc}
 cflags=${CFLAGS-"-O3"}
 # to force the asm version use: CFLAGS="-O3 -DASMV" ./configure
 case "$cc" in
-  *gcc*) gcc=1;;
+  *gcc*) gcc=1 ;;
 esac
 
 if test "$gcc" -eq 1 && ($cc -c $cflags $test.c) 2>/dev/null; then
   CC="$cc"
-  SFLAGS=${CFLAGS-"-fPIC -O3"}
-  CFLAGS="$cflags"
-  case `(uname -s || echo unknown) 2>/dev/null` in
-  Linux | linux | GNU | GNU/*) LDSHARED=${LDSHARED-"$cc -shared -Wl,-soname,libz.so.1"};;
-  CYGWIN* | Cygwin* | cygwin* | OS/2* )
-             EXE='.exe';;
+  SFLAGS="${CFLAGS--O3} -fPIC"
+  CFLAGS="${CFLAGS--O3}"
+  if test $build64 -eq 1; then
+    CFLAGS="${CFLAGS} -m64"
+    SFLAGS="${SFLAGS} -m64"
+  fi
+  if test "${ZLIBGCCWARN}" = "YES"; then
+    CFLAGS="${CFLAGS} -Wall -Wextra -pedantic"
+  fi
+  if test -z "$uname"; then
+    uname=`(uname -s || echo unknown) 2>/dev/null`
+  fi
+  case "$uname" in
+  Linux* | linux* | GNU | GNU/* | *BSD | DragonFly) LDSHARED=${LDSHARED-"$cc -shared -Wl,-soname,libz.so.1,--version-script,zlib.map"} ;;
+  CYGWIN* | Cygwin* | cygwin* | OS/2*)
+        EXE='.exe' ;;
+  MINGW*|mingw*)
+# temporary bypass
+        rm -f $test.[co] $test $test$shared_ext
+        echo "Please use win32/Makefile.gcc instead."
+        exit 1
+        LDSHARED=${LDSHARED-"$cc -shared"}
+        LDSHAREDLIBC=""
+        EXE='.exe' ;;
   QNX*)  # This is for QNX6. I suppose that the QNX rule below is for QNX2,QNX4
          # (alain.bonnefoy at icbt.com)
-                 LDSHARED=${LDSHARED-"$cc -shared -Wl,-hlibz.so.1"};;
+                 LDSHARED=${LDSHARED-"$cc -shared -Wl,-hlibz.so.1"} ;;
   HP-UX*)
          LDSHARED=${LDSHARED-"$cc -shared $SFLAGS"}
          case `(uname -m || echo unknown) 2>/dev/null` in
          ia64)
                  shared_ext='.so'
-                 SHAREDLIB='libz.so';;
+                 SHAREDLIB='libz.so' ;;
          *)
                  shared_ext='.sl'
-                 SHAREDLIB='libz.sl';;
-         esac;;
+                 SHAREDLIB='libz.sl' ;;
+         esac ;;
   Darwin*)   shared_ext='.dylib'
              SHAREDLIB=libz$shared_ext
              SHAREDLIBV=libz.$VER$shared_ext
              SHAREDLIBM=libz.$VER1$shared_ext
-             LDSHARED=${LDSHARED-"$cc -dynamiclib -install_name $libdir/$SHAREDLIBM -compatibility_version $VER1 -current_version $VER"};;
-  *)             LDSHARED=${LDSHARED-"$cc -shared"};;
+             LDSHARED=${LDSHARED-"$cc -dynamiclib -install_name $libdir/$SHAREDLIBM -compatibility_version $VER1 -current_version $VER3"} ;;
+  *)             LDSHARED=${LDSHARED-"$cc -shared"} ;;
   esac
 else
   # find system name and corresponding cc options
   CC=${CC-cc}
-  case `(uname -sr || echo unknown) 2>/dev/null` in
+  gcc=0
+  if test -z "$uname"; then
+    uname=`(uname -sr || echo unknown) 2>/dev/null`
+  fi
+  case "$uname" in
   HP-UX*)    SFLAGS=${CFLAGS-"-O +z"}
              CFLAGS=${CFLAGS-"-O"}
 #            LDSHARED=${LDSHARED-"ld -b +vnocompatwarnings"}
@@ -110,57 +164,64 @@
          case `(uname -m || echo unknown) 2>/dev/null` in
          ia64)
              shared_ext='.so'
-             SHAREDLIB='libz.so';;
+             SHAREDLIB='libz.so' ;;
          *)
              shared_ext='.sl'
-             SHAREDLIB='libz.sl';;
-         esac;;
+             SHAREDLIB='libz.sl' ;;
+         esac ;;
   IRIX*)     SFLAGS=${CFLAGS-"-ansi -O2 -rpath ."}
              CFLAGS=${CFLAGS-"-ansi -O2"}
-             LDSHARED=${LDSHARED-"cc -shared"};;
+             LDSHARED=${LDSHARED-"cc -shared -Wl,-soname,libz.so.1"} ;;
   OSF1\ V4*) SFLAGS=${CFLAGS-"-O -std1"}
              CFLAGS=${CFLAGS-"-O -std1"}
-             LDSHARED=${LDSHARED-"cc -shared  -Wl,-soname,libz.so -Wl,-msym -Wl,-rpath,$(libdir) -Wl,-set_version,${VER}:1.0"};;
+             LDFLAGS="${LDFLAGS} -Wl,-rpath,."
+             LDSHARED=${LDSHARED-"cc -shared  -Wl,-soname,libz.so -Wl,-msym -Wl,-rpath,$(libdir) -Wl,-set_version,${VER}:1.0"} ;;
   OSF1*)     SFLAGS=${CFLAGS-"-O -std1"}
              CFLAGS=${CFLAGS-"-O -std1"}
-             LDSHARED=${LDSHARED-"cc -shared"};;
+             LDSHARED=${LDSHARED-"cc -shared -Wl,-soname,libz.so.1"} ;;
   QNX*)      SFLAGS=${CFLAGS-"-4 -O"}
              CFLAGS=${CFLAGS-"-4 -O"}
              LDSHARED=${LDSHARED-"cc"}
              RANLIB=${RANLIB-"true"}
-             AR="cc -A";;
+             AR_RC="cc -A" ;;
   SCO_SV\ 3.2*) SFLAGS=${CFLAGS-"-O3 -dy -KPIC "}
              CFLAGS=${CFLAGS-"-O3"}
-             LDSHARED=${LDSHARED-"cc -dy -KPIC -G"};;
-  SunOS\ 5*) SFLAGS=${CFLAGS-"-fast -xcg89 -KPIC -R."}
-             CFLAGS=${CFLAGS-"-fast -xcg89"}
-             LDSHARED=${LDSHARED-"cc -G"};;
+             LDSHARED=${LDSHARED-"cc -dy -KPIC -G"} ;;
+  SunOS\ 5*) LDSHARED=${LDSHARED-"cc -G"}
+         case `(uname -m || echo unknown) 2>/dev/null` in
+         i86*)
+             SFLAGS=${CFLAGS-"-xpentium -fast -KPIC -R."}
+             CFLAGS=${CFLAGS-"-xpentium -fast"} ;;
+         *)
+             SFLAGS=${CFLAGS-"-fast -xcg92 -KPIC -R."}
+             CFLAGS=${CFLAGS-"-fast -xcg92"} ;;
+         esac ;;
   SunOS\ 4*) SFLAGS=${CFLAGS-"-O2 -PIC"}
              CFLAGS=${CFLAGS-"-O2"}
-             LDSHARED=${LDSHARED-"ld"};;
-  SunStudio\ 9*) SFLAGS=${CFLAGS-"-DUSE_MMAP -fast -xcode=pic32 -xtarget=ultra3 -xarch=v9b"}
-             CFLAGS=${CFLAGS-"-DUSE_MMAP -fast -xtarget=ultra3 -xarch=v9b"}
-             LDSHARED=${LDSHARED-"cc -xarch=v9b"};;
+             LDSHARED=${LDSHARED-"ld"} ;;
+  SunStudio\ 9*) SFLAGS=${CFLAGS-"-fast -xcode=pic32 -xtarget=ultra3 -xarch=v9b"}
+             CFLAGS=${CFLAGS-"-fast -xtarget=ultra3 -xarch=v9b"}
+             LDSHARED=${LDSHARED-"cc -xarch=v9b"} ;;
   UNIX_System_V\ 4.2.0)
              SFLAGS=${CFLAGS-"-KPIC -O"}
              CFLAGS=${CFLAGS-"-O"}
-             LDSHARED=${LDSHARED-"cc -G"};;
+             LDSHARED=${LDSHARED-"cc -G"} ;;
   UNIX_SV\ 4.2MP)
              SFLAGS=${CFLAGS-"-Kconform_pic -O"}
              CFLAGS=${CFLAGS-"-O"}
-             LDSHARED=${LDSHARED-"cc -G"};;
+             LDSHARED=${LDSHARED-"cc -G"} ;;
   OpenUNIX\ 5)
              SFLAGS=${CFLAGS-"-KPIC -O"}
              CFLAGS=${CFLAGS-"-O"}
-             LDSHARED=${LDSHARED-"cc -G"};;
+             LDSHARED=${LDSHARED-"cc -G"} ;;
   AIX*)  # Courtesy of dbakker at arrayasolutions.com
              SFLAGS=${CFLAGS-"-O -qmaxmem=8192"}
              CFLAGS=${CFLAGS-"-O -qmaxmem=8192"}
-             LDSHARED=${LDSHARED-"xlc -G"};;
-  # send working options for other systems to support at gzip.org
+             LDSHARED=${LDSHARED-"xlc -G"} ;;
+  # send working options for other systems to zlib at gzip.org
   *)         SFLAGS=${CFLAGS-"-O"}
              CFLAGS=${CFLAGS-"-O"}
-             LDSHARED=${LDSHARED-"cc -shared"};;
+             LDSHARED=${LDSHARED-"cc -shared"} ;;
   esac
 fi
 
@@ -171,38 +232,83 @@
 if test $shared -eq 1; then
   echo Checking for shared library support...
   # we must test in two steps (cc then ld), required at least on SunOS 4.x
-  if test "`($CC -c $SFLAGS $test.c) 2>&1`" = "" &&
-     test "`($LDSHARED -o $test$shared_ext $test.o) 2>&1`" = ""; then
-    CFLAGS="$SFLAGS"
-    LIBS="$SHAREDLIBV"
+  if test "`($CC -w -c $SFLAGS $test.c) 2>&1`" = "" &&
+     test "`($LDSHARED $SFLAGS -o $test$shared_ext $test.o) 2>&1`" = ""; then
     echo Building shared library $SHAREDLIBV with $CC.
   elif test -z "$old_cc" -a -z "$old_cflags"; then
     echo No shared library support.
     shared=0;
   else
+    echo Tested $CC -w -c $SFLAGS $test.c
+    $CC -w -c $SFLAGS $test.c
+    echo Tested $LDSHARED $SFLAGS -o $test$shared_ext $test.o
+    $LDSHARED $SFLAGS -o $test$shared_ext $test.o
     echo 'No shared library support; try without defining CC and CFLAGS'
     shared=0;
   fi
 fi
 if test $shared -eq 0; then
   LDSHARED="$CC"
-  echo Building static library $LIBS version $VER with $CC.
+  ALL="static"
+  TEST="all teststatic"
+  SHAREDLIB=""
+  SHAREDLIBV=""
+  SHAREDLIBM=""
+  echo Building static library $STATICLIB version $VER with $CC.
 else
-  LDFLAGS="-L. ${SHAREDLIBV}"
+  ALL="static shared"
+  TEST="all teststatic testshared"
 fi
 
 cat > $test.c <<EOF
+#include <sys/types.h>
+off64_t dummy = 0;
+EOF
+if test "`($CC -c $CFLAGS -D_LARGEFILE64_SOURCE=1 $test.c) 2>&1`" = ""; then
+  CFLAGS="${CFLAGS} -D_LARGEFILE64_SOURCE=1"
+  SFLAGS="${SFLAGS} -D_LARGEFILE64_SOURCE=1"
+  ALL="${ALL} all64"
+  TEST="${TEST} test64"
+  echo "Checking for off64_t... Yes."
+  echo "Checking for fseeko... Yes."
+else
+  echo "Checking for off64_t... No."
+  cat > $test.c <<EOF
+#include <stdio.h>
+int main(void) {
+  fseeko(NULL, 0, 0);
+  return 0;
+}
+EOF
+  if test "`($CC $CFLAGS -o $test $test.c) 2>&1`" = ""; then
+    echo "Checking for fseeko... Yes."
+  else
+    CFLAGS="${CFLAGS} -DNO_FSEEKO"
+    SFLAGS="${SFLAGS} -DNO_FSEEKO"
+    echo "Checking for fseeko... No."
+  fi
+fi
+
+cp -p zconf.h.in zconf.h
+
+cat > $test.c <<EOF
 #include <unistd.h>
 int main() { return 0; }
 EOF
 if test "`($CC -c $CFLAGS $test.c) 2>&1`" = ""; then
-  sed < zconf.in.h "/HAVE_UNISTD_H/s%0%1%" > zconf.h
+  sed < zconf.h "/^#ifdef HAVE_UNISTD_H.* may be/s/def HAVE_UNISTD_H\(.*\) may be/ 1\1 was/" > zconf.temp.h
+  mv zconf.temp.h zconf.h
   echo "Checking for unistd.h... Yes."
 else
-  cp -p zconf.in.h zconf.h
   echo "Checking for unistd.h... No."
 fi
 
+if test $zprefix -eq 1; then
+  sed < zconf.h "/#ifdef Z_PREFIX.* may be/s/def Z_PREFIX\(.*\) may be/ 1\1 was/" > zconf.temp.h
+  mv zconf.temp.h zconf.h
+  echo "Using z_ prefix on all symbols."
+fi
+
 cat > $test.c <<EOF
 #include <stdio.h>
 #include <stdarg.h>
@@ -219,13 +325,13 @@
 EOF
 
 if test "`($CC -c $CFLAGS $test.c) 2>&1`" = ""; then
-  echo "Checking whether to use vs[n]printf() or s[n]printf()... using vs[n]printf()"
+  echo "Checking whether to use vs[n]printf() or s[n]printf()... using vs[n]printf()."
 
   cat > $test.c <<EOF
 #include <stdio.h>
 #include <stdarg.h>
 
-int mytest(char *fmt, ...)
+int mytest(const char *fmt, ...)
 {
   char buf[20];
   va_list ap;
@@ -249,7 +355,7 @@
 #include <stdio.h>
 #include <stdarg.h>
 
-int mytest(char *fmt, ...)
+int mytest(const char *fmt, ...)
 {
   int n;
   char buf[20];
@@ -271,6 +377,7 @@
       echo "Checking for return value of vsnprintf()... Yes."
     else
       CFLAGS="$CFLAGS -DHAS_vsnprintf_void"
+      SFLAGS="$SFLAGS -DHAS_vsnprintf_void"
       echo "Checking for return value of vsnprintf()... No."
       echo "  WARNING: apparently vsnprintf() does not return a value. zlib"
       echo "  can build but will be open to possible string-format security"
@@ -278,6 +385,7 @@
     fi
   else
     CFLAGS="$CFLAGS -DNO_vsnprintf"
+    SFLAGS="$SFLAGS -DNO_vsnprintf"
     echo "Checking for vsnprintf() in stdio.h... No."
     echo "  WARNING: vsnprintf() not found, falling back to vsprintf(). zlib"
     echo "  can build but will be open to possible buffer-overflow security"
@@ -287,7 +395,7 @@
 #include <stdio.h>
 #include <stdarg.h>
 
-int mytest(char *fmt, ...)
+int mytest(const char *fmt, ...)
 {
   int n;
   char buf[20];
@@ -309,6 +417,7 @@
       echo "Checking for return value of vsprintf()... Yes."
     else
       CFLAGS="$CFLAGS -DHAS_vsprintf_void"
+      SFLAGS="$SFLAGS -DHAS_vsprintf_void"
       echo "Checking for return value of vsprintf()... No."
       echo "  WARNING: apparently vsprintf() does not return a value. zlib"
       echo "  can build but will be open to possible string-format security"
@@ -316,7 +425,7 @@
     fi
   fi
 else
-  echo "Checking whether to use vs[n]printf() or s[n]printf()... using s[n]printf()"
+  echo "Checking whether to use vs[n]printf() or s[n]printf()... using s[n]printf()."
 
   cat >$test.c <<EOF
 #include <stdio.h>
@@ -358,6 +467,7 @@
       echo "Checking for return value of snprintf()... Yes."
     else
       CFLAGS="$CFLAGS -DHAS_snprintf_void"
+      SFLAGS="$SFLAGS -DHAS_snprintf_void"
       echo "Checking for return value of snprintf()... No."
       echo "  WARNING: apparently snprintf() does not return a value. zlib"
       echo "  can build but will be open to possible string-format security"
@@ -365,6 +475,7 @@
     fi
   else
     CFLAGS="$CFLAGS -DNO_snprintf"
+    SFLAGS="$SFLAGS -DNO_snprintf"
     echo "Checking for snprintf() in stdio.h... No."
     echo "  WARNING: snprintf() not found, falling back to sprintf(). zlib"
     echo "  can build but will be open to possible buffer-overflow security"
@@ -390,6 +501,7 @@
       echo "Checking for return value of sprintf()... Yes."
     else
       CFLAGS="$CFLAGS -DHAS_sprintf_void"
+      SFLAGS="$SFLAGS -DHAS_sprintf_void"
       echo "Checking for return value of sprintf()... No."
       echo "  WARNING: apparently sprintf() does not return a value. zlib"
       echo "  can build but will be open to possible string-format security"
@@ -398,41 +510,37 @@
   fi
 fi
 
-cat >$test.c <<EOF
-#include <errno.h>
-int main() { return 0; }
-EOF
-if test "`($CC -c $CFLAGS $test.c) 2>&1`" = ""; then
-  echo "Checking for errno.h... Yes."
-else
-  echo "Checking for errno.h... No."
-  CFLAGS="$CFLAGS -DNO_ERRNO_H"
-fi
-
-cat > $test.c <<EOF
-#include <sys/types.h>
-#include <sys/mman.h>
-#include <sys/stat.h>
-caddr_t hello() {
-  return mmap((caddr_t)0, (off_t)0, PROT_READ, MAP_SHARED, 0, (off_t)0);
+if test "$gcc" -eq 1; then
+  cat > $test.c <<EOF
+#if ((__GNUC__-0) * 10 + __GNUC_MINOR__-0 >= 33)
+#  define ZLIB_INTERNAL __attribute__((visibility ("hidden")))
+#else
+#  define ZLIB_INTERNAL
+#endif
+int ZLIB_INTERNAL foo;
+int main()
+{
+  return 0;
 }
 EOF
-if test "`($CC -c $CFLAGS $test.c) 2>&1`" = ""; then
-  CFLAGS="$CFLAGS -DUSE_MMAP"
-  echo Checking for mmap support... Yes.
-else
-  echo Checking for mmap support... No.
+  if test "`($CC -c $CFLAGS $test.c) 2>&1`" = ""; then
+    echo "Checking for attribute(visibility) support... Yes."
+  else
+    CFLAGS="$CFLAGS -DNO_VIZ"
+    SFLAGS="$SFLAGS -DNO_VIZ"
+    echo "Checking for attribute(visibility) support... No."
+  fi
 fi
 
 CPP=${CPP-"$CC -E"}
 case $CFLAGS in
   *ASMV*)
-    if test "`nm $test.o | grep _hello`" = ""; then
+    if test "`$NM $test.o | grep _hello`" = ""; then
       CPP="$CPP -DNO_UNDERLINE"
       echo Checking for underline in external names... No.
     else
       echo Checking for underline in external names... Yes.
-    fi;;
+    fi ;;
 esac
 
 rm -f $test.[co] $test $test$shared_ext
@@ -441,19 +549,48 @@
 sed < Makefile.in "
 /^CC *=/s#=.*#=$CC#
 /^CFLAGS *=/s#=.*#=$CFLAGS#
+/^SFLAGS *=/s#=.*#=$SFLAGS#
+/^LDFLAGS *=/s#=.*#=$LDFLAGS#
+/^LDSHARED *=/s#=.*#=$LDSHARED#
+/^CPP *=/s#=.*#=$CPP#
+/^STATICLIB *=/s#=.*#=$STATICLIB#
+/^SHAREDLIB *=/s#=.*#=$SHAREDLIB#
+/^SHAREDLIBV *=/s#=.*#=$SHAREDLIBV#
+/^SHAREDLIBM *=/s#=.*#=$SHAREDLIBM#
+/^AR *=/s#=.*#=$AR_RC#
+/^RANLIB *=/s#=.*#=$RANLIB#
+/^LDCONFIG *=/s#=.*#=$LDCONFIG#
+/^LDSHAREDLIBC *=/s#=.*#=$LDSHAREDLIBC#
+/^EXE *=/s#=.*#=$EXE#
+/^prefix *=/s#=.*#=$prefix#
+/^exec_prefix *=/s#=.*#=$exec_prefix#
+/^libdir *=/s#=.*#=$libdir#
+/^sharedlibdir *=/s#=.*#=$sharedlibdir#
+/^includedir *=/s#=.*#=$includedir#
+/^mandir *=/s#=.*#=$mandir#
+/^all: */s#:.*#: $ALL#
+/^test: */s#:.*#: $TEST#
+" > Makefile
+
+sed < zlib.pc.in "
+/^CC *=/s#=.*#=$CC#
+/^CFLAGS *=/s#=.*#=$CFLAGS#
 /^CPP *=/s#=.*#=$CPP#
 /^LDSHARED *=/s#=.*#=$LDSHARED#
-/^LIBS *=/s#=.*#=$LIBS#
+/^STATICLIB *=/s#=.*#=$STATICLIB#
 /^SHAREDLIB *=/s#=.*#=$SHAREDLIB#
 /^SHAREDLIBV *=/s#=.*#=$SHAREDLIBV#
 /^SHAREDLIBM *=/s#=.*#=$SHAREDLIBM#
-/^AR *=/s#=.*#=$AR#
+/^AR *=/s#=.*#=$AR_RC#
 /^RANLIB *=/s#=.*#=$RANLIB#
 /^EXE *=/s#=.*#=$EXE#
 /^prefix *=/s#=.*#=$prefix#
 /^exec_prefix *=/s#=.*#=$exec_prefix#
 /^libdir *=/s#=.*#=$libdir#
+/^sharedlibdir *=/s#=.*#=$sharedlibdir#
 /^includedir *=/s#=.*#=$includedir#
 /^mandir *=/s#=.*#=$mandir#
 /^LDFLAGS *=/s#=.*#=$LDFLAGS#
-" > Makefile
+" | sed -e "
+s/\@VERSION\@/$VER/g;
+" > zlib.pc

Modified: python/branches/py3k/Modules/zlib/crc32.c
==============================================================================
--- python/branches/py3k/Modules/zlib/crc32.c	(original)
+++ python/branches/py3k/Modules/zlib/crc32.c	Fri Jul 30 22:03:17 2010
@@ -1,5 +1,5 @@
 /* crc32.c -- compute the CRC-32 of a data stream
- * Copyright (C) 1995-2005 Mark Adler
+ * Copyright (C) 1995-2006, 2010 Mark Adler
  * For conditions of distribution and use, see copyright notice in zlib.h
  *
  * Thanks to Rodney Brown <rbrown64 at csc.com.au> for his contribution of faster
@@ -53,7 +53,7 @@
 
 /* Definitions for doing the crc four data bytes at a time. */
 #ifdef BYFOUR
-#  define REV(w) (((w)>>24)+(((w)>>8)&0xff00)+ \
+#  define REV(w) ((((w)>>24)&0xff)+(((w)>>8)&0xff00)+ \
                 (((w)&0xff00)<<8)+(((w)&0xff)<<24))
    local unsigned long crc32_little OF((unsigned long,
                         const unsigned char FAR *, unsigned));
@@ -68,6 +68,8 @@
 local unsigned long gf2_matrix_times OF((unsigned long *mat,
                                          unsigned long vec));
 local void gf2_matrix_square OF((unsigned long *square, unsigned long *mat));
+local uLong crc32_combine_(uLong crc1, uLong crc2, z_off64_t len2);
+
 
 #ifdef DYNAMIC_CRC_TABLE
 
@@ -219,7 +221,7 @@
 unsigned long ZEXPORT crc32(crc, buf, len)
     unsigned long crc;
     const unsigned char FAR *buf;
-    unsigned len;
+    uInt len;
 {
     if (buf == Z_NULL) return 0UL;
 
@@ -367,22 +369,22 @@
 }
 
 /* ========================================================================= */
-uLong ZEXPORT crc32_combine(crc1, crc2, len2)
+local uLong crc32_combine_(crc1, crc2, len2)
     uLong crc1;
     uLong crc2;
-    z_off_t len2;
+    z_off64_t len2;
 {
     int n;
     unsigned long row;
     unsigned long even[GF2_DIM];    /* even-power-of-two zeros operator */
     unsigned long odd[GF2_DIM];     /* odd-power-of-two zeros operator */
 
-    /* degenerate case */
-    if (len2 == 0)
+    /* degenerate case (also disallow negative lengths) */
+    if (len2 <= 0)
         return crc1;
 
     /* put operator for one zero bit in odd */
-    odd[0] = 0xedb88320L;           /* CRC-32 polynomial */
+    odd[0] = 0xedb88320UL;          /* CRC-32 polynomial */
     row = 1;
     for (n = 1; n < GF2_DIM; n++) {
         odd[n] = row;
@@ -421,3 +423,20 @@
     crc1 ^= crc2;
     return crc1;
 }
+
+/* ========================================================================= */
+uLong ZEXPORT crc32_combine(crc1, crc2, len2)
+    uLong crc1;
+    uLong crc2;
+    z_off_t len2;
+{
+    return crc32_combine_(crc1, crc2, len2);
+}
+
+uLong ZEXPORT crc32_combine64(crc1, crc2, len2)
+    uLong crc1;
+    uLong crc2;
+    z_off64_t len2;
+{
+    return crc32_combine_(crc1, crc2, len2);
+}

Modified: python/branches/py3k/Modules/zlib/deflate.c
==============================================================================
--- python/branches/py3k/Modules/zlib/deflate.c	(original)
+++ python/branches/py3k/Modules/zlib/deflate.c	Fri Jul 30 22:03:17 2010
@@ -1,5 +1,5 @@
 /* deflate.c -- compress data using the deflation algorithm
- * Copyright (C) 1995-2005 Jean-loup Gailly.
+ * Copyright (C) 1995-2010 Jean-loup Gailly and Mark Adler
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -52,7 +52,7 @@
 #include "deflate.h"
 
 const char deflate_copyright[] =
-   " deflate 1.2.3 Copyright 1995-2005 Jean-loup Gailly ";
+   " deflate 1.2.5 Copyright 1995-2010 Jean-loup Gailly and Mark Adler ";
 /*
   If you use the zlib library in a product, an acknowledgment is welcome
   in the documentation of your product. If for some reason you cannot
@@ -79,19 +79,18 @@
 #ifndef FASTEST
 local block_state deflate_slow   OF((deflate_state *s, int flush));
 #endif
+local block_state deflate_rle    OF((deflate_state *s, int flush));
+local block_state deflate_huff   OF((deflate_state *s, int flush));
 local void lm_init        OF((deflate_state *s));
 local void putShortMSB    OF((deflate_state *s, uInt b));
 local void flush_pending  OF((z_streamp strm));
 local int read_buf        OF((z_streamp strm, Bytef *buf, unsigned size));
-#ifndef FASTEST
 #ifdef ASMV
       void match_init OF((void)); /* asm code initialization */
       uInt longest_match  OF((deflate_state *s, IPos cur_match));
 #else
 local uInt longest_match  OF((deflate_state *s, IPos cur_match));
 #endif
-#endif
-local uInt longest_match_fast OF((deflate_state *s, IPos cur_match));
 
 #ifdef DEBUG
 local  void check_match OF((deflate_state *s, IPos start, IPos match,
@@ -110,11 +109,6 @@
 #endif
 /* Matches of length 3 are discarded if their distance exceeds TOO_FAR */
 
-#define MIN_LOOKAHEAD (MAX_MATCH+MIN_MATCH+1)
-/* Minimum amount of lookahead, except at the end of the input file.
- * See deflate.c for comments about the MIN_MATCH+1.
- */
-
 /* Values for max_lazy_match, good_match and max_chain_length, depending on
  * the desired pack level (0..9). The values given below have been tuned to
  * exclude worst case performance for pathological files. Better values may be
@@ -288,6 +282,8 @@
     s->prev   = (Posf *)  ZALLOC(strm, s->w_size, sizeof(Pos));
     s->head   = (Posf *)  ZALLOC(strm, s->hash_size, sizeof(Pos));
 
+    s->high_water = 0;      /* nothing written to s->window yet */
+
     s->lit_bufsize = 1 << (memLevel + 6); /* 16K elements by default */
 
     overlay = (ushf *) ZALLOC(strm, s->lit_bufsize, sizeof(ush)+2);
@@ -332,8 +328,8 @@
         strm->adler = adler32(strm->adler, dictionary, dictLength);
 
     if (length < MIN_MATCH) return Z_OK;
-    if (length > MAX_DIST(s)) {
-        length = MAX_DIST(s);
+    if (length > s->w_size) {
+        length = s->w_size;
         dictionary += dictLength - length; /* use the tail of the dictionary */
     }
     zmemcpy(s->window, dictionary, length);
@@ -435,9 +431,10 @@
     }
     func = configuration_table[s->level].func;
 
-    if (func != configuration_table[level].func && strm->total_in != 0) {
+    if ((strategy != s->strategy || func != configuration_table[level].func) &&
+        strm->total_in != 0) {
         /* Flush the last buffer: */
-        err = deflate(strm, Z_PARTIAL_FLUSH);
+        err = deflate(strm, Z_BLOCK);
     }
     if (s->level != level) {
         s->level = level;
@@ -481,33 +478,66 @@
  * resulting from using fixed blocks instead of stored blocks, which deflate
  * can emit on compressed data for some combinations of the parameters.
  *
- * This function could be more sophisticated to provide closer upper bounds
- * for every combination of windowBits and memLevel, as well as wrap.
- * But even the conservative upper bound of about 14% expansion does not
- * seem onerous for output buffer allocation.
+ * This function could be more sophisticated to provide closer upper bounds for
+ * every combination of windowBits and memLevel.  But even the conservative
+ * upper bound of about 14% expansion does not seem onerous for output buffer
+ * allocation.
  */
 uLong ZEXPORT deflateBound(strm, sourceLen)
     z_streamp strm;
     uLong sourceLen;
 {
     deflate_state *s;
-    uLong destLen;
+    uLong complen, wraplen;
+    Bytef *str;
 
-    /* conservative upper bound */
-    destLen = sourceLen +
-              ((sourceLen + 7) >> 3) + ((sourceLen + 63) >> 6) + 11;
+    /* conservative upper bound for compressed data */
+    complen = sourceLen +
+              ((sourceLen + 7) >> 3) + ((sourceLen + 63) >> 6) + 5;
 
-    /* if can't get parameters, return conservative bound */
+    /* if can't get parameters, return conservative bound plus zlib wrapper */
     if (strm == Z_NULL || strm->state == Z_NULL)
-        return destLen;
+        return complen + 6;
 
-    /* if not default parameters, return conservative bound */
+    /* compute wrapper length */
     s = strm->state;
+    switch (s->wrap) {
+    case 0:                                 /* raw deflate */
+        wraplen = 0;
+        break;
+    case 1:                                 /* zlib wrapper */
+        wraplen = 6 + (s->strstart ? 4 : 0);
+        break;
+    case 2:                                 /* gzip wrapper */
+        wraplen = 18;
+        if (s->gzhead != Z_NULL) {          /* user-supplied gzip header */
+            if (s->gzhead->extra != Z_NULL)
+                wraplen += 2 + s->gzhead->extra_len;
+            str = s->gzhead->name;
+            if (str != Z_NULL)
+                do {
+                    wraplen++;
+                } while (*str++);
+            str = s->gzhead->comment;
+            if (str != Z_NULL)
+                do {
+                    wraplen++;
+                } while (*str++);
+            if (s->gzhead->hcrc)
+                wraplen += 2;
+        }
+        break;
+    default:                                /* for compiler happiness */
+        wraplen = 6;
+    }
+
+    /* if not default parameters, return conservative bound */
     if (s->w_bits != 15 || s->hash_bits != 8 + 7)
-        return destLen;
+        return complen + wraplen;
 
     /* default settings: return tight bound for that case */
-    return compressBound(sourceLen);
+    return sourceLen + (sourceLen >> 12) + (sourceLen >> 14) +
+           (sourceLen >> 25) + 13 - 6 + wraplen;
 }
 
 /* =========================================================================
@@ -557,7 +587,7 @@
     deflate_state *s;
 
     if (strm == Z_NULL || strm->state == Z_NULL ||
-        flush > Z_FINISH || flush < 0) {
+        flush > Z_BLOCK || flush < 0) {
         return Z_STREAM_ERROR;
     }
     s = strm->state;
@@ -581,7 +611,7 @@
             put_byte(s, 31);
             put_byte(s, 139);
             put_byte(s, 8);
-            if (s->gzhead == NULL) {
+            if (s->gzhead == Z_NULL) {
                 put_byte(s, 0);
                 put_byte(s, 0);
                 put_byte(s, 0);
@@ -608,7 +638,7 @@
                             (s->strategy >= Z_HUFFMAN_ONLY || s->level < 2 ?
                              4 : 0));
                 put_byte(s, s->gzhead->os & 0xff);
-                if (s->gzhead->extra != NULL) {
+                if (s->gzhead->extra != Z_NULL) {
                     put_byte(s, s->gzhead->extra_len & 0xff);
                     put_byte(s, (s->gzhead->extra_len >> 8) & 0xff);
                 }
@@ -650,7 +680,7 @@
     }
 #ifdef GZIP
     if (s->status == EXTRA_STATE) {
-        if (s->gzhead->extra != NULL) {
+        if (s->gzhead->extra != Z_NULL) {
             uInt beg = s->pending;  /* start of bytes to update crc */
 
             while (s->gzindex < (s->gzhead->extra_len & 0xffff)) {
@@ -678,7 +708,7 @@
             s->status = NAME_STATE;
     }
     if (s->status == NAME_STATE) {
-        if (s->gzhead->name != NULL) {
+        if (s->gzhead->name != Z_NULL) {
             uInt beg = s->pending;  /* start of bytes to update crc */
             int val;
 
@@ -709,7 +739,7 @@
             s->status = COMMENT_STATE;
     }
     if (s->status == COMMENT_STATE) {
-        if (s->gzhead->comment != NULL) {
+        if (s->gzhead->comment != Z_NULL) {
             uInt beg = s->pending;  /* start of bytes to update crc */
             int val;
 
@@ -787,7 +817,9 @@
         (flush != Z_NO_FLUSH && s->status != FINISH_STATE)) {
         block_state bstate;
 
-        bstate = (*(configuration_table[s->level].func))(s, flush);
+        bstate = s->strategy == Z_HUFFMAN_ONLY ? deflate_huff(s, flush) :
+                    (s->strategy == Z_RLE ? deflate_rle(s, flush) :
+                        (*(configuration_table[s->level].func))(s, flush));
 
         if (bstate == finish_started || bstate == finish_done) {
             s->status = FINISH_STATE;
@@ -808,13 +840,17 @@
         if (bstate == block_done) {
             if (flush == Z_PARTIAL_FLUSH) {
                 _tr_align(s);
-            } else { /* FULL_FLUSH or SYNC_FLUSH */
+            } else if (flush != Z_BLOCK) { /* FULL_FLUSH or SYNC_FLUSH */
                 _tr_stored_block(s, (char*)0, 0L, 0);
                 /* For a full flush, this empty block will be recognized
                  * as a special marker by inflate_sync().
                  */
                 if (flush == Z_FULL_FLUSH) {
                     CLEAR_HASH(s);             /* forget history */
+                    if (s->lookahead == 0) {
+                        s->strstart = 0;
+                        s->block_start = 0L;
+                    }
                 }
             }
             flush_pending(strm);
@@ -1167,12 +1203,13 @@
     return s->lookahead;
 }
 #endif /* ASMV */
-#endif /* FASTEST */
+
+#else /* FASTEST */
 
 /* ---------------------------------------------------------------------------
- * Optimized version for level == 1 or strategy == Z_RLE only
+ * Optimized version for FASTEST only
  */
-local uInt longest_match_fast(s, cur_match)
+local uInt longest_match(s, cur_match)
     deflate_state *s;
     IPos cur_match;                             /* current match */
 {
@@ -1225,6 +1262,8 @@
     return (uInt)len <= s->lookahead ? (uInt)len : s->lookahead;
 }
 
+#endif /* FASTEST */
+
 #ifdef DEBUG
 /* ===========================================================================
  * Check that the match at match_start is indeed a match.
@@ -1303,7 +1342,6 @@
                later. (Using level 0 permanently is not an optimal usage of
                zlib, so we don't care about this pathological case.)
              */
-            /* %%% avoid this when Z_RLE */
             n = s->hash_size;
             p = &s->head[n];
             do {
@@ -1355,27 +1393,61 @@
          */
 
     } while (s->lookahead < MIN_LOOKAHEAD && s->strm->avail_in != 0);
+
+    /* If the WIN_INIT bytes after the end of the current data have never been
+     * written, then zero those bytes in order to avoid memory check reports of
+     * the use of uninitialized (or uninitialised as Julian writes) bytes by
+     * the longest match routines.  Update the high water mark for the next
+     * time through here.  WIN_INIT is set to MAX_MATCH since the longest match
+     * routines allow scanning to strstart + MAX_MATCH, ignoring lookahead.
+     */
+    if (s->high_water < s->window_size) {
+        ulg curr = s->strstart + (ulg)(s->lookahead);
+        ulg init;
+
+        if (s->high_water < curr) {
+            /* Previous high water mark below current data -- zero WIN_INIT
+             * bytes or up to end of window, whichever is less.
+             */
+            init = s->window_size - curr;
+            if (init > WIN_INIT)
+                init = WIN_INIT;
+            zmemzero(s->window + curr, (unsigned)init);
+            s->high_water = curr + init;
+        }
+        else if (s->high_water < (ulg)curr + WIN_INIT) {
+            /* High water mark at or above current data, but below current data
+             * plus WIN_INIT -- zero out to current data plus WIN_INIT, or up
+             * to end of window, whichever is less.
+             */
+            init = (ulg)curr + WIN_INIT - s->high_water;
+            if (init > s->window_size - s->high_water)
+                init = s->window_size - s->high_water;
+            zmemzero(s->window + s->high_water, (unsigned)init);
+            s->high_water += init;
+        }
+    }
 }
 
 /* ===========================================================================
  * Flush the current block, with given end-of-file flag.
  * IN assertion: strstart is set to the end of the current match.
  */
-#define FLUSH_BLOCK_ONLY(s, eof) { \
+#define FLUSH_BLOCK_ONLY(s, last) { \
    _tr_flush_block(s, (s->block_start >= 0L ? \
                    (charf *)&s->window[(unsigned)s->block_start] : \
                    (charf *)Z_NULL), \
                 (ulg)((long)s->strstart - s->block_start), \
-                (eof)); \
+                (last)); \
    s->block_start = s->strstart; \
    flush_pending(s->strm); \
    Tracev((stderr,"[FLUSH]")); \
 }
 
 /* Same but force premature exit if necessary. */
-#define FLUSH_BLOCK(s, eof) { \
-   FLUSH_BLOCK_ONLY(s, eof); \
-   if (s->strm->avail_out == 0) return (eof) ? finish_started : need_more; \
+#define FLUSH_BLOCK(s, last) { \
+   FLUSH_BLOCK_ONLY(s, last); \
+   if (s->strm->avail_out == 0) return (last) ? finish_started : need_more; \
 }
 
 /* ===========================================================================
@@ -1449,7 +1521,7 @@
     deflate_state *s;
     int flush;
 {
-    IPos hash_head = NIL; /* head of the hash chain */
+    IPos hash_head;       /* head of the hash chain */
     int bflush;           /* set if current block must be flushed */
 
     for (;;) {
@@ -1469,6 +1541,7 @@
         /* Insert the string window[strstart .. strstart+2] in the
          * dictionary, and set hash_head to the head of the hash chain:
          */
+        hash_head = NIL;
         if (s->lookahead >= MIN_MATCH) {
             INSERT_STRING(s, s->strstart, hash_head);
         }
@@ -1481,19 +1554,8 @@
              * of window index 0 (in particular we have to avoid a match
              * of the string with itself at the start of the input file).
              */
-#ifdef FASTEST
-            if ((s->strategy != Z_HUFFMAN_ONLY && s->strategy != Z_RLE) ||
-                (s->strategy == Z_RLE && s->strstart - hash_head == 1)) {
-                s->match_length = longest_match_fast (s, hash_head);
-            }
-#else
-            if (s->strategy != Z_HUFFMAN_ONLY && s->strategy != Z_RLE) {
-                s->match_length = longest_match (s, hash_head);
-            } else if (s->strategy == Z_RLE && s->strstart - hash_head == 1) {
-                s->match_length = longest_match_fast (s, hash_head);
-            }
-#endif
-            /* longest_match() or longest_match_fast() sets match_start */
+            s->match_length = longest_match (s, hash_head);
+            /* longest_match() sets match_start */
         }
         if (s->match_length >= MIN_MATCH) {
             check_match(s, s->strstart, s->match_start, s->match_length);
@@ -1555,7 +1617,7 @@
     deflate_state *s;
     int flush;
 {
-    IPos hash_head = NIL;    /* head of hash chain */
+    IPos hash_head;          /* head of hash chain */
     int bflush;              /* set if current block must be flushed */
 
     /* Process the input block. */
@@ -1576,6 +1638,7 @@
         /* Insert the string window[strstart .. strstart+2] in the
          * dictionary, and set hash_head to the head of the hash chain:
          */
+        hash_head = NIL;
         if (s->lookahead >= MIN_MATCH) {
             INSERT_STRING(s, s->strstart, hash_head);
         }
@@ -1591,12 +1654,8 @@
              * of window index 0 (in particular we have to avoid a match
              * of the string with itself at the start of the input file).
              */
-            if (s->strategy != Z_HUFFMAN_ONLY && s->strategy != Z_RLE) {
-                s->match_length = longest_match (s, hash_head);
-            } else if (s->strategy == Z_RLE && s->strstart - hash_head == 1) {
-                s->match_length = longest_match_fast (s, hash_head);
-            }
-            /* longest_match() or longest_match_fast() sets match_start */
+            s->match_length = longest_match (s, hash_head);
+            /* longest_match() sets match_start */
 
             if (s->match_length <= 5 && (s->strategy == Z_FILTERED
 #if TOO_FAR <= 32767
@@ -1674,7 +1733,6 @@
 }
 #endif /* FASTEST */
 
-#if 0
 /* ===========================================================================
  * For Z_RLE, simply look for runs of bytes, generate matches only of distance
  * one.  Do not maintain a hash table.  (It will be regenerated if this run of
@@ -1684,11 +1742,9 @@
     deflate_state *s;
     int flush;
 {
-    int bflush;         /* set if current block must be flushed */
-    uInt run;           /* length of run */
-    uInt max;           /* maximum length of run */
-    uInt prev;          /* byte at distance one to match */
-    Bytef *scan;        /* scan for end of run */
+    int bflush;             /* set if current block must be flushed */
+    uInt prev;              /* byte at distance one to match */
+    Bytef *scan, *strend;   /* scan goes up to strend for length of run */
 
     for (;;) {
         /* Make sure that we always have enough lookahead, except
@@ -1704,23 +1760,33 @@
         }
 
         /* See how many times the previous byte repeats */
-        run = 0;
-        if (s->strstart > 0) {      /* if there is a previous byte, that is */
-            max = s->lookahead < MAX_MATCH ? s->lookahead : MAX_MATCH;
+        s->match_length = 0;
+        if (s->lookahead >= MIN_MATCH && s->strstart > 0) {
             scan = s->window + s->strstart - 1;
-            prev = *scan++;
-            do {
-                if (*scan++ != prev)
-                    break;
-            } while (++run < max);
+            prev = *scan;
+            if (prev == *++scan && prev == *++scan && prev == *++scan) {
+                strend = s->window + s->strstart + MAX_MATCH;
+                do {
+                } while (prev == *++scan && prev == *++scan &&
+                         prev == *++scan && prev == *++scan &&
+                         prev == *++scan && prev == *++scan &&
+                         prev == *++scan && prev == *++scan &&
+                         scan < strend);
+                s->match_length = MAX_MATCH - (int)(strend - scan);
+                if (s->match_length > s->lookahead)
+                    s->match_length = s->lookahead;
+            }
         }
 
         /* Emit match if have run of MIN_MATCH or longer, else emit literal */
-        if (run >= MIN_MATCH) {
-            check_match(s, s->strstart, s->strstart - 1, run);
-            _tr_tally_dist(s, 1, run - MIN_MATCH, bflush);
-            s->lookahead -= run;
-            s->strstart += run;
+        if (s->match_length >= MIN_MATCH) {
+            check_match(s, s->strstart, s->strstart - 1, s->match_length);
+
+            _tr_tally_dist(s, 1, s->match_length - MIN_MATCH, bflush);
+
+            s->lookahead -= s->match_length;
+            s->strstart += s->match_length;
+            s->match_length = 0;
         } else {
             /* No match, output a literal byte */
             Tracevv((stderr,"%c", s->window[s->strstart]));
@@ -1733,4 +1799,36 @@
     FLUSH_BLOCK(s, flush == Z_FINISH);
     return flush == Z_FINISH ? finish_done : block_done;
 }
-#endif
+
+/* ===========================================================================
+ * For Z_HUFFMAN_ONLY, do not look for matches.  Do not maintain a hash table.
+ * (It will be regenerated if this run of deflate switches away from Huffman.)
+ */
+local block_state deflate_huff(s, flush)
+    deflate_state *s;
+    int flush;
+{
+    int bflush;             /* set if current block must be flushed */
+
+    for (;;) {
+        /* Make sure that we have a literal to write. */
+        if (s->lookahead == 0) {
+            fill_window(s);
+            if (s->lookahead == 0) {
+                if (flush == Z_NO_FLUSH)
+                    return need_more;
+                break;      /* flush the current block */
+            }
+        }
+
+        /* Output a literal byte */
+        s->match_length = 0;
+        Tracevv((stderr,"%c", s->window[s->strstart]));
+        _tr_tally_lit (s, s->window[s->strstart], bflush);
+        s->lookahead--;
+        s->strstart++;
+        if (bflush) FLUSH_BLOCK(s, 0);
+    }
+    FLUSH_BLOCK(s, flush == Z_FINISH);
+    return flush == Z_FINISH ? finish_done : block_done;
+}

Modified: python/branches/py3k/Modules/zlib/deflate.h
==============================================================================
--- python/branches/py3k/Modules/zlib/deflate.h	(original)
+++ python/branches/py3k/Modules/zlib/deflate.h	Fri Jul 30 22:03:17 2010
@@ -1,5 +1,5 @@
 /* deflate.h -- internal compression state
- * Copyright (C) 1995-2004 Jean-loup Gailly
+ * Copyright (C) 1995-2010 Jean-loup Gailly
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -260,6 +260,13 @@
      * are always zero.
      */
 
+    ulg high_water;
+    /* High water mark offset in window for initialized bytes -- bytes above
+     * this are set to zero in order to avoid memory check warnings when
+     * longest match routines access bytes past the input.  This is then
+     * updated to the new high water mark.
+     */
+
 } FAR deflate_state;
 
 /* Output a byte on the stream.
@@ -278,14 +285,18 @@
  * distances are limited to MAX_DIST instead of WSIZE.
  */
 
+#define WIN_INIT MAX_MATCH
+/* Number of bytes after end of data in window to initialize in order to avoid
+   memory checker errors from longest match routines */
+
         /* in trees.c */
-void _tr_init         OF((deflate_state *s));
-int  _tr_tally        OF((deflate_state *s, unsigned dist, unsigned lc));
-void _tr_flush_block  OF((deflate_state *s, charf *buf, ulg stored_len,
-                          int eof));
-void _tr_align        OF((deflate_state *s));
-void _tr_stored_block OF((deflate_state *s, charf *buf, ulg stored_len,
-                          int eof));
+void ZLIB_INTERNAL _tr_init OF((deflate_state *s));
+int ZLIB_INTERNAL _tr_tally OF((deflate_state *s, unsigned dist, unsigned lc));
+void ZLIB_INTERNAL _tr_flush_block OF((deflate_state *s, charf *buf,
+                        ulg stored_len, int last));
+void ZLIB_INTERNAL _tr_align OF((deflate_state *s));
+void ZLIB_INTERNAL _tr_stored_block OF((deflate_state *s, charf *buf,
+                        ulg stored_len, int last));
 
 #define d_code(dist) \
    ((dist) < 256 ? _dist_code[dist] : _dist_code[256+((dist)>>7)])
@@ -298,11 +309,11 @@
 /* Inline versions of _tr_tally for speed: */
 
 #if defined(GEN_TREES_H) || !defined(STDC)
-  extern uch _length_code[];
-  extern uch _dist_code[];
+  extern uch ZLIB_INTERNAL _length_code[];
+  extern uch ZLIB_INTERNAL _dist_code[];
 #else
-  extern const uch _length_code[];
-  extern const uch _dist_code[];
+  extern const uch ZLIB_INTERNAL _length_code[];
+  extern const uch ZLIB_INTERNAL _dist_code[];
 #endif
 
 # define _tr_tally_lit(s, c, flush) \

Modified: python/branches/py3k/Modules/zlib/example.c
==============================================================================
--- python/branches/py3k/Modules/zlib/example.c	(original)
+++ python/branches/py3k/Modules/zlib/example.c	Fri Jul 30 22:03:17 2010
@@ -1,19 +1,19 @@
 /* example.c -- usage example of the zlib compression library
- * Copyright (C) 1995-2004 Jean-loup Gailly.
+ * Copyright (C) 1995-2006 Jean-loup Gailly.
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
 /* @(#) $Id$ */
 
-#include <stdio.h>
 #include "zlib.h"
+#include <stdio.h>
 
 #ifdef STDC
 #  include <string.h>
 #  include <stdlib.h>
 #endif
 
-#if defined(VMS)
+#if defined(VMS) || defined(RISCOS)
 #  define TESTFILE "foo-gz"
 #else
 #  define TESTFILE "foo.gz"

Modified: python/branches/py3k/Modules/zlib/infback.c
==============================================================================
--- python/branches/py3k/Modules/zlib/infback.c	(original)
+++ python/branches/py3k/Modules/zlib/infback.c	Fri Jul 30 22:03:17 2010
@@ -1,5 +1,5 @@
 /* infback.c -- inflate using a call-back interface
- * Copyright (C) 1995-2005 Mark Adler
+ * Copyright (C) 1995-2009 Mark Adler
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -55,7 +55,7 @@
     state->wbits = windowBits;
     state->wsize = 1U << windowBits;
     state->window = window;
-    state->write = 0;
+    state->wnext = 0;
     state->whave = 0;
     return Z_OK;
 }
@@ -253,7 +253,7 @@
     unsigned bits;              /* bits in bit buffer */
     unsigned copy;              /* number of stored or match bytes to copy */
     unsigned char FAR *from;    /* where to copy match bytes from */
-    code this;                  /* current decoding table entry */
+    code here;                  /* current decoding table entry */
     code last;                  /* parent table entry */
     unsigned len;               /* length to copy for repeats, bits to drop */
     int ret;                    /* return code */
@@ -389,19 +389,19 @@
             state->have = 0;
             while (state->have < state->nlen + state->ndist) {
                 for (;;) {
-                    this = state->lencode[BITS(state->lenbits)];
-                    if ((unsigned)(this.bits) <= bits) break;
+                    here = state->lencode[BITS(state->lenbits)];
+                    if ((unsigned)(here.bits) <= bits) break;
                     PULLBYTE();
                 }
-                if (this.val < 16) {
-                    NEEDBITS(this.bits);
-                    DROPBITS(this.bits);
-                    state->lens[state->have++] = this.val;
+                if (here.val < 16) {
+                    NEEDBITS(here.bits);
+                    DROPBITS(here.bits);
+                    state->lens[state->have++] = here.val;
                 }
                 else {
-                    if (this.val == 16) {
-                        NEEDBITS(this.bits + 2);
-                        DROPBITS(this.bits);
+                    if (here.val == 16) {
+                        NEEDBITS(here.bits + 2);
+                        DROPBITS(here.bits);
                         if (state->have == 0) {
                             strm->msg = (char *)"invalid bit length repeat";
                             state->mode = BAD;
@@ -411,16 +411,16 @@
                         copy = 3 + BITS(2);
                         DROPBITS(2);
                     }
-                    else if (this.val == 17) {
-                        NEEDBITS(this.bits + 3);
-                        DROPBITS(this.bits);
+                    else if (here.val == 17) {
+                        NEEDBITS(here.bits + 3);
+                        DROPBITS(here.bits);
                         len = 0;
                         copy = 3 + BITS(3);
                         DROPBITS(3);
                     }
                     else {
-                        NEEDBITS(this.bits + 7);
-                        DROPBITS(this.bits);
+                        NEEDBITS(here.bits + 7);
+                        DROPBITS(here.bits);
                         len = 0;
                         copy = 11 + BITS(7);
                         DROPBITS(7);
@@ -438,7 +438,16 @@
             /* handle error breaks in while */
             if (state->mode == BAD) break;
 
-            /* build code tables */
+            /* check for end-of-block code (better have one) */
+            if (state->lens[256] == 0) {
+                strm->msg = (char *)"invalid code -- missing end-of-block";
+                state->mode = BAD;
+                break;
+            }
+
+            /* build code tables -- note: do not change the lenbits or distbits
+               values here (9 and 6) without reading the comments in inftrees.h
+               concerning the ENOUGH constants, which depend on those values */
             state->next = state->codes;
             state->lencode = (code const FAR *)(state->next);
             state->lenbits = 9;
@@ -474,28 +483,28 @@
 
             /* get a literal, length, or end-of-block code */
             for (;;) {
-                this = state->lencode[BITS(state->lenbits)];
-                if ((unsigned)(this.bits) <= bits) break;
+                here = state->lencode[BITS(state->lenbits)];
+                if ((unsigned)(here.bits) <= bits) break;
                 PULLBYTE();
             }
-            if (this.op && (this.op & 0xf0) == 0) {
-                last = this;
+            if (here.op && (here.op & 0xf0) == 0) {
+                last = here;
                 for (;;) {
-                    this = state->lencode[last.val +
+                    here = state->lencode[last.val +
                             (BITS(last.bits + last.op) >> last.bits)];
-                    if ((unsigned)(last.bits + this.bits) <= bits) break;
+                    if ((unsigned)(last.bits + here.bits) <= bits) break;
                     PULLBYTE();
                 }
                 DROPBITS(last.bits);
             }
-            DROPBITS(this.bits);
-            state->length = (unsigned)this.val;
+            DROPBITS(here.bits);
+            state->length = (unsigned)here.val;
 
             /* process literal */
-            if (this.op == 0) {
-                Tracevv((stderr, this.val >= 0x20 && this.val < 0x7f ?
+            if (here.op == 0) {
+                Tracevv((stderr, here.val >= 0x20 && here.val < 0x7f ?
                         "inflate:         literal '%c'\n" :
-                        "inflate:         literal 0x%02x\n", this.val));
+                        "inflate:         literal 0x%02x\n", here.val));
                 ROOM();
                 *put++ = (unsigned char)(state->length);
                 left--;
@@ -504,21 +513,21 @@
             }
 
             /* process end of block */
-            if (this.op & 32) {
+            if (here.op & 32) {
                 Tracevv((stderr, "inflate:         end of block\n"));
                 state->mode = TYPE;
                 break;
             }
 
             /* invalid code */
-            if (this.op & 64) {
+            if (here.op & 64) {
                 strm->msg = (char *)"invalid literal/length code";
                 state->mode = BAD;
                 break;
             }
 
             /* length code -- get extra bits, if any */
-            state->extra = (unsigned)(this.op) & 15;
+            state->extra = (unsigned)(here.op) & 15;
             if (state->extra != 0) {
                 NEEDBITS(state->extra);
                 state->length += BITS(state->extra);
@@ -528,30 +537,30 @@
 
             /* get distance code */
             for (;;) {
-                this = state->distcode[BITS(state->distbits)];
-                if ((unsigned)(this.bits) <= bits) break;
+                here = state->distcode[BITS(state->distbits)];
+                if ((unsigned)(here.bits) <= bits) break;
                 PULLBYTE();
             }
-            if ((this.op & 0xf0) == 0) {
-                last = this;
+            if ((here.op & 0xf0) == 0) {
+                last = here;
                 for (;;) {
-                    this = state->distcode[last.val +
+                    here = state->distcode[last.val +
                             (BITS(last.bits + last.op) >> last.bits)];
-                    if ((unsigned)(last.bits + this.bits) <= bits) break;
+                    if ((unsigned)(last.bits + here.bits) <= bits) break;
                     PULLBYTE();
                 }
                 DROPBITS(last.bits);
             }
-            DROPBITS(this.bits);
-            if (this.op & 64) {
+            DROPBITS(here.bits);
+            if (here.op & 64) {
                 strm->msg = (char *)"invalid distance code";
                 state->mode = BAD;
                 break;
             }
-            state->offset = (unsigned)this.val;
+            state->offset = (unsigned)here.val;
 
             /* get distance extra bits, if any */
-            state->extra = (unsigned)(this.op) & 15;
+            state->extra = (unsigned)(here.op) & 15;
             if (state->extra != 0) {
                 NEEDBITS(state->extra);
                 state->offset += BITS(state->extra);

Modified: python/branches/py3k/Modules/zlib/inffast.c
==============================================================================
--- python/branches/py3k/Modules/zlib/inffast.c	(original)
+++ python/branches/py3k/Modules/zlib/inffast.c	Fri Jul 30 22:03:17 2010
@@ -1,5 +1,5 @@
 /* inffast.c -- fast decoding
- * Copyright (C) 1995-2004 Mark Adler
+ * Copyright (C) 1995-2008, 2010 Mark Adler
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -64,7 +64,7 @@
       requires strm->avail_out >= 258 for each loop to avoid checking for
       output space.
  */
-void inflate_fast(strm, start)
+void ZLIB_INTERNAL inflate_fast(strm, start)
 z_streamp strm;
 unsigned start;         /* inflate()'s starting value for strm->avail_out */
 {
@@ -79,7 +79,7 @@
 #endif
     unsigned wsize;             /* window size or zero if not using window */
     unsigned whave;             /* valid bytes in the window */
-    unsigned write;             /* window write index */
+    unsigned wnext;             /* window write index */
     unsigned char FAR *window;  /* allocated sliding window, if wsize != 0 */
     unsigned long hold;         /* local strm->hold */
     unsigned bits;              /* local strm->bits */
@@ -87,7 +87,7 @@
     code const FAR *dcode;      /* local strm->distcode */
     unsigned lmask;             /* mask for first level of length codes */
     unsigned dmask;             /* mask for first level of distance codes */
-    code this;                  /* retrieved table entry */
+    code here;                  /* retrieved table entry */
     unsigned op;                /* code bits, operation, extra bits, or */
                                 /*  window position, window bytes to copy */
     unsigned len;               /* match length, unused bytes */
@@ -106,7 +106,7 @@
 #endif
     wsize = state->wsize;
     whave = state->whave;
-    write = state->write;
+    wnext = state->wnext;
     window = state->window;
     hold = state->hold;
     bits = state->bits;
@@ -124,20 +124,20 @@
             hold += (unsigned long)(PUP(in)) << bits;
             bits += 8;
         }
-        this = lcode[hold & lmask];
+        here = lcode[hold & lmask];
       dolen:
-        op = (unsigned)(this.bits);
+        op = (unsigned)(here.bits);
         hold >>= op;
         bits -= op;
-        op = (unsigned)(this.op);
+        op = (unsigned)(here.op);
         if (op == 0) {                          /* literal */
-            Tracevv((stderr, this.val >= 0x20 && this.val < 0x7f ?
+            Tracevv((stderr, here.val >= 0x20 && here.val < 0x7f ?
                     "inflate:         literal '%c'\n" :
-                    "inflate:         literal 0x%02x\n", this.val));
-            PUP(out) = (unsigned char)(this.val);
+                    "inflate:         literal 0x%02x\n", here.val));
+            PUP(out) = (unsigned char)(here.val);
         }
         else if (op & 16) {                     /* length base */
-            len = (unsigned)(this.val);
+            len = (unsigned)(here.val);
             op &= 15;                           /* number of extra bits */
             if (op) {
                 if (bits < op) {
@@ -155,14 +155,14 @@
                 hold += (unsigned long)(PUP(in)) << bits;
                 bits += 8;
             }
-            this = dcode[hold & dmask];
+            here = dcode[hold & dmask];
           dodist:
-            op = (unsigned)(this.bits);
+            op = (unsigned)(here.bits);
             hold >>= op;
             bits -= op;
-            op = (unsigned)(this.op);
+            op = (unsigned)(here.op);
             if (op & 16) {                      /* distance base */
-                dist = (unsigned)(this.val);
+                dist = (unsigned)(here.val);
                 op &= 15;                       /* number of extra bits */
                 if (bits < op) {
                     hold += (unsigned long)(PUP(in)) << bits;
@@ -187,12 +187,34 @@
                 if (dist > op) {                /* see if copy from window */
                     op = dist - op;             /* distance back in window */
                     if (op > whave) {
-                        strm->msg = (char *)"invalid distance too far back";
-                        state->mode = BAD;
-                        break;
+                        if (state->sane) {
+                            strm->msg =
+                                (char *)"invalid distance too far back";
+                            state->mode = BAD;
+                            break;
+                        }
+#ifdef INFLATE_ALLOW_INVALID_DISTANCE_TOOFAR_ARRR
+                        if (len <= op - whave) {
+                            do {
+                                PUP(out) = 0;
+                            } while (--len);
+                            continue;
+                        }
+                        len -= op - whave;
+                        do {
+                            PUP(out) = 0;
+                        } while (--op > whave);
+                        if (op == 0) {
+                            from = out - dist;
+                            do {
+                                PUP(out) = PUP(from);
+                            } while (--len);
+                            continue;
+                        }
+#endif
                     }
                     from = window - OFF;
-                    if (write == 0) {           /* very common case */
+                    if (wnext == 0) {           /* very common case */
                         from += wsize - op;
                         if (op < len) {         /* some from window */
                             len -= op;
@@ -202,17 +224,17 @@
                             from = out - dist;  /* rest from output */
                         }
                     }
-                    else if (write < op) {      /* wrap around window */
-                        from += wsize + write - op;
-                        op -= write;
+                    else if (wnext < op) {      /* wrap around window */
+                        from += wsize + wnext - op;
+                        op -= wnext;
                         if (op < len) {         /* some from end of window */
                             len -= op;
                             do {
                                 PUP(out) = PUP(from);
                             } while (--op);
                             from = window - OFF;
-                            if (write < len) {  /* some from start of window */
-                                op = write;
+                            if (wnext < len) {  /* some from start of window */
+                                op = wnext;
                                 len -= op;
                                 do {
                                     PUP(out) = PUP(from);
@@ -222,7 +244,7 @@
                         }
                     }
                     else {                      /* contiguous in window */
-                        from += write - op;
+                        from += wnext - op;
                         if (op < len) {         /* some from window */
                             len -= op;
                             do {
@@ -259,7 +281,7 @@
                 }
             }
             else if ((op & 64) == 0) {          /* 2nd level distance code */
-                this = dcode[this.val + (hold & ((1U << op) - 1))];
+                here = dcode[here.val + (hold & ((1U << op) - 1))];
                 goto dodist;
             }
             else {
@@ -269,7 +291,7 @@
             }
         }
         else if ((op & 64) == 0) {              /* 2nd level length code */
-            this = lcode[this.val + (hold & ((1U << op) - 1))];
+            here = lcode[here.val + (hold & ((1U << op) - 1))];
             goto dolen;
         }
         else if (op & 32) {                     /* end-of-block */
@@ -305,7 +327,7 @@
    inflate_fast() speedups that turned out slower (on a PowerPC G3 750CXe):
    - Using bit fields for code structure
    - Different op definition to avoid & for extra bits (do & for table bits)
-   - Three separate decoding do-loops for direct, window, and write == 0
+   - Three separate decoding do-loops for direct, window, and wnext == 0
    - Special case for distance > 1 copies to do overlapped load and store copy
    - Explicit branch predictions (based on measured branch probabilities)
    - Deferring match copy and interspersed it with decoding subsequent codes

Modified: python/branches/py3k/Modules/zlib/inffast.h
==============================================================================
--- python/branches/py3k/Modules/zlib/inffast.h	(original)
+++ python/branches/py3k/Modules/zlib/inffast.h	Fri Jul 30 22:03:17 2010
@@ -1,5 +1,5 @@
 /* inffast.h -- header to use inffast.c
- * Copyright (C) 1995-2003 Mark Adler
+ * Copyright (C) 1995-2003, 2010 Mark Adler
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -8,4 +8,4 @@
    subject to change. Applications should only use zlib.h.
  */
 
-void inflate_fast OF((z_streamp strm, unsigned start));
+void ZLIB_INTERNAL inflate_fast OF((z_streamp strm, unsigned start));

Modified: python/branches/py3k/Modules/zlib/inflate.c
==============================================================================
--- python/branches/py3k/Modules/zlib/inflate.c	(original)
+++ python/branches/py3k/Modules/zlib/inflate.c	Fri Jul 30 22:03:17 2010
@@ -1,5 +1,5 @@
 /* inflate.c -- zlib decompression
- * Copyright (C) 1995-2005 Mark Adler
+ * Copyright (C) 1995-2010 Mark Adler
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -45,7 +45,7 @@
  * - Rearrange window copies in inflate_fast() for speed and simplification
  * - Unroll last copy for window match in inflate_fast()
  * - Use local copies of window variables in inflate_fast() for speed
- * - Pull out common write == 0 case for speed in inflate_fast()
+ * - Pull out common wnext == 0 case for speed in inflate_fast()
  * - Make op and len in inflate_fast() unsigned for consistency
  * - Add FAR to lcode and dcode declarations in inflate_fast()
  * - Simplified bad distance check in inflate_fast()
@@ -117,28 +117,52 @@
     state->head = Z_NULL;
     state->wsize = 0;
     state->whave = 0;
-    state->write = 0;
+    state->wnext = 0;
     state->hold = 0;
     state->bits = 0;
     state->lencode = state->distcode = state->next = state->codes;
+    state->sane = 1;
+    state->back = -1;
     Tracev((stderr, "inflate: reset\n"));
     return Z_OK;
 }
 
-int ZEXPORT inflatePrime(strm, bits, value)
+int ZEXPORT inflateReset2(strm, windowBits)
 z_streamp strm;
-int bits;
-int value;
+int windowBits;
 {
+    int wrap;
     struct inflate_state FAR *state;
 
+    /* get the state */
     if (strm == Z_NULL || strm->state == Z_NULL) return Z_STREAM_ERROR;
     state = (struct inflate_state FAR *)strm->state;
-    if (bits > 16 || state->bits + bits > 32) return Z_STREAM_ERROR;
-    value &= (1L << bits) - 1;
-    state->hold += value << state->bits;
-    state->bits += bits;
-    return Z_OK;
+
+    /* extract wrap request from windowBits parameter */
+    if (windowBits < 0) {
+        wrap = 0;
+        windowBits = -windowBits;
+    }
+    else {
+        wrap = (windowBits >> 4) + 1;
+#ifdef GUNZIP
+        if (windowBits < 48)
+            windowBits &= 15;
+#endif
+    }
+
+    /* set number of window bits, free window if different */
+    if (windowBits && (windowBits < 8 || windowBits > 15))
+        return Z_STREAM_ERROR;
+    if (state->window != Z_NULL && state->wbits != (unsigned)windowBits) {
+        ZFREE(strm, state->window);
+        state->window = Z_NULL;
+    }
+
+    /* update state and reset the rest of it */
+    state->wrap = wrap;
+    state->wbits = (unsigned)windowBits;
+    return inflateReset(strm);
 }
 
 int ZEXPORT inflateInit2_(strm, windowBits, version, stream_size)
@@ -147,6 +171,7 @@
 const char *version;
 int stream_size;
 {
+    int ret;
     struct inflate_state FAR *state;
 
     if (version == Z_NULL || version[0] != ZLIB_VERSION[0] ||
@@ -164,24 +189,13 @@
     if (state == Z_NULL) return Z_MEM_ERROR;
     Tracev((stderr, "inflate: allocated\n"));
     strm->state = (struct internal_state FAR *)state;
-    if (windowBits < 0) {
-        state->wrap = 0;
-        windowBits = -windowBits;
-    }
-    else {
-        state->wrap = (windowBits >> 4) + 1;
-#ifdef GUNZIP
-        if (windowBits < 48) windowBits &= 15;
-#endif
-    }
-    if (windowBits < 8 || windowBits > 15) {
+    state->window = Z_NULL;
+    ret = inflateReset2(strm, windowBits);
+    if (ret != Z_OK) {
         ZFREE(strm, state);
         strm->state = Z_NULL;
-        return Z_STREAM_ERROR;
     }
-    state->wbits = (unsigned)windowBits;
-    state->window = Z_NULL;
-    return inflateReset(strm);
+    return ret;
 }
 
 int ZEXPORT inflateInit_(strm, version, stream_size)
@@ -192,6 +206,27 @@
     return inflateInit2_(strm, DEF_WBITS, version, stream_size);
 }
 
+int ZEXPORT inflatePrime(strm, bits, value)
+z_streamp strm;
+int bits;
+int value;
+{
+    struct inflate_state FAR *state;
+
+    if (strm == Z_NULL || strm->state == Z_NULL) return Z_STREAM_ERROR;
+    state = (struct inflate_state FAR *)strm->state;
+    if (bits < 0) {
+        state->hold = 0;
+        state->bits = 0;
+        return Z_OK;
+    }
+    if (bits > 16 || state->bits + bits > 32) return Z_STREAM_ERROR;
+    value &= (1L << bits) - 1;
+    state->hold += value << state->bits;
+    state->bits += bits;
+    return Z_OK;
+}
+
 /*
    Return state with length and distance decoding tables and index sizes set to
    fixed code decoding.  Normally this returns fixed tables from inffixed.h.
@@ -340,7 +375,7 @@
     /* if window not in use yet, initialize */
     if (state->wsize == 0) {
         state->wsize = 1U << state->wbits;
-        state->write = 0;
+        state->wnext = 0;
         state->whave = 0;
     }
 
@@ -348,22 +383,22 @@
     copy = out - strm->avail_out;
     if (copy >= state->wsize) {
         zmemcpy(state->window, strm->next_out - state->wsize, state->wsize);
-        state->write = 0;
+        state->wnext = 0;
         state->whave = state->wsize;
     }
     else {
-        dist = state->wsize - state->write;
+        dist = state->wsize - state->wnext;
         if (dist > copy) dist = copy;
-        zmemcpy(state->window + state->write, strm->next_out - copy, dist);
+        zmemcpy(state->window + state->wnext, strm->next_out - copy, dist);
         copy -= dist;
         if (copy) {
             zmemcpy(state->window, strm->next_out - copy, copy);
-            state->write = copy;
+            state->wnext = copy;
             state->whave = state->wsize;
         }
         else {
-            state->write += dist;
-            if (state->write == state->wsize) state->write = 0;
+            state->wnext += dist;
+            if (state->wnext == state->wsize) state->wnext = 0;
             if (state->whave < state->wsize) state->whave += dist;
         }
     }
@@ -564,7 +599,7 @@
     unsigned in, out;           /* save starting available input and output */
     unsigned copy;              /* number of stored or match bytes to copy */
     unsigned char FAR *from;    /* where to copy match bytes from */
-    code this;                  /* current decoding table entry */
+    code here;                  /* current decoding table entry */
     code last;                  /* parent table entry */
     unsigned len;               /* length to copy for repeats, bits to drop */
     int ret;                    /* return code */
@@ -619,7 +654,9 @@
             }
             DROPBITS(4);
             len = BITS(4) + 8;
-            if (len > state->wbits) {
+            if (state->wbits == 0)
+                state->wbits = len;
+            else if (len > state->wbits) {
                 strm->msg = (char *)"invalid window size";
                 state->mode = BAD;
                 break;
@@ -771,7 +808,7 @@
             strm->adler = state->check = adler32(0L, Z_NULL, 0);
             state->mode = TYPE;
         case TYPE:
-            if (flush == Z_BLOCK) goto inf_leave;
+            if (flush == Z_BLOCK || flush == Z_TREES) goto inf_leave;
         case TYPEDO:
             if (state->last) {
                 BYTEBITS();
@@ -791,7 +828,11 @@
                 fixedtables(state);
                 Tracev((stderr, "inflate:     fixed codes block%s\n",
                         state->last ? " (last)" : ""));
-                state->mode = LEN;              /* decode codes */
+                state->mode = LEN_;             /* decode codes */
+                if (flush == Z_TREES) {
+                    DROPBITS(2);
+                    goto inf_leave;
+                }
                 break;
             case 2:                             /* dynamic block */
                 Tracev((stderr, "inflate:     dynamic codes block%s\n",
@@ -816,6 +857,9 @@
             Tracev((stderr, "inflate:       stored length %u\n",
                     state->length));
             INITBITS();
+            state->mode = COPY_;
+            if (flush == Z_TREES) goto inf_leave;
+        case COPY_:
             state->mode = COPY;
         case COPY:
             copy = state->length;
@@ -876,19 +920,19 @@
         case CODELENS:
             while (state->have < state->nlen + state->ndist) {
                 for (;;) {
-                    this = state->lencode[BITS(state->lenbits)];
-                    if ((unsigned)(this.bits) <= bits) break;
+                    here = state->lencode[BITS(state->lenbits)];
+                    if ((unsigned)(here.bits) <= bits) break;
                     PULLBYTE();
                 }
-                if (this.val < 16) {
-                    NEEDBITS(this.bits);
-                    DROPBITS(this.bits);
-                    state->lens[state->have++] = this.val;
+                if (here.val < 16) {
+                    NEEDBITS(here.bits);
+                    DROPBITS(here.bits);
+                    state->lens[state->have++] = here.val;
                 }
                 else {
-                    if (this.val == 16) {
-                        NEEDBITS(this.bits + 2);
-                        DROPBITS(this.bits);
+                    if (here.val == 16) {
+                        NEEDBITS(here.bits + 2);
+                        DROPBITS(here.bits);
                         if (state->have == 0) {
                             strm->msg = (char *)"invalid bit length repeat";
                             state->mode = BAD;
@@ -898,16 +942,16 @@
                         copy = 3 + BITS(2);
                         DROPBITS(2);
                     }
-                    else if (this.val == 17) {
-                        NEEDBITS(this.bits + 3);
-                        DROPBITS(this.bits);
+                    else if (here.val == 17) {
+                        NEEDBITS(here.bits + 3);
+                        DROPBITS(here.bits);
                         len = 0;
                         copy = 3 + BITS(3);
                         DROPBITS(3);
                     }
                     else {
-                        NEEDBITS(this.bits + 7);
-                        DROPBITS(this.bits);
+                        NEEDBITS(here.bits + 7);
+                        DROPBITS(here.bits);
                         len = 0;
                         copy = 11 + BITS(7);
                         DROPBITS(7);
@@ -925,7 +969,16 @@
             /* handle error breaks in while */
             if (state->mode == BAD) break;
 
-            /* build code tables */
+            /* check for end-of-block code (better have one) */
+            if (state->lens[256] == 0) {
+                strm->msg = (char *)"invalid code -- missing end-of-block";
+                state->mode = BAD;
+                break;
+            }
+
+            /* build code tables -- note: do not change the lenbits or distbits
+               values here (9 and 6) without reading the comments in inftrees.h
+               concerning the ENOUGH constants, which depend on those values */
             state->next = state->codes;
             state->lencode = (code const FAR *)(state->next);
             state->lenbits = 9;
@@ -946,88 +999,102 @@
                 break;
             }
             Tracev((stderr, "inflate:       codes ok\n"));
+            state->mode = LEN_;
+            if (flush == Z_TREES) goto inf_leave;
+        case LEN_:
             state->mode = LEN;
         case LEN:
             if (have >= 6 && left >= 258) {
                 RESTORE();
                 inflate_fast(strm, out);
                 LOAD();
+                if (state->mode == TYPE)
+                    state->back = -1;
                 break;
             }
+            state->back = 0;
             for (;;) {
-                this = state->lencode[BITS(state->lenbits)];
-                if ((unsigned)(this.bits) <= bits) break;
+                here = state->lencode[BITS(state->lenbits)];
+                if ((unsigned)(here.bits) <= bits) break;
                 PULLBYTE();
             }
-            if (this.op && (this.op & 0xf0) == 0) {
-                last = this;
+            if (here.op && (here.op & 0xf0) == 0) {
+                last = here;
                 for (;;) {
-                    this = state->lencode[last.val +
+                    here = state->lencode[last.val +
                             (BITS(last.bits + last.op) >> last.bits)];
-                    if ((unsigned)(last.bits + this.bits) <= bits) break;
+                    if ((unsigned)(last.bits + here.bits) <= bits) break;
                     PULLBYTE();
                 }
                 DROPBITS(last.bits);
+                state->back += last.bits;
             }
-            DROPBITS(this.bits);
-            state->length = (unsigned)this.val;
-            if ((int)(this.op) == 0) {
-                Tracevv((stderr, this.val >= 0x20 && this.val < 0x7f ?
+            DROPBITS(here.bits);
+            state->back += here.bits;
+            state->length = (unsigned)here.val;
+            if ((int)(here.op) == 0) {
+                Tracevv((stderr, here.val >= 0x20 && here.val < 0x7f ?
                         "inflate:         literal '%c'\n" :
-                        "inflate:         literal 0x%02x\n", this.val));
+                        "inflate:         literal 0x%02x\n", here.val));
                 state->mode = LIT;
                 break;
             }
-            if (this.op & 32) {
+            if (here.op & 32) {
                 Tracevv((stderr, "inflate:         end of block\n"));
+                state->back = -1;
                 state->mode = TYPE;
                 break;
             }
-            if (this.op & 64) {
+            if (here.op & 64) {
                 strm->msg = (char *)"invalid literal/length code";
                 state->mode = BAD;
                 break;
             }
-            state->extra = (unsigned)(this.op) & 15;
+            state->extra = (unsigned)(here.op) & 15;
             state->mode = LENEXT;
         case LENEXT:
             if (state->extra) {
                 NEEDBITS(state->extra);
                 state->length += BITS(state->extra);
                 DROPBITS(state->extra);
+                state->back += state->extra;
             }
             Tracevv((stderr, "inflate:         length %u\n", state->length));
+            state->was = state->length;
             state->mode = DIST;
         case DIST:
             for (;;) {
-                this = state->distcode[BITS(state->distbits)];
-                if ((unsigned)(this.bits) <= bits) break;
+                here = state->distcode[BITS(state->distbits)];
+                if ((unsigned)(here.bits) <= bits) break;
                 PULLBYTE();
             }
-            if ((this.op & 0xf0) == 0) {
-                last = this;
+            if ((here.op & 0xf0) == 0) {
+                last = here;
                 for (;;) {
-                    this = state->distcode[last.val +
+                    here = state->distcode[last.val +
                             (BITS(last.bits + last.op) >> last.bits)];
-                    if ((unsigned)(last.bits + this.bits) <= bits) break;
+                    if ((unsigned)(last.bits + here.bits) <= bits) break;
                     PULLBYTE();
                 }
                 DROPBITS(last.bits);
+                state->back += last.bits;
             }
-            DROPBITS(this.bits);
-            if (this.op & 64) {
+            DROPBITS(here.bits);
+            state->back += here.bits;
+            if (here.op & 64) {
                 strm->msg = (char *)"invalid distance code";
                 state->mode = BAD;
                 break;
             }
-            state->offset = (unsigned)this.val;
-            state->extra = (unsigned)(this.op) & 15;
+            state->offset = (unsigned)here.val;
+            state->extra = (unsigned)(here.op) & 15;
             state->mode = DISTEXT;
         case DISTEXT:
             if (state->extra) {
                 NEEDBITS(state->extra);
                 state->offset += BITS(state->extra);
                 DROPBITS(state->extra);
+                state->back += state->extra;
             }
 #ifdef INFLATE_STRICT
             if (state->offset > state->dmax) {
@@ -1036,11 +1103,6 @@
                 break;
             }
 #endif
-            if (state->offset > state->whave + out - left) {
-                strm->msg = (char *)"invalid distance too far back";
-                state->mode = BAD;
-                break;
-            }
             Tracevv((stderr, "inflate:         distance %u\n", state->offset));
             state->mode = MATCH;
         case MATCH:
@@ -1048,12 +1110,32 @@
             copy = out - left;
             if (state->offset > copy) {         /* copy from window */
                 copy = state->offset - copy;
-                if (copy > state->write) {
-                    copy -= state->write;
+                if (copy > state->whave) {
+                    if (state->sane) {
+                        strm->msg = (char *)"invalid distance too far back";
+                        state->mode = BAD;
+                        break;
+                    }
+#ifdef INFLATE_ALLOW_INVALID_DISTANCE_TOOFAR_ARRR
+                    Trace((stderr, "inflate.c too far\n"));
+                    copy -= state->whave;
+                    if (copy > state->length) copy = state->length;
+                    if (copy > left) copy = left;
+                    left -= copy;
+                    state->length -= copy;
+                    do {
+                        *put++ = 0;
+                    } while (--copy);
+                    if (state->length == 0) state->mode = LEN;
+                    break;
+#endif
+                }
+                if (copy > state->wnext) {
+                    copy -= state->wnext;
                     from = state->window + (state->wsize - copy);
                 }
                 else
-                    from = state->window + (state->write - copy);
+                    from = state->window + (state->wnext - copy);
                 if (copy > state->length) copy = state->length;
             }
             else {                              /* copy from output */
@@ -1146,7 +1228,8 @@
         strm->adler = state->check =
             UPDATE(state->check, strm->next_out - out, out);
     strm->data_type = state->bits + (state->last ? 64 : 0) +
-                      (state->mode == TYPE ? 128 : 0);
+                      (state->mode == TYPE ? 128 : 0) +
+                      (state->mode == LEN_ || state->mode == COPY_ ? 256 : 0);
     if (((in == 0 && out == 0) || flush == Z_FINISH) && ret == Z_OK)
         ret = Z_BUF_ERROR;
     return ret;
@@ -1366,3 +1449,32 @@
     dest->state = (struct internal_state FAR *)copy;
     return Z_OK;
 }
+
+int ZEXPORT inflateUndermine(strm, subvert)
+z_streamp strm;
+int subvert;
+{
+    struct inflate_state FAR *state;
+
+    if (strm == Z_NULL || strm->state == Z_NULL) return Z_STREAM_ERROR;
+    state = (struct inflate_state FAR *)strm->state;
+    state->sane = !subvert;
+#ifdef INFLATE_ALLOW_INVALID_DISTANCE_TOOFAR_ARRR
+    return Z_OK;
+#else
+    state->sane = 1;
+    return Z_DATA_ERROR;
+#endif
+}
+
+long ZEXPORT inflateMark(strm)
+z_streamp strm;
+{
+    struct inflate_state FAR *state;
+
+    if (strm == Z_NULL || strm->state == Z_NULL) return -1L << 16;
+    state = (struct inflate_state FAR *)strm->state;
+    return ((long)(state->back) << 16) +
+        (state->mode == COPY ? state->length :
+            (state->mode == MATCH ? state->was - state->length : 0));
+}

Modified: python/branches/py3k/Modules/zlib/inflate.h
==============================================================================
--- python/branches/py3k/Modules/zlib/inflate.h	(original)
+++ python/branches/py3k/Modules/zlib/inflate.h	Fri Jul 30 22:03:17 2010
@@ -1,5 +1,5 @@
 /* inflate.h -- internal inflate state definition
- * Copyright (C) 1995-2004 Mark Adler
+ * Copyright (C) 1995-2009 Mark Adler
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -32,11 +32,13 @@
         TYPE,       /* i: waiting for type bits, including last-flag bit */
         TYPEDO,     /* i: same, but skip check to exit inflate on new block */
         STORED,     /* i: waiting for stored size (length and complement) */
+        COPY_,      /* i/o: same as COPY below, but only first time in */
         COPY,       /* i/o: waiting for input or output to copy stored block */
         TABLE,      /* i: waiting for dynamic block table lengths */
         LENLENS,    /* i: waiting for code length code lengths */
         CODELENS,   /* i: waiting for length/lit and distance code lengths */
-            LEN,        /* i: waiting for length/lit code */
+            LEN_,       /* i: same as LEN below, but only first time in */
+            LEN,        /* i: waiting for length/lit/eob code */
             LENEXT,     /* i: waiting for length extra bits */
             DIST,       /* i: waiting for distance code */
             DISTEXT,    /* i: waiting for distance extra bits */
@@ -53,19 +55,21 @@
 /*
     State transitions between above modes -
 
-    (most modes can go to the BAD or MEM mode -- not shown for clarity)
+    (most modes can go to BAD or MEM on error -- not shown for clarity)
 
     Process header:
-        HEAD -> (gzip) or (zlib)
-        (gzip) -> FLAGS -> TIME -> OS -> EXLEN -> EXTRA -> NAME
-        NAME -> COMMENT -> HCRC -> TYPE
+        HEAD -> (gzip) or (zlib) or (raw)
+        (gzip) -> FLAGS -> TIME -> OS -> EXLEN -> EXTRA -> NAME -> COMMENT ->
+                  HCRC -> TYPE
         (zlib) -> DICTID or TYPE
         DICTID -> DICT -> TYPE
+        (raw) -> TYPEDO
     Read deflate blocks:
-            TYPE -> STORED or TABLE or LEN or CHECK
-            STORED -> COPY -> TYPE
-            TABLE -> LENLENS -> CODELENS -> LEN
-    Read deflate codes:
+            TYPE -> TYPEDO -> STORED or TABLE or LEN_ or CHECK
+            STORED -> COPY_ -> COPY -> TYPE
+            TABLE -> LENLENS -> CODELENS -> LEN_
+            LEN_ -> LEN
+    Read deflate codes in fixed or dynamic block:
                 LEN -> LENEXT or LIT or TYPE
                 LENEXT -> DIST -> DISTEXT -> MATCH -> LEN
                 LIT -> LEN
@@ -73,7 +77,7 @@
         CHECK -> LENGTH -> DONE
  */
 
-/* state maintained between inflate() calls.  Approximately 7K bytes. */
+/* state maintained between inflate() calls.  Approximately 10K bytes. */
 struct inflate_state {
     inflate_mode mode;          /* current inflate mode */
     int last;                   /* true if processing last block */
@@ -88,7 +92,7 @@
     unsigned wbits;             /* log base 2 of requested window size */
     unsigned wsize;             /* window size or zero if not using window */
     unsigned whave;             /* valid bytes in the window */
-    unsigned write;             /* window write index */
+    unsigned wnext;             /* window write index */
     unsigned char FAR *window;  /* allocated sliding window, if needed */
         /* bit accumulator */
     unsigned long hold;         /* input bit accumulator */
@@ -112,4 +116,7 @@
     unsigned short lens[320];   /* temporary storage for code lengths */
     unsigned short work[288];   /* work area for code table building */
     code codes[ENOUGH];         /* space for code tables */
+    int sane;                   /* if false, allow invalid distance too far */
+    int back;                   /* bits back of last unprocessed length/lit */
+    unsigned was;               /* initial length of match */
 };

Modified: python/branches/py3k/Modules/zlib/inftrees.c
==============================================================================
--- python/branches/py3k/Modules/zlib/inftrees.c	(original)
+++ python/branches/py3k/Modules/zlib/inftrees.c	Fri Jul 30 22:03:17 2010
@@ -1,5 +1,5 @@
 /* inftrees.c -- generate Huffman trees for efficient decoding
- * Copyright (C) 1995-2005 Mark Adler
+ * Copyright (C) 1995-2010 Mark Adler
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -9,7 +9,7 @@
 #define MAXBITS 15
 
 const char inflate_copyright[] =
-   " inflate 1.2.3 Copyright 1995-2005 Mark Adler ";
+   " inflate 1.2.5 Copyright 1995-2010 Mark Adler ";
 /*
   If you use the zlib library in a product, an acknowledgment is welcome
   in the documentation of your product. If for some reason you cannot
@@ -29,7 +29,7 @@
    table index bits.  It will differ if the request is greater than the
    longest code or if it is less than the shortest code.
  */
-int inflate_table(type, lens, codes, table, bits, work)
+int ZLIB_INTERNAL inflate_table(type, lens, codes, table, bits, work)
 codetype type;
 unsigned short FAR *lens;
 unsigned codes;
@@ -50,7 +50,7 @@
     unsigned fill;              /* index for replicating entries */
     unsigned low;               /* low bits for current root entry */
     unsigned mask;              /* mask for low root bits */
-    code this;                  /* table entry for duplication */
+    code here;                  /* table entry for duplication */
     code FAR *next;             /* next available space in table */
     const unsigned short FAR *base;     /* base value table to use */
     const unsigned short FAR *extra;    /* extra bits table to use */
@@ -62,7 +62,7 @@
         35, 43, 51, 59, 67, 83, 99, 115, 131, 163, 195, 227, 258, 0, 0};
     static const unsigned short lext[31] = { /* Length codes 257..285 extra */
         16, 16, 16, 16, 16, 16, 16, 16, 17, 17, 17, 17, 18, 18, 18, 18,
-        19, 19, 19, 19, 20, 20, 20, 20, 21, 21, 21, 21, 16, 201, 196};
+        19, 19, 19, 19, 20, 20, 20, 20, 21, 21, 21, 21, 16, 73, 195};
     static const unsigned short dbase[32] = { /* Distance codes 0..29 base */
         1, 2, 3, 4, 5, 7, 9, 13, 17, 25, 33, 49, 65, 97, 129, 193,
         257, 385, 513, 769, 1025, 1537, 2049, 3073, 4097, 6145,
@@ -115,15 +115,15 @@
         if (count[max] != 0) break;
     if (root > max) root = max;
     if (max == 0) {                     /* no symbols to code at all */
-        this.op = (unsigned char)64;    /* invalid code marker */
-        this.bits = (unsigned char)1;
-        this.val = (unsigned short)0;
-        *(*table)++ = this;             /* make a table to force an error */
-        *(*table)++ = this;
+        here.op = (unsigned char)64;    /* invalid code marker */
+        here.bits = (unsigned char)1;
+        here.val = (unsigned short)0;
+        *(*table)++ = here;             /* make a table to force an error */
+        *(*table)++ = here;
         *bits = 1;
         return 0;     /* no symbols, but wait for decoding to report error */
     }
-    for (min = 1; min <= MAXBITS; min++)
+    for (min = 1; min < max; min++)
         if (count[min] != 0) break;
     if (root < min) root = min;
 
@@ -166,11 +166,10 @@
        entered in the tables.
 
        used keeps track of how many table entries have been allocated from the
-       provided *table space.  It is checked when a LENS table is being made
-       against the space in *table, ENOUGH, minus the maximum space needed by
-       the worst case distance code, MAXD.  This should never happen, but the
-       sufficiency of ENOUGH has not been proven exhaustively, hence the check.
-       This assumes that when type == LENS, bits == 9.
+       provided *table space.  It is checked for LENS and DIST tables against
+       the constants ENOUGH_LENS and ENOUGH_DISTS to guard against changes in
+       the initial root table size constants.  See the comments in inftrees.h
+       for more information.
 
        sym increments through all symbols, and the loop terminates when
        all codes of length max, i.e. all codes, have been processed.  This
@@ -209,24 +208,25 @@
     mask = used - 1;            /* mask for comparing low */
 
     /* check available table space */
-    if (type == LENS && used >= ENOUGH - MAXD)
+    if ((type == LENS && used >= ENOUGH_LENS) ||
+        (type == DISTS && used >= ENOUGH_DISTS))
         return 1;
 
     /* process all codes and make table entries */
     for (;;) {
         /* create table entry */
-        this.bits = (unsigned char)(len - drop);
+        here.bits = (unsigned char)(len - drop);
         if ((int)(work[sym]) < end) {
-            this.op = (unsigned char)0;
-            this.val = work[sym];
+            here.op = (unsigned char)0;
+            here.val = work[sym];
         }
         else if ((int)(work[sym]) > end) {
-            this.op = (unsigned char)(extra[work[sym]]);
-            this.val = base[work[sym]];
+            here.op = (unsigned char)(extra[work[sym]]);
+            here.val = base[work[sym]];
         }
         else {
-            this.op = (unsigned char)(32 + 64);         /* end of block */
-            this.val = 0;
+            here.op = (unsigned char)(32 + 64);         /* end of block */
+            here.val = 0;
         }
 
         /* replicate for those indices with low len bits equal to huff */
@@ -235,7 +235,7 @@
         min = fill;                 /* save offset to next table */
         do {
             fill -= incr;
-            next[(huff >> drop) + fill] = this;
+            next[(huff >> drop) + fill] = here;
         } while (fill != 0);
 
         /* backwards increment the len-bit code huff */
@@ -277,7 +277,8 @@
 
             /* check for enough space */
             used += 1U << curr;
-            if (type == LENS && used >= ENOUGH - MAXD)
+            if ((type == LENS && used >= ENOUGH_LENS) ||
+                (type == DISTS && used >= ENOUGH_DISTS))
                 return 1;
 
             /* point entry in root table to sub-table */
@@ -295,20 +296,20 @@
        through high index bits.  When the current sub-table is filled, the loop
        drops back to the root table to fill in any remaining entries there.
      */
-    this.op = (unsigned char)64;                /* invalid code marker */
-    this.bits = (unsigned char)(len - drop);
-    this.val = (unsigned short)0;
+    here.op = (unsigned char)64;                /* invalid code marker */
+    here.bits = (unsigned char)(len - drop);
+    here.val = (unsigned short)0;
     while (huff != 0) {
         /* when done with sub-table, drop back to root table */
         if (drop != 0 && (huff & mask) != low) {
             drop = 0;
             len = root;
             next = *table;
-            this.bits = (unsigned char)len;
+            here.bits = (unsigned char)len;
         }
 
         /* put invalid code marker in table */
-        next[huff >> drop] = this;
+        next[huff >> drop] = here;
 
         /* backwards increment the len-bit code huff */
         incr = 1U << (len - 1);

Modified: python/branches/py3k/Modules/zlib/inftrees.h
==============================================================================
--- python/branches/py3k/Modules/zlib/inftrees.h	(original)
+++ python/branches/py3k/Modules/zlib/inftrees.h	Fri Jul 30 22:03:17 2010
@@ -1,5 +1,5 @@
 /* inftrees.h -- header to use inftrees.c
- * Copyright (C) 1995-2005 Mark Adler
+ * Copyright (C) 1995-2005, 2010 Mark Adler
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -35,21 +35,28 @@
     01000000 - invalid code
  */
 
-/* Maximum size of dynamic tree.  The maximum found in a long but non-
-   exhaustive search was 1444 code structures (852 for length/literals
-   and 592 for distances, the latter actually the result of an
-   exhaustive search).  The true maximum is not known, but the value
-   below is more than safe. */
-#define ENOUGH 2048
-#define MAXD 592
+/* Maximum size of the dynamic table.  The maximum number of code structures is
+   1444, which is the sum of 852 for literal/length codes and 592 for distance
+   codes.  These values were found by exhaustive searches using the program
+   examples/enough.c found in the zlib distribtution.  The arguments to that
+   program are the number of symbols, the initial root table size, and the
+   maximum bit length of a code.  "enough 286 9 15" for literal/length codes
+   returns returns 852, and "enough 30 6 15" for distance codes returns 592.
+   The initial root table size (9 or 6) is found in the fifth argument of the
+   inflate_table() calls in inflate.c and infback.c.  If the root table size is
+   changed, then these maximum sizes would be need to be recalculated and
+   updated. */
+#define ENOUGH_LENS 852
+#define ENOUGH_DISTS 592
+#define ENOUGH (ENOUGH_LENS+ENOUGH_DISTS)
 
-/* Type of code to build for inftable() */
+/* Type of code to build for inflate_table() */
 typedef enum {
     CODES,
     LENS,
     DISTS
 } codetype;
 
-extern int inflate_table OF((codetype type, unsigned short FAR *lens,
+int ZLIB_INTERNAL inflate_table OF((codetype type, unsigned short FAR *lens,
                              unsigned codes, code FAR * FAR *table,
                              unsigned FAR *bits, unsigned short FAR *work));

Modified: python/branches/py3k/Modules/zlib/make_vms.com
==============================================================================
--- python/branches/py3k/Modules/zlib/make_vms.com	(original)
+++ python/branches/py3k/Modules/zlib/make_vms.com	Fri Jul 30 22:03:17 2010
@@ -1,33 +1,57 @@
 $! make libz under VMS written by
 $! Martin P.J. Zinser
-$! <zinser at zinser.no-ip.info or zinser at sysdev.deutsche-boerse.com>
 $!
-$ on error then goto err_exit
+$! In case of problems with the install you might contact me at
+$! zinser at zinser.no-ip.info(preferred) or
+$! zinser at sysdev.deutsche-boerse.com (work)
+$!
+$! Make procedure history for Zlib
 $!
+$!------------------------------------------------------------------------------
+$! Version history
+$! 0.01 20060120 First version to receive a number
+$! 0.02 20061008 Adapt to new Makefile.in
+$! 0.03 20091224 Add support for large file check
+$! 0.04 20100110 Add new gzclose, gzlib, gzread, gzwrite
+$! 0.05 20100221 Exchange zlibdefs.h by zconf.h.in
 $!
-$! Just some general constants...
+$ on error then goto err_exit
+$ set proc/parse=ext
 $!
 $ true  = 1
 $ false = 0
 $ tmpnam = "temp_" + f$getjpi("","pid")
-$ SAY = "WRITE SYS$OUTPUT"
+$ tt = tmpnam + ".txt"
+$ tc = tmpnam + ".c"
+$ th = tmpnam + ".h"
+$ define/nolog tconfig 'th'
+$ its_decc = false
+$ its_vaxc = false
+$ its_gnuc = false
+$ s_case   = False
 $!
 $! Setup variables holding "config" information
 $!
-$ Make     = ""
+$ Make    = ""
 $ name     = "Zlib"
 $ version  = "?.?.?"
 $ v_string = "ZLIB_VERSION"
 $ v_file   = "zlib.h"
-$ ccopt    = ""
-$ lopts    = ""
+$ ccopt   = ""
+$ lopts   = ""
+$ dnsrl   = ""
+$ aconf_in_file = "zconf.h.in#zconf.h_in"
+$ conf_check_string = ""
 $ linkonly = false
 $ optfile  = name + ".opt"
-$ its_decc = false
-$ its_vaxc = false
-$ its_gnuc = false
-$ axp      = f$getsyi("HW_MODEL").ge.1024
-$ s_case   = false
+$ libdefs  = ""
+$ axp      = f$getsyi("HW_MODEL").ge.1024 .and. f$getsyi("HW_MODEL").lt.4096
+$!
+$ whoami = f$parse(f$enviornment("Procedure"),,,,"NO_CONCEAL")
+$ mydef  = F$parse(whoami,,,"DEVICE")
+$ mydir  = f$parse(whoami,,,"DIRECTORY") - "]["
+$ myproc = f$parse(whoami,,,"Name") + f$parse(whoami,,,"type")
+$!
 $! Check for MMK/MMS
 $!
 $ If F$Search ("Sys$System:MMS.EXE") .nes. "" Then Make = "MMS"
@@ -36,11 +60,16 @@
 $!
 $ gosub find_version
 $!
+$  open/write topt tmp.opt
+$  open/write optf 'optfile'
+$!
 $ gosub check_opts
 $!
 $! Look for the compiler used
 $!
 $ gosub check_compiler
+$ close topt
+$!
 $ if its_decc
 $ then
 $   ccopt = "/prefix=all" + ccopt
@@ -60,6 +89,52 @@
 $    if f$trnlnm("SYS").eqs."" then define sys sys$library:
 $ endif
 $!
+$! Build a fake configure input header
+$!
+$ open/write conf_hin config.hin
+$ write conf_hin "#undef _LARGEFILE64_SOURCE"
+$ close conf_hin
+$!
+$!
+$ i = 0
+$FIND_ACONF:
+$ fname = f$element(i,"#",aconf_in_file)
+$ if fname .eqs. "#" then goto AMISS_ERR
+$ if f$search(fname) .eqs. ""
+$ then
+$   i = i + 1
+$   goto find_aconf
+$ endif
+$ open/read/err=aconf_err aconf_in 'fname'
+$ open/write aconf zconf.h
+$ACONF_LOOP:
+$ read/end_of_file=aconf_exit aconf_in line
+$ work = f$edit(line, "compress,trim")
+$ if f$extract(0,6,work) .nes. "#undef"
+$ then
+$   if f$extract(0,12,work) .nes. "#cmakedefine"
+$   then
+$       write aconf line
+$   endif
+$ else
+$   cdef = f$element(1," ",work)
+$   gosub check_config
+$ endif
+$ goto aconf_loop
+$ACONF_EXIT:
+$ write aconf "#define VMS 1"
+$ write aconf "#include <unistd.h>"
+$ write aconf "#include <unixio.h>"
+$ write aconf "#ifdef _LARGEFILE"
+$ write aconf "#define off64_t __off64_t"
+$ write aconf "#define fopen64 fopen"
+$ write aconf "#define fseeko64 fseeko"
+$ write aconf "#define lseek64 lseek"
+$ write aconf "#define ftello64 ftell"
+$ write aconf "#endif"
+$ close aconf_in
+$ close aconf
+$ if f$search("''th'") .nes. "" then delete 'th';*
 $! Build the thing plain or with mms
 $!
 $ write sys$output "Compiling Zlib sources ..."
@@ -74,8 +149,14 @@
                 crc32.c zlib.h zconf.h
 $   CALL MAKE deflate.OBJ "CC ''CCOPT' deflate" -
                 deflate.c deflate.h zutil.h zlib.h zconf.h
-$   CALL MAKE gzio.OBJ "CC ''CCOPT' gzio" -
-                gzio.c zutil.h zlib.h zconf.h
+$   CALL MAKE gzclose.OBJ "CC ''CCOPT' gzclose" -
+                gzclose.c zutil.h zlib.h zconf.h
+$   CALL MAKE gzlib.OBJ "CC ''CCOPT' gzlib" -
+                gzlib.c zutil.h zlib.h zconf.h
+$   CALL MAKE gzread.OBJ "CC ''CCOPT' gzread" -
+                gzread.c zutil.h zlib.h zconf.h
+$   CALL MAKE gzwrite.OBJ "CC ''CCOPT' gzwrite" -
+                gzwrite.c zutil.h zlib.h zconf.h
 $   CALL MAKE infback.OBJ "CC ''CCOPT' infback" -
                 infback.c zutil.h inftrees.h inflate.h inffast.h inffixed.h
 $   CALL MAKE inffast.OBJ "CC ''CCOPT' inffast" -
@@ -107,7 +188,7 @@
 $   endif
 $  else
 $   gosub crea_mms
-$   SAY "Make ''name' ''version' with ''Make' "
+$   write sys$output "Make ''name' ''version' with ''Make' "
 $   'make'
 $  endif
 $!
@@ -133,6 +214,15 @@
 $ goto err_exit
 $ERR_EXIT:
 $ set message/facil/ident/sever/text
+$ close/nolog optf
+$ close/nolog topt
+$ close/nolog conf_hin
+$ close/nolog aconf_in
+$ close/nolog aconf
+$ close/nolog out
+$ close/nolog min
+$ close/nolog mod
+$ close/nolog h_in
 $ write sys$output "Exiting..."
 $ exit 2
 $!
@@ -180,61 +270,72 @@
 $!
 $! Check command line options and set symbols accordingly
 $!
+$!------------------------------------------------------------------------------
+$! Version history
+$! 0.01 20041206 First version to receive a number
+$! 0.02 20060126 Add new "HELP" target
 $ CHECK_OPTS:
 $ i = 1
 $ OPT_LOOP:
 $ if i .lt. 9
 $ then
 $   cparm = f$edit(p'i',"upcase")
-$   if cparm .eqs. "DEBUG"
-$   then
-$     ccopt = ccopt + "/noopt/deb"
-$     lopts = lopts + "/deb"
-$   endif
-$   if f$locate("CCOPT=",cparm) .lt. f$length(cparm)
-$   then
-$     start = f$locate("=",cparm) + 1
-$     len   = f$length(cparm) - start
-$     ccopt = ccopt + f$extract(start,len,cparm)
-$     if f$locate("AS_IS",f$edit(ccopt,"UPCASE")) .lt. f$length(ccopt) -
-         then s_case = true
-$   endif
-$   if cparm .eqs. "LINK" then linkonly = true
-$   if f$locate("LOPTS=",cparm) .lt. f$length(cparm)
-$   then
-$     start = f$locate("=",cparm) + 1
-$     len   = f$length(cparm) - start
-$     lopts = lopts + f$extract(start,len,cparm)
-$   endif
-$   if f$locate("CC=",cparm) .lt. f$length(cparm)
+$!
+$! Check if parameter actually contains something
+$!
+$   if f$edit(cparm,"trim") .nes. ""
 $   then
-$     start  = f$locate("=",cparm) + 1
-$     len    = f$length(cparm) - start
-$     cc_com = f$extract(start,len,cparm)
-      if (cc_com .nes. "DECC") .and. -
-         (cc_com .nes. "VAXC") .and. -
-	 (cc_com .nes. "GNUC")
+$     if cparm .eqs. "DEBUG"
 $     then
-$       write sys$output "Unsupported compiler choice ''cc_com' ignored"
-$       write sys$output "Use DECC, VAXC, or GNUC instead"
-$     else
-$     	if cc_com .eqs. "DECC" then its_decc = true
-$     	if cc_com .eqs. "VAXC" then its_vaxc = true
-$     	if cc_com .eqs. "GNUC" then its_gnuc = true
+$       ccopt = ccopt + "/noopt/deb"
+$       lopts = lopts + "/deb"
 $     endif
-$   endif
-$   if f$locate("MAKE=",cparm) .lt. f$length(cparm)
-$   then
-$     start  = f$locate("=",cparm) + 1
-$     len    = f$length(cparm) - start
-$     mmks = f$extract(start,len,cparm)
-$     if (mmks .eqs. "MMK") .or. (mmks .eqs. "MMS")
+$     if f$locate("CCOPT=",cparm) .lt. f$length(cparm)
 $     then
-$       make = mmks
-$     else
-$       write sys$output "Unsupported make choice ''mmks' ignored"
-$       write sys$output "Use MMK or MMS instead"
+$       start = f$locate("=",cparm) + 1
+$       len   = f$length(cparm) - start
+$       ccopt = ccopt + f$extract(start,len,cparm)
+$       if f$locate("AS_IS",f$edit(ccopt,"UPCASE")) .lt. f$length(ccopt) -
+          then s_case = true
 $     endif
+$     if cparm .eqs. "LINK" then linkonly = true
+$     if f$locate("LOPTS=",cparm) .lt. f$length(cparm)
+$     then
+$       start = f$locate("=",cparm) + 1
+$       len   = f$length(cparm) - start
+$       lopts = lopts + f$extract(start,len,cparm)
+$     endif
+$     if f$locate("CC=",cparm) .lt. f$length(cparm)
+$     then
+$       start  = f$locate("=",cparm) + 1
+$       len    = f$length(cparm) - start
+$       cc_com = f$extract(start,len,cparm)
+        if (cc_com .nes. "DECC") .and. -
+           (cc_com .nes. "VAXC") .and. -
+           (cc_com .nes. "GNUC")
+$       then
+$         write sys$output "Unsupported compiler choice ''cc_com' ignored"
+$         write sys$output "Use DECC, VAXC, or GNUC instead"
+$       else
+$         if cc_com .eqs. "DECC" then its_decc = true
+$         if cc_com .eqs. "VAXC" then its_vaxc = true
+$         if cc_com .eqs. "GNUC" then its_gnuc = true
+$       endif
+$     endif
+$     if f$locate("MAKE=",cparm) .lt. f$length(cparm)
+$     then
+$       start  = f$locate("=",cparm) + 1
+$       len    = f$length(cparm) - start
+$       mmks = f$extract(start,len,cparm)
+$       if (mmks .eqs. "MMK") .or. (mmks .eqs. "MMS")
+$       then
+$         make = mmks
+$       else
+$         write sys$output "Unsupported make choice ''mmks' ignored"
+$         write sys$output "Use MMK or MMS instead"
+$       endif
+$     endif
+$     if cparm .eqs. "HELP" then gosub bhelp
 $   endif
 $   i = i + 1
 $   goto opt_loop
@@ -244,6 +345,11 @@
 $!
 $! Look for the compiler used
 $!
+$! Version history
+$! 0.01 20040223 First version to receive a number
+$! 0.02 20040229 Save/set value of decc$no_rooted_search_lists
+$! 0.03 20060202 Extend handling of GNU C
+$! 0.04 20090402 Compaq -> hp
 $CHECK_COMPILER:
 $ if (.not. (its_decc .or. its_vaxc .or. its_gnuc))
 $ then
@@ -257,9 +363,26 @@
 $ if (.not. (its_decc .or. its_vaxc .or. its_gnuc))
 $ then goto CC_ERR
 $ else
-$   if its_decc then write sys$output "CC compiler check ... Compaq C"
-$   if its_vaxc then write sys$output "CC compiler check ... VAX C"
-$   if its_gnuc then write sys$output "CC compiler check ... GNU C"
+$   if its_decc
+$   then
+$     write sys$output "CC compiler check ... hp C"
+$     if f$trnlnm("decc$no_rooted_search_lists") .nes. ""
+$     then
+$       dnrsl = f$trnlnm("decc$no_rooted_search_lists")
+$     endif
+$     define/nolog decc$no_rooted_search_lists 1
+$   else
+$     if its_vaxc then write sys$output "CC compiler check ... VAX C"
+$     if its_gnuc
+$     then
+$         write sys$output "CC compiler check ... GNU C"
+$         if f$trnlnm(topt) then write topt "gnu_cc:[000000]gcclib.olb/lib"
+$         if f$trnlnm(optf) then write optf "gnu_cc:[000000]gcclib.olb/lib"
+$         cc = "gcc"
+$     endif
+$     if f$trnlnm(topt) then write topt "sys$share:vaxcrtl.exe/share"
+$     if f$trnlnm(optf) then write optf "sys$share:vaxcrtl.exe/share"
+$   endif
 $ endif
 $ return
 $!------------------------------------------------------------------------------
@@ -276,7 +399,8 @@
 # written by Martin P.J. Zinser
 # <zinser at zinser.no-ip.info or zinser at sysdev.deutsche-boerse.com>
 
-OBJS = adler32.obj, compress.obj, crc32.obj, gzio.obj, uncompr.obj, infback.obj\
+OBJS = adler32.obj, compress.obj, crc32.obj, gzclose.obj, gzlib.obj\
+       gzread.obj, gzwrite.obj, uncompr.obj, infback.obj\
        deflate.obj, trees.obj, zutil.obj, inflate.obj, \
        inftrees.obj, inffast.obj
 
@@ -308,7 +432,10 @@
 crc32.obj    : crc32.c zutil.h zlib.h zconf.h
 deflate.obj  : deflate.c deflate.h zutil.h zlib.h zconf.h
 example.obj  : example.c zlib.h zconf.h
-gzio.obj     : gzio.c zutil.h zlib.h zconf.h
+gzclose.obj  : gzclose.c zutil.h zlib.h zconf.h
+gzlib.obj    : gzlib.c zutil.h zlib.h zconf.h
+gzread.obj   : gzread.c zutil.h zlib.h zconf.h
+gzwrite.obj  : gzwrite.c zutil.h zlib.h zconf.h
 inffast.obj  : inffast.c zutil.h zlib.h zconf.h inftrees.h inffast.h
 inflate.obj  : inflate.c zutil.h zlib.h zconf.h
 inftrees.obj : inftrees.c zutil.h zlib.h zconf.h inftrees.h
@@ -328,7 +455,7 @@
 $CREA_OLIST:
 $ open/read min makefile.in
 $ open/write mod modules.opt
-$ src_check = "OBJS ="
+$ src_check = "OBJC ="
 $MRLOOP:
 $ read/end=mrdone min rec
 $ if (f$extract(0,6,rec) .nes. src_check) then goto mrloop
@@ -382,13 +509,182 @@
 $ return
 $!------------------------------------------------------------------------------
 $!
+$CHECK_CONFIG:
+$!
+$ in_ldef = f$locate(cdef,libdefs)
+$ if (in_ldef .lt. f$length(libdefs))
+$ then
+$   write aconf "#define ''cdef' 1"
+$   libdefs = f$extract(0,in_ldef,libdefs) + -
+              f$extract(in_ldef + f$length(cdef) + 1, -
+                        f$length(libdefs) - in_ldef - f$length(cdef) - 1, -
+                        libdefs)
+$ else
+$   if (f$type('cdef') .eqs. "INTEGER")
+$   then
+$     write aconf "#define ''cdef' ", 'cdef'
+$   else
+$     if (f$type('cdef') .eqs. "STRING")
+$     then
+$       write aconf "#define ''cdef' ", """", '''cdef'', """"
+$     else
+$       gosub check_cc_def
+$     endif
+$   endif
+$ endif
+$ return
+$!------------------------------------------------------------------------------
+$!
+$! Check if this is a define relating to the properties of the C/C++
+$! compiler
+$!
+$ CHECK_CC_DEF:
+$ if (cdef .eqs. "_LARGEFILE64_SOURCE")
+$ then
+$   copy sys$input: 'tc'
+$   deck
+#include "tconfig"
+#define _LARGEFILE
+#include <stdio.h>
+
+int main(){
+FILE *fp;
+  fp = fopen("temp.txt","r");
+  fseeko(fp,1,SEEK_SET);
+  fclose(fp);
+}
+
+$   eod
+$   test_inv = false
+$   comm_h = false
+$   gosub cc_prop_check
+$   return
+$ endif
+$ write aconf "/* ", line, " */"
+$ return
+$!------------------------------------------------------------------------------
+$!
+$! Check for properties of C/C++ compiler
+$!
+$! Version history
+$! 0.01 20031020 First version to receive a number
+$! 0.02 20031022 Added logic for defines with value
+$! 0.03 20040309 Make sure local config file gets not deleted
+$! 0.04 20041230 Also write include for configure run
+$! 0.05 20050103 Add processing of "comment defines"
+$CC_PROP_CHECK:
+$ cc_prop = true
+$ is_need = false
+$ is_need = (f$extract(0,4,cdef) .eqs. "NEED") .or. (test_inv .eq. true)
+$ if f$search(th) .eqs. "" then create 'th'
+$ set message/nofac/noident/nosever/notext
+$ on error then continue
+$ cc 'tmpnam'
+$ if .not. ($status)  then cc_prop = false
+$ on error then continue
+$! The headers might lie about the capabilities of the RTL
+$ link 'tmpnam',tmp.opt/opt
+$ if .not. ($status)  then cc_prop = false
+$ set message/fac/ident/sever/text
+$ on error then goto err_exit
+$ delete/nolog 'tmpnam'.*;*/exclude='th'
+$ if (cc_prop .and. .not. is_need) .or. -
+     (.not. cc_prop .and. is_need)
+$ then
+$   write sys$output "Checking for ''cdef'... yes"
+$   if f$type('cdef_val'_yes) .nes. ""
+$   then
+$     if f$type('cdef_val'_yes) .eqs. "INTEGER" -
+         then call write_config f$fao("#define !AS !UL",cdef,'cdef_val'_yes)
+$     if f$type('cdef_val'_yes) .eqs. "STRING" -
+         then call write_config f$fao("#define !AS !AS",cdef,'cdef_val'_yes)
+$   else
+$     call write_config f$fao("#define !AS 1",cdef)
+$   endif
+$   if (cdef .eqs. "HAVE_FSEEKO") .or. (cdef .eqs. "_LARGE_FILES") .or. -
+       (cdef .eqs. "_LARGEFILE64_SOURCE") then -
+      call write_config f$string("#define _LARGEFILE 1")
+$ else
+$   write sys$output "Checking for ''cdef'... no"
+$   if (comm_h)
+$   then
+      call write_config f$fao("/* !AS */",line)
+$   else
+$     if f$type('cdef_val'_no) .nes. ""
+$     then
+$       if f$type('cdef_val'_no) .eqs. "INTEGER" -
+           then call write_config f$fao("#define !AS !UL",cdef,'cdef_val'_no)
+$       if f$type('cdef_val'_no) .eqs. "STRING" -
+           then call write_config f$fao("#define !AS !AS",cdef,'cdef_val'_no)
+$     else
+$       call write_config f$fao("#undef !AS",cdef)
+$     endif
+$   endif
+$ endif
+$ return
+$!------------------------------------------------------------------------------
+$!
+$! Check for properties of C/C++ compiler with multiple result values
+$!
+$! Version history
+$! 0.01 20040127 First version
+$! 0.02 20050103 Reconcile changes from cc_prop up to version 0.05
+$CC_MPROP_CHECK:
+$ cc_prop = true
+$ i    = 1
+$ idel = 1
+$ MT_LOOP:
+$ if f$type(result_'i') .eqs. "STRING"
+$ then
+$   set message/nofac/noident/nosever/notext
+$   on error then continue
+$   cc 'tmpnam'_'i'
+$   if .not. ($status)  then cc_prop = false
+$   on error then continue
+$! The headers might lie about the capabilities of the RTL
+$   link 'tmpnam'_'i',tmp.opt/opt
+$   if .not. ($status)  then cc_prop = false
+$   set message/fac/ident/sever/text
+$   on error then goto err_exit
+$   delete/nolog 'tmpnam'_'i'.*;*
+$   if (cc_prop)
+$   then
+$     write sys$output "Checking for ''cdef'... ", mdef_'i'
+$     if f$type(mdef_'i') .eqs. "INTEGER" -
+         then call write_config f$fao("#define !AS !UL",cdef,mdef_'i')
+$     if f$type('cdef_val'_yes) .eqs. "STRING" -
+         then call write_config f$fao("#define !AS !AS",cdef,mdef_'i')
+$     goto msym_clean
+$   else
+$     i = i + 1
+$     goto mt_loop
+$   endif
+$ endif
+$ write sys$output "Checking for ''cdef'... no"
+$ call write_config f$fao("#undef !AS",cdef)
+$ MSYM_CLEAN:
+$ if (idel .le. msym_max)
+$ then
+$   delete/sym mdef_'idel'
+$   idel = idel + 1
+$   goto msym_clean
+$ endif
+$ return
+$!------------------------------------------------------------------------------
+$!
 $! Analyze Object files for OpenVMS AXP to extract Procedure and Data
 $! information to build a symbol vector for a shareable image
 $! All the "brains" of this logic was suggested by Hartmut Becker
 $! (Hartmut.Becker at compaq.com). All the bugs were introduced by me
-$! (zinser at decus.de), so if you do have problem reports please do not
+$! (zinser at zinser.no-ip.info), so if you do have problem reports please do not
 $! bother Hartmut/HP, but get in touch with me
 $!
+$! Version history
+$! 0.01 20040406 Skip over shareable images in option file
+$! 0.02 20041109 Fix option file for shareable images with case_sensitive=YES
+$! 0.03 20050107 Skip over Identification labels in option file
+$! 0.04 20060117 Add uppercase alias to code compiled with /name=as_is
+$!
 $ ANAL_OBJ_AXP: Subroutine
 $ V = 'F$Verify(0)
 $ SAY := "WRITE_ SYS$OUTPUT"
@@ -409,6 +705,17 @@
 $ open/append atmp a.tmp
 $ loop:
 $ read/end=end_loop in line
+$ if f$locate("/SHARE",f$edit(line,"upcase")) .lt. f$length(line)
+$ then
+$   write sys$output "ANAL_SKP_SHR-i-skipshare, ''line'"
+$   goto loop
+$ endif
+$ if f$locate("IDENTIFICATION=",f$edit(line,"upcase")) .lt. f$length(line)
+$ then
+$   write sys$output "ANAL_OBJ_AXP-i-ident: Identification ", -
+                     f$element(1,"=",line)
+$   goto loop
+$ endif
 $ f= f$search(line)
 $ if f .eqs. ""
 $ then
@@ -450,8 +757,31 @@
 sub/symbol: "/symbol_vector=(/whole
 sub/"/=DATA)/whole
 exit
-$ sort/nodupl d.tmp,f.tmp 'p2'
-$ delete a.tmp;*,b.tmp;*,c.tmp;*,d.tmp;*,e.tmp;*,f.tmp;*
+$ sort/nodupl d.tmp,f.tmp g.tmp
+$ open/read raw_vector g.tmp
+$ open/write case_vector 'p2'
+$ RAWLOOP:
+$ read/end=end_rawloop raw_vector raw_element
+$ write case_vector raw_element
+$ if f$locate("=PROCEDURE)",raw_element) .lt. f$length(raw_element)
+$ then
+$     name = f$element(1,"=",raw_element) - "("
+$     if f$edit(name,"UPCASE") .nes. name then -
+          write case_vector f$fao(" symbol_vector=(!AS/!AS=PROCEDURE)", -
+	                          f$edit(name,"UPCASE"), name)
+$ endif
+$ if f$locate("=DATA)",raw_element) .lt. f$length(raw_element)
+$ then
+$     name = f$element(1,"=",raw_element) - "("
+$     if f$edit(name,"UPCASE") .nes. name then -
+          write case_vector f$fao(" symbol_vector=(!AS/!AS=DATA)", -
+	                          f$edit(name,"UPCASE"), name)
+$ endif
+$ goto rawloop
+$ END_RAWLOOP:
+$ close raw_vector
+$ close case_vector
+$ delete a.tmp;*,b.tmp;*,c.tmp;*,d.tmp;*,e.tmp;*,f.tmp;*,g.tmp;*
 $ if f$search("x.tmp") .nes. "" -
 	then $ delete x.tmp;*
 $!
@@ -459,3 +789,16 @@
 $ if V then set verify
 $ endsubroutine
 $!------------------------------------------------------------------------------
+$!
+$! Write configuration to both permanent and temporary config file
+$!
+$! Version history
+$! 0.01 20031029 First version to receive a number
+$!
+$WRITE_CONFIG: SUBROUTINE
+$  write aconf 'p1'
+$  open/append confh 'th'
+$  write confh 'p1'
+$  close confh
+$ENDSUBROUTINE
+$!------------------------------------------------------------------------------

Modified: python/branches/py3k/Modules/zlib/minigzip.c
==============================================================================
--- python/branches/py3k/Modules/zlib/minigzip.c	(original)
+++ python/branches/py3k/Modules/zlib/minigzip.c	Fri Jul 30 22:03:17 2010
@@ -1,5 +1,5 @@
 /* minigzip.c -- simulate gzip using the zlib compression library
- * Copyright (C) 1995-2005 Jean-loup Gailly.
+ * Copyright (C) 1995-2006, 2010 Jean-loup Gailly.
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -15,8 +15,8 @@
 
 /* @(#) $Id$ */
 
-#include <stdio.h>
 #include "zlib.h"
+#include <stdio.h>
 
 #ifdef STDC
 #  include <string.h>
@@ -32,6 +32,9 @@
 #if defined(MSDOS) || defined(OS2) || defined(WIN32) || defined(__CYGWIN__)
 #  include <fcntl.h>
 #  include <io.h>
+#  ifdef UNDER_CE
+#    include <stdlib.h>
+#  endif
 #  define SET_BINARY_MODE(file) setmode(fileno(file), O_BINARY)
 #else
 #  define SET_BINARY_MODE(file)
@@ -41,13 +44,84 @@
 #  define unlink delete
 #  define GZ_SUFFIX "-gz"
 #endif
+#ifdef RISCOS
+#  define unlink remove
+#  define GZ_SUFFIX "-gz"
+#  define fileno(file) file->__file
+#endif
 #if defined(__MWERKS__) && __dest_os != __be_os && __dest_os != __win32_os
 #  include <unix.h> /* for fileno */
 #endif
 
+#if !defined(Z_HAVE_UNISTD_H) && !defined(_LARGEFILE64_SOURCE)
 #ifndef WIN32 /* unlink already in stdio.h for WIN32 */
   extern int unlink OF((const char *));
 #endif
+#endif
+
+#if defined(UNDER_CE)
+#  include <windows.h>
+#  define perror(s) pwinerror(s)
+
+/* Map the Windows error number in ERROR to a locale-dependent error
+   message string and return a pointer to it.  Typically, the values
+   for ERROR come from GetLastError.
+
+   The string pointed to shall not be modified by the application,
+   but may be overwritten by a subsequent call to strwinerror
+
+   The strwinerror function does not change the current setting
+   of GetLastError.  */
+
+static char *strwinerror (error)
+     DWORD error;
+{
+    static char buf[1024];
+
+    wchar_t *msgbuf;
+    DWORD lasterr = GetLastError();
+    DWORD chars = FormatMessage(FORMAT_MESSAGE_FROM_SYSTEM
+        | FORMAT_MESSAGE_ALLOCATE_BUFFER,
+        NULL,
+        error,
+        0, /* Default language */
+        (LPVOID)&msgbuf,
+        0,
+        NULL);
+    if (chars != 0) {
+        /* If there is an \r\n appended, zap it.  */
+        if (chars >= 2
+            && msgbuf[chars - 2] == '\r' && msgbuf[chars - 1] == '\n') {
+            chars -= 2;
+            msgbuf[chars] = 0;
+        }
+
+        if (chars > sizeof (buf) - 1) {
+            chars = sizeof (buf) - 1;
+            msgbuf[chars] = 0;
+        }
+
+        wcstombs(buf, msgbuf, chars + 1);
+        LocalFree(msgbuf);
+    }
+    else {
+        sprintf(buf, "unknown win32 error (%ld)", error);
+    }
+
+    SetLastError(lasterr);
+    return buf;
+}
+
+static void pwinerror (s)
+    const char *s;
+{
+    if (s && *s)
+        fprintf(stderr, "%s: %s\n", s, strwinerror(GetLastError ()));
+    else
+        fprintf(stderr, "%s\n", strwinerror(GetLastError ()));
+}
+
+#endif /* UNDER_CE */
 
 #ifndef GZ_SUFFIX
 #  define GZ_SUFFIX ".gz"
@@ -193,6 +267,11 @@
     FILE  *in;
     gzFile out;
 
+    if (strlen(file) + strlen(GZ_SUFFIX) >= sizeof(outfile)) {
+        fprintf(stderr, "%s: filename too long\n", prog);
+        exit(1);
+    }
+
     strcpy(outfile, file);
     strcat(outfile, GZ_SUFFIX);
 
@@ -222,7 +301,12 @@
     char *infile, *outfile;
     FILE  *out;
     gzFile in;
-    uInt len = (uInt)strlen(file);
+    size_t len = strlen(file);
+
+    if (len + strlen(GZ_SUFFIX) >= sizeof(buf)) {
+        fprintf(stderr, "%s: filename too long\n", prog);
+        exit(1);
+    }
 
     strcpy(buf, file);
 
@@ -253,7 +337,8 @@
 
 
 /* ===========================================================================
- * Usage:  minigzip [-d] [-f] [-h] [-r] [-1 to -9] [files...]
+ * Usage:  minigzip [-c] [-d] [-f] [-h] [-r] [-1 to -9] [files...]
+ *   -c : write to standard output
  *   -d : decompress
  *   -f : compress with Z_FILTERED
  *   -h : compress with Z_HUFFMAN_ONLY
@@ -265,17 +350,30 @@
     int argc;
     char *argv[];
 {
+    int copyout = 0;
     int uncompr = 0;
     gzFile file;
-    char outmode[20];
+    char *bname, outmode[20];
 
     strcpy(outmode, "wb6 ");
 
     prog = argv[0];
+    bname = strrchr(argv[0], '/');
+    if (bname)
+      bname++;
+    else
+      bname = argv[0];
     argc--, argv++;
 
+    if (!strcmp(bname, "gunzip"))
+      uncompr = 1;
+    else if (!strcmp(bname, "zcat"))
+      copyout = uncompr = 1;
+
     while (argc > 0) {
-      if (strcmp(*argv, "-d") == 0)
+      if (strcmp(*argv, "-c") == 0)
+        copyout = 1;
+      else if (strcmp(*argv, "-d") == 0)
         uncompr = 1;
       else if (strcmp(*argv, "-f") == 0)
         outmode[3] = 'f';
@@ -305,11 +403,36 @@
             gz_compress(stdin, file);
         }
     } else {
+        if (copyout) {
+            SET_BINARY_MODE(stdout);
+        }
         do {
             if (uncompr) {
-                file_uncompress(*argv);
+                if (copyout) {
+                    file = gzopen(*argv, "rb");
+                    if (file == NULL)
+                        fprintf(stderr, "%s: can't gzopen %s\n", prog, *argv);
+                    else
+                        gz_uncompress(file, stdout);
+                } else {
+                    file_uncompress(*argv);
+                }
             } else {
-                file_compress(*argv, outmode);
+                if (copyout) {
+                    FILE * in = fopen(*argv, "rb");
+
+                    if (in == NULL) {
+                        perror(*argv);
+                    } else {
+                        file = gzdopen(fileno(stdout), outmode);
+                        if (file == NULL) error("can't gzdopen stdout");
+
+                        gz_compress(in, file);
+                    }
+
+                } else {
+                    file_compress(*argv, outmode);
+                }
             }
         } while (argv++, --argc);
     }

Modified: python/branches/py3k/Modules/zlib/trees.c
==============================================================================
--- python/branches/py3k/Modules/zlib/trees.c	(original)
+++ python/branches/py3k/Modules/zlib/trees.c	Fri Jul 30 22:03:17 2010
@@ -1,5 +1,6 @@
 /* trees.c -- output deflated data using Huffman coding
- * Copyright (C) 1995-2005 Jean-loup Gailly
+ * Copyright (C) 1995-2010 Jean-loup Gailly
+ * detect_data_type() function provided freely by Cosmin Truta, 2006
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -152,7 +153,7 @@
                               int blcodes));
 local void compress_block OF((deflate_state *s, ct_data *ltree,
                               ct_data *dtree));
-local void set_data_type  OF((deflate_state *s));
+local int  detect_data_type OF((deflate_state *s));
 local unsigned bi_reverse OF((unsigned value, int length));
 local void bi_windup      OF((deflate_state *s));
 local void bi_flush       OF((deflate_state *s));
@@ -203,12 +204,12 @@
      * unused bits in value.
      */
     if (s->bi_valid > (int)Buf_size - length) {
-        s->bi_buf |= (value << s->bi_valid);
+        s->bi_buf |= (ush)value << s->bi_valid;
         put_short(s, s->bi_buf);
         s->bi_buf = (ush)value >> (Buf_size - s->bi_valid);
         s->bi_valid += length - Buf_size;
     } else {
-        s->bi_buf |= value << s->bi_valid;
+        s->bi_buf |= (ush)value << s->bi_valid;
         s->bi_valid += length;
     }
 }
@@ -218,12 +219,12 @@
 { int len = length;\
   if (s->bi_valid > (int)Buf_size - len) {\
     int val = value;\
-    s->bi_buf |= (val << s->bi_valid);\
+    s->bi_buf |= (ush)val << s->bi_valid;\
     put_short(s, s->bi_buf);\
     s->bi_buf = (ush)val >> (Buf_size - s->bi_valid);\
     s->bi_valid += len - Buf_size;\
   } else {\
-    s->bi_buf |= (value) << s->bi_valid;\
+    s->bi_buf |= (ush)(value) << s->bi_valid;\
     s->bi_valid += len;\
   }\
 }
@@ -250,11 +251,13 @@
     if (static_init_done) return;
 
     /* For some embedded targets, global variables are not initialized: */
+#ifdef NO_INIT_GLOBAL_POINTERS
     static_l_desc.static_tree = static_ltree;
     static_l_desc.extra_bits = extra_lbits;
     static_d_desc.static_tree = static_dtree;
     static_d_desc.extra_bits = extra_dbits;
     static_bl_desc.extra_bits = extra_blbits;
+#endif
 
     /* Initialize the mapping length (0..255) -> length code (0..28) */
     length = 0;
@@ -348,13 +351,14 @@
                 static_dtree[i].Len, SEPARATOR(i, D_CODES-1, 5));
     }
 
-    fprintf(header, "const uch _dist_code[DIST_CODE_LEN] = {\n");
+    fprintf(header, "const uch ZLIB_INTERNAL _dist_code[DIST_CODE_LEN] = {\n");
     for (i = 0; i < DIST_CODE_LEN; i++) {
         fprintf(header, "%2u%s", _dist_code[i],
                 SEPARATOR(i, DIST_CODE_LEN-1, 20));
     }
 
-    fprintf(header, "const uch _length_code[MAX_MATCH-MIN_MATCH+1]= {\n");
+    fprintf(header,
+        "const uch ZLIB_INTERNAL _length_code[MAX_MATCH-MIN_MATCH+1]= {\n");
     for (i = 0; i < MAX_MATCH-MIN_MATCH+1; i++) {
         fprintf(header, "%2u%s", _length_code[i],
                 SEPARATOR(i, MAX_MATCH-MIN_MATCH, 20));
@@ -379,7 +383,7 @@
 /* ===========================================================================
  * Initialize the tree data structures for a new zlib stream.
  */
-void _tr_init(s)
+void ZLIB_INTERNAL _tr_init(s)
     deflate_state *s;
 {
     tr_static_init();
@@ -864,13 +868,13 @@
 /* ===========================================================================
  * Send a stored block
  */
-void _tr_stored_block(s, buf, stored_len, eof)
+void ZLIB_INTERNAL _tr_stored_block(s, buf, stored_len, last)
     deflate_state *s;
     charf *buf;       /* input block */
     ulg stored_len;   /* length of input block */
-    int eof;          /* true if this is the last block for a file */
+    int last;         /* one if this is the last block for a file */
 {
-    send_bits(s, (STORED_BLOCK<<1)+eof, 3);  /* send block type */
+    send_bits(s, (STORED_BLOCK<<1)+last, 3);    /* send block type */
 #ifdef DEBUG
     s->compressed_len = (s->compressed_len + 3 + 7) & (ulg)~7L;
     s->compressed_len += (stored_len + 4) << 3;
@@ -889,7 +893,7 @@
  * To simplify the code, we assume the worst case of last real code encoded
  * on one bit only.
  */
-void _tr_align(s)
+void ZLIB_INTERNAL _tr_align(s)
     deflate_state *s;
 {
     send_bits(s, STATIC_TREES<<1, 3);
@@ -918,11 +922,11 @@
  * Determine the best encoding for the current block: dynamic trees, static
  * trees or store, and output the encoded block to the zip file.
  */
-void _tr_flush_block(s, buf, stored_len, eof)
+void ZLIB_INTERNAL _tr_flush_block(s, buf, stored_len, last)
     deflate_state *s;
     charf *buf;       /* input block, or NULL if too old */
     ulg stored_len;   /* length of input block */
-    int eof;          /* true if this is the last block for a file */
+    int last;         /* one if this is the last block for a file */
 {
     ulg opt_lenb, static_lenb; /* opt_len and static_len in bytes */
     int max_blindex = 0;  /* index of last bit length code of non zero freq */
@@ -931,8 +935,8 @@
     if (s->level > 0) {
 
         /* Check if the file is binary or text */
-        if (stored_len > 0 && s->strm->data_type == Z_UNKNOWN)
-            set_data_type(s);
+        if (s->strm->data_type == Z_UNKNOWN)
+            s->strm->data_type = detect_data_type(s);
 
         /* Construct the literal and distance trees */
         build_tree(s, (tree_desc *)(&(s->l_desc)));
@@ -978,20 +982,20 @@
          * successful. If LIT_BUFSIZE <= WSIZE, it is never too late to
          * transform a block into a stored block.
          */
-        _tr_stored_block(s, buf, stored_len, eof);
+        _tr_stored_block(s, buf, stored_len, last);
 
 #ifdef FORCE_STATIC
     } else if (static_lenb >= 0) { /* force static trees */
 #else
     } else if (s->strategy == Z_FIXED || static_lenb == opt_lenb) {
 #endif
-        send_bits(s, (STATIC_TREES<<1)+eof, 3);
+        send_bits(s, (STATIC_TREES<<1)+last, 3);
         compress_block(s, (ct_data *)static_ltree, (ct_data *)static_dtree);
 #ifdef DEBUG
         s->compressed_len += 3 + s->static_len;
 #endif
     } else {
-        send_bits(s, (DYN_TREES<<1)+eof, 3);
+        send_bits(s, (DYN_TREES<<1)+last, 3);
         send_all_trees(s, s->l_desc.max_code+1, s->d_desc.max_code+1,
                        max_blindex+1);
         compress_block(s, (ct_data *)s->dyn_ltree, (ct_data *)s->dyn_dtree);
@@ -1005,21 +1009,21 @@
      */
     init_block(s);
 
-    if (eof) {
+    if (last) {
         bi_windup(s);
 #ifdef DEBUG
         s->compressed_len += 7;  /* align on byte boundary */
 #endif
     }
     Tracev((stderr,"\ncomprlen %lu(%lu) ", s->compressed_len>>3,
-           s->compressed_len-7*eof));
+           s->compressed_len-7*last));
 }
 
 /* ===========================================================================
  * Save the match info and tally the frequency counts. Return true if
  * the current block must be flushed.
  */
-int _tr_tally (s, dist, lc)
+int ZLIB_INTERNAL _tr_tally (s, dist, lc)
     deflate_state *s;
     unsigned dist;  /* distance of matched string */
     unsigned lc;    /* match length-MIN_MATCH or unmatched char (if dist==0) */
@@ -1118,24 +1122,45 @@
 }
 
 /* ===========================================================================
- * Set the data type to BINARY or TEXT, using a crude approximation:
- * set it to Z_TEXT if all symbols are either printable characters (33 to 255)
- * or white spaces (9 to 13, or 32); or set it to Z_BINARY otherwise.
+ * Check if the data type is TEXT or BINARY, using the following algorithm:
+ * - TEXT if the two conditions below are satisfied:
+ *    a) There are no non-portable control characters belonging to the
+ *       "black list" (0..6, 14..25, 28..31).
+ *    b) There is at least one printable character belonging to the
+ *       "white list" (9 {TAB}, 10 {LF}, 13 {CR}, 32..255).
+ * - BINARY otherwise.
+ * - The following partially-portable control characters form a
+ *   "gray list" that is ignored in this detection algorithm:
+ *   (7 {BEL}, 8 {BS}, 11 {VT}, 12 {FF}, 26 {SUB}, 27 {ESC}).
  * IN assertion: the fields Freq of dyn_ltree are set.
  */
-local void set_data_type(s)
+local int detect_data_type(s)
     deflate_state *s;
 {
+    /* black_mask is the bit mask of black-listed bytes
+     * set bits 0..6, 14..25, and 28..31
+     * 0xf3ffc07f = binary 11110011111111111100000001111111
+     */
+    unsigned long black_mask = 0xf3ffc07fUL;
     int n;
 
-    for (n = 0; n < 9; n++)
+    /* Check for non-textual ("black-listed") bytes. */
+    for (n = 0; n <= 31; n++, black_mask >>= 1)
+        if ((black_mask & 1) && (s->dyn_ltree[n].Freq != 0))
+            return Z_BINARY;
+
+    /* Check for textual ("white-listed") bytes. */
+    if (s->dyn_ltree[9].Freq != 0 || s->dyn_ltree[10].Freq != 0
+            || s->dyn_ltree[13].Freq != 0)
+        return Z_TEXT;
+    for (n = 32; n < LITERALS; n++)
         if (s->dyn_ltree[n].Freq != 0)
-            break;
-    if (n == 9)
-        for (n = 14; n < 32; n++)
-            if (s->dyn_ltree[n].Freq != 0)
-                break;
-    s->strm->data_type = (n == 32) ? Z_TEXT : Z_BINARY;
+            return Z_TEXT;
+
+    /* There are no "black-listed" or "white-listed" bytes:
+     * this stream either is empty or has tolerated ("gray-listed") bytes only.
+     */
+    return Z_BINARY;
 }
 
 /* ===========================================================================

Modified: python/branches/py3k/Modules/zlib/trees.h
==============================================================================
--- python/branches/py3k/Modules/zlib/trees.h	(original)
+++ python/branches/py3k/Modules/zlib/trees.h	Fri Jul 30 22:03:17 2010
@@ -70,7 +70,7 @@
 {{19},{ 5}}, {{11},{ 5}}, {{27},{ 5}}, {{ 7},{ 5}}, {{23},{ 5}}
 };
 
-const uch _dist_code[DIST_CODE_LEN] = {
+const uch ZLIB_INTERNAL _dist_code[DIST_CODE_LEN] = {
  0,  1,  2,  3,  4,  4,  5,  5,  6,  6,  6,  6,  7,  7,  7,  7,  8,  8,  8,  8,
  8,  8,  8,  8,  9,  9,  9,  9,  9,  9,  9,  9, 10, 10, 10, 10, 10, 10, 10, 10,
 10, 10, 10, 10, 10, 10, 10, 10, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11,
@@ -99,7 +99,7 @@
 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29
 };
 
-const uch _length_code[MAX_MATCH-MIN_MATCH+1]= {
+const uch ZLIB_INTERNAL _length_code[MAX_MATCH-MIN_MATCH+1]= {
  0,  1,  2,  3,  4,  5,  6,  7,  8,  8,  9,  9, 10, 10, 11, 11, 12, 12, 12, 12,
 13, 13, 13, 13, 14, 14, 14, 14, 15, 15, 15, 15, 16, 16, 16, 16, 16, 16, 16, 16,
 17, 17, 17, 17, 17, 17, 17, 17, 18, 18, 18, 18, 18, 18, 18, 18, 19, 19, 19, 19,

Modified: python/branches/py3k/Modules/zlib/uncompr.c
==============================================================================
--- python/branches/py3k/Modules/zlib/uncompr.c	(original)
+++ python/branches/py3k/Modules/zlib/uncompr.c	Fri Jul 30 22:03:17 2010
@@ -1,5 +1,5 @@
 /* uncompr.c -- decompress a memory buffer
- * Copyright (C) 1995-2003 Jean-loup Gailly.
+ * Copyright (C) 1995-2003, 2010 Jean-loup Gailly.
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -16,8 +16,6 @@
    been saved previously by the compressor and transmitted to the decompressor
    by some mechanism outside the scope of this compression library.)
    Upon exit, destLen is the actual size of the compressed buffer.
-     This function can be used to decompress a whole file at once if the
-   input file is mmap'ed.
 
      uncompress returns Z_OK if success, Z_MEM_ERROR if there was not
    enough memory, Z_BUF_ERROR if there was not enough room in the output

Modified: python/branches/py3k/Modules/zlib/zconf.h
==============================================================================
--- python/branches/py3k/Modules/zlib/zconf.h	(original)
+++ python/branches/py3k/Modules/zlib/zconf.h	Fri Jul 30 22:03:17 2010
@@ -1,5 +1,5 @@
 /* zconf.h -- configuration of the zlib compression library
- * Copyright (C) 1995-2005 Jean-loup Gailly.
+ * Copyright (C) 1995-2010 Jean-loup Gailly.
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -11,52 +11,124 @@
 /*
  * If you *really* need a unique prefix for all types and library functions,
  * compile with -DZ_PREFIX. The "standard" zlib should be compiled without it.
+ * Even better than compiling with -DZ_PREFIX would be to use configure to set
+ * this permanently in zconf.h using "./configure --zprefix".
  */
-#ifdef Z_PREFIX
-#  define deflateInit_          z_deflateInit_
+#ifdef Z_PREFIX     /* may be set to #if 1 by ./configure */
+
+/* all linked symbols */
+#  define _dist_code            z__dist_code
+#  define _length_code          z__length_code
+#  define _tr_align             z__tr_align
+#  define _tr_flush_block       z__tr_flush_block
+#  define _tr_init              z__tr_init
+#  define _tr_stored_block      z__tr_stored_block
+#  define _tr_tally             z__tr_tally
+#  define adler32               z_adler32
+#  define adler32_combine       z_adler32_combine
+#  define adler32_combine64     z_adler32_combine64
+#  define compress              z_compress
+#  define compress2             z_compress2
+#  define compressBound         z_compressBound
+#  define crc32                 z_crc32
+#  define crc32_combine         z_crc32_combine
+#  define crc32_combine64       z_crc32_combine64
 #  define deflate               z_deflate
+#  define deflateBound          z_deflateBound
+#  define deflateCopy           z_deflateCopy
 #  define deflateEnd            z_deflateEnd
-#  define inflateInit_          z_inflateInit_
-#  define inflate               z_inflate
-#  define inflateEnd            z_inflateEnd
 #  define deflateInit2_         z_deflateInit2_
-#  define deflateSetDictionary  z_deflateSetDictionary
-#  define deflateCopy           z_deflateCopy
-#  define deflateReset          z_deflateReset
+#  define deflateInit_          z_deflateInit_
 #  define deflateParams         z_deflateParams
-#  define deflateBound          z_deflateBound
 #  define deflatePrime          z_deflatePrime
+#  define deflateReset          z_deflateReset
+#  define deflateSetDictionary  z_deflateSetDictionary
+#  define deflateSetHeader      z_deflateSetHeader
+#  define deflateTune           z_deflateTune
+#  define deflate_copyright     z_deflate_copyright
+#  define get_crc_table         z_get_crc_table
+#  define gz_error              z_gz_error
+#  define gz_intmax             z_gz_intmax
+#  define gz_strwinerror        z_gz_strwinerror
+#  define gzbuffer              z_gzbuffer
+#  define gzclearerr            z_gzclearerr
+#  define gzclose               z_gzclose
+#  define gzclose_r             z_gzclose_r
+#  define gzclose_w             z_gzclose_w
+#  define gzdirect              z_gzdirect
+#  define gzdopen               z_gzdopen
+#  define gzeof                 z_gzeof
+#  define gzerror               z_gzerror
+#  define gzflush               z_gzflush
+#  define gzgetc                z_gzgetc
+#  define gzgets                z_gzgets
+#  define gzoffset              z_gzoffset
+#  define gzoffset64            z_gzoffset64
+#  define gzopen                z_gzopen
+#  define gzopen64              z_gzopen64
+#  define gzprintf              z_gzprintf
+#  define gzputc                z_gzputc
+#  define gzputs                z_gzputs
+#  define gzread                z_gzread
+#  define gzrewind              z_gzrewind
+#  define gzseek                z_gzseek
+#  define gzseek64              z_gzseek64
+#  define gzsetparams           z_gzsetparams
+#  define gztell                z_gztell
+#  define gztell64              z_gztell64
+#  define gzungetc              z_gzungetc
+#  define gzwrite               z_gzwrite
+#  define inflate               z_inflate
+#  define inflateBack           z_inflateBack
+#  define inflateBackEnd        z_inflateBackEnd
+#  define inflateBackInit_      z_inflateBackInit_
+#  define inflateCopy           z_inflateCopy
+#  define inflateEnd            z_inflateEnd
+#  define inflateGetHeader      z_inflateGetHeader
 #  define inflateInit2_         z_inflateInit2_
+#  define inflateInit_          z_inflateInit_
+#  define inflateMark           z_inflateMark
+#  define inflatePrime          z_inflatePrime
+#  define inflateReset          z_inflateReset
+#  define inflateReset2         z_inflateReset2
 #  define inflateSetDictionary  z_inflateSetDictionary
 #  define inflateSync           z_inflateSync
 #  define inflateSyncPoint      z_inflateSyncPoint
-#  define inflateCopy           z_inflateCopy
-#  define inflateReset          z_inflateReset
-#  define inflateBack           z_inflateBack
-#  define inflateBackEnd        z_inflateBackEnd
-#  define compress              z_compress
-#  define compress2             z_compress2
-#  define compressBound         z_compressBound
+#  define inflateUndermine      z_inflateUndermine
+#  define inflate_copyright     z_inflate_copyright
+#  define inflate_fast          z_inflate_fast
+#  define inflate_table         z_inflate_table
 #  define uncompress            z_uncompress
-#  define adler32               z_adler32
-#  define crc32                 z_crc32
-#  define get_crc_table         z_get_crc_table
 #  define zError                z_zError
+#  define zcalloc               z_zcalloc
+#  define zcfree                z_zcfree
+#  define zlibCompileFlags      z_zlibCompileFlags
+#  define zlibVersion           z_zlibVersion
 
+/* all zlib typedefs in zlib.h and zconf.h */
+#  define Byte                  z_Byte
+#  define Bytef                 z_Bytef
 #  define alloc_func            z_alloc_func
+#  define charf                 z_charf
 #  define free_func             z_free_func
+#  define gzFile                z_gzFile
+#  define gz_header             z_gz_header
+#  define gz_headerp            z_gz_headerp
 #  define in_func               z_in_func
+#  define intf                  z_intf
 #  define out_func              z_out_func
-#  define Byte                  z_Byte
 #  define uInt                  z_uInt
-#  define uLong                 z_uLong
-#  define Bytef                 z_Bytef
-#  define charf                 z_charf
-#  define intf                  z_intf
 #  define uIntf                 z_uIntf
+#  define uLong                 z_uLong
 #  define uLongf                z_uLongf
-#  define voidpf                z_voidpf
 #  define voidp                 z_voidp
+#  define voidpc                z_voidpc
+#  define voidpf                z_voidpf
+
+/* all zlib structs in zlib.h and zconf.h */
+#  define gz_header_s           z_gz_header_s
+#  define internal_state        z_internal_state
+
 #endif
 
 #if defined(__MSDOS__) && !defined(MSDOS)
@@ -284,49 +356,73 @@
    typedef Byte       *voidp;
 #endif
 
-#if 0           /* HAVE_UNISTD_H -- this line is updated by ./configure */
-#  include <sys/types.h> /* for off_t */
-#  include <unistd.h>    /* for SEEK_* and off_t */
+#ifdef HAVE_UNISTD_H    /* may be set to #if 1 by ./configure */
+#  define Z_HAVE_UNISTD_H
+#endif
+
+#ifdef STDC
+#  include <sys/types.h>    /* for off_t */
+#endif
+
+/* a little trick to accommodate both "#define _LARGEFILE64_SOURCE" and
+ * "#define _LARGEFILE64_SOURCE 1" as requesting 64-bit operations, (even
+ * though the former does not conform to the LFS document), but considering
+ * both "#undef _LARGEFILE64_SOURCE" and "#define _LARGEFILE64_SOURCE 0" as
+ * equivalently requesting no 64-bit operations
+ */
+#if -_LARGEFILE64_SOURCE - -1 == 1
+#  undef _LARGEFILE64_SOURCE
+#endif
+
+#if defined(Z_HAVE_UNISTD_H) || defined(_LARGEFILE64_SOURCE)
+#  include <unistd.h>       /* for SEEK_* and off_t */
 #  ifdef VMS
-#    include <unixio.h>   /* for off_t */
+#    include <unixio.h>     /* for off_t */
+#  endif
+#  ifndef z_off_t
+#    define z_off_t off_t
 #  endif
-#  define z_off_t off_t
 #endif
+
 #ifndef SEEK_SET
 #  define SEEK_SET        0       /* Seek from beginning of file.  */
 #  define SEEK_CUR        1       /* Seek from current position.  */
 #  define SEEK_END        2       /* Set file pointer to EOF plus "offset" */
 #endif
+
 #ifndef z_off_t
 #  define z_off_t long
 #endif
 
+#if defined(_LARGEFILE64_SOURCE) && _LFS64_LARGEFILE-0
+#  define z_off64_t off64_t
+#else
+#  define z_off64_t z_off_t
+#endif
+
 #if defined(__OS400__)
 #  define NO_vsnprintf
 #endif
 
 #if defined(__MVS__)
 #  define NO_vsnprintf
-#  ifdef FAR
-#    undef FAR
-#  endif
 #endif
 
 /* MVS linker does not support external names larger than 8 bytes */
 #if defined(__MVS__)
-#   pragma map(deflateInit_,"DEIN")
-#   pragma map(deflateInit2_,"DEIN2")
-#   pragma map(deflateEnd,"DEEND")
-#   pragma map(deflateBound,"DEBND")
-#   pragma map(inflateInit_,"ININ")
-#   pragma map(inflateInit2_,"ININ2")
-#   pragma map(inflateEnd,"INEND")
-#   pragma map(inflateSync,"INSY")
-#   pragma map(inflateSetDictionary,"INSEDI")
-#   pragma map(compressBound,"CMBND")
-#   pragma map(inflate_table,"INTABL")
-#   pragma map(inflate_fast,"INFA")
-#   pragma map(inflate_copyright,"INCOPY")
+  #pragma map(deflateInit_,"DEIN")
+  #pragma map(deflateInit2_,"DEIN2")
+  #pragma map(deflateEnd,"DEEND")
+  #pragma map(deflateBound,"DEBND")
+  #pragma map(inflateInit_,"ININ")
+  #pragma map(inflateInit2_,"ININ2")
+  #pragma map(inflateEnd,"INEND")
+  #pragma map(inflateSync,"INSY")
+  #pragma map(inflateSetDictionary,"INSEDI")
+  #pragma map(compressBound,"CMBND")
+  #pragma map(inflate_table,"INTABL")
+  #pragma map(inflate_fast,"INFA")
+  #pragma map(inflate_copyright,"INCOPY")
 #endif
 
 #endif /* ZCONF_H */

Modified: python/branches/py3k/Modules/zlib/zlib.3
==============================================================================
--- python/branches/py3k/Modules/zlib/zlib.3	(original)
+++ python/branches/py3k/Modules/zlib/zlib.3	Fri Jul 30 22:03:17 2010
@@ -1,4 +1,4 @@
-.TH ZLIB 3 "18 July 2005"
+.TH ZLIB 3 "19 Apr 2010"
 .SH NAME
 zlib \- compression/decompression library
 .SH SYNOPSIS
@@ -9,15 +9,15 @@
 The
 .I zlib
 library is a general purpose data compression library.
-The code is thread safe.
+The code is thread safe, assuming that the standard library functions
+used are thread safe, such as memory allocation routines.
 It provides in-memory compression and decompression functions,
 including integrity checks of the uncompressed data.
 This version of the library supports only one compression method (deflation)
-but other algorithms will be added later
-and will have the same stream interface.
+but other algorithms may be added later
+with the same stream interface.
 .LP
 Compression can be done in a single step if the buffers are large enough
-(for example if an input file is mmap'ed),
 or can be done by repeated calls of the compression function.
 In the latter case,
 the application must provide more input and/or consume the output
@@ -30,7 +30,7 @@
 .LP
 The library does not install any signal handler.
 The decoder checks the consistency of the compressed data,
-so the library should never crash even in case of corrupted input.
+so the library should never crash even in the case of corrupted input.
 .LP
 All functions of the compression library are documented in the file
 .IR zlib.h .
@@ -38,18 +38,19 @@
 in the files
 .I example.c
 and
-.IR minigzip.c .
+.IR minigzip.c,
+as well as other examples in the
+.IR examples/
+directory.
 .LP
 Changes to this version are documented in the file
 .I ChangeLog
-that accompanies the source,
-and are concerned primarily with bug fixes and portability enhancements.
+that accompanies the source.
 .LP
-A Java implementation of
 .I zlib
-is available in the Java Development Kit 1.1:
+is available in Java using the java.util.zip package:
 .IP
-http://www.javasoft.com/products/JDK/1.1/docs/api/Package-java.util.zip.html
+http://java.sun.com/developer/technicalArticles/Programming/compression/
 .LP
 A Perl interface to
 .IR zlib ,
@@ -57,7 +58,7 @@
 is available at CPAN (Comprehensive Perl Archive Network) sites,
 including:
 .IP
-http://www.cpan.org/modules/by-module/Compress/
+http://search.cpan.org/~pmqs/IO-Compress-Zlib/
 .LP
 A Python interface to
 .IR zlib ,
@@ -66,14 +67,11 @@
 .IP
 http://www.python.org/doc/lib/module-zlib.html
 .LP
-A
 .I zlib
-binding for
-.IR tcl (1),
-written by Andreas Kupries (a.kupries at westend.com),
-is availlable at:
+is built into
+.IR tcl:
 .IP
-http://www.westend.com/~kupries/doc/trf/man/man.html
+http://wiki.tcl.tk/4610
 .LP
 An experimental package to read and write files in .zip format,
 written on top of
@@ -81,40 +79,34 @@
 by Gilles Vollant (info at winimage.com),
 is available at:
 .IP
-http://www.winimage.com/zLibDll/unzip.html
+http://www.winimage.com/zLibDll/minizip.html
 and also in the
 .I contrib/minizip
 directory of the main
 .I zlib
-web site.
+source distribution.
 .SH "SEE ALSO"
 The
 .I zlib
-web site can be found at either of these locations:
+web site can be found at:
 .IP
-http://www.zlib.org
-.br
-http://www.gzip.org/zlib/
+http://zlib.net/
 .LP
 The data format used by the zlib library is described by RFC
 (Request for Comments) 1950 to 1952 in the files:
 .IP
-http://www.ietf.org/rfc/rfc1950.txt (concerning zlib format)
+http://www.ietf.org/rfc/rfc1950.txt (for the zlib header and trailer format)
 .br
-http://www.ietf.org/rfc/rfc1951.txt (concerning deflate format)
+http://www.ietf.org/rfc/rfc1951.txt (for the deflate compressed data format)
 .br
-http://www.ietf.org/rfc/rfc1952.txt (concerning gzip format)
-.LP
-These documents are also available in other formats from:
-.IP
-ftp://ftp.uu.net/graphics/png/documents/zlib/zdoc-index.html
+http://www.ietf.org/rfc/rfc1952.txt (for the gzip header and trailer format)
 .LP
-Mark Nelson (markn at ieee.org) wrote an article about
+Mark Nelson wrote an article about
 .I zlib
 for the Jan. 1997 issue of  Dr. Dobb's Journal;
 a copy of the article is available at:
 .IP
-http://dogma.net/markn/articles/zlibtool/zlibtool.htm
+http://marknelson.us/1997/01/01/zlib-engine/
 .SH "REPORTING PROBLEMS"
 Before reporting a problem,
 please check the
@@ -127,14 +119,14 @@
 .I zlib
 FAQ at:
 .IP
-http://www.gzip.org/zlib/zlib_faq.html
+http://zlib.net/zlib_faq.html
 .LP
 before asking for help.
 Send questions and/or comments to zlib at gzip.org,
 or (for the Windows DLL version) to Gilles Vollant (info at winimage.com).
 .SH AUTHORS
-Version 1.2.3
-Copyright (C) 1995-2005 Jean-loup Gailly (jloup at gzip.org)
+Version 1.2.5
+Copyright (C) 1995-2010 Jean-loup Gailly (jloup at gzip.org)
 and Mark Adler (madler at alumni.caltech.edu).
 .LP
 This software is provided "as-is,"

Modified: python/branches/py3k/Modules/zlib/zlib.h
==============================================================================
--- python/branches/py3k/Modules/zlib/zlib.h	(original)
+++ python/branches/py3k/Modules/zlib/zlib.h	Fri Jul 30 22:03:17 2010
@@ -1,7 +1,7 @@
 /* zlib.h -- interface of the 'zlib' general purpose compression library
-  version 1.2.3, July 18th, 2005
+  version 1.2.5, April 19th, 2010
 
-  Copyright (C) 1995-2005 Jean-loup Gailly and Mark Adler
+  Copyright (C) 1995-2010 Jean-loup Gailly and Mark Adler
 
   This software is provided 'as-is', without any express or implied
   warranty.  In no event will the authors be held liable for any damages
@@ -37,41 +37,44 @@
 extern "C" {
 #endif
 
-#define ZLIB_VERSION "1.2.3"
-#define ZLIB_VERNUM 0x1230
-
-/*
-     The 'zlib' compression library provides in-memory compression and
-  decompression functions, including integrity checks of the uncompressed
-  data.  This version of the library supports only one compression method
-  (deflation) but other algorithms will be added later and will have the same
-  stream interface.
-
-     Compression can be done in a single step if the buffers are large
-  enough (for example if an input file is mmap'ed), or can be done by
-  repeated calls of the compression function.  In the latter case, the
-  application must provide more input and/or consume the output
+#define ZLIB_VERSION "1.2.5"
+#define ZLIB_VERNUM 0x1250
+#define ZLIB_VER_MAJOR 1
+#define ZLIB_VER_MINOR 2
+#define ZLIB_VER_REVISION 5
+#define ZLIB_VER_SUBREVISION 0
+
+/*
+    The 'zlib' compression library provides in-memory compression and
+  decompression functions, including integrity checks of the uncompressed data.
+  This version of the library supports only one compression method (deflation)
+  but other algorithms will be added later and will have the same stream
+  interface.
+
+    Compression can be done in a single step if the buffers are large enough,
+  or can be done by repeated calls of the compression function.  In the latter
+  case, the application must provide more input and/or consume the output
   (providing more output space) before each call.
 
-     The compressed data format used by default by the in-memory functions is
+    The compressed data format used by default by the in-memory functions is
   the zlib format, which is a zlib wrapper documented in RFC 1950, wrapped
   around a deflate stream, which is itself documented in RFC 1951.
 
-     The library also supports reading and writing files in gzip (.gz) format
+    The library also supports reading and writing files in gzip (.gz) format
   with an interface similar to that of stdio using the functions that start
   with "gz".  The gzip format is different from the zlib format.  gzip is a
   gzip wrapper, documented in RFC 1952, wrapped around a deflate stream.
 
-     This library can optionally read and write gzip streams in memory as well.
+    This library can optionally read and write gzip streams in memory as well.
 
-     The zlib format was designed to be compact and fast for use in memory
+    The zlib format was designed to be compact and fast for use in memory
   and on communications channels.  The gzip format was designed for single-
   file compression on file systems, has a larger header than zlib to maintain
   directory information, and uses a different, slower check method than zlib.
 
-     The library does not install any signal handler. The decoder checks
-  the consistency of the compressed data, so the library should never
-  crash even in case of corrupted input.
+    The library does not install any signal handler.  The decoder checks
+  the consistency of the compressed data, so the library should never crash
+  even in case of corrupted input.
 */
 
 typedef voidpf (*alloc_func) OF((voidpf opaque, uInt items, uInt size));
@@ -126,45 +129,45 @@
 typedef gz_header FAR *gz_headerp;
 
 /*
-   The application must update next_in and avail_in when avail_in has
-   dropped to zero. It must update next_out and avail_out when avail_out
-   has dropped to zero. The application must initialize zalloc, zfree and
-   opaque before calling the init function. All other fields are set by the
-   compression library and must not be updated by the application.
-
-   The opaque value provided by the application will be passed as the first
-   parameter for calls of zalloc and zfree. This can be useful for custom
-   memory management. The compression library attaches no meaning to the
+     The application must update next_in and avail_in when avail_in has dropped
+   to zero.  It must update next_out and avail_out when avail_out has dropped
+   to zero.  The application must initialize zalloc, zfree and opaque before
+   calling the init function.  All other fields are set by the compression
+   library and must not be updated by the application.
+
+     The opaque value provided by the application will be passed as the first
+   parameter for calls of zalloc and zfree.  This can be useful for custom
+   memory management.  The compression library attaches no meaning to the
    opaque value.
 
-   zalloc must return Z_NULL if there is not enough memory for the object.
+     zalloc must return Z_NULL if there is not enough memory for the object.
    If zlib is used in a multi-threaded application, zalloc and zfree must be
    thread safe.
 
-   On 16-bit systems, the functions zalloc and zfree must be able to allocate
-   exactly 65536 bytes, but will not be required to allocate more than this
-   if the symbol MAXSEG_64K is defined (see zconf.h). WARNING: On MSDOS,
-   pointers returned by zalloc for objects of exactly 65536 bytes *must*
-   have their offset normalized to zero. The default allocation function
-   provided by this library ensures this (see zutil.c). To reduce memory
-   requirements and avoid any allocation of 64K objects, at the expense of
-   compression ratio, compile the library with -DMAX_WBITS=14 (see zconf.h).
-
-   The fields total_in and total_out can be used for statistics or
-   progress reports. After compression, total_in holds the total size of
-   the uncompressed data and may be saved for use in the decompressor
-   (particularly if the decompressor wants to decompress everything in
-   a single step).
+     On 16-bit systems, the functions zalloc and zfree must be able to allocate
+   exactly 65536 bytes, but will not be required to allocate more than this if
+   the symbol MAXSEG_64K is defined (see zconf.h).  WARNING: On MSDOS, pointers
+   returned by zalloc for objects of exactly 65536 bytes *must* have their
+   offset normalized to zero.  The default allocation function provided by this
+   library ensures this (see zutil.c).  To reduce memory requirements and avoid
+   any allocation of 64K objects, at the expense of compression ratio, compile
+   the library with -DMAX_WBITS=14 (see zconf.h).
+
+     The fields total_in and total_out can be used for statistics or progress
+   reports.  After compression, total_in holds the total size of the
+   uncompressed data and may be saved for use in the decompressor (particularly
+   if the decompressor wants to decompress everything in a single step).
 */
 
                         /* constants */
 
 #define Z_NO_FLUSH      0
-#define Z_PARTIAL_FLUSH 1 /* will be removed, use Z_SYNC_FLUSH instead */
+#define Z_PARTIAL_FLUSH 1
 #define Z_SYNC_FLUSH    2
 #define Z_FULL_FLUSH    3
 #define Z_FINISH        4
 #define Z_BLOCK         5
+#define Z_TREES         6
 /* Allowed flush values; see deflate() and inflate() below for details */
 
 #define Z_OK            0
@@ -176,8 +179,8 @@
 #define Z_MEM_ERROR    (-4)
 #define Z_BUF_ERROR    (-5)
 #define Z_VERSION_ERROR (-6)
-/* Return codes for the compression/decompression functions. Negative
- * values are errors, positive values are used for special but normal events.
+/* Return codes for the compression/decompression functions. Negative values
+ * are errors, positive values are used for special but normal events.
  */
 
 #define Z_NO_COMPRESSION         0
@@ -207,119 +210,140 @@
 #define zlib_version zlibVersion()
 /* for compatibility with versions < 1.0.2 */
 
+
                         /* basic functions */
 
 ZEXTERN const char * ZEXPORT zlibVersion OF((void));
 /* The application can compare zlibVersion and ZLIB_VERSION for consistency.
-   If the first character differs, the library code actually used is
-   not compatible with the zlib.h header file used by the application.
-   This check is automatically made by deflateInit and inflateInit.
+   If the first character differs, the library code actually used is not
+   compatible with the zlib.h header file used by the application.  This check
+   is automatically made by deflateInit and inflateInit.
  */
 
 /*
 ZEXTERN int ZEXPORT deflateInit OF((z_streamp strm, int level));
 
-     Initializes the internal stream state for compression. The fields
-   zalloc, zfree and opaque must be initialized before by the caller.
-   If zalloc and zfree are set to Z_NULL, deflateInit updates them to
-   use default allocation functions.
+     Initializes the internal stream state for compression.  The fields
+   zalloc, zfree and opaque must be initialized before by the caller.  If
+   zalloc and zfree are set to Z_NULL, deflateInit updates them to use default
+   allocation functions.
 
      The compression level must be Z_DEFAULT_COMPRESSION, or between 0 and 9:
-   1 gives best speed, 9 gives best compression, 0 gives no compression at
-   all (the input data is simply copied a block at a time).
-   Z_DEFAULT_COMPRESSION requests a default compromise between speed and
-   compression (currently equivalent to level 6).
+   1 gives best speed, 9 gives best compression, 0 gives no compression at all
+   (the input data is simply copied a block at a time).  Z_DEFAULT_COMPRESSION
+   requests a default compromise between speed and compression (currently
+   equivalent to level 6).
 
-     deflateInit returns Z_OK if success, Z_MEM_ERROR if there was not
-   enough memory, Z_STREAM_ERROR if level is not a valid compression level,
+     deflateInit returns Z_OK if success, Z_MEM_ERROR if there was not enough
+   memory, Z_STREAM_ERROR if level is not a valid compression level, or
    Z_VERSION_ERROR if the zlib library version (zlib_version) is incompatible
-   with the version assumed by the caller (ZLIB_VERSION).
-   msg is set to null if there is no error message.  deflateInit does not
-   perform any compression: this will be done by deflate().
+   with the version assumed by the caller (ZLIB_VERSION).  msg is set to null
+   if there is no error message.  deflateInit does not perform any compression:
+   this will be done by deflate().
 */
 
 
 ZEXTERN int ZEXPORT deflate OF((z_streamp strm, int flush));
 /*
     deflate compresses as much data as possible, and stops when the input
-  buffer becomes empty or the output buffer becomes full. It may introduce some
-  output latency (reading input without producing any output) except when
+  buffer becomes empty or the output buffer becomes full.  It may introduce
+  some output latency (reading input without producing any output) except when
   forced to flush.
 
-    The detailed semantics are as follows. deflate performs one or both of the
+    The detailed semantics are as follows.  deflate performs one or both of the
   following actions:
 
   - Compress more input starting at next_in and update next_in and avail_in
-    accordingly. If not all input can be processed (because there is not
+    accordingly.  If not all input can be processed (because there is not
     enough room in the output buffer), next_in and avail_in are updated and
     processing will resume at this point for the next call of deflate().
 
   - Provide more output starting at next_out and update next_out and avail_out
-    accordingly. This action is forced if the parameter flush is non zero.
+    accordingly.  This action is forced if the parameter flush is non zero.
     Forcing flush frequently degrades the compression ratio, so this parameter
-    should be set only when necessary (in interactive applications).
-    Some output may be provided even if flush is not set.
+    should be set only when necessary (in interactive applications).  Some
+    output may be provided even if flush is not set.
 
-  Before the call of deflate(), the application should ensure that at least
-  one of the actions is possible, by providing more input and/or consuming
-  more output, and updating avail_in or avail_out accordingly; avail_out
-  should never be zero before the call. The application can consume the
-  compressed output when it wants, for example when the output buffer is full
-  (avail_out == 0), or after each call of deflate(). If deflate returns Z_OK
-  and with zero avail_out, it must be called again after making room in the
-  output buffer because there might be more output pending.
+    Before the call of deflate(), the application should ensure that at least
+  one of the actions is possible, by providing more input and/or consuming more
+  output, and updating avail_in or avail_out accordingly; avail_out should
+  never be zero before the call.  The application can consume the compressed
+  output when it wants, for example when the output buffer is full (avail_out
+  == 0), or after each call of deflate().  If deflate returns Z_OK and with
+  zero avail_out, it must be called again after making room in the output
+  buffer because there might be more output pending.
 
     Normally the parameter flush is set to Z_NO_FLUSH, which allows deflate to
-  decide how much data to accumualte before producing output, in order to
+  decide how much data to accumulate before producing output, in order to
   maximize compression.
 
     If the parameter flush is set to Z_SYNC_FLUSH, all pending output is
   flushed to the output buffer and the output is aligned on a byte boundary, so
-  that the decompressor can get all input data available so far. (In particular
-  avail_in is zero after the call if enough output space has been provided
-  before the call.)  Flushing may degrade compression for some compression
-  algorithms and so it should be used only when necessary.
+  that the decompressor can get all input data available so far.  (In
+  particular avail_in is zero after the call if enough output space has been
+  provided before the call.) Flushing may degrade compression for some
+  compression algorithms and so it should be used only when necessary.  This
+  completes the current deflate block and follows it with an empty stored block
+  that is three bits plus filler bits to the next byte, followed by four bytes
+  (00 00 ff ff).
+
+    If flush is set to Z_PARTIAL_FLUSH, all pending output is flushed to the
+  output buffer, but the output is not aligned to a byte boundary.  All of the
+  input data so far will be available to the decompressor, as for Z_SYNC_FLUSH.
+  This completes the current deflate block and follows it with an empty fixed
+  codes block that is 10 bits long.  This assures that enough bytes are output
+  in order for the decompressor to finish the block before the empty fixed code
+  block.
+
+    If flush is set to Z_BLOCK, a deflate block is completed and emitted, as
+  for Z_SYNC_FLUSH, but the output is not aligned on a byte boundary, and up to
+  seven bits of the current block are held to be written as the next byte after
+  the next deflate block is completed.  In this case, the decompressor may not
+  be provided enough bits at this point in order to complete decompression of
+  the data provided so far to the compressor.  It may need to wait for the next
+  block to be emitted.  This is for advanced applications that need to control
+  the emission of deflate blocks.
 
     If flush is set to Z_FULL_FLUSH, all output is flushed as with
   Z_SYNC_FLUSH, and the compression state is reset so that decompression can
   restart from this point if previous compressed data has been damaged or if
-  random access is desired. Using Z_FULL_FLUSH too often can seriously degrade
+  random access is desired.  Using Z_FULL_FLUSH too often can seriously degrade
   compression.
 
     If deflate returns with avail_out == 0, this function must be called again
   with the same value of the flush parameter and more output space (updated
   avail_out), until the flush is complete (deflate returns with non-zero
-  avail_out). In the case of a Z_FULL_FLUSH or Z_SYNC_FLUSH, make sure that
+  avail_out).  In the case of a Z_FULL_FLUSH or Z_SYNC_FLUSH, make sure that
   avail_out is greater than six to avoid repeated flush markers due to
   avail_out == 0 on return.
 
     If the parameter flush is set to Z_FINISH, pending input is processed,
-  pending output is flushed and deflate returns with Z_STREAM_END if there
-  was enough output space; if deflate returns with Z_OK, this function must be
+  pending output is flushed and deflate returns with Z_STREAM_END if there was
+  enough output space; if deflate returns with Z_OK, this function must be
   called again with Z_FINISH and more output space (updated avail_out) but no
-  more input data, until it returns with Z_STREAM_END or an error. After
-  deflate has returned Z_STREAM_END, the only possible operations on the
-  stream are deflateReset or deflateEnd.
+  more input data, until it returns with Z_STREAM_END or an error.  After
+  deflate has returned Z_STREAM_END, the only possible operations on the stream
+  are deflateReset or deflateEnd.
 
     Z_FINISH can be used immediately after deflateInit if all the compression
-  is to be done in a single step. In this case, avail_out must be at least
-  the value returned by deflateBound (see below). If deflate does not return
+  is to be done in a single step.  In this case, avail_out must be at least the
+  value returned by deflateBound (see below).  If deflate does not return
   Z_STREAM_END, then it must be called again as described above.
 
     deflate() sets strm->adler to the adler32 checksum of all input read
   so far (that is, total_in bytes).
 
     deflate() may update strm->data_type if it can make a good guess about
-  the input data type (Z_BINARY or Z_TEXT). In doubt, the data is considered
-  binary. This field is only for information purposes and does not affect
-  the compression algorithm in any manner.
+  the input data type (Z_BINARY or Z_TEXT).  In doubt, the data is considered
+  binary.  This field is only for information purposes and does not affect the
+  compression algorithm in any manner.
 
     deflate() returns Z_OK if some progress has been made (more input
   processed or more output produced), Z_STREAM_END if all input has been
   consumed and all output has been produced (only when flush is set to
   Z_FINISH), Z_STREAM_ERROR if the stream state was inconsistent (for example
-  if next_in or next_out was NULL), Z_BUF_ERROR if no progress is possible
-  (for example avail_in or avail_out was zero). Note that Z_BUF_ERROR is not
+  if next_in or next_out was Z_NULL), Z_BUF_ERROR if no progress is possible
+  (for example avail_in or avail_out was zero).  Note that Z_BUF_ERROR is not
   fatal, and deflate() can be called again with more input and more output
   space to continue compressing.
 */
@@ -328,13 +352,13 @@
 ZEXTERN int ZEXPORT deflateEnd OF((z_streamp strm));
 /*
      All dynamically allocated data structures for this stream are freed.
-   This function discards any unprocessed input and does not flush any
-   pending output.
+   This function discards any unprocessed input and does not flush any pending
+   output.
 
      deflateEnd returns Z_OK if success, Z_STREAM_ERROR if the
    stream state was inconsistent, Z_DATA_ERROR if the stream was freed
-   prematurely (some input or output was discarded). In the error case,
-   msg may be set but then points to a static string (which must not be
+   prematurely (some input or output was discarded).  In the error case, msg
+   may be set but then points to a static string (which must not be
    deallocated).
 */
 
@@ -342,10 +366,10 @@
 /*
 ZEXTERN int ZEXPORT inflateInit OF((z_streamp strm));
 
-     Initializes the internal stream state for decompression. The fields
+     Initializes the internal stream state for decompression.  The fields
    next_in, avail_in, zalloc, zfree and opaque must be initialized before by
-   the caller. If next_in is not Z_NULL and avail_in is large enough (the exact
-   value depends on the compression method), inflateInit determines the
+   the caller.  If next_in is not Z_NULL and avail_in is large enough (the
+   exact value depends on the compression method), inflateInit determines the
    compression method from the zlib header and allocates all data structures
    accordingly; otherwise the allocation will be deferred to the first call of
    inflate.  If zalloc and zfree are set to Z_NULL, inflateInit updates them to
@@ -353,95 +377,108 @@
 
      inflateInit returns Z_OK if success, Z_MEM_ERROR if there was not enough
    memory, Z_VERSION_ERROR if the zlib library version is incompatible with the
-   version assumed by the caller.  msg is set to null if there is no error
-   message. inflateInit does not perform any decompression apart from reading
-   the zlib header if present: this will be done by inflate().  (So next_in and
-   avail_in may be modified, but next_out and avail_out are unchanged.)
+   version assumed by the caller, or Z_STREAM_ERROR if the parameters are
+   invalid, such as a null pointer to the structure.  msg is set to null if
+   there is no error message.  inflateInit does not perform any decompression
+   apart from possibly reading the zlib header if present: actual decompression
+   will be done by inflate().  (So next_in and avail_in may be modified, but
+   next_out and avail_out are unused and unchanged.) The current implementation
+   of inflateInit() does not process any header information -- that is deferred
+   until inflate() is called.
 */
 
 
 ZEXTERN int ZEXPORT inflate OF((z_streamp strm, int flush));
 /*
     inflate decompresses as much data as possible, and stops when the input
-  buffer becomes empty or the output buffer becomes full. It may introduce
+  buffer becomes empty or the output buffer becomes full.  It may introduce
   some output latency (reading input without producing any output) except when
   forced to flush.
 
-  The detailed semantics are as follows. inflate performs one or both of the
+  The detailed semantics are as follows.  inflate performs one or both of the
   following actions:
 
   - Decompress more input starting at next_in and update next_in and avail_in
-    accordingly. If not all input can be processed (because there is not
-    enough room in the output buffer), next_in is updated and processing
-    will resume at this point for the next call of inflate().
+    accordingly.  If not all input can be processed (because there is not
+    enough room in the output buffer), next_in is updated and processing will
+    resume at this point for the next call of inflate().
 
   - Provide more output starting at next_out and update next_out and avail_out
-    accordingly.  inflate() provides as much output as possible, until there
-    is no more input data or no more space in the output buffer (see below
-    about the flush parameter).
-
-  Before the call of inflate(), the application should ensure that at least
-  one of the actions is possible, by providing more input and/or consuming
-  more output, and updating the next_* and avail_* values accordingly.
-  The application can consume the uncompressed output when it wants, for
-  example when the output buffer is full (avail_out == 0), or after each
-  call of inflate(). If inflate returns Z_OK and with zero avail_out, it
-  must be called again after making room in the output buffer because there
-  might be more output pending.
-
-    The flush parameter of inflate() can be Z_NO_FLUSH, Z_SYNC_FLUSH,
-  Z_FINISH, or Z_BLOCK. Z_SYNC_FLUSH requests that inflate() flush as much
-  output as possible to the output buffer. Z_BLOCK requests that inflate() stop
-  if and when it gets to the next deflate block boundary. When decoding the
-  zlib or gzip format, this will cause inflate() to return immediately after
-  the header and before the first block. When doing a raw inflate, inflate()
-  will go ahead and process the first block, and will return when it gets to
-  the end of that block, or when it runs out of data.
+    accordingly.  inflate() provides as much output as possible, until there is
+    no more input data or no more space in the output buffer (see below about
+    the flush parameter).
+
+    Before the call of inflate(), the application should ensure that at least
+  one of the actions is possible, by providing more input and/or consuming more
+  output, and updating the next_* and avail_* values accordingly.  The
+  application can consume the uncompressed output when it wants, for example
+  when the output buffer is full (avail_out == 0), or after each call of
+  inflate().  If inflate returns Z_OK and with zero avail_out, it must be
+  called again after making room in the output buffer because there might be
+  more output pending.
+
+    The flush parameter of inflate() can be Z_NO_FLUSH, Z_SYNC_FLUSH, Z_FINISH,
+  Z_BLOCK, or Z_TREES.  Z_SYNC_FLUSH requests that inflate() flush as much
+  output as possible to the output buffer.  Z_BLOCK requests that inflate()
+  stop if and when it gets to the next deflate block boundary.  When decoding
+  the zlib or gzip format, this will cause inflate() to return immediately
+  after the header and before the first block.  When doing a raw inflate,
+  inflate() will go ahead and process the first block, and will return when it
+  gets to the end of that block, or when it runs out of data.
 
     The Z_BLOCK option assists in appending to or combining deflate streams.
   Also to assist in this, on return inflate() will set strm->data_type to the
-  number of unused bits in the last byte taken from strm->next_in, plus 64
-  if inflate() is currently decoding the last block in the deflate stream,
-  plus 128 if inflate() returned immediately after decoding an end-of-block
-  code or decoding the complete header up to just before the first byte of the
-  deflate stream. The end-of-block will not be indicated until all of the
-  uncompressed data from that block has been written to strm->next_out.  The
-  number of unused bits may in general be greater than seven, except when
-  bit 7 of data_type is set, in which case the number of unused bits will be
-  less than eight.
+  number of unused bits in the last byte taken from strm->next_in, plus 64 if
+  inflate() is currently decoding the last block in the deflate stream, plus
+  128 if inflate() returned immediately after decoding an end-of-block code or
+  decoding the complete header up to just before the first byte of the deflate
+  stream.  The end-of-block will not be indicated until all of the uncompressed
+  data from that block has been written to strm->next_out.  The number of
+  unused bits may in general be greater than seven, except when bit 7 of
+  data_type is set, in which case the number of unused bits will be less than
+  eight.  data_type is set as noted here every time inflate() returns for all
+  flush options, and so can be used to determine the amount of currently
+  consumed input in bits.
+
+    The Z_TREES option behaves as Z_BLOCK does, but it also returns when the
+  end of each deflate block header is reached, before any actual data in that
+  block is decoded.  This allows the caller to determine the length of the
+  deflate block header for later use in random access within a deflate block.
+  256 is added to the value of strm->data_type when inflate() returns
+  immediately after reaching the end of the deflate block header.
 
     inflate() should normally be called until it returns Z_STREAM_END or an
-  error. However if all decompression is to be performed in a single step
-  (a single call of inflate), the parameter flush should be set to
-  Z_FINISH. In this case all pending input is processed and all pending
-  output is flushed; avail_out must be large enough to hold all the
-  uncompressed data. (The size of the uncompressed data may have been saved
-  by the compressor for this purpose.) The next operation on this stream must
-  be inflateEnd to deallocate the decompression state. The use of Z_FINISH
-  is never required, but can be used to inform inflate that a faster approach
-  may be used for the single inflate() call.
+  error.  However if all decompression is to be performed in a single step (a
+  single call of inflate), the parameter flush should be set to Z_FINISH.  In
+  this case all pending input is processed and all pending output is flushed;
+  avail_out must be large enough to hold all the uncompressed data.  (The size
+  of the uncompressed data may have been saved by the compressor for this
+  purpose.) The next operation on this stream must be inflateEnd to deallocate
+  the decompression state.  The use of Z_FINISH is never required, but can be
+  used to inform inflate that a faster approach may be used for the single
+  inflate() call.
 
      In this implementation, inflate() always flushes as much output as
   possible to the output buffer, and always uses the faster approach on the
-  first call. So the only effect of the flush parameter in this implementation
+  first call.  So the only effect of the flush parameter in this implementation
   is on the return value of inflate(), as noted below, or when it returns early
-  because Z_BLOCK is used.
+  because Z_BLOCK or Z_TREES is used.
 
      If a preset dictionary is needed after this call (see inflateSetDictionary
   below), inflate sets strm->adler to the adler32 checksum of the dictionary
   chosen by the compressor and returns Z_NEED_DICT; otherwise it sets
   strm->adler to the adler32 checksum of all output produced so far (that is,
   total_out bytes) and returns Z_OK, Z_STREAM_END or an error code as described
-  below. At the end of the stream, inflate() checks that its computed adler32
+  below.  At the end of the stream, inflate() checks that its computed adler32
   checksum is equal to that saved by the compressor and returns Z_STREAM_END
   only if the checksum is correct.
 
-    inflate() will decompress and check either zlib-wrapped or gzip-wrapped
-  deflate data.  The header type is detected automatically.  Any information
-  contained in the gzip header is not retained, so applications that need that
-  information should instead use raw inflate, see inflateInit2() below, or
-  inflateBack() and perform their own processing of the gzip header and
-  trailer.
+    inflate() can decompress and check either zlib-wrapped or gzip-wrapped
+  deflate data.  The header type is detected automatically, if requested when
+  initializing with inflateInit2().  Any information contained in the gzip
+  header is not retained, so applications that need that information should
+  instead use raw inflate, see inflateInit2() below, or inflateBack() and
+  perform their own processing of the gzip header and trailer.
 
     inflate() returns Z_OK if some progress has been made (more input processed
   or more output produced), Z_STREAM_END if the end of the compressed data has
@@ -449,27 +486,28 @@
   preset dictionary is needed at this point, Z_DATA_ERROR if the input data was
   corrupted (input stream not conforming to the zlib format or incorrect check
   value), Z_STREAM_ERROR if the stream structure was inconsistent (for example
-  if next_in or next_out was NULL), Z_MEM_ERROR if there was not enough memory,
+  next_in or next_out was Z_NULL), Z_MEM_ERROR if there was not enough memory,
   Z_BUF_ERROR if no progress is possible or if there was not enough room in the
-  output buffer when Z_FINISH is used. Note that Z_BUF_ERROR is not fatal, and
+  output buffer when Z_FINISH is used.  Note that Z_BUF_ERROR is not fatal, and
   inflate() can be called again with more input and more output space to
-  continue decompressing. If Z_DATA_ERROR is returned, the application may then
-  call inflateSync() to look for a good compression block if a partial recovery
-  of the data is desired.
+  continue decompressing.  If Z_DATA_ERROR is returned, the application may
+  then call inflateSync() to look for a good compression block if a partial
+  recovery of the data is desired.
 */
 
 
 ZEXTERN int ZEXPORT inflateEnd OF((z_streamp strm));
 /*
      All dynamically allocated data structures for this stream are freed.
-   This function discards any unprocessed input and does not flush any
-   pending output.
+   This function discards any unprocessed input and does not flush any pending
+   output.
 
      inflateEnd returns Z_OK if success, Z_STREAM_ERROR if the stream state
-   was inconsistent. In the error case, msg may be set but then points to a
+   was inconsistent.  In the error case, msg may be set but then points to a
    static string (which must not be deallocated).
 */
 
+
                         /* Advanced functions */
 
 /*
@@ -484,55 +522,57 @@
                                      int  memLevel,
                                      int  strategy));
 
-     This is another version of deflateInit with more compression options. The
-   fields next_in, zalloc, zfree and opaque must be initialized before by
-   the caller.
+     This is another version of deflateInit with more compression options.  The
+   fields next_in, zalloc, zfree and opaque must be initialized before by the
+   caller.
 
-     The method parameter is the compression method. It must be Z_DEFLATED in
+     The method parameter is the compression method.  It must be Z_DEFLATED in
    this version of the library.
 
      The windowBits parameter is the base two logarithm of the window size
-   (the size of the history buffer). It should be in the range 8..15 for this
-   version of the library. Larger values of this parameter result in better
-   compression at the expense of memory usage. The default value is 15 if
+   (the size of the history buffer).  It should be in the range 8..15 for this
+   version of the library.  Larger values of this parameter result in better
+   compression at the expense of memory usage.  The default value is 15 if
    deflateInit is used instead.
 
-     windowBits can also be -8..-15 for raw deflate. In this case, -windowBits
-   determines the window size. deflate() will then generate raw deflate data
+     windowBits can also be -8..-15 for raw deflate.  In this case, -windowBits
+   determines the window size.  deflate() will then generate raw deflate data
    with no zlib header or trailer, and will not compute an adler32 check value.
 
-     windowBits can also be greater than 15 for optional gzip encoding. Add
+     windowBits can also be greater than 15 for optional gzip encoding.  Add
    16 to windowBits to write a simple gzip header and trailer around the
-   compressed data instead of a zlib wrapper. The gzip header will have no
-   file name, no extra data, no comment, no modification time (set to zero),
-   no header crc, and the operating system will be set to 255 (unknown).  If a
+   compressed data instead of a zlib wrapper.  The gzip header will have no
+   file name, no extra data, no comment, no modification time (set to zero), no
+   header crc, and the operating system will be set to 255 (unknown).  If a
    gzip stream is being written, strm->adler is a crc32 instead of an adler32.
 
      The memLevel parameter specifies how much memory should be allocated
-   for the internal compression state. memLevel=1 uses minimum memory but
-   is slow and reduces compression ratio; memLevel=9 uses maximum memory
-   for optimal speed. The default value is 8. See zconf.h for total memory
-   usage as a function of windowBits and memLevel.
+   for the internal compression state.  memLevel=1 uses minimum memory but is
+   slow and reduces compression ratio; memLevel=9 uses maximum memory for
+   optimal speed.  The default value is 8.  See zconf.h for total memory usage
+   as a function of windowBits and memLevel.
 
-     The strategy parameter is used to tune the compression algorithm. Use the
+     The strategy parameter is used to tune the compression algorithm.  Use the
    value Z_DEFAULT_STRATEGY for normal data, Z_FILTERED for data produced by a
    filter (or predictor), Z_HUFFMAN_ONLY to force Huffman encoding only (no
    string match), or Z_RLE to limit match distances to one (run-length
-   encoding). Filtered data consists mostly of small values with a somewhat
-   random distribution. In this case, the compression algorithm is tuned to
-   compress them better. The effect of Z_FILTERED is to force more Huffman
+   encoding).  Filtered data consists mostly of small values with a somewhat
+   random distribution.  In this case, the compression algorithm is tuned to
+   compress them better.  The effect of Z_FILTERED is to force more Huffman
    coding and less string matching; it is somewhat intermediate between
-   Z_DEFAULT and Z_HUFFMAN_ONLY. Z_RLE is designed to be almost as fast as
-   Z_HUFFMAN_ONLY, but give better compression for PNG image data. The strategy
-   parameter only affects the compression ratio but not the correctness of the
-   compressed output even if it is not set appropriately.  Z_FIXED prevents the
-   use of dynamic Huffman codes, allowing for a simpler decoder for special
-   applications.
-
-      deflateInit2 returns Z_OK if success, Z_MEM_ERROR if there was not enough
-   memory, Z_STREAM_ERROR if a parameter is invalid (such as an invalid
-   method). msg is set to null if there is no error message.  deflateInit2 does
-   not perform any compression: this will be done by deflate().
+   Z_DEFAULT_STRATEGY and Z_HUFFMAN_ONLY.  Z_RLE is designed to be almost as
+   fast as Z_HUFFMAN_ONLY, but give better compression for PNG image data.  The
+   strategy parameter only affects the compression ratio but not the
+   correctness of the compressed output even if it is not set appropriately.
+   Z_FIXED prevents the use of dynamic Huffman codes, allowing for a simpler
+   decoder for special applications.
+
+     deflateInit2 returns Z_OK if success, Z_MEM_ERROR if there was not enough
+   memory, Z_STREAM_ERROR if any parameter is invalid (such as an invalid
+   method), or Z_VERSION_ERROR if the zlib library version (zlib_version) is
+   incompatible with the version assumed by the caller (ZLIB_VERSION).  msg is
+   set to null if there is no error message.  deflateInit2 does not perform any
+   compression: this will be done by deflate().
 */
 
 ZEXTERN int ZEXPORT deflateSetDictionary OF((z_streamp strm,
@@ -540,37 +580,37 @@
                                              uInt  dictLength));
 /*
      Initializes the compression dictionary from the given byte sequence
-   without producing any compressed output. This function must be called
-   immediately after deflateInit, deflateInit2 or deflateReset, before any
-   call of deflate. The compressor and decompressor must use exactly the same
+   without producing any compressed output.  This function must be called
+   immediately after deflateInit, deflateInit2 or deflateReset, before any call
+   of deflate.  The compressor and decompressor must use exactly the same
    dictionary (see inflateSetDictionary).
 
      The dictionary should consist of strings (byte sequences) that are likely
    to be encountered later in the data to be compressed, with the most commonly
-   used strings preferably put towards the end of the dictionary. Using a
+   used strings preferably put towards the end of the dictionary.  Using a
    dictionary is most useful when the data to be compressed is short and can be
    predicted with good accuracy; the data can then be compressed better than
    with the default empty dictionary.
 
      Depending on the size of the compression data structures selected by
    deflateInit or deflateInit2, a part of the dictionary may in effect be
-   discarded, for example if the dictionary is larger than the window size in
-   deflate or deflate2. Thus the strings most likely to be useful should be
-   put at the end of the dictionary, not at the front. In addition, the
-   current implementation of deflate will use at most the window size minus
-   262 bytes of the provided dictionary.
+   discarded, for example if the dictionary is larger than the window size
+   provided in deflateInit or deflateInit2.  Thus the strings most likely to be
+   useful should be put at the end of the dictionary, not at the front.  In
+   addition, the current implementation of deflate will use at most the window
+   size minus 262 bytes of the provided dictionary.
 
      Upon return of this function, strm->adler is set to the adler32 value
    of the dictionary; the decompressor may later use this value to determine
-   which dictionary has been used by the compressor. (The adler32 value
+   which dictionary has been used by the compressor.  (The adler32 value
    applies to the whole dictionary even if only a subset of the dictionary is
    actually used by the compressor.) If a raw deflate was requested, then the
    adler32 value is not computed and strm->adler is not set.
 
      deflateSetDictionary returns Z_OK if success, or Z_STREAM_ERROR if a
-   parameter is invalid (such as NULL dictionary) or the stream state is
+   parameter is invalid (e.g.  dictionary being Z_NULL) or the stream state is
    inconsistent (for example if deflate has already been called for this stream
-   or if the compression method is bsort). deflateSetDictionary does not
+   or if the compression method is bsort).  deflateSetDictionary does not
    perform any compression: this will be done by deflate().
 */
 
@@ -581,26 +621,26 @@
 
      This function can be useful when several compression strategies will be
    tried, for example when there are several ways of pre-processing the input
-   data with a filter. The streams that will be discarded should then be freed
+   data with a filter.  The streams that will be discarded should then be freed
    by calling deflateEnd.  Note that deflateCopy duplicates the internal
-   compression state which can be quite large, so this strategy is slow and
-   can consume lots of memory.
+   compression state which can be quite large, so this strategy is slow and can
+   consume lots of memory.
 
      deflateCopy returns Z_OK if success, Z_MEM_ERROR if there was not
    enough memory, Z_STREAM_ERROR if the source stream state was inconsistent
-   (such as zalloc being NULL). msg is left unchanged in both source and
+   (such as zalloc being Z_NULL).  msg is left unchanged in both source and
    destination.
 */
 
 ZEXTERN int ZEXPORT deflateReset OF((z_streamp strm));
 /*
      This function is equivalent to deflateEnd followed by deflateInit,
-   but does not free and reallocate all the internal compression state.
-   The stream will keep the same compression level and any other attributes
-   that may have been set by deflateInit2.
+   but does not free and reallocate all the internal compression state.  The
+   stream will keep the same compression level and any other attributes that
+   may have been set by deflateInit2.
 
-      deflateReset returns Z_OK if success, or Z_STREAM_ERROR if the source
-   stream state was inconsistent (such as zalloc or state being NULL).
+     deflateReset returns Z_OK if success, or Z_STREAM_ERROR if the source
+   stream state was inconsistent (such as zalloc or state being Z_NULL).
 */
 
 ZEXTERN int ZEXPORT deflateParams OF((z_streamp strm,
@@ -610,18 +650,18 @@
      Dynamically update the compression level and compression strategy.  The
    interpretation of level and strategy is as in deflateInit2.  This can be
    used to switch between compression and straight copy of the input data, or
-   to switch to a different kind of input data requiring a different
-   strategy. If the compression level is changed, the input available so far
-   is compressed with the old level (and may be flushed); the new level will
-   take effect only at the next call of deflate().
+   to switch to a different kind of input data requiring a different strategy.
+   If the compression level is changed, the input available so far is
+   compressed with the old level (and may be flushed); the new level will take
+   effect only at the next call of deflate().
 
      Before the call of deflateParams, the stream state must be set as for
-   a call of deflate(), since the currently available input may have to
-   be compressed and flushed. In particular, strm->avail_out must be non-zero.
+   a call of deflate(), since the currently available input may have to be
+   compressed and flushed.  In particular, strm->avail_out must be non-zero.
 
      deflateParams returns Z_OK if success, Z_STREAM_ERROR if the source
-   stream state was inconsistent or if a parameter was invalid, Z_BUF_ERROR
-   if strm->avail_out was zero.
+   stream state was inconsistent or if a parameter was invalid, Z_BUF_ERROR if
+   strm->avail_out was zero.
 */
 
 ZEXTERN int ZEXPORT deflateTune OF((z_streamp strm,
@@ -645,9 +685,10 @@
                                        uLong sourceLen));
 /*
      deflateBound() returns an upper bound on the compressed size after
-   deflation of sourceLen bytes.  It must be called after deflateInit()
-   or deflateInit2().  This would be used to allocate an output buffer
-   for deflation in a single pass, and so would be called before deflate().
+   deflation of sourceLen bytes.  It must be called after deflateInit() or
+   deflateInit2(), and after deflateSetHeader(), if used.  This would be used
+   to allocate an output buffer for deflation in a single pass, and so would be
+   called before deflate().
 */
 
 ZEXTERN int ZEXPORT deflatePrime OF((z_streamp strm,
@@ -655,21 +696,21 @@
                                      int value));
 /*
      deflatePrime() inserts bits in the deflate output stream.  The intent
-  is that this function is used to start off the deflate output with the
-  bits leftover from a previous deflate stream when appending to it.  As such,
-  this function can only be used for raw deflate, and must be used before the
-  first deflate() call after a deflateInit2() or deflateReset().  bits must be
-  less than or equal to 16, and that many of the least significant bits of
-  value will be inserted in the output.
+   is that this function is used to start off the deflate output with the bits
+   leftover from a previous deflate stream when appending to it.  As such, this
+   function can only be used for raw deflate, and must be used before the first
+   deflate() call after a deflateInit2() or deflateReset().  bits must be less
+   than or equal to 16, and that many of the least significant bits of value
+   will be inserted in the output.
 
-      deflatePrime returns Z_OK if success, or Z_STREAM_ERROR if the source
+     deflatePrime returns Z_OK if success, or Z_STREAM_ERROR if the source
    stream state was inconsistent.
 */
 
 ZEXTERN int ZEXPORT deflateSetHeader OF((z_streamp strm,
                                          gz_headerp head));
 /*
-      deflateSetHeader() provides gzip header information for when a gzip
+     deflateSetHeader() provides gzip header information for when a gzip
    stream is requested by deflateInit2().  deflateSetHeader() may be called
    after deflateInit2() or deflateReset() and before the first call of
    deflate().  The text, time, os, extra field, name, and comment information
@@ -682,11 +723,11 @@
    1.3.x) do not support header crc's, and will report that it is a "multi-part
    gzip file" and give up.
 
-      If deflateSetHeader is not used, the default gzip header has text false,
+     If deflateSetHeader is not used, the default gzip header has text false,
    the time set to zero, and os set to 255, with no extra, name, or comment
    fields.  The gzip header is returned to the default state by deflateReset().
 
-      deflateSetHeader returns Z_OK if success, or Z_STREAM_ERROR if the source
+     deflateSetHeader returns Z_OK if success, or Z_STREAM_ERROR if the source
    stream state was inconsistent.
 */
 
@@ -694,43 +735,50 @@
 ZEXTERN int ZEXPORT inflateInit2 OF((z_streamp strm,
                                      int  windowBits));
 
-     This is another version of inflateInit with an extra parameter. The
+     This is another version of inflateInit with an extra parameter.  The
    fields next_in, avail_in, zalloc, zfree and opaque must be initialized
    before by the caller.
 
      The windowBits parameter is the base two logarithm of the maximum window
    size (the size of the history buffer).  It should be in the range 8..15 for
-   this version of the library. The default value is 15 if inflateInit is used
-   instead. windowBits must be greater than or equal to the windowBits value
+   this version of the library.  The default value is 15 if inflateInit is used
+   instead.  windowBits must be greater than or equal to the windowBits value
    provided to deflateInit2() while compressing, or it must be equal to 15 if
-   deflateInit2() was not used. If a compressed stream with a larger window
+   deflateInit2() was not used.  If a compressed stream with a larger window
    size is given as input, inflate() will return with the error code
    Z_DATA_ERROR instead of trying to allocate a larger window.
 
-     windowBits can also be -8..-15 for raw inflate. In this case, -windowBits
-   determines the window size. inflate() will then process raw deflate data,
+     windowBits can also be zero to request that inflate use the window size in
+   the zlib header of the compressed stream.
+
+     windowBits can also be -8..-15 for raw inflate.  In this case, -windowBits
+   determines the window size.  inflate() will then process raw deflate data,
    not looking for a zlib or gzip header, not generating a check value, and not
-   looking for any check values for comparison at the end of the stream. This
+   looking for any check values for comparison at the end of the stream.  This
    is for use with other formats that use the deflate compressed data format
-   such as zip.  Those formats provide their own check values. If a custom
+   such as zip.  Those formats provide their own check values.  If a custom
    format is developed using the raw deflate format for compressed data, it is
    recommended that a check value such as an adler32 or a crc32 be applied to
    the uncompressed data as is done in the zlib, gzip, and zip formats.  For
-   most applications, the zlib format should be used as is. Note that comments
+   most applications, the zlib format should be used as is.  Note that comments
    above on the use in deflateInit2() applies to the magnitude of windowBits.
 
-     windowBits can also be greater than 15 for optional gzip decoding. Add
+     windowBits can also be greater than 15 for optional gzip decoding.  Add
    32 to windowBits to enable zlib and gzip decoding with automatic header
    detection, or add 16 to decode only the gzip format (the zlib format will
-   return a Z_DATA_ERROR).  If a gzip stream is being decoded, strm->adler is
-   a crc32 instead of an adler32.
+   return a Z_DATA_ERROR).  If a gzip stream is being decoded, strm->adler is a
+   crc32 instead of an adler32.
 
      inflateInit2 returns Z_OK if success, Z_MEM_ERROR if there was not enough
-   memory, Z_STREAM_ERROR if a parameter is invalid (such as a null strm). msg
-   is set to null if there is no error message.  inflateInit2 does not perform
-   any decompression apart from reading the zlib header if present: this will
-   be done by inflate(). (So next_in and avail_in may be modified, but next_out
-   and avail_out are unchanged.)
+   memory, Z_VERSION_ERROR if the zlib library version is incompatible with the
+   version assumed by the caller, or Z_STREAM_ERROR if the parameters are
+   invalid, such as a null pointer to the structure.  msg is set to null if
+   there is no error message.  inflateInit2 does not perform any decompression
+   apart from possibly reading the zlib header if present: actual decompression
+   will be done by inflate().  (So next_in and avail_in may be modified, but
+   next_out and avail_out are unused and unchanged.) The current implementation
+   of inflateInit2() does not process any header information -- that is
+   deferred until inflate() is called.
 */
 
 ZEXTERN int ZEXPORT inflateSetDictionary OF((z_streamp strm,
@@ -738,8 +786,8 @@
                                              uInt  dictLength));
 /*
      Initializes the decompression dictionary from the given uncompressed byte
-   sequence. This function must be called immediately after a call of inflate,
-   if that call returned Z_NEED_DICT. The dictionary chosen by the compressor
+   sequence.  This function must be called immediately after a call of inflate,
+   if that call returned Z_NEED_DICT.  The dictionary chosen by the compressor
    can be determined from the adler32 value returned by that call of inflate.
    The compressor and decompressor must use exactly the same dictionary (see
    deflateSetDictionary).  For raw inflate, this function can be called
@@ -748,26 +796,26 @@
    dictionary that was used for compression is provided.
 
      inflateSetDictionary returns Z_OK if success, Z_STREAM_ERROR if a
-   parameter is invalid (such as NULL dictionary) or the stream state is
+   parameter is invalid (e.g.  dictionary being Z_NULL) or the stream state is
    inconsistent, Z_DATA_ERROR if the given dictionary doesn't match the
-   expected one (incorrect adler32 value). inflateSetDictionary does not
+   expected one (incorrect adler32 value).  inflateSetDictionary does not
    perform any decompression: this will be done by subsequent calls of
    inflate().
 */
 
 ZEXTERN int ZEXPORT inflateSync OF((z_streamp strm));
 /*
-    Skips invalid compressed data until a full flush point (see above the
-  description of deflate with Z_FULL_FLUSH) can be found, or until all
-  available input is skipped. No output is provided.
-
-    inflateSync returns Z_OK if a full flush point has been found, Z_BUF_ERROR
-  if no more input was provided, Z_DATA_ERROR if no flush point has been found,
-  or Z_STREAM_ERROR if the stream structure was inconsistent. In the success
-  case, the application may save the current current value of total_in which
-  indicates where valid compressed data was found. In the error case, the
-  application may repeatedly call inflateSync, providing more input each time,
-  until success or end of the input data.
+     Skips invalid compressed data until a full flush point (see above the
+   description of deflate with Z_FULL_FLUSH) can be found, or until all
+   available input is skipped.  No output is provided.
+
+     inflateSync returns Z_OK if a full flush point has been found, Z_BUF_ERROR
+   if no more input was provided, Z_DATA_ERROR if no flush point has been
+   found, or Z_STREAM_ERROR if the stream structure was inconsistent.  In the
+   success case, the application may save the current current value of total_in
+   which indicates where valid compressed data was found.  In the error case,
+   the application may repeatedly call inflateSync, providing more input each
+   time, until success or end of the input data.
 */
 
 ZEXTERN int ZEXPORT inflateCopy OF((z_streamp dest,
@@ -782,18 +830,30 @@
 
      inflateCopy returns Z_OK if success, Z_MEM_ERROR if there was not
    enough memory, Z_STREAM_ERROR if the source stream state was inconsistent
-   (such as zalloc being NULL). msg is left unchanged in both source and
+   (such as zalloc being Z_NULL).  msg is left unchanged in both source and
    destination.
 */
 
 ZEXTERN int ZEXPORT inflateReset OF((z_streamp strm));
 /*
      This function is equivalent to inflateEnd followed by inflateInit,
-   but does not free and reallocate all the internal decompression state.
-   The stream will keep attributes that may have been set by inflateInit2.
+   but does not free and reallocate all the internal decompression state.  The
+   stream will keep attributes that may have been set by inflateInit2.
 
-      inflateReset returns Z_OK if success, or Z_STREAM_ERROR if the source
-   stream state was inconsistent (such as zalloc or state being NULL).
+     inflateReset returns Z_OK if success, or Z_STREAM_ERROR if the source
+   stream state was inconsistent (such as zalloc or state being Z_NULL).
+*/
+
+ZEXTERN int ZEXPORT inflateReset2 OF((z_streamp strm,
+                                      int windowBits));
+/*
+     This function is the same as inflateReset, but it also permits changing
+   the wrap and window size requests.  The windowBits parameter is interpreted
+   the same as it is for inflateInit2.
+
+     inflateReset2 returns Z_OK if success, or Z_STREAM_ERROR if the source
+   stream state was inconsistent (such as zalloc or state being Z_NULL), or if
+   the windowBits parameter is invalid.
 */
 
 ZEXTERN int ZEXPORT inflatePrime OF((z_streamp strm,
@@ -801,54 +861,87 @@
                                      int value));
 /*
      This function inserts bits in the inflate input stream.  The intent is
-  that this function is used to start inflating at a bit position in the
-  middle of a byte.  The provided bits will be used before any bytes are used
-  from next_in.  This function should only be used with raw inflate, and
-  should be used before the first inflate() call after inflateInit2() or
-  inflateReset().  bits must be less than or equal to 16, and that many of the
-  least significant bits of value will be inserted in the input.
+   that this function is used to start inflating at a bit position in the
+   middle of a byte.  The provided bits will be used before any bytes are used
+   from next_in.  This function should only be used with raw inflate, and
+   should be used before the first inflate() call after inflateInit2() or
+   inflateReset().  bits must be less than or equal to 16, and that many of the
+   least significant bits of value will be inserted in the input.
+
+     If bits is negative, then the input stream bit buffer is emptied.  Then
+   inflatePrime() can be called again to put bits in the buffer.  This is used
+   to clear out bits leftover after feeding inflate a block description prior
+   to feeding inflate codes.
 
-      inflatePrime returns Z_OK if success, or Z_STREAM_ERROR if the source
+     inflatePrime returns Z_OK if success, or Z_STREAM_ERROR if the source
    stream state was inconsistent.
 */
 
+ZEXTERN long ZEXPORT inflateMark OF((z_streamp strm));
+/*
+     This function returns two values, one in the lower 16 bits of the return
+   value, and the other in the remaining upper bits, obtained by shifting the
+   return value down 16 bits.  If the upper value is -1 and the lower value is
+   zero, then inflate() is currently decoding information outside of a block.
+   If the upper value is -1 and the lower value is non-zero, then inflate is in
+   the middle of a stored block, with the lower value equaling the number of
+   bytes from the input remaining to copy.  If the upper value is not -1, then
+   it is the number of bits back from the current bit position in the input of
+   the code (literal or length/distance pair) currently being processed.  In
+   that case the lower value is the number of bytes already emitted for that
+   code.
+
+     A code is being processed if inflate is waiting for more input to complete
+   decoding of the code, or if it has completed decoding but is waiting for
+   more output space to write the literal or match data.
+
+     inflateMark() is used to mark locations in the input data for random
+   access, which may be at bit positions, and to note those cases where the
+   output of a code may span boundaries of random access blocks.  The current
+   location in the input stream can be determined from avail_in and data_type
+   as noted in the description for the Z_BLOCK flush parameter for inflate.
+
+     inflateMark returns the value noted above or -1 << 16 if the provided
+   source stream state was inconsistent.
+*/
+
 ZEXTERN int ZEXPORT inflateGetHeader OF((z_streamp strm,
                                          gz_headerp head));
 /*
-      inflateGetHeader() requests that gzip header information be stored in the
+     inflateGetHeader() requests that gzip header information be stored in the
    provided gz_header structure.  inflateGetHeader() may be called after
    inflateInit2() or inflateReset(), and before the first call of inflate().
    As inflate() processes the gzip stream, head->done is zero until the header
    is completed, at which time head->done is set to one.  If a zlib stream is
    being decoded, then head->done is set to -1 to indicate that there will be
-   no gzip header information forthcoming.  Note that Z_BLOCK can be used to
-   force inflate() to return immediately after header processing is complete
-   and before any actual data is decompressed.
+   no gzip header information forthcoming.  Note that Z_BLOCK or Z_TREES can be
+   used to force inflate() to return immediately after header processing is
+   complete and before any actual data is decompressed.
 
-      The text, time, xflags, and os fields are filled in with the gzip header
+     The text, time, xflags, and os fields are filled in with the gzip header
    contents.  hcrc is set to true if there is a header CRC.  (The header CRC
-   was valid if done is set to one.)  If extra is not Z_NULL, then extra_max
+   was valid if done is set to one.) If extra is not Z_NULL, then extra_max
    contains the maximum number of bytes to write to extra.  Once done is true,
    extra_len contains the actual extra field length, and extra contains the
    extra field, or that field truncated if extra_max is less than extra_len.
    If name is not Z_NULL, then up to name_max characters are written there,
    terminated with a zero unless the length is greater than name_max.  If
    comment is not Z_NULL, then up to comm_max characters are written there,
-   terminated with a zero unless the length is greater than comm_max.  When
-   any of extra, name, or comment are not Z_NULL and the respective field is
-   not present in the header, then that field is set to Z_NULL to signal its
+   terminated with a zero unless the length is greater than comm_max.  When any
+   of extra, name, or comment are not Z_NULL and the respective field is not
+   present in the header, then that field is set to Z_NULL to signal its
    absence.  This allows the use of deflateSetHeader() with the returned
    structure to duplicate the header.  However if those fields are set to
    allocated memory, then the application will need to save those pointers
    elsewhere so that they can be eventually freed.
 
-      If inflateGetHeader is not used, then the header information is simply
+     If inflateGetHeader is not used, then the header information is simply
    discarded.  The header is always checked for validity, including the header
    CRC if present.  inflateReset() will reset the process to discard the header
    information.  The application would need to call inflateGetHeader() again to
    retrieve the header from the next gzip stream.
 
-      inflateGetHeader returns Z_OK if success, or Z_STREAM_ERROR if the source
+     inflateGetHeader returns Z_OK if success, or Z_STREAM_ERROR if the source
    stream state was inconsistent.
 */
 
@@ -869,9 +962,9 @@
      See inflateBack() for the usage of these routines.
 
      inflateBackInit will return Z_OK on success, Z_STREAM_ERROR if any of
-   the paramaters are invalid, Z_MEM_ERROR if the internal state could not
-   be allocated, or Z_VERSION_ERROR if the version of the library does not
-   match the version of the header file.
+   the paramaters are invalid, Z_MEM_ERROR if the internal state could not be
+   allocated, or Z_VERSION_ERROR if the version of the library does not match
+   the version of the header file.
 */
 
 typedef unsigned (*in_func) OF((void FAR *, unsigned char FAR * FAR *));
@@ -891,15 +984,15 @@
      inflateBackInit() must be called first to allocate the internal state
    and to initialize the state with the user-provided window buffer.
    inflateBack() may then be used multiple times to inflate a complete, raw
-   deflate stream with each call.  inflateBackEnd() is then called to free
-   the allocated state.
+   deflate stream with each call.  inflateBackEnd() is then called to free the
+   allocated state.
 
      A raw deflate stream is one with no zlib or gzip header or trailer.
    This routine would normally be used in a utility that reads zip or gzip
    files and writes out uncompressed files.  The utility would decode the
-   header and process the trailer on its own, hence this routine expects
-   only the raw deflate stream to decompress.  This is different from the
-   normal behavior of inflate(), which expects either a zlib or gzip header and
+   header and process the trailer on its own, hence this routine expects only
+   the raw deflate stream to decompress.  This is different from the normal
+   behavior of inflate(), which expects either a zlib or gzip header and
    trailer around the deflate stream.
 
      inflateBack() uses two subroutines supplied by the caller that are then
@@ -925,7 +1018,7 @@
    calling inflateBack().  If strm->next_in is Z_NULL, then in() will be called
    immediately for input.  If strm->next_in is not Z_NULL, then strm->avail_in
    must also be initialized, and then if strm->avail_in is not zero, input will
-   initially be taken from strm->next_in[0 .. strm->avail_in - 1].
+   initially be taken from strm->next_in[0 ..  strm->avail_in - 1].
 
      The in_desc and out_desc parameters of inflateBack() is passed as the
    first parameter of in() and out() respectively when they are called.  These
@@ -935,15 +1028,15 @@
      On return, inflateBack() will set strm->next_in and strm->avail_in to
    pass back any unused input that was provided by the last in() call.  The
    return values of inflateBack() can be Z_STREAM_END on success, Z_BUF_ERROR
-   if in() or out() returned an error, Z_DATA_ERROR if there was a format
-   error in the deflate stream (in which case strm->msg is set to indicate the
-   nature of the error), or Z_STREAM_ERROR if the stream was not properly
-   initialized.  In the case of Z_BUF_ERROR, an input or output error can be
-   distinguished using strm->next_in which will be Z_NULL only if in() returned
-   an error.  If strm->next is not Z_NULL, then the Z_BUF_ERROR was due to
-   out() returning non-zero.  (in() will always be called before out(), so
-   strm->next_in is assured to be defined if out() returns non-zero.)  Note
-   that inflateBack() cannot return Z_OK.
+   if in() or out() returned an error, Z_DATA_ERROR if there was a format error
+   in the deflate stream (in which case strm->msg is set to indicate the nature
+   of the error), or Z_STREAM_ERROR if the stream was not properly initialized.
+   In the case of Z_BUF_ERROR, an input or output error can be distinguished
+   using strm->next_in which will be Z_NULL only if in() returned an error.  If
+   strm->next_in is not Z_NULL, then the Z_BUF_ERROR was due to out() returning
+   non-zero.  (in() will always be called before out(), so strm->next_in is
+   assured to be defined if out() returns non-zero.) Note that inflateBack()
+   cannot return Z_OK.
 */
 
 ZEXTERN int ZEXPORT inflateBackEnd OF((z_streamp strm));
@@ -999,23 +1092,22 @@
                         /* utility functions */
 
 /*
-     The following utility functions are implemented on top of the
-   basic stream-oriented functions. To simplify the interface, some
-   default options are assumed (compression level and memory usage,
-   standard memory allocation functions). The source code of these
-   utility functions can easily be modified if you need special options.
+     The following utility functions are implemented on top of the basic
+   stream-oriented functions.  To simplify the interface, some default options
+   are assumed (compression level and memory usage, standard memory allocation
+   functions).  The source code of these utility functions can be modified if
+   you need special options.
 */
 
 ZEXTERN int ZEXPORT compress OF((Bytef *dest,   uLongf *destLen,
                                  const Bytef *source, uLong sourceLen));
 /*
      Compresses the source buffer into the destination buffer.  sourceLen is
-   the byte length of the source buffer. Upon entry, destLen is the total
-   size of the destination buffer, which must be at least the value returned
-   by compressBound(sourceLen). Upon exit, destLen is the actual size of the
+   the byte length of the source buffer.  Upon entry, destLen is the total size
+   of the destination buffer, which must be at least the value returned by
+   compressBound(sourceLen).  Upon exit, destLen is the actual size of the
    compressed buffer.
-     This function can be used to compress a whole file at once if the
-   input file is mmap'ed.
+
      compress returns Z_OK if success, Z_MEM_ERROR if there was not
    enough memory, Z_BUF_ERROR if there was not enough room in the output
    buffer.
@@ -1025,11 +1117,11 @@
                                   const Bytef *source, uLong sourceLen,
                                   int level));
 /*
-     Compresses the source buffer into the destination buffer. The level
+     Compresses the source buffer into the destination buffer.  The level
    parameter has the same meaning as in deflateInit.  sourceLen is the byte
-   length of the source buffer. Upon entry, destLen is the total size of the
+   length of the source buffer.  Upon entry, destLen is the total size of the
    destination buffer, which must be at least the value returned by
-   compressBound(sourceLen). Upon exit, destLen is the actual size of the
+   compressBound(sourceLen).  Upon exit, destLen is the actual size of the
    compressed buffer.
 
      compress2 returns Z_OK if success, Z_MEM_ERROR if there was not enough
@@ -1040,22 +1132,20 @@
 ZEXTERN uLong ZEXPORT compressBound OF((uLong sourceLen));
 /*
      compressBound() returns an upper bound on the compressed size after
-   compress() or compress2() on sourceLen bytes.  It would be used before
-   a compress() or compress2() call to allocate the destination buffer.
+   compress() or compress2() on sourceLen bytes.  It would be used before a
+   compress() or compress2() call to allocate the destination buffer.
 */
 
 ZEXTERN int ZEXPORT uncompress OF((Bytef *dest,   uLongf *destLen,
                                    const Bytef *source, uLong sourceLen));
 /*
      Decompresses the source buffer into the destination buffer.  sourceLen is
-   the byte length of the source buffer. Upon entry, destLen is the total
-   size of the destination buffer, which must be large enough to hold the
-   entire uncompressed data. (The size of the uncompressed data must have
-   been saved previously by the compressor and transmitted to the decompressor
-   by some mechanism outside the scope of this compression library.)
-   Upon exit, destLen is the actual size of the compressed buffer.
-     This function can be used to decompress a whole file at once if the
-   input file is mmap'ed.
+   the byte length of the source buffer.  Upon entry, destLen is the total size
+   of the destination buffer, which must be large enough to hold the entire
+   uncompressed data.  (The size of the uncompressed data must have been saved
+   previously by the compressor and transmitted to the decompressor by some
+   mechanism outside the scope of this compression library.) Upon exit, destLen
+   is the actual size of the uncompressed buffer.
 
      uncompress returns Z_OK if success, Z_MEM_ERROR if there was not
    enough memory, Z_BUF_ERROR if there was not enough room in the output
@@ -1063,136 +1153,199 @@
 */
 
 
-typedef voidp gzFile;
+                        /* gzip file access functions */
+
+/*
+     This library supports reading and writing files in gzip (.gz) format with
+   an interface similar to that of stdio, using the functions that start with
+   "gz".  The gzip format is different from the zlib format.  gzip is a gzip
+   wrapper, documented in RFC 1952, wrapped around a deflate stream.
+*/
+
+typedef voidp gzFile;       /* opaque gzip file descriptor */
 
-ZEXTERN gzFile ZEXPORT gzopen  OF((const char *path, const char *mode));
 /*
-     Opens a gzip (.gz) file for reading or writing. The mode parameter
-   is as in fopen ("rb" or "wb") but can also include a compression level
-   ("wb9") or a strategy: 'f' for filtered data as in "wb6f", 'h' for
-   Huffman only compression as in "wb1h", or 'R' for run-length encoding
-   as in "wb1R". (See the description of deflateInit2 for more information
-   about the strategy parameter.)
+ZEXTERN gzFile ZEXPORT gzopen OF((const char *path, const char *mode));
+
+     Opens a gzip (.gz) file for reading or writing.  The mode parameter is as
+   in fopen ("rb" or "wb") but can also include a compression level ("wb9") or
+   a strategy: 'f' for filtered data as in "wb6f", 'h' for Huffman-only
+   compression as in "wb1h", 'R' for run-length encoding as in "wb1R", or 'F'
+   for fixed code compression as in "wb9F".  (See the description of
+   deflateInit2 for more information about the strategy parameter.) Also "a"
+   can be used instead of "w" to request that the gzip stream that will be
+   written be appended to the file.  "+" will result in an error, since reading
+   and writing to the same gzip file is not supported.
 
      gzopen can be used to read a file which is not in gzip format; in this
    case gzread will directly read from the file without decompression.
 
-     gzopen returns NULL if the file could not be opened or if there was
-   insufficient memory to allocate the (de)compression state; errno
-   can be checked to distinguish the two cases (if errno is zero, the
-   zlib error is Z_MEM_ERROR).  */
-
-ZEXTERN gzFile ZEXPORT gzdopen  OF((int fd, const char *mode));
-/*
-     gzdopen() associates a gzFile with the file descriptor fd.  File
-   descriptors are obtained from calls like open, dup, creat, pipe or
-   fileno (in the file has been previously opened with fopen).
-   The mode parameter is as in gzopen.
-     The next call of gzclose on the returned gzFile will also close the
-   file descriptor fd, just like fclose(fdopen(fd), mode) closes the file
-   descriptor fd. If you want to keep fd open, use gzdopen(dup(fd), mode).
-     gzdopen returns NULL if there was insufficient memory to allocate
-   the (de)compression state.
+     gzopen returns NULL if the file could not be opened, if there was
+   insufficient memory to allocate the gzFile state, or if an invalid mode was
+   specified (an 'r', 'w', or 'a' was not provided, or '+' was provided).
+   errno can be checked to determine if the reason gzopen failed was that the
+   file could not be opened.
+*/
+
+ZEXTERN gzFile ZEXPORT gzdopen OF((int fd, const char *mode));
+/*
+     gzdopen associates a gzFile with the file descriptor fd.  File descriptors
+   are obtained from calls like open, dup, creat, pipe or fileno (if the file
+   has been previously opened with fopen).  The mode parameter is as in gzopen.
+
+     The next call of gzclose on the returned gzFile will also close the file
+   descriptor fd, just like fclose(fdopen(fd, mode)) closes the file descriptor
+   fd.  If you want to keep fd open, use fd = dup(fd_keep); gz = gzdopen(fd,
+   mode);.  The duplicated descriptor should be saved to avoid a leak, since
+   gzdopen does not close fd if it fails.
+
+     gzdopen returns NULL if there was insufficient memory to allocate the
+   gzFile state, if an invalid mode was specified (an 'r', 'w', or 'a' was not
+   provided, or '+' was provided), or if fd is -1.  The file descriptor is not
+   used until the next gz* read, write, seek, or close operation, so gzdopen
+   will not detect if fd is invalid (unless fd is -1).
+*/
+
+ZEXTERN int ZEXPORT gzbuffer OF((gzFile file, unsigned size));
+/*
+     Set the internal buffer size used by this library's functions.  The
+   default buffer size is 8192 bytes.  This function must be called after
+   gzopen() or gzdopen(), and before any other calls that read or write the
+   file.  The buffer memory allocation is always deferred to the first read or
+   write.  Two buffers are allocated, either both of the specified size when
+   writing, or one of the specified size and the other twice that size when
+   reading.  A larger buffer size of, for example, 64K or 128K bytes will
+   noticeably increase the speed of decompression (reading).
+
+     The new buffer size also affects the maximum length for gzprintf().
+
+     gzbuffer() returns 0 on success, or -1 on failure, such as being called
+   too late.
 */
 
 ZEXTERN int ZEXPORT gzsetparams OF((gzFile file, int level, int strategy));
 /*
-     Dynamically update the compression level or strategy. See the description
+     Dynamically update the compression level or strategy.  See the description
    of deflateInit2 for the meaning of these parameters.
+
      gzsetparams returns Z_OK if success, or Z_STREAM_ERROR if the file was not
    opened for writing.
 */
 
-ZEXTERN int ZEXPORT    gzread  OF((gzFile file, voidp buf, unsigned len));
+ZEXTERN int ZEXPORT gzread OF((gzFile file, voidp buf, unsigned len));
 /*
-     Reads the given number of uncompressed bytes from the compressed file.
-   If the input file was not in gzip format, gzread copies the given number
-   of bytes into the buffer.
-     gzread returns the number of uncompressed bytes actually read (0 for
-   end of file, -1 for error). */
+     Reads the given number of uncompressed bytes from the compressed file.  If
+   the input file was not in gzip format, gzread copies the given number of
+   bytes into the buffer.
+
+     After reaching the end of a gzip stream in the input, gzread will continue
+   to read, looking for another gzip stream, or failing that, reading the rest
+   of the input file directly without decompression.  The entire input file
+   will be read if gzread is called until it returns less than the requested
+   len.
 
-ZEXTERN int ZEXPORT    gzwrite OF((gzFile file,
-                                   voidpc buf, unsigned len));
+     gzread returns the number of uncompressed bytes actually read, less than
+   len for end of file, or -1 for error.
+*/
+
+ZEXTERN int ZEXPORT gzwrite OF((gzFile file,
+                                voidpc buf, unsigned len));
 /*
      Writes the given number of uncompressed bytes into the compressed file.
-   gzwrite returns the number of uncompressed bytes actually written
-   (0 in case of error).
+   gzwrite returns the number of uncompressed bytes written or 0 in case of
+   error.
 */
 
-ZEXTERN int ZEXPORTVA   gzprintf OF((gzFile file, const char *format, ...));
+ZEXTERN int ZEXPORTVA gzprintf OF((gzFile file, const char *format, ...));
 /*
-     Converts, formats, and writes the args to the compressed file under
-   control of the format string, as in fprintf. gzprintf returns the number of
-   uncompressed bytes actually written (0 in case of error).  The number of
-   uncompressed bytes written is limited to 4095. The caller should assure that
-   this limit is not exceeded. If it is exceeded, then gzprintf() will return
-   return an error (0) with nothing written. In this case, there may also be a
-   buffer overflow with unpredictable consequences, which is possible only if
-   zlib was compiled with the insecure functions sprintf() or vsprintf()
-   because the secure snprintf() or vsnprintf() functions were not available.
+     Converts, formats, and writes the arguments to the compressed file under
+   control of the format string, as in fprintf.  gzprintf returns the number of
+   uncompressed bytes actually written, or 0 in case of error.  The number of
+   uncompressed bytes written is limited to 8191, or one less than the buffer
+   size given to gzbuffer().  The caller should assure that this limit is not
+   exceeded.  If it is exceeded, then gzprintf() will return an error (0) with
+   nothing written.  In this case, there may also be a buffer overflow with
+   unpredictable consequences, which is possible only if zlib was compiled with
+   the insecure functions sprintf() or vsprintf() because the secure snprintf()
+   or vsnprintf() functions were not available.  This can be determined using
+   zlibCompileFlags().
 */
 
 ZEXTERN int ZEXPORT gzputs OF((gzFile file, const char *s));
 /*
-      Writes the given null-terminated string to the compressed file, excluding
+     Writes the given null-terminated string to the compressed file, excluding
    the terminating null character.
-      gzputs returns the number of characters written, or -1 in case of error.
+
+     gzputs returns the number of characters written, or -1 in case of error.
 */
 
 ZEXTERN char * ZEXPORT gzgets OF((gzFile file, char *buf, int len));
 /*
-      Reads bytes from the compressed file until len-1 characters are read, or
-   a newline character is read and transferred to buf, or an end-of-file
-   condition is encountered.  The string is then terminated with a null
-   character.
-      gzgets returns buf, or Z_NULL in case of error.
+     Reads bytes from the compressed file until len-1 characters are read, or a
+   newline character is read and transferred to buf, or an end-of-file
+   condition is encountered.  If any characters are read or if len == 1, the
+   string is terminated with a null character.  If no characters are read due
+   to an end-of-file or len < 1, then the buffer is left untouched.
+
+     gzgets returns buf which is a null-terminated string, or it returns NULL
+   for end-of-file or in case of error.  If there was an error, the contents at
+   buf are indeterminate.
 */
 
-ZEXTERN int ZEXPORT    gzputc OF((gzFile file, int c));
+ZEXTERN int ZEXPORT gzputc OF((gzFile file, int c));
 /*
-      Writes c, converted to an unsigned char, into the compressed file.
-   gzputc returns the value that was written, or -1 in case of error.
+     Writes c, converted to an unsigned char, into the compressed file.  gzputc
+   returns the value that was written, or -1 in case of error.
 */
 
-ZEXTERN int ZEXPORT    gzgetc OF((gzFile file));
+ZEXTERN int ZEXPORT gzgetc OF((gzFile file));
 /*
-      Reads one byte from the compressed file. gzgetc returns this byte
-   or -1 in case of end of file or error.
+     Reads one byte from the compressed file.  gzgetc returns this byte or -1
+   in case of end of file or error.
 */
 
-ZEXTERN int ZEXPORT    gzungetc OF((int c, gzFile file));
+ZEXTERN int ZEXPORT gzungetc OF((int c, gzFile file));
 /*
-      Push one character back onto the stream to be read again later.
-   Only one character of push-back is allowed.  gzungetc() returns the
-   character pushed, or -1 on failure.  gzungetc() will fail if a
-   character has been pushed but not read yet, or if c is -1. The pushed
-   character will be discarded if the stream is repositioned with gzseek()
-   or gzrewind().
+     Push one character back onto the stream to be read as the first character
+   on the next read.  At least one character of push-back is allowed.
+   gzungetc() returns the character pushed, or -1 on failure.  gzungetc() will
+   fail if c is -1, and may fail if a character has been pushed but not read
+   yet.  If gzungetc is used immediately after gzopen or gzdopen, at least the
+   output buffer size of pushed characters is allowed.  (See gzbuffer above.)
+   The pushed character will be discarded if the stream is repositioned with
+   gzseek() or gzrewind().
 */
 
-ZEXTERN int ZEXPORT    gzflush OF((gzFile file, int flush));
+ZEXTERN int ZEXPORT gzflush OF((gzFile file, int flush));
 /*
-     Flushes all pending output into the compressed file. The parameter
-   flush is as in the deflate() function. The return value is the zlib
-   error number (see function gzerror below). gzflush returns Z_OK if
-   the flush parameter is Z_FINISH and all output could be flushed.
-     gzflush should be called only when strictly necessary because it can
-   degrade compression.
+     Flushes all pending output into the compressed file.  The parameter flush
+   is as in the deflate() function.  The return value is the zlib error number
+   (see function gzerror below).  gzflush is only permitted when writing.
+
+     If the flush parameter is Z_FINISH, the remaining data is written and the
+   gzip stream is completed in the output.  If gzwrite() is called again, a new
+   gzip stream will be started in the output.  gzread() is able to read such
+   concatented gzip streams.
+
+     gzflush should be called only when strictly necessary because it will
+   degrade compression if called too often.
 */
 
-ZEXTERN z_off_t ZEXPORT    gzseek OF((gzFile file,
-                                      z_off_t offset, int whence));
 /*
-      Sets the starting position for the next gzread or gzwrite on the
-   given compressed file. The offset represents a number of bytes in the
-   uncompressed data stream. The whence parameter is defined as in lseek(2);
+ZEXTERN z_off_t ZEXPORT gzseek OF((gzFile file,
+                                   z_off_t offset, int whence));
+
+     Sets the starting position for the next gzread or gzwrite on the given
+   compressed file.  The offset represents a number of bytes in the
+   uncompressed data stream.  The whence parameter is defined as in lseek(2);
    the value SEEK_END is not supported.
+
      If the file is opened for reading, this function is emulated but can be
-   extremely slow. If the file is opened for writing, only forward seeks are
+   extremely slow.  If the file is opened for writing, only forward seeks are
    supported; gzseek then compresses a sequence of zeroes up to the new
    starting position.
 
-      gzseek returns the resulting offset location as measured in bytes from
+     gzseek returns the resulting offset location as measured in bytes from
    the beginning of the uncompressed stream, or -1 in case of error, in
    particular if the file is opened for writing and the new starting position
    would be before the current position.
@@ -1202,68 +1355,127 @@
 /*
      Rewinds the given file. This function is supported only for reading.
 
-   gzrewind(file) is equivalent to (int)gzseek(file, 0L, SEEK_SET)
+     gzrewind(file) is equivalent to (int)gzseek(file, 0L, SEEK_SET)
 */
 
+/*
 ZEXTERN z_off_t ZEXPORT    gztell OF((gzFile file));
+
+     Returns the starting position for the next gzread or gzwrite on the given
+   compressed file.  This position represents a number of bytes in the
+   uncompressed data stream, and is zero when starting, even if appending or
+   reading a gzip stream from the middle of a file using gzdopen().
+
+     gztell(file) is equivalent to gzseek(file, 0L, SEEK_CUR)
+*/
+
 /*
-     Returns the starting position for the next gzread or gzwrite on the
-   given compressed file. This position represents a number of bytes in the
-   uncompressed data stream.
+ZEXTERN z_off_t ZEXPORT gzoffset OF((gzFile file));
 
-   gztell(file) is equivalent to gzseek(file, 0L, SEEK_CUR)
+     Returns the current offset in the file being read or written.  This offset
+   includes the count of bytes that precede the gzip stream, for example when
+   appending or when using gzdopen() for reading.  When reading, the offset
+   does not include as yet unused buffered input.  This information can be used
+   for a progress indicator.  On error, gzoffset() returns -1.
 */
 
 ZEXTERN int ZEXPORT gzeof OF((gzFile file));
 /*
-     Returns 1 when EOF has previously been detected reading the given
-   input stream, otherwise zero.
+     Returns true (1) if the end-of-file indicator has been set while reading,
+   false (0) otherwise.  Note that the end-of-file indicator is set only if the
+   read tried to go past the end of the input, but came up short.  Therefore,
+   just like feof(), gzeof() may return false even if there is no more data to
+   read, in the event that the last read request was for the exact number of
+   bytes remaining in the input file.  This will happen if the input file size
+   is an exact multiple of the buffer size.
+
+     If gzeof() returns true, then the read functions will return no more data,
+   unless the end-of-file indicator is reset by gzclearerr() and the input file
+   has grown since the previous end of file was detected.
 */
 
 ZEXTERN int ZEXPORT gzdirect OF((gzFile file));
 /*
-     Returns 1 if file is being read directly without decompression, otherwise
-   zero.
+     Returns true (1) if file is being copied directly while reading, or false
+   (0) if file is a gzip stream being decompressed.  This state can change from
+   false to true while reading the input file if the end of a gzip stream is
+   reached, but is followed by data that is not another gzip stream.
+
+     If the input file is empty, gzdirect() will return true, since the input
+   does not contain a gzip stream.
+
+     If gzdirect() is used immediately after gzopen() or gzdopen() it will
+   cause buffers to be allocated to allow reading the file to determine if it
+   is a gzip file.  Therefore if gzbuffer() is used, it should be called before
+   gzdirect().
 */
 
 ZEXTERN int ZEXPORT    gzclose OF((gzFile file));
 /*
-     Flushes all pending output if necessary, closes the compressed file
-   and deallocates all the (de)compression state. The return value is the zlib
-   error number (see function gzerror below).
+     Flushes all pending output if necessary, closes the compressed file and
+   deallocates the (de)compression state.  Note that once file is closed, you
+   cannot call gzerror with file, since its structures have been deallocated.
+   gzclose must not be called more than once on the same file, just as free
+   must not be called more than once on the same allocation.
+
+     gzclose will return Z_STREAM_ERROR if file is not valid, Z_ERRNO on a
+   file operation error, or Z_OK on success.
+*/
+
+ZEXTERN int ZEXPORT gzclose_r OF((gzFile file));
+ZEXTERN int ZEXPORT gzclose_w OF((gzFile file));
+/*
+     Same as gzclose(), but gzclose_r() is only for use when reading, and
+   gzclose_w() is only for use when writing or appending.  The advantage to
+   using these instead of gzclose() is that they avoid linking in zlib
+   compression or decompression code that is not used when only reading or only
+   writing respectively.  If gzclose() is used, then both compression and
+   decompression code will be included the application when linking to a static
+   zlib library.
 */
 
 ZEXTERN const char * ZEXPORT gzerror OF((gzFile file, int *errnum));
 /*
-     Returns the error message for the last error which occurred on the
-   given compressed file. errnum is set to zlib error number. If an
-   error occurred in the file system and not in the compression library,
-   errnum is set to Z_ERRNO and the application may consult errno
-   to get the exact error code.
+     Returns the error message for the last error which occurred on the given
+   compressed file.  errnum is set to zlib error number.  If an error occurred
+   in the file system and not in the compression library, errnum is set to
+   Z_ERRNO and the application may consult errno to get the exact error code.
+
+     The application must not modify the returned string.  Future calls to
+   this function may invalidate the previously returned string.  If file is
+   closed, then the string previously returned by gzerror will no longer be
+   available.
+
+     gzerror() should be used to distinguish errors from end-of-file for those
+   functions above that do not distinguish those cases in their return values.
 */
 
 ZEXTERN void ZEXPORT gzclearerr OF((gzFile file));
 /*
-     Clears the error and end-of-file flags for file. This is analogous to the
-   clearerr() function in stdio. This is useful for continuing to read a gzip
+     Clears the error and end-of-file flags for file.  This is analogous to the
+   clearerr() function in stdio.  This is useful for continuing to read a gzip
    file that is being written concurrently.
 */
 
+
                         /* checksum functions */
 
 /*
      These functions are not related to compression but are exported
-   anyway because they might be useful in applications using the
-   compression library.
+   anyway because they might be useful in applications using the compression
+   library.
 */
 
 ZEXTERN uLong ZEXPORT adler32 OF((uLong adler, const Bytef *buf, uInt len));
 /*
      Update a running Adler-32 checksum with the bytes buf[0..len-1] and
-   return the updated checksum. If buf is NULL, this function returns
-   the required initial value for the checksum.
-   An Adler-32 checksum is almost as reliable as a CRC32 but can be computed
-   much faster. Usage example:
+   return the updated checksum.  If buf is Z_NULL, this function returns the
+   required initial value for the checksum.
+
+     An Adler-32 checksum is almost as reliable as a CRC32 but can be computed
+   much faster.
+
+   Usage example:
 
      uLong adler = adler32(0L, Z_NULL, 0);
 
@@ -1273,9 +1485,10 @@
      if (adler != original_adler) error();
 */
 
+/*
 ZEXTERN uLong ZEXPORT adler32_combine OF((uLong adler1, uLong adler2,
                                           z_off_t len2));
-/*
+
      Combine two Adler-32 checksums into one.  For two sequences of bytes, seq1
    and seq2 with lengths len1 and len2, Adler-32 checksums were calculated for
    each, adler1 and adler2.  adler32_combine() returns the Adler-32 checksum of
@@ -1285,9 +1498,11 @@
 ZEXTERN uLong ZEXPORT crc32   OF((uLong crc, const Bytef *buf, uInt len));
 /*
      Update a running CRC-32 with the bytes buf[0..len-1] and return the
-   updated CRC-32. If buf is NULL, this function returns the required initial
-   value for the for the crc. Pre- and post-conditioning (one's complement) is
-   performed within this function so it shouldn't be done by the application.
+   updated CRC-32.  If buf is Z_NULL, this function returns the required
+   initial value for the for the crc.  Pre- and post-conditioning (one's
+   complement) is performed within this function so it shouldn't be done by the
+   application.
+
    Usage example:
 
      uLong crc = crc32(0L, Z_NULL, 0);
@@ -1298,9 +1513,9 @@
      if (crc != original_crc) error();
 */
 
+/*
 ZEXTERN uLong ZEXPORT crc32_combine OF((uLong crc1, uLong crc2, z_off_t len2));
 
-/*
      Combine two CRC-32 check values into one.  For two sequences of bytes,
    seq1 and seq2 with lengths len1 and len2, CRC-32 check values were
    calculated for each, crc1 and crc2.  crc32_combine() returns the CRC-32
@@ -1339,16 +1554,57 @@
         inflateInit2_((strm), (windowBits), ZLIB_VERSION, sizeof(z_stream))
 #define inflateBackInit(strm, windowBits, window) \
         inflateBackInit_((strm), (windowBits), (window), \
-        ZLIB_VERSION, sizeof(z_stream))
+                                            ZLIB_VERSION, sizeof(z_stream))
+
+/* provide 64-bit offset functions if _LARGEFILE64_SOURCE defined, and/or
+ * change the regular functions to 64 bits if _FILE_OFFSET_BITS is 64 (if
+ * both are true, the application gets the *64 functions, and the regular
+ * functions are changed to 64 bits) -- in case these are set on systems
+ * without large file support, _LFS64_LARGEFILE must also be true
+ */
+#if defined(_LARGEFILE64_SOURCE) && _LFS64_LARGEFILE-0
+   ZEXTERN gzFile ZEXPORT gzopen64 OF((const char *, const char *));
+   ZEXTERN z_off64_t ZEXPORT gzseek64 OF((gzFile, z_off64_t, int));
+   ZEXTERN z_off64_t ZEXPORT gztell64 OF((gzFile));
+   ZEXTERN z_off64_t ZEXPORT gzoffset64 OF((gzFile));
+   ZEXTERN uLong ZEXPORT adler32_combine64 OF((uLong, uLong, z_off64_t));
+   ZEXTERN uLong ZEXPORT crc32_combine64 OF((uLong, uLong, z_off64_t));
+#endif
 
+#if !defined(ZLIB_INTERNAL) && _FILE_OFFSET_BITS-0 == 64 && _LFS64_LARGEFILE-0
+#  define gzopen gzopen64
+#  define gzseek gzseek64
+#  define gztell gztell64
+#  define gzoffset gzoffset64
+#  define adler32_combine adler32_combine64
+#  define crc32_combine crc32_combine64
+#  ifdef _LARGEFILE64_SOURCE
+     ZEXTERN gzFile ZEXPORT gzopen64 OF((const char *, const char *));
+     ZEXTERN z_off_t ZEXPORT gzseek64 OF((gzFile, z_off_t, int));
+     ZEXTERN z_off_t ZEXPORT gztell64 OF((gzFile));
+     ZEXTERN z_off_t ZEXPORT gzoffset64 OF((gzFile));
+     ZEXTERN uLong ZEXPORT adler32_combine64 OF((uLong, uLong, z_off_t));
+     ZEXTERN uLong ZEXPORT crc32_combine64 OF((uLong, uLong, z_off_t));
+#  endif
+#else
+   ZEXTERN gzFile ZEXPORT gzopen OF((const char *, const char *));
+   ZEXTERN z_off_t ZEXPORT gzseek OF((gzFile, z_off_t, int));
+   ZEXTERN z_off_t ZEXPORT gztell OF((gzFile));
+   ZEXTERN z_off_t ZEXPORT gzoffset OF((gzFile));
+   ZEXTERN uLong ZEXPORT adler32_combine OF((uLong, uLong, z_off_t));
+   ZEXTERN uLong ZEXPORT crc32_combine OF((uLong, uLong, z_off_t));
+#endif
 
+/* hack for buggy compilers */
 #if !defined(ZUTIL_H) && !defined(NO_DUMMY_DECL)
-    struct internal_state {int dummy;}; /* hack for buggy compilers */
+    struct internal_state {int dummy;};
 #endif
 
+/* undocumented functions */
 ZEXTERN const char   * ZEXPORT zError           OF((int));
-ZEXTERN int            ZEXPORT inflateSyncPoint OF((z_streamp z));
+ZEXTERN int            ZEXPORT inflateSyncPoint OF((z_streamp));
 ZEXTERN const uLongf * ZEXPORT get_crc_table    OF((void));
+ZEXTERN int            ZEXPORT inflateUndermine OF((z_streamp, int));
 
 #ifdef __cplusplus
 }

Modified: python/branches/py3k/Modules/zlib/zutil.c
==============================================================================
--- python/branches/py3k/Modules/zlib/zutil.c	(original)
+++ python/branches/py3k/Modules/zlib/zutil.c	Fri Jul 30 22:03:17 2010
@@ -1,5 +1,5 @@
 /* zutil.c -- target dependent utility functions for the compression library
- * Copyright (C) 1995-2005 Jean-loup Gailly.
+ * Copyright (C) 1995-2005, 2010 Jean-loup Gailly.
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -34,25 +34,25 @@
     uLong flags;
 
     flags = 0;
-    switch (sizeof(uInt)) {
+    switch ((int)(sizeof(uInt))) {
     case 2:     break;
     case 4:     flags += 1;     break;
     case 8:     flags += 2;     break;
     default:    flags += 3;
     }
-    switch (sizeof(uLong)) {
+    switch ((int)(sizeof(uLong))) {
     case 2:     break;
     case 4:     flags += 1 << 2;        break;
     case 8:     flags += 2 << 2;        break;
     default:    flags += 3 << 2;
     }
-    switch (sizeof(voidpf)) {
+    switch ((int)(sizeof(voidpf))) {
     case 2:     break;
     case 4:     flags += 1 << 4;        break;
     case 8:     flags += 2 << 4;        break;
     default:    flags += 3 << 4;
     }
-    switch (sizeof(z_off_t)) {
+    switch ((int)(sizeof(z_off_t))) {
     case 2:     break;
     case 4:     flags += 1 << 6;        break;
     case 8:     flags += 2 << 6;        break;
@@ -117,9 +117,9 @@
 #  ifndef verbose
 #    define verbose 0
 #  endif
-int z_verbose = verbose;
+int ZLIB_INTERNAL z_verbose = verbose;
 
-void z_error (m)
+void ZLIB_INTERNAL z_error (m)
     char *m;
 {
     fprintf(stderr, "%s\n", m);
@@ -146,7 +146,7 @@
 
 #ifndef HAVE_MEMCPY
 
-void zmemcpy(dest, source, len)
+void ZLIB_INTERNAL zmemcpy(dest, source, len)
     Bytef* dest;
     const Bytef* source;
     uInt  len;
@@ -157,7 +157,7 @@
     } while (--len != 0);
 }
 
-int zmemcmp(s1, s2, len)
+int ZLIB_INTERNAL zmemcmp(s1, s2, len)
     const Bytef* s1;
     const Bytef* s2;
     uInt  len;
@@ -170,7 +170,7 @@
     return 0;
 }
 
-void zmemzero(dest, len)
+void ZLIB_INTERNAL zmemzero(dest, len)
     Bytef* dest;
     uInt  len;
 {
@@ -213,7 +213,7 @@
  * a protected system like OS/2. Use Microsoft C instead.
  */
 
-voidpf zcalloc (voidpf opaque, unsigned items, unsigned size)
+voidpf ZLIB_INTERNAL zcalloc (voidpf opaque, unsigned items, unsigned size)
 {
     voidpf buf = opaque; /* just to make some compilers happy */
     ulg bsize = (ulg)items*size;
@@ -237,7 +237,7 @@
     return buf;
 }
 
-void  zcfree (voidpf opaque, voidpf ptr)
+void ZLIB_INTERNAL zcfree (voidpf opaque, voidpf ptr)
 {
     int n;
     if (*(ush*)&ptr != 0) { /* object < 64K */
@@ -272,13 +272,13 @@
 #  define _hfree   hfree
 #endif
 
-voidpf zcalloc (voidpf opaque, unsigned items, unsigned size)
+voidpf ZLIB_INTERNAL zcalloc (voidpf opaque, uInt items, uInt size)
 {
     if (opaque) opaque = 0; /* to make compiler happy */
     return _halloc((long)items, size);
 }
 
-void  zcfree (voidpf opaque, voidpf ptr)
+void ZLIB_INTERNAL zcfree (voidpf opaque, voidpf ptr)
 {
     if (opaque) opaque = 0; /* to make compiler happy */
     _hfree(ptr);
@@ -297,7 +297,7 @@
 extern void   free   OF((voidpf ptr));
 #endif
 
-voidpf zcalloc (opaque, items, size)
+voidpf ZLIB_INTERNAL zcalloc (opaque, items, size)
     voidpf opaque;
     unsigned items;
     unsigned size;
@@ -307,7 +307,7 @@
                               (voidpf)calloc(items, size);
 }
 
-void  zcfree (opaque, ptr)
+void ZLIB_INTERNAL zcfree (opaque, ptr)
     voidpf opaque;
     voidpf ptr;
 {

Modified: python/branches/py3k/Modules/zlib/zutil.h
==============================================================================
--- python/branches/py3k/Modules/zlib/zutil.h	(original)
+++ python/branches/py3k/Modules/zlib/zutil.h	Fri Jul 30 22:03:17 2010
@@ -1,5 +1,5 @@
 /* zutil.h -- internal interface and configuration of the compression library
- * Copyright (C) 1995-2005 Jean-loup Gailly.
+ * Copyright (C) 1995-2010 Jean-loup Gailly.
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -13,31 +13,21 @@
 #ifndef ZUTIL_H
 #define ZUTIL_H
 
-#define ZLIB_INTERNAL
+#if ((__GNUC__-0) * 10 + __GNUC_MINOR__-0 >= 33) && !defined(NO_VIZ)
+#  define ZLIB_INTERNAL __attribute__((visibility ("hidden")))
+#else
+#  define ZLIB_INTERNAL
+#endif
+
 #include "zlib.h"
 
 #ifdef STDC
-#  ifndef _WIN32_WCE
+#  if !(defined(_WIN32_WCE) && defined(_MSC_VER))
 #    include <stddef.h>
 #  endif
 #  include <string.h>
 #  include <stdlib.h>
 #endif
-#ifdef NO_ERRNO_H
-#   ifdef _WIN32_WCE
-      /* The Microsoft C Run-Time Library for Windows CE doesn't have
-       * errno.  We define it as a global variable to simplify porting.
-       * Its value is always 0 and should not be used.  We rename it to
-       * avoid conflict with other libraries that use the same workaround.
-       */
-#     define errno z_errno
-#   endif
-    extern int errno;
-#else
-#  ifndef _WIN32_WCE
-#    include <errno.h>
-#  endif
-#endif
 
 #ifndef local
 #  define local static
@@ -89,7 +79,7 @@
 #if defined(MSDOS) || (defined(WINDOWS) && !defined(WIN32))
 #  define OS_CODE  0x00
 #  if defined(__TURBOC__) || defined(__BORLANDC__)
-#    if(__STDC__ == 1) && (defined(__LARGE__) || defined(__COMPACT__))
+#    if (__STDC__ == 1) && (defined(__LARGE__) || defined(__COMPACT__))
        /* Allow compilation with ANSI keywords only enabled */
        void _Cdecl farfree( void *block );
        void *_Cdecl farmalloc( unsigned long nbytes );
@@ -118,7 +108,7 @@
 #ifdef OS2
 #  define OS_CODE  0x06
 #  ifdef M_I86
-     #include <malloc.h>
+#    include <malloc.h>
 #  endif
 #endif
 
@@ -147,11 +137,11 @@
 #  define OS_CODE  0x0f
 #endif
 
-#if defined(_BEOS_)
+#if defined(_BEOS_) || defined(RISCOS)
 #  define fdopen(fd,mode) NULL /* No fdopen() */
 #endif
 
-#if (defined(_MSC_VER) && (_MSC_VER > 600))
+#if (defined(_MSC_VER) && (_MSC_VER > 600)) && !defined __INTERIX
 #  if defined(_WIN32_WCE)
 #    define fdopen(fd,mode) NULL /* No fdopen() */
 #    ifndef _PTRDIFF_T_DEFINED
@@ -163,6 +153,18 @@
 #  endif
 #endif
 
+#if defined(__BORLANDC__)
+  #pragma warn -8004
+  #pragma warn -8008
+  #pragma warn -8066
+#endif
+
+/* provide prototypes for these when building zlib without LFS */
+#if !defined(_LARGEFILE64_SOURCE) || _LFS64_LARGEFILE-0 == 0
+    ZEXTERN uLong ZEXPORT adler32_combine64 OF((uLong, uLong, z_off_t));
+    ZEXTERN uLong ZEXPORT crc32_combine64 OF((uLong, uLong, z_off_t));
+#endif
+
         /* common defaults */
 
 #ifndef OS_CODE
@@ -197,7 +199,9 @@
 #  ifdef WIN32
      /* In Win32, vsnprintf is available as the "non-ANSI" _vsnprintf. */
 #    if !defined(vsnprintf) && !defined(NO_vsnprintf)
-#      define vsnprintf _vsnprintf
+#      if !defined(_MSC_VER) || ( defined(_MSC_VER) && _MSC_VER < 1500 )
+#         define vsnprintf _vsnprintf
+#      endif
 #    endif
 #  endif
 #  ifdef __SASC
@@ -232,16 +236,16 @@
 #    define zmemzero(dest, len) memset(dest, 0, len)
 #  endif
 #else
-   extern void zmemcpy  OF((Bytef* dest, const Bytef* source, uInt len));
-   extern int  zmemcmp  OF((const Bytef* s1, const Bytef* s2, uInt len));
-   extern void zmemzero OF((Bytef* dest, uInt len));
+   void ZLIB_INTERNAL zmemcpy OF((Bytef* dest, const Bytef* source, uInt len));
+   int ZLIB_INTERNAL zmemcmp OF((const Bytef* s1, const Bytef* s2, uInt len));
+   void ZLIB_INTERNAL zmemzero OF((Bytef* dest, uInt len));
 #endif
 
 /* Diagnostic functions */
 #ifdef DEBUG
 #  include <stdio.h>
-   extern int z_verbose;
-   extern void z_error    OF((char *m));
+   extern int ZLIB_INTERNAL z_verbose;
+   extern void ZLIB_INTERNAL z_error OF((char *m));
 #  define Assert(cond,msg) {if(!(cond)) z_error(msg);}
 #  define Trace(x) {if (z_verbose>=0) fprintf x ;}
 #  define Tracev(x) {if (z_verbose>0) fprintf x ;}
@@ -258,8 +262,9 @@
 #endif
 
 
-voidpf zcalloc OF((voidpf opaque, unsigned items, unsigned size));
-void   zcfree  OF((voidpf opaque, voidpf ptr));
+voidpf ZLIB_INTERNAL zcalloc OF((voidpf opaque, unsigned items,
+                        unsigned size));
+void ZLIB_INTERNAL zcfree  OF((voidpf opaque, voidpf ptr));
 
 #define ZALLOC(strm, items, size) \
            (*((strm)->zalloc))((strm)->opaque, (items), (size))

From python-checkins at python.org  Fri Jul 30 22:05:54 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Fri, 30 Jul 2010 22:05:54 +0200 (CEST)
Subject: [Python-checkins] r83296 - svn:log
Message-ID: <20100730200554.9821EEEA4A@mail.python.org>

Author: martin.v.loewis
Revision: 83296
Property Name: svn:log
Action: modified

Property diff:
--- old property value
+++ new property value
@@ -1 +1 @@
-Import files from zlib 1.0.5.
+Import files from zlib 1.2.5.

From python-checkins at python.org  Fri Jul 30 22:08:12 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 22:08:12 +0200 (CEST)
Subject: [Python-checkins] r83297 -
	python/branches/import_unicode/Modules/main.c
Message-ID: <20100730200812.C81F9EEA4C@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 22:08:12 2010
New Revision: 83297

Log:
Fix _Py_wchar2char(): unescape char in U+DC00..U+DCFF

instead of range U+D800..U+DFFF

Modified:
   python/branches/import_unicode/Modules/main.c

Modified: python/branches/import_unicode/Modules/main.c
==============================================================================
--- python/branches/import_unicode/Modules/main.c	(original)
+++ python/branches/import_unicode/Modules/main.c	Fri Jul 30 22:08:12 2010
@@ -744,7 +744,7 @@
     bytes = result;
     for (i=0; i < len; i++) {
         c = text[i];
-        if (c >= 0xd800 && c <= 0xdfff) {
+        if (c >= 0xdc00 && c <= 0xdcff) {
             /* Surrogate character */
             *bytes++ = c - 0xdc00;
             size--;

From python-checkins at python.org  Fri Jul 30 22:13:37 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 22:13:37 +0200 (CEST)
Subject: [Python-checkins] r83298 -
	python/branches/import_unicode/Objects/codeobject.c
Message-ID: <20100730201337.AC1E2EE98D@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 22:13:37 2010
New Revision: 83298

Log:
document and simplify _Py_code_object_list

Modified:
   python/branches/import_unicode/Objects/codeobject.c

Modified: python/branches/import_unicode/Objects/codeobject.c
==============================================================================
--- python/branches/import_unicode/Objects/codeobject.c	(original)
+++ python/branches/import_unicode/Objects/codeobject.c	Fri Jul 30 22:13:37 2010
@@ -5,6 +5,9 @@
 #define NAME_CHARS \
     "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ_abcdefghijklmnopqrstuvwxyz"
 
+/* List of all code objects used to reencode code filenames at startup when the
+ * file system encoding is changed. After startup, the list is no more used (it
+ * is set to NULL). */
 PyObject *_Py_code_object_list = NULL;
 
 /* all_name_chars(s): true iff all chars in s are valid NAME_CHARS */
@@ -46,10 +49,7 @@
 
 void _PyCode_ClearList(void)
 {
-    if (_Py_code_object_list == NULL)
-        return;
-    Py_DECREF(_Py_code_object_list);
-    _Py_code_object_list = NULL;
+    Py_CLEAR(_Py_code_object_list);
 }
 
 PyCodeObject *

From python-checkins at python.org  Fri Jul 30 22:20:43 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 22:20:43 +0200 (CEST)
Subject: [Python-checkins] r83299 - in python/branches/import_unicode:
	Include/code.h Objects/object.c
Message-ID: <20100730202043.802BCEE98D@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 22:20:43 2010
New Revision: 83299

Log:
move _Py_code_object_list declaration to code.h

Modified:
   python/branches/import_unicode/Include/code.h
   python/branches/import_unicode/Objects/object.c

Modified: python/branches/import_unicode/Include/code.h
==============================================================================
--- python/branches/import_unicode/Include/code.h	(original)
+++ python/branches/import_unicode/Include/code.h	Fri Jul 30 22:20:43 2010
@@ -72,7 +72,7 @@
 PyAPI_FUNC(PyCodeObject *) PyCode_New(
 	int, int, int, int, int, PyObject *, PyObject *,
 	PyObject *, PyObject *, PyObject *, PyObject *,
-	PyObject *, PyObject *, int, PyObject *); 
+	PyObject *, PyObject *, int, PyObject *);
         /* same as struct above */
 
 /* Creates a new empty code object with the specified source location. */
@@ -99,6 +99,8 @@
 PyAPI_FUNC(PyObject*) PyCode_Optimize(PyObject *code, PyObject* consts,
                                       PyObject *names, PyObject *lineno_obj);
 
+extern PyObject *_Py_code_object_list;
+
 #ifdef __cplusplus
 }
 #endif

Modified: python/branches/import_unicode/Objects/object.c
==============================================================================
--- python/branches/import_unicode/Objects/object.c	(original)
+++ python/branches/import_unicode/Objects/object.c	Fri Jul 30 22:20:43 2010
@@ -9,8 +9,6 @@
 extern "C" {
 #endif
 
-extern PyObject *_Py_code_object_list;
-
 #ifdef Py_REF_DEBUG
 Py_ssize_t _Py_RefTotal;
 

From python-checkins at python.org  Fri Jul 30 22:26:57 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 22:26:57 +0200 (CEST)
Subject: [Python-checkins] r83300 - in python/branches/import_unicode:
	Include/Python.h Modules/zipimport.c Python/import.c
Message-ID: <20100730202657.D343DEEA2B@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 22:26:57 2010
New Revision: 83300

Log:
declare _Py_fopen() in Python.h

mark _Py_stat() as static

Modified:
   python/branches/import_unicode/Include/Python.h
   python/branches/import_unicode/Modules/zipimport.c
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Include/Python.h
==============================================================================
--- python/branches/import_unicode/Include/Python.h	(original)
+++ python/branches/import_unicode/Include/Python.h	Fri Jul 30 22:26:57 2010
@@ -132,6 +132,9 @@
 PyAPI_FUNC(wchar_t *) _Py_char2wchar(char *);
 PyAPI_FUNC(char*) _Py_wchar2char(const wchar_t *text);
 
+/* _Py_fopen lives in import.c */
+PyAPI_FUNC(FILE*) _Py_fopen(PyObject *unicode, const char *mode);
+
 #ifdef __cplusplus
 }
 #endif

Modified: python/branches/import_unicode/Modules/zipimport.c
==============================================================================
--- python/branches/import_unicode/Modules/zipimport.c	(original)
+++ python/branches/import_unicode/Modules/zipimport.c	Fri Jul 30 22:26:57 2010
@@ -44,7 +44,6 @@
 static PyObject *zip_directory_cache = NULL;
 
 /* forward decls */
-extern FILE* _Py_fopen(PyObject *unicode, const char *mode);
 static PyObject *read_directory(PyObject *archive);
 static PyObject *get_data(PyObject *archive, PyObject *toc_entry);
 static PyObject *get_module_code(ZipImporter *self, char *fullname,

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul 30 22:26:57 2010
@@ -139,7 +139,7 @@
 /* Forward */
 FILE* _Py_fopen(PyObject *unicode, const char *mode);
 #ifdef HAVE_STAT
-int _Py_stat(PyObject *unicode, struct stat *statbuf);
+static int _Py_stat(PyObject *unicode, struct stat *statbuf);
 static int find_init_module(PyObject *);
 #endif
 
@@ -2161,7 +2161,7 @@
 
 
 #ifdef HAVE_STAT
-int
+static int
 _Py_stat(PyObject *unicode, struct stat *statbuf)
 {
 #ifdef MS_WINDOWS

From python-checkins at python.org  Fri Jul 30 22:30:05 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 22:30:05 +0200 (CEST)
Subject: [Python-checkins] r83301 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100730203005.2A3E4D6A4@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 22:30:05 2010
New Revision: 83301

Log:
make_source_pathname() raises a MemoryError on strdup() failure

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul 30 22:30:05 2010
@@ -1060,7 +1060,7 @@
     pathname = strdup(PyBytes_AsString(pathbytes));
     Py_DECREF(pathbytes);
     if (pathname == NULL)
-        return NULL;
+        return PyErr_NoMemory();
 
     cpathname = _make_source_pathname(pathname, buf);
     free(pathname);

From python-checkins at python.org  Fri Jul 30 22:33:59 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 22:33:59 +0200 (CEST)
Subject: [Python-checkins] r83302 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100730203359.0F905EE98D@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 22:33:58 2010
New Revision: 83302

Log:
parse_source_module() free the arena on error

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul 30 22:33:58 2010
@@ -1173,8 +1173,10 @@
         return NULL;
 
     pathbytes = PyUnicode_EncodeFSDefault(pathobj);
-    if (pathbytes == NULL)
+    if (pathbytes == NULL) {
+        PyArena_Free(arena);
         return NULL;
+    }
     pathname = PyBytes_AS_STRING(pathbytes);
 
     flags.cf_flags = 0;

From python-checkins at python.org  Fri Jul 30 22:35:57 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 22:35:57 +0200 (CEST)
Subject: [Python-checkins] r83303 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100730203557.2BF05EE98D@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 22:35:57 2010
New Revision: 83303

Log:
fix filename match in get_sourcefile()

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul 30 22:35:57 2010
@@ -1452,10 +1452,10 @@
     file = PyUnicode_AS_UNICODE(fileobj);
 
     /* match '*.py?' */
-    if (4 <= len
-        && file[len-3] == '.'
-        && (file[len-2] == 'p' || file[len-2] == 'P')
-        && (file[len-1] == 'y' || file[len-1] == 'Y')) {
+    if (5 <= len
+        && file[len-4] == '.'
+        && (file[len-3] == 'p' || file[len-3] == 'P')
+        && (file[len-2] == 'y' || file[len-2] == 'Y')) {
         Py_INCREF(fileobj);
         return fileobj;
     }

From python-checkins at python.org  Fri Jul 30 23:04:11 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 23:04:11 +0200 (CEST)
Subject: [Python-checkins] r83304 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100730210411.90321EEA4C@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 23:04:11 2010
New Revision: 83304

Log:
re-fix filename match in get_sourcefile()

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul 30 23:04:11 2010
@@ -1452,10 +1452,10 @@
     file = PyUnicode_AS_UNICODE(fileobj);
 
     /* match '*.py?' */
-    if (5 <= len
-        && file[len-4] == '.'
-        && (file[len-3] == 'p' || file[len-3] == 'P')
-        && (file[len-2] == 'y' || file[len-2] == 'Y')) {
+    if (len < 5
+        || file[len-4] != '.'
+        || (file[len-3] != 'p' && file[len-3] != 'P')
+        || (file[len-2] != 'y' && file[len-2] != 'Y')) {
         Py_INCREF(fileobj);
         return fileobj;
     }

From python-checkins at python.org  Fri Jul 30 23:08:30 2010
From: python-checkins at python.org (victor.stinner)
Date: Fri, 30 Jul 2010 23:08:30 +0200 (CEST)
Subject: [Python-checkins] r83305 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100730210830.D0941EE98D@mail.python.org>

Author: victor.stinner
Date: Fri Jul 30 23:08:30 2010
New Revision: 83305

Log:
Fix a refleak in import.c

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Fri Jul 30 23:08:30 2010
@@ -1912,6 +1912,7 @@
             unicode_without_suffix = saved_unicode;
             namelen = saved_namelen;
 #endif
+            Py_CLEAR(unicode);
         }
         Py_CLEAR(unicode_without_suffix);
 #if defined(PYOS_OS2)
@@ -1919,7 +1920,6 @@
 #endif
         if (fp != NULL)
             break;
-        Py_DECREF(unicode);
     }
     Py_XDECREF(unicode_without_suffix);
     if (fp == NULL) {

From python-checkins at python.org  Fri Jul 30 23:27:18 2010
From: python-checkins at python.org (matthias.klose)
Date: Fri, 30 Jul 2010 23:27:18 +0200 (CEST)
Subject: [Python-checkins] r83306 - in python/branches/release27-maint:
	Misc/NEWS Modules/_cursesmodule.c
Message-ID: <20100730212718.C6C47EE9CC@mail.python.org>

Author: matthias.klose
Date: Fri Jul 30 23:27:18 2010
New Revision: 83306

Log:
- Issue #7567: Don't call `setupterm' twice.


Modified:
   python/branches/release27-maint/Misc/NEWS
   python/branches/release27-maint/Modules/_cursesmodule.c

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Fri Jul 30 23:27:18 2010
@@ -78,6 +78,8 @@
   then that exception wasn't properly propagated on machines where
   char is unsigned.
 
+- Issue #7567: Don't call `setupterm' twice.
+
 Build
 -----
 

Modified: python/branches/release27-maint/Modules/_cursesmodule.c
==============================================================================
--- python/branches/release27-maint/Modules/_cursesmodule.c	(original)
+++ python/branches/release27-maint/Modules/_cursesmodule.c	Fri Jul 30 23:27:18 2010
@@ -2047,7 +2047,7 @@
         }
     }
 
-    if (setupterm(termstr,fd,&err) == ERR) {
+    if (!initialised_setupterm && setupterm(termstr,fd,&err) == ERR) {
         char* s = "setupterm: unknown error";
 
         if (err == 0) {

From python-checkins at python.org  Fri Jul 30 23:40:57 2010
From: python-checkins at python.org (matthias.klose)
Date: Fri, 30 Jul 2010 23:40:57 +0200 (CEST)
Subject: [Python-checkins] r83307 - in python/branches/py3k: Misc/NEWS
	Modules/_cursesmodule.c
Message-ID: <20100730214057.E3ABDEE9E7@mail.python.org>

Author: matthias.klose
Date: Fri Jul 30 23:40:57 2010
New Revision: 83307

Log:
- Issue #7567: PyCurses_setupterm: Don't call `setupterm' twice.


Modified:
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/Modules/_cursesmodule.c

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Fri Jul 30 23:40:57 2010
@@ -1673,6 +1673,8 @@
 - Issue #7347: winreg: Add CreateKeyEx and DeleteKeyEx, as well as fix a
   bug in the return value of QueryReflectionKey.
 
+- Issue #7567: PyCurses_setupterm: Don't call `setupterm' twice.
+
 Build
 -----
 

Modified: python/branches/py3k/Modules/_cursesmodule.c
==============================================================================
--- python/branches/py3k/Modules/_cursesmodule.c	(original)
+++ python/branches/py3k/Modules/_cursesmodule.c	Fri Jul 30 23:40:57 2010
@@ -2126,7 +2126,7 @@
         }
     }
 
-    if (setupterm(termstr,fd,&err) == ERR) {
+    if (!initialised_setupterm && setupterm(termstr,fd,&err) == ERR) {
         char* s = "setupterm: unknown error";
 
         if (err == 0) {

From python-checkins at python.org  Sat Jul 31 00:20:16 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 00:20:16 +0200 (CEST)
Subject: [Python-checkins] r83308 - in python/branches/py3k: Lib/pdb.py
	Misc/NEWS
Message-ID: <20100730222016.64520EE9E7@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 00:20:16 2010
New Revision: 83308

Log:
Part of #7245: when KeyboardInterrupt is raised while defining commands, restore the old commands instead of producing a traceback.

Modified:
   python/branches/py3k/Lib/pdb.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/pdb.py
==============================================================================
--- python/branches/py3k/Lib/pdb.py	(original)
+++ python/branches/py3k/Lib/pdb.py	Sat Jul 31 00:20:16 2010
@@ -371,7 +371,7 @@
         else:
             return self.handle_command_def(line)
 
-    def handle_command_def(self,line):
+    def handle_command_def(self, line):
         """Handles one command line during command list definition."""
         cmd, arg, line = self.parseline(line)
         if not cmd:
@@ -457,14 +457,33 @@
                 self.error("Usage: commands [bnum]\n        ...\n        end")
                 return
         self.commands_bnum = bnum
+        # Save old definitions for the case of a keyboard interrupt.
+        if bnum in self.commands:
+            old_command_defs = (self.commands[bnum],
+                                self.commands_doprompt[bnum],
+                                self.commands_silent[bnum])
+        else:
+            old_command_defs = None
         self.commands[bnum] = []
         self.commands_doprompt[bnum] = True
         self.commands_silent[bnum] = False
+
         prompt_back = self.prompt
         self.prompt = '(com) '
         self.commands_defining = True
         try:
             self.cmdloop()
+        except KeyboardInterrupt:
+            # Restore old definitions.
+            if old_command_defs:
+                self.commands[bnum] = old_command_defs[0]
+                self.commands_doprompt[bnum] = old_command_defs[1]
+                self.commands_silent[bnum] = old_command_defs[2]
+            else:
+                del self.commands[bnum]
+                del self.commands_doprompt[bnum]
+                del self.commands_silent[bnum]
+            self.error('command definition aborted, old commands restored')
         finally:
             self.commands_defining = False
             self.prompt = prompt_back

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sat Jul 31 00:20:16 2010
@@ -475,6 +475,9 @@
 Library
 -------
 
+- In pdb, when Ctrl-C is entered while defining commands for a
+  breakpoint, the old commands are restored.
+
 - For traceback debugging, the pdb listing now also shows the locations
   where the exception was originally (re)raised, if it differs from the
   last line executed (e.g. in case of finally clauses).

From python-checkins at python.org  Sat Jul 31 00:26:55 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 00:26:55 +0200 (CEST)
Subject: [Python-checkins] r83309 - peps/trunk/pep-0392.txt
Message-ID: <20100730222655.1F2E9EEA20@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 00:26:54 2010
New Revision: 83309

Log:
Update date.

Modified:
   peps/trunk/pep-0392.txt

Modified: peps/trunk/pep-0392.txt
==============================================================================
--- peps/trunk/pep-0392.txt	(original)
+++ peps/trunk/pep-0392.txt	Sat Jul 31 00:26:54 2010
@@ -34,7 +34,7 @@
 
 The current schedule is:
 
-- 3.2 alpha 1: August 7, 2010
+- 3.2 alpha 1: August 1, 2010
 - 3.2 alpha 2: September 4, 2010
 - 3.2 alpha 3: October 2, 2010
 - 3.2 beta 1: October 30, 2010

From python-checkins at python.org  Sat Jul 31 01:43:46 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Sat, 31 Jul 2010 01:43:46 +0200 (CEST)
Subject: [Python-checkins] r83310 - peps/trunk/pep-3151.txt
Message-ID: <20100730234346.0E49FEEA1F@mail.python.org>

Author: antoine.pitrou
Date: Sat Jul 31 01:43:45 2010
New Revision: 83310

Log:
Mention VMSError



Modified:
   peps/trunk/pep-3151.txt

Modified: peps/trunk/pep-3151.txt
==============================================================================
--- peps/trunk/pep-3151.txt	(original)
+++ peps/trunk/pep-3151.txt	Sat Jul 31 01:43:45 2010
@@ -48,6 +48,7 @@
             +-- io.UnsupportedOperation (also inherits from ValueError)
             +-- socket.error
         +-- OSError
+            +-- VMSError
             +-- WindowsError
         +-- mmap.error
     +-- select.error
@@ -206,7 +207,7 @@
 
 * alias both socket.error and select.error to IOError
 * alias mmap.error to OSError
-* alias WindowsError to OSError
+* alias both WindowsError and VMSError to OSError
 * alias OSError to IOError
 * coalesce EnvironmentError into IOError
 
@@ -543,6 +544,14 @@
 Appendix B: Survey of raised OS and IO errors
 =============================================
 
+About VMSError
+--------------
+
+VMSError is completely unused by the interpreter core and the standard
+library.  It was added as part of the OpenVMS patches submitted in 2002
+by Jean-Fran?ois Pi?ronne [4]_; the motivation for including VMSError was that
+it could be raised by third-party packages.
+
 Interpreter core
 ----------------
 
@@ -797,6 +806,8 @@
    ``EnvironmentError``: `around 3000 results
    <http://www.google.com/codesearch?q=lang%3Apython%20EnvironmentError>`_
 
+.. [4] http://bugs.python.org/issue614055
+
 Copyright
 =========
 

From python-checkins at python.org  Sat Jul 31 01:49:00 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Sat, 31 Jul 2010 01:49:00 +0200 (CEST)
Subject: [Python-checkins] r83311 -
	tracker/instances/python-dev/lib/openid2rp.py
Message-ID: <20100730234900.F2E06EEA1F@mail.python.org>

Author: martin.v.loewis
Date: Sat Jul 31 01:49:00 2010
New Revision: 83311

Log:
Incorporate upstream fixes.


Modified:
   tracker/instances/python-dev/lib/openid2rp.py

Modified: tracker/instances/python-dev/lib/openid2rp.py
==============================================================================
--- tracker/instances/python-dev/lib/openid2rp.py	(original)
+++ tracker/instances/python-dev/lib/openid2rp.py	Sat Jul 31 01:49:00 2010
@@ -146,7 +146,10 @@
 
     def do_link(self, attrs):
         attrs = dict(attrs)
-        self.links[attrs['rel']] = attrs['href']
+        try:
+            self.links[attrs['rel']] = attrs['href']
+        except KeyError:
+            pass
 
     def do_meta(self, attrs):
         attrs = dict(attrs)
@@ -457,7 +460,7 @@
 
 def _prepare_response(response):
     if isinstance(response, str):
-        return cgi.parse_qs(response)
+        return cgi.parse_qs(response, keep_blank_values=True)
     # backwards compatibility: allow caller to pass parse_qs result
     # already
     pass

From python-checkins at python.org  Sat Jul 31 09:03:43 2010
From: python-checkins at python.org (raymond.hettinger)
Date: Sat, 31 Jul 2010 09:03:43 +0200 (CEST)
Subject: [Python-checkins] r83312 -
	python/branches/release31-maint/Lib/_abcoll.py
Message-ID: <20100731070343.261E7EEA7B@mail.python.org>

Author: raymond.hettinger
Date: Sat Jul 31 09:03:42 2010
New Revision: 83312

Log:
Only expose the abstract base classes.
The concrete types are for internal use (registration).
Keep from accidentally resurrecting the types module in collections.



Modified:
   python/branches/release31-maint/Lib/_abcoll.py

Modified: python/branches/release31-maint/Lib/_abcoll.py
==============================================================================
--- python/branches/release31-maint/Lib/_abcoll.py	(original)
+++ python/branches/release31-maint/Lib/_abcoll.py	Sat Jul 31 09:03:42 2010
@@ -18,11 +18,6 @@
            "MappingView", "KeysView", "ItemsView", "ValuesView",
            "Sequence", "MutableSequence",
            "ByteString",
-           "bytearray_iterator", "bytes_iterator", "dict_itemiterator",
-           "dict_items", "dict_keyiterator", "dict_keys", "dict_proxy",
-           "dict_valueiterator", "dict_values", "list_iterator",
-           "list_reverseiterator", "range_iterator", "set_iterator",
-           "str_iterator", "tuple_iterator", "zip_iterator",
            ]
 
 

From python-checkins at python.org  Sat Jul 31 09:12:50 2010
From: python-checkins at python.org (raymond.hettinger)
Date: Sat, 31 Jul 2010 09:12:50 +0200 (CEST)
Subject: [Python-checkins] r83313 - python/branches/py3k/Lib/_abcoll.py
Message-ID: <20100731071250.C5559EEA45@mail.python.org>

Author: raymond.hettinger
Date: Sat Jul 31 09:12:50 2010
New Revision: 83313

Log:
Only expose the abstract base classes.
The concrete types are for internal use (registration).
We are not trying to resurrect the types module
in collections.



Modified:
   python/branches/py3k/Lib/_abcoll.py

Modified: python/branches/py3k/Lib/_abcoll.py
==============================================================================
--- python/branches/py3k/Lib/_abcoll.py	(original)
+++ python/branches/py3k/Lib/_abcoll.py	Sat Jul 31 09:12:50 2010
@@ -18,11 +18,6 @@
            "MappingView", "KeysView", "ItemsView", "ValuesView",
            "Sequence", "MutableSequence",
            "ByteString",
-           "bytearray_iterator", "bytes_iterator", "dict_itemiterator",
-           "dict_items", "dict_keyiterator", "dict_keys", "dict_proxy",
-           "dict_valueiterator", "dict_values", "list_iterator",
-           "list_reverseiterator", "range_iterator", "set_iterator",
-           "str_iterator", "tuple_iterator", "zip_iterator",
            ]
 
 

From python-checkins at python.org  Sat Jul 31 10:00:13 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 10:00:13 +0200 (CEST)
Subject: [Python-checkins] r83314 -
	python/branches/release26-maint/Doc/library/difflib.rst
Message-ID: <20100731080013.79F8EEEAB2@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 10:00:13 2010
New Revision: 83314

Log:
#2986: document SequenceMatcher heuristic.

Modified:
   python/branches/release26-maint/Doc/library/difflib.rst

Modified: python/branches/release26-maint/Doc/library/difflib.rst
==============================================================================
--- python/branches/release26-maint/Doc/library/difflib.rst	(original)
+++ python/branches/release26-maint/Doc/library/difflib.rst	Sat Jul 31 10:00:13 2010
@@ -37,6 +37,11 @@
    complicated way on how many elements the sequences have in common; best case
    time is linear.
 
+   **Heuristic:** To speed-up matching, items that appear more than 1% of the
+   time in sequences of at least 200 items are treated as junk.  This has the
+   unfortunate side-effect of giving bad results for sequences constructed from
+   a small set of items.  An option to turn off the heuristic will be added to a
+   future version.
 
 .. class:: Differ
 

From python-checkins at python.org  Sat Jul 31 10:14:16 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 10:14:16 +0200 (CEST)
Subject: [Python-checkins] r83315 - python/branches/py3k/Lib/test/test_pdb.py
Message-ID: <20100731081416.D6581EEA3C@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 10:14:16 2010
New Revision: 83315

Log:
Fix pdb test failures on the buildbots.

Modified:
   python/branches/py3k/Lib/test/test_pdb.py

Modified: python/branches/py3k/Lib/test/test_pdb.py
==============================================================================
--- python/branches/py3k/Lib/test/test_pdb.py	(original)
+++ python/branches/py3k/Lib/test/test_pdb.py	Sat Jul 31 10:14:16 2010
@@ -345,7 +345,7 @@
     -> test_pdb.do_nothing()
     (Pdb) step
     --Call--
-    > /home/gbr/devel/python/Lib/test/test_pdb.py(260)do_nothing()
+    > ...test_pdb.py(...)do_nothing()
     -> def do_nothing():
     (Pdb) longlist
     ...  ->     def do_nothing():
@@ -558,9 +558,9 @@
 def test_pdb_run_with_code_object():
     """Testing run and runeval with code object as a first argument.
 
-    >>> with PdbTestInput(['step','x', 'continue']):
+    >>> with PdbTestInput(['step','x', 'continue']):  # doctest: +ELLIPSIS
     ...     pdb_invoke('run', compile('x=1', '<string>', 'exec'))
-    > <string>(1)<module>()
+    > <string>(1)<module>()...
     (Pdb) step
     --Return--
     > <string>(1)<module>()->None

From python-checkins at python.org  Sat Jul 31 10:20:03 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 10:20:03 +0200 (CEST)
Subject: [Python-checkins] r83316 -
	python/branches/py3k/Lib/test/test_urllib.py
Message-ID: <20100731082003.0E7FEEEA72@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 10:20:02 2010
New Revision: 83316

Log:
Make urllib tests pass for now.  Will figure out what the correct semantics should be after release.

Modified:
   python/branches/py3k/Lib/test/test_urllib.py

Modified: python/branches/py3k/Lib/test/test_urllib.py
==============================================================================
--- python/branches/py3k/Lib/test/test_urllib.py	(original)
+++ python/branches/py3k/Lib/test/test_urllib.py	Sat Jul 31 10:20:02 2010
@@ -555,8 +555,8 @@
         self.assertEqual(result.count('%'), 1,
                          "using unquote(): not all characters escaped: "
                          "%s" % result)
-        self.assertRaises(TypeError, urllib.parse.unquote, None)
-        self.assertRaises(TypeError, urllib.parse.unquote, ())
+        self.assertRaises((TypeError, AttributeError), urllib.parse.unquote, None)
+        self.assertRaises((TypeError, AttributeError), urllib.parse.unquote, ())
 
     def test_unquoting_badpercent(self):
         # Test unquoting on bad percent-escapes
@@ -591,8 +591,8 @@
         result = urllib.parse.unquote_to_bytes(given)
         self.assertEqual(expect, result, "using unquote_to_bytes(): %r != %r"
                          % (expect, result))
-        self.assertRaises(TypeError, urllib.parse.unquote_to_bytes, None)
-        self.assertRaises(TypeError, urllib.parse.unquote_to_bytes, ())
+        self.assertRaises((TypeError, AttributeError), urllib.parse.unquote_to_bytes, None)
+        self.assertRaises((TypeError, AttributeError), urllib.parse.unquote_to_bytes, ())
 
     def test_unquoting_mixed_case(self):
         # Test unquoting on mixed-case hex digits in the percent-escapes

From python-checkins at python.org  Sat Jul 31 10:27:47 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 10:27:47 +0200 (CEST)
Subject: [Python-checkins] r83317 - in python/branches/py3k:
	Doc/tools/sphinxext/pyspecific.py Lib/pydoc_data/topics.py
Message-ID: <20100731082747.4D77BDA36@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 10:27:46 2010
New Revision: 83317

Log:
Update pydoc topics and adapt Topics builder to Sphinx 1.0.

Modified:
   python/branches/py3k/Doc/tools/sphinxext/pyspecific.py
   python/branches/py3k/Lib/pydoc_data/topics.py

Modified: python/branches/py3k/Doc/tools/sphinxext/pyspecific.py
==============================================================================
--- python/branches/py3k/Doc/tools/sphinxext/pyspecific.py	(original)
+++ python/branches/py3k/Doc/tools/sphinxext/pyspecific.py	Sat Jul 31 10:27:46 2010
@@ -145,10 +145,10 @@
         for label in self.status_iterator(pydoc_topic_labels,
                                           'building topics... ',
                                           length=len(pydoc_topic_labels)):
-            if label not in self.env.labels:
+            if label not in self.env.domaindata['std']['labels']:
                 self.warn('label %r not in documentation' % label)
                 continue
-            docname, labelid, sectname = self.env.labels[label]
+            docname, labelid, sectname = self.env.domaindata['std']['labels'][label]
             doctree = self.env.get_and_resolve_doctree(docname, self)
             document = new_document('<section node>')
             document.append(doctree.ids[labelid])

Modified: python/branches/py3k/Lib/pydoc_data/topics.py
==============================================================================
--- python/branches/py3k/Lib/pydoc_data/topics.py	(original)
+++ python/branches/py3k/Lib/pydoc_data/topics.py	Sat Jul 31 10:27:46 2010
@@ -1,78 +1,77 @@
-# Autogenerated by Sphinx on Fri Jun 26 08:03:32 2009
+# Autogenerated by Sphinx on Sat Jul 31 10:22:58 2010
 topics = {'assert': '\nThe ``assert`` statement\n************************\n\nAssert statements are a convenient way to insert debugging assertions\ninto a program:\n\n   assert_stmt ::= "assert" expression ["," expression]\n\nThe simple form, ``assert expression``, is equivalent to\n\n   if __debug__:\n      if not expression: raise AssertionError\n\nThe extended form, ``assert expression1, expression2``, is equivalent\nto\n\n   if __debug__:\n      if not expression1: raise AssertionError(expression2)\n\nThese equivalences assume that ``__debug__`` and ``AssertionError``\nrefer to the built-in variables with those names.  In the current\nimplementation, the built-in variable ``__debug__`` is ``True`` under\nnormal circumstances, ``False`` when optimization is requested\n(command line option -O).  The current code generator emits no code\nfor an assert statement when optimization is requested at compile\ntime.  Note that it is unnecessary to include the source code for the\nexpression that failed in the error message; it will be displayed as\npart of the stack trace.\n\nAssignments to ``__debug__`` are illegal.  The value for the built-in\nvariable is determined when the interpreter starts.\n',
- 'assignment': '\nAssignment statements\n*********************\n\nAssignment statements are used to (re)bind names to values and to\nmodify attributes or items of mutable objects:\n\n   assignment_stmt ::= (target_list "=")+ (expression_list | yield_expression)\n   target_list     ::= target ("," target)* [","]\n   target          ::= identifier\n              | "(" target_list ")"\n              | "[" target_list "]"\n              | attributeref\n              | subscription\n              | slicing\n              | "*" target\n\n(See section *Primaries* for the syntax definitions for the last three\nsymbols.)\n\nAn assignment statement evaluates the expression list (remember that\nthis can be a single expression or a comma-separated list, the latter\nyielding a tuple) and assigns the single resulting object to each of\nthe target lists, from left to right.\n\nAssignment is defined recursively depending on the form of the target\n(list). When a target is part of a mutable object (an attribute\nreference, subscription or slicing), the mutable object must\nultimately perform the assignment and decide about its validity, and\nmay raise an exception if the assignment is unacceptable.  The rules\nobserved by various types and the exceptions raised are given with the\ndefinition of the object types (see section *The standard type\nhierarchy*).\n\nAssignment of an object to a target list, optionally enclosed in\nparentheses or square brackets, is recursively defined as follows.\n\n* If the target list is a single target: The object is assigned to\n  that target.\n\n* If the target list is a comma-separated list of targets: The object\n  must be an iterable with the same number of items as there are\n  targets in the target list, and the items are assigned, from left to\n  right, to the corresponding targets. (This rule is relaxed as of\n  Python 1.5; in earlier versions, the object had to be a tuple.\n  Since strings are sequences, an assignment like ``a, b = "xy"`` is\n  now legal as long as the string has the right length.)\n\n  * If the target list contains one target prefixed with an asterisk,\n    called a "starred" target: The object must be a sequence with at\n    least as many items as there are targets in the target list, minus\n    one.  The first items of the sequence are assigned, from left to\n    right, to the targets before the starred target.  The final items\n    of the sequence are assigned to the targets after the starred\n    target.  A list of the remaining items in the sequence is then\n    assigned to the starred target (the list can be empty).\n\n  * Else: The object must be a sequence with the same number of items\n    as there are targets in the target list, and the items are\n    assigned, from left to right, to the corresponding targets.\n\nAssignment of an object to a single target is recursively defined as\nfollows.\n\n* If the target is an identifier (name):\n\n  * If the name does not occur in a ``global`` or ``nonlocal``\n    statement in the current code block: the name is bound to the\n    object in the current local namespace.\n\n  * Otherwise: the name is bound to the object in the global namespace\n    or the outer namespace determined by ``nonlocal``, respectively.\n\n  The name is rebound if it was already bound.  This may cause the\n  reference count for the object previously bound to the name to reach\n  zero, causing the object to be deallocated and its destructor (if it\n  has one) to be called.\n\n  The name is rebound if it was already bound.  This may cause the\n  reference count for the object previously bound to the name to reach\n  zero, causing the object to be deallocated and its destructor (if it\n  has one) to be called.\n\n* If the target is a target list enclosed in parentheses or in square\n  brackets: The object must be an iterable with the same number of\n  items as there are targets in the target list, and its items are\n  assigned, from left to right, to the corresponding targets.\n\n* If the target is an attribute reference: The primary expression in\n  the reference is evaluated.  It should yield an object with\n  assignable attributes; if this is not the case, ``TypeError`` is\n  raised.  That object is then asked to assign the assigned object to\n  the given attribute; if it cannot perform the assignment, it raises\n  an exception (usually but not necessarily ``AttributeError``).\n\n* If the target is a subscription: The primary expression in the\n  reference is evaluated.  It should yield either a mutable sequence\n  object (such as a list) or a mapping object (such as a dictionary).\n  Next, the subscript expression is evaluated.\n\n  If the primary is a mutable sequence object (such as a list), the\n  subscript must yield an integer.  If it is negative, the sequence\'s\n  length is added to it.  The resulting value must be a nonnegative\n  integer less than the sequence\'s length, and the sequence is asked\n  to assign the assigned object to its item with that index.  If the\n  index is out of range, ``IndexError`` is raised (assignment to a\n  subscripted sequence cannot add new items to a list).\n\n  If the primary is a mapping object (such as a dictionary), the\n  subscript must have a type compatible with the mapping\'s key type,\n  and the mapping is then asked to create a key/datum pair which maps\n  the subscript to the assigned object.  This can either replace an\n  existing key/value pair with the same key value, or insert a new\n  key/value pair (if no key with the same value existed).\n\n  For user-defined objects, the ``__setitem__()`` method is called\n  with appropriate arguments.\n\n* If the target is a slicing: The primary expression in the reference\n  is evaluated.  It should yield a mutable sequence object (such as a\n  list).  The assigned object should be a sequence object of the same\n  type.  Next, the lower and upper bound expressions are evaluated,\n  insofar they are present; defaults are zero and the sequence\'s\n  length.  The bounds should evaluate to integers. If either bound is\n  negative, the sequence\'s length is added to it.  The resulting\n  bounds are clipped to lie between zero and the sequence\'s length,\n  inclusive.  Finally, the sequence object is asked to replace the\n  slice with the items of the assigned sequence.  The length of the\n  slice may be different from the length of the assigned sequence,\n  thus changing the length of the target sequence, if the object\n  allows it.\n\n(In the current implementation, the syntax for targets is taken to be\nthe same as for expressions, and invalid syntax is rejected during the\ncode generation phase, causing less detailed error messages.)\n\nWARNING: Although the definition of assignment implies that overlaps\nbetween the left-hand side and the right-hand side are \'safe\' (for\nexample ``a, b = b, a`` swaps two variables), overlaps *within* the\ncollection of assigned-to variables are not safe!  For instance, the\nfollowing program prints ``[0, 2]``:\n\n   x = [0, 1]\n   i = 0\n   i, x[i] = 1, 2\n   print(x)\n\nSee also:\n\n   **PEP 3132** - Extended Iterable Unpacking\n      The specification for the ``*target`` feature.\n\n\nAugmented assignment statements\n===============================\n\nAugmented assignment is the combination, in a single statement, of a\nbinary operation and an assignment statement:\n\n   augmented_assignment_stmt ::= augtarget augop (expression_list | yield_expression)\n   augtarget                 ::= identifier | attributeref | subscription | slicing\n   augop                     ::= "+=" | "-=" | "*=" | "/=" | "//=" | "%=" | "**="\n             | ">>=" | "<<=" | "&=" | "^=" | "|="\n\n(See section *Primaries* for the syntax definitions for the last three\nsymbols.)\n\nAn augmented assignment evaluates the target (which, unlike normal\nassignment statements, cannot be an unpacking) and the expression\nlist, performs the binary operation specific to the type of assignment\non the two operands, and assigns the result to the original target.\nThe target is only evaluated once.\n\nAn augmented assignment expression like ``x += 1`` can be rewritten as\n``x = x + 1`` to achieve a similar, but not exactly equal effect. In\nthe augmented version, ``x`` is only evaluated once. Also, when\npossible, the actual operation is performed *in-place*, meaning that\nrather than creating a new object and assigning that to the target,\nthe old object is modified instead.\n\nWith the exception of assigning to tuples and multiple targets in a\nsingle statement, the assignment done by augmented assignment\nstatements is handled the same way as normal assignments. Similarly,\nwith the exception of the possible *in-place* behavior, the binary\noperation performed by augmented assignment is the same as the normal\nbinary operations.\n\nFor targets which are attribute references, the initial value is\nretrieved with a ``getattr()`` and the result is assigned with a\n``setattr()``.  Notice that the two methods do not necessarily refer\nto the same variable.  When ``getattr()`` refers to a class variable,\n``setattr()`` still writes to an instance variable. For example:\n\n   class A:\n       x = 3    # class variable\n   a = A()\n   a.x += 1     # writes a.x as 4 leaving A.x as 3\n',
+ 'assignment': '\nAssignment statements\n*********************\n\nAssignment statements are used to (re)bind names to values and to\nmodify attributes or items of mutable objects:\n\n   assignment_stmt ::= (target_list "=")+ (expression_list | yield_expression)\n   target_list     ::= target ("," target)* [","]\n   target          ::= identifier\n              | "(" target_list ")"\n              | "[" target_list "]"\n              | attributeref\n              | subscription\n              | slicing\n              | "*" target\n\n(See section *Primaries* for the syntax definitions for the last three\nsymbols.)\n\nAn assignment statement evaluates the expression list (remember that\nthis can be a single expression or a comma-separated list, the latter\nyielding a tuple) and assigns the single resulting object to each of\nthe target lists, from left to right.\n\nAssignment is defined recursively depending on the form of the target\n(list). When a target is part of a mutable object (an attribute\nreference, subscription or slicing), the mutable object must\nultimately perform the assignment and decide about its validity, and\nmay raise an exception if the assignment is unacceptable.  The rules\nobserved by various types and the exceptions raised are given with the\ndefinition of the object types (see section *The standard type\nhierarchy*).\n\nAssignment of an object to a target list, optionally enclosed in\nparentheses or square brackets, is recursively defined as follows.\n\n* If the target list is a single target: The object is assigned to\n  that target.\n\n* If the target list is a comma-separated list of targets: The object\n  must be an iterable with the same number of items as there are\n  targets in the target list, and the items are assigned, from left to\n  right, to the corresponding targets. (This rule is relaxed as of\n  Python 1.5; in earlier versions, the object had to be a tuple.\n  Since strings are sequences, an assignment like ``a, b = "xy"`` is\n  now legal as long as the string has the right length.)\n\n  * If the target list contains one target prefixed with an asterisk,\n    called a "starred" target: The object must be a sequence with at\n    least as many items as there are targets in the target list, minus\n    one.  The first items of the sequence are assigned, from left to\n    right, to the targets before the starred target.  The final items\n    of the sequence are assigned to the targets after the starred\n    target.  A list of the remaining items in the sequence is then\n    assigned to the starred target (the list can be empty).\n\n  * Else: The object must be a sequence with the same number of items\n    as there are targets in the target list, and the items are\n    assigned, from left to right, to the corresponding targets.\n\nAssignment of an object to a single target is recursively defined as\nfollows.\n\n* If the target is an identifier (name):\n\n  * If the name does not occur in a ``global`` or ``nonlocal``\n    statement in the current code block: the name is bound to the\n    object in the current local namespace.\n\n  * Otherwise: the name is bound to the object in the global namespace\n    or the outer namespace determined by ``nonlocal``, respectively.\n\n  The name is rebound if it was already bound.  This may cause the\n  reference count for the object previously bound to the name to reach\n  zero, causing the object to be deallocated and its destructor (if it\n  has one) to be called.\n\n* If the target is a target list enclosed in parentheses or in square\n  brackets: The object must be an iterable with the same number of\n  items as there are targets in the target list, and its items are\n  assigned, from left to right, to the corresponding targets.\n\n* If the target is an attribute reference: The primary expression in\n  the reference is evaluated.  It should yield an object with\n  assignable attributes; if this is not the case, ``TypeError`` is\n  raised.  That object is then asked to assign the assigned object to\n  the given attribute; if it cannot perform the assignment, it raises\n  an exception (usually but not necessarily ``AttributeError``).\n\n  Note: If the object is a class instance and the attribute reference\n  occurs on both sides of the assignment operator, the RHS expression,\n  ``a.x`` can access either an instance attribute or (if no instance\n  attribute exists) a class attribute.  The LHS target ``a.x`` is\n  always set as an instance attribute, creating it if necessary.\n  Thus, the two occurrences of ``a.x`` do not necessarily refer to the\n  same attribute: if the RHS expression refers to a class attribute,\n  the LHS creates a new instance attribute as the target of the\n  assignment:\n\n     class Cls:\n         x = 3             # class variable\n     inst = Cls()\n     inst.x = inst.x + 1   # writes inst.x as 4 leaving Cls.x as 3\n\n  This description does not necessarily apply to descriptor\n  attributes, such as properties created with ``property()``.\n\n* If the target is a subscription: The primary expression in the\n  reference is evaluated.  It should yield either a mutable sequence\n  object (such as a list) or a mapping object (such as a dictionary).\n  Next, the subscript expression is evaluated.\n\n  If the primary is a mutable sequence object (such as a list), the\n  subscript must yield an integer.  If it is negative, the sequence\'s\n  length is added to it.  The resulting value must be a nonnegative\n  integer less than the sequence\'s length, and the sequence is asked\n  to assign the assigned object to its item with that index.  If the\n  index is out of range, ``IndexError`` is raised (assignment to a\n  subscripted sequence cannot add new items to a list).\n\n  If the primary is a mapping object (such as a dictionary), the\n  subscript must have a type compatible with the mapping\'s key type,\n  and the mapping is then asked to create a key/datum pair which maps\n  the subscript to the assigned object.  This can either replace an\n  existing key/value pair with the same key value, or insert a new\n  key/value pair (if no key with the same value existed).\n\n  For user-defined objects, the ``__setitem__()`` method is called\n  with appropriate arguments.\n\n* If the target is a slicing: The primary expression in the reference\n  is evaluated.  It should yield a mutable sequence object (such as a\n  list).  The assigned object should be a sequence object of the same\n  type.  Next, the lower and upper bound expressions are evaluated,\n  insofar they are present; defaults are zero and the sequence\'s\n  length.  The bounds should evaluate to integers. If either bound is\n  negative, the sequence\'s length is added to it.  The resulting\n  bounds are clipped to lie between zero and the sequence\'s length,\n  inclusive.  Finally, the sequence object is asked to replace the\n  slice with the items of the assigned sequence.  The length of the\n  slice may be different from the length of the assigned sequence,\n  thus changing the length of the target sequence, if the object\n  allows it.\n\n**CPython implementation detail:** In the current implementation, the\nsyntax for targets is taken to be the same as for expressions, and\ninvalid syntax is rejected during the code generation phase, causing\nless detailed error messages.\n\nWARNING: Although the definition of assignment implies that overlaps\nbetween the left-hand side and the right-hand side are \'safe\' (for\nexample ``a, b = b, a`` swaps two variables), overlaps *within* the\ncollection of assigned-to variables are not safe!  For instance, the\nfollowing program prints ``[0, 2]``:\n\n   x = [0, 1]\n   i = 0\n   i, x[i] = 1, 2\n   print(x)\n\nSee also:\n\n   **PEP 3132** - Extended Iterable Unpacking\n      The specification for the ``*target`` feature.\n\n\nAugmented assignment statements\n===============================\n\nAugmented assignment is the combination, in a single statement, of a\nbinary operation and an assignment statement:\n\n   augmented_assignment_stmt ::= augtarget augop (expression_list | yield_expression)\n   augtarget                 ::= identifier | attributeref | subscription | slicing\n   augop                     ::= "+=" | "-=" | "*=" | "/=" | "//=" | "%=" | "**="\n             | ">>=" | "<<=" | "&=" | "^=" | "|="\n\n(See section *Primaries* for the syntax definitions for the last three\nsymbols.)\n\nAn augmented assignment evaluates the target (which, unlike normal\nassignment statements, cannot be an unpacking) and the expression\nlist, performs the binary operation specific to the type of assignment\non the two operands, and assigns the result to the original target.\nThe target is only evaluated once.\n\nAn augmented assignment expression like ``x += 1`` can be rewritten as\n``x = x + 1`` to achieve a similar, but not exactly equal effect. In\nthe augmented version, ``x`` is only evaluated once. Also, when\npossible, the actual operation is performed *in-place*, meaning that\nrather than creating a new object and assigning that to the target,\nthe old object is modified instead.\n\nWith the exception of assigning to tuples and multiple targets in a\nsingle statement, the assignment done by augmented assignment\nstatements is handled the same way as normal assignments. Similarly,\nwith the exception of the possible *in-place* behavior, the binary\noperation performed by augmented assignment is the same as the normal\nbinary operations.\n\nFor targets which are attribute references, the same *caveat about\nclass and instance attributes* applies as for regular assignments.\n',
  'atom-identifiers': '\nIdentifiers (Names)\n*******************\n\nAn identifier occurring as an atom is a name.  See section\n*Identifiers and keywords* for lexical definition and section *Naming\nand binding* for documentation of naming and binding.\n\nWhen the name is bound to an object, evaluation of the atom yields\nthat object. When a name is not bound, an attempt to evaluate it\nraises a ``NameError`` exception.\n\n**Private name mangling:** When an identifier that textually occurs in\na class definition begins with two or more underscore characters and\ndoes not end in two or more underscores, it is considered a *private\nname* of that class. Private names are transformed to a longer form\nbefore code is generated for them.  The transformation inserts the\nclass name in front of the name, with leading underscores removed, and\na single underscore inserted in front of the class name.  For example,\nthe identifier ``__spam`` occurring in a class named ``Ham`` will be\ntransformed to ``_Ham__spam``.  This transformation is independent of\nthe syntactical context in which the identifier is used.  If the\ntransformed name is extremely long (longer than 255 characters),\nimplementation defined truncation may happen.  If the class name\nconsists only of underscores, no transformation is done.\n',
  'atom-literals': "\nLiterals\n********\n\nPython supports string and bytes literals and various numeric\nliterals:\n\n   literal ::= stringliteral | bytesliteral\n               | integer | floatnumber | imagnumber\n\nEvaluation of a literal yields an object of the given type (string,\nbytes, integer, floating point number, complex number) with the given\nvalue.  The value may be approximated in the case of floating point\nand imaginary (complex) literals.  See section *Literals* for details.\n\nWith the exception of bytes literals, these all correspond to\nimmutable data types, and hence the object's identity is less\nimportant than its value. Multiple evaluations of literals with the\nsame value (either the same occurrence in the program text or a\ndifferent occurrence) may obtain the same object or a different object\nwith the same value.\n",
- 'attribute-access': '\nCustomizing attribute access\n****************************\n\nThe following methods can be defined to customize the meaning of\nattribute access (use of, assignment to, or deletion of ``x.name``)\nfor class instances.\n\nobject.__getattr__(self, name)\n\n   Called when an attribute lookup has not found the attribute in the\n   usual places (i.e. it is not an instance attribute nor is it found\n   in the class tree for ``self``).  ``name`` is the attribute name.\n   This method should return the (computed) attribute value or raise\n   an ``AttributeError`` exception.\n\n   Note that if the attribute is found through the normal mechanism,\n   ``__getattr__()`` is not called.  (This is an intentional asymmetry\n   between ``__getattr__()`` and ``__setattr__()``.) This is done both\n   for efficiency reasons and because otherwise ``__getattr__()``\n   would have no way to access other attributes of the instance.  Note\n   that at least for instance variables, you can fake total control by\n   not inserting any values in the instance attribute dictionary (but\n   instead inserting them in another object).  See the\n   ``__getattribute__()`` method below for a way to actually get total\n   control over attribute access.\n\nobject.__getattribute__(self, name)\n\n   Called unconditionally to implement attribute accesses for\n   instances of the class. If the class also defines\n   ``__getattr__()``, the latter will not be called unless\n   ``__getattribute__()`` either calls it explicitly or raises an\n   ``AttributeError``. This method should return the (computed)\n   attribute value or raise an ``AttributeError`` exception. In order\n   to avoid infinite recursion in this method, its implementation\n   should always call the base class method with the same name to\n   access any attributes it needs, for example,\n   ``object.__getattribute__(self, name)``.\n\n   Note: This method may still be bypassed when looking up special methods\n     as the result of implicit invocation via language syntax or\n     builtin functions. See *Special method lookup*.\n\nobject.__setattr__(self, name, value)\n\n   Called when an attribute assignment is attempted.  This is called\n   instead of the normal mechanism (i.e. store the value in the\n   instance dictionary). *name* is the attribute name, *value* is the\n   value to be assigned to it.\n\n   If ``__setattr__()`` wants to assign to an instance attribute, it\n   should call the base class method with the same name, for example,\n   ``object.__setattr__(self, name, value)``.\n\nobject.__delattr__(self, name)\n\n   Like ``__setattr__()`` but for attribute deletion instead of\n   assignment.  This should only be implemented if ``del obj.name`` is\n   meaningful for the object.\n\nobject.__dir__(self)\n\n   Called when ``dir()`` is called on the object.  A list must be\n   returned.\n\n\nImplementing Descriptors\n========================\n\nThe following methods only apply when an instance of the class\ncontaining the method (a so-called *descriptor* class) appears in the\nclass dictionary of another class, known as the *owner* class.  In the\nexamples below, "the attribute" refers to the attribute whose name is\nthe key of the property in the owner class\' ``__dict__``.\n\nobject.__get__(self, instance, owner)\n\n   Called to get the attribute of the owner class (class attribute\n   access) or of an instance of that class (instance attribute\n   access). *owner* is always the owner class, while *instance* is the\n   instance that the attribute was accessed through, or ``None`` when\n   the attribute is accessed through the *owner*.  This method should\n   return the (computed) attribute value or raise an\n   ``AttributeError`` exception.\n\nobject.__set__(self, instance, value)\n\n   Called to set the attribute on an instance *instance* of the owner\n   class to a new value, *value*.\n\nobject.__delete__(self, instance)\n\n   Called to delete the attribute on an instance *instance* of the\n   owner class.\n\n\nInvoking Descriptors\n====================\n\nIn general, a descriptor is an object attribute with "binding\nbehavior", one whose attribute access has been overridden by methods\nin the descriptor protocol:  ``__get__()``, ``__set__()``, and\n``__delete__()``. If any of those methods are defined for an object,\nit is said to be a descriptor.\n\nThe default behavior for attribute access is to get, set, or delete\nthe attribute from an object\'s dictionary. For instance, ``a.x`` has a\nlookup chain starting with ``a.__dict__[\'x\']``, then\n``type(a).__dict__[\'x\']``, and continuing through the base classes of\n``type(a)`` excluding metaclasses.\n\nHowever, if the looked-up value is an object defining one of the\ndescriptor methods, then Python may override the default behavior and\ninvoke the descriptor method instead.  Where this occurs in the\nprecedence chain depends on which descriptor methods were defined and\nhow they were called.\n\nThe starting point for descriptor invocation is a binding, ``a.x``.\nHow the arguments are assembled depends on ``a``:\n\nDirect Call\n   The simplest and least common call is when user code directly\n   invokes a descriptor method:    ``x.__get__(a)``.\n\nInstance Binding\n   If binding to an object instance, ``a.x`` is transformed into the\n   call: ``type(a).__dict__[\'x\'].__get__(a, type(a))``.\n\nClass Binding\n   If binding to a class, ``A.x`` is transformed into the call:\n   ``A.__dict__[\'x\'].__get__(None, A)``.\n\nSuper Binding\n   If ``a`` is an instance of ``super``, then the binding ``super(B,\n   obj).m()`` searches ``obj.__class__.__mro__`` for the base class\n   ``A`` immediately preceding ``B`` and then invokes the descriptor\n   with the call: ``A.__dict__[\'m\'].__get__(obj, A)``.\n\nFor instance bindings, the precedence of descriptor invocation depends\non the which descriptor methods are defined.  Normally, data\ndescriptors define both ``__get__()`` and ``__set__()``, while non-\ndata descriptors have just the ``__get__()`` method.  Data descriptors\nalways override a redefinition in an instance dictionary.  In\ncontrast, non-data descriptors can be overridden by instances. [2]\n\nPython methods (including ``staticmethod()`` and ``classmethod()``)\nare implemented as non-data descriptors.  Accordingly, instances can\nredefine and override methods.  This allows individual instances to\nacquire behaviors that differ from other instances of the same class.\n\nThe ``property()`` function is implemented as a data descriptor.\nAccordingly, instances cannot override the behavior of a property.\n\n\n__slots__\n=========\n\nBy default, instances of classes have a dictionary for attribute\nstorage.  This wastes space for objects having very few instance\nvariables.  The space consumption can become acute when creating large\nnumbers of instances.\n\nThe default can be overridden by defining *__slots__* in a class\ndefinition. The *__slots__* declaration takes a sequence of instance\nvariables and reserves just enough space in each instance to hold a\nvalue for each variable.  Space is saved because *__dict__* is not\ncreated for each instance.\n\nobject.__slots__\n\n   This class variable can be assigned a string, iterable, or sequence\n   of strings with variable names used by instances.  If defined in a\n   class, *__slots__* reserves space for the declared variables and\n   prevents the automatic creation of *__dict__* and *__weakref__* for\n   each instance.\n\n\nNotes on using *__slots__*\n--------------------------\n\n* When inheriting from a class without *__slots__*, the *__dict__*\n  attribute of that class will always be accessible, so a *__slots__*\n  definition in the subclass is meaningless.\n\n* Without a *__dict__* variable, instances cannot be assigned new\n  variables not listed in the *__slots__* definition.  Attempts to\n  assign to an unlisted variable name raises ``AttributeError``. If\n  dynamic assignment of new variables is desired, then add\n  ``\'__dict__\'`` to the sequence of strings in the *__slots__*\n  declaration.\n\n* Without a *__weakref__* variable for each instance, classes defining\n  *__slots__* do not support weak references to its instances. If weak\n  reference support is needed, then add ``\'__weakref__\'`` to the\n  sequence of strings in the *__slots__* declaration.\n\n* *__slots__* are implemented at the class level by creating\n  descriptors (*Implementing Descriptors*) for each variable name.  As\n  a result, class attributes cannot be used to set default values for\n  instance variables defined by *__slots__*; otherwise, the class\n  attribute would overwrite the descriptor assignment.\n\n* If a class defines a slot also defined in a base class, the instance\n  variable defined by the base class slot is inaccessible (except by\n  retrieving its descriptor directly from the base class). This\n  renders the meaning of the program undefined.  In the future, a\n  check may be added to prevent this.\n\n* The action of a *__slots__* declaration is limited to the class\n  where it is defined.  As a result, subclasses will have a *__dict__*\n  unless they also define *__slots__*.\n\n* Nonempty *__slots__* does not work for classes derived from\n  "variable-length" built-in types such as ``int``, ``str`` and\n  ``tuple``.\n\n* Any non-string iterable may be assigned to *__slots__*. Mappings may\n  also be used; however, in the future, special meaning may be\n  assigned to the values corresponding to each key.\n\n* *__class__* assignment works only if both classes have the same\n  *__slots__*.\n',
+ 'attribute-access': '\nCustomizing attribute access\n****************************\n\nThe following methods can be defined to customize the meaning of\nattribute access (use of, assignment to, or deletion of ``x.name``)\nfor class instances.\n\nobject.__getattr__(self, name)\n\n   Called when an attribute lookup has not found the attribute in the\n   usual places (i.e. it is not an instance attribute nor is it found\n   in the class tree for ``self``).  ``name`` is the attribute name.\n   This method should return the (computed) attribute value or raise\n   an ``AttributeError`` exception.\n\n   Note that if the attribute is found through the normal mechanism,\n   ``__getattr__()`` is not called.  (This is an intentional asymmetry\n   between ``__getattr__()`` and ``__setattr__()``.) This is done both\n   for efficiency reasons and because otherwise ``__getattr__()``\n   would have no way to access other attributes of the instance.  Note\n   that at least for instance variables, you can fake total control by\n   not inserting any values in the instance attribute dictionary (but\n   instead inserting them in another object).  See the\n   ``__getattribute__()`` method below for a way to actually get total\n   control over attribute access.\n\nobject.__getattribute__(self, name)\n\n   Called unconditionally to implement attribute accesses for\n   instances of the class. If the class also defines\n   ``__getattr__()``, the latter will not be called unless\n   ``__getattribute__()`` either calls it explicitly or raises an\n   ``AttributeError``. This method should return the (computed)\n   attribute value or raise an ``AttributeError`` exception. In order\n   to avoid infinite recursion in this method, its implementation\n   should always call the base class method with the same name to\n   access any attributes it needs, for example,\n   ``object.__getattribute__(self, name)``.\n\n   Note: This method may still be bypassed when looking up special methods\n     as the result of implicit invocation via language syntax or\n     built-in functions. See *Special method lookup*.\n\nobject.__setattr__(self, name, value)\n\n   Called when an attribute assignment is attempted.  This is called\n   instead of the normal mechanism (i.e. store the value in the\n   instance dictionary). *name* is the attribute name, *value* is the\n   value to be assigned to it.\n\n   If ``__setattr__()`` wants to assign to an instance attribute, it\n   should call the base class method with the same name, for example,\n   ``object.__setattr__(self, name, value)``.\n\nobject.__delattr__(self, name)\n\n   Like ``__setattr__()`` but for attribute deletion instead of\n   assignment.  This should only be implemented if ``del obj.name`` is\n   meaningful for the object.\n\nobject.__dir__(self)\n\n   Called when ``dir()`` is called on the object.  A list must be\n   returned.\n\n\nImplementing Descriptors\n========================\n\nThe following methods only apply when an instance of the class\ncontaining the method (a so-called *descriptor* class) appears in the\nclass dictionary of another class, known as the *owner* class.  In the\nexamples below, "the attribute" refers to the attribute whose name is\nthe key of the property in the owner class\' ``__dict__``.\n\nobject.__get__(self, instance, owner)\n\n   Called to get the attribute of the owner class (class attribute\n   access) or of an instance of that class (instance attribute\n   access). *owner* is always the owner class, while *instance* is the\n   instance that the attribute was accessed through, or ``None`` when\n   the attribute is accessed through the *owner*.  This method should\n   return the (computed) attribute value or raise an\n   ``AttributeError`` exception.\n\nobject.__set__(self, instance, value)\n\n   Called to set the attribute on an instance *instance* of the owner\n   class to a new value, *value*.\n\nobject.__delete__(self, instance)\n\n   Called to delete the attribute on an instance *instance* of the\n   owner class.\n\n\nInvoking Descriptors\n====================\n\nIn general, a descriptor is an object attribute with "binding\nbehavior", one whose attribute access has been overridden by methods\nin the descriptor protocol:  ``__get__()``, ``__set__()``, and\n``__delete__()``. If any of those methods are defined for an object,\nit is said to be a descriptor.\n\nThe default behavior for attribute access is to get, set, or delete\nthe attribute from an object\'s dictionary. For instance, ``a.x`` has a\nlookup chain starting with ``a.__dict__[\'x\']``, then\n``type(a).__dict__[\'x\']``, and continuing through the base classes of\n``type(a)`` excluding metaclasses.\n\nHowever, if the looked-up value is an object defining one of the\ndescriptor methods, then Python may override the default behavior and\ninvoke the descriptor method instead.  Where this occurs in the\nprecedence chain depends on which descriptor methods were defined and\nhow they were called.\n\nThe starting point for descriptor invocation is a binding, ``a.x``.\nHow the arguments are assembled depends on ``a``:\n\nDirect Call\n   The simplest and least common call is when user code directly\n   invokes a descriptor method:    ``x.__get__(a)``.\n\nInstance Binding\n   If binding to an object instance, ``a.x`` is transformed into the\n   call: ``type(a).__dict__[\'x\'].__get__(a, type(a))``.\n\nClass Binding\n   If binding to a class, ``A.x`` is transformed into the call:\n   ``A.__dict__[\'x\'].__get__(None, A)``.\n\nSuper Binding\n   If ``a`` is an instance of ``super``, then the binding ``super(B,\n   obj).m()`` searches ``obj.__class__.__mro__`` for the base class\n   ``A`` immediately preceding ``B`` and then invokes the descriptor\n   with the call: ``A.__dict__[\'m\'].__get__(obj, A)``.\n\nFor instance bindings, the precedence of descriptor invocation depends\non the which descriptor methods are defined.  A descriptor can define\nany combination of ``__get__()``, ``__set__()`` and ``__delete__()``.\nIf it does not define ``__get__()``, then accessing the attribute will\nreturn the descriptor object itself unless there is a value in the\nobject\'s instance dictionary.  If the descriptor defines ``__set__()``\nand/or ``__delete__()``, it is a data descriptor; if it defines\nneither, it is a non-data descriptor.  Normally, data descriptors\ndefine both ``__get__()`` and ``__set__()``, while non-data\ndescriptors have just the ``__get__()`` method.  Data descriptors with\n``__set__()`` and ``__get__()`` defined always override a redefinition\nin an instance dictionary.  In contrast, non-data descriptors can be\noverridden by instances.\n\nPython methods (including ``staticmethod()`` and ``classmethod()``)\nare implemented as non-data descriptors.  Accordingly, instances can\nredefine and override methods.  This allows individual instances to\nacquire behaviors that differ from other instances of the same class.\n\nThe ``property()`` function is implemented as a data descriptor.\nAccordingly, instances cannot override the behavior of a property.\n\n\n__slots__\n=========\n\nBy default, instances of classes have a dictionary for attribute\nstorage.  This wastes space for objects having very few instance\nvariables.  The space consumption can become acute when creating large\nnumbers of instances.\n\nThe default can be overridden by defining *__slots__* in a class\ndefinition. The *__slots__* declaration takes a sequence of instance\nvariables and reserves just enough space in each instance to hold a\nvalue for each variable.  Space is saved because *__dict__* is not\ncreated for each instance.\n\nobject.__slots__\n\n   This class variable can be assigned a string, iterable, or sequence\n   of strings with variable names used by instances.  If defined in a\n   class, *__slots__* reserves space for the declared variables and\n   prevents the automatic creation of *__dict__* and *__weakref__* for\n   each instance.\n\n\nNotes on using *__slots__*\n--------------------------\n\n* When inheriting from a class without *__slots__*, the *__dict__*\n  attribute of that class will always be accessible, so a *__slots__*\n  definition in the subclass is meaningless.\n\n* Without a *__dict__* variable, instances cannot be assigned new\n  variables not listed in the *__slots__* definition.  Attempts to\n  assign to an unlisted variable name raises ``AttributeError``. If\n  dynamic assignment of new variables is desired, then add\n  ``\'__dict__\'`` to the sequence of strings in the *__slots__*\n  declaration.\n\n* Without a *__weakref__* variable for each instance, classes defining\n  *__slots__* do not support weak references to its instances. If weak\n  reference support is needed, then add ``\'__weakref__\'`` to the\n  sequence of strings in the *__slots__* declaration.\n\n* *__slots__* are implemented at the class level by creating\n  descriptors (*Implementing Descriptors*) for each variable name.  As\n  a result, class attributes cannot be used to set default values for\n  instance variables defined by *__slots__*; otherwise, the class\n  attribute would overwrite the descriptor assignment.\n\n* The action of a *__slots__* declaration is limited to the class\n  where it is defined.  As a result, subclasses will have a *__dict__*\n  unless they also define *__slots__* (which must only contain names\n  of any *additional* slots).\n\n* If a class defines a slot also defined in a base class, the instance\n  variable defined by the base class slot is inaccessible (except by\n  retrieving its descriptor directly from the base class). This\n  renders the meaning of the program undefined.  In the future, a\n  check may be added to prevent this.\n\n* Nonempty *__slots__* does not work for classes derived from\n  "variable-length" built-in types such as ``int``, ``str`` and\n  ``tuple``.\n\n* Any non-string iterable may be assigned to *__slots__*. Mappings may\n  also be used; however, in the future, special meaning may be\n  assigned to the values corresponding to each key.\n\n* *__class__* assignment works only if both classes have the same\n  *__slots__*.\n',
  'attribute-references': '\nAttribute references\n********************\n\nAn attribute reference is a primary followed by a period and a name:\n\n   attributeref ::= primary "." identifier\n\nThe primary must evaluate to an object of a type that supports\nattribute references, which most objects do.  This object is then\nasked to produce the attribute whose name is the identifier (which can\nbe customized by overriding the ``__getattr__()`` method).  If this\nattribute is not available, the exception ``AttributeError`` is\nraised.  Otherwise, the type and value of the object produced is\ndetermined by the object.  Multiple evaluations of the same attribute\nreference may yield different objects.\n',
- 'augassign': '\nAugmented assignment statements\n*******************************\n\nAugmented assignment is the combination, in a single statement, of a\nbinary operation and an assignment statement:\n\n   augmented_assignment_stmt ::= augtarget augop (expression_list | yield_expression)\n   augtarget                 ::= identifier | attributeref | subscription | slicing\n   augop                     ::= "+=" | "-=" | "*=" | "/=" | "//=" | "%=" | "**="\n             | ">>=" | "<<=" | "&=" | "^=" | "|="\n\n(See section *Primaries* for the syntax definitions for the last three\nsymbols.)\n\nAn augmented assignment evaluates the target (which, unlike normal\nassignment statements, cannot be an unpacking) and the expression\nlist, performs the binary operation specific to the type of assignment\non the two operands, and assigns the result to the original target.\nThe target is only evaluated once.\n\nAn augmented assignment expression like ``x += 1`` can be rewritten as\n``x = x + 1`` to achieve a similar, but not exactly equal effect. In\nthe augmented version, ``x`` is only evaluated once. Also, when\npossible, the actual operation is performed *in-place*, meaning that\nrather than creating a new object and assigning that to the target,\nthe old object is modified instead.\n\nWith the exception of assigning to tuples and multiple targets in a\nsingle statement, the assignment done by augmented assignment\nstatements is handled the same way as normal assignments. Similarly,\nwith the exception of the possible *in-place* behavior, the binary\noperation performed by augmented assignment is the same as the normal\nbinary operations.\n\nFor targets which are attribute references, the initial value is\nretrieved with a ``getattr()`` and the result is assigned with a\n``setattr()``.  Notice that the two methods do not necessarily refer\nto the same variable.  When ``getattr()`` refers to a class variable,\n``setattr()`` still writes to an instance variable. For example:\n\n   class A:\n       x = 3    # class variable\n   a = A()\n   a.x += 1     # writes a.x as 4 leaving A.x as 3\n',
+ 'augassign': '\nAugmented assignment statements\n*******************************\n\nAugmented assignment is the combination, in a single statement, of a\nbinary operation and an assignment statement:\n\n   augmented_assignment_stmt ::= augtarget augop (expression_list | yield_expression)\n   augtarget                 ::= identifier | attributeref | subscription | slicing\n   augop                     ::= "+=" | "-=" | "*=" | "/=" | "//=" | "%=" | "**="\n             | ">>=" | "<<=" | "&=" | "^=" | "|="\n\n(See section *Primaries* for the syntax definitions for the last three\nsymbols.)\n\nAn augmented assignment evaluates the target (which, unlike normal\nassignment statements, cannot be an unpacking) and the expression\nlist, performs the binary operation specific to the type of assignment\non the two operands, and assigns the result to the original target.\nThe target is only evaluated once.\n\nAn augmented assignment expression like ``x += 1`` can be rewritten as\n``x = x + 1`` to achieve a similar, but not exactly equal effect. In\nthe augmented version, ``x`` is only evaluated once. Also, when\npossible, the actual operation is performed *in-place*, meaning that\nrather than creating a new object and assigning that to the target,\nthe old object is modified instead.\n\nWith the exception of assigning to tuples and multiple targets in a\nsingle statement, the assignment done by augmented assignment\nstatements is handled the same way as normal assignments. Similarly,\nwith the exception of the possible *in-place* behavior, the binary\noperation performed by augmented assignment is the same as the normal\nbinary operations.\n\nFor targets which are attribute references, the same *caveat about\nclass and instance attributes* applies as for regular assignments.\n',
  'binary': '\nBinary arithmetic operations\n****************************\n\nThe binary arithmetic operations have the conventional priority\nlevels.  Note that some of these operations also apply to certain non-\nnumeric types.  Apart from the power operator, there are only two\nlevels, one for multiplicative operators and one for additive\noperators:\n\n   m_expr ::= u_expr | m_expr "*" u_expr | m_expr "//" u_expr | m_expr "/" u_expr\n              | m_expr "%" u_expr\n   a_expr ::= m_expr | a_expr "+" m_expr | a_expr "-" m_expr\n\nThe ``*`` (multiplication) operator yields the product of its\narguments.  The arguments must either both be numbers, or one argument\nmust be an integer and the other must be a sequence. In the former\ncase, the numbers are converted to a common type and then multiplied\ntogether.  In the latter case, sequence repetition is performed; a\nnegative repetition factor yields an empty sequence.\n\nThe ``/`` (division) and ``//`` (floor division) operators yield the\nquotient of their arguments.  The numeric arguments are first\nconverted to a common type. Integer division yields a float, while\nfloor division of integers results in an integer; the result is that\nof mathematical division with the \'floor\' function applied to the\nresult.  Division by zero raises the ``ZeroDivisionError`` exception.\n\nThe ``%`` (modulo) operator yields the remainder from the division of\nthe first argument by the second.  The numeric arguments are first\nconverted to a common type.  A zero right argument raises the\n``ZeroDivisionError`` exception.  The arguments may be floating point\nnumbers, e.g., ``3.14%0.7`` equals ``0.34`` (since ``3.14`` equals\n``4*0.7 + 0.34``.)  The modulo operator always yields a result with\nthe same sign as its second operand (or zero); the absolute value of\nthe result is strictly smaller than the absolute value of the second\noperand [1].\n\nThe floor division and modulo operators are connected by the following\nidentity: ``x == (x//y)*y + (x%y)``.  Floor division and modulo are\nalso connected with the built-in function ``divmod()``: ``divmod(x, y)\n== (x//y, x%y)``. [2].\n\nIn addition to performing the modulo operation on numbers, the ``%``\noperator is also overloaded by string objects to perform old-style\nstring formatting (also known as interpolation).  The syntax for\nstring formatting is described in the Python Library Reference,\nsection *Old String Formatting Operations*.\n\nThe floor division operator, the modulo operator, and the ``divmod()``\nfunction are not defined for complex numbers.  Instead, convert to a\nfloating point number using the ``abs()`` function if appropriate.\n\nThe ``+`` (addition) operator yields the sum of its arguments.  The\narguments must either both be numbers or both sequences of the same\ntype.  In the former case, the numbers are converted to a common type\nand then added together.  In the latter case, the sequences are\nconcatenated.\n\nThe ``-`` (subtraction) operator yields the difference of its\narguments.  The numeric arguments are first converted to a common\ntype.\n',
  'bitwise': '\nBinary bitwise operations\n*************************\n\nEach of the three bitwise operations has a different priority level:\n\n   and_expr ::= shift_expr | and_expr "&" shift_expr\n   xor_expr ::= and_expr | xor_expr "^" and_expr\n   or_expr  ::= xor_expr | or_expr "|" xor_expr\n\nThe ``&`` operator yields the bitwise AND of its arguments, which must\nbe integers.\n\nThe ``^`` operator yields the bitwise XOR (exclusive OR) of its\narguments, which must be integers.\n\nThe ``|`` operator yields the bitwise (inclusive) OR of its arguments,\nwhich must be integers.\n',
  'bltin-code-objects': '\nCode Objects\n************\n\nCode objects are used by the implementation to represent "pseudo-\ncompiled" executable Python code such as a function body. They differ\nfrom function objects because they don\'t contain a reference to their\nglobal execution environment.  Code objects are returned by the built-\nin ``compile()`` function and can be extracted from function objects\nthrough their ``__code__`` attribute. See also the ``code`` module.\n\nA code object can be executed or evaluated by passing it (instead of a\nsource string) to the ``exec()`` or ``eval()``  built-in functions.\n\nSee *The standard type hierarchy* for more information.\n',
  'bltin-ellipsis-object': '\nThe Ellipsis Object\n*******************\n\nThis object is commonly used by slicing (see *Slicings*).  It supports\nno special operations.  There is exactly one ellipsis object, named\n``Ellipsis`` (a built-in name).\n\nIt is written as ``Ellipsis`` or ``...``.\n',
- 'bltin-file-objects': '\nFile Objects\n************\n\nFile objects are implemented using C\'s ``stdio`` package and can be\ncreated with the built-in ``open()`` function.  File objects are also\nreturned by some other built-in functions and methods, such as\n``os.popen()`` and ``os.fdopen()`` and the ``makefile()`` method of\nsocket objects. Temporary files can be created using the ``tempfile``\nmodule, and high-level file operations such as copying, moving, and\ndeleting files and directories can be achieved with the ``shutil``\nmodule.\n\nWhen a file operation fails for an I/O-related reason, the exception\n``IOError`` is raised.  This includes situations where the operation\nis not defined for some reason, like ``seek()`` on a tty device or\nwriting a file opened for reading.\n\nFiles have the following methods:\n\nfile.close()\n\n   Close the file.  A closed file cannot be read or written any more.\n   Any operation which requires that the file be open will raise a\n   ``ValueError`` after the file has been closed.  Calling ``close()``\n   more than once is allowed.\n\n   You can avoid having to call this method explicitly if you use the\n   ``with`` statement.  For example, the following code will\n   automatically close *f* when the ``with`` block is exited:\n\n      from __future__ import with_statement # This isn\'t required in Python 2.6\n\n      with open("hello.txt") as f:\n          for line in f:\n              print(line)\n\n   In older versions of Python, you would have needed to do this to\n   get the same effect:\n\n      f = open("hello.txt")\n      try:\n          for line in f:\n              print(line)\n      finally:\n          f.close()\n\n   Note: Not all "file-like" types in Python support use as a context\n     manager for the ``with`` statement.  If your code is intended to\n     work with any file-like object, you can use the function\n     ``contextlib.closing()`` instead of using the object directly.\n\nfile.flush()\n\n   Flush the internal buffer, like ``stdio``\'s ``fflush()``.  This may\n   be a no-op on some file-like objects.\n\n   Note: ``flush()`` does not necessarily write the file\'s data to disk.\n     Use ``flush()`` followed by ``os.fsync()`` to ensure this\n     behavior.\n\nfile.fileno()\n\n   Return the integer "file descriptor" that is used by the underlying\n   implementation to request I/O operations from the operating system.\n   This can be useful for other, lower level interfaces that use file\n   descriptors, such as the ``fcntl`` module or ``os.read()`` and\n   friends.\n\n   Note: File-like objects which do not have a real file descriptor should\n     *not* provide this method!\n\nfile.isatty()\n\n   Return ``True`` if the file is connected to a tty(-like) device,\n   else ``False``.\n\n   Note: If a file-like object is not associated with a real file, this\n     method should *not* be implemented.\n\nfile.__next__()\n\n   A file object is its own iterator, for example ``iter(f)`` returns\n   *f* (unless *f* is closed).  When a file is used as an iterator,\n   typically in a ``for`` loop (for example, ``for line in f:\n   print(line)``), the ``__next__()`` method is called repeatedly.\n   This method returns the next input line, or raises\n   ``StopIteration`` when EOF is hit when the file is open for reading\n   (behavior is undefined when the file is open for writing).  In\n   order to make a ``for`` loop the most efficient way of looping over\n   the lines of a file (a very common operation), the ``__next__()``\n   method uses a hidden read-ahead buffer.  As a consequence of using\n   a read-ahead buffer, combining ``__next__()`` with other file\n   methods (like ``readline()``) does not work right.  However, using\n   ``seek()`` to reposition the file to an absolute position will\n   flush the read-ahead buffer.\n\nfile.read([size])\n\n   Read at most *size* bytes from the file (less if the read hits EOF\n   before obtaining *size* bytes).  If the *size* argument is negative\n   or omitted, read all data until EOF is reached.  The bytes are\n   returned as a string object.  An empty string is returned when EOF\n   is encountered immediately.  (For certain files, like ttys, it\n   makes sense to continue reading after an EOF is hit.)  Note that\n   this method may call the underlying C function ``fread()`` more\n   than once in an effort to acquire as close to *size* bytes as\n   possible. Also note that when in non-blocking mode, less data than\n   was requested may be returned, even if no *size* parameter was\n   given.\n\nfile.readline([size])\n\n   Read one entire line from the file.  A trailing newline character\n   is kept in the string (but may be absent when a file ends with an\n   incomplete line). [5]  If the *size* argument is present and non-\n   negative, it is a maximum byte count (including the trailing\n   newline) and an incomplete line may be returned. An empty string is\n   returned *only* when EOF is encountered immediately.\n\n   Note: Unlike ``stdio``\'s ``fgets()``, the returned string contains null\n     characters (``\'\\0\'``) if they occurred in the input.\n\nfile.readlines([sizehint])\n\n   Read until EOF using ``readline()`` and return a list containing\n   the lines thus read.  If the optional *sizehint* argument is\n   present, instead of reading up to EOF, whole lines totalling\n   approximately *sizehint* bytes (possibly after rounding up to an\n   internal buffer size) are read.  Objects implementing a file-like\n   interface may choose to ignore *sizehint* if it cannot be\n   implemented, or cannot be implemented efficiently.\n\nfile.seek(offset[, whence])\n\n   Set the file\'s current position, like ``stdio``\'s ``fseek()``. The\n   *whence* argument is optional and defaults to  ``os.SEEK_SET`` or\n   ``0`` (absolute file positioning); other values are ``os.SEEK_CUR``\n   or ``1`` (seek relative to the current position) and\n   ``os.SEEK_END`` or ``2``  (seek relative to the file\'s end).  There\n   is no return value.\n\n   For example, ``f.seek(2, os.SEEK_CUR)`` advances the position by\n   two and ``f.seek(-3, os.SEEK_END)`` sets the position to the third\n   to last.\n\n   Note that if the file is opened for appending (mode ``\'a\'`` or\n   ``\'a+\'``), any ``seek()`` operations will be undone at the next\n   write.  If the file is only opened for writing in append mode (mode\n   ``\'a\'``), this method is essentially a no-op, but it remains useful\n   for files opened in append mode with reading enabled (mode\n   ``\'a+\'``).  If the file is opened in text mode (without ``\'b\'``),\n   only offsets returned by ``tell()`` are legal.  Use of other\n   offsets causes undefined behavior.\n\n   Note that not all file objects are seekable.\n\nfile.tell()\n\n   Return the file\'s current position, like ``stdio``\'s ``ftell()``.\n\n   Note: On Windows, ``tell()`` can return illegal values (after an\n     ``fgets()``) when reading files with Unix-style line-endings. Use\n     binary mode (``\'rb\'``) to circumvent this problem.\n\nfile.truncate([size])\n\n   Truncate the file\'s size.  If the optional *size* argument is\n   present, the file is truncated to (at most) that size.  The size\n   defaults to the current position. The current file position is not\n   changed.  Note that if a specified size exceeds the file\'s current\n   size, the result is platform-dependent:  possibilities include that\n   the file may remain unchanged, increase to the specified size as if\n   zero-filled, or increase to the specified size with undefined new\n   content. Availability:  Windows, many Unix variants.\n\nfile.write(str)\n\n   Write a string to the file.  Due to buffering, the string may not\n   actually show up in the file until the ``flush()`` or ``close()``\n   method is called.\n\n   The meaning of the return value is not defined for every file-like\n   object. Some (mostly low-level) file-like objects may return the\n   number of bytes actually written, others return ``None``.\n\nfile.writelines(sequence)\n\n   Write a sequence of strings to the file.  The sequence can be any\n   iterable object producing strings, typically a list of strings.\n   There is no return value. (The name is intended to match\n   ``readlines()``; ``writelines()`` does not add line separators.)\n\nFiles support the iterator protocol.  Each iteration returns the same\nresult as ``file.readline()``, and iteration ends when the\n``readline()`` method returns an empty string.\n\nFile objects also offer a number of other interesting attributes.\nThese are not required for file-like objects, but should be\nimplemented if they make sense for the particular object.\n\nfile.closed\n\n   bool indicating the current state of the file object.  This is a\n   read-only attribute; the ``close()`` method changes the value. It\n   may not be available on all file-like objects.\n\nfile.encoding\n\n   The encoding that this file uses. When strings are written to a\n   file, they will be converted to byte strings using this encoding.\n   In addition, when the file is connected to a terminal, the\n   attribute gives the encoding that the terminal is likely to use\n   (that  information might be incorrect if the user has misconfigured\n   the  terminal). The attribute is read-only and may not be present\n   on all file-like objects. It may also be ``None``, in which case\n   the file uses the system default encoding for converting strings.\n\nfile.errors\n\n   The Unicode error handler used along with the encoding.\n\nfile.mode\n\n   The I/O mode for the file.  If the file was created using the\n   ``open()`` built-in function, this will be the value of the *mode*\n   parameter.  This is a read-only attribute and may not be present on\n   all file-like objects.\n\nfile.name\n\n   If the file object was created using ``open()``, the name of the\n   file. Otherwise, some string that indicates the source of the file\n   object, of the form ``<...>``.  This is a read-only attribute and\n   may not be present on all file-like objects.\n\nfile.newlines\n\n   If Python was built with the *--with-universal-newlines* option to\n   **configure** (the default) this read-only attribute exists, and\n   for files opened in universal newline read mode it keeps track of\n   the types of newlines encountered while reading the file. The\n   values it can take are ``\'\\r\'``, ``\'\\n\'``, ``\'\\r\\n\'``, ``None``\n   (unknown, no newlines read yet) or a tuple containing all the\n   newline types seen, to indicate that multiple newline conventions\n   were encountered. For files not opened in universal newline read\n   mode the value of this attribute will be ``None``.\n',
  'bltin-null-object': "\nThe Null Object\n***************\n\nThis object is returned by functions that don't explicitly return a\nvalue.  It supports no special operations.  There is exactly one null\nobject, named ``None`` (a built-in name).\n\nIt is written as ``None``.\n",
  'bltin-type-objects': "\nType Objects\n************\n\nType objects represent the various object types.  An object's type is\naccessed by the built-in function ``type()``.  There are no special\noperations on types.  The standard module ``types`` defines names for\nall standard built-in types.\n\nTypes are written like this: ``<class 'int'>``.\n",
- 'booleans': '\nBoolean operations\n******************\n\nBoolean operations have the lowest priority of all Python operations:\n\n   expression             ::= conditional_expression | lambda_form\n   expression_nocond      ::= or_test | lambda_form_nocond\n   conditional_expression ::= or_test ["if" or_test "else" expression]\n   or_test                ::= and_test | or_test "or" and_test\n   and_test               ::= not_test | and_test "and" not_test\n   not_test               ::= comparison | "not" not_test\n\nIn the context of Boolean operations, and also when expressions are\nused by control flow statements, the following values are interpreted\nas false: ``False``, ``None``, numeric zero of all types, and empty\nstrings and containers (including strings, tuples, lists,\ndictionaries, sets and frozensets).  All other values are interpreted\nas true.  User-defined objects can customize their truth value by\nproviding a ``__bool__()`` method.\n\nThe operator ``not`` yields ``True`` if its argument is false,\n``False`` otherwise.\n\nThe expression ``x if C else y`` first evaluates *C* (*not* *x*); if\n*C* is true, *x* is evaluated and its value is returned; otherwise,\n*y* is evaluated and its value is returned.\n\nThe expression ``x and y`` first evaluates *x*; if *x* is false, its\nvalue is returned; otherwise, *y* is evaluated and the resulting value\nis returned.\n\nThe expression ``x or y`` first evaluates *x*; if *x* is true, its\nvalue is returned; otherwise, *y* is evaluated and the resulting value\nis returned.\n\n(Note that neither ``and`` nor ``or`` restrict the value and type they\nreturn to ``False`` and ``True``, but rather return the last evaluated\nargument.  This is sometimes useful, e.g., if ``s`` is a string that\nshould be replaced by a default value if it is empty, the expression\n``s or \'foo\'`` yields the desired value.  Because ``not`` has to\ninvent a value anyway, it does not bother to return a value of the\nsame type as its argument, so e.g., ``not \'foo\'`` yields ``False``,\nnot ``\'\'``.)\n',
+ 'booleans': '\nBoolean operations\n******************\n\n   or_test  ::= and_test | or_test "or" and_test\n   and_test ::= not_test | and_test "and" not_test\n   not_test ::= comparison | "not" not_test\n\nIn the context of Boolean operations, and also when expressions are\nused by control flow statements, the following values are interpreted\nas false: ``False``, ``None``, numeric zero of all types, and empty\nstrings and containers (including strings, tuples, lists,\ndictionaries, sets and frozensets).  All other values are interpreted\nas true.  User-defined objects can customize their truth value by\nproviding a ``__bool__()`` method.\n\nThe operator ``not`` yields ``True`` if its argument is false,\n``False`` otherwise.\n\nThe expression ``x and y`` first evaluates *x*; if *x* is false, its\nvalue is returned; otherwise, *y* is evaluated and the resulting value\nis returned.\n\nThe expression ``x or y`` first evaluates *x*; if *x* is true, its\nvalue is returned; otherwise, *y* is evaluated and the resulting value\nis returned.\n\n(Note that neither ``and`` nor ``or`` restrict the value and type they\nreturn to ``False`` and ``True``, but rather return the last evaluated\nargument.  This is sometimes useful, e.g., if ``s`` is a string that\nshould be replaced by a default value if it is empty, the expression\n``s or \'foo\'`` yields the desired value.  Because ``not`` has to\ninvent a value anyway, it does not bother to return a value of the\nsame type as its argument, so e.g., ``not \'foo\'`` yields ``False``,\nnot ``\'\'``.)\n',
  'break': '\nThe ``break`` statement\n***********************\n\n   break_stmt ::= "break"\n\n``break`` may only occur syntactically nested in a ``for`` or\n``while`` loop, but not nested in a function or class definition\nwithin that loop.\n\nIt terminates the nearest enclosing loop, skipping the optional\n``else`` clause if the loop has one.\n\nIf a ``for`` loop is terminated by ``break``, the loop control target\nkeeps its current value.\n\nWhen ``break`` passes control out of a ``try`` statement with a\n``finally`` clause, that ``finally`` clause is executed before really\nleaving the loop.\n',
  'callable-types': '\nEmulating callable objects\n**************************\n\nobject.__call__(self[, args...])\n\n   Called when the instance is "called" as a function; if this method\n   is defined, ``x(arg1, arg2, ...)`` is a shorthand for\n   ``x.__call__(arg1, arg2, ...)``.\n',
- 'calls': '\nCalls\n*****\n\nA call calls a callable object (e.g., a function) with a possibly\nempty series of arguments:\n\n   call                 ::= primary "(" [argument_list [","] | comprehension] ")"\n   argument_list        ::= positional_arguments ["," keyword_arguments]\n                       ["," "*" expression] ["," keyword_arguments]\n                       ["," "**" expression]\n                     | keyword_arguments ["," "*" expression]\n                       ["," keyword_arguments] ["," "**" expression]\n                     | "*" expression ["," keyword_arguments] ["," "**" expression]\n                     | "**" expression\n   positional_arguments ::= expression ("," expression)*\n   keyword_arguments    ::= keyword_item ("," keyword_item)*\n   keyword_item         ::= identifier "=" expression\n\nA trailing comma may be present after the positional and keyword\narguments but does not affect the semantics.\n\nThe primary must evaluate to a callable object (user-defined\nfunctions, built-in functions, methods of built-in objects, class\nobjects, methods of class instances, and all objects having a\n``__call__()`` method are callable).  All argument expressions are\nevaluated before the call is attempted.  Please refer to section\n*Function definitions* for the syntax of formal parameter lists.\n\nIf keyword arguments are present, they are first converted to\npositional arguments, as follows.  First, a list of unfilled slots is\ncreated for the formal parameters.  If there are N positional\narguments, they are placed in the first N slots.  Next, for each\nkeyword argument, the identifier is used to determine the\ncorresponding slot (if the identifier is the same as the first formal\nparameter name, the first slot is used, and so on).  If the slot is\nalready filled, a ``TypeError`` exception is raised. Otherwise, the\nvalue of the argument is placed in the slot, filling it (even if the\nexpression is ``None``, it fills the slot).  When all arguments have\nbeen processed, the slots that are still unfilled are filled with the\ncorresponding default value from the function definition.  (Default\nvalues are calculated, once, when the function is defined; thus, a\nmutable object such as a list or dictionary used as default value will\nbe shared by all calls that don\'t specify an argument value for the\ncorresponding slot; this should usually be avoided.)  If there are any\nunfilled slots for which no default value is specified, a\n``TypeError`` exception is raised.  Otherwise, the list of filled\nslots is used as the argument list for the call.\n\nNote: An implementation may provide builtin functions whose positional\n  parameters do not have names, even if they are \'named\' for the\n  purpose of documentation, and which therefore cannot be supplied by\n  keyword.  In CPython, this is the case for functions implemented in\n  C that use ``PyArg_ParseTuple()`` to parse their arguments.\n\nIf there are more positional arguments than there are formal parameter\nslots, a ``TypeError`` exception is raised, unless a formal parameter\nusing the syntax ``*identifier`` is present; in this case, that formal\nparameter receives a tuple containing the excess positional arguments\n(or an empty tuple if there were no excess positional arguments).\n\nIf any keyword argument does not correspond to a formal parameter\nname, a ``TypeError`` exception is raised, unless a formal parameter\nusing the syntax ``**identifier`` is present; in this case, that\nformal parameter receives a dictionary containing the excess keyword\narguments (using the keywords as keys and the argument values as\ncorresponding values), or a (new) empty dictionary if there were no\nexcess keyword arguments.\n\nIf the syntax ``*expression`` appears in the function call,\n``expression`` must evaluate to a sequence.  Elements from this\nsequence are treated as if they were additional positional arguments;\nif there are positional arguments *x1*,..., *xN*, and ``expression``\nevaluates to a sequence *y1*, ..., *yM*, this is equivalent to a call\nwith M+N positional arguments *x1*, ..., *xN*, *y1*, ..., *yM*.\n\nA consequence of this is that although the ``*expression`` syntax may\nappear *after* some keyword arguments, it is processed *before* the\nkeyword arguments (and the ``**expression`` argument, if any -- see\nbelow).  So:\n\n   >>> def f(a, b):\n   ...  print(a, b)\n   ...\n   >>> f(b=1, *(2,))\n   2 1\n   >>> f(a=1, *(2,))\n   Traceback (most recent call last):\n     File "<stdin>", line 1, in ?\n   TypeError: f() got multiple values for keyword argument \'a\'\n   >>> f(1, *(2,))\n   1 2\n\nIt is unusual for both keyword arguments and the ``*expression``\nsyntax to be used in the same call, so in practice this confusion does\nnot arise.\n\nIf the syntax ``**expression`` appears in the function call,\n``expression`` must evaluate to a mapping, the contents of which are\ntreated as additional keyword arguments.  In the case of a keyword\nappearing in both ``expression`` and as an explicit keyword argument,\na ``TypeError`` exception is raised.\n\nFormal parameters using the syntax ``*identifier`` or ``**identifier``\ncannot be used as positional argument slots or as keyword argument\nnames.\n\nA call always returns some value, possibly ``None``, unless it raises\nan exception.  How this value is computed depends on the type of the\ncallable object.\n\nIf it is---\n\na user-defined function:\n   The code block for the function is executed, passing it the\n   argument list.  The first thing the code block will do is bind the\n   formal parameters to the arguments; this is described in section\n   *Function definitions*.  When the code block executes a ``return``\n   statement, this specifies the return value of the function call.\n\na built-in function or method:\n   The result is up to the interpreter; see *Built-in Functions* for\n   the descriptions of built-in functions and methods.\n\na class object:\n   A new instance of that class is returned.\n\na class instance method:\n   The corresponding user-defined function is called, with an argument\n   list that is one longer than the argument list of the call: the\n   instance becomes the first argument.\n\na class instance:\n   The class must define a ``__call__()`` method; the effect is then\n   the same as if that method was called.\n',
+ 'calls': '\nCalls\n*****\n\nA call calls a callable object (e.g., a function) with a possibly\nempty series of arguments:\n\n   call                 ::= primary "(" [argument_list [","] | comprehension] ")"\n   argument_list        ::= positional_arguments ["," keyword_arguments]\n                       ["," "*" expression] ["," keyword_arguments]\n                       ["," "**" expression]\n                     | keyword_arguments ["," "*" expression]\n                       ["," keyword_arguments] ["," "**" expression]\n                     | "*" expression ["," keyword_arguments] ["," "**" expression]\n                     | "**" expression\n   positional_arguments ::= expression ("," expression)*\n   keyword_arguments    ::= keyword_item ("," keyword_item)*\n   keyword_item         ::= identifier "=" expression\n\nA trailing comma may be present after the positional and keyword\narguments but does not affect the semantics.\n\nThe primary must evaluate to a callable object (user-defined\nfunctions, built-in functions, methods of built-in objects, class\nobjects, methods of class instances, and all objects having a\n``__call__()`` method are callable).  All argument expressions are\nevaluated before the call is attempted.  Please refer to section\n*Function definitions* for the syntax of formal parameter lists.\n\nIf keyword arguments are present, they are first converted to\npositional arguments, as follows.  First, a list of unfilled slots is\ncreated for the formal parameters.  If there are N positional\narguments, they are placed in the first N slots.  Next, for each\nkeyword argument, the identifier is used to determine the\ncorresponding slot (if the identifier is the same as the first formal\nparameter name, the first slot is used, and so on).  If the slot is\nalready filled, a ``TypeError`` exception is raised. Otherwise, the\nvalue of the argument is placed in the slot, filling it (even if the\nexpression is ``None``, it fills the slot).  When all arguments have\nbeen processed, the slots that are still unfilled are filled with the\ncorresponding default value from the function definition.  (Default\nvalues are calculated, once, when the function is defined; thus, a\nmutable object such as a list or dictionary used as default value will\nbe shared by all calls that don\'t specify an argument value for the\ncorresponding slot; this should usually be avoided.)  If there are any\nunfilled slots for which no default value is specified, a\n``TypeError`` exception is raised.  Otherwise, the list of filled\nslots is used as the argument list for the call.\n\n**CPython implementation detail:** An implementation may provide\nbuilt-in functions whose positional parameters do not have names, even\nif they are \'named\' for the purpose of documentation, and which\ntherefore cannot be supplied by keyword.  In CPython, this is the case\nfor functions implemented in C that use ``PyArg_ParseTuple()`` to\nparse their arguments.\n\nIf there are more positional arguments than there are formal parameter\nslots, a ``TypeError`` exception is raised, unless a formal parameter\nusing the syntax ``*identifier`` is present; in this case, that formal\nparameter receives a tuple containing the excess positional arguments\n(or an empty tuple if there were no excess positional arguments).\n\nIf any keyword argument does not correspond to a formal parameter\nname, a ``TypeError`` exception is raised, unless a formal parameter\nusing the syntax ``**identifier`` is present; in this case, that\nformal parameter receives a dictionary containing the excess keyword\narguments (using the keywords as keys and the argument values as\ncorresponding values), or a (new) empty dictionary if there were no\nexcess keyword arguments.\n\nIf the syntax ``*expression`` appears in the function call,\n``expression`` must evaluate to a sequence.  Elements from this\nsequence are treated as if they were additional positional arguments;\nif there are positional arguments *x1*,..., *xN*, and ``expression``\nevaluates to a sequence *y1*, ..., *yM*, this is equivalent to a call\nwith M+N positional arguments *x1*, ..., *xN*, *y1*, ..., *yM*.\n\nA consequence of this is that although the ``*expression`` syntax may\nappear *after* some keyword arguments, it is processed *before* the\nkeyword arguments (and the ``**expression`` argument, if any -- see\nbelow).  So:\n\n   >>> def f(a, b):\n   ...  print(a, b)\n   ...\n   >>> f(b=1, *(2,))\n   2 1\n   >>> f(a=1, *(2,))\n   Traceback (most recent call last):\n     File "<stdin>", line 1, in ?\n   TypeError: f() got multiple values for keyword argument \'a\'\n   >>> f(1, *(2,))\n   1 2\n\nIt is unusual for both keyword arguments and the ``*expression``\nsyntax to be used in the same call, so in practice this confusion does\nnot arise.\n\nIf the syntax ``**expression`` appears in the function call,\n``expression`` must evaluate to a mapping, the contents of which are\ntreated as additional keyword arguments.  In the case of a keyword\nappearing in both ``expression`` and as an explicit keyword argument,\na ``TypeError`` exception is raised.\n\nFormal parameters using the syntax ``*identifier`` or ``**identifier``\ncannot be used as positional argument slots or as keyword argument\nnames.\n\nA call always returns some value, possibly ``None``, unless it raises\nan exception.  How this value is computed depends on the type of the\ncallable object.\n\nIf it is---\n\na user-defined function:\n   The code block for the function is executed, passing it the\n   argument list.  The first thing the code block will do is bind the\n   formal parameters to the arguments; this is described in section\n   *Function definitions*.  When the code block executes a ``return``\n   statement, this specifies the return value of the function call.\n\na built-in function or method:\n   The result is up to the interpreter; see *Built-in Functions* for\n   the descriptions of built-in functions and methods.\n\na class object:\n   A new instance of that class is returned.\n\na class instance method:\n   The corresponding user-defined function is called, with an argument\n   list that is one longer than the argument list of the call: the\n   instance becomes the first argument.\n\na class instance:\n   The class must define a ``__call__()`` method; the effect is then\n   the same as if that method was called.\n',
  'class': '\nClass definitions\n*****************\n\nA class definition defines a class object (see section *The standard\ntype hierarchy*):\n\n   classdef    ::= [decorators] "class" classname [inheritance] ":" suite\n   inheritance ::= "(" [expression_list] ")"\n   classname   ::= identifier\n\nA class definition is an executable statement.  It first evaluates the\ninheritance list, if present.  Each item in the inheritance list\nshould evaluate to a class object or class type which allows\nsubclassing.  The class\'s suite is then executed in a new execution\nframe (see section *Naming and binding*), using a newly created local\nnamespace and the original global namespace. (Usually, the suite\ncontains only function definitions.)  When the class\'s suite finishes\nexecution, its execution frame is discarded but its local namespace is\nsaved. [4] A class object is then created using the inheritance list\nfor the base classes and the saved local namespace for the attribute\ndictionary.  The class name is bound to this class object in the\noriginal local namespace.\n\nClasses can also be decorated; as with functions,\n\n   @f1(arg)\n   @f2\n   class Foo: pass\n\nis equivalent to\n\n   class Foo: pass\n   Foo = f1(arg)(f2(Foo))\n\n**Programmer\'s note:** Variables defined in the class definition are\nclass variables; they are shared by instances. Instance variables can\nbe set in a method with ``self.name = value``.  Both class and\ninstance variables are accessible through the notation\n"``self.name``", and an instance variable hides a class variable with\nthe same name when accessed in this way.  Class variables can be used\nas defaults for instance variables, but using mutable values there can\nlead to unexpected results.  Descriptors can be used to create\ninstance variables with different implementation details.\n\nSee also:\n\n   **PEP 3129** - Class Decorators\n\nClass definitions, like function definitions, may be wrapped by one or\nmore *decorator* expressions.  The evaluation rules for the decorator\nexpressions are the same as for functions.  The result must be a class\nobject, which is then bound to the class name.\n\n-[ Footnotes ]-\n\n[1] The exception is propagated to the invocation stack only if there\n    is no ``finally`` clause that negates the exception.\n\n[2] Currently, control "flows off the end" except in the case of an\n    exception or the execution of a ``return``, ``continue``, or\n    ``break`` statement.\n\n[3] A string literal appearing as the first statement in the function\n    body is transformed into the function\'s ``__doc__`` attribute and\n    therefore the function\'s *docstring*.\n\n[4] A string literal appearing as the first statement in the class\n    body is transformed into the namespace\'s ``__doc__`` item and\n    therefore the class\'s *docstring*.\n',
- 'comparisons': '\nComparisons\n***********\n\nUnlike C, all comparison operations in Python have the same priority,\nwhich is lower than that of any arithmetic, shifting or bitwise\noperation.  Also unlike C, expressions like ``a < b < c`` have the\ninterpretation that is conventional in mathematics:\n\n   comparison    ::= or_expr ( comp_operator or_expr )*\n   comp_operator ::= "<" | ">" | "==" | ">=" | "<=" | "!="\n                     | "is" ["not"] | ["not"] "in"\n\nComparisons yield boolean values: ``True`` or ``False``.\n\nComparisons can be chained arbitrarily, e.g., ``x < y <= z`` is\nequivalent to ``x < y and y <= z``, except that ``y`` is evaluated\nonly once (but in both cases ``z`` is not evaluated at all when ``x <\ny`` is found to be false).\n\nFormally, if *a*, *b*, *c*, ..., *y*, *z* are expressions and *op1*,\n*op2*, ..., *opN* are comparison operators, then ``a op1 b op2 c ... y\nopN z`` is equivalent to ``a op1 b and b op2 c and ... y opN z``,\nexcept that each expression is evaluated at most once.\n\nNote that ``a op1 b op2 c`` doesn\'t imply any kind of comparison\nbetween *a* and *c*, so that, e.g., ``x < y > z`` is perfectly legal\n(though perhaps not pretty).\n\nThe operators ``<``, ``>``, ``==``, ``>=``, ``<=``, and ``!=`` compare\nthe values of two objects.  The objects need not have the same type.\nIf both are numbers, they are converted to a common type.  Otherwise,\nthe ``==`` and ``!=`` operators *always* consider objects of different\ntypes to be unequal, while the ``<``, ``>``, ``>=`` and ``<=``\noperators raise a ``TypeError`` when comparing objects of different\ntypes that do not implement these operators for the given pair of\ntypes.  You can control comparison behavior of objects of non-builtin\ntypes by defining rich comparison methods like ``__gt__()``, described\nin section *Basic customization*.\n\nComparison of objects of the same type depends on the type:\n\n* Numbers are compared arithmetically.\n\n* The values ``float(\'NaN\')`` and ``Decimal(\'NaN\')`` are special. The\n  are identical to themselves, ``x is x`` but are not equal to\n  themselves, ``x != x``.  Additionally, comparing any value to a\n  not-a-number value will return ``False``.  For example, both ``3 <\n  float(\'NaN\')`` and ``float(\'NaN\') < 3`` will return ``False``.\n\n* Bytes objects are compared lexicographically using the numeric\n  values of their elements.\n\n* Strings are compared lexicographically using the numeric equivalents\n  (the result of the built-in function ``ord()``) of their characters.\n  [3] String and bytes object can\'t be compared!\n\n* Tuples and lists are compared lexicographically using comparison of\n  corresponding elements.  This means that to compare equal, each\n  element must compare equal and the two sequences must be of the same\n  type and have the same length.\n\n  If not equal, the sequences are ordered the same as their first\n  differing elements.  For example, ``[1,2,x] <= [1,2,y]`` has the\n  same value as ``x <= y``.  If the corresponding element does not\n  exist, the shorter sequence is ordered first (for example, ``[1,2] <\n  [1,2,3]``).\n\n* Mappings (dictionaries) compare equal if and only if their sorted\n  ``(key, value)`` lists compare equal. [4] Outcomes other than\n  equality are resolved consistently, but are not otherwise defined.\n  [5]\n\n* Sets and frozensets define comparison operators to mean subset and\n  superset tests.  Those relations do not define total orderings (the\n  two sets ``{1,2}`` and {2,3} are not equal, nor subsets of one\n  another, nor supersets of one another).  Accordingly, sets are not\n  appropriate arguments for functions which depend on total ordering.\n  For example, ``min()``, ``max()``, and ``sorted()`` produce\n  undefined results given a list of sets as inputs.\n\n* Most other objects of builtin types compare unequal unless they are\n  the same object; the choice whether one object is considered smaller\n  or larger than another one is made arbitrarily but consistently\n  within one execution of a program.\n\nComparison of objects of the differing types depends on whether either\nof the types provide explicit support for the comparison.  Most\nnumeric types can be compared with one another, but comparisons of\n``float`` and ``Decimal`` are not supported to avoid the inevitable\nconfusion arising from representation issues such as ``float(\'1.1\')``\nbeing inexactly represented and therefore not exactly equal to\n``Decimal(\'1.1\')`` which is.  When cross-type comparison is not\nsupported, the comparison method returns ``NotImplemented``.  This can\ncreate the illusion of non-transitivity between supported cross-type\ncomparisons and unsupported comparisons.  For example, ``Decimal(2) ==\n2`` and *2 == float(2)`* but ``Decimal(2) != float(2)``.\n\nThe operators ``in`` and ``not in`` test for membership.  ``x in s``\nevaluates to true if *x* is a member of *s*, and false otherwise.  ``x\nnot in s`` returns the negation of ``x in s``.  All built-in sequences\nand set types support this as well as dictionary, for which ``in``\ntests whether a the dictionary has a given key. For container types\nsuch as list, tuple, set, frozenset, dict, or collections.deque, the\nexpression ``x in y`` is equivalent to ``any(x is e or x == e for val\ne in y)``.\n\nFor the string and bytes types, ``x in y`` is true if and only if *x*\nis a substring of *y*.  An equivalent test is ``y.find(x) != -1``.\nEmpty strings are always considered to be a substring of any other\nstring, so ``"" in "abc"`` will return ``True``.\n\nFor user-defined classes which define the ``__contains__()`` method,\n``x in y`` is true if and only if ``y.__contains__(x)`` is true.\n\nFor user-defined classes which do not define ``__contains__()`` and do\ndefine ``__getitem__()``, ``x in y`` is true if and only if there is a\nnon-negative integer index *i* such that ``x == y[i]``, and all lower\ninteger indices do not raise ``IndexError`` exception.  (If any other\nexception is raised, it is as if ``in`` raised that exception).\n\nThe operator ``not in`` is defined to have the inverse true value of\n``in``.\n\nThe operators ``is`` and ``is not`` test for object identity: ``x is\ny`` is true if and only if *x* and *y* are the same object.  ``x is\nnot y`` yields the inverse truth value. [6]\n',
- 'compound': '\nCompound statements\n*******************\n\nCompound statements contain (groups of) other statements; they affect\nor control the execution of those other statements in some way.  In\ngeneral, compound statements span multiple lines, although in simple\nincarnations a whole compound statement may be contained in one line.\n\nThe ``if``, ``while`` and ``for`` statements implement traditional\ncontrol flow constructs.  ``try`` specifies exception handlers and/or\ncleanup code for a group of statements, while the ``with`` statement\nallows the execution of initialization and finalization code around a\nblock of code.  Function and class definitions are also syntactically\ncompound statements.\n\nCompound statements consist of one or more \'clauses.\'  A clause\nconsists of a header and a \'suite.\'  The clause headers of a\nparticular compound statement are all at the same indentation level.\nEach clause header begins with a uniquely identifying keyword and ends\nwith a colon.  A suite is a group of statements controlled by a\nclause.  A suite can be one or more semicolon-separated simple\nstatements on the same line as the header, following the header\'s\ncolon, or it can be one or more indented statements on subsequent\nlines.  Only the latter form of suite can contain nested compound\nstatements; the following is illegal, mostly because it wouldn\'t be\nclear to which ``if`` clause a following ``else`` clause would belong:\n\n   if test1: if test2: print(x)\n\nAlso note that the semicolon binds tighter than the colon in this\ncontext, so that in the following example, either all or none of the\n``print()`` calls are executed:\n\n   if x < y < z: print(x); print(y); print(z)\n\nSummarizing:\n\n   compound_stmt ::= if_stmt\n                     | while_stmt\n                     | for_stmt\n                     | try_stmt\n                     | with_stmt\n                     | funcdef\n                     | classdef\n   suite         ::= stmt_list NEWLINE | NEWLINE INDENT statement+ DEDENT\n   statement     ::= stmt_list NEWLINE | compound_stmt\n   stmt_list     ::= simple_stmt (";" simple_stmt)* [";"]\n\nNote that statements always end in a ``NEWLINE`` possibly followed by\na ``DEDENT``.  Also note that optional continuation clauses always\nbegin with a keyword that cannot start a statement, thus there are no\nambiguities (the \'dangling ``else``\' problem is solved in Python by\nrequiring nested ``if`` statements to be indented).\n\nThe formatting of the grammar rules in the following sections places\neach clause on a separate line for clarity.\n\n\nThe ``if`` statement\n====================\n\nThe ``if`` statement is used for conditional execution:\n\n   if_stmt ::= "if" expression ":" suite\n               ( "elif" expression ":" suite )*\n               ["else" ":" suite]\n\nIt selects exactly one of the suites by evaluating the expressions one\nby one until one is found to be true (see section *Boolean operations*\nfor the definition of true and false); then that suite is executed\n(and no other part of the ``if`` statement is executed or evaluated).\nIf all expressions are false, the suite of the ``else`` clause, if\npresent, is executed.\n\n\nThe ``while`` statement\n=======================\n\nThe ``while`` statement is used for repeated execution as long as an\nexpression is true:\n\n   while_stmt ::= "while" expression ":" suite\n                  ["else" ":" suite]\n\nThis repeatedly tests the expression and, if it is true, executes the\nfirst suite; if the expression is false (which may be the first time\nit is tested) the suite of the ``else`` clause, if present, is\nexecuted and the loop terminates.\n\nA ``break`` statement executed in the first suite terminates the loop\nwithout executing the ``else`` clause\'s suite.  A ``continue``\nstatement executed in the first suite skips the rest of the suite and\ngoes back to testing the expression.\n\n\nThe ``for`` statement\n=====================\n\nThe ``for`` statement is used to iterate over the elements of a\nsequence (such as a string, tuple or list) or other iterable object:\n\n   for_stmt ::= "for" target_list "in" expression_list ":" suite\n                ["else" ":" suite]\n\nThe expression list is evaluated once; it should yield an iterable\nobject.  An iterator is created for the result of the\n``expression_list``.  The suite is then executed once for each item\nprovided by the iterator, in the order of ascending indices.  Each\nitem in turn is assigned to the target list using the standard rules\nfor assignments (see *Assignment statements*), and then the suite is\nexecuted.  When the items are exhausted (which is immediately when the\nsequence is empty or an iterator raises a ``StopIteration``\nexception), the suite in the ``else`` clause, if present, is executed,\nand the loop terminates.\n\nA ``break`` statement executed in the first suite terminates the loop\nwithout executing the ``else`` clause\'s suite.  A ``continue``\nstatement executed in the first suite skips the rest of the suite and\ncontinues with the next item, or with the ``else`` clause if there was\nno next item.\n\nThe suite may assign to the variable(s) in the target list; this does\nnot affect the next item assigned to it.\n\nNames in the target list are not deleted when the loop is finished,\nbut if the sequence is empty, it will not have been assigned to at all\nby the loop.  Hint: the built-in function ``range()`` returns an\niterator of integers suitable to emulate the effect of Pascal\'s ``for\ni := a to b do``; e.g., ``list(range(3))`` returns the list ``[0, 1,\n2]``.\n\nNote: There is a subtlety when the sequence is being modified by the loop\n  (this can only occur for mutable sequences, i.e. lists).  An\n  internal counter is used to keep track of which item is used next,\n  and this is incremented on each iteration.  When this counter has\n  reached the length of the sequence the loop terminates.  This means\n  that if the suite deletes the current (or a previous) item from the\n  sequence, the next item will be skipped (since it gets the index of\n  the current item which has already been treated).  Likewise, if the\n  suite inserts an item in the sequence before the current item, the\n  current item will be treated again the next time through the loop.\n  This can lead to nasty bugs that can be avoided by making a\n  temporary copy using a slice of the whole sequence, e.g.,\n\n     for x in a[:]:\n         if x < 0: a.remove(x)\n\n\nThe ``try`` statement\n=====================\n\nThe ``try`` statement specifies exception handlers and/or cleanup code\nfor a group of statements:\n\n   try_stmt  ::= try1_stmt | try2_stmt\n   try1_stmt ::= "try" ":" suite\n                 ("except" [expression ["as" target]] ":" suite)+\n                 ["else" ":" suite]\n                 ["finally" ":" suite]\n   try2_stmt ::= "try" ":" suite\n                 "finally" ":" suite\n\nThe ``except`` clause(s) specify one or more exception handlers. When\nno exception occurs in the ``try`` clause, no exception handler is\nexecuted. When an exception occurs in the ``try`` suite, a search for\nan exception handler is started.  This search inspects the except\nclauses in turn until one is found that matches the exception.  An\nexpression-less except clause, if present, must be last; it matches\nany exception.  For an except clause with an expression, that\nexpression is evaluated, and the clause matches the exception if the\nresulting object is "compatible" with the exception.  An object is\ncompatible with an exception if it is the class or a base class of the\nexception object or a tuple containing an item compatible with the\nexception.\n\nIf no except clause matches the exception, the search for an exception\nhandler continues in the surrounding code and on the invocation stack.\n[1]\n\nIf the evaluation of an expression in the header of an except clause\nraises an exception, the original search for a handler is canceled and\na search starts for the new exception in the surrounding code and on\nthe call stack (it is treated as if the entire ``try`` statement\nraised the exception).\n\nWhen a matching except clause is found, the exception is assigned to\nthe target specified after the ``as`` keyword in that except clause,\nif present, and the except clause\'s suite is executed.  All except\nclauses must have an executable block.  When the end of this block is\nreached, execution continues normally after the entire try statement.\n(This means that if two nested handlers exist for the same exception,\nand the exception occurs in the try clause of the inner handler, the\nouter handler will not handle the exception.)\n\nWhen an exception has been assigned using ``as target``, it is cleared\nat the end of the except clause.  This is as if\n\n   except E as N:\n       foo\n\nwas translated to\n\n   except E as N:\n       try:\n           foo\n       finally:\n           N = None\n           del N\n\nThat means that you have to assign the exception to a different name\nif you want to be able to refer to it after the except clause.  The\nreason for this is that with the traceback attached to them,\nexceptions will form a reference cycle with the stack frame, keeping\nall locals in that frame alive until the next garbage collection\noccurs.\n\nBefore an except clause\'s suite is executed, details about the\nexception are stored in the ``sys`` module and can be access via\n``sys.exc_info()``. ``sys.exc_info()`` returns a 3-tuple consisting\nof: ``exc_type``, the exception class; ``exc_value``, the exception\ninstance; ``exc_traceback``, a traceback object (see section *The\nstandard type hierarchy*) identifying the point in the program where\nthe exception occurred. ``sys.exc_info()`` values are restored to\ntheir previous values (before the call) when returning from a function\nthat handled an exception.\n\nThe optional ``else`` clause is executed if and when control flows off\nthe end of the ``try`` clause. [2] Exceptions in the ``else`` clause\nare not handled by the preceding ``except`` clauses.\n\nIf ``finally`` is present, it specifies a \'cleanup\' handler.  The\n``try`` clause is executed, including any ``except`` and ``else``\nclauses.  If an exception occurs in any of the clauses and is not\nhandled, the exception is temporarily saved. The ``finally`` clause is\nexecuted.  If there is a saved exception, it is re-raised at the end\nof the ``finally`` clause. If the ``finally`` clause raises another\nexception or executes a ``return`` or ``break`` statement, the saved\nexception is lost.  The exception information is not available to the\nprogram during execution of the ``finally`` clause.\n\nWhen a ``return``, ``break`` or ``continue`` statement is executed in\nthe ``try`` suite of a ``try``...``finally`` statement, the\n``finally`` clause is also executed \'on the way out.\' A ``continue``\nstatement is illegal in the ``finally`` clause. (The reason is a\nproblem with the current implementation --- this restriction may be\nlifted in the future).\n\nAdditional information on exceptions can be found in section\n*Exceptions*, and information on using the ``raise`` statement to\ngenerate exceptions may be found in section *The raise statement*.\n\n\nThe ``with`` statement\n======================\n\nThe ``with`` statement is used to wrap the execution of a block with\nmethods defined by a context manager (see section *With Statement\nContext Managers*). This allows common\n``try``...``except``...``finally`` usage patterns to be encapsulated\nfor convenient reuse.\n\n   with_stmt ::= "with" with_item ("," with_item)* ":" suite\n   with_item ::= expression ["as" target]\n\nThe execution of the ``with`` statement with one "item" proceeds as\nfollows:\n\n1. The context expression is evaluated to obtain a context manager.\n\n2. The context manager\'s ``__enter__()`` method is invoked.\n\n3. If a target was included in the ``with`` statement, the return\n   value from ``__enter__()`` is assigned to it.\n\n   Note: The ``with`` statement guarantees that if the ``__enter__()``\n     method returns without an error, then ``__exit__()`` will always\n     be called.  Thus, if an error occurs during the assignment to the\n     target list, it will be treated the same as an error occurring\n     within the suite would be.  See step 5 below.\n\n4. The suite is executed.\n\n5. The context manager\'s ``__exit__()`` method is invoked.  If an\n   exception caused the suite to be exited, its type, value, and\n   traceback are passed as arguments to ``__exit__()``. Otherwise,\n   three ``None`` arguments are supplied.\n\n   If the suite was exited due to an exception, and the return value\n   from the ``__exit__()`` method was false, the exception is\n   reraised.  If the return value was true, the exception is\n   suppressed, and execution continues with the statement following\n   the ``with`` statement.\n\n   If the suite was exited for any reason other than an exception, the\n   return value from ``__exit__()`` is ignored, and execution proceeds\n   at the normal location for the kind of exit that was taken.\n\nWith more than one item, the context managers are processed as if\nmultiple ``with`` statements were nested:\n\n   with A() as a, B() as b:\n       suite\n\nis equivalent to\n\n   with A() as a:\n       with B() as b:\n           suite\n\nChanged in version 3.1: Support for multiple context expressions.\n\nSee also:\n\n   **PEP 0343** - The "with" statement\n      The specification, background, and examples for the Python\n      ``with`` statement.\n\n\nFunction definitions\n====================\n\nA function definition defines a user-defined function object (see\nsection *The standard type hierarchy*):\n\n   funcdef        ::= [decorators] "def" funcname "(" [parameter_list] ")" ["->" expression] ":" suite\n   decorators     ::= decorator+\n   decorator      ::= "@" dotted_name ["(" [argument_list [","]] ")"] NEWLINE\n   dotted_name    ::= identifier ("." identifier)*\n   parameter_list ::= (defparameter ",")*\n                      (  "*" [parameter] ("," defparameter)*\n                      [, "**" parameter]\n                      | "**" parameter\n                      | defparameter [","] )\n   parameter      ::= identifier [":" expression]\n   defparameter   ::= parameter ["=" expression]\n   funcname       ::= identifier\n\nA function definition is an executable statement.  Its execution binds\nthe function name in the current local namespace to a function object\n(a wrapper around the executable code for the function).  This\nfunction object contains a reference to the current global namespace\nas the global namespace to be used when the function is called.\n\nThe function definition does not execute the function body; this gets\nexecuted only when the function is called. [3]\n\nA function definition may be wrapped by one or more *decorator*\nexpressions. Decorator expressions are evaluated when the function is\ndefined, in the scope that contains the function definition.  The\nresult must be a callable, which is invoked with the function object\nas the only argument. The returned value is bound to the function name\ninstead of the function object.  Multiple decorators are applied in\nnested fashion. For example, the following code\n\n   @f1(arg)\n   @f2\n   def func(): pass\n\nis equivalent to\n\n   def func(): pass\n   func = f1(arg)(f2(func))\n\nWhen one or more parameters have the form *parameter* ``=``\n*expression*, the function is said to have "default parameter values."\nFor a parameter with a default value, the corresponding argument may\nbe omitted from a call, in which case the parameter\'s default value is\nsubstituted.  If a parameter has a default value, all following\nparameters up until the "``*``" must also have a default value ---\nthis is a syntactic restriction that is not expressed by the grammar.\n\n**Default parameter values are evaluated when the function definition\nis executed.** This means that the expression is evaluated once, when\nthe function is defined, and that that same "pre-computed" value is\nused for each call.  This is especially important to understand when a\ndefault parameter is a mutable object, such as a list or a dictionary:\nif the function modifies the object (e.g. by appending an item to a\nlist), the default value is in effect modified. This is generally not\nwhat was intended.  A way around this is to use ``None`` as the\ndefault, and explicitly test for it in the body of the function, e.g.:\n\n   def whats_on_the_telly(penguin=None):\n       if penguin is None:\n           penguin = []\n       penguin.append("property of the zoo")\n       return penguin\n\nFunction call semantics are described in more detail in section\n*Calls*. A function call always assigns values to all parameters\nmentioned in the parameter list, either from position arguments, from\nkeyword arguments, or from default values.  If the form\n"``*identifier``" is present, it is initialized to a tuple receiving\nany excess positional parameters, defaulting to the empty tuple.  If\nthe form "``**identifier``" is present, it is initialized to a new\ndictionary receiving any excess keyword arguments, defaulting to a new\nempty dictionary. Parameters after "``*``" or "``*identifier``" are\nkeyword-only parameters and may only be passed used keyword arguments.\n\nParameters may have annotations of the form "``: expression``"\nfollowing the parameter name.  Any parameter may have an annotation\neven those of the form ``*identifier`` or ``**identifier``.  Functions\nmay have "return" annotation of the form "``-> expression``" after the\nparameter list.  These annotations can be any valid Python expression\nand are evaluated when the function definition is executed.\nAnnotations may be evaluated in a different order than they appear in\nthe source code.  The presence of annotations does not change the\nsemantics of a function.  The annotation values are available as\nvalues of a dictionary keyed by the parameters\' names in the\n``__annotations__`` attribute of the function object.\n\nIt is also possible to create anonymous functions (functions not bound\nto a name), for immediate use in expressions.  This uses lambda forms,\ndescribed in section *Lambdas*.  Note that the lambda form is merely a\nshorthand for a simplified function definition; a function defined in\na "``def``" statement can be passed around or assigned to another name\njust like a function defined by a lambda form.  The "``def``" form is\nactually more powerful since it allows the execution of multiple\nstatements and annotations.\n\n**Programmer\'s note:** Functions are first-class objects.  A "``def``"\nform executed inside a function definition defines a local function\nthat can be returned or passed around.  Free variables used in the\nnested function can access the local variables of the function\ncontaining the def.  See section *Naming and binding* for details.\n\n\nClass definitions\n=================\n\nA class definition defines a class object (see section *The standard\ntype hierarchy*):\n\n   classdef    ::= [decorators] "class" classname [inheritance] ":" suite\n   inheritance ::= "(" [expression_list] ")"\n   classname   ::= identifier\n\nA class definition is an executable statement.  It first evaluates the\ninheritance list, if present.  Each item in the inheritance list\nshould evaluate to a class object or class type which allows\nsubclassing.  The class\'s suite is then executed in a new execution\nframe (see section *Naming and binding*), using a newly created local\nnamespace and the original global namespace. (Usually, the suite\ncontains only function definitions.)  When the class\'s suite finishes\nexecution, its execution frame is discarded but its local namespace is\nsaved. [4] A class object is then created using the inheritance list\nfor the base classes and the saved local namespace for the attribute\ndictionary.  The class name is bound to this class object in the\noriginal local namespace.\n\nClasses can also be decorated; as with functions,\n\n   @f1(arg)\n   @f2\n   class Foo: pass\n\nis equivalent to\n\n   class Foo: pass\n   Foo = f1(arg)(f2(Foo))\n\n**Programmer\'s note:** Variables defined in the class definition are\nclass variables; they are shared by instances. Instance variables can\nbe set in a method with ``self.name = value``.  Both class and\ninstance variables are accessible through the notation\n"``self.name``", and an instance variable hides a class variable with\nthe same name when accessed in this way.  Class variables can be used\nas defaults for instance variables, but using mutable values there can\nlead to unexpected results.  Descriptors can be used to create\ninstance variables with different implementation details.\n\nSee also:\n\n   **PEP 3129** - Class Decorators\n\nClass definitions, like function definitions, may be wrapped by one or\nmore *decorator* expressions.  The evaluation rules for the decorator\nexpressions are the same as for functions.  The result must be a class\nobject, which is then bound to the class name.\n\n-[ Footnotes ]-\n\n[1] The exception is propagated to the invocation stack only if there\n    is no ``finally`` clause that negates the exception.\n\n[2] Currently, control "flows off the end" except in the case of an\n    exception or the execution of a ``return``, ``continue``, or\n    ``break`` statement.\n\n[3] A string literal appearing as the first statement in the function\n    body is transformed into the function\'s ``__doc__`` attribute and\n    therefore the function\'s *docstring*.\n\n[4] A string literal appearing as the first statement in the class\n    body is transformed into the namespace\'s ``__doc__`` item and\n    therefore the class\'s *docstring*.\n',
+ 'comparisons': '\nComparisons\n***********\n\nUnlike C, all comparison operations in Python have the same priority,\nwhich is lower than that of any arithmetic, shifting or bitwise\noperation.  Also unlike C, expressions like ``a < b < c`` have the\ninterpretation that is conventional in mathematics:\n\n   comparison    ::= or_expr ( comp_operator or_expr )*\n   comp_operator ::= "<" | ">" | "==" | ">=" | "<=" | "!="\n                     | "is" ["not"] | ["not"] "in"\n\nComparisons yield boolean values: ``True`` or ``False``.\n\nComparisons can be chained arbitrarily, e.g., ``x < y <= z`` is\nequivalent to ``x < y and y <= z``, except that ``y`` is evaluated\nonly once (but in both cases ``z`` is not evaluated at all when ``x <\ny`` is found to be false).\n\nFormally, if *a*, *b*, *c*, ..., *y*, *z* are expressions and *op1*,\n*op2*, ..., *opN* are comparison operators, then ``a op1 b op2 c ... y\nopN z`` is equivalent to ``a op1 b and b op2 c and ... y opN z``,\nexcept that each expression is evaluated at most once.\n\nNote that ``a op1 b op2 c`` doesn\'t imply any kind of comparison\nbetween *a* and *c*, so that, e.g., ``x < y > z`` is perfectly legal\n(though perhaps not pretty).\n\nThe operators ``<``, ``>``, ``==``, ``>=``, ``<=``, and ``!=`` compare\nthe values of two objects.  The objects need not have the same type.\nIf both are numbers, they are converted to a common type.  Otherwise,\nthe ``==`` and ``!=`` operators *always* consider objects of different\ntypes to be unequal, while the ``<``, ``>``, ``>=`` and ``<=``\noperators raise a ``TypeError`` when comparing objects of different\ntypes that do not implement these operators for the given pair of\ntypes.  You can control comparison behavior of objects of non-built-in\ntypes by defining rich comparison methods like ``__gt__()``, described\nin section *Basic customization*.\n\nComparison of objects of the same type depends on the type:\n\n* Numbers are compared arithmetically.\n\n* The values ``float(\'NaN\')`` and ``Decimal(\'NaN\')`` are special. The\n  are identical to themselves, ``x is x`` but are not equal to\n  themselves, ``x != x``.  Additionally, comparing any value to a\n  not-a-number value will return ``False``.  For example, both ``3 <\n  float(\'NaN\')`` and ``float(\'NaN\') < 3`` will return ``False``.\n\n* Bytes objects are compared lexicographically using the numeric\n  values of their elements.\n\n* Strings are compared lexicographically using the numeric equivalents\n  (the result of the built-in function ``ord()``) of their characters.\n  [3] String and bytes object can\'t be compared!\n\n* Tuples and lists are compared lexicographically using comparison of\n  corresponding elements.  This means that to compare equal, each\n  element must compare equal and the two sequences must be of the same\n  type and have the same length.\n\n  If not equal, the sequences are ordered the same as their first\n  differing elements.  For example, ``[1,2,x] <= [1,2,y]`` has the\n  same value as ``x <= y``.  If the corresponding element does not\n  exist, the shorter sequence is ordered first (for example, ``[1,2] <\n  [1,2,3]``).\n\n* Mappings (dictionaries) compare equal if and only if they have the\n  same ``(key, value)`` pairs. Order comparisons ``(\'<\', \'<=\', \'>=\',\n  \'>\')`` raise ``TypeError``.\n\n* Sets and frozensets define comparison operators to mean subset and\n  superset tests.  Those relations do not define total orderings (the\n  two sets ``{1,2}`` and {2,3} are not equal, nor subsets of one\n  another, nor supersets of one another).  Accordingly, sets are not\n  appropriate arguments for functions which depend on total ordering.\n  For example, ``min()``, ``max()``, and ``sorted()`` produce\n  undefined results given a list of sets as inputs.\n\n* Most other objects of built-in types compare unequal unless they are\n  the same object; the choice whether one object is considered smaller\n  or larger than another one is made arbitrarily but consistently\n  within one execution of a program.\n\nComparison of objects of the differing types depends on whether either\nof the types provide explicit support for the comparison.  Most\nnumeric types can be compared with one another, but comparisons of\n``float`` and ``Decimal`` are not supported to avoid the inevitable\nconfusion arising from representation issues such as ``float(\'1.1\')``\nbeing inexactly represented and therefore not exactly equal to\n``Decimal(\'1.1\')`` which is.  When cross-type comparison is not\nsupported, the comparison method returns ``NotImplemented``.  This can\ncreate the illusion of non-transitivity between supported cross-type\ncomparisons and unsupported comparisons.  For example, ``Decimal(2) ==\n2`` and *2 == float(2)`* but ``Decimal(2) != float(2)``.\n\nThe operators ``in`` and ``not in`` test for membership.  ``x in s``\nevaluates to true if *x* is a member of *s*, and false otherwise.  ``x\nnot in s`` returns the negation of ``x in s``.  All built-in sequences\nand set types support this as well as dictionary, for which ``in``\ntests whether a the dictionary has a given key. For container types\nsuch as list, tuple, set, frozenset, dict, or collections.deque, the\nexpression ``x in y`` is equivalent to ``any(x is e or x == e for e in\ny)``.\n\nFor the string and bytes types, ``x in y`` is true if and only if *x*\nis a substring of *y*.  An equivalent test is ``y.find(x) != -1``.\nEmpty strings are always considered to be a substring of any other\nstring, so ``"" in "abc"`` will return ``True``.\n\nFor user-defined classes which define the ``__contains__()`` method,\n``x in y`` is true if and only if ``y.__contains__(x)`` is true.\n\nFor user-defined classes which do not define ``__contains__()`` but do\ndefine ``__iter__()``, ``x in y`` is true if some value ``z`` with ``x\n== z`` is produced while iterating over ``y``.  If an exception is\nraised during the iteration, it is as if ``in`` raised that exception.\n\nLastly, the old-style iteration protocol is tried: if a class defines\n``__getitem__()``, ``x in y`` is true if and only if there is a non-\nnegative integer index *i* such that ``x == y[i]``, and all lower\ninteger indices do not raise ``IndexError`` exception.  (If any other\nexception is raised, it is as if ``in`` raised that exception).\n\nThe operator ``not in`` is defined to have the inverse true value of\n``in``.\n\nThe operators ``is`` and ``is not`` test for object identity: ``x is\ny`` is true if and only if *x* and *y* are the same object.  ``x is\nnot y`` yields the inverse truth value. [4]\n',
+ 'compound': '\nCompound statements\n*******************\n\nCompound statements contain (groups of) other statements; they affect\nor control the execution of those other statements in some way.  In\ngeneral, compound statements span multiple lines, although in simple\nincarnations a whole compound statement may be contained in one line.\n\nThe ``if``, ``while`` and ``for`` statements implement traditional\ncontrol flow constructs.  ``try`` specifies exception handlers and/or\ncleanup code for a group of statements, while the ``with`` statement\nallows the execution of initialization and finalization code around a\nblock of code.  Function and class definitions are also syntactically\ncompound statements.\n\nCompound statements consist of one or more \'clauses.\'  A clause\nconsists of a header and a \'suite.\'  The clause headers of a\nparticular compound statement are all at the same indentation level.\nEach clause header begins with a uniquely identifying keyword and ends\nwith a colon.  A suite is a group of statements controlled by a\nclause.  A suite can be one or more semicolon-separated simple\nstatements on the same line as the header, following the header\'s\ncolon, or it can be one or more indented statements on subsequent\nlines.  Only the latter form of suite can contain nested compound\nstatements; the following is illegal, mostly because it wouldn\'t be\nclear to which ``if`` clause a following ``else`` clause would belong:\n\n   if test1: if test2: print(x)\n\nAlso note that the semicolon binds tighter than the colon in this\ncontext, so that in the following example, either all or none of the\n``print()`` calls are executed:\n\n   if x < y < z: print(x); print(y); print(z)\n\nSummarizing:\n\n   compound_stmt ::= if_stmt\n                     | while_stmt\n                     | for_stmt\n                     | try_stmt\n                     | with_stmt\n                     | funcdef\n                     | classdef\n   suite         ::= stmt_list NEWLINE | NEWLINE INDENT statement+ DEDENT\n   statement     ::= stmt_list NEWLINE | compound_stmt\n   stmt_list     ::= simple_stmt (";" simple_stmt)* [";"]\n\nNote that statements always end in a ``NEWLINE`` possibly followed by\na ``DEDENT``.  Also note that optional continuation clauses always\nbegin with a keyword that cannot start a statement, thus there are no\nambiguities (the \'dangling ``else``\' problem is solved in Python by\nrequiring nested ``if`` statements to be indented).\n\nThe formatting of the grammar rules in the following sections places\neach clause on a separate line for clarity.\n\n\nThe ``if`` statement\n====================\n\nThe ``if`` statement is used for conditional execution:\n\n   if_stmt ::= "if" expression ":" suite\n               ( "elif" expression ":" suite )*\n               ["else" ":" suite]\n\nIt selects exactly one of the suites by evaluating the expressions one\nby one until one is found to be true (see section *Boolean operations*\nfor the definition of true and false); then that suite is executed\n(and no other part of the ``if`` statement is executed or evaluated).\nIf all expressions are false, the suite of the ``else`` clause, if\npresent, is executed.\n\n\nThe ``while`` statement\n=======================\n\nThe ``while`` statement is used for repeated execution as long as an\nexpression is true:\n\n   while_stmt ::= "while" expression ":" suite\n                  ["else" ":" suite]\n\nThis repeatedly tests the expression and, if it is true, executes the\nfirst suite; if the expression is false (which may be the first time\nit is tested) the suite of the ``else`` clause, if present, is\nexecuted and the loop terminates.\n\nA ``break`` statement executed in the first suite terminates the loop\nwithout executing the ``else`` clause\'s suite.  A ``continue``\nstatement executed in the first suite skips the rest of the suite and\ngoes back to testing the expression.\n\n\nThe ``for`` statement\n=====================\n\nThe ``for`` statement is used to iterate over the elements of a\nsequence (such as a string, tuple or list) or other iterable object:\n\n   for_stmt ::= "for" target_list "in" expression_list ":" suite\n                ["else" ":" suite]\n\nThe expression list is evaluated once; it should yield an iterable\nobject.  An iterator is created for the result of the\n``expression_list``.  The suite is then executed once for each item\nprovided by the iterator, in the order of ascending indices.  Each\nitem in turn is assigned to the target list using the standard rules\nfor assignments (see *Assignment statements*), and then the suite is\nexecuted.  When the items are exhausted (which is immediately when the\nsequence is empty or an iterator raises a ``StopIteration``\nexception), the suite in the ``else`` clause, if present, is executed,\nand the loop terminates.\n\nA ``break`` statement executed in the first suite terminates the loop\nwithout executing the ``else`` clause\'s suite.  A ``continue``\nstatement executed in the first suite skips the rest of the suite and\ncontinues with the next item, or with the ``else`` clause if there was\nno next item.\n\nThe suite may assign to the variable(s) in the target list; this does\nnot affect the next item assigned to it.\n\nNames in the target list are not deleted when the loop is finished,\nbut if the sequence is empty, it will not have been assigned to at all\nby the loop.  Hint: the built-in function ``range()`` returns an\niterator of integers suitable to emulate the effect of Pascal\'s ``for\ni := a to b do``; e.g., ``list(range(3))`` returns the list ``[0, 1,\n2]``.\n\nNote: There is a subtlety when the sequence is being modified by the loop\n  (this can only occur for mutable sequences, i.e. lists).  An\n  internal counter is used to keep track of which item is used next,\n  and this is incremented on each iteration.  When this counter has\n  reached the length of the sequence the loop terminates.  This means\n  that if the suite deletes the current (or a previous) item from the\n  sequence, the next item will be skipped (since it gets the index of\n  the current item which has already been treated).  Likewise, if the\n  suite inserts an item in the sequence before the current item, the\n  current item will be treated again the next time through the loop.\n  This can lead to nasty bugs that can be avoided by making a\n  temporary copy using a slice of the whole sequence, e.g.,\n\n     for x in a[:]:\n         if x < 0: a.remove(x)\n\n\nThe ``try`` statement\n=====================\n\nThe ``try`` statement specifies exception handlers and/or cleanup code\nfor a group of statements:\n\n   try_stmt  ::= try1_stmt | try2_stmt\n   try1_stmt ::= "try" ":" suite\n                 ("except" [expression ["as" target]] ":" suite)+\n                 ["else" ":" suite]\n                 ["finally" ":" suite]\n   try2_stmt ::= "try" ":" suite\n                 "finally" ":" suite\n\nThe ``except`` clause(s) specify one or more exception handlers. When\nno exception occurs in the ``try`` clause, no exception handler is\nexecuted. When an exception occurs in the ``try`` suite, a search for\nan exception handler is started.  This search inspects the except\nclauses in turn until one is found that matches the exception.  An\nexpression-less except clause, if present, must be last; it matches\nany exception.  For an except clause with an expression, that\nexpression is evaluated, and the clause matches the exception if the\nresulting object is "compatible" with the exception.  An object is\ncompatible with an exception if it is the class or a base class of the\nexception object or a tuple containing an item compatible with the\nexception.\n\nIf no except clause matches the exception, the search for an exception\nhandler continues in the surrounding code and on the invocation stack.\n[1]\n\nIf the evaluation of an expression in the header of an except clause\nraises an exception, the original search for a handler is canceled and\na search starts for the new exception in the surrounding code and on\nthe call stack (it is treated as if the entire ``try`` statement\nraised the exception).\n\nWhen a matching except clause is found, the exception is assigned to\nthe target specified after the ``as`` keyword in that except clause,\nif present, and the except clause\'s suite is executed.  All except\nclauses must have an executable block.  When the end of this block is\nreached, execution continues normally after the entire try statement.\n(This means that if two nested handlers exist for the same exception,\nand the exception occurs in the try clause of the inner handler, the\nouter handler will not handle the exception.)\n\nWhen an exception has been assigned using ``as target``, it is cleared\nat the end of the except clause.  This is as if\n\n   except E as N:\n       foo\n\nwas translated to\n\n   except E as N:\n       try:\n           foo\n       finally:\n           del N\n\nThis means the exception must be assigned to a different name to be\nable to refer to it after the except clause.  Exceptions are cleared\nbecause with the traceback attached to them, they form a reference\ncycle with the stack frame, keeping all locals in that frame alive\nuntil the next garbage collection occurs.\n\nBefore an except clause\'s suite is executed, details about the\nexception are stored in the ``sys`` module and can be access via\n``sys.exc_info()``. ``sys.exc_info()`` returns a 3-tuple consisting\nof: ``exc_type``, the exception class; ``exc_value``, the exception\ninstance; ``exc_traceback``, a traceback object (see section *The\nstandard type hierarchy*) identifying the point in the program where\nthe exception occurred. ``sys.exc_info()`` values are restored to\ntheir previous values (before the call) when returning from a function\nthat handled an exception.\n\nThe optional ``else`` clause is executed if and when control flows off\nthe end of the ``try`` clause. [2] Exceptions in the ``else`` clause\nare not handled by the preceding ``except`` clauses.\n\nIf ``finally`` is present, it specifies a \'cleanup\' handler.  The\n``try`` clause is executed, including any ``except`` and ``else``\nclauses.  If an exception occurs in any of the clauses and is not\nhandled, the exception is temporarily saved. The ``finally`` clause is\nexecuted.  If there is a saved exception, it is re-raised at the end\nof the ``finally`` clause. If the ``finally`` clause raises another\nexception or executes a ``return`` or ``break`` statement, the saved\nexception is lost.  The exception information is not available to the\nprogram during execution of the ``finally`` clause.\n\nWhen a ``return``, ``break`` or ``continue`` statement is executed in\nthe ``try`` suite of a ``try``...``finally`` statement, the\n``finally`` clause is also executed \'on the way out.\' A ``continue``\nstatement is illegal in the ``finally`` clause. (The reason is a\nproblem with the current implementation --- this restriction may be\nlifted in the future).\n\nAdditional information on exceptions can be found in section\n*Exceptions*, and information on using the ``raise`` statement to\ngenerate exceptions may be found in section *The raise statement*.\n\n\nThe ``with`` statement\n======================\n\nThe ``with`` statement is used to wrap the execution of a block with\nmethods defined by a context manager (see section *With Statement\nContext Managers*). This allows common\n``try``...``except``...``finally`` usage patterns to be encapsulated\nfor convenient reuse.\n\n   with_stmt ::= "with" with_item ("," with_item)* ":" suite\n   with_item ::= expression ["as" target]\n\nThe execution of the ``with`` statement with one "item" proceeds as\nfollows:\n\n1. The context expression is evaluated to obtain a context manager.\n\n2. The context manager\'s ``__exit__()`` is loaded for later use.\n\n3. The context manager\'s ``__enter__()`` method is invoked.\n\n4. If a target was included in the ``with`` statement, the return\n   value from ``__enter__()`` is assigned to it.\n\n   Note: The ``with`` statement guarantees that if the ``__enter__()``\n     method returns without an error, then ``__exit__()`` will always\n     be called. Thus, if an error occurs during the assignment to the\n     target list, it will be treated the same as an error occurring\n     within the suite would be. See step 6 below.\n\n5. The suite is executed.\n\n6. The context manager\'s ``__exit__()`` method is invoked.  If an\n   exception caused the suite to be exited, its type, value, and\n   traceback are passed as arguments to ``__exit__()``. Otherwise,\n   three ``None`` arguments are supplied.\n\n   If the suite was exited due to an exception, and the return value\n   from the ``__exit__()`` method was false, the exception is\n   reraised.  If the return value was true, the exception is\n   suppressed, and execution continues with the statement following\n   the ``with`` statement.\n\n   If the suite was exited for any reason other than an exception, the\n   return value from ``__exit__()`` is ignored, and execution proceeds\n   at the normal location for the kind of exit that was taken.\n\nWith more than one item, the context managers are processed as if\nmultiple ``with`` statements were nested:\n\n   with A() as a, B() as b:\n       suite\n\nis equivalent to\n\n   with A() as a:\n       with B() as b:\n           suite\n\nChanged in version 3.1: Support for multiple context expressions.\n\nSee also:\n\n   **PEP 0343** - The "with" statement\n      The specification, background, and examples for the Python\n      ``with`` statement.\n\n\nFunction definitions\n====================\n\nA function definition defines a user-defined function object (see\nsection *The standard type hierarchy*):\n\n   funcdef        ::= [decorators] "def" funcname "(" [parameter_list] ")" ["->" expression] ":" suite\n   decorators     ::= decorator+\n   decorator      ::= "@" dotted_name ["(" [argument_list [","]] ")"] NEWLINE\n   dotted_name    ::= identifier ("." identifier)*\n   parameter_list ::= (defparameter ",")*\n                      (  "*" [parameter] ("," defparameter)*\n                      [, "**" parameter]\n                      | "**" parameter\n                      | defparameter [","] )\n   parameter      ::= identifier [":" expression]\n   defparameter   ::= parameter ["=" expression]\n   funcname       ::= identifier\n\nA function definition is an executable statement.  Its execution binds\nthe function name in the current local namespace to a function object\n(a wrapper around the executable code for the function).  This\nfunction object contains a reference to the current global namespace\nas the global namespace to be used when the function is called.\n\nThe function definition does not execute the function body; this gets\nexecuted only when the function is called. [3]\n\nA function definition may be wrapped by one or more *decorator*\nexpressions. Decorator expressions are evaluated when the function is\ndefined, in the scope that contains the function definition.  The\nresult must be a callable, which is invoked with the function object\nas the only argument. The returned value is bound to the function name\ninstead of the function object.  Multiple decorators are applied in\nnested fashion. For example, the following code\n\n   @f1(arg)\n   @f2\n   def func(): pass\n\nis equivalent to\n\n   def func(): pass\n   func = f1(arg)(f2(func))\n\nWhen one or more parameters have the form *parameter* ``=``\n*expression*, the function is said to have "default parameter values."\nFor a parameter with a default value, the corresponding argument may\nbe omitted from a call, in which case the parameter\'s default value is\nsubstituted.  If a parameter has a default value, all following\nparameters up until the "``*``" must also have a default value ---\nthis is a syntactic restriction that is not expressed by the grammar.\n\n**Default parameter values are evaluated when the function definition\nis executed.** This means that the expression is evaluated once, when\nthe function is defined, and that that same "pre-computed" value is\nused for each call.  This is especially important to understand when a\ndefault parameter is a mutable object, such as a list or a dictionary:\nif the function modifies the object (e.g. by appending an item to a\nlist), the default value is in effect modified. This is generally not\nwhat was intended.  A way around this is to use ``None`` as the\ndefault, and explicitly test for it in the body of the function, e.g.:\n\n   def whats_on_the_telly(penguin=None):\n       if penguin is None:\n           penguin = []\n       penguin.append("property of the zoo")\n       return penguin\n\nFunction call semantics are described in more detail in section\n*Calls*. A function call always assigns values to all parameters\nmentioned in the parameter list, either from position arguments, from\nkeyword arguments, or from default values.  If the form\n"``*identifier``" is present, it is initialized to a tuple receiving\nany excess positional parameters, defaulting to the empty tuple.  If\nthe form "``**identifier``" is present, it is initialized to a new\ndictionary receiving any excess keyword arguments, defaulting to a new\nempty dictionary. Parameters after "``*``" or "``*identifier``" are\nkeyword-only parameters and may only be passed used keyword arguments.\n\nParameters may have annotations of the form "``: expression``"\nfollowing the parameter name.  Any parameter may have an annotation\neven those of the form ``*identifier`` or ``**identifier``.  Functions\nmay have "return" annotation of the form "``-> expression``" after the\nparameter list.  These annotations can be any valid Python expression\nand are evaluated when the function definition is executed.\nAnnotations may be evaluated in a different order than they appear in\nthe source code.  The presence of annotations does not change the\nsemantics of a function.  The annotation values are available as\nvalues of a dictionary keyed by the parameters\' names in the\n``__annotations__`` attribute of the function object.\n\nIt is also possible to create anonymous functions (functions not bound\nto a name), for immediate use in expressions.  This uses lambda forms,\ndescribed in section *Lambdas*.  Note that the lambda form is merely a\nshorthand for a simplified function definition; a function defined in\na "``def``" statement can be passed around or assigned to another name\njust like a function defined by a lambda form.  The "``def``" form is\nactually more powerful since it allows the execution of multiple\nstatements and annotations.\n\n**Programmer\'s note:** Functions are first-class objects.  A "``def``"\nform executed inside a function definition defines a local function\nthat can be returned or passed around.  Free variables used in the\nnested function can access the local variables of the function\ncontaining the def.  See section *Naming and binding* for details.\n\n\nClass definitions\n=================\n\nA class definition defines a class object (see section *The standard\ntype hierarchy*):\n\n   classdef    ::= [decorators] "class" classname [inheritance] ":" suite\n   inheritance ::= "(" [expression_list] ")"\n   classname   ::= identifier\n\nA class definition is an executable statement.  It first evaluates the\ninheritance list, if present.  Each item in the inheritance list\nshould evaluate to a class object or class type which allows\nsubclassing.  The class\'s suite is then executed in a new execution\nframe (see section *Naming and binding*), using a newly created local\nnamespace and the original global namespace. (Usually, the suite\ncontains only function definitions.)  When the class\'s suite finishes\nexecution, its execution frame is discarded but its local namespace is\nsaved. [4] A class object is then created using the inheritance list\nfor the base classes and the saved local namespace for the attribute\ndictionary.  The class name is bound to this class object in the\noriginal local namespace.\n\nClasses can also be decorated; as with functions,\n\n   @f1(arg)\n   @f2\n   class Foo: pass\n\nis equivalent to\n\n   class Foo: pass\n   Foo = f1(arg)(f2(Foo))\n\n**Programmer\'s note:** Variables defined in the class definition are\nclass variables; they are shared by instances. Instance variables can\nbe set in a method with ``self.name = value``.  Both class and\ninstance variables are accessible through the notation\n"``self.name``", and an instance variable hides a class variable with\nthe same name when accessed in this way.  Class variables can be used\nas defaults for instance variables, but using mutable values there can\nlead to unexpected results.  Descriptors can be used to create\ninstance variables with different implementation details.\n\nSee also:\n\n   **PEP 3129** - Class Decorators\n\nClass definitions, like function definitions, may be wrapped by one or\nmore *decorator* expressions.  The evaluation rules for the decorator\nexpressions are the same as for functions.  The result must be a class\nobject, which is then bound to the class name.\n\n-[ Footnotes ]-\n\n[1] The exception is propagated to the invocation stack only if there\n    is no ``finally`` clause that negates the exception.\n\n[2] Currently, control "flows off the end" except in the case of an\n    exception or the execution of a ``return``, ``continue``, or\n    ``break`` statement.\n\n[3] A string literal appearing as the first statement in the function\n    body is transformed into the function\'s ``__doc__`` attribute and\n    therefore the function\'s *docstring*.\n\n[4] A string literal appearing as the first statement in the class\n    body is transformed into the namespace\'s ``__doc__`` item and\n    therefore the class\'s *docstring*.\n',
  'context-managers': '\nWith Statement Context Managers\n*******************************\n\nA *context manager* is an object that defines the runtime context to\nbe established when executing a ``with`` statement. The context\nmanager handles the entry into, and the exit from, the desired runtime\ncontext for the execution of the block of code.  Context managers are\nnormally invoked using the ``with`` statement (described in section\n*The with statement*), but can also be used by directly invoking their\nmethods.\n\nTypical uses of context managers include saving and restoring various\nkinds of global state, locking and unlocking resources, closing opened\nfiles, etc.\n\nFor more information on context managers, see *Context Manager Types*.\n\nobject.__enter__(self)\n\n   Enter the runtime context related to this object. The ``with``\n   statement will bind this method\'s return value to the target(s)\n   specified in the ``as`` clause of the statement, if any.\n\nobject.__exit__(self, exc_type, exc_value, traceback)\n\n   Exit the runtime context related to this object. The parameters\n   describe the exception that caused the context to be exited. If the\n   context was exited without an exception, all three arguments will\n   be ``None``.\n\n   If an exception is supplied, and the method wishes to suppress the\n   exception (i.e., prevent it from being propagated), it should\n   return a true value. Otherwise, the exception will be processed\n   normally upon exit from this method.\n\n   Note that ``__exit__()`` methods should not reraise the passed-in\n   exception; this is the caller\'s responsibility.\n\nSee also:\n\n   **PEP 0343** - The "with" statement\n      The specification, background, and examples for the Python\n      ``with`` statement.\n',
  'continue': '\nThe ``continue`` statement\n**************************\n\n   continue_stmt ::= "continue"\n\n``continue`` may only occur syntactically nested in a ``for`` or\n``while`` loop, but not nested in a function or class definition or\n``finally`` clause within that loop.  It continues with the next cycle\nof the nearest enclosing loop.\n\nWhen ``continue`` passes control out of a ``try`` statement with a\n``finally`` clause, that ``finally`` clause is executed before really\nstarting the next loop cycle.\n',
  'conversions': '\nArithmetic conversions\n**********************\n\nWhen a description of an arithmetic operator below uses the phrase\n"the numeric arguments are converted to a common type," this means\nthat the operator implementation for built-in types works that way:\n\n* If either argument is a complex number, the other is converted to\n  complex;\n\n* otherwise, if either argument is a floating point number, the other\n  is converted to floating point;\n\n* otherwise, both must be integers and no conversion is necessary.\n\nSome additional rules apply for certain operators (e.g., a string left\nargument to the \'%\' operator).  Extensions must define their own\nconversion behavior.\n',
- 'customization': '\nBasic customization\n*******************\n\nobject.__new__(cls[, ...])\n\n   Called to create a new instance of class *cls*.  ``__new__()`` is a\n   static method (special-cased so you need not declare it as such)\n   that takes the class of which an instance was requested as its\n   first argument.  The remaining arguments are those passed to the\n   object constructor expression (the call to the class).  The return\n   value of ``__new__()`` should be the new object instance (usually\n   an instance of *cls*).\n\n   Typical implementations create a new instance of the class by\n   invoking the superclass\'s ``__new__()`` method using\n   ``super(currentclass, cls).__new__(cls[, ...])`` with appropriate\n   arguments and then modifying the newly-created instance as\n   necessary before returning it.\n\n   If ``__new__()`` returns an instance of *cls*, then the new\n   instance\'s ``__init__()`` method will be invoked like\n   ``__init__(self[, ...])``, where *self* is the new instance and the\n   remaining arguments are the same as were passed to ``__new__()``.\n\n   If ``__new__()`` does not return an instance of *cls*, then the new\n   instance\'s ``__init__()`` method will not be invoked.\n\n   ``__new__()`` is intended mainly to allow subclasses of immutable\n   types (like int, str, or tuple) to customize instance creation.  It\n   is also commonly overridden in custom metaclasses in order to\n   customize class creation.\n\nobject.__init__(self[, ...])\n\n   Called when the instance is created.  The arguments are those\n   passed to the class constructor expression.  If a base class has an\n   ``__init__()`` method, the derived class\'s ``__init__()`` method,\n   if any, must explicitly call it to ensure proper initialization of\n   the base class part of the instance; for example:\n   ``BaseClass.__init__(self, [args...])``.  As a special constraint\n   on constructors, no value may be returned; doing so will cause a\n   ``TypeError`` to be raised at runtime.\n\nobject.__del__(self)\n\n   Called when the instance is about to be destroyed.  This is also\n   called a destructor.  If a base class has a ``__del__()`` method,\n   the derived class\'s ``__del__()`` method, if any, must explicitly\n   call it to ensure proper deletion of the base class part of the\n   instance.  Note that it is possible (though not recommended!) for\n   the ``__del__()`` method to postpone destruction of the instance by\n   creating a new reference to it.  It may then be called at a later\n   time when this new reference is deleted.  It is not guaranteed that\n   ``__del__()`` methods are called for objects that still exist when\n   the interpreter exits.\n\n   Note: ``del x`` doesn\'t directly call ``x.__del__()`` --- the former\n     decrements the reference count for ``x`` by one, and the latter\n     is only called when ``x``\'s reference count reaches zero.  Some\n     common situations that may prevent the reference count of an\n     object from going to zero include: circular references between\n     objects (e.g., a doubly-linked list or a tree data structure with\n     parent and child pointers); a reference to the object on the\n     stack frame of a function that caught an exception (the traceback\n     stored in ``sys.exc_info()[2]`` keeps the stack frame alive); or\n     a reference to the object on the stack frame that raised an\n     unhandled exception in interactive mode (the traceback stored in\n     ``sys.last_traceback`` keeps the stack frame alive).  The first\n     situation can only be remedied by explicitly breaking the cycles;\n     the latter two situations can be resolved by storing ``None`` in\n     ``sys.last_traceback``. Circular references which are garbage are\n     detected when the option cycle detector is enabled (it\'s on by\n     default), but can only be cleaned up if there are no Python-\n     level ``__del__()`` methods involved. Refer to the documentation\n     for the ``gc`` module for more information about how\n     ``__del__()`` methods are handled by the cycle detector,\n     particularly the description of the ``garbage`` value.\n\n   Warning: Due to the precarious circumstances under which ``__del__()``\n     methods are invoked, exceptions that occur during their execution\n     are ignored, and a warning is printed to ``sys.stderr`` instead.\n     Also, when ``__del__()`` is invoked in response to a module being\n     deleted (e.g., when execution of the program is done), other\n     globals referenced by the ``__del__()`` method may already have\n     been deleted or in the process of being torn down (e.g. the\n     import machinery shutting down).  For this reason, ``__del__()``\n     methods should do the absolute minimum needed to maintain\n     external invariants.  Starting with version 1.5, Python\n     guarantees that globals whose name begins with a single\n     underscore are deleted from their module before other globals are\n     deleted; if no other references to such globals exist, this may\n     help in assuring that imported modules are still available at the\n     time when the ``__del__()`` method is called.\n\nobject.__repr__(self)\n\n   Called by the ``repr()`` built-in function to compute the\n   "official" string representation of an object.  If at all possible,\n   this should look like a valid Python expression that could be used\n   to recreate an object with the same value (given an appropriate\n   environment).  If this is not possible, a string of the form\n   ``<...some useful description...>`` should be returned. The return\n   value must be a string object. If a class defines ``__repr__()``\n   but not ``__str__()``, then ``__repr__()`` is also used when an\n   "informal" string representation of instances of that class is\n   required.\n\n   This is typically used for debugging, so it is important that the\n   representation is information-rich and unambiguous.\n\nobject.__str__(self)\n\n   Called by the ``str()`` built-in function and by the ``print()``\n   function to compute the "informal" string representation of an\n   object.  This differs from ``__repr__()`` in that it does not have\n   to be a valid Python expression: a more convenient or concise\n   representation may be used instead. The return value must be a\n   string object.\n\nobject.__format__(self, format_spec)\n\n   Called by the ``format()`` built-in function (and by extension, the\n   ``format()`` method of class ``str``) to produce a "formatted"\n   string representation of an object. The ``format_spec`` argument is\n   a string that contains a description of the formatting options\n   desired. The interpretation of the ``format_spec`` argument is up\n   to the type implementing ``__format__()``, however most classes\n   will either delegate formatting to one of the built-in types, or\n   use a similar formatting option syntax.\n\n   See *Format Specification Mini-Language* for a description of the\n   standard formatting syntax.\n\n   The return value must be a string object.\n\nobject.__lt__(self, other)\nobject.__le__(self, other)\nobject.__eq__(self, other)\nobject.__ne__(self, other)\nobject.__gt__(self, other)\nobject.__ge__(self, other)\n\n   These are the so-called "rich comparison" methods. The\n   correspondence between operator symbols and method names is as\n   follows: ``x<y`` calls ``x.__lt__(y)``, ``x<=y`` calls\n   ``x.__le__(y)``, ``x==y`` calls ``x.__eq__(y)``, ``x!=y`` calls\n   ``x.__ne__(y)``, ``x>y`` calls ``x.__gt__(y)``, and ``x>=y`` calls\n   ``x.__ge__(y)``.\n\n   A rich comparison method may return the singleton\n   ``NotImplemented`` if it does not implement the operation for a\n   given pair of arguments. By convention, ``False`` and ``True`` are\n   returned for a successful comparison. However, these methods can\n   return any value, so if the comparison operator is used in a\n   Boolean context (e.g., in the condition of an ``if`` statement),\n   Python will call ``bool()`` on the value to determine if the result\n   is true or false.\n\n   There are no implied relationships among the comparison operators.\n   The truth of ``x==y`` does not imply that ``x!=y`` is false.\n   Accordingly, when defining ``__eq__()``, one should also define\n   ``__ne__()`` so that the operators will behave as expected.  See\n   the paragraph on ``__hash__()`` for some important notes on\n   creating *hashable* objects which support custom comparison\n   operations and are usable as dictionary keys.\n\n   There are no swapped-argument versions of these methods (to be used\n   when the left argument does not support the operation but the right\n   argument does); rather, ``__lt__()`` and ``__gt__()`` are each\n   other\'s reflection, ``__le__()`` and ``__ge__()`` are each other\'s\n   reflection, and ``__eq__()`` and ``__ne__()`` are their own\n   reflection.\n\n   Arguments to rich comparison methods are never coerced.\n\n   To automatically generate ordering operations from a single root\n   operation, see the Total Ordering recipe in the ASPN cookbook.\n\nobject.__hash__(self)\n\n   Called by built-in function ``hash()`` and for operations on\n   members of hashed collections including ``set``, ``frozenset``, and\n   ``dict``.  ``__hash__()`` should return an integer.  The only\n   required property is that objects which compare equal have the same\n   hash value; it is advised to somehow mix together (e.g. using\n   exclusive or) the hash values for the components of the object that\n   also play a part in comparison of objects.\n\n   If a class does not define an ``__eq__()`` method it should not\n   define a ``__hash__()`` operation either; if it defines\n   ``__eq__()`` but not ``__hash__()``, its instances will not be\n   usable as items in hashable collections.  If a class defines\n   mutable objects and implements an ``__eq__()`` method, it should\n   not implement ``__hash__()``, since the implementation of hashable\n   collections requires that a key\'s hash value is immutable (if the\n   object\'s hash value changes, it will be in the wrong hash bucket).\n\n   User-defined classes have ``__eq__()`` and ``__hash__()`` methods\n   by default; with them, all objects compare unequal (except with\n   themselves) and ``x.__hash__()`` returns ``id(x)``.\n\n   Classes which inherit a ``__hash__()`` method from a parent class\n   but change the meaning of ``__eq__()`` such that the hash value\n   returned is no longer appropriate (e.g. by switching to a value-\n   based concept of equality instead of the default identity based\n   equality) can explicitly flag themselves as being unhashable by\n   setting ``__hash__ = None`` in the class definition. Doing so means\n   that not only will instances of the class raise an appropriate\n   ``TypeError`` when a program attempts to retrieve their hash value,\n   but they will also be correctly identified as unhashable when\n   checking ``isinstance(obj, collections.Hashable)`` (unlike classes\n   which define their own ``__hash__()`` to explicitly raise\n   ``TypeError``).\n\n   If a class that overrides ``__eq__()`` needs to retain the\n   implementation of ``__hash__()`` from a parent class, the\n   interpreter must be told this explicitly by setting ``__hash__ =\n   <ParentClass>.__hash__``. Otherwise the inheritance of\n   ``__hash__()`` will be blocked, just as if ``__hash__`` had been\n   explicitly set to ``None``.\n\nobject.__bool__(self)\n\n   Called to implement truth value testing and the built-in operation\n   ``bool()``; should return ``False`` or ``True``, or their integer\n   equivalents ``0`` or ``1``.  When this method is not defined,\n   ``__len__()`` is called, if it is defined, and the object is\n   considered true if its result is nonzero. If a class defines\n   neither ``__len__()`` nor ``__bool__()``, all its instances are\n   considered true.\n',
- 'debugger': '\n``pdb`` --- The Python Debugger\n*******************************\n\nThe module ``pdb`` defines an interactive source code debugger for\nPython programs.  It supports setting (conditional) breakpoints and\nsingle stepping at the source line level, inspection of stack frames,\nsource code listing, and evaluation of arbitrary Python code in the\ncontext of any stack frame.  It also supports post-mortem debugging\nand can be called under program control.\n\nThe debugger is extensible --- it is actually defined as the class\n``Pdb``. This is currently undocumented but easily understood by\nreading the source.  The extension interface uses the modules ``bdb``\n(undocumented) and ``cmd``.\n\nThe debugger\'s prompt is ``(Pdb)``. Typical usage to run a program\nunder control of the debugger is:\n\n   >>> import pdb\n   >>> import mymodule\n   >>> pdb.run(\'mymodule.test()\')\n   > <string>(0)?()\n   (Pdb) continue\n   > <string>(1)?()\n   (Pdb) continue\n   NameError: \'spam\'\n   > <string>(1)?()\n   (Pdb)\n\n``pdb.py`` can also be invoked as a script to debug other scripts.\nFor example:\n\n   python -m pdb myscript.py\n\nWhen invoked as a script, pdb will automatically enter post-mortem\ndebugging if the program being debugged exits abnormally. After post-\nmortem debugging (or after normal exit of the program), pdb will\nrestart the program. Automatic restarting preserves pdb\'s state (such\nas breakpoints) and in most cases is more useful than quitting the\ndebugger upon program\'s exit.\n\nThe typical usage to break into the debugger from a running program is\nto insert\n\n   import pdb; pdb.set_trace()\n\nat the location you want to break into the debugger.  You can then\nstep through the code following this statement, and continue running\nwithout debugger using the ``c`` command.\n\nThe typical usage to inspect a crashed program is:\n\n   >>> import pdb\n   >>> import mymodule\n   >>> mymodule.test()\n   Traceback (most recent call last):\n     File "<stdin>", line 1, in ?\n     File "./mymodule.py", line 4, in test\n       test2()\n     File "./mymodule.py", line 3, in test2\n       print(spam)\n   NameError: spam\n   >>> pdb.pm()\n   > ./mymodule.py(3)test2()\n   -> print(spam)\n   (Pdb)\n\nThe module defines the following functions; each enters the debugger\nin a slightly different way:\n\npdb.run(statement[, globals[, locals]])\n\n   Execute the *statement* (given as a string) under debugger control.\n   The debugger prompt appears before any code is executed; you can\n   set breakpoints and type ``continue``, or you can step through the\n   statement using ``step`` or ``next`` (all these commands are\n   explained below).  The optional *globals* and *locals* arguments\n   specify the environment in which the code is executed; by default\n   the dictionary of the module ``__main__`` is used.  (See the\n   explanation of the built-in ``exec()`` or ``eval()`` functions.)\n\npdb.runeval(expression[, globals[, locals]])\n\n   Evaluate the *expression* (given as a string) under debugger\n   control.  When ``runeval()`` returns, it returns the value of the\n   expression.  Otherwise this function is similar to ``run()``.\n\npdb.runcall(function[, argument, ...])\n\n   Call the *function* (a function or method object, not a string)\n   with the given arguments.  When ``runcall()`` returns, it returns\n   whatever the function call returned.  The debugger prompt appears\n   as soon as the function is entered.\n\npdb.set_trace()\n\n   Enter the debugger at the calling stack frame.  This is useful to\n   hard-code a breakpoint at a given point in a program, even if the\n   code is not otherwise being debugged (e.g. when an assertion\n   fails).\n\npdb.post_mortem([traceback])\n\n   Enter post-mortem debugging of the given *traceback* object.  If no\n   *traceback* is given, it uses the one of the exception that is\n   currently being handled (an exception must be being handled if the\n   default is to be used).\n\npdb.pm()\n\n   Enter post-mortem debugging of the traceback found in\n   ``sys.last_traceback``.\n\nThe ``run_*`` functions and ``set_trace()`` are aliases for\ninstantiating the ``Pdb`` class and calling the method of the same\nname.  If you want to access further features, you have to do this\nyourself:\n\nclass class pdb.Pdb(completekey=\'tab\', stdin=None, stdout=None, skip=None)\n\n   ``Pdb`` is the debugger class.\n\n   The *completekey*, *stdin* and *stdout* arguments are passed to the\n   underlying ``cmd.Cmd`` class; see the description there.\n\n   The *skip* argument, if given, must be an iterable of glob-style\n   module name patterns.  The debugger will not step into frames that\n   originate in a module that matches one of these patterns. [1]\n\n   Example call to enable tracing with *skip*:\n\n      import pdb; pdb.Pdb(skip=[\'django.*\']).set_trace()\n\n   New in version 3.1: The *skip* argument.\n\n   run(statement[, globals[, locals]])\n   runeval(expression[, globals[, locals]])\n   runcall(function[, argument, ...])\n   set_trace()\n\n      See the documentation for the functions explained above.\n',
+ 'customization': '\nBasic customization\n*******************\n\nobject.__new__(cls[, ...])\n\n   Called to create a new instance of class *cls*.  ``__new__()`` is a\n   static method (special-cased so you need not declare it as such)\n   that takes the class of which an instance was requested as its\n   first argument.  The remaining arguments are those passed to the\n   object constructor expression (the call to the class).  The return\n   value of ``__new__()`` should be the new object instance (usually\n   an instance of *cls*).\n\n   Typical implementations create a new instance of the class by\n   invoking the superclass\'s ``__new__()`` method using\n   ``super(currentclass, cls).__new__(cls[, ...])`` with appropriate\n   arguments and then modifying the newly-created instance as\n   necessary before returning it.\n\n   If ``__new__()`` returns an instance of *cls*, then the new\n   instance\'s ``__init__()`` method will be invoked like\n   ``__init__(self[, ...])``, where *self* is the new instance and the\n   remaining arguments are the same as were passed to ``__new__()``.\n\n   If ``__new__()`` does not return an instance of *cls*, then the new\n   instance\'s ``__init__()`` method will not be invoked.\n\n   ``__new__()`` is intended mainly to allow subclasses of immutable\n   types (like int, str, or tuple) to customize instance creation.  It\n   is also commonly overridden in custom metaclasses in order to\n   customize class creation.\n\nobject.__init__(self[, ...])\n\n   Called when the instance is created.  The arguments are those\n   passed to the class constructor expression.  If a base class has an\n   ``__init__()`` method, the derived class\'s ``__init__()`` method,\n   if any, must explicitly call it to ensure proper initialization of\n   the base class part of the instance; for example:\n   ``BaseClass.__init__(self, [args...])``.  As a special constraint\n   on constructors, no value may be returned; doing so will cause a\n   ``TypeError`` to be raised at runtime.\n\nobject.__del__(self)\n\n   Called when the instance is about to be destroyed.  This is also\n   called a destructor.  If a base class has a ``__del__()`` method,\n   the derived class\'s ``__del__()`` method, if any, must explicitly\n   call it to ensure proper deletion of the base class part of the\n   instance.  Note that it is possible (though not recommended!) for\n   the ``__del__()`` method to postpone destruction of the instance by\n   creating a new reference to it.  It may then be called at a later\n   time when this new reference is deleted.  It is not guaranteed that\n   ``__del__()`` methods are called for objects that still exist when\n   the interpreter exits.\n\n   Note: ``del x`` doesn\'t directly call ``x.__del__()`` --- the former\n     decrements the reference count for ``x`` by one, and the latter\n     is only called when ``x``\'s reference count reaches zero.  Some\n     common situations that may prevent the reference count of an\n     object from going to zero include: circular references between\n     objects (e.g., a doubly-linked list or a tree data structure with\n     parent and child pointers); a reference to the object on the\n     stack frame of a function that caught an exception (the traceback\n     stored in ``sys.exc_info()[2]`` keeps the stack frame alive); or\n     a reference to the object on the stack frame that raised an\n     unhandled exception in interactive mode (the traceback stored in\n     ``sys.last_traceback`` keeps the stack frame alive).  The first\n     situation can only be remedied by explicitly breaking the cycles;\n     the latter two situations can be resolved by storing ``None`` in\n     ``sys.last_traceback``. Circular references which are garbage are\n     detected when the option cycle detector is enabled (it\'s on by\n     default), but can only be cleaned up if there are no Python-\n     level ``__del__()`` methods involved. Refer to the documentation\n     for the ``gc`` module for more information about how\n     ``__del__()`` methods are handled by the cycle detector,\n     particularly the description of the ``garbage`` value.\n\n   Warning: Due to the precarious circumstances under which ``__del__()``\n     methods are invoked, exceptions that occur during their execution\n     are ignored, and a warning is printed to ``sys.stderr`` instead.\n     Also, when ``__del__()`` is invoked in response to a module being\n     deleted (e.g., when execution of the program is done), other\n     globals referenced by the ``__del__()`` method may already have\n     been deleted or in the process of being torn down (e.g. the\n     import machinery shutting down).  For this reason, ``__del__()``\n     methods should do the absolute minimum needed to maintain\n     external invariants.  Starting with version 1.5, Python\n     guarantees that globals whose name begins with a single\n     underscore are deleted from their module before other globals are\n     deleted; if no other references to such globals exist, this may\n     help in assuring that imported modules are still available at the\n     time when the ``__del__()`` method is called.\n\nobject.__repr__(self)\n\n   Called by the ``repr()`` built-in function to compute the\n   "official" string representation of an object.  If at all possible,\n   this should look like a valid Python expression that could be used\n   to recreate an object with the same value (given an appropriate\n   environment).  If this is not possible, a string of the form\n   ``<...some useful description...>`` should be returned. The return\n   value must be a string object. If a class defines ``__repr__()``\n   but not ``__str__()``, then ``__repr__()`` is also used when an\n   "informal" string representation of instances of that class is\n   required.\n\n   This is typically used for debugging, so it is important that the\n   representation is information-rich and unambiguous.\n\nobject.__str__(self)\n\n   Called by the ``str()`` built-in function and by the ``print()``\n   function to compute the "informal" string representation of an\n   object.  This differs from ``__repr__()`` in that it does not have\n   to be a valid Python expression: a more convenient or concise\n   representation may be used instead. The return value must be a\n   string object.\n\nobject.__format__(self, format_spec)\n\n   Called by the ``format()`` built-in function (and by extension, the\n   ``format()`` method of class ``str``) to produce a "formatted"\n   string representation of an object. The ``format_spec`` argument is\n   a string that contains a description of the formatting options\n   desired. The interpretation of the ``format_spec`` argument is up\n   to the type implementing ``__format__()``, however most classes\n   will either delegate formatting to one of the built-in types, or\n   use a similar formatting option syntax.\n\n   See *Format Specification Mini-Language* for a description of the\n   standard formatting syntax.\n\n   The return value must be a string object.\n\nobject.__lt__(self, other)\nobject.__le__(self, other)\nobject.__eq__(self, other)\nobject.__ne__(self, other)\nobject.__gt__(self, other)\nobject.__ge__(self, other)\n\n   These are the so-called "rich comparison" methods. The\n   correspondence between operator symbols and method names is as\n   follows: ``x<y`` calls ``x.__lt__(y)``, ``x<=y`` calls\n   ``x.__le__(y)``, ``x==y`` calls ``x.__eq__(y)``, ``x!=y`` calls\n   ``x.__ne__(y)``, ``x>y`` calls ``x.__gt__(y)``, and ``x>=y`` calls\n   ``x.__ge__(y)``.\n\n   A rich comparison method may return the singleton\n   ``NotImplemented`` if it does not implement the operation for a\n   given pair of arguments. By convention, ``False`` and ``True`` are\n   returned for a successful comparison. However, these methods can\n   return any value, so if the comparison operator is used in a\n   Boolean context (e.g., in the condition of an ``if`` statement),\n   Python will call ``bool()`` on the value to determine if the result\n   is true or false.\n\n   There are no implied relationships among the comparison operators.\n   The truth of ``x==y`` does not imply that ``x!=y`` is false.\n   Accordingly, when defining ``__eq__()``, one should also define\n   ``__ne__()`` so that the operators will behave as expected.  See\n   the paragraph on ``__hash__()`` for some important notes on\n   creating *hashable* objects which support custom comparison\n   operations and are usable as dictionary keys.\n\n   There are no swapped-argument versions of these methods (to be used\n   when the left argument does not support the operation but the right\n   argument does); rather, ``__lt__()`` and ``__gt__()`` are each\n   other\'s reflection, ``__le__()`` and ``__ge__()`` are each other\'s\n   reflection, and ``__eq__()`` and ``__ne__()`` are their own\n   reflection.\n\n   Arguments to rich comparison methods are never coerced.\n\n   To automatically generate ordering operations from a single root\n   operation, see ``functools.total_ordering()``.\n\nobject.__hash__(self)\n\n   Called by built-in function ``hash()`` and for operations on\n   members of hashed collections including ``set``, ``frozenset``, and\n   ``dict``.  ``__hash__()`` should return an integer.  The only\n   required property is that objects which compare equal have the same\n   hash value; it is advised to somehow mix together (e.g. using\n   exclusive or) the hash values for the components of the object that\n   also play a part in comparison of objects.\n\n   If a class does not define an ``__eq__()`` method it should not\n   define a ``__hash__()`` operation either; if it defines\n   ``__eq__()`` but not ``__hash__()``, its instances will not be\n   usable as items in hashable collections.  If a class defines\n   mutable objects and implements an ``__eq__()`` method, it should\n   not implement ``__hash__()``, since the implementation of hashable\n   collections requires that a key\'s hash value is immutable (if the\n   object\'s hash value changes, it will be in the wrong hash bucket).\n\n   User-defined classes have ``__eq__()`` and ``__hash__()`` methods\n   by default; with them, all objects compare unequal (except with\n   themselves) and ``x.__hash__()`` returns ``id(x)``.\n\n   Classes which inherit a ``__hash__()`` method from a parent class\n   but change the meaning of ``__eq__()`` such that the hash value\n   returned is no longer appropriate (e.g. by switching to a value-\n   based concept of equality instead of the default identity based\n   equality) can explicitly flag themselves as being unhashable by\n   setting ``__hash__ = None`` in the class definition. Doing so means\n   that not only will instances of the class raise an appropriate\n   ``TypeError`` when a program attempts to retrieve their hash value,\n   but they will also be correctly identified as unhashable when\n   checking ``isinstance(obj, collections.Hashable)`` (unlike classes\n   which define their own ``__hash__()`` to explicitly raise\n   ``TypeError``).\n\n   If a class that overrides ``__eq__()`` needs to retain the\n   implementation of ``__hash__()`` from a parent class, the\n   interpreter must be told this explicitly by setting ``__hash__ =\n   <ParentClass>.__hash__``. Otherwise the inheritance of\n   ``__hash__()`` will be blocked, just as if ``__hash__`` had been\n   explicitly set to ``None``.\n\nobject.__bool__(self)\n\n   Called to implement truth value testing and the built-in operation\n   ``bool()``; should return ``False`` or ``True``.  When this method\n   is not defined, ``__len__()`` is called, if it is defined, and the\n   object is considered true if its result is nonzero.  If a class\n   defines neither ``__len__()`` nor ``__bool__()``, all its instances\n   are considered true.\n',
+ 'debugger': '\n``pdb`` --- The Python Debugger\n*******************************\n\nThe module ``pdb`` defines an interactive source code debugger for\nPython programs.  It supports setting (conditional) breakpoints and\nsingle stepping at the source line level, inspection of stack frames,\nsource code listing, and evaluation of arbitrary Python code in the\ncontext of any stack frame.  It also supports post-mortem debugging\nand can be called under program control.\n\nThe debugger is extensible -- it is actually defined as the class\n``Pdb``. This is currently undocumented but easily understood by\nreading the source.  The extension interface uses the modules ``bdb``\nand ``cmd``.\n\nThe debugger\'s prompt is ``(Pdb)``. Typical usage to run a program\nunder control of the debugger is:\n\n   >>> import pdb\n   >>> import mymodule\n   >>> pdb.run(\'mymodule.test()\')\n   > <string>(0)?()\n   (Pdb) continue\n   > <string>(1)?()\n   (Pdb) continue\n   NameError: \'spam\'\n   > <string>(1)?()\n   (Pdb)\n\n``pdb.py`` can also be invoked as a script to debug other scripts.\nFor example:\n\n   python3 -m pdb myscript.py\n\nWhen invoked as a script, pdb will automatically enter post-mortem\ndebugging if the program being debugged exits abnormally.  After post-\nmortem debugging (or after normal exit of the program), pdb will\nrestart the program.  Automatic restarting preserves pdb\'s state (such\nas breakpoints) and in most cases is more useful than quitting the\ndebugger upon program\'s exit.\n\nNew in version 3.2: ``pdb.py`` now accepts a ``-c`` option that\nexecutes commands as if given in a ``.pdbrc`` file, see *Debugger\nCommands*.\n\nThe typical usage to break into the debugger from a running program is\nto insert\n\n   import pdb; pdb.set_trace()\n\nat the location you want to break into the debugger.  You can then\nstep through the code following this statement, and continue running\nwithout the debugger using the ``continue`` command.\n\nThe typical usage to inspect a crashed program is:\n\n   >>> import pdb\n   >>> import mymodule\n   >>> mymodule.test()\n   Traceback (most recent call last):\n     File "<stdin>", line 1, in ?\n     File "./mymodule.py", line 4, in test\n       test2()\n     File "./mymodule.py", line 3, in test2\n       print(spam)\n   NameError: spam\n   >>> pdb.pm()\n   > ./mymodule.py(3)test2()\n   -> print(spam)\n   (Pdb)\n\nThe module defines the following functions; each enters the debugger\nin a slightly different way:\n\npdb.run(statement, globals=None, locals=None)\n\n   Execute the *statement* (given as a string or a code object) under\n   debugger control.  The debugger prompt appears before any code is\n   executed; you can set breakpoints and type ``continue``, or you can\n   step through the statement using ``step`` or ``next`` (all these\n   commands are explained below).  The optional *globals* and *locals*\n   arguments specify the environment in which the code is executed; by\n   default the dictionary of the module ``__main__`` is used.  (See\n   the explanation of the built-in ``exec()`` or ``eval()``\n   functions.)\n\npdb.runeval(expression, globals=None, locals=None)\n\n   Evaluate the *expression* (given as a string or a code object)\n   under debugger control.  When ``runeval()`` returns, it returns the\n   value of the expression.  Otherwise this function is similar to\n   ``run()``.\n\npdb.runcall(function, *args, **kwds)\n\n   Call the *function* (a function or method object, not a string)\n   with the given arguments.  When ``runcall()`` returns, it returns\n   whatever the function call returned.  The debugger prompt appears\n   as soon as the function is entered.\n\npdb.set_trace()\n\n   Enter the debugger at the calling stack frame.  This is useful to\n   hard-code a breakpoint at a given point in a program, even if the\n   code is not otherwise being debugged (e.g. when an assertion\n   fails).\n\npdb.post_mortem(traceback=None)\n\n   Enter post-mortem debugging of the given *traceback* object.  If no\n   *traceback* is given, it uses the one of the exception that is\n   currently being handled (an exception must be being handled if the\n   default is to be used).\n\npdb.pm()\n\n   Enter post-mortem debugging of the traceback found in\n   ``sys.last_traceback``.\n\nThe ``run_*`` functions and ``set_trace()`` are aliases for\ninstantiating the ``Pdb`` class and calling the method of the same\nname.  If you want to access further features, you have to do this\nyourself:\n\nclass class pdb.Pdb(completekey=\'tab\', stdin=None, stdout=None, skip=None)\n\n   ``Pdb`` is the debugger class.\n\n   The *completekey*, *stdin* and *stdout* arguments are passed to the\n   underlying ``cmd.Cmd`` class; see the description there.\n\n   The *skip* argument, if given, must be an iterable of glob-style\n   module name patterns.  The debugger will not step into frames that\n   originate in a module that matches one of these patterns. [1]\n\n   Example call to enable tracing with *skip*:\n\n      import pdb; pdb.Pdb(skip=[\'django.*\']).set_trace()\n\n   New in version 3.1: The *skip* argument.\n\n   run(statement, globals=None, locals=None)\n   runeval(expression, globals=None, locals=None)\n   runcall(function, *args, **kwds)\n   set_trace()\n\n      See the documentation for the functions explained above.\n',
  'del': '\nThe ``del`` statement\n*********************\n\n   del_stmt ::= "del" target_list\n\nDeletion is recursively defined very similar to the way assignment is\ndefined. Rather that spelling it out in full details, here are some\nhints.\n\nDeletion of a target list recursively deletes each target, from left\nto right.\n\nDeletion of a name removes the binding of that name from the local or\nglobal namespace, depending on whether the name occurs in a ``global``\nstatement in the same code block.  If the name is unbound, a\n``NameError`` exception will be raised.\n\nIt is illegal to delete a name from the local namespace if it occurs\nas a free variable in a nested block.\n\nDeletion of attribute references, subscriptions and slicings is passed\nto the primary object involved; deletion of a slicing is in general\nequivalent to assignment of an empty slice of the right type (but even\nthis is determined by the sliced object).\n',
  'dict': '\nDictionary displays\n*******************\n\nA dictionary display is a possibly empty series of key/datum pairs\nenclosed in curly braces:\n\n   dict_display       ::= "{" [key_datum_list | dict_comprehension] "}"\n   key_datum_list     ::= key_datum ("," key_datum)* [","]\n   key_datum          ::= expression ":" expression\n   dict_comprehension ::= expression ":" expression comp_for\n\nA dictionary display yields a new dictionary object.\n\nIf a comma-separated sequence of key/datum pairs is given, they are\nevaluated from left to right to define the entries of the dictionary:\neach key object is used as a key into the dictionary to store the\ncorresponding datum.  This means that you can specify the same key\nmultiple times in the key/datum list, and the final dictionary\'s value\nfor that key will be the last one given.\n\nA dict comprehension, in contrast to list and set comprehensions,\nneeds two expressions separated with a colon followed by the usual\n"for" and "if" clauses. When the comprehension is run, the resulting\nkey and value elements are inserted in the new dictionary in the order\nthey are produced.\n\nRestrictions on the types of the key values are listed earlier in\nsection *The standard type hierarchy*.  (To summarize, the key type\nshould be *hashable*, which excludes all mutable objects.)  Clashes\nbetween duplicate keys are not detected; the last datum (textually\nrightmost in the display) stored for a given key value prevails.\n',
  'dynamic-features': '\nInteraction with dynamic features\n*********************************\n\nThere are several cases where Python statements are illegal when used\nin conjunction with nested scopes that contain free variables.\n\nIf a variable is referenced in an enclosing scope, it is illegal to\ndelete the name.  An error will be reported at compile time.\n\nIf the wild card form of import --- ``import *`` --- is used in a\nfunction and the function contains or is a nested block with free\nvariables, the compiler will raise a ``SyntaxError``.\n\nThe ``eval()`` and ``exec()`` functions do not have access to the full\nenvironment for resolving names.  Names may be resolved in the local\nand global namespaces of the caller.  Free variables are not resolved\nin the nearest enclosing namespace, but in the global namespace.  [1]\nThe ``exec()`` and ``eval()`` functions have optional arguments to\noverride the global and local namespace.  If only one namespace is\nspecified, it is used for both.\n',
  'else': '\nThe ``if`` statement\n********************\n\nThe ``if`` statement is used for conditional execution:\n\n   if_stmt ::= "if" expression ":" suite\n               ( "elif" expression ":" suite )*\n               ["else" ":" suite]\n\nIt selects exactly one of the suites by evaluating the expressions one\nby one until one is found to be true (see section *Boolean operations*\nfor the definition of true and false); then that suite is executed\n(and no other part of the ``if`` statement is executed or evaluated).\nIf all expressions are false, the suite of the ``else`` clause, if\npresent, is executed.\n',
  'exceptions': '\nExceptions\n**********\n\nExceptions are a means of breaking out of the normal flow of control\nof a code block in order to handle errors or other exceptional\nconditions.  An exception is *raised* at the point where the error is\ndetected; it may be *handled* by the surrounding code block or by any\ncode block that directly or indirectly invoked the code block where\nthe error occurred.\n\nThe Python interpreter raises an exception when it detects a run-time\nerror (such as division by zero).  A Python program can also\nexplicitly raise an exception with the ``raise`` statement. Exception\nhandlers are specified with the ``try`` ... ``except`` statement.  The\n``finally`` clause of such a statement can be used to specify cleanup\ncode which does not handle the exception, but is executed whether an\nexception occurred or not in the preceding code.\n\nPython uses the "termination" model of error handling: an exception\nhandler can find out what happened and continue execution at an outer\nlevel, but it cannot repair the cause of the error and retry the\nfailing operation (except by re-entering the offending piece of code\nfrom the top).\n\nWhen an exception is not handled at all, the interpreter terminates\nexecution of the program, or returns to its interactive main loop.  In\neither case, it prints a stack backtrace, except when the exception is\n``SystemExit``.\n\nExceptions are identified by class instances.  The ``except`` clause\nis selected depending on the class of the instance: it must reference\nthe class of the instance or a base class thereof.  The instance can\nbe received by the handler and can carry additional information about\nthe exceptional condition.\n\nNote: Exception messages are not part of the Python API.  Their contents\n  may change from one version of Python to the next without warning\n  and should not be relied on by code which will run under multiple\n  versions of the interpreter.\n\nSee also the description of the ``try`` statement in section *The try\nstatement* and ``raise`` statement in section *The raise statement*.\n\n-[ Footnotes ]-\n\n[1] This limitation occurs because the code that is executed by these\n    operations is not available at the time the module is compiled.\n',
- 'execmodel': '\nExecution model\n***************\n\n\nNaming and binding\n==================\n\n*Names* refer to objects.  Names are introduced by name binding\noperations. Each occurrence of a name in the program text refers to\nthe *binding* of that name established in the innermost function block\ncontaining the use.\n\nA *block* is a piece of Python program text that is executed as a\nunit. The following are blocks: a module, a function body, and a class\ndefinition. Each command typed interactively is a block.  A script\nfile (a file given as standard input to the interpreter or specified\non the interpreter command line the first argument) is a code block.\nA script command (a command specified on the interpreter command line\nwith the \'**-c**\' option) is a code block.  The string argument passed\nto the built-in functions ``eval()`` and ``exec()`` is a code block.\n\nA code block is executed in an *execution frame*.  A frame contains\nsome administrative information (used for debugging) and determines\nwhere and how execution continues after the code block\'s execution has\ncompleted.\n\nA *scope* defines the visibility of a name within a block.  If a local\nvariable is defined in a block, its scope includes that block.  If the\ndefinition occurs in a function block, the scope extends to any blocks\ncontained within the defining one, unless a contained block introduces\na different binding for the name.  The scope of names defined in a\nclass block is limited to the class block; it does not extend to the\ncode blocks of methods -- this includes comprehensions and generator\nexpressions since they are implemented using a function scope.  This\nmeans that the following will fail:\n\n   class A:\n       a = 42\n       b = list(a + i for i in range(10))\n\nWhen a name is used in a code block, it is resolved using the nearest\nenclosing scope.  The set of all such scopes visible to a code block\nis called the block\'s *environment*.\n\nIf a name is bound in a block, it is a local variable of that block,\nunless declared as ``nonlocal``.  If a name is bound at the module\nlevel, it is a global variable.  (The variables of the module code\nblock are local and global.)  If a variable is used in a code block\nbut not defined there, it is a *free variable*.\n\nWhen a name is not found at all, a ``NameError`` exception is raised.\nIf the name refers to a local variable that has not been bound, a\n``UnboundLocalError`` exception is raised.  ``UnboundLocalError`` is a\nsubclass of ``NameError``.\n\nThe following constructs bind names: formal parameters to functions,\n``import`` statements, class and function definitions (these bind the\nclass or function name in the defining block), and targets that are\nidentifiers if occurring in an assignment, ``for`` loop header, or\nafter ``as`` in a ``with`` statement or :keyword.`except` clause. The\n``import`` statement of the form ``from ... import *`` binds all names\ndefined in the imported module, except those beginning with an\nunderscore.  This form may only be used at the module level.\n\nA target occurring in a ``del`` statement is also considered bound for\nthis purpose (though the actual semantics are to unbind the name).  It\nis illegal to unbind a name that is referenced by an enclosing scope;\nthe compiler will report a ``SyntaxError``.\n\nEach assignment or import statement occurs within a block defined by a\nclass or function definition or at the module level (the top-level\ncode block).\n\nIf a name binding operation occurs anywhere within a code block, all\nuses of the name within the block are treated as references to the\ncurrent block.  This can lead to errors when a name is used within a\nblock before it is bound.  This rule is subtle.  Python lacks\ndeclarations and allows name binding operations to occur anywhere\nwithin a code block.  The local variables of a code block can be\ndetermined by scanning the entire text of the block for name binding\noperations.\n\nIf the ``global`` statement occurs within a block, all uses of the\nname specified in the statement refer to the binding of that name in\nthe top-level namespace.  Names are resolved in the top-level\nnamespace by searching the global namespace, i.e. the namespace of the\nmodule containing the code block, and the builtin namespace, the\nnamespace of the module ``builtins``.  The global namespace is\nsearched first.  If the name is not found there, the builtin namespace\nis searched.  The global statement must precede all uses of the name.\n\nThe built-in namespace associated with the execution of a code block\nis actually found by looking up the name ``__builtins__`` in its\nglobal namespace; this should be a dictionary or a module (in the\nlatter case the module\'s dictionary is used).  By default, when in the\n``__main__`` module, ``__builtins__`` is the built-in module\n``builtins``; when in any other module, ``__builtins__`` is an alias\nfor the dictionary of the ``builtins`` module itself.\n``__builtins__`` can be set to a user-created dictionary to create a\nweak form of restricted execution.\n\nNote: Users should not touch ``__builtins__``; it is strictly an\n  implementation detail.  Users wanting to override values in the\n  built-in namespace should ``import`` the ``builtins`` module and\n  modify its attributes appropriately.\n\nThe namespace for a module is automatically created the first time a\nmodule is imported.  The main module for a script is always called\n``__main__``.\n\nThe global statement has the same scope as a name binding operation in\nthe same block.  If the nearest enclosing scope for a free variable\ncontains a global statement, the free variable is treated as a global.\n\nA class definition is an executable statement that may use and define\nnames. These references follow the normal rules for name resolution.\nThe namespace of the class definition becomes the attribute dictionary\nof the class.  Names defined at the class scope are not visible in\nmethods.\n\n\nInteraction with dynamic features\n---------------------------------\n\nThere are several cases where Python statements are illegal when used\nin conjunction with nested scopes that contain free variables.\n\nIf a variable is referenced in an enclosing scope, it is illegal to\ndelete the name.  An error will be reported at compile time.\n\nIf the wild card form of import --- ``import *`` --- is used in a\nfunction and the function contains or is a nested block with free\nvariables, the compiler will raise a ``SyntaxError``.\n\nThe ``eval()`` and ``exec()`` functions do not have access to the full\nenvironment for resolving names.  Names may be resolved in the local\nand global namespaces of the caller.  Free variables are not resolved\nin the nearest enclosing namespace, but in the global namespace.  [1]\nThe ``exec()`` and ``eval()`` functions have optional arguments to\noverride the global and local namespace.  If only one namespace is\nspecified, it is used for both.\n\n\nExceptions\n==========\n\nExceptions are a means of breaking out of the normal flow of control\nof a code block in order to handle errors or other exceptional\nconditions.  An exception is *raised* at the point where the error is\ndetected; it may be *handled* by the surrounding code block or by any\ncode block that directly or indirectly invoked the code block where\nthe error occurred.\n\nThe Python interpreter raises an exception when it detects a run-time\nerror (such as division by zero).  A Python program can also\nexplicitly raise an exception with the ``raise`` statement. Exception\nhandlers are specified with the ``try`` ... ``except`` statement.  The\n``finally`` clause of such a statement can be used to specify cleanup\ncode which does not handle the exception, but is executed whether an\nexception occurred or not in the preceding code.\n\nPython uses the "termination" model of error handling: an exception\nhandler can find out what happened and continue execution at an outer\nlevel, but it cannot repair the cause of the error and retry the\nfailing operation (except by re-entering the offending piece of code\nfrom the top).\n\nWhen an exception is not handled at all, the interpreter terminates\nexecution of the program, or returns to its interactive main loop.  In\neither case, it prints a stack backtrace, except when the exception is\n``SystemExit``.\n\nExceptions are identified by class instances.  The ``except`` clause\nis selected depending on the class of the instance: it must reference\nthe class of the instance or a base class thereof.  The instance can\nbe received by the handler and can carry additional information about\nthe exceptional condition.\n\nNote: Exception messages are not part of the Python API.  Their contents\n  may change from one version of Python to the next without warning\n  and should not be relied on by code which will run under multiple\n  versions of the interpreter.\n\nSee also the description of the ``try`` statement in section *The try\nstatement* and ``raise`` statement in section *The raise statement*.\n\n-[ Footnotes ]-\n\n[1] This limitation occurs because the code that is executed by these\n    operations is not available at the time the module is compiled.\n',
+ 'execmodel': '\nExecution model\n***************\n\n\nNaming and binding\n==================\n\n*Names* refer to objects.  Names are introduced by name binding\noperations. Each occurrence of a name in the program text refers to\nthe *binding* of that name established in the innermost function block\ncontaining the use.\n\nA *block* is a piece of Python program text that is executed as a\nunit. The following are blocks: a module, a function body, and a class\ndefinition. Each command typed interactively is a block.  A script\nfile (a file given as standard input to the interpreter or specified\non the interpreter command line the first argument) is a code block.\nA script command (a command specified on the interpreter command line\nwith the \'**-c**\' option) is a code block.  The string argument passed\nto the built-in functions ``eval()`` and ``exec()`` is a code block.\n\nA code block is executed in an *execution frame*.  A frame contains\nsome administrative information (used for debugging) and determines\nwhere and how execution continues after the code block\'s execution has\ncompleted.\n\nA *scope* defines the visibility of a name within a block.  If a local\nvariable is defined in a block, its scope includes that block.  If the\ndefinition occurs in a function block, the scope extends to any blocks\ncontained within the defining one, unless a contained block introduces\na different binding for the name.  The scope of names defined in a\nclass block is limited to the class block; it does not extend to the\ncode blocks of methods -- this includes comprehensions and generator\nexpressions since they are implemented using a function scope.  This\nmeans that the following will fail:\n\n   class A:\n       a = 42\n       b = list(a + i for i in range(10))\n\nWhen a name is used in a code block, it is resolved using the nearest\nenclosing scope.  The set of all such scopes visible to a code block\nis called the block\'s *environment*.\n\nIf a name is bound in a block, it is a local variable of that block,\nunless declared as ``nonlocal``.  If a name is bound at the module\nlevel, it is a global variable.  (The variables of the module code\nblock are local and global.)  If a variable is used in a code block\nbut not defined there, it is a *free variable*.\n\nWhen a name is not found at all, a ``NameError`` exception is raised.\nIf the name refers to a local variable that has not been bound, a\n``UnboundLocalError`` exception is raised.  ``UnboundLocalError`` is a\nsubclass of ``NameError``.\n\nThe following constructs bind names: formal parameters to functions,\n``import`` statements, class and function definitions (these bind the\nclass or function name in the defining block), and targets that are\nidentifiers if occurring in an assignment, ``for`` loop header, or\nafter ``as`` in a ``with`` statement or :keyword.`except` clause. The\n``import`` statement of the form ``from ... import *`` binds all names\ndefined in the imported module, except those beginning with an\nunderscore.  This form may only be used at the module level.\n\nA target occurring in a ``del`` statement is also considered bound for\nthis purpose (though the actual semantics are to unbind the name).  It\nis illegal to unbind a name that is referenced by an enclosing scope;\nthe compiler will report a ``SyntaxError``.\n\nEach assignment or import statement occurs within a block defined by a\nclass or function definition or at the module level (the top-level\ncode block).\n\nIf a name binding operation occurs anywhere within a code block, all\nuses of the name within the block are treated as references to the\ncurrent block.  This can lead to errors when a name is used within a\nblock before it is bound.  This rule is subtle.  Python lacks\ndeclarations and allows name binding operations to occur anywhere\nwithin a code block.  The local variables of a code block can be\ndetermined by scanning the entire text of the block for name binding\noperations.\n\nIf the ``global`` statement occurs within a block, all uses of the\nname specified in the statement refer to the binding of that name in\nthe top-level namespace.  Names are resolved in the top-level\nnamespace by searching the global namespace, i.e. the namespace of the\nmodule containing the code block, and the builtins namespace, the\nnamespace of the module ``builtins``.  The global namespace is\nsearched first.  If the name is not found there, the builtins\nnamespace is searched.  The global statement must precede all uses of\nthe name.\n\nThe builtins namespace associated with the execution of a code block\nis actually found by looking up the name ``__builtins__`` in its\nglobal namespace; this should be a dictionary or a module (in the\nlatter case the module\'s dictionary is used).  By default, when in the\n``__main__`` module, ``__builtins__`` is the built-in module\n``builtins``; when in any other module, ``__builtins__`` is an alias\nfor the dictionary of the ``builtins`` module itself.\n``__builtins__`` can be set to a user-created dictionary to create a\nweak form of restricted execution.\n\n**CPython implementation detail:** Users should not touch\n``__builtins__``; it is strictly an implementation detail.  Users\nwanting to override values in the builtins namespace should ``import``\nthe ``builtins`` module and modify its attributes appropriately.\n\nThe namespace for a module is automatically created the first time a\nmodule is imported.  The main module for a script is always called\n``__main__``.\n\nThe global statement has the same scope as a name binding operation in\nthe same block.  If the nearest enclosing scope for a free variable\ncontains a global statement, the free variable is treated as a global.\n\nA class definition is an executable statement that may use and define\nnames. These references follow the normal rules for name resolution.\nThe namespace of the class definition becomes the attribute dictionary\nof the class.  Names defined at the class scope are not visible in\nmethods.\n\n\nInteraction with dynamic features\n---------------------------------\n\nThere are several cases where Python statements are illegal when used\nin conjunction with nested scopes that contain free variables.\n\nIf a variable is referenced in an enclosing scope, it is illegal to\ndelete the name.  An error will be reported at compile time.\n\nIf the wild card form of import --- ``import *`` --- is used in a\nfunction and the function contains or is a nested block with free\nvariables, the compiler will raise a ``SyntaxError``.\n\nThe ``eval()`` and ``exec()`` functions do not have access to the full\nenvironment for resolving names.  Names may be resolved in the local\nand global namespaces of the caller.  Free variables are not resolved\nin the nearest enclosing namespace, but in the global namespace.  [1]\nThe ``exec()`` and ``eval()`` functions have optional arguments to\noverride the global and local namespace.  If only one namespace is\nspecified, it is used for both.\n\n\nExceptions\n==========\n\nExceptions are a means of breaking out of the normal flow of control\nof a code block in order to handle errors or other exceptional\nconditions.  An exception is *raised* at the point where the error is\ndetected; it may be *handled* by the surrounding code block or by any\ncode block that directly or indirectly invoked the code block where\nthe error occurred.\n\nThe Python interpreter raises an exception when it detects a run-time\nerror (such as division by zero).  A Python program can also\nexplicitly raise an exception with the ``raise`` statement. Exception\nhandlers are specified with the ``try`` ... ``except`` statement.  The\n``finally`` clause of such a statement can be used to specify cleanup\ncode which does not handle the exception, but is executed whether an\nexception occurred or not in the preceding code.\n\nPython uses the "termination" model of error handling: an exception\nhandler can find out what happened and continue execution at an outer\nlevel, but it cannot repair the cause of the error and retry the\nfailing operation (except by re-entering the offending piece of code\nfrom the top).\n\nWhen an exception is not handled at all, the interpreter terminates\nexecution of the program, or returns to its interactive main loop.  In\neither case, it prints a stack backtrace, except when the exception is\n``SystemExit``.\n\nExceptions are identified by class instances.  The ``except`` clause\nis selected depending on the class of the instance: it must reference\nthe class of the instance or a base class thereof.  The instance can\nbe received by the handler and can carry additional information about\nthe exceptional condition.\n\nNote: Exception messages are not part of the Python API.  Their contents\n  may change from one version of Python to the next without warning\n  and should not be relied on by code which will run under multiple\n  versions of the interpreter.\n\nSee also the description of the ``try`` statement in section *The try\nstatement* and ``raise`` statement in section *The raise statement*.\n\n-[ Footnotes ]-\n\n[1] This limitation occurs because the code that is executed by these\n    operations is not available at the time the module is compiled.\n',
  'exprlists': '\nExpression lists\n****************\n\n   expression_list ::= expression ( "," expression )* [","]\n\nAn expression list containing at least one comma yields a tuple.  The\nlength of the tuple is the number of expressions in the list.  The\nexpressions are evaluated from left to right.\n\nThe trailing comma is required only to create a single tuple (a.k.a. a\n*singleton*); it is optional in all other cases.  A single expression\nwithout a trailing comma doesn\'t create a tuple, but rather yields the\nvalue of that expression. (To create an empty tuple, use an empty pair\nof parentheses: ``()``.)\n',
  'floating': '\nFloating point literals\n***********************\n\nFloating point literals are described by the following lexical\ndefinitions:\n\n   floatnumber   ::= pointfloat | exponentfloat\n   pointfloat    ::= [intpart] fraction | intpart "."\n   exponentfloat ::= (intpart | pointfloat) exponent\n   intpart       ::= digit+\n   fraction      ::= "." digit+\n   exponent      ::= ("e" | "E") ["+" | "-"] digit+\n\nNote that the integer and exponent parts are always interpreted using\nradix 10. For example, ``077e010`` is legal, and denotes the same\nnumber as ``77e10``. The allowed range of floating point literals is\nimplementation-dependent. Some examples of floating point literals:\n\n   3.14    10.    .001    1e100    3.14e-10    0e0\n\nNote that numeric literals do not include a sign; a phrase like ``-1``\nis actually an expression composed of the unary operator ``-`` and the\nliteral ``1``.\n',
  'for': '\nThe ``for`` statement\n*********************\n\nThe ``for`` statement is used to iterate over the elements of a\nsequence (such as a string, tuple or list) or other iterable object:\n\n   for_stmt ::= "for" target_list "in" expression_list ":" suite\n                ["else" ":" suite]\n\nThe expression list is evaluated once; it should yield an iterable\nobject.  An iterator is created for the result of the\n``expression_list``.  The suite is then executed once for each item\nprovided by the iterator, in the order of ascending indices.  Each\nitem in turn is assigned to the target list using the standard rules\nfor assignments (see *Assignment statements*), and then the suite is\nexecuted.  When the items are exhausted (which is immediately when the\nsequence is empty or an iterator raises a ``StopIteration``\nexception), the suite in the ``else`` clause, if present, is executed,\nand the loop terminates.\n\nA ``break`` statement executed in the first suite terminates the loop\nwithout executing the ``else`` clause\'s suite.  A ``continue``\nstatement executed in the first suite skips the rest of the suite and\ncontinues with the next item, or with the ``else`` clause if there was\nno next item.\n\nThe suite may assign to the variable(s) in the target list; this does\nnot affect the next item assigned to it.\n\nNames in the target list are not deleted when the loop is finished,\nbut if the sequence is empty, it will not have been assigned to at all\nby the loop.  Hint: the built-in function ``range()`` returns an\niterator of integers suitable to emulate the effect of Pascal\'s ``for\ni := a to b do``; e.g., ``list(range(3))`` returns the list ``[0, 1,\n2]``.\n\nNote: There is a subtlety when the sequence is being modified by the loop\n  (this can only occur for mutable sequences, i.e. lists).  An\n  internal counter is used to keep track of which item is used next,\n  and this is incremented on each iteration.  When this counter has\n  reached the length of the sequence the loop terminates.  This means\n  that if the suite deletes the current (or a previous) item from the\n  sequence, the next item will be skipped (since it gets the index of\n  the current item which has already been treated).  Likewise, if the\n  suite inserts an item in the sequence before the current item, the\n  current item will be treated again the next time through the loop.\n  This can lead to nasty bugs that can be avoided by making a\n  temporary copy using a slice of the whole sequence, e.g.,\n\n     for x in a[:]:\n         if x < 0: a.remove(x)\n',
- 'formatstrings': '\nFormat String Syntax\n********************\n\nThe ``str.format()`` method and the ``Formatter`` class share the same\nsyntax for format strings (although in the case of ``Formatter``,\nsubclasses can define their own format string syntax.)\n\nFormat strings contain "replacement fields" surrounded by curly braces\n``{}``. Anything that is not contained in braces is considered literal\ntext, which is copied unchanged to the output.  If you need to include\na brace character in the literal text, it can be escaped by doubling:\n``{{`` and ``}}``.\n\nThe grammar for a replacement field is as follows:\n\n      replacement_field ::= "{" field_name ["!" conversion] [":" format_spec] "}"\n      field_name        ::= arg_name ("." attribute_name | "[" element_index "]")*\n      arg_name          ::= (identifier | integer)?\n      attribute_name    ::= identifier\n      element_index     ::= integer\n      conversion        ::= "r" | "s" | "a"\n      format_spec       ::= <described in the next section>\n\nIn less formal terms, the replacement field starts with a *field_name*\nthat specifies the object whose value is to be formatted and inserted\ninto the output instead of the replacement field. The *field_name* is\noptionally followed by a  *conversion* field, which is preceded by an\nexclamation point ``\'!\'``, and a *format_spec*, which is preceded by a\ncolon ``\':\'``.  These specify a non-default format for the replacement\nvalue.\n\nThe *field_name* itself begins with an *arg_name* that is either\neither a number or a keyword.  If it\'s a number, it refers to a\npositional argument, and if it\'s a keyword, it refers to a named\nkeyword argument.  If the numerical arg_names in a format string are\n0, 1, 2, ... in sequence, they can all be omitted (not just some) and\nthe numbers 0, 1, 2, ... will be automatically inserted in that order.\nThe *arg_name* can be followed by any number of index or attribute\nexpressions. An expression of the form ``\'.name\'`` selects the named\nattribute using ``getattr()``, while an expression of the form\n``\'[index]\'`` does an index lookup using ``__getitem__()``.\n\nSome simple format string examples:\n\n   "First, thou shalt count to {0}" # References first positional argument\n   "Bring me a {}"                  # Implicitly references the first positional argument\n   "From {} to {}"                  # Same as "From {0] to {1}"\n   "My quest is {name}"             # References keyword argument \'name\'\n   "Weight in tons {0.weight}"      # \'weight\' attribute of first positional arg\n   "Units destroyed: {players[0]}"  # First element of keyword argument \'players\'.\n\nThe *conversion* field causes a type coercion before formatting.\nNormally, the job of formatting a value is done by the\n``__format__()`` method of the value itself.  However, in some cases\nit is desirable to force a type to be formatted as a string,\noverriding its own definition of formatting.  By converting the value\nto a string before calling ``__format__()``, the normal formatting\nlogic is bypassed.\n\nThree conversion flags are currently supported: ``\'!s\'`` which calls\n``str()`` on the value, ``\'!r\'`` which calls ``repr()`` and ``\'!a\'``\nwhich calls ``ascii()``.\n\nSome examples:\n\n   "Harold\'s a clever {0!s}"        # Calls str() on the argument first\n   "Bring out the holy {name!r}"    # Calls repr() on the argument first\n\nThe *format_spec* field contains a specification of how the value\nshould be presented, including such details as field width, alignment,\npadding, decimal precision and so on.  Each value type can define it\'s\nown "formatting mini-language" or interpretation of the *format_spec*.\n\nMost built-in types support a common formatting mini-language, which\nis described in the next section.\n\nA *format_spec* field can also include nested replacement fields\nwithin it. These nested replacement fields can contain only a field\nname; conversion flags and format specifications are not allowed.  The\nreplacement fields within the format_spec are substituted before the\n*format_spec* string is interpreted. This allows the formatting of a\nvalue to be dynamically specified.\n\nFor example, suppose you wanted to have a replacement field whose\nfield width is determined by another variable:\n\n   "A man with two {0:{1}}".format("noses", 10)\n\nThis would first evaluate the inner replacement field, making the\nformat string effectively:\n\n   "A man with two {0:10}"\n\nThen the outer replacement field would be evaluated, producing:\n\n   "noses     "\n\nWhich is substituted into the string, yielding:\n\n   "A man with two noses     "\n\n(The extra space is because we specified a field width of 10, and\nbecause left alignment is the default for strings.)\n\n\nFormat Specification Mini-Language\n==================================\n\n"Format specifications" are used within replacement fields contained\nwithin a format string to define how individual values are presented\n(see *Format String Syntax*.)  They can also be passed directly to the\nbuiltin ``format()`` function.  Each formattable type may define how\nthe format specification is to be interpreted.\n\nMost built-in types implement the following options for format\nspecifications, although some of the formatting options are only\nsupported by the numeric types.\n\nA general convention is that an empty format string (``""``) produces\nthe same result as if you had called ``str()`` on the value.\n\nThe general form of a *standard format specifier* is:\n\n   format_spec ::= [[fill]align][sign][#][0][width][.precision][type]\n   fill        ::= <a character other than \'}\'>\n   align       ::= "<" | ">" | "=" | "^"\n   sign        ::= "+" | "-" | " "\n   width       ::= integer\n   precision   ::= integer\n   type        ::= "b" | "c" | "d" | "e" | "E" | "f" | "F" | "g" | "G" | "n" | "o" | "x" | "X" | "%"\n\nThe *fill* character can be any character other than \'}\' (which\nsignifies the end of the field).  The presence of a fill character is\nsignaled by the *next* character, which must be one of the alignment\noptions. If the second character of *format_spec* is not a valid\nalignment option, then it is assumed that both the fill character and\nthe alignment option are absent.\n\nThe meaning of the various alignment options is as follows:\n\n   +-----------+------------------------------------------------------------+\n   | Option    | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'<\'``   | Forces the field to be left-aligned within the available   |\n   |           | space (This is the default.)                               |\n   +-----------+------------------------------------------------------------+\n   | ``\'>\'``   | Forces the field to be right-aligned within the available  |\n   |           | space.                                                     |\n   +-----------+------------------------------------------------------------+\n   | ``\'=\'``   | Forces the padding to be placed after the sign (if any)    |\n   |           | but before the digits.  This is used for printing fields   |\n   |           | in the form \'+000000120\'. This alignment option is only    |\n   |           | valid for numeric types.                                   |\n   +-----------+------------------------------------------------------------+\n   | ``\'^\'``   | Forces the field to be centered within the available       |\n   |           | space.                                                     |\n   +-----------+------------------------------------------------------------+\n\nNote that unless a minimum field width is defined, the field width\nwill always be the same size as the data to fill it, so that the\nalignment option has no meaning in this case.\n\nThe *sign* option is only valid for number types, and can be one of\nthe following:\n\n   +-----------+------------------------------------------------------------+\n   | Option    | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'+\'``   | indicates that a sign should be used for both positive as  |\n   |           | well as negative numbers.                                  |\n   +-----------+------------------------------------------------------------+\n   | ``\'-\'``   | indicates that a sign should be used only for negative     |\n   |           | numbers (this is the default behavior).                    |\n   +-----------+------------------------------------------------------------+\n   | space     | indicates that a leading space should be used on positive  |\n   |           | numbers, and a minus sign on negative numbers.             |\n   +-----------+------------------------------------------------------------+\n\nThe ``\'#\'`` option is only valid for integers, and only for binary,\noctal, or hexadecimal output.  If present, it specifies that the\noutput will be prefixed by ``\'0b\'``, ``\'0o\'``, or ``\'0x\'``,\nrespectively.\n\n*width* is a decimal integer defining the minimum field width.  If not\nspecified, then the field width will be determined by the content.\n\nIf the *width* field is preceded by a zero (``\'0\'``) character, this\nenables zero-padding.  This is equivalent to an *alignment* type of\n``\'=\'`` and a *fill* character of ``\'0\'``.\n\nThe *precision* is a decimal number indicating how many digits should\nbe displayed after the decimal point for a floating point value\nformatted with ``\'f\'`` and ``\'F\'``, or before and after the decimal\npoint for a floating point value formatted with ``\'g\'`` or ``\'G\'``.\nFor non-number types the field indicates the maximum field size - in\nother words, how many characters will be used from the field content.\nThe *precision* is not allowed for integer values.\n\nFinally, the *type* determines how the data should be presented.\n\nThe available integer presentation types are:\n\n   +-----------+------------------------------------------------------------+\n   | Type      | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'b\'``   | Binary format. Outputs the number in base 2.               |\n   +-----------+------------------------------------------------------------+\n   | ``\'c\'``   | Character. Converts the integer to the corresponding       |\n   |           | unicode character before printing.                         |\n   +-----------+------------------------------------------------------------+\n   | ``\'d\'``   | Decimal Integer. Outputs the number in base 10.            |\n   +-----------+------------------------------------------------------------+\n   | ``\'o\'``   | Octal format. Outputs the number in base 8.                |\n   +-----------+------------------------------------------------------------+\n   | ``\'x\'``   | Hex format. Outputs the number in base 16, using lower-    |\n   |           | case letters for the digits above 9.                       |\n   +-----------+------------------------------------------------------------+\n   | ``\'X\'``   | Hex format. Outputs the number in base 16, using upper-    |\n   |           | case letters for the digits above 9.                       |\n   +-----------+------------------------------------------------------------+\n   | ``\'n\'``   | Number. This is the same as ``\'d\'``, except that it uses   |\n   |           | the current locale setting to insert the appropriate       |\n   |           | number separator characters.                               |\n   +-----------+------------------------------------------------------------+\n   | None      | The same as ``\'d\'``.                                       |\n   +-----------+------------------------------------------------------------+\n\nThe available presentation types for floating point and decimal values\nare:\n\n   +-----------+------------------------------------------------------------+\n   | Type      | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'e\'``   | Exponent notation. Prints the number in scientific         |\n   |           | notation using the letter \'e\' to indicate the exponent.    |\n   +-----------+------------------------------------------------------------+\n   | ``\'E\'``   | Exponent notation. Same as ``\'e\'`` except it uses an upper |\n   |           | case \'E\' as the separator character.                       |\n   +-----------+------------------------------------------------------------+\n   | ``\'f\'``   | Fixed point. Displays the number as a fixed-point number.  |\n   +-----------+------------------------------------------------------------+\n   | ``\'F\'``   | Fixed point. Same as ``\'f\'``, but converts ``nan`` to      |\n   |           | ``NAN`` and ``inf`` to ``INF``.                            |\n   +-----------+------------------------------------------------------------+\n   | ``\'g\'``   | General format. This prints the number as a fixed-point    |\n   |           | number, unless the number is too large, in which case it   |\n   |           | switches to ``\'e\'`` exponent notation. Infinity and NaN    |\n   |           | values are formatted as ``inf``, ``-inf`` and ``nan``,     |\n   |           | respectively.                                              |\n   +-----------+------------------------------------------------------------+\n   | ``\'G\'``   | General format. Same as ``\'g\'`` except switches to ``\'E\'`` |\n   |           | if the number gets to large. The representations of        |\n   |           | infinity and NaN are uppercased, too.                      |\n   +-----------+------------------------------------------------------------+\n   | ``\'n\'``   | Number. This is the same as ``\'g\'``, except that it uses   |\n   |           | the current locale setting to insert the appropriate       |\n   |           | number separator characters.                               |\n   +-----------+------------------------------------------------------------+\n   | ``\'%\'``   | Percentage. Multiplies the number by 100 and displays in   |\n   |           | fixed (``\'f\'``) format, followed by a percent sign.        |\n   +-----------+------------------------------------------------------------+\n   | None      | Similar to ``\'g\'``, except with at least one digit past    |\n   |           | the decimal point and a default precision of 12. This is   |\n   |           | intended to match ``str()``, except you can add the other  |\n   |           | format modifiers.                                          |\n   +-----------+------------------------------------------------------------+\n',
+ 'formatstrings': '\nFormat String Syntax\n********************\n\nThe ``str.format()`` method and the ``Formatter`` class share the same\nsyntax for format strings (although in the case of ``Formatter``,\nsubclasses can define their own format string syntax).\n\nFormat strings contain "replacement fields" surrounded by curly braces\n``{}``. Anything that is not contained in braces is considered literal\ntext, which is copied unchanged to the output.  If you need to include\na brace character in the literal text, it can be escaped by doubling:\n``{{`` and ``}}``.\n\nThe grammar for a replacement field is as follows:\n\n      replacement_field ::= "{" [field_name] ["!" conversion] [":" format_spec] "}"\n      field_name        ::= arg_name ("." attribute_name | "[" element_index "]")*\n      arg_name          ::= [identifier | integer]\n      attribute_name    ::= identifier\n      element_index     ::= integer | index_string\n      index_string      ::= <any source character except "]"> +\n      conversion        ::= "r" | "s" | "a"\n      format_spec       ::= <described in the next section>\n\nIn less formal terms, the replacement field can start with a\n*field_name* that specifies the object whose value is to be formatted\nand inserted into the output instead of the replacement field. The\n*field_name* is optionally followed by a  *conversion* field, which is\npreceded by an exclamation point ``\'!\'``, and a *format_spec*, which\nis preceded by a colon ``\':\'``.  These specify a non-default format\nfor the replacement value.\n\nSee also the *Format Specification Mini-Language* section.\n\nThe *field_name* itself begins with an *arg_name* that is either\neither a number or a keyword.  If it\'s a number, it refers to a\npositional argument, and if it\'s a keyword, it refers to a named\nkeyword argument.  If the numerical arg_names in a format string are\n0, 1, 2, ... in sequence, they can all be omitted (not just some) and\nthe numbers 0, 1, 2, ... will be automatically inserted in that order.\nThe *arg_name* can be followed by any number of index or attribute\nexpressions. An expression of the form ``\'.name\'`` selects the named\nattribute using ``getattr()``, while an expression of the form\n``\'[index]\'`` does an index lookup using ``__getitem__()``.\n\nChanged in version 3.1: The positional argument specifiers can be\nomitted, so ``\'{} {}\'`` is equivalent to ``\'{0} {1}\'``.\n\nSome simple format string examples:\n\n   "First, thou shalt count to {0}" # References first positional argument\n   "Bring me a {}"                  # Implicitly references the first positional argument\n   "From {} to {}"                  # Same as "From {0} to {1}"\n   "My quest is {name}"             # References keyword argument \'name\'\n   "Weight in tons {0.weight}"      # \'weight\' attribute of first positional arg\n   "Units destroyed: {players[0]}"  # First element of keyword argument \'players\'.\n\nThe *conversion* field causes a type coercion before formatting.\nNormally, the job of formatting a value is done by the\n``__format__()`` method of the value itself.  However, in some cases\nit is desirable to force a type to be formatted as a string,\noverriding its own definition of formatting.  By converting the value\nto a string before calling ``__format__()``, the normal formatting\nlogic is bypassed.\n\nThree conversion flags are currently supported: ``\'!s\'`` which calls\n``str()`` on the value, ``\'!r\'`` which calls ``repr()`` and ``\'!a\'``\nwhich calls ``ascii()``.\n\nSome examples:\n\n   "Harold\'s a clever {0!s}"        # Calls str() on the argument first\n   "Bring out the holy {name!r}"    # Calls repr() on the argument first\n   "More {!a}"                      # Calls ascii() on the argument first\n\nThe *format_spec* field contains a specification of how the value\nshould be presented, including such details as field width, alignment,\npadding, decimal precision and so on.  Each value type can define its\nown "formatting mini-language" or interpretation of the *format_spec*.\n\nMost built-in types support a common formatting mini-language, which\nis described in the next section.\n\nA *format_spec* field can also include nested replacement fields\nwithin it. These nested replacement fields can contain only a field\nname; conversion flags and format specifications are not allowed.  The\nreplacement fields within the format_spec are substituted before the\n*format_spec* string is interpreted. This allows the formatting of a\nvalue to be dynamically specified.\n\nSee the *Format examples* section for some examples.\n\n\nFormat Specification Mini-Language\n==================================\n\n"Format specifications" are used within replacement fields contained\nwithin a format string to define how individual values are presented\n(see *Format String Syntax*).  They can also be passed directly to the\nbuilt-in ``format()`` function.  Each formattable type may define how\nthe format specification is to be interpreted.\n\nMost built-in types implement the following options for format\nspecifications, although some of the formatting options are only\nsupported by the numeric types.\n\nA general convention is that an empty format string (``""``) produces\nthe same result as if you had called ``str()`` on the value. A non-\nempty format string typically modifies the result.\n\nThe general form of a *standard format specifier* is:\n\n   format_spec ::= [[fill]align][sign][#][0][width][,][.precision][type]\n   fill        ::= <a character other than \'}\'>\n   align       ::= "<" | ">" | "=" | "^"\n   sign        ::= "+" | "-" | " "\n   width       ::= integer\n   precision   ::= integer\n   type        ::= "b" | "c" | "d" | "e" | "E" | "f" | "F" | "g" | "G" | "n" | "o" | "s" | "x" | "X" | "%"\n\nThe *fill* character can be any character other than \'}\' (which\nsignifies the end of the field).  The presence of a fill character is\nsignaled by the *next* character, which must be one of the alignment\noptions. If the second character of *format_spec* is not a valid\nalignment option, then it is assumed that both the fill character and\nthe alignment option are absent.\n\nThe meaning of the various alignment options is as follows:\n\n   +-----------+------------------------------------------------------------+\n   | Option    | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'<\'``   | Forces the field to be left-aligned within the available   |\n   |           | space (this is the default).                               |\n   +-----------+------------------------------------------------------------+\n   | ``\'>\'``   | Forces the field to be right-aligned within the available  |\n   |           | space.                                                     |\n   +-----------+------------------------------------------------------------+\n   | ``\'=\'``   | Forces the padding to be placed after the sign (if any)    |\n   |           | but before the digits.  This is used for printing fields   |\n   |           | in the form \'+000000120\'. This alignment option is only    |\n   |           | valid for numeric types.                                   |\n   +-----------+------------------------------------------------------------+\n   | ``\'^\'``   | Forces the field to be centered within the available       |\n   |           | space.                                                     |\n   +-----------+------------------------------------------------------------+\n\nNote that unless a minimum field width is defined, the field width\nwill always be the same size as the data to fill it, so that the\nalignment option has no meaning in this case.\n\nThe *sign* option is only valid for number types, and can be one of\nthe following:\n\n   +-----------+------------------------------------------------------------+\n   | Option    | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'+\'``   | indicates that a sign should be used for both positive as  |\n   |           | well as negative numbers.                                  |\n   +-----------+------------------------------------------------------------+\n   | ``\'-\'``   | indicates that a sign should be used only for negative     |\n   |           | numbers (this is the default behavior).                    |\n   +-----------+------------------------------------------------------------+\n   | space     | indicates that a leading space should be used on positive  |\n   |           | numbers, and a minus sign on negative numbers.             |\n   +-----------+------------------------------------------------------------+\n\nThe ``\'#\'`` option is only valid for integers, and only for binary,\noctal, or hexadecimal output.  If present, it specifies that the\noutput will be prefixed by ``\'0b\'``, ``\'0o\'``, or ``\'0x\'``,\nrespectively.\n\nThe ``\',\'`` option signals the use of a comma for a thousands\nseparator. For a locale aware separator, use the ``\'n\'`` integer\npresentation type instead.\n\nChanged in version 3.1: Added the ``\',\'`` option (see also **PEP\n378**).\n\n*width* is a decimal integer defining the minimum field width.  If not\nspecified, then the field width will be determined by the content.\n\nIf the *width* field is preceded by a zero (``\'0\'``) character, this\nenables zero-padding.  This is equivalent to an *alignment* type of\n``\'=\'`` and a *fill* character of ``\'0\'``.\n\nThe *precision* is a decimal number indicating how many digits should\nbe displayed after the decimal point for a floating point value\nformatted with ``\'f\'`` and ``\'F\'``, or before and after the decimal\npoint for a floating point value formatted with ``\'g\'`` or ``\'G\'``.\nFor non-number types the field indicates the maximum field size - in\nother words, how many characters will be used from the field content.\nThe *precision* is not allowed for integer values.\n\nFinally, the *type* determines how the data should be presented.\n\nThe available string presentation types are:\n\n   +-----------+------------------------------------------------------------+\n   | Type      | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'s\'``   | String format. This is the default type for strings and    |\n   |           | may be omitted.                                            |\n   +-----------+------------------------------------------------------------+\n   | None      | The same as ``\'s\'``.                                       |\n   +-----------+------------------------------------------------------------+\n\nThe available integer presentation types are:\n\n   +-----------+------------------------------------------------------------+\n   | Type      | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'b\'``   | Binary format. Outputs the number in base 2.               |\n   +-----------+------------------------------------------------------------+\n   | ``\'c\'``   | Character. Converts the integer to the corresponding       |\n   |           | unicode character before printing.                         |\n   +-----------+------------------------------------------------------------+\n   | ``\'d\'``   | Decimal Integer. Outputs the number in base 10.            |\n   +-----------+------------------------------------------------------------+\n   | ``\'o\'``   | Octal format. Outputs the number in base 8.                |\n   +-----------+------------------------------------------------------------+\n   | ``\'x\'``   | Hex format. Outputs the number in base 16, using lower-    |\n   |           | case letters for the digits above 9.                       |\n   +-----------+------------------------------------------------------------+\n   | ``\'X\'``   | Hex format. Outputs the number in base 16, using upper-    |\n   |           | case letters for the digits above 9.                       |\n   +-----------+------------------------------------------------------------+\n   | ``\'n\'``   | Number. This is the same as ``\'d\'``, except that it uses   |\n   |           | the current locale setting to insert the appropriate       |\n   |           | number separator characters.                               |\n   +-----------+------------------------------------------------------------+\n   | None      | The same as ``\'d\'``.                                       |\n   +-----------+------------------------------------------------------------+\n\nIn addition to the above presentation types, integers can be formatted\nwith the floating point presentation types listed below (except\n``\'n\'`` and None). When doing so, ``float()`` is used to convert the\ninteger to a floating point number before formatting.\n\nThe available presentation types for floating point and decimal values\nare:\n\n   +-----------+------------------------------------------------------------+\n   | Type      | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'e\'``   | Exponent notation. Prints the number in scientific         |\n   |           | notation using the letter \'e\' to indicate the exponent.    |\n   +-----------+------------------------------------------------------------+\n   | ``\'E\'``   | Exponent notation. Same as ``\'e\'`` except it uses an upper |\n   |           | case \'E\' as the separator character.                       |\n   +-----------+------------------------------------------------------------+\n   | ``\'f\'``   | Fixed point. Displays the number as a fixed-point number.  |\n   +-----------+------------------------------------------------------------+\n   | ``\'F\'``   | Fixed point. Same as ``\'f\'``, but converts ``nan`` to      |\n   |           | ``NAN`` and ``inf`` to ``INF``.                            |\n   +-----------+------------------------------------------------------------+\n   | ``\'g\'``   | General format.  For a given precision ``p >= 1``, this    |\n   |           | rounds the number to ``p`` significant digits and then     |\n   |           | formats the result in either fixed-point format or in      |\n   |           | scientific notation, depending on its magnitude.  The      |\n   |           | precise rules are as follows: suppose that the result      |\n   |           | formatted with presentation type ``\'e\'`` and precision     |\n   |           | ``p-1`` would have exponent ``exp``.  Then if ``-4 <= exp  |\n   |           | < p``, the number is formatted with presentation type      |\n   |           | ``\'f\'`` and precision ``p-1-exp``. Otherwise, the number   |\n   |           | is formatted with presentation type ``\'e\'`` and precision  |\n   |           | ``p-1``. In both cases insignificant trailing zeros are    |\n   |           | removed from the significand, and the decimal point is     |\n   |           | also removed if there are no remaining digits following    |\n   |           | it.  Postive and negative infinity, positive and negative  |\n   |           | zero, and nans, are formatted as ``inf``, ``-inf``, ``0``, |\n   |           | ``-0`` and ``nan`` respectively, regardless of the         |\n   |           | precision.  A precision of ``0`` is treated as equivalent  |\n   |           | to a precision of ``1``.                                   |\n   +-----------+------------------------------------------------------------+\n   | ``\'G\'``   | General format. Same as ``\'g\'`` except switches to ``\'E\'`` |\n   |           | if the number gets too large. The representations of       |\n   |           | infinity and NaN are uppercased, too.                      |\n   +-----------+------------------------------------------------------------+\n   | ``\'n\'``   | Number. This is the same as ``\'g\'``, except that it uses   |\n   |           | the current locale setting to insert the appropriate       |\n   |           | number separator characters.                               |\n   +-----------+------------------------------------------------------------+\n   | ``\'%\'``   | Percentage. Multiplies the number by 100 and displays in   |\n   |           | fixed (``\'f\'``) format, followed by a percent sign.        |\n   +-----------+------------------------------------------------------------+\n   | None      | Similar to ``\'g\'``, except with at least one digit past    |\n   |           | the decimal point and a default precision of 12. This is   |\n   |           | intended to match ``str()``, except you can add the other  |\n   |           | format modifiers.                                          |\n   +-----------+------------------------------------------------------------+\n\n\nFormat examples\n===============\n\nThis section contains examples of the new format syntax and comparison\nwith the old ``%``-formatting.\n\nIn most of the cases the syntax is similar to the old\n``%``-formatting, with the addition of the ``{}`` and with ``:`` used\ninstead of ``%``. For example, ``\'%03.2f\'`` can be translated to\n``\'{:03.2f}\'``.\n\nThe new format syntax also supports new and different options, shown\nin the follow examples.\n\nAccessing arguments by position:\n\n   >>> \'{0}, {1}, {2}\'.format(\'a\', \'b\', \'c\')\n   \'a, b, c\'\n   >>> \'{}, {}, {}\'.format(\'a\', \'b\', \'c\')  # 3.1+ only\n   \'a, b, c\'\n   >>> \'{2}, {1}, {0}\'.format(\'a\', \'b\', \'c\')\n   \'c, b, a\'\n   >>> \'{2}, {1}, {0}\'.format(*\'abc\')      # unpacking argument sequence\n   \'c, b, a\'\n   >>> \'{0}{1}{0}\'.format(\'abra\', \'cad\')   # arguments\' indices can be repeated\n   \'abracadabra\'\n\nAccessing arguments by name:\n\n   >>> \'Coordinates: {latitude}, {longitude}\'.format(latitude=\'37.24N\', longitude=\'-115.81W\')\n   \'Coordinates: 37.24N, -115.81W\'\n   >>> coord = {\'latitude\': \'37.24N\', \'longitude\': \'-115.81W\'}\n   >>> \'Coordinates: {latitude}, {longitude}\'.format(**coord)\n   \'Coordinates: 37.24N, -115.81W\'\n\nAccessing arguments\' attributes:\n\n   >>> c = 3-5j\n   >>> (\'The complex number {0} is formed from the real part {0.real} \'\n   ...  \'and the imaginary part {0.imag}.\').format(c)\n   \'The complex number (3-5j) is formed from the real part 3.0 and the imaginary part -5.0.\'\n   >>> class Point:\n   ...     def __init__(self, x, y):\n   ...         self.x, self.y = x, y\n   ...     def __str__(self):\n   ...         return \'Point({self.x}, {self.y})\'.format(self=self)\n   ...\n   >>> str(Point(4, 2))\n   \'Point(4, 2)\'\n\nAccessing arguments\' items:\n\n   >>> coord = (3, 5)\n   >>> \'X: {0[0]};  Y: {0[1]}\'.format(coord)\n   \'X: 3;  Y: 5\'\n\nReplacing ``%s`` and ``%r``:\n\n   >>> "repr() shows quotes: {!r}; str() doesn\'t: {!s}".format(\'test1\', \'test2\')\n   "repr() shows quotes: \'test1\'; str() doesn\'t: test2"\n\nAligning the text and specifying a width:\n\n   >>> \'{:<30}\'.format(\'left aligned\')\n   \'left aligned                  \'\n   >>> \'{:>30}\'.format(\'right aligned\')\n   \'                 right aligned\'\n   >>> \'{:^30}\'.format(\'centered\')\n   \'           centered           \'\n   >>> \'{:*^30}\'.format(\'centered\')  # use \'*\' as a fill char\n   \'***********centered***********\'\n\nReplacing ``%+f``, ``%-f``, and ``% f`` and specifying a sign:\n\n   >>> \'{:+f}; {:+f}\'.format(3.14, -3.14)  # show it always\n   \'+3.140000; -3.140000\'\n   >>> \'{: f}; {: f}\'.format(3.14, -3.14)  # show a space for positive numbers\n   \' 3.140000; -3.140000\'\n   >>> \'{:-f}; {:-f}\'.format(3.14, -3.14)  # show only the minus -- same as \'{:f}; {:f}\'\n   \'3.140000; -3.140000\'\n\nReplacing ``%x`` and ``%o`` and converting the value to different\nbases:\n\n   >>> # format also supports binary numbers\n   >>> "int: {0:d};  hex: {0:x};  oct: {0:o};  bin: {0:b}".format(42)\n   \'int: 42;  hex: 2a;  oct: 52;  bin: 101010\'\n   >>> # with 0x, 0o, or 0b as prefix:\n   >>> "int: {0:d};  hex: {0:#x};  oct: {0:#o};  bin: {0:#b}".format(42)\n   \'int: 42;  hex: 0x2a;  oct: 0o52;  bin: 0b101010\'\n\nUsing the comma as a thousands separator:\n\n   >>> \'{:,}\'.format(1234567890)\n   \'1,234,567,890\'\n\nExpressing a percentage:\n\n   >>> points = 19\n   >>> total = 22\n   >>> \'Correct answers: {:.2%}.\'.format(points/total)\n   \'Correct answers: 86.36%\'\n\nUsing type-specific formatting:\n\n   >>> import datetime\n   >>> d = datetime.datetime(2010, 7, 4, 12, 15, 58)\n   >>> \'{:%Y-%m-%d %H:%M:%S}\'.format(d)\n   \'2010-07-04 12:15:58\'\n\nNesting arguments and more complex examples:\n\n   >>> for align, text in zip(\'<^>\', [\'left\', \'center\', \'right\']):\n   ...     \'{0:{align}{fill}16}\'.format(text, fill=align, align=align)\n   ...\n   \'left<<<<<<<<<<<<\'\n   \'^^^^^center^^^^^\'\n   \'>>>>>>>>>>>right\'\n   >>>\n   >>> octets = [192, 168, 0, 1]\n   >>> \'{:02X}{:02X}{:02X}{:02X}\'.format(*octets)\n   \'C0A80001\'\n   >>> int(_, 16)\n   3232235521\n   >>>\n   >>> width = 5\n   >>> for num in range(5,12):\n   ...     for base in \'dXob\':\n   ...         print(\'{0:{width}{base}}\'.format(num, base=base, width=width), end=\' \')\n   ...     print()\n   ...\n       5     5     5   101\n       6     6     6   110\n       7     7     7   111\n       8     8    10  1000\n       9     9    11  1001\n      10     A    12  1010\n      11     B    13  1011\n',
  'function': '\nFunction definitions\n********************\n\nA function definition defines a user-defined function object (see\nsection *The standard type hierarchy*):\n\n   funcdef        ::= [decorators] "def" funcname "(" [parameter_list] ")" ["->" expression] ":" suite\n   decorators     ::= decorator+\n   decorator      ::= "@" dotted_name ["(" [argument_list [","]] ")"] NEWLINE\n   dotted_name    ::= identifier ("." identifier)*\n   parameter_list ::= (defparameter ",")*\n                      (  "*" [parameter] ("," defparameter)*\n                      [, "**" parameter]\n                      | "**" parameter\n                      | defparameter [","] )\n   parameter      ::= identifier [":" expression]\n   defparameter   ::= parameter ["=" expression]\n   funcname       ::= identifier\n\nA function definition is an executable statement.  Its execution binds\nthe function name in the current local namespace to a function object\n(a wrapper around the executable code for the function).  This\nfunction object contains a reference to the current global namespace\nas the global namespace to be used when the function is called.\n\nThe function definition does not execute the function body; this gets\nexecuted only when the function is called. [3]\n\nA function definition may be wrapped by one or more *decorator*\nexpressions. Decorator expressions are evaluated when the function is\ndefined, in the scope that contains the function definition.  The\nresult must be a callable, which is invoked with the function object\nas the only argument. The returned value is bound to the function name\ninstead of the function object.  Multiple decorators are applied in\nnested fashion. For example, the following code\n\n   @f1(arg)\n   @f2\n   def func(): pass\n\nis equivalent to\n\n   def func(): pass\n   func = f1(arg)(f2(func))\n\nWhen one or more parameters have the form *parameter* ``=``\n*expression*, the function is said to have "default parameter values."\nFor a parameter with a default value, the corresponding argument may\nbe omitted from a call, in which case the parameter\'s default value is\nsubstituted.  If a parameter has a default value, all following\nparameters up until the "``*``" must also have a default value ---\nthis is a syntactic restriction that is not expressed by the grammar.\n\n**Default parameter values are evaluated when the function definition\nis executed.** This means that the expression is evaluated once, when\nthe function is defined, and that that same "pre-computed" value is\nused for each call.  This is especially important to understand when a\ndefault parameter is a mutable object, such as a list or a dictionary:\nif the function modifies the object (e.g. by appending an item to a\nlist), the default value is in effect modified. This is generally not\nwhat was intended.  A way around this is to use ``None`` as the\ndefault, and explicitly test for it in the body of the function, e.g.:\n\n   def whats_on_the_telly(penguin=None):\n       if penguin is None:\n           penguin = []\n       penguin.append("property of the zoo")\n       return penguin\n\nFunction call semantics are described in more detail in section\n*Calls*. A function call always assigns values to all parameters\nmentioned in the parameter list, either from position arguments, from\nkeyword arguments, or from default values.  If the form\n"``*identifier``" is present, it is initialized to a tuple receiving\nany excess positional parameters, defaulting to the empty tuple.  If\nthe form "``**identifier``" is present, it is initialized to a new\ndictionary receiving any excess keyword arguments, defaulting to a new\nempty dictionary. Parameters after "``*``" or "``*identifier``" are\nkeyword-only parameters and may only be passed used keyword arguments.\n\nParameters may have annotations of the form "``: expression``"\nfollowing the parameter name.  Any parameter may have an annotation\neven those of the form ``*identifier`` or ``**identifier``.  Functions\nmay have "return" annotation of the form "``-> expression``" after the\nparameter list.  These annotations can be any valid Python expression\nand are evaluated when the function definition is executed.\nAnnotations may be evaluated in a different order than they appear in\nthe source code.  The presence of annotations does not change the\nsemantics of a function.  The annotation values are available as\nvalues of a dictionary keyed by the parameters\' names in the\n``__annotations__`` attribute of the function object.\n\nIt is also possible to create anonymous functions (functions not bound\nto a name), for immediate use in expressions.  This uses lambda forms,\ndescribed in section *Lambdas*.  Note that the lambda form is merely a\nshorthand for a simplified function definition; a function defined in\na "``def``" statement can be passed around or assigned to another name\njust like a function defined by a lambda form.  The "``def``" form is\nactually more powerful since it allows the execution of multiple\nstatements and annotations.\n\n**Programmer\'s note:** Functions are first-class objects.  A "``def``"\nform executed inside a function definition defines a local function\nthat can be returned or passed around.  Free variables used in the\nnested function can access the local variables of the function\ncontaining the def.  See section *Naming and binding* for details.\n',
- 'global': '\nThe ``global`` statement\n************************\n\n   global_stmt ::= "global" identifier ("," identifier)*\n\nThe ``global`` statement is a declaration which holds for the entire\ncurrent code block.  It means that the listed identifiers are to be\ninterpreted as globals.  It would be impossible to assign to a global\nvariable without ``global``, although free variables may refer to\nglobals without being declared global.\n\nNames listed in a ``global`` statement must not be used in the same\ncode block textually preceding that ``global`` statement.\n\nNames listed in a ``global`` statement must not be defined as formal\nparameters or in a ``for`` loop control target, ``class`` definition,\nfunction definition, or ``import`` statement.\n\n(The current implementation does not enforce the latter two\nrestrictions, but programs should not abuse this freedom, as future\nimplementations may enforce them or silently change the meaning of the\nprogram.)\n\n**Programmer\'s note:** the ``global`` is a directive to the parser.\nIt applies only to code parsed at the same time as the ``global``\nstatement. In particular, a ``global`` statement contained in a string\nor code object supplied to the builtin ``exec()`` function does not\naffect the code block *containing* the function call, and code\ncontained in such a string is unaffected by ``global`` statements in\nthe code containing the function call.  The same applies to the\n``eval()`` and ``compile()`` functions.\n',
+ 'global': '\nThe ``global`` statement\n************************\n\n   global_stmt ::= "global" identifier ("," identifier)*\n\nThe ``global`` statement is a declaration which holds for the entire\ncurrent code block.  It means that the listed identifiers are to be\ninterpreted as globals.  It would be impossible to assign to a global\nvariable without ``global``, although free variables may refer to\nglobals without being declared global.\n\nNames listed in a ``global`` statement must not be used in the same\ncode block textually preceding that ``global`` statement.\n\nNames listed in a ``global`` statement must not be defined as formal\nparameters or in a ``for`` loop control target, ``class`` definition,\nfunction definition, or ``import`` statement.\n\n**CPython implementation detail:** The current implementation does not\nenforce the latter two restrictions, but programs should not abuse\nthis freedom, as future implementations may enforce them or silently\nchange the meaning of the program.\n\n**Programmer\'s note:** the ``global`` is a directive to the parser.\nIt applies only to code parsed at the same time as the ``global``\nstatement. In particular, a ``global`` statement contained in a string\nor code object supplied to the built-in ``exec()`` function does not\naffect the code block *containing* the function call, and code\ncontained in such a string is unaffected by ``global`` statements in\nthe code containing the function call.  The same applies to the\n``eval()`` and ``compile()`` functions.\n',
  'id-classes': '\nReserved classes of identifiers\n*******************************\n\nCertain classes of identifiers (besides keywords) have special\nmeanings.  These classes are identified by the patterns of leading and\ntrailing underscore characters:\n\n``_*``\n   Not imported by ``from module import *``.  The special identifier\n   ``_`` is used in the interactive interpreter to store the result of\n   the last evaluation; it is stored in the ``builtins`` module.  When\n   not in interactive mode, ``_`` has no special meaning and is not\n   defined. See section *The import statement*.\n\n   Note: The name ``_`` is often used in conjunction with\n     internationalization; refer to the documentation for the\n     ``gettext`` module for more information on this convention.\n\n``__*__``\n   System-defined names.  These names are defined by the interpreter\n   and its implementation (including the standard library);\n   applications should not expect to define additional names using\n   this convention.  The set of names of this class defined by Python\n   may be extended in future versions. See section *Special method\n   names*.\n\n``__*``\n   Class-private names.  Names in this category, when used within the\n   context of a class definition, are re-written to use a mangled form\n   to help avoid name clashes between "private" attributes of base and\n   derived classes. See section *Identifiers (Names)*.\n',
  'identifiers': '\nIdentifiers and keywords\n************************\n\nIdentifiers (also referred to as *names*) are described by the\nfollowing lexical definitions.\n\nThe syntax of identifiers in Python is based on the Unicode standard\nannex UAX-31, with elaboration and changes as defined below; see also\n**PEP 3131** for further details.\n\nWithin the ASCII range (U+0001..U+007F), the valid characters for\nidentifiers are the same as in Python 2.x: the uppercase and lowercase\nletters ``A`` through ``Z``, the underscore ``_`` and, except for the\nfirst character, the digits ``0`` through ``9``.\n\nPython 3.0 introduces additional characters from outside the ASCII\nrange (see **PEP 3131**).  For these characters, the classification\nuses the version of the Unicode Character Database as included in the\n``unicodedata`` module.\n\nIdentifiers are unlimited in length.  Case is significant.\n\n   identifier  ::= id_start id_continue*\n   id_start    ::= <all characters in general categories Lu, Ll, Lt, Lm, Lo, Nl, the underscore, and characters with the Other_ID_Start property>\n   id_continue ::= <all characters in id_start, plus characters in the categories Mn, Mc, Nd, Pc and others with the Other_ID_Continue property>\n\nThe Unicode category codes mentioned above stand for:\n\n* *Lu* - uppercase letters\n\n* *Ll* - lowercase letters\n\n* *Lt* - titlecase letters\n\n* *Lm* - modifier letters\n\n* *Lo* - other letters\n\n* *Nl* - letter numbers\n\n* *Mn* - nonspacing marks\n\n* *Mc* - spacing combining marks\n\n* *Nd* - decimal numbers\n\n* *Pc* - connector punctuations\n\nAll identifiers are converted into the normal form NFC while parsing;\ncomparison of identifiers is based on NFC.\n\nA non-normative HTML file listing all valid identifier characters for\nUnicode 4.1 can be found at http://www.dcl.hpi.uni-\npotsdam.de/home/loewis/table-3131.html.\n\n\nKeywords\n========\n\nThe following identifiers are used as reserved words, or *keywords* of\nthe language, and cannot be used as ordinary identifiers.  They must\nbe spelled exactly as written here:\n\n   False      class      finally    is         return\n   None       continue   for        lambda     try\n   True       def        from       nonlocal   while\n   and        del        global     not        with\n   as         elif       if         or         yield\n   assert     else       import     pass\n   break      except     in         raise\n\n\nReserved classes of identifiers\n===============================\n\nCertain classes of identifiers (besides keywords) have special\nmeanings.  These classes are identified by the patterns of leading and\ntrailing underscore characters:\n\n``_*``\n   Not imported by ``from module import *``.  The special identifier\n   ``_`` is used in the interactive interpreter to store the result of\n   the last evaluation; it is stored in the ``builtins`` module.  When\n   not in interactive mode, ``_`` has no special meaning and is not\n   defined. See section *The import statement*.\n\n   Note: The name ``_`` is often used in conjunction with\n     internationalization; refer to the documentation for the\n     ``gettext`` module for more information on this convention.\n\n``__*__``\n   System-defined names.  These names are defined by the interpreter\n   and its implementation (including the standard library);\n   applications should not expect to define additional names using\n   this convention.  The set of names of this class defined by Python\n   may be extended in future versions. See section *Special method\n   names*.\n\n``__*``\n   Class-private names.  Names in this category, when used within the\n   context of a class definition, are re-written to use a mangled form\n   to help avoid name clashes between "private" attributes of base and\n   derived classes. See section *Identifiers (Names)*.\n',
  'if': '\nThe ``if`` statement\n********************\n\nThe ``if`` statement is used for conditional execution:\n\n   if_stmt ::= "if" expression ":" suite\n               ( "elif" expression ":" suite )*\n               ["else" ":" suite]\n\nIt selects exactly one of the suites by evaluating the expressions one\nby one until one is found to be true (see section *Boolean operations*\nfor the definition of true and false); then that suite is executed\n(and no other part of the ``if`` statement is executed or evaluated).\nIf all expressions are false, the suite of the ``else`` clause, if\npresent, is executed.\n',
  'imaginary': '\nImaginary literals\n******************\n\nImaginary literals are described by the following lexical definitions:\n\n   imagnumber ::= (floatnumber | intpart) ("j" | "J")\n\nAn imaginary literal yields a complex number with a real part of 0.0.\nComplex numbers are represented as a pair of floating point numbers\nand have the same restrictions on their range.  To create a complex\nnumber with a nonzero real part, add a floating point number to it,\ne.g., ``(3+4j)``.  Some examples of imaginary literals:\n\n   3.14j   10.j    10j     .001j   1e100j  3.14e-10j\n',
- 'import': '\nThe ``import`` statement\n************************\n\n   import_stmt     ::= "import" module ["as" name] ( "," module ["as" name] )*\n                   | "from" relative_module "import" identifier ["as" name]\n                   ( "," identifier ["as" name] )*\n                   | "from" relative_module "import" "(" identifier ["as" name]\n                   ( "," identifier ["as" name] )* [","] ")"\n                   | "from" module "import" "*"\n   module          ::= (identifier ".")* identifier\n   relative_module ::= "."* module | "."+\n   name            ::= identifier\n\nImport statements are executed in two steps: (1) find a module, and\ninitialize it if necessary; (2) define a name or names in the local\nnamespace (of the scope where the ``import`` statement occurs). The\nstatement comes in two forms differing on whether it uses the ``from``\nkeyword. The first form (without ``from``) repeats these steps for\neach identifier in the list. The form with ``from`` performs step (1)\nonce, and then performs step (2) repeatedly. For a reference\nimplementation of step (1), see the ``importlib`` module.\n\nTo understand how step (1) occurs, one must first understand how\nPython handles hierarchical naming of modules. To help organize\nmodules and provide a hierarchy in naming, Python has a concept of\npackages. A package can contain other packages and modules while\nmodules cannot contain other modules or packages. From a file system\nperspective, packages are directories and modules are files. The\noriginal specification for packages is still available to read,\nalthough minor details have changed since the writing of that\ndocument.\n\nOnce the name of the module is known (unless otherwise specified, the\nterm "module" will refer to both packages and modules), searching for\nthe module or package can begin. The first place checked is\n``sys.modules``, the cache of all modules that have been imported\npreviously. If the module is found there then it is used in step (2)\nof import.\n\nIf the module is not found in the cache, then ``sys.meta_path`` is\nsearched (the specification for ``sys.meta_path`` can be found in\n**PEP 302**). The object is a list of *finder* objects which are\nqueried in order as to whether they know how to load the module by\ncalling their ``find_module()`` method with the name of the module. If\nthe module happens to be contained within a package (as denoted by the\nexistence of a dot in the name), then a second argument to\n``find_module()`` is given as the value of the ``__path__`` attribute\nfrom the parent package (everything up to the last dot in the name of\nthe module being imported). If a finder can find the module it returns\na *loader* (discussed later) or returns ``None``.\n\nIf none of the finders on ``sys.meta_path`` are able to find the\nmodule then some implicitly defined finders are queried.\nImplementations of Python vary in what implicit meta path finders are\ndefined. The one they all do define, though, is one that handles\n``sys.path_hooks``, ``sys.path_importer_cache``, and ``sys.path``.\n\nThe implicit finder searches for the requested module in the "paths"\nspecified in one of two places ("paths" do not have to be file system\npaths). If the module being imported is supposed to be contained\nwithin a package then the second argument passed to ``find_module()``,\n``__path__`` on the parent package, is used as the source of paths. If\nthe module is not contained in a package then ``sys.path`` is used as\nthe source of paths.\n\nOnce the source of paths is chosen it is iterated over to find a\nfinder that can handle that path. The dict at\n``sys.path_importer_cache`` caches finders for paths and is checked\nfor a finder. If the path does not have a finder cached then\n``sys.path_hooks`` is searched by calling each object in the list with\na single argument of the path, returning a finder or raises\n``ImportError``. If a finder is returned then it is cached in\n``sys.path_importer_cache`` and then used for that path entry. If no\nfinder can be found but the path exists then a value of ``None`` is\nstored in ``sys.path_importer_cache`` to signify that an implicit,\nfile-based finder that handles modules stored as individual files\nshould be used for that path. If the path does not exist then a finder\nwhich always returns ``None`` is placed in the cache for the path.\n\nIf no finder can find the module then ``ImportError`` is raised.\nOtherwise some finder returned a loader whose ``load_module()`` method\nis called with the name of the module to load (see **PEP 302** for the\noriginal definition of loaders). A loader has several responsibilities\nto perform on a module it loads. First, if the module already exists\nin ``sys.modules`` (a possibility if the loader is called outside of\nthe import machinery) then it is to use that module for initialization\nand not a new module. But if the module does not exist in\n``sys.modules`` then it is to be added to that dict before\ninitialization begins. If an error occurs during loading of the module\nand it was added to ``sys.modules`` it is to be removed from the dict.\nIf an error occurs but the module was already in ``sys.modules`` it is\nleft in the dict.\n\nThe loader must set several attributes on the module. ``__name__`` is\nto be set to the name of the module. ``__file__`` is to be the "path"\nto the file unless the module is built-in (and thus listed in\n``sys.builtin_module_names``) in which case the attribute is not set.\nIf what is being imported is a package then ``__path__`` is to be set\nto a list of paths to be searched when looking for modules and\npackages contained within the package being imported. ``__package__``\nis optional but should be set to the name of package that contains the\nmodule or package (the empty string is used for module not contained\nin a package). ``__loader__`` is also optional but should be set to\nthe loader object that is loading the module.\n\nIf an error occurs during loading then the loader raises\n``ImportError`` if some other exception is not already being\npropagated. Otherwise the loader returns the module that was loaded\nand initialized.\n\nWhen step (1) finishes without raising an exception, step (2) can\nbegin.\n\nThe first form of ``import`` statement binds the module name in the\nlocal namespace to the module object, and then goes on to import the\nnext identifier, if any.  If the module name is followed by ``as``,\nthe name following ``as`` is used as the local name for the module.\n\nThe ``from`` form does not bind the module name: it goes through the\nlist of identifiers, looks each one of them up in the module found in\nstep (1), and binds the name in the local namespace to the object thus\nfound.  As with the first form of ``import``, an alternate local name\ncan be supplied by specifying "``as`` localname".  If a name is not\nfound, ``ImportError`` is raised.  If the list of identifiers is\nreplaced by a star (``\'*\'``), all public names defined in the module\nare bound in the local namespace of the ``import`` statement..\n\nThe *public names* defined by a module are determined by checking the\nmodule\'s namespace for a variable named ``__all__``; if defined, it\nmust be a sequence of strings which are names defined or imported by\nthat module.  The names given in ``__all__`` are all considered public\nand are required to exist.  If ``__all__`` is not defined, the set of\npublic names includes all names found in the module\'s namespace which\ndo not begin with an underscore character (``\'_\'``). ``__all__``\nshould contain the entire public API. It is intended to avoid\naccidentally exporting items that are not part of the API (such as\nlibrary modules which were imported and used within the module).\n\nThe ``from`` form with ``*`` may only occur in a module scope.  The\nwild card form of import --- ``import *`` --- is only allowed at the\nmodule level. Attempting to use it in class for function definitions\nwill raise a ``SyntaxError``.\n\nWhen specifying what module to import you do not have to specify the\nabsolute name of the module. When a module or package is contained\nwithin another package it is possible to make a relative import within\nthe same top package without having to mention the package name. By\nusing leading dots in the specified module or package after ``from``\nyou can specify how high to traverse up the current package hierarchy\nwithout specifying exact names. One leading dot means the current\npackage where the module making the import exists. Two dots means up\none package level. Three dots is up two levels, etc. So if you execute\n``from . import mod`` from a module in the ``pkg`` package then you\nwill end up importing ``pkg.mod``. If you execute ``from ..subpkg2\nimprt mod`` from within ``pkg.subpkg1`` you will import\n``pkg.subpkg2.mod``. The specification for relative imports is\ncontained within **PEP 328**.\n\n``importlib.import_module()`` is provided to support applications that\ndetermine which modules need to be loaded dynamically.\n\n\nFuture statements\n=================\n\nA *future statement* is a directive to the compiler that a particular\nmodule should be compiled using syntax or semantics that will be\navailable in a specified future release of Python.  The future\nstatement is intended to ease migration to future versions of Python\nthat introduce incompatible changes to the language.  It allows use of\nthe new features on a per-module basis before the release in which the\nfeature becomes standard.\n\n   future_statement ::= "from" "__future__" "import" feature ["as" name]\n                        ("," feature ["as" name])*\n                        | "from" "__future__" "import" "(" feature ["as" name]\n                        ("," feature ["as" name])* [","] ")"\n   feature          ::= identifier\n   name             ::= identifier\n\nA future statement must appear near the top of the module.  The only\nlines that can appear before a future statement are:\n\n* the module docstring (if any),\n\n* comments,\n\n* blank lines, and\n\n* other future statements.\n\nThe features recognized by Python 3.0 are ``absolute_import``,\n``division``, ``generators``, ``unicode_literals``,\n``print_function``, ``nested_scopes`` and ``with_statement``.  They\nare all redundant because they are always enabled, and only kept for\nbackwards compatibility.\n\nA future statement is recognized and treated specially at compile\ntime: Changes to the semantics of core constructs are often\nimplemented by generating different code.  It may even be the case\nthat a new feature introduces new incompatible syntax (such as a new\nreserved word), in which case the compiler may need to parse the\nmodule differently.  Such decisions cannot be pushed off until\nruntime.\n\nFor any given release, the compiler knows which feature names have\nbeen defined, and raises a compile-time error if a future statement\ncontains a feature not known to it.\n\nThe direct runtime semantics are the same as for any import statement:\nthere is a standard module ``__future__``, described later, and it\nwill be imported in the usual way at the time the future statement is\nexecuted.\n\nThe interesting runtime semantics depend on the specific feature\nenabled by the future statement.\n\nNote that there is nothing special about the statement:\n\n   import __future__ [as name]\n\nThat is not a future statement; it\'s an ordinary import statement with\nno special semantics or syntax restrictions.\n\nCode compiled by calls to the builtin functions ``exec()`` and\n``compile()`` that occur in a module ``M`` containing a future\nstatement will, by default, use the new syntax or semantics associated\nwith the future statement.  This can be controlled by optional\narguments to ``compile()`` --- see the documentation of that function\nfor details.\n\nA future statement typed at an interactive interpreter prompt will\ntake effect for the rest of the interpreter session.  If an\ninterpreter is started with the *-i* option, is passed a script name\nto execute, and the script includes a future statement, it will be in\neffect in the interactive session started after the script is\nexecuted.\n\nSee also:\n\n   **PEP 236** - Back to the __future__\n      The original proposal for the __future__ mechanism.\n',
- 'in': '\nComparisons\n***********\n\nUnlike C, all comparison operations in Python have the same priority,\nwhich is lower than that of any arithmetic, shifting or bitwise\noperation.  Also unlike C, expressions like ``a < b < c`` have the\ninterpretation that is conventional in mathematics:\n\n   comparison    ::= or_expr ( comp_operator or_expr )*\n   comp_operator ::= "<" | ">" | "==" | ">=" | "<=" | "!="\n                     | "is" ["not"] | ["not"] "in"\n\nComparisons yield boolean values: ``True`` or ``False``.\n\nComparisons can be chained arbitrarily, e.g., ``x < y <= z`` is\nequivalent to ``x < y and y <= z``, except that ``y`` is evaluated\nonly once (but in both cases ``z`` is not evaluated at all when ``x <\ny`` is found to be false).\n\nFormally, if *a*, *b*, *c*, ..., *y*, *z* are expressions and *op1*,\n*op2*, ..., *opN* are comparison operators, then ``a op1 b op2 c ... y\nopN z`` is equivalent to ``a op1 b and b op2 c and ... y opN z``,\nexcept that each expression is evaluated at most once.\n\nNote that ``a op1 b op2 c`` doesn\'t imply any kind of comparison\nbetween *a* and *c*, so that, e.g., ``x < y > z`` is perfectly legal\n(though perhaps not pretty).\n\nThe operators ``<``, ``>``, ``==``, ``>=``, ``<=``, and ``!=`` compare\nthe values of two objects.  The objects need not have the same type.\nIf both are numbers, they are converted to a common type.  Otherwise,\nthe ``==`` and ``!=`` operators *always* consider objects of different\ntypes to be unequal, while the ``<``, ``>``, ``>=`` and ``<=``\noperators raise a ``TypeError`` when comparing objects of different\ntypes that do not implement these operators for the given pair of\ntypes.  You can control comparison behavior of objects of non-builtin\ntypes by defining rich comparison methods like ``__gt__()``, described\nin section *Basic customization*.\n\nComparison of objects of the same type depends on the type:\n\n* Numbers are compared arithmetically.\n\n* The values ``float(\'NaN\')`` and ``Decimal(\'NaN\')`` are special. The\n  are identical to themselves, ``x is x`` but are not equal to\n  themselves, ``x != x``.  Additionally, comparing any value to a\n  not-a-number value will return ``False``.  For example, both ``3 <\n  float(\'NaN\')`` and ``float(\'NaN\') < 3`` will return ``False``.\n\n* Bytes objects are compared lexicographically using the numeric\n  values of their elements.\n\n* Strings are compared lexicographically using the numeric equivalents\n  (the result of the built-in function ``ord()``) of their characters.\n  [3] String and bytes object can\'t be compared!\n\n* Tuples and lists are compared lexicographically using comparison of\n  corresponding elements.  This means that to compare equal, each\n  element must compare equal and the two sequences must be of the same\n  type and have the same length.\n\n  If not equal, the sequences are ordered the same as their first\n  differing elements.  For example, ``[1,2,x] <= [1,2,y]`` has the\n  same value as ``x <= y``.  If the corresponding element does not\n  exist, the shorter sequence is ordered first (for example, ``[1,2] <\n  [1,2,3]``).\n\n* Mappings (dictionaries) compare equal if and only if their sorted\n  ``(key, value)`` lists compare equal. [4] Outcomes other than\n  equality are resolved consistently, but are not otherwise defined.\n  [5]\n\n* Sets and frozensets define comparison operators to mean subset and\n  superset tests.  Those relations do not define total orderings (the\n  two sets ``{1,2}`` and {2,3} are not equal, nor subsets of one\n  another, nor supersets of one another).  Accordingly, sets are not\n  appropriate arguments for functions which depend on total ordering.\n  For example, ``min()``, ``max()``, and ``sorted()`` produce\n  undefined results given a list of sets as inputs.\n\n* Most other objects of builtin types compare unequal unless they are\n  the same object; the choice whether one object is considered smaller\n  or larger than another one is made arbitrarily but consistently\n  within one execution of a program.\n\nComparison of objects of the differing types depends on whether either\nof the types provide explicit support for the comparison.  Most\nnumeric types can be compared with one another, but comparisons of\n``float`` and ``Decimal`` are not supported to avoid the inevitable\nconfusion arising from representation issues such as ``float(\'1.1\')``\nbeing inexactly represented and therefore not exactly equal to\n``Decimal(\'1.1\')`` which is.  When cross-type comparison is not\nsupported, the comparison method returns ``NotImplemented``.  This can\ncreate the illusion of non-transitivity between supported cross-type\ncomparisons and unsupported comparisons.  For example, ``Decimal(2) ==\n2`` and *2 == float(2)`* but ``Decimal(2) != float(2)``.\n\nThe operators ``in`` and ``not in`` test for membership.  ``x in s``\nevaluates to true if *x* is a member of *s*, and false otherwise.  ``x\nnot in s`` returns the negation of ``x in s``.  All built-in sequences\nand set types support this as well as dictionary, for which ``in``\ntests whether a the dictionary has a given key. For container types\nsuch as list, tuple, set, frozenset, dict, or collections.deque, the\nexpression ``x in y`` is equivalent to ``any(x is e or x == e for val\ne in y)``.\n\nFor the string and bytes types, ``x in y`` is true if and only if *x*\nis a substring of *y*.  An equivalent test is ``y.find(x) != -1``.\nEmpty strings are always considered to be a substring of any other\nstring, so ``"" in "abc"`` will return ``True``.\n\nFor user-defined classes which define the ``__contains__()`` method,\n``x in y`` is true if and only if ``y.__contains__(x)`` is true.\n\nFor user-defined classes which do not define ``__contains__()`` and do\ndefine ``__getitem__()``, ``x in y`` is true if and only if there is a\nnon-negative integer index *i* such that ``x == y[i]``, and all lower\ninteger indices do not raise ``IndexError`` exception.  (If any other\nexception is raised, it is as if ``in`` raised that exception).\n\nThe operator ``not in`` is defined to have the inverse true value of\n``in``.\n\nThe operators ``is`` and ``is not`` test for object identity: ``x is\ny`` is true if and only if *x* and *y* are the same object.  ``x is\nnot y`` yields the inverse truth value. [6]\n',
+ 'import': '\nThe ``import`` statement\n************************\n\n   import_stmt     ::= "import" module ["as" name] ( "," module ["as" name] )*\n                   | "from" relative_module "import" identifier ["as" name]\n                   ( "," identifier ["as" name] )*\n                   | "from" relative_module "import" "(" identifier ["as" name]\n                   ( "," identifier ["as" name] )* [","] ")"\n                   | "from" module "import" "*"\n   module          ::= (identifier ".")* identifier\n   relative_module ::= "."* module | "."+\n   name            ::= identifier\n\nImport statements are executed in two steps: (1) find a module, and\ninitialize it if necessary; (2) define a name or names in the local\nnamespace (of the scope where the ``import`` statement occurs). The\nstatement comes in two forms differing on whether it uses the ``from``\nkeyword. The first form (without ``from``) repeats these steps for\neach identifier in the list. The form with ``from`` performs step (1)\nonce, and then performs step (2) repeatedly. For a reference\nimplementation of step (1), see the ``importlib`` module.\n\nTo understand how step (1) occurs, one must first understand how\nPython handles hierarchical naming of modules. To help organize\nmodules and provide a hierarchy in naming, Python has a concept of\npackages. A package can contain other packages and modules while\nmodules cannot contain other modules or packages. From a file system\nperspective, packages are directories and modules are files. The\noriginal specification for packages is still available to read,\nalthough minor details have changed since the writing of that\ndocument.\n\nOnce the name of the module is known (unless otherwise specified, the\nterm "module" will refer to both packages and modules), searching for\nthe module or package can begin. The first place checked is\n``sys.modules``, the cache of all modules that have been imported\npreviously. If the module is found there then it is used in step (2)\nof import unless ``None`` is found in ``sys.modules``, in which case\n``ImportError`` is raised.\n\nIf the module is not found in the cache, then ``sys.meta_path`` is\nsearched (the specification for ``sys.meta_path`` can be found in\n**PEP 302**). The object is a list of *finder* objects which are\nqueried in order as to whether they know how to load the module by\ncalling their ``find_module()`` method with the name of the module. If\nthe module happens to be contained within a package (as denoted by the\nexistence of a dot in the name), then a second argument to\n``find_module()`` is given as the value of the ``__path__`` attribute\nfrom the parent package (everything up to the last dot in the name of\nthe module being imported). If a finder can find the module it returns\na *loader* (discussed later) or returns ``None``.\n\nIf none of the finders on ``sys.meta_path`` are able to find the\nmodule then some implicitly defined finders are queried.\nImplementations of Python vary in what implicit meta path finders are\ndefined. The one they all do define, though, is one that handles\n``sys.path_hooks``, ``sys.path_importer_cache``, and ``sys.path``.\n\nThe implicit finder searches for the requested module in the "paths"\nspecified in one of two places ("paths" do not have to be file system\npaths). If the module being imported is supposed to be contained\nwithin a package then the second argument passed to ``find_module()``,\n``__path__`` on the parent package, is used as the source of paths. If\nthe module is not contained in a package then ``sys.path`` is used as\nthe source of paths.\n\nOnce the source of paths is chosen it is iterated over to find a\nfinder that can handle that path. The dict at\n``sys.path_importer_cache`` caches finders for paths and is checked\nfor a finder. If the path does not have a finder cached then\n``sys.path_hooks`` is searched by calling each object in the list with\na single argument of the path, returning a finder or raises\n``ImportError``. If a finder is returned then it is cached in\n``sys.path_importer_cache`` and then used for that path entry. If no\nfinder can be found but the path exists then a value of ``None`` is\nstored in ``sys.path_importer_cache`` to signify that an implicit,\nfile-based finder that handles modules stored as individual files\nshould be used for that path. If the path does not exist then a finder\nwhich always returns ``None`` is placed in the cache for the path.\n\nIf no finder can find the module then ``ImportError`` is raised.\nOtherwise some finder returned a loader whose ``load_module()`` method\nis called with the name of the module to load (see **PEP 302** for the\noriginal definition of loaders). A loader has several responsibilities\nto perform on a module it loads. First, if the module already exists\nin ``sys.modules`` (a possibility if the loader is called outside of\nthe import machinery) then it is to use that module for initialization\nand not a new module. But if the module does not exist in\n``sys.modules`` then it is to be added to that dict before\ninitialization begins. If an error occurs during loading of the module\nand it was added to ``sys.modules`` it is to be removed from the dict.\nIf an error occurs but the module was already in ``sys.modules`` it is\nleft in the dict.\n\nThe loader must set several attributes on the module. ``__name__`` is\nto be set to the name of the module. ``__file__`` is to be the "path"\nto the file unless the module is built-in (and thus listed in\n``sys.builtin_module_names``) in which case the attribute is not set.\nIf what is being imported is a package then ``__path__`` is to be set\nto a list of paths to be searched when looking for modules and\npackages contained within the package being imported. ``__package__``\nis optional but should be set to the name of package that contains the\nmodule or package (the empty string is used for module not contained\nin a package). ``__loader__`` is also optional but should be set to\nthe loader object that is loading the module.\n\nIf an error occurs during loading then the loader raises\n``ImportError`` if some other exception is not already being\npropagated. Otherwise the loader returns the module that was loaded\nand initialized.\n\nWhen step (1) finishes without raising an exception, step (2) can\nbegin.\n\nThe first form of ``import`` statement binds the module name in the\nlocal namespace to the module object, and then goes on to import the\nnext identifier, if any.  If the module name is followed by ``as``,\nthe name following ``as`` is used as the local name for the module.\n\nThe ``from`` form does not bind the module name: it goes through the\nlist of identifiers, looks each one of them up in the module found in\nstep (1), and binds the name in the local namespace to the object thus\nfound.  As with the first form of ``import``, an alternate local name\ncan be supplied by specifying "``as`` localname".  If a name is not\nfound, ``ImportError`` is raised.  If the list of identifiers is\nreplaced by a star (``\'*\'``), all public names defined in the module\nare bound in the local namespace of the ``import`` statement..\n\nThe *public names* defined by a module are determined by checking the\nmodule\'s namespace for a variable named ``__all__``; if defined, it\nmust be a sequence of strings which are names defined or imported by\nthat module.  The names given in ``__all__`` are all considered public\nand are required to exist.  If ``__all__`` is not defined, the set of\npublic names includes all names found in the module\'s namespace which\ndo not begin with an underscore character (``\'_\'``). ``__all__``\nshould contain the entire public API. It is intended to avoid\naccidentally exporting items that are not part of the API (such as\nlibrary modules which were imported and used within the module).\n\nThe ``from`` form with ``*`` may only occur in a module scope.  The\nwild card form of import --- ``import *`` --- is only allowed at the\nmodule level. Attempting to use it in class or function definitions\nwill raise a ``SyntaxError``.\n\nWhen specifying what module to import you do not have to specify the\nabsolute name of the module. When a module or package is contained\nwithin another package it is possible to make a relative import within\nthe same top package without having to mention the package name. By\nusing leading dots in the specified module or package after ``from``\nyou can specify how high to traverse up the current package hierarchy\nwithout specifying exact names. One leading dot means the current\npackage where the module making the import exists. Two dots means up\none package level. Three dots is up two levels, etc. So if you execute\n``from . import mod`` from a module in the ``pkg`` package then you\nwill end up importing ``pkg.mod``. If you execute ``from ..subpkg2\nimprt mod`` from within ``pkg.subpkg1`` you will import\n``pkg.subpkg2.mod``. The specification for relative imports is\ncontained within **PEP 328**.\n\n``importlib.import_module()`` is provided to support applications that\ndetermine which modules need to be loaded dynamically.\n\n\nFuture statements\n=================\n\nA *future statement* is a directive to the compiler that a particular\nmodule should be compiled using syntax or semantics that will be\navailable in a specified future release of Python.  The future\nstatement is intended to ease migration to future versions of Python\nthat introduce incompatible changes to the language.  It allows use of\nthe new features on a per-module basis before the release in which the\nfeature becomes standard.\n\n   future_statement ::= "from" "__future__" "import" feature ["as" name]\n                        ("," feature ["as" name])*\n                        | "from" "__future__" "import" "(" feature ["as" name]\n                        ("," feature ["as" name])* [","] ")"\n   feature          ::= identifier\n   name             ::= identifier\n\nA future statement must appear near the top of the module.  The only\nlines that can appear before a future statement are:\n\n* the module docstring (if any),\n\n* comments,\n\n* blank lines, and\n\n* other future statements.\n\nThe features recognized by Python 3.0 are ``absolute_import``,\n``division``, ``generators``, ``unicode_literals``,\n``print_function``, ``nested_scopes`` and ``with_statement``.  They\nare all redundant because they are always enabled, and only kept for\nbackwards compatibility.\n\nA future statement is recognized and treated specially at compile\ntime: Changes to the semantics of core constructs are often\nimplemented by generating different code.  It may even be the case\nthat a new feature introduces new incompatible syntax (such as a new\nreserved word), in which case the compiler may need to parse the\nmodule differently.  Such decisions cannot be pushed off until\nruntime.\n\nFor any given release, the compiler knows which feature names have\nbeen defined, and raises a compile-time error if a future statement\ncontains a feature not known to it.\n\nThe direct runtime semantics are the same as for any import statement:\nthere is a standard module ``__future__``, described later, and it\nwill be imported in the usual way at the time the future statement is\nexecuted.\n\nThe interesting runtime semantics depend on the specific feature\nenabled by the future statement.\n\nNote that there is nothing special about the statement:\n\n   import __future__ [as name]\n\nThat is not a future statement; it\'s an ordinary import statement with\nno special semantics or syntax restrictions.\n\nCode compiled by calls to the built-in functions ``exec()`` and\n``compile()`` that occur in a module ``M`` containing a future\nstatement will, by default, use the new syntax or semantics associated\nwith the future statement.  This can be controlled by optional\narguments to ``compile()`` --- see the documentation of that function\nfor details.\n\nA future statement typed at an interactive interpreter prompt will\ntake effect for the rest of the interpreter session.  If an\ninterpreter is started with the *-i* option, is passed a script name\nto execute, and the script includes a future statement, it will be in\neffect in the interactive session started after the script is\nexecuted.\n\nSee also:\n\n   **PEP 236** - Back to the __future__\n      The original proposal for the __future__ mechanism.\n',
+ 'in': '\nComparisons\n***********\n\nUnlike C, all comparison operations in Python have the same priority,\nwhich is lower than that of any arithmetic, shifting or bitwise\noperation.  Also unlike C, expressions like ``a < b < c`` have the\ninterpretation that is conventional in mathematics:\n\n   comparison    ::= or_expr ( comp_operator or_expr )*\n   comp_operator ::= "<" | ">" | "==" | ">=" | "<=" | "!="\n                     | "is" ["not"] | ["not"] "in"\n\nComparisons yield boolean values: ``True`` or ``False``.\n\nComparisons can be chained arbitrarily, e.g., ``x < y <= z`` is\nequivalent to ``x < y and y <= z``, except that ``y`` is evaluated\nonly once (but in both cases ``z`` is not evaluated at all when ``x <\ny`` is found to be false).\n\nFormally, if *a*, *b*, *c*, ..., *y*, *z* are expressions and *op1*,\n*op2*, ..., *opN* are comparison operators, then ``a op1 b op2 c ... y\nopN z`` is equivalent to ``a op1 b and b op2 c and ... y opN z``,\nexcept that each expression is evaluated at most once.\n\nNote that ``a op1 b op2 c`` doesn\'t imply any kind of comparison\nbetween *a* and *c*, so that, e.g., ``x < y > z`` is perfectly legal\n(though perhaps not pretty).\n\nThe operators ``<``, ``>``, ``==``, ``>=``, ``<=``, and ``!=`` compare\nthe values of two objects.  The objects need not have the same type.\nIf both are numbers, they are converted to a common type.  Otherwise,\nthe ``==`` and ``!=`` operators *always* consider objects of different\ntypes to be unequal, while the ``<``, ``>``, ``>=`` and ``<=``\noperators raise a ``TypeError`` when comparing objects of different\ntypes that do not implement these operators for the given pair of\ntypes.  You can control comparison behavior of objects of non-built-in\ntypes by defining rich comparison methods like ``__gt__()``, described\nin section *Basic customization*.\n\nComparison of objects of the same type depends on the type:\n\n* Numbers are compared arithmetically.\n\n* The values ``float(\'NaN\')`` and ``Decimal(\'NaN\')`` are special. The\n  are identical to themselves, ``x is x`` but are not equal to\n  themselves, ``x != x``.  Additionally, comparing any value to a\n  not-a-number value will return ``False``.  For example, both ``3 <\n  float(\'NaN\')`` and ``float(\'NaN\') < 3`` will return ``False``.\n\n* Bytes objects are compared lexicographically using the numeric\n  values of their elements.\n\n* Strings are compared lexicographically using the numeric equivalents\n  (the result of the built-in function ``ord()``) of their characters.\n  [3] String and bytes object can\'t be compared!\n\n* Tuples and lists are compared lexicographically using comparison of\n  corresponding elements.  This means that to compare equal, each\n  element must compare equal and the two sequences must be of the same\n  type and have the same length.\n\n  If not equal, the sequences are ordered the same as their first\n  differing elements.  For example, ``[1,2,x] <= [1,2,y]`` has the\n  same value as ``x <= y``.  If the corresponding element does not\n  exist, the shorter sequence is ordered first (for example, ``[1,2] <\n  [1,2,3]``).\n\n* Mappings (dictionaries) compare equal if and only if they have the\n  same ``(key, value)`` pairs. Order comparisons ``(\'<\', \'<=\', \'>=\',\n  \'>\')`` raise ``TypeError``.\n\n* Sets and frozensets define comparison operators to mean subset and\n  superset tests.  Those relations do not define total orderings (the\n  two sets ``{1,2}`` and {2,3} are not equal, nor subsets of one\n  another, nor supersets of one another).  Accordingly, sets are not\n  appropriate arguments for functions which depend on total ordering.\n  For example, ``min()``, ``max()``, and ``sorted()`` produce\n  undefined results given a list of sets as inputs.\n\n* Most other objects of built-in types compare unequal unless they are\n  the same object; the choice whether one object is considered smaller\n  or larger than another one is made arbitrarily but consistently\n  within one execution of a program.\n\nComparison of objects of the differing types depends on whether either\nof the types provide explicit support for the comparison.  Most\nnumeric types can be compared with one another, but comparisons of\n``float`` and ``Decimal`` are not supported to avoid the inevitable\nconfusion arising from representation issues such as ``float(\'1.1\')``\nbeing inexactly represented and therefore not exactly equal to\n``Decimal(\'1.1\')`` which is.  When cross-type comparison is not\nsupported, the comparison method returns ``NotImplemented``.  This can\ncreate the illusion of non-transitivity between supported cross-type\ncomparisons and unsupported comparisons.  For example, ``Decimal(2) ==\n2`` and *2 == float(2)`* but ``Decimal(2) != float(2)``.\n\nThe operators ``in`` and ``not in`` test for membership.  ``x in s``\nevaluates to true if *x* is a member of *s*, and false otherwise.  ``x\nnot in s`` returns the negation of ``x in s``.  All built-in sequences\nand set types support this as well as dictionary, for which ``in``\ntests whether a the dictionary has a given key. For container types\nsuch as list, tuple, set, frozenset, dict, or collections.deque, the\nexpression ``x in y`` is equivalent to ``any(x is e or x == e for e in\ny)``.\n\nFor the string and bytes types, ``x in y`` is true if and only if *x*\nis a substring of *y*.  An equivalent test is ``y.find(x) != -1``.\nEmpty strings are always considered to be a substring of any other\nstring, so ``"" in "abc"`` will return ``True``.\n\nFor user-defined classes which define the ``__contains__()`` method,\n``x in y`` is true if and only if ``y.__contains__(x)`` is true.\n\nFor user-defined classes which do not define ``__contains__()`` but do\ndefine ``__iter__()``, ``x in y`` is true if some value ``z`` with ``x\n== z`` is produced while iterating over ``y``.  If an exception is\nraised during the iteration, it is as if ``in`` raised that exception.\n\nLastly, the old-style iteration protocol is tried: if a class defines\n``__getitem__()``, ``x in y`` is true if and only if there is a non-\nnegative integer index *i* such that ``x == y[i]``, and all lower\ninteger indices do not raise ``IndexError`` exception.  (If any other\nexception is raised, it is as if ``in`` raised that exception).\n\nThe operator ``not in`` is defined to have the inverse true value of\n``in``.\n\nThe operators ``is`` and ``is not`` test for object identity: ``x is\ny`` is true if and only if *x* and *y* are the same object.  ``x is\nnot y`` yields the inverse truth value. [4]\n',
  'integers': '\nInteger literals\n****************\n\nInteger literals are described by the following lexical definitions:\n\n   integer        ::= decimalinteger | octinteger | hexinteger | bininteger\n   decimalinteger ::= nonzerodigit digit* | "0"+\n   nonzerodigit   ::= "1"..."9"\n   digit          ::= "0"..."9"\n   octinteger     ::= "0" ("o" | "O") octdigit+\n   hexinteger     ::= "0" ("x" | "X") hexdigit+\n   bininteger     ::= "0" ("b" | "B") bindigit+\n   octdigit       ::= "0"..."7"\n   hexdigit       ::= digit | "a"..."f" | "A"..."F"\n   bindigit       ::= "0" | "1"\n\nThere is no limit for the length of integer literals apart from what\ncan be stored in available memory.\n\nNote that leading zeros in a non-zero decimal number are not allowed.\nThis is for disambiguation with C-style octal literals, which Python\nused before version 3.0.\n\nSome examples of integer literals:\n\n   7     2147483647                        0o177    0b100110111\n   3     79228162514264337593543950336     0o377    0x100000000\n         79228162514264337593543950336              0xdeadbeef\n',
  'lambda': '\nLambdas\n*******\n\n   lambda_form        ::= "lambda" [parameter_list]: expression\n   lambda_form_nocond ::= "lambda" [parameter_list]: expression_nocond\n\nLambda forms (lambda expressions) have the same syntactic position as\nexpressions.  They are a shorthand to create anonymous functions; the\nexpression ``lambda arguments: expression`` yields a function object.\nThe unnamed object behaves like a function object defined with\n\n   def <lambda>(arguments):\n       return expression\n\nSee section *Function definitions* for the syntax of parameter lists.\nNote that functions created with lambda forms cannot contain\nstatements or annotations.\n',
  'lists': '\nList displays\n*************\n\nA list display is a possibly empty series of expressions enclosed in\nsquare brackets:\n\n   list_display ::= "[" [expression_list | comprehension] "]"\n\nA list display yields a new list object, the contents being specified\nby either a list of expressions or a comprehension.  When a comma-\nseparated list of expressions is supplied, its elements are evaluated\nfrom left to right and placed into the list object in that order.\nWhen a comprehension is supplied, the list is constructed from the\nelements resulting from the comprehension.\n',
- 'naming': "\nNaming and binding\n******************\n\n*Names* refer to objects.  Names are introduced by name binding\noperations. Each occurrence of a name in the program text refers to\nthe *binding* of that name established in the innermost function block\ncontaining the use.\n\nA *block* is a piece of Python program text that is executed as a\nunit. The following are blocks: a module, a function body, and a class\ndefinition. Each command typed interactively is a block.  A script\nfile (a file given as standard input to the interpreter or specified\non the interpreter command line the first argument) is a code block.\nA script command (a command specified on the interpreter command line\nwith the '**-c**' option) is a code block.  The string argument passed\nto the built-in functions ``eval()`` and ``exec()`` is a code block.\n\nA code block is executed in an *execution frame*.  A frame contains\nsome administrative information (used for debugging) and determines\nwhere and how execution continues after the code block's execution has\ncompleted.\n\nA *scope* defines the visibility of a name within a block.  If a local\nvariable is defined in a block, its scope includes that block.  If the\ndefinition occurs in a function block, the scope extends to any blocks\ncontained within the defining one, unless a contained block introduces\na different binding for the name.  The scope of names defined in a\nclass block is limited to the class block; it does not extend to the\ncode blocks of methods -- this includes comprehensions and generator\nexpressions since they are implemented using a function scope.  This\nmeans that the following will fail:\n\n   class A:\n       a = 42\n       b = list(a + i for i in range(10))\n\nWhen a name is used in a code block, it is resolved using the nearest\nenclosing scope.  The set of all such scopes visible to a code block\nis called the block's *environment*.\n\nIf a name is bound in a block, it is a local variable of that block,\nunless declared as ``nonlocal``.  If a name is bound at the module\nlevel, it is a global variable.  (The variables of the module code\nblock are local and global.)  If a variable is used in a code block\nbut not defined there, it is a *free variable*.\n\nWhen a name is not found at all, a ``NameError`` exception is raised.\nIf the name refers to a local variable that has not been bound, a\n``UnboundLocalError`` exception is raised.  ``UnboundLocalError`` is a\nsubclass of ``NameError``.\n\nThe following constructs bind names: formal parameters to functions,\n``import`` statements, class and function definitions (these bind the\nclass or function name in the defining block), and targets that are\nidentifiers if occurring in an assignment, ``for`` loop header, or\nafter ``as`` in a ``with`` statement or :keyword.`except` clause. The\n``import`` statement of the form ``from ... import *`` binds all names\ndefined in the imported module, except those beginning with an\nunderscore.  This form may only be used at the module level.\n\nA target occurring in a ``del`` statement is also considered bound for\nthis purpose (though the actual semantics are to unbind the name).  It\nis illegal to unbind a name that is referenced by an enclosing scope;\nthe compiler will report a ``SyntaxError``.\n\nEach assignment or import statement occurs within a block defined by a\nclass or function definition or at the module level (the top-level\ncode block).\n\nIf a name binding operation occurs anywhere within a code block, all\nuses of the name within the block are treated as references to the\ncurrent block.  This can lead to errors when a name is used within a\nblock before it is bound.  This rule is subtle.  Python lacks\ndeclarations and allows name binding operations to occur anywhere\nwithin a code block.  The local variables of a code block can be\ndetermined by scanning the entire text of the block for name binding\noperations.\n\nIf the ``global`` statement occurs within a block, all uses of the\nname specified in the statement refer to the binding of that name in\nthe top-level namespace.  Names are resolved in the top-level\nnamespace by searching the global namespace, i.e. the namespace of the\nmodule containing the code block, and the builtin namespace, the\nnamespace of the module ``builtins``.  The global namespace is\nsearched first.  If the name is not found there, the builtin namespace\nis searched.  The global statement must precede all uses of the name.\n\nThe built-in namespace associated with the execution of a code block\nis actually found by looking up the name ``__builtins__`` in its\nglobal namespace; this should be a dictionary or a module (in the\nlatter case the module's dictionary is used).  By default, when in the\n``__main__`` module, ``__builtins__`` is the built-in module\n``builtins``; when in any other module, ``__builtins__`` is an alias\nfor the dictionary of the ``builtins`` module itself.\n``__builtins__`` can be set to a user-created dictionary to create a\nweak form of restricted execution.\n\nNote: Users should not touch ``__builtins__``; it is strictly an\n  implementation detail.  Users wanting to override values in the\n  built-in namespace should ``import`` the ``builtins`` module and\n  modify its attributes appropriately.\n\nThe namespace for a module is automatically created the first time a\nmodule is imported.  The main module for a script is always called\n``__main__``.\n\nThe global statement has the same scope as a name binding operation in\nthe same block.  If the nearest enclosing scope for a free variable\ncontains a global statement, the free variable is treated as a global.\n\nA class definition is an executable statement that may use and define\nnames. These references follow the normal rules for name resolution.\nThe namespace of the class definition becomes the attribute dictionary\nof the class.  Names defined at the class scope are not visible in\nmethods.\n\n\nInteraction with dynamic features\n=================================\n\nThere are several cases where Python statements are illegal when used\nin conjunction with nested scopes that contain free variables.\n\nIf a variable is referenced in an enclosing scope, it is illegal to\ndelete the name.  An error will be reported at compile time.\n\nIf the wild card form of import --- ``import *`` --- is used in a\nfunction and the function contains or is a nested block with free\nvariables, the compiler will raise a ``SyntaxError``.\n\nThe ``eval()`` and ``exec()`` functions do not have access to the full\nenvironment for resolving names.  Names may be resolved in the local\nand global namespaces of the caller.  Free variables are not resolved\nin the nearest enclosing namespace, but in the global namespace.  [1]\nThe ``exec()`` and ``eval()`` functions have optional arguments to\noverride the global and local namespace.  If only one namespace is\nspecified, it is used for both.\n",
+ 'naming': "\nNaming and binding\n******************\n\n*Names* refer to objects.  Names are introduced by name binding\noperations. Each occurrence of a name in the program text refers to\nthe *binding* of that name established in the innermost function block\ncontaining the use.\n\nA *block* is a piece of Python program text that is executed as a\nunit. The following are blocks: a module, a function body, and a class\ndefinition. Each command typed interactively is a block.  A script\nfile (a file given as standard input to the interpreter or specified\non the interpreter command line the first argument) is a code block.\nA script command (a command specified on the interpreter command line\nwith the '**-c**' option) is a code block.  The string argument passed\nto the built-in functions ``eval()`` and ``exec()`` is a code block.\n\nA code block is executed in an *execution frame*.  A frame contains\nsome administrative information (used for debugging) and determines\nwhere and how execution continues after the code block's execution has\ncompleted.\n\nA *scope* defines the visibility of a name within a block.  If a local\nvariable is defined in a block, its scope includes that block.  If the\ndefinition occurs in a function block, the scope extends to any blocks\ncontained within the defining one, unless a contained block introduces\na different binding for the name.  The scope of names defined in a\nclass block is limited to the class block; it does not extend to the\ncode blocks of methods -- this includes comprehensions and generator\nexpressions since they are implemented using a function scope.  This\nmeans that the following will fail:\n\n   class A:\n       a = 42\n       b = list(a + i for i in range(10))\n\nWhen a name is used in a code block, it is resolved using the nearest\nenclosing scope.  The set of all such scopes visible to a code block\nis called the block's *environment*.\n\nIf a name is bound in a block, it is a local variable of that block,\nunless declared as ``nonlocal``.  If a name is bound at the module\nlevel, it is a global variable.  (The variables of the module code\nblock are local and global.)  If a variable is used in a code block\nbut not defined there, it is a *free variable*.\n\nWhen a name is not found at all, a ``NameError`` exception is raised.\nIf the name refers to a local variable that has not been bound, a\n``UnboundLocalError`` exception is raised.  ``UnboundLocalError`` is a\nsubclass of ``NameError``.\n\nThe following constructs bind names: formal parameters to functions,\n``import`` statements, class and function definitions (these bind the\nclass or function name in the defining block), and targets that are\nidentifiers if occurring in an assignment, ``for`` loop header, or\nafter ``as`` in a ``with`` statement or :keyword.`except` clause. The\n``import`` statement of the form ``from ... import *`` binds all names\ndefined in the imported module, except those beginning with an\nunderscore.  This form may only be used at the module level.\n\nA target occurring in a ``del`` statement is also considered bound for\nthis purpose (though the actual semantics are to unbind the name).  It\nis illegal to unbind a name that is referenced by an enclosing scope;\nthe compiler will report a ``SyntaxError``.\n\nEach assignment or import statement occurs within a block defined by a\nclass or function definition or at the module level (the top-level\ncode block).\n\nIf a name binding operation occurs anywhere within a code block, all\nuses of the name within the block are treated as references to the\ncurrent block.  This can lead to errors when a name is used within a\nblock before it is bound.  This rule is subtle.  Python lacks\ndeclarations and allows name binding operations to occur anywhere\nwithin a code block.  The local variables of a code block can be\ndetermined by scanning the entire text of the block for name binding\noperations.\n\nIf the ``global`` statement occurs within a block, all uses of the\nname specified in the statement refer to the binding of that name in\nthe top-level namespace.  Names are resolved in the top-level\nnamespace by searching the global namespace, i.e. the namespace of the\nmodule containing the code block, and the builtins namespace, the\nnamespace of the module ``builtins``.  The global namespace is\nsearched first.  If the name is not found there, the builtins\nnamespace is searched.  The global statement must precede all uses of\nthe name.\n\nThe builtins namespace associated with the execution of a code block\nis actually found by looking up the name ``__builtins__`` in its\nglobal namespace; this should be a dictionary or a module (in the\nlatter case the module's dictionary is used).  By default, when in the\n``__main__`` module, ``__builtins__`` is the built-in module\n``builtins``; when in any other module, ``__builtins__`` is an alias\nfor the dictionary of the ``builtins`` module itself.\n``__builtins__`` can be set to a user-created dictionary to create a\nweak form of restricted execution.\n\n**CPython implementation detail:** Users should not touch\n``__builtins__``; it is strictly an implementation detail.  Users\nwanting to override values in the builtins namespace should ``import``\nthe ``builtins`` module and modify its attributes appropriately.\n\nThe namespace for a module is automatically created the first time a\nmodule is imported.  The main module for a script is always called\n``__main__``.\n\nThe global statement has the same scope as a name binding operation in\nthe same block.  If the nearest enclosing scope for a free variable\ncontains a global statement, the free variable is treated as a global.\n\nA class definition is an executable statement that may use and define\nnames. These references follow the normal rules for name resolution.\nThe namespace of the class definition becomes the attribute dictionary\nof the class.  Names defined at the class scope are not visible in\nmethods.\n\n\nInteraction with dynamic features\n=================================\n\nThere are several cases where Python statements are illegal when used\nin conjunction with nested scopes that contain free variables.\n\nIf a variable is referenced in an enclosing scope, it is illegal to\ndelete the name.  An error will be reported at compile time.\n\nIf the wild card form of import --- ``import *`` --- is used in a\nfunction and the function contains or is a nested block with free\nvariables, the compiler will raise a ``SyntaxError``.\n\nThe ``eval()`` and ``exec()`` functions do not have access to the full\nenvironment for resolving names.  Names may be resolved in the local\nand global namespaces of the caller.  Free variables are not resolved\nin the nearest enclosing namespace, but in the global namespace.  [1]\nThe ``exec()`` and ``eval()`` functions have optional arguments to\noverride the global and local namespace.  If only one namespace is\nspecified, it is used for both.\n",
  'numbers': "\nNumeric literals\n****************\n\nThere are three types of numeric literals: integers, floating point\nnumbers, and imaginary numbers.  There are no complex literals\n(complex numbers can be formed by adding a real number and an\nimaginary number).\n\nNote that numeric literals do not include a sign; a phrase like ``-1``\nis actually an expression composed of the unary operator '``-``' and\nthe literal ``1``.\n",
- 'numeric-types': "\nEmulating numeric types\n***********************\n\nThe following methods can be defined to emulate numeric objects.\nMethods corresponding to operations that are not supported by the\nparticular kind of number implemented (e.g., bitwise operations for\nnon-integral numbers) should be left undefined.\n\nobject.__add__(self, other)\nobject.__sub__(self, other)\nobject.__mul__(self, other)\nobject.__truediv__(self, other)\nobject.__floordiv__(self, other)\nobject.__mod__(self, other)\nobject.__divmod__(self, other)\nobject.__pow__(self, other[, modulo])\nobject.__lshift__(self, other)\nobject.__rshift__(self, other)\nobject.__and__(self, other)\nobject.__xor__(self, other)\nobject.__or__(self, other)\n\n   These methods are called to implement the binary arithmetic\n   operations (``+``, ``-``, ``*``, ``/``, ``//``, ``%``,\n   ``divmod()``, ``pow()``, ``**``, ``<<``, ``>>``, ``&``, ``^``,\n   ``|``).  For instance, to evaluate the expression ``x + y``, where\n   *x* is an instance of a class that has an ``__add__()`` method,\n   ``x.__add__(y)`` is called.  The ``__divmod__()`` method should be\n   the equivalent to using ``__floordiv__()`` and ``__mod__()``; it\n   should not be related to ``__truediv__()``.  Note that\n   ``__pow__()`` should be defined to accept an optional third\n   argument if the ternary version of the built-in ``pow()`` function\n   is to be supported.\n\n   If one of those methods does not support the operation with the\n   supplied arguments, it should return ``NotImplemented``.\n\nobject.__radd__(self, other)\nobject.__rsub__(self, other)\nobject.__rmul__(self, other)\nobject.__rtruediv__(self, other)\nobject.__rfloordiv__(self, other)\nobject.__rmod__(self, other)\nobject.__rdivmod__(self, other)\nobject.__rpow__(self, other)\nobject.__rlshift__(self, other)\nobject.__rrshift__(self, other)\nobject.__rand__(self, other)\nobject.__rxor__(self, other)\nobject.__ror__(self, other)\n\n   These methods are called to implement the binary arithmetic\n   operations (``+``, ``-``, ``*``, ``/``, ``//``, ``%``,\n   ``divmod()``, ``pow()``, ``**``, ``<<``, ``>>``, ``&``, ``^``,\n   ``|``) with reflected (swapped) operands. These functions are only\n   called if the left operand does not support the corresponding\n   operation and the operands are of different types. [3]  For\n   instance, to evaluate the expression ``x - y``, where *y* is an\n   instance of a class that has an ``__rsub__()`` method,\n   ``y.__rsub__(x)`` is called if ``x.__sub__(y)`` returns\n   *NotImplemented*.\n\n   Note that ternary ``pow()`` will not try calling ``__rpow__()``\n   (the coercion rules would become too complicated).\n\n   Note: If the right operand's type is a subclass of the left operand's\n     type and that subclass provides the reflected method for the\n     operation, this method will be called before the left operand's\n     non-reflected method.  This behavior allows subclasses to\n     override their ancestors' operations.\n\nobject.__iadd__(self, other)\nobject.__isub__(self, other)\nobject.__imul__(self, other)\nobject.__itruediv__(self, other)\nobject.__ifloordiv__(self, other)\nobject.__imod__(self, other)\nobject.__ipow__(self, other[, modulo])\nobject.__ilshift__(self, other)\nobject.__irshift__(self, other)\nobject.__iand__(self, other)\nobject.__ixor__(self, other)\nobject.__ior__(self, other)\n\n   These methods are called to implement the augmented arithmetic\n   assignments (``+=``, ``-=``, ``*=``, ``/=``, ``//=``, ``%=``,\n   ``**=``, ``<<=``, ``>>=``, ``&=``, ``^=``, ``|=``).  These methods\n   should attempt to do the operation in-place (modifying *self*) and\n   return the result (which could be, but does not have to be,\n   *self*).  If a specific method is not defined, the augmented\n   assignment falls back to the normal methods.  For instance, to\n   execute the statement ``x += y``, where *x* is an instance of a\n   class that has an ``__iadd__()`` method, ``x.__iadd__(y)`` is\n   called.  If *x* is an instance of a class that does not define a\n   ``__iadd__()`` method, ``x.__add__(y)`` and ``y.__radd__(x)`` are\n   considered, as with the evaluation of ``x + y``.\n\nobject.__neg__(self)\nobject.__pos__(self)\nobject.__abs__(self)\nobject.__invert__(self)\n\n   Called to implement the unary arithmetic operations (``-``, ``+``,\n   ``abs()`` and ``~``).\n\nobject.__complex__(self)\nobject.__int__(self)\nobject.__float__(self)\nobject.__round__(self[, n])\n\n   Called to implement the built-in functions ``complex()``,\n   ``int()``, ``float()`` and ``round()``.  Should return a value of\n   the appropriate type.\n\nobject.__index__(self)\n\n   Called to implement ``operator.index()``.  Also called whenever\n   Python needs an integer object (such as in slicing, or in the\n   built-in ``bin()``, ``hex()`` and ``oct()`` functions). Must return\n   an integer.\n",
- 'objects': '\nObjects, values and types\n*************************\n\n*Objects* are Python\'s abstraction for data.  All data in a Python\nprogram is represented by objects or by relations between objects. (In\na sense, and in conformance to Von Neumann\'s model of a "stored\nprogram computer," code is also represented by objects.)\n\nEvery object has an identity, a type and a value.  An object\'s\n*identity* never changes once it has been created; you may think of it\nas the object\'s address in memory.  The \'``is``\' operator compares the\nidentity of two objects; the ``id()`` function returns an integer\nrepresenting its identity (currently implemented as its address). An\nobject\'s *type* is also unchangeable. [1] An object\'s type determines\nthe operations that the object supports (e.g., "does it have a\nlength?") and also defines the possible values for objects of that\ntype.  The ``type()`` function returns an object\'s type (which is an\nobject itself).  The *value* of some objects can change.  Objects\nwhose value can change are said to be *mutable*; objects whose value\nis unchangeable once they are created are called *immutable*. (The\nvalue of an immutable container object that contains a reference to a\nmutable object can change when the latter\'s value is changed; however\nthe container is still considered immutable, because the collection of\nobjects it contains cannot be changed.  So, immutability is not\nstrictly the same as having an unchangeable value, it is more subtle.)\nAn object\'s mutability is determined by its type; for instance,\nnumbers, strings and tuples are immutable, while dictionaries and\nlists are mutable.\n\nObjects are never explicitly destroyed; however, when they become\nunreachable they may be garbage-collected.  An implementation is\nallowed to postpone garbage collection or omit it altogether --- it is\na matter of implementation quality how garbage collection is\nimplemented, as long as no objects are collected that are still\nreachable.  (Implementation note: CPython currently uses a reference-\ncounting scheme with (optional) delayed detection of cyclically linked\ngarbage, which collects most objects as soon as they become\nunreachable, but is not guaranteed to collect garbage containing\ncircular references.  See the documentation of the ``gc`` module for\ninformation on controlling the collection of cyclic garbage.  Other\nimplementations act differently and CPython may change.)\n\nNote that the use of the implementation\'s tracing or debugging\nfacilities may keep objects alive that would normally be collectable.\nAlso note that catching an exception with a \'``try``...``except``\'\nstatement may keep objects alive.\n\nSome objects contain references to "external" resources such as open\nfiles or windows.  It is understood that these resources are freed\nwhen the object is garbage-collected, but since garbage collection is\nnot guaranteed to happen, such objects also provide an explicit way to\nrelease the external resource, usually a ``close()`` method. Programs\nare strongly recommended to explicitly close such objects.  The\n\'``try``...``finally``\' statement and the \'``with``\' statement provide\nconvenient ways to do this.\n\nSome objects contain references to other objects; these are called\n*containers*. Examples of containers are tuples, lists and\ndictionaries.  The references are part of a container\'s value.  In\nmost cases, when we talk about the value of a container, we imply the\nvalues, not the identities of the contained objects; however, when we\ntalk about the mutability of a container, only the identities of the\nimmediately contained objects are implied.  So, if an immutable\ncontainer (like a tuple) contains a reference to a mutable object, its\nvalue changes if that mutable object is changed.\n\nTypes affect almost all aspects of object behavior.  Even the\nimportance of object identity is affected in some sense: for immutable\ntypes, operations that compute new values may actually return a\nreference to any existing object with the same type and value, while\nfor mutable objects this is not allowed.  E.g., after ``a = 1; b =\n1``, ``a`` and ``b`` may or may not refer to the same object with the\nvalue one, depending on the implementation, but after ``c = []; d =\n[]``, ``c`` and ``d`` are guaranteed to refer to two different,\nunique, newly created empty lists. (Note that ``c = d = []`` assigns\nthe same object to both ``c`` and ``d``.)\n',
- 'operator-summary': '\nSummary\n*******\n\nThe following table summarizes the operator precedences in Python,\nfrom lowest precedence (least binding) to highest precedence (most\nbinding).  Operators in the same box have the same precedence.  Unless\nthe syntax is explicitly given, operators are binary.  Operators in\nthe same box group left to right (except for comparisons, including\ntests, which all have the same precedence and chain from left to right\n--- see section *Comparisons* --- and exponentiation, which groups\nfrom right to left).\n\n+-------------------------------------------------+---------------------------------------+\n| Operator                                        | Description                           |\n+=================================================+=======================================+\n| ``lambda``                                      | Lambda expression                     |\n+-------------------------------------------------+---------------------------------------+\n| ``or``                                          | Boolean OR                            |\n+-------------------------------------------------+---------------------------------------+\n| ``and``                                         | Boolean AND                           |\n+-------------------------------------------------+---------------------------------------+\n| ``not`` *x*                                     | Boolean NOT                           |\n+-------------------------------------------------+---------------------------------------+\n| ``in``, ``not`` ``in``, ``is``, ``is not``,     | Comparisons, including membership     |\n| ``<``, ``<=``, ``>``, ``>=``, ``!=``, ``==``    | tests and identity tests,             |\n+-------------------------------------------------+---------------------------------------+\n| ``|``                                           | Bitwise OR                            |\n+-------------------------------------------------+---------------------------------------+\n| ``^``                                           | Bitwise XOR                           |\n+-------------------------------------------------+---------------------------------------+\n| ``&``                                           | Bitwise AND                           |\n+-------------------------------------------------+---------------------------------------+\n| ``<<``, ``>>``                                  | Shifts                                |\n+-------------------------------------------------+---------------------------------------+\n| ``+``, ``-``                                    | Addition and subtraction              |\n+-------------------------------------------------+---------------------------------------+\n| ``*``, ``/``, ``//``, ``%``                     | Multiplication, division, remainder   |\n+-------------------------------------------------+---------------------------------------+\n| ``+x``, ``-x``, ``~x``                          | Positive, negative, bitwise NOT       |\n+-------------------------------------------------+---------------------------------------+\n| ``**``                                          | Exponentiation [7]                    |\n+-------------------------------------------------+---------------------------------------+\n| ``x[index]``, ``x[index:index]``,               | Subscription, slicing, call,          |\n| ``x(arguments...)``, ``x.attribute``            | attribute reference                   |\n+-------------------------------------------------+---------------------------------------+\n| ``(expressions...)``, ``[expressions...]``,     | Binding or tuple display, list        |\n| ``{key:datum...}``,                             | display, dictionary display,          |\n+-------------------------------------------------+---------------------------------------+\n\n-[ Footnotes ]-\n\n[1] While ``abs(x%y) < abs(y)`` is true mathematically, for floats it\n    may not be true numerically due to roundoff.  For example, and\n    assuming a platform on which a Python float is an IEEE 754 double-\n    precision number, in order that ``-1e-100 % 1e100`` have the same\n    sign as ``1e100``, the computed result is ``-1e-100 + 1e100``,\n    which is numerically exactly equal to ``1e100``.  Function\n    ``fmod()`` in the ``math`` module returns a result whose sign\n    matches the sign of the first argument instead, and so returns\n    ``-1e-100`` in this case. Which approach is more appropriate\n    depends on the application.\n\n[2] If x is very close to an exact integer multiple of y, it\'s\n    possible for ``x//y`` to be one larger than ``(x-x%y)//y`` due to\n    rounding.  In such cases, Python returns the latter result, in\n    order to preserve that ``divmod(x,y)[0] * y + x % y`` be very\n    close to ``x``.\n\n[3] While comparisons between strings make sense at the byte level,\n    they may be counter-intuitive to users.  For example, the strings\n    ``"\\u00C7"`` and ``"\\u0327\\u0043"`` compare differently, even\n    though they both represent the same unicode character (LATIN\n    CAPITAL LETTER C WITH CEDILLA).  To compare strings in a human\n    recognizable way, compare using ``unicodedata.normalize()``.\n\n[4] The implementation computes this efficiently, without constructing\n    lists or sorting.\n\n[5] Earlier versions of Python used lexicographic comparison of the\n    sorted (key, value) lists, but this was very expensive for the\n    common case of comparing for equality.  An even earlier version of\n    Python compared dictionaries by identity only, but this caused\n    surprises because people expected to be able to test a dictionary\n    for emptiness by comparing it to ``{}``.\n\n[6] Due to automatic garbage-collection, free lists, and the dynamic\n    nature of descriptors, you may notice seemingly unusual behaviour\n    in certain uses of the ``is`` operator, like those involving\n    comparisons between instance methods, or constants.  Check their\n    documentation for more info.\n\n[7] The power operator ``**`` binds less tightly than an arithmetic or\n    bitwise unary operator on its right, that is, ``2**-1`` is\n    ``0.5``.\n',
+ 'numeric-types': "\nEmulating numeric types\n***********************\n\nThe following methods can be defined to emulate numeric objects.\nMethods corresponding to operations that are not supported by the\nparticular kind of number implemented (e.g., bitwise operations for\nnon-integral numbers) should be left undefined.\n\nobject.__add__(self, other)\nobject.__sub__(self, other)\nobject.__mul__(self, other)\nobject.__truediv__(self, other)\nobject.__floordiv__(self, other)\nobject.__mod__(self, other)\nobject.__divmod__(self, other)\nobject.__pow__(self, other[, modulo])\nobject.__lshift__(self, other)\nobject.__rshift__(self, other)\nobject.__and__(self, other)\nobject.__xor__(self, other)\nobject.__or__(self, other)\n\n   These methods are called to implement the binary arithmetic\n   operations (``+``, ``-``, ``*``, ``/``, ``//``, ``%``,\n   ``divmod()``, ``pow()``, ``**``, ``<<``, ``>>``, ``&``, ``^``,\n   ``|``).  For instance, to evaluate the expression ``x + y``, where\n   *x* is an instance of a class that has an ``__add__()`` method,\n   ``x.__add__(y)`` is called.  The ``__divmod__()`` method should be\n   the equivalent to using ``__floordiv__()`` and ``__mod__()``; it\n   should not be related to ``__truediv__()``.  Note that\n   ``__pow__()`` should be defined to accept an optional third\n   argument if the ternary version of the built-in ``pow()`` function\n   is to be supported.\n\n   If one of those methods does not support the operation with the\n   supplied arguments, it should return ``NotImplemented``.\n\nobject.__radd__(self, other)\nobject.__rsub__(self, other)\nobject.__rmul__(self, other)\nobject.__rtruediv__(self, other)\nobject.__rfloordiv__(self, other)\nobject.__rmod__(self, other)\nobject.__rdivmod__(self, other)\nobject.__rpow__(self, other)\nobject.__rlshift__(self, other)\nobject.__rrshift__(self, other)\nobject.__rand__(self, other)\nobject.__rxor__(self, other)\nobject.__ror__(self, other)\n\n   These methods are called to implement the binary arithmetic\n   operations (``+``, ``-``, ``*``, ``/``, ``//``, ``%``,\n   ``divmod()``, ``pow()``, ``**``, ``<<``, ``>>``, ``&``, ``^``,\n   ``|``) with reflected (swapped) operands. These functions are only\n   called if the left operand does not support the corresponding\n   operation and the operands are of different types. [2]  For\n   instance, to evaluate the expression ``x - y``, where *y* is an\n   instance of a class that has an ``__rsub__()`` method,\n   ``y.__rsub__(x)`` is called if ``x.__sub__(y)`` returns\n   *NotImplemented*.\n\n   Note that ternary ``pow()`` will not try calling ``__rpow__()``\n   (the coercion rules would become too complicated).\n\n   Note: If the right operand's type is a subclass of the left operand's\n     type and that subclass provides the reflected method for the\n     operation, this method will be called before the left operand's\n     non-reflected method.  This behavior allows subclasses to\n     override their ancestors' operations.\n\nobject.__iadd__(self, other)\nobject.__isub__(self, other)\nobject.__imul__(self, other)\nobject.__itruediv__(self, other)\nobject.__ifloordiv__(self, other)\nobject.__imod__(self, other)\nobject.__ipow__(self, other[, modulo])\nobject.__ilshift__(self, other)\nobject.__irshift__(self, other)\nobject.__iand__(self, other)\nobject.__ixor__(self, other)\nobject.__ior__(self, other)\n\n   These methods are called to implement the augmented arithmetic\n   assignments (``+=``, ``-=``, ``*=``, ``/=``, ``//=``, ``%=``,\n   ``**=``, ``<<=``, ``>>=``, ``&=``, ``^=``, ``|=``).  These methods\n   should attempt to do the operation in-place (modifying *self*) and\n   return the result (which could be, but does not have to be,\n   *self*).  If a specific method is not defined, the augmented\n   assignment falls back to the normal methods.  For instance, to\n   execute the statement ``x += y``, where *x* is an instance of a\n   class that has an ``__iadd__()`` method, ``x.__iadd__(y)`` is\n   called.  If *x* is an instance of a class that does not define a\n   ``__iadd__()`` method, ``x.__add__(y)`` and ``y.__radd__(x)`` are\n   considered, as with the evaluation of ``x + y``.\n\nobject.__neg__(self)\nobject.__pos__(self)\nobject.__abs__(self)\nobject.__invert__(self)\n\n   Called to implement the unary arithmetic operations (``-``, ``+``,\n   ``abs()`` and ``~``).\n\nobject.__complex__(self)\nobject.__int__(self)\nobject.__float__(self)\nobject.__round__(self[, n])\n\n   Called to implement the built-in functions ``complex()``,\n   ``int()``, ``float()`` and ``round()``.  Should return a value of\n   the appropriate type.\n\nobject.__index__(self)\n\n   Called to implement ``operator.index()``.  Also called whenever\n   Python needs an integer object (such as in slicing, or in the\n   built-in ``bin()``, ``hex()`` and ``oct()`` functions). Must return\n   an integer.\n",
+ 'objects': '\nObjects, values and types\n*************************\n\n*Objects* are Python\'s abstraction for data.  All data in a Python\nprogram is represented by objects or by relations between objects. (In\na sense, and in conformance to Von Neumann\'s model of a "stored\nprogram computer," code is also represented by objects.)\n\nEvery object has an identity, a type and a value.  An object\'s\n*identity* never changes once it has been created; you may think of it\nas the object\'s address in memory.  The \'``is``\' operator compares the\nidentity of two objects; the ``id()`` function returns an integer\nrepresenting its identity (currently implemented as its address). An\nobject\'s *type* is also unchangeable. [1] An object\'s type determines\nthe operations that the object supports (e.g., "does it have a\nlength?") and also defines the possible values for objects of that\ntype.  The ``type()`` function returns an object\'s type (which is an\nobject itself).  The *value* of some objects can change.  Objects\nwhose value can change are said to be *mutable*; objects whose value\nis unchangeable once they are created are called *immutable*. (The\nvalue of an immutable container object that contains a reference to a\nmutable object can change when the latter\'s value is changed; however\nthe container is still considered immutable, because the collection of\nobjects it contains cannot be changed.  So, immutability is not\nstrictly the same as having an unchangeable value, it is more subtle.)\nAn object\'s mutability is determined by its type; for instance,\nnumbers, strings and tuples are immutable, while dictionaries and\nlists are mutable.\n\nObjects are never explicitly destroyed; however, when they become\nunreachable they may be garbage-collected.  An implementation is\nallowed to postpone garbage collection or omit it altogether --- it is\na matter of implementation quality how garbage collection is\nimplemented, as long as no objects are collected that are still\nreachable.\n\n**CPython implementation detail:** CPython currently uses a reference-\ncounting scheme with (optional) delayed detection of cyclically linked\ngarbage, which collects most objects as soon as they become\nunreachable, but is not guaranteed to collect garbage containing\ncircular references.  See the documentation of the ``gc`` module for\ninformation on controlling the collection of cyclic garbage. Other\nimplementations act differently and CPython may change.\n\nNote that the use of the implementation\'s tracing or debugging\nfacilities may keep objects alive that would normally be collectable.\nAlso note that catching an exception with a \'``try``...``except``\'\nstatement may keep objects alive.\n\nSome objects contain references to "external" resources such as open\nfiles or windows.  It is understood that these resources are freed\nwhen the object is garbage-collected, but since garbage collection is\nnot guaranteed to happen, such objects also provide an explicit way to\nrelease the external resource, usually a ``close()`` method. Programs\nare strongly recommended to explicitly close such objects.  The\n\'``try``...``finally``\' statement and the \'``with``\' statement provide\nconvenient ways to do this.\n\nSome objects contain references to other objects; these are called\n*containers*. Examples of containers are tuples, lists and\ndictionaries.  The references are part of a container\'s value.  In\nmost cases, when we talk about the value of a container, we imply the\nvalues, not the identities of the contained objects; however, when we\ntalk about the mutability of a container, only the identities of the\nimmediately contained objects are implied.  So, if an immutable\ncontainer (like a tuple) contains a reference to a mutable object, its\nvalue changes if that mutable object is changed.\n\nTypes affect almost all aspects of object behavior.  Even the\nimportance of object identity is affected in some sense: for immutable\ntypes, operations that compute new values may actually return a\nreference to any existing object with the same type and value, while\nfor mutable objects this is not allowed.  E.g., after ``a = 1; b =\n1``, ``a`` and ``b`` may or may not refer to the same object with the\nvalue one, depending on the implementation, but after ``c = []; d =\n[]``, ``c`` and ``d`` are guaranteed to refer to two different,\nunique, newly created empty lists. (Note that ``c = d = []`` assigns\nthe same object to both ``c`` and ``d``.)\n',
+ 'operator-summary': '\nSummary\n*******\n\nThe following table summarizes the operator precedences in Python,\nfrom lowest precedence (least binding) to highest precedence (most\nbinding).  Operators in the same box have the same precedence.  Unless\nthe syntax is explicitly given, operators are binary.  Operators in\nthe same box group left to right (except for comparisons, including\ntests, which all have the same precedence and chain from left to right\n--- see section *Comparisons* --- and exponentiation, which groups\nfrom right to left).\n\n+-------------------------------------------------+---------------------------------------+\n| Operator                                        | Description                           |\n+=================================================+=======================================+\n| ``lambda``                                      | Lambda expression                     |\n+-------------------------------------------------+---------------------------------------+\n| ``if`` -- ``else``                              | Conditional expression                |\n+-------------------------------------------------+---------------------------------------+\n| ``or``                                          | Boolean OR                            |\n+-------------------------------------------------+---------------------------------------+\n| ``and``                                         | Boolean AND                           |\n+-------------------------------------------------+---------------------------------------+\n| ``not`` *x*                                     | Boolean NOT                           |\n+-------------------------------------------------+---------------------------------------+\n| ``in``, ``not`` ``in``, ``is``, ``is not``,     | Comparisons, including membership     |\n| ``<``, ``<=``, ``>``, ``>=``, ``!=``, ``==``    | tests and identity tests,             |\n+-------------------------------------------------+---------------------------------------+\n| ``|``                                           | Bitwise OR                            |\n+-------------------------------------------------+---------------------------------------+\n| ``^``                                           | Bitwise XOR                           |\n+-------------------------------------------------+---------------------------------------+\n| ``&``                                           | Bitwise AND                           |\n+-------------------------------------------------+---------------------------------------+\n| ``<<``, ``>>``                                  | Shifts                                |\n+-------------------------------------------------+---------------------------------------+\n| ``+``, ``-``                                    | Addition and subtraction              |\n+-------------------------------------------------+---------------------------------------+\n| ``*``, ``/``, ``//``, ``%``                     | Multiplication, division, remainder   |\n+-------------------------------------------------+---------------------------------------+\n| ``+x``, ``-x``, ``~x``                          | Positive, negative, bitwise NOT       |\n+-------------------------------------------------+---------------------------------------+\n| ``**``                                          | Exponentiation [5]                    |\n+-------------------------------------------------+---------------------------------------+\n| ``x[index]``, ``x[index:index]``,               | Subscription, slicing, call,          |\n| ``x(arguments...)``, ``x.attribute``            | attribute reference                   |\n+-------------------------------------------------+---------------------------------------+\n| ``(expressions...)``, ``[expressions...]``,     | Binding or tuple display, list        |\n| ``{key:datum...}``,                             | display, dictionary display,          |\n+-------------------------------------------------+---------------------------------------+\n\n-[ Footnotes ]-\n\n[1] While ``abs(x%y) < abs(y)`` is true mathematically, for floats it\n    may not be true numerically due to roundoff.  For example, and\n    assuming a platform on which a Python float is an IEEE 754 double-\n    precision number, in order that ``-1e-100 % 1e100`` have the same\n    sign as ``1e100``, the computed result is ``-1e-100 + 1e100``,\n    which is numerically exactly equal to ``1e100``.  Function\n    ``fmod()`` in the ``math`` module returns a result whose sign\n    matches the sign of the first argument instead, and so returns\n    ``-1e-100`` in this case. Which approach is more appropriate\n    depends on the application.\n\n[2] If x is very close to an exact integer multiple of y, it\'s\n    possible for ``x//y`` to be one larger than ``(x-x%y)//y`` due to\n    rounding.  In such cases, Python returns the latter result, in\n    order to preserve that ``divmod(x,y)[0] * y + x % y`` be very\n    close to ``x``.\n\n[3] While comparisons between strings make sense at the byte level,\n    they may be counter-intuitive to users.  For example, the strings\n    ``"\\u00C7"`` and ``"\\u0327\\u0043"`` compare differently, even\n    though they both represent the same unicode character (LATIN\n    CAPITAL LETTER C WITH CEDILLA).  To compare strings in a human\n    recognizable way, compare using ``unicodedata.normalize()``.\n\n[4] Due to automatic garbage-collection, free lists, and the dynamic\n    nature of descriptors, you may notice seemingly unusual behaviour\n    in certain uses of the ``is`` operator, like those involving\n    comparisons between instance methods, or constants.  Check their\n    documentation for more info.\n\n[5] The power operator ``**`` binds less tightly than an arithmetic or\n    bitwise unary operator on its right, that is, ``2**-1`` is\n    ``0.5``.\n',
  'pass': '\nThe ``pass`` statement\n**********************\n\n   pass_stmt ::= "pass"\n\n``pass`` is a null operation --- when it is executed, nothing happens.\nIt is useful as a placeholder when a statement is required\nsyntactically, but no code needs to be executed, for example:\n\n   def f(arg): pass    # a function that does nothing (yet)\n\n   class C: pass       # a class with no methods (yet)\n',
  'power': '\nThe power operator\n******************\n\nThe power operator binds more tightly than unary operators on its\nleft; it binds less tightly than unary operators on its right.  The\nsyntax is:\n\n   power ::= primary ["**" u_expr]\n\nThus, in an unparenthesized sequence of power and unary operators, the\noperators are evaluated from right to left (this does not constrain\nthe evaluation order for the operands): ``-1**2`` results in ``-1``.\n\nThe power operator has the same semantics as the built-in ``pow()``\nfunction, when called with two arguments: it yields its left argument\nraised to the power of its right argument.  The numeric arguments are\nfirst converted to a common type, and the result is of that type.\n\nFor int operands, the result has the same type as the operands unless\nthe second argument is negative; in that case, all arguments are\nconverted to float and a float result is delivered. For example,\n``10**2`` returns ``100``, but ``10**-2`` returns ``0.01``.\n\nRaising ``0.0`` to a negative power results in a\n``ZeroDivisionError``. Raising a negative number to a fractional power\nresults in a ``complex`` number. (In earlier versions it raised a\n``ValueError``.)\n',
  'raise': '\nThe ``raise`` statement\n***********************\n\n   raise_stmt ::= "raise" [expression ["from" expression]]\n\nIf no expressions are present, ``raise`` re-raises the last exception\nthat was active in the current scope.  If no exception is active in\nthe current scope, a ``TypeError`` exception is raised indicating that\nthis is an error (if running under IDLE, a ``queue.Empty`` exception\nis raised instead).\n\nOtherwise, ``raise`` evaluates the first expression as the exception\nobject.  It must be either a subclass or an instance of\n``BaseException``. If it is a class, the exception instance will be\nobtained when needed by instantiating the class with no arguments.\n\nThe *type* of the exception is the exception instance\'s class, the\n*value* is the instance itself.\n\nA traceback object is normally created automatically when an exception\nis raised and attached to it as the ``__traceback__`` attribute, which\nis writable. You can create an exception and set your own traceback in\none step using the ``with_traceback()`` exception method (which\nreturns the same exception instance, with its traceback set to its\nargument), like so:\n\n   raise Exception("foo occurred").with_traceback(tracebackobj)\n\nThe ``from`` clause is used for exception chaining: if given, the\nsecond *expression* must be another exception class or instance, which\nwill then be attached to the raised exception as the ``__cause__``\nattribute (which is writable).  If the raised exception is not\nhandled, both exceptions will be printed:\n\n   >>> try:\n   ...     print(1 / 0)\n   ... except Exception as exc:\n   ...     raise RuntimeError("Something bad happened") from exc\n   ...\n   Traceback (most recent call last):\n     File "<stdin>", line 2, in <module>\n   ZeroDivisionError: int division or modulo by zero\n\n   The above exception was the direct cause of the following exception:\n\n   Traceback (most recent call last):\n     File "<stdin>", line 4, in <module>\n   RuntimeError: Something bad happened\n\nA similar mechanism works implicitly if an exception is raised inside\nan exception handler: the previous exception is then attached as the\nnew exception\'s ``__context__`` attribute:\n\n   >>> try:\n   ...     print(1 / 0)\n   ... except:\n   ...     raise RuntimeError("Something bad happened")\n   ...\n   Traceback (most recent call last):\n     File "<stdin>", line 2, in <module>\n   ZeroDivisionError: int division or modulo by zero\n\n   During handling of the above exception, another exception occurred:\n\n   Traceback (most recent call last):\n     File "<stdin>", line 4, in <module>\n   RuntimeError: Something bad happened\n\nAdditional information on exceptions can be found in section\n*Exceptions*, and information about handling exceptions is in section\n*The try statement*.\n',
- 'return': '\nThe ``return`` statement\n************************\n\n   return_stmt ::= "return" [expression_list]\n\n``return`` may only occur syntactically nested in a function\ndefinition, not within a nested class definition.\n\nIf an expression list is present, it is evaluated, else ``None`` is\nsubstituted.\n\n``return`` leaves the current function call with the expression list\n(or ``None``) as return value.\n\nWhen ``return`` passes control out of a ``try`` statement with a\n``finally`` clause, that ``finally`` clause is executed before really\nleaving the function.\n\nIn a generator function, the ``return`` statement is not allowed to\ninclude an **expression_list**.  In that context, a bare ``return``\nindicates that the generator is done and will cause ``StopIteration``\nto be raised.\n',
- 'sequence-types': "\nEmulating container types\n*************************\n\nThe following methods can be defined to implement container objects.\nContainers usually are sequences (such as lists or tuples) or mappings\n(like dictionaries), but can represent other containers as well.  The\nfirst set of methods is used either to emulate a sequence or to\nemulate a mapping; the difference is that for a sequence, the\nallowable keys should be the integers *k* for which ``0 <= k < N``\nwhere *N* is the length of the sequence, or slice objects, which\ndefine a range of items.  It is also recommended that mappings provide\nthe methods ``keys()``, ``values()``, ``items()``, ``get()``,\n``clear()``, ``setdefault()``, ``pop()``, ``popitem()``, ``copy()``,\nand ``update()`` behaving similar to those for Python's standard\ndictionary objects.  The ``collections`` module provides a\n``MutableMapping`` abstract base class to help create those methods\nfrom a base set of ``__getitem__()``, ``__setitem__()``,\n``__delitem__()``, and ``keys()``. Mutable sequences should provide\nmethods ``append()``, ``count()``, ``index()``, ``extend()``,\n``insert()``, ``pop()``, ``remove()``, ``reverse()`` and ``sort()``,\nlike Python standard list objects.  Finally, sequence types should\nimplement addition (meaning concatenation) and multiplication (meaning\nrepetition) by defining the methods ``__add__()``, ``__radd__()``,\n``__iadd__()``, ``__mul__()``, ``__rmul__()`` and ``__imul__()``\ndescribed below; they should not define other numerical operators.  It\nis recommended that both mappings and sequences implement the\n``__contains__()`` method to allow efficient use of the ``in``\noperator; for mappings, ``in`` should search the mapping's keys; for\nsequences, it should search through the values.  It is further\nrecommended that both mappings and sequences implement the\n``__iter__()`` method to allow efficient iteration through the\ncontainer; for mappings, ``__iter__()`` should be the same as\n``keys()``; for sequences, it should iterate through the values.\n\nobject.__len__(self)\n\n   Called to implement the built-in function ``len()``.  Should return\n   the length of the object, an integer ``>=`` 0.  Also, an object\n   that doesn't define a ``__bool__()`` method and whose ``__len__()``\n   method returns zero is considered to be false in a Boolean context.\n\nNote: Slicing is done exclusively with the following three methods.  A\n  call like\n\n     a[1:2] = b\n\n  is translated to\n\n     a[slice(1, 2, None)] = b\n\n  and so forth.  Missing slice items are always filled in with\n  ``None``.\n\nobject.__getitem__(self, key)\n\n   Called to implement evaluation of ``self[key]``. For sequence\n   types, the accepted keys should be integers and slice objects.\n   Note that the special interpretation of negative indexes (if the\n   class wishes to emulate a sequence type) is up to the\n   ``__getitem__()`` method. If *key* is of an inappropriate type,\n   ``TypeError`` may be raised; if of a value outside the set of\n   indexes for the sequence (after any special interpretation of\n   negative values), ``IndexError`` should be raised. For mapping\n   types, if *key* is missing (not in the container), ``KeyError``\n   should be raised.\n\n   Note: ``for`` loops expect that an ``IndexError`` will be raised for\n     illegal indexes to allow proper detection of the end of the\n     sequence.\n\nobject.__setitem__(self, key, value)\n\n   Called to implement assignment to ``self[key]``.  Same note as for\n   ``__getitem__()``.  This should only be implemented for mappings if\n   the objects support changes to the values for keys, or if new keys\n   can be added, or for sequences if elements can be replaced.  The\n   same exceptions should be raised for improper *key* values as for\n   the ``__getitem__()`` method.\n\nobject.__delitem__(self, key)\n\n   Called to implement deletion of ``self[key]``.  Same note as for\n   ``__getitem__()``.  This should only be implemented for mappings if\n   the objects support removal of keys, or for sequences if elements\n   can be removed from the sequence.  The same exceptions should be\n   raised for improper *key* values as for the ``__getitem__()``\n   method.\n\nobject.__iter__(self)\n\n   This method is called when an iterator is required for a container.\n   This method should return a new iterator object that can iterate\n   over all the objects in the container.  For mappings, it should\n   iterate over the keys of the container, and should also be made\n   available as the method ``keys()``.\n\n   Iterator objects also need to implement this method; they are\n   required to return themselves.  For more information on iterator\n   objects, see *Iterator Types*.\n\nobject.__reversed__(self)\n\n   Called (if present) by the ``reversed()`` builtin to implement\n   reverse iteration.  It should return a new iterator object that\n   iterates over all the objects in the container in reverse order.\n\n   If the ``__reversed__()`` method is not provided, the\n   ``reversed()`` builtin will fall back to using the sequence\n   protocol (``__len__()`` and ``__getitem__()``).  Objects that\n   support the sequence protocol should only provide\n   ``__reversed__()`` if they can provide an implementation that is\n   more efficient than the one provided by ``reversed()``.\n\nThe membership test operators (``in`` and ``not in``) are normally\nimplemented as an iteration through a sequence.  However, container\nobjects can supply the following special method with a more efficient\nimplementation, which also does not require the object be a sequence.\n\nobject.__contains__(self, item)\n\n   Called to implement membership test operators.  Should return true\n   if *item* is in *self*, false otherwise.  For mapping objects, this\n   should consider the keys of the mapping rather than the values or\n   the key-item pairs.\n",
- 'shifting': '\nShifting operations\n*******************\n\nThe shifting operations have lower priority than the arithmetic\noperations:\n\n   shift_expr ::= a_expr | shift_expr ( "<<" | ">>" ) a_expr\n\nThese operators accept integers as arguments.  They shift the first\nargument to the left or right by the number of bits given by the\nsecond argument.\n\nA right shift by *n* bits is defined as division by ``pow(2,n)``.  A\nleft shift by *n* bits is defined as multiplication with ``pow(2,n)``.\n',
+ 'return': '\nThe ``return`` statement\n************************\n\n   return_stmt ::= "return" [expression_list]\n\n``return`` may only occur syntactically nested in a function\ndefinition, not within a nested class definition.\n\nIf an expression list is present, it is evaluated, else ``None`` is\nsubstituted.\n\n``return`` leaves the current function call with the expression list\n(or ``None``) as return value.\n\nWhen ``return`` passes control out of a ``try`` statement with a\n``finally`` clause, that ``finally`` clause is executed before really\nleaving the function.\n\nIn a generator function, the ``return`` statement is not allowed to\ninclude an ``expression_list``.  In that context, a bare ``return``\nindicates that the generator is done and will cause ``StopIteration``\nto be raised.\n',
+ 'sequence-types': "\nEmulating container types\n*************************\n\nThe following methods can be defined to implement container objects.\nContainers usually are sequences (such as lists or tuples) or mappings\n(like dictionaries), but can represent other containers as well.  The\nfirst set of methods is used either to emulate a sequence or to\nemulate a mapping; the difference is that for a sequence, the\nallowable keys should be the integers *k* for which ``0 <= k < N``\nwhere *N* is the length of the sequence, or slice objects, which\ndefine a range of items.  It is also recommended that mappings provide\nthe methods ``keys()``, ``values()``, ``items()``, ``get()``,\n``clear()``, ``setdefault()``, ``pop()``, ``popitem()``, ``copy()``,\nand ``update()`` behaving similar to those for Python's standard\ndictionary objects.  The ``collections`` module provides a\n``MutableMapping`` abstract base class to help create those methods\nfrom a base set of ``__getitem__()``, ``__setitem__()``,\n``__delitem__()``, and ``keys()``. Mutable sequences should provide\nmethods ``append()``, ``count()``, ``index()``, ``extend()``,\n``insert()``, ``pop()``, ``remove()``, ``reverse()`` and ``sort()``,\nlike Python standard list objects.  Finally, sequence types should\nimplement addition (meaning concatenation) and multiplication (meaning\nrepetition) by defining the methods ``__add__()``, ``__radd__()``,\n``__iadd__()``, ``__mul__()``, ``__rmul__()`` and ``__imul__()``\ndescribed below; they should not define other numerical operators.  It\nis recommended that both mappings and sequences implement the\n``__contains__()`` method to allow efficient use of the ``in``\noperator; for mappings, ``in`` should search the mapping's keys; for\nsequences, it should search through the values.  It is further\nrecommended that both mappings and sequences implement the\n``__iter__()`` method to allow efficient iteration through the\ncontainer; for mappings, ``__iter__()`` should be the same as\n``keys()``; for sequences, it should iterate through the values.\n\nobject.__len__(self)\n\n   Called to implement the built-in function ``len()``.  Should return\n   the length of the object, an integer ``>=`` 0.  Also, an object\n   that doesn't define a ``__bool__()`` method and whose ``__len__()``\n   method returns zero is considered to be false in a Boolean context.\n\nNote: Slicing is done exclusively with the following three methods.  A\n  call like\n\n     a[1:2] = b\n\n  is translated to\n\n     a[slice(1, 2, None)] = b\n\n  and so forth.  Missing slice items are always filled in with\n  ``None``.\n\nobject.__getitem__(self, key)\n\n   Called to implement evaluation of ``self[key]``. For sequence\n   types, the accepted keys should be integers and slice objects.\n   Note that the special interpretation of negative indexes (if the\n   class wishes to emulate a sequence type) is up to the\n   ``__getitem__()`` method. If *key* is of an inappropriate type,\n   ``TypeError`` may be raised; if of a value outside the set of\n   indexes for the sequence (after any special interpretation of\n   negative values), ``IndexError`` should be raised. For mapping\n   types, if *key* is missing (not in the container), ``KeyError``\n   should be raised.\n\n   Note: ``for`` loops expect that an ``IndexError`` will be raised for\n     illegal indexes to allow proper detection of the end of the\n     sequence.\n\nobject.__setitem__(self, key, value)\n\n   Called to implement assignment to ``self[key]``.  Same note as for\n   ``__getitem__()``.  This should only be implemented for mappings if\n   the objects support changes to the values for keys, or if new keys\n   can be added, or for sequences if elements can be replaced.  The\n   same exceptions should be raised for improper *key* values as for\n   the ``__getitem__()`` method.\n\nobject.__delitem__(self, key)\n\n   Called to implement deletion of ``self[key]``.  Same note as for\n   ``__getitem__()``.  This should only be implemented for mappings if\n   the objects support removal of keys, or for sequences if elements\n   can be removed from the sequence.  The same exceptions should be\n   raised for improper *key* values as for the ``__getitem__()``\n   method.\n\nobject.__iter__(self)\n\n   This method is called when an iterator is required for a container.\n   This method should return a new iterator object that can iterate\n   over all the objects in the container.  For mappings, it should\n   iterate over the keys of the container, and should also be made\n   available as the method ``keys()``.\n\n   Iterator objects also need to implement this method; they are\n   required to return themselves.  For more information on iterator\n   objects, see *Iterator Types*.\n\nobject.__reversed__(self)\n\n   Called (if present) by the ``reversed()`` built-in to implement\n   reverse iteration.  It should return a new iterator object that\n   iterates over all the objects in the container in reverse order.\n\n   If the ``__reversed__()`` method is not provided, the\n   ``reversed()`` built-in will fall back to using the sequence\n   protocol (``__len__()`` and ``__getitem__()``).  Objects that\n   support the sequence protocol should only provide\n   ``__reversed__()`` if they can provide an implementation that is\n   more efficient than the one provided by ``reversed()``.\n\nThe membership test operators (``in`` and ``not in``) are normally\nimplemented as an iteration through a sequence.  However, container\nobjects can supply the following special method with a more efficient\nimplementation, which also does not require the object be a sequence.\n\nobject.__contains__(self, item)\n\n   Called to implement membership test operators.  Should return true\n   if *item* is in *self*, false otherwise.  For mapping objects, this\n   should consider the keys of the mapping rather than the values or\n   the key-item pairs.\n\n   For objects that don't define ``__contains__()``, the membership\n   test first tries iteration via ``__iter__()``, then the old\n   sequence iteration protocol via ``__getitem__()``, see *this\n   section in the language reference*.\n",
+ 'shifting': '\nShifting operations\n*******************\n\nThe shifting operations have lower priority than the arithmetic\noperations:\n\n   shift_expr ::= a_expr | shift_expr ( "<<" | ">>" ) a_expr\n\nThese operators accept integers as arguments.  They shift the first\nargument to the left or right by the number of bits given by the\nsecond argument.\n\nA right shift by *n* bits is defined as division by ``pow(2,n)``.  A\nleft shift by *n* bits is defined as multiplication with ``pow(2,n)``.\n\nNote: In the current implementation, the right-hand operand is required to\n  be at most ``sys.maxsize``.  If the right-hand operand is larger\n  than ``sys.maxsize`` an ``OverflowError`` exception is raised.\n',
  'slicings': '\nSlicings\n********\n\nA slicing selects a range of items in a sequence object (e.g., a\nstring, tuple or list).  Slicings may be used as expressions or as\ntargets in assignment or ``del`` statements.  The syntax for a\nslicing:\n\n   slicing      ::= primary "[" slice_list "]"\n   slice_list   ::= slice_item ("," slice_item)* [","]\n   slice_item   ::= expression | proper_slice\n   proper_slice ::= [lower_bound] ":" [upper_bound] [ ":" [stride] ]\n   lower_bound  ::= expression\n   upper_bound  ::= expression\n   stride       ::= expression\n\nThere is ambiguity in the formal syntax here: anything that looks like\nan expression list also looks like a slice list, so any subscription\ncan be interpreted as a slicing.  Rather than further complicating the\nsyntax, this is disambiguated by defining that in this case the\ninterpretation as a subscription takes priority over the\ninterpretation as a slicing (this is the case if the slice list\ncontains no proper slice).\n\nThe semantics for a slicing are as follows.  The primary must evaluate\nto a mapping object, and it is indexed (using the same\n``__getitem__()`` method as normal subscription) with a key that is\nconstructed from the slice list, as follows.  If the slice list\ncontains at least one comma, the key is a tuple containing the\nconversion of the slice items; otherwise, the conversion of the lone\nslice item is the key.  The conversion of a slice item that is an\nexpression is that expression.  The conversion of a proper slice is a\nslice object (see section *The standard type hierarchy*) whose\n``start``, ``stop`` and ``step`` attributes are the values of the\nexpressions given as lower bound, upper bound and stride,\nrespectively, substituting ``None`` for missing expressions.\n',
- 'specialattrs': "\nSpecial Attributes\n******************\n\nThe implementation adds a few special read-only attributes to several\nobject types, where they are relevant.  Some of these are not reported\nby the ``dir()`` built-in function.\n\nobject.__dict__\n\n   A dictionary or other mapping object used to store an object's\n   (writable) attributes.\n\ninstance.__class__\n\n   The class to which a class instance belongs.\n\nclass.__bases__\n\n   The tuple of base classes of a class object.  If there are no base\n   classes, this will be an empty tuple.\n\nclass.__name__\n\n   The name of the class or type.\n\nThe following attributes are only supported by *new-style class*es.\n\nclass.__mro__\n\n   This attribute is a tuple of classes that are considered when\n   looking for base classes during method resolution.\n\nclass.mro()\n\n   This method can be overridden by a metaclass to customize the\n   method resolution order for its instances.  It is called at class\n   instantiation, and its result is stored in ``__mro__``.\n\nclass.__subclasses__()\n\n   Each new-style class keeps a list of weak references to its\n   immediate subclasses.  This method returns a list of all those\n   references still alive. Example:\n\n      >>> int.__subclasses__()\n      [<type 'bool'>]\n\n-[ Footnotes ]-\n\n[1] Additional information on these special methods may be found in\n    the Python Reference Manual (*Basic customization*).\n\n[2] As a consequence, the list ``[1, 2]`` is considered equal to\n    ``[1.0, 2.0]``, and similarly for tuples.\n\n[3] They must have since the parser can't tell the type of the\n    operands.\n\n[4] To format only a tuple you should therefore provide a singleton\n    tuple whose only element is the tuple to be formatted.\n\n[5] The advantage of leaving the newline on is that returning an empty\n    string is then an unambiguous EOF indication.  It is also possible\n    (in cases where it might matter, for example, if you want to make\n    an exact copy of a file while scanning its lines) to tell whether\n    the last line of a file ended in a newline or not (yes this\n    happens!).\n",
- 'specialnames': '\nSpecial method names\n********************\n\nA class can implement certain operations that are invoked by special\nsyntax (such as arithmetic operations or subscripting and slicing) by\ndefining methods with special names. This is Python\'s approach to\n*operator overloading*, allowing classes to define their own behavior\nwith respect to language operators.  For instance, if a class defines\na method named ``__getitem__()``, and ``x`` is an instance of this\nclass, then ``x[i]`` is roughly equivalent to ``type(x).__getitem__(x,\ni)``.  Except where mentioned, attempts to execute an operation raise\nan exception when no appropriate method is defined (typically\n``AttributeError`` or ``TypeError``).\n\nWhen implementing a class that emulates any built-in type, it is\nimportant that the emulation only be implemented to the degree that it\nmakes sense for the object being modelled.  For example, some\nsequences may work well with retrieval of individual elements, but\nextracting a slice may not make sense.  (One example of this is the\n``NodeList`` interface in the W3C\'s Document Object Model.)\n\n\nBasic customization\n===================\n\nobject.__new__(cls[, ...])\n\n   Called to create a new instance of class *cls*.  ``__new__()`` is a\n   static method (special-cased so you need not declare it as such)\n   that takes the class of which an instance was requested as its\n   first argument.  The remaining arguments are those passed to the\n   object constructor expression (the call to the class).  The return\n   value of ``__new__()`` should be the new object instance (usually\n   an instance of *cls*).\n\n   Typical implementations create a new instance of the class by\n   invoking the superclass\'s ``__new__()`` method using\n   ``super(currentclass, cls).__new__(cls[, ...])`` with appropriate\n   arguments and then modifying the newly-created instance as\n   necessary before returning it.\n\n   If ``__new__()`` returns an instance of *cls*, then the new\n   instance\'s ``__init__()`` method will be invoked like\n   ``__init__(self[, ...])``, where *self* is the new instance and the\n   remaining arguments are the same as were passed to ``__new__()``.\n\n   If ``__new__()`` does not return an instance of *cls*, then the new\n   instance\'s ``__init__()`` method will not be invoked.\n\n   ``__new__()`` is intended mainly to allow subclasses of immutable\n   types (like int, str, or tuple) to customize instance creation.  It\n   is also commonly overridden in custom metaclasses in order to\n   customize class creation.\n\nobject.__init__(self[, ...])\n\n   Called when the instance is created.  The arguments are those\n   passed to the class constructor expression.  If a base class has an\n   ``__init__()`` method, the derived class\'s ``__init__()`` method,\n   if any, must explicitly call it to ensure proper initialization of\n   the base class part of the instance; for example:\n   ``BaseClass.__init__(self, [args...])``.  As a special constraint\n   on constructors, no value may be returned; doing so will cause a\n   ``TypeError`` to be raised at runtime.\n\nobject.__del__(self)\n\n   Called when the instance is about to be destroyed.  This is also\n   called a destructor.  If a base class has a ``__del__()`` method,\n   the derived class\'s ``__del__()`` method, if any, must explicitly\n   call it to ensure proper deletion of the base class part of the\n   instance.  Note that it is possible (though not recommended!) for\n   the ``__del__()`` method to postpone destruction of the instance by\n   creating a new reference to it.  It may then be called at a later\n   time when this new reference is deleted.  It is not guaranteed that\n   ``__del__()`` methods are called for objects that still exist when\n   the interpreter exits.\n\n   Note: ``del x`` doesn\'t directly call ``x.__del__()`` --- the former\n     decrements the reference count for ``x`` by one, and the latter\n     is only called when ``x``\'s reference count reaches zero.  Some\n     common situations that may prevent the reference count of an\n     object from going to zero include: circular references between\n     objects (e.g., a doubly-linked list or a tree data structure with\n     parent and child pointers); a reference to the object on the\n     stack frame of a function that caught an exception (the traceback\n     stored in ``sys.exc_info()[2]`` keeps the stack frame alive); or\n     a reference to the object on the stack frame that raised an\n     unhandled exception in interactive mode (the traceback stored in\n     ``sys.last_traceback`` keeps the stack frame alive).  The first\n     situation can only be remedied by explicitly breaking the cycles;\n     the latter two situations can be resolved by storing ``None`` in\n     ``sys.last_traceback``. Circular references which are garbage are\n     detected when the option cycle detector is enabled (it\'s on by\n     default), but can only be cleaned up if there are no Python-\n     level ``__del__()`` methods involved. Refer to the documentation\n     for the ``gc`` module for more information about how\n     ``__del__()`` methods are handled by the cycle detector,\n     particularly the description of the ``garbage`` value.\n\n   Warning: Due to the precarious circumstances under which ``__del__()``\n     methods are invoked, exceptions that occur during their execution\n     are ignored, and a warning is printed to ``sys.stderr`` instead.\n     Also, when ``__del__()`` is invoked in response to a module being\n     deleted (e.g., when execution of the program is done), other\n     globals referenced by the ``__del__()`` method may already have\n     been deleted or in the process of being torn down (e.g. the\n     import machinery shutting down).  For this reason, ``__del__()``\n     methods should do the absolute minimum needed to maintain\n     external invariants.  Starting with version 1.5, Python\n     guarantees that globals whose name begins with a single\n     underscore are deleted from their module before other globals are\n     deleted; if no other references to such globals exist, this may\n     help in assuring that imported modules are still available at the\n     time when the ``__del__()`` method is called.\n\nobject.__repr__(self)\n\n   Called by the ``repr()`` built-in function to compute the\n   "official" string representation of an object.  If at all possible,\n   this should look like a valid Python expression that could be used\n   to recreate an object with the same value (given an appropriate\n   environment).  If this is not possible, a string of the form\n   ``<...some useful description...>`` should be returned. The return\n   value must be a string object. If a class defines ``__repr__()``\n   but not ``__str__()``, then ``__repr__()`` is also used when an\n   "informal" string representation of instances of that class is\n   required.\n\n   This is typically used for debugging, so it is important that the\n   representation is information-rich and unambiguous.\n\nobject.__str__(self)\n\n   Called by the ``str()`` built-in function and by the ``print()``\n   function to compute the "informal" string representation of an\n   object.  This differs from ``__repr__()`` in that it does not have\n   to be a valid Python expression: a more convenient or concise\n   representation may be used instead. The return value must be a\n   string object.\n\nobject.__format__(self, format_spec)\n\n   Called by the ``format()`` built-in function (and by extension, the\n   ``format()`` method of class ``str``) to produce a "formatted"\n   string representation of an object. The ``format_spec`` argument is\n   a string that contains a description of the formatting options\n   desired. The interpretation of the ``format_spec`` argument is up\n   to the type implementing ``__format__()``, however most classes\n   will either delegate formatting to one of the built-in types, or\n   use a similar formatting option syntax.\n\n   See *Format Specification Mini-Language* for a description of the\n   standard formatting syntax.\n\n   The return value must be a string object.\n\nobject.__lt__(self, other)\nobject.__le__(self, other)\nobject.__eq__(self, other)\nobject.__ne__(self, other)\nobject.__gt__(self, other)\nobject.__ge__(self, other)\n\n   These are the so-called "rich comparison" methods. The\n   correspondence between operator symbols and method names is as\n   follows: ``x<y`` calls ``x.__lt__(y)``, ``x<=y`` calls\n   ``x.__le__(y)``, ``x==y`` calls ``x.__eq__(y)``, ``x!=y`` calls\n   ``x.__ne__(y)``, ``x>y`` calls ``x.__gt__(y)``, and ``x>=y`` calls\n   ``x.__ge__(y)``.\n\n   A rich comparison method may return the singleton\n   ``NotImplemented`` if it does not implement the operation for a\n   given pair of arguments. By convention, ``False`` and ``True`` are\n   returned for a successful comparison. However, these methods can\n   return any value, so if the comparison operator is used in a\n   Boolean context (e.g., in the condition of an ``if`` statement),\n   Python will call ``bool()`` on the value to determine if the result\n   is true or false.\n\n   There are no implied relationships among the comparison operators.\n   The truth of ``x==y`` does not imply that ``x!=y`` is false.\n   Accordingly, when defining ``__eq__()``, one should also define\n   ``__ne__()`` so that the operators will behave as expected.  See\n   the paragraph on ``__hash__()`` for some important notes on\n   creating *hashable* objects which support custom comparison\n   operations and are usable as dictionary keys.\n\n   There are no swapped-argument versions of these methods (to be used\n   when the left argument does not support the operation but the right\n   argument does); rather, ``__lt__()`` and ``__gt__()`` are each\n   other\'s reflection, ``__le__()`` and ``__ge__()`` are each other\'s\n   reflection, and ``__eq__()`` and ``__ne__()`` are their own\n   reflection.\n\n   Arguments to rich comparison methods are never coerced.\n\n   To automatically generate ordering operations from a single root\n   operation, see the Total Ordering recipe in the ASPN cookbook.\n\nobject.__hash__(self)\n\n   Called by built-in function ``hash()`` and for operations on\n   members of hashed collections including ``set``, ``frozenset``, and\n   ``dict``.  ``__hash__()`` should return an integer.  The only\n   required property is that objects which compare equal have the same\n   hash value; it is advised to somehow mix together (e.g. using\n   exclusive or) the hash values for the components of the object that\n   also play a part in comparison of objects.\n\n   If a class does not define an ``__eq__()`` method it should not\n   define a ``__hash__()`` operation either; if it defines\n   ``__eq__()`` but not ``__hash__()``, its instances will not be\n   usable as items in hashable collections.  If a class defines\n   mutable objects and implements an ``__eq__()`` method, it should\n   not implement ``__hash__()``, since the implementation of hashable\n   collections requires that a key\'s hash value is immutable (if the\n   object\'s hash value changes, it will be in the wrong hash bucket).\n\n   User-defined classes have ``__eq__()`` and ``__hash__()`` methods\n   by default; with them, all objects compare unequal (except with\n   themselves) and ``x.__hash__()`` returns ``id(x)``.\n\n   Classes which inherit a ``__hash__()`` method from a parent class\n   but change the meaning of ``__eq__()`` such that the hash value\n   returned is no longer appropriate (e.g. by switching to a value-\n   based concept of equality instead of the default identity based\n   equality) can explicitly flag themselves as being unhashable by\n   setting ``__hash__ = None`` in the class definition. Doing so means\n   that not only will instances of the class raise an appropriate\n   ``TypeError`` when a program attempts to retrieve their hash value,\n   but they will also be correctly identified as unhashable when\n   checking ``isinstance(obj, collections.Hashable)`` (unlike classes\n   which define their own ``__hash__()`` to explicitly raise\n   ``TypeError``).\n\n   If a class that overrides ``__eq__()`` needs to retain the\n   implementation of ``__hash__()`` from a parent class, the\n   interpreter must be told this explicitly by setting ``__hash__ =\n   <ParentClass>.__hash__``. Otherwise the inheritance of\n   ``__hash__()`` will be blocked, just as if ``__hash__`` had been\n   explicitly set to ``None``.\n\nobject.__bool__(self)\n\n   Called to implement truth value testing and the built-in operation\n   ``bool()``; should return ``False`` or ``True``, or their integer\n   equivalents ``0`` or ``1``.  When this method is not defined,\n   ``__len__()`` is called, if it is defined, and the object is\n   considered true if its result is nonzero. If a class defines\n   neither ``__len__()`` nor ``__bool__()``, all its instances are\n   considered true.\n\n\nCustomizing attribute access\n============================\n\nThe following methods can be defined to customize the meaning of\nattribute access (use of, assignment to, or deletion of ``x.name``)\nfor class instances.\n\nobject.__getattr__(self, name)\n\n   Called when an attribute lookup has not found the attribute in the\n   usual places (i.e. it is not an instance attribute nor is it found\n   in the class tree for ``self``).  ``name`` is the attribute name.\n   This method should return the (computed) attribute value or raise\n   an ``AttributeError`` exception.\n\n   Note that if the attribute is found through the normal mechanism,\n   ``__getattr__()`` is not called.  (This is an intentional asymmetry\n   between ``__getattr__()`` and ``__setattr__()``.) This is done both\n   for efficiency reasons and because otherwise ``__getattr__()``\n   would have no way to access other attributes of the instance.  Note\n   that at least for instance variables, you can fake total control by\n   not inserting any values in the instance attribute dictionary (but\n   instead inserting them in another object).  See the\n   ``__getattribute__()`` method below for a way to actually get total\n   control over attribute access.\n\nobject.__getattribute__(self, name)\n\n   Called unconditionally to implement attribute accesses for\n   instances of the class. If the class also defines\n   ``__getattr__()``, the latter will not be called unless\n   ``__getattribute__()`` either calls it explicitly or raises an\n   ``AttributeError``. This method should return the (computed)\n   attribute value or raise an ``AttributeError`` exception. In order\n   to avoid infinite recursion in this method, its implementation\n   should always call the base class method with the same name to\n   access any attributes it needs, for example,\n   ``object.__getattribute__(self, name)``.\n\n   Note: This method may still be bypassed when looking up special methods\n     as the result of implicit invocation via language syntax or\n     builtin functions. See *Special method lookup*.\n\nobject.__setattr__(self, name, value)\n\n   Called when an attribute assignment is attempted.  This is called\n   instead of the normal mechanism (i.e. store the value in the\n   instance dictionary). *name* is the attribute name, *value* is the\n   value to be assigned to it.\n\n   If ``__setattr__()`` wants to assign to an instance attribute, it\n   should call the base class method with the same name, for example,\n   ``object.__setattr__(self, name, value)``.\n\nobject.__delattr__(self, name)\n\n   Like ``__setattr__()`` but for attribute deletion instead of\n   assignment.  This should only be implemented if ``del obj.name`` is\n   meaningful for the object.\n\nobject.__dir__(self)\n\n   Called when ``dir()`` is called on the object.  A list must be\n   returned.\n\n\nImplementing Descriptors\n------------------------\n\nThe following methods only apply when an instance of the class\ncontaining the method (a so-called *descriptor* class) appears in the\nclass dictionary of another class, known as the *owner* class.  In the\nexamples below, "the attribute" refers to the attribute whose name is\nthe key of the property in the owner class\' ``__dict__``.\n\nobject.__get__(self, instance, owner)\n\n   Called to get the attribute of the owner class (class attribute\n   access) or of an instance of that class (instance attribute\n   access). *owner* is always the owner class, while *instance* is the\n   instance that the attribute was accessed through, or ``None`` when\n   the attribute is accessed through the *owner*.  This method should\n   return the (computed) attribute value or raise an\n   ``AttributeError`` exception.\n\nobject.__set__(self, instance, value)\n\n   Called to set the attribute on an instance *instance* of the owner\n   class to a new value, *value*.\n\nobject.__delete__(self, instance)\n\n   Called to delete the attribute on an instance *instance* of the\n   owner class.\n\n\nInvoking Descriptors\n--------------------\n\nIn general, a descriptor is an object attribute with "binding\nbehavior", one whose attribute access has been overridden by methods\nin the descriptor protocol:  ``__get__()``, ``__set__()``, and\n``__delete__()``. If any of those methods are defined for an object,\nit is said to be a descriptor.\n\nThe default behavior for attribute access is to get, set, or delete\nthe attribute from an object\'s dictionary. For instance, ``a.x`` has a\nlookup chain starting with ``a.__dict__[\'x\']``, then\n``type(a).__dict__[\'x\']``, and continuing through the base classes of\n``type(a)`` excluding metaclasses.\n\nHowever, if the looked-up value is an object defining one of the\ndescriptor methods, then Python may override the default behavior and\ninvoke the descriptor method instead.  Where this occurs in the\nprecedence chain depends on which descriptor methods were defined and\nhow they were called.\n\nThe starting point for descriptor invocation is a binding, ``a.x``.\nHow the arguments are assembled depends on ``a``:\n\nDirect Call\n   The simplest and least common call is when user code directly\n   invokes a descriptor method:    ``x.__get__(a)``.\n\nInstance Binding\n   If binding to an object instance, ``a.x`` is transformed into the\n   call: ``type(a).__dict__[\'x\'].__get__(a, type(a))``.\n\nClass Binding\n   If binding to a class, ``A.x`` is transformed into the call:\n   ``A.__dict__[\'x\'].__get__(None, A)``.\n\nSuper Binding\n   If ``a`` is an instance of ``super``, then the binding ``super(B,\n   obj).m()`` searches ``obj.__class__.__mro__`` for the base class\n   ``A`` immediately preceding ``B`` and then invokes the descriptor\n   with the call: ``A.__dict__[\'m\'].__get__(obj, A)``.\n\nFor instance bindings, the precedence of descriptor invocation depends\non the which descriptor methods are defined.  Normally, data\ndescriptors define both ``__get__()`` and ``__set__()``, while non-\ndata descriptors have just the ``__get__()`` method.  Data descriptors\nalways override a redefinition in an instance dictionary.  In\ncontrast, non-data descriptors can be overridden by instances. [2]\n\nPython methods (including ``staticmethod()`` and ``classmethod()``)\nare implemented as non-data descriptors.  Accordingly, instances can\nredefine and override methods.  This allows individual instances to\nacquire behaviors that differ from other instances of the same class.\n\nThe ``property()`` function is implemented as a data descriptor.\nAccordingly, instances cannot override the behavior of a property.\n\n\n__slots__\n---------\n\nBy default, instances of classes have a dictionary for attribute\nstorage.  This wastes space for objects having very few instance\nvariables.  The space consumption can become acute when creating large\nnumbers of instances.\n\nThe default can be overridden by defining *__slots__* in a class\ndefinition. The *__slots__* declaration takes a sequence of instance\nvariables and reserves just enough space in each instance to hold a\nvalue for each variable.  Space is saved because *__dict__* is not\ncreated for each instance.\n\nobject.__slots__\n\n   This class variable can be assigned a string, iterable, or sequence\n   of strings with variable names used by instances.  If defined in a\n   class, *__slots__* reserves space for the declared variables and\n   prevents the automatic creation of *__dict__* and *__weakref__* for\n   each instance.\n\n\nNotes on using *__slots__*\n~~~~~~~~~~~~~~~~~~~~~~~~~~\n\n* When inheriting from a class without *__slots__*, the *__dict__*\n  attribute of that class will always be accessible, so a *__slots__*\n  definition in the subclass is meaningless.\n\n* Without a *__dict__* variable, instances cannot be assigned new\n  variables not listed in the *__slots__* definition.  Attempts to\n  assign to an unlisted variable name raises ``AttributeError``. If\n  dynamic assignment of new variables is desired, then add\n  ``\'__dict__\'`` to the sequence of strings in the *__slots__*\n  declaration.\n\n* Without a *__weakref__* variable for each instance, classes defining\n  *__slots__* do not support weak references to its instances. If weak\n  reference support is needed, then add ``\'__weakref__\'`` to the\n  sequence of strings in the *__slots__* declaration.\n\n* *__slots__* are implemented at the class level by creating\n  descriptors (*Implementing Descriptors*) for each variable name.  As\n  a result, class attributes cannot be used to set default values for\n  instance variables defined by *__slots__*; otherwise, the class\n  attribute would overwrite the descriptor assignment.\n\n* If a class defines a slot also defined in a base class, the instance\n  variable defined by the base class slot is inaccessible (except by\n  retrieving its descriptor directly from the base class). This\n  renders the meaning of the program undefined.  In the future, a\n  check may be added to prevent this.\n\n* The action of a *__slots__* declaration is limited to the class\n  where it is defined.  As a result, subclasses will have a *__dict__*\n  unless they also define *__slots__*.\n\n* Nonempty *__slots__* does not work for classes derived from\n  "variable-length" built-in types such as ``int``, ``str`` and\n  ``tuple``.\n\n* Any non-string iterable may be assigned to *__slots__*. Mappings may\n  also be used; however, in the future, special meaning may be\n  assigned to the values corresponding to each key.\n\n* *__class__* assignment works only if both classes have the same\n  *__slots__*.\n\n\nCustomizing class creation\n==========================\n\nBy default, classes are constructed using ``type()``. A class\ndefinition is read into a separate namespace and the value of class\nname is bound to the result of ``type(name, bases, dict)``.\n\nWhen the class definition is read, if a callable ``metaclass`` keyword\nargument is passed after the bases in the class definition, the\ncallable given will be called instead of ``type()``.  If other keyword\narguments are passed, they will also be passed to the metaclass.  This\nallows classes or functions to be written which monitor or alter the\nclass creation process:\n\n* Modifying the class dictionary prior to the class being created.\n\n* Returning an instance of another class -- essentially performing the\n  role of a factory function.\n\nThese steps will have to be performed in the metaclass\'s ``__new__()``\nmethod -- ``type.__new__()`` can then be called from this method to\ncreate a class with different properties.  This example adds a new\nelement to the class dictionary before creating the class:\n\n   class metacls(type):\n       def __new__(mcs, name, bases, dict):\n           dict[\'foo\'] = \'metacls was here\'\n           return type.__new__(mcs, name, bases, dict)\n\nYou can of course also override other class methods (or add new\nmethods); for example defining a custom ``__call__()`` method in the\nmetaclass allows custom behavior when the class is called, e.g. not\nalways creating a new instance.\n\nIf the metaclass has a ``__prepare__()`` attribute (usually\nimplemented as a class or static method), it is called before the\nclass body is evaluated with the name of the class and a tuple of its\nbases for arguments.  It should return an object that supports the\nmapping interface that will be used to store the namespace of the\nclass.  The default is a plain dictionary.  This could be used, for\nexample, to keep track of the order that class attributes are declared\nin by returning an ordered dictionary.\n\nThe appropriate metaclass is determined by the following precedence\nrules:\n\n* If the ``metaclass`` keyword argument is based with the bases, it is\n  used.\n\n* Otherwise, if there is at least one base class, its metaclass is\n  used.\n\n* Otherwise, the default metaclass (``type``) is used.\n\nThe potential uses for metaclasses are boundless. Some ideas that have\nbeen explored including logging, interface checking, automatic\ndelegation, automatic property creation, proxies, frameworks, and\nautomatic resource locking/synchronization.\n\nHere is an example of a metaclass that uses an\n``collections.OrderedDict`` to remember the order that class members\nwere defined:\n\n   class OrderedClass(type):\n\n        @classmethod\n        def __prepare__(metacls, name, bases, **kwds):\n           return collections.OrderedDict()\n\n        def __new__(cls, name, bases, classdict):\n           result = type.__new__(cls, name, bases, dict(classdict))\n           result.members = tuple(classdict)\n           return result\n\n   class A(metaclass=OrderedClass):\n       def one(self): pass\n       def two(self): pass\n       def three(self): pass\n       def four(self): pass\n\n   >>> A.members\n   (\'__module__\', \'one\', \'two\', \'three\', \'four\')\n\nWhen the class definition for *A* gets executed, the process begins\nwith calling the metaclass\'s ``__prepare__()`` method which returns an\nempty ``collections.OrderedDict``.  That mapping records the methods\nand attributes of *A* as they are defined within the body of the class\nstatement. Once those definitions are executed, the ordered dictionary\nis fully populated and the metaclass\'s ``__new__()`` method gets\ninvoked.  That method builds the new type and it saves the ordered\ndictionary keys in an attribute called *members*.\n\n\nEmulating callable objects\n==========================\n\nobject.__call__(self[, args...])\n\n   Called when the instance is "called" as a function; if this method\n   is defined, ``x(arg1, arg2, ...)`` is a shorthand for\n   ``x.__call__(arg1, arg2, ...)``.\n\n\nEmulating container types\n=========================\n\nThe following methods can be defined to implement container objects.\nContainers usually are sequences (such as lists or tuples) or mappings\n(like dictionaries), but can represent other containers as well.  The\nfirst set of methods is used either to emulate a sequence or to\nemulate a mapping; the difference is that for a sequence, the\nallowable keys should be the integers *k* for which ``0 <= k < N``\nwhere *N* is the length of the sequence, or slice objects, which\ndefine a range of items.  It is also recommended that mappings provide\nthe methods ``keys()``, ``values()``, ``items()``, ``get()``,\n``clear()``, ``setdefault()``, ``pop()``, ``popitem()``, ``copy()``,\nand ``update()`` behaving similar to those for Python\'s standard\ndictionary objects.  The ``collections`` module provides a\n``MutableMapping`` abstract base class to help create those methods\nfrom a base set of ``__getitem__()``, ``__setitem__()``,\n``__delitem__()``, and ``keys()``. Mutable sequences should provide\nmethods ``append()``, ``count()``, ``index()``, ``extend()``,\n``insert()``, ``pop()``, ``remove()``, ``reverse()`` and ``sort()``,\nlike Python standard list objects.  Finally, sequence types should\nimplement addition (meaning concatenation) and multiplication (meaning\nrepetition) by defining the methods ``__add__()``, ``__radd__()``,\n``__iadd__()``, ``__mul__()``, ``__rmul__()`` and ``__imul__()``\ndescribed below; they should not define other numerical operators.  It\nis recommended that both mappings and sequences implement the\n``__contains__()`` method to allow efficient use of the ``in``\noperator; for mappings, ``in`` should search the mapping\'s keys; for\nsequences, it should search through the values.  It is further\nrecommended that both mappings and sequences implement the\n``__iter__()`` method to allow efficient iteration through the\ncontainer; for mappings, ``__iter__()`` should be the same as\n``keys()``; for sequences, it should iterate through the values.\n\nobject.__len__(self)\n\n   Called to implement the built-in function ``len()``.  Should return\n   the length of the object, an integer ``>=`` 0.  Also, an object\n   that doesn\'t define a ``__bool__()`` method and whose ``__len__()``\n   method returns zero is considered to be false in a Boolean context.\n\nNote: Slicing is done exclusively with the following three methods.  A\n  call like\n\n     a[1:2] = b\n\n  is translated to\n\n     a[slice(1, 2, None)] = b\n\n  and so forth.  Missing slice items are always filled in with\n  ``None``.\n\nobject.__getitem__(self, key)\n\n   Called to implement evaluation of ``self[key]``. For sequence\n   types, the accepted keys should be integers and slice objects.\n   Note that the special interpretation of negative indexes (if the\n   class wishes to emulate a sequence type) is up to the\n   ``__getitem__()`` method. If *key* is of an inappropriate type,\n   ``TypeError`` may be raised; if of a value outside the set of\n   indexes for the sequence (after any special interpretation of\n   negative values), ``IndexError`` should be raised. For mapping\n   types, if *key* is missing (not in the container), ``KeyError``\n   should be raised.\n\n   Note: ``for`` loops expect that an ``IndexError`` will be raised for\n     illegal indexes to allow proper detection of the end of the\n     sequence.\n\nobject.__setitem__(self, key, value)\n\n   Called to implement assignment to ``self[key]``.  Same note as for\n   ``__getitem__()``.  This should only be implemented for mappings if\n   the objects support changes to the values for keys, or if new keys\n   can be added, or for sequences if elements can be replaced.  The\n   same exceptions should be raised for improper *key* values as for\n   the ``__getitem__()`` method.\n\nobject.__delitem__(self, key)\n\n   Called to implement deletion of ``self[key]``.  Same note as for\n   ``__getitem__()``.  This should only be implemented for mappings if\n   the objects support removal of keys, or for sequences if elements\n   can be removed from the sequence.  The same exceptions should be\n   raised for improper *key* values as for the ``__getitem__()``\n   method.\n\nobject.__iter__(self)\n\n   This method is called when an iterator is required for a container.\n   This method should return a new iterator object that can iterate\n   over all the objects in the container.  For mappings, it should\n   iterate over the keys of the container, and should also be made\n   available as the method ``keys()``.\n\n   Iterator objects also need to implement this method; they are\n   required to return themselves.  For more information on iterator\n   objects, see *Iterator Types*.\n\nobject.__reversed__(self)\n\n   Called (if present) by the ``reversed()`` builtin to implement\n   reverse iteration.  It should return a new iterator object that\n   iterates over all the objects in the container in reverse order.\n\n   If the ``__reversed__()`` method is not provided, the\n   ``reversed()`` builtin will fall back to using the sequence\n   protocol (``__len__()`` and ``__getitem__()``).  Objects that\n   support the sequence protocol should only provide\n   ``__reversed__()`` if they can provide an implementation that is\n   more efficient than the one provided by ``reversed()``.\n\nThe membership test operators (``in`` and ``not in``) are normally\nimplemented as an iteration through a sequence.  However, container\nobjects can supply the following special method with a more efficient\nimplementation, which also does not require the object be a sequence.\n\nobject.__contains__(self, item)\n\n   Called to implement membership test operators.  Should return true\n   if *item* is in *self*, false otherwise.  For mapping objects, this\n   should consider the keys of the mapping rather than the values or\n   the key-item pairs.\n\n\nEmulating numeric types\n=======================\n\nThe following methods can be defined to emulate numeric objects.\nMethods corresponding to operations that are not supported by the\nparticular kind of number implemented (e.g., bitwise operations for\nnon-integral numbers) should be left undefined.\n\nobject.__add__(self, other)\nobject.__sub__(self, other)\nobject.__mul__(self, other)\nobject.__truediv__(self, other)\nobject.__floordiv__(self, other)\nobject.__mod__(self, other)\nobject.__divmod__(self, other)\nobject.__pow__(self, other[, modulo])\nobject.__lshift__(self, other)\nobject.__rshift__(self, other)\nobject.__and__(self, other)\nobject.__xor__(self, other)\nobject.__or__(self, other)\n\n   These methods are called to implement the binary arithmetic\n   operations (``+``, ``-``, ``*``, ``/``, ``//``, ``%``,\n   ``divmod()``, ``pow()``, ``**``, ``<<``, ``>>``, ``&``, ``^``,\n   ``|``).  For instance, to evaluate the expression ``x + y``, where\n   *x* is an instance of a class that has an ``__add__()`` method,\n   ``x.__add__(y)`` is called.  The ``__divmod__()`` method should be\n   the equivalent to using ``__floordiv__()`` and ``__mod__()``; it\n   should not be related to ``__truediv__()``.  Note that\n   ``__pow__()`` should be defined to accept an optional third\n   argument if the ternary version of the built-in ``pow()`` function\n   is to be supported.\n\n   If one of those methods does not support the operation with the\n   supplied arguments, it should return ``NotImplemented``.\n\nobject.__radd__(self, other)\nobject.__rsub__(self, other)\nobject.__rmul__(self, other)\nobject.__rtruediv__(self, other)\nobject.__rfloordiv__(self, other)\nobject.__rmod__(self, other)\nobject.__rdivmod__(self, other)\nobject.__rpow__(self, other)\nobject.__rlshift__(self, other)\nobject.__rrshift__(self, other)\nobject.__rand__(self, other)\nobject.__rxor__(self, other)\nobject.__ror__(self, other)\n\n   These methods are called to implement the binary arithmetic\n   operations (``+``, ``-``, ``*``, ``/``, ``//``, ``%``,\n   ``divmod()``, ``pow()``, ``**``, ``<<``, ``>>``, ``&``, ``^``,\n   ``|``) with reflected (swapped) operands. These functions are only\n   called if the left operand does not support the corresponding\n   operation and the operands are of different types. [3]  For\n   instance, to evaluate the expression ``x - y``, where *y* is an\n   instance of a class that has an ``__rsub__()`` method,\n   ``y.__rsub__(x)`` is called if ``x.__sub__(y)`` returns\n   *NotImplemented*.\n\n   Note that ternary ``pow()`` will not try calling ``__rpow__()``\n   (the coercion rules would become too complicated).\n\n   Note: If the right operand\'s type is a subclass of the left operand\'s\n     type and that subclass provides the reflected method for the\n     operation, this method will be called before the left operand\'s\n     non-reflected method.  This behavior allows subclasses to\n     override their ancestors\' operations.\n\nobject.__iadd__(self, other)\nobject.__isub__(self, other)\nobject.__imul__(self, other)\nobject.__itruediv__(self, other)\nobject.__ifloordiv__(self, other)\nobject.__imod__(self, other)\nobject.__ipow__(self, other[, modulo])\nobject.__ilshift__(self, other)\nobject.__irshift__(self, other)\nobject.__iand__(self, other)\nobject.__ixor__(self, other)\nobject.__ior__(self, other)\n\n   These methods are called to implement the augmented arithmetic\n   assignments (``+=``, ``-=``, ``*=``, ``/=``, ``//=``, ``%=``,\n   ``**=``, ``<<=``, ``>>=``, ``&=``, ``^=``, ``|=``).  These methods\n   should attempt to do the operation in-place (modifying *self*) and\n   return the result (which could be, but does not have to be,\n   *self*).  If a specific method is not defined, the augmented\n   assignment falls back to the normal methods.  For instance, to\n   execute the statement ``x += y``, where *x* is an instance of a\n   class that has an ``__iadd__()`` method, ``x.__iadd__(y)`` is\n   called.  If *x* is an instance of a class that does not define a\n   ``__iadd__()`` method, ``x.__add__(y)`` and ``y.__radd__(x)`` are\n   considered, as with the evaluation of ``x + y``.\n\nobject.__neg__(self)\nobject.__pos__(self)\nobject.__abs__(self)\nobject.__invert__(self)\n\n   Called to implement the unary arithmetic operations (``-``, ``+``,\n   ``abs()`` and ``~``).\n\nobject.__complex__(self)\nobject.__int__(self)\nobject.__float__(self)\nobject.__round__(self[, n])\n\n   Called to implement the built-in functions ``complex()``,\n   ``int()``, ``float()`` and ``round()``.  Should return a value of\n   the appropriate type.\n\nobject.__index__(self)\n\n   Called to implement ``operator.index()``.  Also called whenever\n   Python needs an integer object (such as in slicing, or in the\n   built-in ``bin()``, ``hex()`` and ``oct()`` functions). Must return\n   an integer.\n\n\nWith Statement Context Managers\n===============================\n\nA *context manager* is an object that defines the runtime context to\nbe established when executing a ``with`` statement. The context\nmanager handles the entry into, and the exit from, the desired runtime\ncontext for the execution of the block of code.  Context managers are\nnormally invoked using the ``with`` statement (described in section\n*The with statement*), but can also be used by directly invoking their\nmethods.\n\nTypical uses of context managers include saving and restoring various\nkinds of global state, locking and unlocking resources, closing opened\nfiles, etc.\n\nFor more information on context managers, see *Context Manager Types*.\n\nobject.__enter__(self)\n\n   Enter the runtime context related to this object. The ``with``\n   statement will bind this method\'s return value to the target(s)\n   specified in the ``as`` clause of the statement, if any.\n\nobject.__exit__(self, exc_type, exc_value, traceback)\n\n   Exit the runtime context related to this object. The parameters\n   describe the exception that caused the context to be exited. If the\n   context was exited without an exception, all three arguments will\n   be ``None``.\n\n   If an exception is supplied, and the method wishes to suppress the\n   exception (i.e., prevent it from being propagated), it should\n   return a true value. Otherwise, the exception will be processed\n   normally upon exit from this method.\n\n   Note that ``__exit__()`` methods should not reraise the passed-in\n   exception; this is the caller\'s responsibility.\n\nSee also:\n\n   **PEP 0343** - The "with" statement\n      The specification, background, and examples for the Python\n      ``with`` statement.\n\n\nSpecial method lookup\n=====================\n\nFor custom classes, implicit invocations of special methods are only\nguaranteed to work correctly if defined on an object\'s type, not in\nthe object\'s instance dictionary.  That behaviour is the reason why\nthe following code raises an exception:\n\n   >>> class C(object):\n   ...     pass\n   ...\n   >>> c = C()\n   >>> c.__len__ = lambda: 5\n   >>> len(c)\n   Traceback (most recent call last):\n     File "<stdin>", line 1, in <module>\n   TypeError: object of type \'C\' has no len()\n\nThe rationale behind this behaviour lies with a number of special\nmethods such as ``__hash__()`` and ``__repr__()`` that are implemented\nby all objects, including type objects. If the implicit lookup of\nthese methods used the conventional lookup process, they would fail\nwhen invoked on the type object itself:\n\n   >>> 1 .__hash__() == hash(1)\n   True\n   >>> int.__hash__() == hash(int)\n   Traceback (most recent call last):\n     File "<stdin>", line 1, in <module>\n   TypeError: descriptor \'__hash__\' of \'int\' object needs an argument\n\nIncorrectly attempting to invoke an unbound method of a class in this\nway is sometimes referred to as \'metaclass confusion\', and is avoided\nby bypassing the instance when looking up special methods:\n\n   >>> type(1).__hash__(1) == hash(1)\n   True\n   >>> type(int).__hash__(int) == hash(int)\n   True\n\nIn addition to bypassing any instance attributes in the interest of\ncorrectness, implicit special method lookup generally also bypasses\nthe ``__getattribute__()`` method even of the object\'s metaclass:\n\n   >>> class Meta(type):\n   ...    def __getattribute__(*args):\n   ...       print("Metaclass getattribute invoked")\n   ...       return type.__getattribute__(*args)\n   ...\n   >>> class C(object, metaclass=Meta):\n   ...     def __len__(self):\n   ...         return 10\n   ...     def __getattribute__(*args):\n   ...         print("Class getattribute invoked")\n   ...         return object.__getattribute__(*args)\n   ...\n   >>> c = C()\n   >>> c.__len__()                 # Explicit lookup via instance\n   Class getattribute invoked\n   10\n   >>> type(c).__len__(c)          # Explicit lookup via type\n   Metaclass getattribute invoked\n   10\n   >>> len(c)                      # Implicit lookup\n   10\n\nBypassing the ``__getattribute__()`` machinery in this fashion\nprovides significant scope for speed optimisations within the\ninterpreter, at the cost of some flexibility in the handling of\nspecial methods (the special method *must* be set on the class object\nitself in order to be consistently invoked by the interpreter).\n\n-[ Footnotes ]-\n\n[1] It *is* possible in some cases to change an object\'s type, under\n    certain controlled conditions. It generally isn\'t a good idea\n    though, since it can lead to some very strange behaviour if it is\n    handled incorrectly.\n\n[2] A descriptor can define any combination of ``__get__()``,\n    ``__set__()`` and ``__delete__()``.  If it does not define\n    ``__get__()``, then accessing the attribute even on an instance\n    will return the descriptor object itself.  If the descriptor\n    defines ``__set__()`` and/or ``__delete__()``, it is a data\n    descriptor; if it defines neither, it is a non-data descriptor.\n\n[3] For operands of the same type, it is assumed that if the non-\n    reflected method (such as ``__add__()``) fails the operation is\n    not supported, which is why the reflected method is not called.\n',
- 'string-methods': '\nString Methods\n**************\n\nString objects support the methods listed below.  Note that none of\nthese methods take keyword arguments.\n\nIn addition, Python\'s strings support the sequence type methods\ndescribed in the *Sequence Types --- str, bytes, bytearray, list,\ntuple, range* section. To output formatted strings, see the *String\nFormatting* section. Also, see the ``re`` module for string functions\nbased on regular expressions.\n\nstr.capitalize()\n\n   Return a copy of the string with only its first character\n   capitalized.\n\nstr.center(width[, fillchar])\n\n   Return centered in a string of length *width*. Padding is done\n   using the specified *fillchar* (default is a space).\n\nstr.count(sub[, start[, end]])\n\n   Return the number of non-overlapping occurrences of substring *sub*\n   in the range [*start*, *end*].  Optional arguments *start* and\n   *end* are interpreted as in slice notation.\n\nstr.encode([encoding[, errors]])\n\n   Return an encoded version of the string as a bytes object.  Default\n   encoding is the current default string encoding.  *errors* may be\n   given to set a different error handling scheme.  The default for\n   *errors* is ``\'strict\'``, meaning that encoding errors raise a\n   ``UnicodeError``.  Other possible values are ``\'ignore\'``,\n   ``\'replace\'``, ``\'xmlcharrefreplace\'``, ``\'backslashreplace\'`` and\n   any other name registered via ``codecs.register_error()``, see\n   section *Codec Base Classes*. For a list of possible encodings, see\n   section *Standard Encodings*.\n\nstr.endswith(suffix[, start[, end]])\n\n   Return ``True`` if the string ends with the specified *suffix*,\n   otherwise return ``False``.  *suffix* can also be a tuple of\n   suffixes to look for.  With optional *start*, test beginning at\n   that position.  With optional *end*, stop comparing at that\n   position.\n\nstr.expandtabs([tabsize])\n\n   Return a copy of the string where all tab characters are replaced\n   by one or more spaces, depending on the current column and the\n   given tab size.  The column number is reset to zero after each\n   newline occurring in the string. If *tabsize* is not given, a tab\n   size of ``8`` characters is assumed.  This doesn\'t understand other\n   non-printing characters or escape sequences.\n\nstr.find(sub[, start[, end]])\n\n   Return the lowest index in the string where substring *sub* is\n   found, such that *sub* is contained in the range [*start*, *end*].\n   Optional arguments *start* and *end* are interpreted as in slice\n   notation.  Return ``-1`` if *sub* is not found.\n\nstr.format(*args, **kwargs)\n\n   Perform a string formatting operation.  The *format_string*\n   argument can contain literal text or replacement fields delimited\n   by braces ``{}``.  Each replacement field contains either the\n   numeric index of a positional argument, or the name of a keyword\n   argument.  Returns a copy of *format_string* where each replacement\n   field is replaced with the string value of the corresponding\n   argument.\n\n   >>> "The sum of 1 + 2 is {0}".format(1+2)\n   \'The sum of 1 + 2 is 3\'\n\n   See *Format String Syntax* for a description of the various\n   formatting options that can be specified in format strings.\n\nstr.index(sub[, start[, end]])\n\n   Like ``find()``, but raise ``ValueError`` when the substring is not\n   found.\n\nstr.isalnum()\n\n   Return true if all characters in the string are alphanumeric and\n   there is at least one character, false otherwise.\n\nstr.isalpha()\n\n   Return true if all characters in the string are alphabetic and\n   there is at least one character, false otherwise.\n\nstr.isdecimal()\n\n   Return true if all characters in the string are decimal characters\n   and there is at least one character, false otherwise. Decimal\n   characters include digit characters, and all characters that that\n   can be used to form decimal-radix numbers, e.g. U+0660, ARABIC-\n   INDIC DIGIT ZERO.\n\nstr.isdigit()\n\n   Return true if all characters in the string are digits and there is\n   at least one character, false otherwise.\n\nstr.isidentifier()\n\n   Return true if the string is a valid identifier according to the\n   language definition, section *Identifiers and keywords*.\n\nstr.islower()\n\n   Return true if all cased characters in the string are lowercase and\n   there is at least one cased character, false otherwise.\n\nstr.isnumeric()\n\n   Return true if all characters in the string are numeric characters,\n   and there is at least one character, false otherwise. Numeric\n   characters include digit characters, and all characters that have\n   the Unicode numeric value property, e.g. U+2155, VULGAR FRACTION\n   ONE FIFTH.\n\nstr.isprintable()\n\n   Return true if all characters in the string are printable or the\n   string is empty, false otherwise.  Nonprintable characters are\n   those characters defined in the Unicode character database as\n   "Other" or "Separator", excepting the ASCII space (0x20) which is\n   considered printable.  (Note that printable characters in this\n   context are those which should not be escaped when ``repr()`` is\n   invoked on a string.  It has no bearing on the handling of strings\n   written to ``sys.stdout`` or ``sys.stderr``.)\n\nstr.isspace()\n\n   Return true if there are only whitespace characters in the string\n   and there is at least one character, false otherwise.\n\nstr.istitle()\n\n   Return true if the string is a titlecased string and there is at\n   least one character, for example uppercase characters may only\n   follow uncased characters and lowercase characters only cased ones.\n   Return false otherwise.\n\nstr.isupper()\n\n   Return true if all cased characters in the string are uppercase and\n   there is at least one cased character, false otherwise.\n\nstr.join(seq)\n\n   Return a string which is the concatenation of the strings in the\n   sequence *seq*.  A ``TypeError`` will be raised if there are any\n   non-string values in *seq*, including ``bytes`` objects.  The\n   separator between elements is the string providing this method.\n\nstr.ljust(width[, fillchar])\n\n   Return the string left justified in a string of length *width*.\n   Padding is done using the specified *fillchar* (default is a\n   space).  The original string is returned if *width* is less than\n   ``len(s)``.\n\nstr.lower()\n\n   Return a copy of the string converted to lowercase.\n\nstr.lstrip([chars])\n\n   Return a copy of the string with leading characters removed.  The\n   *chars* argument is a string specifying the set of characters to be\n   removed.  If omitted or ``None``, the *chars* argument defaults to\n   removing whitespace.  The *chars* argument is not a prefix; rather,\n   all combinations of its values are stripped:\n\n   >>> \'   spacious   \'.lstrip()\n   \'spacious   \'\n   >>> \'www.example.com\'.lstrip(\'cmowz.\')\n   \'example.com\'\n\nstatic str.maketrans(x[, y[, z]])\n\n   This static method returns a translation table usable for\n   ``str.translate()``.\n\n   If there is only one argument, it must be a dictionary mapping\n   Unicode ordinals (integers) or characters (strings of length 1) to\n   Unicode ordinals, strings (of arbitrary lengths) or None.\n   Character keys will then be converted to ordinals.\n\n   If there are two arguments, they must be strings of equal length,\n   and in the resulting dictionary, each character in x will be mapped\n   to the character at the same position in y.  If there is a third\n   argument, it must be a string, whose characters will be mapped to\n   None in the result.\n\nstr.partition(sep)\n\n   Split the string at the first occurrence of *sep*, and return a\n   3-tuple containing the part before the separator, the separator\n   itself, and the part after the separator.  If the separator is not\n   found, return a 3-tuple containing the string itself, followed by\n   two empty strings.\n\nstr.replace(old, new[, count])\n\n   Return a copy of the string with all occurrences of substring *old*\n   replaced by *new*.  If the optional argument *count* is given, only\n   the first *count* occurrences are replaced.\n\nstr.rfind(sub[, start[, end]])\n\n   Return the highest index in the string where substring *sub* is\n   found, such that *sub* is contained within s[start,end].  Optional\n   arguments *start* and *end* are interpreted as in slice notation.\n   Return ``-1`` on failure.\n\nstr.rindex(sub[, start[, end]])\n\n   Like ``rfind()`` but raises ``ValueError`` when the substring *sub*\n   is not found.\n\nstr.rjust(width[, fillchar])\n\n   Return the string right justified in a string of length *width*.\n   Padding is done using the specified *fillchar* (default is a\n   space). The original string is returned if *width* is less than\n   ``len(s)``.\n\nstr.rpartition(sep)\n\n   Split the string at the last occurrence of *sep*, and return a\n   3-tuple containing the part before the separator, the separator\n   itself, and the part after the separator.  If the separator is not\n   found, return a 3-tuple containing two empty strings, followed by\n   the string itself.\n\nstr.rsplit([sep[, maxsplit]])\n\n   Return a list of the words in the string, using *sep* as the\n   delimiter string. If *maxsplit* is given, at most *maxsplit* splits\n   are done, the *rightmost* ones.  If *sep* is not specified or\n   ``None``, any whitespace string is a separator.  Except for\n   splitting from the right, ``rsplit()`` behaves like ``split()``\n   which is described in detail below.\n\nstr.rstrip([chars])\n\n   Return a copy of the string with trailing characters removed.  The\n   *chars* argument is a string specifying the set of characters to be\n   removed.  If omitted or ``None``, the *chars* argument defaults to\n   removing whitespace.  The *chars* argument is not a suffix; rather,\n   all combinations of its values are stripped:\n\n   >>> \'   spacious   \'.rstrip()\n   \'   spacious\'\n   >>> \'mississippi\'.rstrip(\'ipz\')\n   \'mississ\'\n\nstr.split([sep[, maxsplit]])\n\n   Return a list of the words in the string, using *sep* as the\n   delimiter string.  If *maxsplit* is given, at most *maxsplit*\n   splits are done (thus, the list will have at most ``maxsplit+1``\n   elements).  If *maxsplit* is not specified, then there is no limit\n   on the number of splits (all possible splits are made).\n\n   If *sep* is given, consecutive delimiters are not grouped together\n   and are deemed to delimit empty strings (for example,\n   ``\'1,,2\'.split(\',\')`` returns ``[\'1\', \'\', \'2\']``).  The *sep*\n   argument may consist of multiple characters (for example,\n   ``\'1<>2<>3\'.split(\'<>\')`` returns ``[\'1\', \'2\', \'3\']``). Splitting\n   an empty string with a specified separator returns ``[\'\']``.\n\n   If *sep* is not specified or is ``None``, a different splitting\n   algorithm is applied: runs of consecutive whitespace are regarded\n   as a single separator, and the result will contain no empty strings\n   at the start or end if the string has leading or trailing\n   whitespace.  Consequently, splitting an empty string or a string\n   consisting of just whitespace with a ``None`` separator returns\n   ``[]``.\n\n   For example, ``\' 1  2   3  \'.split()`` returns ``[\'1\', \'2\', \'3\']``,\n   and ``\'  1  2   3  \'.split(None, 1)`` returns ``[\'1\', \'2   3  \']``.\n\nstr.splitlines([keepends])\n\n   Return a list of the lines in the string, breaking at line\n   boundaries.  Line breaks are not included in the resulting list\n   unless *keepends* is given and true.\n\nstr.startswith(prefix[, start[, end]])\n\n   Return ``True`` if string starts with the *prefix*, otherwise\n   return ``False``. *prefix* can also be a tuple of prefixes to look\n   for.  With optional *start*, test string beginning at that\n   position.  With optional *end*, stop comparing string at that\n   position.\n\nstr.strip([chars])\n\n   Return a copy of the string with the leading and trailing\n   characters removed. The *chars* argument is a string specifying the\n   set of characters to be removed. If omitted or ``None``, the\n   *chars* argument defaults to removing whitespace. The *chars*\n   argument is not a prefix or suffix; rather, all combinations of its\n   values are stripped:\n\n   >>> \'   spacious   \'.strip()\n   \'spacious\'\n   >>> \'www.example.com\'.strip(\'cmowz.\')\n   \'example\'\n\nstr.swapcase()\n\n   Return a copy of the string with uppercase characters converted to\n   lowercase and vice versa.\n\nstr.title()\n\n   Return a titlecased version of the string: words start with\n   uppercase characters, all remaining cased characters are lowercase.\n\nstr.translate(map)\n\n   Return a copy of the *s* where all characters have been mapped\n   through the *map* which must be a dictionary of Unicode ordinals\n   (integers) to Unicode ordinals, strings or ``None``.  Unmapped\n   characters are left untouched. Characters mapped to ``None`` are\n   deleted.\n\n   You can use ``str.maketrans()`` to create a translation map from\n   character-to-character mappings in different formats.\n\n   Note: An even more flexible approach is to create a custom character\n     mapping codec using the ``codecs`` module (see\n     ``encodings.cp1251`` for an example).\n\nstr.upper()\n\n   Return a copy of the string converted to uppercase.\n\nstr.zfill(width)\n\n   Return the numeric string left filled with zeros in a string of\n   length *width*.  A sign prefix is handled correctly.  The original\n   string is returned if *width* is less than ``len(s)``.\n',
- 'strings': '\nString and Bytes literals\n*************************\n\nString literals are described by the following lexical definitions:\n\n   stringliteral   ::= [stringprefix](shortstring | longstring)\n   stringprefix    ::= "r" | "R"\n   shortstring     ::= "\'" shortstringitem* "\'" | \'"\' shortstringitem* \'"\'\n   longstring      ::= "\'\'\'" longstringitem* "\'\'\'" | \'"""\' longstringitem* \'"""\'\n   shortstringitem ::= shortstringchar | stringescapeseq\n   longstringitem  ::= longstringchar | stringescapeseq\n   shortstringchar ::= <any source character except "\\" or newline or the quote>\n   longstringchar  ::= <any source character except "\\">\n   stringescapeseq ::= "\\" <any source character>\n\n   bytesliteral   ::= bytesprefix(shortbytes | longbytes)\n   bytesprefix    ::= "b" | "B"\n   shortbytes     ::= "\'" shortbytesitem* "\'" | \'"\' shortbytesitem* \'"\'\n   longbytes      ::= "\'\'\'" longbytesitem* "\'\'\'" | \'"""\' longbytesitem* \'"""\'\n   shortbytesitem ::= shortbyteschar | bytesescapeseq\n   longbytesitem  ::= longbyteschar | bytesescapeseq\n   shortbyteschar ::= <any ASCII character except "\\" or newline or the quote>\n   longbyteschar  ::= <any ASCII character except "\\">\n   bytesescapeseq ::= "\\" <any ASCII character>\n\nOne syntactic restriction not indicated by these productions is that\nwhitespace is not allowed between the **stringprefix** or\n**bytesprefix** and the rest of the literal. The source character set\nis defined by the encoding declaration; it is UTF-8 if no encoding\ndeclaration is given in the source file; see section *Encoding\ndeclarations*.\n\nIn plain English: Both types of literals can be enclosed in matching\nsingle quotes (``\'``) or double quotes (``"``).  They can also be\nenclosed in matching groups of three single or double quotes (these\nare generally referred to as *triple-quoted strings*).  The backslash\n(``\\``) character is used to escape characters that otherwise have a\nspecial meaning, such as newline, backslash itself, or the quote\ncharacter.\n\nString literals may optionally be prefixed with a letter ``\'r\'`` or\n``\'R\'``; such strings are called *raw strings* and treat backslashes\nas literal characters.  As a result, ``\'\\U\'`` and ``\'\\u\'`` escapes in\nraw strings are not treated specially.\n\nBytes literals are always prefixed with ``\'b\'`` or ``\'B\'``; they\nproduce an instance of the ``bytes`` type instead of the ``str`` type.\nThey may only contain ASCII characters; bytes with a numeric value of\n128 or greater must be expressed with escapes.\n\nIn triple-quoted strings, unescaped newlines and quotes are allowed\n(and are retained), except that three unescaped quotes in a row\nterminate the string.  (A "quote" is the character used to open the\nstring, i.e. either ``\'`` or ``"``.)\n\nUnless an ``\'r\'`` or ``\'R\'`` prefix is present, escape sequences in\nstrings are interpreted according to rules similar to those used by\nStandard C.  The recognized escape sequences are:\n\n+-------------------+-----------------------------------+---------+\n| Escape Sequence   | Meaning                           | Notes   |\n+===================+===================================+=========+\n| ``\\newline``      | Backslash and newline ignored     |         |\n+-------------------+-----------------------------------+---------+\n| ``\\\\``            | Backslash (``\\``)                 |         |\n+-------------------+-----------------------------------+---------+\n| ``\\\'``            | Single quote (``\'``)              |         |\n+-------------------+-----------------------------------+---------+\n| ``\\"``            | Double quote (``"``)              |         |\n+-------------------+-----------------------------------+---------+\n| ``\\a``            | ASCII Bell (BEL)                  |         |\n+-------------------+-----------------------------------+---------+\n| ``\\b``            | ASCII Backspace (BS)              |         |\n+-------------------+-----------------------------------+---------+\n| ``\\f``            | ASCII Formfeed (FF)               |         |\n+-------------------+-----------------------------------+---------+\n| ``\\n``            | ASCII Linefeed (LF)               |         |\n+-------------------+-----------------------------------+---------+\n| ``\\r``            | ASCII Carriage Return (CR)        |         |\n+-------------------+-----------------------------------+---------+\n| ``\\t``            | ASCII Horizontal Tab (TAB)        |         |\n+-------------------+-----------------------------------+---------+\n| ``\\v``            | ASCII Vertical Tab (VT)           |         |\n+-------------------+-----------------------------------+---------+\n| ``\\ooo``          | Character with octal value *ooo*  | (1,3)   |\n+-------------------+-----------------------------------+---------+\n| ``\\xhh``          | Character with hex value *hh*     | (2,3)   |\n+-------------------+-----------------------------------+---------+\n\nEscape sequences only recognized in string literals are:\n\n+-------------------+-----------------------------------+---------+\n| Escape Sequence   | Meaning                           | Notes   |\n+===================+===================================+=========+\n| ``\\N{name}``      | Character named *name* in the     |         |\n|                   | Unicode database                  |         |\n+-------------------+-----------------------------------+---------+\n| ``\\uxxxx``        | Character with 16-bit hex value   | (4)     |\n|                   | *xxxx*                            |         |\n+-------------------+-----------------------------------+---------+\n| ``\\Uxxxxxxxx``    | Character with 32-bit hex value   | (5)     |\n|                   | *xxxxxxxx*                        |         |\n+-------------------+-----------------------------------+---------+\n\nNotes:\n\n1. As in Standard C, up to three octal digits are accepted.\n\n2. Unlike in Standard C, at most two hex digits are accepted.\n\n3. In a bytes literal, hexadecimal and octal escapes denote the byte\n   with the given value. In a string literal, these escapes denote a\n   Unicode character with the given value.\n\n4. Individual code units which form parts of a surrogate pair can be\n   encoded using this escape sequence. Unlike in Standard C, exactly\n   two hex digits are required.\n\n5. Any Unicode character can be encoded this way, but characters\n   outside the Basic Multilingual Plane (BMP) will be encoded using a\n   surrogate pair if Python is compiled to use 16-bit code units (the\n   default).  Individual code units which form parts of a surrogate\n   pair can be encoded using this escape sequence.\n\nUnlike Standard C, all unrecognized escape sequences are left in the\nstring unchanged, i.e., *the backslash is left in the string*.  (This\nbehavior is useful when debugging: if an escape sequence is mistyped,\nthe resulting output is more easily recognized as broken.)  It is also\nimportant to note that the escape sequences only recognized in string\nliterals fall into the category of unrecognized escapes for bytes\nliterals.\n\nEven in a raw string, string quotes can be escaped with a backslash,\nbut the backslash remains in the string; for example, ``r"\\""`` is a\nvalid string literal consisting of two characters: a backslash and a\ndouble quote; ``r"\\"`` is not a valid string literal (even a raw\nstring cannot end in an odd number of backslashes).  Specifically, *a\nraw string cannot end in a single backslash* (since the backslash\nwould escape the following quote character).  Note also that a single\nbackslash followed by a newline is interpreted as those two characters\nas part of the string, *not* as a line continuation.\n',
+ 'specialattrs': "\nSpecial Attributes\n******************\n\nThe implementation adds a few special read-only attributes to several\nobject types, where they are relevant.  Some of these are not reported\nby the ``dir()`` built-in function.\n\nobject.__dict__\n\n   A dictionary or other mapping object used to store an object's\n   (writable) attributes.\n\ninstance.__class__\n\n   The class to which a class instance belongs.\n\nclass.__bases__\n\n   The tuple of base classes of a class object.\n\nclass.__name__\n\n   The name of the class or type.\n\nThe following attributes are only supported by *new-style class*es.\n\nclass.__mro__\n\n   This attribute is a tuple of classes that are considered when\n   looking for base classes during method resolution.\n\nclass.mro()\n\n   This method can be overridden by a metaclass to customize the\n   method resolution order for its instances.  It is called at class\n   instantiation, and its result is stored in ``__mro__``.\n\nclass.__subclasses__()\n\n   Each new-style class keeps a list of weak references to its\n   immediate subclasses.  This method returns a list of all those\n   references still alive. Example:\n\n      >>> int.__subclasses__()\n      [<type 'bool'>]\n\n-[ Footnotes ]-\n\n[1] Additional information on these special methods may be found in\n    the Python Reference Manual (*Basic customization*).\n\n[2] As a consequence, the list ``[1, 2]`` is considered equal to\n    ``[1.0, 2.0]``, and similarly for tuples.\n\n[3] They must have since the parser can't tell the type of the\n    operands.\n\n[4] To format only a tuple you should therefore provide a singleton\n    tuple whose only element is the tuple to be formatted.\n",
+ 'specialnames': '\nSpecial method names\n********************\n\nA class can implement certain operations that are invoked by special\nsyntax (such as arithmetic operations or subscripting and slicing) by\ndefining methods with special names. This is Python\'s approach to\n*operator overloading*, allowing classes to define their own behavior\nwith respect to language operators.  For instance, if a class defines\na method named ``__getitem__()``, and ``x`` is an instance of this\nclass, then ``x[i]`` is roughly equivalent to ``type(x).__getitem__(x,\ni)``.  Except where mentioned, attempts to execute an operation raise\nan exception when no appropriate method is defined (typically\n``AttributeError`` or ``TypeError``).\n\nWhen implementing a class that emulates any built-in type, it is\nimportant that the emulation only be implemented to the degree that it\nmakes sense for the object being modelled.  For example, some\nsequences may work well with retrieval of individual elements, but\nextracting a slice may not make sense.  (One example of this is the\n``NodeList`` interface in the W3C\'s Document Object Model.)\n\n\nBasic customization\n===================\n\nobject.__new__(cls[, ...])\n\n   Called to create a new instance of class *cls*.  ``__new__()`` is a\n   static method (special-cased so you need not declare it as such)\n   that takes the class of which an instance was requested as its\n   first argument.  The remaining arguments are those passed to the\n   object constructor expression (the call to the class).  The return\n   value of ``__new__()`` should be the new object instance (usually\n   an instance of *cls*).\n\n   Typical implementations create a new instance of the class by\n   invoking the superclass\'s ``__new__()`` method using\n   ``super(currentclass, cls).__new__(cls[, ...])`` with appropriate\n   arguments and then modifying the newly-created instance as\n   necessary before returning it.\n\n   If ``__new__()`` returns an instance of *cls*, then the new\n   instance\'s ``__init__()`` method will be invoked like\n   ``__init__(self[, ...])``, where *self* is the new instance and the\n   remaining arguments are the same as were passed to ``__new__()``.\n\n   If ``__new__()`` does not return an instance of *cls*, then the new\n   instance\'s ``__init__()`` method will not be invoked.\n\n   ``__new__()`` is intended mainly to allow subclasses of immutable\n   types (like int, str, or tuple) to customize instance creation.  It\n   is also commonly overridden in custom metaclasses in order to\n   customize class creation.\n\nobject.__init__(self[, ...])\n\n   Called when the instance is created.  The arguments are those\n   passed to the class constructor expression.  If a base class has an\n   ``__init__()`` method, the derived class\'s ``__init__()`` method,\n   if any, must explicitly call it to ensure proper initialization of\n   the base class part of the instance; for example:\n   ``BaseClass.__init__(self, [args...])``.  As a special constraint\n   on constructors, no value may be returned; doing so will cause a\n   ``TypeError`` to be raised at runtime.\n\nobject.__del__(self)\n\n   Called when the instance is about to be destroyed.  This is also\n   called a destructor.  If a base class has a ``__del__()`` method,\n   the derived class\'s ``__del__()`` method, if any, must explicitly\n   call it to ensure proper deletion of the base class part of the\n   instance.  Note that it is possible (though not recommended!) for\n   the ``__del__()`` method to postpone destruction of the instance by\n   creating a new reference to it.  It may then be called at a later\n   time when this new reference is deleted.  It is not guaranteed that\n   ``__del__()`` methods are called for objects that still exist when\n   the interpreter exits.\n\n   Note: ``del x`` doesn\'t directly call ``x.__del__()`` --- the former\n     decrements the reference count for ``x`` by one, and the latter\n     is only called when ``x``\'s reference count reaches zero.  Some\n     common situations that may prevent the reference count of an\n     object from going to zero include: circular references between\n     objects (e.g., a doubly-linked list or a tree data structure with\n     parent and child pointers); a reference to the object on the\n     stack frame of a function that caught an exception (the traceback\n     stored in ``sys.exc_info()[2]`` keeps the stack frame alive); or\n     a reference to the object on the stack frame that raised an\n     unhandled exception in interactive mode (the traceback stored in\n     ``sys.last_traceback`` keeps the stack frame alive).  The first\n     situation can only be remedied by explicitly breaking the cycles;\n     the latter two situations can be resolved by storing ``None`` in\n     ``sys.last_traceback``. Circular references which are garbage are\n     detected when the option cycle detector is enabled (it\'s on by\n     default), but can only be cleaned up if there are no Python-\n     level ``__del__()`` methods involved. Refer to the documentation\n     for the ``gc`` module for more information about how\n     ``__del__()`` methods are handled by the cycle detector,\n     particularly the description of the ``garbage`` value.\n\n   Warning: Due to the precarious circumstances under which ``__del__()``\n     methods are invoked, exceptions that occur during their execution\n     are ignored, and a warning is printed to ``sys.stderr`` instead.\n     Also, when ``__del__()`` is invoked in response to a module being\n     deleted (e.g., when execution of the program is done), other\n     globals referenced by the ``__del__()`` method may already have\n     been deleted or in the process of being torn down (e.g. the\n     import machinery shutting down).  For this reason, ``__del__()``\n     methods should do the absolute minimum needed to maintain\n     external invariants.  Starting with version 1.5, Python\n     guarantees that globals whose name begins with a single\n     underscore are deleted from their module before other globals are\n     deleted; if no other references to such globals exist, this may\n     help in assuring that imported modules are still available at the\n     time when the ``__del__()`` method is called.\n\nobject.__repr__(self)\n\n   Called by the ``repr()`` built-in function to compute the\n   "official" string representation of an object.  If at all possible,\n   this should look like a valid Python expression that could be used\n   to recreate an object with the same value (given an appropriate\n   environment).  If this is not possible, a string of the form\n   ``<...some useful description...>`` should be returned. The return\n   value must be a string object. If a class defines ``__repr__()``\n   but not ``__str__()``, then ``__repr__()`` is also used when an\n   "informal" string representation of instances of that class is\n   required.\n\n   This is typically used for debugging, so it is important that the\n   representation is information-rich and unambiguous.\n\nobject.__str__(self)\n\n   Called by the ``str()`` built-in function and by the ``print()``\n   function to compute the "informal" string representation of an\n   object.  This differs from ``__repr__()`` in that it does not have\n   to be a valid Python expression: a more convenient or concise\n   representation may be used instead. The return value must be a\n   string object.\n\nobject.__format__(self, format_spec)\n\n   Called by the ``format()`` built-in function (and by extension, the\n   ``format()`` method of class ``str``) to produce a "formatted"\n   string representation of an object. The ``format_spec`` argument is\n   a string that contains a description of the formatting options\n   desired. The interpretation of the ``format_spec`` argument is up\n   to the type implementing ``__format__()``, however most classes\n   will either delegate formatting to one of the built-in types, or\n   use a similar formatting option syntax.\n\n   See *Format Specification Mini-Language* for a description of the\n   standard formatting syntax.\n\n   The return value must be a string object.\n\nobject.__lt__(self, other)\nobject.__le__(self, other)\nobject.__eq__(self, other)\nobject.__ne__(self, other)\nobject.__gt__(self, other)\nobject.__ge__(self, other)\n\n   These are the so-called "rich comparison" methods. The\n   correspondence between operator symbols and method names is as\n   follows: ``x<y`` calls ``x.__lt__(y)``, ``x<=y`` calls\n   ``x.__le__(y)``, ``x==y`` calls ``x.__eq__(y)``, ``x!=y`` calls\n   ``x.__ne__(y)``, ``x>y`` calls ``x.__gt__(y)``, and ``x>=y`` calls\n   ``x.__ge__(y)``.\n\n   A rich comparison method may return the singleton\n   ``NotImplemented`` if it does not implement the operation for a\n   given pair of arguments. By convention, ``False`` and ``True`` are\n   returned for a successful comparison. However, these methods can\n   return any value, so if the comparison operator is used in a\n   Boolean context (e.g., in the condition of an ``if`` statement),\n   Python will call ``bool()`` on the value to determine if the result\n   is true or false.\n\n   There are no implied relationships among the comparison operators.\n   The truth of ``x==y`` does not imply that ``x!=y`` is false.\n   Accordingly, when defining ``__eq__()``, one should also define\n   ``__ne__()`` so that the operators will behave as expected.  See\n   the paragraph on ``__hash__()`` for some important notes on\n   creating *hashable* objects which support custom comparison\n   operations and are usable as dictionary keys.\n\n   There are no swapped-argument versions of these methods (to be used\n   when the left argument does not support the operation but the right\n   argument does); rather, ``__lt__()`` and ``__gt__()`` are each\n   other\'s reflection, ``__le__()`` and ``__ge__()`` are each other\'s\n   reflection, and ``__eq__()`` and ``__ne__()`` are their own\n   reflection.\n\n   Arguments to rich comparison methods are never coerced.\n\n   To automatically generate ordering operations from a single root\n   operation, see ``functools.total_ordering()``.\n\nobject.__hash__(self)\n\n   Called by built-in function ``hash()`` and for operations on\n   members of hashed collections including ``set``, ``frozenset``, and\n   ``dict``.  ``__hash__()`` should return an integer.  The only\n   required property is that objects which compare equal have the same\n   hash value; it is advised to somehow mix together (e.g. using\n   exclusive or) the hash values for the components of the object that\n   also play a part in comparison of objects.\n\n   If a class does not define an ``__eq__()`` method it should not\n   define a ``__hash__()`` operation either; if it defines\n   ``__eq__()`` but not ``__hash__()``, its instances will not be\n   usable as items in hashable collections.  If a class defines\n   mutable objects and implements an ``__eq__()`` method, it should\n   not implement ``__hash__()``, since the implementation of hashable\n   collections requires that a key\'s hash value is immutable (if the\n   object\'s hash value changes, it will be in the wrong hash bucket).\n\n   User-defined classes have ``__eq__()`` and ``__hash__()`` methods\n   by default; with them, all objects compare unequal (except with\n   themselves) and ``x.__hash__()`` returns ``id(x)``.\n\n   Classes which inherit a ``__hash__()`` method from a parent class\n   but change the meaning of ``__eq__()`` such that the hash value\n   returned is no longer appropriate (e.g. by switching to a value-\n   based concept of equality instead of the default identity based\n   equality) can explicitly flag themselves as being unhashable by\n   setting ``__hash__ = None`` in the class definition. Doing so means\n   that not only will instances of the class raise an appropriate\n   ``TypeError`` when a program attempts to retrieve their hash value,\n   but they will also be correctly identified as unhashable when\n   checking ``isinstance(obj, collections.Hashable)`` (unlike classes\n   which define their own ``__hash__()`` to explicitly raise\n   ``TypeError``).\n\n   If a class that overrides ``__eq__()`` needs to retain the\n   implementation of ``__hash__()`` from a parent class, the\n   interpreter must be told this explicitly by setting ``__hash__ =\n   <ParentClass>.__hash__``. Otherwise the inheritance of\n   ``__hash__()`` will be blocked, just as if ``__hash__`` had been\n   explicitly set to ``None``.\n\nobject.__bool__(self)\n\n   Called to implement truth value testing and the built-in operation\n   ``bool()``; should return ``False`` or ``True``.  When this method\n   is not defined, ``__len__()`` is called, if it is defined, and the\n   object is considered true if its result is nonzero.  If a class\n   defines neither ``__len__()`` nor ``__bool__()``, all its instances\n   are considered true.\n\n\nCustomizing attribute access\n============================\n\nThe following methods can be defined to customize the meaning of\nattribute access (use of, assignment to, or deletion of ``x.name``)\nfor class instances.\n\nobject.__getattr__(self, name)\n\n   Called when an attribute lookup has not found the attribute in the\n   usual places (i.e. it is not an instance attribute nor is it found\n   in the class tree for ``self``).  ``name`` is the attribute name.\n   This method should return the (computed) attribute value or raise\n   an ``AttributeError`` exception.\n\n   Note that if the attribute is found through the normal mechanism,\n   ``__getattr__()`` is not called.  (This is an intentional asymmetry\n   between ``__getattr__()`` and ``__setattr__()``.) This is done both\n   for efficiency reasons and because otherwise ``__getattr__()``\n   would have no way to access other attributes of the instance.  Note\n   that at least for instance variables, you can fake total control by\n   not inserting any values in the instance attribute dictionary (but\n   instead inserting them in another object).  See the\n   ``__getattribute__()`` method below for a way to actually get total\n   control over attribute access.\n\nobject.__getattribute__(self, name)\n\n   Called unconditionally to implement attribute accesses for\n   instances of the class. If the class also defines\n   ``__getattr__()``, the latter will not be called unless\n   ``__getattribute__()`` either calls it explicitly or raises an\n   ``AttributeError``. This method should return the (computed)\n   attribute value or raise an ``AttributeError`` exception. In order\n   to avoid infinite recursion in this method, its implementation\n   should always call the base class method with the same name to\n   access any attributes it needs, for example,\n   ``object.__getattribute__(self, name)``.\n\n   Note: This method may still be bypassed when looking up special methods\n     as the result of implicit invocation via language syntax or\n     built-in functions. See *Special method lookup*.\n\nobject.__setattr__(self, name, value)\n\n   Called when an attribute assignment is attempted.  This is called\n   instead of the normal mechanism (i.e. store the value in the\n   instance dictionary). *name* is the attribute name, *value* is the\n   value to be assigned to it.\n\n   If ``__setattr__()`` wants to assign to an instance attribute, it\n   should call the base class method with the same name, for example,\n   ``object.__setattr__(self, name, value)``.\n\nobject.__delattr__(self, name)\n\n   Like ``__setattr__()`` but for attribute deletion instead of\n   assignment.  This should only be implemented if ``del obj.name`` is\n   meaningful for the object.\n\nobject.__dir__(self)\n\n   Called when ``dir()`` is called on the object.  A list must be\n   returned.\n\n\nImplementing Descriptors\n------------------------\n\nThe following methods only apply when an instance of the class\ncontaining the method (a so-called *descriptor* class) appears in the\nclass dictionary of another class, known as the *owner* class.  In the\nexamples below, "the attribute" refers to the attribute whose name is\nthe key of the property in the owner class\' ``__dict__``.\n\nobject.__get__(self, instance, owner)\n\n   Called to get the attribute of the owner class (class attribute\n   access) or of an instance of that class (instance attribute\n   access). *owner* is always the owner class, while *instance* is the\n   instance that the attribute was accessed through, or ``None`` when\n   the attribute is accessed through the *owner*.  This method should\n   return the (computed) attribute value or raise an\n   ``AttributeError`` exception.\n\nobject.__set__(self, instance, value)\n\n   Called to set the attribute on an instance *instance* of the owner\n   class to a new value, *value*.\n\nobject.__delete__(self, instance)\n\n   Called to delete the attribute on an instance *instance* of the\n   owner class.\n\n\nInvoking Descriptors\n--------------------\n\nIn general, a descriptor is an object attribute with "binding\nbehavior", one whose attribute access has been overridden by methods\nin the descriptor protocol:  ``__get__()``, ``__set__()``, and\n``__delete__()``. If any of those methods are defined for an object,\nit is said to be a descriptor.\n\nThe default behavior for attribute access is to get, set, or delete\nthe attribute from an object\'s dictionary. For instance, ``a.x`` has a\nlookup chain starting with ``a.__dict__[\'x\']``, then\n``type(a).__dict__[\'x\']``, and continuing through the base classes of\n``type(a)`` excluding metaclasses.\n\nHowever, if the looked-up value is an object defining one of the\ndescriptor methods, then Python may override the default behavior and\ninvoke the descriptor method instead.  Where this occurs in the\nprecedence chain depends on which descriptor methods were defined and\nhow they were called.\n\nThe starting point for descriptor invocation is a binding, ``a.x``.\nHow the arguments are assembled depends on ``a``:\n\nDirect Call\n   The simplest and least common call is when user code directly\n   invokes a descriptor method:    ``x.__get__(a)``.\n\nInstance Binding\n   If binding to an object instance, ``a.x`` is transformed into the\n   call: ``type(a).__dict__[\'x\'].__get__(a, type(a))``.\n\nClass Binding\n   If binding to a class, ``A.x`` is transformed into the call:\n   ``A.__dict__[\'x\'].__get__(None, A)``.\n\nSuper Binding\n   If ``a`` is an instance of ``super``, then the binding ``super(B,\n   obj).m()`` searches ``obj.__class__.__mro__`` for the base class\n   ``A`` immediately preceding ``B`` and then invokes the descriptor\n   with the call: ``A.__dict__[\'m\'].__get__(obj, A)``.\n\nFor instance bindings, the precedence of descriptor invocation depends\non the which descriptor methods are defined.  A descriptor can define\nany combination of ``__get__()``, ``__set__()`` and ``__delete__()``.\nIf it does not define ``__get__()``, then accessing the attribute will\nreturn the descriptor object itself unless there is a value in the\nobject\'s instance dictionary.  If the descriptor defines ``__set__()``\nand/or ``__delete__()``, it is a data descriptor; if it defines\nneither, it is a non-data descriptor.  Normally, data descriptors\ndefine both ``__get__()`` and ``__set__()``, while non-data\ndescriptors have just the ``__get__()`` method.  Data descriptors with\n``__set__()`` and ``__get__()`` defined always override a redefinition\nin an instance dictionary.  In contrast, non-data descriptors can be\noverridden by instances.\n\nPython methods (including ``staticmethod()`` and ``classmethod()``)\nare implemented as non-data descriptors.  Accordingly, instances can\nredefine and override methods.  This allows individual instances to\nacquire behaviors that differ from other instances of the same class.\n\nThe ``property()`` function is implemented as a data descriptor.\nAccordingly, instances cannot override the behavior of a property.\n\n\n__slots__\n---------\n\nBy default, instances of classes have a dictionary for attribute\nstorage.  This wastes space for objects having very few instance\nvariables.  The space consumption can become acute when creating large\nnumbers of instances.\n\nThe default can be overridden by defining *__slots__* in a class\ndefinition. The *__slots__* declaration takes a sequence of instance\nvariables and reserves just enough space in each instance to hold a\nvalue for each variable.  Space is saved because *__dict__* is not\ncreated for each instance.\n\nobject.__slots__\n\n   This class variable can be assigned a string, iterable, or sequence\n   of strings with variable names used by instances.  If defined in a\n   class, *__slots__* reserves space for the declared variables and\n   prevents the automatic creation of *__dict__* and *__weakref__* for\n   each instance.\n\n\nNotes on using *__slots__*\n~~~~~~~~~~~~~~~~~~~~~~~~~~\n\n* When inheriting from a class without *__slots__*, the *__dict__*\n  attribute of that class will always be accessible, so a *__slots__*\n  definition in the subclass is meaningless.\n\n* Without a *__dict__* variable, instances cannot be assigned new\n  variables not listed in the *__slots__* definition.  Attempts to\n  assign to an unlisted variable name raises ``AttributeError``. If\n  dynamic assignment of new variables is desired, then add\n  ``\'__dict__\'`` to the sequence of strings in the *__slots__*\n  declaration.\n\n* Without a *__weakref__* variable for each instance, classes defining\n  *__slots__* do not support weak references to its instances. If weak\n  reference support is needed, then add ``\'__weakref__\'`` to the\n  sequence of strings in the *__slots__* declaration.\n\n* *__slots__* are implemented at the class level by creating\n  descriptors (*Implementing Descriptors*) for each variable name.  As\n  a result, class attributes cannot be used to set default values for\n  instance variables defined by *__slots__*; otherwise, the class\n  attribute would overwrite the descriptor assignment.\n\n* The action of a *__slots__* declaration is limited to the class\n  where it is defined.  As a result, subclasses will have a *__dict__*\n  unless they also define *__slots__* (which must only contain names\n  of any *additional* slots).\n\n* If a class defines a slot also defined in a base class, the instance\n  variable defined by the base class slot is inaccessible (except by\n  retrieving its descriptor directly from the base class). This\n  renders the meaning of the program undefined.  In the future, a\n  check may be added to prevent this.\n\n* Nonempty *__slots__* does not work for classes derived from\n  "variable-length" built-in types such as ``int``, ``str`` and\n  ``tuple``.\n\n* Any non-string iterable may be assigned to *__slots__*. Mappings may\n  also be used; however, in the future, special meaning may be\n  assigned to the values corresponding to each key.\n\n* *__class__* assignment works only if both classes have the same\n  *__slots__*.\n\n\nCustomizing class creation\n==========================\n\nBy default, classes are constructed using ``type()``. A class\ndefinition is read into a separate namespace and the value of class\nname is bound to the result of ``type(name, bases, dict)``.\n\nWhen the class definition is read, if a callable ``metaclass`` keyword\nargument is passed after the bases in the class definition, the\ncallable given will be called instead of ``type()``.  If other keyword\narguments are passed, they will also be passed to the metaclass.  This\nallows classes or functions to be written which monitor or alter the\nclass creation process:\n\n* Modifying the class dictionary prior to the class being created.\n\n* Returning an instance of another class -- essentially performing the\n  role of a factory function.\n\nThese steps will have to be performed in the metaclass\'s ``__new__()``\nmethod -- ``type.__new__()`` can then be called from this method to\ncreate a class with different properties.  This example adds a new\nelement to the class dictionary before creating the class:\n\n   class metacls(type):\n       def __new__(mcs, name, bases, dict):\n           dict[\'foo\'] = \'metacls was here\'\n           return type.__new__(mcs, name, bases, dict)\n\nYou can of course also override other class methods (or add new\nmethods); for example defining a custom ``__call__()`` method in the\nmetaclass allows custom behavior when the class is called, e.g. not\nalways creating a new instance.\n\nIf the metaclass has a ``__prepare__()`` attribute (usually\nimplemented as a class or static method), it is called before the\nclass body is evaluated with the name of the class and a tuple of its\nbases for arguments.  It should return an object that supports the\nmapping interface that will be used to store the namespace of the\nclass.  The default is a plain dictionary.  This could be used, for\nexample, to keep track of the order that class attributes are declared\nin by returning an ordered dictionary.\n\nThe appropriate metaclass is determined by the following precedence\nrules:\n\n* If the ``metaclass`` keyword argument is passed with the bases, it\n  is used.\n\n* Otherwise, if there is at least one base class, its metaclass is\n  used.\n\n* Otherwise, the default metaclass (``type``) is used.\n\nThe potential uses for metaclasses are boundless. Some ideas that have\nbeen explored including logging, interface checking, automatic\ndelegation, automatic property creation, proxies, frameworks, and\nautomatic resource locking/synchronization.\n\nHere is an example of a metaclass that uses an\n``collections.OrderedDict`` to remember the order that class members\nwere defined:\n\n   class OrderedClass(type):\n\n        @classmethod\n        def __prepare__(metacls, name, bases, **kwds):\n           return collections.OrderedDict()\n\n        def __new__(cls, name, bases, classdict):\n           result = type.__new__(cls, name, bases, dict(classdict))\n           result.members = tuple(classdict)\n           return result\n\n   class A(metaclass=OrderedClass):\n       def one(self): pass\n       def two(self): pass\n       def three(self): pass\n       def four(self): pass\n\n   >>> A.members\n   (\'__module__\', \'one\', \'two\', \'three\', \'four\')\n\nWhen the class definition for *A* gets executed, the process begins\nwith calling the metaclass\'s ``__prepare__()`` method which returns an\nempty ``collections.OrderedDict``.  That mapping records the methods\nand attributes of *A* as they are defined within the body of the class\nstatement. Once those definitions are executed, the ordered dictionary\nis fully populated and the metaclass\'s ``__new__()`` method gets\ninvoked.  That method builds the new type and it saves the ordered\ndictionary keys in an attribute called *members*.\n\n\nCustomizing instance and subclass checks\n========================================\n\nThe following methods are used to override the default behavior of the\n``isinstance()`` and ``issubclass()`` built-in functions.\n\nIn particular, the metaclass ``abc.ABCMeta`` implements these methods\nin order to allow the addition of Abstract Base Classes (ABCs) as\n"virtual base classes" to any class or type (including built-in\ntypes), including other ABCs.\n\nclass.__instancecheck__(self, instance)\n\n   Return true if *instance* should be considered a (direct or\n   indirect) instance of *class*. If defined, called to implement\n   ``isinstance(instance, class)``.\n\nclass.__subclasscheck__(self, subclass)\n\n   Return true if *subclass* should be considered a (direct or\n   indirect) subclass of *class*.  If defined, called to implement\n   ``issubclass(subclass, class)``.\n\nNote that these methods are looked up on the type (metaclass) of a\nclass.  They cannot be defined as class methods in the actual class.\nThis is consistent with the lookup of special methods that are called\non instances, only in this case the instance is itself a class.\n\nSee also:\n\n   **PEP 3119** - Introducing Abstract Base Classes\n      Includes the specification for customizing ``isinstance()`` and\n      ``issubclass()`` behavior through ``__instancecheck__()`` and\n      ``__subclasscheck__()``, with motivation for this functionality\n      in the context of adding Abstract Base Classes (see the ``abc``\n      module) to the language.\n\n\nEmulating callable objects\n==========================\n\nobject.__call__(self[, args...])\n\n   Called when the instance is "called" as a function; if this method\n   is defined, ``x(arg1, arg2, ...)`` is a shorthand for\n   ``x.__call__(arg1, arg2, ...)``.\n\n\nEmulating container types\n=========================\n\nThe following methods can be defined to implement container objects.\nContainers usually are sequences (such as lists or tuples) or mappings\n(like dictionaries), but can represent other containers as well.  The\nfirst set of methods is used either to emulate a sequence or to\nemulate a mapping; the difference is that for a sequence, the\nallowable keys should be the integers *k* for which ``0 <= k < N``\nwhere *N* is the length of the sequence, or slice objects, which\ndefine a range of items.  It is also recommended that mappings provide\nthe methods ``keys()``, ``values()``, ``items()``, ``get()``,\n``clear()``, ``setdefault()``, ``pop()``, ``popitem()``, ``copy()``,\nand ``update()`` behaving similar to those for Python\'s standard\ndictionary objects.  The ``collections`` module provides a\n``MutableMapping`` abstract base class to help create those methods\nfrom a base set of ``__getitem__()``, ``__setitem__()``,\n``__delitem__()``, and ``keys()``. Mutable sequences should provide\nmethods ``append()``, ``count()``, ``index()``, ``extend()``,\n``insert()``, ``pop()``, ``remove()``, ``reverse()`` and ``sort()``,\nlike Python standard list objects.  Finally, sequence types should\nimplement addition (meaning concatenation) and multiplication (meaning\nrepetition) by defining the methods ``__add__()``, ``__radd__()``,\n``__iadd__()``, ``__mul__()``, ``__rmul__()`` and ``__imul__()``\ndescribed below; they should not define other numerical operators.  It\nis recommended that both mappings and sequences implement the\n``__contains__()`` method to allow efficient use of the ``in``\noperator; for mappings, ``in`` should search the mapping\'s keys; for\nsequences, it should search through the values.  It is further\nrecommended that both mappings and sequences implement the\n``__iter__()`` method to allow efficient iteration through the\ncontainer; for mappings, ``__iter__()`` should be the same as\n``keys()``; for sequences, it should iterate through the values.\n\nobject.__len__(self)\n\n   Called to implement the built-in function ``len()``.  Should return\n   the length of the object, an integer ``>=`` 0.  Also, an object\n   that doesn\'t define a ``__bool__()`` method and whose ``__len__()``\n   method returns zero is considered to be false in a Boolean context.\n\nNote: Slicing is done exclusively with the following three methods.  A\n  call like\n\n     a[1:2] = b\n\n  is translated to\n\n     a[slice(1, 2, None)] = b\n\n  and so forth.  Missing slice items are always filled in with\n  ``None``.\n\nobject.__getitem__(self, key)\n\n   Called to implement evaluation of ``self[key]``. For sequence\n   types, the accepted keys should be integers and slice objects.\n   Note that the special interpretation of negative indexes (if the\n   class wishes to emulate a sequence type) is up to the\n   ``__getitem__()`` method. If *key* is of an inappropriate type,\n   ``TypeError`` may be raised; if of a value outside the set of\n   indexes for the sequence (after any special interpretation of\n   negative values), ``IndexError`` should be raised. For mapping\n   types, if *key* is missing (not in the container), ``KeyError``\n   should be raised.\n\n   Note: ``for`` loops expect that an ``IndexError`` will be raised for\n     illegal indexes to allow proper detection of the end of the\n     sequence.\n\nobject.__setitem__(self, key, value)\n\n   Called to implement assignment to ``self[key]``.  Same note as for\n   ``__getitem__()``.  This should only be implemented for mappings if\n   the objects support changes to the values for keys, or if new keys\n   can be added, or for sequences if elements can be replaced.  The\n   same exceptions should be raised for improper *key* values as for\n   the ``__getitem__()`` method.\n\nobject.__delitem__(self, key)\n\n   Called to implement deletion of ``self[key]``.  Same note as for\n   ``__getitem__()``.  This should only be implemented for mappings if\n   the objects support removal of keys, or for sequences if elements\n   can be removed from the sequence.  The same exceptions should be\n   raised for improper *key* values as for the ``__getitem__()``\n   method.\n\nobject.__iter__(self)\n\n   This method is called when an iterator is required for a container.\n   This method should return a new iterator object that can iterate\n   over all the objects in the container.  For mappings, it should\n   iterate over the keys of the container, and should also be made\n   available as the method ``keys()``.\n\n   Iterator objects also need to implement this method; they are\n   required to return themselves.  For more information on iterator\n   objects, see *Iterator Types*.\n\nobject.__reversed__(self)\n\n   Called (if present) by the ``reversed()`` built-in to implement\n   reverse iteration.  It should return a new iterator object that\n   iterates over all the objects in the container in reverse order.\n\n   If the ``__reversed__()`` method is not provided, the\n   ``reversed()`` built-in will fall back to using the sequence\n   protocol (``__len__()`` and ``__getitem__()``).  Objects that\n   support the sequence protocol should only provide\n   ``__reversed__()`` if they can provide an implementation that is\n   more efficient than the one provided by ``reversed()``.\n\nThe membership test operators (``in`` and ``not in``) are normally\nimplemented as an iteration through a sequence.  However, container\nobjects can supply the following special method with a more efficient\nimplementation, which also does not require the object be a sequence.\n\nobject.__contains__(self, item)\n\n   Called to implement membership test operators.  Should return true\n   if *item* is in *self*, false otherwise.  For mapping objects, this\n   should consider the keys of the mapping rather than the values or\n   the key-item pairs.\n\n   For objects that don\'t define ``__contains__()``, the membership\n   test first tries iteration via ``__iter__()``, then the old\n   sequence iteration protocol via ``__getitem__()``, see *this\n   section in the language reference*.\n\n\nEmulating numeric types\n=======================\n\nThe following methods can be defined to emulate numeric objects.\nMethods corresponding to operations that are not supported by the\nparticular kind of number implemented (e.g., bitwise operations for\nnon-integral numbers) should be left undefined.\n\nobject.__add__(self, other)\nobject.__sub__(self, other)\nobject.__mul__(self, other)\nobject.__truediv__(self, other)\nobject.__floordiv__(self, other)\nobject.__mod__(self, other)\nobject.__divmod__(self, other)\nobject.__pow__(self, other[, modulo])\nobject.__lshift__(self, other)\nobject.__rshift__(self, other)\nobject.__and__(self, other)\nobject.__xor__(self, other)\nobject.__or__(self, other)\n\n   These methods are called to implement the binary arithmetic\n   operations (``+``, ``-``, ``*``, ``/``, ``//``, ``%``,\n   ``divmod()``, ``pow()``, ``**``, ``<<``, ``>>``, ``&``, ``^``,\n   ``|``).  For instance, to evaluate the expression ``x + y``, where\n   *x* is an instance of a class that has an ``__add__()`` method,\n   ``x.__add__(y)`` is called.  The ``__divmod__()`` method should be\n   the equivalent to using ``__floordiv__()`` and ``__mod__()``; it\n   should not be related to ``__truediv__()``.  Note that\n   ``__pow__()`` should be defined to accept an optional third\n   argument if the ternary version of the built-in ``pow()`` function\n   is to be supported.\n\n   If one of those methods does not support the operation with the\n   supplied arguments, it should return ``NotImplemented``.\n\nobject.__radd__(self, other)\nobject.__rsub__(self, other)\nobject.__rmul__(self, other)\nobject.__rtruediv__(self, other)\nobject.__rfloordiv__(self, other)\nobject.__rmod__(self, other)\nobject.__rdivmod__(self, other)\nobject.__rpow__(self, other)\nobject.__rlshift__(self, other)\nobject.__rrshift__(self, other)\nobject.__rand__(self, other)\nobject.__rxor__(self, other)\nobject.__ror__(self, other)\n\n   These methods are called to implement the binary arithmetic\n   operations (``+``, ``-``, ``*``, ``/``, ``//``, ``%``,\n   ``divmod()``, ``pow()``, ``**``, ``<<``, ``>>``, ``&``, ``^``,\n   ``|``) with reflected (swapped) operands. These functions are only\n   called if the left operand does not support the corresponding\n   operation and the operands are of different types. [2]  For\n   instance, to evaluate the expression ``x - y``, where *y* is an\n   instance of a class that has an ``__rsub__()`` method,\n   ``y.__rsub__(x)`` is called if ``x.__sub__(y)`` returns\n   *NotImplemented*.\n\n   Note that ternary ``pow()`` will not try calling ``__rpow__()``\n   (the coercion rules would become too complicated).\n\n   Note: If the right operand\'s type is a subclass of the left operand\'s\n     type and that subclass provides the reflected method for the\n     operation, this method will be called before the left operand\'s\n     non-reflected method.  This behavior allows subclasses to\n     override their ancestors\' operations.\n\nobject.__iadd__(self, other)\nobject.__isub__(self, other)\nobject.__imul__(self, other)\nobject.__itruediv__(self, other)\nobject.__ifloordiv__(self, other)\nobject.__imod__(self, other)\nobject.__ipow__(self, other[, modulo])\nobject.__ilshift__(self, other)\nobject.__irshift__(self, other)\nobject.__iand__(self, other)\nobject.__ixor__(self, other)\nobject.__ior__(self, other)\n\n   These methods are called to implement the augmented arithmetic\n   assignments (``+=``, ``-=``, ``*=``, ``/=``, ``//=``, ``%=``,\n   ``**=``, ``<<=``, ``>>=``, ``&=``, ``^=``, ``|=``).  These methods\n   should attempt to do the operation in-place (modifying *self*) and\n   return the result (which could be, but does not have to be,\n   *self*).  If a specific method is not defined, the augmented\n   assignment falls back to the normal methods.  For instance, to\n   execute the statement ``x += y``, where *x* is an instance of a\n   class that has an ``__iadd__()`` method, ``x.__iadd__(y)`` is\n   called.  If *x* is an instance of a class that does not define a\n   ``__iadd__()`` method, ``x.__add__(y)`` and ``y.__radd__(x)`` are\n   considered, as with the evaluation of ``x + y``.\n\nobject.__neg__(self)\nobject.__pos__(self)\nobject.__abs__(self)\nobject.__invert__(self)\n\n   Called to implement the unary arithmetic operations (``-``, ``+``,\n   ``abs()`` and ``~``).\n\nobject.__complex__(self)\nobject.__int__(self)\nobject.__float__(self)\nobject.__round__(self[, n])\n\n   Called to implement the built-in functions ``complex()``,\n   ``int()``, ``float()`` and ``round()``.  Should return a value of\n   the appropriate type.\n\nobject.__index__(self)\n\n   Called to implement ``operator.index()``.  Also called whenever\n   Python needs an integer object (such as in slicing, or in the\n   built-in ``bin()``, ``hex()`` and ``oct()`` functions). Must return\n   an integer.\n\n\nWith Statement Context Managers\n===============================\n\nA *context manager* is an object that defines the runtime context to\nbe established when executing a ``with`` statement. The context\nmanager handles the entry into, and the exit from, the desired runtime\ncontext for the execution of the block of code.  Context managers are\nnormally invoked using the ``with`` statement (described in section\n*The with statement*), but can also be used by directly invoking their\nmethods.\n\nTypical uses of context managers include saving and restoring various\nkinds of global state, locking and unlocking resources, closing opened\nfiles, etc.\n\nFor more information on context managers, see *Context Manager Types*.\n\nobject.__enter__(self)\n\n   Enter the runtime context related to this object. The ``with``\n   statement will bind this method\'s return value to the target(s)\n   specified in the ``as`` clause of the statement, if any.\n\nobject.__exit__(self, exc_type, exc_value, traceback)\n\n   Exit the runtime context related to this object. The parameters\n   describe the exception that caused the context to be exited. If the\n   context was exited without an exception, all three arguments will\n   be ``None``.\n\n   If an exception is supplied, and the method wishes to suppress the\n   exception (i.e., prevent it from being propagated), it should\n   return a true value. Otherwise, the exception will be processed\n   normally upon exit from this method.\n\n   Note that ``__exit__()`` methods should not reraise the passed-in\n   exception; this is the caller\'s responsibility.\n\nSee also:\n\n   **PEP 0343** - The "with" statement\n      The specification, background, and examples for the Python\n      ``with`` statement.\n\n\nSpecial method lookup\n=====================\n\nFor custom classes, implicit invocations of special methods are only\nguaranteed to work correctly if defined on an object\'s type, not in\nthe object\'s instance dictionary.  That behaviour is the reason why\nthe following code raises an exception:\n\n   >>> class C(object):\n   ...     pass\n   ...\n   >>> c = C()\n   >>> c.__len__ = lambda: 5\n   >>> len(c)\n   Traceback (most recent call last):\n     File "<stdin>", line 1, in <module>\n   TypeError: object of type \'C\' has no len()\n\nThe rationale behind this behaviour lies with a number of special\nmethods such as ``__hash__()`` and ``__repr__()`` that are implemented\nby all objects, including type objects. If the implicit lookup of\nthese methods used the conventional lookup process, they would fail\nwhen invoked on the type object itself:\n\n   >>> 1 .__hash__() == hash(1)\n   True\n   >>> int.__hash__() == hash(int)\n   Traceback (most recent call last):\n     File "<stdin>", line 1, in <module>\n   TypeError: descriptor \'__hash__\' of \'int\' object needs an argument\n\nIncorrectly attempting to invoke an unbound method of a class in this\nway is sometimes referred to as \'metaclass confusion\', and is avoided\nby bypassing the instance when looking up special methods:\n\n   >>> type(1).__hash__(1) == hash(1)\n   True\n   >>> type(int).__hash__(int) == hash(int)\n   True\n\nIn addition to bypassing any instance attributes in the interest of\ncorrectness, implicit special method lookup generally also bypasses\nthe ``__getattribute__()`` method even of the object\'s metaclass:\n\n   >>> class Meta(type):\n   ...    def __getattribute__(*args):\n   ...       print("Metaclass getattribute invoked")\n   ...       return type.__getattribute__(*args)\n   ...\n   >>> class C(object, metaclass=Meta):\n   ...     def __len__(self):\n   ...         return 10\n   ...     def __getattribute__(*args):\n   ...         print("Class getattribute invoked")\n   ...         return object.__getattribute__(*args)\n   ...\n   >>> c = C()\n   >>> c.__len__()                 # Explicit lookup via instance\n   Class getattribute invoked\n   10\n   >>> type(c).__len__(c)          # Explicit lookup via type\n   Metaclass getattribute invoked\n   10\n   >>> len(c)                      # Implicit lookup\n   10\n\nBypassing the ``__getattribute__()`` machinery in this fashion\nprovides significant scope for speed optimisations within the\ninterpreter, at the cost of some flexibility in the handling of\nspecial methods (the special method *must* be set on the class object\nitself in order to be consistently invoked by the interpreter).\n\n-[ Footnotes ]-\n\n[1] It *is* possible in some cases to change an object\'s type, under\n    certain controlled conditions. It generally isn\'t a good idea\n    though, since it can lead to some very strange behaviour if it is\n    handled incorrectly.\n\n[2] For operands of the same type, it is assumed that if the non-\n    reflected method (such as ``__add__()``) fails the operation is\n    not supported, which is why the reflected method is not called.\n',
+ 'string-methods': '\nString Methods\n**************\n\nString objects support the methods listed below.\n\nIn addition, Python\'s strings support the sequence type methods\ndescribed in the *Sequence Types --- str, bytes, bytearray, list,\ntuple, range* section. To output formatted strings, see the *String\nFormatting* section. Also, see the ``re`` module for string functions\nbased on regular expressions.\n\nstr.capitalize()\n\n   Return a copy of the string with its first character capitalized\n   and the rest lowercased.\n\nstr.center(width[, fillchar])\n\n   Return centered in a string of length *width*. Padding is done\n   using the specified *fillchar* (default is a space).\n\nstr.count(sub[, start[, end]])\n\n   Return the number of non-overlapping occurrences of substring *sub*\n   in the range [*start*, *end*].  Optional arguments *start* and\n   *end* are interpreted as in slice notation.\n\nstr.encode(encoding=sys.getdefaultencoding(), errors="strict")\n\n   Return an encoded version of the string as a bytes object.  Default\n   encoding is the current default string encoding.  *errors* may be\n   given to set a different error handling scheme.  The default for\n   *errors* is ``\'strict\'``, meaning that encoding errors raise a\n   ``UnicodeError``.  Other possible values are ``\'ignore\'``,\n   ``\'replace\'``, ``\'xmlcharrefreplace\'``, ``\'backslashreplace\'`` and\n   any other name registered via ``codecs.register_error()``, see\n   section *Codec Base Classes*. For a list of possible encodings, see\n   section *Standard Encodings*.\n\n   Changed in version 3.1: Added support for keyword arguments added.\n\nstr.endswith(suffix[, start[, end]])\n\n   Return ``True`` if the string ends with the specified *suffix*,\n   otherwise return ``False``.  *suffix* can also be a tuple of\n   suffixes to look for.  With optional *start*, test beginning at\n   that position.  With optional *end*, stop comparing at that\n   position.\n\nstr.expandtabs([tabsize])\n\n   Return a copy of the string where all tab characters are replaced\n   by one or more spaces, depending on the current column and the\n   given tab size.  The column number is reset to zero after each\n   newline occurring in the string. If *tabsize* is not given, a tab\n   size of ``8`` characters is assumed.  This doesn\'t understand other\n   non-printing characters or escape sequences.\n\nstr.find(sub[, start[, end]])\n\n   Return the lowest index in the string where substring *sub* is\n   found, such that *sub* is contained in the slice ``s[start:end]``.\n   Optional arguments *start* and *end* are interpreted as in slice\n   notation.  Return ``-1`` if *sub* is not found.\n\nstr.format(*args, **kwargs)\n\n   Perform a string formatting operation.  The string on which this\n   method is called can contain literal text or replacement fields\n   delimited by braces ``{}``.  Each replacement field contains either\n   the numeric index of a positional argument, or the name of a\n   keyword argument.  Returns a copy of the string where each\n   replacement field is replaced with the string value of the\n   corresponding argument.\n\n   >>> "The sum of 1 + 2 is {0}".format(1+2)\n   \'The sum of 1 + 2 is 3\'\n\n   See *Format String Syntax* for a description of the various\n   formatting options that can be specified in format strings.\n\nstr.index(sub[, start[, end]])\n\n   Like ``find()``, but raise ``ValueError`` when the substring is not\n   found.\n\nstr.isalnum()\n\n   Return true if all characters in the string are alphanumeric and\n   there is at least one character, false otherwise.\n\nstr.isalpha()\n\n   Return true if all characters in the string are alphabetic and\n   there is at least one character, false otherwise.\n\nstr.isdecimal()\n\n   Return true if all characters in the string are decimal characters\n   and there is at least one character, false otherwise. Decimal\n   characters include digit characters, and all characters that that\n   can be used to form decimal-radix numbers, e.g. U+0660, ARABIC-\n   INDIC DIGIT ZERO.\n\nstr.isdigit()\n\n   Return true if all characters in the string are digits and there is\n   at least one character, false otherwise.\n\nstr.isidentifier()\n\n   Return true if the string is a valid identifier according to the\n   language definition, section *Identifiers and keywords*.\n\nstr.islower()\n\n   Return true if all cased characters in the string are lowercase and\n   there is at least one cased character, false otherwise.\n\nstr.isnumeric()\n\n   Return true if all characters in the string are numeric characters,\n   and there is at least one character, false otherwise. Numeric\n   characters include digit characters, and all characters that have\n   the Unicode numeric value property, e.g. U+2155, VULGAR FRACTION\n   ONE FIFTH.\n\nstr.isprintable()\n\n   Return true if all characters in the string are printable or the\n   string is empty, false otherwise.  Nonprintable characters are\n   those characters defined in the Unicode character database as\n   "Other" or "Separator", excepting the ASCII space (0x20) which is\n   considered printable.  (Note that printable characters in this\n   context are those which should not be escaped when ``repr()`` is\n   invoked on a string.  It has no bearing on the handling of strings\n   written to ``sys.stdout`` or ``sys.stderr``.)\n\nstr.isspace()\n\n   Return true if there are only whitespace characters in the string\n   and there is at least one character, false otherwise.\n\nstr.istitle()\n\n   Return true if the string is a titlecased string and there is at\n   least one character, for example uppercase characters may only\n   follow uncased characters and lowercase characters only cased ones.\n   Return false otherwise.\n\nstr.isupper()\n\n   Return true if all cased characters in the string are uppercase and\n   there is at least one cased character, false otherwise.\n\nstr.join(iterable)\n\n   Return a string which is the concatenation of the strings in the\n   *iterable* *iterable*.  A ``TypeError`` will be raised if there are\n   any non-string values in *seq*, including ``bytes`` objects.  The\n   separator between elements is the string providing this method.\n\nstr.ljust(width[, fillchar])\n\n   Return the string left justified in a string of length *width*.\n   Padding is done using the specified *fillchar* (default is a\n   space).  The original string is returned if *width* is less than\n   ``len(s)``.\n\nstr.lower()\n\n   Return a copy of the string converted to lowercase.\n\nstr.lstrip([chars])\n\n   Return a copy of the string with leading characters removed.  The\n   *chars* argument is a string specifying the set of characters to be\n   removed.  If omitted or ``None``, the *chars* argument defaults to\n   removing whitespace.  The *chars* argument is not a prefix; rather,\n   all combinations of its values are stripped:\n\n   >>> \'   spacious   \'.lstrip()\n   \'spacious   \'\n   >>> \'www.example.com\'.lstrip(\'cmowz.\')\n   \'example.com\'\n\nstatic str.maketrans(x[, y[, z]])\n\n   This static method returns a translation table usable for\n   ``str.translate()``.\n\n   If there is only one argument, it must be a dictionary mapping\n   Unicode ordinals (integers) or characters (strings of length 1) to\n   Unicode ordinals, strings (of arbitrary lengths) or None.\n   Character keys will then be converted to ordinals.\n\n   If there are two arguments, they must be strings of equal length,\n   and in the resulting dictionary, each character in x will be mapped\n   to the character at the same position in y.  If there is a third\n   argument, it must be a string, whose characters will be mapped to\n   None in the result.\n\nstr.partition(sep)\n\n   Split the string at the first occurrence of *sep*, and return a\n   3-tuple containing the part before the separator, the separator\n   itself, and the part after the separator.  If the separator is not\n   found, return a 3-tuple containing the string itself, followed by\n   two empty strings.\n\nstr.replace(old, new[, count])\n\n   Return a copy of the string with all occurrences of substring *old*\n   replaced by *new*.  If the optional argument *count* is given, only\n   the first *count* occurrences are replaced.\n\nstr.rfind(sub[, start[, end]])\n\n   Return the highest index in the string where substring *sub* is\n   found, such that *sub* is contained within ``s[start:end]``.\n   Optional arguments *start* and *end* are interpreted as in slice\n   notation.  Return ``-1`` on failure.\n\nstr.rindex(sub[, start[, end]])\n\n   Like ``rfind()`` but raises ``ValueError`` when the substring *sub*\n   is not found.\n\nstr.rjust(width[, fillchar])\n\n   Return the string right justified in a string of length *width*.\n   Padding is done using the specified *fillchar* (default is a\n   space). The original string is returned if *width* is less than\n   ``len(s)``.\n\nstr.rpartition(sep)\n\n   Split the string at the last occurrence of *sep*, and return a\n   3-tuple containing the part before the separator, the separator\n   itself, and the part after the separator.  If the separator is not\n   found, return a 3-tuple containing two empty strings, followed by\n   the string itself.\n\nstr.rsplit([sep[, maxsplit]])\n\n   Return a list of the words in the string, using *sep* as the\n   delimiter string. If *maxsplit* is given, at most *maxsplit* splits\n   are done, the *rightmost* ones.  If *sep* is not specified or\n   ``None``, any whitespace string is a separator.  Except for\n   splitting from the right, ``rsplit()`` behaves like ``split()``\n   which is described in detail below.\n\nstr.rstrip([chars])\n\n   Return a copy of the string with trailing characters removed.  The\n   *chars* argument is a string specifying the set of characters to be\n   removed.  If omitted or ``None``, the *chars* argument defaults to\n   removing whitespace.  The *chars* argument is not a suffix; rather,\n   all combinations of its values are stripped:\n\n   >>> \'   spacious   \'.rstrip()\n   \'   spacious\'\n   >>> \'mississippi\'.rstrip(\'ipz\')\n   \'mississ\'\n\nstr.split([sep[, maxsplit]])\n\n   Return a list of the words in the string, using *sep* as the\n   delimiter string.  If *maxsplit* is given, at most *maxsplit*\n   splits are done (thus, the list will have at most ``maxsplit+1``\n   elements).  If *maxsplit* is not specified, then there is no limit\n   on the number of splits (all possible splits are made).\n\n   If *sep* is given, consecutive delimiters are not grouped together\n   and are deemed to delimit empty strings (for example,\n   ``\'1,,2\'.split(\',\')`` returns ``[\'1\', \'\', \'2\']``).  The *sep*\n   argument may consist of multiple characters (for example,\n   ``\'1<>2<>3\'.split(\'<>\')`` returns ``[\'1\', \'2\', \'3\']``). Splitting\n   an empty string with a specified separator returns ``[\'\']``.\n\n   If *sep* is not specified or is ``None``, a different splitting\n   algorithm is applied: runs of consecutive whitespace are regarded\n   as a single separator, and the result will contain no empty strings\n   at the start or end if the string has leading or trailing\n   whitespace.  Consequently, splitting an empty string or a string\n   consisting of just whitespace with a ``None`` separator returns\n   ``[]``.\n\n   For example, ``\' 1  2   3  \'.split()`` returns ``[\'1\', \'2\', \'3\']``,\n   and ``\'  1  2   3  \'.split(None, 1)`` returns ``[\'1\', \'2   3  \']``.\n\nstr.splitlines([keepends])\n\n   Return a list of the lines in the string, breaking at line\n   boundaries.  Line breaks are not included in the resulting list\n   unless *keepends* is given and true.\n\nstr.startswith(prefix[, start[, end]])\n\n   Return ``True`` if string starts with the *prefix*, otherwise\n   return ``False``. *prefix* can also be a tuple of prefixes to look\n   for.  With optional *start*, test string beginning at that\n   position.  With optional *end*, stop comparing string at that\n   position.\n\nstr.strip([chars])\n\n   Return a copy of the string with the leading and trailing\n   characters removed. The *chars* argument is a string specifying the\n   set of characters to be removed. If omitted or ``None``, the\n   *chars* argument defaults to removing whitespace. The *chars*\n   argument is not a prefix or suffix; rather, all combinations of its\n   values are stripped:\n\n   >>> \'   spacious   \'.strip()\n   \'spacious\'\n   >>> \'www.example.com\'.strip(\'cmowz.\')\n   \'example\'\n\nstr.swapcase()\n\n   Return a copy of the string with uppercase characters converted to\n   lowercase and vice versa.\n\nstr.title()\n\n   Return a titlecased version of the string where words start with an\n   uppercase character and the remaining characters are lowercase.\n\n   The algorithm uses a simple language-independent definition of a\n   word as groups of consecutive letters.  The definition works in\n   many contexts but it means that apostrophes in contractions and\n   possessives form word boundaries, which may not be the desired\n   result:\n\n      >>> "they\'re bill\'s friends from the UK".title()\n      "They\'Re Bill\'S Friends From The Uk"\n\n   A workaround for apostrophes can be constructed using regular\n   expressions:\n\n      >>> import re\n      >>> def titlecase(s):\n              return re.sub(r"[A-Za-z]+(\'[A-Za-z]+)?",\n                            lambda mo: mo.group(0)[0].upper() +\n                                       mo.group(0)[1:].lower(),\n                            s)\n\n      >>> titlecase("they\'re bill\'s friends.")\n      "They\'re Bill\'s Friends."\n\nstr.translate(map)\n\n   Return a copy of the *s* where all characters have been mapped\n   through the *map* which must be a dictionary of Unicode ordinals\n   (integers) to Unicode ordinals, strings or ``None``.  Unmapped\n   characters are left untouched. Characters mapped to ``None`` are\n   deleted.\n\n   You can use ``str.maketrans()`` to create a translation map from\n   character-to-character mappings in different formats.\n\n   You can use the ``maketrans()`` helper function in the ``string``\n   module to create a translation table. For string objects, set the\n   *table* argument to ``None`` for translations that only delete\n   characters:\n\n   Note: An even more flexible approach is to create a custom character\n     mapping codec using the ``codecs`` module (see\n     ``encodings.cp1251`` for an example).\n\nstr.upper()\n\n   Return a copy of the string converted to uppercase.\n\nstr.zfill(width)\n\n   Return the numeric string left filled with zeros in a string of\n   length *width*.  A sign prefix is handled correctly.  The original\n   string is returned if *width* is less than ``len(s)``.\n',
+ 'strings': '\nString and Bytes literals\n*************************\n\nString literals are described by the following lexical definitions:\n\n   stringliteral   ::= [stringprefix](shortstring | longstring)\n   stringprefix    ::= "r" | "R"\n   shortstring     ::= "\'" shortstringitem* "\'" | \'"\' shortstringitem* \'"\'\n   longstring      ::= "\'\'\'" longstringitem* "\'\'\'" | \'"""\' longstringitem* \'"""\'\n   shortstringitem ::= shortstringchar | stringescapeseq\n   longstringitem  ::= longstringchar | stringescapeseq\n   shortstringchar ::= <any source character except "\\" or newline or the quote>\n   longstringchar  ::= <any source character except "\\">\n   stringescapeseq ::= "\\" <any source character>\n\n   bytesliteral   ::= bytesprefix(shortbytes | longbytes)\n   bytesprefix    ::= "b" | "B" | "br" | "Br" | "bR" | "BR"\n   shortbytes     ::= "\'" shortbytesitem* "\'" | \'"\' shortbytesitem* \'"\'\n   longbytes      ::= "\'\'\'" longbytesitem* "\'\'\'" | \'"""\' longbytesitem* \'"""\'\n   shortbytesitem ::= shortbyteschar | bytesescapeseq\n   longbytesitem  ::= longbyteschar | bytesescapeseq\n   shortbyteschar ::= <any ASCII character except "\\" or newline or the quote>\n   longbyteschar  ::= <any ASCII character except "\\">\n   bytesescapeseq ::= "\\" <any ASCII character>\n\nOne syntactic restriction not indicated by these productions is that\nwhitespace is not allowed between the ``stringprefix`` or\n``bytesprefix`` and the rest of the literal. The source character set\nis defined by the encoding declaration; it is UTF-8 if no encoding\ndeclaration is given in the source file; see section *Encoding\ndeclarations*.\n\nIn plain English: Both types of literals can be enclosed in matching\nsingle quotes (``\'``) or double quotes (``"``).  They can also be\nenclosed in matching groups of three single or double quotes (these\nare generally referred to as *triple-quoted strings*).  The backslash\n(``\\``) character is used to escape characters that otherwise have a\nspecial meaning, such as newline, backslash itself, or the quote\ncharacter.\n\nBytes literals are always prefixed with ``\'b\'`` or ``\'B\'``; they\nproduce an instance of the ``bytes`` type instead of the ``str`` type.\nThey may only contain ASCII characters; bytes with a numeric value of\n128 or greater must be expressed with escapes.\n\nBoth string and bytes literals may optionally be prefixed with a\nletter ``\'r\'`` or ``\'R\'``; such strings are called *raw strings* and\ntreat backslashes as literal characters.  As a result, in string\nliterals, ``\'\\U\'`` and ``\'\\u\'`` escapes in raw strings are not treated\nspecially.\n\nIn triple-quoted strings, unescaped newlines and quotes are allowed\n(and are retained), except that three unescaped quotes in a row\nterminate the string.  (A "quote" is the character used to open the\nstring, i.e. either ``\'`` or ``"``.)\n\nUnless an ``\'r\'`` or ``\'R\'`` prefix is present, escape sequences in\nstrings are interpreted according to rules similar to those used by\nStandard C.  The recognized escape sequences are:\n\n+-------------------+-----------------------------------+---------+\n| Escape Sequence   | Meaning                           | Notes   |\n+===================+===================================+=========+\n| ``\\newline``      | Backslash and newline ignored     |         |\n+-------------------+-----------------------------------+---------+\n| ``\\\\``            | Backslash (``\\``)                 |         |\n+-------------------+-----------------------------------+---------+\n| ``\\\'``            | Single quote (``\'``)              |         |\n+-------------------+-----------------------------------+---------+\n| ``\\"``            | Double quote (``"``)              |         |\n+-------------------+-----------------------------------+---------+\n| ``\\a``            | ASCII Bell (BEL)                  |         |\n+-------------------+-----------------------------------+---------+\n| ``\\b``            | ASCII Backspace (BS)              |         |\n+-------------------+-----------------------------------+---------+\n| ``\\f``            | ASCII Formfeed (FF)               |         |\n+-------------------+-----------------------------------+---------+\n| ``\\n``            | ASCII Linefeed (LF)               |         |\n+-------------------+-----------------------------------+---------+\n| ``\\r``            | ASCII Carriage Return (CR)        |         |\n+-------------------+-----------------------------------+---------+\n| ``\\t``            | ASCII Horizontal Tab (TAB)        |         |\n+-------------------+-----------------------------------+---------+\n| ``\\v``            | ASCII Vertical Tab (VT)           |         |\n+-------------------+-----------------------------------+---------+\n| ``\\ooo``          | Character with octal value *ooo*  | (1,3)   |\n+-------------------+-----------------------------------+---------+\n| ``\\xhh``          | Character with hex value *hh*     | (2,3)   |\n+-------------------+-----------------------------------+---------+\n\nEscape sequences only recognized in string literals are:\n\n+-------------------+-----------------------------------+---------+\n| Escape Sequence   | Meaning                           | Notes   |\n+===================+===================================+=========+\n| ``\\N{name}``      | Character named *name* in the     |         |\n|                   | Unicode database                  |         |\n+-------------------+-----------------------------------+---------+\n| ``\\uxxxx``        | Character with 16-bit hex value   | (4)     |\n|                   | *xxxx*                            |         |\n+-------------------+-----------------------------------+---------+\n| ``\\Uxxxxxxxx``    | Character with 32-bit hex value   | (5)     |\n|                   | *xxxxxxxx*                        |         |\n+-------------------+-----------------------------------+---------+\n\nNotes:\n\n1. As in Standard C, up to three octal digits are accepted.\n\n2. Unlike in Standard C, exactly two hex digits are required.\n\n3. In a bytes literal, hexadecimal and octal escapes denote the byte\n   with the given value. In a string literal, these escapes denote a\n   Unicode character with the given value.\n\n4. Individual code units which form parts of a surrogate pair can be\n   encoded using this escape sequence.  Exactly four hex digits are\n   required.\n\n5. Any Unicode character can be encoded this way, but characters\n   outside the Basic Multilingual Plane (BMP) will be encoded using a\n   surrogate pair if Python is compiled to use 16-bit code units (the\n   default).  Exactly eight hex digits are required.\n\nUnlike Standard C, all unrecognized escape sequences are left in the\nstring unchanged, i.e., *the backslash is left in the string*.  (This\nbehavior is useful when debugging: if an escape sequence is mistyped,\nthe resulting output is more easily recognized as broken.)  It is also\nimportant to note that the escape sequences only recognized in string\nliterals fall into the category of unrecognized escapes for bytes\nliterals.\n\nEven in a raw string, string quotes can be escaped with a backslash,\nbut the backslash remains in the string; for example, ``r"\\""`` is a\nvalid string literal consisting of two characters: a backslash and a\ndouble quote; ``r"\\"`` is not a valid string literal (even a raw\nstring cannot end in an odd number of backslashes).  Specifically, *a\nraw string cannot end in a single backslash* (since the backslash\nwould escape the following quote character).  Note also that a single\nbackslash followed by a newline is interpreted as those two characters\nas part of the string, *not* as a line continuation.\n',
  'subscriptions': '\nSubscriptions\n*************\n\nA subscription selects an item of a sequence (string, tuple or list)\nor mapping (dictionary) object:\n\n   subscription ::= primary "[" expression_list "]"\n\nThe primary must evaluate to an object that supports subscription,\ne.g. a list or dictionary.  User-defined objects can support\nsubscription by defining a ``__getitem__()`` method.\n\nFor built-in objects, there are two types of objects that support\nsubscription:\n\nIf the primary is a mapping, the expression list must evaluate to an\nobject whose value is one of the keys of the mapping, and the\nsubscription selects the value in the mapping that corresponds to that\nkey.  (The expression list is a tuple except if it has exactly one\nitem.)\n\nIf the primary is a sequence, the expression (list) must evaluate to\nan integer. If this value is negative, the length of the sequence is\nadded to it (so that, e.g., ``x[-1]`` selects the last item of ``x``.)\nThe resulting value must be a nonnegative integer less than the number\nof items in the sequence, and the subscription selects the item whose\nindex is that value (counting from zero).\n\nA string\'s items are characters.  A character is not a separate data\ntype but a string of exactly one character.\n',
  'truth': "\nTruth Value Testing\n*******************\n\nAny object can be tested for truth value, for use in an ``if`` or\n``while`` condition or as operand of the Boolean operations below. The\nfollowing values are considered false:\n\n* ``None``\n\n* ``False``\n\n* zero of any numeric type, for example, ``0``, ``0.0``, ``0j``.\n\n* any empty sequence, for example, ``''``, ``()``, ``[]``.\n\n* any empty mapping, for example, ``{}``.\n\n* instances of user-defined classes, if the class defines a\n  ``__bool__()`` or ``__len__()`` method, when that method returns the\n  integer zero or ``bool`` value ``False``. [1]\n\nAll other values are considered true --- so objects of many types are\nalways true.\n\nOperations and built-in functions that have a Boolean result always\nreturn ``0`` or ``False`` for false and ``1`` or ``True`` for true,\nunless otherwise stated. (Important exception: the Boolean operations\n``or`` and ``and`` always return one of their operands.)\n",
- 'try': '\nThe ``try`` statement\n*********************\n\nThe ``try`` statement specifies exception handlers and/or cleanup code\nfor a group of statements:\n\n   try_stmt  ::= try1_stmt | try2_stmt\n   try1_stmt ::= "try" ":" suite\n                 ("except" [expression ["as" target]] ":" suite)+\n                 ["else" ":" suite]\n                 ["finally" ":" suite]\n   try2_stmt ::= "try" ":" suite\n                 "finally" ":" suite\n\nThe ``except`` clause(s) specify one or more exception handlers. When\nno exception occurs in the ``try`` clause, no exception handler is\nexecuted. When an exception occurs in the ``try`` suite, a search for\nan exception handler is started.  This search inspects the except\nclauses in turn until one is found that matches the exception.  An\nexpression-less except clause, if present, must be last; it matches\nany exception.  For an except clause with an expression, that\nexpression is evaluated, and the clause matches the exception if the\nresulting object is "compatible" with the exception.  An object is\ncompatible with an exception if it is the class or a base class of the\nexception object or a tuple containing an item compatible with the\nexception.\n\nIf no except clause matches the exception, the search for an exception\nhandler continues in the surrounding code and on the invocation stack.\n[1]\n\nIf the evaluation of an expression in the header of an except clause\nraises an exception, the original search for a handler is canceled and\na search starts for the new exception in the surrounding code and on\nthe call stack (it is treated as if the entire ``try`` statement\nraised the exception).\n\nWhen a matching except clause is found, the exception is assigned to\nthe target specified after the ``as`` keyword in that except clause,\nif present, and the except clause\'s suite is executed.  All except\nclauses must have an executable block.  When the end of this block is\nreached, execution continues normally after the entire try statement.\n(This means that if two nested handlers exist for the same exception,\nand the exception occurs in the try clause of the inner handler, the\nouter handler will not handle the exception.)\n\nWhen an exception has been assigned using ``as target``, it is cleared\nat the end of the except clause.  This is as if\n\n   except E as N:\n       foo\n\nwas translated to\n\n   except E as N:\n       try:\n           foo\n       finally:\n           N = None\n           del N\n\nThat means that you have to assign the exception to a different name\nif you want to be able to refer to it after the except clause.  The\nreason for this is that with the traceback attached to them,\nexceptions will form a reference cycle with the stack frame, keeping\nall locals in that frame alive until the next garbage collection\noccurs.\n\nBefore an except clause\'s suite is executed, details about the\nexception are stored in the ``sys`` module and can be access via\n``sys.exc_info()``. ``sys.exc_info()`` returns a 3-tuple consisting\nof: ``exc_type``, the exception class; ``exc_value``, the exception\ninstance; ``exc_traceback``, a traceback object (see section *The\nstandard type hierarchy*) identifying the point in the program where\nthe exception occurred. ``sys.exc_info()`` values are restored to\ntheir previous values (before the call) when returning from a function\nthat handled an exception.\n\nThe optional ``else`` clause is executed if and when control flows off\nthe end of the ``try`` clause. [2] Exceptions in the ``else`` clause\nare not handled by the preceding ``except`` clauses.\n\nIf ``finally`` is present, it specifies a \'cleanup\' handler.  The\n``try`` clause is executed, including any ``except`` and ``else``\nclauses.  If an exception occurs in any of the clauses and is not\nhandled, the exception is temporarily saved. The ``finally`` clause is\nexecuted.  If there is a saved exception, it is re-raised at the end\nof the ``finally`` clause. If the ``finally`` clause raises another\nexception or executes a ``return`` or ``break`` statement, the saved\nexception is lost.  The exception information is not available to the\nprogram during execution of the ``finally`` clause.\n\nWhen a ``return``, ``break`` or ``continue`` statement is executed in\nthe ``try`` suite of a ``try``...``finally`` statement, the\n``finally`` clause is also executed \'on the way out.\' A ``continue``\nstatement is illegal in the ``finally`` clause. (The reason is a\nproblem with the current implementation --- this restriction may be\nlifted in the future).\n\nAdditional information on exceptions can be found in section\n*Exceptions*, and information on using the ``raise`` statement to\ngenerate exceptions may be found in section *The raise statement*.\n',
- 'types': '\nThe standard type hierarchy\n***************************\n\nBelow is a list of the types that are built into Python.  Extension\nmodules (written in C, Java, or other languages, depending on the\nimplementation) can define additional types.  Future versions of\nPython may add types to the type hierarchy (e.g., rational numbers,\nefficiently stored arrays of integers, etc.), although such additions\nwill often be provided via the standard library instead.\n\nSome of the type descriptions below contain a paragraph listing\n\'special attributes.\'  These are attributes that provide access to the\nimplementation and are not intended for general use.  Their definition\nmay change in the future.\n\nNone\n   This type has a single value.  There is a single object with this\n   value. This object is accessed through the built-in name ``None``.\n   It is used to signify the absence of a value in many situations,\n   e.g., it is returned from functions that don\'t explicitly return\n   anything. Its truth value is false.\n\nNotImplemented\n   This type has a single value.  There is a single object with this\n   value. This object is accessed through the built-in name\n   ``NotImplemented``. Numeric methods and rich comparison methods may\n   return this value if they do not implement the operation for the\n   operands provided.  (The interpreter will then try the reflected\n   operation, or some other fallback, depending on the operator.)  Its\n   truth value is true.\n\nEllipsis\n   This type has a single value.  There is a single object with this\n   value. This object is accessed through the literal ``...`` or the\n   built-in name ``Ellipsis``.  Its truth value is true.\n\n``numbers.Number``\n   These are created by numeric literals and returned as results by\n   arithmetic operators and arithmetic built-in functions.  Numeric\n   objects are immutable; once created their value never changes.\n   Python numbers are of course strongly related to mathematical\n   numbers, but subject to the limitations of numerical representation\n   in computers.\n\n   Python distinguishes between integers, floating point numbers, and\n   complex numbers:\n\n   ``numbers.Integral``\n      These represent elements from the mathematical set of integers\n      (positive and negative).\n\n      There are two types of integers:\n\n      Integers (``int``)\n\n         These represent numbers in an unlimited range, subject to\n         available (virtual) memory only.  For the purpose of shift\n         and mask operations, a binary representation is assumed, and\n         negative numbers are represented in a variant of 2\'s\n         complement which gives the illusion of an infinite string of\n         sign bits extending to the left.\n\n      Booleans (``bool``)\n         These represent the truth values False and True.  The two\n         objects representing the values False and True are the only\n         Boolean objects. The Boolean type is a subtype of the integer\n         type, and Boolean values behave like the values 0 and 1,\n         respectively, in almost all contexts, the exception being\n         that when converted to a string, the strings ``"False"`` or\n         ``"True"`` are returned, respectively.\n\n      The rules for integer representation are intended to give the\n      most meaningful interpretation of shift and mask operations\n      involving negative integers.\n\n   ``numbers.Real`` (``float``)\n      These represent machine-level double precision floating point\n      numbers. You are at the mercy of the underlying machine\n      architecture (and C or Java implementation) for the accepted\n      range and handling of overflow. Python does not support single-\n      precision floating point numbers; the savings in processor and\n      memory usage that are usually the reason for using these is\n      dwarfed by the overhead of using objects in Python, so there is\n      no reason to complicate the language with two kinds of floating\n      point numbers.\n\n   ``numbers.Complex`` (``complex``)\n      These represent complex numbers as a pair of machine-level\n      double precision floating point numbers.  The same caveats apply\n      as for floating point numbers. The real and imaginary parts of a\n      complex number ``z`` can be retrieved through the read-only\n      attributes ``z.real`` and ``z.imag``.\n\nSequences\n   These represent finite ordered sets indexed by non-negative\n   numbers. The built-in function ``len()`` returns the number of\n   items of a sequence. When the length of a sequence is *n*, the\n   index set contains the numbers 0, 1, ..., *n*-1.  Item *i* of\n   sequence *a* is selected by ``a[i]``.\n\n   Sequences also support slicing: ``a[i:j]`` selects all items with\n   index *k* such that *i* ``<=`` *k* ``<`` *j*.  When used as an\n   expression, a slice is a sequence of the same type.  This implies\n   that the index set is renumbered so that it starts at 0.\n\n   Some sequences also support "extended slicing" with a third "step"\n   parameter: ``a[i:j:k]`` selects all items of *a* with index *x*\n   where ``x = i + n*k``, *n* ``>=`` ``0`` and *i* ``<=`` *x* ``<``\n   *j*.\n\n   Sequences are distinguished according to their mutability:\n\n   Immutable sequences\n      An object of an immutable sequence type cannot change once it is\n      created.  (If the object contains references to other objects,\n      these other objects may be mutable and may be changed; however,\n      the collection of objects directly referenced by an immutable\n      object cannot change.)\n\n      The following types are immutable sequences:\n\n      Strings\n         The items of a string object are Unicode code units.  A\n         Unicode code unit is represented by a string object of one\n         item and can hold either a 16-bit or 32-bit value\n         representing a Unicode ordinal (the maximum value for the\n         ordinal is given in ``sys.maxunicode``, and depends on how\n         Python is configured at compile time).  Surrogate pairs may\n         be present in the Unicode object, and will be reported as two\n         separate items.  The built-in functions ``chr()`` and\n         ``ord()`` convert between code units and nonnegative integers\n         representing the Unicode ordinals as defined in the Unicode\n         Standard 3.0. Conversion from and to other encodings are\n         possible through the string method ``encode()``.\n\n      Tuples\n         The items of a tuple are arbitrary Python objects. Tuples of\n         two or more items are formed by comma-separated lists of\n         expressions.  A tuple of one item (a \'singleton\') can be\n         formed by affixing a comma to an expression (an expression by\n         itself does not create a tuple, since parentheses must be\n         usable for grouping of expressions).  An empty tuple can be\n         formed by an empty pair of parentheses.\n\n      Bytes\n         A bytes object is an immutable array.  The items are 8-bit\n         bytes, represented by integers in the range 0 <= x < 256.\n         Bytes literals (like ``b\'abc\'`` and the built-in function\n         ``bytes()`` can be used to construct bytes objects.  Also,\n         bytes objects can be decoded to strings via the ``decode()``\n         method.\n\n   Mutable sequences\n      Mutable sequences can be changed after they are created.  The\n      subscription and slicing notations can be used as the target of\n      assignment and ``del`` (delete) statements.\n\n      There are currently two intrinsic mutable sequence types:\n\n      Lists\n         The items of a list are arbitrary Python objects.  Lists are\n         formed by placing a comma-separated list of expressions in\n         square brackets. (Note that there are no special cases needed\n         to form lists of length 0 or 1.)\n\n      Byte Arrays\n         A bytearray object is a mutable array. They are created by\n         the built-in ``bytearray()`` constructor.  Aside from being\n         mutable (and hence unhashable), byte arrays otherwise provide\n         the same interface and functionality as immutable bytes\n         objects.\n\n      The extension module ``array`` provides an additional example of\n      a mutable sequence type, as does the ``collections`` module.\n\nSet types\n   These represent unordered, finite sets of unique, immutable\n   objects. As such, they cannot be indexed by any subscript. However,\n   they can be iterated over, and the built-in function ``len()``\n   returns the number of items in a set. Common uses for sets are fast\n   membership testing, removing duplicates from a sequence, and\n   computing mathematical operations such as intersection, union,\n   difference, and symmetric difference.\n\n   For set elements, the same immutability rules apply as for\n   dictionary keys. Note that numeric types obey the normal rules for\n   numeric comparison: if two numbers compare equal (e.g., ``1`` and\n   ``1.0``), only one of them can be contained in a set.\n\n   There are currently two intrinsic set types:\n\n   Sets\n      These represent a mutable set. They are created by the built-in\n      ``set()`` constructor and can be modified afterwards by several\n      methods, such as ``add()``.\n\n   Frozen sets\n      These represent an immutable set.  They are created by the\n      built-in ``frozenset()`` constructor.  As a frozenset is\n      immutable and *hashable*, it can be used again as an element of\n      another set, or as a dictionary key.\n\nMappings\n   These represent finite sets of objects indexed by arbitrary index\n   sets. The subscript notation ``a[k]`` selects the item indexed by\n   ``k`` from the mapping ``a``; this can be used in expressions and\n   as the target of assignments or ``del`` statements. The built-in\n   function ``len()`` returns the number of items in a mapping.\n\n   There is currently a single intrinsic mapping type:\n\n   Dictionaries\n      These represent finite sets of objects indexed by nearly\n      arbitrary values.  The only types of values not acceptable as\n      keys are values containing lists or dictionaries or other\n      mutable types that are compared by value rather than by object\n      identity, the reason being that the efficient implementation of\n      dictionaries requires a key\'s hash value to remain constant.\n      Numeric types used for keys obey the normal rules for numeric\n      comparison: if two numbers compare equal (e.g., ``1`` and\n      ``1.0``) then they can be used interchangeably to index the same\n      dictionary entry.\n\n      Dictionaries are mutable; they can be created by the ``{...}``\n      notation (see section *Dictionary displays*).\n\n      The extension modules ``dbm.ndbm`` and ``dbm.gnu`` provide\n      additional examples of mapping types, as does the\n      ``collections`` module.\n\nCallable types\n   These are the types to which the function call operation (see\n   section *Calls*) can be applied:\n\n   User-defined functions\n      A user-defined function object is created by a function\n      definition (see section *Function definitions*).  It should be\n      called with an argument list containing the same number of items\n      as the function\'s formal parameter list.\n\n      Special attributes:\n\n      +---------------------------+---------------------------------+-------------+\n      | Attribute                 | Meaning                         |             |\n      +===========================+=================================+=============+\n      | ``__doc__``               | The function\'s documentation    | Writable    |\n      |                           | string, or ``None`` if          |             |\n      |                           | unavailable                     |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__name__``              | The function\'s name             | Writable    |\n      +---------------------------+---------------------------------+-------------+\n      | ``__module__``            | The name of the module the      | Writable    |\n      |                           | function was defined in, or     |             |\n      |                           | ``None`` if unavailable.        |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__defaults__``          | A tuple containing default      | Writable    |\n      |                           | argument values for those       |             |\n      |                           | arguments that have defaults,   |             |\n      |                           | or ``None`` if no arguments     |             |\n      |                           | have a default value            |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__code__``              | The code object representing    | Writable    |\n      |                           | the compiled function body.     |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__globals__``           | A reference to the dictionary   | Read-only   |\n      |                           | that holds the function\'s       |             |\n      |                           | global variables --- the global |             |\n      |                           | namespace of the module in      |             |\n      |                           | which the function was defined. |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__dict__``              | The namespace supporting        | Writable    |\n      |                           | arbitrary function attributes.  |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__closure__``           | ``None`` or a tuple of cells    | Read-only   |\n      |                           | that contain bindings for the   |             |\n      |                           | function\'s free variables.      |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__annotations__``       | A dict containing annotations   | Writable    |\n      |                           | of parameters.  The keys of the |             |\n      |                           | dict are the parameter names,   |             |\n      |                           | or ``\'return\'`` for the return  |             |\n      |                           | annotation, if provided.        |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__kwdefaults__``        | A dict containing defaults for  | Writable    |\n      |                           | keyword-only parameters.        |             |\n      +---------------------------+---------------------------------+-------------+\n\n      Most of the attributes labelled "Writable" check the type of the\n      assigned value.\n\n      Function objects also support getting and setting arbitrary\n      attributes, which can be used, for example, to attach metadata\n      to functions.  Regular attribute dot-notation is used to get and\n      set such attributes. *Note that the current implementation only\n      supports function attributes on user-defined functions. Function\n      attributes on built-in functions may be supported in the\n      future.*\n\n      Additional information about a function\'s definition can be\n      retrieved from its code object; see the description of internal\n      types below.\n\n   Instance methods\n      An instance method object combines a class, a class instance and\n      any callable object (normally a user-defined function).\n\n      Special read-only attributes: ``__self__`` is the class instance\n      object, ``__func__`` is the function object; ``__doc__`` is the\n      method\'s documentation (same as ``__func__.__doc__``);\n      ``__name__`` is the method name (same as ``__func__.__name__``);\n      ``__module__`` is the name of the module the method was defined\n      in, or ``None`` if unavailable.\n\n      Methods also support accessing (but not setting) the arbitrary\n      function attributes on the underlying function object.\n\n      User-defined method objects may be created when getting an\n      attribute of a class (perhaps via an instance of that class), if\n      that attribute is a user-defined function object or a class\n      method object.\n\n      When an instance method object is created by retrieving a user-\n      defined function object from a class via one of its instances,\n      its ``__self__`` attribute is the instance, and the method\n      object is said to be bound.  The new method\'s ``__func__``\n      attribute is the original function object.\n\n      When a user-defined method object is created by retrieving\n      another method object from a class or instance, the behaviour is\n      the same as for a function object, except that the ``__func__``\n      attribute of the new instance is not the original method object\n      but its ``__func__`` attribute.\n\n      When an instance method object is created by retrieving a class\n      method object from a class or instance, its ``__self__``\n      attribute is the class itself, and its ``__func__`` attribute is\n      the function object underlying the class method.\n\n      When an instance method object is called, the underlying\n      function (``__func__``) is called, inserting the class instance\n      (``__self__``) in front of the argument list.  For instance,\n      when ``C`` is a class which contains a definition for a function\n      ``f()``, and ``x`` is an instance of ``C``, calling ``x.f(1)``\n      is equivalent to calling ``C.f(x, 1)``.\n\n      When an instance method object is derived from a class method\n      object, the "class instance" stored in ``__self__`` will\n      actually be the class itself, so that calling either ``x.f(1)``\n      or ``C.f(1)`` is equivalent to calling ``f(C,1)`` where ``f`` is\n      the underlying function.\n\n      Note that the transformation from function object to instance\n      method object happens each time the attribute is retrieved from\n      the instance.  In some cases, a fruitful optimization is to\n      assign the attribute to a local variable and call that local\n      variable. Also notice that this transformation only happens for\n      user-defined functions; other callable objects (and all non-\n      callable objects) are retrieved without transformation.  It is\n      also important to note that user-defined functions which are\n      attributes of a class instance are not converted to bound\n      methods; this *only* happens when the function is an attribute\n      of the class.\n\n   Generator functions\n      A function or method which uses the ``yield`` statement (see\n      section *The yield statement*) is called a *generator function*.\n      Such a function, when called, always returns an iterator object\n      which can be used to execute the body of the function:  calling\n      the iterator\'s ``__next__()`` method will cause the function to\n      execute until it provides a value using the ``yield`` statement.\n      When the function executes a ``return`` statement or falls off\n      the end, a ``StopIteration`` exception is raised and the\n      iterator will have reached the end of the set of values to be\n      returned.\n\n   Built-in functions\n      A built-in function object is a wrapper around a C function.\n      Examples of built-in functions are ``len()`` and ``math.sin()``\n      (``math`` is a standard built-in module). The number and type of\n      the arguments are determined by the C function. Special read-\n      only attributes: ``__doc__`` is the function\'s documentation\n      string, or ``None`` if unavailable; ``__name__`` is the\n      function\'s name; ``__self__`` is set to ``None`` (but see the\n      next item); ``__module__`` is the name of the module the\n      function was defined in or ``None`` if unavailable.\n\n   Built-in methods\n      This is really a different disguise of a built-in function, this\n      time containing an object passed to the C function as an\n      implicit extra argument.  An example of a built-in method is\n      ``alist.append()``, assuming *alist* is a list object. In this\n      case, the special read-only attribute ``__self__`` is set to the\n      object denoted by *list*.\n\n   Classes\n      Classes are callable.  These objects normally act as factories\n      for new instances of themselves, but variations are possible for\n      class types that override ``__new__()``.  The arguments of the\n      call are passed to ``__new__()`` and, in the typical case, to\n      ``__init__()`` to initialize the new instance.\n\n   Class Instances\n      Instances of arbitrary classes can be made callable by defining\n      a ``__call__()`` method in their class.\n\nModules\n   Modules are imported by the ``import`` statement (see section *The\n   import statement*). A module object has a namespace implemented by\n   a dictionary object (this is the dictionary referenced by the\n   __globals__ attribute of functions defined in the module).\n   Attribute references are translated to lookups in this dictionary,\n   e.g., ``m.x`` is equivalent to ``m.__dict__["x"]``. A module object\n   does not contain the code object used to initialize the module\n   (since it isn\'t needed once the initialization is done).\n\n   Attribute assignment updates the module\'s namespace dictionary,\n   e.g., ``m.x = 1`` is equivalent to ``m.__dict__["x"] = 1``.\n\n   Special read-only attribute: ``__dict__`` is the module\'s namespace\n   as a dictionary object.\n\n   Predefined (writable) attributes: ``__name__`` is the module\'s\n   name; ``__doc__`` is the module\'s documentation string, or ``None``\n   if unavailable; ``__file__`` is the pathname of the file from which\n   the module was loaded, if it was loaded from a file. The\n   ``__file__`` attribute is not present for C modules that are\n   statically linked into the interpreter; for extension modules\n   loaded dynamically from a shared library, it is the pathname of the\n   shared library file.\n\nCustom classes\n   Custon class types are typically created by class definitions (see\n   section *Class definitions*).  A class has a namespace implemented\n   by a dictionary object. Class attribute references are translated\n   to lookups in this dictionary, e.g., ``C.x`` is translated to\n   ``C.__dict__["x"]`` (although there are a number of hooks which\n   allow for other means of locating attributes). When the attribute\n   name is not found there, the attribute search continues in the base\n   classes. This search of the base classes uses the C3 method\n   resolution order which behaves correctly even in the presence of\n   \'diamond\' inheritance structures where there are multiple\n   inheritance paths leading back to a common ancestor. Additional\n   details on the C3 MRO used by Python can be found in the\n   documentation accompanying the 2.3 release at\n   http://www.python.org/download/releases/2.3/mro/.\n\n   When a class attribute reference (for class ``C``, say) would yield\n   a class method object, it is transformed into an instance method\n   object whose ``__self__`` attributes is ``C``.  When it would yield\n   a static method object, it is transformed into the object wrapped\n   by the static method object. See section *Implementing Descriptors*\n   for another way in which attributes retrieved from a class may\n   differ from those actually contained in its ``__dict__``.\n\n   Class attribute assignments update the class\'s dictionary, never\n   the dictionary of a base class.\n\n   A class object can be called (see above) to yield a class instance\n   (see below).\n\n   Special attributes: ``__name__`` is the class name; ``__module__``\n   is the module name in which the class was defined; ``__dict__`` is\n   the dictionary containing the class\'s namespace; ``__bases__`` is a\n   tuple (possibly empty or a singleton) containing the base classes,\n   in the order of their occurrence in the base class list;\n   ``__doc__`` is the class\'s documentation string, or None if\n   undefined.\n\nClass instances\n   A class instance is created by calling a class object (see above).\n   A class instance has a namespace implemented as a dictionary which\n   is the first place in which attribute references are searched.\n   When an attribute is not found there, and the instance\'s class has\n   an attribute by that name, the search continues with the class\n   attributes.  If a class attribute is found that is a user-defined\n   function object, it is transformed into an instance method object\n   whose ``__self__`` attribute is the instance.  Static method and\n   class method objects are also transformed; see above under\n   "Classes".  See section *Implementing Descriptors* for another way\n   in which attributes of a class retrieved via its instances may\n   differ from the objects actually stored in the class\'s\n   ``__dict__``.  If no class attribute is found, and the object\'s\n   class has a ``__getattr__()`` method, that is called to satisfy the\n   lookup.\n\n   Attribute assignments and deletions update the instance\'s\n   dictionary, never a class\'s dictionary.  If the class has a\n   ``__setattr__()`` or ``__delattr__()`` method, this is called\n   instead of updating the instance dictionary directly.\n\n   Class instances can pretend to be numbers, sequences, or mappings\n   if they have methods with certain special names.  See section\n   *Special method names*.\n\n   Special attributes: ``__dict__`` is the attribute dictionary;\n   ``__class__`` is the instance\'s class.\n\nFiles\n   A file object represents an open file.  File objects are created by\n   the ``open()`` built-in function, and also by ``os.popen()``,\n   ``os.fdopen()``, and the ``makefile()`` method of socket objects\n   (and perhaps by other functions or methods provided by extension\n   modules).  The objects ``sys.stdin``, ``sys.stdout`` and\n   ``sys.stderr`` are initialized to file objects corresponding to the\n   interpreter\'s standard input, output and error streams.  See *File\n   Objects* for complete documentation of file objects.\n\nInternal types\n   A few types used internally by the interpreter are exposed to the\n   user. Their definitions may change with future versions of the\n   interpreter, but they are mentioned here for completeness.\n\n   Code objects\n      Code objects represent *byte-compiled* executable Python code,\n      or *bytecode*. The difference between a code object and a\n      function object is that the function object contains an explicit\n      reference to the function\'s globals (the module in which it was\n      defined), while a code object contains no context; also the\n      default argument values are stored in the function object, not\n      in the code object (because they represent values calculated at\n      run-time).  Unlike function objects, code objects are immutable\n      and contain no references (directly or indirectly) to mutable\n      objects.\n\n      Special read-only attributes: ``co_name`` gives the function\n      name; ``co_argcount`` is the number of positional arguments\n      (including arguments with default values); ``co_nlocals`` is the\n      number of local variables used by the function (including\n      arguments); ``co_varnames`` is a tuple containing the names of\n      the local variables (starting with the argument names);\n      ``co_cellvars`` is a tuple containing the names of local\n      variables that are referenced by nested functions;\n      ``co_freevars`` is a tuple containing the names of free\n      variables; ``co_code`` is a string representing the sequence of\n      bytecode instructions; ``co_consts`` is a tuple containing the\n      literals used by the bytecode; ``co_names`` is a tuple\n      containing the names used by the bytecode; ``co_filename`` is\n      the filename from which the code was compiled;\n      ``co_firstlineno`` is the first line number of the function;\n      ``co_lnotab`` is a string encoding the mapping from bytecode\n      offsets to line numbers (for details see the source code of the\n      interpreter); ``co_stacksize`` is the required stack size\n      (including local variables); ``co_flags`` is an integer encoding\n      a number of flags for the interpreter.\n\n      The following flag bits are defined for ``co_flags``: bit\n      ``0x04`` is set if the function uses the ``*arguments`` syntax\n      to accept an arbitrary number of positional arguments; bit\n      ``0x08`` is set if the function uses the ``**keywords`` syntax\n      to accept arbitrary keyword arguments; bit ``0x20`` is set if\n      the function is a generator.\n\n      Future feature declarations (``from __future__ import\n      division``) also use bits in ``co_flags`` to indicate whether a\n      code object was compiled with a particular feature enabled: bit\n      ``0x2000`` is set if the function was compiled with future\n      division enabled; bits ``0x10`` and ``0x1000`` were used in\n      earlier versions of Python.\n\n      Other bits in ``co_flags`` are reserved for internal use.\n\n      If a code object represents a function, the first item in\n      ``co_consts`` is the documentation string of the function, or\n      ``None`` if undefined.\n\n   Frame objects\n      Frame objects represent execution frames.  They may occur in\n      traceback objects (see below).\n\n      Special read-only attributes: ``f_back`` is to the previous\n      stack frame (towards the caller), or ``None`` if this is the\n      bottom stack frame; ``f_code`` is the code object being executed\n      in this frame; ``f_locals`` is the dictionary used to look up\n      local variables; ``f_globals`` is used for global variables;\n      ``f_builtins`` is used for built-in (intrinsic) names;\n      ``f_lasti`` gives the precise instruction (this is an index into\n      the bytecode string of the code object).\n\n      Special writable attributes: ``f_trace``, if not ``None``, is a\n      function called at the start of each source code line (this is\n      used by the debugger); ``f_lineno`` is the current line number\n      of the frame --- writing to this from within a trace function\n      jumps to the given line (only for the bottom-most frame).  A\n      debugger can implement a Jump command (aka Set Next Statement)\n      by writing to f_lineno.\n\n   Traceback objects\n      Traceback objects represent a stack trace of an exception.  A\n      traceback object is created when an exception occurs.  When the\n      search for an exception handler unwinds the execution stack, at\n      each unwound level a traceback object is inserted in front of\n      the current traceback.  When an exception handler is entered,\n      the stack trace is made available to the program. (See section\n      *The try statement*.) It is accessible as the third item of the\n      tuple returned by ``sys.exc_info()``. When the program contains\n      no suitable handler, the stack trace is written (nicely\n      formatted) to the standard error stream; if the interpreter is\n      interactive, it is also made available to the user as\n      ``sys.last_traceback``.\n\n      Special read-only attributes: ``tb_next`` is the next level in\n      the stack trace (towards the frame where the exception\n      occurred), or ``None`` if there is no next level; ``tb_frame``\n      points to the execution frame of the current level;\n      ``tb_lineno`` gives the line number where the exception\n      occurred; ``tb_lasti`` indicates the precise instruction.  The\n      line number and last instruction in the traceback may differ\n      from the line number of its frame object if the exception\n      occurred in a ``try`` statement with no matching except clause\n      or with a finally clause.\n\n   Slice objects\n      Slice objects are used to represent slices for ``__getitem__()``\n      methods.  They are also created by the built-in ``slice()``\n      function.\n\n      Special read-only attributes: ``start`` is the lower bound;\n      ``stop`` is the upper bound; ``step`` is the step value; each is\n      ``None`` if omitted. These attributes can have any type.\n\n      Slice objects support one method:\n\n      slice.indices(self, length)\n\n         This method takes a single integer argument *length* and\n         computes information about the slice that the slice object\n         would describe if applied to a sequence of *length* items.\n         It returns a tuple of three integers; respectively these are\n         the *start* and *stop* indices and the *step* or stride\n         length of the slice. Missing or out-of-bounds indices are\n         handled in a manner consistent with regular slices.\n\n   Static method objects\n      Static method objects provide a way of defeating the\n      transformation of function objects to method objects described\n      above. A static method object is a wrapper around any other\n      object, usually a user-defined method object. When a static\n      method object is retrieved from a class or a class instance, the\n      object actually returned is the wrapped object, which is not\n      subject to any further transformation. Static method objects are\n      not themselves callable, although the objects they wrap usually\n      are. Static method objects are created by the built-in\n      ``staticmethod()`` constructor.\n\n   Class method objects\n      A class method object, like a static method object, is a wrapper\n      around another object that alters the way in which that object\n      is retrieved from classes and class instances. The behaviour of\n      class method objects upon such retrieval is described above,\n      under "User-defined methods". Class method objects are created\n      by the built-in ``classmethod()`` constructor.\n',
+ 'try': '\nThe ``try`` statement\n*********************\n\nThe ``try`` statement specifies exception handlers and/or cleanup code\nfor a group of statements:\n\n   try_stmt  ::= try1_stmt | try2_stmt\n   try1_stmt ::= "try" ":" suite\n                 ("except" [expression ["as" target]] ":" suite)+\n                 ["else" ":" suite]\n                 ["finally" ":" suite]\n   try2_stmt ::= "try" ":" suite\n                 "finally" ":" suite\n\nThe ``except`` clause(s) specify one or more exception handlers. When\nno exception occurs in the ``try`` clause, no exception handler is\nexecuted. When an exception occurs in the ``try`` suite, a search for\nan exception handler is started.  This search inspects the except\nclauses in turn until one is found that matches the exception.  An\nexpression-less except clause, if present, must be last; it matches\nany exception.  For an except clause with an expression, that\nexpression is evaluated, and the clause matches the exception if the\nresulting object is "compatible" with the exception.  An object is\ncompatible with an exception if it is the class or a base class of the\nexception object or a tuple containing an item compatible with the\nexception.\n\nIf no except clause matches the exception, the search for an exception\nhandler continues in the surrounding code and on the invocation stack.\n[1]\n\nIf the evaluation of an expression in the header of an except clause\nraises an exception, the original search for a handler is canceled and\na search starts for the new exception in the surrounding code and on\nthe call stack (it is treated as if the entire ``try`` statement\nraised the exception).\n\nWhen a matching except clause is found, the exception is assigned to\nthe target specified after the ``as`` keyword in that except clause,\nif present, and the except clause\'s suite is executed.  All except\nclauses must have an executable block.  When the end of this block is\nreached, execution continues normally after the entire try statement.\n(This means that if two nested handlers exist for the same exception,\nand the exception occurs in the try clause of the inner handler, the\nouter handler will not handle the exception.)\n\nWhen an exception has been assigned using ``as target``, it is cleared\nat the end of the except clause.  This is as if\n\n   except E as N:\n       foo\n\nwas translated to\n\n   except E as N:\n       try:\n           foo\n       finally:\n           del N\n\nThis means the exception must be assigned to a different name to be\nable to refer to it after the except clause.  Exceptions are cleared\nbecause with the traceback attached to them, they form a reference\ncycle with the stack frame, keeping all locals in that frame alive\nuntil the next garbage collection occurs.\n\nBefore an except clause\'s suite is executed, details about the\nexception are stored in the ``sys`` module and can be access via\n``sys.exc_info()``. ``sys.exc_info()`` returns a 3-tuple consisting\nof: ``exc_type``, the exception class; ``exc_value``, the exception\ninstance; ``exc_traceback``, a traceback object (see section *The\nstandard type hierarchy*) identifying the point in the program where\nthe exception occurred. ``sys.exc_info()`` values are restored to\ntheir previous values (before the call) when returning from a function\nthat handled an exception.\n\nThe optional ``else`` clause is executed if and when control flows off\nthe end of the ``try`` clause. [2] Exceptions in the ``else`` clause\nare not handled by the preceding ``except`` clauses.\n\nIf ``finally`` is present, it specifies a \'cleanup\' handler.  The\n``try`` clause is executed, including any ``except`` and ``else``\nclauses.  If an exception occurs in any of the clauses and is not\nhandled, the exception is temporarily saved. The ``finally`` clause is\nexecuted.  If there is a saved exception, it is re-raised at the end\nof the ``finally`` clause. If the ``finally`` clause raises another\nexception or executes a ``return`` or ``break`` statement, the saved\nexception is lost.  The exception information is not available to the\nprogram during execution of the ``finally`` clause.\n\nWhen a ``return``, ``break`` or ``continue`` statement is executed in\nthe ``try`` suite of a ``try``...``finally`` statement, the\n``finally`` clause is also executed \'on the way out.\' A ``continue``\nstatement is illegal in the ``finally`` clause. (The reason is a\nproblem with the current implementation --- this restriction may be\nlifted in the future).\n\nAdditional information on exceptions can be found in section\n*Exceptions*, and information on using the ``raise`` statement to\ngenerate exceptions may be found in section *The raise statement*.\n',
+ 'types': '\nThe standard type hierarchy\n***************************\n\nBelow is a list of the types that are built into Python.  Extension\nmodules (written in C, Java, or other languages, depending on the\nimplementation) can define additional types.  Future versions of\nPython may add types to the type hierarchy (e.g., rational numbers,\nefficiently stored arrays of integers, etc.), although such additions\nwill often be provided via the standard library instead.\n\nSome of the type descriptions below contain a paragraph listing\n\'special attributes.\'  These are attributes that provide access to the\nimplementation and are not intended for general use.  Their definition\nmay change in the future.\n\nNone\n   This type has a single value.  There is a single object with this\n   value. This object is accessed through the built-in name ``None``.\n   It is used to signify the absence of a value in many situations,\n   e.g., it is returned from functions that don\'t explicitly return\n   anything. Its truth value is false.\n\nNotImplemented\n   This type has a single value.  There is a single object with this\n   value. This object is accessed through the built-in name\n   ``NotImplemented``. Numeric methods and rich comparison methods may\n   return this value if they do not implement the operation for the\n   operands provided.  (The interpreter will then try the reflected\n   operation, or some other fallback, depending on the operator.)  Its\n   truth value is true.\n\nEllipsis\n   This type has a single value.  There is a single object with this\n   value. This object is accessed through the literal ``...`` or the\n   built-in name ``Ellipsis``.  Its truth value is true.\n\n``numbers.Number``\n   These are created by numeric literals and returned as results by\n   arithmetic operators and arithmetic built-in functions.  Numeric\n   objects are immutable; once created their value never changes.\n   Python numbers are of course strongly related to mathematical\n   numbers, but subject to the limitations of numerical representation\n   in computers.\n\n   Python distinguishes between integers, floating point numbers, and\n   complex numbers:\n\n   ``numbers.Integral``\n      These represent elements from the mathematical set of integers\n      (positive and negative).\n\n      There are two types of integers:\n\n      Integers (``int``)\n\n         These represent numbers in an unlimited range, subject to\n         available (virtual) memory only.  For the purpose of shift\n         and mask operations, a binary representation is assumed, and\n         negative numbers are represented in a variant of 2\'s\n         complement which gives the illusion of an infinite string of\n         sign bits extending to the left.\n\n      Booleans (``bool``)\n         These represent the truth values False and True.  The two\n         objects representing the values False and True are the only\n         Boolean objects. The Boolean type is a subtype of the integer\n         type, and Boolean values behave like the values 0 and 1,\n         respectively, in almost all contexts, the exception being\n         that when converted to a string, the strings ``"False"`` or\n         ``"True"`` are returned, respectively.\n\n      The rules for integer representation are intended to give the\n      most meaningful interpretation of shift and mask operations\n      involving negative integers.\n\n   ``numbers.Real`` (``float``)\n      These represent machine-level double precision floating point\n      numbers. You are at the mercy of the underlying machine\n      architecture (and C or Java implementation) for the accepted\n      range and handling of overflow. Python does not support single-\n      precision floating point numbers; the savings in processor and\n      memory usage that are usually the reason for using these is\n      dwarfed by the overhead of using objects in Python, so there is\n      no reason to complicate the language with two kinds of floating\n      point numbers.\n\n   ``numbers.Complex`` (``complex``)\n      These represent complex numbers as a pair of machine-level\n      double precision floating point numbers.  The same caveats apply\n      as for floating point numbers. The real and imaginary parts of a\n      complex number ``z`` can be retrieved through the read-only\n      attributes ``z.real`` and ``z.imag``.\n\nSequences\n   These represent finite ordered sets indexed by non-negative\n   numbers. The built-in function ``len()`` returns the number of\n   items of a sequence. When the length of a sequence is *n*, the\n   index set contains the numbers 0, 1, ..., *n*-1.  Item *i* of\n   sequence *a* is selected by ``a[i]``.\n\n   Sequences also support slicing: ``a[i:j]`` selects all items with\n   index *k* such that *i* ``<=`` *k* ``<`` *j*.  When used as an\n   expression, a slice is a sequence of the same type.  This implies\n   that the index set is renumbered so that it starts at 0.\n\n   Some sequences also support "extended slicing" with a third "step"\n   parameter: ``a[i:j:k]`` selects all items of *a* with index *x*\n   where ``x = i + n*k``, *n* ``>=`` ``0`` and *i* ``<=`` *x* ``<``\n   *j*.\n\n   Sequences are distinguished according to their mutability:\n\n   Immutable sequences\n      An object of an immutable sequence type cannot change once it is\n      created.  (If the object contains references to other objects,\n      these other objects may be mutable and may be changed; however,\n      the collection of objects directly referenced by an immutable\n      object cannot change.)\n\n      The following types are immutable sequences:\n\n      Strings\n         The items of a string object are Unicode code units.  A\n         Unicode code unit is represented by a string object of one\n         item and can hold either a 16-bit or 32-bit value\n         representing a Unicode ordinal (the maximum value for the\n         ordinal is given in ``sys.maxunicode``, and depends on how\n         Python is configured at compile time).  Surrogate pairs may\n         be present in the Unicode object, and will be reported as two\n         separate items.  The built-in functions ``chr()`` and\n         ``ord()`` convert between code units and nonnegative integers\n         representing the Unicode ordinals as defined in the Unicode\n         Standard 3.0. Conversion from and to other encodings are\n         possible through the string method ``encode()``.\n\n      Tuples\n         The items of a tuple are arbitrary Python objects. Tuples of\n         two or more items are formed by comma-separated lists of\n         expressions.  A tuple of one item (a \'singleton\') can be\n         formed by affixing a comma to an expression (an expression by\n         itself does not create a tuple, since parentheses must be\n         usable for grouping of expressions).  An empty tuple can be\n         formed by an empty pair of parentheses.\n\n      Bytes\n         A bytes object is an immutable array.  The items are 8-bit\n         bytes, represented by integers in the range 0 <= x < 256.\n         Bytes literals (like ``b\'abc\'`` and the built-in function\n         ``bytes()`` can be used to construct bytes objects.  Also,\n         bytes objects can be decoded to strings via the ``decode()``\n         method.\n\n   Mutable sequences\n      Mutable sequences can be changed after they are created.  The\n      subscription and slicing notations can be used as the target of\n      assignment and ``del`` (delete) statements.\n\n      There are currently two intrinsic mutable sequence types:\n\n      Lists\n         The items of a list are arbitrary Python objects.  Lists are\n         formed by placing a comma-separated list of expressions in\n         square brackets. (Note that there are no special cases needed\n         to form lists of length 0 or 1.)\n\n      Byte Arrays\n         A bytearray object is a mutable array. They are created by\n         the built-in ``bytearray()`` constructor.  Aside from being\n         mutable (and hence unhashable), byte arrays otherwise provide\n         the same interface and functionality as immutable bytes\n         objects.\n\n      The extension module ``array`` provides an additional example of\n      a mutable sequence type, as does the ``collections`` module.\n\nSet types\n   These represent unordered, finite sets of unique, immutable\n   objects. As such, they cannot be indexed by any subscript. However,\n   they can be iterated over, and the built-in function ``len()``\n   returns the number of items in a set. Common uses for sets are fast\n   membership testing, removing duplicates from a sequence, and\n   computing mathematical operations such as intersection, union,\n   difference, and symmetric difference.\n\n   For set elements, the same immutability rules apply as for\n   dictionary keys. Note that numeric types obey the normal rules for\n   numeric comparison: if two numbers compare equal (e.g., ``1`` and\n   ``1.0``), only one of them can be contained in a set.\n\n   There are currently two intrinsic set types:\n\n   Sets\n      These represent a mutable set. They are created by the built-in\n      ``set()`` constructor and can be modified afterwards by several\n      methods, such as ``add()``.\n\n   Frozen sets\n      These represent an immutable set.  They are created by the\n      built-in ``frozenset()`` constructor.  As a frozenset is\n      immutable and *hashable*, it can be used again as an element of\n      another set, or as a dictionary key.\n\nMappings\n   These represent finite sets of objects indexed by arbitrary index\n   sets. The subscript notation ``a[k]`` selects the item indexed by\n   ``k`` from the mapping ``a``; this can be used in expressions and\n   as the target of assignments or ``del`` statements. The built-in\n   function ``len()`` returns the number of items in a mapping.\n\n   There is currently a single intrinsic mapping type:\n\n   Dictionaries\n      These represent finite sets of objects indexed by nearly\n      arbitrary values.  The only types of values not acceptable as\n      keys are values containing lists or dictionaries or other\n      mutable types that are compared by value rather than by object\n      identity, the reason being that the efficient implementation of\n      dictionaries requires a key\'s hash value to remain constant.\n      Numeric types used for keys obey the normal rules for numeric\n      comparison: if two numbers compare equal (e.g., ``1`` and\n      ``1.0``) then they can be used interchangeably to index the same\n      dictionary entry.\n\n      Dictionaries are mutable; they can be created by the ``{...}``\n      notation (see section *Dictionary displays*).\n\n      The extension modules ``dbm.ndbm`` and ``dbm.gnu`` provide\n      additional examples of mapping types, as does the\n      ``collections`` module.\n\nCallable types\n   These are the types to which the function call operation (see\n   section *Calls*) can be applied:\n\n   User-defined functions\n      A user-defined function object is created by a function\n      definition (see section *Function definitions*).  It should be\n      called with an argument list containing the same number of items\n      as the function\'s formal parameter list.\n\n      Special attributes:\n\n      +---------------------------+---------------------------------+-------------+\n      | Attribute                 | Meaning                         |             |\n      +===========================+=================================+=============+\n      | ``__doc__``               | The function\'s documentation    | Writable    |\n      |                           | string, or ``None`` if          |             |\n      |                           | unavailable                     |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__name__``              | The function\'s name             | Writable    |\n      +---------------------------+---------------------------------+-------------+\n      | ``__module__``            | The name of the module the      | Writable    |\n      |                           | function was defined in, or     |             |\n      |                           | ``None`` if unavailable.        |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__defaults__``          | A tuple containing default      | Writable    |\n      |                           | argument values for those       |             |\n      |                           | arguments that have defaults,   |             |\n      |                           | or ``None`` if no arguments     |             |\n      |                           | have a default value            |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__code__``              | The code object representing    | Writable    |\n      |                           | the compiled function body.     |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__globals__``           | A reference to the dictionary   | Read-only   |\n      |                           | that holds the function\'s       |             |\n      |                           | global variables --- the global |             |\n      |                           | namespace of the module in      |             |\n      |                           | which the function was defined. |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__dict__``              | The namespace supporting        | Writable    |\n      |                           | arbitrary function attributes.  |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__closure__``           | ``None`` or a tuple of cells    | Read-only   |\n      |                           | that contain bindings for the   |             |\n      |                           | function\'s free variables.      |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__annotations__``       | A dict containing annotations   | Writable    |\n      |                           | of parameters.  The keys of the |             |\n      |                           | dict are the parameter names,   |             |\n      |                           | or ``\'return\'`` for the return  |             |\n      |                           | annotation, if provided.        |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__kwdefaults__``        | A dict containing defaults for  | Writable    |\n      |                           | keyword-only parameters.        |             |\n      +---------------------------+---------------------------------+-------------+\n\n      Most of the attributes labelled "Writable" check the type of the\n      assigned value.\n\n      Function objects also support getting and setting arbitrary\n      attributes, which can be used, for example, to attach metadata\n      to functions.  Regular attribute dot-notation is used to get and\n      set such attributes. *Note that the current implementation only\n      supports function attributes on user-defined functions. Function\n      attributes on built-in functions may be supported in the\n      future.*\n\n      Additional information about a function\'s definition can be\n      retrieved from its code object; see the description of internal\n      types below.\n\n   Instance methods\n      An instance method object combines a class, a class instance and\n      any callable object (normally a user-defined function).\n\n      Special read-only attributes: ``__self__`` is the class instance\n      object, ``__func__`` is the function object; ``__doc__`` is the\n      method\'s documentation (same as ``__func__.__doc__``);\n      ``__name__`` is the method name (same as ``__func__.__name__``);\n      ``__module__`` is the name of the module the method was defined\n      in, or ``None`` if unavailable.\n\n      Methods also support accessing (but not setting) the arbitrary\n      function attributes on the underlying function object.\n\n      User-defined method objects may be created when getting an\n      attribute of a class (perhaps via an instance of that class), if\n      that attribute is a user-defined function object or a class\n      method object.\n\n      When an instance method object is created by retrieving a user-\n      defined function object from a class via one of its instances,\n      its ``__self__`` attribute is the instance, and the method\n      object is said to be bound.  The new method\'s ``__func__``\n      attribute is the original function object.\n\n      When a user-defined method object is created by retrieving\n      another method object from a class or instance, the behaviour is\n      the same as for a function object, except that the ``__func__``\n      attribute of the new instance is not the original method object\n      but its ``__func__`` attribute.\n\n      When an instance method object is created by retrieving a class\n      method object from a class or instance, its ``__self__``\n      attribute is the class itself, and its ``__func__`` attribute is\n      the function object underlying the class method.\n\n      When an instance method object is called, the underlying\n      function (``__func__``) is called, inserting the class instance\n      (``__self__``) in front of the argument list.  For instance,\n      when ``C`` is a class which contains a definition for a function\n      ``f()``, and ``x`` is an instance of ``C``, calling ``x.f(1)``\n      is equivalent to calling ``C.f(x, 1)``.\n\n      When an instance method object is derived from a class method\n      object, the "class instance" stored in ``__self__`` will\n      actually be the class itself, so that calling either ``x.f(1)``\n      or ``C.f(1)`` is equivalent to calling ``f(C,1)`` where ``f`` is\n      the underlying function.\n\n      Note that the transformation from function object to instance\n      method object happens each time the attribute is retrieved from\n      the instance.  In some cases, a fruitful optimization is to\n      assign the attribute to a local variable and call that local\n      variable. Also notice that this transformation only happens for\n      user-defined functions; other callable objects (and all non-\n      callable objects) are retrieved without transformation.  It is\n      also important to note that user-defined functions which are\n      attributes of a class instance are not converted to bound\n      methods; this *only* happens when the function is an attribute\n      of the class.\n\n   Generator functions\n      A function or method which uses the ``yield`` statement (see\n      section *The yield statement*) is called a *generator function*.\n      Such a function, when called, always returns an iterator object\n      which can be used to execute the body of the function:  calling\n      the iterator\'s ``__next__()`` method will cause the function to\n      execute until it provides a value using the ``yield`` statement.\n      When the function executes a ``return`` statement or falls off\n      the end, a ``StopIteration`` exception is raised and the\n      iterator will have reached the end of the set of values to be\n      returned.\n\n   Built-in functions\n      A built-in function object is a wrapper around a C function.\n      Examples of built-in functions are ``len()`` and ``math.sin()``\n      (``math`` is a standard built-in module). The number and type of\n      the arguments are determined by the C function. Special read-\n      only attributes: ``__doc__`` is the function\'s documentation\n      string, or ``None`` if unavailable; ``__name__`` is the\n      function\'s name; ``__self__`` is set to ``None`` (but see the\n      next item); ``__module__`` is the name of the module the\n      function was defined in or ``None`` if unavailable.\n\n   Built-in methods\n      This is really a different disguise of a built-in function, this\n      time containing an object passed to the C function as an\n      implicit extra argument.  An example of a built-in method is\n      ``alist.append()``, assuming *alist* is a list object. In this\n      case, the special read-only attribute ``__self__`` is set to the\n      object denoted by *list*.\n\n   Classes\n      Classes are callable.  These objects normally act as factories\n      for new instances of themselves, but variations are possible for\n      class types that override ``__new__()``.  The arguments of the\n      call are passed to ``__new__()`` and, in the typical case, to\n      ``__init__()`` to initialize the new instance.\n\n   Class Instances\n      Instances of arbitrary classes can be made callable by defining\n      a ``__call__()`` method in their class.\n\nModules\n   Modules are imported by the ``import`` statement (see section *The\n   import statement*). A module object has a namespace implemented by\n   a dictionary object (this is the dictionary referenced by the\n   __globals__ attribute of functions defined in the module).\n   Attribute references are translated to lookups in this dictionary,\n   e.g., ``m.x`` is equivalent to ``m.__dict__["x"]``. A module object\n   does not contain the code object used to initialize the module\n   (since it isn\'t needed once the initialization is done).\n\n   Attribute assignment updates the module\'s namespace dictionary,\n   e.g., ``m.x = 1`` is equivalent to ``m.__dict__["x"] = 1``.\n\n   Special read-only attribute: ``__dict__`` is the module\'s namespace\n   as a dictionary object.\n\n   Predefined (writable) attributes: ``__name__`` is the module\'s\n   name; ``__doc__`` is the module\'s documentation string, or ``None``\n   if unavailable; ``__file__`` is the pathname of the file from which\n   the module was loaded, if it was loaded from a file. The\n   ``__file__`` attribute is not present for C modules that are\n   statically linked into the interpreter; for extension modules\n   loaded dynamically from a shared library, it is the pathname of the\n   shared library file.\n\nCustom classes\n   Custom class types are typically created by class definitions (see\n   section *Class definitions*).  A class has a namespace implemented\n   by a dictionary object. Class attribute references are translated\n   to lookups in this dictionary, e.g., ``C.x`` is translated to\n   ``C.__dict__["x"]`` (although there are a number of hooks which\n   allow for other means of locating attributes). When the attribute\n   name is not found there, the attribute search continues in the base\n   classes. This search of the base classes uses the C3 method\n   resolution order which behaves correctly even in the presence of\n   \'diamond\' inheritance structures where there are multiple\n   inheritance paths leading back to a common ancestor. Additional\n   details on the C3 MRO used by Python can be found in the\n   documentation accompanying the 2.3 release at\n   http://www.python.org/download/releases/2.3/mro/.\n\n   When a class attribute reference (for class ``C``, say) would yield\n   a class method object, it is transformed into an instance method\n   object whose ``__self__`` attributes is ``C``.  When it would yield\n   a static method object, it is transformed into the object wrapped\n   by the static method object. See section *Implementing Descriptors*\n   for another way in which attributes retrieved from a class may\n   differ from those actually contained in its ``__dict__``.\n\n   Class attribute assignments update the class\'s dictionary, never\n   the dictionary of a base class.\n\n   A class object can be called (see above) to yield a class instance\n   (see below).\n\n   Special attributes: ``__name__`` is the class name; ``__module__``\n   is the module name in which the class was defined; ``__dict__`` is\n   the dictionary containing the class\'s namespace; ``__bases__`` is a\n   tuple (possibly empty or a singleton) containing the base classes,\n   in the order of their occurrence in the base class list;\n   ``__doc__`` is the class\'s documentation string, or None if\n   undefined.\n\nClass instances\n   A class instance is created by calling a class object (see above).\n   A class instance has a namespace implemented as a dictionary which\n   is the first place in which attribute references are searched.\n   When an attribute is not found there, and the instance\'s class has\n   an attribute by that name, the search continues with the class\n   attributes.  If a class attribute is found that is a user-defined\n   function object, it is transformed into an instance method object\n   whose ``__self__`` attribute is the instance.  Static method and\n   class method objects are also transformed; see above under\n   "Classes".  See section *Implementing Descriptors* for another way\n   in which attributes of a class retrieved via its instances may\n   differ from the objects actually stored in the class\'s\n   ``__dict__``.  If no class attribute is found, and the object\'s\n   class has a ``__getattr__()`` method, that is called to satisfy the\n   lookup.\n\n   Attribute assignments and deletions update the instance\'s\n   dictionary, never a class\'s dictionary.  If the class has a\n   ``__setattr__()`` or ``__delattr__()`` method, this is called\n   instead of updating the instance dictionary directly.\n\n   Class instances can pretend to be numbers, sequences, or mappings\n   if they have methods with certain special names.  See section\n   *Special method names*.\n\n   Special attributes: ``__dict__`` is the attribute dictionary;\n   ``__class__`` is the instance\'s class.\n\nI/O objects (also known as file objects)\n   A file object represents an open file.  Various shortcuts are\n   available to create file objects: the ``open()`` built-in function,\n   and also ``os.popen()``, ``os.fdopen()``, and the ``makefile()``\n   method of socket objects (and perhaps by other functions or methods\n   provided by extension modules).\n\n   The objects ``sys.stdin``, ``sys.stdout`` and ``sys.stderr`` are\n   initialized to file objects corresponding to the interpreter\'s\n   standard input, output and error streams; they are all open in text\n   mode and therefore follow the interface defined by the\n   ``io.TextIOBase`` abstract class.\n\nInternal types\n   A few types used internally by the interpreter are exposed to the\n   user. Their definitions may change with future versions of the\n   interpreter, but they are mentioned here for completeness.\n\n   Code objects\n      Code objects represent *byte-compiled* executable Python code,\n      or *bytecode*. The difference between a code object and a\n      function object is that the function object contains an explicit\n      reference to the function\'s globals (the module in which it was\n      defined), while a code object contains no context; also the\n      default argument values are stored in the function object, not\n      in the code object (because they represent values calculated at\n      run-time).  Unlike function objects, code objects are immutable\n      and contain no references (directly or indirectly) to mutable\n      objects.\n\n      Special read-only attributes: ``co_name`` gives the function\n      name; ``co_argcount`` is the number of positional arguments\n      (including arguments with default values); ``co_nlocals`` is the\n      number of local variables used by the function (including\n      arguments); ``co_varnames`` is a tuple containing the names of\n      the local variables (starting with the argument names);\n      ``co_cellvars`` is a tuple containing the names of local\n      variables that are referenced by nested functions;\n      ``co_freevars`` is a tuple containing the names of free\n      variables; ``co_code`` is a string representing the sequence of\n      bytecode instructions; ``co_consts`` is a tuple containing the\n      literals used by the bytecode; ``co_names`` is a tuple\n      containing the names used by the bytecode; ``co_filename`` is\n      the filename from which the code was compiled;\n      ``co_firstlineno`` is the first line number of the function;\n      ``co_lnotab`` is a string encoding the mapping from bytecode\n      offsets to line numbers (for details see the source code of the\n      interpreter); ``co_stacksize`` is the required stack size\n      (including local variables); ``co_flags`` is an integer encoding\n      a number of flags for the interpreter.\n\n      The following flag bits are defined for ``co_flags``: bit\n      ``0x04`` is set if the function uses the ``*arguments`` syntax\n      to accept an arbitrary number of positional arguments; bit\n      ``0x08`` is set if the function uses the ``**keywords`` syntax\n      to accept arbitrary keyword arguments; bit ``0x20`` is set if\n      the function is a generator.\n\n      Future feature declarations (``from __future__ import\n      division``) also use bits in ``co_flags`` to indicate whether a\n      code object was compiled with a particular feature enabled: bit\n      ``0x2000`` is set if the function was compiled with future\n      division enabled; bits ``0x10`` and ``0x1000`` were used in\n      earlier versions of Python.\n\n      Other bits in ``co_flags`` are reserved for internal use.\n\n      If a code object represents a function, the first item in\n      ``co_consts`` is the documentation string of the function, or\n      ``None`` if undefined.\n\n   Frame objects\n      Frame objects represent execution frames.  They may occur in\n      traceback objects (see below).\n\n      Special read-only attributes: ``f_back`` is to the previous\n      stack frame (towards the caller), or ``None`` if this is the\n      bottom stack frame; ``f_code`` is the code object being executed\n      in this frame; ``f_locals`` is the dictionary used to look up\n      local variables; ``f_globals`` is used for global variables;\n      ``f_builtins`` is used for built-in (intrinsic) names;\n      ``f_lasti`` gives the precise instruction (this is an index into\n      the bytecode string of the code object).\n\n      Special writable attributes: ``f_trace``, if not ``None``, is a\n      function called at the start of each source code line (this is\n      used by the debugger); ``f_lineno`` is the current line number\n      of the frame --- writing to this from within a trace function\n      jumps to the given line (only for the bottom-most frame).  A\n      debugger can implement a Jump command (aka Set Next Statement)\n      by writing to f_lineno.\n\n   Traceback objects\n      Traceback objects represent a stack trace of an exception.  A\n      traceback object is created when an exception occurs.  When the\n      search for an exception handler unwinds the execution stack, at\n      each unwound level a traceback object is inserted in front of\n      the current traceback.  When an exception handler is entered,\n      the stack trace is made available to the program. (See section\n      *The try statement*.) It is accessible as the third item of the\n      tuple returned by ``sys.exc_info()``. When the program contains\n      no suitable handler, the stack trace is written (nicely\n      formatted) to the standard error stream; if the interpreter is\n      interactive, it is also made available to the user as\n      ``sys.last_traceback``.\n\n      Special read-only attributes: ``tb_next`` is the next level in\n      the stack trace (towards the frame where the exception\n      occurred), or ``None`` if there is no next level; ``tb_frame``\n      points to the execution frame of the current level;\n      ``tb_lineno`` gives the line number where the exception\n      occurred; ``tb_lasti`` indicates the precise instruction.  The\n      line number and last instruction in the traceback may differ\n      from the line number of its frame object if the exception\n      occurred in a ``try`` statement with no matching except clause\n      or with a finally clause.\n\n   Slice objects\n      Slice objects are used to represent slices for ``__getitem__()``\n      methods.  They are also created by the built-in ``slice()``\n      function.\n\n      Special read-only attributes: ``start`` is the lower bound;\n      ``stop`` is the upper bound; ``step`` is the step value; each is\n      ``None`` if omitted. These attributes can have any type.\n\n      Slice objects support one method:\n\n      slice.indices(self, length)\n\n         This method takes a single integer argument *length* and\n         computes information about the slice that the slice object\n         would describe if applied to a sequence of *length* items.\n         It returns a tuple of three integers; respectively these are\n         the *start* and *stop* indices and the *step* or stride\n         length of the slice. Missing or out-of-bounds indices are\n         handled in a manner consistent with regular slices.\n\n   Static method objects\n      Static method objects provide a way of defeating the\n      transformation of function objects to method objects described\n      above. A static method object is a wrapper around any other\n      object, usually a user-defined method object. When a static\n      method object is retrieved from a class or a class instance, the\n      object actually returned is the wrapped object, which is not\n      subject to any further transformation. Static method objects are\n      not themselves callable, although the objects they wrap usually\n      are. Static method objects are created by the built-in\n      ``staticmethod()`` constructor.\n\n   Class method objects\n      A class method object, like a static method object, is a wrapper\n      around another object that alters the way in which that object\n      is retrieved from classes and class instances. The behaviour of\n      class method objects upon such retrieval is described above,\n      under "User-defined methods". Class method objects are created\n      by the built-in ``classmethod()`` constructor.\n',
  'typesfunctions': '\nFunctions\n*********\n\nFunction objects are created by function definitions.  The only\noperation on a function object is to call it: ``func(argument-list)``.\n\nThere are really two flavors of function objects: built-in functions\nand user-defined functions.  Both support the same operation (to call\nthe function), but the implementation is different, hence the\ndifferent object types.\n\nSee *Function definitions* for more information.\n',
- 'typesmapping': '\nMapping Types --- ``dict``\n**************************\n\nA *mapping* object maps *hashable* values to arbitrary objects.\nMappings are mutable objects.  There is currently only one standard\nmapping type, the *dictionary*.  (For other containers see the built\nin ``list``, ``set``, and ``tuple`` classes, and the ``collections``\nmodule.)\n\nA dictionary\'s keys are *almost* arbitrary values.  Values that are\nnot *hashable*, that is, values containing lists, dictionaries or\nother mutable types (that are compared by value rather than by object\nidentity) may not be used as keys.  Numeric types used for keys obey\nthe normal rules for numeric comparison: if two numbers compare equal\n(such as ``1`` and ``1.0``) then they can be used interchangeably to\nindex the same dictionary entry.  (Note however, that since computers\nstore floating-point numbers as approximations it is usually unwise to\nuse them as dictionary keys.)\n\nDictionaries can be created by placing a comma-separated list of\n``key: value`` pairs within braces, for example: ``{\'jack\': 4098,\n\'sjoerd\': 4127}`` or ``{4098: \'jack\', 4127: \'sjoerd\'}``, or by the\n``dict`` constructor.\n\nclass class dict([arg])\n\n   Return a new dictionary initialized from an optional positional\n   argument or from a set of keyword arguments.  If no arguments are\n   given, return a new empty dictionary.  If the positional argument\n   *arg* is a mapping object, return a dictionary mapping the same\n   keys to the same values as does the mapping object.  Otherwise the\n   positional argument must be a sequence, a container that supports\n   iteration, or an iterator object.  The elements of the argument\n   must each also be of one of those kinds, and each must in turn\n   contain exactly two objects.  The first is used as a key in the new\n   dictionary, and the second as the key\'s value.  If a given key is\n   seen more than once, the last value associated with it is retained\n   in the new dictionary.\n\n   If keyword arguments are given, the keywords themselves with their\n   associated values are added as items to the dictionary.  If a key\n   is specified both in the positional argument and as a keyword\n   argument, the value associated with the keyword is retained in the\n   dictionary.  For example, these all return a dictionary equal to\n   ``{"one": 2, "two": 3}``:\n\n   * ``dict(one=2, two=3)``\n\n   * ``dict({\'one\': 2, \'two\': 3})``\n\n   * ``dict(zip((\'one\', \'two\'), (2, 3)))``\n\n   * ``dict([[\'two\', 3], [\'one\', 2]])``\n\n   The first example only works for keys that are valid Python\n   identifiers; the others work with any valid keys.\n\n   These are the operations that dictionaries support (and therefore,\n   custom mapping types should support too):\n\n   len(d)\n\n      Return the number of items in the dictionary *d*.\n\n   d[key]\n\n      Return the item of *d* with key *key*.  Raises a ``KeyError`` if\n      *key* is not in the map.\n\n      If a subclass of dict defines a method ``__missing__()``, if the\n      key *key* is not present, the ``d[key]`` operation calls that\n      method with the key *key* as argument.  The ``d[key]`` operation\n      then returns or raises whatever is returned or raised by the\n      ``__missing__(key)`` call if the key is not present. No other\n      operations or methods invoke ``__missing__()``. If\n      ``__missing__()`` is not defined, ``KeyError`` is raised.\n      ``__missing__()`` must be a method; it cannot be an instance\n      variable. For an example, see ``collections.defaultdict``.\n\n   d[key] = value\n\n      Set ``d[key]`` to *value*.\n\n   del d[key]\n\n      Remove ``d[key]`` from *d*.  Raises a ``KeyError`` if *key* is\n      not in the map.\n\n   key in d\n\n      Return ``True`` if *d* has a key *key*, else ``False``.\n\n   key not in d\n\n      Equivalent to ``not key in d``.\n\n   iter(d)\n\n      Return an iterator over the keys of the dictionary.  This is a\n      shortcut for ``iterkeys()``.\n\n   clear()\n\n      Remove all items from the dictionary.\n\n   copy()\n\n      Return a shallow copy of the dictionary.\n\n   classmethod fromkeys(seq[, value])\n\n      Create a new dictionary with keys from *seq* and values set to\n      *value*.\n\n      ``fromkeys()`` is a class method that returns a new dictionary.\n      *value* defaults to ``None``.\n\n   get(key[, default])\n\n      Return the value for *key* if *key* is in the dictionary, else\n      *default*. If *default* is not given, it defaults to ``None``,\n      so that this method never raises a ``KeyError``.\n\n   items()\n\n      Return a new view of the dictionary\'s items (``(key, value)``\n      pairs).  See below for documentation of view objects.\n\n   keys()\n\n      Return a new view of the dictionary\'s keys.  See below for\n      documentation of view objects.\n\n   pop(key[, default])\n\n      If *key* is in the dictionary, remove it and return its value,\n      else return *default*.  If *default* is not given and *key* is\n      not in the dictionary, a ``KeyError`` is raised.\n\n   popitem()\n\n      Remove and return an arbitrary ``(key, value)`` pair from the\n      dictionary.\n\n      ``popitem()`` is useful to destructively iterate over a\n      dictionary, as often used in set algorithms.  If the dictionary\n      is empty, calling ``popitem()`` raises a ``KeyError``.\n\n   setdefault(key[, default])\n\n      If *key* is in the dictionary, return its value.  If not, insert\n      *key* with a value of *default* and return *default*.  *default*\n      defaults to ``None``.\n\n   update([other])\n\n      Update the dictionary with the key/value pairs from *other*,\n      overwriting existing keys.  Return ``None``.\n\n         ``update()`` accepts either another dictionary object or an\n         iterable of key/value pairs (as a tuple or other iterable of\n         length two).  If keyword arguments are specified, the\n         dictionary is then is updated with those key/value pairs:\n         ``d.update(red=1, blue=2)``.\n\n   values()\n\n      Return a new view of the dictionary\'s values.  See below for\n      documentation of view objects.\n\n\nDictionary view objects\n=======================\n\nThe objects returned by ``dict.keys()``, ``dict.values()`` and\n``dict.items()`` are *view objects*.  They provide a dynamic view on\nthe dictionary\'s entries, which means that when the dictionary\nchanges, the view reflects these changes.\n\nDictionary views can be iterated over to yield their respective data,\nand support membership tests:\n\nlen(dictview)\n\n   Return the number of entries in the dictionary.\n\niter(dictview)\n\n   Return an iterator over the keys, values or items (represented as\n   tuples of ``(key, value)``) in the dictionary.\n\n   Keys and values are iterated over in an arbitrary order which is\n   non-random, varies across Python implementations, and depends on\n   the dictionary\'s history of insertions and deletions. If keys,\n   values and items views are iterated over with no intervening\n   modifications to the dictionary, the order of items will directly\n   correspond.  This allows the creation of ``(value, key)`` pairs\n   using ``zip()``: ``pairs = zip(d.values(), d.keys())``.  Another\n   way to create the same list is ``pairs = [(v, k) for (k, v) in\n   d.items()]``.\n\n   Iterating views while adding or deleting entries in the dictionary\n   may raise a ``RuntimeError`` or fail to iterate over all entries.\n\nx in dictview\n\n   Return ``True`` if *x* is in the underlying dictionary\'s keys,\n   values or items (in the latter case, *x* should be a ``(key,\n   value)`` tuple).\n\nKeys views are set-like since their entries are unique and hashable.\nIf all values are hashable, so that (key, value) pairs are unique and\nhashable, then the items view is also set-like.  (Values views are not\ntreated as set-like since the entries are generally not unique.)  Then\nthese set operations are available ("other" refers either to another\nview or a set):\n\ndictview & other\n\n   Return the intersection of the dictview and the other object as a\n   new set.\n\ndictview | other\n\n   Return the union of the dictview and the other object as a new set.\n\ndictview - other\n\n   Return the difference between the dictview and the other object\n   (all elements in *dictview* that aren\'t in *other*) as a new set.\n\ndictview ^ other\n\n   Return the symmetric difference (all elements either in *dictview*\n   or *other*, but not in both) of the dictview and the other object\n   as a new set.\n\nAn example of dictionary view usage:\n\n   >>> dishes = {\'eggs\': 2, \'sausage\': 1, \'bacon\': 1, \'spam\': 500}\n   >>> keys = dishes.keys()\n   >>> values = dishes.values()\n\n   >>> # iteration\n   >>> n = 0\n   >>> for val in values:\n   ...     n += val\n   >>> print(n)\n   504\n\n   >>> # keys and values are iterated over in the same order\n   >>> list(keys)\n   [\'eggs\', \'bacon\', \'sausage\', \'spam\']\n   >>> list(values)\n   [2, 1, 1, 500]\n\n   >>> # view objects are dynamic and reflect dict changes\n   >>> del dishes[\'eggs\']\n   >>> del dishes[\'sausage\']\n   >>> list(keys)\n   [\'spam\', \'bacon\']\n\n   >>> # set operations\n   >>> keys & {\'eggs\', \'bacon\', \'salad\'}\n   {\'bacon\'}\n',
+ 'typesmapping': '\nMapping Types --- ``dict``\n**************************\n\nA *mapping* object maps *hashable* values to arbitrary objects.\nMappings are mutable objects.  There is currently only one standard\nmapping type, the *dictionary*.  (For other containers see the built\nin ``list``, ``set``, and ``tuple`` classes, and the ``collections``\nmodule.)\n\nA dictionary\'s keys are *almost* arbitrary values.  Values that are\nnot *hashable*, that is, values containing lists, dictionaries or\nother mutable types (that are compared by value rather than by object\nidentity) may not be used as keys.  Numeric types used for keys obey\nthe normal rules for numeric comparison: if two numbers compare equal\n(such as ``1`` and ``1.0``) then they can be used interchangeably to\nindex the same dictionary entry.  (Note however, that since computers\nstore floating-point numbers as approximations it is usually unwise to\nuse them as dictionary keys.)\n\nDictionaries can be created by placing a comma-separated list of\n``key: value`` pairs within braces, for example: ``{\'jack\': 4098,\n\'sjoerd\': 4127}`` or ``{4098: \'jack\', 4127: \'sjoerd\'}``, or by the\n``dict`` constructor.\n\nclass class dict([arg])\n\n   Return a new dictionary initialized from an optional positional\n   argument or from a set of keyword arguments.  If no arguments are\n   given, return a new empty dictionary.  If the positional argument\n   *arg* is a mapping object, return a dictionary mapping the same\n   keys to the same values as does the mapping object.  Otherwise the\n   positional argument must be a sequence, a container that supports\n   iteration, or an iterator object.  The elements of the argument\n   must each also be of one of those kinds, and each must in turn\n   contain exactly two objects.  The first is used as a key in the new\n   dictionary, and the second as the key\'s value.  If a given key is\n   seen more than once, the last value associated with it is retained\n   in the new dictionary.\n\n   If keyword arguments are given, the keywords themselves with their\n   associated values are added as items to the dictionary.  If a key\n   is specified both in the positional argument and as a keyword\n   argument, the value associated with the keyword is retained in the\n   dictionary.  For example, these all return a dictionary equal to\n   ``{"one": 2, "two": 3}``:\n\n   * ``dict(one=2, two=3)``\n\n   * ``dict({\'one\': 2, \'two\': 3})``\n\n   * ``dict(zip((\'one\', \'two\'), (2, 3)))``\n\n   * ``dict([[\'two\', 3], [\'one\', 2]])``\n\n   The first example only works for keys that are valid Python\n   identifiers; the others work with any valid keys.\n\n   These are the operations that dictionaries support (and therefore,\n   custom mapping types should support too):\n\n   len(d)\n\n      Return the number of items in the dictionary *d*.\n\n   d[key]\n\n      Return the item of *d* with key *key*.  Raises a ``KeyError`` if\n      *key* is not in the map.\n\n      If a subclass of dict defines a method ``__missing__()``, if the\n      key *key* is not present, the ``d[key]`` operation calls that\n      method with the key *key* as argument.  The ``d[key]`` operation\n      then returns or raises whatever is returned or raised by the\n      ``__missing__(key)`` call if the key is not present. No other\n      operations or methods invoke ``__missing__()``. If\n      ``__missing__()`` is not defined, ``KeyError`` is raised.\n      ``__missing__()`` must be a method; it cannot be an instance\n      variable. For an example, see ``collections.defaultdict``.\n\n   d[key] = value\n\n      Set ``d[key]`` to *value*.\n\n   del d[key]\n\n      Remove ``d[key]`` from *d*.  Raises a ``KeyError`` if *key* is\n      not in the map.\n\n   key in d\n\n      Return ``True`` if *d* has a key *key*, else ``False``.\n\n   key not in d\n\n      Equivalent to ``not key in d``.\n\n   iter(d)\n\n      Return an iterator over the keys of the dictionary.  This is a\n      shortcut for ``iter(d.keys())``.\n\n   clear()\n\n      Remove all items from the dictionary.\n\n   copy()\n\n      Return a shallow copy of the dictionary.\n\n   classmethod fromkeys(seq[, value])\n\n      Create a new dictionary with keys from *seq* and values set to\n      *value*.\n\n      ``fromkeys()`` is a class method that returns a new dictionary.\n      *value* defaults to ``None``.\n\n   get(key[, default])\n\n      Return the value for *key* if *key* is in the dictionary, else\n      *default*. If *default* is not given, it defaults to ``None``,\n      so that this method never raises a ``KeyError``.\n\n   items()\n\n      Return a new view of the dictionary\'s items (``(key, value)``\n      pairs).  See below for documentation of view objects.\n\n   keys()\n\n      Return a new view of the dictionary\'s keys.  See below for\n      documentation of view objects.\n\n   pop(key[, default])\n\n      If *key* is in the dictionary, remove it and return its value,\n      else return *default*.  If *default* is not given and *key* is\n      not in the dictionary, a ``KeyError`` is raised.\n\n   popitem()\n\n      Remove and return an arbitrary ``(key, value)`` pair from the\n      dictionary.\n\n      ``popitem()`` is useful to destructively iterate over a\n      dictionary, as often used in set algorithms.  If the dictionary\n      is empty, calling ``popitem()`` raises a ``KeyError``.\n\n   setdefault(key[, default])\n\n      If *key* is in the dictionary, return its value.  If not, insert\n      *key* with a value of *default* and return *default*.  *default*\n      defaults to ``None``.\n\n   update([other])\n\n      Update the dictionary with the key/value pairs from *other*,\n      overwriting existing keys.  Return ``None``.\n\n         ``update()`` accepts either another dictionary object or an\n         iterable of key/value pairs (as a tuple or other iterable of\n         length two).  If keyword arguments are specified, the\n         dictionary is then updated with those key/value pairs:\n         ``d.update(red=1, blue=2)``.\n\n   values()\n\n      Return a new view of the dictionary\'s values.  See below for\n      documentation of view objects.\n\n\nDictionary view objects\n=======================\n\nThe objects returned by ``dict.keys()``, ``dict.values()`` and\n``dict.items()`` are *view objects*.  They provide a dynamic view on\nthe dictionary\'s entries, which means that when the dictionary\nchanges, the view reflects these changes.\n\nDictionary views can be iterated over to yield their respective data,\nand support membership tests:\n\nlen(dictview)\n\n   Return the number of entries in the dictionary.\n\niter(dictview)\n\n   Return an iterator over the keys, values or items (represented as\n   tuples of ``(key, value)``) in the dictionary.\n\n   Keys and values are iterated over in an arbitrary order which is\n   non-random, varies across Python implementations, and depends on\n   the dictionary\'s history of insertions and deletions. If keys,\n   values and items views are iterated over with no intervening\n   modifications to the dictionary, the order of items will directly\n   correspond.  This allows the creation of ``(value, key)`` pairs\n   using ``zip()``: ``pairs = zip(d.values(), d.keys())``.  Another\n   way to create the same list is ``pairs = [(v, k) for (k, v) in\n   d.items()]``.\n\n   Iterating views while adding or deleting entries in the dictionary\n   may raise a ``RuntimeError`` or fail to iterate over all entries.\n\nx in dictview\n\n   Return ``True`` if *x* is in the underlying dictionary\'s keys,\n   values or items (in the latter case, *x* should be a ``(key,\n   value)`` tuple).\n\nKeys views are set-like since their entries are unique and hashable.\nIf all values are hashable, so that (key, value) pairs are unique and\nhashable, then the items view is also set-like.  (Values views are not\ntreated as set-like since the entries are generally not unique.)  Then\nthese set operations are available ("other" refers either to another\nview or a set):\n\ndictview & other\n\n   Return the intersection of the dictview and the other object as a\n   new set.\n\ndictview | other\n\n   Return the union of the dictview and the other object as a new set.\n\ndictview - other\n\n   Return the difference between the dictview and the other object\n   (all elements in *dictview* that aren\'t in *other*) as a new set.\n\ndictview ^ other\n\n   Return the symmetric difference (all elements either in *dictview*\n   or *other*, but not in both) of the dictview and the other object\n   as a new set.\n\nAn example of dictionary view usage:\n\n   >>> dishes = {\'eggs\': 2, \'sausage\': 1, \'bacon\': 1, \'spam\': 500}\n   >>> keys = dishes.keys()\n   >>> values = dishes.values()\n\n   >>> # iteration\n   >>> n = 0\n   >>> for val in values:\n   ...     n += val\n   >>> print(n)\n   504\n\n   >>> # keys and values are iterated over in the same order\n   >>> list(keys)\n   [\'eggs\', \'bacon\', \'sausage\', \'spam\']\n   >>> list(values)\n   [2, 1, 1, 500]\n\n   >>> # view objects are dynamic and reflect dict changes\n   >>> del dishes[\'eggs\']\n   >>> del dishes[\'sausage\']\n   >>> list(keys)\n   [\'spam\', \'bacon\']\n\n   >>> # set operations\n   >>> keys & {\'eggs\', \'bacon\', \'salad\'}\n   {\'bacon\'}\n',
  'typesmethods': "\nMethods\n*******\n\nMethods are functions that are called using the attribute notation.\nThere are two flavors: built-in methods (such as ``append()`` on\nlists) and class instance methods.  Built-in methods are described\nwith the types that support them.\n\nIf you access a method (a function defined in a class namespace)\nthrough an instance, you get a special object: a *bound method* (also\ncalled *instance method*) object. When called, it will add the\n``self`` argument to the argument list.  Bound methods have two\nspecial read-only attributes: ``m.__self__`` is the object on which\nthe method operates, and ``m.__func__`` is the function implementing\nthe method.  Calling ``m(arg-1, arg-2, ..., arg-n)`` is completely\nequivalent to calling ``m.__func__(m.__self__, arg-1, arg-2, ...,\narg-n)``.\n\nLike function objects, bound method objects support getting arbitrary\nattributes.  However, since method attributes are actually stored on\nthe underlying function object (``meth.__func__``), setting method\nattributes on bound methods is disallowed.  Attempting to set a method\nattribute results in a ``TypeError`` being raised.  In order to set a\nmethod attribute, you need to explicitly set it on the underlying\nfunction object:\n\n   class C:\n       def method(self):\n           pass\n\n   c = C()\n   c.method.__func__.whoami = 'my name is c'\n\nSee *The standard type hierarchy* for more information.\n",
  'typesmodules': "\nModules\n*******\n\nThe only special operation on a module is attribute access:\n``m.name``, where *m* is a module and *name* accesses a name defined\nin *m*'s symbol table. Module attributes can be assigned to.  (Note\nthat the ``import`` statement is not, strictly speaking, an operation\non a module object; ``import foo`` does not require a module object\nnamed *foo* to exist, rather it requires an (external) *definition*\nfor a module named *foo* somewhere.)\n\nA special member of every module is ``__dict__``. This is the\ndictionary containing the module's symbol table. Modifying this\ndictionary will actually change the module's symbol table, but direct\nassignment to the ``__dict__`` attribute is not possible (you can\nwrite ``m.__dict__['a'] = 1``, which defines ``m.a`` to be ``1``, but\nyou can't write ``m.__dict__ = {}``).  Modifying ``__dict__`` directly\nis not recommended.\n\nModules built into the interpreter are written like this: ``<module\n'sys' (built-in)>``.  If loaded from a file, they are written as\n``<module 'os' from '/usr/local/lib/pythonX.Y/os.pyc'>``.\n",
- 'typesseq': '\nSequence Types --- ``str``, ``bytes``, ``bytearray``, ``list``, ``tuple``, ``range``\n************************************************************************************\n\nThere are six sequence types: strings, byte sequences (``bytes``\nobjects), byte arrays (``bytearray`` objects), lists, tuples, and\nrange objects.  For other containers see the built in ``dict`` and\n``set`` classes, and the ``collections`` module.\n\nStrings contain Unicode characters.  Their literals are written in\nsingle or double quotes: ``\'xyzzy\'``, ``"frobozz"``.  See *String and\nBytes literals* for more about string literals.  In addition to the\nfunctionality described here, there are also string-specific methods\ndescribed in the *String Methods* section.\n\nBytes and bytearray objects contain single bytes -- the former is\nimmutable while the latter is a mutable sequence.  Bytes objects can\nbe constructed the constructor, ``bytes()``, and from literals; use a\n``b`` prefix with normal string syntax: ``b\'xyzzy\'``.  To construct\nbyte arrays, use the ``bytearray()`` function.\n\nWarning: While string objects are sequences of characters (represented by\n  strings of length 1), bytes and bytearray objects are sequences of\n  *integers* (between 0 and 255), representing the ASCII value of\n  single bytes.  That means that for a bytes or bytearray object *b*,\n  ``b[0]`` will be an integer, while ``b[0:1]`` will be a bytes or\n  bytearray object of length 1.  The representation of bytes objects\n  uses the literal format (``b\'...\'``) since it is generally more\n  useful than e.g. ``bytes([50, 19, 100])``.  You can always convert a\n  bytes object into a list of integers using ``list(b)``.Also, while\n  in previous Python versions, byte strings and Unicode strings could\n  be exchanged for each other rather freely (barring encoding issues),\n  strings and bytes are now completely separate concepts.  There\'s no\n  implicit en-/decoding if you pass and object of the wrong type.  A\n  string always compares unequal to a bytes or bytearray object.\n\nLists are constructed with square brackets, separating items with\ncommas: ``[a, b, c]``.  Tuples are constructed by the comma operator\n(not within square brackets), with or without enclosing parentheses,\nbut an empty tuple must have the enclosing parentheses, such as ``a,\nb, c`` or ``()``.  A single item tuple must have a trailing comma,\nsuch as ``(d,)``.\n\nObjects of type range are created using the ``range()`` function.\nThey don\'t support slicing, concatenation or repetition, and using\n``in``, ``not in``, ``min()`` or ``max()`` on them is inefficient.\n\nMost sequence types support the following operations.  The ``in`` and\n``not in`` operations have the same priorities as the comparison\noperations.  The ``+`` and ``*`` operations have the same priority as\nthe corresponding numeric operations. [3] Additional methods are\nprovided for *Mutable Sequence Types*.\n\nThis table lists the sequence operations sorted in ascending priority\n(operations in the same box have the same priority).  In the table,\n*s* and *t* are sequences of the same type; *n*, *i* and *j* are\nintegers:\n\n+--------------------+----------------------------------+------------+\n| Operation          | Result                           | Notes      |\n+====================+==================================+============+\n| ``x in s``         | ``True`` if an item of *s* is    | (1)        |\n|                    | equal to *x*, else ``False``     |            |\n+--------------------+----------------------------------+------------+\n| ``x not in s``     | ``False`` if an item of *s* is   | (1)        |\n|                    | equal to *x*, else ``True``      |            |\n+--------------------+----------------------------------+------------+\n| ``s + t``          | the concatenation of *s* and *t* | (6)        |\n+--------------------+----------------------------------+------------+\n| ``s * n, n * s``   | *n* shallow copies of *s*        | (2)        |\n|                    | concatenated                     |            |\n+--------------------+----------------------------------+------------+\n| ``s[i]``           | *i*\'th item of *s*, origin 0     | (3)        |\n+--------------------+----------------------------------+------------+\n| ``s[i:j]``         | slice of *s* from *i* to *j*     | (3)(4)     |\n+--------------------+----------------------------------+------------+\n| ``s[i:j:k]``       | slice of *s* from *i* to *j*     | (3)(5)     |\n|                    | with step *k*                    |            |\n+--------------------+----------------------------------+------------+\n| ``len(s)``         | length of *s*                    |            |\n+--------------------+----------------------------------+------------+\n| ``min(s)``         | smallest item of *s*             |            |\n+--------------------+----------------------------------+------------+\n| ``max(s)``         | largest item of *s*              |            |\n+--------------------+----------------------------------+------------+\n\nSequence types also support comparisons.  In particular, tuples and\nlists are compared lexicographically by comparing corresponding\nelements.  This means that to compare equal, every element must\ncompare equal and the two sequences must be of the same type and have\nthe same length.  (For full details see *Comparisons* in the language\nreference.)\n\nNotes:\n\n1. When *s* is a string object, the ``in`` and ``not in`` operations\n   act like a substring test.\n\n2. Values of *n* less than ``0`` are treated as ``0`` (which yields an\n   empty sequence of the same type as *s*).  Note also that the copies\n   are shallow; nested structures are not copied.  This often haunts\n   new Python programmers; consider:\n\n   >>> lists = [[]] * 3\n   >>> lists\n   [[], [], []]\n   >>> lists[0].append(3)\n   >>> lists\n   [[3], [3], [3]]\n\n   What has happened is that ``[[]]`` is a one-element list containing\n   an empty list, so all three elements of ``[[]] * 3`` are (pointers\n   to) this single empty list.  Modifying any of the elements of\n   ``lists`` modifies this single list. You can create a list of\n   different lists this way:\n\n   >>> lists = [[] for i in range(3)]\n   >>> lists[0].append(3)\n   >>> lists[1].append(5)\n   >>> lists[2].append(7)\n   >>> lists\n   [[3], [5], [7]]\n\n3. If *i* or *j* is negative, the index is relative to the end of the\n   string: ``len(s) + i`` or ``len(s) + j`` is substituted.  But note\n   that ``-0`` is still ``0``.\n\n4. The slice of *s* from *i* to *j* is defined as the sequence of\n   items with index *k* such that ``i <= k < j``.  If *i* or *j* is\n   greater than ``len(s)``, use ``len(s)``.  If *i* is omitted or\n   ``None``, use ``0``.  If *j* is omitted or ``None``, use\n   ``len(s)``.  If *i* is greater than or equal to *j*, the slice is\n   empty.\n\n5. The slice of *s* from *i* to *j* with step *k* is defined as the\n   sequence of items with index  ``x = i + n*k`` such that ``0 <= n <\n   (j-i)/k``.  In other words, the indices are ``i``, ``i+k``,\n   ``i+2*k``, ``i+3*k`` and so on, stopping when *j* is reached (but\n   never including *j*).  If *i* or *j* is greater than ``len(s)``,\n   use ``len(s)``.  If *i* or *j* are omitted or ``None``, they become\n   "end" values (which end depends on the sign of *k*).  Note, *k*\n   cannot be zero. If *k* is ``None``, it is treated like ``1``.\n\n6. If *s* and *t* are both strings, some Python implementations such\n   as CPython can usually perform an in-place optimization for\n   assignments of the form ``s=s+t`` or ``s+=t``.  When applicable,\n   this optimization makes quadratic run-time much less likely.  This\n   optimization is both version and implementation dependent. For\n   performance sensitive code, it is preferable to use the\n   ``str.join()`` method which assures consistent linear concatenation\n   performance across versions and implementations.\n\n\nString Methods\n==============\n\nString objects support the methods listed below.  Note that none of\nthese methods take keyword arguments.\n\nIn addition, Python\'s strings support the sequence type methods\ndescribed in the *Sequence Types --- str, bytes, bytearray, list,\ntuple, range* section. To output formatted strings, see the *String\nFormatting* section. Also, see the ``re`` module for string functions\nbased on regular expressions.\n\nstr.capitalize()\n\n   Return a copy of the string with only its first character\n   capitalized.\n\nstr.center(width[, fillchar])\n\n   Return centered in a string of length *width*. Padding is done\n   using the specified *fillchar* (default is a space).\n\nstr.count(sub[, start[, end]])\n\n   Return the number of non-overlapping occurrences of substring *sub*\n   in the range [*start*, *end*].  Optional arguments *start* and\n   *end* are interpreted as in slice notation.\n\nstr.encode([encoding[, errors]])\n\n   Return an encoded version of the string as a bytes object.  Default\n   encoding is the current default string encoding.  *errors* may be\n   given to set a different error handling scheme.  The default for\n   *errors* is ``\'strict\'``, meaning that encoding errors raise a\n   ``UnicodeError``.  Other possible values are ``\'ignore\'``,\n   ``\'replace\'``, ``\'xmlcharrefreplace\'``, ``\'backslashreplace\'`` and\n   any other name registered via ``codecs.register_error()``, see\n   section *Codec Base Classes*. For a list of possible encodings, see\n   section *Standard Encodings*.\n\nstr.endswith(suffix[, start[, end]])\n\n   Return ``True`` if the string ends with the specified *suffix*,\n   otherwise return ``False``.  *suffix* can also be a tuple of\n   suffixes to look for.  With optional *start*, test beginning at\n   that position.  With optional *end*, stop comparing at that\n   position.\n\nstr.expandtabs([tabsize])\n\n   Return a copy of the string where all tab characters are replaced\n   by one or more spaces, depending on the current column and the\n   given tab size.  The column number is reset to zero after each\n   newline occurring in the string. If *tabsize* is not given, a tab\n   size of ``8`` characters is assumed.  This doesn\'t understand other\n   non-printing characters or escape sequences.\n\nstr.find(sub[, start[, end]])\n\n   Return the lowest index in the string where substring *sub* is\n   found, such that *sub* is contained in the range [*start*, *end*].\n   Optional arguments *start* and *end* are interpreted as in slice\n   notation.  Return ``-1`` if *sub* is not found.\n\nstr.format(*args, **kwargs)\n\n   Perform a string formatting operation.  The *format_string*\n   argument can contain literal text or replacement fields delimited\n   by braces ``{}``.  Each replacement field contains either the\n   numeric index of a positional argument, or the name of a keyword\n   argument.  Returns a copy of *format_string* where each replacement\n   field is replaced with the string value of the corresponding\n   argument.\n\n   >>> "The sum of 1 + 2 is {0}".format(1+2)\n   \'The sum of 1 + 2 is 3\'\n\n   See *Format String Syntax* for a description of the various\n   formatting options that can be specified in format strings.\n\nstr.index(sub[, start[, end]])\n\n   Like ``find()``, but raise ``ValueError`` when the substring is not\n   found.\n\nstr.isalnum()\n\n   Return true if all characters in the string are alphanumeric and\n   there is at least one character, false otherwise.\n\nstr.isalpha()\n\n   Return true if all characters in the string are alphabetic and\n   there is at least one character, false otherwise.\n\nstr.isdecimal()\n\n   Return true if all characters in the string are decimal characters\n   and there is at least one character, false otherwise. Decimal\n   characters include digit characters, and all characters that that\n   can be used to form decimal-radix numbers, e.g. U+0660, ARABIC-\n   INDIC DIGIT ZERO.\n\nstr.isdigit()\n\n   Return true if all characters in the string are digits and there is\n   at least one character, false otherwise.\n\nstr.isidentifier()\n\n   Return true if the string is a valid identifier according to the\n   language definition, section *Identifiers and keywords*.\n\nstr.islower()\n\n   Return true if all cased characters in the string are lowercase and\n   there is at least one cased character, false otherwise.\n\nstr.isnumeric()\n\n   Return true if all characters in the string are numeric characters,\n   and there is at least one character, false otherwise. Numeric\n   characters include digit characters, and all characters that have\n   the Unicode numeric value property, e.g. U+2155, VULGAR FRACTION\n   ONE FIFTH.\n\nstr.isprintable()\n\n   Return true if all characters in the string are printable or the\n   string is empty, false otherwise.  Nonprintable characters are\n   those characters defined in the Unicode character database as\n   "Other" or "Separator", excepting the ASCII space (0x20) which is\n   considered printable.  (Note that printable characters in this\n   context are those which should not be escaped when ``repr()`` is\n   invoked on a string.  It has no bearing on the handling of strings\n   written to ``sys.stdout`` or ``sys.stderr``.)\n\nstr.isspace()\n\n   Return true if there are only whitespace characters in the string\n   and there is at least one character, false otherwise.\n\nstr.istitle()\n\n   Return true if the string is a titlecased string and there is at\n   least one character, for example uppercase characters may only\n   follow uncased characters and lowercase characters only cased ones.\n   Return false otherwise.\n\nstr.isupper()\n\n   Return true if all cased characters in the string are uppercase and\n   there is at least one cased character, false otherwise.\n\nstr.join(seq)\n\n   Return a string which is the concatenation of the strings in the\n   sequence *seq*.  A ``TypeError`` will be raised if there are any\n   non-string values in *seq*, including ``bytes`` objects.  The\n   separator between elements is the string providing this method.\n\nstr.ljust(width[, fillchar])\n\n   Return the string left justified in a string of length *width*.\n   Padding is done using the specified *fillchar* (default is a\n   space).  The original string is returned if *width* is less than\n   ``len(s)``.\n\nstr.lower()\n\n   Return a copy of the string converted to lowercase.\n\nstr.lstrip([chars])\n\n   Return a copy of the string with leading characters removed.  The\n   *chars* argument is a string specifying the set of characters to be\n   removed.  If omitted or ``None``, the *chars* argument defaults to\n   removing whitespace.  The *chars* argument is not a prefix; rather,\n   all combinations of its values are stripped:\n\n   >>> \'   spacious   \'.lstrip()\n   \'spacious   \'\n   >>> \'www.example.com\'.lstrip(\'cmowz.\')\n   \'example.com\'\n\nstatic str.maketrans(x[, y[, z]])\n\n   This static method returns a translation table usable for\n   ``str.translate()``.\n\n   If there is only one argument, it must be a dictionary mapping\n   Unicode ordinals (integers) or characters (strings of length 1) to\n   Unicode ordinals, strings (of arbitrary lengths) or None.\n   Character keys will then be converted to ordinals.\n\n   If there are two arguments, they must be strings of equal length,\n   and in the resulting dictionary, each character in x will be mapped\n   to the character at the same position in y.  If there is a third\n   argument, it must be a string, whose characters will be mapped to\n   None in the result.\n\nstr.partition(sep)\n\n   Split the string at the first occurrence of *sep*, and return a\n   3-tuple containing the part before the separator, the separator\n   itself, and the part after the separator.  If the separator is not\n   found, return a 3-tuple containing the string itself, followed by\n   two empty strings.\n\nstr.replace(old, new[, count])\n\n   Return a copy of the string with all occurrences of substring *old*\n   replaced by *new*.  If the optional argument *count* is given, only\n   the first *count* occurrences are replaced.\n\nstr.rfind(sub[, start[, end]])\n\n   Return the highest index in the string where substring *sub* is\n   found, such that *sub* is contained within s[start,end].  Optional\n   arguments *start* and *end* are interpreted as in slice notation.\n   Return ``-1`` on failure.\n\nstr.rindex(sub[, start[, end]])\n\n   Like ``rfind()`` but raises ``ValueError`` when the substring *sub*\n   is not found.\n\nstr.rjust(width[, fillchar])\n\n   Return the string right justified in a string of length *width*.\n   Padding is done using the specified *fillchar* (default is a\n   space). The original string is returned if *width* is less than\n   ``len(s)``.\n\nstr.rpartition(sep)\n\n   Split the string at the last occurrence of *sep*, and return a\n   3-tuple containing the part before the separator, the separator\n   itself, and the part after the separator.  If the separator is not\n   found, return a 3-tuple containing two empty strings, followed by\n   the string itself.\n\nstr.rsplit([sep[, maxsplit]])\n\n   Return a list of the words in the string, using *sep* as the\n   delimiter string. If *maxsplit* is given, at most *maxsplit* splits\n   are done, the *rightmost* ones.  If *sep* is not specified or\n   ``None``, any whitespace string is a separator.  Except for\n   splitting from the right, ``rsplit()`` behaves like ``split()``\n   which is described in detail below.\n\nstr.rstrip([chars])\n\n   Return a copy of the string with trailing characters removed.  The\n   *chars* argument is a string specifying the set of characters to be\n   removed.  If omitted or ``None``, the *chars* argument defaults to\n   removing whitespace.  The *chars* argument is not a suffix; rather,\n   all combinations of its values are stripped:\n\n   >>> \'   spacious   \'.rstrip()\n   \'   spacious\'\n   >>> \'mississippi\'.rstrip(\'ipz\')\n   \'mississ\'\n\nstr.split([sep[, maxsplit]])\n\n   Return a list of the words in the string, using *sep* as the\n   delimiter string.  If *maxsplit* is given, at most *maxsplit*\n   splits are done (thus, the list will have at most ``maxsplit+1``\n   elements).  If *maxsplit* is not specified, then there is no limit\n   on the number of splits (all possible splits are made).\n\n   If *sep* is given, consecutive delimiters are not grouped together\n   and are deemed to delimit empty strings (for example,\n   ``\'1,,2\'.split(\',\')`` returns ``[\'1\', \'\', \'2\']``).  The *sep*\n   argument may consist of multiple characters (for example,\n   ``\'1<>2<>3\'.split(\'<>\')`` returns ``[\'1\', \'2\', \'3\']``). Splitting\n   an empty string with a specified separator returns ``[\'\']``.\n\n   If *sep* is not specified or is ``None``, a different splitting\n   algorithm is applied: runs of consecutive whitespace are regarded\n   as a single separator, and the result will contain no empty strings\n   at the start or end if the string has leading or trailing\n   whitespace.  Consequently, splitting an empty string or a string\n   consisting of just whitespace with a ``None`` separator returns\n   ``[]``.\n\n   For example, ``\' 1  2   3  \'.split()`` returns ``[\'1\', \'2\', \'3\']``,\n   and ``\'  1  2   3  \'.split(None, 1)`` returns ``[\'1\', \'2   3  \']``.\n\nstr.splitlines([keepends])\n\n   Return a list of the lines in the string, breaking at line\n   boundaries.  Line breaks are not included in the resulting list\n   unless *keepends* is given and true.\n\nstr.startswith(prefix[, start[, end]])\n\n   Return ``True`` if string starts with the *prefix*, otherwise\n   return ``False``. *prefix* can also be a tuple of prefixes to look\n   for.  With optional *start*, test string beginning at that\n   position.  With optional *end*, stop comparing string at that\n   position.\n\nstr.strip([chars])\n\n   Return a copy of the string with the leading and trailing\n   characters removed. The *chars* argument is a string specifying the\n   set of characters to be removed. If omitted or ``None``, the\n   *chars* argument defaults to removing whitespace. The *chars*\n   argument is not a prefix or suffix; rather, all combinations of its\n   values are stripped:\n\n   >>> \'   spacious   \'.strip()\n   \'spacious\'\n   >>> \'www.example.com\'.strip(\'cmowz.\')\n   \'example\'\n\nstr.swapcase()\n\n   Return a copy of the string with uppercase characters converted to\n   lowercase and vice versa.\n\nstr.title()\n\n   Return a titlecased version of the string: words start with\n   uppercase characters, all remaining cased characters are lowercase.\n\nstr.translate(map)\n\n   Return a copy of the *s* where all characters have been mapped\n   through the *map* which must be a dictionary of Unicode ordinals\n   (integers) to Unicode ordinals, strings or ``None``.  Unmapped\n   characters are left untouched. Characters mapped to ``None`` are\n   deleted.\n\n   You can use ``str.maketrans()`` to create a translation map from\n   character-to-character mappings in different formats.\n\n   Note: An even more flexible approach is to create a custom character\n     mapping codec using the ``codecs`` module (see\n     ``encodings.cp1251`` for an example).\n\nstr.upper()\n\n   Return a copy of the string converted to uppercase.\n\nstr.zfill(width)\n\n   Return the numeric string left filled with zeros in a string of\n   length *width*.  A sign prefix is handled correctly.  The original\n   string is returned if *width* is less than ``len(s)``.\n\n\nOld String Formatting Operations\n================================\n\nNote: The formatting operations described here are obsolete and may go\n  away in future versions of Python.  Use the new *String Formatting*\n  in new code.\n\nString objects have one unique built-in operation: the ``%`` operator\n(modulo). This is also known as the string *formatting* or\n*interpolation* operator. Given ``format % values`` (where *format* is\na string), ``%`` conversion specifications in *format* are replaced\nwith zero or more elements of *values*. The effect is similar to the\nusing ``sprintf()`` in the C language.\n\nIf *format* requires a single argument, *values* may be a single non-\ntuple object. [4]  Otherwise, *values* must be a tuple with exactly\nthe number of items specified by the format string, or a single\nmapping object (for example, a dictionary).\n\nA conversion specifier contains two or more characters and has the\nfollowing components, which must occur in this order:\n\n1. The ``\'%\'`` character, which marks the start of the specifier.\n\n2. Mapping key (optional), consisting of a parenthesised sequence of\n   characters (for example, ``(somename)``).\n\n3. Conversion flags (optional), which affect the result of some\n   conversion types.\n\n4. Minimum field width (optional).  If specified as an ``\'*\'``\n   (asterisk), the actual width is read from the next element of the\n   tuple in *values*, and the object to convert comes after the\n   minimum field width and optional precision.\n\n5. Precision (optional), given as a ``\'.\'`` (dot) followed by the\n   precision.  If specified as ``\'*\'`` (an asterisk), the actual width\n   is read from the next element of the tuple in *values*, and the\n   value to convert comes after the precision.\n\n6. Length modifier (optional).\n\n7. Conversion type.\n\nWhen the right argument is a dictionary (or other mapping type), then\nthe formats in the string *must* include a parenthesised mapping key\ninto that dictionary inserted immediately after the ``\'%\'`` character.\nThe mapping key selects the value to be formatted from the mapping.\nFor example:\n\n>>> print(\'%(language)s has %(#)03d quote types.\' % \\\n...       {\'language\': "Python", "#": 2})\nPython has 002 quote types.\n\nIn this case no ``*`` specifiers may occur in a format (since they\nrequire a sequential parameter list).\n\nThe conversion flag characters are:\n\n+-----------+-----------------------------------------------------------------------+\n| Flag      | Meaning                                                               |\n+===========+=======================================================================+\n| ``\'#\'``   | The value conversion will use the "alternate form" (where defined     |\n|           | below).                                                               |\n+-----------+-----------------------------------------------------------------------+\n| ``\'0\'``   | The conversion will be zero padded for numeric values.                |\n+-----------+-----------------------------------------------------------------------+\n| ``\'-\'``   | The converted value is left adjusted (overrides the ``\'0\'``           |\n|           | conversion if both are given).                                        |\n+-----------+-----------------------------------------------------------------------+\n| ``\' \'``   | (a space) A blank should be left before a positive number (or empty   |\n|           | string) produced by a signed conversion.                              |\n+-----------+-----------------------------------------------------------------------+\n| ``\'+\'``   | A sign character (``\'+\'`` or ``\'-\'``) will precede the conversion     |\n|           | (overrides a "space" flag).                                           |\n+-----------+-----------------------------------------------------------------------+\n\nA length modifier (``h``, ``l``, or ``L``) may be present, but is\nignored as it is not necessary for Python -- so e.g. ``%ld`` is\nidentical to ``%d``.\n\nThe conversion types are:\n\n+--------------+-------------------------------------------------------+---------+\n| Conversion   | Meaning                                               | Notes   |\n+==============+=======================================================+=========+\n| ``\'d\'``      | Signed integer decimal.                               |         |\n+--------------+-------------------------------------------------------+---------+\n| ``\'i\'``      | Signed integer decimal.                               |         |\n+--------------+-------------------------------------------------------+---------+\n| ``\'o\'``      | Signed octal value.                                   | (1)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'u\'``      | Obsolete type -- it is identical to ``\'d\'``.          | (7)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'x\'``      | Signed hexadecimal (lowercase).                       | (2)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'X\'``      | Signed hexadecimal (uppercase).                       | (2)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'e\'``      | Floating point exponential format (lowercase).        | (3)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'E\'``      | Floating point exponential format (uppercase).        | (3)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'f\'``      | Floating point decimal format.                        | (3)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'F\'``      | Floating point decimal format.                        | (3)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'g\'``      | Floating point format. Uses lowercase exponential     | (4)     |\n|              | format if exponent is less than -4 or not less than   |         |\n|              | precision, decimal format otherwise.                  |         |\n+--------------+-------------------------------------------------------+---------+\n| ``\'G\'``      | Floating point format. Uses uppercase exponential     | (4)     |\n|              | format if exponent is less than -4 or not less than   |         |\n|              | precision, decimal format otherwise.                  |         |\n+--------------+-------------------------------------------------------+---------+\n| ``\'c\'``      | Single character (accepts integer or single character |         |\n|              | string).                                              |         |\n+--------------+-------------------------------------------------------+---------+\n| ``\'r\'``      | String (converts any python object using ``repr()``). | (5)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'s\'``      | String (converts any python object using ``str()``).  |         |\n+--------------+-------------------------------------------------------+---------+\n| ``\'%\'``      | No argument is converted, results in a ``\'%\'``        |         |\n|              | character in the result.                              |         |\n+--------------+-------------------------------------------------------+---------+\n\nNotes:\n\n1. The alternate form causes a leading zero (``\'0\'``) to be inserted\n   between left-hand padding and the formatting of the number if the\n   leading character of the result is not already a zero.\n\n2. The alternate form causes a leading ``\'0x\'`` or ``\'0X\'`` (depending\n   on whether the ``\'x\'`` or ``\'X\'`` format was used) to be inserted\n   between left-hand padding and the formatting of the number if the\n   leading character of the result is not already a zero.\n\n3. The alternate form causes the result to always contain a decimal\n   point, even if no digits follow it.\n\n   The precision determines the number of digits after the decimal\n   point and defaults to 6.\n\n4. The alternate form causes the result to always contain a decimal\n   point, and trailing zeroes are not removed as they would otherwise\n   be.\n\n   The precision determines the number of significant digits before\n   and after the decimal point and defaults to 6.\n\n5. The precision determines the maximal number of characters used.\n\n1. See **PEP 237**.\n\nSince Python strings have an explicit length, ``%s`` conversions do\nnot assume that ``\'\\0\'`` is the end of the string.\n\nChanged in version 3.1: ``%f`` conversions for numbers whose absolute\nvalue is over 1e50 are no longer replaced by ``%g`` conversions.\n\nAdditional string operations are defined in standard modules\n``string`` and ``re``.\n\n\nRange Type\n==========\n\nThe ``range`` type is an immutable sequence which is commonly used for\nlooping.  The advantage of the ``range`` type is that an ``range``\nobject will always take the same amount of memory, no matter the size\nof the range it represents.  There are no consistent performance\nadvantages.\n\nRange objects have very little behavior: they only support indexing,\niteration, and the ``len()`` function.\n\n\nMutable Sequence Types\n======================\n\nList and bytearray objects support additional operations that allow\nin-place modification of the object.  Other mutable sequence types\n(when added to the language) should also support these operations.\nStrings and tuples are immutable sequence types: such objects cannot\nbe modified once created. The following operations are defined on\nmutable sequence types (where *x* is an arbitrary object).\n\nNote that while lists allow their items to be of any type, bytearray\nobject "items" are all integers in the range 0 <= x < 256.\n\n+--------------------------------+----------------------------------+-----------------------+\n| Operation                      | Result                           | Notes                 |\n+================================+==================================+=======================+\n| ``s[i] = x``                   | item *i* of *s* is replaced by   |                       |\n|                                | *x*                              |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s[i:j] = t``                 | slice of *s* from *i* to *j* is  |                       |\n|                                | replaced by the contents of the  |                       |\n|                                | iterable *t*                     |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``del s[i:j]``                 | same as ``s[i:j] = []``          |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s[i:j:k] = t``               | the elements of ``s[i:j:k]`` are | (1)                   |\n|                                | replaced by those of *t*         |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``del s[i:j:k]``               | removes the elements of          |                       |\n|                                | ``s[i:j:k]`` from the list       |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.append(x)``                | same as ``s[len(s):len(s)] =     |                       |\n|                                | [x]``                            |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.extend(x)``                | same as ``s[len(s):len(s)] = x`` | (2)                   |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.count(x)``                 | return number of *i*\'s for which |                       |\n|                                | ``s[i] == x``                    |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.index(x[, i[, j]])``       | return smallest *k* such that    | (3)                   |\n|                                | ``s[k] == x`` and ``i <= k < j`` |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.insert(i, x)``             | same as ``s[i:i] = [x]``         | (4)                   |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.pop([i])``                 | same as ``x = s[i]; del s[i];    | (5)                   |\n|                                | return x``                       |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.remove(x)``                | same as ``del s[s.index(x)]``    | (3)                   |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.reverse()``                | reverses the items of *s* in     | (6)                   |\n|                                | place                            |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.sort([key[, reverse]])``   | sort the items of *s* in place   | (6), (7), (8)         |\n+--------------------------------+----------------------------------+-----------------------+\n\nNotes:\n\n1. *t* must have the same length as the slice it is replacing.\n\n2. *x* can be any iterable object.\n\n3. Raises ``ValueError`` when *x* is not found in *s*. When a negative\n   index is passed as the second or third parameter to the ``index()``\n   method, the sequence length is added, as for slice indices.  If it\n   is still negative, it is truncated to zero, as for slice indices.\n\n4. When a negative index is passed as the first parameter to the\n   ``insert()`` method, the sequence length is added, as for slice\n   indices.  If it is still negative, it is truncated to zero, as for\n   slice indices.\n\n5. The optional argument *i* defaults to ``-1``, so that by default\n   the last item is removed and returned.\n\n6. The ``sort()`` and ``reverse()`` methods modify the sequence in\n   place for economy of space when sorting or reversing a large\n   sequence.  To remind you that they operate by side effect, they\n   don\'t return the sorted or reversed sequence.\n\n7. The ``sort()`` method takes optional arguments for controlling the\n   comparisons.  Each must be specified as a keyword argument.\n\n   *key* specifies a function of one argument that is used to extract\n   a comparison key from each list element: ``key=str.lower``.  The\n   default value is ``None``.\n\n   *reverse* is a boolean value.  If set to ``True``, then the list\n   elements are sorted as if each comparison were reversed.\n\n   The ``sort()`` method is guaranteed to be stable.  A sort is stable\n   if it guarantees not to change the relative order of elements that\n   compare equal --- this is helpful for sorting in multiple passes\n   (for example, sort by department, then by salary grade).\n\n   While a list is being sorted, the effect of attempting to mutate,\n   or even inspect, the list is undefined.  The C implementation makes\n   the list appear empty for the duration, and raises ``ValueError``\n   if it can detect that the list has been mutated during a sort.\n\n8. ``sort()`` is not supported by ``bytearray`` objects.\n\n\nBytes and Byte Array Methods\n============================\n\nBytes and bytearray objects, being "strings of bytes", have all\nmethods found on strings, with the exception of ``encode()``,\n``format()`` and ``isidentifier()``, which do not make sense with\nthese types.  For converting the objects to strings, they have a\n``decode()`` method.\n\nWherever one of these methods needs to interpret the bytes as\ncharacters (e.g. the ``is...()`` methods), the ASCII character set is\nassumed.\n\nNote: The methods on bytes and bytearray objects don\'t accept strings as\n  their arguments, just as the methods on strings don\'t accept bytes\n  as their arguments.  For example, you have to write\n\n     a = "abc"\n     b = a.replace("a", "f")\n\n  and\n\n     a = b"abc"\n     b = a.replace(b"a", b"f")\n\nbytes.decode([encoding[, errors]])\nbytearray.decode([encoding[, errors]])\n\n   Return a string decoded from the given bytes.  Default encoding is\n   the current default string encoding.  *errors* may be given to set\n   a different error handling scheme.  The default for *errors* is\n   ``\'strict\'``, meaning that encoding errors raise a\n   ``UnicodeError``.  Other possible values are ``\'ignore\'``,\n   ``\'replace\'`` and any other name registered via\n   ``codecs.register_error()``, see section *Codec Base Classes*. For\n   a list of possible encodings, see section *Standard Encodings*.\n\nThe bytes and bytearray types have an additional class method:\n\nclassmethod bytes.fromhex(string)\nclassmethod bytearray.fromhex(string)\n\n   This ``bytes`` class method returns a bytes or bytearray object,\n   decoding the given string object.  The string must contain two\n   hexadecimal digits per byte, spaces are ignored.\n\n   >>> bytes.fromhex(\'f0 f1f2  \')\n   b\'\\xf0\\xf1\\xf2\'\n\nThe maketrans and translate methods differ in semantics from the\nversions available on strings:\n\nbytes.translate(table[, delete])\nbytearray.translate(table[, delete])\n\n   Return a copy of the bytes or bytearray object where all bytes\n   occurring in the optional argument *delete* are removed, and the\n   remaining bytes have been mapped through the given translation\n   table, which must be a bytes object of length 256.\n\n   You can use the ``bytes.maketrans()`` method to create a\n   translation table.\n\n   Set the *table* argument to ``None`` for translations that only\n   delete characters:\n\n      >>> b\'read this short text\'.translate(None, b\'aeiou\')\n      b\'rd ths shrt txt\'\n\nstatic bytes.maketrans(from, to)\nstatic bytearray.maketrans(from, to)\n\n   This static method returns a translation table usable for\n   ``bytes.translate()`` that will map each character in *from* into\n   the character at the same position in *to*; *from* and *to* must be\n   bytes objects and have the same length.\n\n   New in version 3.1.\n',
- 'typesseq-mutable': '\nMutable Sequence Types\n**********************\n\nList and bytearray objects support additional operations that allow\nin-place modification of the object.  Other mutable sequence types\n(when added to the language) should also support these operations.\nStrings and tuples are immutable sequence types: such objects cannot\nbe modified once created. The following operations are defined on\nmutable sequence types (where *x* is an arbitrary object).\n\nNote that while lists allow their items to be of any type, bytearray\nobject "items" are all integers in the range 0 <= x < 256.\n\n+--------------------------------+----------------------------------+-----------------------+\n| Operation                      | Result                           | Notes                 |\n+================================+==================================+=======================+\n| ``s[i] = x``                   | item *i* of *s* is replaced by   |                       |\n|                                | *x*                              |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s[i:j] = t``                 | slice of *s* from *i* to *j* is  |                       |\n|                                | replaced by the contents of the  |                       |\n|                                | iterable *t*                     |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``del s[i:j]``                 | same as ``s[i:j] = []``          |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s[i:j:k] = t``               | the elements of ``s[i:j:k]`` are | (1)                   |\n|                                | replaced by those of *t*         |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``del s[i:j:k]``               | removes the elements of          |                       |\n|                                | ``s[i:j:k]`` from the list       |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.append(x)``                | same as ``s[len(s):len(s)] =     |                       |\n|                                | [x]``                            |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.extend(x)``                | same as ``s[len(s):len(s)] = x`` | (2)                   |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.count(x)``                 | return number of *i*\'s for which |                       |\n|                                | ``s[i] == x``                    |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.index(x[, i[, j]])``       | return smallest *k* such that    | (3)                   |\n|                                | ``s[k] == x`` and ``i <= k < j`` |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.insert(i, x)``             | same as ``s[i:i] = [x]``         | (4)                   |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.pop([i])``                 | same as ``x = s[i]; del s[i];    | (5)                   |\n|                                | return x``                       |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.remove(x)``                | same as ``del s[s.index(x)]``    | (3)                   |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.reverse()``                | reverses the items of *s* in     | (6)                   |\n|                                | place                            |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.sort([key[, reverse]])``   | sort the items of *s* in place   | (6), (7), (8)         |\n+--------------------------------+----------------------------------+-----------------------+\n\nNotes:\n\n1. *t* must have the same length as the slice it is replacing.\n\n2. *x* can be any iterable object.\n\n3. Raises ``ValueError`` when *x* is not found in *s*. When a negative\n   index is passed as the second or third parameter to the ``index()``\n   method, the sequence length is added, as for slice indices.  If it\n   is still negative, it is truncated to zero, as for slice indices.\n\n4. When a negative index is passed as the first parameter to the\n   ``insert()`` method, the sequence length is added, as for slice\n   indices.  If it is still negative, it is truncated to zero, as for\n   slice indices.\n\n5. The optional argument *i* defaults to ``-1``, so that by default\n   the last item is removed and returned.\n\n6. The ``sort()`` and ``reverse()`` methods modify the sequence in\n   place for economy of space when sorting or reversing a large\n   sequence.  To remind you that they operate by side effect, they\n   don\'t return the sorted or reversed sequence.\n\n7. The ``sort()`` method takes optional arguments for controlling the\n   comparisons.  Each must be specified as a keyword argument.\n\n   *key* specifies a function of one argument that is used to extract\n   a comparison key from each list element: ``key=str.lower``.  The\n   default value is ``None``.\n\n   *reverse* is a boolean value.  If set to ``True``, then the list\n   elements are sorted as if each comparison were reversed.\n\n   The ``sort()`` method is guaranteed to be stable.  A sort is stable\n   if it guarantees not to change the relative order of elements that\n   compare equal --- this is helpful for sorting in multiple passes\n   (for example, sort by department, then by salary grade).\n\n   While a list is being sorted, the effect of attempting to mutate,\n   or even inspect, the list is undefined.  The C implementation makes\n   the list appear empty for the duration, and raises ``ValueError``\n   if it can detect that the list has been mutated during a sort.\n\n8. ``sort()`` is not supported by ``bytearray`` objects.\n',
+ 'typesseq': '\nSequence Types --- ``str``, ``bytes``, ``bytearray``, ``list``, ``tuple``, ``range``\n************************************************************************************\n\nThere are six sequence types: strings, byte sequences (``bytes``\nobjects), byte arrays (``bytearray`` objects), lists, tuples, and\nrange objects.  For other containers see the built in ``dict`` and\n``set`` classes, and the ``collections`` module.\n\nStrings contain Unicode characters.  Their literals are written in\nsingle or double quotes: ``\'xyzzy\'``, ``"frobozz"``.  See *String and\nBytes literals* for more about string literals.  In addition to the\nfunctionality described here, there are also string-specific methods\ndescribed in the *String Methods* section.\n\nBytes and bytearray objects contain single bytes -- the former is\nimmutable while the latter is a mutable sequence.  Bytes objects can\nbe constructed the constructor, ``bytes()``, and from literals; use a\n``b`` prefix with normal string syntax: ``b\'xyzzy\'``.  To construct\nbyte arrays, use the ``bytearray()`` function.\n\nWarning: While string objects are sequences of characters (represented by\n  strings of length 1), bytes and bytearray objects are sequences of\n  *integers* (between 0 and 255), representing the ASCII value of\n  single bytes.  That means that for a bytes or bytearray object *b*,\n  ``b[0]`` will be an integer, while ``b[0:1]`` will be a bytes or\n  bytearray object of length 1.  The representation of bytes objects\n  uses the literal format (``b\'...\'``) since it is generally more\n  useful than e.g. ``bytes([50, 19, 100])``.  You can always convert a\n  bytes object into a list of integers using ``list(b)``.Also, while\n  in previous Python versions, byte strings and Unicode strings could\n  be exchanged for each other rather freely (barring encoding issues),\n  strings and bytes are now completely separate concepts.  There\'s no\n  implicit en-/decoding if you pass an object of the wrong type.  A\n  string always compares unequal to a bytes or bytearray object.\n\nLists are constructed with square brackets, separating items with\ncommas: ``[a, b, c]``.  Tuples are constructed by the comma operator\n(not within square brackets), with or without enclosing parentheses,\nbut an empty tuple must have the enclosing parentheses, such as ``a,\nb, c`` or ``()``.  A single item tuple must have a trailing comma,\nsuch as ``(d,)``.\n\nObjects of type range are created using the ``range()`` function.\nThey don\'t support slicing, concatenation or repetition, and using\n``in``, ``not in``, ``min()`` or ``max()`` on them is inefficient.\n\nMost sequence types support the following operations.  The ``in`` and\n``not in`` operations have the same priorities as the comparison\noperations.  The ``+`` and ``*`` operations have the same priority as\nthe corresponding numeric operations. [3] Additional methods are\nprovided for *Mutable Sequence Types*.\n\nThis table lists the sequence operations sorted in ascending priority\n(operations in the same box have the same priority).  In the table,\n*s* and *t* are sequences of the same type; *n*, *i* and *j* are\nintegers:\n\n+--------------------+----------------------------------+------------+\n| Operation          | Result                           | Notes      |\n+====================+==================================+============+\n| ``x in s``         | ``True`` if an item of *s* is    | (1)        |\n|                    | equal to *x*, else ``False``     |            |\n+--------------------+----------------------------------+------------+\n| ``x not in s``     | ``False`` if an item of *s* is   | (1)        |\n|                    | equal to *x*, else ``True``      |            |\n+--------------------+----------------------------------+------------+\n| ``s + t``          | the concatenation of *s* and *t* | (6)        |\n+--------------------+----------------------------------+------------+\n| ``s * n, n * s``   | *n* shallow copies of *s*        | (2)        |\n|                    | concatenated                     |            |\n+--------------------+----------------------------------+------------+\n| ``s[i]``           | *i*\'th item of *s*, origin 0     | (3)        |\n+--------------------+----------------------------------+------------+\n| ``s[i:j]``         | slice of *s* from *i* to *j*     | (3)(4)     |\n+--------------------+----------------------------------+------------+\n| ``s[i:j:k]``       | slice of *s* from *i* to *j*     | (3)(5)     |\n|                    | with step *k*                    |            |\n+--------------------+----------------------------------+------------+\n| ``len(s)``         | length of *s*                    |            |\n+--------------------+----------------------------------+------------+\n| ``min(s)``         | smallest item of *s*             |            |\n+--------------------+----------------------------------+------------+\n| ``max(s)``         | largest item of *s*              |            |\n+--------------------+----------------------------------+------------+\n\nSequence types also support comparisons.  In particular, tuples and\nlists are compared lexicographically by comparing corresponding\nelements.  This means that to compare equal, every element must\ncompare equal and the two sequences must be of the same type and have\nthe same length.  (For full details see *Comparisons* in the language\nreference.)\n\nNotes:\n\n1. When *s* is a string object, the ``in`` and ``not in`` operations\n   act like a substring test.\n\n2. Values of *n* less than ``0`` are treated as ``0`` (which yields an\n   empty sequence of the same type as *s*).  Note also that the copies\n   are shallow; nested structures are not copied.  This often haunts\n   new Python programmers; consider:\n\n   >>> lists = [[]] * 3\n   >>> lists\n   [[], [], []]\n   >>> lists[0].append(3)\n   >>> lists\n   [[3], [3], [3]]\n\n   What has happened is that ``[[]]`` is a one-element list containing\n   an empty list, so all three elements of ``[[]] * 3`` are (pointers\n   to) this single empty list.  Modifying any of the elements of\n   ``lists`` modifies this single list. You can create a list of\n   different lists this way:\n\n   >>> lists = [[] for i in range(3)]\n   >>> lists[0].append(3)\n   >>> lists[1].append(5)\n   >>> lists[2].append(7)\n   >>> lists\n   [[3], [5], [7]]\n\n3. If *i* or *j* is negative, the index is relative to the end of the\n   string: ``len(s) + i`` or ``len(s) + j`` is substituted.  But note\n   that ``-0`` is still ``0``.\n\n4. The slice of *s* from *i* to *j* is defined as the sequence of\n   items with index *k* such that ``i <= k < j``.  If *i* or *j* is\n   greater than ``len(s)``, use ``len(s)``.  If *i* is omitted or\n   ``None``, use ``0``.  If *j* is omitted or ``None``, use\n   ``len(s)``.  If *i* is greater than or equal to *j*, the slice is\n   empty.\n\n5. The slice of *s* from *i* to *j* with step *k* is defined as the\n   sequence of items with index  ``x = i + n*k`` such that ``0 <= n <\n   (j-i)/k``.  In other words, the indices are ``i``, ``i+k``,\n   ``i+2*k``, ``i+3*k`` and so on, stopping when *j* is reached (but\n   never including *j*).  If *i* or *j* is greater than ``len(s)``,\n   use ``len(s)``.  If *i* or *j* are omitted or ``None``, they become\n   "end" values (which end depends on the sign of *k*).  Note, *k*\n   cannot be zero. If *k* is ``None``, it is treated like ``1``.\n\n6. **CPython implementation detail:** If *s* and *t* are both strings,\n   some Python implementations such as CPython can usually perform an\n   in-place optimization for assignments of the form ``s = s + t`` or\n   ``s += t``.  When applicable, this optimization makes quadratic\n   run-time much less likely.  This optimization is both version and\n   implementation dependent.  For performance sensitive code, it is\n   preferable to use the ``str.join()`` method which assures\n   consistent linear concatenation performance across versions and\n   implementations.\n\n\nString Methods\n==============\n\nString objects support the methods listed below.\n\nIn addition, Python\'s strings support the sequence type methods\ndescribed in the *Sequence Types --- str, bytes, bytearray, list,\ntuple, range* section. To output formatted strings, see the *String\nFormatting* section. Also, see the ``re`` module for string functions\nbased on regular expressions.\n\nstr.capitalize()\n\n   Return a copy of the string with its first character capitalized\n   and the rest lowercased.\n\nstr.center(width[, fillchar])\n\n   Return centered in a string of length *width*. Padding is done\n   using the specified *fillchar* (default is a space).\n\nstr.count(sub[, start[, end]])\n\n   Return the number of non-overlapping occurrences of substring *sub*\n   in the range [*start*, *end*].  Optional arguments *start* and\n   *end* are interpreted as in slice notation.\n\nstr.encode(encoding=sys.getdefaultencoding(), errors="strict")\n\n   Return an encoded version of the string as a bytes object.  Default\n   encoding is the current default string encoding.  *errors* may be\n   given to set a different error handling scheme.  The default for\n   *errors* is ``\'strict\'``, meaning that encoding errors raise a\n   ``UnicodeError``.  Other possible values are ``\'ignore\'``,\n   ``\'replace\'``, ``\'xmlcharrefreplace\'``, ``\'backslashreplace\'`` and\n   any other name registered via ``codecs.register_error()``, see\n   section *Codec Base Classes*. For a list of possible encodings, see\n   section *Standard Encodings*.\n\n   Changed in version 3.1: Added support for keyword arguments added.\n\nstr.endswith(suffix[, start[, end]])\n\n   Return ``True`` if the string ends with the specified *suffix*,\n   otherwise return ``False``.  *suffix* can also be a tuple of\n   suffixes to look for.  With optional *start*, test beginning at\n   that position.  With optional *end*, stop comparing at that\n   position.\n\nstr.expandtabs([tabsize])\n\n   Return a copy of the string where all tab characters are replaced\n   by one or more spaces, depending on the current column and the\n   given tab size.  The column number is reset to zero after each\n   newline occurring in the string. If *tabsize* is not given, a tab\n   size of ``8`` characters is assumed.  This doesn\'t understand other\n   non-printing characters or escape sequences.\n\nstr.find(sub[, start[, end]])\n\n   Return the lowest index in the string where substring *sub* is\n   found, such that *sub* is contained in the slice ``s[start:end]``.\n   Optional arguments *start* and *end* are interpreted as in slice\n   notation.  Return ``-1`` if *sub* is not found.\n\nstr.format(*args, **kwargs)\n\n   Perform a string formatting operation.  The string on which this\n   method is called can contain literal text or replacement fields\n   delimited by braces ``{}``.  Each replacement field contains either\n   the numeric index of a positional argument, or the name of a\n   keyword argument.  Returns a copy of the string where each\n   replacement field is replaced with the string value of the\n   corresponding argument.\n\n   >>> "The sum of 1 + 2 is {0}".format(1+2)\n   \'The sum of 1 + 2 is 3\'\n\n   See *Format String Syntax* for a description of the various\n   formatting options that can be specified in format strings.\n\nstr.index(sub[, start[, end]])\n\n   Like ``find()``, but raise ``ValueError`` when the substring is not\n   found.\n\nstr.isalnum()\n\n   Return true if all characters in the string are alphanumeric and\n   there is at least one character, false otherwise.\n\nstr.isalpha()\n\n   Return true if all characters in the string are alphabetic and\n   there is at least one character, false otherwise.\n\nstr.isdecimal()\n\n   Return true if all characters in the string are decimal characters\n   and there is at least one character, false otherwise. Decimal\n   characters include digit characters, and all characters that that\n   can be used to form decimal-radix numbers, e.g. U+0660, ARABIC-\n   INDIC DIGIT ZERO.\n\nstr.isdigit()\n\n   Return true if all characters in the string are digits and there is\n   at least one character, false otherwise.\n\nstr.isidentifier()\n\n   Return true if the string is a valid identifier according to the\n   language definition, section *Identifiers and keywords*.\n\nstr.islower()\n\n   Return true if all cased characters in the string are lowercase and\n   there is at least one cased character, false otherwise.\n\nstr.isnumeric()\n\n   Return true if all characters in the string are numeric characters,\n   and there is at least one character, false otherwise. Numeric\n   characters include digit characters, and all characters that have\n   the Unicode numeric value property, e.g. U+2155, VULGAR FRACTION\n   ONE FIFTH.\n\nstr.isprintable()\n\n   Return true if all characters in the string are printable or the\n   string is empty, false otherwise.  Nonprintable characters are\n   those characters defined in the Unicode character database as\n   "Other" or "Separator", excepting the ASCII space (0x20) which is\n   considered printable.  (Note that printable characters in this\n   context are those which should not be escaped when ``repr()`` is\n   invoked on a string.  It has no bearing on the handling of strings\n   written to ``sys.stdout`` or ``sys.stderr``.)\n\nstr.isspace()\n\n   Return true if there are only whitespace characters in the string\n   and there is at least one character, false otherwise.\n\nstr.istitle()\n\n   Return true if the string is a titlecased string and there is at\n   least one character, for example uppercase characters may only\n   follow uncased characters and lowercase characters only cased ones.\n   Return false otherwise.\n\nstr.isupper()\n\n   Return true if all cased characters in the string are uppercase and\n   there is at least one cased character, false otherwise.\n\nstr.join(iterable)\n\n   Return a string which is the concatenation of the strings in the\n   *iterable* *iterable*.  A ``TypeError`` will be raised if there are\n   any non-string values in *seq*, including ``bytes`` objects.  The\n   separator between elements is the string providing this method.\n\nstr.ljust(width[, fillchar])\n\n   Return the string left justified in a string of length *width*.\n   Padding is done using the specified *fillchar* (default is a\n   space).  The original string is returned if *width* is less than\n   ``len(s)``.\n\nstr.lower()\n\n   Return a copy of the string converted to lowercase.\n\nstr.lstrip([chars])\n\n   Return a copy of the string with leading characters removed.  The\n   *chars* argument is a string specifying the set of characters to be\n   removed.  If omitted or ``None``, the *chars* argument defaults to\n   removing whitespace.  The *chars* argument is not a prefix; rather,\n   all combinations of its values are stripped:\n\n   >>> \'   spacious   \'.lstrip()\n   \'spacious   \'\n   >>> \'www.example.com\'.lstrip(\'cmowz.\')\n   \'example.com\'\n\nstatic str.maketrans(x[, y[, z]])\n\n   This static method returns a translation table usable for\n   ``str.translate()``.\n\n   If there is only one argument, it must be a dictionary mapping\n   Unicode ordinals (integers) or characters (strings of length 1) to\n   Unicode ordinals, strings (of arbitrary lengths) or None.\n   Character keys will then be converted to ordinals.\n\n   If there are two arguments, they must be strings of equal length,\n   and in the resulting dictionary, each character in x will be mapped\n   to the character at the same position in y.  If there is a third\n   argument, it must be a string, whose characters will be mapped to\n   None in the result.\n\nstr.partition(sep)\n\n   Split the string at the first occurrence of *sep*, and return a\n   3-tuple containing the part before the separator, the separator\n   itself, and the part after the separator.  If the separator is not\n   found, return a 3-tuple containing the string itself, followed by\n   two empty strings.\n\nstr.replace(old, new[, count])\n\n   Return a copy of the string with all occurrences of substring *old*\n   replaced by *new*.  If the optional argument *count* is given, only\n   the first *count* occurrences are replaced.\n\nstr.rfind(sub[, start[, end]])\n\n   Return the highest index in the string where substring *sub* is\n   found, such that *sub* is contained within ``s[start:end]``.\n   Optional arguments *start* and *end* are interpreted as in slice\n   notation.  Return ``-1`` on failure.\n\nstr.rindex(sub[, start[, end]])\n\n   Like ``rfind()`` but raises ``ValueError`` when the substring *sub*\n   is not found.\n\nstr.rjust(width[, fillchar])\n\n   Return the string right justified in a string of length *width*.\n   Padding is done using the specified *fillchar* (default is a\n   space). The original string is returned if *width* is less than\n   ``len(s)``.\n\nstr.rpartition(sep)\n\n   Split the string at the last occurrence of *sep*, and return a\n   3-tuple containing the part before the separator, the separator\n   itself, and the part after the separator.  If the separator is not\n   found, return a 3-tuple containing two empty strings, followed by\n   the string itself.\n\nstr.rsplit([sep[, maxsplit]])\n\n   Return a list of the words in the string, using *sep* as the\n   delimiter string. If *maxsplit* is given, at most *maxsplit* splits\n   are done, the *rightmost* ones.  If *sep* is not specified or\n   ``None``, any whitespace string is a separator.  Except for\n   splitting from the right, ``rsplit()`` behaves like ``split()``\n   which is described in detail below.\n\nstr.rstrip([chars])\n\n   Return a copy of the string with trailing characters removed.  The\n   *chars* argument is a string specifying the set of characters to be\n   removed.  If omitted or ``None``, the *chars* argument defaults to\n   removing whitespace.  The *chars* argument is not a suffix; rather,\n   all combinations of its values are stripped:\n\n   >>> \'   spacious   \'.rstrip()\n   \'   spacious\'\n   >>> \'mississippi\'.rstrip(\'ipz\')\n   \'mississ\'\n\nstr.split([sep[, maxsplit]])\n\n   Return a list of the words in the string, using *sep* as the\n   delimiter string.  If *maxsplit* is given, at most *maxsplit*\n   splits are done (thus, the list will have at most ``maxsplit+1``\n   elements).  If *maxsplit* is not specified, then there is no limit\n   on the number of splits (all possible splits are made).\n\n   If *sep* is given, consecutive delimiters are not grouped together\n   and are deemed to delimit empty strings (for example,\n   ``\'1,,2\'.split(\',\')`` returns ``[\'1\', \'\', \'2\']``).  The *sep*\n   argument may consist of multiple characters (for example,\n   ``\'1<>2<>3\'.split(\'<>\')`` returns ``[\'1\', \'2\', \'3\']``). Splitting\n   an empty string with a specified separator returns ``[\'\']``.\n\n   If *sep* is not specified or is ``None``, a different splitting\n   algorithm is applied: runs of consecutive whitespace are regarded\n   as a single separator, and the result will contain no empty strings\n   at the start or end if the string has leading or trailing\n   whitespace.  Consequently, splitting an empty string or a string\n   consisting of just whitespace with a ``None`` separator returns\n   ``[]``.\n\n   For example, ``\' 1  2   3  \'.split()`` returns ``[\'1\', \'2\', \'3\']``,\n   and ``\'  1  2   3  \'.split(None, 1)`` returns ``[\'1\', \'2   3  \']``.\n\nstr.splitlines([keepends])\n\n   Return a list of the lines in the string, breaking at line\n   boundaries.  Line breaks are not included in the resulting list\n   unless *keepends* is given and true.\n\nstr.startswith(prefix[, start[, end]])\n\n   Return ``True`` if string starts with the *prefix*, otherwise\n   return ``False``. *prefix* can also be a tuple of prefixes to look\n   for.  With optional *start*, test string beginning at that\n   position.  With optional *end*, stop comparing string at that\n   position.\n\nstr.strip([chars])\n\n   Return a copy of the string with the leading and trailing\n   characters removed. The *chars* argument is a string specifying the\n   set of characters to be removed. If omitted or ``None``, the\n   *chars* argument defaults to removing whitespace. The *chars*\n   argument is not a prefix or suffix; rather, all combinations of its\n   values are stripped:\n\n   >>> \'   spacious   \'.strip()\n   \'spacious\'\n   >>> \'www.example.com\'.strip(\'cmowz.\')\n   \'example\'\n\nstr.swapcase()\n\n   Return a copy of the string with uppercase characters converted to\n   lowercase and vice versa.\n\nstr.title()\n\n   Return a titlecased version of the string where words start with an\n   uppercase character and the remaining characters are lowercase.\n\n   The algorithm uses a simple language-independent definition of a\n   word as groups of consecutive letters.  The definition works in\n   many contexts but it means that apostrophes in contractions and\n   possessives form word boundaries, which may not be the desired\n   result:\n\n      >>> "they\'re bill\'s friends from the UK".title()\n      "They\'Re Bill\'S Friends From The Uk"\n\n   A workaround for apostrophes can be constructed using regular\n   expressions:\n\n      >>> import re\n      >>> def titlecase(s):\n              return re.sub(r"[A-Za-z]+(\'[A-Za-z]+)?",\n                            lambda mo: mo.group(0)[0].upper() +\n                                       mo.group(0)[1:].lower(),\n                            s)\n\n      >>> titlecase("they\'re bill\'s friends.")\n      "They\'re Bill\'s Friends."\n\nstr.translate(map)\n\n   Return a copy of the *s* where all characters have been mapped\n   through the *map* which must be a dictionary of Unicode ordinals\n   (integers) to Unicode ordinals, strings or ``None``.  Unmapped\n   characters are left untouched. Characters mapped to ``None`` are\n   deleted.\n\n   You can use ``str.maketrans()`` to create a translation map from\n   character-to-character mappings in different formats.\n\n   You can use the ``maketrans()`` helper function in the ``string``\n   module to create a translation table. For string objects, set the\n   *table* argument to ``None`` for translations that only delete\n   characters:\n\n   Note: An even more flexible approach is to create a custom character\n     mapping codec using the ``codecs`` module (see\n     ``encodings.cp1251`` for an example).\n\nstr.upper()\n\n   Return a copy of the string converted to uppercase.\n\nstr.zfill(width)\n\n   Return the numeric string left filled with zeros in a string of\n   length *width*.  A sign prefix is handled correctly.  The original\n   string is returned if *width* is less than ``len(s)``.\n\n\nOld String Formatting Operations\n================================\n\nNote: The formatting operations described here are obsolete and may go\n  away in future versions of Python.  Use the new *String Formatting*\n  in new code.\n\nString objects have one unique built-in operation: the ``%`` operator\n(modulo). This is also known as the string *formatting* or\n*interpolation* operator. Given ``format % values`` (where *format* is\na string), ``%`` conversion specifications in *format* are replaced\nwith zero or more elements of *values*. The effect is similar to the\nusing ``sprintf()`` in the C language.\n\nIf *format* requires a single argument, *values* may be a single non-\ntuple object. [4]  Otherwise, *values* must be a tuple with exactly\nthe number of items specified by the format string, or a single\nmapping object (for example, a dictionary).\n\nA conversion specifier contains two or more characters and has the\nfollowing components, which must occur in this order:\n\n1. The ``\'%\'`` character, which marks the start of the specifier.\n\n2. Mapping key (optional), consisting of a parenthesised sequence of\n   characters (for example, ``(somename)``).\n\n3. Conversion flags (optional), which affect the result of some\n   conversion types.\n\n4. Minimum field width (optional).  If specified as an ``\'*\'``\n   (asterisk), the actual width is read from the next element of the\n   tuple in *values*, and the object to convert comes after the\n   minimum field width and optional precision.\n\n5. Precision (optional), given as a ``\'.\'`` (dot) followed by the\n   precision.  If specified as ``\'*\'`` (an asterisk), the actual width\n   is read from the next element of the tuple in *values*, and the\n   value to convert comes after the precision.\n\n6. Length modifier (optional).\n\n7. Conversion type.\n\nWhen the right argument is a dictionary (or other mapping type), then\nthe formats in the string *must* include a parenthesised mapping key\ninto that dictionary inserted immediately after the ``\'%\'`` character.\nThe mapping key selects the value to be formatted from the mapping.\nFor example:\n\n>>> print(\'%(language)s has %(#)03d quote types.\' % \\\n...       {\'language\': "Python", "#": 2})\nPython has 002 quote types.\n\nIn this case no ``*`` specifiers may occur in a format (since they\nrequire a sequential parameter list).\n\nThe conversion flag characters are:\n\n+-----------+-----------------------------------------------------------------------+\n| Flag      | Meaning                                                               |\n+===========+=======================================================================+\n| ``\'#\'``   | The value conversion will use the "alternate form" (where defined     |\n|           | below).                                                               |\n+-----------+-----------------------------------------------------------------------+\n| ``\'0\'``   | The conversion will be zero padded for numeric values.                |\n+-----------+-----------------------------------------------------------------------+\n| ``\'-\'``   | The converted value is left adjusted (overrides the ``\'0\'``           |\n|           | conversion if both are given).                                        |\n+-----------+-----------------------------------------------------------------------+\n| ``\' \'``   | (a space) A blank should be left before a positive number (or empty   |\n|           | string) produced by a signed conversion.                              |\n+-----------+-----------------------------------------------------------------------+\n| ``\'+\'``   | A sign character (``\'+\'`` or ``\'-\'``) will precede the conversion     |\n|           | (overrides a "space" flag).                                           |\n+-----------+-----------------------------------------------------------------------+\n\nA length modifier (``h``, ``l``, or ``L``) may be present, but is\nignored as it is not necessary for Python -- so e.g. ``%ld`` is\nidentical to ``%d``.\n\nThe conversion types are:\n\n+--------------+-------------------------------------------------------+---------+\n| Conversion   | Meaning                                               | Notes   |\n+==============+=======================================================+=========+\n| ``\'d\'``      | Signed integer decimal.                               |         |\n+--------------+-------------------------------------------------------+---------+\n| ``\'i\'``      | Signed integer decimal.                               |         |\n+--------------+-------------------------------------------------------+---------+\n| ``\'o\'``      | Signed octal value.                                   | (1)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'u\'``      | Obsolete type -- it is identical to ``\'d\'``.          | (7)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'x\'``      | Signed hexadecimal (lowercase).                       | (2)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'X\'``      | Signed hexadecimal (uppercase).                       | (2)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'e\'``      | Floating point exponential format (lowercase).        | (3)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'E\'``      | Floating point exponential format (uppercase).        | (3)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'f\'``      | Floating point decimal format.                        | (3)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'F\'``      | Floating point decimal format.                        | (3)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'g\'``      | Floating point format. Uses lowercase exponential     | (4)     |\n|              | format if exponent is less than -4 or not less than   |         |\n|              | precision, decimal format otherwise.                  |         |\n+--------------+-------------------------------------------------------+---------+\n| ``\'G\'``      | Floating point format. Uses uppercase exponential     | (4)     |\n|              | format if exponent is less than -4 or not less than   |         |\n|              | precision, decimal format otherwise.                  |         |\n+--------------+-------------------------------------------------------+---------+\n| ``\'c\'``      | Single character (accepts integer or single character |         |\n|              | string).                                              |         |\n+--------------+-------------------------------------------------------+---------+\n| ``\'r\'``      | String (converts any Python object using ``repr()``). | (5)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'s\'``      | String (converts any Python object using ``str()``).  |         |\n+--------------+-------------------------------------------------------+---------+\n| ``\'%\'``      | No argument is converted, results in a ``\'%\'``        |         |\n|              | character in the result.                              |         |\n+--------------+-------------------------------------------------------+---------+\n\nNotes:\n\n1. The alternate form causes a leading zero (``\'0\'``) to be inserted\n   between left-hand padding and the formatting of the number if the\n   leading character of the result is not already a zero.\n\n2. The alternate form causes a leading ``\'0x\'`` or ``\'0X\'`` (depending\n   on whether the ``\'x\'`` or ``\'X\'`` format was used) to be inserted\n   between left-hand padding and the formatting of the number if the\n   leading character of the result is not already a zero.\n\n3. The alternate form causes the result to always contain a decimal\n   point, even if no digits follow it.\n\n   The precision determines the number of digits after the decimal\n   point and defaults to 6.\n\n4. The alternate form causes the result to always contain a decimal\n   point, and trailing zeroes are not removed as they would otherwise\n   be.\n\n   The precision determines the number of significant digits before\n   and after the decimal point and defaults to 6.\n\n5. The precision determines the maximal number of characters used.\n\n1. See **PEP 237**.\n\nSince Python strings have an explicit length, ``%s`` conversions do\nnot assume that ``\'\\0\'`` is the end of the string.\n\nChanged in version 3.1: ``%f`` conversions for numbers whose absolute\nvalue is over 1e50 are no longer replaced by ``%g`` conversions.\n\nAdditional string operations are defined in standard modules\n``string`` and ``re``.\n\n\nRange Type\n==========\n\nThe ``range`` type is an immutable sequence which is commonly used for\nlooping.  The advantage of the ``range`` type is that an ``range``\nobject will always take the same amount of memory, no matter the size\nof the range it represents.  There are no consistent performance\nadvantages.\n\nRange objects have very little behavior: they only support indexing,\niteration, and the ``len()`` function.\n\n\nMutable Sequence Types\n======================\n\nList and bytearray objects support additional operations that allow\nin-place modification of the object.  Other mutable sequence types\n(when added to the language) should also support these operations.\nStrings and tuples are immutable sequence types: such objects cannot\nbe modified once created. The following operations are defined on\nmutable sequence types (where *x* is an arbitrary object).\n\nNote that while lists allow their items to be of any type, bytearray\nobject "items" are all integers in the range 0 <= x < 256.\n\n+--------------------------------+----------------------------------+-----------------------+\n| Operation                      | Result                           | Notes                 |\n+================================+==================================+=======================+\n| ``s[i] = x``                   | item *i* of *s* is replaced by   |                       |\n|                                | *x*                              |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s[i:j] = t``                 | slice of *s* from *i* to *j* is  |                       |\n|                                | replaced by the contents of the  |                       |\n|                                | iterable *t*                     |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``del s[i:j]``                 | same as ``s[i:j] = []``          |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s[i:j:k] = t``               | the elements of ``s[i:j:k]`` are | (1)                   |\n|                                | replaced by those of *t*         |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``del s[i:j:k]``               | removes the elements of          |                       |\n|                                | ``s[i:j:k]`` from the list       |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.append(x)``                | same as ``s[len(s):len(s)] =     |                       |\n|                                | [x]``                            |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.extend(x)``                | same as ``s[len(s):len(s)] = x`` | (2)                   |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.count(x)``                 | return number of *i*\'s for which |                       |\n|                                | ``s[i] == x``                    |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.index(x[, i[, j]])``       | return smallest *k* such that    | (3)                   |\n|                                | ``s[k] == x`` and ``i <= k < j`` |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.insert(i, x)``             | same as ``s[i:i] = [x]``         | (4)                   |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.pop([i])``                 | same as ``x = s[i]; del s[i];    | (5)                   |\n|                                | return x``                       |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.remove(x)``                | same as ``del s[s.index(x)]``    | (3)                   |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.reverse()``                | reverses the items of *s* in     | (6)                   |\n|                                | place                            |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.sort([key[, reverse]])``   | sort the items of *s* in place   | (6), (7), (8)         |\n+--------------------------------+----------------------------------+-----------------------+\n\nNotes:\n\n1. *t* must have the same length as the slice it is replacing.\n\n2. *x* can be any iterable object.\n\n3. Raises ``ValueError`` when *x* is not found in *s*. When a negative\n   index is passed as the second or third parameter to the ``index()``\n   method, the sequence length is added, as for slice indices.  If it\n   is still negative, it is truncated to zero, as for slice indices.\n\n4. When a negative index is passed as the first parameter to the\n   ``insert()`` method, the sequence length is added, as for slice\n   indices.  If it is still negative, it is truncated to zero, as for\n   slice indices.\n\n5. The optional argument *i* defaults to ``-1``, so that by default\n   the last item is removed and returned.\n\n6. The ``sort()`` and ``reverse()`` methods modify the sequence in\n   place for economy of space when sorting or reversing a large\n   sequence.  To remind you that they operate by side effect, they\n   don\'t return the sorted or reversed sequence.\n\n7. The ``sort()`` method takes optional arguments for controlling the\n   comparisons.  Each must be specified as a keyword argument.\n\n   *key* specifies a function of one argument that is used to extract\n   a comparison key from each list element: ``key=str.lower``.  The\n   default value is ``None``. Use ``functools.cmp_to_key()`` to\n   convert an old-style *cmp* function to a *key* function.\n\n   *reverse* is a boolean value.  If set to ``True``, then the list\n   elements are sorted as if each comparison were reversed.\n\n   The ``sort()`` method is guaranteed to be stable.  A sort is stable\n   if it guarantees not to change the relative order of elements that\n   compare equal --- this is helpful for sorting in multiple passes\n   (for example, sort by department, then by salary grade).\n\n   **CPython implementation detail:** While a list is being sorted,\n   the effect of attempting to mutate, or even inspect, the list is\n   undefined.  The C implementation of Python makes the list appear\n   empty for the duration, and raises ``ValueError`` if it can detect\n   that the list has been mutated during a sort.\n\n8. ``sort()`` is not supported by ``bytearray`` objects.\n\n\nBytes and Byte Array Methods\n============================\n\nBytes and bytearray objects, being "strings of bytes", have all\nmethods found on strings, with the exception of ``encode()``,\n``format()`` and ``isidentifier()``, which do not make sense with\nthese types.  For converting the objects to strings, they have a\n``decode()`` method.\n\nWherever one of these methods needs to interpret the bytes as\ncharacters (e.g. the ``is...()`` methods), the ASCII character set is\nassumed.\n\nNote: The methods on bytes and bytearray objects don\'t accept strings as\n  their arguments, just as the methods on strings don\'t accept bytes\n  as their arguments.  For example, you have to write\n\n     a = "abc"\n     b = a.replace("a", "f")\n\n  and\n\n     a = b"abc"\n     b = a.replace(b"a", b"f")\n\nbytes.decode(encoding=sys.getdefaultencoding(), errors="strict")\nbytearray.decode(encoding=sys.getdefaultencoding(), errors="strict")\n\n   Return a string decoded from the given bytes.  Default encoding is\n   the current default string encoding.  *errors* may be given to set\n   a different error handling scheme.  The default for *errors* is\n   ``\'strict\'``, meaning that encoding errors raise a\n   ``UnicodeError``.  Other possible values are ``\'ignore\'``,\n   ``\'replace\'`` and any other name registered via\n   ``codecs.register_error()``, see section *Codec Base Classes*. For\n   a list of possible encodings, see section *Standard Encodings*.\n\n   Changed in version 3.1: Added support for keyword arguments.\n\nThe bytes and bytearray types have an additional class method:\n\nclassmethod bytes.fromhex(string)\nclassmethod bytearray.fromhex(string)\n\n   This ``bytes`` class method returns a bytes or bytearray object,\n   decoding the given string object.  The string must contain two\n   hexadecimal digits per byte, spaces are ignored.\n\n   >>> bytes.fromhex(\'f0 f1f2  \')\n   b\'\\xf0\\xf1\\xf2\'\n\nThe maketrans and translate methods differ in semantics from the\nversions available on strings:\n\nbytes.translate(table[, delete])\nbytearray.translate(table[, delete])\n\n   Return a copy of the bytes or bytearray object where all bytes\n   occurring in the optional argument *delete* are removed, and the\n   remaining bytes have been mapped through the given translation\n   table, which must be a bytes object of length 256.\n\n   You can use the ``bytes.maketrans()`` method to create a\n   translation table.\n\n   Set the *table* argument to ``None`` for translations that only\n   delete characters:\n\n      >>> b\'read this short text\'.translate(None, b\'aeiou\')\n      b\'rd ths shrt txt\'\n\nstatic bytes.maketrans(from, to)\nstatic bytearray.maketrans(from, to)\n\n   This static method returns a translation table usable for\n   ``bytes.translate()`` that will map each character in *from* into\n   the character at the same position in *to*; *from* and *to* must be\n   bytes objects and have the same length.\n\n   New in version 3.1.\n',
+ 'typesseq-mutable': '\nMutable Sequence Types\n**********************\n\nList and bytearray objects support additional operations that allow\nin-place modification of the object.  Other mutable sequence types\n(when added to the language) should also support these operations.\nStrings and tuples are immutable sequence types: such objects cannot\nbe modified once created. The following operations are defined on\nmutable sequence types (where *x* is an arbitrary object).\n\nNote that while lists allow their items to be of any type, bytearray\nobject "items" are all integers in the range 0 <= x < 256.\n\n+--------------------------------+----------------------------------+-----------------------+\n| Operation                      | Result                           | Notes                 |\n+================================+==================================+=======================+\n| ``s[i] = x``                   | item *i* of *s* is replaced by   |                       |\n|                                | *x*                              |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s[i:j] = t``                 | slice of *s* from *i* to *j* is  |                       |\n|                                | replaced by the contents of the  |                       |\n|                                | iterable *t*                     |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``del s[i:j]``                 | same as ``s[i:j] = []``          |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s[i:j:k] = t``               | the elements of ``s[i:j:k]`` are | (1)                   |\n|                                | replaced by those of *t*         |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``del s[i:j:k]``               | removes the elements of          |                       |\n|                                | ``s[i:j:k]`` from the list       |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.append(x)``                | same as ``s[len(s):len(s)] =     |                       |\n|                                | [x]``                            |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.extend(x)``                | same as ``s[len(s):len(s)] = x`` | (2)                   |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.count(x)``                 | return number of *i*\'s for which |                       |\n|                                | ``s[i] == x``                    |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.index(x[, i[, j]])``       | return smallest *k* such that    | (3)                   |\n|                                | ``s[k] == x`` and ``i <= k < j`` |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.insert(i, x)``             | same as ``s[i:i] = [x]``         | (4)                   |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.pop([i])``                 | same as ``x = s[i]; del s[i];    | (5)                   |\n|                                | return x``                       |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.remove(x)``                | same as ``del s[s.index(x)]``    | (3)                   |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.reverse()``                | reverses the items of *s* in     | (6)                   |\n|                                | place                            |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.sort([key[, reverse]])``   | sort the items of *s* in place   | (6), (7), (8)         |\n+--------------------------------+----------------------------------+-----------------------+\n\nNotes:\n\n1. *t* must have the same length as the slice it is replacing.\n\n2. *x* can be any iterable object.\n\n3. Raises ``ValueError`` when *x* is not found in *s*. When a negative\n   index is passed as the second or third parameter to the ``index()``\n   method, the sequence length is added, as for slice indices.  If it\n   is still negative, it is truncated to zero, as for slice indices.\n\n4. When a negative index is passed as the first parameter to the\n   ``insert()`` method, the sequence length is added, as for slice\n   indices.  If it is still negative, it is truncated to zero, as for\n   slice indices.\n\n5. The optional argument *i* defaults to ``-1``, so that by default\n   the last item is removed and returned.\n\n6. The ``sort()`` and ``reverse()`` methods modify the sequence in\n   place for economy of space when sorting or reversing a large\n   sequence.  To remind you that they operate by side effect, they\n   don\'t return the sorted or reversed sequence.\n\n7. The ``sort()`` method takes optional arguments for controlling the\n   comparisons.  Each must be specified as a keyword argument.\n\n   *key* specifies a function of one argument that is used to extract\n   a comparison key from each list element: ``key=str.lower``.  The\n   default value is ``None``. Use ``functools.cmp_to_key()`` to\n   convert an old-style *cmp* function to a *key* function.\n\n   *reverse* is a boolean value.  If set to ``True``, then the list\n   elements are sorted as if each comparison were reversed.\n\n   The ``sort()`` method is guaranteed to be stable.  A sort is stable\n   if it guarantees not to change the relative order of elements that\n   compare equal --- this is helpful for sorting in multiple passes\n   (for example, sort by department, then by salary grade).\n\n   **CPython implementation detail:** While a list is being sorted,\n   the effect of attempting to mutate, or even inspect, the list is\n   undefined.  The C implementation of Python makes the list appear\n   empty for the duration, and raises ``ValueError`` if it can detect\n   that the list has been mutated during a sort.\n\n8. ``sort()`` is not supported by ``bytearray`` objects.\n',
  'unary': '\nUnary arithmetic and bitwise operations\n***************************************\n\nAll unary arithmetic and bitwise operations have the same priority:\n\n   u_expr ::= power | "-" u_expr | "+" u_expr | "~" u_expr\n\nThe unary ``-`` (minus) operator yields the negation of its numeric\nargument.\n\nThe unary ``+`` (plus) operator yields its numeric argument unchanged.\n\nThe unary ``~`` (invert) operator yields the bitwise inversion of its\ninteger argument.  The bitwise inversion of ``x`` is defined as\n``-(x+1)``.  It only applies to integral numbers.\n\nIn all three cases, if the argument does not have the proper type, a\n``TypeError`` exception is raised.\n',
  'while': '\nThe ``while`` statement\n***********************\n\nThe ``while`` statement is used for repeated execution as long as an\nexpression is true:\n\n   while_stmt ::= "while" expression ":" suite\n                  ["else" ":" suite]\n\nThis repeatedly tests the expression and, if it is true, executes the\nfirst suite; if the expression is false (which may be the first time\nit is tested) the suite of the ``else`` clause, if present, is\nexecuted and the loop terminates.\n\nA ``break`` statement executed in the first suite terminates the loop\nwithout executing the ``else`` clause\'s suite.  A ``continue``\nstatement executed in the first suite skips the rest of the suite and\ngoes back to testing the expression.\n',
- 'with': '\nThe ``with`` statement\n**********************\n\nThe ``with`` statement is used to wrap the execution of a block with\nmethods defined by a context manager (see section *With Statement\nContext Managers*). This allows common\n``try``...``except``...``finally`` usage patterns to be encapsulated\nfor convenient reuse.\n\n   with_stmt ::= "with" with_item ("," with_item)* ":" suite\n   with_item ::= expression ["as" target]\n\nThe execution of the ``with`` statement with one "item" proceeds as\nfollows:\n\n1. The context expression is evaluated to obtain a context manager.\n\n2. The context manager\'s ``__enter__()`` method is invoked.\n\n3. If a target was included in the ``with`` statement, the return\n   value from ``__enter__()`` is assigned to it.\n\n   Note: The ``with`` statement guarantees that if the ``__enter__()``\n     method returns without an error, then ``__exit__()`` will always\n     be called.  Thus, if an error occurs during the assignment to the\n     target list, it will be treated the same as an error occurring\n     within the suite would be.  See step 5 below.\n\n4. The suite is executed.\n\n5. The context manager\'s ``__exit__()`` method is invoked.  If an\n   exception caused the suite to be exited, its type, value, and\n   traceback are passed as arguments to ``__exit__()``. Otherwise,\n   three ``None`` arguments are supplied.\n\n   If the suite was exited due to an exception, and the return value\n   from the ``__exit__()`` method was false, the exception is\n   reraised.  If the return value was true, the exception is\n   suppressed, and execution continues with the statement following\n   the ``with`` statement.\n\n   If the suite was exited for any reason other than an exception, the\n   return value from ``__exit__()`` is ignored, and execution proceeds\n   at the normal location for the kind of exit that was taken.\n\nWith more than one item, the context managers are processed as if\nmultiple ``with`` statements were nested:\n\n   with A() as a, B() as b:\n       suite\n\nis equivalent to\n\n   with A() as a:\n       with B() as b:\n           suite\n\nChanged in version 3.1: Support for multiple context expressions.\n\nSee also:\n\n   **PEP 0343** - The "with" statement\n      The specification, background, and examples for the Python\n      ``with`` statement.\n',
- 'yield': '\nThe ``yield`` statement\n***********************\n\n   yield_stmt ::= yield_expression\n\nThe ``yield`` statement is only used when defining a generator\nfunction, and is only used in the body of the generator function.\nUsing a ``yield`` statement in a function definition is sufficient to\ncause that definition to create a generator function instead of a\nnormal function. When a generator function is called, it returns an\niterator known as a generator iterator, or more commonly, a generator.\nThe body of the generator function is executed by calling the\n``next()`` function on the generator repeatedly until it raises an\nexception.\n\nWhen a ``yield`` statement is executed, the state of the generator is\nfrozen and the value of **expression_list** is returned to\n``next()``\'s caller.  By "frozen" we mean that all local state is\nretained, including the current bindings of local variables, the\ninstruction pointer, and the internal evaluation stack: enough\ninformation is saved so that the next time ``next()`` is invoked, the\nfunction can proceed exactly as if the ``yield`` statement were just\nanother external call.\n\nThe ``yield`` statement is allowed in the ``try`` clause of a ``try``\n...  ``finally`` construct.  If the generator is not resumed before it\nis finalized (by reaching a zero reference count or by being garbage\ncollected), the generator-iterator\'s ``close()`` method will be\ncalled, allowing any pending ``finally`` clauses to execute.\n\nSee also:\n\n   **PEP 0255** - Simple Generators\n      The proposal for adding generators and the ``yield`` statement\n      to Python.\n\n   **PEP 0342** - Coroutines via Enhanced Generators\n      The proposal that, among other generator enhancements, proposed\n      allowing ``yield`` to appear inside a ``try`` ... ``finally``\n      block.\n'}
+ 'with': '\nThe ``with`` statement\n**********************\n\nThe ``with`` statement is used to wrap the execution of a block with\nmethods defined by a context manager (see section *With Statement\nContext Managers*). This allows common\n``try``...``except``...``finally`` usage patterns to be encapsulated\nfor convenient reuse.\n\n   with_stmt ::= "with" with_item ("," with_item)* ":" suite\n   with_item ::= expression ["as" target]\n\nThe execution of the ``with`` statement with one "item" proceeds as\nfollows:\n\n1. The context expression is evaluated to obtain a context manager.\n\n2. The context manager\'s ``__exit__()`` is loaded for later use.\n\n3. The context manager\'s ``__enter__()`` method is invoked.\n\n4. If a target was included in the ``with`` statement, the return\n   value from ``__enter__()`` is assigned to it.\n\n   Note: The ``with`` statement guarantees that if the ``__enter__()``\n     method returns without an error, then ``__exit__()`` will always\n     be called. Thus, if an error occurs during the assignment to the\n     target list, it will be treated the same as an error occurring\n     within the suite would be. See step 6 below.\n\n5. The suite is executed.\n\n6. The context manager\'s ``__exit__()`` method is invoked.  If an\n   exception caused the suite to be exited, its type, value, and\n   traceback are passed as arguments to ``__exit__()``. Otherwise,\n   three ``None`` arguments are supplied.\n\n   If the suite was exited due to an exception, and the return value\n   from the ``__exit__()`` method was false, the exception is\n   reraised.  If the return value was true, the exception is\n   suppressed, and execution continues with the statement following\n   the ``with`` statement.\n\n   If the suite was exited for any reason other than an exception, the\n   return value from ``__exit__()`` is ignored, and execution proceeds\n   at the normal location for the kind of exit that was taken.\n\nWith more than one item, the context managers are processed as if\nmultiple ``with`` statements were nested:\n\n   with A() as a, B() as b:\n       suite\n\nis equivalent to\n\n   with A() as a:\n       with B() as b:\n           suite\n\nChanged in version 3.1: Support for multiple context expressions.\n\nSee also:\n\n   **PEP 0343** - The "with" statement\n      The specification, background, and examples for the Python\n      ``with`` statement.\n',
+ 'yield': '\nThe ``yield`` statement\n***********************\n\n   yield_stmt ::= yield_expression\n\nThe ``yield`` statement is only used when defining a generator\nfunction, and is only used in the body of the generator function.\nUsing a ``yield`` statement in a function definition is sufficient to\ncause that definition to create a generator function instead of a\nnormal function. When a generator function is called, it returns an\niterator known as a generator iterator, or more commonly, a generator.\nThe body of the generator function is executed by calling the\n``next()`` function on the generator repeatedly until it raises an\nexception.\n\nWhen a ``yield`` statement is executed, the state of the generator is\nfrozen and the value of ``expression_list`` is returned to\n``next()``\'s caller.  By "frozen" we mean that all local state is\nretained, including the current bindings of local variables, the\ninstruction pointer, and the internal evaluation stack: enough\ninformation is saved so that the next time ``next()`` is invoked, the\nfunction can proceed exactly as if the ``yield`` statement were just\nanother external call.\n\nThe ``yield`` statement is allowed in the ``try`` clause of a ``try``\n...  ``finally`` construct.  If the generator is not resumed before it\nis finalized (by reaching a zero reference count or by being garbage\ncollected), the generator-iterator\'s ``close()`` method will be\ncalled, allowing any pending ``finally`` clauses to execute.\n\nSee also:\n\n   **PEP 0255** - Simple Generators\n      The proposal for adding generators and the ``yield`` statement\n      to Python.\n\n   **PEP 0342** - Coroutines via Enhanced Generators\n      The proposal that, among other generator enhancements, proposed\n      allowing ``yield`` to appear inside a ``try`` ... ``finally``\n      block.\n'}

From python-checkins at python.org  Sat Jul 31 10:56:11 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 10:56:11 +0200 (CEST)
Subject: [Python-checkins] r83318 - in python/branches/py3k:
	Include/patchlevel.h Lib/distutils/__init__.py
	Lib/idlelib/idlever.py Misc/NEWS Misc/RPM/python-3.2.spec README
Message-ID: <20100731085611.55C45F36D@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 10:56:11 2010
New Revision: 83318

Log:
Bump versions and review NEWS file.

Modified:
   python/branches/py3k/Include/patchlevel.h
   python/branches/py3k/Lib/distutils/__init__.py
   python/branches/py3k/Lib/idlelib/idlever.py
   python/branches/py3k/Misc/NEWS
   python/branches/py3k/Misc/RPM/python-3.2.spec
   python/branches/py3k/README

Modified: python/branches/py3k/Include/patchlevel.h
==============================================================================
--- python/branches/py3k/Include/patchlevel.h	(original)
+++ python/branches/py3k/Include/patchlevel.h	Sat Jul 31 10:56:11 2010
@@ -20,10 +20,10 @@
 #define PY_MINOR_VERSION	2
 #define PY_MICRO_VERSION	0
 #define PY_RELEASE_LEVEL	PY_RELEASE_LEVEL_ALPHA
-#define PY_RELEASE_SERIAL	0
+#define PY_RELEASE_SERIAL	1
 
 /* Version as a string */
-#define PY_VERSION      	"3.2a0"
+#define PY_VERSION      	"3.2a1"
 /*--end constants--*/
 
 /* Subversion Revision number of this file (not of the repository) */

Modified: python/branches/py3k/Lib/distutils/__init__.py
==============================================================================
--- python/branches/py3k/Lib/distutils/__init__.py	(original)
+++ python/branches/py3k/Lib/distutils/__init__.py	Sat Jul 31 10:56:11 2010
@@ -15,5 +15,5 @@
 # Updated automatically by the Python release process.
 #
 #--start constants--
-__version__ = "3.2a0"
+__version__ = "3.2a1"
 #--end constants--

Modified: python/branches/py3k/Lib/idlelib/idlever.py
==============================================================================
--- python/branches/py3k/Lib/idlelib/idlever.py	(original)
+++ python/branches/py3k/Lib/idlelib/idlever.py	Sat Jul 31 10:56:11 2010
@@ -1 +1 @@
-IDLE_VERSION = "3.2a0"
+IDLE_VERSION = "3.2a1"

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sat Jul 31 10:56:11 2010
@@ -7,7 +7,7 @@
 What's New in Python 3.2 Alpha 1?
 =================================
 
-*Release date: XX-XXX-XXX*
+*Release date: 01-Aug-2010*
 
 Core and Builtins
 -----------------
@@ -20,16 +20,15 @@
 - Issue #8413: structsequence now subclasses tuple.
 
 - Issue #8271: during the decoding of an invalid UTF-8 byte sequence, only the
-  start byte and the continuation byte(s) are now considered invalid, instead
-  of the number of bytes specified by the start byte.
-  E.g.: '\xf1\x80AB'.decode('utf-8', 'replace') now returns u'\ufffdAB' and
-  replaces with U+FFFD only the start byte ('\xf1') and the continuation byte
-  ('\x80') even if '\xf1' is the start byte of a 4-bytes sequence.
-  Previous versions returned a single u'\ufffd'.
-
-- Issue #9011: A negated imaginary literal (e.g., "-7j") now has real
-  part -0.0 rather than 0.0.  So "-7j" is now exactly equivalent to
-  "-(7j)".
+  start byte and the continuation byte(s) are now considered invalid, instead of
+  the number of bytes specified by the start byte.  E.g.:
+  '\xf1\x80AB'.decode('utf-8', 'replace') now returns u'\ufffdAB' and replaces
+  with U+FFFD only the start byte ('\xf1') and the continuation byte ('\x80')
+  even if '\xf1' is the start byte of a 4-bytes sequence.  Previous versions
+  returned a single u'\ufffd'.
+
+- Issue #9011: A negated imaginary literal (e.g., "-7j") now has real part -0.0
+  rather than 0.0.  So "-7j" is now exactly equivalent to "-(7j)".
 
 - Be more specific in error messages about positional arguments.
 
@@ -37,27 +36,26 @@
   objects, as described in the documentation.
 
 - Issue #6543: Write the traceback in the terminal encoding instead of utf-8.
-  Fix the encoding of the modules filename. Patch written by Amaury Forgeot
+  Fix the encoding of the modules filename.  Patch written by Amaury Forgeot
   d'Arc.
 
-- Issue #9011: Remove buggy and unnecessary (in 3.x) ST->AST
-  compilation code dealing with unary minus applied to a constant.
-  The removed code was mutating the ST, causing a second compilation
-  to fail.
+- Issue #9011: Remove buggy and unnecessary (in 3.x) ST->AST compilation code
+  dealing with unary minus applied to a constant.  The removed code was mutating
+  the ST, causing a second compilation to fail.
 
 - Issue #850997: mbcs encoding (Windows only) handles errors argument: strict
-  mode raises unicode errors. The encoder only supports "strict" and "replace"
+  mode raises unicode errors.  The encoder only supports "strict" and "replace"
   error handlers, the decoder only supports "strict" and "ignore" error
-  handlers. Patch written by Mark Hammond.
+  handlers.  Patch written by Mark Hammond.
 
 - Issue #8850: Remove "w" and "w#" formats from PyArg_Parse*() functions, use
   "w*" format instead. Add tests for "w*" format.
 
 - Issue #8592: PyArg_Parse*() functions raise a TypeError for "y", "u" and "Z"
-  formats if the string contains a null byte/character. Write unit tests for
+  formats if the string contains a null byte/character.  Write unit tests for
   string formats.
 
-- Issue #7490: to facilitate sharing of doctests between 2.x and 3.x test
+- Issue #7490: To facilitate sharing of doctests between 2.x and 3.x test
   suites, the IGNORE_EXCEPTION_DETAIL directive now also ignores the module
   location of the raised exception.
 
@@ -66,94 +64,93 @@
 
 - Issue #9058: Remove assertions about INT_MAX in UnicodeDecodeError.
 
-- Issue #8941: decoding big endian UTF-32 data in UCS-2 builds could crash
-  the interpreter with characters outside the Basic Multilingual Plane
-  (higher than 0x10000).
-
-- Issue #8950: (See also issue #5080).  Py_ArgParse*() functions now
-  raise TypeError instead of giving a DeprecationWarning when a float
-  is parsed using the 'L' code (for long long).  (All other integer
-  codes already raise TypeError in this case.)
+- Issue #8941: Decoding big endian UTF-32 data in UCS-2 builds could crash the
+  interpreter with characters outside the Basic Multilingual Plane (higher than
+  0x10000).
+
+- Issue #8950: (See also issue #5080).  Py_ArgParse*() functions now raise
+  TypeError instead of giving a DeprecationWarning when a float is parsed using
+  the 'L' code (for long long).  (All other integer codes already raise
+  TypeError in this case.)
 
 - Issue #8922: Normalize the encoding name in PyUnicode_AsEncodedString() to
   enable shortcuts for upper case encoding name. Add also a shortcut for
   "iso-8859-1" in PyUnicode_AsEncodedString() and PyUnicode_Decode().
 
-- Issue #8838: Remove codecs.charbuffer_encode() function. The buffer protocol
-  doesn't support "char buffer" anymore in Python3.
+- Issue #8838: Remove codecs.charbuffer_encode() function.  The buffer protocol
+  doesn't support "char buffer" anymore in Python 3.
 
 - Issue #8339: Remove "t#" format of PyArg_Parse*() functions, use "s#" or "s*"
-  instead. codecs.charbuffer_encode() now accepts modifiable buffer objects
+  instead.  codecs.charbuffer_encode() now accepts modifiable buffer objects
   like bytearray.
 
-- Issue #8837: Remove "O?" format of PyArg_Parse*() functions. The format is no
+- Issue #8837: Remove "O?" format of PyArg_Parse*() functions.  The format is no
   used anymore and it was never documented.
 
-- In the str.format(), raise a ValueError when indexes to arguments are too
-  large.
+- In str.format(), raise a ValueError when indexes to arguments are too large.
 
-- Issue #2844: Make int('42', n) consistently raise ValueError for
-  invalid integers n (including n = -909).
+- Issue #2844: Make int('42', n) consistently raise ValueError for invalid
+  integers n (including n = -909).
 
-- Issue #8188: Introduce a new scheme for computing hashes of numbers
-  (instances of int, float, complex, decimal.Decimal and
-  fractions.Fraction) that makes it easy to maintain the invariant
-  that hash(x) == hash(y) whenever x and y have equal value.
+- Issue #8188: Introduce a new scheme for computing hashes of numbers (instances
+  of int, float, complex, decimal.Decimal and fractions.Fraction) that makes it
+  easy to maintain the invariant that hash(x) == hash(y) whenever x and y have
+  equal value.
 
 - Issue #8748: Fix two issues with comparisons between complex and integer
   objects.  (1) The comparison could incorrectly return True in some cases
-  (2**53+1 == complex(2**53) == 2**53), breaking transivity of equality.
-  (2) The comparison raised an OverflowError for large integers, leading
-  to unpredictable exceptions when combining integers and complex objects
-  in sets or dicts.
+  (2**53+1 == complex(2**53) == 2**53), breaking transitivity of equality.
+  (2) The comparison raised an OverflowError for large integers, leading to
+  unpredictable exceptions when combining integers and complex objects in sets
+  or dicts.
 
 - Issue #8766: Initialize _warnings module before importing the first module.
   Fix a crash if an empty directory called "encodings" exists in sys.path.
 
 - Issue #8589: Decode PYTHONWARNINGS environment variable with the file system
   encoding and surrogateespace error handler instead of the locale encoding to
-  be consistent with os.environ. Add PySys_AddWarnOptionUnicode() function.
+  be consistent with os.environ.  Add PySys_AddWarnOptionUnicode() function.
 
-- PyObject_Dump() encodes unicode objects to utf8 with backslashreplace
-  (instead of strict) error handler to escape surrogates
+- PyObject_Dump() encodes unicode objects to utf8 with backslashreplace (instead
+  of strict) error handler to escape surrogates.
 
 - Issue #8715: Create PyUnicode_EncodeFSDefault() function: Encode a Unicode
   object to Py_FileSystemDefaultEncoding with the "surrogateescape" error
-  handler, and return bytes. If Py_FileSystemDefaultEncoding is not set, fall
+  handler, and return bytes.  If Py_FileSystemDefaultEncoding is not set, fall
   back to UTF-8.
 
 - Enable shortcuts for common encodings in PyUnicode_AsEncodedString() for any
-  error handler, not only the default error handler (strict)
+  error handler, not only the default error handler (strict).
 
 - Issue #8610: Load file system codec at startup, and display a fatal error on
-  failure. Set the file system encoding to utf-8 (instead of None) if getting
+  failure.  Set the file system encoding to utf-8 (instead of None) if getting
   the locale encoding failed, or if nl_langinfo(CODESET) function is missing.
 
 - PyFile_FromFd() uses PyUnicode_DecodeFSDefault() instead of
-  PyUnicode_FromString() to support surrogates in the filename and use the
-  right encoding
+  PyUnicode_FromString() to support surrogates in the filename and use the right
+  encoding.
 
 - Issue #7507: Quote "!" in pipes.quote(); it is special to some shells.
 
-- PyUnicode_DecodeFSDefaultAndSize() uses surrogateescape error handler
+- PyUnicode_DecodeFSDefaultAndSize() uses surrogateescape error handler.
 
 - Issue #8419: Prevent the dict constructor from accepting non-string keyword
   arguments.
 
 - Issue #8124: PySys_WriteStdout() and PySys_WriteStderr() don't execute
-  indirectly Python signal handlers anymore because mywrite() ignores
-  exceptions (KeyboardInterrupt)
+  indirectly Python signal handlers anymore because mywrite() ignores exceptions
+  (KeyboardInterrupt).
 
 - Issue #8092: Fix PyUnicode_EncodeUTF8() to support error handler producing
-  unicode string (eg. backslashreplace)
+  unicode string (eg. backslashreplace).
 
-- Issue #8485: PyUnicode_FSConverter() doesn't accept bytearray object anymore,
-  you have to convert your bytearray filenames to bytes
+- Issue #8485: PyUnicode_FSConverter() doesn't accept byteearray objects
+  anymore, you have to convert your bytearray filenames to bytes.
 
 - Issue #7332: Remove the 16KB stack-based buffer in
-  PyMarshal_ReadLastObjectFromFile, which doesn't bring any noticeable
-  benefit compared to the dynamic memory allocation fallback.  Patch by
-  Charles-Fran?ois Natali.
+  PyMarshal_ReadLastObjectFromFile, which doesn't bring any noticeable benefit
+  compared to the dynamic memory allocation fallback.  Patch by Charles-Fran?ois
+  Natali.
 
 - Issue #8417: Raise an OverflowError when an integer larger than sys.maxsize is
   passed to bytes or bytearray.
@@ -163,9 +160,9 @@
 - Issue #8329: Don't return the same lists from select.select when no fds are
   changed.
 
-- Issue #8259: 1L << (2**31) no longer produces an 'outrageous shift error'
-  on 64-bit machines.  The shift count for either left or right shift is
-  permitted to be up to sys.maxsize.
+- Issue #8259: 1L << (2**31) no longer produces an 'outrageous shift error' on
+  64-bit machines.  The shift count for either left or right shift is permitted
+  to be up to sys.maxsize.
 
 - Ensure that tokenization of identifiers is not affected by locale.
 
@@ -173,16 +170,16 @@
 
 - Raise a TypeError when trying to delete a T_STRING_INPLACE struct member.
 
-- Issue #8211: Save/restore CFLAGS around AC_PROG_CC in configure.in, in
-  case it is set.
+- Issue #8211: Save/restore CFLAGS around AC_PROG_CC in configure.in, in case it
+  is set.
 
 - Issue #8226: sys.setfilesystemencoding() raises a LookupError if the encoding
-  is unknown
+  is unknown.
 
-- Issue #1583863: An str subclass can now override the __str__ method
+- Issue #1583863: A str subclass can now override the __str__ method.
 
 - Issue #8014: Setting a T_UINT or T_PYSSIZET attribute of an object with
-  PyMemberDefs could produce an internal error;  raise TypeError instead.
+  PyMemberDefs could produce an internal error; raise TypeError instead.
 
 - Issue #7845: Rich comparison methods on the complex type now return
   NotImplemented rather than raising a TypeError when comparing with an
@@ -191,72 +188,66 @@
 
 - Issue #3137: Don't ignore errors at startup, especially a keyboard interrupt
   (SIGINT). If an error occurs while importing the site module, the error is
-  printed and Python exits. Initialize the GIL before importing the site
-  module.
+  printed and Python exits. Initialize the GIL before importing the site module.
 
 - Issue #7173: Generator finalization could invalidate sys.exc_info().
 
-- Issue #7544: Preallocate thread memory before creating the thread to avoid
-  a fatal error in low memory condition.
+- Issue #7544: Preallocate thread memory before creating the thread to avoid a
+  fatal error in low memory condition.
 
-- Issue #7820: The parser tokenizer restores all bytes in the right if
-  the BOM check fails.
+- Issue #7820: The parser tokenizer restores all bytes in the right if the BOM
+  check fails.
 
 - Handle errors from looking up __prepare__ correctly.
 
-- Issue #5939: Add additional runtime checking to ensure a valid capsule
-  in Modules/_ctypes/callproc.c.
+- Issue #5939: Add additional runtime checking to ensure a valid capsule in
+  Modules/_ctypes/callproc.c.
 
 - Issue #7309: Fix unchecked attribute access when converting
-  UnicodeEncodeError, UnicodeDecodeError, and UnicodeTranslateError to
-  strings.
+  UnicodeEncodeError, UnicodeDecodeError, and UnicodeTranslateError to strings.
 
-- Issue #6902: Fix problem with built-in types format incorrectly with
-  0 padding.
+- Issue #6902: Fix problem with built-in types format incorrectly with 0
+  padding.
 
-- Issue #7988: Fix default alignment to be right aligned for
-  complex.__format__. Now it matches other numeric types.
+- Issue #7988: Fix default alignment to be right aligned for complex.__format__.
+  Now it matches other numeric types.
 
 - Issue #5988: Remove deprecated functions PyOS_ascii_formatd,
-  PyOS_ascii_strtod, and PyOS_ascii_atof. Use PyOS_double_to_string
-  and PyOS_string_to_double instead. See issue #5835 for the original
-  deprecations.
-
-- Issue #7385: Fix a crash in `MemoryView_FromObject` when
-  `PyObject_GetBuffer` fails.  Patch by Florent Xicluna.
-
-- Issue #7788: Fix an interpreter crash produced by deleting a list
-  slice with very large step value.
-
-- Issue #7766: Change sys.getwindowsversion() return value to a named
-  tuple and add the additional members returned in an OSVERSIONINFOEX
-  structure. The new members are service_pack_major, service_pack_minor,
-  suite_mask, and product_type.
-
-- Issue #7561: Operations on empty bytearrays (such as `int(bytearray())`)
-  could crash in many places because of the PyByteArray_AS_STRING() macro
-  returning NULL.  The macro now returns a statically allocated empty
-  string instead.
+  PyOS_ascii_strtod, and PyOS_ascii_atof.  Use PyOS_double_to_string and
+  PyOS_string_to_double instead.  See issue #5835 for the original deprecations.
+
+- Issue #7385: Fix a crash in `MemoryView_FromObject` when `PyObject_GetBuffer`
+  fails.  Patch by Florent Xicluna.
+
+- Issue #7788: Fix an interpreter crash produced by deleting a list slice with
+  very large step value.
+
+- Issue #7766: Change sys.getwindowsversion() return value to a named tuple and
+  add the additional members returned in an OSVERSIONINFOEX structure.  The new
+  members are service_pack_major, service_pack_minor, suite_mask, and
+  product_type.
+
+- Issue #7561: Operations on empty bytearrays (such as `int(bytearray())`) could
+  crash in many places because of the PyByteArray_AS_STRING() macro returning
+  NULL.  The macro now returns a statically allocated empty string instead.
 
 - Issue #6690: Optimize the bytecode for expressions such as `x in {1, 2, 3}`,
-  where the right hand operand is a set of constants, by turning the set into
-  a frozenset and pre-building it as a constant.  The comparison operation
-  is made against the constant instead of building a new set each time it is
-  executed (a similar optimization already existed which turned a list of
-  constants into a pre-built tuple).  Patch and additional tests by Dave
-  Malcolm.
-
-- Issue #7622: Improve the split(), rsplit(), splitlines() and replace()
-  methods of bytes, bytearray and unicode objects by using a common
-  implementation based on stringlib's fast search.  Patch by Florent Xicluna.
-
-- Issue #7632: Fix various str -> float conversion bugs present in 2.7
-  alpha 2, including: (1) a serious 'wrong output' bug that could
-  occur for long (> 40 digit) input strings, (2) a crash in dtoa.c
-  that occurred in debug builds when parsing certain long numeric
-  strings corresponding to subnormal values, (3) a memory leak for
-  some values large enough to cause overflow, and (4) a number of
-  flaws that could lead to incorrectly rounded results.
+  where the right hand operand is a set of constants, by turning the set into a
+  frozenset and pre-building it as a constant.  The comparison operation is made
+  against the constant instead of building a new set each time it is executed (a
+  similar optimization already existed which turned a list of constants into a
+  pre-built tuple).  Patch and additional tests by Dave Malcolm.
+
+- Issue #7622: Improve the split(), rsplit(), splitlines() and replace() methods
+  of bytes, bytearray and unicode objects by using a common implementation based
+  on stringlib's fast search.  Patch by Florent Xicluna.
+
+- Issue #7632: Fix various str -> float conversion bugs present in 2.7 alpha 2,
+  including: (1) a serious 'wrong output' bug that could occur for long (> 40
+  digit) input strings, (2) a crash in dtoa.c that occurred in debug builds when
+  parsing certain long numeric strings corresponding to subnormal values, (3) a
+  memory leak for some values large enough to cause overflow, and (4) a number
+  of flaws that could lead to incorrectly rounded results.
 
 - The __complex__ method is now looked up on the class of instances to make it
   consistent with other special methods.
@@ -267,24 +258,23 @@
 - Issue #7604: Deleting an unset slotted attribute did not raise an
   AttributeError.
 
-- Issue #7534: Fix handling of IEEE specials (infinities, nans,
-  negative zero) in ** operator.  The behaviour now conforms to that
-  described in C99 Annex F.
-
-- Issue #1811: improve accuracy and cross-platform consistency for
-  true division of integers: the result of a/b is now correctly
-  rounded for ints a and b (at least on IEEE 754 platforms), and in
-  particular does not depend on the internal representation of an int.
+- Issue #7534: Fix handling of IEEE specials (infinities, nans, negative zero)
+  in ** operator.  The behaviour now conforms to that described in C99 Annex F.
+
+- Issue #1811: improve accuracy and cross-platform consistency for true division
+  of integers: the result of a/b is now correctly rounded for ints a and b (at
+  least on IEEE 754 platforms), and in particular does not depend on the
+  internal representation of an int.
 
 - Issue #6834: replace the implementation for the 'python' and 'pythonw'
   executables on OSX.
 
-  These executables now work properly with the arch(1) command:
-  ``arch -ppc python`` will start a universal binary version of python
-  in PPC mode (unlike previous releases).
+  These executables now work properly with the arch(1) command: ``arch -ppc
+  python`` will start a universal binary version of python in PPC mode (unlike
+  previous releases).
 
-- Issue #7466: segmentation fault when the garbage collector is called
-  in the middle of populating a tuple.  Patch by Florent Xicluna.
+- Issue #7466: Segmentation fault when the garbage collector is called in the
+  middle of populating a tuple.  Patch by Florent Xicluna.
 
 - Issue #7419: setlocale() could crash the interpreter on Windows when called
   with invalid values.
@@ -292,22 +282,22 @@
 - Issue #6077: On Windows, files opened with tempfile.TemporaryFile in "wt+"
   mode would appear truncated on the first '0x1a' byte (aka. Ctrl+Z).
 
-- Issue #7085: Fix crash when importing some extensions in a thread
-  on MacOSX 10.6.
+- Issue #7085: Fix crash when importing some extensions in a thread on MacOSX
+  10.6.
 
 - Issue #1757126: Fix the cyrillic-asian alias for the ptcp154 encoding.
 
 - Issue #6970: Remove redundant calls when comparing objects that don't
   implement the relevant rich comparison methods.
 
-- Issue #7298: fixes for range and reversed(range(...)).  Iteration
-  over range(a, b, c) incorrectly gave an empty iterator when a, b and
-  c fit in C long but the length of the range did not.  Also fix
-  several cases where reversed(range(a, b, c)) gave wrong results, and
-  fix a refleak for reversed(range(a, b, c)) with large arguments.
+- Issue #7298: Fixes for range and reversed(range(...)).  Iteration over
+  range(a, b, c) incorrectly gave an empty iterator when a, b and c fit in C
+  long but the length of the range did not.  Also fix several cases where
+  reversed(range(a, b, c)) gave wrong results, and fix a refleak for
+  reversed(range(a, b, c)) with large arguments.
 
-- Issue #7244: itertools.izip_longest() no longer ignores exceptions
-  raised during the formation of an output tuple.
+- Issue #7244: itertools.izip_longest() no longer ignores exceptions raised
+  during the formation of an output tuple.
 
 - Issue #3297: On wide unicode builds, do not split unicode characters into
   surrogates.
@@ -324,7 +314,7 @@
   will be entirely removed in 3.3.
 
 - Support for OSF* has been disabled. If nobody stands up, support will be
-  removed in 3.3. See http://bugs.python.org/issue8606 .
+  removed in 3.3. See <http://bugs.python.org/issue8606>.
 
 - Peephole constant folding had missed UNARY_POSITIVE.
 
@@ -335,26 +325,26 @@
 - Issue #7147: Remove support for compiling Python without complex number
   support.
 
-- Issue #7120: logging: Removed import of multiprocessing which is causing
-  crash in GAE.
+- Issue #7120: logging: Removed import of multiprocessing which is causing crash
+  in GAE.
 
-- Issue #1754094: Improve the stack depth calculation in the compiler.
-  There should be no other effect than a small decrease in memory use.
-  Patch by Christopher Tur Lesniewski-Laas.
+- Issue #1754094: Improve the stack depth calculation in the compiler.  There
+  should be no other effect than a small decrease in memory use.  Patch by
+  Christopher Tur Lesniewski-Laas.
 
-- Issue #7065: Fix a crash in bytes.maketrans and bytearray.maketrans when
-  using byte values greater than 127.  Patch by Derk Drukker.
+- Issue #7065: Fix a crash in bytes.maketrans and bytearray.maketrans when using
+  byte values greater than 127.  Patch by Derk Drukker.
 
 - Issue #1571184: The Unicode database contains properties for more characters.
   The tables for code points representing numeric values, white spaces or line
   breaks are now generated from the official Unicode Character Database files,
   and include information from the Unihan.txt file.
 
-- Issue #7019: Raise ValueError when unmarshalling bad long data, instead
-  of producing internally inconsistent Python longs.
+- Issue #7019: Raise ValueError when unmarshalling bad long data, instead of
+  producing internally inconsistent Python longs.
 
-- Issue #6990: Fix threading.local subclasses leaving old state around
-  after a reference cycle GC which could be recycled by new locals.
+- Issue #6990: Fix threading.local subclasses leaving old state around after a
+  reference cycle GC which could be recycled by new locals.
 
 - Issue #5460: Fix an ambiguity in the grammar.
 
@@ -364,8 +354,8 @@
 
 - Issue #6846: Fix bug where bytearray.pop() returns negative integers.
 
-- Issue #6750: A text file opened with io.open() could duplicate its output
-  when writing from multiple threads at the same time.
+- Issue #6750: A text file opened with io.open() could duplicate its output when
+  writing from multiple threads at the same time.
 
 - Issue #6707: dir() on an uninitialized module caused a crash.
 
@@ -374,85 +364,81 @@
 - Issue #6573: set.union() stopped processing inputs if an instance of self
   occurred in the argument chain.
 
-- Issue #6070: On posix platforms import no longer copies the execute bit
-  from the .py file to the .pyc file if it is set.
+- Issue #6070: On posix platforms import no longer copies the execute bit from
+  the .py file to the .pyc file if it is set.
 
 - Issue #1616979: Added the cp720 (Arabic DOS) encoding.
 
-- Issue #6428: Since Python 3.0, the __bool__ method must return a bool
-  object, and not an int.  Fix the corresponding error message, and the
-  documentation.
+- Issue #6428: Since Python 3.0, the __bool__ method must return a bool object,
+  and not an int.  Fix the corresponding error message, and the documentation.
 
 - The deprecated PyCObject has been removed.
 
-- Issue #6347: Include inttypes.h as well as stdint.h in pyport.h.
-  This fixes a build failure on HP-UX: int32_t and uint32_t are
-  defined in inttypes.h instead of stdint.h on that platform.
+- Issue #6347: Include inttypes.h as well as stdint.h in pyport.h.  This fixes a
+  build failure on HP-UX: int32_t and uint32_t are defined in inttypes.h instead
+  of stdint.h on that platform.
 
-- Issue #6373: Fixed a SystemError when encoding with the latin-1 codec and
-  the 'surrogateescape' error handler, a string which contains unpaired
-  surrogates.
+- Issue #6373: Fixed a SystemError when encoding with the latin-1 codec and the
+  'surrogateescape' error handler, a string which contains unpaired surrogates.
 
 - Issue #4856: Remove checks for win NT.
 
-- Issue #6687: PyBytes_FromObject() no longer accepts an integer as its
-  argument to construct a null-initialized bytes object.
+- Issue #6687: PyBytes_FromObject() no longer accepts an integer as its argument
+  to construct a null-initialized bytes object.
 
-- Issue #1023290: Add from_bytes() and to_bytes() methods to integers.
-  These methods allow the conversion of integers to bytes, and vice-versa.
+- Issue #1023290: Add from_bytes() and to_bytes() methods to integers.  These
+  methods allow the conversion of integers to bytes, and vice-versa.
 
-- Issue #7382: Fix bug in bytes.__getnewargs__ that prevented bytes
-  instances from being copied with copy.copy(), and bytes subclasses
-  from being pickled properly.
+- Issue #7382: Fix bug in bytes.__getnewargs__ that prevented bytes instances
+  from being copied with copy.copy(), and bytes subclasses from being pickled
+  properly.
 
 - Code objects now support weak references.
 
-- Issue #7072: isspace(0xa0) is true on Mac OS X
+- Issue #7072: isspace(0xa0) is true on Mac OS X.
 
-- Issue #8084: PEP 370 now conforms to system conventions for framework
-  builds on MacOS X. That is, "python setup.py install --user" will install
+- Issue #8084: PEP 370 now conforms to system conventions for framework builds
+  on MacOS X. That is, "python setup.py install --user" will install
     into "~/Library/Python/2.7" instead of "~/.local".
 
 C-API
 -----
 
-- Issue #5753: A new C API function, :cfunc:`PySys_SetArgvEx`, allows
-  embedders of the interpreter to set sys.argv without also modifying
-  sys.path.  This helps fix `CVE-2008-5983
+- Issue #5753: A new C API function, `PySys_SetArgvEx`, allows embedders of the
+  interpreter to set sys.argv without also modifying sys.path.  This helps fix
+  `CVE-2008-5983
   <http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2008-5983>`_.
 
 - Add PyArg_ValidateKeywordArguments, which checks if all keyword arguments are
   strings in an efficient manner.
 
 - Issue #8276: PyEval_CallObject() is now only available in macro form.  The
-  function declaration, which was kept for backwards compatibility reasons,
-  is now removed (the macro was introduced in 1997!).
+  function declaration, which was kept for backwards compatibility reasons, is
+  now removed (the macro was introduced in 1997!).
 
-- Issue #7767: New function PyLong_AsLongLongAndOverflow added,
-  analogous to PyLong_AsLongAndOverflow.
+- Issue #7767: New function PyLong_AsLongLongAndOverflow added, analogous to
+  PyLong_AsLongAndOverflow.
 
 - Make PyUnicode_CompareWithASCIIString return not equal if the Python string
   has '\0' at the end.
 
 - Issue #5080: The argument parsing functions PyArg_ParseTuple,
-  PyArg_ParseTupleAndKeywords, PyArg_VaParse,
-  PyArg_VaParseTupleAndKeywords and PyArg_Parse now raise a
-  DeprecationWarning for float arguments passed with the 'L' format
-  code.  This will become a TypeError in a future version of Python,
+  PyArg_ParseTupleAndKeywords, PyArg_VaParse, PyArg_VaParseTupleAndKeywords and
+  PyArg_Parse now raise a DeprecationWarning for float arguments passed with the
+  'L' format code.  This will become a TypeError in a future version of Python,
   to match the behaviour of the other integer format codes.
 
-- Issue #7033: function ``PyErr_NewExceptionWithDoc()`` added.
+- Issue #7033: Function ``PyErr_NewExceptionWithDoc()`` added.
 
-- Issue #7414: 'C' code wasn't being skipped properly (for keyword arguments)
-  in PyArg_ParseTupleAndKeywords.
+- Issue #7414: 'C' code wasn't being skipped properly (for keyword arguments) in
+  PyArg_ParseTupleAndKeywords.
 
-- Issue #7228: Add '%lld' and '%llu' support to PyString_FromFormat(V)
-  and PyErr_Format, on machines with HAVE_LONG_LONG defined.
+- Issue #7228: Add '%lld' and '%llu' support to PyString_FromFormat(V) and
+  PyErr_Format, on machines with HAVE_LONG_LONG defined.
 
-- Issue #6151: Made PyDescr_COMMON conform to standard C (like PyObject_HEAD
-  in PEP 3123).  The PyDescr_TYPE and PyDescr_NAME macros should be
-  should used for accessing the d_type and d_name members of structures
-  using PyDescr_COMMON.
+- Issue #6151: Made PyDescr_COMMON conform to standard C (like PyObject_HEAD in
+  PEP 3123).  The PyDescr_TYPE and PyDescr_NAME macros should be should used for
+  accessing the d_type and d_name members of structures using PyDescr_COMMON.
 
 - Issue #6405: Remove duplicate type declarations in descrobject.h.
 
@@ -467,56 +453,56 @@
 - Issue #1419652: Change the first argument to PyImport_AppendInittab() to
   ``const char *`` as the string is stored beyond the call.
 
-- Issue #2422: When compiled with the ``--with-valgrind`` option, the
-  pymalloc allocator will be automatically disabled when running under
-  Valgrind.  This gives improved memory leak detection when running
-  under Valgrind, while taking advantage of pymalloc at other times.
+- Issue #2422: When compiled with the ``--with-valgrind`` option, the pymalloc
+  allocator will be automatically disabled when running under Valgrind.  This
+  gives improved memory leak detection when running under Valgrind, while taking
+  advantage of pymalloc at other times.
 
 Library
 -------
 
-- In pdb, when Ctrl-C is entered while defining commands for a
-  breakpoint, the old commands are restored.
+- In pdb, when Ctrl-C is entered while defining commands for a breakpoint, the
+  old commands are restored.
 
-- For traceback debugging, the pdb listing now also shows the locations
-  where the exception was originally (re)raised, if it differs from the
-  last line executed (e.g. in case of finally clauses).
+- For traceback debugging, the pdb listing now also shows the locations where
+  the exception was originally (re)raised, if it differs from the last line
+  executed (e.g. in case of finally clauses).
 
-- The pdb command "source" has been added.  It displays the source
-  code for a given object, if possible.
+- The pdb command "source" has been added.  It displays the source code for a
+  given object, if possible.
 
-- The pdb command "longlist" has been added.  It displays the whole
-  source code for the current function.
+- The pdb command "longlist" has been added.  It displays the whole source code
+  for the current function.
 
-- Issue #1503502: Make pdb.Pdb easier to subclass by putting message
-  and error output into methods.
+- Issue #1503502: Make pdb.Pdb easier to subclass by putting message and error
+  output into methods.
 
-- Issue #809887: Make the output of pdb's breakpoint deletions more
-  consistent; emit a message when a breakpoint is enabled or disabled.
+- Issue #809887: Make the output of pdb's breakpoint deletions more consistent;
+  emit a message when a breakpoint is enabled or disabled.
 
-- Issue #5294: Fix the behavior of pdb's "continue" command when called
-  in the top-level debugged frame.
+- Issue #5294: Fix the behavior of pdb's "continue" command when called in the
+  top-level debugged frame.
 
-- Issue #5727: Restore the ability to use readline when calling into pdb
-  in doctests.
+- Issue #5727: Restore the ability to use readline when calling into pdb in
+  doctests.
 
-- Issue #6719: In pdb, do not stop somewhere in the encodings machinery
-  if the source file to be debugged is in a non-builtin encoding.
+- Issue #6719: In pdb, do not stop somewhere in the encodings machinery if the
+  source file to be debugged is in a non-builtin encoding.
 
-- Issue #8048: Prevent doctests from failing when sys.displayhook has
-  been reassigned.
+- Issue #8048: Prevent doctests from failing when sys.displayhook has been
+  reassigned.
 
-- Issue #8015: In pdb, do not crash when an empty line is entered as
-  a breakpoint command.
+- Issue #8015: In pdb, do not crash when an empty line is entered as a
+  breakpoint command.
 
 - In pdb, allow giving a line number to the "until" command.
 
-- Issue #1437051: For pdb, allow "continue" and related commands in
-  .pdbrc files.  Also, add a command-line option "-c" that runs a
-  command as if given in .pdbrc.
+- Issue #1437051: For pdb, allow "continue" and related commands in .pdbrc
+  files.  Also, add a command-line option "-c" that runs a command as if given
+  in .pdbrc.
 
-- Issue #4179: In pdb, allow "list ." as a command to return to the
-  currently debugged line.
+- Issue #4179: In pdb, allow "list ." as a command to return to the currently
+  debugged line.
 
 - Issue #4108: In urllib.robotparser, if there are multiple 'User-agent: *'
   entries, consider the first one.
@@ -524,74 +510,72 @@
 - Issue #6630: Allow customizing regex flags when subclassing the
   string.Template class.
 
-- Issue #9411: Allow specifying an encoding for config files in the
-  configparser module.
+- Issue #9411: Allow specifying an encoding for config files in the configparser
+  module.
 
-- Issue #1682942: Improvements to configparser: support alternate
-  delimiters, alternate comment prefixes and empty lines in values.
+- Issue #1682942: Improvements to configparser: support alternate delimiters,
+  alternate comment prefixes and empty lines in values.
 
 - Issue #9354: Provide getsockopt() in asyncore's file_wrapper.
 
 - Issue #8966: ctypes: Remove implicit bytes-unicode conversion.
 
-- Issue #9378: python -m pickle <pickle file> will now load and
-  display the first object in the pickle file.
+- Issue #9378: python -m pickle <pickle file> will now load and display the
+  first object in the pickle file.
 
 - Issue #4770: Restrict binascii module to accept only bytes (as specified).
-  And fix the email package to encode to ASCII instead of
-  ``raw-unicode-escape`` before ASCII-to-binary decoding.
+  And fix the email package to encode to ASCII instead of ``raw-unicode-escape``
+  before ASCII-to-binary decoding.
 
-- Issue #9384: python -m tkinter will now display a simple demo applet.
+- Issue #9384: ``python -m tkinter`` will now display a simple demo applet.
 
-- The default size of the re module's compiled regular expression cache has
-  been increased from 100 to 500 and the cache replacement policy has changed
-  from simply clearing the entire cache on overflow to randomly forgetting 20%
-  of the existing cached compiled regular expressions.  This is a performance
-  win for applications that use a lot of regular expressions and limits the
-  impact of the performance hit anytime the cache is exceeded.
+- The default size of the re module's compiled regular expression cache has been
+  increased from 100 to 500 and the cache replacement policy has changed from
+  simply clearing the entire cache on overflow to randomly forgetting 20% of the
+  existing cached compiled regular expressions.  This is a performance win for
+  applications that use a lot of regular expressions and limits the impact of
+  the performance hit anytime the cache is exceeded.
 
 - Issue #7113: Speed up loading in configparser. Patch by ?ukasz Langa.
 
-- Issue #9032: XML-RPC client retries the request on EPIPE error. The EPIPE
+- Issue #9032: XML-RPC client retries the request on EPIPE error.  The EPIPE
   error occurs when the server closes the socket and the client sends a big
   XML-RPC request.
 
-- Issue #4629: getopt raises an error if an argument ends with = whereas getopt
-  doesn't except a value (eg. --help= is rejected if getopt uses ['help='] long
-  options).
-
-- Issue #7989: Added pure python implementation of the `datetime`
-  module.  The C module is renamed to `_datetime` and if available,
-  overrides all classes defined in datetime with fast C impementation.
-  Python implementation is based on the original python prototype for
-  the datetime module by Tim Peters with minor modifications by the
-  PyPy project.  The test suite now tests `datetime` module with and
-  without `_datetime` acceleration using the same test cases.
+- Issue #4629: getopt raises an error if an argument ends with "=", whereas
+  getopt doesn't accept a value (eg. --help= is rejected if getopt uses
+  ['help='] long options).
+
+- Issue #7989: Added pure python implementation of the `datetime` module.  The C
+  module is renamed to `_datetime` and if available, overrides all classes
+  defined in datetime with fast C impementation.  Python implementation is based
+  on the original python prototype for the datetime module by Tim Peters with
+  minor modifications by the PyPy project.  The test suite now tests `datetime`
+  module with and without `_datetime` acceleration using the same test cases.
 
-- Issue #7895: platform.mac_ver() no longer crashes after calling os.fork()
+- Issue #7895: platform.mac_ver() no longer crashes after calling os.fork().
 
-- Issue #9323: Fixed a bug in trace.py that resulted in loosing the
-  name of the script being traced.  Patch by Eli Bendersky.
+- Issue #9323: Fixed a bug in trace.py that resulted in loosing the name of the
+  script being traced.  Patch by Eli Bendersky.
 
-- Issue #9282: Fixed --listfuncs option of trace.py.  Thanks Eli
-  Bendersky for the patch.
+- Issue #9282: Fixed --listfuncs option of trace.py.  Thanks Eli Bendersky for
+  the patch.
 
 - Issue #3704: http.cookiejar was not properly handling URLs with a / in the
   parameters.
 
-- Issue #9268: ``pickletools.dis()`` now has an optional *annotate*
-  argument which controls printing of opcode descriptions in ``dis()``
-  output.
+- Issue #9268: ``pickletools.dis()`` now has an optional *annotate* argument
+  which controls printing of opcode descriptions in ``dis()`` output.
 
-- Issue #1555570: email no longer inserts extra blank lines when a \r\n
-  combo crosses an 8192 byte boundary.
+- Issue #1555570: email no longer inserts extra blank lines when a \r\n combo
+  crosses an 8192 byte boundary.
 
 - Issue #9243: Fix sndhdr module and add unit tests, contributed by James Lee.
 
 - ``ast.literal_eval()`` now allows byte literals.
 
-- Issue #9137: Fix issue in MutableMapping.update, which incorrectly
-  treated keyword arguments called 'self' or 'other' specially.
+- Issue #9137: Fix issue in MutableMapping.update, which incorrectly treated
+  keyword arguments called 'self' or 'other' specially.
 
 - ``ast.literal_eval()`` now allows set literals.
 
@@ -599,13 +583,13 @@
 
 - Issue #7646: The fnmatch pattern cache no longer grows without bound.
 
-- Issue #9136: Fix 'dictionary changed size during iteration'
-  RuntimeError produced when profiling the decimal module.  This was
-  due to a dangerous iteration over 'locals()' in Context.__init__.
-
-- Fix extreme speed issue in Decimal.pow when the base is an exact
-  power of 10 and the exponent is tiny (for example,
-  Decimal(10) ** Decimal('1e-999999999')).
+- Issue #9136: Fix 'dictionary changed size during iteration' RuntimeError
+  produced when profiling the decimal module.  This was due to a dangerous
+  iteration over 'locals()' in Context.__init__.
+
+- Fix extreme speed issue in Decimal.pow when the base is an exact power of 10
+  and the exponent is tiny (for example, ``Decimal(10) **
+  Decimal('1e-999999999')``).
 
 - Issue #9186: Fix math.log1p(-1.0) to raise ValueError, not OverflowError.
 
@@ -613,18 +597,18 @@
 
 - Issue #9128: Fix validation of class decorators in parser module.
 
-- Issue #9094: python -m pickletools will now disassemble pickle files
-  listed in the command line arguments.  See output of python -m
-  pickletools -h for more details.
+- Issue #9094: python -m pickletools will now disassemble pickle files listed in
+  the command line arguments.  See output of python -m pickletools -h for more
+  details.
 
 - Issue #5468: urlencode to handle bytes type and other encodings in its query
   parameter. Patch by Dan Mahn.
 
-- Issue #7673: Fix security vulnerability (CVE-2010-2089) in the audioop
-  module, ensure that the input string length is a multiple of the frame size
+- Issue #7673: Fix security vulnerability (CVE-2010-2089) in the audioop module,
+  ensure that the input string length is a multiple of the frame size.
 
-- Issue #6507: Accept source strings in dis.dis(). Original patch by
-  Daniel Urban.
+- Issue #6507: Accept source strings in dis.dis().  Original patch by Daniel
+  Urban.
 
 - Issue #7829: Clearly document that the dis module is exposing an
   implementation detail that is not stable between Python VMs or releases.
@@ -633,73 +617,72 @@
   raises an exception.
 
 - Issue #9110: Addition of ContextDecorator to contextlib, for creating APIs
-  that act as both context managers and decorators. contextmanager changes
-  to use ContextDecorator.
+  that act as both context managers and decorators. contextmanager changes to
+  use ContextDecorator.
 
 - Implement importlib.abc.SourceLoader and deprecate PyLoader and PyPycLoader
   for removal in Python 3.4.
 
-- Issue #9064: pdb's "up" and "down" commands now accept an optional argument.
+- Issue #9064: pdb's "up" and "down" commands now accept an optional argument
+  giving the number of frames to go.
 
-- Issue #9018: os.path.normcase() now raises a TypeError if the argument is
-  not ``str`` or ``bytes``.
+- Issue #9018: os.path.normcase() now raises a TypeError if the argument is not
+  ``str`` or ``bytes``.
 
-- Issue #9075: In the ssl module, remove the setting of a ``debug`` flag
-  on an OpenSSL structure.
+- Issue #9075: In the ssl module, remove the setting of a ``debug`` flag on an
+  OpenSSL structure.
 
-- Issue #8682: The ssl module now temporary increments the reference count of
-  a socket object got through ``PyWeakref_GetObject``, so as to avoid possible
+- Issue #8682: The ssl module now temporary increments the reference count of a
+  socket object got through ``PyWeakref_GetObject``, so as to avoid possible
   deallocation while the object is still being used.
 
 - Issue #1368368: FancyURLOpener class changed to throw an Exception on wrong
-  password instead of presenting an interactive prompt. Older behavior can be
+  password instead of presenting an interactive prompt.  Older behavior can be
   obtained by passing retry=True to http_error_xxx methods of FancyURLOpener.
 
-- Issue #8720: fix regression caused by fix for #4050 by making getsourcefile
+- Issue #8720: Fix regression caused by fix for #4050 by making getsourcefile
   smart enough to find source files in the linecache.
 
-- Issue #5610: feedparser no longer eats extra characters at the end of
-  a body part if the body part ends with a \r\n.
+- Issue #5610: feedparser no longer eats extra characters at the end of a body
+  part if the body part ends with a ``\r\n``.
 
-- Issue #8986: math.erfc was incorrectly raising OverflowError for
-  values between -27.3 and -30.0 on some platforms.
+- Issue #8986: math.erfc was incorrectly raising OverflowError for values
+  between -27.3 and -30.0 on some platforms.
 
 - Issue #8784: Set tarfile default encoding to 'utf-8' on Windows.
 
 - Issue #8966: If a ctypes structure field is an array of c_char, convert its
   value to bytes instead of str (as done for c_char and c_char_p).
 
-- Issue #8188: Comparisons between Decimal and Fraction objects are
-  now permitted, returning a result based on the exact numerical
-  values of the operands.  This builds on issue #2531, which allowed
-  Decimal-to-float comparisons;  all comparisons involving numeric
-  types (bool, int, float, complex, Decimal, Fraction) should now
-  act as expected.
+- Issue #8188: Comparisons between Decimal and Fraction objects are now
+  permitted, returning a result based on the exact numerical values of the
+  operands.  This builds on issue #2531, which allowed Decimal-to-float
+  comparisons; all comparisons involving numeric types (bool, int, float,
+  complex, Decimal, Fraction) should now act as expected.
 
-- Issue #8897: Fix sunau module, use bytes to write the header. Patch written
-  by Thomas Jollans.
+- Issue #8897: Fix sunau module, use bytes to write the header. Patch written by
+  Thomas Jollans.
 
 - Issue #8899: time.struct_time now has class and atribute docstrings.
 
 - Issue #6470: Drop UNC prefix in FixTk.
 
-- Issue #4768: base64 encoded email body parts were incorrectly stored as
-  binary strings.  They are now correctly converted to strings.
+- Issue #4768: base64 encoded email body parts were incorrectly stored as binary
+  strings.  They are now correctly converted to strings.
 
 - Issue #8833: tarfile created hard link entries with a size field != 0 by
   mistake.
 
-- Charset.body_encode now correctly handles base64 encoding by encoding
-  with the output_charset before calling base64mime.encode.  Passes the
-  tests from 2.x issue 1368247.
+- Charset.body_encode now correctly handles base64 encoding by encoding with the
+  output_charset before calling base64mime.encode.  Passes the tests from 2.x
+  issue 1368247.
 
 - Issue #8845: sqlite3 Connection objects now have a read-only in_transaction
   attribute that is True iff there are uncommitted changes.
 
-- Issue #1289118: datetime.timedelta objects can now be multiplied by float
-  and divided by float and int objects.  Results are rounded to the nearest
-  multiple of timedelta.resolution with ties resolved using round-half-to-even
-  method.
+- Issue #1289118: datetime.timedelta objects can now be multiplied by float and
+  divided by float and int objects.  Results are rounded to the nearest multiple
+  of timedelta.resolution with ties resolved using round-half-to-even method.
 
 - Issue #7150: Raise OverflowError if the result of adding or subtracting
   timedelta from date or datetime falls outside of the MINYEAR:MAXYEAR range.
@@ -708,25 +691,24 @@
 
 - Issue #4769: Fix main() function of the base64 module, use sys.stdin.buffer
   and sys.stdout.buffer (instead of sys.stdin and sys.stdout) to use the bytes
-  API
+  API.
 
-- Issue #8770: now sysconfig displays information when it's called as
-  a script. Initial idea by Sridhar Ratnakumar.
+- Issue #8770: Now sysconfig displays information when it's called as a script.
+  Initial idea by Sridhar Ratnakumar.
 
 - Issue #6662: Fix parsing of malformatted charref (&#bad;), patch written by
-  Fredrik H??rd
+  Fredrik H??rd.
 
 - Issue #8540: Decimal module: rename the Context._clamp attribute to
-  Context.clamp and make it public.  This is useful in creating
-  contexts that correspond to the decimal interchange formats
-  specified in IEEE 754.
+  Context.clamp and make it public.  This is useful in creating contexts that
+  correspond to the decimal interchange formats specified in IEEE 754.
 
 - Issue #6268: Fix seek() method of codecs.open(), don't read or write the BOM
   twice after seek(0). Fix also reset() method of codecs, UTF-16, UTF-32 and
   StreamWriter classes.
 
 - Issue #3798: sys.exit(message) writes the message to sys.stderr file, instead
-  of the C file stderr, to use stderr encoding and error handler
+  of the C file stderr, to use stderr encoding and error handler.
 
 - Issue #8782: Add a trailing newline in linecache.updatecache to the last line
   of files without one.
@@ -735,11 +717,11 @@
   comparing to a non-mapping.
 
 - Issue #8774: tabnanny uses the encoding cookie (#coding:...) to use the
-  correct encoding
+  correct encoding.
 
-- Issue #4870: Add an `options` attribute to SSL contexts, as well as
-  several ``OP_*`` constants to the `ssl` module.  This allows to selectively
-  disable protocol versions, when used in combination with `PROTOCOL_SSLv23`.
+- Issue #4870: Add an `options` attribute to SSL contexts, as well as several
+  ``OP_*`` constants to the `ssl` module.  This allows to selectively disable
+  protocol versions, when used in combination with `PROTOCOL_SSLv23`.
 
 - Issue #8759: Fixed user paths in sysconfig for posix and os2 schemes.
 
@@ -761,15 +743,15 @@
 - Issue #8688: Distutils now recalculates MANIFEST everytime.
 
 - Issue #8477: ssl.RAND_egd() and ssl._test_decode_cert() support str with
-  surrogates and bytes for the filename
+  surrogates and bytes for the filename.
 
 - Issue #8550: Add first class ``SSLContext`` objects to the ssl module.
 
-- Issue #8681: Make the zlib module's error messages more informative when
-  the zlib itself doesn't give any detailed explanation.
+- Issue #8681: Make the zlib module's error messages more informative when the
+  zlib itself doesn't give any detailed explanation.
 
-- The audioop module now supports sound fragments of length greater
-  than 2**31 bytes on 64-bit machines, and is PY_SSIZE_T_CLEAN.
+- The audioop module now supports sound fragments of length greater than 2**31
+  bytes on 64-bit machines, and is PY_SSIZE_T_CLEAN.
 
 - Issue #4972: Add support for the context manager protocol to the ftplib.FTP
   class.
@@ -780,9 +762,8 @@
 - Issue #8514: Add os.fsencode() function (Unix only): encode a string to bytes
   for use in the file system, environment variables or the command line.
 
-- Issue #8571: Fix an internal error when compressing or decompressing a
-  chunk larger than 1GB with the zlib module's compressor and decompressor
-  objects.
+- Issue #8571: Fix an internal error when compressing or decompressing a chunk
+  larger than 1GB with the zlib module's compressor and decompressor objects.
 
 - Issue #8603: Support bytes environmental variables on Unix: Add os.environb
   mapping and os.getenvb() function. os.unsetenv() encodes str argument to the
@@ -791,46 +772,46 @@
 
 - Issue #8573: asyncore _strerror() function might throw ValueError.
 
-- Issue #8483: asyncore.dispatcher's __getattr__ method produced confusing
-  error messages when accessing undefined class attributes because of the cheap
-  inheritance with the underlying socket object.
-  The cheap inheritance has been deprecated.
+- Issue #8483: asyncore.dispatcher's __getattr__ method produced confusing error
+  messages when accessing undefined class attributes because of the cheap
+  inheritance with the underlying socket object.  The cheap inheritance has been
+  deprecated.
 
 - Issue #4265: shutil.copyfile() was leaking file descriptors when disk fills.
   Patch by Tres Seaver.
 
 - Issue #8390: tarfile uses surrogateespace as the default error handler
-  (instead of replace in read mode or strict in write mode)
+  (instead of replace in read mode or strict in write mode).
 
 - Issue #7755: Use an unencumbered audio file for tests.
 
-- Issue #8621: uuid.uuid4() returned the same sequence of values in the
-  parent and any children created using ``os.fork`` on MacOS X 10.6.
+- Issue #8621: uuid.uuid4() returned the same sequence of values in the parent
+  and any children created using ``os.fork`` on MacOS X 10.6.
 
-- Issue #8567: Fix precedence of signals in Decimal module: when a
-  Decimal operation raises multiple signals and more than one of those
-  signals is trapped, the specification determines the order in which
-  the signals should be handled.  In many cases this order wasn't
-  being followed, leading to the wrong Python exception being raised.
+- Issue #8567: Fix precedence of signals in Decimal module: when a Decimal
+  operation raises multiple signals and more than one of those signals is
+  trapped, the specification determines the order in which the signals should be
+  handled.  In many cases this order wasn't being followed, leading to the wrong
+  Python exception being raised.
 
 - Issue #7865: The close() method of :mod:`io` objects should not swallow
-  exceptions raised by the implicit flush().  Also ensure that calling
-  close() several times is supported.  Patch by Pascal Chambon.
+  exceptions raised by the implicit flush().  Also qensure that calling close()
+  several times is supported.  Patch by Pascal Chambon.
 
 - Issue #4687: Fix accuracy of garbage collection runtimes displayed with
   gc.DEBUG_STATS.
 
-- Issue #8354: The siginterrupt setting is now preserved for all signals,
-  not just SIGCHLD.
+- Issue #8354: The siginterrupt setting is now preserved for all signals, not
+  just SIGCHLD.
 
 - Issue #7192: webbrowser.get("firefox") now works on Mac OS X, as does
   webbrowser.get("safari").
 
-- Issue #8464: tarfile no longer creates files with execute permissions set
-  when mode="w|" is used.
+- Issue #8464: tarfile no longer creates files with execute permissions set when
+  mode="w|" is used.
 
-- Issue #7834: Fix connect() of Bluetooth L2CAP sockets with recent versions
-  of the Linux kernel.  Patch by Yaniv Aknin.
+- Issue #7834: Fix connect() of Bluetooth L2CAP sockets with recent versions of
+  the Linux kernel.  Patch by Yaniv Aknin.
 
 - Issue #8295: Added shutil.unpack_archive.
 
@@ -845,36 +826,36 @@
 - Issue #6656: fix locale.format_string to handle escaped percents
   and mappings.
 
-- Issue #2302: Fix a race condition in SocketServer.BaseServer.shutdown,
-  where the method could block indefinitely if called just before the
-  event loop started running.  This also fixes the occasional freezes
-  witnessed in test_httpservers.
-
-- Issue #8524: When creating an SSL socket, the timeout value of the
-  original socket wasn't retained (instead, a socket with a positive timeout
-  would be turned into a non-blocking SSL socket).
+- Issue #2302: Fix a race condition in SocketServer.BaseServer.shutdown, where
+  the method could block indefinitely if called just before the event loop
+  started running.  This also fixes the occasional freezes witnessed in
+  test_httpservers.
+
+- Issue #8524: When creating an SSL socket, the timeout value of the original
+  socket wasn't retained (instead, a socket with a positive timeout would be
+  turned into a non-blocking SSL socket).
 
 - Issue #5103: SSL handshake would ignore the socket timeout and block
   indefinitely if the other end didn't respond.
 
-- The do_handshake() method of SSL objects now adjusts the blocking mode of
-  the SSL structure if necessary (as other methods already do).
+- The do_handshake() method of SSL objects now adjusts the blocking mode of the
+  SSL structure if necessary (as other methods already do).
 
-- Issue #8391: os.execvpe() and os.getenv() supports unicode with surrogates
-  and bytes strings for environment keys and values
+- Issue #8391: os.execvpe() and os.getenv() supports unicode with surrogates and
+  bytes strings for environment keys and values.
 
 - Issue #8467: Pure Python implementation of subprocess encodes the error
   message using surrogatepass error handler to support surrogates in the
-  message
+  message.
 
-- Issue #8468: bz2.BZ2File() accepts str with surrogates and bytes filenames
+- Issue #8468: bz2.BZ2File() accepts str with surrogates and bytes filenames.
 
-- Issue #8451: Syslog module now uses basename(sys.argv[0]) instead of
-  the string "python" as the *ident*.  openlog() arguments are all optional
-  and keywords.
+- Issue #8451: Syslog module now uses basename(sys.argv[0]) instead of the
+  string "python" as the *ident*.  openlog() arguments are all optional and
+  keywords.
 
-- Issue #8108: Fix the unwrap() method of SSL objects when the socket has
-  a non-infinite timeout.  Also make that method friendlier with applications
+- Issue #8108: Fix the unwrap() method of SSL objects when the socket has a
+  non-infinite timeout.  Also make that method friendlier with applications
   wanting to continue using the socket in clear-text mode, by disabling
   OpenSSL's internal readahead.  Thanks to Darryl Miles for guidance.
 
@@ -884,25 +865,24 @@
 - Issue #8195: Fix a crash in sqlite Connection.create_collation() if the
   collation name contains a surrogate character.
 
-- Issue #8484: Load all ciphers and digest algorithms when initializing
-  the _ssl extension, such that verification of some SSL certificates
-  doesn't fail because of an "unknown algorithm".
+- Issue #8484: Load all ciphers and digest algorithms when initializing the _ssl
+  extension, such that verification of some SSL certificates doesn't fail
+  because of an "unknown algorithm".
 
 - Issue #6547: Added the ignore_dangling_symlinks option to shutil.copytree.
 
-- Issue #1540112: Now allowing the choice of a copy function in
-  shutil.copytree.
+- Issue #1540112: Now allowing the choice of a copy function in shutil.copytree.
 
 - Issue #4814: timeout parameter is now applied also for connections resulting
   from PORT/EPRT commands.
 
 - Issue #8463: added missing reference to bztar in shutil's documentation.
 
-- Issue #7154: urllib.request can now detect the proxy settings on OSX 10.6
-  (as long as the user didn't specify 'automatic proxy configuration').
+- Issue #7154: urllib.request can now detect the proxy settings on OSX 10.6 (as
+  long as the user didn't specify 'automatic proxy configuration').
 
-- Issue #3817: ftplib.FTP.abort() method now considers 225 a valid response
-  code as stated in RFC-959 at chapter 5.4.
+- Issue #3817: ftplib.FTP.abort() method now considers 225 a valid response code
+  as stated in RFC-959 at chapter 5.4.
 
 - Issue #8394: _ctypes.dlopen() accepts bytes, bytearray and str with
   surrogates.
@@ -917,95 +897,92 @@
   the current working directory.
 
 - Issue #7606: XML-RPC traceback stored in X-traceback is now encoded to ASCII
-  using backslashreplace error handler
+  using backslashreplace error handler.
 
-- Issue #8412: os.system() now accepts bytes, bytearray and str with
-  surrogates.
+- Issue #8412: os.system() now accepts bytes, bytearray and str with surrogates.
 
 - Issue #2987: RFC2732 support for urlparse (IPv6 addresses). Patch by Tony
   Locke and Hans Ulrich Niedermann.
 
 - Issue #5277: Fix quote counting when parsing RFC 2231 encoded parameters.
 
-- Issue #7316: the acquire() method of lock objects in the :mod:`threading`
+- Issue #7316: The acquire() method of lock objects in the :mod:`threading`
   module now takes an optional timeout argument in seconds.  Timeout support
-  relies on the system threading library, so as to avoid a semi-busy wait
-  loop.
+  relies on the system threading library, so as to avoid a semi-busy wait loop.
 
 - Issue #8383: pickle and pickletools use surrogatepass error handler when
   encoding unicode as utf8 to support lone surrogates and stay compatible with
-  Python 2.x and 3.0
+  Python 2.x and 3.x.
 
 - Issue #7585: difflib context and unified diffs now place a tab between
-  filename and date, conforming to the 'standards' they were originally
-  designed to follow.  This improves compatibility with patch tools.
+  filename and date, conforming to the 'standards' they were originally designed
+  to follow.  This improves compatibility with patch tools.
 
 - Issue #7472: Fixed typo in email.encoders module; messages using ISO-2022
-  character sets will now consistently use a Content-Transfer-Encoding of
-  7bit rather than sometimes being marked as 8bit.
+  character sets will now consistently use a Content-Transfer-Encoding of 7bit
+  rather than sometimes being marked as 8bit.
 
 - Issue #8375: test_distutils now checks if the temporary directory are still
   present before it cleans them.
 
-- Issue #8374: Update the internal alias table in the :mod:`locale` module
-  to cover recent locale changes and additions.
+- Issue #8374: Update the internal alias table in the :mod:`locale` module to
+  cover recent locale changes and additions.
 
 - Issue #8321: Give access to OpenSSL version numbers from the `ssl` module,
-  using the new attributes `ssl.OPENSSL_VERSION`, `ssl.OPENSSL_VERSION_INFO`
-  and `ssl.OPENSSL_VERSION_NUMBER`.
+  using the new attributes `ssl.OPENSSL_VERSION`, `ssl.OPENSSL_VERSION_INFO` and
+  `ssl.OPENSSL_VERSION_NUMBER`.
 
 - Add functools.total_ordering() and functools.cmp_to_key().
 
-- Issue #8257: The Decimal construct now accepts a float instance
-  directly, converting that float to a Decimal of equal value:
+- Issue #8257: The Decimal construct now accepts a float instance directly,
+  converting that float to a Decimal of equal value:
 
      >>> Decimal(1.1)
      Decimal('1.100000000000000088817841970012523233890533447265625')
 
-- Issue #8294: The Fraction constructor now accepts Decimal and float
-  instances directly.
+- Issue #8294: The Fraction constructor now accepts Decimal and float instances
+  directly.
 
-- Issue #7279: Comparisons involving a Decimal signaling NaN now
-  signal InvalidOperation instead of returning False.  (Comparisons
-  involving a quiet NaN are unchanged.)  Also, Decimal quiet NaNs
-  are now hashable;  Decimal signaling NaNs remain unhashable.
-
-- Issue #2531: Comparison operations between floats and Decimal
-  instances now return a result based on the numeric values of the
-  operands;  previously they returned an arbitrary result based on
-  the relative ordering of id(float) and id(Decimal).  See also
-  issue #8188, which adds Decimal-to-Fraction comparisons.
+- Issue #7279: Comparisons involving a Decimal signaling NaN now signal
+  InvalidOperation instead of returning False.  (Comparisons involving a quiet
+  NaN are unchanged.)  Also, Decimal quiet NaNs are now hashable; Decimal
+  signaling NaNs remain unhashable.
+
+- Issue #2531: Comparison operations between floats and Decimal instances now
+  return a result based on the numeric values of the operands; previously they
+  returned an arbitrary result based on the relative ordering of id(float) and
+  id(Decimal).  See also issue #8188, which adds Decimal-to-Fraction
+  comparisons.
 
 - Added a subtract() method to collections.Counter().
 
 - Issue #8233: When run as a script, py_compile.py optionally takes a single
-  argument `-` which tells it to read files to compile from stdin.  Each line
-  is read on demand and the named file is compiled immediately.  (Original
-  patch by Piotr O?arowski).
+  argument `-` which tells it to read files to compile from stdin.  Each line is
+  read on demand and the named file is compiled immediately.  (Original patch by
+  Piotr O?arowski).
 
 - Backwards incompatible change: Unicode codepoints line tabulation (0x0B) and
   form feed (0x0C) are now considered linebreaks, as specified in Unicode
-  Standard Annex #14.  See issue #7643.
-  http://www.unicode.org/reports/tr14/
+  Standard Annex #14.  See issue #7643.  http://www.unicode.org/reports/tr14/
 
-- Comparisons using one of <, <=, >, >= between a complex instance and
-  a Fractions instance now raise TypeError instead of returning
-  True/False.  This makes Fraction <=> complex comparisons consistent with
-  int <=> complex, float <=> complex, and complex <=> complex comparisons.
-
-- Issue #8139: ossaudiodev didn't initialize its types properly, therefore
-  some methods (such as oss_mixer_device.fileno()) were not available.
-  Initial patch by Bertrand Janin.
+- Comparisons using one of <, <=, >, >= between a complex instance and a
+  Fractions instance now raise TypeError instead of returning True/False.  This
+  makes Fraction <=> complex comparisons consistent with int <=> complex, float
+  <=> complex, and complex <=> complex comparisons.
+
+- Issue #8139: ossaudiodev didn't initialize its types properly, therefore some
+  methods (such as oss_mixer_device.fileno()) were not available.  Initial patch
+  by Bertrand Janin.
 
 - Issue #8205: Remove the "Modules" directory from sys.path when Python is
   running from the build directory (POSIX only).
 
 - Issue #7512: shutil.copystat() could raise an OSError when the filesystem
-  didn't support chflags() (for example ZFS under FreeBSD).  The error is
-  now silenced.
+  didn't support chflags() (for example ZFS under FreeBSD).  The error is now
+  silenced.
 
-- Issue #7860: platform.uname now reports the correct 'machine' type
-  when Python is running in WOW64 mode on 64 bit Windows.
+- Issue #7860: platform.uname now reports the correct 'machine' type when Python
+  is running in WOW64 mode on 64 bit Windows.
 
 - Issue #3890, #8222: Fix recv() and recv_into() on non-blocking SSL sockets.
   Also, enable the SSL_MODE_AUTO_RETRY flag on SSL sockets, so that blocking
@@ -1029,31 +1006,31 @@
 - Issue #4961: Inconsistent/wrong result of askyesno function in tkMessageBox
   with Tcl/Tk-8.5.
 
-- Issue #8140: extend compileall to compile single files. Add -i option.
+- Issue #8140: extend compileall to compile single files.  Add -i option.
 
-- Issue #7356: ctypes.util: Make parsing of ldconfig output independent of
-  the locale.
+- Issue #7356: ctypes.util: Make parsing of ldconfig output independent of the
+  locale.
 
-- The internals of the subprocess module on POSIX systems have been replaced
-  by an extension module (_posixsubprocess) so that the fork()+exec() can be
-  done safely without the possibility of deadlock in multithreaded applications.
-
-- subprocess.Popen now has restore_signals and start_new_session features.
-  The default of restore_signals=True is a new behavior compared to earlier
-  Python versions.  This means that signals such as SIGPIPE are not ignored
-  by default in subprocesses launched by Python (Issue #1652).
+- The internals of the subprocess module on POSIX systems have been replaced by
+  an extension module (_posixsubprocess) so that the fork()+exec() can be done
+  safely without the possibility of deadlock in multithreaded applications.
+
+- subprocess.Popen now has restore_signals and start_new_session features.  The
+  default of restore_signals=True is a new behavior compared to earlier Python
+  versions.  This means that signals such as SIGPIPE are not ignored by default
+  in subprocesses launched by Python (Issue #1652).
 
 - Issue #6472: The xml.etree package is updated to ElementTree 1.3.  The
   cElementTree module is updated too.
 
 - Issue #7774: Set sys.executable to an empty string if argv[0] has been set to
-  an non existent program name and Python is unable to retrieve the real
-  program name
+  an non existent program name and Python is unable to retrieve the real program
+  name.
 
 - Issue #7880: Fix sysconfig when the python executable is a symbolic link.
 
-- Issue #6509: fix re.sub to work properly when the pattern, the string, and
-  the replacement were all bytes. Patch by Antoine Pitrou.
+- Issue #6509: fix re.sub to work properly when the pattern, the string, and the
+  replacement were all bytes.  Patch by Antoine Pitrou.
 
 - The sqlite3 module was updated to pysqlite 2.6.0. This fixes several obscure
   bugs and allows loading SQLite extensions from shared libraries.
@@ -1083,16 +1060,16 @@
 
 - Issue #1537721: Add a writeheader() method to csv.DictWriter.
 
-- Issue #7959: ctypes callback functions are now registered correctly
-  with the cycle garbage collector.
+- Issue #7959: ctypes callback functions are now registered correctly with the
+  cycle garbage collector.
 
 - Issue #5801: removed spurious empty lines in wsgiref.
 
-- Issue #6666: fix bug in trace.py that applied the list of directories
-  to be ignored only to the first file.  Noted by Bogdan Opanchuk.
+- Issue #6666: fix bug in trace.py that applied the list of directories to be
+  ignored only to the first file.  Noted by Bogdan Opanchuk.
 
-- Issue #7597: curses.use_env() can now be called before initscr().
-  Noted by Kan-Ru Chen.
+- Issue #7597: curses.use_env() can now be called before initscr().  Noted by
+  Kan-Ru Chen.
 
 - Issue #7310: fix the __repr__ of os.environ to show the environment variables.
 
@@ -1102,48 +1079,46 @@
 - Issue #7361: Importlib was not properly checking the number of bytes in
   bytecode file when it was less then 8 bytes.
 
-- Issue #7633: In the decimal module, Context class methods (with the
-  exception of canonical and is_canonical) now accept instances of int
-  and long wherever a Decimal instance is accepted, and implicitly
-  convert that argument to Decimal.  Previously only some arguments
-  were converted.
+- Issue #7633: In the decimal module, Context class methods (with the exception
+  of canonical and is_canonical) now accept instances of int and long wherever a
+  Decimal instance is accepted, and implicitly convert that argument to Decimal.
+  Previously only some arguments were converted.
 
 - Issue #7835: shelve should no longer produce mysterious warnings during
   interpreter shutdown.
 
-- Issue #2746: Don't escape ampersands and angle brackets ("&", "<", ">")
-  in XML processing instructions and comments.  These raw characters are
-  allowed by the XML specification, and are necessary when outputting e.g.
-  PHP code in a processing instruction.  Patch by Neil Muller.
-
-- Issue #6233: ElementTree failed converting unicode characters to XML
-  entities when they could't be represented in the requested output
-  encoding.  Patch by Jerry Chen.
+- Issue #2746: Don't escape ampersands and angle brackets ("&", "<", ">") in XML
+  processing instructions and comments.  These raw characters are allowed by the
+  XML specification, and are necessary when outputting e.g.  PHP code in a
+  processing instruction.  Patch by Neil Muller.
+
+- Issue #6233: ElementTree failed converting unicode characters to XML entities
+  when they could't be represented in the requested output encoding.  Patch by
+  Jerry Chen.
 
-- Issue #6003: add an argument to ``zipfile.Zipfile.writestr`` to
-  specify the compression type.
+- Issue #6003: add an argument to ``zipfile.Zipfile.writestr`` to specify the
+  compression type.
 
 - Issue #4772: Raise a ValueError when an unknown Bluetooth protocol is
   specified, rather than fall through to AF_PACKET (in the `socket` module).
-  Also, raise ValueError rather than TypeError when an unknown TIPC address
-  type is specified.  Patch by Brian Curtin.
+  Also, raise ValueError rather than TypeError when an unknown TIPC address type
+  is specified.  Patch by Brian Curtin.
 
-- Issue #6939: Fix file I/O objects in the `io` module to keep the original
-  file position when calling `truncate()`.  It would previously change the
-  file position to the given argument, which goes against the tradition of
+- Issue #6939: Fix file I/O objects in the `io` module to keep the original file
+  position when calling `truncate()`.  It would previously change the file
+  position to the given argument, which goes against the tradition of
   ftruncate() and other truncation APIs.  Patch by Pascal Chambon.
 
 - Issue #7610: Reworked implementation of the internal
-  :class:`zipfile.ZipExtFile` class used to represent files stored inside
-  an archive.  The new implementation is significantly faster and can
-  be wrapped in a :class:`io.BufferedReader` object for more speedups.
-  It also solves an issue where interleaved calls to `read()` and
-  `readline()` give wrong results.  Patch by Nir Aides.
+  :class:`zipfile.ZipExtFile` class used to represent files stored inside an
+  archive.  The new implementation is significantly faster and can be wrapped in
+  a :class:`io.BufferedReader` object for more speedups.  It also solves an
+  issue where interleaved calls to `read()` and `readline()` give wrong results.
+  Patch by Nir Aides.
 
 - Issue #6963: Added "maxtasksperchild" argument to multiprocessing.Pool,
-  allowing for a maximum number of tasks within the pool to be completed by
-  the worker before that worker is terminated, and a new one created to
-  replace it.
+  allowing for a maximum number of tasks within the pool to be completed by the
+  worker before that worker is terminated, and a new one created to replace it.
 
 - Issue #7792: Registering non-classes to ABCs raised an obscure error.
 
@@ -1157,21 +1132,21 @@
 - Issue #7561: Fix crashes when using bytearray objects with the posix
   module.
 
-- Issue #1670765: Prevent email.generator.Generator from re-wrapping
-  headers in multipart/signed MIME parts, which fixes one of the sources of
-  invalid modifications to such parts by Generator.
+- Issue #1670765: Prevent email.generator.Generator from re-wrapping headers in
+  multipart/signed MIME parts, which fixes one of the sources of invalid
+  modifications to such parts by Generator.
 
-- Issue #7703: Add support for the new buffer API to `binascii.a2bhqx`.
-  Patch by Florent Xicluna, along with some additional tests.
+- Issue #7703: Add support for the new buffer API to `binascii.a2bhqx`.  Patch
+  by Florent Xicluna, along with some additional tests.
 
-- Issue #7701: Fix crash in binascii.b2a_uu() in debug mode when given a
-  1-byte argument.  Patch by Victor Stinner.
+- Issue #7701: Fix crash in binascii.b2a_uu() in debug mode when given a 1-byte
+  argument.  Patch by Victor Stinner.
 
-- Issue #3299: Fix possible crash in the _sre module when given bad
-  argument values in debug mode.  Patch by Victor Stinner.
+- Issue #3299: Fix possible crash in the _sre module when given bad argument
+  values in debug mode.  Patch by Victor Stinner.
 
-- Issue #2846: Add support for gzip.GzipFile reading zero-padded files.
-  Patch by Brian Curtin.
+- Issue #2846: Add support for gzip.GzipFile reading zero-padded files.  Patch
+  by Brian Curtin.
 
 - Issue #7681: Use floor division in appropiate places in the wave module.
 
@@ -1179,63 +1154,61 @@
   Extension extra options may change the output without changing the .c
   file). Initial patch by Collin Winter.
 
-- Issue #7617: Make sure distutils.unixccompiler.UnixCCompiler recognizes
-  gcc when it has a fully qualified configuration prefix. Initial patch
-  by Arfrever.
+- Issue #7617: Make sure distutils.unixccompiler.UnixCCompiler recognizes gcc
+  when it has a fully qualified configuration prefix. Initial patch by Arfrever.
 
-- Issue #7105: Make WeakKeyDictionary and WeakValueDictionary robust against
-  the destruction of weakref'ed objects while iterating.
+- Issue #7105: Make WeakKeyDictionary and WeakValueDictionary robust against the
+  destruction of weakref'ed objects while iterating.
 
-- Issue #7455: Fix possible crash in cPickle on invalid input.  Patch by
-  Victor Stinner.
+- Issue #7455: Fix possible crash in cPickle on invalid input.  Patch by Victor
+  Stinner.
 
 - Issue #1628205: Socket file objects returned by socket.socket.makefile() now
-  properly handles EINTR within the read, readline, write & flush methods.
-  The socket.sendall() method now properly handles interrupted system calls.
+  properly handles EINTR within the read, readline, write & flush methods.  The
+  socket.sendall() method now properly handles interrupted system calls.
 
-- Issue #7471: Improve the performance of GzipFile's buffering mechanism,
-  and make it implement the `io.BufferedIOBase` ABC to allow for further
-  speedups by wrapping it in an `io.BufferedReader`.  Patch by Nir Aides.
+- Issue #7471: Improve the performance of GzipFile's buffering mechanism, and
+  make it implement the `io.BufferedIOBase` ABC to allow for further speedups by
+  wrapping it in an `io.BufferedReader`.  Patch by Nir Aides.
 
 - Issue #3972: http.client.HTTPConnection now accepts an optional source_address
   parameter to allow specifying where your connections come from.
 
 - socket.create_connection now accepts an optional source_address parameter.
 
-- Issue #5511: now zipfile.ZipFile can be used as a context manager.
-  Initial patch by Brian Curtin.
+- Issue #5511: now zipfile.ZipFile can be used as a context manager.  Initial
+  patch by Brian Curtin.
 
-- Issue #7556: Make sure Distutils' msvc9compile reads and writes the
-  MSVC XML Manifest file in text mode so string patterns can be used
-  in regular expressions.
-
-- Issue #7552: Removed line feed in the base64 Authorization header in
-  the Distutils upload command to avoid an error when PyPI reads it.
-  This occurs on long passwords. Initial patch by JP St. Pierre.
+- Issue #7556: Make sure Distutils' msvc9compile reads and writes the MSVC XML
+  Manifest file in text mode so string patterns can be used in regular
+  expressions.
+
+- Issue #7552: Removed line feed in the base64 Authorization header in the
+  Distutils upload command to avoid an error when PyPI reads it.  This occurs on
+  long passwords. Initial patch by JP St. Pierre.
 
 - Issue #7231: urllib2 cannot handle https with proxy requiring auth.  Patch by
   Tatsuhiro Tsujikawa.
 
-- Issue #4757: `zlib.compress` and other methods in the zlib module now
-  raise a TypeError when given an `str` object (rather than a `bytes`-like
-  object).  Patch by Victor Stinner and Florent Xicluna.
+- Issue #4757: `zlib.compress` and other methods in the zlib module now raise a
+  TypeError when given an `str` object (rather than a `bytes`-like object).
+  Patch by Victor Stinner and Florent Xicluna.
 
 - Issue #7349: Make methods of file objects in the io module accept None as an
   argument where file-like objects (ie StringIO and BytesIO) accept them to mean
   the same as passing no argument.
 
-- Issue #7357: tarfile no longer suppresses fatal extraction errors by
-  default.
+- Issue #7357: tarfile no longer suppresses fatal extraction errors by default.
 
-- Issue #5949: added check for correct lineends in input from IMAP server
-  in imaplib.
+- Issue #5949: added check for correct lineends in input from IMAP server in
+  imaplib.
 
 - Add count() and reverse() methods to collections.deque().
 
 - Fix variations of extending deques:  d.extend(d)  d.extendleft(d)  d+=d
 
-- Issue #6986: Fix crash in the JSON C accelerator when called with the
-  wrong parameter types.  Patch by Victor Stinner.
+- Issue #6986: Fix crash in the JSON C accelerator when called with the wrong
+  parameter types.  Patch by Victor Stinner.
 
 - Issue #7457: added a read_pkg_file method to
   distutils.dist.DistributionMetadata.
@@ -1243,26 +1216,26 @@
 - logging: Added optional `secure` parameter to SMTPHandler, to enable use of
   TLS with authentication credentials.
 
-- Issue #1923: Fixed the removal of meaningful spaces when PKG-INFO is
-  generated in Distutils. Patch by Stephen Emslie.
+- Issue #1923: Fixed the removal of meaningful spaces when PKG-INFO is generated
+  in Distutils.  Patch by Stephen Emslie.
 
 - Issue #4120: Drop reference to CRT from manifest when building extensions with
   msvc9compiler.
 
 - Issue #7333: The `posix` module gains an `initgroups()` function providing
-  access to the initgroups(3) C library call on Unix systems which implement
-  it.  Patch by Jean-Paul Calderone.
+  access to the initgroups(3) C library call on Unix systems which implement it.
+  Patch by Jean-Paul Calderone.
 
 - Issue #7408: Fixed distutils.tests.sdist so it doesn't check for group
   ownership when the group is not forced, because the group may be different
   from the user's group and inherit from its container when the test is run.
 
 - Issue #4486: When an exception has an explicit cause, do not print its
-  implicit context too.  This affects the `traceback` module as well as
-  built-in exception printing.
+  implicit context too.  This affects the `traceback` module as well as built-in
+  exception printing.
 
-- Issue #1515: Enable use of deepcopy() with instance methods.  Patch by
-  Robert Collins.
+- Issue #1515: Enable use of deepcopy() with instance methods.  Patch by Robert
+  Collins.
 
 - Issue #7403: logging: Fixed possible race condition in lock creation.
 
@@ -1270,50 +1243,48 @@
   `storbinary()` method of FTP and FTP_TLS objects gains an optional `rest`
   argument.  Patch by Pablo Mouzo.
 
-- Issue #5788: `datetime.timedelta` objects get a new `total_seconds()`
-  method returning the total number of seconds in the duration.  Patch by
-  Brian Quinlan.
+- Issue #5788: `datetime.timedelta` objects get a new `total_seconds()` method
+  returning the total number of seconds in the duration.  Patch by Brian
+  Quinlan.
 
 - Issue #7133: SSL objects now support the new buffer API.
 
-- Issue #1488943: difflib.Differ() doesn't always add hints for tab characters
+- Issue #1488943: difflib.Differ() doesn't always add hints for tab characters.
 
 - Issue #6123: tarfile now opens empty archives correctly and consistently
   raises ReadError on empty files.
 
-- Issue #7354: distutils.tests.test_msvc9compiler - dragfullwindows can
-  be 2.
+- Issue #7354: distutils.tests.test_msvc9compiler - dragfullwindows can be 2.
 
 - Issue #5037: Proxy the __bytes__ special method instead to __bytes__ instead
   of __str__.
 
-- Issue #7341: Close the internal file object in the TarFile constructor in
-  case of an error.
+- Issue #7341: Close the internal file object in the TarFile constructor in case
+  of an error.
 
 - Issue #7293: distutils.test_msvc9compiler is fixed to work on any fresh
   Windows box. Help provided by David Bolen.
 
-- Issue #2054: ftplib now provides an FTP_TLS class to do secure FTP using
-  TLS or SSL.  Patch by Giampaolo Rodola'.
+- Issue #2054: ftplib now provides an FTP_TLS class to do secure FTP using TLS
+  or SSL.  Patch by Giampaolo Rodola'.
 
-- Issue #7328: pydoc no longer corrupts sys.path when run with the '-m' switch
+- Issue #7328: pydoc no longer corrupts sys.path when run with the '-m' switch.
 
-- Issue #4969: The mimetypes module now reads the MIME database from
-  the registry under Windows.  Patch by Gabriel Genellina.
+- Issue #4969: The mimetypes module now reads the MIME database from the
+  registry under Windows.  Patch by Gabriel Genellina.
 
-- Issue #6816: runpy now provides a run_path function that allows Python code
-  to execute file paths that refer to source or compiled Python files as well
-  as zipfiles, directories and other valid sys.path entries that contain a
-  __main__.py file. This allows applications that run other Python scripts to
+- Issue #6816: runpy now provides a run_path function that allows Python code to
+  execute file paths that refer to source or compiled Python files as well as
+  zipfiles, directories and other valid sys.path entries that contain a
+  __main__.py file.  This allows applications that run other Python scripts to
   support the same flexibility as the CPython command line itself.
 
-- Issue #7318: multiprocessing now uses a timeout when it fails to establish
-  a connection with another process, rather than looping endlessly. The
-  default timeout is 20 seconds, which should be amply sufficient for
-  local connections.
+- Issue #7318: multiprocessing now uses a timeout when it fails to establish a
+  connection with another process, rather than looping endlessly.  The default
+  timeout is 20 seconds, which should be amply sufficient for local connections.
 
 - Issue #7197: Allow unittest.TextTestRunner objects to be pickled and
-  unpickled. This fixes crashes under Windows when trying to run
+  unpickled.  This fixes crashes under Windows when trying to run
   test_multiprocessing in verbose mode.
 
 - Issue #7893: ``unittest.TextTestResult`` is made public and a ``resultclass``
@@ -1324,12 +1295,12 @@
   name in failure reports even if the test has a docstring.
 
 - Issue #3001: Add a C implementation of recursive locks which is used by
-  default when instantiating a `threading.RLock` object. This makes
-  recursive locks as fast as regular non-recursive locks (previously,
-  they were slower by 10x to 15x).
+  default when instantiating a `threading.RLock` object. This makes recursive
+  locks as fast as regular non-recursive locks (previously, they were slower by
+  10x to 15x).
 
-- Issue #7282: Fix a memory leak when an RLock was used in a thread other
-  than those started through `threading.Thread` (for example, using
+- Issue #7282: Fix a memory leak when an RLock was used in a thread other than
+  those started through `threading.Thread` (for example, using
   `_thread.start_new_thread()`).
 
 - Issue #7187: Importlib would not silence the IOError raised when trying to
@@ -1341,20 +1312,19 @@
 - Issue #7211: Allow 64-bit values for the `ident` and `data` fields of kevent
   objects on 64-bit systems.  Patch by Michael Broghton.
 
-- Issue #6896: mailbox.Maildir now invalidates its internal cache each time
-  a modification is done through it.  This fixes inconsistencies and test
-  failures on systems with slightly bogus mtime behaviour.
-
-- Issue #7246 & Issue #7208: getpass now properly flushes input before
-  reading from stdin so that existing input does not confuse it and
-  lead to incorrect entry or an IOError.  It also properly flushes it
-  afterwards to avoid the terminal echoing the input afterwards on
-  OSes such as Solaris.
-
-- Issue #7233: Fix a number of two-argument Decimal methods to make
-  sure that they accept an int or long as the second argument.  Also
-  fix buggy handling of large arguments (those with coefficient longer
-  than the current precision) in shift and rotate.
+- Issue #6896: mailbox.Maildir now invalidates its internal cache each time a
+  modification is done through it.  This fixes inconsistencies and test failures
+  on systems with slightly bogus mtime behaviour.
+
+- Issue #7246 & Issue #7208: getpass now properly flushes input before reading
+  from stdin so that existing input does not confuse it and lead to incorrect
+  entry or an IOError.  It also properly flushes it afterwards to avoid the
+  terminal echoing the input afterwards on OSes such as Solaris.
+
+- Issue #7233: Fix a number of two-argument Decimal methods to make sure that
+  they accept an int or long as the second argument.  Also fix buggy handling of
+  large arguments (those with coefficient longer than the current precision) in
+  shift and rotate.
 
 - Issue #4750: Store the basename of the original filename in the gzip FNAME
   header as required by RFC 1952.
@@ -1376,31 +1346,31 @@
 - Issue #7080: locale.strxfrm() raises a MemoryError on 64-bit non-Windows
   platforms, and assorted locale fixes by Derk Drukker.
 
-- Issue #5833: Fix extra space character in readline completion with the
-  GNU readline library version 6.0.
+- Issue #5833: Fix extra space character in readline completion with the GNU
+  readline library version 6.0.
 
-- Issue #6894: Fixed the issue urllib2 doesn't respect "no_proxy" environment
+- Issue #6894: Fixed the issue urllib2 doesn't respect "no_proxy" environment.
 
 - Issue #7086: Added TCP support to SysLogHandler, and tidied up some
   anachronisms in the code which were a relic of 1.5.2 compatibility.
 
-- Issue #7082: When falling back to the MIME 'name' parameter, the
-  correct place to look for it is the Content-Type header.
+- Issue #7082: When falling back to the MIME 'name' parameter, the correct place
+  to look for it is the Content-Type header.
 
 - Make tokenize.detect_coding() normalize utf-8 and iso-8859-1 variants like the
   builtin tokenizer.
 
-- Issue #7048: Force Decimal.logb to round its result when that result
-  is too large to fit in the current precision.
+- Issue #7048: Force Decimal.logb to round its result when that result is too
+  large to fit in the current precision.
 
-- Issue #6236, #6348: Fix various failures in the I/O library under AIX
-  and other platforms, when using a non-gcc compiler. Patch by Derk Drukker.
+- Issue #6236, #6348: Fix various failures in the I/O library under AIX and
+  other platforms, when using a non-gcc compiler. Patch by Derk Drukker.
 
-- Issue #4606: Passing 'None' if ctypes argtype is set to POINTER(...)
-  does now always result in NULL.
+- Issue #4606: Passing 'None' if ctypes argtype is set to POINTER(...)  does now
+  always result in NULL.
 
-- Issue #5042: Structure sub-subclass does now initialize correctly
-  with base class positional arguments.
+- Issue #5042: Structure sub-subclass does now initialize correctly with base
+  class positional arguments.
 
 - Issue #6882: Import uuid creates zombies processes.
 
@@ -1410,22 +1380,21 @@
 
 - Issue #6888: pdb's alias command was broken when no arguments were given.
 
-- Issue #6857: Default format() alignment should be '>' for Decimal
-  instances.
+- Issue #6857: Default format() alignment should be '>' for Decimal instances.
 
-- Issue #6795: int(Decimal('nan')) now raises ValueError instead of
-  returning NaN or raising InvalidContext.  Also, fix infinite recursion
-  in long(Decimal('nan')).
+- Issue #6795: int(Decimal('nan')) now raises ValueError instead of returning
+  NaN or raising InvalidContext.  Also, fix infinite recursion in
+  long(Decimal('nan')).
 
-- Issue #6850: Fix bug in Decimal._parse_format_specifier for formats
-  with no type specifier.
+- Issue #6850: Fix bug in Decimal._parse_format_specifier for formats with no
+  type specifier.
 
 - Issue #6239: ctypes.c_char_p return value must return bytes.
 
-- Issue #6838: Use a list to accumulate the value instead of
-  repeatedly concatenating strings in http.client's
-  HTTPResponse._read_chunked providing a significant speed increase
-  when downloading large files servend with a Transfer-Encoding of 'chunked'.
+- Issue #6838: Use a list to accumulate the value instead of repeatedly
+  concatenating strings in http.client's HTTPResponse._read_chunked providing a
+  significant speed increase when downloading large files servend with a
+  Transfer-Encoding of 'chunked'.
 
 - Trying to import a submodule from a module that is not a package, ImportError
   should be raised, not AttributeError.
@@ -1434,8 +1403,7 @@
   fall back to computing what __package__ should be instead of giving up.
 
 - Raise a TypeError when the name of a module to be imported for
-  importlib.__import__ is not a string (was raising an
-  AttributeError before).
+  importlib.__import__ is not a string (was raising an AttributeError before).
 
 - Allow the fromlist passed into importlib.__import__ to be any iterable.
 
@@ -1457,50 +1425,50 @@
 
 - Issue #6665: Fix fnmatch to properly match filenames with newlines in them.
 
-- Issue #1135: Add the XView and YView mix-ins to avoid duplicating
-  the xview* and yview* methods.
+- Issue #1135: Add the XView and YView mix-ins to avoid duplicating the xview*
+  and yview* methods.
 
 - Issue #6629: Fix a data corruption issue in the new I/O library, which could
   occur when writing to a BufferedRandom object (e.g. a file opened in "rb+" or
   "wb+" mode) after having buffered a certain amount of data for reading. This
   bug was not present in the pure Python implementation.
 
-- Issue #6622: Fix "local variable 'secret' referenced before
-  assignment" bug in POP3.apop.
+- Issue #6622: Fix "local variable 'secret' referenced before assignment" bug in
+  POP3.apop.
 
 - Issue #2715: Remove remnants of Carbon.File from binhex module.
 
-- Issue #6595: The Decimal constructor now allows arbitrary Unicode
-  decimal digits in input, as recommended by the standard.  Previously
-  it was restricted to accepting [0-9].
+- Issue #6595: The Decimal constructor now allows arbitrary Unicode decimal
+  digits in input, as recommended by the standard.  Previously it was restricted
+  to accepting [0-9].
 
 - Issue #6106: telnetlib.Telnet.process_rawq doesn't handle default WILL/WONT
   DO/DONT correctly.
 
 - Issue #1424152: Fix for http.client, urllib.request to support SSL while
-  working through proxy. Original patch by Christopher Li, changes made by
-  Senthil Kumaran
+  working through proxy.  Original patch by Christopher Li, changes made by
+  Senthil Kumaran.
 
 - Add importlib.abc.ExecutionLoader to represent the PEP 302 protocol for
-  loaders that allow for modules to be executed. Both importlib.abc.PyLoader
-  and PyPycLoader inherit from this class and provide implementations in
-  relation to other methods required by the ABCs.
+  loaders that allow for modules to be executed. Both importlib.abc.PyLoader and
+  PyPycLoader inherit from this class and provide implementations in relation to
+  other methods required by the ABCs.
 
 - importlib.abc.PyLoader did not inherit from importlib.abc.ResourceLoader like
   the documentation said it did even though the code in PyLoader relied on the
   abstract method required by ResourceLoader.
 
-- Issue #6431: Make Fraction type return NotImplemented when it doesn't
-  know how to handle a comparison without loss of precision.  Also add
-  correct handling of infinities and nans for comparisons with float.
+- Issue #6431: Make Fraction type return NotImplemented when it doesn't know how
+  to handle a comparison without loss of precision.  Also add correct handling
+  of infinities and nans for comparisons with float.
 
 - Issue #6415: Fixed warnings.warn segfault on bad formatted string.
 
-- Issue #6358: The exit status of a command started with os.popen() was
-  reported differently than it did with python 2.x.
+- Issue #6358: The exit status of a command started with os.popen() was reported
+  differently than it did with python 2.x.
 
-- Issue #6323: The pdb debugger did not exit when running a script with a
-  syntax error.
+- Issue #6323: The pdb debugger did not exit when running a script with a syntax
+  error.
 
 - Issue #3392: The subprocess communicate() method no longer fails in select()
   when file descriptors are large; communicate() now uses poll() when possible.
@@ -1516,21 +1484,21 @@
 
 - Issue #6218: io.StringIO and io.BytesIO instances are now picklable.
 
-- The os.get_exec_path() function to return the list of directories that will
-  be searched for an executable when launching a subprocess was added.
+- The os.get_exec_path() function to return the list of directories that will be
+  searched for an executable when launching a subprocess was added.
 
 - Issue #7481: When a threading.Thread failed to start it would leave the
   instance stuck in initial state and present in threading.enumerate().
 
-- Issue #1068268: The subprocess module now handles EINTR in internal
-  os.waitpid and os.read system calls where appropriate.
+- Issue #1068268: The subprocess module now handles EINTR in internal os.waitpid
+  and os.read system calls where appropriate.
 
 - Issue #6729: Added ctypes.c_ssize_t to represent ssize_t.
 
 - Issue #6247: The argparse module has been added to the standard library.
 
-- Issue #8235: _socket: Add the constant ``SO_SETFIB``.  SO_SETFIB is
-  a socket option available on FreeBSD 7.1 and newer.
+- Issue #8235: _socket: Add the constant ``SO_SETFIB``.  SO_SETFIB is a socket
+  option available on FreeBSD 7.1 and newer.
 
 Extension Modules
 -----------------
@@ -1538,27 +1506,25 @@
 - Issue #9422: Fix memory leak when re-initializing a struct.Struct object.
 
 - Issue #7900: The getgroups(2) system call on MacOSX behaves rather oddly
-  compared to other unix systems. In particular, os.getgroups() does
-  not reflect any changes made using os.setgroups() but basicly always
-  returns the same information as the id command.
-
-  os.getgroups() can now return more than 16 groups on MacOSX.
+  compared to other unix systems. In particular, os.getgroups() does not reflect
+  any changes made using os.setgroups() but basicly always returns the same
+  information as the id command. os.getgroups() can now return more than 16
+  groups on MacOSX.
 
 - Issue #6095: Make directory argument to os.listdir optional.
 
-- Issue #9277: Fix bug in struct.pack for bools in standard mode
-  (e.g., struct.pack('>?')):  if conversion to bool raised an exception
-  then that exception wasn't properly propagated on machines where
-  char is unsigned.
+- Issue #9277: Fix bug in struct.pack for bools in standard mode (e.g.,
+  struct.pack('>?')): if conversion to bool raised an exception then that
+  exception wasn't properly propagated on machines where char is unsigned.
 
-- Issue #5180: Fixed a bug that prevented loading 2.x pickles in 3.x
-  python when they contain instances of old-style classes.
+- Issue #5180: Fixed a bug that prevented loading 2.x pickles in 3.x python when
+  they contain instances of old-style classes.
 
-- Issue #9165: Add new functions math.isfinite and cmath.isfinite, to
-  accompany existing isinf and isnan functions.
+- Issue #9165: Add new functions math.isfinite and cmath.isfinite, to accompany
+  existing isinf and isnan functions.
 
-- Issue #1578269: Implement os.symlink for Windows 6.0+. Patch by
-  Jason R. Coombs
+- Issue #1578269: Implement os.symlink for Windows 6.0+.  Patch by Jason
+  R. Coombs.
 
 - In struct.pack, correctly propogate exceptions from computing the truth of an
   object in the '?' format.
@@ -1568,58 +1534,56 @@
 - In the math module, correctly lookup __trunc__, __ceil__, and __floor__ as
   special methods.
 
-- Issue #9005: Prevent utctimetuple() from producing year 0 or year
-  10,000.  Prior to this change, timezone adjustment in utctimetuple()
-  could produce tm_year value of 0 or 10,000.  Now an OverflowError is
-  raised in these edge cases.
-
-- Issue #6641: The ``datetime.strptime`` method now supports the
-  ``%z`` directive.  When the ``%z`` directive is present in the
-  format string, an aware ``datetime`` object is returned with
-  ``tzinfo`` bound to a ``datetime.timezone`` instance constructed
-  from the parsed offset.  If both ``%z`` and ``%Z`` are present, the
-  data in ``%Z`` field is used for timezone name, but ``%Z`` data
-  without ``%z`` is discarded.
+- Issue #9005: Prevent utctimetuple() from producing year 0 or year 10,000.
+  Prior to this change, timezone adjustment in utctimetuple() could produce
+  tm_year value of 0 or 10,000.  Now an OverflowError is raised in these edge
+  cases.
+
+- Issue #6641: The ``datetime.strptime`` method now supports the ``%z``
+  directive.  When the ``%z`` directive is present in the format string, an
+  aware ``datetime`` object is returned with ``tzinfo`` bound to a
+  ``datetime.timezone`` instance constructed from the parsed offset.  If both
+  ``%z`` and ``%Z`` are present, the data in ``%Z`` field is used for timezone
+  name, but ``%Z`` data without ``%z`` is discarded.
 
 - Issue #5094: The ``datetime`` module now has a simple concrete class
-  implementing ``datetime.tzinfo`` interface.  Instances of the new
-  class, ``datetime.timezone``, return fixed name and UTC offset from
-  their ``tzname(dt)`` and ``utcoffset(dt)`` methods.  The ``dst(dt)``
-  method always returns ``None``.  A class attribute, ``utc`` contains
-  an instance representing the UTC timezone.  Original patch by Rafe
-  Kaplan.
+  implementing ``datetime.tzinfo`` interface.  Instances of the new class,
+  ``datetime.timezone``, return fixed name and UTC offset from their
+  ``tzname(dt)`` and ``utcoffset(dt)`` methods.  The ``dst(dt)`` method always
+  returns ``None``.  A class attribute, ``utc`` contains an instance
+  representing the UTC timezone.  Original patch by Rafe Kaplan.
 
-- Issue #8973: Add __all__ to struct module; this ensures that
-  help(struct) includes documentation for the struct.Struct class.
+- Issue #8973: Add __all__ to struct module; this ensures that help(struct)
+  includes documentation for the struct.Struct class.
 
 - Issue #3129: Trailing digits in struct format string are no longer ignored.
   For example, "1" or "ilib123" are now invalid formats and cause
   ``struct.error`` to be raised.  Patch by Caleb Deveraux.
 
-- Issue #7384: If the system readline library is linked against ncurses,
-  the curses module must be linked against ncurses as well. Otherwise it
-  is not safe to load both the readline and curses modules in an application.
-
-- Issue #2810: Fix cases where the Windows registry API returns
-  ERROR_MORE_DATA, requiring a re-try in order to get the complete result.
-
-- Issue #8692: Optimize math.factorial: replace the previous naive
-  algorithm with an improved 'binary-split' algorithm that uses fewer
-  multiplications and allows many of the multiplications to be
-  performed using plain C integer arithmetic instead of PyLong
-  arithmetic.  Also uses a lookup table for small arguments.
+- Issue #7384: If the system readline library is linked against ncurses, the
+  curses module must be linked against ncurses as well. Otherwise it is not safe
+  to load both the readline and curses modules in an application.
+
+- Issue #2810: Fix cases where the Windows registry API returns ERROR_MORE_DATA,
+  requiring a re-try in order to get the complete result.
+
+- Issue #8692: Optimize math.factorial: replace the previous naive algorithm
+  with an improved 'binary-split' algorithm that uses fewer multiplications and
+  allows many of the multiplications to be performed using plain C integer
+  arithmetic instead of PyLong arithmetic.  Also uses a lookup table for small
+  arguments.
 
 - Issue #8674: Fixed a number of incorrect or undefined-behaviour-inducing
   overflow checks in the audioop module.
 
 - Issue #8644: The accuracy of td.total_seconds() has been improved (by
   calculating with integer arithmetic instead of float arithmetic internally):
-  the result is now always correctly rounded, and is equivalent to td /
-  timedelta(seconds=1).
+  the result is now always correctly rounded, and is equivalent to ``td /
+  timedelta(seconds=1)``.
 
-- Issue #2706: Allow division of a timedelta by another timedelta:
-  timedelta / timedelta, timedelta % timedelta, timedelta // timedelta
-  and divmod(timedelta, timedelta) are all supported.
+- Issue #2706: Allow division of a timedelta by another timedelta: timedelta /
+  timedelta, timedelta % timedelta, timedelta // timedelta and divmod(timedelta,
+  timedelta) are all supported.
 
 - Issue #8314: Fix unsigned long long bug in libffi on Sparc v8.
 
@@ -1632,7 +1596,7 @@
 
 - Issue #6949: Allow the _dbm extension to be built with db 4.8.x.
 
-- Issue #6544: fix a reference leak in the kqueue implementation's error
+- Issue #6544: Fix a reference leak in the kqueue implementation's error
   handling.
 
 - Stop providing crtassem.h symbols when compiling with Visual Studio 2010, as
@@ -1644,37 +1608,36 @@
 
 - Issue #3366: Add erf, erfc, expm1, gamma, lgamma functions to math module.
 
-- Issue #6877: It is now possible to link the readline extension to the
-  libedit readline emulation on OSX 10.5 or later.
+- Issue #6877: It is now possible to link the readline extension to the libedit
+  readline emulation on OSX 10.5 or later.
 
 - Issue #6848: Fix curses module build failure on OS X 10.6.
 
 - Fix a segfault that could be triggered by expat with specially formed input.
 
-- Issue #6561: '\d' in a regex now matches only characters with
-  Unicode category 'Nd' (Number, Decimal Digit).  Previously it also
-  matched characters with category 'No'.
-
-- Issue #4509: Array objects are no longer modified after an operation
-  failing due to the resize restriction in-place when the object has exported
-  buffers.
+- Issue #6561: '\d' in a regex now matches only characters with Unicode category
+  'Nd' (Number, Decimal Digit).  Previously it also matched characters with
+  category 'No'.
+
+- Issue #4509: Array objects are no longer modified after an operation failing
+  due to the resize restriction in-place when the object has exported buffers.
 
 - Issue #2389: Array objects are now pickled in a portable manner.
 
 - Expat: Fix DoS via XML document with malformed UTF-8 sequences
   (CVE_2009_3560).
 
-- Issue #7242: On Solaris 9 and earlier calling os.fork() from within a
-  thread could raise an incorrect RuntimeError about not holding the import
-  lock.  The import lock is now reinitialized after fork.
+- Issue #7242: On Solaris 9 and earlier calling os.fork() from within a thread
+  could raise an incorrect RuntimeError about not holding the import lock.  The
+  import lock is now reinitialized after fork.
 
 - Issue #7999: os.setreuid() and os.setregid() would refuse to accept a -1
   parameter on some platforms such as OS X.
 
 - Build the ossaudio extension on GNU/kFreeBSD.
 
-- Issue #7347: winreg: Add CreateKeyEx and DeleteKeyEx, as well as fix a
-  bug in the return value of QueryReflectionKey.
+- Issue #7347: winreg: Add CreateKeyEx and DeleteKeyEx, as well as fix a bug in
+  the return value of QueryReflectionKey.
 
 - Issue #7567: PyCurses_setupterm: Don't call `setupterm' twice.
 
@@ -1685,43 +1648,42 @@
 
 - Issue #9280: Make sharedinstall depend on sharedmods.
 
-- Issue #9189: Make a user-specified CFLAGS, CPPFLAGS, or LDFLAGS
-  setting override the configure and makefile defaults, without
-  deleting options the user didn't intend to override.  Developers
-  should no longer need to specify OPT or EXTRA_CFLAGS, although those
-  variables are still present for backward-compatibility.
+- Issue #9189: Make a user-specified CFLAGS, CPPFLAGS, or LDFLAGS setting
+  override the configure and makefile defaults, without deleting options the
+  user didn't intend to override.  Developers should no longer need to specify
+  OPT or EXTRA_CFLAGS, although those variables are still present for
+  backward-compatibility.
 
 - Issue #8854: Fix finding Visual Studio 2008 on Windows x64.
 
 - Issue #1759169, #8864: Drop _XOPEN_SOURCE on Solaris, define it for
   multiprocessing only.
 
-- Issue #8625: Turn off optimization in --with-pydebug builds with
-  gcc.  (Optimization was unintentionally turned on in gcc
-  --with-pydebug builds as a result of the issue #1628484 fix,
-  combined with autoconf's strange choice of default CFLAGS produced
-  by AC_PROG_CC for gcc.)
+- Issue #8625: Turn off optimization in --with-pydebug builds with gcc.
+  (Optimization was unintentionally turned on in gcc --with-pydebug builds as a
+  result of the issue #1628484 fix, combined with autoconf's strange choice of
+  default CFLAGS produced by AC_PROG_CC for gcc.)
 
-- Issue #3646: It is now easily possible to install a Python framework into
-  your home directory on MacOSX, see Mac/README for more information.
+- Issue #3646: It is now easily possible to install a Python framework into your
+  home directory on MacOSX, see Mac/README for more information.
 
 - Issue #3928: os.mknod() now available in Solaris, also.
 
 - Issue #3326: Build Python without -fno-strict-aliasing when the gcc does not
   give false warnings.
 
-- Issue #1628484: The Makefile doesn't ignore the CFLAGS environment
-  variable anymore.  It also forwards the LDFLAGS settings to the linker
-  when building a shared library.
+- Issue #1628484: The Makefile doesn't ignore the CFLAGS environment variable
+  anymore.  It also forwards the LDFLAGS settings to the linker when building a
+  shared library.
 
-- Issue #6716: Quote -x arguments of compileall in MSI installer.
-  Exclude 2to3 tests from compileall.
+- Issue #6716: Quote -x arguments of compileall in MSI installer.  Exclude 2to3
+  tests from compileall.
 
 - Issue #3920, #7903: Define _BSD_SOURCE on OpenBSD 4.4 through 4.9.
 
-- Issue #7632: When Py_USING_MEMORY_DEBUGGER is defined, disable the
-  private memory allocation scheme in dtoa.c and use PyMem_Malloc and
-  PyMem_Free instead.  Also disable caching of powers of 5.
+- Issue #7632: When Py_USING_MEMORY_DEBUGGER is defined, disable the private
+  memory allocation scheme in dtoa.c and use PyMem_Malloc and PyMem_Free
+  instead.  Also disable caching of powers of 5.
 
 - Issue #6491: Allow --with-dbmliborder to specify that no dbms will be built.
 
@@ -1740,21 +1702,21 @@
 - Issue #5792: Extend the short float repr support to x86 systems using
   icc or suncc.
 
-- Issue #6603: Change READ_TIMESTAMP macro in ceval.c so that it
-  compiles correctly under gcc on x86-64.  This fixes a reported
-  problem with the --with-tsc build on x86-64.
+- Issue #6603: Change READ_TIMESTAMP macro in ceval.c so that it compiles
+  correctly under gcc on x86-64.  This fixes a reported problem with the
+  --with-tsc build on x86-64.
 
-- Issue #6802: Fix build issues on MacOSX 10.6
+- Issue #6802: Fix build issues on MacOSX 10.6.
 
 - Issue #6244: Allow detect_tkinter to look for Tcl/Tk 8.6.
 
 - Issue 4601: 'make install' did not set the appropriate permissions on
   directories.
 
-- Issue 5390: Add uninstall icon independent of whether file
-  extensions are installed.
+- Issue 5390: Add uninstall icon independent of whether file extensions are
+  installed.
 
-- Issue #7541: when using ``python-config`` with a framework install the
+- Issue #7541: When using ``python-config`` with a framework install the
   compiler might use the wrong library.
 
 - python-config now supports multiple options on the same command line.
@@ -1763,24 +1725,23 @@
 
 - Issue #8510: Update to autoconf2.65.
 
-
 Documentation
 ------------
 
 - Issue 9255: Document that the 'test' package is meant for internal Python use
   only.
 
-- A small wsgi server was added as Tools/scripts/serve.py, and is used to
+- A small WSGI server was added as Tools/scripts/serve.py, and is used to
   implement a local documentation server via 'make serve' in the doc directory.
 
-- Updating `Using Python` documentation to include description of CPython's
-  -J and -X options.
+- Updating `Using Python` documentation to include description of CPython's -J
+  and -X options.
 
 - Document that importing a module that has None in sys.modules triggers an
   ImportError.
 
-- Issue #6556: Fixed the Distutils configuration files location explanation
-  for Windows.
+- Issue #6556: Fixed the Distutils configuration files location explanation for
+  Windows.
 
 - Update python manual page (options -B, -O0, -s, environment variables
   PYTHONDONTWRITEBYTECODE, PYTHONNOUSERSITE).
@@ -1791,19 +1752,19 @@
 Tests
 -----
 
-- Issue #9251: test_threaded_import didn't fail when run through regrtest
-  if the import lock was disabled.
+- Issue #9251: test_threaded_import didn't fail when run through regrtest if the
+  import lock was disabled.
 
 - Issue #8605: Skip test_gdb if Python is compiled with optimizations.
 
-- Issue #7449: Skip test_socketserver if threading support is disabled
+- Issue #7449: Skip test_socketserver if threading support is disabled.
 
-- Issue #8672: Add a zlib test ensuring that an incomplete stream can be
-  handled by a decompressor object without errors (it returns incomplete
-  uncompressed data).
+- Issue #8672: Add a zlib test ensuring that an incomplete stream can be handled
+  by a decompressor object without errors (it returns incomplete uncompressed
+  data).
 
 - Issue #8533: regrtest uses backslashreplace error handler for stdout to avoid
-  UnicodeEncodeError (write non-ASCII character to stdout using ASCII encoding)
+  UnicodeEncodeError (write non-ASCII character to stdout using ASCII encoding).
 
 - Issue #8576: Remove use of find_unused_port() in test_smtplib and
   test_multiprocessing.  Patch by Paul Moore.
@@ -1811,14 +1772,14 @@
 - Issue #7449: Fix many tests to support Python compiled without thread
   support. Patches written by Jerry Seutter.
 
-- Issue #8108: test_ftplib's non-blocking SSL server now has proper handling
-  of SSL shutdowns.
+- Issue #8108: test_ftplib's non-blocking SSL server now has proper handling of
+  SSL shutdowns.
 
-- Issues #8279, #8330, #8437, #8480, #8495: Fix test_gdb failures, patch
-  written by Dave Malcolm
+- Issues #8279, #8330, #8437, #8480, #8495: Fix test_gdb failures, patch written
+  by Dave Malcolm.
 
-- Issue #3864: Skip three test_signal tests on freebsd6 because they fail
-  if any thread was previously started, most likely due to a platform bug.
+- Issue #3864: Skip three test_signal tests on freebsd6 because they fail if any
+  thread was previously started, most likely due to a platform bug.
 
 - Issue #8193: Fix test_zlib failure with zlib 1.2.4.
 
@@ -1830,37 +1791,37 @@
 - Issue #8180 and #8207: Fix test_pep277 on OS X and add more tests for special
   Unicode normalization cases.
 
-- Issue #7783: test.support.open_urlresource invalidates the outdated files
-  from the local cache.
+- Issue #7783: test.support.open_urlresource invalidates the outdated files from
+  the local cache.
 
 - Issue #7849: Now the utility ``check_warnings`` verifies if the warnings are
   effectively raised.
 
-- The four path modules (genericpath, macpath, ntpath, posixpath) share a
-  common TestCase for some tests: test_genericpath.CommonTest.
+- The four path modules (genericpath, macpath, ntpath, posixpath) share a common
+  TestCase for some tests: test_genericpath.CommonTest.
 
-- Print platform information when running the whole test suite, or using
-  the --verbose flag.
+- Print platform information when running the whole test suite, or using the
+  --verbose flag.
 
 - Issue #767675: enable test_pep277 on POSIX platforms with Unicode-friendly
   filesystem encoding.
 
-- Issue #6292: for the moment at least, the test suite runs cleanly if python
-  is run with the -OO flag.  Tests requiring docstrings are skipped.
+- Issue #6292: for the moment at least, the test suite runs cleanly if python is
+  run with the -OO flag.  Tests requiring docstrings are skipped.
 
-- Issue #7712: test.support gained a new `temp_cwd` context manager which is
-  now also used by regrtest to run all the tests in a temporary directory.
-  The original CWD is saved in `support.SAVEDCWD`.
-  Thanks to Florent Xicluna who helped with the patch.
-
-- Issue #7924: Fix an intermittent 'XXX undetected error' failure in
-  test_capi (only seen so far on platforms where the curses module
-  wasn't built), due to an uncleared exception.
+- Issue #7712: test.support gained a new `temp_cwd` context manager which is now
+  also used by regrtest to run all the tests in a temporary directory.  The
+  original CWD is saved in `support.SAVEDCWD`.  Thanks to Florent Xicluna who
+  helped with the patch.
+
+- Issue #7924: Fix an intermittent 'XXX undetected error' failure in test_capi
+  (only seen so far on platforms where the curses module wasn't built), due to
+  an uncleared exception.
 
 - Issue #7728: test_timeout was changed to use support.bind_port instead of a
   hard coded port.
 
-- Issue #7376: instead of running a self-test (which was failing) when called
+- Issue #7376: Instead of running a self-test (which was failing) when called
   with no arguments, doctest.py now gives a usage message.
 
 - Issue #7396: fix regrtest -s, which was broken by the -j enhancement.
@@ -1868,20 +1829,20 @@
 - Issue #7498: test_multiprocessing now uses test.support.find_unused_port
   instead of a hardcoded port number in test_rapid_restart.
 
-- Issue #7431: use TESTFN in test_linecache instead of trying to create a
-  file in the Lib/test directory, which might be read-only for the
-  user running the tests.
-
-- Issue #7324: add a sanity check to regrtest argument parsing to
-  catch the case of an option with no handler.
-
-- Issue #7312: Add a -F flag to run the selected tests in a loop until
-  a test fails.  Can be combined with -j.
-
-- Issue #6551: test_zipimport could import and then destroy some modules of
-  the encodings package, which would make other tests fail further down
-  the road because the internally cached encoders and decoders would point
-  to empty global variables.
+- Issue #7431: Use TESTFN in test_linecache instead of trying to create a file
+  in the Lib/test directory, which might be read-only for the user running the
+  tests.
+
+- Issue #7324: Add a sanity check to regrtest argument parsing to catch the case
+  of an option with no handler.
+
+- Issue #7312: Add a -F flag to run the selected tests in a loop until a test
+  fails.  Can be combined with -j.
+
+- Issue #6551: test_zipimport could import and then destroy some modules of the
+  encodings package, which would make other tests fail further down the road
+  because the internally cached encoders and decoders would point to empty
+  global variables.
 
 - Issue #7295: Do not use a hardcoded file name in test_tarfile.
 
@@ -1897,29 +1858,29 @@
   calling methods on the object.
 
 - Issue #7222: Make thread "reaping" more reliable so that reference
-  leak-chasing test runs give sensible results. The previous method of
-  reaping threads could return successfully while some Thread objects were
-  still referenced. This also introduces a new private function:
+  leak-chasing test runs give sensible results. The previous method of reaping
+  threads could return successfully while some Thread objects were still
+  referenced. This also introduces a new private function:
   :func:`_thread._count()`.
 
-- Issue #7151: fixed regrtest -j so that output to stderr from a test no
-  longer runs the risk of causing the worker thread to fail.
+- Issue #7151: Fixed regrtest -j so that output to stderr from a test no longer
+  runs the risk of causing the worker thread to fail.
 
 - Issue #7055: test___all__ now greedily detects all modules which have an
   __all__ attribute, rather than using a hardcoded and incomplete list.
 
 - Issue #7058: Added save/restore for things like sys.argv and cwd to
-  runtest_inner in regrtest, with warnings if the called test modifies them,
-  and a new section in the summary report at the end.
+  runtest_inner in regrtest, with warnings if the called test modifies them, and
+  a new section in the summary report at the end.
 
 - Issue #7042: Fix test_signal (test_itimer_virtual) failure on OS X 10.6.
 
-- Fixed tests in importlib.test.source.test_abc_loader that were masking
-  the proper exceptions that should be raised for missing or improper code
-  object bytecode.
+- Fixed tests in importlib.test.source.test_abc_loader that were masking the
+  proper exceptions that should be raised for missing or improper code object
+  bytecode.
 
 - Removed importlib's custom test discovery code and switched to
-    unittest.TestLoader.discover().
+  unittest.TestLoader.discover().
 
 Tools/Demos
 -----------
@@ -1927,8 +1888,7 @@
 - Issue #5464, #8974: Implement plural forms in msgfmt.py.
 
 - iobench (a file I/O benchmark) and ccbench (a concurrency benchmark) were
-  added to the `Tools/` directory.  They were previously living in the
-  sandbox.
+  added to the `Tools/` directory.  They were previously living in the sandbox.
 
 
 What's New in Python 3.1?

Modified: python/branches/py3k/Misc/RPM/python-3.2.spec
==============================================================================
--- python/branches/py3k/Misc/RPM/python-3.2.spec	(original)
+++ python/branches/py3k/Misc/RPM/python-3.2.spec	Sat Jul 31 10:56:11 2010
@@ -39,8 +39,8 @@
 
 %define name python
 #--start constants--
-%define version 3.2a0
-%define libvers 3.2
+%define version 3.2a1
+%define libver 3.2
 #--end constants--
 %define release 1pydotorg
 %define __prefix /usr

Modified: python/branches/py3k/README
==============================================================================
--- python/branches/py3k/README	(original)
+++ python/branches/py3k/README	Sat Jul 31 10:56:11 2010
@@ -1,5 +1,5 @@
-This is Python version 3.2
-==========================
+This is Python version 3.2 alpha 1
+==================================
 
 Copyright (c) 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010
 Python Software Foundation.
@@ -53,9 +53,9 @@
 ----------
 
 We try to have a comprehensive overview of the changes in the "What's New in
-Python 3.1" document, found at
+Python 3.2" document, found at
 
-    http://docs.python.org/dev/3.1/whatsnew/3.1.html
+    http://docs.python.org/dev/3.2/whatsnew/3.2.html
 
 For a more detailed change log, read Misc/NEWS (though this file, too,
 is incomplete, and also doesn't list anything merged in from the 2.7
@@ -68,30 +68,27 @@
 Documentation
 -------------
 
-Documentation for Python 3.1 is online, updated twice a day:
+Documentation for Python 3.2 is online, updated daily:
 
-    http://docs.python.org/dev/3.1/
+    http://docs.python.org/dev/
 
-All documentation is also available online at the Python web site
-(http://docs.python.org/, see below).  It is available online for
-occasional reference, or can be downloaded in many formats for faster
-access.  The documentation is downloadable in HTML, PostScript, PDF,
-LaTeX (through 2.5), and reStructuredText (2.6+) formats; the LaTeX and
-reStructuredText versions are primarily for documentation authors,
-translators, and people with special formatting requirements.
+It can also be downloaded in many formats for faster access.  The documentation
+is downloadable in HTML, PDF, and reStructuredText formats; the latter version
+is primarily for documentation authors, translators, and people with special
+formatting requirements.
 
 
 Converting From Python 2.x to 3.x
 ---------------------------------
 
-Python starting with 2.6 will contain features to help locating code that
+Python starting with 2.6 contains features to help locating code that
 needs to be changed, such as optional warnings when deprecated features are
 used, and backported versions of certain key Python 3.x features.
 
 A source-to-source translation tool, "2to3", can take care of the mundane task
 of converting large amounts of source code.  It is not a complete solution but
 is complemented by the deprecation warnings in 2.6.  See
-http://docs.python.org/dev/py3k/library/2to3.html for more information.
+http://docs.python.org/dev/library/2to3.html for more information.
 
 
 Testing
@@ -135,7 +132,7 @@
 version (if any) is your "primary" version.  Install that version using
 "make install".  Install all other versions using "make altinstall".
 
-For example, if you want to install Python 2.5, 2.6 and 3.0 with 2.6 being
+For example, if you want to install Python 2.5, 2.6 and 3.2 with 2.6 being
 the primary version, you would execute "make install" in your 2.6 build
 directory and "make altinstall" in the others.
 

From python-checkins at python.org  Sat Jul 31 10:56:55 2010
From: python-checkins at python.org (florent.xicluna)
Date: Sat, 31 Jul 2010 10:56:55 +0200 (CEST)
Subject: [Python-checkins] r83319 - in python/branches/py3k/Lib:
	test/test_urllib.py urllib/parse.py
Message-ID: <20100731085655.7BA2FEEAAE@mail.python.org>

Author: florent.xicluna
Date: Sat Jul 31 10:56:55 2010
New Revision: 83319

Log:
Fix an oversight in r83294.  unquote() should reject bytes.  Issue #9301.


Modified:
   python/branches/py3k/Lib/test/test_urllib.py
   python/branches/py3k/Lib/urllib/parse.py

Modified: python/branches/py3k/Lib/test/test_urllib.py
==============================================================================
--- python/branches/py3k/Lib/test/test_urllib.py	(original)
+++ python/branches/py3k/Lib/test/test_urllib.py	Sat Jul 31 10:56:55 2010
@@ -557,6 +557,7 @@
                          "%s" % result)
         self.assertRaises((TypeError, AttributeError), urllib.parse.unquote, None)
         self.assertRaises((TypeError, AttributeError), urllib.parse.unquote, ())
+        self.assertRaises((TypeError, AttributeError), urllib.parse.unquote, b'')
 
     def test_unquoting_badpercent(self):
         # Test unquoting on bad percent-escapes

Modified: python/branches/py3k/Lib/urllib/parse.py
==============================================================================
--- python/branches/py3k/Lib/urllib/parse.py	(original)
+++ python/branches/py3k/Lib/urllib/parse.py	Sat Jul 31 10:56:55 2010
@@ -338,7 +338,7 @@
 
     unquote('abc%20def') -> 'abc def'.
     """
-    if string in (b'', ''):
+    if string == '':
         return string
     res = string.split('%')
     if len(res) == 1:

From python-checkins at python.org  Sat Jul 31 11:01:16 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 11:01:16 +0200 (CEST)
Subject: [Python-checkins] r83320 - in python/branches/py3k: Doc/README.txt
	Doc/license.rst Doc/make.bat LICENSE PC/python_nt.rc
	Python/getcopyright.c
Message-ID: <20100731090116.673D0EEA75@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 11:01:16 2010
New Revision: 83320

Log:
Update copyright years and add releases to release list.  Also update Sphinx version number.

Modified:
   python/branches/py3k/Doc/README.txt
   python/branches/py3k/Doc/license.rst
   python/branches/py3k/Doc/make.bat
   python/branches/py3k/LICENSE
   python/branches/py3k/PC/python_nt.rc
   python/branches/py3k/Python/getcopyright.c

Modified: python/branches/py3k/Doc/README.txt
==============================================================================
--- python/branches/py3k/Doc/README.txt	(original)
+++ python/branches/py3k/Doc/README.txt	Sat Jul 31 11:01:16 2010
@@ -78,7 +78,7 @@
 
 You'll need to install the Sphinx package, either by checking it out via ::
 
-   svn co http://svn.python.org/projects/external/Sphinx-0.6.5/sphinx tools/sphinx
+   svn co http://svn.python.org/projects/external/Sphinx-1.0.1/sphinx tools/sphinx
 
 or by installing it from PyPI.
 
@@ -132,7 +132,7 @@
 as long as you don't change or remove the copyright notice:
 
 ----------------------------------------------------------------------
-Copyright (c) 2000-2008 Python Software Foundation.
+Copyright (c) 2000-2010 Python Software Foundation.
 All rights reserved.
 
 Copyright (c) 2000 BeOpen.com.

Modified: python/branches/py3k/Doc/license.rst
==============================================================================
--- python/branches/py3k/Doc/license.rst	(original)
+++ python/branches/py3k/Doc/license.rst	Sat Jul 31 11:01:16 2010
@@ -106,6 +106,10 @@
 +----------------+--------------+------------+------------+-----------------+
 | 3.1.1          | 3.1          | 2009       | PSF        | yes             |
 +----------------+--------------+------------+------------+-----------------+
+| 3.1.2          | 3.1          | 2010       | PSF        | yes             |
++----------------+--------------+------------+------------+-----------------+
+| 3.2            | 3.1          | 2010       | PSF        | yes             |
++----------------+--------------+------------+------------+-----------------+
 
 .. note::
 

Modified: python/branches/py3k/Doc/make.bat
==============================================================================
--- python/branches/py3k/Doc/make.bat	(original)
+++ python/branches/py3k/Doc/make.bat	Sat Jul 31 11:01:16 2010
@@ -34,7 +34,7 @@
 goto end
 
 :checkout
-svn co %SVNROOT%/external/Sphinx-0.6.5/sphinx tools/sphinx
+svn co %SVNROOT%/external/Sphinx-1.0.1/sphinx tools/sphinx
 svn co %SVNROOT%/external/docutils-0.6/docutils tools/docutils
 svn co %SVNROOT%/external/Jinja-2.3.1/jinja2 tools/jinja2
 svn co %SVNROOT%/external/Pygments-1.3.1/pygments tools/pygments

Modified: python/branches/py3k/LICENSE
==============================================================================
--- python/branches/py3k/LICENSE	(original)
+++ python/branches/py3k/LICENSE	Sat Jul 31 11:01:16 2010
@@ -68,6 +68,7 @@
     3.1             3.0.1       2009        PSF         yes
     3.1.1           3.1         2009        PSF         yes
     3.1.2           3.1         2010        PSF         yes
+    3.2             3.1         2010        PSF         yes
 
 Footnotes:
 

Modified: python/branches/py3k/PC/python_nt.rc
==============================================================================
--- python/branches/py3k/PC/python_nt.rc	(original)
+++ python/branches/py3k/PC/python_nt.rc	Sat Jul 31 11:01:16 2010
@@ -61,7 +61,7 @@
             VALUE "FileDescription", "Python Core\0"
             VALUE "FileVersion", PYTHON_VERSION
             VALUE "InternalName", "Python DLL\0"
-            VALUE "LegalCopyright", "Copyright ? 2001-2008 Python Software Foundation. Copyright ? 2000 BeOpen.com. Copyright ? 1995-2001 CNRI. Copyright ? 1991-1995 SMC.\0"
+            VALUE "LegalCopyright", "Copyright ? 2001-2010 Python Software Foundation. Copyright ? 2000 BeOpen.com. Copyright ? 1995-2001 CNRI. Copyright ? 1991-1995 SMC.\0"
             VALUE "OriginalFilename", PYTHON_DLL_NAME "\0"
             VALUE "ProductName", "Python\0"
             VALUE "ProductVersion", PYTHON_VERSION

Modified: python/branches/py3k/Python/getcopyright.c
==============================================================================
--- python/branches/py3k/Python/getcopyright.c	(original)
+++ python/branches/py3k/Python/getcopyright.c	Sat Jul 31 11:01:16 2010
@@ -2,7 +2,7 @@
 
 #include "Python.h"
 
-static char cprt[] = 
+static char cprt[] =
 "\
 Copyright (c) 2001-2010 Python Software Foundation.\n\
 All Rights Reserved.\n\

From python-checkins at python.org  Sat Jul 31 11:03:30 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 11:03:30 +0200 (CEST)
Subject: [Python-checkins] r83321 - python/branches/py3k/Lib/idlelib/NEWS.txt
Message-ID: <20100731090330.4530AEEA83@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 11:03:30 2010
New Revision: 83321

Log:
At least give IDLE 3.1 a release date.  No further entries there for 3.2 though.

Modified:
   python/branches/py3k/Lib/idlelib/NEWS.txt

Modified: python/branches/py3k/Lib/idlelib/NEWS.txt
==============================================================================
--- python/branches/py3k/Lib/idlelib/NEWS.txt	(original)
+++ python/branches/py3k/Lib/idlelib/NEWS.txt	Sat Jul 31 11:03:30 2010
@@ -1,7 +1,7 @@
-What's New in IDLE 3.1b1?
+What's New in IDLE 3.1?
 =========================
 
-*Release date: XX-XXX-09*
+*Release date: 27-Jun-09*
 
 - Use of 'filter' in keybindingDialog.py was causing custom key assignment to
   fail.  Patch 5707 amaury.forgeotdarc.

From python-checkins at python.org  Sat Jul 31 11:10:51 2010
From: python-checkins at python.org (tarek.ziade)
Date: Sat, 31 Jul 2010 11:10:51 +0200 (CEST)
Subject: [Python-checkins] r83322 - in python/branches/py3k/Doc/distutils:
	apiref.rst builtdist.rst commandref.rst examples.rst
	extending.rst setupscript.rst sourcedist.rst uploading.rst
Message-ID: <20100731091051.8491FEDFE@mail.python.org>

Author: tarek.ziade
Date: Sat Jul 31 11:10:51 2010
New Revision: 83322

Log:
reverted distutils doc to its 3.1 state

Modified:
   python/branches/py3k/Doc/distutils/apiref.rst
   python/branches/py3k/Doc/distutils/builtdist.rst
   python/branches/py3k/Doc/distutils/commandref.rst
   python/branches/py3k/Doc/distutils/examples.rst
   python/branches/py3k/Doc/distutils/extending.rst
   python/branches/py3k/Doc/distutils/setupscript.rst
   python/branches/py3k/Doc/distutils/sourcedist.rst
   python/branches/py3k/Doc/distutils/uploading.rst

Modified: python/branches/py3k/Doc/distutils/apiref.rst
==============================================================================
--- python/branches/py3k/Doc/distutils/apiref.rst	(original)
+++ python/branches/py3k/Doc/distutils/apiref.rst	Sat Jul 31 11:10:51 2010
@@ -21,9 +21,7 @@
 .. function:: setup(arguments)
 
    The basic do-everything function that does most everything you could ever ask
-   for from a Distutils method.
-
-   .. See XXXXX
+   for from a Distutils method. See XXXXX
 
    The setup function takes a large number of arguments. These are laid out in the
    following table.
@@ -149,11 +147,11 @@
 In addition, the :mod:`distutils.core` module exposed a number of  classes that
 live elsewhere.
 
-* :class:`~distutils.extension.Extension` from :mod:`distutils.extension`
+* :class:`Extension` from :mod:`distutils.extension`
 
-* :class:`~distutils.cmd.Command` from :mod:`distutils.cmd`
+* :class:`Command` from :mod:`distutils.cmd`
 
-* :class:`~distutils.dist.Distribution` from :mod:`distutils.dist`
+* :class:`Distribution` from :mod:`distutils.dist`
 
 A short description of each of these follows, but see the relevant module for
 the full reference.
@@ -997,7 +995,7 @@
    errors are ignored (apart from being reported to ``sys.stdout`` if *verbose* is
    true).
 
-.. XXX Some of this could be replaced with the shutil module?
+**\*\*** Some of this could be replaced with the shutil module? **\*\***
 
 
 :mod:`distutils.file_util` --- Single file operations
@@ -1313,7 +1311,8 @@
   the "negative alias" of :option:`--verbose`, then :option:`--quiet` on the
   command line sets *verbose* to false.
 
-.. XXX Should be replaced with :mod:`optparse`.
+**\*\*** Should be replaced with :mod:`optik` (which is also now known as
+:mod:`optparse` in Python 2.3 and later). **\*\***
 
 
 .. function:: fancy_getopt(options, negative_opt, object, args)
@@ -1681,8 +1680,8 @@
 ===================================================================
 
 .. module:: distutils.cmd
-   :synopsis: This module provides the abstract base class Command. This class
-              is subclassed by the modules in the distutils.command subpackage.
+   :synopsis: This module provides the abstract base class Command. This class is subclassed
+              by the modules in the distutils.command  subpackage.
 
 
 This module supplies the abstract base class :class:`Command`.
@@ -1692,84 +1691,20 @@
 
    Abstract base class for defining command classes, the "worker bees" of the
    Distutils.  A useful analogy for command classes is to think of them as
-   subroutines with local variables called *options*.  The options are declared
-   in :meth:`initialize_options` and defined (given their final values) in
-   :meth:`finalize_options`, both of which must be defined by every command
-   class.  The distinction between the two is necessary because option values
-   might come from the outside world (command line, config file, ...), and any
-   options dependent on other options must be computed after these outside
-   influences have been processed --- hence :meth:`finalize_options`.  The body
-   of the subroutine, where it does all its work based on the values of its
-   options, is the :meth:`run` method, which must also be implemented by every
-   command class.
+   subroutines with local variables called *options*.  The options are declared in
+   :meth:`initialize_options` and defined (given their final values) in
+   :meth:`finalize_options`, both of which must be defined by every command class.
+   The distinction between the two is necessary because option values might come
+   from the outside world (command line, config file, ...), and any options
+   dependent on other options must be computed after these outside influences have
+   been processed --- hence :meth:`finalize_options`.  The body of the subroutine,
+   where it does all its work based on the values of its options, is the
+   :meth:`run` method, which must also be implemented by every command class.
 
-   The class constructor takes a single argument *dist*, a :class:`Distribution`
+   The class constructor takes a single argument *dist*, a  :class:`Distribution`
    instance.
 
 
-Creating a new Distutils command
-================================
-
-This section outlines the steps to create a new Distutils command.
-
-A new command lives in a module in the :mod:`distutils.command` package. There
-is a sample template in that directory called :file:`command_template`.  Copy
-this file to a new module with the same name as the new command you're
-implementing.  This module should implement a class with the same name as the
-module (and the command).  So, for instance, to create the command
-``peel_banana`` (so that users can run ``setup.py peel_banana``), you'd copy
-:file:`command_template` to :file:`distutils/command/peel_banana.py`, then edit
-it so that it's implementing the class :class:`peel_banana`, a subclass of
-:class:`distutils.cmd.Command`.
-
-Subclasses of :class:`Command` must define the following methods.
-
-.. method:: Command.initialize_options()
-
-   Set default values for all the options that this command supports.  Note that
-   these defaults may be overridden by other commands, by the setup script, by
-   config files, or by the command-line.  Thus, this is not the place to code
-   dependencies between options; generally, :meth:`initialize_options`
-   implementations are just a bunch of ``self.foo = None`` assignments.
-
-
-.. method:: Command.finalize_options()
-
-   Set final values for all the options that this command supports. This is
-   always called as late as possible, ie.  after any option assignments from the
-   command-line or from other commands have been done.  Thus, this is the place
-   to to code option dependencies: if *foo* depends on *bar*, then it is safe to
-   set *foo* from *bar* as long as *foo* still has the same value it was
-   assigned in :meth:`initialize_options`.
-
-
-.. method:: Command.run()
-
-   A command's raison d'etre: carry out the action it exists to perform,
-   controlled by the options initialized in :meth:`initialize_options`,
-   customized by other commands, the setup script, the command-line, and config
-   files, and finalized in :meth:`finalize_options`.  All terminal output and
-   filesystem interaction should be done by :meth:`run`.
-
-
-.. attribute:: Command.sub_commands
-
-   *sub_commands* formalizes the notion of a "family" of commands,
-   e.g. ``install`` as the parent with sub-commands ``install_lib``,
-   ``install_headers``, etc.  The parent of a family of commands defines
-   *sub_commands* as a class attribute; it's a list of 2-tuples ``(command_name,
-   predicate)``, with *command_name* a string and *predicate* a function, a
-   string or ``None``.  *predicate* is a method of the parent command that
-   determines whether the corresponding command is applicable in the current
-   situation.  (E.g. we ``install_headers`` is only applicable if we have any C
-   header files to install.)  If *predicate* is ``None``, that command is always
-   applicable.
-
-   *sub_commands* is usually defined at the *end* of a class, because
-   predicates can be methods of the class, so they must already have been
-   defined.  The canonical example is the :command:`install` command.
-
-
 :mod:`distutils.command` --- Individual Distutils commands
 ==========================================================
 
@@ -2008,3 +1943,76 @@
 This is described in more detail in :pep:`301`.
 
 .. % todo
+
+:mod:`distutils.command.check` --- Check the meta-data of a package
+===================================================================
+
+.. module:: distutils.command.check
+   :synopsis: Check the metadata of a package
+
+
+The ``check`` command performs some tests on the meta-data of a package.
+For example, it verifies that all required meta-data are provided as
+the arguments passed to the :func:`setup` function.
+
+.. % todo
+
+
+Creating a new Distutils command
+================================
+
+This section outlines the steps to create a new Distutils command.
+
+A new command lives in a module in the :mod:`distutils.command` package. There
+is a sample template in that directory called  :file:`command_template`. Copy
+this file to a new module with the same name as the new command you're
+implementing. This module should implement a class with the same name as the
+module (and the command). So, for instance, to create the command
+``peel_banana`` (so that users can run ``setup.py peel_banana``), you'd copy
+:file:`command_template`  to :file:`distutils/command/peel_banana.py`, then edit
+it so that it's implementing the class :class:`peel_banana`, a subclass of
+:class:`distutils.cmd.Command`.
+
+Subclasses of :class:`Command` must define the following methods.
+
+
+.. method:: Command.initialize_options()
+
+   Set default values for all the options that this command supports.  Note that
+   these defaults may be overridden by other commands, by the setup script, by
+   config files, or by the command-line.  Thus, this is not the place to code
+   dependencies between options; generally, :meth:`initialize_options`
+   implementations are just a bunch of ``self.foo = None`` assignments.
+
+
+.. method:: Command.finalize_options()
+
+   Set final values for all the options that this command supports. This is
+   always called as late as possible, ie.  after any option assignments from the
+   command-line or from other commands have been done.  Thus, this is the place
+   to to code option dependencies: if *foo* depends on *bar*, then it is safe to
+   set *foo* from *bar* as long as *foo* still has the same value it was
+   assigned in :meth:`initialize_options`.
+
+
+.. method:: Command.run()
+
+   A command's raison d'etre: carry out the action it exists to perform, controlled
+   by the options initialized in :meth:`initialize_options`, customized by other
+   commands, the setup script, the command-line, and config files, and finalized in
+   :meth:`finalize_options`.  All terminal output and filesystem interaction should
+   be done by :meth:`run`.
+
+*sub_commands* formalizes the notion of a "family" of commands, eg. ``install``
+as the parent with sub-commands ``install_lib``, ``install_headers``, etc.  The
+parent of a family of commands defines *sub_commands* as a class attribute; it's
+a list of 2-tuples ``(command_name, predicate)``, with *command_name* a string
+and *predicate* a function, a string or None. *predicate* is a method of
+the parent command that determines whether the corresponding command is
+applicable in the current situation.  (Eg. we ``install_headers`` is only
+applicable if we have any C header files to install.)  If *predicate* is None,
+that command is always applicable.
+
+*sub_commands* is usually defined at the \*end\* of a class, because predicates
+can be methods of the class, so they must already have been defined.  The
+canonical example is the :command:`install` command.

Modified: python/branches/py3k/Doc/distutils/builtdist.rst
==============================================================================
--- python/branches/py3k/Doc/distutils/builtdist.rst	(original)
+++ python/branches/py3k/Doc/distutils/builtdist.rst	Sat Jul 31 11:10:51 2010
@@ -146,8 +146,8 @@
 Creating dumb built distributions
 =================================
 
-.. XXX Need to document absolute vs. prefix-relative packages here, but first
-       I have to implement it!
+**\*\*** Need to document absolute vs. prefix-relative packages here, but first
+I have to implement it! **\*\***
 
 
 .. _creating-rpms:
@@ -241,8 +241,7 @@
 configuration file, :file:`setup.cfg`\ ---see section :ref:`setup-config`.  If
 you distribute or package many Python module distributions, you might want to
 put options that apply to all of them in your personal Distutils configuration
-file (:file:`~/.pydistutils.cfg`).  If you want to temporarily disable
-this file, you can pass the --no-user-cfg option to setup.py.
+file (:file:`~/.pydistutils.cfg`).
 
 There are three steps to building a binary RPM package, all of which are
 handled automatically by the Distutils:

Modified: python/branches/py3k/Doc/distutils/commandref.rst
==============================================================================
--- python/branches/py3k/Doc/distutils/commandref.rst	(original)
+++ python/branches/py3k/Doc/distutils/commandref.rst	Sat Jul 31 11:10:51 2010
@@ -48,6 +48,50 @@
 .. % \label{clean-cmd}
 
 
+.. _sdist-cmd:
+
+Creating a source distribution: the :command:`sdist` command
+============================================================
+
+**\*\*** fragment moved down from above: needs context! **\*\***
+
+The manifest template commands are:
+
++-------------------------------------------+-----------------------------------------------+
+| Command                                   | Description                                   |
++===========================================+===============================================+
+| :command:`include pat1 pat2 ...`          | include all files matching any of the listed  |
+|                                           | patterns                                      |
++-------------------------------------------+-----------------------------------------------+
+| :command:`exclude pat1 pat2 ...`          | exclude all files matching any of the listed  |
+|                                           | patterns                                      |
++-------------------------------------------+-----------------------------------------------+
+| :command:`recursive-include dir pat1 pat2 | include all files under *dir* matching any of |
+| ...`                                      | the listed patterns                           |
++-------------------------------------------+-----------------------------------------------+
+| :command:`recursive-exclude dir pat1 pat2 | exclude all files under *dir* matching any of |
+| ...`                                      | the listed patterns                           |
++-------------------------------------------+-----------------------------------------------+
+| :command:`global-include pat1 pat2 ...`   | include all files anywhere in the source tree |
+|                                           | matching --- & any of the listed patterns     |
++-------------------------------------------+-----------------------------------------------+
+| :command:`global-exclude pat1 pat2 ...`   | exclude all files anywhere in the source tree |
+|                                           | matching --- & any of the listed patterns     |
++-------------------------------------------+-----------------------------------------------+
+| :command:`prune dir`                      | exclude all files under *dir*                 |
++-------------------------------------------+-----------------------------------------------+
+| :command:`graft dir`                      | include all files under *dir*                 |
++-------------------------------------------+-----------------------------------------------+
+
+The patterns here are Unix-style "glob" patterns: ``*`` matches any sequence of
+regular filename characters, ``?`` matches any single regular filename
+character, and ``[range]`` matches any of the characters in *range* (e.g.,
+``a-z``, ``a-zA-Z``, ``a-f0-9_.``).  The definition of "regular filename
+character" is platform-specific: on Unix it is anything except slash; on Windows
+anything except backslash or colon.
+
+**\*\*** Windows support not there yet **\*\***
+
 .. % \section{Creating a built distribution: the
 .. % \protect\command{bdist} command family}
 .. % \label{bdist-cmds}

Modified: python/branches/py3k/Doc/distutils/examples.rst
==============================================================================
--- python/branches/py3k/Doc/distutils/examples.rst	(original)
+++ python/branches/py3k/Doc/distutils/examples.rst	Sat Jul 31 11:10:51 2010
@@ -233,6 +233,58 @@
          ext_modules=[Extension('foopkg.foo', ['foo.c'])],
          )
 
+Checking a package
+==================
+
+The ``check`` command allows you to verify if your package meta-data
+meet the minimum requirements to build a distribution.
+
+To run it, just call it using your :file:`setup.py` script. If something is
+missing, ``check`` will display a warning.
+
+Let's take an example with a simple script::
+
+    from distutils.core import setup
+
+    setup(name='foobar')
+
+Running the ``check`` command will display some warnings::
+
+    $ python setup.py check
+    running check
+    warning: check: missing required meta-data: version, url
+    warning: check: missing meta-data: either (author and author_email) or
+             (maintainer and maintainer_email) must be supplied
+
+
+If you use the reStructuredText syntax in the `long_description` field and
+`docutils <http://docutils.sourceforge.net/>`_ is installed you can check if
+the syntax is fine with the ``check`` command, using the `restructuredtext`
+option.
+
+For example, if the :file:`setup.py` script is changed like this::
+
+    from distutils.core import setup
+
+    desc = """\
+    My description
+    =============
+
+    This is the description of the ``foobar`` package.
+    """
+
+    setup(name='foobar', version='1', author='tarek',
+        author_email='tarek at ziade.org',
+        url='http://example.com', long_description=desc)
+
+Where the long description is broken, ``check`` will be able to detect it
+by using the `docutils` parser::
+
+    $ pythontrunk setup.py check --restructuredtext
+    running check
+    warning: check: Title underline too short. (line 2)
+    warning: check: Could not finish the parsing.
+
 .. % \section{Multiple extension modules}
 .. % \label{multiple-ext}
 

Modified: python/branches/py3k/Doc/distutils/extending.rst
==============================================================================
--- python/branches/py3k/Doc/distutils/extending.rst	(original)
+++ python/branches/py3k/Doc/distutils/extending.rst	Sat Jul 31 11:10:51 2010
@@ -15,8 +15,8 @@
 should be copied into packages in addition to :file:`.py` files as a
 convenience.
 
-Most distutils command implementations are subclasses of the
-:class:`distutils.cmd.Command` class.  New commands may directly inherit from
+Most distutils command implementations are subclasses of the :class:`Command`
+class from :mod:`distutils.cmd`.  New commands may directly inherit from
 :class:`Command`, while replacements often derive from :class:`Command`
 indirectly, directly subclassing the command they are replacing.  Commands are
 required to derive from :class:`Command`.

Modified: python/branches/py3k/Doc/distutils/setupscript.rst
==============================================================================
--- python/branches/py3k/Doc/distutils/setupscript.rst	(original)
+++ python/branches/py3k/Doc/distutils/setupscript.rst	Sat Jul 31 11:10:51 2010
@@ -207,7 +207,7 @@
 SWIG on the interface file and compile the resulting C/C++ file into your
 extension.
 
-.. XXX SWIG support is rough around the edges and largely untested!
+**\*\*** SWIG support is rough around the edges and largely untested! **\*\***
 
 This warning notwithstanding, options to SWIG can be currently passed like
 this::
@@ -326,7 +326,7 @@
 (Again, this sort of non-portable construct should be avoided if you intend to
 distribute your code.)
 
-.. XXX Should mention clib libraries here or somewhere else!
+**\*\*** Should mention clib libraries here or somewhere else! **\*\***
 
 
 Other options

Modified: python/branches/py3k/Doc/distutils/sourcedist.rst
==============================================================================
--- python/branches/py3k/Doc/distutils/sourcedist.rst	(original)
+++ python/branches/py3k/Doc/distutils/sourcedist.rst	Sat Jul 31 11:10:51 2010
@@ -26,16 +26,16 @@
 +===========+=========================+=========+
 | ``zip``   | zip file (:file:`.zip`) | (1),(3) |
 +-----------+-------------------------+---------+
-| ``gztar`` | gzip'ed tar file        | \(2)    |
+| ``gztar`` | gzip'ed tar file        | (2),(4) |
 |           | (:file:`.tar.gz`)       |         |
 +-----------+-------------------------+---------+
-| ``bztar`` | bzip2'ed tar file       |         |
+| ``bztar`` | bzip2'ed tar file       | \(4)    |
 |           | (:file:`.tar.bz2`)      |         |
 +-----------+-------------------------+---------+
 | ``ztar``  | compressed tar file     | \(4)    |
 |           | (:file:`.tar.Z`)        |         |
 +-----------+-------------------------+---------+
-| ``tar``   | tar file (:file:`.tar`) |         |
+| ``tar``   | tar file (:file:`.tar`) | \(4)    |
 +-----------+-------------------------+---------+
 
 Notes:
@@ -51,16 +51,8 @@
    of the standard Python library since Python 1.6)
 
 (4)
-   requires the :program:`compress` program. Notice that this format is now
-   pending for deprecation and will be removed in the future versions of Python.
-
-When using any ``tar`` format (``gztar``, ``bztar``, ``ztar`` or
-``tar``) under Unix, you can specify the ``owner`` and ``group`` names
-that will be set for each member of the archive.
-
-For example, if you want all files of the archive to be owned by root::
-
-    python setup.py sdist --owner=root --group=root
+   requires external utilities: :program:`tar` and possibly one of :program:`gzip`,
+   :program:`bzip2`, or :program:`compress`
 
 
 .. _manifest:
@@ -76,10 +68,10 @@
   :option:`packages` options
 
 * all C source files mentioned in the :option:`ext_modules` or
-  :option:`libraries` options
+  :option:`libraries` options (
 
-  .. XXX Getting C library sources is currently broken -- no
-     :meth:`get_source_files` method in :file:`build_clib.py`!
+  **\*\*** getting C library sources currently broken---no
+  :meth:`get_source_files` method in :file:`build_clib.py`! **\*\***)
 
 * scripts identified by the :option:`scripts` option
   See :ref:`distutils-installing-scripts`.
@@ -111,60 +103,9 @@
 :file:`MANIFEST`, you must specify everything: the default set of files
 described above does not apply in this case.
 
-See :ref:`manifest_template` section for a syntax reference.
-
-.. _manifest-options:
-
-Manifest-related options
-========================
-
-The normal course of operations for the :command:`sdist` command is as follows:
-
-* if the manifest file, :file:`MANIFEST` doesn't exist, read :file:`MANIFEST.in`
-  and create the manifest
-
-* if neither :file:`MANIFEST` nor :file:`MANIFEST.in` exist, create a manifest
-  with just the default file set
-
-* if either :file:`MANIFEST.in` or the setup script (:file:`setup.py`) are more
-  recent than :file:`MANIFEST`, recreate :file:`MANIFEST` by reading
-  :file:`MANIFEST.in`
-
-* use the list of files now in :file:`MANIFEST` (either just generated or read
-  in) to create the source distribution archive(s)
-
-There are a couple of options that modify this behaviour.  First, use the
-:option:`--no-defaults` and :option:`--no-prune` to disable the standard
-"include" and "exclude" sets.
-
-Second, you might just want to (re)generate the manifest, but not create a
-source distribution::
-
-   python setup.py sdist --manifest-only
-
-:option:`-o` is a sortcut for :option:`--manifest-only`.
-
-.. _manifest_template:
-
-The MANIFEST.in template
-========================
-
-A :file:`MANIFEST.in` file can be added in a project to define the list of
-files to include in the distribution built by the :command:`sdist` command.
-
-When :command:`sdist` is run, it will look for the :file:`MANIFEST.in` file
-and interpret it to generate the :file:`MANIFEST` file that contains the
-list of files that will be included in the package.
-
-This mechanism can be used when the default list of files is not enough.
-(See :ref:`manifest`).
-
-Principle
----------
-
 The manifest template has one command per line, where each command specifies a
 set of files to include or exclude from the source distribution.  For an
-example, let's look at the Distutils' own manifest template::
+example, again we turn to the Distutils' own manifest template::
 
    include *.txt
    recursive-include examples *.txt *.py
@@ -176,7 +117,9 @@
 :file:`examples/sample?/build`.  All of this is done *after* the standard
 include set, so you can exclude files from the standard set with explicit
 instructions in the manifest template.  (Or, you can use the
-:option:`--no-defaults` option to disable the standard set entirely.)
+:option:`--no-defaults` option to disable the standard set entirely.)  There are
+several other commands available in the manifest template mini-language; see
+section :ref:`sdist-cmd`.
 
 The order of commands in the manifest template matters: initially, we have the
 list of default files as described above, and each command in the template adds
@@ -230,41 +173,36 @@
 them to the standard representation on your platform. That way, the manifest
 template is portable across operating systems.
 
-Commands
---------
 
-The manifest template commands are:
+.. _manifest-options:
+
+Manifest-related options
+========================
+
+The normal course of operations for the :command:`sdist` command is as follows:
+
+* if the manifest file, :file:`MANIFEST` doesn't exist, read :file:`MANIFEST.in`
+  and create the manifest
+
+* if neither :file:`MANIFEST` nor :file:`MANIFEST.in` exist, create a manifest
+  with just the default file set
+
+* if either :file:`MANIFEST.in` or the setup script (:file:`setup.py`) are more
+  recent than :file:`MANIFEST`, recreate :file:`MANIFEST` by reading
+  :file:`MANIFEST.in`
+
+* use the list of files now in :file:`MANIFEST` (either just generated or read
+  in) to create the source distribution archive(s)
+
+There are a couple of options that modify this behaviour.  First, use the
+:option:`--no-defaults` and :option:`--no-prune` to disable the standard
+"include" and "exclude" sets.
+
+Second, you might just want to (re)generate the manifest, but not create a source
+distribution::
+
+   python setup.py sdist --manifest-only
+
+:option:`-o` is a shortcut for :option:`--manifest-only`.
 
-+-------------------------------------------+-----------------------------------------------+
-| Command                                   | Description                                   |
-+===========================================+===============================================+
-| :command:`include pat1 pat2 ...`          | include all files matching any of the listed  |
-|                                           | patterns                                      |
-+-------------------------------------------+-----------------------------------------------+
-| :command:`exclude pat1 pat2 ...`          | exclude all files matching any of the listed  |
-|                                           | patterns                                      |
-+-------------------------------------------+-----------------------------------------------+
-| :command:`recursive-include dir pat1 pat2 | include all files under *dir* matching any of |
-| ...`                                      | the listed patterns                           |
-+-------------------------------------------+-----------------------------------------------+
-| :command:`recursive-exclude dir pat1 pat2 | exclude all files under *dir* matching any of |
-| ...`                                      | the listed patterns                           |
-+-------------------------------------------+-----------------------------------------------+
-| :command:`global-include pat1 pat2 ...`   | include all files anywhere in the source tree |
-|                                           | matching --- & any of the listed patterns     |
-+-------------------------------------------+-----------------------------------------------+
-| :command:`global-exclude pat1 pat2 ...`   | exclude all files anywhere in the source tree |
-|                                           | matching --- & any of the listed patterns     |
-+-------------------------------------------+-----------------------------------------------+
-| :command:`prune dir`                      | exclude all files under *dir*                 |
-+-------------------------------------------+-----------------------------------------------+
-| :command:`graft dir`                      | include all files under *dir*                 |
-+-------------------------------------------+-----------------------------------------------+
-
-The patterns here are Unix-style "glob" patterns: ``*`` matches any sequence of
-regular filename characters, ``?`` matches any single regular filename
-character, and ``[range]`` matches any of the characters in *range* (e.g.,
-``a-z``, ``a-zA-Z``, ``a-f0-9_.``).  The definition of "regular filename
-character" is platform-specific: on Unix it is anything except slash; on Windows
-anything except backslash or colon.
 

Modified: python/branches/py3k/Doc/distutils/uploading.rst
==============================================================================
--- python/branches/py3k/Doc/distutils/uploading.rst	(original)
+++ python/branches/py3k/Doc/distutils/uploading.rst	Sat Jul 31 11:10:51 2010
@@ -60,13 +60,13 @@
     setup(name='Distutils',
           long_description=open('README.txt'))
 
-In that case, :file:`README.txt` is a regular reStructuredText text file located
-in the root of the package besides :file:`setup.py`.
+In that case, `README.txt` is a regular reStructuredText text file located
+in the root of the package besides `setup.py`.
 
 To prevent registering broken reStructuredText content, you can use the
-:program:`rst2html` program that is provided by the :mod:`docutils` package
+:program:`rst2html` program that is provided by the `docutils` package
 and check the ``long_description`` from the command line::
 
     $ python setup.py --long-description | rst2html.py > output.html
 
-:mod:`docutils` will display a warning if there's something wrong with your syntax.
+`docutils` will display a warning if there's something wrong with your syntax.

From python-checkins at python.org  Sat Jul 31 11:15:10 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 11:15:10 +0200 (CEST)
Subject: [Python-checkins] r83323 - in python/branches/py3k/Doc/distutils:
	apiref.rst builtdist.rst commandref.rst setupscript.rst
	sourcedist.rst
Message-ID: <20100731091510.B9EFFEEA83@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 11:15:10 2010
New Revision: 83323

Log:
After distutils doc reversal, change back **bold todo** items to XXX comments.

Modified:
   python/branches/py3k/Doc/distutils/apiref.rst
   python/branches/py3k/Doc/distutils/builtdist.rst
   python/branches/py3k/Doc/distutils/commandref.rst
   python/branches/py3k/Doc/distutils/setupscript.rst
   python/branches/py3k/Doc/distutils/sourcedist.rst

Modified: python/branches/py3k/Doc/distutils/apiref.rst
==============================================================================
--- python/branches/py3k/Doc/distutils/apiref.rst	(original)
+++ python/branches/py3k/Doc/distutils/apiref.rst	Sat Jul 31 11:15:10 2010
@@ -995,7 +995,7 @@
    errors are ignored (apart from being reported to ``sys.stdout`` if *verbose* is
    true).
 
-**\*\*** Some of this could be replaced with the shutil module? **\*\***
+.. XXX Some of this could be replaced with the shutil module?
 
 
 :mod:`distutils.file_util` --- Single file operations
@@ -1311,9 +1311,7 @@
   the "negative alias" of :option:`--verbose`, then :option:`--quiet` on the
   command line sets *verbose* to false.
 
-**\*\*** Should be replaced with :mod:`optik` (which is also now known as
-:mod:`optparse` in Python 2.3 and later). **\*\***
-
+.. XXX Should be replaced with optparse
 
 .. function:: fancy_getopt(options, negative_opt, object, args)
 

Modified: python/branches/py3k/Doc/distutils/builtdist.rst
==============================================================================
--- python/branches/py3k/Doc/distutils/builtdist.rst	(original)
+++ python/branches/py3k/Doc/distutils/builtdist.rst	Sat Jul 31 11:15:10 2010
@@ -141,13 +141,13 @@
 commands.
 
 
-.. _creating-dumb:
+.. .. _creating-dumb:
 
-Creating dumb built distributions
-=================================
+.. Creating dumb built distributions
+.. =================================
 
-**\*\*** Need to document absolute vs. prefix-relative packages here, but first
-I have to implement it! **\*\***
+.. XXX Need to document absolute vs. prefix-relative packages here, but first
+   I have to implement it!
 
 
 .. _creating-rpms:

Modified: python/branches/py3k/Doc/distutils/commandref.rst
==============================================================================
--- python/branches/py3k/Doc/distutils/commandref.rst	(original)
+++ python/branches/py3k/Doc/distutils/commandref.rst	Sat Jul 31 11:15:10 2010
@@ -53,7 +53,7 @@
 Creating a source distribution: the :command:`sdist` command
 ============================================================
 
-**\*\*** fragment moved down from above: needs context! **\*\***
+.. XXX fragment moved down from above: needs context!
 
 The manifest template commands are:
 
@@ -90,7 +90,7 @@
 character" is platform-specific: on Unix it is anything except slash; on Windows
 anything except backslash or colon.
 
-**\*\*** Windows support not there yet **\*\***
+.. XXX Windows support not there yet
 
 .. % \section{Creating a built distribution: the
 .. % \protect\command{bdist} command family}

Modified: python/branches/py3k/Doc/distutils/setupscript.rst
==============================================================================
--- python/branches/py3k/Doc/distutils/setupscript.rst	(original)
+++ python/branches/py3k/Doc/distutils/setupscript.rst	Sat Jul 31 11:15:10 2010
@@ -207,7 +207,7 @@
 SWIG on the interface file and compile the resulting C/C++ file into your
 extension.
 
-**\*\*** SWIG support is rough around the edges and largely untested! **\*\***
+.. XXX SWIG support is rough around the edges and largely untested!
 
 This warning notwithstanding, options to SWIG can be currently passed like
 this::
@@ -326,7 +326,7 @@
 (Again, this sort of non-portable construct should be avoided if you intend to
 distribute your code.)
 
-**\*\*** Should mention clib libraries here or somewhere else! **\*\***
+.. XXX Should mention clib libraries here or somewhere else!
 
 
 Other options

Modified: python/branches/py3k/Doc/distutils/sourcedist.rst
==============================================================================
--- python/branches/py3k/Doc/distutils/sourcedist.rst	(original)
+++ python/branches/py3k/Doc/distutils/sourcedist.rst	Sat Jul 31 11:15:10 2010
@@ -70,8 +70,8 @@
 * all C source files mentioned in the :option:`ext_modules` or
   :option:`libraries` options (
 
-  **\*\*** getting C library sources currently broken---no
-  :meth:`get_source_files` method in :file:`build_clib.py`! **\*\***)
+  .. XXX getting C library sources currently broken---no
+         :meth:`get_source_files` method in :file:`build_clib.py`!
 
 * scripts identified by the :option:`scripts` option
   See :ref:`distutils-installing-scripts`.

From python-checkins at python.org  Sat Jul 31 11:21:51 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Sat, 31 Jul 2010 11:21:51 +0200 (CEST)
Subject: [Python-checkins] r83324 -
	python/branches/py3k/Tools/buildbot/external-common.bat
Message-ID: <20100731092151.EF0BEEEA86@mail.python.org>

Author: martin.v.loewis
Date: Sat Jul 31 11:21:51 2010
New Revision: 83324

Log:
Update OpenSSL version.


Modified:
   python/branches/py3k/Tools/buildbot/external-common.bat

Modified: python/branches/py3k/Tools/buildbot/external-common.bat
==============================================================================
--- python/branches/py3k/Tools/buildbot/external-common.bat	(original)
+++ python/branches/py3k/Tools/buildbot/external-common.bat	Sat Jul 31 11:21:51 2010
@@ -14,7 +14,7 @@
 @rem if exist tk8.4.16 rd /s/q tk8.4.16
 @rem if exist tk-8.4.18.1 rd /s/q tk-8.4.18.1
 @rem if exist db-4.4.20 rd /s/q db-4.4.20
- at rem if exist openssl-0.9.8l rd /s/q openssl-0.9.8l
+ at rem if exist openssl-1.0.0a rd /s/q openssl-1.0.0a
 @rem if exist sqlite-3.6.21 rd /s/q sqlite-3.6.21    
 
 @rem bzip
@@ -27,7 +27,7 @@
 if not exist db-4.4.20 svn export http://svn.python.org/projects/external/db-4.4.20-vs9 db-4.4.20
 
 @rem OpenSSL
-if not exist openssl-0.9.8l svn export http://svn.python.org/projects/external/openssl-0.9.8l
+if not exist openssl-1.0.0a svn export http://svn.python.org/projects/external/openssl-1.0.0a
 
 @rem tcl/tk
 if not exist tcl-8.5.2.1 (

From python-checkins at python.org  Sat Jul 31 11:37:03 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 11:37:03 +0200 (CEST)
Subject: [Python-checkins] r83325 -
	python/branches/py3k/Modules/_ctypes/ctypes.h
Message-ID: <20100731093703.4355DEE994@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 11:37:03 2010
New Revision: 83325

Log:
Copy Sun-specific inclusion of <alloca.h> from 2.7 maint to trunk; it seems to not have been merged to py3k.

Modified:
   python/branches/py3k/Modules/_ctypes/ctypes.h

Modified: python/branches/py3k/Modules/_ctypes/ctypes.h
==============================================================================
--- python/branches/py3k/Modules/_ctypes/ctypes.h	(original)
+++ python/branches/py3k/Modules/_ctypes/ctypes.h	Sat Jul 31 11:37:03 2010
@@ -1,3 +1,7 @@
+#if defined (__SVR4) && defined (__sun)
+#   include <alloca.h>
+#endif
+
 #ifndef MS_WIN32
 #define max(a, b) ((a) > (b) ? (a) : (b))
 #define min(a, b) ((a) < (b) ? (a) : (b))

From python-checkins at python.org  Sat Jul 31 12:08:09 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 12:08:09 +0200 (CEST)
Subject: [Python-checkins] r83326 - in python/branches/py3k/Lib: smtpd.py
	test/test_smtpd.py
Message-ID: <20100731100809.49D00EEAC0@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 12:08:09 2010
New Revision: 83326

Log:
Avoid triggering DeprecationWarnings in test_smtpd and smtpd.

Modified:
   python/branches/py3k/Lib/smtpd.py
   python/branches/py3k/Lib/test/test_smtpd.py

Modified: python/branches/py3k/Lib/smtpd.py
==============================================================================
--- python/branches/py3k/Lib/smtpd.py	(original)
+++ python/branches/py3k/Lib/smtpd.py	Sat Jul 31 12:08:09 2010
@@ -293,10 +293,10 @@
                 else:
                     data.append(text)
             self.received_data = NEWLINE.join(data)
-            status = self.__server.process_message(self.peer,
-                                                   self.mailfrom,
-                                                   self.rcpttos,
-                                                   self.received_data)
+            status = self.smtp_server.process_message(self.peer,
+                                                      self.mailfrom,
+                                                      self.rcpttos,
+                                                      self.received_data)
             self.rcpttos = []
             self.mailfrom = None
             self.smtp_state = self.COMMAND

Modified: python/branches/py3k/Lib/test/test_smtpd.py
==============================================================================
--- python/branches/py3k/Lib/test/test_smtpd.py	(original)
+++ python/branches/py3k/Lib/test/test_smtpd.py	Sat Jul 31 12:08:09 2010
@@ -79,101 +79,102 @@
 
     def test_missing_data(self):
         self.write_line(b'')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
                          b'500 Error: bad syntax\r\n')
 
     def test_EHLO_not_implemented(self):
         self.write_line(b'EHLO test.example')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
                          b'502 Error: command "EHLO" not implemented\r\n')
 
     def test_HELO(self):
         name = socket.getfqdn()
         self.write_line(b'HELO test.example')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
                          '250 {}\r\n'.format(name).encode('ascii'))
 
     def test_HELO_bad_syntax(self):
         self.write_line(b'HELO')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
                          b'501 Syntax: HELO hostname\r\n')
 
     def test_HELO_duplicate(self):
         self.write_line(b'HELO test.example')
         self.write_line(b'HELO test.example')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
                          b'503 Duplicate HELO/EHLO\r\n')
 
     def test_NOOP(self):
         self.write_line(b'NOOP')
-        self.assertEqual(self.channel.last, b'250 Ok\r\n')
+        self.assertEqual(self.channel.socket.last, b'250 Ok\r\n')
 
     def test_NOOP_bad_syntax(self):
         self.write_line(b'NOOP hi')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
                          b'501 Syntax: NOOP\r\n')
 
     def test_QUIT(self):
         self.write_line(b'QUIT')
-        self.assertEqual(self.channel.last, b'221 Bye\r\n')
+        self.assertEqual(self.channel.socket.last, b'221 Bye\r\n')
 
     def test_QUIT_arg_ignored(self):
         self.write_line(b'QUIT bye bye')
-        self.assertEqual(self.channel.last, b'221 Bye\r\n')
+        self.assertEqual(self.channel.socket.last, b'221 Bye\r\n')
 
     def test_bad_state(self):
-        self.channel._SMTPChannel__state = 'BAD STATE'
+        self.channel.smtp_state = 'BAD STATE'
         self.write_line(b'HELO')
-        self.assertEqual(self.channel.last, b'451 Internal confusion\r\n')
+        self.assertEqual(self.channel.socket.last,
+                         b'451 Internal confusion\r\n')
 
     def test_need_MAIL(self):
         self.write_line(b'RCPT to:spam at example')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
             b'503 Error: need MAIL command\r\n')
 
     def test_MAIL_syntax(self):
         self.write_line(b'MAIL from eggs at example')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
             b'501 Syntax: MAIL FROM:<address>\r\n')
 
     def test_MAIL_missing_from(self):
         self.write_line(b'MAIL from:')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
             b'501 Syntax: MAIL FROM:<address>\r\n')
 
     def test_MAIL_chevrons(self):
         self.write_line(b'MAIL from:<eggs at example>')
-        self.assertEqual(self.channel.last, b'250 Ok\r\n')
+        self.assertEqual(self.channel.socket.last, b'250 Ok\r\n')
 
     def test_nested_MAIL(self):
         self.write_line(b'MAIL from:eggs at example')
         self.write_line(b'MAIL from:spam at example')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
             b'503 Error: nested MAIL command\r\n')
 
     def test_need_RCPT(self):
         self.write_line(b'MAIL From:eggs at example')
         self.write_line(b'DATA')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
             b'503 Error: need RCPT command\r\n')
 
     def test_RCPT_syntax(self):
         self.write_line(b'MAIL From:eggs at example')
         self.write_line(b'RCPT to eggs at example')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
             b'501 Syntax: RCPT TO: <address>\r\n')
 
     def test_data_dialog(self):
         self.write_line(b'MAIL From:eggs at example')
-        self.assertEqual(self.channel.last, b'250 Ok\r\n')
+        self.assertEqual(self.channel.socket.last, b'250 Ok\r\n')
         self.write_line(b'RCPT To:spam at example')
-        self.assertEqual(self.channel.last, b'250 Ok\r\n')
+        self.assertEqual(self.channel.socket.last, b'250 Ok\r\n')
 
         self.write_line(b'DATA')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
             b'354 End data with <CR><LF>.<CR><LF>\r\n')
         self.write_line(b'data\r\nmore\r\n.')
-        self.assertEqual(self.channel.last, b'250 Ok\r\n')
+        self.assertEqual(self.channel.socket.last, b'250 Ok\r\n')
         self.assertEqual(self.server.messages[-1],
             ('peer', 'eggs at example', ['spam at example'], 'data\nmore'))
 
@@ -181,7 +182,7 @@
         self.write_line(b'MAIL From:eggs at example')
         self.write_line(b'RCPT To:spam at example')
         self.write_line(b'DATA spam')
-        self.assertEqual(self.channel.last, b'501 Syntax: DATA\r\n')
+        self.assertEqual(self.channel.socket.last, b'501 Syntax: DATA\r\n')
 
     def test_multiple_RCPT(self):
         self.write_line(b'MAIL From:eggs at example')
@@ -197,13 +198,13 @@
         self.write_line(b'RCPT To:spam at example')
         self.write_line(b'DATA')
         self.write_line(b'return status\r\n.')
-        self.assertEqual(self.channel.last, b'250 Okish\r\n')
+        self.assertEqual(self.channel.socket.last, b'250 Okish\r\n')
 
     def test_RSET(self):
         self.write_line(b'MAIL From:eggs at example')
         self.write_line(b'RCPT To:spam at example')
         self.write_line(b'RSET')
-        self.assertEqual(self.channel.last, b'250 Ok\r\n')
+        self.assertEqual(self.channel.socket.last, b'250 Ok\r\n')
         self.write_line(b'MAIL From:foo at example')
         self.write_line(b'RCPT To:eggs at example')
         self.write_line(b'DATA')
@@ -213,7 +214,7 @@
 
     def test_RSET_syntax(self):
         self.write_line(b'RSET hi')
-        self.assertEqual(self.channel.last, b'501 Syntax: RSET\r\n')
+        self.assertEqual(self.channel.socket.last, b'501 Syntax: RSET\r\n')
 
 
 def test_main():

From python-checkins at python.org  Sat Jul 31 12:11:39 2010
From: python-checkins at python.org (raymond.hettinger)
Date: Sat, 31 Jul 2010 12:11:39 +0200 (CEST)
Subject: [Python-checkins] r83327 - in python/branches/py3k:
	Doc/library/functools.rst Lib/functools.py
	Lib/test/test_functools.py Misc/NEWS
Message-ID: <20100731101139.ED8C1EEADA@mail.python.org>

Author: raymond.hettinger
Date: Sat Jul 31 12:11:39 2010
New Revision: 83327

Log:
Add functools.lfu_cache() and functools.lru_cache().

Modified:
   python/branches/py3k/Doc/library/functools.rst
   python/branches/py3k/Lib/functools.py
   python/branches/py3k/Lib/test/test_functools.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Doc/library/functools.rst
==============================================================================
--- python/branches/py3k/Doc/library/functools.rst	(original)
+++ python/branches/py3k/Doc/library/functools.rst	Sat Jul 31 12:11:39 2010
@@ -37,6 +37,57 @@
 
    .. versionadded:: 3.2
 
+.. decorator:: lfu_cache(maxsize)
+
+   Decorator to wrap a function with a memoizing callable that saves up to the
+   *maxsize* most frequent calls.  It can save time when an expensive or I/O
+   bound function is periodically called with the same arguments.
+
+   The *maxsize* parameter defaults to 100.  Since a dictionary is used to cache
+   results, the positional and keyword arguments to the function must be
+   hashable.
+
+   The wrapped function is instrumented with two attributes, :attr:`hits`
+   and :attr:`misses` which count the number of successful or unsuccessful
+   cache lookups.  These statistics are helpful for tuning the *maxsize*
+   parameter and for measuring the cache's effectiveness.
+
+   The wrapped function also has a :attr:`clear` attribute which can be
+   called (with no arguments) to clear the cache.
+
+   A `LFU (least frequently used) cache
+   <http://en.wikipedia.org/wiki/Cache_algorithms#Least-Frequently_Used>`_
+   is indicated when the pattern of calls does not change over time, when
+   more the most common calls already seen are the best predictors of the
+   most common upcoming calls.
+
+   .. versionadded:: 3.2
+
+.. decorator:: lru_cache(maxsize)
+
+   Decorator to wrap a function with a memoizing callable that saves up to the
+   *maxsize* most recent calls.  It can save time when an expensive or I/O bound
+   function is periodically called with the same arguments.
+
+   The *maxsize* parameter defaults to 100.  Since a dictionary is used to cache
+   results, the positional and keyword arguments to the function must be
+   hashable.
+
+   The wrapped function is instrumented with two attributes, :attr:`hits`
+   and :attr:`misses` which count the number of successful or unsuccessful
+   cache lookups.  These statistics are helpful for tuning the *maxsize*
+   parameter and for measuring the cache's effectiveness.
+
+   The wrapped function also has a :attr:`clear` attribute which can be
+   called (with no arguments) to clear the cache.
+
+   A `LRU (least recently used) cache
+   <http://en.wikipedia.org/wiki/Cache_algorithms#Least_Recently_Used>`_
+   is indicated when the pattern of calls changes over time, such as
+   when more recent calls are the best predictors of upcoming calls.
+
+   .. versionadded:: 3.2
+
 .. decorator:: total_ordering
 
    Given a class defining one or more rich comparison ordering methods, this

Modified: python/branches/py3k/Lib/functools.py
==============================================================================
--- python/branches/py3k/Lib/functools.py	(original)
+++ python/branches/py3k/Lib/functools.py	Sat Jul 31 12:11:39 2010
@@ -4,10 +4,17 @@
 # to allow utilities written in Python to be added
 # to the functools module.
 # Written by Nick Coghlan <ncoghlan at gmail.com>
-#   Copyright (C) 2006 Python Software Foundation.
+# and Raymond Hettinger <python at rcn.com>
+#   Copyright (C) 2006-2010 Python Software Foundation.
 # See C source code for _functools credits/copyright
 
+__all__ = ['update_wrapper', 'wraps', 'WRAPPER_ASSIGNMENTS', 'WRAPPER_UPDATES',
+           'total_ordering', 'cmp_to_key', 'lfu_cache', 'lru_cache']
+
 from _functools import partial, reduce
+from collections import OrderedDict, Counter
+from heapq import nsmallest
+from operator import itemgetter
 
 # update_wrapper() and wraps() are tools to help write
 # wrapper functions that can handle naive introspection
@@ -97,3 +104,88 @@
         def __hash__(self):
             raise TypeError('hash not implemented')
     return K
+
+def lfu_cache(maxsize=100):
+    '''Least-frequently-used cache decorator.
+
+    Arguments to the cached function must be hashable.
+    Cache performance statistics stored in f.hits and f.misses.
+    Clear the cache using f.clear().
+    http://en.wikipedia.org/wiki/Cache_algorithms#Least-Frequently_Used
+
+    '''
+    def decorating_function(user_function):
+        cache = {}                      # mapping of args to results
+        use_count = Counter()           # times each key has been accessed
+        kwd_mark = object()             # separate positional and keyword args
+
+        @wraps(user_function)
+        def wrapper(*args, **kwds):
+            key = args
+            if kwds:
+                key += (kwd_mark,) + tuple(sorted(kwds.items()))
+            use_count[key] += 1         # count a use of this key
+            try:
+                result = cache[key]
+                wrapper.hits += 1
+            except KeyError:
+                result = user_function(*args, **kwds)
+                cache[key] = result
+                wrapper.misses += 1
+                if len(cache) > maxsize:
+                    # purge the 10% least frequently used entries
+                    for key, _ in nsmallest(maxsize // 10,
+                                            use_count.items(),
+                                            key=itemgetter(1)):
+                        del cache[key], use_count[key]
+            return result
+
+        def clear():
+            'Clear the cache and cache statistics'
+            cache.clear()
+            use_count.clear()
+            wrapper.hits = wrapper.misses = 0
+
+        wrapper.hits = wrapper.misses = 0
+        wrapper.clear = clear
+        return wrapper
+    return decorating_function
+
+def lru_cache(maxsize=100):
+    '''Least-recently-used cache decorator.
+
+    Arguments to the cached function must be hashable.
+    Cache performance statistics stored in f.hits and f.misses.
+    Clear the cache using f.clear().
+    http://en.wikipedia.org/wiki/Cache_algorithms#Least_Recently_Used
+
+    '''
+    def decorating_function(user_function):
+        cache = OrderedDict()           # ordered least recent to most recent
+        kwd_mark = object()             # separate positional and keyword args
+
+        @wraps(user_function)
+        def wrapper(*args, **kwds):
+            key = args
+            if kwds:
+                key += (kwd_mark,) + tuple(sorted(kwds.items()))
+            try:
+                result = cache.pop(key)
+                wrapper.hits += 1
+            except KeyError:
+                result = user_function(*args, **kwds)
+                wrapper.misses += 1
+                if len(cache) >= maxsize:
+                    cache.popitem(0)    # purge least recently used cache entry
+            cache[key] = result         # record recent use of this key
+            return result
+
+        def clear():
+            'Clear the cache and cache statistics'
+            cache.clear()
+            wrapper.hits = wrapper.misses = 0
+
+        wrapper.hits = wrapper.misses = 0
+        wrapper.clear = clear
+        return wrapper
+    return decorating_function

Modified: python/branches/py3k/Lib/test/test_functools.py
==============================================================================
--- python/branches/py3k/Lib/test/test_functools.py	(original)
+++ python/branches/py3k/Lib/test/test_functools.py	Sat Jul 31 12:11:39 2010
@@ -4,6 +4,7 @@
 from test import support
 from weakref import proxy
 import pickle
+from random import choice
 
 @staticmethod
 def PythonPartial(func, *args, **keywords):
@@ -454,6 +455,50 @@
             class A:
                 pass
 
+class TestLRU(unittest.TestCase):
+
+    def test_lru(self):
+        def orig(x, y):
+            return 3*x+y
+        f = functools.lru_cache(maxsize=20)(orig)
+
+        domain = range(5)
+        for i in range(1000):
+            x, y = choice(domain), choice(domain)
+            actual = f(x, y)
+            expected = orig(x, y)
+            self.assertEquals(actual, expected)
+        self.assert_(f.hits > f.misses)
+        self.assertEquals(f.hits + f.misses, 1000)
+
+        f.clear()   # test clearing
+        self.assertEqual(f.hits, 0)
+        self.assertEqual(f.misses, 0)
+        f(x, y)
+        self.assertEqual(f.hits, 0)
+        self.assertEqual(f.misses, 1)
+
+    def test_lfu(self):
+        def orig(x, y):
+            return 3*x+y
+        f = functools.lfu_cache(maxsize=20)(orig)
+
+        domain = range(5)
+        for i in range(1000):
+            x, y = choice(domain), choice(domain)
+            actual = f(x, y)
+            expected = orig(x, y)
+            self.assertEquals(actual, expected)
+        self.assert_(f.hits > f.misses)
+        self.assertEquals(f.hits + f.misses, 1000)
+
+        f.clear()   # test clearing
+        self.assertEqual(f.hits, 0)
+        self.assertEqual(f.misses, 0)
+        f(x, y)
+        self.assertEqual(f.hits, 0)
+        self.assertEqual(f.misses, 1)
+
 def test_main(verbose=None):
     test_classes = (
         TestPartial,
@@ -461,7 +506,8 @@
         TestPythonPartial,
         TestUpdateWrapper,
         TestWraps,
-        TestReduce
+        TestReduce,
+        TestLRU,
     )
     support.run_unittest(*test_classes)
 

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sat Jul 31 12:11:39 2010
@@ -504,6 +504,8 @@
 - Issue #4179: In pdb, allow "list ." as a command to return to the currently
   debugged line.
 
+- Add lfu_cache() and lru_cache() decorators to the functools module.
+
 - Issue #4108: In urllib.robotparser, if there are multiple 'User-agent: *'
   entries, consider the first one.
 

From python-checkins at python.org  Sat Jul 31 12:14:41 2010
From: python-checkins at python.org (raymond.hettinger)
Date: Sat, 31 Jul 2010 12:14:41 +0200 (CEST)
Subject: [Python-checkins] r83328 -
	python/branches/py3k/Doc/library/collections.rst
Message-ID: <20100731101441.C8498EEAE2@mail.python.org>

Author: raymond.hettinger
Date: Sat Jul 31 12:14:41 2010
New Revision: 83328

Log:
Document how to change OrderedDict update order from first to last.

Modified:
   python/branches/py3k/Doc/library/collections.rst

Modified: python/branches/py3k/Doc/library/collections.rst
==============================================================================
--- python/branches/py3k/Doc/library/collections.rst	(original)
+++ python/branches/py3k/Doc/library/collections.rst	Sat Jul 31 12:14:41 2010
@@ -938,6 +938,18 @@
 are deleted.  But when new keys are added, the keys are appended
 to the end and the sort is not maintained.
 
+It is also straight-forward to create an ordered dictionary variant
+that the remembers the order the keys were *last* inserted.
+If a new entry overwrites an existing entry, the
+original insertion position is changed and moved to the end::
+
+    class LastUpdatedOrderedDict(OrderedDict):
+        'Store items is the order the keys were last added'
+        def __setitem__(self, key, value):
+            if key in self:
+                del self[key]
+            OrderedDict.__setitem__(self, key, value)
+
 
 :class:`UserDict` objects
 -------------------------

From python-checkins at python.org  Sat Jul 31 12:16:21 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 12:16:21 +0200 (CEST)
Subject: [Python-checkins] r83329 - in python/branches/py3k:
	Doc/library/functools.rst Lib/functools.py
	Lib/test/test_functools.py Misc/NEWS
Message-ID: <20100731101621.3F57CEEAD8@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 12:16:21 2010
New Revision: 83329

Log:
Revert r83327.  This will have to wait until after the alpha1 release.

Modified:
   python/branches/py3k/Doc/library/functools.rst
   python/branches/py3k/Lib/functools.py
   python/branches/py3k/Lib/test/test_functools.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Doc/library/functools.rst
==============================================================================
--- python/branches/py3k/Doc/library/functools.rst	(original)
+++ python/branches/py3k/Doc/library/functools.rst	Sat Jul 31 12:16:21 2010
@@ -37,57 +37,6 @@
 
    .. versionadded:: 3.2
 
-.. decorator:: lfu_cache(maxsize)
-
-   Decorator to wrap a function with a memoizing callable that saves up to the
-   *maxsize* most frequent calls.  It can save time when an expensive or I/O
-   bound function is periodically called with the same arguments.
-
-   The *maxsize* parameter defaults to 100.  Since a dictionary is used to cache
-   results, the positional and keyword arguments to the function must be
-   hashable.
-
-   The wrapped function is instrumented with two attributes, :attr:`hits`
-   and :attr:`misses` which count the number of successful or unsuccessful
-   cache lookups.  These statistics are helpful for tuning the *maxsize*
-   parameter and for measuring the cache's effectiveness.
-
-   The wrapped function also has a :attr:`clear` attribute which can be
-   called (with no arguments) to clear the cache.
-
-   A `LFU (least frequently used) cache
-   <http://en.wikipedia.org/wiki/Cache_algorithms#Least-Frequently_Used>`_
-   is indicated when the pattern of calls does not change over time, when
-   more the most common calls already seen are the best predictors of the
-   most common upcoming calls.
-
-   .. versionadded:: 3.2
-
-.. decorator:: lru_cache(maxsize)
-
-   Decorator to wrap a function with a memoizing callable that saves up to the
-   *maxsize* most recent calls.  It can save time when an expensive or I/O bound
-   function is periodically called with the same arguments.
-
-   The *maxsize* parameter defaults to 100.  Since a dictionary is used to cache
-   results, the positional and keyword arguments to the function must be
-   hashable.
-
-   The wrapped function is instrumented with two attributes, :attr:`hits`
-   and :attr:`misses` which count the number of successful or unsuccessful
-   cache lookups.  These statistics are helpful for tuning the *maxsize*
-   parameter and for measuring the cache's effectiveness.
-
-   The wrapped function also has a :attr:`clear` attribute which can be
-   called (with no arguments) to clear the cache.
-
-   A `LRU (least recently used) cache
-   <http://en.wikipedia.org/wiki/Cache_algorithms#Least_Recently_Used>`_
-   is indicated when the pattern of calls changes over time, such as
-   when more recent calls are the best predictors of upcoming calls.
-
-   .. versionadded:: 3.2
-
 .. decorator:: total_ordering
 
    Given a class defining one or more rich comparison ordering methods, this

Modified: python/branches/py3k/Lib/functools.py
==============================================================================
--- python/branches/py3k/Lib/functools.py	(original)
+++ python/branches/py3k/Lib/functools.py	Sat Jul 31 12:16:21 2010
@@ -4,17 +4,10 @@
 # to allow utilities written in Python to be added
 # to the functools module.
 # Written by Nick Coghlan <ncoghlan at gmail.com>
-# and Raymond Hettinger <python at rcn.com>
-#   Copyright (C) 2006-2010 Python Software Foundation.
+#   Copyright (C) 2006 Python Software Foundation.
 # See C source code for _functools credits/copyright
 
-__all__ = ['update_wrapper', 'wraps', 'WRAPPER_ASSIGNMENTS', 'WRAPPER_UPDATES',
-           'total_ordering', 'cmp_to_key', 'lfu_cache', 'lru_cache']
-
 from _functools import partial, reduce
-from collections import OrderedDict, Counter
-from heapq import nsmallest
-from operator import itemgetter
 
 # update_wrapper() and wraps() are tools to help write
 # wrapper functions that can handle naive introspection
@@ -104,88 +97,3 @@
         def __hash__(self):
             raise TypeError('hash not implemented')
     return K
-
-def lfu_cache(maxsize=100):
-    '''Least-frequently-used cache decorator.
-
-    Arguments to the cached function must be hashable.
-    Cache performance statistics stored in f.hits and f.misses.
-    Clear the cache using f.clear().
-    http://en.wikipedia.org/wiki/Cache_algorithms#Least-Frequently_Used
-
-    '''
-    def decorating_function(user_function):
-        cache = {}                      # mapping of args to results
-        use_count = Counter()           # times each key has been accessed
-        kwd_mark = object()             # separate positional and keyword args
-
-        @wraps(user_function)
-        def wrapper(*args, **kwds):
-            key = args
-            if kwds:
-                key += (kwd_mark,) + tuple(sorted(kwds.items()))
-            use_count[key] += 1         # count a use of this key
-            try:
-                result = cache[key]
-                wrapper.hits += 1
-            except KeyError:
-                result = user_function(*args, **kwds)
-                cache[key] = result
-                wrapper.misses += 1
-                if len(cache) > maxsize:
-                    # purge the 10% least frequently used entries
-                    for key, _ in nsmallest(maxsize // 10,
-                                            use_count.items(),
-                                            key=itemgetter(1)):
-                        del cache[key], use_count[key]
-            return result
-
-        def clear():
-            'Clear the cache and cache statistics'
-            cache.clear()
-            use_count.clear()
-            wrapper.hits = wrapper.misses = 0
-
-        wrapper.hits = wrapper.misses = 0
-        wrapper.clear = clear
-        return wrapper
-    return decorating_function
-
-def lru_cache(maxsize=100):
-    '''Least-recently-used cache decorator.
-
-    Arguments to the cached function must be hashable.
-    Cache performance statistics stored in f.hits and f.misses.
-    Clear the cache using f.clear().
-    http://en.wikipedia.org/wiki/Cache_algorithms#Least_Recently_Used
-
-    '''
-    def decorating_function(user_function):
-        cache = OrderedDict()           # ordered least recent to most recent
-        kwd_mark = object()             # separate positional and keyword args
-
-        @wraps(user_function)
-        def wrapper(*args, **kwds):
-            key = args
-            if kwds:
-                key += (kwd_mark,) + tuple(sorted(kwds.items()))
-            try:
-                result = cache.pop(key)
-                wrapper.hits += 1
-            except KeyError:
-                result = user_function(*args, **kwds)
-                wrapper.misses += 1
-                if len(cache) >= maxsize:
-                    cache.popitem(0)    # purge least recently used cache entry
-            cache[key] = result         # record recent use of this key
-            return result
-
-        def clear():
-            'Clear the cache and cache statistics'
-            cache.clear()
-            wrapper.hits = wrapper.misses = 0
-
-        wrapper.hits = wrapper.misses = 0
-        wrapper.clear = clear
-        return wrapper
-    return decorating_function

Modified: python/branches/py3k/Lib/test/test_functools.py
==============================================================================
--- python/branches/py3k/Lib/test/test_functools.py	(original)
+++ python/branches/py3k/Lib/test/test_functools.py	Sat Jul 31 12:16:21 2010
@@ -4,7 +4,6 @@
 from test import support
 from weakref import proxy
 import pickle
-from random import choice
 
 @staticmethod
 def PythonPartial(func, *args, **keywords):
@@ -455,50 +454,6 @@
             class A:
                 pass
 
-class TestLRU(unittest.TestCase):
-
-    def test_lru(self):
-        def orig(x, y):
-            return 3*x+y
-        f = functools.lru_cache(maxsize=20)(orig)
-
-        domain = range(5)
-        for i in range(1000):
-            x, y = choice(domain), choice(domain)
-            actual = f(x, y)
-            expected = orig(x, y)
-            self.assertEquals(actual, expected)
-        self.assert_(f.hits > f.misses)
-        self.assertEquals(f.hits + f.misses, 1000)
-
-        f.clear()   # test clearing
-        self.assertEqual(f.hits, 0)
-        self.assertEqual(f.misses, 0)
-        f(x, y)
-        self.assertEqual(f.hits, 0)
-        self.assertEqual(f.misses, 1)
-
-    def test_lfu(self):
-        def orig(x, y):
-            return 3*x+y
-        f = functools.lfu_cache(maxsize=20)(orig)
-
-        domain = range(5)
-        for i in range(1000):
-            x, y = choice(domain), choice(domain)
-            actual = f(x, y)
-            expected = orig(x, y)
-            self.assertEquals(actual, expected)
-        self.assert_(f.hits > f.misses)
-        self.assertEquals(f.hits + f.misses, 1000)
-
-        f.clear()   # test clearing
-        self.assertEqual(f.hits, 0)
-        self.assertEqual(f.misses, 0)
-        f(x, y)
-        self.assertEqual(f.hits, 0)
-        self.assertEqual(f.misses, 1)
-
 def test_main(verbose=None):
     test_classes = (
         TestPartial,
@@ -506,8 +461,7 @@
         TestPythonPartial,
         TestUpdateWrapper,
         TestWraps,
-        TestReduce,
-        TestLRU,
+        TestReduce
     )
     support.run_unittest(*test_classes)
 

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sat Jul 31 12:16:21 2010
@@ -504,8 +504,6 @@
 - Issue #4179: In pdb, allow "list ." as a command to return to the currently
   debugged line.
 
-- Add lfu_cache() and lru_cache() decorators to the functools module.
-
 - Issue #4108: In urllib.robotparser, if there are multiple 'User-agent: *'
   entries, consider the first one.
 

From python-checkins at python.org  Sat Jul 31 12:16:57 2010
From: python-checkins at python.org (raymond.hettinger)
Date: Sat, 31 Jul 2010 12:16:57 +0200 (CEST)
Subject: [Python-checkins] r83330 -
	python/branches/release31-maint/Doc/library/collections.rst
Message-ID: <20100731101657.A0209EEAE5@mail.python.org>

Author: raymond.hettinger
Date: Sat Jul 31 12:16:57 2010
New Revision: 83330

Log:
Document how to change OrderedDict update order from first to last.

Modified:
   python/branches/release31-maint/Doc/library/collections.rst

Modified: python/branches/release31-maint/Doc/library/collections.rst
==============================================================================
--- python/branches/release31-maint/Doc/library/collections.rst	(original)
+++ python/branches/release31-maint/Doc/library/collections.rst	Sat Jul 31 12:16:57 2010
@@ -911,6 +911,18 @@
 are deleted.  But when new keys are added, the keys are appended
 to the end and the sort is not maintained.
 
+It is also straight-forward to create an ordered dictionary variant
+that the remembers the order the keys were *last* inserted.
+If a new entry overwrites an existing entry, the
+original insertion position is changed and moved to the end::
+
+    class LastUpdatedOrderedDict(OrderedDict):
+        'Store items is the order the keys were last added'
+        def __setitem__(self, key, value):
+            if key in self:
+                del self[key]
+            OrderedDict.__setitem__(self, key, value)
+
 
 :class:`UserDict` objects
 -------------------------

From python-checkins at python.org  Sat Jul 31 12:31:17 2010
From: python-checkins at python.org (victor.stinner)
Date: Sat, 31 Jul 2010 12:31:17 +0200 (CEST)
Subject: [Python-checkins] r83332 -
	python/branches/import_unicode/Python/import.c
Message-ID: <20100731103117.74130EEA7B@mail.python.org>

Author: victor.stinner
Date: Sat Jul 31 12:31:17 2010
New Revision: 83332

Log:
Fix usage of PyUnicode_AsWideChar() and sizeof() in import.c

Last argument is the number of characters, not the number of bytes.

Modified:
   python/branches/import_unicode/Python/import.c

Modified: python/branches/import_unicode/Python/import.c
==============================================================================
--- python/branches/import_unicode/Python/import.c	(original)
+++ python/branches/import_unicode/Python/import.c	Sat Jul 31 12:31:17 2010
@@ -2002,7 +2002,8 @@
     if (Py_GETENV("PYTHONCASEOK") != NULL)
         return 1;
 
-    len = PyUnicode_AsWideChar((PyUnicodeObject*)fullpath_obj, buffer, sizeof(buffer));
+    len = PyUnicode_AsWideChar((PyUnicodeObject*)fullpath_obj, buffer,
+                               sizeof(buffer) / sizeof(buffer[0]));
     if (len == -1)
         return 0;
 
@@ -2138,7 +2139,8 @@
     Py_ssize_t len;
     int usize;
 
-    len = PyUnicode_AsWideChar((PyUnicodeObject*)unicode, path, sizeof(path));
+    len = PyUnicode_AsWideChar((PyUnicodeObject*)unicode, path,
+                               sizeof(path) / sizeof(path[0]));
     if (len == -1)
         return NULL;
 
@@ -2172,7 +2174,8 @@
     int err;
     struct _stat wstatbuf;
 
-    len = PyUnicode_AsWideChar((PyUnicodeObject*)unicode, path, sizeof(path));
+    len = PyUnicode_AsWideChar((PyUnicodeObject*)unicode, path,
+                               sizeof(path) / sizeof(path[0]));
     if (len == -1)
         return -1;
     err = _wstat(path, &wstatbuf);
@@ -3874,7 +3877,7 @@
         wchar_t path[MAXPATHLEN+1];
         Py_ssize_t len;
         len = PyUnicode_AsWideChar((PyUnicodeObject*)pathobj,
-                                   path, sizeof(path));
+                                   path, sizeof(path) / sizeof(path[0]));
         if (len == -1)
             return -1;
         /* see issue1293 and issue3677:

From python-checkins at python.org  Sat Jul 31 12:32:43 2010
From: python-checkins at python.org (victor.stinner)
Date: Sat, 31 Jul 2010 12:32:43 +0200 (CEST)
Subject: [Python-checkins] r83333 -
	python/branches/import_unicode/Modules/main.c
Message-ID: <20100731103243.454DFEEA45@mail.python.org>

Author: victor.stinner
Date: Sat Jul 31 12:32:43 2010
New Revision: 83333

Log:
run_command(): encode command to utf-8, not file system encoding

Modified:
   python/branches/import_unicode/Modules/main.c

Modified: python/branches/import_unicode/Modules/main.c
==============================================================================
--- python/branches/import_unicode/Modules/main.c	(original)
+++ python/branches/import_unicode/Modules/main.c	Sat Jul 31 12:32:43 2010
@@ -265,7 +265,7 @@
     commandObj = PyUnicode_FromWideChar(command, -1);
     if (commandObj == NULL)
         goto error;
-    command_bytes = PyUnicode_EncodeFSDefault(commandObj);
+    command_bytes = PyUnicode_AsUTF8String(commandObj);
     Py_DECREF(commandObj);
     if (command_bytes == NULL)
         goto error;

From python-checkins at python.org  Sat Jul 31 12:49:53 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Sat, 31 Jul 2010 12:49:53 +0200 (CEST)
Subject: [Python-checkins] r83335 - python/branches/py3k/PCbuild/build_ssl.py
Message-ID: <20100731104953.E8B26D37F@mail.python.org>

Author: martin.v.loewis
Date: Sat Jul 31 12:49:53 2010
New Revision: 83335

Log:
Copy asm files into place.

Modified:
   python/branches/py3k/PCbuild/build_ssl.py

Modified: python/branches/py3k/PCbuild/build_ssl.py
==============================================================================
--- python/branches/py3k/PCbuild/build_ssl.py	(original)
+++ python/branches/py3k/PCbuild/build_ssl.py	Sat Jul 31 12:49:53 2010
@@ -163,12 +163,14 @@
         do_script = "ms\\do_nasm"
         makefile="ms\\nt.mak"
         m32 = makefile
+        dirsuffix = "32"
     elif sys.argv[2] == "x64":
         arch="amd64"
         configure = "VC-WIN64A"
         do_script = "ms\\do_win64a"
         makefile = "ms\\nt64.mak"
         m32 = makefile.replace('64', '')
+        dirsuffix = "64"
         #os.environ["VSEXTCOMP_USECL"] = "MS_OPTERON"
     else:
         raise ValueError(str(sys.argv))
@@ -225,6 +227,13 @@
             shutil.copy(r"crypto\buildinf.h", r"crypto\buildinf_%s.h" % arch)
             shutil.copy(r"crypto\opensslconf.h", r"crypto\opensslconf_%s.h" % arch)
 
+        # If the assembler files don't exist in tmpXX, copy them there
+        if not os.path.exists("tmp"+dirsuffix):
+            os.mkdir("tmp"+dirsuffix)
+        for f in os.listdir("asm"+dirsuffix):
+            if not f.endswith(".asm"): continue
+            shutil.copy(r"asm%s\%s" % (dirsuffix, f), "tmp"+dirsuffix)
+
         # Now run make.
         if arch == "amd64":
             rc = os.system("ml64 -c -Foms\\uptable.obj ms\\uptable.asm")

From python-checkins at python.org  Sat Jul 31 12:50:17 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Sat, 31 Jul 2010 12:50:17 +0200 (CEST)
Subject: [Python-checkins] r83336 -
	python/branches/py3k/Tools/buildbot/external-common.bat
Message-ID: <20100731105017.13368EEAC6@mail.python.org>

Author: martin.v.loewis
Date: Sat Jul 31 12:50:16 2010
New Revision: 83336

Log:
Delete openssl checkouts.


Modified:
   python/branches/py3k/Tools/buildbot/external-common.bat

Modified: python/branches/py3k/Tools/buildbot/external-common.bat
==============================================================================
--- python/branches/py3k/Tools/buildbot/external-common.bat	(original)
+++ python/branches/py3k/Tools/buildbot/external-common.bat	Sat Jul 31 12:50:16 2010
@@ -14,7 +14,8 @@
 @rem if exist tk8.4.16 rd /s/q tk8.4.16
 @rem if exist tk-8.4.18.1 rd /s/q tk-8.4.18.1
 @rem if exist db-4.4.20 rd /s/q db-4.4.20
- at rem if exist openssl-1.0.0a rd /s/q openssl-1.0.0a
+rem if exist openssl-0.9.8l rd /s/q openssl-0.9.8l
+rem if exist openssl-1.0.0a rd /s/q openssl-1.0.0a
 @rem if exist sqlite-3.6.21 rd /s/q sqlite-3.6.21    
 
 @rem bzip

From python-checkins at python.org  Sat Jul 31 12:52:56 2010
From: python-checkins at python.org (victor.stinner)
Date: Sat, 31 Jul 2010 12:52:56 +0200 (CEST)
Subject: [Python-checkins] r83337 - in python/branches/py3k/Lib/ctypes/test:
	test_bytes.py test_loading.py test_random_things.py
Message-ID: <20100731105256.414D1EE994@mail.python.org>

Author: victor.stinner
Date: Sat Jul 31 12:52:56 2010
New Revision: 83337

Log:
Issue #8966: Fix ctypes tests for Windows

I removed the implicit conversion from str to bytes.


Modified:
   python/branches/py3k/Lib/ctypes/test/test_bytes.py
   python/branches/py3k/Lib/ctypes/test/test_loading.py
   python/branches/py3k/Lib/ctypes/test/test_random_things.py

Modified: python/branches/py3k/Lib/ctypes/test/test_bytes.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_bytes.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_bytes.py	Sat Jul 31 12:52:56 2010
@@ -45,7 +45,6 @@
                 _type_ = "X"
 
             BSTR("abc")
-            BSTR(b"abc")
 
 if __name__ == '__main__':
     unittest.main()

Modified: python/branches/py3k/Lib/ctypes/test/test_loading.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_loading.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_loading.py	Sat Jul 31 12:52:56 2010
@@ -97,7 +97,7 @@
             self.assertEqual(0, advapi32.CloseEventLog(None))
             windll.kernel32.GetProcAddress.argtypes = c_void_p, c_char_p
             windll.kernel32.GetProcAddress.restype = c_void_p
-            proc = windll.kernel32.GetProcAddress(advapi32._handle, "CloseEventLog")
+            proc = windll.kernel32.GetProcAddress(advapi32._handle, b"CloseEventLog")
             self.assertTrue(proc)
             # This is the real test: call the function via 'call_function'
             self.assertEqual(0, call_function(proc, (None,)))

Modified: python/branches/py3k/Lib/ctypes/test/test_random_things.py
==============================================================================
--- python/branches/py3k/Lib/ctypes/test/test_random_things.py	(original)
+++ python/branches/py3k/Lib/ctypes/test/test_random_things.py	Sat Jul 31 12:52:56 2010
@@ -18,7 +18,7 @@
             windll.kernel32.GetProcAddress.restype = c_void_p
 
             hdll = windll.kernel32.LoadLibraryA(b"kernel32")
-            funcaddr = windll.kernel32.GetProcAddress(hdll, "GetModuleHandleA")
+            funcaddr = windll.kernel32.GetProcAddress(hdll, b"GetModuleHandleA")
 
             self.assertEqual(call_function(funcaddr, (None,)),
                                  windll.kernel32.GetModuleHandleA(None))

From python-checkins at python.org  Sat Jul 31 12:56:54 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Sat, 31 Jul 2010 12:56:54 +0200 (CEST)
Subject: [Python-checkins] r83338 -
	python/branches/py3k/Tools/buildbot/external-common.bat
Message-ID: <20100731105654.10232EE994@mail.python.org>

Author: martin.v.loewis
Date: Sat Jul 31 12:56:53 2010
New Revision: 83338

Log:
Truly uncomment rm lines.


Modified:
   python/branches/py3k/Tools/buildbot/external-common.bat

Modified: python/branches/py3k/Tools/buildbot/external-common.bat
==============================================================================
--- python/branches/py3k/Tools/buildbot/external-common.bat	(original)
+++ python/branches/py3k/Tools/buildbot/external-common.bat	Sat Jul 31 12:56:53 2010
@@ -14,8 +14,8 @@
 @rem if exist tk8.4.16 rd /s/q tk8.4.16
 @rem if exist tk-8.4.18.1 rd /s/q tk-8.4.18.1
 @rem if exist db-4.4.20 rd /s/q db-4.4.20
-rem if exist openssl-0.9.8l rd /s/q openssl-0.9.8l
-rem if exist openssl-1.0.0a rd /s/q openssl-1.0.0a
+if exist openssl-0.9.8l rd /s/q openssl-0.9.8l
+if exist openssl-1.0.0a rd /s/q openssl-1.0.0a
 @rem if exist sqlite-3.6.21 rd /s/q sqlite-3.6.21    
 
 @rem bzip

From python-checkins at python.org  Sat Jul 31 13:00:48 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 13:00:48 +0200 (CEST)
Subject: [Python-checkins] r83339 - python/branches/py3k/README
Message-ID: <20100731110048.10244EE994@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 13:00:47 2010
New Revision: 83339

Log:
Rewrap.

Modified:
   python/branches/py3k/README

Modified: python/branches/py3k/README
==============================================================================
--- python/branches/py3k/README	(original)
+++ python/branches/py3k/README	Sat Jul 31 13:00:47 2010
@@ -5,11 +5,10 @@
 Python Software Foundation.
 All rights reserved.
 
-Python 3.x is a new version of the language, which is incompatible with the
-2.x line of releases.  The language is mostly the same, but many details,
-especially how built-in objects like dictionaries and strings work, have
-changed considerably, and a lot of deprecated features have finally been
-removed.
+Python 3.x is a new version of the language, which is incompatible with the 2.x
+line of releases.  The language is mostly the same, but many details, especially
+how built-in objects like dictionaries and strings work, have changed
+considerably, and a lot of deprecated features have finally been removed.
 
 
 Build Instructions
@@ -24,20 +23,19 @@
 
 This will install Python as python3.
 
-You can pass many options to the configure script; run "./configure
---help" to find out more.  On OSX and Cygwin, the executable is called
-python.exe; elsewhere it's just python.
-
-On Mac OS X, if you have configured Python with --enable-framework,
-you should use "make frameworkinstall" to do the installation.  Note
-that this installs the Python executable in a place that is not
-normally on your PATH, you may want to set up a symlink in
-/usr/local/bin.
+You can pass many options to the configure script; run "./configure --help" to
+find out more.  On OSX and Cygwin, the executable is called python.exe;
+elsewhere it's just python.
+
+On Mac OS X, if you have configured Python with --enable-framework, you should
+use "make frameworkinstall" to do the installation.  Note that this installs the
+Python executable in a place that is not normally on your PATH, you may want to
+set up a symlink in /usr/local/bin.
 
 On Windows, see PCbuild/readme.txt.
 
-If you wish, you can create a subdirectory and invoke configure from
-there.  For example:
+If you wish, you can create a subdirectory and invoke configure from there.  For
+example:
 
     mkdir debug
     cd debug
@@ -45,8 +43,8 @@
     make
     make test
 
-(This will fail if you *also* built at the top-level directory.  You
-should do a "make clean" at the toplevel first.)
+(This will fail if you *also* built at the top-level directory.  You should do a
+"make clean" at the toplevel first.)
 
 
 What's New
@@ -57,9 +55,9 @@
 
     http://docs.python.org/dev/3.2/whatsnew/3.2.html
 
-For a more detailed change log, read Misc/NEWS (though this file, too,
-is incomplete, and also doesn't list anything merged in from the 2.7
-release under development).
+For a more detailed change log, read Misc/NEWS (though this file, too, is
+incomplete, and also doesn't list anything merged in from the 2.7 release under
+development).
 
 If you want to install multiple versions of Python see the section below
 entitled "Installing multiple versions".
@@ -81,9 +79,9 @@
 Converting From Python 2.x to 3.x
 ---------------------------------
 
-Python starting with 2.6 contains features to help locating code that
-needs to be changed, such as optional warnings when deprecated features are
-used, and backported versions of certain key Python 3.x features.
+Python starting with 2.6 contains features to help locating code that needs to
+be changed, such as optional warnings when deprecated features are used, and
+backported versions of certain key Python 3.x features.
 
 A source-to-source translation tool, "2to3", can take care of the mundane task
 of converting large amounts of source code.  It is not a complete solution but
@@ -94,60 +92,58 @@
 Testing
 -------
 
-To test the interpreter, type "make test" in the top-level directory.
-This runs the test set twice (once with no compiled files, once with
-the compiled files left by the previous test run).  The test set
-produces some output.  You can generally ignore the messages about
-skipped tests due to optional features which can't be imported.
-If a message is printed about a failed test or a traceback or core
-dump is produced, something is wrong.  On some Linux systems (those
-that are not yet using glibc 6), test_strftime fails due to a
-non-standard implementation of strftime() in the C library. Please
-ignore this, or upgrade to glibc version 6.
+To test the interpreter, type "make test" in the top-level directory.  This runs
+the test set twice (once with no compiled files, once with the compiled files
+left by the previous test run).  The test set produces some output.  You can
+generally ignore the messages about skipped tests due to optional features which
+can't be imported.  If a message is printed about a failed test or a traceback
+or core dump is produced, something is wrong.  On some Linux systems (those that
+are not yet using glibc 6), test_strftime fails due to a non-standard
+implementation of strftime() in the C library. Please ignore this, or upgrade to
+glibc version 6.
 
 By default, tests are prevented from overusing resources like disk space and
 memory.  To enable these tests, run "make testall".
 
-IMPORTANT: If the tests fail and you decide to mail a bug report,
-*don't* include the output of "make test".  It is useless.  Run the
-failing test manually, as follows:
+IMPORTANT: If the tests fail and you decide to mail a bug report, *don't*
+include the output of "make test".  It is useless.  Run the failing test
+manually, as follows:
 
         ./python Lib/test/regrtest.py -v test_whatever
 
-(substituting the top of the source tree for '.' if you built in a
-different directory).  This runs the test in verbose mode.
+(substituting the top of the source tree for '.' if you built in a different
+directory).  This runs the test in verbose mode.
 
 
 Installing multiple versions
 ----------------------------
 
 On Unix and Mac systems if you intend to install multiple versions of Python
-using the same installation prefix (--prefix argument to the configure
-script) you must take care that your primary python executable is not
-overwritten by the installation of a different version.  All files and
-directories installed using "make altinstall" contain the major and minor
-version and can thus live side-by-side.  "make install" also creates
-${prefix}/bin/python3 which refers to ${prefix}/bin/pythonX.Y.  If you intend
-to install multiple versions using the same prefix you must decide which
-version (if any) is your "primary" version.  Install that version using
-"make install".  Install all other versions using "make altinstall".
-
-For example, if you want to install Python 2.5, 2.6 and 3.2 with 2.6 being
-the primary version, you would execute "make install" in your 2.6 build
-directory and "make altinstall" in the others.
+using the same installation prefix (--prefix argument to the configure script)
+you must take care that your primary python executable is not overwritten by the
+installation of a different version.  All files and directories installed using
+"make altinstall" contain the major and minor version and can thus live
+side-by-side.  "make install" also creates ${prefix}/bin/python3 which refers to
+${prefix}/bin/pythonX.Y.  If you intend to install multiple versions using the
+same prefix you must decide which version (if any) is your "primary" version.
+Install that version using "make install".  Install all other versions using
+"make altinstall".
+
+For example, if you want to install Python 2.5, 2.6 and 3.2 with 2.6 being the
+primary version, you would execute "make install" in your 2.6 build directory
+and "make altinstall" in the others.
 
 
 Issue Tracker and Mailing List
 ------------------------------
 
-We're soliciting bug reports about all aspects of the language.  Fixes
-are also welcome, preferable in unified diff format.  Please use the
-issue tracker:
+We're soliciting bug reports about all aspects of the language.  Fixes are also
+welcome, preferable in unified diff format.  Please use the issue tracker:
 
     http://bugs.python.org/
 
-If you're not sure whether you're dealing with a bug or a feature, use
-the mailing list:
+If you're not sure whether you're dealing with a bug or a feature, use the
+mailing list:
 
     python-dev at python.org
 

From python-checkins at python.org  Sat Jul 31 13:01:09 2010
From: python-checkins at python.org (victor.stinner)
Date: Sat, 31 Jul 2010 13:01:09 +0200 (CEST)
Subject: [Python-checkins] r83340 - in python/branches/import_unicode:
	Doc/README.txt Doc/distutils/apiref.rst
	Doc/distutils/builtdist.rst Doc/distutils/commandref.rst
	Doc/distutils/examples.rst Doc/distutils/extending.rst
	Doc/distutils/sourcedist.rst Doc/distutils/uploading.rst
	Doc/library/collections.rst Doc/license.rst Doc/make.bat
	Doc/tools/sphinxext/pyspecific.py Include/patchlevel.h
	LICENSE Lib/_abcoll.py Lib/ctypes/test/test_bytes.py
	Lib/ctypes/test/test_loading.py
	Lib/ctypes/test/test_random_things.py
	Lib/distutils/__init__.py Lib/idlelib/NEWS.txt
	Lib/idlelib/idlever.py Lib/pdb.py Lib/pydoc_data/topics.py
	Lib/smtpd.py Lib/test/test_pdb.py Lib/test/test_smtpd.py
	Lib/test/test_urllib.py Lib/urllib/parse.py Misc/NEWS
	Misc/RPM/python-3.2.spec Modules/_ctypes/ctypes.h
	Modules/_cursesmodule.c Modules/zlib/ChangeLog
	Modules/zlib/FAQ Modules/zlib/INDEX Modules/zlib/Makefile
	Modules/zlib/Makefile.in Modules/zlib/README
	Modules/zlib/adler32.c Modules/zlib/compress.c
	Modules/zlib/configure Modules/zlib/crc32.c
	Modules/zlib/deflate.c Modules/zlib/deflate.h
	Modules/zlib/example.c Modules/zlib/infback.c
	Modules/zlib/inffast.c Modules/zlib/inffast.h
	Modules/zlib/inflate.c Modules/zlib/inflate.h
	Modules/zlib/inftrees.c Modules/zlib/inftrees.h
	Modules/zlib/make_vms.com Modules/zlib/minigzip.c
	Modules/zlib/trees.c Modules/zlib/trees.h
	Modules/zlib/uncompr.c Modules/zlib/zconf.h
	Modules/zlib/zlib.3 Modules/zlib/zlib.h Modules/zlib/zutil.c
	Modules/zlib/zutil.h PC/python_nt.rc PCbuild/build_ssl.py
	Python/getcopyright.c README Tools/buildbot/external-common.bat
Message-ID: <20100731110109.11679EE994@mail.python.org>

Author: victor.stinner
Date: Sat Jul 31 13:01:07 2010
New Revision: 83340

Log:
Merge with py3k

svn merge -r 83295:HEAD svn+ssh://pythondev at svn.python.org/python/branches/py3k


Modified:
   python/branches/import_unicode/Doc/README.txt
   python/branches/import_unicode/Doc/distutils/apiref.rst
   python/branches/import_unicode/Doc/distutils/builtdist.rst
   python/branches/import_unicode/Doc/distutils/commandref.rst
   python/branches/import_unicode/Doc/distutils/examples.rst
   python/branches/import_unicode/Doc/distutils/extending.rst
   python/branches/import_unicode/Doc/distutils/sourcedist.rst
   python/branches/import_unicode/Doc/distutils/uploading.rst
   python/branches/import_unicode/Doc/library/collections.rst
   python/branches/import_unicode/Doc/license.rst
   python/branches/import_unicode/Doc/make.bat
   python/branches/import_unicode/Doc/tools/sphinxext/pyspecific.py
   python/branches/import_unicode/Include/patchlevel.h
   python/branches/import_unicode/LICENSE
   python/branches/import_unicode/Lib/_abcoll.py
   python/branches/import_unicode/Lib/ctypes/test/test_bytes.py
   python/branches/import_unicode/Lib/ctypes/test/test_loading.py
   python/branches/import_unicode/Lib/ctypes/test/test_random_things.py
   python/branches/import_unicode/Lib/distutils/__init__.py
   python/branches/import_unicode/Lib/idlelib/NEWS.txt
   python/branches/import_unicode/Lib/idlelib/idlever.py
   python/branches/import_unicode/Lib/pdb.py
   python/branches/import_unicode/Lib/pydoc_data/topics.py
   python/branches/import_unicode/Lib/smtpd.py
   python/branches/import_unicode/Lib/test/test_pdb.py
   python/branches/import_unicode/Lib/test/test_smtpd.py
   python/branches/import_unicode/Lib/test/test_urllib.py
   python/branches/import_unicode/Lib/urllib/parse.py
   python/branches/import_unicode/Misc/NEWS
   python/branches/import_unicode/Misc/RPM/python-3.2.spec
   python/branches/import_unicode/Modules/_ctypes/ctypes.h
   python/branches/import_unicode/Modules/_cursesmodule.c
   python/branches/import_unicode/Modules/zlib/ChangeLog
   python/branches/import_unicode/Modules/zlib/FAQ
   python/branches/import_unicode/Modules/zlib/INDEX
   python/branches/import_unicode/Modules/zlib/Makefile
   python/branches/import_unicode/Modules/zlib/Makefile.in
   python/branches/import_unicode/Modules/zlib/README
   python/branches/import_unicode/Modules/zlib/adler32.c
   python/branches/import_unicode/Modules/zlib/compress.c
   python/branches/import_unicode/Modules/zlib/configure
   python/branches/import_unicode/Modules/zlib/crc32.c
   python/branches/import_unicode/Modules/zlib/deflate.c
   python/branches/import_unicode/Modules/zlib/deflate.h
   python/branches/import_unicode/Modules/zlib/example.c
   python/branches/import_unicode/Modules/zlib/infback.c
   python/branches/import_unicode/Modules/zlib/inffast.c
   python/branches/import_unicode/Modules/zlib/inffast.h
   python/branches/import_unicode/Modules/zlib/inflate.c
   python/branches/import_unicode/Modules/zlib/inflate.h
   python/branches/import_unicode/Modules/zlib/inftrees.c
   python/branches/import_unicode/Modules/zlib/inftrees.h
   python/branches/import_unicode/Modules/zlib/make_vms.com
   python/branches/import_unicode/Modules/zlib/minigzip.c
   python/branches/import_unicode/Modules/zlib/trees.c
   python/branches/import_unicode/Modules/zlib/trees.h
   python/branches/import_unicode/Modules/zlib/uncompr.c
   python/branches/import_unicode/Modules/zlib/zconf.h
   python/branches/import_unicode/Modules/zlib/zlib.3
   python/branches/import_unicode/Modules/zlib/zlib.h
   python/branches/import_unicode/Modules/zlib/zutil.c
   python/branches/import_unicode/Modules/zlib/zutil.h
   python/branches/import_unicode/PC/python_nt.rc
   python/branches/import_unicode/PCbuild/build_ssl.py
   python/branches/import_unicode/Python/getcopyright.c
   python/branches/import_unicode/README
   python/branches/import_unicode/Tools/buildbot/external-common.bat

Modified: python/branches/import_unicode/Doc/README.txt
==============================================================================
--- python/branches/import_unicode/Doc/README.txt	(original)
+++ python/branches/import_unicode/Doc/README.txt	Sat Jul 31 13:01:07 2010
@@ -78,7 +78,7 @@
 
 You'll need to install the Sphinx package, either by checking it out via ::
 
-   svn co http://svn.python.org/projects/external/Sphinx-0.6.5/sphinx tools/sphinx
+   svn co http://svn.python.org/projects/external/Sphinx-1.0.1/sphinx tools/sphinx
 
 or by installing it from PyPI.
 
@@ -132,7 +132,7 @@
 as long as you don't change or remove the copyright notice:
 
 ----------------------------------------------------------------------
-Copyright (c) 2000-2008 Python Software Foundation.
+Copyright (c) 2000-2010 Python Software Foundation.
 All rights reserved.
 
 Copyright (c) 2000 BeOpen.com.

Modified: python/branches/import_unicode/Doc/distutils/apiref.rst
==============================================================================
--- python/branches/import_unicode/Doc/distutils/apiref.rst	(original)
+++ python/branches/import_unicode/Doc/distutils/apiref.rst	Sat Jul 31 13:01:07 2010
@@ -21,9 +21,7 @@
 .. function:: setup(arguments)
 
    The basic do-everything function that does most everything you could ever ask
-   for from a Distutils method.
-
-   .. See XXXXX
+   for from a Distutils method. See XXXXX
 
    The setup function takes a large number of arguments. These are laid out in the
    following table.
@@ -149,11 +147,11 @@
 In addition, the :mod:`distutils.core` module exposed a number of  classes that
 live elsewhere.
 
-* :class:`~distutils.extension.Extension` from :mod:`distutils.extension`
+* :class:`Extension` from :mod:`distutils.extension`
 
-* :class:`~distutils.cmd.Command` from :mod:`distutils.cmd`
+* :class:`Command` from :mod:`distutils.cmd`
 
-* :class:`~distutils.dist.Distribution` from :mod:`distutils.dist`
+* :class:`Distribution` from :mod:`distutils.dist`
 
 A short description of each of these follows, but see the relevant module for
 the full reference.
@@ -1313,8 +1311,7 @@
   the "negative alias" of :option:`--verbose`, then :option:`--quiet` on the
   command line sets *verbose* to false.
 
-.. XXX Should be replaced with :mod:`optparse`.
-
+.. XXX Should be replaced with optparse
 
 .. function:: fancy_getopt(options, negative_opt, object, args)
 
@@ -1681,8 +1678,8 @@
 ===================================================================
 
 .. module:: distutils.cmd
-   :synopsis: This module provides the abstract base class Command. This class
-              is subclassed by the modules in the distutils.command subpackage.
+   :synopsis: This module provides the abstract base class Command. This class is subclassed
+              by the modules in the distutils.command  subpackage.
 
 
 This module supplies the abstract base class :class:`Command`.
@@ -1692,84 +1689,20 @@
 
    Abstract base class for defining command classes, the "worker bees" of the
    Distutils.  A useful analogy for command classes is to think of them as
-   subroutines with local variables called *options*.  The options are declared
-   in :meth:`initialize_options` and defined (given their final values) in
-   :meth:`finalize_options`, both of which must be defined by every command
-   class.  The distinction between the two is necessary because option values
-   might come from the outside world (command line, config file, ...), and any
-   options dependent on other options must be computed after these outside
-   influences have been processed --- hence :meth:`finalize_options`.  The body
-   of the subroutine, where it does all its work based on the values of its
-   options, is the :meth:`run` method, which must also be implemented by every
-   command class.
+   subroutines with local variables called *options*.  The options are declared in
+   :meth:`initialize_options` and defined (given their final values) in
+   :meth:`finalize_options`, both of which must be defined by every command class.
+   The distinction between the two is necessary because option values might come
+   from the outside world (command line, config file, ...), and any options
+   dependent on other options must be computed after these outside influences have
+   been processed --- hence :meth:`finalize_options`.  The body of the subroutine,
+   where it does all its work based on the values of its options, is the
+   :meth:`run` method, which must also be implemented by every command class.
 
-   The class constructor takes a single argument *dist*, a :class:`Distribution`
+   The class constructor takes a single argument *dist*, a  :class:`Distribution`
    instance.
 
 
-Creating a new Distutils command
-================================
-
-This section outlines the steps to create a new Distutils command.
-
-A new command lives in a module in the :mod:`distutils.command` package. There
-is a sample template in that directory called :file:`command_template`.  Copy
-this file to a new module with the same name as the new command you're
-implementing.  This module should implement a class with the same name as the
-module (and the command).  So, for instance, to create the command
-``peel_banana`` (so that users can run ``setup.py peel_banana``), you'd copy
-:file:`command_template` to :file:`distutils/command/peel_banana.py`, then edit
-it so that it's implementing the class :class:`peel_banana`, a subclass of
-:class:`distutils.cmd.Command`.
-
-Subclasses of :class:`Command` must define the following methods.
-
-.. method:: Command.initialize_options()
-
-   Set default values for all the options that this command supports.  Note that
-   these defaults may be overridden by other commands, by the setup script, by
-   config files, or by the command-line.  Thus, this is not the place to code
-   dependencies between options; generally, :meth:`initialize_options`
-   implementations are just a bunch of ``self.foo = None`` assignments.
-
-
-.. method:: Command.finalize_options()
-
-   Set final values for all the options that this command supports. This is
-   always called as late as possible, ie.  after any option assignments from the
-   command-line or from other commands have been done.  Thus, this is the place
-   to to code option dependencies: if *foo* depends on *bar*, then it is safe to
-   set *foo* from *bar* as long as *foo* still has the same value it was
-   assigned in :meth:`initialize_options`.
-
-
-.. method:: Command.run()
-
-   A command's raison d'etre: carry out the action it exists to perform,
-   controlled by the options initialized in :meth:`initialize_options`,
-   customized by other commands, the setup script, the command-line, and config
-   files, and finalized in :meth:`finalize_options`.  All terminal output and
-   filesystem interaction should be done by :meth:`run`.
-
-
-.. attribute:: Command.sub_commands
-
-   *sub_commands* formalizes the notion of a "family" of commands,
-   e.g. ``install`` as the parent with sub-commands ``install_lib``,
-   ``install_headers``, etc.  The parent of a family of commands defines
-   *sub_commands* as a class attribute; it's a list of 2-tuples ``(command_name,
-   predicate)``, with *command_name* a string and *predicate* a function, a
-   string or ``None``.  *predicate* is a method of the parent command that
-   determines whether the corresponding command is applicable in the current
-   situation.  (E.g. we ``install_headers`` is only applicable if we have any C
-   header files to install.)  If *predicate* is ``None``, that command is always
-   applicable.
-
-   *sub_commands* is usually defined at the *end* of a class, because
-   predicates can be methods of the class, so they must already have been
-   defined.  The canonical example is the :command:`install` command.
-
-
 :mod:`distutils.command` --- Individual Distutils commands
 ==========================================================
 
@@ -2008,3 +1941,76 @@
 This is described in more detail in :pep:`301`.
 
 .. % todo
+
+:mod:`distutils.command.check` --- Check the meta-data of a package
+===================================================================
+
+.. module:: distutils.command.check
+   :synopsis: Check the metadata of a package
+
+
+The ``check`` command performs some tests on the meta-data of a package.
+For example, it verifies that all required meta-data are provided as
+the arguments passed to the :func:`setup` function.
+
+.. % todo
+
+
+Creating a new Distutils command
+================================
+
+This section outlines the steps to create a new Distutils command.
+
+A new command lives in a module in the :mod:`distutils.command` package. There
+is a sample template in that directory called  :file:`command_template`. Copy
+this file to a new module with the same name as the new command you're
+implementing. This module should implement a class with the same name as the
+module (and the command). So, for instance, to create the command
+``peel_banana`` (so that users can run ``setup.py peel_banana``), you'd copy
+:file:`command_template`  to :file:`distutils/command/peel_banana.py`, then edit
+it so that it's implementing the class :class:`peel_banana`, a subclass of
+:class:`distutils.cmd.Command`.
+
+Subclasses of :class:`Command` must define the following methods.
+
+
+.. method:: Command.initialize_options()
+
+   Set default values for all the options that this command supports.  Note that
+   these defaults may be overridden by other commands, by the setup script, by
+   config files, or by the command-line.  Thus, this is not the place to code
+   dependencies between options; generally, :meth:`initialize_options`
+   implementations are just a bunch of ``self.foo = None`` assignments.
+
+
+.. method:: Command.finalize_options()
+
+   Set final values for all the options that this command supports. This is
+   always called as late as possible, ie.  after any option assignments from the
+   command-line or from other commands have been done.  Thus, this is the place
+   to to code option dependencies: if *foo* depends on *bar*, then it is safe to
+   set *foo* from *bar* as long as *foo* still has the same value it was
+   assigned in :meth:`initialize_options`.
+
+
+.. method:: Command.run()
+
+   A command's raison d'etre: carry out the action it exists to perform, controlled
+   by the options initialized in :meth:`initialize_options`, customized by other
+   commands, the setup script, the command-line, and config files, and finalized in
+   :meth:`finalize_options`.  All terminal output and filesystem interaction should
+   be done by :meth:`run`.
+
+*sub_commands* formalizes the notion of a "family" of commands, eg. ``install``
+as the parent with sub-commands ``install_lib``, ``install_headers``, etc.  The
+parent of a family of commands defines *sub_commands* as a class attribute; it's
+a list of 2-tuples ``(command_name, predicate)``, with *command_name* a string
+and *predicate* a function, a string or None. *predicate* is a method of
+the parent command that determines whether the corresponding command is
+applicable in the current situation.  (Eg. we ``install_headers`` is only
+applicable if we have any C header files to install.)  If *predicate* is None,
+that command is always applicable.
+
+*sub_commands* is usually defined at the \*end\* of a class, because predicates
+can be methods of the class, so they must already have been defined.  The
+canonical example is the :command:`install` command.

Modified: python/branches/import_unicode/Doc/distutils/builtdist.rst
==============================================================================
--- python/branches/import_unicode/Doc/distutils/builtdist.rst	(original)
+++ python/branches/import_unicode/Doc/distutils/builtdist.rst	Sat Jul 31 13:01:07 2010
@@ -141,13 +141,13 @@
 commands.
 
 
-.. _creating-dumb:
+.. .. _creating-dumb:
 
-Creating dumb built distributions
-=================================
+.. Creating dumb built distributions
+.. =================================
 
 .. XXX Need to document absolute vs. prefix-relative packages here, but first
-       I have to implement it!
+   I have to implement it!
 
 
 .. _creating-rpms:
@@ -241,8 +241,7 @@
 configuration file, :file:`setup.cfg`\ ---see section :ref:`setup-config`.  If
 you distribute or package many Python module distributions, you might want to
 put options that apply to all of them in your personal Distutils configuration
-file (:file:`~/.pydistutils.cfg`).  If you want to temporarily disable
-this file, you can pass the --no-user-cfg option to setup.py.
+file (:file:`~/.pydistutils.cfg`).
 
 There are three steps to building a binary RPM package, all of which are
 handled automatically by the Distutils:

Modified: python/branches/import_unicode/Doc/distutils/commandref.rst
==============================================================================
--- python/branches/import_unicode/Doc/distutils/commandref.rst	(original)
+++ python/branches/import_unicode/Doc/distutils/commandref.rst	Sat Jul 31 13:01:07 2010
@@ -48,6 +48,50 @@
 .. % \label{clean-cmd}
 
 
+.. _sdist-cmd:
+
+Creating a source distribution: the :command:`sdist` command
+============================================================
+
+.. XXX fragment moved down from above: needs context!
+
+The manifest template commands are:
+
++-------------------------------------------+-----------------------------------------------+
+| Command                                   | Description                                   |
++===========================================+===============================================+
+| :command:`include pat1 pat2 ...`          | include all files matching any of the listed  |
+|                                           | patterns                                      |
++-------------------------------------------+-----------------------------------------------+
+| :command:`exclude pat1 pat2 ...`          | exclude all files matching any of the listed  |
+|                                           | patterns                                      |
++-------------------------------------------+-----------------------------------------------+
+| :command:`recursive-include dir pat1 pat2 | include all files under *dir* matching any of |
+| ...`                                      | the listed patterns                           |
++-------------------------------------------+-----------------------------------------------+
+| :command:`recursive-exclude dir pat1 pat2 | exclude all files under *dir* matching any of |
+| ...`                                      | the listed patterns                           |
++-------------------------------------------+-----------------------------------------------+
+| :command:`global-include pat1 pat2 ...`   | include all files anywhere in the source tree |
+|                                           | matching --- & any of the listed patterns     |
++-------------------------------------------+-----------------------------------------------+
+| :command:`global-exclude pat1 pat2 ...`   | exclude all files anywhere in the source tree |
+|                                           | matching --- & any of the listed patterns     |
++-------------------------------------------+-----------------------------------------------+
+| :command:`prune dir`                      | exclude all files under *dir*                 |
++-------------------------------------------+-----------------------------------------------+
+| :command:`graft dir`                      | include all files under *dir*                 |
++-------------------------------------------+-----------------------------------------------+
+
+The patterns here are Unix-style "glob" patterns: ``*`` matches any sequence of
+regular filename characters, ``?`` matches any single regular filename
+character, and ``[range]`` matches any of the characters in *range* (e.g.,
+``a-z``, ``a-zA-Z``, ``a-f0-9_.``).  The definition of "regular filename
+character" is platform-specific: on Unix it is anything except slash; on Windows
+anything except backslash or colon.
+
+.. XXX Windows support not there yet
+
 .. % \section{Creating a built distribution: the
 .. % \protect\command{bdist} command family}
 .. % \label{bdist-cmds}

Modified: python/branches/import_unicode/Doc/distutils/examples.rst
==============================================================================
--- python/branches/import_unicode/Doc/distutils/examples.rst	(original)
+++ python/branches/import_unicode/Doc/distutils/examples.rst	Sat Jul 31 13:01:07 2010
@@ -233,6 +233,58 @@
          ext_modules=[Extension('foopkg.foo', ['foo.c'])],
          )
 
+Checking a package
+==================
+
+The ``check`` command allows you to verify if your package meta-data
+meet the minimum requirements to build a distribution.
+
+To run it, just call it using your :file:`setup.py` script. If something is
+missing, ``check`` will display a warning.
+
+Let's take an example with a simple script::
+
+    from distutils.core import setup
+
+    setup(name='foobar')
+
+Running the ``check`` command will display some warnings::
+
+    $ python setup.py check
+    running check
+    warning: check: missing required meta-data: version, url
+    warning: check: missing meta-data: either (author and author_email) or
+             (maintainer and maintainer_email) must be supplied
+
+
+If you use the reStructuredText syntax in the `long_description` field and
+`docutils <http://docutils.sourceforge.net/>`_ is installed you can check if
+the syntax is fine with the ``check`` command, using the `restructuredtext`
+option.
+
+For example, if the :file:`setup.py` script is changed like this::
+
+    from distutils.core import setup
+
+    desc = """\
+    My description
+    =============
+
+    This is the description of the ``foobar`` package.
+    """
+
+    setup(name='foobar', version='1', author='tarek',
+        author_email='tarek at ziade.org',
+        url='http://example.com', long_description=desc)
+
+Where the long description is broken, ``check`` will be able to detect it
+by using the `docutils` parser::
+
+    $ pythontrunk setup.py check --restructuredtext
+    running check
+    warning: check: Title underline too short. (line 2)
+    warning: check: Could not finish the parsing.
+
 .. % \section{Multiple extension modules}
 .. % \label{multiple-ext}
 

Modified: python/branches/import_unicode/Doc/distutils/extending.rst
==============================================================================
--- python/branches/import_unicode/Doc/distutils/extending.rst	(original)
+++ python/branches/import_unicode/Doc/distutils/extending.rst	Sat Jul 31 13:01:07 2010
@@ -15,8 +15,8 @@
 should be copied into packages in addition to :file:`.py` files as a
 convenience.
 
-Most distutils command implementations are subclasses of the
-:class:`distutils.cmd.Command` class.  New commands may directly inherit from
+Most distutils command implementations are subclasses of the :class:`Command`
+class from :mod:`distutils.cmd`.  New commands may directly inherit from
 :class:`Command`, while replacements often derive from :class:`Command`
 indirectly, directly subclassing the command they are replacing.  Commands are
 required to derive from :class:`Command`.

Modified: python/branches/import_unicode/Doc/distutils/sourcedist.rst
==============================================================================
--- python/branches/import_unicode/Doc/distutils/sourcedist.rst	(original)
+++ python/branches/import_unicode/Doc/distutils/sourcedist.rst	Sat Jul 31 13:01:07 2010
@@ -26,16 +26,16 @@
 +===========+=========================+=========+
 | ``zip``   | zip file (:file:`.zip`) | (1),(3) |
 +-----------+-------------------------+---------+
-| ``gztar`` | gzip'ed tar file        | \(2)    |
+| ``gztar`` | gzip'ed tar file        | (2),(4) |
 |           | (:file:`.tar.gz`)       |         |
 +-----------+-------------------------+---------+
-| ``bztar`` | bzip2'ed tar file       |         |
+| ``bztar`` | bzip2'ed tar file       | \(4)    |
 |           | (:file:`.tar.bz2`)      |         |
 +-----------+-------------------------+---------+
 | ``ztar``  | compressed tar file     | \(4)    |
 |           | (:file:`.tar.Z`)        |         |
 +-----------+-------------------------+---------+
-| ``tar``   | tar file (:file:`.tar`) |         |
+| ``tar``   | tar file (:file:`.tar`) | \(4)    |
 +-----------+-------------------------+---------+
 
 Notes:
@@ -51,16 +51,8 @@
    of the standard Python library since Python 1.6)
 
 (4)
-   requires the :program:`compress` program. Notice that this format is now
-   pending for deprecation and will be removed in the future versions of Python.
-
-When using any ``tar`` format (``gztar``, ``bztar``, ``ztar`` or
-``tar``) under Unix, you can specify the ``owner`` and ``group`` names
-that will be set for each member of the archive.
-
-For example, if you want all files of the archive to be owned by root::
-
-    python setup.py sdist --owner=root --group=root
+   requires external utilities: :program:`tar` and possibly one of :program:`gzip`,
+   :program:`bzip2`, or :program:`compress`
 
 
 .. _manifest:
@@ -76,10 +68,10 @@
   :option:`packages` options
 
 * all C source files mentioned in the :option:`ext_modules` or
-  :option:`libraries` options
+  :option:`libraries` options (
 
-  .. XXX Getting C library sources is currently broken -- no
-     :meth:`get_source_files` method in :file:`build_clib.py`!
+  .. XXX getting C library sources currently broken---no
+         :meth:`get_source_files` method in :file:`build_clib.py`!
 
 * scripts identified by the :option:`scripts` option
   See :ref:`distutils-installing-scripts`.
@@ -111,60 +103,9 @@
 :file:`MANIFEST`, you must specify everything: the default set of files
 described above does not apply in this case.
 
-See :ref:`manifest_template` section for a syntax reference.
-
-.. _manifest-options:
-
-Manifest-related options
-========================
-
-The normal course of operations for the :command:`sdist` command is as follows:
-
-* if the manifest file, :file:`MANIFEST` doesn't exist, read :file:`MANIFEST.in`
-  and create the manifest
-
-* if neither :file:`MANIFEST` nor :file:`MANIFEST.in` exist, create a manifest
-  with just the default file set
-
-* if either :file:`MANIFEST.in` or the setup script (:file:`setup.py`) are more
-  recent than :file:`MANIFEST`, recreate :file:`MANIFEST` by reading
-  :file:`MANIFEST.in`
-
-* use the list of files now in :file:`MANIFEST` (either just generated or read
-  in) to create the source distribution archive(s)
-
-There are a couple of options that modify this behaviour.  First, use the
-:option:`--no-defaults` and :option:`--no-prune` to disable the standard
-"include" and "exclude" sets.
-
-Second, you might just want to (re)generate the manifest, but not create a
-source distribution::
-
-   python setup.py sdist --manifest-only
-
-:option:`-o` is a sortcut for :option:`--manifest-only`.
-
-.. _manifest_template:
-
-The MANIFEST.in template
-========================
-
-A :file:`MANIFEST.in` file can be added in a project to define the list of
-files to include in the distribution built by the :command:`sdist` command.
-
-When :command:`sdist` is run, it will look for the :file:`MANIFEST.in` file
-and interpret it to generate the :file:`MANIFEST` file that contains the
-list of files that will be included in the package.
-
-This mechanism can be used when the default list of files is not enough.
-(See :ref:`manifest`).
-
-Principle
----------
-
 The manifest template has one command per line, where each command specifies a
 set of files to include or exclude from the source distribution.  For an
-example, let's look at the Distutils' own manifest template::
+example, again we turn to the Distutils' own manifest template::
 
    include *.txt
    recursive-include examples *.txt *.py
@@ -176,7 +117,9 @@
 :file:`examples/sample?/build`.  All of this is done *after* the standard
 include set, so you can exclude files from the standard set with explicit
 instructions in the manifest template.  (Or, you can use the
-:option:`--no-defaults` option to disable the standard set entirely.)
+:option:`--no-defaults` option to disable the standard set entirely.)  There are
+several other commands available in the manifest template mini-language; see
+section :ref:`sdist-cmd`.
 
 The order of commands in the manifest template matters: initially, we have the
 list of default files as described above, and each command in the template adds
@@ -230,41 +173,36 @@
 them to the standard representation on your platform. That way, the manifest
 template is portable across operating systems.
 
-Commands
---------
 
-The manifest template commands are:
+.. _manifest-options:
+
+Manifest-related options
+========================
+
+The normal course of operations for the :command:`sdist` command is as follows:
+
+* if the manifest file, :file:`MANIFEST` doesn't exist, read :file:`MANIFEST.in`
+  and create the manifest
+
+* if neither :file:`MANIFEST` nor :file:`MANIFEST.in` exist, create a manifest
+  with just the default file set
+
+* if either :file:`MANIFEST.in` or the setup script (:file:`setup.py`) are more
+  recent than :file:`MANIFEST`, recreate :file:`MANIFEST` by reading
+  :file:`MANIFEST.in`
+
+* use the list of files now in :file:`MANIFEST` (either just generated or read
+  in) to create the source distribution archive(s)
+
+There are a couple of options that modify this behaviour.  First, use the
+:option:`--no-defaults` and :option:`--no-prune` to disable the standard
+"include" and "exclude" sets.
+
+Second, you might just want to (re)generate the manifest, but not create a source
+distribution::
+
+   python setup.py sdist --manifest-only
+
+:option:`-o` is a shortcut for :option:`--manifest-only`.
 
-+-------------------------------------------+-----------------------------------------------+
-| Command                                   | Description                                   |
-+===========================================+===============================================+
-| :command:`include pat1 pat2 ...`          | include all files matching any of the listed  |
-|                                           | patterns                                      |
-+-------------------------------------------+-----------------------------------------------+
-| :command:`exclude pat1 pat2 ...`          | exclude all files matching any of the listed  |
-|                                           | patterns                                      |
-+-------------------------------------------+-----------------------------------------------+
-| :command:`recursive-include dir pat1 pat2 | include all files under *dir* matching any of |
-| ...`                                      | the listed patterns                           |
-+-------------------------------------------+-----------------------------------------------+
-| :command:`recursive-exclude dir pat1 pat2 | exclude all files under *dir* matching any of |
-| ...`                                      | the listed patterns                           |
-+-------------------------------------------+-----------------------------------------------+
-| :command:`global-include pat1 pat2 ...`   | include all files anywhere in the source tree |
-|                                           | matching --- & any of the listed patterns     |
-+-------------------------------------------+-----------------------------------------------+
-| :command:`global-exclude pat1 pat2 ...`   | exclude all files anywhere in the source tree |
-|                                           | matching --- & any of the listed patterns     |
-+-------------------------------------------+-----------------------------------------------+
-| :command:`prune dir`                      | exclude all files under *dir*                 |
-+-------------------------------------------+-----------------------------------------------+
-| :command:`graft dir`                      | include all files under *dir*                 |
-+-------------------------------------------+-----------------------------------------------+
-
-The patterns here are Unix-style "glob" patterns: ``*`` matches any sequence of
-regular filename characters, ``?`` matches any single regular filename
-character, and ``[range]`` matches any of the characters in *range* (e.g.,
-``a-z``, ``a-zA-Z``, ``a-f0-9_.``).  The definition of "regular filename
-character" is platform-specific: on Unix it is anything except slash; on Windows
-anything except backslash or colon.
 

Modified: python/branches/import_unicode/Doc/distutils/uploading.rst
==============================================================================
--- python/branches/import_unicode/Doc/distutils/uploading.rst	(original)
+++ python/branches/import_unicode/Doc/distutils/uploading.rst	Sat Jul 31 13:01:07 2010
@@ -60,13 +60,13 @@
     setup(name='Distutils',
           long_description=open('README.txt'))
 
-In that case, :file:`README.txt` is a regular reStructuredText text file located
-in the root of the package besides :file:`setup.py`.
+In that case, `README.txt` is a regular reStructuredText text file located
+in the root of the package besides `setup.py`.
 
 To prevent registering broken reStructuredText content, you can use the
-:program:`rst2html` program that is provided by the :mod:`docutils` package
+:program:`rst2html` program that is provided by the `docutils` package
 and check the ``long_description`` from the command line::
 
     $ python setup.py --long-description | rst2html.py > output.html
 
-:mod:`docutils` will display a warning if there's something wrong with your syntax.
+`docutils` will display a warning if there's something wrong with your syntax.

Modified: python/branches/import_unicode/Doc/library/collections.rst
==============================================================================
--- python/branches/import_unicode/Doc/library/collections.rst	(original)
+++ python/branches/import_unicode/Doc/library/collections.rst	Sat Jul 31 13:01:07 2010
@@ -938,6 +938,18 @@
 are deleted.  But when new keys are added, the keys are appended
 to the end and the sort is not maintained.
 
+It is also straight-forward to create an ordered dictionary variant
+that the remembers the order the keys were *last* inserted.
+If a new entry overwrites an existing entry, the
+original insertion position is changed and moved to the end::
+
+    class LastUpdatedOrderedDict(OrderedDict):
+        'Store items is the order the keys were last added'
+        def __setitem__(self, key, value):
+            if key in self:
+                del self[key]
+            OrderedDict.__setitem__(self, key, value)
+
 
 :class:`UserDict` objects
 -------------------------

Modified: python/branches/import_unicode/Doc/license.rst
==============================================================================
--- python/branches/import_unicode/Doc/license.rst	(original)
+++ python/branches/import_unicode/Doc/license.rst	Sat Jul 31 13:01:07 2010
@@ -106,6 +106,10 @@
 +----------------+--------------+------------+------------+-----------------+
 | 3.1.1          | 3.1          | 2009       | PSF        | yes             |
 +----------------+--------------+------------+------------+-----------------+
+| 3.1.2          | 3.1          | 2010       | PSF        | yes             |
++----------------+--------------+------------+------------+-----------------+
+| 3.2            | 3.1          | 2010       | PSF        | yes             |
++----------------+--------------+------------+------------+-----------------+
 
 .. note::
 

Modified: python/branches/import_unicode/Doc/make.bat
==============================================================================
--- python/branches/import_unicode/Doc/make.bat	(original)
+++ python/branches/import_unicode/Doc/make.bat	Sat Jul 31 13:01:07 2010
@@ -34,7 +34,7 @@
 goto end
 
 :checkout
-svn co %SVNROOT%/external/Sphinx-0.6.5/sphinx tools/sphinx
+svn co %SVNROOT%/external/Sphinx-1.0.1/sphinx tools/sphinx
 svn co %SVNROOT%/external/docutils-0.6/docutils tools/docutils
 svn co %SVNROOT%/external/Jinja-2.3.1/jinja2 tools/jinja2
 svn co %SVNROOT%/external/Pygments-1.3.1/pygments tools/pygments

Modified: python/branches/import_unicode/Doc/tools/sphinxext/pyspecific.py
==============================================================================
--- python/branches/import_unicode/Doc/tools/sphinxext/pyspecific.py	(original)
+++ python/branches/import_unicode/Doc/tools/sphinxext/pyspecific.py	Sat Jul 31 13:01:07 2010
@@ -145,10 +145,10 @@
         for label in self.status_iterator(pydoc_topic_labels,
                                           'building topics... ',
                                           length=len(pydoc_topic_labels)):
-            if label not in self.env.labels:
+            if label not in self.env.domaindata['std']['labels']:
                 self.warn('label %r not in documentation' % label)
                 continue
-            docname, labelid, sectname = self.env.labels[label]
+            docname, labelid, sectname = self.env.domaindata['std']['labels'][label]
             doctree = self.env.get_and_resolve_doctree(docname, self)
             document = new_document('<section node>')
             document.append(doctree.ids[labelid])

Modified: python/branches/import_unicode/Include/patchlevel.h
==============================================================================
--- python/branches/import_unicode/Include/patchlevel.h	(original)
+++ python/branches/import_unicode/Include/patchlevel.h	Sat Jul 31 13:01:07 2010
@@ -20,10 +20,10 @@
 #define PY_MINOR_VERSION	2
 #define PY_MICRO_VERSION	0
 #define PY_RELEASE_LEVEL	PY_RELEASE_LEVEL_ALPHA
-#define PY_RELEASE_SERIAL	0
+#define PY_RELEASE_SERIAL	1
 
 /* Version as a string */
-#define PY_VERSION      	"3.2a0"
+#define PY_VERSION      	"3.2a1"
 /*--end constants--*/
 
 /* Subversion Revision number of this file (not of the repository) */

Modified: python/branches/import_unicode/LICENSE
==============================================================================
--- python/branches/import_unicode/LICENSE	(original)
+++ python/branches/import_unicode/LICENSE	Sat Jul 31 13:01:07 2010
@@ -68,6 +68,7 @@
     3.1             3.0.1       2009        PSF         yes
     3.1.1           3.1         2009        PSF         yes
     3.1.2           3.1         2010        PSF         yes
+    3.2             3.1         2010        PSF         yes
 
 Footnotes:
 

Modified: python/branches/import_unicode/Lib/_abcoll.py
==============================================================================
--- python/branches/import_unicode/Lib/_abcoll.py	(original)
+++ python/branches/import_unicode/Lib/_abcoll.py	Sat Jul 31 13:01:07 2010
@@ -18,11 +18,6 @@
            "MappingView", "KeysView", "ItemsView", "ValuesView",
            "Sequence", "MutableSequence",
            "ByteString",
-           "bytearray_iterator", "bytes_iterator", "dict_itemiterator",
-           "dict_items", "dict_keyiterator", "dict_keys", "dict_proxy",
-           "dict_valueiterator", "dict_values", "list_iterator",
-           "list_reverseiterator", "range_iterator", "set_iterator",
-           "str_iterator", "tuple_iterator", "zip_iterator",
            ]
 
 

Modified: python/branches/import_unicode/Lib/ctypes/test/test_bytes.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/test/test_bytes.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/test/test_bytes.py	Sat Jul 31 13:01:07 2010
@@ -45,7 +45,6 @@
                 _type_ = "X"
 
             BSTR("abc")
-            BSTR(b"abc")
 
 if __name__ == '__main__':
     unittest.main()

Modified: python/branches/import_unicode/Lib/ctypes/test/test_loading.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/test/test_loading.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/test/test_loading.py	Sat Jul 31 13:01:07 2010
@@ -97,7 +97,7 @@
             self.assertEqual(0, advapi32.CloseEventLog(None))
             windll.kernel32.GetProcAddress.argtypes = c_void_p, c_char_p
             windll.kernel32.GetProcAddress.restype = c_void_p
-            proc = windll.kernel32.GetProcAddress(advapi32._handle, "CloseEventLog")
+            proc = windll.kernel32.GetProcAddress(advapi32._handle, b"CloseEventLog")
             self.assertTrue(proc)
             # This is the real test: call the function via 'call_function'
             self.assertEqual(0, call_function(proc, (None,)))

Modified: python/branches/import_unicode/Lib/ctypes/test/test_random_things.py
==============================================================================
--- python/branches/import_unicode/Lib/ctypes/test/test_random_things.py	(original)
+++ python/branches/import_unicode/Lib/ctypes/test/test_random_things.py	Sat Jul 31 13:01:07 2010
@@ -18,7 +18,7 @@
             windll.kernel32.GetProcAddress.restype = c_void_p
 
             hdll = windll.kernel32.LoadLibraryA(b"kernel32")
-            funcaddr = windll.kernel32.GetProcAddress(hdll, "GetModuleHandleA")
+            funcaddr = windll.kernel32.GetProcAddress(hdll, b"GetModuleHandleA")
 
             self.assertEqual(call_function(funcaddr, (None,)),
                                  windll.kernel32.GetModuleHandleA(None))

Modified: python/branches/import_unicode/Lib/distutils/__init__.py
==============================================================================
--- python/branches/import_unicode/Lib/distutils/__init__.py	(original)
+++ python/branches/import_unicode/Lib/distutils/__init__.py	Sat Jul 31 13:01:07 2010
@@ -15,5 +15,5 @@
 # Updated automatically by the Python release process.
 #
 #--start constants--
-__version__ = "3.2a0"
+__version__ = "3.2a1"
 #--end constants--

Modified: python/branches/import_unicode/Lib/idlelib/NEWS.txt
==============================================================================
--- python/branches/import_unicode/Lib/idlelib/NEWS.txt	(original)
+++ python/branches/import_unicode/Lib/idlelib/NEWS.txt	Sat Jul 31 13:01:07 2010
@@ -1,7 +1,7 @@
-What's New in IDLE 3.1b1?
+What's New in IDLE 3.1?
 =========================
 
-*Release date: XX-XXX-09*
+*Release date: 27-Jun-09*
 
 - Use of 'filter' in keybindingDialog.py was causing custom key assignment to
   fail.  Patch 5707 amaury.forgeotdarc.

Modified: python/branches/import_unicode/Lib/idlelib/idlever.py
==============================================================================
--- python/branches/import_unicode/Lib/idlelib/idlever.py	(original)
+++ python/branches/import_unicode/Lib/idlelib/idlever.py	Sat Jul 31 13:01:07 2010
@@ -1 +1 @@
-IDLE_VERSION = "3.2a0"
+IDLE_VERSION = "3.2a1"

Modified: python/branches/import_unicode/Lib/pdb.py
==============================================================================
--- python/branches/import_unicode/Lib/pdb.py	(original)
+++ python/branches/import_unicode/Lib/pdb.py	Sat Jul 31 13:01:07 2010
@@ -62,178 +62,28 @@
 Debugger commands
 =================
 
-h(elp)
-        Without argument, print the list of available commands.  With
-        a command name as argument, print help about that command.
-
-w(here)
-        Print a stack trace, with the most recent frame at the bottom.
-        An arrow indicates the "current frame", which determines the
-        context of most commands.
-
-d(own) [ count ]
-        Move the current frame count (default one) levels down in the
-        stack trace (to a newer frame).
-
-u(p) [ count ]
-        Move the current frame count (default one) levels up in the
-        stack trace (to an older frame).
-
-b(reak) [ ([filename:]lineno | function) [, condition] ]
-        With a filename:lineno argument, set a break there.  If
-        filename is omitted, use the current file.  With a function
-        name, set a break at the first executable line of that
-        function.  Without argument, list all breaks.  Each breakpoint
-        is assigned a number to which all the other breakpoint
-        commands refer.
-
-        The condition argument, if present, is a string which must
-        evaluate to true in order for the breakpoint to be honored.
-
-tbreak [ ([filename:]lineno | function) [, condition] ]
-        Temporary breakpoint, which is removed automatically when it
-        is first hit.  The arguments are the same as for break.
-
-cl(ear) [bpnumber [bpnumber ...] ]
-        With a space separated list of breakpoint numbers, clear those
-        breakpoints.  Without argument, clear all breaks (but first
-        ask confirmation).
-
-disable bpnumber [bpnumber ...]
-        Disable the breakpoints given as a space separated list of
-        breakpoint numbers.  Disabling a breakpoint means it cannot
-        cause the program to stop execution, but unlike clearing a
-        breakpoint, it remains in the list of breakpoints and can be
-        (re-)enabled.
-
-enable bpnumber [bpnumber ...]
-        Enable the breakpoints specified.
-
-ignore bpnumber [count]
-        Set the ignore count for the given breakpoint number.  If
-        count is omitted, the ignore count is set to 0.  A breakpoint
-        becomes active when the ignore count is zero.  When non-zero,
-        the count is decremented each time the breakpoint is reached
-        and the breakpoint is not disabled and any associated
-        condition evaluates to true.
-
-condition bpnumber [condition]
-        Set a new condition for the breakpoint, an expression which
-        must evaluate to true before the breakpoint is honored.  If
-        condition is absent, any existing condition is removed; i.e.,
-        the breakpoint is made unconditional.
-
-commands [bpnumber]
-        Specify a list of commands for the breakpoint.  Type a line
-        containing just 'end' to terminate the commands.  The commands
-        are executed when the breakpoint is hit.
-
-        With no breakpoint number argument, refers to the last
-        breakpoint set.
-
-s(tep)
-        Execute the current line, stop at the first possible occasion
-        (either in a function that is called or in the current
-        function).
-
-n(ext)
-        Continue execution until the next line in the current function
-        is reached or it returns.
-
-unt(il)
-        Continue execution until the line with a number greater than
-        the current one is reached or until the current frame returns.
-
-r(eturn)
-        Continue execution until the current function returns.
-
-run [args...]
-        Restart the debugged python program. If a string is supplied
-        it is splitted with "shlex", and the result is used as the new
-        sys.argv.  History, breakpoints, actions and debugger options
-        are preserved.  "restart" is an alias for "run".
-
-c(ont(inue))
-        Continue execution, only stop when a breakpoint is encountered.
-
-l(ist) [first [,last]]
-        List source code for the current file.
-        Without arguments, list 11 lines around the current line
-        or continue the previous listing.
-        With one argument, list 11 lines starting at that line.
-        With two arguments, list the given range;
-        if the second argument is less than the first, it is a count.
-
-a(rgs)
-        Print the argument list of the current function.
-
-p expression
-        Print the value of the expression.
-
-(!) statement
-        Execute the (one-line) statement in the context of the current
-        stack frame.  The exclamation point can be omitted unless the
-        first word of the statement resembles a debugger command.  To
-        assign to a global variable you must always prefix the command
-        with a 'global' command, e.g.:
-        (Pdb) global list_options; list_options = ['-l']
-        (Pdb)
-
-
-whatis arg
-        Print the type of the argument.
-
-alias [name [command]]
-        Creates an alias called 'name' that executes 'command'.  The
-        command must *not* be enclosed in quotes.  Replaceable
-        parameters can be indicated by %1, %2, and so on, while %* is
-        replaced by all the parameters.  If no command is given, the
-        current alias for name is shown. If no name is given, all
-        aliases are listed.
-
-        Aliases may be nested and can contain anything that can be
-        legally typed at the pdb prompt.  Note!  You *can* override
-        internal pdb commands with aliases!  Those internal commands
-        are then hidden until the alias is removed.  Aliasing is
-        recursively applied to the first word of the command line; all
-        other words in the line are left alone.
-
-        As an example, here are two useful aliases (especially when
-        placed in the .pdbrc file):
-
-        # Print instance variables (usage "pi classInst")
-        alias pi for k in %1.__dict__.keys(): print "%1.",k,"=",%1.__dict__[k]
-        # Print instance variables in self
-        alias ps pi self
-
-unalias name
-        Delete the specified alias.
-
-q(uit)
-        Quit from the debugger. The program being executed is aborted.
 """
+# NOTE: the actual command documentation is collected from docstrings of the
+# commands and is appended to __doc__ after the class has been defined.
 
 import sys
 import linecache
 import cmd
 import bdb
-from reprlib import Repr
+import dis
 import os
 import re
+import code
 import pprint
 import traceback
+import inspect
+import types
 
 
 class Restart(Exception):
     """Causes a debugger to be restarted for the debugged python program."""
     pass
 
-# Create a custom safe Repr instance and increase its maxstring.
-# The default of 30 truncates error messages too easily.
-_repr = Repr()
-_repr.maxstring = 200
-_saferepr = _repr.repr
-
 __all__ = ["run", "pm", "Pdb", "runeval", "runctx", "runcall", "set_trace",
            "post_mortem", "help"]
 
@@ -257,6 +107,23 @@
     fp.close()
     return answer
 
+def getsourcelines(obj):
+    lines, lineno = inspect.findsource(obj)
+    if inspect.isframe(obj) and obj.f_globals is obj.f_locals:
+        # must be a module frame: do not try to cut a block out of it
+        return lines, 1
+    elif inspect.ismodule(obj):
+        return lines, 1
+    return inspect.getblock(lines[lineno:]), lineno+1
+
+def lasti2lineno(code, lasti):
+    linestarts = list(dis.findlinestarts(code))
+    linestarts.reverse()
+    for i, lineno in linestarts:
+        if lasti >= i:
+            return lineno
+    return 0
+
 
 # Interaction prompt line will separate file and call info from code
 # text using value of line_prefix string.  A newline and arrow may
@@ -276,6 +143,7 @@
         self.aliases = {}
         self.mainpyfile = ''
         self._wait_for_mainpyfile = 0
+        self.tb_lineno = {}
         # Try to load readline if it exists
         try:
             import readline
@@ -322,28 +190,43 @@
         self.stack = []
         self.curindex = 0
         self.curframe = None
+        self.tb_lineno.clear()
 
-    def setup(self, f, t):
+    def setup(self, f, tb):
         self.forget()
-        self.stack, self.curindex = self.get_stack(f, t)
+        self.stack, self.curindex = self.get_stack(f, tb)
+        while tb:
+            # when setting up post-mortem debugging with a traceback, save all
+            # the original line numbers to be displayed along the current line
+            # numbers (which can be different, e.g. due to finally clauses)
+            lineno = lasti2lineno(tb.tb_frame.f_code, tb.tb_lasti)
+            self.tb_lineno[tb.tb_frame] = lineno
+            tb = tb.tb_next
         self.curframe = self.stack[self.curindex][0]
         # The f_locals dictionary is updated from the actual frame
         # locals whenever the .f_locals accessor is called, so we
         # cache it here to ensure that modifications are not overwritten.
         self.curframe_locals = self.curframe.f_locals
-        self.execRcLines()
+        return self.execRcLines()
 
     # Can be executed earlier than 'setup' if desired
     def execRcLines(self):
-        if self.rcLines:
-            # Make local copy because of recursion
-            rcLines = self.rcLines
-            # executed only once
-            self.rcLines = []
-            for line in rcLines:
-                line = line[:-1]
-                if len(line) > 0 and line[0] != '#':
-                    self.onecmd(line)
+        if not self.rcLines:
+            return
+        # local copy because of recursion
+        rcLines = self.rcLines
+        rcLines.reverse()
+        # execute every line only once
+        self.rcLines = []
+        while rcLines:
+            line = rcLines.pop().strip()
+            if line and line[0] != '#':
+                if self.onecmd(line):
+                    # if onecmd returns True, the command wants to exit
+                    # from the interaction, save leftover rc lines
+                    # to execute before next interaction
+                    self.rcLines += reversed(rcLines)
+                    return True
 
     # Override Bdb methods
 
@@ -353,7 +236,7 @@
         if self._wait_for_mainpyfile:
             return
         if self.stop_here(frame):
-            print('--Call--', file=self.stdout)
+            self.message('--Call--')
             self.interaction(frame, None)
 
     def user_line(self, frame):
@@ -366,7 +249,7 @@
         if self.bp_commands(frame):
             self.interaction(frame, None)
 
-    def bp_commands(self,frame):
+    def bp_commands(self, frame):
         """Call every command that was set for the current active breakpoint
         (if there is one).
 
@@ -392,23 +275,31 @@
 
     def user_return(self, frame, return_value):
         """This function is called when a return trap is set here."""
+        if self._wait_for_mainpyfile:
+            return
         frame.f_locals['__return__'] = return_value
-        print('--Return--', file=self.stdout)
+        self.message('--Return--')
         self.interaction(frame, None)
 
     def user_exception(self, frame, exc_info):
         """This function is called if an exception occurs,
         but only if we are to stop at or just below this level."""
+        if self._wait_for_mainpyfile:
+            return
         exc_type, exc_value, exc_traceback = exc_info
         frame.f_locals['__exception__'] = exc_type, exc_value
-        exc_type_name = exc_type.__name__
-        print(exc_type_name + ':', _saferepr(exc_value), file=self.stdout)
+        self.message(traceback.format_exception_only(exc_type,
+                                                     exc_value)[-1].strip())
         self.interaction(frame, exc_traceback)
 
     # General interaction function
 
     def interaction(self, frame, traceback):
-        self.setup(frame, traceback)
+        if self.setup(frame, traceback):
+            # no interaction desired at this time (happens if .pdbrc contains
+            # a command like "continue")
+            self.forget()
+            return
         self.print_stack_entry(self.stack[self.curindex])
         self.cmdloop()
         self.forget()
@@ -419,7 +310,7 @@
         """
         # reproduce the behavior of the standard displayhook, not printing None
         if obj is not None:
-            print(repr(obj))
+            self.message(repr(obj))
 
     def default(self, line):
         if line[:1] == '!': line = line[1:]
@@ -440,11 +331,8 @@
                 sys.stdin = save_stdin
                 sys.displayhook = save_displayhook
         except:
-            t, v = sys.exc_info()[:2]
-            if type(t) == type(''):
-                exc_type_name = t
-            else: exc_type_name = t.__name__
-            print('***', exc_type_name + ':', v, file=self.stdout)
+            exc_info = sys.exc_info()[:2]
+            self.error(traceback.format_exception_only(*exc_info)[-1].strip())
 
     def precmd(self, line):
         """Handle alias expansion and ';;' separator."""
@@ -483,9 +371,11 @@
         else:
             return self.handle_command_def(line)
 
-    def handle_command_def(self,line):
-        """ Handles one command line during command list definition. """
+    def handle_command_def(self, line):
+        """Handles one command line during command list definition."""
         cmd, arg, line = self.parseline(line)
+        if not cmd:
+            return
         if cmd == 'silent':
             self.commands_silent[self.commands_bnum] = True
             return # continue to handle other cmd def in the cmd list
@@ -493,7 +383,7 @@
             self.cmdqueue = []
             return 1 # end of cmd list
         cmdlist = self.commands[self.commands_bnum]
-        if (arg):
+        if arg:
             cmdlist.append(cmd+' '+arg)
         else:
             cmdlist.append(cmd)
@@ -509,45 +399,116 @@
             return 1
         return
 
+    # interface abstraction functions
+
+    def message(self, msg):
+        print(msg, file=self.stdout)
+
+    def error(self, msg):
+        print('***', msg, file=self.stdout)
+
     # Command definitions, called by cmdloop()
     # The argument is the remaining string on the command line
     # Return true to exit from the command loop
 
-    do_h = cmd.Cmd.do_help
-
     def do_commands(self, arg):
-        """Defines a list of commands associated to a breakpoint.
+        """commands [bpnumber]
+        (com) ...
+        (com) end
+        (Pdb)
+
+        Specify a list of commands for breakpoint number bpnumber.
+        The commands themselves are entered on the following lines.
+        Type a line containing just 'end' to terminate the commands.
+        The commands are executed when the breakpoint is hit.
+
+        To remove all commands from a breakpoint, type commands and
+        follow it immediately with end; that is, give no commands.
+
+        With no bpnumber argument, commands refers to the last
+        breakpoint set.
 
-        Those commands will be executed whenever the breakpoint causes
-        the program to stop execution."""
+        You can use breakpoint commands to start your program up
+        again.  Simply use the continue command, or step, or any other
+        command that resumes execution.
+
+        Specifying any command resuming execution (currently continue,
+        step, next, return, jump, quit and their abbreviations)
+        terminates the command list (as if that command was
+        immediately followed by end).  This is because any time you
+        resume execution (even with a simple next or step), you may
+        encounter another breakpoint -- which could have its own
+        command list, leading to ambiguities about which list to
+        execute.
+
+        If you use the 'silent' command in the command list, the usual
+        message about stopping at a breakpoint is not printed.  This
+        may be desirable for breakpoints that are to print a specific
+        message and then continue.  If none of the other commands
+        print anything, you will see no sign that the breakpoint was
+        reached.
+        """
         if not arg:
-            bnum = len(bdb.Breakpoint.bpbynumber)-1
+            bnum = len(bdb.Breakpoint.bpbynumber) - 1
         else:
             try:
                 bnum = int(arg)
             except:
-                print("Usage : commands [bnum]\n        ...\n        end",
-                      file=self.stdout)
+                self.error("Usage: commands [bnum]\n        ...\n        end")
                 return
         self.commands_bnum = bnum
+        # Save old definitions for the case of a keyboard interrupt.
+        if bnum in self.commands:
+            old_command_defs = (self.commands[bnum],
+                                self.commands_doprompt[bnum],
+                                self.commands_silent[bnum])
+        else:
+            old_command_defs = None
         self.commands[bnum] = []
         self.commands_doprompt[bnum] = True
         self.commands_silent[bnum] = False
+
         prompt_back = self.prompt
         self.prompt = '(com) '
         self.commands_defining = True
-        self.cmdloop()
-        self.commands_defining = False
-        self.prompt = prompt_back
+        try:
+            self.cmdloop()
+        except KeyboardInterrupt:
+            # Restore old definitions.
+            if old_command_defs:
+                self.commands[bnum] = old_command_defs[0]
+                self.commands_doprompt[bnum] = old_command_defs[1]
+                self.commands_silent[bnum] = old_command_defs[2]
+            else:
+                del self.commands[bnum]
+                del self.commands_doprompt[bnum]
+                del self.commands_silent[bnum]
+            self.error('command definition aborted, old commands restored')
+        finally:
+            self.commands_defining = False
+            self.prompt = prompt_back
 
     def do_break(self, arg, temporary = 0):
-        # break [ ([filename:]lineno | function) [, "condition"] ]
+        """b(reak) [ ([filename:]lineno | function) [, condition] ]
+        Without argument, list all breaks.
+
+        With a line number argument, set a break at this line in the
+        current file.  With a function name, set a break at the first
+        executable line of that function.  If a second argument is
+        present, it is a string specifying an expression which must
+        evaluate to true before the breakpoint is honored.
+
+        The line number may be prefixed with a filename and a colon,
+        to specify a breakpoint in another file (probably one that
+        hasn't been loaded yet).  The file is searched for on
+        sys.path; the .py suffix may be omitted.
+        """
         if not arg:
             if self.breaks:  # There's at least one
-                print("Num Type         Disp Enb   Where", file=self.stdout)
+                self.message("Num Type         Disp Enb   Where")
                 for bp in bdb.Breakpoint.bpbynumber:
                     if bp:
-                        bp.bpprint(self.stdout)
+                        self.message(bp.bpformat())
             return
         # parse arguments; comma has lowest precedence
         # and cannot occur in filename
@@ -566,8 +527,7 @@
             filename = arg[:colon].rstrip()
             f = self.lookupmodule(filename)
             if not f:
-                print('*** ', repr(filename), end=' ', file=self.stdout)
-                print('not found from sys.path', file=self.stdout)
+                self.error('%r not found from sys.path' % filename)
                 return
             else:
                 filename = f
@@ -575,7 +535,7 @@
             try:
                 lineno = int(arg)
             except ValueError as msg:
-                print('*** Bad lineno:', arg, file=self.stdout)
+                self.error('Bad lineno: %s' % arg)
                 return
         else:
             # no colon; can be lineno or function
@@ -601,10 +561,8 @@
                     # last thing to try
                     (ok, filename, ln) = self.lineinfo(arg)
                     if not ok:
-                        print('*** The specified object', end=' ', file=self.stdout)
-                        print(repr(arg), end=' ', file=self.stdout)
-                        print('is not a function', file=self.stdout)
-                        print('or was not found along sys.path.', file=self.stdout)
+                        self.error('The specified object %r is not a function '
+                                   'or was not found along sys.path.' % arg)
                         return
                     funcname = ok # ok contains a function name
                     lineno = int(ln)
@@ -615,12 +573,12 @@
         if line:
             # now set the break point
             err = self.set_break(filename, line, temporary, cond, funcname)
-            if err: print('***', err, file=self.stdout)
+            if err:
+                self.error(err, file=self.stdout)
             else:
                 bp = self.get_breaks(filename, line)[-1]
-                print("Breakpoint %d at %s:%d" % (bp.number,
-                                                                 bp.file,
-                                                                 bp.line), file=self.stdout)
+                self.message("Breakpoint %d at %s:%d" %
+                             (bp.number, bp.file, bp.line))
 
     # To be overridden in derived debuggers
     def defaultFile(self):
@@ -633,6 +591,10 @@
     do_b = do_break
 
     def do_tbreak(self, arg):
+        """tbreak [ ([filename:]lineno | function) [, condition] ]
+        Same arguments as break, but sets a temporary breakpoint: it
+        is automatically deleted when first hit.
+        """
         self.do_break(arg, 1)
 
     def lineinfo(self, identifier):
@@ -674,114 +636,115 @@
         Return `lineno` if it is, 0 if not (e.g. a docstring, comment, blank
         line or EOF). Warning: testing is not comprehensive.
         """
-        line = linecache.getline(filename, lineno, self.curframe.f_globals)
+        # this method should be callable before starting debugging, so default
+        # to "no globals" if there is no current frame
+        globs = self.curframe.f_globals if hasattr(self, 'curframe') else None
+        line = linecache.getline(filename, lineno, globs)
         if not line:
-            print('End of file', file=self.stdout)
+            self.message('End of file')
             return 0
         line = line.strip()
         # Don't allow setting breakpoint at a blank line
         if (not line or (line[0] == '#') or
              (line[:3] == '"""') or line[:3] == "'''"):
-            print('*** Blank or comment', file=self.stdout)
+            self.error('Blank or comment')
             return 0
         return lineno
 
     def do_enable(self, arg):
+        """enable bpnumber [bpnumber ...]
+        Enables the breakpoints given as a space separated list of
+        breakpoint numbers.
+        """
         args = arg.split()
         for i in args:
             try:
-                i = int(i)
-            except ValueError:
-                print('Breakpoint index %r is not a number' % i, file=self.stdout)
-                continue
-
-            if not (0 <= i < len(bdb.Breakpoint.bpbynumber)):
-                print('No breakpoint numbered', i, file=self.stdout)
-                continue
-
-            bp = bdb.Breakpoint.bpbynumber[i]
-            if bp:
+                bp = self.get_bpbynumber(i)
+            except ValueError as err:
+                self.error(err)
+            else:
                 bp.enable()
+                self.message('Enabled %s' % bp)
 
     def do_disable(self, arg):
+        """disable bpnumber [bpnumber ...]
+        Disables the breakpoints given as a space separated list of
+        breakpoint numbers.  Disabling a breakpoint means it cannot
+        cause the program to stop execution, but unlike clearing a
+        breakpoint, it remains in the list of breakpoints and can be
+        (re-)enabled.
+        """
         args = arg.split()
         for i in args:
             try:
-                i = int(i)
-            except ValueError:
-                print('Breakpoint index %r is not a number' % i, file=self.stdout)
-                continue
-
-            if not (0 <= i < len(bdb.Breakpoint.bpbynumber)):
-                print('No breakpoint numbered', i, file=self.stdout)
-                continue
-
-            bp = bdb.Breakpoint.bpbynumber[i]
-            if bp:
+                bp = self.get_bpbynumber(i)
+            except ValueError as err:
+                self.error(err)
+            else:
                 bp.disable()
+                self.message('Disabled %s' % bp)
 
     def do_condition(self, arg):
-        # arg is breakpoint number and condition
+        """condition bpnumber [condition]
+        Set a new condition for the breakpoint, an expression which
+        must evaluate to true before the breakpoint is honored.  If
+        condition is absent, any existing condition is removed; i.e.,
+        the breakpoint is made unconditional.
+        """
         args = arg.split(' ', 1)
         try:
-            bpnum = int(args[0].strip())
-        except ValueError:
-            # something went wrong
-            print('Breakpoint index %r is not a number' % args[0], file=self.stdout)
-            return
-        try:
             cond = args[1]
-        except:
+        except IndexError:
             cond = None
         try:
-            bp = bdb.Breakpoint.bpbynumber[bpnum]
-        except IndexError:
-            print('Breakpoint index %r is not valid' % args[0],
-                  file=self.stdout)
-            return
-        if bp:
+            bp = self.get_bpbynumber(args[0].strip())
+        except ValueError as err:
+            self.error(err)
+        else:
             bp.cond = cond
             if not cond:
-                print('Breakpoint', bpnum, end=' ', file=self.stdout)
-                print('is now unconditional.', file=self.stdout)
+                self.message('Breakpoint %d is now unconditional.' % bp.number)
+            else:
+                self.message('New condition set for breakpoint %d.' % bp.number)
 
-    def do_ignore(self,arg):
-        """arg is bp number followed by ignore count."""
+    def do_ignore(self, arg):
+        """ignore bpnumber [count]
+        Set the ignore count for the given breakpoint number.  If
+        count is omitted, the ignore count is set to 0.  A breakpoint
+        becomes active when the ignore count is zero.  When non-zero,
+        the count is decremented each time the breakpoint is reached
+        and the breakpoint is not disabled and any associated
+        condition evaluates to true.
+        """
         args = arg.split()
         try:
-            bpnum = int(args[0].strip())
-        except ValueError:
-            # something went wrong
-            print('Breakpoint index %r is not a number' % args[0], file=self.stdout)
-            return
-        try:
             count = int(args[1].strip())
         except:
             count = 0
         try:
-            bp = bdb.Breakpoint.bpbynumber[bpnum]
-        except IndexError:
-            print('Breakpoint index %r is not valid' % args[0],
-                  file=self.stdout)
-            return
-        if bp:
+            bp = self.get_bpbynumber(args[0].strip())
+        except ValueError as err:
+            self.error(err)
+        else:
             bp.ignore = count
             if count > 0:
-                reply = 'Will ignore next '
                 if count > 1:
-                    reply = reply + '%d crossings' % count
+                    countstr = '%d crossings' % count
                 else:
-                    reply = reply + '1 crossing'
-                print(reply + ' of breakpoint %d.' % bpnum, file=self.stdout)
+                    countstr = '1 crossing'
+                self.message('Will ignore next %s of breakpoint %d.' %
+                             (countstr, bp.number))
             else:
-                print('Will stop next time breakpoint', end=' ', file=self.stdout)
-                print(bpnum, 'is reached.', file=self.stdout)
+                self.message('Will stop next time breakpoint %d is reached.'
+                             % bp.number)
 
     def do_clear(self, arg):
-        """Three possibilities, tried in this order:
-        clear -> clear all breaks, ask for confirmation
-        clear file:lineno -> clear all breaks at file:lineno
-        clear bpno bpno ... -> clear breakpoints by number"""
+        """cl(ear) filename:lineno\ncl(ear) [bpnumber [bpnumber...]]
+        With a space separated list of breakpoint numbers, clear
+        those breakpoints.  Without argument, clear all breaks (but
+        first ask confirmation).  With a filename:lineno argument,
+        clear all breaks at that line in that file.
+        """
         if not arg:
             try:
                 reply = input('Clear all breaks? ')
@@ -789,7 +752,10 @@
                 reply = 'no'
             reply = reply.strip().lower()
             if reply in ('y', 'yes'):
+                bplist = [bp for bp in bdb.Breakpoint.bpbynumber if bp]
                 self.clear_all_breaks()
+                for bp in bplist:
+                    self.message('Deleted %s' % bp)
             return
         if ':' in arg:
             # Make sure it works for "clear C:\foo\bar.py:12"
@@ -801,28 +767,31 @@
             except ValueError:
                 err = "Invalid line number (%s)" % arg
             else:
+                bplist = self.get_breaks(filename, lineno)
                 err = self.clear_break(filename, lineno)
-            if err: print('***', err, file=self.stdout)
+            if err:
+                self.error(err)
+            else:
+                for bp in bplist:
+                    self.message('Deleted %s' % bp)
             return
         numberlist = arg.split()
         for i in numberlist:
             try:
-                i = int(i)
-            except ValueError:
-                print('Breakpoint index %r is not a number' % i, file=self.stdout)
-                continue
-
-            if not (0 <= i < len(bdb.Breakpoint.bpbynumber)):
-                print('No breakpoint numbered', i, file=self.stdout)
-                continue
-            err = self.clear_bpbynumber(i)
-            if err:
-                print('***', err, file=self.stdout)
+                bp = self.get_bpbynumber(i)
+            except ValueError as err:
+                self.error(err)
             else:
-                print('Deleted breakpoint', i, file=self.stdout)
+                self.clear_break(bp.file, bp.line)
+                self.message('Deleted %s' % bp)
     do_cl = do_clear # 'c' is already an abbreviation for 'continue'
 
     def do_where(self, arg):
+        """w(here)
+        Print a stack trace, with the most recent frame at the bottom.
+        An arrow indicates the "current frame", which determines the
+        context of most commands.  'bt' is an alias for this command.
+        """
         self.print_stack_trace()
     do_w = do_where
     do_bt = do_where
@@ -836,13 +805,17 @@
         self.lineno = None
 
     def do_up(self, arg):
+        """u(p) [count]
+        Move the current frame count (default one) levels up in the
+        stack trace (to an older frame).
+        """
         if self.curindex == 0:
-            print('*** Oldest frame', file=self.stdout)
+            self.error('Oldest frame')
             return
         try:
             count = int(arg or 1)
         except ValueError:
-            print('*** Invalid frame count (%s)' % arg, file=self.stdout)
+            self.error('Invalid frame count (%s)' % arg)
             return
         if count < 0:
             newframe = 0
@@ -852,13 +825,17 @@
     do_u = do_up
 
     def do_down(self, arg):
+        """d(own) [count]
+        Move the current frame count (default one) levels down in the
+        stack trace (to a newer frame).
+        """
         if self.curindex + 1 == len(self.stack):
-            print('*** Newest frame', file=self.stdout)
+            self.error('Newest frame')
             return
         try:
             count = int(arg or 1)
         except ValueError:
-            print('*** Invalid frame count (%s)' % arg, file=self.stdout)
+            self.error('Invalid frame count (%s)' % arg)
             return
         if count < 0:
             newframe = len(self.stack) - 1
@@ -868,50 +845,99 @@
     do_d = do_down
 
     def do_until(self, arg):
-        self.set_until(self.curframe)
+        """unt(il) [lineno]
+        Without argument, continue execution until the line with a
+        number greater than the current one is reached.  With a line
+        number, continue execution until a line with a number greater
+        or equal to that is reached.  In both cases, also stop when
+        the current frame returns.
+        """
+        if arg:
+            try:
+                lineno = int(arg)
+            except ValueError:
+                self.error('Error in argument: %r' % arg)
+                return
+            if lineno <= self.curframe.f_lineno:
+                self.error('"until" line number is smaller than current '
+                           'line number')
+                return
+        else:
+            lineno = None
+        self.set_until(self.curframe, lineno)
         return 1
     do_unt = do_until
 
     def do_step(self, arg):
+        """s(tep)
+        Execute the current line, stop at the first possible occasion
+        (either in a function that is called or in the current
+        function).
+        """
         self.set_step()
         return 1
     do_s = do_step
 
     def do_next(self, arg):
+        """n(ext)
+        Continue execution until the next line in the current function
+        is reached or it returns.
+        """
         self.set_next(self.curframe)
         return 1
     do_n = do_next
 
     def do_run(self, arg):
-        """Restart program by raising an exception to be caught in the main
-        debugger loop.  If arguments were given, set them in sys.argv."""
+        """run [args...]
+        Restart the debugged python program. If a string is supplied
+        it is splitted with "shlex", and the result is used as the new
+        sys.argv.  History, breakpoints, actions and debugger options
+        are preserved.  "restart" is an alias for "run".
+        """
         if arg:
             import shlex
             argv0 = sys.argv[0:1]
             sys.argv = shlex.split(arg)
             sys.argv[:0] = argv0
+        # this is caught in the main debugger loop
         raise Restart
 
     do_restart = do_run
 
     def do_return(self, arg):
+        """r(eturn)
+        Continue execution until the current function returns.
+        """
         self.set_return(self.curframe)
         return 1
     do_r = do_return
 
     def do_continue(self, arg):
+        """c(ont(inue))
+        Continue execution, only stop when a breakpoint is encountered.
+        """
         self.set_continue()
         return 1
     do_c = do_cont = do_continue
 
     def do_jump(self, arg):
+        """j(ump) lineno
+        Set the next line that will be executed.  Only available in
+        the bottom-most frame.  This lets you jump back and execute
+        code again, or jump forward to skip code that you don't want
+        to run.
+
+        It should be noted that not all jumps are allowed -- for
+        instance it is not possible to jump into the middle of a
+        for loop or out of a finally clause.
+        """
         if self.curindex + 1 != len(self.stack):
-            print("*** You can only jump within the bottom frame", file=self.stdout)
+            self.error('You can only jump within the bottom frame')
             return
         try:
             arg = int(arg)
         except ValueError:
-            print("*** The 'jump' command requires a line number.", file=self.stdout)
+            self.error("The 'jump' command requires a line number")
         else:
             try:
                 # Do the jump, fix up our copy of the stack, and display the
@@ -920,22 +946,30 @@
                 self.stack[self.curindex] = self.stack[self.curindex][0], arg
                 self.print_stack_entry(self.stack[self.curindex])
             except ValueError as e:
-                print('*** Jump failed:', e, file=self.stdout)
+                self.error('Jump failed: %s' % e)
     do_j = do_jump
 
     def do_debug(self, arg):
+        """debug code
+        Enter a recursive debugger that steps through the code
+        argument (which is an arbitrary expression or statement to be
+        executed in the current environment).
+        """
         sys.settrace(None)
         globals = self.curframe.f_globals
         locals = self.curframe_locals
         p = Pdb(self.completekey, self.stdin, self.stdout)
         p.prompt = "(%s) " % self.prompt.strip()
-        print("ENTERING RECURSIVE DEBUGGER", file=self.stdout)
+        self.message("ENTERING RECURSIVE DEBUGGER")
         sys.call_tracing(p.run, (arg, globals, locals))
-        print("LEAVING RECURSIVE DEBUGGER", file=self.stdout)
+        self.message("LEAVING RECURSIVE DEBUGGER")
         sys.settrace(self.trace_dispatch)
         self.lastcmd = p.lastcmd
 
     def do_quit(self, arg):
+        """q(uit)\nexit
+        Quit from the debugger. The program being executed is aborted.
+        """
         self._user_requested_quit = 1
         self.set_quit()
         return 1
@@ -944,12 +978,18 @@
     do_exit = do_quit
 
     def do_EOF(self, arg):
-        print(file=self.stdout)
+        """EOF
+        Handles the receipt of EOF as a command.
+        """
+        self.message('')
         self._user_requested_quit = 1
         self.set_quit()
         return 1
 
     def do_args(self, arg):
+        """a(rgs)
+        Print the argument list of the current function.
+        """
         co = self.curframe.f_code
         dict = self.curframe_locals
         n = co.co_argcount
@@ -957,62 +997,83 @@
         if co.co_flags & 8: n = n+1
         for i in range(n):
             name = co.co_varnames[i]
-            print(name, '=', end=' ', file=self.stdout)
-            if name in dict: print(dict[name], file=self.stdout)
-            else: print("*** undefined ***", file=self.stdout)
+            if name in dict:
+                self.message('%s = %r' % (name, dict[name]))
+            else:
+                self.message('%s = *** undefined ***' % (name,))
     do_a = do_args
 
     def do_retval(self, arg):
+        """retval
+        Print the return value for the last return of a function.
+        """
         if '__return__' in self.curframe_locals:
-            print(self.curframe_locals['__return__'], file=self.stdout)
+            self.message(repr(self.curframe_locals['__return__']))
         else:
-            print('*** Not yet returned!', file=self.stdout)
+            self.error('Not yet returned!')
     do_rv = do_retval
 
     def _getval(self, arg):
         try:
             return eval(arg, self.curframe.f_globals, self.curframe_locals)
         except:
-            t, v = sys.exc_info()[:2]
-            if isinstance(t, str):
-                exc_type_name = t
-            else: exc_type_name = t.__name__
-            print('***', exc_type_name + ':', repr(v), file=self.stdout)
+            exc_info = sys.exc_info()[:2]
+            self.error(traceback.format_exception_only(*exc_info)[-1].strip())
             raise
 
     def do_p(self, arg):
+        """p(rint) expression
+        Print the value of the expression.
+        """
         try:
-            print(repr(self._getval(arg)), file=self.stdout)
+            self.message(repr(self._getval(arg)))
         except:
             pass
     # make "print" an alias of "p" since print isn't a Python statement anymore
     do_print = do_p
 
     def do_pp(self, arg):
+        """pp expression
+        Pretty-print the value of the expression.
+        """
         try:
-            pprint.pprint(self._getval(arg), self.stdout)
+            self.message(pprint.pformat(self._getval(arg)))
         except:
             pass
 
     def do_list(self, arg):
+        """l(ist) [first [,last] | .]
+
+        List source code for the current file.  Without arguments,
+        list 11 lines around the current line or continue the previous
+        listing.  With . as argument, list 11 lines around the current
+        line.  With one argument, list 11 lines starting at that line.
+        With two arguments, list the given range; if the second
+        argument is less than the first, it is a count.
+
+        The current line in the current frame is indicated by "->".
+        If an exception is being debugged, the line where the
+        exception was originally raised or propagated is indicated by
+        ">>", if it differs from the current line.
+        """
         self.lastcmd = 'list'
         last = None
-        if arg:
+        if arg and arg != '.':
             try:
-                x = eval(arg, {}, {})
-                if type(x) == type(()):
-                    first, last = x
-                    first = int(first)
-                    last = int(last)
+                if ',' in arg:
+                    first, last = arg.split(',')
+                    first = int(first.strip())
+                    last = int(last.strip())
                     if last < first:
-                        # Assume it's a count
+                        # assume it's a count
                         last = first + last
                 else:
-                    first = max(1, int(x) - 5)
-            except:
-                print('*** Error in argument:', repr(arg), file=self.stdout)
+                    first = int(arg.strip())
+                    first = max(1, first - 5)
+            except ValueError:
+                self.error('Error in argument: %r' % arg)
                 return
-        elif self.lineno is None:
+        elif self.lineno is None or arg == '.':
             first = max(1, self.curframe.f_lineno - 5)
         else:
             first = self.lineno + 1
@@ -1021,71 +1082,144 @@
         filename = self.curframe.f_code.co_filename
         breaklist = self.get_file_breaks(filename)
         try:
-            for lineno in range(first, last+1):
-                line = linecache.getline(filename, lineno,
-                                         self.curframe.f_globals)
-                if not line:
-                    print('[EOF]', file=self.stdout)
-                    break
-                else:
-                    s = repr(lineno).rjust(3)
-                    if len(s) < 4: s = s + ' '
-                    if lineno in breaklist: s = s + 'B'
-                    else: s = s + ' '
-                    if lineno == self.curframe.f_lineno:
-                        s = s + '->'
-                    print(s + '\t' + line, end='', file=self.stdout)
-                    self.lineno = lineno
+            lines = linecache.getlines(filename, self.curframe.f_globals)
+            self._print_lines(lines[first-1:last], first, breaklist,
+                              self.curframe)
+            self.lineno = min(last, len(lines))
+            if len(lines) < last:
+                self.message('[EOF]')
         except KeyboardInterrupt:
             pass
     do_l = do_list
 
+    def do_longlist(self, arg):
+        """longlist | ll
+        List the whole source code for the current function or frame.
+        """
+        filename = self.curframe.f_code.co_filename
+        breaklist = self.get_file_breaks(filename)
+        try:
+            lines, lineno = getsourcelines(self.curframe)
+        except IOError as err:
+            self.error(err)
+            return
+        self._print_lines(lines, lineno, breaklist, self.curframe)
+    do_ll = do_longlist
+
+    def do_source(self, arg):
+        """source expression
+        Try to get source code for the given object and display it.
+        """
+        try:
+            obj = self._getval(arg)
+        except:
+            return
+        try:
+            lines, lineno = getsourcelines(obj)
+        except (IOError, TypeError) as err:
+            self.error(err)
+            return
+        self._print_lines(lines, lineno)
+
+    def _print_lines(self, lines, start, breaks=(), frame=None):
+        """Print a range of lines."""
+        if frame:
+            current_lineno = frame.f_lineno
+            exc_lineno = self.tb_lineno.get(frame, -1)
+        else:
+            current_lineno = exc_lineno = -1
+        for lineno, line in enumerate(lines, start):
+            s = str(lineno).rjust(3)
+            if len(s) < 4:
+                s += ' '
+            if lineno in breaks:
+                s += 'B'
+            else:
+                s += ' '
+            if lineno == current_lineno:
+                s += '->'
+            elif lineno == exc_lineno:
+                s += '>>'
+            self.message(s + '\t' + line.rstrip())
+
     def do_whatis(self, arg):
+        """whatis arg
+        Print the type of the argument.
+        """
         try:
-            value = eval(arg, self.curframe.f_globals,
-                            self.curframe_locals)
+            value = self._getval(arg)
         except:
-            t, v = sys.exc_info()[:2]
-            if type(t) == type(''):
-                exc_type_name = t
-            else: exc_type_name = t.__name__
-            print('***', exc_type_name + ':', repr(v), file=self.stdout)
+            # _getval() already printed the error
             return
         code = None
         # Is it a function?
-        try: code = value.__code__
-        except: pass
+        try:
+            code = value.__code__
+        except Exception:
+            pass
         if code:
-            print('Function', code.co_name, file=self.stdout)
+            self.message('Function %s' % code.co_name)
             return
         # Is it an instance method?
-        try: code = value.__func__.__code__
-        except: pass
+        try:
+            code = value.__func__.__code__
+        except Exception:
+            pass
         if code:
-            print('Method', code.co_name, file=self.stdout)
+            self.message('Method %s' % code.co_name)
+            return
+        # Is it a class?
+        if value.__class__ is type:
+            self.message('Class %s.%s' % (value.__module__, value.__name__))
             return
         # None of the above...
-        print(type(value), file=self.stdout)
+        self.message(type(value))
 
     def do_alias(self, arg):
+        """alias [name [command [parameter parameter ...] ]]
+        Create an alias called 'name' that executes 'command'.  The
+        command must *not* be enclosed in quotes.  Replaceable
+        parameters can be indicated by %1, %2, and so on, while %* is
+        replaced by all the parameters.  If no command is given, the
+        current alias for name is shown. If no name is given, all
+        aliases are listed.
+
+        Aliases may be nested and can contain anything that can be
+        legally typed at the pdb prompt.  Note!  You *can* override
+        internal pdb commands with aliases!  Those internal commands
+        are then hidden until the alias is removed.  Aliasing is
+        recursively applied to the first word of the command line; all
+        other words in the line are left alone.
+
+        As an example, here are two useful aliases (especially when
+        placed in the .pdbrc file):
+
+        # Print instance variables (usage "pi classInst")
+        alias pi for k in %1.__dict__.keys(): print "%1.",k,"=",%1.__dict__[k]
+        # Print instance variables in self
+        alias ps pi self
+        """
         args = arg.split()
         if len(args) == 0:
             keys = sorted(self.aliases.keys())
             for alias in keys:
-                print("%s = %s" % (alias, self.aliases[alias]), file=self.stdout)
+                self.message("%s = %s" % (alias, self.aliases[alias]))
             return
         if args[0] in self.aliases and len(args) == 1:
-            print("%s = %s" % (args[0], self.aliases[args[0]]), file=self.stdout)
+            self.message("%s = %s" % (args[0], self.aliases[args[0]]))
         else:
             self.aliases[args[0]] = ' '.join(args[1:])
 
     def do_unalias(self, arg):
+        """unalias name
+        Delete the specified alias.
+        """
         args = arg.split()
         if len(args) == 0: return
         if args[0] in self.aliases:
             del self.aliases[args[0]]
 
-    #list of all the commands making the program resume execution.
+    # List of all the commands making the program resume execution.
     commands_resuming = ['do_continue', 'do_step', 'do_next', 'do_return',
                          'do_quit', 'do_jump']
 
@@ -1107,292 +1241,53 @@
     def print_stack_entry(self, frame_lineno, prompt_prefix=line_prefix):
         frame, lineno = frame_lineno
         if frame is self.curframe:
-            print('>', end=' ', file=self.stdout)
+            prefix = '> '
         else:
-            print(' ', end=' ', file=self.stdout)
-        print(self.format_stack_entry(frame_lineno,
-                                                     prompt_prefix), file=self.stdout)
-
-
-    # Help methods (derived from docstring)
-
-    def help_help(self):
-        self.help_h()
-
-    def help_h(self):
-        print("""h(elp)
-Without argument, print the list of available commands.
-With a command name as argument, print help about that command
-"help pdb" shows the full pdb documentation
-"help exec" gives help on the ! command""", file=self.stdout)
-
-    def help_where(self):
-        self.help_w()
-
-    def help_w(self):
-        print("""w(here)
-Print a stack trace, with the most recent frame at the bottom.
-An arrow indicates the "current frame", which determines the
-context of most commands.  'bt' is an alias for this command.""", file=self.stdout)
-
-    help_bt = help_w
-
-    def help_down(self):
-        self.help_d()
-
-    def help_d(self):
-        print("""d(own)
-Move the current frame one level down in the stack trace
-(to a newer frame).""", file=self.stdout)
-
-    def help_up(self):
-        self.help_u()
-
-    def help_u(self):
-        print("""u(p)
-Move the current frame one level up in the stack trace
-(to an older frame).""", file=self.stdout)
-
-    def help_break(self):
-        self.help_b()
-
-    def help_b(self):
-        print("""b(reak) ([file:]lineno | function) [, condition]
-With a line number argument, set a break there in the current
-file.  With a function name, set a break at first executable line
-of that function.  Without argument, list all breaks.  If a second
-argument is present, it is a string specifying an expression
-which must evaluate to true before the breakpoint is honored.
-
-The line number may be prefixed with a filename and a colon,
-to specify a breakpoint in another file (probably one that
-hasn't been loaded yet).  The file is searched for on sys.path;
-the .py suffix may be omitted.""", file=self.stdout)
-
-    def help_clear(self):
-        self.help_cl()
-
-    def help_cl(self):
-        print("cl(ear) filename:lineno", file=self.stdout)
-        print("""cl(ear) [bpnumber [bpnumber...]]
-With a space separated list of breakpoint numbers, clear
-those breakpoints.  Without argument, clear all breaks (but
-first ask confirmation).  With a filename:lineno argument,
-clear all breaks at that line in that file.""", file=self.stdout)
-
-    def help_tbreak(self):
-        print("""tbreak  same arguments as break, but breakpoint is
-removed when first hit.""", file=self.stdout)
-
-    def help_enable(self):
-        print("""enable bpnumber [bpnumber ...]
-Enables the breakpoints given as a space separated list of
-bp numbers.""", file=self.stdout)
-
-    def help_disable(self):
-        print("""disable bpnumber [bpnumber ...]
-Disables the breakpoints given as a space separated list of
-bp numbers.""", file=self.stdout)
-
-    def help_ignore(self):
-        print("""ignore bpnumber count
-Sets the ignore count for the given breakpoint number.  A breakpoint
-becomes active when the ignore count is zero.  When non-zero, the
-count is decremented each time the breakpoint is reached and the
-breakpoint is not disabled and any associated condition evaluates
-to true.""", file=self.stdout)
-
-    def help_condition(self):
-        print("""condition bpnumber str_condition
-str_condition is a string specifying an expression which
-must evaluate to true before the breakpoint is honored.
-If str_condition is absent, any existing condition is removed;
-i.e., the breakpoint is made unconditional.""", file=self.stdout)
-
-    def help_step(self):
-        self.help_s()
-
-    def help_s(self):
-        print("""s(tep)
-Execute the current line, stop at the first possible occasion
-(either in a function that is called or in the current function).""", file=self.stdout)
-
-    def help_until(self):
-        self.help_unt()
-
-    def help_unt(self):
-        print("""unt(il)
-Continue execution until the line with a number greater than the current
-one is reached or until the current frame returns""")
-
-    def help_next(self):
-        self.help_n()
-
-    def help_n(self):
-        print("""n(ext)
-Continue execution until the next line in the current function
-is reached or it returns.""", file=self.stdout)
-
-    def help_return(self):
-        self.help_r()
-
-    def help_r(self):
-        print("""r(eturn)
-Continue execution until the current function returns.""", file=self.stdout)
-
-    def help_continue(self):
-        self.help_c()
-
-    def help_cont(self):
-        self.help_c()
-
-    def help_c(self):
-        print("""c(ont(inue))
-Continue execution, only stop when a breakpoint is encountered.""", file=self.stdout)
-
-    def help_jump(self):
-        self.help_j()
-
-    def help_j(self):
-        print("""j(ump) lineno
-Set the next line that will be executed.""", file=self.stdout)
-
-    def help_debug(self):
-        print("""debug code
-Enter a recursive debugger that steps through the code argument
-(which is an arbitrary expression or statement to be executed
-in the current environment).""", file=self.stdout)
-
-    def help_list(self):
-        self.help_l()
-
-    def help_l(self):
-        print("""l(ist) [first [,last]]
-List source code for the current file.
-Without arguments, list 11 lines around the current line
-or continue the previous listing.
-With one argument, list 11 lines starting at that line.
-With two arguments, list the given range;
-if the second argument is less than the first, it is a count.""", file=self.stdout)
-
-    def help_args(self):
-        self.help_a()
-
-    def help_a(self):
-        print("""a(rgs)
-Print the arguments of the current function.""", file=self.stdout)
-
-    def help_p(self):
-        print("""p(rint) expression
-Print the value of the expression.""", file=self.stdout)
-
-    def help_pp(self):
-        print("""pp expression
-Pretty-print the value of the expression.""", file=self.stdout)
+            prefix = '  '
+        self.message(prefix +
+                     self.format_stack_entry(frame_lineno, prompt_prefix))
+
+    # Provide help
+
+    def do_help(self, arg):
+        """h(elp)
+        Without argument, print the list of available commands.
+        With a command name as argument, print help about that command.
+        "help pdb" shows the full pdb documentation.
+        "help exec" gives help on the ! command.
+        """
+        if not arg:
+            return cmd.Cmd.do_help(self, arg)
+        try:
+            try:
+                topic = getattr(self, 'help_' + arg)
+                return topic()
+            except AttributeError:
+                command = getattr(self, 'do_' + arg)
+        except AttributeError:
+            self.error('No help for %r' % arg)
+        else:
+            self.message(command.__doc__.rstrip())
+
+    do_h = do_help
 
     def help_exec(self):
-        print("""(!) statement
-Execute the (one-line) statement in the context of
-the current stack frame.
-The exclamation point can be omitted unless the first word
-of the statement resembles a debugger command.
-To assign to a global variable you must always prefix the
-command with a 'global' command, e.g.:
-(Pdb) global list_options; list_options = ['-l']
-(Pdb)""", file=self.stdout)
-
-    def help_run(self):
-        print("""run [args...]
-Restart the debugged python program. If a string is supplied, it is
-splitted with "shlex" and the result is used as the new sys.argv.
-History, breakpoints, actions and debugger options are preserved.
-"restart" is an alias for "run".""")
-
-    help_restart = help_run
-
-    def help_quit(self):
-        self.help_q()
-
-    def help_q(self):
-        print("""q(uit) or exit - Quit from the debugger.
-The program being executed is aborted.""", file=self.stdout)
-
-    help_exit = help_q
-
-    def help_whatis(self):
-        print("""whatis arg
-Prints the type of the argument.""", file=self.stdout)
-
-    def help_EOF(self):
-        print("""EOF
-Handles the receipt of EOF as a command.""", file=self.stdout)
-
-    def help_alias(self):
-        print("""alias [name [command [parameter parameter ...] ]]
-Creates an alias called 'name' the executes 'command'.  The command
-must *not* be enclosed in quotes.  Replaceable parameters are
-indicated by %1, %2, and so on, while %* is replaced by all the
-parameters.  If no command is given, the current alias for name
-is shown. If no name is given, all aliases are listed.
-
-Aliases may be nested and can contain anything that can be
-legally typed at the pdb prompt.  Note!  You *can* override
-internal pdb commands with aliases!  Those internal commands
-are then hidden until the alias is removed.  Aliasing is recursively
-applied to the first word of the command line; all other words
-in the line are left alone.
-
-Some useful aliases (especially when placed in the .pdbrc file) are:
-
-#Print instance variables (usage "pi classInst")
-alias pi for k in %1.__dict__.keys(): print "%1.",k,"=",%1.__dict__[k]
-
-#Print instance variables in self
-alias ps pi self
-""", file=self.stdout)
-
-    def help_unalias(self):
-        print("""unalias name
-Deletes the specified alias.""", file=self.stdout)
-
-    def help_commands(self):
-        print("""commands [bpnumber]
-(com) ...
-(com) end
-(Pdb)
-
-Specify a list of commands for breakpoint number bpnumber.  The
-commands themselves appear on the following lines.  Type a line
-containing just 'end' to terminate the commands.
-
-To remove all commands from a breakpoint, type commands and
-follow it immediately with  end; that is, give no commands.
-
-With no bpnumber argument, commands refers to the last
-breakpoint set.
-
-You can use breakpoint commands to start your program up again.
-Simply use the continue command, or step, or any other
-command that resumes execution.
-
-Specifying any command resuming execution (currently continue,
-step, next, return, jump, quit and their abbreviations) terminates
-the command list (as if that command was immediately followed by end).
-This is because any time you resume execution
-(even with a simple next or step), you may encounter
-another breakpoint--which could have its own command list, leading to
-ambiguities about which list to execute.
-
-   If you use the 'silent' command in the command list, the
-usual message about stopping at a breakpoint is not printed.  This may
-be desirable for breakpoints that are to print a specific message and
-then continue.  If none of the other commands print anything, you
-see no sign that the breakpoint was reached.
-""", file=self.stdout)
+        """(!) statement
+        Execute the (one-line) statement in the context of the current
+        stack frame.  The exclamation point can be omitted unless the
+        first word of the statement resembles a debugger command.  To
+        assign to a global variable you must always prefix the command
+        with a 'global' command, e.g.:
+        (Pdb) global list_options; list_options = ['-l']
+        (Pdb)
+        """
+        self.message(self.help_exec.__doc__.strip())
 
     def help_pdb(self):
         help()
 
+    # other helper functions
+
     def lookupmodule(self, filename):
         """Helper function for break/clear parsing -- may be overridden.
 
@@ -1443,6 +1338,24 @@
                         (fp.read(), self.mainpyfile)
         self.run(statement)
 
+# Collect all command help into docstring
+
+# unfortunately we can't guess this order from the class definition
+_help_order = [
+    'help', 'where', 'down', 'up', 'break', 'tbreak', 'clear', 'disable',
+    'enable', 'ignore', 'condition', 'commands', 'step', 'next', 'until',
+    'jump', 'return', 'retval', 'run', 'continue', 'list', 'longlist',
+    'args', 'print', 'pp', 'whatis', 'source', 'alias', 'unalias',
+    'debug', 'quit',
+]
+
+docs = set()
+for _command in _help_order:
+    __doc__ += getattr(Pdb, 'do_' + _command).__doc__.strip() + '\n\n'
+__doc__ += Pdb.help_exec.__doc__
+
+del _help_order, _command
+
 # Simplified interface
 
 def run(statement, globals=None, locals=None):
@@ -1469,9 +1382,9 @@
         # sys.exc_info() returns (type, value, traceback) if an exception is
         # being handled, otherwise it returns None
         t = sys.exc_info()[2]
-        if t is None:
-            raise ValueError("A valid traceback must be passed if no "
-                                               "exception is being handled")
+    if t is None:
+        raise ValueError("A valid traceback must be passed if no "
+                         "exception is being handled")
 
     p = Pdb()
     p.reset()
@@ -1493,17 +1406,42 @@
     import pydoc
     pydoc.pager(__doc__)
 
+_usage = """\
+usage: pdb.py [-c command] ... pyfile [arg] ...
+
+Debug the Python program given by pyfile.
+
+Initial commands are read from .pdbrc files in your home directory
+and in the current directory, if they exist.  Commands supplied with
+-c are executed after commands from .pdbrc files.
+
+To let the script run until an exception occurs, use "-c continue".
+To let the script run up to a given line X in the debugged file, use
+"-c 'until X'"."""
+
 def main():
-    if not sys.argv[1:] or sys.argv[1] in ("--help", "-h"):
-        print("usage: pdb.py scriptfile [arg] ...")
+    import getopt
+
+    opts, args = getopt.getopt(sys.argv[1:], 'hc:', ['--help', '--command='])
+
+    if not args:
+        print(_usage)
         sys.exit(2)
 
-    mainpyfile =  sys.argv[1]     # Get script filename
+    commands = []
+    for opt, optarg in opts:
+        if opt in ['-h', '--help']:
+            print(_usage)
+            sys.exit()
+        elif opt in ['-c', '--command']:
+            commands.append(optarg)
+
+    mainpyfile = args[0]     # Get script filename
     if not os.path.exists(mainpyfile):
         print('Error:', mainpyfile, 'does not exist')
         sys.exit(1)
 
-    del sys.argv[0]         # Hide "pdb.py" from argument list
+    sys.argv[:] = args      # Hide "pdb.py" and pdb options from argument list
 
     # Replace pdb's dir with script's dir in front of module search path.
     sys.path[0] = os.path.dirname(mainpyfile)
@@ -1513,7 +1451,8 @@
     # changed by the user from the command line. There is a "restart" command
     # which allows explicit specification of command line arguments.
     pdb = Pdb()
-    while 1:
+    pdb.rcLines.extend(commands)
+    while True:
         try:
             pdb._runscript(mainpyfile)
             if pdb._user_requested_quit:
@@ -1521,10 +1460,10 @@
             print("The program finished and will be restarted")
         except Restart:
             print("Restarting", mainpyfile, "with arguments:")
-            print("\t" + " ".join(sys.argv[1:]))
+            print("\t" + " ".join(args))
         except SystemExit:
             # In most cases SystemExit does not warrant a post-mortem session.
-            print("The program exited via sys.exit(). Exit status: ", end=' ')
+            print("The program exited via sys.exit(). Exit status:", end=' ')
             print(sys.exc_info()[1])
         except:
             traceback.print_exc()

Modified: python/branches/import_unicode/Lib/pydoc_data/topics.py
==============================================================================
--- python/branches/import_unicode/Lib/pydoc_data/topics.py	(original)
+++ python/branches/import_unicode/Lib/pydoc_data/topics.py	Sat Jul 31 13:01:07 2010
@@ -1,78 +1,77 @@
-# Autogenerated by Sphinx on Fri Jun 26 08:03:32 2009
+# Autogenerated by Sphinx on Sat Jul 31 10:22:58 2010
 topics = {'assert': '\nThe ``assert`` statement\n************************\n\nAssert statements are a convenient way to insert debugging assertions\ninto a program:\n\n   assert_stmt ::= "assert" expression ["," expression]\n\nThe simple form, ``assert expression``, is equivalent to\n\n   if __debug__:\n      if not expression: raise AssertionError\n\nThe extended form, ``assert expression1, expression2``, is equivalent\nto\n\n   if __debug__:\n      if not expression1: raise AssertionError(expression2)\n\nThese equivalences assume that ``__debug__`` and ``AssertionError``\nrefer to the built-in variables with those names.  In the current\nimplementation, the built-in variable ``__debug__`` is ``True`` under\nnormal circumstances, ``False`` when optimization is requested\n(command line option -O).  The current code generator emits no code\nfor an assert statement when optimization is requested at compile\ntime.  Note that it is unnecessary to include the source code for the\nexpression that failed in the error message; it will be displayed as\npart of the stack trace.\n\nAssignments to ``__debug__`` are illegal.  The value for the built-in\nvariable is determined when the interpreter starts.\n',
- 'assignment': '\nAssignment statements\n*********************\n\nAssignment statements are used to (re)bind names to values and to\nmodify attributes or items of mutable objects:\n\n   assignment_stmt ::= (target_list "=")+ (expression_list | yield_expression)\n   target_list     ::= target ("," target)* [","]\n   target          ::= identifier\n              | "(" target_list ")"\n              | "[" target_list "]"\n              | attributeref\n              | subscription\n              | slicing\n              | "*" target\n\n(See section *Primaries* for the syntax definitions for the last three\nsymbols.)\n\nAn assignment statement evaluates the expression list (remember that\nthis can be a single expression or a comma-separated list, the latter\nyielding a tuple) and assigns the single resulting object to each of\nthe target lists, from left to right.\n\nAssignment is defined recursively depending on the form of the target\n(list). When a target is part of a mutable object (an attribute\nreference, subscription or slicing), the mutable object must\nultimately perform the assignment and decide about its validity, and\nmay raise an exception if the assignment is unacceptable.  The rules\nobserved by various types and the exceptions raised are given with the\ndefinition of the object types (see section *The standard type\nhierarchy*).\n\nAssignment of an object to a target list, optionally enclosed in\nparentheses or square brackets, is recursively defined as follows.\n\n* If the target list is a single target: The object is assigned to\n  that target.\n\n* If the target list is a comma-separated list of targets: The object\n  must be an iterable with the same number of items as there are\n  targets in the target list, and the items are assigned, from left to\n  right, to the corresponding targets. (This rule is relaxed as of\n  Python 1.5; in earlier versions, the object had to be a tuple.\n  Since strings are sequences, an assignment like ``a, b = "xy"`` is\n  now legal as long as the string has the right length.)\n\n  * If the target list contains one target prefixed with an asterisk,\n    called a "starred" target: The object must be a sequence with at\n    least as many items as there are targets in the target list, minus\n    one.  The first items of the sequence are assigned, from left to\n    right, to the targets before the starred target.  The final items\n    of the sequence are assigned to the targets after the starred\n    target.  A list of the remaining items in the sequence is then\n    assigned to the starred target (the list can be empty).\n\n  * Else: The object must be a sequence with the same number of items\n    as there are targets in the target list, and the items are\n    assigned, from left to right, to the corresponding targets.\n\nAssignment of an object to a single target is recursively defined as\nfollows.\n\n* If the target is an identifier (name):\n\n  * If the name does not occur in a ``global`` or ``nonlocal``\n    statement in the current code block: the name is bound to the\n    object in the current local namespace.\n\n  * Otherwise: the name is bound to the object in the global namespace\n    or the outer namespace determined by ``nonlocal``, respectively.\n\n  The name is rebound if it was already bound.  This may cause the\n  reference count for the object previously bound to the name to reach\n  zero, causing the object to be deallocated and its destructor (if it\n  has one) to be called.\n\n  The name is rebound if it was already bound.  This may cause the\n  reference count for the object previously bound to the name to reach\n  zero, causing the object to be deallocated and its destructor (if it\n  has one) to be called.\n\n* If the target is a target list enclosed in parentheses or in square\n  brackets: The object must be an iterable with the same number of\n  items as there are targets in the target list, and its items are\n  assigned, from left to right, to the corresponding targets.\n\n* If the target is an attribute reference: The primary expression in\n  the reference is evaluated.  It should yield an object with\n  assignable attributes; if this is not the case, ``TypeError`` is\n  raised.  That object is then asked to assign the assigned object to\n  the given attribute; if it cannot perform the assignment, it raises\n  an exception (usually but not necessarily ``AttributeError``).\n\n* If the target is a subscription: The primary expression in the\n  reference is evaluated.  It should yield either a mutable sequence\n  object (such as a list) or a mapping object (such as a dictionary).\n  Next, the subscript expression is evaluated.\n\n  If the primary is a mutable sequence object (such as a list), the\n  subscript must yield an integer.  If it is negative, the sequence\'s\n  length is added to it.  The resulting value must be a nonnegative\n  integer less than the sequence\'s length, and the sequence is asked\n  to assign the assigned object to its item with that index.  If the\n  index is out of range, ``IndexError`` is raised (assignment to a\n  subscripted sequence cannot add new items to a list).\n\n  If the primary is a mapping object (such as a dictionary), the\n  subscript must have a type compatible with the mapping\'s key type,\n  and the mapping is then asked to create a key/datum pair which maps\n  the subscript to the assigned object.  This can either replace an\n  existing key/value pair with the same key value, or insert a new\n  key/value pair (if no key with the same value existed).\n\n  For user-defined objects, the ``__setitem__()`` method is called\n  with appropriate arguments.\n\n* If the target is a slicing: The primary expression in the reference\n  is evaluated.  It should yield a mutable sequence object (such as a\n  list).  The assigned object should be a sequence object of the same\n  type.  Next, the lower and upper bound expressions are evaluated,\n  insofar they are present; defaults are zero and the sequence\'s\n  length.  The bounds should evaluate to integers. If either bound is\n  negative, the sequence\'s length is added to it.  The resulting\n  bounds are clipped to lie between zero and the sequence\'s length,\n  inclusive.  Finally, the sequence object is asked to replace the\n  slice with the items of the assigned sequence.  The length of the\n  slice may be different from the length of the assigned sequence,\n  thus changing the length of the target sequence, if the object\n  allows it.\n\n(In the current implementation, the syntax for targets is taken to be\nthe same as for expressions, and invalid syntax is rejected during the\ncode generation phase, causing less detailed error messages.)\n\nWARNING: Although the definition of assignment implies that overlaps\nbetween the left-hand side and the right-hand side are \'safe\' (for\nexample ``a, b = b, a`` swaps two variables), overlaps *within* the\ncollection of assigned-to variables are not safe!  For instance, the\nfollowing program prints ``[0, 2]``:\n\n   x = [0, 1]\n   i = 0\n   i, x[i] = 1, 2\n   print(x)\n\nSee also:\n\n   **PEP 3132** - Extended Iterable Unpacking\n      The specification for the ``*target`` feature.\n\n\nAugmented assignment statements\n===============================\n\nAugmented assignment is the combination, in a single statement, of a\nbinary operation and an assignment statement:\n\n   augmented_assignment_stmt ::= augtarget augop (expression_list | yield_expression)\n   augtarget                 ::= identifier | attributeref | subscription | slicing\n   augop                     ::= "+=" | "-=" | "*=" | "/=" | "//=" | "%=" | "**="\n             | ">>=" | "<<=" | "&=" | "^=" | "|="\n\n(See section *Primaries* for the syntax definitions for the last three\nsymbols.)\n\nAn augmented assignment evaluates the target (which, unlike normal\nassignment statements, cannot be an unpacking) and the expression\nlist, performs the binary operation specific to the type of assignment\non the two operands, and assigns the result to the original target.\nThe target is only evaluated once.\n\nAn augmented assignment expression like ``x += 1`` can be rewritten as\n``x = x + 1`` to achieve a similar, but not exactly equal effect. In\nthe augmented version, ``x`` is only evaluated once. Also, when\npossible, the actual operation is performed *in-place*, meaning that\nrather than creating a new object and assigning that to the target,\nthe old object is modified instead.\n\nWith the exception of assigning to tuples and multiple targets in a\nsingle statement, the assignment done by augmented assignment\nstatements is handled the same way as normal assignments. Similarly,\nwith the exception of the possible *in-place* behavior, the binary\noperation performed by augmented assignment is the same as the normal\nbinary operations.\n\nFor targets which are attribute references, the initial value is\nretrieved with a ``getattr()`` and the result is assigned with a\n``setattr()``.  Notice that the two methods do not necessarily refer\nto the same variable.  When ``getattr()`` refers to a class variable,\n``setattr()`` still writes to an instance variable. For example:\n\n   class A:\n       x = 3    # class variable\n   a = A()\n   a.x += 1     # writes a.x as 4 leaving A.x as 3\n',
+ 'assignment': '\nAssignment statements\n*********************\n\nAssignment statements are used to (re)bind names to values and to\nmodify attributes or items of mutable objects:\n\n   assignment_stmt ::= (target_list "=")+ (expression_list | yield_expression)\n   target_list     ::= target ("," target)* [","]\n   target          ::= identifier\n              | "(" target_list ")"\n              | "[" target_list "]"\n              | attributeref\n              | subscription\n              | slicing\n              | "*" target\n\n(See section *Primaries* for the syntax definitions for the last three\nsymbols.)\n\nAn assignment statement evaluates the expression list (remember that\nthis can be a single expression or a comma-separated list, the latter\nyielding a tuple) and assigns the single resulting object to each of\nthe target lists, from left to right.\n\nAssignment is defined recursively depending on the form of the target\n(list). When a target is part of a mutable object (an attribute\nreference, subscription or slicing), the mutable object must\nultimately perform the assignment and decide about its validity, and\nmay raise an exception if the assignment is unacceptable.  The rules\nobserved by various types and the exceptions raised are given with the\ndefinition of the object types (see section *The standard type\nhierarchy*).\n\nAssignment of an object to a target list, optionally enclosed in\nparentheses or square brackets, is recursively defined as follows.\n\n* If the target list is a single target: The object is assigned to\n  that target.\n\n* If the target list is a comma-separated list of targets: The object\n  must be an iterable with the same number of items as there are\n  targets in the target list, and the items are assigned, from left to\n  right, to the corresponding targets. (This rule is relaxed as of\n  Python 1.5; in earlier versions, the object had to be a tuple.\n  Since strings are sequences, an assignment like ``a, b = "xy"`` is\n  now legal as long as the string has the right length.)\n\n  * If the target list contains one target prefixed with an asterisk,\n    called a "starred" target: The object must be a sequence with at\n    least as many items as there are targets in the target list, minus\n    one.  The first items of the sequence are assigned, from left to\n    right, to the targets before the starred target.  The final items\n    of the sequence are assigned to the targets after the starred\n    target.  A list of the remaining items in the sequence is then\n    assigned to the starred target (the list can be empty).\n\n  * Else: The object must be a sequence with the same number of items\n    as there are targets in the target list, and the items are\n    assigned, from left to right, to the corresponding targets.\n\nAssignment of an object to a single target is recursively defined as\nfollows.\n\n* If the target is an identifier (name):\n\n  * If the name does not occur in a ``global`` or ``nonlocal``\n    statement in the current code block: the name is bound to the\n    object in the current local namespace.\n\n  * Otherwise: the name is bound to the object in the global namespace\n    or the outer namespace determined by ``nonlocal``, respectively.\n\n  The name is rebound if it was already bound.  This may cause the\n  reference count for the object previously bound to the name to reach\n  zero, causing the object to be deallocated and its destructor (if it\n  has one) to be called.\n\n* If the target is a target list enclosed in parentheses or in square\n  brackets: The object must be an iterable with the same number of\n  items as there are targets in the target list, and its items are\n  assigned, from left to right, to the corresponding targets.\n\n* If the target is an attribute reference: The primary expression in\n  the reference is evaluated.  It should yield an object with\n  assignable attributes; if this is not the case, ``TypeError`` is\n  raised.  That object is then asked to assign the assigned object to\n  the given attribute; if it cannot perform the assignment, it raises\n  an exception (usually but not necessarily ``AttributeError``).\n\n  Note: If the object is a class instance and the attribute reference\n  occurs on both sides of the assignment operator, the RHS expression,\n  ``a.x`` can access either an instance attribute or (if no instance\n  attribute exists) a class attribute.  The LHS target ``a.x`` is\n  always set as an instance attribute, creating it if necessary.\n  Thus, the two occurrences of ``a.x`` do not necessarily refer to the\n  same attribute: if the RHS expression refers to a class attribute,\n  the LHS creates a new instance attribute as the target of the\n  assignment:\n\n     class Cls:\n         x = 3             # class variable\n     inst = Cls()\n     inst.x = inst.x + 1   # writes inst.x as 4 leaving Cls.x as 3\n\n  This description does not necessarily apply to descriptor\n  attributes, such as properties created with ``property()``.\n\n* If the target is a subscription: The primary expression in the\n  reference is evaluated.  It should yield either a mutable sequence\n  object (such as a list) or a mapping object (such as a dictionary).\n  Next, the subscript expression is evaluated.\n\n  If the primary is a mutable sequence object (such as a list), the\n  subscript must yield an integer.  If it is negative, the sequence\'s\n  length is added to it.  The resulting value must be a nonnegative\n  integer less than the sequence\'s length, and the sequence is asked\n  to assign the assigned object to its item with that index.  If the\n  index is out of range, ``IndexError`` is raised (assignment to a\n  subscripted sequence cannot add new items to a list).\n\n  If the primary is a mapping object (such as a dictionary), the\n  subscript must have a type compatible with the mapping\'s key type,\n  and the mapping is then asked to create a key/datum pair which maps\n  the subscript to the assigned object.  This can either replace an\n  existing key/value pair with the same key value, or insert a new\n  key/value pair (if no key with the same value existed).\n\n  For user-defined objects, the ``__setitem__()`` method is called\n  with appropriate arguments.\n\n* If the target is a slicing: The primary expression in the reference\n  is evaluated.  It should yield a mutable sequence object (such as a\n  list).  The assigned object should be a sequence object of the same\n  type.  Next, the lower and upper bound expressions are evaluated,\n  insofar they are present; defaults are zero and the sequence\'s\n  length.  The bounds should evaluate to integers. If either bound is\n  negative, the sequence\'s length is added to it.  The resulting\n  bounds are clipped to lie between zero and the sequence\'s length,\n  inclusive.  Finally, the sequence object is asked to replace the\n  slice with the items of the assigned sequence.  The length of the\n  slice may be different from the length of the assigned sequence,\n  thus changing the length of the target sequence, if the object\n  allows it.\n\n**CPython implementation detail:** In the current implementation, the\nsyntax for targets is taken to be the same as for expressions, and\ninvalid syntax is rejected during the code generation phase, causing\nless detailed error messages.\n\nWARNING: Although the definition of assignment implies that overlaps\nbetween the left-hand side and the right-hand side are \'safe\' (for\nexample ``a, b = b, a`` swaps two variables), overlaps *within* the\ncollection of assigned-to variables are not safe!  For instance, the\nfollowing program prints ``[0, 2]``:\n\n   x = [0, 1]\n   i = 0\n   i, x[i] = 1, 2\n   print(x)\n\nSee also:\n\n   **PEP 3132** - Extended Iterable Unpacking\n      The specification for the ``*target`` feature.\n\n\nAugmented assignment statements\n===============================\n\nAugmented assignment is the combination, in a single statement, of a\nbinary operation and an assignment statement:\n\n   augmented_assignment_stmt ::= augtarget augop (expression_list | yield_expression)\n   augtarget                 ::= identifier | attributeref | subscription | slicing\n   augop                     ::= "+=" | "-=" | "*=" | "/=" | "//=" | "%=" | "**="\n             | ">>=" | "<<=" | "&=" | "^=" | "|="\n\n(See section *Primaries* for the syntax definitions for the last three\nsymbols.)\n\nAn augmented assignment evaluates the target (which, unlike normal\nassignment statements, cannot be an unpacking) and the expression\nlist, performs the binary operation specific to the type of assignment\non the two operands, and assigns the result to the original target.\nThe target is only evaluated once.\n\nAn augmented assignment expression like ``x += 1`` can be rewritten as\n``x = x + 1`` to achieve a similar, but not exactly equal effect. In\nthe augmented version, ``x`` is only evaluated once. Also, when\npossible, the actual operation is performed *in-place*, meaning that\nrather than creating a new object and assigning that to the target,\nthe old object is modified instead.\n\nWith the exception of assigning to tuples and multiple targets in a\nsingle statement, the assignment done by augmented assignment\nstatements is handled the same way as normal assignments. Similarly,\nwith the exception of the possible *in-place* behavior, the binary\noperation performed by augmented assignment is the same as the normal\nbinary operations.\n\nFor targets which are attribute references, the same *caveat about\nclass and instance attributes* applies as for regular assignments.\n',
  'atom-identifiers': '\nIdentifiers (Names)\n*******************\n\nAn identifier occurring as an atom is a name.  See section\n*Identifiers and keywords* for lexical definition and section *Naming\nand binding* for documentation of naming and binding.\n\nWhen the name is bound to an object, evaluation of the atom yields\nthat object. When a name is not bound, an attempt to evaluate it\nraises a ``NameError`` exception.\n\n**Private name mangling:** When an identifier that textually occurs in\na class definition begins with two or more underscore characters and\ndoes not end in two or more underscores, it is considered a *private\nname* of that class. Private names are transformed to a longer form\nbefore code is generated for them.  The transformation inserts the\nclass name in front of the name, with leading underscores removed, and\na single underscore inserted in front of the class name.  For example,\nthe identifier ``__spam`` occurring in a class named ``Ham`` will be\ntransformed to ``_Ham__spam``.  This transformation is independent of\nthe syntactical context in which the identifier is used.  If the\ntransformed name is extremely long (longer than 255 characters),\nimplementation defined truncation may happen.  If the class name\nconsists only of underscores, no transformation is done.\n',
  'atom-literals': "\nLiterals\n********\n\nPython supports string and bytes literals and various numeric\nliterals:\n\n   literal ::= stringliteral | bytesliteral\n               | integer | floatnumber | imagnumber\n\nEvaluation of a literal yields an object of the given type (string,\nbytes, integer, floating point number, complex number) with the given\nvalue.  The value may be approximated in the case of floating point\nand imaginary (complex) literals.  See section *Literals* for details.\n\nWith the exception of bytes literals, these all correspond to\nimmutable data types, and hence the object's identity is less\nimportant than its value. Multiple evaluations of literals with the\nsame value (either the same occurrence in the program text or a\ndifferent occurrence) may obtain the same object or a different object\nwith the same value.\n",
- 'attribute-access': '\nCustomizing attribute access\n****************************\n\nThe following methods can be defined to customize the meaning of\nattribute access (use of, assignment to, or deletion of ``x.name``)\nfor class instances.\n\nobject.__getattr__(self, name)\n\n   Called when an attribute lookup has not found the attribute in the\n   usual places (i.e. it is not an instance attribute nor is it found\n   in the class tree for ``self``).  ``name`` is the attribute name.\n   This method should return the (computed) attribute value or raise\n   an ``AttributeError`` exception.\n\n   Note that if the attribute is found through the normal mechanism,\n   ``__getattr__()`` is not called.  (This is an intentional asymmetry\n   between ``__getattr__()`` and ``__setattr__()``.) This is done both\n   for efficiency reasons and because otherwise ``__getattr__()``\n   would have no way to access other attributes of the instance.  Note\n   that at least for instance variables, you can fake total control by\n   not inserting any values in the instance attribute dictionary (but\n   instead inserting them in another object).  See the\n   ``__getattribute__()`` method below for a way to actually get total\n   control over attribute access.\n\nobject.__getattribute__(self, name)\n\n   Called unconditionally to implement attribute accesses for\n   instances of the class. If the class also defines\n   ``__getattr__()``, the latter will not be called unless\n   ``__getattribute__()`` either calls it explicitly or raises an\n   ``AttributeError``. This method should return the (computed)\n   attribute value or raise an ``AttributeError`` exception. In order\n   to avoid infinite recursion in this method, its implementation\n   should always call the base class method with the same name to\n   access any attributes it needs, for example,\n   ``object.__getattribute__(self, name)``.\n\n   Note: This method may still be bypassed when looking up special methods\n     as the result of implicit invocation via language syntax or\n     builtin functions. See *Special method lookup*.\n\nobject.__setattr__(self, name, value)\n\n   Called when an attribute assignment is attempted.  This is called\n   instead of the normal mechanism (i.e. store the value in the\n   instance dictionary). *name* is the attribute name, *value* is the\n   value to be assigned to it.\n\n   If ``__setattr__()`` wants to assign to an instance attribute, it\n   should call the base class method with the same name, for example,\n   ``object.__setattr__(self, name, value)``.\n\nobject.__delattr__(self, name)\n\n   Like ``__setattr__()`` but for attribute deletion instead of\n   assignment.  This should only be implemented if ``del obj.name`` is\n   meaningful for the object.\n\nobject.__dir__(self)\n\n   Called when ``dir()`` is called on the object.  A list must be\n   returned.\n\n\nImplementing Descriptors\n========================\n\nThe following methods only apply when an instance of the class\ncontaining the method (a so-called *descriptor* class) appears in the\nclass dictionary of another class, known as the *owner* class.  In the\nexamples below, "the attribute" refers to the attribute whose name is\nthe key of the property in the owner class\' ``__dict__``.\n\nobject.__get__(self, instance, owner)\n\n   Called to get the attribute of the owner class (class attribute\n   access) or of an instance of that class (instance attribute\n   access). *owner* is always the owner class, while *instance* is the\n   instance that the attribute was accessed through, or ``None`` when\n   the attribute is accessed through the *owner*.  This method should\n   return the (computed) attribute value or raise an\n   ``AttributeError`` exception.\n\nobject.__set__(self, instance, value)\n\n   Called to set the attribute on an instance *instance* of the owner\n   class to a new value, *value*.\n\nobject.__delete__(self, instance)\n\n   Called to delete the attribute on an instance *instance* of the\n   owner class.\n\n\nInvoking Descriptors\n====================\n\nIn general, a descriptor is an object attribute with "binding\nbehavior", one whose attribute access has been overridden by methods\nin the descriptor protocol:  ``__get__()``, ``__set__()``, and\n``__delete__()``. If any of those methods are defined for an object,\nit is said to be a descriptor.\n\nThe default behavior for attribute access is to get, set, or delete\nthe attribute from an object\'s dictionary. For instance, ``a.x`` has a\nlookup chain starting with ``a.__dict__[\'x\']``, then\n``type(a).__dict__[\'x\']``, and continuing through the base classes of\n``type(a)`` excluding metaclasses.\n\nHowever, if the looked-up value is an object defining one of the\ndescriptor methods, then Python may override the default behavior and\ninvoke the descriptor method instead.  Where this occurs in the\nprecedence chain depends on which descriptor methods were defined and\nhow they were called.\n\nThe starting point for descriptor invocation is a binding, ``a.x``.\nHow the arguments are assembled depends on ``a``:\n\nDirect Call\n   The simplest and least common call is when user code directly\n   invokes a descriptor method:    ``x.__get__(a)``.\n\nInstance Binding\n   If binding to an object instance, ``a.x`` is transformed into the\n   call: ``type(a).__dict__[\'x\'].__get__(a, type(a))``.\n\nClass Binding\n   If binding to a class, ``A.x`` is transformed into the call:\n   ``A.__dict__[\'x\'].__get__(None, A)``.\n\nSuper Binding\n   If ``a`` is an instance of ``super``, then the binding ``super(B,\n   obj).m()`` searches ``obj.__class__.__mro__`` for the base class\n   ``A`` immediately preceding ``B`` and then invokes the descriptor\n   with the call: ``A.__dict__[\'m\'].__get__(obj, A)``.\n\nFor instance bindings, the precedence of descriptor invocation depends\non the which descriptor methods are defined.  Normally, data\ndescriptors define both ``__get__()`` and ``__set__()``, while non-\ndata descriptors have just the ``__get__()`` method.  Data descriptors\nalways override a redefinition in an instance dictionary.  In\ncontrast, non-data descriptors can be overridden by instances. [2]\n\nPython methods (including ``staticmethod()`` and ``classmethod()``)\nare implemented as non-data descriptors.  Accordingly, instances can\nredefine and override methods.  This allows individual instances to\nacquire behaviors that differ from other instances of the same class.\n\nThe ``property()`` function is implemented as a data descriptor.\nAccordingly, instances cannot override the behavior of a property.\n\n\n__slots__\n=========\n\nBy default, instances of classes have a dictionary for attribute\nstorage.  This wastes space for objects having very few instance\nvariables.  The space consumption can become acute when creating large\nnumbers of instances.\n\nThe default can be overridden by defining *__slots__* in a class\ndefinition. The *__slots__* declaration takes a sequence of instance\nvariables and reserves just enough space in each instance to hold a\nvalue for each variable.  Space is saved because *__dict__* is not\ncreated for each instance.\n\nobject.__slots__\n\n   This class variable can be assigned a string, iterable, or sequence\n   of strings with variable names used by instances.  If defined in a\n   class, *__slots__* reserves space for the declared variables and\n   prevents the automatic creation of *__dict__* and *__weakref__* for\n   each instance.\n\n\nNotes on using *__slots__*\n--------------------------\n\n* When inheriting from a class without *__slots__*, the *__dict__*\n  attribute of that class will always be accessible, so a *__slots__*\n  definition in the subclass is meaningless.\n\n* Without a *__dict__* variable, instances cannot be assigned new\n  variables not listed in the *__slots__* definition.  Attempts to\n  assign to an unlisted variable name raises ``AttributeError``. If\n  dynamic assignment of new variables is desired, then add\n  ``\'__dict__\'`` to the sequence of strings in the *__slots__*\n  declaration.\n\n* Without a *__weakref__* variable for each instance, classes defining\n  *__slots__* do not support weak references to its instances. If weak\n  reference support is needed, then add ``\'__weakref__\'`` to the\n  sequence of strings in the *__slots__* declaration.\n\n* *__slots__* are implemented at the class level by creating\n  descriptors (*Implementing Descriptors*) for each variable name.  As\n  a result, class attributes cannot be used to set default values for\n  instance variables defined by *__slots__*; otherwise, the class\n  attribute would overwrite the descriptor assignment.\n\n* If a class defines a slot also defined in a base class, the instance\n  variable defined by the base class slot is inaccessible (except by\n  retrieving its descriptor directly from the base class). This\n  renders the meaning of the program undefined.  In the future, a\n  check may be added to prevent this.\n\n* The action of a *__slots__* declaration is limited to the class\n  where it is defined.  As a result, subclasses will have a *__dict__*\n  unless they also define *__slots__*.\n\n* Nonempty *__slots__* does not work for classes derived from\n  "variable-length" built-in types such as ``int``, ``str`` and\n  ``tuple``.\n\n* Any non-string iterable may be assigned to *__slots__*. Mappings may\n  also be used; however, in the future, special meaning may be\n  assigned to the values corresponding to each key.\n\n* *__class__* assignment works only if both classes have the same\n  *__slots__*.\n',
+ 'attribute-access': '\nCustomizing attribute access\n****************************\n\nThe following methods can be defined to customize the meaning of\nattribute access (use of, assignment to, or deletion of ``x.name``)\nfor class instances.\n\nobject.__getattr__(self, name)\n\n   Called when an attribute lookup has not found the attribute in the\n   usual places (i.e. it is not an instance attribute nor is it found\n   in the class tree for ``self``).  ``name`` is the attribute name.\n   This method should return the (computed) attribute value or raise\n   an ``AttributeError`` exception.\n\n   Note that if the attribute is found through the normal mechanism,\n   ``__getattr__()`` is not called.  (This is an intentional asymmetry\n   between ``__getattr__()`` and ``__setattr__()``.) This is done both\n   for efficiency reasons and because otherwise ``__getattr__()``\n   would have no way to access other attributes of the instance.  Note\n   that at least for instance variables, you can fake total control by\n   not inserting any values in the instance attribute dictionary (but\n   instead inserting them in another object).  See the\n   ``__getattribute__()`` method below for a way to actually get total\n   control over attribute access.\n\nobject.__getattribute__(self, name)\n\n   Called unconditionally to implement attribute accesses for\n   instances of the class. If the class also defines\n   ``__getattr__()``, the latter will not be called unless\n   ``__getattribute__()`` either calls it explicitly or raises an\n   ``AttributeError``. This method should return the (computed)\n   attribute value or raise an ``AttributeError`` exception. In order\n   to avoid infinite recursion in this method, its implementation\n   should always call the base class method with the same name to\n   access any attributes it needs, for example,\n   ``object.__getattribute__(self, name)``.\n\n   Note: This method may still be bypassed when looking up special methods\n     as the result of implicit invocation via language syntax or\n     built-in functions. See *Special method lookup*.\n\nobject.__setattr__(self, name, value)\n\n   Called when an attribute assignment is attempted.  This is called\n   instead of the normal mechanism (i.e. store the value in the\n   instance dictionary). *name* is the attribute name, *value* is the\n   value to be assigned to it.\n\n   If ``__setattr__()`` wants to assign to an instance attribute, it\n   should call the base class method with the same name, for example,\n   ``object.__setattr__(self, name, value)``.\n\nobject.__delattr__(self, name)\n\n   Like ``__setattr__()`` but for attribute deletion instead of\n   assignment.  This should only be implemented if ``del obj.name`` is\n   meaningful for the object.\n\nobject.__dir__(self)\n\n   Called when ``dir()`` is called on the object.  A list must be\n   returned.\n\n\nImplementing Descriptors\n========================\n\nThe following methods only apply when an instance of the class\ncontaining the method (a so-called *descriptor* class) appears in the\nclass dictionary of another class, known as the *owner* class.  In the\nexamples below, "the attribute" refers to the attribute whose name is\nthe key of the property in the owner class\' ``__dict__``.\n\nobject.__get__(self, instance, owner)\n\n   Called to get the attribute of the owner class (class attribute\n   access) or of an instance of that class (instance attribute\n   access). *owner* is always the owner class, while *instance* is the\n   instance that the attribute was accessed through, or ``None`` when\n   the attribute is accessed through the *owner*.  This method should\n   return the (computed) attribute value or raise an\n   ``AttributeError`` exception.\n\nobject.__set__(self, instance, value)\n\n   Called to set the attribute on an instance *instance* of the owner\n   class to a new value, *value*.\n\nobject.__delete__(self, instance)\n\n   Called to delete the attribute on an instance *instance* of the\n   owner class.\n\n\nInvoking Descriptors\n====================\n\nIn general, a descriptor is an object attribute with "binding\nbehavior", one whose attribute access has been overridden by methods\nin the descriptor protocol:  ``__get__()``, ``__set__()``, and\n``__delete__()``. If any of those methods are defined for an object,\nit is said to be a descriptor.\n\nThe default behavior for attribute access is to get, set, or delete\nthe attribute from an object\'s dictionary. For instance, ``a.x`` has a\nlookup chain starting with ``a.__dict__[\'x\']``, then\n``type(a).__dict__[\'x\']``, and continuing through the base classes of\n``type(a)`` excluding metaclasses.\n\nHowever, if the looked-up value is an object defining one of the\ndescriptor methods, then Python may override the default behavior and\ninvoke the descriptor method instead.  Where this occurs in the\nprecedence chain depends on which descriptor methods were defined and\nhow they were called.\n\nThe starting point for descriptor invocation is a binding, ``a.x``.\nHow the arguments are assembled depends on ``a``:\n\nDirect Call\n   The simplest and least common call is when user code directly\n   invokes a descriptor method:    ``x.__get__(a)``.\n\nInstance Binding\n   If binding to an object instance, ``a.x`` is transformed into the\n   call: ``type(a).__dict__[\'x\'].__get__(a, type(a))``.\n\nClass Binding\n   If binding to a class, ``A.x`` is transformed into the call:\n   ``A.__dict__[\'x\'].__get__(None, A)``.\n\nSuper Binding\n   If ``a`` is an instance of ``super``, then the binding ``super(B,\n   obj).m()`` searches ``obj.__class__.__mro__`` for the base class\n   ``A`` immediately preceding ``B`` and then invokes the descriptor\n   with the call: ``A.__dict__[\'m\'].__get__(obj, A)``.\n\nFor instance bindings, the precedence of descriptor invocation depends\non the which descriptor methods are defined.  A descriptor can define\nany combination of ``__get__()``, ``__set__()`` and ``__delete__()``.\nIf it does not define ``__get__()``, then accessing the attribute will\nreturn the descriptor object itself unless there is a value in the\nobject\'s instance dictionary.  If the descriptor defines ``__set__()``\nand/or ``__delete__()``, it is a data descriptor; if it defines\nneither, it is a non-data descriptor.  Normally, data descriptors\ndefine both ``__get__()`` and ``__set__()``, while non-data\ndescriptors have just the ``__get__()`` method.  Data descriptors with\n``__set__()`` and ``__get__()`` defined always override a redefinition\nin an instance dictionary.  In contrast, non-data descriptors can be\noverridden by instances.\n\nPython methods (including ``staticmethod()`` and ``classmethod()``)\nare implemented as non-data descriptors.  Accordingly, instances can\nredefine and override methods.  This allows individual instances to\nacquire behaviors that differ from other instances of the same class.\n\nThe ``property()`` function is implemented as a data descriptor.\nAccordingly, instances cannot override the behavior of a property.\n\n\n__slots__\n=========\n\nBy default, instances of classes have a dictionary for attribute\nstorage.  This wastes space for objects having very few instance\nvariables.  The space consumption can become acute when creating large\nnumbers of instances.\n\nThe default can be overridden by defining *__slots__* in a class\ndefinition. The *__slots__* declaration takes a sequence of instance\nvariables and reserves just enough space in each instance to hold a\nvalue for each variable.  Space is saved because *__dict__* is not\ncreated for each instance.\n\nobject.__slots__\n\n   This class variable can be assigned a string, iterable, or sequence\n   of strings with variable names used by instances.  If defined in a\n   class, *__slots__* reserves space for the declared variables and\n   prevents the automatic creation of *__dict__* and *__weakref__* for\n   each instance.\n\n\nNotes on using *__slots__*\n--------------------------\n\n* When inheriting from a class without *__slots__*, the *__dict__*\n  attribute of that class will always be accessible, so a *__slots__*\n  definition in the subclass is meaningless.\n\n* Without a *__dict__* variable, instances cannot be assigned new\n  variables not listed in the *__slots__* definition.  Attempts to\n  assign to an unlisted variable name raises ``AttributeError``. If\n  dynamic assignment of new variables is desired, then add\n  ``\'__dict__\'`` to the sequence of strings in the *__slots__*\n  declaration.\n\n* Without a *__weakref__* variable for each instance, classes defining\n  *__slots__* do not support weak references to its instances. If weak\n  reference support is needed, then add ``\'__weakref__\'`` to the\n  sequence of strings in the *__slots__* declaration.\n\n* *__slots__* are implemented at the class level by creating\n  descriptors (*Implementing Descriptors*) for each variable name.  As\n  a result, class attributes cannot be used to set default values for\n  instance variables defined by *__slots__*; otherwise, the class\n  attribute would overwrite the descriptor assignment.\n\n* The action of a *__slots__* declaration is limited to the class\n  where it is defined.  As a result, subclasses will have a *__dict__*\n  unless they also define *__slots__* (which must only contain names\n  of any *additional* slots).\n\n* If a class defines a slot also defined in a base class, the instance\n  variable defined by the base class slot is inaccessible (except by\n  retrieving its descriptor directly from the base class). This\n  renders the meaning of the program undefined.  In the future, a\n  check may be added to prevent this.\n\n* Nonempty *__slots__* does not work for classes derived from\n  "variable-length" built-in types such as ``int``, ``str`` and\n  ``tuple``.\n\n* Any non-string iterable may be assigned to *__slots__*. Mappings may\n  also be used; however, in the future, special meaning may be\n  assigned to the values corresponding to each key.\n\n* *__class__* assignment works only if both classes have the same\n  *__slots__*.\n',
  'attribute-references': '\nAttribute references\n********************\n\nAn attribute reference is a primary followed by a period and a name:\n\n   attributeref ::= primary "." identifier\n\nThe primary must evaluate to an object of a type that supports\nattribute references, which most objects do.  This object is then\nasked to produce the attribute whose name is the identifier (which can\nbe customized by overriding the ``__getattr__()`` method).  If this\nattribute is not available, the exception ``AttributeError`` is\nraised.  Otherwise, the type and value of the object produced is\ndetermined by the object.  Multiple evaluations of the same attribute\nreference may yield different objects.\n',
- 'augassign': '\nAugmented assignment statements\n*******************************\n\nAugmented assignment is the combination, in a single statement, of a\nbinary operation and an assignment statement:\n\n   augmented_assignment_stmt ::= augtarget augop (expression_list | yield_expression)\n   augtarget                 ::= identifier | attributeref | subscription | slicing\n   augop                     ::= "+=" | "-=" | "*=" | "/=" | "//=" | "%=" | "**="\n             | ">>=" | "<<=" | "&=" | "^=" | "|="\n\n(See section *Primaries* for the syntax definitions for the last three\nsymbols.)\n\nAn augmented assignment evaluates the target (which, unlike normal\nassignment statements, cannot be an unpacking) and the expression\nlist, performs the binary operation specific to the type of assignment\non the two operands, and assigns the result to the original target.\nThe target is only evaluated once.\n\nAn augmented assignment expression like ``x += 1`` can be rewritten as\n``x = x + 1`` to achieve a similar, but not exactly equal effect. In\nthe augmented version, ``x`` is only evaluated once. Also, when\npossible, the actual operation is performed *in-place*, meaning that\nrather than creating a new object and assigning that to the target,\nthe old object is modified instead.\n\nWith the exception of assigning to tuples and multiple targets in a\nsingle statement, the assignment done by augmented assignment\nstatements is handled the same way as normal assignments. Similarly,\nwith the exception of the possible *in-place* behavior, the binary\noperation performed by augmented assignment is the same as the normal\nbinary operations.\n\nFor targets which are attribute references, the initial value is\nretrieved with a ``getattr()`` and the result is assigned with a\n``setattr()``.  Notice that the two methods do not necessarily refer\nto the same variable.  When ``getattr()`` refers to a class variable,\n``setattr()`` still writes to an instance variable. For example:\n\n   class A:\n       x = 3    # class variable\n   a = A()\n   a.x += 1     # writes a.x as 4 leaving A.x as 3\n',
+ 'augassign': '\nAugmented assignment statements\n*******************************\n\nAugmented assignment is the combination, in a single statement, of a\nbinary operation and an assignment statement:\n\n   augmented_assignment_stmt ::= augtarget augop (expression_list | yield_expression)\n   augtarget                 ::= identifier | attributeref | subscription | slicing\n   augop                     ::= "+=" | "-=" | "*=" | "/=" | "//=" | "%=" | "**="\n             | ">>=" | "<<=" | "&=" | "^=" | "|="\n\n(See section *Primaries* for the syntax definitions for the last three\nsymbols.)\n\nAn augmented assignment evaluates the target (which, unlike normal\nassignment statements, cannot be an unpacking) and the expression\nlist, performs the binary operation specific to the type of assignment\non the two operands, and assigns the result to the original target.\nThe target is only evaluated once.\n\nAn augmented assignment expression like ``x += 1`` can be rewritten as\n``x = x + 1`` to achieve a similar, but not exactly equal effect. In\nthe augmented version, ``x`` is only evaluated once. Also, when\npossible, the actual operation is performed *in-place*, meaning that\nrather than creating a new object and assigning that to the target,\nthe old object is modified instead.\n\nWith the exception of assigning to tuples and multiple targets in a\nsingle statement, the assignment done by augmented assignment\nstatements is handled the same way as normal assignments. Similarly,\nwith the exception of the possible *in-place* behavior, the binary\noperation performed by augmented assignment is the same as the normal\nbinary operations.\n\nFor targets which are attribute references, the same *caveat about\nclass and instance attributes* applies as for regular assignments.\n',
  'binary': '\nBinary arithmetic operations\n****************************\n\nThe binary arithmetic operations have the conventional priority\nlevels.  Note that some of these operations also apply to certain non-\nnumeric types.  Apart from the power operator, there are only two\nlevels, one for multiplicative operators and one for additive\noperators:\n\n   m_expr ::= u_expr | m_expr "*" u_expr | m_expr "//" u_expr | m_expr "/" u_expr\n              | m_expr "%" u_expr\n   a_expr ::= m_expr | a_expr "+" m_expr | a_expr "-" m_expr\n\nThe ``*`` (multiplication) operator yields the product of its\narguments.  The arguments must either both be numbers, or one argument\nmust be an integer and the other must be a sequence. In the former\ncase, the numbers are converted to a common type and then multiplied\ntogether.  In the latter case, sequence repetition is performed; a\nnegative repetition factor yields an empty sequence.\n\nThe ``/`` (division) and ``//`` (floor division) operators yield the\nquotient of their arguments.  The numeric arguments are first\nconverted to a common type. Integer division yields a float, while\nfloor division of integers results in an integer; the result is that\nof mathematical division with the \'floor\' function applied to the\nresult.  Division by zero raises the ``ZeroDivisionError`` exception.\n\nThe ``%`` (modulo) operator yields the remainder from the division of\nthe first argument by the second.  The numeric arguments are first\nconverted to a common type.  A zero right argument raises the\n``ZeroDivisionError`` exception.  The arguments may be floating point\nnumbers, e.g., ``3.14%0.7`` equals ``0.34`` (since ``3.14`` equals\n``4*0.7 + 0.34``.)  The modulo operator always yields a result with\nthe same sign as its second operand (or zero); the absolute value of\nthe result is strictly smaller than the absolute value of the second\noperand [1].\n\nThe floor division and modulo operators are connected by the following\nidentity: ``x == (x//y)*y + (x%y)``.  Floor division and modulo are\nalso connected with the built-in function ``divmod()``: ``divmod(x, y)\n== (x//y, x%y)``. [2].\n\nIn addition to performing the modulo operation on numbers, the ``%``\noperator is also overloaded by string objects to perform old-style\nstring formatting (also known as interpolation).  The syntax for\nstring formatting is described in the Python Library Reference,\nsection *Old String Formatting Operations*.\n\nThe floor division operator, the modulo operator, and the ``divmod()``\nfunction are not defined for complex numbers.  Instead, convert to a\nfloating point number using the ``abs()`` function if appropriate.\n\nThe ``+`` (addition) operator yields the sum of its arguments.  The\narguments must either both be numbers or both sequences of the same\ntype.  In the former case, the numbers are converted to a common type\nand then added together.  In the latter case, the sequences are\nconcatenated.\n\nThe ``-`` (subtraction) operator yields the difference of its\narguments.  The numeric arguments are first converted to a common\ntype.\n',
  'bitwise': '\nBinary bitwise operations\n*************************\n\nEach of the three bitwise operations has a different priority level:\n\n   and_expr ::= shift_expr | and_expr "&" shift_expr\n   xor_expr ::= and_expr | xor_expr "^" and_expr\n   or_expr  ::= xor_expr | or_expr "|" xor_expr\n\nThe ``&`` operator yields the bitwise AND of its arguments, which must\nbe integers.\n\nThe ``^`` operator yields the bitwise XOR (exclusive OR) of its\narguments, which must be integers.\n\nThe ``|`` operator yields the bitwise (inclusive) OR of its arguments,\nwhich must be integers.\n',
  'bltin-code-objects': '\nCode Objects\n************\n\nCode objects are used by the implementation to represent "pseudo-\ncompiled" executable Python code such as a function body. They differ\nfrom function objects because they don\'t contain a reference to their\nglobal execution environment.  Code objects are returned by the built-\nin ``compile()`` function and can be extracted from function objects\nthrough their ``__code__`` attribute. See also the ``code`` module.\n\nA code object can be executed or evaluated by passing it (instead of a\nsource string) to the ``exec()`` or ``eval()``  built-in functions.\n\nSee *The standard type hierarchy* for more information.\n',
  'bltin-ellipsis-object': '\nThe Ellipsis Object\n*******************\n\nThis object is commonly used by slicing (see *Slicings*).  It supports\nno special operations.  There is exactly one ellipsis object, named\n``Ellipsis`` (a built-in name).\n\nIt is written as ``Ellipsis`` or ``...``.\n',
- 'bltin-file-objects': '\nFile Objects\n************\n\nFile objects are implemented using C\'s ``stdio`` package and can be\ncreated with the built-in ``open()`` function.  File objects are also\nreturned by some other built-in functions and methods, such as\n``os.popen()`` and ``os.fdopen()`` and the ``makefile()`` method of\nsocket objects. Temporary files can be created using the ``tempfile``\nmodule, and high-level file operations such as copying, moving, and\ndeleting files and directories can be achieved with the ``shutil``\nmodule.\n\nWhen a file operation fails for an I/O-related reason, the exception\n``IOError`` is raised.  This includes situations where the operation\nis not defined for some reason, like ``seek()`` on a tty device or\nwriting a file opened for reading.\n\nFiles have the following methods:\n\nfile.close()\n\n   Close the file.  A closed file cannot be read or written any more.\n   Any operation which requires that the file be open will raise a\n   ``ValueError`` after the file has been closed.  Calling ``close()``\n   more than once is allowed.\n\n   You can avoid having to call this method explicitly if you use the\n   ``with`` statement.  For example, the following code will\n   automatically close *f* when the ``with`` block is exited:\n\n      from __future__ import with_statement # This isn\'t required in Python 2.6\n\n      with open("hello.txt") as f:\n          for line in f:\n              print(line)\n\n   In older versions of Python, you would have needed to do this to\n   get the same effect:\n\n      f = open("hello.txt")\n      try:\n          for line in f:\n              print(line)\n      finally:\n          f.close()\n\n   Note: Not all "file-like" types in Python support use as a context\n     manager for the ``with`` statement.  If your code is intended to\n     work with any file-like object, you can use the function\n     ``contextlib.closing()`` instead of using the object directly.\n\nfile.flush()\n\n   Flush the internal buffer, like ``stdio``\'s ``fflush()``.  This may\n   be a no-op on some file-like objects.\n\n   Note: ``flush()`` does not necessarily write the file\'s data to disk.\n     Use ``flush()`` followed by ``os.fsync()`` to ensure this\n     behavior.\n\nfile.fileno()\n\n   Return the integer "file descriptor" that is used by the underlying\n   implementation to request I/O operations from the operating system.\n   This can be useful for other, lower level interfaces that use file\n   descriptors, such as the ``fcntl`` module or ``os.read()`` and\n   friends.\n\n   Note: File-like objects which do not have a real file descriptor should\n     *not* provide this method!\n\nfile.isatty()\n\n   Return ``True`` if the file is connected to a tty(-like) device,\n   else ``False``.\n\n   Note: If a file-like object is not associated with a real file, this\n     method should *not* be implemented.\n\nfile.__next__()\n\n   A file object is its own iterator, for example ``iter(f)`` returns\n   *f* (unless *f* is closed).  When a file is used as an iterator,\n   typically in a ``for`` loop (for example, ``for line in f:\n   print(line)``), the ``__next__()`` method is called repeatedly.\n   This method returns the next input line, or raises\n   ``StopIteration`` when EOF is hit when the file is open for reading\n   (behavior is undefined when the file is open for writing).  In\n   order to make a ``for`` loop the most efficient way of looping over\n   the lines of a file (a very common operation), the ``__next__()``\n   method uses a hidden read-ahead buffer.  As a consequence of using\n   a read-ahead buffer, combining ``__next__()`` with other file\n   methods (like ``readline()``) does not work right.  However, using\n   ``seek()`` to reposition the file to an absolute position will\n   flush the read-ahead buffer.\n\nfile.read([size])\n\n   Read at most *size* bytes from the file (less if the read hits EOF\n   before obtaining *size* bytes).  If the *size* argument is negative\n   or omitted, read all data until EOF is reached.  The bytes are\n   returned as a string object.  An empty string is returned when EOF\n   is encountered immediately.  (For certain files, like ttys, it\n   makes sense to continue reading after an EOF is hit.)  Note that\n   this method may call the underlying C function ``fread()`` more\n   than once in an effort to acquire as close to *size* bytes as\n   possible. Also note that when in non-blocking mode, less data than\n   was requested may be returned, even if no *size* parameter was\n   given.\n\nfile.readline([size])\n\n   Read one entire line from the file.  A trailing newline character\n   is kept in the string (but may be absent when a file ends with an\n   incomplete line). [5]  If the *size* argument is present and non-\n   negative, it is a maximum byte count (including the trailing\n   newline) and an incomplete line may be returned. An empty string is\n   returned *only* when EOF is encountered immediately.\n\n   Note: Unlike ``stdio``\'s ``fgets()``, the returned string contains null\n     characters (``\'\\0\'``) if they occurred in the input.\n\nfile.readlines([sizehint])\n\n   Read until EOF using ``readline()`` and return a list containing\n   the lines thus read.  If the optional *sizehint* argument is\n   present, instead of reading up to EOF, whole lines totalling\n   approximately *sizehint* bytes (possibly after rounding up to an\n   internal buffer size) are read.  Objects implementing a file-like\n   interface may choose to ignore *sizehint* if it cannot be\n   implemented, or cannot be implemented efficiently.\n\nfile.seek(offset[, whence])\n\n   Set the file\'s current position, like ``stdio``\'s ``fseek()``. The\n   *whence* argument is optional and defaults to  ``os.SEEK_SET`` or\n   ``0`` (absolute file positioning); other values are ``os.SEEK_CUR``\n   or ``1`` (seek relative to the current position) and\n   ``os.SEEK_END`` or ``2``  (seek relative to the file\'s end).  There\n   is no return value.\n\n   For example, ``f.seek(2, os.SEEK_CUR)`` advances the position by\n   two and ``f.seek(-3, os.SEEK_END)`` sets the position to the third\n   to last.\n\n   Note that if the file is opened for appending (mode ``\'a\'`` or\n   ``\'a+\'``), any ``seek()`` operations will be undone at the next\n   write.  If the file is only opened for writing in append mode (mode\n   ``\'a\'``), this method is essentially a no-op, but it remains useful\n   for files opened in append mode with reading enabled (mode\n   ``\'a+\'``).  If the file is opened in text mode (without ``\'b\'``),\n   only offsets returned by ``tell()`` are legal.  Use of other\n   offsets causes undefined behavior.\n\n   Note that not all file objects are seekable.\n\nfile.tell()\n\n   Return the file\'s current position, like ``stdio``\'s ``ftell()``.\n\n   Note: On Windows, ``tell()`` can return illegal values (after an\n     ``fgets()``) when reading files with Unix-style line-endings. Use\n     binary mode (``\'rb\'``) to circumvent this problem.\n\nfile.truncate([size])\n\n   Truncate the file\'s size.  If the optional *size* argument is\n   present, the file is truncated to (at most) that size.  The size\n   defaults to the current position. The current file position is not\n   changed.  Note that if a specified size exceeds the file\'s current\n   size, the result is platform-dependent:  possibilities include that\n   the file may remain unchanged, increase to the specified size as if\n   zero-filled, or increase to the specified size with undefined new\n   content. Availability:  Windows, many Unix variants.\n\nfile.write(str)\n\n   Write a string to the file.  Due to buffering, the string may not\n   actually show up in the file until the ``flush()`` or ``close()``\n   method is called.\n\n   The meaning of the return value is not defined for every file-like\n   object. Some (mostly low-level) file-like objects may return the\n   number of bytes actually written, others return ``None``.\n\nfile.writelines(sequence)\n\n   Write a sequence of strings to the file.  The sequence can be any\n   iterable object producing strings, typically a list of strings.\n   There is no return value. (The name is intended to match\n   ``readlines()``; ``writelines()`` does not add line separators.)\n\nFiles support the iterator protocol.  Each iteration returns the same\nresult as ``file.readline()``, and iteration ends when the\n``readline()`` method returns an empty string.\n\nFile objects also offer a number of other interesting attributes.\nThese are not required for file-like objects, but should be\nimplemented if they make sense for the particular object.\n\nfile.closed\n\n   bool indicating the current state of the file object.  This is a\n   read-only attribute; the ``close()`` method changes the value. It\n   may not be available on all file-like objects.\n\nfile.encoding\n\n   The encoding that this file uses. When strings are written to a\n   file, they will be converted to byte strings using this encoding.\n   In addition, when the file is connected to a terminal, the\n   attribute gives the encoding that the terminal is likely to use\n   (that  information might be incorrect if the user has misconfigured\n   the  terminal). The attribute is read-only and may not be present\n   on all file-like objects. It may also be ``None``, in which case\n   the file uses the system default encoding for converting strings.\n\nfile.errors\n\n   The Unicode error handler used along with the encoding.\n\nfile.mode\n\n   The I/O mode for the file.  If the file was created using the\n   ``open()`` built-in function, this will be the value of the *mode*\n   parameter.  This is a read-only attribute and may not be present on\n   all file-like objects.\n\nfile.name\n\n   If the file object was created using ``open()``, the name of the\n   file. Otherwise, some string that indicates the source of the file\n   object, of the form ``<...>``.  This is a read-only attribute and\n   may not be present on all file-like objects.\n\nfile.newlines\n\n   If Python was built with the *--with-universal-newlines* option to\n   **configure** (the default) this read-only attribute exists, and\n   for files opened in universal newline read mode it keeps track of\n   the types of newlines encountered while reading the file. The\n   values it can take are ``\'\\r\'``, ``\'\\n\'``, ``\'\\r\\n\'``, ``None``\n   (unknown, no newlines read yet) or a tuple containing all the\n   newline types seen, to indicate that multiple newline conventions\n   were encountered. For files not opened in universal newline read\n   mode the value of this attribute will be ``None``.\n',
  'bltin-null-object': "\nThe Null Object\n***************\n\nThis object is returned by functions that don't explicitly return a\nvalue.  It supports no special operations.  There is exactly one null\nobject, named ``None`` (a built-in name).\n\nIt is written as ``None``.\n",
  'bltin-type-objects': "\nType Objects\n************\n\nType objects represent the various object types.  An object's type is\naccessed by the built-in function ``type()``.  There are no special\noperations on types.  The standard module ``types`` defines names for\nall standard built-in types.\n\nTypes are written like this: ``<class 'int'>``.\n",
- 'booleans': '\nBoolean operations\n******************\n\nBoolean operations have the lowest priority of all Python operations:\n\n   expression             ::= conditional_expression | lambda_form\n   expression_nocond      ::= or_test | lambda_form_nocond\n   conditional_expression ::= or_test ["if" or_test "else" expression]\n   or_test                ::= and_test | or_test "or" and_test\n   and_test               ::= not_test | and_test "and" not_test\n   not_test               ::= comparison | "not" not_test\n\nIn the context of Boolean operations, and also when expressions are\nused by control flow statements, the following values are interpreted\nas false: ``False``, ``None``, numeric zero of all types, and empty\nstrings and containers (including strings, tuples, lists,\ndictionaries, sets and frozensets).  All other values are interpreted\nas true.  User-defined objects can customize their truth value by\nproviding a ``__bool__()`` method.\n\nThe operator ``not`` yields ``True`` if its argument is false,\n``False`` otherwise.\n\nThe expression ``x if C else y`` first evaluates *C* (*not* *x*); if\n*C* is true, *x* is evaluated and its value is returned; otherwise,\n*y* is evaluated and its value is returned.\n\nThe expression ``x and y`` first evaluates *x*; if *x* is false, its\nvalue is returned; otherwise, *y* is evaluated and the resulting value\nis returned.\n\nThe expression ``x or y`` first evaluates *x*; if *x* is true, its\nvalue is returned; otherwise, *y* is evaluated and the resulting value\nis returned.\n\n(Note that neither ``and`` nor ``or`` restrict the value and type they\nreturn to ``False`` and ``True``, but rather return the last evaluated\nargument.  This is sometimes useful, e.g., if ``s`` is a string that\nshould be replaced by a default value if it is empty, the expression\n``s or \'foo\'`` yields the desired value.  Because ``not`` has to\ninvent a value anyway, it does not bother to return a value of the\nsame type as its argument, so e.g., ``not \'foo\'`` yields ``False``,\nnot ``\'\'``.)\n',
+ 'booleans': '\nBoolean operations\n******************\n\n   or_test  ::= and_test | or_test "or" and_test\n   and_test ::= not_test | and_test "and" not_test\n   not_test ::= comparison | "not" not_test\n\nIn the context of Boolean operations, and also when expressions are\nused by control flow statements, the following values are interpreted\nas false: ``False``, ``None``, numeric zero of all types, and empty\nstrings and containers (including strings, tuples, lists,\ndictionaries, sets and frozensets).  All other values are interpreted\nas true.  User-defined objects can customize their truth value by\nproviding a ``__bool__()`` method.\n\nThe operator ``not`` yields ``True`` if its argument is false,\n``False`` otherwise.\n\nThe expression ``x and y`` first evaluates *x*; if *x* is false, its\nvalue is returned; otherwise, *y* is evaluated and the resulting value\nis returned.\n\nThe expression ``x or y`` first evaluates *x*; if *x* is true, its\nvalue is returned; otherwise, *y* is evaluated and the resulting value\nis returned.\n\n(Note that neither ``and`` nor ``or`` restrict the value and type they\nreturn to ``False`` and ``True``, but rather return the last evaluated\nargument.  This is sometimes useful, e.g., if ``s`` is a string that\nshould be replaced by a default value if it is empty, the expression\n``s or \'foo\'`` yields the desired value.  Because ``not`` has to\ninvent a value anyway, it does not bother to return a value of the\nsame type as its argument, so e.g., ``not \'foo\'`` yields ``False``,\nnot ``\'\'``.)\n',
  'break': '\nThe ``break`` statement\n***********************\n\n   break_stmt ::= "break"\n\n``break`` may only occur syntactically nested in a ``for`` or\n``while`` loop, but not nested in a function or class definition\nwithin that loop.\n\nIt terminates the nearest enclosing loop, skipping the optional\n``else`` clause if the loop has one.\n\nIf a ``for`` loop is terminated by ``break``, the loop control target\nkeeps its current value.\n\nWhen ``break`` passes control out of a ``try`` statement with a\n``finally`` clause, that ``finally`` clause is executed before really\nleaving the loop.\n',
  'callable-types': '\nEmulating callable objects\n**************************\n\nobject.__call__(self[, args...])\n\n   Called when the instance is "called" as a function; if this method\n   is defined, ``x(arg1, arg2, ...)`` is a shorthand for\n   ``x.__call__(arg1, arg2, ...)``.\n',
- 'calls': '\nCalls\n*****\n\nA call calls a callable object (e.g., a function) with a possibly\nempty series of arguments:\n\n   call                 ::= primary "(" [argument_list [","] | comprehension] ")"\n   argument_list        ::= positional_arguments ["," keyword_arguments]\n                       ["," "*" expression] ["," keyword_arguments]\n                       ["," "**" expression]\n                     | keyword_arguments ["," "*" expression]\n                       ["," keyword_arguments] ["," "**" expression]\n                     | "*" expression ["," keyword_arguments] ["," "**" expression]\n                     | "**" expression\n   positional_arguments ::= expression ("," expression)*\n   keyword_arguments    ::= keyword_item ("," keyword_item)*\n   keyword_item         ::= identifier "=" expression\n\nA trailing comma may be present after the positional and keyword\narguments but does not affect the semantics.\n\nThe primary must evaluate to a callable object (user-defined\nfunctions, built-in functions, methods of built-in objects, class\nobjects, methods of class instances, and all objects having a\n``__call__()`` method are callable).  All argument expressions are\nevaluated before the call is attempted.  Please refer to section\n*Function definitions* for the syntax of formal parameter lists.\n\nIf keyword arguments are present, they are first converted to\npositional arguments, as follows.  First, a list of unfilled slots is\ncreated for the formal parameters.  If there are N positional\narguments, they are placed in the first N slots.  Next, for each\nkeyword argument, the identifier is used to determine the\ncorresponding slot (if the identifier is the same as the first formal\nparameter name, the first slot is used, and so on).  If the slot is\nalready filled, a ``TypeError`` exception is raised. Otherwise, the\nvalue of the argument is placed in the slot, filling it (even if the\nexpression is ``None``, it fills the slot).  When all arguments have\nbeen processed, the slots that are still unfilled are filled with the\ncorresponding default value from the function definition.  (Default\nvalues are calculated, once, when the function is defined; thus, a\nmutable object such as a list or dictionary used as default value will\nbe shared by all calls that don\'t specify an argument value for the\ncorresponding slot; this should usually be avoided.)  If there are any\nunfilled slots for which no default value is specified, a\n``TypeError`` exception is raised.  Otherwise, the list of filled\nslots is used as the argument list for the call.\n\nNote: An implementation may provide builtin functions whose positional\n  parameters do not have names, even if they are \'named\' for the\n  purpose of documentation, and which therefore cannot be supplied by\n  keyword.  In CPython, this is the case for functions implemented in\n  C that use ``PyArg_ParseTuple()`` to parse their arguments.\n\nIf there are more positional arguments than there are formal parameter\nslots, a ``TypeError`` exception is raised, unless a formal parameter\nusing the syntax ``*identifier`` is present; in this case, that formal\nparameter receives a tuple containing the excess positional arguments\n(or an empty tuple if there were no excess positional arguments).\n\nIf any keyword argument does not correspond to a formal parameter\nname, a ``TypeError`` exception is raised, unless a formal parameter\nusing the syntax ``**identifier`` is present; in this case, that\nformal parameter receives a dictionary containing the excess keyword\narguments (using the keywords as keys and the argument values as\ncorresponding values), or a (new) empty dictionary if there were no\nexcess keyword arguments.\n\nIf the syntax ``*expression`` appears in the function call,\n``expression`` must evaluate to a sequence.  Elements from this\nsequence are treated as if they were additional positional arguments;\nif there are positional arguments *x1*,..., *xN*, and ``expression``\nevaluates to a sequence *y1*, ..., *yM*, this is equivalent to a call\nwith M+N positional arguments *x1*, ..., *xN*, *y1*, ..., *yM*.\n\nA consequence of this is that although the ``*expression`` syntax may\nappear *after* some keyword arguments, it is processed *before* the\nkeyword arguments (and the ``**expression`` argument, if any -- see\nbelow).  So:\n\n   >>> def f(a, b):\n   ...  print(a, b)\n   ...\n   >>> f(b=1, *(2,))\n   2 1\n   >>> f(a=1, *(2,))\n   Traceback (most recent call last):\n     File "<stdin>", line 1, in ?\n   TypeError: f() got multiple values for keyword argument \'a\'\n   >>> f(1, *(2,))\n   1 2\n\nIt is unusual for both keyword arguments and the ``*expression``\nsyntax to be used in the same call, so in practice this confusion does\nnot arise.\n\nIf the syntax ``**expression`` appears in the function call,\n``expression`` must evaluate to a mapping, the contents of which are\ntreated as additional keyword arguments.  In the case of a keyword\nappearing in both ``expression`` and as an explicit keyword argument,\na ``TypeError`` exception is raised.\n\nFormal parameters using the syntax ``*identifier`` or ``**identifier``\ncannot be used as positional argument slots or as keyword argument\nnames.\n\nA call always returns some value, possibly ``None``, unless it raises\nan exception.  How this value is computed depends on the type of the\ncallable object.\n\nIf it is---\n\na user-defined function:\n   The code block for the function is executed, passing it the\n   argument list.  The first thing the code block will do is bind the\n   formal parameters to the arguments; this is described in section\n   *Function definitions*.  When the code block executes a ``return``\n   statement, this specifies the return value of the function call.\n\na built-in function or method:\n   The result is up to the interpreter; see *Built-in Functions* for\n   the descriptions of built-in functions and methods.\n\na class object:\n   A new instance of that class is returned.\n\na class instance method:\n   The corresponding user-defined function is called, with an argument\n   list that is one longer than the argument list of the call: the\n   instance becomes the first argument.\n\na class instance:\n   The class must define a ``__call__()`` method; the effect is then\n   the same as if that method was called.\n',
+ 'calls': '\nCalls\n*****\n\nA call calls a callable object (e.g., a function) with a possibly\nempty series of arguments:\n\n   call                 ::= primary "(" [argument_list [","] | comprehension] ")"\n   argument_list        ::= positional_arguments ["," keyword_arguments]\n                       ["," "*" expression] ["," keyword_arguments]\n                       ["," "**" expression]\n                     | keyword_arguments ["," "*" expression]\n                       ["," keyword_arguments] ["," "**" expression]\n                     | "*" expression ["," keyword_arguments] ["," "**" expression]\n                     | "**" expression\n   positional_arguments ::= expression ("," expression)*\n   keyword_arguments    ::= keyword_item ("," keyword_item)*\n   keyword_item         ::= identifier "=" expression\n\nA trailing comma may be present after the positional and keyword\narguments but does not affect the semantics.\n\nThe primary must evaluate to a callable object (user-defined\nfunctions, built-in functions, methods of built-in objects, class\nobjects, methods of class instances, and all objects having a\n``__call__()`` method are callable).  All argument expressions are\nevaluated before the call is attempted.  Please refer to section\n*Function definitions* for the syntax of formal parameter lists.\n\nIf keyword arguments are present, they are first converted to\npositional arguments, as follows.  First, a list of unfilled slots is\ncreated for the formal parameters.  If there are N positional\narguments, they are placed in the first N slots.  Next, for each\nkeyword argument, the identifier is used to determine the\ncorresponding slot (if the identifier is the same as the first formal\nparameter name, the first slot is used, and so on).  If the slot is\nalready filled, a ``TypeError`` exception is raised. Otherwise, the\nvalue of the argument is placed in the slot, filling it (even if the\nexpression is ``None``, it fills the slot).  When all arguments have\nbeen processed, the slots that are still unfilled are filled with the\ncorresponding default value from the function definition.  (Default\nvalues are calculated, once, when the function is defined; thus, a\nmutable object such as a list or dictionary used as default value will\nbe shared by all calls that don\'t specify an argument value for the\ncorresponding slot; this should usually be avoided.)  If there are any\nunfilled slots for which no default value is specified, a\n``TypeError`` exception is raised.  Otherwise, the list of filled\nslots is used as the argument list for the call.\n\n**CPython implementation detail:** An implementation may provide\nbuilt-in functions whose positional parameters do not have names, even\nif they are \'named\' for the purpose of documentation, and which\ntherefore cannot be supplied by keyword.  In CPython, this is the case\nfor functions implemented in C that use ``PyArg_ParseTuple()`` to\nparse their arguments.\n\nIf there are more positional arguments than there are formal parameter\nslots, a ``TypeError`` exception is raised, unless a formal parameter\nusing the syntax ``*identifier`` is present; in this case, that formal\nparameter receives a tuple containing the excess positional arguments\n(or an empty tuple if there were no excess positional arguments).\n\nIf any keyword argument does not correspond to a formal parameter\nname, a ``TypeError`` exception is raised, unless a formal parameter\nusing the syntax ``**identifier`` is present; in this case, that\nformal parameter receives a dictionary containing the excess keyword\narguments (using the keywords as keys and the argument values as\ncorresponding values), or a (new) empty dictionary if there were no\nexcess keyword arguments.\n\nIf the syntax ``*expression`` appears in the function call,\n``expression`` must evaluate to a sequence.  Elements from this\nsequence are treated as if they were additional positional arguments;\nif there are positional arguments *x1*,..., *xN*, and ``expression``\nevaluates to a sequence *y1*, ..., *yM*, this is equivalent to a call\nwith M+N positional arguments *x1*, ..., *xN*, *y1*, ..., *yM*.\n\nA consequence of this is that although the ``*expression`` syntax may\nappear *after* some keyword arguments, it is processed *before* the\nkeyword arguments (and the ``**expression`` argument, if any -- see\nbelow).  So:\n\n   >>> def f(a, b):\n   ...  print(a, b)\n   ...\n   >>> f(b=1, *(2,))\n   2 1\n   >>> f(a=1, *(2,))\n   Traceback (most recent call last):\n     File "<stdin>", line 1, in ?\n   TypeError: f() got multiple values for keyword argument \'a\'\n   >>> f(1, *(2,))\n   1 2\n\nIt is unusual for both keyword arguments and the ``*expression``\nsyntax to be used in the same call, so in practice this confusion does\nnot arise.\n\nIf the syntax ``**expression`` appears in the function call,\n``expression`` must evaluate to a mapping, the contents of which are\ntreated as additional keyword arguments.  In the case of a keyword\nappearing in both ``expression`` and as an explicit keyword argument,\na ``TypeError`` exception is raised.\n\nFormal parameters using the syntax ``*identifier`` or ``**identifier``\ncannot be used as positional argument slots or as keyword argument\nnames.\n\nA call always returns some value, possibly ``None``, unless it raises\nan exception.  How this value is computed depends on the type of the\ncallable object.\n\nIf it is---\n\na user-defined function:\n   The code block for the function is executed, passing it the\n   argument list.  The first thing the code block will do is bind the\n   formal parameters to the arguments; this is described in section\n   *Function definitions*.  When the code block executes a ``return``\n   statement, this specifies the return value of the function call.\n\na built-in function or method:\n   The result is up to the interpreter; see *Built-in Functions* for\n   the descriptions of built-in functions and methods.\n\na class object:\n   A new instance of that class is returned.\n\na class instance method:\n   The corresponding user-defined function is called, with an argument\n   list that is one longer than the argument list of the call: the\n   instance becomes the first argument.\n\na class instance:\n   The class must define a ``__call__()`` method; the effect is then\n   the same as if that method was called.\n',
  'class': '\nClass definitions\n*****************\n\nA class definition defines a class object (see section *The standard\ntype hierarchy*):\n\n   classdef    ::= [decorators] "class" classname [inheritance] ":" suite\n   inheritance ::= "(" [expression_list] ")"\n   classname   ::= identifier\n\nA class definition is an executable statement.  It first evaluates the\ninheritance list, if present.  Each item in the inheritance list\nshould evaluate to a class object or class type which allows\nsubclassing.  The class\'s suite is then executed in a new execution\nframe (see section *Naming and binding*), using a newly created local\nnamespace and the original global namespace. (Usually, the suite\ncontains only function definitions.)  When the class\'s suite finishes\nexecution, its execution frame is discarded but its local namespace is\nsaved. [4] A class object is then created using the inheritance list\nfor the base classes and the saved local namespace for the attribute\ndictionary.  The class name is bound to this class object in the\noriginal local namespace.\n\nClasses can also be decorated; as with functions,\n\n   @f1(arg)\n   @f2\n   class Foo: pass\n\nis equivalent to\n\n   class Foo: pass\n   Foo = f1(arg)(f2(Foo))\n\n**Programmer\'s note:** Variables defined in the class definition are\nclass variables; they are shared by instances. Instance variables can\nbe set in a method with ``self.name = value``.  Both class and\ninstance variables are accessible through the notation\n"``self.name``", and an instance variable hides a class variable with\nthe same name when accessed in this way.  Class variables can be used\nas defaults for instance variables, but using mutable values there can\nlead to unexpected results.  Descriptors can be used to create\ninstance variables with different implementation details.\n\nSee also:\n\n   **PEP 3129** - Class Decorators\n\nClass definitions, like function definitions, may be wrapped by one or\nmore *decorator* expressions.  The evaluation rules for the decorator\nexpressions are the same as for functions.  The result must be a class\nobject, which is then bound to the class name.\n\n-[ Footnotes ]-\n\n[1] The exception is propagated to the invocation stack only if there\n    is no ``finally`` clause that negates the exception.\n\n[2] Currently, control "flows off the end" except in the case of an\n    exception or the execution of a ``return``, ``continue``, or\n    ``break`` statement.\n\n[3] A string literal appearing as the first statement in the function\n    body is transformed into the function\'s ``__doc__`` attribute and\n    therefore the function\'s *docstring*.\n\n[4] A string literal appearing as the first statement in the class\n    body is transformed into the namespace\'s ``__doc__`` item and\n    therefore the class\'s *docstring*.\n',
- 'comparisons': '\nComparisons\n***********\n\nUnlike C, all comparison operations in Python have the same priority,\nwhich is lower than that of any arithmetic, shifting or bitwise\noperation.  Also unlike C, expressions like ``a < b < c`` have the\ninterpretation that is conventional in mathematics:\n\n   comparison    ::= or_expr ( comp_operator or_expr )*\n   comp_operator ::= "<" | ">" | "==" | ">=" | "<=" | "!="\n                     | "is" ["not"] | ["not"] "in"\n\nComparisons yield boolean values: ``True`` or ``False``.\n\nComparisons can be chained arbitrarily, e.g., ``x < y <= z`` is\nequivalent to ``x < y and y <= z``, except that ``y`` is evaluated\nonly once (but in both cases ``z`` is not evaluated at all when ``x <\ny`` is found to be false).\n\nFormally, if *a*, *b*, *c*, ..., *y*, *z* are expressions and *op1*,\n*op2*, ..., *opN* are comparison operators, then ``a op1 b op2 c ... y\nopN z`` is equivalent to ``a op1 b and b op2 c and ... y opN z``,\nexcept that each expression is evaluated at most once.\n\nNote that ``a op1 b op2 c`` doesn\'t imply any kind of comparison\nbetween *a* and *c*, so that, e.g., ``x < y > z`` is perfectly legal\n(though perhaps not pretty).\n\nThe operators ``<``, ``>``, ``==``, ``>=``, ``<=``, and ``!=`` compare\nthe values of two objects.  The objects need not have the same type.\nIf both are numbers, they are converted to a common type.  Otherwise,\nthe ``==`` and ``!=`` operators *always* consider objects of different\ntypes to be unequal, while the ``<``, ``>``, ``>=`` and ``<=``\noperators raise a ``TypeError`` when comparing objects of different\ntypes that do not implement these operators for the given pair of\ntypes.  You can control comparison behavior of objects of non-builtin\ntypes by defining rich comparison methods like ``__gt__()``, described\nin section *Basic customization*.\n\nComparison of objects of the same type depends on the type:\n\n* Numbers are compared arithmetically.\n\n* The values ``float(\'NaN\')`` and ``Decimal(\'NaN\')`` are special. The\n  are identical to themselves, ``x is x`` but are not equal to\n  themselves, ``x != x``.  Additionally, comparing any value to a\n  not-a-number value will return ``False``.  For example, both ``3 <\n  float(\'NaN\')`` and ``float(\'NaN\') < 3`` will return ``False``.\n\n* Bytes objects are compared lexicographically using the numeric\n  values of their elements.\n\n* Strings are compared lexicographically using the numeric equivalents\n  (the result of the built-in function ``ord()``) of their characters.\n  [3] String and bytes object can\'t be compared!\n\n* Tuples and lists are compared lexicographically using comparison of\n  corresponding elements.  This means that to compare equal, each\n  element must compare equal and the two sequences must be of the same\n  type and have the same length.\n\n  If not equal, the sequences are ordered the same as their first\n  differing elements.  For example, ``[1,2,x] <= [1,2,y]`` has the\n  same value as ``x <= y``.  If the corresponding element does not\n  exist, the shorter sequence is ordered first (for example, ``[1,2] <\n  [1,2,3]``).\n\n* Mappings (dictionaries) compare equal if and only if their sorted\n  ``(key, value)`` lists compare equal. [4] Outcomes other than\n  equality are resolved consistently, but are not otherwise defined.\n  [5]\n\n* Sets and frozensets define comparison operators to mean subset and\n  superset tests.  Those relations do not define total orderings (the\n  two sets ``{1,2}`` and {2,3} are not equal, nor subsets of one\n  another, nor supersets of one another).  Accordingly, sets are not\n  appropriate arguments for functions which depend on total ordering.\n  For example, ``min()``, ``max()``, and ``sorted()`` produce\n  undefined results given a list of sets as inputs.\n\n* Most other objects of builtin types compare unequal unless they are\n  the same object; the choice whether one object is considered smaller\n  or larger than another one is made arbitrarily but consistently\n  within one execution of a program.\n\nComparison of objects of the differing types depends on whether either\nof the types provide explicit support for the comparison.  Most\nnumeric types can be compared with one another, but comparisons of\n``float`` and ``Decimal`` are not supported to avoid the inevitable\nconfusion arising from representation issues such as ``float(\'1.1\')``\nbeing inexactly represented and therefore not exactly equal to\n``Decimal(\'1.1\')`` which is.  When cross-type comparison is not\nsupported, the comparison method returns ``NotImplemented``.  This can\ncreate the illusion of non-transitivity between supported cross-type\ncomparisons and unsupported comparisons.  For example, ``Decimal(2) ==\n2`` and *2 == float(2)`* but ``Decimal(2) != float(2)``.\n\nThe operators ``in`` and ``not in`` test for membership.  ``x in s``\nevaluates to true if *x* is a member of *s*, and false otherwise.  ``x\nnot in s`` returns the negation of ``x in s``.  All built-in sequences\nand set types support this as well as dictionary, for which ``in``\ntests whether a the dictionary has a given key. For container types\nsuch as list, tuple, set, frozenset, dict, or collections.deque, the\nexpression ``x in y`` is equivalent to ``any(x is e or x == e for val\ne in y)``.\n\nFor the string and bytes types, ``x in y`` is true if and only if *x*\nis a substring of *y*.  An equivalent test is ``y.find(x) != -1``.\nEmpty strings are always considered to be a substring of any other\nstring, so ``"" in "abc"`` will return ``True``.\n\nFor user-defined classes which define the ``__contains__()`` method,\n``x in y`` is true if and only if ``y.__contains__(x)`` is true.\n\nFor user-defined classes which do not define ``__contains__()`` and do\ndefine ``__getitem__()``, ``x in y`` is true if and only if there is a\nnon-negative integer index *i* such that ``x == y[i]``, and all lower\ninteger indices do not raise ``IndexError`` exception.  (If any other\nexception is raised, it is as if ``in`` raised that exception).\n\nThe operator ``not in`` is defined to have the inverse true value of\n``in``.\n\nThe operators ``is`` and ``is not`` test for object identity: ``x is\ny`` is true if and only if *x* and *y* are the same object.  ``x is\nnot y`` yields the inverse truth value. [6]\n',
- 'compound': '\nCompound statements\n*******************\n\nCompound statements contain (groups of) other statements; they affect\nor control the execution of those other statements in some way.  In\ngeneral, compound statements span multiple lines, although in simple\nincarnations a whole compound statement may be contained in one line.\n\nThe ``if``, ``while`` and ``for`` statements implement traditional\ncontrol flow constructs.  ``try`` specifies exception handlers and/or\ncleanup code for a group of statements, while the ``with`` statement\nallows the execution of initialization and finalization code around a\nblock of code.  Function and class definitions are also syntactically\ncompound statements.\n\nCompound statements consist of one or more \'clauses.\'  A clause\nconsists of a header and a \'suite.\'  The clause headers of a\nparticular compound statement are all at the same indentation level.\nEach clause header begins with a uniquely identifying keyword and ends\nwith a colon.  A suite is a group of statements controlled by a\nclause.  A suite can be one or more semicolon-separated simple\nstatements on the same line as the header, following the header\'s\ncolon, or it can be one or more indented statements on subsequent\nlines.  Only the latter form of suite can contain nested compound\nstatements; the following is illegal, mostly because it wouldn\'t be\nclear to which ``if`` clause a following ``else`` clause would belong:\n\n   if test1: if test2: print(x)\n\nAlso note that the semicolon binds tighter than the colon in this\ncontext, so that in the following example, either all or none of the\n``print()`` calls are executed:\n\n   if x < y < z: print(x); print(y); print(z)\n\nSummarizing:\n\n   compound_stmt ::= if_stmt\n                     | while_stmt\n                     | for_stmt\n                     | try_stmt\n                     | with_stmt\n                     | funcdef\n                     | classdef\n   suite         ::= stmt_list NEWLINE | NEWLINE INDENT statement+ DEDENT\n   statement     ::= stmt_list NEWLINE | compound_stmt\n   stmt_list     ::= simple_stmt (";" simple_stmt)* [";"]\n\nNote that statements always end in a ``NEWLINE`` possibly followed by\na ``DEDENT``.  Also note that optional continuation clauses always\nbegin with a keyword that cannot start a statement, thus there are no\nambiguities (the \'dangling ``else``\' problem is solved in Python by\nrequiring nested ``if`` statements to be indented).\n\nThe formatting of the grammar rules in the following sections places\neach clause on a separate line for clarity.\n\n\nThe ``if`` statement\n====================\n\nThe ``if`` statement is used for conditional execution:\n\n   if_stmt ::= "if" expression ":" suite\n               ( "elif" expression ":" suite )*\n               ["else" ":" suite]\n\nIt selects exactly one of the suites by evaluating the expressions one\nby one until one is found to be true (see section *Boolean operations*\nfor the definition of true and false); then that suite is executed\n(and no other part of the ``if`` statement is executed or evaluated).\nIf all expressions are false, the suite of the ``else`` clause, if\npresent, is executed.\n\n\nThe ``while`` statement\n=======================\n\nThe ``while`` statement is used for repeated execution as long as an\nexpression is true:\n\n   while_stmt ::= "while" expression ":" suite\n                  ["else" ":" suite]\n\nThis repeatedly tests the expression and, if it is true, executes the\nfirst suite; if the expression is false (which may be the first time\nit is tested) the suite of the ``else`` clause, if present, is\nexecuted and the loop terminates.\n\nA ``break`` statement executed in the first suite terminates the loop\nwithout executing the ``else`` clause\'s suite.  A ``continue``\nstatement executed in the first suite skips the rest of the suite and\ngoes back to testing the expression.\n\n\nThe ``for`` statement\n=====================\n\nThe ``for`` statement is used to iterate over the elements of a\nsequence (such as a string, tuple or list) or other iterable object:\n\n   for_stmt ::= "for" target_list "in" expression_list ":" suite\n                ["else" ":" suite]\n\nThe expression list is evaluated once; it should yield an iterable\nobject.  An iterator is created for the result of the\n``expression_list``.  The suite is then executed once for each item\nprovided by the iterator, in the order of ascending indices.  Each\nitem in turn is assigned to the target list using the standard rules\nfor assignments (see *Assignment statements*), and then the suite is\nexecuted.  When the items are exhausted (which is immediately when the\nsequence is empty or an iterator raises a ``StopIteration``\nexception), the suite in the ``else`` clause, if present, is executed,\nand the loop terminates.\n\nA ``break`` statement executed in the first suite terminates the loop\nwithout executing the ``else`` clause\'s suite.  A ``continue``\nstatement executed in the first suite skips the rest of the suite and\ncontinues with the next item, or with the ``else`` clause if there was\nno next item.\n\nThe suite may assign to the variable(s) in the target list; this does\nnot affect the next item assigned to it.\n\nNames in the target list are not deleted when the loop is finished,\nbut if the sequence is empty, it will not have been assigned to at all\nby the loop.  Hint: the built-in function ``range()`` returns an\niterator of integers suitable to emulate the effect of Pascal\'s ``for\ni := a to b do``; e.g., ``list(range(3))`` returns the list ``[0, 1,\n2]``.\n\nNote: There is a subtlety when the sequence is being modified by the loop\n  (this can only occur for mutable sequences, i.e. lists).  An\n  internal counter is used to keep track of which item is used next,\n  and this is incremented on each iteration.  When this counter has\n  reached the length of the sequence the loop terminates.  This means\n  that if the suite deletes the current (or a previous) item from the\n  sequence, the next item will be skipped (since it gets the index of\n  the current item which has already been treated).  Likewise, if the\n  suite inserts an item in the sequence before the current item, the\n  current item will be treated again the next time through the loop.\n  This can lead to nasty bugs that can be avoided by making a\n  temporary copy using a slice of the whole sequence, e.g.,\n\n     for x in a[:]:\n         if x < 0: a.remove(x)\n\n\nThe ``try`` statement\n=====================\n\nThe ``try`` statement specifies exception handlers and/or cleanup code\nfor a group of statements:\n\n   try_stmt  ::= try1_stmt | try2_stmt\n   try1_stmt ::= "try" ":" suite\n                 ("except" [expression ["as" target]] ":" suite)+\n                 ["else" ":" suite]\n                 ["finally" ":" suite]\n   try2_stmt ::= "try" ":" suite\n                 "finally" ":" suite\n\nThe ``except`` clause(s) specify one or more exception handlers. When\nno exception occurs in the ``try`` clause, no exception handler is\nexecuted. When an exception occurs in the ``try`` suite, a search for\nan exception handler is started.  This search inspects the except\nclauses in turn until one is found that matches the exception.  An\nexpression-less except clause, if present, must be last; it matches\nany exception.  For an except clause with an expression, that\nexpression is evaluated, and the clause matches the exception if the\nresulting object is "compatible" with the exception.  An object is\ncompatible with an exception if it is the class or a base class of the\nexception object or a tuple containing an item compatible with the\nexception.\n\nIf no except clause matches the exception, the search for an exception\nhandler continues in the surrounding code and on the invocation stack.\n[1]\n\nIf the evaluation of an expression in the header of an except clause\nraises an exception, the original search for a handler is canceled and\na search starts for the new exception in the surrounding code and on\nthe call stack (it is treated as if the entire ``try`` statement\nraised the exception).\n\nWhen a matching except clause is found, the exception is assigned to\nthe target specified after the ``as`` keyword in that except clause,\nif present, and the except clause\'s suite is executed.  All except\nclauses must have an executable block.  When the end of this block is\nreached, execution continues normally after the entire try statement.\n(This means that if two nested handlers exist for the same exception,\nand the exception occurs in the try clause of the inner handler, the\nouter handler will not handle the exception.)\n\nWhen an exception has been assigned using ``as target``, it is cleared\nat the end of the except clause.  This is as if\n\n   except E as N:\n       foo\n\nwas translated to\n\n   except E as N:\n       try:\n           foo\n       finally:\n           N = None\n           del N\n\nThat means that you have to assign the exception to a different name\nif you want to be able to refer to it after the except clause.  The\nreason for this is that with the traceback attached to them,\nexceptions will form a reference cycle with the stack frame, keeping\nall locals in that frame alive until the next garbage collection\noccurs.\n\nBefore an except clause\'s suite is executed, details about the\nexception are stored in the ``sys`` module and can be access via\n``sys.exc_info()``. ``sys.exc_info()`` returns a 3-tuple consisting\nof: ``exc_type``, the exception class; ``exc_value``, the exception\ninstance; ``exc_traceback``, a traceback object (see section *The\nstandard type hierarchy*) identifying the point in the program where\nthe exception occurred. ``sys.exc_info()`` values are restored to\ntheir previous values (before the call) when returning from a function\nthat handled an exception.\n\nThe optional ``else`` clause is executed if and when control flows off\nthe end of the ``try`` clause. [2] Exceptions in the ``else`` clause\nare not handled by the preceding ``except`` clauses.\n\nIf ``finally`` is present, it specifies a \'cleanup\' handler.  The\n``try`` clause is executed, including any ``except`` and ``else``\nclauses.  If an exception occurs in any of the clauses and is not\nhandled, the exception is temporarily saved. The ``finally`` clause is\nexecuted.  If there is a saved exception, it is re-raised at the end\nof the ``finally`` clause. If the ``finally`` clause raises another\nexception or executes a ``return`` or ``break`` statement, the saved\nexception is lost.  The exception information is not available to the\nprogram during execution of the ``finally`` clause.\n\nWhen a ``return``, ``break`` or ``continue`` statement is executed in\nthe ``try`` suite of a ``try``...``finally`` statement, the\n``finally`` clause is also executed \'on the way out.\' A ``continue``\nstatement is illegal in the ``finally`` clause. (The reason is a\nproblem with the current implementation --- this restriction may be\nlifted in the future).\n\nAdditional information on exceptions can be found in section\n*Exceptions*, and information on using the ``raise`` statement to\ngenerate exceptions may be found in section *The raise statement*.\n\n\nThe ``with`` statement\n======================\n\nThe ``with`` statement is used to wrap the execution of a block with\nmethods defined by a context manager (see section *With Statement\nContext Managers*). This allows common\n``try``...``except``...``finally`` usage patterns to be encapsulated\nfor convenient reuse.\n\n   with_stmt ::= "with" with_item ("," with_item)* ":" suite\n   with_item ::= expression ["as" target]\n\nThe execution of the ``with`` statement with one "item" proceeds as\nfollows:\n\n1. The context expression is evaluated to obtain a context manager.\n\n2. The context manager\'s ``__enter__()`` method is invoked.\n\n3. If a target was included in the ``with`` statement, the return\n   value from ``__enter__()`` is assigned to it.\n\n   Note: The ``with`` statement guarantees that if the ``__enter__()``\n     method returns without an error, then ``__exit__()`` will always\n     be called.  Thus, if an error occurs during the assignment to the\n     target list, it will be treated the same as an error occurring\n     within the suite would be.  See step 5 below.\n\n4. The suite is executed.\n\n5. The context manager\'s ``__exit__()`` method is invoked.  If an\n   exception caused the suite to be exited, its type, value, and\n   traceback are passed as arguments to ``__exit__()``. Otherwise,\n   three ``None`` arguments are supplied.\n\n   If the suite was exited due to an exception, and the return value\n   from the ``__exit__()`` method was false, the exception is\n   reraised.  If the return value was true, the exception is\n   suppressed, and execution continues with the statement following\n   the ``with`` statement.\n\n   If the suite was exited for any reason other than an exception, the\n   return value from ``__exit__()`` is ignored, and execution proceeds\n   at the normal location for the kind of exit that was taken.\n\nWith more than one item, the context managers are processed as if\nmultiple ``with`` statements were nested:\n\n   with A() as a, B() as b:\n       suite\n\nis equivalent to\n\n   with A() as a:\n       with B() as b:\n           suite\n\nChanged in version 3.1: Support for multiple context expressions.\n\nSee also:\n\n   **PEP 0343** - The "with" statement\n      The specification, background, and examples for the Python\n      ``with`` statement.\n\n\nFunction definitions\n====================\n\nA function definition defines a user-defined function object (see\nsection *The standard type hierarchy*):\n\n   funcdef        ::= [decorators] "def" funcname "(" [parameter_list] ")" ["->" expression] ":" suite\n   decorators     ::= decorator+\n   decorator      ::= "@" dotted_name ["(" [argument_list [","]] ")"] NEWLINE\n   dotted_name    ::= identifier ("." identifier)*\n   parameter_list ::= (defparameter ",")*\n                      (  "*" [parameter] ("," defparameter)*\n                      [, "**" parameter]\n                      | "**" parameter\n                      | defparameter [","] )\n   parameter      ::= identifier [":" expression]\n   defparameter   ::= parameter ["=" expression]\n   funcname       ::= identifier\n\nA function definition is an executable statement.  Its execution binds\nthe function name in the current local namespace to a function object\n(a wrapper around the executable code for the function).  This\nfunction object contains a reference to the current global namespace\nas the global namespace to be used when the function is called.\n\nThe function definition does not execute the function body; this gets\nexecuted only when the function is called. [3]\n\nA function definition may be wrapped by one or more *decorator*\nexpressions. Decorator expressions are evaluated when the function is\ndefined, in the scope that contains the function definition.  The\nresult must be a callable, which is invoked with the function object\nas the only argument. The returned value is bound to the function name\ninstead of the function object.  Multiple decorators are applied in\nnested fashion. For example, the following code\n\n   @f1(arg)\n   @f2\n   def func(): pass\n\nis equivalent to\n\n   def func(): pass\n   func = f1(arg)(f2(func))\n\nWhen one or more parameters have the form *parameter* ``=``\n*expression*, the function is said to have "default parameter values."\nFor a parameter with a default value, the corresponding argument may\nbe omitted from a call, in which case the parameter\'s default value is\nsubstituted.  If a parameter has a default value, all following\nparameters up until the "``*``" must also have a default value ---\nthis is a syntactic restriction that is not expressed by the grammar.\n\n**Default parameter values are evaluated when the function definition\nis executed.** This means that the expression is evaluated once, when\nthe function is defined, and that that same "pre-computed" value is\nused for each call.  This is especially important to understand when a\ndefault parameter is a mutable object, such as a list or a dictionary:\nif the function modifies the object (e.g. by appending an item to a\nlist), the default value is in effect modified. This is generally not\nwhat was intended.  A way around this is to use ``None`` as the\ndefault, and explicitly test for it in the body of the function, e.g.:\n\n   def whats_on_the_telly(penguin=None):\n       if penguin is None:\n           penguin = []\n       penguin.append("property of the zoo")\n       return penguin\n\nFunction call semantics are described in more detail in section\n*Calls*. A function call always assigns values to all parameters\nmentioned in the parameter list, either from position arguments, from\nkeyword arguments, or from default values.  If the form\n"``*identifier``" is present, it is initialized to a tuple receiving\nany excess positional parameters, defaulting to the empty tuple.  If\nthe form "``**identifier``" is present, it is initialized to a new\ndictionary receiving any excess keyword arguments, defaulting to a new\nempty dictionary. Parameters after "``*``" or "``*identifier``" are\nkeyword-only parameters and may only be passed used keyword arguments.\n\nParameters may have annotations of the form "``: expression``"\nfollowing the parameter name.  Any parameter may have an annotation\neven those of the form ``*identifier`` or ``**identifier``.  Functions\nmay have "return" annotation of the form "``-> expression``" after the\nparameter list.  These annotations can be any valid Python expression\nand are evaluated when the function definition is executed.\nAnnotations may be evaluated in a different order than they appear in\nthe source code.  The presence of annotations does not change the\nsemantics of a function.  The annotation values are available as\nvalues of a dictionary keyed by the parameters\' names in the\n``__annotations__`` attribute of the function object.\n\nIt is also possible to create anonymous functions (functions not bound\nto a name), for immediate use in expressions.  This uses lambda forms,\ndescribed in section *Lambdas*.  Note that the lambda form is merely a\nshorthand for a simplified function definition; a function defined in\na "``def``" statement can be passed around or assigned to another name\njust like a function defined by a lambda form.  The "``def``" form is\nactually more powerful since it allows the execution of multiple\nstatements and annotations.\n\n**Programmer\'s note:** Functions are first-class objects.  A "``def``"\nform executed inside a function definition defines a local function\nthat can be returned or passed around.  Free variables used in the\nnested function can access the local variables of the function\ncontaining the def.  See section *Naming and binding* for details.\n\n\nClass definitions\n=================\n\nA class definition defines a class object (see section *The standard\ntype hierarchy*):\n\n   classdef    ::= [decorators] "class" classname [inheritance] ":" suite\n   inheritance ::= "(" [expression_list] ")"\n   classname   ::= identifier\n\nA class definition is an executable statement.  It first evaluates the\ninheritance list, if present.  Each item in the inheritance list\nshould evaluate to a class object or class type which allows\nsubclassing.  The class\'s suite is then executed in a new execution\nframe (see section *Naming and binding*), using a newly created local\nnamespace and the original global namespace. (Usually, the suite\ncontains only function definitions.)  When the class\'s suite finishes\nexecution, its execution frame is discarded but its local namespace is\nsaved. [4] A class object is then created using the inheritance list\nfor the base classes and the saved local namespace for the attribute\ndictionary.  The class name is bound to this class object in the\noriginal local namespace.\n\nClasses can also be decorated; as with functions,\n\n   @f1(arg)\n   @f2\n   class Foo: pass\n\nis equivalent to\n\n   class Foo: pass\n   Foo = f1(arg)(f2(Foo))\n\n**Programmer\'s note:** Variables defined in the class definition are\nclass variables; they are shared by instances. Instance variables can\nbe set in a method with ``self.name = value``.  Both class and\ninstance variables are accessible through the notation\n"``self.name``", and an instance variable hides a class variable with\nthe same name when accessed in this way.  Class variables can be used\nas defaults for instance variables, but using mutable values there can\nlead to unexpected results.  Descriptors can be used to create\ninstance variables with different implementation details.\n\nSee also:\n\n   **PEP 3129** - Class Decorators\n\nClass definitions, like function definitions, may be wrapped by one or\nmore *decorator* expressions.  The evaluation rules for the decorator\nexpressions are the same as for functions.  The result must be a class\nobject, which is then bound to the class name.\n\n-[ Footnotes ]-\n\n[1] The exception is propagated to the invocation stack only if there\n    is no ``finally`` clause that negates the exception.\n\n[2] Currently, control "flows off the end" except in the case of an\n    exception or the execution of a ``return``, ``continue``, or\n    ``break`` statement.\n\n[3] A string literal appearing as the first statement in the function\n    body is transformed into the function\'s ``__doc__`` attribute and\n    therefore the function\'s *docstring*.\n\n[4] A string literal appearing as the first statement in the class\n    body is transformed into the namespace\'s ``__doc__`` item and\n    therefore the class\'s *docstring*.\n',
+ 'comparisons': '\nComparisons\n***********\n\nUnlike C, all comparison operations in Python have the same priority,\nwhich is lower than that of any arithmetic, shifting or bitwise\noperation.  Also unlike C, expressions like ``a < b < c`` have the\ninterpretation that is conventional in mathematics:\n\n   comparison    ::= or_expr ( comp_operator or_expr )*\n   comp_operator ::= "<" | ">" | "==" | ">=" | "<=" | "!="\n                     | "is" ["not"] | ["not"] "in"\n\nComparisons yield boolean values: ``True`` or ``False``.\n\nComparisons can be chained arbitrarily, e.g., ``x < y <= z`` is\nequivalent to ``x < y and y <= z``, except that ``y`` is evaluated\nonly once (but in both cases ``z`` is not evaluated at all when ``x <\ny`` is found to be false).\n\nFormally, if *a*, *b*, *c*, ..., *y*, *z* are expressions and *op1*,\n*op2*, ..., *opN* are comparison operators, then ``a op1 b op2 c ... y\nopN z`` is equivalent to ``a op1 b and b op2 c and ... y opN z``,\nexcept that each expression is evaluated at most once.\n\nNote that ``a op1 b op2 c`` doesn\'t imply any kind of comparison\nbetween *a* and *c*, so that, e.g., ``x < y > z`` is perfectly legal\n(though perhaps not pretty).\n\nThe operators ``<``, ``>``, ``==``, ``>=``, ``<=``, and ``!=`` compare\nthe values of two objects.  The objects need not have the same type.\nIf both are numbers, they are converted to a common type.  Otherwise,\nthe ``==`` and ``!=`` operators *always* consider objects of different\ntypes to be unequal, while the ``<``, ``>``, ``>=`` and ``<=``\noperators raise a ``TypeError`` when comparing objects of different\ntypes that do not implement these operators for the given pair of\ntypes.  You can control comparison behavior of objects of non-built-in\ntypes by defining rich comparison methods like ``__gt__()``, described\nin section *Basic customization*.\n\nComparison of objects of the same type depends on the type:\n\n* Numbers are compared arithmetically.\n\n* The values ``float(\'NaN\')`` and ``Decimal(\'NaN\')`` are special. The\n  are identical to themselves, ``x is x`` but are not equal to\n  themselves, ``x != x``.  Additionally, comparing any value to a\n  not-a-number value will return ``False``.  For example, both ``3 <\n  float(\'NaN\')`` and ``float(\'NaN\') < 3`` will return ``False``.\n\n* Bytes objects are compared lexicographically using the numeric\n  values of their elements.\n\n* Strings are compared lexicographically using the numeric equivalents\n  (the result of the built-in function ``ord()``) of their characters.\n  [3] String and bytes object can\'t be compared!\n\n* Tuples and lists are compared lexicographically using comparison of\n  corresponding elements.  This means that to compare equal, each\n  element must compare equal and the two sequences must be of the same\n  type and have the same length.\n\n  If not equal, the sequences are ordered the same as their first\n  differing elements.  For example, ``[1,2,x] <= [1,2,y]`` has the\n  same value as ``x <= y``.  If the corresponding element does not\n  exist, the shorter sequence is ordered first (for example, ``[1,2] <\n  [1,2,3]``).\n\n* Mappings (dictionaries) compare equal if and only if they have the\n  same ``(key, value)`` pairs. Order comparisons ``(\'<\', \'<=\', \'>=\',\n  \'>\')`` raise ``TypeError``.\n\n* Sets and frozensets define comparison operators to mean subset and\n  superset tests.  Those relations do not define total orderings (the\n  two sets ``{1,2}`` and {2,3} are not equal, nor subsets of one\n  another, nor supersets of one another).  Accordingly, sets are not\n  appropriate arguments for functions which depend on total ordering.\n  For example, ``min()``, ``max()``, and ``sorted()`` produce\n  undefined results given a list of sets as inputs.\n\n* Most other objects of built-in types compare unequal unless they are\n  the same object; the choice whether one object is considered smaller\n  or larger than another one is made arbitrarily but consistently\n  within one execution of a program.\n\nComparison of objects of the differing types depends on whether either\nof the types provide explicit support for the comparison.  Most\nnumeric types can be compared with one another, but comparisons of\n``float`` and ``Decimal`` are not supported to avoid the inevitable\nconfusion arising from representation issues such as ``float(\'1.1\')``\nbeing inexactly represented and therefore not exactly equal to\n``Decimal(\'1.1\')`` which is.  When cross-type comparison is not\nsupported, the comparison method returns ``NotImplemented``.  This can\ncreate the illusion of non-transitivity between supported cross-type\ncomparisons and unsupported comparisons.  For example, ``Decimal(2) ==\n2`` and *2 == float(2)`* but ``Decimal(2) != float(2)``.\n\nThe operators ``in`` and ``not in`` test for membership.  ``x in s``\nevaluates to true if *x* is a member of *s*, and false otherwise.  ``x\nnot in s`` returns the negation of ``x in s``.  All built-in sequences\nand set types support this as well as dictionary, for which ``in``\ntests whether a the dictionary has a given key. For container types\nsuch as list, tuple, set, frozenset, dict, or collections.deque, the\nexpression ``x in y`` is equivalent to ``any(x is e or x == e for e in\ny)``.\n\nFor the string and bytes types, ``x in y`` is true if and only if *x*\nis a substring of *y*.  An equivalent test is ``y.find(x) != -1``.\nEmpty strings are always considered to be a substring of any other\nstring, so ``"" in "abc"`` will return ``True``.\n\nFor user-defined classes which define the ``__contains__()`` method,\n``x in y`` is true if and only if ``y.__contains__(x)`` is true.\n\nFor user-defined classes which do not define ``__contains__()`` but do\ndefine ``__iter__()``, ``x in y`` is true if some value ``z`` with ``x\n== z`` is produced while iterating over ``y``.  If an exception is\nraised during the iteration, it is as if ``in`` raised that exception.\n\nLastly, the old-style iteration protocol is tried: if a class defines\n``__getitem__()``, ``x in y`` is true if and only if there is a non-\nnegative integer index *i* such that ``x == y[i]``, and all lower\ninteger indices do not raise ``IndexError`` exception.  (If any other\nexception is raised, it is as if ``in`` raised that exception).\n\nThe operator ``not in`` is defined to have the inverse true value of\n``in``.\n\nThe operators ``is`` and ``is not`` test for object identity: ``x is\ny`` is true if and only if *x* and *y* are the same object.  ``x is\nnot y`` yields the inverse truth value. [4]\n',
+ 'compound': '\nCompound statements\n*******************\n\nCompound statements contain (groups of) other statements; they affect\nor control the execution of those other statements in some way.  In\ngeneral, compound statements span multiple lines, although in simple\nincarnations a whole compound statement may be contained in one line.\n\nThe ``if``, ``while`` and ``for`` statements implement traditional\ncontrol flow constructs.  ``try`` specifies exception handlers and/or\ncleanup code for a group of statements, while the ``with`` statement\nallows the execution of initialization and finalization code around a\nblock of code.  Function and class definitions are also syntactically\ncompound statements.\n\nCompound statements consist of one or more \'clauses.\'  A clause\nconsists of a header and a \'suite.\'  The clause headers of a\nparticular compound statement are all at the same indentation level.\nEach clause header begins with a uniquely identifying keyword and ends\nwith a colon.  A suite is a group of statements controlled by a\nclause.  A suite can be one or more semicolon-separated simple\nstatements on the same line as the header, following the header\'s\ncolon, or it can be one or more indented statements on subsequent\nlines.  Only the latter form of suite can contain nested compound\nstatements; the following is illegal, mostly because it wouldn\'t be\nclear to which ``if`` clause a following ``else`` clause would belong:\n\n   if test1: if test2: print(x)\n\nAlso note that the semicolon binds tighter than the colon in this\ncontext, so that in the following example, either all or none of the\n``print()`` calls are executed:\n\n   if x < y < z: print(x); print(y); print(z)\n\nSummarizing:\n\n   compound_stmt ::= if_stmt\n                     | while_stmt\n                     | for_stmt\n                     | try_stmt\n                     | with_stmt\n                     | funcdef\n                     | classdef\n   suite         ::= stmt_list NEWLINE | NEWLINE INDENT statement+ DEDENT\n   statement     ::= stmt_list NEWLINE | compound_stmt\n   stmt_list     ::= simple_stmt (";" simple_stmt)* [";"]\n\nNote that statements always end in a ``NEWLINE`` possibly followed by\na ``DEDENT``.  Also note that optional continuation clauses always\nbegin with a keyword that cannot start a statement, thus there are no\nambiguities (the \'dangling ``else``\' problem is solved in Python by\nrequiring nested ``if`` statements to be indented).\n\nThe formatting of the grammar rules in the following sections places\neach clause on a separate line for clarity.\n\n\nThe ``if`` statement\n====================\n\nThe ``if`` statement is used for conditional execution:\n\n   if_stmt ::= "if" expression ":" suite\n               ( "elif" expression ":" suite )*\n               ["else" ":" suite]\n\nIt selects exactly one of the suites by evaluating the expressions one\nby one until one is found to be true (see section *Boolean operations*\nfor the definition of true and false); then that suite is executed\n(and no other part of the ``if`` statement is executed or evaluated).\nIf all expressions are false, the suite of the ``else`` clause, if\npresent, is executed.\n\n\nThe ``while`` statement\n=======================\n\nThe ``while`` statement is used for repeated execution as long as an\nexpression is true:\n\n   while_stmt ::= "while" expression ":" suite\n                  ["else" ":" suite]\n\nThis repeatedly tests the expression and, if it is true, executes the\nfirst suite; if the expression is false (which may be the first time\nit is tested) the suite of the ``else`` clause, if present, is\nexecuted and the loop terminates.\n\nA ``break`` statement executed in the first suite terminates the loop\nwithout executing the ``else`` clause\'s suite.  A ``continue``\nstatement executed in the first suite skips the rest of the suite and\ngoes back to testing the expression.\n\n\nThe ``for`` statement\n=====================\n\nThe ``for`` statement is used to iterate over the elements of a\nsequence (such as a string, tuple or list) or other iterable object:\n\n   for_stmt ::= "for" target_list "in" expression_list ":" suite\n                ["else" ":" suite]\n\nThe expression list is evaluated once; it should yield an iterable\nobject.  An iterator is created for the result of the\n``expression_list``.  The suite is then executed once for each item\nprovided by the iterator, in the order of ascending indices.  Each\nitem in turn is assigned to the target list using the standard rules\nfor assignments (see *Assignment statements*), and then the suite is\nexecuted.  When the items are exhausted (which is immediately when the\nsequence is empty or an iterator raises a ``StopIteration``\nexception), the suite in the ``else`` clause, if present, is executed,\nand the loop terminates.\n\nA ``break`` statement executed in the first suite terminates the loop\nwithout executing the ``else`` clause\'s suite.  A ``continue``\nstatement executed in the first suite skips the rest of the suite and\ncontinues with the next item, or with the ``else`` clause if there was\nno next item.\n\nThe suite may assign to the variable(s) in the target list; this does\nnot affect the next item assigned to it.\n\nNames in the target list are not deleted when the loop is finished,\nbut if the sequence is empty, it will not have been assigned to at all\nby the loop.  Hint: the built-in function ``range()`` returns an\niterator of integers suitable to emulate the effect of Pascal\'s ``for\ni := a to b do``; e.g., ``list(range(3))`` returns the list ``[0, 1,\n2]``.\n\nNote: There is a subtlety when the sequence is being modified by the loop\n  (this can only occur for mutable sequences, i.e. lists).  An\n  internal counter is used to keep track of which item is used next,\n  and this is incremented on each iteration.  When this counter has\n  reached the length of the sequence the loop terminates.  This means\n  that if the suite deletes the current (or a previous) item from the\n  sequence, the next item will be skipped (since it gets the index of\n  the current item which has already been treated).  Likewise, if the\n  suite inserts an item in the sequence before the current item, the\n  current item will be treated again the next time through the loop.\n  This can lead to nasty bugs that can be avoided by making a\n  temporary copy using a slice of the whole sequence, e.g.,\n\n     for x in a[:]:\n         if x < 0: a.remove(x)\n\n\nThe ``try`` statement\n=====================\n\nThe ``try`` statement specifies exception handlers and/or cleanup code\nfor a group of statements:\n\n   try_stmt  ::= try1_stmt | try2_stmt\n   try1_stmt ::= "try" ":" suite\n                 ("except" [expression ["as" target]] ":" suite)+\n                 ["else" ":" suite]\n                 ["finally" ":" suite]\n   try2_stmt ::= "try" ":" suite\n                 "finally" ":" suite\n\nThe ``except`` clause(s) specify one or more exception handlers. When\nno exception occurs in the ``try`` clause, no exception handler is\nexecuted. When an exception occurs in the ``try`` suite, a search for\nan exception handler is started.  This search inspects the except\nclauses in turn until one is found that matches the exception.  An\nexpression-less except clause, if present, must be last; it matches\nany exception.  For an except clause with an expression, that\nexpression is evaluated, and the clause matches the exception if the\nresulting object is "compatible" with the exception.  An object is\ncompatible with an exception if it is the class or a base class of the\nexception object or a tuple containing an item compatible with the\nexception.\n\nIf no except clause matches the exception, the search for an exception\nhandler continues in the surrounding code and on the invocation stack.\n[1]\n\nIf the evaluation of an expression in the header of an except clause\nraises an exception, the original search for a handler is canceled and\na search starts for the new exception in the surrounding code and on\nthe call stack (it is treated as if the entire ``try`` statement\nraised the exception).\n\nWhen a matching except clause is found, the exception is assigned to\nthe target specified after the ``as`` keyword in that except clause,\nif present, and the except clause\'s suite is executed.  All except\nclauses must have an executable block.  When the end of this block is\nreached, execution continues normally after the entire try statement.\n(This means that if two nested handlers exist for the same exception,\nand the exception occurs in the try clause of the inner handler, the\nouter handler will not handle the exception.)\n\nWhen an exception has been assigned using ``as target``, it is cleared\nat the end of the except clause.  This is as if\n\n   except E as N:\n       foo\n\nwas translated to\n\n   except E as N:\n       try:\n           foo\n       finally:\n           del N\n\nThis means the exception must be assigned to a different name to be\nable to refer to it after the except clause.  Exceptions are cleared\nbecause with the traceback attached to them, they form a reference\ncycle with the stack frame, keeping all locals in that frame alive\nuntil the next garbage collection occurs.\n\nBefore an except clause\'s suite is executed, details about the\nexception are stored in the ``sys`` module and can be access via\n``sys.exc_info()``. ``sys.exc_info()`` returns a 3-tuple consisting\nof: ``exc_type``, the exception class; ``exc_value``, the exception\ninstance; ``exc_traceback``, a traceback object (see section *The\nstandard type hierarchy*) identifying the point in the program where\nthe exception occurred. ``sys.exc_info()`` values are restored to\ntheir previous values (before the call) when returning from a function\nthat handled an exception.\n\nThe optional ``else`` clause is executed if and when control flows off\nthe end of the ``try`` clause. [2] Exceptions in the ``else`` clause\nare not handled by the preceding ``except`` clauses.\n\nIf ``finally`` is present, it specifies a \'cleanup\' handler.  The\n``try`` clause is executed, including any ``except`` and ``else``\nclauses.  If an exception occurs in any of the clauses and is not\nhandled, the exception is temporarily saved. The ``finally`` clause is\nexecuted.  If there is a saved exception, it is re-raised at the end\nof the ``finally`` clause. If the ``finally`` clause raises another\nexception or executes a ``return`` or ``break`` statement, the saved\nexception is lost.  The exception information is not available to the\nprogram during execution of the ``finally`` clause.\n\nWhen a ``return``, ``break`` or ``continue`` statement is executed in\nthe ``try`` suite of a ``try``...``finally`` statement, the\n``finally`` clause is also executed \'on the way out.\' A ``continue``\nstatement is illegal in the ``finally`` clause. (The reason is a\nproblem with the current implementation --- this restriction may be\nlifted in the future).\n\nAdditional information on exceptions can be found in section\n*Exceptions*, and information on using the ``raise`` statement to\ngenerate exceptions may be found in section *The raise statement*.\n\n\nThe ``with`` statement\n======================\n\nThe ``with`` statement is used to wrap the execution of a block with\nmethods defined by a context manager (see section *With Statement\nContext Managers*). This allows common\n``try``...``except``...``finally`` usage patterns to be encapsulated\nfor convenient reuse.\n\n   with_stmt ::= "with" with_item ("," with_item)* ":" suite\n   with_item ::= expression ["as" target]\n\nThe execution of the ``with`` statement with one "item" proceeds as\nfollows:\n\n1. The context expression is evaluated to obtain a context manager.\n\n2. The context manager\'s ``__exit__()`` is loaded for later use.\n\n3. The context manager\'s ``__enter__()`` method is invoked.\n\n4. If a target was included in the ``with`` statement, the return\n   value from ``__enter__()`` is assigned to it.\n\n   Note: The ``with`` statement guarantees that if the ``__enter__()``\n     method returns without an error, then ``__exit__()`` will always\n     be called. Thus, if an error occurs during the assignment to the\n     target list, it will be treated the same as an error occurring\n     within the suite would be. See step 6 below.\n\n5. The suite is executed.\n\n6. The context manager\'s ``__exit__()`` method is invoked.  If an\n   exception caused the suite to be exited, its type, value, and\n   traceback are passed as arguments to ``__exit__()``. Otherwise,\n   three ``None`` arguments are supplied.\n\n   If the suite was exited due to an exception, and the return value\n   from the ``__exit__()`` method was false, the exception is\n   reraised.  If the return value was true, the exception is\n   suppressed, and execution continues with the statement following\n   the ``with`` statement.\n\n   If the suite was exited for any reason other than an exception, the\n   return value from ``__exit__()`` is ignored, and execution proceeds\n   at the normal location for the kind of exit that was taken.\n\nWith more than one item, the context managers are processed as if\nmultiple ``with`` statements were nested:\n\n   with A() as a, B() as b:\n       suite\n\nis equivalent to\n\n   with A() as a:\n       with B() as b:\n           suite\n\nChanged in version 3.1: Support for multiple context expressions.\n\nSee also:\n\n   **PEP 0343** - The "with" statement\n      The specification, background, and examples for the Python\n      ``with`` statement.\n\n\nFunction definitions\n====================\n\nA function definition defines a user-defined function object (see\nsection *The standard type hierarchy*):\n\n   funcdef        ::= [decorators] "def" funcname "(" [parameter_list] ")" ["->" expression] ":" suite\n   decorators     ::= decorator+\n   decorator      ::= "@" dotted_name ["(" [argument_list [","]] ")"] NEWLINE\n   dotted_name    ::= identifier ("." identifier)*\n   parameter_list ::= (defparameter ",")*\n                      (  "*" [parameter] ("," defparameter)*\n                      [, "**" parameter]\n                      | "**" parameter\n                      | defparameter [","] )\n   parameter      ::= identifier [":" expression]\n   defparameter   ::= parameter ["=" expression]\n   funcname       ::= identifier\n\nA function definition is an executable statement.  Its execution binds\nthe function name in the current local namespace to a function object\n(a wrapper around the executable code for the function).  This\nfunction object contains a reference to the current global namespace\nas the global namespace to be used when the function is called.\n\nThe function definition does not execute the function body; this gets\nexecuted only when the function is called. [3]\n\nA function definition may be wrapped by one or more *decorator*\nexpressions. Decorator expressions are evaluated when the function is\ndefined, in the scope that contains the function definition.  The\nresult must be a callable, which is invoked with the function object\nas the only argument. The returned value is bound to the function name\ninstead of the function object.  Multiple decorators are applied in\nnested fashion. For example, the following code\n\n   @f1(arg)\n   @f2\n   def func(): pass\n\nis equivalent to\n\n   def func(): pass\n   func = f1(arg)(f2(func))\n\nWhen one or more parameters have the form *parameter* ``=``\n*expression*, the function is said to have "default parameter values."\nFor a parameter with a default value, the corresponding argument may\nbe omitted from a call, in which case the parameter\'s default value is\nsubstituted.  If a parameter has a default value, all following\nparameters up until the "``*``" must also have a default value ---\nthis is a syntactic restriction that is not expressed by the grammar.\n\n**Default parameter values are evaluated when the function definition\nis executed.** This means that the expression is evaluated once, when\nthe function is defined, and that that same "pre-computed" value is\nused for each call.  This is especially important to understand when a\ndefault parameter is a mutable object, such as a list or a dictionary:\nif the function modifies the object (e.g. by appending an item to a\nlist), the default value is in effect modified. This is generally not\nwhat was intended.  A way around this is to use ``None`` as the\ndefault, and explicitly test for it in the body of the function, e.g.:\n\n   def whats_on_the_telly(penguin=None):\n       if penguin is None:\n           penguin = []\n       penguin.append("property of the zoo")\n       return penguin\n\nFunction call semantics are described in more detail in section\n*Calls*. A function call always assigns values to all parameters\nmentioned in the parameter list, either from position arguments, from\nkeyword arguments, or from default values.  If the form\n"``*identifier``" is present, it is initialized to a tuple receiving\nany excess positional parameters, defaulting to the empty tuple.  If\nthe form "``**identifier``" is present, it is initialized to a new\ndictionary receiving any excess keyword arguments, defaulting to a new\nempty dictionary. Parameters after "``*``" or "``*identifier``" are\nkeyword-only parameters and may only be passed used keyword arguments.\n\nParameters may have annotations of the form "``: expression``"\nfollowing the parameter name.  Any parameter may have an annotation\neven those of the form ``*identifier`` or ``**identifier``.  Functions\nmay have "return" annotation of the form "``-> expression``" after the\nparameter list.  These annotations can be any valid Python expression\nand are evaluated when the function definition is executed.\nAnnotations may be evaluated in a different order than they appear in\nthe source code.  The presence of annotations does not change the\nsemantics of a function.  The annotation values are available as\nvalues of a dictionary keyed by the parameters\' names in the\n``__annotations__`` attribute of the function object.\n\nIt is also possible to create anonymous functions (functions not bound\nto a name), for immediate use in expressions.  This uses lambda forms,\ndescribed in section *Lambdas*.  Note that the lambda form is merely a\nshorthand for a simplified function definition; a function defined in\na "``def``" statement can be passed around or assigned to another name\njust like a function defined by a lambda form.  The "``def``" form is\nactually more powerful since it allows the execution of multiple\nstatements and annotations.\n\n**Programmer\'s note:** Functions are first-class objects.  A "``def``"\nform executed inside a function definition defines a local function\nthat can be returned or passed around.  Free variables used in the\nnested function can access the local variables of the function\ncontaining the def.  See section *Naming and binding* for details.\n\n\nClass definitions\n=================\n\nA class definition defines a class object (see section *The standard\ntype hierarchy*):\n\n   classdef    ::= [decorators] "class" classname [inheritance] ":" suite\n   inheritance ::= "(" [expression_list] ")"\n   classname   ::= identifier\n\nA class definition is an executable statement.  It first evaluates the\ninheritance list, if present.  Each item in the inheritance list\nshould evaluate to a class object or class type which allows\nsubclassing.  The class\'s suite is then executed in a new execution\nframe (see section *Naming and binding*), using a newly created local\nnamespace and the original global namespace. (Usually, the suite\ncontains only function definitions.)  When the class\'s suite finishes\nexecution, its execution frame is discarded but its local namespace is\nsaved. [4] A class object is then created using the inheritance list\nfor the base classes and the saved local namespace for the attribute\ndictionary.  The class name is bound to this class object in the\noriginal local namespace.\n\nClasses can also be decorated; as with functions,\n\n   @f1(arg)\n   @f2\n   class Foo: pass\n\nis equivalent to\n\n   class Foo: pass\n   Foo = f1(arg)(f2(Foo))\n\n**Programmer\'s note:** Variables defined in the class definition are\nclass variables; they are shared by instances. Instance variables can\nbe set in a method with ``self.name = value``.  Both class and\ninstance variables are accessible through the notation\n"``self.name``", and an instance variable hides a class variable with\nthe same name when accessed in this way.  Class variables can be used\nas defaults for instance variables, but using mutable values there can\nlead to unexpected results.  Descriptors can be used to create\ninstance variables with different implementation details.\n\nSee also:\n\n   **PEP 3129** - Class Decorators\n\nClass definitions, like function definitions, may be wrapped by one or\nmore *decorator* expressions.  The evaluation rules for the decorator\nexpressions are the same as for functions.  The result must be a class\nobject, which is then bound to the class name.\n\n-[ Footnotes ]-\n\n[1] The exception is propagated to the invocation stack only if there\n    is no ``finally`` clause that negates the exception.\n\n[2] Currently, control "flows off the end" except in the case of an\n    exception or the execution of a ``return``, ``continue``, or\n    ``break`` statement.\n\n[3] A string literal appearing as the first statement in the function\n    body is transformed into the function\'s ``__doc__`` attribute and\n    therefore the function\'s *docstring*.\n\n[4] A string literal appearing as the first statement in the class\n    body is transformed into the namespace\'s ``__doc__`` item and\n    therefore the class\'s *docstring*.\n',
  'context-managers': '\nWith Statement Context Managers\n*******************************\n\nA *context manager* is an object that defines the runtime context to\nbe established when executing a ``with`` statement. The context\nmanager handles the entry into, and the exit from, the desired runtime\ncontext for the execution of the block of code.  Context managers are\nnormally invoked using the ``with`` statement (described in section\n*The with statement*), but can also be used by directly invoking their\nmethods.\n\nTypical uses of context managers include saving and restoring various\nkinds of global state, locking and unlocking resources, closing opened\nfiles, etc.\n\nFor more information on context managers, see *Context Manager Types*.\n\nobject.__enter__(self)\n\n   Enter the runtime context related to this object. The ``with``\n   statement will bind this method\'s return value to the target(s)\n   specified in the ``as`` clause of the statement, if any.\n\nobject.__exit__(self, exc_type, exc_value, traceback)\n\n   Exit the runtime context related to this object. The parameters\n   describe the exception that caused the context to be exited. If the\n   context was exited without an exception, all three arguments will\n   be ``None``.\n\n   If an exception is supplied, and the method wishes to suppress the\n   exception (i.e., prevent it from being propagated), it should\n   return a true value. Otherwise, the exception will be processed\n   normally upon exit from this method.\n\n   Note that ``__exit__()`` methods should not reraise the passed-in\n   exception; this is the caller\'s responsibility.\n\nSee also:\n\n   **PEP 0343** - The "with" statement\n      The specification, background, and examples for the Python\n      ``with`` statement.\n',
  'continue': '\nThe ``continue`` statement\n**************************\n\n   continue_stmt ::= "continue"\n\n``continue`` may only occur syntactically nested in a ``for`` or\n``while`` loop, but not nested in a function or class definition or\n``finally`` clause within that loop.  It continues with the next cycle\nof the nearest enclosing loop.\n\nWhen ``continue`` passes control out of a ``try`` statement with a\n``finally`` clause, that ``finally`` clause is executed before really\nstarting the next loop cycle.\n',
  'conversions': '\nArithmetic conversions\n**********************\n\nWhen a description of an arithmetic operator below uses the phrase\n"the numeric arguments are converted to a common type," this means\nthat the operator implementation for built-in types works that way:\n\n* If either argument is a complex number, the other is converted to\n  complex;\n\n* otherwise, if either argument is a floating point number, the other\n  is converted to floating point;\n\n* otherwise, both must be integers and no conversion is necessary.\n\nSome additional rules apply for certain operators (e.g., a string left\nargument to the \'%\' operator).  Extensions must define their own\nconversion behavior.\n',
- 'customization': '\nBasic customization\n*******************\n\nobject.__new__(cls[, ...])\n\n   Called to create a new instance of class *cls*.  ``__new__()`` is a\n   static method (special-cased so you need not declare it as such)\n   that takes the class of which an instance was requested as its\n   first argument.  The remaining arguments are those passed to the\n   object constructor expression (the call to the class).  The return\n   value of ``__new__()`` should be the new object instance (usually\n   an instance of *cls*).\n\n   Typical implementations create a new instance of the class by\n   invoking the superclass\'s ``__new__()`` method using\n   ``super(currentclass, cls).__new__(cls[, ...])`` with appropriate\n   arguments and then modifying the newly-created instance as\n   necessary before returning it.\n\n   If ``__new__()`` returns an instance of *cls*, then the new\n   instance\'s ``__init__()`` method will be invoked like\n   ``__init__(self[, ...])``, where *self* is the new instance and the\n   remaining arguments are the same as were passed to ``__new__()``.\n\n   If ``__new__()`` does not return an instance of *cls*, then the new\n   instance\'s ``__init__()`` method will not be invoked.\n\n   ``__new__()`` is intended mainly to allow subclasses of immutable\n   types (like int, str, or tuple) to customize instance creation.  It\n   is also commonly overridden in custom metaclasses in order to\n   customize class creation.\n\nobject.__init__(self[, ...])\n\n   Called when the instance is created.  The arguments are those\n   passed to the class constructor expression.  If a base class has an\n   ``__init__()`` method, the derived class\'s ``__init__()`` method,\n   if any, must explicitly call it to ensure proper initialization of\n   the base class part of the instance; for example:\n   ``BaseClass.__init__(self, [args...])``.  As a special constraint\n   on constructors, no value may be returned; doing so will cause a\n   ``TypeError`` to be raised at runtime.\n\nobject.__del__(self)\n\n   Called when the instance is about to be destroyed.  This is also\n   called a destructor.  If a base class has a ``__del__()`` method,\n   the derived class\'s ``__del__()`` method, if any, must explicitly\n   call it to ensure proper deletion of the base class part of the\n   instance.  Note that it is possible (though not recommended!) for\n   the ``__del__()`` method to postpone destruction of the instance by\n   creating a new reference to it.  It may then be called at a later\n   time when this new reference is deleted.  It is not guaranteed that\n   ``__del__()`` methods are called for objects that still exist when\n   the interpreter exits.\n\n   Note: ``del x`` doesn\'t directly call ``x.__del__()`` --- the former\n     decrements the reference count for ``x`` by one, and the latter\n     is only called when ``x``\'s reference count reaches zero.  Some\n     common situations that may prevent the reference count of an\n     object from going to zero include: circular references between\n     objects (e.g., a doubly-linked list or a tree data structure with\n     parent and child pointers); a reference to the object on the\n     stack frame of a function that caught an exception (the traceback\n     stored in ``sys.exc_info()[2]`` keeps the stack frame alive); or\n     a reference to the object on the stack frame that raised an\n     unhandled exception in interactive mode (the traceback stored in\n     ``sys.last_traceback`` keeps the stack frame alive).  The first\n     situation can only be remedied by explicitly breaking the cycles;\n     the latter two situations can be resolved by storing ``None`` in\n     ``sys.last_traceback``. Circular references which are garbage are\n     detected when the option cycle detector is enabled (it\'s on by\n     default), but can only be cleaned up if there are no Python-\n     level ``__del__()`` methods involved. Refer to the documentation\n     for the ``gc`` module for more information about how\n     ``__del__()`` methods are handled by the cycle detector,\n     particularly the description of the ``garbage`` value.\n\n   Warning: Due to the precarious circumstances under which ``__del__()``\n     methods are invoked, exceptions that occur during their execution\n     are ignored, and a warning is printed to ``sys.stderr`` instead.\n     Also, when ``__del__()`` is invoked in response to a module being\n     deleted (e.g., when execution of the program is done), other\n     globals referenced by the ``__del__()`` method may already have\n     been deleted or in the process of being torn down (e.g. the\n     import machinery shutting down).  For this reason, ``__del__()``\n     methods should do the absolute minimum needed to maintain\n     external invariants.  Starting with version 1.5, Python\n     guarantees that globals whose name begins with a single\n     underscore are deleted from their module before other globals are\n     deleted; if no other references to such globals exist, this may\n     help in assuring that imported modules are still available at the\n     time when the ``__del__()`` method is called.\n\nobject.__repr__(self)\n\n   Called by the ``repr()`` built-in function to compute the\n   "official" string representation of an object.  If at all possible,\n   this should look like a valid Python expression that could be used\n   to recreate an object with the same value (given an appropriate\n   environment).  If this is not possible, a string of the form\n   ``<...some useful description...>`` should be returned. The return\n   value must be a string object. If a class defines ``__repr__()``\n   but not ``__str__()``, then ``__repr__()`` is also used when an\n   "informal" string representation of instances of that class is\n   required.\n\n   This is typically used for debugging, so it is important that the\n   representation is information-rich and unambiguous.\n\nobject.__str__(self)\n\n   Called by the ``str()`` built-in function and by the ``print()``\n   function to compute the "informal" string representation of an\n   object.  This differs from ``__repr__()`` in that it does not have\n   to be a valid Python expression: a more convenient or concise\n   representation may be used instead. The return value must be a\n   string object.\n\nobject.__format__(self, format_spec)\n\n   Called by the ``format()`` built-in function (and by extension, the\n   ``format()`` method of class ``str``) to produce a "formatted"\n   string representation of an object. The ``format_spec`` argument is\n   a string that contains a description of the formatting options\n   desired. The interpretation of the ``format_spec`` argument is up\n   to the type implementing ``__format__()``, however most classes\n   will either delegate formatting to one of the built-in types, or\n   use a similar formatting option syntax.\n\n   See *Format Specification Mini-Language* for a description of the\n   standard formatting syntax.\n\n   The return value must be a string object.\n\nobject.__lt__(self, other)\nobject.__le__(self, other)\nobject.__eq__(self, other)\nobject.__ne__(self, other)\nobject.__gt__(self, other)\nobject.__ge__(self, other)\n\n   These are the so-called "rich comparison" methods. The\n   correspondence between operator symbols and method names is as\n   follows: ``x<y`` calls ``x.__lt__(y)``, ``x<=y`` calls\n   ``x.__le__(y)``, ``x==y`` calls ``x.__eq__(y)``, ``x!=y`` calls\n   ``x.__ne__(y)``, ``x>y`` calls ``x.__gt__(y)``, and ``x>=y`` calls\n   ``x.__ge__(y)``.\n\n   A rich comparison method may return the singleton\n   ``NotImplemented`` if it does not implement the operation for a\n   given pair of arguments. By convention, ``False`` and ``True`` are\n   returned for a successful comparison. However, these methods can\n   return any value, so if the comparison operator is used in a\n   Boolean context (e.g., in the condition of an ``if`` statement),\n   Python will call ``bool()`` on the value to determine if the result\n   is true or false.\n\n   There are no implied relationships among the comparison operators.\n   The truth of ``x==y`` does not imply that ``x!=y`` is false.\n   Accordingly, when defining ``__eq__()``, one should also define\n   ``__ne__()`` so that the operators will behave as expected.  See\n   the paragraph on ``__hash__()`` for some important notes on\n   creating *hashable* objects which support custom comparison\n   operations and are usable as dictionary keys.\n\n   There are no swapped-argument versions of these methods (to be used\n   when the left argument does not support the operation but the right\n   argument does); rather, ``__lt__()`` and ``__gt__()`` are each\n   other\'s reflection, ``__le__()`` and ``__ge__()`` are each other\'s\n   reflection, and ``__eq__()`` and ``__ne__()`` are their own\n   reflection.\n\n   Arguments to rich comparison methods are never coerced.\n\n   To automatically generate ordering operations from a single root\n   operation, see the Total Ordering recipe in the ASPN cookbook.\n\nobject.__hash__(self)\n\n   Called by built-in function ``hash()`` and for operations on\n   members of hashed collections including ``set``, ``frozenset``, and\n   ``dict``.  ``__hash__()`` should return an integer.  The only\n   required property is that objects which compare equal have the same\n   hash value; it is advised to somehow mix together (e.g. using\n   exclusive or) the hash values for the components of the object that\n   also play a part in comparison of objects.\n\n   If a class does not define an ``__eq__()`` method it should not\n   define a ``__hash__()`` operation either; if it defines\n   ``__eq__()`` but not ``__hash__()``, its instances will not be\n   usable as items in hashable collections.  If a class defines\n   mutable objects and implements an ``__eq__()`` method, it should\n   not implement ``__hash__()``, since the implementation of hashable\n   collections requires that a key\'s hash value is immutable (if the\n   object\'s hash value changes, it will be in the wrong hash bucket).\n\n   User-defined classes have ``__eq__()`` and ``__hash__()`` methods\n   by default; with them, all objects compare unequal (except with\n   themselves) and ``x.__hash__()`` returns ``id(x)``.\n\n   Classes which inherit a ``__hash__()`` method from a parent class\n   but change the meaning of ``__eq__()`` such that the hash value\n   returned is no longer appropriate (e.g. by switching to a value-\n   based concept of equality instead of the default identity based\n   equality) can explicitly flag themselves as being unhashable by\n   setting ``__hash__ = None`` in the class definition. Doing so means\n   that not only will instances of the class raise an appropriate\n   ``TypeError`` when a program attempts to retrieve their hash value,\n   but they will also be correctly identified as unhashable when\n   checking ``isinstance(obj, collections.Hashable)`` (unlike classes\n   which define their own ``__hash__()`` to explicitly raise\n   ``TypeError``).\n\n   If a class that overrides ``__eq__()`` needs to retain the\n   implementation of ``__hash__()`` from a parent class, the\n   interpreter must be told this explicitly by setting ``__hash__ =\n   <ParentClass>.__hash__``. Otherwise the inheritance of\n   ``__hash__()`` will be blocked, just as if ``__hash__`` had been\n   explicitly set to ``None``.\n\nobject.__bool__(self)\n\n   Called to implement truth value testing and the built-in operation\n   ``bool()``; should return ``False`` or ``True``, or their integer\n   equivalents ``0`` or ``1``.  When this method is not defined,\n   ``__len__()`` is called, if it is defined, and the object is\n   considered true if its result is nonzero. If a class defines\n   neither ``__len__()`` nor ``__bool__()``, all its instances are\n   considered true.\n',
- 'debugger': '\n``pdb`` --- The Python Debugger\n*******************************\n\nThe module ``pdb`` defines an interactive source code debugger for\nPython programs.  It supports setting (conditional) breakpoints and\nsingle stepping at the source line level, inspection of stack frames,\nsource code listing, and evaluation of arbitrary Python code in the\ncontext of any stack frame.  It also supports post-mortem debugging\nand can be called under program control.\n\nThe debugger is extensible --- it is actually defined as the class\n``Pdb``. This is currently undocumented but easily understood by\nreading the source.  The extension interface uses the modules ``bdb``\n(undocumented) and ``cmd``.\n\nThe debugger\'s prompt is ``(Pdb)``. Typical usage to run a program\nunder control of the debugger is:\n\n   >>> import pdb\n   >>> import mymodule\n   >>> pdb.run(\'mymodule.test()\')\n   > <string>(0)?()\n   (Pdb) continue\n   > <string>(1)?()\n   (Pdb) continue\n   NameError: \'spam\'\n   > <string>(1)?()\n   (Pdb)\n\n``pdb.py`` can also be invoked as a script to debug other scripts.\nFor example:\n\n   python -m pdb myscript.py\n\nWhen invoked as a script, pdb will automatically enter post-mortem\ndebugging if the program being debugged exits abnormally. After post-\nmortem debugging (or after normal exit of the program), pdb will\nrestart the program. Automatic restarting preserves pdb\'s state (such\nas breakpoints) and in most cases is more useful than quitting the\ndebugger upon program\'s exit.\n\nThe typical usage to break into the debugger from a running program is\nto insert\n\n   import pdb; pdb.set_trace()\n\nat the location you want to break into the debugger.  You can then\nstep through the code following this statement, and continue running\nwithout debugger using the ``c`` command.\n\nThe typical usage to inspect a crashed program is:\n\n   >>> import pdb\n   >>> import mymodule\n   >>> mymodule.test()\n   Traceback (most recent call last):\n     File "<stdin>", line 1, in ?\n     File "./mymodule.py", line 4, in test\n       test2()\n     File "./mymodule.py", line 3, in test2\n       print(spam)\n   NameError: spam\n   >>> pdb.pm()\n   > ./mymodule.py(3)test2()\n   -> print(spam)\n   (Pdb)\n\nThe module defines the following functions; each enters the debugger\nin a slightly different way:\n\npdb.run(statement[, globals[, locals]])\n\n   Execute the *statement* (given as a string) under debugger control.\n   The debugger prompt appears before any code is executed; you can\n   set breakpoints and type ``continue``, or you can step through the\n   statement using ``step`` or ``next`` (all these commands are\n   explained below).  The optional *globals* and *locals* arguments\n   specify the environment in which the code is executed; by default\n   the dictionary of the module ``__main__`` is used.  (See the\n   explanation of the built-in ``exec()`` or ``eval()`` functions.)\n\npdb.runeval(expression[, globals[, locals]])\n\n   Evaluate the *expression* (given as a string) under debugger\n   control.  When ``runeval()`` returns, it returns the value of the\n   expression.  Otherwise this function is similar to ``run()``.\n\npdb.runcall(function[, argument, ...])\n\n   Call the *function* (a function or method object, not a string)\n   with the given arguments.  When ``runcall()`` returns, it returns\n   whatever the function call returned.  The debugger prompt appears\n   as soon as the function is entered.\n\npdb.set_trace()\n\n   Enter the debugger at the calling stack frame.  This is useful to\n   hard-code a breakpoint at a given point in a program, even if the\n   code is not otherwise being debugged (e.g. when an assertion\n   fails).\n\npdb.post_mortem([traceback])\n\n   Enter post-mortem debugging of the given *traceback* object.  If no\n   *traceback* is given, it uses the one of the exception that is\n   currently being handled (an exception must be being handled if the\n   default is to be used).\n\npdb.pm()\n\n   Enter post-mortem debugging of the traceback found in\n   ``sys.last_traceback``.\n\nThe ``run_*`` functions and ``set_trace()`` are aliases for\ninstantiating the ``Pdb`` class and calling the method of the same\nname.  If you want to access further features, you have to do this\nyourself:\n\nclass class pdb.Pdb(completekey=\'tab\', stdin=None, stdout=None, skip=None)\n\n   ``Pdb`` is the debugger class.\n\n   The *completekey*, *stdin* and *stdout* arguments are passed to the\n   underlying ``cmd.Cmd`` class; see the description there.\n\n   The *skip* argument, if given, must be an iterable of glob-style\n   module name patterns.  The debugger will not step into frames that\n   originate in a module that matches one of these patterns. [1]\n\n   Example call to enable tracing with *skip*:\n\n      import pdb; pdb.Pdb(skip=[\'django.*\']).set_trace()\n\n   New in version 3.1: The *skip* argument.\n\n   run(statement[, globals[, locals]])\n   runeval(expression[, globals[, locals]])\n   runcall(function[, argument, ...])\n   set_trace()\n\n      See the documentation for the functions explained above.\n',
+ 'customization': '\nBasic customization\n*******************\n\nobject.__new__(cls[, ...])\n\n   Called to create a new instance of class *cls*.  ``__new__()`` is a\n   static method (special-cased so you need not declare it as such)\n   that takes the class of which an instance was requested as its\n   first argument.  The remaining arguments are those passed to the\n   object constructor expression (the call to the class).  The return\n   value of ``__new__()`` should be the new object instance (usually\n   an instance of *cls*).\n\n   Typical implementations create a new instance of the class by\n   invoking the superclass\'s ``__new__()`` method using\n   ``super(currentclass, cls).__new__(cls[, ...])`` with appropriate\n   arguments and then modifying the newly-created instance as\n   necessary before returning it.\n\n   If ``__new__()`` returns an instance of *cls*, then the new\n   instance\'s ``__init__()`` method will be invoked like\n   ``__init__(self[, ...])``, where *self* is the new instance and the\n   remaining arguments are the same as were passed to ``__new__()``.\n\n   If ``__new__()`` does not return an instance of *cls*, then the new\n   instance\'s ``__init__()`` method will not be invoked.\n\n   ``__new__()`` is intended mainly to allow subclasses of immutable\n   types (like int, str, or tuple) to customize instance creation.  It\n   is also commonly overridden in custom metaclasses in order to\n   customize class creation.\n\nobject.__init__(self[, ...])\n\n   Called when the instance is created.  The arguments are those\n   passed to the class constructor expression.  If a base class has an\n   ``__init__()`` method, the derived class\'s ``__init__()`` method,\n   if any, must explicitly call it to ensure proper initialization of\n   the base class part of the instance; for example:\n   ``BaseClass.__init__(self, [args...])``.  As a special constraint\n   on constructors, no value may be returned; doing so will cause a\n   ``TypeError`` to be raised at runtime.\n\nobject.__del__(self)\n\n   Called when the instance is about to be destroyed.  This is also\n   called a destructor.  If a base class has a ``__del__()`` method,\n   the derived class\'s ``__del__()`` method, if any, must explicitly\n   call it to ensure proper deletion of the base class part of the\n   instance.  Note that it is possible (though not recommended!) for\n   the ``__del__()`` method to postpone destruction of the instance by\n   creating a new reference to it.  It may then be called at a later\n   time when this new reference is deleted.  It is not guaranteed that\n   ``__del__()`` methods are called for objects that still exist when\n   the interpreter exits.\n\n   Note: ``del x`` doesn\'t directly call ``x.__del__()`` --- the former\n     decrements the reference count for ``x`` by one, and the latter\n     is only called when ``x``\'s reference count reaches zero.  Some\n     common situations that may prevent the reference count of an\n     object from going to zero include: circular references between\n     objects (e.g., a doubly-linked list or a tree data structure with\n     parent and child pointers); a reference to the object on the\n     stack frame of a function that caught an exception (the traceback\n     stored in ``sys.exc_info()[2]`` keeps the stack frame alive); or\n     a reference to the object on the stack frame that raised an\n     unhandled exception in interactive mode (the traceback stored in\n     ``sys.last_traceback`` keeps the stack frame alive).  The first\n     situation can only be remedied by explicitly breaking the cycles;\n     the latter two situations can be resolved by storing ``None`` in\n     ``sys.last_traceback``. Circular references which are garbage are\n     detected when the option cycle detector is enabled (it\'s on by\n     default), but can only be cleaned up if there are no Python-\n     level ``__del__()`` methods involved. Refer to the documentation\n     for the ``gc`` module for more information about how\n     ``__del__()`` methods are handled by the cycle detector,\n     particularly the description of the ``garbage`` value.\n\n   Warning: Due to the precarious circumstances under which ``__del__()``\n     methods are invoked, exceptions that occur during their execution\n     are ignored, and a warning is printed to ``sys.stderr`` instead.\n     Also, when ``__del__()`` is invoked in response to a module being\n     deleted (e.g., when execution of the program is done), other\n     globals referenced by the ``__del__()`` method may already have\n     been deleted or in the process of being torn down (e.g. the\n     import machinery shutting down).  For this reason, ``__del__()``\n     methods should do the absolute minimum needed to maintain\n     external invariants.  Starting with version 1.5, Python\n     guarantees that globals whose name begins with a single\n     underscore are deleted from their module before other globals are\n     deleted; if no other references to such globals exist, this may\n     help in assuring that imported modules are still available at the\n     time when the ``__del__()`` method is called.\n\nobject.__repr__(self)\n\n   Called by the ``repr()`` built-in function to compute the\n   "official" string representation of an object.  If at all possible,\n   this should look like a valid Python expression that could be used\n   to recreate an object with the same value (given an appropriate\n   environment).  If this is not possible, a string of the form\n   ``<...some useful description...>`` should be returned. The return\n   value must be a string object. If a class defines ``__repr__()``\n   but not ``__str__()``, then ``__repr__()`` is also used when an\n   "informal" string representation of instances of that class is\n   required.\n\n   This is typically used for debugging, so it is important that the\n   representation is information-rich and unambiguous.\n\nobject.__str__(self)\n\n   Called by the ``str()`` built-in function and by the ``print()``\n   function to compute the "informal" string representation of an\n   object.  This differs from ``__repr__()`` in that it does not have\n   to be a valid Python expression: a more convenient or concise\n   representation may be used instead. The return value must be a\n   string object.\n\nobject.__format__(self, format_spec)\n\n   Called by the ``format()`` built-in function (and by extension, the\n   ``format()`` method of class ``str``) to produce a "formatted"\n   string representation of an object. The ``format_spec`` argument is\n   a string that contains a description of the formatting options\n   desired. The interpretation of the ``format_spec`` argument is up\n   to the type implementing ``__format__()``, however most classes\n   will either delegate formatting to one of the built-in types, or\n   use a similar formatting option syntax.\n\n   See *Format Specification Mini-Language* for a description of the\n   standard formatting syntax.\n\n   The return value must be a string object.\n\nobject.__lt__(self, other)\nobject.__le__(self, other)\nobject.__eq__(self, other)\nobject.__ne__(self, other)\nobject.__gt__(self, other)\nobject.__ge__(self, other)\n\n   These are the so-called "rich comparison" methods. The\n   correspondence between operator symbols and method names is as\n   follows: ``x<y`` calls ``x.__lt__(y)``, ``x<=y`` calls\n   ``x.__le__(y)``, ``x==y`` calls ``x.__eq__(y)``, ``x!=y`` calls\n   ``x.__ne__(y)``, ``x>y`` calls ``x.__gt__(y)``, and ``x>=y`` calls\n   ``x.__ge__(y)``.\n\n   A rich comparison method may return the singleton\n   ``NotImplemented`` if it does not implement the operation for a\n   given pair of arguments. By convention, ``False`` and ``True`` are\n   returned for a successful comparison. However, these methods can\n   return any value, so if the comparison operator is used in a\n   Boolean context (e.g., in the condition of an ``if`` statement),\n   Python will call ``bool()`` on the value to determine if the result\n   is true or false.\n\n   There are no implied relationships among the comparison operators.\n   The truth of ``x==y`` does not imply that ``x!=y`` is false.\n   Accordingly, when defining ``__eq__()``, one should also define\n   ``__ne__()`` so that the operators will behave as expected.  See\n   the paragraph on ``__hash__()`` for some important notes on\n   creating *hashable* objects which support custom comparison\n   operations and are usable as dictionary keys.\n\n   There are no swapped-argument versions of these methods (to be used\n   when the left argument does not support the operation but the right\n   argument does); rather, ``__lt__()`` and ``__gt__()`` are each\n   other\'s reflection, ``__le__()`` and ``__ge__()`` are each other\'s\n   reflection, and ``__eq__()`` and ``__ne__()`` are their own\n   reflection.\n\n   Arguments to rich comparison methods are never coerced.\n\n   To automatically generate ordering operations from a single root\n   operation, see ``functools.total_ordering()``.\n\nobject.__hash__(self)\n\n   Called by built-in function ``hash()`` and for operations on\n   members of hashed collections including ``set``, ``frozenset``, and\n   ``dict``.  ``__hash__()`` should return an integer.  The only\n   required property is that objects which compare equal have the same\n   hash value; it is advised to somehow mix together (e.g. using\n   exclusive or) the hash values for the components of the object that\n   also play a part in comparison of objects.\n\n   If a class does not define an ``__eq__()`` method it should not\n   define a ``__hash__()`` operation either; if it defines\n   ``__eq__()`` but not ``__hash__()``, its instances will not be\n   usable as items in hashable collections.  If a class defines\n   mutable objects and implements an ``__eq__()`` method, it should\n   not implement ``__hash__()``, since the implementation of hashable\n   collections requires that a key\'s hash value is immutable (if the\n   object\'s hash value changes, it will be in the wrong hash bucket).\n\n   User-defined classes have ``__eq__()`` and ``__hash__()`` methods\n   by default; with them, all objects compare unequal (except with\n   themselves) and ``x.__hash__()`` returns ``id(x)``.\n\n   Classes which inherit a ``__hash__()`` method from a parent class\n   but change the meaning of ``__eq__()`` such that the hash value\n   returned is no longer appropriate (e.g. by switching to a value-\n   based concept of equality instead of the default identity based\n   equality) can explicitly flag themselves as being unhashable by\n   setting ``__hash__ = None`` in the class definition. Doing so means\n   that not only will instances of the class raise an appropriate\n   ``TypeError`` when a program attempts to retrieve their hash value,\n   but they will also be correctly identified as unhashable when\n   checking ``isinstance(obj, collections.Hashable)`` (unlike classes\n   which define their own ``__hash__()`` to explicitly raise\n   ``TypeError``).\n\n   If a class that overrides ``__eq__()`` needs to retain the\n   implementation of ``__hash__()`` from a parent class, the\n   interpreter must be told this explicitly by setting ``__hash__ =\n   <ParentClass>.__hash__``. Otherwise the inheritance of\n   ``__hash__()`` will be blocked, just as if ``__hash__`` had been\n   explicitly set to ``None``.\n\nobject.__bool__(self)\n\n   Called to implement truth value testing and the built-in operation\n   ``bool()``; should return ``False`` or ``True``.  When this method\n   is not defined, ``__len__()`` is called, if it is defined, and the\n   object is considered true if its result is nonzero.  If a class\n   defines neither ``__len__()`` nor ``__bool__()``, all its instances\n   are considered true.\n',
+ 'debugger': '\n``pdb`` --- The Python Debugger\n*******************************\n\nThe module ``pdb`` defines an interactive source code debugger for\nPython programs.  It supports setting (conditional) breakpoints and\nsingle stepping at the source line level, inspection of stack frames,\nsource code listing, and evaluation of arbitrary Python code in the\ncontext of any stack frame.  It also supports post-mortem debugging\nand can be called under program control.\n\nThe debugger is extensible -- it is actually defined as the class\n``Pdb``. This is currently undocumented but easily understood by\nreading the source.  The extension interface uses the modules ``bdb``\nand ``cmd``.\n\nThe debugger\'s prompt is ``(Pdb)``. Typical usage to run a program\nunder control of the debugger is:\n\n   >>> import pdb\n   >>> import mymodule\n   >>> pdb.run(\'mymodule.test()\')\n   > <string>(0)?()\n   (Pdb) continue\n   > <string>(1)?()\n   (Pdb) continue\n   NameError: \'spam\'\n   > <string>(1)?()\n   (Pdb)\n\n``pdb.py`` can also be invoked as a script to debug other scripts.\nFor example:\n\n   python3 -m pdb myscript.py\n\nWhen invoked as a script, pdb will automatically enter post-mortem\ndebugging if the program being debugged exits abnormally.  After post-\nmortem debugging (or after normal exit of the program), pdb will\nrestart the program.  Automatic restarting preserves pdb\'s state (such\nas breakpoints) and in most cases is more useful than quitting the\ndebugger upon program\'s exit.\n\nNew in version 3.2: ``pdb.py`` now accepts a ``-c`` option that\nexecutes commands as if given in a ``.pdbrc`` file, see *Debugger\nCommands*.\n\nThe typical usage to break into the debugger from a running program is\nto insert\n\n   import pdb; pdb.set_trace()\n\nat the location you want to break into the debugger.  You can then\nstep through the code following this statement, and continue running\nwithout the debugger using the ``continue`` command.\n\nThe typical usage to inspect a crashed program is:\n\n   >>> import pdb\n   >>> import mymodule\n   >>> mymodule.test()\n   Traceback (most recent call last):\n     File "<stdin>", line 1, in ?\n     File "./mymodule.py", line 4, in test\n       test2()\n     File "./mymodule.py", line 3, in test2\n       print(spam)\n   NameError: spam\n   >>> pdb.pm()\n   > ./mymodule.py(3)test2()\n   -> print(spam)\n   (Pdb)\n\nThe module defines the following functions; each enters the debugger\nin a slightly different way:\n\npdb.run(statement, globals=None, locals=None)\n\n   Execute the *statement* (given as a string or a code object) under\n   debugger control.  The debugger prompt appears before any code is\n   executed; you can set breakpoints and type ``continue``, or you can\n   step through the statement using ``step`` or ``next`` (all these\n   commands are explained below).  The optional *globals* and *locals*\n   arguments specify the environment in which the code is executed; by\n   default the dictionary of the module ``__main__`` is used.  (See\n   the explanation of the built-in ``exec()`` or ``eval()``\n   functions.)\n\npdb.runeval(expression, globals=None, locals=None)\n\n   Evaluate the *expression* (given as a string or a code object)\n   under debugger control.  When ``runeval()`` returns, it returns the\n   value of the expression.  Otherwise this function is similar to\n   ``run()``.\n\npdb.runcall(function, *args, **kwds)\n\n   Call the *function* (a function or method object, not a string)\n   with the given arguments.  When ``runcall()`` returns, it returns\n   whatever the function call returned.  The debugger prompt appears\n   as soon as the function is entered.\n\npdb.set_trace()\n\n   Enter the debugger at the calling stack frame.  This is useful to\n   hard-code a breakpoint at a given point in a program, even if the\n   code is not otherwise being debugged (e.g. when an assertion\n   fails).\n\npdb.post_mortem(traceback=None)\n\n   Enter post-mortem debugging of the given *traceback* object.  If no\n   *traceback* is given, it uses the one of the exception that is\n   currently being handled (an exception must be being handled if the\n   default is to be used).\n\npdb.pm()\n\n   Enter post-mortem debugging of the traceback found in\n   ``sys.last_traceback``.\n\nThe ``run_*`` functions and ``set_trace()`` are aliases for\ninstantiating the ``Pdb`` class and calling the method of the same\nname.  If you want to access further features, you have to do this\nyourself:\n\nclass class pdb.Pdb(completekey=\'tab\', stdin=None, stdout=None, skip=None)\n\n   ``Pdb`` is the debugger class.\n\n   The *completekey*, *stdin* and *stdout* arguments are passed to the\n   underlying ``cmd.Cmd`` class; see the description there.\n\n   The *skip* argument, if given, must be an iterable of glob-style\n   module name patterns.  The debugger will not step into frames that\n   originate in a module that matches one of these patterns. [1]\n\n   Example call to enable tracing with *skip*:\n\n      import pdb; pdb.Pdb(skip=[\'django.*\']).set_trace()\n\n   New in version 3.1: The *skip* argument.\n\n   run(statement, globals=None, locals=None)\n   runeval(expression, globals=None, locals=None)\n   runcall(function, *args, **kwds)\n   set_trace()\n\n      See the documentation for the functions explained above.\n',
  'del': '\nThe ``del`` statement\n*********************\n\n   del_stmt ::= "del" target_list\n\nDeletion is recursively defined very similar to the way assignment is\ndefined. Rather that spelling it out in full details, here are some\nhints.\n\nDeletion of a target list recursively deletes each target, from left\nto right.\n\nDeletion of a name removes the binding of that name from the local or\nglobal namespace, depending on whether the name occurs in a ``global``\nstatement in the same code block.  If the name is unbound, a\n``NameError`` exception will be raised.\n\nIt is illegal to delete a name from the local namespace if it occurs\nas a free variable in a nested block.\n\nDeletion of attribute references, subscriptions and slicings is passed\nto the primary object involved; deletion of a slicing is in general\nequivalent to assignment of an empty slice of the right type (but even\nthis is determined by the sliced object).\n',
  'dict': '\nDictionary displays\n*******************\n\nA dictionary display is a possibly empty series of key/datum pairs\nenclosed in curly braces:\n\n   dict_display       ::= "{" [key_datum_list | dict_comprehension] "}"\n   key_datum_list     ::= key_datum ("," key_datum)* [","]\n   key_datum          ::= expression ":" expression\n   dict_comprehension ::= expression ":" expression comp_for\n\nA dictionary display yields a new dictionary object.\n\nIf a comma-separated sequence of key/datum pairs is given, they are\nevaluated from left to right to define the entries of the dictionary:\neach key object is used as a key into the dictionary to store the\ncorresponding datum.  This means that you can specify the same key\nmultiple times in the key/datum list, and the final dictionary\'s value\nfor that key will be the last one given.\n\nA dict comprehension, in contrast to list and set comprehensions,\nneeds two expressions separated with a colon followed by the usual\n"for" and "if" clauses. When the comprehension is run, the resulting\nkey and value elements are inserted in the new dictionary in the order\nthey are produced.\n\nRestrictions on the types of the key values are listed earlier in\nsection *The standard type hierarchy*.  (To summarize, the key type\nshould be *hashable*, which excludes all mutable objects.)  Clashes\nbetween duplicate keys are not detected; the last datum (textually\nrightmost in the display) stored for a given key value prevails.\n',
  'dynamic-features': '\nInteraction with dynamic features\n*********************************\n\nThere are several cases where Python statements are illegal when used\nin conjunction with nested scopes that contain free variables.\n\nIf a variable is referenced in an enclosing scope, it is illegal to\ndelete the name.  An error will be reported at compile time.\n\nIf the wild card form of import --- ``import *`` --- is used in a\nfunction and the function contains or is a nested block with free\nvariables, the compiler will raise a ``SyntaxError``.\n\nThe ``eval()`` and ``exec()`` functions do not have access to the full\nenvironment for resolving names.  Names may be resolved in the local\nand global namespaces of the caller.  Free variables are not resolved\nin the nearest enclosing namespace, but in the global namespace.  [1]\nThe ``exec()`` and ``eval()`` functions have optional arguments to\noverride the global and local namespace.  If only one namespace is\nspecified, it is used for both.\n',
  'else': '\nThe ``if`` statement\n********************\n\nThe ``if`` statement is used for conditional execution:\n\n   if_stmt ::= "if" expression ":" suite\n               ( "elif" expression ":" suite )*\n               ["else" ":" suite]\n\nIt selects exactly one of the suites by evaluating the expressions one\nby one until one is found to be true (see section *Boolean operations*\nfor the definition of true and false); then that suite is executed\n(and no other part of the ``if`` statement is executed or evaluated).\nIf all expressions are false, the suite of the ``else`` clause, if\npresent, is executed.\n',
  'exceptions': '\nExceptions\n**********\n\nExceptions are a means of breaking out of the normal flow of control\nof a code block in order to handle errors or other exceptional\nconditions.  An exception is *raised* at the point where the error is\ndetected; it may be *handled* by the surrounding code block or by any\ncode block that directly or indirectly invoked the code block where\nthe error occurred.\n\nThe Python interpreter raises an exception when it detects a run-time\nerror (such as division by zero).  A Python program can also\nexplicitly raise an exception with the ``raise`` statement. Exception\nhandlers are specified with the ``try`` ... ``except`` statement.  The\n``finally`` clause of such a statement can be used to specify cleanup\ncode which does not handle the exception, but is executed whether an\nexception occurred or not in the preceding code.\n\nPython uses the "termination" model of error handling: an exception\nhandler can find out what happened and continue execution at an outer\nlevel, but it cannot repair the cause of the error and retry the\nfailing operation (except by re-entering the offending piece of code\nfrom the top).\n\nWhen an exception is not handled at all, the interpreter terminates\nexecution of the program, or returns to its interactive main loop.  In\neither case, it prints a stack backtrace, except when the exception is\n``SystemExit``.\n\nExceptions are identified by class instances.  The ``except`` clause\nis selected depending on the class of the instance: it must reference\nthe class of the instance or a base class thereof.  The instance can\nbe received by the handler and can carry additional information about\nthe exceptional condition.\n\nNote: Exception messages are not part of the Python API.  Their contents\n  may change from one version of Python to the next without warning\n  and should not be relied on by code which will run under multiple\n  versions of the interpreter.\n\nSee also the description of the ``try`` statement in section *The try\nstatement* and ``raise`` statement in section *The raise statement*.\n\n-[ Footnotes ]-\n\n[1] This limitation occurs because the code that is executed by these\n    operations is not available at the time the module is compiled.\n',
- 'execmodel': '\nExecution model\n***************\n\n\nNaming and binding\n==================\n\n*Names* refer to objects.  Names are introduced by name binding\noperations. Each occurrence of a name in the program text refers to\nthe *binding* of that name established in the innermost function block\ncontaining the use.\n\nA *block* is a piece of Python program text that is executed as a\nunit. The following are blocks: a module, a function body, and a class\ndefinition. Each command typed interactively is a block.  A script\nfile (a file given as standard input to the interpreter or specified\non the interpreter command line the first argument) is a code block.\nA script command (a command specified on the interpreter command line\nwith the \'**-c**\' option) is a code block.  The string argument passed\nto the built-in functions ``eval()`` and ``exec()`` is a code block.\n\nA code block is executed in an *execution frame*.  A frame contains\nsome administrative information (used for debugging) and determines\nwhere and how execution continues after the code block\'s execution has\ncompleted.\n\nA *scope* defines the visibility of a name within a block.  If a local\nvariable is defined in a block, its scope includes that block.  If the\ndefinition occurs in a function block, the scope extends to any blocks\ncontained within the defining one, unless a contained block introduces\na different binding for the name.  The scope of names defined in a\nclass block is limited to the class block; it does not extend to the\ncode blocks of methods -- this includes comprehensions and generator\nexpressions since they are implemented using a function scope.  This\nmeans that the following will fail:\n\n   class A:\n       a = 42\n       b = list(a + i for i in range(10))\n\nWhen a name is used in a code block, it is resolved using the nearest\nenclosing scope.  The set of all such scopes visible to a code block\nis called the block\'s *environment*.\n\nIf a name is bound in a block, it is a local variable of that block,\nunless declared as ``nonlocal``.  If a name is bound at the module\nlevel, it is a global variable.  (The variables of the module code\nblock are local and global.)  If a variable is used in a code block\nbut not defined there, it is a *free variable*.\n\nWhen a name is not found at all, a ``NameError`` exception is raised.\nIf the name refers to a local variable that has not been bound, a\n``UnboundLocalError`` exception is raised.  ``UnboundLocalError`` is a\nsubclass of ``NameError``.\n\nThe following constructs bind names: formal parameters to functions,\n``import`` statements, class and function definitions (these bind the\nclass or function name in the defining block), and targets that are\nidentifiers if occurring in an assignment, ``for`` loop header, or\nafter ``as`` in a ``with`` statement or :keyword.`except` clause. The\n``import`` statement of the form ``from ... import *`` binds all names\ndefined in the imported module, except those beginning with an\nunderscore.  This form may only be used at the module level.\n\nA target occurring in a ``del`` statement is also considered bound for\nthis purpose (though the actual semantics are to unbind the name).  It\nis illegal to unbind a name that is referenced by an enclosing scope;\nthe compiler will report a ``SyntaxError``.\n\nEach assignment or import statement occurs within a block defined by a\nclass or function definition or at the module level (the top-level\ncode block).\n\nIf a name binding operation occurs anywhere within a code block, all\nuses of the name within the block are treated as references to the\ncurrent block.  This can lead to errors when a name is used within a\nblock before it is bound.  This rule is subtle.  Python lacks\ndeclarations and allows name binding operations to occur anywhere\nwithin a code block.  The local variables of a code block can be\ndetermined by scanning the entire text of the block for name binding\noperations.\n\nIf the ``global`` statement occurs within a block, all uses of the\nname specified in the statement refer to the binding of that name in\nthe top-level namespace.  Names are resolved in the top-level\nnamespace by searching the global namespace, i.e. the namespace of the\nmodule containing the code block, and the builtin namespace, the\nnamespace of the module ``builtins``.  The global namespace is\nsearched first.  If the name is not found there, the builtin namespace\nis searched.  The global statement must precede all uses of the name.\n\nThe built-in namespace associated with the execution of a code block\nis actually found by looking up the name ``__builtins__`` in its\nglobal namespace; this should be a dictionary or a module (in the\nlatter case the module\'s dictionary is used).  By default, when in the\n``__main__`` module, ``__builtins__`` is the built-in module\n``builtins``; when in any other module, ``__builtins__`` is an alias\nfor the dictionary of the ``builtins`` module itself.\n``__builtins__`` can be set to a user-created dictionary to create a\nweak form of restricted execution.\n\nNote: Users should not touch ``__builtins__``; it is strictly an\n  implementation detail.  Users wanting to override values in the\n  built-in namespace should ``import`` the ``builtins`` module and\n  modify its attributes appropriately.\n\nThe namespace for a module is automatically created the first time a\nmodule is imported.  The main module for a script is always called\n``__main__``.\n\nThe global statement has the same scope as a name binding operation in\nthe same block.  If the nearest enclosing scope for a free variable\ncontains a global statement, the free variable is treated as a global.\n\nA class definition is an executable statement that may use and define\nnames. These references follow the normal rules for name resolution.\nThe namespace of the class definition becomes the attribute dictionary\nof the class.  Names defined at the class scope are not visible in\nmethods.\n\n\nInteraction with dynamic features\n---------------------------------\n\nThere are several cases where Python statements are illegal when used\nin conjunction with nested scopes that contain free variables.\n\nIf a variable is referenced in an enclosing scope, it is illegal to\ndelete the name.  An error will be reported at compile time.\n\nIf the wild card form of import --- ``import *`` --- is used in a\nfunction and the function contains or is a nested block with free\nvariables, the compiler will raise a ``SyntaxError``.\n\nThe ``eval()`` and ``exec()`` functions do not have access to the full\nenvironment for resolving names.  Names may be resolved in the local\nand global namespaces of the caller.  Free variables are not resolved\nin the nearest enclosing namespace, but in the global namespace.  [1]\nThe ``exec()`` and ``eval()`` functions have optional arguments to\noverride the global and local namespace.  If only one namespace is\nspecified, it is used for both.\n\n\nExceptions\n==========\n\nExceptions are a means of breaking out of the normal flow of control\nof a code block in order to handle errors or other exceptional\nconditions.  An exception is *raised* at the point where the error is\ndetected; it may be *handled* by the surrounding code block or by any\ncode block that directly or indirectly invoked the code block where\nthe error occurred.\n\nThe Python interpreter raises an exception when it detects a run-time\nerror (such as division by zero).  A Python program can also\nexplicitly raise an exception with the ``raise`` statement. Exception\nhandlers are specified with the ``try`` ... ``except`` statement.  The\n``finally`` clause of such a statement can be used to specify cleanup\ncode which does not handle the exception, but is executed whether an\nexception occurred or not in the preceding code.\n\nPython uses the "termination" model of error handling: an exception\nhandler can find out what happened and continue execution at an outer\nlevel, but it cannot repair the cause of the error and retry the\nfailing operation (except by re-entering the offending piece of code\nfrom the top).\n\nWhen an exception is not handled at all, the interpreter terminates\nexecution of the program, or returns to its interactive main loop.  In\neither case, it prints a stack backtrace, except when the exception is\n``SystemExit``.\n\nExceptions are identified by class instances.  The ``except`` clause\nis selected depending on the class of the instance: it must reference\nthe class of the instance or a base class thereof.  The instance can\nbe received by the handler and can carry additional information about\nthe exceptional condition.\n\nNote: Exception messages are not part of the Python API.  Their contents\n  may change from one version of Python to the next without warning\n  and should not be relied on by code which will run under multiple\n  versions of the interpreter.\n\nSee also the description of the ``try`` statement in section *The try\nstatement* and ``raise`` statement in section *The raise statement*.\n\n-[ Footnotes ]-\n\n[1] This limitation occurs because the code that is executed by these\n    operations is not available at the time the module is compiled.\n',
+ 'execmodel': '\nExecution model\n***************\n\n\nNaming and binding\n==================\n\n*Names* refer to objects.  Names are introduced by name binding\noperations. Each occurrence of a name in the program text refers to\nthe *binding* of that name established in the innermost function block\ncontaining the use.\n\nA *block* is a piece of Python program text that is executed as a\nunit. The following are blocks: a module, a function body, and a class\ndefinition. Each command typed interactively is a block.  A script\nfile (a file given as standard input to the interpreter or specified\non the interpreter command line the first argument) is a code block.\nA script command (a command specified on the interpreter command line\nwith the \'**-c**\' option) is a code block.  The string argument passed\nto the built-in functions ``eval()`` and ``exec()`` is a code block.\n\nA code block is executed in an *execution frame*.  A frame contains\nsome administrative information (used for debugging) and determines\nwhere and how execution continues after the code block\'s execution has\ncompleted.\n\nA *scope* defines the visibility of a name within a block.  If a local\nvariable is defined in a block, its scope includes that block.  If the\ndefinition occurs in a function block, the scope extends to any blocks\ncontained within the defining one, unless a contained block introduces\na different binding for the name.  The scope of names defined in a\nclass block is limited to the class block; it does not extend to the\ncode blocks of methods -- this includes comprehensions and generator\nexpressions since they are implemented using a function scope.  This\nmeans that the following will fail:\n\n   class A:\n       a = 42\n       b = list(a + i for i in range(10))\n\nWhen a name is used in a code block, it is resolved using the nearest\nenclosing scope.  The set of all such scopes visible to a code block\nis called the block\'s *environment*.\n\nIf a name is bound in a block, it is a local variable of that block,\nunless declared as ``nonlocal``.  If a name is bound at the module\nlevel, it is a global variable.  (The variables of the module code\nblock are local and global.)  If a variable is used in a code block\nbut not defined there, it is a *free variable*.\n\nWhen a name is not found at all, a ``NameError`` exception is raised.\nIf the name refers to a local variable that has not been bound, a\n``UnboundLocalError`` exception is raised.  ``UnboundLocalError`` is a\nsubclass of ``NameError``.\n\nThe following constructs bind names: formal parameters to functions,\n``import`` statements, class and function definitions (these bind the\nclass or function name in the defining block), and targets that are\nidentifiers if occurring in an assignment, ``for`` loop header, or\nafter ``as`` in a ``with`` statement or :keyword.`except` clause. The\n``import`` statement of the form ``from ... import *`` binds all names\ndefined in the imported module, except those beginning with an\nunderscore.  This form may only be used at the module level.\n\nA target occurring in a ``del`` statement is also considered bound for\nthis purpose (though the actual semantics are to unbind the name).  It\nis illegal to unbind a name that is referenced by an enclosing scope;\nthe compiler will report a ``SyntaxError``.\n\nEach assignment or import statement occurs within a block defined by a\nclass or function definition or at the module level (the top-level\ncode block).\n\nIf a name binding operation occurs anywhere within a code block, all\nuses of the name within the block are treated as references to the\ncurrent block.  This can lead to errors when a name is used within a\nblock before it is bound.  This rule is subtle.  Python lacks\ndeclarations and allows name binding operations to occur anywhere\nwithin a code block.  The local variables of a code block can be\ndetermined by scanning the entire text of the block for name binding\noperations.\n\nIf the ``global`` statement occurs within a block, all uses of the\nname specified in the statement refer to the binding of that name in\nthe top-level namespace.  Names are resolved in the top-level\nnamespace by searching the global namespace, i.e. the namespace of the\nmodule containing the code block, and the builtins namespace, the\nnamespace of the module ``builtins``.  The global namespace is\nsearched first.  If the name is not found there, the builtins\nnamespace is searched.  The global statement must precede all uses of\nthe name.\n\nThe builtins namespace associated with the execution of a code block\nis actually found by looking up the name ``__builtins__`` in its\nglobal namespace; this should be a dictionary or a module (in the\nlatter case the module\'s dictionary is used).  By default, when in the\n``__main__`` module, ``__builtins__`` is the built-in module\n``builtins``; when in any other module, ``__builtins__`` is an alias\nfor the dictionary of the ``builtins`` module itself.\n``__builtins__`` can be set to a user-created dictionary to create a\nweak form of restricted execution.\n\n**CPython implementation detail:** Users should not touch\n``__builtins__``; it is strictly an implementation detail.  Users\nwanting to override values in the builtins namespace should ``import``\nthe ``builtins`` module and modify its attributes appropriately.\n\nThe namespace for a module is automatically created the first time a\nmodule is imported.  The main module for a script is always called\n``__main__``.\n\nThe global statement has the same scope as a name binding operation in\nthe same block.  If the nearest enclosing scope for a free variable\ncontains a global statement, the free variable is treated as a global.\n\nA class definition is an executable statement that may use and define\nnames. These references follow the normal rules for name resolution.\nThe namespace of the class definition becomes the attribute dictionary\nof the class.  Names defined at the class scope are not visible in\nmethods.\n\n\nInteraction with dynamic features\n---------------------------------\n\nThere are several cases where Python statements are illegal when used\nin conjunction with nested scopes that contain free variables.\n\nIf a variable is referenced in an enclosing scope, it is illegal to\ndelete the name.  An error will be reported at compile time.\n\nIf the wild card form of import --- ``import *`` --- is used in a\nfunction and the function contains or is a nested block with free\nvariables, the compiler will raise a ``SyntaxError``.\n\nThe ``eval()`` and ``exec()`` functions do not have access to the full\nenvironment for resolving names.  Names may be resolved in the local\nand global namespaces of the caller.  Free variables are not resolved\nin the nearest enclosing namespace, but in the global namespace.  [1]\nThe ``exec()`` and ``eval()`` functions have optional arguments to\noverride the global and local namespace.  If only one namespace is\nspecified, it is used for both.\n\n\nExceptions\n==========\n\nExceptions are a means of breaking out of the normal flow of control\nof a code block in order to handle errors or other exceptional\nconditions.  An exception is *raised* at the point where the error is\ndetected; it may be *handled* by the surrounding code block or by any\ncode block that directly or indirectly invoked the code block where\nthe error occurred.\n\nThe Python interpreter raises an exception when it detects a run-time\nerror (such as division by zero).  A Python program can also\nexplicitly raise an exception with the ``raise`` statement. Exception\nhandlers are specified with the ``try`` ... ``except`` statement.  The\n``finally`` clause of such a statement can be used to specify cleanup\ncode which does not handle the exception, but is executed whether an\nexception occurred or not in the preceding code.\n\nPython uses the "termination" model of error handling: an exception\nhandler can find out what happened and continue execution at an outer\nlevel, but it cannot repair the cause of the error and retry the\nfailing operation (except by re-entering the offending piece of code\nfrom the top).\n\nWhen an exception is not handled at all, the interpreter terminates\nexecution of the program, or returns to its interactive main loop.  In\neither case, it prints a stack backtrace, except when the exception is\n``SystemExit``.\n\nExceptions are identified by class instances.  The ``except`` clause\nis selected depending on the class of the instance: it must reference\nthe class of the instance or a base class thereof.  The instance can\nbe received by the handler and can carry additional information about\nthe exceptional condition.\n\nNote: Exception messages are not part of the Python API.  Their contents\n  may change from one version of Python to the next without warning\n  and should not be relied on by code which will run under multiple\n  versions of the interpreter.\n\nSee also the description of the ``try`` statement in section *The try\nstatement* and ``raise`` statement in section *The raise statement*.\n\n-[ Footnotes ]-\n\n[1] This limitation occurs because the code that is executed by these\n    operations is not available at the time the module is compiled.\n',
  'exprlists': '\nExpression lists\n****************\n\n   expression_list ::= expression ( "," expression )* [","]\n\nAn expression list containing at least one comma yields a tuple.  The\nlength of the tuple is the number of expressions in the list.  The\nexpressions are evaluated from left to right.\n\nThe trailing comma is required only to create a single tuple (a.k.a. a\n*singleton*); it is optional in all other cases.  A single expression\nwithout a trailing comma doesn\'t create a tuple, but rather yields the\nvalue of that expression. (To create an empty tuple, use an empty pair\nof parentheses: ``()``.)\n',
  'floating': '\nFloating point literals\n***********************\n\nFloating point literals are described by the following lexical\ndefinitions:\n\n   floatnumber   ::= pointfloat | exponentfloat\n   pointfloat    ::= [intpart] fraction | intpart "."\n   exponentfloat ::= (intpart | pointfloat) exponent\n   intpart       ::= digit+\n   fraction      ::= "." digit+\n   exponent      ::= ("e" | "E") ["+" | "-"] digit+\n\nNote that the integer and exponent parts are always interpreted using\nradix 10. For example, ``077e010`` is legal, and denotes the same\nnumber as ``77e10``. The allowed range of floating point literals is\nimplementation-dependent. Some examples of floating point literals:\n\n   3.14    10.    .001    1e100    3.14e-10    0e0\n\nNote that numeric literals do not include a sign; a phrase like ``-1``\nis actually an expression composed of the unary operator ``-`` and the\nliteral ``1``.\n',
  'for': '\nThe ``for`` statement\n*********************\n\nThe ``for`` statement is used to iterate over the elements of a\nsequence (such as a string, tuple or list) or other iterable object:\n\n   for_stmt ::= "for" target_list "in" expression_list ":" suite\n                ["else" ":" suite]\n\nThe expression list is evaluated once; it should yield an iterable\nobject.  An iterator is created for the result of the\n``expression_list``.  The suite is then executed once for each item\nprovided by the iterator, in the order of ascending indices.  Each\nitem in turn is assigned to the target list using the standard rules\nfor assignments (see *Assignment statements*), and then the suite is\nexecuted.  When the items are exhausted (which is immediately when the\nsequence is empty or an iterator raises a ``StopIteration``\nexception), the suite in the ``else`` clause, if present, is executed,\nand the loop terminates.\n\nA ``break`` statement executed in the first suite terminates the loop\nwithout executing the ``else`` clause\'s suite.  A ``continue``\nstatement executed in the first suite skips the rest of the suite and\ncontinues with the next item, or with the ``else`` clause if there was\nno next item.\n\nThe suite may assign to the variable(s) in the target list; this does\nnot affect the next item assigned to it.\n\nNames in the target list are not deleted when the loop is finished,\nbut if the sequence is empty, it will not have been assigned to at all\nby the loop.  Hint: the built-in function ``range()`` returns an\niterator of integers suitable to emulate the effect of Pascal\'s ``for\ni := a to b do``; e.g., ``list(range(3))`` returns the list ``[0, 1,\n2]``.\n\nNote: There is a subtlety when the sequence is being modified by the loop\n  (this can only occur for mutable sequences, i.e. lists).  An\n  internal counter is used to keep track of which item is used next,\n  and this is incremented on each iteration.  When this counter has\n  reached the length of the sequence the loop terminates.  This means\n  that if the suite deletes the current (or a previous) item from the\n  sequence, the next item will be skipped (since it gets the index of\n  the current item which has already been treated).  Likewise, if the\n  suite inserts an item in the sequence before the current item, the\n  current item will be treated again the next time through the loop.\n  This can lead to nasty bugs that can be avoided by making a\n  temporary copy using a slice of the whole sequence, e.g.,\n\n     for x in a[:]:\n         if x < 0: a.remove(x)\n',
- 'formatstrings': '\nFormat String Syntax\n********************\n\nThe ``str.format()`` method and the ``Formatter`` class share the same\nsyntax for format strings (although in the case of ``Formatter``,\nsubclasses can define their own format string syntax.)\n\nFormat strings contain "replacement fields" surrounded by curly braces\n``{}``. Anything that is not contained in braces is considered literal\ntext, which is copied unchanged to the output.  If you need to include\na brace character in the literal text, it can be escaped by doubling:\n``{{`` and ``}}``.\n\nThe grammar for a replacement field is as follows:\n\n      replacement_field ::= "{" field_name ["!" conversion] [":" format_spec] "}"\n      field_name        ::= arg_name ("." attribute_name | "[" element_index "]")*\n      arg_name          ::= (identifier | integer)?\n      attribute_name    ::= identifier\n      element_index     ::= integer\n      conversion        ::= "r" | "s" | "a"\n      format_spec       ::= <described in the next section>\n\nIn less formal terms, the replacement field starts with a *field_name*\nthat specifies the object whose value is to be formatted and inserted\ninto the output instead of the replacement field. The *field_name* is\noptionally followed by a  *conversion* field, which is preceded by an\nexclamation point ``\'!\'``, and a *format_spec*, which is preceded by a\ncolon ``\':\'``.  These specify a non-default format for the replacement\nvalue.\n\nThe *field_name* itself begins with an *arg_name* that is either\neither a number or a keyword.  If it\'s a number, it refers to a\npositional argument, and if it\'s a keyword, it refers to a named\nkeyword argument.  If the numerical arg_names in a format string are\n0, 1, 2, ... in sequence, they can all be omitted (not just some) and\nthe numbers 0, 1, 2, ... will be automatically inserted in that order.\nThe *arg_name* can be followed by any number of index or attribute\nexpressions. An expression of the form ``\'.name\'`` selects the named\nattribute using ``getattr()``, while an expression of the form\n``\'[index]\'`` does an index lookup using ``__getitem__()``.\n\nSome simple format string examples:\n\n   "First, thou shalt count to {0}" # References first positional argument\n   "Bring me a {}"                  # Implicitly references the first positional argument\n   "From {} to {}"                  # Same as "From {0] to {1}"\n   "My quest is {name}"             # References keyword argument \'name\'\n   "Weight in tons {0.weight}"      # \'weight\' attribute of first positional arg\n   "Units destroyed: {players[0]}"  # First element of keyword argument \'players\'.\n\nThe *conversion* field causes a type coercion before formatting.\nNormally, the job of formatting a value is done by the\n``__format__()`` method of the value itself.  However, in some cases\nit is desirable to force a type to be formatted as a string,\noverriding its own definition of formatting.  By converting the value\nto a string before calling ``__format__()``, the normal formatting\nlogic is bypassed.\n\nThree conversion flags are currently supported: ``\'!s\'`` which calls\n``str()`` on the value, ``\'!r\'`` which calls ``repr()`` and ``\'!a\'``\nwhich calls ``ascii()``.\n\nSome examples:\n\n   "Harold\'s a clever {0!s}"        # Calls str() on the argument first\n   "Bring out the holy {name!r}"    # Calls repr() on the argument first\n\nThe *format_spec* field contains a specification of how the value\nshould be presented, including such details as field width, alignment,\npadding, decimal precision and so on.  Each value type can define it\'s\nown "formatting mini-language" or interpretation of the *format_spec*.\n\nMost built-in types support a common formatting mini-language, which\nis described in the next section.\n\nA *format_spec* field can also include nested replacement fields\nwithin it. These nested replacement fields can contain only a field\nname; conversion flags and format specifications are not allowed.  The\nreplacement fields within the format_spec are substituted before the\n*format_spec* string is interpreted. This allows the formatting of a\nvalue to be dynamically specified.\n\nFor example, suppose you wanted to have a replacement field whose\nfield width is determined by another variable:\n\n   "A man with two {0:{1}}".format("noses", 10)\n\nThis would first evaluate the inner replacement field, making the\nformat string effectively:\n\n   "A man with two {0:10}"\n\nThen the outer replacement field would be evaluated, producing:\n\n   "noses     "\n\nWhich is substituted into the string, yielding:\n\n   "A man with two noses     "\n\n(The extra space is because we specified a field width of 10, and\nbecause left alignment is the default for strings.)\n\n\nFormat Specification Mini-Language\n==================================\n\n"Format specifications" are used within replacement fields contained\nwithin a format string to define how individual values are presented\n(see *Format String Syntax*.)  They can also be passed directly to the\nbuiltin ``format()`` function.  Each formattable type may define how\nthe format specification is to be interpreted.\n\nMost built-in types implement the following options for format\nspecifications, although some of the formatting options are only\nsupported by the numeric types.\n\nA general convention is that an empty format string (``""``) produces\nthe same result as if you had called ``str()`` on the value.\n\nThe general form of a *standard format specifier* is:\n\n   format_spec ::= [[fill]align][sign][#][0][width][.precision][type]\n   fill        ::= <a character other than \'}\'>\n   align       ::= "<" | ">" | "=" | "^"\n   sign        ::= "+" | "-" | " "\n   width       ::= integer\n   precision   ::= integer\n   type        ::= "b" | "c" | "d" | "e" | "E" | "f" | "F" | "g" | "G" | "n" | "o" | "x" | "X" | "%"\n\nThe *fill* character can be any character other than \'}\' (which\nsignifies the end of the field).  The presence of a fill character is\nsignaled by the *next* character, which must be one of the alignment\noptions. If the second character of *format_spec* is not a valid\nalignment option, then it is assumed that both the fill character and\nthe alignment option are absent.\n\nThe meaning of the various alignment options is as follows:\n\n   +-----------+------------------------------------------------------------+\n   | Option    | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'<\'``   | Forces the field to be left-aligned within the available   |\n   |           | space (This is the default.)                               |\n   +-----------+------------------------------------------------------------+\n   | ``\'>\'``   | Forces the field to be right-aligned within the available  |\n   |           | space.                                                     |\n   +-----------+------------------------------------------------------------+\n   | ``\'=\'``   | Forces the padding to be placed after the sign (if any)    |\n   |           | but before the digits.  This is used for printing fields   |\n   |           | in the form \'+000000120\'. This alignment option is only    |\n   |           | valid for numeric types.                                   |\n   +-----------+------------------------------------------------------------+\n   | ``\'^\'``   | Forces the field to be centered within the available       |\n   |           | space.                                                     |\n   +-----------+------------------------------------------------------------+\n\nNote that unless a minimum field width is defined, the field width\nwill always be the same size as the data to fill it, so that the\nalignment option has no meaning in this case.\n\nThe *sign* option is only valid for number types, and can be one of\nthe following:\n\n   +-----------+------------------------------------------------------------+\n   | Option    | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'+\'``   | indicates that a sign should be used for both positive as  |\n   |           | well as negative numbers.                                  |\n   +-----------+------------------------------------------------------------+\n   | ``\'-\'``   | indicates that a sign should be used only for negative     |\n   |           | numbers (this is the default behavior).                    |\n   +-----------+------------------------------------------------------------+\n   | space     | indicates that a leading space should be used on positive  |\n   |           | numbers, and a minus sign on negative numbers.             |\n   +-----------+------------------------------------------------------------+\n\nThe ``\'#\'`` option is only valid for integers, and only for binary,\noctal, or hexadecimal output.  If present, it specifies that the\noutput will be prefixed by ``\'0b\'``, ``\'0o\'``, or ``\'0x\'``,\nrespectively.\n\n*width* is a decimal integer defining the minimum field width.  If not\nspecified, then the field width will be determined by the content.\n\nIf the *width* field is preceded by a zero (``\'0\'``) character, this\nenables zero-padding.  This is equivalent to an *alignment* type of\n``\'=\'`` and a *fill* character of ``\'0\'``.\n\nThe *precision* is a decimal number indicating how many digits should\nbe displayed after the decimal point for a floating point value\nformatted with ``\'f\'`` and ``\'F\'``, or before and after the decimal\npoint for a floating point value formatted with ``\'g\'`` or ``\'G\'``.\nFor non-number types the field indicates the maximum field size - in\nother words, how many characters will be used from the field content.\nThe *precision* is not allowed for integer values.\n\nFinally, the *type* determines how the data should be presented.\n\nThe available integer presentation types are:\n\n   +-----------+------------------------------------------------------------+\n   | Type      | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'b\'``   | Binary format. Outputs the number in base 2.               |\n   +-----------+------------------------------------------------------------+\n   | ``\'c\'``   | Character. Converts the integer to the corresponding       |\n   |           | unicode character before printing.                         |\n   +-----------+------------------------------------------------------------+\n   | ``\'d\'``   | Decimal Integer. Outputs the number in base 10.            |\n   +-----------+------------------------------------------------------------+\n   | ``\'o\'``   | Octal format. Outputs the number in base 8.                |\n   +-----------+------------------------------------------------------------+\n   | ``\'x\'``   | Hex format. Outputs the number in base 16, using lower-    |\n   |           | case letters for the digits above 9.                       |\n   +-----------+------------------------------------------------------------+\n   | ``\'X\'``   | Hex format. Outputs the number in base 16, using upper-    |\n   |           | case letters for the digits above 9.                       |\n   +-----------+------------------------------------------------------------+\n   | ``\'n\'``   | Number. This is the same as ``\'d\'``, except that it uses   |\n   |           | the current locale setting to insert the appropriate       |\n   |           | number separator characters.                               |\n   +-----------+------------------------------------------------------------+\n   | None      | The same as ``\'d\'``.                                       |\n   +-----------+------------------------------------------------------------+\n\nThe available presentation types for floating point and decimal values\nare:\n\n   +-----------+------------------------------------------------------------+\n   | Type      | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'e\'``   | Exponent notation. Prints the number in scientific         |\n   |           | notation using the letter \'e\' to indicate the exponent.    |\n   +-----------+------------------------------------------------------------+\n   | ``\'E\'``   | Exponent notation. Same as ``\'e\'`` except it uses an upper |\n   |           | case \'E\' as the separator character.                       |\n   +-----------+------------------------------------------------------------+\n   | ``\'f\'``   | Fixed point. Displays the number as a fixed-point number.  |\n   +-----------+------------------------------------------------------------+\n   | ``\'F\'``   | Fixed point. Same as ``\'f\'``, but converts ``nan`` to      |\n   |           | ``NAN`` and ``inf`` to ``INF``.                            |\n   +-----------+------------------------------------------------------------+\n   | ``\'g\'``   | General format. This prints the number as a fixed-point    |\n   |           | number, unless the number is too large, in which case it   |\n   |           | switches to ``\'e\'`` exponent notation. Infinity and NaN    |\n   |           | values are formatted as ``inf``, ``-inf`` and ``nan``,     |\n   |           | respectively.                                              |\n   +-----------+------------------------------------------------------------+\n   | ``\'G\'``   | General format. Same as ``\'g\'`` except switches to ``\'E\'`` |\n   |           | if the number gets to large. The representations of        |\n   |           | infinity and NaN are uppercased, too.                      |\n   +-----------+------------------------------------------------------------+\n   | ``\'n\'``   | Number. This is the same as ``\'g\'``, except that it uses   |\n   |           | the current locale setting to insert the appropriate       |\n   |           | number separator characters.                               |\n   +-----------+------------------------------------------------------------+\n   | ``\'%\'``   | Percentage. Multiplies the number by 100 and displays in   |\n   |           | fixed (``\'f\'``) format, followed by a percent sign.        |\n   +-----------+------------------------------------------------------------+\n   | None      | Similar to ``\'g\'``, except with at least one digit past    |\n   |           | the decimal point and a default precision of 12. This is   |\n   |           | intended to match ``str()``, except you can add the other  |\n   |           | format modifiers.                                          |\n   +-----------+------------------------------------------------------------+\n',
+ 'formatstrings': '\nFormat String Syntax\n********************\n\nThe ``str.format()`` method and the ``Formatter`` class share the same\nsyntax for format strings (although in the case of ``Formatter``,\nsubclasses can define their own format string syntax).\n\nFormat strings contain "replacement fields" surrounded by curly braces\n``{}``. Anything that is not contained in braces is considered literal\ntext, which is copied unchanged to the output.  If you need to include\na brace character in the literal text, it can be escaped by doubling:\n``{{`` and ``}}``.\n\nThe grammar for a replacement field is as follows:\n\n      replacement_field ::= "{" [field_name] ["!" conversion] [":" format_spec] "}"\n      field_name        ::= arg_name ("." attribute_name | "[" element_index "]")*\n      arg_name          ::= [identifier | integer]\n      attribute_name    ::= identifier\n      element_index     ::= integer | index_string\n      index_string      ::= <any source character except "]"> +\n      conversion        ::= "r" | "s" | "a"\n      format_spec       ::= <described in the next section>\n\nIn less formal terms, the replacement field can start with a\n*field_name* that specifies the object whose value is to be formatted\nand inserted into the output instead of the replacement field. The\n*field_name* is optionally followed by a  *conversion* field, which is\npreceded by an exclamation point ``\'!\'``, and a *format_spec*, which\nis preceded by a colon ``\':\'``.  These specify a non-default format\nfor the replacement value.\n\nSee also the *Format Specification Mini-Language* section.\n\nThe *field_name* itself begins with an *arg_name* that is either\neither a number or a keyword.  If it\'s a number, it refers to a\npositional argument, and if it\'s a keyword, it refers to a named\nkeyword argument.  If the numerical arg_names in a format string are\n0, 1, 2, ... in sequence, they can all be omitted (not just some) and\nthe numbers 0, 1, 2, ... will be automatically inserted in that order.\nThe *arg_name* can be followed by any number of index or attribute\nexpressions. An expression of the form ``\'.name\'`` selects the named\nattribute using ``getattr()``, while an expression of the form\n``\'[index]\'`` does an index lookup using ``__getitem__()``.\n\nChanged in version 3.1: The positional argument specifiers can be\nomitted, so ``\'{} {}\'`` is equivalent to ``\'{0} {1}\'``.\n\nSome simple format string examples:\n\n   "First, thou shalt count to {0}" # References first positional argument\n   "Bring me a {}"                  # Implicitly references the first positional argument\n   "From {} to {}"                  # Same as "From {0} to {1}"\n   "My quest is {name}"             # References keyword argument \'name\'\n   "Weight in tons {0.weight}"      # \'weight\' attribute of first positional arg\n   "Units destroyed: {players[0]}"  # First element of keyword argument \'players\'.\n\nThe *conversion* field causes a type coercion before formatting.\nNormally, the job of formatting a value is done by the\n``__format__()`` method of the value itself.  However, in some cases\nit is desirable to force a type to be formatted as a string,\noverriding its own definition of formatting.  By converting the value\nto a string before calling ``__format__()``, the normal formatting\nlogic is bypassed.\n\nThree conversion flags are currently supported: ``\'!s\'`` which calls\n``str()`` on the value, ``\'!r\'`` which calls ``repr()`` and ``\'!a\'``\nwhich calls ``ascii()``.\n\nSome examples:\n\n   "Harold\'s a clever {0!s}"        # Calls str() on the argument first\n   "Bring out the holy {name!r}"    # Calls repr() on the argument first\n   "More {!a}"                      # Calls ascii() on the argument first\n\nThe *format_spec* field contains a specification of how the value\nshould be presented, including such details as field width, alignment,\npadding, decimal precision and so on.  Each value type can define its\nown "formatting mini-language" or interpretation of the *format_spec*.\n\nMost built-in types support a common formatting mini-language, which\nis described in the next section.\n\nA *format_spec* field can also include nested replacement fields\nwithin it. These nested replacement fields can contain only a field\nname; conversion flags and format specifications are not allowed.  The\nreplacement fields within the format_spec are substituted before the\n*format_spec* string is interpreted. This allows the formatting of a\nvalue to be dynamically specified.\n\nSee the *Format examples* section for some examples.\n\n\nFormat Specification Mini-Language\n==================================\n\n"Format specifications" are used within replacement fields contained\nwithin a format string to define how individual values are presented\n(see *Format String Syntax*).  They can also be passed directly to the\nbuilt-in ``format()`` function.  Each formattable type may define how\nthe format specification is to be interpreted.\n\nMost built-in types implement the following options for format\nspecifications, although some of the formatting options are only\nsupported by the numeric types.\n\nA general convention is that an empty format string (``""``) produces\nthe same result as if you had called ``str()`` on the value. A non-\nempty format string typically modifies the result.\n\nThe general form of a *standard format specifier* is:\n\n   format_spec ::= [[fill]align][sign][#][0][width][,][.precision][type]\n   fill        ::= <a character other than \'}\'>\n   align       ::= "<" | ">" | "=" | "^"\n   sign        ::= "+" | "-" | " "\n   width       ::= integer\n   precision   ::= integer\n   type        ::= "b" | "c" | "d" | "e" | "E" | "f" | "F" | "g" | "G" | "n" | "o" | "s" | "x" | "X" | "%"\n\nThe *fill* character can be any character other than \'}\' (which\nsignifies the end of the field).  The presence of a fill character is\nsignaled by the *next* character, which must be one of the alignment\noptions. If the second character of *format_spec* is not a valid\nalignment option, then it is assumed that both the fill character and\nthe alignment option are absent.\n\nThe meaning of the various alignment options is as follows:\n\n   +-----------+------------------------------------------------------------+\n   | Option    | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'<\'``   | Forces the field to be left-aligned within the available   |\n   |           | space (this is the default).                               |\n   +-----------+------------------------------------------------------------+\n   | ``\'>\'``   | Forces the field to be right-aligned within the available  |\n   |           | space.                                                     |\n   +-----------+------------------------------------------------------------+\n   | ``\'=\'``   | Forces the padding to be placed after the sign (if any)    |\n   |           | but before the digits.  This is used for printing fields   |\n   |           | in the form \'+000000120\'. This alignment option is only    |\n   |           | valid for numeric types.                                   |\n   +-----------+------------------------------------------------------------+\n   | ``\'^\'``   | Forces the field to be centered within the available       |\n   |           | space.                                                     |\n   +-----------+------------------------------------------------------------+\n\nNote that unless a minimum field width is defined, the field width\nwill always be the same size as the data to fill it, so that the\nalignment option has no meaning in this case.\n\nThe *sign* option is only valid for number types, and can be one of\nthe following:\n\n   +-----------+------------------------------------------------------------+\n   | Option    | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'+\'``   | indicates that a sign should be used for both positive as  |\n   |           | well as negative numbers.                                  |\n   +-----------+------------------------------------------------------------+\n   | ``\'-\'``   | indicates that a sign should be used only for negative     |\n   |           | numbers (this is the default behavior).                    |\n   +-----------+------------------------------------------------------------+\n   | space     | indicates that a leading space should be used on positive  |\n   |           | numbers, and a minus sign on negative numbers.             |\n   +-----------+------------------------------------------------------------+\n\nThe ``\'#\'`` option is only valid for integers, and only for binary,\noctal, or hexadecimal output.  If present, it specifies that the\noutput will be prefixed by ``\'0b\'``, ``\'0o\'``, or ``\'0x\'``,\nrespectively.\n\nThe ``\',\'`` option signals the use of a comma for a thousands\nseparator. For a locale aware separator, use the ``\'n\'`` integer\npresentation type instead.\n\nChanged in version 3.1: Added the ``\',\'`` option (see also **PEP\n378**).\n\n*width* is a decimal integer defining the minimum field width.  If not\nspecified, then the field width will be determined by the content.\n\nIf the *width* field is preceded by a zero (``\'0\'``) character, this\nenables zero-padding.  This is equivalent to an *alignment* type of\n``\'=\'`` and a *fill* character of ``\'0\'``.\n\nThe *precision* is a decimal number indicating how many digits should\nbe displayed after the decimal point for a floating point value\nformatted with ``\'f\'`` and ``\'F\'``, or before and after the decimal\npoint for a floating point value formatted with ``\'g\'`` or ``\'G\'``.\nFor non-number types the field indicates the maximum field size - in\nother words, how many characters will be used from the field content.\nThe *precision* is not allowed for integer values.\n\nFinally, the *type* determines how the data should be presented.\n\nThe available string presentation types are:\n\n   +-----------+------------------------------------------------------------+\n   | Type      | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'s\'``   | String format. This is the default type for strings and    |\n   |           | may be omitted.                                            |\n   +-----------+------------------------------------------------------------+\n   | None      | The same as ``\'s\'``.                                       |\n   +-----------+------------------------------------------------------------+\n\nThe available integer presentation types are:\n\n   +-----------+------------------------------------------------------------+\n   | Type      | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'b\'``   | Binary format. Outputs the number in base 2.               |\n   +-----------+------------------------------------------------------------+\n   | ``\'c\'``   | Character. Converts the integer to the corresponding       |\n   |           | unicode character before printing.                         |\n   +-----------+------------------------------------------------------------+\n   | ``\'d\'``   | Decimal Integer. Outputs the number in base 10.            |\n   +-----------+------------------------------------------------------------+\n   | ``\'o\'``   | Octal format. Outputs the number in base 8.                |\n   +-----------+------------------------------------------------------------+\n   | ``\'x\'``   | Hex format. Outputs the number in base 16, using lower-    |\n   |           | case letters for the digits above 9.                       |\n   +-----------+------------------------------------------------------------+\n   | ``\'X\'``   | Hex format. Outputs the number in base 16, using upper-    |\n   |           | case letters for the digits above 9.                       |\n   +-----------+------------------------------------------------------------+\n   | ``\'n\'``   | Number. This is the same as ``\'d\'``, except that it uses   |\n   |           | the current locale setting to insert the appropriate       |\n   |           | number separator characters.                               |\n   +-----------+------------------------------------------------------------+\n   | None      | The same as ``\'d\'``.                                       |\n   +-----------+------------------------------------------------------------+\n\nIn addition to the above presentation types, integers can be formatted\nwith the floating point presentation types listed below (except\n``\'n\'`` and None). When doing so, ``float()`` is used to convert the\ninteger to a floating point number before formatting.\n\nThe available presentation types for floating point and decimal values\nare:\n\n   +-----------+------------------------------------------------------------+\n   | Type      | Meaning                                                    |\n   +===========+============================================================+\n   | ``\'e\'``   | Exponent notation. Prints the number in scientific         |\n   |           | notation using the letter \'e\' to indicate the exponent.    |\n   +-----------+------------------------------------------------------------+\n   | ``\'E\'``   | Exponent notation. Same as ``\'e\'`` except it uses an upper |\n   |           | case \'E\' as the separator character.                       |\n   +-----------+------------------------------------------------------------+\n   | ``\'f\'``   | Fixed point. Displays the number as a fixed-point number.  |\n   +-----------+------------------------------------------------------------+\n   | ``\'F\'``   | Fixed point. Same as ``\'f\'``, but converts ``nan`` to      |\n   |           | ``NAN`` and ``inf`` to ``INF``.                            |\n   +-----------+------------------------------------------------------------+\n   | ``\'g\'``   | General format.  For a given precision ``p >= 1``, this    |\n   |           | rounds the number to ``p`` significant digits and then     |\n   |           | formats the result in either fixed-point format or in      |\n   |           | scientific notation, depending on its magnitude.  The      |\n   |           | precise rules are as follows: suppose that the result      |\n   |           | formatted with presentation type ``\'e\'`` and precision     |\n   |           | ``p-1`` would have exponent ``exp``.  Then if ``-4 <= exp  |\n   |           | < p``, the number is formatted with presentation type      |\n   |           | ``\'f\'`` and precision ``p-1-exp``. Otherwise, the number   |\n   |           | is formatted with presentation type ``\'e\'`` and precision  |\n   |           | ``p-1``. In both cases insignificant trailing zeros are    |\n   |           | removed from the significand, and the decimal point is     |\n   |           | also removed if there are no remaining digits following    |\n   |           | it.  Postive and negative infinity, positive and negative  |\n   |           | zero, and nans, are formatted as ``inf``, ``-inf``, ``0``, |\n   |           | ``-0`` and ``nan`` respectively, regardless of the         |\n   |           | precision.  A precision of ``0`` is treated as equivalent  |\n   |           | to a precision of ``1``.                                   |\n   +-----------+------------------------------------------------------------+\n   | ``\'G\'``   | General format. Same as ``\'g\'`` except switches to ``\'E\'`` |\n   |           | if the number gets too large. The representations of       |\n   |           | infinity and NaN are uppercased, too.                      |\n   +-----------+------------------------------------------------------------+\n   | ``\'n\'``   | Number. This is the same as ``\'g\'``, except that it uses   |\n   |           | the current locale setting to insert the appropriate       |\n   |           | number separator characters.                               |\n   +-----------+------------------------------------------------------------+\n   | ``\'%\'``   | Percentage. Multiplies the number by 100 and displays in   |\n   |           | fixed (``\'f\'``) format, followed by a percent sign.        |\n   +-----------+------------------------------------------------------------+\n   | None      | Similar to ``\'g\'``, except with at least one digit past    |\n   |           | the decimal point and a default precision of 12. This is   |\n   |           | intended to match ``str()``, except you can add the other  |\n   |           | format modifiers.                                          |\n   +-----------+------------------------------------------------------------+\n\n\nFormat examples\n===============\n\nThis section contains examples of the new format syntax and comparison\nwith the old ``%``-formatting.\n\nIn most of the cases the syntax is similar to the old\n``%``-formatting, with the addition of the ``{}`` and with ``:`` used\ninstead of ``%``. For example, ``\'%03.2f\'`` can be translated to\n``\'{:03.2f}\'``.\n\nThe new format syntax also supports new and different options, shown\nin the follow examples.\n\nAccessing arguments by position:\n\n   >>> \'{0}, {1}, {2}\'.format(\'a\', \'b\', \'c\')\n   \'a, b, c\'\n   >>> \'{}, {}, {}\'.format(\'a\', \'b\', \'c\')  # 3.1+ only\n   \'a, b, c\'\n   >>> \'{2}, {1}, {0}\'.format(\'a\', \'b\', \'c\')\n   \'c, b, a\'\n   >>> \'{2}, {1}, {0}\'.format(*\'abc\')      # unpacking argument sequence\n   \'c, b, a\'\n   >>> \'{0}{1}{0}\'.format(\'abra\', \'cad\')   # arguments\' indices can be repeated\n   \'abracadabra\'\n\nAccessing arguments by name:\n\n   >>> \'Coordinates: {latitude}, {longitude}\'.format(latitude=\'37.24N\', longitude=\'-115.81W\')\n   \'Coordinates: 37.24N, -115.81W\'\n   >>> coord = {\'latitude\': \'37.24N\', \'longitude\': \'-115.81W\'}\n   >>> \'Coordinates: {latitude}, {longitude}\'.format(**coord)\n   \'Coordinates: 37.24N, -115.81W\'\n\nAccessing arguments\' attributes:\n\n   >>> c = 3-5j\n   >>> (\'The complex number {0} is formed from the real part {0.real} \'\n   ...  \'and the imaginary part {0.imag}.\').format(c)\n   \'The complex number (3-5j) is formed from the real part 3.0 and the imaginary part -5.0.\'\n   >>> class Point:\n   ...     def __init__(self, x, y):\n   ...         self.x, self.y = x, y\n   ...     def __str__(self):\n   ...         return \'Point({self.x}, {self.y})\'.format(self=self)\n   ...\n   >>> str(Point(4, 2))\n   \'Point(4, 2)\'\n\nAccessing arguments\' items:\n\n   >>> coord = (3, 5)\n   >>> \'X: {0[0]};  Y: {0[1]}\'.format(coord)\n   \'X: 3;  Y: 5\'\n\nReplacing ``%s`` and ``%r``:\n\n   >>> "repr() shows quotes: {!r}; str() doesn\'t: {!s}".format(\'test1\', \'test2\')\n   "repr() shows quotes: \'test1\'; str() doesn\'t: test2"\n\nAligning the text and specifying a width:\n\n   >>> \'{:<30}\'.format(\'left aligned\')\n   \'left aligned                  \'\n   >>> \'{:>30}\'.format(\'right aligned\')\n   \'                 right aligned\'\n   >>> \'{:^30}\'.format(\'centered\')\n   \'           centered           \'\n   >>> \'{:*^30}\'.format(\'centered\')  # use \'*\' as a fill char\n   \'***********centered***********\'\n\nReplacing ``%+f``, ``%-f``, and ``% f`` and specifying a sign:\n\n   >>> \'{:+f}; {:+f}\'.format(3.14, -3.14)  # show it always\n   \'+3.140000; -3.140000\'\n   >>> \'{: f}; {: f}\'.format(3.14, -3.14)  # show a space for positive numbers\n   \' 3.140000; -3.140000\'\n   >>> \'{:-f}; {:-f}\'.format(3.14, -3.14)  # show only the minus -- same as \'{:f}; {:f}\'\n   \'3.140000; -3.140000\'\n\nReplacing ``%x`` and ``%o`` and converting the value to different\nbases:\n\n   >>> # format also supports binary numbers\n   >>> "int: {0:d};  hex: {0:x};  oct: {0:o};  bin: {0:b}".format(42)\n   \'int: 42;  hex: 2a;  oct: 52;  bin: 101010\'\n   >>> # with 0x, 0o, or 0b as prefix:\n   >>> "int: {0:d};  hex: {0:#x};  oct: {0:#o};  bin: {0:#b}".format(42)\n   \'int: 42;  hex: 0x2a;  oct: 0o52;  bin: 0b101010\'\n\nUsing the comma as a thousands separator:\n\n   >>> \'{:,}\'.format(1234567890)\n   \'1,234,567,890\'\n\nExpressing a percentage:\n\n   >>> points = 19\n   >>> total = 22\n   >>> \'Correct answers: {:.2%}.\'.format(points/total)\n   \'Correct answers: 86.36%\'\n\nUsing type-specific formatting:\n\n   >>> import datetime\n   >>> d = datetime.datetime(2010, 7, 4, 12, 15, 58)\n   >>> \'{:%Y-%m-%d %H:%M:%S}\'.format(d)\n   \'2010-07-04 12:15:58\'\n\nNesting arguments and more complex examples:\n\n   >>> for align, text in zip(\'<^>\', [\'left\', \'center\', \'right\']):\n   ...     \'{0:{align}{fill}16}\'.format(text, fill=align, align=align)\n   ...\n   \'left<<<<<<<<<<<<\'\n   \'^^^^^center^^^^^\'\n   \'>>>>>>>>>>>right\'\n   >>>\n   >>> octets = [192, 168, 0, 1]\n   >>> \'{:02X}{:02X}{:02X}{:02X}\'.format(*octets)\n   \'C0A80001\'\n   >>> int(_, 16)\n   3232235521\n   >>>\n   >>> width = 5\n   >>> for num in range(5,12):\n   ...     for base in \'dXob\':\n   ...         print(\'{0:{width}{base}}\'.format(num, base=base, width=width), end=\' \')\n   ...     print()\n   ...\n       5     5     5   101\n       6     6     6   110\n       7     7     7   111\n       8     8    10  1000\n       9     9    11  1001\n      10     A    12  1010\n      11     B    13  1011\n',
  'function': '\nFunction definitions\n********************\n\nA function definition defines a user-defined function object (see\nsection *The standard type hierarchy*):\n\n   funcdef        ::= [decorators] "def" funcname "(" [parameter_list] ")" ["->" expression] ":" suite\n   decorators     ::= decorator+\n   decorator      ::= "@" dotted_name ["(" [argument_list [","]] ")"] NEWLINE\n   dotted_name    ::= identifier ("." identifier)*\n   parameter_list ::= (defparameter ",")*\n                      (  "*" [parameter] ("," defparameter)*\n                      [, "**" parameter]\n                      | "**" parameter\n                      | defparameter [","] )\n   parameter      ::= identifier [":" expression]\n   defparameter   ::= parameter ["=" expression]\n   funcname       ::= identifier\n\nA function definition is an executable statement.  Its execution binds\nthe function name in the current local namespace to a function object\n(a wrapper around the executable code for the function).  This\nfunction object contains a reference to the current global namespace\nas the global namespace to be used when the function is called.\n\nThe function definition does not execute the function body; this gets\nexecuted only when the function is called. [3]\n\nA function definition may be wrapped by one or more *decorator*\nexpressions. Decorator expressions are evaluated when the function is\ndefined, in the scope that contains the function definition.  The\nresult must be a callable, which is invoked with the function object\nas the only argument. The returned value is bound to the function name\ninstead of the function object.  Multiple decorators are applied in\nnested fashion. For example, the following code\n\n   @f1(arg)\n   @f2\n   def func(): pass\n\nis equivalent to\n\n   def func(): pass\n   func = f1(arg)(f2(func))\n\nWhen one or more parameters have the form *parameter* ``=``\n*expression*, the function is said to have "default parameter values."\nFor a parameter with a default value, the corresponding argument may\nbe omitted from a call, in which case the parameter\'s default value is\nsubstituted.  If a parameter has a default value, all following\nparameters up until the "``*``" must also have a default value ---\nthis is a syntactic restriction that is not expressed by the grammar.\n\n**Default parameter values are evaluated when the function definition\nis executed.** This means that the expression is evaluated once, when\nthe function is defined, and that that same "pre-computed" value is\nused for each call.  This is especially important to understand when a\ndefault parameter is a mutable object, such as a list or a dictionary:\nif the function modifies the object (e.g. by appending an item to a\nlist), the default value is in effect modified. This is generally not\nwhat was intended.  A way around this is to use ``None`` as the\ndefault, and explicitly test for it in the body of the function, e.g.:\n\n   def whats_on_the_telly(penguin=None):\n       if penguin is None:\n           penguin = []\n       penguin.append("property of the zoo")\n       return penguin\n\nFunction call semantics are described in more detail in section\n*Calls*. A function call always assigns values to all parameters\nmentioned in the parameter list, either from position arguments, from\nkeyword arguments, or from default values.  If the form\n"``*identifier``" is present, it is initialized to a tuple receiving\nany excess positional parameters, defaulting to the empty tuple.  If\nthe form "``**identifier``" is present, it is initialized to a new\ndictionary receiving any excess keyword arguments, defaulting to a new\nempty dictionary. Parameters after "``*``" or "``*identifier``" are\nkeyword-only parameters and may only be passed used keyword arguments.\n\nParameters may have annotations of the form "``: expression``"\nfollowing the parameter name.  Any parameter may have an annotation\neven those of the form ``*identifier`` or ``**identifier``.  Functions\nmay have "return" annotation of the form "``-> expression``" after the\nparameter list.  These annotations can be any valid Python expression\nand are evaluated when the function definition is executed.\nAnnotations may be evaluated in a different order than they appear in\nthe source code.  The presence of annotations does not change the\nsemantics of a function.  The annotation values are available as\nvalues of a dictionary keyed by the parameters\' names in the\n``__annotations__`` attribute of the function object.\n\nIt is also possible to create anonymous functions (functions not bound\nto a name), for immediate use in expressions.  This uses lambda forms,\ndescribed in section *Lambdas*.  Note that the lambda form is merely a\nshorthand for a simplified function definition; a function defined in\na "``def``" statement can be passed around or assigned to another name\njust like a function defined by a lambda form.  The "``def``" form is\nactually more powerful since it allows the execution of multiple\nstatements and annotations.\n\n**Programmer\'s note:** Functions are first-class objects.  A "``def``"\nform executed inside a function definition defines a local function\nthat can be returned or passed around.  Free variables used in the\nnested function can access the local variables of the function\ncontaining the def.  See section *Naming and binding* for details.\n',
- 'global': '\nThe ``global`` statement\n************************\n\n   global_stmt ::= "global" identifier ("," identifier)*\n\nThe ``global`` statement is a declaration which holds for the entire\ncurrent code block.  It means that the listed identifiers are to be\ninterpreted as globals.  It would be impossible to assign to a global\nvariable without ``global``, although free variables may refer to\nglobals without being declared global.\n\nNames listed in a ``global`` statement must not be used in the same\ncode block textually preceding that ``global`` statement.\n\nNames listed in a ``global`` statement must not be defined as formal\nparameters or in a ``for`` loop control target, ``class`` definition,\nfunction definition, or ``import`` statement.\n\n(The current implementation does not enforce the latter two\nrestrictions, but programs should not abuse this freedom, as future\nimplementations may enforce them or silently change the meaning of the\nprogram.)\n\n**Programmer\'s note:** the ``global`` is a directive to the parser.\nIt applies only to code parsed at the same time as the ``global``\nstatement. In particular, a ``global`` statement contained in a string\nor code object supplied to the builtin ``exec()`` function does not\naffect the code block *containing* the function call, and code\ncontained in such a string is unaffected by ``global`` statements in\nthe code containing the function call.  The same applies to the\n``eval()`` and ``compile()`` functions.\n',
+ 'global': '\nThe ``global`` statement\n************************\n\n   global_stmt ::= "global" identifier ("," identifier)*\n\nThe ``global`` statement is a declaration which holds for the entire\ncurrent code block.  It means that the listed identifiers are to be\ninterpreted as globals.  It would be impossible to assign to a global\nvariable without ``global``, although free variables may refer to\nglobals without being declared global.\n\nNames listed in a ``global`` statement must not be used in the same\ncode block textually preceding that ``global`` statement.\n\nNames listed in a ``global`` statement must not be defined as formal\nparameters or in a ``for`` loop control target, ``class`` definition,\nfunction definition, or ``import`` statement.\n\n**CPython implementation detail:** The current implementation does not\nenforce the latter two restrictions, but programs should not abuse\nthis freedom, as future implementations may enforce them or silently\nchange the meaning of the program.\n\n**Programmer\'s note:** the ``global`` is a directive to the parser.\nIt applies only to code parsed at the same time as the ``global``\nstatement. In particular, a ``global`` statement contained in a string\nor code object supplied to the built-in ``exec()`` function does not\naffect the code block *containing* the function call, and code\ncontained in such a string is unaffected by ``global`` statements in\nthe code containing the function call.  The same applies to the\n``eval()`` and ``compile()`` functions.\n',
  'id-classes': '\nReserved classes of identifiers\n*******************************\n\nCertain classes of identifiers (besides keywords) have special\nmeanings.  These classes are identified by the patterns of leading and\ntrailing underscore characters:\n\n``_*``\n   Not imported by ``from module import *``.  The special identifier\n   ``_`` is used in the interactive interpreter to store the result of\n   the last evaluation; it is stored in the ``builtins`` module.  When\n   not in interactive mode, ``_`` has no special meaning and is not\n   defined. See section *The import statement*.\n\n   Note: The name ``_`` is often used in conjunction with\n     internationalization; refer to the documentation for the\n     ``gettext`` module for more information on this convention.\n\n``__*__``\n   System-defined names.  These names are defined by the interpreter\n   and its implementation (including the standard library);\n   applications should not expect to define additional names using\n   this convention.  The set of names of this class defined by Python\n   may be extended in future versions. See section *Special method\n   names*.\n\n``__*``\n   Class-private names.  Names in this category, when used within the\n   context of a class definition, are re-written to use a mangled form\n   to help avoid name clashes between "private" attributes of base and\n   derived classes. See section *Identifiers (Names)*.\n',
  'identifiers': '\nIdentifiers and keywords\n************************\n\nIdentifiers (also referred to as *names*) are described by the\nfollowing lexical definitions.\n\nThe syntax of identifiers in Python is based on the Unicode standard\nannex UAX-31, with elaboration and changes as defined below; see also\n**PEP 3131** for further details.\n\nWithin the ASCII range (U+0001..U+007F), the valid characters for\nidentifiers are the same as in Python 2.x: the uppercase and lowercase\nletters ``A`` through ``Z``, the underscore ``_`` and, except for the\nfirst character, the digits ``0`` through ``9``.\n\nPython 3.0 introduces additional characters from outside the ASCII\nrange (see **PEP 3131**).  For these characters, the classification\nuses the version of the Unicode Character Database as included in the\n``unicodedata`` module.\n\nIdentifiers are unlimited in length.  Case is significant.\n\n   identifier  ::= id_start id_continue*\n   id_start    ::= <all characters in general categories Lu, Ll, Lt, Lm, Lo, Nl, the underscore, and characters with the Other_ID_Start property>\n   id_continue ::= <all characters in id_start, plus characters in the categories Mn, Mc, Nd, Pc and others with the Other_ID_Continue property>\n\nThe Unicode category codes mentioned above stand for:\n\n* *Lu* - uppercase letters\n\n* *Ll* - lowercase letters\n\n* *Lt* - titlecase letters\n\n* *Lm* - modifier letters\n\n* *Lo* - other letters\n\n* *Nl* - letter numbers\n\n* *Mn* - nonspacing marks\n\n* *Mc* - spacing combining marks\n\n* *Nd* - decimal numbers\n\n* *Pc* - connector punctuations\n\nAll identifiers are converted into the normal form NFC while parsing;\ncomparison of identifiers is based on NFC.\n\nA non-normative HTML file listing all valid identifier characters for\nUnicode 4.1 can be found at http://www.dcl.hpi.uni-\npotsdam.de/home/loewis/table-3131.html.\n\n\nKeywords\n========\n\nThe following identifiers are used as reserved words, or *keywords* of\nthe language, and cannot be used as ordinary identifiers.  They must\nbe spelled exactly as written here:\n\n   False      class      finally    is         return\n   None       continue   for        lambda     try\n   True       def        from       nonlocal   while\n   and        del        global     not        with\n   as         elif       if         or         yield\n   assert     else       import     pass\n   break      except     in         raise\n\n\nReserved classes of identifiers\n===============================\n\nCertain classes of identifiers (besides keywords) have special\nmeanings.  These classes are identified by the patterns of leading and\ntrailing underscore characters:\n\n``_*``\n   Not imported by ``from module import *``.  The special identifier\n   ``_`` is used in the interactive interpreter to store the result of\n   the last evaluation; it is stored in the ``builtins`` module.  When\n   not in interactive mode, ``_`` has no special meaning and is not\n   defined. See section *The import statement*.\n\n   Note: The name ``_`` is often used in conjunction with\n     internationalization; refer to the documentation for the\n     ``gettext`` module for more information on this convention.\n\n``__*__``\n   System-defined names.  These names are defined by the interpreter\n   and its implementation (including the standard library);\n   applications should not expect to define additional names using\n   this convention.  The set of names of this class defined by Python\n   may be extended in future versions. See section *Special method\n   names*.\n\n``__*``\n   Class-private names.  Names in this category, when used within the\n   context of a class definition, are re-written to use a mangled form\n   to help avoid name clashes between "private" attributes of base and\n   derived classes. See section *Identifiers (Names)*.\n',
  'if': '\nThe ``if`` statement\n********************\n\nThe ``if`` statement is used for conditional execution:\n\n   if_stmt ::= "if" expression ":" suite\n               ( "elif" expression ":" suite )*\n               ["else" ":" suite]\n\nIt selects exactly one of the suites by evaluating the expressions one\nby one until one is found to be true (see section *Boolean operations*\nfor the definition of true and false); then that suite is executed\n(and no other part of the ``if`` statement is executed or evaluated).\nIf all expressions are false, the suite of the ``else`` clause, if\npresent, is executed.\n',
  'imaginary': '\nImaginary literals\n******************\n\nImaginary literals are described by the following lexical definitions:\n\n   imagnumber ::= (floatnumber | intpart) ("j" | "J")\n\nAn imaginary literal yields a complex number with a real part of 0.0.\nComplex numbers are represented as a pair of floating point numbers\nand have the same restrictions on their range.  To create a complex\nnumber with a nonzero real part, add a floating point number to it,\ne.g., ``(3+4j)``.  Some examples of imaginary literals:\n\n   3.14j   10.j    10j     .001j   1e100j  3.14e-10j\n',
- 'import': '\nThe ``import`` statement\n************************\n\n   import_stmt     ::= "import" module ["as" name] ( "," module ["as" name] )*\n                   | "from" relative_module "import" identifier ["as" name]\n                   ( "," identifier ["as" name] )*\n                   | "from" relative_module "import" "(" identifier ["as" name]\n                   ( "," identifier ["as" name] )* [","] ")"\n                   | "from" module "import" "*"\n   module          ::= (identifier ".")* identifier\n   relative_module ::= "."* module | "."+\n   name            ::= identifier\n\nImport statements are executed in two steps: (1) find a module, and\ninitialize it if necessary; (2) define a name or names in the local\nnamespace (of the scope where the ``import`` statement occurs). The\nstatement comes in two forms differing on whether it uses the ``from``\nkeyword. The first form (without ``from``) repeats these steps for\neach identifier in the list. The form with ``from`` performs step (1)\nonce, and then performs step (2) repeatedly. For a reference\nimplementation of step (1), see the ``importlib`` module.\n\nTo understand how step (1) occurs, one must first understand how\nPython handles hierarchical naming of modules. To help organize\nmodules and provide a hierarchy in naming, Python has a concept of\npackages. A package can contain other packages and modules while\nmodules cannot contain other modules or packages. From a file system\nperspective, packages are directories and modules are files. The\noriginal specification for packages is still available to read,\nalthough minor details have changed since the writing of that\ndocument.\n\nOnce the name of the module is known (unless otherwise specified, the\nterm "module" will refer to both packages and modules), searching for\nthe module or package can begin. The first place checked is\n``sys.modules``, the cache of all modules that have been imported\npreviously. If the module is found there then it is used in step (2)\nof import.\n\nIf the module is not found in the cache, then ``sys.meta_path`` is\nsearched (the specification for ``sys.meta_path`` can be found in\n**PEP 302**). The object is a list of *finder* objects which are\nqueried in order as to whether they know how to load the module by\ncalling their ``find_module()`` method with the name of the module. If\nthe module happens to be contained within a package (as denoted by the\nexistence of a dot in the name), then a second argument to\n``find_module()`` is given as the value of the ``__path__`` attribute\nfrom the parent package (everything up to the last dot in the name of\nthe module being imported). If a finder can find the module it returns\na *loader* (discussed later) or returns ``None``.\n\nIf none of the finders on ``sys.meta_path`` are able to find the\nmodule then some implicitly defined finders are queried.\nImplementations of Python vary in what implicit meta path finders are\ndefined. The one they all do define, though, is one that handles\n``sys.path_hooks``, ``sys.path_importer_cache``, and ``sys.path``.\n\nThe implicit finder searches for the requested module in the "paths"\nspecified in one of two places ("paths" do not have to be file system\npaths). If the module being imported is supposed to be contained\nwithin a package then the second argument passed to ``find_module()``,\n``__path__`` on the parent package, is used as the source of paths. If\nthe module is not contained in a package then ``sys.path`` is used as\nthe source of paths.\n\nOnce the source of paths is chosen it is iterated over to find a\nfinder that can handle that path. The dict at\n``sys.path_importer_cache`` caches finders for paths and is checked\nfor a finder. If the path does not have a finder cached then\n``sys.path_hooks`` is searched by calling each object in the list with\na single argument of the path, returning a finder or raises\n``ImportError``. If a finder is returned then it is cached in\n``sys.path_importer_cache`` and then used for that path entry. If no\nfinder can be found but the path exists then a value of ``None`` is\nstored in ``sys.path_importer_cache`` to signify that an implicit,\nfile-based finder that handles modules stored as individual files\nshould be used for that path. If the path does not exist then a finder\nwhich always returns ``None`` is placed in the cache for the path.\n\nIf no finder can find the module then ``ImportError`` is raised.\nOtherwise some finder returned a loader whose ``load_module()`` method\nis called with the name of the module to load (see **PEP 302** for the\noriginal definition of loaders). A loader has several responsibilities\nto perform on a module it loads. First, if the module already exists\nin ``sys.modules`` (a possibility if the loader is called outside of\nthe import machinery) then it is to use that module for initialization\nand not a new module. But if the module does not exist in\n``sys.modules`` then it is to be added to that dict before\ninitialization begins. If an error occurs during loading of the module\nand it was added to ``sys.modules`` it is to be removed from the dict.\nIf an error occurs but the module was already in ``sys.modules`` it is\nleft in the dict.\n\nThe loader must set several attributes on the module. ``__name__`` is\nto be set to the name of the module. ``__file__`` is to be the "path"\nto the file unless the module is built-in (and thus listed in\n``sys.builtin_module_names``) in which case the attribute is not set.\nIf what is being imported is a package then ``__path__`` is to be set\nto a list of paths to be searched when looking for modules and\npackages contained within the package being imported. ``__package__``\nis optional but should be set to the name of package that contains the\nmodule or package (the empty string is used for module not contained\nin a package). ``__loader__`` is also optional but should be set to\nthe loader object that is loading the module.\n\nIf an error occurs during loading then the loader raises\n``ImportError`` if some other exception is not already being\npropagated. Otherwise the loader returns the module that was loaded\nand initialized.\n\nWhen step (1) finishes without raising an exception, step (2) can\nbegin.\n\nThe first form of ``import`` statement binds the module name in the\nlocal namespace to the module object, and then goes on to import the\nnext identifier, if any.  If the module name is followed by ``as``,\nthe name following ``as`` is used as the local name for the module.\n\nThe ``from`` form does not bind the module name: it goes through the\nlist of identifiers, looks each one of them up in the module found in\nstep (1), and binds the name in the local namespace to the object thus\nfound.  As with the first form of ``import``, an alternate local name\ncan be supplied by specifying "``as`` localname".  If a name is not\nfound, ``ImportError`` is raised.  If the list of identifiers is\nreplaced by a star (``\'*\'``), all public names defined in the module\nare bound in the local namespace of the ``import`` statement..\n\nThe *public names* defined by a module are determined by checking the\nmodule\'s namespace for a variable named ``__all__``; if defined, it\nmust be a sequence of strings which are names defined or imported by\nthat module.  The names given in ``__all__`` are all considered public\nand are required to exist.  If ``__all__`` is not defined, the set of\npublic names includes all names found in the module\'s namespace which\ndo not begin with an underscore character (``\'_\'``). ``__all__``\nshould contain the entire public API. It is intended to avoid\naccidentally exporting items that are not part of the API (such as\nlibrary modules which were imported and used within the module).\n\nThe ``from`` form with ``*`` may only occur in a module scope.  The\nwild card form of import --- ``import *`` --- is only allowed at the\nmodule level. Attempting to use it in class for function definitions\nwill raise a ``SyntaxError``.\n\nWhen specifying what module to import you do not have to specify the\nabsolute name of the module. When a module or package is contained\nwithin another package it is possible to make a relative import within\nthe same top package without having to mention the package name. By\nusing leading dots in the specified module or package after ``from``\nyou can specify how high to traverse up the current package hierarchy\nwithout specifying exact names. One leading dot means the current\npackage where the module making the import exists. Two dots means up\none package level. Three dots is up two levels, etc. So if you execute\n``from . import mod`` from a module in the ``pkg`` package then you\nwill end up importing ``pkg.mod``. If you execute ``from ..subpkg2\nimprt mod`` from within ``pkg.subpkg1`` you will import\n``pkg.subpkg2.mod``. The specification for relative imports is\ncontained within **PEP 328**.\n\n``importlib.import_module()`` is provided to support applications that\ndetermine which modules need to be loaded dynamically.\n\n\nFuture statements\n=================\n\nA *future statement* is a directive to the compiler that a particular\nmodule should be compiled using syntax or semantics that will be\navailable in a specified future release of Python.  The future\nstatement is intended to ease migration to future versions of Python\nthat introduce incompatible changes to the language.  It allows use of\nthe new features on a per-module basis before the release in which the\nfeature becomes standard.\n\n   future_statement ::= "from" "__future__" "import" feature ["as" name]\n                        ("," feature ["as" name])*\n                        | "from" "__future__" "import" "(" feature ["as" name]\n                        ("," feature ["as" name])* [","] ")"\n   feature          ::= identifier\n   name             ::= identifier\n\nA future statement must appear near the top of the module.  The only\nlines that can appear before a future statement are:\n\n* the module docstring (if any),\n\n* comments,\n\n* blank lines, and\n\n* other future statements.\n\nThe features recognized by Python 3.0 are ``absolute_import``,\n``division``, ``generators``, ``unicode_literals``,\n``print_function``, ``nested_scopes`` and ``with_statement``.  They\nare all redundant because they are always enabled, and only kept for\nbackwards compatibility.\n\nA future statement is recognized and treated specially at compile\ntime: Changes to the semantics of core constructs are often\nimplemented by generating different code.  It may even be the case\nthat a new feature introduces new incompatible syntax (such as a new\nreserved word), in which case the compiler may need to parse the\nmodule differently.  Such decisions cannot be pushed off until\nruntime.\n\nFor any given release, the compiler knows which feature names have\nbeen defined, and raises a compile-time error if a future statement\ncontains a feature not known to it.\n\nThe direct runtime semantics are the same as for any import statement:\nthere is a standard module ``__future__``, described later, and it\nwill be imported in the usual way at the time the future statement is\nexecuted.\n\nThe interesting runtime semantics depend on the specific feature\nenabled by the future statement.\n\nNote that there is nothing special about the statement:\n\n   import __future__ [as name]\n\nThat is not a future statement; it\'s an ordinary import statement with\nno special semantics or syntax restrictions.\n\nCode compiled by calls to the builtin functions ``exec()`` and\n``compile()`` that occur in a module ``M`` containing a future\nstatement will, by default, use the new syntax or semantics associated\nwith the future statement.  This can be controlled by optional\narguments to ``compile()`` --- see the documentation of that function\nfor details.\n\nA future statement typed at an interactive interpreter prompt will\ntake effect for the rest of the interpreter session.  If an\ninterpreter is started with the *-i* option, is passed a script name\nto execute, and the script includes a future statement, it will be in\neffect in the interactive session started after the script is\nexecuted.\n\nSee also:\n\n   **PEP 236** - Back to the __future__\n      The original proposal for the __future__ mechanism.\n',
- 'in': '\nComparisons\n***********\n\nUnlike C, all comparison operations in Python have the same priority,\nwhich is lower than that of any arithmetic, shifting or bitwise\noperation.  Also unlike C, expressions like ``a < b < c`` have the\ninterpretation that is conventional in mathematics:\n\n   comparison    ::= or_expr ( comp_operator or_expr )*\n   comp_operator ::= "<" | ">" | "==" | ">=" | "<=" | "!="\n                     | "is" ["not"] | ["not"] "in"\n\nComparisons yield boolean values: ``True`` or ``False``.\n\nComparisons can be chained arbitrarily, e.g., ``x < y <= z`` is\nequivalent to ``x < y and y <= z``, except that ``y`` is evaluated\nonly once (but in both cases ``z`` is not evaluated at all when ``x <\ny`` is found to be false).\n\nFormally, if *a*, *b*, *c*, ..., *y*, *z* are expressions and *op1*,\n*op2*, ..., *opN* are comparison operators, then ``a op1 b op2 c ... y\nopN z`` is equivalent to ``a op1 b and b op2 c and ... y opN z``,\nexcept that each expression is evaluated at most once.\n\nNote that ``a op1 b op2 c`` doesn\'t imply any kind of comparison\nbetween *a* and *c*, so that, e.g., ``x < y > z`` is perfectly legal\n(though perhaps not pretty).\n\nThe operators ``<``, ``>``, ``==``, ``>=``, ``<=``, and ``!=`` compare\nthe values of two objects.  The objects need not have the same type.\nIf both are numbers, they are converted to a common type.  Otherwise,\nthe ``==`` and ``!=`` operators *always* consider objects of different\ntypes to be unequal, while the ``<``, ``>``, ``>=`` and ``<=``\noperators raise a ``TypeError`` when comparing objects of different\ntypes that do not implement these operators for the given pair of\ntypes.  You can control comparison behavior of objects of non-builtin\ntypes by defining rich comparison methods like ``__gt__()``, described\nin section *Basic customization*.\n\nComparison of objects of the same type depends on the type:\n\n* Numbers are compared arithmetically.\n\n* The values ``float(\'NaN\')`` and ``Decimal(\'NaN\')`` are special. The\n  are identical to themselves, ``x is x`` but are not equal to\n  themselves, ``x != x``.  Additionally, comparing any value to a\n  not-a-number value will return ``False``.  For example, both ``3 <\n  float(\'NaN\')`` and ``float(\'NaN\') < 3`` will return ``False``.\n\n* Bytes objects are compared lexicographically using the numeric\n  values of their elements.\n\n* Strings are compared lexicographically using the numeric equivalents\n  (the result of the built-in function ``ord()``) of their characters.\n  [3] String and bytes object can\'t be compared!\n\n* Tuples and lists are compared lexicographically using comparison of\n  corresponding elements.  This means that to compare equal, each\n  element must compare equal and the two sequences must be of the same\n  type and have the same length.\n\n  If not equal, the sequences are ordered the same as their first\n  differing elements.  For example, ``[1,2,x] <= [1,2,y]`` has the\n  same value as ``x <= y``.  If the corresponding element does not\n  exist, the shorter sequence is ordered first (for example, ``[1,2] <\n  [1,2,3]``).\n\n* Mappings (dictionaries) compare equal if and only if their sorted\n  ``(key, value)`` lists compare equal. [4] Outcomes other than\n  equality are resolved consistently, but are not otherwise defined.\n  [5]\n\n* Sets and frozensets define comparison operators to mean subset and\n  superset tests.  Those relations do not define total orderings (the\n  two sets ``{1,2}`` and {2,3} are not equal, nor subsets of one\n  another, nor supersets of one another).  Accordingly, sets are not\n  appropriate arguments for functions which depend on total ordering.\n  For example, ``min()``, ``max()``, and ``sorted()`` produce\n  undefined results given a list of sets as inputs.\n\n* Most other objects of builtin types compare unequal unless they are\n  the same object; the choice whether one object is considered smaller\n  or larger than another one is made arbitrarily but consistently\n  within one execution of a program.\n\nComparison of objects of the differing types depends on whether either\nof the types provide explicit support for the comparison.  Most\nnumeric types can be compared with one another, but comparisons of\n``float`` and ``Decimal`` are not supported to avoid the inevitable\nconfusion arising from representation issues such as ``float(\'1.1\')``\nbeing inexactly represented and therefore not exactly equal to\n``Decimal(\'1.1\')`` which is.  When cross-type comparison is not\nsupported, the comparison method returns ``NotImplemented``.  This can\ncreate the illusion of non-transitivity between supported cross-type\ncomparisons and unsupported comparisons.  For example, ``Decimal(2) ==\n2`` and *2 == float(2)`* but ``Decimal(2) != float(2)``.\n\nThe operators ``in`` and ``not in`` test for membership.  ``x in s``\nevaluates to true if *x* is a member of *s*, and false otherwise.  ``x\nnot in s`` returns the negation of ``x in s``.  All built-in sequences\nand set types support this as well as dictionary, for which ``in``\ntests whether a the dictionary has a given key. For container types\nsuch as list, tuple, set, frozenset, dict, or collections.deque, the\nexpression ``x in y`` is equivalent to ``any(x is e or x == e for val\ne in y)``.\n\nFor the string and bytes types, ``x in y`` is true if and only if *x*\nis a substring of *y*.  An equivalent test is ``y.find(x) != -1``.\nEmpty strings are always considered to be a substring of any other\nstring, so ``"" in "abc"`` will return ``True``.\n\nFor user-defined classes which define the ``__contains__()`` method,\n``x in y`` is true if and only if ``y.__contains__(x)`` is true.\n\nFor user-defined classes which do not define ``__contains__()`` and do\ndefine ``__getitem__()``, ``x in y`` is true if and only if there is a\nnon-negative integer index *i* such that ``x == y[i]``, and all lower\ninteger indices do not raise ``IndexError`` exception.  (If any other\nexception is raised, it is as if ``in`` raised that exception).\n\nThe operator ``not in`` is defined to have the inverse true value of\n``in``.\n\nThe operators ``is`` and ``is not`` test for object identity: ``x is\ny`` is true if and only if *x* and *y* are the same object.  ``x is\nnot y`` yields the inverse truth value. [6]\n',
+ 'import': '\nThe ``import`` statement\n************************\n\n   import_stmt     ::= "import" module ["as" name] ( "," module ["as" name] )*\n                   | "from" relative_module "import" identifier ["as" name]\n                   ( "," identifier ["as" name] )*\n                   | "from" relative_module "import" "(" identifier ["as" name]\n                   ( "," identifier ["as" name] )* [","] ")"\n                   | "from" module "import" "*"\n   module          ::= (identifier ".")* identifier\n   relative_module ::= "."* module | "."+\n   name            ::= identifier\n\nImport statements are executed in two steps: (1) find a module, and\ninitialize it if necessary; (2) define a name or names in the local\nnamespace (of the scope where the ``import`` statement occurs). The\nstatement comes in two forms differing on whether it uses the ``from``\nkeyword. The first form (without ``from``) repeats these steps for\neach identifier in the list. The form with ``from`` performs step (1)\nonce, and then performs step (2) repeatedly. For a reference\nimplementation of step (1), see the ``importlib`` module.\n\nTo understand how step (1) occurs, one must first understand how\nPython handles hierarchical naming of modules. To help organize\nmodules and provide a hierarchy in naming, Python has a concept of\npackages. A package can contain other packages and modules while\nmodules cannot contain other modules or packages. From a file system\nperspective, packages are directories and modules are files. The\noriginal specification for packages is still available to read,\nalthough minor details have changed since the writing of that\ndocument.\n\nOnce the name of the module is known (unless otherwise specified, the\nterm "module" will refer to both packages and modules), searching for\nthe module or package can begin. The first place checked is\n``sys.modules``, the cache of all modules that have been imported\npreviously. If the module is found there then it is used in step (2)\nof import unless ``None`` is found in ``sys.modules``, in which case\n``ImportError`` is raised.\n\nIf the module is not found in the cache, then ``sys.meta_path`` is\nsearched (the specification for ``sys.meta_path`` can be found in\n**PEP 302**). The object is a list of *finder* objects which are\nqueried in order as to whether they know how to load the module by\ncalling their ``find_module()`` method with the name of the module. If\nthe module happens to be contained within a package (as denoted by the\nexistence of a dot in the name), then a second argument to\n``find_module()`` is given as the value of the ``__path__`` attribute\nfrom the parent package (everything up to the last dot in the name of\nthe module being imported). If a finder can find the module it returns\na *loader* (discussed later) or returns ``None``.\n\nIf none of the finders on ``sys.meta_path`` are able to find the\nmodule then some implicitly defined finders are queried.\nImplementations of Python vary in what implicit meta path finders are\ndefined. The one they all do define, though, is one that handles\n``sys.path_hooks``, ``sys.path_importer_cache``, and ``sys.path``.\n\nThe implicit finder searches for the requested module in the "paths"\nspecified in one of two places ("paths" do not have to be file system\npaths). If the module being imported is supposed to be contained\nwithin a package then the second argument passed to ``find_module()``,\n``__path__`` on the parent package, is used as the source of paths. If\nthe module is not contained in a package then ``sys.path`` is used as\nthe source of paths.\n\nOnce the source of paths is chosen it is iterated over to find a\nfinder that can handle that path. The dict at\n``sys.path_importer_cache`` caches finders for paths and is checked\nfor a finder. If the path does not have a finder cached then\n``sys.path_hooks`` is searched by calling each object in the list with\na single argument of the path, returning a finder or raises\n``ImportError``. If a finder is returned then it is cached in\n``sys.path_importer_cache`` and then used for that path entry. If no\nfinder can be found but the path exists then a value of ``None`` is\nstored in ``sys.path_importer_cache`` to signify that an implicit,\nfile-based finder that handles modules stored as individual files\nshould be used for that path. If the path does not exist then a finder\nwhich always returns ``None`` is placed in the cache for the path.\n\nIf no finder can find the module then ``ImportError`` is raised.\nOtherwise some finder returned a loader whose ``load_module()`` method\nis called with the name of the module to load (see **PEP 302** for the\noriginal definition of loaders). A loader has several responsibilities\nto perform on a module it loads. First, if the module already exists\nin ``sys.modules`` (a possibility if the loader is called outside of\nthe import machinery) then it is to use that module for initialization\nand not a new module. But if the module does not exist in\n``sys.modules`` then it is to be added to that dict before\ninitialization begins. If an error occurs during loading of the module\nand it was added to ``sys.modules`` it is to be removed from the dict.\nIf an error occurs but the module was already in ``sys.modules`` it is\nleft in the dict.\n\nThe loader must set several attributes on the module. ``__name__`` is\nto be set to the name of the module. ``__file__`` is to be the "path"\nto the file unless the module is built-in (and thus listed in\n``sys.builtin_module_names``) in which case the attribute is not set.\nIf what is being imported is a package then ``__path__`` is to be set\nto a list of paths to be searched when looking for modules and\npackages contained within the package being imported. ``__package__``\nis optional but should be set to the name of package that contains the\nmodule or package (the empty string is used for module not contained\nin a package). ``__loader__`` is also optional but should be set to\nthe loader object that is loading the module.\n\nIf an error occurs during loading then the loader raises\n``ImportError`` if some other exception is not already being\npropagated. Otherwise the loader returns the module that was loaded\nand initialized.\n\nWhen step (1) finishes without raising an exception, step (2) can\nbegin.\n\nThe first form of ``import`` statement binds the module name in the\nlocal namespace to the module object, and then goes on to import the\nnext identifier, if any.  If the module name is followed by ``as``,\nthe name following ``as`` is used as the local name for the module.\n\nThe ``from`` form does not bind the module name: it goes through the\nlist of identifiers, looks each one of them up in the module found in\nstep (1), and binds the name in the local namespace to the object thus\nfound.  As with the first form of ``import``, an alternate local name\ncan be supplied by specifying "``as`` localname".  If a name is not\nfound, ``ImportError`` is raised.  If the list of identifiers is\nreplaced by a star (``\'*\'``), all public names defined in the module\nare bound in the local namespace of the ``import`` statement..\n\nThe *public names* defined by a module are determined by checking the\nmodule\'s namespace for a variable named ``__all__``; if defined, it\nmust be a sequence of strings which are names defined or imported by\nthat module.  The names given in ``__all__`` are all considered public\nand are required to exist.  If ``__all__`` is not defined, the set of\npublic names includes all names found in the module\'s namespace which\ndo not begin with an underscore character (``\'_\'``). ``__all__``\nshould contain the entire public API. It is intended to avoid\naccidentally exporting items that are not part of the API (such as\nlibrary modules which were imported and used within the module).\n\nThe ``from`` form with ``*`` may only occur in a module scope.  The\nwild card form of import --- ``import *`` --- is only allowed at the\nmodule level. Attempting to use it in class or function definitions\nwill raise a ``SyntaxError``.\n\nWhen specifying what module to import you do not have to specify the\nabsolute name of the module. When a module or package is contained\nwithin another package it is possible to make a relative import within\nthe same top package without having to mention the package name. By\nusing leading dots in the specified module or package after ``from``\nyou can specify how high to traverse up the current package hierarchy\nwithout specifying exact names. One leading dot means the current\npackage where the module making the import exists. Two dots means up\none package level. Three dots is up two levels, etc. So if you execute\n``from . import mod`` from a module in the ``pkg`` package then you\nwill end up importing ``pkg.mod``. If you execute ``from ..subpkg2\nimprt mod`` from within ``pkg.subpkg1`` you will import\n``pkg.subpkg2.mod``. The specification for relative imports is\ncontained within **PEP 328**.\n\n``importlib.import_module()`` is provided to support applications that\ndetermine which modules need to be loaded dynamically.\n\n\nFuture statements\n=================\n\nA *future statement* is a directive to the compiler that a particular\nmodule should be compiled using syntax or semantics that will be\navailable in a specified future release of Python.  The future\nstatement is intended to ease migration to future versions of Python\nthat introduce incompatible changes to the language.  It allows use of\nthe new features on a per-module basis before the release in which the\nfeature becomes standard.\n\n   future_statement ::= "from" "__future__" "import" feature ["as" name]\n                        ("," feature ["as" name])*\n                        | "from" "__future__" "import" "(" feature ["as" name]\n                        ("," feature ["as" name])* [","] ")"\n   feature          ::= identifier\n   name             ::= identifier\n\nA future statement must appear near the top of the module.  The only\nlines that can appear before a future statement are:\n\n* the module docstring (if any),\n\n* comments,\n\n* blank lines, and\n\n* other future statements.\n\nThe features recognized by Python 3.0 are ``absolute_import``,\n``division``, ``generators``, ``unicode_literals``,\n``print_function``, ``nested_scopes`` and ``with_statement``.  They\nare all redundant because they are always enabled, and only kept for\nbackwards compatibility.\n\nA future statement is recognized and treated specially at compile\ntime: Changes to the semantics of core constructs are often\nimplemented by generating different code.  It may even be the case\nthat a new feature introduces new incompatible syntax (such as a new\nreserved word), in which case the compiler may need to parse the\nmodule differently.  Such decisions cannot be pushed off until\nruntime.\n\nFor any given release, the compiler knows which feature names have\nbeen defined, and raises a compile-time error if a future statement\ncontains a feature not known to it.\n\nThe direct runtime semantics are the same as for any import statement:\nthere is a standard module ``__future__``, described later, and it\nwill be imported in the usual way at the time the future statement is\nexecuted.\n\nThe interesting runtime semantics depend on the specific feature\nenabled by the future statement.\n\nNote that there is nothing special about the statement:\n\n   import __future__ [as name]\n\nThat is not a future statement; it\'s an ordinary import statement with\nno special semantics or syntax restrictions.\n\nCode compiled by calls to the built-in functions ``exec()`` and\n``compile()`` that occur in a module ``M`` containing a future\nstatement will, by default, use the new syntax or semantics associated\nwith the future statement.  This can be controlled by optional\narguments to ``compile()`` --- see the documentation of that function\nfor details.\n\nA future statement typed at an interactive interpreter prompt will\ntake effect for the rest of the interpreter session.  If an\ninterpreter is started with the *-i* option, is passed a script name\nto execute, and the script includes a future statement, it will be in\neffect in the interactive session started after the script is\nexecuted.\n\nSee also:\n\n   **PEP 236** - Back to the __future__\n      The original proposal for the __future__ mechanism.\n',
+ 'in': '\nComparisons\n***********\n\nUnlike C, all comparison operations in Python have the same priority,\nwhich is lower than that of any arithmetic, shifting or bitwise\noperation.  Also unlike C, expressions like ``a < b < c`` have the\ninterpretation that is conventional in mathematics:\n\n   comparison    ::= or_expr ( comp_operator or_expr )*\n   comp_operator ::= "<" | ">" | "==" | ">=" | "<=" | "!="\n                     | "is" ["not"] | ["not"] "in"\n\nComparisons yield boolean values: ``True`` or ``False``.\n\nComparisons can be chained arbitrarily, e.g., ``x < y <= z`` is\nequivalent to ``x < y and y <= z``, except that ``y`` is evaluated\nonly once (but in both cases ``z`` is not evaluated at all when ``x <\ny`` is found to be false).\n\nFormally, if *a*, *b*, *c*, ..., *y*, *z* are expressions and *op1*,\n*op2*, ..., *opN* are comparison operators, then ``a op1 b op2 c ... y\nopN z`` is equivalent to ``a op1 b and b op2 c and ... y opN z``,\nexcept that each expression is evaluated at most once.\n\nNote that ``a op1 b op2 c`` doesn\'t imply any kind of comparison\nbetween *a* and *c*, so that, e.g., ``x < y > z`` is perfectly legal\n(though perhaps not pretty).\n\nThe operators ``<``, ``>``, ``==``, ``>=``, ``<=``, and ``!=`` compare\nthe values of two objects.  The objects need not have the same type.\nIf both are numbers, they are converted to a common type.  Otherwise,\nthe ``==`` and ``!=`` operators *always* consider objects of different\ntypes to be unequal, while the ``<``, ``>``, ``>=`` and ``<=``\noperators raise a ``TypeError`` when comparing objects of different\ntypes that do not implement these operators for the given pair of\ntypes.  You can control comparison behavior of objects of non-built-in\ntypes by defining rich comparison methods like ``__gt__()``, described\nin section *Basic customization*.\n\nComparison of objects of the same type depends on the type:\n\n* Numbers are compared arithmetically.\n\n* The values ``float(\'NaN\')`` and ``Decimal(\'NaN\')`` are special. The\n  are identical to themselves, ``x is x`` but are not equal to\n  themselves, ``x != x``.  Additionally, comparing any value to a\n  not-a-number value will return ``False``.  For example, both ``3 <\n  float(\'NaN\')`` and ``float(\'NaN\') < 3`` will return ``False``.\n\n* Bytes objects are compared lexicographically using the numeric\n  values of their elements.\n\n* Strings are compared lexicographically using the numeric equivalents\n  (the result of the built-in function ``ord()``) of their characters.\n  [3] String and bytes object can\'t be compared!\n\n* Tuples and lists are compared lexicographically using comparison of\n  corresponding elements.  This means that to compare equal, each\n  element must compare equal and the two sequences must be of the same\n  type and have the same length.\n\n  If not equal, the sequences are ordered the same as their first\n  differing elements.  For example, ``[1,2,x] <= [1,2,y]`` has the\n  same value as ``x <= y``.  If the corresponding element does not\n  exist, the shorter sequence is ordered first (for example, ``[1,2] <\n  [1,2,3]``).\n\n* Mappings (dictionaries) compare equal if and only if they have the\n  same ``(key, value)`` pairs. Order comparisons ``(\'<\', \'<=\', \'>=\',\n  \'>\')`` raise ``TypeError``.\n\n* Sets and frozensets define comparison operators to mean subset and\n  superset tests.  Those relations do not define total orderings (the\n  two sets ``{1,2}`` and {2,3} are not equal, nor subsets of one\n  another, nor supersets of one another).  Accordingly, sets are not\n  appropriate arguments for functions which depend on total ordering.\n  For example, ``min()``, ``max()``, and ``sorted()`` produce\n  undefined results given a list of sets as inputs.\n\n* Most other objects of built-in types compare unequal unless they are\n  the same object; the choice whether one object is considered smaller\n  or larger than another one is made arbitrarily but consistently\n  within one execution of a program.\n\nComparison of objects of the differing types depends on whether either\nof the types provide explicit support for the comparison.  Most\nnumeric types can be compared with one another, but comparisons of\n``float`` and ``Decimal`` are not supported to avoid the inevitable\nconfusion arising from representation issues such as ``float(\'1.1\')``\nbeing inexactly represented and therefore not exactly equal to\n``Decimal(\'1.1\')`` which is.  When cross-type comparison is not\nsupported, the comparison method returns ``NotImplemented``.  This can\ncreate the illusion of non-transitivity between supported cross-type\ncomparisons and unsupported comparisons.  For example, ``Decimal(2) ==\n2`` and *2 == float(2)`* but ``Decimal(2) != float(2)``.\n\nThe operators ``in`` and ``not in`` test for membership.  ``x in s``\nevaluates to true if *x* is a member of *s*, and false otherwise.  ``x\nnot in s`` returns the negation of ``x in s``.  All built-in sequences\nand set types support this as well as dictionary, for which ``in``\ntests whether a the dictionary has a given key. For container types\nsuch as list, tuple, set, frozenset, dict, or collections.deque, the\nexpression ``x in y`` is equivalent to ``any(x is e or x == e for e in\ny)``.\n\nFor the string and bytes types, ``x in y`` is true if and only if *x*\nis a substring of *y*.  An equivalent test is ``y.find(x) != -1``.\nEmpty strings are always considered to be a substring of any other\nstring, so ``"" in "abc"`` will return ``True``.\n\nFor user-defined classes which define the ``__contains__()`` method,\n``x in y`` is true if and only if ``y.__contains__(x)`` is true.\n\nFor user-defined classes which do not define ``__contains__()`` but do\ndefine ``__iter__()``, ``x in y`` is true if some value ``z`` with ``x\n== z`` is produced while iterating over ``y``.  If an exception is\nraised during the iteration, it is as if ``in`` raised that exception.\n\nLastly, the old-style iteration protocol is tried: if a class defines\n``__getitem__()``, ``x in y`` is true if and only if there is a non-\nnegative integer index *i* such that ``x == y[i]``, and all lower\ninteger indices do not raise ``IndexError`` exception.  (If any other\nexception is raised, it is as if ``in`` raised that exception).\n\nThe operator ``not in`` is defined to have the inverse true value of\n``in``.\n\nThe operators ``is`` and ``is not`` test for object identity: ``x is\ny`` is true if and only if *x* and *y* are the same object.  ``x is\nnot y`` yields the inverse truth value. [4]\n',
  'integers': '\nInteger literals\n****************\n\nInteger literals are described by the following lexical definitions:\n\n   integer        ::= decimalinteger | octinteger | hexinteger | bininteger\n   decimalinteger ::= nonzerodigit digit* | "0"+\n   nonzerodigit   ::= "1"..."9"\n   digit          ::= "0"..."9"\n   octinteger     ::= "0" ("o" | "O") octdigit+\n   hexinteger     ::= "0" ("x" | "X") hexdigit+\n   bininteger     ::= "0" ("b" | "B") bindigit+\n   octdigit       ::= "0"..."7"\n   hexdigit       ::= digit | "a"..."f" | "A"..."F"\n   bindigit       ::= "0" | "1"\n\nThere is no limit for the length of integer literals apart from what\ncan be stored in available memory.\n\nNote that leading zeros in a non-zero decimal number are not allowed.\nThis is for disambiguation with C-style octal literals, which Python\nused before version 3.0.\n\nSome examples of integer literals:\n\n   7     2147483647                        0o177    0b100110111\n   3     79228162514264337593543950336     0o377    0x100000000\n         79228162514264337593543950336              0xdeadbeef\n',
  'lambda': '\nLambdas\n*******\n\n   lambda_form        ::= "lambda" [parameter_list]: expression\n   lambda_form_nocond ::= "lambda" [parameter_list]: expression_nocond\n\nLambda forms (lambda expressions) have the same syntactic position as\nexpressions.  They are a shorthand to create anonymous functions; the\nexpression ``lambda arguments: expression`` yields a function object.\nThe unnamed object behaves like a function object defined with\n\n   def <lambda>(arguments):\n       return expression\n\nSee section *Function definitions* for the syntax of parameter lists.\nNote that functions created with lambda forms cannot contain\nstatements or annotations.\n',
  'lists': '\nList displays\n*************\n\nA list display is a possibly empty series of expressions enclosed in\nsquare brackets:\n\n   list_display ::= "[" [expression_list | comprehension] "]"\n\nA list display yields a new list object, the contents being specified\nby either a list of expressions or a comprehension.  When a comma-\nseparated list of expressions is supplied, its elements are evaluated\nfrom left to right and placed into the list object in that order.\nWhen a comprehension is supplied, the list is constructed from the\nelements resulting from the comprehension.\n',
- 'naming': "\nNaming and binding\n******************\n\n*Names* refer to objects.  Names are introduced by name binding\noperations. Each occurrence of a name in the program text refers to\nthe *binding* of that name established in the innermost function block\ncontaining the use.\n\nA *block* is a piece of Python program text that is executed as a\nunit. The following are blocks: a module, a function body, and a class\ndefinition. Each command typed interactively is a block.  A script\nfile (a file given as standard input to the interpreter or specified\non the interpreter command line the first argument) is a code block.\nA script command (a command specified on the interpreter command line\nwith the '**-c**' option) is a code block.  The string argument passed\nto the built-in functions ``eval()`` and ``exec()`` is a code block.\n\nA code block is executed in an *execution frame*.  A frame contains\nsome administrative information (used for debugging) and determines\nwhere and how execution continues after the code block's execution has\ncompleted.\n\nA *scope* defines the visibility of a name within a block.  If a local\nvariable is defined in a block, its scope includes that block.  If the\ndefinition occurs in a function block, the scope extends to any blocks\ncontained within the defining one, unless a contained block introduces\na different binding for the name.  The scope of names defined in a\nclass block is limited to the class block; it does not extend to the\ncode blocks of methods -- this includes comprehensions and generator\nexpressions since they are implemented using a function scope.  This\nmeans that the following will fail:\n\n   class A:\n       a = 42\n       b = list(a + i for i in range(10))\n\nWhen a name is used in a code block, it is resolved using the nearest\nenclosing scope.  The set of all such scopes visible to a code block\nis called the block's *environment*.\n\nIf a name is bound in a block, it is a local variable of that block,\nunless declared as ``nonlocal``.  If a name is bound at the module\nlevel, it is a global variable.  (The variables of the module code\nblock are local and global.)  If a variable is used in a code block\nbut not defined there, it is a *free variable*.\n\nWhen a name is not found at all, a ``NameError`` exception is raised.\nIf the name refers to a local variable that has not been bound, a\n``UnboundLocalError`` exception is raised.  ``UnboundLocalError`` is a\nsubclass of ``NameError``.\n\nThe following constructs bind names: formal parameters to functions,\n``import`` statements, class and function definitions (these bind the\nclass or function name in the defining block), and targets that are\nidentifiers if occurring in an assignment, ``for`` loop header, or\nafter ``as`` in a ``with`` statement or :keyword.`except` clause. The\n``import`` statement of the form ``from ... import *`` binds all names\ndefined in the imported module, except those beginning with an\nunderscore.  This form may only be used at the module level.\n\nA target occurring in a ``del`` statement is also considered bound for\nthis purpose (though the actual semantics are to unbind the name).  It\nis illegal to unbind a name that is referenced by an enclosing scope;\nthe compiler will report a ``SyntaxError``.\n\nEach assignment or import statement occurs within a block defined by a\nclass or function definition or at the module level (the top-level\ncode block).\n\nIf a name binding operation occurs anywhere within a code block, all\nuses of the name within the block are treated as references to the\ncurrent block.  This can lead to errors when a name is used within a\nblock before it is bound.  This rule is subtle.  Python lacks\ndeclarations and allows name binding operations to occur anywhere\nwithin a code block.  The local variables of a code block can be\ndetermined by scanning the entire text of the block for name binding\noperations.\n\nIf the ``global`` statement occurs within a block, all uses of the\nname specified in the statement refer to the binding of that name in\nthe top-level namespace.  Names are resolved in the top-level\nnamespace by searching the global namespace, i.e. the namespace of the\nmodule containing the code block, and the builtin namespace, the\nnamespace of the module ``builtins``.  The global namespace is\nsearched first.  If the name is not found there, the builtin namespace\nis searched.  The global statement must precede all uses of the name.\n\nThe built-in namespace associated with the execution of a code block\nis actually found by looking up the name ``__builtins__`` in its\nglobal namespace; this should be a dictionary or a module (in the\nlatter case the module's dictionary is used).  By default, when in the\n``__main__`` module, ``__builtins__`` is the built-in module\n``builtins``; when in any other module, ``__builtins__`` is an alias\nfor the dictionary of the ``builtins`` module itself.\n``__builtins__`` can be set to a user-created dictionary to create a\nweak form of restricted execution.\n\nNote: Users should not touch ``__builtins__``; it is strictly an\n  implementation detail.  Users wanting to override values in the\n  built-in namespace should ``import`` the ``builtins`` module and\n  modify its attributes appropriately.\n\nThe namespace for a module is automatically created the first time a\nmodule is imported.  The main module for a script is always called\n``__main__``.\n\nThe global statement has the same scope as a name binding operation in\nthe same block.  If the nearest enclosing scope for a free variable\ncontains a global statement, the free variable is treated as a global.\n\nA class definition is an executable statement that may use and define\nnames. These references follow the normal rules for name resolution.\nThe namespace of the class definition becomes the attribute dictionary\nof the class.  Names defined at the class scope are not visible in\nmethods.\n\n\nInteraction with dynamic features\n=================================\n\nThere are several cases where Python statements are illegal when used\nin conjunction with nested scopes that contain free variables.\n\nIf a variable is referenced in an enclosing scope, it is illegal to\ndelete the name.  An error will be reported at compile time.\n\nIf the wild card form of import --- ``import *`` --- is used in a\nfunction and the function contains or is a nested block with free\nvariables, the compiler will raise a ``SyntaxError``.\n\nThe ``eval()`` and ``exec()`` functions do not have access to the full\nenvironment for resolving names.  Names may be resolved in the local\nand global namespaces of the caller.  Free variables are not resolved\nin the nearest enclosing namespace, but in the global namespace.  [1]\nThe ``exec()`` and ``eval()`` functions have optional arguments to\noverride the global and local namespace.  If only one namespace is\nspecified, it is used for both.\n",
+ 'naming': "\nNaming and binding\n******************\n\n*Names* refer to objects.  Names are introduced by name binding\noperations. Each occurrence of a name in the program text refers to\nthe *binding* of that name established in the innermost function block\ncontaining the use.\n\nA *block* is a piece of Python program text that is executed as a\nunit. The following are blocks: a module, a function body, and a class\ndefinition. Each command typed interactively is a block.  A script\nfile (a file given as standard input to the interpreter or specified\non the interpreter command line the first argument) is a code block.\nA script command (a command specified on the interpreter command line\nwith the '**-c**' option) is a code block.  The string argument passed\nto the built-in functions ``eval()`` and ``exec()`` is a code block.\n\nA code block is executed in an *execution frame*.  A frame contains\nsome administrative information (used for debugging) and determines\nwhere and how execution continues after the code block's execution has\ncompleted.\n\nA *scope* defines the visibility of a name within a block.  If a local\nvariable is defined in a block, its scope includes that block.  If the\ndefinition occurs in a function block, the scope extends to any blocks\ncontained within the defining one, unless a contained block introduces\na different binding for the name.  The scope of names defined in a\nclass block is limited to the class block; it does not extend to the\ncode blocks of methods -- this includes comprehensions and generator\nexpressions since they are implemented using a function scope.  This\nmeans that the following will fail:\n\n   class A:\n       a = 42\n       b = list(a + i for i in range(10))\n\nWhen a name is used in a code block, it is resolved using the nearest\nenclosing scope.  The set of all such scopes visible to a code block\nis called the block's *environment*.\n\nIf a name is bound in a block, it is a local variable of that block,\nunless declared as ``nonlocal``.  If a name is bound at the module\nlevel, it is a global variable.  (The variables of the module code\nblock are local and global.)  If a variable is used in a code block\nbut not defined there, it is a *free variable*.\n\nWhen a name is not found at all, a ``NameError`` exception is raised.\nIf the name refers to a local variable that has not been bound, a\n``UnboundLocalError`` exception is raised.  ``UnboundLocalError`` is a\nsubclass of ``NameError``.\n\nThe following constructs bind names: formal parameters to functions,\n``import`` statements, class and function definitions (these bind the\nclass or function name in the defining block), and targets that are\nidentifiers if occurring in an assignment, ``for`` loop header, or\nafter ``as`` in a ``with`` statement or :keyword.`except` clause. The\n``import`` statement of the form ``from ... import *`` binds all names\ndefined in the imported module, except those beginning with an\nunderscore.  This form may only be used at the module level.\n\nA target occurring in a ``del`` statement is also considered bound for\nthis purpose (though the actual semantics are to unbind the name).  It\nis illegal to unbind a name that is referenced by an enclosing scope;\nthe compiler will report a ``SyntaxError``.\n\nEach assignment or import statement occurs within a block defined by a\nclass or function definition or at the module level (the top-level\ncode block).\n\nIf a name binding operation occurs anywhere within a code block, all\nuses of the name within the block are treated as references to the\ncurrent block.  This can lead to errors when a name is used within a\nblock before it is bound.  This rule is subtle.  Python lacks\ndeclarations and allows name binding operations to occur anywhere\nwithin a code block.  The local variables of a code block can be\ndetermined by scanning the entire text of the block for name binding\noperations.\n\nIf the ``global`` statement occurs within a block, all uses of the\nname specified in the statement refer to the binding of that name in\nthe top-level namespace.  Names are resolved in the top-level\nnamespace by searching the global namespace, i.e. the namespace of the\nmodule containing the code block, and the builtins namespace, the\nnamespace of the module ``builtins``.  The global namespace is\nsearched first.  If the name is not found there, the builtins\nnamespace is searched.  The global statement must precede all uses of\nthe name.\n\nThe builtins namespace associated with the execution of a code block\nis actually found by looking up the name ``__builtins__`` in its\nglobal namespace; this should be a dictionary or a module (in the\nlatter case the module's dictionary is used).  By default, when in the\n``__main__`` module, ``__builtins__`` is the built-in module\n``builtins``; when in any other module, ``__builtins__`` is an alias\nfor the dictionary of the ``builtins`` module itself.\n``__builtins__`` can be set to a user-created dictionary to create a\nweak form of restricted execution.\n\n**CPython implementation detail:** Users should not touch\n``__builtins__``; it is strictly an implementation detail.  Users\nwanting to override values in the builtins namespace should ``import``\nthe ``builtins`` module and modify its attributes appropriately.\n\nThe namespace for a module is automatically created the first time a\nmodule is imported.  The main module for a script is always called\n``__main__``.\n\nThe global statement has the same scope as a name binding operation in\nthe same block.  If the nearest enclosing scope for a free variable\ncontains a global statement, the free variable is treated as a global.\n\nA class definition is an executable statement that may use and define\nnames. These references follow the normal rules for name resolution.\nThe namespace of the class definition becomes the attribute dictionary\nof the class.  Names defined at the class scope are not visible in\nmethods.\n\n\nInteraction with dynamic features\n=================================\n\nThere are several cases where Python statements are illegal when used\nin conjunction with nested scopes that contain free variables.\n\nIf a variable is referenced in an enclosing scope, it is illegal to\ndelete the name.  An error will be reported at compile time.\n\nIf the wild card form of import --- ``import *`` --- is used in a\nfunction and the function contains or is a nested block with free\nvariables, the compiler will raise a ``SyntaxError``.\n\nThe ``eval()`` and ``exec()`` functions do not have access to the full\nenvironment for resolving names.  Names may be resolved in the local\nand global namespaces of the caller.  Free variables are not resolved\nin the nearest enclosing namespace, but in the global namespace.  [1]\nThe ``exec()`` and ``eval()`` functions have optional arguments to\noverride the global and local namespace.  If only one namespace is\nspecified, it is used for both.\n",
  'numbers': "\nNumeric literals\n****************\n\nThere are three types of numeric literals: integers, floating point\nnumbers, and imaginary numbers.  There are no complex literals\n(complex numbers can be formed by adding a real number and an\nimaginary number).\n\nNote that numeric literals do not include a sign; a phrase like ``-1``\nis actually an expression composed of the unary operator '``-``' and\nthe literal ``1``.\n",
- 'numeric-types': "\nEmulating numeric types\n***********************\n\nThe following methods can be defined to emulate numeric objects.\nMethods corresponding to operations that are not supported by the\nparticular kind of number implemented (e.g., bitwise operations for\nnon-integral numbers) should be left undefined.\n\nobject.__add__(self, other)\nobject.__sub__(self, other)\nobject.__mul__(self, other)\nobject.__truediv__(self, other)\nobject.__floordiv__(self, other)\nobject.__mod__(self, other)\nobject.__divmod__(self, other)\nobject.__pow__(self, other[, modulo])\nobject.__lshift__(self, other)\nobject.__rshift__(self, other)\nobject.__and__(self, other)\nobject.__xor__(self, other)\nobject.__or__(self, other)\n\n   These methods are called to implement the binary arithmetic\n   operations (``+``, ``-``, ``*``, ``/``, ``//``, ``%``,\n   ``divmod()``, ``pow()``, ``**``, ``<<``, ``>>``, ``&``, ``^``,\n   ``|``).  For instance, to evaluate the expression ``x + y``, where\n   *x* is an instance of a class that has an ``__add__()`` method,\n   ``x.__add__(y)`` is called.  The ``__divmod__()`` method should be\n   the equivalent to using ``__floordiv__()`` and ``__mod__()``; it\n   should not be related to ``__truediv__()``.  Note that\n   ``__pow__()`` should be defined to accept an optional third\n   argument if the ternary version of the built-in ``pow()`` function\n   is to be supported.\n\n   If one of those methods does not support the operation with the\n   supplied arguments, it should return ``NotImplemented``.\n\nobject.__radd__(self, other)\nobject.__rsub__(self, other)\nobject.__rmul__(self, other)\nobject.__rtruediv__(self, other)\nobject.__rfloordiv__(self, other)\nobject.__rmod__(self, other)\nobject.__rdivmod__(self, other)\nobject.__rpow__(self, other)\nobject.__rlshift__(self, other)\nobject.__rrshift__(self, other)\nobject.__rand__(self, other)\nobject.__rxor__(self, other)\nobject.__ror__(self, other)\n\n   These methods are called to implement the binary arithmetic\n   operations (``+``, ``-``, ``*``, ``/``, ``//``, ``%``,\n   ``divmod()``, ``pow()``, ``**``, ``<<``, ``>>``, ``&``, ``^``,\n   ``|``) with reflected (swapped) operands. These functions are only\n   called if the left operand does not support the corresponding\n   operation and the operands are of different types. [3]  For\n   instance, to evaluate the expression ``x - y``, where *y* is an\n   instance of a class that has an ``__rsub__()`` method,\n   ``y.__rsub__(x)`` is called if ``x.__sub__(y)`` returns\n   *NotImplemented*.\n\n   Note that ternary ``pow()`` will not try calling ``__rpow__()``\n   (the coercion rules would become too complicated).\n\n   Note: If the right operand's type is a subclass of the left operand's\n     type and that subclass provides the reflected method for the\n     operation, this method will be called before the left operand's\n     non-reflected method.  This behavior allows subclasses to\n     override their ancestors' operations.\n\nobject.__iadd__(self, other)\nobject.__isub__(self, other)\nobject.__imul__(self, other)\nobject.__itruediv__(self, other)\nobject.__ifloordiv__(self, other)\nobject.__imod__(self, other)\nobject.__ipow__(self, other[, modulo])\nobject.__ilshift__(self, other)\nobject.__irshift__(self, other)\nobject.__iand__(self, other)\nobject.__ixor__(self, other)\nobject.__ior__(self, other)\n\n   These methods are called to implement the augmented arithmetic\n   assignments (``+=``, ``-=``, ``*=``, ``/=``, ``//=``, ``%=``,\n   ``**=``, ``<<=``, ``>>=``, ``&=``, ``^=``, ``|=``).  These methods\n   should attempt to do the operation in-place (modifying *self*) and\n   return the result (which could be, but does not have to be,\n   *self*).  If a specific method is not defined, the augmented\n   assignment falls back to the normal methods.  For instance, to\n   execute the statement ``x += y``, where *x* is an instance of a\n   class that has an ``__iadd__()`` method, ``x.__iadd__(y)`` is\n   called.  If *x* is an instance of a class that does not define a\n   ``__iadd__()`` method, ``x.__add__(y)`` and ``y.__radd__(x)`` are\n   considered, as with the evaluation of ``x + y``.\n\nobject.__neg__(self)\nobject.__pos__(self)\nobject.__abs__(self)\nobject.__invert__(self)\n\n   Called to implement the unary arithmetic operations (``-``, ``+``,\n   ``abs()`` and ``~``).\n\nobject.__complex__(self)\nobject.__int__(self)\nobject.__float__(self)\nobject.__round__(self[, n])\n\n   Called to implement the built-in functions ``complex()``,\n   ``int()``, ``float()`` and ``round()``.  Should return a value of\n   the appropriate type.\n\nobject.__index__(self)\n\n   Called to implement ``operator.index()``.  Also called whenever\n   Python needs an integer object (such as in slicing, or in the\n   built-in ``bin()``, ``hex()`` and ``oct()`` functions). Must return\n   an integer.\n",
- 'objects': '\nObjects, values and types\n*************************\n\n*Objects* are Python\'s abstraction for data.  All data in a Python\nprogram is represented by objects or by relations between objects. (In\na sense, and in conformance to Von Neumann\'s model of a "stored\nprogram computer," code is also represented by objects.)\n\nEvery object has an identity, a type and a value.  An object\'s\n*identity* never changes once it has been created; you may think of it\nas the object\'s address in memory.  The \'``is``\' operator compares the\nidentity of two objects; the ``id()`` function returns an integer\nrepresenting its identity (currently implemented as its address). An\nobject\'s *type* is also unchangeable. [1] An object\'s type determines\nthe operations that the object supports (e.g., "does it have a\nlength?") and also defines the possible values for objects of that\ntype.  The ``type()`` function returns an object\'s type (which is an\nobject itself).  The *value* of some objects can change.  Objects\nwhose value can change are said to be *mutable*; objects whose value\nis unchangeable once they are created are called *immutable*. (The\nvalue of an immutable container object that contains a reference to a\nmutable object can change when the latter\'s value is changed; however\nthe container is still considered immutable, because the collection of\nobjects it contains cannot be changed.  So, immutability is not\nstrictly the same as having an unchangeable value, it is more subtle.)\nAn object\'s mutability is determined by its type; for instance,\nnumbers, strings and tuples are immutable, while dictionaries and\nlists are mutable.\n\nObjects are never explicitly destroyed; however, when they become\nunreachable they may be garbage-collected.  An implementation is\nallowed to postpone garbage collection or omit it altogether --- it is\na matter of implementation quality how garbage collection is\nimplemented, as long as no objects are collected that are still\nreachable.  (Implementation note: CPython currently uses a reference-\ncounting scheme with (optional) delayed detection of cyclically linked\ngarbage, which collects most objects as soon as they become\nunreachable, but is not guaranteed to collect garbage containing\ncircular references.  See the documentation of the ``gc`` module for\ninformation on controlling the collection of cyclic garbage.  Other\nimplementations act differently and CPython may change.)\n\nNote that the use of the implementation\'s tracing or debugging\nfacilities may keep objects alive that would normally be collectable.\nAlso note that catching an exception with a \'``try``...``except``\'\nstatement may keep objects alive.\n\nSome objects contain references to "external" resources such as open\nfiles or windows.  It is understood that these resources are freed\nwhen the object is garbage-collected, but since garbage collection is\nnot guaranteed to happen, such objects also provide an explicit way to\nrelease the external resource, usually a ``close()`` method. Programs\nare strongly recommended to explicitly close such objects.  The\n\'``try``...``finally``\' statement and the \'``with``\' statement provide\nconvenient ways to do this.\n\nSome objects contain references to other objects; these are called\n*containers*. Examples of containers are tuples, lists and\ndictionaries.  The references are part of a container\'s value.  In\nmost cases, when we talk about the value of a container, we imply the\nvalues, not the identities of the contained objects; however, when we\ntalk about the mutability of a container, only the identities of the\nimmediately contained objects are implied.  So, if an immutable\ncontainer (like a tuple) contains a reference to a mutable object, its\nvalue changes if that mutable object is changed.\n\nTypes affect almost all aspects of object behavior.  Even the\nimportance of object identity is affected in some sense: for immutable\ntypes, operations that compute new values may actually return a\nreference to any existing object with the same type and value, while\nfor mutable objects this is not allowed.  E.g., after ``a = 1; b =\n1``, ``a`` and ``b`` may or may not refer to the same object with the\nvalue one, depending on the implementation, but after ``c = []; d =\n[]``, ``c`` and ``d`` are guaranteed to refer to two different,\nunique, newly created empty lists. (Note that ``c = d = []`` assigns\nthe same object to both ``c`` and ``d``.)\n',
- 'operator-summary': '\nSummary\n*******\n\nThe following table summarizes the operator precedences in Python,\nfrom lowest precedence (least binding) to highest precedence (most\nbinding).  Operators in the same box have the same precedence.  Unless\nthe syntax is explicitly given, operators are binary.  Operators in\nthe same box group left to right (except for comparisons, including\ntests, which all have the same precedence and chain from left to right\n--- see section *Comparisons* --- and exponentiation, which groups\nfrom right to left).\n\n+-------------------------------------------------+---------------------------------------+\n| Operator                                        | Description                           |\n+=================================================+=======================================+\n| ``lambda``                                      | Lambda expression                     |\n+-------------------------------------------------+---------------------------------------+\n| ``or``                                          | Boolean OR                            |\n+-------------------------------------------------+---------------------------------------+\n| ``and``                                         | Boolean AND                           |\n+-------------------------------------------------+---------------------------------------+\n| ``not`` *x*                                     | Boolean NOT                           |\n+-------------------------------------------------+---------------------------------------+\n| ``in``, ``not`` ``in``, ``is``, ``is not``,     | Comparisons, including membership     |\n| ``<``, ``<=``, ``>``, ``>=``, ``!=``, ``==``    | tests and identity tests,             |\n+-------------------------------------------------+---------------------------------------+\n| ``|``                                           | Bitwise OR                            |\n+-------------------------------------------------+---------------------------------------+\n| ``^``                                           | Bitwise XOR                           |\n+-------------------------------------------------+---------------------------------------+\n| ``&``                                           | Bitwise AND                           |\n+-------------------------------------------------+---------------------------------------+\n| ``<<``, ``>>``                                  | Shifts                                |\n+-------------------------------------------------+---------------------------------------+\n| ``+``, ``-``                                    | Addition and subtraction              |\n+-------------------------------------------------+---------------------------------------+\n| ``*``, ``/``, ``//``, ``%``                     | Multiplication, division, remainder   |\n+-------------------------------------------------+---------------------------------------+\n| ``+x``, ``-x``, ``~x``                          | Positive, negative, bitwise NOT       |\n+-------------------------------------------------+---------------------------------------+\n| ``**``                                          | Exponentiation [7]                    |\n+-------------------------------------------------+---------------------------------------+\n| ``x[index]``, ``x[index:index]``,               | Subscription, slicing, call,          |\n| ``x(arguments...)``, ``x.attribute``            | attribute reference                   |\n+-------------------------------------------------+---------------------------------------+\n| ``(expressions...)``, ``[expressions...]``,     | Binding or tuple display, list        |\n| ``{key:datum...}``,                             | display, dictionary display,          |\n+-------------------------------------------------+---------------------------------------+\n\n-[ Footnotes ]-\n\n[1] While ``abs(x%y) < abs(y)`` is true mathematically, for floats it\n    may not be true numerically due to roundoff.  For example, and\n    assuming a platform on which a Python float is an IEEE 754 double-\n    precision number, in order that ``-1e-100 % 1e100`` have the same\n    sign as ``1e100``, the computed result is ``-1e-100 + 1e100``,\n    which is numerically exactly equal to ``1e100``.  Function\n    ``fmod()`` in the ``math`` module returns a result whose sign\n    matches the sign of the first argument instead, and so returns\n    ``-1e-100`` in this case. Which approach is more appropriate\n    depends on the application.\n\n[2] If x is very close to an exact integer multiple of y, it\'s\n    possible for ``x//y`` to be one larger than ``(x-x%y)//y`` due to\n    rounding.  In such cases, Python returns the latter result, in\n    order to preserve that ``divmod(x,y)[0] * y + x % y`` be very\n    close to ``x``.\n\n[3] While comparisons between strings make sense at the byte level,\n    they may be counter-intuitive to users.  For example, the strings\n    ``"\\u00C7"`` and ``"\\u0327\\u0043"`` compare differently, even\n    though they both represent the same unicode character (LATIN\n    CAPITAL LETTER C WITH CEDILLA).  To compare strings in a human\n    recognizable way, compare using ``unicodedata.normalize()``.\n\n[4] The implementation computes this efficiently, without constructing\n    lists or sorting.\n\n[5] Earlier versions of Python used lexicographic comparison of the\n    sorted (key, value) lists, but this was very expensive for the\n    common case of comparing for equality.  An even earlier version of\n    Python compared dictionaries by identity only, but this caused\n    surprises because people expected to be able to test a dictionary\n    for emptiness by comparing it to ``{}``.\n\n[6] Due to automatic garbage-collection, free lists, and the dynamic\n    nature of descriptors, you may notice seemingly unusual behaviour\n    in certain uses of the ``is`` operator, like those involving\n    comparisons between instance methods, or constants.  Check their\n    documentation for more info.\n\n[7] The power operator ``**`` binds less tightly than an arithmetic or\n    bitwise unary operator on its right, that is, ``2**-1`` is\n    ``0.5``.\n',
+ 'numeric-types': "\nEmulating numeric types\n***********************\n\nThe following methods can be defined to emulate numeric objects.\nMethods corresponding to operations that are not supported by the\nparticular kind of number implemented (e.g., bitwise operations for\nnon-integral numbers) should be left undefined.\n\nobject.__add__(self, other)\nobject.__sub__(self, other)\nobject.__mul__(self, other)\nobject.__truediv__(self, other)\nobject.__floordiv__(self, other)\nobject.__mod__(self, other)\nobject.__divmod__(self, other)\nobject.__pow__(self, other[, modulo])\nobject.__lshift__(self, other)\nobject.__rshift__(self, other)\nobject.__and__(self, other)\nobject.__xor__(self, other)\nobject.__or__(self, other)\n\n   These methods are called to implement the binary arithmetic\n   operations (``+``, ``-``, ``*``, ``/``, ``//``, ``%``,\n   ``divmod()``, ``pow()``, ``**``, ``<<``, ``>>``, ``&``, ``^``,\n   ``|``).  For instance, to evaluate the expression ``x + y``, where\n   *x* is an instance of a class that has an ``__add__()`` method,\n   ``x.__add__(y)`` is called.  The ``__divmod__()`` method should be\n   the equivalent to using ``__floordiv__()`` and ``__mod__()``; it\n   should not be related to ``__truediv__()``.  Note that\n   ``__pow__()`` should be defined to accept an optional third\n   argument if the ternary version of the built-in ``pow()`` function\n   is to be supported.\n\n   If one of those methods does not support the operation with the\n   supplied arguments, it should return ``NotImplemented``.\n\nobject.__radd__(self, other)\nobject.__rsub__(self, other)\nobject.__rmul__(self, other)\nobject.__rtruediv__(self, other)\nobject.__rfloordiv__(self, other)\nobject.__rmod__(self, other)\nobject.__rdivmod__(self, other)\nobject.__rpow__(self, other)\nobject.__rlshift__(self, other)\nobject.__rrshift__(self, other)\nobject.__rand__(self, other)\nobject.__rxor__(self, other)\nobject.__ror__(self, other)\n\n   These methods are called to implement the binary arithmetic\n   operations (``+``, ``-``, ``*``, ``/``, ``//``, ``%``,\n   ``divmod()``, ``pow()``, ``**``, ``<<``, ``>>``, ``&``, ``^``,\n   ``|``) with reflected (swapped) operands. These functions are only\n   called if the left operand does not support the corresponding\n   operation and the operands are of different types. [2]  For\n   instance, to evaluate the expression ``x - y``, where *y* is an\n   instance of a class that has an ``__rsub__()`` method,\n   ``y.__rsub__(x)`` is called if ``x.__sub__(y)`` returns\n   *NotImplemented*.\n\n   Note that ternary ``pow()`` will not try calling ``__rpow__()``\n   (the coercion rules would become too complicated).\n\n   Note: If the right operand's type is a subclass of the left operand's\n     type and that subclass provides the reflected method for the\n     operation, this method will be called before the left operand's\n     non-reflected method.  This behavior allows subclasses to\n     override their ancestors' operations.\n\nobject.__iadd__(self, other)\nobject.__isub__(self, other)\nobject.__imul__(self, other)\nobject.__itruediv__(self, other)\nobject.__ifloordiv__(self, other)\nobject.__imod__(self, other)\nobject.__ipow__(self, other[, modulo])\nobject.__ilshift__(self, other)\nobject.__irshift__(self, other)\nobject.__iand__(self, other)\nobject.__ixor__(self, other)\nobject.__ior__(self, other)\n\n   These methods are called to implement the augmented arithmetic\n   assignments (``+=``, ``-=``, ``*=``, ``/=``, ``//=``, ``%=``,\n   ``**=``, ``<<=``, ``>>=``, ``&=``, ``^=``, ``|=``).  These methods\n   should attempt to do the operation in-place (modifying *self*) and\n   return the result (which could be, but does not have to be,\n   *self*).  If a specific method is not defined, the augmented\n   assignment falls back to the normal methods.  For instance, to\n   execute the statement ``x += y``, where *x* is an instance of a\n   class that has an ``__iadd__()`` method, ``x.__iadd__(y)`` is\n   called.  If *x* is an instance of a class that does not define a\n   ``__iadd__()`` method, ``x.__add__(y)`` and ``y.__radd__(x)`` are\n   considered, as with the evaluation of ``x + y``.\n\nobject.__neg__(self)\nobject.__pos__(self)\nobject.__abs__(self)\nobject.__invert__(self)\n\n   Called to implement the unary arithmetic operations (``-``, ``+``,\n   ``abs()`` and ``~``).\n\nobject.__complex__(self)\nobject.__int__(self)\nobject.__float__(self)\nobject.__round__(self[, n])\n\n   Called to implement the built-in functions ``complex()``,\n   ``int()``, ``float()`` and ``round()``.  Should return a value of\n   the appropriate type.\n\nobject.__index__(self)\n\n   Called to implement ``operator.index()``.  Also called whenever\n   Python needs an integer object (such as in slicing, or in the\n   built-in ``bin()``, ``hex()`` and ``oct()`` functions). Must return\n   an integer.\n",
+ 'objects': '\nObjects, values and types\n*************************\n\n*Objects* are Python\'s abstraction for data.  All data in a Python\nprogram is represented by objects or by relations between objects. (In\na sense, and in conformance to Von Neumann\'s model of a "stored\nprogram computer," code is also represented by objects.)\n\nEvery object has an identity, a type and a value.  An object\'s\n*identity* never changes once it has been created; you may think of it\nas the object\'s address in memory.  The \'``is``\' operator compares the\nidentity of two objects; the ``id()`` function returns an integer\nrepresenting its identity (currently implemented as its address). An\nobject\'s *type* is also unchangeable. [1] An object\'s type determines\nthe operations that the object supports (e.g., "does it have a\nlength?") and also defines the possible values for objects of that\ntype.  The ``type()`` function returns an object\'s type (which is an\nobject itself).  The *value* of some objects can change.  Objects\nwhose value can change are said to be *mutable*; objects whose value\nis unchangeable once they are created are called *immutable*. (The\nvalue of an immutable container object that contains a reference to a\nmutable object can change when the latter\'s value is changed; however\nthe container is still considered immutable, because the collection of\nobjects it contains cannot be changed.  So, immutability is not\nstrictly the same as having an unchangeable value, it is more subtle.)\nAn object\'s mutability is determined by its type; for instance,\nnumbers, strings and tuples are immutable, while dictionaries and\nlists are mutable.\n\nObjects are never explicitly destroyed; however, when they become\nunreachable they may be garbage-collected.  An implementation is\nallowed to postpone garbage collection or omit it altogether --- it is\na matter of implementation quality how garbage collection is\nimplemented, as long as no objects are collected that are still\nreachable.\n\n**CPython implementation detail:** CPython currently uses a reference-\ncounting scheme with (optional) delayed detection of cyclically linked\ngarbage, which collects most objects as soon as they become\nunreachable, but is not guaranteed to collect garbage containing\ncircular references.  See the documentation of the ``gc`` module for\ninformation on controlling the collection of cyclic garbage. Other\nimplementations act differently and CPython may change.\n\nNote that the use of the implementation\'s tracing or debugging\nfacilities may keep objects alive that would normally be collectable.\nAlso note that catching an exception with a \'``try``...``except``\'\nstatement may keep objects alive.\n\nSome objects contain references to "external" resources such as open\nfiles or windows.  It is understood that these resources are freed\nwhen the object is garbage-collected, but since garbage collection is\nnot guaranteed to happen, such objects also provide an explicit way to\nrelease the external resource, usually a ``close()`` method. Programs\nare strongly recommended to explicitly close such objects.  The\n\'``try``...``finally``\' statement and the \'``with``\' statement provide\nconvenient ways to do this.\n\nSome objects contain references to other objects; these are called\n*containers*. Examples of containers are tuples, lists and\ndictionaries.  The references are part of a container\'s value.  In\nmost cases, when we talk about the value of a container, we imply the\nvalues, not the identities of the contained objects; however, when we\ntalk about the mutability of a container, only the identities of the\nimmediately contained objects are implied.  So, if an immutable\ncontainer (like a tuple) contains a reference to a mutable object, its\nvalue changes if that mutable object is changed.\n\nTypes affect almost all aspects of object behavior.  Even the\nimportance of object identity is affected in some sense: for immutable\ntypes, operations that compute new values may actually return a\nreference to any existing object with the same type and value, while\nfor mutable objects this is not allowed.  E.g., after ``a = 1; b =\n1``, ``a`` and ``b`` may or may not refer to the same object with the\nvalue one, depending on the implementation, but after ``c = []; d =\n[]``, ``c`` and ``d`` are guaranteed to refer to two different,\nunique, newly created empty lists. (Note that ``c = d = []`` assigns\nthe same object to both ``c`` and ``d``.)\n',
+ 'operator-summary': '\nSummary\n*******\n\nThe following table summarizes the operator precedences in Python,\nfrom lowest precedence (least binding) to highest precedence (most\nbinding).  Operators in the same box have the same precedence.  Unless\nthe syntax is explicitly given, operators are binary.  Operators in\nthe same box group left to right (except for comparisons, including\ntests, which all have the same precedence and chain from left to right\n--- see section *Comparisons* --- and exponentiation, which groups\nfrom right to left).\n\n+-------------------------------------------------+---------------------------------------+\n| Operator                                        | Description                           |\n+=================================================+=======================================+\n| ``lambda``                                      | Lambda expression                     |\n+-------------------------------------------------+---------------------------------------+\n| ``if`` -- ``else``                              | Conditional expression                |\n+-------------------------------------------------+---------------------------------------+\n| ``or``                                          | Boolean OR                            |\n+-------------------------------------------------+---------------------------------------+\n| ``and``                                         | Boolean AND                           |\n+-------------------------------------------------+---------------------------------------+\n| ``not`` *x*                                     | Boolean NOT                           |\n+-------------------------------------------------+---------------------------------------+\n| ``in``, ``not`` ``in``, ``is``, ``is not``,     | Comparisons, including membership     |\n| ``<``, ``<=``, ``>``, ``>=``, ``!=``, ``==``    | tests and identity tests,             |\n+-------------------------------------------------+---------------------------------------+\n| ``|``                                           | Bitwise OR                            |\n+-------------------------------------------------+---------------------------------------+\n| ``^``                                           | Bitwise XOR                           |\n+-------------------------------------------------+---------------------------------------+\n| ``&``                                           | Bitwise AND                           |\n+-------------------------------------------------+---------------------------------------+\n| ``<<``, ``>>``                                  | Shifts                                |\n+-------------------------------------------------+---------------------------------------+\n| ``+``, ``-``                                    | Addition and subtraction              |\n+-------------------------------------------------+---------------------------------------+\n| ``*``, ``/``, ``//``, ``%``                     | Multiplication, division, remainder   |\n+-------------------------------------------------+---------------------------------------+\n| ``+x``, ``-x``, ``~x``                          | Positive, negative, bitwise NOT       |\n+-------------------------------------------------+---------------------------------------+\n| ``**``                                          | Exponentiation [5]                    |\n+-------------------------------------------------+---------------------------------------+\n| ``x[index]``, ``x[index:index]``,               | Subscription, slicing, call,          |\n| ``x(arguments...)``, ``x.attribute``            | attribute reference                   |\n+-------------------------------------------------+---------------------------------------+\n| ``(expressions...)``, ``[expressions...]``,     | Binding or tuple display, list        |\n| ``{key:datum...}``,                             | display, dictionary display,          |\n+-------------------------------------------------+---------------------------------------+\n\n-[ Footnotes ]-\n\n[1] While ``abs(x%y) < abs(y)`` is true mathematically, for floats it\n    may not be true numerically due to roundoff.  For example, and\n    assuming a platform on which a Python float is an IEEE 754 double-\n    precision number, in order that ``-1e-100 % 1e100`` have the same\n    sign as ``1e100``, the computed result is ``-1e-100 + 1e100``,\n    which is numerically exactly equal to ``1e100``.  Function\n    ``fmod()`` in the ``math`` module returns a result whose sign\n    matches the sign of the first argument instead, and so returns\n    ``-1e-100`` in this case. Which approach is more appropriate\n    depends on the application.\n\n[2] If x is very close to an exact integer multiple of y, it\'s\n    possible for ``x//y`` to be one larger than ``(x-x%y)//y`` due to\n    rounding.  In such cases, Python returns the latter result, in\n    order to preserve that ``divmod(x,y)[0] * y + x % y`` be very\n    close to ``x``.\n\n[3] While comparisons between strings make sense at the byte level,\n    they may be counter-intuitive to users.  For example, the strings\n    ``"\\u00C7"`` and ``"\\u0327\\u0043"`` compare differently, even\n    though they both represent the same unicode character (LATIN\n    CAPITAL LETTER C WITH CEDILLA).  To compare strings in a human\n    recognizable way, compare using ``unicodedata.normalize()``.\n\n[4] Due to automatic garbage-collection, free lists, and the dynamic\n    nature of descriptors, you may notice seemingly unusual behaviour\n    in certain uses of the ``is`` operator, like those involving\n    comparisons between instance methods, or constants.  Check their\n    documentation for more info.\n\n[5] The power operator ``**`` binds less tightly than an arithmetic or\n    bitwise unary operator on its right, that is, ``2**-1`` is\n    ``0.5``.\n',
  'pass': '\nThe ``pass`` statement\n**********************\n\n   pass_stmt ::= "pass"\n\n``pass`` is a null operation --- when it is executed, nothing happens.\nIt is useful as a placeholder when a statement is required\nsyntactically, but no code needs to be executed, for example:\n\n   def f(arg): pass    # a function that does nothing (yet)\n\n   class C: pass       # a class with no methods (yet)\n',
  'power': '\nThe power operator\n******************\n\nThe power operator binds more tightly than unary operators on its\nleft; it binds less tightly than unary operators on its right.  The\nsyntax is:\n\n   power ::= primary ["**" u_expr]\n\nThus, in an unparenthesized sequence of power and unary operators, the\noperators are evaluated from right to left (this does not constrain\nthe evaluation order for the operands): ``-1**2`` results in ``-1``.\n\nThe power operator has the same semantics as the built-in ``pow()``\nfunction, when called with two arguments: it yields its left argument\nraised to the power of its right argument.  The numeric arguments are\nfirst converted to a common type, and the result is of that type.\n\nFor int operands, the result has the same type as the operands unless\nthe second argument is negative; in that case, all arguments are\nconverted to float and a float result is delivered. For example,\n``10**2`` returns ``100``, but ``10**-2`` returns ``0.01``.\n\nRaising ``0.0`` to a negative power results in a\n``ZeroDivisionError``. Raising a negative number to a fractional power\nresults in a ``complex`` number. (In earlier versions it raised a\n``ValueError``.)\n',
  'raise': '\nThe ``raise`` statement\n***********************\n\n   raise_stmt ::= "raise" [expression ["from" expression]]\n\nIf no expressions are present, ``raise`` re-raises the last exception\nthat was active in the current scope.  If no exception is active in\nthe current scope, a ``TypeError`` exception is raised indicating that\nthis is an error (if running under IDLE, a ``queue.Empty`` exception\nis raised instead).\n\nOtherwise, ``raise`` evaluates the first expression as the exception\nobject.  It must be either a subclass or an instance of\n``BaseException``. If it is a class, the exception instance will be\nobtained when needed by instantiating the class with no arguments.\n\nThe *type* of the exception is the exception instance\'s class, the\n*value* is the instance itself.\n\nA traceback object is normally created automatically when an exception\nis raised and attached to it as the ``__traceback__`` attribute, which\nis writable. You can create an exception and set your own traceback in\none step using the ``with_traceback()`` exception method (which\nreturns the same exception instance, with its traceback set to its\nargument), like so:\n\n   raise Exception("foo occurred").with_traceback(tracebackobj)\n\nThe ``from`` clause is used for exception chaining: if given, the\nsecond *expression* must be another exception class or instance, which\nwill then be attached to the raised exception as the ``__cause__``\nattribute (which is writable).  If the raised exception is not\nhandled, both exceptions will be printed:\n\n   >>> try:\n   ...     print(1 / 0)\n   ... except Exception as exc:\n   ...     raise RuntimeError("Something bad happened") from exc\n   ...\n   Traceback (most recent call last):\n     File "<stdin>", line 2, in <module>\n   ZeroDivisionError: int division or modulo by zero\n\n   The above exception was the direct cause of the following exception:\n\n   Traceback (most recent call last):\n     File "<stdin>", line 4, in <module>\n   RuntimeError: Something bad happened\n\nA similar mechanism works implicitly if an exception is raised inside\nan exception handler: the previous exception is then attached as the\nnew exception\'s ``__context__`` attribute:\n\n   >>> try:\n   ...     print(1 / 0)\n   ... except:\n   ...     raise RuntimeError("Something bad happened")\n   ...\n   Traceback (most recent call last):\n     File "<stdin>", line 2, in <module>\n   ZeroDivisionError: int division or modulo by zero\n\n   During handling of the above exception, another exception occurred:\n\n   Traceback (most recent call last):\n     File "<stdin>", line 4, in <module>\n   RuntimeError: Something bad happened\n\nAdditional information on exceptions can be found in section\n*Exceptions*, and information about handling exceptions is in section\n*The try statement*.\n',
- 'return': '\nThe ``return`` statement\n************************\n\n   return_stmt ::= "return" [expression_list]\n\n``return`` may only occur syntactically nested in a function\ndefinition, not within a nested class definition.\n\nIf an expression list is present, it is evaluated, else ``None`` is\nsubstituted.\n\n``return`` leaves the current function call with the expression list\n(or ``None``) as return value.\n\nWhen ``return`` passes control out of a ``try`` statement with a\n``finally`` clause, that ``finally`` clause is executed before really\nleaving the function.\n\nIn a generator function, the ``return`` statement is not allowed to\ninclude an **expression_list**.  In that context, a bare ``return``\nindicates that the generator is done and will cause ``StopIteration``\nto be raised.\n',
- 'sequence-types': "\nEmulating container types\n*************************\n\nThe following methods can be defined to implement container objects.\nContainers usually are sequences (such as lists or tuples) or mappings\n(like dictionaries), but can represent other containers as well.  The\nfirst set of methods is used either to emulate a sequence or to\nemulate a mapping; the difference is that for a sequence, the\nallowable keys should be the integers *k* for which ``0 <= k < N``\nwhere *N* is the length of the sequence, or slice objects, which\ndefine a range of items.  It is also recommended that mappings provide\nthe methods ``keys()``, ``values()``, ``items()``, ``get()``,\n``clear()``, ``setdefault()``, ``pop()``, ``popitem()``, ``copy()``,\nand ``update()`` behaving similar to those for Python's standard\ndictionary objects.  The ``collections`` module provides a\n``MutableMapping`` abstract base class to help create those methods\nfrom a base set of ``__getitem__()``, ``__setitem__()``,\n``__delitem__()``, and ``keys()``. Mutable sequences should provide\nmethods ``append()``, ``count()``, ``index()``, ``extend()``,\n``insert()``, ``pop()``, ``remove()``, ``reverse()`` and ``sort()``,\nlike Python standard list objects.  Finally, sequence types should\nimplement addition (meaning concatenation) and multiplication (meaning\nrepetition) by defining the methods ``__add__()``, ``__radd__()``,\n``__iadd__()``, ``__mul__()``, ``__rmul__()`` and ``__imul__()``\ndescribed below; they should not define other numerical operators.  It\nis recommended that both mappings and sequences implement the\n``__contains__()`` method to allow efficient use of the ``in``\noperator; for mappings, ``in`` should search the mapping's keys; for\nsequences, it should search through the values.  It is further\nrecommended that both mappings and sequences implement the\n``__iter__()`` method to allow efficient iteration through the\ncontainer; for mappings, ``__iter__()`` should be the same as\n``keys()``; for sequences, it should iterate through the values.\n\nobject.__len__(self)\n\n   Called to implement the built-in function ``len()``.  Should return\n   the length of the object, an integer ``>=`` 0.  Also, an object\n   that doesn't define a ``__bool__()`` method and whose ``__len__()``\n   method returns zero is considered to be false in a Boolean context.\n\nNote: Slicing is done exclusively with the following three methods.  A\n  call like\n\n     a[1:2] = b\n\n  is translated to\n\n     a[slice(1, 2, None)] = b\n\n  and so forth.  Missing slice items are always filled in with\n  ``None``.\n\nobject.__getitem__(self, key)\n\n   Called to implement evaluation of ``self[key]``. For sequence\n   types, the accepted keys should be integers and slice objects.\n   Note that the special interpretation of negative indexes (if the\n   class wishes to emulate a sequence type) is up to the\n   ``__getitem__()`` method. If *key* is of an inappropriate type,\n   ``TypeError`` may be raised; if of a value outside the set of\n   indexes for the sequence (after any special interpretation of\n   negative values), ``IndexError`` should be raised. For mapping\n   types, if *key* is missing (not in the container), ``KeyError``\n   should be raised.\n\n   Note: ``for`` loops expect that an ``IndexError`` will be raised for\n     illegal indexes to allow proper detection of the end of the\n     sequence.\n\nobject.__setitem__(self, key, value)\n\n   Called to implement assignment to ``self[key]``.  Same note as for\n   ``__getitem__()``.  This should only be implemented for mappings if\n   the objects support changes to the values for keys, or if new keys\n   can be added, or for sequences if elements can be replaced.  The\n   same exceptions should be raised for improper *key* values as for\n   the ``__getitem__()`` method.\n\nobject.__delitem__(self, key)\n\n   Called to implement deletion of ``self[key]``.  Same note as for\n   ``__getitem__()``.  This should only be implemented for mappings if\n   the objects support removal of keys, or for sequences if elements\n   can be removed from the sequence.  The same exceptions should be\n   raised for improper *key* values as for the ``__getitem__()``\n   method.\n\nobject.__iter__(self)\n\n   This method is called when an iterator is required for a container.\n   This method should return a new iterator object that can iterate\n   over all the objects in the container.  For mappings, it should\n   iterate over the keys of the container, and should also be made\n   available as the method ``keys()``.\n\n   Iterator objects also need to implement this method; they are\n   required to return themselves.  For more information on iterator\n   objects, see *Iterator Types*.\n\nobject.__reversed__(self)\n\n   Called (if present) by the ``reversed()`` builtin to implement\n   reverse iteration.  It should return a new iterator object that\n   iterates over all the objects in the container in reverse order.\n\n   If the ``__reversed__()`` method is not provided, the\n   ``reversed()`` builtin will fall back to using the sequence\n   protocol (``__len__()`` and ``__getitem__()``).  Objects that\n   support the sequence protocol should only provide\n   ``__reversed__()`` if they can provide an implementation that is\n   more efficient than the one provided by ``reversed()``.\n\nThe membership test operators (``in`` and ``not in``) are normally\nimplemented as an iteration through a sequence.  However, container\nobjects can supply the following special method with a more efficient\nimplementation, which also does not require the object be a sequence.\n\nobject.__contains__(self, item)\n\n   Called to implement membership test operators.  Should return true\n   if *item* is in *self*, false otherwise.  For mapping objects, this\n   should consider the keys of the mapping rather than the values or\n   the key-item pairs.\n",
- 'shifting': '\nShifting operations\n*******************\n\nThe shifting operations have lower priority than the arithmetic\noperations:\n\n   shift_expr ::= a_expr | shift_expr ( "<<" | ">>" ) a_expr\n\nThese operators accept integers as arguments.  They shift the first\nargument to the left or right by the number of bits given by the\nsecond argument.\n\nA right shift by *n* bits is defined as division by ``pow(2,n)``.  A\nleft shift by *n* bits is defined as multiplication with ``pow(2,n)``.\n',
+ 'return': '\nThe ``return`` statement\n************************\n\n   return_stmt ::= "return" [expression_list]\n\n``return`` may only occur syntactically nested in a function\ndefinition, not within a nested class definition.\n\nIf an expression list is present, it is evaluated, else ``None`` is\nsubstituted.\n\n``return`` leaves the current function call with the expression list\n(or ``None``) as return value.\n\nWhen ``return`` passes control out of a ``try`` statement with a\n``finally`` clause, that ``finally`` clause is executed before really\nleaving the function.\n\nIn a generator function, the ``return`` statement is not allowed to\ninclude an ``expression_list``.  In that context, a bare ``return``\nindicates that the generator is done and will cause ``StopIteration``\nto be raised.\n',
+ 'sequence-types': "\nEmulating container types\n*************************\n\nThe following methods can be defined to implement container objects.\nContainers usually are sequences (such as lists or tuples) or mappings\n(like dictionaries), but can represent other containers as well.  The\nfirst set of methods is used either to emulate a sequence or to\nemulate a mapping; the difference is that for a sequence, the\nallowable keys should be the integers *k* for which ``0 <= k < N``\nwhere *N* is the length of the sequence, or slice objects, which\ndefine a range of items.  It is also recommended that mappings provide\nthe methods ``keys()``, ``values()``, ``items()``, ``get()``,\n``clear()``, ``setdefault()``, ``pop()``, ``popitem()``, ``copy()``,\nand ``update()`` behaving similar to those for Python's standard\ndictionary objects.  The ``collections`` module provides a\n``MutableMapping`` abstract base class to help create those methods\nfrom a base set of ``__getitem__()``, ``__setitem__()``,\n``__delitem__()``, and ``keys()``. Mutable sequences should provide\nmethods ``append()``, ``count()``, ``index()``, ``extend()``,\n``insert()``, ``pop()``, ``remove()``, ``reverse()`` and ``sort()``,\nlike Python standard list objects.  Finally, sequence types should\nimplement addition (meaning concatenation) and multiplication (meaning\nrepetition) by defining the methods ``__add__()``, ``__radd__()``,\n``__iadd__()``, ``__mul__()``, ``__rmul__()`` and ``__imul__()``\ndescribed below; they should not define other numerical operators.  It\nis recommended that both mappings and sequences implement the\n``__contains__()`` method to allow efficient use of the ``in``\noperator; for mappings, ``in`` should search the mapping's keys; for\nsequences, it should search through the values.  It is further\nrecommended that both mappings and sequences implement the\n``__iter__()`` method to allow efficient iteration through the\ncontainer; for mappings, ``__iter__()`` should be the same as\n``keys()``; for sequences, it should iterate through the values.\n\nobject.__len__(self)\n\n   Called to implement the built-in function ``len()``.  Should return\n   the length of the object, an integer ``>=`` 0.  Also, an object\n   that doesn't define a ``__bool__()`` method and whose ``__len__()``\n   method returns zero is considered to be false in a Boolean context.\n\nNote: Slicing is done exclusively with the following three methods.  A\n  call like\n\n     a[1:2] = b\n\n  is translated to\n\n     a[slice(1, 2, None)] = b\n\n  and so forth.  Missing slice items are always filled in with\n  ``None``.\n\nobject.__getitem__(self, key)\n\n   Called to implement evaluation of ``self[key]``. For sequence\n   types, the accepted keys should be integers and slice objects.\n   Note that the special interpretation of negative indexes (if the\n   class wishes to emulate a sequence type) is up to the\n   ``__getitem__()`` method. If *key* is of an inappropriate type,\n   ``TypeError`` may be raised; if of a value outside the set of\n   indexes for the sequence (after any special interpretation of\n   negative values), ``IndexError`` should be raised. For mapping\n   types, if *key* is missing (not in the container), ``KeyError``\n   should be raised.\n\n   Note: ``for`` loops expect that an ``IndexError`` will be raised for\n     illegal indexes to allow proper detection of the end of the\n     sequence.\n\nobject.__setitem__(self, key, value)\n\n   Called to implement assignment to ``self[key]``.  Same note as for\n   ``__getitem__()``.  This should only be implemented for mappings if\n   the objects support changes to the values for keys, or if new keys\n   can be added, or for sequences if elements can be replaced.  The\n   same exceptions should be raised for improper *key* values as for\n   the ``__getitem__()`` method.\n\nobject.__delitem__(self, key)\n\n   Called to implement deletion of ``self[key]``.  Same note as for\n   ``__getitem__()``.  This should only be implemented for mappings if\n   the objects support removal of keys, or for sequences if elements\n   can be removed from the sequence.  The same exceptions should be\n   raised for improper *key* values as for the ``__getitem__()``\n   method.\n\nobject.__iter__(self)\n\n   This method is called when an iterator is required for a container.\n   This method should return a new iterator object that can iterate\n   over all the objects in the container.  For mappings, it should\n   iterate over the keys of the container, and should also be made\n   available as the method ``keys()``.\n\n   Iterator objects also need to implement this method; they are\n   required to return themselves.  For more information on iterator\n   objects, see *Iterator Types*.\n\nobject.__reversed__(self)\n\n   Called (if present) by the ``reversed()`` built-in to implement\n   reverse iteration.  It should return a new iterator object that\n   iterates over all the objects in the container in reverse order.\n\n   If the ``__reversed__()`` method is not provided, the\n   ``reversed()`` built-in will fall back to using the sequence\n   protocol (``__len__()`` and ``__getitem__()``).  Objects that\n   support the sequence protocol should only provide\n   ``__reversed__()`` if they can provide an implementation that is\n   more efficient than the one provided by ``reversed()``.\n\nThe membership test operators (``in`` and ``not in``) are normally\nimplemented as an iteration through a sequence.  However, container\nobjects can supply the following special method with a more efficient\nimplementation, which also does not require the object be a sequence.\n\nobject.__contains__(self, item)\n\n   Called to implement membership test operators.  Should return true\n   if *item* is in *self*, false otherwise.  For mapping objects, this\n   should consider the keys of the mapping rather than the values or\n   the key-item pairs.\n\n   For objects that don't define ``__contains__()``, the membership\n   test first tries iteration via ``__iter__()``, then the old\n   sequence iteration protocol via ``__getitem__()``, see *this\n   section in the language reference*.\n",
+ 'shifting': '\nShifting operations\n*******************\n\nThe shifting operations have lower priority than the arithmetic\noperations:\n\n   shift_expr ::= a_expr | shift_expr ( "<<" | ">>" ) a_expr\n\nThese operators accept integers as arguments.  They shift the first\nargument to the left or right by the number of bits given by the\nsecond argument.\n\nA right shift by *n* bits is defined as division by ``pow(2,n)``.  A\nleft shift by *n* bits is defined as multiplication with ``pow(2,n)``.\n\nNote: In the current implementation, the right-hand operand is required to\n  be at most ``sys.maxsize``.  If the right-hand operand is larger\n  than ``sys.maxsize`` an ``OverflowError`` exception is raised.\n',
  'slicings': '\nSlicings\n********\n\nA slicing selects a range of items in a sequence object (e.g., a\nstring, tuple or list).  Slicings may be used as expressions or as\ntargets in assignment or ``del`` statements.  The syntax for a\nslicing:\n\n   slicing      ::= primary "[" slice_list "]"\n   slice_list   ::= slice_item ("," slice_item)* [","]\n   slice_item   ::= expression | proper_slice\n   proper_slice ::= [lower_bound] ":" [upper_bound] [ ":" [stride] ]\n   lower_bound  ::= expression\n   upper_bound  ::= expression\n   stride       ::= expression\n\nThere is ambiguity in the formal syntax here: anything that looks like\nan expression list also looks like a slice list, so any subscription\ncan be interpreted as a slicing.  Rather than further complicating the\nsyntax, this is disambiguated by defining that in this case the\ninterpretation as a subscription takes priority over the\ninterpretation as a slicing (this is the case if the slice list\ncontains no proper slice).\n\nThe semantics for a slicing are as follows.  The primary must evaluate\nto a mapping object, and it is indexed (using the same\n``__getitem__()`` method as normal subscription) with a key that is\nconstructed from the slice list, as follows.  If the slice list\ncontains at least one comma, the key is a tuple containing the\nconversion of the slice items; otherwise, the conversion of the lone\nslice item is the key.  The conversion of a slice item that is an\nexpression is that expression.  The conversion of a proper slice is a\nslice object (see section *The standard type hierarchy*) whose\n``start``, ``stop`` and ``step`` attributes are the values of the\nexpressions given as lower bound, upper bound and stride,\nrespectively, substituting ``None`` for missing expressions.\n',
- 'specialattrs': "\nSpecial Attributes\n******************\n\nThe implementation adds a few special read-only attributes to several\nobject types, where they are relevant.  Some of these are not reported\nby the ``dir()`` built-in function.\n\nobject.__dict__\n\n   A dictionary or other mapping object used to store an object's\n   (writable) attributes.\n\ninstance.__class__\n\n   The class to which a class instance belongs.\n\nclass.__bases__\n\n   The tuple of base classes of a class object.  If there are no base\n   classes, this will be an empty tuple.\n\nclass.__name__\n\n   The name of the class or type.\n\nThe following attributes are only supported by *new-style class*es.\n\nclass.__mro__\n\n   This attribute is a tuple of classes that are considered when\n   looking for base classes during method resolution.\n\nclass.mro()\n\n   This method can be overridden by a metaclass to customize the\n   method resolution order for its instances.  It is called at class\n   instantiation, and its result is stored in ``__mro__``.\n\nclass.__subclasses__()\n\n   Each new-style class keeps a list of weak references to its\n   immediate subclasses.  This method returns a list of all those\n   references still alive. Example:\n\n      >>> int.__subclasses__()\n      [<type 'bool'>]\n\n-[ Footnotes ]-\n\n[1] Additional information on these special methods may be found in\n    the Python Reference Manual (*Basic customization*).\n\n[2] As a consequence, the list ``[1, 2]`` is considered equal to\n    ``[1.0, 2.0]``, and similarly for tuples.\n\n[3] They must have since the parser can't tell the type of the\n    operands.\n\n[4] To format only a tuple you should therefore provide a singleton\n    tuple whose only element is the tuple to be formatted.\n\n[5] The advantage of leaving the newline on is that returning an empty\n    string is then an unambiguous EOF indication.  It is also possible\n    (in cases where it might matter, for example, if you want to make\n    an exact copy of a file while scanning its lines) to tell whether\n    the last line of a file ended in a newline or not (yes this\n    happens!).\n",
- 'specialnames': '\nSpecial method names\n********************\n\nA class can implement certain operations that are invoked by special\nsyntax (such as arithmetic operations or subscripting and slicing) by\ndefining methods with special names. This is Python\'s approach to\n*operator overloading*, allowing classes to define their own behavior\nwith respect to language operators.  For instance, if a class defines\na method named ``__getitem__()``, and ``x`` is an instance of this\nclass, then ``x[i]`` is roughly equivalent to ``type(x).__getitem__(x,\ni)``.  Except where mentioned, attempts to execute an operation raise\nan exception when no appropriate method is defined (typically\n``AttributeError`` or ``TypeError``).\n\nWhen implementing a class that emulates any built-in type, it is\nimportant that the emulation only be implemented to the degree that it\nmakes sense for the object being modelled.  For example, some\nsequences may work well with retrieval of individual elements, but\nextracting a slice may not make sense.  (One example of this is the\n``NodeList`` interface in the W3C\'s Document Object Model.)\n\n\nBasic customization\n===================\n\nobject.__new__(cls[, ...])\n\n   Called to create a new instance of class *cls*.  ``__new__()`` is a\n   static method (special-cased so you need not declare it as such)\n   that takes the class of which an instance was requested as its\n   first argument.  The remaining arguments are those passed to the\n   object constructor expression (the call to the class).  The return\n   value of ``__new__()`` should be the new object instance (usually\n   an instance of *cls*).\n\n   Typical implementations create a new instance of the class by\n   invoking the superclass\'s ``__new__()`` method using\n   ``super(currentclass, cls).__new__(cls[, ...])`` with appropriate\n   arguments and then modifying the newly-created instance as\n   necessary before returning it.\n\n   If ``__new__()`` returns an instance of *cls*, then the new\n   instance\'s ``__init__()`` method will be invoked like\n   ``__init__(self[, ...])``, where *self* is the new instance and the\n   remaining arguments are the same as were passed to ``__new__()``.\n\n   If ``__new__()`` does not return an instance of *cls*, then the new\n   instance\'s ``__init__()`` method will not be invoked.\n\n   ``__new__()`` is intended mainly to allow subclasses of immutable\n   types (like int, str, or tuple) to customize instance creation.  It\n   is also commonly overridden in custom metaclasses in order to\n   customize class creation.\n\nobject.__init__(self[, ...])\n\n   Called when the instance is created.  The arguments are those\n   passed to the class constructor expression.  If a base class has an\n   ``__init__()`` method, the derived class\'s ``__init__()`` method,\n   if any, must explicitly call it to ensure proper initialization of\n   the base class part of the instance; for example:\n   ``BaseClass.__init__(self, [args...])``.  As a special constraint\n   on constructors, no value may be returned; doing so will cause a\n   ``TypeError`` to be raised at runtime.\n\nobject.__del__(self)\n\n   Called when the instance is about to be destroyed.  This is also\n   called a destructor.  If a base class has a ``__del__()`` method,\n   the derived class\'s ``__del__()`` method, if any, must explicitly\n   call it to ensure proper deletion of the base class part of the\n   instance.  Note that it is possible (though not recommended!) for\n   the ``__del__()`` method to postpone destruction of the instance by\n   creating a new reference to it.  It may then be called at a later\n   time when this new reference is deleted.  It is not guaranteed that\n   ``__del__()`` methods are called for objects that still exist when\n   the interpreter exits.\n\n   Note: ``del x`` doesn\'t directly call ``x.__del__()`` --- the former\n     decrements the reference count for ``x`` by one, and the latter\n     is only called when ``x``\'s reference count reaches zero.  Some\n     common situations that may prevent the reference count of an\n     object from going to zero include: circular references between\n     objects (e.g., a doubly-linked list or a tree data structure with\n     parent and child pointers); a reference to the object on the\n     stack frame of a function that caught an exception (the traceback\n     stored in ``sys.exc_info()[2]`` keeps the stack frame alive); or\n     a reference to the object on the stack frame that raised an\n     unhandled exception in interactive mode (the traceback stored in\n     ``sys.last_traceback`` keeps the stack frame alive).  The first\n     situation can only be remedied by explicitly breaking the cycles;\n     the latter two situations can be resolved by storing ``None`` in\n     ``sys.last_traceback``. Circular references which are garbage are\n     detected when the option cycle detector is enabled (it\'s on by\n     default), but can only be cleaned up if there are no Python-\n     level ``__del__()`` methods involved. Refer to the documentation\n     for the ``gc`` module for more information about how\n     ``__del__()`` methods are handled by the cycle detector,\n     particularly the description of the ``garbage`` value.\n\n   Warning: Due to the precarious circumstances under which ``__del__()``\n     methods are invoked, exceptions that occur during their execution\n     are ignored, and a warning is printed to ``sys.stderr`` instead.\n     Also, when ``__del__()`` is invoked in response to a module being\n     deleted (e.g., when execution of the program is done), other\n     globals referenced by the ``__del__()`` method may already have\n     been deleted or in the process of being torn down (e.g. the\n     import machinery shutting down).  For this reason, ``__del__()``\n     methods should do the absolute minimum needed to maintain\n     external invariants.  Starting with version 1.5, Python\n     guarantees that globals whose name begins with a single\n     underscore are deleted from their module before other globals are\n     deleted; if no other references to such globals exist, this may\n     help in assuring that imported modules are still available at the\n     time when the ``__del__()`` method is called.\n\nobject.__repr__(self)\n\n   Called by the ``repr()`` built-in function to compute the\n   "official" string representation of an object.  If at all possible,\n   this should look like a valid Python expression that could be used\n   to recreate an object with the same value (given an appropriate\n   environment).  If this is not possible, a string of the form\n   ``<...some useful description...>`` should be returned. The return\n   value must be a string object. If a class defines ``__repr__()``\n   but not ``__str__()``, then ``__repr__()`` is also used when an\n   "informal" string representation of instances of that class is\n   required.\n\n   This is typically used for debugging, so it is important that the\n   representation is information-rich and unambiguous.\n\nobject.__str__(self)\n\n   Called by the ``str()`` built-in function and by the ``print()``\n   function to compute the "informal" string representation of an\n   object.  This differs from ``__repr__()`` in that it does not have\n   to be a valid Python expression: a more convenient or concise\n   representation may be used instead. The return value must be a\n   string object.\n\nobject.__format__(self, format_spec)\n\n   Called by the ``format()`` built-in function (and by extension, the\n   ``format()`` method of class ``str``) to produce a "formatted"\n   string representation of an object. The ``format_spec`` argument is\n   a string that contains a description of the formatting options\n   desired. The interpretation of the ``format_spec`` argument is up\n   to the type implementing ``__format__()``, however most classes\n   will either delegate formatting to one of the built-in types, or\n   use a similar formatting option syntax.\n\n   See *Format Specification Mini-Language* for a description of the\n   standard formatting syntax.\n\n   The return value must be a string object.\n\nobject.__lt__(self, other)\nobject.__le__(self, other)\nobject.__eq__(self, other)\nobject.__ne__(self, other)\nobject.__gt__(self, other)\nobject.__ge__(self, other)\n\n   These are the so-called "rich comparison" methods. The\n   correspondence between operator symbols and method names is as\n   follows: ``x<y`` calls ``x.__lt__(y)``, ``x<=y`` calls\n   ``x.__le__(y)``, ``x==y`` calls ``x.__eq__(y)``, ``x!=y`` calls\n   ``x.__ne__(y)``, ``x>y`` calls ``x.__gt__(y)``, and ``x>=y`` calls\n   ``x.__ge__(y)``.\n\n   A rich comparison method may return the singleton\n   ``NotImplemented`` if it does not implement the operation for a\n   given pair of arguments. By convention, ``False`` and ``True`` are\n   returned for a successful comparison. However, these methods can\n   return any value, so if the comparison operator is used in a\n   Boolean context (e.g., in the condition of an ``if`` statement),\n   Python will call ``bool()`` on the value to determine if the result\n   is true or false.\n\n   There are no implied relationships among the comparison operators.\n   The truth of ``x==y`` does not imply that ``x!=y`` is false.\n   Accordingly, when defining ``__eq__()``, one should also define\n   ``__ne__()`` so that the operators will behave as expected.  See\n   the paragraph on ``__hash__()`` for some important notes on\n   creating *hashable* objects which support custom comparison\n   operations and are usable as dictionary keys.\n\n   There are no swapped-argument versions of these methods (to be used\n   when the left argument does not support the operation but the right\n   argument does); rather, ``__lt__()`` and ``__gt__()`` are each\n   other\'s reflection, ``__le__()`` and ``__ge__()`` are each other\'s\n   reflection, and ``__eq__()`` and ``__ne__()`` are their own\n   reflection.\n\n   Arguments to rich comparison methods are never coerced.\n\n   To automatically generate ordering operations from a single root\n   operation, see the Total Ordering recipe in the ASPN cookbook.\n\nobject.__hash__(self)\n\n   Called by built-in function ``hash()`` and for operations on\n   members of hashed collections including ``set``, ``frozenset``, and\n   ``dict``.  ``__hash__()`` should return an integer.  The only\n   required property is that objects which compare equal have the same\n   hash value; it is advised to somehow mix together (e.g. using\n   exclusive or) the hash values for the components of the object that\n   also play a part in comparison of objects.\n\n   If a class does not define an ``__eq__()`` method it should not\n   define a ``__hash__()`` operation either; if it defines\n   ``__eq__()`` but not ``__hash__()``, its instances will not be\n   usable as items in hashable collections.  If a class defines\n   mutable objects and implements an ``__eq__()`` method, it should\n   not implement ``__hash__()``, since the implementation of hashable\n   collections requires that a key\'s hash value is immutable (if the\n   object\'s hash value changes, it will be in the wrong hash bucket).\n\n   User-defined classes have ``__eq__()`` and ``__hash__()`` methods\n   by default; with them, all objects compare unequal (except with\n   themselves) and ``x.__hash__()`` returns ``id(x)``.\n\n   Classes which inherit a ``__hash__()`` method from a parent class\n   but change the meaning of ``__eq__()`` such that the hash value\n   returned is no longer appropriate (e.g. by switching to a value-\n   based concept of equality instead of the default identity based\n   equality) can explicitly flag themselves as being unhashable by\n   setting ``__hash__ = None`` in the class definition. Doing so means\n   that not only will instances of the class raise an appropriate\n   ``TypeError`` when a program attempts to retrieve their hash value,\n   but they will also be correctly identified as unhashable when\n   checking ``isinstance(obj, collections.Hashable)`` (unlike classes\n   which define their own ``__hash__()`` to explicitly raise\n   ``TypeError``).\n\n   If a class that overrides ``__eq__()`` needs to retain the\n   implementation of ``__hash__()`` from a parent class, the\n   interpreter must be told this explicitly by setting ``__hash__ =\n   <ParentClass>.__hash__``. Otherwise the inheritance of\n   ``__hash__()`` will be blocked, just as if ``__hash__`` had been\n   explicitly set to ``None``.\n\nobject.__bool__(self)\n\n   Called to implement truth value testing and the built-in operation\n   ``bool()``; should return ``False`` or ``True``, or their integer\n   equivalents ``0`` or ``1``.  When this method is not defined,\n   ``__len__()`` is called, if it is defined, and the object is\n   considered true if its result is nonzero. If a class defines\n   neither ``__len__()`` nor ``__bool__()``, all its instances are\n   considered true.\n\n\nCustomizing attribute access\n============================\n\nThe following methods can be defined to customize the meaning of\nattribute access (use of, assignment to, or deletion of ``x.name``)\nfor class instances.\n\nobject.__getattr__(self, name)\n\n   Called when an attribute lookup has not found the attribute in the\n   usual places (i.e. it is not an instance attribute nor is it found\n   in the class tree for ``self``).  ``name`` is the attribute name.\n   This method should return the (computed) attribute value or raise\n   an ``AttributeError`` exception.\n\n   Note that if the attribute is found through the normal mechanism,\n   ``__getattr__()`` is not called.  (This is an intentional asymmetry\n   between ``__getattr__()`` and ``__setattr__()``.) This is done both\n   for efficiency reasons and because otherwise ``__getattr__()``\n   would have no way to access other attributes of the instance.  Note\n   that at least for instance variables, you can fake total control by\n   not inserting any values in the instance attribute dictionary (but\n   instead inserting them in another object).  See the\n   ``__getattribute__()`` method below for a way to actually get total\n   control over attribute access.\n\nobject.__getattribute__(self, name)\n\n   Called unconditionally to implement attribute accesses for\n   instances of the class. If the class also defines\n   ``__getattr__()``, the latter will not be called unless\n   ``__getattribute__()`` either calls it explicitly or raises an\n   ``AttributeError``. This method should return the (computed)\n   attribute value or raise an ``AttributeError`` exception. In order\n   to avoid infinite recursion in this method, its implementation\n   should always call the base class method with the same name to\n   access any attributes it needs, for example,\n   ``object.__getattribute__(self, name)``.\n\n   Note: This method may still be bypassed when looking up special methods\n     as the result of implicit invocation via language syntax or\n     builtin functions. See *Special method lookup*.\n\nobject.__setattr__(self, name, value)\n\n   Called when an attribute assignment is attempted.  This is called\n   instead of the normal mechanism (i.e. store the value in the\n   instance dictionary). *name* is the attribute name, *value* is the\n   value to be assigned to it.\n\n   If ``__setattr__()`` wants to assign to an instance attribute, it\n   should call the base class method with the same name, for example,\n   ``object.__setattr__(self, name, value)``.\n\nobject.__delattr__(self, name)\n\n   Like ``__setattr__()`` but for attribute deletion instead of\n   assignment.  This should only be implemented if ``del obj.name`` is\n   meaningful for the object.\n\nobject.__dir__(self)\n\n   Called when ``dir()`` is called on the object.  A list must be\n   returned.\n\n\nImplementing Descriptors\n------------------------\n\nThe following methods only apply when an instance of the class\ncontaining the method (a so-called *descriptor* class) appears in the\nclass dictionary of another class, known as the *owner* class.  In the\nexamples below, "the attribute" refers to the attribute whose name is\nthe key of the property in the owner class\' ``__dict__``.\n\nobject.__get__(self, instance, owner)\n\n   Called to get the attribute of the owner class (class attribute\n   access) or of an instance of that class (instance attribute\n   access). *owner* is always the owner class, while *instance* is the\n   instance that the attribute was accessed through, or ``None`` when\n   the attribute is accessed through the *owner*.  This method should\n   return the (computed) attribute value or raise an\n   ``AttributeError`` exception.\n\nobject.__set__(self, instance, value)\n\n   Called to set the attribute on an instance *instance* of the owner\n   class to a new value, *value*.\n\nobject.__delete__(self, instance)\n\n   Called to delete the attribute on an instance *instance* of the\n   owner class.\n\n\nInvoking Descriptors\n--------------------\n\nIn general, a descriptor is an object attribute with "binding\nbehavior", one whose attribute access has been overridden by methods\nin the descriptor protocol:  ``__get__()``, ``__set__()``, and\n``__delete__()``. If any of those methods are defined for an object,\nit is said to be a descriptor.\n\nThe default behavior for attribute access is to get, set, or delete\nthe attribute from an object\'s dictionary. For instance, ``a.x`` has a\nlookup chain starting with ``a.__dict__[\'x\']``, then\n``type(a).__dict__[\'x\']``, and continuing through the base classes of\n``type(a)`` excluding metaclasses.\n\nHowever, if the looked-up value is an object defining one of the\ndescriptor methods, then Python may override the default behavior and\ninvoke the descriptor method instead.  Where this occurs in the\nprecedence chain depends on which descriptor methods were defined and\nhow they were called.\n\nThe starting point for descriptor invocation is a binding, ``a.x``.\nHow the arguments are assembled depends on ``a``:\n\nDirect Call\n   The simplest and least common call is when user code directly\n   invokes a descriptor method:    ``x.__get__(a)``.\n\nInstance Binding\n   If binding to an object instance, ``a.x`` is transformed into the\n   call: ``type(a).__dict__[\'x\'].__get__(a, type(a))``.\n\nClass Binding\n   If binding to a class, ``A.x`` is transformed into the call:\n   ``A.__dict__[\'x\'].__get__(None, A)``.\n\nSuper Binding\n   If ``a`` is an instance of ``super``, then the binding ``super(B,\n   obj).m()`` searches ``obj.__class__.__mro__`` for the base class\n   ``A`` immediately preceding ``B`` and then invokes the descriptor\n   with the call: ``A.__dict__[\'m\'].__get__(obj, A)``.\n\nFor instance bindings, the precedence of descriptor invocation depends\non the which descriptor methods are defined.  Normally, data\ndescriptors define both ``__get__()`` and ``__set__()``, while non-\ndata descriptors have just the ``__get__()`` method.  Data descriptors\nalways override a redefinition in an instance dictionary.  In\ncontrast, non-data descriptors can be overridden by instances. [2]\n\nPython methods (including ``staticmethod()`` and ``classmethod()``)\nare implemented as non-data descriptors.  Accordingly, instances can\nredefine and override methods.  This allows individual instances to\nacquire behaviors that differ from other instances of the same class.\n\nThe ``property()`` function is implemented as a data descriptor.\nAccordingly, instances cannot override the behavior of a property.\n\n\n__slots__\n---------\n\nBy default, instances of classes have a dictionary for attribute\nstorage.  This wastes space for objects having very few instance\nvariables.  The space consumption can become acute when creating large\nnumbers of instances.\n\nThe default can be overridden by defining *__slots__* in a class\ndefinition. The *__slots__* declaration takes a sequence of instance\nvariables and reserves just enough space in each instance to hold a\nvalue for each variable.  Space is saved because *__dict__* is not\ncreated for each instance.\n\nobject.__slots__\n\n   This class variable can be assigned a string, iterable, or sequence\n   of strings with variable names used by instances.  If defined in a\n   class, *__slots__* reserves space for the declared variables and\n   prevents the automatic creation of *__dict__* and *__weakref__* for\n   each instance.\n\n\nNotes on using *__slots__*\n~~~~~~~~~~~~~~~~~~~~~~~~~~\n\n* When inheriting from a class without *__slots__*, the *__dict__*\n  attribute of that class will always be accessible, so a *__slots__*\n  definition in the subclass is meaningless.\n\n* Without a *__dict__* variable, instances cannot be assigned new\n  variables not listed in the *__slots__* definition.  Attempts to\n  assign to an unlisted variable name raises ``AttributeError``. If\n  dynamic assignment of new variables is desired, then add\n  ``\'__dict__\'`` to the sequence of strings in the *__slots__*\n  declaration.\n\n* Without a *__weakref__* variable for each instance, classes defining\n  *__slots__* do not support weak references to its instances. If weak\n  reference support is needed, then add ``\'__weakref__\'`` to the\n  sequence of strings in the *__slots__* declaration.\n\n* *__slots__* are implemented at the class level by creating\n  descriptors (*Implementing Descriptors*) for each variable name.  As\n  a result, class attributes cannot be used to set default values for\n  instance variables defined by *__slots__*; otherwise, the class\n  attribute would overwrite the descriptor assignment.\n\n* If a class defines a slot also defined in a base class, the instance\n  variable defined by the base class slot is inaccessible (except by\n  retrieving its descriptor directly from the base class). This\n  renders the meaning of the program undefined.  In the future, a\n  check may be added to prevent this.\n\n* The action of a *__slots__* declaration is limited to the class\n  where it is defined.  As a result, subclasses will have a *__dict__*\n  unless they also define *__slots__*.\n\n* Nonempty *__slots__* does not work for classes derived from\n  "variable-length" built-in types such as ``int``, ``str`` and\n  ``tuple``.\n\n* Any non-string iterable may be assigned to *__slots__*. Mappings may\n  also be used; however, in the future, special meaning may be\n  assigned to the values corresponding to each key.\n\n* *__class__* assignment works only if both classes have the same\n  *__slots__*.\n\n\nCustomizing class creation\n==========================\n\nBy default, classes are constructed using ``type()``. A class\ndefinition is read into a separate namespace and the value of class\nname is bound to the result of ``type(name, bases, dict)``.\n\nWhen the class definition is read, if a callable ``metaclass`` keyword\nargument is passed after the bases in the class definition, the\ncallable given will be called instead of ``type()``.  If other keyword\narguments are passed, they will also be passed to the metaclass.  This\nallows classes or functions to be written which monitor or alter the\nclass creation process:\n\n* Modifying the class dictionary prior to the class being created.\n\n* Returning an instance of another class -- essentially performing the\n  role of a factory function.\n\nThese steps will have to be performed in the metaclass\'s ``__new__()``\nmethod -- ``type.__new__()`` can then be called from this method to\ncreate a class with different properties.  This example adds a new\nelement to the class dictionary before creating the class:\n\n   class metacls(type):\n       def __new__(mcs, name, bases, dict):\n           dict[\'foo\'] = \'metacls was here\'\n           return type.__new__(mcs, name, bases, dict)\n\nYou can of course also override other class methods (or add new\nmethods); for example defining a custom ``__call__()`` method in the\nmetaclass allows custom behavior when the class is called, e.g. not\nalways creating a new instance.\n\nIf the metaclass has a ``__prepare__()`` attribute (usually\nimplemented as a class or static method), it is called before the\nclass body is evaluated with the name of the class and a tuple of its\nbases for arguments.  It should return an object that supports the\nmapping interface that will be used to store the namespace of the\nclass.  The default is a plain dictionary.  This could be used, for\nexample, to keep track of the order that class attributes are declared\nin by returning an ordered dictionary.\n\nThe appropriate metaclass is determined by the following precedence\nrules:\n\n* If the ``metaclass`` keyword argument is based with the bases, it is\n  used.\n\n* Otherwise, if there is at least one base class, its metaclass is\n  used.\n\n* Otherwise, the default metaclass (``type``) is used.\n\nThe potential uses for metaclasses are boundless. Some ideas that have\nbeen explored including logging, interface checking, automatic\ndelegation, automatic property creation, proxies, frameworks, and\nautomatic resource locking/synchronization.\n\nHere is an example of a metaclass that uses an\n``collections.OrderedDict`` to remember the order that class members\nwere defined:\n\n   class OrderedClass(type):\n\n        @classmethod\n        def __prepare__(metacls, name, bases, **kwds):\n           return collections.OrderedDict()\n\n        def __new__(cls, name, bases, classdict):\n           result = type.__new__(cls, name, bases, dict(classdict))\n           result.members = tuple(classdict)\n           return result\n\n   class A(metaclass=OrderedClass):\n       def one(self): pass\n       def two(self): pass\n       def three(self): pass\n       def four(self): pass\n\n   >>> A.members\n   (\'__module__\', \'one\', \'two\', \'three\', \'four\')\n\nWhen the class definition for *A* gets executed, the process begins\nwith calling the metaclass\'s ``__prepare__()`` method which returns an\nempty ``collections.OrderedDict``.  That mapping records the methods\nand attributes of *A* as they are defined within the body of the class\nstatement. Once those definitions are executed, the ordered dictionary\nis fully populated and the metaclass\'s ``__new__()`` method gets\ninvoked.  That method builds the new type and it saves the ordered\ndictionary keys in an attribute called *members*.\n\n\nEmulating callable objects\n==========================\n\nobject.__call__(self[, args...])\n\n   Called when the instance is "called" as a function; if this method\n   is defined, ``x(arg1, arg2, ...)`` is a shorthand for\n   ``x.__call__(arg1, arg2, ...)``.\n\n\nEmulating container types\n=========================\n\nThe following methods can be defined to implement container objects.\nContainers usually are sequences (such as lists or tuples) or mappings\n(like dictionaries), but can represent other containers as well.  The\nfirst set of methods is used either to emulate a sequence or to\nemulate a mapping; the difference is that for a sequence, the\nallowable keys should be the integers *k* for which ``0 <= k < N``\nwhere *N* is the length of the sequence, or slice objects, which\ndefine a range of items.  It is also recommended that mappings provide\nthe methods ``keys()``, ``values()``, ``items()``, ``get()``,\n``clear()``, ``setdefault()``, ``pop()``, ``popitem()``, ``copy()``,\nand ``update()`` behaving similar to those for Python\'s standard\ndictionary objects.  The ``collections`` module provides a\n``MutableMapping`` abstract base class to help create those methods\nfrom a base set of ``__getitem__()``, ``__setitem__()``,\n``__delitem__()``, and ``keys()``. Mutable sequences should provide\nmethods ``append()``, ``count()``, ``index()``, ``extend()``,\n``insert()``, ``pop()``, ``remove()``, ``reverse()`` and ``sort()``,\nlike Python standard list objects.  Finally, sequence types should\nimplement addition (meaning concatenation) and multiplication (meaning\nrepetition) by defining the methods ``__add__()``, ``__radd__()``,\n``__iadd__()``, ``__mul__()``, ``__rmul__()`` and ``__imul__()``\ndescribed below; they should not define other numerical operators.  It\nis recommended that both mappings and sequences implement the\n``__contains__()`` method to allow efficient use of the ``in``\noperator; for mappings, ``in`` should search the mapping\'s keys; for\nsequences, it should search through the values.  It is further\nrecommended that both mappings and sequences implement the\n``__iter__()`` method to allow efficient iteration through the\ncontainer; for mappings, ``__iter__()`` should be the same as\n``keys()``; for sequences, it should iterate through the values.\n\nobject.__len__(self)\n\n   Called to implement the built-in function ``len()``.  Should return\n   the length of the object, an integer ``>=`` 0.  Also, an object\n   that doesn\'t define a ``__bool__()`` method and whose ``__len__()``\n   method returns zero is considered to be false in a Boolean context.\n\nNote: Slicing is done exclusively with the following three methods.  A\n  call like\n\n     a[1:2] = b\n\n  is translated to\n\n     a[slice(1, 2, None)] = b\n\n  and so forth.  Missing slice items are always filled in with\n  ``None``.\n\nobject.__getitem__(self, key)\n\n   Called to implement evaluation of ``self[key]``. For sequence\n   types, the accepted keys should be integers and slice objects.\n   Note that the special interpretation of negative indexes (if the\n   class wishes to emulate a sequence type) is up to the\n   ``__getitem__()`` method. If *key* is of an inappropriate type,\n   ``TypeError`` may be raised; if of a value outside the set of\n   indexes for the sequence (after any special interpretation of\n   negative values), ``IndexError`` should be raised. For mapping\n   types, if *key* is missing (not in the container), ``KeyError``\n   should be raised.\n\n   Note: ``for`` loops expect that an ``IndexError`` will be raised for\n     illegal indexes to allow proper detection of the end of the\n     sequence.\n\nobject.__setitem__(self, key, value)\n\n   Called to implement assignment to ``self[key]``.  Same note as for\n   ``__getitem__()``.  This should only be implemented for mappings if\n   the objects support changes to the values for keys, or if new keys\n   can be added, or for sequences if elements can be replaced.  The\n   same exceptions should be raised for improper *key* values as for\n   the ``__getitem__()`` method.\n\nobject.__delitem__(self, key)\n\n   Called to implement deletion of ``self[key]``.  Same note as for\n   ``__getitem__()``.  This should only be implemented for mappings if\n   the objects support removal of keys, or for sequences if elements\n   can be removed from the sequence.  The same exceptions should be\n   raised for improper *key* values as for the ``__getitem__()``\n   method.\n\nobject.__iter__(self)\n\n   This method is called when an iterator is required for a container.\n   This method should return a new iterator object that can iterate\n   over all the objects in the container.  For mappings, it should\n   iterate over the keys of the container, and should also be made\n   available as the method ``keys()``.\n\n   Iterator objects also need to implement this method; they are\n   required to return themselves.  For more information on iterator\n   objects, see *Iterator Types*.\n\nobject.__reversed__(self)\n\n   Called (if present) by the ``reversed()`` builtin to implement\n   reverse iteration.  It should return a new iterator object that\n   iterates over all the objects in the container in reverse order.\n\n   If the ``__reversed__()`` method is not provided, the\n   ``reversed()`` builtin will fall back to using the sequence\n   protocol (``__len__()`` and ``__getitem__()``).  Objects that\n   support the sequence protocol should only provide\n   ``__reversed__()`` if they can provide an implementation that is\n   more efficient than the one provided by ``reversed()``.\n\nThe membership test operators (``in`` and ``not in``) are normally\nimplemented as an iteration through a sequence.  However, container\nobjects can supply the following special method with a more efficient\nimplementation, which also does not require the object be a sequence.\n\nobject.__contains__(self, item)\n\n   Called to implement membership test operators.  Should return true\n   if *item* is in *self*, false otherwise.  For mapping objects, this\n   should consider the keys of the mapping rather than the values or\n   the key-item pairs.\n\n\nEmulating numeric types\n=======================\n\nThe following methods can be defined to emulate numeric objects.\nMethods corresponding to operations that are not supported by the\nparticular kind of number implemented (e.g., bitwise operations for\nnon-integral numbers) should be left undefined.\n\nobject.__add__(self, other)\nobject.__sub__(self, other)\nobject.__mul__(self, other)\nobject.__truediv__(self, other)\nobject.__floordiv__(self, other)\nobject.__mod__(self, other)\nobject.__divmod__(self, other)\nobject.__pow__(self, other[, modulo])\nobject.__lshift__(self, other)\nobject.__rshift__(self, other)\nobject.__and__(self, other)\nobject.__xor__(self, other)\nobject.__or__(self, other)\n\n   These methods are called to implement the binary arithmetic\n   operations (``+``, ``-``, ``*``, ``/``, ``//``, ``%``,\n   ``divmod()``, ``pow()``, ``**``, ``<<``, ``>>``, ``&``, ``^``,\n   ``|``).  For instance, to evaluate the expression ``x + y``, where\n   *x* is an instance of a class that has an ``__add__()`` method,\n   ``x.__add__(y)`` is called.  The ``__divmod__()`` method should be\n   the equivalent to using ``__floordiv__()`` and ``__mod__()``; it\n   should not be related to ``__truediv__()``.  Note that\n   ``__pow__()`` should be defined to accept an optional third\n   argument if the ternary version of the built-in ``pow()`` function\n   is to be supported.\n\n   If one of those methods does not support the operation with the\n   supplied arguments, it should return ``NotImplemented``.\n\nobject.__radd__(self, other)\nobject.__rsub__(self, other)\nobject.__rmul__(self, other)\nobject.__rtruediv__(self, other)\nobject.__rfloordiv__(self, other)\nobject.__rmod__(self, other)\nobject.__rdivmod__(self, other)\nobject.__rpow__(self, other)\nobject.__rlshift__(self, other)\nobject.__rrshift__(self, other)\nobject.__rand__(self, other)\nobject.__rxor__(self, other)\nobject.__ror__(self, other)\n\n   These methods are called to implement the binary arithmetic\n   operations (``+``, ``-``, ``*``, ``/``, ``//``, ``%``,\n   ``divmod()``, ``pow()``, ``**``, ``<<``, ``>>``, ``&``, ``^``,\n   ``|``) with reflected (swapped) operands. These functions are only\n   called if the left operand does not support the corresponding\n   operation and the operands are of different types. [3]  For\n   instance, to evaluate the expression ``x - y``, where *y* is an\n   instance of a class that has an ``__rsub__()`` method,\n   ``y.__rsub__(x)`` is called if ``x.__sub__(y)`` returns\n   *NotImplemented*.\n\n   Note that ternary ``pow()`` will not try calling ``__rpow__()``\n   (the coercion rules would become too complicated).\n\n   Note: If the right operand\'s type is a subclass of the left operand\'s\n     type and that subclass provides the reflected method for the\n     operation, this method will be called before the left operand\'s\n     non-reflected method.  This behavior allows subclasses to\n     override their ancestors\' operations.\n\nobject.__iadd__(self, other)\nobject.__isub__(self, other)\nobject.__imul__(self, other)\nobject.__itruediv__(self, other)\nobject.__ifloordiv__(self, other)\nobject.__imod__(self, other)\nobject.__ipow__(self, other[, modulo])\nobject.__ilshift__(self, other)\nobject.__irshift__(self, other)\nobject.__iand__(self, other)\nobject.__ixor__(self, other)\nobject.__ior__(self, other)\n\n   These methods are called to implement the augmented arithmetic\n   assignments (``+=``, ``-=``, ``*=``, ``/=``, ``//=``, ``%=``,\n   ``**=``, ``<<=``, ``>>=``, ``&=``, ``^=``, ``|=``).  These methods\n   should attempt to do the operation in-place (modifying *self*) and\n   return the result (which could be, but does not have to be,\n   *self*).  If a specific method is not defined, the augmented\n   assignment falls back to the normal methods.  For instance, to\n   execute the statement ``x += y``, where *x* is an instance of a\n   class that has an ``__iadd__()`` method, ``x.__iadd__(y)`` is\n   called.  If *x* is an instance of a class that does not define a\n   ``__iadd__()`` method, ``x.__add__(y)`` and ``y.__radd__(x)`` are\n   considered, as with the evaluation of ``x + y``.\n\nobject.__neg__(self)\nobject.__pos__(self)\nobject.__abs__(self)\nobject.__invert__(self)\n\n   Called to implement the unary arithmetic operations (``-``, ``+``,\n   ``abs()`` and ``~``).\n\nobject.__complex__(self)\nobject.__int__(self)\nobject.__float__(self)\nobject.__round__(self[, n])\n\n   Called to implement the built-in functions ``complex()``,\n   ``int()``, ``float()`` and ``round()``.  Should return a value of\n   the appropriate type.\n\nobject.__index__(self)\n\n   Called to implement ``operator.index()``.  Also called whenever\n   Python needs an integer object (such as in slicing, or in the\n   built-in ``bin()``, ``hex()`` and ``oct()`` functions). Must return\n   an integer.\n\n\nWith Statement Context Managers\n===============================\n\nA *context manager* is an object that defines the runtime context to\nbe established when executing a ``with`` statement. The context\nmanager handles the entry into, and the exit from, the desired runtime\ncontext for the execution of the block of code.  Context managers are\nnormally invoked using the ``with`` statement (described in section\n*The with statement*), but can also be used by directly invoking their\nmethods.\n\nTypical uses of context managers include saving and restoring various\nkinds of global state, locking and unlocking resources, closing opened\nfiles, etc.\n\nFor more information on context managers, see *Context Manager Types*.\n\nobject.__enter__(self)\n\n   Enter the runtime context related to this object. The ``with``\n   statement will bind this method\'s return value to the target(s)\n   specified in the ``as`` clause of the statement, if any.\n\nobject.__exit__(self, exc_type, exc_value, traceback)\n\n   Exit the runtime context related to this object. The parameters\n   describe the exception that caused the context to be exited. If the\n   context was exited without an exception, all three arguments will\n   be ``None``.\n\n   If an exception is supplied, and the method wishes to suppress the\n   exception (i.e., prevent it from being propagated), it should\n   return a true value. Otherwise, the exception will be processed\n   normally upon exit from this method.\n\n   Note that ``__exit__()`` methods should not reraise the passed-in\n   exception; this is the caller\'s responsibility.\n\nSee also:\n\n   **PEP 0343** - The "with" statement\n      The specification, background, and examples for the Python\n      ``with`` statement.\n\n\nSpecial method lookup\n=====================\n\nFor custom classes, implicit invocations of special methods are only\nguaranteed to work correctly if defined on an object\'s type, not in\nthe object\'s instance dictionary.  That behaviour is the reason why\nthe following code raises an exception:\n\n   >>> class C(object):\n   ...     pass\n   ...\n   >>> c = C()\n   >>> c.__len__ = lambda: 5\n   >>> len(c)\n   Traceback (most recent call last):\n     File "<stdin>", line 1, in <module>\n   TypeError: object of type \'C\' has no len()\n\nThe rationale behind this behaviour lies with a number of special\nmethods such as ``__hash__()`` and ``__repr__()`` that are implemented\nby all objects, including type objects. If the implicit lookup of\nthese methods used the conventional lookup process, they would fail\nwhen invoked on the type object itself:\n\n   >>> 1 .__hash__() == hash(1)\n   True\n   >>> int.__hash__() == hash(int)\n   Traceback (most recent call last):\n     File "<stdin>", line 1, in <module>\n   TypeError: descriptor \'__hash__\' of \'int\' object needs an argument\n\nIncorrectly attempting to invoke an unbound method of a class in this\nway is sometimes referred to as \'metaclass confusion\', and is avoided\nby bypassing the instance when looking up special methods:\n\n   >>> type(1).__hash__(1) == hash(1)\n   True\n   >>> type(int).__hash__(int) == hash(int)\n   True\n\nIn addition to bypassing any instance attributes in the interest of\ncorrectness, implicit special method lookup generally also bypasses\nthe ``__getattribute__()`` method even of the object\'s metaclass:\n\n   >>> class Meta(type):\n   ...    def __getattribute__(*args):\n   ...       print("Metaclass getattribute invoked")\n   ...       return type.__getattribute__(*args)\n   ...\n   >>> class C(object, metaclass=Meta):\n   ...     def __len__(self):\n   ...         return 10\n   ...     def __getattribute__(*args):\n   ...         print("Class getattribute invoked")\n   ...         return object.__getattribute__(*args)\n   ...\n   >>> c = C()\n   >>> c.__len__()                 # Explicit lookup via instance\n   Class getattribute invoked\n   10\n   >>> type(c).__len__(c)          # Explicit lookup via type\n   Metaclass getattribute invoked\n   10\n   >>> len(c)                      # Implicit lookup\n   10\n\nBypassing the ``__getattribute__()`` machinery in this fashion\nprovides significant scope for speed optimisations within the\ninterpreter, at the cost of some flexibility in the handling of\nspecial methods (the special method *must* be set on the class object\nitself in order to be consistently invoked by the interpreter).\n\n-[ Footnotes ]-\n\n[1] It *is* possible in some cases to change an object\'s type, under\n    certain controlled conditions. It generally isn\'t a good idea\n    though, since it can lead to some very strange behaviour if it is\n    handled incorrectly.\n\n[2] A descriptor can define any combination of ``__get__()``,\n    ``__set__()`` and ``__delete__()``.  If it does not define\n    ``__get__()``, then accessing the attribute even on an instance\n    will return the descriptor object itself.  If the descriptor\n    defines ``__set__()`` and/or ``__delete__()``, it is a data\n    descriptor; if it defines neither, it is a non-data descriptor.\n\n[3] For operands of the same type, it is assumed that if the non-\n    reflected method (such as ``__add__()``) fails the operation is\n    not supported, which is why the reflected method is not called.\n',
- 'string-methods': '\nString Methods\n**************\n\nString objects support the methods listed below.  Note that none of\nthese methods take keyword arguments.\n\nIn addition, Python\'s strings support the sequence type methods\ndescribed in the *Sequence Types --- str, bytes, bytearray, list,\ntuple, range* section. To output formatted strings, see the *String\nFormatting* section. Also, see the ``re`` module for string functions\nbased on regular expressions.\n\nstr.capitalize()\n\n   Return a copy of the string with only its first character\n   capitalized.\n\nstr.center(width[, fillchar])\n\n   Return centered in a string of length *width*. Padding is done\n   using the specified *fillchar* (default is a space).\n\nstr.count(sub[, start[, end]])\n\n   Return the number of non-overlapping occurrences of substring *sub*\n   in the range [*start*, *end*].  Optional arguments *start* and\n   *end* are interpreted as in slice notation.\n\nstr.encode([encoding[, errors]])\n\n   Return an encoded version of the string as a bytes object.  Default\n   encoding is the current default string encoding.  *errors* may be\n   given to set a different error handling scheme.  The default for\n   *errors* is ``\'strict\'``, meaning that encoding errors raise a\n   ``UnicodeError``.  Other possible values are ``\'ignore\'``,\n   ``\'replace\'``, ``\'xmlcharrefreplace\'``, ``\'backslashreplace\'`` and\n   any other name registered via ``codecs.register_error()``, see\n   section *Codec Base Classes*. For a list of possible encodings, see\n   section *Standard Encodings*.\n\nstr.endswith(suffix[, start[, end]])\n\n   Return ``True`` if the string ends with the specified *suffix*,\n   otherwise return ``False``.  *suffix* can also be a tuple of\n   suffixes to look for.  With optional *start*, test beginning at\n   that position.  With optional *end*, stop comparing at that\n   position.\n\nstr.expandtabs([tabsize])\n\n   Return a copy of the string where all tab characters are replaced\n   by one or more spaces, depending on the current column and the\n   given tab size.  The column number is reset to zero after each\n   newline occurring in the string. If *tabsize* is not given, a tab\n   size of ``8`` characters is assumed.  This doesn\'t understand other\n   non-printing characters or escape sequences.\n\nstr.find(sub[, start[, end]])\n\n   Return the lowest index in the string where substring *sub* is\n   found, such that *sub* is contained in the range [*start*, *end*].\n   Optional arguments *start* and *end* are interpreted as in slice\n   notation.  Return ``-1`` if *sub* is not found.\n\nstr.format(*args, **kwargs)\n\n   Perform a string formatting operation.  The *format_string*\n   argument can contain literal text or replacement fields delimited\n   by braces ``{}``.  Each replacement field contains either the\n   numeric index of a positional argument, or the name of a keyword\n   argument.  Returns a copy of *format_string* where each replacement\n   field is replaced with the string value of the corresponding\n   argument.\n\n   >>> "The sum of 1 + 2 is {0}".format(1+2)\n   \'The sum of 1 + 2 is 3\'\n\n   See *Format String Syntax* for a description of the various\n   formatting options that can be specified in format strings.\n\nstr.index(sub[, start[, end]])\n\n   Like ``find()``, but raise ``ValueError`` when the substring is not\n   found.\n\nstr.isalnum()\n\n   Return true if all characters in the string are alphanumeric and\n   there is at least one character, false otherwise.\n\nstr.isalpha()\n\n   Return true if all characters in the string are alphabetic and\n   there is at least one character, false otherwise.\n\nstr.isdecimal()\n\n   Return true if all characters in the string are decimal characters\n   and there is at least one character, false otherwise. Decimal\n   characters include digit characters, and all characters that that\n   can be used to form decimal-radix numbers, e.g. U+0660, ARABIC-\n   INDIC DIGIT ZERO.\n\nstr.isdigit()\n\n   Return true if all characters in the string are digits and there is\n   at least one character, false otherwise.\n\nstr.isidentifier()\n\n   Return true if the string is a valid identifier according to the\n   language definition, section *Identifiers and keywords*.\n\nstr.islower()\n\n   Return true if all cased characters in the string are lowercase and\n   there is at least one cased character, false otherwise.\n\nstr.isnumeric()\n\n   Return true if all characters in the string are numeric characters,\n   and there is at least one character, false otherwise. Numeric\n   characters include digit characters, and all characters that have\n   the Unicode numeric value property, e.g. U+2155, VULGAR FRACTION\n   ONE FIFTH.\n\nstr.isprintable()\n\n   Return true if all characters in the string are printable or the\n   string is empty, false otherwise.  Nonprintable characters are\n   those characters defined in the Unicode character database as\n   "Other" or "Separator", excepting the ASCII space (0x20) which is\n   considered printable.  (Note that printable characters in this\n   context are those which should not be escaped when ``repr()`` is\n   invoked on a string.  It has no bearing on the handling of strings\n   written to ``sys.stdout`` or ``sys.stderr``.)\n\nstr.isspace()\n\n   Return true if there are only whitespace characters in the string\n   and there is at least one character, false otherwise.\n\nstr.istitle()\n\n   Return true if the string is a titlecased string and there is at\n   least one character, for example uppercase characters may only\n   follow uncased characters and lowercase characters only cased ones.\n   Return false otherwise.\n\nstr.isupper()\n\n   Return true if all cased characters in the string are uppercase and\n   there is at least one cased character, false otherwise.\n\nstr.join(seq)\n\n   Return a string which is the concatenation of the strings in the\n   sequence *seq*.  A ``TypeError`` will be raised if there are any\n   non-string values in *seq*, including ``bytes`` objects.  The\n   separator between elements is the string providing this method.\n\nstr.ljust(width[, fillchar])\n\n   Return the string left justified in a string of length *width*.\n   Padding is done using the specified *fillchar* (default is a\n   space).  The original string is returned if *width* is less than\n   ``len(s)``.\n\nstr.lower()\n\n   Return a copy of the string converted to lowercase.\n\nstr.lstrip([chars])\n\n   Return a copy of the string with leading characters removed.  The\n   *chars* argument is a string specifying the set of characters to be\n   removed.  If omitted or ``None``, the *chars* argument defaults to\n   removing whitespace.  The *chars* argument is not a prefix; rather,\n   all combinations of its values are stripped:\n\n   >>> \'   spacious   \'.lstrip()\n   \'spacious   \'\n   >>> \'www.example.com\'.lstrip(\'cmowz.\')\n   \'example.com\'\n\nstatic str.maketrans(x[, y[, z]])\n\n   This static method returns a translation table usable for\n   ``str.translate()``.\n\n   If there is only one argument, it must be a dictionary mapping\n   Unicode ordinals (integers) or characters (strings of length 1) to\n   Unicode ordinals, strings (of arbitrary lengths) or None.\n   Character keys will then be converted to ordinals.\n\n   If there are two arguments, they must be strings of equal length,\n   and in the resulting dictionary, each character in x will be mapped\n   to the character at the same position in y.  If there is a third\n   argument, it must be a string, whose characters will be mapped to\n   None in the result.\n\nstr.partition(sep)\n\n   Split the string at the first occurrence of *sep*, and return a\n   3-tuple containing the part before the separator, the separator\n   itself, and the part after the separator.  If the separator is not\n   found, return a 3-tuple containing the string itself, followed by\n   two empty strings.\n\nstr.replace(old, new[, count])\n\n   Return a copy of the string with all occurrences of substring *old*\n   replaced by *new*.  If the optional argument *count* is given, only\n   the first *count* occurrences are replaced.\n\nstr.rfind(sub[, start[, end]])\n\n   Return the highest index in the string where substring *sub* is\n   found, such that *sub* is contained within s[start,end].  Optional\n   arguments *start* and *end* are interpreted as in slice notation.\n   Return ``-1`` on failure.\n\nstr.rindex(sub[, start[, end]])\n\n   Like ``rfind()`` but raises ``ValueError`` when the substring *sub*\n   is not found.\n\nstr.rjust(width[, fillchar])\n\n   Return the string right justified in a string of length *width*.\n   Padding is done using the specified *fillchar* (default is a\n   space). The original string is returned if *width* is less than\n   ``len(s)``.\n\nstr.rpartition(sep)\n\n   Split the string at the last occurrence of *sep*, and return a\n   3-tuple containing the part before the separator, the separator\n   itself, and the part after the separator.  If the separator is not\n   found, return a 3-tuple containing two empty strings, followed by\n   the string itself.\n\nstr.rsplit([sep[, maxsplit]])\n\n   Return a list of the words in the string, using *sep* as the\n   delimiter string. If *maxsplit* is given, at most *maxsplit* splits\n   are done, the *rightmost* ones.  If *sep* is not specified or\n   ``None``, any whitespace string is a separator.  Except for\n   splitting from the right, ``rsplit()`` behaves like ``split()``\n   which is described in detail below.\n\nstr.rstrip([chars])\n\n   Return a copy of the string with trailing characters removed.  The\n   *chars* argument is a string specifying the set of characters to be\n   removed.  If omitted or ``None``, the *chars* argument defaults to\n   removing whitespace.  The *chars* argument is not a suffix; rather,\n   all combinations of its values are stripped:\n\n   >>> \'   spacious   \'.rstrip()\n   \'   spacious\'\n   >>> \'mississippi\'.rstrip(\'ipz\')\n   \'mississ\'\n\nstr.split([sep[, maxsplit]])\n\n   Return a list of the words in the string, using *sep* as the\n   delimiter string.  If *maxsplit* is given, at most *maxsplit*\n   splits are done (thus, the list will have at most ``maxsplit+1``\n   elements).  If *maxsplit* is not specified, then there is no limit\n   on the number of splits (all possible splits are made).\n\n   If *sep* is given, consecutive delimiters are not grouped together\n   and are deemed to delimit empty strings (for example,\n   ``\'1,,2\'.split(\',\')`` returns ``[\'1\', \'\', \'2\']``).  The *sep*\n   argument may consist of multiple characters (for example,\n   ``\'1<>2<>3\'.split(\'<>\')`` returns ``[\'1\', \'2\', \'3\']``). Splitting\n   an empty string with a specified separator returns ``[\'\']``.\n\n   If *sep* is not specified or is ``None``, a different splitting\n   algorithm is applied: runs of consecutive whitespace are regarded\n   as a single separator, and the result will contain no empty strings\n   at the start or end if the string has leading or trailing\n   whitespace.  Consequently, splitting an empty string or a string\n   consisting of just whitespace with a ``None`` separator returns\n   ``[]``.\n\n   For example, ``\' 1  2   3  \'.split()`` returns ``[\'1\', \'2\', \'3\']``,\n   and ``\'  1  2   3  \'.split(None, 1)`` returns ``[\'1\', \'2   3  \']``.\n\nstr.splitlines([keepends])\n\n   Return a list of the lines in the string, breaking at line\n   boundaries.  Line breaks are not included in the resulting list\n   unless *keepends* is given and true.\n\nstr.startswith(prefix[, start[, end]])\n\n   Return ``True`` if string starts with the *prefix*, otherwise\n   return ``False``. *prefix* can also be a tuple of prefixes to look\n   for.  With optional *start*, test string beginning at that\n   position.  With optional *end*, stop comparing string at that\n   position.\n\nstr.strip([chars])\n\n   Return a copy of the string with the leading and trailing\n   characters removed. The *chars* argument is a string specifying the\n   set of characters to be removed. If omitted or ``None``, the\n   *chars* argument defaults to removing whitespace. The *chars*\n   argument is not a prefix or suffix; rather, all combinations of its\n   values are stripped:\n\n   >>> \'   spacious   \'.strip()\n   \'spacious\'\n   >>> \'www.example.com\'.strip(\'cmowz.\')\n   \'example\'\n\nstr.swapcase()\n\n   Return a copy of the string with uppercase characters converted to\n   lowercase and vice versa.\n\nstr.title()\n\n   Return a titlecased version of the string: words start with\n   uppercase characters, all remaining cased characters are lowercase.\n\nstr.translate(map)\n\n   Return a copy of the *s* where all characters have been mapped\n   through the *map* which must be a dictionary of Unicode ordinals\n   (integers) to Unicode ordinals, strings or ``None``.  Unmapped\n   characters are left untouched. Characters mapped to ``None`` are\n   deleted.\n\n   You can use ``str.maketrans()`` to create a translation map from\n   character-to-character mappings in different formats.\n\n   Note: An even more flexible approach is to create a custom character\n     mapping codec using the ``codecs`` module (see\n     ``encodings.cp1251`` for an example).\n\nstr.upper()\n\n   Return a copy of the string converted to uppercase.\n\nstr.zfill(width)\n\n   Return the numeric string left filled with zeros in a string of\n   length *width*.  A sign prefix is handled correctly.  The original\n   string is returned if *width* is less than ``len(s)``.\n',
- 'strings': '\nString and Bytes literals\n*************************\n\nString literals are described by the following lexical definitions:\n\n   stringliteral   ::= [stringprefix](shortstring | longstring)\n   stringprefix    ::= "r" | "R"\n   shortstring     ::= "\'" shortstringitem* "\'" | \'"\' shortstringitem* \'"\'\n   longstring      ::= "\'\'\'" longstringitem* "\'\'\'" | \'"""\' longstringitem* \'"""\'\n   shortstringitem ::= shortstringchar | stringescapeseq\n   longstringitem  ::= longstringchar | stringescapeseq\n   shortstringchar ::= <any source character except "\\" or newline or the quote>\n   longstringchar  ::= <any source character except "\\">\n   stringescapeseq ::= "\\" <any source character>\n\n   bytesliteral   ::= bytesprefix(shortbytes | longbytes)\n   bytesprefix    ::= "b" | "B"\n   shortbytes     ::= "\'" shortbytesitem* "\'" | \'"\' shortbytesitem* \'"\'\n   longbytes      ::= "\'\'\'" longbytesitem* "\'\'\'" | \'"""\' longbytesitem* \'"""\'\n   shortbytesitem ::= shortbyteschar | bytesescapeseq\n   longbytesitem  ::= longbyteschar | bytesescapeseq\n   shortbyteschar ::= <any ASCII character except "\\" or newline or the quote>\n   longbyteschar  ::= <any ASCII character except "\\">\n   bytesescapeseq ::= "\\" <any ASCII character>\n\nOne syntactic restriction not indicated by these productions is that\nwhitespace is not allowed between the **stringprefix** or\n**bytesprefix** and the rest of the literal. The source character set\nis defined by the encoding declaration; it is UTF-8 if no encoding\ndeclaration is given in the source file; see section *Encoding\ndeclarations*.\n\nIn plain English: Both types of literals can be enclosed in matching\nsingle quotes (``\'``) or double quotes (``"``).  They can also be\nenclosed in matching groups of three single or double quotes (these\nare generally referred to as *triple-quoted strings*).  The backslash\n(``\\``) character is used to escape characters that otherwise have a\nspecial meaning, such as newline, backslash itself, or the quote\ncharacter.\n\nString literals may optionally be prefixed with a letter ``\'r\'`` or\n``\'R\'``; such strings are called *raw strings* and treat backslashes\nas literal characters.  As a result, ``\'\\U\'`` and ``\'\\u\'`` escapes in\nraw strings are not treated specially.\n\nBytes literals are always prefixed with ``\'b\'`` or ``\'B\'``; they\nproduce an instance of the ``bytes`` type instead of the ``str`` type.\nThey may only contain ASCII characters; bytes with a numeric value of\n128 or greater must be expressed with escapes.\n\nIn triple-quoted strings, unescaped newlines and quotes are allowed\n(and are retained), except that three unescaped quotes in a row\nterminate the string.  (A "quote" is the character used to open the\nstring, i.e. either ``\'`` or ``"``.)\n\nUnless an ``\'r\'`` or ``\'R\'`` prefix is present, escape sequences in\nstrings are interpreted according to rules similar to those used by\nStandard C.  The recognized escape sequences are:\n\n+-------------------+-----------------------------------+---------+\n| Escape Sequence   | Meaning                           | Notes   |\n+===================+===================================+=========+\n| ``\\newline``      | Backslash and newline ignored     |         |\n+-------------------+-----------------------------------+---------+\n| ``\\\\``            | Backslash (``\\``)                 |         |\n+-------------------+-----------------------------------+---------+\n| ``\\\'``            | Single quote (``\'``)              |         |\n+-------------------+-----------------------------------+---------+\n| ``\\"``            | Double quote (``"``)              |         |\n+-------------------+-----------------------------------+---------+\n| ``\\a``            | ASCII Bell (BEL)                  |         |\n+-------------------+-----------------------------------+---------+\n| ``\\b``            | ASCII Backspace (BS)              |         |\n+-------------------+-----------------------------------+---------+\n| ``\\f``            | ASCII Formfeed (FF)               |         |\n+-------------------+-----------------------------------+---------+\n| ``\\n``            | ASCII Linefeed (LF)               |         |\n+-------------------+-----------------------------------+---------+\n| ``\\r``            | ASCII Carriage Return (CR)        |         |\n+-------------------+-----------------------------------+---------+\n| ``\\t``            | ASCII Horizontal Tab (TAB)        |         |\n+-------------------+-----------------------------------+---------+\n| ``\\v``            | ASCII Vertical Tab (VT)           |         |\n+-------------------+-----------------------------------+---------+\n| ``\\ooo``          | Character with octal value *ooo*  | (1,3)   |\n+-------------------+-----------------------------------+---------+\n| ``\\xhh``          | Character with hex value *hh*     | (2,3)   |\n+-------------------+-----------------------------------+---------+\n\nEscape sequences only recognized in string literals are:\n\n+-------------------+-----------------------------------+---------+\n| Escape Sequence   | Meaning                           | Notes   |\n+===================+===================================+=========+\n| ``\\N{name}``      | Character named *name* in the     |         |\n|                   | Unicode database                  |         |\n+-------------------+-----------------------------------+---------+\n| ``\\uxxxx``        | Character with 16-bit hex value   | (4)     |\n|                   | *xxxx*                            |         |\n+-------------------+-----------------------------------+---------+\n| ``\\Uxxxxxxxx``    | Character with 32-bit hex value   | (5)     |\n|                   | *xxxxxxxx*                        |         |\n+-------------------+-----------------------------------+---------+\n\nNotes:\n\n1. As in Standard C, up to three octal digits are accepted.\n\n2. Unlike in Standard C, at most two hex digits are accepted.\n\n3. In a bytes literal, hexadecimal and octal escapes denote the byte\n   with the given value. In a string literal, these escapes denote a\n   Unicode character with the given value.\n\n4. Individual code units which form parts of a surrogate pair can be\n   encoded using this escape sequence. Unlike in Standard C, exactly\n   two hex digits are required.\n\n5. Any Unicode character can be encoded this way, but characters\n   outside the Basic Multilingual Plane (BMP) will be encoded using a\n   surrogate pair if Python is compiled to use 16-bit code units (the\n   default).  Individual code units which form parts of a surrogate\n   pair can be encoded using this escape sequence.\n\nUnlike Standard C, all unrecognized escape sequences are left in the\nstring unchanged, i.e., *the backslash is left in the string*.  (This\nbehavior is useful when debugging: if an escape sequence is mistyped,\nthe resulting output is more easily recognized as broken.)  It is also\nimportant to note that the escape sequences only recognized in string\nliterals fall into the category of unrecognized escapes for bytes\nliterals.\n\nEven in a raw string, string quotes can be escaped with a backslash,\nbut the backslash remains in the string; for example, ``r"\\""`` is a\nvalid string literal consisting of two characters: a backslash and a\ndouble quote; ``r"\\"`` is not a valid string literal (even a raw\nstring cannot end in an odd number of backslashes).  Specifically, *a\nraw string cannot end in a single backslash* (since the backslash\nwould escape the following quote character).  Note also that a single\nbackslash followed by a newline is interpreted as those two characters\nas part of the string, *not* as a line continuation.\n',
+ 'specialattrs': "\nSpecial Attributes\n******************\n\nThe implementation adds a few special read-only attributes to several\nobject types, where they are relevant.  Some of these are not reported\nby the ``dir()`` built-in function.\n\nobject.__dict__\n\n   A dictionary or other mapping object used to store an object's\n   (writable) attributes.\n\ninstance.__class__\n\n   The class to which a class instance belongs.\n\nclass.__bases__\n\n   The tuple of base classes of a class object.\n\nclass.__name__\n\n   The name of the class or type.\n\nThe following attributes are only supported by *new-style class*es.\n\nclass.__mro__\n\n   This attribute is a tuple of classes that are considered when\n   looking for base classes during method resolution.\n\nclass.mro()\n\n   This method can be overridden by a metaclass to customize the\n   method resolution order for its instances.  It is called at class\n   instantiation, and its result is stored in ``__mro__``.\n\nclass.__subclasses__()\n\n   Each new-style class keeps a list of weak references to its\n   immediate subclasses.  This method returns a list of all those\n   references still alive. Example:\n\n      >>> int.__subclasses__()\n      [<type 'bool'>]\n\n-[ Footnotes ]-\n\n[1] Additional information on these special methods may be found in\n    the Python Reference Manual (*Basic customization*).\n\n[2] As a consequence, the list ``[1, 2]`` is considered equal to\n    ``[1.0, 2.0]``, and similarly for tuples.\n\n[3] They must have since the parser can't tell the type of the\n    operands.\n\n[4] To format only a tuple you should therefore provide a singleton\n    tuple whose only element is the tuple to be formatted.\n",
+ 'specialnames': '\nSpecial method names\n********************\n\nA class can implement certain operations that are invoked by special\nsyntax (such as arithmetic operations or subscripting and slicing) by\ndefining methods with special names. This is Python\'s approach to\n*operator overloading*, allowing classes to define their own behavior\nwith respect to language operators.  For instance, if a class defines\na method named ``__getitem__()``, and ``x`` is an instance of this\nclass, then ``x[i]`` is roughly equivalent to ``type(x).__getitem__(x,\ni)``.  Except where mentioned, attempts to execute an operation raise\nan exception when no appropriate method is defined (typically\n``AttributeError`` or ``TypeError``).\n\nWhen implementing a class that emulates any built-in type, it is\nimportant that the emulation only be implemented to the degree that it\nmakes sense for the object being modelled.  For example, some\nsequences may work well with retrieval of individual elements, but\nextracting a slice may not make sense.  (One example of this is the\n``NodeList`` interface in the W3C\'s Document Object Model.)\n\n\nBasic customization\n===================\n\nobject.__new__(cls[, ...])\n\n   Called to create a new instance of class *cls*.  ``__new__()`` is a\n   static method (special-cased so you need not declare it as such)\n   that takes the class of which an instance was requested as its\n   first argument.  The remaining arguments are those passed to the\n   object constructor expression (the call to the class).  The return\n   value of ``__new__()`` should be the new object instance (usually\n   an instance of *cls*).\n\n   Typical implementations create a new instance of the class by\n   invoking the superclass\'s ``__new__()`` method using\n   ``super(currentclass, cls).__new__(cls[, ...])`` with appropriate\n   arguments and then modifying the newly-created instance as\n   necessary before returning it.\n\n   If ``__new__()`` returns an instance of *cls*, then the new\n   instance\'s ``__init__()`` method will be invoked like\n   ``__init__(self[, ...])``, where *self* is the new instance and the\n   remaining arguments are the same as were passed to ``__new__()``.\n\n   If ``__new__()`` does not return an instance of *cls*, then the new\n   instance\'s ``__init__()`` method will not be invoked.\n\n   ``__new__()`` is intended mainly to allow subclasses of immutable\n   types (like int, str, or tuple) to customize instance creation.  It\n   is also commonly overridden in custom metaclasses in order to\n   customize class creation.\n\nobject.__init__(self[, ...])\n\n   Called when the instance is created.  The arguments are those\n   passed to the class constructor expression.  If a base class has an\n   ``__init__()`` method, the derived class\'s ``__init__()`` method,\n   if any, must explicitly call it to ensure proper initialization of\n   the base class part of the instance; for example:\n   ``BaseClass.__init__(self, [args...])``.  As a special constraint\n   on constructors, no value may be returned; doing so will cause a\n   ``TypeError`` to be raised at runtime.\n\nobject.__del__(self)\n\n   Called when the instance is about to be destroyed.  This is also\n   called a destructor.  If a base class has a ``__del__()`` method,\n   the derived class\'s ``__del__()`` method, if any, must explicitly\n   call it to ensure proper deletion of the base class part of the\n   instance.  Note that it is possible (though not recommended!) for\n   the ``__del__()`` method to postpone destruction of the instance by\n   creating a new reference to it.  It may then be called at a later\n   time when this new reference is deleted.  It is not guaranteed that\n   ``__del__()`` methods are called for objects that still exist when\n   the interpreter exits.\n\n   Note: ``del x`` doesn\'t directly call ``x.__del__()`` --- the former\n     decrements the reference count for ``x`` by one, and the latter\n     is only called when ``x``\'s reference count reaches zero.  Some\n     common situations that may prevent the reference count of an\n     object from going to zero include: circular references between\n     objects (e.g., a doubly-linked list or a tree data structure with\n     parent and child pointers); a reference to the object on the\n     stack frame of a function that caught an exception (the traceback\n     stored in ``sys.exc_info()[2]`` keeps the stack frame alive); or\n     a reference to the object on the stack frame that raised an\n     unhandled exception in interactive mode (the traceback stored in\n     ``sys.last_traceback`` keeps the stack frame alive).  The first\n     situation can only be remedied by explicitly breaking the cycles;\n     the latter two situations can be resolved by storing ``None`` in\n     ``sys.last_traceback``. Circular references which are garbage are\n     detected when the option cycle detector is enabled (it\'s on by\n     default), but can only be cleaned up if there are no Python-\n     level ``__del__()`` methods involved. Refer to the documentation\n     for the ``gc`` module for more information about how\n     ``__del__()`` methods are handled by the cycle detector,\n     particularly the description of the ``garbage`` value.\n\n   Warning: Due to the precarious circumstances under which ``__del__()``\n     methods are invoked, exceptions that occur during their execution\n     are ignored, and a warning is printed to ``sys.stderr`` instead.\n     Also, when ``__del__()`` is invoked in response to a module being\n     deleted (e.g., when execution of the program is done), other\n     globals referenced by the ``__del__()`` method may already have\n     been deleted or in the process of being torn down (e.g. the\n     import machinery shutting down).  For this reason, ``__del__()``\n     methods should do the absolute minimum needed to maintain\n     external invariants.  Starting with version 1.5, Python\n     guarantees that globals whose name begins with a single\n     underscore are deleted from their module before other globals are\n     deleted; if no other references to such globals exist, this may\n     help in assuring that imported modules are still available at the\n     time when the ``__del__()`` method is called.\n\nobject.__repr__(self)\n\n   Called by the ``repr()`` built-in function to compute the\n   "official" string representation of an object.  If at all possible,\n   this should look like a valid Python expression that could be used\n   to recreate an object with the same value (given an appropriate\n   environment).  If this is not possible, a string of the form\n   ``<...some useful description...>`` should be returned. The return\n   value must be a string object. If a class defines ``__repr__()``\n   but not ``__str__()``, then ``__repr__()`` is also used when an\n   "informal" string representation of instances of that class is\n   required.\n\n   This is typically used for debugging, so it is important that the\n   representation is information-rich and unambiguous.\n\nobject.__str__(self)\n\n   Called by the ``str()`` built-in function and by the ``print()``\n   function to compute the "informal" string representation of an\n   object.  This differs from ``__repr__()`` in that it does not have\n   to be a valid Python expression: a more convenient or concise\n   representation may be used instead. The return value must be a\n   string object.\n\nobject.__format__(self, format_spec)\n\n   Called by the ``format()`` built-in function (and by extension, the\n   ``format()`` method of class ``str``) to produce a "formatted"\n   string representation of an object. The ``format_spec`` argument is\n   a string that contains a description of the formatting options\n   desired. The interpretation of the ``format_spec`` argument is up\n   to the type implementing ``__format__()``, however most classes\n   will either delegate formatting to one of the built-in types, or\n   use a similar formatting option syntax.\n\n   See *Format Specification Mini-Language* for a description of the\n   standard formatting syntax.\n\n   The return value must be a string object.\n\nobject.__lt__(self, other)\nobject.__le__(self, other)\nobject.__eq__(self, other)\nobject.__ne__(self, other)\nobject.__gt__(self, other)\nobject.__ge__(self, other)\n\n   These are the so-called "rich comparison" methods. The\n   correspondence between operator symbols and method names is as\n   follows: ``x<y`` calls ``x.__lt__(y)``, ``x<=y`` calls\n   ``x.__le__(y)``, ``x==y`` calls ``x.__eq__(y)``, ``x!=y`` calls\n   ``x.__ne__(y)``, ``x>y`` calls ``x.__gt__(y)``, and ``x>=y`` calls\n   ``x.__ge__(y)``.\n\n   A rich comparison method may return the singleton\n   ``NotImplemented`` if it does not implement the operation for a\n   given pair of arguments. By convention, ``False`` and ``True`` are\n   returned for a successful comparison. However, these methods can\n   return any value, so if the comparison operator is used in a\n   Boolean context (e.g., in the condition of an ``if`` statement),\n   Python will call ``bool()`` on the value to determine if the result\n   is true or false.\n\n   There are no implied relationships among the comparison operators.\n   The truth of ``x==y`` does not imply that ``x!=y`` is false.\n   Accordingly, when defining ``__eq__()``, one should also define\n   ``__ne__()`` so that the operators will behave as expected.  See\n   the paragraph on ``__hash__()`` for some important notes on\n   creating *hashable* objects which support custom comparison\n   operations and are usable as dictionary keys.\n\n   There are no swapped-argument versions of these methods (to be used\n   when the left argument does not support the operation but the right\n   argument does); rather, ``__lt__()`` and ``__gt__()`` are each\n   other\'s reflection, ``__le__()`` and ``__ge__()`` are each other\'s\n   reflection, and ``__eq__()`` and ``__ne__()`` are their own\n   reflection.\n\n   Arguments to rich comparison methods are never coerced.\n\n   To automatically generate ordering operations from a single root\n   operation, see ``functools.total_ordering()``.\n\nobject.__hash__(self)\n\n   Called by built-in function ``hash()`` and for operations on\n   members of hashed collections including ``set``, ``frozenset``, and\n   ``dict``.  ``__hash__()`` should return an integer.  The only\n   required property is that objects which compare equal have the same\n   hash value; it is advised to somehow mix together (e.g. using\n   exclusive or) the hash values for the components of the object that\n   also play a part in comparison of objects.\n\n   If a class does not define an ``__eq__()`` method it should not\n   define a ``__hash__()`` operation either; if it defines\n   ``__eq__()`` but not ``__hash__()``, its instances will not be\n   usable as items in hashable collections.  If a class defines\n   mutable objects and implements an ``__eq__()`` method, it should\n   not implement ``__hash__()``, since the implementation of hashable\n   collections requires that a key\'s hash value is immutable (if the\n   object\'s hash value changes, it will be in the wrong hash bucket).\n\n   User-defined classes have ``__eq__()`` and ``__hash__()`` methods\n   by default; with them, all objects compare unequal (except with\n   themselves) and ``x.__hash__()`` returns ``id(x)``.\n\n   Classes which inherit a ``__hash__()`` method from a parent class\n   but change the meaning of ``__eq__()`` such that the hash value\n   returned is no longer appropriate (e.g. by switching to a value-\n   based concept of equality instead of the default identity based\n   equality) can explicitly flag themselves as being unhashable by\n   setting ``__hash__ = None`` in the class definition. Doing so means\n   that not only will instances of the class raise an appropriate\n   ``TypeError`` when a program attempts to retrieve their hash value,\n   but they will also be correctly identified as unhashable when\n   checking ``isinstance(obj, collections.Hashable)`` (unlike classes\n   which define their own ``__hash__()`` to explicitly raise\n   ``TypeError``).\n\n   If a class that overrides ``__eq__()`` needs to retain the\n   implementation of ``__hash__()`` from a parent class, the\n   interpreter must be told this explicitly by setting ``__hash__ =\n   <ParentClass>.__hash__``. Otherwise the inheritance of\n   ``__hash__()`` will be blocked, just as if ``__hash__`` had been\n   explicitly set to ``None``.\n\nobject.__bool__(self)\n\n   Called to implement truth value testing and the built-in operation\n   ``bool()``; should return ``False`` or ``True``.  When this method\n   is not defined, ``__len__()`` is called, if it is defined, and the\n   object is considered true if its result is nonzero.  If a class\n   defines neither ``__len__()`` nor ``__bool__()``, all its instances\n   are considered true.\n\n\nCustomizing attribute access\n============================\n\nThe following methods can be defined to customize the meaning of\nattribute access (use of, assignment to, or deletion of ``x.name``)\nfor class instances.\n\nobject.__getattr__(self, name)\n\n   Called when an attribute lookup has not found the attribute in the\n   usual places (i.e. it is not an instance attribute nor is it found\n   in the class tree for ``self``).  ``name`` is the attribute name.\n   This method should return the (computed) attribute value or raise\n   an ``AttributeError`` exception.\n\n   Note that if the attribute is found through the normal mechanism,\n   ``__getattr__()`` is not called.  (This is an intentional asymmetry\n   between ``__getattr__()`` and ``__setattr__()``.) This is done both\n   for efficiency reasons and because otherwise ``__getattr__()``\n   would have no way to access other attributes of the instance.  Note\n   that at least for instance variables, you can fake total control by\n   not inserting any values in the instance attribute dictionary (but\n   instead inserting them in another object).  See the\n   ``__getattribute__()`` method below for a way to actually get total\n   control over attribute access.\n\nobject.__getattribute__(self, name)\n\n   Called unconditionally to implement attribute accesses for\n   instances of the class. If the class also defines\n   ``__getattr__()``, the latter will not be called unless\n   ``__getattribute__()`` either calls it explicitly or raises an\n   ``AttributeError``. This method should return the (computed)\n   attribute value or raise an ``AttributeError`` exception. In order\n   to avoid infinite recursion in this method, its implementation\n   should always call the base class method with the same name to\n   access any attributes it needs, for example,\n   ``object.__getattribute__(self, name)``.\n\n   Note: This method may still be bypassed when looking up special methods\n     as the result of implicit invocation via language syntax or\n     built-in functions. See *Special method lookup*.\n\nobject.__setattr__(self, name, value)\n\n   Called when an attribute assignment is attempted.  This is called\n   instead of the normal mechanism (i.e. store the value in the\n   instance dictionary). *name* is the attribute name, *value* is the\n   value to be assigned to it.\n\n   If ``__setattr__()`` wants to assign to an instance attribute, it\n   should call the base class method with the same name, for example,\n   ``object.__setattr__(self, name, value)``.\n\nobject.__delattr__(self, name)\n\n   Like ``__setattr__()`` but for attribute deletion instead of\n   assignment.  This should only be implemented if ``del obj.name`` is\n   meaningful for the object.\n\nobject.__dir__(self)\n\n   Called when ``dir()`` is called on the object.  A list must be\n   returned.\n\n\nImplementing Descriptors\n------------------------\n\nThe following methods only apply when an instance of the class\ncontaining the method (a so-called *descriptor* class) appears in the\nclass dictionary of another class, known as the *owner* class.  In the\nexamples below, "the attribute" refers to the attribute whose name is\nthe key of the property in the owner class\' ``__dict__``.\n\nobject.__get__(self, instance, owner)\n\n   Called to get the attribute of the owner class (class attribute\n   access) or of an instance of that class (instance attribute\n   access). *owner* is always the owner class, while *instance* is the\n   instance that the attribute was accessed through, or ``None`` when\n   the attribute is accessed through the *owner*.  This method should\n   return the (computed) attribute value or raise an\n   ``AttributeError`` exception.\n\nobject.__set__(self, instance, value)\n\n   Called to set the attribute on an instance *instance* of the owner\n   class to a new value, *value*.\n\nobject.__delete__(self, instance)\n\n   Called to delete the attribute on an instance *instance* of the\n   owner class.\n\n\nInvoking Descriptors\n--------------------\n\nIn general, a descriptor is an object attribute with "binding\nbehavior", one whose attribute access has been overridden by methods\nin the descriptor protocol:  ``__get__()``, ``__set__()``, and\n``__delete__()``. If any of those methods are defined for an object,\nit is said to be a descriptor.\n\nThe default behavior for attribute access is to get, set, or delete\nthe attribute from an object\'s dictionary. For instance, ``a.x`` has a\nlookup chain starting with ``a.__dict__[\'x\']``, then\n``type(a).__dict__[\'x\']``, and continuing through the base classes of\n``type(a)`` excluding metaclasses.\n\nHowever, if the looked-up value is an object defining one of the\ndescriptor methods, then Python may override the default behavior and\ninvoke the descriptor method instead.  Where this occurs in the\nprecedence chain depends on which descriptor methods were defined and\nhow they were called.\n\nThe starting point for descriptor invocation is a binding, ``a.x``.\nHow the arguments are assembled depends on ``a``:\n\nDirect Call\n   The simplest and least common call is when user code directly\n   invokes a descriptor method:    ``x.__get__(a)``.\n\nInstance Binding\n   If binding to an object instance, ``a.x`` is transformed into the\n   call: ``type(a).__dict__[\'x\'].__get__(a, type(a))``.\n\nClass Binding\n   If binding to a class, ``A.x`` is transformed into the call:\n   ``A.__dict__[\'x\'].__get__(None, A)``.\n\nSuper Binding\n   If ``a`` is an instance of ``super``, then the binding ``super(B,\n   obj).m()`` searches ``obj.__class__.__mro__`` for the base class\n   ``A`` immediately preceding ``B`` and then invokes the descriptor\n   with the call: ``A.__dict__[\'m\'].__get__(obj, A)``.\n\nFor instance bindings, the precedence of descriptor invocation depends\non the which descriptor methods are defined.  A descriptor can define\nany combination of ``__get__()``, ``__set__()`` and ``__delete__()``.\nIf it does not define ``__get__()``, then accessing the attribute will\nreturn the descriptor object itself unless there is a value in the\nobject\'s instance dictionary.  If the descriptor defines ``__set__()``\nand/or ``__delete__()``, it is a data descriptor; if it defines\nneither, it is a non-data descriptor.  Normally, data descriptors\ndefine both ``__get__()`` and ``__set__()``, while non-data\ndescriptors have just the ``__get__()`` method.  Data descriptors with\n``__set__()`` and ``__get__()`` defined always override a redefinition\nin an instance dictionary.  In contrast, non-data descriptors can be\noverridden by instances.\n\nPython methods (including ``staticmethod()`` and ``classmethod()``)\nare implemented as non-data descriptors.  Accordingly, instances can\nredefine and override methods.  This allows individual instances to\nacquire behaviors that differ from other instances of the same class.\n\nThe ``property()`` function is implemented as a data descriptor.\nAccordingly, instances cannot override the behavior of a property.\n\n\n__slots__\n---------\n\nBy default, instances of classes have a dictionary for attribute\nstorage.  This wastes space for objects having very few instance\nvariables.  The space consumption can become acute when creating large\nnumbers of instances.\n\nThe default can be overridden by defining *__slots__* in a class\ndefinition. The *__slots__* declaration takes a sequence of instance\nvariables and reserves just enough space in each instance to hold a\nvalue for each variable.  Space is saved because *__dict__* is not\ncreated for each instance.\n\nobject.__slots__\n\n   This class variable can be assigned a string, iterable, or sequence\n   of strings with variable names used by instances.  If defined in a\n   class, *__slots__* reserves space for the declared variables and\n   prevents the automatic creation of *__dict__* and *__weakref__* for\n   each instance.\n\n\nNotes on using *__slots__*\n~~~~~~~~~~~~~~~~~~~~~~~~~~\n\n* When inheriting from a class without *__slots__*, the *__dict__*\n  attribute of that class will always be accessible, so a *__slots__*\n  definition in the subclass is meaningless.\n\n* Without a *__dict__* variable, instances cannot be assigned new\n  variables not listed in the *__slots__* definition.  Attempts to\n  assign to an unlisted variable name raises ``AttributeError``. If\n  dynamic assignment of new variables is desired, then add\n  ``\'__dict__\'`` to the sequence of strings in the *__slots__*\n  declaration.\n\n* Without a *__weakref__* variable for each instance, classes defining\n  *__slots__* do not support weak references to its instances. If weak\n  reference support is needed, then add ``\'__weakref__\'`` to the\n  sequence of strings in the *__slots__* declaration.\n\n* *__slots__* are implemented at the class level by creating\n  descriptors (*Implementing Descriptors*) for each variable name.  As\n  a result, class attributes cannot be used to set default values for\n  instance variables defined by *__slots__*; otherwise, the class\n  attribute would overwrite the descriptor assignment.\n\n* The action of a *__slots__* declaration is limited to the class\n  where it is defined.  As a result, subclasses will have a *__dict__*\n  unless they also define *__slots__* (which must only contain names\n  of any *additional* slots).\n\n* If a class defines a slot also defined in a base class, the instance\n  variable defined by the base class slot is inaccessible (except by\n  retrieving its descriptor directly from the base class). This\n  renders the meaning of the program undefined.  In the future, a\n  check may be added to prevent this.\n\n* Nonempty *__slots__* does not work for classes derived from\n  "variable-length" built-in types such as ``int``, ``str`` and\n  ``tuple``.\n\n* Any non-string iterable may be assigned to *__slots__*. Mappings may\n  also be used; however, in the future, special meaning may be\n  assigned to the values corresponding to each key.\n\n* *__class__* assignment works only if both classes have the same\n  *__slots__*.\n\n\nCustomizing class creation\n==========================\n\nBy default, classes are constructed using ``type()``. A class\ndefinition is read into a separate namespace and the value of class\nname is bound to the result of ``type(name, bases, dict)``.\n\nWhen the class definition is read, if a callable ``metaclass`` keyword\nargument is passed after the bases in the class definition, the\ncallable given will be called instead of ``type()``.  If other keyword\narguments are passed, they will also be passed to the metaclass.  This\nallows classes or functions to be written which monitor or alter the\nclass creation process:\n\n* Modifying the class dictionary prior to the class being created.\n\n* Returning an instance of another class -- essentially performing the\n  role of a factory function.\n\nThese steps will have to be performed in the metaclass\'s ``__new__()``\nmethod -- ``type.__new__()`` can then be called from this method to\ncreate a class with different properties.  This example adds a new\nelement to the class dictionary before creating the class:\n\n   class metacls(type):\n       def __new__(mcs, name, bases, dict):\n           dict[\'foo\'] = \'metacls was here\'\n           return type.__new__(mcs, name, bases, dict)\n\nYou can of course also override other class methods (or add new\nmethods); for example defining a custom ``__call__()`` method in the\nmetaclass allows custom behavior when the class is called, e.g. not\nalways creating a new instance.\n\nIf the metaclass has a ``__prepare__()`` attribute (usually\nimplemented as a class or static method), it is called before the\nclass body is evaluated with the name of the class and a tuple of its\nbases for arguments.  It should return an object that supports the\nmapping interface that will be used to store the namespace of the\nclass.  The default is a plain dictionary.  This could be used, for\nexample, to keep track of the order that class attributes are declared\nin by returning an ordered dictionary.\n\nThe appropriate metaclass is determined by the following precedence\nrules:\n\n* If the ``metaclass`` keyword argument is passed with the bases, it\n  is used.\n\n* Otherwise, if there is at least one base class, its metaclass is\n  used.\n\n* Otherwise, the default metaclass (``type``) is used.\n\nThe potential uses for metaclasses are boundless. Some ideas that have\nbeen explored including logging, interface checking, automatic\ndelegation, automatic property creation, proxies, frameworks, and\nautomatic resource locking/synchronization.\n\nHere is an example of a metaclass that uses an\n``collections.OrderedDict`` to remember the order that class members\nwere defined:\n\n   class OrderedClass(type):\n\n        @classmethod\n        def __prepare__(metacls, name, bases, **kwds):\n           return collections.OrderedDict()\n\n        def __new__(cls, name, bases, classdict):\n           result = type.__new__(cls, name, bases, dict(classdict))\n           result.members = tuple(classdict)\n           return result\n\n   class A(metaclass=OrderedClass):\n       def one(self): pass\n       def two(self): pass\n       def three(self): pass\n       def four(self): pass\n\n   >>> A.members\n   (\'__module__\', \'one\', \'two\', \'three\', \'four\')\n\nWhen the class definition for *A* gets executed, the process begins\nwith calling the metaclass\'s ``__prepare__()`` method which returns an\nempty ``collections.OrderedDict``.  That mapping records the methods\nand attributes of *A* as they are defined within the body of the class\nstatement. Once those definitions are executed, the ordered dictionary\nis fully populated and the metaclass\'s ``__new__()`` method gets\ninvoked.  That method builds the new type and it saves the ordered\ndictionary keys in an attribute called *members*.\n\n\nCustomizing instance and subclass checks\n========================================\n\nThe following methods are used to override the default behavior of the\n``isinstance()`` and ``issubclass()`` built-in functions.\n\nIn particular, the metaclass ``abc.ABCMeta`` implements these methods\nin order to allow the addition of Abstract Base Classes (ABCs) as\n"virtual base classes" to any class or type (including built-in\ntypes), including other ABCs.\n\nclass.__instancecheck__(self, instance)\n\n   Return true if *instance* should be considered a (direct or\n   indirect) instance of *class*. If defined, called to implement\n   ``isinstance(instance, class)``.\n\nclass.__subclasscheck__(self, subclass)\n\n   Return true if *subclass* should be considered a (direct or\n   indirect) subclass of *class*.  If defined, called to implement\n   ``issubclass(subclass, class)``.\n\nNote that these methods are looked up on the type (metaclass) of a\nclass.  They cannot be defined as class methods in the actual class.\nThis is consistent with the lookup of special methods that are called\non instances, only in this case the instance is itself a class.\n\nSee also:\n\n   **PEP 3119** - Introducing Abstract Base Classes\n      Includes the specification for customizing ``isinstance()`` and\n      ``issubclass()`` behavior through ``__instancecheck__()`` and\n      ``__subclasscheck__()``, with motivation for this functionality\n      in the context of adding Abstract Base Classes (see the ``abc``\n      module) to the language.\n\n\nEmulating callable objects\n==========================\n\nobject.__call__(self[, args...])\n\n   Called when the instance is "called" as a function; if this method\n   is defined, ``x(arg1, arg2, ...)`` is a shorthand for\n   ``x.__call__(arg1, arg2, ...)``.\n\n\nEmulating container types\n=========================\n\nThe following methods can be defined to implement container objects.\nContainers usually are sequences (such as lists or tuples) or mappings\n(like dictionaries), but can represent other containers as well.  The\nfirst set of methods is used either to emulate a sequence or to\nemulate a mapping; the difference is that for a sequence, the\nallowable keys should be the integers *k* for which ``0 <= k < N``\nwhere *N* is the length of the sequence, or slice objects, which\ndefine a range of items.  It is also recommended that mappings provide\nthe methods ``keys()``, ``values()``, ``items()``, ``get()``,\n``clear()``, ``setdefault()``, ``pop()``, ``popitem()``, ``copy()``,\nand ``update()`` behaving similar to those for Python\'s standard\ndictionary objects.  The ``collections`` module provides a\n``MutableMapping`` abstract base class to help create those methods\nfrom a base set of ``__getitem__()``, ``__setitem__()``,\n``__delitem__()``, and ``keys()``. Mutable sequences should provide\nmethods ``append()``, ``count()``, ``index()``, ``extend()``,\n``insert()``, ``pop()``, ``remove()``, ``reverse()`` and ``sort()``,\nlike Python standard list objects.  Finally, sequence types should\nimplement addition (meaning concatenation) and multiplication (meaning\nrepetition) by defining the methods ``__add__()``, ``__radd__()``,\n``__iadd__()``, ``__mul__()``, ``__rmul__()`` and ``__imul__()``\ndescribed below; they should not define other numerical operators.  It\nis recommended that both mappings and sequences implement the\n``__contains__()`` method to allow efficient use of the ``in``\noperator; for mappings, ``in`` should search the mapping\'s keys; for\nsequences, it should search through the values.  It is further\nrecommended that both mappings and sequences implement the\n``__iter__()`` method to allow efficient iteration through the\ncontainer; for mappings, ``__iter__()`` should be the same as\n``keys()``; for sequences, it should iterate through the values.\n\nobject.__len__(self)\n\n   Called to implement the built-in function ``len()``.  Should return\n   the length of the object, an integer ``>=`` 0.  Also, an object\n   that doesn\'t define a ``__bool__()`` method and whose ``__len__()``\n   method returns zero is considered to be false in a Boolean context.\n\nNote: Slicing is done exclusively with the following three methods.  A\n  call like\n\n     a[1:2] = b\n\n  is translated to\n\n     a[slice(1, 2, None)] = b\n\n  and so forth.  Missing slice items are always filled in with\n  ``None``.\n\nobject.__getitem__(self, key)\n\n   Called to implement evaluation of ``self[key]``. For sequence\n   types, the accepted keys should be integers and slice objects.\n   Note that the special interpretation of negative indexes (if the\n   class wishes to emulate a sequence type) is up to the\n   ``__getitem__()`` method. If *key* is of an inappropriate type,\n   ``TypeError`` may be raised; if of a value outside the set of\n   indexes for the sequence (after any special interpretation of\n   negative values), ``IndexError`` should be raised. For mapping\n   types, if *key* is missing (not in the container), ``KeyError``\n   should be raised.\n\n   Note: ``for`` loops expect that an ``IndexError`` will be raised for\n     illegal indexes to allow proper detection of the end of the\n     sequence.\n\nobject.__setitem__(self, key, value)\n\n   Called to implement assignment to ``self[key]``.  Same note as for\n   ``__getitem__()``.  This should only be implemented for mappings if\n   the objects support changes to the values for keys, or if new keys\n   can be added, or for sequences if elements can be replaced.  The\n   same exceptions should be raised for improper *key* values as for\n   the ``__getitem__()`` method.\n\nobject.__delitem__(self, key)\n\n   Called to implement deletion of ``self[key]``.  Same note as for\n   ``__getitem__()``.  This should only be implemented for mappings if\n   the objects support removal of keys, or for sequences if elements\n   can be removed from the sequence.  The same exceptions should be\n   raised for improper *key* values as for the ``__getitem__()``\n   method.\n\nobject.__iter__(self)\n\n   This method is called when an iterator is required for a container.\n   This method should return a new iterator object that can iterate\n   over all the objects in the container.  For mappings, it should\n   iterate over the keys of the container, and should also be made\n   available as the method ``keys()``.\n\n   Iterator objects also need to implement this method; they are\n   required to return themselves.  For more information on iterator\n   objects, see *Iterator Types*.\n\nobject.__reversed__(self)\n\n   Called (if present) by the ``reversed()`` built-in to implement\n   reverse iteration.  It should return a new iterator object that\n   iterates over all the objects in the container in reverse order.\n\n   If the ``__reversed__()`` method is not provided, the\n   ``reversed()`` built-in will fall back to using the sequence\n   protocol (``__len__()`` and ``__getitem__()``).  Objects that\n   support the sequence protocol should only provide\n   ``__reversed__()`` if they can provide an implementation that is\n   more efficient than the one provided by ``reversed()``.\n\nThe membership test operators (``in`` and ``not in``) are normally\nimplemented as an iteration through a sequence.  However, container\nobjects can supply the following special method with a more efficient\nimplementation, which also does not require the object be a sequence.\n\nobject.__contains__(self, item)\n\n   Called to implement membership test operators.  Should return true\n   if *item* is in *self*, false otherwise.  For mapping objects, this\n   should consider the keys of the mapping rather than the values or\n   the key-item pairs.\n\n   For objects that don\'t define ``__contains__()``, the membership\n   test first tries iteration via ``__iter__()``, then the old\n   sequence iteration protocol via ``__getitem__()``, see *this\n   section in the language reference*.\n\n\nEmulating numeric types\n=======================\n\nThe following methods can be defined to emulate numeric objects.\nMethods corresponding to operations that are not supported by the\nparticular kind of number implemented (e.g., bitwise operations for\nnon-integral numbers) should be left undefined.\n\nobject.__add__(self, other)\nobject.__sub__(self, other)\nobject.__mul__(self, other)\nobject.__truediv__(self, other)\nobject.__floordiv__(self, other)\nobject.__mod__(self, other)\nobject.__divmod__(self, other)\nobject.__pow__(self, other[, modulo])\nobject.__lshift__(self, other)\nobject.__rshift__(self, other)\nobject.__and__(self, other)\nobject.__xor__(self, other)\nobject.__or__(self, other)\n\n   These methods are called to implement the binary arithmetic\n   operations (``+``, ``-``, ``*``, ``/``, ``//``, ``%``,\n   ``divmod()``, ``pow()``, ``**``, ``<<``, ``>>``, ``&``, ``^``,\n   ``|``).  For instance, to evaluate the expression ``x + y``, where\n   *x* is an instance of a class that has an ``__add__()`` method,\n   ``x.__add__(y)`` is called.  The ``__divmod__()`` method should be\n   the equivalent to using ``__floordiv__()`` and ``__mod__()``; it\n   should not be related to ``__truediv__()``.  Note that\n   ``__pow__()`` should be defined to accept an optional third\n   argument if the ternary version of the built-in ``pow()`` function\n   is to be supported.\n\n   If one of those methods does not support the operation with the\n   supplied arguments, it should return ``NotImplemented``.\n\nobject.__radd__(self, other)\nobject.__rsub__(self, other)\nobject.__rmul__(self, other)\nobject.__rtruediv__(self, other)\nobject.__rfloordiv__(self, other)\nobject.__rmod__(self, other)\nobject.__rdivmod__(self, other)\nobject.__rpow__(self, other)\nobject.__rlshift__(self, other)\nobject.__rrshift__(self, other)\nobject.__rand__(self, other)\nobject.__rxor__(self, other)\nobject.__ror__(self, other)\n\n   These methods are called to implement the binary arithmetic\n   operations (``+``, ``-``, ``*``, ``/``, ``//``, ``%``,\n   ``divmod()``, ``pow()``, ``**``, ``<<``, ``>>``, ``&``, ``^``,\n   ``|``) with reflected (swapped) operands. These functions are only\n   called if the left operand does not support the corresponding\n   operation and the operands are of different types. [2]  For\n   instance, to evaluate the expression ``x - y``, where *y* is an\n   instance of a class that has an ``__rsub__()`` method,\n   ``y.__rsub__(x)`` is called if ``x.__sub__(y)`` returns\n   *NotImplemented*.\n\n   Note that ternary ``pow()`` will not try calling ``__rpow__()``\n   (the coercion rules would become too complicated).\n\n   Note: If the right operand\'s type is a subclass of the left operand\'s\n     type and that subclass provides the reflected method for the\n     operation, this method will be called before the left operand\'s\n     non-reflected method.  This behavior allows subclasses to\n     override their ancestors\' operations.\n\nobject.__iadd__(self, other)\nobject.__isub__(self, other)\nobject.__imul__(self, other)\nobject.__itruediv__(self, other)\nobject.__ifloordiv__(self, other)\nobject.__imod__(self, other)\nobject.__ipow__(self, other[, modulo])\nobject.__ilshift__(self, other)\nobject.__irshift__(self, other)\nobject.__iand__(self, other)\nobject.__ixor__(self, other)\nobject.__ior__(self, other)\n\n   These methods are called to implement the augmented arithmetic\n   assignments (``+=``, ``-=``, ``*=``, ``/=``, ``//=``, ``%=``,\n   ``**=``, ``<<=``, ``>>=``, ``&=``, ``^=``, ``|=``).  These methods\n   should attempt to do the operation in-place (modifying *self*) and\n   return the result (which could be, but does not have to be,\n   *self*).  If a specific method is not defined, the augmented\n   assignment falls back to the normal methods.  For instance, to\n   execute the statement ``x += y``, where *x* is an instance of a\n   class that has an ``__iadd__()`` method, ``x.__iadd__(y)`` is\n   called.  If *x* is an instance of a class that does not define a\n   ``__iadd__()`` method, ``x.__add__(y)`` and ``y.__radd__(x)`` are\n   considered, as with the evaluation of ``x + y``.\n\nobject.__neg__(self)\nobject.__pos__(self)\nobject.__abs__(self)\nobject.__invert__(self)\n\n   Called to implement the unary arithmetic operations (``-``, ``+``,\n   ``abs()`` and ``~``).\n\nobject.__complex__(self)\nobject.__int__(self)\nobject.__float__(self)\nobject.__round__(self[, n])\n\n   Called to implement the built-in functions ``complex()``,\n   ``int()``, ``float()`` and ``round()``.  Should return a value of\n   the appropriate type.\n\nobject.__index__(self)\n\n   Called to implement ``operator.index()``.  Also called whenever\n   Python needs an integer object (such as in slicing, or in the\n   built-in ``bin()``, ``hex()`` and ``oct()`` functions). Must return\n   an integer.\n\n\nWith Statement Context Managers\n===============================\n\nA *context manager* is an object that defines the runtime context to\nbe established when executing a ``with`` statement. The context\nmanager handles the entry into, and the exit from, the desired runtime\ncontext for the execution of the block of code.  Context managers are\nnormally invoked using the ``with`` statement (described in section\n*The with statement*), but can also be used by directly invoking their\nmethods.\n\nTypical uses of context managers include saving and restoring various\nkinds of global state, locking and unlocking resources, closing opened\nfiles, etc.\n\nFor more information on context managers, see *Context Manager Types*.\n\nobject.__enter__(self)\n\n   Enter the runtime context related to this object. The ``with``\n   statement will bind this method\'s return value to the target(s)\n   specified in the ``as`` clause of the statement, if any.\n\nobject.__exit__(self, exc_type, exc_value, traceback)\n\n   Exit the runtime context related to this object. The parameters\n   describe the exception that caused the context to be exited. If the\n   context was exited without an exception, all three arguments will\n   be ``None``.\n\n   If an exception is supplied, and the method wishes to suppress the\n   exception (i.e., prevent it from being propagated), it should\n   return a true value. Otherwise, the exception will be processed\n   normally upon exit from this method.\n\n   Note that ``__exit__()`` methods should not reraise the passed-in\n   exception; this is the caller\'s responsibility.\n\nSee also:\n\n   **PEP 0343** - The "with" statement\n      The specification, background, and examples for the Python\n      ``with`` statement.\n\n\nSpecial method lookup\n=====================\n\nFor custom classes, implicit invocations of special methods are only\nguaranteed to work correctly if defined on an object\'s type, not in\nthe object\'s instance dictionary.  That behaviour is the reason why\nthe following code raises an exception:\n\n   >>> class C(object):\n   ...     pass\n   ...\n   >>> c = C()\n   >>> c.__len__ = lambda: 5\n   >>> len(c)\n   Traceback (most recent call last):\n     File "<stdin>", line 1, in <module>\n   TypeError: object of type \'C\' has no len()\n\nThe rationale behind this behaviour lies with a number of special\nmethods such as ``__hash__()`` and ``__repr__()`` that are implemented\nby all objects, including type objects. If the implicit lookup of\nthese methods used the conventional lookup process, they would fail\nwhen invoked on the type object itself:\n\n   >>> 1 .__hash__() == hash(1)\n   True\n   >>> int.__hash__() == hash(int)\n   Traceback (most recent call last):\n     File "<stdin>", line 1, in <module>\n   TypeError: descriptor \'__hash__\' of \'int\' object needs an argument\n\nIncorrectly attempting to invoke an unbound method of a class in this\nway is sometimes referred to as \'metaclass confusion\', and is avoided\nby bypassing the instance when looking up special methods:\n\n   >>> type(1).__hash__(1) == hash(1)\n   True\n   >>> type(int).__hash__(int) == hash(int)\n   True\n\nIn addition to bypassing any instance attributes in the interest of\ncorrectness, implicit special method lookup generally also bypasses\nthe ``__getattribute__()`` method even of the object\'s metaclass:\n\n   >>> class Meta(type):\n   ...    def __getattribute__(*args):\n   ...       print("Metaclass getattribute invoked")\n   ...       return type.__getattribute__(*args)\n   ...\n   >>> class C(object, metaclass=Meta):\n   ...     def __len__(self):\n   ...         return 10\n   ...     def __getattribute__(*args):\n   ...         print("Class getattribute invoked")\n   ...         return object.__getattribute__(*args)\n   ...\n   >>> c = C()\n   >>> c.__len__()                 # Explicit lookup via instance\n   Class getattribute invoked\n   10\n   >>> type(c).__len__(c)          # Explicit lookup via type\n   Metaclass getattribute invoked\n   10\n   >>> len(c)                      # Implicit lookup\n   10\n\nBypassing the ``__getattribute__()`` machinery in this fashion\nprovides significant scope for speed optimisations within the\ninterpreter, at the cost of some flexibility in the handling of\nspecial methods (the special method *must* be set on the class object\nitself in order to be consistently invoked by the interpreter).\n\n-[ Footnotes ]-\n\n[1] It *is* possible in some cases to change an object\'s type, under\n    certain controlled conditions. It generally isn\'t a good idea\n    though, since it can lead to some very strange behaviour if it is\n    handled incorrectly.\n\n[2] For operands of the same type, it is assumed that if the non-\n    reflected method (such as ``__add__()``) fails the operation is\n    not supported, which is why the reflected method is not called.\n',
+ 'string-methods': '\nString Methods\n**************\n\nString objects support the methods listed below.\n\nIn addition, Python\'s strings support the sequence type methods\ndescribed in the *Sequence Types --- str, bytes, bytearray, list,\ntuple, range* section. To output formatted strings, see the *String\nFormatting* section. Also, see the ``re`` module for string functions\nbased on regular expressions.\n\nstr.capitalize()\n\n   Return a copy of the string with its first character capitalized\n   and the rest lowercased.\n\nstr.center(width[, fillchar])\n\n   Return centered in a string of length *width*. Padding is done\n   using the specified *fillchar* (default is a space).\n\nstr.count(sub[, start[, end]])\n\n   Return the number of non-overlapping occurrences of substring *sub*\n   in the range [*start*, *end*].  Optional arguments *start* and\n   *end* are interpreted as in slice notation.\n\nstr.encode(encoding=sys.getdefaultencoding(), errors="strict")\n\n   Return an encoded version of the string as a bytes object.  Default\n   encoding is the current default string encoding.  *errors* may be\n   given to set a different error handling scheme.  The default for\n   *errors* is ``\'strict\'``, meaning that encoding errors raise a\n   ``UnicodeError``.  Other possible values are ``\'ignore\'``,\n   ``\'replace\'``, ``\'xmlcharrefreplace\'``, ``\'backslashreplace\'`` and\n   any other name registered via ``codecs.register_error()``, see\n   section *Codec Base Classes*. For a list of possible encodings, see\n   section *Standard Encodings*.\n\n   Changed in version 3.1: Added support for keyword arguments added.\n\nstr.endswith(suffix[, start[, end]])\n\n   Return ``True`` if the string ends with the specified *suffix*,\n   otherwise return ``False``.  *suffix* can also be a tuple of\n   suffixes to look for.  With optional *start*, test beginning at\n   that position.  With optional *end*, stop comparing at that\n   position.\n\nstr.expandtabs([tabsize])\n\n   Return a copy of the string where all tab characters are replaced\n   by one or more spaces, depending on the current column and the\n   given tab size.  The column number is reset to zero after each\n   newline occurring in the string. If *tabsize* is not given, a tab\n   size of ``8`` characters is assumed.  This doesn\'t understand other\n   non-printing characters or escape sequences.\n\nstr.find(sub[, start[, end]])\n\n   Return the lowest index in the string where substring *sub* is\n   found, such that *sub* is contained in the slice ``s[start:end]``.\n   Optional arguments *start* and *end* are interpreted as in slice\n   notation.  Return ``-1`` if *sub* is not found.\n\nstr.format(*args, **kwargs)\n\n   Perform a string formatting operation.  The string on which this\n   method is called can contain literal text or replacement fields\n   delimited by braces ``{}``.  Each replacement field contains either\n   the numeric index of a positional argument, or the name of a\n   keyword argument.  Returns a copy of the string where each\n   replacement field is replaced with the string value of the\n   corresponding argument.\n\n   >>> "The sum of 1 + 2 is {0}".format(1+2)\n   \'The sum of 1 + 2 is 3\'\n\n   See *Format String Syntax* for a description of the various\n   formatting options that can be specified in format strings.\n\nstr.index(sub[, start[, end]])\n\n   Like ``find()``, but raise ``ValueError`` when the substring is not\n   found.\n\nstr.isalnum()\n\n   Return true if all characters in the string are alphanumeric and\n   there is at least one character, false otherwise.\n\nstr.isalpha()\n\n   Return true if all characters in the string are alphabetic and\n   there is at least one character, false otherwise.\n\nstr.isdecimal()\n\n   Return true if all characters in the string are decimal characters\n   and there is at least one character, false otherwise. Decimal\n   characters include digit characters, and all characters that that\n   can be used to form decimal-radix numbers, e.g. U+0660, ARABIC-\n   INDIC DIGIT ZERO.\n\nstr.isdigit()\n\n   Return true if all characters in the string are digits and there is\n   at least one character, false otherwise.\n\nstr.isidentifier()\n\n   Return true if the string is a valid identifier according to the\n   language definition, section *Identifiers and keywords*.\n\nstr.islower()\n\n   Return true if all cased characters in the string are lowercase and\n   there is at least one cased character, false otherwise.\n\nstr.isnumeric()\n\n   Return true if all characters in the string are numeric characters,\n   and there is at least one character, false otherwise. Numeric\n   characters include digit characters, and all characters that have\n   the Unicode numeric value property, e.g. U+2155, VULGAR FRACTION\n   ONE FIFTH.\n\nstr.isprintable()\n\n   Return true if all characters in the string are printable or the\n   string is empty, false otherwise.  Nonprintable characters are\n   those characters defined in the Unicode character database as\n   "Other" or "Separator", excepting the ASCII space (0x20) which is\n   considered printable.  (Note that printable characters in this\n   context are those which should not be escaped when ``repr()`` is\n   invoked on a string.  It has no bearing on the handling of strings\n   written to ``sys.stdout`` or ``sys.stderr``.)\n\nstr.isspace()\n\n   Return true if there are only whitespace characters in the string\n   and there is at least one character, false otherwise.\n\nstr.istitle()\n\n   Return true if the string is a titlecased string and there is at\n   least one character, for example uppercase characters may only\n   follow uncased characters and lowercase characters only cased ones.\n   Return false otherwise.\n\nstr.isupper()\n\n   Return true if all cased characters in the string are uppercase and\n   there is at least one cased character, false otherwise.\n\nstr.join(iterable)\n\n   Return a string which is the concatenation of the strings in the\n   *iterable* *iterable*.  A ``TypeError`` will be raised if there are\n   any non-string values in *seq*, including ``bytes`` objects.  The\n   separator between elements is the string providing this method.\n\nstr.ljust(width[, fillchar])\n\n   Return the string left justified in a string of length *width*.\n   Padding is done using the specified *fillchar* (default is a\n   space).  The original string is returned if *width* is less than\n   ``len(s)``.\n\nstr.lower()\n\n   Return a copy of the string converted to lowercase.\n\nstr.lstrip([chars])\n\n   Return a copy of the string with leading characters removed.  The\n   *chars* argument is a string specifying the set of characters to be\n   removed.  If omitted or ``None``, the *chars* argument defaults to\n   removing whitespace.  The *chars* argument is not a prefix; rather,\n   all combinations of its values are stripped:\n\n   >>> \'   spacious   \'.lstrip()\n   \'spacious   \'\n   >>> \'www.example.com\'.lstrip(\'cmowz.\')\n   \'example.com\'\n\nstatic str.maketrans(x[, y[, z]])\n\n   This static method returns a translation table usable for\n   ``str.translate()``.\n\n   If there is only one argument, it must be a dictionary mapping\n   Unicode ordinals (integers) or characters (strings of length 1) to\n   Unicode ordinals, strings (of arbitrary lengths) or None.\n   Character keys will then be converted to ordinals.\n\n   If there are two arguments, they must be strings of equal length,\n   and in the resulting dictionary, each character in x will be mapped\n   to the character at the same position in y.  If there is a third\n   argument, it must be a string, whose characters will be mapped to\n   None in the result.\n\nstr.partition(sep)\n\n   Split the string at the first occurrence of *sep*, and return a\n   3-tuple containing the part before the separator, the separator\n   itself, and the part after the separator.  If the separator is not\n   found, return a 3-tuple containing the string itself, followed by\n   two empty strings.\n\nstr.replace(old, new[, count])\n\n   Return a copy of the string with all occurrences of substring *old*\n   replaced by *new*.  If the optional argument *count* is given, only\n   the first *count* occurrences are replaced.\n\nstr.rfind(sub[, start[, end]])\n\n   Return the highest index in the string where substring *sub* is\n   found, such that *sub* is contained within ``s[start:end]``.\n   Optional arguments *start* and *end* are interpreted as in slice\n   notation.  Return ``-1`` on failure.\n\nstr.rindex(sub[, start[, end]])\n\n   Like ``rfind()`` but raises ``ValueError`` when the substring *sub*\n   is not found.\n\nstr.rjust(width[, fillchar])\n\n   Return the string right justified in a string of length *width*.\n   Padding is done using the specified *fillchar* (default is a\n   space). The original string is returned if *width* is less than\n   ``len(s)``.\n\nstr.rpartition(sep)\n\n   Split the string at the last occurrence of *sep*, and return a\n   3-tuple containing the part before the separator, the separator\n   itself, and the part after the separator.  If the separator is not\n   found, return a 3-tuple containing two empty strings, followed by\n   the string itself.\n\nstr.rsplit([sep[, maxsplit]])\n\n   Return a list of the words in the string, using *sep* as the\n   delimiter string. If *maxsplit* is given, at most *maxsplit* splits\n   are done, the *rightmost* ones.  If *sep* is not specified or\n   ``None``, any whitespace string is a separator.  Except for\n   splitting from the right, ``rsplit()`` behaves like ``split()``\n   which is described in detail below.\n\nstr.rstrip([chars])\n\n   Return a copy of the string with trailing characters removed.  The\n   *chars* argument is a string specifying the set of characters to be\n   removed.  If omitted or ``None``, the *chars* argument defaults to\n   removing whitespace.  The *chars* argument is not a suffix; rather,\n   all combinations of its values are stripped:\n\n   >>> \'   spacious   \'.rstrip()\n   \'   spacious\'\n   >>> \'mississippi\'.rstrip(\'ipz\')\n   \'mississ\'\n\nstr.split([sep[, maxsplit]])\n\n   Return a list of the words in the string, using *sep* as the\n   delimiter string.  If *maxsplit* is given, at most *maxsplit*\n   splits are done (thus, the list will have at most ``maxsplit+1``\n   elements).  If *maxsplit* is not specified, then there is no limit\n   on the number of splits (all possible splits are made).\n\n   If *sep* is given, consecutive delimiters are not grouped together\n   and are deemed to delimit empty strings (for example,\n   ``\'1,,2\'.split(\',\')`` returns ``[\'1\', \'\', \'2\']``).  The *sep*\n   argument may consist of multiple characters (for example,\n   ``\'1<>2<>3\'.split(\'<>\')`` returns ``[\'1\', \'2\', \'3\']``). Splitting\n   an empty string with a specified separator returns ``[\'\']``.\n\n   If *sep* is not specified or is ``None``, a different splitting\n   algorithm is applied: runs of consecutive whitespace are regarded\n   as a single separator, and the result will contain no empty strings\n   at the start or end if the string has leading or trailing\n   whitespace.  Consequently, splitting an empty string or a string\n   consisting of just whitespace with a ``None`` separator returns\n   ``[]``.\n\n   For example, ``\' 1  2   3  \'.split()`` returns ``[\'1\', \'2\', \'3\']``,\n   and ``\'  1  2   3  \'.split(None, 1)`` returns ``[\'1\', \'2   3  \']``.\n\nstr.splitlines([keepends])\n\n   Return a list of the lines in the string, breaking at line\n   boundaries.  Line breaks are not included in the resulting list\n   unless *keepends* is given and true.\n\nstr.startswith(prefix[, start[, end]])\n\n   Return ``True`` if string starts with the *prefix*, otherwise\n   return ``False``. *prefix* can also be a tuple of prefixes to look\n   for.  With optional *start*, test string beginning at that\n   position.  With optional *end*, stop comparing string at that\n   position.\n\nstr.strip([chars])\n\n   Return a copy of the string with the leading and trailing\n   characters removed. The *chars* argument is a string specifying the\n   set of characters to be removed. If omitted or ``None``, the\n   *chars* argument defaults to removing whitespace. The *chars*\n   argument is not a prefix or suffix; rather, all combinations of its\n   values are stripped:\n\n   >>> \'   spacious   \'.strip()\n   \'spacious\'\n   >>> \'www.example.com\'.strip(\'cmowz.\')\n   \'example\'\n\nstr.swapcase()\n\n   Return a copy of the string with uppercase characters converted to\n   lowercase and vice versa.\n\nstr.title()\n\n   Return a titlecased version of the string where words start with an\n   uppercase character and the remaining characters are lowercase.\n\n   The algorithm uses a simple language-independent definition of a\n   word as groups of consecutive letters.  The definition works in\n   many contexts but it means that apostrophes in contractions and\n   possessives form word boundaries, which may not be the desired\n   result:\n\n      >>> "they\'re bill\'s friends from the UK".title()\n      "They\'Re Bill\'S Friends From The Uk"\n\n   A workaround for apostrophes can be constructed using regular\n   expressions:\n\n      >>> import re\n      >>> def titlecase(s):\n              return re.sub(r"[A-Za-z]+(\'[A-Za-z]+)?",\n                            lambda mo: mo.group(0)[0].upper() +\n                                       mo.group(0)[1:].lower(),\n                            s)\n\n      >>> titlecase("they\'re bill\'s friends.")\n      "They\'re Bill\'s Friends."\n\nstr.translate(map)\n\n   Return a copy of the *s* where all characters have been mapped\n   through the *map* which must be a dictionary of Unicode ordinals\n   (integers) to Unicode ordinals, strings or ``None``.  Unmapped\n   characters are left untouched. Characters mapped to ``None`` are\n   deleted.\n\n   You can use ``str.maketrans()`` to create a translation map from\n   character-to-character mappings in different formats.\n\n   You can use the ``maketrans()`` helper function in the ``string``\n   module to create a translation table. For string objects, set the\n   *table* argument to ``None`` for translations that only delete\n   characters:\n\n   Note: An even more flexible approach is to create a custom character\n     mapping codec using the ``codecs`` module (see\n     ``encodings.cp1251`` for an example).\n\nstr.upper()\n\n   Return a copy of the string converted to uppercase.\n\nstr.zfill(width)\n\n   Return the numeric string left filled with zeros in a string of\n   length *width*.  A sign prefix is handled correctly.  The original\n   string is returned if *width* is less than ``len(s)``.\n',
+ 'strings': '\nString and Bytes literals\n*************************\n\nString literals are described by the following lexical definitions:\n\n   stringliteral   ::= [stringprefix](shortstring | longstring)\n   stringprefix    ::= "r" | "R"\n   shortstring     ::= "\'" shortstringitem* "\'" | \'"\' shortstringitem* \'"\'\n   longstring      ::= "\'\'\'" longstringitem* "\'\'\'" | \'"""\' longstringitem* \'"""\'\n   shortstringitem ::= shortstringchar | stringescapeseq\n   longstringitem  ::= longstringchar | stringescapeseq\n   shortstringchar ::= <any source character except "\\" or newline or the quote>\n   longstringchar  ::= <any source character except "\\">\n   stringescapeseq ::= "\\" <any source character>\n\n   bytesliteral   ::= bytesprefix(shortbytes | longbytes)\n   bytesprefix    ::= "b" | "B" | "br" | "Br" | "bR" | "BR"\n   shortbytes     ::= "\'" shortbytesitem* "\'" | \'"\' shortbytesitem* \'"\'\n   longbytes      ::= "\'\'\'" longbytesitem* "\'\'\'" | \'"""\' longbytesitem* \'"""\'\n   shortbytesitem ::= shortbyteschar | bytesescapeseq\n   longbytesitem  ::= longbyteschar | bytesescapeseq\n   shortbyteschar ::= <any ASCII character except "\\" or newline or the quote>\n   longbyteschar  ::= <any ASCII character except "\\">\n   bytesescapeseq ::= "\\" <any ASCII character>\n\nOne syntactic restriction not indicated by these productions is that\nwhitespace is not allowed between the ``stringprefix`` or\n``bytesprefix`` and the rest of the literal. The source character set\nis defined by the encoding declaration; it is UTF-8 if no encoding\ndeclaration is given in the source file; see section *Encoding\ndeclarations*.\n\nIn plain English: Both types of literals can be enclosed in matching\nsingle quotes (``\'``) or double quotes (``"``).  They can also be\nenclosed in matching groups of three single or double quotes (these\nare generally referred to as *triple-quoted strings*).  The backslash\n(``\\``) character is used to escape characters that otherwise have a\nspecial meaning, such as newline, backslash itself, or the quote\ncharacter.\n\nBytes literals are always prefixed with ``\'b\'`` or ``\'B\'``; they\nproduce an instance of the ``bytes`` type instead of the ``str`` type.\nThey may only contain ASCII characters; bytes with a numeric value of\n128 or greater must be expressed with escapes.\n\nBoth string and bytes literals may optionally be prefixed with a\nletter ``\'r\'`` or ``\'R\'``; such strings are called *raw strings* and\ntreat backslashes as literal characters.  As a result, in string\nliterals, ``\'\\U\'`` and ``\'\\u\'`` escapes in raw strings are not treated\nspecially.\n\nIn triple-quoted strings, unescaped newlines and quotes are allowed\n(and are retained), except that three unescaped quotes in a row\nterminate the string.  (A "quote" is the character used to open the\nstring, i.e. either ``\'`` or ``"``.)\n\nUnless an ``\'r\'`` or ``\'R\'`` prefix is present, escape sequences in\nstrings are interpreted according to rules similar to those used by\nStandard C.  The recognized escape sequences are:\n\n+-------------------+-----------------------------------+---------+\n| Escape Sequence   | Meaning                           | Notes   |\n+===================+===================================+=========+\n| ``\\newline``      | Backslash and newline ignored     |         |\n+-------------------+-----------------------------------+---------+\n| ``\\\\``            | Backslash (``\\``)                 |         |\n+-------------------+-----------------------------------+---------+\n| ``\\\'``            | Single quote (``\'``)              |         |\n+-------------------+-----------------------------------+---------+\n| ``\\"``            | Double quote (``"``)              |         |\n+-------------------+-----------------------------------+---------+\n| ``\\a``            | ASCII Bell (BEL)                  |         |\n+-------------------+-----------------------------------+---------+\n| ``\\b``            | ASCII Backspace (BS)              |         |\n+-------------------+-----------------------------------+---------+\n| ``\\f``            | ASCII Formfeed (FF)               |         |\n+-------------------+-----------------------------------+---------+\n| ``\\n``            | ASCII Linefeed (LF)               |         |\n+-------------------+-----------------------------------+---------+\n| ``\\r``            | ASCII Carriage Return (CR)        |         |\n+-------------------+-----------------------------------+---------+\n| ``\\t``            | ASCII Horizontal Tab (TAB)        |         |\n+-------------------+-----------------------------------+---------+\n| ``\\v``            | ASCII Vertical Tab (VT)           |         |\n+-------------------+-----------------------------------+---------+\n| ``\\ooo``          | Character with octal value *ooo*  | (1,3)   |\n+-------------------+-----------------------------------+---------+\n| ``\\xhh``          | Character with hex value *hh*     | (2,3)   |\n+-------------------+-----------------------------------+---------+\n\nEscape sequences only recognized in string literals are:\n\n+-------------------+-----------------------------------+---------+\n| Escape Sequence   | Meaning                           | Notes   |\n+===================+===================================+=========+\n| ``\\N{name}``      | Character named *name* in the     |         |\n|                   | Unicode database                  |         |\n+-------------------+-----------------------------------+---------+\n| ``\\uxxxx``        | Character with 16-bit hex value   | (4)     |\n|                   | *xxxx*                            |         |\n+-------------------+-----------------------------------+---------+\n| ``\\Uxxxxxxxx``    | Character with 32-bit hex value   | (5)     |\n|                   | *xxxxxxxx*                        |         |\n+-------------------+-----------------------------------+---------+\n\nNotes:\n\n1. As in Standard C, up to three octal digits are accepted.\n\n2. Unlike in Standard C, exactly two hex digits are required.\n\n3. In a bytes literal, hexadecimal and octal escapes denote the byte\n   with the given value. In a string literal, these escapes denote a\n   Unicode character with the given value.\n\n4. Individual code units which form parts of a surrogate pair can be\n   encoded using this escape sequence.  Exactly four hex digits are\n   required.\n\n5. Any Unicode character can be encoded this way, but characters\n   outside the Basic Multilingual Plane (BMP) will be encoded using a\n   surrogate pair if Python is compiled to use 16-bit code units (the\n   default).  Exactly eight hex digits are required.\n\nUnlike Standard C, all unrecognized escape sequences are left in the\nstring unchanged, i.e., *the backslash is left in the string*.  (This\nbehavior is useful when debugging: if an escape sequence is mistyped,\nthe resulting output is more easily recognized as broken.)  It is also\nimportant to note that the escape sequences only recognized in string\nliterals fall into the category of unrecognized escapes for bytes\nliterals.\n\nEven in a raw string, string quotes can be escaped with a backslash,\nbut the backslash remains in the string; for example, ``r"\\""`` is a\nvalid string literal consisting of two characters: a backslash and a\ndouble quote; ``r"\\"`` is not a valid string literal (even a raw\nstring cannot end in an odd number of backslashes).  Specifically, *a\nraw string cannot end in a single backslash* (since the backslash\nwould escape the following quote character).  Note also that a single\nbackslash followed by a newline is interpreted as those two characters\nas part of the string, *not* as a line continuation.\n',
  'subscriptions': '\nSubscriptions\n*************\n\nA subscription selects an item of a sequence (string, tuple or list)\nor mapping (dictionary) object:\n\n   subscription ::= primary "[" expression_list "]"\n\nThe primary must evaluate to an object that supports subscription,\ne.g. a list or dictionary.  User-defined objects can support\nsubscription by defining a ``__getitem__()`` method.\n\nFor built-in objects, there are two types of objects that support\nsubscription:\n\nIf the primary is a mapping, the expression list must evaluate to an\nobject whose value is one of the keys of the mapping, and the\nsubscription selects the value in the mapping that corresponds to that\nkey.  (The expression list is a tuple except if it has exactly one\nitem.)\n\nIf the primary is a sequence, the expression (list) must evaluate to\nan integer. If this value is negative, the length of the sequence is\nadded to it (so that, e.g., ``x[-1]`` selects the last item of ``x``.)\nThe resulting value must be a nonnegative integer less than the number\nof items in the sequence, and the subscription selects the item whose\nindex is that value (counting from zero).\n\nA string\'s items are characters.  A character is not a separate data\ntype but a string of exactly one character.\n',
  'truth': "\nTruth Value Testing\n*******************\n\nAny object can be tested for truth value, for use in an ``if`` or\n``while`` condition or as operand of the Boolean operations below. The\nfollowing values are considered false:\n\n* ``None``\n\n* ``False``\n\n* zero of any numeric type, for example, ``0``, ``0.0``, ``0j``.\n\n* any empty sequence, for example, ``''``, ``()``, ``[]``.\n\n* any empty mapping, for example, ``{}``.\n\n* instances of user-defined classes, if the class defines a\n  ``__bool__()`` or ``__len__()`` method, when that method returns the\n  integer zero or ``bool`` value ``False``. [1]\n\nAll other values are considered true --- so objects of many types are\nalways true.\n\nOperations and built-in functions that have a Boolean result always\nreturn ``0`` or ``False`` for false and ``1`` or ``True`` for true,\nunless otherwise stated. (Important exception: the Boolean operations\n``or`` and ``and`` always return one of their operands.)\n",
- 'try': '\nThe ``try`` statement\n*********************\n\nThe ``try`` statement specifies exception handlers and/or cleanup code\nfor a group of statements:\n\n   try_stmt  ::= try1_stmt | try2_stmt\n   try1_stmt ::= "try" ":" suite\n                 ("except" [expression ["as" target]] ":" suite)+\n                 ["else" ":" suite]\n                 ["finally" ":" suite]\n   try2_stmt ::= "try" ":" suite\n                 "finally" ":" suite\n\nThe ``except`` clause(s) specify one or more exception handlers. When\nno exception occurs in the ``try`` clause, no exception handler is\nexecuted. When an exception occurs in the ``try`` suite, a search for\nan exception handler is started.  This search inspects the except\nclauses in turn until one is found that matches the exception.  An\nexpression-less except clause, if present, must be last; it matches\nany exception.  For an except clause with an expression, that\nexpression is evaluated, and the clause matches the exception if the\nresulting object is "compatible" with the exception.  An object is\ncompatible with an exception if it is the class or a base class of the\nexception object or a tuple containing an item compatible with the\nexception.\n\nIf no except clause matches the exception, the search for an exception\nhandler continues in the surrounding code and on the invocation stack.\n[1]\n\nIf the evaluation of an expression in the header of an except clause\nraises an exception, the original search for a handler is canceled and\na search starts for the new exception in the surrounding code and on\nthe call stack (it is treated as if the entire ``try`` statement\nraised the exception).\n\nWhen a matching except clause is found, the exception is assigned to\nthe target specified after the ``as`` keyword in that except clause,\nif present, and the except clause\'s suite is executed.  All except\nclauses must have an executable block.  When the end of this block is\nreached, execution continues normally after the entire try statement.\n(This means that if two nested handlers exist for the same exception,\nand the exception occurs in the try clause of the inner handler, the\nouter handler will not handle the exception.)\n\nWhen an exception has been assigned using ``as target``, it is cleared\nat the end of the except clause.  This is as if\n\n   except E as N:\n       foo\n\nwas translated to\n\n   except E as N:\n       try:\n           foo\n       finally:\n           N = None\n           del N\n\nThat means that you have to assign the exception to a different name\nif you want to be able to refer to it after the except clause.  The\nreason for this is that with the traceback attached to them,\nexceptions will form a reference cycle with the stack frame, keeping\nall locals in that frame alive until the next garbage collection\noccurs.\n\nBefore an except clause\'s suite is executed, details about the\nexception are stored in the ``sys`` module and can be access via\n``sys.exc_info()``. ``sys.exc_info()`` returns a 3-tuple consisting\nof: ``exc_type``, the exception class; ``exc_value``, the exception\ninstance; ``exc_traceback``, a traceback object (see section *The\nstandard type hierarchy*) identifying the point in the program where\nthe exception occurred. ``sys.exc_info()`` values are restored to\ntheir previous values (before the call) when returning from a function\nthat handled an exception.\n\nThe optional ``else`` clause is executed if and when control flows off\nthe end of the ``try`` clause. [2] Exceptions in the ``else`` clause\nare not handled by the preceding ``except`` clauses.\n\nIf ``finally`` is present, it specifies a \'cleanup\' handler.  The\n``try`` clause is executed, including any ``except`` and ``else``\nclauses.  If an exception occurs in any of the clauses and is not\nhandled, the exception is temporarily saved. The ``finally`` clause is\nexecuted.  If there is a saved exception, it is re-raised at the end\nof the ``finally`` clause. If the ``finally`` clause raises another\nexception or executes a ``return`` or ``break`` statement, the saved\nexception is lost.  The exception information is not available to the\nprogram during execution of the ``finally`` clause.\n\nWhen a ``return``, ``break`` or ``continue`` statement is executed in\nthe ``try`` suite of a ``try``...``finally`` statement, the\n``finally`` clause is also executed \'on the way out.\' A ``continue``\nstatement is illegal in the ``finally`` clause. (The reason is a\nproblem with the current implementation --- this restriction may be\nlifted in the future).\n\nAdditional information on exceptions can be found in section\n*Exceptions*, and information on using the ``raise`` statement to\ngenerate exceptions may be found in section *The raise statement*.\n',
- 'types': '\nThe standard type hierarchy\n***************************\n\nBelow is a list of the types that are built into Python.  Extension\nmodules (written in C, Java, or other languages, depending on the\nimplementation) can define additional types.  Future versions of\nPython may add types to the type hierarchy (e.g., rational numbers,\nefficiently stored arrays of integers, etc.), although such additions\nwill often be provided via the standard library instead.\n\nSome of the type descriptions below contain a paragraph listing\n\'special attributes.\'  These are attributes that provide access to the\nimplementation and are not intended for general use.  Their definition\nmay change in the future.\n\nNone\n   This type has a single value.  There is a single object with this\n   value. This object is accessed through the built-in name ``None``.\n   It is used to signify the absence of a value in many situations,\n   e.g., it is returned from functions that don\'t explicitly return\n   anything. Its truth value is false.\n\nNotImplemented\n   This type has a single value.  There is a single object with this\n   value. This object is accessed through the built-in name\n   ``NotImplemented``. Numeric methods and rich comparison methods may\n   return this value if they do not implement the operation for the\n   operands provided.  (The interpreter will then try the reflected\n   operation, or some other fallback, depending on the operator.)  Its\n   truth value is true.\n\nEllipsis\n   This type has a single value.  There is a single object with this\n   value. This object is accessed through the literal ``...`` or the\n   built-in name ``Ellipsis``.  Its truth value is true.\n\n``numbers.Number``\n   These are created by numeric literals and returned as results by\n   arithmetic operators and arithmetic built-in functions.  Numeric\n   objects are immutable; once created their value never changes.\n   Python numbers are of course strongly related to mathematical\n   numbers, but subject to the limitations of numerical representation\n   in computers.\n\n   Python distinguishes between integers, floating point numbers, and\n   complex numbers:\n\n   ``numbers.Integral``\n      These represent elements from the mathematical set of integers\n      (positive and negative).\n\n      There are two types of integers:\n\n      Integers (``int``)\n\n         These represent numbers in an unlimited range, subject to\n         available (virtual) memory only.  For the purpose of shift\n         and mask operations, a binary representation is assumed, and\n         negative numbers are represented in a variant of 2\'s\n         complement which gives the illusion of an infinite string of\n         sign bits extending to the left.\n\n      Booleans (``bool``)\n         These represent the truth values False and True.  The two\n         objects representing the values False and True are the only\n         Boolean objects. The Boolean type is a subtype of the integer\n         type, and Boolean values behave like the values 0 and 1,\n         respectively, in almost all contexts, the exception being\n         that when converted to a string, the strings ``"False"`` or\n         ``"True"`` are returned, respectively.\n\n      The rules for integer representation are intended to give the\n      most meaningful interpretation of shift and mask operations\n      involving negative integers.\n\n   ``numbers.Real`` (``float``)\n      These represent machine-level double precision floating point\n      numbers. You are at the mercy of the underlying machine\n      architecture (and C or Java implementation) for the accepted\n      range and handling of overflow. Python does not support single-\n      precision floating point numbers; the savings in processor and\n      memory usage that are usually the reason for using these is\n      dwarfed by the overhead of using objects in Python, so there is\n      no reason to complicate the language with two kinds of floating\n      point numbers.\n\n   ``numbers.Complex`` (``complex``)\n      These represent complex numbers as a pair of machine-level\n      double precision floating point numbers.  The same caveats apply\n      as for floating point numbers. The real and imaginary parts of a\n      complex number ``z`` can be retrieved through the read-only\n      attributes ``z.real`` and ``z.imag``.\n\nSequences\n   These represent finite ordered sets indexed by non-negative\n   numbers. The built-in function ``len()`` returns the number of\n   items of a sequence. When the length of a sequence is *n*, the\n   index set contains the numbers 0, 1, ..., *n*-1.  Item *i* of\n   sequence *a* is selected by ``a[i]``.\n\n   Sequences also support slicing: ``a[i:j]`` selects all items with\n   index *k* such that *i* ``<=`` *k* ``<`` *j*.  When used as an\n   expression, a slice is a sequence of the same type.  This implies\n   that the index set is renumbered so that it starts at 0.\n\n   Some sequences also support "extended slicing" with a third "step"\n   parameter: ``a[i:j:k]`` selects all items of *a* with index *x*\n   where ``x = i + n*k``, *n* ``>=`` ``0`` and *i* ``<=`` *x* ``<``\n   *j*.\n\n   Sequences are distinguished according to their mutability:\n\n   Immutable sequences\n      An object of an immutable sequence type cannot change once it is\n      created.  (If the object contains references to other objects,\n      these other objects may be mutable and may be changed; however,\n      the collection of objects directly referenced by an immutable\n      object cannot change.)\n\n      The following types are immutable sequences:\n\n      Strings\n         The items of a string object are Unicode code units.  A\n         Unicode code unit is represented by a string object of one\n         item and can hold either a 16-bit or 32-bit value\n         representing a Unicode ordinal (the maximum value for the\n         ordinal is given in ``sys.maxunicode``, and depends on how\n         Python is configured at compile time).  Surrogate pairs may\n         be present in the Unicode object, and will be reported as two\n         separate items.  The built-in functions ``chr()`` and\n         ``ord()`` convert between code units and nonnegative integers\n         representing the Unicode ordinals as defined in the Unicode\n         Standard 3.0. Conversion from and to other encodings are\n         possible through the string method ``encode()``.\n\n      Tuples\n         The items of a tuple are arbitrary Python objects. Tuples of\n         two or more items are formed by comma-separated lists of\n         expressions.  A tuple of one item (a \'singleton\') can be\n         formed by affixing a comma to an expression (an expression by\n         itself does not create a tuple, since parentheses must be\n         usable for grouping of expressions).  An empty tuple can be\n         formed by an empty pair of parentheses.\n\n      Bytes\n         A bytes object is an immutable array.  The items are 8-bit\n         bytes, represented by integers in the range 0 <= x < 256.\n         Bytes literals (like ``b\'abc\'`` and the built-in function\n         ``bytes()`` can be used to construct bytes objects.  Also,\n         bytes objects can be decoded to strings via the ``decode()``\n         method.\n\n   Mutable sequences\n      Mutable sequences can be changed after they are created.  The\n      subscription and slicing notations can be used as the target of\n      assignment and ``del`` (delete) statements.\n\n      There are currently two intrinsic mutable sequence types:\n\n      Lists\n         The items of a list are arbitrary Python objects.  Lists are\n         formed by placing a comma-separated list of expressions in\n         square brackets. (Note that there are no special cases needed\n         to form lists of length 0 or 1.)\n\n      Byte Arrays\n         A bytearray object is a mutable array. They are created by\n         the built-in ``bytearray()`` constructor.  Aside from being\n         mutable (and hence unhashable), byte arrays otherwise provide\n         the same interface and functionality as immutable bytes\n         objects.\n\n      The extension module ``array`` provides an additional example of\n      a mutable sequence type, as does the ``collections`` module.\n\nSet types\n   These represent unordered, finite sets of unique, immutable\n   objects. As such, they cannot be indexed by any subscript. However,\n   they can be iterated over, and the built-in function ``len()``\n   returns the number of items in a set. Common uses for sets are fast\n   membership testing, removing duplicates from a sequence, and\n   computing mathematical operations such as intersection, union,\n   difference, and symmetric difference.\n\n   For set elements, the same immutability rules apply as for\n   dictionary keys. Note that numeric types obey the normal rules for\n   numeric comparison: if two numbers compare equal (e.g., ``1`` and\n   ``1.0``), only one of them can be contained in a set.\n\n   There are currently two intrinsic set types:\n\n   Sets\n      These represent a mutable set. They are created by the built-in\n      ``set()`` constructor and can be modified afterwards by several\n      methods, such as ``add()``.\n\n   Frozen sets\n      These represent an immutable set.  They are created by the\n      built-in ``frozenset()`` constructor.  As a frozenset is\n      immutable and *hashable*, it can be used again as an element of\n      another set, or as a dictionary key.\n\nMappings\n   These represent finite sets of objects indexed by arbitrary index\n   sets. The subscript notation ``a[k]`` selects the item indexed by\n   ``k`` from the mapping ``a``; this can be used in expressions and\n   as the target of assignments or ``del`` statements. The built-in\n   function ``len()`` returns the number of items in a mapping.\n\n   There is currently a single intrinsic mapping type:\n\n   Dictionaries\n      These represent finite sets of objects indexed by nearly\n      arbitrary values.  The only types of values not acceptable as\n      keys are values containing lists or dictionaries or other\n      mutable types that are compared by value rather than by object\n      identity, the reason being that the efficient implementation of\n      dictionaries requires a key\'s hash value to remain constant.\n      Numeric types used for keys obey the normal rules for numeric\n      comparison: if two numbers compare equal (e.g., ``1`` and\n      ``1.0``) then they can be used interchangeably to index the same\n      dictionary entry.\n\n      Dictionaries are mutable; they can be created by the ``{...}``\n      notation (see section *Dictionary displays*).\n\n      The extension modules ``dbm.ndbm`` and ``dbm.gnu`` provide\n      additional examples of mapping types, as does the\n      ``collections`` module.\n\nCallable types\n   These are the types to which the function call operation (see\n   section *Calls*) can be applied:\n\n   User-defined functions\n      A user-defined function object is created by a function\n      definition (see section *Function definitions*).  It should be\n      called with an argument list containing the same number of items\n      as the function\'s formal parameter list.\n\n      Special attributes:\n\n      +---------------------------+---------------------------------+-------------+\n      | Attribute                 | Meaning                         |             |\n      +===========================+=================================+=============+\n      | ``__doc__``               | The function\'s documentation    | Writable    |\n      |                           | string, or ``None`` if          |             |\n      |                           | unavailable                     |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__name__``              | The function\'s name             | Writable    |\n      +---------------------------+---------------------------------+-------------+\n      | ``__module__``            | The name of the module the      | Writable    |\n      |                           | function was defined in, or     |             |\n      |                           | ``None`` if unavailable.        |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__defaults__``          | A tuple containing default      | Writable    |\n      |                           | argument values for those       |             |\n      |                           | arguments that have defaults,   |             |\n      |                           | or ``None`` if no arguments     |             |\n      |                           | have a default value            |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__code__``              | The code object representing    | Writable    |\n      |                           | the compiled function body.     |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__globals__``           | A reference to the dictionary   | Read-only   |\n      |                           | that holds the function\'s       |             |\n      |                           | global variables --- the global |             |\n      |                           | namespace of the module in      |             |\n      |                           | which the function was defined. |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__dict__``              | The namespace supporting        | Writable    |\n      |                           | arbitrary function attributes.  |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__closure__``           | ``None`` or a tuple of cells    | Read-only   |\n      |                           | that contain bindings for the   |             |\n      |                           | function\'s free variables.      |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__annotations__``       | A dict containing annotations   | Writable    |\n      |                           | of parameters.  The keys of the |             |\n      |                           | dict are the parameter names,   |             |\n      |                           | or ``\'return\'`` for the return  |             |\n      |                           | annotation, if provided.        |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__kwdefaults__``        | A dict containing defaults for  | Writable    |\n      |                           | keyword-only parameters.        |             |\n      +---------------------------+---------------------------------+-------------+\n\n      Most of the attributes labelled "Writable" check the type of the\n      assigned value.\n\n      Function objects also support getting and setting arbitrary\n      attributes, which can be used, for example, to attach metadata\n      to functions.  Regular attribute dot-notation is used to get and\n      set such attributes. *Note that the current implementation only\n      supports function attributes on user-defined functions. Function\n      attributes on built-in functions may be supported in the\n      future.*\n\n      Additional information about a function\'s definition can be\n      retrieved from its code object; see the description of internal\n      types below.\n\n   Instance methods\n      An instance method object combines a class, a class instance and\n      any callable object (normally a user-defined function).\n\n      Special read-only attributes: ``__self__`` is the class instance\n      object, ``__func__`` is the function object; ``__doc__`` is the\n      method\'s documentation (same as ``__func__.__doc__``);\n      ``__name__`` is the method name (same as ``__func__.__name__``);\n      ``__module__`` is the name of the module the method was defined\n      in, or ``None`` if unavailable.\n\n      Methods also support accessing (but not setting) the arbitrary\n      function attributes on the underlying function object.\n\n      User-defined method objects may be created when getting an\n      attribute of a class (perhaps via an instance of that class), if\n      that attribute is a user-defined function object or a class\n      method object.\n\n      When an instance method object is created by retrieving a user-\n      defined function object from a class via one of its instances,\n      its ``__self__`` attribute is the instance, and the method\n      object is said to be bound.  The new method\'s ``__func__``\n      attribute is the original function object.\n\n      When a user-defined method object is created by retrieving\n      another method object from a class or instance, the behaviour is\n      the same as for a function object, except that the ``__func__``\n      attribute of the new instance is not the original method object\n      but its ``__func__`` attribute.\n\n      When an instance method object is created by retrieving a class\n      method object from a class or instance, its ``__self__``\n      attribute is the class itself, and its ``__func__`` attribute is\n      the function object underlying the class method.\n\n      When an instance method object is called, the underlying\n      function (``__func__``) is called, inserting the class instance\n      (``__self__``) in front of the argument list.  For instance,\n      when ``C`` is a class which contains a definition for a function\n      ``f()``, and ``x`` is an instance of ``C``, calling ``x.f(1)``\n      is equivalent to calling ``C.f(x, 1)``.\n\n      When an instance method object is derived from a class method\n      object, the "class instance" stored in ``__self__`` will\n      actually be the class itself, so that calling either ``x.f(1)``\n      or ``C.f(1)`` is equivalent to calling ``f(C,1)`` where ``f`` is\n      the underlying function.\n\n      Note that the transformation from function object to instance\n      method object happens each time the attribute is retrieved from\n      the instance.  In some cases, a fruitful optimization is to\n      assign the attribute to a local variable and call that local\n      variable. Also notice that this transformation only happens for\n      user-defined functions; other callable objects (and all non-\n      callable objects) are retrieved without transformation.  It is\n      also important to note that user-defined functions which are\n      attributes of a class instance are not converted to bound\n      methods; this *only* happens when the function is an attribute\n      of the class.\n\n   Generator functions\n      A function or method which uses the ``yield`` statement (see\n      section *The yield statement*) is called a *generator function*.\n      Such a function, when called, always returns an iterator object\n      which can be used to execute the body of the function:  calling\n      the iterator\'s ``__next__()`` method will cause the function to\n      execute until it provides a value using the ``yield`` statement.\n      When the function executes a ``return`` statement or falls off\n      the end, a ``StopIteration`` exception is raised and the\n      iterator will have reached the end of the set of values to be\n      returned.\n\n   Built-in functions\n      A built-in function object is a wrapper around a C function.\n      Examples of built-in functions are ``len()`` and ``math.sin()``\n      (``math`` is a standard built-in module). The number and type of\n      the arguments are determined by the C function. Special read-\n      only attributes: ``__doc__`` is the function\'s documentation\n      string, or ``None`` if unavailable; ``__name__`` is the\n      function\'s name; ``__self__`` is set to ``None`` (but see the\n      next item); ``__module__`` is the name of the module the\n      function was defined in or ``None`` if unavailable.\n\n   Built-in methods\n      This is really a different disguise of a built-in function, this\n      time containing an object passed to the C function as an\n      implicit extra argument.  An example of a built-in method is\n      ``alist.append()``, assuming *alist* is a list object. In this\n      case, the special read-only attribute ``__self__`` is set to the\n      object denoted by *list*.\n\n   Classes\n      Classes are callable.  These objects normally act as factories\n      for new instances of themselves, but variations are possible for\n      class types that override ``__new__()``.  The arguments of the\n      call are passed to ``__new__()`` and, in the typical case, to\n      ``__init__()`` to initialize the new instance.\n\n   Class Instances\n      Instances of arbitrary classes can be made callable by defining\n      a ``__call__()`` method in their class.\n\nModules\n   Modules are imported by the ``import`` statement (see section *The\n   import statement*). A module object has a namespace implemented by\n   a dictionary object (this is the dictionary referenced by the\n   __globals__ attribute of functions defined in the module).\n   Attribute references are translated to lookups in this dictionary,\n   e.g., ``m.x`` is equivalent to ``m.__dict__["x"]``. A module object\n   does not contain the code object used to initialize the module\n   (since it isn\'t needed once the initialization is done).\n\n   Attribute assignment updates the module\'s namespace dictionary,\n   e.g., ``m.x = 1`` is equivalent to ``m.__dict__["x"] = 1``.\n\n   Special read-only attribute: ``__dict__`` is the module\'s namespace\n   as a dictionary object.\n\n   Predefined (writable) attributes: ``__name__`` is the module\'s\n   name; ``__doc__`` is the module\'s documentation string, or ``None``\n   if unavailable; ``__file__`` is the pathname of the file from which\n   the module was loaded, if it was loaded from a file. The\n   ``__file__`` attribute is not present for C modules that are\n   statically linked into the interpreter; for extension modules\n   loaded dynamically from a shared library, it is the pathname of the\n   shared library file.\n\nCustom classes\n   Custon class types are typically created by class definitions (see\n   section *Class definitions*).  A class has a namespace implemented\n   by a dictionary object. Class attribute references are translated\n   to lookups in this dictionary, e.g., ``C.x`` is translated to\n   ``C.__dict__["x"]`` (although there are a number of hooks which\n   allow for other means of locating attributes). When the attribute\n   name is not found there, the attribute search continues in the base\n   classes. This search of the base classes uses the C3 method\n   resolution order which behaves correctly even in the presence of\n   \'diamond\' inheritance structures where there are multiple\n   inheritance paths leading back to a common ancestor. Additional\n   details on the C3 MRO used by Python can be found in the\n   documentation accompanying the 2.3 release at\n   http://www.python.org/download/releases/2.3/mro/.\n\n   When a class attribute reference (for class ``C``, say) would yield\n   a class method object, it is transformed into an instance method\n   object whose ``__self__`` attributes is ``C``.  When it would yield\n   a static method object, it is transformed into the object wrapped\n   by the static method object. See section *Implementing Descriptors*\n   for another way in which attributes retrieved from a class may\n   differ from those actually contained in its ``__dict__``.\n\n   Class attribute assignments update the class\'s dictionary, never\n   the dictionary of a base class.\n\n   A class object can be called (see above) to yield a class instance\n   (see below).\n\n   Special attributes: ``__name__`` is the class name; ``__module__``\n   is the module name in which the class was defined; ``__dict__`` is\n   the dictionary containing the class\'s namespace; ``__bases__`` is a\n   tuple (possibly empty or a singleton) containing the base classes,\n   in the order of their occurrence in the base class list;\n   ``__doc__`` is the class\'s documentation string, or None if\n   undefined.\n\nClass instances\n   A class instance is created by calling a class object (see above).\n   A class instance has a namespace implemented as a dictionary which\n   is the first place in which attribute references are searched.\n   When an attribute is not found there, and the instance\'s class has\n   an attribute by that name, the search continues with the class\n   attributes.  If a class attribute is found that is a user-defined\n   function object, it is transformed into an instance method object\n   whose ``__self__`` attribute is the instance.  Static method and\n   class method objects are also transformed; see above under\n   "Classes".  See section *Implementing Descriptors* for another way\n   in which attributes of a class retrieved via its instances may\n   differ from the objects actually stored in the class\'s\n   ``__dict__``.  If no class attribute is found, and the object\'s\n   class has a ``__getattr__()`` method, that is called to satisfy the\n   lookup.\n\n   Attribute assignments and deletions update the instance\'s\n   dictionary, never a class\'s dictionary.  If the class has a\n   ``__setattr__()`` or ``__delattr__()`` method, this is called\n   instead of updating the instance dictionary directly.\n\n   Class instances can pretend to be numbers, sequences, or mappings\n   if they have methods with certain special names.  See section\n   *Special method names*.\n\n   Special attributes: ``__dict__`` is the attribute dictionary;\n   ``__class__`` is the instance\'s class.\n\nFiles\n   A file object represents an open file.  File objects are created by\n   the ``open()`` built-in function, and also by ``os.popen()``,\n   ``os.fdopen()``, and the ``makefile()`` method of socket objects\n   (and perhaps by other functions or methods provided by extension\n   modules).  The objects ``sys.stdin``, ``sys.stdout`` and\n   ``sys.stderr`` are initialized to file objects corresponding to the\n   interpreter\'s standard input, output and error streams.  See *File\n   Objects* for complete documentation of file objects.\n\nInternal types\n   A few types used internally by the interpreter are exposed to the\n   user. Their definitions may change with future versions of the\n   interpreter, but they are mentioned here for completeness.\n\n   Code objects\n      Code objects represent *byte-compiled* executable Python code,\n      or *bytecode*. The difference between a code object and a\n      function object is that the function object contains an explicit\n      reference to the function\'s globals (the module in which it was\n      defined), while a code object contains no context; also the\n      default argument values are stored in the function object, not\n      in the code object (because they represent values calculated at\n      run-time).  Unlike function objects, code objects are immutable\n      and contain no references (directly or indirectly) to mutable\n      objects.\n\n      Special read-only attributes: ``co_name`` gives the function\n      name; ``co_argcount`` is the number of positional arguments\n      (including arguments with default values); ``co_nlocals`` is the\n      number of local variables used by the function (including\n      arguments); ``co_varnames`` is a tuple containing the names of\n      the local variables (starting with the argument names);\n      ``co_cellvars`` is a tuple containing the names of local\n      variables that are referenced by nested functions;\n      ``co_freevars`` is a tuple containing the names of free\n      variables; ``co_code`` is a string representing the sequence of\n      bytecode instructions; ``co_consts`` is a tuple containing the\n      literals used by the bytecode; ``co_names`` is a tuple\n      containing the names used by the bytecode; ``co_filename`` is\n      the filename from which the code was compiled;\n      ``co_firstlineno`` is the first line number of the function;\n      ``co_lnotab`` is a string encoding the mapping from bytecode\n      offsets to line numbers (for details see the source code of the\n      interpreter); ``co_stacksize`` is the required stack size\n      (including local variables); ``co_flags`` is an integer encoding\n      a number of flags for the interpreter.\n\n      The following flag bits are defined for ``co_flags``: bit\n      ``0x04`` is set if the function uses the ``*arguments`` syntax\n      to accept an arbitrary number of positional arguments; bit\n      ``0x08`` is set if the function uses the ``**keywords`` syntax\n      to accept arbitrary keyword arguments; bit ``0x20`` is set if\n      the function is a generator.\n\n      Future feature declarations (``from __future__ import\n      division``) also use bits in ``co_flags`` to indicate whether a\n      code object was compiled with a particular feature enabled: bit\n      ``0x2000`` is set if the function was compiled with future\n      division enabled; bits ``0x10`` and ``0x1000`` were used in\n      earlier versions of Python.\n\n      Other bits in ``co_flags`` are reserved for internal use.\n\n      If a code object represents a function, the first item in\n      ``co_consts`` is the documentation string of the function, or\n      ``None`` if undefined.\n\n   Frame objects\n      Frame objects represent execution frames.  They may occur in\n      traceback objects (see below).\n\n      Special read-only attributes: ``f_back`` is to the previous\n      stack frame (towards the caller), or ``None`` if this is the\n      bottom stack frame; ``f_code`` is the code object being executed\n      in this frame; ``f_locals`` is the dictionary used to look up\n      local variables; ``f_globals`` is used for global variables;\n      ``f_builtins`` is used for built-in (intrinsic) names;\n      ``f_lasti`` gives the precise instruction (this is an index into\n      the bytecode string of the code object).\n\n      Special writable attributes: ``f_trace``, if not ``None``, is a\n      function called at the start of each source code line (this is\n      used by the debugger); ``f_lineno`` is the current line number\n      of the frame --- writing to this from within a trace function\n      jumps to the given line (only for the bottom-most frame).  A\n      debugger can implement a Jump command (aka Set Next Statement)\n      by writing to f_lineno.\n\n   Traceback objects\n      Traceback objects represent a stack trace of an exception.  A\n      traceback object is created when an exception occurs.  When the\n      search for an exception handler unwinds the execution stack, at\n      each unwound level a traceback object is inserted in front of\n      the current traceback.  When an exception handler is entered,\n      the stack trace is made available to the program. (See section\n      *The try statement*.) It is accessible as the third item of the\n      tuple returned by ``sys.exc_info()``. When the program contains\n      no suitable handler, the stack trace is written (nicely\n      formatted) to the standard error stream; if the interpreter is\n      interactive, it is also made available to the user as\n      ``sys.last_traceback``.\n\n      Special read-only attributes: ``tb_next`` is the next level in\n      the stack trace (towards the frame where the exception\n      occurred), or ``None`` if there is no next level; ``tb_frame``\n      points to the execution frame of the current level;\n      ``tb_lineno`` gives the line number where the exception\n      occurred; ``tb_lasti`` indicates the precise instruction.  The\n      line number and last instruction in the traceback may differ\n      from the line number of its frame object if the exception\n      occurred in a ``try`` statement with no matching except clause\n      or with a finally clause.\n\n   Slice objects\n      Slice objects are used to represent slices for ``__getitem__()``\n      methods.  They are also created by the built-in ``slice()``\n      function.\n\n      Special read-only attributes: ``start`` is the lower bound;\n      ``stop`` is the upper bound; ``step`` is the step value; each is\n      ``None`` if omitted. These attributes can have any type.\n\n      Slice objects support one method:\n\n      slice.indices(self, length)\n\n         This method takes a single integer argument *length* and\n         computes information about the slice that the slice object\n         would describe if applied to a sequence of *length* items.\n         It returns a tuple of three integers; respectively these are\n         the *start* and *stop* indices and the *step* or stride\n         length of the slice. Missing or out-of-bounds indices are\n         handled in a manner consistent with regular slices.\n\n   Static method objects\n      Static method objects provide a way of defeating the\n      transformation of function objects to method objects described\n      above. A static method object is a wrapper around any other\n      object, usually a user-defined method object. When a static\n      method object is retrieved from a class or a class instance, the\n      object actually returned is the wrapped object, which is not\n      subject to any further transformation. Static method objects are\n      not themselves callable, although the objects they wrap usually\n      are. Static method objects are created by the built-in\n      ``staticmethod()`` constructor.\n\n   Class method objects\n      A class method object, like a static method object, is a wrapper\n      around another object that alters the way in which that object\n      is retrieved from classes and class instances. The behaviour of\n      class method objects upon such retrieval is described above,\n      under "User-defined methods". Class method objects are created\n      by the built-in ``classmethod()`` constructor.\n',
+ 'try': '\nThe ``try`` statement\n*********************\n\nThe ``try`` statement specifies exception handlers and/or cleanup code\nfor a group of statements:\n\n   try_stmt  ::= try1_stmt | try2_stmt\n   try1_stmt ::= "try" ":" suite\n                 ("except" [expression ["as" target]] ":" suite)+\n                 ["else" ":" suite]\n                 ["finally" ":" suite]\n   try2_stmt ::= "try" ":" suite\n                 "finally" ":" suite\n\nThe ``except`` clause(s) specify one or more exception handlers. When\nno exception occurs in the ``try`` clause, no exception handler is\nexecuted. When an exception occurs in the ``try`` suite, a search for\nan exception handler is started.  This search inspects the except\nclauses in turn until one is found that matches the exception.  An\nexpression-less except clause, if present, must be last; it matches\nany exception.  For an except clause with an expression, that\nexpression is evaluated, and the clause matches the exception if the\nresulting object is "compatible" with the exception.  An object is\ncompatible with an exception if it is the class or a base class of the\nexception object or a tuple containing an item compatible with the\nexception.\n\nIf no except clause matches the exception, the search for an exception\nhandler continues in the surrounding code and on the invocation stack.\n[1]\n\nIf the evaluation of an expression in the header of an except clause\nraises an exception, the original search for a handler is canceled and\na search starts for the new exception in the surrounding code and on\nthe call stack (it is treated as if the entire ``try`` statement\nraised the exception).\n\nWhen a matching except clause is found, the exception is assigned to\nthe target specified after the ``as`` keyword in that except clause,\nif present, and the except clause\'s suite is executed.  All except\nclauses must have an executable block.  When the end of this block is\nreached, execution continues normally after the entire try statement.\n(This means that if two nested handlers exist for the same exception,\nand the exception occurs in the try clause of the inner handler, the\nouter handler will not handle the exception.)\n\nWhen an exception has been assigned using ``as target``, it is cleared\nat the end of the except clause.  This is as if\n\n   except E as N:\n       foo\n\nwas translated to\n\n   except E as N:\n       try:\n           foo\n       finally:\n           del N\n\nThis means the exception must be assigned to a different name to be\nable to refer to it after the except clause.  Exceptions are cleared\nbecause with the traceback attached to them, they form a reference\ncycle with the stack frame, keeping all locals in that frame alive\nuntil the next garbage collection occurs.\n\nBefore an except clause\'s suite is executed, details about the\nexception are stored in the ``sys`` module and can be access via\n``sys.exc_info()``. ``sys.exc_info()`` returns a 3-tuple consisting\nof: ``exc_type``, the exception class; ``exc_value``, the exception\ninstance; ``exc_traceback``, a traceback object (see section *The\nstandard type hierarchy*) identifying the point in the program where\nthe exception occurred. ``sys.exc_info()`` values are restored to\ntheir previous values (before the call) when returning from a function\nthat handled an exception.\n\nThe optional ``else`` clause is executed if and when control flows off\nthe end of the ``try`` clause. [2] Exceptions in the ``else`` clause\nare not handled by the preceding ``except`` clauses.\n\nIf ``finally`` is present, it specifies a \'cleanup\' handler.  The\n``try`` clause is executed, including any ``except`` and ``else``\nclauses.  If an exception occurs in any of the clauses and is not\nhandled, the exception is temporarily saved. The ``finally`` clause is\nexecuted.  If there is a saved exception, it is re-raised at the end\nof the ``finally`` clause. If the ``finally`` clause raises another\nexception or executes a ``return`` or ``break`` statement, the saved\nexception is lost.  The exception information is not available to the\nprogram during execution of the ``finally`` clause.\n\nWhen a ``return``, ``break`` or ``continue`` statement is executed in\nthe ``try`` suite of a ``try``...``finally`` statement, the\n``finally`` clause is also executed \'on the way out.\' A ``continue``\nstatement is illegal in the ``finally`` clause. (The reason is a\nproblem with the current implementation --- this restriction may be\nlifted in the future).\n\nAdditional information on exceptions can be found in section\n*Exceptions*, and information on using the ``raise`` statement to\ngenerate exceptions may be found in section *The raise statement*.\n',
+ 'types': '\nThe standard type hierarchy\n***************************\n\nBelow is a list of the types that are built into Python.  Extension\nmodules (written in C, Java, or other languages, depending on the\nimplementation) can define additional types.  Future versions of\nPython may add types to the type hierarchy (e.g., rational numbers,\nefficiently stored arrays of integers, etc.), although such additions\nwill often be provided via the standard library instead.\n\nSome of the type descriptions below contain a paragraph listing\n\'special attributes.\'  These are attributes that provide access to the\nimplementation and are not intended for general use.  Their definition\nmay change in the future.\n\nNone\n   This type has a single value.  There is a single object with this\n   value. This object is accessed through the built-in name ``None``.\n   It is used to signify the absence of a value in many situations,\n   e.g., it is returned from functions that don\'t explicitly return\n   anything. Its truth value is false.\n\nNotImplemented\n   This type has a single value.  There is a single object with this\n   value. This object is accessed through the built-in name\n   ``NotImplemented``. Numeric methods and rich comparison methods may\n   return this value if they do not implement the operation for the\n   operands provided.  (The interpreter will then try the reflected\n   operation, or some other fallback, depending on the operator.)  Its\n   truth value is true.\n\nEllipsis\n   This type has a single value.  There is a single object with this\n   value. This object is accessed through the literal ``...`` or the\n   built-in name ``Ellipsis``.  Its truth value is true.\n\n``numbers.Number``\n   These are created by numeric literals and returned as results by\n   arithmetic operators and arithmetic built-in functions.  Numeric\n   objects are immutable; once created their value never changes.\n   Python numbers are of course strongly related to mathematical\n   numbers, but subject to the limitations of numerical representation\n   in computers.\n\n   Python distinguishes between integers, floating point numbers, and\n   complex numbers:\n\n   ``numbers.Integral``\n      These represent elements from the mathematical set of integers\n      (positive and negative).\n\n      There are two types of integers:\n\n      Integers (``int``)\n\n         These represent numbers in an unlimited range, subject to\n         available (virtual) memory only.  For the purpose of shift\n         and mask operations, a binary representation is assumed, and\n         negative numbers are represented in a variant of 2\'s\n         complement which gives the illusion of an infinite string of\n         sign bits extending to the left.\n\n      Booleans (``bool``)\n         These represent the truth values False and True.  The two\n         objects representing the values False and True are the only\n         Boolean objects. The Boolean type is a subtype of the integer\n         type, and Boolean values behave like the values 0 and 1,\n         respectively, in almost all contexts, the exception being\n         that when converted to a string, the strings ``"False"`` or\n         ``"True"`` are returned, respectively.\n\n      The rules for integer representation are intended to give the\n      most meaningful interpretation of shift and mask operations\n      involving negative integers.\n\n   ``numbers.Real`` (``float``)\n      These represent machine-level double precision floating point\n      numbers. You are at the mercy of the underlying machine\n      architecture (and C or Java implementation) for the accepted\n      range and handling of overflow. Python does not support single-\n      precision floating point numbers; the savings in processor and\n      memory usage that are usually the reason for using these is\n      dwarfed by the overhead of using objects in Python, so there is\n      no reason to complicate the language with two kinds of floating\n      point numbers.\n\n   ``numbers.Complex`` (``complex``)\n      These represent complex numbers as a pair of machine-level\n      double precision floating point numbers.  The same caveats apply\n      as for floating point numbers. The real and imaginary parts of a\n      complex number ``z`` can be retrieved through the read-only\n      attributes ``z.real`` and ``z.imag``.\n\nSequences\n   These represent finite ordered sets indexed by non-negative\n   numbers. The built-in function ``len()`` returns the number of\n   items of a sequence. When the length of a sequence is *n*, the\n   index set contains the numbers 0, 1, ..., *n*-1.  Item *i* of\n   sequence *a* is selected by ``a[i]``.\n\n   Sequences also support slicing: ``a[i:j]`` selects all items with\n   index *k* such that *i* ``<=`` *k* ``<`` *j*.  When used as an\n   expression, a slice is a sequence of the same type.  This implies\n   that the index set is renumbered so that it starts at 0.\n\n   Some sequences also support "extended slicing" with a third "step"\n   parameter: ``a[i:j:k]`` selects all items of *a* with index *x*\n   where ``x = i + n*k``, *n* ``>=`` ``0`` and *i* ``<=`` *x* ``<``\n   *j*.\n\n   Sequences are distinguished according to their mutability:\n\n   Immutable sequences\n      An object of an immutable sequence type cannot change once it is\n      created.  (If the object contains references to other objects,\n      these other objects may be mutable and may be changed; however,\n      the collection of objects directly referenced by an immutable\n      object cannot change.)\n\n      The following types are immutable sequences:\n\n      Strings\n         The items of a string object are Unicode code units.  A\n         Unicode code unit is represented by a string object of one\n         item and can hold either a 16-bit or 32-bit value\n         representing a Unicode ordinal (the maximum value for the\n         ordinal is given in ``sys.maxunicode``, and depends on how\n         Python is configured at compile time).  Surrogate pairs may\n         be present in the Unicode object, and will be reported as two\n         separate items.  The built-in functions ``chr()`` and\n         ``ord()`` convert between code units and nonnegative integers\n         representing the Unicode ordinals as defined in the Unicode\n         Standard 3.0. Conversion from and to other encodings are\n         possible through the string method ``encode()``.\n\n      Tuples\n         The items of a tuple are arbitrary Python objects. Tuples of\n         two or more items are formed by comma-separated lists of\n         expressions.  A tuple of one item (a \'singleton\') can be\n         formed by affixing a comma to an expression (an expression by\n         itself does not create a tuple, since parentheses must be\n         usable for grouping of expressions).  An empty tuple can be\n         formed by an empty pair of parentheses.\n\n      Bytes\n         A bytes object is an immutable array.  The items are 8-bit\n         bytes, represented by integers in the range 0 <= x < 256.\n         Bytes literals (like ``b\'abc\'`` and the built-in function\n         ``bytes()`` can be used to construct bytes objects.  Also,\n         bytes objects can be decoded to strings via the ``decode()``\n         method.\n\n   Mutable sequences\n      Mutable sequences can be changed after they are created.  The\n      subscription and slicing notations can be used as the target of\n      assignment and ``del`` (delete) statements.\n\n      There are currently two intrinsic mutable sequence types:\n\n      Lists\n         The items of a list are arbitrary Python objects.  Lists are\n         formed by placing a comma-separated list of expressions in\n         square brackets. (Note that there are no special cases needed\n         to form lists of length 0 or 1.)\n\n      Byte Arrays\n         A bytearray object is a mutable array. They are created by\n         the built-in ``bytearray()`` constructor.  Aside from being\n         mutable (and hence unhashable), byte arrays otherwise provide\n         the same interface and functionality as immutable bytes\n         objects.\n\n      The extension module ``array`` provides an additional example of\n      a mutable sequence type, as does the ``collections`` module.\n\nSet types\n   These represent unordered, finite sets of unique, immutable\n   objects. As such, they cannot be indexed by any subscript. However,\n   they can be iterated over, and the built-in function ``len()``\n   returns the number of items in a set. Common uses for sets are fast\n   membership testing, removing duplicates from a sequence, and\n   computing mathematical operations such as intersection, union,\n   difference, and symmetric difference.\n\n   For set elements, the same immutability rules apply as for\n   dictionary keys. Note that numeric types obey the normal rules for\n   numeric comparison: if two numbers compare equal (e.g., ``1`` and\n   ``1.0``), only one of them can be contained in a set.\n\n   There are currently two intrinsic set types:\n\n   Sets\n      These represent a mutable set. They are created by the built-in\n      ``set()`` constructor and can be modified afterwards by several\n      methods, such as ``add()``.\n\n   Frozen sets\n      These represent an immutable set.  They are created by the\n      built-in ``frozenset()`` constructor.  As a frozenset is\n      immutable and *hashable*, it can be used again as an element of\n      another set, or as a dictionary key.\n\nMappings\n   These represent finite sets of objects indexed by arbitrary index\n   sets. The subscript notation ``a[k]`` selects the item indexed by\n   ``k`` from the mapping ``a``; this can be used in expressions and\n   as the target of assignments or ``del`` statements. The built-in\n   function ``len()`` returns the number of items in a mapping.\n\n   There is currently a single intrinsic mapping type:\n\n   Dictionaries\n      These represent finite sets of objects indexed by nearly\n      arbitrary values.  The only types of values not acceptable as\n      keys are values containing lists or dictionaries or other\n      mutable types that are compared by value rather than by object\n      identity, the reason being that the efficient implementation of\n      dictionaries requires a key\'s hash value to remain constant.\n      Numeric types used for keys obey the normal rules for numeric\n      comparison: if two numbers compare equal (e.g., ``1`` and\n      ``1.0``) then they can be used interchangeably to index the same\n      dictionary entry.\n\n      Dictionaries are mutable; they can be created by the ``{...}``\n      notation (see section *Dictionary displays*).\n\n      The extension modules ``dbm.ndbm`` and ``dbm.gnu`` provide\n      additional examples of mapping types, as does the\n      ``collections`` module.\n\nCallable types\n   These are the types to which the function call operation (see\n   section *Calls*) can be applied:\n\n   User-defined functions\n      A user-defined function object is created by a function\n      definition (see section *Function definitions*).  It should be\n      called with an argument list containing the same number of items\n      as the function\'s formal parameter list.\n\n      Special attributes:\n\n      +---------------------------+---------------------------------+-------------+\n      | Attribute                 | Meaning                         |             |\n      +===========================+=================================+=============+\n      | ``__doc__``               | The function\'s documentation    | Writable    |\n      |                           | string, or ``None`` if          |             |\n      |                           | unavailable                     |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__name__``              | The function\'s name             | Writable    |\n      +---------------------------+---------------------------------+-------------+\n      | ``__module__``            | The name of the module the      | Writable    |\n      |                           | function was defined in, or     |             |\n      |                           | ``None`` if unavailable.        |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__defaults__``          | A tuple containing default      | Writable    |\n      |                           | argument values for those       |             |\n      |                           | arguments that have defaults,   |             |\n      |                           | or ``None`` if no arguments     |             |\n      |                           | have a default value            |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__code__``              | The code object representing    | Writable    |\n      |                           | the compiled function body.     |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__globals__``           | A reference to the dictionary   | Read-only   |\n      |                           | that holds the function\'s       |             |\n      |                           | global variables --- the global |             |\n      |                           | namespace of the module in      |             |\n      |                           | which the function was defined. |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__dict__``              | The namespace supporting        | Writable    |\n      |                           | arbitrary function attributes.  |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__closure__``           | ``None`` or a tuple of cells    | Read-only   |\n      |                           | that contain bindings for the   |             |\n      |                           | function\'s free variables.      |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__annotations__``       | A dict containing annotations   | Writable    |\n      |                           | of parameters.  The keys of the |             |\n      |                           | dict are the parameter names,   |             |\n      |                           | or ``\'return\'`` for the return  |             |\n      |                           | annotation, if provided.        |             |\n      +---------------------------+---------------------------------+-------------+\n      | ``__kwdefaults__``        | A dict containing defaults for  | Writable    |\n      |                           | keyword-only parameters.        |             |\n      +---------------------------+---------------------------------+-------------+\n\n      Most of the attributes labelled "Writable" check the type of the\n      assigned value.\n\n      Function objects also support getting and setting arbitrary\n      attributes, which can be used, for example, to attach metadata\n      to functions.  Regular attribute dot-notation is used to get and\n      set such attributes. *Note that the current implementation only\n      supports function attributes on user-defined functions. Function\n      attributes on built-in functions may be supported in the\n      future.*\n\n      Additional information about a function\'s definition can be\n      retrieved from its code object; see the description of internal\n      types below.\n\n   Instance methods\n      An instance method object combines a class, a class instance and\n      any callable object (normally a user-defined function).\n\n      Special read-only attributes: ``__self__`` is the class instance\n      object, ``__func__`` is the function object; ``__doc__`` is the\n      method\'s documentation (same as ``__func__.__doc__``);\n      ``__name__`` is the method name (same as ``__func__.__name__``);\n      ``__module__`` is the name of the module the method was defined\n      in, or ``None`` if unavailable.\n\n      Methods also support accessing (but not setting) the arbitrary\n      function attributes on the underlying function object.\n\n      User-defined method objects may be created when getting an\n      attribute of a class (perhaps via an instance of that class), if\n      that attribute is a user-defined function object or a class\n      method object.\n\n      When an instance method object is created by retrieving a user-\n      defined function object from a class via one of its instances,\n      its ``__self__`` attribute is the instance, and the method\n      object is said to be bound.  The new method\'s ``__func__``\n      attribute is the original function object.\n\n      When a user-defined method object is created by retrieving\n      another method object from a class or instance, the behaviour is\n      the same as for a function object, except that the ``__func__``\n      attribute of the new instance is not the original method object\n      but its ``__func__`` attribute.\n\n      When an instance method object is created by retrieving a class\n      method object from a class or instance, its ``__self__``\n      attribute is the class itself, and its ``__func__`` attribute is\n      the function object underlying the class method.\n\n      When an instance method object is called, the underlying\n      function (``__func__``) is called, inserting the class instance\n      (``__self__``) in front of the argument list.  For instance,\n      when ``C`` is a class which contains a definition for a function\n      ``f()``, and ``x`` is an instance of ``C``, calling ``x.f(1)``\n      is equivalent to calling ``C.f(x, 1)``.\n\n      When an instance method object is derived from a class method\n      object, the "class instance" stored in ``__self__`` will\n      actually be the class itself, so that calling either ``x.f(1)``\n      or ``C.f(1)`` is equivalent to calling ``f(C,1)`` where ``f`` is\n      the underlying function.\n\n      Note that the transformation from function object to instance\n      method object happens each time the attribute is retrieved from\n      the instance.  In some cases, a fruitful optimization is to\n      assign the attribute to a local variable and call that local\n      variable. Also notice that this transformation only happens for\n      user-defined functions; other callable objects (and all non-\n      callable objects) are retrieved without transformation.  It is\n      also important to note that user-defined functions which are\n      attributes of a class instance are not converted to bound\n      methods; this *only* happens when the function is an attribute\n      of the class.\n\n   Generator functions\n      A function or method which uses the ``yield`` statement (see\n      section *The yield statement*) is called a *generator function*.\n      Such a function, when called, always returns an iterator object\n      which can be used to execute the body of the function:  calling\n      the iterator\'s ``__next__()`` method will cause the function to\n      execute until it provides a value using the ``yield`` statement.\n      When the function executes a ``return`` statement or falls off\n      the end, a ``StopIteration`` exception is raised and the\n      iterator will have reached the end of the set of values to be\n      returned.\n\n   Built-in functions\n      A built-in function object is a wrapper around a C function.\n      Examples of built-in functions are ``len()`` and ``math.sin()``\n      (``math`` is a standard built-in module). The number and type of\n      the arguments are determined by the C function. Special read-\n      only attributes: ``__doc__`` is the function\'s documentation\n      string, or ``None`` if unavailable; ``__name__`` is the\n      function\'s name; ``__self__`` is set to ``None`` (but see the\n      next item); ``__module__`` is the name of the module the\n      function was defined in or ``None`` if unavailable.\n\n   Built-in methods\n      This is really a different disguise of a built-in function, this\n      time containing an object passed to the C function as an\n      implicit extra argument.  An example of a built-in method is\n      ``alist.append()``, assuming *alist* is a list object. In this\n      case, the special read-only attribute ``__self__`` is set to the\n      object denoted by *list*.\n\n   Classes\n      Classes are callable.  These objects normally act as factories\n      for new instances of themselves, but variations are possible for\n      class types that override ``__new__()``.  The arguments of the\n      call are passed to ``__new__()`` and, in the typical case, to\n      ``__init__()`` to initialize the new instance.\n\n   Class Instances\n      Instances of arbitrary classes can be made callable by defining\n      a ``__call__()`` method in their class.\n\nModules\n   Modules are imported by the ``import`` statement (see section *The\n   import statement*). A module object has a namespace implemented by\n   a dictionary object (this is the dictionary referenced by the\n   __globals__ attribute of functions defined in the module).\n   Attribute references are translated to lookups in this dictionary,\n   e.g., ``m.x`` is equivalent to ``m.__dict__["x"]``. A module object\n   does not contain the code object used to initialize the module\n   (since it isn\'t needed once the initialization is done).\n\n   Attribute assignment updates the module\'s namespace dictionary,\n   e.g., ``m.x = 1`` is equivalent to ``m.__dict__["x"] = 1``.\n\n   Special read-only attribute: ``__dict__`` is the module\'s namespace\n   as a dictionary object.\n\n   Predefined (writable) attributes: ``__name__`` is the module\'s\n   name; ``__doc__`` is the module\'s documentation string, or ``None``\n   if unavailable; ``__file__`` is the pathname of the file from which\n   the module was loaded, if it was loaded from a file. The\n   ``__file__`` attribute is not present for C modules that are\n   statically linked into the interpreter; for extension modules\n   loaded dynamically from a shared library, it is the pathname of the\n   shared library file.\n\nCustom classes\n   Custom class types are typically created by class definitions (see\n   section *Class definitions*).  A class has a namespace implemented\n   by a dictionary object. Class attribute references are translated\n   to lookups in this dictionary, e.g., ``C.x`` is translated to\n   ``C.__dict__["x"]`` (although there are a number of hooks which\n   allow for other means of locating attributes). When the attribute\n   name is not found there, the attribute search continues in the base\n   classes. This search of the base classes uses the C3 method\n   resolution order which behaves correctly even in the presence of\n   \'diamond\' inheritance structures where there are multiple\n   inheritance paths leading back to a common ancestor. Additional\n   details on the C3 MRO used by Python can be found in the\n   documentation accompanying the 2.3 release at\n   http://www.python.org/download/releases/2.3/mro/.\n\n   When a class attribute reference (for class ``C``, say) would yield\n   a class method object, it is transformed into an instance method\n   object whose ``__self__`` attributes is ``C``.  When it would yield\n   a static method object, it is transformed into the object wrapped\n   by the static method object. See section *Implementing Descriptors*\n   for another way in which attributes retrieved from a class may\n   differ from those actually contained in its ``__dict__``.\n\n   Class attribute assignments update the class\'s dictionary, never\n   the dictionary of a base class.\n\n   A class object can be called (see above) to yield a class instance\n   (see below).\n\n   Special attributes: ``__name__`` is the class name; ``__module__``\n   is the module name in which the class was defined; ``__dict__`` is\n   the dictionary containing the class\'s namespace; ``__bases__`` is a\n   tuple (possibly empty or a singleton) containing the base classes,\n   in the order of their occurrence in the base class list;\n   ``__doc__`` is the class\'s documentation string, or None if\n   undefined.\n\nClass instances\n   A class instance is created by calling a class object (see above).\n   A class instance has a namespace implemented as a dictionary which\n   is the first place in which attribute references are searched.\n   When an attribute is not found there, and the instance\'s class has\n   an attribute by that name, the search continues with the class\n   attributes.  If a class attribute is found that is a user-defined\n   function object, it is transformed into an instance method object\n   whose ``__self__`` attribute is the instance.  Static method and\n   class method objects are also transformed; see above under\n   "Classes".  See section *Implementing Descriptors* for another way\n   in which attributes of a class retrieved via its instances may\n   differ from the objects actually stored in the class\'s\n   ``__dict__``.  If no class attribute is found, and the object\'s\n   class has a ``__getattr__()`` method, that is called to satisfy the\n   lookup.\n\n   Attribute assignments and deletions update the instance\'s\n   dictionary, never a class\'s dictionary.  If the class has a\n   ``__setattr__()`` or ``__delattr__()`` method, this is called\n   instead of updating the instance dictionary directly.\n\n   Class instances can pretend to be numbers, sequences, or mappings\n   if they have methods with certain special names.  See section\n   *Special method names*.\n\n   Special attributes: ``__dict__`` is the attribute dictionary;\n   ``__class__`` is the instance\'s class.\n\nI/O objects (also known as file objects)\n   A file object represents an open file.  Various shortcuts are\n   available to create file objects: the ``open()`` built-in function,\n   and also ``os.popen()``, ``os.fdopen()``, and the ``makefile()``\n   method of socket objects (and perhaps by other functions or methods\n   provided by extension modules).\n\n   The objects ``sys.stdin``, ``sys.stdout`` and ``sys.stderr`` are\n   initialized to file objects corresponding to the interpreter\'s\n   standard input, output and error streams; they are all open in text\n   mode and therefore follow the interface defined by the\n   ``io.TextIOBase`` abstract class.\n\nInternal types\n   A few types used internally by the interpreter are exposed to the\n   user. Their definitions may change with future versions of the\n   interpreter, but they are mentioned here for completeness.\n\n   Code objects\n      Code objects represent *byte-compiled* executable Python code,\n      or *bytecode*. The difference between a code object and a\n      function object is that the function object contains an explicit\n      reference to the function\'s globals (the module in which it was\n      defined), while a code object contains no context; also the\n      default argument values are stored in the function object, not\n      in the code object (because they represent values calculated at\n      run-time).  Unlike function objects, code objects are immutable\n      and contain no references (directly or indirectly) to mutable\n      objects.\n\n      Special read-only attributes: ``co_name`` gives the function\n      name; ``co_argcount`` is the number of positional arguments\n      (including arguments with default values); ``co_nlocals`` is the\n      number of local variables used by the function (including\n      arguments); ``co_varnames`` is a tuple containing the names of\n      the local variables (starting with the argument names);\n      ``co_cellvars`` is a tuple containing the names of local\n      variables that are referenced by nested functions;\n      ``co_freevars`` is a tuple containing the names of free\n      variables; ``co_code`` is a string representing the sequence of\n      bytecode instructions; ``co_consts`` is a tuple containing the\n      literals used by the bytecode; ``co_names`` is a tuple\n      containing the names used by the bytecode; ``co_filename`` is\n      the filename from which the code was compiled;\n      ``co_firstlineno`` is the first line number of the function;\n      ``co_lnotab`` is a string encoding the mapping from bytecode\n      offsets to line numbers (for details see the source code of the\n      interpreter); ``co_stacksize`` is the required stack size\n      (including local variables); ``co_flags`` is an integer encoding\n      a number of flags for the interpreter.\n\n      The following flag bits are defined for ``co_flags``: bit\n      ``0x04`` is set if the function uses the ``*arguments`` syntax\n      to accept an arbitrary number of positional arguments; bit\n      ``0x08`` is set if the function uses the ``**keywords`` syntax\n      to accept arbitrary keyword arguments; bit ``0x20`` is set if\n      the function is a generator.\n\n      Future feature declarations (``from __future__ import\n      division``) also use bits in ``co_flags`` to indicate whether a\n      code object was compiled with a particular feature enabled: bit\n      ``0x2000`` is set if the function was compiled with future\n      division enabled; bits ``0x10`` and ``0x1000`` were used in\n      earlier versions of Python.\n\n      Other bits in ``co_flags`` are reserved for internal use.\n\n      If a code object represents a function, the first item in\n      ``co_consts`` is the documentation string of the function, or\n      ``None`` if undefined.\n\n   Frame objects\n      Frame objects represent execution frames.  They may occur in\n      traceback objects (see below).\n\n      Special read-only attributes: ``f_back`` is to the previous\n      stack frame (towards the caller), or ``None`` if this is the\n      bottom stack frame; ``f_code`` is the code object being executed\n      in this frame; ``f_locals`` is the dictionary used to look up\n      local variables; ``f_globals`` is used for global variables;\n      ``f_builtins`` is used for built-in (intrinsic) names;\n      ``f_lasti`` gives the precise instruction (this is an index into\n      the bytecode string of the code object).\n\n      Special writable attributes: ``f_trace``, if not ``None``, is a\n      function called at the start of each source code line (this is\n      used by the debugger); ``f_lineno`` is the current line number\n      of the frame --- writing to this from within a trace function\n      jumps to the given line (only for the bottom-most frame).  A\n      debugger can implement a Jump command (aka Set Next Statement)\n      by writing to f_lineno.\n\n   Traceback objects\n      Traceback objects represent a stack trace of an exception.  A\n      traceback object is created when an exception occurs.  When the\n      search for an exception handler unwinds the execution stack, at\n      each unwound level a traceback object is inserted in front of\n      the current traceback.  When an exception handler is entered,\n      the stack trace is made available to the program. (See section\n      *The try statement*.) It is accessible as the third item of the\n      tuple returned by ``sys.exc_info()``. When the program contains\n      no suitable handler, the stack trace is written (nicely\n      formatted) to the standard error stream; if the interpreter is\n      interactive, it is also made available to the user as\n      ``sys.last_traceback``.\n\n      Special read-only attributes: ``tb_next`` is the next level in\n      the stack trace (towards the frame where the exception\n      occurred), or ``None`` if there is no next level; ``tb_frame``\n      points to the execution frame of the current level;\n      ``tb_lineno`` gives the line number where the exception\n      occurred; ``tb_lasti`` indicates the precise instruction.  The\n      line number and last instruction in the traceback may differ\n      from the line number of its frame object if the exception\n      occurred in a ``try`` statement with no matching except clause\n      or with a finally clause.\n\n   Slice objects\n      Slice objects are used to represent slices for ``__getitem__()``\n      methods.  They are also created by the built-in ``slice()``\n      function.\n\n      Special read-only attributes: ``start`` is the lower bound;\n      ``stop`` is the upper bound; ``step`` is the step value; each is\n      ``None`` if omitted. These attributes can have any type.\n\n      Slice objects support one method:\n\n      slice.indices(self, length)\n\n         This method takes a single integer argument *length* and\n         computes information about the slice that the slice object\n         would describe if applied to a sequence of *length* items.\n         It returns a tuple of three integers; respectively these are\n         the *start* and *stop* indices and the *step* or stride\n         length of the slice. Missing or out-of-bounds indices are\n         handled in a manner consistent with regular slices.\n\n   Static method objects\n      Static method objects provide a way of defeating the\n      transformation of function objects to method objects described\n      above. A static method object is a wrapper around any other\n      object, usually a user-defined method object. When a static\n      method object is retrieved from a class or a class instance, the\n      object actually returned is the wrapped object, which is not\n      subject to any further transformation. Static method objects are\n      not themselves callable, although the objects they wrap usually\n      are. Static method objects are created by the built-in\n      ``staticmethod()`` constructor.\n\n   Class method objects\n      A class method object, like a static method object, is a wrapper\n      around another object that alters the way in which that object\n      is retrieved from classes and class instances. The behaviour of\n      class method objects upon such retrieval is described above,\n      under "User-defined methods". Class method objects are created\n      by the built-in ``classmethod()`` constructor.\n',
  'typesfunctions': '\nFunctions\n*********\n\nFunction objects are created by function definitions.  The only\noperation on a function object is to call it: ``func(argument-list)``.\n\nThere are really two flavors of function objects: built-in functions\nand user-defined functions.  Both support the same operation (to call\nthe function), but the implementation is different, hence the\ndifferent object types.\n\nSee *Function definitions* for more information.\n',
- 'typesmapping': '\nMapping Types --- ``dict``\n**************************\n\nA *mapping* object maps *hashable* values to arbitrary objects.\nMappings are mutable objects.  There is currently only one standard\nmapping type, the *dictionary*.  (For other containers see the built\nin ``list``, ``set``, and ``tuple`` classes, and the ``collections``\nmodule.)\n\nA dictionary\'s keys are *almost* arbitrary values.  Values that are\nnot *hashable*, that is, values containing lists, dictionaries or\nother mutable types (that are compared by value rather than by object\nidentity) may not be used as keys.  Numeric types used for keys obey\nthe normal rules for numeric comparison: if two numbers compare equal\n(such as ``1`` and ``1.0``) then they can be used interchangeably to\nindex the same dictionary entry.  (Note however, that since computers\nstore floating-point numbers as approximations it is usually unwise to\nuse them as dictionary keys.)\n\nDictionaries can be created by placing a comma-separated list of\n``key: value`` pairs within braces, for example: ``{\'jack\': 4098,\n\'sjoerd\': 4127}`` or ``{4098: \'jack\', 4127: \'sjoerd\'}``, or by the\n``dict`` constructor.\n\nclass class dict([arg])\n\n   Return a new dictionary initialized from an optional positional\n   argument or from a set of keyword arguments.  If no arguments are\n   given, return a new empty dictionary.  If the positional argument\n   *arg* is a mapping object, return a dictionary mapping the same\n   keys to the same values as does the mapping object.  Otherwise the\n   positional argument must be a sequence, a container that supports\n   iteration, or an iterator object.  The elements of the argument\n   must each also be of one of those kinds, and each must in turn\n   contain exactly two objects.  The first is used as a key in the new\n   dictionary, and the second as the key\'s value.  If a given key is\n   seen more than once, the last value associated with it is retained\n   in the new dictionary.\n\n   If keyword arguments are given, the keywords themselves with their\n   associated values are added as items to the dictionary.  If a key\n   is specified both in the positional argument and as a keyword\n   argument, the value associated with the keyword is retained in the\n   dictionary.  For example, these all return a dictionary equal to\n   ``{"one": 2, "two": 3}``:\n\n   * ``dict(one=2, two=3)``\n\n   * ``dict({\'one\': 2, \'two\': 3})``\n\n   * ``dict(zip((\'one\', \'two\'), (2, 3)))``\n\n   * ``dict([[\'two\', 3], [\'one\', 2]])``\n\n   The first example only works for keys that are valid Python\n   identifiers; the others work with any valid keys.\n\n   These are the operations that dictionaries support (and therefore,\n   custom mapping types should support too):\n\n   len(d)\n\n      Return the number of items in the dictionary *d*.\n\n   d[key]\n\n      Return the item of *d* with key *key*.  Raises a ``KeyError`` if\n      *key* is not in the map.\n\n      If a subclass of dict defines a method ``__missing__()``, if the\n      key *key* is not present, the ``d[key]`` operation calls that\n      method with the key *key* as argument.  The ``d[key]`` operation\n      then returns or raises whatever is returned or raised by the\n      ``__missing__(key)`` call if the key is not present. No other\n      operations or methods invoke ``__missing__()``. If\n      ``__missing__()`` is not defined, ``KeyError`` is raised.\n      ``__missing__()`` must be a method; it cannot be an instance\n      variable. For an example, see ``collections.defaultdict``.\n\n   d[key] = value\n\n      Set ``d[key]`` to *value*.\n\n   del d[key]\n\n      Remove ``d[key]`` from *d*.  Raises a ``KeyError`` if *key* is\n      not in the map.\n\n   key in d\n\n      Return ``True`` if *d* has a key *key*, else ``False``.\n\n   key not in d\n\n      Equivalent to ``not key in d``.\n\n   iter(d)\n\n      Return an iterator over the keys of the dictionary.  This is a\n      shortcut for ``iterkeys()``.\n\n   clear()\n\n      Remove all items from the dictionary.\n\n   copy()\n\n      Return a shallow copy of the dictionary.\n\n   classmethod fromkeys(seq[, value])\n\n      Create a new dictionary with keys from *seq* and values set to\n      *value*.\n\n      ``fromkeys()`` is a class method that returns a new dictionary.\n      *value* defaults to ``None``.\n\n   get(key[, default])\n\n      Return the value for *key* if *key* is in the dictionary, else\n      *default*. If *default* is not given, it defaults to ``None``,\n      so that this method never raises a ``KeyError``.\n\n   items()\n\n      Return a new view of the dictionary\'s items (``(key, value)``\n      pairs).  See below for documentation of view objects.\n\n   keys()\n\n      Return a new view of the dictionary\'s keys.  See below for\n      documentation of view objects.\n\n   pop(key[, default])\n\n      If *key* is in the dictionary, remove it and return its value,\n      else return *default*.  If *default* is not given and *key* is\n      not in the dictionary, a ``KeyError`` is raised.\n\n   popitem()\n\n      Remove and return an arbitrary ``(key, value)`` pair from the\n      dictionary.\n\n      ``popitem()`` is useful to destructively iterate over a\n      dictionary, as often used in set algorithms.  If the dictionary\n      is empty, calling ``popitem()`` raises a ``KeyError``.\n\n   setdefault(key[, default])\n\n      If *key* is in the dictionary, return its value.  If not, insert\n      *key* with a value of *default* and return *default*.  *default*\n      defaults to ``None``.\n\n   update([other])\n\n      Update the dictionary with the key/value pairs from *other*,\n      overwriting existing keys.  Return ``None``.\n\n         ``update()`` accepts either another dictionary object or an\n         iterable of key/value pairs (as a tuple or other iterable of\n         length two).  If keyword arguments are specified, the\n         dictionary is then is updated with those key/value pairs:\n         ``d.update(red=1, blue=2)``.\n\n   values()\n\n      Return a new view of the dictionary\'s values.  See below for\n      documentation of view objects.\n\n\nDictionary view objects\n=======================\n\nThe objects returned by ``dict.keys()``, ``dict.values()`` and\n``dict.items()`` are *view objects*.  They provide a dynamic view on\nthe dictionary\'s entries, which means that when the dictionary\nchanges, the view reflects these changes.\n\nDictionary views can be iterated over to yield their respective data,\nand support membership tests:\n\nlen(dictview)\n\n   Return the number of entries in the dictionary.\n\niter(dictview)\n\n   Return an iterator over the keys, values or items (represented as\n   tuples of ``(key, value)``) in the dictionary.\n\n   Keys and values are iterated over in an arbitrary order which is\n   non-random, varies across Python implementations, and depends on\n   the dictionary\'s history of insertions and deletions. If keys,\n   values and items views are iterated over with no intervening\n   modifications to the dictionary, the order of items will directly\n   correspond.  This allows the creation of ``(value, key)`` pairs\n   using ``zip()``: ``pairs = zip(d.values(), d.keys())``.  Another\n   way to create the same list is ``pairs = [(v, k) for (k, v) in\n   d.items()]``.\n\n   Iterating views while adding or deleting entries in the dictionary\n   may raise a ``RuntimeError`` or fail to iterate over all entries.\n\nx in dictview\n\n   Return ``True`` if *x* is in the underlying dictionary\'s keys,\n   values or items (in the latter case, *x* should be a ``(key,\n   value)`` tuple).\n\nKeys views are set-like since their entries are unique and hashable.\nIf all values are hashable, so that (key, value) pairs are unique and\nhashable, then the items view is also set-like.  (Values views are not\ntreated as set-like since the entries are generally not unique.)  Then\nthese set operations are available ("other" refers either to another\nview or a set):\n\ndictview & other\n\n   Return the intersection of the dictview and the other object as a\n   new set.\n\ndictview | other\n\n   Return the union of the dictview and the other object as a new set.\n\ndictview - other\n\n   Return the difference between the dictview and the other object\n   (all elements in *dictview* that aren\'t in *other*) as a new set.\n\ndictview ^ other\n\n   Return the symmetric difference (all elements either in *dictview*\n   or *other*, but not in both) of the dictview and the other object\n   as a new set.\n\nAn example of dictionary view usage:\n\n   >>> dishes = {\'eggs\': 2, \'sausage\': 1, \'bacon\': 1, \'spam\': 500}\n   >>> keys = dishes.keys()\n   >>> values = dishes.values()\n\n   >>> # iteration\n   >>> n = 0\n   >>> for val in values:\n   ...     n += val\n   >>> print(n)\n   504\n\n   >>> # keys and values are iterated over in the same order\n   >>> list(keys)\n   [\'eggs\', \'bacon\', \'sausage\', \'spam\']\n   >>> list(values)\n   [2, 1, 1, 500]\n\n   >>> # view objects are dynamic and reflect dict changes\n   >>> del dishes[\'eggs\']\n   >>> del dishes[\'sausage\']\n   >>> list(keys)\n   [\'spam\', \'bacon\']\n\n   >>> # set operations\n   >>> keys & {\'eggs\', \'bacon\', \'salad\'}\n   {\'bacon\'}\n',
+ 'typesmapping': '\nMapping Types --- ``dict``\n**************************\n\nA *mapping* object maps *hashable* values to arbitrary objects.\nMappings are mutable objects.  There is currently only one standard\nmapping type, the *dictionary*.  (For other containers see the built\nin ``list``, ``set``, and ``tuple`` classes, and the ``collections``\nmodule.)\n\nA dictionary\'s keys are *almost* arbitrary values.  Values that are\nnot *hashable*, that is, values containing lists, dictionaries or\nother mutable types (that are compared by value rather than by object\nidentity) may not be used as keys.  Numeric types used for keys obey\nthe normal rules for numeric comparison: if two numbers compare equal\n(such as ``1`` and ``1.0``) then they can be used interchangeably to\nindex the same dictionary entry.  (Note however, that since computers\nstore floating-point numbers as approximations it is usually unwise to\nuse them as dictionary keys.)\n\nDictionaries can be created by placing a comma-separated list of\n``key: value`` pairs within braces, for example: ``{\'jack\': 4098,\n\'sjoerd\': 4127}`` or ``{4098: \'jack\', 4127: \'sjoerd\'}``, or by the\n``dict`` constructor.\n\nclass class dict([arg])\n\n   Return a new dictionary initialized from an optional positional\n   argument or from a set of keyword arguments.  If no arguments are\n   given, return a new empty dictionary.  If the positional argument\n   *arg* is a mapping object, return a dictionary mapping the same\n   keys to the same values as does the mapping object.  Otherwise the\n   positional argument must be a sequence, a container that supports\n   iteration, or an iterator object.  The elements of the argument\n   must each also be of one of those kinds, and each must in turn\n   contain exactly two objects.  The first is used as a key in the new\n   dictionary, and the second as the key\'s value.  If a given key is\n   seen more than once, the last value associated with it is retained\n   in the new dictionary.\n\n   If keyword arguments are given, the keywords themselves with their\n   associated values are added as items to the dictionary.  If a key\n   is specified both in the positional argument and as a keyword\n   argument, the value associated with the keyword is retained in the\n   dictionary.  For example, these all return a dictionary equal to\n   ``{"one": 2, "two": 3}``:\n\n   * ``dict(one=2, two=3)``\n\n   * ``dict({\'one\': 2, \'two\': 3})``\n\n   * ``dict(zip((\'one\', \'two\'), (2, 3)))``\n\n   * ``dict([[\'two\', 3], [\'one\', 2]])``\n\n   The first example only works for keys that are valid Python\n   identifiers; the others work with any valid keys.\n\n   These are the operations that dictionaries support (and therefore,\n   custom mapping types should support too):\n\n   len(d)\n\n      Return the number of items in the dictionary *d*.\n\n   d[key]\n\n      Return the item of *d* with key *key*.  Raises a ``KeyError`` if\n      *key* is not in the map.\n\n      If a subclass of dict defines a method ``__missing__()``, if the\n      key *key* is not present, the ``d[key]`` operation calls that\n      method with the key *key* as argument.  The ``d[key]`` operation\n      then returns or raises whatever is returned or raised by the\n      ``__missing__(key)`` call if the key is not present. No other\n      operations or methods invoke ``__missing__()``. If\n      ``__missing__()`` is not defined, ``KeyError`` is raised.\n      ``__missing__()`` must be a method; it cannot be an instance\n      variable. For an example, see ``collections.defaultdict``.\n\n   d[key] = value\n\n      Set ``d[key]`` to *value*.\n\n   del d[key]\n\n      Remove ``d[key]`` from *d*.  Raises a ``KeyError`` if *key* is\n      not in the map.\n\n   key in d\n\n      Return ``True`` if *d* has a key *key*, else ``False``.\n\n   key not in d\n\n      Equivalent to ``not key in d``.\n\n   iter(d)\n\n      Return an iterator over the keys of the dictionary.  This is a\n      shortcut for ``iter(d.keys())``.\n\n   clear()\n\n      Remove all items from the dictionary.\n\n   copy()\n\n      Return a shallow copy of the dictionary.\n\n   classmethod fromkeys(seq[, value])\n\n      Create a new dictionary with keys from *seq* and values set to\n      *value*.\n\n      ``fromkeys()`` is a class method that returns a new dictionary.\n      *value* defaults to ``None``.\n\n   get(key[, default])\n\n      Return the value for *key* if *key* is in the dictionary, else\n      *default*. If *default* is not given, it defaults to ``None``,\n      so that this method never raises a ``KeyError``.\n\n   items()\n\n      Return a new view of the dictionary\'s items (``(key, value)``\n      pairs).  See below for documentation of view objects.\n\n   keys()\n\n      Return a new view of the dictionary\'s keys.  See below for\n      documentation of view objects.\n\n   pop(key[, default])\n\n      If *key* is in the dictionary, remove it and return its value,\n      else return *default*.  If *default* is not given and *key* is\n      not in the dictionary, a ``KeyError`` is raised.\n\n   popitem()\n\n      Remove and return an arbitrary ``(key, value)`` pair from the\n      dictionary.\n\n      ``popitem()`` is useful to destructively iterate over a\n      dictionary, as often used in set algorithms.  If the dictionary\n      is empty, calling ``popitem()`` raises a ``KeyError``.\n\n   setdefault(key[, default])\n\n      If *key* is in the dictionary, return its value.  If not, insert\n      *key* with a value of *default* and return *default*.  *default*\n      defaults to ``None``.\n\n   update([other])\n\n      Update the dictionary with the key/value pairs from *other*,\n      overwriting existing keys.  Return ``None``.\n\n         ``update()`` accepts either another dictionary object or an\n         iterable of key/value pairs (as a tuple or other iterable of\n         length two).  If keyword arguments are specified, the\n         dictionary is then updated with those key/value pairs:\n         ``d.update(red=1, blue=2)``.\n\n   values()\n\n      Return a new view of the dictionary\'s values.  See below for\n      documentation of view objects.\n\n\nDictionary view objects\n=======================\n\nThe objects returned by ``dict.keys()``, ``dict.values()`` and\n``dict.items()`` are *view objects*.  They provide a dynamic view on\nthe dictionary\'s entries, which means that when the dictionary\nchanges, the view reflects these changes.\n\nDictionary views can be iterated over to yield their respective data,\nand support membership tests:\n\nlen(dictview)\n\n   Return the number of entries in the dictionary.\n\niter(dictview)\n\n   Return an iterator over the keys, values or items (represented as\n   tuples of ``(key, value)``) in the dictionary.\n\n   Keys and values are iterated over in an arbitrary order which is\n   non-random, varies across Python implementations, and depends on\n   the dictionary\'s history of insertions and deletions. If keys,\n   values and items views are iterated over with no intervening\n   modifications to the dictionary, the order of items will directly\n   correspond.  This allows the creation of ``(value, key)`` pairs\n   using ``zip()``: ``pairs = zip(d.values(), d.keys())``.  Another\n   way to create the same list is ``pairs = [(v, k) for (k, v) in\n   d.items()]``.\n\n   Iterating views while adding or deleting entries in the dictionary\n   may raise a ``RuntimeError`` or fail to iterate over all entries.\n\nx in dictview\n\n   Return ``True`` if *x* is in the underlying dictionary\'s keys,\n   values or items (in the latter case, *x* should be a ``(key,\n   value)`` tuple).\n\nKeys views are set-like since their entries are unique and hashable.\nIf all values are hashable, so that (key, value) pairs are unique and\nhashable, then the items view is also set-like.  (Values views are not\ntreated as set-like since the entries are generally not unique.)  Then\nthese set operations are available ("other" refers either to another\nview or a set):\n\ndictview & other\n\n   Return the intersection of the dictview and the other object as a\n   new set.\n\ndictview | other\n\n   Return the union of the dictview and the other object as a new set.\n\ndictview - other\n\n   Return the difference between the dictview and the other object\n   (all elements in *dictview* that aren\'t in *other*) as a new set.\n\ndictview ^ other\n\n   Return the symmetric difference (all elements either in *dictview*\n   or *other*, but not in both) of the dictview and the other object\n   as a new set.\n\nAn example of dictionary view usage:\n\n   >>> dishes = {\'eggs\': 2, \'sausage\': 1, \'bacon\': 1, \'spam\': 500}\n   >>> keys = dishes.keys()\n   >>> values = dishes.values()\n\n   >>> # iteration\n   >>> n = 0\n   >>> for val in values:\n   ...     n += val\n   >>> print(n)\n   504\n\n   >>> # keys and values are iterated over in the same order\n   >>> list(keys)\n   [\'eggs\', \'bacon\', \'sausage\', \'spam\']\n   >>> list(values)\n   [2, 1, 1, 500]\n\n   >>> # view objects are dynamic and reflect dict changes\n   >>> del dishes[\'eggs\']\n   >>> del dishes[\'sausage\']\n   >>> list(keys)\n   [\'spam\', \'bacon\']\n\n   >>> # set operations\n   >>> keys & {\'eggs\', \'bacon\', \'salad\'}\n   {\'bacon\'}\n',
  'typesmethods': "\nMethods\n*******\n\nMethods are functions that are called using the attribute notation.\nThere are two flavors: built-in methods (such as ``append()`` on\nlists) and class instance methods.  Built-in methods are described\nwith the types that support them.\n\nIf you access a method (a function defined in a class namespace)\nthrough an instance, you get a special object: a *bound method* (also\ncalled *instance method*) object. When called, it will add the\n``self`` argument to the argument list.  Bound methods have two\nspecial read-only attributes: ``m.__self__`` is the object on which\nthe method operates, and ``m.__func__`` is the function implementing\nthe method.  Calling ``m(arg-1, arg-2, ..., arg-n)`` is completely\nequivalent to calling ``m.__func__(m.__self__, arg-1, arg-2, ...,\narg-n)``.\n\nLike function objects, bound method objects support getting arbitrary\nattributes.  However, since method attributes are actually stored on\nthe underlying function object (``meth.__func__``), setting method\nattributes on bound methods is disallowed.  Attempting to set a method\nattribute results in a ``TypeError`` being raised.  In order to set a\nmethod attribute, you need to explicitly set it on the underlying\nfunction object:\n\n   class C:\n       def method(self):\n           pass\n\n   c = C()\n   c.method.__func__.whoami = 'my name is c'\n\nSee *The standard type hierarchy* for more information.\n",
  'typesmodules': "\nModules\n*******\n\nThe only special operation on a module is attribute access:\n``m.name``, where *m* is a module and *name* accesses a name defined\nin *m*'s symbol table. Module attributes can be assigned to.  (Note\nthat the ``import`` statement is not, strictly speaking, an operation\non a module object; ``import foo`` does not require a module object\nnamed *foo* to exist, rather it requires an (external) *definition*\nfor a module named *foo* somewhere.)\n\nA special member of every module is ``__dict__``. This is the\ndictionary containing the module's symbol table. Modifying this\ndictionary will actually change the module's symbol table, but direct\nassignment to the ``__dict__`` attribute is not possible (you can\nwrite ``m.__dict__['a'] = 1``, which defines ``m.a`` to be ``1``, but\nyou can't write ``m.__dict__ = {}``).  Modifying ``__dict__`` directly\nis not recommended.\n\nModules built into the interpreter are written like this: ``<module\n'sys' (built-in)>``.  If loaded from a file, they are written as\n``<module 'os' from '/usr/local/lib/pythonX.Y/os.pyc'>``.\n",
- 'typesseq': '\nSequence Types --- ``str``, ``bytes``, ``bytearray``, ``list``, ``tuple``, ``range``\n************************************************************************************\n\nThere are six sequence types: strings, byte sequences (``bytes``\nobjects), byte arrays (``bytearray`` objects), lists, tuples, and\nrange objects.  For other containers see the built in ``dict`` and\n``set`` classes, and the ``collections`` module.\n\nStrings contain Unicode characters.  Their literals are written in\nsingle or double quotes: ``\'xyzzy\'``, ``"frobozz"``.  See *String and\nBytes literals* for more about string literals.  In addition to the\nfunctionality described here, there are also string-specific methods\ndescribed in the *String Methods* section.\n\nBytes and bytearray objects contain single bytes -- the former is\nimmutable while the latter is a mutable sequence.  Bytes objects can\nbe constructed the constructor, ``bytes()``, and from literals; use a\n``b`` prefix with normal string syntax: ``b\'xyzzy\'``.  To construct\nbyte arrays, use the ``bytearray()`` function.\n\nWarning: While string objects are sequences of characters (represented by\n  strings of length 1), bytes and bytearray objects are sequences of\n  *integers* (between 0 and 255), representing the ASCII value of\n  single bytes.  That means that for a bytes or bytearray object *b*,\n  ``b[0]`` will be an integer, while ``b[0:1]`` will be a bytes or\n  bytearray object of length 1.  The representation of bytes objects\n  uses the literal format (``b\'...\'``) since it is generally more\n  useful than e.g. ``bytes([50, 19, 100])``.  You can always convert a\n  bytes object into a list of integers using ``list(b)``.Also, while\n  in previous Python versions, byte strings and Unicode strings could\n  be exchanged for each other rather freely (barring encoding issues),\n  strings and bytes are now completely separate concepts.  There\'s no\n  implicit en-/decoding if you pass and object of the wrong type.  A\n  string always compares unequal to a bytes or bytearray object.\n\nLists are constructed with square brackets, separating items with\ncommas: ``[a, b, c]``.  Tuples are constructed by the comma operator\n(not within square brackets), with or without enclosing parentheses,\nbut an empty tuple must have the enclosing parentheses, such as ``a,\nb, c`` or ``()``.  A single item tuple must have a trailing comma,\nsuch as ``(d,)``.\n\nObjects of type range are created using the ``range()`` function.\nThey don\'t support slicing, concatenation or repetition, and using\n``in``, ``not in``, ``min()`` or ``max()`` on them is inefficient.\n\nMost sequence types support the following operations.  The ``in`` and\n``not in`` operations have the same priorities as the comparison\noperations.  The ``+`` and ``*`` operations have the same priority as\nthe corresponding numeric operations. [3] Additional methods are\nprovided for *Mutable Sequence Types*.\n\nThis table lists the sequence operations sorted in ascending priority\n(operations in the same box have the same priority).  In the table,\n*s* and *t* are sequences of the same type; *n*, *i* and *j* are\nintegers:\n\n+--------------------+----------------------------------+------------+\n| Operation          | Result                           | Notes      |\n+====================+==================================+============+\n| ``x in s``         | ``True`` if an item of *s* is    | (1)        |\n|                    | equal to *x*, else ``False``     |            |\n+--------------------+----------------------------------+------------+\n| ``x not in s``     | ``False`` if an item of *s* is   | (1)        |\n|                    | equal to *x*, else ``True``      |            |\n+--------------------+----------------------------------+------------+\n| ``s + t``          | the concatenation of *s* and *t* | (6)        |\n+--------------------+----------------------------------+------------+\n| ``s * n, n * s``   | *n* shallow copies of *s*        | (2)        |\n|                    | concatenated                     |            |\n+--------------------+----------------------------------+------------+\n| ``s[i]``           | *i*\'th item of *s*, origin 0     | (3)        |\n+--------------------+----------------------------------+------------+\n| ``s[i:j]``         | slice of *s* from *i* to *j*     | (3)(4)     |\n+--------------------+----------------------------------+------------+\n| ``s[i:j:k]``       | slice of *s* from *i* to *j*     | (3)(5)     |\n|                    | with step *k*                    |            |\n+--------------------+----------------------------------+------------+\n| ``len(s)``         | length of *s*                    |            |\n+--------------------+----------------------------------+------------+\n| ``min(s)``         | smallest item of *s*             |            |\n+--------------------+----------------------------------+------------+\n| ``max(s)``         | largest item of *s*              |            |\n+--------------------+----------------------------------+------------+\n\nSequence types also support comparisons.  In particular, tuples and\nlists are compared lexicographically by comparing corresponding\nelements.  This means that to compare equal, every element must\ncompare equal and the two sequences must be of the same type and have\nthe same length.  (For full details see *Comparisons* in the language\nreference.)\n\nNotes:\n\n1. When *s* is a string object, the ``in`` and ``not in`` operations\n   act like a substring test.\n\n2. Values of *n* less than ``0`` are treated as ``0`` (which yields an\n   empty sequence of the same type as *s*).  Note also that the copies\n   are shallow; nested structures are not copied.  This often haunts\n   new Python programmers; consider:\n\n   >>> lists = [[]] * 3\n   >>> lists\n   [[], [], []]\n   >>> lists[0].append(3)\n   >>> lists\n   [[3], [3], [3]]\n\n   What has happened is that ``[[]]`` is a one-element list containing\n   an empty list, so all three elements of ``[[]] * 3`` are (pointers\n   to) this single empty list.  Modifying any of the elements of\n   ``lists`` modifies this single list. You can create a list of\n   different lists this way:\n\n   >>> lists = [[] for i in range(3)]\n   >>> lists[0].append(3)\n   >>> lists[1].append(5)\n   >>> lists[2].append(7)\n   >>> lists\n   [[3], [5], [7]]\n\n3. If *i* or *j* is negative, the index is relative to the end of the\n   string: ``len(s) + i`` or ``len(s) + j`` is substituted.  But note\n   that ``-0`` is still ``0``.\n\n4. The slice of *s* from *i* to *j* is defined as the sequence of\n   items with index *k* such that ``i <= k < j``.  If *i* or *j* is\n   greater than ``len(s)``, use ``len(s)``.  If *i* is omitted or\n   ``None``, use ``0``.  If *j* is omitted or ``None``, use\n   ``len(s)``.  If *i* is greater than or equal to *j*, the slice is\n   empty.\n\n5. The slice of *s* from *i* to *j* with step *k* is defined as the\n   sequence of items with index  ``x = i + n*k`` such that ``0 <= n <\n   (j-i)/k``.  In other words, the indices are ``i``, ``i+k``,\n   ``i+2*k``, ``i+3*k`` and so on, stopping when *j* is reached (but\n   never including *j*).  If *i* or *j* is greater than ``len(s)``,\n   use ``len(s)``.  If *i* or *j* are omitted or ``None``, they become\n   "end" values (which end depends on the sign of *k*).  Note, *k*\n   cannot be zero. If *k* is ``None``, it is treated like ``1``.\n\n6. If *s* and *t* are both strings, some Python implementations such\n   as CPython can usually perform an in-place optimization for\n   assignments of the form ``s=s+t`` or ``s+=t``.  When applicable,\n   this optimization makes quadratic run-time much less likely.  This\n   optimization is both version and implementation dependent. For\n   performance sensitive code, it is preferable to use the\n   ``str.join()`` method which assures consistent linear concatenation\n   performance across versions and implementations.\n\n\nString Methods\n==============\n\nString objects support the methods listed below.  Note that none of\nthese methods take keyword arguments.\n\nIn addition, Python\'s strings support the sequence type methods\ndescribed in the *Sequence Types --- str, bytes, bytearray, list,\ntuple, range* section. To output formatted strings, see the *String\nFormatting* section. Also, see the ``re`` module for string functions\nbased on regular expressions.\n\nstr.capitalize()\n\n   Return a copy of the string with only its first character\n   capitalized.\n\nstr.center(width[, fillchar])\n\n   Return centered in a string of length *width*. Padding is done\n   using the specified *fillchar* (default is a space).\n\nstr.count(sub[, start[, end]])\n\n   Return the number of non-overlapping occurrences of substring *sub*\n   in the range [*start*, *end*].  Optional arguments *start* and\n   *end* are interpreted as in slice notation.\n\nstr.encode([encoding[, errors]])\n\n   Return an encoded version of the string as a bytes object.  Default\n   encoding is the current default string encoding.  *errors* may be\n   given to set a different error handling scheme.  The default for\n   *errors* is ``\'strict\'``, meaning that encoding errors raise a\n   ``UnicodeError``.  Other possible values are ``\'ignore\'``,\n   ``\'replace\'``, ``\'xmlcharrefreplace\'``, ``\'backslashreplace\'`` and\n   any other name registered via ``codecs.register_error()``, see\n   section *Codec Base Classes*. For a list of possible encodings, see\n   section *Standard Encodings*.\n\nstr.endswith(suffix[, start[, end]])\n\n   Return ``True`` if the string ends with the specified *suffix*,\n   otherwise return ``False``.  *suffix* can also be a tuple of\n   suffixes to look for.  With optional *start*, test beginning at\n   that position.  With optional *end*, stop comparing at that\n   position.\n\nstr.expandtabs([tabsize])\n\n   Return a copy of the string where all tab characters are replaced\n   by one or more spaces, depending on the current column and the\n   given tab size.  The column number is reset to zero after each\n   newline occurring in the string. If *tabsize* is not given, a tab\n   size of ``8`` characters is assumed.  This doesn\'t understand other\n   non-printing characters or escape sequences.\n\nstr.find(sub[, start[, end]])\n\n   Return the lowest index in the string where substring *sub* is\n   found, such that *sub* is contained in the range [*start*, *end*].\n   Optional arguments *start* and *end* are interpreted as in slice\n   notation.  Return ``-1`` if *sub* is not found.\n\nstr.format(*args, **kwargs)\n\n   Perform a string formatting operation.  The *format_string*\n   argument can contain literal text or replacement fields delimited\n   by braces ``{}``.  Each replacement field contains either the\n   numeric index of a positional argument, or the name of a keyword\n   argument.  Returns a copy of *format_string* where each replacement\n   field is replaced with the string value of the corresponding\n   argument.\n\n   >>> "The sum of 1 + 2 is {0}".format(1+2)\n   \'The sum of 1 + 2 is 3\'\n\n   See *Format String Syntax* for a description of the various\n   formatting options that can be specified in format strings.\n\nstr.index(sub[, start[, end]])\n\n   Like ``find()``, but raise ``ValueError`` when the substring is not\n   found.\n\nstr.isalnum()\n\n   Return true if all characters in the string are alphanumeric and\n   there is at least one character, false otherwise.\n\nstr.isalpha()\n\n   Return true if all characters in the string are alphabetic and\n   there is at least one character, false otherwise.\n\nstr.isdecimal()\n\n   Return true if all characters in the string are decimal characters\n   and there is at least one character, false otherwise. Decimal\n   characters include digit characters, and all characters that that\n   can be used to form decimal-radix numbers, e.g. U+0660, ARABIC-\n   INDIC DIGIT ZERO.\n\nstr.isdigit()\n\n   Return true if all characters in the string are digits and there is\n   at least one character, false otherwise.\n\nstr.isidentifier()\n\n   Return true if the string is a valid identifier according to the\n   language definition, section *Identifiers and keywords*.\n\nstr.islower()\n\n   Return true if all cased characters in the string are lowercase and\n   there is at least one cased character, false otherwise.\n\nstr.isnumeric()\n\n   Return true if all characters in the string are numeric characters,\n   and there is at least one character, false otherwise. Numeric\n   characters include digit characters, and all characters that have\n   the Unicode numeric value property, e.g. U+2155, VULGAR FRACTION\n   ONE FIFTH.\n\nstr.isprintable()\n\n   Return true if all characters in the string are printable or the\n   string is empty, false otherwise.  Nonprintable characters are\n   those characters defined in the Unicode character database as\n   "Other" or "Separator", excepting the ASCII space (0x20) which is\n   considered printable.  (Note that printable characters in this\n   context are those which should not be escaped when ``repr()`` is\n   invoked on a string.  It has no bearing on the handling of strings\n   written to ``sys.stdout`` or ``sys.stderr``.)\n\nstr.isspace()\n\n   Return true if there are only whitespace characters in the string\n   and there is at least one character, false otherwise.\n\nstr.istitle()\n\n   Return true if the string is a titlecased string and there is at\n   least one character, for example uppercase characters may only\n   follow uncased characters and lowercase characters only cased ones.\n   Return false otherwise.\n\nstr.isupper()\n\n   Return true if all cased characters in the string are uppercase and\n   there is at least one cased character, false otherwise.\n\nstr.join(seq)\n\n   Return a string which is the concatenation of the strings in the\n   sequence *seq*.  A ``TypeError`` will be raised if there are any\n   non-string values in *seq*, including ``bytes`` objects.  The\n   separator between elements is the string providing this method.\n\nstr.ljust(width[, fillchar])\n\n   Return the string left justified in a string of length *width*.\n   Padding is done using the specified *fillchar* (default is a\n   space).  The original string is returned if *width* is less than\n   ``len(s)``.\n\nstr.lower()\n\n   Return a copy of the string converted to lowercase.\n\nstr.lstrip([chars])\n\n   Return a copy of the string with leading characters removed.  The\n   *chars* argument is a string specifying the set of characters to be\n   removed.  If omitted or ``None``, the *chars* argument defaults to\n   removing whitespace.  The *chars* argument is not a prefix; rather,\n   all combinations of its values are stripped:\n\n   >>> \'   spacious   \'.lstrip()\n   \'spacious   \'\n   >>> \'www.example.com\'.lstrip(\'cmowz.\')\n   \'example.com\'\n\nstatic str.maketrans(x[, y[, z]])\n\n   This static method returns a translation table usable for\n   ``str.translate()``.\n\n   If there is only one argument, it must be a dictionary mapping\n   Unicode ordinals (integers) or characters (strings of length 1) to\n   Unicode ordinals, strings (of arbitrary lengths) or None.\n   Character keys will then be converted to ordinals.\n\n   If there are two arguments, they must be strings of equal length,\n   and in the resulting dictionary, each character in x will be mapped\n   to the character at the same position in y.  If there is a third\n   argument, it must be a string, whose characters will be mapped to\n   None in the result.\n\nstr.partition(sep)\n\n   Split the string at the first occurrence of *sep*, and return a\n   3-tuple containing the part before the separator, the separator\n   itself, and the part after the separator.  If the separator is not\n   found, return a 3-tuple containing the string itself, followed by\n   two empty strings.\n\nstr.replace(old, new[, count])\n\n   Return a copy of the string with all occurrences of substring *old*\n   replaced by *new*.  If the optional argument *count* is given, only\n   the first *count* occurrences are replaced.\n\nstr.rfind(sub[, start[, end]])\n\n   Return the highest index in the string where substring *sub* is\n   found, such that *sub* is contained within s[start,end].  Optional\n   arguments *start* and *end* are interpreted as in slice notation.\n   Return ``-1`` on failure.\n\nstr.rindex(sub[, start[, end]])\n\n   Like ``rfind()`` but raises ``ValueError`` when the substring *sub*\n   is not found.\n\nstr.rjust(width[, fillchar])\n\n   Return the string right justified in a string of length *width*.\n   Padding is done using the specified *fillchar* (default is a\n   space). The original string is returned if *width* is less than\n   ``len(s)``.\n\nstr.rpartition(sep)\n\n   Split the string at the last occurrence of *sep*, and return a\n   3-tuple containing the part before the separator, the separator\n   itself, and the part after the separator.  If the separator is not\n   found, return a 3-tuple containing two empty strings, followed by\n   the string itself.\n\nstr.rsplit([sep[, maxsplit]])\n\n   Return a list of the words in the string, using *sep* as the\n   delimiter string. If *maxsplit* is given, at most *maxsplit* splits\n   are done, the *rightmost* ones.  If *sep* is not specified or\n   ``None``, any whitespace string is a separator.  Except for\n   splitting from the right, ``rsplit()`` behaves like ``split()``\n   which is described in detail below.\n\nstr.rstrip([chars])\n\n   Return a copy of the string with trailing characters removed.  The\n   *chars* argument is a string specifying the set of characters to be\n   removed.  If omitted or ``None``, the *chars* argument defaults to\n   removing whitespace.  The *chars* argument is not a suffix; rather,\n   all combinations of its values are stripped:\n\n   >>> \'   spacious   \'.rstrip()\n   \'   spacious\'\n   >>> \'mississippi\'.rstrip(\'ipz\')\n   \'mississ\'\n\nstr.split([sep[, maxsplit]])\n\n   Return a list of the words in the string, using *sep* as the\n   delimiter string.  If *maxsplit* is given, at most *maxsplit*\n   splits are done (thus, the list will have at most ``maxsplit+1``\n   elements).  If *maxsplit* is not specified, then there is no limit\n   on the number of splits (all possible splits are made).\n\n   If *sep* is given, consecutive delimiters are not grouped together\n   and are deemed to delimit empty strings (for example,\n   ``\'1,,2\'.split(\',\')`` returns ``[\'1\', \'\', \'2\']``).  The *sep*\n   argument may consist of multiple characters (for example,\n   ``\'1<>2<>3\'.split(\'<>\')`` returns ``[\'1\', \'2\', \'3\']``). Splitting\n   an empty string with a specified separator returns ``[\'\']``.\n\n   If *sep* is not specified or is ``None``, a different splitting\n   algorithm is applied: runs of consecutive whitespace are regarded\n   as a single separator, and the result will contain no empty strings\n   at the start or end if the string has leading or trailing\n   whitespace.  Consequently, splitting an empty string or a string\n   consisting of just whitespace with a ``None`` separator returns\n   ``[]``.\n\n   For example, ``\' 1  2   3  \'.split()`` returns ``[\'1\', \'2\', \'3\']``,\n   and ``\'  1  2   3  \'.split(None, 1)`` returns ``[\'1\', \'2   3  \']``.\n\nstr.splitlines([keepends])\n\n   Return a list of the lines in the string, breaking at line\n   boundaries.  Line breaks are not included in the resulting list\n   unless *keepends* is given and true.\n\nstr.startswith(prefix[, start[, end]])\n\n   Return ``True`` if string starts with the *prefix*, otherwise\n   return ``False``. *prefix* can also be a tuple of prefixes to look\n   for.  With optional *start*, test string beginning at that\n   position.  With optional *end*, stop comparing string at that\n   position.\n\nstr.strip([chars])\n\n   Return a copy of the string with the leading and trailing\n   characters removed. The *chars* argument is a string specifying the\n   set of characters to be removed. If omitted or ``None``, the\n   *chars* argument defaults to removing whitespace. The *chars*\n   argument is not a prefix or suffix; rather, all combinations of its\n   values are stripped:\n\n   >>> \'   spacious   \'.strip()\n   \'spacious\'\n   >>> \'www.example.com\'.strip(\'cmowz.\')\n   \'example\'\n\nstr.swapcase()\n\n   Return a copy of the string with uppercase characters converted to\n   lowercase and vice versa.\n\nstr.title()\n\n   Return a titlecased version of the string: words start with\n   uppercase characters, all remaining cased characters are lowercase.\n\nstr.translate(map)\n\n   Return a copy of the *s* where all characters have been mapped\n   through the *map* which must be a dictionary of Unicode ordinals\n   (integers) to Unicode ordinals, strings or ``None``.  Unmapped\n   characters are left untouched. Characters mapped to ``None`` are\n   deleted.\n\n   You can use ``str.maketrans()`` to create a translation map from\n   character-to-character mappings in different formats.\n\n   Note: An even more flexible approach is to create a custom character\n     mapping codec using the ``codecs`` module (see\n     ``encodings.cp1251`` for an example).\n\nstr.upper()\n\n   Return a copy of the string converted to uppercase.\n\nstr.zfill(width)\n\n   Return the numeric string left filled with zeros in a string of\n   length *width*.  A sign prefix is handled correctly.  The original\n   string is returned if *width* is less than ``len(s)``.\n\n\nOld String Formatting Operations\n================================\n\nNote: The formatting operations described here are obsolete and may go\n  away in future versions of Python.  Use the new *String Formatting*\n  in new code.\n\nString objects have one unique built-in operation: the ``%`` operator\n(modulo). This is also known as the string *formatting* or\n*interpolation* operator. Given ``format % values`` (where *format* is\na string), ``%`` conversion specifications in *format* are replaced\nwith zero or more elements of *values*. The effect is similar to the\nusing ``sprintf()`` in the C language.\n\nIf *format* requires a single argument, *values* may be a single non-\ntuple object. [4]  Otherwise, *values* must be a tuple with exactly\nthe number of items specified by the format string, or a single\nmapping object (for example, a dictionary).\n\nA conversion specifier contains two or more characters and has the\nfollowing components, which must occur in this order:\n\n1. The ``\'%\'`` character, which marks the start of the specifier.\n\n2. Mapping key (optional), consisting of a parenthesised sequence of\n   characters (for example, ``(somename)``).\n\n3. Conversion flags (optional), which affect the result of some\n   conversion types.\n\n4. Minimum field width (optional).  If specified as an ``\'*\'``\n   (asterisk), the actual width is read from the next element of the\n   tuple in *values*, and the object to convert comes after the\n   minimum field width and optional precision.\n\n5. Precision (optional), given as a ``\'.\'`` (dot) followed by the\n   precision.  If specified as ``\'*\'`` (an asterisk), the actual width\n   is read from the next element of the tuple in *values*, and the\n   value to convert comes after the precision.\n\n6. Length modifier (optional).\n\n7. Conversion type.\n\nWhen the right argument is a dictionary (or other mapping type), then\nthe formats in the string *must* include a parenthesised mapping key\ninto that dictionary inserted immediately after the ``\'%\'`` character.\nThe mapping key selects the value to be formatted from the mapping.\nFor example:\n\n>>> print(\'%(language)s has %(#)03d quote types.\' % \\\n...       {\'language\': "Python", "#": 2})\nPython has 002 quote types.\n\nIn this case no ``*`` specifiers may occur in a format (since they\nrequire a sequential parameter list).\n\nThe conversion flag characters are:\n\n+-----------+-----------------------------------------------------------------------+\n| Flag      | Meaning                                                               |\n+===========+=======================================================================+\n| ``\'#\'``   | The value conversion will use the "alternate form" (where defined     |\n|           | below).                                                               |\n+-----------+-----------------------------------------------------------------------+\n| ``\'0\'``   | The conversion will be zero padded for numeric values.                |\n+-----------+-----------------------------------------------------------------------+\n| ``\'-\'``   | The converted value is left adjusted (overrides the ``\'0\'``           |\n|           | conversion if both are given).                                        |\n+-----------+-----------------------------------------------------------------------+\n| ``\' \'``   | (a space) A blank should be left before a positive number (or empty   |\n|           | string) produced by a signed conversion.                              |\n+-----------+-----------------------------------------------------------------------+\n| ``\'+\'``   | A sign character (``\'+\'`` or ``\'-\'``) will precede the conversion     |\n|           | (overrides a "space" flag).                                           |\n+-----------+-----------------------------------------------------------------------+\n\nA length modifier (``h``, ``l``, or ``L``) may be present, but is\nignored as it is not necessary for Python -- so e.g. ``%ld`` is\nidentical to ``%d``.\n\nThe conversion types are:\n\n+--------------+-------------------------------------------------------+---------+\n| Conversion   | Meaning                                               | Notes   |\n+==============+=======================================================+=========+\n| ``\'d\'``      | Signed integer decimal.                               |         |\n+--------------+-------------------------------------------------------+---------+\n| ``\'i\'``      | Signed integer decimal.                               |         |\n+--------------+-------------------------------------------------------+---------+\n| ``\'o\'``      | Signed octal value.                                   | (1)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'u\'``      | Obsolete type -- it is identical to ``\'d\'``.          | (7)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'x\'``      | Signed hexadecimal (lowercase).                       | (2)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'X\'``      | Signed hexadecimal (uppercase).                       | (2)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'e\'``      | Floating point exponential format (lowercase).        | (3)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'E\'``      | Floating point exponential format (uppercase).        | (3)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'f\'``      | Floating point decimal format.                        | (3)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'F\'``      | Floating point decimal format.                        | (3)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'g\'``      | Floating point format. Uses lowercase exponential     | (4)     |\n|              | format if exponent is less than -4 or not less than   |         |\n|              | precision, decimal format otherwise.                  |         |\n+--------------+-------------------------------------------------------+---------+\n| ``\'G\'``      | Floating point format. Uses uppercase exponential     | (4)     |\n|              | format if exponent is less than -4 or not less than   |         |\n|              | precision, decimal format otherwise.                  |         |\n+--------------+-------------------------------------------------------+---------+\n| ``\'c\'``      | Single character (accepts integer or single character |         |\n|              | string).                                              |         |\n+--------------+-------------------------------------------------------+---------+\n| ``\'r\'``      | String (converts any python object using ``repr()``). | (5)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'s\'``      | String (converts any python object using ``str()``).  |         |\n+--------------+-------------------------------------------------------+---------+\n| ``\'%\'``      | No argument is converted, results in a ``\'%\'``        |         |\n|              | character in the result.                              |         |\n+--------------+-------------------------------------------------------+---------+\n\nNotes:\n\n1. The alternate form causes a leading zero (``\'0\'``) to be inserted\n   between left-hand padding and the formatting of the number if the\n   leading character of the result is not already a zero.\n\n2. The alternate form causes a leading ``\'0x\'`` or ``\'0X\'`` (depending\n   on whether the ``\'x\'`` or ``\'X\'`` format was used) to be inserted\n   between left-hand padding and the formatting of the number if the\n   leading character of the result is not already a zero.\n\n3. The alternate form causes the result to always contain a decimal\n   point, even if no digits follow it.\n\n   The precision determines the number of digits after the decimal\n   point and defaults to 6.\n\n4. The alternate form causes the result to always contain a decimal\n   point, and trailing zeroes are not removed as they would otherwise\n   be.\n\n   The precision determines the number of significant digits before\n   and after the decimal point and defaults to 6.\n\n5. The precision determines the maximal number of characters used.\n\n1. See **PEP 237**.\n\nSince Python strings have an explicit length, ``%s`` conversions do\nnot assume that ``\'\\0\'`` is the end of the string.\n\nChanged in version 3.1: ``%f`` conversions for numbers whose absolute\nvalue is over 1e50 are no longer replaced by ``%g`` conversions.\n\nAdditional string operations are defined in standard modules\n``string`` and ``re``.\n\n\nRange Type\n==========\n\nThe ``range`` type is an immutable sequence which is commonly used for\nlooping.  The advantage of the ``range`` type is that an ``range``\nobject will always take the same amount of memory, no matter the size\nof the range it represents.  There are no consistent performance\nadvantages.\n\nRange objects have very little behavior: they only support indexing,\niteration, and the ``len()`` function.\n\n\nMutable Sequence Types\n======================\n\nList and bytearray objects support additional operations that allow\nin-place modification of the object.  Other mutable sequence types\n(when added to the language) should also support these operations.\nStrings and tuples are immutable sequence types: such objects cannot\nbe modified once created. The following operations are defined on\nmutable sequence types (where *x* is an arbitrary object).\n\nNote that while lists allow their items to be of any type, bytearray\nobject "items" are all integers in the range 0 <= x < 256.\n\n+--------------------------------+----------------------------------+-----------------------+\n| Operation                      | Result                           | Notes                 |\n+================================+==================================+=======================+\n| ``s[i] = x``                   | item *i* of *s* is replaced by   |                       |\n|                                | *x*                              |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s[i:j] = t``                 | slice of *s* from *i* to *j* is  |                       |\n|                                | replaced by the contents of the  |                       |\n|                                | iterable *t*                     |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``del s[i:j]``                 | same as ``s[i:j] = []``          |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s[i:j:k] = t``               | the elements of ``s[i:j:k]`` are | (1)                   |\n|                                | replaced by those of *t*         |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``del s[i:j:k]``               | removes the elements of          |                       |\n|                                | ``s[i:j:k]`` from the list       |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.append(x)``                | same as ``s[len(s):len(s)] =     |                       |\n|                                | [x]``                            |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.extend(x)``                | same as ``s[len(s):len(s)] = x`` | (2)                   |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.count(x)``                 | return number of *i*\'s for which |                       |\n|                                | ``s[i] == x``                    |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.index(x[, i[, j]])``       | return smallest *k* such that    | (3)                   |\n|                                | ``s[k] == x`` and ``i <= k < j`` |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.insert(i, x)``             | same as ``s[i:i] = [x]``         | (4)                   |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.pop([i])``                 | same as ``x = s[i]; del s[i];    | (5)                   |\n|                                | return x``                       |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.remove(x)``                | same as ``del s[s.index(x)]``    | (3)                   |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.reverse()``                | reverses the items of *s* in     | (6)                   |\n|                                | place                            |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.sort([key[, reverse]])``   | sort the items of *s* in place   | (6), (7), (8)         |\n+--------------------------------+----------------------------------+-----------------------+\n\nNotes:\n\n1. *t* must have the same length as the slice it is replacing.\n\n2. *x* can be any iterable object.\n\n3. Raises ``ValueError`` when *x* is not found in *s*. When a negative\n   index is passed as the second or third parameter to the ``index()``\n   method, the sequence length is added, as for slice indices.  If it\n   is still negative, it is truncated to zero, as for slice indices.\n\n4. When a negative index is passed as the first parameter to the\n   ``insert()`` method, the sequence length is added, as for slice\n   indices.  If it is still negative, it is truncated to zero, as for\n   slice indices.\n\n5. The optional argument *i* defaults to ``-1``, so that by default\n   the last item is removed and returned.\n\n6. The ``sort()`` and ``reverse()`` methods modify the sequence in\n   place for economy of space when sorting or reversing a large\n   sequence.  To remind you that they operate by side effect, they\n   don\'t return the sorted or reversed sequence.\n\n7. The ``sort()`` method takes optional arguments for controlling the\n   comparisons.  Each must be specified as a keyword argument.\n\n   *key* specifies a function of one argument that is used to extract\n   a comparison key from each list element: ``key=str.lower``.  The\n   default value is ``None``.\n\n   *reverse* is a boolean value.  If set to ``True``, then the list\n   elements are sorted as if each comparison were reversed.\n\n   The ``sort()`` method is guaranteed to be stable.  A sort is stable\n   if it guarantees not to change the relative order of elements that\n   compare equal --- this is helpful for sorting in multiple passes\n   (for example, sort by department, then by salary grade).\n\n   While a list is being sorted, the effect of attempting to mutate,\n   or even inspect, the list is undefined.  The C implementation makes\n   the list appear empty for the duration, and raises ``ValueError``\n   if it can detect that the list has been mutated during a sort.\n\n8. ``sort()`` is not supported by ``bytearray`` objects.\n\n\nBytes and Byte Array Methods\n============================\n\nBytes and bytearray objects, being "strings of bytes", have all\nmethods found on strings, with the exception of ``encode()``,\n``format()`` and ``isidentifier()``, which do not make sense with\nthese types.  For converting the objects to strings, they have a\n``decode()`` method.\n\nWherever one of these methods needs to interpret the bytes as\ncharacters (e.g. the ``is...()`` methods), the ASCII character set is\nassumed.\n\nNote: The methods on bytes and bytearray objects don\'t accept strings as\n  their arguments, just as the methods on strings don\'t accept bytes\n  as their arguments.  For example, you have to write\n\n     a = "abc"\n     b = a.replace("a", "f")\n\n  and\n\n     a = b"abc"\n     b = a.replace(b"a", b"f")\n\nbytes.decode([encoding[, errors]])\nbytearray.decode([encoding[, errors]])\n\n   Return a string decoded from the given bytes.  Default encoding is\n   the current default string encoding.  *errors* may be given to set\n   a different error handling scheme.  The default for *errors* is\n   ``\'strict\'``, meaning that encoding errors raise a\n   ``UnicodeError``.  Other possible values are ``\'ignore\'``,\n   ``\'replace\'`` and any other name registered via\n   ``codecs.register_error()``, see section *Codec Base Classes*. For\n   a list of possible encodings, see section *Standard Encodings*.\n\nThe bytes and bytearray types have an additional class method:\n\nclassmethod bytes.fromhex(string)\nclassmethod bytearray.fromhex(string)\n\n   This ``bytes`` class method returns a bytes or bytearray object,\n   decoding the given string object.  The string must contain two\n   hexadecimal digits per byte, spaces are ignored.\n\n   >>> bytes.fromhex(\'f0 f1f2  \')\n   b\'\\xf0\\xf1\\xf2\'\n\nThe maketrans and translate methods differ in semantics from the\nversions available on strings:\n\nbytes.translate(table[, delete])\nbytearray.translate(table[, delete])\n\n   Return a copy of the bytes or bytearray object where all bytes\n   occurring in the optional argument *delete* are removed, and the\n   remaining bytes have been mapped through the given translation\n   table, which must be a bytes object of length 256.\n\n   You can use the ``bytes.maketrans()`` method to create a\n   translation table.\n\n   Set the *table* argument to ``None`` for translations that only\n   delete characters:\n\n      >>> b\'read this short text\'.translate(None, b\'aeiou\')\n      b\'rd ths shrt txt\'\n\nstatic bytes.maketrans(from, to)\nstatic bytearray.maketrans(from, to)\n\n   This static method returns a translation table usable for\n   ``bytes.translate()`` that will map each character in *from* into\n   the character at the same position in *to*; *from* and *to* must be\n   bytes objects and have the same length.\n\n   New in version 3.1.\n',
- 'typesseq-mutable': '\nMutable Sequence Types\n**********************\n\nList and bytearray objects support additional operations that allow\nin-place modification of the object.  Other mutable sequence types\n(when added to the language) should also support these operations.\nStrings and tuples are immutable sequence types: such objects cannot\nbe modified once created. The following operations are defined on\nmutable sequence types (where *x* is an arbitrary object).\n\nNote that while lists allow their items to be of any type, bytearray\nobject "items" are all integers in the range 0 <= x < 256.\n\n+--------------------------------+----------------------------------+-----------------------+\n| Operation                      | Result                           | Notes                 |\n+================================+==================================+=======================+\n| ``s[i] = x``                   | item *i* of *s* is replaced by   |                       |\n|                                | *x*                              |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s[i:j] = t``                 | slice of *s* from *i* to *j* is  |                       |\n|                                | replaced by the contents of the  |                       |\n|                                | iterable *t*                     |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``del s[i:j]``                 | same as ``s[i:j] = []``          |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s[i:j:k] = t``               | the elements of ``s[i:j:k]`` are | (1)                   |\n|                                | replaced by those of *t*         |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``del s[i:j:k]``               | removes the elements of          |                       |\n|                                | ``s[i:j:k]`` from the list       |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.append(x)``                | same as ``s[len(s):len(s)] =     |                       |\n|                                | [x]``                            |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.extend(x)``                | same as ``s[len(s):len(s)] = x`` | (2)                   |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.count(x)``                 | return number of *i*\'s for which |                       |\n|                                | ``s[i] == x``                    |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.index(x[, i[, j]])``       | return smallest *k* such that    | (3)                   |\n|                                | ``s[k] == x`` and ``i <= k < j`` |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.insert(i, x)``             | same as ``s[i:i] = [x]``         | (4)                   |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.pop([i])``                 | same as ``x = s[i]; del s[i];    | (5)                   |\n|                                | return x``                       |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.remove(x)``                | same as ``del s[s.index(x)]``    | (3)                   |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.reverse()``                | reverses the items of *s* in     | (6)                   |\n|                                | place                            |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.sort([key[, reverse]])``   | sort the items of *s* in place   | (6), (7), (8)         |\n+--------------------------------+----------------------------------+-----------------------+\n\nNotes:\n\n1. *t* must have the same length as the slice it is replacing.\n\n2. *x* can be any iterable object.\n\n3. Raises ``ValueError`` when *x* is not found in *s*. When a negative\n   index is passed as the second or third parameter to the ``index()``\n   method, the sequence length is added, as for slice indices.  If it\n   is still negative, it is truncated to zero, as for slice indices.\n\n4. When a negative index is passed as the first parameter to the\n   ``insert()`` method, the sequence length is added, as for slice\n   indices.  If it is still negative, it is truncated to zero, as for\n   slice indices.\n\n5. The optional argument *i* defaults to ``-1``, so that by default\n   the last item is removed and returned.\n\n6. The ``sort()`` and ``reverse()`` methods modify the sequence in\n   place for economy of space when sorting or reversing a large\n   sequence.  To remind you that they operate by side effect, they\n   don\'t return the sorted or reversed sequence.\n\n7. The ``sort()`` method takes optional arguments for controlling the\n   comparisons.  Each must be specified as a keyword argument.\n\n   *key* specifies a function of one argument that is used to extract\n   a comparison key from each list element: ``key=str.lower``.  The\n   default value is ``None``.\n\n   *reverse* is a boolean value.  If set to ``True``, then the list\n   elements are sorted as if each comparison were reversed.\n\n   The ``sort()`` method is guaranteed to be stable.  A sort is stable\n   if it guarantees not to change the relative order of elements that\n   compare equal --- this is helpful for sorting in multiple passes\n   (for example, sort by department, then by salary grade).\n\n   While a list is being sorted, the effect of attempting to mutate,\n   or even inspect, the list is undefined.  The C implementation makes\n   the list appear empty for the duration, and raises ``ValueError``\n   if it can detect that the list has been mutated during a sort.\n\n8. ``sort()`` is not supported by ``bytearray`` objects.\n',
+ 'typesseq': '\nSequence Types --- ``str``, ``bytes``, ``bytearray``, ``list``, ``tuple``, ``range``\n************************************************************************************\n\nThere are six sequence types: strings, byte sequences (``bytes``\nobjects), byte arrays (``bytearray`` objects), lists, tuples, and\nrange objects.  For other containers see the built in ``dict`` and\n``set`` classes, and the ``collections`` module.\n\nStrings contain Unicode characters.  Their literals are written in\nsingle or double quotes: ``\'xyzzy\'``, ``"frobozz"``.  See *String and\nBytes literals* for more about string literals.  In addition to the\nfunctionality described here, there are also string-specific methods\ndescribed in the *String Methods* section.\n\nBytes and bytearray objects contain single bytes -- the former is\nimmutable while the latter is a mutable sequence.  Bytes objects can\nbe constructed the constructor, ``bytes()``, and from literals; use a\n``b`` prefix with normal string syntax: ``b\'xyzzy\'``.  To construct\nbyte arrays, use the ``bytearray()`` function.\n\nWarning: While string objects are sequences of characters (represented by\n  strings of length 1), bytes and bytearray objects are sequences of\n  *integers* (between 0 and 255), representing the ASCII value of\n  single bytes.  That means that for a bytes or bytearray object *b*,\n  ``b[0]`` will be an integer, while ``b[0:1]`` will be a bytes or\n  bytearray object of length 1.  The representation of bytes objects\n  uses the literal format (``b\'...\'``) since it is generally more\n  useful than e.g. ``bytes([50, 19, 100])``.  You can always convert a\n  bytes object into a list of integers using ``list(b)``.Also, while\n  in previous Python versions, byte strings and Unicode strings could\n  be exchanged for each other rather freely (barring encoding issues),\n  strings and bytes are now completely separate concepts.  There\'s no\n  implicit en-/decoding if you pass an object of the wrong type.  A\n  string always compares unequal to a bytes or bytearray object.\n\nLists are constructed with square brackets, separating items with\ncommas: ``[a, b, c]``.  Tuples are constructed by the comma operator\n(not within square brackets), with or without enclosing parentheses,\nbut an empty tuple must have the enclosing parentheses, such as ``a,\nb, c`` or ``()``.  A single item tuple must have a trailing comma,\nsuch as ``(d,)``.\n\nObjects of type range are created using the ``range()`` function.\nThey don\'t support slicing, concatenation or repetition, and using\n``in``, ``not in``, ``min()`` or ``max()`` on them is inefficient.\n\nMost sequence types support the following operations.  The ``in`` and\n``not in`` operations have the same priorities as the comparison\noperations.  The ``+`` and ``*`` operations have the same priority as\nthe corresponding numeric operations. [3] Additional methods are\nprovided for *Mutable Sequence Types*.\n\nThis table lists the sequence operations sorted in ascending priority\n(operations in the same box have the same priority).  In the table,\n*s* and *t* are sequences of the same type; *n*, *i* and *j* are\nintegers:\n\n+--------------------+----------------------------------+------------+\n| Operation          | Result                           | Notes      |\n+====================+==================================+============+\n| ``x in s``         | ``True`` if an item of *s* is    | (1)        |\n|                    | equal to *x*, else ``False``     |            |\n+--------------------+----------------------------------+------------+\n| ``x not in s``     | ``False`` if an item of *s* is   | (1)        |\n|                    | equal to *x*, else ``True``      |            |\n+--------------------+----------------------------------+------------+\n| ``s + t``          | the concatenation of *s* and *t* | (6)        |\n+--------------------+----------------------------------+------------+\n| ``s * n, n * s``   | *n* shallow copies of *s*        | (2)        |\n|                    | concatenated                     |            |\n+--------------------+----------------------------------+------------+\n| ``s[i]``           | *i*\'th item of *s*, origin 0     | (3)        |\n+--------------------+----------------------------------+------------+\n| ``s[i:j]``         | slice of *s* from *i* to *j*     | (3)(4)     |\n+--------------------+----------------------------------+------------+\n| ``s[i:j:k]``       | slice of *s* from *i* to *j*     | (3)(5)     |\n|                    | with step *k*                    |            |\n+--------------------+----------------------------------+------------+\n| ``len(s)``         | length of *s*                    |            |\n+--------------------+----------------------------------+------------+\n| ``min(s)``         | smallest item of *s*             |            |\n+--------------------+----------------------------------+------------+\n| ``max(s)``         | largest item of *s*              |            |\n+--------------------+----------------------------------+------------+\n\nSequence types also support comparisons.  In particular, tuples and\nlists are compared lexicographically by comparing corresponding\nelements.  This means that to compare equal, every element must\ncompare equal and the two sequences must be of the same type and have\nthe same length.  (For full details see *Comparisons* in the language\nreference.)\n\nNotes:\n\n1. When *s* is a string object, the ``in`` and ``not in`` operations\n   act like a substring test.\n\n2. Values of *n* less than ``0`` are treated as ``0`` (which yields an\n   empty sequence of the same type as *s*).  Note also that the copies\n   are shallow; nested structures are not copied.  This often haunts\n   new Python programmers; consider:\n\n   >>> lists = [[]] * 3\n   >>> lists\n   [[], [], []]\n   >>> lists[0].append(3)\n   >>> lists\n   [[3], [3], [3]]\n\n   What has happened is that ``[[]]`` is a one-element list containing\n   an empty list, so all three elements of ``[[]] * 3`` are (pointers\n   to) this single empty list.  Modifying any of the elements of\n   ``lists`` modifies this single list. You can create a list of\n   different lists this way:\n\n   >>> lists = [[] for i in range(3)]\n   >>> lists[0].append(3)\n   >>> lists[1].append(5)\n   >>> lists[2].append(7)\n   >>> lists\n   [[3], [5], [7]]\n\n3. If *i* or *j* is negative, the index is relative to the end of the\n   string: ``len(s) + i`` or ``len(s) + j`` is substituted.  But note\n   that ``-0`` is still ``0``.\n\n4. The slice of *s* from *i* to *j* is defined as the sequence of\n   items with index *k* such that ``i <= k < j``.  If *i* or *j* is\n   greater than ``len(s)``, use ``len(s)``.  If *i* is omitted or\n   ``None``, use ``0``.  If *j* is omitted or ``None``, use\n   ``len(s)``.  If *i* is greater than or equal to *j*, the slice is\n   empty.\n\n5. The slice of *s* from *i* to *j* with step *k* is defined as the\n   sequence of items with index  ``x = i + n*k`` such that ``0 <= n <\n   (j-i)/k``.  In other words, the indices are ``i``, ``i+k``,\n   ``i+2*k``, ``i+3*k`` and so on, stopping when *j* is reached (but\n   never including *j*).  If *i* or *j* is greater than ``len(s)``,\n   use ``len(s)``.  If *i* or *j* are omitted or ``None``, they become\n   "end" values (which end depends on the sign of *k*).  Note, *k*\n   cannot be zero. If *k* is ``None``, it is treated like ``1``.\n\n6. **CPython implementation detail:** If *s* and *t* are both strings,\n   some Python implementations such as CPython can usually perform an\n   in-place optimization for assignments of the form ``s = s + t`` or\n   ``s += t``.  When applicable, this optimization makes quadratic\n   run-time much less likely.  This optimization is both version and\n   implementation dependent.  For performance sensitive code, it is\n   preferable to use the ``str.join()`` method which assures\n   consistent linear concatenation performance across versions and\n   implementations.\n\n\nString Methods\n==============\n\nString objects support the methods listed below.\n\nIn addition, Python\'s strings support the sequence type methods\ndescribed in the *Sequence Types --- str, bytes, bytearray, list,\ntuple, range* section. To output formatted strings, see the *String\nFormatting* section. Also, see the ``re`` module for string functions\nbased on regular expressions.\n\nstr.capitalize()\n\n   Return a copy of the string with its first character capitalized\n   and the rest lowercased.\n\nstr.center(width[, fillchar])\n\n   Return centered in a string of length *width*. Padding is done\n   using the specified *fillchar* (default is a space).\n\nstr.count(sub[, start[, end]])\n\n   Return the number of non-overlapping occurrences of substring *sub*\n   in the range [*start*, *end*].  Optional arguments *start* and\n   *end* are interpreted as in slice notation.\n\nstr.encode(encoding=sys.getdefaultencoding(), errors="strict")\n\n   Return an encoded version of the string as a bytes object.  Default\n   encoding is the current default string encoding.  *errors* may be\n   given to set a different error handling scheme.  The default for\n   *errors* is ``\'strict\'``, meaning that encoding errors raise a\n   ``UnicodeError``.  Other possible values are ``\'ignore\'``,\n   ``\'replace\'``, ``\'xmlcharrefreplace\'``, ``\'backslashreplace\'`` and\n   any other name registered via ``codecs.register_error()``, see\n   section *Codec Base Classes*. For a list of possible encodings, see\n   section *Standard Encodings*.\n\n   Changed in version 3.1: Added support for keyword arguments added.\n\nstr.endswith(suffix[, start[, end]])\n\n   Return ``True`` if the string ends with the specified *suffix*,\n   otherwise return ``False``.  *suffix* can also be a tuple of\n   suffixes to look for.  With optional *start*, test beginning at\n   that position.  With optional *end*, stop comparing at that\n   position.\n\nstr.expandtabs([tabsize])\n\n   Return a copy of the string where all tab characters are replaced\n   by one or more spaces, depending on the current column and the\n   given tab size.  The column number is reset to zero after each\n   newline occurring in the string. If *tabsize* is not given, a tab\n   size of ``8`` characters is assumed.  This doesn\'t understand other\n   non-printing characters or escape sequences.\n\nstr.find(sub[, start[, end]])\n\n   Return the lowest index in the string where substring *sub* is\n   found, such that *sub* is contained in the slice ``s[start:end]``.\n   Optional arguments *start* and *end* are interpreted as in slice\n   notation.  Return ``-1`` if *sub* is not found.\n\nstr.format(*args, **kwargs)\n\n   Perform a string formatting operation.  The string on which this\n   method is called can contain literal text or replacement fields\n   delimited by braces ``{}``.  Each replacement field contains either\n   the numeric index of a positional argument, or the name of a\n   keyword argument.  Returns a copy of the string where each\n   replacement field is replaced with the string value of the\n   corresponding argument.\n\n   >>> "The sum of 1 + 2 is {0}".format(1+2)\n   \'The sum of 1 + 2 is 3\'\n\n   See *Format String Syntax* for a description of the various\n   formatting options that can be specified in format strings.\n\nstr.index(sub[, start[, end]])\n\n   Like ``find()``, but raise ``ValueError`` when the substring is not\n   found.\n\nstr.isalnum()\n\n   Return true if all characters in the string are alphanumeric and\n   there is at least one character, false otherwise.\n\nstr.isalpha()\n\n   Return true if all characters in the string are alphabetic and\n   there is at least one character, false otherwise.\n\nstr.isdecimal()\n\n   Return true if all characters in the string are decimal characters\n   and there is at least one character, false otherwise. Decimal\n   characters include digit characters, and all characters that that\n   can be used to form decimal-radix numbers, e.g. U+0660, ARABIC-\n   INDIC DIGIT ZERO.\n\nstr.isdigit()\n\n   Return true if all characters in the string are digits and there is\n   at least one character, false otherwise.\n\nstr.isidentifier()\n\n   Return true if the string is a valid identifier according to the\n   language definition, section *Identifiers and keywords*.\n\nstr.islower()\n\n   Return true if all cased characters in the string are lowercase and\n   there is at least one cased character, false otherwise.\n\nstr.isnumeric()\n\n   Return true if all characters in the string are numeric characters,\n   and there is at least one character, false otherwise. Numeric\n   characters include digit characters, and all characters that have\n   the Unicode numeric value property, e.g. U+2155, VULGAR FRACTION\n   ONE FIFTH.\n\nstr.isprintable()\n\n   Return true if all characters in the string are printable or the\n   string is empty, false otherwise.  Nonprintable characters are\n   those characters defined in the Unicode character database as\n   "Other" or "Separator", excepting the ASCII space (0x20) which is\n   considered printable.  (Note that printable characters in this\n   context are those which should not be escaped when ``repr()`` is\n   invoked on a string.  It has no bearing on the handling of strings\n   written to ``sys.stdout`` or ``sys.stderr``.)\n\nstr.isspace()\n\n   Return true if there are only whitespace characters in the string\n   and there is at least one character, false otherwise.\n\nstr.istitle()\n\n   Return true if the string is a titlecased string and there is at\n   least one character, for example uppercase characters may only\n   follow uncased characters and lowercase characters only cased ones.\n   Return false otherwise.\n\nstr.isupper()\n\n   Return true if all cased characters in the string are uppercase and\n   there is at least one cased character, false otherwise.\n\nstr.join(iterable)\n\n   Return a string which is the concatenation of the strings in the\n   *iterable* *iterable*.  A ``TypeError`` will be raised if there are\n   any non-string values in *seq*, including ``bytes`` objects.  The\n   separator between elements is the string providing this method.\n\nstr.ljust(width[, fillchar])\n\n   Return the string left justified in a string of length *width*.\n   Padding is done using the specified *fillchar* (default is a\n   space).  The original string is returned if *width* is less than\n   ``len(s)``.\n\nstr.lower()\n\n   Return a copy of the string converted to lowercase.\n\nstr.lstrip([chars])\n\n   Return a copy of the string with leading characters removed.  The\n   *chars* argument is a string specifying the set of characters to be\n   removed.  If omitted or ``None``, the *chars* argument defaults to\n   removing whitespace.  The *chars* argument is not a prefix; rather,\n   all combinations of its values are stripped:\n\n   >>> \'   spacious   \'.lstrip()\n   \'spacious   \'\n   >>> \'www.example.com\'.lstrip(\'cmowz.\')\n   \'example.com\'\n\nstatic str.maketrans(x[, y[, z]])\n\n   This static method returns a translation table usable for\n   ``str.translate()``.\n\n   If there is only one argument, it must be a dictionary mapping\n   Unicode ordinals (integers) or characters (strings of length 1) to\n   Unicode ordinals, strings (of arbitrary lengths) or None.\n   Character keys will then be converted to ordinals.\n\n   If there are two arguments, they must be strings of equal length,\n   and in the resulting dictionary, each character in x will be mapped\n   to the character at the same position in y.  If there is a third\n   argument, it must be a string, whose characters will be mapped to\n   None in the result.\n\nstr.partition(sep)\n\n   Split the string at the first occurrence of *sep*, and return a\n   3-tuple containing the part before the separator, the separator\n   itself, and the part after the separator.  If the separator is not\n   found, return a 3-tuple containing the string itself, followed by\n   two empty strings.\n\nstr.replace(old, new[, count])\n\n   Return a copy of the string with all occurrences of substring *old*\n   replaced by *new*.  If the optional argument *count* is given, only\n   the first *count* occurrences are replaced.\n\nstr.rfind(sub[, start[, end]])\n\n   Return the highest index in the string where substring *sub* is\n   found, such that *sub* is contained within ``s[start:end]``.\n   Optional arguments *start* and *end* are interpreted as in slice\n   notation.  Return ``-1`` on failure.\n\nstr.rindex(sub[, start[, end]])\n\n   Like ``rfind()`` but raises ``ValueError`` when the substring *sub*\n   is not found.\n\nstr.rjust(width[, fillchar])\n\n   Return the string right justified in a string of length *width*.\n   Padding is done using the specified *fillchar* (default is a\n   space). The original string is returned if *width* is less than\n   ``len(s)``.\n\nstr.rpartition(sep)\n\n   Split the string at the last occurrence of *sep*, and return a\n   3-tuple containing the part before the separator, the separator\n   itself, and the part after the separator.  If the separator is not\n   found, return a 3-tuple containing two empty strings, followed by\n   the string itself.\n\nstr.rsplit([sep[, maxsplit]])\n\n   Return a list of the words in the string, using *sep* as the\n   delimiter string. If *maxsplit* is given, at most *maxsplit* splits\n   are done, the *rightmost* ones.  If *sep* is not specified or\n   ``None``, any whitespace string is a separator.  Except for\n   splitting from the right, ``rsplit()`` behaves like ``split()``\n   which is described in detail below.\n\nstr.rstrip([chars])\n\n   Return a copy of the string with trailing characters removed.  The\n   *chars* argument is a string specifying the set of characters to be\n   removed.  If omitted or ``None``, the *chars* argument defaults to\n   removing whitespace.  The *chars* argument is not a suffix; rather,\n   all combinations of its values are stripped:\n\n   >>> \'   spacious   \'.rstrip()\n   \'   spacious\'\n   >>> \'mississippi\'.rstrip(\'ipz\')\n   \'mississ\'\n\nstr.split([sep[, maxsplit]])\n\n   Return a list of the words in the string, using *sep* as the\n   delimiter string.  If *maxsplit* is given, at most *maxsplit*\n   splits are done (thus, the list will have at most ``maxsplit+1``\n   elements).  If *maxsplit* is not specified, then there is no limit\n   on the number of splits (all possible splits are made).\n\n   If *sep* is given, consecutive delimiters are not grouped together\n   and are deemed to delimit empty strings (for example,\n   ``\'1,,2\'.split(\',\')`` returns ``[\'1\', \'\', \'2\']``).  The *sep*\n   argument may consist of multiple characters (for example,\n   ``\'1<>2<>3\'.split(\'<>\')`` returns ``[\'1\', \'2\', \'3\']``). Splitting\n   an empty string with a specified separator returns ``[\'\']``.\n\n   If *sep* is not specified or is ``None``, a different splitting\n   algorithm is applied: runs of consecutive whitespace are regarded\n   as a single separator, and the result will contain no empty strings\n   at the start or end if the string has leading or trailing\n   whitespace.  Consequently, splitting an empty string or a string\n   consisting of just whitespace with a ``None`` separator returns\n   ``[]``.\n\n   For example, ``\' 1  2   3  \'.split()`` returns ``[\'1\', \'2\', \'3\']``,\n   and ``\'  1  2   3  \'.split(None, 1)`` returns ``[\'1\', \'2   3  \']``.\n\nstr.splitlines([keepends])\n\n   Return a list of the lines in the string, breaking at line\n   boundaries.  Line breaks are not included in the resulting list\n   unless *keepends* is given and true.\n\nstr.startswith(prefix[, start[, end]])\n\n   Return ``True`` if string starts with the *prefix*, otherwise\n   return ``False``. *prefix* can also be a tuple of prefixes to look\n   for.  With optional *start*, test string beginning at that\n   position.  With optional *end*, stop comparing string at that\n   position.\n\nstr.strip([chars])\n\n   Return a copy of the string with the leading and trailing\n   characters removed. The *chars* argument is a string specifying the\n   set of characters to be removed. If omitted or ``None``, the\n   *chars* argument defaults to removing whitespace. The *chars*\n   argument is not a prefix or suffix; rather, all combinations of its\n   values are stripped:\n\n   >>> \'   spacious   \'.strip()\n   \'spacious\'\n   >>> \'www.example.com\'.strip(\'cmowz.\')\n   \'example\'\n\nstr.swapcase()\n\n   Return a copy of the string with uppercase characters converted to\n   lowercase and vice versa.\n\nstr.title()\n\n   Return a titlecased version of the string where words start with an\n   uppercase character and the remaining characters are lowercase.\n\n   The algorithm uses a simple language-independent definition of a\n   word as groups of consecutive letters.  The definition works in\n   many contexts but it means that apostrophes in contractions and\n   possessives form word boundaries, which may not be the desired\n   result:\n\n      >>> "they\'re bill\'s friends from the UK".title()\n      "They\'Re Bill\'S Friends From The Uk"\n\n   A workaround for apostrophes can be constructed using regular\n   expressions:\n\n      >>> import re\n      >>> def titlecase(s):\n              return re.sub(r"[A-Za-z]+(\'[A-Za-z]+)?",\n                            lambda mo: mo.group(0)[0].upper() +\n                                       mo.group(0)[1:].lower(),\n                            s)\n\n      >>> titlecase("they\'re bill\'s friends.")\n      "They\'re Bill\'s Friends."\n\nstr.translate(map)\n\n   Return a copy of the *s* where all characters have been mapped\n   through the *map* which must be a dictionary of Unicode ordinals\n   (integers) to Unicode ordinals, strings or ``None``.  Unmapped\n   characters are left untouched. Characters mapped to ``None`` are\n   deleted.\n\n   You can use ``str.maketrans()`` to create a translation map from\n   character-to-character mappings in different formats.\n\n   You can use the ``maketrans()`` helper function in the ``string``\n   module to create a translation table. For string objects, set the\n   *table* argument to ``None`` for translations that only delete\n   characters:\n\n   Note: An even more flexible approach is to create a custom character\n     mapping codec using the ``codecs`` module (see\n     ``encodings.cp1251`` for an example).\n\nstr.upper()\n\n   Return a copy of the string converted to uppercase.\n\nstr.zfill(width)\n\n   Return the numeric string left filled with zeros in a string of\n   length *width*.  A sign prefix is handled correctly.  The original\n   string is returned if *width* is less than ``len(s)``.\n\n\nOld String Formatting Operations\n================================\n\nNote: The formatting operations described here are obsolete and may go\n  away in future versions of Python.  Use the new *String Formatting*\n  in new code.\n\nString objects have one unique built-in operation: the ``%`` operator\n(modulo). This is also known as the string *formatting* or\n*interpolation* operator. Given ``format % values`` (where *format* is\na string), ``%`` conversion specifications in *format* are replaced\nwith zero or more elements of *values*. The effect is similar to the\nusing ``sprintf()`` in the C language.\n\nIf *format* requires a single argument, *values* may be a single non-\ntuple object. [4]  Otherwise, *values* must be a tuple with exactly\nthe number of items specified by the format string, or a single\nmapping object (for example, a dictionary).\n\nA conversion specifier contains two or more characters and has the\nfollowing components, which must occur in this order:\n\n1. The ``\'%\'`` character, which marks the start of the specifier.\n\n2. Mapping key (optional), consisting of a parenthesised sequence of\n   characters (for example, ``(somename)``).\n\n3. Conversion flags (optional), which affect the result of some\n   conversion types.\n\n4. Minimum field width (optional).  If specified as an ``\'*\'``\n   (asterisk), the actual width is read from the next element of the\n   tuple in *values*, and the object to convert comes after the\n   minimum field width and optional precision.\n\n5. Precision (optional), given as a ``\'.\'`` (dot) followed by the\n   precision.  If specified as ``\'*\'`` (an asterisk), the actual width\n   is read from the next element of the tuple in *values*, and the\n   value to convert comes after the precision.\n\n6. Length modifier (optional).\n\n7. Conversion type.\n\nWhen the right argument is a dictionary (or other mapping type), then\nthe formats in the string *must* include a parenthesised mapping key\ninto that dictionary inserted immediately after the ``\'%\'`` character.\nThe mapping key selects the value to be formatted from the mapping.\nFor example:\n\n>>> print(\'%(language)s has %(#)03d quote types.\' % \\\n...       {\'language\': "Python", "#": 2})\nPython has 002 quote types.\n\nIn this case no ``*`` specifiers may occur in a format (since they\nrequire a sequential parameter list).\n\nThe conversion flag characters are:\n\n+-----------+-----------------------------------------------------------------------+\n| Flag      | Meaning                                                               |\n+===========+=======================================================================+\n| ``\'#\'``   | The value conversion will use the "alternate form" (where defined     |\n|           | below).                                                               |\n+-----------+-----------------------------------------------------------------------+\n| ``\'0\'``   | The conversion will be zero padded for numeric values.                |\n+-----------+-----------------------------------------------------------------------+\n| ``\'-\'``   | The converted value is left adjusted (overrides the ``\'0\'``           |\n|           | conversion if both are given).                                        |\n+-----------+-----------------------------------------------------------------------+\n| ``\' \'``   | (a space) A blank should be left before a positive number (or empty   |\n|           | string) produced by a signed conversion.                              |\n+-----------+-----------------------------------------------------------------------+\n| ``\'+\'``   | A sign character (``\'+\'`` or ``\'-\'``) will precede the conversion     |\n|           | (overrides a "space" flag).                                           |\n+-----------+-----------------------------------------------------------------------+\n\nA length modifier (``h``, ``l``, or ``L``) may be present, but is\nignored as it is not necessary for Python -- so e.g. ``%ld`` is\nidentical to ``%d``.\n\nThe conversion types are:\n\n+--------------+-------------------------------------------------------+---------+\n| Conversion   | Meaning                                               | Notes   |\n+==============+=======================================================+=========+\n| ``\'d\'``      | Signed integer decimal.                               |         |\n+--------------+-------------------------------------------------------+---------+\n| ``\'i\'``      | Signed integer decimal.                               |         |\n+--------------+-------------------------------------------------------+---------+\n| ``\'o\'``      | Signed octal value.                                   | (1)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'u\'``      | Obsolete type -- it is identical to ``\'d\'``.          | (7)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'x\'``      | Signed hexadecimal (lowercase).                       | (2)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'X\'``      | Signed hexadecimal (uppercase).                       | (2)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'e\'``      | Floating point exponential format (lowercase).        | (3)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'E\'``      | Floating point exponential format (uppercase).        | (3)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'f\'``      | Floating point decimal format.                        | (3)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'F\'``      | Floating point decimal format.                        | (3)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'g\'``      | Floating point format. Uses lowercase exponential     | (4)     |\n|              | format if exponent is less than -4 or not less than   |         |\n|              | precision, decimal format otherwise.                  |         |\n+--------------+-------------------------------------------------------+---------+\n| ``\'G\'``      | Floating point format. Uses uppercase exponential     | (4)     |\n|              | format if exponent is less than -4 or not less than   |         |\n|              | precision, decimal format otherwise.                  |         |\n+--------------+-------------------------------------------------------+---------+\n| ``\'c\'``      | Single character (accepts integer or single character |         |\n|              | string).                                              |         |\n+--------------+-------------------------------------------------------+---------+\n| ``\'r\'``      | String (converts any Python object using ``repr()``). | (5)     |\n+--------------+-------------------------------------------------------+---------+\n| ``\'s\'``      | String (converts any Python object using ``str()``).  |         |\n+--------------+-------------------------------------------------------+---------+\n| ``\'%\'``      | No argument is converted, results in a ``\'%\'``        |         |\n|              | character in the result.                              |         |\n+--------------+-------------------------------------------------------+---------+\n\nNotes:\n\n1. The alternate form causes a leading zero (``\'0\'``) to be inserted\n   between left-hand padding and the formatting of the number if the\n   leading character of the result is not already a zero.\n\n2. The alternate form causes a leading ``\'0x\'`` or ``\'0X\'`` (depending\n   on whether the ``\'x\'`` or ``\'X\'`` format was used) to be inserted\n   between left-hand padding and the formatting of the number if the\n   leading character of the result is not already a zero.\n\n3. The alternate form causes the result to always contain a decimal\n   point, even if no digits follow it.\n\n   The precision determines the number of digits after the decimal\n   point and defaults to 6.\n\n4. The alternate form causes the result to always contain a decimal\n   point, and trailing zeroes are not removed as they would otherwise\n   be.\n\n   The precision determines the number of significant digits before\n   and after the decimal point and defaults to 6.\n\n5. The precision determines the maximal number of characters used.\n\n1. See **PEP 237**.\n\nSince Python strings have an explicit length, ``%s`` conversions do\nnot assume that ``\'\\0\'`` is the end of the string.\n\nChanged in version 3.1: ``%f`` conversions for numbers whose absolute\nvalue is over 1e50 are no longer replaced by ``%g`` conversions.\n\nAdditional string operations are defined in standard modules\n``string`` and ``re``.\n\n\nRange Type\n==========\n\nThe ``range`` type is an immutable sequence which is commonly used for\nlooping.  The advantage of the ``range`` type is that an ``range``\nobject will always take the same amount of memory, no matter the size\nof the range it represents.  There are no consistent performance\nadvantages.\n\nRange objects have very little behavior: they only support indexing,\niteration, and the ``len()`` function.\n\n\nMutable Sequence Types\n======================\n\nList and bytearray objects support additional operations that allow\nin-place modification of the object.  Other mutable sequence types\n(when added to the language) should also support these operations.\nStrings and tuples are immutable sequence types: such objects cannot\nbe modified once created. The following operations are defined on\nmutable sequence types (where *x* is an arbitrary object).\n\nNote that while lists allow their items to be of any type, bytearray\nobject "items" are all integers in the range 0 <= x < 256.\n\n+--------------------------------+----------------------------------+-----------------------+\n| Operation                      | Result                           | Notes                 |\n+================================+==================================+=======================+\n| ``s[i] = x``                   | item *i* of *s* is replaced by   |                       |\n|                                | *x*                              |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s[i:j] = t``                 | slice of *s* from *i* to *j* is  |                       |\n|                                | replaced by the contents of the  |                       |\n|                                | iterable *t*                     |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``del s[i:j]``                 | same as ``s[i:j] = []``          |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s[i:j:k] = t``               | the elements of ``s[i:j:k]`` are | (1)                   |\n|                                | replaced by those of *t*         |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``del s[i:j:k]``               | removes the elements of          |                       |\n|                                | ``s[i:j:k]`` from the list       |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.append(x)``                | same as ``s[len(s):len(s)] =     |                       |\n|                                | [x]``                            |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.extend(x)``                | same as ``s[len(s):len(s)] = x`` | (2)                   |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.count(x)``                 | return number of *i*\'s for which |                       |\n|                                | ``s[i] == x``                    |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.index(x[, i[, j]])``       | return smallest *k* such that    | (3)                   |\n|                                | ``s[k] == x`` and ``i <= k < j`` |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.insert(i, x)``             | same as ``s[i:i] = [x]``         | (4)                   |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.pop([i])``                 | same as ``x = s[i]; del s[i];    | (5)                   |\n|                                | return x``                       |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.remove(x)``                | same as ``del s[s.index(x)]``    | (3)                   |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.reverse()``                | reverses the items of *s* in     | (6)                   |\n|                                | place                            |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.sort([key[, reverse]])``   | sort the items of *s* in place   | (6), (7), (8)         |\n+--------------------------------+----------------------------------+-----------------------+\n\nNotes:\n\n1. *t* must have the same length as the slice it is replacing.\n\n2. *x* can be any iterable object.\n\n3. Raises ``ValueError`` when *x* is not found in *s*. When a negative\n   index is passed as the second or third parameter to the ``index()``\n   method, the sequence length is added, as for slice indices.  If it\n   is still negative, it is truncated to zero, as for slice indices.\n\n4. When a negative index is passed as the first parameter to the\n   ``insert()`` method, the sequence length is added, as for slice\n   indices.  If it is still negative, it is truncated to zero, as for\n   slice indices.\n\n5. The optional argument *i* defaults to ``-1``, so that by default\n   the last item is removed and returned.\n\n6. The ``sort()`` and ``reverse()`` methods modify the sequence in\n   place for economy of space when sorting or reversing a large\n   sequence.  To remind you that they operate by side effect, they\n   don\'t return the sorted or reversed sequence.\n\n7. The ``sort()`` method takes optional arguments for controlling the\n   comparisons.  Each must be specified as a keyword argument.\n\n   *key* specifies a function of one argument that is used to extract\n   a comparison key from each list element: ``key=str.lower``.  The\n   default value is ``None``. Use ``functools.cmp_to_key()`` to\n   convert an old-style *cmp* function to a *key* function.\n\n   *reverse* is a boolean value.  If set to ``True``, then the list\n   elements are sorted as if each comparison were reversed.\n\n   The ``sort()`` method is guaranteed to be stable.  A sort is stable\n   if it guarantees not to change the relative order of elements that\n   compare equal --- this is helpful for sorting in multiple passes\n   (for example, sort by department, then by salary grade).\n\n   **CPython implementation detail:** While a list is being sorted,\n   the effect of attempting to mutate, or even inspect, the list is\n   undefined.  The C implementation of Python makes the list appear\n   empty for the duration, and raises ``ValueError`` if it can detect\n   that the list has been mutated during a sort.\n\n8. ``sort()`` is not supported by ``bytearray`` objects.\n\n\nBytes and Byte Array Methods\n============================\n\nBytes and bytearray objects, being "strings of bytes", have all\nmethods found on strings, with the exception of ``encode()``,\n``format()`` and ``isidentifier()``, which do not make sense with\nthese types.  For converting the objects to strings, they have a\n``decode()`` method.\n\nWherever one of these methods needs to interpret the bytes as\ncharacters (e.g. the ``is...()`` methods), the ASCII character set is\nassumed.\n\nNote: The methods on bytes and bytearray objects don\'t accept strings as\n  their arguments, just as the methods on strings don\'t accept bytes\n  as their arguments.  For example, you have to write\n\n     a = "abc"\n     b = a.replace("a", "f")\n\n  and\n\n     a = b"abc"\n     b = a.replace(b"a", b"f")\n\nbytes.decode(encoding=sys.getdefaultencoding(), errors="strict")\nbytearray.decode(encoding=sys.getdefaultencoding(), errors="strict")\n\n   Return a string decoded from the given bytes.  Default encoding is\n   the current default string encoding.  *errors* may be given to set\n   a different error handling scheme.  The default for *errors* is\n   ``\'strict\'``, meaning that encoding errors raise a\n   ``UnicodeError``.  Other possible values are ``\'ignore\'``,\n   ``\'replace\'`` and any other name registered via\n   ``codecs.register_error()``, see section *Codec Base Classes*. For\n   a list of possible encodings, see section *Standard Encodings*.\n\n   Changed in version 3.1: Added support for keyword arguments.\n\nThe bytes and bytearray types have an additional class method:\n\nclassmethod bytes.fromhex(string)\nclassmethod bytearray.fromhex(string)\n\n   This ``bytes`` class method returns a bytes or bytearray object,\n   decoding the given string object.  The string must contain two\n   hexadecimal digits per byte, spaces are ignored.\n\n   >>> bytes.fromhex(\'f0 f1f2  \')\n   b\'\\xf0\\xf1\\xf2\'\n\nThe maketrans and translate methods differ in semantics from the\nversions available on strings:\n\nbytes.translate(table[, delete])\nbytearray.translate(table[, delete])\n\n   Return a copy of the bytes or bytearray object where all bytes\n   occurring in the optional argument *delete* are removed, and the\n   remaining bytes have been mapped through the given translation\n   table, which must be a bytes object of length 256.\n\n   You can use the ``bytes.maketrans()`` method to create a\n   translation table.\n\n   Set the *table* argument to ``None`` for translations that only\n   delete characters:\n\n      >>> b\'read this short text\'.translate(None, b\'aeiou\')\n      b\'rd ths shrt txt\'\n\nstatic bytes.maketrans(from, to)\nstatic bytearray.maketrans(from, to)\n\n   This static method returns a translation table usable for\n   ``bytes.translate()`` that will map each character in *from* into\n   the character at the same position in *to*; *from* and *to* must be\n   bytes objects and have the same length.\n\n   New in version 3.1.\n',
+ 'typesseq-mutable': '\nMutable Sequence Types\n**********************\n\nList and bytearray objects support additional operations that allow\nin-place modification of the object.  Other mutable sequence types\n(when added to the language) should also support these operations.\nStrings and tuples are immutable sequence types: such objects cannot\nbe modified once created. The following operations are defined on\nmutable sequence types (where *x* is an arbitrary object).\n\nNote that while lists allow their items to be of any type, bytearray\nobject "items" are all integers in the range 0 <= x < 256.\n\n+--------------------------------+----------------------------------+-----------------------+\n| Operation                      | Result                           | Notes                 |\n+================================+==================================+=======================+\n| ``s[i] = x``                   | item *i* of *s* is replaced by   |                       |\n|                                | *x*                              |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s[i:j] = t``                 | slice of *s* from *i* to *j* is  |                       |\n|                                | replaced by the contents of the  |                       |\n|                                | iterable *t*                     |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``del s[i:j]``                 | same as ``s[i:j] = []``          |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s[i:j:k] = t``               | the elements of ``s[i:j:k]`` are | (1)                   |\n|                                | replaced by those of *t*         |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``del s[i:j:k]``               | removes the elements of          |                       |\n|                                | ``s[i:j:k]`` from the list       |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.append(x)``                | same as ``s[len(s):len(s)] =     |                       |\n|                                | [x]``                            |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.extend(x)``                | same as ``s[len(s):len(s)] = x`` | (2)                   |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.count(x)``                 | return number of *i*\'s for which |                       |\n|                                | ``s[i] == x``                    |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.index(x[, i[, j]])``       | return smallest *k* such that    | (3)                   |\n|                                | ``s[k] == x`` and ``i <= k < j`` |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.insert(i, x)``             | same as ``s[i:i] = [x]``         | (4)                   |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.pop([i])``                 | same as ``x = s[i]; del s[i];    | (5)                   |\n|                                | return x``                       |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.remove(x)``                | same as ``del s[s.index(x)]``    | (3)                   |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.reverse()``                | reverses the items of *s* in     | (6)                   |\n|                                | place                            |                       |\n+--------------------------------+----------------------------------+-----------------------+\n| ``s.sort([key[, reverse]])``   | sort the items of *s* in place   | (6), (7), (8)         |\n+--------------------------------+----------------------------------+-----------------------+\n\nNotes:\n\n1. *t* must have the same length as the slice it is replacing.\n\n2. *x* can be any iterable object.\n\n3. Raises ``ValueError`` when *x* is not found in *s*. When a negative\n   index is passed as the second or third parameter to the ``index()``\n   method, the sequence length is added, as for slice indices.  If it\n   is still negative, it is truncated to zero, as for slice indices.\n\n4. When a negative index is passed as the first parameter to the\n   ``insert()`` method, the sequence length is added, as for slice\n   indices.  If it is still negative, it is truncated to zero, as for\n   slice indices.\n\n5. The optional argument *i* defaults to ``-1``, so that by default\n   the last item is removed and returned.\n\n6. The ``sort()`` and ``reverse()`` methods modify the sequence in\n   place for economy of space when sorting or reversing a large\n   sequence.  To remind you that they operate by side effect, they\n   don\'t return the sorted or reversed sequence.\n\n7. The ``sort()`` method takes optional arguments for controlling the\n   comparisons.  Each must be specified as a keyword argument.\n\n   *key* specifies a function of one argument that is used to extract\n   a comparison key from each list element: ``key=str.lower``.  The\n   default value is ``None``. Use ``functools.cmp_to_key()`` to\n   convert an old-style *cmp* function to a *key* function.\n\n   *reverse* is a boolean value.  If set to ``True``, then the list\n   elements are sorted as if each comparison were reversed.\n\n   The ``sort()`` method is guaranteed to be stable.  A sort is stable\n   if it guarantees not to change the relative order of elements that\n   compare equal --- this is helpful for sorting in multiple passes\n   (for example, sort by department, then by salary grade).\n\n   **CPython implementation detail:** While a list is being sorted,\n   the effect of attempting to mutate, or even inspect, the list is\n   undefined.  The C implementation of Python makes the list appear\n   empty for the duration, and raises ``ValueError`` if it can detect\n   that the list has been mutated during a sort.\n\n8. ``sort()`` is not supported by ``bytearray`` objects.\n',
  'unary': '\nUnary arithmetic and bitwise operations\n***************************************\n\nAll unary arithmetic and bitwise operations have the same priority:\n\n   u_expr ::= power | "-" u_expr | "+" u_expr | "~" u_expr\n\nThe unary ``-`` (minus) operator yields the negation of its numeric\nargument.\n\nThe unary ``+`` (plus) operator yields its numeric argument unchanged.\n\nThe unary ``~`` (invert) operator yields the bitwise inversion of its\ninteger argument.  The bitwise inversion of ``x`` is defined as\n``-(x+1)``.  It only applies to integral numbers.\n\nIn all three cases, if the argument does not have the proper type, a\n``TypeError`` exception is raised.\n',
  'while': '\nThe ``while`` statement\n***********************\n\nThe ``while`` statement is used for repeated execution as long as an\nexpression is true:\n\n   while_stmt ::= "while" expression ":" suite\n                  ["else" ":" suite]\n\nThis repeatedly tests the expression and, if it is true, executes the\nfirst suite; if the expression is false (which may be the first time\nit is tested) the suite of the ``else`` clause, if present, is\nexecuted and the loop terminates.\n\nA ``break`` statement executed in the first suite terminates the loop\nwithout executing the ``else`` clause\'s suite.  A ``continue``\nstatement executed in the first suite skips the rest of the suite and\ngoes back to testing the expression.\n',
- 'with': '\nThe ``with`` statement\n**********************\n\nThe ``with`` statement is used to wrap the execution of a block with\nmethods defined by a context manager (see section *With Statement\nContext Managers*). This allows common\n``try``...``except``...``finally`` usage patterns to be encapsulated\nfor convenient reuse.\n\n   with_stmt ::= "with" with_item ("," with_item)* ":" suite\n   with_item ::= expression ["as" target]\n\nThe execution of the ``with`` statement with one "item" proceeds as\nfollows:\n\n1. The context expression is evaluated to obtain a context manager.\n\n2. The context manager\'s ``__enter__()`` method is invoked.\n\n3. If a target was included in the ``with`` statement, the return\n   value from ``__enter__()`` is assigned to it.\n\n   Note: The ``with`` statement guarantees that if the ``__enter__()``\n     method returns without an error, then ``__exit__()`` will always\n     be called.  Thus, if an error occurs during the assignment to the\n     target list, it will be treated the same as an error occurring\n     within the suite would be.  See step 5 below.\n\n4. The suite is executed.\n\n5. The context manager\'s ``__exit__()`` method is invoked.  If an\n   exception caused the suite to be exited, its type, value, and\n   traceback are passed as arguments to ``__exit__()``. Otherwise,\n   three ``None`` arguments are supplied.\n\n   If the suite was exited due to an exception, and the return value\n   from the ``__exit__()`` method was false, the exception is\n   reraised.  If the return value was true, the exception is\n   suppressed, and execution continues with the statement following\n   the ``with`` statement.\n\n   If the suite was exited for any reason other than an exception, the\n   return value from ``__exit__()`` is ignored, and execution proceeds\n   at the normal location for the kind of exit that was taken.\n\nWith more than one item, the context managers are processed as if\nmultiple ``with`` statements were nested:\n\n   with A() as a, B() as b:\n       suite\n\nis equivalent to\n\n   with A() as a:\n       with B() as b:\n           suite\n\nChanged in version 3.1: Support for multiple context expressions.\n\nSee also:\n\n   **PEP 0343** - The "with" statement\n      The specification, background, and examples for the Python\n      ``with`` statement.\n',
- 'yield': '\nThe ``yield`` statement\n***********************\n\n   yield_stmt ::= yield_expression\n\nThe ``yield`` statement is only used when defining a generator\nfunction, and is only used in the body of the generator function.\nUsing a ``yield`` statement in a function definition is sufficient to\ncause that definition to create a generator function instead of a\nnormal function. When a generator function is called, it returns an\niterator known as a generator iterator, or more commonly, a generator.\nThe body of the generator function is executed by calling the\n``next()`` function on the generator repeatedly until it raises an\nexception.\n\nWhen a ``yield`` statement is executed, the state of the generator is\nfrozen and the value of **expression_list** is returned to\n``next()``\'s caller.  By "frozen" we mean that all local state is\nretained, including the current bindings of local variables, the\ninstruction pointer, and the internal evaluation stack: enough\ninformation is saved so that the next time ``next()`` is invoked, the\nfunction can proceed exactly as if the ``yield`` statement were just\nanother external call.\n\nThe ``yield`` statement is allowed in the ``try`` clause of a ``try``\n...  ``finally`` construct.  If the generator is not resumed before it\nis finalized (by reaching a zero reference count or by being garbage\ncollected), the generator-iterator\'s ``close()`` method will be\ncalled, allowing any pending ``finally`` clauses to execute.\n\nSee also:\n\n   **PEP 0255** - Simple Generators\n      The proposal for adding generators and the ``yield`` statement\n      to Python.\n\n   **PEP 0342** - Coroutines via Enhanced Generators\n      The proposal that, among other generator enhancements, proposed\n      allowing ``yield`` to appear inside a ``try`` ... ``finally``\n      block.\n'}
+ 'with': '\nThe ``with`` statement\n**********************\n\nThe ``with`` statement is used to wrap the execution of a block with\nmethods defined by a context manager (see section *With Statement\nContext Managers*). This allows common\n``try``...``except``...``finally`` usage patterns to be encapsulated\nfor convenient reuse.\n\n   with_stmt ::= "with" with_item ("," with_item)* ":" suite\n   with_item ::= expression ["as" target]\n\nThe execution of the ``with`` statement with one "item" proceeds as\nfollows:\n\n1. The context expression is evaluated to obtain a context manager.\n\n2. The context manager\'s ``__exit__()`` is loaded for later use.\n\n3. The context manager\'s ``__enter__()`` method is invoked.\n\n4. If a target was included in the ``with`` statement, the return\n   value from ``__enter__()`` is assigned to it.\n\n   Note: The ``with`` statement guarantees that if the ``__enter__()``\n     method returns without an error, then ``__exit__()`` will always\n     be called. Thus, if an error occurs during the assignment to the\n     target list, it will be treated the same as an error occurring\n     within the suite would be. See step 6 below.\n\n5. The suite is executed.\n\n6. The context manager\'s ``__exit__()`` method is invoked.  If an\n   exception caused the suite to be exited, its type, value, and\n   traceback are passed as arguments to ``__exit__()``. Otherwise,\n   three ``None`` arguments are supplied.\n\n   If the suite was exited due to an exception, and the return value\n   from the ``__exit__()`` method was false, the exception is\n   reraised.  If the return value was true, the exception is\n   suppressed, and execution continues with the statement following\n   the ``with`` statement.\n\n   If the suite was exited for any reason other than an exception, the\n   return value from ``__exit__()`` is ignored, and execution proceeds\n   at the normal location for the kind of exit that was taken.\n\nWith more than one item, the context managers are processed as if\nmultiple ``with`` statements were nested:\n\n   with A() as a, B() as b:\n       suite\n\nis equivalent to\n\n   with A() as a:\n       with B() as b:\n           suite\n\nChanged in version 3.1: Support for multiple context expressions.\n\nSee also:\n\n   **PEP 0343** - The "with" statement\n      The specification, background, and examples for the Python\n      ``with`` statement.\n',
+ 'yield': '\nThe ``yield`` statement\n***********************\n\n   yield_stmt ::= yield_expression\n\nThe ``yield`` statement is only used when defining a generator\nfunction, and is only used in the body of the generator function.\nUsing a ``yield`` statement in a function definition is sufficient to\ncause that definition to create a generator function instead of a\nnormal function. When a generator function is called, it returns an\niterator known as a generator iterator, or more commonly, a generator.\nThe body of the generator function is executed by calling the\n``next()`` function on the generator repeatedly until it raises an\nexception.\n\nWhen a ``yield`` statement is executed, the state of the generator is\nfrozen and the value of ``expression_list`` is returned to\n``next()``\'s caller.  By "frozen" we mean that all local state is\nretained, including the current bindings of local variables, the\ninstruction pointer, and the internal evaluation stack: enough\ninformation is saved so that the next time ``next()`` is invoked, the\nfunction can proceed exactly as if the ``yield`` statement were just\nanother external call.\n\nThe ``yield`` statement is allowed in the ``try`` clause of a ``try``\n...  ``finally`` construct.  If the generator is not resumed before it\nis finalized (by reaching a zero reference count or by being garbage\ncollected), the generator-iterator\'s ``close()`` method will be\ncalled, allowing any pending ``finally`` clauses to execute.\n\nSee also:\n\n   **PEP 0255** - Simple Generators\n      The proposal for adding generators and the ``yield`` statement\n      to Python.\n\n   **PEP 0342** - Coroutines via Enhanced Generators\n      The proposal that, among other generator enhancements, proposed\n      allowing ``yield`` to appear inside a ``try`` ... ``finally``\n      block.\n'}

Modified: python/branches/import_unicode/Lib/smtpd.py
==============================================================================
--- python/branches/import_unicode/Lib/smtpd.py	(original)
+++ python/branches/import_unicode/Lib/smtpd.py	Sat Jul 31 13:01:07 2010
@@ -293,10 +293,10 @@
                 else:
                     data.append(text)
             self.received_data = NEWLINE.join(data)
-            status = self.__server.process_message(self.peer,
-                                                   self.mailfrom,
-                                                   self.rcpttos,
-                                                   self.received_data)
+            status = self.smtp_server.process_message(self.peer,
+                                                      self.mailfrom,
+                                                      self.rcpttos,
+                                                      self.received_data)
             self.rcpttos = []
             self.mailfrom = None
             self.smtp_state = self.COMMAND

Modified: python/branches/import_unicode/Lib/test/test_pdb.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_pdb.py	(original)
+++ python/branches/import_unicode/Lib/test/test_pdb.py	Sat Jul 31 13:01:07 2010
@@ -1,8 +1,10 @@
-# A test suite for pdb; at the moment, this only validates skipping of
-# specified test modules (RFE #5142).
+# A test suite for pdb; not very comprehensive at the moment.
 
 import imp
+import pdb
 import sys
+import unittest
+import subprocess
 
 from test import support
 # This little helper class is essential for testing pdb under doctest.
@@ -52,6 +54,373 @@
     """
 
 
+def test_pdb_basic_commands():
+    """Test the basic commands of pdb.
+
+    >>> def test_function_2(foo, bar='default'):
+    ...     print(foo)
+    ...     for i in range(5):
+    ...         print(i)
+    ...     print(bar)
+    ...     for i in range(10):
+    ...         never_executed
+    ...     print('after for')
+    ...     print('...')
+    ...     return foo.upper()
+
+    >>> def test_function():
+    ...     import pdb; pdb.Pdb().set_trace()
+    ...     ret = test_function_2('baz')
+    ...     print(ret)
+
+    >>> with PdbTestInput([  # doctest: +ELLIPSIS, +NORMALIZE_WHITESPACE
+    ...     'step',       # entering the function call
+    ...     'args',       # display function args
+    ...     'list',       # list function source
+    ...     'bt',         # display backtrace
+    ...     'up',         # step up to test_function()
+    ...     'down',       # step down to test_function_2() again
+    ...     'next',       # stepping to print(foo)
+    ...     'next',       # stepping to the for loop
+    ...     'step',       # stepping into the for loop
+    ...     'until',      # continuing until out of the for loop
+    ...     'next',       # executing the print(bar)
+    ...     'jump 8',     # jump over second for loop
+    ...     'return',     # return out of function
+    ...     'retval',     # display return value
+    ...     'continue',
+    ... ]):
+    ...    test_function()
+    > <doctest test.test_pdb.test_pdb_basic_commands[1]>(3)test_function()
+    -> ret = test_function_2('baz')
+    (Pdb) step
+    --Call--
+    > <doctest test.test_pdb.test_pdb_basic_commands[0]>(1)test_function_2()
+    -> def test_function_2(foo, bar='default'):
+    (Pdb) args
+    foo = 'baz'
+    bar = 'default'
+    (Pdb) list
+      1  ->     def test_function_2(foo, bar='default'):
+      2             print(foo)
+      3             for i in range(5):
+      4                 print(i)
+      5             print(bar)
+      6             for i in range(10):
+      7                 never_executed
+      8             print('after for')
+      9             print('...')
+     10             return foo.upper()
+    [EOF]
+    (Pdb) bt
+    ...
+      <doctest test.test_pdb.test_pdb_basic_commands[2]>(18)<module>()
+    -> test_function()
+      <doctest test.test_pdb.test_pdb_basic_commands[1]>(3)test_function()
+    -> ret = test_function_2('baz')
+    > <doctest test.test_pdb.test_pdb_basic_commands[0]>(1)test_function_2()
+    -> def test_function_2(foo, bar='default'):
+    (Pdb) up
+    > <doctest test.test_pdb.test_pdb_basic_commands[1]>(3)test_function()
+    -> ret = test_function_2('baz')
+    (Pdb) down
+    > <doctest test.test_pdb.test_pdb_basic_commands[0]>(1)test_function_2()
+    -> def test_function_2(foo, bar='default'):
+    (Pdb) next
+    > <doctest test.test_pdb.test_pdb_basic_commands[0]>(2)test_function_2()
+    -> print(foo)
+    (Pdb) next
+    baz
+    > <doctest test.test_pdb.test_pdb_basic_commands[0]>(3)test_function_2()
+    -> for i in range(5):
+    (Pdb) step
+    > <doctest test.test_pdb.test_pdb_basic_commands[0]>(4)test_function_2()
+    -> print(i)
+    (Pdb) until
+    0
+    1
+    2
+    3
+    4
+    > <doctest test.test_pdb.test_pdb_basic_commands[0]>(5)test_function_2()
+    -> print(bar)
+    (Pdb) next
+    default
+    > <doctest test.test_pdb.test_pdb_basic_commands[0]>(6)test_function_2()
+    -> for i in range(10):
+    (Pdb) jump 8
+    > <doctest test.test_pdb.test_pdb_basic_commands[0]>(8)test_function_2()
+    -> print('after for')
+    (Pdb) return
+    after for
+    ...
+    --Return--
+    > <doctest test.test_pdb.test_pdb_basic_commands[0]>(10)test_function_2()->'BAZ'
+    -> return foo.upper()
+    (Pdb) retval
+    'BAZ'
+    (Pdb) continue
+    BAZ
+    """
+
+
+def test_pdb_breakpoint_commands():
+    """Test basic commands related to breakpoints.
+
+    >>> def test_function():
+    ...     import pdb; pdb.Pdb().set_trace()
+    ...     print(1)
+    ...     print(2)
+    ...     print(3)
+    ...     print(4)
+
+    First, need to clear bdb state that might be left over from previous tests.
+    Otherwise, the new breakpoints might get assigned different numbers.
+
+    >>> from bdb import Breakpoint
+    >>> Breakpoint.next = 1
+    >>> Breakpoint.bplist = {}
+    >>> Breakpoint.bpbynumber = [None]
+
+    Now test the breakpoint commands.  NORMALIZE_WHITESPACE is needed because
+    the breakpoint list outputs a tab for the "stop only" and "ignore next"
+    lines, which we don't want to put in here.
+
+    >>> with PdbTestInput([  # doctest: +NORMALIZE_WHITESPACE
+    ...     'break 3',
+    ...     'disable 1',
+    ...     'ignore 1 10',
+    ...     'condition 1 1 < 2',
+    ...     'break 4',
+    ...     'break',
+    ...     'condition 1',
+    ...     'enable 1',
+    ...     'clear 1',
+    ...     'commands 2',
+    ...     'print 42',
+    ...     'end',
+    ...     'continue',  # will stop at breakpoint 2 (line 4)
+    ...     'clear',     # clear all!
+    ...     'y',
+    ...     'tbreak 5',
+    ...     'continue',  # will stop at temporary breakpoint
+    ...     'break',     # make sure breakpoint is gone
+    ...     'continue',
+    ... ]):
+    ...    test_function()
+    > <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>(3)test_function()
+    -> print(1)
+    (Pdb) break 3
+    Breakpoint 1 at <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>:3
+    (Pdb) disable 1
+    Disabled breakpoint 1 at <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>:3
+    (Pdb) ignore 1 10
+    Will ignore next 10 crossings of breakpoint 1.
+    (Pdb) condition 1 1 < 2
+    New condition set for breakpoint 1.
+    (Pdb) break 4
+    Breakpoint 2 at <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>:4
+    (Pdb) break
+    Num Type         Disp Enb   Where
+    1   breakpoint   keep no    at <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>:3
+            stop only if 1 < 2
+            ignore next 10 hits
+    2   breakpoint   keep yes   at <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>:4
+    (Pdb) condition 1
+    Breakpoint 1 is now unconditional.
+    (Pdb) enable 1
+    Enabled breakpoint 1 at <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>:3
+    (Pdb) clear 1
+    Deleted breakpoint 1 at <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>:3
+    (Pdb) commands 2
+    (com) print 42
+    (com) end
+    (Pdb) continue
+    1
+    42
+    > <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>(4)test_function()
+    -> print(2)
+    (Pdb) clear
+    Clear all breaks? y
+    Deleted breakpoint 2 at <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>:4
+    (Pdb) tbreak 5
+    Breakpoint 3 at <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>:5
+    (Pdb) continue
+    2
+    Deleted breakpoint 3 at <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>:5
+    > <doctest test.test_pdb.test_pdb_breakpoint_commands[0]>(5)test_function()
+    -> print(3)
+    (Pdb) break
+    (Pdb) continue
+    3
+    4
+    """
+
+
+def do_nothing():
+    pass
+
+def do_something():
+    print(42)
+
+def test_list_commands():
+    """Test the list and source commands of pdb.
+
+    >>> def test_function_2(foo):
+    ...     import test_pdb
+    ...     test_pdb.do_nothing()
+    ...     'some...'
+    ...     'more...'
+    ...     'code...'
+    ...     'to...'
+    ...     'make...'
+    ...     'a...'
+    ...     'long...'
+    ...     'listing...'
+    ...     'useful...'
+    ...     '...'
+    ...     '...'
+    ...     return foo
+
+    >>> def test_function():
+    ...     import pdb; pdb.Pdb().set_trace()
+    ...     ret = test_function_2('baz')
+
+    >>> with PdbTestInput([  # doctest: +ELLIPSIS, +NORMALIZE_WHITESPACE
+    ...     'list',      # list first function
+    ...     'step',      # step into second function
+    ...     'list',      # list second function
+    ...     'list',      # continue listing to EOF
+    ...     'list 1,3',  # list specific lines
+    ...     'list x',    # invalid argument
+    ...     'next',      # step to import
+    ...     'next',      # step over import
+    ...     'step',      # step into do_nothing
+    ...     'longlist',  # list all lines
+    ...     'source do_something',  # list all lines of function
+    ...     'source fooxxx',        # something that doesn't exit
+    ...     'continue',
+    ... ]):
+    ...    test_function()
+    > <doctest test.test_pdb.test_list_commands[1]>(3)test_function()
+    -> ret = test_function_2('baz')
+    (Pdb) list
+      1         def test_function():
+      2             import pdb; pdb.Pdb().set_trace()
+      3  ->         ret = test_function_2('baz')
+    [EOF]
+    (Pdb) step
+    --Call--
+    > <doctest test.test_pdb.test_list_commands[0]>(1)test_function_2()
+    -> def test_function_2(foo):
+    (Pdb) list
+      1  ->     def test_function_2(foo):
+      2             import test_pdb
+      3             test_pdb.do_nothing()
+      4             'some...'
+      5             'more...'
+      6             'code...'
+      7             'to...'
+      8             'make...'
+      9             'a...'
+     10             'long...'
+     11             'listing...'
+    (Pdb) list
+     12             'useful...'
+     13             '...'
+     14             '...'
+     15             return foo
+    [EOF]
+    (Pdb) list 1,3
+      1  ->     def test_function_2(foo):
+      2             import test_pdb
+      3             test_pdb.do_nothing()
+    (Pdb) list x
+    *** ...
+    (Pdb) next
+    > <doctest test.test_pdb.test_list_commands[0]>(2)test_function_2()
+    -> import test_pdb
+    (Pdb) next
+    > <doctest test.test_pdb.test_list_commands[0]>(3)test_function_2()
+    -> test_pdb.do_nothing()
+    (Pdb) step
+    --Call--
+    > ...test_pdb.py(...)do_nothing()
+    -> def do_nothing():
+    (Pdb) longlist
+    ...  ->     def do_nothing():
+    ...             pass
+    (Pdb) source do_something
+    ...         def do_something():
+    ...             print(42)
+    (Pdb) source fooxxx
+    *** ...
+    (Pdb) continue
+    """
+
+
+def test_post_mortem():
+    """Test post mortem traceback debugging.
+
+    >>> def test_function_2():
+    ...     try:
+    ...         1/0
+    ...     finally:
+    ...         print('Exception!')
+
+    >>> def test_function():
+    ...     import pdb; pdb.Pdb().set_trace()
+    ...     test_function_2()
+    ...     print('Not reached.')
+
+    >>> with PdbTestInput([  # doctest: +ELLIPSIS, +NORMALIZE_WHITESPACE
+    ...     'next',      # step over exception-raising call
+    ...     'bt',        # get a backtrace
+    ...     'list',      # list code of test_function()
+    ...     'down',      # step into test_function_2()
+    ...     'list',      # list code of test_function_2()
+    ...     'continue',
+    ... ]):
+    ...    try:
+    ...        test_function()
+    ...    except ZeroDivisionError:
+    ...        print('Correctly reraised.')
+    > <doctest test.test_pdb.test_post_mortem[1]>(3)test_function()
+    -> test_function_2()
+    (Pdb) next
+    Exception!
+    ZeroDivisionError: division by zero
+    > <doctest test.test_pdb.test_post_mortem[1]>(3)test_function()
+    -> test_function_2()
+    (Pdb) bt
+    ...
+      <doctest test.test_pdb.test_post_mortem[2]>(10)<module>()
+    -> test_function()
+    > <doctest test.test_pdb.test_post_mortem[1]>(3)test_function()
+    -> test_function_2()
+      <doctest test.test_pdb.test_post_mortem[0]>(3)test_function_2()
+    -> 1/0
+    (Pdb) list
+      1         def test_function():
+      2             import pdb; pdb.Pdb().set_trace()
+      3  ->         test_function_2()
+      4             print('Not reached.')
+    [EOF]
+    (Pdb) down
+    > <doctest test.test_pdb.test_post_mortem[0]>(3)test_function_2()
+    -> 1/0
+    (Pdb) list
+      1         def test_function_2():
+      2             try:
+      3  >>             1/0
+      4             finally:
+      5  ->             print('Exception!')
+    [EOF]
+    (Pdb) continue
+    Correctly reraised.
+    """
+
+
 def test_pdb_skip_modules():
     """This illustrates the simple case of module skipping.
 
@@ -123,9 +492,116 @@
     """
 
 
+def test_pdb_continue_in_bottomframe():
+    """Test that "continue" and "next" work properly in bottom frame (issue #5294).
+
+    >>> def test_function():
+    ...     import pdb, sys; inst = pdb.Pdb()
+    ...     inst.set_trace()
+    ...     inst.botframe = sys._getframe()  # hackery to get the right botframe
+    ...     print(1)
+    ...     print(2)
+    ...     print(3)
+    ...     print(4)
+
+    >>> with PdbTestInput([  # doctest: +ELLIPSIS
+    ...     'next',
+    ...     'break 7',
+    ...     'continue',
+    ...     'next',
+    ...     'continue',
+    ...     'continue',
+    ... ]):
+    ...    test_function()
+    > <doctest test.test_pdb.test_pdb_continue_in_bottomframe[0]>(4)test_function()
+    -> inst.botframe = sys._getframe()  # hackery to get the right botframe
+    (Pdb) next
+    > <doctest test.test_pdb.test_pdb_continue_in_bottomframe[0]>(5)test_function()
+    -> print(1)
+    (Pdb) break 7
+    Breakpoint ... at <doctest test.test_pdb.test_pdb_continue_in_bottomframe[0]>:7
+    (Pdb) continue
+    1
+    2
+    > <doctest test.test_pdb.test_pdb_continue_in_bottomframe[0]>(7)test_function()
+    -> print(3)
+    (Pdb) next
+    3
+    > <doctest test.test_pdb.test_pdb_continue_in_bottomframe[0]>(8)test_function()
+    -> print(4)
+    (Pdb) continue
+    4
+    """
+
+
+def pdb_invoke(method, arg):
+    """Run pdb.method(arg)."""
+    import pdb; getattr(pdb, method)(arg)
+
+
+def test_pdb_run_with_incorrect_argument():
+    """Testing run and runeval with incorrect first argument.
+
+    >>> pti = PdbTestInput(['continue',])
+    >>> with pti:
+    ...     pdb_invoke('run', lambda x: x)
+    Traceback (most recent call last):
+    TypeError: exec() arg 1 must be a string, bytes or code object
+
+    >>> with pti:
+    ...     pdb_invoke('runeval', lambda x: x)
+    Traceback (most recent call last):
+    TypeError: eval() arg 1 must be a string, bytes or code object
+    """
+
+
+def test_pdb_run_with_code_object():
+    """Testing run and runeval with code object as a first argument.
+
+    >>> with PdbTestInput(['step','x', 'continue']):  # doctest: +ELLIPSIS
+    ...     pdb_invoke('run', compile('x=1', '<string>', 'exec'))
+    > <string>(1)<module>()...
+    (Pdb) step
+    --Return--
+    > <string>(1)<module>()->None
+    (Pdb) x
+    1
+    (Pdb) continue
+
+    >>> with PdbTestInput(['x', 'continue']):
+    ...     x=0
+    ...     pdb_invoke('runeval', compile('x+1', '<string>', 'eval'))
+    > <string>(1)<module>()->None
+    (Pdb) x
+    1
+    (Pdb) continue
+    """
+
+
+class PdbTestCase(unittest.TestCase):
+
+    def test_issue7964(self):
+        # open the file as binary so we can force \r\n newline
+        with open(support.TESTFN, 'wb') as f:
+            f.write(b'print("testing my pdb")\r\n')
+        cmd = [sys.executable, '-m', 'pdb', support.TESTFN]
+        proc = subprocess.Popen(cmd,
+            stdout=subprocess.PIPE,
+            stdin=subprocess.PIPE,
+            stderr=subprocess.STDOUT,
+            )
+        stdout, stderr = proc.communicate(b'quit\n')
+        self.assertNotIn(b'SyntaxError', stdout,
+                         "Got a syntax error running test script under PDB")
+
+    def tearDown(self):
+        support.unlink(support.TESTFN)
+
+
 def test_main():
     from test import test_pdb
     support.run_doctest(test_pdb, verbosity=True)
+    support.run_unittest(PdbTestCase)
 
 
 if __name__ == '__main__':

Modified: python/branches/import_unicode/Lib/test/test_smtpd.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_smtpd.py	(original)
+++ python/branches/import_unicode/Lib/test/test_smtpd.py	Sat Jul 31 13:01:07 2010
@@ -79,101 +79,102 @@
 
     def test_missing_data(self):
         self.write_line(b'')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
                          b'500 Error: bad syntax\r\n')
 
     def test_EHLO_not_implemented(self):
         self.write_line(b'EHLO test.example')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
                          b'502 Error: command "EHLO" not implemented\r\n')
 
     def test_HELO(self):
         name = socket.getfqdn()
         self.write_line(b'HELO test.example')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
                          '250 {}\r\n'.format(name).encode('ascii'))
 
     def test_HELO_bad_syntax(self):
         self.write_line(b'HELO')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
                          b'501 Syntax: HELO hostname\r\n')
 
     def test_HELO_duplicate(self):
         self.write_line(b'HELO test.example')
         self.write_line(b'HELO test.example')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
                          b'503 Duplicate HELO/EHLO\r\n')
 
     def test_NOOP(self):
         self.write_line(b'NOOP')
-        self.assertEqual(self.channel.last, b'250 Ok\r\n')
+        self.assertEqual(self.channel.socket.last, b'250 Ok\r\n')
 
     def test_NOOP_bad_syntax(self):
         self.write_line(b'NOOP hi')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
                          b'501 Syntax: NOOP\r\n')
 
     def test_QUIT(self):
         self.write_line(b'QUIT')
-        self.assertEqual(self.channel.last, b'221 Bye\r\n')
+        self.assertEqual(self.channel.socket.last, b'221 Bye\r\n')
 
     def test_QUIT_arg_ignored(self):
         self.write_line(b'QUIT bye bye')
-        self.assertEqual(self.channel.last, b'221 Bye\r\n')
+        self.assertEqual(self.channel.socket.last, b'221 Bye\r\n')
 
     def test_bad_state(self):
-        self.channel._SMTPChannel__state = 'BAD STATE'
+        self.channel.smtp_state = 'BAD STATE'
         self.write_line(b'HELO')
-        self.assertEqual(self.channel.last, b'451 Internal confusion\r\n')
+        self.assertEqual(self.channel.socket.last,
+                         b'451 Internal confusion\r\n')
 
     def test_need_MAIL(self):
         self.write_line(b'RCPT to:spam at example')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
             b'503 Error: need MAIL command\r\n')
 
     def test_MAIL_syntax(self):
         self.write_line(b'MAIL from eggs at example')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
             b'501 Syntax: MAIL FROM:<address>\r\n')
 
     def test_MAIL_missing_from(self):
         self.write_line(b'MAIL from:')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
             b'501 Syntax: MAIL FROM:<address>\r\n')
 
     def test_MAIL_chevrons(self):
         self.write_line(b'MAIL from:<eggs at example>')
-        self.assertEqual(self.channel.last, b'250 Ok\r\n')
+        self.assertEqual(self.channel.socket.last, b'250 Ok\r\n')
 
     def test_nested_MAIL(self):
         self.write_line(b'MAIL from:eggs at example')
         self.write_line(b'MAIL from:spam at example')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
             b'503 Error: nested MAIL command\r\n')
 
     def test_need_RCPT(self):
         self.write_line(b'MAIL From:eggs at example')
         self.write_line(b'DATA')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
             b'503 Error: need RCPT command\r\n')
 
     def test_RCPT_syntax(self):
         self.write_line(b'MAIL From:eggs at example')
         self.write_line(b'RCPT to eggs at example')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
             b'501 Syntax: RCPT TO: <address>\r\n')
 
     def test_data_dialog(self):
         self.write_line(b'MAIL From:eggs at example')
-        self.assertEqual(self.channel.last, b'250 Ok\r\n')
+        self.assertEqual(self.channel.socket.last, b'250 Ok\r\n')
         self.write_line(b'RCPT To:spam at example')
-        self.assertEqual(self.channel.last, b'250 Ok\r\n')
+        self.assertEqual(self.channel.socket.last, b'250 Ok\r\n')
 
         self.write_line(b'DATA')
-        self.assertEqual(self.channel.last,
+        self.assertEqual(self.channel.socket.last,
             b'354 End data with <CR><LF>.<CR><LF>\r\n')
         self.write_line(b'data\r\nmore\r\n.')
-        self.assertEqual(self.channel.last, b'250 Ok\r\n')
+        self.assertEqual(self.channel.socket.last, b'250 Ok\r\n')
         self.assertEqual(self.server.messages[-1],
             ('peer', 'eggs at example', ['spam at example'], 'data\nmore'))
 
@@ -181,7 +182,7 @@
         self.write_line(b'MAIL From:eggs at example')
         self.write_line(b'RCPT To:spam at example')
         self.write_line(b'DATA spam')
-        self.assertEqual(self.channel.last, b'501 Syntax: DATA\r\n')
+        self.assertEqual(self.channel.socket.last, b'501 Syntax: DATA\r\n')
 
     def test_multiple_RCPT(self):
         self.write_line(b'MAIL From:eggs at example')
@@ -197,13 +198,13 @@
         self.write_line(b'RCPT To:spam at example')
         self.write_line(b'DATA')
         self.write_line(b'return status\r\n.')
-        self.assertEqual(self.channel.last, b'250 Okish\r\n')
+        self.assertEqual(self.channel.socket.last, b'250 Okish\r\n')
 
     def test_RSET(self):
         self.write_line(b'MAIL From:eggs at example')
         self.write_line(b'RCPT To:spam at example')
         self.write_line(b'RSET')
-        self.assertEqual(self.channel.last, b'250 Ok\r\n')
+        self.assertEqual(self.channel.socket.last, b'250 Ok\r\n')
         self.write_line(b'MAIL From:foo at example')
         self.write_line(b'RCPT To:eggs at example')
         self.write_line(b'DATA')
@@ -213,7 +214,7 @@
 
     def test_RSET_syntax(self):
         self.write_line(b'RSET hi')
-        self.assertEqual(self.channel.last, b'501 Syntax: RSET\r\n')
+        self.assertEqual(self.channel.socket.last, b'501 Syntax: RSET\r\n')
 
 
 def test_main():

Modified: python/branches/import_unicode/Lib/test/test_urllib.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_urllib.py	(original)
+++ python/branches/import_unicode/Lib/test/test_urllib.py	Sat Jul 31 13:01:07 2010
@@ -531,6 +531,7 @@
         self.assertEqual(expect, result,
                          "using quote_plus(): %r != %r" % (expect, result))
 
+
 class UnquotingTests(unittest.TestCase):
     """Tests for unquote() and unquote_plus()
 
@@ -558,7 +559,9 @@
         self.assertEqual(result.count('%'), 1,
                          "using unquote(): not all characters escaped: "
                          "%s" % result)
-        self.assertRaises(TypeError, urllib.parse.unquote, None)
+        self.assertRaises((TypeError, AttributeError), urllib.parse.unquote, None)
+        self.assertRaises((TypeError, AttributeError), urllib.parse.unquote, ())
+        self.assertRaises((TypeError, AttributeError), urllib.parse.unquote, b'')
 
     def test_unquoting_badpercent(self):
         # Test unquoting on bad percent-escapes
@@ -593,8 +596,8 @@
         result = urllib.parse.unquote_to_bytes(given)
         self.assertEqual(expect, result, "using unquote_to_bytes(): %r != %r"
                          % (expect, result))
-
-        self.assertRaises(TypeError, urllib.parse.unquote_to_bytes, None)
+        self.assertRaises((TypeError, AttributeError), urllib.parse.unquote_to_bytes, None)
+        self.assertRaises((TypeError, AttributeError), urllib.parse.unquote_to_bytes, ())
 
     def test_unquoting_mixed_case(self):
         # Test unquoting on mixed-case hex digits in the percent-escapes

Modified: python/branches/import_unicode/Lib/urllib/parse.py
==============================================================================
--- python/branches/import_unicode/Lib/urllib/parse.py	(original)
+++ python/branches/import_unicode/Lib/urllib/parse.py	Sat Jul 31 13:01:07 2010
@@ -313,9 +313,7 @@
     """unquote_to_bytes('abc%20def') -> b'abc def'."""
     # Note: strings are encoded as UTF-8. This is only an issue if it contains
     # unescaped non-ASCII characters, which URIs should not.
-    if not string:
-        if string is None:
-            raise TypeError('None object is invalid for unquote_to_bytes()')
+    if string in (b'', ''):
         return b''
     if isinstance(string, str):
         string = string.encode('utf-8')
@@ -340,9 +338,7 @@
 
     unquote('abc%20def') -> 'abc def'.
     """
-    if not string:
-        if string is None:
-            raise TypeError('None object is invalid for unquote() function.')
+    if string == '':
         return string
     res = string.split('%')
     if len(res) == 1:

Modified: python/branches/import_unicode/Misc/NEWS
==============================================================================
--- python/branches/import_unicode/Misc/NEWS	(original)
+++ python/branches/import_unicode/Misc/NEWS	Sat Jul 31 13:01:07 2010
@@ -7,7 +7,7 @@
 What's New in Python 3.2 Alpha 1?
 =================================
 
-*Release date: XX-XXX-XXX*
+*Release date: 01-Aug-2010*
 
 Core and Builtins
 -----------------
@@ -20,16 +20,15 @@
 - Issue #8413: structsequence now subclasses tuple.
 
 - Issue #8271: during the decoding of an invalid UTF-8 byte sequence, only the
-  start byte and the continuation byte(s) are now considered invalid, instead
-  of the number of bytes specified by the start byte.
-  E.g.: '\xf1\x80AB'.decode('utf-8', 'replace') now returns u'\ufffdAB' and
-  replaces with U+FFFD only the start byte ('\xf1') and the continuation byte
-  ('\x80') even if '\xf1' is the start byte of a 4-bytes sequence.
-  Previous versions returned a single u'\ufffd'.
-
-- Issue #9011: A negated imaginary literal (e.g., "-7j") now has real
-  part -0.0 rather than 0.0.  So "-7j" is now exactly equivalent to
-  "-(7j)".
+  start byte and the continuation byte(s) are now considered invalid, instead of
+  the number of bytes specified by the start byte.  E.g.:
+  '\xf1\x80AB'.decode('utf-8', 'replace') now returns u'\ufffdAB' and replaces
+  with U+FFFD only the start byte ('\xf1') and the continuation byte ('\x80')
+  even if '\xf1' is the start byte of a 4-bytes sequence.  Previous versions
+  returned a single u'\ufffd'.
+
+- Issue #9011: A negated imaginary literal (e.g., "-7j") now has real part -0.0
+  rather than 0.0.  So "-7j" is now exactly equivalent to "-(7j)".
 
 - Be more specific in error messages about positional arguments.
 
@@ -37,27 +36,26 @@
   objects, as described in the documentation.
 
 - Issue #6543: Write the traceback in the terminal encoding instead of utf-8.
-  Fix the encoding of the modules filename. Patch written by Amaury Forgeot
+  Fix the encoding of the modules filename.  Patch written by Amaury Forgeot
   d'Arc.
 
-- Issue #9011: Remove buggy and unnecessary (in 3.x) ST->AST
-  compilation code dealing with unary minus applied to a constant.
-  The removed code was mutating the ST, causing a second compilation
-  to fail.
+- Issue #9011: Remove buggy and unnecessary (in 3.x) ST->AST compilation code
+  dealing with unary minus applied to a constant.  The removed code was mutating
+  the ST, causing a second compilation to fail.
 
 - Issue #850997: mbcs encoding (Windows only) handles errors argument: strict
-  mode raises unicode errors. The encoder only supports "strict" and "replace"
+  mode raises unicode errors.  The encoder only supports "strict" and "replace"
   error handlers, the decoder only supports "strict" and "ignore" error
-  handlers. Patch written by Mark Hammond.
+  handlers.  Patch written by Mark Hammond.
 
 - Issue #8850: Remove "w" and "w#" formats from PyArg_Parse*() functions, use
   "w*" format instead. Add tests for "w*" format.
 
 - Issue #8592: PyArg_Parse*() functions raise a TypeError for "y", "u" and "Z"
-  formats if the string contains a null byte/character. Write unit tests for
+  formats if the string contains a null byte/character.  Write unit tests for
   string formats.
 
-- Issue #7490: to facilitate sharing of doctests between 2.x and 3.x test
+- Issue #7490: To facilitate sharing of doctests between 2.x and 3.x test
   suites, the IGNORE_EXCEPTION_DETAIL directive now also ignores the module
   location of the raised exception.
 
@@ -66,94 +64,93 @@
 
 - Issue #9058: Remove assertions about INT_MAX in UnicodeDecodeError.
 
-- Issue #8941: decoding big endian UTF-32 data in UCS-2 builds could crash
-  the interpreter with characters outside the Basic Multilingual Plane
-  (higher than 0x10000).
-
-- Issue #8950: (See also issue #5080).  Py_ArgParse*() functions now
-  raise TypeError instead of giving a DeprecationWarning when a float
-  is parsed using the 'L' code (for long long).  (All other integer
-  codes already raise TypeError in this case.)
+- Issue #8941: Decoding big endian UTF-32 data in UCS-2 builds could crash the
+  interpreter with characters outside the Basic Multilingual Plane (higher than
+  0x10000).
+
+- Issue #8950: (See also issue #5080).  Py_ArgParse*() functions now raise
+  TypeError instead of giving a DeprecationWarning when a float is parsed using
+  the 'L' code (for long long).  (All other integer codes already raise
+  TypeError in this case.)
 
 - Issue #8922: Normalize the encoding name in PyUnicode_AsEncodedString() to
   enable shortcuts for upper case encoding name. Add also a shortcut for
   "iso-8859-1" in PyUnicode_AsEncodedString() and PyUnicode_Decode().
 
-- Issue #8838: Remove codecs.charbuffer_encode() function. The buffer protocol
-  doesn't support "char buffer" anymore in Python3.
+- Issue #8838: Remove codecs.charbuffer_encode() function.  The buffer protocol
+  doesn't support "char buffer" anymore in Python 3.
 
 - Issue #8339: Remove "t#" format of PyArg_Parse*() functions, use "s#" or "s*"
-  instead. codecs.charbuffer_encode() now accepts modifiable buffer objects
+  instead.  codecs.charbuffer_encode() now accepts modifiable buffer objects
   like bytearray.
 
-- Issue #8837: Remove "O?" format of PyArg_Parse*() functions. The format is no
+- Issue #8837: Remove "O?" format of PyArg_Parse*() functions.  The format is no
   used anymore and it was never documented.
 
-- In the str.format(), raise a ValueError when indexes to arguments are too
-  large.
+- In str.format(), raise a ValueError when indexes to arguments are too large.
 
-- Issue #2844: Make int('42', n) consistently raise ValueError for
-  invalid integers n (including n = -909).
+- Issue #2844: Make int('42', n) consistently raise ValueError for invalid
+  integers n (including n = -909).
 
-- Issue #8188: Introduce a new scheme for computing hashes of numbers
-  (instances of int, float, complex, decimal.Decimal and
-  fractions.Fraction) that makes it easy to maintain the invariant
-  that hash(x) == hash(y) whenever x and y have equal value.
+- Issue #8188: Introduce a new scheme for computing hashes of numbers (instances
+  of int, float, complex, decimal.Decimal and fractions.Fraction) that makes it
+  easy to maintain the invariant that hash(x) == hash(y) whenever x and y have
+  equal value.
 
 - Issue #8748: Fix two issues with comparisons between complex and integer
   objects.  (1) The comparison could incorrectly return True in some cases
-  (2**53+1 == complex(2**53) == 2**53), breaking transivity of equality.
-  (2) The comparison raised an OverflowError for large integers, leading
-  to unpredictable exceptions when combining integers and complex objects
-  in sets or dicts.
+  (2**53+1 == complex(2**53) == 2**53), breaking transitivity of equality.
+  (2) The comparison raised an OverflowError for large integers, leading to
+  unpredictable exceptions when combining integers and complex objects in sets
+  or dicts.
 
 - Issue #8766: Initialize _warnings module before importing the first module.
   Fix a crash if an empty directory called "encodings" exists in sys.path.
 
 - Issue #8589: Decode PYTHONWARNINGS environment variable with the file system
   encoding and surrogateespace error handler instead of the locale encoding to
-  be consistent with os.environ. Add PySys_AddWarnOptionUnicode() function.
+  be consistent with os.environ.  Add PySys_AddWarnOptionUnicode() function.
 
-- PyObject_Dump() encodes unicode objects to utf8 with backslashreplace
-  (instead of strict) error handler to escape surrogates
+- PyObject_Dump() encodes unicode objects to utf8 with backslashreplace (instead
+  of strict) error handler to escape surrogates.
 
 - Issue #8715: Create PyUnicode_EncodeFSDefault() function: Encode a Unicode
   object to Py_FileSystemDefaultEncoding with the "surrogateescape" error
-  handler, and return bytes. If Py_FileSystemDefaultEncoding is not set, fall
+  handler, and return bytes.  If Py_FileSystemDefaultEncoding is not set, fall
   back to UTF-8.
 
 - Enable shortcuts for common encodings in PyUnicode_AsEncodedString() for any
-  error handler, not only the default error handler (strict)
+  error handler, not only the default error handler (strict).
 
 - Issue #8610: Load file system codec at startup, and display a fatal error on
-  failure. Set the file system encoding to utf-8 (instead of None) if getting
+  failure.  Set the file system encoding to utf-8 (instead of None) if getting
   the locale encoding failed, or if nl_langinfo(CODESET) function is missing.
 
 - PyFile_FromFd() uses PyUnicode_DecodeFSDefault() instead of
-  PyUnicode_FromString() to support surrogates in the filename and use the
-  right encoding
+  PyUnicode_FromString() to support surrogates in the filename and use the right
+  encoding.
 
 - Issue #7507: Quote "!" in pipes.quote(); it is special to some shells.
 
-- PyUnicode_DecodeFSDefaultAndSize() uses surrogateescape error handler
+- PyUnicode_DecodeFSDefaultAndSize() uses surrogateescape error handler.
 
 - Issue #8419: Prevent the dict constructor from accepting non-string keyword
   arguments.
 
 - Issue #8124: PySys_WriteStdout() and PySys_WriteStderr() don't execute
-  indirectly Python signal handlers anymore because mywrite() ignores
-  exceptions (KeyboardInterrupt)
+  indirectly Python signal handlers anymore because mywrite() ignores exceptions
+  (KeyboardInterrupt).
 
 - Issue #8092: Fix PyUnicode_EncodeUTF8() to support error handler producing
-  unicode string (eg. backslashreplace)
+  unicode string (eg. backslashreplace).
 
-- Issue #8485: PyUnicode_FSConverter() doesn't accept bytearray object anymore,
-  you have to convert your bytearray filenames to bytes
+- Issue #8485: PyUnicode_FSConverter() doesn't accept byteearray objects
+  anymore, you have to convert your bytearray filenames to bytes.
 
 - Issue #7332: Remove the 16KB stack-based buffer in
-  PyMarshal_ReadLastObjectFromFile, which doesn't bring any noticeable
-  benefit compared to the dynamic memory allocation fallback.  Patch by
-  Charles-Fran??ois Natali.
+  PyMarshal_ReadLastObjectFromFile, which doesn't bring any noticeable benefit
+  compared to the dynamic memory allocation fallback.  Patch by Charles-Fran??ois
+  Natali.
 
 - Issue #8417: Raise an OverflowError when an integer larger than sys.maxsize is
   passed to bytes or bytearray.
@@ -163,9 +160,9 @@
 - Issue #8329: Don't return the same lists from select.select when no fds are
   changed.
 
-- Issue #8259: 1L << (2**31) no longer produces an 'outrageous shift error'
-  on 64-bit machines.  The shift count for either left or right shift is
-  permitted to be up to sys.maxsize.
+- Issue #8259: 1L << (2**31) no longer produces an 'outrageous shift error' on
+  64-bit machines.  The shift count for either left or right shift is permitted
+  to be up to sys.maxsize.
 
 - Ensure that tokenization of identifiers is not affected by locale.
 
@@ -173,16 +170,16 @@
 
 - Raise a TypeError when trying to delete a T_STRING_INPLACE struct member.
 
-- Issue #8211: Save/restore CFLAGS around AC_PROG_CC in configure.in, in
-  case it is set.
+- Issue #8211: Save/restore CFLAGS around AC_PROG_CC in configure.in, in case it
+  is set.
 
 - Issue #8226: sys.setfilesystemencoding() raises a LookupError if the encoding
-  is unknown
+  is unknown.
 
-- Issue #1583863: An str subclass can now override the __str__ method
+- Issue #1583863: A str subclass can now override the __str__ method.
 
 - Issue #8014: Setting a T_UINT or T_PYSSIZET attribute of an object with
-  PyMemberDefs could produce an internal error;  raise TypeError instead.
+  PyMemberDefs could produce an internal error; raise TypeError instead.
 
 - Issue #7845: Rich comparison methods on the complex type now return
   NotImplemented rather than raising a TypeError when comparing with an
@@ -191,72 +188,66 @@
 
 - Issue #3137: Don't ignore errors at startup, especially a keyboard interrupt
   (SIGINT). If an error occurs while importing the site module, the error is
-  printed and Python exits. Initialize the GIL before importing the site
-  module.
+  printed and Python exits. Initialize the GIL before importing the site module.
 
 - Issue #7173: Generator finalization could invalidate sys.exc_info().
 
-- Issue #7544: Preallocate thread memory before creating the thread to avoid
-  a fatal error in low memory condition.
+- Issue #7544: Preallocate thread memory before creating the thread to avoid a
+  fatal error in low memory condition.
 
-- Issue #7820: The parser tokenizer restores all bytes in the right if
-  the BOM check fails.
+- Issue #7820: The parser tokenizer restores all bytes in the right if the BOM
+  check fails.
 
 - Handle errors from looking up __prepare__ correctly.
 
-- Issue #5939: Add additional runtime checking to ensure a valid capsule
-  in Modules/_ctypes/callproc.c.
+- Issue #5939: Add additional runtime checking to ensure a valid capsule in
+  Modules/_ctypes/callproc.c.
 
 - Issue #7309: Fix unchecked attribute access when converting
-  UnicodeEncodeError, UnicodeDecodeError, and UnicodeTranslateError to
-  strings.
+  UnicodeEncodeError, UnicodeDecodeError, and UnicodeTranslateError to strings.
 
-- Issue #6902: Fix problem with built-in types format incorrectly with
-  0 padding.
+- Issue #6902: Fix problem with built-in types format incorrectly with 0
+  padding.
 
-- Issue #7988: Fix default alignment to be right aligned for
-  complex.__format__. Now it matches other numeric types.
+- Issue #7988: Fix default alignment to be right aligned for complex.__format__.
+  Now it matches other numeric types.
 
 - Issue #5988: Remove deprecated functions PyOS_ascii_formatd,
-  PyOS_ascii_strtod, and PyOS_ascii_atof. Use PyOS_double_to_string
-  and PyOS_string_to_double instead. See issue #5835 for the original
-  deprecations.
-
-- Issue #7385: Fix a crash in `MemoryView_FromObject` when
-  `PyObject_GetBuffer` fails.  Patch by Florent Xicluna.
-
-- Issue #7788: Fix an interpreter crash produced by deleting a list
-  slice with very large step value.
-
-- Issue #7766: Change sys.getwindowsversion() return value to a named
-  tuple and add the additional members returned in an OSVERSIONINFOEX
-  structure. The new members are service_pack_major, service_pack_minor,
-  suite_mask, and product_type.
-
-- Issue #7561: Operations on empty bytearrays (such as `int(bytearray())`)
-  could crash in many places because of the PyByteArray_AS_STRING() macro
-  returning NULL.  The macro now returns a statically allocated empty
-  string instead.
+  PyOS_ascii_strtod, and PyOS_ascii_atof.  Use PyOS_double_to_string and
+  PyOS_string_to_double instead.  See issue #5835 for the original deprecations.
+
+- Issue #7385: Fix a crash in `MemoryView_FromObject` when `PyObject_GetBuffer`
+  fails.  Patch by Florent Xicluna.
+
+- Issue #7788: Fix an interpreter crash produced by deleting a list slice with
+  very large step value.
+
+- Issue #7766: Change sys.getwindowsversion() return value to a named tuple and
+  add the additional members returned in an OSVERSIONINFOEX structure.  The new
+  members are service_pack_major, service_pack_minor, suite_mask, and
+  product_type.
+
+- Issue #7561: Operations on empty bytearrays (such as `int(bytearray())`) could
+  crash in many places because of the PyByteArray_AS_STRING() macro returning
+  NULL.  The macro now returns a statically allocated empty string instead.
 
 - Issue #6690: Optimize the bytecode for expressions such as `x in {1, 2, 3}`,
-  where the right hand operand is a set of constants, by turning the set into
-  a frozenset and pre-building it as a constant.  The comparison operation
-  is made against the constant instead of building a new set each time it is
-  executed (a similar optimization already existed which turned a list of
-  constants into a pre-built tuple).  Patch and additional tests by Dave
-  Malcolm.
-
-- Issue #7622: Improve the split(), rsplit(), splitlines() and replace()
-  methods of bytes, bytearray and unicode objects by using a common
-  implementation based on stringlib's fast search.  Patch by Florent Xicluna.
-
-- Issue #7632: Fix various str -> float conversion bugs present in 2.7
-  alpha 2, including: (1) a serious 'wrong output' bug that could
-  occur for long (> 40 digit) input strings, (2) a crash in dtoa.c
-  that occurred in debug builds when parsing certain long numeric
-  strings corresponding to subnormal values, (3) a memory leak for
-  some values large enough to cause overflow, and (4) a number of
-  flaws that could lead to incorrectly rounded results.
+  where the right hand operand is a set of constants, by turning the set into a
+  frozenset and pre-building it as a constant.  The comparison operation is made
+  against the constant instead of building a new set each time it is executed (a
+  similar optimization already existed which turned a list of constants into a
+  pre-built tuple).  Patch and additional tests by Dave Malcolm.
+
+- Issue #7622: Improve the split(), rsplit(), splitlines() and replace() methods
+  of bytes, bytearray and unicode objects by using a common implementation based
+  on stringlib's fast search.  Patch by Florent Xicluna.
+
+- Issue #7632: Fix various str -> float conversion bugs present in 2.7 alpha 2,
+  including: (1) a serious 'wrong output' bug that could occur for long (> 40
+  digit) input strings, (2) a crash in dtoa.c that occurred in debug builds when
+  parsing certain long numeric strings corresponding to subnormal values, (3) a
+  memory leak for some values large enough to cause overflow, and (4) a number
+  of flaws that could lead to incorrectly rounded results.
 
 - The __complex__ method is now looked up on the class of instances to make it
   consistent with other special methods.
@@ -267,24 +258,23 @@
 - Issue #7604: Deleting an unset slotted attribute did not raise an
   AttributeError.
 
-- Issue #7534: Fix handling of IEEE specials (infinities, nans,
-  negative zero) in ** operator.  The behaviour now conforms to that
-  described in C99 Annex F.
-
-- Issue #1811: improve accuracy and cross-platform consistency for
-  true division of integers: the result of a/b is now correctly
-  rounded for ints a and b (at least on IEEE 754 platforms), and in
-  particular does not depend on the internal representation of an int.
+- Issue #7534: Fix handling of IEEE specials (infinities, nans, negative zero)
+  in ** operator.  The behaviour now conforms to that described in C99 Annex F.
+
+- Issue #1811: improve accuracy and cross-platform consistency for true division
+  of integers: the result of a/b is now correctly rounded for ints a and b (at
+  least on IEEE 754 platforms), and in particular does not depend on the
+  internal representation of an int.
 
 - Issue #6834: replace the implementation for the 'python' and 'pythonw'
   executables on OSX.
 
-  These executables now work properly with the arch(1) command:
-  ``arch -ppc python`` will start a universal binary version of python
-  in PPC mode (unlike previous releases).
+  These executables now work properly with the arch(1) command: ``arch -ppc
+  python`` will start a universal binary version of python in PPC mode (unlike
+  previous releases).
 
-- Issue #7466: segmentation fault when the garbage collector is called
-  in the middle of populating a tuple.  Patch by Florent Xicluna.
+- Issue #7466: Segmentation fault when the garbage collector is called in the
+  middle of populating a tuple.  Patch by Florent Xicluna.
 
 - Issue #7419: setlocale() could crash the interpreter on Windows when called
   with invalid values.
@@ -292,22 +282,22 @@
 - Issue #6077: On Windows, files opened with tempfile.TemporaryFile in "wt+"
   mode would appear truncated on the first '0x1a' byte (aka. Ctrl+Z).
 
-- Issue #7085: Fix crash when importing some extensions in a thread
-  on MacOSX 10.6.
+- Issue #7085: Fix crash when importing some extensions in a thread on MacOSX
+  10.6.
 
 - Issue #1757126: Fix the cyrillic-asian alias for the ptcp154 encoding.
 
 - Issue #6970: Remove redundant calls when comparing objects that don't
   implement the relevant rich comparison methods.
 
-- Issue #7298: fixes for range and reversed(range(...)).  Iteration
-  over range(a, b, c) incorrectly gave an empty iterator when a, b and
-  c fit in C long but the length of the range did not.  Also fix
-  several cases where reversed(range(a, b, c)) gave wrong results, and
-  fix a refleak for reversed(range(a, b, c)) with large arguments.
+- Issue #7298: Fixes for range and reversed(range(...)).  Iteration over
+  range(a, b, c) incorrectly gave an empty iterator when a, b and c fit in C
+  long but the length of the range did not.  Also fix several cases where
+  reversed(range(a, b, c)) gave wrong results, and fix a refleak for
+  reversed(range(a, b, c)) with large arguments.
 
-- Issue #7244: itertools.izip_longest() no longer ignores exceptions
-  raised during the formation of an output tuple.
+- Issue #7244: itertools.izip_longest() no longer ignores exceptions raised
+  during the formation of an output tuple.
 
 - Issue #3297: On wide unicode builds, do not split unicode characters into
   surrogates.
@@ -324,7 +314,7 @@
   will be entirely removed in 3.3.
 
 - Support for OSF* has been disabled. If nobody stands up, support will be
-  removed in 3.3. See http://bugs.python.org/issue8606 .
+  removed in 3.3. See <http://bugs.python.org/issue8606>.
 
 - Peephole constant folding had missed UNARY_POSITIVE.
 
@@ -335,26 +325,26 @@
 - Issue #7147: Remove support for compiling Python without complex number
   support.
 
-- Issue #7120: logging: Removed import of multiprocessing which is causing
-  crash in GAE.
+- Issue #7120: logging: Removed import of multiprocessing which is causing crash
+  in GAE.
 
-- Issue #1754094: Improve the stack depth calculation in the compiler.
-  There should be no other effect than a small decrease in memory use.
-  Patch by Christopher Tur Lesniewski-Laas.
+- Issue #1754094: Improve the stack depth calculation in the compiler.  There
+  should be no other effect than a small decrease in memory use.  Patch by
+  Christopher Tur Lesniewski-Laas.
 
-- Issue #7065: Fix a crash in bytes.maketrans and bytearray.maketrans when
-  using byte values greater than 127.  Patch by Derk Drukker.
+- Issue #7065: Fix a crash in bytes.maketrans and bytearray.maketrans when using
+  byte values greater than 127.  Patch by Derk Drukker.
 
 - Issue #1571184: The Unicode database contains properties for more characters.
   The tables for code points representing numeric values, white spaces or line
   breaks are now generated from the official Unicode Character Database files,
   and include information from the Unihan.txt file.
 
-- Issue #7019: Raise ValueError when unmarshalling bad long data, instead
-  of producing internally inconsistent Python longs.
+- Issue #7019: Raise ValueError when unmarshalling bad long data, instead of
+  producing internally inconsistent Python longs.
 
-- Issue #6990: Fix threading.local subclasses leaving old state around
-  after a reference cycle GC which could be recycled by new locals.
+- Issue #6990: Fix threading.local subclasses leaving old state around after a
+  reference cycle GC which could be recycled by new locals.
 
 - Issue #5460: Fix an ambiguity in the grammar.
 
@@ -364,8 +354,8 @@
 
 - Issue #6846: Fix bug where bytearray.pop() returns negative integers.
 
-- Issue #6750: A text file opened with io.open() could duplicate its output
-  when writing from multiple threads at the same time.
+- Issue #6750: A text file opened with io.open() could duplicate its output when
+  writing from multiple threads at the same time.
 
 - Issue #6707: dir() on an uninitialized module caused a crash.
 
@@ -374,85 +364,81 @@
 - Issue #6573: set.union() stopped processing inputs if an instance of self
   occurred in the argument chain.
 
-- Issue #6070: On posix platforms import no longer copies the execute bit
-  from the .py file to the .pyc file if it is set.
+- Issue #6070: On posix platforms import no longer copies the execute bit from
+  the .py file to the .pyc file if it is set.
 
 - Issue #1616979: Added the cp720 (Arabic DOS) encoding.
 
-- Issue #6428: Since Python 3.0, the __bool__ method must return a bool
-  object, and not an int.  Fix the corresponding error message, and the
-  documentation.
+- Issue #6428: Since Python 3.0, the __bool__ method must return a bool object,
+  and not an int.  Fix the corresponding error message, and the documentation.
 
 - The deprecated PyCObject has been removed.
 
-- Issue #6347: Include inttypes.h as well as stdint.h in pyport.h.
-  This fixes a build failure on HP-UX: int32_t and uint32_t are
-  defined in inttypes.h instead of stdint.h on that platform.
+- Issue #6347: Include inttypes.h as well as stdint.h in pyport.h.  This fixes a
+  build failure on HP-UX: int32_t and uint32_t are defined in inttypes.h instead
+  of stdint.h on that platform.
 
-- Issue #6373: Fixed a SystemError when encoding with the latin-1 codec and
-  the 'surrogateescape' error handler, a string which contains unpaired
-  surrogates.
+- Issue #6373: Fixed a SystemError when encoding with the latin-1 codec and the
+  'surrogateescape' error handler, a string which contains unpaired surrogates.
 
 - Issue #4856: Remove checks for win NT.
 
-- Issue #6687: PyBytes_FromObject() no longer accepts an integer as its
-  argument to construct a null-initialized bytes object.
+- Issue #6687: PyBytes_FromObject() no longer accepts an integer as its argument
+  to construct a null-initialized bytes object.
 
-- Issue #1023290: Add from_bytes() and to_bytes() methods to integers.
-  These methods allow the conversion of integers to bytes, and vice-versa.
+- Issue #1023290: Add from_bytes() and to_bytes() methods to integers.  These
+  methods allow the conversion of integers to bytes, and vice-versa.
 
-- Issue #7382: Fix bug in bytes.__getnewargs__ that prevented bytes
-  instances from being copied with copy.copy(), and bytes subclasses
-  from being pickled properly.
+- Issue #7382: Fix bug in bytes.__getnewargs__ that prevented bytes instances
+  from being copied with copy.copy(), and bytes subclasses from being pickled
+  properly.
 
 - Code objects now support weak references.
 
-- Issue #7072: isspace(0xa0) is true on Mac OS X
+- Issue #7072: isspace(0xa0) is true on Mac OS X.
 
-- Issue #8084: PEP 370 now conforms to system conventions for framework
-  builds on MacOS X. That is, "python setup.py install --user" will install
+- Issue #8084: PEP 370 now conforms to system conventions for framework builds
+  on MacOS X. That is, "python setup.py install --user" will install
     into "~/Library/Python/2.7" instead of "~/.local".
 
 C-API
 -----
 
-- Issue #5753: A new C API function, :cfunc:`PySys_SetArgvEx`, allows
-  embedders of the interpreter to set sys.argv without also modifying
-  sys.path.  This helps fix `CVE-2008-5983
+- Issue #5753: A new C API function, `PySys_SetArgvEx`, allows embedders of the
+  interpreter to set sys.argv without also modifying sys.path.  This helps fix
+  `CVE-2008-5983
   <http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2008-5983>`_.
 
 - Add PyArg_ValidateKeywordArguments, which checks if all keyword arguments are
   strings in an efficient manner.
 
 - Issue #8276: PyEval_CallObject() is now only available in macro form.  The
-  function declaration, which was kept for backwards compatibility reasons,
-  is now removed (the macro was introduced in 1997!).
+  function declaration, which was kept for backwards compatibility reasons, is
+  now removed (the macro was introduced in 1997!).
 
-- Issue #7767: New function PyLong_AsLongLongAndOverflow added,
-  analogous to PyLong_AsLongAndOverflow.
+- Issue #7767: New function PyLong_AsLongLongAndOverflow added, analogous to
+  PyLong_AsLongAndOverflow.
 
 - Make PyUnicode_CompareWithASCIIString return not equal if the Python string
   has '\0' at the end.
 
 - Issue #5080: The argument parsing functions PyArg_ParseTuple,
-  PyArg_ParseTupleAndKeywords, PyArg_VaParse,
-  PyArg_VaParseTupleAndKeywords and PyArg_Parse now raise a
-  DeprecationWarning for float arguments passed with the 'L' format
-  code.  This will become a TypeError in a future version of Python,
+  PyArg_ParseTupleAndKeywords, PyArg_VaParse, PyArg_VaParseTupleAndKeywords and
+  PyArg_Parse now raise a DeprecationWarning for float arguments passed with the
+  'L' format code.  This will become a TypeError in a future version of Python,
   to match the behaviour of the other integer format codes.
 
-- Issue #7033: function ``PyErr_NewExceptionWithDoc()`` added.
+- Issue #7033: Function ``PyErr_NewExceptionWithDoc()`` added.
 
-- Issue #7414: 'C' code wasn't being skipped properly (for keyword arguments)
-  in PyArg_ParseTupleAndKeywords.
+- Issue #7414: 'C' code wasn't being skipped properly (for keyword arguments) in
+  PyArg_ParseTupleAndKeywords.
 
-- Issue #7228: Add '%lld' and '%llu' support to PyString_FromFormat(V)
-  and PyErr_Format, on machines with HAVE_LONG_LONG defined.
+- Issue #7228: Add '%lld' and '%llu' support to PyString_FromFormat(V) and
+  PyErr_Format, on machines with HAVE_LONG_LONG defined.
 
-- Issue #6151: Made PyDescr_COMMON conform to standard C (like PyObject_HEAD
-  in PEP 3123).  The PyDescr_TYPE and PyDescr_NAME macros should be
-  should used for accessing the d_type and d_name members of structures
-  using PyDescr_COMMON.
+- Issue #6151: Made PyDescr_COMMON conform to standard C (like PyObject_HEAD in
+  PEP 3123).  The PyDescr_TYPE and PyDescr_NAME macros should be should used for
+  accessing the d_type and d_name members of structures using PyDescr_COMMON.
 
 - Issue #6405: Remove duplicate type declarations in descrobject.h.
 
@@ -467,88 +453,129 @@
 - Issue #1419652: Change the first argument to PyImport_AppendInittab() to
   ``const char *`` as the string is stored beyond the call.
 
-- Issue #2422: When compiled with the ``--with-valgrind`` option, the
-  pymalloc allocator will be automatically disabled when running under
-  Valgrind.  This gives improved memory leak detection when running
-  under Valgrind, while taking advantage of pymalloc at other times.
+- Issue #2422: When compiled with the ``--with-valgrind`` option, the pymalloc
+  allocator will be automatically disabled when running under Valgrind.  This
+  gives improved memory leak detection when running under Valgrind, while taking
+  advantage of pymalloc at other times.
 
 Library
 -------
 
+- In pdb, when Ctrl-C is entered while defining commands for a breakpoint, the
+  old commands are restored.
+
+- For traceback debugging, the pdb listing now also shows the locations where
+  the exception was originally (re)raised, if it differs from the last line
+  executed (e.g. in case of finally clauses).
+
+- The pdb command "source" has been added.  It displays the source code for a
+  given object, if possible.
+
+- The pdb command "longlist" has been added.  It displays the whole source code
+  for the current function.
+
+- Issue #1503502: Make pdb.Pdb easier to subclass by putting message and error
+  output into methods.
+
+- Issue #809887: Make the output of pdb's breakpoint deletions more consistent;
+  emit a message when a breakpoint is enabled or disabled.
+
+- Issue #5294: Fix the behavior of pdb's "continue" command when called in the
+  top-level debugged frame.
+
+- Issue #5727: Restore the ability to use readline when calling into pdb in
+  doctests.
+
+- Issue #6719: In pdb, do not stop somewhere in the encodings machinery if the
+  source file to be debugged is in a non-builtin encoding.
+
+- Issue #8048: Prevent doctests from failing when sys.displayhook has been
+  reassigned.
+
+- Issue #8015: In pdb, do not crash when an empty line is entered as a
+  breakpoint command.
+
+- In pdb, allow giving a line number to the "until" command.
+
+- Issue #1437051: For pdb, allow "continue" and related commands in .pdbrc
+  files.  Also, add a command-line option "-c" that runs a command as if given
+  in .pdbrc.
+
+- Issue #4179: In pdb, allow "list ." as a command to return to the currently
+  debugged line.
+
 - Issue #4108: In urllib.robotparser, if there are multiple 'User-agent: *'
   entries, consider the first one.
 
 - Issue #6630: Allow customizing regex flags when subclassing the
   string.Template class.
 
-- Issue #9411: Allow specifying an encoding for config files in the
-  configparser module.
+- Issue #9411: Allow specifying an encoding for config files in the configparser
+  module.
 
-- Issue #1682942: Improvements to configparser: support alternate
-  delimiters, alternate comment prefixes and empty lines in values.
+- Issue #1682942: Improvements to configparser: support alternate delimiters,
+  alternate comment prefixes and empty lines in values.
 
 - Issue #9354: Provide getsockopt() in asyncore's file_wrapper.
 
 - Issue #8966: ctypes: Remove implicit bytes-unicode conversion.
 
-- Issue #9378: python -m pickle <pickle file> will now load and
-  display the first object in the pickle file.
+- Issue #9378: python -m pickle <pickle file> will now load and display the
+  first object in the pickle file.
 
 - Issue #4770: Restrict binascii module to accept only bytes (as specified).
-  And fix the email package to encode to ASCII instead of
-  ``raw-unicode-escape`` before ASCII-to-binary decoding.
+  And fix the email package to encode to ASCII instead of ``raw-unicode-escape``
+  before ASCII-to-binary decoding.
 
-- Issue #9384: python -m tkinter will now display a simple demo applet.
+- Issue #9384: ``python -m tkinter`` will now display a simple demo applet.
 
-- The default size of the re module's compiled regular expression cache has
-  been increased from 100 to 500 and the cache replacement policy has changed
-  from simply clearing the entire cache on overflow to randomly forgetting 20%
-  of the existing cached compiled regular expressions.  This is a performance
-  win for applications that use a lot of regular expressions and limits the
-  impact of the performance hit anytime the cache is exceeded.
+- The default size of the re module's compiled regular expression cache has been
+  increased from 100 to 500 and the cache replacement policy has changed from
+  simply clearing the entire cache on overflow to randomly forgetting 20% of the
+  existing cached compiled regular expressions.  This is a performance win for
+  applications that use a lot of regular expressions and limits the impact of
+  the performance hit anytime the cache is exceeded.
 
 - Issue #7113: Speed up loading in configparser. Patch by ??ukasz Langa.
 
-- Issue #9032: XML-RPC client retries the request on EPIPE error. The EPIPE
+- Issue #9032: XML-RPC client retries the request on EPIPE error.  The EPIPE
   error occurs when the server closes the socket and the client sends a big
   XML-RPC request.
 
-- Issue #4629: getopt raises an error if an argument ends with = whereas getopt
-  doesn't except a value (eg. --help= is rejected if getopt uses ['help='] long
-  options).
-
-- Issue #7989: Added pure python implementation of the `datetime`
-  module.  The C module is renamed to `_datetime` and if available,
-  overrides all classes defined in datetime with fast C impementation.
-  Python implementation is based on the original python prototype for
-  the datetime module by Tim Peters with minor modifications by the
-  PyPy project.  The test suite now tests `datetime` module with and
-  without `_datetime` acceleration using the same test cases.
+- Issue #4629: getopt raises an error if an argument ends with "=", whereas
+  getopt doesn't accept a value (eg. --help= is rejected if getopt uses
+  ['help='] long options).
+
+- Issue #7989: Added pure python implementation of the `datetime` module.  The C
+  module is renamed to `_datetime` and if available, overrides all classes
+  defined in datetime with fast C impementation.  Python implementation is based
+  on the original python prototype for the datetime module by Tim Peters with
+  minor modifications by the PyPy project.  The test suite now tests `datetime`
+  module with and without `_datetime` acceleration using the same test cases.
 
-- Issue #7895: platform.mac_ver() no longer crashes after calling os.fork()
+- Issue #7895: platform.mac_ver() no longer crashes after calling os.fork().
 
-- Issue #9323: Fixed a bug in trace.py that resulted in loosing the
-  name of the script being traced.  Patch by Eli Bendersky.
+- Issue #9323: Fixed a bug in trace.py that resulted in loosing the name of the
+  script being traced.  Patch by Eli Bendersky.
 
-- Issue #9282: Fixed --listfuncs option of trace.py.  Thanks Eli
-  Bendersky for the patch.
+- Issue #9282: Fixed --listfuncs option of trace.py.  Thanks Eli Bendersky for
+  the patch.
 
 - Issue #3704: http.cookiejar was not properly handling URLs with a / in the
   parameters.
 
-- Issue #9268: ``pickletools.dis()`` now has an optional *annotate*
-  argument which controls printing of opcode descriptions in ``dis()``
-  output.
+- Issue #9268: ``pickletools.dis()`` now has an optional *annotate* argument
+  which controls printing of opcode descriptions in ``dis()`` output.
 
-- Issue #1555570: email no longer inserts extra blank lines when a \r\n
-  combo crosses an 8192 byte boundary.
+- Issue #1555570: email no longer inserts extra blank lines when a \r\n combo
+  crosses an 8192 byte boundary.
 
 - Issue #9243: Fix sndhdr module and add unit tests, contributed by James Lee.
 
 - ``ast.literal_eval()`` now allows byte literals.
 
-- Issue #9137: Fix issue in MutableMapping.update, which incorrectly
-  treated keyword arguments called 'self' or 'other' specially.
+- Issue #9137: Fix issue in MutableMapping.update, which incorrectly treated
+  keyword arguments called 'self' or 'other' specially.
 
 - ``ast.literal_eval()`` now allows set literals.
 
@@ -556,13 +583,13 @@
 
 - Issue #7646: The fnmatch pattern cache no longer grows without bound.
 
-- Issue #9136: Fix 'dictionary changed size during iteration'
-  RuntimeError produced when profiling the decimal module.  This was
-  due to a dangerous iteration over 'locals()' in Context.__init__.
-
-- Fix extreme speed issue in Decimal.pow when the base is an exact
-  power of 10 and the exponent is tiny (for example,
-  Decimal(10) ** Decimal('1e-999999999')).
+- Issue #9136: Fix 'dictionary changed size during iteration' RuntimeError
+  produced when profiling the decimal module.  This was due to a dangerous
+  iteration over 'locals()' in Context.__init__.
+
+- Fix extreme speed issue in Decimal.pow when the base is an exact power of 10
+  and the exponent is tiny (for example, ``Decimal(10) **
+  Decimal('1e-999999999')``).
 
 - Issue #9186: Fix math.log1p(-1.0) to raise ValueError, not OverflowError.
 
@@ -570,18 +597,18 @@
 
 - Issue #9128: Fix validation of class decorators in parser module.
 
-- Issue #9094: python -m pickletools will now disassemble pickle files
-  listed in the command line arguments.  See output of python -m
-  pickletools -h for more details.
+- Issue #9094: python -m pickletools will now disassemble pickle files listed in
+  the command line arguments.  See output of python -m pickletools -h for more
+  details.
 
 - Issue #5468: urlencode to handle bytes type and other encodings in its query
   parameter. Patch by Dan Mahn.
 
-- Issue #7673: Fix security vulnerability (CVE-2010-2089) in the audioop
-  module, ensure that the input string length is a multiple of the frame size
+- Issue #7673: Fix security vulnerability (CVE-2010-2089) in the audioop module,
+  ensure that the input string length is a multiple of the frame size.
 
-- Issue #6507: Accept source strings in dis.dis(). Original patch by
-  Daniel Urban.
+- Issue #6507: Accept source strings in dis.dis().  Original patch by Daniel
+  Urban.
 
 - Issue #7829: Clearly document that the dis module is exposing an
   implementation detail that is not stable between Python VMs or releases.
@@ -590,73 +617,72 @@
   raises an exception.
 
 - Issue #9110: Addition of ContextDecorator to contextlib, for creating APIs
-  that act as both context managers and decorators. contextmanager changes
-  to use ContextDecorator.
+  that act as both context managers and decorators. contextmanager changes to
+  use ContextDecorator.
 
 - Implement importlib.abc.SourceLoader and deprecate PyLoader and PyPycLoader
   for removal in Python 3.4.
 
-- Issue #9064: pdb's "up" and "down" commands now accept an optional argument.
+- Issue #9064: pdb's "up" and "down" commands now accept an optional argument
+  giving the number of frames to go.
 
-- Issue #9018: os.path.normcase() now raises a TypeError if the argument is
-  not ``str`` or ``bytes``.
+- Issue #9018: os.path.normcase() now raises a TypeError if the argument is not
+  ``str`` or ``bytes``.
 
-- Issue #9075: In the ssl module, remove the setting of a ``debug`` flag
-  on an OpenSSL structure.
+- Issue #9075: In the ssl module, remove the setting of a ``debug`` flag on an
+  OpenSSL structure.
 
-- Issue #8682: The ssl module now temporary increments the reference count of
-  a socket object got through ``PyWeakref_GetObject``, so as to avoid possible
+- Issue #8682: The ssl module now temporary increments the reference count of a
+  socket object got through ``PyWeakref_GetObject``, so as to avoid possible
   deallocation while the object is still being used.
 
 - Issue #1368368: FancyURLOpener class changed to throw an Exception on wrong
-  password instead of presenting an interactive prompt. Older behavior can be
+  password instead of presenting an interactive prompt.  Older behavior can be
   obtained by passing retry=True to http_error_xxx methods of FancyURLOpener.
 
-- Issue #8720: fix regression caused by fix for #4050 by making getsourcefile
+- Issue #8720: Fix regression caused by fix for #4050 by making getsourcefile
   smart enough to find source files in the linecache.
 
-- Issue #5610: feedparser no longer eats extra characters at the end of
-  a body part if the body part ends with a \r\n.
+- Issue #5610: feedparser no longer eats extra characters at the end of a body
+  part if the body part ends with a ``\r\n``.
 
-- Issue #8986: math.erfc was incorrectly raising OverflowError for
-  values between -27.3 and -30.0 on some platforms.
+- Issue #8986: math.erfc was incorrectly raising OverflowError for values
+  between -27.3 and -30.0 on some platforms.
 
 - Issue #8784: Set tarfile default encoding to 'utf-8' on Windows.
 
 - Issue #8966: If a ctypes structure field is an array of c_char, convert its
   value to bytes instead of str (as done for c_char and c_char_p).
 
-- Issue #8188: Comparisons between Decimal and Fraction objects are
-  now permitted, returning a result based on the exact numerical
-  values of the operands.  This builds on issue #2531, which allowed
-  Decimal-to-float comparisons;  all comparisons involving numeric
-  types (bool, int, float, complex, Decimal, Fraction) should now
-  act as expected.
+- Issue #8188: Comparisons between Decimal and Fraction objects are now
+  permitted, returning a result based on the exact numerical values of the
+  operands.  This builds on issue #2531, which allowed Decimal-to-float
+  comparisons; all comparisons involving numeric types (bool, int, float,
+  complex, Decimal, Fraction) should now act as expected.
 
-- Issue #8897: Fix sunau module, use bytes to write the header. Patch written
-  by Thomas Jollans.
+- Issue #8897: Fix sunau module, use bytes to write the header. Patch written by
+  Thomas Jollans.
 
 - Issue #8899: time.struct_time now has class and atribute docstrings.
 
 - Issue #6470: Drop UNC prefix in FixTk.
 
-- Issue #4768: base64 encoded email body parts were incorrectly stored as
-  binary strings.  They are now correctly converted to strings.
+- Issue #4768: base64 encoded email body parts were incorrectly stored as binary
+  strings.  They are now correctly converted to strings.
 
 - Issue #8833: tarfile created hard link entries with a size field != 0 by
   mistake.
 
-- Charset.body_encode now correctly handles base64 encoding by encoding
-  with the output_charset before calling base64mime.encode.  Passes the
-  tests from 2.x issue 1368247.
+- Charset.body_encode now correctly handles base64 encoding by encoding with the
+  output_charset before calling base64mime.encode.  Passes the tests from 2.x
+  issue 1368247.
 
 - Issue #8845: sqlite3 Connection objects now have a read-only in_transaction
   attribute that is True iff there are uncommitted changes.
 
-- Issue #1289118: datetime.timedelta objects can now be multiplied by float
-  and divided by float and int objects.  Results are rounded to the nearest
-  multiple of timedelta.resolution with ties resolved using round-half-to-even
-  method.
+- Issue #1289118: datetime.timedelta objects can now be multiplied by float and
+  divided by float and int objects.  Results are rounded to the nearest multiple
+  of timedelta.resolution with ties resolved using round-half-to-even method.
 
 - Issue #7150: Raise OverflowError if the result of adding or subtracting
   timedelta from date or datetime falls outside of the MINYEAR:MAXYEAR range.
@@ -665,25 +691,24 @@
 
 - Issue #4769: Fix main() function of the base64 module, use sys.stdin.buffer
   and sys.stdout.buffer (instead of sys.stdin and sys.stdout) to use the bytes
-  API
+  API.
 
-- Issue #8770: now sysconfig displays information when it's called as
-  a script. Initial idea by Sridhar Ratnakumar.
+- Issue #8770: Now sysconfig displays information when it's called as a script.
+  Initial idea by Sridhar Ratnakumar.
 
 - Issue #6662: Fix parsing of malformatted charref (&#bad;), patch written by
-  Fredrik H????rd
+  Fredrik H????rd.
 
 - Issue #8540: Decimal module: rename the Context._clamp attribute to
-  Context.clamp and make it public.  This is useful in creating
-  contexts that correspond to the decimal interchange formats
-  specified in IEEE 754.
+  Context.clamp and make it public.  This is useful in creating contexts that
+  correspond to the decimal interchange formats specified in IEEE 754.
 
 - Issue #6268: Fix seek() method of codecs.open(), don't read or write the BOM
   twice after seek(0). Fix also reset() method of codecs, UTF-16, UTF-32 and
   StreamWriter classes.
 
 - Issue #3798: sys.exit(message) writes the message to sys.stderr file, instead
-  of the C file stderr, to use stderr encoding and error handler
+  of the C file stderr, to use stderr encoding and error handler.
 
 - Issue #8782: Add a trailing newline in linecache.updatecache to the last line
   of files without one.
@@ -692,11 +717,11 @@
   comparing to a non-mapping.
 
 - Issue #8774: tabnanny uses the encoding cookie (#coding:...) to use the
-  correct encoding
+  correct encoding.
 
-- Issue #4870: Add an `options` attribute to SSL contexts, as well as
-  several ``OP_*`` constants to the `ssl` module.  This allows to selectively
-  disable protocol versions, when used in combination with `PROTOCOL_SSLv23`.
+- Issue #4870: Add an `options` attribute to SSL contexts, as well as several
+  ``OP_*`` constants to the `ssl` module.  This allows to selectively disable
+  protocol versions, when used in combination with `PROTOCOL_SSLv23`.
 
 - Issue #8759: Fixed user paths in sysconfig for posix and os2 schemes.
 
@@ -718,15 +743,15 @@
 - Issue #8688: Distutils now recalculates MANIFEST everytime.
 
 - Issue #8477: ssl.RAND_egd() and ssl._test_decode_cert() support str with
-  surrogates and bytes for the filename
+  surrogates and bytes for the filename.
 
 - Issue #8550: Add first class ``SSLContext`` objects to the ssl module.
 
-- Issue #8681: Make the zlib module's error messages more informative when
-  the zlib itself doesn't give any detailed explanation.
+- Issue #8681: Make the zlib module's error messages more informative when the
+  zlib itself doesn't give any detailed explanation.
 
-- The audioop module now supports sound fragments of length greater
-  than 2**31 bytes on 64-bit machines, and is PY_SSIZE_T_CLEAN.
+- The audioop module now supports sound fragments of length greater than 2**31
+  bytes on 64-bit machines, and is PY_SSIZE_T_CLEAN.
 
 - Issue #4972: Add support for the context manager protocol to the ftplib.FTP
   class.
@@ -737,9 +762,8 @@
 - Issue #8514: Add os.fsencode() function (Unix only): encode a string to bytes
   for use in the file system, environment variables or the command line.
 
-- Issue #8571: Fix an internal error when compressing or decompressing a
-  chunk larger than 1GB with the zlib module's compressor and decompressor
-  objects.
+- Issue #8571: Fix an internal error when compressing or decompressing a chunk
+  larger than 1GB with the zlib module's compressor and decompressor objects.
 
 - Issue #8603: Support bytes environmental variables on Unix: Add os.environb
   mapping and os.getenvb() function. os.unsetenv() encodes str argument to the
@@ -748,46 +772,46 @@
 
 - Issue #8573: asyncore _strerror() function might throw ValueError.
 
-- Issue #8483: asyncore.dispatcher's __getattr__ method produced confusing
-  error messages when accessing undefined class attributes because of the cheap
-  inheritance with the underlying socket object.
-  The cheap inheritance has been deprecated.
+- Issue #8483: asyncore.dispatcher's __getattr__ method produced confusing error
+  messages when accessing undefined class attributes because of the cheap
+  inheritance with the underlying socket object.  The cheap inheritance has been
+  deprecated.
 
 - Issue #4265: shutil.copyfile() was leaking file descriptors when disk fills.
   Patch by Tres Seaver.
 
 - Issue #8390: tarfile uses surrogateespace as the default error handler
-  (instead of replace in read mode or strict in write mode)
+  (instead of replace in read mode or strict in write mode).
 
 - Issue #7755: Use an unencumbered audio file for tests.
 
-- Issue #8621: uuid.uuid4() returned the same sequence of values in the
-  parent and any children created using ``os.fork`` on MacOS X 10.6.
+- Issue #8621: uuid.uuid4() returned the same sequence of values in the parent
+  and any children created using ``os.fork`` on MacOS X 10.6.
 
-- Issue #8567: Fix precedence of signals in Decimal module: when a
-  Decimal operation raises multiple signals and more than one of those
-  signals is trapped, the specification determines the order in which
-  the signals should be handled.  In many cases this order wasn't
-  being followed, leading to the wrong Python exception being raised.
+- Issue #8567: Fix precedence of signals in Decimal module: when a Decimal
+  operation raises multiple signals and more than one of those signals is
+  trapped, the specification determines the order in which the signals should be
+  handled.  In many cases this order wasn't being followed, leading to the wrong
+  Python exception being raised.
 
 - Issue #7865: The close() method of :mod:`io` objects should not swallow
-  exceptions raised by the implicit flush().  Also ensure that calling
-  close() several times is supported.  Patch by Pascal Chambon.
+  exceptions raised by the implicit flush().  Also qensure that calling close()
+  several times is supported.  Patch by Pascal Chambon.
 
 - Issue #4687: Fix accuracy of garbage collection runtimes displayed with
   gc.DEBUG_STATS.
 
-- Issue #8354: The siginterrupt setting is now preserved for all signals,
-  not just SIGCHLD.
+- Issue #8354: The siginterrupt setting is now preserved for all signals, not
+  just SIGCHLD.
 
 - Issue #7192: webbrowser.get("firefox") now works on Mac OS X, as does
   webbrowser.get("safari").
 
-- Issue #8464: tarfile no longer creates files with execute permissions set
-  when mode="w|" is used.
+- Issue #8464: tarfile no longer creates files with execute permissions set when
+  mode="w|" is used.
 
-- Issue #7834: Fix connect() of Bluetooth L2CAP sockets with recent versions
-  of the Linux kernel.  Patch by Yaniv Aknin.
+- Issue #7834: Fix connect() of Bluetooth L2CAP sockets with recent versions of
+  the Linux kernel.  Patch by Yaniv Aknin.
 
 - Issue #8295: Added shutil.unpack_archive.
 
@@ -802,36 +826,36 @@
 - Issue #6656: fix locale.format_string to handle escaped percents
   and mappings.
 
-- Issue #2302: Fix a race condition in SocketServer.BaseServer.shutdown,
-  where the method could block indefinitely if called just before the
-  event loop started running.  This also fixes the occasional freezes
-  witnessed in test_httpservers.
-
-- Issue #8524: When creating an SSL socket, the timeout value of the
-  original socket wasn't retained (instead, a socket with a positive timeout
-  would be turned into a non-blocking SSL socket).
+- Issue #2302: Fix a race condition in SocketServer.BaseServer.shutdown, where
+  the method could block indefinitely if called just before the event loop
+  started running.  This also fixes the occasional freezes witnessed in
+  test_httpservers.
+
+- Issue #8524: When creating an SSL socket, the timeout value of the original
+  socket wasn't retained (instead, a socket with a positive timeout would be
+  turned into a non-blocking SSL socket).
 
 - Issue #5103: SSL handshake would ignore the socket timeout and block
   indefinitely if the other end didn't respond.
 
-- The do_handshake() method of SSL objects now adjusts the blocking mode of
-  the SSL structure if necessary (as other methods already do).
+- The do_handshake() method of SSL objects now adjusts the blocking mode of the
+  SSL structure if necessary (as other methods already do).
 
-- Issue #8391: os.execvpe() and os.getenv() supports unicode with surrogates
-  and bytes strings for environment keys and values
+- Issue #8391: os.execvpe() and os.getenv() supports unicode with surrogates and
+  bytes strings for environment keys and values.
 
 - Issue #8467: Pure Python implementation of subprocess encodes the error
   message using surrogatepass error handler to support surrogates in the
-  message
+  message.
 
-- Issue #8468: bz2.BZ2File() accepts str with surrogates and bytes filenames
+- Issue #8468: bz2.BZ2File() accepts str with surrogates and bytes filenames.
 
-- Issue #8451: Syslog module now uses basename(sys.argv[0]) instead of
-  the string "python" as the *ident*.  openlog() arguments are all optional
-  and keywords.
+- Issue #8451: Syslog module now uses basename(sys.argv[0]) instead of the
+  string "python" as the *ident*.  openlog() arguments are all optional and
+  keywords.
 
-- Issue #8108: Fix the unwrap() method of SSL objects when the socket has
-  a non-infinite timeout.  Also make that method friendlier with applications
+- Issue #8108: Fix the unwrap() method of SSL objects when the socket has a
+  non-infinite timeout.  Also make that method friendlier with applications
   wanting to continue using the socket in clear-text mode, by disabling
   OpenSSL's internal readahead.  Thanks to Darryl Miles for guidance.
 
@@ -841,25 +865,24 @@
 - Issue #8195: Fix a crash in sqlite Connection.create_collation() if the
   collation name contains a surrogate character.
 
-- Issue #8484: Load all ciphers and digest algorithms when initializing
-  the _ssl extension, such that verification of some SSL certificates
-  doesn't fail because of an "unknown algorithm".
+- Issue #8484: Load all ciphers and digest algorithms when initializing the _ssl
+  extension, such that verification of some SSL certificates doesn't fail
+  because of an "unknown algorithm".
 
 - Issue #6547: Added the ignore_dangling_symlinks option to shutil.copytree.
 
-- Issue #1540112: Now allowing the choice of a copy function in
-  shutil.copytree.
+- Issue #1540112: Now allowing the choice of a copy function in shutil.copytree.
 
 - Issue #4814: timeout parameter is now applied also for connections resulting
   from PORT/EPRT commands.
 
 - Issue #8463: added missing reference to bztar in shutil's documentation.
 
-- Issue #7154: urllib.request can now detect the proxy settings on OSX 10.6
-  (as long as the user didn't specify 'automatic proxy configuration').
+- Issue #7154: urllib.request can now detect the proxy settings on OSX 10.6 (as
+  long as the user didn't specify 'automatic proxy configuration').
 
-- Issue #3817: ftplib.FTP.abort() method now considers 225 a valid response
-  code as stated in RFC-959 at chapter 5.4.
+- Issue #3817: ftplib.FTP.abort() method now considers 225 a valid response code
+  as stated in RFC-959 at chapter 5.4.
 
 - Issue #8394: _ctypes.dlopen() accepts bytes, bytearray and str with
   surrogates.
@@ -874,95 +897,92 @@
   the current working directory.
 
 - Issue #7606: XML-RPC traceback stored in X-traceback is now encoded to ASCII
-  using backslashreplace error handler
+  using backslashreplace error handler.
 
-- Issue #8412: os.system() now accepts bytes, bytearray and str with
-  surrogates.
+- Issue #8412: os.system() now accepts bytes, bytearray and str with surrogates.
 
 - Issue #2987: RFC2732 support for urlparse (IPv6 addresses). Patch by Tony
   Locke and Hans Ulrich Niedermann.
 
 - Issue #5277: Fix quote counting when parsing RFC 2231 encoded parameters.
 
-- Issue #7316: the acquire() method of lock objects in the :mod:`threading`
+- Issue #7316: The acquire() method of lock objects in the :mod:`threading`
   module now takes an optional timeout argument in seconds.  Timeout support
-  relies on the system threading library, so as to avoid a semi-busy wait
-  loop.
+  relies on the system threading library, so as to avoid a semi-busy wait loop.
 
 - Issue #8383: pickle and pickletools use surrogatepass error handler when
   encoding unicode as utf8 to support lone surrogates and stay compatible with
-  Python 2.x and 3.0
+  Python 2.x and 3.x.
 
 - Issue #7585: difflib context and unified diffs now place a tab between
-  filename and date, conforming to the 'standards' they were originally
-  designed to follow.  This improves compatibility with patch tools.
+  filename and date, conforming to the 'standards' they were originally designed
+  to follow.  This improves compatibility with patch tools.
 
 - Issue #7472: Fixed typo in email.encoders module; messages using ISO-2022
-  character sets will now consistently use a Content-Transfer-Encoding of
-  7bit rather than sometimes being marked as 8bit.
+  character sets will now consistently use a Content-Transfer-Encoding of 7bit
+  rather than sometimes being marked as 8bit.
 
 - Issue #8375: test_distutils now checks if the temporary directory are still
   present before it cleans them.
 
-- Issue #8374: Update the internal alias table in the :mod:`locale` module
-  to cover recent locale changes and additions.
+- Issue #8374: Update the internal alias table in the :mod:`locale` module to
+  cover recent locale changes and additions.
 
 - Issue #8321: Give access to OpenSSL version numbers from the `ssl` module,
-  using the new attributes `ssl.OPENSSL_VERSION`, `ssl.OPENSSL_VERSION_INFO`
-  and `ssl.OPENSSL_VERSION_NUMBER`.
+  using the new attributes `ssl.OPENSSL_VERSION`, `ssl.OPENSSL_VERSION_INFO` and
+  `ssl.OPENSSL_VERSION_NUMBER`.
 
 - Add functools.total_ordering() and functools.cmp_to_key().
 
-- Issue #8257: The Decimal construct now accepts a float instance
-  directly, converting that float to a Decimal of equal value:
+- Issue #8257: The Decimal construct now accepts a float instance directly,
+  converting that float to a Decimal of equal value:
 
      >>> Decimal(1.1)
      Decimal('1.100000000000000088817841970012523233890533447265625')
 
-- Issue #8294: The Fraction constructor now accepts Decimal and float
-  instances directly.
+- Issue #8294: The Fraction constructor now accepts Decimal and float instances
+  directly.
 
-- Issue #7279: Comparisons involving a Decimal signaling NaN now
-  signal InvalidOperation instead of returning False.  (Comparisons
-  involving a quiet NaN are unchanged.)  Also, Decimal quiet NaNs
-  are now hashable;  Decimal signaling NaNs remain unhashable.
-
-- Issue #2531: Comparison operations between floats and Decimal
-  instances now return a result based on the numeric values of the
-  operands;  previously they returned an arbitrary result based on
-  the relative ordering of id(float) and id(Decimal).  See also
-  issue #8188, which adds Decimal-to-Fraction comparisons.
+- Issue #7279: Comparisons involving a Decimal signaling NaN now signal
+  InvalidOperation instead of returning False.  (Comparisons involving a quiet
+  NaN are unchanged.)  Also, Decimal quiet NaNs are now hashable; Decimal
+  signaling NaNs remain unhashable.
+
+- Issue #2531: Comparison operations between floats and Decimal instances now
+  return a result based on the numeric values of the operands; previously they
+  returned an arbitrary result based on the relative ordering of id(float) and
+  id(Decimal).  See also issue #8188, which adds Decimal-to-Fraction
+  comparisons.
 
 - Added a subtract() method to collections.Counter().
 
 - Issue #8233: When run as a script, py_compile.py optionally takes a single
-  argument `-` which tells it to read files to compile from stdin.  Each line
-  is read on demand and the named file is compiled immediately.  (Original
-  patch by Piotr O??arowski).
+  argument `-` which tells it to read files to compile from stdin.  Each line is
+  read on demand and the named file is compiled immediately.  (Original patch by
+  Piotr O??arowski).
 
 - Backwards incompatible change: Unicode codepoints line tabulation (0x0B) and
   form feed (0x0C) are now considered linebreaks, as specified in Unicode
-  Standard Annex #14.  See issue #7643.
-  http://www.unicode.org/reports/tr14/
+  Standard Annex #14.  See issue #7643.  http://www.unicode.org/reports/tr14/
 
-- Comparisons using one of <, <=, >, >= between a complex instance and
-  a Fractions instance now raise TypeError instead of returning
-  True/False.  This makes Fraction <=> complex comparisons consistent with
-  int <=> complex, float <=> complex, and complex <=> complex comparisons.
-
-- Issue #8139: ossaudiodev didn't initialize its types properly, therefore
-  some methods (such as oss_mixer_device.fileno()) were not available.
-  Initial patch by Bertrand Janin.
+- Comparisons using one of <, <=, >, >= between a complex instance and a
+  Fractions instance now raise TypeError instead of returning True/False.  This
+  makes Fraction <=> complex comparisons consistent with int <=> complex, float
+  <=> complex, and complex <=> complex comparisons.
+
+- Issue #8139: ossaudiodev didn't initialize its types properly, therefore some
+  methods (such as oss_mixer_device.fileno()) were not available.  Initial patch
+  by Bertrand Janin.
 
 - Issue #8205: Remove the "Modules" directory from sys.path when Python is
   running from the build directory (POSIX only).
 
 - Issue #7512: shutil.copystat() could raise an OSError when the filesystem
-  didn't support chflags() (for example ZFS under FreeBSD).  The error is
-  now silenced.
+  didn't support chflags() (for example ZFS under FreeBSD).  The error is now
+  silenced.
 
-- Issue #7860: platform.uname now reports the correct 'machine' type
-  when Python is running in WOW64 mode on 64 bit Windows.
+- Issue #7860: platform.uname now reports the correct 'machine' type when Python
+  is running in WOW64 mode on 64 bit Windows.
 
 - Issue #3890, #8222: Fix recv() and recv_into() on non-blocking SSL sockets.
   Also, enable the SSL_MODE_AUTO_RETRY flag on SSL sockets, so that blocking
@@ -986,31 +1006,31 @@
 - Issue #4961: Inconsistent/wrong result of askyesno function in tkMessageBox
   with Tcl/Tk-8.5.
 
-- Issue #8140: extend compileall to compile single files. Add -i option.
+- Issue #8140: extend compileall to compile single files.  Add -i option.
 
-- Issue #7356: ctypes.util: Make parsing of ldconfig output independent of
-  the locale.
+- Issue #7356: ctypes.util: Make parsing of ldconfig output independent of the
+  locale.
 
-- The internals of the subprocess module on POSIX systems have been replaced
-  by an extension module (_posixsubprocess) so that the fork()+exec() can be
-  done safely without the possibility of deadlock in multithreaded applications.
-
-- subprocess.Popen now has restore_signals and start_new_session features.
-  The default of restore_signals=True is a new behavior compared to earlier
-  Python versions.  This means that signals such as SIGPIPE are not ignored
-  by default in subprocesses launched by Python (Issue #1652).
+- The internals of the subprocess module on POSIX systems have been replaced by
+  an extension module (_posixsubprocess) so that the fork()+exec() can be done
+  safely without the possibility of deadlock in multithreaded applications.
+
+- subprocess.Popen now has restore_signals and start_new_session features.  The
+  default of restore_signals=True is a new behavior compared to earlier Python
+  versions.  This means that signals such as SIGPIPE are not ignored by default
+  in subprocesses launched by Python (Issue #1652).
 
 - Issue #6472: The xml.etree package is updated to ElementTree 1.3.  The
   cElementTree module is updated too.
 
 - Issue #7774: Set sys.executable to an empty string if argv[0] has been set to
-  an non existent program name and Python is unable to retrieve the real
-  program name
+  an non existent program name and Python is unable to retrieve the real program
+  name.
 
 - Issue #7880: Fix sysconfig when the python executable is a symbolic link.
 
-- Issue #6509: fix re.sub to work properly when the pattern, the string, and
-  the replacement were all bytes. Patch by Antoine Pitrou.
+- Issue #6509: fix re.sub to work properly when the pattern, the string, and the
+  replacement were all bytes.  Patch by Antoine Pitrou.
 
 - The sqlite3 module was updated to pysqlite 2.6.0. This fixes several obscure
   bugs and allows loading SQLite extensions from shared libraries.
@@ -1040,16 +1060,16 @@
 
 - Issue #1537721: Add a writeheader() method to csv.DictWriter.
 
-- Issue #7959: ctypes callback functions are now registered correctly
-  with the cycle garbage collector.
+- Issue #7959: ctypes callback functions are now registered correctly with the
+  cycle garbage collector.
 
 - Issue #5801: removed spurious empty lines in wsgiref.
 
-- Issue #6666: fix bug in trace.py that applied the list of directories
-  to be ignored only to the first file.  Noted by Bogdan Opanchuk.
+- Issue #6666: fix bug in trace.py that applied the list of directories to be
+  ignored only to the first file.  Noted by Bogdan Opanchuk.
 
-- Issue #7597: curses.use_env() can now be called before initscr().
-  Noted by Kan-Ru Chen.
+- Issue #7597: curses.use_env() can now be called before initscr().  Noted by
+  Kan-Ru Chen.
 
 - Issue #7310: fix the __repr__ of os.environ to show the environment variables.
 
@@ -1059,48 +1079,46 @@
 - Issue #7361: Importlib was not properly checking the number of bytes in
   bytecode file when it was less then 8 bytes.
 
-- Issue #7633: In the decimal module, Context class methods (with the
-  exception of canonical and is_canonical) now accept instances of int
-  and long wherever a Decimal instance is accepted, and implicitly
-  convert that argument to Decimal.  Previously only some arguments
-  were converted.
+- Issue #7633: In the decimal module, Context class methods (with the exception
+  of canonical and is_canonical) now accept instances of int and long wherever a
+  Decimal instance is accepted, and implicitly convert that argument to Decimal.
+  Previously only some arguments were converted.
 
 - Issue #7835: shelve should no longer produce mysterious warnings during
   interpreter shutdown.
 
-- Issue #2746: Don't escape ampersands and angle brackets ("&", "<", ">")
-  in XML processing instructions and comments.  These raw characters are
-  allowed by the XML specification, and are necessary when outputting e.g.
-  PHP code in a processing instruction.  Patch by Neil Muller.
-
-- Issue #6233: ElementTree failed converting unicode characters to XML
-  entities when they could't be represented in the requested output
-  encoding.  Patch by Jerry Chen.
+- Issue #2746: Don't escape ampersands and angle brackets ("&", "<", ">") in XML
+  processing instructions and comments.  These raw characters are allowed by the
+  XML specification, and are necessary when outputting e.g.  PHP code in a
+  processing instruction.  Patch by Neil Muller.
+
+- Issue #6233: ElementTree failed converting unicode characters to XML entities
+  when they could't be represented in the requested output encoding.  Patch by
+  Jerry Chen.
 
-- Issue #6003: add an argument to ``zipfile.Zipfile.writestr`` to
-  specify the compression type.
+- Issue #6003: add an argument to ``zipfile.Zipfile.writestr`` to specify the
+  compression type.
 
 - Issue #4772: Raise a ValueError when an unknown Bluetooth protocol is
   specified, rather than fall through to AF_PACKET (in the `socket` module).
-  Also, raise ValueError rather than TypeError when an unknown TIPC address
-  type is specified.  Patch by Brian Curtin.
+  Also, raise ValueError rather than TypeError when an unknown TIPC address type
+  is specified.  Patch by Brian Curtin.
 
-- Issue #6939: Fix file I/O objects in the `io` module to keep the original
-  file position when calling `truncate()`.  It would previously change the
-  file position to the given argument, which goes against the tradition of
+- Issue #6939: Fix file I/O objects in the `io` module to keep the original file
+  position when calling `truncate()`.  It would previously change the file
+  position to the given argument, which goes against the tradition of
   ftruncate() and other truncation APIs.  Patch by Pascal Chambon.
 
 - Issue #7610: Reworked implementation of the internal
-  :class:`zipfile.ZipExtFile` class used to represent files stored inside
-  an archive.  The new implementation is significantly faster and can
-  be wrapped in a :class:`io.BufferedReader` object for more speedups.
-  It also solves an issue where interleaved calls to `read()` and
-  `readline()` give wrong results.  Patch by Nir Aides.
+  :class:`zipfile.ZipExtFile` class used to represent files stored inside an
+  archive.  The new implementation is significantly faster and can be wrapped in
+  a :class:`io.BufferedReader` object for more speedups.  It also solves an
+  issue where interleaved calls to `read()` and `readline()` give wrong results.
+  Patch by Nir Aides.
 
 - Issue #6963: Added "maxtasksperchild" argument to multiprocessing.Pool,
-  allowing for a maximum number of tasks within the pool to be completed by
-  the worker before that worker is terminated, and a new one created to
-  replace it.
+  allowing for a maximum number of tasks within the pool to be completed by the
+  worker before that worker is terminated, and a new one created to replace it.
 
 - Issue #7792: Registering non-classes to ABCs raised an obscure error.
 
@@ -1114,21 +1132,21 @@
 - Issue #7561: Fix crashes when using bytearray objects with the posix
   module.
 
-- Issue #1670765: Prevent email.generator.Generator from re-wrapping
-  headers in multipart/signed MIME parts, which fixes one of the sources of
-  invalid modifications to such parts by Generator.
+- Issue #1670765: Prevent email.generator.Generator from re-wrapping headers in
+  multipart/signed MIME parts, which fixes one of the sources of invalid
+  modifications to such parts by Generator.
 
-- Issue #7703: Add support for the new buffer API to `binascii.a2bhqx`.
-  Patch by Florent Xicluna, along with some additional tests.
+- Issue #7703: Add support for the new buffer API to `binascii.a2bhqx`.  Patch
+  by Florent Xicluna, along with some additional tests.
 
-- Issue #7701: Fix crash in binascii.b2a_uu() in debug mode when given a
-  1-byte argument.  Patch by Victor Stinner.
+- Issue #7701: Fix crash in binascii.b2a_uu() in debug mode when given a 1-byte
+  argument.  Patch by Victor Stinner.
 
-- Issue #3299: Fix possible crash in the _sre module when given bad
-  argument values in debug mode.  Patch by Victor Stinner.
+- Issue #3299: Fix possible crash in the _sre module when given bad argument
+  values in debug mode.  Patch by Victor Stinner.
 
-- Issue #2846: Add support for gzip.GzipFile reading zero-padded files.
-  Patch by Brian Curtin.
+- Issue #2846: Add support for gzip.GzipFile reading zero-padded files.  Patch
+  by Brian Curtin.
 
 - Issue #7681: Use floor division in appropiate places in the wave module.
 
@@ -1136,63 +1154,61 @@
   Extension extra options may change the output without changing the .c
   file). Initial patch by Collin Winter.
 
-- Issue #7617: Make sure distutils.unixccompiler.UnixCCompiler recognizes
-  gcc when it has a fully qualified configuration prefix. Initial patch
-  by Arfrever.
+- Issue #7617: Make sure distutils.unixccompiler.UnixCCompiler recognizes gcc
+  when it has a fully qualified configuration prefix. Initial patch by Arfrever.
 
-- Issue #7105: Make WeakKeyDictionary and WeakValueDictionary robust against
-  the destruction of weakref'ed objects while iterating.
+- Issue #7105: Make WeakKeyDictionary and WeakValueDictionary robust against the
+  destruction of weakref'ed objects while iterating.
 
-- Issue #7455: Fix possible crash in cPickle on invalid input.  Patch by
-  Victor Stinner.
+- Issue #7455: Fix possible crash in cPickle on invalid input.  Patch by Victor
+  Stinner.
 
 - Issue #1628205: Socket file objects returned by socket.socket.makefile() now
-  properly handles EINTR within the read, readline, write & flush methods.
-  The socket.sendall() method now properly handles interrupted system calls.
+  properly handles EINTR within the read, readline, write & flush methods.  The
+  socket.sendall() method now properly handles interrupted system calls.
 
-- Issue #7471: Improve the performance of GzipFile's buffering mechanism,
-  and make it implement the `io.BufferedIOBase` ABC to allow for further
-  speedups by wrapping it in an `io.BufferedReader`.  Patch by Nir Aides.
+- Issue #7471: Improve the performance of GzipFile's buffering mechanism, and
+  make it implement the `io.BufferedIOBase` ABC to allow for further speedups by
+  wrapping it in an `io.BufferedReader`.  Patch by Nir Aides.
 
 - Issue #3972: http.client.HTTPConnection now accepts an optional source_address
   parameter to allow specifying where your connections come from.
 
 - socket.create_connection now accepts an optional source_address parameter.
 
-- Issue #5511: now zipfile.ZipFile can be used as a context manager.
-  Initial patch by Brian Curtin.
+- Issue #5511: now zipfile.ZipFile can be used as a context manager.  Initial
+  patch by Brian Curtin.
 
-- Issue #7556: Make sure Distutils' msvc9compile reads and writes the
-  MSVC XML Manifest file in text mode so string patterns can be used
-  in regular expressions.
-
-- Issue #7552: Removed line feed in the base64 Authorization header in
-  the Distutils upload command to avoid an error when PyPI reads it.
-  This occurs on long passwords. Initial patch by JP St. Pierre.
+- Issue #7556: Make sure Distutils' msvc9compile reads and writes the MSVC XML
+  Manifest file in text mode so string patterns can be used in regular
+  expressions.
+
+- Issue #7552: Removed line feed in the base64 Authorization header in the
+  Distutils upload command to avoid an error when PyPI reads it.  This occurs on
+  long passwords. Initial patch by JP St. Pierre.
 
 - Issue #7231: urllib2 cannot handle https with proxy requiring auth.  Patch by
   Tatsuhiro Tsujikawa.
 
-- Issue #4757: `zlib.compress` and other methods in the zlib module now
-  raise a TypeError when given an `str` object (rather than a `bytes`-like
-  object).  Patch by Victor Stinner and Florent Xicluna.
+- Issue #4757: `zlib.compress` and other methods in the zlib module now raise a
+  TypeError when given an `str` object (rather than a `bytes`-like object).
+  Patch by Victor Stinner and Florent Xicluna.
 
 - Issue #7349: Make methods of file objects in the io module accept None as an
   argument where file-like objects (ie StringIO and BytesIO) accept them to mean
   the same as passing no argument.
 
-- Issue #7357: tarfile no longer suppresses fatal extraction errors by
-  default.
+- Issue #7357: tarfile no longer suppresses fatal extraction errors by default.
 
-- Issue #5949: added check for correct lineends in input from IMAP server
-  in imaplib.
+- Issue #5949: added check for correct lineends in input from IMAP server in
+  imaplib.
 
 - Add count() and reverse() methods to collections.deque().
 
 - Fix variations of extending deques:  d.extend(d)  d.extendleft(d)  d+=d
 
-- Issue #6986: Fix crash in the JSON C accelerator when called with the
-  wrong parameter types.  Patch by Victor Stinner.
+- Issue #6986: Fix crash in the JSON C accelerator when called with the wrong
+  parameter types.  Patch by Victor Stinner.
 
 - Issue #7457: added a read_pkg_file method to
   distutils.dist.DistributionMetadata.
@@ -1200,26 +1216,26 @@
 - logging: Added optional `secure` parameter to SMTPHandler, to enable use of
   TLS with authentication credentials.
 
-- Issue #1923: Fixed the removal of meaningful spaces when PKG-INFO is
-  generated in Distutils. Patch by Stephen Emslie.
+- Issue #1923: Fixed the removal of meaningful spaces when PKG-INFO is generated
+  in Distutils.  Patch by Stephen Emslie.
 
 - Issue #4120: Drop reference to CRT from manifest when building extensions with
   msvc9compiler.
 
 - Issue #7333: The `posix` module gains an `initgroups()` function providing
-  access to the initgroups(3) C library call on Unix systems which implement
-  it.  Patch by Jean-Paul Calderone.
+  access to the initgroups(3) C library call on Unix systems which implement it.
+  Patch by Jean-Paul Calderone.
 
 - Issue #7408: Fixed distutils.tests.sdist so it doesn't check for group
   ownership when the group is not forced, because the group may be different
   from the user's group and inherit from its container when the test is run.
 
 - Issue #4486: When an exception has an explicit cause, do not print its
-  implicit context too.  This affects the `traceback` module as well as
-  built-in exception printing.
+  implicit context too.  This affects the `traceback` module as well as built-in
+  exception printing.
 
-- Issue #1515: Enable use of deepcopy() with instance methods.  Patch by
-  Robert Collins.
+- Issue #1515: Enable use of deepcopy() with instance methods.  Patch by Robert
+  Collins.
 
 - Issue #7403: logging: Fixed possible race condition in lock creation.
 
@@ -1227,50 +1243,48 @@
   `storbinary()` method of FTP and FTP_TLS objects gains an optional `rest`
   argument.  Patch by Pablo Mouzo.
 
-- Issue #5788: `datetime.timedelta` objects get a new `total_seconds()`
-  method returning the total number of seconds in the duration.  Patch by
-  Brian Quinlan.
+- Issue #5788: `datetime.timedelta` objects get a new `total_seconds()` method
+  returning the total number of seconds in the duration.  Patch by Brian
+  Quinlan.
 
 - Issue #7133: SSL objects now support the new buffer API.
 
-- Issue #1488943: difflib.Differ() doesn't always add hints for tab characters
+- Issue #1488943: difflib.Differ() doesn't always add hints for tab characters.
 
 - Issue #6123: tarfile now opens empty archives correctly and consistently
   raises ReadError on empty files.
 
-- Issue #7354: distutils.tests.test_msvc9compiler - dragfullwindows can
-  be 2.
+- Issue #7354: distutils.tests.test_msvc9compiler - dragfullwindows can be 2.
 
 - Issue #5037: Proxy the __bytes__ special method instead to __bytes__ instead
   of __str__.
 
-- Issue #7341: Close the internal file object in the TarFile constructor in
-  case of an error.
+- Issue #7341: Close the internal file object in the TarFile constructor in case
+  of an error.
 
 - Issue #7293: distutils.test_msvc9compiler is fixed to work on any fresh
   Windows box. Help provided by David Bolen.
 
-- Issue #2054: ftplib now provides an FTP_TLS class to do secure FTP using
-  TLS or SSL.  Patch by Giampaolo Rodola'.
+- Issue #2054: ftplib now provides an FTP_TLS class to do secure FTP using TLS
+  or SSL.  Patch by Giampaolo Rodola'.
 
-- Issue #7328: pydoc no longer corrupts sys.path when run with the '-m' switch
+- Issue #7328: pydoc no longer corrupts sys.path when run with the '-m' switch.
 
-- Issue #4969: The mimetypes module now reads the MIME database from
-  the registry under Windows.  Patch by Gabriel Genellina.
+- Issue #4969: The mimetypes module now reads the MIME database from the
+  registry under Windows.  Patch by Gabriel Genellina.
 
-- Issue #6816: runpy now provides a run_path function that allows Python code
-  to execute file paths that refer to source or compiled Python files as well
-  as zipfiles, directories and other valid sys.path entries that contain a
-  __main__.py file. This allows applications that run other Python scripts to
+- Issue #6816: runpy now provides a run_path function that allows Python code to
+  execute file paths that refer to source or compiled Python files as well as
+  zipfiles, directories and other valid sys.path entries that contain a
+  __main__.py file.  This allows applications that run other Python scripts to
   support the same flexibility as the CPython command line itself.
 
-- Issue #7318: multiprocessing now uses a timeout when it fails to establish
-  a connection with another process, rather than looping endlessly. The
-  default timeout is 20 seconds, which should be amply sufficient for
-  local connections.
+- Issue #7318: multiprocessing now uses a timeout when it fails to establish a
+  connection with another process, rather than looping endlessly.  The default
+  timeout is 20 seconds, which should be amply sufficient for local connections.
 
 - Issue #7197: Allow unittest.TextTestRunner objects to be pickled and
-  unpickled. This fixes crashes under Windows when trying to run
+  unpickled.  This fixes crashes under Windows when trying to run
   test_multiprocessing in verbose mode.
 
 - Issue #7893: ``unittest.TextTestResult`` is made public and a ``resultclass``
@@ -1281,12 +1295,12 @@
   name in failure reports even if the test has a docstring.
 
 - Issue #3001: Add a C implementation of recursive locks which is used by
-  default when instantiating a `threading.RLock` object. This makes
-  recursive locks as fast as regular non-recursive locks (previously,
-  they were slower by 10x to 15x).
+  default when instantiating a `threading.RLock` object. This makes recursive
+  locks as fast as regular non-recursive locks (previously, they were slower by
+  10x to 15x).
 
-- Issue #7282: Fix a memory leak when an RLock was used in a thread other
-  than those started through `threading.Thread` (for example, using
+- Issue #7282: Fix a memory leak when an RLock was used in a thread other than
+  those started through `threading.Thread` (for example, using
   `_thread.start_new_thread()`).
 
 - Issue #7187: Importlib would not silence the IOError raised when trying to
@@ -1298,20 +1312,19 @@
 - Issue #7211: Allow 64-bit values for the `ident` and `data` fields of kevent
   objects on 64-bit systems.  Patch by Michael Broghton.
 
-- Issue #6896: mailbox.Maildir now invalidates its internal cache each time
-  a modification is done through it.  This fixes inconsistencies and test
-  failures on systems with slightly bogus mtime behaviour.
-
-- Issue #7246 & Issue #7208: getpass now properly flushes input before
-  reading from stdin so that existing input does not confuse it and
-  lead to incorrect entry or an IOError.  It also properly flushes it
-  afterwards to avoid the terminal echoing the input afterwards on
-  OSes such as Solaris.
-
-- Issue #7233: Fix a number of two-argument Decimal methods to make
-  sure that they accept an int or long as the second argument.  Also
-  fix buggy handling of large arguments (those with coefficient longer
-  than the current precision) in shift and rotate.
+- Issue #6896: mailbox.Maildir now invalidates its internal cache each time a
+  modification is done through it.  This fixes inconsistencies and test failures
+  on systems with slightly bogus mtime behaviour.
+
+- Issue #7246 & Issue #7208: getpass now properly flushes input before reading
+  from stdin so that existing input does not confuse it and lead to incorrect
+  entry or an IOError.  It also properly flushes it afterwards to avoid the
+  terminal echoing the input afterwards on OSes such as Solaris.
+
+- Issue #7233: Fix a number of two-argument Decimal methods to make sure that
+  they accept an int or long as the second argument.  Also fix buggy handling of
+  large arguments (those with coefficient longer than the current precision) in
+  shift and rotate.
 
 - Issue #4750: Store the basename of the original filename in the gzip FNAME
   header as required by RFC 1952.
@@ -1333,31 +1346,31 @@
 - Issue #7080: locale.strxfrm() raises a MemoryError on 64-bit non-Windows
   platforms, and assorted locale fixes by Derk Drukker.
 
-- Issue #5833: Fix extra space character in readline completion with the
-  GNU readline library version 6.0.
+- Issue #5833: Fix extra space character in readline completion with the GNU
+  readline library version 6.0.
 
-- Issue #6894: Fixed the issue urllib2 doesn't respect "no_proxy" environment
+- Issue #6894: Fixed the issue urllib2 doesn't respect "no_proxy" environment.
 
 - Issue #7086: Added TCP support to SysLogHandler, and tidied up some
   anachronisms in the code which were a relic of 1.5.2 compatibility.
 
-- Issue #7082: When falling back to the MIME 'name' parameter, the
-  correct place to look for it is the Content-Type header.
+- Issue #7082: When falling back to the MIME 'name' parameter, the correct place
+  to look for it is the Content-Type header.
 
 - Make tokenize.detect_coding() normalize utf-8 and iso-8859-1 variants like the
   builtin tokenizer.
 
-- Issue #7048: Force Decimal.logb to round its result when that result
-  is too large to fit in the current precision.
+- Issue #7048: Force Decimal.logb to round its result when that result is too
+  large to fit in the current precision.
 
-- Issue #6236, #6348: Fix various failures in the I/O library under AIX
-  and other platforms, when using a non-gcc compiler. Patch by Derk Drukker.
+- Issue #6236, #6348: Fix various failures in the I/O library under AIX and
+  other platforms, when using a non-gcc compiler. Patch by Derk Drukker.
 
-- Issue #4606: Passing 'None' if ctypes argtype is set to POINTER(...)
-  does now always result in NULL.
+- Issue #4606: Passing 'None' if ctypes argtype is set to POINTER(...)  does now
+  always result in NULL.
 
-- Issue #5042: Structure sub-subclass does now initialize correctly
-  with base class positional arguments.
+- Issue #5042: Structure sub-subclass does now initialize correctly with base
+  class positional arguments.
 
 - Issue #6882: Import uuid creates zombies processes.
 
@@ -1367,22 +1380,21 @@
 
 - Issue #6888: pdb's alias command was broken when no arguments were given.
 
-- Issue #6857: Default format() alignment should be '>' for Decimal
-  instances.
+- Issue #6857: Default format() alignment should be '>' for Decimal instances.
 
-- Issue #6795: int(Decimal('nan')) now raises ValueError instead of
-  returning NaN or raising InvalidContext.  Also, fix infinite recursion
-  in long(Decimal('nan')).
+- Issue #6795: int(Decimal('nan')) now raises ValueError instead of returning
+  NaN or raising InvalidContext.  Also, fix infinite recursion in
+  long(Decimal('nan')).
 
-- Issue #6850: Fix bug in Decimal._parse_format_specifier for formats
-  with no type specifier.
+- Issue #6850: Fix bug in Decimal._parse_format_specifier for formats with no
+  type specifier.
 
 - Issue #6239: ctypes.c_char_p return value must return bytes.
 
-- Issue #6838: Use a list to accumulate the value instead of
-  repeatedly concatenating strings in http.client's
-  HTTPResponse._read_chunked providing a significant speed increase
-  when downloading large files servend with a Transfer-Encoding of 'chunked'.
+- Issue #6838: Use a list to accumulate the value instead of repeatedly
+  concatenating strings in http.client's HTTPResponse._read_chunked providing a
+  significant speed increase when downloading large files servend with a
+  Transfer-Encoding of 'chunked'.
 
 - Trying to import a submodule from a module that is not a package, ImportError
   should be raised, not AttributeError.
@@ -1391,8 +1403,7 @@
   fall back to computing what __package__ should be instead of giving up.
 
 - Raise a TypeError when the name of a module to be imported for
-  importlib.__import__ is not a string (was raising an
-  AttributeError before).
+  importlib.__import__ is not a string (was raising an AttributeError before).
 
 - Allow the fromlist passed into importlib.__import__ to be any iterable.
 
@@ -1414,50 +1425,50 @@
 
 - Issue #6665: Fix fnmatch to properly match filenames with newlines in them.
 
-- Issue #1135: Add the XView and YView mix-ins to avoid duplicating
-  the xview* and yview* methods.
+- Issue #1135: Add the XView and YView mix-ins to avoid duplicating the xview*
+  and yview* methods.
 
 - Issue #6629: Fix a data corruption issue in the new I/O library, which could
   occur when writing to a BufferedRandom object (e.g. a file opened in "rb+" or
   "wb+" mode) after having buffered a certain amount of data for reading. This
   bug was not present in the pure Python implementation.
 
-- Issue #6622: Fix "local variable 'secret' referenced before
-  assignment" bug in POP3.apop.
+- Issue #6622: Fix "local variable 'secret' referenced before assignment" bug in
+  POP3.apop.
 
 - Issue #2715: Remove remnants of Carbon.File from binhex module.
 
-- Issue #6595: The Decimal constructor now allows arbitrary Unicode
-  decimal digits in input, as recommended by the standard.  Previously
-  it was restricted to accepting [0-9].
+- Issue #6595: The Decimal constructor now allows arbitrary Unicode decimal
+  digits in input, as recommended by the standard.  Previously it was restricted
+  to accepting [0-9].
 
 - Issue #6106: telnetlib.Telnet.process_rawq doesn't handle default WILL/WONT
   DO/DONT correctly.
 
 - Issue #1424152: Fix for http.client, urllib.request to support SSL while
-  working through proxy. Original patch by Christopher Li, changes made by
-  Senthil Kumaran
+  working through proxy.  Original patch by Christopher Li, changes made by
+  Senthil Kumaran.
 
 - Add importlib.abc.ExecutionLoader to represent the PEP 302 protocol for
-  loaders that allow for modules to be executed. Both importlib.abc.PyLoader
-  and PyPycLoader inherit from this class and provide implementations in
-  relation to other methods required by the ABCs.
+  loaders that allow for modules to be executed. Both importlib.abc.PyLoader and
+  PyPycLoader inherit from this class and provide implementations in relation to
+  other methods required by the ABCs.
 
 - importlib.abc.PyLoader did not inherit from importlib.abc.ResourceLoader like
   the documentation said it did even though the code in PyLoader relied on the
   abstract method required by ResourceLoader.
 
-- Issue #6431: Make Fraction type return NotImplemented when it doesn't
-  know how to handle a comparison without loss of precision.  Also add
-  correct handling of infinities and nans for comparisons with float.
+- Issue #6431: Make Fraction type return NotImplemented when it doesn't know how
+  to handle a comparison without loss of precision.  Also add correct handling
+  of infinities and nans for comparisons with float.
 
 - Issue #6415: Fixed warnings.warn segfault on bad formatted string.
 
-- Issue #6358: The exit status of a command started with os.popen() was
-  reported differently than it did with python 2.x.
+- Issue #6358: The exit status of a command started with os.popen() was reported
+  differently than it did with python 2.x.
 
-- Issue #6323: The pdb debugger did not exit when running a script with a
-  syntax error.
+- Issue #6323: The pdb debugger did not exit when running a script with a syntax
+  error.
 
 - Issue #3392: The subprocess communicate() method no longer fails in select()
   when file descriptors are large; communicate() now uses poll() when possible.
@@ -1473,21 +1484,21 @@
 
 - Issue #6218: io.StringIO and io.BytesIO instances are now picklable.
 
-- The os.get_exec_path() function to return the list of directories that will
-  be searched for an executable when launching a subprocess was added.
+- The os.get_exec_path() function to return the list of directories that will be
+  searched for an executable when launching a subprocess was added.
 
 - Issue #7481: When a threading.Thread failed to start it would leave the
   instance stuck in initial state and present in threading.enumerate().
 
-- Issue #1068268: The subprocess module now handles EINTR in internal
-  os.waitpid and os.read system calls where appropriate.
+- Issue #1068268: The subprocess module now handles EINTR in internal os.waitpid
+  and os.read system calls where appropriate.
 
 - Issue #6729: Added ctypes.c_ssize_t to represent ssize_t.
 
 - Issue #6247: The argparse module has been added to the standard library.
 
-- Issue #8235: _socket: Add the constant ``SO_SETFIB``.  SO_SETFIB is
-  a socket option available on FreeBSD 7.1 and newer.
+- Issue #8235: _socket: Add the constant ``SO_SETFIB``.  SO_SETFIB is a socket
+  option available on FreeBSD 7.1 and newer.
 
 Extension Modules
 -----------------
@@ -1495,27 +1506,25 @@
 - Issue #9422: Fix memory leak when re-initializing a struct.Struct object.
 
 - Issue #7900: The getgroups(2) system call on MacOSX behaves rather oddly
-  compared to other unix systems. In particular, os.getgroups() does
-  not reflect any changes made using os.setgroups() but basicly always
-  returns the same information as the id command.
-
-  os.getgroups() can now return more than 16 groups on MacOSX.
+  compared to other unix systems. In particular, os.getgroups() does not reflect
+  any changes made using os.setgroups() but basicly always returns the same
+  information as the id command. os.getgroups() can now return more than 16
+  groups on MacOSX.
 
 - Issue #6095: Make directory argument to os.listdir optional.
 
-- Issue #9277: Fix bug in struct.pack for bools in standard mode
-  (e.g., struct.pack('>?')):  if conversion to bool raised an exception
-  then that exception wasn't properly propagated on machines where
-  char is unsigned.
+- Issue #9277: Fix bug in struct.pack for bools in standard mode (e.g.,
+  struct.pack('>?')): if conversion to bool raised an exception then that
+  exception wasn't properly propagated on machines where char is unsigned.
 
-- Issue #5180: Fixed a bug that prevented loading 2.x pickles in 3.x
-  python when they contain instances of old-style classes.
+- Issue #5180: Fixed a bug that prevented loading 2.x pickles in 3.x python when
+  they contain instances of old-style classes.
 
-- Issue #9165: Add new functions math.isfinite and cmath.isfinite, to
-  accompany existing isinf and isnan functions.
+- Issue #9165: Add new functions math.isfinite and cmath.isfinite, to accompany
+  existing isinf and isnan functions.
 
-- Issue #1578269: Implement os.symlink for Windows 6.0+. Patch by
-  Jason R. Coombs
+- Issue #1578269: Implement os.symlink for Windows 6.0+.  Patch by Jason
+  R. Coombs.
 
 - In struct.pack, correctly propogate exceptions from computing the truth of an
   object in the '?' format.
@@ -1525,58 +1534,56 @@
 - In the math module, correctly lookup __trunc__, __ceil__, and __floor__ as
   special methods.
 
-- Issue #9005: Prevent utctimetuple() from producing year 0 or year
-  10,000.  Prior to this change, timezone adjustment in utctimetuple()
-  could produce tm_year value of 0 or 10,000.  Now an OverflowError is
-  raised in these edge cases.
-
-- Issue #6641: The ``datetime.strptime`` method now supports the
-  ``%z`` directive.  When the ``%z`` directive is present in the
-  format string, an aware ``datetime`` object is returned with
-  ``tzinfo`` bound to a ``datetime.timezone`` instance constructed
-  from the parsed offset.  If both ``%z`` and ``%Z`` are present, the
-  data in ``%Z`` field is used for timezone name, but ``%Z`` data
-  without ``%z`` is discarded.
+- Issue #9005: Prevent utctimetuple() from producing year 0 or year 10,000.
+  Prior to this change, timezone adjustment in utctimetuple() could produce
+  tm_year value of 0 or 10,000.  Now an OverflowError is raised in these edge
+  cases.
+
+- Issue #6641: The ``datetime.strptime`` method now supports the ``%z``
+  directive.  When the ``%z`` directive is present in the format string, an
+  aware ``datetime`` object is returned with ``tzinfo`` bound to a
+  ``datetime.timezone`` instance constructed from the parsed offset.  If both
+  ``%z`` and ``%Z`` are present, the data in ``%Z`` field is used for timezone
+  name, but ``%Z`` data without ``%z`` is discarded.
 
 - Issue #5094: The ``datetime`` module now has a simple concrete class
-  implementing ``datetime.tzinfo`` interface.  Instances of the new
-  class, ``datetime.timezone``, return fixed name and UTC offset from
-  their ``tzname(dt)`` and ``utcoffset(dt)`` methods.  The ``dst(dt)``
-  method always returns ``None``.  A class attribute, ``utc`` contains
-  an instance representing the UTC timezone.  Original patch by Rafe
-  Kaplan.
+  implementing ``datetime.tzinfo`` interface.  Instances of the new class,
+  ``datetime.timezone``, return fixed name and UTC offset from their
+  ``tzname(dt)`` and ``utcoffset(dt)`` methods.  The ``dst(dt)`` method always
+  returns ``None``.  A class attribute, ``utc`` contains an instance
+  representing the UTC timezone.  Original patch by Rafe Kaplan.
 
-- Issue #8973: Add __all__ to struct module; this ensures that
-  help(struct) includes documentation for the struct.Struct class.
+- Issue #8973: Add __all__ to struct module; this ensures that help(struct)
+  includes documentation for the struct.Struct class.
 
 - Issue #3129: Trailing digits in struct format string are no longer ignored.
   For example, "1" or "ilib123" are now invalid formats and cause
   ``struct.error`` to be raised.  Patch by Caleb Deveraux.
 
-- Issue #7384: If the system readline library is linked against ncurses,
-  the curses module must be linked against ncurses as well. Otherwise it
-  is not safe to load both the readline and curses modules in an application.
-
-- Issue #2810: Fix cases where the Windows registry API returns
-  ERROR_MORE_DATA, requiring a re-try in order to get the complete result.
-
-- Issue #8692: Optimize math.factorial: replace the previous naive
-  algorithm with an improved 'binary-split' algorithm that uses fewer
-  multiplications and allows many of the multiplications to be
-  performed using plain C integer arithmetic instead of PyLong
-  arithmetic.  Also uses a lookup table for small arguments.
+- Issue #7384: If the system readline library is linked against ncurses, the
+  curses module must be linked against ncurses as well. Otherwise it is not safe
+  to load both the readline and curses modules in an application.
+
+- Issue #2810: Fix cases where the Windows registry API returns ERROR_MORE_DATA,
+  requiring a re-try in order to get the complete result.
+
+- Issue #8692: Optimize math.factorial: replace the previous naive algorithm
+  with an improved 'binary-split' algorithm that uses fewer multiplications and
+  allows many of the multiplications to be performed using plain C integer
+  arithmetic instead of PyLong arithmetic.  Also uses a lookup table for small
+  arguments.
 
 - Issue #8674: Fixed a number of incorrect or undefined-behaviour-inducing
   overflow checks in the audioop module.
 
 - Issue #8644: The accuracy of td.total_seconds() has been improved (by
   calculating with integer arithmetic instead of float arithmetic internally):
-  the result is now always correctly rounded, and is equivalent to td /
-  timedelta(seconds=1).
+  the result is now always correctly rounded, and is equivalent to ``td /
+  timedelta(seconds=1)``.
 
-- Issue #2706: Allow division of a timedelta by another timedelta:
-  timedelta / timedelta, timedelta % timedelta, timedelta // timedelta
-  and divmod(timedelta, timedelta) are all supported.
+- Issue #2706: Allow division of a timedelta by another timedelta: timedelta /
+  timedelta, timedelta % timedelta, timedelta // timedelta and divmod(timedelta,
+  timedelta) are all supported.
 
 - Issue #8314: Fix unsigned long long bug in libffi on Sparc v8.
 
@@ -1589,7 +1596,7 @@
 
 - Issue #6949: Allow the _dbm extension to be built with db 4.8.x.
 
-- Issue #6544: fix a reference leak in the kqueue implementation's error
+- Issue #6544: Fix a reference leak in the kqueue implementation's error
   handling.
 
 - Stop providing crtassem.h symbols when compiling with Visual Studio 2010, as
@@ -1601,80 +1608,82 @@
 
 - Issue #3366: Add erf, erfc, expm1, gamma, lgamma functions to math module.
 
-- Issue #6877: It is now possible to link the readline extension to the
-  libedit readline emulation on OSX 10.5 or later.
+- Issue #6877: It is now possible to link the readline extension to the libedit
+  readline emulation on OSX 10.5 or later.
 
 - Issue #6848: Fix curses module build failure on OS X 10.6.
 
 - Fix a segfault that could be triggered by expat with specially formed input.
 
-- Issue #6561: '\d' in a regex now matches only characters with
-  Unicode category 'Nd' (Number, Decimal Digit).  Previously it also
-  matched characters with category 'No'.
-
-- Issue #4509: Array objects are no longer modified after an operation
-  failing due to the resize restriction in-place when the object has exported
-  buffers.
+- Issue #6561: '\d' in a regex now matches only characters with Unicode category
+  'Nd' (Number, Decimal Digit).  Previously it also matched characters with
+  category 'No'.
+
+- Issue #4509: Array objects are no longer modified after an operation failing
+  due to the resize restriction in-place when the object has exported buffers.
 
 - Issue #2389: Array objects are now pickled in a portable manner.
 
 - Expat: Fix DoS via XML document with malformed UTF-8 sequences
   (CVE_2009_3560).
 
-- Issue #7242: On Solaris 9 and earlier calling os.fork() from within a
-  thread could raise an incorrect RuntimeError about not holding the import
-  lock.  The import lock is now reinitialized after fork.
+- Issue #7242: On Solaris 9 and earlier calling os.fork() from within a thread
+  could raise an incorrect RuntimeError about not holding the import lock.  The
+  import lock is now reinitialized after fork.
 
 - Issue #7999: os.setreuid() and os.setregid() would refuse to accept a -1
   parameter on some platforms such as OS X.
 
 - Build the ossaudio extension on GNU/kFreeBSD.
 
-- Issue #7347: winreg: Add CreateKeyEx and DeleteKeyEx, as well as fix a
-  bug in the return value of QueryReflectionKey.
+- Issue #7347: winreg: Add CreateKeyEx and DeleteKeyEx, as well as fix a bug in
+  the return value of QueryReflectionKey.
+
+- Issue #7567: PyCurses_setupterm: Don't call `setupterm' twice.
 
 Build
 -----
 
+- Use OpenSSL 1.0.0a on Windows.
+
 - Issue #9280: Make sharedinstall depend on sharedmods.
 
-- Issue #9189: Make a user-specified CFLAGS, CPPFLAGS, or LDFLAGS
-  setting override the configure and makefile defaults, without
-  deleting options the user didn't intend to override.  Developers
-  should no longer need to specify OPT or EXTRA_CFLAGS, although those
-  variables are still present for backward-compatibility.
+- Issue #9189: Make a user-specified CFLAGS, CPPFLAGS, or LDFLAGS setting
+  override the configure and makefile defaults, without deleting options the
+  user didn't intend to override.  Developers should no longer need to specify
+  OPT or EXTRA_CFLAGS, although those variables are still present for
+  backward-compatibility.
 
 - Issue #8854: Fix finding Visual Studio 2008 on Windows x64.
 
 - Issue #1759169, #8864: Drop _XOPEN_SOURCE on Solaris, define it for
   multiprocessing only.
 
-- Issue #8625: Turn off optimization in --with-pydebug builds with
-  gcc.  (Optimization was unintentionally turned on in gcc
-  --with-pydebug builds as a result of the issue #1628484 fix,
-  combined with autoconf's strange choice of default CFLAGS produced
-  by AC_PROG_CC for gcc.)
+- Issue #8625: Turn off optimization in --with-pydebug builds with gcc.
+  (Optimization was unintentionally turned on in gcc --with-pydebug builds as a
+  result of the issue #1628484 fix, combined with autoconf's strange choice of
+  default CFLAGS produced by AC_PROG_CC for gcc.)
 
-- Issue #3646: It is now easily possible to install a Python framework into
-  your home directory on MacOSX, see Mac/README for more information.
+- Issue #3646: It is now easily possible to install a Python framework into your
+  home directory on MacOSX, see Mac/README for more information.
 
 - Issue #3928: os.mknod() now available in Solaris, also.
 
 - Issue #3326: Build Python without -fno-strict-aliasing when the gcc does not
   give false warnings.
 
-- Issue #1628484: The Makefile doesn't ignore the CFLAGS environment
-  variable anymore.  It also forwards the LDFLAGS settings to the linker
-  when building a shared library.
+- Issue #1628484: The Makefile doesn't ignore the CFLAGS environment variable
+  anymore.  It also forwards the LDFLAGS settings to the linker when building a
+  shared library.
 
-- Issue #6716: Quote -x arguments of compileall in MSI installer.
-  Exclude 2to3 tests from compileall.
+- Issue #6716: Quote -x arguments of compileall in MSI installer.  Exclude 2to3
+  tests from compileall.
 
 - Issue #3920, #7903: Define _BSD_SOURCE on OpenBSD 4.4 through 4.9.
 
-- Issue #7632: When Py_USING_MEMORY_DEBUGGER is defined, disable the
-  private memory allocation scheme in dtoa.c and use PyMem_Malloc and
-  PyMem_Free instead.  Also disable caching of powers of 5.
+- Issue #7632: When Py_USING_MEMORY_DEBUGGER is defined, disable the private
+  memory allocation scheme in dtoa.c and use PyMem_Malloc and PyMem_Free
+  instead.  Also disable caching of powers of 5.
 
 - Issue #6491: Allow --with-dbmliborder to specify that no dbms will be built.
 
@@ -1693,21 +1702,21 @@
 - Issue #5792: Extend the short float repr support to x86 systems using
   icc or suncc.
 
-- Issue #6603: Change READ_TIMESTAMP macro in ceval.c so that it
-  compiles correctly under gcc on x86-64.  This fixes a reported
-  problem with the --with-tsc build on x86-64.
+- Issue #6603: Change READ_TIMESTAMP macro in ceval.c so that it compiles
+  correctly under gcc on x86-64.  This fixes a reported problem with the
+  --with-tsc build on x86-64.
 
-- Issue #6802: Fix build issues on MacOSX 10.6
+- Issue #6802: Fix build issues on MacOSX 10.6.
 
 - Issue #6244: Allow detect_tkinter to look for Tcl/Tk 8.6.
 
 - Issue 4601: 'make install' did not set the appropriate permissions on
   directories.
 
-- Issue 5390: Add uninstall icon independent of whether file
-  extensions are installed.
+- Issue 5390: Add uninstall icon independent of whether file extensions are
+  installed.
 
-- Issue #7541: when using ``python-config`` with a framework install the
+- Issue #7541: When using ``python-config`` with a framework install the
   compiler might use the wrong library.
 
 - python-config now supports multiple options on the same command line.
@@ -1716,24 +1725,23 @@
 
 - Issue #8510: Update to autoconf2.65.
 
-
 Documentation
 ------------
 
 - Issue 9255: Document that the 'test' package is meant for internal Python use
   only.
 
-- A small wsgi server was added as Tools/scripts/serve.py, and is used to
+- A small WSGI server was added as Tools/scripts/serve.py, and is used to
   implement a local documentation server via 'make serve' in the doc directory.
 
-- Updating `Using Python` documentation to include description of CPython's
-  -J and -X options.
+- Updating `Using Python` documentation to include description of CPython's -J
+  and -X options.
 
 - Document that importing a module that has None in sys.modules triggers an
   ImportError.
 
-- Issue #6556: Fixed the Distutils configuration files location explanation
-  for Windows.
+- Issue #6556: Fixed the Distutils configuration files location explanation for
+  Windows.
 
 - Update python manual page (options -B, -O0, -s, environment variables
   PYTHONDONTWRITEBYTECODE, PYTHONNOUSERSITE).
@@ -1744,19 +1752,19 @@
 Tests
 -----
 
-- Issue #9251: test_threaded_import didn't fail when run through regrtest
-  if the import lock was disabled.
+- Issue #9251: test_threaded_import didn't fail when run through regrtest if the
+  import lock was disabled.
 
 - Issue #8605: Skip test_gdb if Python is compiled with optimizations.
 
-- Issue #7449: Skip test_socketserver if threading support is disabled
+- Issue #7449: Skip test_socketserver if threading support is disabled.
 
-- Issue #8672: Add a zlib test ensuring that an incomplete stream can be
-  handled by a decompressor object without errors (it returns incomplete
-  uncompressed data).
+- Issue #8672: Add a zlib test ensuring that an incomplete stream can be handled
+  by a decompressor object without errors (it returns incomplete uncompressed
+  data).
 
 - Issue #8533: regrtest uses backslashreplace error handler for stdout to avoid
-  UnicodeEncodeError (write non-ASCII character to stdout using ASCII encoding)
+  UnicodeEncodeError (write non-ASCII character to stdout using ASCII encoding).
 
 - Issue #8576: Remove use of find_unused_port() in test_smtplib and
   test_multiprocessing.  Patch by Paul Moore.
@@ -1764,14 +1772,14 @@
 - Issue #7449: Fix many tests to support Python compiled without thread
   support. Patches written by Jerry Seutter.
 
-- Issue #8108: test_ftplib's non-blocking SSL server now has proper handling
-  of SSL shutdowns.
+- Issue #8108: test_ftplib's non-blocking SSL server now has proper handling of
+  SSL shutdowns.
 
-- Issues #8279, #8330, #8437, #8480, #8495: Fix test_gdb failures, patch
-  written by Dave Malcolm
+- Issues #8279, #8330, #8437, #8480, #8495: Fix test_gdb failures, patch written
+  by Dave Malcolm.
 
-- Issue #3864: Skip three test_signal tests on freebsd6 because they fail
-  if any thread was previously started, most likely due to a platform bug.
+- Issue #3864: Skip three test_signal tests on freebsd6 because they fail if any
+  thread was previously started, most likely due to a platform bug.
 
 - Issue #8193: Fix test_zlib failure with zlib 1.2.4.
 
@@ -1783,37 +1791,37 @@
 - Issue #8180 and #8207: Fix test_pep277 on OS X and add more tests for special
   Unicode normalization cases.
 
-- Issue #7783: test.support.open_urlresource invalidates the outdated files
-  from the local cache.
+- Issue #7783: test.support.open_urlresource invalidates the outdated files from
+  the local cache.
 
 - Issue #7849: Now the utility ``check_warnings`` verifies if the warnings are
   effectively raised.
 
-- The four path modules (genericpath, macpath, ntpath, posixpath) share a
-  common TestCase for some tests: test_genericpath.CommonTest.
+- The four path modules (genericpath, macpath, ntpath, posixpath) share a common
+  TestCase for some tests: test_genericpath.CommonTest.
 
-- Print platform information when running the whole test suite, or using
-  the --verbose flag.
+- Print platform information when running the whole test suite, or using the
+  --verbose flag.
 
 - Issue #767675: enable test_pep277 on POSIX platforms with Unicode-friendly
   filesystem encoding.
 
-- Issue #6292: for the moment at least, the test suite runs cleanly if python
-  is run with the -OO flag.  Tests requiring docstrings are skipped.
+- Issue #6292: for the moment at least, the test suite runs cleanly if python is
+  run with the -OO flag.  Tests requiring docstrings are skipped.
 
-- Issue #7712: test.support gained a new `temp_cwd` context manager which is
-  now also used by regrtest to run all the tests in a temporary directory.
-  The original CWD is saved in `support.SAVEDCWD`.
-  Thanks to Florent Xicluna who helped with the patch.
-
-- Issue #7924: Fix an intermittent 'XXX undetected error' failure in
-  test_capi (only seen so far on platforms where the curses module
-  wasn't built), due to an uncleared exception.
+- Issue #7712: test.support gained a new `temp_cwd` context manager which is now
+  also used by regrtest to run all the tests in a temporary directory.  The
+  original CWD is saved in `support.SAVEDCWD`.  Thanks to Florent Xicluna who
+  helped with the patch.
+
+- Issue #7924: Fix an intermittent 'XXX undetected error' failure in test_capi
+  (only seen so far on platforms where the curses module wasn't built), due to
+  an uncleared exception.
 
 - Issue #7728: test_timeout was changed to use support.bind_port instead of a
   hard coded port.
 
-- Issue #7376: instead of running a self-test (which was failing) when called
+- Issue #7376: Instead of running a self-test (which was failing) when called
   with no arguments, doctest.py now gives a usage message.
 
 - Issue #7396: fix regrtest -s, which was broken by the -j enhancement.
@@ -1821,20 +1829,20 @@
 - Issue #7498: test_multiprocessing now uses test.support.find_unused_port
   instead of a hardcoded port number in test_rapid_restart.
 
-- Issue #7431: use TESTFN in test_linecache instead of trying to create a
-  file in the Lib/test directory, which might be read-only for the
-  user running the tests.
-
-- Issue #7324: add a sanity check to regrtest argument parsing to
-  catch the case of an option with no handler.
-
-- Issue #7312: Add a -F flag to run the selected tests in a loop until
-  a test fails.  Can be combined with -j.
-
-- Issue #6551: test_zipimport could import and then destroy some modules of
-  the encodings package, which would make other tests fail further down
-  the road because the internally cached encoders and decoders would point
-  to empty global variables.
+- Issue #7431: Use TESTFN in test_linecache instead of trying to create a file
+  in the Lib/test directory, which might be read-only for the user running the
+  tests.
+
+- Issue #7324: Add a sanity check to regrtest argument parsing to catch the case
+  of an option with no handler.
+
+- Issue #7312: Add a -F flag to run the selected tests in a loop until a test
+  fails.  Can be combined with -j.
+
+- Issue #6551: test_zipimport could import and then destroy some modules of the
+  encodings package, which would make other tests fail further down the road
+  because the internally cached encoders and decoders would point to empty
+  global variables.
 
 - Issue #7295: Do not use a hardcoded file name in test_tarfile.
 
@@ -1850,29 +1858,29 @@
   calling methods on the object.
 
 - Issue #7222: Make thread "reaping" more reliable so that reference
-  leak-chasing test runs give sensible results. The previous method of
-  reaping threads could return successfully while some Thread objects were
-  still referenced. This also introduces a new private function:
+  leak-chasing test runs give sensible results. The previous method of reaping
+  threads could return successfully while some Thread objects were still
+  referenced. This also introduces a new private function:
   :func:`_thread._count()`.
 
-- Issue #7151: fixed regrtest -j so that output to stderr from a test no
-  longer runs the risk of causing the worker thread to fail.
+- Issue #7151: Fixed regrtest -j so that output to stderr from a test no longer
+  runs the risk of causing the worker thread to fail.
 
 - Issue #7055: test___all__ now greedily detects all modules which have an
   __all__ attribute, rather than using a hardcoded and incomplete list.
 
 - Issue #7058: Added save/restore for things like sys.argv and cwd to
-  runtest_inner in regrtest, with warnings if the called test modifies them,
-  and a new section in the summary report at the end.
+  runtest_inner in regrtest, with warnings if the called test modifies them, and
+  a new section in the summary report at the end.
 
 - Issue #7042: Fix test_signal (test_itimer_virtual) failure on OS X 10.6.
 
-- Fixed tests in importlib.test.source.test_abc_loader that were masking
-  the proper exceptions that should be raised for missing or improper code
-  object bytecode.
+- Fixed tests in importlib.test.source.test_abc_loader that were masking the
+  proper exceptions that should be raised for missing or improper code object
+  bytecode.
 
 - Removed importlib's custom test discovery code and switched to
-    unittest.TestLoader.discover().
+  unittest.TestLoader.discover().
 
 Tools/Demos
 -----------
@@ -1880,8 +1888,7 @@
 - Issue #5464, #8974: Implement plural forms in msgfmt.py.
 
 - iobench (a file I/O benchmark) and ccbench (a concurrency benchmark) were
-  added to the `Tools/` directory.  They were previously living in the
-  sandbox.
+  added to the `Tools/` directory.  They were previously living in the sandbox.
 
 
 What's New in Python 3.1?

Modified: python/branches/import_unicode/Misc/RPM/python-3.2.spec
==============================================================================
--- python/branches/import_unicode/Misc/RPM/python-3.2.spec	(original)
+++ python/branches/import_unicode/Misc/RPM/python-3.2.spec	Sat Jul 31 13:01:07 2010
@@ -39,8 +39,8 @@
 
 %define name python
 #--start constants--
-%define version 3.2a0
-%define libvers 3.2
+%define version 3.2a1
+%define libver 3.2
 #--end constants--
 %define release 1pydotorg
 %define __prefix /usr

Modified: python/branches/import_unicode/Modules/_ctypes/ctypes.h
==============================================================================
--- python/branches/import_unicode/Modules/_ctypes/ctypes.h	(original)
+++ python/branches/import_unicode/Modules/_ctypes/ctypes.h	Sat Jul 31 13:01:07 2010
@@ -1,3 +1,7 @@
+#if defined (__SVR4) && defined (__sun)
+#   include <alloca.h>
+#endif
+
 #ifndef MS_WIN32
 #define max(a, b) ((a) > (b) ? (a) : (b))
 #define min(a, b) ((a) < (b) ? (a) : (b))

Modified: python/branches/import_unicode/Modules/_cursesmodule.c
==============================================================================
--- python/branches/import_unicode/Modules/_cursesmodule.c	(original)
+++ python/branches/import_unicode/Modules/_cursesmodule.c	Sat Jul 31 13:01:07 2010
@@ -2126,7 +2126,7 @@
         }
     }
 
-    if (setupterm(termstr,fd,&err) == ERR) {
+    if (!initialised_setupterm && setupterm(termstr,fd,&err) == ERR) {
         char* s = "setupterm: unknown error";
 
         if (err == 0) {

Modified: python/branches/import_unicode/Modules/zlib/ChangeLog
==============================================================================
--- python/branches/import_unicode/Modules/zlib/ChangeLog	(original)
+++ python/branches/import_unicode/Modules/zlib/ChangeLog	Sat Jul 31 13:01:07 2010
@@ -1,6 +1,359 @@
 
                 ChangeLog file for zlib
 
+Changes in 1.2.5 (19 Apr 2010)
+- Disable visibility attribute in win32/Makefile.gcc [Bar-Lev]
+- Default to libdir as sharedlibdir in configure [Nieder]
+- Update copyright dates on modified source files
+- Update trees.c to be able to generate modified trees.h
+- Exit configure for MinGW, suggesting win32/Makefile.gcc
+
+Changes in 1.2.4.5 (18 Apr 2010)
+- Set sharedlibdir in configure [Torok]
+- Set LDFLAGS in Makefile.in [Bar-Lev]
+- Avoid mkdir objs race condition in Makefile.in [Bowler]
+- Add ZLIB_INTERNAL in front of internal inter-module functions and arrays
+- Define ZLIB_INTERNAL to hide internal functions and arrays for GNU C
+- Don't use hidden attribute when it is a warning generator (e.g. Solaris)
+
+Changes in 1.2.4.4 (18 Apr 2010)
+- Fix CROSS_PREFIX executable testing, CHOST extract, mingw* [Torok]
+- Undefine _LARGEFILE64_SOURCE in zconf.h if it is zero, but not if empty
+- Try to use bash or ksh regardless of functionality of /bin/sh
+- Fix configure incompatibility with NetBSD sh
+- Remove attempt to run under bash or ksh since have better NetBSD fix
+- Fix win32/Makefile.gcc for MinGW [Bar-Lev]
+- Add diagnostic messages when using CROSS_PREFIX in configure
+- Added --sharedlibdir option to configure [Weigelt]
+- Use hidden visibility attribute when available [Frysinger]
+
+Changes in 1.2.4.3 (10 Apr 2010)
+- Only use CROSS_PREFIX in configure for ar and ranlib if they exist
+- Use CROSS_PREFIX for nm [Bar-Lev]
+- Assume _LARGEFILE64_SOURCE defined is equivalent to true
+- Avoid use of undefined symbols in #if with && and ||
+- Make *64 prototypes in gzguts.h consistent with functions
+- Add -shared load option for MinGW in configure [Bowler]
+- Move z_off64_t to public interface, use instead of off64_t
+- Remove ! from shell test in configure (not portable to Solaris)
+- Change +0 macro tests to -0 for possibly increased portability
+
+Changes in 1.2.4.2 (9 Apr 2010)
+- Add consistent carriage returns to readme.txt's in masmx86 and masmx64
+- Really provide prototypes for *64 functions when building without LFS
+- Only define unlink() in minigzip.c if unistd.h not included
+- Update README to point to contrib/vstudio project files
+- Move projects/vc6 to old/ and remove projects/
+- Include stdlib.h in minigzip.c for setmode() definition under WinCE
+- Clean up assembler builds in win32/Makefile.msc [Rowe]
+- Include sys/types.h for Microsoft for off_t definition
+- Fix memory leak on error in gz_open()
+- Symbolize nm as $NM in configure [Weigelt]
+- Use TEST_LDSHARED instead of LDSHARED to link test programs [Weigelt]
+- Add +0 to _FILE_OFFSET_BITS and _LFS64_LARGEFILE in case not defined
+- Fix bug in gzeof() to take into account unused input data
+- Avoid initialization of structures with variables in puff.c
+- Updated win32/README-WIN32.txt [Rowe]
+
+Changes in 1.2.4.1 (28 Mar 2010)
+- Remove the use of [a-z] constructs for sed in configure [gentoo 310225]
+- Remove $(SHAREDLIB) from LIBS in Makefile.in [Creech]
+- Restore "for debugging" comment on sprintf() in gzlib.c
+- Remove fdopen for MVS from gzguts.h
+- Put new README-WIN32.txt in win32 [Rowe]
+- Add check for shell to configure and invoke another shell if needed
+- Fix big fat stinking bug in gzseek() on uncompressed files
+- Remove vestigial F_OPEN64 define in zutil.h
+- Set and check the value of _LARGEFILE_SOURCE and _LARGEFILE64_SOURCE
+- Avoid errors on non-LFS systems when applications define LFS macros
+- Set EXE to ".exe" in configure for MINGW [Kahle]
+- Match crc32() in crc32.c exactly to the prototype in zlib.h [Sherrill]
+- Add prefix for cross-compilation in win32/makefile.gcc [Bar-Lev]
+- Add DLL install in win32/makefile.gcc [Bar-Lev]
+- Allow Linux* or linux* from uname in configure [Bar-Lev]
+- Allow ldconfig to be redefined in configure and Makefile.in [Bar-Lev]
+- Add cross-compilation prefixes to configure [Bar-Lev]
+- Match type exactly in gz_load() invocation in gzread.c
+- Match type exactly of zcalloc() in zutil.c to zlib.h alloc_func
+- Provide prototypes for *64 functions when building zlib without LFS
+- Don't use -lc when linking shared library on MinGW
+- Remove errno.h check in configure and vestigial errno code in zutil.h
+
+Changes in 1.2.4 (14 Mar 2010)
+- Fix VER3 extraction in configure for no fourth subversion
+- Update zlib.3, add docs to Makefile.in to make .pdf out of it
+- Add zlib.3.pdf to distribution
+- Don't set error code in gzerror() if passed pointer is NULL
+- Apply destination directory fixes to CMakeLists.txt [Lowman]
+- Move #cmakedefine's to a new zconf.in.cmakein
+- Restore zconf.h for builds that don't use configure or cmake
+- Add distclean to dummy Makefile for convenience
+- Update and improve INDEX, README, and FAQ
+- Update CMakeLists.txt for the return of zconf.h [Lowman]
+- Update contrib/vstudio/vc9 and vc10 [Vollant]
+- Change libz.dll.a back to libzdll.a in win32/Makefile.gcc
+- Apply license and readme changes to contrib/asm686 [Raiter]
+- Check file name lengths and add -c option in minigzip.c [Li]
+- Update contrib/amd64 and contrib/masmx86/ [Vollant]
+- Avoid use of "eof" parameter in trees.c to not shadow library variable
+- Update make_vms.com for removal of zlibdefs.h [Zinser]
+- Update assembler code and vstudio projects in contrib [Vollant]
+- Remove outdated assembler code contrib/masm686 and contrib/asm586
+- Remove old vc7 and vc8 from contrib/vstudio
+- Update win32/Makefile.msc, add ZLIB_VER_SUBREVISION [Rowe]
+- Fix memory leaks in gzclose_r() and gzclose_w(), file leak in gz_open()
+- Add contrib/gcc_gvmat64 for longest_match and inflate_fast [Vollant]
+- Remove *64 functions from win32/zlib.def (they're not 64-bit yet)
+- Fix bug in void-returning vsprintf() case in gzwrite.c
+- Fix name change from inflate.h in contrib/inflate86/inffas86.c
+- Check if temporary file exists before removing in make_vms.com [Zinser]
+- Fix make install and uninstall for --static option
+- Fix usage of _MSC_VER in gzguts.h and zutil.h [Truta]
+- Update readme.txt in contrib/masmx64 and masmx86 to assemble
+
+Changes in 1.2.3.9 (21 Feb 2010)
+- Expunge gzio.c
+- Move as400 build information to old
+- Fix updates in contrib/minizip and contrib/vstudio
+- Add const to vsnprintf test in configure to avoid warnings [Weigelt]
+- Delete zconf.h (made by configure) [Weigelt]
+- Change zconf.in.h to zconf.h.in per convention [Weigelt]
+- Check for NULL buf in gzgets()
+- Return empty string for gzgets() with len == 1 (like fgets())
+- Fix description of gzgets() in zlib.h for end-of-file, NULL return
+- Update minizip to 1.1 [Vollant]
+- Avoid MSVC loss of data warnings in gzread.c, gzwrite.c
+- Note in zlib.h that gzerror() should be used to distinguish from EOF
+- Remove use of snprintf() from gzlib.c
+- Fix bug in gzseek()
+- Update contrib/vstudio, adding vc9 and vc10 [Kuno, Vollant]
+- Fix zconf.h generation in CMakeLists.txt [Lowman]
+- Improve comments in zconf.h where modified by configure
+
+Changes in 1.2.3.8 (13 Feb 2010)
+- Clean up text files (tabs, trailing whitespace, etc.) [Oberhumer]
+- Use z_off64_t in gz_zero() and gz_skip() to match state->skip
+- Avoid comparison problem when sizeof(int) == sizeof(z_off64_t)
+- Revert to Makefile.in from 1.2.3.6 (live with the clutter)
+- Fix missing error return in gzflush(), add zlib.h note
+- Add *64 functions to zlib.map [Levin]
+- Fix signed/unsigned comparison in gz_comp()
+- Use SFLAGS when testing shared linking in configure
+- Add --64 option to ./configure to use -m64 with gcc
+- Fix ./configure --help to correctly name options
+- Have make fail if a test fails [Levin]
+- Avoid buffer overrun in contrib/masmx64/gvmat64.asm [Simpson]
+- Remove assembler object files from contrib
+
+Changes in 1.2.3.7 (24 Jan 2010)
+- Always gzopen() with O_LARGEFILE if available
+- Fix gzdirect() to work immediately after gzopen() or gzdopen()
+- Make gzdirect() more precise when the state changes while reading
+- Improve zlib.h documentation in many places
+- Catch memory allocation failure in gz_open()
+- Complete close operation if seek forward in gzclose_w() fails
+- Return Z_ERRNO from gzclose_r() if close() fails
+- Return Z_STREAM_ERROR instead of EOF for gzclose() being passed NULL
+- Return zero for gzwrite() errors to match zlib.h description
+- Return -1 on gzputs() error to match zlib.h description
+- Add zconf.in.h to allow recovery from configure modification [Weigelt]
+- Fix static library permissions in Makefile.in [Weigelt]
+- Avoid warnings in configure tests that hide functionality [Weigelt]
+- Add *BSD and DragonFly to Linux case in configure [gentoo 123571]
+- Change libzdll.a to libz.dll.a in win32/Makefile.gcc [gentoo 288212]
+- Avoid access of uninitialized data for first inflateReset2 call [Gomes]
+- Keep object files in subdirectories to reduce the clutter somewhat
+- Remove default Makefile and zlibdefs.h, add dummy Makefile
+- Add new external functions to Z_PREFIX, remove duplicates, z_z_ -> z_
+- Remove zlibdefs.h completely -- modify zconf.h instead
+
+Changes in 1.2.3.6 (17 Jan 2010)
+- Avoid void * arithmetic in gzread.c and gzwrite.c
+- Make compilers happier with const char * for gz_error message
+- Avoid unused parameter warning in inflate.c
+- Avoid signed-unsigned comparison warning in inflate.c
+- Indent #pragma's for traditional C
+- Fix usage of strwinerror() in glib.c, change to gz_strwinerror()
+- Correct email address in configure for system options
+- Update make_vms.com and add make_vms.com to contrib/minizip [Zinser]
+- Update zlib.map [Brown]
+- Fix Makefile.in for Solaris 10 make of example64 and minizip64 [Torok]
+- Apply various fixes to CMakeLists.txt [Lowman]
+- Add checks on len in gzread() and gzwrite()
+- Add error message for no more room for gzungetc()
+- Remove zlib version check in gzwrite()
+- Defer compression of gzprintf() result until need to
+- Use snprintf() in gzdopen() if available
+- Remove USE_MMAP configuration determination (only used by minigzip)
+- Remove examples/pigz.c (available separately)
+- Update examples/gun.c to 1.6
+
+Changes in 1.2.3.5 (8 Jan 2010)
+- Add space after #if in zutil.h for some compilers
+- Fix relatively harmless bug in deflate_fast() [Exarevsky]
+- Fix same problem in deflate_slow()
+- Add $(SHAREDLIBV) to LIBS in Makefile.in [Brown]
+- Add deflate_rle() for faster Z_RLE strategy run-length encoding
+- Add deflate_huff() for faster Z_HUFFMAN_ONLY encoding
+- Change name of "write" variable in inffast.c to avoid library collisions
+- Fix premature EOF from gzread() in gzio.c [Brown]
+- Use zlib header window size if windowBits is 0 in inflateInit2()
+- Remove compressBound() call in deflate.c to avoid linking compress.o
+- Replace use of errno in gz* with functions, support WinCE [Alves]
+- Provide alternative to perror() in minigzip.c for WinCE [Alves]
+- Don't use _vsnprintf on later versions of MSVC [Lowman]
+- Add CMake build script and input file [Lowman]
+- Update contrib/minizip to 1.1 [Svensson, Vollant]
+- Moved nintendods directory from contrib to .
+- Replace gzio.c with a new set of routines with the same functionality
+- Add gzbuffer(), gzoffset(), gzclose_r(), gzclose_w() as part of above
+- Update contrib/minizip to 1.1b
+- Change gzeof() to return 0 on error instead of -1 to agree with zlib.h
+
+Changes in 1.2.3.4 (21 Dec 2009)
+- Use old school .SUFFIXES in Makefile.in for FreeBSD compatibility
+- Update comments in configure and Makefile.in for default --shared
+- Fix test -z's in configure [Marquess]
+- Build examplesh and minigzipsh when not testing
+- Change NULL's to Z_NULL's in deflate.c and in comments in zlib.h
+- Import LDFLAGS from the environment in configure
+- Fix configure to populate SFLAGS with discovered CFLAGS options
+- Adapt make_vms.com to the new Makefile.in [Zinser]
+- Add zlib2ansi script for C++ compilation [Marquess]
+- Add _FILE_OFFSET_BITS=64 test to make test (when applicable)
+- Add AMD64 assembler code for longest match to contrib [Teterin]
+- Include options from $SFLAGS when doing $LDSHARED
+- Simplify 64-bit file support by introducing z_off64_t type
+- Make shared object files in objs directory to work around old Sun cc
+- Use only three-part version number for Darwin shared compiles
+- Add rc option to ar in Makefile.in for when ./configure not run
+- Add -WI,-rpath,. to LDFLAGS for OSF 1 V4*
+- Set LD_LIBRARYN32_PATH for SGI IRIX shared compile
+- Protect against _FILE_OFFSET_BITS being defined when compiling zlib
+- Rename Makefile.in targets allstatic to static and allshared to shared
+- Fix static and shared Makefile.in targets to be independent
+- Correct error return bug in gz_open() by setting state [Brown]
+- Put spaces before ;;'s in configure for better sh compatibility
+- Add pigz.c (parallel implementation of gzip) to examples/
+- Correct constant in crc32.c to UL [Leventhal]
+- Reject negative lengths in crc32_combine()
+- Add inflateReset2() function to work like inflateEnd()/inflateInit2()
+- Include sys/types.h for _LARGEFILE64_SOURCE [Brown]
+- Correct typo in doc/algorithm.txt [Janik]
+- Fix bug in adler32_combine() [Zhu]
+- Catch missing-end-of-block-code error in all inflates and in puff
+    Assures that random input to inflate eventually results in an error
+- Added enough.c (calculation of ENOUGH for inftrees.h) to examples/
+- Update ENOUGH and its usage to reflect discovered bounds
+- Fix gzerror() error report on empty input file [Brown]
+- Add ush casts in trees.c to avoid pedantic runtime errors
+- Fix typo in zlib.h uncompress() description [Reiss]
+- Correct inflate() comments with regard to automatic header detection
+- Remove deprecation comment on Z_PARTIAL_FLUSH (it stays)
+- Put new version of gzlog (2.0) in examples with interruption recovery
+- Add puff compile option to permit invalid distance-too-far streams
+- Add puff TEST command options, ability to read piped input
+- Prototype the *64 functions in zlib.h when _FILE_OFFSET_BITS == 64, but
+  _LARGEFILE64_SOURCE not defined
+- Fix Z_FULL_FLUSH to truly erase the past by resetting s->strstart
+- Fix deflateSetDictionary() to use all 32K for output consistency
+- Remove extraneous #define MIN_LOOKAHEAD in deflate.c (in deflate.h)
+- Clear bytes after deflate lookahead to avoid use of uninitialized data
+- Change a limit in inftrees.c to be more transparent to Coverity Prevent
+- Update win32/zlib.def with exported symbols from zlib.h
+- Correct spelling error in zlib.h [Willem]
+- Allow Z_BLOCK for deflate() to force a new block
+- Allow negative bits in inflatePrime() to delete existing bit buffer
+- Add Z_TREES flush option to inflate() to return at end of trees
+- Add inflateMark() to return current state information for random access
+- Add Makefile for NintendoDS to contrib [Costa]
+- Add -w in configure compile tests to avoid spurious warnings [Beucler]
+- Fix typos in zlib.h comments for deflateSetDictionary()
+- Fix EOF detection in transparent gzread() [Maier]
+
+Changes in 1.2.3.3 (2 October 2006)
+- Make --shared the default for configure, add a --static option
+- Add compile option to permit invalid distance-too-far streams
+- Add inflateUndermine() function which is required to enable above
+- Remove use of "this" variable name for C++ compatibility [Marquess]
+- Add testing of shared library in make test, if shared library built
+- Use ftello() and fseeko() if available instead of ftell() and fseek()
+- Provide two versions of all functions that use the z_off_t type for
+  binary compatibility -- a normal version and a 64-bit offset version,
+  per the Large File Support Extension when _LARGEFILE64_SOURCE is
+  defined; use the 64-bit versions by default when _FILE_OFFSET_BITS
+  is defined to be 64
+- Add a --uname= option to configure to perhaps help with cross-compiling
+
+Changes in 1.2.3.2 (3 September 2006)
+- Turn off silly Borland warnings [Hay]
+- Use off64_t and define _LARGEFILE64_SOURCE when present
+- Fix missing dependency on inffixed.h in Makefile.in
+- Rig configure --shared to build both shared and static [Teredesai, Truta]
+- Remove zconf.in.h and instead create a new zlibdefs.h file
+- Fix contrib/minizip/unzip.c non-encrypted after encrypted [Vollant]
+- Add treebuild.xml (see http://treebuild.metux.de/) [Weigelt]
+
+Changes in 1.2.3.1 (16 August 2006)
+- Add watcom directory with OpenWatcom make files [Daniel]
+- Remove #undef of FAR in zconf.in.h for MVS [Fedtke]
+- Update make_vms.com [Zinser]
+- Use -fPIC for shared build in configure [Teredesai, Nicholson]
+- Use only major version number for libz.so on IRIX and OSF1 [Reinholdtsen]
+- Use fdopen() (not _fdopen()) for Interix in zutil.h [B?ck]
+- Add some FAQ entries about the contrib directory
+- Update the MVS question in the FAQ
+- Avoid extraneous reads after EOF in gzio.c [Brown]
+- Correct spelling of "successfully" in gzio.c [Randers-Pehrson]
+- Add comments to zlib.h about gzerror() usage [Brown]
+- Set extra flags in gzip header in gzopen() like deflate() does
+- Make configure options more compatible with double-dash conventions
+  [Weigelt]
+- Clean up compilation under Solaris SunStudio cc [Rowe, Reinholdtsen]
+- Fix uninstall target in Makefile.in [Truta]
+- Add pkgconfig support [Weigelt]
+- Use $(DESTDIR) macro in Makefile.in [Reinholdtsen, Weigelt]
+- Replace set_data_type() with a more accurate detect_data_type() in
+  trees.c, according to the txtvsbin.txt document [Truta]
+- Swap the order of #include <stdio.h> and #include "zlib.h" in
+  gzio.c, example.c and minigzip.c [Truta]
+- Shut up annoying VS2005 warnings about standard C deprecation [Rowe,
+  Truta] (where?)
+- Fix target "clean" from win32/Makefile.bor [Truta]
+- Create .pdb and .manifest files in win32/makefile.msc [Ziegler, Rowe]
+- Update zlib www home address in win32/DLL_FAQ.txt [Truta]
+- Update contrib/masmx86/inffas32.asm for VS2005 [Vollant, Van Wassenhove]
+- Enable browse info in the "Debug" and "ASM Debug" configurations in
+  the Visual C++ 6 project, and set (non-ASM) "Debug" as default [Truta]
+- Add pkgconfig support [Weigelt]
+- Add ZLIB_VER_MAJOR, ZLIB_VER_MINOR and ZLIB_VER_REVISION in zlib.h,
+  for use in win32/zlib1.rc [Polushin, Rowe, Truta]
+- Add a document that explains the new text detection scheme to
+  doc/txtvsbin.txt [Truta]
+- Add rfc1950.txt, rfc1951.txt and rfc1952.txt to doc/ [Truta]
+- Move algorithm.txt into doc/ [Truta]
+- Synchronize FAQ with website
+- Fix compressBound(), was low for some pathological cases [Fearnley]
+- Take into account wrapper variations in deflateBound()
+- Set examples/zpipe.c input and output to binary mode for Windows
+- Update examples/zlib_how.html with new zpipe.c (also web site)
+- Fix some warnings in examples/gzlog.c and examples/zran.c (it seems
+  that gcc became pickier in 4.0)
+- Add zlib.map for Linux: "All symbols from zlib-1.1.4 remain
+  un-versioned, the patch adds versioning only for symbols introduced in
+  zlib-1.2.0 or later.  It also declares as local those symbols which are
+  not designed to be exported." [Levin]
+- Update Z_PREFIX list in zconf.in.h, add --zprefix option to configure
+- Do not initialize global static by default in trees.c, add a response
+  NO_INIT_GLOBAL_POINTERS to initialize them if needed [Marquess]
+- Don't use strerror() in gzio.c under WinCE [Yakimov]
+- Don't use errno.h in zutil.h under WinCE [Yakimov]
+- Move arguments for AR to its usage to allow replacing ar [Marot]
+- Add HAVE_VISIBILITY_PRAGMA in zconf.in.h for Mozilla [Randers-Pehrson]
+- Improve inflateInit() and inflateInit2() documentation
+- Fix structure size comment in inflate.h
+- Change configure help option from --h* to --help [Santos]
+
 Changes in 1.2.3 (18 July 2005)
 - Apply security vulnerability fixes to contrib/infback9 as well
 - Clean up some text files (carriage returns, trailing space)
@@ -13,7 +366,7 @@
   compile
 - Fix some spelling errors in comments [Betts]
 - Correct inflateInit2() error return documentation in zlib.h
-- Added zran.c example of compressed data random access to examples
+- Add zran.c example of compressed data random access to examples
   directory, shows use of inflatePrime()
 - Fix cast for assignments to strm->state in inflate.c and infback.c
 - Fix zlibCompileFlags() in zutil.c to use 1L for long shifts [Oberhumer]

Modified: python/branches/import_unicode/Modules/zlib/FAQ
==============================================================================
--- python/branches/import_unicode/Modules/zlib/FAQ	(original)
+++ python/branches/import_unicode/Modules/zlib/FAQ	Sat Jul 31 13:01:07 2010
@@ -3,8 +3,8 @@
 
 
 If your question is not there, please check the zlib home page
-http://www.zlib.org which may have more recent information.
-The lastest zlib FAQ is at http://www.gzip.org/zlib/zlib_faq.html
+http://zlib.net/ which may have more recent information.
+The lastest zlib FAQ is at http://zlib.net/zlib_faq.html
 
 
  1. Is zlib Y2K-compliant?
@@ -13,54 +13,51 @@
 
  2. Where can I get a Windows DLL version?
 
-    The zlib sources can be compiled without change to produce a DLL.
-    See the file win32/DLL_FAQ.txt in the zlib distribution.
-    Pointers to the precompiled DLL are found in the zlib web site at
-    http://www.zlib.org.
+    The zlib sources can be compiled without change to produce a DLL.  See the
+    file win32/DLL_FAQ.txt in the zlib distribution.  Pointers to the
+    precompiled DLL are found in the zlib web site at http://zlib.net/ .
 
  3. Where can I get a Visual Basic interface to zlib?
 
     See
-        * http://www.dogma.net/markn/articles/zlibtool/zlibtool.htm
-        * contrib/visual-basic.txt in the zlib distribution
+        * http://marknelson.us/1997/01/01/zlib-engine/
         * win32/DLL_FAQ.txt in the zlib distribution
 
  4. compress() returns Z_BUF_ERROR.
 
-    Make sure that before the call of compress, the length of the compressed
-    buffer is equal to the total size of the compressed buffer and not
-    zero. For Visual Basic, check that this parameter is passed by reference
+    Make sure that before the call of compress(), the length of the compressed
+    buffer is equal to the available size of the compressed buffer and not
+    zero.  For Visual Basic, check that this parameter is passed by reference
     ("as any"), not by value ("as long").
 
  5. deflate() or inflate() returns Z_BUF_ERROR.
 
-    Before making the call, make sure that avail_in and avail_out are not
-    zero. When setting the parameter flush equal to Z_FINISH, also make sure
-    that avail_out is big enough to allow processing all pending input.
-    Note that a Z_BUF_ERROR is not fatal--another call to deflate() or
-    inflate() can be made with more input or output space. A Z_BUF_ERROR
-    may in fact be unavoidable depending on how the functions are used, since
-    it is not possible to tell whether or not there is more output pending
-    when strm.avail_out returns with zero.
+    Before making the call, make sure that avail_in and avail_out are not zero.
+    When setting the parameter flush equal to Z_FINISH, also make sure that
+    avail_out is big enough to allow processing all pending input.  Note that a
+    Z_BUF_ERROR is not fatal--another call to deflate() or inflate() can be
+    made with more input or output space.  A Z_BUF_ERROR may in fact be
+    unavoidable depending on how the functions are used, since it is not
+    possible to tell whether or not there is more output pending when
+    strm.avail_out returns with zero.  See http://zlib.net/zlib_how.html for a
+    heavily annotated example.
 
  6. Where's the zlib documentation (man pages, etc.)?
 
-    It's in zlib.h for the moment, and Francis S. Lin has converted it to a
-    web page zlib.html. Volunteers to transform this to Unix-style man pages,
-    please contact us (zlib at gzip.org). Examples of zlib usage are in the files
-    example.c and minigzip.c.
+    It's in zlib.h .  Examples of zlib usage are in the files example.c and
+    minigzip.c, with more in examples/ .
 
  7. Why don't you use GNU autoconf or libtool or ...?
 
-    Because we would like to keep zlib as a very small and simple
-    package. zlib is rather portable and doesn't need much configuration.
+    Because we would like to keep zlib as a very small and simple package.
+    zlib is rather portable and doesn't need much configuration.
 
  8. I found a bug in zlib.
 
-    Most of the time, such problems are due to an incorrect usage of
-    zlib. Please try to reproduce the problem with a small program and send
-    the corresponding source to us at zlib at gzip.org . Do not send
-    multi-megabyte data files without prior agreement.
+    Most of the time, such problems are due to an incorrect usage of zlib.
+    Please try to reproduce the problem with a small program and send the
+    corresponding source to us at zlib at gzip.org .  Do not send multi-megabyte
+    data files without prior agreement.
 
  9. Why do I get "undefined reference to gzputc"?
 
@@ -82,7 +79,7 @@
 
 12. Can zlib handle .Z files?
 
-    No, sorry. You have to spawn an uncompress or gunzip subprocess, or adapt
+    No, sorry.  You have to spawn an uncompress or gunzip subprocess, or adapt
     the code of uncompress on your own.
 
 13. How can I make a Unix shared library?
@@ -99,8 +96,10 @@
 
     However, many flavors of Unix come with a shared zlib already installed.
     Before going to the trouble of compiling a shared version of zlib and
-    trying to install it, you may want to check if it's already there! If you
-    can #include <zlib.h>, it's there. The -lz option will probably link to it.
+    trying to install it, you may want to check if it's already there!  If you
+    can #include <zlib.h>, it's there.  The -lz option will probably link to
+    it.  You can check the version at the top of zlib.h or with the
+    ZLIB_VERSION symbol defined in zlib.h .
 
 15. I have a question about OttoPDF.
 
@@ -109,8 +108,8 @@
 
 16. Can zlib decode Flate data in an Adobe PDF file?
 
-    Yes. See http://www.fastio.com/ (ClibPDF), or http://www.pdflib.com/ .
-    To modify PDF forms, see http://sourceforge.net/projects/acroformtool/ .
+    Yes. See http://www.pdflib.com/ . To modify PDF forms, see
+    http://sourceforge.net/projects/acroformtool/ .
 
 17. Why am I getting this "register_frame_info not found" error on Solaris?
 
@@ -121,67 +120,67 @@
         symbol __register_frame_info: referenced symbol not found
 
     The symbol __register_frame_info is not part of zlib, it is generated by
-    the C compiler (cc or gcc). You must recompile applications using zlib
-    which have this problem. This problem is specific to Solaris. See
+    the C compiler (cc or gcc).  You must recompile applications using zlib
+    which have this problem.  This problem is specific to Solaris.  See
     http://www.sunfreeware.com for Solaris versions of zlib and applications
     using zlib.
 
 18. Why does gzip give an error on a file I make with compress/deflate?
 
     The compress and deflate functions produce data in the zlib format, which
-    is different and incompatible with the gzip format. The gz* functions in
-    zlib on the other hand use the gzip format. Both the zlib and gzip
-    formats use the same compressed data format internally, but have different
-    headers and trailers around the compressed data.
+    is different and incompatible with the gzip format.  The gz* functions in
+    zlib on the other hand use the gzip format.  Both the zlib and gzip formats
+    use the same compressed data format internally, but have different headers
+    and trailers around the compressed data.
 
 19. Ok, so why are there two different formats?
 
-    The gzip format was designed to retain the directory information about
-    a single file, such as the name and last modification date. The zlib
-    format on the other hand was designed for in-memory and communication
-    channel applications, and has a much more compact header and trailer and
-    uses a faster integrity check than gzip.
+    The gzip format was designed to retain the directory information about a
+    single file, such as the name and last modification date.  The zlib format
+    on the other hand was designed for in-memory and communication channel
+    applications, and has a much more compact header and trailer and uses a
+    faster integrity check than gzip.
 
 20. Well that's nice, but how do I make a gzip file in memory?
 
     You can request that deflate write the gzip format instead of the zlib
-    format using deflateInit2(). You can also request that inflate decode
-    the gzip format using inflateInit2(). Read zlib.h for more details.
+    format using deflateInit2().  You can also request that inflate decode the
+    gzip format using inflateInit2().  Read zlib.h for more details.
 
 21. Is zlib thread-safe?
 
-    Yes. However any library routines that zlib uses and any application-
-    provided memory allocation routines must also be thread-safe. zlib's gz*
+    Yes.  However any library routines that zlib uses and any application-
+    provided memory allocation routines must also be thread-safe.  zlib's gz*
     functions use stdio library routines, and most of zlib's functions use the
-    library memory allocation routines by default. zlib's Init functions allow
-    for the application to provide custom memory allocation routines.
+    library memory allocation routines by default.  zlib's *Init* functions
+    allow for the application to provide custom memory allocation routines.
 
     Of course, you should only operate on any given zlib or gzip stream from a
     single thread at a time.
 
 22. Can I use zlib in my commercial application?
 
-    Yes. Please read the license in zlib.h.
+    Yes.  Please read the license in zlib.h.
 
 23. Is zlib under the GNU license?
 
-    No. Please read the license in zlib.h.
+    No.  Please read the license in zlib.h.
 
 24. The license says that altered source versions must be "plainly marked". So
     what exactly do I need to do to meet that requirement?
 
-    You need to change the ZLIB_VERSION and ZLIB_VERNUM #defines in zlib.h. In
+    You need to change the ZLIB_VERSION and ZLIB_VERNUM #defines in zlib.h.  In
     particular, the final version number needs to be changed to "f", and an
-    identification string should be appended to ZLIB_VERSION. Version numbers
+    identification string should be appended to ZLIB_VERSION.  Version numbers
     x.x.x.f are reserved for modifications to zlib by others than the zlib
-    maintainers. For example, if the version of the base zlib you are altering
+    maintainers.  For example, if the version of the base zlib you are altering
     is "1.2.3.4", then in zlib.h you should change ZLIB_VERNUM to 0x123f, and
-    ZLIB_VERSION to something like "1.2.3.f-zachary-mods-v3". You can also
+    ZLIB_VERSION to something like "1.2.3.f-zachary-mods-v3".  You can also
     update the version strings in deflate.c and inftrees.c.
 
     For altered source distributions, you should also note the origin and
     nature of the changes in zlib.h, as well as in ChangeLog and README, along
-    with the dates of the alterations. The origin should include at least your
+    with the dates of the alterations.  The origin should include at least your
     name (or your company's name), and an email address to contact for help or
     issues with the library.
 
@@ -197,105 +196,112 @@
 
 26. Will zlib work on a 64-bit machine?
 
-    It should. It has been tested on 64-bit machines, and has no dependence
-    on any data types being limited to 32-bits in length. If you have any
+    Yes.  It has been tested on 64-bit machines, and has no dependence on any
+    data types being limited to 32-bits in length.  If you have any
     difficulties, please provide a complete problem report to zlib at gzip.org
 
 27. Will zlib decompress data from the PKWare Data Compression Library?
 
-    No. The PKWare DCL uses a completely different compressed data format
-    than does PKZIP and zlib. However, you can look in zlib's contrib/blast
+    No.  The PKWare DCL uses a completely different compressed data format than
+    does PKZIP and zlib.  However, you can look in zlib's contrib/blast
     directory for a possible solution to your problem.
 
 28. Can I access data randomly in a compressed stream?
 
-    No, not without some preparation. If when compressing you periodically
-    use Z_FULL_FLUSH, carefully write all the pending data at those points,
-    and keep an index of those locations, then you can start decompression
-    at those points. You have to be careful to not use Z_FULL_FLUSH too
-    often, since it can significantly degrade compression.
+    No, not without some preparation.  If when compressing you periodically use
+    Z_FULL_FLUSH, carefully write all the pending data at those points, and
+    keep an index of those locations, then you can start decompression at those
+    points.  You have to be careful to not use Z_FULL_FLUSH too often, since it
+    can significantly degrade compression.  Alternatively, you can scan a
+    deflate stream once to generate an index, and then use that index for
+    random access.  See examples/zran.c .
 
 29. Does zlib work on MVS, OS/390, CICS, etc.?
 
-    We don't know for sure. We have heard occasional reports of success on
-    these systems. If you do use it on one of these, please provide us with
-    a report, instructions, and patches that we can reference when we get
-    these questions. Thanks.
+    It has in the past, but we have not heard of any recent evidence.  There
+    were working ports of zlib 1.1.4 to MVS, but those links no longer work.
+    If you know of recent, successful applications of zlib on these operating
+    systems, please let us know.  Thanks.
 
-30. Is there some simpler, easier to read version of inflate I can look at
-    to understand the deflate format?
+30. Is there some simpler, easier to read version of inflate I can look at to
+    understand the deflate format?
 
-    First off, you should read RFC 1951. Second, yes. Look in zlib's
+    First off, you should read RFC 1951.  Second, yes.  Look in zlib's
     contrib/puff directory.
 
 31. Does zlib infringe on any patents?
 
-    As far as we know, no. In fact, that was originally the whole point behind
-    zlib. Look here for some more information:
+    As far as we know, no.  In fact, that was originally the whole point behind
+    zlib.  Look here for some more information:
 
     http://www.gzip.org/#faq11
 
 32. Can zlib work with greater than 4 GB of data?
 
-    Yes. inflate() and deflate() will process any amount of data correctly.
+    Yes.  inflate() and deflate() will process any amount of data correctly.
     Each call of inflate() or deflate() is limited to input and output chunks
     of the maximum value that can be stored in the compiler's "unsigned int"
-    type, but there is no limit to the number of chunks. Note however that the
-    strm.total_in and strm_total_out counters may be limited to 4 GB. These
+    type, but there is no limit to the number of chunks.  Note however that the
+    strm.total_in and strm_total_out counters may be limited to 4 GB.  These
     counters are provided as a convenience and are not used internally by
-    inflate() or deflate(). The application can easily set up its own counters
+    inflate() or deflate().  The application can easily set up its own counters
     updated after each call of inflate() or deflate() to count beyond 4 GB.
     compress() and uncompress() may be limited to 4 GB, since they operate in a
-    single call. gzseek() and gztell() may be limited to 4 GB depending on how
-    zlib is compiled. See the zlibCompileFlags() function in zlib.h.
+    single call.  gzseek() and gztell() may be limited to 4 GB depending on how
+    zlib is compiled.  See the zlibCompileFlags() function in zlib.h.
 
-    The word "may" appears several times above since there is a 4 GB limit
-    only if the compiler's "long" type is 32 bits. If the compiler's "long"
-    type is 64 bits, then the limit is 16 exabytes.
+    The word "may" appears several times above since there is a 4 GB limit only
+    if the compiler's "long" type is 32 bits.  If the compiler's "long" type is
+    64 bits, then the limit is 16 exabytes.
 
 33. Does zlib have any security vulnerabilities?
 
-    The only one that we are aware of is potentially in gzprintf(). If zlib
-    is compiled to use sprintf() or vsprintf(), then there is no protection
-    against a buffer overflow of a 4K string space, other than the caller of
-    gzprintf() assuring that the output will not exceed 4K. On the other
-    hand, if zlib is compiled to use snprintf() or vsnprintf(), which should
-    normally be the case, then there is no vulnerability. The ./configure
-    script will display warnings if an insecure variation of sprintf() will
-    be used by gzprintf(). Also the zlibCompileFlags() function will return
-    information on what variant of sprintf() is used by gzprintf().
+    The only one that we are aware of is potentially in gzprintf().  If zlib is
+    compiled to use sprintf() or vsprintf(), then there is no protection
+    against a buffer overflow of an 8K string space (or other value as set by
+    gzbuffer()), other than the caller of gzprintf() assuring that the output
+    will not exceed 8K.  On the other hand, if zlib is compiled to use
+    snprintf() or vsnprintf(), which should normally be the case, then there is
+    no vulnerability.  The ./configure script will display warnings if an
+    insecure variation of sprintf() will be used by gzprintf().  Also the
+    zlibCompileFlags() function will return information on what variant of
+    sprintf() is used by gzprintf().
 
     If you don't have snprintf() or vsnprintf() and would like one, you can
     find a portable implementation here:
 
         http://www.ijs.si/software/snprintf/
 
-    Note that you should be using the most recent version of zlib. Versions
-    1.1.3 and before were subject to a double-free vulnerability.
+    Note that you should be using the most recent version of zlib.  Versions
+    1.1.3 and before were subject to a double-free vulnerability, and versions
+    1.2.1 and 1.2.2 were subject to an access exception when decompressing
+    invalid compressed data.
 
 34. Is there a Java version of zlib?
 
     Probably what you want is to use zlib in Java. zlib is already included
     as part of the Java SDK in the java.util.zip package. If you really want
     a version of zlib written in the Java language, look on the zlib home
-    page for links: http://www.zlib.org/
+    page for links: http://zlib.net/ .
 
 35. I get this or that compiler or source-code scanner warning when I crank it
     up to maximally-pedantic. Can't you guys write proper code?
 
     Many years ago, we gave up attempting to avoid warnings on every compiler
-    in the universe. It just got to be a waste of time, and some compilers
-    were downright silly. So now, we simply make sure that the code always
-    works.
+    in the universe.  It just got to be a waste of time, and some compilers
+    were downright silly as well as contradicted each other.  So now, we simply
+    make sure that the code always works.
 
 36. Valgrind (or some similar memory access checker) says that deflate is
     performing a conditional jump that depends on an uninitialized value.
     Isn't that a bug?
 
-    No.  That is intentional for performance reasons, and the output of
-    deflate is not affected.  This only started showing up recently since
-    zlib 1.2.x uses malloc() by default for allocations, whereas earlier
-    versions used calloc(), which zeros out the allocated memory.
+    No.  That is intentional for performance reasons, and the output of deflate
+    is not affected.  This only started showing up recently since zlib 1.2.x
+    uses malloc() by default for allocations, whereas earlier versions used
+    calloc(), which zeros out the allocated memory.  Even though the code was
+    correct, versions 1.2.4 and later was changed to not stimulate these
+    checkers.
 
 37. Will zlib read the (insert any ancient or arcane format here) compressed
     data format?
@@ -305,20 +311,21 @@
 
 38. How can I encrypt/decrypt zip files with zlib?
 
-    zlib doesn't support encryption. The original PKZIP encryption is very weak
-    and can be broken with freely available programs. To get strong encryption,
-    use GnuPG, http://www.gnupg.org/ , which already includes zlib compression.
-    For PKZIP compatible "encryption", look at http://www.info-zip.org/
+    zlib doesn't support encryption.  The original PKZIP encryption is very
+    weak and can be broken with freely available programs.  To get strong
+    encryption, use GnuPG, http://www.gnupg.org/ , which already includes zlib
+    compression.  For PKZIP compatible "encryption", look at
+    http://www.info-zip.org/
 
 39. What's the difference between the "gzip" and "deflate" HTTP 1.1 encodings?
 
-    "gzip" is the gzip format, and "deflate" is the zlib format. They should
-    probably have called the second one "zlib" instead to avoid confusion
-    with the raw deflate compressed data format. While the HTTP 1.1 RFC 2616
+    "gzip" is the gzip format, and "deflate" is the zlib format.  They should
+    probably have called the second one "zlib" instead to avoid confusion with
+    the raw deflate compressed data format.  While the HTTP 1.1 RFC 2616
     correctly points to the zlib specification in RFC 1950 for the "deflate"
     transfer encoding, there have been reports of servers and browsers that
     incorrectly produce or expect raw deflate data per the deflate
-    specficiation in RFC 1951, most notably Microsoft. So even though the
+    specficiation in RFC 1951, most notably Microsoft.  So even though the
     "deflate" transfer encoding using the zlib format would be the more
     efficient approach (and in fact exactly what the zlib format was designed
     for), using the "gzip" transfer encoding is probably more reliable due to
@@ -328,12 +335,32 @@
 
 40. Does zlib support the new "Deflate64" format introduced by PKWare?
 
-    No. PKWare has apparently decided to keep that format proprietary, since
-    they have not documented it as they have previous compression formats.
-    In any case, the compression improvements are so modest compared to other
-    more modern approaches, that it's not worth the effort to implement.
+    No.  PKWare has apparently decided to keep that format proprietary, since
+    they have not documented it as they have previous compression formats.  In
+    any case, the compression improvements are so modest compared to other more
+    modern approaches, that it's not worth the effort to implement.
 
-41. Can you please sign these lengthy legal documents and fax them back to us
+41. I'm having a problem with the zip functions in zlib, can you help?
+
+    There are no zip functions in zlib.  You are probably using minizip by
+    Giles Vollant, which is found in the contrib directory of zlib.  It is not
+    part of zlib.  In fact none of the stuff in contrib is part of zlib.  The
+    files in there are not supported by the zlib authors.  You need to contact
+    the authors of the respective contribution for help.
+
+42. The match.asm code in contrib is under the GNU General Public License.
+    Since it's part of zlib, doesn't that mean that all of zlib falls under the
+    GNU GPL?
+
+    No.  The files in contrib are not part of zlib.  They were contributed by
+    other authors and are provided as a convenience to the user within the zlib
+    distribution.  Each item in contrib has its own license.
+
+43. Is zlib subject to export controls?  What is its ECCN?
+
+    zlib is not subject to export controls, and so is classified as EAR99.
+
+44. Can you please sign these lengthy legal documents and fax them back to us
     so that we can use your software in our product?
 
     No. Go away. Shoo.

Modified: python/branches/import_unicode/Modules/zlib/INDEX
==============================================================================
--- python/branches/import_unicode/Modules/zlib/INDEX	(original)
+++ python/branches/import_unicode/Modules/zlib/INDEX	Sat Jul 31 13:01:07 2010
@@ -1,23 +1,32 @@
+CMakeLists.txt  cmake build file
 ChangeLog       history of changes
 FAQ             Frequently Asked Questions about zlib
 INDEX           this file
-Makefile        makefile for Unix (generated by configure)
-Makefile.in     makefile for Unix (template for configure)
+Makefile        dummy Makefile that tells you to ./configure
+Makefile.in     template for Unix Makefile
 README          guess what
-algorithm.txt   description of the (de)compression algorithm
 configure       configure script for Unix
-zconf.in.h      template for zconf.h (used by configure)
+make_vms.com    makefile for VMS
+treebuild.xml   XML description of source file dependencies
+zconf.h.cmakein zconf.h template for cmake
+zconf.h.in      zconf.h template for configure
+zlib.3          Man page for zlib
+zlib.3.pdf      Man page in PDF format
+zlib.map        Linux symbol information
+zlib.pc.in      Template for pkg-config descriptor
+zlib2ansi       perl script to convert source files for C++ compilation
 
 amiga/          makefiles for Amiga SAS C
-as400/          makefiles for IBM AS/400
+doc/            documentation for formats and algorithms
 msdos/          makefiles for MSDOS
+nintendods/     makefile for Nintendo DS
 old/            makefiles for various architectures and zlib documentation
                 files that have not yet been updated for zlib 1.2.x
-projects/       projects for various Integrated Development Environments
 qnx/            makefiles for QNX
+watcom/         makefiles for OpenWatcom
 win32/          makefiles for Windows
 
-                zlib public header files (must be kept):
+                zlib public header files (required for library use):
 zconf.h
 zlib.h
 
@@ -28,7 +37,11 @@
 crc32.h
 deflate.c
 deflate.h
-gzio.c
+gzclose.c
+gzguts.h
+gzlib.c
+gzread.c
+gzwrite.c
 infback.c
 inffast.c
 inffast.h
@@ -46,6 +59,7 @@
                 source files for sample programs:
 example.c
 minigzip.c
+See examples/README.examples for more
 
                 unsupported contribution by third parties
 See contrib/README.contrib

Modified: python/branches/import_unicode/Modules/zlib/Makefile
==============================================================================
--- python/branches/import_unicode/Modules/zlib/Makefile	(original)
+++ python/branches/import_unicode/Modules/zlib/Makefile	Sat Jul 31 13:01:07 2010
@@ -1,154 +1,5 @@
-# Makefile for zlib
-# Copyright (C) 1995-2005 Jean-loup Gailly.
-# For conditions of distribution and use, see copyright notice in zlib.h
+all:
+	- at echo "Please use ./configure first.  Thank you."
 
-# To compile and test, type:
-#    ./configure; make test
-# The call of configure is optional if you don't have special requirements
-# If you wish to build zlib as a shared library, use: ./configure -s
-
-# To use the asm code, type:
-#    cp contrib/asm?86/match.S ./match.S
-#    make LOC=-DASMV OBJA=match.o
-
-# To install /usr/local/lib/libz.* and /usr/local/include/zlib.h, type:
-#    make install
-# To install in $HOME instead of /usr/local, use:
-#    make install prefix=$HOME
-
-CC=cc
-
-CFLAGS=-O
-#CFLAGS=-O -DMAX_WBITS=14 -DMAX_MEM_LEVEL=7
-#CFLAGS=-g -DDEBUG
-#CFLAGS=-O3 -Wall -Wwrite-strings -Wpointer-arith -Wconversion \
-#           -Wstrict-prototypes -Wmissing-prototypes
-
-LDFLAGS=libz.a
-LDSHARED=$(CC)
-CPP=$(CC) -E
-
-LIBS=libz.a
-SHAREDLIB=libz.so
-SHAREDLIBV=libz.so.1.2.3
-SHAREDLIBM=libz.so.1
-
-AR=ar rc
-RANLIB=ranlib
-TAR=tar
-SHELL=/bin/sh
-EXE=
-
-prefix = /usr/local
-exec_prefix = ${prefix}
-libdir = ${exec_prefix}/lib
-includedir = ${prefix}/include
-mandir = ${prefix}/share/man
-man3dir = ${mandir}/man3
-
-OBJS = adler32.o compress.o crc32.o gzio.o uncompr.o deflate.o trees.o \
-       zutil.o inflate.o infback.o inftrees.o inffast.o
-
-OBJA =
-# to use the asm code: make OBJA=match.o
-
-TEST_OBJS = example.o minigzip.o
-
-all: example$(EXE) minigzip$(EXE)
-
-check: test
-test: all
-	@LD_LIBRARY_PATH=.:$(LD_LIBRARY_PATH) ; export LD_LIBRARY_PATH; \
-	echo hello world | ./minigzip | ./minigzip -d || \
-	  echo '		*** minigzip test FAILED ***' ; \
-	if ./example; then \
-	  echo '		*** zlib test OK ***'; \
-	else \
-	  echo '		*** zlib test FAILED ***'; \
-	fi
-
-libz.a: $(OBJS) $(OBJA)
-	$(AR) $@ $(OBJS) $(OBJA)
-	-@ ($(RANLIB) $@ || true) >/dev/null 2>&1
-
-match.o: match.S
-	$(CPP) match.S > _match.s
-	$(CC) -c _match.s
-	mv _match.o match.o
-	rm -f _match.s
-
-$(SHAREDLIBV): $(OBJS)
-	$(LDSHARED) -o $@ $(OBJS)
-	rm -f $(SHAREDLIB) $(SHAREDLIBM)
-	ln -s $@ $(SHAREDLIB)
-	ln -s $@ $(SHAREDLIBM)
-
-example$(EXE): example.o $(LIBS)
-	$(CC) $(CFLAGS) -o $@ example.o $(LDFLAGS)
-
-minigzip$(EXE): minigzip.o $(LIBS)
-	$(CC) $(CFLAGS) -o $@ minigzip.o $(LDFLAGS)
-
-install: $(LIBS)
-	- at if [ ! -d $(exec_prefix) ]; then mkdir -p $(exec_prefix); fi
-	- at if [ ! -d $(includedir)  ]; then mkdir -p $(includedir); fi
-	- at if [ ! -d $(libdir)      ]; then mkdir -p $(libdir); fi
-	- at if [ ! -d $(man3dir)     ]; then mkdir -p $(man3dir); fi
-	cp zlib.h zconf.h $(includedir)
-	chmod 644 $(includedir)/zlib.h $(includedir)/zconf.h
-	cp $(LIBS) $(libdir)
-	cd $(libdir); chmod 755 $(LIBS)
-	-@(cd $(libdir); $(RANLIB) libz.a || true) >/dev/null 2>&1
-	cd $(libdir); if test -f $(SHAREDLIBV); then \
-	  rm -f $(SHAREDLIB) $(SHAREDLIBM); \
-	  ln -s $(SHAREDLIBV) $(SHAREDLIB); \
-	  ln -s $(SHAREDLIBV) $(SHAREDLIBM); \
-	  (ldconfig || true)  >/dev/null 2>&1; \
-	fi
-	cp zlib.3 $(man3dir)
-	chmod 644 $(man3dir)/zlib.3
-# The ranlib in install is needed on NeXTSTEP which checks file times
-# ldconfig is for Linux
-
-uninstall:
-	cd $(includedir); \
-	cd $(libdir); rm -f libz.a; \
-	if test -f $(SHAREDLIBV); then \
-	  rm -f $(SHAREDLIBV) $(SHAREDLIB) $(SHAREDLIBM); \
-	fi
-	cd $(man3dir); rm -f zlib.3
-
-mostlyclean: clean
-clean:
-	rm -f *.o *~ example$(EXE) minigzip$(EXE) \
-	   libz.* foo.gz so_locations \
-	   _match.s maketree contrib/infback9/*.o
-
-maintainer-clean: distclean
-distclean: clean
-	cp -p Makefile.in Makefile
-	cp -p zconf.in.h zconf.h
-	rm -f .DS_Store
-
-tags:
-	etags *.[ch]
-
-depend:
-	makedepend -- $(CFLAGS) -- *.[ch]
-
-# DO NOT DELETE THIS LINE -- make depend depends on it.
-
-adler32.o: zlib.h zconf.h
-compress.o: zlib.h zconf.h
-crc32.o: crc32.h zlib.h zconf.h
-deflate.o: deflate.h zutil.h zlib.h zconf.h
-example.o: zlib.h zconf.h
-gzio.o: zutil.h zlib.h zconf.h
-inffast.o: zutil.h zlib.h zconf.h inftrees.h inflate.h inffast.h
-inflate.o: zutil.h zlib.h zconf.h inftrees.h inflate.h inffast.h
-infback.o: zutil.h zlib.h zconf.h inftrees.h inflate.h inffast.h
-inftrees.o: zutil.h zlib.h zconf.h inftrees.h
-minigzip.o: zlib.h zconf.h
-trees.o: deflate.h zutil.h zlib.h zconf.h trees.h
-uncompr.o: zlib.h zconf.h
-zutil.o: zutil.h zlib.h zconf.h
+distclean:
+	make -f Makefile.in distclean

Modified: python/branches/import_unicode/Modules/zlib/Makefile.in
==============================================================================
--- python/branches/import_unicode/Modules/zlib/Makefile.in	(original)
+++ python/branches/import_unicode/Modules/zlib/Makefile.in	Sat Jul 31 13:01:07 2010
@@ -1,11 +1,11 @@
 # Makefile for zlib
-# Copyright (C) 1995-2005 Jean-loup Gailly.
+# Copyright (C) 1995-2010 Jean-loup Gailly.
 # For conditions of distribution and use, see copyright notice in zlib.h
 
 # To compile and test, type:
 #    ./configure; make test
-# The call of configure is optional if you don't have special requirements
-# If you wish to build zlib as a shared library, use: ./configure -s
+# Normally configure builds both a static and a shared library.
+# If you want to build just a static library, use: ./configure --static
 
 # To use the asm code, type:
 #    cp contrib/asm?86/match.S ./match.S
@@ -24,17 +24,22 @@
 #CFLAGS=-O3 -Wall -Wwrite-strings -Wpointer-arith -Wconversion \
 #           -Wstrict-prototypes -Wmissing-prototypes
 
-LDFLAGS=libz.a
+SFLAGS=-O
+LDFLAGS=
+TEST_LDFLAGS=-L. libz.a
 LDSHARED=$(CC)
 CPP=$(CC) -E
 
-LIBS=libz.a
+STATICLIB=libz.a
 SHAREDLIB=libz.so
-SHAREDLIBV=libz.so.1.2.3
+SHAREDLIBV=libz.so.1.2.5
 SHAREDLIBM=libz.so.1
+LIBS=$(STATICLIB) $(SHAREDLIBV)
 
 AR=ar rc
 RANLIB=ranlib
+LDCONFIG=ldconfig
+LDSHAREDLIBC=-lc
 TAR=tar
 SHELL=/bin/sh
 EXE=
@@ -42,33 +47,68 @@
 prefix = /usr/local
 exec_prefix = ${prefix}
 libdir = ${exec_prefix}/lib
+sharedlibdir = ${libdir}
 includedir = ${prefix}/include
 mandir = ${prefix}/share/man
 man3dir = ${mandir}/man3
+pkgconfigdir = ${libdir}/pkgconfig
 
-OBJS = adler32.o compress.o crc32.o gzio.o uncompr.o deflate.o trees.o \
-       zutil.o inflate.o infback.o inftrees.o inffast.o
+OBJC = adler32.o compress.o crc32.o deflate.o gzclose.o gzlib.o gzread.o \
+	gzwrite.o infback.o inffast.o inflate.o inftrees.o trees.o uncompr.o zutil.o
 
+PIC_OBJC = adler32.lo compress.lo crc32.lo deflate.lo gzclose.lo gzlib.lo gzread.lo \
+	gzwrite.lo infback.lo inffast.lo inflate.lo inftrees.lo trees.lo uncompr.lo zutil.lo
+
+# to use the asm code: make OBJA=match.o, PIC_OBJA=match.lo
 OBJA =
-# to use the asm code: make OBJA=match.o
+PIC_OBJA =
+
+OBJS = $(OBJC) $(OBJA)
+
+PIC_OBJS = $(PIC_OBJC) $(PIC_OBJA)
+
+all: static shared
 
-TEST_OBJS = example.o minigzip.o
+static: example$(EXE) minigzip$(EXE)
 
-all: example$(EXE) minigzip$(EXE)
+shared: examplesh$(EXE) minigzipsh$(EXE)
+
+all64: example64$(EXE) minigzip64$(EXE)
 
 check: test
-test: all
-	@LD_LIBRARY_PATH=.:$(LD_LIBRARY_PATH) ; export LD_LIBRARY_PATH; \
-	echo hello world | ./minigzip | ./minigzip -d || \
-	  echo '		*** minigzip test FAILED ***' ; \
-	if ./example; then \
+
+test: all teststatic testshared
+
+teststatic: static
+	@if echo hello world | ./minigzip | ./minigzip -d && ./example; then \
 	  echo '		*** zlib test OK ***'; \
 	else \
-	  echo '		*** zlib test FAILED ***'; \
+	  echo '		*** zlib test FAILED ***'; false; \
+	fi
+	- at rm -f foo.gz
+
+testshared: shared
+	@LD_LIBRARY_PATH=`pwd`:$(LD_LIBRARY_PATH) ; export LD_LIBRARY_PATH; \
+	LD_LIBRARYN32_PATH=`pwd`:$(LD_LIBRARYN32_PATH) ; export LD_LIBRARYN32_PATH; \
+	DYLD_LIBRARY_PATH=`pwd`:$(DYLD_LIBRARY_PATH) ; export DYLD_LIBRARY_PATH; \
+	SHLIB_PATH=`pwd`:$(SHLIB_PATH) ; export SHLIB_PATH; \
+	if echo hello world | ./minigzipsh | ./minigzipsh -d && ./examplesh; then \
+	  echo '		*** zlib shared test OK ***'; \
+	else \
+	  echo '		*** zlib shared test FAILED ***'; false; \
+	fi
+	- at rm -f foo.gz
+
+test64: all64
+	@if echo hello world | ./minigzip64 | ./minigzip64 -d && ./example64; then \
+	  echo '		*** zlib 64-bit test OK ***'; \
+	else \
+	  echo '		*** zlib 64-bit test FAILED ***'; false; \
 	fi
+	- at rm -f foo.gz
 
-libz.a: $(OBJS) $(OBJA)
-	$(AR) $@ $(OBJS) $(OBJA)
+libz.a: $(OBJS)
+	$(AR) $@ $(OBJS)
 	-@ ($(RANLIB) $@ || true) >/dev/null 2>&1
 
 match.o: match.S
@@ -77,58 +117,116 @@
 	mv _match.o match.o
 	rm -f _match.s
 
-$(SHAREDLIBV): $(OBJS)
-	$(LDSHARED) -o $@ $(OBJS)
+match.lo: match.S
+	$(CPP) match.S > _match.s
+	$(CC) -c -fPIC _match.s
+	mv _match.o match.lo
+	rm -f _match.s
+
+example64.o: example.c zlib.h zconf.h
+	$(CC) $(CFLAGS) -D_FILE_OFFSET_BITS=64 -c -o $@ example.c
+
+minigzip64.o: minigzip.c zlib.h zconf.h
+	$(CC) $(CFLAGS) -D_FILE_OFFSET_BITS=64 -c -o $@ minigzip.c
+
+.SUFFIXES: .lo
+
+.c.lo:
+	- at mkdir objs 2>/dev/null || test -d objs
+	$(CC) $(SFLAGS) -DPIC -c -o objs/$*.o $<
+	- at mv objs/$*.o $@
+
+$(SHAREDLIBV): $(PIC_OBJS)
+	$(LDSHARED) $(SFLAGS) -o $@ $(PIC_OBJS) $(LDSHAREDLIBC) $(LDFLAGS)
 	rm -f $(SHAREDLIB) $(SHAREDLIBM)
 	ln -s $@ $(SHAREDLIB)
 	ln -s $@ $(SHAREDLIBM)
+	- at rmdir objs
 
-example$(EXE): example.o $(LIBS)
-	$(CC) $(CFLAGS) -o $@ example.o $(LDFLAGS)
+example$(EXE): example.o $(STATICLIB)
+	$(CC) $(CFLAGS) -o $@ example.o $(TEST_LDFLAGS)
 
-minigzip$(EXE): minigzip.o $(LIBS)
-	$(CC) $(CFLAGS) -o $@ minigzip.o $(LDFLAGS)
+minigzip$(EXE): minigzip.o $(STATICLIB)
+	$(CC) $(CFLAGS) -o $@ minigzip.o $(TEST_LDFLAGS)
 
-install: $(LIBS)
-	- at if [ ! -d $(exec_prefix) ]; then mkdir -p $(exec_prefix); fi
-	- at if [ ! -d $(includedir)  ]; then mkdir -p $(includedir); fi
-	- at if [ ! -d $(libdir)      ]; then mkdir -p $(libdir); fi
-	- at if [ ! -d $(man3dir)     ]; then mkdir -p $(man3dir); fi
-	cp zlib.h zconf.h $(includedir)
-	chmod 644 $(includedir)/zlib.h $(includedir)/zconf.h
-	cp $(LIBS) $(libdir)
-	cd $(libdir); chmod 755 $(LIBS)
-	-@(cd $(libdir); $(RANLIB) libz.a || true) >/dev/null 2>&1
-	cd $(libdir); if test -f $(SHAREDLIBV); then \
+examplesh$(EXE): example.o $(SHAREDLIBV)
+	$(CC) $(CFLAGS) -o $@ example.o -L. $(SHAREDLIBV)
+
+minigzipsh$(EXE): minigzip.o $(SHAREDLIBV)
+	$(CC) $(CFLAGS) -o $@ minigzip.o -L. $(SHAREDLIBV)
+
+example64$(EXE): example64.o $(STATICLIB)
+	$(CC) $(CFLAGS) -o $@ example64.o $(TEST_LDFLAGS)
+
+minigzip64$(EXE): minigzip64.o $(STATICLIB)
+	$(CC) $(CFLAGS) -o $@ minigzip64.o $(TEST_LDFLAGS)
+
+install-libs: $(LIBS)
+	- at if [ ! -d $(DESTDIR)$(exec_prefix)  ]; then mkdir -p $(DESTDIR)$(exec_prefix); fi
+	- at if [ ! -d $(DESTDIR)$(libdir)       ]; then mkdir -p $(DESTDIR)$(libdir); fi
+	- at if [ ! -d $(DESTDIR)$(sharedlibdir) ]; then mkdir -p $(DESTDIR)$(sharedlibdir); fi
+	- at if [ ! -d $(DESTDIR)$(man3dir)      ]; then mkdir -p $(DESTDIR)$(man3dir); fi
+	- at if [ ! -d $(DESTDIR)$(pkgconfigdir) ]; then mkdir -p $(DESTDIR)$(pkgconfigdir); fi
+	cp $(STATICLIB) $(DESTDIR)$(libdir)
+	cp $(SHAREDLIBV) $(DESTDIR)$(sharedlibdir)
+	cd $(DESTDIR)$(libdir); chmod u=rw,go=r $(STATICLIB)
+	-@(cd $(DESTDIR)$(libdir); $(RANLIB) libz.a || true) >/dev/null 2>&1
+	- at cd $(DESTDIR)$(sharedlibdir); if test "$(SHAREDLIBV)" -a -f $(SHAREDLIBV); then \
+	  chmod 755 $(SHAREDLIBV); \
 	  rm -f $(SHAREDLIB) $(SHAREDLIBM); \
 	  ln -s $(SHAREDLIBV) $(SHAREDLIB); \
 	  ln -s $(SHAREDLIBV) $(SHAREDLIBM); \
-	  (ldconfig || true)  >/dev/null 2>&1; \
+	  ($(LDCONFIG) || true)  >/dev/null 2>&1; \
 	fi
-	cp zlib.3 $(man3dir)
-	chmod 644 $(man3dir)/zlib.3
+	cp zlib.3 $(DESTDIR)$(man3dir)
+	chmod 644 $(DESTDIR)$(man3dir)/zlib.3
+	cp zlib.pc $(DESTDIR)$(pkgconfigdir)
+	chmod 644 $(DESTDIR)$(pkgconfigdir)/zlib.pc
 # The ranlib in install is needed on NeXTSTEP which checks file times
 # ldconfig is for Linux
 
+install: install-libs
+	- at if [ ! -d $(DESTDIR)$(includedir)   ]; then mkdir -p $(DESTDIR)$(includedir); fi
+	cp zlib.h zconf.h $(DESTDIR)$(includedir)
+	chmod 644 $(DESTDIR)$(includedir)/zlib.h $(DESTDIR)$(includedir)/zconf.h
+
 uninstall:
-	cd $(includedir); \
-	cd $(libdir); rm -f libz.a; \
-	if test -f $(SHAREDLIBV); then \
+	cd $(DESTDIR)$(includedir); rm -f zlib.h zconf.h
+	cd $(DESTDIR)$(libdir); rm -f libz.a; \
+	if test "$(SHAREDLIBV)" -a -f $(SHAREDLIBV); then \
 	  rm -f $(SHAREDLIBV) $(SHAREDLIB) $(SHAREDLIBM); \
 	fi
-	cd $(man3dir); rm -f zlib.3
+	cd $(DESTDIR)$(man3dir); rm -f zlib.3
+	cd $(DESTDIR)$(pkgconfigdir); rm -f zlib.pc
+
+docs: zlib.3.pdf
+
+zlib.3.pdf: zlib.3
+	groff -mandoc -f H -T ps zlib.3 | ps2pdf - zlib.3.pdf
+
+zconf.h.in: zconf.h.cmakein
+	sed "/^#cmakedefine/D" < zconf.h.cmakein > zconf.h.in
+	touch -r zconf.h.cmakein zconf.h.in
+
+zconf: zconf.h.in
+	cp -p zconf.h.in zconf.h
 
 mostlyclean: clean
 clean:
-	rm -f *.o *~ example$(EXE) minigzip$(EXE) \
+	rm -f *.o *.lo *~ \
+	   example$(EXE) minigzip$(EXE) examplesh$(EXE) minigzipsh$(EXE) \
+	   example64$(EXE) minigzip64$(EXE) \
 	   libz.* foo.gz so_locations \
 	   _match.s maketree contrib/infback9/*.o
+	rm -rf objs
 
 maintainer-clean: distclean
-distclean: clean
-	cp -p Makefile.in Makefile
-	cp -p zconf.in.h zconf.h
-	rm -f .DS_Store
+distclean: clean zconf docs
+	rm -f Makefile zlib.pc
+	- at rm -f .DS_Store
+	- at printf 'all:\n\t- at echo "Please use ./configure first.  Thank you."\n' > Makefile
+	- at printf '\ndistclean:\n\tmake -f Makefile.in distclean\n' >> Makefile
+	- at touch -r Makefile.in Makefile
 
 tags:
 	etags *.[ch]
@@ -138,17 +236,22 @@
 
 # DO NOT DELETE THIS LINE -- make depend depends on it.
 
-adler32.o: zlib.h zconf.h
-compress.o: zlib.h zconf.h
-crc32.o: crc32.h zlib.h zconf.h
+adler32.o zutil.o: zutil.h zlib.h zconf.h
+gzclose.o gzlib.o gzread.o gzwrite.o: zlib.h zconf.h gzguts.h
+compress.o example.o minigzip.o uncompr.o: zlib.h zconf.h
+crc32.o: zutil.h zlib.h zconf.h crc32.h
 deflate.o: deflate.h zutil.h zlib.h zconf.h
-example.o: zlib.h zconf.h
-gzio.o: zutil.h zlib.h zconf.h
+infback.o inflate.o: zutil.h zlib.h zconf.h inftrees.h inflate.h inffast.h inffixed.h
 inffast.o: zutil.h zlib.h zconf.h inftrees.h inflate.h inffast.h
-inflate.o: zutil.h zlib.h zconf.h inftrees.h inflate.h inffast.h
-infback.o: zutil.h zlib.h zconf.h inftrees.h inflate.h inffast.h
 inftrees.o: zutil.h zlib.h zconf.h inftrees.h
-minigzip.o: zlib.h zconf.h
 trees.o: deflate.h zutil.h zlib.h zconf.h trees.h
-uncompr.o: zlib.h zconf.h
-zutil.o: zutil.h zlib.h zconf.h
+
+adler32.lo zutil.lo: zutil.h zlib.h zconf.h
+gzclose.lo gzlib.lo gzread.lo gzwrite.lo: zlib.h zconf.h gzguts.h
+compress.lo example.lo minigzip.lo uncompr.lo: zlib.h zconf.h
+crc32.lo: zutil.h zlib.h zconf.h crc32.h
+deflate.lo: deflate.h zutil.h zlib.h zconf.h
+infback.lo inflate.lo: zutil.h zlib.h zconf.h inftrees.h inflate.h inffast.h inffixed.h
+inffast.lo: zutil.h zlib.h zconf.h inftrees.h inflate.h inffast.h
+inftrees.lo: zutil.h zlib.h zconf.h inftrees.h
+trees.lo: deflate.h zutil.h zlib.h zconf.h trees.h

Modified: python/branches/import_unicode/Modules/zlib/README
==============================================================================
--- python/branches/import_unicode/Modules/zlib/README	(original)
+++ python/branches/import_unicode/Modules/zlib/README	Sat Jul 31 13:01:07 2010
@@ -1,56 +1,52 @@
 ZLIB DATA COMPRESSION LIBRARY
 
-zlib 1.2.3 is a general purpose data compression library.  All the code is
+zlib 1.2.5 is a general purpose data compression library.  All the code is
 thread safe.  The data format used by the zlib library is described by RFCs
 (Request for Comments) 1950 to 1952 in the files
 http://www.ietf.org/rfc/rfc1950.txt (zlib format), rfc1951.txt (deflate format)
-and rfc1952.txt (gzip format). These documents are also available in other
-formats from ftp://ftp.uu.net/graphics/png/documents/zlib/zdoc-index.html
+and rfc1952.txt (gzip format).
 
 All functions of the compression library are documented in the file zlib.h
-(volunteer to write man pages welcome, contact zlib at gzip.org). A usage example
+(volunteer to write man pages welcome, contact zlib at gzip.org).  A usage example
 of the library is given in the file example.c which also tests that the library
-is working correctly. Another example is given in the file minigzip.c. The
+is working correctly.  Another example is given in the file minigzip.c.  The
 compression library itself is composed of all source files except example.c and
 minigzip.c.
 
 To compile all files and run the test program, follow the instructions given at
-the top of Makefile. In short "make test; make install" should work for most
-machines. For Unix: "./configure; make test; make install". For MSDOS, use one
-of the special makefiles such as Makefile.msc. For VMS, use make_vms.com.
+the top of Makefile.in.  In short "./configure; make test", and if that goes
+well, "make install" should work for most flavors of Unix.  For Windows, use one
+of the special makefiles in win32/ or contrib/vstudio/ .  For VMS, use
+make_vms.com.
 
 Questions about zlib should be sent to <zlib at gzip.org>, or to Gilles Vollant
-<info at winimage.com> for the Windows DLL version. The zlib home page is
-http://www.zlib.org or http://www.gzip.org/zlib/ Before reporting a problem,
-please check this site to verify that you have the latest version of zlib;
-otherwise get the latest version and check whether the problem still exists or
-not.
-
-PLEASE read the zlib FAQ http://www.gzip.org/zlib/zlib_faq.html before asking
-for help.
-
-Mark Nelson <markn at ieee.org> wrote an article about zlib for the Jan. 1997
-issue of  Dr. Dobb's Journal; a copy of the article is available in
-http://dogma.net/markn/articles/zlibtool/zlibtool.htm
-
-The changes made in version 1.2.3 are documented in the file ChangeLog.
-
-Unsupported third party contributions are provided in directory "contrib".
-
-A Java implementation of zlib is available in the Java Development Kit
-http://java.sun.com/j2se/1.4.2/docs/api/java/util/zip/package-summary.html
-See the zlib home page http://www.zlib.org for details.
-
-A Perl interface to zlib written by Paul Marquess <pmqs at cpan.org> is in the
-CPAN (Comprehensive Perl Archive Network) sites
-http://www.cpan.org/modules/by-module/Compress/
+<info at winimage.com> for the Windows DLL version.  The zlib home page is
+http://zlib.net/ .  Before reporting a problem, please check this site to
+verify that you have the latest version of zlib; otherwise get the latest
+version and check whether the problem still exists or not.
+
+PLEASE read the zlib FAQ http://zlib.net/zlib_faq.html before asking for help.
+
+Mark Nelson <markn at ieee.org> wrote an article about zlib for the Jan.  1997
+issue of Dr.  Dobb's Journal; a copy of the article is available at
+http://marknelson.us/1997/01/01/zlib-engine/ .
+
+The changes made in version 1.2.5 are documented in the file ChangeLog.
+
+Unsupported third party contributions are provided in directory contrib/ .
+
+zlib is available in Java using the java.util.zip package, documented at
+http://java.sun.com/developer/technicalArticles/Programming/compression/ .
+
+A Perl interface to zlib written by Paul Marquess <pmqs at cpan.org> is available
+at CPAN (Comprehensive Perl Archive Network) sites, including
+http://search.cpan.org/~pmqs/IO-Compress-Zlib/ .
 
 A Python interface to zlib written by A.M. Kuchling <amk at amk.ca> is
 available in Python 1.5 and later versions, see
-http://www.python.org/doc/lib/module-zlib.html
+http://www.python.org/doc/lib/module-zlib.html .
 
-A zlib binding for TCL written by Andreas Kupries <a.kupries at westend.com> is
-availlable at http://www.oche.de/~akupries/soft/trf/trf_zip.html
+zlib is built into tcl: http://wiki.tcl.tk/4610 .
 
 An experimental package to read and write files in .zip format, written on top
 of zlib by Gilles Vollant <info at winimage.com>, is available in the
@@ -74,25 +70,21 @@
 - zlib doesn't work on HP-UX 9.05 with some versions of /bin/cc. It works with
   other compilers. Use "make test" to check your compiler.
 
-- gzdopen is not supported on BEOS and by some Mac compilers.
+- gzdopen is not supported on RISCOS or BEOS.
 
 - For PalmOs, see http://palmzlib.sourceforge.net/
 
-- When building a shared, i.e. dynamic library on Mac OS X, the library must be
-  installed before testing (do "make install" before "make test"), since the
-  library location is specified in the library.
-
 
 Acknowledgments:
 
-  The deflate format used by zlib was defined by Phil Katz. The deflate
-  and zlib specifications were written by L. Peter Deutsch. Thanks to all the
-  people who reported problems and suggested various improvements in zlib;
-  they are too numerous to cite here.
+  The deflate format used by zlib was defined by Phil Katz.  The deflate and
+  zlib specifications were written by L.  Peter Deutsch.  Thanks to all the
+  people who reported problems and suggested various improvements in zlib; they
+  are too numerous to cite here.
 
 Copyright notice:
 
- (C) 1995-2004 Jean-loup Gailly and Mark Adler
+ (C) 1995-2010 Jean-loup Gailly and Mark Adler
 
   This software is provided 'as-is', without any express or implied
   warranty.  In no event will the authors be held liable for any damages
@@ -113,13 +105,11 @@
   Jean-loup Gailly        Mark Adler
   jloup at gzip.org          madler at alumni.caltech.edu
 
-If you use the zlib library in a product, we would appreciate *not*
-receiving lengthy legal documents to sign. The sources are provided
-for free but without warranty of any kind.  The library has been
-entirely written by Jean-loup Gailly and Mark Adler; it does not
-include third-party code.
-
-If you redistribute modified sources, we would appreciate that you include
-in the file ChangeLog history information documenting your changes. Please
-read the FAQ for more information on the distribution of modified source
-versions.
+If you use the zlib library in a product, we would appreciate *not* receiving
+lengthy legal documents to sign.  The sources are provided for free but without
+warranty of any kind.  The library has been entirely written by Jean-loup
+Gailly and Mark Adler; it does not include third-party code.
+
+If you redistribute modified sources, we would appreciate that you include in
+the file ChangeLog history information documenting your changes.  Please read
+the FAQ for more information on the distribution of modified source versions.

Modified: python/branches/import_unicode/Modules/zlib/adler32.c
==============================================================================
--- python/branches/import_unicode/Modules/zlib/adler32.c	(original)
+++ python/branches/import_unicode/Modules/zlib/adler32.c	Sat Jul 31 13:01:07 2010
@@ -1,12 +1,15 @@
 /* adler32.c -- compute the Adler-32 checksum of a data stream
- * Copyright (C) 1995-2004 Mark Adler
+ * Copyright (C) 1995-2007 Mark Adler
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
 /* @(#) $Id$ */
 
-#define ZLIB_INTERNAL
-#include "zlib.h"
+#include "zutil.h"
+
+#define local static
+
+local uLong adler32_combine_(uLong adler1, uLong adler2, z_off64_t len2);
 
 #define BASE 65521UL    /* largest prime smaller than 65536 */
 #define NMAX 5552
@@ -125,10 +128,10 @@
 }
 
 /* ========================================================================= */
-uLong ZEXPORT adler32_combine(adler1, adler2, len2)
+local uLong adler32_combine_(adler1, adler2, len2)
     uLong adler1;
     uLong adler2;
-    z_off_t len2;
+    z_off64_t len2;
 {
     unsigned long sum1;
     unsigned long sum2;
@@ -141,9 +144,26 @@
     MOD(sum2);
     sum1 += (adler2 & 0xffff) + BASE - 1;
     sum2 += ((adler1 >> 16) & 0xffff) + ((adler2 >> 16) & 0xffff) + BASE - rem;
-    if (sum1 > BASE) sum1 -= BASE;
-    if (sum1 > BASE) sum1 -= BASE;
-    if (sum2 > (BASE << 1)) sum2 -= (BASE << 1);
-    if (sum2 > BASE) sum2 -= BASE;
+    if (sum1 >= BASE) sum1 -= BASE;
+    if (sum1 >= BASE) sum1 -= BASE;
+    if (sum2 >= (BASE << 1)) sum2 -= (BASE << 1);
+    if (sum2 >= BASE) sum2 -= BASE;
     return sum1 | (sum2 << 16);
 }
+
+/* ========================================================================= */
+uLong ZEXPORT adler32_combine(adler1, adler2, len2)
+    uLong adler1;
+    uLong adler2;
+    z_off_t len2;
+{
+    return adler32_combine_(adler1, adler2, len2);
+}
+
+uLong ZEXPORT adler32_combine64(adler1, adler2, len2)
+    uLong adler1;
+    uLong adler2;
+    z_off64_t len2;
+{
+    return adler32_combine_(adler1, adler2, len2);
+}

Modified: python/branches/import_unicode/Modules/zlib/compress.c
==============================================================================
--- python/branches/import_unicode/Modules/zlib/compress.c	(original)
+++ python/branches/import_unicode/Modules/zlib/compress.c	Sat Jul 31 13:01:07 2010
@@ -1,5 +1,5 @@
 /* compress.c -- compress a memory buffer
- * Copyright (C) 1995-2003 Jean-loup Gailly.
+ * Copyright (C) 1995-2005 Jean-loup Gailly.
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -75,5 +75,6 @@
 uLong ZEXPORT compressBound (sourceLen)
     uLong sourceLen;
 {
-    return sourceLen + (sourceLen >> 12) + (sourceLen >> 14) + 11;
+    return sourceLen + (sourceLen >> 12) + (sourceLen >> 14) +
+           (sourceLen >> 25) + 13;
 }

Modified: python/branches/import_unicode/Modules/zlib/configure
==============================================================================
--- python/branches/import_unicode/Modules/zlib/configure	(original)
+++ python/branches/import_unicode/Modules/zlib/configure	Sat Jul 31 13:01:07 2010
@@ -1,37 +1,61 @@
 #!/bin/sh
-# configure script for zlib. This script is needed only if
-# you wish to build a shared library and your system supports them,
-# of if you need special compiler, flags or install directory.
-# Otherwise, you can just use directly "make test; make install"
+# configure script for zlib.
 #
-# To create a shared library, use "configure --shared"; by default a static
-# library is created. If the primitive shared library support provided here
-# does not work, use ftp://prep.ai.mit.edu/pub/gnu/libtool-*.tar.gz
+# Normally configure builds both a static and a shared library.
+# If you want to build just a static library, use: ./configure --static
 #
 # To impose specific compiler or flags or install directory, use for example:
 #    prefix=$HOME CC=cc CFLAGS="-O4" ./configure
 # or for csh/tcsh users:
 #    (setenv prefix $HOME; setenv CC cc; setenv CFLAGS "-O4"; ./configure)
-# LDSHARED is the command to be used to create a shared library
 
 # Incorrect settings of CC or CFLAGS may prevent creating a shared library.
 # If you have problems, try without defining CC and CFLAGS before reporting
 # an error.
 
-LIBS=libz.a
-LDFLAGS="-L. ${LIBS}"
+if [ -n "${CHOST}" ]; then
+    uname="$(echo "${CHOST}" | sed -e 's/^[^-]*-\([^-]*\)$/\1/' -e 's/^[^-]*-[^-]*-\([^-]*\)$/\1/' -e 's/^[^-]*-[^-]*-\([^-]*\)-.*$/\1/')"
+    CROSS_PREFIX="${CHOST}-"
+fi
+
+STATICLIB=libz.a
+LDFLAGS="${LDFLAGS} -L. ${STATICLIB}"
 VER=`sed -n -e '/VERSION "/s/.*"\(.*\)".*/\1/p' < zlib.h`
+VER3=`sed -n -e '/VERSION "/s/.*"\([0-9]*\\.[0-9]*\\.[0-9]*\).*/\1/p' < zlib.h`
 VER2=`sed -n -e '/VERSION "/s/.*"\([0-9]*\\.[0-9]*\)\\..*/\1/p' < zlib.h`
 VER1=`sed -n -e '/VERSION "/s/.*"\([0-9]*\)\\..*/\1/p' < zlib.h`
-AR=${AR-"ar rc"}
-RANLIB=${RANLIB-"ranlib"}
+if "${CROSS_PREFIX}ar" --version >/dev/null 2>/dev/null || test $? -lt 126; then
+    AR=${AR-"${CROSS_PREFIX}ar"}
+    test -n "${CROSS_PREFIX}" && echo Using ${AR}
+else
+    AR=${AR-"ar"}
+    test -n "${CROSS_PREFIX}" && echo Using ${AR}
+fi
+AR_RC="${AR} rc"
+if "${CROSS_PREFIX}ranlib" --version >/dev/null 2>/dev/null || test $? -lt 126; then
+    RANLIB=${RANLIB-"${CROSS_PREFIX}ranlib"}
+    test -n "${CROSS_PREFIX}" && echo Using ${RANLIB}
+else
+    RANLIB=${RANLIB-"ranlib"}
+fi
+if "${CROSS_PREFIX}nm" --version >/dev/null 2>/dev/null || test $? -lt 126; then
+    NM=${NM-"${CROSS_PREFIX}nm"}
+    test -n "${CROSS_PREFIX}" && echo Using ${NM}
+else
+    NM=${NM-"nm"}
+fi
+LDCONFIG=${LDCONFIG-"ldconfig"}
+LDSHAREDLIBC="${LDSHAREDLIBC--lc}"
 prefix=${prefix-/usr/local}
 exec_prefix=${exec_prefix-'${prefix}'}
 libdir=${libdir-'${exec_prefix}/lib'}
+sharedlibdir=${sharedlibdir-'${libdir}'}
 includedir=${includedir-'${prefix}/include'}
 mandir=${mandir-'${prefix}/share/man'}
 shared_ext='.so'
-shared=0
+shared=1
+zprefix=0
+build64=0
 gcc=0
 old_cc="$CC"
 old_cflags="$CFLAGS"
@@ -39,21 +63,29 @@
 while test $# -ge 1
 do
 case "$1" in
-    -h* | --h*)
+    -h* | --help)
       echo 'usage:'
-      echo '  configure [--shared] [--prefix=PREFIX]  [--exec_prefix=EXPREFIX]'
-      echo '     [--libdir=LIBDIR] [--includedir=INCLUDEDIR]'
-        exit 0;;
-    -p*=* | --p*=*) prefix=`echo $1 | sed 's/[-a-z_]*=//'`; shift;;
-    -e*=* | --e*=*) exec_prefix=`echo $1 | sed 's/[-a-z_]*=//'`; shift;;
-    -l*=* | --libdir=*) libdir=`echo $1 | sed 's/[-a-z_]*=//'`; shift;;
-    -i*=* | --includedir=*) includedir=`echo $1 | sed 's/[-a-z_]*=//'`;shift;;
-    -p* | --p*) prefix="$2"; shift; shift;;
-    -e* | --e*) exec_prefix="$2"; shift; shift;;
-    -l* | --l*) libdir="$2"; shift; shift;;
-    -i* | --i*) includedir="$2"; shift; shift;;
-    -s* | --s*) shared=1; shift;;
-    *) echo "unknown option: $1"; echo "$0 --help for help"; exit 1;;
+      echo '  configure [--zprefix] [--prefix=PREFIX]  [--eprefix=EXPREFIX]'
+      echo '    [--static] [--64] [--libdir=LIBDIR] [--sharedlibdir=LIBDIR]'
+      echo '    [--includedir=INCLUDEDIR]'
+        exit 0 ;;
+    -p*=* | --prefix=*) prefix=`echo $1 | sed 's/.*=//'`; shift ;;
+    -e*=* | --eprefix=*) exec_prefix=`echo $1 | sed 's/.*=//'`; shift ;;
+    -l*=* | --libdir=*) libdir=`echo $1 | sed 's/.*=//'`; shift ;;
+    --sharedlibdir=*) sharedlibdir=`echo $1 | sed 's/.*=//'`; shift ;;
+    -i*=* | --includedir=*) includedir=`echo $1 | sed 's/.*=//'`;shift ;;
+    -u*=* | --uname=*) uname=`echo $1 | sed 's/.*=//'`;shift ;;
+    -p* | --prefix) prefix="$2"; shift; shift ;;
+    -e* | --eprefix) exec_prefix="$2"; shift; shift ;;
+    -l* | --libdir) libdir="$2"; shift; shift ;;
+    -i* | --includedir) includedir="$2"; shift; shift ;;
+    -s* | --shared | --enable-shared) shared=1; shift ;;
+    -t | --static) shared=0; shift ;;
+    -z* | --zprefix) zprefix=1; shift ;;
+    -6* | --64) build64=1; shift ;;
+    --sysconfdir=*) echo "ignored option: --sysconfdir"; shift ;;
+    --localstatedir=*) echo "ignored option: --localstatedir"; shift ;;
+    *) echo "unknown option: $1"; echo "$0 --help for help"; exit 1 ;;
     esac
 done
 
@@ -63,46 +95,68 @@
 int hello() {return getchar();}
 EOF
 
-test -z "$CC" && echo Checking for gcc...
-cc=${CC-gcc}
+test -z "$CC" && echo Checking for ${CROSS_PREFIX}gcc...
+cc=${CC-${CROSS_PREFIX}gcc}
 cflags=${CFLAGS-"-O3"}
 # to force the asm version use: CFLAGS="-O3 -DASMV" ./configure
 case "$cc" in
-  *gcc*) gcc=1;;
+  *gcc*) gcc=1 ;;
 esac
 
 if test "$gcc" -eq 1 && ($cc -c $cflags $test.c) 2>/dev/null; then
   CC="$cc"
-  SFLAGS=${CFLAGS-"-fPIC -O3"}
-  CFLAGS="$cflags"
-  case `(uname -s || echo unknown) 2>/dev/null` in
-  Linux | linux | GNU | GNU/*) LDSHARED=${LDSHARED-"$cc -shared -Wl,-soname,libz.so.1"};;
-  CYGWIN* | Cygwin* | cygwin* | OS/2* )
-             EXE='.exe';;
+  SFLAGS="${CFLAGS--O3} -fPIC"
+  CFLAGS="${CFLAGS--O3}"
+  if test $build64 -eq 1; then
+    CFLAGS="${CFLAGS} -m64"
+    SFLAGS="${SFLAGS} -m64"
+  fi
+  if test "${ZLIBGCCWARN}" = "YES"; then
+    CFLAGS="${CFLAGS} -Wall -Wextra -pedantic"
+  fi
+  if test -z "$uname"; then
+    uname=`(uname -s || echo unknown) 2>/dev/null`
+  fi
+  case "$uname" in
+  Linux* | linux* | GNU | GNU/* | *BSD | DragonFly) LDSHARED=${LDSHARED-"$cc -shared -Wl,-soname,libz.so.1,--version-script,zlib.map"} ;;
+  CYGWIN* | Cygwin* | cygwin* | OS/2*)
+        EXE='.exe' ;;
+  MINGW*|mingw*)
+# temporary bypass
+        rm -f $test.[co] $test $test$shared_ext
+        echo "Please use win32/Makefile.gcc instead."
+        exit 1
+        LDSHARED=${LDSHARED-"$cc -shared"}
+        LDSHAREDLIBC=""
+        EXE='.exe' ;;
   QNX*)  # This is for QNX6. I suppose that the QNX rule below is for QNX2,QNX4
          # (alain.bonnefoy at icbt.com)
-                 LDSHARED=${LDSHARED-"$cc -shared -Wl,-hlibz.so.1"};;
+                 LDSHARED=${LDSHARED-"$cc -shared -Wl,-hlibz.so.1"} ;;
   HP-UX*)
          LDSHARED=${LDSHARED-"$cc -shared $SFLAGS"}
          case `(uname -m || echo unknown) 2>/dev/null` in
          ia64)
                  shared_ext='.so'
-                 SHAREDLIB='libz.so';;
+                 SHAREDLIB='libz.so' ;;
          *)
                  shared_ext='.sl'
-                 SHAREDLIB='libz.sl';;
-         esac;;
+                 SHAREDLIB='libz.sl' ;;
+         esac ;;
   Darwin*)   shared_ext='.dylib'
              SHAREDLIB=libz$shared_ext
              SHAREDLIBV=libz.$VER$shared_ext
              SHAREDLIBM=libz.$VER1$shared_ext
-             LDSHARED=${LDSHARED-"$cc -dynamiclib -install_name $libdir/$SHAREDLIBM -compatibility_version $VER1 -current_version $VER"};;
-  *)             LDSHARED=${LDSHARED-"$cc -shared"};;
+             LDSHARED=${LDSHARED-"$cc -dynamiclib -install_name $libdir/$SHAREDLIBM -compatibility_version $VER1 -current_version $VER3"} ;;
+  *)             LDSHARED=${LDSHARED-"$cc -shared"} ;;
   esac
 else
   # find system name and corresponding cc options
   CC=${CC-cc}
-  case `(uname -sr || echo unknown) 2>/dev/null` in
+  gcc=0
+  if test -z "$uname"; then
+    uname=`(uname -sr || echo unknown) 2>/dev/null`
+  fi
+  case "$uname" in
   HP-UX*)    SFLAGS=${CFLAGS-"-O +z"}
              CFLAGS=${CFLAGS-"-O"}
 #            LDSHARED=${LDSHARED-"ld -b +vnocompatwarnings"}
@@ -110,57 +164,64 @@
          case `(uname -m || echo unknown) 2>/dev/null` in
          ia64)
              shared_ext='.so'
-             SHAREDLIB='libz.so';;
+             SHAREDLIB='libz.so' ;;
          *)
              shared_ext='.sl'
-             SHAREDLIB='libz.sl';;
-         esac;;
+             SHAREDLIB='libz.sl' ;;
+         esac ;;
   IRIX*)     SFLAGS=${CFLAGS-"-ansi -O2 -rpath ."}
              CFLAGS=${CFLAGS-"-ansi -O2"}
-             LDSHARED=${LDSHARED-"cc -shared"};;
+             LDSHARED=${LDSHARED-"cc -shared -Wl,-soname,libz.so.1"} ;;
   OSF1\ V4*) SFLAGS=${CFLAGS-"-O -std1"}
              CFLAGS=${CFLAGS-"-O -std1"}
-             LDSHARED=${LDSHARED-"cc -shared  -Wl,-soname,libz.so -Wl,-msym -Wl,-rpath,$(libdir) -Wl,-set_version,${VER}:1.0"};;
+             LDFLAGS="${LDFLAGS} -Wl,-rpath,."
+             LDSHARED=${LDSHARED-"cc -shared  -Wl,-soname,libz.so -Wl,-msym -Wl,-rpath,$(libdir) -Wl,-set_version,${VER}:1.0"} ;;
   OSF1*)     SFLAGS=${CFLAGS-"-O -std1"}
              CFLAGS=${CFLAGS-"-O -std1"}
-             LDSHARED=${LDSHARED-"cc -shared"};;
+             LDSHARED=${LDSHARED-"cc -shared -Wl,-soname,libz.so.1"} ;;
   QNX*)      SFLAGS=${CFLAGS-"-4 -O"}
              CFLAGS=${CFLAGS-"-4 -O"}
              LDSHARED=${LDSHARED-"cc"}
              RANLIB=${RANLIB-"true"}
-             AR="cc -A";;
+             AR_RC="cc -A" ;;
   SCO_SV\ 3.2*) SFLAGS=${CFLAGS-"-O3 -dy -KPIC "}
              CFLAGS=${CFLAGS-"-O3"}
-             LDSHARED=${LDSHARED-"cc -dy -KPIC -G"};;
-  SunOS\ 5*) SFLAGS=${CFLAGS-"-fast -xcg89 -KPIC -R."}
-             CFLAGS=${CFLAGS-"-fast -xcg89"}
-             LDSHARED=${LDSHARED-"cc -G"};;
+             LDSHARED=${LDSHARED-"cc -dy -KPIC -G"} ;;
+  SunOS\ 5*) LDSHARED=${LDSHARED-"cc -G"}
+         case `(uname -m || echo unknown) 2>/dev/null` in
+         i86*)
+             SFLAGS=${CFLAGS-"-xpentium -fast -KPIC -R."}
+             CFLAGS=${CFLAGS-"-xpentium -fast"} ;;
+         *)
+             SFLAGS=${CFLAGS-"-fast -xcg92 -KPIC -R."}
+             CFLAGS=${CFLAGS-"-fast -xcg92"} ;;
+         esac ;;
   SunOS\ 4*) SFLAGS=${CFLAGS-"-O2 -PIC"}
              CFLAGS=${CFLAGS-"-O2"}
-             LDSHARED=${LDSHARED-"ld"};;
-  SunStudio\ 9*) SFLAGS=${CFLAGS-"-DUSE_MMAP -fast -xcode=pic32 -xtarget=ultra3 -xarch=v9b"}
-             CFLAGS=${CFLAGS-"-DUSE_MMAP -fast -xtarget=ultra3 -xarch=v9b"}
-             LDSHARED=${LDSHARED-"cc -xarch=v9b"};;
+             LDSHARED=${LDSHARED-"ld"} ;;
+  SunStudio\ 9*) SFLAGS=${CFLAGS-"-fast -xcode=pic32 -xtarget=ultra3 -xarch=v9b"}
+             CFLAGS=${CFLAGS-"-fast -xtarget=ultra3 -xarch=v9b"}
+             LDSHARED=${LDSHARED-"cc -xarch=v9b"} ;;
   UNIX_System_V\ 4.2.0)
              SFLAGS=${CFLAGS-"-KPIC -O"}
              CFLAGS=${CFLAGS-"-O"}
-             LDSHARED=${LDSHARED-"cc -G"};;
+             LDSHARED=${LDSHARED-"cc -G"} ;;
   UNIX_SV\ 4.2MP)
              SFLAGS=${CFLAGS-"-Kconform_pic -O"}
              CFLAGS=${CFLAGS-"-O"}
-             LDSHARED=${LDSHARED-"cc -G"};;
+             LDSHARED=${LDSHARED-"cc -G"} ;;
   OpenUNIX\ 5)
              SFLAGS=${CFLAGS-"-KPIC -O"}
              CFLAGS=${CFLAGS-"-O"}
-             LDSHARED=${LDSHARED-"cc -G"};;
+             LDSHARED=${LDSHARED-"cc -G"} ;;
   AIX*)  # Courtesy of dbakker at arrayasolutions.com
              SFLAGS=${CFLAGS-"-O -qmaxmem=8192"}
              CFLAGS=${CFLAGS-"-O -qmaxmem=8192"}
-             LDSHARED=${LDSHARED-"xlc -G"};;
-  # send working options for other systems to support at gzip.org
+             LDSHARED=${LDSHARED-"xlc -G"} ;;
+  # send working options for other systems to zlib at gzip.org
   *)         SFLAGS=${CFLAGS-"-O"}
              CFLAGS=${CFLAGS-"-O"}
-             LDSHARED=${LDSHARED-"cc -shared"};;
+             LDSHARED=${LDSHARED-"cc -shared"} ;;
   esac
 fi
 
@@ -171,38 +232,83 @@
 if test $shared -eq 1; then
   echo Checking for shared library support...
   # we must test in two steps (cc then ld), required at least on SunOS 4.x
-  if test "`($CC -c $SFLAGS $test.c) 2>&1`" = "" &&
-     test "`($LDSHARED -o $test$shared_ext $test.o) 2>&1`" = ""; then
-    CFLAGS="$SFLAGS"
-    LIBS="$SHAREDLIBV"
+  if test "`($CC -w -c $SFLAGS $test.c) 2>&1`" = "" &&
+     test "`($LDSHARED $SFLAGS -o $test$shared_ext $test.o) 2>&1`" = ""; then
     echo Building shared library $SHAREDLIBV with $CC.
   elif test -z "$old_cc" -a -z "$old_cflags"; then
     echo No shared library support.
     shared=0;
   else
+    echo Tested $CC -w -c $SFLAGS $test.c
+    $CC -w -c $SFLAGS $test.c
+    echo Tested $LDSHARED $SFLAGS -o $test$shared_ext $test.o
+    $LDSHARED $SFLAGS -o $test$shared_ext $test.o
     echo 'No shared library support; try without defining CC and CFLAGS'
     shared=0;
   fi
 fi
 if test $shared -eq 0; then
   LDSHARED="$CC"
-  echo Building static library $LIBS version $VER with $CC.
+  ALL="static"
+  TEST="all teststatic"
+  SHAREDLIB=""
+  SHAREDLIBV=""
+  SHAREDLIBM=""
+  echo Building static library $STATICLIB version $VER with $CC.
 else
-  LDFLAGS="-L. ${SHAREDLIBV}"
+  ALL="static shared"
+  TEST="all teststatic testshared"
 fi
 
 cat > $test.c <<EOF
+#include <sys/types.h>
+off64_t dummy = 0;
+EOF
+if test "`($CC -c $CFLAGS -D_LARGEFILE64_SOURCE=1 $test.c) 2>&1`" = ""; then
+  CFLAGS="${CFLAGS} -D_LARGEFILE64_SOURCE=1"
+  SFLAGS="${SFLAGS} -D_LARGEFILE64_SOURCE=1"
+  ALL="${ALL} all64"
+  TEST="${TEST} test64"
+  echo "Checking for off64_t... Yes."
+  echo "Checking for fseeko... Yes."
+else
+  echo "Checking for off64_t... No."
+  cat > $test.c <<EOF
+#include <stdio.h>
+int main(void) {
+  fseeko(NULL, 0, 0);
+  return 0;
+}
+EOF
+  if test "`($CC $CFLAGS -o $test $test.c) 2>&1`" = ""; then
+    echo "Checking for fseeko... Yes."
+  else
+    CFLAGS="${CFLAGS} -DNO_FSEEKO"
+    SFLAGS="${SFLAGS} -DNO_FSEEKO"
+    echo "Checking for fseeko... No."
+  fi
+fi
+
+cp -p zconf.h.in zconf.h
+
+cat > $test.c <<EOF
 #include <unistd.h>
 int main() { return 0; }
 EOF
 if test "`($CC -c $CFLAGS $test.c) 2>&1`" = ""; then
-  sed < zconf.in.h "/HAVE_UNISTD_H/s%0%1%" > zconf.h
+  sed < zconf.h "/^#ifdef HAVE_UNISTD_H.* may be/s/def HAVE_UNISTD_H\(.*\) may be/ 1\1 was/" > zconf.temp.h
+  mv zconf.temp.h zconf.h
   echo "Checking for unistd.h... Yes."
 else
-  cp -p zconf.in.h zconf.h
   echo "Checking for unistd.h... No."
 fi
 
+if test $zprefix -eq 1; then
+  sed < zconf.h "/#ifdef Z_PREFIX.* may be/s/def Z_PREFIX\(.*\) may be/ 1\1 was/" > zconf.temp.h
+  mv zconf.temp.h zconf.h
+  echo "Using z_ prefix on all symbols."
+fi
+
 cat > $test.c <<EOF
 #include <stdio.h>
 #include <stdarg.h>
@@ -219,13 +325,13 @@
 EOF
 
 if test "`($CC -c $CFLAGS $test.c) 2>&1`" = ""; then
-  echo "Checking whether to use vs[n]printf() or s[n]printf()... using vs[n]printf()"
+  echo "Checking whether to use vs[n]printf() or s[n]printf()... using vs[n]printf()."
 
   cat > $test.c <<EOF
 #include <stdio.h>
 #include <stdarg.h>
 
-int mytest(char *fmt, ...)
+int mytest(const char *fmt, ...)
 {
   char buf[20];
   va_list ap;
@@ -249,7 +355,7 @@
 #include <stdio.h>
 #include <stdarg.h>
 
-int mytest(char *fmt, ...)
+int mytest(const char *fmt, ...)
 {
   int n;
   char buf[20];
@@ -271,6 +377,7 @@
       echo "Checking for return value of vsnprintf()... Yes."
     else
       CFLAGS="$CFLAGS -DHAS_vsnprintf_void"
+      SFLAGS="$SFLAGS -DHAS_vsnprintf_void"
       echo "Checking for return value of vsnprintf()... No."
       echo "  WARNING: apparently vsnprintf() does not return a value. zlib"
       echo "  can build but will be open to possible string-format security"
@@ -278,6 +385,7 @@
     fi
   else
     CFLAGS="$CFLAGS -DNO_vsnprintf"
+    SFLAGS="$SFLAGS -DNO_vsnprintf"
     echo "Checking for vsnprintf() in stdio.h... No."
     echo "  WARNING: vsnprintf() not found, falling back to vsprintf(). zlib"
     echo "  can build but will be open to possible buffer-overflow security"
@@ -287,7 +395,7 @@
 #include <stdio.h>
 #include <stdarg.h>
 
-int mytest(char *fmt, ...)
+int mytest(const char *fmt, ...)
 {
   int n;
   char buf[20];
@@ -309,6 +417,7 @@
       echo "Checking for return value of vsprintf()... Yes."
     else
       CFLAGS="$CFLAGS -DHAS_vsprintf_void"
+      SFLAGS="$SFLAGS -DHAS_vsprintf_void"
       echo "Checking for return value of vsprintf()... No."
       echo "  WARNING: apparently vsprintf() does not return a value. zlib"
       echo "  can build but will be open to possible string-format security"
@@ -316,7 +425,7 @@
     fi
   fi
 else
-  echo "Checking whether to use vs[n]printf() or s[n]printf()... using s[n]printf()"
+  echo "Checking whether to use vs[n]printf() or s[n]printf()... using s[n]printf()."
 
   cat >$test.c <<EOF
 #include <stdio.h>
@@ -358,6 +467,7 @@
       echo "Checking for return value of snprintf()... Yes."
     else
       CFLAGS="$CFLAGS -DHAS_snprintf_void"
+      SFLAGS="$SFLAGS -DHAS_snprintf_void"
       echo "Checking for return value of snprintf()... No."
       echo "  WARNING: apparently snprintf() does not return a value. zlib"
       echo "  can build but will be open to possible string-format security"
@@ -365,6 +475,7 @@
     fi
   else
     CFLAGS="$CFLAGS -DNO_snprintf"
+    SFLAGS="$SFLAGS -DNO_snprintf"
     echo "Checking for snprintf() in stdio.h... No."
     echo "  WARNING: snprintf() not found, falling back to sprintf(). zlib"
     echo "  can build but will be open to possible buffer-overflow security"
@@ -390,6 +501,7 @@
       echo "Checking for return value of sprintf()... Yes."
     else
       CFLAGS="$CFLAGS -DHAS_sprintf_void"
+      SFLAGS="$SFLAGS -DHAS_sprintf_void"
       echo "Checking for return value of sprintf()... No."
       echo "  WARNING: apparently sprintf() does not return a value. zlib"
       echo "  can build but will be open to possible string-format security"
@@ -398,41 +510,37 @@
   fi
 fi
 
-cat >$test.c <<EOF
-#include <errno.h>
-int main() { return 0; }
-EOF
-if test "`($CC -c $CFLAGS $test.c) 2>&1`" = ""; then
-  echo "Checking for errno.h... Yes."
-else
-  echo "Checking for errno.h... No."
-  CFLAGS="$CFLAGS -DNO_ERRNO_H"
-fi
-
-cat > $test.c <<EOF
-#include <sys/types.h>
-#include <sys/mman.h>
-#include <sys/stat.h>
-caddr_t hello() {
-  return mmap((caddr_t)0, (off_t)0, PROT_READ, MAP_SHARED, 0, (off_t)0);
+if test "$gcc" -eq 1; then
+  cat > $test.c <<EOF
+#if ((__GNUC__-0) * 10 + __GNUC_MINOR__-0 >= 33)
+#  define ZLIB_INTERNAL __attribute__((visibility ("hidden")))
+#else
+#  define ZLIB_INTERNAL
+#endif
+int ZLIB_INTERNAL foo;
+int main()
+{
+  return 0;
 }
 EOF
-if test "`($CC -c $CFLAGS $test.c) 2>&1`" = ""; then
-  CFLAGS="$CFLAGS -DUSE_MMAP"
-  echo Checking for mmap support... Yes.
-else
-  echo Checking for mmap support... No.
+  if test "`($CC -c $CFLAGS $test.c) 2>&1`" = ""; then
+    echo "Checking for attribute(visibility) support... Yes."
+  else
+    CFLAGS="$CFLAGS -DNO_VIZ"
+    SFLAGS="$SFLAGS -DNO_VIZ"
+    echo "Checking for attribute(visibility) support... No."
+  fi
 fi
 
 CPP=${CPP-"$CC -E"}
 case $CFLAGS in
   *ASMV*)
-    if test "`nm $test.o | grep _hello`" = ""; then
+    if test "`$NM $test.o | grep _hello`" = ""; then
       CPP="$CPP -DNO_UNDERLINE"
       echo Checking for underline in external names... No.
     else
       echo Checking for underline in external names... Yes.
-    fi;;
+    fi ;;
 esac
 
 rm -f $test.[co] $test $test$shared_ext
@@ -441,19 +549,48 @@
 sed < Makefile.in "
 /^CC *=/s#=.*#=$CC#
 /^CFLAGS *=/s#=.*#=$CFLAGS#
+/^SFLAGS *=/s#=.*#=$SFLAGS#
+/^LDFLAGS *=/s#=.*#=$LDFLAGS#
+/^LDSHARED *=/s#=.*#=$LDSHARED#
+/^CPP *=/s#=.*#=$CPP#
+/^STATICLIB *=/s#=.*#=$STATICLIB#
+/^SHAREDLIB *=/s#=.*#=$SHAREDLIB#
+/^SHAREDLIBV *=/s#=.*#=$SHAREDLIBV#
+/^SHAREDLIBM *=/s#=.*#=$SHAREDLIBM#
+/^AR *=/s#=.*#=$AR_RC#
+/^RANLIB *=/s#=.*#=$RANLIB#
+/^LDCONFIG *=/s#=.*#=$LDCONFIG#
+/^LDSHAREDLIBC *=/s#=.*#=$LDSHAREDLIBC#
+/^EXE *=/s#=.*#=$EXE#
+/^prefix *=/s#=.*#=$prefix#
+/^exec_prefix *=/s#=.*#=$exec_prefix#
+/^libdir *=/s#=.*#=$libdir#
+/^sharedlibdir *=/s#=.*#=$sharedlibdir#
+/^includedir *=/s#=.*#=$includedir#
+/^mandir *=/s#=.*#=$mandir#
+/^all: */s#:.*#: $ALL#
+/^test: */s#:.*#: $TEST#
+" > Makefile
+
+sed < zlib.pc.in "
+/^CC *=/s#=.*#=$CC#
+/^CFLAGS *=/s#=.*#=$CFLAGS#
 /^CPP *=/s#=.*#=$CPP#
 /^LDSHARED *=/s#=.*#=$LDSHARED#
-/^LIBS *=/s#=.*#=$LIBS#
+/^STATICLIB *=/s#=.*#=$STATICLIB#
 /^SHAREDLIB *=/s#=.*#=$SHAREDLIB#
 /^SHAREDLIBV *=/s#=.*#=$SHAREDLIBV#
 /^SHAREDLIBM *=/s#=.*#=$SHAREDLIBM#
-/^AR *=/s#=.*#=$AR#
+/^AR *=/s#=.*#=$AR_RC#
 /^RANLIB *=/s#=.*#=$RANLIB#
 /^EXE *=/s#=.*#=$EXE#
 /^prefix *=/s#=.*#=$prefix#
 /^exec_prefix *=/s#=.*#=$exec_prefix#
 /^libdir *=/s#=.*#=$libdir#
+/^sharedlibdir *=/s#=.*#=$sharedlibdir#
 /^includedir *=/s#=.*#=$includedir#
 /^mandir *=/s#=.*#=$mandir#
 /^LDFLAGS *=/s#=.*#=$LDFLAGS#
-" > Makefile
+" | sed -e "
+s/\@VERSION\@/$VER/g;
+" > zlib.pc

Modified: python/branches/import_unicode/Modules/zlib/crc32.c
==============================================================================
--- python/branches/import_unicode/Modules/zlib/crc32.c	(original)
+++ python/branches/import_unicode/Modules/zlib/crc32.c	Sat Jul 31 13:01:07 2010
@@ -1,5 +1,5 @@
 /* crc32.c -- compute the CRC-32 of a data stream
- * Copyright (C) 1995-2005 Mark Adler
+ * Copyright (C) 1995-2006, 2010 Mark Adler
  * For conditions of distribution and use, see copyright notice in zlib.h
  *
  * Thanks to Rodney Brown <rbrown64 at csc.com.au> for his contribution of faster
@@ -53,7 +53,7 @@
 
 /* Definitions for doing the crc four data bytes at a time. */
 #ifdef BYFOUR
-#  define REV(w) (((w)>>24)+(((w)>>8)&0xff00)+ \
+#  define REV(w) ((((w)>>24)&0xff)+(((w)>>8)&0xff00)+ \
                 (((w)&0xff00)<<8)+(((w)&0xff)<<24))
    local unsigned long crc32_little OF((unsigned long,
                         const unsigned char FAR *, unsigned));
@@ -68,6 +68,8 @@
 local unsigned long gf2_matrix_times OF((unsigned long *mat,
                                          unsigned long vec));
 local void gf2_matrix_square OF((unsigned long *square, unsigned long *mat));
+local uLong crc32_combine_(uLong crc1, uLong crc2, z_off64_t len2);
+
 
 #ifdef DYNAMIC_CRC_TABLE
 
@@ -219,7 +221,7 @@
 unsigned long ZEXPORT crc32(crc, buf, len)
     unsigned long crc;
     const unsigned char FAR *buf;
-    unsigned len;
+    uInt len;
 {
     if (buf == Z_NULL) return 0UL;
 
@@ -367,22 +369,22 @@
 }
 
 /* ========================================================================= */
-uLong ZEXPORT crc32_combine(crc1, crc2, len2)
+local uLong crc32_combine_(crc1, crc2, len2)
     uLong crc1;
     uLong crc2;
-    z_off_t len2;
+    z_off64_t len2;
 {
     int n;
     unsigned long row;
     unsigned long even[GF2_DIM];    /* even-power-of-two zeros operator */
     unsigned long odd[GF2_DIM];     /* odd-power-of-two zeros operator */
 
-    /* degenerate case */
-    if (len2 == 0)
+    /* degenerate case (also disallow negative lengths) */
+    if (len2 <= 0)
         return crc1;
 
     /* put operator for one zero bit in odd */
-    odd[0] = 0xedb88320L;           /* CRC-32 polynomial */
+    odd[0] = 0xedb88320UL;          /* CRC-32 polynomial */
     row = 1;
     for (n = 1; n < GF2_DIM; n++) {
         odd[n] = row;
@@ -421,3 +423,20 @@
     crc1 ^= crc2;
     return crc1;
 }
+
+/* ========================================================================= */
+uLong ZEXPORT crc32_combine(crc1, crc2, len2)
+    uLong crc1;
+    uLong crc2;
+    z_off_t len2;
+{
+    return crc32_combine_(crc1, crc2, len2);
+}
+
+uLong ZEXPORT crc32_combine64(crc1, crc2, len2)
+    uLong crc1;
+    uLong crc2;
+    z_off64_t len2;
+{
+    return crc32_combine_(crc1, crc2, len2);
+}

Modified: python/branches/import_unicode/Modules/zlib/deflate.c
==============================================================================
--- python/branches/import_unicode/Modules/zlib/deflate.c	(original)
+++ python/branches/import_unicode/Modules/zlib/deflate.c	Sat Jul 31 13:01:07 2010
@@ -1,5 +1,5 @@
 /* deflate.c -- compress data using the deflation algorithm
- * Copyright (C) 1995-2005 Jean-loup Gailly.
+ * Copyright (C) 1995-2010 Jean-loup Gailly and Mark Adler
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -52,7 +52,7 @@
 #include "deflate.h"
 
 const char deflate_copyright[] =
-   " deflate 1.2.3 Copyright 1995-2005 Jean-loup Gailly ";
+   " deflate 1.2.5 Copyright 1995-2010 Jean-loup Gailly and Mark Adler ";
 /*
   If you use the zlib library in a product, an acknowledgment is welcome
   in the documentation of your product. If for some reason you cannot
@@ -79,19 +79,18 @@
 #ifndef FASTEST
 local block_state deflate_slow   OF((deflate_state *s, int flush));
 #endif
+local block_state deflate_rle    OF((deflate_state *s, int flush));
+local block_state deflate_huff   OF((deflate_state *s, int flush));
 local void lm_init        OF((deflate_state *s));
 local void putShortMSB    OF((deflate_state *s, uInt b));
 local void flush_pending  OF((z_streamp strm));
 local int read_buf        OF((z_streamp strm, Bytef *buf, unsigned size));
-#ifndef FASTEST
 #ifdef ASMV
       void match_init OF((void)); /* asm code initialization */
       uInt longest_match  OF((deflate_state *s, IPos cur_match));
 #else
 local uInt longest_match  OF((deflate_state *s, IPos cur_match));
 #endif
-#endif
-local uInt longest_match_fast OF((deflate_state *s, IPos cur_match));
 
 #ifdef DEBUG
 local  void check_match OF((deflate_state *s, IPos start, IPos match,
@@ -110,11 +109,6 @@
 #endif
 /* Matches of length 3 are discarded if their distance exceeds TOO_FAR */
 
-#define MIN_LOOKAHEAD (MAX_MATCH+MIN_MATCH+1)
-/* Minimum amount of lookahead, except at the end of the input file.
- * See deflate.c for comments about the MIN_MATCH+1.
- */
-
 /* Values for max_lazy_match, good_match and max_chain_length, depending on
  * the desired pack level (0..9). The values given below have been tuned to
  * exclude worst case performance for pathological files. Better values may be
@@ -288,6 +282,8 @@
     s->prev   = (Posf *)  ZALLOC(strm, s->w_size, sizeof(Pos));
     s->head   = (Posf *)  ZALLOC(strm, s->hash_size, sizeof(Pos));
 
+    s->high_water = 0;      /* nothing written to s->window yet */
+
     s->lit_bufsize = 1 << (memLevel + 6); /* 16K elements by default */
 
     overlay = (ushf *) ZALLOC(strm, s->lit_bufsize, sizeof(ush)+2);
@@ -332,8 +328,8 @@
         strm->adler = adler32(strm->adler, dictionary, dictLength);
 
     if (length < MIN_MATCH) return Z_OK;
-    if (length > MAX_DIST(s)) {
-        length = MAX_DIST(s);
+    if (length > s->w_size) {
+        length = s->w_size;
         dictionary += dictLength - length; /* use the tail of the dictionary */
     }
     zmemcpy(s->window, dictionary, length);
@@ -435,9 +431,10 @@
     }
     func = configuration_table[s->level].func;
 
-    if (func != configuration_table[level].func && strm->total_in != 0) {
+    if ((strategy != s->strategy || func != configuration_table[level].func) &&
+        strm->total_in != 0) {
         /* Flush the last buffer: */
-        err = deflate(strm, Z_PARTIAL_FLUSH);
+        err = deflate(strm, Z_BLOCK);
     }
     if (s->level != level) {
         s->level = level;
@@ -481,33 +478,66 @@
  * resulting from using fixed blocks instead of stored blocks, which deflate
  * can emit on compressed data for some combinations of the parameters.
  *
- * This function could be more sophisticated to provide closer upper bounds
- * for every combination of windowBits and memLevel, as well as wrap.
- * But even the conservative upper bound of about 14% expansion does not
- * seem onerous for output buffer allocation.
+ * This function could be more sophisticated to provide closer upper bounds for
+ * every combination of windowBits and memLevel.  But even the conservative
+ * upper bound of about 14% expansion does not seem onerous for output buffer
+ * allocation.
  */
 uLong ZEXPORT deflateBound(strm, sourceLen)
     z_streamp strm;
     uLong sourceLen;
 {
     deflate_state *s;
-    uLong destLen;
+    uLong complen, wraplen;
+    Bytef *str;
 
-    /* conservative upper bound */
-    destLen = sourceLen +
-              ((sourceLen + 7) >> 3) + ((sourceLen + 63) >> 6) + 11;
+    /* conservative upper bound for compressed data */
+    complen = sourceLen +
+              ((sourceLen + 7) >> 3) + ((sourceLen + 63) >> 6) + 5;
 
-    /* if can't get parameters, return conservative bound */
+    /* if can't get parameters, return conservative bound plus zlib wrapper */
     if (strm == Z_NULL || strm->state == Z_NULL)
-        return destLen;
+        return complen + 6;
 
-    /* if not default parameters, return conservative bound */
+    /* compute wrapper length */
     s = strm->state;
+    switch (s->wrap) {
+    case 0:                                 /* raw deflate */
+        wraplen = 0;
+        break;
+    case 1:                                 /* zlib wrapper */
+        wraplen = 6 + (s->strstart ? 4 : 0);
+        break;
+    case 2:                                 /* gzip wrapper */
+        wraplen = 18;
+        if (s->gzhead != Z_NULL) {          /* user-supplied gzip header */
+            if (s->gzhead->extra != Z_NULL)
+                wraplen += 2 + s->gzhead->extra_len;
+            str = s->gzhead->name;
+            if (str != Z_NULL)
+                do {
+                    wraplen++;
+                } while (*str++);
+            str = s->gzhead->comment;
+            if (str != Z_NULL)
+                do {
+                    wraplen++;
+                } while (*str++);
+            if (s->gzhead->hcrc)
+                wraplen += 2;
+        }
+        break;
+    default:                                /* for compiler happiness */
+        wraplen = 6;
+    }
+
+    /* if not default parameters, return conservative bound */
     if (s->w_bits != 15 || s->hash_bits != 8 + 7)
-        return destLen;
+        return complen + wraplen;
 
     /* default settings: return tight bound for that case */
-    return compressBound(sourceLen);
+    return sourceLen + (sourceLen >> 12) + (sourceLen >> 14) +
+           (sourceLen >> 25) + 13 - 6 + wraplen;
 }
 
 /* =========================================================================
@@ -557,7 +587,7 @@
     deflate_state *s;
 
     if (strm == Z_NULL || strm->state == Z_NULL ||
-        flush > Z_FINISH || flush < 0) {
+        flush > Z_BLOCK || flush < 0) {
         return Z_STREAM_ERROR;
     }
     s = strm->state;
@@ -581,7 +611,7 @@
             put_byte(s, 31);
             put_byte(s, 139);
             put_byte(s, 8);
-            if (s->gzhead == NULL) {
+            if (s->gzhead == Z_NULL) {
                 put_byte(s, 0);
                 put_byte(s, 0);
                 put_byte(s, 0);
@@ -608,7 +638,7 @@
                             (s->strategy >= Z_HUFFMAN_ONLY || s->level < 2 ?
                              4 : 0));
                 put_byte(s, s->gzhead->os & 0xff);
-                if (s->gzhead->extra != NULL) {
+                if (s->gzhead->extra != Z_NULL) {
                     put_byte(s, s->gzhead->extra_len & 0xff);
                     put_byte(s, (s->gzhead->extra_len >> 8) & 0xff);
                 }
@@ -650,7 +680,7 @@
     }
 #ifdef GZIP
     if (s->status == EXTRA_STATE) {
-        if (s->gzhead->extra != NULL) {
+        if (s->gzhead->extra != Z_NULL) {
             uInt beg = s->pending;  /* start of bytes to update crc */
 
             while (s->gzindex < (s->gzhead->extra_len & 0xffff)) {
@@ -678,7 +708,7 @@
             s->status = NAME_STATE;
     }
     if (s->status == NAME_STATE) {
-        if (s->gzhead->name != NULL) {
+        if (s->gzhead->name != Z_NULL) {
             uInt beg = s->pending;  /* start of bytes to update crc */
             int val;
 
@@ -709,7 +739,7 @@
             s->status = COMMENT_STATE;
     }
     if (s->status == COMMENT_STATE) {
-        if (s->gzhead->comment != NULL) {
+        if (s->gzhead->comment != Z_NULL) {
             uInt beg = s->pending;  /* start of bytes to update crc */
             int val;
 
@@ -787,7 +817,9 @@
         (flush != Z_NO_FLUSH && s->status != FINISH_STATE)) {
         block_state bstate;
 
-        bstate = (*(configuration_table[s->level].func))(s, flush);
+        bstate = s->strategy == Z_HUFFMAN_ONLY ? deflate_huff(s, flush) :
+                    (s->strategy == Z_RLE ? deflate_rle(s, flush) :
+                        (*(configuration_table[s->level].func))(s, flush));
 
         if (bstate == finish_started || bstate == finish_done) {
             s->status = FINISH_STATE;
@@ -808,13 +840,17 @@
         if (bstate == block_done) {
             if (flush == Z_PARTIAL_FLUSH) {
                 _tr_align(s);
-            } else { /* FULL_FLUSH or SYNC_FLUSH */
+            } else if (flush != Z_BLOCK) { /* FULL_FLUSH or SYNC_FLUSH */
                 _tr_stored_block(s, (char*)0, 0L, 0);
                 /* For a full flush, this empty block will be recognized
                  * as a special marker by inflate_sync().
                  */
                 if (flush == Z_FULL_FLUSH) {
                     CLEAR_HASH(s);             /* forget history */
+                    if (s->lookahead == 0) {
+                        s->strstart = 0;
+                        s->block_start = 0L;
+                    }
                 }
             }
             flush_pending(strm);
@@ -1167,12 +1203,13 @@
     return s->lookahead;
 }
 #endif /* ASMV */
-#endif /* FASTEST */
+
+#else /* FASTEST */
 
 /* ---------------------------------------------------------------------------
- * Optimized version for level == 1 or strategy == Z_RLE only
+ * Optimized version for FASTEST only
  */
-local uInt longest_match_fast(s, cur_match)
+local uInt longest_match(s, cur_match)
     deflate_state *s;
     IPos cur_match;                             /* current match */
 {
@@ -1225,6 +1262,8 @@
     return (uInt)len <= s->lookahead ? (uInt)len : s->lookahead;
 }
 
+#endif /* FASTEST */
+
 #ifdef DEBUG
 /* ===========================================================================
  * Check that the match at match_start is indeed a match.
@@ -1303,7 +1342,6 @@
                later. (Using level 0 permanently is not an optimal usage of
                zlib, so we don't care about this pathological case.)
              */
-            /* %%% avoid this when Z_RLE */
             n = s->hash_size;
             p = &s->head[n];
             do {
@@ -1355,27 +1393,61 @@
          */
 
     } while (s->lookahead < MIN_LOOKAHEAD && s->strm->avail_in != 0);
+
+    /* If the WIN_INIT bytes after the end of the current data have never been
+     * written, then zero those bytes in order to avoid memory check reports of
+     * the use of uninitialized (or uninitialised as Julian writes) bytes by
+     * the longest match routines.  Update the high water mark for the next
+     * time through here.  WIN_INIT is set to MAX_MATCH since the longest match
+     * routines allow scanning to strstart + MAX_MATCH, ignoring lookahead.
+     */
+    if (s->high_water < s->window_size) {
+        ulg curr = s->strstart + (ulg)(s->lookahead);
+        ulg init;
+
+        if (s->high_water < curr) {
+            /* Previous high water mark below current data -- zero WIN_INIT
+             * bytes or up to end of window, whichever is less.
+             */
+            init = s->window_size - curr;
+            if (init > WIN_INIT)
+                init = WIN_INIT;
+            zmemzero(s->window + curr, (unsigned)init);
+            s->high_water = curr + init;
+        }
+        else if (s->high_water < (ulg)curr + WIN_INIT) {
+            /* High water mark at or above current data, but below current data
+             * plus WIN_INIT -- zero out to current data plus WIN_INIT, or up
+             * to end of window, whichever is less.
+             */
+            init = (ulg)curr + WIN_INIT - s->high_water;
+            if (init > s->window_size - s->high_water)
+                init = s->window_size - s->high_water;
+            zmemzero(s->window + s->high_water, (unsigned)init);
+            s->high_water += init;
+        }
+    }
 }
 
 /* ===========================================================================
  * Flush the current block, with given end-of-file flag.
  * IN assertion: strstart is set to the end of the current match.
  */
-#define FLUSH_BLOCK_ONLY(s, eof) { \
+#define FLUSH_BLOCK_ONLY(s, last) { \
    _tr_flush_block(s, (s->block_start >= 0L ? \
                    (charf *)&s->window[(unsigned)s->block_start] : \
                    (charf *)Z_NULL), \
                 (ulg)((long)s->strstart - s->block_start), \
-                (eof)); \
+                (last)); \
    s->block_start = s->strstart; \
    flush_pending(s->strm); \
    Tracev((stderr,"[FLUSH]")); \
 }
 
 /* Same but force premature exit if necessary. */
-#define FLUSH_BLOCK(s, eof) { \
-   FLUSH_BLOCK_ONLY(s, eof); \
-   if (s->strm->avail_out == 0) return (eof) ? finish_started : need_more; \
+#define FLUSH_BLOCK(s, last) { \
+   FLUSH_BLOCK_ONLY(s, last); \
+   if (s->strm->avail_out == 0) return (last) ? finish_started : need_more; \
 }
 
 /* ===========================================================================
@@ -1449,7 +1521,7 @@
     deflate_state *s;
     int flush;
 {
-    IPos hash_head = NIL; /* head of the hash chain */
+    IPos hash_head;       /* head of the hash chain */
     int bflush;           /* set if current block must be flushed */
 
     for (;;) {
@@ -1469,6 +1541,7 @@
         /* Insert the string window[strstart .. strstart+2] in the
          * dictionary, and set hash_head to the head of the hash chain:
          */
+        hash_head = NIL;
         if (s->lookahead >= MIN_MATCH) {
             INSERT_STRING(s, s->strstart, hash_head);
         }
@@ -1481,19 +1554,8 @@
              * of window index 0 (in particular we have to avoid a match
              * of the string with itself at the start of the input file).
              */
-#ifdef FASTEST
-            if ((s->strategy != Z_HUFFMAN_ONLY && s->strategy != Z_RLE) ||
-                (s->strategy == Z_RLE && s->strstart - hash_head == 1)) {
-                s->match_length = longest_match_fast (s, hash_head);
-            }
-#else
-            if (s->strategy != Z_HUFFMAN_ONLY && s->strategy != Z_RLE) {
-                s->match_length = longest_match (s, hash_head);
-            } else if (s->strategy == Z_RLE && s->strstart - hash_head == 1) {
-                s->match_length = longest_match_fast (s, hash_head);
-            }
-#endif
-            /* longest_match() or longest_match_fast() sets match_start */
+            s->match_length = longest_match (s, hash_head);
+            /* longest_match() sets match_start */
         }
         if (s->match_length >= MIN_MATCH) {
             check_match(s, s->strstart, s->match_start, s->match_length);
@@ -1555,7 +1617,7 @@
     deflate_state *s;
     int flush;
 {
-    IPos hash_head = NIL;    /* head of hash chain */
+    IPos hash_head;          /* head of hash chain */
     int bflush;              /* set if current block must be flushed */
 
     /* Process the input block. */
@@ -1576,6 +1638,7 @@
         /* Insert the string window[strstart .. strstart+2] in the
          * dictionary, and set hash_head to the head of the hash chain:
          */
+        hash_head = NIL;
         if (s->lookahead >= MIN_MATCH) {
             INSERT_STRING(s, s->strstart, hash_head);
         }
@@ -1591,12 +1654,8 @@
              * of window index 0 (in particular we have to avoid a match
              * of the string with itself at the start of the input file).
              */
-            if (s->strategy != Z_HUFFMAN_ONLY && s->strategy != Z_RLE) {
-                s->match_length = longest_match (s, hash_head);
-            } else if (s->strategy == Z_RLE && s->strstart - hash_head == 1) {
-                s->match_length = longest_match_fast (s, hash_head);
-            }
-            /* longest_match() or longest_match_fast() sets match_start */
+            s->match_length = longest_match (s, hash_head);
+            /* longest_match() sets match_start */
 
             if (s->match_length <= 5 && (s->strategy == Z_FILTERED
 #if TOO_FAR <= 32767
@@ -1674,7 +1733,6 @@
 }
 #endif /* FASTEST */
 
-#if 0
 /* ===========================================================================
  * For Z_RLE, simply look for runs of bytes, generate matches only of distance
  * one.  Do not maintain a hash table.  (It will be regenerated if this run of
@@ -1684,11 +1742,9 @@
     deflate_state *s;
     int flush;
 {
-    int bflush;         /* set if current block must be flushed */
-    uInt run;           /* length of run */
-    uInt max;           /* maximum length of run */
-    uInt prev;          /* byte at distance one to match */
-    Bytef *scan;        /* scan for end of run */
+    int bflush;             /* set if current block must be flushed */
+    uInt prev;              /* byte at distance one to match */
+    Bytef *scan, *strend;   /* scan goes up to strend for length of run */
 
     for (;;) {
         /* Make sure that we always have enough lookahead, except
@@ -1704,23 +1760,33 @@
         }
 
         /* See how many times the previous byte repeats */
-        run = 0;
-        if (s->strstart > 0) {      /* if there is a previous byte, that is */
-            max = s->lookahead < MAX_MATCH ? s->lookahead : MAX_MATCH;
+        s->match_length = 0;
+        if (s->lookahead >= MIN_MATCH && s->strstart > 0) {
             scan = s->window + s->strstart - 1;
-            prev = *scan++;
-            do {
-                if (*scan++ != prev)
-                    break;
-            } while (++run < max);
+            prev = *scan;
+            if (prev == *++scan && prev == *++scan && prev == *++scan) {
+                strend = s->window + s->strstart + MAX_MATCH;
+                do {
+                } while (prev == *++scan && prev == *++scan &&
+                         prev == *++scan && prev == *++scan &&
+                         prev == *++scan && prev == *++scan &&
+                         prev == *++scan && prev == *++scan &&
+                         scan < strend);
+                s->match_length = MAX_MATCH - (int)(strend - scan);
+                if (s->match_length > s->lookahead)
+                    s->match_length = s->lookahead;
+            }
         }
 
         /* Emit match if have run of MIN_MATCH or longer, else emit literal */
-        if (run >= MIN_MATCH) {
-            check_match(s, s->strstart, s->strstart - 1, run);
-            _tr_tally_dist(s, 1, run - MIN_MATCH, bflush);
-            s->lookahead -= run;
-            s->strstart += run;
+        if (s->match_length >= MIN_MATCH) {
+            check_match(s, s->strstart, s->strstart - 1, s->match_length);
+
+            _tr_tally_dist(s, 1, s->match_length - MIN_MATCH, bflush);
+
+            s->lookahead -= s->match_length;
+            s->strstart += s->match_length;
+            s->match_length = 0;
         } else {
             /* No match, output a literal byte */
             Tracevv((stderr,"%c", s->window[s->strstart]));
@@ -1733,4 +1799,36 @@
     FLUSH_BLOCK(s, flush == Z_FINISH);
     return flush == Z_FINISH ? finish_done : block_done;
 }
-#endif
+
+/* ===========================================================================
+ * For Z_HUFFMAN_ONLY, do not look for matches.  Do not maintain a hash table.
+ * (It will be regenerated if this run of deflate switches away from Huffman.)
+ */
+local block_state deflate_huff(s, flush)
+    deflate_state *s;
+    int flush;
+{
+    int bflush;             /* set if current block must be flushed */
+
+    for (;;) {
+        /* Make sure that we have a literal to write. */
+        if (s->lookahead == 0) {
+            fill_window(s);
+            if (s->lookahead == 0) {
+                if (flush == Z_NO_FLUSH)
+                    return need_more;
+                break;      /* flush the current block */
+            }
+        }
+
+        /* Output a literal byte */
+        s->match_length = 0;
+        Tracevv((stderr,"%c", s->window[s->strstart]));
+        _tr_tally_lit (s, s->window[s->strstart], bflush);
+        s->lookahead--;
+        s->strstart++;
+        if (bflush) FLUSH_BLOCK(s, 0);
+    }
+    FLUSH_BLOCK(s, flush == Z_FINISH);
+    return flush == Z_FINISH ? finish_done : block_done;
+}

Modified: python/branches/import_unicode/Modules/zlib/deflate.h
==============================================================================
--- python/branches/import_unicode/Modules/zlib/deflate.h	(original)
+++ python/branches/import_unicode/Modules/zlib/deflate.h	Sat Jul 31 13:01:07 2010
@@ -1,5 +1,5 @@
 /* deflate.h -- internal compression state
- * Copyright (C) 1995-2004 Jean-loup Gailly
+ * Copyright (C) 1995-2010 Jean-loup Gailly
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -260,6 +260,13 @@
      * are always zero.
      */
 
+    ulg high_water;
+    /* High water mark offset in window for initialized bytes -- bytes above
+     * this are set to zero in order to avoid memory check warnings when
+     * longest match routines access bytes past the input.  This is then
+     * updated to the new high water mark.
+     */
+
 } FAR deflate_state;
 
 /* Output a byte on the stream.
@@ -278,14 +285,18 @@
  * distances are limited to MAX_DIST instead of WSIZE.
  */
 
+#define WIN_INIT MAX_MATCH
+/* Number of bytes after end of data in window to initialize in order to avoid
+   memory checker errors from longest match routines */
+
         /* in trees.c */
-void _tr_init         OF((deflate_state *s));
-int  _tr_tally        OF((deflate_state *s, unsigned dist, unsigned lc));
-void _tr_flush_block  OF((deflate_state *s, charf *buf, ulg stored_len,
-                          int eof));
-void _tr_align        OF((deflate_state *s));
-void _tr_stored_block OF((deflate_state *s, charf *buf, ulg stored_len,
-                          int eof));
+void ZLIB_INTERNAL _tr_init OF((deflate_state *s));
+int ZLIB_INTERNAL _tr_tally OF((deflate_state *s, unsigned dist, unsigned lc));
+void ZLIB_INTERNAL _tr_flush_block OF((deflate_state *s, charf *buf,
+                        ulg stored_len, int last));
+void ZLIB_INTERNAL _tr_align OF((deflate_state *s));
+void ZLIB_INTERNAL _tr_stored_block OF((deflate_state *s, charf *buf,
+                        ulg stored_len, int last));
 
 #define d_code(dist) \
    ((dist) < 256 ? _dist_code[dist] : _dist_code[256+((dist)>>7)])
@@ -298,11 +309,11 @@
 /* Inline versions of _tr_tally for speed: */
 
 #if defined(GEN_TREES_H) || !defined(STDC)
-  extern uch _length_code[];
-  extern uch _dist_code[];
+  extern uch ZLIB_INTERNAL _length_code[];
+  extern uch ZLIB_INTERNAL _dist_code[];
 #else
-  extern const uch _length_code[];
-  extern const uch _dist_code[];
+  extern const uch ZLIB_INTERNAL _length_code[];
+  extern const uch ZLIB_INTERNAL _dist_code[];
 #endif
 
 # define _tr_tally_lit(s, c, flush) \

Modified: python/branches/import_unicode/Modules/zlib/example.c
==============================================================================
--- python/branches/import_unicode/Modules/zlib/example.c	(original)
+++ python/branches/import_unicode/Modules/zlib/example.c	Sat Jul 31 13:01:07 2010
@@ -1,19 +1,19 @@
 /* example.c -- usage example of the zlib compression library
- * Copyright (C) 1995-2004 Jean-loup Gailly.
+ * Copyright (C) 1995-2006 Jean-loup Gailly.
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
 /* @(#) $Id$ */
 
-#include <stdio.h>
 #include "zlib.h"
+#include <stdio.h>
 
 #ifdef STDC
 #  include <string.h>
 #  include <stdlib.h>
 #endif
 
-#if defined(VMS)
+#if defined(VMS) || defined(RISCOS)
 #  define TESTFILE "foo-gz"
 #else
 #  define TESTFILE "foo.gz"

Modified: python/branches/import_unicode/Modules/zlib/infback.c
==============================================================================
--- python/branches/import_unicode/Modules/zlib/infback.c	(original)
+++ python/branches/import_unicode/Modules/zlib/infback.c	Sat Jul 31 13:01:07 2010
@@ -1,5 +1,5 @@
 /* infback.c -- inflate using a call-back interface
- * Copyright (C) 1995-2005 Mark Adler
+ * Copyright (C) 1995-2009 Mark Adler
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -55,7 +55,7 @@
     state->wbits = windowBits;
     state->wsize = 1U << windowBits;
     state->window = window;
-    state->write = 0;
+    state->wnext = 0;
     state->whave = 0;
     return Z_OK;
 }
@@ -253,7 +253,7 @@
     unsigned bits;              /* bits in bit buffer */
     unsigned copy;              /* number of stored or match bytes to copy */
     unsigned char FAR *from;    /* where to copy match bytes from */
-    code this;                  /* current decoding table entry */
+    code here;                  /* current decoding table entry */
     code last;                  /* parent table entry */
     unsigned len;               /* length to copy for repeats, bits to drop */
     int ret;                    /* return code */
@@ -389,19 +389,19 @@
             state->have = 0;
             while (state->have < state->nlen + state->ndist) {
                 for (;;) {
-                    this = state->lencode[BITS(state->lenbits)];
-                    if ((unsigned)(this.bits) <= bits) break;
+                    here = state->lencode[BITS(state->lenbits)];
+                    if ((unsigned)(here.bits) <= bits) break;
                     PULLBYTE();
                 }
-                if (this.val < 16) {
-                    NEEDBITS(this.bits);
-                    DROPBITS(this.bits);
-                    state->lens[state->have++] = this.val;
+                if (here.val < 16) {
+                    NEEDBITS(here.bits);
+                    DROPBITS(here.bits);
+                    state->lens[state->have++] = here.val;
                 }
                 else {
-                    if (this.val == 16) {
-                        NEEDBITS(this.bits + 2);
-                        DROPBITS(this.bits);
+                    if (here.val == 16) {
+                        NEEDBITS(here.bits + 2);
+                        DROPBITS(here.bits);
                         if (state->have == 0) {
                             strm->msg = (char *)"invalid bit length repeat";
                             state->mode = BAD;
@@ -411,16 +411,16 @@
                         copy = 3 + BITS(2);
                         DROPBITS(2);
                     }
-                    else if (this.val == 17) {
-                        NEEDBITS(this.bits + 3);
-                        DROPBITS(this.bits);
+                    else if (here.val == 17) {
+                        NEEDBITS(here.bits + 3);
+                        DROPBITS(here.bits);
                         len = 0;
                         copy = 3 + BITS(3);
                         DROPBITS(3);
                     }
                     else {
-                        NEEDBITS(this.bits + 7);
-                        DROPBITS(this.bits);
+                        NEEDBITS(here.bits + 7);
+                        DROPBITS(here.bits);
                         len = 0;
                         copy = 11 + BITS(7);
                         DROPBITS(7);
@@ -438,7 +438,16 @@
             /* handle error breaks in while */
             if (state->mode == BAD) break;
 
-            /* build code tables */
+            /* check for end-of-block code (better have one) */
+            if (state->lens[256] == 0) {
+                strm->msg = (char *)"invalid code -- missing end-of-block";
+                state->mode = BAD;
+                break;
+            }
+
+            /* build code tables -- note: do not change the lenbits or distbits
+               values here (9 and 6) without reading the comments in inftrees.h
+               concerning the ENOUGH constants, which depend on those values */
             state->next = state->codes;
             state->lencode = (code const FAR *)(state->next);
             state->lenbits = 9;
@@ -474,28 +483,28 @@
 
             /* get a literal, length, or end-of-block code */
             for (;;) {
-                this = state->lencode[BITS(state->lenbits)];
-                if ((unsigned)(this.bits) <= bits) break;
+                here = state->lencode[BITS(state->lenbits)];
+                if ((unsigned)(here.bits) <= bits) break;
                 PULLBYTE();
             }
-            if (this.op && (this.op & 0xf0) == 0) {
-                last = this;
+            if (here.op && (here.op & 0xf0) == 0) {
+                last = here;
                 for (;;) {
-                    this = state->lencode[last.val +
+                    here = state->lencode[last.val +
                             (BITS(last.bits + last.op) >> last.bits)];
-                    if ((unsigned)(last.bits + this.bits) <= bits) break;
+                    if ((unsigned)(last.bits + here.bits) <= bits) break;
                     PULLBYTE();
                 }
                 DROPBITS(last.bits);
             }
-            DROPBITS(this.bits);
-            state->length = (unsigned)this.val;
+            DROPBITS(here.bits);
+            state->length = (unsigned)here.val;
 
             /* process literal */
-            if (this.op == 0) {
-                Tracevv((stderr, this.val >= 0x20 && this.val < 0x7f ?
+            if (here.op == 0) {
+                Tracevv((stderr, here.val >= 0x20 && here.val < 0x7f ?
                         "inflate:         literal '%c'\n" :
-                        "inflate:         literal 0x%02x\n", this.val));
+                        "inflate:         literal 0x%02x\n", here.val));
                 ROOM();
                 *put++ = (unsigned char)(state->length);
                 left--;
@@ -504,21 +513,21 @@
             }
 
             /* process end of block */
-            if (this.op & 32) {
+            if (here.op & 32) {
                 Tracevv((stderr, "inflate:         end of block\n"));
                 state->mode = TYPE;
                 break;
             }
 
             /* invalid code */
-            if (this.op & 64) {
+            if (here.op & 64) {
                 strm->msg = (char *)"invalid literal/length code";
                 state->mode = BAD;
                 break;
             }
 
             /* length code -- get extra bits, if any */
-            state->extra = (unsigned)(this.op) & 15;
+            state->extra = (unsigned)(here.op) & 15;
             if (state->extra != 0) {
                 NEEDBITS(state->extra);
                 state->length += BITS(state->extra);
@@ -528,30 +537,30 @@
 
             /* get distance code */
             for (;;) {
-                this = state->distcode[BITS(state->distbits)];
-                if ((unsigned)(this.bits) <= bits) break;
+                here = state->distcode[BITS(state->distbits)];
+                if ((unsigned)(here.bits) <= bits) break;
                 PULLBYTE();
             }
-            if ((this.op & 0xf0) == 0) {
-                last = this;
+            if ((here.op & 0xf0) == 0) {
+                last = here;
                 for (;;) {
-                    this = state->distcode[last.val +
+                    here = state->distcode[last.val +
                             (BITS(last.bits + last.op) >> last.bits)];
-                    if ((unsigned)(last.bits + this.bits) <= bits) break;
+                    if ((unsigned)(last.bits + here.bits) <= bits) break;
                     PULLBYTE();
                 }
                 DROPBITS(last.bits);
             }
-            DROPBITS(this.bits);
-            if (this.op & 64) {
+            DROPBITS(here.bits);
+            if (here.op & 64) {
                 strm->msg = (char *)"invalid distance code";
                 state->mode = BAD;
                 break;
             }
-            state->offset = (unsigned)this.val;
+            state->offset = (unsigned)here.val;
 
             /* get distance extra bits, if any */
-            state->extra = (unsigned)(this.op) & 15;
+            state->extra = (unsigned)(here.op) & 15;
             if (state->extra != 0) {
                 NEEDBITS(state->extra);
                 state->offset += BITS(state->extra);

Modified: python/branches/import_unicode/Modules/zlib/inffast.c
==============================================================================
--- python/branches/import_unicode/Modules/zlib/inffast.c	(original)
+++ python/branches/import_unicode/Modules/zlib/inffast.c	Sat Jul 31 13:01:07 2010
@@ -1,5 +1,5 @@
 /* inffast.c -- fast decoding
- * Copyright (C) 1995-2004 Mark Adler
+ * Copyright (C) 1995-2008, 2010 Mark Adler
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -64,7 +64,7 @@
       requires strm->avail_out >= 258 for each loop to avoid checking for
       output space.
  */
-void inflate_fast(strm, start)
+void ZLIB_INTERNAL inflate_fast(strm, start)
 z_streamp strm;
 unsigned start;         /* inflate()'s starting value for strm->avail_out */
 {
@@ -79,7 +79,7 @@
 #endif
     unsigned wsize;             /* window size or zero if not using window */
     unsigned whave;             /* valid bytes in the window */
-    unsigned write;             /* window write index */
+    unsigned wnext;             /* window write index */
     unsigned char FAR *window;  /* allocated sliding window, if wsize != 0 */
     unsigned long hold;         /* local strm->hold */
     unsigned bits;              /* local strm->bits */
@@ -87,7 +87,7 @@
     code const FAR *dcode;      /* local strm->distcode */
     unsigned lmask;             /* mask for first level of length codes */
     unsigned dmask;             /* mask for first level of distance codes */
-    code this;                  /* retrieved table entry */
+    code here;                  /* retrieved table entry */
     unsigned op;                /* code bits, operation, extra bits, or */
                                 /*  window position, window bytes to copy */
     unsigned len;               /* match length, unused bytes */
@@ -106,7 +106,7 @@
 #endif
     wsize = state->wsize;
     whave = state->whave;
-    write = state->write;
+    wnext = state->wnext;
     window = state->window;
     hold = state->hold;
     bits = state->bits;
@@ -124,20 +124,20 @@
             hold += (unsigned long)(PUP(in)) << bits;
             bits += 8;
         }
-        this = lcode[hold & lmask];
+        here = lcode[hold & lmask];
       dolen:
-        op = (unsigned)(this.bits);
+        op = (unsigned)(here.bits);
         hold >>= op;
         bits -= op;
-        op = (unsigned)(this.op);
+        op = (unsigned)(here.op);
         if (op == 0) {                          /* literal */
-            Tracevv((stderr, this.val >= 0x20 && this.val < 0x7f ?
+            Tracevv((stderr, here.val >= 0x20 && here.val < 0x7f ?
                     "inflate:         literal '%c'\n" :
-                    "inflate:         literal 0x%02x\n", this.val));
-            PUP(out) = (unsigned char)(this.val);
+                    "inflate:         literal 0x%02x\n", here.val));
+            PUP(out) = (unsigned char)(here.val);
         }
         else if (op & 16) {                     /* length base */
-            len = (unsigned)(this.val);
+            len = (unsigned)(here.val);
             op &= 15;                           /* number of extra bits */
             if (op) {
                 if (bits < op) {
@@ -155,14 +155,14 @@
                 hold += (unsigned long)(PUP(in)) << bits;
                 bits += 8;
             }
-            this = dcode[hold & dmask];
+            here = dcode[hold & dmask];
           dodist:
-            op = (unsigned)(this.bits);
+            op = (unsigned)(here.bits);
             hold >>= op;
             bits -= op;
-            op = (unsigned)(this.op);
+            op = (unsigned)(here.op);
             if (op & 16) {                      /* distance base */
-                dist = (unsigned)(this.val);
+                dist = (unsigned)(here.val);
                 op &= 15;                       /* number of extra bits */
                 if (bits < op) {
                     hold += (unsigned long)(PUP(in)) << bits;
@@ -187,12 +187,34 @@
                 if (dist > op) {                /* see if copy from window */
                     op = dist - op;             /* distance back in window */
                     if (op > whave) {
-                        strm->msg = (char *)"invalid distance too far back";
-                        state->mode = BAD;
-                        break;
+                        if (state->sane) {
+                            strm->msg =
+                                (char *)"invalid distance too far back";
+                            state->mode = BAD;
+                            break;
+                        }
+#ifdef INFLATE_ALLOW_INVALID_DISTANCE_TOOFAR_ARRR
+                        if (len <= op - whave) {
+                            do {
+                                PUP(out) = 0;
+                            } while (--len);
+                            continue;
+                        }
+                        len -= op - whave;
+                        do {
+                            PUP(out) = 0;
+                        } while (--op > whave);
+                        if (op == 0) {
+                            from = out - dist;
+                            do {
+                                PUP(out) = PUP(from);
+                            } while (--len);
+                            continue;
+                        }
+#endif
                     }
                     from = window - OFF;
-                    if (write == 0) {           /* very common case */
+                    if (wnext == 0) {           /* very common case */
                         from += wsize - op;
                         if (op < len) {         /* some from window */
                             len -= op;
@@ -202,17 +224,17 @@
                             from = out - dist;  /* rest from output */
                         }
                     }
-                    else if (write < op) {      /* wrap around window */
-                        from += wsize + write - op;
-                        op -= write;
+                    else if (wnext < op) {      /* wrap around window */
+                        from += wsize + wnext - op;
+                        op -= wnext;
                         if (op < len) {         /* some from end of window */
                             len -= op;
                             do {
                                 PUP(out) = PUP(from);
                             } while (--op);
                             from = window - OFF;
-                            if (write < len) {  /* some from start of window */
-                                op = write;
+                            if (wnext < len) {  /* some from start of window */
+                                op = wnext;
                                 len -= op;
                                 do {
                                     PUP(out) = PUP(from);
@@ -222,7 +244,7 @@
                         }
                     }
                     else {                      /* contiguous in window */
-                        from += write - op;
+                        from += wnext - op;
                         if (op < len) {         /* some from window */
                             len -= op;
                             do {
@@ -259,7 +281,7 @@
                 }
             }
             else if ((op & 64) == 0) {          /* 2nd level distance code */
-                this = dcode[this.val + (hold & ((1U << op) - 1))];
+                here = dcode[here.val + (hold & ((1U << op) - 1))];
                 goto dodist;
             }
             else {
@@ -269,7 +291,7 @@
             }
         }
         else if ((op & 64) == 0) {              /* 2nd level length code */
-            this = lcode[this.val + (hold & ((1U << op) - 1))];
+            here = lcode[here.val + (hold & ((1U << op) - 1))];
             goto dolen;
         }
         else if (op & 32) {                     /* end-of-block */
@@ -305,7 +327,7 @@
    inflate_fast() speedups that turned out slower (on a PowerPC G3 750CXe):
    - Using bit fields for code structure
    - Different op definition to avoid & for extra bits (do & for table bits)
-   - Three separate decoding do-loops for direct, window, and write == 0
+   - Three separate decoding do-loops for direct, window, and wnext == 0
    - Special case for distance > 1 copies to do overlapped load and store copy
    - Explicit branch predictions (based on measured branch probabilities)
    - Deferring match copy and interspersed it with decoding subsequent codes

Modified: python/branches/import_unicode/Modules/zlib/inffast.h
==============================================================================
--- python/branches/import_unicode/Modules/zlib/inffast.h	(original)
+++ python/branches/import_unicode/Modules/zlib/inffast.h	Sat Jul 31 13:01:07 2010
@@ -1,5 +1,5 @@
 /* inffast.h -- header to use inffast.c
- * Copyright (C) 1995-2003 Mark Adler
+ * Copyright (C) 1995-2003, 2010 Mark Adler
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -8,4 +8,4 @@
    subject to change. Applications should only use zlib.h.
  */
 
-void inflate_fast OF((z_streamp strm, unsigned start));
+void ZLIB_INTERNAL inflate_fast OF((z_streamp strm, unsigned start));

Modified: python/branches/import_unicode/Modules/zlib/inflate.c
==============================================================================
--- python/branches/import_unicode/Modules/zlib/inflate.c	(original)
+++ python/branches/import_unicode/Modules/zlib/inflate.c	Sat Jul 31 13:01:07 2010
@@ -1,5 +1,5 @@
 /* inflate.c -- zlib decompression
- * Copyright (C) 1995-2005 Mark Adler
+ * Copyright (C) 1995-2010 Mark Adler
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -45,7 +45,7 @@
  * - Rearrange window copies in inflate_fast() for speed and simplification
  * - Unroll last copy for window match in inflate_fast()
  * - Use local copies of window variables in inflate_fast() for speed
- * - Pull out common write == 0 case for speed in inflate_fast()
+ * - Pull out common wnext == 0 case for speed in inflate_fast()
  * - Make op and len in inflate_fast() unsigned for consistency
  * - Add FAR to lcode and dcode declarations in inflate_fast()
  * - Simplified bad distance check in inflate_fast()
@@ -117,28 +117,52 @@
     state->head = Z_NULL;
     state->wsize = 0;
     state->whave = 0;
-    state->write = 0;
+    state->wnext = 0;
     state->hold = 0;
     state->bits = 0;
     state->lencode = state->distcode = state->next = state->codes;
+    state->sane = 1;
+    state->back = -1;
     Tracev((stderr, "inflate: reset\n"));
     return Z_OK;
 }
 
-int ZEXPORT inflatePrime(strm, bits, value)
+int ZEXPORT inflateReset2(strm, windowBits)
 z_streamp strm;
-int bits;
-int value;
+int windowBits;
 {
+    int wrap;
     struct inflate_state FAR *state;
 
+    /* get the state */
     if (strm == Z_NULL || strm->state == Z_NULL) return Z_STREAM_ERROR;
     state = (struct inflate_state FAR *)strm->state;
-    if (bits > 16 || state->bits + bits > 32) return Z_STREAM_ERROR;
-    value &= (1L << bits) - 1;
-    state->hold += value << state->bits;
-    state->bits += bits;
-    return Z_OK;
+
+    /* extract wrap request from windowBits parameter */
+    if (windowBits < 0) {
+        wrap = 0;
+        windowBits = -windowBits;
+    }
+    else {
+        wrap = (windowBits >> 4) + 1;
+#ifdef GUNZIP
+        if (windowBits < 48)
+            windowBits &= 15;
+#endif
+    }
+
+    /* set number of window bits, free window if different */
+    if (windowBits && (windowBits < 8 || windowBits > 15))
+        return Z_STREAM_ERROR;
+    if (state->window != Z_NULL && state->wbits != (unsigned)windowBits) {
+        ZFREE(strm, state->window);
+        state->window = Z_NULL;
+    }
+
+    /* update state and reset the rest of it */
+    state->wrap = wrap;
+    state->wbits = (unsigned)windowBits;
+    return inflateReset(strm);
 }
 
 int ZEXPORT inflateInit2_(strm, windowBits, version, stream_size)
@@ -147,6 +171,7 @@
 const char *version;
 int stream_size;
 {
+    int ret;
     struct inflate_state FAR *state;
 
     if (version == Z_NULL || version[0] != ZLIB_VERSION[0] ||
@@ -164,24 +189,13 @@
     if (state == Z_NULL) return Z_MEM_ERROR;
     Tracev((stderr, "inflate: allocated\n"));
     strm->state = (struct internal_state FAR *)state;
-    if (windowBits < 0) {
-        state->wrap = 0;
-        windowBits = -windowBits;
-    }
-    else {
-        state->wrap = (windowBits >> 4) + 1;
-#ifdef GUNZIP
-        if (windowBits < 48) windowBits &= 15;
-#endif
-    }
-    if (windowBits < 8 || windowBits > 15) {
+    state->window = Z_NULL;
+    ret = inflateReset2(strm, windowBits);
+    if (ret != Z_OK) {
         ZFREE(strm, state);
         strm->state = Z_NULL;
-        return Z_STREAM_ERROR;
     }
-    state->wbits = (unsigned)windowBits;
-    state->window = Z_NULL;
-    return inflateReset(strm);
+    return ret;
 }
 
 int ZEXPORT inflateInit_(strm, version, stream_size)
@@ -192,6 +206,27 @@
     return inflateInit2_(strm, DEF_WBITS, version, stream_size);
 }
 
+int ZEXPORT inflatePrime(strm, bits, value)
+z_streamp strm;
+int bits;
+int value;
+{
+    struct inflate_state FAR *state;
+
+    if (strm == Z_NULL || strm->state == Z_NULL) return Z_STREAM_ERROR;
+    state = (struct inflate_state FAR *)strm->state;
+    if (bits < 0) {
+        state->hold = 0;
+        state->bits = 0;
+        return Z_OK;
+    }
+    if (bits > 16 || state->bits + bits > 32) return Z_STREAM_ERROR;
+    value &= (1L << bits) - 1;
+    state->hold += value << state->bits;
+    state->bits += bits;
+    return Z_OK;
+}
+
 /*
    Return state with length and distance decoding tables and index sizes set to
    fixed code decoding.  Normally this returns fixed tables from inffixed.h.
@@ -340,7 +375,7 @@
     /* if window not in use yet, initialize */
     if (state->wsize == 0) {
         state->wsize = 1U << state->wbits;
-        state->write = 0;
+        state->wnext = 0;
         state->whave = 0;
     }
 
@@ -348,22 +383,22 @@
     copy = out - strm->avail_out;
     if (copy >= state->wsize) {
         zmemcpy(state->window, strm->next_out - state->wsize, state->wsize);
-        state->write = 0;
+        state->wnext = 0;
         state->whave = state->wsize;
     }
     else {
-        dist = state->wsize - state->write;
+        dist = state->wsize - state->wnext;
         if (dist > copy) dist = copy;
-        zmemcpy(state->window + state->write, strm->next_out - copy, dist);
+        zmemcpy(state->window + state->wnext, strm->next_out - copy, dist);
         copy -= dist;
         if (copy) {
             zmemcpy(state->window, strm->next_out - copy, copy);
-            state->write = copy;
+            state->wnext = copy;
             state->whave = state->wsize;
         }
         else {
-            state->write += dist;
-            if (state->write == state->wsize) state->write = 0;
+            state->wnext += dist;
+            if (state->wnext == state->wsize) state->wnext = 0;
             if (state->whave < state->wsize) state->whave += dist;
         }
     }
@@ -564,7 +599,7 @@
     unsigned in, out;           /* save starting available input and output */
     unsigned copy;              /* number of stored or match bytes to copy */
     unsigned char FAR *from;    /* where to copy match bytes from */
-    code this;                  /* current decoding table entry */
+    code here;                  /* current decoding table entry */
     code last;                  /* parent table entry */
     unsigned len;               /* length to copy for repeats, bits to drop */
     int ret;                    /* return code */
@@ -619,7 +654,9 @@
             }
             DROPBITS(4);
             len = BITS(4) + 8;
-            if (len > state->wbits) {
+            if (state->wbits == 0)
+                state->wbits = len;
+            else if (len > state->wbits) {
                 strm->msg = (char *)"invalid window size";
                 state->mode = BAD;
                 break;
@@ -771,7 +808,7 @@
             strm->adler = state->check = adler32(0L, Z_NULL, 0);
             state->mode = TYPE;
         case TYPE:
-            if (flush == Z_BLOCK) goto inf_leave;
+            if (flush == Z_BLOCK || flush == Z_TREES) goto inf_leave;
         case TYPEDO:
             if (state->last) {
                 BYTEBITS();
@@ -791,7 +828,11 @@
                 fixedtables(state);
                 Tracev((stderr, "inflate:     fixed codes block%s\n",
                         state->last ? " (last)" : ""));
-                state->mode = LEN;              /* decode codes */
+                state->mode = LEN_;             /* decode codes */
+                if (flush == Z_TREES) {
+                    DROPBITS(2);
+                    goto inf_leave;
+                }
                 break;
             case 2:                             /* dynamic block */
                 Tracev((stderr, "inflate:     dynamic codes block%s\n",
@@ -816,6 +857,9 @@
             Tracev((stderr, "inflate:       stored length %u\n",
                     state->length));
             INITBITS();
+            state->mode = COPY_;
+            if (flush == Z_TREES) goto inf_leave;
+        case COPY_:
             state->mode = COPY;
         case COPY:
             copy = state->length;
@@ -876,19 +920,19 @@
         case CODELENS:
             while (state->have < state->nlen + state->ndist) {
                 for (;;) {
-                    this = state->lencode[BITS(state->lenbits)];
-                    if ((unsigned)(this.bits) <= bits) break;
+                    here = state->lencode[BITS(state->lenbits)];
+                    if ((unsigned)(here.bits) <= bits) break;
                     PULLBYTE();
                 }
-                if (this.val < 16) {
-                    NEEDBITS(this.bits);
-                    DROPBITS(this.bits);
-                    state->lens[state->have++] = this.val;
+                if (here.val < 16) {
+                    NEEDBITS(here.bits);
+                    DROPBITS(here.bits);
+                    state->lens[state->have++] = here.val;
                 }
                 else {
-                    if (this.val == 16) {
-                        NEEDBITS(this.bits + 2);
-                        DROPBITS(this.bits);
+                    if (here.val == 16) {
+                        NEEDBITS(here.bits + 2);
+                        DROPBITS(here.bits);
                         if (state->have == 0) {
                             strm->msg = (char *)"invalid bit length repeat";
                             state->mode = BAD;
@@ -898,16 +942,16 @@
                         copy = 3 + BITS(2);
                         DROPBITS(2);
                     }
-                    else if (this.val == 17) {
-                        NEEDBITS(this.bits + 3);
-                        DROPBITS(this.bits);
+                    else if (here.val == 17) {
+                        NEEDBITS(here.bits + 3);
+                        DROPBITS(here.bits);
                         len = 0;
                         copy = 3 + BITS(3);
                         DROPBITS(3);
                     }
                     else {
-                        NEEDBITS(this.bits + 7);
-                        DROPBITS(this.bits);
+                        NEEDBITS(here.bits + 7);
+                        DROPBITS(here.bits);
                         len = 0;
                         copy = 11 + BITS(7);
                         DROPBITS(7);
@@ -925,7 +969,16 @@
             /* handle error breaks in while */
             if (state->mode == BAD) break;
 
-            /* build code tables */
+            /* check for end-of-block code (better have one) */
+            if (state->lens[256] == 0) {
+                strm->msg = (char *)"invalid code -- missing end-of-block";
+                state->mode = BAD;
+                break;
+            }
+
+            /* build code tables -- note: do not change the lenbits or distbits
+               values here (9 and 6) without reading the comments in inftrees.h
+               concerning the ENOUGH constants, which depend on those values */
             state->next = state->codes;
             state->lencode = (code const FAR *)(state->next);
             state->lenbits = 9;
@@ -946,88 +999,102 @@
                 break;
             }
             Tracev((stderr, "inflate:       codes ok\n"));
+            state->mode = LEN_;
+            if (flush == Z_TREES) goto inf_leave;
+        case LEN_:
             state->mode = LEN;
         case LEN:
             if (have >= 6 && left >= 258) {
                 RESTORE();
                 inflate_fast(strm, out);
                 LOAD();
+                if (state->mode == TYPE)
+                    state->back = -1;
                 break;
             }
+            state->back = 0;
             for (;;) {
-                this = state->lencode[BITS(state->lenbits)];
-                if ((unsigned)(this.bits) <= bits) break;
+                here = state->lencode[BITS(state->lenbits)];
+                if ((unsigned)(here.bits) <= bits) break;
                 PULLBYTE();
             }
-            if (this.op && (this.op & 0xf0) == 0) {
-                last = this;
+            if (here.op && (here.op & 0xf0) == 0) {
+                last = here;
                 for (;;) {
-                    this = state->lencode[last.val +
+                    here = state->lencode[last.val +
                             (BITS(last.bits + last.op) >> last.bits)];
-                    if ((unsigned)(last.bits + this.bits) <= bits) break;
+                    if ((unsigned)(last.bits + here.bits) <= bits) break;
                     PULLBYTE();
                 }
                 DROPBITS(last.bits);
+                state->back += last.bits;
             }
-            DROPBITS(this.bits);
-            state->length = (unsigned)this.val;
-            if ((int)(this.op) == 0) {
-                Tracevv((stderr, this.val >= 0x20 && this.val < 0x7f ?
+            DROPBITS(here.bits);
+            state->back += here.bits;
+            state->length = (unsigned)here.val;
+            if ((int)(here.op) == 0) {
+                Tracevv((stderr, here.val >= 0x20 && here.val < 0x7f ?
                         "inflate:         literal '%c'\n" :
-                        "inflate:         literal 0x%02x\n", this.val));
+                        "inflate:         literal 0x%02x\n", here.val));
                 state->mode = LIT;
                 break;
             }
-            if (this.op & 32) {
+            if (here.op & 32) {
                 Tracevv((stderr, "inflate:         end of block\n"));
+                state->back = -1;
                 state->mode = TYPE;
                 break;
             }
-            if (this.op & 64) {
+            if (here.op & 64) {
                 strm->msg = (char *)"invalid literal/length code";
                 state->mode = BAD;
                 break;
             }
-            state->extra = (unsigned)(this.op) & 15;
+            state->extra = (unsigned)(here.op) & 15;
             state->mode = LENEXT;
         case LENEXT:
             if (state->extra) {
                 NEEDBITS(state->extra);
                 state->length += BITS(state->extra);
                 DROPBITS(state->extra);
+                state->back += state->extra;
             }
             Tracevv((stderr, "inflate:         length %u\n", state->length));
+            state->was = state->length;
             state->mode = DIST;
         case DIST:
             for (;;) {
-                this = state->distcode[BITS(state->distbits)];
-                if ((unsigned)(this.bits) <= bits) break;
+                here = state->distcode[BITS(state->distbits)];
+                if ((unsigned)(here.bits) <= bits) break;
                 PULLBYTE();
             }
-            if ((this.op & 0xf0) == 0) {
-                last = this;
+            if ((here.op & 0xf0) == 0) {
+                last = here;
                 for (;;) {
-                    this = state->distcode[last.val +
+                    here = state->distcode[last.val +
                             (BITS(last.bits + last.op) >> last.bits)];
-                    if ((unsigned)(last.bits + this.bits) <= bits) break;
+                    if ((unsigned)(last.bits + here.bits) <= bits) break;
                     PULLBYTE();
                 }
                 DROPBITS(last.bits);
+                state->back += last.bits;
             }
-            DROPBITS(this.bits);
-            if (this.op & 64) {
+            DROPBITS(here.bits);
+            state->back += here.bits;
+            if (here.op & 64) {
                 strm->msg = (char *)"invalid distance code";
                 state->mode = BAD;
                 break;
             }
-            state->offset = (unsigned)this.val;
-            state->extra = (unsigned)(this.op) & 15;
+            state->offset = (unsigned)here.val;
+            state->extra = (unsigned)(here.op) & 15;
             state->mode = DISTEXT;
         case DISTEXT:
             if (state->extra) {
                 NEEDBITS(state->extra);
                 state->offset += BITS(state->extra);
                 DROPBITS(state->extra);
+                state->back += state->extra;
             }
 #ifdef INFLATE_STRICT
             if (state->offset > state->dmax) {
@@ -1036,11 +1103,6 @@
                 break;
             }
 #endif
-            if (state->offset > state->whave + out - left) {
-                strm->msg = (char *)"invalid distance too far back";
-                state->mode = BAD;
-                break;
-            }
             Tracevv((stderr, "inflate:         distance %u\n", state->offset));
             state->mode = MATCH;
         case MATCH:
@@ -1048,12 +1110,32 @@
             copy = out - left;
             if (state->offset > copy) {         /* copy from window */
                 copy = state->offset - copy;
-                if (copy > state->write) {
-                    copy -= state->write;
+                if (copy > state->whave) {
+                    if (state->sane) {
+                        strm->msg = (char *)"invalid distance too far back";
+                        state->mode = BAD;
+                        break;
+                    }
+#ifdef INFLATE_ALLOW_INVALID_DISTANCE_TOOFAR_ARRR
+                    Trace((stderr, "inflate.c too far\n"));
+                    copy -= state->whave;
+                    if (copy > state->length) copy = state->length;
+                    if (copy > left) copy = left;
+                    left -= copy;
+                    state->length -= copy;
+                    do {
+                        *put++ = 0;
+                    } while (--copy);
+                    if (state->length == 0) state->mode = LEN;
+                    break;
+#endif
+                }
+                if (copy > state->wnext) {
+                    copy -= state->wnext;
                     from = state->window + (state->wsize - copy);
                 }
                 else
-                    from = state->window + (state->write - copy);
+                    from = state->window + (state->wnext - copy);
                 if (copy > state->length) copy = state->length;
             }
             else {                              /* copy from output */
@@ -1146,7 +1228,8 @@
         strm->adler = state->check =
             UPDATE(state->check, strm->next_out - out, out);
     strm->data_type = state->bits + (state->last ? 64 : 0) +
-                      (state->mode == TYPE ? 128 : 0);
+                      (state->mode == TYPE ? 128 : 0) +
+                      (state->mode == LEN_ || state->mode == COPY_ ? 256 : 0);
     if (((in == 0 && out == 0) || flush == Z_FINISH) && ret == Z_OK)
         ret = Z_BUF_ERROR;
     return ret;
@@ -1366,3 +1449,32 @@
     dest->state = (struct internal_state FAR *)copy;
     return Z_OK;
 }
+
+int ZEXPORT inflateUndermine(strm, subvert)
+z_streamp strm;
+int subvert;
+{
+    struct inflate_state FAR *state;
+
+    if (strm == Z_NULL || strm->state == Z_NULL) return Z_STREAM_ERROR;
+    state = (struct inflate_state FAR *)strm->state;
+    state->sane = !subvert;
+#ifdef INFLATE_ALLOW_INVALID_DISTANCE_TOOFAR_ARRR
+    return Z_OK;
+#else
+    state->sane = 1;
+    return Z_DATA_ERROR;
+#endif
+}
+
+long ZEXPORT inflateMark(strm)
+z_streamp strm;
+{
+    struct inflate_state FAR *state;
+
+    if (strm == Z_NULL || strm->state == Z_NULL) return -1L << 16;
+    state = (struct inflate_state FAR *)strm->state;
+    return ((long)(state->back) << 16) +
+        (state->mode == COPY ? state->length :
+            (state->mode == MATCH ? state->was - state->length : 0));
+}

Modified: python/branches/import_unicode/Modules/zlib/inflate.h
==============================================================================
--- python/branches/import_unicode/Modules/zlib/inflate.h	(original)
+++ python/branches/import_unicode/Modules/zlib/inflate.h	Sat Jul 31 13:01:07 2010
@@ -1,5 +1,5 @@
 /* inflate.h -- internal inflate state definition
- * Copyright (C) 1995-2004 Mark Adler
+ * Copyright (C) 1995-2009 Mark Adler
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -32,11 +32,13 @@
         TYPE,       /* i: waiting for type bits, including last-flag bit */
         TYPEDO,     /* i: same, but skip check to exit inflate on new block */
         STORED,     /* i: waiting for stored size (length and complement) */
+        COPY_,      /* i/o: same as COPY below, but only first time in */
         COPY,       /* i/o: waiting for input or output to copy stored block */
         TABLE,      /* i: waiting for dynamic block table lengths */
         LENLENS,    /* i: waiting for code length code lengths */
         CODELENS,   /* i: waiting for length/lit and distance code lengths */
-            LEN,        /* i: waiting for length/lit code */
+            LEN_,       /* i: same as LEN below, but only first time in */
+            LEN,        /* i: waiting for length/lit/eob code */
             LENEXT,     /* i: waiting for length extra bits */
             DIST,       /* i: waiting for distance code */
             DISTEXT,    /* i: waiting for distance extra bits */
@@ -53,19 +55,21 @@
 /*
     State transitions between above modes -
 
-    (most modes can go to the BAD or MEM mode -- not shown for clarity)
+    (most modes can go to BAD or MEM on error -- not shown for clarity)
 
     Process header:
-        HEAD -> (gzip) or (zlib)
-        (gzip) -> FLAGS -> TIME -> OS -> EXLEN -> EXTRA -> NAME
-        NAME -> COMMENT -> HCRC -> TYPE
+        HEAD -> (gzip) or (zlib) or (raw)
+        (gzip) -> FLAGS -> TIME -> OS -> EXLEN -> EXTRA -> NAME -> COMMENT ->
+                  HCRC -> TYPE
         (zlib) -> DICTID or TYPE
         DICTID -> DICT -> TYPE
+        (raw) -> TYPEDO
     Read deflate blocks:
-            TYPE -> STORED or TABLE or LEN or CHECK
-            STORED -> COPY -> TYPE
-            TABLE -> LENLENS -> CODELENS -> LEN
-    Read deflate codes:
+            TYPE -> TYPEDO -> STORED or TABLE or LEN_ or CHECK
+            STORED -> COPY_ -> COPY -> TYPE
+            TABLE -> LENLENS -> CODELENS -> LEN_
+            LEN_ -> LEN
+    Read deflate codes in fixed or dynamic block:
                 LEN -> LENEXT or LIT or TYPE
                 LENEXT -> DIST -> DISTEXT -> MATCH -> LEN
                 LIT -> LEN
@@ -73,7 +77,7 @@
         CHECK -> LENGTH -> DONE
  */
 
-/* state maintained between inflate() calls.  Approximately 7K bytes. */
+/* state maintained between inflate() calls.  Approximately 10K bytes. */
 struct inflate_state {
     inflate_mode mode;          /* current inflate mode */
     int last;                   /* true if processing last block */
@@ -88,7 +92,7 @@
     unsigned wbits;             /* log base 2 of requested window size */
     unsigned wsize;             /* window size or zero if not using window */
     unsigned whave;             /* valid bytes in the window */
-    unsigned write;             /* window write index */
+    unsigned wnext;             /* window write index */
     unsigned char FAR *window;  /* allocated sliding window, if needed */
         /* bit accumulator */
     unsigned long hold;         /* input bit accumulator */
@@ -112,4 +116,7 @@
     unsigned short lens[320];   /* temporary storage for code lengths */
     unsigned short work[288];   /* work area for code table building */
     code codes[ENOUGH];         /* space for code tables */
+    int sane;                   /* if false, allow invalid distance too far */
+    int back;                   /* bits back of last unprocessed length/lit */
+    unsigned was;               /* initial length of match */
 };

Modified: python/branches/import_unicode/Modules/zlib/inftrees.c
==============================================================================
--- python/branches/import_unicode/Modules/zlib/inftrees.c	(original)
+++ python/branches/import_unicode/Modules/zlib/inftrees.c	Sat Jul 31 13:01:07 2010
@@ -1,5 +1,5 @@
 /* inftrees.c -- generate Huffman trees for efficient decoding
- * Copyright (C) 1995-2005 Mark Adler
+ * Copyright (C) 1995-2010 Mark Adler
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -9,7 +9,7 @@
 #define MAXBITS 15
 
 const char inflate_copyright[] =
-   " inflate 1.2.3 Copyright 1995-2005 Mark Adler ";
+   " inflate 1.2.5 Copyright 1995-2010 Mark Adler ";
 /*
   If you use the zlib library in a product, an acknowledgment is welcome
   in the documentation of your product. If for some reason you cannot
@@ -29,7 +29,7 @@
    table index bits.  It will differ if the request is greater than the
    longest code or if it is less than the shortest code.
  */
-int inflate_table(type, lens, codes, table, bits, work)
+int ZLIB_INTERNAL inflate_table(type, lens, codes, table, bits, work)
 codetype type;
 unsigned short FAR *lens;
 unsigned codes;
@@ -50,7 +50,7 @@
     unsigned fill;              /* index for replicating entries */
     unsigned low;               /* low bits for current root entry */
     unsigned mask;              /* mask for low root bits */
-    code this;                  /* table entry for duplication */
+    code here;                  /* table entry for duplication */
     code FAR *next;             /* next available space in table */
     const unsigned short FAR *base;     /* base value table to use */
     const unsigned short FAR *extra;    /* extra bits table to use */
@@ -62,7 +62,7 @@
         35, 43, 51, 59, 67, 83, 99, 115, 131, 163, 195, 227, 258, 0, 0};
     static const unsigned short lext[31] = { /* Length codes 257..285 extra */
         16, 16, 16, 16, 16, 16, 16, 16, 17, 17, 17, 17, 18, 18, 18, 18,
-        19, 19, 19, 19, 20, 20, 20, 20, 21, 21, 21, 21, 16, 201, 196};
+        19, 19, 19, 19, 20, 20, 20, 20, 21, 21, 21, 21, 16, 73, 195};
     static const unsigned short dbase[32] = { /* Distance codes 0..29 base */
         1, 2, 3, 4, 5, 7, 9, 13, 17, 25, 33, 49, 65, 97, 129, 193,
         257, 385, 513, 769, 1025, 1537, 2049, 3073, 4097, 6145,
@@ -115,15 +115,15 @@
         if (count[max] != 0) break;
     if (root > max) root = max;
     if (max == 0) {                     /* no symbols to code at all */
-        this.op = (unsigned char)64;    /* invalid code marker */
-        this.bits = (unsigned char)1;
-        this.val = (unsigned short)0;
-        *(*table)++ = this;             /* make a table to force an error */
-        *(*table)++ = this;
+        here.op = (unsigned char)64;    /* invalid code marker */
+        here.bits = (unsigned char)1;
+        here.val = (unsigned short)0;
+        *(*table)++ = here;             /* make a table to force an error */
+        *(*table)++ = here;
         *bits = 1;
         return 0;     /* no symbols, but wait for decoding to report error */
     }
-    for (min = 1; min <= MAXBITS; min++)
+    for (min = 1; min < max; min++)
         if (count[min] != 0) break;
     if (root < min) root = min;
 
@@ -166,11 +166,10 @@
        entered in the tables.
 
        used keeps track of how many table entries have been allocated from the
-       provided *table space.  It is checked when a LENS table is being made
-       against the space in *table, ENOUGH, minus the maximum space needed by
-       the worst case distance code, MAXD.  This should never happen, but the
-       sufficiency of ENOUGH has not been proven exhaustively, hence the check.
-       This assumes that when type == LENS, bits == 9.
+       provided *table space.  It is checked for LENS and DIST tables against
+       the constants ENOUGH_LENS and ENOUGH_DISTS to guard against changes in
+       the initial root table size constants.  See the comments in inftrees.h
+       for more information.
 
        sym increments through all symbols, and the loop terminates when
        all codes of length max, i.e. all codes, have been processed.  This
@@ -209,24 +208,25 @@
     mask = used - 1;            /* mask for comparing low */
 
     /* check available table space */
-    if (type == LENS && used >= ENOUGH - MAXD)
+    if ((type == LENS && used >= ENOUGH_LENS) ||
+        (type == DISTS && used >= ENOUGH_DISTS))
         return 1;
 
     /* process all codes and make table entries */
     for (;;) {
         /* create table entry */
-        this.bits = (unsigned char)(len - drop);
+        here.bits = (unsigned char)(len - drop);
         if ((int)(work[sym]) < end) {
-            this.op = (unsigned char)0;
-            this.val = work[sym];
+            here.op = (unsigned char)0;
+            here.val = work[sym];
         }
         else if ((int)(work[sym]) > end) {
-            this.op = (unsigned char)(extra[work[sym]]);
-            this.val = base[work[sym]];
+            here.op = (unsigned char)(extra[work[sym]]);
+            here.val = base[work[sym]];
         }
         else {
-            this.op = (unsigned char)(32 + 64);         /* end of block */
-            this.val = 0;
+            here.op = (unsigned char)(32 + 64);         /* end of block */
+            here.val = 0;
         }
 
         /* replicate for those indices with low len bits equal to huff */
@@ -235,7 +235,7 @@
         min = fill;                 /* save offset to next table */
         do {
             fill -= incr;
-            next[(huff >> drop) + fill] = this;
+            next[(huff >> drop) + fill] = here;
         } while (fill != 0);
 
         /* backwards increment the len-bit code huff */
@@ -277,7 +277,8 @@
 
             /* check for enough space */
             used += 1U << curr;
-            if (type == LENS && used >= ENOUGH - MAXD)
+            if ((type == LENS && used >= ENOUGH_LENS) ||
+                (type == DISTS && used >= ENOUGH_DISTS))
                 return 1;
 
             /* point entry in root table to sub-table */
@@ -295,20 +296,20 @@
        through high index bits.  When the current sub-table is filled, the loop
        drops back to the root table to fill in any remaining entries there.
      */
-    this.op = (unsigned char)64;                /* invalid code marker */
-    this.bits = (unsigned char)(len - drop);
-    this.val = (unsigned short)0;
+    here.op = (unsigned char)64;                /* invalid code marker */
+    here.bits = (unsigned char)(len - drop);
+    here.val = (unsigned short)0;
     while (huff != 0) {
         /* when done with sub-table, drop back to root table */
         if (drop != 0 && (huff & mask) != low) {
             drop = 0;
             len = root;
             next = *table;
-            this.bits = (unsigned char)len;
+            here.bits = (unsigned char)len;
         }
 
         /* put invalid code marker in table */
-        next[huff >> drop] = this;
+        next[huff >> drop] = here;
 
         /* backwards increment the len-bit code huff */
         incr = 1U << (len - 1);

Modified: python/branches/import_unicode/Modules/zlib/inftrees.h
==============================================================================
--- python/branches/import_unicode/Modules/zlib/inftrees.h	(original)
+++ python/branches/import_unicode/Modules/zlib/inftrees.h	Sat Jul 31 13:01:07 2010
@@ -1,5 +1,5 @@
 /* inftrees.h -- header to use inftrees.c
- * Copyright (C) 1995-2005 Mark Adler
+ * Copyright (C) 1995-2005, 2010 Mark Adler
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -35,21 +35,28 @@
     01000000 - invalid code
  */
 
-/* Maximum size of dynamic tree.  The maximum found in a long but non-
-   exhaustive search was 1444 code structures (852 for length/literals
-   and 592 for distances, the latter actually the result of an
-   exhaustive search).  The true maximum is not known, but the value
-   below is more than safe. */
-#define ENOUGH 2048
-#define MAXD 592
+/* Maximum size of the dynamic table.  The maximum number of code structures is
+   1444, which is the sum of 852 for literal/length codes and 592 for distance
+   codes.  These values were found by exhaustive searches using the program
+   examples/enough.c found in the zlib distribtution.  The arguments to that
+   program are the number of symbols, the initial root table size, and the
+   maximum bit length of a code.  "enough 286 9 15" for literal/length codes
+   returns returns 852, and "enough 30 6 15" for distance codes returns 592.
+   The initial root table size (9 or 6) is found in the fifth argument of the
+   inflate_table() calls in inflate.c and infback.c.  If the root table size is
+   changed, then these maximum sizes would be need to be recalculated and
+   updated. */
+#define ENOUGH_LENS 852
+#define ENOUGH_DISTS 592
+#define ENOUGH (ENOUGH_LENS+ENOUGH_DISTS)
 
-/* Type of code to build for inftable() */
+/* Type of code to build for inflate_table() */
 typedef enum {
     CODES,
     LENS,
     DISTS
 } codetype;
 
-extern int inflate_table OF((codetype type, unsigned short FAR *lens,
+int ZLIB_INTERNAL inflate_table OF((codetype type, unsigned short FAR *lens,
                              unsigned codes, code FAR * FAR *table,
                              unsigned FAR *bits, unsigned short FAR *work));

Modified: python/branches/import_unicode/Modules/zlib/make_vms.com
==============================================================================
--- python/branches/import_unicode/Modules/zlib/make_vms.com	(original)
+++ python/branches/import_unicode/Modules/zlib/make_vms.com	Sat Jul 31 13:01:07 2010
@@ -1,33 +1,57 @@
 $! make libz under VMS written by
 $! Martin P.J. Zinser
-$! <zinser at zinser.no-ip.info or zinser at sysdev.deutsche-boerse.com>
 $!
-$ on error then goto err_exit
+$! In case of problems with the install you might contact me at
+$! zinser at zinser.no-ip.info(preferred) or
+$! zinser at sysdev.deutsche-boerse.com (work)
+$!
+$! Make procedure history for Zlib
 $!
+$!------------------------------------------------------------------------------
+$! Version history
+$! 0.01 20060120 First version to receive a number
+$! 0.02 20061008 Adapt to new Makefile.in
+$! 0.03 20091224 Add support for large file check
+$! 0.04 20100110 Add new gzclose, gzlib, gzread, gzwrite
+$! 0.05 20100221 Exchange zlibdefs.h by zconf.h.in
 $!
-$! Just some general constants...
+$ on error then goto err_exit
+$ set proc/parse=ext
 $!
 $ true  = 1
 $ false = 0
 $ tmpnam = "temp_" + f$getjpi("","pid")
-$ SAY = "WRITE SYS$OUTPUT"
+$ tt = tmpnam + ".txt"
+$ tc = tmpnam + ".c"
+$ th = tmpnam + ".h"
+$ define/nolog tconfig 'th'
+$ its_decc = false
+$ its_vaxc = false
+$ its_gnuc = false
+$ s_case   = False
 $!
 $! Setup variables holding "config" information
 $!
-$ Make     = ""
+$ Make    = ""
 $ name     = "Zlib"
 $ version  = "?.?.?"
 $ v_string = "ZLIB_VERSION"
 $ v_file   = "zlib.h"
-$ ccopt    = ""
-$ lopts    = ""
+$ ccopt   = ""
+$ lopts   = ""
+$ dnsrl   = ""
+$ aconf_in_file = "zconf.h.in#zconf.h_in"
+$ conf_check_string = ""
 $ linkonly = false
 $ optfile  = name + ".opt"
-$ its_decc = false
-$ its_vaxc = false
-$ its_gnuc = false
-$ axp      = f$getsyi("HW_MODEL").ge.1024
-$ s_case   = false
+$ libdefs  = ""
+$ axp      = f$getsyi("HW_MODEL").ge.1024 .and. f$getsyi("HW_MODEL").lt.4096
+$!
+$ whoami = f$parse(f$enviornment("Procedure"),,,,"NO_CONCEAL")
+$ mydef  = F$parse(whoami,,,"DEVICE")
+$ mydir  = f$parse(whoami,,,"DIRECTORY") - "]["
+$ myproc = f$parse(whoami,,,"Name") + f$parse(whoami,,,"type")
+$!
 $! Check for MMK/MMS
 $!
 $ If F$Search ("Sys$System:MMS.EXE") .nes. "" Then Make = "MMS"
@@ -36,11 +60,16 @@
 $!
 $ gosub find_version
 $!
+$  open/write topt tmp.opt
+$  open/write optf 'optfile'
+$!
 $ gosub check_opts
 $!
 $! Look for the compiler used
 $!
 $ gosub check_compiler
+$ close topt
+$!
 $ if its_decc
 $ then
 $   ccopt = "/prefix=all" + ccopt
@@ -60,6 +89,52 @@
 $    if f$trnlnm("SYS").eqs."" then define sys sys$library:
 $ endif
 $!
+$! Build a fake configure input header
+$!
+$ open/write conf_hin config.hin
+$ write conf_hin "#undef _LARGEFILE64_SOURCE"
+$ close conf_hin
+$!
+$!
+$ i = 0
+$FIND_ACONF:
+$ fname = f$element(i,"#",aconf_in_file)
+$ if fname .eqs. "#" then goto AMISS_ERR
+$ if f$search(fname) .eqs. ""
+$ then
+$   i = i + 1
+$   goto find_aconf
+$ endif
+$ open/read/err=aconf_err aconf_in 'fname'
+$ open/write aconf zconf.h
+$ACONF_LOOP:
+$ read/end_of_file=aconf_exit aconf_in line
+$ work = f$edit(line, "compress,trim")
+$ if f$extract(0,6,work) .nes. "#undef"
+$ then
+$   if f$extract(0,12,work) .nes. "#cmakedefine"
+$   then
+$       write aconf line
+$   endif
+$ else
+$   cdef = f$element(1," ",work)
+$   gosub check_config
+$ endif
+$ goto aconf_loop
+$ACONF_EXIT:
+$ write aconf "#define VMS 1"
+$ write aconf "#include <unistd.h>"
+$ write aconf "#include <unixio.h>"
+$ write aconf "#ifdef _LARGEFILE"
+$ write aconf "#define off64_t __off64_t"
+$ write aconf "#define fopen64 fopen"
+$ write aconf "#define fseeko64 fseeko"
+$ write aconf "#define lseek64 lseek"
+$ write aconf "#define ftello64 ftell"
+$ write aconf "#endif"
+$ close aconf_in
+$ close aconf
+$ if f$search("''th'") .nes. "" then delete 'th';*
 $! Build the thing plain or with mms
 $!
 $ write sys$output "Compiling Zlib sources ..."
@@ -74,8 +149,14 @@
                 crc32.c zlib.h zconf.h
 $   CALL MAKE deflate.OBJ "CC ''CCOPT' deflate" -
                 deflate.c deflate.h zutil.h zlib.h zconf.h
-$   CALL MAKE gzio.OBJ "CC ''CCOPT' gzio" -
-                gzio.c zutil.h zlib.h zconf.h
+$   CALL MAKE gzclose.OBJ "CC ''CCOPT' gzclose" -
+                gzclose.c zutil.h zlib.h zconf.h
+$   CALL MAKE gzlib.OBJ "CC ''CCOPT' gzlib" -
+                gzlib.c zutil.h zlib.h zconf.h
+$   CALL MAKE gzread.OBJ "CC ''CCOPT' gzread" -
+                gzread.c zutil.h zlib.h zconf.h
+$   CALL MAKE gzwrite.OBJ "CC ''CCOPT' gzwrite" -
+                gzwrite.c zutil.h zlib.h zconf.h
 $   CALL MAKE infback.OBJ "CC ''CCOPT' infback" -
                 infback.c zutil.h inftrees.h inflate.h inffast.h inffixed.h
 $   CALL MAKE inffast.OBJ "CC ''CCOPT' inffast" -
@@ -107,7 +188,7 @@
 $   endif
 $  else
 $   gosub crea_mms
-$   SAY "Make ''name' ''version' with ''Make' "
+$   write sys$output "Make ''name' ''version' with ''Make' "
 $   'make'
 $  endif
 $!
@@ -133,6 +214,15 @@
 $ goto err_exit
 $ERR_EXIT:
 $ set message/facil/ident/sever/text
+$ close/nolog optf
+$ close/nolog topt
+$ close/nolog conf_hin
+$ close/nolog aconf_in
+$ close/nolog aconf
+$ close/nolog out
+$ close/nolog min
+$ close/nolog mod
+$ close/nolog h_in
 $ write sys$output "Exiting..."
 $ exit 2
 $!
@@ -180,61 +270,72 @@
 $!
 $! Check command line options and set symbols accordingly
 $!
+$!------------------------------------------------------------------------------
+$! Version history
+$! 0.01 20041206 First version to receive a number
+$! 0.02 20060126 Add new "HELP" target
 $ CHECK_OPTS:
 $ i = 1
 $ OPT_LOOP:
 $ if i .lt. 9
 $ then
 $   cparm = f$edit(p'i',"upcase")
-$   if cparm .eqs. "DEBUG"
-$   then
-$     ccopt = ccopt + "/noopt/deb"
-$     lopts = lopts + "/deb"
-$   endif
-$   if f$locate("CCOPT=",cparm) .lt. f$length(cparm)
-$   then
-$     start = f$locate("=",cparm) + 1
-$     len   = f$length(cparm) - start
-$     ccopt = ccopt + f$extract(start,len,cparm)
-$     if f$locate("AS_IS",f$edit(ccopt,"UPCASE")) .lt. f$length(ccopt) -
-         then s_case = true
-$   endif
-$   if cparm .eqs. "LINK" then linkonly = true
-$   if f$locate("LOPTS=",cparm) .lt. f$length(cparm)
-$   then
-$     start = f$locate("=",cparm) + 1
-$     len   = f$length(cparm) - start
-$     lopts = lopts + f$extract(start,len,cparm)
-$   endif
-$   if f$locate("CC=",cparm) .lt. f$length(cparm)
+$!
+$! Check if parameter actually contains something
+$!
+$   if f$edit(cparm,"trim") .nes. ""
 $   then
-$     start  = f$locate("=",cparm) + 1
-$     len    = f$length(cparm) - start
-$     cc_com = f$extract(start,len,cparm)
-      if (cc_com .nes. "DECC") .and. -
-         (cc_com .nes. "VAXC") .and. -
-	 (cc_com .nes. "GNUC")
+$     if cparm .eqs. "DEBUG"
 $     then
-$       write sys$output "Unsupported compiler choice ''cc_com' ignored"
-$       write sys$output "Use DECC, VAXC, or GNUC instead"
-$     else
-$     	if cc_com .eqs. "DECC" then its_decc = true
-$     	if cc_com .eqs. "VAXC" then its_vaxc = true
-$     	if cc_com .eqs. "GNUC" then its_gnuc = true
+$       ccopt = ccopt + "/noopt/deb"
+$       lopts = lopts + "/deb"
 $     endif
-$   endif
-$   if f$locate("MAKE=",cparm) .lt. f$length(cparm)
-$   then
-$     start  = f$locate("=",cparm) + 1
-$     len    = f$length(cparm) - start
-$     mmks = f$extract(start,len,cparm)
-$     if (mmks .eqs. "MMK") .or. (mmks .eqs. "MMS")
+$     if f$locate("CCOPT=",cparm) .lt. f$length(cparm)
 $     then
-$       make = mmks
-$     else
-$       write sys$output "Unsupported make choice ''mmks' ignored"
-$       write sys$output "Use MMK or MMS instead"
+$       start = f$locate("=",cparm) + 1
+$       len   = f$length(cparm) - start
+$       ccopt = ccopt + f$extract(start,len,cparm)
+$       if f$locate("AS_IS",f$edit(ccopt,"UPCASE")) .lt. f$length(ccopt) -
+          then s_case = true
 $     endif
+$     if cparm .eqs. "LINK" then linkonly = true
+$     if f$locate("LOPTS=",cparm) .lt. f$length(cparm)
+$     then
+$       start = f$locate("=",cparm) + 1
+$       len   = f$length(cparm) - start
+$       lopts = lopts + f$extract(start,len,cparm)
+$     endif
+$     if f$locate("CC=",cparm) .lt. f$length(cparm)
+$     then
+$       start  = f$locate("=",cparm) + 1
+$       len    = f$length(cparm) - start
+$       cc_com = f$extract(start,len,cparm)
+        if (cc_com .nes. "DECC") .and. -
+           (cc_com .nes. "VAXC") .and. -
+           (cc_com .nes. "GNUC")
+$       then
+$         write sys$output "Unsupported compiler choice ''cc_com' ignored"
+$         write sys$output "Use DECC, VAXC, or GNUC instead"
+$       else
+$         if cc_com .eqs. "DECC" then its_decc = true
+$         if cc_com .eqs. "VAXC" then its_vaxc = true
+$         if cc_com .eqs. "GNUC" then its_gnuc = true
+$       endif
+$     endif
+$     if f$locate("MAKE=",cparm) .lt. f$length(cparm)
+$     then
+$       start  = f$locate("=",cparm) + 1
+$       len    = f$length(cparm) - start
+$       mmks = f$extract(start,len,cparm)
+$       if (mmks .eqs. "MMK") .or. (mmks .eqs. "MMS")
+$       then
+$         make = mmks
+$       else
+$         write sys$output "Unsupported make choice ''mmks' ignored"
+$         write sys$output "Use MMK or MMS instead"
+$       endif
+$     endif
+$     if cparm .eqs. "HELP" then gosub bhelp
 $   endif
 $   i = i + 1
 $   goto opt_loop
@@ -244,6 +345,11 @@
 $!
 $! Look for the compiler used
 $!
+$! Version history
+$! 0.01 20040223 First version to receive a number
+$! 0.02 20040229 Save/set value of decc$no_rooted_search_lists
+$! 0.03 20060202 Extend handling of GNU C
+$! 0.04 20090402 Compaq -> hp
 $CHECK_COMPILER:
 $ if (.not. (its_decc .or. its_vaxc .or. its_gnuc))
 $ then
@@ -257,9 +363,26 @@
 $ if (.not. (its_decc .or. its_vaxc .or. its_gnuc))
 $ then goto CC_ERR
 $ else
-$   if its_decc then write sys$output "CC compiler check ... Compaq C"
-$   if its_vaxc then write sys$output "CC compiler check ... VAX C"
-$   if its_gnuc then write sys$output "CC compiler check ... GNU C"
+$   if its_decc
+$   then
+$     write sys$output "CC compiler check ... hp C"
+$     if f$trnlnm("decc$no_rooted_search_lists") .nes. ""
+$     then
+$       dnrsl = f$trnlnm("decc$no_rooted_search_lists")
+$     endif
+$     define/nolog decc$no_rooted_search_lists 1
+$   else
+$     if its_vaxc then write sys$output "CC compiler check ... VAX C"
+$     if its_gnuc
+$     then
+$         write sys$output "CC compiler check ... GNU C"
+$         if f$trnlnm(topt) then write topt "gnu_cc:[000000]gcclib.olb/lib"
+$         if f$trnlnm(optf) then write optf "gnu_cc:[000000]gcclib.olb/lib"
+$         cc = "gcc"
+$     endif
+$     if f$trnlnm(topt) then write topt "sys$share:vaxcrtl.exe/share"
+$     if f$trnlnm(optf) then write optf "sys$share:vaxcrtl.exe/share"
+$   endif
 $ endif
 $ return
 $!------------------------------------------------------------------------------
@@ -276,7 +399,8 @@
 # written by Martin P.J. Zinser
 # <zinser at zinser.no-ip.info or zinser at sysdev.deutsche-boerse.com>
 
-OBJS = adler32.obj, compress.obj, crc32.obj, gzio.obj, uncompr.obj, infback.obj\
+OBJS = adler32.obj, compress.obj, crc32.obj, gzclose.obj, gzlib.obj\
+       gzread.obj, gzwrite.obj, uncompr.obj, infback.obj\
        deflate.obj, trees.obj, zutil.obj, inflate.obj, \
        inftrees.obj, inffast.obj
 
@@ -308,7 +432,10 @@
 crc32.obj    : crc32.c zutil.h zlib.h zconf.h
 deflate.obj  : deflate.c deflate.h zutil.h zlib.h zconf.h
 example.obj  : example.c zlib.h zconf.h
-gzio.obj     : gzio.c zutil.h zlib.h zconf.h
+gzclose.obj  : gzclose.c zutil.h zlib.h zconf.h
+gzlib.obj    : gzlib.c zutil.h zlib.h zconf.h
+gzread.obj   : gzread.c zutil.h zlib.h zconf.h
+gzwrite.obj  : gzwrite.c zutil.h zlib.h zconf.h
 inffast.obj  : inffast.c zutil.h zlib.h zconf.h inftrees.h inffast.h
 inflate.obj  : inflate.c zutil.h zlib.h zconf.h
 inftrees.obj : inftrees.c zutil.h zlib.h zconf.h inftrees.h
@@ -328,7 +455,7 @@
 $CREA_OLIST:
 $ open/read min makefile.in
 $ open/write mod modules.opt
-$ src_check = "OBJS ="
+$ src_check = "OBJC ="
 $MRLOOP:
 $ read/end=mrdone min rec
 $ if (f$extract(0,6,rec) .nes. src_check) then goto mrloop
@@ -382,13 +509,182 @@
 $ return
 $!------------------------------------------------------------------------------
 $!
+$CHECK_CONFIG:
+$!
+$ in_ldef = f$locate(cdef,libdefs)
+$ if (in_ldef .lt. f$length(libdefs))
+$ then
+$   write aconf "#define ''cdef' 1"
+$   libdefs = f$extract(0,in_ldef,libdefs) + -
+              f$extract(in_ldef + f$length(cdef) + 1, -
+                        f$length(libdefs) - in_ldef - f$length(cdef) - 1, -
+                        libdefs)
+$ else
+$   if (f$type('cdef') .eqs. "INTEGER")
+$   then
+$     write aconf "#define ''cdef' ", 'cdef'
+$   else
+$     if (f$type('cdef') .eqs. "STRING")
+$     then
+$       write aconf "#define ''cdef' ", """", '''cdef'', """"
+$     else
+$       gosub check_cc_def
+$     endif
+$   endif
+$ endif
+$ return
+$!------------------------------------------------------------------------------
+$!
+$! Check if this is a define relating to the properties of the C/C++
+$! compiler
+$!
+$ CHECK_CC_DEF:
+$ if (cdef .eqs. "_LARGEFILE64_SOURCE")
+$ then
+$   copy sys$input: 'tc'
+$   deck
+#include "tconfig"
+#define _LARGEFILE
+#include <stdio.h>
+
+int main(){
+FILE *fp;
+  fp = fopen("temp.txt","r");
+  fseeko(fp,1,SEEK_SET);
+  fclose(fp);
+}
+
+$   eod
+$   test_inv = false
+$   comm_h = false
+$   gosub cc_prop_check
+$   return
+$ endif
+$ write aconf "/* ", line, " */"
+$ return
+$!------------------------------------------------------------------------------
+$!
+$! Check for properties of C/C++ compiler
+$!
+$! Version history
+$! 0.01 20031020 First version to receive a number
+$! 0.02 20031022 Added logic for defines with value
+$! 0.03 20040309 Make sure local config file gets not deleted
+$! 0.04 20041230 Also write include for configure run
+$! 0.05 20050103 Add processing of "comment defines"
+$CC_PROP_CHECK:
+$ cc_prop = true
+$ is_need = false
+$ is_need = (f$extract(0,4,cdef) .eqs. "NEED") .or. (test_inv .eq. true)
+$ if f$search(th) .eqs. "" then create 'th'
+$ set message/nofac/noident/nosever/notext
+$ on error then continue
+$ cc 'tmpnam'
+$ if .not. ($status)  then cc_prop = false
+$ on error then continue
+$! The headers might lie about the capabilities of the RTL
+$ link 'tmpnam',tmp.opt/opt
+$ if .not. ($status)  then cc_prop = false
+$ set message/fac/ident/sever/text
+$ on error then goto err_exit
+$ delete/nolog 'tmpnam'.*;*/exclude='th'
+$ if (cc_prop .and. .not. is_need) .or. -
+     (.not. cc_prop .and. is_need)
+$ then
+$   write sys$output "Checking for ''cdef'... yes"
+$   if f$type('cdef_val'_yes) .nes. ""
+$   then
+$     if f$type('cdef_val'_yes) .eqs. "INTEGER" -
+         then call write_config f$fao("#define !AS !UL",cdef,'cdef_val'_yes)
+$     if f$type('cdef_val'_yes) .eqs. "STRING" -
+         then call write_config f$fao("#define !AS !AS",cdef,'cdef_val'_yes)
+$   else
+$     call write_config f$fao("#define !AS 1",cdef)
+$   endif
+$   if (cdef .eqs. "HAVE_FSEEKO") .or. (cdef .eqs. "_LARGE_FILES") .or. -
+       (cdef .eqs. "_LARGEFILE64_SOURCE") then -
+      call write_config f$string("#define _LARGEFILE 1")
+$ else
+$   write sys$output "Checking for ''cdef'... no"
+$   if (comm_h)
+$   then
+      call write_config f$fao("/* !AS */",line)
+$   else
+$     if f$type('cdef_val'_no) .nes. ""
+$     then
+$       if f$type('cdef_val'_no) .eqs. "INTEGER" -
+           then call write_config f$fao("#define !AS !UL",cdef,'cdef_val'_no)
+$       if f$type('cdef_val'_no) .eqs. "STRING" -
+           then call write_config f$fao("#define !AS !AS",cdef,'cdef_val'_no)
+$     else
+$       call write_config f$fao("#undef !AS",cdef)
+$     endif
+$   endif
+$ endif
+$ return
+$!------------------------------------------------------------------------------
+$!
+$! Check for properties of C/C++ compiler with multiple result values
+$!
+$! Version history
+$! 0.01 20040127 First version
+$! 0.02 20050103 Reconcile changes from cc_prop up to version 0.05
+$CC_MPROP_CHECK:
+$ cc_prop = true
+$ i    = 1
+$ idel = 1
+$ MT_LOOP:
+$ if f$type(result_'i') .eqs. "STRING"
+$ then
+$   set message/nofac/noident/nosever/notext
+$   on error then continue
+$   cc 'tmpnam'_'i'
+$   if .not. ($status)  then cc_prop = false
+$   on error then continue
+$! The headers might lie about the capabilities of the RTL
+$   link 'tmpnam'_'i',tmp.opt/opt
+$   if .not. ($status)  then cc_prop = false
+$   set message/fac/ident/sever/text
+$   on error then goto err_exit
+$   delete/nolog 'tmpnam'_'i'.*;*
+$   if (cc_prop)
+$   then
+$     write sys$output "Checking for ''cdef'... ", mdef_'i'
+$     if f$type(mdef_'i') .eqs. "INTEGER" -
+         then call write_config f$fao("#define !AS !UL",cdef,mdef_'i')
+$     if f$type('cdef_val'_yes) .eqs. "STRING" -
+         then call write_config f$fao("#define !AS !AS",cdef,mdef_'i')
+$     goto msym_clean
+$   else
+$     i = i + 1
+$     goto mt_loop
+$   endif
+$ endif
+$ write sys$output "Checking for ''cdef'... no"
+$ call write_config f$fao("#undef !AS",cdef)
+$ MSYM_CLEAN:
+$ if (idel .le. msym_max)
+$ then
+$   delete/sym mdef_'idel'
+$   idel = idel + 1
+$   goto msym_clean
+$ endif
+$ return
+$!------------------------------------------------------------------------------
+$!
 $! Analyze Object files for OpenVMS AXP to extract Procedure and Data
 $! information to build a symbol vector for a shareable image
 $! All the "brains" of this logic was suggested by Hartmut Becker
 $! (Hartmut.Becker at compaq.com). All the bugs were introduced by me
-$! (zinser at decus.de), so if you do have problem reports please do not
+$! (zinser at zinser.no-ip.info), so if you do have problem reports please do not
 $! bother Hartmut/HP, but get in touch with me
 $!
+$! Version history
+$! 0.01 20040406 Skip over shareable images in option file
+$! 0.02 20041109 Fix option file for shareable images with case_sensitive=YES
+$! 0.03 20050107 Skip over Identification labels in option file
+$! 0.04 20060117 Add uppercase alias to code compiled with /name=as_is
+$!
 $ ANAL_OBJ_AXP: Subroutine
 $ V = 'F$Verify(0)
 $ SAY := "WRITE_ SYS$OUTPUT"
@@ -409,6 +705,17 @@
 $ open/append atmp a.tmp
 $ loop:
 $ read/end=end_loop in line
+$ if f$locate("/SHARE",f$edit(line,"upcase")) .lt. f$length(line)
+$ then
+$   write sys$output "ANAL_SKP_SHR-i-skipshare, ''line'"
+$   goto loop
+$ endif
+$ if f$locate("IDENTIFICATION=",f$edit(line,"upcase")) .lt. f$length(line)
+$ then
+$   write sys$output "ANAL_OBJ_AXP-i-ident: Identification ", -
+                     f$element(1,"=",line)
+$   goto loop
+$ endif
 $ f= f$search(line)
 $ if f .eqs. ""
 $ then
@@ -450,8 +757,31 @@
 sub/symbol: "/symbol_vector=(/whole
 sub/"/=DATA)/whole
 exit
-$ sort/nodupl d.tmp,f.tmp 'p2'
-$ delete a.tmp;*,b.tmp;*,c.tmp;*,d.tmp;*,e.tmp;*,f.tmp;*
+$ sort/nodupl d.tmp,f.tmp g.tmp
+$ open/read raw_vector g.tmp
+$ open/write case_vector 'p2'
+$ RAWLOOP:
+$ read/end=end_rawloop raw_vector raw_element
+$ write case_vector raw_element
+$ if f$locate("=PROCEDURE)",raw_element) .lt. f$length(raw_element)
+$ then
+$     name = f$element(1,"=",raw_element) - "("
+$     if f$edit(name,"UPCASE") .nes. name then -
+          write case_vector f$fao(" symbol_vector=(!AS/!AS=PROCEDURE)", -
+	                          f$edit(name,"UPCASE"), name)
+$ endif
+$ if f$locate("=DATA)",raw_element) .lt. f$length(raw_element)
+$ then
+$     name = f$element(1,"=",raw_element) - "("
+$     if f$edit(name,"UPCASE") .nes. name then -
+          write case_vector f$fao(" symbol_vector=(!AS/!AS=DATA)", -
+	                          f$edit(name,"UPCASE"), name)
+$ endif
+$ goto rawloop
+$ END_RAWLOOP:
+$ close raw_vector
+$ close case_vector
+$ delete a.tmp;*,b.tmp;*,c.tmp;*,d.tmp;*,e.tmp;*,f.tmp;*,g.tmp;*
 $ if f$search("x.tmp") .nes. "" -
 	then $ delete x.tmp;*
 $!
@@ -459,3 +789,16 @@
 $ if V then set verify
 $ endsubroutine
 $!------------------------------------------------------------------------------
+$!
+$! Write configuration to both permanent and temporary config file
+$!
+$! Version history
+$! 0.01 20031029 First version to receive a number
+$!
+$WRITE_CONFIG: SUBROUTINE
+$  write aconf 'p1'
+$  open/append confh 'th'
+$  write confh 'p1'
+$  close confh
+$ENDSUBROUTINE
+$!------------------------------------------------------------------------------

Modified: python/branches/import_unicode/Modules/zlib/minigzip.c
==============================================================================
--- python/branches/import_unicode/Modules/zlib/minigzip.c	(original)
+++ python/branches/import_unicode/Modules/zlib/minigzip.c	Sat Jul 31 13:01:07 2010
@@ -1,5 +1,5 @@
 /* minigzip.c -- simulate gzip using the zlib compression library
- * Copyright (C) 1995-2005 Jean-loup Gailly.
+ * Copyright (C) 1995-2006, 2010 Jean-loup Gailly.
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -15,8 +15,8 @@
 
 /* @(#) $Id$ */
 
-#include <stdio.h>
 #include "zlib.h"
+#include <stdio.h>
 
 #ifdef STDC
 #  include <string.h>
@@ -32,6 +32,9 @@
 #if defined(MSDOS) || defined(OS2) || defined(WIN32) || defined(__CYGWIN__)
 #  include <fcntl.h>
 #  include <io.h>
+#  ifdef UNDER_CE
+#    include <stdlib.h>
+#  endif
 #  define SET_BINARY_MODE(file) setmode(fileno(file), O_BINARY)
 #else
 #  define SET_BINARY_MODE(file)
@@ -41,13 +44,84 @@
 #  define unlink delete
 #  define GZ_SUFFIX "-gz"
 #endif
+#ifdef RISCOS
+#  define unlink remove
+#  define GZ_SUFFIX "-gz"
+#  define fileno(file) file->__file
+#endif
 #if defined(__MWERKS__) && __dest_os != __be_os && __dest_os != __win32_os
 #  include <unix.h> /* for fileno */
 #endif
 
+#if !defined(Z_HAVE_UNISTD_H) && !defined(_LARGEFILE64_SOURCE)
 #ifndef WIN32 /* unlink already in stdio.h for WIN32 */
   extern int unlink OF((const char *));
 #endif
+#endif
+
+#if defined(UNDER_CE)
+#  include <windows.h>
+#  define perror(s) pwinerror(s)
+
+/* Map the Windows error number in ERROR to a locale-dependent error
+   message string and return a pointer to it.  Typically, the values
+   for ERROR come from GetLastError.
+
+   The string pointed to shall not be modified by the application,
+   but may be overwritten by a subsequent call to strwinerror
+
+   The strwinerror function does not change the current setting
+   of GetLastError.  */
+
+static char *strwinerror (error)
+     DWORD error;
+{
+    static char buf[1024];
+
+    wchar_t *msgbuf;
+    DWORD lasterr = GetLastError();
+    DWORD chars = FormatMessage(FORMAT_MESSAGE_FROM_SYSTEM
+        | FORMAT_MESSAGE_ALLOCATE_BUFFER,
+        NULL,
+        error,
+        0, /* Default language */
+        (LPVOID)&msgbuf,
+        0,
+        NULL);
+    if (chars != 0) {
+        /* If there is an \r\n appended, zap it.  */
+        if (chars >= 2
+            && msgbuf[chars - 2] == '\r' && msgbuf[chars - 1] == '\n') {
+            chars -= 2;
+            msgbuf[chars] = 0;
+        }
+
+        if (chars > sizeof (buf) - 1) {
+            chars = sizeof (buf) - 1;
+            msgbuf[chars] = 0;
+        }
+
+        wcstombs(buf, msgbuf, chars + 1);
+        LocalFree(msgbuf);
+    }
+    else {
+        sprintf(buf, "unknown win32 error (%ld)", error);
+    }
+
+    SetLastError(lasterr);
+    return buf;
+}
+
+static void pwinerror (s)
+    const char *s;
+{
+    if (s && *s)
+        fprintf(stderr, "%s: %s\n", s, strwinerror(GetLastError ()));
+    else
+        fprintf(stderr, "%s\n", strwinerror(GetLastError ()));
+}
+
+#endif /* UNDER_CE */
 
 #ifndef GZ_SUFFIX
 #  define GZ_SUFFIX ".gz"
@@ -193,6 +267,11 @@
     FILE  *in;
     gzFile out;
 
+    if (strlen(file) + strlen(GZ_SUFFIX) >= sizeof(outfile)) {
+        fprintf(stderr, "%s: filename too long\n", prog);
+        exit(1);
+    }
+
     strcpy(outfile, file);
     strcat(outfile, GZ_SUFFIX);
 
@@ -222,7 +301,12 @@
     char *infile, *outfile;
     FILE  *out;
     gzFile in;
-    uInt len = (uInt)strlen(file);
+    size_t len = strlen(file);
+
+    if (len + strlen(GZ_SUFFIX) >= sizeof(buf)) {
+        fprintf(stderr, "%s: filename too long\n", prog);
+        exit(1);
+    }
 
     strcpy(buf, file);
 
@@ -253,7 +337,8 @@
 
 
 /* ===========================================================================
- * Usage:  minigzip [-d] [-f] [-h] [-r] [-1 to -9] [files...]
+ * Usage:  minigzip [-c] [-d] [-f] [-h] [-r] [-1 to -9] [files...]
+ *   -c : write to standard output
  *   -d : decompress
  *   -f : compress with Z_FILTERED
  *   -h : compress with Z_HUFFMAN_ONLY
@@ -265,17 +350,30 @@
     int argc;
     char *argv[];
 {
+    int copyout = 0;
     int uncompr = 0;
     gzFile file;
-    char outmode[20];
+    char *bname, outmode[20];
 
     strcpy(outmode, "wb6 ");
 
     prog = argv[0];
+    bname = strrchr(argv[0], '/');
+    if (bname)
+      bname++;
+    else
+      bname = argv[0];
     argc--, argv++;
 
+    if (!strcmp(bname, "gunzip"))
+      uncompr = 1;
+    else if (!strcmp(bname, "zcat"))
+      copyout = uncompr = 1;
+
     while (argc > 0) {
-      if (strcmp(*argv, "-d") == 0)
+      if (strcmp(*argv, "-c") == 0)
+        copyout = 1;
+      else if (strcmp(*argv, "-d") == 0)
         uncompr = 1;
       else if (strcmp(*argv, "-f") == 0)
         outmode[3] = 'f';
@@ -305,11 +403,36 @@
             gz_compress(stdin, file);
         }
     } else {
+        if (copyout) {
+            SET_BINARY_MODE(stdout);
+        }
         do {
             if (uncompr) {
-                file_uncompress(*argv);
+                if (copyout) {
+                    file = gzopen(*argv, "rb");
+                    if (file == NULL)
+                        fprintf(stderr, "%s: can't gzopen %s\n", prog, *argv);
+                    else
+                        gz_uncompress(file, stdout);
+                } else {
+                    file_uncompress(*argv);
+                }
             } else {
-                file_compress(*argv, outmode);
+                if (copyout) {
+                    FILE * in = fopen(*argv, "rb");
+
+                    if (in == NULL) {
+                        perror(*argv);
+                    } else {
+                        file = gzdopen(fileno(stdout), outmode);
+                        if (file == NULL) error("can't gzdopen stdout");
+
+                        gz_compress(in, file);
+                    }
+
+                } else {
+                    file_compress(*argv, outmode);
+                }
             }
         } while (argv++, --argc);
     }

Modified: python/branches/import_unicode/Modules/zlib/trees.c
==============================================================================
--- python/branches/import_unicode/Modules/zlib/trees.c	(original)
+++ python/branches/import_unicode/Modules/zlib/trees.c	Sat Jul 31 13:01:07 2010
@@ -1,5 +1,6 @@
 /* trees.c -- output deflated data using Huffman coding
- * Copyright (C) 1995-2005 Jean-loup Gailly
+ * Copyright (C) 1995-2010 Jean-loup Gailly
+ * detect_data_type() function provided freely by Cosmin Truta, 2006
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -152,7 +153,7 @@
                               int blcodes));
 local void compress_block OF((deflate_state *s, ct_data *ltree,
                               ct_data *dtree));
-local void set_data_type  OF((deflate_state *s));
+local int  detect_data_type OF((deflate_state *s));
 local unsigned bi_reverse OF((unsigned value, int length));
 local void bi_windup      OF((deflate_state *s));
 local void bi_flush       OF((deflate_state *s));
@@ -203,12 +204,12 @@
      * unused bits in value.
      */
     if (s->bi_valid > (int)Buf_size - length) {
-        s->bi_buf |= (value << s->bi_valid);
+        s->bi_buf |= (ush)value << s->bi_valid;
         put_short(s, s->bi_buf);
         s->bi_buf = (ush)value >> (Buf_size - s->bi_valid);
         s->bi_valid += length - Buf_size;
     } else {
-        s->bi_buf |= value << s->bi_valid;
+        s->bi_buf |= (ush)value << s->bi_valid;
         s->bi_valid += length;
     }
 }
@@ -218,12 +219,12 @@
 { int len = length;\
   if (s->bi_valid > (int)Buf_size - len) {\
     int val = value;\
-    s->bi_buf |= (val << s->bi_valid);\
+    s->bi_buf |= (ush)val << s->bi_valid;\
     put_short(s, s->bi_buf);\
     s->bi_buf = (ush)val >> (Buf_size - s->bi_valid);\
     s->bi_valid += len - Buf_size;\
   } else {\
-    s->bi_buf |= (value) << s->bi_valid;\
+    s->bi_buf |= (ush)(value) << s->bi_valid;\
     s->bi_valid += len;\
   }\
 }
@@ -250,11 +251,13 @@
     if (static_init_done) return;
 
     /* For some embedded targets, global variables are not initialized: */
+#ifdef NO_INIT_GLOBAL_POINTERS
     static_l_desc.static_tree = static_ltree;
     static_l_desc.extra_bits = extra_lbits;
     static_d_desc.static_tree = static_dtree;
     static_d_desc.extra_bits = extra_dbits;
     static_bl_desc.extra_bits = extra_blbits;
+#endif
 
     /* Initialize the mapping length (0..255) -> length code (0..28) */
     length = 0;
@@ -348,13 +351,14 @@
                 static_dtree[i].Len, SEPARATOR(i, D_CODES-1, 5));
     }
 
-    fprintf(header, "const uch _dist_code[DIST_CODE_LEN] = {\n");
+    fprintf(header, "const uch ZLIB_INTERNAL _dist_code[DIST_CODE_LEN] = {\n");
     for (i = 0; i < DIST_CODE_LEN; i++) {
         fprintf(header, "%2u%s", _dist_code[i],
                 SEPARATOR(i, DIST_CODE_LEN-1, 20));
     }
 
-    fprintf(header, "const uch _length_code[MAX_MATCH-MIN_MATCH+1]= {\n");
+    fprintf(header,
+        "const uch ZLIB_INTERNAL _length_code[MAX_MATCH-MIN_MATCH+1]= {\n");
     for (i = 0; i < MAX_MATCH-MIN_MATCH+1; i++) {
         fprintf(header, "%2u%s", _length_code[i],
                 SEPARATOR(i, MAX_MATCH-MIN_MATCH, 20));
@@ -379,7 +383,7 @@
 /* ===========================================================================
  * Initialize the tree data structures for a new zlib stream.
  */
-void _tr_init(s)
+void ZLIB_INTERNAL _tr_init(s)
     deflate_state *s;
 {
     tr_static_init();
@@ -864,13 +868,13 @@
 /* ===========================================================================
  * Send a stored block
  */
-void _tr_stored_block(s, buf, stored_len, eof)
+void ZLIB_INTERNAL _tr_stored_block(s, buf, stored_len, last)
     deflate_state *s;
     charf *buf;       /* input block */
     ulg stored_len;   /* length of input block */
-    int eof;          /* true if this is the last block for a file */
+    int last;         /* one if this is the last block for a file */
 {
-    send_bits(s, (STORED_BLOCK<<1)+eof, 3);  /* send block type */
+    send_bits(s, (STORED_BLOCK<<1)+last, 3);    /* send block type */
 #ifdef DEBUG
     s->compressed_len = (s->compressed_len + 3 + 7) & (ulg)~7L;
     s->compressed_len += (stored_len + 4) << 3;
@@ -889,7 +893,7 @@
  * To simplify the code, we assume the worst case of last real code encoded
  * on one bit only.
  */
-void _tr_align(s)
+void ZLIB_INTERNAL _tr_align(s)
     deflate_state *s;
 {
     send_bits(s, STATIC_TREES<<1, 3);
@@ -918,11 +922,11 @@
  * Determine the best encoding for the current block: dynamic trees, static
  * trees or store, and output the encoded block to the zip file.
  */
-void _tr_flush_block(s, buf, stored_len, eof)
+void ZLIB_INTERNAL _tr_flush_block(s, buf, stored_len, last)
     deflate_state *s;
     charf *buf;       /* input block, or NULL if too old */
     ulg stored_len;   /* length of input block */
-    int eof;          /* true if this is the last block for a file */
+    int last;         /* one if this is the last block for a file */
 {
     ulg opt_lenb, static_lenb; /* opt_len and static_len in bytes */
     int max_blindex = 0;  /* index of last bit length code of non zero freq */
@@ -931,8 +935,8 @@
     if (s->level > 0) {
 
         /* Check if the file is binary or text */
-        if (stored_len > 0 && s->strm->data_type == Z_UNKNOWN)
-            set_data_type(s);
+        if (s->strm->data_type == Z_UNKNOWN)
+            s->strm->data_type = detect_data_type(s);
 
         /* Construct the literal and distance trees */
         build_tree(s, (tree_desc *)(&(s->l_desc)));
@@ -978,20 +982,20 @@
          * successful. If LIT_BUFSIZE <= WSIZE, it is never too late to
          * transform a block into a stored block.
          */
-        _tr_stored_block(s, buf, stored_len, eof);
+        _tr_stored_block(s, buf, stored_len, last);
 
 #ifdef FORCE_STATIC
     } else if (static_lenb >= 0) { /* force static trees */
 #else
     } else if (s->strategy == Z_FIXED || static_lenb == opt_lenb) {
 #endif
-        send_bits(s, (STATIC_TREES<<1)+eof, 3);
+        send_bits(s, (STATIC_TREES<<1)+last, 3);
         compress_block(s, (ct_data *)static_ltree, (ct_data *)static_dtree);
 #ifdef DEBUG
         s->compressed_len += 3 + s->static_len;
 #endif
     } else {
-        send_bits(s, (DYN_TREES<<1)+eof, 3);
+        send_bits(s, (DYN_TREES<<1)+last, 3);
         send_all_trees(s, s->l_desc.max_code+1, s->d_desc.max_code+1,
                        max_blindex+1);
         compress_block(s, (ct_data *)s->dyn_ltree, (ct_data *)s->dyn_dtree);
@@ -1005,21 +1009,21 @@
      */
     init_block(s);
 
-    if (eof) {
+    if (last) {
         bi_windup(s);
 #ifdef DEBUG
         s->compressed_len += 7;  /* align on byte boundary */
 #endif
     }
     Tracev((stderr,"\ncomprlen %lu(%lu) ", s->compressed_len>>3,
-           s->compressed_len-7*eof));
+           s->compressed_len-7*last));
 }
 
 /* ===========================================================================
  * Save the match info and tally the frequency counts. Return true if
  * the current block must be flushed.
  */
-int _tr_tally (s, dist, lc)
+int ZLIB_INTERNAL _tr_tally (s, dist, lc)
     deflate_state *s;
     unsigned dist;  /* distance of matched string */
     unsigned lc;    /* match length-MIN_MATCH or unmatched char (if dist==0) */
@@ -1118,24 +1122,45 @@
 }
 
 /* ===========================================================================
- * Set the data type to BINARY or TEXT, using a crude approximation:
- * set it to Z_TEXT if all symbols are either printable characters (33 to 255)
- * or white spaces (9 to 13, or 32); or set it to Z_BINARY otherwise.
+ * Check if the data type is TEXT or BINARY, using the following algorithm:
+ * - TEXT if the two conditions below are satisfied:
+ *    a) There are no non-portable control characters belonging to the
+ *       "black list" (0..6, 14..25, 28..31).
+ *    b) There is at least one printable character belonging to the
+ *       "white list" (9 {TAB}, 10 {LF}, 13 {CR}, 32..255).
+ * - BINARY otherwise.
+ * - The following partially-portable control characters form a
+ *   "gray list" that is ignored in this detection algorithm:
+ *   (7 {BEL}, 8 {BS}, 11 {VT}, 12 {FF}, 26 {SUB}, 27 {ESC}).
  * IN assertion: the fields Freq of dyn_ltree are set.
  */
-local void set_data_type(s)
+local int detect_data_type(s)
     deflate_state *s;
 {
+    /* black_mask is the bit mask of black-listed bytes
+     * set bits 0..6, 14..25, and 28..31
+     * 0xf3ffc07f = binary 11110011111111111100000001111111
+     */
+    unsigned long black_mask = 0xf3ffc07fUL;
     int n;
 
-    for (n = 0; n < 9; n++)
+    /* Check for non-textual ("black-listed") bytes. */
+    for (n = 0; n <= 31; n++, black_mask >>= 1)
+        if ((black_mask & 1) && (s->dyn_ltree[n].Freq != 0))
+            return Z_BINARY;
+
+    /* Check for textual ("white-listed") bytes. */
+    if (s->dyn_ltree[9].Freq != 0 || s->dyn_ltree[10].Freq != 0
+            || s->dyn_ltree[13].Freq != 0)
+        return Z_TEXT;
+    for (n = 32; n < LITERALS; n++)
         if (s->dyn_ltree[n].Freq != 0)
-            break;
-    if (n == 9)
-        for (n = 14; n < 32; n++)
-            if (s->dyn_ltree[n].Freq != 0)
-                break;
-    s->strm->data_type = (n == 32) ? Z_TEXT : Z_BINARY;
+            return Z_TEXT;
+
+    /* There are no "black-listed" or "white-listed" bytes:
+     * this stream either is empty or has tolerated ("gray-listed") bytes only.
+     */
+    return Z_BINARY;
 }
 
 /* ===========================================================================

Modified: python/branches/import_unicode/Modules/zlib/trees.h
==============================================================================
--- python/branches/import_unicode/Modules/zlib/trees.h	(original)
+++ python/branches/import_unicode/Modules/zlib/trees.h	Sat Jul 31 13:01:07 2010
@@ -70,7 +70,7 @@
 {{19},{ 5}}, {{11},{ 5}}, {{27},{ 5}}, {{ 7},{ 5}}, {{23},{ 5}}
 };
 
-const uch _dist_code[DIST_CODE_LEN] = {
+const uch ZLIB_INTERNAL _dist_code[DIST_CODE_LEN] = {
  0,  1,  2,  3,  4,  4,  5,  5,  6,  6,  6,  6,  7,  7,  7,  7,  8,  8,  8,  8,
  8,  8,  8,  8,  9,  9,  9,  9,  9,  9,  9,  9, 10, 10, 10, 10, 10, 10, 10, 10,
 10, 10, 10, 10, 10, 10, 10, 10, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11,
@@ -99,7 +99,7 @@
 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29
 };
 
-const uch _length_code[MAX_MATCH-MIN_MATCH+1]= {
+const uch ZLIB_INTERNAL _length_code[MAX_MATCH-MIN_MATCH+1]= {
  0,  1,  2,  3,  4,  5,  6,  7,  8,  8,  9,  9, 10, 10, 11, 11, 12, 12, 12, 12,
 13, 13, 13, 13, 14, 14, 14, 14, 15, 15, 15, 15, 16, 16, 16, 16, 16, 16, 16, 16,
 17, 17, 17, 17, 17, 17, 17, 17, 18, 18, 18, 18, 18, 18, 18, 18, 19, 19, 19, 19,

Modified: python/branches/import_unicode/Modules/zlib/uncompr.c
==============================================================================
--- python/branches/import_unicode/Modules/zlib/uncompr.c	(original)
+++ python/branches/import_unicode/Modules/zlib/uncompr.c	Sat Jul 31 13:01:07 2010
@@ -1,5 +1,5 @@
 /* uncompr.c -- decompress a memory buffer
- * Copyright (C) 1995-2003 Jean-loup Gailly.
+ * Copyright (C) 1995-2003, 2010 Jean-loup Gailly.
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -16,8 +16,6 @@
    been saved previously by the compressor and transmitted to the decompressor
    by some mechanism outside the scope of this compression library.)
    Upon exit, destLen is the actual size of the compressed buffer.
-     This function can be used to decompress a whole file at once if the
-   input file is mmap'ed.
 
      uncompress returns Z_OK if success, Z_MEM_ERROR if there was not
    enough memory, Z_BUF_ERROR if there was not enough room in the output

Modified: python/branches/import_unicode/Modules/zlib/zconf.h
==============================================================================
--- python/branches/import_unicode/Modules/zlib/zconf.h	(original)
+++ python/branches/import_unicode/Modules/zlib/zconf.h	Sat Jul 31 13:01:07 2010
@@ -1,5 +1,5 @@
 /* zconf.h -- configuration of the zlib compression library
- * Copyright (C) 1995-2005 Jean-loup Gailly.
+ * Copyright (C) 1995-2010 Jean-loup Gailly.
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -11,52 +11,124 @@
 /*
  * If you *really* need a unique prefix for all types and library functions,
  * compile with -DZ_PREFIX. The "standard" zlib should be compiled without it.
+ * Even better than compiling with -DZ_PREFIX would be to use configure to set
+ * this permanently in zconf.h using "./configure --zprefix".
  */
-#ifdef Z_PREFIX
-#  define deflateInit_          z_deflateInit_
+#ifdef Z_PREFIX     /* may be set to #if 1 by ./configure */
+
+/* all linked symbols */
+#  define _dist_code            z__dist_code
+#  define _length_code          z__length_code
+#  define _tr_align             z__tr_align
+#  define _tr_flush_block       z__tr_flush_block
+#  define _tr_init              z__tr_init
+#  define _tr_stored_block      z__tr_stored_block
+#  define _tr_tally             z__tr_tally
+#  define adler32               z_adler32
+#  define adler32_combine       z_adler32_combine
+#  define adler32_combine64     z_adler32_combine64
+#  define compress              z_compress
+#  define compress2             z_compress2
+#  define compressBound         z_compressBound
+#  define crc32                 z_crc32
+#  define crc32_combine         z_crc32_combine
+#  define crc32_combine64       z_crc32_combine64
 #  define deflate               z_deflate
+#  define deflateBound          z_deflateBound
+#  define deflateCopy           z_deflateCopy
 #  define deflateEnd            z_deflateEnd
-#  define inflateInit_          z_inflateInit_
-#  define inflate               z_inflate
-#  define inflateEnd            z_inflateEnd
 #  define deflateInit2_         z_deflateInit2_
-#  define deflateSetDictionary  z_deflateSetDictionary
-#  define deflateCopy           z_deflateCopy
-#  define deflateReset          z_deflateReset
+#  define deflateInit_          z_deflateInit_
 #  define deflateParams         z_deflateParams
-#  define deflateBound          z_deflateBound
 #  define deflatePrime          z_deflatePrime
+#  define deflateReset          z_deflateReset
+#  define deflateSetDictionary  z_deflateSetDictionary
+#  define deflateSetHeader      z_deflateSetHeader
+#  define deflateTune           z_deflateTune
+#  define deflate_copyright     z_deflate_copyright
+#  define get_crc_table         z_get_crc_table
+#  define gz_error              z_gz_error
+#  define gz_intmax             z_gz_intmax
+#  define gz_strwinerror        z_gz_strwinerror
+#  define gzbuffer              z_gzbuffer
+#  define gzclearerr            z_gzclearerr
+#  define gzclose               z_gzclose
+#  define gzclose_r             z_gzclose_r
+#  define gzclose_w             z_gzclose_w
+#  define gzdirect              z_gzdirect
+#  define gzdopen               z_gzdopen
+#  define gzeof                 z_gzeof
+#  define gzerror               z_gzerror
+#  define gzflush               z_gzflush
+#  define gzgetc                z_gzgetc
+#  define gzgets                z_gzgets
+#  define gzoffset              z_gzoffset
+#  define gzoffset64            z_gzoffset64
+#  define gzopen                z_gzopen
+#  define gzopen64              z_gzopen64
+#  define gzprintf              z_gzprintf
+#  define gzputc                z_gzputc
+#  define gzputs                z_gzputs
+#  define gzread                z_gzread
+#  define gzrewind              z_gzrewind
+#  define gzseek                z_gzseek
+#  define gzseek64              z_gzseek64
+#  define gzsetparams           z_gzsetparams
+#  define gztell                z_gztell
+#  define gztell64              z_gztell64
+#  define gzungetc              z_gzungetc
+#  define gzwrite               z_gzwrite
+#  define inflate               z_inflate
+#  define inflateBack           z_inflateBack
+#  define inflateBackEnd        z_inflateBackEnd
+#  define inflateBackInit_      z_inflateBackInit_
+#  define inflateCopy           z_inflateCopy
+#  define inflateEnd            z_inflateEnd
+#  define inflateGetHeader      z_inflateGetHeader
 #  define inflateInit2_         z_inflateInit2_
+#  define inflateInit_          z_inflateInit_
+#  define inflateMark           z_inflateMark
+#  define inflatePrime          z_inflatePrime
+#  define inflateReset          z_inflateReset
+#  define inflateReset2         z_inflateReset2
 #  define inflateSetDictionary  z_inflateSetDictionary
 #  define inflateSync           z_inflateSync
 #  define inflateSyncPoint      z_inflateSyncPoint
-#  define inflateCopy           z_inflateCopy
-#  define inflateReset          z_inflateReset
-#  define inflateBack           z_inflateBack
-#  define inflateBackEnd        z_inflateBackEnd
-#  define compress              z_compress
-#  define compress2             z_compress2
-#  define compressBound         z_compressBound
+#  define inflateUndermine      z_inflateUndermine
+#  define inflate_copyright     z_inflate_copyright
+#  define inflate_fast          z_inflate_fast
+#  define inflate_table         z_inflate_table
 #  define uncompress            z_uncompress
-#  define adler32               z_adler32
-#  define crc32                 z_crc32
-#  define get_crc_table         z_get_crc_table
 #  define zError                z_zError
+#  define zcalloc               z_zcalloc
+#  define zcfree                z_zcfree
+#  define zlibCompileFlags      z_zlibCompileFlags
+#  define zlibVersion           z_zlibVersion
 
+/* all zlib typedefs in zlib.h and zconf.h */
+#  define Byte                  z_Byte
+#  define Bytef                 z_Bytef
 #  define alloc_func            z_alloc_func
+#  define charf                 z_charf
 #  define free_func             z_free_func
+#  define gzFile                z_gzFile
+#  define gz_header             z_gz_header
+#  define gz_headerp            z_gz_headerp
 #  define in_func               z_in_func
+#  define intf                  z_intf
 #  define out_func              z_out_func
-#  define Byte                  z_Byte
 #  define uInt                  z_uInt
-#  define uLong                 z_uLong
-#  define Bytef                 z_Bytef
-#  define charf                 z_charf
-#  define intf                  z_intf
 #  define uIntf                 z_uIntf
+#  define uLong                 z_uLong
 #  define uLongf                z_uLongf
-#  define voidpf                z_voidpf
 #  define voidp                 z_voidp
+#  define voidpc                z_voidpc
+#  define voidpf                z_voidpf
+
+/* all zlib structs in zlib.h and zconf.h */
+#  define gz_header_s           z_gz_header_s
+#  define internal_state        z_internal_state
+
 #endif
 
 #if defined(__MSDOS__) && !defined(MSDOS)
@@ -284,49 +356,73 @@
    typedef Byte       *voidp;
 #endif
 
-#if 0           /* HAVE_UNISTD_H -- this line is updated by ./configure */
-#  include <sys/types.h> /* for off_t */
-#  include <unistd.h>    /* for SEEK_* and off_t */
+#ifdef HAVE_UNISTD_H    /* may be set to #if 1 by ./configure */
+#  define Z_HAVE_UNISTD_H
+#endif
+
+#ifdef STDC
+#  include <sys/types.h>    /* for off_t */
+#endif
+
+/* a little trick to accommodate both "#define _LARGEFILE64_SOURCE" and
+ * "#define _LARGEFILE64_SOURCE 1" as requesting 64-bit operations, (even
+ * though the former does not conform to the LFS document), but considering
+ * both "#undef _LARGEFILE64_SOURCE" and "#define _LARGEFILE64_SOURCE 0" as
+ * equivalently requesting no 64-bit operations
+ */
+#if -_LARGEFILE64_SOURCE - -1 == 1
+#  undef _LARGEFILE64_SOURCE
+#endif
+
+#if defined(Z_HAVE_UNISTD_H) || defined(_LARGEFILE64_SOURCE)
+#  include <unistd.h>       /* for SEEK_* and off_t */
 #  ifdef VMS
-#    include <unixio.h>   /* for off_t */
+#    include <unixio.h>     /* for off_t */
+#  endif
+#  ifndef z_off_t
+#    define z_off_t off_t
 #  endif
-#  define z_off_t off_t
 #endif
+
 #ifndef SEEK_SET
 #  define SEEK_SET        0       /* Seek from beginning of file.  */
 #  define SEEK_CUR        1       /* Seek from current position.  */
 #  define SEEK_END        2       /* Set file pointer to EOF plus "offset" */
 #endif
+
 #ifndef z_off_t
 #  define z_off_t long
 #endif
 
+#if defined(_LARGEFILE64_SOURCE) && _LFS64_LARGEFILE-0
+#  define z_off64_t off64_t
+#else
+#  define z_off64_t z_off_t
+#endif
+
 #if defined(__OS400__)
 #  define NO_vsnprintf
 #endif
 
 #if defined(__MVS__)
 #  define NO_vsnprintf
-#  ifdef FAR
-#    undef FAR
-#  endif
 #endif
 
 /* MVS linker does not support external names larger than 8 bytes */
 #if defined(__MVS__)
-#   pragma map(deflateInit_,"DEIN")
-#   pragma map(deflateInit2_,"DEIN2")
-#   pragma map(deflateEnd,"DEEND")
-#   pragma map(deflateBound,"DEBND")
-#   pragma map(inflateInit_,"ININ")
-#   pragma map(inflateInit2_,"ININ2")
-#   pragma map(inflateEnd,"INEND")
-#   pragma map(inflateSync,"INSY")
-#   pragma map(inflateSetDictionary,"INSEDI")
-#   pragma map(compressBound,"CMBND")
-#   pragma map(inflate_table,"INTABL")
-#   pragma map(inflate_fast,"INFA")
-#   pragma map(inflate_copyright,"INCOPY")
+  #pragma map(deflateInit_,"DEIN")
+  #pragma map(deflateInit2_,"DEIN2")
+  #pragma map(deflateEnd,"DEEND")
+  #pragma map(deflateBound,"DEBND")
+  #pragma map(inflateInit_,"ININ")
+  #pragma map(inflateInit2_,"ININ2")
+  #pragma map(inflateEnd,"INEND")
+  #pragma map(inflateSync,"INSY")
+  #pragma map(inflateSetDictionary,"INSEDI")
+  #pragma map(compressBound,"CMBND")
+  #pragma map(inflate_table,"INTABL")
+  #pragma map(inflate_fast,"INFA")
+  #pragma map(inflate_copyright,"INCOPY")
 #endif
 
 #endif /* ZCONF_H */

Modified: python/branches/import_unicode/Modules/zlib/zlib.3
==============================================================================
--- python/branches/import_unicode/Modules/zlib/zlib.3	(original)
+++ python/branches/import_unicode/Modules/zlib/zlib.3	Sat Jul 31 13:01:07 2010
@@ -1,4 +1,4 @@
-.TH ZLIB 3 "18 July 2005"
+.TH ZLIB 3 "19 Apr 2010"
 .SH NAME
 zlib \- compression/decompression library
 .SH SYNOPSIS
@@ -9,15 +9,15 @@
 The
 .I zlib
 library is a general purpose data compression library.
-The code is thread safe.
+The code is thread safe, assuming that the standard library functions
+used are thread safe, such as memory allocation routines.
 It provides in-memory compression and decompression functions,
 including integrity checks of the uncompressed data.
 This version of the library supports only one compression method (deflation)
-but other algorithms will be added later
-and will have the same stream interface.
+but other algorithms may be added later
+with the same stream interface.
 .LP
 Compression can be done in a single step if the buffers are large enough
-(for example if an input file is mmap'ed),
 or can be done by repeated calls of the compression function.
 In the latter case,
 the application must provide more input and/or consume the output
@@ -30,7 +30,7 @@
 .LP
 The library does not install any signal handler.
 The decoder checks the consistency of the compressed data,
-so the library should never crash even in case of corrupted input.
+so the library should never crash even in the case of corrupted input.
 .LP
 All functions of the compression library are documented in the file
 .IR zlib.h .
@@ -38,18 +38,19 @@
 in the files
 .I example.c
 and
-.IR minigzip.c .
+.IR minigzip.c,
+as well as other examples in the
+.IR examples/
+directory.
 .LP
 Changes to this version are documented in the file
 .I ChangeLog
-that accompanies the source,
-and are concerned primarily with bug fixes and portability enhancements.
+that accompanies the source.
 .LP
-A Java implementation of
 .I zlib
-is available in the Java Development Kit 1.1:
+is available in Java using the java.util.zip package:
 .IP
-http://www.javasoft.com/products/JDK/1.1/docs/api/Package-java.util.zip.html
+http://java.sun.com/developer/technicalArticles/Programming/compression/
 .LP
 A Perl interface to
 .IR zlib ,
@@ -57,7 +58,7 @@
 is available at CPAN (Comprehensive Perl Archive Network) sites,
 including:
 .IP
-http://www.cpan.org/modules/by-module/Compress/
+http://search.cpan.org/~pmqs/IO-Compress-Zlib/
 .LP
 A Python interface to
 .IR zlib ,
@@ -66,14 +67,11 @@
 .IP
 http://www.python.org/doc/lib/module-zlib.html
 .LP
-A
 .I zlib
-binding for
-.IR tcl (1),
-written by Andreas Kupries (a.kupries at westend.com),
-is availlable at:
+is built into
+.IR tcl:
 .IP
-http://www.westend.com/~kupries/doc/trf/man/man.html
+http://wiki.tcl.tk/4610
 .LP
 An experimental package to read and write files in .zip format,
 written on top of
@@ -81,40 +79,34 @@
 by Gilles Vollant (info at winimage.com),
 is available at:
 .IP
-http://www.winimage.com/zLibDll/unzip.html
+http://www.winimage.com/zLibDll/minizip.html
 and also in the
 .I contrib/minizip
 directory of the main
 .I zlib
-web site.
+source distribution.
 .SH "SEE ALSO"
 The
 .I zlib
-web site can be found at either of these locations:
+web site can be found at:
 .IP
-http://www.zlib.org
-.br
-http://www.gzip.org/zlib/
+http://zlib.net/
 .LP
 The data format used by the zlib library is described by RFC
 (Request for Comments) 1950 to 1952 in the files:
 .IP
-http://www.ietf.org/rfc/rfc1950.txt (concerning zlib format)
+http://www.ietf.org/rfc/rfc1950.txt (for the zlib header and trailer format)
 .br
-http://www.ietf.org/rfc/rfc1951.txt (concerning deflate format)
+http://www.ietf.org/rfc/rfc1951.txt (for the deflate compressed data format)
 .br
-http://www.ietf.org/rfc/rfc1952.txt (concerning gzip format)
-.LP
-These documents are also available in other formats from:
-.IP
-ftp://ftp.uu.net/graphics/png/documents/zlib/zdoc-index.html
+http://www.ietf.org/rfc/rfc1952.txt (for the gzip header and trailer format)
 .LP
-Mark Nelson (markn at ieee.org) wrote an article about
+Mark Nelson wrote an article about
 .I zlib
 for the Jan. 1997 issue of  Dr. Dobb's Journal;
 a copy of the article is available at:
 .IP
-http://dogma.net/markn/articles/zlibtool/zlibtool.htm
+http://marknelson.us/1997/01/01/zlib-engine/
 .SH "REPORTING PROBLEMS"
 Before reporting a problem,
 please check the
@@ -127,14 +119,14 @@
 .I zlib
 FAQ at:
 .IP
-http://www.gzip.org/zlib/zlib_faq.html
+http://zlib.net/zlib_faq.html
 .LP
 before asking for help.
 Send questions and/or comments to zlib at gzip.org,
 or (for the Windows DLL version) to Gilles Vollant (info at winimage.com).
 .SH AUTHORS
-Version 1.2.3
-Copyright (C) 1995-2005 Jean-loup Gailly (jloup at gzip.org)
+Version 1.2.5
+Copyright (C) 1995-2010 Jean-loup Gailly (jloup at gzip.org)
 and Mark Adler (madler at alumni.caltech.edu).
 .LP
 This software is provided "as-is,"

Modified: python/branches/import_unicode/Modules/zlib/zlib.h
==============================================================================
--- python/branches/import_unicode/Modules/zlib/zlib.h	(original)
+++ python/branches/import_unicode/Modules/zlib/zlib.h	Sat Jul 31 13:01:07 2010
@@ -1,7 +1,7 @@
 /* zlib.h -- interface of the 'zlib' general purpose compression library
-  version 1.2.3, July 18th, 2005
+  version 1.2.5, April 19th, 2010
 
-  Copyright (C) 1995-2005 Jean-loup Gailly and Mark Adler
+  Copyright (C) 1995-2010 Jean-loup Gailly and Mark Adler
 
   This software is provided 'as-is', without any express or implied
   warranty.  In no event will the authors be held liable for any damages
@@ -37,41 +37,44 @@
 extern "C" {
 #endif
 
-#define ZLIB_VERSION "1.2.3"
-#define ZLIB_VERNUM 0x1230
-
-/*
-     The 'zlib' compression library provides in-memory compression and
-  decompression functions, including integrity checks of the uncompressed
-  data.  This version of the library supports only one compression method
-  (deflation) but other algorithms will be added later and will have the same
-  stream interface.
-
-     Compression can be done in a single step if the buffers are large
-  enough (for example if an input file is mmap'ed), or can be done by
-  repeated calls of the compression function.  In the latter case, the
-  application must provide more input and/or consume the output
+#define ZLIB_VERSION "1.2.5"
+#define ZLIB_VERNUM 0x1250
+#define ZLIB_VER_MAJOR 1
+#define ZLIB_VER_MINOR 2
+#define ZLIB_VER_REVISION 5
+#define ZLIB_VER_SUBREVISION 0
+
+/*
+    The 'zlib' compression library provides in-memory compression and
+  decompression functions, including integrity checks of the uncompressed data.
+  This version of the library supports only one compression method (deflation)
+  but other algorithms will be added later and will have the same stream
+  interface.
+
+    Compression can be done in a single step if the buffers are large enough,
+  or can be done by repeated calls of the compression function.  In the latter
+  case, the application must provide more input and/or consume the output
   (providing more output space) before each call.
 
-     The compressed data format used by default by the in-memory functions is
+    The compressed data format used by default by the in-memory functions is
   the zlib format, which is a zlib wrapper documented in RFC 1950, wrapped
   around a deflate stream, which is itself documented in RFC 1951.
 
-     The library also supports reading and writing files in gzip (.gz) format
+    The library also supports reading and writing files in gzip (.gz) format
   with an interface similar to that of stdio using the functions that start
   with "gz".  The gzip format is different from the zlib format.  gzip is a
   gzip wrapper, documented in RFC 1952, wrapped around a deflate stream.
 
-     This library can optionally read and write gzip streams in memory as well.
+    This library can optionally read and write gzip streams in memory as well.
 
-     The zlib format was designed to be compact and fast for use in memory
+    The zlib format was designed to be compact and fast for use in memory
   and on communications channels.  The gzip format was designed for single-
   file compression on file systems, has a larger header than zlib to maintain
   directory information, and uses a different, slower check method than zlib.
 
-     The library does not install any signal handler. The decoder checks
-  the consistency of the compressed data, so the library should never
-  crash even in case of corrupted input.
+    The library does not install any signal handler.  The decoder checks
+  the consistency of the compressed data, so the library should never crash
+  even in case of corrupted input.
 */
 
 typedef voidpf (*alloc_func) OF((voidpf opaque, uInt items, uInt size));
@@ -126,45 +129,45 @@
 typedef gz_header FAR *gz_headerp;
 
 /*
-   The application must update next_in and avail_in when avail_in has
-   dropped to zero. It must update next_out and avail_out when avail_out
-   has dropped to zero. The application must initialize zalloc, zfree and
-   opaque before calling the init function. All other fields are set by the
-   compression library and must not be updated by the application.
-
-   The opaque value provided by the application will be passed as the first
-   parameter for calls of zalloc and zfree. This can be useful for custom
-   memory management. The compression library attaches no meaning to the
+     The application must update next_in and avail_in when avail_in has dropped
+   to zero.  It must update next_out and avail_out when avail_out has dropped
+   to zero.  The application must initialize zalloc, zfree and opaque before
+   calling the init function.  All other fields are set by the compression
+   library and must not be updated by the application.
+
+     The opaque value provided by the application will be passed as the first
+   parameter for calls of zalloc and zfree.  This can be useful for custom
+   memory management.  The compression library attaches no meaning to the
    opaque value.
 
-   zalloc must return Z_NULL if there is not enough memory for the object.
+     zalloc must return Z_NULL if there is not enough memory for the object.
    If zlib is used in a multi-threaded application, zalloc and zfree must be
    thread safe.
 
-   On 16-bit systems, the functions zalloc and zfree must be able to allocate
-   exactly 65536 bytes, but will not be required to allocate more than this
-   if the symbol MAXSEG_64K is defined (see zconf.h). WARNING: On MSDOS,
-   pointers returned by zalloc for objects of exactly 65536 bytes *must*
-   have their offset normalized to zero. The default allocation function
-   provided by this library ensures this (see zutil.c). To reduce memory
-   requirements and avoid any allocation of 64K objects, at the expense of
-   compression ratio, compile the library with -DMAX_WBITS=14 (see zconf.h).
-
-   The fields total_in and total_out can be used for statistics or
-   progress reports. After compression, total_in holds the total size of
-   the uncompressed data and may be saved for use in the decompressor
-   (particularly if the decompressor wants to decompress everything in
-   a single step).
+     On 16-bit systems, the functions zalloc and zfree must be able to allocate
+   exactly 65536 bytes, but will not be required to allocate more than this if
+   the symbol MAXSEG_64K is defined (see zconf.h).  WARNING: On MSDOS, pointers
+   returned by zalloc for objects of exactly 65536 bytes *must* have their
+   offset normalized to zero.  The default allocation function provided by this
+   library ensures this (see zutil.c).  To reduce memory requirements and avoid
+   any allocation of 64K objects, at the expense of compression ratio, compile
+   the library with -DMAX_WBITS=14 (see zconf.h).
+
+     The fields total_in and total_out can be used for statistics or progress
+   reports.  After compression, total_in holds the total size of the
+   uncompressed data and may be saved for use in the decompressor (particularly
+   if the decompressor wants to decompress everything in a single step).
 */
 
                         /* constants */
 
 #define Z_NO_FLUSH      0
-#define Z_PARTIAL_FLUSH 1 /* will be removed, use Z_SYNC_FLUSH instead */
+#define Z_PARTIAL_FLUSH 1
 #define Z_SYNC_FLUSH    2
 #define Z_FULL_FLUSH    3
 #define Z_FINISH        4
 #define Z_BLOCK         5
+#define Z_TREES         6
 /* Allowed flush values; see deflate() and inflate() below for details */
 
 #define Z_OK            0
@@ -176,8 +179,8 @@
 #define Z_MEM_ERROR    (-4)
 #define Z_BUF_ERROR    (-5)
 #define Z_VERSION_ERROR (-6)
-/* Return codes for the compression/decompression functions. Negative
- * values are errors, positive values are used for special but normal events.
+/* Return codes for the compression/decompression functions. Negative values
+ * are errors, positive values are used for special but normal events.
  */
 
 #define Z_NO_COMPRESSION         0
@@ -207,119 +210,140 @@
 #define zlib_version zlibVersion()
 /* for compatibility with versions < 1.0.2 */
 
+
                         /* basic functions */
 
 ZEXTERN const char * ZEXPORT zlibVersion OF((void));
 /* The application can compare zlibVersion and ZLIB_VERSION for consistency.
-   If the first character differs, the library code actually used is
-   not compatible with the zlib.h header file used by the application.
-   This check is automatically made by deflateInit and inflateInit.
+   If the first character differs, the library code actually used is not
+   compatible with the zlib.h header file used by the application.  This check
+   is automatically made by deflateInit and inflateInit.
  */
 
 /*
 ZEXTERN int ZEXPORT deflateInit OF((z_streamp strm, int level));
 
-     Initializes the internal stream state for compression. The fields
-   zalloc, zfree and opaque must be initialized before by the caller.
-   If zalloc and zfree are set to Z_NULL, deflateInit updates them to
-   use default allocation functions.
+     Initializes the internal stream state for compression.  The fields
+   zalloc, zfree and opaque must be initialized before by the caller.  If
+   zalloc and zfree are set to Z_NULL, deflateInit updates them to use default
+   allocation functions.
 
      The compression level must be Z_DEFAULT_COMPRESSION, or between 0 and 9:
-   1 gives best speed, 9 gives best compression, 0 gives no compression at
-   all (the input data is simply copied a block at a time).
-   Z_DEFAULT_COMPRESSION requests a default compromise between speed and
-   compression (currently equivalent to level 6).
+   1 gives best speed, 9 gives best compression, 0 gives no compression at all
+   (the input data is simply copied a block at a time).  Z_DEFAULT_COMPRESSION
+   requests a default compromise between speed and compression (currently
+   equivalent to level 6).
 
-     deflateInit returns Z_OK if success, Z_MEM_ERROR if there was not
-   enough memory, Z_STREAM_ERROR if level is not a valid compression level,
+     deflateInit returns Z_OK if success, Z_MEM_ERROR if there was not enough
+   memory, Z_STREAM_ERROR if level is not a valid compression level, or
    Z_VERSION_ERROR if the zlib library version (zlib_version) is incompatible
-   with the version assumed by the caller (ZLIB_VERSION).
-   msg is set to null if there is no error message.  deflateInit does not
-   perform any compression: this will be done by deflate().
+   with the version assumed by the caller (ZLIB_VERSION).  msg is set to null
+   if there is no error message.  deflateInit does not perform any compression:
+   this will be done by deflate().
 */
 
 
 ZEXTERN int ZEXPORT deflate OF((z_streamp strm, int flush));
 /*
     deflate compresses as much data as possible, and stops when the input
-  buffer becomes empty or the output buffer becomes full. It may introduce some
-  output latency (reading input without producing any output) except when
+  buffer becomes empty or the output buffer becomes full.  It may introduce
+  some output latency (reading input without producing any output) except when
   forced to flush.
 
-    The detailed semantics are as follows. deflate performs one or both of the
+    The detailed semantics are as follows.  deflate performs one or both of the
   following actions:
 
   - Compress more input starting at next_in and update next_in and avail_in
-    accordingly. If not all input can be processed (because there is not
+    accordingly.  If not all input can be processed (because there is not
     enough room in the output buffer), next_in and avail_in are updated and
     processing will resume at this point for the next call of deflate().
 
   - Provide more output starting at next_out and update next_out and avail_out
-    accordingly. This action is forced if the parameter flush is non zero.
+    accordingly.  This action is forced if the parameter flush is non zero.
     Forcing flush frequently degrades the compression ratio, so this parameter
-    should be set only when necessary (in interactive applications).
-    Some output may be provided even if flush is not set.
+    should be set only when necessary (in interactive applications).  Some
+    output may be provided even if flush is not set.
 
-  Before the call of deflate(), the application should ensure that at least
-  one of the actions is possible, by providing more input and/or consuming
-  more output, and updating avail_in or avail_out accordingly; avail_out
-  should never be zero before the call. The application can consume the
-  compressed output when it wants, for example when the output buffer is full
-  (avail_out == 0), or after each call of deflate(). If deflate returns Z_OK
-  and with zero avail_out, it must be called again after making room in the
-  output buffer because there might be more output pending.
+    Before the call of deflate(), the application should ensure that at least
+  one of the actions is possible, by providing more input and/or consuming more
+  output, and updating avail_in or avail_out accordingly; avail_out should
+  never be zero before the call.  The application can consume the compressed
+  output when it wants, for example when the output buffer is full (avail_out
+  == 0), or after each call of deflate().  If deflate returns Z_OK and with
+  zero avail_out, it must be called again after making room in the output
+  buffer because there might be more output pending.
 
     Normally the parameter flush is set to Z_NO_FLUSH, which allows deflate to
-  decide how much data to accumualte before producing output, in order to
+  decide how much data to accumulate before producing output, in order to
   maximize compression.
 
     If the parameter flush is set to Z_SYNC_FLUSH, all pending output is
   flushed to the output buffer and the output is aligned on a byte boundary, so
-  that the decompressor can get all input data available so far. (In particular
-  avail_in is zero after the call if enough output space has been provided
-  before the call.)  Flushing may degrade compression for some compression
-  algorithms and so it should be used only when necessary.
+  that the decompressor can get all input data available so far.  (In
+  particular avail_in is zero after the call if enough output space has been
+  provided before the call.) Flushing may degrade compression for some
+  compression algorithms and so it should be used only when necessary.  This
+  completes the current deflate block and follows it with an empty stored block
+  that is three bits plus filler bits to the next byte, followed by four bytes
+  (00 00 ff ff).
+
+    If flush is set to Z_PARTIAL_FLUSH, all pending output is flushed to the
+  output buffer, but the output is not aligned to a byte boundary.  All of the
+  input data so far will be available to the decompressor, as for Z_SYNC_FLUSH.
+  This completes the current deflate block and follows it with an empty fixed
+  codes block that is 10 bits long.  This assures that enough bytes are output
+  in order for the decompressor to finish the block before the empty fixed code
+  block.
+
+    If flush is set to Z_BLOCK, a deflate block is completed and emitted, as
+  for Z_SYNC_FLUSH, but the output is not aligned on a byte boundary, and up to
+  seven bits of the current block are held to be written as the next byte after
+  the next deflate block is completed.  In this case, the decompressor may not
+  be provided enough bits at this point in order to complete decompression of
+  the data provided so far to the compressor.  It may need to wait for the next
+  block to be emitted.  This is for advanced applications that need to control
+  the emission of deflate blocks.
 
     If flush is set to Z_FULL_FLUSH, all output is flushed as with
   Z_SYNC_FLUSH, and the compression state is reset so that decompression can
   restart from this point if previous compressed data has been damaged or if
-  random access is desired. Using Z_FULL_FLUSH too often can seriously degrade
+  random access is desired.  Using Z_FULL_FLUSH too often can seriously degrade
   compression.
 
     If deflate returns with avail_out == 0, this function must be called again
   with the same value of the flush parameter and more output space (updated
   avail_out), until the flush is complete (deflate returns with non-zero
-  avail_out). In the case of a Z_FULL_FLUSH or Z_SYNC_FLUSH, make sure that
+  avail_out).  In the case of a Z_FULL_FLUSH or Z_SYNC_FLUSH, make sure that
   avail_out is greater than six to avoid repeated flush markers due to
   avail_out == 0 on return.
 
     If the parameter flush is set to Z_FINISH, pending input is processed,
-  pending output is flushed and deflate returns with Z_STREAM_END if there
-  was enough output space; if deflate returns with Z_OK, this function must be
+  pending output is flushed and deflate returns with Z_STREAM_END if there was
+  enough output space; if deflate returns with Z_OK, this function must be
   called again with Z_FINISH and more output space (updated avail_out) but no
-  more input data, until it returns with Z_STREAM_END or an error. After
-  deflate has returned Z_STREAM_END, the only possible operations on the
-  stream are deflateReset or deflateEnd.
+  more input data, until it returns with Z_STREAM_END or an error.  After
+  deflate has returned Z_STREAM_END, the only possible operations on the stream
+  are deflateReset or deflateEnd.
 
     Z_FINISH can be used immediately after deflateInit if all the compression
-  is to be done in a single step. In this case, avail_out must be at least
-  the value returned by deflateBound (see below). If deflate does not return
+  is to be done in a single step.  In this case, avail_out must be at least the
+  value returned by deflateBound (see below).  If deflate does not return
   Z_STREAM_END, then it must be called again as described above.
 
     deflate() sets strm->adler to the adler32 checksum of all input read
   so far (that is, total_in bytes).
 
     deflate() may update strm->data_type if it can make a good guess about
-  the input data type (Z_BINARY or Z_TEXT). In doubt, the data is considered
-  binary. This field is only for information purposes and does not affect
-  the compression algorithm in any manner.
+  the input data type (Z_BINARY or Z_TEXT).  In doubt, the data is considered
+  binary.  This field is only for information purposes and does not affect the
+  compression algorithm in any manner.
 
     deflate() returns Z_OK if some progress has been made (more input
   processed or more output produced), Z_STREAM_END if all input has been
   consumed and all output has been produced (only when flush is set to
   Z_FINISH), Z_STREAM_ERROR if the stream state was inconsistent (for example
-  if next_in or next_out was NULL), Z_BUF_ERROR if no progress is possible
-  (for example avail_in or avail_out was zero). Note that Z_BUF_ERROR is not
+  if next_in or next_out was Z_NULL), Z_BUF_ERROR if no progress is possible
+  (for example avail_in or avail_out was zero).  Note that Z_BUF_ERROR is not
   fatal, and deflate() can be called again with more input and more output
   space to continue compressing.
 */
@@ -328,13 +352,13 @@
 ZEXTERN int ZEXPORT deflateEnd OF((z_streamp strm));
 /*
      All dynamically allocated data structures for this stream are freed.
-   This function discards any unprocessed input and does not flush any
-   pending output.
+   This function discards any unprocessed input and does not flush any pending
+   output.
 
      deflateEnd returns Z_OK if success, Z_STREAM_ERROR if the
    stream state was inconsistent, Z_DATA_ERROR if the stream was freed
-   prematurely (some input or output was discarded). In the error case,
-   msg may be set but then points to a static string (which must not be
+   prematurely (some input or output was discarded).  In the error case, msg
+   may be set but then points to a static string (which must not be
    deallocated).
 */
 
@@ -342,10 +366,10 @@
 /*
 ZEXTERN int ZEXPORT inflateInit OF((z_streamp strm));
 
-     Initializes the internal stream state for decompression. The fields
+     Initializes the internal stream state for decompression.  The fields
    next_in, avail_in, zalloc, zfree and opaque must be initialized before by
-   the caller. If next_in is not Z_NULL and avail_in is large enough (the exact
-   value depends on the compression method), inflateInit determines the
+   the caller.  If next_in is not Z_NULL and avail_in is large enough (the
+   exact value depends on the compression method), inflateInit determines the
    compression method from the zlib header and allocates all data structures
    accordingly; otherwise the allocation will be deferred to the first call of
    inflate.  If zalloc and zfree are set to Z_NULL, inflateInit updates them to
@@ -353,95 +377,108 @@
 
      inflateInit returns Z_OK if success, Z_MEM_ERROR if there was not enough
    memory, Z_VERSION_ERROR if the zlib library version is incompatible with the
-   version assumed by the caller.  msg is set to null if there is no error
-   message. inflateInit does not perform any decompression apart from reading
-   the zlib header if present: this will be done by inflate().  (So next_in and
-   avail_in may be modified, but next_out and avail_out are unchanged.)
+   version assumed by the caller, or Z_STREAM_ERROR if the parameters are
+   invalid, such as a null pointer to the structure.  msg is set to null if
+   there is no error message.  inflateInit does not perform any decompression
+   apart from possibly reading the zlib header if present: actual decompression
+   will be done by inflate().  (So next_in and avail_in may be modified, but
+   next_out and avail_out are unused and unchanged.) The current implementation
+   of inflateInit() does not process any header information -- that is deferred
+   until inflate() is called.
 */
 
 
 ZEXTERN int ZEXPORT inflate OF((z_streamp strm, int flush));
 /*
     inflate decompresses as much data as possible, and stops when the input
-  buffer becomes empty or the output buffer becomes full. It may introduce
+  buffer becomes empty or the output buffer becomes full.  It may introduce
   some output latency (reading input without producing any output) except when
   forced to flush.
 
-  The detailed semantics are as follows. inflate performs one or both of the
+  The detailed semantics are as follows.  inflate performs one or both of the
   following actions:
 
   - Decompress more input starting at next_in and update next_in and avail_in
-    accordingly. If not all input can be processed (because there is not
-    enough room in the output buffer), next_in is updated and processing
-    will resume at this point for the next call of inflate().
+    accordingly.  If not all input can be processed (because there is not
+    enough room in the output buffer), next_in is updated and processing will
+    resume at this point for the next call of inflate().
 
   - Provide more output starting at next_out and update next_out and avail_out
-    accordingly.  inflate() provides as much output as possible, until there
-    is no more input data or no more space in the output buffer (see below
-    about the flush parameter).
-
-  Before the call of inflate(), the application should ensure that at least
-  one of the actions is possible, by providing more input and/or consuming
-  more output, and updating the next_* and avail_* values accordingly.
-  The application can consume the uncompressed output when it wants, for
-  example when the output buffer is full (avail_out == 0), or after each
-  call of inflate(). If inflate returns Z_OK and with zero avail_out, it
-  must be called again after making room in the output buffer because there
-  might be more output pending.
-
-    The flush parameter of inflate() can be Z_NO_FLUSH, Z_SYNC_FLUSH,
-  Z_FINISH, or Z_BLOCK. Z_SYNC_FLUSH requests that inflate() flush as much
-  output as possible to the output buffer. Z_BLOCK requests that inflate() stop
-  if and when it gets to the next deflate block boundary. When decoding the
-  zlib or gzip format, this will cause inflate() to return immediately after
-  the header and before the first block. When doing a raw inflate, inflate()
-  will go ahead and process the first block, and will return when it gets to
-  the end of that block, or when it runs out of data.
+    accordingly.  inflate() provides as much output as possible, until there is
+    no more input data or no more space in the output buffer (see below about
+    the flush parameter).
+
+    Before the call of inflate(), the application should ensure that at least
+  one of the actions is possible, by providing more input and/or consuming more
+  output, and updating the next_* and avail_* values accordingly.  The
+  application can consume the uncompressed output when it wants, for example
+  when the output buffer is full (avail_out == 0), or after each call of
+  inflate().  If inflate returns Z_OK and with zero avail_out, it must be
+  called again after making room in the output buffer because there might be
+  more output pending.
+
+    The flush parameter of inflate() can be Z_NO_FLUSH, Z_SYNC_FLUSH, Z_FINISH,
+  Z_BLOCK, or Z_TREES.  Z_SYNC_FLUSH requests that inflate() flush as much
+  output as possible to the output buffer.  Z_BLOCK requests that inflate()
+  stop if and when it gets to the next deflate block boundary.  When decoding
+  the zlib or gzip format, this will cause inflate() to return immediately
+  after the header and before the first block.  When doing a raw inflate,
+  inflate() will go ahead and process the first block, and will return when it
+  gets to the end of that block, or when it runs out of data.
 
     The Z_BLOCK option assists in appending to or combining deflate streams.
   Also to assist in this, on return inflate() will set strm->data_type to the
-  number of unused bits in the last byte taken from strm->next_in, plus 64
-  if inflate() is currently decoding the last block in the deflate stream,
-  plus 128 if inflate() returned immediately after decoding an end-of-block
-  code or decoding the complete header up to just before the first byte of the
-  deflate stream. The end-of-block will not be indicated until all of the
-  uncompressed data from that block has been written to strm->next_out.  The
-  number of unused bits may in general be greater than seven, except when
-  bit 7 of data_type is set, in which case the number of unused bits will be
-  less than eight.
+  number of unused bits in the last byte taken from strm->next_in, plus 64 if
+  inflate() is currently decoding the last block in the deflate stream, plus
+  128 if inflate() returned immediately after decoding an end-of-block code or
+  decoding the complete header up to just before the first byte of the deflate
+  stream.  The end-of-block will not be indicated until all of the uncompressed
+  data from that block has been written to strm->next_out.  The number of
+  unused bits may in general be greater than seven, except when bit 7 of
+  data_type is set, in which case the number of unused bits will be less than
+  eight.  data_type is set as noted here every time inflate() returns for all
+  flush options, and so can be used to determine the amount of currently
+  consumed input in bits.
+
+    The Z_TREES option behaves as Z_BLOCK does, but it also returns when the
+  end of each deflate block header is reached, before any actual data in that
+  block is decoded.  This allows the caller to determine the length of the
+  deflate block header for later use in random access within a deflate block.
+  256 is added to the value of strm->data_type when inflate() returns
+  immediately after reaching the end of the deflate block header.
 
     inflate() should normally be called until it returns Z_STREAM_END or an
-  error. However if all decompression is to be performed in a single step
-  (a single call of inflate), the parameter flush should be set to
-  Z_FINISH. In this case all pending input is processed and all pending
-  output is flushed; avail_out must be large enough to hold all the
-  uncompressed data. (The size of the uncompressed data may have been saved
-  by the compressor for this purpose.) The next operation on this stream must
-  be inflateEnd to deallocate the decompression state. The use of Z_FINISH
-  is never required, but can be used to inform inflate that a faster approach
-  may be used for the single inflate() call.
+  error.  However if all decompression is to be performed in a single step (a
+  single call of inflate), the parameter flush should be set to Z_FINISH.  In
+  this case all pending input is processed and all pending output is flushed;
+  avail_out must be large enough to hold all the uncompressed data.  (The size
+  of the uncompressed data may have been saved by the compressor for this
+  purpose.) The next operation on this stream must be inflateEnd to deallocate
+  the decompression state.  The use of Z_FINISH is never required, but can be
+  used to inform inflate that a faster approach may be used for the single
+  inflate() call.
 
      In this implementation, inflate() always flushes as much output as
   possible to the output buffer, and always uses the faster approach on the
-  first call. So the only effect of the flush parameter in this implementation
+  first call.  So the only effect of the flush parameter in this implementation
   is on the return value of inflate(), as noted below, or when it returns early
-  because Z_BLOCK is used.
+  because Z_BLOCK or Z_TREES is used.
 
      If a preset dictionary is needed after this call (see inflateSetDictionary
   below), inflate sets strm->adler to the adler32 checksum of the dictionary
   chosen by the compressor and returns Z_NEED_DICT; otherwise it sets
   strm->adler to the adler32 checksum of all output produced so far (that is,
   total_out bytes) and returns Z_OK, Z_STREAM_END or an error code as described
-  below. At the end of the stream, inflate() checks that its computed adler32
+  below.  At the end of the stream, inflate() checks that its computed adler32
   checksum is equal to that saved by the compressor and returns Z_STREAM_END
   only if the checksum is correct.
 
-    inflate() will decompress and check either zlib-wrapped or gzip-wrapped
-  deflate data.  The header type is detected automatically.  Any information
-  contained in the gzip header is not retained, so applications that need that
-  information should instead use raw inflate, see inflateInit2() below, or
-  inflateBack() and perform their own processing of the gzip header and
-  trailer.
+    inflate() can decompress and check either zlib-wrapped or gzip-wrapped
+  deflate data.  The header type is detected automatically, if requested when
+  initializing with inflateInit2().  Any information contained in the gzip
+  header is not retained, so applications that need that information should
+  instead use raw inflate, see inflateInit2() below, or inflateBack() and
+  perform their own processing of the gzip header and trailer.
 
     inflate() returns Z_OK if some progress has been made (more input processed
   or more output produced), Z_STREAM_END if the end of the compressed data has
@@ -449,27 +486,28 @@
   preset dictionary is needed at this point, Z_DATA_ERROR if the input data was
   corrupted (input stream not conforming to the zlib format or incorrect check
   value), Z_STREAM_ERROR if the stream structure was inconsistent (for example
-  if next_in or next_out was NULL), Z_MEM_ERROR if there was not enough memory,
+  next_in or next_out was Z_NULL), Z_MEM_ERROR if there was not enough memory,
   Z_BUF_ERROR if no progress is possible or if there was not enough room in the
-  output buffer when Z_FINISH is used. Note that Z_BUF_ERROR is not fatal, and
+  output buffer when Z_FINISH is used.  Note that Z_BUF_ERROR is not fatal, and
   inflate() can be called again with more input and more output space to
-  continue decompressing. If Z_DATA_ERROR is returned, the application may then
-  call inflateSync() to look for a good compression block if a partial recovery
-  of the data is desired.
+  continue decompressing.  If Z_DATA_ERROR is returned, the application may
+  then call inflateSync() to look for a good compression block if a partial
+  recovery of the data is desired.
 */
 
 
 ZEXTERN int ZEXPORT inflateEnd OF((z_streamp strm));
 /*
      All dynamically allocated data structures for this stream are freed.
-   This function discards any unprocessed input and does not flush any
-   pending output.
+   This function discards any unprocessed input and does not flush any pending
+   output.
 
      inflateEnd returns Z_OK if success, Z_STREAM_ERROR if the stream state
-   was inconsistent. In the error case, msg may be set but then points to a
+   was inconsistent.  In the error case, msg may be set but then points to a
    static string (which must not be deallocated).
 */
 
+
                         /* Advanced functions */
 
 /*
@@ -484,55 +522,57 @@
                                      int  memLevel,
                                      int  strategy));
 
-     This is another version of deflateInit with more compression options. The
-   fields next_in, zalloc, zfree and opaque must be initialized before by
-   the caller.
+     This is another version of deflateInit with more compression options.  The
+   fields next_in, zalloc, zfree and opaque must be initialized before by the
+   caller.
 
-     The method parameter is the compression method. It must be Z_DEFLATED in
+     The method parameter is the compression method.  It must be Z_DEFLATED in
    this version of the library.
 
      The windowBits parameter is the base two logarithm of the window size
-   (the size of the history buffer). It should be in the range 8..15 for this
-   version of the library. Larger values of this parameter result in better
-   compression at the expense of memory usage. The default value is 15 if
+   (the size of the history buffer).  It should be in the range 8..15 for this
+   version of the library.  Larger values of this parameter result in better
+   compression at the expense of memory usage.  The default value is 15 if
    deflateInit is used instead.
 
-     windowBits can also be -8..-15 for raw deflate. In this case, -windowBits
-   determines the window size. deflate() will then generate raw deflate data
+     windowBits can also be -8..-15 for raw deflate.  In this case, -windowBits
+   determines the window size.  deflate() will then generate raw deflate data
    with no zlib header or trailer, and will not compute an adler32 check value.
 
-     windowBits can also be greater than 15 for optional gzip encoding. Add
+     windowBits can also be greater than 15 for optional gzip encoding.  Add
    16 to windowBits to write a simple gzip header and trailer around the
-   compressed data instead of a zlib wrapper. The gzip header will have no
-   file name, no extra data, no comment, no modification time (set to zero),
-   no header crc, and the operating system will be set to 255 (unknown).  If a
+   compressed data instead of a zlib wrapper.  The gzip header will have no
+   file name, no extra data, no comment, no modification time (set to zero), no
+   header crc, and the operating system will be set to 255 (unknown).  If a
    gzip stream is being written, strm->adler is a crc32 instead of an adler32.
 
      The memLevel parameter specifies how much memory should be allocated
-   for the internal compression state. memLevel=1 uses minimum memory but
-   is slow and reduces compression ratio; memLevel=9 uses maximum memory
-   for optimal speed. The default value is 8. See zconf.h for total memory
-   usage as a function of windowBits and memLevel.
+   for the internal compression state.  memLevel=1 uses minimum memory but is
+   slow and reduces compression ratio; memLevel=9 uses maximum memory for
+   optimal speed.  The default value is 8.  See zconf.h for total memory usage
+   as a function of windowBits and memLevel.
 
-     The strategy parameter is used to tune the compression algorithm. Use the
+     The strategy parameter is used to tune the compression algorithm.  Use the
    value Z_DEFAULT_STRATEGY for normal data, Z_FILTERED for data produced by a
    filter (or predictor), Z_HUFFMAN_ONLY to force Huffman encoding only (no
    string match), or Z_RLE to limit match distances to one (run-length
-   encoding). Filtered data consists mostly of small values with a somewhat
-   random distribution. In this case, the compression algorithm is tuned to
-   compress them better. The effect of Z_FILTERED is to force more Huffman
+   encoding).  Filtered data consists mostly of small values with a somewhat
+   random distribution.  In this case, the compression algorithm is tuned to
+   compress them better.  The effect of Z_FILTERED is to force more Huffman
    coding and less string matching; it is somewhat intermediate between
-   Z_DEFAULT and Z_HUFFMAN_ONLY. Z_RLE is designed to be almost as fast as
-   Z_HUFFMAN_ONLY, but give better compression for PNG image data. The strategy
-   parameter only affects the compression ratio but not the correctness of the
-   compressed output even if it is not set appropriately.  Z_FIXED prevents the
-   use of dynamic Huffman codes, allowing for a simpler decoder for special
-   applications.
-
-      deflateInit2 returns Z_OK if success, Z_MEM_ERROR if there was not enough
-   memory, Z_STREAM_ERROR if a parameter is invalid (such as an invalid
-   method). msg is set to null if there is no error message.  deflateInit2 does
-   not perform any compression: this will be done by deflate().
+   Z_DEFAULT_STRATEGY and Z_HUFFMAN_ONLY.  Z_RLE is designed to be almost as
+   fast as Z_HUFFMAN_ONLY, but give better compression for PNG image data.  The
+   strategy parameter only affects the compression ratio but not the
+   correctness of the compressed output even if it is not set appropriately.
+   Z_FIXED prevents the use of dynamic Huffman codes, allowing for a simpler
+   decoder for special applications.
+
+     deflateInit2 returns Z_OK if success, Z_MEM_ERROR if there was not enough
+   memory, Z_STREAM_ERROR if any parameter is invalid (such as an invalid
+   method), or Z_VERSION_ERROR if the zlib library version (zlib_version) is
+   incompatible with the version assumed by the caller (ZLIB_VERSION).  msg is
+   set to null if there is no error message.  deflateInit2 does not perform any
+   compression: this will be done by deflate().
 */
 
 ZEXTERN int ZEXPORT deflateSetDictionary OF((z_streamp strm,
@@ -540,37 +580,37 @@
                                              uInt  dictLength));
 /*
      Initializes the compression dictionary from the given byte sequence
-   without producing any compressed output. This function must be called
-   immediately after deflateInit, deflateInit2 or deflateReset, before any
-   call of deflate. The compressor and decompressor must use exactly the same
+   without producing any compressed output.  This function must be called
+   immediately after deflateInit, deflateInit2 or deflateReset, before any call
+   of deflate.  The compressor and decompressor must use exactly the same
    dictionary (see inflateSetDictionary).
 
      The dictionary should consist of strings (byte sequences) that are likely
    to be encountered later in the data to be compressed, with the most commonly
-   used strings preferably put towards the end of the dictionary. Using a
+   used strings preferably put towards the end of the dictionary.  Using a
    dictionary is most useful when the data to be compressed is short and can be
    predicted with good accuracy; the data can then be compressed better than
    with the default empty dictionary.
 
      Depending on the size of the compression data structures selected by
    deflateInit or deflateInit2, a part of the dictionary may in effect be
-   discarded, for example if the dictionary is larger than the window size in
-   deflate or deflate2. Thus the strings most likely to be useful should be
-   put at the end of the dictionary, not at the front. In addition, the
-   current implementation of deflate will use at most the window size minus
-   262 bytes of the provided dictionary.
+   discarded, for example if the dictionary is larger than the window size
+   provided in deflateInit or deflateInit2.  Thus the strings most likely to be
+   useful should be put at the end of the dictionary, not at the front.  In
+   addition, the current implementation of deflate will use at most the window
+   size minus 262 bytes of the provided dictionary.
 
      Upon return of this function, strm->adler is set to the adler32 value
    of the dictionary; the decompressor may later use this value to determine
-   which dictionary has been used by the compressor. (The adler32 value
+   which dictionary has been used by the compressor.  (The adler32 value
    applies to the whole dictionary even if only a subset of the dictionary is
    actually used by the compressor.) If a raw deflate was requested, then the
    adler32 value is not computed and strm->adler is not set.
 
      deflateSetDictionary returns Z_OK if success, or Z_STREAM_ERROR if a
-   parameter is invalid (such as NULL dictionary) or the stream state is
+   parameter is invalid (e.g.  dictionary being Z_NULL) or the stream state is
    inconsistent (for example if deflate has already been called for this stream
-   or if the compression method is bsort). deflateSetDictionary does not
+   or if the compression method is bsort).  deflateSetDictionary does not
    perform any compression: this will be done by deflate().
 */
 
@@ -581,26 +621,26 @@
 
      This function can be useful when several compression strategies will be
    tried, for example when there are several ways of pre-processing the input
-   data with a filter. The streams that will be discarded should then be freed
+   data with a filter.  The streams that will be discarded should then be freed
    by calling deflateEnd.  Note that deflateCopy duplicates the internal
-   compression state which can be quite large, so this strategy is slow and
-   can consume lots of memory.
+   compression state which can be quite large, so this strategy is slow and can
+   consume lots of memory.
 
      deflateCopy returns Z_OK if success, Z_MEM_ERROR if there was not
    enough memory, Z_STREAM_ERROR if the source stream state was inconsistent
-   (such as zalloc being NULL). msg is left unchanged in both source and
+   (such as zalloc being Z_NULL).  msg is left unchanged in both source and
    destination.
 */
 
 ZEXTERN int ZEXPORT deflateReset OF((z_streamp strm));
 /*
      This function is equivalent to deflateEnd followed by deflateInit,
-   but does not free and reallocate all the internal compression state.
-   The stream will keep the same compression level and any other attributes
-   that may have been set by deflateInit2.
+   but does not free and reallocate all the internal compression state.  The
+   stream will keep the same compression level and any other attributes that
+   may have been set by deflateInit2.
 
-      deflateReset returns Z_OK if success, or Z_STREAM_ERROR if the source
-   stream state was inconsistent (such as zalloc or state being NULL).
+     deflateReset returns Z_OK if success, or Z_STREAM_ERROR if the source
+   stream state was inconsistent (such as zalloc or state being Z_NULL).
 */
 
 ZEXTERN int ZEXPORT deflateParams OF((z_streamp strm,
@@ -610,18 +650,18 @@
      Dynamically update the compression level and compression strategy.  The
    interpretation of level and strategy is as in deflateInit2.  This can be
    used to switch between compression and straight copy of the input data, or
-   to switch to a different kind of input data requiring a different
-   strategy. If the compression level is changed, the input available so far
-   is compressed with the old level (and may be flushed); the new level will
-   take effect only at the next call of deflate().
+   to switch to a different kind of input data requiring a different strategy.
+   If the compression level is changed, the input available so far is
+   compressed with the old level (and may be flushed); the new level will take
+   effect only at the next call of deflate().
 
      Before the call of deflateParams, the stream state must be set as for
-   a call of deflate(), since the currently available input may have to
-   be compressed and flushed. In particular, strm->avail_out must be non-zero.
+   a call of deflate(), since the currently available input may have to be
+   compressed and flushed.  In particular, strm->avail_out must be non-zero.
 
      deflateParams returns Z_OK if success, Z_STREAM_ERROR if the source
-   stream state was inconsistent or if a parameter was invalid, Z_BUF_ERROR
-   if strm->avail_out was zero.
+   stream state was inconsistent or if a parameter was invalid, Z_BUF_ERROR if
+   strm->avail_out was zero.
 */
 
 ZEXTERN int ZEXPORT deflateTune OF((z_streamp strm,
@@ -645,9 +685,10 @@
                                        uLong sourceLen));
 /*
      deflateBound() returns an upper bound on the compressed size after
-   deflation of sourceLen bytes.  It must be called after deflateInit()
-   or deflateInit2().  This would be used to allocate an output buffer
-   for deflation in a single pass, and so would be called before deflate().
+   deflation of sourceLen bytes.  It must be called after deflateInit() or
+   deflateInit2(), and after deflateSetHeader(), if used.  This would be used
+   to allocate an output buffer for deflation in a single pass, and so would be
+   called before deflate().
 */
 
 ZEXTERN int ZEXPORT deflatePrime OF((z_streamp strm,
@@ -655,21 +696,21 @@
                                      int value));
 /*
      deflatePrime() inserts bits in the deflate output stream.  The intent
-  is that this function is used to start off the deflate output with the
-  bits leftover from a previous deflate stream when appending to it.  As such,
-  this function can only be used for raw deflate, and must be used before the
-  first deflate() call after a deflateInit2() or deflateReset().  bits must be
-  less than or equal to 16, and that many of the least significant bits of
-  value will be inserted in the output.
+   is that this function is used to start off the deflate output with the bits
+   leftover from a previous deflate stream when appending to it.  As such, this
+   function can only be used for raw deflate, and must be used before the first
+   deflate() call after a deflateInit2() or deflateReset().  bits must be less
+   than or equal to 16, and that many of the least significant bits of value
+   will be inserted in the output.
 
-      deflatePrime returns Z_OK if success, or Z_STREAM_ERROR if the source
+     deflatePrime returns Z_OK if success, or Z_STREAM_ERROR if the source
    stream state was inconsistent.
 */
 
 ZEXTERN int ZEXPORT deflateSetHeader OF((z_streamp strm,
                                          gz_headerp head));
 /*
-      deflateSetHeader() provides gzip header information for when a gzip
+     deflateSetHeader() provides gzip header information for when a gzip
    stream is requested by deflateInit2().  deflateSetHeader() may be called
    after deflateInit2() or deflateReset() and before the first call of
    deflate().  The text, time, os, extra field, name, and comment information
@@ -682,11 +723,11 @@
    1.3.x) do not support header crc's, and will report that it is a "multi-part
    gzip file" and give up.
 
-      If deflateSetHeader is not used, the default gzip header has text false,
+     If deflateSetHeader is not used, the default gzip header has text false,
    the time set to zero, and os set to 255, with no extra, name, or comment
    fields.  The gzip header is returned to the default state by deflateReset().
 
-      deflateSetHeader returns Z_OK if success, or Z_STREAM_ERROR if the source
+     deflateSetHeader returns Z_OK if success, or Z_STREAM_ERROR if the source
    stream state was inconsistent.
 */
 
@@ -694,43 +735,50 @@
 ZEXTERN int ZEXPORT inflateInit2 OF((z_streamp strm,
                                      int  windowBits));
 
-     This is another version of inflateInit with an extra parameter. The
+     This is another version of inflateInit with an extra parameter.  The
    fields next_in, avail_in, zalloc, zfree and opaque must be initialized
    before by the caller.
 
      The windowBits parameter is the base two logarithm of the maximum window
    size (the size of the history buffer).  It should be in the range 8..15 for
-   this version of the library. The default value is 15 if inflateInit is used
-   instead. windowBits must be greater than or equal to the windowBits value
+   this version of the library.  The default value is 15 if inflateInit is used
+   instead.  windowBits must be greater than or equal to the windowBits value
    provided to deflateInit2() while compressing, or it must be equal to 15 if
-   deflateInit2() was not used. If a compressed stream with a larger window
+   deflateInit2() was not used.  If a compressed stream with a larger window
    size is given as input, inflate() will return with the error code
    Z_DATA_ERROR instead of trying to allocate a larger window.
 
-     windowBits can also be -8..-15 for raw inflate. In this case, -windowBits
-   determines the window size. inflate() will then process raw deflate data,
+     windowBits can also be zero to request that inflate use the window size in
+   the zlib header of the compressed stream.
+
+     windowBits can also be -8..-15 for raw inflate.  In this case, -windowBits
+   determines the window size.  inflate() will then process raw deflate data,
    not looking for a zlib or gzip header, not generating a check value, and not
-   looking for any check values for comparison at the end of the stream. This
+   looking for any check values for comparison at the end of the stream.  This
    is for use with other formats that use the deflate compressed data format
-   such as zip.  Those formats provide their own check values. If a custom
+   such as zip.  Those formats provide their own check values.  If a custom
    format is developed using the raw deflate format for compressed data, it is
    recommended that a check value such as an adler32 or a crc32 be applied to
    the uncompressed data as is done in the zlib, gzip, and zip formats.  For
-   most applications, the zlib format should be used as is. Note that comments
+   most applications, the zlib format should be used as is.  Note that comments
    above on the use in deflateInit2() applies to the magnitude of windowBits.
 
-     windowBits can also be greater than 15 for optional gzip decoding. Add
+     windowBits can also be greater than 15 for optional gzip decoding.  Add
    32 to windowBits to enable zlib and gzip decoding with automatic header
    detection, or add 16 to decode only the gzip format (the zlib format will
-   return a Z_DATA_ERROR).  If a gzip stream is being decoded, strm->adler is
-   a crc32 instead of an adler32.
+   return a Z_DATA_ERROR).  If a gzip stream is being decoded, strm->adler is a
+   crc32 instead of an adler32.
 
      inflateInit2 returns Z_OK if success, Z_MEM_ERROR if there was not enough
-   memory, Z_STREAM_ERROR if a parameter is invalid (such as a null strm). msg
-   is set to null if there is no error message.  inflateInit2 does not perform
-   any decompression apart from reading the zlib header if present: this will
-   be done by inflate(). (So next_in and avail_in may be modified, but next_out
-   and avail_out are unchanged.)
+   memory, Z_VERSION_ERROR if the zlib library version is incompatible with the
+   version assumed by the caller, or Z_STREAM_ERROR if the parameters are
+   invalid, such as a null pointer to the structure.  msg is set to null if
+   there is no error message.  inflateInit2 does not perform any decompression
+   apart from possibly reading the zlib header if present: actual decompression
+   will be done by inflate().  (So next_in and avail_in may be modified, but
+   next_out and avail_out are unused and unchanged.) The current implementation
+   of inflateInit2() does not process any header information -- that is
+   deferred until inflate() is called.
 */
 
 ZEXTERN int ZEXPORT inflateSetDictionary OF((z_streamp strm,
@@ -738,8 +786,8 @@
                                              uInt  dictLength));
 /*
      Initializes the decompression dictionary from the given uncompressed byte
-   sequence. This function must be called immediately after a call of inflate,
-   if that call returned Z_NEED_DICT. The dictionary chosen by the compressor
+   sequence.  This function must be called immediately after a call of inflate,
+   if that call returned Z_NEED_DICT.  The dictionary chosen by the compressor
    can be determined from the adler32 value returned by that call of inflate.
    The compressor and decompressor must use exactly the same dictionary (see
    deflateSetDictionary).  For raw inflate, this function can be called
@@ -748,26 +796,26 @@
    dictionary that was used for compression is provided.
 
      inflateSetDictionary returns Z_OK if success, Z_STREAM_ERROR if a
-   parameter is invalid (such as NULL dictionary) or the stream state is
+   parameter is invalid (e.g.  dictionary being Z_NULL) or the stream state is
    inconsistent, Z_DATA_ERROR if the given dictionary doesn't match the
-   expected one (incorrect adler32 value). inflateSetDictionary does not
+   expected one (incorrect adler32 value).  inflateSetDictionary does not
    perform any decompression: this will be done by subsequent calls of
    inflate().
 */
 
 ZEXTERN int ZEXPORT inflateSync OF((z_streamp strm));
 /*
-    Skips invalid compressed data until a full flush point (see above the
-  description of deflate with Z_FULL_FLUSH) can be found, or until all
-  available input is skipped. No output is provided.
-
-    inflateSync returns Z_OK if a full flush point has been found, Z_BUF_ERROR
-  if no more input was provided, Z_DATA_ERROR if no flush point has been found,
-  or Z_STREAM_ERROR if the stream structure was inconsistent. In the success
-  case, the application may save the current current value of total_in which
-  indicates where valid compressed data was found. In the error case, the
-  application may repeatedly call inflateSync, providing more input each time,
-  until success or end of the input data.
+     Skips invalid compressed data until a full flush point (see above the
+   description of deflate with Z_FULL_FLUSH) can be found, or until all
+   available input is skipped.  No output is provided.
+
+     inflateSync returns Z_OK if a full flush point has been found, Z_BUF_ERROR
+   if no more input was provided, Z_DATA_ERROR if no flush point has been
+   found, or Z_STREAM_ERROR if the stream structure was inconsistent.  In the
+   success case, the application may save the current current value of total_in
+   which indicates where valid compressed data was found.  In the error case,
+   the application may repeatedly call inflateSync, providing more input each
+   time, until success or end of the input data.
 */
 
 ZEXTERN int ZEXPORT inflateCopy OF((z_streamp dest,
@@ -782,18 +830,30 @@
 
      inflateCopy returns Z_OK if success, Z_MEM_ERROR if there was not
    enough memory, Z_STREAM_ERROR if the source stream state was inconsistent
-   (such as zalloc being NULL). msg is left unchanged in both source and
+   (such as zalloc being Z_NULL).  msg is left unchanged in both source and
    destination.
 */
 
 ZEXTERN int ZEXPORT inflateReset OF((z_streamp strm));
 /*
      This function is equivalent to inflateEnd followed by inflateInit,
-   but does not free and reallocate all the internal decompression state.
-   The stream will keep attributes that may have been set by inflateInit2.
+   but does not free and reallocate all the internal decompression state.  The
+   stream will keep attributes that may have been set by inflateInit2.
 
-      inflateReset returns Z_OK if success, or Z_STREAM_ERROR if the source
-   stream state was inconsistent (such as zalloc or state being NULL).
+     inflateReset returns Z_OK if success, or Z_STREAM_ERROR if the source
+   stream state was inconsistent (such as zalloc or state being Z_NULL).
+*/
+
+ZEXTERN int ZEXPORT inflateReset2 OF((z_streamp strm,
+                                      int windowBits));
+/*
+     This function is the same as inflateReset, but it also permits changing
+   the wrap and window size requests.  The windowBits parameter is interpreted
+   the same as it is for inflateInit2.
+
+     inflateReset2 returns Z_OK if success, or Z_STREAM_ERROR if the source
+   stream state was inconsistent (such as zalloc or state being Z_NULL), or if
+   the windowBits parameter is invalid.
 */
 
 ZEXTERN int ZEXPORT inflatePrime OF((z_streamp strm,
@@ -801,54 +861,87 @@
                                      int value));
 /*
      This function inserts bits in the inflate input stream.  The intent is
-  that this function is used to start inflating at a bit position in the
-  middle of a byte.  The provided bits will be used before any bytes are used
-  from next_in.  This function should only be used with raw inflate, and
-  should be used before the first inflate() call after inflateInit2() or
-  inflateReset().  bits must be less than or equal to 16, and that many of the
-  least significant bits of value will be inserted in the input.
+   that this function is used to start inflating at a bit position in the
+   middle of a byte.  The provided bits will be used before any bytes are used
+   from next_in.  This function should only be used with raw inflate, and
+   should be used before the first inflate() call after inflateInit2() or
+   inflateReset().  bits must be less than or equal to 16, and that many of the
+   least significant bits of value will be inserted in the input.
+
+     If bits is negative, then the input stream bit buffer is emptied.  Then
+   inflatePrime() can be called again to put bits in the buffer.  This is used
+   to clear out bits leftover after feeding inflate a block description prior
+   to feeding inflate codes.
 
-      inflatePrime returns Z_OK if success, or Z_STREAM_ERROR if the source
+     inflatePrime returns Z_OK if success, or Z_STREAM_ERROR if the source
    stream state was inconsistent.
 */
 
+ZEXTERN long ZEXPORT inflateMark OF((z_streamp strm));
+/*
+     This function returns two values, one in the lower 16 bits of the return
+   value, and the other in the remaining upper bits, obtained by shifting the
+   return value down 16 bits.  If the upper value is -1 and the lower value is
+   zero, then inflate() is currently decoding information outside of a block.
+   If the upper value is -1 and the lower value is non-zero, then inflate is in
+   the middle of a stored block, with the lower value equaling the number of
+   bytes from the input remaining to copy.  If the upper value is not -1, then
+   it is the number of bits back from the current bit position in the input of
+   the code (literal or length/distance pair) currently being processed.  In
+   that case the lower value is the number of bytes already emitted for that
+   code.
+
+     A code is being processed if inflate is waiting for more input to complete
+   decoding of the code, or if it has completed decoding but is waiting for
+   more output space to write the literal or match data.
+
+     inflateMark() is used to mark locations in the input data for random
+   access, which may be at bit positions, and to note those cases where the
+   output of a code may span boundaries of random access blocks.  The current
+   location in the input stream can be determined from avail_in and data_type
+   as noted in the description for the Z_BLOCK flush parameter for inflate.
+
+     inflateMark returns the value noted above or -1 << 16 if the provided
+   source stream state was inconsistent.
+*/
+
 ZEXTERN int ZEXPORT inflateGetHeader OF((z_streamp strm,
                                          gz_headerp head));
 /*
-      inflateGetHeader() requests that gzip header information be stored in the
+     inflateGetHeader() requests that gzip header information be stored in the
    provided gz_header structure.  inflateGetHeader() may be called after
    inflateInit2() or inflateReset(), and before the first call of inflate().
    As inflate() processes the gzip stream, head->done is zero until the header
    is completed, at which time head->done is set to one.  If a zlib stream is
    being decoded, then head->done is set to -1 to indicate that there will be
-   no gzip header information forthcoming.  Note that Z_BLOCK can be used to
-   force inflate() to return immediately after header processing is complete
-   and before any actual data is decompressed.
+   no gzip header information forthcoming.  Note that Z_BLOCK or Z_TREES can be
+   used to force inflate() to return immediately after header processing is
+   complete and before any actual data is decompressed.
 
-      The text, time, xflags, and os fields are filled in with the gzip header
+     The text, time, xflags, and os fields are filled in with the gzip header
    contents.  hcrc is set to true if there is a header CRC.  (The header CRC
-   was valid if done is set to one.)  If extra is not Z_NULL, then extra_max
+   was valid if done is set to one.) If extra is not Z_NULL, then extra_max
    contains the maximum number of bytes to write to extra.  Once done is true,
    extra_len contains the actual extra field length, and extra contains the
    extra field, or that field truncated if extra_max is less than extra_len.
    If name is not Z_NULL, then up to name_max characters are written there,
    terminated with a zero unless the length is greater than name_max.  If
    comment is not Z_NULL, then up to comm_max characters are written there,
-   terminated with a zero unless the length is greater than comm_max.  When
-   any of extra, name, or comment are not Z_NULL and the respective field is
-   not present in the header, then that field is set to Z_NULL to signal its
+   terminated with a zero unless the length is greater than comm_max.  When any
+   of extra, name, or comment are not Z_NULL and the respective field is not
+   present in the header, then that field is set to Z_NULL to signal its
    absence.  This allows the use of deflateSetHeader() with the returned
    structure to duplicate the header.  However if those fields are set to
    allocated memory, then the application will need to save those pointers
    elsewhere so that they can be eventually freed.
 
-      If inflateGetHeader is not used, then the header information is simply
+     If inflateGetHeader is not used, then the header information is simply
    discarded.  The header is always checked for validity, including the header
    CRC if present.  inflateReset() will reset the process to discard the header
    information.  The application would need to call inflateGetHeader() again to
    retrieve the header from the next gzip stream.
 
-      inflateGetHeader returns Z_OK if success, or Z_STREAM_ERROR if the source
+     inflateGetHeader returns Z_OK if success, or Z_STREAM_ERROR if the source
    stream state was inconsistent.
 */
 
@@ -869,9 +962,9 @@
      See inflateBack() for the usage of these routines.
 
      inflateBackInit will return Z_OK on success, Z_STREAM_ERROR if any of
-   the paramaters are invalid, Z_MEM_ERROR if the internal state could not
-   be allocated, or Z_VERSION_ERROR if the version of the library does not
-   match the version of the header file.
+   the paramaters are invalid, Z_MEM_ERROR if the internal state could not be
+   allocated, or Z_VERSION_ERROR if the version of the library does not match
+   the version of the header file.
 */
 
 typedef unsigned (*in_func) OF((void FAR *, unsigned char FAR * FAR *));
@@ -891,15 +984,15 @@
      inflateBackInit() must be called first to allocate the internal state
    and to initialize the state with the user-provided window buffer.
    inflateBack() may then be used multiple times to inflate a complete, raw
-   deflate stream with each call.  inflateBackEnd() is then called to free
-   the allocated state.
+   deflate stream with each call.  inflateBackEnd() is then called to free the
+   allocated state.
 
      A raw deflate stream is one with no zlib or gzip header or trailer.
    This routine would normally be used in a utility that reads zip or gzip
    files and writes out uncompressed files.  The utility would decode the
-   header and process the trailer on its own, hence this routine expects
-   only the raw deflate stream to decompress.  This is different from the
-   normal behavior of inflate(), which expects either a zlib or gzip header and
+   header and process the trailer on its own, hence this routine expects only
+   the raw deflate stream to decompress.  This is different from the normal
+   behavior of inflate(), which expects either a zlib or gzip header and
    trailer around the deflate stream.
 
      inflateBack() uses two subroutines supplied by the caller that are then
@@ -925,7 +1018,7 @@
    calling inflateBack().  If strm->next_in is Z_NULL, then in() will be called
    immediately for input.  If strm->next_in is not Z_NULL, then strm->avail_in
    must also be initialized, and then if strm->avail_in is not zero, input will
-   initially be taken from strm->next_in[0 .. strm->avail_in - 1].
+   initially be taken from strm->next_in[0 ..  strm->avail_in - 1].
 
      The in_desc and out_desc parameters of inflateBack() is passed as the
    first parameter of in() and out() respectively when they are called.  These
@@ -935,15 +1028,15 @@
      On return, inflateBack() will set strm->next_in and strm->avail_in to
    pass back any unused input that was provided by the last in() call.  The
    return values of inflateBack() can be Z_STREAM_END on success, Z_BUF_ERROR
-   if in() or out() returned an error, Z_DATA_ERROR if there was a format
-   error in the deflate stream (in which case strm->msg is set to indicate the
-   nature of the error), or Z_STREAM_ERROR if the stream was not properly
-   initialized.  In the case of Z_BUF_ERROR, an input or output error can be
-   distinguished using strm->next_in which will be Z_NULL only if in() returned
-   an error.  If strm->next is not Z_NULL, then the Z_BUF_ERROR was due to
-   out() returning non-zero.  (in() will always be called before out(), so
-   strm->next_in is assured to be defined if out() returns non-zero.)  Note
-   that inflateBack() cannot return Z_OK.
+   if in() or out() returned an error, Z_DATA_ERROR if there was a format error
+   in the deflate stream (in which case strm->msg is set to indicate the nature
+   of the error), or Z_STREAM_ERROR if the stream was not properly initialized.
+   In the case of Z_BUF_ERROR, an input or output error can be distinguished
+   using strm->next_in which will be Z_NULL only if in() returned an error.  If
+   strm->next_in is not Z_NULL, then the Z_BUF_ERROR was due to out() returning
+   non-zero.  (in() will always be called before out(), so strm->next_in is
+   assured to be defined if out() returns non-zero.) Note that inflateBack()
+   cannot return Z_OK.
 */
 
 ZEXTERN int ZEXPORT inflateBackEnd OF((z_streamp strm));
@@ -999,23 +1092,22 @@
                         /* utility functions */
 
 /*
-     The following utility functions are implemented on top of the
-   basic stream-oriented functions. To simplify the interface, some
-   default options are assumed (compression level and memory usage,
-   standard memory allocation functions). The source code of these
-   utility functions can easily be modified if you need special options.
+     The following utility functions are implemented on top of the basic
+   stream-oriented functions.  To simplify the interface, some default options
+   are assumed (compression level and memory usage, standard memory allocation
+   functions).  The source code of these utility functions can be modified if
+   you need special options.
 */
 
 ZEXTERN int ZEXPORT compress OF((Bytef *dest,   uLongf *destLen,
                                  const Bytef *source, uLong sourceLen));
 /*
      Compresses the source buffer into the destination buffer.  sourceLen is
-   the byte length of the source buffer. Upon entry, destLen is the total
-   size of the destination buffer, which must be at least the value returned
-   by compressBound(sourceLen). Upon exit, destLen is the actual size of the
+   the byte length of the source buffer.  Upon entry, destLen is the total size
+   of the destination buffer, which must be at least the value returned by
+   compressBound(sourceLen).  Upon exit, destLen is the actual size of the
    compressed buffer.
-     This function can be used to compress a whole file at once if the
-   input file is mmap'ed.
+
      compress returns Z_OK if success, Z_MEM_ERROR if there was not
    enough memory, Z_BUF_ERROR if there was not enough room in the output
    buffer.
@@ -1025,11 +1117,11 @@
                                   const Bytef *source, uLong sourceLen,
                                   int level));
 /*
-     Compresses the source buffer into the destination buffer. The level
+     Compresses the source buffer into the destination buffer.  The level
    parameter has the same meaning as in deflateInit.  sourceLen is the byte
-   length of the source buffer. Upon entry, destLen is the total size of the
+   length of the source buffer.  Upon entry, destLen is the total size of the
    destination buffer, which must be at least the value returned by
-   compressBound(sourceLen). Upon exit, destLen is the actual size of the
+   compressBound(sourceLen).  Upon exit, destLen is the actual size of the
    compressed buffer.
 
      compress2 returns Z_OK if success, Z_MEM_ERROR if there was not enough
@@ -1040,22 +1132,20 @@
 ZEXTERN uLong ZEXPORT compressBound OF((uLong sourceLen));
 /*
      compressBound() returns an upper bound on the compressed size after
-   compress() or compress2() on sourceLen bytes.  It would be used before
-   a compress() or compress2() call to allocate the destination buffer.
+   compress() or compress2() on sourceLen bytes.  It would be used before a
+   compress() or compress2() call to allocate the destination buffer.
 */
 
 ZEXTERN int ZEXPORT uncompress OF((Bytef *dest,   uLongf *destLen,
                                    const Bytef *source, uLong sourceLen));
 /*
      Decompresses the source buffer into the destination buffer.  sourceLen is
-   the byte length of the source buffer. Upon entry, destLen is the total
-   size of the destination buffer, which must be large enough to hold the
-   entire uncompressed data. (The size of the uncompressed data must have
-   been saved previously by the compressor and transmitted to the decompressor
-   by some mechanism outside the scope of this compression library.)
-   Upon exit, destLen is the actual size of the compressed buffer.
-     This function can be used to decompress a whole file at once if the
-   input file is mmap'ed.
+   the byte length of the source buffer.  Upon entry, destLen is the total size
+   of the destination buffer, which must be large enough to hold the entire
+   uncompressed data.  (The size of the uncompressed data must have been saved
+   previously by the compressor and transmitted to the decompressor by some
+   mechanism outside the scope of this compression library.) Upon exit, destLen
+   is the actual size of the uncompressed buffer.
 
      uncompress returns Z_OK if success, Z_MEM_ERROR if there was not
    enough memory, Z_BUF_ERROR if there was not enough room in the output
@@ -1063,136 +1153,199 @@
 */
 
 
-typedef voidp gzFile;
+                        /* gzip file access functions */
+
+/*
+     This library supports reading and writing files in gzip (.gz) format with
+   an interface similar to that of stdio, using the functions that start with
+   "gz".  The gzip format is different from the zlib format.  gzip is a gzip
+   wrapper, documented in RFC 1952, wrapped around a deflate stream.
+*/
+
+typedef voidp gzFile;       /* opaque gzip file descriptor */
 
-ZEXTERN gzFile ZEXPORT gzopen  OF((const char *path, const char *mode));
 /*
-     Opens a gzip (.gz) file for reading or writing. The mode parameter
-   is as in fopen ("rb" or "wb") but can also include a compression level
-   ("wb9") or a strategy: 'f' for filtered data as in "wb6f", 'h' for
-   Huffman only compression as in "wb1h", or 'R' for run-length encoding
-   as in "wb1R". (See the description of deflateInit2 for more information
-   about the strategy parameter.)
+ZEXTERN gzFile ZEXPORT gzopen OF((const char *path, const char *mode));
+
+     Opens a gzip (.gz) file for reading or writing.  The mode parameter is as
+   in fopen ("rb" or "wb") but can also include a compression level ("wb9") or
+   a strategy: 'f' for filtered data as in "wb6f", 'h' for Huffman-only
+   compression as in "wb1h", 'R' for run-length encoding as in "wb1R", or 'F'
+   for fixed code compression as in "wb9F".  (See the description of
+   deflateInit2 for more information about the strategy parameter.) Also "a"
+   can be used instead of "w" to request that the gzip stream that will be
+   written be appended to the file.  "+" will result in an error, since reading
+   and writing to the same gzip file is not supported.
 
      gzopen can be used to read a file which is not in gzip format; in this
    case gzread will directly read from the file without decompression.
 
-     gzopen returns NULL if the file could not be opened or if there was
-   insufficient memory to allocate the (de)compression state; errno
-   can be checked to distinguish the two cases (if errno is zero, the
-   zlib error is Z_MEM_ERROR).  */
-
-ZEXTERN gzFile ZEXPORT gzdopen  OF((int fd, const char *mode));
-/*
-     gzdopen() associates a gzFile with the file descriptor fd.  File
-   descriptors are obtained from calls like open, dup, creat, pipe or
-   fileno (in the file has been previously opened with fopen).
-   The mode parameter is as in gzopen.
-     The next call of gzclose on the returned gzFile will also close the
-   file descriptor fd, just like fclose(fdopen(fd), mode) closes the file
-   descriptor fd. If you want to keep fd open, use gzdopen(dup(fd), mode).
-     gzdopen returns NULL if there was insufficient memory to allocate
-   the (de)compression state.
+     gzopen returns NULL if the file could not be opened, if there was
+   insufficient memory to allocate the gzFile state, or if an invalid mode was
+   specified (an 'r', 'w', or 'a' was not provided, or '+' was provided).
+   errno can be checked to determine if the reason gzopen failed was that the
+   file could not be opened.
+*/
+
+ZEXTERN gzFile ZEXPORT gzdopen OF((int fd, const char *mode));
+/*
+     gzdopen associates a gzFile with the file descriptor fd.  File descriptors
+   are obtained from calls like open, dup, creat, pipe or fileno (if the file
+   has been previously opened with fopen).  The mode parameter is as in gzopen.
+
+     The next call of gzclose on the returned gzFile will also close the file
+   descriptor fd, just like fclose(fdopen(fd, mode)) closes the file descriptor
+   fd.  If you want to keep fd open, use fd = dup(fd_keep); gz = gzdopen(fd,
+   mode);.  The duplicated descriptor should be saved to avoid a leak, since
+   gzdopen does not close fd if it fails.
+
+     gzdopen returns NULL if there was insufficient memory to allocate the
+   gzFile state, if an invalid mode was specified (an 'r', 'w', or 'a' was not
+   provided, or '+' was provided), or if fd is -1.  The file descriptor is not
+   used until the next gz* read, write, seek, or close operation, so gzdopen
+   will not detect if fd is invalid (unless fd is -1).
+*/
+
+ZEXTERN int ZEXPORT gzbuffer OF((gzFile file, unsigned size));
+/*
+     Set the internal buffer size used by this library's functions.  The
+   default buffer size is 8192 bytes.  This function must be called after
+   gzopen() or gzdopen(), and before any other calls that read or write the
+   file.  The buffer memory allocation is always deferred to the first read or
+   write.  Two buffers are allocated, either both of the specified size when
+   writing, or one of the specified size and the other twice that size when
+   reading.  A larger buffer size of, for example, 64K or 128K bytes will
+   noticeably increase the speed of decompression (reading).
+
+     The new buffer size also affects the maximum length for gzprintf().
+
+     gzbuffer() returns 0 on success, or -1 on failure, such as being called
+   too late.
 */
 
 ZEXTERN int ZEXPORT gzsetparams OF((gzFile file, int level, int strategy));
 /*
-     Dynamically update the compression level or strategy. See the description
+     Dynamically update the compression level or strategy.  See the description
    of deflateInit2 for the meaning of these parameters.
+
      gzsetparams returns Z_OK if success, or Z_STREAM_ERROR if the file was not
    opened for writing.
 */
 
-ZEXTERN int ZEXPORT    gzread  OF((gzFile file, voidp buf, unsigned len));
+ZEXTERN int ZEXPORT gzread OF((gzFile file, voidp buf, unsigned len));
 /*
-     Reads the given number of uncompressed bytes from the compressed file.
-   If the input file was not in gzip format, gzread copies the given number
-   of bytes into the buffer.
-     gzread returns the number of uncompressed bytes actually read (0 for
-   end of file, -1 for error). */
+     Reads the given number of uncompressed bytes from the compressed file.  If
+   the input file was not in gzip format, gzread copies the given number of
+   bytes into the buffer.
+
+     After reaching the end of a gzip stream in the input, gzread will continue
+   to read, looking for another gzip stream, or failing that, reading the rest
+   of the input file directly without decompression.  The entire input file
+   will be read if gzread is called until it returns less than the requested
+   len.
 
-ZEXTERN int ZEXPORT    gzwrite OF((gzFile file,
-                                   voidpc buf, unsigned len));
+     gzread returns the number of uncompressed bytes actually read, less than
+   len for end of file, or -1 for error.
+*/
+
+ZEXTERN int ZEXPORT gzwrite OF((gzFile file,
+                                voidpc buf, unsigned len));
 /*
      Writes the given number of uncompressed bytes into the compressed file.
-   gzwrite returns the number of uncompressed bytes actually written
-   (0 in case of error).
+   gzwrite returns the number of uncompressed bytes written or 0 in case of
+   error.
 */
 
-ZEXTERN int ZEXPORTVA   gzprintf OF((gzFile file, const char *format, ...));
+ZEXTERN int ZEXPORTVA gzprintf OF((gzFile file, const char *format, ...));
 /*
-     Converts, formats, and writes the args to the compressed file under
-   control of the format string, as in fprintf. gzprintf returns the number of
-   uncompressed bytes actually written (0 in case of error).  The number of
-   uncompressed bytes written is limited to 4095. The caller should assure that
-   this limit is not exceeded. If it is exceeded, then gzprintf() will return
-   return an error (0) with nothing written. In this case, there may also be a
-   buffer overflow with unpredictable consequences, which is possible only if
-   zlib was compiled with the insecure functions sprintf() or vsprintf()
-   because the secure snprintf() or vsnprintf() functions were not available.
+     Converts, formats, and writes the arguments to the compressed file under
+   control of the format string, as in fprintf.  gzprintf returns the number of
+   uncompressed bytes actually written, or 0 in case of error.  The number of
+   uncompressed bytes written is limited to 8191, or one less than the buffer
+   size given to gzbuffer().  The caller should assure that this limit is not
+   exceeded.  If it is exceeded, then gzprintf() will return an error (0) with
+   nothing written.  In this case, there may also be a buffer overflow with
+   unpredictable consequences, which is possible only if zlib was compiled with
+   the insecure functions sprintf() or vsprintf() because the secure snprintf()
+   or vsnprintf() functions were not available.  This can be determined using
+   zlibCompileFlags().
 */
 
 ZEXTERN int ZEXPORT gzputs OF((gzFile file, const char *s));
 /*
-      Writes the given null-terminated string to the compressed file, excluding
+     Writes the given null-terminated string to the compressed file, excluding
    the terminating null character.
-      gzputs returns the number of characters written, or -1 in case of error.
+
+     gzputs returns the number of characters written, or -1 in case of error.
 */
 
 ZEXTERN char * ZEXPORT gzgets OF((gzFile file, char *buf, int len));
 /*
-      Reads bytes from the compressed file until len-1 characters are read, or
-   a newline character is read and transferred to buf, or an end-of-file
-   condition is encountered.  The string is then terminated with a null
-   character.
-      gzgets returns buf, or Z_NULL in case of error.
+     Reads bytes from the compressed file until len-1 characters are read, or a
+   newline character is read and transferred to buf, or an end-of-file
+   condition is encountered.  If any characters are read or if len == 1, the
+   string is terminated with a null character.  If no characters are read due
+   to an end-of-file or len < 1, then the buffer is left untouched.
+
+     gzgets returns buf which is a null-terminated string, or it returns NULL
+   for end-of-file or in case of error.  If there was an error, the contents at
+   buf are indeterminate.
 */
 
-ZEXTERN int ZEXPORT    gzputc OF((gzFile file, int c));
+ZEXTERN int ZEXPORT gzputc OF((gzFile file, int c));
 /*
-      Writes c, converted to an unsigned char, into the compressed file.
-   gzputc returns the value that was written, or -1 in case of error.
+     Writes c, converted to an unsigned char, into the compressed file.  gzputc
+   returns the value that was written, or -1 in case of error.
 */
 
-ZEXTERN int ZEXPORT    gzgetc OF((gzFile file));
+ZEXTERN int ZEXPORT gzgetc OF((gzFile file));
 /*
-      Reads one byte from the compressed file. gzgetc returns this byte
-   or -1 in case of end of file or error.
+     Reads one byte from the compressed file.  gzgetc returns this byte or -1
+   in case of end of file or error.
 */
 
-ZEXTERN int ZEXPORT    gzungetc OF((int c, gzFile file));
+ZEXTERN int ZEXPORT gzungetc OF((int c, gzFile file));
 /*
-      Push one character back onto the stream to be read again later.
-   Only one character of push-back is allowed.  gzungetc() returns the
-   character pushed, or -1 on failure.  gzungetc() will fail if a
-   character has been pushed but not read yet, or if c is -1. The pushed
-   character will be discarded if the stream is repositioned with gzseek()
-   or gzrewind().
+     Push one character back onto the stream to be read as the first character
+   on the next read.  At least one character of push-back is allowed.
+   gzungetc() returns the character pushed, or -1 on failure.  gzungetc() will
+   fail if c is -1, and may fail if a character has been pushed but not read
+   yet.  If gzungetc is used immediately after gzopen or gzdopen, at least the
+   output buffer size of pushed characters is allowed.  (See gzbuffer above.)
+   The pushed character will be discarded if the stream is repositioned with
+   gzseek() or gzrewind().
 */
 
-ZEXTERN int ZEXPORT    gzflush OF((gzFile file, int flush));
+ZEXTERN int ZEXPORT gzflush OF((gzFile file, int flush));
 /*
-     Flushes all pending output into the compressed file. The parameter
-   flush is as in the deflate() function. The return value is the zlib
-   error number (see function gzerror below). gzflush returns Z_OK if
-   the flush parameter is Z_FINISH and all output could be flushed.
-     gzflush should be called only when strictly necessary because it can
-   degrade compression.
+     Flushes all pending output into the compressed file.  The parameter flush
+   is as in the deflate() function.  The return value is the zlib error number
+   (see function gzerror below).  gzflush is only permitted when writing.
+
+     If the flush parameter is Z_FINISH, the remaining data is written and the
+   gzip stream is completed in the output.  If gzwrite() is called again, a new
+   gzip stream will be started in the output.  gzread() is able to read such
+   concatented gzip streams.
+
+     gzflush should be called only when strictly necessary because it will
+   degrade compression if called too often.
 */
 
-ZEXTERN z_off_t ZEXPORT    gzseek OF((gzFile file,
-                                      z_off_t offset, int whence));
 /*
-      Sets the starting position for the next gzread or gzwrite on the
-   given compressed file. The offset represents a number of bytes in the
-   uncompressed data stream. The whence parameter is defined as in lseek(2);
+ZEXTERN z_off_t ZEXPORT gzseek OF((gzFile file,
+                                   z_off_t offset, int whence));
+
+     Sets the starting position for the next gzread or gzwrite on the given
+   compressed file.  The offset represents a number of bytes in the
+   uncompressed data stream.  The whence parameter is defined as in lseek(2);
    the value SEEK_END is not supported.
+
      If the file is opened for reading, this function is emulated but can be
-   extremely slow. If the file is opened for writing, only forward seeks are
+   extremely slow.  If the file is opened for writing, only forward seeks are
    supported; gzseek then compresses a sequence of zeroes up to the new
    starting position.
 
-      gzseek returns the resulting offset location as measured in bytes from
+     gzseek returns the resulting offset location as measured in bytes from
    the beginning of the uncompressed stream, or -1 in case of error, in
    particular if the file is opened for writing and the new starting position
    would be before the current position.
@@ -1202,68 +1355,127 @@
 /*
      Rewinds the given file. This function is supported only for reading.
 
-   gzrewind(file) is equivalent to (int)gzseek(file, 0L, SEEK_SET)
+     gzrewind(file) is equivalent to (int)gzseek(file, 0L, SEEK_SET)
 */
 
+/*
 ZEXTERN z_off_t ZEXPORT    gztell OF((gzFile file));
+
+     Returns the starting position for the next gzread or gzwrite on the given
+   compressed file.  This position represents a number of bytes in the
+   uncompressed data stream, and is zero when starting, even if appending or
+   reading a gzip stream from the middle of a file using gzdopen().
+
+     gztell(file) is equivalent to gzseek(file, 0L, SEEK_CUR)
+*/
+
 /*
-     Returns the starting position for the next gzread or gzwrite on the
-   given compressed file. This position represents a number of bytes in the
-   uncompressed data stream.
+ZEXTERN z_off_t ZEXPORT gzoffset OF((gzFile file));
 
-   gztell(file) is equivalent to gzseek(file, 0L, SEEK_CUR)
+     Returns the current offset in the file being read or written.  This offset
+   includes the count of bytes that precede the gzip stream, for example when
+   appending or when using gzdopen() for reading.  When reading, the offset
+   does not include as yet unused buffered input.  This information can be used
+   for a progress indicator.  On error, gzoffset() returns -1.
 */
 
 ZEXTERN int ZEXPORT gzeof OF((gzFile file));
 /*
-     Returns 1 when EOF has previously been detected reading the given
-   input stream, otherwise zero.
+     Returns true (1) if the end-of-file indicator has been set while reading,
+   false (0) otherwise.  Note that the end-of-file indicator is set only if the
+   read tried to go past the end of the input, but came up short.  Therefore,
+   just like feof(), gzeof() may return false even if there is no more data to
+   read, in the event that the last read request was for the exact number of
+   bytes remaining in the input file.  This will happen if the input file size
+   is an exact multiple of the buffer size.
+
+     If gzeof() returns true, then the read functions will return no more data,
+   unless the end-of-file indicator is reset by gzclearerr() and the input file
+   has grown since the previous end of file was detected.
 */
 
 ZEXTERN int ZEXPORT gzdirect OF((gzFile file));
 /*
-     Returns 1 if file is being read directly without decompression, otherwise
-   zero.
+     Returns true (1) if file is being copied directly while reading, or false
+   (0) if file is a gzip stream being decompressed.  This state can change from
+   false to true while reading the input file if the end of a gzip stream is
+   reached, but is followed by data that is not another gzip stream.
+
+     If the input file is empty, gzdirect() will return true, since the input
+   does not contain a gzip stream.
+
+     If gzdirect() is used immediately after gzopen() or gzdopen() it will
+   cause buffers to be allocated to allow reading the file to determine if it
+   is a gzip file.  Therefore if gzbuffer() is used, it should be called before
+   gzdirect().
 */
 
 ZEXTERN int ZEXPORT    gzclose OF((gzFile file));
 /*
-     Flushes all pending output if necessary, closes the compressed file
-   and deallocates all the (de)compression state. The return value is the zlib
-   error number (see function gzerror below).
+     Flushes all pending output if necessary, closes the compressed file and
+   deallocates the (de)compression state.  Note that once file is closed, you
+   cannot call gzerror with file, since its structures have been deallocated.
+   gzclose must not be called more than once on the same file, just as free
+   must not be called more than once on the same allocation.
+
+     gzclose will return Z_STREAM_ERROR if file is not valid, Z_ERRNO on a
+   file operation error, or Z_OK on success.
+*/
+
+ZEXTERN int ZEXPORT gzclose_r OF((gzFile file));
+ZEXTERN int ZEXPORT gzclose_w OF((gzFile file));
+/*
+     Same as gzclose(), but gzclose_r() is only for use when reading, and
+   gzclose_w() is only for use when writing or appending.  The advantage to
+   using these instead of gzclose() is that they avoid linking in zlib
+   compression or decompression code that is not used when only reading or only
+   writing respectively.  If gzclose() is used, then both compression and
+   decompression code will be included the application when linking to a static
+   zlib library.
 */
 
 ZEXTERN const char * ZEXPORT gzerror OF((gzFile file, int *errnum));
 /*
-     Returns the error message for the last error which occurred on the
-   given compressed file. errnum is set to zlib error number. If an
-   error occurred in the file system and not in the compression library,
-   errnum is set to Z_ERRNO and the application may consult errno
-   to get the exact error code.
+     Returns the error message for the last error which occurred on the given
+   compressed file.  errnum is set to zlib error number.  If an error occurred
+   in the file system and not in the compression library, errnum is set to
+   Z_ERRNO and the application may consult errno to get the exact error code.
+
+     The application must not modify the returned string.  Future calls to
+   this function may invalidate the previously returned string.  If file is
+   closed, then the string previously returned by gzerror will no longer be
+   available.
+
+     gzerror() should be used to distinguish errors from end-of-file for those
+   functions above that do not distinguish those cases in their return values.
 */
 
 ZEXTERN void ZEXPORT gzclearerr OF((gzFile file));
 /*
-     Clears the error and end-of-file flags for file. This is analogous to the
-   clearerr() function in stdio. This is useful for continuing to read a gzip
+     Clears the error and end-of-file flags for file.  This is analogous to the
+   clearerr() function in stdio.  This is useful for continuing to read a gzip
    file that is being written concurrently.
 */
 
+
                         /* checksum functions */
 
 /*
      These functions are not related to compression but are exported
-   anyway because they might be useful in applications using the
-   compression library.
+   anyway because they might be useful in applications using the compression
+   library.
 */
 
 ZEXTERN uLong ZEXPORT adler32 OF((uLong adler, const Bytef *buf, uInt len));
 /*
      Update a running Adler-32 checksum with the bytes buf[0..len-1] and
-   return the updated checksum. If buf is NULL, this function returns
-   the required initial value for the checksum.
-   An Adler-32 checksum is almost as reliable as a CRC32 but can be computed
-   much faster. Usage example:
+   return the updated checksum.  If buf is Z_NULL, this function returns the
+   required initial value for the checksum.
+
+     An Adler-32 checksum is almost as reliable as a CRC32 but can be computed
+   much faster.
+
+   Usage example:
 
      uLong adler = adler32(0L, Z_NULL, 0);
 
@@ -1273,9 +1485,10 @@
      if (adler != original_adler) error();
 */
 
+/*
 ZEXTERN uLong ZEXPORT adler32_combine OF((uLong adler1, uLong adler2,
                                           z_off_t len2));
-/*
+
      Combine two Adler-32 checksums into one.  For two sequences of bytes, seq1
    and seq2 with lengths len1 and len2, Adler-32 checksums were calculated for
    each, adler1 and adler2.  adler32_combine() returns the Adler-32 checksum of
@@ -1285,9 +1498,11 @@
 ZEXTERN uLong ZEXPORT crc32   OF((uLong crc, const Bytef *buf, uInt len));
 /*
      Update a running CRC-32 with the bytes buf[0..len-1] and return the
-   updated CRC-32. If buf is NULL, this function returns the required initial
-   value for the for the crc. Pre- and post-conditioning (one's complement) is
-   performed within this function so it shouldn't be done by the application.
+   updated CRC-32.  If buf is Z_NULL, this function returns the required
+   initial value for the for the crc.  Pre- and post-conditioning (one's
+   complement) is performed within this function so it shouldn't be done by the
+   application.
+
    Usage example:
 
      uLong crc = crc32(0L, Z_NULL, 0);
@@ -1298,9 +1513,9 @@
      if (crc != original_crc) error();
 */
 
+/*
 ZEXTERN uLong ZEXPORT crc32_combine OF((uLong crc1, uLong crc2, z_off_t len2));
 
-/*
      Combine two CRC-32 check values into one.  For two sequences of bytes,
    seq1 and seq2 with lengths len1 and len2, CRC-32 check values were
    calculated for each, crc1 and crc2.  crc32_combine() returns the CRC-32
@@ -1339,16 +1554,57 @@
         inflateInit2_((strm), (windowBits), ZLIB_VERSION, sizeof(z_stream))
 #define inflateBackInit(strm, windowBits, window) \
         inflateBackInit_((strm), (windowBits), (window), \
-        ZLIB_VERSION, sizeof(z_stream))
+                                            ZLIB_VERSION, sizeof(z_stream))
+
+/* provide 64-bit offset functions if _LARGEFILE64_SOURCE defined, and/or
+ * change the regular functions to 64 bits if _FILE_OFFSET_BITS is 64 (if
+ * both are true, the application gets the *64 functions, and the regular
+ * functions are changed to 64 bits) -- in case these are set on systems
+ * without large file support, _LFS64_LARGEFILE must also be true
+ */
+#if defined(_LARGEFILE64_SOURCE) && _LFS64_LARGEFILE-0
+   ZEXTERN gzFile ZEXPORT gzopen64 OF((const char *, const char *));
+   ZEXTERN z_off64_t ZEXPORT gzseek64 OF((gzFile, z_off64_t, int));
+   ZEXTERN z_off64_t ZEXPORT gztell64 OF((gzFile));
+   ZEXTERN z_off64_t ZEXPORT gzoffset64 OF((gzFile));
+   ZEXTERN uLong ZEXPORT adler32_combine64 OF((uLong, uLong, z_off64_t));
+   ZEXTERN uLong ZEXPORT crc32_combine64 OF((uLong, uLong, z_off64_t));
+#endif
 
+#if !defined(ZLIB_INTERNAL) && _FILE_OFFSET_BITS-0 == 64 && _LFS64_LARGEFILE-0
+#  define gzopen gzopen64
+#  define gzseek gzseek64
+#  define gztell gztell64
+#  define gzoffset gzoffset64
+#  define adler32_combine adler32_combine64
+#  define crc32_combine crc32_combine64
+#  ifdef _LARGEFILE64_SOURCE
+     ZEXTERN gzFile ZEXPORT gzopen64 OF((const char *, const char *));
+     ZEXTERN z_off_t ZEXPORT gzseek64 OF((gzFile, z_off_t, int));
+     ZEXTERN z_off_t ZEXPORT gztell64 OF((gzFile));
+     ZEXTERN z_off_t ZEXPORT gzoffset64 OF((gzFile));
+     ZEXTERN uLong ZEXPORT adler32_combine64 OF((uLong, uLong, z_off_t));
+     ZEXTERN uLong ZEXPORT crc32_combine64 OF((uLong, uLong, z_off_t));
+#  endif
+#else
+   ZEXTERN gzFile ZEXPORT gzopen OF((const char *, const char *));
+   ZEXTERN z_off_t ZEXPORT gzseek OF((gzFile, z_off_t, int));
+   ZEXTERN z_off_t ZEXPORT gztell OF((gzFile));
+   ZEXTERN z_off_t ZEXPORT gzoffset OF((gzFile));
+   ZEXTERN uLong ZEXPORT adler32_combine OF((uLong, uLong, z_off_t));
+   ZEXTERN uLong ZEXPORT crc32_combine OF((uLong, uLong, z_off_t));
+#endif
 
+/* hack for buggy compilers */
 #if !defined(ZUTIL_H) && !defined(NO_DUMMY_DECL)
-    struct internal_state {int dummy;}; /* hack for buggy compilers */
+    struct internal_state {int dummy;};
 #endif
 
+/* undocumented functions */
 ZEXTERN const char   * ZEXPORT zError           OF((int));
-ZEXTERN int            ZEXPORT inflateSyncPoint OF((z_streamp z));
+ZEXTERN int            ZEXPORT inflateSyncPoint OF((z_streamp));
 ZEXTERN const uLongf * ZEXPORT get_crc_table    OF((void));
+ZEXTERN int            ZEXPORT inflateUndermine OF((z_streamp, int));
 
 #ifdef __cplusplus
 }

Modified: python/branches/import_unicode/Modules/zlib/zutil.c
==============================================================================
--- python/branches/import_unicode/Modules/zlib/zutil.c	(original)
+++ python/branches/import_unicode/Modules/zlib/zutil.c	Sat Jul 31 13:01:07 2010
@@ -1,5 +1,5 @@
 /* zutil.c -- target dependent utility functions for the compression library
- * Copyright (C) 1995-2005 Jean-loup Gailly.
+ * Copyright (C) 1995-2005, 2010 Jean-loup Gailly.
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -34,25 +34,25 @@
     uLong flags;
 
     flags = 0;
-    switch (sizeof(uInt)) {
+    switch ((int)(sizeof(uInt))) {
     case 2:     break;
     case 4:     flags += 1;     break;
     case 8:     flags += 2;     break;
     default:    flags += 3;
     }
-    switch (sizeof(uLong)) {
+    switch ((int)(sizeof(uLong))) {
     case 2:     break;
     case 4:     flags += 1 << 2;        break;
     case 8:     flags += 2 << 2;        break;
     default:    flags += 3 << 2;
     }
-    switch (sizeof(voidpf)) {
+    switch ((int)(sizeof(voidpf))) {
     case 2:     break;
     case 4:     flags += 1 << 4;        break;
     case 8:     flags += 2 << 4;        break;
     default:    flags += 3 << 4;
     }
-    switch (sizeof(z_off_t)) {
+    switch ((int)(sizeof(z_off_t))) {
     case 2:     break;
     case 4:     flags += 1 << 6;        break;
     case 8:     flags += 2 << 6;        break;
@@ -117,9 +117,9 @@
 #  ifndef verbose
 #    define verbose 0
 #  endif
-int z_verbose = verbose;
+int ZLIB_INTERNAL z_verbose = verbose;
 
-void z_error (m)
+void ZLIB_INTERNAL z_error (m)
     char *m;
 {
     fprintf(stderr, "%s\n", m);
@@ -146,7 +146,7 @@
 
 #ifndef HAVE_MEMCPY
 
-void zmemcpy(dest, source, len)
+void ZLIB_INTERNAL zmemcpy(dest, source, len)
     Bytef* dest;
     const Bytef* source;
     uInt  len;
@@ -157,7 +157,7 @@
     } while (--len != 0);
 }
 
-int zmemcmp(s1, s2, len)
+int ZLIB_INTERNAL zmemcmp(s1, s2, len)
     const Bytef* s1;
     const Bytef* s2;
     uInt  len;
@@ -170,7 +170,7 @@
     return 0;
 }
 
-void zmemzero(dest, len)
+void ZLIB_INTERNAL zmemzero(dest, len)
     Bytef* dest;
     uInt  len;
 {
@@ -213,7 +213,7 @@
  * a protected system like OS/2. Use Microsoft C instead.
  */
 
-voidpf zcalloc (voidpf opaque, unsigned items, unsigned size)
+voidpf ZLIB_INTERNAL zcalloc (voidpf opaque, unsigned items, unsigned size)
 {
     voidpf buf = opaque; /* just to make some compilers happy */
     ulg bsize = (ulg)items*size;
@@ -237,7 +237,7 @@
     return buf;
 }
 
-void  zcfree (voidpf opaque, voidpf ptr)
+void ZLIB_INTERNAL zcfree (voidpf opaque, voidpf ptr)
 {
     int n;
     if (*(ush*)&ptr != 0) { /* object < 64K */
@@ -272,13 +272,13 @@
 #  define _hfree   hfree
 #endif
 
-voidpf zcalloc (voidpf opaque, unsigned items, unsigned size)
+voidpf ZLIB_INTERNAL zcalloc (voidpf opaque, uInt items, uInt size)
 {
     if (opaque) opaque = 0; /* to make compiler happy */
     return _halloc((long)items, size);
 }
 
-void  zcfree (voidpf opaque, voidpf ptr)
+void ZLIB_INTERNAL zcfree (voidpf opaque, voidpf ptr)
 {
     if (opaque) opaque = 0; /* to make compiler happy */
     _hfree(ptr);
@@ -297,7 +297,7 @@
 extern void   free   OF((voidpf ptr));
 #endif
 
-voidpf zcalloc (opaque, items, size)
+voidpf ZLIB_INTERNAL zcalloc (opaque, items, size)
     voidpf opaque;
     unsigned items;
     unsigned size;
@@ -307,7 +307,7 @@
                               (voidpf)calloc(items, size);
 }
 
-void  zcfree (opaque, ptr)
+void ZLIB_INTERNAL zcfree (opaque, ptr)
     voidpf opaque;
     voidpf ptr;
 {

Modified: python/branches/import_unicode/Modules/zlib/zutil.h
==============================================================================
--- python/branches/import_unicode/Modules/zlib/zutil.h	(original)
+++ python/branches/import_unicode/Modules/zlib/zutil.h	Sat Jul 31 13:01:07 2010
@@ -1,5 +1,5 @@
 /* zutil.h -- internal interface and configuration of the compression library
- * Copyright (C) 1995-2005 Jean-loup Gailly.
+ * Copyright (C) 1995-2010 Jean-loup Gailly.
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
@@ -13,31 +13,21 @@
 #ifndef ZUTIL_H
 #define ZUTIL_H
 
-#define ZLIB_INTERNAL
+#if ((__GNUC__-0) * 10 + __GNUC_MINOR__-0 >= 33) && !defined(NO_VIZ)
+#  define ZLIB_INTERNAL __attribute__((visibility ("hidden")))
+#else
+#  define ZLIB_INTERNAL
+#endif
+
 #include "zlib.h"
 
 #ifdef STDC
-#  ifndef _WIN32_WCE
+#  if !(defined(_WIN32_WCE) && defined(_MSC_VER))
 #    include <stddef.h>
 #  endif
 #  include <string.h>
 #  include <stdlib.h>
 #endif
-#ifdef NO_ERRNO_H
-#   ifdef _WIN32_WCE
-      /* The Microsoft C Run-Time Library for Windows CE doesn't have
-       * errno.  We define it as a global variable to simplify porting.
-       * Its value is always 0 and should not be used.  We rename it to
-       * avoid conflict with other libraries that use the same workaround.
-       */
-#     define errno z_errno
-#   endif
-    extern int errno;
-#else
-#  ifndef _WIN32_WCE
-#    include <errno.h>
-#  endif
-#endif
 
 #ifndef local
 #  define local static
@@ -89,7 +79,7 @@
 #if defined(MSDOS) || (defined(WINDOWS) && !defined(WIN32))
 #  define OS_CODE  0x00
 #  if defined(__TURBOC__) || defined(__BORLANDC__)
-#    if(__STDC__ == 1) && (defined(__LARGE__) || defined(__COMPACT__))
+#    if (__STDC__ == 1) && (defined(__LARGE__) || defined(__COMPACT__))
        /* Allow compilation with ANSI keywords only enabled */
        void _Cdecl farfree( void *block );
        void *_Cdecl farmalloc( unsigned long nbytes );
@@ -118,7 +108,7 @@
 #ifdef OS2
 #  define OS_CODE  0x06
 #  ifdef M_I86
-     #include <malloc.h>
+#    include <malloc.h>
 #  endif
 #endif
 
@@ -147,11 +137,11 @@
 #  define OS_CODE  0x0f
 #endif
 
-#if defined(_BEOS_)
+#if defined(_BEOS_) || defined(RISCOS)
 #  define fdopen(fd,mode) NULL /* No fdopen() */
 #endif
 
-#if (defined(_MSC_VER) && (_MSC_VER > 600))
+#if (defined(_MSC_VER) && (_MSC_VER > 600)) && !defined __INTERIX
 #  if defined(_WIN32_WCE)
 #    define fdopen(fd,mode) NULL /* No fdopen() */
 #    ifndef _PTRDIFF_T_DEFINED
@@ -163,6 +153,18 @@
 #  endif
 #endif
 
+#if defined(__BORLANDC__)
+  #pragma warn -8004
+  #pragma warn -8008
+  #pragma warn -8066
+#endif
+
+/* provide prototypes for these when building zlib without LFS */
+#if !defined(_LARGEFILE64_SOURCE) || _LFS64_LARGEFILE-0 == 0
+    ZEXTERN uLong ZEXPORT adler32_combine64 OF((uLong, uLong, z_off_t));
+    ZEXTERN uLong ZEXPORT crc32_combine64 OF((uLong, uLong, z_off_t));
+#endif
+
         /* common defaults */
 
 #ifndef OS_CODE
@@ -197,7 +199,9 @@
 #  ifdef WIN32
      /* In Win32, vsnprintf is available as the "non-ANSI" _vsnprintf. */
 #    if !defined(vsnprintf) && !defined(NO_vsnprintf)
-#      define vsnprintf _vsnprintf
+#      if !defined(_MSC_VER) || ( defined(_MSC_VER) && _MSC_VER < 1500 )
+#         define vsnprintf _vsnprintf
+#      endif
 #    endif
 #  endif
 #  ifdef __SASC
@@ -232,16 +236,16 @@
 #    define zmemzero(dest, len) memset(dest, 0, len)
 #  endif
 #else
-   extern void zmemcpy  OF((Bytef* dest, const Bytef* source, uInt len));
-   extern int  zmemcmp  OF((const Bytef* s1, const Bytef* s2, uInt len));
-   extern void zmemzero OF((Bytef* dest, uInt len));
+   void ZLIB_INTERNAL zmemcpy OF((Bytef* dest, const Bytef* source, uInt len));
+   int ZLIB_INTERNAL zmemcmp OF((const Bytef* s1, const Bytef* s2, uInt len));
+   void ZLIB_INTERNAL zmemzero OF((Bytef* dest, uInt len));
 #endif
 
 /* Diagnostic functions */
 #ifdef DEBUG
 #  include <stdio.h>
-   extern int z_verbose;
-   extern void z_error    OF((char *m));
+   extern int ZLIB_INTERNAL z_verbose;
+   extern void ZLIB_INTERNAL z_error OF((char *m));
 #  define Assert(cond,msg) {if(!(cond)) z_error(msg);}
 #  define Trace(x) {if (z_verbose>=0) fprintf x ;}
 #  define Tracev(x) {if (z_verbose>0) fprintf x ;}
@@ -258,8 +262,9 @@
 #endif
 
 
-voidpf zcalloc OF((voidpf opaque, unsigned items, unsigned size));
-void   zcfree  OF((voidpf opaque, voidpf ptr));
+voidpf ZLIB_INTERNAL zcalloc OF((voidpf opaque, unsigned items,
+                        unsigned size));
+void ZLIB_INTERNAL zcfree  OF((voidpf opaque, voidpf ptr));
 
 #define ZALLOC(strm, items, size) \
            (*((strm)->zalloc))((strm)->opaque, (items), (size))

Modified: python/branches/import_unicode/PC/python_nt.rc
==============================================================================
--- python/branches/import_unicode/PC/python_nt.rc	(original)
+++ python/branches/import_unicode/PC/python_nt.rc	Sat Jul 31 13:01:07 2010
@@ -61,7 +61,7 @@
             VALUE "FileDescription", "Python Core\0"
             VALUE "FileVersion", PYTHON_VERSION
             VALUE "InternalName", "Python DLL\0"
-            VALUE "LegalCopyright", "Copyright ? 2001-2008 Python Software Foundation. Copyright ? 2000 BeOpen.com. Copyright ? 1995-2001 CNRI. Copyright ? 1991-1995 SMC.\0"
+            VALUE "LegalCopyright", "Copyright ? 2001-2010 Python Software Foundation. Copyright ? 2000 BeOpen.com. Copyright ? 1995-2001 CNRI. Copyright ? 1991-1995 SMC.\0"
             VALUE "OriginalFilename", PYTHON_DLL_NAME "\0"
             VALUE "ProductName", "Python\0"
             VALUE "ProductVersion", PYTHON_VERSION

Modified: python/branches/import_unicode/PCbuild/build_ssl.py
==============================================================================
--- python/branches/import_unicode/PCbuild/build_ssl.py	(original)
+++ python/branches/import_unicode/PCbuild/build_ssl.py	Sat Jul 31 13:01:07 2010
@@ -166,12 +166,14 @@
         do_script = "ms\\do_nasm"
         makefile="ms\\nt.mak"
         m32 = makefile
+        dirsuffix = "32"
     elif sys.argv[2] == "x64":
         arch="amd64"
         configure = "VC-WIN64A"
         do_script = "ms\\do_win64a"
         makefile = "ms\\nt64.mak"
         m32 = makefile.replace('64', '')
+        dirsuffix = "64"
         #os.environ["VSEXTCOMP_USECL"] = "MS_OPTERON"
     else:
         raise ValueError(str(sys.argv))
@@ -228,6 +230,13 @@
             shutil.copy(r"crypto\buildinf.h", r"crypto\buildinf_%s.h" % arch)
             shutil.copy(r"crypto\opensslconf.h", r"crypto\opensslconf_%s.h" % arch)
 
+        # If the assembler files don't exist in tmpXX, copy them there
+        if not os.path.exists("tmp"+dirsuffix):
+            os.mkdir("tmp"+dirsuffix)
+        for f in os.listdir("asm"+dirsuffix):
+            if not f.endswith(".asm"): continue
+            shutil.copy(r"asm%s\%s" % (dirsuffix, f), "tmp"+dirsuffix)
+
         # Now run make.
         if arch == "amd64":
             rc = os.system("ml64 -c -Foms\\uptable.obj ms\\uptable.asm")

Modified: python/branches/import_unicode/Python/getcopyright.c
==============================================================================
--- python/branches/import_unicode/Python/getcopyright.c	(original)
+++ python/branches/import_unicode/Python/getcopyright.c	Sat Jul 31 13:01:07 2010
@@ -2,7 +2,7 @@
 
 #include "Python.h"
 
-static char cprt[] = 
+static char cprt[] =
 "\
 Copyright (c) 2001-2010 Python Software Foundation.\n\
 All Rights Reserved.\n\

Modified: python/branches/import_unicode/README
==============================================================================
--- python/branches/import_unicode/README	(original)
+++ python/branches/import_unicode/README	Sat Jul 31 13:01:07 2010
@@ -1,5 +1,5 @@
-This is Python version 3.2
-==========================
+This is Python version 3.2 alpha 1
+==================================
 
 Copyright (c) 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010
 Python Software Foundation.
@@ -53,9 +53,9 @@
 ----------
 
 We try to have a comprehensive overview of the changes in the "What's New in
-Python 3.1" document, found at
+Python 3.2" document, found at
 
-    http://docs.python.org/dev/3.1/whatsnew/3.1.html
+    http://docs.python.org/dev/3.2/whatsnew/3.2.html
 
 For a more detailed change log, read Misc/NEWS (though this file, too,
 is incomplete, and also doesn't list anything merged in from the 2.7
@@ -68,30 +68,27 @@
 Documentation
 -------------
 
-Documentation for Python 3.1 is online, updated twice a day:
+Documentation for Python 3.2 is online, updated daily:
 
-    http://docs.python.org/dev/3.1/
+    http://docs.python.org/dev/
 
-All documentation is also available online at the Python web site
-(http://docs.python.org/, see below).  It is available online for
-occasional reference, or can be downloaded in many formats for faster
-access.  The documentation is downloadable in HTML, PostScript, PDF,
-LaTeX (through 2.5), and reStructuredText (2.6+) formats; the LaTeX and
-reStructuredText versions are primarily for documentation authors,
-translators, and people with special formatting requirements.
+It can also be downloaded in many formats for faster access.  The documentation
+is downloadable in HTML, PDF, and reStructuredText formats; the latter version
+is primarily for documentation authors, translators, and people with special
+formatting requirements.
 
 
 Converting From Python 2.x to 3.x
 ---------------------------------
 
-Python starting with 2.6 will contain features to help locating code that
+Python starting with 2.6 contains features to help locating code that
 needs to be changed, such as optional warnings when deprecated features are
 used, and backported versions of certain key Python 3.x features.
 
 A source-to-source translation tool, "2to3", can take care of the mundane task
 of converting large amounts of source code.  It is not a complete solution but
 is complemented by the deprecation warnings in 2.6.  See
-http://docs.python.org/dev/py3k/library/2to3.html for more information.
+http://docs.python.org/dev/library/2to3.html for more information.
 
 
 Testing
@@ -135,7 +132,7 @@
 version (if any) is your "primary" version.  Install that version using
 "make install".  Install all other versions using "make altinstall".
 
-For example, if you want to install Python 2.5, 2.6 and 3.0 with 2.6 being
+For example, if you want to install Python 2.5, 2.6 and 3.2 with 2.6 being
 the primary version, you would execute "make install" in your 2.6 build
 directory and "make altinstall" in the others.
 

Modified: python/branches/import_unicode/Tools/buildbot/external-common.bat
==============================================================================
--- python/branches/import_unicode/Tools/buildbot/external-common.bat	(original)
+++ python/branches/import_unicode/Tools/buildbot/external-common.bat	Sat Jul 31 13:01:07 2010
@@ -14,7 +14,8 @@
 @rem if exist tk8.4.16 rd /s/q tk8.4.16
 @rem if exist tk-8.4.18.1 rd /s/q tk-8.4.18.1
 @rem if exist db-4.4.20 rd /s/q db-4.4.20
- at rem if exist openssl-0.9.8l rd /s/q openssl-0.9.8l
+if exist openssl-0.9.8l rd /s/q openssl-0.9.8l
+if exist openssl-1.0.0a rd /s/q openssl-1.0.0a
 @rem if exist sqlite-3.6.21 rd /s/q sqlite-3.6.21    
 
 @rem bzip
@@ -27,7 +28,7 @@
 if not exist db-4.4.20 svn export http://svn.python.org/projects/external/db-4.4.20-vs9 db-4.4.20
 
 @rem OpenSSL
-if not exist openssl-0.9.8l svn export http://svn.python.org/projects/external/openssl-0.9.8l
+if not exist openssl-1.0.0a svn export http://svn.python.org/projects/external/openssl-1.0.0a
 
 @rem tcl/tk
 if not exist tcl-8.5.2.1 (

From python-checkins at python.org  Sat Jul 31 13:40:07 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 13:40:07 +0200 (CEST)
Subject: [Python-checkins] r83341 -
	python/branches/py3k/Doc/library/datetime.rst
Message-ID: <20100731114007.B3233EE9C7@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 13:40:07 2010
New Revision: 83341

Log:
#9430: document timedelta str() and repr().

Modified:
   python/branches/py3k/Doc/library/datetime.rst

Modified: python/branches/py3k/Doc/library/datetime.rst
==============================================================================
--- python/branches/py3k/Doc/library/datetime.rst	(original)
+++ python/branches/py3k/Doc/library/datetime.rst	Sat Jul 31 13:40:07 2010
@@ -265,6 +265,14 @@
 | ``abs(t)``                     | equivalent to +\ *t* when ``t.days >= 0``, and|
 |                                | to -*t* when ``t.days < 0``. (2)              |
 +--------------------------------+-----------------------------------------------+
+| ``str(t)``                     | Returns a string in the form                  |
+|                                | ``[D day[s], ][H]H:MM:SS[.UUUUUU]``, where D  |
+|                                | is negative for negative ``t``. (5)           |
++--------------------------------+-----------------------------------------------+
+| ``repr(t)``                    | Returns a string in the form                  |
+|                                | ``datetime.timedelta(D[, S[, U]])``, where D  |
+|                                | is negative for negative ``t``. (5)           |
++--------------------------------+-----------------------------------------------+
 
 Notes:
 
@@ -280,6 +288,16 @@
 (4)
    -*timedelta.max* is not representable as a :class:`timedelta` object.
 
+(5)
+  String representations of :class:`timedelta` objects are normalized
+  similarly to their internal representation.  This leads to somewhat
+  unusual results for negative timedeltas.  For example:
+
+  >>> timedelta(hours=-5)
+  datetime.timedelta(-1, 68400)
+  >>> print(_)
+  -1 day, 19:00:00
+
 In addition to the operations listed above :class:`timedelta` objects support
 certain additions and subtractions with :class:`date` and :class:`datetime`
 objects (see below).

From python-checkins at python.org  Sat Jul 31 13:52:47 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 13:52:47 +0200 (CEST)
Subject: [Python-checkins] r83342 - python/branches/py3k/Lib/test/test_pdb.py
Message-ID: <20100731115247.2173EEE9C7@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 13:52:46 2010
New Revision: 83342

Log:
Import test_pdb with its full name, so that running python -m test.test_pdb succeeds.

Modified:
   python/branches/py3k/Lib/test/test_pdb.py

Modified: python/branches/py3k/Lib/test/test_pdb.py
==============================================================================
--- python/branches/py3k/Lib/test/test_pdb.py	(original)
+++ python/branches/py3k/Lib/test/test_pdb.py	Sat Jul 31 13:52:46 2010
@@ -267,8 +267,8 @@
     """Test the list and source commands of pdb.
 
     >>> def test_function_2(foo):
-    ...     import test_pdb
-    ...     test_pdb.do_nothing()
+    ...     import test.test_pdb
+    ...     test.test_pdb.do_nothing()
     ...     'some...'
     ...     'more...'
     ...     'code...'
@@ -315,8 +315,8 @@
     -> def test_function_2(foo):
     (Pdb) list
       1  ->     def test_function_2(foo):
-      2             import test_pdb
-      3             test_pdb.do_nothing()
+      2             import test.test_pdb
+      3             test.test_pdb.do_nothing()
       4             'some...'
       5             'more...'
       6             'code...'
@@ -333,16 +333,16 @@
     [EOF]
     (Pdb) list 1,3
       1  ->     def test_function_2(foo):
-      2             import test_pdb
-      3             test_pdb.do_nothing()
+      2             import test.test_pdb
+      3             test.test_pdb.do_nothing()
     (Pdb) list x
     *** ...
     (Pdb) next
     > <doctest test.test_pdb.test_list_commands[0]>(2)test_function_2()
-    -> import test_pdb
+    -> import test.test_pdb
     (Pdb) next
     > <doctest test.test_pdb.test_list_commands[0]>(3)test_function_2()
-    -> test_pdb.do_nothing()
+    -> test.test_pdb.do_nothing()
     (Pdb) step
     --Call--
     > ...test_pdb.py(...)do_nothing()

From python-checkins at python.org  Sat Jul 31 14:06:51 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 14:06:51 +0200 (CEST)
Subject: [Python-checkins] r83343 - python/branches/py3k/Tools/msi/uuids.py
Message-ID: <20100731120651.AB43FEEA45@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 14:06:51 2010
New Revision: 83343

Log:
>From Martin: New UUIDs for the 3.2 release series.

Modified:
   python/branches/py3k/Tools/msi/uuids.py

Modified: python/branches/py3k/Tools/msi/uuids.py
==============================================================================
--- python/branches/py3k/Tools/msi/uuids.py	(original)
+++ python/branches/py3k/Tools/msi/uuids.py	Sat Jul 31 14:06:51 2010
@@ -77,4 +77,12 @@
     '3.1.1150':'{7ff90460-89b7-435b-b583-b37b2815ccc7}', # 3.1.1
     '3.1.2121':'{ec45624a-378c-43be-91f3-3f7a59b0d90c}', # 3.1.2c1
     '3.1.2150':'{d40af016-506c-43fb-a738-bd54fa8c1e85}', # 3.1.2
+    '3.2.101' :'{b411f168-7a36-4fff-902c-a554d1c78a4f}', # 3.2a1
+    '3.2.102' :'{79ff73b7-8359-410f-b9c5-152d2026f8c8}', # 3.2a2
+    '3.2.103' :'{e7635c65-c221-4b9b-b70a-5611b8369d77}', # 3.2a3
+    '3.2.111' :'{20bfc16f-c7cd-4fc0-8f96-9914614a3c50}', # 3.2b1
+    '3.2.112' :'{0e350c98-8d73-4993-b686-cfe87160046e}', # 3.2b2
+    '3.2.121' :'{2094968d-7583-47f6-a7fd-22304532e09f}', # 3.2rc1
+    '3.2.122' :'{4f3edfa6-cf70-469a-825f-e1206aa7f412}', # 3.2rc2
+    '3.2.150' :'{b2042d5e-986d-44ec-aee3-afe4108ccc93}', # 3.2.0
 }

From python-checkins at python.org  Sat Jul 31 15:26:15 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 15:26:15 +0200 (CEST)
Subject: [Python-checkins] r83344 - peps/trunk/pep-0101.txt
Message-ID: <20100731132615.80451EEAAE@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 15:26:15 2010
New Revision: 83344

Log:
Consistent checkboxes.

Modified:
   peps/trunk/pep-0101.txt

Modified: peps/trunk/pep-0101.txt
==============================================================================
--- peps/trunk/pep-0101.txt	(original)
+++ peps/trunk/pep-0101.txt	Sat Jul 31 15:26:15 2010
@@ -574,15 +574,15 @@
 
 What Next?
 
-  __ Verify!  Pretend you're a user: download the files from python.org, and
-     make Python from it. This step is too easy to overlook, and on several
-     occasions we've had useless release files.  Once a general server problem
-     caused mysterious corruption of all files; once the source tarball got
-     built incorrectly; more than once the file upload process on SF truncated
-     files; and so on.
+  ___ Verify!  Pretend you're a user: download the files from python.org, and
+      make Python from it. This step is too easy to overlook, and on several
+      occasions we've had useless release files.  Once a general server problem
+      caused mysterious corruption of all files; once the source tarball got
+      built incorrectly; more than once the file upload process on SF truncated
+      files; and so on.
 
-  __ Rejoice.  Drink.  Be Merry.  Write a PEP like this one.  Or be
-     like unto Guido and take A Vacation.
+  ___ Rejoice.  Drink.  Be Merry.  Write a PEP like this one.  Or be
+      like unto Guido and take A Vacation.
 
   You've just made a Python release!
 

From python-checkins at python.org  Sat Jul 31 15:53:30 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 15:53:30 +0200 (CEST)
Subject: [Python-checkins] r83345 - python/tags/r32a1
Message-ID: <20100731135330.05951EEA45@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 15:53:29 2010
New Revision: 83345

Log:
Tagging release 3.2a1.


Added:
   python/tags/r32a1/
      - copied from r83344, /python/branches/py3k/

From python-checkins at python.org  Sat Jul 31 16:13:40 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 16:13:40 +0200 (CEST)
Subject: [Python-checkins] r83346 - sandbox/trunk/release/release.py
Message-ID: <20100731141340.12018EEA75@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 16:13:39 2010
New Revision: 83346

Log:
Allow selecting the release signing private key.

Modified:
   sandbox/trunk/release/release.py

Modified: sandbox/trunk/release/release.py
==============================================================================
--- sandbox/trunk/release/release.py	(original)
+++ sandbox/trunk/release/release.py	Sat Jul 31 16:13:39 2010
@@ -240,8 +240,11 @@
         fp.write(checksum_bz2.hexdigest())
 
     print('Signing tarballs')
-    os.system('gpg -bas ' + tgz)
-    os.system('gpg -bas ' + bz)
+    print('List of available private keys:')
+    run_cmd(['gpg -K | grep -A 1 "^sec"'], silent=True)
+    uid = input('Please enter key ID to use for signing: ')
+    os.system('gpg -bas -u ' + uid + ' ' + tgz)
+    os.system('gpg -bas -u ' + uid + ' ' + bz)
 
 
 def export(tag):

From python-checkins at python.org  Sat Jul 31 19:44:39 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Sat, 31 Jul 2010 19:44:39 +0200 (CEST)
Subject: [Python-checkins] r83347 - peps/trunk/pep-3151.txt
Message-ID: <20100731174439.B7B1AEEA75@mail.python.org>

Author: antoine.pitrou
Date: Sat Jul 31 19:44:39 2010
New Revision: 83347

Log:
Suggesting a mechanism for deprecation of old exception names



Modified:
   peps/trunk/pep-3151.txt

Modified: peps/trunk/pep-3151.txt
==============================================================================
--- peps/trunk/pep-3151.txt	(original)
+++ peps/trunk/pep-3151.txt	Sat Jul 31 19:44:39 2010
@@ -251,9 +251,34 @@
 --------------------
 
 It is not yet decided whether the old names will be deprecated (then removed)
-or all alternative names will continue living in the root namespace.
-Deprecation of names from the root namespace presents some implementation
-challenges, especially where performance is important.
+or all names will continue living forever in the builtins namespace.
+
+built-in exceptions
+'''''''''''''''''''
+
+Deprecating the old built-in exceptions cannot be done in a straightforward
+fashion by intercepting all lookups in the builtins namespace, since these
+are performance-critical.  We also cannot work at the object level, since
+the deprecated names will be aliased to non-deprecated objects.
+
+A solution is to recognize these names at compilation time, and
+then emit a separate ``LOAD_OLD_GLOBAL`` opcode instead of the regular
+``LOAD_GLOBAL``.  This specialized opcode will handle the output of a
+DeprecationWarning (or PendingDeprecationWarning, depending on the policy
+decided upon) when the name doesn't exist in the globals namespace, but
+only in the builtins one.  This will be enough to avoid false positives
+(for example if someone defines their own ``OSError`` in a module), and
+false negatives will be rare (for example when someone accesses ``OSError``
+through the ``builtins`` module rather than directly).
+
+module-level exceptions
+'''''''''''''''''''''''
+
+The above approach cannot be used easily, since it would require
+special-casing some modules when compiling code objects.  However, these
+names are by construction much less visible (they don't appear in the
+builtins namespace), and lesser-known too, so we might decide to let them
+live in their own namespaces.
 
 
 .. _Step 2:

From python-checkins at python.org  Sat Jul 31 20:05:35 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 20:05:35 +0200 (CEST)
Subject: [Python-checkins] r83348 - in python/branches/py3k:
	Include/patchlevel.h Misc/NEWS
Message-ID: <20100731180535.EE3C2EEAF8@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 20:05:35 2010
New Revision: 83348

Log:
Post-release updates.

Modified:
   python/branches/py3k/Include/patchlevel.h
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Include/patchlevel.h
==============================================================================
--- python/branches/py3k/Include/patchlevel.h	(original)
+++ python/branches/py3k/Include/patchlevel.h	Sat Jul 31 20:05:35 2010
@@ -23,7 +23,7 @@
 #define PY_RELEASE_SERIAL	1
 
 /* Version as a string */
-#define PY_VERSION      	"3.2a1"
+#define PY_VERSION      	"3.2a1+"
 /*--end constants--*/
 
 /* Subversion Revision number of this file (not of the repository) */

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sat Jul 31 20:05:35 2010
@@ -4,6 +4,18 @@
 
 (editors: check NEWS.help for information about editing NEWS using ReST.)
 
+What's New in Python 3.2 Alpha 2?
+=================================
+
+*Release date: XX-Sep-2010*
+
+Core and Builtins
+-----------------
+
+Library
+-------
+
+
 What's New in Python 3.2 Alpha 1?
 =================================
 

From python-checkins at python.org  Sat Jul 31 20:08:33 2010
From: python-checkins at python.org (antoine.pitrou)
Date: Sat, 31 Jul 2010 20:08:33 +0200 (CEST)
Subject: [Python-checkins] r83349 - python/branches/py3k/Doc/whatsnew/3.2.rst
Message-ID: <20100731180833.A712BEE994@mail.python.org>

Author: antoine.pitrou
Date: Sat Jul 31 20:08:33 2010
New Revision: 83349

Log:
Add ssl changes to the 3.2 "what's new".



Modified:
   python/branches/py3k/Doc/whatsnew/3.2.rst

Modified: python/branches/py3k/Doc/whatsnew/3.2.rst
==============================================================================
--- python/branches/py3k/Doc/whatsnew/3.2.rst	(original)
+++ python/branches/py3k/Doc/whatsnew/3.2.rst	Sat Jul 31 20:08:33 2010
@@ -69,6 +69,37 @@
 * The :class:`ftplib.FTP` class now supports the context manager protocol
   (Contributed by Tarek Ziad? and Giampaolo Rodol?; :issue:`4972`.)
 
+* The :mod:`ssl` module has a new class, :class:`~ssl.SSLContext` which
+  serves as a container for various persistent SSL data, such as protocol
+  settings, certificates, private keys, and various other options.
+  The :meth:`~ssl.SSLContext.wrap_socket` method allows to create an
+  SSL socket from such an SSL context.
+  (Added by Antoine Pitrou; :issue:`8550`.)
+
+  The :func:`ssl.wrap_socket` constructor function now takes a
+  *ciphers* argument that's a string listing the encryption algorithms
+  to be allowed; the format of the string is described
+  `in the OpenSSL documentation
+  <http://www.openssl.org/docs/apps/ciphers.html#CIPHER_LIST_FORMAT>`__.
+  (Added by Antoine Pitrou; :issue:`8322`.)
+
+  Various options have been added to the :mod:`ssl` module, such as
+  :data:`~ssl.OP_NO_SSLv2` which allows to force disabling of the insecure
+  and obsolete SSLv2 protocol.
+  (Added by Antoine Pitrou; :issue:`4870`.)
+
+  Another change makes the extension load all of OpenSSL's ciphers and
+  digest algorithms so that they're all available.  Some SSL
+  certificates couldn't be verified, reporting an "unknown algorithm"
+  error.  (Reported by Beda Kosata, and fixed by Antoine Pitrou;
+  :issue:`8484`.)
+
+  The version of OpenSSL being used is now available as the module
+  attributes :data:`ssl.OPENSSL_VERSION` (a string),
+  :data:`ssl.OPENSSL_VERSION_INFO` (a 5-tuple), and
+  :data:`ssl.OPENSSL_VERSION_NUMBER` (an integer).  (Added by Antoine
+  Pitrou; :issue:`8321`.)
+
 * The previously deprecated :func:`string.maketrans` function has been
   removed in favor of the static methods, :meth:`bytes.maketrans` and
   :meth:`bytearray.maketrans`.  This change solves the confusion around which

From python-checkins at python.org  Sat Jul 31 20:09:23 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 20:09:23 +0200 (CEST)
Subject: [Python-checkins] r83350 - in python/branches/py3k:
	Doc/library/functools.rst Lib/functools.py
	Lib/test/test_functools.py Misc/NEWS
Message-ID: <20100731180923.49DC2EE994@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 20:09:23 2010
New Revision: 83350

Log:
Re-commit r83327 now that the release is done.

Modified:
   python/branches/py3k/Doc/library/functools.rst
   python/branches/py3k/Lib/functools.py
   python/branches/py3k/Lib/test/test_functools.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Doc/library/functools.rst
==============================================================================
--- python/branches/py3k/Doc/library/functools.rst	(original)
+++ python/branches/py3k/Doc/library/functools.rst	Sat Jul 31 20:09:23 2010
@@ -37,6 +37,57 @@
 
    .. versionadded:: 3.2
 
+.. decorator:: lfu_cache(maxsize)
+
+   Decorator to wrap a function with a memoizing callable that saves up to the
+   *maxsize* most frequent calls.  It can save time when an expensive or I/O
+   bound function is periodically called with the same arguments.
+
+   The *maxsize* parameter defaults to 100.  Since a dictionary is used to cache
+   results, the positional and keyword arguments to the function must be
+   hashable.
+
+   The wrapped function is instrumented with two attributes, :attr:`hits`
+   and :attr:`misses` which count the number of successful or unsuccessful
+   cache lookups.  These statistics are helpful for tuning the *maxsize*
+   parameter and for measuring the cache's effectiveness.
+
+   The wrapped function also has a :attr:`clear` attribute which can be
+   called (with no arguments) to clear the cache.
+
+   A `LFU (least frequently used) cache
+   <http://en.wikipedia.org/wiki/Cache_algorithms#Least-Frequently_Used>`_
+   is indicated when the pattern of calls does not change over time, when
+   more the most common calls already seen are the best predictors of the
+   most common upcoming calls.
+
+   .. versionadded:: 3.2
+
+.. decorator:: lru_cache(maxsize)
+
+   Decorator to wrap a function with a memoizing callable that saves up to the
+   *maxsize* most recent calls.  It can save time when an expensive or I/O bound
+   function is periodically called with the same arguments.
+
+   The *maxsize* parameter defaults to 100.  Since a dictionary is used to cache
+   results, the positional and keyword arguments to the function must be
+   hashable.
+
+   The wrapped function is instrumented with two attributes, :attr:`hits`
+   and :attr:`misses` which count the number of successful or unsuccessful
+   cache lookups.  These statistics are helpful for tuning the *maxsize*
+   parameter and for measuring the cache's effectiveness.
+
+   The wrapped function also has a :attr:`clear` attribute which can be
+   called (with no arguments) to clear the cache.
+
+   A `LRU (least recently used) cache
+   <http://en.wikipedia.org/wiki/Cache_algorithms#Least_Recently_Used>`_
+   is indicated when the pattern of calls changes over time, such as
+   when more recent calls are the best predictors of upcoming calls.
+
+   .. versionadded:: 3.2
+
 .. decorator:: total_ordering
 
    Given a class defining one or more rich comparison ordering methods, this

Modified: python/branches/py3k/Lib/functools.py
==============================================================================
--- python/branches/py3k/Lib/functools.py	(original)
+++ python/branches/py3k/Lib/functools.py	Sat Jul 31 20:09:23 2010
@@ -4,10 +4,17 @@
 # to allow utilities written in Python to be added
 # to the functools module.
 # Written by Nick Coghlan <ncoghlan at gmail.com>
-#   Copyright (C) 2006 Python Software Foundation.
+# and Raymond Hettinger <python at rcn.com>
+#   Copyright (C) 2006-2010 Python Software Foundation.
 # See C source code for _functools credits/copyright
 
+__all__ = ['update_wrapper', 'wraps', 'WRAPPER_ASSIGNMENTS', 'WRAPPER_UPDATES',
+           'total_ordering', 'cmp_to_key', 'lfu_cache', 'lru_cache']
+
 from _functools import partial, reduce
+from collections import OrderedDict, Counter
+from heapq import nsmallest
+from operator import itemgetter
 
 # update_wrapper() and wraps() are tools to help write
 # wrapper functions that can handle naive introspection
@@ -97,3 +104,88 @@
         def __hash__(self):
             raise TypeError('hash not implemented')
     return K
+
+def lfu_cache(maxsize=100):
+    '''Least-frequently-used cache decorator.
+
+    Arguments to the cached function must be hashable.
+    Cache performance statistics stored in f.hits and f.misses.
+    Clear the cache using f.clear().
+    http://en.wikipedia.org/wiki/Cache_algorithms#Least-Frequently_Used
+
+    '''
+    def decorating_function(user_function):
+        cache = {}                      # mapping of args to results
+        use_count = Counter()           # times each key has been accessed
+        kwd_mark = object()             # separate positional and keyword args
+
+        @wraps(user_function)
+        def wrapper(*args, **kwds):
+            key = args
+            if kwds:
+                key += (kwd_mark,) + tuple(sorted(kwds.items()))
+            use_count[key] += 1         # count a use of this key
+            try:
+                result = cache[key]
+                wrapper.hits += 1
+            except KeyError:
+                result = user_function(*args, **kwds)
+                cache[key] = result
+                wrapper.misses += 1
+                if len(cache) > maxsize:
+                    # purge the 10% least frequently used entries
+                    for key, _ in nsmallest(maxsize // 10,
+                                            use_count.items(),
+                                            key=itemgetter(1)):
+                        del cache[key], use_count[key]
+            return result
+
+        def clear():
+            'Clear the cache and cache statistics'
+            cache.clear()
+            use_count.clear()
+            wrapper.hits = wrapper.misses = 0
+
+        wrapper.hits = wrapper.misses = 0
+        wrapper.clear = clear
+        return wrapper
+    return decorating_function
+
+def lru_cache(maxsize=100):
+    '''Least-recently-used cache decorator.
+
+    Arguments to the cached function must be hashable.
+    Cache performance statistics stored in f.hits and f.misses.
+    Clear the cache using f.clear().
+    http://en.wikipedia.org/wiki/Cache_algorithms#Least_Recently_Used
+
+    '''
+    def decorating_function(user_function):
+        cache = OrderedDict()           # ordered least recent to most recent
+        kwd_mark = object()             # separate positional and keyword args
+
+        @wraps(user_function)
+        def wrapper(*args, **kwds):
+            key = args
+            if kwds:
+                key += (kwd_mark,) + tuple(sorted(kwds.items()))
+            try:
+                result = cache.pop(key)
+                wrapper.hits += 1
+            except KeyError:
+                result = user_function(*args, **kwds)
+                wrapper.misses += 1
+                if len(cache) >= maxsize:
+                    cache.popitem(0)    # purge least recently used cache entry
+            cache[key] = result         # record recent use of this key
+            return result
+
+        def clear():
+            'Clear the cache and cache statistics'
+            cache.clear()
+            wrapper.hits = wrapper.misses = 0
+
+        wrapper.hits = wrapper.misses = 0
+        wrapper.clear = clear
+        return wrapper
+    return decorating_function

Modified: python/branches/py3k/Lib/test/test_functools.py
==============================================================================
--- python/branches/py3k/Lib/test/test_functools.py	(original)
+++ python/branches/py3k/Lib/test/test_functools.py	Sat Jul 31 20:09:23 2010
@@ -4,6 +4,7 @@
 from test import support
 from weakref import proxy
 import pickle
+from random import choice
 
 @staticmethod
 def PythonPartial(func, *args, **keywords):
@@ -454,6 +455,50 @@
             class A:
                 pass
 
+class TestLRU(unittest.TestCase):
+
+    def test_lru(self):
+        def orig(x, y):
+            return 3*x+y
+        f = functools.lru_cache(maxsize=20)(orig)
+
+        domain = range(5)
+        for i in range(1000):
+            x, y = choice(domain), choice(domain)
+            actual = f(x, y)
+            expected = orig(x, y)
+            self.assertEquals(actual, expected)
+        self.assert_(f.hits > f.misses)
+        self.assertEquals(f.hits + f.misses, 1000)
+
+        f.clear()   # test clearing
+        self.assertEqual(f.hits, 0)
+        self.assertEqual(f.misses, 0)
+        f(x, y)
+        self.assertEqual(f.hits, 0)
+        self.assertEqual(f.misses, 1)
+
+    def test_lfu(self):
+        def orig(x, y):
+            return 3*x+y
+        f = functools.lfu_cache(maxsize=20)(orig)
+
+        domain = range(5)
+        for i in range(1000):
+            x, y = choice(domain), choice(domain)
+            actual = f(x, y)
+            expected = orig(x, y)
+            self.assertEquals(actual, expected)
+        self.assert_(f.hits > f.misses)
+        self.assertEquals(f.hits + f.misses, 1000)
+
+        f.clear()   # test clearing
+        self.assertEqual(f.hits, 0)
+        self.assertEqual(f.misses, 0)
+        f(x, y)
+        self.assertEqual(f.hits, 0)
+        self.assertEqual(f.misses, 1)
+
 def test_main(verbose=None):
     test_classes = (
         TestPartial,
@@ -461,7 +506,8 @@
         TestPythonPartial,
         TestUpdateWrapper,
         TestWraps,
-        TestReduce
+        TestReduce,
+        TestLRU,
     )
     support.run_unittest(*test_classes)
 

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sat Jul 31 20:09:23 2010
@@ -516,6 +516,8 @@
 - Issue #4179: In pdb, allow "list ." as a command to return to the currently
   debugged line.
 
+- Add lfu_cache() and lru_cache() decorators to the functools module.
+
 - Issue #4108: In urllib.robotparser, if there are multiple 'User-agent: *'
   entries, consider the first one.
 

From python-checkins at python.org  Sat Jul 31 20:09:46 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 20:09:46 +0200 (CEST)
Subject: [Python-checkins] r83351 - python/branches/py3k/Misc/NEWS
Message-ID: <20100731180946.5FCDBEE994@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 20:09:46 2010
New Revision: 83351

Log:
Move news item to the correct position.

Modified:
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sat Jul 31 20:09:46 2010
@@ -15,6 +15,8 @@
 Library
 -------
 
+- Add lfu_cache() and lru_cache() decorators to the functools module.
+
 
 What's New in Python 3.2 Alpha 1?
 =================================
@@ -516,8 +518,6 @@
 - Issue #4179: In pdb, allow "list ." as a command to return to the currently
   debugged line.
 
-- Add lfu_cache() and lru_cache() decorators to the functools module.
-
 - Issue #4108: In urllib.robotparser, if there are multiple 'User-agent: *'
   entries, consider the first one.
 

From python-checkins at python.org  Sat Jul 31 20:11:07 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 20:11:07 +0200 (CEST)
Subject: [Python-checkins] r83352 -
	python/branches/py3k/Lib/test/datetimetester.py
Message-ID: <20100731181107.D29D4EEB1B@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 20:11:07 2010
New Revision: 83352

Log:
#9440: Remove borderline test case that fails based on unpredictable conditions such as compiler flags.

Modified:
   python/branches/py3k/Lib/test/datetimetester.py

Modified: python/branches/py3k/Lib/test/datetimetester.py
==============================================================================
--- python/branches/py3k/Lib/test/datetimetester.py	(original)
+++ python/branches/py3k/Lib/test/datetimetester.py	Sat Jul 31 20:11:07 2010
@@ -1733,10 +1733,6 @@
                          self.theclass.fromtimestamp(1))
         self.assertEqual(self.theclass.fromtimestamp(0.99999949).microsecond,
                          999999)
-        # XXX Arguably incorrect behavior.  Since round(0.6112295, 6)
-        # returns 0.611229, we should see 611229 us below, not 611230
-        self.assertEqual(self.theclass.fromtimestamp(0.6112295).microsecond,
-                         611230)
 
     def test_insane_fromtimestamp(self):
         # It's possible that some platform maps time_t to double,

From python-checkins at python.org  Sat Jul 31 20:59:20 2010
From: python-checkins at python.org (martin.v.loewis)
Date: Sat, 31 Jul 2010 20:59:20 +0200 (CEST)
Subject: [Python-checkins] r83353 - python/branches/py3k/Tools/msi/msi.py
Message-ID: <20100731185920.75555EEB21@mail.python.org>

Author: martin.v.loewis
Date: Sat Jul 31 20:59:20 2010
New Revision: 83353

Log:
Drop webchecker and BerkeleyDB license.

Modified:
   python/branches/py3k/Tools/msi/msi.py

Modified: python/branches/py3k/Tools/msi/msi.py
==============================================================================
--- python/branches/py3k/Tools/msi/msi.py	(original)
+++ python/branches/py3k/Tools/msi/msi.py	Sat Jul 31 20:59:20 2010
@@ -879,7 +879,6 @@
     shutil.copyfileobj(open(os.path.join(srcdir, "LICENSE")), out)
     shutil.copyfileobj(open("crtlicense.txt"), out)
     for name, pat, file in (("bzip2","bzip2-*", "LICENSE"),
-                      ("Berkeley DB", "db-*", "LICENSE"),
                       ("openssl", "openssl-*", "LICENSE"),
                       ("Tcl", "tcl8*", "license.terms"),
                       ("Tk", "tk8*", "license.terms"),
@@ -1122,7 +1121,7 @@
     # Add tools
     tools.set_current()
     tooldir = PyDirectory(db, cab, root, "Tools", "Tools", "TOOLS|Tools")
-    for f in ['i18n', 'pynche', 'Scripts', 'versioncheck', 'webchecker']:
+    for f in ['i18n', 'pynche', 'Scripts', 'versioncheck']:
         lib = PyDirectory(db, cab, tooldir, f, f, "%s|%s" % (tooldir.make_short(f), f))
         lib.glob("*.py")
         lib.glob("*.pyw", exclude=['pydocgui.pyw'])

From python-checkins at python.org  Sat Jul 31 21:07:37 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 21:07:37 +0200 (CEST)
Subject: [Python-checkins] r83354 -
	python/branches/release27-maint/Objects/stringobject.c
Message-ID: <20100731190737.BCEC6EE994@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 21:07:37 2010
New Revision: 83354

Log:
#9328: string format methods return strings.

Modified:
   python/branches/release27-maint/Objects/stringobject.c

Modified: python/branches/release27-maint/Objects/stringobject.c
==============================================================================
--- python/branches/release27-maint/Objects/stringobject.c	(original)
+++ python/branches/release27-maint/Objects/stringobject.c	Sat Jul 31 21:07:37 2010
@@ -3580,7 +3580,7 @@
 #include "stringlib/string_format.h"
 
 PyDoc_STRVAR(format__doc__,
-"S.format(*args, **kwargs) -> unicode\n\
+"S.format(*args, **kwargs) -> string\n\
 \n\
 ");
 
@@ -3614,7 +3614,7 @@
 }
 
 PyDoc_STRVAR(p_format__doc__,
-"S.__format__(format_spec) -> unicode\n\
+"S.__format__(format_spec) -> string\n\
 \n\
 ");
 

From python-checkins at python.org  Sat Jul 31 21:17:11 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 21:17:11 +0200 (CEST)
Subject: [Python-checkins] r83355 - python/branches/py3k/Lib/test/test_tcl.py
Message-ID: <20100731191711.ED457F2F7@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 21:17:11 2010
New Revision: 83355

Log:
Fix bad merge: test_support -> support.

Modified:
   python/branches/py3k/Lib/test/test_tcl.py

Modified: python/branches/py3k/Lib/test/test_tcl.py
==============================================================================
--- python/branches/py3k/Lib/test/test_tcl.py	(original)
+++ python/branches/py3k/Lib/test/test_tcl.py	Sat Jul 31 21:17:11 2010
@@ -143,7 +143,7 @@
                                     fullname[0],
                                     fullname[3:])
 
-        with test_support.EnvironmentVarGuard() as env:
+        with support.EnvironmentVarGuard() as env:
             env.unset("TCL_LIBRARY")
             f = os.popen('%s -c "import Tkinter; print Tkinter"' % (unc_name,))
 

From python-checkins at python.org  Sat Jul 31 21:29:15 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 21:29:15 +0200 (CEST)
Subject: [Python-checkins] r83356 - python/branches/py3k/Python/getversion.c
Message-ID: <20100731192915.9FC6DEEA7B@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 21:29:15 2010
New Revision: 83356

Log:
Remove trailing whitespace.

Modified:
   python/branches/py3k/Python/getversion.c

Modified: python/branches/py3k/Python/getversion.c
==============================================================================
--- python/branches/py3k/Python/getversion.c	(original)
+++ python/branches/py3k/Python/getversion.c	Sat Jul 31 21:29:15 2010
@@ -9,7 +9,7 @@
 Py_GetVersion(void)
 {
 	static char version[250];
-	PyOS_snprintf(version, sizeof(version), "%.80s (%.80s) %.80s", 
+	PyOS_snprintf(version, sizeof(version), "%.80s (%.80s) %.80s",
 		      PY_VERSION, Py_GetBuildInfo(), Py_GetCompiler());
 	return version;
 }

From python-checkins at python.org  Sat Jul 31 21:59:55 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 21:59:55 +0200 (CEST)
Subject: [Python-checkins] r83357 - python/branches/py3k/Doc/library/sys.rst
Message-ID: <20100731195955.4D982EE994@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 21:59:55 2010
New Revision: 83357

Log:
#5778: document that sys.version can contain a newline.

Modified:
   python/branches/py3k/Doc/library/sys.rst

Modified: python/branches/py3k/Doc/library/sys.rst
==============================================================================
--- python/branches/py3k/Doc/library/sys.rst	(original)
+++ python/branches/py3k/Doc/library/sys.rst	Sat Jul 31 21:59:55 2010
@@ -936,8 +936,9 @@
 
    A string containing the version number of the Python interpreter plus additional
    information on the build number and compiler used. It has a value of the form
-   ``'version (#build_number, build_date, build_time) [compiler]'``.  The first
-   three characters are used to identify the version in the installation
+   ``'version (#build_number, build_date, build_time) [compiler]'``.  There may be
+   a newline character preceding the opening bracket with of the compiler info.
+   The first three characters are used to identify the version in the installation
    directories (where appropriate on each platform).  An example::
 
       >>> import sys

From python-checkins at python.org  Sat Jul 31 22:05:32 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 22:05:32 +0200 (CEST)
Subject: [Python-checkins] r83358 - python/branches/py3k/Doc/library/sys.rst
Message-ID: <20100731200532.17BD4EE994@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 22:05:31 2010
New Revision: 83358

Log:
#9442: do not document a specific format for sys.version; rather refer to version_info and the platform module.

Modified:
   python/branches/py3k/Doc/library/sys.rst

Modified: python/branches/py3k/Doc/library/sys.rst
==============================================================================
--- python/branches/py3k/Doc/library/sys.rst	(original)
+++ python/branches/py3k/Doc/library/sys.rst	Sat Jul 31 22:05:31 2010
@@ -935,15 +935,10 @@
 .. data:: version
 
    A string containing the version number of the Python interpreter plus additional
-   information on the build number and compiler used. It has a value of the form
-   ``'version (#build_number, build_date, build_time) [compiler]'``.  There may be
-   a newline character preceding the opening bracket with of the compiler info.
-   The first three characters are used to identify the version in the installation
-   directories (where appropriate on each platform).  An example::
-
-      >>> import sys
-      >>> sys.version
-      '1.5.2 (#0 Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)]'
+   information on the build number and compiler used.  This string is displayed
+   when the interactive interpreter is started.  Do not extract version information
+   out of it, rather, use :data:`version_info` and the functions provided by the
+   :mod:`platform` module.
 
 
 .. data:: api_version

From python-checkins at python.org  Sat Jul 31 22:08:16 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 22:08:16 +0200 (CEST)
Subject: [Python-checkins] r83359 - in python/branches/py3k:
	Doc/library/fileinput.rst Lib/fileinput.py
	Lib/test/test_fileinput.py Misc/NEWS
Message-ID: <20100731200816.19200EEB3F@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 22:08:15 2010
New Revision: 83359

Log:
#1286: allow using fileinput.FileInput as context manager.

Modified:
   python/branches/py3k/Doc/library/fileinput.rst
   python/branches/py3k/Lib/fileinput.py
   python/branches/py3k/Lib/test/test_fileinput.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Doc/library/fileinput.rst
==============================================================================
--- python/branches/py3k/Doc/library/fileinput.rst	(original)
+++ python/branches/py3k/Doc/library/fileinput.rst	Sat Jul 31 22:08:15 2010
@@ -24,7 +24,7 @@
 
 All files are opened in text mode by default, but you can override this by
 specifying the *mode* parameter in the call to :func:`.input` or
-:class:`FileInput()`.  If an I/O error occurs during opening or reading a file,
+:class:`FileInput`.  If an I/O error occurs during opening or reading a file,
 :exc:`IOError` is raised.
 
 If ``sys.stdin`` is used more than once, the second and further use will return
@@ -54,6 +54,16 @@
    during iteration.  The parameters to this function will be passed along to the
    constructor of the :class:`FileInput` class.
 
+   The :class:`FileInput` instance can be used as a context manager in the
+   :keyword:`with` statement.  In this example, *input* is closed after the
+   :keyword:`with` statement is exited, even if an exception occurs::
+
+      with fileinput.input(files=('spam.txt', 'eggs.txt')) as input:
+          process(input)
+
+   .. versionchanged:: 3.2
+      Can be used as a context manager.
+
 
 The following functions use the global state created by :func:`fileinput.input`;
 if there is no active state, :exc:`RuntimeError` is raised.
@@ -132,13 +142,23 @@
    *filename* and *mode*, and returns an accordingly opened file-like object. You
    cannot use *inplace* and *openhook* together.
 
-
-**Optional in-place filtering:** if the keyword argument ``inplace=1`` is passed
-to :func:`fileinput.input` or to the :class:`FileInput` constructor, the file is
-moved to a backup file and standard output is directed to the input file (if a
-file of the same name as the backup file already exists, it will be replaced
-silently).  This makes it possible to write a filter that rewrites its input
-file in place.  If the *backup* parameter is given (typically as
+   A :class:`FileInput` instance can be used as a context manager in the
+   :keyword:`with` statement.  In this example, *input* is closed after the
+   :keyword:`with` statement is exited, even if an exception occurs::
+
+      with FileInput(files=('spam.txt', 'eggs.txt')) as input:
+          process(input)
+
+   .. versionchanged:: 3.2
+      Can be used as a context manager.
+
+
+**Optional in-place filtering:** if the keyword argument ``inplace=True`` is
+passed to :func:`fileinput.input` or to the :class:`FileInput` constructor, the
+file is moved to a backup file and standard output is directed to the input file
+(if a file of the same name as the backup file already exists, it will be
+replaced silently).  This makes it possible to write a filter that rewrites its
+input file in place.  If the *backup* parameter is given (typically as
 ``backup='.<some extension>'``), it specifies the extension for the backup file,
 and the backup file remains around; by default, the extension is ``'.bak'`` and
 it is deleted when the output file is closed.  In-place filtering is disabled

Modified: python/branches/py3k/Lib/fileinput.py
==============================================================================
--- python/branches/py3k/Lib/fileinput.py	(original)
+++ python/branches/py3k/Lib/fileinput.py	Sat Jul 31 22:08:15 2010
@@ -238,6 +238,12 @@
         self.nextfile()
         self._files = ()
 
+    def __enter__(self):
+        return self
+
+    def __exit__(self, type, value, traceback):
+        self.close()
+
     def __iter__(self):
         return self
 

Modified: python/branches/py3k/Lib/test/test_fileinput.py
==============================================================================
--- python/branches/py3k/Lib/test/test_fileinput.py	(original)
+++ python/branches/py3k/Lib/test/test_fileinput.py	Sat Jul 31 22:08:15 2010
@@ -231,6 +231,30 @@
 ##         finally:
 ##             remove_tempfiles(t1)
 
+    def test_context_manager(self):
+        try:
+            t1 = writeTmp(1, ["A\nB\nC"])
+            t2 = writeTmp(2, ["D\nE\nF"])
+            with FileInput(files=(t1, t2)) as fi:
+                lines = list(fi)
+            self.assertEqual(lines, ["A\n", "B\n", "C", "D\n", "E\n", "F"])
+            self.assertEqual(fi.filelineno(), 3)
+            self.assertEqual(fi.lineno(), 6)
+            self.assertEqual(fi._files, ())
+        finally:
+            remove_tempfiles(t1, t2)
+
+    def test_close_on_exception(self):
+        try:
+            t1 = writeTmp(1, [""])
+            with FileInput(files=t1) as fi:
+                raise IOError
+        except IOError:
+            self.assertEqual(fi._files, ())
+        finally:
+            remove_tempfiles(t1)
+
+
 def test_main():
     run_unittest(BufferSizesTests, FileInputTests)
 

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sat Jul 31 22:08:15 2010
@@ -15,6 +15,8 @@
 Library
 -------
 
+- Issue #1286: Allow using fileinput.FileInput as a context manager.
+
 - Add lfu_cache() and lru_cache() decorators to the functools module.
 
 

From python-checkins at python.org  Sat Jul 31 22:13:44 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 22:13:44 +0200 (CEST)
Subject: [Python-checkins] r83360 -
	python/branches/py3k/Lib/test/test_cfgparser.py
Message-ID: <20100731201344.C6C42EE994@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 22:13:44 2010
New Revision: 83360

Log:
Clarify comment in comments test case explaining comment semantics.

Modified:
   python/branches/py3k/Lib/test/test_cfgparser.py

Modified: python/branches/py3k/Lib/test/test_cfgparser.py
==============================================================================
--- python/branches/py3k/Lib/test/test_cfgparser.py	(original)
+++ python/branches/py3k/Lib/test/test_cfgparser.py	Sat Jul 31 22:13:44 2010
@@ -676,7 +676,7 @@
         # but this is a comment
         ; another comment
         quirk: this;is not a comment
-        ; a space must precede a comment character
+        ; a space must precede an inline comment
         """)
         cf = self.fromstring(config_string)
         self.assertEqual(cf.get('Commented Bar', 'foo'), 'bar # not a comment!')

From python-checkins at python.org  Sat Jul 31 23:04:00 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 23:04:00 +0200 (CEST)
Subject: [Python-checkins] r83361 - in python/branches/py3k/Lib:
	http/cookies.py test/test_http_cookies.py
Message-ID: <20100731210400.4B4C9EEB3A@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 23:04:00 2010
New Revision: 83361

Log:
#3788: more tests for http.cookies, now at 95% coverage.  Also bring coding style in the module up to PEP 8, where it does not break backwards compatibility.

Modified:
   python/branches/py3k/Lib/http/cookies.py
   python/branches/py3k/Lib/test/test_http_cookies.py

Modified: python/branches/py3k/Lib/http/cookies.py
==============================================================================
--- python/branches/py3k/Lib/http/cookies.py	(original)
+++ python/branches/py3k/Lib/http/cookies.py	Sat Jul 31 23:04:00 2010
@@ -46,7 +46,7 @@
 The Basics
 ----------
 
-Importing is easy..
+Importing is easy...
 
    >>> from http import cookies
 
@@ -127,19 +127,14 @@
    'Set-Cookie: number=7\r\nSet-Cookie: string=seven'
 
 Finis.
-"""  #"
-#     ^
-#     |----helps out font-lock
+"""
 
 #
 # Import our required modules
 #
+import re
 import string
 
-from pickle import dumps, loads
-
-import re, warnings
-
 __all__ = ["CookieError", "BaseCookie", "SimpleCookie"]
 
 _nulljoin = ''.join
@@ -235,7 +230,7 @@
     if all(c in LegalChars for c in str):
         return str
     else:
-        return '"' + _nulljoin( map(_Translator.get, str, str) ) + '"'
+        return '"' + _nulljoin(map(_Translator.get, str, str)) + '"'
 
 
 _OctalPatt = re.compile(r"\\[0-3][0-7][0-7]")
@@ -244,7 +239,7 @@
 def _unquote(str):
     # If there aren't any doublequotes,
     # then there can't be any special characters.  See RFC 2109.
-    if  len(str) < 2:
+    if len(str) < 2:
         return str
     if str[0] != '"' or str[-1] != '"':
         return str
@@ -263,31 +258,32 @@
     n = len(str)
     res = []
     while 0 <= i < n:
-        Omatch = _OctalPatt.search(str, i)
-        Qmatch = _QuotePatt.search(str, i)
-        if not Omatch and not Qmatch:              # Neither matched
+        o_match = _OctalPatt.search(str, i)
+        q_match = _QuotePatt.search(str, i)
+        if not o_match and not q_match:              # Neither matched
             res.append(str[i:])
             break
         # else:
         j = k = -1
-        if Omatch: j = Omatch.start(0)
-        if Qmatch: k = Qmatch.start(0)
-        if Qmatch and ( not Omatch or k < j ):     # QuotePatt matched
+        if o_match:
+            j = o_match.start(0)
+        if q_match:
+            k = q_match.start(0)
+        if q_match and (not o_match or k < j):     # QuotePatt matched
             res.append(str[i:k])
             res.append(str[k+1])
-            i = k+2
+            i = k + 2
         else:                                      # OctalPatt matched
             res.append(str[i:j])
-            res.append( chr( int(str[j+1:j+4], 8) ) )
-            i = j+4
+            res.append(chr(int(str[j+1:j+4], 8)))
+            i = j + 4
     return _nulljoin(res)
 
-# The _getdate() routine is used to set the expiration time in
-# the cookie's HTTP header.      By default, _getdate() returns the
-# current time in the appropriate "expires" format for a
-# Set-Cookie header.     The one optional argument is an offset from
-# now, in seconds.      For example, an offset of -3600 means "one hour ago".
-# The offset may be a floating point number.
+# The _getdate() routine is used to set the expiration time in the cookie's HTTP
+# header.  By default, _getdate() returns the current time in the appropriate
+# "expires" format for a Set-Cookie header.  The one optional argument is an
+# offset from now, in seconds.  For example, an offset of -3600 means "one hour
+# ago".  The offset may be a floating point number.
 #
 
 _weekdayname = ['Mon', 'Tue', 'Wed', 'Thu', 'Fri', 'Sat', 'Sun']
@@ -305,7 +301,7 @@
 
 
 class Morsel(dict):
-    """A class to hold ONE key,value pair.
+    """A class to hold ONE (key, value) pair.
 
     In a cookie, each such pair may have several attributes, so this class is
     used to keep the attributes associated with the appropriate key,value pair.
@@ -326,23 +322,24 @@
     # This dictionary provides a mapping from the lowercase
     # variant on the left to the appropriate traditional
     # formatting on the right.
-    _reserved = { "expires" : "expires",
-                   "path"        : "Path",
-                   "comment" : "Comment",
-                   "domain"      : "Domain",
-                   "max-age" : "Max-Age",
-                   "secure"      : "secure",
-                   "httponly"  : "httponly",
-                   "version" : "Version",
-                   }
+    _reserved = {
+        "expires"  : "expires",
+        "path"     : "Path",
+        "comment"  : "Comment",
+        "domain"   : "Domain",
+        "max-age"  : "Max-Age",
+        "secure"   : "secure",
+        "httponly" : "httponly",
+        "version"  : "Version",
+    }
 
     def __init__(self):
         # Set defaults
         self.key = self.value = self.coded_value = None
 
         # Set default attributes
-        for K in self._reserved:
-            dict.__setitem__(self, K, "")
+        for key in self._reserved:
+            dict.__setitem__(self, key, "")
 
     def __setitem__(self, K, V):
         K = K.lower()
@@ -362,18 +359,18 @@
             raise CookieError("Illegal key value: %s" % key)
 
         # It's a good key, so save it.
-        self.key                 = key
-        self.value               = val
-        self.coded_value         = coded_val
+        self.key = key
+        self.value = val
+        self.coded_value = coded_val
 
-    def output(self, attrs=None, header = "Set-Cookie:"):
-        return "%s %s" % ( header, self.OutputString(attrs) )
+    def output(self, attrs=None, header="Set-Cookie:"):
+        return "%s %s" % (header, self.OutputString(attrs))
 
     __str__ = output
 
     def __repr__(self):
         return '<%s: %s=%s>' % (self.__class__.__name__,
-                                self.key, repr(self.value) )
+                                self.key, repr(self.value))
 
     def js_output(self, attrs=None):
         # Print javascript
@@ -383,34 +380,36 @@
         document.cookie = \"%s\";
         // end hiding -->
         </script>
-        """ % ( self.OutputString(attrs).replace('"',r'\"'))
+        """ % (self.OutputString(attrs).replace('"', r'\"'))
 
     def OutputString(self, attrs=None):
         # Build up our result
         #
         result = []
-        RA = result.append
+        append = result.append
 
         # First, the key=value pair
-        RA("%s=%s" % (self.key, self.coded_value))
+        append("%s=%s" % (self.key, self.coded_value))
 
         # Now add any defined attributes
         if attrs is None:
             attrs = self._reserved
         items = sorted(self.items())
-        for K,V in items:
-            if V == "": continue
-            if K not in attrs: continue
-            if K == "expires" and type(V) == type(1):
-                RA("%s=%s" % (self._reserved[K], _getdate(V)))
-            elif K == "max-age" and type(V) == type(1):
-                RA("%s=%d" % (self._reserved[K], V))
-            elif K == "secure":
-                RA(str(self._reserved[K]))
-            elif K == "httponly":
-                RA(str(self._reserved[K]))
+        for key, value in items:
+            if value == "":
+                continue
+            if key not in attrs:
+                continue
+            if key == "expires" and isinstance(value, int):
+                append("%s=%s" % (self._reserved[key], _getdate(value)))
+            elif key == "max-age" and isinstance(value, int):
+                append("%s=%d" % (self._reserved[key], value))
+            elif key == "secure":
+                append(str(self._reserved[key]))
+            elif key == "httponly":
+                append(str(self._reserved[key]))
             else:
-                RA("%s=%s" % (self._reserved[K], V))
+                append("%s=%s" % (self._reserved[key], value))
 
         # Return the result
         return _semispacejoin(result)
@@ -426,24 +425,23 @@
 #
 
 _LegalCharsPatt  = r"[\w\d!#%&'~_`><@,:/\$\*\+\-\.\^\|\)\(\?\}\{\=]"
-_CookiePattern = re.compile(
-    r"(?x)"                       # This is a Verbose pattern
-    r"(?P<key>"                   # Start of group 'key'
-    ""+ _LegalCharsPatt +"+?"     # Any word of at least one letter, nongreedy
-    r")"                          # End of group 'key'
-    r"\s*=\s*"                    # Equal Sign
-    r"(?P<val>"                   # Start of group 'val'
-    r'"(?:[^\\"]|\\.)*"'            # Any doublequoted string
-    r"|"                            # or
-    ""+ _LegalCharsPatt +"*"        # Any word or empty string
-    r")"                          # End of group 'val'
-    r"\s*;?"                      # Probably ending in a semi-colon
-    , re.ASCII)                   # May be removed if safe.
-
-
-# At long last, here is the cookie class.
-#   Using this class is almost just like using a dictionary.
-# See this module's docstring for example usage.
+_CookiePattern = re.compile(r"""
+    (?x)                           # This is a verbose pattern
+    (?P<key>                       # Start of group 'key'
+    """ + _LegalCharsPatt + r"""+?   # Any word of at least one letter
+    )                              # End of group 'key'
+    \s*=\s*                        # Equal Sign
+    (?P<val>                       # Start of group 'val'
+    "(?:[^\\"]|\\.)*"                # Any doublequoted string
+    |                                # or
+    """ + _LegalCharsPatt + r"""*    # Any word or empty string
+    )                              # End of group 'val'
+    \s*;?                          # Probably ending in a semi-colon
+    """, re.ASCII)                 # May be removed if safe.
+
+
+# At long last, here is the cookie class.  Using this class is almost just like
+# using a dictionary.  See this module's docstring for example usage.
 #
 class BaseCookie(dict):
     """A container class for a set of Morsels."""
@@ -467,7 +465,8 @@
         return strval, strval
 
     def __init__(self, input=None):
-        if input: self.load(input)
+        if input:
+            self.load(input)
 
     def __set(self, key, real_value, coded_value):
         """Private method for setting a cookie's value"""
@@ -484,25 +483,25 @@
         """Return a string suitable for HTTP."""
         result = []
         items = sorted(self.items())
-        for K,V in items:
-            result.append( V.output(attrs, header) )
+        for key, value in items:
+            result.append(value.output(attrs, header))
         return sep.join(result)
 
     __str__ = output
 
     def __repr__(self):
-        L = []
+        l = []
         items = sorted(self.items())
-        for K,V in items:
-            L.append( '%s=%s' % (K,repr(V.value) ) )
-        return '<%s: %s>' % (self.__class__.__name__, _spacejoin(L))
+        for key, value in items:
+            l.append('%s=%s' % (key, repr(value.value)))
+        return '<%s: %s>' % (self.__class__.__name__, _spacejoin(l))
 
     def js_output(self, attrs=None):
         """Return a string suitable for JavaScript."""
         result = []
         items = sorted(self.items())
-        for K,V in items:
-            result.append( V.js_output(attrs) )
+        for key, value in items:
+            result.append(value.js_output(attrs))
         return _nulljoin(result)
 
     def load(self, rawdata):
@@ -511,15 +510,15 @@
         is equivalent to calling:
             map(Cookie.__setitem__, d.keys(), d.values())
         """
-        if type(rawdata) == type(""):
-            self.__ParseString(rawdata)
+        if isinstance(rawdata, str):
+            self.__parse_string(rawdata)
         else:
             # self.update() wouldn't call our custom __setitem__
-            for k, v in rawdata.items():
-                self[k] = v
+            for key, value in rawdata.items():
+                self[key] = value
         return
 
-    def __ParseString(self, str, patt=_CookiePattern):
+    def __parse_string(self, str, patt=_CookiePattern):
         i = 0            # Our starting point
         n = len(str)     # Length of string
         M = None         # current morsel
@@ -527,25 +526,27 @@
         while 0 <= i < n:
             # Start looking for a cookie
             match = patt.search(str, i)
-            if not match: break          # No more cookies
+            if not match:
+                # No more cookies
+                break
 
-            K,V = match.group("key"), match.group("val")
+            key, value = match.group("key"), match.group("val")
             i = match.end(0)
 
             # Parse the key, value in case it's metainfo
-            if K[0] == "$":
+            if key[0] == "$":
                 # We ignore attributes which pertain to the cookie
                 # mechanism as a whole.  See RFC 2109.
                 # (Does anyone care?)
                 if M:
-                    M[ K[1:] ] = V
-            elif K.lower() in Morsel._reserved:
+                    M[key[1:]] = value
+            elif key.lower() in Morsel._reserved:
                 if M:
-                    M[ K ] = _unquote(V)
+                    M[key] = _unquote(value)
             else:
-                rval, cval = self.value_decode(V)
-                self.__set(K, rval, cval)
-                M = self[K]
+                rval, cval = self.value_decode(value)
+                self.__set(key, rval, cval)
+                M = self[key]
 
 
 class SimpleCookie(BaseCookie):
@@ -556,16 +557,8 @@
     received from HTTP are kept as strings.
     """
     def value_decode(self, val):
-        return _unquote( val ), val
+        return _unquote(val), val
+
     def value_encode(self, val):
         strval = str(val)
-        return strval, _quote( strval )
-
-###########################################################
-
-def _test():
-    import doctest, http.cookies
-    return doctest.testmod(http.cookies)
-
-if __name__ == "__main__":
-    _test()
+        return strval, _quote(strval)

Modified: python/branches/py3k/Lib/test/test_http_cookies.py
==============================================================================
--- python/branches/py3k/Lib/test/test_http_cookies.py	(original)
+++ python/branches/py3k/Lib/test/test_http_cookies.py	Sat Jul 31 23:04:00 2010
@@ -19,24 +19,21 @@
 
     def test_basic(self):
         cases = [
-            { 'data': 'chips=ahoy; vienna=finger',
-              'dict': {'chips':'ahoy', 'vienna':'finger'},
-              'repr': "<SimpleCookie: chips='ahoy' vienna='finger'>",
-              'output': 'Set-Cookie: chips=ahoy\nSet-Cookie: vienna=finger',
-            },
-
-            { 'data': 'keebler="E=mc2; L=\\"Loves\\"; fudge=\\012;"',
-              'dict': {'keebler' : 'E=mc2; L="Loves"; fudge=\012;'},
-              'repr': '''<SimpleCookie: keebler='E=mc2; L="Loves"; fudge=\\n;'>''',
-              'output': 'Set-Cookie: keebler="E=mc2; L=\\"Loves\\"; fudge=\\012;"',
-            },
+            {'data': 'chips=ahoy; vienna=finger',
+             'dict': {'chips':'ahoy', 'vienna':'finger'},
+             'repr': "<SimpleCookie: chips='ahoy' vienna='finger'>",
+             'output': 'Set-Cookie: chips=ahoy\nSet-Cookie: vienna=finger'},
+
+            {'data': 'keebler="E=mc2; L=\\"Loves\\"; fudge=\\012;"',
+             'dict': {'keebler' : 'E=mc2; L="Loves"; fudge=\012;'},
+             'repr': '''<SimpleCookie: keebler='E=mc2; L="Loves"; fudge=\\n;'>''',
+             'output': 'Set-Cookie: keebler="E=mc2; L=\\"Loves\\"; fudge=\\012;"'},
 
             # Check illegal cookies that have an '=' char in an unquoted value
-            { 'data': 'keebler=E=mc2',
-              'dict': {'keebler' : 'E=mc2'},
-              'repr': "<SimpleCookie: keebler='E=mc2'>",
-              'output': 'Set-Cookie: keebler=E=mc2',
-            }
+            {'data': 'keebler=E=mc2',
+             'dict': {'keebler' : 'E=mc2'},
+             'repr': "<SimpleCookie: keebler='E=mc2'>",
+             'output': 'Set-Cookie: keebler=E=mc2'},
         ]
 
         for case in cases:
@@ -72,6 +69,26 @@
         </script>
         """)
 
+    def test_special_attrs(self):
+        # 'expires'
+        C = cookies.SimpleCookie('Customer="WILE_E_COYOTE"')
+        C['Customer']['expires'] = 0
+        # can't test exact output, it always depends on current date/time
+        self.assertTrue(C.output().endswith('GMT'))
+
+        # 'max-age'
+        C = cookies.SimpleCookie('Customer="WILE_E_COYOTE"')
+        C['Customer']['max-age'] = 10
+        self.assertEqual(C.output(),
+                         'Set-Cookie: Customer="WILE_E_COYOTE"; Max-Age=10')
+
+        # others
+        C = cookies.SimpleCookie('Customer="WILE_E_COYOTE"')
+        C['Customer']['secure'] = True
+        C['Customer']['httponly'] = True
+        self.assertEqual(C.output(),
+            'Set-Cookie: Customer="WILE_E_COYOTE"; httponly; secure')
+
     def test_quoted_meta(self):
         # Try cookie with quoted meta-data
         C = cookies.SimpleCookie()
@@ -80,8 +97,72 @@
         self.assertEqual(C['Customer']['version'], '1')
         self.assertEqual(C['Customer']['path'], '/acme')
 
+        self.assertEqual(C.output(['path']),
+                         'Set-Cookie: Customer="WILE_E_COYOTE"; Path=/acme')
+        self.assertEqual(C.js_output(), r"""
+        <script type="text/javascript">
+        <!-- begin hiding
+        document.cookie = "Customer=\"WILE_E_COYOTE\"; Path=/acme; Version=1";
+        // end hiding -->
+        </script>
+        """)
+        self.assertEqual(C.js_output(['path']), r"""
+        <script type="text/javascript">
+        <!-- begin hiding
+        document.cookie = "Customer=\"WILE_E_COYOTE\"; Path=/acme";
+        // end hiding -->
+        </script>
+        """)
+
+class MorselTests(unittest.TestCase):
+    """Tests for the Morsel object."""
+
+    def test_reserved_keys(self):
+        M = cookies.Morsel()
+        # tests valid and invalid reserved keys for Morsels
+        for i in M._reserved:
+            # Test that all valid keys are reported as reserved and set them
+            self.assertTrue(M.isReservedKey(i))
+            M[i] = '%s_value' % i
+        for i in M._reserved:
+            # Test that valid key values come out fine
+            self.assertEqual(M[i], '%s_value' % i)
+        for i in "the holy hand grenade".split():
+            # Test that invalid keys raise CookieError
+            self.assertRaises(cookies.CookieError,
+                              M.__setitem__, i, '%s_value' % i)
+
+    def test_setter(self):
+        M = cookies.Morsel()
+        # tests the .set method to set keys and their values
+        for i in M._reserved:
+            # Makes sure that all reserved keys can't be set this way
+            self.assertRaises(cookies.CookieError,
+                              M.set, i, '%s_value' % i, '%s_value' % i)
+        for i in "thou cast _the- !holy! ^hand| +*grenade~".split():
+            # Try typical use case. Setting decent values.
+            # Check output and js_output.
+            M['path'] = '/foo' # Try a reserved key as well
+            M.set(i, "%s_val" % i, "%s_coded_val" % i)
+            self.assertEqual(
+                M.output(),
+                "Set-Cookie: %s=%s; Path=/foo" % (i, "%s_coded_val" % i))
+            expected_js_output = """
+        <script type="text/javascript">
+        <!-- begin hiding
+        document.cookie = "%s=%s; Path=/foo";
+        // end hiding -->
+        </script>
+        """ % (i, "%s_coded_val" % i)
+            self.assertEqual(M.js_output(), expected_js_output)
+        for i in ["foo bar", "foo at bar"]:
+            # Try some illegal characters
+            self.assertRaises(cookies.CookieError,
+                              M.set, i, '%s_value' % i, '%s_value' % i)
+
+
 def test_main():
-    run_unittest(CookieTests)
+    run_unittest(CookieTests, MorselTests)
     run_doctest(cookies)
 
 if __name__ == '__main__':

From python-checkins at python.org  Sat Jul 31 23:12:15 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 23:12:15 +0200 (CEST)
Subject: [Python-checkins] r83362 - python/branches/py3k/Lib/test/data/README
Message-ID: <20100731211215.EA599EEAC9@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 23:12:15 2010
New Revision: 83362

Log:
#8910: add a file explaining why Lib/test/data is there.

Added:
   python/branches/py3k/Lib/test/data/README

Added: python/branches/py3k/Lib/test/data/README
==============================================================================
--- (empty file)
+++ python/branches/py3k/Lib/test/data/README	Sat Jul 31 23:12:15 2010
@@ -0,0 +1,2 @@
+This empty directory serves as destination for temporary files
+created by some tests.

From python-checkins at python.org  Sat Jul 31 23:21:40 2010
From: python-checkins at python.org (victor.stinner)
Date: Sat, 31 Jul 2010 23:21:40 +0200 (CEST)
Subject: [Python-checkins] r83363 -
	python/branches/import_unicode/Lib/test/test_xml_etree.py
Message-ID: <20100731212140.0ACE7EEA7B@mail.python.org>

Author: victor.stinner
Date: Sat Jul 31 23:21:39 2010
New Revision: 83363

Log:
fix a typo in a comment: undecodable => unencodable

Modified:
   python/branches/import_unicode/Lib/test/test_xml_etree.py

Modified: python/branches/import_unicode/Lib/test/test_xml_etree.py
==============================================================================
--- python/branches/import_unicode/Lib/test/test_xml_etree.py	(original)
+++ python/branches/import_unicode/Lib/test/test_xml_etree.py	Sat Jul 31 23:21:39 2010
@@ -24,7 +24,7 @@
 try:
     SIMPLE_XMLFILE.encode("utf8")
 except UnicodeEncodeError:
-    # ignore all tests if the current working directory is undecodable
+    # ignore all tests if the current working directory is unencodable
     raise unittest.SkipTest("filename is not encodable to utf8")
 SIMPLE_NS_XMLFILE = findfile("simple-ns.xml", subdir="xmltestdata")
 

From python-checkins at python.org  Sat Jul 31 23:21:41 2010
From: python-checkins at python.org (victor.stinner)
Date: Sat, 31 Jul 2010 23:21:41 +0200 (CEST)
Subject: [Python-checkins] r83364 -
	python/branches/import_unicode/Modules/main.c
Message-ID: <20100731212141.830EFEEA7B@mail.python.org>

Author: victor.stinner
Date: Sat Jul 31 23:21:41 2010
New Revision: 83364

Log:
catch PyUnicode_EncodeFSDefault() error

Create run_file() subfunction

Modified:
   python/branches/import_unicode/Modules/main.c

Modified: python/branches/import_unicode/Modules/main.c
==============================================================================
--- python/branches/import_unicode/Modules/main.c	(original)
+++ python/branches/import_unicode/Modules/main.c	Sat Jul 31 23:21:41 2010
@@ -279,6 +279,38 @@
     return 1;
 }
 
+static int
+run_file(FILE *fp, const wchar_t *filename, PyCompilerFlags *p_cf)
+{
+    PyObject *filenameObj, *filename_bytes = NULL;
+    char *filename_str = "<stdin>";
+    int sts;
+
+    /* call pending calls like signal handlers (SIGINT) */
+    if (Py_MakePendingCalls() == -1) {
+        PyErr_Print();
+        return 1;
+    }
+
+    if (filename) {
+        filenameObj = PyUnicode_FromWideChar(filename, wcslen(filename));
+        if (filenameObj != NULL) {
+            filename_bytes = PyUnicode_EncodeFSDefault(filenameObj);
+            Py_DECREF(filenameObj);
+        }
+        if (filename_bytes != NULL)
+            filename_str = PyBytes_AsString(filename_bytes);
+        else {
+            PyErr_Clear();
+            filename_str = "<decoding error>";
+        }
+    } else
+        filename_str = "<stdin>";
+
+    sts = PyRun_AnyFileExFlags(fp, filename_str, filename != NULL, p_cf) != 0;
+    Py_XDECREF(filename_bytes);
+    return sts;
+}
 
 /* Main program */
 
@@ -648,34 +680,8 @@
             }
         }
 
-        if (sts==-1) {
-            PyObject *filenameObj = NULL;
-            PyObject *filename_bytes = NULL;
-            char *p_cfilename = "<stdin>";
-            if (filename) {
-                filenameObj = PyUnicode_FromWideChar(
-                    filename, wcslen(filename));
-                if (filenameObj != NULL) {
-                    filename_bytes = PyUnicode_EncodeFSDefault(filenameObj);
-                    p_cfilename = PyBytes_AsString(filename_bytes);
-                }
-                else
-                    p_cfilename = "<decoding error>";
-            }
-            /* call pending calls like signal handlers (SIGINT) */
-            if (Py_MakePendingCalls() == -1) {
-                PyErr_Print();
-                sts = 1;
-            } else {
-                sts = PyRun_AnyFileExFlags(
-                    fp,
-                    p_cfilename,
-                    filename != NULL, &cf) != 0;
-            }
-            Py_XDECREF(filename_bytes);
-            Py_XDECREF(filenameObj);
-        }
-
+        if (sts==-1)
+            sts = run_file(fp, filename, &cf);
     }
 
     /* Check this environment variable at the end, to give programs the

From python-checkins at python.org  Sat Jul 31 23:22:36 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 23:22:36 +0200 (CEST)
Subject: [Python-checkins] r83365 - in python/branches/release27-maint:
	Lib/hotshot/log.py Lib/test/test_hotshot.py Misc/NEWS
Message-ID: <20100731212236.89783EEA7B@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 23:22:36 2010
New Revision: 83365

Log:
#1019882: if start() and stop() were not in the same stack frame, stats.load() would crash with IndexError.

Modified:
   python/branches/release27-maint/Lib/hotshot/log.py
   python/branches/release27-maint/Lib/test/test_hotshot.py
   python/branches/release27-maint/Misc/NEWS

Modified: python/branches/release27-maint/Lib/hotshot/log.py
==============================================================================
--- python/branches/release27-maint/Lib/hotshot/log.py	(original)
+++ python/branches/release27-maint/Lib/hotshot/log.py	Sat Jul 31 23:22:36 2010
@@ -106,7 +106,10 @@
                 return what, t, tdelta
 
             if what == WHAT_EXIT:
-                return what, self._pop(), tdelta
+                try:
+                    return what, self._pop(), tdelta
+                except IndexError:
+                    raise StopIteration
 
             if what == WHAT_LINENO:
                 filename, firstlineno, funcname = self._stack[-1]

Modified: python/branches/release27-maint/Lib/test/test_hotshot.py
==============================================================================
--- python/branches/release27-maint/Lib/test/test_hotshot.py	(original)
+++ python/branches/release27-maint/Lib/test/test_hotshot.py	Sat Jul 31 23:22:36 2010
@@ -10,6 +10,7 @@
 # Silence Py3k warning
 hotshot = test_support.import_module('hotshot', deprecated=True)
 from hotshot.log import ENTER, EXIT, LINE
+from hotshot import stats
 
 
 def shortfilename(fn):
@@ -136,6 +137,19 @@
             emptyfile.close()
         gc.collect()
 
+    def test_load_stats(self):
+        def start(prof):
+            prof.start()
+        # Make sure stats can be loaded when start and stop of profiler
+        # are not executed in the same stack frame.
+        profiler = self.new_profiler()
+        start(profiler)
+        profiler.stop()
+        profiler.close()
+        stats.load(self.logfn)
+        os.unlink(self.logfn)
+
+
 def test_main():
     test_support.run_unittest(HotShotTestCase)
 

Modified: python/branches/release27-maint/Misc/NEWS
==============================================================================
--- python/branches/release27-maint/Misc/NEWS	(original)
+++ python/branches/release27-maint/Misc/NEWS	Sat Jul 31 23:22:36 2010
@@ -64,6 +64,8 @@
 Extension Modules
 -----------------
 
+- Issue #1019882: Fix IndexError when loading certain hotshot stats.
+
 - Issue #9422: Fix memory leak when re-initializing a struct.Struct object.
 
 - Issue #7900: The getgroups(2) system call on MacOSX behaves rather oddly

From python-checkins at python.org  Sat Jul 31 23:26:40 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 23:26:40 +0200 (CEST)
Subject: [Python-checkins] r83366 -
	python/branches/py3k/Lib/test/test_optparse.py
Message-ID: <20100731212640.B46ABEE98B@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 23:26:40 2010
New Revision: 83366

Log:
There always is a False and True now.

Modified:
   python/branches/py3k/Lib/test/test_optparse.py

Modified: python/branches/py3k/Lib/test/test_optparse.py
==============================================================================
--- python/branches/py3k/Lib/test/test_optparse.py	(original)
+++ python/branches/py3k/Lib/test/test_optparse.py	Sat Jul 31 23:26:40 2010
@@ -781,15 +781,13 @@
         (options, args) = self.assertParseOK(["-q"],
                                              {'verbose': 0},
                                              [])
-        if hasattr(__builtins__, 'False'):
-            self.assertTrue(options.verbose is False)
+        self.assertTrue(options.verbose is False)
 
     def test_bool_true(self):
         (options, args) = self.assertParseOK(["-v"],
                                              {'verbose': 1},
                                              [])
-        if hasattr(__builtins__, 'True'):
-            self.assertTrue(options.verbose is True)
+        self.assertTrue(options.verbose is True)
 
     def test_bool_flicker_on_and_off(self):
         self.assertParseOK(["-qvq", "-q", "-v"],

From python-checkins at python.org  Sat Jul 31 23:33:40 2010
From: python-checkins at python.org (victor.stinner)
Date: Sat, 31 Jul 2010 23:33:40 +0200 (CEST)
Subject: [Python-checkins] r83367 -
	python/branches/import_unicode/Parser/tokenizer.c
Message-ID: <20100731213340.D15D6EE98B@mail.python.org>

Author: victor.stinner
Date: Sat Jul 31 23:33:40 2010
New Revision: 83367

Log:
PyObject_CallMethod(): use 'N' format

instead of 'O' format to avoid Py_DECREF().

Modified:
   python/branches/import_unicode/Parser/tokenizer.c

Modified: python/branches/import_unicode/Parser/tokenizer.c
==============================================================================
--- python/branches/import_unicode/Parser/tokenizer.c	(original)
+++ python/branches/import_unicode/Parser/tokenizer.c	Sat Jul 31 23:33:40 2010
@@ -471,9 +471,8 @@
         PyObject *filename = PyUnicode_DecodeFSDefault(tok->filename);
         if (filename == NULL)
             goto cleanup;
-        stream = PyObject_CallMethod(io, "open", "Osis",
+        stream = PyObject_CallMethod(io, "open", "Nsis",
                                      filename, "r", -1, enc);
-        Py_DECREF(filename);
     }
     else
         stream = PyObject_CallMethod(io, "open", "isisOOO",

From python-checkins at python.org  Sat Jul 31 23:40:15 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 23:40:15 +0200 (CEST)
Subject: [Python-checkins] r83368 - in python/branches/py3k: Lib/ntpath.py
	Lib/test/test_ntpath.py Misc/NEWS
Message-ID: <20100731214015.D10B0D435@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 23:40:15 2010
New Revision: 83368

Log:
#7909: the prefixes \\.\ and \\?\ indicate special Windows paths, do not try to manipulate them.  See http://msdn.microsoft.com/en-us/library/aa365247%28VS.85%29.aspx for details.

Modified:
   python/branches/py3k/Lib/ntpath.py
   python/branches/py3k/Lib/test/test_ntpath.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/ntpath.py
==============================================================================
--- python/branches/py3k/Lib/ntpath.py	(original)
+++ python/branches/py3k/Lib/ntpath.py	Sat Jul 31 23:40:15 2010
@@ -71,6 +71,12 @@
     else:
         return ':'
 
+def _get_special(path):
+    if isinstance(path, bytes):
+        return (b'\\\\.\\', b'\\\\?\\')
+    else:
+        return ('\\\\.\\', '\\\\?\\')
+
 # Normalize the case of a pathname and map slashes to backslashes.
 # Other normalizations (such as optimizing '../' away) are not done
 # (this is done by normpath).
@@ -524,6 +530,13 @@
     """Normalize path, eliminating double slashes, etc."""
     sep = _get_sep(path)
     dotdot = _get_dot(path) * 2
+    special_prefixes = _get_special(path)
+    if path.startswith(special_prefixes):
+        # in the case of paths with these prefixes:
+        # \\.\ -> device names
+        # \\?\ -> literal paths
+        # do not do any normalization, but return the path unchanged
+        return path
     path = path.replace(_get_altsep(path), sep)
     prefix, path = splitdrive(path)
 

Modified: python/branches/py3k/Lib/test/test_ntpath.py
==============================================================================
--- python/branches/py3k/Lib/test/test_ntpath.py	(original)
+++ python/branches/py3k/Lib/test/test_ntpath.py	Sat Jul 31 23:40:15 2010
@@ -174,6 +174,9 @@
         tester("ntpath.normpath('C:////a/b')", r'C:\a\b')
         tester("ntpath.normpath('//machine/share//a/b')", r'\\machine\share\a\b')
 
+        tester("ntpath.normpath('\\\\.\\NUL')", r'\\.\NUL')
+        tester("ntpath.normpath('\\\\?\\D:/XY\\Z')", r'\\?\D:/XY\Z')
+
     def test_expandvars(self):
         with support.EnvironmentVarGuard() as env:
             env.clear()

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sat Jul 31 23:40:15 2010
@@ -15,6 +15,9 @@
 Library
 -------
 
+- Issue #7909: Do not touch paths with the special prefixes ``\\.\``
+  or ``\\?\`` in ntpath.normpath().
+
 - Issue #1286: Allow using fileinput.FileInput as a context manager.
 
 - Add lfu_cache() and lru_cache() decorators to the functools module.

From python-checkins at python.org  Sat Jul 31 23:41:43 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 23:41:43 +0200 (CEST)
Subject: [Python-checkins] r83369 - in python/branches/py3k:
	Doc/library/dbm.rst Misc/NEWS
Message-ID: <20100731214143.00911EE98B@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 23:41:42 2010
New Revision: 83369

Log:
Fix "Berkeley" name.

Modified:
   python/branches/py3k/Doc/library/dbm.rst
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Doc/library/dbm.rst
==============================================================================
--- python/branches/py3k/Doc/library/dbm.rst	(original)
+++ python/branches/py3k/Doc/library/dbm.rst	Sat Jul 31 23:41:42 2010
@@ -8,7 +8,7 @@
 :mod:`dbm.gnu` or :mod:`dbm.ndbm`.  If none of these modules is installed, the
 slow-but-simple implementation in module :mod:`dbm.dumb` will be used.  There
 is a `third party interface <http://www.jcea.es/programacion/pybsddb.htm>`_ to
-the Oracle Berkely DB.
+the Oracle Berkeley DB.
 
 
 .. exception:: error

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sat Jul 31 23:41:42 2010
@@ -2393,8 +2393,8 @@
   an error.  The _PY_STRUCT_FLOAT_COERCE constant has been removed.
   The version number has been bumped to 0.3.
 
-- Issue #5359: Readd the Berkley-DB detection code to allow _dbm be built
-  using Berkley-DB.
+- Issue #5359: Readd the Berkeley DB detection code to allow _dbm be built
+  using Berkeley DB.
 
 Tests
 -----

From python-checkins at python.org  Sat Jul 31 23:51:48 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 23:51:48 +0200 (CEST)
Subject: [Python-checkins] r83370 - in python/branches/py3k: Lib/pydoc.py
	Misc/NEWS
Message-ID: <20100731215148.48D90EE994@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 23:51:48 2010
New Revision: 83370

Log:
#8198: the Helper class should not save the stdin and stdout objects
at import time, rather by default use the current streams like the
other APIs that output help.



Modified:
   python/branches/py3k/Lib/pydoc.py
   python/branches/py3k/Misc/NEWS

Modified: python/branches/py3k/Lib/pydoc.py
==============================================================================
--- python/branches/py3k/Lib/pydoc.py	(original)
+++ python/branches/py3k/Lib/pydoc.py	Sat Jul 31 23:51:48 2010
@@ -1695,9 +1695,17 @@
         'CONTEXTMANAGERS': ('context-managers', 'with'),
     }
 
-    def __init__(self, input, output):
-        self.input = input
-        self.output = output
+    def __init__(self, input=None, output=None):
+        self._input = input
+        self._output = output
+
+    @property
+    def input(self):
+        return self._input or sys.stdin
+
+    @property
+    def output(self):
+        return self._output or sys.stdout
 
     def __repr__(self):
         if inspect.stack()[1][3] == '?':
@@ -1874,7 +1882,7 @@
 for modules whose descriptions contain the word "spam".
 ''')
 
-help = Helper(sys.stdin, sys.stdout)
+help = Helper()
 
 class Scanner:
     """A generic tree iterator."""

Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS	(original)
+++ python/branches/py3k/Misc/NEWS	Sat Jul 31 23:51:48 2010
@@ -15,6 +15,9 @@
 Library
 -------
 
+- Issue #8198: In pydoc, output all help text to the correct stream
+  when sys.stdout is reassigned.
+
 - Issue #7909: Do not touch paths with the special prefixes ``\\.\``
   or ``\\?\`` in ntpath.normpath().
 

From python-checkins at python.org  Sat Jul 31 23:54:25 2010
From: python-checkins at python.org (georg.brandl)
Date: Sat, 31 Jul 2010 23:54:25 +0200 (CEST)
Subject: [Python-checkins] r83371 - in python/branches/py3k/Lib:
	distutils/command/sdist.py http/server.py platform.py
Message-ID: <20100731215425.27C53EE994@mail.python.org>

Author: georg.brandl
Date: Sat Jul 31 23:54:24 2010
New Revision: 83371

Log:
#8292: Fix three instances of truth tests on return values of filter() (which is always true in Python 3).

Modified:
   python/branches/py3k/Lib/distutils/command/sdist.py
   python/branches/py3k/Lib/http/server.py
   python/branches/py3k/Lib/platform.py

Modified: python/branches/py3k/Lib/distutils/command/sdist.py
==============================================================================
--- python/branches/py3k/Lib/distutils/command/sdist.py	(original)
+++ python/branches/py3k/Lib/distutils/command/sdist.py	Sat Jul 31 23:54:24 2010
@@ -240,8 +240,7 @@
         optional = ['test/test*.py', 'setup.cfg']
         for pattern in optional:
             files = filter(os.path.isfile, glob(pattern))
-            if files:
-                self.filelist.extend(files)
+            self.filelist.extend(files)
 
         # build_py is used to get:
         #  - python modules

Modified: python/branches/py3k/Lib/http/server.py
==============================================================================
--- python/branches/py3k/Lib/http/server.py	(original)
+++ python/branches/py3k/Lib/http/server.py	Sat Jul 31 23:54:24 2010
@@ -1023,8 +1023,9 @@
         if ua:
             env['HTTP_USER_AGENT'] = ua
         co = filter(None, self.headers.get_all('cookie', []))
-        if co:
-            env['HTTP_COOKIE'] = ', '.join(co)
+        cookie_str = ', '.join(co)
+        if cookie_str:
+            env['HTTP_COOKIE'] = cookie_str
         # XXX Other HTTP_* headers
         # Since we're setting the env in the parent, provide empty
         # values to override previously set values

Modified: python/branches/py3k/Lib/platform.py
==============================================================================
--- python/branches/py3k/Lib/platform.py	(original)
+++ python/branches/py3k/Lib/platform.py	Sat Jul 31 23:54:24 2010
@@ -1137,7 +1137,7 @@
     except AttributeError:
         no_os_uname = 1
 
-    if no_os_uname or not filter(None, (system, node, release, version, machine)):
+    if no_os_uname or not list(filter(None, (system, node, release, version, machine))):
         # Hmm, no there is either no uname or uname has returned
         #'unknowns'... we'll have to poke around the system then.
         if no_os_uname:

From richardjones at optushome.com.au  Wed Jul 28 04:03:31 2010
From: richardjones at optushome.com.au (Richard Jones)
Date: Wed, 28 Jul 2010 02:03:31 -0000
Subject: [Python-checkins] [Pypi-checkins] r829 - trunk/pypi
In-Reply-To: <20100726154841.D3F32EE9C5@mail.python.org>
References: <20100726154841.D3F32EE9C5@mail.python.org>
Message-ID: <AANLkTikPwTxkb3b+4_4755bWwkabyqAydSXaV9vb93rD@mail.gmail.com>

Hi Georg,

I just wanted to check that you are able to update the live code when
changes like this need to be made.


     Richard

On Tue, Jul 27, 2010 at 1:48 AM, georg.brandl
<python-checkins at python.org> wrote:
> Author: georg.brandl
> Date: Mon Jul 26 17:48:41 2010
> New Revision: 829
>
> Modified:
> ? trunk/pypi/webui.py
> Log:
> Check for sufficient user rights on role form submit.
>
> Modified: trunk/pypi/webui.py
> ==============================================================================
> --- trunk/pypi/webui.py (original)
> +++ trunk/pypi/webui.py Mon Jul 26 17:48:41 2010
> @@ -935,6 +935,11 @@
> ? ? ? ? user_name = self.form['user_name']
> ? ? ? ? role_name = self.form['role_name']
>
> + ? ? ? ?# make sure only admins and owners can add roles
> + ? ? ? ?if not (self.store.has_role('Admin', package_name) or
> + ? ? ? ? ? ? ? ?self.store.has_role('Owner', package_name)):
> + ? ? ? ? ? ?raise Unauthorised
> +
> ? ? ? ? # further vali:dation
> ? ? ? ? if role_name not in ('Owner', 'Maintainer'):
> ? ? ? ? ? ? raise FormError, 'role_name not Owner or Maintainer'
> _______________________________________________
> Pypi-checkins mailing list
> Pypi-checkins at python.org
> http://mail.python.org/mailman/listinfo/pypi-checkins
>